1529 files changed, 37406 insertions, 27610 deletions

diff --git a/Documentation/cgroup-v2.txt b/Documentation/cgroup-v2.txt
index e8d25e784214..ff49cf901148 100644
--- a/Documentation/cgroup-v2.txt
+++ b/Documentation/cgroup-v2.txt
@@ -7,7 +7,7 @@ This is the authoritative documentation on the design, interface and
 conventions of cgroup v2.  It describes all userland-visible aspects
 of cgroup including core and specific controller behaviors.  All
 future changes must be reflected in this document.  Documentation for
-v1 is available under Documentation/cgroup-legacy/.
+v1 is available under Documentation/cgroup-v1/.
 
 CONTENTS
 
diff --git a/Documentation/devicetree/bindings/arm/omap/omap.txt b/Documentation/devicetree/bindings/arm/omap/omap.txt
index a2bd593881ca..66422d663184 100644
--- a/Documentation/devicetree/bindings/arm/omap/omap.txt
+++ b/Documentation/devicetree/bindings/arm/omap/omap.txt
@@ -23,6 +23,7 @@ Optional properties:
   during suspend.
 - ti,no-reset-on-init: When present, the module should not be reset at init
 - ti,no-idle-on-init: When present, the module should not be idled at init
+- ti,no-idle: When present, the module is never allowed to idle.
 
 Example:
 
diff --git a/Documentation/devicetree/bindings/clock/rockchip,rk3036-cru.txt b/Documentation/devicetree/bindings/clock/rockchip,rk3036-cru.txt
index ace05992a262..20df350b9ef3 100644
--- a/Documentation/devicetree/bindings/clock/rockchip,rk3036-cru.txt
+++ b/Documentation/devicetree/bindings/clock/rockchip,rk3036-cru.txt
@@ -30,7 +30,7 @@ that they are defined using standard clock bindings with following
 clock-output-names:
  - "xin24m" - crystal input - required,
  - "ext_i2s" - external I2S clock - optional,
- - "ext_gmac" - external GMAC clock - optional
+ - "rmii_clkin" - external EMAC clock - optional
 
 Example: Clock controller node:
 
diff --git a/Documentation/devicetree/bindings/display/brcm,bcm-vc4.txt b/Documentation/devicetree/bindings/display/brcm,bcm-vc4.txt
index 56a961aa5061..9f97df4d5152 100644
--- a/Documentation/devicetree/bindings/display/brcm,bcm-vc4.txt
+++ b/Documentation/devicetree/bindings/display/brcm,bcm-vc4.txt
@@ -35,6 +35,12 @@ Optional properties for HDMI:
 		  as an interrupt/status bit in the HDMI controller
 		  itself).  See bindings/pinctrl/brcm,bcm2835-gpio.txt
 
+Required properties for V3D:
+- compatible:	Should be "brcm,bcm2835-v3d"
+- reg:		Physical base address and length of the V3D's registers
+- interrupts:	The interrupt number
+		  See bindings/interrupt-controller/brcm,bcm2835-armctrl-ic.txt
+
 Example:
 pixelvalve@7e807000 {
 	compatible = "brcm,bcm2835-pixelvalve2";
@@ -60,6 +66,12 @@ hdmi: hdmi@7e902000 {
 	clock-names = "pixel", "hdmi";
 };
 
+v3d: v3d@7ec00000 {
+	compatible = "brcm,bcm2835-v3d";
+	reg = <0x7ec00000 0x1000>;
+	interrupts = <1 10>;
+};
+
 vc4: gpu {
 	compatible = "brcm,bcm2835-vc4";
 };
diff --git a/Documentation/devicetree/bindings/display/exynos/exynos_dsim.txt b/Documentation/devicetree/bindings/display/exynos/exynos_dsim.txt
index 0e6f0c024858..22756b3dede2 100644
--- a/Documentation/devicetree/bindings/display/exynos/exynos_dsim.txt
+++ b/Documentation/devicetree/bindings/display/exynos/exynos_dsim.txt
@@ -6,6 +6,7 @@ Required properties:
 		"samsung,exynos4210-mipi-dsi" /* for Exynos4 SoCs */
 		"samsung,exynos4415-mipi-dsi" /* for Exynos4415 SoC */
 		"samsung,exynos5410-mipi-dsi" /* for Exynos5410/5420/5440 SoCs */
+		"samsung,exynos5422-mipi-dsi" /* for Exynos5422/5800 SoCs */
 		"samsung,exynos5433-mipi-dsi" /* for Exynos5433 SoCs */
   - reg: physical base address and length of the registers set for the device
   - interrupts: should contain DSI interrupt
diff --git a/Documentation/devicetree/bindings/display/exynos/samsung-fimd.txt b/Documentation/devicetree/bindings/display/exynos/samsung-fimd.txt
index 27c3ce0db16a..c7c6b9af87ac 100644
--- a/Documentation/devicetree/bindings/display/exynos/samsung-fimd.txt
+++ b/Documentation/devicetree/bindings/display/exynos/samsung-fimd.txt
@@ -12,7 +12,8 @@ Required properties:
 		"samsung,exynos3250-fimd"; /* for Exynos3250/3472 SoCs */
 		"samsung,exynos4210-fimd"; /* for Exynos4 SoCs */
 		"samsung,exynos4415-fimd"; /* for Exynos4415 SoC */
-		"samsung,exynos5250-fimd"; /* for Exynos5 SoCs */
+		"samsung,exynos5250-fimd"; /* for Exynos5250 SoCs */
+		"samsung,exynos5420-fimd"; /* for Exynos5420/5422/5800 SoCs */
 
 - reg: physical base address and length of the FIMD registers set.
 
diff --git a/Documentation/devicetree/bindings/display/msm/dsi.txt b/Documentation/devicetree/bindings/display/msm/dsi.txt
index e7423bea1424..f5948c48b9a2 100644
--- a/Documentation/devicetree/bindings/display/msm/dsi.txt
+++ b/Documentation/devicetree/bindings/display/msm/dsi.txt
@@ -44,9 +44,34 @@ Optional properties:
 - pinctrl-names: the pin control state names; should contain "default"
 - pinctrl-0: the default pinctrl state (active)
 - pinctrl-n: the "sleep" pinctrl state
-- port: DSI controller output port. This contains one endpoint subnode, with its
-  remote-endpoint set to the phandle of the connected panel's endpoint.
-  See Documentation/devicetree/bindings/graph.txt for device graph info.
+- port: DSI controller output port, containing one endpoint subnode.
+
+  DSI Endpoint properties:
+  - remote-endpoint: set to phandle of the connected panel's endpoint.
+    See Documentation/devicetree/bindings/graph.txt for device graph info.
+  - qcom,data-lane-map: this describes how the logical DSI lanes are mapped
+    to the physical lanes on the given platform. The value contained in
+    index n describes what logical data lane is mapped to the physical data
+    lane n (DATAn, where n lies between 0 and 3).
+
+    For example:
+
+    qcom,data-lane-map = <3 0 1 2>;
+
+    The above mapping describes that the logical data lane DATA3 is mapped to
+    the physical data lane DATA0, logical DATA0 to physical DATA1, logic DATA1
+    to phys DATA2 and logic DATA2 to phys DATA3.
+
+    There are only a limited number of physical to logical mappings possible:
+
+    "0123": Logic 0->Phys 0; Logic 1->Phys 1; Logic 2->Phys 2; Logic 3->Phys 3;
+    "3012": Logic 3->Phys 0; Logic 0->Phys 1; Logic 1->Phys 2; Logic 2->Phys 3;
+    "2301": Logic 2->Phys 0; Logic 3->Phys 1; Logic 0->Phys 2; Logic 1->Phys 3;
+    "1230": Logic 1->Phys 0; Logic 2->Phys 1; Logic 3->Phys 2; Logic 0->Phys 3;
+    "0321": Logic 0->Phys 0; Logic 3->Phys 1; Logic 2->Phys 2; Logic 1->Phys 3;
+    "1032": Logic 1->Phys 0; Logic 0->Phys 1; Logic 3->Phys 2; Logic 2->Phys 3;
+    "2103": Logic 2->Phys 0; Logic 1->Phys 1; Logic 0->Phys 2; Logic 3->Phys 3;
+    "3210": Logic 3->Phys 0; Logic 2->Phys 1; Logic 1->Phys 2; Logic 0->Phys 3;
 
 DSI PHY:
 Required properties:
@@ -131,6 +156,7 @@ Example:
 		port {
 			dsi0_out: endpoint {
 				remote-endpoint = <&panel_in>;
+				lanes = <0 1 2 3>;
 			};
 		};
 	};
diff --git a/Documentation/devicetree/bindings/display/msm/hdmi.txt b/Documentation/devicetree/bindings/display/msm/hdmi.txt
index 379ee2ea9a3d..b63f614e0c04 100644
--- a/Documentation/devicetree/bindings/display/msm/hdmi.txt
+++ b/Documentation/devicetree/bindings/display/msm/hdmi.txt
@@ -11,6 +11,7 @@ Required properties:
 - reg: Physical base address and length of the controller's registers
 - reg-names: "core_physical"
 - interrupts: The interrupt signal from the hdmi block.
+- power-domains: Should be <&mmcc MDSS_GDSC>.
 - clocks: device clocks
   See ../clocks/clock-bindings.txt for details.
 - qcom,hdmi-tx-ddc-clk-gpio: ddc clk pin
@@ -18,6 +19,8 @@ Required properties:
 - qcom,hdmi-tx-hpd-gpio: hpd pin
 - core-vdda-supply: phandle to supply regulator
 - hdmi-mux-supply: phandle to mux regulator
+- phys: the phandle for the HDMI PHY device
+- phy-names: the name of the corresponding PHY device
 
 Optional properties:
 - qcom,hdmi-tx-mux-en-gpio: hdmi mux enable pin
@@ -27,15 +30,38 @@ Optional properties:
 - pinctrl-0: the default pinctrl state (active)
 - pinctrl-1: the "sleep" pinctrl state
 
+HDMI PHY:
+Required properties:
+- compatible: Could be the following
+  * "qcom,hdmi-phy-8660"
+  * "qcom,hdmi-phy-8960"
+  * "qcom,hdmi-phy-8974"
+  * "qcom,hdmi-phy-8084"
+  * "qcom,hdmi-phy-8996"
+- #phy-cells: Number of cells in a PHY specifier; Should be 0.
+- reg: Physical base address and length of the registers of the PHY sub blocks.
+- reg-names: The names of register regions. The following regions are required:
+  * "hdmi_phy"
+  * "hdmi_pll"
+  For HDMI PHY on msm8996, these additional register regions are required:
+    * "hdmi_tx_l0"
+    * "hdmi_tx_l1"
+    * "hdmi_tx_l3"
+    * "hdmi_tx_l4"
+- power-domains: Should be <&mmcc MDSS_GDSC>.
+- clocks: device clocks
+  See Documentation/devicetree/bindings/clocks/clock-bindings.txt for details.
+- core-vdda-supply: phandle to vdda regulator device node
+
 Example:
 
 / {
 	...
 
-	hdmi: qcom,hdmi-tx-8960@4a00000 {
+	hdmi: hdmi@4a00000 {
 		compatible = "qcom,hdmi-tx-8960";
 		reg-names = "core_physical";
-		reg = <0x04a00000 0x1000>;
+		reg = <0x04a00000 0x2f0>;
 		interrupts = <GIC_SPI 79 0>;
 		power-domains = <&mmcc MDSS_GDSC>;
 		clock-names =
@@ -54,5 +80,21 @@ Example:
 		pinctrl-names = "default", "sleep";
 		pinctrl-0 = <&hpd_active  &ddc_active  &cec_active>;
 		pinctrl-1 = <&hpd_suspend &ddc_suspend &cec_suspend>;
+
+		phys = <&hdmi_phy>;
+		phy-names = "hdmi_phy";
+	};
+
+	hdmi_phy: phy@4a00400 {
+		compatible = "qcom,hdmi-phy-8960";
+		reg-names = "hdmi_phy",
+			    "hdmi_pll";
+		reg = <0x4a00400 0x60>,
+		      <0x4a00500 0x100>;
+		#phy-cells = <0>;
+		power-domains = <&mmcc MDSS_GDSC>;
+		clock-names = "slave_iface_clk";
+		clocks = <&mmcc HDMI_S_AHB_CLK>;
+		core-vdda-supply = <&pm8921_hdmi_mvs>;
 	};
 };
diff --git a/Documentation/devicetree/bindings/display/panel/lg,lp120up1.txt b/Documentation/devicetree/bindings/display/panel/lg,lp120up1.txt
new file mode 100644
index 000000000000..8c5de692c55c
--- /dev/null
+++ b/Documentation/devicetree/bindings/display/panel/lg,lp120up1.txt
@@ -0,0 +1,7 @@
+LG 12.0" (1920x1280 pixels) TFT LCD panel
+
+Required properties:
+- compatible: should be "lg,lp120up1"
+
+This binding is compatible with the simple-panel binding, which is specified
+in simple-panel.txt in this directory.
diff --git a/Documentation/devicetree/bindings/display/panel/urt,umsh-8596md.txt b/Documentation/devicetree/bindings/display/panel/urt,umsh-8596md.txt
new file mode 100644
index 000000000000..088a6cea5015
--- /dev/null
+++ b/Documentation/devicetree/bindings/display/panel/urt,umsh-8596md.txt
@@ -0,0 +1,16 @@
+United Radiant Technology UMSH-8596MD-xT 7.0" WVGA TFT LCD panel
+
+Supported are LVDS versions (-11T, -19T) and parallel ones
+(-T, -1T, -7T, -20T).
+
+Required properties:
+- compatible: should be one of:
+  "urt,umsh-8596md-t",
+  "urt,umsh-8596md-1t",
+  "urt,umsh-8596md-7t",
+  "urt,umsh-8596md-11t",
+  "urt,umsh-8596md-19t",
+  "urt,umsh-8596md-20t".
+
+This binding is compatible with the simple-panel binding, which is specified
+in simple-panel.txt in this directory.
diff --git a/Documentation/devicetree/bindings/interrupt-controller/arm,gic-v3.txt b/Documentation/devicetree/bindings/interrupt-controller/arm,gic-v3.txt
index 7803e77d85cb..007a5b46256a 100644
--- a/Documentation/devicetree/bindings/interrupt-controller/arm,gic-v3.txt
+++ b/Documentation/devicetree/bindings/interrupt-controller/arm,gic-v3.txt
@@ -24,9 +24,8 @@ Main node required properties:
 		1 = edge triggered
 		4 = level triggered
 
-  Cells 4 and beyond are reserved for future use. When the 1st cell
-  has a value of 0 or 1, cells 4 and beyond act as padding, and may be
-  ignored. It is recommended that padding cells have a value of 0.
+  Cells 4 and beyond are reserved for future use and must have a value
+  of 0 if present.
 
 - reg : Specifies base physical address(s) and size of the GIC
   registers, in the following order:
diff --git a/Documentation/devicetree/bindings/net/renesas,ravb.txt b/Documentation/devicetree/bindings/net/renesas,ravb.txt
index 81a9f9e6b45f..c8ac222eac67 100644
--- a/Documentation/devicetree/bindings/net/renesas,ravb.txt
+++ b/Documentation/devicetree/bindings/net/renesas,ravb.txt
@@ -82,8 +82,8 @@ Example:
 				  "ch16", "ch17", "ch18", "ch19",
 				  "ch20", "ch21", "ch22", "ch23",
 				  "ch24";
-		clocks = <&mstp8_clks R8A7795_CLK_ETHERAVB>;
-		power-domains = <&cpg_clocks>;
+		clocks = <&cpg CPG_MOD 812>;
+		power-domains = <&cpg>;
 		phy-mode = "rgmii-id";
 		phy-handle = <&phy0>;
 
diff --git a/Documentation/devicetree/bindings/pci/pci-rcar-gen2.txt b/Documentation/devicetree/bindings/pci/pci-rcar-gen2.txt
index 4e8b90e43dd8..07a75094c5a8 100644
--- a/Documentation/devicetree/bindings/pci/pci-rcar-gen2.txt
+++ b/Documentation/devicetree/bindings/pci/pci-rcar-gen2.txt
@@ -8,6 +8,7 @@ OHCI and EHCI controllers.
 Required properties:
 - compatible: "renesas,pci-r8a7790" for the R8A7790 SoC;
 	      "renesas,pci-r8a7791" for the R8A7791 SoC;
+	      "renesas,pci-r8a7793" for the R8A7793 SoC;
 	      "renesas,pci-r8a7794" for the R8A7794 SoC;
 	      "renesas,pci-rcar-gen2" for a generic R-Car Gen2 compatible device
 
diff --git a/Documentation/devicetree/bindings/pci/rcar-pci.txt b/Documentation/devicetree/bindings/pci/rcar-pci.txt
index 558fe528ae19..6cf99690eef9 100644
--- a/Documentation/devicetree/bindings/pci/rcar-pci.txt
+++ b/Documentation/devicetree/bindings/pci/rcar-pci.txt
@@ -4,6 +4,7 @@ Required properties:
 compatible: "renesas,pcie-r8a7779" for the R8A7779 SoC;
 	    "renesas,pcie-r8a7790" for the R8A7790 SoC;
 	    "renesas,pcie-r8a7791" for the R8A7791 SoC;
+	    "renesas,pcie-r8a7793" for the R8A7793 SoC;
 	    "renesas,pcie-r8a7795" for the R8A7795 SoC;
 	    "renesas,pcie-rcar-gen2" for a generic R-Car Gen2 compatible device.
 
diff --git a/Documentation/devicetree/bindings/regulator/tps65217.txt b/Documentation/devicetree/bindings/regulator/tps65217.txt
index d18109657da6..4f05d208c95c 100644
--- a/Documentation/devicetree/bindings/regulator/tps65217.txt
+++ b/Documentation/devicetree/bindings/regulator/tps65217.txt
@@ -26,11 +26,7 @@ Example:
 		ti,pmic-shutdown-controller;
 
 		regulators {
-			#address-cells = <1>;
-			#size-cells = <0>;
-
 			dcdc1_reg: dcdc1 {
-				reg = <0>;
 				regulator-min-microvolt = <900000>;
 				regulator-max-microvolt = <1800000>;
 				regulator-boot-on;
@@ -38,7 +34,6 @@ Example:
 			};
 
 			dcdc2_reg: dcdc2 {
-				reg = <1>;
 				regulator-min-microvolt = <900000>;
 				regulator-max-microvolt = <3300000>;
 				regulator-boot-on;
@@ -46,7 +41,6 @@ Example:
 			};
 
 			dcdc3_reg: dcc3 {
-				reg = <2>;
 				regulator-min-microvolt = <900000>;
 				regulator-max-microvolt = <1500000>;
 				regulator-boot-on;
@@ -54,7 +48,6 @@ Example:
 			};
 
 			ldo1_reg: ldo1 {
-				reg = <3>;
 				regulator-min-microvolt = <1000000>;
 				regulator-max-microvolt = <3300000>;
 				regulator-boot-on;
@@ -62,7 +55,6 @@ Example:
 			};
 
 			ldo2_reg: ldo2 {
-				reg = <4>;
 				regulator-min-microvolt = <900000>;
 				regulator-max-microvolt = <3300000>;
 				regulator-boot-on;
@@ -70,7 +62,6 @@ Example:
 			};
 
 			ldo3_reg: ldo3 {
-				reg = <5>;
 				regulator-min-microvolt = <1800000>;
 				regulator-max-microvolt = <3300000>;
 				regulator-boot-on;
@@ -78,7 +69,6 @@ Example:
 			};
 
 			ldo4_reg: ldo4 {
-				reg = <6>;
 				regulator-min-microvolt = <1800000>;
 				regulator-max-microvolt = <3300000>;
 				regulator-boot-on;
diff --git a/Documentation/devicetree/bindings/rtc/s3c-rtc.txt b/Documentation/devicetree/bindings/rtc/s3c-rtc.txt
index ac2fcd6ff4b8..1068ffce9f91 100644
--- a/Documentation/devicetree/bindings/rtc/s3c-rtc.txt
+++ b/Documentation/devicetree/bindings/rtc/s3c-rtc.txt
@@ -14,6 +14,10 @@ Required properties:
   interrupt number is the rtc alarm interrupt and second interrupt number
   is the rtc tick interrupt. The number of cells representing a interrupt
   depends on the parent interrupt controller.
+- clocks: Must contain a list of phandle and clock specifier for the rtc
+          and source clocks.
+- clock-names: Must contain "rtc" and "rtc_src" entries sorted in the
+               same order as the clocks property.
 
 Example:
 
@@ -21,4 +25,6 @@ Example:
 		compatible = "samsung,s3c6410-rtc";
 		reg = <0x10070000 0x100>;
 		interrupts = <44 0 45 0>;
+		clocks = <&clock CLK_RTC>, <&s2mps11_osc S2MPS11_CLK_AP>;
+		clock-names = "rtc", "rtc_src";
 	};
diff --git a/Documentation/devicetree/bindings/serial/fsl-imx-uart.txt b/Documentation/devicetree/bindings/serial/fsl-imx-uart.txt
index 35ae1fb3537f..ed94c217c98d 100644
--- a/Documentation/devicetree/bindings/serial/fsl-imx-uart.txt
+++ b/Documentation/devicetree/bindings/serial/fsl-imx-uart.txt
@@ -9,7 +9,7 @@ Optional properties:
 - fsl,uart-has-rtscts : Indicate the uart has rts and cts
 - fsl,irda-mode : Indicate the uart supports irda mode
 - fsl,dte-mode : Indicate the uart works in DTE mode. The uart works
-                  is DCE mode by default.
+                  in DCE mode by default.
 
 Note: Each uart controller should have an alias correctly numbered
 in "aliases" node.
diff --git a/Documentation/devicetree/bindings/sound/fsl-asoc-card.txt b/Documentation/devicetree/bindings/sound/fsl-asoc-card.txt
index ce55c0a6f757..4da41bf1888e 100644
--- a/Documentation/devicetree/bindings/sound/fsl-asoc-card.txt
+++ b/Documentation/devicetree/bindings/sound/fsl-asoc-card.txt
@@ -30,6 +30,8 @@ The compatible list for this generic sound card currently:
  "fsl,imx-audio-sgtl5000"
  (compatible with Documentation/devicetree/bindings/sound/imx-audio-sgtl5000.txt)
 
+ "fsl,imx-audio-wm8960"
+
 Required properties:
 
   - compatible		: Contains one of entries in the compatible list.
diff --git a/Documentation/devicetree/bindings/thermal/rcar-thermal.txt b/Documentation/devicetree/bindings/thermal/rcar-thermal.txt
index 332e625f6ed0..e5ee3f159893 100644
--- a/Documentation/devicetree/bindings/thermal/rcar-thermal.txt
+++ b/Documentation/devicetree/bindings/thermal/rcar-thermal.txt
@@ -1,8 +1,9 @@
 * Renesas R-Car Thermal
 
 Required properties:
-- compatible		: "renesas,thermal-<soctype>", "renesas,rcar-thermal"
-			  as fallback.
+- compatible		: "renesas,thermal-<soctype>",
+			   "renesas,rcar-gen2-thermal" (with thermal-zone) or
+			   "renesas,rcar-thermal" (without thermal-zone) as fallback.
 			  Examples with soctypes are:
 			    - "renesas,thermal-r8a73a4" (R-Mobile APE6)
 			    - "renesas,thermal-r8a7779" (R-Car H1)
@@ -36,3 +37,35 @@ thermal@e61f0000 {
 		0xe61f0300 0x38>;
 	interrupts = <0 69 IRQ_TYPE_LEVEL_HIGH>;
 };
+
+Example (with thermal-zone):
+
+thermal-zones {
+	cpu_thermal: cpu-thermal {
+		polling-delay-passive	= <1000>;
+		polling-delay		= <5000>;
+
+		thermal-sensors = <&thermal>;
+
+		trips {
+			cpu-crit {
+				temperature	= <115000>;
+				hysteresis	= <0>;
+				type		= "critical";
+			};
+		};
+		cooling-maps {
+		};
+	};
+};
+
+thermal: thermal@e61f0000 {
+	compatible =	"renesas,thermal-r8a7790",
+			"renesas,rcar-gen2-thermal",
+			"renesas,rcar-thermal";
+	reg = <0 0xe61f0000 0 0x14>, <0 0xe61f0100 0 0x38>;
+	interrupts = <0 69 IRQ_TYPE_LEVEL_HIGH>;
+	clocks = <&mstp5_clks R8A7790_CLK_THERMAL>;
+	power-domains = <&cpg_clocks>;
+	#thermal-sensor-cells = <0>;
+};
diff --git a/Documentation/devicetree/bindings/vendor-prefixes.txt b/Documentation/devicetree/bindings/vendor-prefixes.txt
index 72e2c5a2b327..1bcef7300b37 100644
--- a/Documentation/devicetree/bindings/vendor-prefixes.txt
+++ b/Documentation/devicetree/bindings/vendor-prefixes.txt
@@ -240,6 +240,7 @@ tplink	TP-LINK Technologies Co., Ltd.
 tronfy	Tronfy
 truly	Truly Semiconductors Limited
 upisemi	uPI Semiconductor Corp.
+urt	United Radiant Technology Corporation
 usi	Universal Scientific Industrial Co., Ltd.
 v3	V3 Semiconductor
 variscite	Variscite Ltd.
diff --git a/Documentation/filesystems/efivarfs.txt b/Documentation/filesystems/efivarfs.txt
index c477af086e65..686a64bba775 100644
--- a/Documentation/filesystems/efivarfs.txt
+++ b/Documentation/filesystems/efivarfs.txt
@@ -14,3 +14,10 @@ filesystem.
 efivarfs is typically mounted like this,
 
 	mount -t efivarfs none /sys/firmware/efi/efivars
+
+Due to the presence of numerous firmware bugs where removing non-standard
+UEFI variables causes the system firmware to fail to POST, efivarfs
+files that are not well-known standardized variables are created
+as immutable files.  This doesn't prevent removal - "chattr -i" will work -
+but it does prevent this kind of failure from being accomplished
+accidentally.
diff --git a/Documentation/kernel-parameters.txt b/Documentation/kernel-parameters.txt
index 551ecf09c8dd..9a53c929f017 100644
--- a/Documentation/kernel-parameters.txt
+++ b/Documentation/kernel-parameters.txt
@@ -4235,6 +4235,17 @@ bytes respectively. Such letter suffixes can also be entirely omitted.
 			The default value of this parameter is determined by
 			the config option CONFIG_WQ_POWER_EFFICIENT_DEFAULT.
 
+	workqueue.debug_force_rr_cpu
+			Workqueue used to implicitly guarantee that work
+			items queued without explicit CPU specified are put
+			on the local CPU.  This guarantee is no longer true
+			and while local CPU is still preferred work items
+			may be put on foreign CPUs.  This debug option
+			forces round-robin CPU selection to flush out
+			usages which depend on the now broken guarantee.
+			When enabled, memory and cache locality will be
+			impacted.
+
 	x2apic_phys	[X86-64,APIC] Use x2apic physical mode instead of
 			default x2apic cluster mode on platforms
 			supporting x2apic.
diff --git a/Documentation/timers/hpet.txt b/Documentation/timers/hpet.txt
index 767392ffd31e..a484d2c109d7 100644
--- a/Documentation/timers/hpet.txt
+++ b/Documentation/timers/hpet.txt
@@ -1,9 +1,7 @@
 		High Precision Event Timer Driver for Linux
 
 The High Precision Event Timer (HPET) hardware follows a specification
-by Intel and Microsoft which can be found at
-
-	http://www.intel.com/hardwaredesign/hpetspec_1.pdf
+by Intel and Microsoft, revision 1.
 
 Each HPET has one fixed-rate counter (at 10+ MHz, hence "High Precision")
 and up to 32 comparators.  Normally three or more comparators are provided,
diff --git a/Documentation/virtual/kvm/mmu.txt b/Documentation/virtual/kvm/mmu.txt
index daf9c0f742d2..c81731096a43 100644
--- a/Documentation/virtual/kvm/mmu.txt
+++ b/Documentation/virtual/kvm/mmu.txt
@@ -358,7 +358,8 @@ In the first case there are two additional complications:
 - if CR4.SMEP is enabled: since we've turned the page into a kernel page,
   the kernel may now execute it.  We handle this by also setting spte.nx.
   If we get a user fetch or read fault, we'll change spte.u=1 and
-  spte.nx=gpte.nx back.
+  spte.nx=gpte.nx back.  For this to work, KVM forces EFER.NX to 1 when
+  shadow paging is in use.
 - if CR4.SMAP is disabled: since the page has been changed to a kernel
   page, it can not be reused when CR4.SMAP is enabled. We set
   CR4.SMAP && !CR0.WP into shadow page's role to avoid this case. Note,
diff --git a/Documentation/watchdog/watchdog-parameters.txt b/Documentation/watchdog/watchdog-parameters.txt
index 9f9ec9f76039..4e4b6f10d841 100644
--- a/Documentation/watchdog/watchdog-parameters.txt
+++ b/Documentation/watchdog/watchdog-parameters.txt
@@ -400,3 +400,7 @@ wm8350_wdt:
 nowayout: Watchdog cannot be stopped once started
 	(default=kernel config parameter)
 -------------------------------------------------
+sun4v_wdt:
+timeout_ms: Watchdog timeout in milliseconds 1..180000, default=60000)
+nowayout: Watchdog cannot be stopped once started
+-------------------------------------------------
diff --git a/MAINTAINERS b/MAINTAINERS
index ea1d1deaf153..a32d2ffd81a8 100644
--- a/MAINTAINERS
+++ b/MAINTAINERS
@@ -926,17 +926,24 @@ M:	Emilio López <emilio@elopez.com.ar>
 S:	Maintained
 F:	drivers/clk/sunxi/
 
-ARM/Amlogic MesonX SoC support
+ARM/Amlogic Meson SoC support
 M:	Carlo Caione <carlo@caione.org>
 L:	linux-arm-kernel@lists.infradead.org (moderated for non-subscribers)
+L:	linux-meson@googlegroups.com
+W:	http://linux-meson.com/
 S:	Maintained
-F:	drivers/media/rc/meson-ir.c
-N:	meson[x68]
+F:	arch/arm/mach-meson/
+F:	arch/arm/boot/dts/meson*
+N:	meson
 
 ARM/Annapurna Labs ALPINE ARCHITECTURE
 M:	Tsahee Zidenberg <tsahee@annapurnalabs.com>
+M:	Antoine Tenart <antoine.tenart@free-electrons.com>
 S:	Maintained
 F:	arch/arm/mach-alpine/
+F:	arch/arm/boot/dts/alpine*
+F:	arch/arm64/boot/dts/al/
+F:	drivers/*/*alpine*
 
 ARM/ATMEL AT91RM9200, AT91SAM9 AND SAMA5 SOC SUPPORT
 M:	Nicolas Ferre <nicolas.ferre@atmel.com>
@@ -1448,8 +1455,8 @@ S:	Maintained
 ARM/RENESAS ARM64 ARCHITECTURE
 M:	Simon Horman <horms@verge.net.au>
 M:	Magnus Damm <magnus.damm@gmail.com>
-L:	linux-sh@vger.kernel.org
-Q:	http://patchwork.kernel.org/project/linux-sh/list/
+L:	linux-renesas-soc@vger.kernel.org
+Q:	http://patchwork.kernel.org/project/linux-renesas-soc/list/
 T:	git git://git.kernel.org/pub/scm/linux/kernel/git/horms/renesas.git next
 S:	Supported
 F:	arch/arm64/boot/dts/renesas/
@@ -2368,14 +2375,6 @@ T:	git git://git.kernel.org/pub/scm/linux/kernel/git/rpi/linux-rpi.git
 S:	Maintained
 N:	bcm2835
 
-BROADCOM BCM33XX MIPS ARCHITECTURE
-M:	Kevin Cernekee <cernekee@gmail.com>
-L:	linux-mips@linux-mips.org
-S:	Maintained
-F:	arch/mips/bcm3384/*
-F:	arch/mips/include/asm/mach-bcm3384/*
-F:	arch/mips/kernel/*bmips*
-
 BROADCOM BCM47XX MIPS ARCHITECTURE
 M:	Hauke Mehrtens <hauke@hauke-m.de>
 M:	Rafał Miłecki <zajec5@gmail.com>
@@ -3458,7 +3457,6 @@ F:	drivers/usb/dwc2/
 DESIGNWARE USB3 DRD IP DRIVER
 M:	Felipe Balbi <balbi@kernel.org>
 L:	linux-usb@vger.kernel.org
-L:	linux-omap@vger.kernel.org
 T:	git git://git.kernel.org/pub/scm/linux/kernel/git/balbi/usb.git
 S:	Maintained
 F:	drivers/usb/dwc3/
@@ -3712,7 +3710,7 @@ F:	drivers/gpu/vga/
 F:	include/drm/
 F:	include/uapi/drm/
 
-RADEON DRM DRIVERS
+RADEON and AMDGPU DRM DRIVERS
 M:	Alex Deucher <alexander.deucher@amd.com>
 M:	Christian König <christian.koenig@amd.com>
 L:	dri-devel@lists.freedesktop.org
@@ -3720,6 +3718,8 @@ T:	git git://people.freedesktop.org/~agd5f/linux
 S:	Supported
 F:	drivers/gpu/drm/radeon/
 F:	include/uapi/drm/radeon*
+F:	drivers/gpu/drm/amd/
+F:	include/uapi/drm/amdgpu*
 
 DRM PANEL DRIVERS
 M:	Thierry Reding <thierry.reding@gmail.com>
@@ -4526,6 +4526,12 @@ L:	linuxppc-dev@lists.ozlabs.org
 S:	Maintained
 F:	drivers/dma/fsldma.*
 
+FREESCALE GPMI NAND DRIVER
+M:	Han Xu <han.xu@nxp.com>
+L:	linux-mtd@lists.infradead.org
+S:	Maintained
+F:	drivers/mtd/nand/gpmi-nand/*
+
 FREESCALE I2C CPM DRIVER
 M:	Jochen Friedrich <jochen@scram.de>
 L:	linuxppc-dev@lists.ozlabs.org
@@ -4542,7 +4548,7 @@ F:	include/linux/platform_data/video-imxfb.h
 F:	drivers/video/fbdev/imxfb.c
 
 FREESCALE QUAD SPI DRIVER
-M:	Han Xu <han.xu@freescale.com>
+M:	Han Xu <han.xu@nxp.com>
 L:	linux-mtd@lists.infradead.org
 S:	Maintained
 F:	drivers/mtd/spi-nor/fsl-quadspi.c
@@ -4556,6 +4562,15 @@ S:	Maintained
 F:	drivers/net/ethernet/freescale/fs_enet/
 F:	include/linux/fs_enet_pd.h
 
+FREESCALE IMX / MXC FEC DRIVER
+M:	Fugang Duan <fugang.duan@nxp.com>
+L:	netdev@vger.kernel.org
+S:	Maintained
+F:	drivers/net/ethernet/freescale/fec_main.c
+F:	drivers/net/ethernet/freescale/fec_ptp.c
+F:	drivers/net/ethernet/freescale/fec.h
+F:	Documentation/devicetree/bindings/net/fsl-fec.txt
+
 FREESCALE QUICC ENGINE LIBRARY
 L:	linuxppc-dev@lists.ozlabs.org
 S:	Orphan
@@ -6142,7 +6157,7 @@ F:	include/uapi/linux/sunrpc/
 
 KERNEL SELFTEST FRAMEWORK
 M:	Shuah Khan <shuahkh@osg.samsung.com>
-L:	linux-api@vger.kernel.org
+L:	linux-kselftest@vger.kernel.org
 T:	git git://git.kernel.org/pub/scm/shuah/linux-kselftest
 S:	Maintained
 F:	tools/testing/selftests
@@ -6772,6 +6787,7 @@ S:	Maintained
 F:	Documentation/networking/mac80211-injection.txt
 F:	include/net/mac80211.h
 F:	net/mac80211/
+F:	drivers/net/wireless/mac80211_hwsim.[ch]
 
 MACVLAN DRIVER
 M:	Patrick McHardy <kaber@trash.net>
@@ -7368,7 +7384,7 @@ F:	drivers/tty/isicom.c
 F:	include/linux/isicom.h
 
 MUSB MULTIPOINT HIGH SPEED DUAL-ROLE CONTROLLER
-M:	Felipe Balbi <balbi@kernel.org>
+M:	Bin Liu <b-liu@ti.com>
 L:	linux-usb@vger.kernel.org
 T:	git git://git.kernel.org/pub/scm/linux/kernel/git/balbi/usb.git
 S:	Maintained
@@ -7700,13 +7716,13 @@ S:	Maintained
 F:	arch/nios2/
 
 NOKIA N900 POWER SUPPLY DRIVERS
-M:	Pali Rohár <pali.rohar@gmail.com>
-S:	Maintained
+R:	Pali Rohár <pali.rohar@gmail.com>
 F:	include/linux/power/bq2415x_charger.h
 F:	include/linux/power/bq27xxx_battery.h
 F:	include/linux/power/isp1704_charger.h
 F:	drivers/power/bq2415x_charger.c
 F:	drivers/power/bq27xxx_battery.c
+F:	drivers/power/bq27xxx_battery_i2c.c
 F:	drivers/power/isp1704_charger.c
 F:	drivers/power/rx51_battery.c
 
@@ -7937,11 +7953,9 @@ F:	drivers/media/platform/omap3isp/
 F:	drivers/staging/media/omap4iss/
 
 OMAP USB SUPPORT
-M:	Felipe Balbi <balbi@kernel.org>
 L:	linux-usb@vger.kernel.org
 L:	linux-omap@vger.kernel.org
-T:	git git://git.kernel.org/pub/scm/linux/kernel/git/balbi/usb.git
-S:	Maintained
+S:	Orphan
 F:	drivers/usb/*/*omap*
 F:	arch/arm/*omap*/usb*
 
@@ -9572,6 +9586,12 @@ M:	Andreas Noever <andreas.noever@gmail.com>
 S:	Maintained
 F:	drivers/thunderbolt/
 
+TI BQ27XXX POWER SUPPLY DRIVER
+R:	Andrew F. Davis <afd@ti.com>
+F:	include/linux/power/bq27xxx_battery.h
+F:	drivers/power/bq27xxx_battery.c
+F:	drivers/power/bq27xxx_battery_i2c.c
+
 TIMEKEEPING, CLOCKSOURCE CORE, NTP, ALARMTIMER
 M:	John Stultz <john.stultz@linaro.org>
 M:	Thomas Gleixner <tglx@linutronix.de>
@@ -9793,10 +9813,11 @@ S:	Supported
 F:	drivers/scsi/be2iscsi/
 
 Emulex 10Gbps NIC BE2, BE3-R, Lancer, Skyhawk-R DRIVER
-M:	Sathya Perla <sathya.perla@avagotech.com>
-M:	Ajit Khaparde <ajit.khaparde@avagotech.com>
-M:	Padmanabh Ratnakar <padmanabh.ratnakar@avagotech.com>
-M:	Sriharsha Basavapatna <sriharsha.basavapatna@avagotech.com>
+M:	Sathya Perla <sathya.perla@broadcom.com>
+M:	Ajit Khaparde <ajit.khaparde@broadcom.com>
+M:	Padmanabh Ratnakar <padmanabh.ratnakar@broadcom.com>
+M:	Sriharsha Basavapatna <sriharsha.basavapatna@broadcom.com>
+M:	Somnath Kotur <somnath.kotur@broadcom.com>
 L:	netdev@vger.kernel.org
 W:	http://www.emulex.com
 S:	Supported
@@ -12034,7 +12055,6 @@ F:	arch/arm64/xen/
 F:	arch/arm64/include/asm/xen/
 
 XEN NETWORK BACKEND DRIVER
-M:	Ian Campbell <ian.campbell@citrix.com>
 M:	Wei Liu <wei.liu2@citrix.com>
 L:	xen-devel@lists.xenproject.org (moderated for non-subscribers)
 L:	netdev@vger.kernel.org
diff --git a/Makefile b/Makefile
index 682840850f58..2d519d2fb3a9 100644
--- a/Makefile
+++ b/Makefile
@@ -1,7 +1,7 @@
 VERSION = 4
 PATCHLEVEL = 5
 SUBLEVEL = 0
-EXTRAVERSION = -rc3
+EXTRAVERSION = -rc7
 NAME = Blurry Fish Butt
 
 # *DOCUMENTATION*
diff --git a/arch/arc/Kconfig b/arch/arc/Kconfig
index 76dde9db7934..8a188bc1786a 100644
--- a/arch/arc/Kconfig
+++ b/arch/arc/Kconfig
@@ -12,8 +12,6 @@ config ARC
 	select BUILDTIME_EXTABLE_SORT
 	select COMMON_CLK
 	select CLONE_BACKWARDS
-	# ARC Busybox based initramfs absolutely relies on DEVTMPFS for /dev
-	select DEVTMPFS if !INITRAMFS_SOURCE=""
 	select GENERIC_ATOMIC64
 	select GENERIC_CLOCKEVENTS
 	select GENERIC_FIND_FIRST_BIT
@@ -275,14 +273,6 @@ config ARC_DCCM_BASE
 	default "0xA0000000"
 	depends on ARC_HAS_DCCM
 
-config ARC_HAS_HW_MPY
-	bool "Use Hardware Multiplier (Normal or Faster XMAC)"
-	default y
-	help
-	  Influences how gcc generates code for MPY operations.
-	  If enabled, MPYxx insns are generated, provided by Standard/XMAC
-	  Multipler. Otherwise software multipy lib is used
-
 choice
 	prompt "MMU Version"
 	default ARC_MMU_V3 if ARC_CPU_770
@@ -338,6 +328,19 @@ config ARC_PAGE_SIZE_4K
 
 endchoice
 
+choice
+	prompt "MMU Super Page Size"
+	depends on ISA_ARCV2 && TRANSPARENT_HUGEPAGE
+	default ARC_HUGEPAGE_2M
+
+config ARC_HUGEPAGE_2M
+	bool "2MB"
+
+config ARC_HUGEPAGE_16M
+	bool "16MB"
+
+endchoice
+
 if ISA_ARCOMPACT
 
 config ARC_COMPACT_IRQ_LEVELS
@@ -410,7 +413,7 @@ config ARC_HAS_RTC
 	default n
 	depends on !SMP
 
-config ARC_HAS_GRTC
+config ARC_HAS_GFRC
 	bool "SMP synchronized 64-bit cycle counter"
 	default y
 	depends on SMP
@@ -529,14 +532,6 @@ config ARC_DBG_TLB_MISS_COUNT
 	  Counts number of I and D TLB Misses and exports them via Debugfs
 	  The counters can be cleared via Debugfs as well
 
-if SMP
-
-config ARC_IPI_DBG
-	bool "Debug Inter Core interrupts"
-	default n
-
-endif
-
 endif
 
 config ARC_UBOOT_SUPPORT
@@ -566,6 +561,12 @@ endmenu
 endmenu	 # "ARC Architecture Configuration"
 
 source "mm/Kconfig"
+
+config FORCE_MAX_ZONEORDER
+	int "Maximum zone order"
+	default "12" if ARC_HUGEPAGE_16M
+	default "11"
+
 source "net/Kconfig"
 source "drivers/Kconfig"
 source "fs/Kconfig"
diff --git a/arch/arc/Makefile b/arch/arc/Makefile
index aeb19021099e..c8230f3395f2 100644
--- a/arch/arc/Makefile
+++ b/arch/arc/Makefile
@@ -74,10 +74,6 @@ ldflags-$(CONFIG_CPU_BIG_ENDIAN)	+= -EB
 # --build-id w/o "-marclinux". Default arc-elf32-ld is OK
 ldflags-$(upto_gcc44)			+= -marclinux
 
-ifndef CONFIG_ARC_HAS_HW_MPY
-	cflags-y	+= -mno-mpy
-endif
-
 LIBGCC	:= $(shell $(CC) $(cflags-y) --print-libgcc-file-name)
 
 # Modules with short calls might break for calls into builtin-kernel
diff --git a/arch/arc/configs/axs101_defconfig b/arch/arc/configs/axs101_defconfig
index f1ac9818b751..5d4e2a07ad3e 100644
--- a/arch/arc/configs/axs101_defconfig
+++ b/arch/arc/configs/axs101_defconfig
@@ -39,6 +39,7 @@ CONFIG_IP_PNP_RARP=y
 # CONFIG_INET_XFRM_MODE_TUNNEL is not set
 # CONFIG_INET_XFRM_MODE_BEET is not set
 # CONFIG_IPV6 is not set
+CONFIG_DEVTMPFS=y
 # CONFIG_STANDALONE is not set
 # CONFIG_PREVENT_FIRMWARE_BUILD is not set
 # CONFIG_FIRMWARE_IN_KERNEL is not set
@@ -73,7 +74,6 @@ CONFIG_I2C_CHARDEV=y
 CONFIG_I2C_DESIGNWARE_PLATFORM=y
 # CONFIG_HWMON is not set
 CONFIG_FB=y
-# CONFIG_VGA_CONSOLE is not set
 CONFIG_FRAMEBUFFER_CONSOLE=y
 CONFIG_FRAMEBUFFER_CONSOLE_DETECT_PRIMARY=y
 CONFIG_LOGO=y
@@ -91,12 +91,10 @@ CONFIG_MMC_SDHCI_PLTFM=y
 CONFIG_MMC_DW=y
 # CONFIG_IOMMU_SUPPORT is not set
 CONFIG_EXT3_FS=y
-CONFIG_EXT4_FS=y
 CONFIG_MSDOS_FS=y
 CONFIG_VFAT_FS=y
 CONFIG_NTFS_FS=y
 CONFIG_TMPFS=y
-CONFIG_JFFS2_FS=y
 CONFIG_NFS_FS=y
 CONFIG_NLS_CODEPAGE_437=y
 CONFIG_NLS_ISO8859_1=y
diff --git a/arch/arc/configs/axs103_defconfig b/arch/arc/configs/axs103_defconfig
index 323486d6ee83..87ee46b237ef 100644
--- a/arch/arc/configs/axs103_defconfig
+++ b/arch/arc/configs/axs103_defconfig
@@ -39,14 +39,10 @@ CONFIG_IP_PNP_RARP=y
 # CONFIG_INET_XFRM_MODE_TUNNEL is not set
 # CONFIG_INET_XFRM_MODE_BEET is not set
 # CONFIG_IPV6 is not set
+CONFIG_DEVTMPFS=y
 # CONFIG_STANDALONE is not set
 # CONFIG_PREVENT_FIRMWARE_BUILD is not set
 # CONFIG_FIRMWARE_IN_KERNEL is not set
-CONFIG_MTD=y
-CONFIG_MTD_CMDLINE_PARTS=y
-CONFIG_MTD_BLOCK=y
-CONFIG_MTD_NAND=y
-CONFIG_MTD_NAND_AXS=y
 CONFIG_SCSI=y
 CONFIG_BLK_DEV_SD=y
 CONFIG_NETDEVICES=y
@@ -78,14 +74,12 @@ CONFIG_I2C_CHARDEV=y
 CONFIG_I2C_DESIGNWARE_PLATFORM=y
 # CONFIG_HWMON is not set
 CONFIG_FB=y
-# CONFIG_VGA_CONSOLE is not set
 CONFIG_FRAMEBUFFER_CONSOLE=y
 CONFIG_FRAMEBUFFER_CONSOLE_DETECT_PRIMARY=y
 CONFIG_LOGO=y
 # CONFIG_LOGO_LINUX_MONO is not set
 # CONFIG_LOGO_LINUX_VGA16 is not set
 # CONFIG_LOGO_LINUX_CLUT224 is not set
-CONFIG_USB=y
 CONFIG_USB_EHCI_HCD=y
 CONFIG_USB_EHCI_HCD_PLATFORM=y
 CONFIG_USB_OHCI_HCD=y
@@ -97,12 +91,10 @@ CONFIG_MMC_SDHCI_PLTFM=y
 CONFIG_MMC_DW=y
 # CONFIG_IOMMU_SUPPORT is not set
 CONFIG_EXT3_FS=y
-CONFIG_EXT4_FS=y
 CONFIG_MSDOS_FS=y
 CONFIG_VFAT_FS=y
 CONFIG_NTFS_FS=y
 CONFIG_TMPFS=y
-CONFIG_JFFS2_FS=y
 CONFIG_NFS_FS=y
 CONFIG_NLS_CODEPAGE_437=y
 CONFIG_NLS_ISO8859_1=y
diff --git a/arch/arc/configs/axs103_smp_defconfig b/arch/arc/configs/axs103_smp_defconfig
index 66191cd0447e..d80daf4f7e73 100644
--- a/arch/arc/configs/axs103_smp_defconfig
+++ b/arch/arc/configs/axs103_smp_defconfig
@@ -40,14 +40,10 @@ CONFIG_IP_PNP_RARP=y
 # CONFIG_INET_XFRM_MODE_TUNNEL is not set
 # CONFIG_INET_XFRM_MODE_BEET is not set
 # CONFIG_IPV6 is not set
+CONFIG_DEVTMPFS=y
 # CONFIG_STANDALONE is not set
 # CONFIG_PREVENT_FIRMWARE_BUILD is not set
 # CONFIG_FIRMWARE_IN_KERNEL is not set
-CONFIG_MTD=y
-CONFIG_MTD_CMDLINE_PARTS=y
-CONFIG_MTD_BLOCK=y
-CONFIG_MTD_NAND=y
-CONFIG_MTD_NAND_AXS=y
 CONFIG_SCSI=y
 CONFIG_BLK_DEV_SD=y
 CONFIG_NETDEVICES=y
@@ -79,14 +75,12 @@ CONFIG_I2C_CHARDEV=y
 CONFIG_I2C_DESIGNWARE_PLATFORM=y
 # CONFIG_HWMON is not set
 CONFIG_FB=y
-# CONFIG_VGA_CONSOLE is not set
 CONFIG_FRAMEBUFFER_CONSOLE=y
 CONFIG_FRAMEBUFFER_CONSOLE_DETECT_PRIMARY=y
 CONFIG_LOGO=y
 # CONFIG_LOGO_LINUX_MONO is not set
 # CONFIG_LOGO_LINUX_VGA16 is not set
 # CONFIG_LOGO_LINUX_CLUT224 is not set
-CONFIG_USB=y
 CONFIG_USB_EHCI_HCD=y
 CONFIG_USB_EHCI_HCD_PLATFORM=y
 CONFIG_USB_OHCI_HCD=y
@@ -98,12 +92,10 @@ CONFIG_MMC_SDHCI_PLTFM=y
 CONFIG_MMC_DW=y
 # CONFIG_IOMMU_SUPPORT is not set
 CONFIG_EXT3_FS=y
-CONFIG_EXT4_FS=y
 CONFIG_MSDOS_FS=y
 CONFIG_VFAT_FS=y
 CONFIG_NTFS_FS=y
 CONFIG_TMPFS=y
-CONFIG_JFFS2_FS=y
 CONFIG_NFS_FS=y
 CONFIG_NLS_CODEPAGE_437=y
 CONFIG_NLS_ISO8859_1=y
diff --git a/arch/arc/configs/nsim_700_defconfig b/arch/arc/configs/nsim_700_defconfig
index 138f9d887957..f41095340b6a 100644
--- a/arch/arc/configs/nsim_700_defconfig
+++ b/arch/arc/configs/nsim_700_defconfig
@@ -4,6 +4,7 @@ CONFIG_DEFAULT_HOSTNAME="ARCLinux"
 # CONFIG_SWAP is not set
 CONFIG_SYSVIPC=y
 CONFIG_POSIX_MQUEUE=y
+# CONFIG_CROSS_MEMORY_ATTACH is not set
 CONFIG_HIGH_RES_TIMERS=y
 CONFIG_IKCONFIG=y
 CONFIG_IKCONFIG_PROC=y
@@ -26,7 +27,6 @@ CONFIG_ARC_PLAT_SIM=y
 CONFIG_ARC_BUILTIN_DTB_NAME="nsim_700"
 CONFIG_PREEMPT=y
 # CONFIG_COMPACTION is not set
-# CONFIG_CROSS_MEMORY_ATTACH is not set
 CONFIG_NET=y
 CONFIG_PACKET=y
 CONFIG_UNIX=y
@@ -34,6 +34,7 @@ CONFIG_UNIX_DIAG=y
 CONFIG_NET_KEY=y
 CONFIG_INET=y
 # CONFIG_IPV6 is not set
+CONFIG_DEVTMPFS=y
 # CONFIG_STANDALONE is not set
 # CONFIG_PREVENT_FIRMWARE_BUILD is not set
 # CONFIG_FIRMWARE_IN_KERNEL is not set
@@ -51,7 +52,6 @@ CONFIG_SERIAL_ARC=y
 CONFIG_SERIAL_ARC_CONSOLE=y
 # CONFIG_HW_RANDOM is not set
 # CONFIG_HWMON is not set
-# CONFIG_VGA_CONSOLE is not set
 # CONFIG_HID is not set
 # CONFIG_USB_SUPPORT is not set
 # CONFIG_IOMMU_SUPPORT is not set
@@ -63,4 +63,3 @@ CONFIG_NFS_FS=y
 # CONFIG_ENABLE_WARN_DEPRECATED is not set
 # CONFIG_ENABLE_MUST_CHECK is not set
 # CONFIG_DEBUG_PREEMPT is not set
-CONFIG_XZ_DEC=y
diff --git a/arch/arc/configs/nsim_hs_defconfig b/arch/arc/configs/nsim_hs_defconfig
index f68838e8068a..cfaa33cb5921 100644
--- a/arch/arc/configs/nsim_hs_defconfig
+++ b/arch/arc/configs/nsim_hs_defconfig
@@ -35,6 +35,7 @@ CONFIG_UNIX_DIAG=y
 CONFIG_NET_KEY=y
 CONFIG_INET=y
 # CONFIG_IPV6 is not set
+CONFIG_DEVTMPFS=y
 # CONFIG_STANDALONE is not set
 # CONFIG_PREVENT_FIRMWARE_BUILD is not set
 # CONFIG_FIRMWARE_IN_KERNEL is not set
@@ -49,7 +50,6 @@ CONFIG_SERIAL_ARC=y
 CONFIG_SERIAL_ARC_CONSOLE=y
 # CONFIG_HW_RANDOM is not set
 # CONFIG_HWMON is not set
-# CONFIG_VGA_CONSOLE is not set
 # CONFIG_HID is not set
 # CONFIG_USB_SUPPORT is not set
 # CONFIG_IOMMU_SUPPORT is not set
@@ -61,4 +61,3 @@ CONFIG_NFS_FS=y
 # CONFIG_ENABLE_WARN_DEPRECATED is not set
 # CONFIG_ENABLE_MUST_CHECK is not set
 # CONFIG_DEBUG_PREEMPT is not set
-CONFIG_XZ_DEC=y
diff --git a/arch/arc/configs/nsim_hs_smp_defconfig b/arch/arc/configs/nsim_hs_smp_defconfig
index 96bd1c20fb0b..bb2a8dc778b5 100644
--- a/arch/arc/configs/nsim_hs_smp_defconfig
+++ b/arch/arc/configs/nsim_hs_smp_defconfig
@@ -2,6 +2,7 @@ CONFIG_CROSS_COMPILE="arc-linux-"
 # CONFIG_LOCALVERSION_AUTO is not set
 CONFIG_DEFAULT_HOSTNAME="ARCLinux"
 # CONFIG_SWAP is not set
+# CONFIG_CROSS_MEMORY_ATTACH is not set
 CONFIG_HIGH_RES_TIMERS=y
 CONFIG_IKCONFIG=y
 CONFIG_IKCONFIG_PROC=y
@@ -21,13 +22,11 @@ CONFIG_MODULES=y
 # CONFIG_IOSCHED_DEADLINE is not set
 # CONFIG_IOSCHED_CFQ is not set
 CONFIG_ARC_PLAT_SIM=y
-CONFIG_ARC_BOARD_ML509=y
 CONFIG_ISA_ARCV2=y
 CONFIG_SMP=y
 CONFIG_ARC_BUILTIN_DTB_NAME="nsim_hs_idu"
 CONFIG_PREEMPT=y
 # CONFIG_COMPACTION is not set
-# CONFIG_CROSS_MEMORY_ATTACH is not set
 CONFIG_NET=y
 CONFIG_PACKET=y
 CONFIG_UNIX=y
@@ -35,6 +34,7 @@ CONFIG_UNIX_DIAG=y
 CONFIG_NET_KEY=y
 CONFIG_INET=y
 # CONFIG_IPV6 is not set
+CONFIG_DEVTMPFS=y
 # CONFIG_STANDALONE is not set
 # CONFIG_PREVENT_FIRMWARE_BUILD is not set
 # CONFIG_FIRMWARE_IN_KERNEL is not set
@@ -49,7 +49,6 @@ CONFIG_SERIAL_ARC=y
 CONFIG_SERIAL_ARC_CONSOLE=y
 # CONFIG_HW_RANDOM is not set
 # CONFIG_HWMON is not set
-# CONFIG_VGA_CONSOLE is not set
 # CONFIG_HID is not set
 # CONFIG_USB_SUPPORT is not set
 # CONFIG_IOMMU_SUPPORT is not set
@@ -60,4 +59,3 @@ CONFIG_TMPFS=y
 CONFIG_NFS_FS=y
 # CONFIG_ENABLE_WARN_DEPRECATED is not set
 # CONFIG_ENABLE_MUST_CHECK is not set
-CONFIG_XZ_DEC=y
diff --git a/arch/arc/configs/nsimosci_defconfig b/arch/arc/configs/nsimosci_defconfig
index 31e1d95764ff..646182e93753 100644
--- a/arch/arc/configs/nsimosci_defconfig
+++ b/arch/arc/configs/nsimosci_defconfig
@@ -33,6 +33,7 @@ CONFIG_UNIX_DIAG=y
 CONFIG_NET_KEY=y
 CONFIG_INET=y
 # CONFIG_IPV6 is not set
+CONFIG_DEVTMPFS=y
 # CONFIG_STANDALONE is not set
 # CONFIG_PREVENT_FIRMWARE_BUILD is not set
 # CONFIG_FIRMWARE_IN_KERNEL is not set
@@ -58,7 +59,6 @@ CONFIG_SERIAL_OF_PLATFORM=y
 # CONFIG_HW_RANDOM is not set
 # CONFIG_HWMON is not set
 CONFIG_FB=y
-# CONFIG_VGA_CONSOLE is not set
 CONFIG_FRAMEBUFFER_CONSOLE=y
 CONFIG_LOGO=y
 # CONFIG_HID is not set
diff --git a/arch/arc/configs/nsimosci_hs_defconfig b/arch/arc/configs/nsimosci_hs_defconfig
index fcae66683ca0..ceca2541950d 100644
--- a/arch/arc/configs/nsimosci_hs_defconfig
+++ b/arch/arc/configs/nsimosci_hs_defconfig
@@ -34,12 +34,12 @@ CONFIG_UNIX_DIAG=y
 CONFIG_NET_KEY=y
 CONFIG_INET=y
 # CONFIG_IPV6 is not set
+CONFIG_DEVTMPFS=y
 # CONFIG_STANDALONE is not set
 # CONFIG_PREVENT_FIRMWARE_BUILD is not set
 # CONFIG_FIRMWARE_IN_KERNEL is not set
 # CONFIG_BLK_DEV is not set
 CONFIG_NETDEVICES=y
-CONFIG_NET_OSCI_LAN=y
 CONFIG_INPUT_EVDEV=y
 # CONFIG_MOUSE_PS2_ALPS is not set
 # CONFIG_MOUSE_PS2_LOGIPS2PP is not set
@@ -58,7 +58,6 @@ CONFIG_SERIAL_OF_PLATFORM=y
 # CONFIG_HW_RANDOM is not set
 # CONFIG_HWMON is not set
 CONFIG_FB=y
-# CONFIG_VGA_CONSOLE is not set
 CONFIG_FRAMEBUFFER_CONSOLE=y
 CONFIG_LOGO=y
 # CONFIG_HID is not set
diff --git a/arch/arc/configs/nsimosci_hs_smp_defconfig b/arch/arc/configs/nsimosci_hs_smp_defconfig
index b01b659168ea..4b6da90f6f26 100644
--- a/arch/arc/configs/nsimosci_hs_smp_defconfig
+++ b/arch/arc/configs/nsimosci_hs_smp_defconfig
@@ -2,6 +2,7 @@ CONFIG_CROSS_COMPILE="arc-linux-"
 CONFIG_DEFAULT_HOSTNAME="ARCLinux"
 # CONFIG_SWAP is not set
 CONFIG_SYSVIPC=y
+# CONFIG_CROSS_MEMORY_ATTACH is not set
 CONFIG_NO_HZ=y
 CONFIG_HIGH_RES_TIMERS=y
 CONFIG_IKCONFIG=y
@@ -18,15 +19,11 @@ CONFIG_MODULES=y
 # CONFIG_IOSCHED_DEADLINE is not set
 # CONFIG_IOSCHED_CFQ is not set
 CONFIG_ARC_PLAT_SIM=y
-CONFIG_ARC_BOARD_ML509=y
 CONFIG_ISA_ARCV2=y
 CONFIG_SMP=y
-CONFIG_ARC_HAS_LL64=y
-# CONFIG_ARC_HAS_RTSC is not set
 CONFIG_ARC_BUILTIN_DTB_NAME="nsimosci_hs_idu"
 CONFIG_PREEMPT=y
 # CONFIG_COMPACTION is not set
-# CONFIG_CROSS_MEMORY_ATTACH is not set
 CONFIG_NET=y
 CONFIG_PACKET=y
 CONFIG_PACKET_DIAG=y
@@ -40,6 +37,7 @@ CONFIG_INET=y
 # CONFIG_INET_LRO is not set
 # CONFIG_IPV6 is not set
 # CONFIG_WIRELESS is not set
+CONFIG_DEVTMPFS=y
 # CONFIG_STANDALONE is not set
 # CONFIG_PREVENT_FIRMWARE_BUILD is not set
 # CONFIG_FIRMWARE_IN_KERNEL is not set
@@ -56,14 +54,11 @@ CONFIG_NETDEVICES=y
 # CONFIG_NET_VENDOR_STMICRO is not set
 # CONFIG_NET_VENDOR_VIA is not set
 # CONFIG_NET_VENDOR_WIZNET is not set
-CONFIG_NET_OSCI_LAN=y
 # CONFIG_WLAN is not set
 CONFIG_INPUT_EVDEV=y
 CONFIG_MOUSE_PS2_TOUCHKIT=y
 # CONFIG_SERIO_SERPORT is not set
-CONFIG_SERIO_LIBPS2=y
 CONFIG_SERIO_ARC_PS2=y
-CONFIG_VT_HW_CONSOLE_BINDING=y
 # CONFIG_LEGACY_PTYS is not set
 # CONFIG_DEVKMEM is not set
 CONFIG_SERIAL_8250=y
@@ -75,9 +70,6 @@ CONFIG_SERIAL_OF_PLATFORM=y
 # CONFIG_HW_RANDOM is not set
 # CONFIG_HWMON is not set
 CONFIG_FB=y
-CONFIG_ARCPGU_RGB888=y
-CONFIG_ARCPGU_DISPTYPE=0
-# CONFIG_VGA_CONSOLE is not set
 CONFIG_FRAMEBUFFER_CONSOLE=y
 CONFIG_LOGO=y
 # CONFIG_HID is not set
diff --git a/arch/arc/configs/tb10x_defconfig b/arch/arc/configs/tb10x_defconfig
index 3b4dc9cebcf1..9b342eaf95ae 100644
--- a/arch/arc/configs/tb10x_defconfig
+++ b/arch/arc/configs/tb10x_defconfig
@@ -3,6 +3,7 @@ CONFIG_CROSS_COMPILE="arc-linux-"
 CONFIG_DEFAULT_HOSTNAME="tb10x"
 CONFIG_SYSVIPC=y
 CONFIG_POSIX_MQUEUE=y
+# CONFIG_CROSS_MEMORY_ATTACH is not set
 CONFIG_HIGH_RES_TIMERS=y
 CONFIG_BSD_PROCESS_ACCT=y
 CONFIG_BSD_PROCESS_ACCT_V3=y
@@ -26,12 +27,10 @@ CONFIG_MODULE_UNLOAD=y
 # CONFIG_BLOCK is not set
 CONFIG_ARC_PLAT_TB10X=y
 CONFIG_ARC_CACHE_LINE_SHIFT=5
-CONFIG_ARC_STACK_NONEXEC=y
 CONFIG_HZ=250
 CONFIG_ARC_BUILTIN_DTB_NAME="abilis_tb100_dvk"
 CONFIG_PREEMPT_VOLUNTARY=y
 # CONFIG_COMPACTION is not set
-# CONFIG_CROSS_MEMORY_ATTACH is not set
 CONFIG_NET=y
 CONFIG_PACKET=y
 CONFIG_UNIX=y
@@ -44,8 +43,8 @@ CONFIG_IP_MULTICAST=y
 # CONFIG_INET_DIAG is not set
 # CONFIG_IPV6 is not set
 # CONFIG_WIRELESS is not set
+CONFIG_DEVTMPFS=y
 # CONFIG_FIRMWARE_IN_KERNEL is not set
-CONFIG_PROC_DEVICETREE=y
 CONFIG_NETDEVICES=y
 # CONFIG_NET_CADENCE is not set
 # CONFIG_NET_VENDOR_BROADCOM is not set
@@ -55,9 +54,6 @@ CONFIG_NETDEVICES=y
 # CONFIG_NET_VENDOR_NATSEMI is not set
 # CONFIG_NET_VENDOR_SEEQ is not set
 CONFIG_STMMAC_ETH=y
-CONFIG_STMMAC_DEBUG_FS=y
-CONFIG_STMMAC_DA=y
-CONFIG_STMMAC_CHAINED=y
 # CONFIG_NET_VENDOR_WIZNET is not set
 # CONFIG_WLAN is not set
 # CONFIG_INPUT is not set
@@ -91,7 +87,6 @@ CONFIG_LEDS_TRIGGER_DEFAULT_ON=y
 CONFIG_LEDS_TRIGGER_TRANSIENT=y
 CONFIG_DMADEVICES=y
 CONFIG_DW_DMAC=y
-CONFIG_NET_DMA=y
 CONFIG_ASYNC_TX_DMA=y
 # CONFIG_IOMMU_SUPPORT is not set
 # CONFIG_DNOTIFY is not set
@@ -100,17 +95,16 @@ CONFIG_TMPFS=y
 CONFIG_CONFIGFS_FS=y
 # CONFIG_MISC_FILESYSTEMS is not set
 # CONFIG_NETWORK_FILESYSTEMS is not set
+CONFIG_DEBUG_INFO=y
 # CONFIG_ENABLE_WARN_DEPRECATED is not set
-CONFIG_MAGIC_SYSRQ=y
 CONFIG_STRIP_ASM_SYMS=y
 CONFIG_DEBUG_FS=y
 CONFIG_HEADERS_CHECK=y
 CONFIG_DEBUG_SECTION_MISMATCH=y
+CONFIG_MAGIC_SYSRQ=y
+CONFIG_DEBUG_MEMORY_INIT=y
+CONFIG_DEBUG_STACKOVERFLOW=y
 CONFIG_DETECT_HUNG_TASK=y
 CONFIG_SCHEDSTATS=y
 CONFIG_TIMER_STATS=y
-CONFIG_DEBUG_INFO=y
-CONFIG_DEBUG_MEMORY_INIT=y
-CONFIG_DEBUG_STACKOVERFLOW=y
-# CONFIG_CRYPTO_ANSI_CPRNG is not set
 # CONFIG_CRYPTO_HW is not set
diff --git a/arch/arc/configs/vdk_hs38_smp_defconfig b/arch/arc/configs/vdk_hs38_smp_defconfig
index f36c047b33ca..735985974a31 100644
--- a/arch/arc/configs/vdk_hs38_smp_defconfig
+++ b/arch/arc/configs/vdk_hs38_smp_defconfig
@@ -16,7 +16,7 @@ CONFIG_ARC_PLAT_AXS10X=y
 CONFIG_AXS103=y
 CONFIG_ISA_ARCV2=y
 CONFIG_SMP=y
-# CONFIG_ARC_HAS_GRTC is not set
+# CONFIG_ARC_HAS_GFRC is not set
 CONFIG_ARC_UBOOT_SUPPORT=y
 CONFIG_ARC_BUILTIN_DTB_NAME="vdk_hs38_smp"
 CONFIG_PREEMPT=y
diff --git a/arch/arc/include/asm/arcregs.h b/arch/arc/include/asm/arcregs.h
index 7fac7d85ed6a..f9f4c6f59fdb 100644
--- a/arch/arc/include/asm/arcregs.h
+++ b/arch/arc/include/asm/arcregs.h
@@ -10,7 +10,8 @@
 #define _ASM_ARC_ARCREGS_H
 
 /* Build Configuration Registers */
-#define ARC_REG_DCCMBASE_BCR	0x61	/* DCCM Base Addr */
+#define ARC_REG_AUX_DCCM	0x18	/* DCCM Base Addr ARCv2 */
+#define ARC_REG_DCCM_BASE_BUILD	0x61	/* DCCM Base Addr ARCompact */
 #define ARC_REG_CRC_BCR		0x62
 #define ARC_REG_VECBASE_BCR	0x68
 #define ARC_REG_PERIBASE_BCR	0x69
@@ -18,10 +19,10 @@
 #define ARC_REG_DPFP_BCR	0x6C	/* ARCompact: Dbl Precision FPU */
 #define ARC_REG_FP_V2_BCR	0xc8	/* ARCv2 FPU */
 #define ARC_REG_SLC_BCR		0xce
-#define ARC_REG_DCCM_BCR	0x74	/* DCCM Present + SZ */
+#define ARC_REG_DCCM_BUILD	0x74	/* DCCM size (common) */
 #define ARC_REG_TIMERS_BCR	0x75
 #define ARC_REG_AP_BCR		0x76
-#define ARC_REG_ICCM_BCR	0x78
+#define ARC_REG_ICCM_BUILD	0x78	/* ICCM size (common) */
 #define ARC_REG_XY_MEM_BCR	0x79
 #define ARC_REG_MAC_BCR		0x7a
 #define ARC_REG_MUL_BCR		0x7b
@@ -36,6 +37,7 @@
 #define ARC_REG_IRQ_BCR		0xF3
 #define ARC_REG_SMART_BCR	0xFF
 #define ARC_REG_CLUSTER_BCR	0xcf
+#define ARC_REG_AUX_ICCM	0x208	/* ICCM Base Addr (ARCv2) */
 
 /* status32 Bits Positions */
 #define STATUS_AE_BIT		5	/* Exception active */
@@ -246,7 +248,7 @@ struct bcr_perip {
 #endif
 };
 
-struct bcr_iccm {
+struct bcr_iccm_arcompact {
 #ifdef CONFIG_CPU_BIG_ENDIAN
 	unsigned int base:16, pad:5, sz:3, ver:8;
 #else
@@ -254,17 +256,15 @@ struct bcr_iccm {
 #endif
 };
 
-/* DCCM Base Address Register: ARC_REG_DCCMBASE_BCR */
-struct bcr_dccm_base {
+struct bcr_iccm_arcv2 {
 #ifdef CONFIG_CPU_BIG_ENDIAN
-	unsigned int addr:24, ver:8;
+	unsigned int pad:8, sz11:4, sz01:4, sz10:4, sz00:4, ver:8;
 #else
-	unsigned int ver:8, addr:24;
+	unsigned int ver:8, sz00:4, sz10:4, sz01:4, sz11:4, pad:8;
 #endif
 };
 
-/* DCCM RAM Configuration Register: ARC_REG_DCCM_BCR */
-struct bcr_dccm {
+struct bcr_dccm_arcompact {
 #ifdef CONFIG_CPU_BIG_ENDIAN
 	unsigned int res:21, sz:3, ver:8;
 #else
@@ -272,6 +272,14 @@ struct bcr_dccm {
 #endif
 };
 
+struct bcr_dccm_arcv2 {
+#ifdef CONFIG_CPU_BIG_ENDIAN
+	unsigned int pad2:12, cyc:3, pad1:1, sz1:4, sz0:4, ver:8;
+#else
+	unsigned int ver:8, sz0:4, sz1:4, pad1:1, cyc:3, pad2:12;
+#endif
+};
+
 /* ARCompact: Both SP and DP FPU BCRs have same format */
 struct bcr_fp_arcompact {
 #ifdef CONFIG_CPU_BIG_ENDIAN
@@ -315,9 +323,9 @@ struct bcr_bpu_arcv2 {
 
 struct bcr_generic {
 #ifdef CONFIG_CPU_BIG_ENDIAN
-	unsigned int pad:24, ver:8;
+	unsigned int info:24, ver:8;
 #else
-	unsigned int ver:8, pad:24;
+	unsigned int ver:8, info:24;
 #endif
 };
 
@@ -349,14 +357,13 @@ struct cpuinfo_arc {
 	struct cpuinfo_arc_bpu bpu;
 	struct bcr_identity core;
 	struct bcr_isa isa;
-	struct bcr_timer timers;
 	unsigned int vec_base;
 	struct cpuinfo_arc_ccm iccm, dccm;
 	struct {
 		unsigned int swap:1, norm:1, minmax:1, barrel:1, crc:1, pad1:3,
 			     fpu_sp:1, fpu_dp:1, pad2:6,
 			     debug:1, ap:1, smart:1, rtt:1, pad3:4,
-			     pad4:8;
+			     timer0:1, timer1:1, rtc:1, gfrc:1, pad4:4;
 	} extn;
 	struct bcr_mpy extn_mpy;
 	struct bcr_extn_xymem extn_xymem;
diff --git a/arch/arc/include/asm/irq.h b/arch/arc/include/asm/irq.h
index 4fd7d62a6e30..49014f0ef36d 100644
--- a/arch/arc/include/asm/irq.h
+++ b/arch/arc/include/asm/irq.h
@@ -16,11 +16,9 @@
 #ifdef CONFIG_ISA_ARCOMPACT
 #define TIMER0_IRQ      3
 #define TIMER1_IRQ      4
-#define IPI_IRQ		(NR_CPU_IRQS-1) /* dummy to enable SMP build for up hardware */
 #else
 #define TIMER0_IRQ      16
 #define TIMER1_IRQ      17
-#define IPI_IRQ         19
 #endif
 
 #include <linux/interrupt.h>
diff --git a/arch/arc/include/asm/irqflags-arcv2.h b/arch/arc/include/asm/irqflags-arcv2.h
index 258b0e5ad332..37c2f751eebf 100644
--- a/arch/arc/include/asm/irqflags-arcv2.h
+++ b/arch/arc/include/asm/irqflags-arcv2.h
@@ -22,6 +22,7 @@
 #define AUX_IRQ_CTRL		0x00E
 #define AUX_IRQ_ACT		0x043	/* Active Intr across all levels */
 #define AUX_IRQ_LVL_PEND	0x200	/* Pending Intr across all levels */
+#define AUX_IRQ_HINT		0x201	/* For generating Soft Interrupts */
 #define AUX_IRQ_PRIORITY	0x206
 #define ICAUSE			0x40a
 #define AUX_IRQ_SELECT		0x40b
@@ -30,8 +31,11 @@
 /* Was Intr taken in User Mode */
 #define AUX_IRQ_ACT_BIT_U	31
 
-/* 0 is highest level, but taken by FIRQs, if present in design */
-#define ARCV2_IRQ_DEF_PRIO		0
+/*
+ * User space should be interruptable even by lowest prio interrupt
+ * Safe even if actual interrupt priorities is fewer or even one
+ */
+#define ARCV2_IRQ_DEF_PRIO	15
 
 /* seed value for status register */
 #define ISA_INIT_STATUS_BITS	(STATUS_IE_MASK | STATUS_AD_MASK | \
@@ -112,6 +116,16 @@ static inline int arch_irqs_disabled(void)
 	return arch_irqs_disabled_flags(arch_local_save_flags());
 }
 
+static inline void arc_softirq_trigger(int irq)
+{
+	write_aux_reg(AUX_IRQ_HINT, irq);
+}
+
+static inline void arc_softirq_clear(int irq)
+{
+	write_aux_reg(AUX_IRQ_HINT, 0);
+}
+
 #else
 
 .macro IRQ_DISABLE  scratch
diff --git a/arch/arc/include/asm/mcip.h b/arch/arc/include/asm/mcip.h
index 46f4e5351b2a..847e3bbe387f 100644
--- a/arch/arc/include/asm/mcip.h
+++ b/arch/arc/include/asm/mcip.h
@@ -39,8 +39,8 @@ struct mcip_cmd {
 #define CMD_DEBUG_SET_MASK		0x34
 #define CMD_DEBUG_SET_SELECT		0x36
 
-#define CMD_GRTC_READ_LO		0x42
-#define CMD_GRTC_READ_HI		0x43
+#define CMD_GFRC_READ_LO		0x42
+#define CMD_GFRC_READ_HI		0x43
 
 #define CMD_IDU_ENABLE			0x71
 #define CMD_IDU_DISABLE			0x72
diff --git a/arch/arc/include/asm/pgtable.h b/arch/arc/include/asm/pgtable.h
index 57af2f05ae84..d426d4215513 100644
--- a/arch/arc/include/asm/pgtable.h
+++ b/arch/arc/include/asm/pgtable.h
@@ -179,37 +179,44 @@
 #define __S111  PAGE_U_X_W_R
 
 /****************************************************************
- * Page Table Lookup split
+ * 2 tier (PGD:PTE) software page walker
  *
- * We implement 2 tier paging and since this is all software, we are free
- * to customize the span of a PGD / PTE entry to suit us
- *
- *			32 bit virtual address
+ * [31]		    32 bit virtual address              [0]
  * -------------------------------------------------------
- * | BITS_FOR_PGD    |  BITS_FOR_PTE    |  BITS_IN_PAGE  |
+ * |               | <------------ PGDIR_SHIFT ----------> |
+ * |		   |					 |
+ * | BITS_FOR_PGD  |  BITS_FOR_PTE  | <-- PAGE_SHIFT --> |
  * -------------------------------------------------------
  *       |                  |                |
  *       |                  |                --> off in page frame
- *       |		    |
  *       |                  ---> index into Page Table
- *       |
  *       ----> index into Page Directory
+ *
+ * In a single page size configuration, only PAGE_SHIFT is fixed
+ * So both PGD and PTE sizing can be tweaked
+ *  e.g. 8K page (PAGE_SHIFT 13) can have
+ *  - PGDIR_SHIFT 21  -> 11:8:13 address split
+ *  - PGDIR_SHIFT 24  -> 8:11:13 address split
+ *
+ * If Super Page is configured, PGDIR_SHIFT becomes fixed too,
+ * so the sizing flexibility is gone.
  */
 
-#define BITS_IN_PAGE	PAGE_SHIFT
-
-/* Optimal Sizing of Pg Tbl - based on MMU page size */
-#if defined(CONFIG_ARC_PAGE_SIZE_8K)
-#define BITS_FOR_PTE	8		/* 11:8:13 */
-#elif defined(CONFIG_ARC_PAGE_SIZE_16K)
-#define BITS_FOR_PTE	8		/* 10:8:14 */
-#elif defined(CONFIG_ARC_PAGE_SIZE_4K)
-#define BITS_FOR_PTE	9		/* 11:9:12 */
+#if defined(CONFIG_ARC_HUGEPAGE_16M)
+#define PGDIR_SHIFT	24
+#elif defined(CONFIG_ARC_HUGEPAGE_2M)
+#define PGDIR_SHIFT	21
+#else
+/*
+ * Only Normal page support so "hackable" (see comment above)
+ * Default value provides 11:8:13 (8K), 11:9:12 (4K)
+ */
+#define PGDIR_SHIFT	21
 #endif
 
-#define BITS_FOR_PGD	(32 - BITS_FOR_PTE - BITS_IN_PAGE)
+#define BITS_FOR_PTE	(PGDIR_SHIFT - PAGE_SHIFT)
+#define BITS_FOR_PGD	(32 - PGDIR_SHIFT)
 
-#define PGDIR_SHIFT	(32 - BITS_FOR_PGD)
 #define PGDIR_SIZE	(1UL << PGDIR_SHIFT)	/* vaddr span, not PDG sz */
 #define PGDIR_MASK	(~(PGDIR_SIZE-1))
 
diff --git a/arch/arc/kernel/entry-arcv2.S b/arch/arc/kernel/entry-arcv2.S
index cbfec79137bf..c1264607bbff 100644
--- a/arch/arc/kernel/entry-arcv2.S
+++ b/arch/arc/kernel/entry-arcv2.S
@@ -45,11 +45,12 @@ VECTOR	reserved		; Reserved slots
 VECTOR	handle_interrupt	; (16) Timer0
 VECTOR	handle_interrupt	; unused (Timer1)
 VECTOR	handle_interrupt	; unused (WDT)
-VECTOR	handle_interrupt	; (19) ICI (inter core interrupt)
-VECTOR	handle_interrupt
-VECTOR	handle_interrupt
-VECTOR	handle_interrupt
-VECTOR	handle_interrupt	; (23) End of fixed IRQs
+VECTOR	handle_interrupt	; (19) Inter core Interrupt (IPI)
+VECTOR	handle_interrupt	; (20) perf Interrupt
+VECTOR	handle_interrupt	; (21) Software Triggered Intr (Self IPI)
+VECTOR	handle_interrupt	; unused
+VECTOR	handle_interrupt	; (23) unused
+# End of fixed IRQs
 
 .rept CONFIG_ARC_NUMBER_OF_INTERRUPTS - 8
 	VECTOR	handle_interrupt
@@ -211,7 +212,11 @@ debug_marker_syscall:
 ; (since IRQ NOT allowed in DS in ARCv2, this can only happen if orig
 ; entry was via Exception in DS which got preempted in kernel).
 ;
-; IRQ RTIE won't reliably restore DE bit and/or BTA, needs handling
+; IRQ RTIE won't reliably restore DE bit and/or BTA, needs workaround
+;
+; Solution is return from Intr w/o any delay slot quirks into a kernel trampoline
+; and from pure kernel mode return to delay slot which handles DS bit/BTA correctly
+
 .Lintr_ret_to_delay_slot:
 debug_marker_ds:
 
@@ -222,18 +227,23 @@ debug_marker_ds:
 	ld	r2, [sp, PT_ret]
 	ld	r3, [sp, PT_status32]
 
+	; STAT32 for Int return created from scratch
+	; (No delay dlot, disable Further intr in trampoline)
+
 	bic  	r0, r3, STATUS_U_MASK|STATUS_DE_MASK|STATUS_IE_MASK|STATUS_L_MASK
 	st	r0, [sp, PT_status32]
 
 	mov	r1, .Lintr_ret_to_delay_slot_2
 	st	r1, [sp, PT_ret]
 
+	; Orig exception PC/STAT32 safekept @orig_r0 and @event stack slots
 	st	r2, [sp, 0]
 	st	r3, [sp, 4]
 
 	b	.Lisr_ret_fast_path
 
 .Lintr_ret_to_delay_slot_2:
+	; Trampoline to restore orig exception PC/STAT32/BTA/AUX_USER_SP
 	sub	sp, sp, SZ_PT_REGS
 	st	r9, [sp, -4]
 
@@ -243,11 +253,19 @@ debug_marker_ds:
 	ld	r9, [sp, 4]
 	sr	r9, [erstatus]
 
+	; restore AUX_USER_SP if returning to U mode
+	bbit0	r9, STATUS_U_BIT, 1f
+	ld	r9, [sp, PT_sp]
+	sr	r9, [AUX_USER_SP]
+
+1:
 	ld	r9, [sp, 8]
 	sr	r9, [erbta]
 
 	ld	r9, [sp, -4]
 	add	sp, sp, SZ_PT_REGS
+
+	; return from pure kernel mode to delay slot
 	rtie
 
 END(ret_from_exception)
diff --git a/arch/arc/kernel/intc-arcv2.c b/arch/arc/kernel/intc-arcv2.c
index 0394f9f61b46..942526322ae7 100644
--- a/arch/arc/kernel/intc-arcv2.c
+++ b/arch/arc/kernel/intc-arcv2.c
@@ -14,6 +14,8 @@
 #include <linux/irqchip.h>
 #include <asm/irq.h>
 
+static int irq_prio;
+
 /*
  * Early Hardware specific Interrupt setup
  * -Called very early (start_kernel -> setup_arch -> setup_processor)
@@ -24,6 +26,14 @@ void arc_init_IRQ(void)
 {
 	unsigned int tmp;
 
+	struct irq_build {
+#ifdef CONFIG_CPU_BIG_ENDIAN
+		unsigned int pad:3, firq:1, prio:4, exts:8, irqs:8, ver:8;
+#else
+		unsigned int ver:8, irqs:8, exts:8, prio:4, firq:1, pad:3;
+#endif
+	} irq_bcr;
+
 	struct aux_irq_ctrl {
 #ifdef CONFIG_CPU_BIG_ENDIAN
 		unsigned int res3:18, save_idx_regs:1, res2:1,
@@ -46,28 +56,25 @@ void arc_init_IRQ(void)
 
 	WRITE_AUX(AUX_IRQ_CTRL, ictrl);
 
-	/* setup status32, don't enable intr yet as kernel doesn't want */
-	tmp = read_aux_reg(0xa);
-	tmp |= ISA_INIT_STATUS_BITS;
-	tmp &= ~STATUS_IE_MASK;
-	asm volatile("flag %0	\n"::"r"(tmp));
-
 	/*
 	 * ARCv2 core intc provides multiple interrupt priorities (upto 16).
 	 * Typical builds though have only two levels (0-high, 1-low)
 	 * Linux by default uses lower prio 1 for most irqs, reserving 0 for
 	 * NMI style interrupts in future (say perf)
-	 *
-	 * Read the intc BCR to confirm that Linux default priority is avail
-	 * in h/w
-	 *
-	 * Note:
-	 *  IRQ_BCR[27..24] contains N-1 (for N priority levels) and prio level
-	 *  is 0 based.
 	 */
-	tmp = (read_aux_reg(ARC_REG_IRQ_BCR) >> 24 ) & 0xF;
-	if (ARCV2_IRQ_DEF_PRIO > tmp)
-		panic("Linux default irq prio incorrect\n");
+
+	READ_BCR(ARC_REG_IRQ_BCR, irq_bcr);
+
+	irq_prio = irq_bcr.prio;	/* Encoded as N-1 for N levels */
+	pr_info("archs-intc\t: %d priority levels (default %d)%s\n",
+		irq_prio + 1, irq_prio,
+		irq_bcr.firq ? " FIRQ (not used)":"");
+
+	/* setup status32, don't enable intr yet as kernel doesn't want */
+	tmp = read_aux_reg(0xa);
+	tmp |= STATUS_AD_MASK | (irq_prio << 1);
+	tmp &= ~STATUS_IE_MASK;
+	asm volatile("flag %0	\n"::"r"(tmp));
 }
 
 static void arcv2_irq_mask(struct irq_data *data)
@@ -86,7 +93,7 @@ void arcv2_irq_enable(struct irq_data *data)
 {
 	/* set default priority */
 	write_aux_reg(AUX_IRQ_SELECT, data->irq);
-	write_aux_reg(AUX_IRQ_PRIORITY, ARCV2_IRQ_DEF_PRIO);
+	write_aux_reg(AUX_IRQ_PRIORITY, irq_prio);
 
 	/*
 	 * hw auto enables (linux unmask) all by default
diff --git a/arch/arc/kernel/intc-compact.c b/arch/arc/kernel/intc-compact.c
index 06bcedf19b62..224d1c3aa9c4 100644
--- a/arch/arc/kernel/intc-compact.c
+++ b/arch/arc/kernel/intc-compact.c
@@ -81,9 +81,6 @@ static int arc_intc_domain_map(struct irq_domain *d, unsigned int irq,
 {
 	switch (irq) {
 	case TIMER0_IRQ:
-#ifdef CONFIG_SMP
-	case IPI_IRQ:
-#endif
 		irq_set_chip_and_handler(irq, &onchip_intc, handle_percpu_irq);
 		break;
 	default:
diff --git a/arch/arc/kernel/mcip.c b/arch/arc/kernel/mcip.c
index bd237acdf4f2..c41c364b926c 100644
--- a/arch/arc/kernel/mcip.c
+++ b/arch/arc/kernel/mcip.c
@@ -11,9 +11,13 @@
 #include <linux/smp.h>
 #include <linux/irq.h>
 #include <linux/spinlock.h>
+#include <asm/irqflags-arcv2.h>
 #include <asm/mcip.h>
 #include <asm/setup.h>
 
+#define IPI_IRQ		19
+#define SOFTIRQ_IRQ	21
+
 static char smp_cpuinfo_buf[128];
 static int idu_detected;
 
@@ -22,6 +26,7 @@ static DEFINE_RAW_SPINLOCK(mcip_lock);
 static void mcip_setup_per_cpu(int cpu)
 {
 	smp_ipi_irq_setup(cpu, IPI_IRQ);
+	smp_ipi_irq_setup(cpu, SOFTIRQ_IRQ);
 }
 
 static void mcip_ipi_send(int cpu)
@@ -29,46 +34,44 @@ static void mcip_ipi_send(int cpu)
 	unsigned long flags;
 	int ipi_was_pending;
 
+	/* ARConnect can only send IPI to others */
+	if (unlikely(cpu == raw_smp_processor_id())) {
+		arc_softirq_trigger(SOFTIRQ_IRQ);
+		return;
+	}
+
+	raw_spin_lock_irqsave(&mcip_lock, flags);
+
 	/*
-	 * NOTE: We must spin here if the other cpu hasn't yet
-	 * serviced a previous message. This can burn lots
-	 * of time, but we MUST follows this protocol or
-	 * ipi messages can be lost!!!
-	 * Also, we must release the lock in this loop because
-	 * the other side may get to this same loop and not
-	 * be able to ack -- thus causing deadlock.
+	 * If receiver already has a pending interrupt, elide sending this one.
+	 * Linux cross core calling works well with concurrent IPIs
+	 * coalesced into one
+	 * see arch/arc/kernel/smp.c: ipi_send_msg_one()
 	 */
+	__mcip_cmd(CMD_INTRPT_READ_STATUS, cpu);
+	ipi_was_pending = read_aux_reg(ARC_REG_MCIP_READBACK);
+	if (!ipi_was_pending)
+		__mcip_cmd(CMD_INTRPT_GENERATE_IRQ, cpu);
 
-	do {
-		raw_spin_lock_irqsave(&mcip_lock, flags);
-		__mcip_cmd(CMD_INTRPT_READ_STATUS, cpu);
-		ipi_was_pending = read_aux_reg(ARC_REG_MCIP_READBACK);
-		if (ipi_was_pending == 0)
-			break; /* break out but keep lock */
-		raw_spin_unlock_irqrestore(&mcip_lock, flags);
-	} while (1);
-
-	__mcip_cmd(CMD_INTRPT_GENERATE_IRQ, cpu);
 	raw_spin_unlock_irqrestore(&mcip_lock, flags);
-
-#ifdef CONFIG_ARC_IPI_DBG
-	if (ipi_was_pending)
-		pr_info("IPI ACK delayed from cpu %d\n", cpu);
-#endif
 }
 
 static void mcip_ipi_clear(int irq)
 {
 	unsigned int cpu, c;
 	unsigned long flags;
-	unsigned int __maybe_unused copy;
+
+	if (unlikely(irq == SOFTIRQ_IRQ)) {
+		arc_softirq_clear(irq);
+		return;
+	}
 
 	raw_spin_lock_irqsave(&mcip_lock, flags);
 
 	/* Who sent the IPI */
 	__mcip_cmd(CMD_INTRPT_CHECK_SOURCE, 0);
 
-	copy = cpu = read_aux_reg(ARC_REG_MCIP_READBACK);	/* 1,2,4,8... */
+	cpu = read_aux_reg(ARC_REG_MCIP_READBACK);	/* 1,2,4,8... */
 
 	/*
 	 * In rare case, multiple concurrent IPIs sent to same target can
@@ -82,12 +85,6 @@ static void mcip_ipi_clear(int irq)
 	} while (cpu);
 
 	raw_spin_unlock_irqrestore(&mcip_lock, flags);
-
-#ifdef CONFIG_ARC_IPI_DBG
-	if (c != __ffs(copy))
-		pr_info("IPIs from %x coalesced to %x\n",
-			copy, raw_smp_processor_id());
-#endif
 }
 
 static void mcip_probe_n_setup(void)
@@ -96,13 +93,13 @@ static void mcip_probe_n_setup(void)
 #ifdef CONFIG_CPU_BIG_ENDIAN
 		unsigned int pad3:8,
 			     idu:1, llm:1, num_cores:6,
-			     iocoh:1,  grtc:1, dbg:1, pad2:1,
+			     iocoh:1,  gfrc:1, dbg:1, pad2:1,
 			     msg:1, sem:1, ipi:1, pad:1,
 			     ver:8;
 #else
 		unsigned int ver:8,
 			     pad:1, ipi:1, sem:1, msg:1,
-			     pad2:1, dbg:1, grtc:1, iocoh:1,
+			     pad2:1, dbg:1, gfrc:1, iocoh:1,
 			     num_cores:6, llm:1, idu:1,
 			     pad3:8;
 #endif
@@ -111,12 +108,13 @@ static void mcip_probe_n_setup(void)
 	READ_BCR(ARC_REG_MCIP_BCR, mp);
 
 	sprintf(smp_cpuinfo_buf,
-		"Extn [SMP]\t: ARConnect (v%d): %d cores with %s%s%s%s\n",
+		"Extn [SMP]\t: ARConnect (v%d): %d cores with %s%s%s%s%s\n",
 		mp.ver, mp.num_cores,
 		IS_AVAIL1(mp.ipi, "IPI "),
 		IS_AVAIL1(mp.idu, "IDU "),
+		IS_AVAIL1(mp.llm, "LLM "),
 		IS_AVAIL1(mp.dbg, "DEBUG "),
-		IS_AVAIL1(mp.grtc, "GRTC"));
+		IS_AVAIL1(mp.gfrc, "GFRC"));
 
 	idu_detected = mp.idu;
 
@@ -125,8 +123,8 @@ static void mcip_probe_n_setup(void)
 		__mcip_cmd_data(CMD_DEBUG_SET_MASK, 0xf, 0xf);
 	}
 
-	if (IS_ENABLED(CONFIG_ARC_HAS_GRTC) && !mp.grtc)
-		panic("kernel trying to use non-existent GRTC\n");
+	if (IS_ENABLED(CONFIG_ARC_HAS_GFRC) && !mp.gfrc)
+		panic("kernel trying to use non-existent GFRC\n");
 }
 
 struct plat_smp_ops plat_smp_ops = {
diff --git a/arch/arc/kernel/setup.c b/arch/arc/kernel/setup.c
index e1b87444ea9a..cdc821df1809 100644
--- a/arch/arc/kernel/setup.c
+++ b/arch/arc/kernel/setup.c
@@ -42,9 +42,57 @@ struct task_struct *_current_task[NR_CPUS];	/* For stack switching */
 
 struct cpuinfo_arc cpuinfo_arc700[NR_CPUS];
 
+static void read_decode_ccm_bcr(struct cpuinfo_arc *cpu)
+{
+	if (is_isa_arcompact()) {
+		struct bcr_iccm_arcompact iccm;
+		struct bcr_dccm_arcompact dccm;
+
+		READ_BCR(ARC_REG_ICCM_BUILD, iccm);
+		if (iccm.ver) {
+			cpu->iccm.sz = 4096 << iccm.sz;	/* 8K to 512K */
+			cpu->iccm.base_addr = iccm.base << 16;
+		}
+
+		READ_BCR(ARC_REG_DCCM_BUILD, dccm);
+		if (dccm.ver) {
+			unsigned long base;
+			cpu->dccm.sz = 2048 << dccm.sz;	/* 2K to 256K */
+
+			base = read_aux_reg(ARC_REG_DCCM_BASE_BUILD);
+			cpu->dccm.base_addr = base & ~0xF;
+		}
+	} else {
+		struct bcr_iccm_arcv2 iccm;
+		struct bcr_dccm_arcv2 dccm;
+		unsigned long region;
+
+		READ_BCR(ARC_REG_ICCM_BUILD, iccm);
+		if (iccm.ver) {
+			cpu->iccm.sz = 256 << iccm.sz00;	/* 512B to 16M */
+			if (iccm.sz00 == 0xF && iccm.sz01 > 0)
+				cpu->iccm.sz <<= iccm.sz01;
+
+			region = read_aux_reg(ARC_REG_AUX_ICCM);
+			cpu->iccm.base_addr = region & 0xF0000000;
+		}
+
+		READ_BCR(ARC_REG_DCCM_BUILD, dccm);
+		if (dccm.ver) {
+			cpu->dccm.sz = 256 << dccm.sz0;
+			if (dccm.sz0 == 0xF && dccm.sz1 > 0)
+				cpu->dccm.sz <<= dccm.sz1;
+
+			region = read_aux_reg(ARC_REG_AUX_DCCM);
+			cpu->dccm.base_addr = region & 0xF0000000;
+		}
+	}
+}
+
 static void read_arc_build_cfg_regs(void)
 {
 	struct bcr_perip uncached_space;
+	struct bcr_timer timer;
 	struct bcr_generic bcr;
 	struct cpuinfo_arc *cpu = &cpuinfo_arc700[smp_processor_id()];
 	unsigned long perip_space;
@@ -53,7 +101,11 @@ static void read_arc_build_cfg_regs(void)
 	READ_BCR(AUX_IDENTITY, cpu->core);
 	READ_BCR(ARC_REG_ISA_CFG_BCR, cpu->isa);
 
-	READ_BCR(ARC_REG_TIMERS_BCR, cpu->timers);
+	READ_BCR(ARC_REG_TIMERS_BCR, timer);
+	cpu->extn.timer0 = timer.t0;
+	cpu->extn.timer1 = timer.t1;
+	cpu->extn.rtc = timer.rtc;
+
 	cpu->vec_base = read_aux_reg(AUX_INTR_VEC_BASE);
 
 	READ_BCR(ARC_REG_D_UNCACH_BCR, uncached_space);
@@ -71,36 +123,11 @@ static void read_arc_build_cfg_regs(void)
 	cpu->extn.swap = read_aux_reg(ARC_REG_SWAP_BCR) ? 1 : 0;        /* 1,3 */
 	cpu->extn.crc = read_aux_reg(ARC_REG_CRC_BCR) ? 1 : 0;
 	cpu->extn.minmax = read_aux_reg(ARC_REG_MIXMAX_BCR) > 1 ? 1 : 0; /* 2 */
-
-	/* Note that we read the CCM BCRs independent of kernel config
-	 * This is to catch the cases where user doesn't know that
-	 * CCMs are present in hardware build
-	 */
-	{
-		struct bcr_iccm iccm;
-		struct bcr_dccm dccm;
-		struct bcr_dccm_base dccm_base;
-		unsigned int bcr_32bit_val;
-
-		bcr_32bit_val = read_aux_reg(ARC_REG_ICCM_BCR);
-		if (bcr_32bit_val) {
-			iccm = *((struct bcr_iccm *)&bcr_32bit_val);
-			cpu->iccm.base_addr = iccm.base << 16;
-			cpu->iccm.sz = 0x2000 << (iccm.sz - 1);
-		}
-
-		bcr_32bit_val = read_aux_reg(ARC_REG_DCCM_BCR);
-		if (bcr_32bit_val) {
-			dccm = *((struct bcr_dccm *)&bcr_32bit_val);
-			cpu->dccm.sz = 0x800 << (dccm.sz);
-
-			READ_BCR(ARC_REG_DCCMBASE_BCR, dccm_base);
-			cpu->dccm.base_addr = dccm_base.addr << 8;
-		}
-	}
-
 	READ_BCR(ARC_REG_XY_MEM_BCR, cpu->extn_xymem);
 
+	/* Read CCM BCRs for boot reporting even if not enabled in Kconfig */
+	read_decode_ccm_bcr(cpu);
+
 	read_decode_mmu_bcr();
 	read_decode_cache_bcr();
 
@@ -208,9 +235,9 @@ static char *arc_cpu_mumbojumbo(int cpu_id, char *buf, int len)
 		       (unsigned int)(arc_get_core_freq() / 10000) % 100);
 
 	n += scnprintf(buf + n, len - n, "Timers\t\t: %s%s%s%s\nISA Extn\t: ",
-		       IS_AVAIL1(cpu->timers.t0, "Timer0 "),
-		       IS_AVAIL1(cpu->timers.t1, "Timer1 "),
-		       IS_AVAIL2(cpu->timers.rtc, "64-bit RTC ",
+		       IS_AVAIL1(cpu->extn.timer0, "Timer0 "),
+		       IS_AVAIL1(cpu->extn.timer1, "Timer1 "),
+		       IS_AVAIL2(cpu->extn.rtc, "Local-64-bit-Ctr ",
 				 CONFIG_ARC_HAS_RTC));
 
 	n += i = scnprintf(buf + n, len - n, "%s%s%s%s%s",
@@ -232,8 +259,6 @@ static char *arc_cpu_mumbojumbo(int cpu_id, char *buf, int len)
 
 			n += scnprintf(buf + n, len - n, "mpy[opt %d] ", opt);
 		}
-		n += scnprintf(buf + n, len - n, "%s",
-			       IS_USED_CFG(CONFIG_ARC_HAS_HW_MPY));
 	}
 
 	n += scnprintf(buf + n, len - n, "%s%s%s%s%s%s%s%s\n",
@@ -293,13 +318,13 @@ static void arc_chk_core_config(void)
 	struct cpuinfo_arc *cpu = &cpuinfo_arc700[smp_processor_id()];
 	int fpu_enabled;
 
-	if (!cpu->timers.t0)
+	if (!cpu->extn.timer0)
 		panic("Timer0 is not present!\n");
 
-	if (!cpu->timers.t1)
+	if (!cpu->extn.timer1)
 		panic("Timer1 is not present!\n");
 
-	if (IS_ENABLED(CONFIG_ARC_HAS_RTC) && !cpu->timers.rtc)
+	if (IS_ENABLED(CONFIG_ARC_HAS_RTC) && !cpu->extn.rtc)
 		panic("RTC is not present\n");
 
 #ifdef CONFIG_ARC_HAS_DCCM
@@ -334,6 +359,7 @@ static void arc_chk_core_config(void)
 		panic("FPU non-existent, disable CONFIG_ARC_FPU_SAVE_RESTORE\n");
 
 	if (is_isa_arcv2() && IS_ENABLED(CONFIG_SMP) && cpu->isa.atomic &&
+	    IS_ENABLED(CONFIG_ARC_HAS_LLSC) &&
 	    !IS_ENABLED(CONFIG_ARC_STAR_9000923308))
 		panic("llock/scond livelock workaround missing\n");
 }
diff --git a/arch/arc/kernel/smp.c b/arch/arc/kernel/smp.c
index ef6e9e15b82a..424e937da5c8 100644
--- a/arch/arc/kernel/smp.c
+++ b/arch/arc/kernel/smp.c
@@ -336,11 +336,8 @@ irqreturn_t do_IPI(int irq, void *dev_id)
 		int rc;
 
 		rc = __do_IPI(msg);
-#ifdef CONFIG_ARC_IPI_DBG
-		/* IPI received but no valid @msg */
 		if (rc)
 			pr_info("IPI with bogus msg %ld in %ld\n", msg, copy);
-#endif
 		pending &= ~(1U << msg);
 	} while (pending);
 
diff --git a/arch/arc/kernel/time.c b/arch/arc/kernel/time.c
index dfad287f1db1..156d9833ff84 100644
--- a/arch/arc/kernel/time.c
+++ b/arch/arc/kernel/time.c
@@ -62,7 +62,7 @@
 
 /********** Clock Source Device *********/
 
-#ifdef CONFIG_ARC_HAS_GRTC
+#ifdef CONFIG_ARC_HAS_GFRC
 
 static int arc_counter_setup(void)
 {
@@ -83,10 +83,10 @@ static cycle_t arc_counter_read(struct clocksource *cs)
 
 	local_irq_save(flags);
 
-	__mcip_cmd(CMD_GRTC_READ_LO, 0);
+	__mcip_cmd(CMD_GFRC_READ_LO, 0);
 	stamp.l = read_aux_reg(ARC_REG_MCIP_READBACK);
 
-	__mcip_cmd(CMD_GRTC_READ_HI, 0);
+	__mcip_cmd(CMD_GFRC_READ_HI, 0);
 	stamp.h = read_aux_reg(ARC_REG_MCIP_READBACK);
 
 	local_irq_restore(flags);
@@ -95,7 +95,7 @@ static cycle_t arc_counter_read(struct clocksource *cs)
 }
 
 static struct clocksource arc_counter = {
-	.name   = "ARConnect GRTC",
+	.name   = "ARConnect GFRC",
 	.rating = 400,
 	.read   = arc_counter_read,
 	.mask   = CLOCKSOURCE_MASK(64),
diff --git a/arch/arm/boot/compressed/Makefile b/arch/arm/boot/compressed/Makefile
index 7a6a58ef8aaf..43788b1a1ac5 100644
--- a/arch/arm/boot/compressed/Makefile
+++ b/arch/arm/boot/compressed/Makefile
@@ -195,5 +195,7 @@ CFLAGS_font.o := -Dstatic=
 $(obj)/font.c: $(FONTC)
 	$(call cmd,shipped)
 
+AFLAGS_hyp-stub.o := -Wa,-march=armv7-a
+
 $(obj)/hyp-stub.S: $(srctree)/arch/$(SRCARCH)/kernel/hyp-stub.S
 	$(call cmd,shipped)
diff --git a/arch/arm/boot/dts/am335x-bone-common.dtsi b/arch/arm/boot/dts/am335x-bone-common.dtsi
index f3db13d2d90e..0cc150b87b86 100644
--- a/arch/arm/boot/dts/am335x-bone-common.dtsi
+++ b/arch/arm/boot/dts/am335x-bone-common.dtsi
@@ -285,8 +285,10 @@
 	};
 };
 
+
+/include/ "tps65217.dtsi"
+
 &tps {
-	compatible = "ti,tps65217";
 	/*
 	 * Configure pmic to enter OFF-state instead of SLEEP-state ("RTC-only
 	 * mode") at poweroff.  Most BeagleBone versions do not support RTC-only
@@ -307,17 +309,12 @@
 	ti,pmic-shutdown-controller;
 
 	regulators {
-		#address-cells = <1>;
-		#size-cells = <0>;
-
 		dcdc1_reg: regulator@0 {
-			reg = <0>;
 			regulator-name = "vdds_dpr";
 			regulator-always-on;
 		};
 
 		dcdc2_reg: regulator@1 {
-			reg = <1>;
 			/* VDD_MPU voltage limits 0.95V - 1.26V with +/-4% tolerance */
 			regulator-name = "vdd_mpu";
 			regulator-min-microvolt = <925000>;
@@ -327,7 +324,6 @@
 		};
 
 		dcdc3_reg: regulator@2 {
-			reg = <2>;
 			/* VDD_CORE voltage limits 0.95V - 1.1V with +/-4% tolerance */
 			regulator-name = "vdd_core";
 			regulator-min-microvolt = <925000>;
@@ -337,25 +333,21 @@
 		};
 
 		ldo1_reg: regulator@3 {
-			reg = <3>;
 			regulator-name = "vio,vrtc,vdds";
 			regulator-always-on;
 		};
 
 		ldo2_reg: regulator@4 {
-			reg = <4>;
 			regulator-name = "vdd_3v3aux";
 			regulator-always-on;
 		};
 
 		ldo3_reg: regulator@5 {
-			reg = <5>;
 			regulator-name = "vdd_1v8";
 			regulator-always-on;
 		};
 
 		ldo4_reg: regulator@6 {
-			reg = <6>;
 			regulator-name = "vdd_3v3a";
 			regulator-always-on;
 		};
diff --git a/arch/arm/boot/dts/am335x-chilisom.dtsi b/arch/arm/boot/dts/am335x-chilisom.dtsi
index fda457b07e15..857d9894103a 100644
--- a/arch/arm/boot/dts/am335x-chilisom.dtsi
+++ b/arch/arm/boot/dts/am335x-chilisom.dtsi
@@ -128,21 +128,16 @@
 
 };
 
-&tps {
-	compatible = "ti,tps65217";
+/include/ "tps65217.dtsi"
 
+&tps {
 	regulators {
-		#address-cells = <1>;
-		#size-cells = <0>;
-
 		dcdc1_reg: regulator@0 {
-			reg = <0>;
 			regulator-name = "vdds_dpr";
 			regulator-always-on;
 		};
 
 		dcdc2_reg: regulator@1 {
-			reg = <1>;
 			/* VDD_MPU voltage limits 0.95V - 1.26V with +/-4% tolerance */
 			regulator-name = "vdd_mpu";
 			regulator-min-microvolt = <925000>;
@@ -152,7 +147,6 @@
 		};
 
 		dcdc3_reg: regulator@2 {
-			reg = <2>;
 			/* VDD_CORE voltage limits 0.95V - 1.1V with +/-4% tolerance */
 			regulator-name = "vdd_core";
 			regulator-min-microvolt = <925000>;
@@ -162,28 +156,24 @@
 		};
 
 		ldo1_reg: regulator@3 {
-			reg = <3>;
 			regulator-name = "vio,vrtc,vdds";
 			regulator-boot-on;
 			regulator-always-on;
 		};
 
 		ldo2_reg: regulator@4 {
-			reg = <4>;
 			regulator-name = "vdd_3v3aux";
 			regulator-boot-on;
 			regulator-always-on;
 		};
 
 		ldo3_reg: regulator@5 {
-			reg = <5>;
 			regulator-name = "vdd_1v8";
 			regulator-boot-on;
 			regulator-always-on;
 		};
 
 		ldo4_reg: regulator@6 {
-			reg = <6>;
 			regulator-name = "vdd_3v3d";
 			regulator-boot-on;
 			regulator-always-on;
diff --git a/arch/arm/boot/dts/am335x-nano.dts b/arch/arm/boot/dts/am335x-nano.dts
index 77559a1ded60..f313999c503e 100644
--- a/arch/arm/boot/dts/am335x-nano.dts
+++ b/arch/arm/boot/dts/am335x-nano.dts
@@ -375,15 +375,11 @@
 	wp-gpios = <&gpio3 18 0>;
 };
 
-&tps {
-	compatible = "ti,tps65217";
+#include "tps65217.dtsi"
 
+&tps {
 	regulators {
-		#address-cells = <1>;
-		#size-cells = <0>;
-
 		dcdc1_reg: regulator@0 {
-			reg = <0>;
 			/* +1.5V voltage with ±4% tolerance */
 			regulator-min-microvolt = <1450000>;
 			regulator-max-microvolt = <1550000>;
@@ -392,7 +388,6 @@
 		};
 
 		dcdc2_reg: regulator@1 {
-			reg = <1>;
 			/* VDD_MPU voltage limits 0.95V - 1.1V with ±4% tolerance */
 			regulator-name = "vdd_mpu";
 			regulator-min-microvolt = <915000>;
@@ -402,7 +397,6 @@
 		};
 
 		dcdc3_reg: regulator@2 {
-			reg = <2>;
 			/* VDD_CORE voltage limits 0.95V - 1.1V with ±4% tolerance */
 			regulator-name = "vdd_core";
 			regulator-min-microvolt = <915000>;
@@ -412,7 +406,6 @@
 		};
 
 		ldo1_reg: regulator@3 {
-			reg = <3>;
 			/* +1.8V voltage with ±4% tolerance */
 			regulator-min-microvolt = <1750000>;
 			regulator-max-microvolt = <1870000>;
@@ -421,7 +414,6 @@
 		};
 
 		ldo2_reg: regulator@4 {
-			reg = <4>;
 			/* +3.3V voltage with ±4% tolerance */
 			regulator-min-microvolt = <3175000>;
 			regulator-max-microvolt = <3430000>;
@@ -430,7 +422,6 @@
 		};
 
 		ldo3_reg: regulator@5 {
-			reg = <5>;
 			/* +1.8V voltage with ±4% tolerance */
 			regulator-min-microvolt = <1750000>;
 			regulator-max-microvolt = <1870000>;
@@ -439,7 +430,6 @@
 		};
 
 		ldo4_reg: regulator@6 {
-			reg = <6>;
 			/* +3.3V voltage with ±4% tolerance */
 			regulator-min-microvolt = <3175000>;
 			regulator-max-microvolt = <3430000>;
diff --git a/arch/arm/boot/dts/am335x-pepper.dts b/arch/arm/boot/dts/am335x-pepper.dts
index 471a3a70ea1f..8867aaaec54d 100644
--- a/arch/arm/boot/dts/am335x-pepper.dts
+++ b/arch/arm/boot/dts/am335x-pepper.dts
@@ -420,9 +420,9 @@
 	vin-supply = <&vbat>;
 };
 
-&tps {
-	compatible = "ti,tps65217";
+/include/ "tps65217.dtsi"
 
+&tps {
 	backlight {
 		isel = <1>; /* ISET1 */
 		fdim = <200>; /* TPS65217_BL_FDIM_200HZ */
@@ -430,17 +430,12 @@
 	};
 
 	regulators {
-		#address-cells = <1>;
-		#size-cells = <0>;
-
 		dcdc1_reg: regulator@0 {
-			reg = <0>;
 			/* VDD_1V8 system supply */
 			regulator-always-on;
 		};
 
 		dcdc2_reg: regulator@1 {
-			reg = <1>;
 			/* VDD_CORE voltage limits 0.95V - 1.26V with +/-4% tolerance */
 			regulator-name = "vdd_core";
 			regulator-min-microvolt = <925000>;
@@ -450,7 +445,6 @@
 		};
 
 		dcdc3_reg: regulator@2 {
-			reg = <2>;
 			/* VDD_MPU voltage limits 0.95V - 1.1V with +/-4% tolerance */
 			regulator-name = "vdd_mpu";
 			regulator-min-microvolt = <925000>;
@@ -460,21 +454,18 @@
 		};
 
 		ldo1_reg: regulator@3 {
-			reg = <3>;
 			/* VRTC 1.8V always-on supply */
 			regulator-name = "vrtc,vdds";
 			regulator-always-on;
 		};
 
 		ldo2_reg: regulator@4 {
-			reg = <4>;
 			/* 3.3V rail */
 			regulator-name = "vdd_3v3aux";
 			regulator-always-on;
 		};
 
 		ldo3_reg: regulator@5 {
-			reg = <5>;
 			/* VDD_3V3A 3.3V rail */
 			regulator-name = "vdd_3v3a";
 			regulator-min-microvolt = <3300000>;
@@ -482,7 +473,6 @@
 		};
 
 		ldo4_reg: regulator@6 {
-			reg = <6>;
 			/* VDD_3V3B 3.3V rail */
 			regulator-name = "vdd_3v3b";
 			regulator-always-on;
diff --git a/arch/arm/boot/dts/am335x-shc.dts b/arch/arm/boot/dts/am335x-shc.dts
index 1b5b044fcd91..865de8500f1c 100644
--- a/arch/arm/boot/dts/am335x-shc.dts
+++ b/arch/arm/boot/dts/am335x-shc.dts
@@ -46,7 +46,7 @@
 			gpios = <&gpio1 29 GPIO_ACTIVE_HIGH>;
 			linux,code = <KEY_BACK>;
 			debounce-interval = <1000>;
-			gpio-key,wakeup;
+			wakeup-source;
 		};
 
 		front_button {
@@ -54,7 +54,7 @@
 			gpios = <&gpio1 25 GPIO_ACTIVE_HIGH>;
 			linux,code = <KEY_FRONT>;
 			debounce-interval = <1000>;
-			gpio-key,wakeup;
+			wakeup-source;
 		};
 	};
 
diff --git a/arch/arm/boot/dts/am335x-sl50.dts b/arch/arm/boot/dts/am335x-sl50.dts
index d38edfa53bb9..3303c281697b 100644
--- a/arch/arm/boot/dts/am335x-sl50.dts
+++ b/arch/arm/boot/dts/am335x-sl50.dts
@@ -375,19 +375,16 @@
 	pinctrl-0 = <&uart4_pins>;
 };
 
+#include "tps65217.dtsi"
+
 &tps {
-	compatible = "ti,tps65217";
 	ti,pmic-shutdown-controller;
 
 	interrupt-parent = <&intc>;
 	interrupts = <7>;	/* NNMI */
 
 	regulators {
-		#address-cells = <1>;
-		#size-cells = <0>;
-
 		dcdc1_reg: regulator@0 {
-			reg = <0>;
 			/* VDDS_DDR */
 			regulator-min-microvolt = <1500000>;
 			regulator-max-microvolt = <1500000>;
@@ -395,7 +392,6 @@
 		};
 
 		dcdc2_reg: regulator@1 {
-			reg = <1>;
 			/* VDD_MPU voltage limits 0.95V - 1.26V with +/-4% tolerance */
 			regulator-name = "vdd_mpu";
 			regulator-min-microvolt = <925000>;
@@ -405,7 +401,6 @@
 		};
 
 		dcdc3_reg: regulator@2 {
-			reg = <2>;
 			/* VDD_CORE voltage limits 0.95V - 1.1V with +/-4% tolerance */
 			regulator-name = "vdd_core";
 			regulator-min-microvolt = <925000>;
@@ -415,7 +410,6 @@
 		};
 
 		ldo1_reg: regulator@3 {
-			reg = <3>;
 			/* VRTC / VIO / VDDS*/
 			regulator-always-on;
 			regulator-min-microvolt = <1800000>;
@@ -423,7 +417,6 @@
 		};
 
 		ldo2_reg: regulator@4 {
-			reg = <4>;
 			/* VDD_3V3AUX */
 			regulator-always-on;
 			regulator-min-microvolt = <3300000>;
@@ -431,7 +424,6 @@
 		};
 
 		ldo3_reg: regulator@5 {
-			reg = <5>;
 			/* VDD_1V8 */
 			regulator-min-microvolt = <1800000>;
 			regulator-max-microvolt = <1800000>;
@@ -439,7 +431,6 @@
 		};
 
 		ldo4_reg: regulator@6 {
-			reg = <6>;
 			/* VDD_3V3A */
 			regulator-min-microvolt = <3300000>;
 			regulator-max-microvolt = <3300000>;
diff --git a/arch/arm/boot/dts/am57xx-beagle-x15.dts b/arch/arm/boot/dts/am57xx-beagle-x15.dts
index 36c0fa6c362a..a0986c65be0c 100644
--- a/arch/arm/boot/dts/am57xx-beagle-x15.dts
+++ b/arch/arm/boot/dts/am57xx-beagle-x15.dts
@@ -173,6 +173,8 @@
 
 		sound0_master: simple-audio-card,codec {
 			sound-dai = <&tlv320aic3104>;
+			assigned-clocks = <&clkoutmux2_clk_mux>;
+			assigned-clock-parents = <&sys_clk2_dclk_div>;
 			clocks = <&clkout2_clk>;
 		};
 	};
@@ -796,6 +798,8 @@
 	pinctrl-names = "default", "sleep";
 	pinctrl-0 = <&mcasp3_pins_default>;
 	pinctrl-1 = <&mcasp3_pins_sleep>;
+	assigned-clocks = <&mcasp3_ahclkx_mux>;
+	assigned-clock-parents = <&sys_clkin2>;
 	status = "okay";
 
 	op-mode = <0>;	/* MCASP_IIS_MODE */
diff --git a/arch/arm/boot/dts/am57xx-cl-som-am57x.dts b/arch/arm/boot/dts/am57xx-cl-som-am57x.dts
index 8d93882dc8d5..1c06cb76da07 100644
--- a/arch/arm/boot/dts/am57xx-cl-som-am57x.dts
+++ b/arch/arm/boot/dts/am57xx-cl-som-am57x.dts
@@ -545,7 +545,7 @@
 		ti,debounce-tol = /bits/ 16 <10>;
 		ti,debounce-rep = /bits/ 16 <1>;
 
-		linux,wakeup;
+		wakeup-source;
 	};
 };
 
diff --git a/arch/arm/boot/dts/armada-xp-axpwifiap.dts b/arch/arm/boot/dts/armada-xp-axpwifiap.dts
index 23fc670c0427..5c21b236721f 100644
--- a/arch/arm/boot/dts/armada-xp-axpwifiap.dts
+++ b/arch/arm/boot/dts/armada-xp-axpwifiap.dts
@@ -70,8 +70,8 @@
 	soc {
 		ranges = <MBUS_ID(0xf0, 0x01) 0 0 0xf1000000 0x100000
 			  MBUS_ID(0x01, 0x1d) 0 0 0xfff00000 0x100000
-			  MBUS_ID(0x09, 0x09) 0 0 0xf8100000 0x10000
-			  MBUS_ID(0x09, 0x05) 0 0 0xf8110000 0x10000>;
+			  MBUS_ID(0x09, 0x09) 0 0 0xf1100000 0x10000
+			  MBUS_ID(0x09, 0x05) 0 0 0xf1110000 0x10000>;
 
 		pcie-controller {
 			status = "okay";
diff --git a/arch/arm/boot/dts/armada-xp-db.dts b/arch/arm/boot/dts/armada-xp-db.dts
index f774101416a5..ebe1d267406d 100644
--- a/arch/arm/boot/dts/armada-xp-db.dts
+++ b/arch/arm/boot/dts/armada-xp-db.dts
@@ -76,8 +76,8 @@
 		ranges = <MBUS_ID(0xf0, 0x01) 0 0 0xf1000000 0x100000
 			  MBUS_ID(0x01, 0x1d) 0 0 0xfff00000 0x100000
 			  MBUS_ID(0x01, 0x2f) 0 0 0xf0000000 0x1000000
-			  MBUS_ID(0x09, 0x09) 0 0 0xf8100000 0x10000
-			  MBUS_ID(0x09, 0x05) 0 0 0xf8110000 0x10000>;
+			  MBUS_ID(0x09, 0x09) 0 0 0xf1100000 0x10000
+			  MBUS_ID(0x09, 0x05) 0 0 0xf1110000 0x10000>;
 
 		devbus-bootcs {
 			status = "okay";
diff --git a/arch/arm/boot/dts/armada-xp-gp.dts b/arch/arm/boot/dts/armada-xp-gp.dts
index 4878d7353069..5730b875c4f5 100644
--- a/arch/arm/boot/dts/armada-xp-gp.dts
+++ b/arch/arm/boot/dts/armada-xp-gp.dts
@@ -95,8 +95,8 @@
 		ranges = <MBUS_ID(0xf0, 0x01) 0 0 0xf1000000 0x100000
 			  MBUS_ID(0x01, 0x1d) 0 0 0xfff00000 0x100000
 			  MBUS_ID(0x01, 0x2f) 0 0 0xf0000000 0x1000000
-			  MBUS_ID(0x09, 0x09) 0 0 0xf8100000 0x10000
-			  MBUS_ID(0x09, 0x05) 0 0 0xf8110000 0x10000>;
+			  MBUS_ID(0x09, 0x09) 0 0 0xf1100000 0x10000
+			  MBUS_ID(0x09, 0x05) 0 0 0xf1110000 0x10000>;
 
 		devbus-bootcs {
 			status = "okay";
diff --git a/arch/arm/boot/dts/armada-xp-lenovo-ix4-300d.dts b/arch/arm/boot/dts/armada-xp-lenovo-ix4-300d.dts
index fb9e1bbf2338..8af463f26ea1 100644
--- a/arch/arm/boot/dts/armada-xp-lenovo-ix4-300d.dts
+++ b/arch/arm/boot/dts/armada-xp-lenovo-ix4-300d.dts
@@ -65,8 +65,8 @@
 	soc {
 		ranges = <MBUS_ID(0xf0, 0x01) 0 0 0xd0000000 0x100000
 			MBUS_ID(0x01, 0x1d) 0 0 0xfff00000 0x100000
-			MBUS_ID(0x09, 0x09) 0 0 0xf8100000 0x10000
-			MBUS_ID(0x09, 0x05) 0 0 0xf8110000 0x10000>;
+			MBUS_ID(0x09, 0x09) 0 0 0xf1100000 0x10000
+			MBUS_ID(0x09, 0x05) 0 0 0xf1110000 0x10000>;
 
 		pcie-controller {
 			status = "okay";
diff --git a/arch/arm/boot/dts/armada-xp-linksys-mamba.dts b/arch/arm/boot/dts/armada-xp-linksys-mamba.dts
index 6e9820e141f8..b89e6cf1271a 100644
--- a/arch/arm/boot/dts/armada-xp-linksys-mamba.dts
+++ b/arch/arm/boot/dts/armada-xp-linksys-mamba.dts
@@ -70,8 +70,8 @@
 	soc {
 		ranges = <MBUS_ID(0xf0, 0x01) 0 0 0xf1000000 0x100000
 			  MBUS_ID(0x01, 0x1d) 0 0 0xfff00000 0x100000
-			  MBUS_ID(0x09, 0x09) 0 0 0xf8100000 0x10000
-			  MBUS_ID(0x09, 0x05) 0 0 0xf8110000 0x10000>;
+			  MBUS_ID(0x09, 0x09) 0 0 0xf1100000 0x10000
+			  MBUS_ID(0x09, 0x05) 0 0 0xf1110000 0x10000>;
 
 		pcie-controller {
 			status = "okay";
diff --git a/arch/arm/boot/dts/armada-xp-matrix.dts b/arch/arm/boot/dts/armada-xp-matrix.dts
index 6ab33837a2b6..6522b04f4a8e 100644
--- a/arch/arm/boot/dts/armada-xp-matrix.dts
+++ b/arch/arm/boot/dts/armada-xp-matrix.dts
@@ -68,8 +68,8 @@
 	soc {
 		ranges = <MBUS_ID(0xf0, 0x01) 0 0 0xf1000000 0x100000
 			  MBUS_ID(0x01, 0x1d) 0 0 0xfff00000 0x100000
-			  MBUS_ID(0x09, 0x09) 0 0 0xf8100000 0x10000
-			  MBUS_ID(0x09, 0x05) 0 0 0xf8110000 0x10000>;
+			  MBUS_ID(0x09, 0x09) 0 0 0xf1100000 0x10000
+			  MBUS_ID(0x09, 0x05) 0 0 0xf1110000 0x10000>;
 
 		internal-regs {
 			serial@12000 {
diff --git a/arch/arm/boot/dts/armada-xp-netgear-rn2120.dts b/arch/arm/boot/dts/armada-xp-netgear-rn2120.dts
index 62175a8848bc..d19f44c70925 100644
--- a/arch/arm/boot/dts/armada-xp-netgear-rn2120.dts
+++ b/arch/arm/boot/dts/armada-xp-netgear-rn2120.dts
@@ -64,8 +64,8 @@
 	soc {
 		ranges = <MBUS_ID(0xf0, 0x01) 0 0 0xd0000000 0x100000
 			  MBUS_ID(0x01, 0x1d) 0 0 0xfff00000 0x100000
-			  MBUS_ID(0x09, 0x09) 0 0 0xf8100000 0x10000
-			  MBUS_ID(0x09, 0x05) 0 0 0xf8110000 0x10000>;
+			  MBUS_ID(0x09, 0x09) 0 0 0xf1100000 0x10000
+			  MBUS_ID(0x09, 0x05) 0 0 0xf1110000 0x10000>;
 
 		pcie-controller {
 			status = "okay";
diff --git a/arch/arm/boot/dts/armada-xp-openblocks-ax3-4.dts b/arch/arm/boot/dts/armada-xp-openblocks-ax3-4.dts
index a5db17782e08..853bd392a4fe 100644
--- a/arch/arm/boot/dts/armada-xp-openblocks-ax3-4.dts
+++ b/arch/arm/boot/dts/armada-xp-openblocks-ax3-4.dts
@@ -65,9 +65,9 @@
 	soc {
 		ranges = <MBUS_ID(0xf0, 0x01) 0 0 0xd0000000 0x100000
 			  MBUS_ID(0x01, 0x1d) 0 0 0xfff00000 0x100000
-			  MBUS_ID(0x01, 0x2f) 0 0 0xf0000000 0x8000000
-			  MBUS_ID(0x09, 0x09) 0 0 0xf8100000 0x10000
-			  MBUS_ID(0x09, 0x05) 0 0 0xf8110000 0x10000>;
+			  MBUS_ID(0x01, 0x2f) 0 0 0xe8000000 0x8000000
+			  MBUS_ID(0x09, 0x09) 0 0 0xf1100000 0x10000
+			  MBUS_ID(0x09, 0x05) 0 0 0xf1110000 0x10000>;
 
 		devbus-bootcs {
 			status = "okay";
diff --git a/arch/arm/boot/dts/armada-xp-synology-ds414.dts b/arch/arm/boot/dts/armada-xp-synology-ds414.dts
index 2391b11dc546..d17dab0a6f51 100644
--- a/arch/arm/boot/dts/armada-xp-synology-ds414.dts
+++ b/arch/arm/boot/dts/armada-xp-synology-ds414.dts
@@ -78,8 +78,8 @@
 	soc {
 		ranges = <MBUS_ID(0xf0, 0x01) 0 0 0xf1000000 0x100000
 			  MBUS_ID(0x01, 0x1d) 0 0 0xfff00000 0x100000
-			  MBUS_ID(0x09, 0x09) 0 0 0xf8100000 0x10000
-			  MBUS_ID(0x09, 0x05) 0 0 0xf8110000 0x10000>;
+			  MBUS_ID(0x09, 0x09) 0 0 0xf1100000 0x10000
+			  MBUS_ID(0x09, 0x05) 0 0 0xf1110000 0x10000>;
 
 		pcie-controller {
 			status = "okay";
diff --git a/arch/arm/boot/dts/dra7.dtsi b/arch/arm/boot/dts/dra7.dtsi
index c4d9175b90dc..f82aa44c3cee 100644
--- a/arch/arm/boot/dts/dra7.dtsi
+++ b/arch/arm/boot/dts/dra7.dtsi
@@ -1500,6 +1500,16 @@
 			       0x48485200 0x2E00>;
 			#address-cells = <1>;
 			#size-cells = <1>;
+
+			/*
+			 * Do not allow gating of cpsw clock as workaround
+			 * for errata i877. Keeping internal clock disabled
+			 * causes the device switching characteristics
+			 * to degrade over time and eventually fail to meet
+			 * the data manual delay time/skew specs.
+			 */
+			ti,no-idle;
+
 			/*
 			 * rx_thresh_pend
 			 * rx_pend
diff --git a/arch/arm/boot/dts/imx6qdl.dtsi b/arch/arm/boot/dts/imx6qdl.dtsi
index 4f6ae921656f..f74d3db4846d 100644
--- a/arch/arm/boot/dts/imx6qdl.dtsi
+++ b/arch/arm/boot/dts/imx6qdl.dtsi
@@ -896,7 +896,6 @@
 				#size-cells = <1>;
 				reg = <0x2100000 0x10000>;
 				ranges = <0 0x2100000 0x10000>;
-				interrupt-parent = <&intc>;
 				clocks = <&clks IMX6QDL_CLK_CAAM_MEM>,
 					 <&clks IMX6QDL_CLK_CAAM_ACLK>,
 					 <&clks IMX6QDL_CLK_CAAM_IPG>,
diff --git a/arch/arm/boot/dts/kirkwood-ds112.dts b/arch/arm/boot/dts/kirkwood-ds112.dts
index bf4143c6cb8f..b84af3da8c84 100644
--- a/arch/arm/boot/dts/kirkwood-ds112.dts
+++ b/arch/arm/boot/dts/kirkwood-ds112.dts
@@ -14,7 +14,7 @@
 #include "kirkwood-synology.dtsi"
 
 / {
-	model = "Synology DS111";
+	model = "Synology DS112";
 	compatible = "synology,ds111", "marvell,kirkwood";
 
 	memory {
diff --git a/arch/arm/boot/dts/orion5x-linkstation-lswtgl.dts b/arch/arm/boot/dts/orion5x-linkstation-lswtgl.dts
index 420788229e6f..aae8a7aceab7 100644
--- a/arch/arm/boot/dts/orion5x-linkstation-lswtgl.dts
+++ b/arch/arm/boot/dts/orion5x-linkstation-lswtgl.dts
@@ -228,6 +228,37 @@
 	};
 };
 
+&devbus_bootcs {
+	status = "okay";
+	devbus,keep-config;
+
+	flash@0 {
+		compatible = "jedec-flash";
+		reg = <0 0x40000>;
+		bank-width = <1>;
+
+		partitions {
+			compatible = "fixed-partitions";
+			#address-cells = <1>;
+			#size-cells = <1>;
+
+			header@0 {
+				reg = <0 0x30000>;
+				read-only;
+			};
+
+			uboot@30000 {
+				reg = <0x30000 0xF000>;
+				read-only;
+			};
+
+			uboot_env@3F000 {
+				reg = <0x3F000 0x1000>;
+			};
+		};
+	};
+};
+
 &mdio {
 	status = "okay";
 
diff --git a/arch/arm/boot/dts/r8a7791-porter.dts b/arch/arm/boot/dts/r8a7791-porter.dts
index 6713b1ea732b..01d239c3eaaa 100644
--- a/arch/arm/boot/dts/r8a7791-porter.dts
+++ b/arch/arm/boot/dts/r8a7791-porter.dts
@@ -283,7 +283,6 @@
 	pinctrl-names = "default";
 
 	status = "okay";
-	renesas,enable-gpio = <&gpio5 31 GPIO_ACTIVE_HIGH>;
 };
 
 &usbphy {
diff --git a/arch/arm/boot/dts/sama5d2-pinfunc.h b/arch/arm/boot/dts/sama5d2-pinfunc.h
index 1afe24629d1f..b0c912feaa2f 100644
--- a/arch/arm/boot/dts/sama5d2-pinfunc.h
+++ b/arch/arm/boot/dts/sama5d2-pinfunc.h
@@ -90,7 +90,7 @@
 #define PIN_PA14__I2SC1_MCK		PINMUX_PIN(PIN_PA14, 4, 2)
 #define PIN_PA14__FLEXCOM3_IO2		PINMUX_PIN(PIN_PA14, 5, 1)
 #define PIN_PA14__D9			PINMUX_PIN(PIN_PA14, 6, 2)
-#define PIN_PA15			14
+#define PIN_PA15			15
 #define PIN_PA15__GPIO			PINMUX_PIN(PIN_PA15, 0, 0)
 #define PIN_PA15__SPI0_MOSI		PINMUX_PIN(PIN_PA15, 1, 1)
 #define PIN_PA15__TF1			PINMUX_PIN(PIN_PA15, 2, 1)
diff --git a/arch/arm/boot/dts/tps65217.dtsi b/arch/arm/boot/dts/tps65217.dtsi
new file mode 100644
index 000000000000..a63272422d76
--- /dev/null
+++ b/arch/arm/boot/dts/tps65217.dtsi
@@ -0,0 +1,56 @@
+/*
+ * Copyright (C) 2012 Texas Instruments Incorporated - http://www.ti.com/
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+
+/*
+ * Integrated Power Management Chip
+ * http://www.ti.com/lit/ds/symlink/tps65217.pdf
+ */
+
+&tps {
+	compatible = "ti,tps65217";
+
+	regulators {
+		#address-cells = <1>;
+		#size-cells = <0>;
+
+		dcdc1_reg: regulator@0 {
+			reg = <0>;
+			regulator-compatible = "dcdc1";
+		};
+
+		dcdc2_reg: regulator@1 {
+			reg = <1>;
+			regulator-compatible = "dcdc2";
+		};
+
+		dcdc3_reg: regulator@2 {
+			reg = <2>;
+			regulator-compatible = "dcdc3";
+		};
+
+		ldo1_reg: regulator@3 {
+			reg = <3>;
+			regulator-compatible = "ldo1";
+		};
+
+		ldo2_reg: regulator@4 {
+			reg = <4>;
+			regulator-compatible = "ldo2";
+		};
+
+		ldo3_reg: regulator@5 {
+			reg = <5>;
+			regulator-compatible = "ldo3";
+		};
+
+		ldo4_reg: regulator@6 {
+			reg = <6>;
+			regulator-compatible = "ldo4";
+		};
+	};
+};
diff --git a/arch/arm/common/icst.c b/arch/arm/common/icst.c
index 2dc6da70ae59..d7ed252708c5 100644
--- a/arch/arm/common/icst.c
+++ b/arch/arm/common/icst.c
@@ -16,7 +16,7 @@
  */
 #include <linux/module.h>
 #include <linux/kernel.h>
-
+#include <asm/div64.h>
 #include <asm/hardware/icst.h>
 
 /*
@@ -29,7 +29,11 @@ EXPORT_SYMBOL(icst525_s2div);
 
 unsigned long icst_hz(const struct icst_params *p, struct icst_vco vco)
 {
-	return p->ref * 2 * (vco.v + 8) / ((vco.r + 2) * p->s2div[vco.s]);
+	u64 dividend = p->ref * 2 * (u64)(vco.v + 8);
+	u32 divisor = (vco.r + 2) * p->s2div[vco.s];
+
+	do_div(dividend, divisor);
+	return (unsigned long)dividend;
 }
 
 EXPORT_SYMBOL(icst_hz);
@@ -58,6 +62,7 @@ icst_hz_to_vco(const struct icst_params *p, unsigned long freq)
 
 		if (f > p->vco_min && f <= p->vco_max)
 			break;
+		i++;
 	} while (i < 8);
 
 	if (i >= 8)
diff --git a/arch/arm/configs/omap2plus_defconfig b/arch/arm/configs/omap2plus_defconfig
index a7151744b85c..d18d6b42fcf5 100644
--- a/arch/arm/configs/omap2plus_defconfig
+++ b/arch/arm/configs/omap2plus_defconfig
@@ -292,24 +292,23 @@ CONFIG_FB=y
 CONFIG_FIRMWARE_EDID=y
 CONFIG_FB_MODE_HELPERS=y
 CONFIG_FB_TILEBLITTING=y
-CONFIG_OMAP2_DSS=m
-CONFIG_OMAP5_DSS_HDMI=y
-CONFIG_OMAP2_DSS_SDI=y
-CONFIG_OMAP2_DSS_DSI=y
+CONFIG_FB_OMAP5_DSS_HDMI=y
+CONFIG_FB_OMAP2_DSS_SDI=y
+CONFIG_FB_OMAP2_DSS_DSI=y
 CONFIG_FB_OMAP2=m
-CONFIG_DISPLAY_ENCODER_TFP410=m
-CONFIG_DISPLAY_ENCODER_TPD12S015=m
-CONFIG_DISPLAY_CONNECTOR_DVI=m
-CONFIG_DISPLAY_CONNECTOR_HDMI=m
-CONFIG_DISPLAY_CONNECTOR_ANALOG_TV=m
-CONFIG_DISPLAY_PANEL_DPI=m
-CONFIG_DISPLAY_PANEL_DSI_CM=m
-CONFIG_DISPLAY_PANEL_SONY_ACX565AKM=m
-CONFIG_DISPLAY_PANEL_LGPHILIPS_LB035Q02=m
-CONFIG_DISPLAY_PANEL_SHARP_LS037V7DW01=m
-CONFIG_DISPLAY_PANEL_TPO_TD028TTEC1=m
-CONFIG_DISPLAY_PANEL_TPO_TD043MTEA1=m
-CONFIG_DISPLAY_PANEL_NEC_NL8048HL11=m
+CONFIG_FB_OMAP2_ENCODER_TFP410=m
+CONFIG_FB_OMAP2_ENCODER_TPD12S015=m
+CONFIG_FB_OMAP2_CONNECTOR_DVI=m
+CONFIG_FB_OMAP2_CONNECTOR_HDMI=m
+CONFIG_FB_OMAP2_CONNECTOR_ANALOG_TV=m
+CONFIG_FB_OMAP2_PANEL_DPI=m
+CONFIG_FB_OMAP2_PANEL_DSI_CM=m
+CONFIG_FB_OMAP2_PANEL_SONY_ACX565AKM=m
+CONFIG_FB_OMAP2_PANEL_LGPHILIPS_LB035Q02=m
+CONFIG_FB_OMAP2_PANEL_SHARP_LS037V7DW01=m
+CONFIG_FB_OMAP2_PANEL_TPO_TD028TTEC1=m
+CONFIG_FB_OMAP2_PANEL_TPO_TD043MTEA1=m
+CONFIG_FB_OMAP2_PANEL_NEC_NL8048HL11=m
 CONFIG_BACKLIGHT_LCD_SUPPORT=y
 CONFIG_LCD_CLASS_DEVICE=y
 CONFIG_LCD_PLATFORM=y
diff --git a/arch/arm/crypto/aes-ce-glue.c b/arch/arm/crypto/aes-ce-glue.c
index b445a5d56f43..89a3a3e592d6 100644
--- a/arch/arm/crypto/aes-ce-glue.c
+++ b/arch/arm/crypto/aes-ce-glue.c
@@ -364,7 +364,7 @@ static struct crypto_alg aes_algs[] = { {
 	.cra_blkcipher = {
 		.min_keysize	= AES_MIN_KEY_SIZE,
 		.max_keysize	= AES_MAX_KEY_SIZE,
-		.ivsize		= AES_BLOCK_SIZE,
+		.ivsize		= 0,
 		.setkey		= ce_aes_setkey,
 		.encrypt	= ecb_encrypt,
 		.decrypt	= ecb_decrypt,
@@ -441,7 +441,7 @@ static struct crypto_alg aes_algs[] = { {
 	.cra_ablkcipher = {
 		.min_keysize	= AES_MIN_KEY_SIZE,
 		.max_keysize	= AES_MAX_KEY_SIZE,
-		.ivsize		= AES_BLOCK_SIZE,
+		.ivsize		= 0,
 		.setkey		= ablk_set_key,
 		.encrypt	= ablk_encrypt,
 		.decrypt	= ablk_decrypt,
diff --git a/arch/arm/include/asm/arch_gicv3.h b/arch/arm/include/asm/arch_gicv3.h
index 7da5503c0591..e08d15184056 100644
--- a/arch/arm/include/asm/arch_gicv3.h
+++ b/arch/arm/include/asm/arch_gicv3.h
@@ -117,6 +117,7 @@ static inline u32 gic_read_iar(void)
 	u32 irqstat;
 
 	asm volatile("mrc " __stringify(ICC_IAR1) : "=r" (irqstat));
+	dsb(sy);
 	return irqstat;
 }
 
diff --git a/arch/arm/include/asm/xen/page-coherent.h b/arch/arm/include/asm/xen/page-coherent.h
index 0375c8caa061..9408a994cc91 100644
--- a/arch/arm/include/asm/xen/page-coherent.h
+++ b/arch/arm/include/asm/xen/page-coherent.h
@@ -35,14 +35,21 @@ static inline void xen_dma_map_page(struct device *hwdev, struct page *page,
 	     dma_addr_t dev_addr, unsigned long offset, size_t size,
 	     enum dma_data_direction dir, struct dma_attrs *attrs)
 {
-	bool local = XEN_PFN_DOWN(dev_addr) == page_to_xen_pfn(page);
+	unsigned long page_pfn = page_to_xen_pfn(page);
+	unsigned long dev_pfn = XEN_PFN_DOWN(dev_addr);
+	unsigned long compound_pages =
+		(1<<compound_order(page)) * XEN_PFN_PER_PAGE;
+	bool local = (page_pfn <= dev_pfn) &&
+		(dev_pfn - page_pfn < compound_pages);
+
 	/*
-	 * Dom0 is mapped 1:1, while the Linux page can be spanned accross
-	 * multiple Xen page, it's not possible to have a mix of local and
-	 * foreign Xen page. So if the first xen_pfn == mfn the page is local
-	 * otherwise it's a foreign page grant-mapped in dom0. If the page is
-	 * local we can safely call the native dma_ops function, otherwise we
-	 * call the xen specific function.
+	 * Dom0 is mapped 1:1, while the Linux page can span across
+	 * multiple Xen pages, it's not possible for it to contain a
+	 * mix of local and foreign Xen pages. So if the first xen_pfn
+	 * == mfn the page is local otherwise it's a foreign page
+	 * grant-mapped in dom0. If the page is local we can safely
+	 * call the native dma_ops function, otherwise we call the xen
+	 * specific function.
 	 */
 	if (local)
 		__generic_dma_ops(hwdev)->map_page(hwdev, page, offset, size, dir, attrs);
diff --git a/arch/arm/kernel/Makefile b/arch/arm/kernel/Makefile
index 2c5f160be65e..ad325a8c7e1e 100644
--- a/arch/arm/kernel/Makefile
+++ b/arch/arm/kernel/Makefile
@@ -88,6 +88,7 @@ obj-$(CONFIG_DEBUG_LL)	+= debug.o
 obj-$(CONFIG_EARLY_PRINTK)	+= early_printk.o
 
 obj-$(CONFIG_ARM_VIRT_EXT)	+= hyp-stub.o
+AFLAGS_hyp-stub.o		:=-Wa,-march=armv7-a
 ifeq ($(CONFIG_ARM_PSCI),y)
 obj-$(CONFIG_SMP)		+= psci_smp.o
 endif
diff --git a/arch/arm/kvm/guest.c b/arch/arm/kvm/guest.c
index 5fa69d7bae58..99361f11354a 100644
--- a/arch/arm/kvm/guest.c
+++ b/arch/arm/kvm/guest.c
@@ -161,7 +161,7 @@ static int get_timer_reg(struct kvm_vcpu *vcpu, const struct kvm_one_reg *reg)
 	u64 val;
 
 	val = kvm_arm_timer_get_reg(vcpu, reg->id);
-	return copy_to_user(uaddr, &val, KVM_REG_SIZE(reg->id));
+	return copy_to_user(uaddr, &val, KVM_REG_SIZE(reg->id)) ? -EFAULT : 0;
 }
 
 static unsigned long num_core_regs(void)
diff --git a/arch/arm/kvm/mmio.c b/arch/arm/kvm/mmio.c
index 7f33b2056ae6..0f6600f05137 100644
--- a/arch/arm/kvm/mmio.c
+++ b/arch/arm/kvm/mmio.c
@@ -206,7 +206,8 @@ int io_mem_abort(struct kvm_vcpu *vcpu, struct kvm_run *run,
 	run->mmio.is_write	= is_write;
 	run->mmio.phys_addr	= fault_ipa;
 	run->mmio.len		= len;
-	memcpy(run->mmio.data, data_buf, len);
+	if (is_write)
+		memcpy(run->mmio.data, data_buf, len);
 
 	if (!ret) {
 		/* We handled the access successfully in the kernel. */
diff --git a/arch/arm/mach-omap2/board-generic.c b/arch/arm/mach-omap2/board-generic.c
index 809827265fb3..bab814d2f37d 100644
--- a/arch/arm/mach-omap2/board-generic.c
+++ b/arch/arm/mach-omap2/board-generic.c
@@ -18,6 +18,7 @@
 
 #include <asm/setup.h>
 #include <asm/mach/arch.h>
+#include <asm/system_info.h>
 
 #include "common.h"
 
@@ -77,12 +78,31 @@ static const char *const n900_boards_compat[] __initconst = {
 	NULL,
 };
 
+/* Set system_rev from atags */
+static void __init rx51_set_system_rev(const struct tag *tags)
+{
+	const struct tag *tag;
+
+	if (tags->hdr.tag != ATAG_CORE)
+		return;
+
+	for_each_tag(tag, tags) {
+		if (tag->hdr.tag == ATAG_REVISION) {
+			system_rev = tag->u.revision.rev;
+			break;
+		}
+	}
+}
+
 /* Legacy userspace on Nokia N900 needs ATAGS exported in /proc/atags,
  * save them while the data is still not overwritten
  */
 static void __init rx51_reserve(void)
 {
-	save_atags((const struct tag *)(PAGE_OFFSET + 0x100));
+	const struct tag *tags = (const struct tag *)(PAGE_OFFSET + 0x100);
+
+	save_atags(tags);
+	rx51_set_system_rev(tags);
 	omap_reserve();
 }
 
diff --git a/arch/arm/mach-omap2/gpmc-onenand.c b/arch/arm/mach-omap2/gpmc-onenand.c
index 7b76ce01c21d..8633c703546a 100644
--- a/arch/arm/mach-omap2/gpmc-onenand.c
+++ b/arch/arm/mach-omap2/gpmc-onenand.c
@@ -101,10 +101,8 @@ static void omap2_onenand_set_async_mode(void __iomem *onenand_base)
 
 static void set_onenand_cfg(void __iomem *onenand_base)
 {
-	u32 reg;
+	u32 reg = ONENAND_SYS_CFG1_RDY | ONENAND_SYS_CFG1_INT;
 
-	reg = readw(onenand_base + ONENAND_REG_SYS_CFG1);
-	reg &= ~((0x7 << ONENAND_SYS_CFG1_BRL_SHIFT) | (0x7 << 9));
 	reg |=	(latency << ONENAND_SYS_CFG1_BRL_SHIFT) |
 		ONENAND_SYS_CFG1_BL_16;
 	if (onenand_flags & ONENAND_FLAG_SYNCREAD)
@@ -123,6 +121,7 @@ static void set_onenand_cfg(void __iomem *onenand_base)
 		reg |= ONENAND_SYS_CFG1_VHF;
 	else
 		reg &= ~ONENAND_SYS_CFG1_VHF;
+
 	writew(reg, onenand_base + ONENAND_REG_SYS_CFG1);
 }
 
@@ -289,6 +288,7 @@ static int omap2_onenand_setup_async(void __iomem *onenand_base)
 		}
 	}
 
+	onenand_async.sync_write = true;
 	omap2_onenand_calc_async_timings(&t);
 
 	ret = gpmc_cs_program_settings(gpmc_onenand_data->cs, &onenand_async);
diff --git a/arch/arm/mach-omap2/omap_device.c b/arch/arm/mach-omap2/omap_device.c
index 0437537751bc..f7ff3b9dad87 100644
--- a/arch/arm/mach-omap2/omap_device.c
+++ b/arch/arm/mach-omap2/omap_device.c
@@ -191,12 +191,22 @@ static int _omap_device_notifier_call(struct notifier_block *nb,
 {
 	struct platform_device *pdev = to_platform_device(dev);
 	struct omap_device *od;
+	int err;
 
 	switch (event) {
 	case BUS_NOTIFY_DEL_DEVICE:
 		if (pdev->archdata.od)
 			omap_device_delete(pdev->archdata.od);
 		break;
+	case BUS_NOTIFY_UNBOUND_DRIVER:
+		od = to_omap_device(pdev);
+		if (od && (od->_state == OMAP_DEVICE_STATE_ENABLED)) {
+			dev_info(dev, "enabled after unload, idling\n");
+			err = omap_device_idle(pdev);
+			if (err)
+				dev_err(dev, "failed to idle\n");
+		}
+		break;
 	case BUS_NOTIFY_ADD_DEVICE:
 		if (pdev->dev.of_node)
 			omap_device_build_from_dt(pdev);
@@ -602,8 +612,10 @@ static int _od_runtime_resume(struct device *dev)
 	int ret;
 
 	ret = omap_device_enable(pdev);
-	if (ret)
+	if (ret) {
+		dev_err(dev, "use pm_runtime_put_sync_suspend() in driver?\n");
 		return ret;
+	}
 
 	return pm_generic_runtime_resume(dev);
 }
diff --git a/arch/arm/mach-omap2/omap_hwmod.c b/arch/arm/mach-omap2/omap_hwmod.c
index e9f65fec55c0..b6d62e4cdfdd 100644
--- a/arch/arm/mach-omap2/omap_hwmod.c
+++ b/arch/arm/mach-omap2/omap_hwmod.c
@@ -2200,6 +2200,11 @@ static int _enable(struct omap_hwmod *oh)
  */
 static int _idle(struct omap_hwmod *oh)
 {
+	if (oh->flags & HWMOD_NO_IDLE) {
+		oh->_int_flags |= _HWMOD_SKIP_ENABLE;
+		return 0;
+	}
+
 	pr_debug("omap_hwmod: %s: idling\n", oh->name);
 
 	if (oh->_state != _HWMOD_STATE_ENABLED) {
@@ -2504,6 +2509,8 @@ static int __init _init(struct omap_hwmod *oh, void *data)
 			oh->flags |= HWMOD_INIT_NO_RESET;
 		if (of_find_property(np, "ti,no-idle-on-init", NULL))
 			oh->flags |= HWMOD_INIT_NO_IDLE;
+		if (of_find_property(np, "ti,no-idle", NULL))
+			oh->flags |= HWMOD_NO_IDLE;
 	}
 
 	oh->_state = _HWMOD_STATE_INITIALIZED;
@@ -2630,7 +2637,7 @@ static void __init _setup_postsetup(struct omap_hwmod *oh)
 	 * XXX HWMOD_INIT_NO_IDLE does not belong in hwmod data -
 	 * it should be set by the core code as a runtime flag during startup
 	 */
-	if ((oh->flags & HWMOD_INIT_NO_IDLE) &&
+	if ((oh->flags & (HWMOD_INIT_NO_IDLE | HWMOD_NO_IDLE)) &&
 	    (postsetup_state == _HWMOD_STATE_IDLE)) {
 		oh->_int_flags |= _HWMOD_SKIP_ENABLE;
 		postsetup_state = _HWMOD_STATE_ENABLED;
diff --git a/arch/arm/mach-omap2/omap_hwmod.h b/arch/arm/mach-omap2/omap_hwmod.h
index 76bce11c85a4..7c7a31169475 100644
--- a/arch/arm/mach-omap2/omap_hwmod.h
+++ b/arch/arm/mach-omap2/omap_hwmod.h
@@ -525,6 +525,8 @@ struct omap_hwmod_omap4_prcm {
  *     or idled.
  * HWMOD_OPT_CLKS_NEEDED: The optional clocks are needed for the module to
  *     operate and they need to be handled at the same time as the main_clk.
+ * HWMOD_NO_IDLE: Do not idle the hwmod at all. Useful to handle certain
+ *     IPs like CPSW on DRA7, where clocks to this module cannot be disabled.
  */
 #define HWMOD_SWSUP_SIDLE			(1 << 0)
 #define HWMOD_SWSUP_MSTANDBY			(1 << 1)
@@ -541,6 +543,7 @@ struct omap_hwmod_omap4_prcm {
 #define HWMOD_SWSUP_SIDLE_ACT			(1 << 12)
 #define HWMOD_RECONFIG_IO_CHAIN			(1 << 13)
 #define HWMOD_OPT_CLKS_NEEDED			(1 << 14)
+#define HWMOD_NO_IDLE				(1 << 15)
 
 /*
  * omap_hwmod._int_flags definitions
diff --git a/arch/arm/mach-shmobile/common.h b/arch/arm/mach-shmobile/common.h
index 9cb11215ceba..b3a4ed5289ec 100644
--- a/arch/arm/mach-shmobile/common.h
+++ b/arch/arm/mach-shmobile/common.h
@@ -4,7 +4,6 @@
 extern void shmobile_init_delay(void);
 extern void shmobile_boot_vector(void);
 extern unsigned long shmobile_boot_fn;
-extern unsigned long shmobile_boot_arg;
 extern unsigned long shmobile_boot_size;
 extern void shmobile_smp_boot(void);
 extern void shmobile_smp_sleep(void);
diff --git a/arch/arm/mach-shmobile/headsmp-scu.S b/arch/arm/mach-shmobile/headsmp-scu.S
index fa5248c52399..5e503d91ad70 100644
--- a/arch/arm/mach-shmobile/headsmp-scu.S
+++ b/arch/arm/mach-shmobile/headsmp-scu.S
@@ -38,9 +38,3 @@ ENTRY(shmobile_boot_scu)
 
 	b	secondary_startup
 ENDPROC(shmobile_boot_scu)
-
-	.text
-	.align	2
-	.globl	shmobile_scu_base
-shmobile_scu_base:
-	.space	4
diff --git a/arch/arm/mach-shmobile/headsmp.S b/arch/arm/mach-shmobile/headsmp.S
index 330c1fc63197..32e0bf6e3ccb 100644
--- a/arch/arm/mach-shmobile/headsmp.S
+++ b/arch/arm/mach-shmobile/headsmp.S
@@ -24,7 +24,6 @@
 	.arm
 	.align  12
 ENTRY(shmobile_boot_vector)
-	ldr     r0, 2f
 	ldr     r1, 1f
 	bx	r1
 
@@ -34,9 +33,6 @@ ENDPROC(shmobile_boot_vector)
 	.globl	shmobile_boot_fn
 shmobile_boot_fn:
 1:	.space	4
-	.globl	shmobile_boot_arg
-shmobile_boot_arg:
-2:	.space	4
 	.globl	shmobile_boot_size
 shmobile_boot_size:
 	.long	. - shmobile_boot_vector
@@ -46,13 +42,15 @@ shmobile_boot_size:
  */
 
 ENTRY(shmobile_smp_boot)
-						@ r0 = MPIDR_HWID_BITMASK
 	mrc	p15, 0, r1, c0, c0, 5		@ r1 = MPIDR
-	and	r0, r1, r0			@ r0 = cpu_logical_map() value
+	and	r0, r1, #0xffffff		@ MPIDR_HWID_BITMASK
+						@ r0 = cpu_logical_map() value
 	mov	r1, #0				@ r1 = CPU index
-	adr	r5, 1f				@ array of per-cpu mpidr values
-	adr	r6, 2f				@ array of per-cpu functions
-	adr	r7, 3f				@ array of per-cpu arguments
+	adr	r2, 1f
+	ldmia	r2, {r5, r6, r7}
+	add	r5, r5, r2			@ array of per-cpu mpidr values
+	add	r6, r6, r2			@ array of per-cpu functions
+	add	r7, r7, r2			@ array of per-cpu arguments
 
 shmobile_smp_boot_find_mpidr:
 	ldr	r8, [r5, r1, lsl #2]
@@ -80,12 +78,18 @@ ENTRY(shmobile_smp_sleep)
 	b	shmobile_smp_boot
 ENDPROC(shmobile_smp_sleep)
 
+	.align	2
+1:	.long	shmobile_smp_mpidr - .
+	.long	shmobile_smp_fn - 1b
+	.long	shmobile_smp_arg - 1b
+
+	.bss
 	.globl	shmobile_smp_mpidr
 shmobile_smp_mpidr:
-1:	.space	NR_CPUS * 4
+	.space	NR_CPUS * 4
 	.globl	shmobile_smp_fn
 shmobile_smp_fn:
-2:	.space	NR_CPUS * 4
+	.space	NR_CPUS * 4
 	.globl	shmobile_smp_arg
 shmobile_smp_arg:
-3:	.space	NR_CPUS * 4
+	.space	NR_CPUS * 4
diff --git a/arch/arm/mach-shmobile/platsmp-apmu.c b/arch/arm/mach-shmobile/platsmp-apmu.c
index 911884f7e28b..aba75c89f9c1 100644
--- a/arch/arm/mach-shmobile/platsmp-apmu.c
+++ b/arch/arm/mach-shmobile/platsmp-apmu.c
@@ -123,7 +123,6 @@ void __init shmobile_smp_apmu_prepare_cpus(unsigned int max_cpus,
 {
 	/* install boot code shared by all CPUs */
 	shmobile_boot_fn = virt_to_phys(shmobile_smp_boot);
-	shmobile_boot_arg = MPIDR_HWID_BITMASK;
 
 	/* perform per-cpu setup */
 	apmu_parse_cfg(apmu_init_cpu, apmu_config, num);
diff --git a/arch/arm/mach-shmobile/platsmp-scu.c b/arch/arm/mach-shmobile/platsmp-scu.c
index 64663110ab6c..081a097c9219 100644
--- a/arch/arm/mach-shmobile/platsmp-scu.c
+++ b/arch/arm/mach-shmobile/platsmp-scu.c
@@ -17,6 +17,9 @@
 #include <asm/smp_scu.h>
 #include "common.h"
 
+
+void __iomem *shmobile_scu_base;
+
 static int shmobile_smp_scu_notifier_call(struct notifier_block *nfb,
 					  unsigned long action, void *hcpu)
 {
@@ -41,7 +44,6 @@ void __init shmobile_smp_scu_prepare_cpus(unsigned int max_cpus)
 {
 	/* install boot code shared by all CPUs */
 	shmobile_boot_fn = virt_to_phys(shmobile_smp_boot);
-	shmobile_boot_arg = MPIDR_HWID_BITMASK;
 
 	/* enable SCU and cache coherency on booting CPU */
 	scu_enable(shmobile_scu_base);
diff --git a/arch/arm/mach-shmobile/smp-r8a7779.c b/arch/arm/mach-shmobile/smp-r8a7779.c
index b854fe2095ad..0b024a9dbd43 100644
--- a/arch/arm/mach-shmobile/smp-r8a7779.c
+++ b/arch/arm/mach-shmobile/smp-r8a7779.c
@@ -92,8 +92,6 @@ static void __init r8a7779_smp_prepare_cpus(unsigned int max_cpus)
 {
 	/* Map the reset vector (in headsmp-scu.S, headsmp.S) */
 	__raw_writel(__pa(shmobile_boot_vector), AVECR);
-	shmobile_boot_fn = virt_to_phys(shmobile_boot_scu);
-	shmobile_boot_arg = (unsigned long)shmobile_scu_base;
 
 	/* setup r8a7779 specific SCU bits */
 	shmobile_scu_base = IOMEM(R8A7779_SCU_BASE);
diff --git a/arch/arm/mm/mmap.c b/arch/arm/mm/mmap.c
index 4b4058db0781..66353caa35b9 100644
--- a/arch/arm/mm/mmap.c
+++ b/arch/arm/mm/mmap.c
@@ -173,7 +173,7 @@ unsigned long arch_mmap_rnd(void)
 {
 	unsigned long rnd;
 
-	rnd = (unsigned long)get_random_int() & ((1 << mmap_rnd_bits) - 1);
+	rnd = get_random_long() & ((1UL << mmap_rnd_bits) - 1);
 
 	return rnd << PAGE_SHIFT;
 }
diff --git a/arch/arm/mm/pageattr.c b/arch/arm/mm/pageattr.c
index cf30daff8932..d19b1ad29b07 100644
--- a/arch/arm/mm/pageattr.c
+++ b/arch/arm/mm/pageattr.c
@@ -49,6 +49,9 @@ static int change_memory_common(unsigned long addr, int numpages,
 		WARN_ON_ONCE(1);
 	}
 
+	if (!numpages)
+		return 0;
+
 	if (start < MODULES_VADDR || start >= MODULES_END)
 		return -EINVAL;
 
diff --git a/arch/arm64/Makefile b/arch/arm64/Makefile
index 307237cfe728..b5e3f6d42b88 100644
--- a/arch/arm64/Makefile
+++ b/arch/arm64/Makefile
@@ -88,7 +88,7 @@ Image: vmlinux
 Image.%: vmlinux
 	$(Q)$(MAKE) $(build)=$(boot) $(boot)/$@
 
-zinstall install: vmlinux
+zinstall install:
 	$(Q)$(MAKE) $(build)=$(boot) $@
 
 %.dtb: scripts
diff --git a/arch/arm64/boot/Makefile b/arch/arm64/boot/Makefile
index abcbba2f01ba..305c552b5ec1 100644
--- a/arch/arm64/boot/Makefile
+++ b/arch/arm64/boot/Makefile
@@ -34,10 +34,10 @@ $(obj)/Image.lzma: $(obj)/Image FORCE
 $(obj)/Image.lzo: $(obj)/Image FORCE
 	$(call if_changed,lzo)
 
-install: $(obj)/Image
+install:
 	$(CONFIG_SHELL) $(srctree)/$(src)/install.sh $(KERNELRELEASE) \
 	$(obj)/Image System.map "$(INSTALL_PATH)"
 
-zinstall: $(obj)/Image.gz
+zinstall:
 	$(CONFIG_SHELL) $(srctree)/$(src)/install.sh $(KERNELRELEASE) \
 	$(obj)/Image.gz System.map "$(INSTALL_PATH)"
diff --git a/arch/arm64/boot/install.sh b/arch/arm64/boot/install.sh
index 12ed78aa6f0c..d91e1f022573 100644
--- a/arch/arm64/boot/install.sh
+++ b/arch/arm64/boot/install.sh
@@ -20,6 +20,20 @@
 #   $4 - default install path (blank if root directory)
 #
 
+verify () {
+	if [ ! -f "$1" ]; then
+		echo ""                                                   1>&2
+		echo " *** Missing file: $1"                              1>&2
+		echo ' *** You need to run "make" before "make install".' 1>&2
+		echo ""                                                   1>&2
+		exit 1
+	fi
+}
+
+# Make sure the files actually exist
+verify "$2"
+verify "$3"
+
 # User may have a custom install script
 if [ -x ~/bin/${INSTALLKERNEL} ]; then exec ~/bin/${INSTALLKERNEL} "$@"; fi
 if [ -x /sbin/${INSTALLKERNEL} ]; then exec /sbin/${INSTALLKERNEL} "$@"; fi
diff --git a/arch/arm64/crypto/aes-glue.c b/arch/arm64/crypto/aes-glue.c
index 05d9e16c0dfd..7a3d22a46faf 100644
--- a/arch/arm64/crypto/aes-glue.c
+++ b/arch/arm64/crypto/aes-glue.c
@@ -294,7 +294,7 @@ static struct crypto_alg aes_algs[] = { {
 	.cra_blkcipher = {
 		.min_keysize	= AES_MIN_KEY_SIZE,
 		.max_keysize	= AES_MAX_KEY_SIZE,
-		.ivsize		= AES_BLOCK_SIZE,
+		.ivsize		= 0,
 		.setkey		= aes_setkey,
 		.encrypt	= ecb_encrypt,
 		.decrypt	= ecb_decrypt,
@@ -371,7 +371,7 @@ static struct crypto_alg aes_algs[] = { {
 	.cra_ablkcipher = {
 		.min_keysize	= AES_MIN_KEY_SIZE,
 		.max_keysize	= AES_MAX_KEY_SIZE,
-		.ivsize		= AES_BLOCK_SIZE,
+		.ivsize		= 0,
 		.setkey		= ablk_set_key,
 		.encrypt	= ablk_encrypt,
 		.decrypt	= ablk_decrypt,
diff --git a/arch/arm64/include/asm/arch_gicv3.h b/arch/arm64/include/asm/arch_gicv3.h
index 2731d3b25ed2..8ec88e5b290f 100644
--- a/arch/arm64/include/asm/arch_gicv3.h
+++ b/arch/arm64/include/asm/arch_gicv3.h
@@ -103,6 +103,7 @@ static inline u64 gic_read_iar_common(void)
 	u64 irqstat;
 
 	asm volatile("mrs_s %0, " __stringify(ICC_IAR1_EL1) : "=r" (irqstat));
+	dsb(sy);
 	return irqstat;
 }
 
diff --git a/arch/arm64/include/asm/kvm_arm.h b/arch/arm64/include/asm/kvm_arm.h
index 738a95f93e49..d201d4b396d1 100644
--- a/arch/arm64/include/asm/kvm_arm.h
+++ b/arch/arm64/include/asm/kvm_arm.h
@@ -107,8 +107,6 @@
 #define TCR_EL2_MASK	(TCR_EL2_TG0 | TCR_EL2_SH0 | \
 			 TCR_EL2_ORGN0 | TCR_EL2_IRGN0 | TCR_EL2_T0SZ)
 
-#define TCR_EL2_FLAGS	(TCR_EL2_RES1 | TCR_EL2_PS_40B)
-
 /* VTCR_EL2 Registers bits */
 #define VTCR_EL2_RES1		(1 << 31)
 #define VTCR_EL2_PS_MASK	(7 << 16)
@@ -182,6 +180,7 @@
 #define CPTR_EL2_TCPAC	(1 << 31)
 #define CPTR_EL2_TTA	(1 << 20)
 #define CPTR_EL2_TFP	(1 << CPTR_EL2_TFP_SHIFT)
+#define CPTR_EL2_DEFAULT	0x000033ff
 
 /* Hyp Debug Configuration Register bits */
 #define MDCR_EL2_TDRA		(1 << 11)
diff --git a/arch/arm64/include/asm/kvm_emulate.h b/arch/arm64/include/asm/kvm_emulate.h
index 3066328cd86b..779a5872a2c5 100644
--- a/arch/arm64/include/asm/kvm_emulate.h
+++ b/arch/arm64/include/asm/kvm_emulate.h
@@ -127,10 +127,14 @@ static inline unsigned long *vcpu_spsr(const struct kvm_vcpu *vcpu)
 
 static inline bool vcpu_mode_priv(const struct kvm_vcpu *vcpu)
 {
-	u32 mode = *vcpu_cpsr(vcpu) & PSR_MODE_MASK;
+	u32 mode;
 
-	if (vcpu_mode_is_32bit(vcpu))
+	if (vcpu_mode_is_32bit(vcpu)) {
+		mode = *vcpu_cpsr(vcpu) & COMPAT_PSR_MODE_MASK;
 		return mode > COMPAT_PSR_MODE_USR;
+	}
+
+	mode = *vcpu_cpsr(vcpu) & PSR_MODE_MASK;
 
 	return mode != PSR_MODE_EL0t;
 }
diff --git a/arch/arm64/include/asm/pgtable.h b/arch/arm64/include/asm/pgtable.h
index bf464de33f52..819aff5d593f 100644
--- a/arch/arm64/include/asm/pgtable.h
+++ b/arch/arm64/include/asm/pgtable.h
@@ -34,7 +34,7 @@
 /*
  * VMALLOC and SPARSEMEM_VMEMMAP ranges.
  *
- * VMEMAP_SIZE: allows the whole VA space to be covered by a struct page array
+ * VMEMAP_SIZE: allows the whole linear region to be covered by a struct page array
  *	(rounded up to PUD_SIZE).
  * VMALLOC_START: beginning of the kernel VA space
  * VMALLOC_END: extends to the available space below vmmemmap, PCI I/O space,
@@ -51,7 +51,9 @@
 
 #define VMALLOC_END		(PAGE_OFFSET - PUD_SIZE - VMEMMAP_SIZE - SZ_64K)
 
-#define vmemmap			((struct page *)(VMALLOC_END + SZ_64K))
+#define VMEMMAP_START		(VMALLOC_END + SZ_64K)
+#define vmemmap			((struct page *)VMEMMAP_START - \
+				 SECTION_ALIGN_DOWN(memstart_addr >> PAGE_SHIFT))
 
 #define FIRST_USER_ADDRESS	0UL
 
diff --git a/arch/arm64/kernel/debug-monitors.c b/arch/arm64/kernel/debug-monitors.c
index 8aee3aeec3e6..c536c9e307b9 100644
--- a/arch/arm64/kernel/debug-monitors.c
+++ b/arch/arm64/kernel/debug-monitors.c
@@ -226,11 +226,28 @@ static int call_step_hook(struct pt_regs *regs, unsigned int esr)
 	return retval;
 }
 
+static void send_user_sigtrap(int si_code)
+{
+	struct pt_regs *regs = current_pt_regs();
+	siginfo_t info = {
+		.si_signo	= SIGTRAP,
+		.si_errno	= 0,
+		.si_code	= si_code,
+		.si_addr	= (void __user *)instruction_pointer(regs),
+	};
+
+	if (WARN_ON(!user_mode(regs)))
+		return;
+
+	if (interrupts_enabled(regs))
+		local_irq_enable();
+
+	force_sig_info(SIGTRAP, &info, current);
+}
+
 static int single_step_handler(unsigned long addr, unsigned int esr,
 			       struct pt_regs *regs)
 {
-	siginfo_t info;
-
 	/*
 	 * If we are stepping a pending breakpoint, call the hw_breakpoint
 	 * handler first.
@@ -239,11 +256,7 @@ static int single_step_handler(unsigned long addr, unsigned int esr,
 		return 0;
 
 	if (user_mode(regs)) {
-		info.si_signo = SIGTRAP;
-		info.si_errno = 0;
-		info.si_code  = TRAP_HWBKPT;
-		info.si_addr  = (void __user *)instruction_pointer(regs);
-		force_sig_info(SIGTRAP, &info, current);
+		send_user_sigtrap(TRAP_HWBKPT);
 
 		/*
 		 * ptrace will disable single step unless explicitly
@@ -307,17 +320,8 @@ static int call_break_hook(struct pt_regs *regs, unsigned int esr)
 static int brk_handler(unsigned long addr, unsigned int esr,
 		       struct pt_regs *regs)
 {
-	siginfo_t info;
-
 	if (user_mode(regs)) {
-		info = (siginfo_t) {
-			.si_signo = SIGTRAP,
-			.si_errno = 0,
-			.si_code  = TRAP_BRKPT,
-			.si_addr  = (void __user *)instruction_pointer(regs),
-		};
-
-		force_sig_info(SIGTRAP, &info, current);
+		send_user_sigtrap(TRAP_BRKPT);
 	} else if (call_break_hook(regs, esr) != DBG_HOOK_HANDLED) {
 		pr_warning("Unexpected kernel BRK exception at EL1\n");
 		return -EFAULT;
@@ -328,7 +332,6 @@ static int brk_handler(unsigned long addr, unsigned int esr,
 
 int aarch32_break_handler(struct pt_regs *regs)
 {
-	siginfo_t info;
 	u32 arm_instr;
 	u16 thumb_instr;
 	bool bp = false;
@@ -359,14 +362,7 @@ int aarch32_break_handler(struct pt_regs *regs)
 	if (!bp)
 		return -EFAULT;
 
-	info = (siginfo_t) {
-		.si_signo = SIGTRAP,
-		.si_errno = 0,
-		.si_code  = TRAP_BRKPT,
-		.si_addr  = pc,
-	};
-
-	force_sig_info(SIGTRAP, &info, current);
+	send_user_sigtrap(TRAP_BRKPT);
 	return 0;
 }
 
diff --git a/arch/arm64/kernel/image.h b/arch/arm64/kernel/image.h
index 999633bd7294..352f7abd91c9 100644
--- a/arch/arm64/kernel/image.h
+++ b/arch/arm64/kernel/image.h
@@ -89,6 +89,7 @@ __efistub_memcpy		= KALLSYMS_HIDE(__pi_memcpy);
 __efistub_memmove		= KALLSYMS_HIDE(__pi_memmove);
 __efistub_memset		= KALLSYMS_HIDE(__pi_memset);
 __efistub_strlen		= KALLSYMS_HIDE(__pi_strlen);
+__efistub_strnlen		= KALLSYMS_HIDE(__pi_strnlen);
 __efistub_strcmp		= KALLSYMS_HIDE(__pi_strcmp);
 __efistub_strncmp		= KALLSYMS_HIDE(__pi_strncmp);
 __efistub___flush_dcache_area	= KALLSYMS_HIDE(__pi___flush_dcache_area);
diff --git a/arch/arm64/kernel/sleep.S b/arch/arm64/kernel/sleep.S
index e33fe33876ab..fd10eb663868 100644
--- a/arch/arm64/kernel/sleep.S
+++ b/arch/arm64/kernel/sleep.S
@@ -145,6 +145,10 @@ ENTRY(cpu_resume_mmu)
 ENDPROC(cpu_resume_mmu)
 	.popsection
 cpu_resume_after_mmu:
+#ifdef CONFIG_KASAN
+	mov	x0, sp
+	bl	kasan_unpoison_remaining_stack
+#endif
 	mov	x0, #0			// return zero on success
 	ldp	x19, x20, [sp, #16]
 	ldp	x21, x22, [sp, #32]
diff --git a/arch/arm64/kernel/stacktrace.c b/arch/arm64/kernel/stacktrace.c
index 4fad9787ab46..d9751a4769e7 100644
--- a/arch/arm64/kernel/stacktrace.c
+++ b/arch/arm64/kernel/stacktrace.c
@@ -44,14 +44,13 @@ int notrace unwind_frame(struct task_struct *tsk, struct stackframe *frame)
 	unsigned long irq_stack_ptr;
 
 	/*
-	 * Use raw_smp_processor_id() to avoid false-positives from
-	 * CONFIG_DEBUG_PREEMPT. get_wchan() calls unwind_frame() on sleeping
-	 * task stacks, we can be pre-empted in this case, so
-	 * {raw_,}smp_processor_id() may give us the wrong value. Sleeping
-	 * tasks can't ever be on an interrupt stack, so regardless of cpu,
-	 * the checks will always fail.
+	 * Switching between stacks is valid when tracing current and in
+	 * non-preemptible context.
 	 */
-	irq_stack_ptr = IRQ_STACK_PTR(raw_smp_processor_id());
+	if (tsk == current && !preemptible())
+		irq_stack_ptr = IRQ_STACK_PTR(smp_processor_id());
+	else
+		irq_stack_ptr = 0;
 
 	low  = frame->sp;
 	/* irq stacks are not THREAD_SIZE aligned */
@@ -64,8 +63,8 @@ int notrace unwind_frame(struct task_struct *tsk, struct stackframe *frame)
 		return -EINVAL;
 
 	frame->sp = fp + 0x10;
-	frame->fp = *(unsigned long *)(fp);
-	frame->pc = *(unsigned long *)(fp + 8);
+	frame->fp = READ_ONCE_NOCHECK(*(unsigned long *)(fp));
+	frame->pc = READ_ONCE_NOCHECK(*(unsigned long *)(fp + 8));
 
 #ifdef CONFIG_FUNCTION_GRAPH_TRACER
 	if (tsk && tsk->ret_stack &&
diff --git a/arch/arm64/kernel/traps.c b/arch/arm64/kernel/traps.c
index cbedd724f48e..c5392081b49b 100644
--- a/arch/arm64/kernel/traps.c
+++ b/arch/arm64/kernel/traps.c
@@ -146,9 +146,18 @@ static void dump_instr(const char *lvl, struct pt_regs *regs)
 static void dump_backtrace(struct pt_regs *regs, struct task_struct *tsk)
 {
 	struct stackframe frame;
-	unsigned long irq_stack_ptr = IRQ_STACK_PTR(smp_processor_id());
+	unsigned long irq_stack_ptr;
 	int skip;
 
+	/*
+	 * Switching between stacks is valid when tracing current and in
+	 * non-preemptible context.
+	 */
+	if (tsk == current && !preemptible())
+		irq_stack_ptr = IRQ_STACK_PTR(smp_processor_id());
+	else
+		irq_stack_ptr = 0;
+
 	pr_debug("%s(regs = %p tsk = %p)\n", __func__, regs, tsk);
 
 	if (!tsk)
diff --git a/arch/arm64/kvm/guest.c b/arch/arm64/kvm/guest.c
index fcb778899a38..9e54ad7c240a 100644
--- a/arch/arm64/kvm/guest.c
+++ b/arch/arm64/kvm/guest.c
@@ -194,7 +194,7 @@ static int get_timer_reg(struct kvm_vcpu *vcpu, const struct kvm_one_reg *reg)
 	u64 val;
 
 	val = kvm_arm_timer_get_reg(vcpu, reg->id);
-	return copy_to_user(uaddr, &val, KVM_REG_SIZE(reg->id));
+	return copy_to_user(uaddr, &val, KVM_REG_SIZE(reg->id)) ? -EFAULT : 0;
 }
 
 /**
diff --git a/arch/arm64/kvm/hyp-init.S b/arch/arm64/kvm/hyp-init.S
index 3e568dcd907b..d073b5a216f7 100644
--- a/arch/arm64/kvm/hyp-init.S
+++ b/arch/arm64/kvm/hyp-init.S
@@ -64,7 +64,7 @@ __do_hyp_init:
 	mrs	x4, tcr_el1
 	ldr	x5, =TCR_EL2_MASK
 	and	x4, x4, x5
-	ldr	x5, =TCR_EL2_FLAGS
+	mov	x5, #TCR_EL2_RES1
 	orr	x4, x4, x5
 
 #ifndef CONFIG_ARM64_VA_BITS_48
@@ -85,15 +85,17 @@ __do_hyp_init:
 	ldr_l	x5, idmap_t0sz
 	bfi	x4, x5, TCR_T0SZ_OFFSET, TCR_TxSZ_WIDTH
 #endif
-	msr	tcr_el2, x4
-
-	ldr	x4, =VTCR_EL2_FLAGS
 	/*
 	 * Read the PARange bits from ID_AA64MMFR0_EL1 and set the PS bits in
-	 * VTCR_EL2.
+	 * TCR_EL2 and VTCR_EL2.
 	 */
 	mrs	x5, ID_AA64MMFR0_EL1
 	bfi	x4, x5, #16, #3
+
+	msr	tcr_el2, x4
+
+	ldr	x4, =VTCR_EL2_FLAGS
+	bfi	x4, x5, #16, #3
 	/*
 	 * Read the VMIDBits bits from ID_AA64MMFR1_EL1 and set the VS bit in
 	 * VTCR_EL2.
diff --git a/arch/arm64/kvm/hyp/switch.c b/arch/arm64/kvm/hyp/switch.c
index ca8f5a5e2f96..f0e7bdfae134 100644
--- a/arch/arm64/kvm/hyp/switch.c
+++ b/arch/arm64/kvm/hyp/switch.c
@@ -36,7 +36,11 @@ static void __hyp_text __activate_traps(struct kvm_vcpu *vcpu)
 	write_sysreg(val, hcr_el2);
 	/* Trap on AArch32 cp15 c15 accesses (EL1 or EL0) */
 	write_sysreg(1 << 15, hstr_el2);
-	write_sysreg(CPTR_EL2_TTA | CPTR_EL2_TFP, cptr_el2);
+
+	val = CPTR_EL2_DEFAULT;
+	val |= CPTR_EL2_TTA | CPTR_EL2_TFP;
+	write_sysreg(val, cptr_el2);
+
 	write_sysreg(vcpu->arch.mdcr_el2, mdcr_el2);
 }
 
@@ -45,7 +49,7 @@ static void __hyp_text __deactivate_traps(struct kvm_vcpu *vcpu)
 	write_sysreg(HCR_RW, hcr_el2);
 	write_sysreg(0, hstr_el2);
 	write_sysreg(read_sysreg(mdcr_el2) & MDCR_EL2_HPMN_MASK, mdcr_el2);
-	write_sysreg(0, cptr_el2);
+	write_sysreg(CPTR_EL2_DEFAULT, cptr_el2);
 }
 
 static void __hyp_text __activate_vm(struct kvm_vcpu *vcpu)
diff --git a/arch/arm64/kvm/hyp/vgic-v3-sr.c b/arch/arm64/kvm/hyp/vgic-v3-sr.c
index 9142e082f5f3..5dd2a26444ec 100644
--- a/arch/arm64/kvm/hyp/vgic-v3-sr.c
+++ b/arch/arm64/kvm/hyp/vgic-v3-sr.c
@@ -149,16 +149,6 @@ void __hyp_text __vgic_v3_restore_state(struct kvm_vcpu *vcpu)
 
 	switch (nr_pri_bits) {
 	case 7:
-		 write_gicreg(cpu_if->vgic_ap1r[3], ICH_AP1R3_EL2);
-		 write_gicreg(cpu_if->vgic_ap1r[2], ICH_AP1R2_EL2);
-	case 6:
-		 write_gicreg(cpu_if->vgic_ap1r[1], ICH_AP1R1_EL2);
-	default:
-		 write_gicreg(cpu_if->vgic_ap1r[0], ICH_AP1R0_EL2);
-	}	 	                           
-		 	                           
-	switch (nr_pri_bits) {
-	case 7:
 		 write_gicreg(cpu_if->vgic_ap0r[3], ICH_AP0R3_EL2);
 		 write_gicreg(cpu_if->vgic_ap0r[2], ICH_AP0R2_EL2);
 	case 6:
@@ -167,6 +157,16 @@ void __hyp_text __vgic_v3_restore_state(struct kvm_vcpu *vcpu)
 		 write_gicreg(cpu_if->vgic_ap0r[0], ICH_AP0R0_EL2);
 	}
 
+	switch (nr_pri_bits) {
+	case 7:
+		 write_gicreg(cpu_if->vgic_ap1r[3], ICH_AP1R3_EL2);
+		 write_gicreg(cpu_if->vgic_ap1r[2], ICH_AP1R2_EL2);
+	case 6:
+		 write_gicreg(cpu_if->vgic_ap1r[1], ICH_AP1R1_EL2);
+	default:
+		 write_gicreg(cpu_if->vgic_ap1r[0], ICH_AP1R0_EL2);
+	}
+
 	switch (max_lr_idx) {
 	case 15:
 		write_gicreg(cpu_if->vgic_lr[VGIC_V3_LR_INDEX(15)], ICH_LR15_EL2);
diff --git a/arch/arm64/kvm/inject_fault.c b/arch/arm64/kvm/inject_fault.c
index 648112e90ed5..4d1ac81870d2 100644
--- a/arch/arm64/kvm/inject_fault.c
+++ b/arch/arm64/kvm/inject_fault.c
@@ -27,7 +27,11 @@
 
 #define PSTATE_FAULT_BITS_64 	(PSR_MODE_EL1h | PSR_A_BIT | PSR_F_BIT | \
 				 PSR_I_BIT | PSR_D_BIT)
-#define EL1_EXCEPT_SYNC_OFFSET	0x200
+
+#define CURRENT_EL_SP_EL0_VECTOR	0x0
+#define CURRENT_EL_SP_ELx_VECTOR	0x200
+#define LOWER_EL_AArch64_VECTOR		0x400
+#define LOWER_EL_AArch32_VECTOR		0x600
 
 static void prepare_fault32(struct kvm_vcpu *vcpu, u32 mode, u32 vect_offset)
 {
@@ -97,6 +101,34 @@ static void inject_abt32(struct kvm_vcpu *vcpu, bool is_pabt,
 		*fsr = 0x14;
 }
 
+enum exception_type {
+	except_type_sync	= 0,
+	except_type_irq		= 0x80,
+	except_type_fiq		= 0x100,
+	except_type_serror	= 0x180,
+};
+
+static u64 get_except_vector(struct kvm_vcpu *vcpu, enum exception_type type)
+{
+	u64 exc_offset;
+
+	switch (*vcpu_cpsr(vcpu) & (PSR_MODE_MASK | PSR_MODE32_BIT)) {
+	case PSR_MODE_EL1t:
+		exc_offset = CURRENT_EL_SP_EL0_VECTOR;
+		break;
+	case PSR_MODE_EL1h:
+		exc_offset = CURRENT_EL_SP_ELx_VECTOR;
+		break;
+	case PSR_MODE_EL0t:
+		exc_offset = LOWER_EL_AArch64_VECTOR;
+		break;
+	default:
+		exc_offset = LOWER_EL_AArch32_VECTOR;
+	}
+
+	return vcpu_sys_reg(vcpu, VBAR_EL1) + exc_offset + type;
+}
+
 static void inject_abt64(struct kvm_vcpu *vcpu, bool is_iabt, unsigned long addr)
 {
 	unsigned long cpsr = *vcpu_cpsr(vcpu);
@@ -108,8 +140,8 @@ static void inject_abt64(struct kvm_vcpu *vcpu, bool is_iabt, unsigned long addr
 	*vcpu_spsr(vcpu) = cpsr;
 	*vcpu_elr_el1(vcpu) = *vcpu_pc(vcpu);
 
+	*vcpu_pc(vcpu) = get_except_vector(vcpu, except_type_sync);
 	*vcpu_cpsr(vcpu) = PSTATE_FAULT_BITS_64;
-	*vcpu_pc(vcpu) = vcpu_sys_reg(vcpu, VBAR_EL1) + EL1_EXCEPT_SYNC_OFFSET;
 
 	vcpu_sys_reg(vcpu, FAR_EL1) = addr;
 
@@ -143,8 +175,8 @@ static void inject_undef64(struct kvm_vcpu *vcpu)
 	*vcpu_spsr(vcpu) = cpsr;
 	*vcpu_elr_el1(vcpu) = *vcpu_pc(vcpu);
 
+	*vcpu_pc(vcpu) = get_except_vector(vcpu, except_type_sync);
 	*vcpu_cpsr(vcpu) = PSTATE_FAULT_BITS_64;
-	*vcpu_pc(vcpu) = vcpu_sys_reg(vcpu, VBAR_EL1) + EL1_EXCEPT_SYNC_OFFSET;
 
 	/*
 	 * Build an unknown exception, depending on the instruction
diff --git a/arch/arm64/kvm/sys_regs.c b/arch/arm64/kvm/sys_regs.c
index eec3598b4184..2e90371cfb37 100644
--- a/arch/arm64/kvm/sys_regs.c
+++ b/arch/arm64/kvm/sys_regs.c
@@ -1007,10 +1007,9 @@ static int emulate_cp(struct kvm_vcpu *vcpu,
 		if (likely(r->access(vcpu, params, r))) {
 			/* Skip instruction, since it was emulated */
 			kvm_skip_instr(vcpu, kvm_vcpu_trap_il_is32bit(vcpu));
+			/* Handled */
+			return 0;
 		}
-
-		/* Handled */
-		return 0;
 	}
 
 	/* Not handled */
@@ -1043,7 +1042,7 @@ static void unhandled_cp_access(struct kvm_vcpu *vcpu,
 }
 
 /**
- * kvm_handle_cp_64 -- handles a mrrc/mcrr trap on a guest CP15 access
+ * kvm_handle_cp_64 -- handles a mrrc/mcrr trap on a guest CP14/CP15 access
  * @vcpu: The VCPU pointer
  * @run:  The kvm_run struct
  */
@@ -1095,7 +1094,7 @@ out:
 }
 
 /**
- * kvm_handle_cp15_32 -- handles a mrc/mcr trap on a guest CP15 access
+ * kvm_handle_cp_32 -- handles a mrc/mcr trap on a guest CP14/CP15 access
  * @vcpu: The VCPU pointer
  * @run:  The kvm_run struct
  */
diff --git a/arch/arm64/lib/strnlen.S b/arch/arm64/lib/strnlen.S
index 2ca665711bf2..eae38da6e0bb 100644
--- a/arch/arm64/lib/strnlen.S
+++ b/arch/arm64/lib/strnlen.S
@@ -168,4 +168,4 @@ CPU_LE( lsr	tmp2, tmp2, tmp4 )	/* Shift (tmp1 & 63).  */
 .Lhit_limit:
 	mov	len, limit
 	ret
-ENDPROC(strnlen)
+ENDPIPROC(strnlen)
diff --git a/arch/arm64/mm/dma-mapping.c b/arch/arm64/mm/dma-mapping.c
index 331c4ca6205c..a6e757cbab77 100644
--- a/arch/arm64/mm/dma-mapping.c
+++ b/arch/arm64/mm/dma-mapping.c
@@ -933,6 +933,10 @@ static int __init __iommu_dma_init(void)
 		ret = register_iommu_dma_ops_notifier(&platform_bus_type);
 	if (!ret)
 		ret = register_iommu_dma_ops_notifier(&amba_bustype);
+
+	/* handle devices queued before this arch_initcall */
+	if (!ret)
+		__iommu_attach_notifier(NULL, BUS_NOTIFY_ADD_DEVICE, NULL);
 	return ret;
 }
 arch_initcall(__iommu_dma_init);
diff --git a/arch/arm64/mm/fault.c b/arch/arm64/mm/fault.c
index 92ddac1e8ca2..abe2a9542b3a 100644
--- a/arch/arm64/mm/fault.c
+++ b/arch/arm64/mm/fault.c
@@ -371,6 +371,13 @@ static int __kprobes do_translation_fault(unsigned long addr,
 	return 0;
 }
 
+static int do_alignment_fault(unsigned long addr, unsigned int esr,
+			      struct pt_regs *regs)
+{
+	do_bad_area(addr, esr, regs);
+	return 0;
+}
+
 /*
  * This abort handler always returns "fault".
  */
@@ -418,7 +425,7 @@ static struct fault_info {
 	{ do_bad,		SIGBUS,  0,		"synchronous parity error (translation table walk)" },
 	{ do_bad,		SIGBUS,  0,		"synchronous parity error (translation table walk)" },
 	{ do_bad,		SIGBUS,  0,		"unknown 32"			},
-	{ do_bad,		SIGBUS,  BUS_ADRALN,	"alignment fault"		},
+	{ do_alignment_fault,	SIGBUS,  BUS_ADRALN,	"alignment fault"		},
 	{ do_bad,		SIGBUS,  0,		"unknown 34"			},
 	{ do_bad,		SIGBUS,  0,		"unknown 35"			},
 	{ do_bad,		SIGBUS,  0,		"unknown 36"			},
diff --git a/arch/arm64/mm/hugetlbpage.c b/arch/arm64/mm/hugetlbpage.c
index 82d607c3614e..da30529bb1f6 100644
--- a/arch/arm64/mm/hugetlbpage.c
+++ b/arch/arm64/mm/hugetlbpage.c
@@ -306,10 +306,6 @@ static __init int setup_hugepagesz(char *opt)
 		hugetlb_add_hstate(PMD_SHIFT - PAGE_SHIFT);
 	} else if (ps == PUD_SIZE) {
 		hugetlb_add_hstate(PUD_SHIFT - PAGE_SHIFT);
-	} else if (ps == (PAGE_SIZE * CONT_PTES)) {
-		hugetlb_add_hstate(CONT_PTE_SHIFT);
-	} else if (ps == (PMD_SIZE * CONT_PMDS)) {
-		hugetlb_add_hstate((PMD_SHIFT + CONT_PMD_SHIFT) - PAGE_SHIFT);
 	} else {
 		pr_err("hugepagesz: Unsupported page size %lu K\n", ps >> 10);
 		return 0;
@@ -317,13 +313,3 @@ static __init int setup_hugepagesz(char *opt)
 	return 1;
 }
 __setup("hugepagesz=", setup_hugepagesz);
-
-#ifdef CONFIG_ARM64_64K_PAGES
-static __init int add_default_hugepagesz(void)
-{
-	if (size_to_hstate(CONT_PTES * PAGE_SIZE) == NULL)
-		hugetlb_add_hstate(CONT_PMD_SHIFT);
-	return 0;
-}
-arch_initcall(add_default_hugepagesz);
-#endif
diff --git a/arch/arm64/mm/init.c b/arch/arm64/mm/init.c
index f3b061e67bfe..7802f216a67a 100644
--- a/arch/arm64/mm/init.c
+++ b/arch/arm64/mm/init.c
@@ -319,8 +319,8 @@ void __init mem_init(void)
 #endif
 		  MLG(VMALLOC_START, VMALLOC_END),
 #ifdef CONFIG_SPARSEMEM_VMEMMAP
-		  MLG((unsigned long)vmemmap,
-		      (unsigned long)vmemmap + VMEMMAP_SIZE),
+		  MLG(VMEMMAP_START,
+		      VMEMMAP_START + VMEMMAP_SIZE),
 		  MLM((unsigned long)virt_to_page(PAGE_OFFSET),
 		      (unsigned long)virt_to_page(high_memory)),
 #endif
diff --git a/arch/arm64/mm/mmap.c b/arch/arm64/mm/mmap.c
index 4c893b5189dd..232f787a088a 100644
--- a/arch/arm64/mm/mmap.c
+++ b/arch/arm64/mm/mmap.c
@@ -53,10 +53,10 @@ unsigned long arch_mmap_rnd(void)
 
 #ifdef CONFIG_COMPAT
 	if (test_thread_flag(TIF_32BIT))
-		rnd = (unsigned long)get_random_int() & ((1 << mmap_rnd_compat_bits) - 1);
+		rnd = get_random_long() & ((1UL << mmap_rnd_compat_bits) - 1);
 	else
 #endif
-		rnd = (unsigned long)get_random_int() & ((1 << mmap_rnd_bits) - 1);
+		rnd = get_random_long() & ((1UL << mmap_rnd_bits) - 1);
 	return rnd << PAGE_SHIFT;
 }
 
diff --git a/arch/m68k/configs/amiga_defconfig b/arch/m68k/configs/amiga_defconfig
index fc96e814188e..d1fc4796025e 100644
--- a/arch/m68k/configs/amiga_defconfig
+++ b/arch/m68k/configs/amiga_defconfig
@@ -108,6 +108,8 @@ CONFIG_NFT_NAT=m
 CONFIG_NFT_QUEUE=m
 CONFIG_NFT_REJECT=m
 CONFIG_NFT_COMPAT=m
+CONFIG_NFT_DUP_NETDEV=m
+CONFIG_NFT_FWD_NETDEV=m
 CONFIG_NETFILTER_XT_SET=m
 CONFIG_NETFILTER_XT_TARGET_CHECKSUM=m
 CONFIG_NETFILTER_XT_TARGET_CLASSIFY=m
@@ -266,6 +268,12 @@ CONFIG_L2TP=m
 CONFIG_BRIDGE=m
 CONFIG_ATALK=m
 CONFIG_6LOWPAN=m
+CONFIG_6LOWPAN_GHC_EXT_HDR_HOP=m
+CONFIG_6LOWPAN_GHC_UDP=m
+CONFIG_6LOWPAN_GHC_ICMPV6=m
+CONFIG_6LOWPAN_GHC_EXT_HDR_DEST=m
+CONFIG_6LOWPAN_GHC_EXT_HDR_FRAG=m
+CONFIG_6LOWPAN_GHC_EXT_HDR_ROUTE=m
 CONFIG_DNS_RESOLVER=y
 CONFIG_BATMAN_ADV=m
 CONFIG_BATMAN_ADV_DAT=y
@@ -366,6 +374,7 @@ CONFIG_ARIADNE=y
 # CONFIG_NET_VENDOR_INTEL is not set
 # CONFIG_NET_VENDOR_MARVELL is not set
 # CONFIG_NET_VENDOR_MICREL is not set
+# CONFIG_NET_VENDOR_NETRONOME is not set
 CONFIG_HYDRA=y
 CONFIG_APNE=y
 CONFIG_ZORRO8390=y
diff --git a/arch/m68k/configs/apollo_defconfig b/arch/m68k/configs/apollo_defconfig
index 05c904f08d9d..9bfe8be3658c 100644
--- a/arch/m68k/configs/apollo_defconfig
+++ b/arch/m68k/configs/apollo_defconfig
@@ -106,6 +106,8 @@ CONFIG_NFT_NAT=m
 CONFIG_NFT_QUEUE=m
 CONFIG_NFT_REJECT=m
 CONFIG_NFT_COMPAT=m
+CONFIG_NFT_DUP_NETDEV=m
+CONFIG_NFT_FWD_NETDEV=m
 CONFIG_NETFILTER_XT_SET=m
 CONFIG_NETFILTER_XT_TARGET_CHECKSUM=m
 CONFIG_NETFILTER_XT_TARGET_CLASSIFY=m
@@ -264,6 +266,12 @@ CONFIG_L2TP=m
 CONFIG_BRIDGE=m
 CONFIG_ATALK=m
 CONFIG_6LOWPAN=m
+CONFIG_6LOWPAN_GHC_EXT_HDR_HOP=m
+CONFIG_6LOWPAN_GHC_UDP=m
+CONFIG_6LOWPAN_GHC_ICMPV6=m
+CONFIG_6LOWPAN_GHC_EXT_HDR_DEST=m
+CONFIG_6LOWPAN_GHC_EXT_HDR_FRAG=m
+CONFIG_6LOWPAN_GHC_EXT_HDR_ROUTE=m
 CONFIG_DNS_RESOLVER=y
 CONFIG_BATMAN_ADV=m
 CONFIG_BATMAN_ADV_DAT=y
@@ -344,6 +352,7 @@ CONFIG_VETH=m
 # CONFIG_NET_VENDOR_MARVELL is not set
 # CONFIG_NET_VENDOR_MICREL is not set
 # CONFIG_NET_VENDOR_NATSEMI is not set
+# CONFIG_NET_VENDOR_NETRONOME is not set
 # CONFIG_NET_VENDOR_QUALCOMM is not set
 # CONFIG_NET_VENDOR_RENESAS is not set
 # CONFIG_NET_VENDOR_ROCKER is not set
diff --git a/arch/m68k/configs/atari_defconfig b/arch/m68k/configs/atari_defconfig
index d572b731c510..ebdcfae55580 100644
--- a/arch/m68k/configs/atari_defconfig
+++ b/arch/m68k/configs/atari_defconfig
@@ -106,6 +106,8 @@ CONFIG_NFT_NAT=m
 CONFIG_NFT_QUEUE=m
 CONFIG_NFT_REJECT=m
 CONFIG_NFT_COMPAT=m
+CONFIG_NFT_DUP_NETDEV=m
+CONFIG_NFT_FWD_NETDEV=m
 CONFIG_NETFILTER_XT_SET=m
 CONFIG_NETFILTER_XT_TARGET_CHECKSUM=m
 CONFIG_NETFILTER_XT_TARGET_CLASSIFY=m
@@ -264,6 +266,12 @@ CONFIG_L2TP=m
 CONFIG_BRIDGE=m
 CONFIG_ATALK=m
 CONFIG_6LOWPAN=m
+CONFIG_6LOWPAN_GHC_EXT_HDR_HOP=m
+CONFIG_6LOWPAN_GHC_UDP=m
+CONFIG_6LOWPAN_GHC_ICMPV6=m
+CONFIG_6LOWPAN_GHC_EXT_HDR_DEST=m
+CONFIG_6LOWPAN_GHC_EXT_HDR_FRAG=m
+CONFIG_6LOWPAN_GHC_EXT_HDR_ROUTE=m
 CONFIG_DNS_RESOLVER=y
 CONFIG_BATMAN_ADV=m
 CONFIG_BATMAN_ADV_DAT=y
@@ -353,6 +361,7 @@ CONFIG_ATARILANCE=y
 # CONFIG_NET_VENDOR_INTEL is not set
 # CONFIG_NET_VENDOR_MARVELL is not set
 # CONFIG_NET_VENDOR_MICREL is not set
+# CONFIG_NET_VENDOR_NETRONOME is not set
 CONFIG_NE2000=y
 # CONFIG_NET_VENDOR_QUALCOMM is not set
 # CONFIG_NET_VENDOR_RENESAS is not set
diff --git a/arch/m68k/configs/bvme6000_defconfig b/arch/m68k/configs/bvme6000_defconfig
index 11a30c65ad44..8acc65e54995 100644
--- a/arch/m68k/configs/bvme6000_defconfig
+++ b/arch/m68k/configs/bvme6000_defconfig
@@ -104,6 +104,8 @@ CONFIG_NFT_NAT=m
 CONFIG_NFT_QUEUE=m
 CONFIG_NFT_REJECT=m
 CONFIG_NFT_COMPAT=m
+CONFIG_NFT_DUP_NETDEV=m
+CONFIG_NFT_FWD_NETDEV=m
 CONFIG_NETFILTER_XT_SET=m
 CONFIG_NETFILTER_XT_TARGET_CHECKSUM=m
 CONFIG_NETFILTER_XT_TARGET_CLASSIFY=m
@@ -262,6 +264,12 @@ CONFIG_L2TP=m
 CONFIG_BRIDGE=m
 CONFIG_ATALK=m
 CONFIG_6LOWPAN=m
+CONFIG_6LOWPAN_GHC_EXT_HDR_HOP=m
+CONFIG_6LOWPAN_GHC_UDP=m
+CONFIG_6LOWPAN_GHC_ICMPV6=m
+CONFIG_6LOWPAN_GHC_EXT_HDR_DEST=m
+CONFIG_6LOWPAN_GHC_EXT_HDR_FRAG=m
+CONFIG_6LOWPAN_GHC_EXT_HDR_ROUTE=m
 CONFIG_DNS_RESOLVER=y
 CONFIG_BATMAN_ADV=m
 CONFIG_BATMAN_ADV_DAT=y
@@ -343,6 +351,7 @@ CONFIG_BVME6000_NET=y
 # CONFIG_NET_VENDOR_MARVELL is not set
 # CONFIG_NET_VENDOR_MICREL is not set
 # CONFIG_NET_VENDOR_NATSEMI is not set
+# CONFIG_NET_VENDOR_NETRONOME is not set
 # CONFIG_NET_VENDOR_QUALCOMM is not set
 # CONFIG_NET_VENDOR_RENESAS is not set
 # CONFIG_NET_VENDOR_ROCKER is not set
diff --git a/arch/m68k/configs/hp300_defconfig b/arch/m68k/configs/hp300_defconfig
index 6630a5154b9d..0c6a3d52b26e 100644
--- a/arch/m68k/configs/hp300_defconfig
+++ b/arch/m68k/configs/hp300_defconfig
@@ -106,6 +106,8 @@ CONFIG_NFT_NAT=m
 CONFIG_NFT_QUEUE=m
 CONFIG_NFT_REJECT=m
 CONFIG_NFT_COMPAT=m
+CONFIG_NFT_DUP_NETDEV=m
+CONFIG_NFT_FWD_NETDEV=m
 CONFIG_NETFILTER_XT_SET=m
 CONFIG_NETFILTER_XT_TARGET_CHECKSUM=m
 CONFIG_NETFILTER_XT_TARGET_CLASSIFY=m
@@ -264,6 +266,12 @@ CONFIG_L2TP=m
 CONFIG_BRIDGE=m
 CONFIG_ATALK=m
 CONFIG_6LOWPAN=m
+CONFIG_6LOWPAN_GHC_EXT_HDR_HOP=m
+CONFIG_6LOWPAN_GHC_UDP=m
+CONFIG_6LOWPAN_GHC_ICMPV6=m
+CONFIG_6LOWPAN_GHC_EXT_HDR_DEST=m
+CONFIG_6LOWPAN_GHC_EXT_HDR_FRAG=m
+CONFIG_6LOWPAN_GHC_EXT_HDR_ROUTE=m
 CONFIG_DNS_RESOLVER=y
 CONFIG_BATMAN_ADV=m
 CONFIG_BATMAN_ADV_DAT=y
@@ -345,6 +353,7 @@ CONFIG_HPLANCE=y
 # CONFIG_NET_VENDOR_MARVELL is not set
 # CONFIG_NET_VENDOR_MICREL is not set
 # CONFIG_NET_VENDOR_NATSEMI is not set
+# CONFIG_NET_VENDOR_NETRONOME is not set
 # CONFIG_NET_VENDOR_QUALCOMM is not set
 # CONFIG_NET_VENDOR_RENESAS is not set
 # CONFIG_NET_VENDOR_ROCKER is not set
diff --git a/arch/m68k/configs/mac_defconfig b/arch/m68k/configs/mac_defconfig
index 1d90b71d0903..12a8a6cb32f4 100644
--- a/arch/m68k/configs/mac_defconfig
+++ b/arch/m68k/configs/mac_defconfig
@@ -105,6 +105,8 @@ CONFIG_NFT_NAT=m
 CONFIG_NFT_QUEUE=m
 CONFIG_NFT_REJECT=m
 CONFIG_NFT_COMPAT=m
+CONFIG_NFT_DUP_NETDEV=m
+CONFIG_NFT_FWD_NETDEV=m
 CONFIG_NETFILTER_XT_SET=m
 CONFIG_NETFILTER_XT_TARGET_CHECKSUM=m
 CONFIG_NETFILTER_XT_TARGET_CLASSIFY=m
@@ -266,6 +268,12 @@ CONFIG_DEV_APPLETALK=m
 CONFIG_IPDDP=m
 CONFIG_IPDDP_ENCAP=y
 CONFIG_6LOWPAN=m
+CONFIG_6LOWPAN_GHC_EXT_HDR_HOP=m
+CONFIG_6LOWPAN_GHC_UDP=m
+CONFIG_6LOWPAN_GHC_ICMPV6=m
+CONFIG_6LOWPAN_GHC_EXT_HDR_DEST=m
+CONFIG_6LOWPAN_GHC_EXT_HDR_FRAG=m
+CONFIG_6LOWPAN_GHC_EXT_HDR_ROUTE=m
 CONFIG_DNS_RESOLVER=y
 CONFIG_BATMAN_ADV=m
 CONFIG_BATMAN_ADV_DAT=y
@@ -362,6 +370,7 @@ CONFIG_MAC89x0=y
 # CONFIG_NET_VENDOR_MARVELL is not set
 # CONFIG_NET_VENDOR_MICREL is not set
 CONFIG_MACSONIC=y
+# CONFIG_NET_VENDOR_NETRONOME is not set
 CONFIG_MAC8390=y
 # CONFIG_NET_VENDOR_QUALCOMM is not set
 # CONFIG_NET_VENDOR_RENESAS is not set
diff --git a/arch/m68k/configs/multi_defconfig b/arch/m68k/configs/multi_defconfig
index 1fd21c1ca87f..64ff2dcb34c8 100644
--- a/arch/m68k/configs/multi_defconfig
+++ b/arch/m68k/configs/multi_defconfig
@@ -115,6 +115,8 @@ CONFIG_NFT_NAT=m
 CONFIG_NFT_QUEUE=m
 CONFIG_NFT_REJECT=m
 CONFIG_NFT_COMPAT=m
+CONFIG_NFT_DUP_NETDEV=m
+CONFIG_NFT_FWD_NETDEV=m
 CONFIG_NETFILTER_XT_SET=m
 CONFIG_NETFILTER_XT_TARGET_CHECKSUM=m
 CONFIG_NETFILTER_XT_TARGET_CLASSIFY=m
@@ -276,6 +278,12 @@ CONFIG_DEV_APPLETALK=m
 CONFIG_IPDDP=m
 CONFIG_IPDDP_ENCAP=y
 CONFIG_6LOWPAN=m
+CONFIG_6LOWPAN_GHC_EXT_HDR_HOP=m
+CONFIG_6LOWPAN_GHC_UDP=m
+CONFIG_6LOWPAN_GHC_ICMPV6=m
+CONFIG_6LOWPAN_GHC_EXT_HDR_DEST=m
+CONFIG_6LOWPAN_GHC_EXT_HDR_FRAG=m
+CONFIG_6LOWPAN_GHC_EXT_HDR_ROUTE=m
 CONFIG_DNS_RESOLVER=y
 CONFIG_BATMAN_ADV=m
 CONFIG_BATMAN_ADV_DAT=y
@@ -404,6 +412,7 @@ CONFIG_MVME16x_NET=y
 # CONFIG_NET_VENDOR_MARVELL is not set
 # CONFIG_NET_VENDOR_MICREL is not set
 CONFIG_MACSONIC=y
+# CONFIG_NET_VENDOR_NETRONOME is not set
 CONFIG_HYDRA=y
 CONFIG_MAC8390=y
 CONFIG_NE2000=y
diff --git a/arch/m68k/configs/mvme147_defconfig b/arch/m68k/configs/mvme147_defconfig
index 74e10f79d7b1..07fc6abcfe0c 100644
--- a/arch/m68k/configs/mvme147_defconfig
+++ b/arch/m68k/configs/mvme147_defconfig
@@ -103,6 +103,8 @@ CONFIG_NFT_NAT=m
 CONFIG_NFT_QUEUE=m
 CONFIG_NFT_REJECT=m
 CONFIG_NFT_COMPAT=m
+CONFIG_NFT_DUP_NETDEV=m
+CONFIG_NFT_FWD_NETDEV=m
 CONFIG_NETFILTER_XT_SET=m
 CONFIG_NETFILTER_XT_TARGET_CHECKSUM=m
 CONFIG_NETFILTER_XT_TARGET_CLASSIFY=m
@@ -261,6 +263,12 @@ CONFIG_L2TP=m
 CONFIG_BRIDGE=m
 CONFIG_ATALK=m
 CONFIG_6LOWPAN=m
+CONFIG_6LOWPAN_GHC_EXT_HDR_HOP=m
+CONFIG_6LOWPAN_GHC_UDP=m
+CONFIG_6LOWPAN_GHC_ICMPV6=m
+CONFIG_6LOWPAN_GHC_EXT_HDR_DEST=m
+CONFIG_6LOWPAN_GHC_EXT_HDR_FRAG=m
+CONFIG_6LOWPAN_GHC_EXT_HDR_ROUTE=m
 CONFIG_DNS_RESOLVER=y
 CONFIG_BATMAN_ADV=m
 CONFIG_BATMAN_ADV_DAT=y
@@ -343,6 +351,7 @@ CONFIG_MVME147_NET=y
 # CONFIG_NET_VENDOR_MARVELL is not set
 # CONFIG_NET_VENDOR_MICREL is not set
 # CONFIG_NET_VENDOR_NATSEMI is not set
+# CONFIG_NET_VENDOR_NETRONOME is not set
 # CONFIG_NET_VENDOR_QUALCOMM is not set
 # CONFIG_NET_VENDOR_RENESAS is not set
 # CONFIG_NET_VENDOR_ROCKER is not set
diff --git a/arch/m68k/configs/mvme16x_defconfig b/arch/m68k/configs/mvme16x_defconfig
index 7034e716f166..69903ded88f7 100644
--- a/arch/m68k/configs/mvme16x_defconfig
+++ b/arch/m68k/configs/mvme16x_defconfig
@@ -104,6 +104,8 @@ CONFIG_NFT_NAT=m
 CONFIG_NFT_QUEUE=m
 CONFIG_NFT_REJECT=m
 CONFIG_NFT_COMPAT=m
+CONFIG_NFT_DUP_NETDEV=m
+CONFIG_NFT_FWD_NETDEV=m
 CONFIG_NETFILTER_XT_SET=m
 CONFIG_NETFILTER_XT_TARGET_CHECKSUM=m
 CONFIG_NETFILTER_XT_TARGET_CLASSIFY=m
@@ -262,6 +264,12 @@ CONFIG_L2TP=m
 CONFIG_BRIDGE=m
 CONFIG_ATALK=m
 CONFIG_6LOWPAN=m
+CONFIG_6LOWPAN_GHC_EXT_HDR_HOP=m
+CONFIG_6LOWPAN_GHC_UDP=m
+CONFIG_6LOWPAN_GHC_ICMPV6=m
+CONFIG_6LOWPAN_GHC_EXT_HDR_DEST=m
+CONFIG_6LOWPAN_GHC_EXT_HDR_FRAG=m
+CONFIG_6LOWPAN_GHC_EXT_HDR_ROUTE=m
 CONFIG_DNS_RESOLVER=y
 CONFIG_BATMAN_ADV=m
 CONFIG_BATMAN_ADV_DAT=y
@@ -343,6 +351,7 @@ CONFIG_MVME16x_NET=y
 # CONFIG_NET_VENDOR_MARVELL is not set
 # CONFIG_NET_VENDOR_MICREL is not set
 # CONFIG_NET_VENDOR_NATSEMI is not set
+# CONFIG_NET_VENDOR_NETRONOME is not set
 # CONFIG_NET_VENDOR_QUALCOMM is not set
 # CONFIG_NET_VENDOR_RENESAS is not set
 # CONFIG_NET_VENDOR_ROCKER is not set
diff --git a/arch/m68k/configs/q40_defconfig b/arch/m68k/configs/q40_defconfig
index f7deb5f702a6..bd8401686dde 100644
--- a/arch/m68k/configs/q40_defconfig
+++ b/arch/m68k/configs/q40_defconfig
@@ -104,6 +104,8 @@ CONFIG_NFT_NAT=m
 CONFIG_NFT_QUEUE=m
 CONFIG_NFT_REJECT=m
 CONFIG_NFT_COMPAT=m
+CONFIG_NFT_DUP_NETDEV=m
+CONFIG_NFT_FWD_NETDEV=m
 CONFIG_NETFILTER_XT_SET=m
 CONFIG_NETFILTER_XT_TARGET_CHECKSUM=m
 CONFIG_NETFILTER_XT_TARGET_CLASSIFY=m
@@ -262,6 +264,12 @@ CONFIG_L2TP=m
 CONFIG_BRIDGE=m
 CONFIG_ATALK=m
 CONFIG_6LOWPAN=m
+CONFIG_6LOWPAN_GHC_EXT_HDR_HOP=m
+CONFIG_6LOWPAN_GHC_UDP=m
+CONFIG_6LOWPAN_GHC_ICMPV6=m
+CONFIG_6LOWPAN_GHC_EXT_HDR_DEST=m
+CONFIG_6LOWPAN_GHC_EXT_HDR_FRAG=m
+CONFIG_6LOWPAN_GHC_EXT_HDR_ROUTE=m
 CONFIG_DNS_RESOLVER=y
 CONFIG_BATMAN_ADV=m
 CONFIG_BATMAN_ADV_DAT=y
@@ -352,6 +360,7 @@ CONFIG_VETH=m
 # CONFIG_NET_VENDOR_INTEL is not set
 # CONFIG_NET_VENDOR_MARVELL is not set
 # CONFIG_NET_VENDOR_MICREL is not set
+# CONFIG_NET_VENDOR_NETRONOME is not set
 CONFIG_NE2000=y
 # CONFIG_NET_VENDOR_QUALCOMM is not set
 # CONFIG_NET_VENDOR_RENESAS is not set
diff --git a/arch/m68k/configs/sun3_defconfig b/arch/m68k/configs/sun3_defconfig
index 0ce79eb0d805..5f9fb3ab9636 100644
--- a/arch/m68k/configs/sun3_defconfig
+++ b/arch/m68k/configs/sun3_defconfig
@@ -101,6 +101,8 @@ CONFIG_NFT_NAT=m
 CONFIG_NFT_QUEUE=m
 CONFIG_NFT_REJECT=m
 CONFIG_NFT_COMPAT=m
+CONFIG_NFT_DUP_NETDEV=m
+CONFIG_NFT_FWD_NETDEV=m
 CONFIG_NETFILTER_XT_SET=m
 CONFIG_NETFILTER_XT_TARGET_CHECKSUM=m
 CONFIG_NETFILTER_XT_TARGET_CLASSIFY=m
@@ -259,6 +261,12 @@ CONFIG_L2TP=m
 CONFIG_BRIDGE=m
 CONFIG_ATALK=m
 CONFIG_6LOWPAN=m
+CONFIG_6LOWPAN_GHC_EXT_HDR_HOP=m
+CONFIG_6LOWPAN_GHC_UDP=m
+CONFIG_6LOWPAN_GHC_ICMPV6=m
+CONFIG_6LOWPAN_GHC_EXT_HDR_DEST=m
+CONFIG_6LOWPAN_GHC_EXT_HDR_FRAG=m
+CONFIG_6LOWPAN_GHC_EXT_HDR_ROUTE=m
 CONFIG_DNS_RESOLVER=y
 CONFIG_BATMAN_ADV=m
 CONFIG_BATMAN_ADV_DAT=y
@@ -340,6 +348,7 @@ CONFIG_SUN3_82586=y
 # CONFIG_NET_VENDOR_MARVELL is not set
 # CONFIG_NET_VENDOR_MICREL is not set
 # CONFIG_NET_VENDOR_NATSEMI is not set
+# CONFIG_NET_VENDOR_NETRONOME is not set
 # CONFIG_NET_VENDOR_QUALCOMM is not set
 # CONFIG_NET_VENDOR_RENESAS is not set
 # CONFIG_NET_VENDOR_ROCKER is not set
diff --git a/arch/m68k/configs/sun3x_defconfig b/arch/m68k/configs/sun3x_defconfig
index 4cb787e4991f..5d1c674530e2 100644
--- a/arch/m68k/configs/sun3x_defconfig
+++ b/arch/m68k/configs/sun3x_defconfig
@@ -101,6 +101,8 @@ CONFIG_NFT_NAT=m
 CONFIG_NFT_QUEUE=m
 CONFIG_NFT_REJECT=m
 CONFIG_NFT_COMPAT=m
+CONFIG_NFT_DUP_NETDEV=m
+CONFIG_NFT_FWD_NETDEV=m
 CONFIG_NETFILTER_XT_SET=m
 CONFIG_NETFILTER_XT_TARGET_CHECKSUM=m
 CONFIG_NETFILTER_XT_TARGET_CLASSIFY=m
@@ -259,6 +261,12 @@ CONFIG_L2TP=m
 CONFIG_BRIDGE=m
 CONFIG_ATALK=m
 CONFIG_6LOWPAN=m
+CONFIG_6LOWPAN_GHC_EXT_HDR_HOP=m
+CONFIG_6LOWPAN_GHC_UDP=m
+CONFIG_6LOWPAN_GHC_ICMPV6=m
+CONFIG_6LOWPAN_GHC_EXT_HDR_DEST=m
+CONFIG_6LOWPAN_GHC_EXT_HDR_FRAG=m
+CONFIG_6LOWPAN_GHC_EXT_HDR_ROUTE=m
 CONFIG_DNS_RESOLVER=y
 CONFIG_BATMAN_ADV=m
 CONFIG_BATMAN_ADV_DAT=y
@@ -341,6 +349,7 @@ CONFIG_SUN3LANCE=y
 # CONFIG_NET_VENDOR_MARVELL is not set
 # CONFIG_NET_VENDOR_MICREL is not set
 # CONFIG_NET_VENDOR_NATSEMI is not set
+# CONFIG_NET_VENDOR_NETRONOME is not set
 # CONFIG_NET_VENDOR_QUALCOMM is not set
 # CONFIG_NET_VENDOR_RENESAS is not set
 # CONFIG_NET_VENDOR_ROCKER is not set
diff --git a/arch/m68k/include/asm/unistd.h b/arch/m68k/include/asm/unistd.h
index f9d96bf86910..bafaff6dcd7b 100644
--- a/arch/m68k/include/asm/unistd.h
+++ b/arch/m68k/include/asm/unistd.h
@@ -4,7 +4,7 @@
 #include <uapi/asm/unistd.h>
 
 
-#define NR_syscalls		376
+#define NR_syscalls		377
 
 #define __ARCH_WANT_OLD_READDIR
 #define __ARCH_WANT_OLD_STAT
diff --git a/arch/m68k/include/uapi/asm/unistd.h b/arch/m68k/include/uapi/asm/unistd.h
index 36cf129de663..0ca729665f29 100644
--- a/arch/m68k/include/uapi/asm/unistd.h
+++ b/arch/m68k/include/uapi/asm/unistd.h
@@ -381,5 +381,6 @@
 #define __NR_userfaultfd	373
 #define __NR_membarrier		374
 #define __NR_mlock2		375
+#define __NR_copy_file_range	376
 
 #endif /* _UAPI_ASM_M68K_UNISTD_H_ */
diff --git a/arch/m68k/kernel/syscalltable.S b/arch/m68k/kernel/syscalltable.S
index 282cd903f4c4..8bb94261ff97 100644
--- a/arch/m68k/kernel/syscalltable.S
+++ b/arch/m68k/kernel/syscalltable.S
@@ -396,3 +396,4 @@ ENTRY(sys_call_table)
 	.long sys_userfaultfd
 	.long sys_membarrier
 	.long sys_mlock2		/* 375 */
+	.long sys_copy_file_range
diff --git a/arch/mips/Kconfig b/arch/mips/Kconfig
index 57a945e832f4..74a3db92da1b 100644
--- a/arch/mips/Kconfig
+++ b/arch/mips/Kconfig
@@ -2085,7 +2085,7 @@ config PAGE_SIZE_32KB
 
 config PAGE_SIZE_64KB
 	bool "64kB"
-	depends on !CPU_R3000 && !CPU_TX39XX
+	depends on !CPU_R3000 && !CPU_TX39XX && !CPU_R6000
 	help
 	  Using 64kB page size will result in higher performance kernel at
 	  the price of higher memory consumption.  This option is available on
diff --git a/arch/mips/boot/dts/brcm/bcm6328.dtsi b/arch/mips/boot/dts/brcm/bcm6328.dtsi
index 459b9b252c3b..d61b1616b604 100644
--- a/arch/mips/boot/dts/brcm/bcm6328.dtsi
+++ b/arch/mips/boot/dts/brcm/bcm6328.dtsi
@@ -74,6 +74,7 @@
 		timer: timer@10000040 {
 			compatible = "syscon";
 			reg = <0x10000040 0x2c>;
+			little-endian;
 		};
 
 		reboot {
diff --git a/arch/mips/boot/dts/brcm/bcm7125.dtsi b/arch/mips/boot/dts/brcm/bcm7125.dtsi
index 4fc7ecee273c..1a7efa883c5e 100644
--- a/arch/mips/boot/dts/brcm/bcm7125.dtsi
+++ b/arch/mips/boot/dts/brcm/bcm7125.dtsi
@@ -98,6 +98,7 @@
 		sun_top_ctrl: syscon@404000 {
 			compatible = "brcm,bcm7125-sun-top-ctrl", "syscon";
 			reg = <0x404000 0x60c>;
+			little-endian;
 		};
 
 		reboot {
diff --git a/arch/mips/boot/dts/brcm/bcm7346.dtsi b/arch/mips/boot/dts/brcm/bcm7346.dtsi
index a3039bb53477..d4bf52cfcf17 100644
--- a/arch/mips/boot/dts/brcm/bcm7346.dtsi
+++ b/arch/mips/boot/dts/brcm/bcm7346.dtsi
@@ -118,6 +118,7 @@
 		sun_top_ctrl: syscon@404000 {
 			compatible = "brcm,bcm7346-sun-top-ctrl", "syscon";
 			reg = <0x404000 0x51c>;
+			little-endian;
 		};
 
 		reboot {
diff --git a/arch/mips/boot/dts/brcm/bcm7358.dtsi b/arch/mips/boot/dts/brcm/bcm7358.dtsi
index 4274ff41ec21..8e2501694d03 100644
--- a/arch/mips/boot/dts/brcm/bcm7358.dtsi
+++ b/arch/mips/boot/dts/brcm/bcm7358.dtsi
@@ -112,6 +112,7 @@
 		sun_top_ctrl: syscon@404000 {
 			compatible = "brcm,bcm7358-sun-top-ctrl", "syscon";
 			reg = <0x404000 0x51c>;
+			little-endian;
 		};
 
 		reboot {
diff --git a/arch/mips/boot/dts/brcm/bcm7360.dtsi b/arch/mips/boot/dts/brcm/bcm7360.dtsi
index 0dcc9163c27b..7e5f76040fb8 100644
--- a/arch/mips/boot/dts/brcm/bcm7360.dtsi
+++ b/arch/mips/boot/dts/brcm/bcm7360.dtsi
@@ -112,6 +112,7 @@
 		sun_top_ctrl: syscon@404000 {
 			compatible = "brcm,bcm7360-sun-top-ctrl", "syscon";
 			reg = <0x404000 0x51c>;
+			little-endian;
 		};
 
 		reboot {
diff --git a/arch/mips/boot/dts/brcm/bcm7362.dtsi b/arch/mips/boot/dts/brcm/bcm7362.dtsi
index 2f3f9fc2c478..c739ea77acb0 100644
--- a/arch/mips/boot/dts/brcm/bcm7362.dtsi
+++ b/arch/mips/boot/dts/brcm/bcm7362.dtsi
@@ -118,6 +118,7 @@
 		sun_top_ctrl: syscon@404000 {
 			compatible = "brcm,bcm7362-sun-top-ctrl", "syscon";
 			reg = <0x404000 0x51c>;
+			little-endian;
 		};
 
 		reboot {
diff --git a/arch/mips/boot/dts/brcm/bcm7420.dtsi b/arch/mips/boot/dts/brcm/bcm7420.dtsi
index bee221b3b568..5f55d0a50a28 100644
--- a/arch/mips/boot/dts/brcm/bcm7420.dtsi
+++ b/arch/mips/boot/dts/brcm/bcm7420.dtsi
@@ -99,6 +99,7 @@
 		sun_top_ctrl: syscon@404000 {
 			compatible = "brcm,bcm7420-sun-top-ctrl", "syscon";
 			reg = <0x404000 0x60c>;
+			little-endian;
 		};
 
 		reboot {
diff --git a/arch/mips/boot/dts/brcm/bcm7425.dtsi b/arch/mips/boot/dts/brcm/bcm7425.dtsi
index 571f30f52e3f..e24d41ab4e30 100644
--- a/arch/mips/boot/dts/brcm/bcm7425.dtsi
+++ b/arch/mips/boot/dts/brcm/bcm7425.dtsi
@@ -100,6 +100,7 @@
 		sun_top_ctrl: syscon@404000 {
 			compatible = "brcm,bcm7425-sun-top-ctrl", "syscon";
 			reg = <0x404000 0x51c>;
+			little-endian;
 		};
 
 		reboot {
diff --git a/arch/mips/boot/dts/brcm/bcm7435.dtsi b/arch/mips/boot/dts/brcm/bcm7435.dtsi
index 614ee211f71a..8b9432cc062b 100644
--- a/arch/mips/boot/dts/brcm/bcm7435.dtsi
+++ b/arch/mips/boot/dts/brcm/bcm7435.dtsi
@@ -114,6 +114,7 @@
 		sun_top_ctrl: syscon@404000 {
 			compatible = "brcm,bcm7425-sun-top-ctrl", "syscon";
 			reg = <0x404000 0x51c>;
+			little-endian;
 		};
 
 		reboot {
diff --git a/arch/mips/include/asm/elf.h b/arch/mips/include/asm/elf.h
index cefb7a596878..e090fc388e02 100644
--- a/arch/mips/include/asm/elf.h
+++ b/arch/mips/include/asm/elf.h
@@ -227,7 +227,7 @@ struct mips_elf_abiflags_v0 {
 	int __res = 1;							\
 	struct elfhdr *__h = (hdr);					\
 									\
-	if (__h->e_machine != EM_MIPS)					\
+	if (!mips_elf_check_machine(__h))				\
 		__res = 0;						\
 	if (__h->e_ident[EI_CLASS] != ELFCLASS32)			\
 		__res = 0;						\
@@ -258,7 +258,7 @@ struct mips_elf_abiflags_v0 {
 	int __res = 1;							\
 	struct elfhdr *__h = (hdr);					\
 									\
-	if (__h->e_machine != EM_MIPS)					\
+	if (!mips_elf_check_machine(__h))				\
 		__res = 0;						\
 	if (__h->e_ident[EI_CLASS] != ELFCLASS64)			\
 		__res = 0;						\
@@ -285,6 +285,11 @@ struct mips_elf_abiflags_v0 {
 
 #endif /* !defined(ELF_ARCH) */
 
+#define mips_elf_check_machine(x) ((x)->e_machine == EM_MIPS)
+
+#define vmcore_elf32_check_arch mips_elf_check_machine
+#define vmcore_elf64_check_arch mips_elf_check_machine
+
 struct mips_abi;
 
 extern struct mips_abi mips_abi;
diff --git a/arch/mips/include/asm/fpu.h b/arch/mips/include/asm/fpu.h
index 9cbf383b8834..f06f97bd62df 100644
--- a/arch/mips/include/asm/fpu.h
+++ b/arch/mips/include/asm/fpu.h
@@ -179,6 +179,10 @@ static inline void lose_fpu_inatomic(int save, struct task_struct *tsk)
 		if (save)
 			_save_fp(tsk);
 		__disable_fpu();
+	} else {
+		/* FPU should not have been left enabled with no owner */
+		WARN(read_c0_status() & ST0_CU1,
+		     "Orphaned FPU left enabled");
 	}
 	KSTK_STATUS(tsk) &= ~ST0_CU1;
 	clear_tsk_thread_flag(tsk, TIF_USEDFPU);
diff --git a/arch/mips/include/asm/octeon/octeon-feature.h b/arch/mips/include/asm/octeon/octeon-feature.h
index 8ebd3f579b84..3ed10a8d7865 100644
--- a/arch/mips/include/asm/octeon/octeon-feature.h
+++ b/arch/mips/include/asm/octeon/octeon-feature.h
@@ -128,7 +128,8 @@ static inline int octeon_has_feature(enum octeon_feature feature)
 	case OCTEON_FEATURE_PCIE:
 		return OCTEON_IS_MODEL(OCTEON_CN56XX)
 			|| OCTEON_IS_MODEL(OCTEON_CN52XX)
-			|| OCTEON_IS_MODEL(OCTEON_CN6XXX);
+			|| OCTEON_IS_MODEL(OCTEON_CN6XXX)
+			|| OCTEON_IS_MODEL(OCTEON_CN7XXX);
 
 	case OCTEON_FEATURE_SRIO:
 		return OCTEON_IS_MODEL(OCTEON_CN63XX)
diff --git a/arch/mips/include/asm/processor.h b/arch/mips/include/asm/processor.h
index 3f832c3dd8f5..041153f5cf93 100644
--- a/arch/mips/include/asm/processor.h
+++ b/arch/mips/include/asm/processor.h
@@ -45,7 +45,7 @@ extern unsigned int vced_count, vcei_count;
  * User space process size: 2GB. This is hardcoded into a few places,
  * so don't change it unless you know what you are doing.
  */
-#define TASK_SIZE	0x7fff8000UL
+#define TASK_SIZE	0x80000000UL
 #endif
 
 #define STACK_TOP_MAX	TASK_SIZE
diff --git a/arch/mips/include/asm/stackframe.h b/arch/mips/include/asm/stackframe.h
index a71da576883c..eebf39549606 100644
--- a/arch/mips/include/asm/stackframe.h
+++ b/arch/mips/include/asm/stackframe.h
@@ -289,7 +289,7 @@
 		.set	reorder
 		.set	noat
 		mfc0	a0, CP0_STATUS
-		li	v1, 0xff00
+		li	v1, ST0_CU1 | ST0_IM
 		ori	a0, STATMASK
 		xori	a0, STATMASK
 		mtc0	a0, CP0_STATUS
@@ -330,7 +330,7 @@
 		ori	a0, STATMASK
 		xori	a0, STATMASK
 		mtc0	a0, CP0_STATUS
-		li	v1, 0xff00
+		li	v1, ST0_CU1 | ST0_FR | ST0_IM
 		and	a0, v1
 		LONG_L	v0, PT_STATUS(sp)
 		nor	v1, $0, v1
diff --git a/arch/mips/include/asm/syscall.h b/arch/mips/include/asm/syscall.h
index 6499d93ae68d..47bc45a67e9b 100644
--- a/arch/mips/include/asm/syscall.h
+++ b/arch/mips/include/asm/syscall.h
@@ -101,10 +101,8 @@ static inline void syscall_get_arguments(struct task_struct *task,
 	/* O32 ABI syscall() - Either 64-bit with O32 or 32-bit */
 	if ((config_enabled(CONFIG_32BIT) ||
 	    test_tsk_thread_flag(task, TIF_32BIT_REGS)) &&
-	    (regs->regs[2] == __NR_syscall)) {
+	    (regs->regs[2] == __NR_syscall))
 		i++;
-		n++;
-	}
 
 	while (n--)
 		ret |= mips_get_syscall_arg(args++, task, regs, i++);
diff --git a/arch/mips/include/uapi/asm/unistd.h b/arch/mips/include/uapi/asm/unistd.h
index 90f03a7da665..3129795de940 100644
--- a/arch/mips/include/uapi/asm/unistd.h
+++ b/arch/mips/include/uapi/asm/unistd.h
@@ -380,16 +380,17 @@
 #define __NR_userfaultfd		(__NR_Linux + 357)
 #define __NR_membarrier			(__NR_Linux + 358)
 #define __NR_mlock2			(__NR_Linux + 359)
+#define __NR_copy_file_range		(__NR_Linux + 360)
 
 /*
  * Offset of the last Linux o32 flavoured syscall
  */
-#define __NR_Linux_syscalls		359
+#define __NR_Linux_syscalls		360
 
 #endif /* _MIPS_SIM == _MIPS_SIM_ABI32 */
 
 #define __NR_O32_Linux			4000
-#define __NR_O32_Linux_syscalls		359
+#define __NR_O32_Linux_syscalls		360
 
 #if _MIPS_SIM == _MIPS_SIM_ABI64
 
@@ -717,16 +718,17 @@
 #define __NR_userfaultfd		(__NR_Linux + 317)
 #define __NR_membarrier			(__NR_Linux + 318)
 #define __NR_mlock2			(__NR_Linux + 319)
+#define __NR_copy_file_range		(__NR_Linux + 320)
 
 /*
  * Offset of the last Linux 64-bit flavoured syscall
  */
-#define __NR_Linux_syscalls		319
+#define __NR_Linux_syscalls		320
 
 #endif /* _MIPS_SIM == _MIPS_SIM_ABI64 */
 
 #define __NR_64_Linux			5000
-#define __NR_64_Linux_syscalls		319
+#define __NR_64_Linux_syscalls		320
 
 #if _MIPS_SIM == _MIPS_SIM_NABI32
 
@@ -1058,15 +1060,16 @@
 #define __NR_userfaultfd		(__NR_Linux + 321)
 #define __NR_membarrier			(__NR_Linux + 322)
 #define __NR_mlock2			(__NR_Linux + 323)
+#define __NR_copy_file_range		(__NR_Linux + 324)
 
 /*
  * Offset of the last N32 flavoured syscall
  */
-#define __NR_Linux_syscalls		323
+#define __NR_Linux_syscalls		324
 
 #endif /* _MIPS_SIM == _MIPS_SIM_NABI32 */
 
 #define __NR_N32_Linux			6000
-#define __NR_N32_Linux_syscalls		323
+#define __NR_N32_Linux_syscalls		324
 
 #endif /* _UAPI_ASM_UNISTD_H */
diff --git a/arch/mips/jz4740/gpio.c b/arch/mips/jz4740/gpio.c
index 8c6d76c9b2d6..d9907e57e9b9 100644
--- a/arch/mips/jz4740/gpio.c
+++ b/arch/mips/jz4740/gpio.c
@@ -270,7 +270,7 @@ uint32_t jz_gpio_port_get_value(int port, uint32_t mask)
 }
 EXPORT_SYMBOL(jz_gpio_port_get_value);
 
-#define IRQ_TO_BIT(irq) BIT(irq_to_gpio(irq) & 0x1f)
+#define IRQ_TO_BIT(irq) BIT((irq - JZ4740_IRQ_GPIO(0)) & 0x1f)
 
 static void jz_gpio_check_trigger_both(struct jz_gpio_chip *chip, unsigned int irq)
 {
diff --git a/arch/mips/kernel/binfmt_elfn32.c b/arch/mips/kernel/binfmt_elfn32.c
index 1188e00bb120..1b992c6e3d8e 100644
--- a/arch/mips/kernel/binfmt_elfn32.c
+++ b/arch/mips/kernel/binfmt_elfn32.c
@@ -35,7 +35,7 @@ typedef elf_fpreg_t elf_fpregset_t[ELF_NFPREG];
 	int __res = 1;							\
 	struct elfhdr *__h = (hdr);					\
 									\
-	if (__h->e_machine != EM_MIPS)					\
+	if (!mips_elf_check_machine(__h))				\
 		__res = 0;						\
 	if (__h->e_ident[EI_CLASS] != ELFCLASS32)			\
 		__res = 0;						\
diff --git a/arch/mips/kernel/binfmt_elfo32.c b/arch/mips/kernel/binfmt_elfo32.c
index 928767858b86..abd3affe5fb3 100644
--- a/arch/mips/kernel/binfmt_elfo32.c
+++ b/arch/mips/kernel/binfmt_elfo32.c
@@ -47,7 +47,7 @@ typedef elf_fpreg_t elf_fpregset_t[ELF_NFPREG];
 	int __res = 1;							\
 	struct elfhdr *__h = (hdr);					\
 									\
-	if (__h->e_machine != EM_MIPS)					\
+	if (!mips_elf_check_machine(__h))				\
 		__res = 0;						\
 	if (__h->e_ident[EI_CLASS] != ELFCLASS32)			\
 		__res = 0;						\
diff --git a/arch/mips/kernel/process.c b/arch/mips/kernel/process.c
index f2975d4d1e44..eddd5fd6fdfa 100644
--- a/arch/mips/kernel/process.c
+++ b/arch/mips/kernel/process.c
@@ -65,12 +65,10 @@ void start_thread(struct pt_regs * regs, unsigned long pc, unsigned long sp)
 	status = regs->cp0_status & ~(ST0_CU0|ST0_CU1|ST0_FR|KU_MASK);
 	status |= KU_USER;
 	regs->cp0_status = status;
+	lose_fpu(0);
+	clear_thread_flag(TIF_MSA_CTX_LIVE);
 	clear_used_math();
-	clear_fpu_owner();
 	init_dsp();
-	clear_thread_flag(TIF_USEDMSA);
-	clear_thread_flag(TIF_MSA_CTX_LIVE);
-	disable_msa();
 	regs->cp0_epc = pc;
 	regs->regs[29] = sp;
 }
diff --git a/arch/mips/kernel/r2300_fpu.S b/arch/mips/kernel/r2300_fpu.S
index 5ce3b746cedc..b4ac6374a38f 100644
--- a/arch/mips/kernel/r2300_fpu.S
+++ b/arch/mips/kernel/r2300_fpu.S
@@ -125,7 +125,7 @@ LEAF(_restore_fp_context)
 	END(_restore_fp_context)
 	.set	reorder
 
-	.type	fault@function
+	.type	fault, @function
 	.ent	fault
 fault:	li	v0, -EFAULT
 	jr	ra
diff --git a/arch/mips/kernel/r4k_fpu.S b/arch/mips/kernel/r4k_fpu.S
index f09546ee2cdc..17732f876eff 100644
--- a/arch/mips/kernel/r4k_fpu.S
+++ b/arch/mips/kernel/r4k_fpu.S
@@ -358,7 +358,7 @@ LEAF(_restore_msa_all_upper)
 
 	.set	reorder
 
-	.type	fault@function
+	.type	fault, @function
 	.ent	fault
 fault:	li	v0, -EFAULT				# failure
 	jr	ra
diff --git a/arch/mips/kernel/scall32-o32.S b/arch/mips/kernel/scall32-o32.S
index 2d23c834ba96..a56317444bda 100644
--- a/arch/mips/kernel/scall32-o32.S
+++ b/arch/mips/kernel/scall32-o32.S
@@ -595,3 +595,4 @@ EXPORT(sys_call_table)
 	PTR	sys_userfaultfd
 	PTR	sys_membarrier
 	PTR	sys_mlock2
+	PTR	sys_copy_file_range		/* 4360 */
diff --git a/arch/mips/kernel/scall64-64.S b/arch/mips/kernel/scall64-64.S
index deac63315d0e..2b2dc14610d0 100644
--- a/arch/mips/kernel/scall64-64.S
+++ b/arch/mips/kernel/scall64-64.S
@@ -433,4 +433,5 @@ EXPORT(sys_call_table)
 	PTR	sys_userfaultfd
 	PTR	sys_membarrier
 	PTR	sys_mlock2
+	PTR	sys_copy_file_range		/* 5320 */
 	.size	sys_call_table,.-sys_call_table
diff --git a/arch/mips/kernel/scall64-n32.S b/arch/mips/kernel/scall64-n32.S
index 5a69eb48d0a8..2bf5c8593d91 100644
--- a/arch/mips/kernel/scall64-n32.S
+++ b/arch/mips/kernel/scall64-n32.S
@@ -423,4 +423,5 @@ EXPORT(sysn32_call_table)
 	PTR	sys_userfaultfd
 	PTR	sys_membarrier
 	PTR	sys_mlock2
+	PTR	sys_copy_file_range
 	.size	sysn32_call_table,.-sysn32_call_table
diff --git a/arch/mips/kernel/scall64-o32.S b/arch/mips/kernel/scall64-o32.S
index e4b6d7c97822..c5b759e584c7 100644
--- a/arch/mips/kernel/scall64-o32.S
+++ b/arch/mips/kernel/scall64-o32.S
@@ -578,4 +578,5 @@ EXPORT(sys32_call_table)
 	PTR	sys_userfaultfd
 	PTR	sys_membarrier
 	PTR	sys_mlock2
+	PTR	sys_copy_file_range		/* 4360 */
 	.size	sys32_call_table,.-sys32_call_table
diff --git a/arch/mips/kernel/setup.c b/arch/mips/kernel/setup.c
index 569a7d5242dd..5fdaf8bdcd2e 100644
--- a/arch/mips/kernel/setup.c
+++ b/arch/mips/kernel/setup.c
@@ -782,6 +782,7 @@ static inline void prefill_possible_map(void) {}
 void __init setup_arch(char **cmdline_p)
 {
 	cpu_probe();
+	mips_cm_probe();
 	prom_init();
 
 	setup_early_fdc_console();
diff --git a/arch/mips/kernel/traps.c b/arch/mips/kernel/traps.c
index bafcb7ad5c85..bf14da9f3e33 100644
--- a/arch/mips/kernel/traps.c
+++ b/arch/mips/kernel/traps.c
@@ -663,7 +663,7 @@ static int simulate_rdhwr_normal(struct pt_regs *regs, unsigned int opcode)
 	return -1;
 }
 
-static int simulate_rdhwr_mm(struct pt_regs *regs, unsigned short opcode)
+static int simulate_rdhwr_mm(struct pt_regs *regs, unsigned int opcode)
 {
 	if ((opcode & MM_POOL32A_FUNC) == MM_RDHWR) {
 		int rd = (opcode & MM_RS) >> 16;
@@ -690,15 +690,15 @@ static int simulate_sync(struct pt_regs *regs, unsigned int opcode)
 asmlinkage void do_ov(struct pt_regs *regs)
 {
 	enum ctx_state prev_state;
-	siginfo_t info;
+	siginfo_t info = {
+		.si_signo = SIGFPE,
+		.si_code = FPE_INTOVF,
+		.si_addr = (void __user *)regs->cp0_epc,
+	};
 
 	prev_state = exception_enter();
 	die_if_kernel("Integer overflow", regs);
 
-	info.si_code = FPE_INTOVF;
-	info.si_signo = SIGFPE;
-	info.si_errno = 0;
-	info.si_addr = (void __user *) regs->cp0_epc;
 	force_sig_info(SIGFPE, &info, current);
 	exception_exit(prev_state);
 }
@@ -874,7 +874,7 @@ out:
 void do_trap_or_bp(struct pt_regs *regs, unsigned int code,
 	const char *str)
 {
-	siginfo_t info;
+	siginfo_t info = { 0 };
 	char b[40];
 
 #ifdef CONFIG_KGDB_LOW_LEVEL_TRAP
@@ -903,7 +903,6 @@ void do_trap_or_bp(struct pt_regs *regs, unsigned int code,
 		else
 			info.si_code = FPE_INTOVF;
 		info.si_signo = SIGFPE;
-		info.si_errno = 0;
 		info.si_addr = (void __user *) regs->cp0_epc;
 		force_sig_info(SIGFPE, &info, current);
 		break;
@@ -1119,11 +1118,12 @@ no_r2_instr:
 	if (get_isa16_mode(regs->cp0_epc)) {
 		unsigned short mmop[2] = { 0 };
 
-		if (unlikely(get_user(mmop[0], epc) < 0))
+		if (unlikely(get_user(mmop[0], (u16 __user *)epc + 0) < 0))
 			status = SIGSEGV;
-		if (unlikely(get_user(mmop[1], epc) < 0))
+		if (unlikely(get_user(mmop[1], (u16 __user *)epc + 1) < 0))
 			status = SIGSEGV;
-		opcode = (mmop[0] << 16) | mmop[1];
+		opcode = mmop[0];
+		opcode = (opcode << 16) | mmop[1];
 
 		if (status < 0)
 			status = simulate_rdhwr_mm(regs, opcode);
@@ -1369,26 +1369,12 @@ asmlinkage void do_cpu(struct pt_regs *regs)
 		if (unlikely(compute_return_epc(regs) < 0))
 			break;
 
-		if (get_isa16_mode(regs->cp0_epc)) {
-			unsigned short mmop[2] = { 0 };
-
-			if (unlikely(get_user(mmop[0], epc) < 0))
-				status = SIGSEGV;
-			if (unlikely(get_user(mmop[1], epc) < 0))
-				status = SIGSEGV;
-			opcode = (mmop[0] << 16) | mmop[1];
-
-			if (status < 0)
-				status = simulate_rdhwr_mm(regs, opcode);
-		} else {
+		if (!get_isa16_mode(regs->cp0_epc)) {
 			if (unlikely(get_user(opcode, epc) < 0))
 				status = SIGSEGV;
 
 			if (!cpu_has_llsc && status < 0)
 				status = simulate_llsc(regs, opcode);
-
-			if (status < 0)
-				status = simulate_rdhwr_normal(regs, opcode);
 		}
 
 		if (status < 0)
diff --git a/arch/mips/kvm/mips.c b/arch/mips/kvm/mips.c
index 8bc3977576e6..3110447ab1e9 100644
--- a/arch/mips/kvm/mips.c
+++ b/arch/mips/kvm/mips.c
@@ -702,7 +702,7 @@ static int kvm_mips_get_reg(struct kvm_vcpu *vcpu,
 	} else if ((reg->id & KVM_REG_SIZE_MASK) == KVM_REG_SIZE_U128) {
 		void __user *uaddr = (void __user *)(long)reg->addr;
 
-		return copy_to_user(uaddr, vs, 16);
+		return copy_to_user(uaddr, vs, 16) ? -EFAULT : 0;
 	} else {
 		return -EINVAL;
 	}
@@ -732,7 +732,7 @@ static int kvm_mips_set_reg(struct kvm_vcpu *vcpu,
 	} else if ((reg->id & KVM_REG_SIZE_MASK) == KVM_REG_SIZE_U128) {
 		void __user *uaddr = (void __user *)(long)reg->addr;
 
-		return copy_from_user(vs, uaddr, 16);
+		return copy_from_user(vs, uaddr, 16) ? -EFAULT : 0;
 	} else {
 		return -EINVAL;
 	}
diff --git a/arch/mips/mm/mmap.c b/arch/mips/mm/mmap.c
index 5c81fdd032c3..353037699512 100644
--- a/arch/mips/mm/mmap.c
+++ b/arch/mips/mm/mmap.c
@@ -146,7 +146,7 @@ unsigned long arch_mmap_rnd(void)
 {
 	unsigned long rnd;
 
-	rnd = (unsigned long)get_random_int();
+	rnd = get_random_long();
 	rnd <<= PAGE_SHIFT;
 	if (TASK_IS_32BIT_ADDR)
 		rnd &= 0xfffffful;
@@ -174,7 +174,7 @@ void arch_pick_mmap_layout(struct mm_struct *mm)
 
 static inline unsigned long brk_rnd(void)
 {
-	unsigned long rnd = get_random_int();
+	unsigned long rnd = get_random_long();
 
 	rnd = rnd << PAGE_SHIFT;
 	/* 8MB for 32bit, 256MB for 64bit */
diff --git a/arch/mips/mm/sc-mips.c b/arch/mips/mm/sc-mips.c
index 3bd0597d9c3d..91dec32c77b7 100644
--- a/arch/mips/mm/sc-mips.c
+++ b/arch/mips/mm/sc-mips.c
@@ -164,11 +164,13 @@ static int __init mips_sc_probe_cm3(void)
 
 	sets = cfg & CM_GCR_L2_CONFIG_SET_SIZE_MSK;
 	sets >>= CM_GCR_L2_CONFIG_SET_SIZE_SHF;
-	c->scache.sets = 64 << sets;
+	if (sets)
+		c->scache.sets = 64 << sets;
 
 	line_sz = cfg & CM_GCR_L2_CONFIG_LINE_SIZE_MSK;
 	line_sz >>= CM_GCR_L2_CONFIG_LINE_SIZE_SHF;
-	c->scache.linesz = 2 << line_sz;
+	if (line_sz)
+		c->scache.linesz = 2 << line_sz;
 
 	assoc = cfg & CM_GCR_L2_CONFIG_ASSOC_MSK;
 	assoc >>= CM_GCR_L2_CONFIG_ASSOC_SHF;
@@ -176,13 +178,12 @@ static int __init mips_sc_probe_cm3(void)
 	c->scache.waysize = c->scache.sets * c->scache.linesz;
 	c->scache.waybit = __ffs(c->scache.waysize);
 
-	c->scache.flags &= ~MIPS_CACHE_NOT_PRESENT;
+	if (c->scache.linesz) {
+		c->scache.flags &= ~MIPS_CACHE_NOT_PRESENT;
+		return 1;
+	}
 
-	return 1;
-}
-
-void __weak platform_early_l2_init(void)
-{
+	return 0;
 }
 
 static inline int __init mips_sc_probe(void)
@@ -194,12 +195,6 @@ static inline int __init mips_sc_probe(void)
 	/* Mark as not present until probe completed */
 	c->scache.flags |= MIPS_CACHE_NOT_PRESENT;
 
-	/*
-	 * Do we need some platform specific probing before
-	 * we configure L2?
-	 */
-	platform_early_l2_init();
-
 	if (mips_cm_revision() >= CM_REV_CM3)
 		return mips_sc_probe_cm3();
 
diff --git a/arch/mips/mti-malta/malta-init.c b/arch/mips/mti-malta/malta-init.c
index 571148c5fd0b..dc2c5214809d 100644
--- a/arch/mips/mti-malta/malta-init.c
+++ b/arch/mips/mti-malta/malta-init.c
@@ -293,7 +293,6 @@ mips_pci_controller:
 	console_config();
 #endif
 	/* Early detection of CMP support */
-	mips_cm_probe();
 	mips_cpc_probe();
 
 	if (!register_cps_smp_ops())
@@ -304,10 +303,3 @@ mips_pci_controller:
 		return;
 	register_up_smp_ops();
 }
-
-void platform_early_l2_init(void)
-{
-	/* L2 configuration lives in the CM3 */
-	if (mips_cm_revision() >= CM_REV_CM3)
-		mips_cm_probe();
-}
diff --git a/arch/mips/pci/pci-mt7620.c b/arch/mips/pci/pci-mt7620.c
index a009ee458934..1ae932c2d78b 100644
--- a/arch/mips/pci/pci-mt7620.c
+++ b/arch/mips/pci/pci-mt7620.c
@@ -297,12 +297,12 @@ static int mt7620_pci_probe(struct platform_device *pdev)
 		return PTR_ERR(rstpcie0);
 
 	bridge_base = devm_ioremap_resource(&pdev->dev, bridge_res);
-	if (!bridge_base)
-		return -ENOMEM;
+	if (IS_ERR(bridge_base))
+		return PTR_ERR(bridge_base);
 
 	pcie_base = devm_ioremap_resource(&pdev->dev, pcie_res);
-	if (!pcie_base)
-		return -ENOMEM;
+	if (IS_ERR(pcie_base))
+		return PTR_ERR(pcie_base);
 
 	iomem_resource.start = 0;
 	iomem_resource.end = ~0;
diff --git a/arch/parisc/include/asm/floppy.h b/arch/parisc/include/asm/floppy.h
index f84ff12574b7..6d8276cd25ca 100644
--- a/arch/parisc/include/asm/floppy.h
+++ b/arch/parisc/include/asm/floppy.h
@@ -33,7 +33,7 @@
  * floppy accesses go through the track buffer.
  */
 #define _CROSS_64KB(a,s,vdma) \
-(!vdma && ((unsigned long)(a)/K_64 != ((unsigned long)(a) + (s) - 1) / K_64))
+(!(vdma) && ((unsigned long)(a)/K_64 != ((unsigned long)(a) + (s) - 1) / K_64))
 
 #define CROSS_64KB(a,s) _CROSS_64KB(a,s,use_virtual_dma & 1)
 
diff --git a/arch/parisc/include/uapi/asm/unistd.h b/arch/parisc/include/uapi/asm/unistd.h
index 35bdccbb2036..b75039f92116 100644
--- a/arch/parisc/include/uapi/asm/unistd.h
+++ b/arch/parisc/include/uapi/asm/unistd.h
@@ -361,8 +361,9 @@
 #define __NR_membarrier		(__NR_Linux + 343)
 #define __NR_userfaultfd	(__NR_Linux + 344)
 #define __NR_mlock2		(__NR_Linux + 345)
+#define __NR_copy_file_range	(__NR_Linux + 346)
 
-#define __NR_Linux_syscalls	(__NR_mlock2 + 1)
+#define __NR_Linux_syscalls	(__NR_copy_file_range + 1)
 
 
 #define __IGNORE_select		/* newselect */
diff --git a/arch/parisc/kernel/ptrace.c b/arch/parisc/kernel/ptrace.c
index 9585c81f755f..ce0b2b4075c7 100644
--- a/arch/parisc/kernel/ptrace.c
+++ b/arch/parisc/kernel/ptrace.c
@@ -269,14 +269,19 @@ long compat_arch_ptrace(struct task_struct *child, compat_long_t request,
 
 long do_syscall_trace_enter(struct pt_regs *regs)
 {
-	long ret = 0;
-
 	/* Do the secure computing check first. */
 	secure_computing_strict(regs->gr[20]);
 
 	if (test_thread_flag(TIF_SYSCALL_TRACE) &&
-	    tracehook_report_syscall_entry(regs))
-		ret = -1L;
+	    tracehook_report_syscall_entry(regs)) {
+		/*
+		 * Tracing decided this syscall should not happen or the
+		 * debugger stored an invalid system call number. Skip
+		 * the system call and the system call restart handling.
+		 */
+		regs->gr[20] = -1UL;
+		goto out;
+	}
 
 #ifdef CONFIG_64BIT
 	if (!is_compat_task())
@@ -290,7 +295,8 @@ long do_syscall_trace_enter(struct pt_regs *regs)
 			regs->gr[24] & 0xffffffff,
 			regs->gr[23] & 0xffffffff);
 
-	return ret ? : regs->gr[20];
+out:
+	return regs->gr[20];
 }
 
 void do_syscall_trace_exit(struct pt_regs *regs)
diff --git a/arch/parisc/kernel/syscall.S b/arch/parisc/kernel/syscall.S
index 3fbd7252a4b2..fbafa0d0e2bf 100644
--- a/arch/parisc/kernel/syscall.S
+++ b/arch/parisc/kernel/syscall.S
@@ -343,7 +343,7 @@ tracesys_next:
 #endif
 
 	comiclr,>>=	__NR_Linux_syscalls, %r20, %r0
-	b,n	.Lsyscall_nosys
+	b,n	.Ltracesys_nosys
 
 	LDREGX  %r20(%r19), %r19
 
@@ -359,6 +359,9 @@ tracesys_next:
 	be      0(%sr7,%r19)
 	ldo	R%tracesys_exit(%r2),%r2
 
+.Ltracesys_nosys:
+	ldo	-ENOSYS(%r0),%r28		/* set errno */
+
 	/* Do *not* call this function on the gateway page, because it
 	makes a direct call to syscall_trace. */
 	
diff --git a/arch/parisc/kernel/syscall_table.S b/arch/parisc/kernel/syscall_table.S
index d4ffcfbc9885..585d50fc75c0 100644
--- a/arch/parisc/kernel/syscall_table.S
+++ b/arch/parisc/kernel/syscall_table.S
@@ -441,6 +441,7 @@
 	ENTRY_SAME(membarrier)
 	ENTRY_SAME(userfaultfd)
 	ENTRY_SAME(mlock2)		/* 345 */
+	ENTRY_SAME(copy_file_range)
 
 
 .ifne (. - 90b) - (__NR_Linux_syscalls * (91b - 90b))
diff --git a/arch/powerpc/Kconfig b/arch/powerpc/Kconfig
index e4824fd04bb7..9faa18c4f3f7 100644
--- a/arch/powerpc/Kconfig
+++ b/arch/powerpc/Kconfig
@@ -557,7 +557,7 @@ choice
 
 config PPC_4K_PAGES
 	bool "4k page size"
-	select HAVE_ARCH_SOFT_DIRTY if CHECKPOINT_RESTORE && PPC_BOOK3S
+	select HAVE_ARCH_SOFT_DIRTY if PPC_BOOK3S_64
 
 config PPC_16K_PAGES
 	bool "16k page size"
@@ -566,7 +566,7 @@ config PPC_16K_PAGES
 config PPC_64K_PAGES
 	bool "64k page size"
 	depends on !PPC_FSL_BOOK3E && (44x || PPC_STD_MMU_64 || PPC_BOOK3E_64)
-	select HAVE_ARCH_SOFT_DIRTY if CHECKPOINT_RESTORE && PPC_BOOK3S
+	select HAVE_ARCH_SOFT_DIRTY if PPC_BOOK3S_64
 
 config PPC_256K_PAGES
 	bool "256k page size"
diff --git a/arch/powerpc/include/asm/book3s/64/pgtable.h b/arch/powerpc/include/asm/book3s/64/pgtable.h
index 8d1c41d28318..ac07a30a7934 100644
--- a/arch/powerpc/include/asm/book3s/64/pgtable.h
+++ b/arch/powerpc/include/asm/book3s/64/pgtable.h
@@ -281,6 +281,10 @@ extern pgtable_t pgtable_trans_huge_withdraw(struct mm_struct *mm, pmd_t *pmdp);
 extern void pmdp_invalidate(struct vm_area_struct *vma, unsigned long address,
 			    pmd_t *pmdp);
 
+#define __HAVE_ARCH_PMDP_HUGE_SPLIT_PREPARE
+extern void pmdp_huge_split_prepare(struct vm_area_struct *vma,
+				    unsigned long address, pmd_t *pmdp);
+
 #define pmd_move_must_withdraw pmd_move_must_withdraw
 struct spinlock;
 static inline int pmd_move_must_withdraw(struct spinlock *new_pmd_ptl,
diff --git a/arch/powerpc/include/asm/eeh.h b/arch/powerpc/include/asm/eeh.h
index c5eb86f3d452..867c39b45df6 100644
--- a/arch/powerpc/include/asm/eeh.h
+++ b/arch/powerpc/include/asm/eeh.h
@@ -81,6 +81,7 @@ struct pci_dn;
 #define EEH_PE_KEEP		(1 << 8)	/* Keep PE on hotplug	*/
 #define EEH_PE_CFG_RESTRICTED	(1 << 9)	/* Block config on error */
 #define EEH_PE_REMOVED		(1 << 10)	/* Removed permanently	*/
+#define EEH_PE_PRI_BUS		(1 << 11)	/* Cached primary bus   */
 
 struct eeh_pe {
 	int type;			/* PE type: PHB/Bus/Device	*/
diff --git a/arch/powerpc/include/asm/trace.h b/arch/powerpc/include/asm/trace.h
index 8e86b48d0369..32e36b16773f 100644
--- a/arch/powerpc/include/asm/trace.h
+++ b/arch/powerpc/include/asm/trace.h
@@ -57,12 +57,14 @@ DEFINE_EVENT(ppc64_interrupt_class, timer_interrupt_exit,
 extern void hcall_tracepoint_regfunc(void);
 extern void hcall_tracepoint_unregfunc(void);
 
-TRACE_EVENT_FN(hcall_entry,
+TRACE_EVENT_FN_COND(hcall_entry,
 
 	TP_PROTO(unsigned long opcode, unsigned long *args),
 
 	TP_ARGS(opcode, args),
 
+	TP_CONDITION(cpu_online(raw_smp_processor_id())),
+
 	TP_STRUCT__entry(
 		__field(unsigned long, opcode)
 	),
@@ -76,13 +78,15 @@ TRACE_EVENT_FN(hcall_entry,
 	hcall_tracepoint_regfunc, hcall_tracepoint_unregfunc
 );
 
-TRACE_EVENT_FN(hcall_exit,
+TRACE_EVENT_FN_COND(hcall_exit,
 
 	TP_PROTO(unsigned long opcode, unsigned long retval,
 		unsigned long *retbuf),
 
 	TP_ARGS(opcode, retval, retbuf),
 
+	TP_CONDITION(cpu_online(raw_smp_processor_id())),
+
 	TP_STRUCT__entry(
 		__field(unsigned long, opcode)
 		__field(unsigned long, retval)
diff --git a/arch/powerpc/kernel/eeh_driver.c b/arch/powerpc/kernel/eeh_driver.c
index 938742135ee0..650cfb31ea3d 100644
--- a/arch/powerpc/kernel/eeh_driver.c
+++ b/arch/powerpc/kernel/eeh_driver.c
@@ -418,8 +418,7 @@ static void *eeh_rmv_device(void *data, void *userdata)
 		eeh_pcid_put(dev);
 		if (driver->err_handler &&
 		    driver->err_handler->error_detected &&
-		    driver->err_handler->slot_reset &&
-		    driver->err_handler->resume)
+		    driver->err_handler->slot_reset)
 			return NULL;
 	}
 
@@ -564,6 +563,7 @@ static int eeh_reset_device(struct eeh_pe *pe, struct pci_bus *bus)
 	 */
 	eeh_pe_state_mark(pe, EEH_PE_KEEP);
 	if (bus) {
+		eeh_pe_state_clear(pe, EEH_PE_PRI_BUS);
 		pci_lock_rescan_remove();
 		pcibios_remove_pci_devices(bus);
 		pci_unlock_rescan_remove();
@@ -803,6 +803,7 @@ perm_error:
 	 * the their PCI config any more.
 	 */
 	if (frozen_bus) {
+		eeh_pe_state_clear(pe, EEH_PE_PRI_BUS);
 		eeh_pe_dev_mode_mark(pe, EEH_DEV_REMOVED);
 
 		pci_lock_rescan_remove();
@@ -886,6 +887,7 @@ static void eeh_handle_special_event(void)
 					continue;
 
 				/* Notify all devices to be down */
+				eeh_pe_state_clear(pe, EEH_PE_PRI_BUS);
 				bus = eeh_pe_bus_get(phb_pe);
 				eeh_pe_dev_traverse(pe,
 					eeh_report_failure, NULL);
diff --git a/arch/powerpc/kernel/eeh_pe.c b/arch/powerpc/kernel/eeh_pe.c
index ca9e5371930e..98f81800e00c 100644
--- a/arch/powerpc/kernel/eeh_pe.c
+++ b/arch/powerpc/kernel/eeh_pe.c
@@ -928,7 +928,7 @@ struct pci_bus *eeh_pe_bus_get(struct eeh_pe *pe)
 		bus = pe->phb->bus;
 	} else if (pe->type & EEH_PE_BUS ||
 		   pe->type & EEH_PE_DEVICE) {
-		if (pe->bus) {
+		if (pe->state & EEH_PE_PRI_BUS) {
 			bus = pe->bus;
 			goto out;
 		}
diff --git a/arch/powerpc/kernel/hw_breakpoint.c b/arch/powerpc/kernel/hw_breakpoint.c
index 05e804cdecaa..aec9a1b1d25b 100644
--- a/arch/powerpc/kernel/hw_breakpoint.c
+++ b/arch/powerpc/kernel/hw_breakpoint.c
@@ -109,8 +109,9 @@ void arch_unregister_hw_breakpoint(struct perf_event *bp)
 	 * If the breakpoint is unregistered between a hw_breakpoint_handler()
 	 * and the single_step_dabr_instruction(), then cleanup the breakpoint
 	 * restoration variables to prevent dangling pointers.
+	 * FIXME, this should not be using bp->ctx at all! Sayeth peterz.
 	 */
-	if (bp->ctx && bp->ctx->task)
+	if (bp->ctx && bp->ctx->task && bp->ctx->task != ((void *)-1L))
 		bp->ctx->task->thread.last_hit_ubp = NULL;
 }
 
diff --git a/arch/powerpc/kernel/module_64.c b/arch/powerpc/kernel/module_64.c
index ac64ffdb52c8..08b7a40de5f8 100644
--- a/arch/powerpc/kernel/module_64.c
+++ b/arch/powerpc/kernel/module_64.c
@@ -340,7 +340,7 @@ static void dedotify(Elf64_Sym *syms, unsigned int numsyms, char *strtab)
 			if (name[0] == '.') {
 				if (strcmp(name+1, "TOC.") == 0)
 					syms[i].st_shndx = SHN_ABS;
-				memmove(name, name+1, strlen(name));
+				syms[i].st_name++;
 			}
 		}
 	}
diff --git a/arch/powerpc/kernel/process.c b/arch/powerpc/kernel/process.c
index dccc87e8fee5..3c5736e52a14 100644
--- a/arch/powerpc/kernel/process.c
+++ b/arch/powerpc/kernel/process.c
@@ -1768,9 +1768,9 @@ static inline unsigned long brk_rnd(void)
 
 	/* 8MB for 32bit, 1GB for 64bit */
 	if (is_32bit_task())
-		rnd = (long)(get_random_int() % (1<<(23-PAGE_SHIFT)));
+		rnd = (get_random_long() % (1UL<<(23-PAGE_SHIFT)));
 	else
-		rnd = (long)(get_random_int() % (1<<(30-PAGE_SHIFT)));
+		rnd = (get_random_long() % (1UL<<(30-PAGE_SHIFT)));
 
 	return rnd << PAGE_SHIFT;
 }
diff --git a/arch/powerpc/kvm/book3s_hv_rmhandlers.S b/arch/powerpc/kvm/book3s_hv_rmhandlers.S
index 6ee26de9a1de..25ae2c9913c3 100644
--- a/arch/powerpc/kvm/book3s_hv_rmhandlers.S
+++ b/arch/powerpc/kvm/book3s_hv_rmhandlers.S
@@ -1370,6 +1370,20 @@ END_FTR_SECTION_IFCLR(CPU_FTR_ARCH_207S)
 	std	r6, VCPU_ACOP(r9)
 	stw	r7, VCPU_GUEST_PID(r9)
 	std	r8, VCPU_WORT(r9)
+	/*
+	 * Restore various registers to 0, where non-zero values
+	 * set by the guest could disrupt the host.
+	 */
+	li	r0, 0
+	mtspr	SPRN_IAMR, r0
+	mtspr	SPRN_CIABR, r0
+	mtspr	SPRN_DAWRX, r0
+	mtspr	SPRN_TCSCR, r0
+	mtspr	SPRN_WORT, r0
+	/* Set MMCRS to 1<<31 to freeze and disable the SPMC counters */
+	li	r0, 1
+	sldi	r0, r0, 31
+	mtspr	SPRN_MMCRS, r0
 8:
 
 	/* Save and reset AMR and UAMOR before turning on the MMU */
diff --git a/arch/powerpc/mm/hash64_64k.c b/arch/powerpc/mm/hash64_64k.c
index 0762c1e08c88..edb09912f0c9 100644
--- a/arch/powerpc/mm/hash64_64k.c
+++ b/arch/powerpc/mm/hash64_64k.c
@@ -111,7 +111,13 @@ int __hash_page_4K(unsigned long ea, unsigned long access, unsigned long vsid,
 	 */
 	if (!(old_pte & _PAGE_COMBO)) {
 		flush_hash_page(vpn, rpte, MMU_PAGE_64K, ssize, flags);
-		old_pte &= ~_PAGE_HASHPTE | _PAGE_F_GIX | _PAGE_F_SECOND;
+		/*
+		 * clear the old slot details from the old and new pte.
+		 * On hash insert failure we use old pte value and we don't
+		 * want slot information there if we have a insert failure.
+		 */
+		old_pte &= ~(_PAGE_HASHPTE | _PAGE_F_GIX | _PAGE_F_SECOND);
+		new_pte &= ~(_PAGE_HASHPTE | _PAGE_F_GIX | _PAGE_F_SECOND);
 		goto htab_insert_hpte;
 	}
 	/*
diff --git a/arch/powerpc/mm/hugepage-hash64.c b/arch/powerpc/mm/hugepage-hash64.c
index 49b152b0f926..eb2accdd76fd 100644
--- a/arch/powerpc/mm/hugepage-hash64.c
+++ b/arch/powerpc/mm/hugepage-hash64.c
@@ -78,9 +78,19 @@ int __hash_page_thp(unsigned long ea, unsigned long access, unsigned long vsid,
 		 * base page size. This is because demote_segment won't flush
 		 * hash page table entries.
 		 */
-		if ((old_pmd & _PAGE_HASHPTE) && !(old_pmd & _PAGE_COMBO))
+		if ((old_pmd & _PAGE_HASHPTE) && !(old_pmd & _PAGE_COMBO)) {
 			flush_hash_hugepage(vsid, ea, pmdp, MMU_PAGE_64K,
 					    ssize, flags);
+			/*
+			 * With THP, we also clear the slot information with
+			 * respect to all the 64K hash pte mapping the 16MB
+			 * page. They are all invalid now. This make sure we
+			 * don't find the slot valid when we fault with 4k
+			 * base page size.
+			 *
+			 */
+			memset(hpte_slot_array, 0, PTE_FRAG_SIZE);
+		}
 	}
 
 	valid = hpte_valid(hpte_slot_array, index);
diff --git a/arch/powerpc/mm/hugetlbpage-book3e.c b/arch/powerpc/mm/hugetlbpage-book3e.c
index 7e6d0880813f..83a8be791e06 100644
--- a/arch/powerpc/mm/hugetlbpage-book3e.c
+++ b/arch/powerpc/mm/hugetlbpage-book3e.c
@@ -8,6 +8,8 @@
 #include <linux/mm.h>
 #include <linux/hugetlb.h>
 
+#include <asm/mmu.h>
+
 #ifdef CONFIG_PPC_FSL_BOOK3E
 #ifdef CONFIG_PPC64
 static inline int tlb1_next(void)
@@ -60,6 +62,14 @@ static inline void book3e_tlb_lock(void)
 	unsigned long tmp;
 	int token = smp_processor_id() + 1;
 
+	/*
+	 * Besides being unnecessary in the absence of SMT, this
+	 * check prevents trying to do lbarx/stbcx. on e5500 which
+	 * doesn't implement either feature.
+	 */
+	if (!cpu_has_feature(CPU_FTR_SMT))
+		return;
+
 	asm volatile("1: lbarx %0, 0, %1;"
 		     "cmpwi %0, 0;"
 		     "bne 2f;"
@@ -80,6 +90,9 @@ static inline void book3e_tlb_unlock(void)
 {
 	struct paca_struct *paca = get_paca();
 
+	if (!cpu_has_feature(CPU_FTR_SMT))
+		return;
+
 	isync();
 	paca->tcd_ptr->lock = 0;
 }
diff --git a/arch/powerpc/mm/mmap.c b/arch/powerpc/mm/mmap.c
index 0f0502e12f6c..4087705ba90f 100644
--- a/arch/powerpc/mm/mmap.c
+++ b/arch/powerpc/mm/mmap.c
@@ -59,9 +59,9 @@ unsigned long arch_mmap_rnd(void)
 
 	/* 8MB for 32bit, 1GB for 64bit */
 	if (is_32bit_task())
-		rnd = (unsigned long)get_random_int() % (1<<(23-PAGE_SHIFT));
+		rnd = get_random_long() % (1<<(23-PAGE_SHIFT));
 	else
-		rnd = (unsigned long)get_random_int() % (1<<(30-PAGE_SHIFT));
+		rnd = get_random_long() % (1UL<<(30-PAGE_SHIFT));
 
 	return rnd << PAGE_SHIFT;
 }
diff --git a/arch/powerpc/mm/pgtable_64.c b/arch/powerpc/mm/pgtable_64.c
index 3124a20d0fab..cdf2123d46db 100644
--- a/arch/powerpc/mm/pgtable_64.c
+++ b/arch/powerpc/mm/pgtable_64.c
@@ -646,6 +646,28 @@ pgtable_t pgtable_trans_huge_withdraw(struct mm_struct *mm, pmd_t *pmdp)
 	return pgtable;
 }
 
+void pmdp_huge_split_prepare(struct vm_area_struct *vma,
+			     unsigned long address, pmd_t *pmdp)
+{
+	VM_BUG_ON(address & ~HPAGE_PMD_MASK);
+	VM_BUG_ON(REGION_ID(address) != USER_REGION_ID);
+
+	/*
+	 * We can't mark the pmd none here, because that will cause a race
+	 * against exit_mmap. We need to continue mark pmd TRANS HUGE, while
+	 * we spilt, but at the same time we wan't rest of the ppc64 code
+	 * not to insert hash pte on this, because we will be modifying
+	 * the deposited pgtable in the caller of this function. Hence
+	 * clear the _PAGE_USER so that we move the fault handling to
+	 * higher level function and that will serialize against ptl.
+	 * We need to flush existing hash pte entries here even though,
+	 * the translation is still valid, because we will withdraw
+	 * pgtable_t after this.
+	 */
+	pmd_hugepage_update(vma->vm_mm, address, pmdp, _PAGE_USER, 0);
+}
+
+
 /*
  * set a new huge pmd. We should not be called for updating
  * an existing pmd entry. That should go via pmd_hugepage_update.
@@ -663,10 +685,20 @@ void set_pmd_at(struct mm_struct *mm, unsigned long addr,
 	return set_pte_at(mm, addr, pmdp_ptep(pmdp), pmd_pte(pmd));
 }
 
+/*
+ * We use this to invalidate a pmdp entry before switching from a
+ * hugepte to regular pmd entry.
+ */
 void pmdp_invalidate(struct vm_area_struct *vma, unsigned long address,
 		     pmd_t *pmdp)
 {
 	pmd_hugepage_update(vma->vm_mm, address, pmdp, _PAGE_PRESENT, 0);
+
+	/*
+	 * This ensures that generic code that rely on IRQ disabling
+	 * to prevent a parallel THP split work as expected.
+	 */
+	kick_all_cpus_sync();
 }
 
 /*
diff --git a/arch/powerpc/platforms/powernv/eeh-powernv.c b/arch/powerpc/platforms/powernv/eeh-powernv.c
index 5f152b95ca0c..87f47e55aab6 100644
--- a/arch/powerpc/platforms/powernv/eeh-powernv.c
+++ b/arch/powerpc/platforms/powernv/eeh-powernv.c
@@ -444,9 +444,12 @@ static void *pnv_eeh_probe(struct pci_dn *pdn, void *data)
 	 * PCI devices of the PE are expected to be removed prior
 	 * to PE reset.
 	 */
-	if (!edev->pe->bus)
+	if (!(edev->pe->state & EEH_PE_PRI_BUS)) {
 		edev->pe->bus = pci_find_bus(hose->global_number,
 					     pdn->busno);
+		if (edev->pe->bus)
+			edev->pe->state |= EEH_PE_PRI_BUS;
+	}
 
 	/*
 	 * Enable EEH explicitly so that we will do EEH check
diff --git a/arch/powerpc/platforms/powernv/pci-ioda.c b/arch/powerpc/platforms/powernv/pci-ioda.c
index 573ae1994097..f90dc04395bf 100644
--- a/arch/powerpc/platforms/powernv/pci-ioda.c
+++ b/arch/powerpc/platforms/powernv/pci-ioda.c
@@ -3180,6 +3180,7 @@ static void pnv_pci_ioda_shutdown(struct pci_controller *hose)
 
 static const struct pci_controller_ops pnv_pci_ioda_controller_ops = {
        .dma_dev_setup = pnv_pci_dma_dev_setup,
+       .dma_bus_setup = pnv_pci_dma_bus_setup,
 #ifdef CONFIG_PCI_MSI
        .setup_msi_irqs = pnv_setup_msi_irqs,
        .teardown_msi_irqs = pnv_teardown_msi_irqs,
diff --git a/arch/powerpc/platforms/powernv/pci.c b/arch/powerpc/platforms/powernv/pci.c
index 2f55c86df703..b1ef84a6c9d1 100644
--- a/arch/powerpc/platforms/powernv/pci.c
+++ b/arch/powerpc/platforms/powernv/pci.c
@@ -599,6 +599,9 @@ int pnv_tce_build(struct iommu_table *tbl, long index, long npages,
 	u64 rpn = __pa(uaddr) >> tbl->it_page_shift;
 	long i;
 
+	if (proto_tce & TCE_PCI_WRITE)
+		proto_tce |= TCE_PCI_READ;
+
 	for (i = 0; i < npages; i++) {
 		unsigned long newtce = proto_tce |
 			((rpn + i) << tbl->it_page_shift);
@@ -620,6 +623,9 @@ int pnv_tce_xchg(struct iommu_table *tbl, long index,
 
 	BUG_ON(*hpa & ~IOMMU_PAGE_MASK(tbl));
 
+	if (newtce & TCE_PCI_WRITE)
+		newtce |= TCE_PCI_READ;
+
 	oldtce = xchg(pnv_tce(tbl, idx), cpu_to_be64(newtce));
 	*hpa = be64_to_cpu(oldtce) & ~(TCE_PCI_READ | TCE_PCI_WRITE);
 	*direction = iommu_tce_direction(oldtce);
@@ -760,6 +766,26 @@ void pnv_pci_dma_dev_setup(struct pci_dev *pdev)
 		phb->dma_dev_setup(phb, pdev);
 }
 
+void pnv_pci_dma_bus_setup(struct pci_bus *bus)
+{
+	struct pci_controller *hose = bus->sysdata;
+	struct pnv_phb *phb = hose->private_data;
+	struct pnv_ioda_pe *pe;
+
+	list_for_each_entry(pe, &phb->ioda.pe_list, list) {
+		if (!(pe->flags & (PNV_IODA_PE_BUS | PNV_IODA_PE_BUS_ALL)))
+			continue;
+
+		if (!pe->pbus)
+			continue;
+
+		if (bus->number == ((pe->rid >> 8) & 0xFF)) {
+			pe->pbus = bus;
+			break;
+		}
+	}
+}
+
 void pnv_pci_shutdown(void)
 {
 	struct pci_controller *hose;
diff --git a/arch/powerpc/platforms/powernv/pci.h b/arch/powerpc/platforms/powernv/pci.h
index 7f56313e8d72..00691a9b99af 100644
--- a/arch/powerpc/platforms/powernv/pci.h
+++ b/arch/powerpc/platforms/powernv/pci.h
@@ -242,6 +242,7 @@ extern void pnv_pci_reset_secondary_bus(struct pci_dev *dev);
 extern int pnv_eeh_phb_reset(struct pci_controller *hose, int option);
 
 extern void pnv_pci_dma_dev_setup(struct pci_dev *pdev);
+extern void pnv_pci_dma_bus_setup(struct pci_bus *bus);
 extern int pnv_setup_msi_irqs(struct pci_dev *pdev, int nvec, int type);
 extern void pnv_teardown_msi_irqs(struct pci_dev *pdev);
 
diff --git a/arch/s390/include/asm/fpu/internal.h b/arch/s390/include/asm/fpu/internal.h
index ea91ddfe54eb..629c90865a07 100644
--- a/arch/s390/include/asm/fpu/internal.h
+++ b/arch/s390/include/asm/fpu/internal.h
@@ -40,6 +40,7 @@ static inline void convert_fp_to_vx(__vector128 *vxrs, freg_t *fprs)
 static inline void fpregs_store(_s390_fp_regs *fpregs, struct fpu *fpu)
 {
 	fpregs->pad = 0;
+	fpregs->fpc = fpu->fpc;
 	if (MACHINE_HAS_VX)
 		convert_vx_to_fp((freg_t *)&fpregs->fprs, fpu->vxrs);
 	else
@@ -49,6 +50,7 @@ static inline void fpregs_store(_s390_fp_regs *fpregs, struct fpu *fpu)
 
 static inline void fpregs_load(_s390_fp_regs *fpregs, struct fpu *fpu)
 {
+	fpu->fpc = fpregs->fpc;
 	if (MACHINE_HAS_VX)
 		convert_fp_to_vx(fpu->vxrs, (freg_t *)&fpregs->fprs);
 	else
diff --git a/arch/s390/include/asm/livepatch.h b/arch/s390/include/asm/livepatch.h
index 7aa799134a11..a52b6cca873d 100644
--- a/arch/s390/include/asm/livepatch.h
+++ b/arch/s390/include/asm/livepatch.h
@@ -37,7 +37,7 @@ static inline void klp_arch_set_pc(struct pt_regs *regs, unsigned long ip)
 	regs->psw.addr = ip;
 }
 #else
-#error Live patching support is disabled; check CONFIG_LIVEPATCH
+#error Include linux/livepatch.h, not asm/livepatch.h
 #endif
 
 #endif
diff --git a/arch/s390/include/asm/mmu_context.h b/arch/s390/include/asm/mmu_context.h
index fb1b93ea3e3f..e485817f7b1a 100644
--- a/arch/s390/include/asm/mmu_context.h
+++ b/arch/s390/include/asm/mmu_context.h
@@ -15,17 +15,25 @@
 static inline int init_new_context(struct task_struct *tsk,
 				   struct mm_struct *mm)
 {
+	spin_lock_init(&mm->context.list_lock);
+	INIT_LIST_HEAD(&mm->context.pgtable_list);
+	INIT_LIST_HEAD(&mm->context.gmap_list);
 	cpumask_clear(&mm->context.cpu_attach_mask);
 	atomic_set(&mm->context.attach_count, 0);
 	mm->context.flush_mm = 0;
-	mm->context.asce_bits = _ASCE_TABLE_LENGTH | _ASCE_USER_BITS;
-	mm->context.asce_bits |= _ASCE_TYPE_REGION3;
 #ifdef CONFIG_PGSTE
 	mm->context.alloc_pgste = page_table_allocate_pgste;
 	mm->context.has_pgste = 0;
 	mm->context.use_skey = 0;
 #endif
-	mm->context.asce_limit = STACK_TOP_MAX;
+	if (mm->context.asce_limit == 0) {
+		/* context created by exec, set asce limit to 4TB */
+		mm->context.asce_bits = _ASCE_TABLE_LENGTH |
+			_ASCE_USER_BITS | _ASCE_TYPE_REGION3;
+		mm->context.asce_limit = STACK_TOP_MAX;
+	} else if (mm->context.asce_limit == (1UL << 31)) {
+		mm_inc_nr_pmds(mm);
+	}
 	crst_table_init((unsigned long *) mm->pgd, pgd_entry_type(mm));
 	return 0;
 }
@@ -111,8 +119,6 @@ static inline void activate_mm(struct mm_struct *prev,
 static inline void arch_dup_mmap(struct mm_struct *oldmm,
 				 struct mm_struct *mm)
 {
-	if (oldmm->context.asce_limit < mm->context.asce_limit)
-		crst_table_downgrade(mm, oldmm->context.asce_limit);
 }
 
 static inline void arch_exit_mmap(struct mm_struct *mm)
diff --git a/arch/s390/include/asm/pgalloc.h b/arch/s390/include/asm/pgalloc.h
index 7b7858f158b4..d7cc79fb6191 100644
--- a/arch/s390/include/asm/pgalloc.h
+++ b/arch/s390/include/asm/pgalloc.h
@@ -100,12 +100,26 @@ static inline void pud_populate(struct mm_struct *mm, pud_t *pud, pmd_t *pmd)
 
 static inline pgd_t *pgd_alloc(struct mm_struct *mm)
 {
-	spin_lock_init(&mm->context.list_lock);
-	INIT_LIST_HEAD(&mm->context.pgtable_list);
-	INIT_LIST_HEAD(&mm->context.gmap_list);
-	return (pgd_t *) crst_table_alloc(mm);
+	unsigned long *table = crst_table_alloc(mm);
+
+	if (!table)
+		return NULL;
+	if (mm->context.asce_limit == (1UL << 31)) {
+		/* Forking a compat process with 2 page table levels */
+		if (!pgtable_pmd_page_ctor(virt_to_page(table))) {
+			crst_table_free(mm, table);
+			return NULL;
+		}
+	}
+	return (pgd_t *) table;
+}
+
+static inline void pgd_free(struct mm_struct *mm, pgd_t *pgd)
+{
+	if (mm->context.asce_limit == (1UL << 31))
+		pgtable_pmd_page_dtor(virt_to_page(pgd));
+	crst_table_free(mm, (unsigned long *) pgd);
 }
-#define pgd_free(mm, pgd) crst_table_free(mm, (unsigned long *) pgd)
 
 static inline void pmd_populate(struct mm_struct *mm,
 				pmd_t *pmd, pgtable_t pte)
diff --git a/arch/s390/kernel/compat_signal.c b/arch/s390/kernel/compat_signal.c
index 66c94417c0ba..4af60374eba0 100644
--- a/arch/s390/kernel/compat_signal.c
+++ b/arch/s390/kernel/compat_signal.c
@@ -271,7 +271,7 @@ static int restore_sigregs_ext32(struct pt_regs *regs,
 
 	/* Restore high gprs from signal stack */
 	if (__copy_from_user(&gprs_high, &sregs_ext->gprs_high,
-			     sizeof(&sregs_ext->gprs_high)))
+			     sizeof(sregs_ext->gprs_high)))
 		return -EFAULT;
 	for (i = 0; i < NUM_GPRS; i++)
 		*(__u32 *)&regs->gprs[i] = gprs_high[i];
diff --git a/arch/s390/kernel/head64.S b/arch/s390/kernel/head64.S
index c5febe84eba6..03c2b469c472 100644
--- a/arch/s390/kernel/head64.S
+++ b/arch/s390/kernel/head64.S
@@ -16,7 +16,7 @@
 
 __HEAD
 ENTRY(startup_continue)
-	tm	__LC_STFLE_FAC_LIST+6,0x80	# LPP available ?
+	tm	__LC_STFLE_FAC_LIST+5,0x80	# LPP available ?
 	jz	0f
 	xc	__LC_LPP+1(7,0),__LC_LPP+1	# clear lpp and current_pid
 	mvi	__LC_LPP,0x80			#   and set LPP_MAGIC
diff --git a/arch/s390/kernel/perf_event.c b/arch/s390/kernel/perf_event.c
index cfcba2dd9bb5..0943b11a2f6e 100644
--- a/arch/s390/kernel/perf_event.c
+++ b/arch/s390/kernel/perf_event.c
@@ -260,12 +260,13 @@ static unsigned long __store_trace(struct perf_callchain_entry *entry,
 void perf_callchain_kernel(struct perf_callchain_entry *entry,
 			   struct pt_regs *regs)
 {
-	unsigned long head;
+	unsigned long head, frame_size;
 	struct stack_frame *head_sf;
 
 	if (user_mode(regs))
 		return;
 
+	frame_size = STACK_FRAME_OVERHEAD + sizeof(struct pt_regs);
 	head = regs->gprs[15];
 	head_sf = (struct stack_frame *) head;
 
@@ -273,8 +274,9 @@ void perf_callchain_kernel(struct perf_callchain_entry *entry,
 		return;
 
 	head = head_sf->back_chain;
-	head = __store_trace(entry, head, S390_lowcore.async_stack - ASYNC_SIZE,
-			     S390_lowcore.async_stack);
+	head = __store_trace(entry, head,
+			     S390_lowcore.async_stack + frame_size - ASYNC_SIZE,
+			     S390_lowcore.async_stack + frame_size);
 
 	__store_trace(entry, head, S390_lowcore.thread_info,
 		      S390_lowcore.thread_info + THREAD_SIZE);
diff --git a/arch/s390/kernel/stacktrace.c b/arch/s390/kernel/stacktrace.c
index 5acba3cb7220..8f64ebd63767 100644
--- a/arch/s390/kernel/stacktrace.c
+++ b/arch/s390/kernel/stacktrace.c
@@ -59,26 +59,32 @@ static unsigned long save_context_stack(struct stack_trace *trace,
 	}
 }
 
-void save_stack_trace(struct stack_trace *trace)
+static void __save_stack_trace(struct stack_trace *trace, unsigned long sp)
 {
-	register unsigned long sp asm ("15");
-	unsigned long orig_sp, new_sp;
+	unsigned long new_sp, frame_size;
 
-	orig_sp = sp;
-	new_sp = save_context_stack(trace, orig_sp,
-				    S390_lowcore.panic_stack - PAGE_SIZE,
-				    S390_lowcore.panic_stack, 1);
-	if (new_sp != orig_sp)
-		return;
+	frame_size = STACK_FRAME_OVERHEAD + sizeof(struct pt_regs);
+	new_sp = save_context_stack(trace, sp,
+			S390_lowcore.panic_stack + frame_size - PAGE_SIZE,
+			S390_lowcore.panic_stack + frame_size, 1);
 	new_sp = save_context_stack(trace, new_sp,
-				    S390_lowcore.async_stack - ASYNC_SIZE,
-				    S390_lowcore.async_stack, 1);
-	if (new_sp != orig_sp)
-		return;
+			S390_lowcore.async_stack + frame_size - ASYNC_SIZE,
+			S390_lowcore.async_stack + frame_size, 1);
 	save_context_stack(trace, new_sp,
 			   S390_lowcore.thread_info,
 			   S390_lowcore.thread_info + THREAD_SIZE, 1);
 }
+
+void save_stack_trace(struct stack_trace *trace)
+{
+	register unsigned long r15 asm ("15");
+	unsigned long sp;
+
+	sp = r15;
+	__save_stack_trace(trace, sp);
+	if (trace->nr_entries < trace->max_entries)
+		trace->entries[trace->nr_entries++] = ULONG_MAX;
+}
 EXPORT_SYMBOL_GPL(save_stack_trace);
 
 void save_stack_trace_tsk(struct task_struct *tsk, struct stack_trace *trace)
@@ -86,6 +92,10 @@ void save_stack_trace_tsk(struct task_struct *tsk, struct stack_trace *trace)
 	unsigned long sp, low, high;
 
 	sp = tsk->thread.ksp;
+	if (tsk == current) {
+		/* Get current stack pointer. */
+		asm volatile("la %0,0(15)" : "=a" (sp));
+	}
 	low = (unsigned long) task_stack_page(tsk);
 	high = (unsigned long) task_pt_regs(tsk);
 	save_context_stack(trace, sp, low, high, 0);
@@ -93,3 +103,14 @@ void save_stack_trace_tsk(struct task_struct *tsk, struct stack_trace *trace)
 		trace->entries[trace->nr_entries++] = ULONG_MAX;
 }
 EXPORT_SYMBOL_GPL(save_stack_trace_tsk);
+
+void save_stack_trace_regs(struct pt_regs *regs, struct stack_trace *trace)
+{
+	unsigned long sp;
+
+	sp = kernel_stack_pointer(regs);
+	__save_stack_trace(trace, sp);
+	if (trace->nr_entries < trace->max_entries)
+		trace->entries[trace->nr_entries++] = ULONG_MAX;
+}
+EXPORT_SYMBOL_GPL(save_stack_trace_regs);
diff --git a/arch/s390/kernel/trace.c b/arch/s390/kernel/trace.c
index 21a5df99552b..dde7654f5c68 100644
--- a/arch/s390/kernel/trace.c
+++ b/arch/s390/kernel/trace.c
@@ -18,6 +18,9 @@ void trace_s390_diagnose_norecursion(int diag_nr)
 	unsigned long flags;
 	unsigned int *depth;
 
+	/* Avoid lockdep recursion. */
+	if (IS_ENABLED(CONFIG_LOCKDEP))
+		return;
 	local_irq_save(flags);
 	depth = this_cpu_ptr(&diagnose_trace_depth);
 	if (*depth == 0) {
diff --git a/arch/s390/kvm/kvm-s390.c b/arch/s390/kvm/kvm-s390.c
index 4af21c771f9b..03dfe9c667f4 100644
--- a/arch/s390/kvm/kvm-s390.c
+++ b/arch/s390/kvm/kvm-s390.c
@@ -2381,7 +2381,7 @@ int kvm_s390_store_status_unloaded(struct kvm_vcpu *vcpu, unsigned long gpa)
 
 	/* manually convert vector registers if necessary */
 	if (MACHINE_HAS_VX) {
-		convert_vx_to_fp(fprs, current->thread.fpu.vxrs);
+		convert_vx_to_fp(fprs, (__vector128 *) vcpu->run->s.regs.vrs);
 		rc = write_guest_abs(vcpu, gpa + __LC_FPREGS_SAVE_AREA,
 				     fprs, 128);
 	} else {
diff --git a/arch/s390/mm/maccess.c b/arch/s390/mm/maccess.c
index fec59c067d0d..792f9c63fbca 100644
--- a/arch/s390/mm/maccess.c
+++ b/arch/s390/mm/maccess.c
@@ -93,15 +93,19 @@ static int __memcpy_real(void *dest, void *src, size_t count)
  */
 int memcpy_real(void *dest, void *src, size_t count)
 {
+	int irqs_disabled, rc;
 	unsigned long flags;
-	int rc;
 
 	if (!count)
 		return 0;
-	local_irq_save(flags);
-	__arch_local_irq_stnsm(0xfbUL);
+	flags = __arch_local_irq_stnsm(0xf8UL);
+	irqs_disabled = arch_irqs_disabled_flags(flags);
+	if (!irqs_disabled)
+		trace_hardirqs_off();
 	rc = __memcpy_real(dest, src, count);
-	local_irq_restore(flags);
+	if (!irqs_disabled)
+		trace_hardirqs_on();
+	__arch_local_irq_ssm(flags);
 	return rc;
 }
 
diff --git a/arch/s390/oprofile/backtrace.c b/arch/s390/oprofile/backtrace.c
index fe0bfe370c45..1884e1759529 100644
--- a/arch/s390/oprofile/backtrace.c
+++ b/arch/s390/oprofile/backtrace.c
@@ -54,12 +54,13 @@ __show_trace(unsigned int *depth, unsigned long sp,
 
 void s390_backtrace(struct pt_regs * const regs, unsigned int depth)
 {
-	unsigned long head;
+	unsigned long head, frame_size;
 	struct stack_frame* head_sf;
 
 	if (user_mode(regs))
 		return;
 
+	frame_size = STACK_FRAME_OVERHEAD + sizeof(struct pt_regs);
 	head = regs->gprs[15];
 	head_sf = (struct stack_frame*)head;
 
@@ -68,8 +69,9 @@ void s390_backtrace(struct pt_regs * const regs, unsigned int depth)
 
 	head = head_sf->back_chain;
 
-	head = __show_trace(&depth, head, S390_lowcore.async_stack - ASYNC_SIZE,
-			    S390_lowcore.async_stack);
+	head = __show_trace(&depth, head,
+			    S390_lowcore.async_stack + frame_size - ASYNC_SIZE,
+			    S390_lowcore.async_stack + frame_size);
 
 	__show_trace(&depth, head, S390_lowcore.thread_info,
 		     S390_lowcore.thread_info + THREAD_SIZE);
diff --git a/arch/sparc/Makefile b/arch/sparc/Makefile
index eaee14637d93..8496a074bd0e 100644
--- a/arch/sparc/Makefile
+++ b/arch/sparc/Makefile
@@ -24,7 +24,13 @@ LDFLAGS        := -m elf32_sparc
 export BITS    := 32
 UTS_MACHINE    := sparc
 
+# We are adding -Wa,-Av8 to KBUILD_CFLAGS to deal with a specs bug in some
+# versions of gcc.  Some gcc versions won't pass -Av8 to binutils when you
+# give -mcpu=v8.  This silently worked with older bintutils versions but
+# does not any more.
 KBUILD_CFLAGS  += -m32 -mcpu=v8 -pipe -mno-fpu -fcall-used-g5 -fcall-used-g7
+KBUILD_CFLAGS  += -Wa,-Av8
+
 KBUILD_AFLAGS  += -m32 -Wa,-Av8
 
 else
diff --git a/arch/sparc/include/uapi/asm/unistd.h b/arch/sparc/include/uapi/asm/unistd.h
index 1c26d440d288..b6de8b10a55b 100644
--- a/arch/sparc/include/uapi/asm/unistd.h
+++ b/arch/sparc/include/uapi/asm/unistd.h
@@ -422,8 +422,9 @@
 #define __NR_listen		354
 #define __NR_setsockopt		355
 #define __NR_mlock2		356
+#define __NR_copy_file_range	357
 
-#define NR_syscalls		357
+#define NR_syscalls		358
 
 /* Bitmask values returned from kern_features system call.  */
 #define KERN_FEATURE_MIXED_MODE_STACK	0x00000001
diff --git a/arch/sparc/kernel/entry.S b/arch/sparc/kernel/entry.S
index 33c02b15f478..a83707c83be8 100644
--- a/arch/sparc/kernel/entry.S
+++ b/arch/sparc/kernel/entry.S
@@ -948,7 +948,24 @@ linux_syscall_trace:
 	cmp	%o0, 0
 	bne	3f
 	 mov	-ENOSYS, %o0
+
+	/* Syscall tracing can modify the registers.  */
+	ld	[%sp + STACKFRAME_SZ + PT_G1], %g1
+	sethi	%hi(sys_call_table), %l7
+	ld	[%sp + STACKFRAME_SZ + PT_I0], %i0
+	or	%l7, %lo(sys_call_table), %l7
+	ld	[%sp + STACKFRAME_SZ + PT_I1], %i1
+	ld	[%sp + STACKFRAME_SZ + PT_I2], %i2
+	ld	[%sp + STACKFRAME_SZ + PT_I3], %i3
+	ld	[%sp + STACKFRAME_SZ + PT_I4], %i4
+	ld	[%sp + STACKFRAME_SZ + PT_I5], %i5
+	cmp	%g1, NR_syscalls
+	bgeu	3f
+	 mov	-ENOSYS, %o0
+
+	sll	%g1, 2, %l4
 	mov	%i0, %o0
+	ld	[%l7 + %l4], %l7
 	mov	%i1, %o1
 	mov	%i2, %o2
 	mov	%i3, %o3
diff --git a/arch/sparc/kernel/hvcalls.S b/arch/sparc/kernel/hvcalls.S
index afbaba52d2f1..d127130bf424 100644
--- a/arch/sparc/kernel/hvcalls.S
+++ b/arch/sparc/kernel/hvcalls.S
@@ -338,8 +338,9 @@ ENTRY(sun4v_mach_set_watchdog)
 	mov	%o1, %o4
 	mov	HV_FAST_MACH_SET_WATCHDOG, %o5
 	ta	HV_FAST_TRAP
+	brnz,a,pn %o4, 0f
 	stx	%o1, [%o4]
-	retl
+0:	retl
 	 nop
 ENDPROC(sun4v_mach_set_watchdog)
 
diff --git a/arch/sparc/kernel/signal_64.c b/arch/sparc/kernel/signal_64.c
index d88beff47bab..39aaec173f66 100644
--- a/arch/sparc/kernel/signal_64.c
+++ b/arch/sparc/kernel/signal_64.c
@@ -52,7 +52,7 @@ asmlinkage void sparc64_set_context(struct pt_regs *regs)
 	unsigned char fenab;
 	int err;
 
-	flush_user_windows();
+	synchronize_user_stack();
 	if (get_thread_wsaved()					||
 	    (((unsigned long)ucp) & (sizeof(unsigned long)-1))	||
 	    (!__access_ok(ucp, sizeof(*ucp))))
diff --git a/arch/sparc/kernel/sparc_ksyms_64.c b/arch/sparc/kernel/sparc_ksyms_64.c
index a92d5d2c46a3..9e034f29dcc5 100644
--- a/arch/sparc/kernel/sparc_ksyms_64.c
+++ b/arch/sparc/kernel/sparc_ksyms_64.c
@@ -37,6 +37,7 @@ EXPORT_SYMBOL(sun4v_niagara_getperf);
 EXPORT_SYMBOL(sun4v_niagara_setperf);
 EXPORT_SYMBOL(sun4v_niagara2_getperf);
 EXPORT_SYMBOL(sun4v_niagara2_setperf);
+EXPORT_SYMBOL(sun4v_mach_set_watchdog);
 
 /* from hweight.S */
 EXPORT_SYMBOL(__arch_hweight8);
diff --git a/arch/sparc/kernel/sys_sparc_64.c b/arch/sparc/kernel/sys_sparc_64.c
index c690c8e16a96..b489e9759518 100644
--- a/arch/sparc/kernel/sys_sparc_64.c
+++ b/arch/sparc/kernel/sys_sparc_64.c
@@ -264,7 +264,7 @@ static unsigned long mmap_rnd(void)
 	unsigned long rnd = 0UL;
 
 	if (current->flags & PF_RANDOMIZE) {
-		unsigned long val = get_random_int();
+		unsigned long val = get_random_long();
 		if (test_thread_flag(TIF_32BIT))
 			rnd = (val % (1UL << (23UL-PAGE_SHIFT)));
 		else
diff --git a/arch/sparc/kernel/syscalls.S b/arch/sparc/kernel/syscalls.S
index bb0008927598..c4a1b5c40e4e 100644
--- a/arch/sparc/kernel/syscalls.S
+++ b/arch/sparc/kernel/syscalls.S
@@ -158,7 +158,25 @@ linux_syscall_trace32:
 	 add	%sp, PTREGS_OFF, %o0
 	brnz,pn	%o0, 3f
 	 mov	-ENOSYS, %o0
+
+	/* Syscall tracing can modify the registers.  */
+	ldx	[%sp + PTREGS_OFF + PT_V9_G1], %g1
+	sethi	%hi(sys_call_table32), %l7
+	ldx	[%sp + PTREGS_OFF + PT_V9_I0], %i0
+	or	%l7, %lo(sys_call_table32), %l7
+	ldx	[%sp + PTREGS_OFF + PT_V9_I1], %i1
+	ldx	[%sp + PTREGS_OFF + PT_V9_I2], %i2
+	ldx	[%sp + PTREGS_OFF + PT_V9_I3], %i3
+	ldx	[%sp + PTREGS_OFF + PT_V9_I4], %i4
+	ldx	[%sp + PTREGS_OFF + PT_V9_I5], %i5
+
+	cmp	%g1, NR_syscalls
+	bgeu,pn	%xcc, 3f
+	 mov	-ENOSYS, %o0
+
+	sll	%g1, 2, %l4
 	srl	%i0, 0, %o0
+	lduw	[%l7 + %l4], %l7
 	srl	%i4, 0, %o4
 	srl	%i1, 0, %o1
 	srl	%i2, 0, %o2
@@ -170,7 +188,25 @@ linux_syscall_trace:
 	 add	%sp, PTREGS_OFF, %o0
 	brnz,pn	%o0, 3f
 	 mov	-ENOSYS, %o0
+
+	/* Syscall tracing can modify the registers.  */
+	ldx	[%sp + PTREGS_OFF + PT_V9_G1], %g1
+	sethi	%hi(sys_call_table64), %l7
+	ldx	[%sp + PTREGS_OFF + PT_V9_I0], %i0
+	or	%l7, %lo(sys_call_table64), %l7
+	ldx	[%sp + PTREGS_OFF + PT_V9_I1], %i1
+	ldx	[%sp + PTREGS_OFF + PT_V9_I2], %i2
+	ldx	[%sp + PTREGS_OFF + PT_V9_I3], %i3
+	ldx	[%sp + PTREGS_OFF + PT_V9_I4], %i4
+	ldx	[%sp + PTREGS_OFF + PT_V9_I5], %i5
+
+	cmp	%g1, NR_syscalls
+	bgeu,pn	%xcc, 3f
+	 mov	-ENOSYS, %o0
+
+	sll	%g1, 2, %l4
 	mov	%i0, %o0
+	lduw	[%l7 + %l4], %l7
 	mov	%i1, %o1
 	mov	%i2, %o2
 	mov	%i3, %o3
diff --git a/arch/sparc/kernel/systbls_32.S b/arch/sparc/kernel/systbls_32.S
index e663b6c78de2..6c3dd6c52f8b 100644
--- a/arch/sparc/kernel/systbls_32.S
+++ b/arch/sparc/kernel/systbls_32.S
@@ -88,4 +88,4 @@ sys_call_table:
 /*340*/	.long sys_ni_syscall, sys_kcmp, sys_finit_module, sys_sched_setattr, sys_sched_getattr
 /*345*/	.long sys_renameat2, sys_seccomp, sys_getrandom, sys_memfd_create, sys_bpf
 /*350*/	.long sys_execveat, sys_membarrier, sys_userfaultfd, sys_bind, sys_listen
-/*355*/	.long sys_setsockopt, sys_mlock2
+/*355*/	.long sys_setsockopt, sys_mlock2, sys_copy_file_range
diff --git a/arch/sparc/kernel/systbls_64.S b/arch/sparc/kernel/systbls_64.S
index 1557121f4cdc..12b524cfcfa0 100644
--- a/arch/sparc/kernel/systbls_64.S
+++ b/arch/sparc/kernel/systbls_64.S
@@ -89,7 +89,7 @@ sys_call_table32:
 /*340*/	.word sys_kern_features, sys_kcmp, sys_finit_module, sys_sched_setattr, sys_sched_getattr
 	.word sys32_renameat2, sys_seccomp, sys_getrandom, sys_memfd_create, sys_bpf
 /*350*/	.word sys32_execveat, sys_membarrier, sys_userfaultfd, sys_bind, sys_listen
-	.word compat_sys_setsockopt, sys_mlock2
+	.word compat_sys_setsockopt, sys_mlock2, sys_copy_file_range
 
 #endif /* CONFIG_COMPAT */
 
@@ -170,4 +170,4 @@ sys_call_table:
 /*340*/	.word sys_kern_features, sys_kcmp, sys_finit_module, sys_sched_setattr, sys_sched_getattr
 	.word sys_renameat2, sys_seccomp, sys_getrandom, sys_memfd_create, sys_bpf
 /*350*/	.word sys64_execveat, sys_membarrier, sys_userfaultfd, sys_bind, sys_listen
-	.word sys_setsockopt, sys_mlock2
+	.word sys_setsockopt, sys_mlock2, sys_copy_file_range
diff --git a/arch/um/kernel/reboot.c b/arch/um/kernel/reboot.c
index 9bdf67a092a5..b60a9f8cda75 100644
--- a/arch/um/kernel/reboot.c
+++ b/arch/um/kernel/reboot.c
@@ -12,6 +12,7 @@
 #include <skas.h>
 
 void (*pm_power_off)(void);
+EXPORT_SYMBOL(pm_power_off);
 
 static void kill_off_processes(void)
 {
diff --git a/arch/um/kernel/signal.c b/arch/um/kernel/signal.c
index fc8be0e3a4ff..57acbd67d85d 100644
--- a/arch/um/kernel/signal.c
+++ b/arch/um/kernel/signal.c
@@ -69,7 +69,7 @@ void do_signal(struct pt_regs *regs)
 	struct ksignal ksig;
 	int handled_sig = 0;
 
-	if (get_signal(&ksig)) {
+	while (get_signal(&ksig)) {
 		handled_sig = 1;
 		/* Whee!  Actually deliver the signal.  */
 		handle_signal(&ksig, regs);
diff --git a/arch/x86/Kconfig b/arch/x86/Kconfig
index 9af2e6338400..c46662f64c39 100644
--- a/arch/x86/Kconfig
+++ b/arch/x86/Kconfig
@@ -475,6 +475,7 @@ config X86_UV
 	depends on X86_64
 	depends on X86_EXTENDED_PLATFORM
 	depends on NUMA
+	depends on EFI
 	depends on X86_X2APIC
 	depends on PCI
 	---help---
@@ -777,8 +778,8 @@ config HPET_TIMER
 	  HPET is the next generation timer replacing legacy 8254s.
 	  The HPET provides a stable time base on SMP
 	  systems, unlike the TSC, but it is more expensive to access,
-	  as it is off-chip.  You can find the HPET spec at
-	  <http://www.intel.com/hardwaredesign/hpetspec_1.pdf>.
+	  as it is off-chip.  The interface used is documented
+	  in the HPET spec, revision 1.
 
 	  You can safely choose Y here.  However, HPET will only be
 	  activated if the platform and the BIOS support this feature.
diff --git a/arch/x86/entry/entry_32.S b/arch/x86/entry/entry_32.S
index 77d8c5112900..bb3e376d0f33 100644
--- a/arch/x86/entry/entry_32.S
+++ b/arch/x86/entry/entry_32.S
@@ -294,6 +294,7 @@ sysenter_past_esp:
 	pushl	$__USER_DS		/* pt_regs->ss */
 	pushl	%ebp			/* pt_regs->sp (stashed in bp) */
 	pushfl				/* pt_regs->flags (except IF = 0) */
+	ASM_CLAC			/* Clear AC after saving FLAGS */
 	orl	$X86_EFLAGS_IF, (%esp)	/* Fix IF */
 	pushl	$__USER_CS		/* pt_regs->cs */
 	pushl	$0			/* pt_regs->ip = 0 (placeholder) */
diff --git a/arch/x86/entry/entry_64_compat.S b/arch/x86/entry/entry_64_compat.S
index ff1c6d61f332..3c990eeee40b 100644
--- a/arch/x86/entry/entry_64_compat.S
+++ b/arch/x86/entry/entry_64_compat.S
@@ -261,6 +261,7 @@ ENTRY(entry_INT80_compat)
 	 * Interrupts are off on entry.
 	 */
 	PARAVIRT_ADJUST_EXCEPTION_FRAME
+	ASM_CLAC			/* Do this early to minimize exposure */
 	SWAPGS
 
 	/*
diff --git a/arch/x86/include/asm/livepatch.h b/arch/x86/include/asm/livepatch.h
index 19c099afa861..e795f5274217 100644
--- a/arch/x86/include/asm/livepatch.h
+++ b/arch/x86/include/asm/livepatch.h
@@ -41,7 +41,7 @@ static inline void klp_arch_set_pc(struct pt_regs *regs, unsigned long ip)
 	regs->ip = ip;
 }
 #else
-#error Live patching support is disabled; check CONFIG_LIVEPATCH
+#error Include linux/livepatch.h, not asm/livepatch.h
 #endif
 
 #endif /* _ASM_X86_LIVEPATCH_H */
diff --git a/arch/x86/include/asm/pci_x86.h b/arch/x86/include/asm/pci_x86.h
index 46873fbd44e1..d08eacd298c2 100644
--- a/arch/x86/include/asm/pci_x86.h
+++ b/arch/x86/include/asm/pci_x86.h
@@ -93,6 +93,8 @@ extern raw_spinlock_t pci_config_lock;
 extern int (*pcibios_enable_irq)(struct pci_dev *dev);
 extern void (*pcibios_disable_irq)(struct pci_dev *dev);
 
+extern bool mp_should_keep_irq(struct device *dev);
+
 struct pci_raw_ops {
 	int (*read)(unsigned int domain, unsigned int bus, unsigned int devfn,
 						int reg, int len, u32 *val);
diff --git a/arch/x86/include/asm/processor.h b/arch/x86/include/asm/processor.h
index 2d5a50cb61a2..20c11d1aa4cc 100644
--- a/arch/x86/include/asm/processor.h
+++ b/arch/x86/include/asm/processor.h
@@ -766,7 +766,7 @@ extern unsigned long thread_saved_pc(struct task_struct *tsk);
  * Return saved PC of a blocked thread.
  * What is this good for? it will be always the scheduler or ret_from_fork.
  */
-#define thread_saved_pc(t)	(*(unsigned long *)((t)->thread.sp - 8))
+#define thread_saved_pc(t)	READ_ONCE_NOCHECK(*(unsigned long *)((t)->thread.sp - 8))
 
 #define task_pt_regs(tsk)	((struct pt_regs *)(tsk)->thread.sp0 - 1)
 extern unsigned long KSTK_ESP(struct task_struct *task);
diff --git a/arch/x86/include/asm/uaccess_32.h b/arch/x86/include/asm/uaccess_32.h
index f5dcb5204dcd..3fe0eac59462 100644
--- a/arch/x86/include/asm/uaccess_32.h
+++ b/arch/x86/include/asm/uaccess_32.h
@@ -48,20 +48,28 @@ __copy_to_user_inatomic(void __user *to, const void *from, unsigned long n)
 
 		switch (n) {
 		case 1:
+			__uaccess_begin();
 			__put_user_size(*(u8 *)from, (u8 __user *)to,
 					1, ret, 1);
+			__uaccess_end();
 			return ret;
 		case 2:
+			__uaccess_begin();
 			__put_user_size(*(u16 *)from, (u16 __user *)to,
 					2, ret, 2);
+			__uaccess_end();
 			return ret;
 		case 4:
+			__uaccess_begin();
 			__put_user_size(*(u32 *)from, (u32 __user *)to,
 					4, ret, 4);
+			__uaccess_end();
 			return ret;
 		case 8:
+			__uaccess_begin();
 			__put_user_size(*(u64 *)from, (u64 __user *)to,
 					8, ret, 8);
+			__uaccess_end();
 			return ret;
 		}
 	}
@@ -103,13 +111,19 @@ __copy_from_user_inatomic(void *to, const void __user *from, unsigned long n)
 
 		switch (n) {
 		case 1:
+			__uaccess_begin();
 			__get_user_size(*(u8 *)to, from, 1, ret, 1);
+			__uaccess_end();
 			return ret;
 		case 2:
+			__uaccess_begin();
 			__get_user_size(*(u16 *)to, from, 2, ret, 2);
+			__uaccess_end();
 			return ret;
 		case 4:
+			__uaccess_begin();
 			__get_user_size(*(u32 *)to, from, 4, ret, 4);
+			__uaccess_end();
 			return ret;
 		}
 	}
@@ -148,13 +162,19 @@ __copy_from_user(void *to, const void __user *from, unsigned long n)
 
 		switch (n) {
 		case 1:
+			__uaccess_begin();
 			__get_user_size(*(u8 *)to, from, 1, ret, 1);
+			__uaccess_end();
 			return ret;
 		case 2:
+			__uaccess_begin();
 			__get_user_size(*(u16 *)to, from, 2, ret, 2);
+			__uaccess_end();
 			return ret;
 		case 4:
+			__uaccess_begin();
 			__get_user_size(*(u32 *)to, from, 4, ret, 4);
+			__uaccess_end();
 			return ret;
 		}
 	}
@@ -170,13 +190,19 @@ static __always_inline unsigned long __copy_from_user_nocache(void *to,
 
 		switch (n) {
 		case 1:
+			__uaccess_begin();
 			__get_user_size(*(u8 *)to, from, 1, ret, 1);
+			__uaccess_end();
 			return ret;
 		case 2:
+			__uaccess_begin();
 			__get_user_size(*(u16 *)to, from, 2, ret, 2);
+			__uaccess_end();
 			return ret;
 		case 4:
+			__uaccess_begin();
 			__get_user_size(*(u32 *)to, from, 4, ret, 4);
+			__uaccess_end();
 			return ret;
 		}
 	}
diff --git a/arch/x86/include/asm/xen/pci.h b/arch/x86/include/asm/xen/pci.h
index 968d57dd54c9..f320ee32d5a1 100644
--- a/arch/x86/include/asm/xen/pci.h
+++ b/arch/x86/include/asm/xen/pci.h
@@ -57,7 +57,7 @@ static inline int xen_pci_frontend_enable_msi(struct pci_dev *dev,
 {
 	if (xen_pci_frontend && xen_pci_frontend->enable_msi)
 		return xen_pci_frontend->enable_msi(dev, vectors);
-	return -ENODEV;
+	return -ENOSYS;
 }
 static inline void xen_pci_frontend_disable_msi(struct pci_dev *dev)
 {
@@ -69,7 +69,7 @@ static inline int xen_pci_frontend_enable_msix(struct pci_dev *dev,
 {
 	if (xen_pci_frontend && xen_pci_frontend->enable_msix)
 		return xen_pci_frontend->enable_msix(dev, vectors, nvec);
-	return -ENODEV;
+	return -ENOSYS;
 }
 static inline void xen_pci_frontend_disable_msix(struct pci_dev *dev)
 {
diff --git a/arch/x86/kernel/acpi/sleep.c b/arch/x86/kernel/acpi/sleep.c
index d1daead5fcdd..adb3eaf8fe2a 100644
--- a/arch/x86/kernel/acpi/sleep.c
+++ b/arch/x86/kernel/acpi/sleep.c
@@ -16,6 +16,7 @@
 #include <asm/cacheflush.h>
 #include <asm/realmode.h>
 
+#include <linux/ftrace.h>
 #include "../../realmode/rm/wakeup.h"
 #include "sleep.h"
 
@@ -107,7 +108,13 @@ int x86_acpi_suspend_lowlevel(void)
        saved_magic = 0x123456789abcdef0L;
 #endif /* CONFIG_64BIT */
 
+	/*
+	 * Pause/unpause graph tracing around do_suspend_lowlevel as it has
+	 * inconsistent call/return info after it jumps to the wakeup vector.
+	 */
+	pause_graph_tracing();
 	do_suspend_lowlevel();
+	unpause_graph_tracing();
 	return 0;
 }
 
diff --git a/arch/x86/kernel/cpu/perf_event_amd_uncore.c b/arch/x86/kernel/cpu/perf_event_amd_uncore.c
index 49742746a6c9..8836fc9fa84b 100644
--- a/arch/x86/kernel/cpu/perf_event_amd_uncore.c
+++ b/arch/x86/kernel/cpu/perf_event_amd_uncore.c
@@ -323,6 +323,8 @@ static int amd_uncore_cpu_up_prepare(unsigned int cpu)
 	return 0;
 
 fail:
+	if (amd_uncore_nb)
+		*per_cpu_ptr(amd_uncore_nb, cpu) = NULL;
 	kfree(uncore_nb);
 	return -ENOMEM;
 }
diff --git a/arch/x86/kvm/emulate.c b/arch/x86/kvm/emulate.c
index 1505587d06e9..b9b09fec173b 100644
--- a/arch/x86/kvm/emulate.c
+++ b/arch/x86/kvm/emulate.c
@@ -650,10 +650,10 @@ static __always_inline int __linearize(struct x86_emulate_ctxt *ctxt,
 	u16 sel;
 
 	la = seg_base(ctxt, addr.seg) + addr.ea;
-	*linear = la;
 	*max_size = 0;
 	switch (mode) {
 	case X86EMUL_MODE_PROT64:
+		*linear = la;
 		if (is_noncanonical_address(la))
 			goto bad;
 
@@ -662,6 +662,7 @@ static __always_inline int __linearize(struct x86_emulate_ctxt *ctxt,
 			goto bad;
 		break;
 	default:
+		*linear = la = (u32)la;
 		usable = ctxt->ops->get_segment(ctxt, &sel, &desc, NULL,
 						addr.seg);
 		if (!usable)
@@ -689,7 +690,6 @@ static __always_inline int __linearize(struct x86_emulate_ctxt *ctxt,
 			if (size > *max_size)
 				goto bad;
 		}
-		la &= (u32)-1;
 		break;
 	}
 	if (insn_aligned(ctxt, size) && ((la & (size - 1)) != 0))
diff --git a/arch/x86/kvm/mmu.c b/arch/x86/kvm/mmu.c
index 95a955de5964..1e7a49bfc94f 100644
--- a/arch/x86/kvm/mmu.c
+++ b/arch/x86/kvm/mmu.c
@@ -3721,13 +3721,15 @@ static void reset_rsvds_bits_mask_ept(struct kvm_vcpu *vcpu,
 void
 reset_shadow_zero_bits_mask(struct kvm_vcpu *vcpu, struct kvm_mmu *context)
 {
+	bool uses_nx = context->nx || context->base_role.smep_andnot_wp;
+
 	/*
 	 * Passing "true" to the last argument is okay; it adds a check
 	 * on bit 8 of the SPTEs which KVM doesn't use anyway.
 	 */
 	__reset_rsvds_bits_mask(vcpu, &context->shadow_zero_check,
 				boot_cpu_data.x86_phys_bits,
-				context->shadow_root_level, context->nx,
+				context->shadow_root_level, uses_nx,
 				guest_cpuid_has_gbpages(vcpu), is_pse(vcpu),
 				true);
 }
diff --git a/arch/x86/kvm/paging_tmpl.h b/arch/x86/kvm/paging_tmpl.h
index 6c9fed957cce..2ce4f05e81d3 100644
--- a/arch/x86/kvm/paging_tmpl.h
+++ b/arch/x86/kvm/paging_tmpl.h
@@ -249,7 +249,7 @@ static int FNAME(update_accessed_dirty_bits)(struct kvm_vcpu *vcpu,
 			return ret;
 
 		kvm_vcpu_mark_page_dirty(vcpu, table_gfn);
-		walker->ptes[level] = pte;
+		walker->ptes[level - 1] = pte;
 	}
 	return 0;
 }
diff --git a/arch/x86/kvm/vmx.c b/arch/x86/kvm/vmx.c
index e2951b6edbbc..9bd8f44baded 100644
--- a/arch/x86/kvm/vmx.c
+++ b/arch/x86/kvm/vmx.c
@@ -596,6 +596,8 @@ struct vcpu_vmx {
 	/* Support for PML */
 #define PML_ENTITY_NUM		512
 	struct page *pml_pg;
+
+	u64 current_tsc_ratio;
 };
 
 enum segment_cache_field {
@@ -1811,6 +1813,13 @@ static void add_atomic_switch_msr(struct vcpu_vmx *vmx, unsigned msr,
 			return;
 		}
 		break;
+	case MSR_IA32_PEBS_ENABLE:
+		/* PEBS needs a quiescent period after being disabled (to write
+		 * a record).  Disabling PEBS through VMX MSR swapping doesn't
+		 * provide that period, so a CPU could write host's record into
+		 * guest's memory.
+		 */
+		wrmsrl(MSR_IA32_PEBS_ENABLE, 0);
 	}
 
 	for (i = 0; i < m->nr; ++i)
@@ -1848,26 +1857,31 @@ static void reload_tss(void)
 
 static bool update_transition_efer(struct vcpu_vmx *vmx, int efer_offset)
 {
-	u64 guest_efer;
-	u64 ignore_bits;
+	u64 guest_efer = vmx->vcpu.arch.efer;
+	u64 ignore_bits = 0;
 
-	guest_efer = vmx->vcpu.arch.efer;
+	if (!enable_ept) {
+		/*
+		 * NX is needed to handle CR0.WP=1, CR4.SMEP=1.  Testing
+		 * host CPUID is more efficient than testing guest CPUID
+		 * or CR4.  Host SMEP is anyway a requirement for guest SMEP.
+		 */
+		if (boot_cpu_has(X86_FEATURE_SMEP))
+			guest_efer |= EFER_NX;
+		else if (!(guest_efer & EFER_NX))
+			ignore_bits |= EFER_NX;
+	}
 
 	/*
-	 * NX is emulated; LMA and LME handled by hardware; SCE meaningless
-	 * outside long mode
+	 * LMA and LME handled by hardware; SCE meaningless outside long mode.
 	 */
-	ignore_bits = EFER_NX | EFER_SCE;
+	ignore_bits |= EFER_SCE;
 #ifdef CONFIG_X86_64
 	ignore_bits |= EFER_LMA | EFER_LME;
 	/* SCE is meaningful only in long mode on Intel */
 	if (guest_efer & EFER_LMA)
 		ignore_bits &= ~(u64)EFER_SCE;
 #endif
-	guest_efer &= ~ignore_bits;
-	guest_efer |= host_efer & ignore_bits;
-	vmx->guest_msrs[efer_offset].data = guest_efer;
-	vmx->guest_msrs[efer_offset].mask = ~ignore_bits;
 
 	clear_atomic_switch_msr(vmx, MSR_EFER);
 
@@ -1878,16 +1892,21 @@ static bool update_transition_efer(struct vcpu_vmx *vmx, int efer_offset)
 	 */
 	if (cpu_has_load_ia32_efer ||
 	    (enable_ept && ((vmx->vcpu.arch.efer ^ host_efer) & EFER_NX))) {
-		guest_efer = vmx->vcpu.arch.efer;
 		if (!(guest_efer & EFER_LMA))
 			guest_efer &= ~EFER_LME;
 		if (guest_efer != host_efer)
 			add_atomic_switch_msr(vmx, MSR_EFER,
 					      guest_efer, host_efer);
 		return false;
-	}
+	} else {
+		guest_efer &= ~ignore_bits;
+		guest_efer |= host_efer & ignore_bits;
 
-	return true;
+		vmx->guest_msrs[efer_offset].data = guest_efer;
+		vmx->guest_msrs[efer_offset].mask = ~ignore_bits;
+
+		return true;
+	}
 }
 
 static unsigned long segment_base(u16 selector)
@@ -2127,14 +2146,16 @@ static void vmx_vcpu_load(struct kvm_vcpu *vcpu, int cpu)
 		rdmsrl(MSR_IA32_SYSENTER_ESP, sysenter_esp);
 		vmcs_writel(HOST_IA32_SYSENTER_ESP, sysenter_esp); /* 22.2.3 */
 
-		/* Setup TSC multiplier */
-		if (cpu_has_vmx_tsc_scaling())
-			vmcs_write64(TSC_MULTIPLIER,
-				     vcpu->arch.tsc_scaling_ratio);
-
 		vmx->loaded_vmcs->cpu = cpu;
 	}
 
+	/* Setup TSC multiplier */
+	if (kvm_has_tsc_control &&
+	    vmx->current_tsc_ratio != vcpu->arch.tsc_scaling_ratio) {
+		vmx->current_tsc_ratio = vcpu->arch.tsc_scaling_ratio;
+		vmcs_write64(TSC_MULTIPLIER, vmx->current_tsc_ratio);
+	}
+
 	vmx_vcpu_pi_load(vcpu, cpu);
 }
 
diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c
index 4244c2baf57d..eaf6ee8c28b8 100644
--- a/arch/x86/kvm/x86.c
+++ b/arch/x86/kvm/x86.c
@@ -6618,12 +6618,12 @@ static int vcpu_enter_guest(struct kvm_vcpu *vcpu)
 	 * KVM_DEBUGREG_WONT_EXIT again.
 	 */
 	if (unlikely(vcpu->arch.switch_db_regs & KVM_DEBUGREG_WONT_EXIT)) {
-		int i;
-
 		WARN_ON(vcpu->guest_debug & KVM_GUESTDBG_USE_HW_BP);
 		kvm_x86_ops->sync_dirty_debug_regs(vcpu);
-		for (i = 0; i < KVM_NR_DB_REGS; i++)
-			vcpu->arch.eff_db[i] = vcpu->arch.db[i];
+		kvm_update_dr0123(vcpu);
+		kvm_update_dr6(vcpu);
+		kvm_update_dr7(vcpu);
+		vcpu->arch.switch_db_regs &= ~KVM_DEBUGREG_RELOAD;
 	}
 
 	/*
diff --git a/arch/x86/lib/copy_user_64.S b/arch/x86/lib/copy_user_64.S
index 982ce34f4a9b..27f89c79a44b 100644
--- a/arch/x86/lib/copy_user_64.S
+++ b/arch/x86/lib/copy_user_64.S
@@ -232,17 +232,31 @@ ENDPROC(copy_user_enhanced_fast_string)
 
 /*
  * copy_user_nocache - Uncached memory copy with exception handling
- * This will force destination/source out of cache for more performance.
+ * This will force destination out of cache for more performance.
+ *
+ * Note: Cached memory copy is used when destination or size is not
+ * naturally aligned. That is:
+ *  - Require 8-byte alignment when size is 8 bytes or larger.
+ *  - Require 4-byte alignment when size is 4 bytes.
  */
 ENTRY(__copy_user_nocache)
 	ASM_STAC
+
+	/* If size is less than 8 bytes, go to 4-byte copy */
 	cmpl $8,%edx
-	jb 20f		/* less then 8 bytes, go to byte copy loop */
+	jb .L_4b_nocache_copy_entry
+
+	/* If destination is not 8-byte aligned, "cache" copy to align it */
 	ALIGN_DESTINATION
+
+	/* Set 4x8-byte copy count and remainder */
 	movl %edx,%ecx
 	andl $63,%edx
 	shrl $6,%ecx
-	jz 17f
+	jz .L_8b_nocache_copy_entry	/* jump if count is 0 */
+
+	/* Perform 4x8-byte nocache loop-copy */
+.L_4x8b_nocache_copy_loop:
 1:	movq (%rsi),%r8
 2:	movq 1*8(%rsi),%r9
 3:	movq 2*8(%rsi),%r10
@@ -262,60 +276,106 @@ ENTRY(__copy_user_nocache)
 	leaq 64(%rsi),%rsi
 	leaq 64(%rdi),%rdi
 	decl %ecx
-	jnz 1b
-17:	movl %edx,%ecx
+	jnz .L_4x8b_nocache_copy_loop
+
+	/* Set 8-byte copy count and remainder */
+.L_8b_nocache_copy_entry:
+	movl %edx,%ecx
 	andl $7,%edx
 	shrl $3,%ecx
-	jz 20f
-18:	movq (%rsi),%r8
-19:	movnti %r8,(%rdi)
+	jz .L_4b_nocache_copy_entry	/* jump if count is 0 */
+
+	/* Perform 8-byte nocache loop-copy */
+.L_8b_nocache_copy_loop:
+20:	movq (%rsi),%r8
+21:	movnti %r8,(%rdi)
 	leaq 8(%rsi),%rsi
 	leaq 8(%rdi),%rdi
 	decl %ecx
-	jnz 18b
-20:	andl %edx,%edx
-	jz 23f
+	jnz .L_8b_nocache_copy_loop
+
+	/* If no byte left, we're done */
+.L_4b_nocache_copy_entry:
+	andl %edx,%edx
+	jz .L_finish_copy
+
+	/* If destination is not 4-byte aligned, go to byte copy: */
+	movl %edi,%ecx
+	andl $3,%ecx
+	jnz .L_1b_cache_copy_entry
+
+	/* Set 4-byte copy count (1 or 0) and remainder */
 	movl %edx,%ecx
-21:	movb (%rsi),%al
-22:	movb %al,(%rdi)
+	andl $3,%edx
+	shrl $2,%ecx
+	jz .L_1b_cache_copy_entry	/* jump if count is 0 */
+
+	/* Perform 4-byte nocache copy: */
+30:	movl (%rsi),%r8d
+31:	movnti %r8d,(%rdi)
+	leaq 4(%rsi),%rsi
+	leaq 4(%rdi),%rdi
+
+	/* If no bytes left, we're done: */
+	andl %edx,%edx
+	jz .L_finish_copy
+
+	/* Perform byte "cache" loop-copy for the remainder */
+.L_1b_cache_copy_entry:
+	movl %edx,%ecx
+.L_1b_cache_copy_loop:
+40:	movb (%rsi),%al
+41:	movb %al,(%rdi)
 	incq %rsi
 	incq %rdi
 	decl %ecx
-	jnz 21b
-23:	xorl %eax,%eax
+	jnz .L_1b_cache_copy_loop
+
+	/* Finished copying; fence the prior stores */
+.L_finish_copy:
+	xorl %eax,%eax
 	ASM_CLAC
 	sfence
 	ret
 
 	.section .fixup,"ax"
-30:	shll $6,%ecx
+.L_fixup_4x8b_copy:
+	shll $6,%ecx
 	addl %ecx,%edx
-	jmp 60f
-40:	lea (%rdx,%rcx,8),%rdx
-	jmp 60f
-50:	movl %ecx,%edx
-60:	sfence
+	jmp .L_fixup_handle_tail
+.L_fixup_8b_copy:
+	lea (%rdx,%rcx,8),%rdx
+	jmp .L_fixup_handle_tail
+.L_fixup_4b_copy:
+	lea (%rdx,%rcx,4),%rdx
+	jmp .L_fixup_handle_tail
+.L_fixup_1b_copy:
+	movl %ecx,%edx
+.L_fixup_handle_tail:
+	sfence
 	jmp copy_user_handle_tail
 	.previous
 
-	_ASM_EXTABLE(1b,30b)
-	_ASM_EXTABLE(2b,30b)
-	_ASM_EXTABLE(3b,30b)
-	_ASM_EXTABLE(4b,30b)
-	_ASM_EXTABLE(5b,30b)
-	_ASM_EXTABLE(6b,30b)
-	_ASM_EXTABLE(7b,30b)
-	_ASM_EXTABLE(8b,30b)
-	_ASM_EXTABLE(9b,30b)
-	_ASM_EXTABLE(10b,30b)
-	_ASM_EXTABLE(11b,30b)
-	_ASM_EXTABLE(12b,30b)
-	_ASM_EXTABLE(13b,30b)
-	_ASM_EXTABLE(14b,30b)
-	_ASM_EXTABLE(15b,30b)
-	_ASM_EXTABLE(16b,30b)
-	_ASM_EXTABLE(18b,40b)
-	_ASM_EXTABLE(19b,40b)
-	_ASM_EXTABLE(21b,50b)
-	_ASM_EXTABLE(22b,50b)
+	_ASM_EXTABLE(1b,.L_fixup_4x8b_copy)
+	_ASM_EXTABLE(2b,.L_fixup_4x8b_copy)
+	_ASM_EXTABLE(3b,.L_fixup_4x8b_copy)
+	_ASM_EXTABLE(4b,.L_fixup_4x8b_copy)
+	_ASM_EXTABLE(5b,.L_fixup_4x8b_copy)
+	_ASM_EXTABLE(6b,.L_fixup_4x8b_copy)
+	_ASM_EXTABLE(7b,.L_fixup_4x8b_copy)
+	_ASM_EXTABLE(8b,.L_fixup_4x8b_copy)
+	_ASM_EXTABLE(9b,.L_fixup_4x8b_copy)
+	_ASM_EXTABLE(10b,.L_fixup_4x8b_copy)
+	_ASM_EXTABLE(11b,.L_fixup_4x8b_copy)
+	_ASM_EXTABLE(12b,.L_fixup_4x8b_copy)
+	_ASM_EXTABLE(13b,.L_fixup_4x8b_copy)
+	_ASM_EXTABLE(14b,.L_fixup_4x8b_copy)
+	_ASM_EXTABLE(15b,.L_fixup_4x8b_copy)
+	_ASM_EXTABLE(16b,.L_fixup_4x8b_copy)
+	_ASM_EXTABLE(20b,.L_fixup_8b_copy)
+	_ASM_EXTABLE(21b,.L_fixup_8b_copy)
+	_ASM_EXTABLE(30b,.L_fixup_4b_copy)
+	_ASM_EXTABLE(31b,.L_fixup_4b_copy)
+	_ASM_EXTABLE(40b,.L_fixup_1b_copy)
+	_ASM_EXTABLE(41b,.L_fixup_1b_copy)
 ENDPROC(__copy_user_nocache)
diff --git a/arch/x86/mm/fault.c b/arch/x86/mm/fault.c
index eef44d9a3f77..e830c71a1323 100644
--- a/arch/x86/mm/fault.c
+++ b/arch/x86/mm/fault.c
@@ -287,6 +287,9 @@ static noinline int vmalloc_fault(unsigned long address)
 	if (!pmd_k)
 		return -1;
 
+	if (pmd_huge(*pmd_k))
+		return 0;
+
 	pte_k = pte_offset_kernel(pmd_k, address);
 	if (!pte_present(*pte_k))
 		return -1;
@@ -360,8 +363,6 @@ void vmalloc_sync_all(void)
  * 64-bit:
  *
  *   Handle a fault on the vmalloc area
- *
- * This assumes no large pages in there.
  */
 static noinline int vmalloc_fault(unsigned long address)
 {
@@ -403,17 +404,23 @@ static noinline int vmalloc_fault(unsigned long address)
 	if (pud_none(*pud_ref))
 		return -1;
 
-	if (pud_none(*pud) || pud_page_vaddr(*pud) != pud_page_vaddr(*pud_ref))
+	if (pud_none(*pud) || pud_pfn(*pud) != pud_pfn(*pud_ref))
 		BUG();
 
+	if (pud_huge(*pud))
+		return 0;
+
 	pmd = pmd_offset(pud, address);
 	pmd_ref = pmd_offset(pud_ref, address);
 	if (pmd_none(*pmd_ref))
 		return -1;
 
-	if (pmd_none(*pmd) || pmd_page(*pmd) != pmd_page(*pmd_ref))
+	if (pmd_none(*pmd) || pmd_pfn(*pmd) != pmd_pfn(*pmd_ref))
 		BUG();
 
+	if (pmd_huge(*pmd))
+		return 0;
+
 	pte_ref = pte_offset_kernel(pmd_ref, address);
 	if (!pte_present(*pte_ref))
 		return -1;
diff --git a/arch/x86/mm/gup.c b/arch/x86/mm/gup.c
index 6d5eb5900372..d8a798d8bf50 100644
--- a/arch/x86/mm/gup.c
+++ b/arch/x86/mm/gup.c
@@ -102,7 +102,6 @@ static noinline int gup_pte_range(pmd_t pmd, unsigned long addr,
 			return 0;
 		}
 
-		page = pte_page(pte);
 		if (pte_devmap(pte)) {
 			pgmap = get_dev_pagemap(pte_pfn(pte), pgmap);
 			if (unlikely(!pgmap)) {
@@ -115,6 +114,7 @@ static noinline int gup_pte_range(pmd_t pmd, unsigned long addr,
 			return 0;
 		}
 		VM_BUG_ON(!pfn_valid(pte_pfn(pte)));
+		page = pte_page(pte);
 		get_page(page);
 		put_dev_pagemap(pgmap);
 		SetPageReferenced(page);
diff --git a/arch/x86/mm/mmap.c b/arch/x86/mm/mmap.c
index 96bd1e2bffaf..72bb52f93c3d 100644
--- a/arch/x86/mm/mmap.c
+++ b/arch/x86/mm/mmap.c
@@ -71,12 +71,12 @@ unsigned long arch_mmap_rnd(void)
 
 	if (mmap_is_ia32())
 #ifdef CONFIG_COMPAT
-		rnd = (unsigned long)get_random_int() & ((1 << mmap_rnd_compat_bits) - 1);
+		rnd = get_random_long() & ((1UL << mmap_rnd_compat_bits) - 1);
 #else
-		rnd = (unsigned long)get_random_int() & ((1 << mmap_rnd_bits) - 1);
+		rnd = get_random_long() & ((1UL << mmap_rnd_bits) - 1);
 #endif
 	else
-		rnd = (unsigned long)get_random_int() & ((1 << mmap_rnd_bits) - 1);
+		rnd = get_random_long() & ((1UL << mmap_rnd_bits) - 1);
 
 	return rnd << PAGE_SHIFT;
 }
diff --git a/arch/x86/mm/mpx.c b/arch/x86/mm/mpx.c
index b2fd67da1701..ef05755a1900 100644
--- a/arch/x86/mm/mpx.c
+++ b/arch/x86/mm/mpx.c
@@ -123,7 +123,7 @@ static int get_reg_offset(struct insn *insn, struct pt_regs *regs,
 		break;
 	}
 
-	if (regno > nr_registers) {
+	if (regno >= nr_registers) {
 		WARN_ONCE(1, "decoded an instruction with an invalid register");
 		return -EINVAL;
 	}
diff --git a/arch/x86/mm/numa.c b/arch/x86/mm/numa.c
index c3b3f653ed0c..d04f8094bc23 100644
--- a/arch/x86/mm/numa.c
+++ b/arch/x86/mm/numa.c
@@ -469,7 +469,7 @@ static void __init numa_clear_kernel_node_hotplug(void)
 {
 	int i, nid;
 	nodemask_t numa_kernel_nodes = NODE_MASK_NONE;
-	unsigned long start, end;
+	phys_addr_t start, end;
 	struct memblock_region *r;
 
 	/*
diff --git a/arch/x86/mm/pageattr.c b/arch/x86/mm/pageattr.c
index 2440814b0069..9cf96d82147a 100644
--- a/arch/x86/mm/pageattr.c
+++ b/arch/x86/mm/pageattr.c
@@ -419,24 +419,30 @@ pmd_t *lookup_pmd_address(unsigned long address)
 phys_addr_t slow_virt_to_phys(void *__virt_addr)
 {
 	unsigned long virt_addr = (unsigned long)__virt_addr;
-	unsigned long phys_addr, offset;
+	phys_addr_t phys_addr;
+	unsigned long offset;
 	enum pg_level level;
 	pte_t *pte;
 
 	pte = lookup_address(virt_addr, &level);
 	BUG_ON(!pte);
 
+	/*
+	 * pXX_pfn() returns unsigned long, which must be cast to phys_addr_t
+	 * before being left-shifted PAGE_SHIFT bits -- this trick is to
+	 * make 32-PAE kernel work correctly.
+	 */
 	switch (level) {
 	case PG_LEVEL_1G:
-		phys_addr = pud_pfn(*(pud_t *)pte) << PAGE_SHIFT;
+		phys_addr = (phys_addr_t)pud_pfn(*(pud_t *)pte) << PAGE_SHIFT;
 		offset = virt_addr & ~PUD_PAGE_MASK;
 		break;
 	case PG_LEVEL_2M:
-		phys_addr = pmd_pfn(*(pmd_t *)pte) << PAGE_SHIFT;
+		phys_addr = (phys_addr_t)pmd_pfn(*(pmd_t *)pte) << PAGE_SHIFT;
 		offset = virt_addr & ~PMD_PAGE_MASK;
 		break;
 	default:
-		phys_addr = pte_pfn(*pte) << PAGE_SHIFT;
+		phys_addr = (phys_addr_t)pte_pfn(*pte) << PAGE_SHIFT;
 		offset = virt_addr & ~PAGE_MASK;
 	}
 
diff --git a/arch/x86/pci/common.c b/arch/x86/pci/common.c
index 2879efc73a96..d34b5118b4e8 100644
--- a/arch/x86/pci/common.c
+++ b/arch/x86/pci/common.c
@@ -711,28 +711,22 @@ int pcibios_add_device(struct pci_dev *dev)
 	return 0;
 }
 
-int pcibios_alloc_irq(struct pci_dev *dev)
+int pcibios_enable_device(struct pci_dev *dev, int mask)
 {
-	/*
-	 * If the PCI device was already claimed by core code and has
-	 * MSI enabled, probing of the pcibios IRQ will overwrite
-	 * dev->irq.  So bail out if MSI is already enabled.
-	 */
-	if (pci_dev_msi_enabled(dev))
-		return -EBUSY;
+	int err;
 
-	return pcibios_enable_irq(dev);
-}
+	if ((err = pci_enable_resources(dev, mask)) < 0)
+		return err;
 
-void pcibios_free_irq(struct pci_dev *dev)
-{
-	if (pcibios_disable_irq)
-		pcibios_disable_irq(dev);
+	if (!pci_dev_msi_enabled(dev))
+		return pcibios_enable_irq(dev);
+	return 0;
 }
 
-int pcibios_enable_device(struct pci_dev *dev, int mask)
+void pcibios_disable_device (struct pci_dev *dev)
 {
-	return pci_enable_resources(dev, mask);
+	if (!pci_dev_msi_enabled(dev) && pcibios_disable_irq)
+		pcibios_disable_irq(dev);
 }
 
 int pci_ext_cfg_avail(void)
diff --git a/arch/x86/pci/intel_mid_pci.c b/arch/x86/pci/intel_mid_pci.c
index 0d24e7c10145..8b93e634af84 100644
--- a/arch/x86/pci/intel_mid_pci.c
+++ b/arch/x86/pci/intel_mid_pci.c
@@ -215,7 +215,7 @@ static int intel_mid_pci_irq_enable(struct pci_dev *dev)
 	int polarity;
 	int ret;
 
-	if (pci_has_managed_irq(dev))
+	if (dev->irq_managed && dev->irq > 0)
 		return 0;
 
 	switch (intel_mid_identify_cpu()) {
@@ -256,13 +256,10 @@ static int intel_mid_pci_irq_enable(struct pci_dev *dev)
 
 static void intel_mid_pci_irq_disable(struct pci_dev *dev)
 {
-	if (pci_has_managed_irq(dev)) {
+	if (!mp_should_keep_irq(&dev->dev) && dev->irq_managed &&
+	    dev->irq > 0) {
 		mp_unmap_irq(dev->irq);
 		dev->irq_managed = 0;
-		/*
-		 * Don't reset dev->irq here, otherwise
-		 * intel_mid_pci_irq_enable() will fail on next call.
-		 */
 	}
 }
 
diff --git a/arch/x86/pci/irq.c b/arch/x86/pci/irq.c
index 32e70343e6fd..9bd115484745 100644
--- a/arch/x86/pci/irq.c
+++ b/arch/x86/pci/irq.c
@@ -1202,7 +1202,7 @@ static int pirq_enable_irq(struct pci_dev *dev)
 			struct pci_dev *temp_dev;
 			int irq;
 
-			if (pci_has_managed_irq(dev))
+			if (dev->irq_managed && dev->irq > 0)
 				return 0;
 
 			irq = IO_APIC_get_PCI_irq_vector(dev->bus->number,
@@ -1230,7 +1230,8 @@ static int pirq_enable_irq(struct pci_dev *dev)
 			}
 			dev = temp_dev;
 			if (irq >= 0) {
-				pci_set_managed_irq(dev, irq);
+				dev->irq_managed = 1;
+				dev->irq = irq;
 				dev_info(&dev->dev, "PCI->APIC IRQ transform: "
 					 "INT %c -> IRQ %d\n", 'A' + pin - 1, irq);
 				return 0;
@@ -1256,10 +1257,24 @@ static int pirq_enable_irq(struct pci_dev *dev)
 	return 0;
 }
 
+bool mp_should_keep_irq(struct device *dev)
+{
+	if (dev->power.is_prepared)
+		return true;
+#ifdef CONFIG_PM
+	if (dev->power.runtime_status == RPM_SUSPENDING)
+		return true;
+#endif
+
+	return false;
+}
+
 static void pirq_disable_irq(struct pci_dev *dev)
 {
-	if (io_apic_assign_pci_irqs && pci_has_managed_irq(dev)) {
+	if (io_apic_assign_pci_irqs && !mp_should_keep_irq(&dev->dev) &&
+	    dev->irq_managed && dev->irq) {
 		mp_unmap_irq(dev->irq);
-		pci_reset_managed_irq(dev);
+		dev->irq = 0;
+		dev->irq_managed = 0;
 	}
 }
diff --git a/arch/x86/pci/xen.c b/arch/x86/pci/xen.c
index ff31ab464213..beac4dfdade6 100644
--- a/arch/x86/pci/xen.c
+++ b/arch/x86/pci/xen.c
@@ -196,7 +196,10 @@ static int xen_setup_msi_irqs(struct pci_dev *dev, int nvec, int type)
 	return 0;
 
 error:
-	dev_err(&dev->dev, "Xen PCI frontend has not registered MSI/MSI-X support!\n");
+	if (ret == -ENOSYS)
+		dev_err(&dev->dev, "Xen PCI frontend has not registered MSI/MSI-X support!\n");
+	else if (ret)
+		dev_err(&dev->dev, "Xen PCI frontend error: %d!\n", ret);
 free:
 	kfree(v);
 	return ret;
diff --git a/arch/x86/platform/intel-quark/imr.c b/arch/x86/platform/intel-quark/imr.c
index c61b6c332e97..bfadcd0f4944 100644
--- a/arch/x86/platform/intel-quark/imr.c
+++ b/arch/x86/platform/intel-quark/imr.c
@@ -592,14 +592,14 @@ static void __init imr_fixup_memmap(struct imr_device *idev)
 	end = (unsigned long)__end_rodata - 1;
 
 	/*
-	 * Setup a locked IMR around the physical extent of the kernel
+	 * Setup an unlocked IMR around the physical extent of the kernel
 	 * from the beginning of the .text secton to the end of the
 	 * .rodata section as one physically contiguous block.
 	 *
 	 * We don't round up @size since it is already PAGE_SIZE aligned.
 	 * See vmlinux.lds.S for details.
 	 */
-	ret = imr_add_range(base, size, IMR_CPU, IMR_CPU, true);
+	ret = imr_add_range(base, size, IMR_CPU, IMR_CPU, false);
 	if (ret < 0) {
 		pr_err("unable to setup IMR for kernel: %zu KiB (%lx - %lx)\n",
 			size / 1024, start, end);
diff --git a/arch/x86/um/os-Linux/task_size.c b/arch/x86/um/os-Linux/task_size.c
index 8502ad30e61b..5adb6a2fd117 100644
--- a/arch/x86/um/os-Linux/task_size.c
+++ b/arch/x86/um/os-Linux/task_size.c
@@ -109,7 +109,7 @@ unsigned long os_get_top_address(void)
 		exit(1);
 	}
 
-	printf("0x%x\n", bottom << UM_KERN_PAGE_SHIFT);
+	printf("0x%lx\n", bottom << UM_KERN_PAGE_SHIFT);
 	printf("Locating the top of the address space ... ");
 	fflush(stdout);
 
@@ -134,7 +134,7 @@ out:
 		exit(1);
 	}
 	top <<= UM_KERN_PAGE_SHIFT;
-	printf("0x%x\n", top);
+	printf("0x%lx\n", top);
 
 	return top;
 }
diff --git a/block/Kconfig b/block/Kconfig
index 161491d0a879..0363cd731320 100644
--- a/block/Kconfig
+++ b/block/Kconfig
@@ -88,6 +88,19 @@ config BLK_DEV_INTEGRITY
 	T10/SCSI Data Integrity Field or the T13/ATA External Path
 	Protection.  If in doubt, say N.
 
+config BLK_DEV_DAX
+	bool "Block device DAX support"
+	depends on FS_DAX
+	depends on BROKEN
+	help
+	  When DAX support is available (CONFIG_FS_DAX) raw block
+	  devices can also support direct userspace access to the
+	  storage capacity via MMAP(2) similar to a file on a
+	  DAX-enabled filesystem.  However, the DAX I/O-path disables
+	  some standard I/O-statistics, and the MMAP(2) path has some
+	  operational differences due to bypassing the page
+	  cache.  If in doubt, say N.
+
 config BLK_DEV_THROTTLING
 	bool "Block layer bio throttling support"
 	depends on BLK_CGROUP=y
diff --git a/block/bio.c b/block/bio.c
index dbabd48b1934..cf7591551b17 100644
--- a/block/bio.c
+++ b/block/bio.c
@@ -874,7 +874,7 @@ int submit_bio_wait(int rw, struct bio *bio)
 	bio->bi_private = &ret;
 	bio->bi_end_io = submit_bio_wait_endio;
 	submit_bio(rw, bio);
-	wait_for_completion(&ret.event);
+	wait_for_completion_io(&ret.event);
 
 	return ret.error;
 }
@@ -1090,9 +1090,12 @@ int bio_uncopy_user(struct bio *bio)
 	if (!bio_flagged(bio, BIO_NULL_MAPPED)) {
 		/*
 		 * if we're in a workqueue, the request is orphaned, so
-		 * don't copy into a random user address space, just free.
+		 * don't copy into a random user address space, just free
+		 * and return -EINTR so user space doesn't expect any data.
 		 */
-		if (current->mm && bio_data_dir(bio) == READ)
+		if (!current->mm)
+			ret = -EINTR;
+		else if (bio_data_dir(bio) == READ)
 			ret = bio_copy_to_iter(bio, bmd->iter);
 		if (bmd->is_our_pages)
 			bio_free_pages(bio);
diff --git a/block/blk-cgroup.c b/block/blk-cgroup.c
index 5a37188b559f..66e6f1aae02e 100644
--- a/block/blk-cgroup.c
+++ b/block/blk-cgroup.c
@@ -788,6 +788,7 @@ int blkg_conf_prep(struct blkcg *blkcg, const struct blkcg_policy *pol,
 {
 	struct gendisk *disk;
 	struct blkcg_gq *blkg;
+	struct module *owner;
 	unsigned int major, minor;
 	int key_len, part, ret;
 	char *body;
@@ -804,7 +805,9 @@ int blkg_conf_prep(struct blkcg *blkcg, const struct blkcg_policy *pol,
 	if (!disk)
 		return -ENODEV;
 	if (part) {
+		owner = disk->fops->owner;
 		put_disk(disk);
+		module_put(owner);
 		return -ENODEV;
 	}
 
@@ -820,7 +823,9 @@ int blkg_conf_prep(struct blkcg *blkcg, const struct blkcg_policy *pol,
 		ret = PTR_ERR(blkg);
 		rcu_read_unlock();
 		spin_unlock_irq(disk->queue->queue_lock);
+		owner = disk->fops->owner;
 		put_disk(disk);
+		module_put(owner);
 		/*
 		 * If queue was bypassing, we should retry.  Do so after a
 		 * short msleep().  It isn't strictly necessary but queue
@@ -851,9 +856,13 @@ EXPORT_SYMBOL_GPL(blkg_conf_prep);
 void blkg_conf_finish(struct blkg_conf_ctx *ctx)
 	__releases(ctx->disk->queue->queue_lock) __releases(rcu)
 {
+	struct module *owner;
+
 	spin_unlock_irq(ctx->disk->queue->queue_lock);
 	rcu_read_unlock();
+	owner = ctx->disk->fops->owner;
 	put_disk(ctx->disk);
+	module_put(owner);
 }
 EXPORT_SYMBOL_GPL(blkg_conf_finish);
 
diff --git a/block/blk-core.c b/block/blk-core.c
index ab51685988c2..b83d29755b5a 100644
--- a/block/blk-core.c
+++ b/block/blk-core.c
@@ -2455,14 +2455,16 @@ struct request *blk_peek_request(struct request_queue *q)
 
 			rq = NULL;
 			break;
-		} else if (ret == BLKPREP_KILL) {
+		} else if (ret == BLKPREP_KILL || ret == BLKPREP_INVALID) {
+			int err = (ret == BLKPREP_INVALID) ? -EREMOTEIO : -EIO;
+
 			rq->cmd_flags |= REQ_QUIET;
 			/*
 			 * Mark this request as started so we don't trigger
 			 * any debug logic in the end I/O path.
 			 */
 			blk_start_request(rq);
-			__blk_end_request_all(rq, -EIO);
+			__blk_end_request_all(rq, err);
 		} else {
 			printk(KERN_ERR "%s: bad return=%d\n", __func__, ret);
 			break;
diff --git a/block/blk-map.c b/block/blk-map.c
index f565e11f465a..a54f0543b956 100644
--- a/block/blk-map.c
+++ b/block/blk-map.c
@@ -57,6 +57,49 @@ static int __blk_rq_unmap_user(struct bio *bio)
 	return ret;
 }
 
+static int __blk_rq_map_user_iov(struct request *rq,
+		struct rq_map_data *map_data, struct iov_iter *iter,
+		gfp_t gfp_mask, bool copy)
+{
+	struct request_queue *q = rq->q;
+	struct bio *bio, *orig_bio;
+	int ret;
+
+	if (copy)
+		bio = bio_copy_user_iov(q, map_data, iter, gfp_mask);
+	else
+		bio = bio_map_user_iov(q, iter, gfp_mask);
+
+	if (IS_ERR(bio))
+		return PTR_ERR(bio);
+
+	if (map_data && map_data->null_mapped)
+		bio_set_flag(bio, BIO_NULL_MAPPED);
+
+	iov_iter_advance(iter, bio->bi_iter.bi_size);
+	if (map_data)
+		map_data->offset += bio->bi_iter.bi_size;
+
+	orig_bio = bio;
+	blk_queue_bounce(q, &bio);
+
+	/*
+	 * We link the bounce buffer in and could have to traverse it
+	 * later so we have to get a ref to prevent it from being freed
+	 */
+	bio_get(bio);
+
+	ret = blk_rq_append_bio(q, rq, bio);
+	if (ret) {
+		bio_endio(bio);
+		__blk_rq_unmap_user(orig_bio);
+		bio_put(bio);
+		return ret;
+	}
+
+	return 0;
+}
+
 /**
  * blk_rq_map_user_iov - map user data to a request, for REQ_TYPE_BLOCK_PC usage
  * @q:		request queue where request should be inserted
@@ -82,10 +125,11 @@ int blk_rq_map_user_iov(struct request_queue *q, struct request *rq,
 			struct rq_map_data *map_data,
 			const struct iov_iter *iter, gfp_t gfp_mask)
 {
-	struct bio *bio;
-	int unaligned = 0;
-	struct iov_iter i;
 	struct iovec iov, prv = {.iov_base = NULL, .iov_len = 0};
+	bool copy = (q->dma_pad_mask & iter->count) || map_data;
+	struct bio *bio = NULL;
+	struct iov_iter i;
+	int ret;
 
 	if (!iter || !iter->count)
 		return -EINVAL;
@@ -101,42 +145,29 @@ int blk_rq_map_user_iov(struct request_queue *q, struct request *rq,
 		 */
 		if ((uaddr & queue_dma_alignment(q)) ||
 		    iovec_gap_to_prv(q, &prv, &iov))
-			unaligned = 1;
+			copy = true;
 
 		prv.iov_base = iov.iov_base;
 		prv.iov_len = iov.iov_len;
 	}
 
-	if (unaligned || (q->dma_pad_mask & iter->count) || map_data)
-		bio = bio_copy_user_iov(q, map_data, iter, gfp_mask);
-	else
-		bio = bio_map_user_iov(q, iter, gfp_mask);
-
-	if (IS_ERR(bio))
-		return PTR_ERR(bio);
-
-	if (map_data && map_data->null_mapped)
-		bio_set_flag(bio, BIO_NULL_MAPPED);
-
-	if (bio->bi_iter.bi_size != iter->count) {
-		/*
-		 * Grab an extra reference to this bio, as bio_unmap_user()
-		 * expects to be able to drop it twice as it happens on the
-		 * normal IO completion path
-		 */
-		bio_get(bio);
-		bio_endio(bio);
-		__blk_rq_unmap_user(bio);
-		return -EINVAL;
-	}
+	i = *iter;
+	do {
+		ret =__blk_rq_map_user_iov(rq, map_data, &i, gfp_mask, copy);
+		if (ret)
+			goto unmap_rq;
+		if (!bio)
+			bio = rq->bio;
+	} while (iov_iter_count(&i));
 
 	if (!bio_flagged(bio, BIO_USER_MAPPED))
 		rq->cmd_flags |= REQ_COPY_USER;
-
-	blk_queue_bounce(q, &bio);
-	bio_get(bio);
-	blk_rq_bio_prep(q, rq, bio);
 	return 0;
+
+unmap_rq:
+	__blk_rq_unmap_user(bio);
+	rq->bio = NULL;
+	return -EINVAL;
 }
 EXPORT_SYMBOL(blk_rq_map_user_iov);
 
diff --git a/block/blk-merge.c b/block/blk-merge.c
index 888a7fec81f7..261353166dcf 100644
--- a/block/blk-merge.c
+++ b/block/blk-merge.c
@@ -304,7 +304,6 @@ static int blk_phys_contig_segment(struct request_queue *q, struct bio *bio,
 				   struct bio *nxt)
 {
 	struct bio_vec end_bv = { NULL }, nxt_bv;
-	struct bvec_iter iter;
 
 	if (!blk_queue_cluster(q))
 		return 0;
@@ -316,11 +315,8 @@ static int blk_phys_contig_segment(struct request_queue *q, struct bio *bio,
 	if (!bio_has_data(bio))
 		return 1;
 
-	bio_for_each_segment(end_bv, bio, iter)
-		if (end_bv.bv_len == iter.bi_size)
-			break;
-
-	nxt_bv = bio_iovec(nxt);
+	bio_get_last_bvec(bio, &end_bv);
+	bio_get_first_bvec(nxt, &nxt_bv);
 
 	if (!BIOVEC_PHYS_MERGEABLE(&end_bv, &nxt_bv))
 		return 0;
diff --git a/block/blk-mq.c b/block/blk-mq.c
index 4c0622fae413..56c0a726b619 100644
--- a/block/blk-mq.c
+++ b/block/blk-mq.c
@@ -599,8 +599,10 @@ static void blk_mq_check_expired(struct blk_mq_hw_ctx *hctx,
 		 * If a request wasn't started before the queue was
 		 * marked dying, kill it here or it'll go unnoticed.
 		 */
-		if (unlikely(blk_queue_dying(rq->q)))
-			blk_mq_complete_request(rq, -EIO);
+		if (unlikely(blk_queue_dying(rq->q))) {
+			rq->errors = -EIO;
+			blk_mq_end_request(rq, rq->errors);
+		}
 		return;
 	}
 
diff --git a/block/blk-settings.c b/block/blk-settings.c
index dd4973583978..c7bb666aafd1 100644
--- a/block/blk-settings.c
+++ b/block/blk-settings.c
@@ -91,8 +91,8 @@ void blk_set_default_limits(struct queue_limits *lim)
 	lim->seg_boundary_mask = BLK_SEG_BOUNDARY_MASK;
 	lim->virt_boundary_mask = 0;
 	lim->max_segment_size = BLK_MAX_SEGMENT_SIZE;
-	lim->max_sectors = lim->max_dev_sectors = lim->max_hw_sectors =
-		BLK_SAFE_MAX_SECTORS;
+	lim->max_sectors = lim->max_hw_sectors = BLK_SAFE_MAX_SECTORS;
+	lim->max_dev_sectors = 0;
 	lim->chunk_sectors = 0;
 	lim->max_write_same_sectors = 0;
 	lim->max_discard_sectors = 0;
diff --git a/block/blk-sysfs.c b/block/blk-sysfs.c
index e140cc487ce1..dd93763057ce 100644
--- a/block/blk-sysfs.c
+++ b/block/blk-sysfs.c
@@ -147,10 +147,9 @@ static ssize_t queue_discard_granularity_show(struct request_queue *q, char *pag
 
 static ssize_t queue_discard_max_hw_show(struct request_queue *q, char *page)
 {
-	unsigned long long val;
 
-	val = q->limits.max_hw_discard_sectors << 9;
-	return sprintf(page, "%llu\n", val);
+	return sprintf(page, "%llu\n",
+		(unsigned long long)q->limits.max_hw_discard_sectors << 9);
 }
 
 static ssize_t queue_discard_max_show(struct request_queue *q, char *page)
diff --git a/block/deadline-iosched.c b/block/deadline-iosched.c
index a753df2b3fc2..d0dd7882d8c7 100644
--- a/block/deadline-iosched.c
+++ b/block/deadline-iosched.c
@@ -39,7 +39,6 @@ struct deadline_data {
 	 */
 	struct request *next_rq[2];
 	unsigned int batching;		/* number of sequential requests made */
-	sector_t last_sector;		/* head position */
 	unsigned int starved;		/* times reads have starved writes */
 
 	/*
@@ -210,8 +209,6 @@ deadline_move_request(struct deadline_data *dd, struct request *rq)
 	dd->next_rq[WRITE] = NULL;
 	dd->next_rq[data_dir] = deadline_latter_request(rq);
 
-	dd->last_sector = rq_end_sector(rq);
-
 	/*
 	 * take it off the sort and fifo list, move
 	 * to dispatch queue
diff --git a/crypto/algif_skcipher.c b/crypto/algif_skcipher.c
index 38c1aa89d3a0..28556fce4267 100644
--- a/crypto/algif_skcipher.c
+++ b/crypto/algif_skcipher.c
@@ -65,18 +65,10 @@ struct skcipher_async_req {
 	struct skcipher_async_rsgl first_sgl;
 	struct list_head list;
 	struct scatterlist *tsg;
-	char iv[];
+	atomic_t *inflight;
+	struct skcipher_request req;
 };
 
-#define GET_SREQ(areq, ctx) (struct skcipher_async_req *)((char *)areq + \
-	crypto_skcipher_reqsize(crypto_skcipher_reqtfm(&ctx->req)))
-
-#define GET_REQ_SIZE(ctx) \
-	crypto_skcipher_reqsize(crypto_skcipher_reqtfm(&ctx->req))
-
-#define GET_IV_SIZE(ctx) \
-	crypto_skcipher_ivsize(crypto_skcipher_reqtfm(&ctx->req))
-
 #define MAX_SGL_ENTS ((4096 - sizeof(struct skcipher_sg_list)) / \
 		      sizeof(struct scatterlist) - 1)
 
@@ -102,15 +94,12 @@ static void skcipher_free_async_sgls(struct skcipher_async_req *sreq)
 
 static void skcipher_async_cb(struct crypto_async_request *req, int err)
 {
-	struct sock *sk = req->data;
-	struct alg_sock *ask = alg_sk(sk);
-	struct skcipher_ctx *ctx = ask->private;
-	struct skcipher_async_req *sreq = GET_SREQ(req, ctx);
+	struct skcipher_async_req *sreq = req->data;
 	struct kiocb *iocb = sreq->iocb;
 
-	atomic_dec(&ctx->inflight);
+	atomic_dec(sreq->inflight);
 	skcipher_free_async_sgls(sreq);
-	kfree(req);
+	kzfree(sreq);
 	iocb->ki_complete(iocb, err, err);
 }
 
@@ -306,8 +295,11 @@ static int skcipher_sendmsg(struct socket *sock, struct msghdr *msg,
 {
 	struct sock *sk = sock->sk;
 	struct alg_sock *ask = alg_sk(sk);
+	struct sock *psk = ask->parent;
+	struct alg_sock *pask = alg_sk(psk);
 	struct skcipher_ctx *ctx = ask->private;
-	struct crypto_skcipher *tfm = crypto_skcipher_reqtfm(&ctx->req);
+	struct skcipher_tfm *skc = pask->private;
+	struct crypto_skcipher *tfm = skc->skcipher;
 	unsigned ivsize = crypto_skcipher_ivsize(tfm);
 	struct skcipher_sg_list *sgl;
 	struct af_alg_control con = {};
@@ -509,37 +501,43 @@ static int skcipher_recvmsg_async(struct socket *sock, struct msghdr *msg,
 {
 	struct sock *sk = sock->sk;
 	struct alg_sock *ask = alg_sk(sk);
+	struct sock *psk = ask->parent;
+	struct alg_sock *pask = alg_sk(psk);
 	struct skcipher_ctx *ctx = ask->private;
+	struct skcipher_tfm *skc = pask->private;
+	struct crypto_skcipher *tfm = skc->skcipher;
 	struct skcipher_sg_list *sgl;
 	struct scatterlist *sg;
 	struct skcipher_async_req *sreq;
 	struct skcipher_request *req;
 	struct skcipher_async_rsgl *last_rsgl = NULL;
-	unsigned int txbufs = 0, len = 0, tx_nents = skcipher_all_sg_nents(ctx);
-	unsigned int reqlen = sizeof(struct skcipher_async_req) +
-				GET_REQ_SIZE(ctx) + GET_IV_SIZE(ctx);
+	unsigned int txbufs = 0, len = 0, tx_nents;
+	unsigned int reqsize = crypto_skcipher_reqsize(tfm);
+	unsigned int ivsize = crypto_skcipher_ivsize(tfm);
 	int err = -ENOMEM;
 	bool mark = false;
+	char *iv;
 
-	lock_sock(sk);
-	req = kmalloc(reqlen, GFP_KERNEL);
-	if (unlikely(!req))
-		goto unlock;
+	sreq = kzalloc(sizeof(*sreq) + reqsize + ivsize, GFP_KERNEL);
+	if (unlikely(!sreq))
+		goto out;
 
-	sreq = GET_SREQ(req, ctx);
+	req = &sreq->req;
+	iv = (char *)(req + 1) + reqsize;
 	sreq->iocb = msg->msg_iocb;
-	memset(&sreq->first_sgl, '\0', sizeof(struct skcipher_async_rsgl));
 	INIT_LIST_HEAD(&sreq->list);
+	sreq->inflight = &ctx->inflight;
+
+	lock_sock(sk);
+	tx_nents = skcipher_all_sg_nents(ctx);
 	sreq->tsg = kcalloc(tx_nents, sizeof(*sg), GFP_KERNEL);
-	if (unlikely(!sreq->tsg)) {
-		kfree(req);
+	if (unlikely(!sreq->tsg))
 		goto unlock;
-	}
 	sg_init_table(sreq->tsg, tx_nents);
-	memcpy(sreq->iv, ctx->iv, GET_IV_SIZE(ctx));
-	skcipher_request_set_tfm(req, crypto_skcipher_reqtfm(&ctx->req));
-	skcipher_request_set_callback(req, CRYPTO_TFM_REQ_MAY_BACKLOG,
-				      skcipher_async_cb, sk);
+	memcpy(iv, ctx->iv, ivsize);
+	skcipher_request_set_tfm(req, tfm);
+	skcipher_request_set_callback(req, CRYPTO_TFM_REQ_MAY_SLEEP,
+				      skcipher_async_cb, sreq);
 
 	while (iov_iter_count(&msg->msg_iter)) {
 		struct skcipher_async_rsgl *rsgl;
@@ -615,20 +613,22 @@ static int skcipher_recvmsg_async(struct socket *sock, struct msghdr *msg,
 		sg_mark_end(sreq->tsg + txbufs - 1);
 
 	skcipher_request_set_crypt(req, sreq->tsg, sreq->first_sgl.sgl.sg,
-				   len, sreq->iv);
+				   len, iv);
 	err = ctx->enc ? crypto_skcipher_encrypt(req) :
 			 crypto_skcipher_decrypt(req);
 	if (err == -EINPROGRESS) {
 		atomic_inc(&ctx->inflight);
 		err = -EIOCBQUEUED;
+		sreq = NULL;
 		goto unlock;
 	}
 free:
 	skcipher_free_async_sgls(sreq);
-	kfree(req);
 unlock:
 	skcipher_wmem_wakeup(sk);
 	release_sock(sk);
+	kzfree(sreq);
+out:
 	return err;
 }
 
@@ -637,9 +637,12 @@ static int skcipher_recvmsg_sync(struct socket *sock, struct msghdr *msg,
 {
 	struct sock *sk = sock->sk;
 	struct alg_sock *ask = alg_sk(sk);
+	struct sock *psk = ask->parent;
+	struct alg_sock *pask = alg_sk(psk);
 	struct skcipher_ctx *ctx = ask->private;
-	unsigned bs = crypto_skcipher_blocksize(crypto_skcipher_reqtfm(
-		&ctx->req));
+	struct skcipher_tfm *skc = pask->private;
+	struct crypto_skcipher *tfm = skc->skcipher;
+	unsigned bs = crypto_skcipher_blocksize(tfm);
 	struct skcipher_sg_list *sgl;
 	struct scatterlist *sg;
 	int err = -EAGAIN;
@@ -947,7 +950,8 @@ static int skcipher_accept_parent_nokey(void *private, struct sock *sk)
 	ask->private = ctx;
 
 	skcipher_request_set_tfm(&ctx->req, skcipher);
-	skcipher_request_set_callback(&ctx->req, CRYPTO_TFM_REQ_MAY_BACKLOG,
+	skcipher_request_set_callback(&ctx->req, CRYPTO_TFM_REQ_MAY_SLEEP |
+						 CRYPTO_TFM_REQ_MAY_BACKLOG,
 				      af_alg_complete, &ctx->completion);
 
 	sk->sk_destruct = skcipher_sock_destruct;
diff --git a/crypto/crypto_user.c b/crypto/crypto_user.c
index 237f3795cfaa..43fe85f20d57 100644
--- a/crypto/crypto_user.c
+++ b/crypto/crypto_user.c
@@ -499,6 +499,7 @@ static int crypto_user_rcv_msg(struct sk_buff *skb, struct nlmsghdr *nlh)
 		if (link->dump == NULL)
 			return -EINVAL;
 
+		down_read(&crypto_alg_sem);
 		list_for_each_entry(alg, &crypto_alg_list, cra_list)
 			dump_alloc += CRYPTO_REPORT_MAXSIZE;
 
@@ -508,8 +509,11 @@ static int crypto_user_rcv_msg(struct sk_buff *skb, struct nlmsghdr *nlh)
 				.done = link->done,
 				.min_dump_alloc = dump_alloc,
 			};
-			return netlink_dump_start(crypto_nlsk, skb, nlh, &c);
+			err = netlink_dump_start(crypto_nlsk, skb, nlh, &c);
 		}
+		up_read(&crypto_alg_sem);
+
+		return err;
 	}
 
 	err = nlmsg_parse(nlh, crypto_msg_min[type], attrs, CRYPTOCFGA_MAX,
diff --git a/drivers/acpi/acpica/psargs.c b/drivers/acpi/acpica/psargs.c
index 305218539df2..d48cbed342c1 100644
--- a/drivers/acpi/acpica/psargs.c
+++ b/drivers/acpi/acpica/psargs.c
@@ -269,8 +269,7 @@ acpi_ps_get_next_namepath(struct acpi_walk_state *walk_state,
 	 */
 	if (ACPI_SUCCESS(status) &&
 	    possible_method_call && (node->type == ACPI_TYPE_METHOD)) {
-		if (GET_CURRENT_ARG_TYPE(walk_state->arg_types) ==
-		    ARGP_SUPERNAME) {
+		if (walk_state->opcode == AML_UNLOAD_OP) {
 			/*
 			 * acpi_ps_get_next_namestring has increased the AML pointer,
 			 * so we need to restore the saved AML pointer for method call.
@@ -697,7 +696,7 @@ static union acpi_parse_object *acpi_ps_get_next_field(struct acpi_parse_state
  *
  * PARAMETERS:  walk_state          - Current state
  *              parser_state        - Current parser state object
- *              arg_type            - The parser argument type (ARGP_*)
+ *              arg_type            - The argument type (AML_*_ARG)
  *              return_arg          - Where the next arg is returned
  *
  * RETURN:      Status, and an op object containing the next argument.
@@ -817,9 +816,9 @@ acpi_ps_get_next_arg(struct acpi_walk_state *walk_state,
 				return_ACPI_STATUS(AE_NO_MEMORY);
 			}
 
-			/* super_name allows argument to be a method call */
+			/* To support super_name arg of Unload */
 
-			if (arg_type == ARGP_SUPERNAME) {
+			if (walk_state->opcode == AML_UNLOAD_OP) {
 				status =
 				    acpi_ps_get_next_namepath(walk_state,
 							      parser_state, arg,
diff --git a/drivers/acpi/nfit.c b/drivers/acpi/nfit.c
index ad6d8c6b777e..35947ac87644 100644
--- a/drivers/acpi/nfit.c
+++ b/drivers/acpi/nfit.c
@@ -469,37 +469,16 @@ static void nfit_mem_find_spa_bdw(struct acpi_nfit_desc *acpi_desc,
 	nfit_mem->bdw = NULL;
 }
 
-static int nfit_mem_add(struct acpi_nfit_desc *acpi_desc,
+static void nfit_mem_init_bdw(struct acpi_nfit_desc *acpi_desc,
 		struct nfit_mem *nfit_mem, struct acpi_nfit_system_address *spa)
 {
 	u16 dcr = __to_nfit_memdev(nfit_mem)->region_index;
 	struct nfit_memdev *nfit_memdev;
 	struct nfit_flush *nfit_flush;
-	struct nfit_dcr *nfit_dcr;
 	struct nfit_bdw *nfit_bdw;
 	struct nfit_idt *nfit_idt;
 	u16 idt_idx, range_index;
 
-	list_for_each_entry(nfit_dcr, &acpi_desc->dcrs, list) {
-		if (nfit_dcr->dcr->region_index != dcr)
-			continue;
-		nfit_mem->dcr = nfit_dcr->dcr;
-		break;
-	}
-
-	if (!nfit_mem->dcr) {
-		dev_dbg(acpi_desc->dev, "SPA %d missing:%s%s\n",
-				spa->range_index, __to_nfit_memdev(nfit_mem)
-				? "" : " MEMDEV", nfit_mem->dcr ? "" : " DCR");
-		return -ENODEV;
-	}
-
-	/*
-	 * We've found enough to create an nvdimm, optionally
-	 * find an associated BDW
-	 */
-	list_add(&nfit_mem->list, &acpi_desc->dimms);
-
 	list_for_each_entry(nfit_bdw, &acpi_desc->bdws, list) {
 		if (nfit_bdw->bdw->region_index != dcr)
 			continue;
@@ -508,12 +487,12 @@ static int nfit_mem_add(struct acpi_nfit_desc *acpi_desc,
 	}
 
 	if (!nfit_mem->bdw)
-		return 0;
+		return;
 
 	nfit_mem_find_spa_bdw(acpi_desc, nfit_mem);
 
 	if (!nfit_mem->spa_bdw)
-		return 0;
+		return;
 
 	range_index = nfit_mem->spa_bdw->range_index;
 	list_for_each_entry(nfit_memdev, &acpi_desc->memdevs, list) {
@@ -538,8 +517,6 @@ static int nfit_mem_add(struct acpi_nfit_desc *acpi_desc,
 		}
 		break;
 	}
-
-	return 0;
 }
 
 static int nfit_mem_dcr_init(struct acpi_nfit_desc *acpi_desc,
@@ -548,7 +525,6 @@ static int nfit_mem_dcr_init(struct acpi_nfit_desc *acpi_desc,
 	struct nfit_mem *nfit_mem, *found;
 	struct nfit_memdev *nfit_memdev;
 	int type = nfit_spa_type(spa);
-	u16 dcr;
 
 	switch (type) {
 	case NFIT_SPA_DCR:
@@ -559,14 +535,18 @@ static int nfit_mem_dcr_init(struct acpi_nfit_desc *acpi_desc,
 	}
 
 	list_for_each_entry(nfit_memdev, &acpi_desc->memdevs, list) {
-		int rc;
+		struct nfit_dcr *nfit_dcr;
+		u32 device_handle;
+		u16 dcr;
 
 		if (nfit_memdev->memdev->range_index != spa->range_index)
 			continue;
 		found = NULL;
 		dcr = nfit_memdev->memdev->region_index;
+		device_handle = nfit_memdev->memdev->device_handle;
 		list_for_each_entry(nfit_mem, &acpi_desc->dimms, list)
-			if (__to_nfit_memdev(nfit_mem)->region_index == dcr) {
+			if (__to_nfit_memdev(nfit_mem)->device_handle
+					== device_handle) {
 				found = nfit_mem;
 				break;
 			}
@@ -579,6 +559,31 @@ static int nfit_mem_dcr_init(struct acpi_nfit_desc *acpi_desc,
 			if (!nfit_mem)
 				return -ENOMEM;
 			INIT_LIST_HEAD(&nfit_mem->list);
+			list_add(&nfit_mem->list, &acpi_desc->dimms);
+		}
+
+		list_for_each_entry(nfit_dcr, &acpi_desc->dcrs, list) {
+			if (nfit_dcr->dcr->region_index != dcr)
+				continue;
+			/*
+			 * Record the control region for the dimm.  For
+			 * the ACPI 6.1 case, where there are separate
+			 * control regions for the pmem vs blk
+			 * interfaces, be sure to record the extended
+			 * blk details.
+			 */
+			if (!nfit_mem->dcr)
+				nfit_mem->dcr = nfit_dcr->dcr;
+			else if (nfit_mem->dcr->windows == 0
+					&& nfit_dcr->dcr->windows)
+				nfit_mem->dcr = nfit_dcr->dcr;
+			break;
+		}
+
+		if (dcr && !nfit_mem->dcr) {
+			dev_err(acpi_desc->dev, "SPA %d missing DCR %d\n",
+					spa->range_index, dcr);
+			return -ENODEV;
 		}
 
 		if (type == NFIT_SPA_DCR) {
@@ -595,6 +600,7 @@ static int nfit_mem_dcr_init(struct acpi_nfit_desc *acpi_desc,
 				nfit_mem->idt_dcr = nfit_idt->idt;
 				break;
 			}
+			nfit_mem_init_bdw(acpi_desc, nfit_mem, spa);
 		} else {
 			/*
 			 * A single dimm may belong to multiple SPA-PM
@@ -603,13 +609,6 @@ static int nfit_mem_dcr_init(struct acpi_nfit_desc *acpi_desc,
 			 */
 			nfit_mem->memdev_pmem = nfit_memdev->memdev;
 		}
-
-		if (found)
-			continue;
-
-		rc = nfit_mem_add(acpi_desc, nfit_mem, spa);
-		if (rc)
-			return rc;
 	}
 
 	return 0;
@@ -1504,9 +1503,7 @@ static int ars_do_start(struct nvdimm_bus_descriptor *nd_desc,
 		case 1:
 			/* ARS unsupported, but we should never get here */
 			return 0;
-		case 2:
-			return -EINVAL;
-		case 3:
+		case 6:
 			/* ARS is in progress */
 			msleep(1000);
 			break;
@@ -1517,13 +1514,13 @@ static int ars_do_start(struct nvdimm_bus_descriptor *nd_desc,
 }
 
 static int ars_get_status(struct nvdimm_bus_descriptor *nd_desc,
-		struct nd_cmd_ars_status *cmd)
+		struct nd_cmd_ars_status *cmd, u32 size)
 {
 	int rc;
 
 	while (1) {
 		rc = nd_desc->ndctl(nd_desc, NULL, ND_CMD_ARS_STATUS, cmd,
-			sizeof(*cmd));
+			size);
 		if (rc || cmd->status & 0xffff)
 			return -ENXIO;
 
@@ -1538,6 +1535,8 @@ static int ars_get_status(struct nvdimm_bus_descriptor *nd_desc,
 		case 2:
 			/* No ARS performed for the current boot */
 			return 0;
+		case 3:
+			/* TODO: error list overflow support */
 		default:
 			return -ENXIO;
 		}
@@ -1581,6 +1580,7 @@ static int acpi_nfit_find_poison(struct acpi_nfit_desc *acpi_desc,
 	struct nd_cmd_ars_start *ars_start = NULL;
 	struct nd_cmd_ars_cap *ars_cap = NULL;
 	u64 start, len, cur, remaining;
+	u32 ars_status_size;
 	int rc;
 
 	ars_cap = kzalloc(sizeof(*ars_cap), GFP_KERNEL);
@@ -1590,14 +1590,21 @@ static int acpi_nfit_find_poison(struct acpi_nfit_desc *acpi_desc,
 	start = ndr_desc->res->start;
 	len = ndr_desc->res->end - ndr_desc->res->start + 1;
 
+	/*
+	 * If ARS is unimplemented, unsupported, or if the 'Persistent Memory
+	 * Scrub' flag in extended status is not set, skip this but continue
+	 * initialization
+	 */
 	rc = ars_get_cap(nd_desc, ars_cap, start, len);
+	if (rc == -ENOTTY) {
+		dev_dbg(acpi_desc->dev,
+			"Address Range Scrub is not implemented, won't create an error list\n");
+		rc = 0;
+		goto out;
+	}
 	if (rc)
 		goto out;
 
-	/*
-	 * If ARS is unsupported, or if the 'Persistent Memory Scrub' flag in
-	 * extended status is not set, skip this but continue initialization
-	 */
 	if ((ars_cap->status & 0xffff) ||
 		!(ars_cap->status >> 16 & ND_ARS_PERSISTENT)) {
 		dev_warn(acpi_desc->dev,
@@ -1610,14 +1617,14 @@ static int acpi_nfit_find_poison(struct acpi_nfit_desc *acpi_desc,
 	 * Check if a full-range ARS has been run. If so, use those results
 	 * without having to start a new ARS.
 	 */
-	ars_status = kzalloc(ars_cap->max_ars_out + sizeof(*ars_status),
-			GFP_KERNEL);
+	ars_status_size = ars_cap->max_ars_out;
+	ars_status = kzalloc(ars_status_size, GFP_KERNEL);
 	if (!ars_status) {
 		rc = -ENOMEM;
 		goto out;
 	}
 
-	rc = ars_get_status(nd_desc, ars_status);
+	rc = ars_get_status(nd_desc, ars_status, ars_status_size);
 	if (rc)
 		goto out;
 
@@ -1647,7 +1654,7 @@ static int acpi_nfit_find_poison(struct acpi_nfit_desc *acpi_desc,
 		if (rc)
 			goto out;
 
-		rc = ars_get_status(nd_desc, ars_status);
+		rc = ars_get_status(nd_desc, ars_status, ars_status_size);
 		if (rc)
 			goto out;
 
diff --git a/drivers/acpi/pci_irq.c b/drivers/acpi/pci_irq.c
index d30184c7f3bc..c8e169e46673 100644
--- a/drivers/acpi/pci_irq.c
+++ b/drivers/acpi/pci_irq.c
@@ -406,7 +406,7 @@ int acpi_pci_irq_enable(struct pci_dev *dev)
 		return 0;
 	}
 
-	if (pci_has_managed_irq(dev))
+	if (dev->irq_managed && dev->irq > 0)
 		return 0;
 
 	entry = acpi_pci_irq_lookup(dev, pin);
@@ -451,7 +451,8 @@ int acpi_pci_irq_enable(struct pci_dev *dev)
 		kfree(entry);
 		return rc;
 	}
-	pci_set_managed_irq(dev, rc);
+	dev->irq = rc;
+	dev->irq_managed = 1;
 
 	if (link)
 		snprintf(link_desc, sizeof(link_desc), " -> Link[%s]", link);
@@ -474,9 +475,17 @@ void acpi_pci_irq_disable(struct pci_dev *dev)
 	u8 pin;
 
 	pin = dev->pin;
-	if (!pin || !pci_has_managed_irq(dev))
+	if (!pin || !dev->irq_managed || dev->irq <= 0)
 		return;
 
+	/* Keep IOAPIC pin configuration when suspending */
+	if (dev->dev.power.is_prepared)
+		return;
+#ifdef	CONFIG_PM
+	if (dev->dev.power.runtime_status == RPM_SUSPENDING)
+		return;
+#endif
+
 	entry = acpi_pci_irq_lookup(dev, pin);
 	if (!entry)
 		return;
@@ -496,6 +505,6 @@ void acpi_pci_irq_disable(struct pci_dev *dev)
 	dev_dbg(&dev->dev, "PCI INT %c disabled\n", pin_name(pin));
 	if (gsi >= 0) {
 		acpi_unregister_gsi(gsi);
-		pci_reset_managed_irq(dev);
+		dev->irq_managed = 0;
 	}
 }
diff --git a/drivers/acpi/pci_link.c b/drivers/acpi/pci_link.c
index fa2863567eed..ededa909df2f 100644
--- a/drivers/acpi/pci_link.c
+++ b/drivers/acpi/pci_link.c
@@ -4,7 +4,6 @@
  *  Copyright (C) 2001, 2002 Andy Grover <andrew.grover@intel.com>
  *  Copyright (C) 2001, 2002 Paul Diefenbaugh <paul.s.diefenbaugh@intel.com>
  *  Copyright (C) 2002       Dominik Brodowski <devel@brodo.de>
- *  Copyright (c) 2015, The Linux Foundation. All rights reserved.
  *
  * ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
  *
@@ -438,6 +437,7 @@ static int acpi_pci_link_set(struct acpi_pci_link *link, int irq)
  * enabled system.
  */
 
+#define ACPI_MAX_IRQS		256
 #define ACPI_MAX_ISA_IRQ	16
 
 #define PIRQ_PENALTY_PCI_AVAILABLE	(0)
@@ -447,7 +447,7 @@ static int acpi_pci_link_set(struct acpi_pci_link *link, int irq)
 #define PIRQ_PENALTY_ISA_USED		(16*16*16*16*16)
 #define PIRQ_PENALTY_ISA_ALWAYS		(16*16*16*16*16*16)
 
-static int acpi_irq_isa_penalty[ACPI_MAX_ISA_IRQ] = {
+static int acpi_irq_penalty[ACPI_MAX_IRQS] = {
 	PIRQ_PENALTY_ISA_ALWAYS,	/* IRQ0 timer */
 	PIRQ_PENALTY_ISA_ALWAYS,	/* IRQ1 keyboard */
 	PIRQ_PENALTY_ISA_ALWAYS,	/* IRQ2 cascade */
@@ -464,68 +464,9 @@ static int acpi_irq_isa_penalty[ACPI_MAX_ISA_IRQ] = {
 	PIRQ_PENALTY_ISA_USED,		/* IRQ13 fpe, sometimes */
 	PIRQ_PENALTY_ISA_USED,		/* IRQ14 ide0 */
 	PIRQ_PENALTY_ISA_USED,		/* IRQ15 ide1 */
+	/* >IRQ15 */
 };
 
-struct irq_penalty_info {
-	int irq;
-	int penalty;
-	struct list_head node;
-};
-
-static LIST_HEAD(acpi_irq_penalty_list);
-
-static int acpi_irq_get_penalty(int irq)
-{
-	struct irq_penalty_info *irq_info;
-
-	if (irq < ACPI_MAX_ISA_IRQ)
-		return acpi_irq_isa_penalty[irq];
-
-	list_for_each_entry(irq_info, &acpi_irq_penalty_list, node) {
-		if (irq_info->irq == irq)
-			return irq_info->penalty;
-	}
-
-	return 0;
-}
-
-static int acpi_irq_set_penalty(int irq, int new_penalty)
-{
-	struct irq_penalty_info *irq_info;
-
-	/* see if this is a ISA IRQ */
-	if (irq < ACPI_MAX_ISA_IRQ) {
-		acpi_irq_isa_penalty[irq] = new_penalty;
-		return 0;
-	}
-
-	/* next, try to locate from the dynamic list */
-	list_for_each_entry(irq_info, &acpi_irq_penalty_list, node) {
-		if (irq_info->irq == irq) {
-			irq_info->penalty  = new_penalty;
-			return 0;
-		}
-	}
-
-	/* nope, let's allocate a slot for this IRQ */
-	irq_info = kzalloc(sizeof(*irq_info), GFP_KERNEL);
-	if (!irq_info)
-		return -ENOMEM;
-
-	irq_info->irq = irq;
-	irq_info->penalty = new_penalty;
-	list_add_tail(&irq_info->node, &acpi_irq_penalty_list);
-
-	return 0;
-}
-
-static void acpi_irq_add_penalty(int irq, int penalty)
-{
-	int curpen = acpi_irq_get_penalty(irq);
-
-	acpi_irq_set_penalty(irq, curpen + penalty);
-}
-
 int __init acpi_irq_penalty_init(void)
 {
 	struct acpi_pci_link *link;
@@ -546,16 +487,15 @@ int __init acpi_irq_penalty_init(void)
 			    link->irq.possible_count;
 
 			for (i = 0; i < link->irq.possible_count; i++) {
-				if (link->irq.possible[i] < ACPI_MAX_ISA_IRQ) {
-					int irqpos = link->irq.possible[i];
-
-					acpi_irq_add_penalty(irqpos, penalty);
-				}
+				if (link->irq.possible[i] < ACPI_MAX_ISA_IRQ)
+					acpi_irq_penalty[link->irq.
+							 possible[i]] +=
+					    penalty;
 			}
 
 		} else if (link->irq.active) {
-			acpi_irq_add_penalty(link->irq.active,
-					     PIRQ_PENALTY_PCI_POSSIBLE);
+			acpi_irq_penalty[link->irq.active] +=
+			    PIRQ_PENALTY_PCI_POSSIBLE;
 		}
 	}
 
@@ -607,12 +547,12 @@ static int acpi_pci_link_allocate(struct acpi_pci_link *link)
 		 * the use of IRQs 9, 10, 11, and >15.
 		 */
 		for (i = (link->irq.possible_count - 1); i >= 0; i--) {
-			if (acpi_irq_get_penalty(irq) >
-			    acpi_irq_get_penalty(link->irq.possible[i]))
+			if (acpi_irq_penalty[irq] >
+			    acpi_irq_penalty[link->irq.possible[i]])
 				irq = link->irq.possible[i];
 		}
 	}
-	if (acpi_irq_get_penalty(irq) >= PIRQ_PENALTY_ISA_ALWAYS) {
+	if (acpi_irq_penalty[irq] >= PIRQ_PENALTY_ISA_ALWAYS) {
 		printk(KERN_ERR PREFIX "No IRQ available for %s [%s]. "
 			    "Try pci=noacpi or acpi=off\n",
 			    acpi_device_name(link->device),
@@ -628,8 +568,7 @@ static int acpi_pci_link_allocate(struct acpi_pci_link *link)
 			    acpi_device_bid(link->device));
 		return -ENODEV;
 	} else {
-		acpi_irq_add_penalty(link->irq.active, PIRQ_PENALTY_PCI_USING);
-
+		acpi_irq_penalty[link->irq.active] += PIRQ_PENALTY_PCI_USING;
 		printk(KERN_WARNING PREFIX "%s [%s] enabled at IRQ %d\n",
 		       acpi_device_name(link->device),
 		       acpi_device_bid(link->device), link->irq.active);
@@ -839,7 +778,7 @@ static void acpi_pci_link_remove(struct acpi_device *device)
 }
 
 /*
- * modify penalty from cmdline
+ * modify acpi_irq_penalty[] from cmdline
  */
 static int __init acpi_irq_penalty_update(char *str, int used)
 {
@@ -857,10 +796,13 @@ static int __init acpi_irq_penalty_update(char *str, int used)
 		if (irq < 0)
 			continue;
 
+		if (irq >= ARRAY_SIZE(acpi_irq_penalty))
+			continue;
+
 		if (used)
-			acpi_irq_add_penalty(irq, PIRQ_PENALTY_ISA_USED);
+			acpi_irq_penalty[irq] += PIRQ_PENALTY_ISA_USED;
 		else
-			acpi_irq_set_penalty(irq, PIRQ_PENALTY_PCI_AVAILABLE);
+			acpi_irq_penalty[irq] = PIRQ_PENALTY_PCI_AVAILABLE;
 
 		if (retval != 2)	/* no next number */
 			break;
@@ -877,15 +819,18 @@ static int __init acpi_irq_penalty_update(char *str, int used)
  */
 void acpi_penalize_isa_irq(int irq, int active)
 {
-	if (irq >= 0)
-		acpi_irq_add_penalty(irq, active ?
-			PIRQ_PENALTY_ISA_USED : PIRQ_PENALTY_PCI_USING);
+	if (irq >= 0 && irq < ARRAY_SIZE(acpi_irq_penalty)) {
+		if (active)
+			acpi_irq_penalty[irq] += PIRQ_PENALTY_ISA_USED;
+		else
+			acpi_irq_penalty[irq] += PIRQ_PENALTY_PCI_USING;
+	}
 }
 
 bool acpi_isa_irq_available(int irq)
 {
-	return irq >= 0 &&
-		(acpi_irq_get_penalty(irq) < PIRQ_PENALTY_ISA_ALWAYS);
+	return irq >= 0 && (irq >= ARRAY_SIZE(acpi_irq_penalty) ||
+			    acpi_irq_penalty[irq] < PIRQ_PENALTY_ISA_ALWAYS);
 }
 
 /*
@@ -895,18 +840,13 @@ bool acpi_isa_irq_available(int irq)
  */
 void acpi_penalize_sci_irq(int irq, int trigger, int polarity)
 {
-	int penalty;
-
-	if (irq < 0)
-		return;
-
-	if (trigger != ACPI_MADT_TRIGGER_LEVEL ||
-	    polarity != ACPI_MADT_POLARITY_ACTIVE_LOW)
-		penalty = PIRQ_PENALTY_ISA_ALWAYS;
-	else
-		penalty = PIRQ_PENALTY_PCI_USING;
-
-	acpi_irq_add_penalty(irq, penalty);
+	if (irq >= 0 && irq < ARRAY_SIZE(acpi_irq_penalty)) {
+		if (trigger != ACPI_MADT_TRIGGER_LEVEL ||
+		    polarity != ACPI_MADT_POLARITY_ACTIVE_LOW)
+			acpi_irq_penalty[irq] += PIRQ_PENALTY_ISA_ALWAYS;
+		else
+			acpi_irq_penalty[irq] += PIRQ_PENALTY_PCI_USING;
+	}
 }
 
 /*
diff --git a/drivers/android/binder.c b/drivers/android/binder.c
index a39e85f9efa9..7d00b7a015ea 100644
--- a/drivers/android/binder.c
+++ b/drivers/android/binder.c
@@ -2074,7 +2074,7 @@ static int binder_thread_write(struct binder_proc *proc,
 			if (get_user(cookie, (binder_uintptr_t __user *)ptr))
 				return -EFAULT;
 
-			ptr += sizeof(void *);
+			ptr += sizeof(cookie);
 			list_for_each_entry(w, &proc->delivered_death, entry) {
 				struct binder_ref_death *tmp_death = container_of(w, struct binder_ref_death, work);
 
diff --git a/drivers/ata/ahci.c b/drivers/ata/ahci.c
index 594fcabd22cd..146dc0b8ec61 100644
--- a/drivers/ata/ahci.c
+++ b/drivers/ata/ahci.c
@@ -264,6 +264,26 @@ static const struct pci_device_id ahci_pci_tbl[] = {
 	{ PCI_VDEVICE(INTEL, 0x3b2b), board_ahci }, /* PCH RAID */
 	{ PCI_VDEVICE(INTEL, 0x3b2c), board_ahci }, /* PCH RAID */
 	{ PCI_VDEVICE(INTEL, 0x3b2f), board_ahci }, /* PCH AHCI */
+	{ PCI_VDEVICE(INTEL, 0x19b0), board_ahci }, /* DNV AHCI */
+	{ PCI_VDEVICE(INTEL, 0x19b1), board_ahci }, /* DNV AHCI */
+	{ PCI_VDEVICE(INTEL, 0x19b2), board_ahci }, /* DNV AHCI */
+	{ PCI_VDEVICE(INTEL, 0x19b3), board_ahci }, /* DNV AHCI */
+	{ PCI_VDEVICE(INTEL, 0x19b4), board_ahci }, /* DNV AHCI */
+	{ PCI_VDEVICE(INTEL, 0x19b5), board_ahci }, /* DNV AHCI */
+	{ PCI_VDEVICE(INTEL, 0x19b6), board_ahci }, /* DNV AHCI */
+	{ PCI_VDEVICE(INTEL, 0x19b7), board_ahci }, /* DNV AHCI */
+	{ PCI_VDEVICE(INTEL, 0x19bE), board_ahci }, /* DNV AHCI */
+	{ PCI_VDEVICE(INTEL, 0x19bF), board_ahci }, /* DNV AHCI */
+	{ PCI_VDEVICE(INTEL, 0x19c0), board_ahci }, /* DNV AHCI */
+	{ PCI_VDEVICE(INTEL, 0x19c1), board_ahci }, /* DNV AHCI */
+	{ PCI_VDEVICE(INTEL, 0x19c2), board_ahci }, /* DNV AHCI */
+	{ PCI_VDEVICE(INTEL, 0x19c3), board_ahci }, /* DNV AHCI */
+	{ PCI_VDEVICE(INTEL, 0x19c4), board_ahci }, /* DNV AHCI */
+	{ PCI_VDEVICE(INTEL, 0x19c5), board_ahci }, /* DNV AHCI */
+	{ PCI_VDEVICE(INTEL, 0x19c6), board_ahci }, /* DNV AHCI */
+	{ PCI_VDEVICE(INTEL, 0x19c7), board_ahci }, /* DNV AHCI */
+	{ PCI_VDEVICE(INTEL, 0x19cE), board_ahci }, /* DNV AHCI */
+	{ PCI_VDEVICE(INTEL, 0x19cF), board_ahci }, /* DNV AHCI */
 	{ PCI_VDEVICE(INTEL, 0x1c02), board_ahci }, /* CPT AHCI */
 	{ PCI_VDEVICE(INTEL, 0x1c03), board_ahci }, /* CPT AHCI */
 	{ PCI_VDEVICE(INTEL, 0x1c04), board_ahci }, /* CPT RAID */
@@ -347,15 +367,21 @@ static const struct pci_device_id ahci_pci_tbl[] = {
 	{ PCI_VDEVICE(INTEL, 0xa107), board_ahci }, /* Sunrise Point-H RAID */
 	{ PCI_VDEVICE(INTEL, 0xa10f), board_ahci }, /* Sunrise Point-H RAID */
 	{ PCI_VDEVICE(INTEL, 0x2822), board_ahci }, /* Lewisburg RAID*/
+	{ PCI_VDEVICE(INTEL, 0x2823), board_ahci }, /* Lewisburg AHCI*/
 	{ PCI_VDEVICE(INTEL, 0x2826), board_ahci }, /* Lewisburg RAID*/
+	{ PCI_VDEVICE(INTEL, 0x2827), board_ahci }, /* Lewisburg RAID*/
 	{ PCI_VDEVICE(INTEL, 0xa182), board_ahci }, /* Lewisburg AHCI*/
 	{ PCI_VDEVICE(INTEL, 0xa184), board_ahci }, /* Lewisburg RAID*/
 	{ PCI_VDEVICE(INTEL, 0xa186), board_ahci }, /* Lewisburg RAID*/
 	{ PCI_VDEVICE(INTEL, 0xa18e), board_ahci }, /* Lewisburg RAID*/
+	{ PCI_VDEVICE(INTEL, 0xa1d2), board_ahci }, /* Lewisburg RAID*/
+	{ PCI_VDEVICE(INTEL, 0xa1d6), board_ahci }, /* Lewisburg RAID*/
 	{ PCI_VDEVICE(INTEL, 0xa202), board_ahci }, /* Lewisburg AHCI*/
 	{ PCI_VDEVICE(INTEL, 0xa204), board_ahci }, /* Lewisburg RAID*/
 	{ PCI_VDEVICE(INTEL, 0xa206), board_ahci }, /* Lewisburg RAID*/
 	{ PCI_VDEVICE(INTEL, 0xa20e), board_ahci }, /* Lewisburg RAID*/
+	{ PCI_VDEVICE(INTEL, 0xa252), board_ahci }, /* Lewisburg RAID*/
+	{ PCI_VDEVICE(INTEL, 0xa256), board_ahci }, /* Lewisburg RAID*/
 
 	/* JMicron 360/1/3/5/6, match class to avoid IDE function */
 	{ PCI_VENDOR_ID_JMICRON, PCI_ANY_ID, PCI_ANY_ID, PCI_ANY_ID,
@@ -1305,6 +1331,44 @@ static inline void ahci_gtf_filter_workaround(struct ata_host *host)
 {}
 #endif
 
+#ifdef CONFIG_ARM64
+/*
+ * Due to ERRATA#22536, ThunderX needs to handle HOST_IRQ_STAT differently.
+ * Workaround is to make sure all pending IRQs are served before leaving
+ * handler.
+ */
+static irqreturn_t ahci_thunderx_irq_handler(int irq, void *dev_instance)
+{
+	struct ata_host *host = dev_instance;
+	struct ahci_host_priv *hpriv;
+	unsigned int rc = 0;
+	void __iomem *mmio;
+	u32 irq_stat, irq_masked;
+	unsigned int handled = 1;
+
+	VPRINTK("ENTER\n");
+	hpriv = host->private_data;
+	mmio = hpriv->mmio;
+	irq_stat = readl(mmio + HOST_IRQ_STAT);
+	if (!irq_stat)
+		return IRQ_NONE;
+
+	do {
+		irq_masked = irq_stat & hpriv->port_map;
+		spin_lock(&host->lock);
+		rc = ahci_handle_port_intr(host, irq_masked);
+		if (!rc)
+			handled = 0;
+		writel(irq_stat, mmio + HOST_IRQ_STAT);
+		irq_stat = readl(mmio + HOST_IRQ_STAT);
+		spin_unlock(&host->lock);
+	} while (irq_stat);
+	VPRINTK("EXIT\n");
+
+	return IRQ_RETVAL(handled);
+}
+#endif
+
 /*
  * ahci_init_msix() - optionally enable per-port MSI-X otherwise defer
  * to single msi.
@@ -1540,6 +1604,11 @@ static int ahci_init_one(struct pci_dev *pdev, const struct pci_device_id *ent)
 	if (ahci_broken_devslp(pdev))
 		hpriv->flags |= AHCI_HFLAG_NO_DEVSLP;
 
+#ifdef CONFIG_ARM64
+	if (pdev->vendor == 0x177d && pdev->device == 0xa01c)
+		hpriv->irq_handler = ahci_thunderx_irq_handler;
+#endif
+
 	/* save initial config */
 	ahci_pci_save_initial_config(pdev, hpriv);
 
diff --git a/drivers/ata/ahci.h b/drivers/ata/ahci.h
index a4faa438889c..167ba7e3b92e 100644
--- a/drivers/ata/ahci.h
+++ b/drivers/ata/ahci.h
@@ -240,8 +240,7 @@ enum {
 						        error-handling stage) */
 	AHCI_HFLAG_NO_DEVSLP		= (1 << 17), /* no device sleep */
 	AHCI_HFLAG_NO_FBS		= (1 << 18), /* no FBS */
-	AHCI_HFLAG_EDGE_IRQ		= (1 << 19), /* HOST_IRQ_STAT behaves as
-							Edge Triggered */
+
 #ifdef CONFIG_PCI_MSI
 	AHCI_HFLAG_MULTI_MSI		= (1 << 20), /* multiple PCI MSIs */
 	AHCI_HFLAG_MULTI_MSIX		= (1 << 21), /* per-port MSI-X */
@@ -250,6 +249,7 @@ enum {
 	AHCI_HFLAG_MULTI_MSI		= 0,
 	AHCI_HFLAG_MULTI_MSIX		= 0,
 #endif
+	AHCI_HFLAG_WAKE_BEFORE_STOP	= (1 << 22), /* wake before DMA stop */
 
 	/* ap->flags bits */
 
@@ -360,6 +360,7 @@ struct ahci_host_priv {
 	 * be overridden anytime before the host is activated.
 	 */
 	void			(*start_engine)(struct ata_port *ap);
+	irqreturn_t 		(*irq_handler)(int irq, void *dev_instance);
 };
 
 #ifdef CONFIG_PCI_MSI
@@ -423,6 +424,7 @@ int ahci_reset_em(struct ata_host *host);
 void ahci_print_info(struct ata_host *host, const char *scc_s);
 int ahci_host_activate(struct ata_host *host, struct scsi_host_template *sht);
 void ahci_error_handler(struct ata_port *ap);
+u32 ahci_handle_port_intr(struct ata_host *host, u32 irq_masked);
 
 static inline void __iomem *__ahci_port_base(struct ata_host *host,
 					     unsigned int port_no)
diff --git a/drivers/ata/ahci_brcmstb.c b/drivers/ata/ahci_brcmstb.c
index b36cae2fd04b..e87bcec0fd7c 100644
--- a/drivers/ata/ahci_brcmstb.c
+++ b/drivers/ata/ahci_brcmstb.c
@@ -317,6 +317,7 @@ static int brcm_ahci_probe(struct platform_device *pdev)
 	if (IS_ERR(hpriv))
 		return PTR_ERR(hpriv);
 	hpriv->plat_data = priv;
+	hpriv->flags = AHCI_HFLAG_WAKE_BEFORE_STOP;
 
 	brcm_sata_alpm_init(hpriv);
 
diff --git a/drivers/ata/ahci_xgene.c b/drivers/ata/ahci_xgene.c
index e2c6d9e0c5ac..8e3f7faf00d3 100644
--- a/drivers/ata/ahci_xgene.c
+++ b/drivers/ata/ahci_xgene.c
@@ -548,6 +548,88 @@ softreset_retry:
 	return rc;
 }
 
+/**
+ * xgene_ahci_handle_broken_edge_irq - Handle the broken irq.
+ * @ata_host: Host that recieved the irq
+ * @irq_masked: HOST_IRQ_STAT value
+ *
+ * For hardware with broken edge trigger latch
+ * the HOST_IRQ_STAT register misses the edge interrupt
+ * when clearing of HOST_IRQ_STAT register and hardware
+ * reporting the PORT_IRQ_STAT register at the
+ * same clock cycle.
+ * As such, the algorithm below outlines the workaround.
+ *
+ * 1. Read HOST_IRQ_STAT register and save the state.
+ * 2. Clear the HOST_IRQ_STAT register.
+ * 3. Read back the HOST_IRQ_STAT register.
+ * 4. If HOST_IRQ_STAT register equals to zero, then
+ *    traverse the rest of port's PORT_IRQ_STAT register
+ *    to check if an interrupt is triggered at that point else
+ *    go to step 6.
+ * 5. If PORT_IRQ_STAT register of rest ports is not equal to zero
+ *    then update the state of HOST_IRQ_STAT saved in step 1.
+ * 6. Handle port interrupts.
+ * 7. Exit
+ */
+static int xgene_ahci_handle_broken_edge_irq(struct ata_host *host,
+					     u32 irq_masked)
+{
+	struct ahci_host_priv *hpriv = host->private_data;
+	void __iomem *port_mmio;
+	int i;
+
+	if (!readl(hpriv->mmio + HOST_IRQ_STAT)) {
+		for (i = 0; i < host->n_ports; i++) {
+			if (irq_masked & (1 << i))
+				continue;
+
+			port_mmio = ahci_port_base(host->ports[i]);
+			if (readl(port_mmio + PORT_IRQ_STAT))
+				irq_masked |= (1 << i);
+		}
+	}
+
+	return ahci_handle_port_intr(host, irq_masked);
+}
+
+static irqreturn_t xgene_ahci_irq_intr(int irq, void *dev_instance)
+{
+	struct ata_host *host = dev_instance;
+	struct ahci_host_priv *hpriv;
+	unsigned int rc = 0;
+	void __iomem *mmio;
+	u32 irq_stat, irq_masked;
+
+	VPRINTK("ENTER\n");
+
+	hpriv = host->private_data;
+	mmio = hpriv->mmio;
+
+	/* sigh.  0xffffffff is a valid return from h/w */
+	irq_stat = readl(mmio + HOST_IRQ_STAT);
+	if (!irq_stat)
+		return IRQ_NONE;
+
+	irq_masked = irq_stat & hpriv->port_map;
+
+	spin_lock(&host->lock);
+
+	/*
+	 * HOST_IRQ_STAT behaves as edge triggered latch meaning that
+	 * it should be cleared before all the port events are cleared.
+	 */
+	writel(irq_stat, mmio + HOST_IRQ_STAT);
+
+	rc = xgene_ahci_handle_broken_edge_irq(host, irq_masked);
+
+	spin_unlock(&host->lock);
+
+	VPRINTK("EXIT\n");
+
+	return IRQ_RETVAL(rc);
+}
+
 static struct ata_port_operations xgene_ahci_v1_ops = {
 	.inherits = &ahci_ops,
 	.host_stop = xgene_ahci_host_stop,
@@ -779,7 +861,8 @@ skip_clk_phy:
 		hpriv->flags = AHCI_HFLAG_NO_NCQ;
 		break;
 	case XGENE_AHCI_V2:
-		hpriv->flags |= AHCI_HFLAG_YES_FBS | AHCI_HFLAG_EDGE_IRQ;
+		hpriv->flags |= AHCI_HFLAG_YES_FBS;
+		hpriv->irq_handler = xgene_ahci_irq_intr;
 		break;
 	default:
 		break;
diff --git a/drivers/ata/libahci.c b/drivers/ata/libahci.c
index d61740e78d6d..85ea5142a095 100644
--- a/drivers/ata/libahci.c
+++ b/drivers/ata/libahci.c
@@ -113,6 +113,7 @@ static ssize_t ahci_store_em_buffer(struct device *dev,
 				    const char *buf, size_t size);
 static ssize_t ahci_show_em_supported(struct device *dev,
 				      struct device_attribute *attr, char *buf);
+static irqreturn_t ahci_single_level_irq_intr(int irq, void *dev_instance);
 
 static DEVICE_ATTR(ahci_host_caps, S_IRUGO, ahci_show_host_caps, NULL);
 static DEVICE_ATTR(ahci_host_cap2, S_IRUGO, ahci_show_host_cap2, NULL);
@@ -496,8 +497,8 @@ void ahci_save_initial_config(struct device *dev, struct ahci_host_priv *hpriv)
 		}
 	}
 
-	/* fabricate port_map from cap.nr_ports */
-	if (!port_map) {
+	/* fabricate port_map from cap.nr_ports for < AHCI 1.3 */
+	if (!port_map && vers < 0x10300) {
 		port_map = (1 << ahci_nr_ports(cap)) - 1;
 		dev_warn(dev, "forcing PORTS_IMPL to 0x%x\n", port_map);
 
@@ -512,6 +513,9 @@ void ahci_save_initial_config(struct device *dev, struct ahci_host_priv *hpriv)
 
 	if (!hpriv->start_engine)
 		hpriv->start_engine = ahci_start_engine;
+
+	if (!hpriv->irq_handler)
+		hpriv->irq_handler = ahci_single_level_irq_intr;
 }
 EXPORT_SYMBOL_GPL(ahci_save_initial_config);
 
@@ -593,8 +597,22 @@ EXPORT_SYMBOL_GPL(ahci_start_engine);
 int ahci_stop_engine(struct ata_port *ap)
 {
 	void __iomem *port_mmio = ahci_port_base(ap);
+	struct ahci_host_priv *hpriv = ap->host->private_data;
 	u32 tmp;
 
+	/*
+	 * On some controllers, stopping a port's DMA engine while the port
+	 * is in ALPM state (partial or slumber) results in failures on
+	 * subsequent DMA engine starts.  For those controllers, put the
+	 * port back in active state before stopping its DMA engine.
+	 */
+	if ((hpriv->flags & AHCI_HFLAG_WAKE_BEFORE_STOP) &&
+	    (ap->link.lpm_policy > ATA_LPM_MAX_POWER) &&
+	    ahci_set_lpm(&ap->link, ATA_LPM_MAX_POWER, ATA_LPM_WAKE_ONLY)) {
+		dev_err(ap->host->dev, "Failed to wake up port before engine stop\n");
+		return -EIO;
+	}
+
 	tmp = readl(port_mmio + PORT_CMD);
 
 	/* check if the HBA is idle */
@@ -689,6 +707,9 @@ static int ahci_set_lpm(struct ata_link *link, enum ata_lpm_policy policy,
 	void __iomem *port_mmio = ahci_port_base(ap);
 
 	if (policy != ATA_LPM_MAX_POWER) {
+		/* wakeup flag only applies to the max power policy */
+		hints &= ~ATA_LPM_WAKE_ONLY;
+
 		/*
 		 * Disable interrupts on Phy Ready. This keeps us from
 		 * getting woken up due to spurious phy ready
@@ -704,7 +725,8 @@ static int ahci_set_lpm(struct ata_link *link, enum ata_lpm_policy policy,
 		u32 cmd = readl(port_mmio + PORT_CMD);
 
 		if (policy == ATA_LPM_MAX_POWER || !(hints & ATA_LPM_HIPM)) {
-			cmd &= ~(PORT_CMD_ASP | PORT_CMD_ALPE);
+			if (!(hints & ATA_LPM_WAKE_ONLY))
+				cmd &= ~(PORT_CMD_ASP | PORT_CMD_ALPE);
 			cmd |= PORT_CMD_ICC_ACTIVE;
 
 			writel(cmd, port_mmio + PORT_CMD);
@@ -712,6 +734,9 @@ static int ahci_set_lpm(struct ata_link *link, enum ata_lpm_policy policy,
 
 			/* wait 10ms to be sure we've come out of LPM state */
 			ata_msleep(ap, 10);
+
+			if (hints & ATA_LPM_WAKE_ONLY)
+				return 0;
 		} else {
 			cmd |= PORT_CMD_ALPE;
 			if (policy == ATA_LPM_MIN_POWER)
@@ -1143,8 +1168,7 @@ static void ahci_port_init(struct device *dev, struct ata_port *ap,
 
 	/* mark esata ports */
 	tmp = readl(port_mmio + PORT_CMD);
-	if ((tmp & PORT_CMD_HPCP) ||
-	    ((tmp & PORT_CMD_ESP) && (hpriv->cap & HOST_CAP_SXS)))
+	if ((tmp & PORT_CMD_ESP) && (hpriv->cap & HOST_CAP_SXS))
 		ap->pflags |= ATA_PFLAG_EXTERNAL;
 }
 
@@ -1825,7 +1849,7 @@ static irqreturn_t ahci_multi_irqs_intr_hard(int irq, void *dev_instance)
 	return IRQ_HANDLED;
 }
 
-static u32 ahci_handle_port_intr(struct ata_host *host, u32 irq_masked)
+u32 ahci_handle_port_intr(struct ata_host *host, u32 irq_masked)
 {
 	unsigned int i, handled = 0;
 
@@ -1851,43 +1875,7 @@ static u32 ahci_handle_port_intr(struct ata_host *host, u32 irq_masked)
 
 	return handled;
 }
-
-static irqreturn_t ahci_single_edge_irq_intr(int irq, void *dev_instance)
-{
-	struct ata_host *host = dev_instance;
-	struct ahci_host_priv *hpriv;
-	unsigned int rc = 0;
-	void __iomem *mmio;
-	u32 irq_stat, irq_masked;
-
-	VPRINTK("ENTER\n");
-
-	hpriv = host->private_data;
-	mmio = hpriv->mmio;
-
-	/* sigh.  0xffffffff is a valid return from h/w */
-	irq_stat = readl(mmio + HOST_IRQ_STAT);
-	if (!irq_stat)
-		return IRQ_NONE;
-
-	irq_masked = irq_stat & hpriv->port_map;
-
-	spin_lock(&host->lock);
-
-	/*
-	 * HOST_IRQ_STAT behaves as edge triggered latch meaning that
-	 * it should be cleared before all the port events are cleared.
-	 */
-	writel(irq_stat, mmio + HOST_IRQ_STAT);
-
-	rc = ahci_handle_port_intr(host, irq_masked);
-
-	spin_unlock(&host->lock);
-
-	VPRINTK("EXIT\n");
-
-	return IRQ_RETVAL(rc);
-}
+EXPORT_SYMBOL_GPL(ahci_handle_port_intr);
 
 static irqreturn_t ahci_single_level_irq_intr(int irq, void *dev_instance)
 {
@@ -2514,14 +2502,18 @@ int ahci_host_activate(struct ata_host *host, struct scsi_host_template *sht)
 	int irq = hpriv->irq;
 	int rc;
 
-	if (hpriv->flags & (AHCI_HFLAG_MULTI_MSI | AHCI_HFLAG_MULTI_MSIX))
+	if (hpriv->flags & (AHCI_HFLAG_MULTI_MSI | AHCI_HFLAG_MULTI_MSIX)) {
+		if (hpriv->irq_handler)
+			dev_warn(host->dev, "both AHCI_HFLAG_MULTI_MSI flag set \
+				 and custom irq handler implemented\n");
+
 		rc = ahci_host_activate_multi_irqs(host, sht);
-	else if (hpriv->flags & AHCI_HFLAG_EDGE_IRQ)
-		rc = ata_host_activate(host, irq, ahci_single_edge_irq_intr,
-				       IRQF_SHARED, sht);
-	else
-		rc = ata_host_activate(host, irq, ahci_single_level_irq_intr,
+	} else {
+		rc = ata_host_activate(host, irq, hpriv->irq_handler,
 				       IRQF_SHARED, sht);
+	}
+
+
 	return rc;
 }
 EXPORT_SYMBOL_GPL(ahci_host_activate);
diff --git a/drivers/ata/libata-core.c b/drivers/ata/libata-core.c
index cbb74719d2c1..55e257c268dd 100644
--- a/drivers/ata/libata-core.c
+++ b/drivers/ata/libata-core.c
@@ -4125,6 +4125,7 @@ static const struct ata_blacklist_entry ata_device_blacklist [] = {
 	{ "SAMSUNG CD-ROM SN-124", "N001",	ATA_HORKAGE_NODMA },
 	{ "Seagate STT20000A", NULL,		ATA_HORKAGE_NODMA },
 	{ " 2GB ATA Flash Disk", "ADMA428M",	ATA_HORKAGE_NODMA },
+	{ "VRFDFC22048UCHC-TE*", NULL,		ATA_HORKAGE_NODMA },
 	/* Odd clown on sil3726/4726 PMPs */
 	{ "Config  Disk",	NULL,		ATA_HORKAGE_DISABLE },
 
diff --git a/drivers/ata/libata-scsi.c b/drivers/ata/libata-scsi.c
index 7e959f90c020..e417e1a1d02c 100644
--- a/drivers/ata/libata-scsi.c
+++ b/drivers/ata/libata-scsi.c
@@ -675,19 +675,18 @@ static int ata_ioc32(struct ata_port *ap)
 int ata_sas_scsi_ioctl(struct ata_port *ap, struct scsi_device *scsidev,
 		     int cmd, void __user *arg)
 {
-	int val = -EINVAL, rc = -EINVAL;
+	unsigned long val;
+	int rc = -EINVAL;
 	unsigned long flags;
 
 	switch (cmd) {
-	case ATA_IOC_GET_IO32:
+	case HDIO_GET_32BIT:
 		spin_lock_irqsave(ap->lock, flags);
 		val = ata_ioc32(ap);
 		spin_unlock_irqrestore(ap->lock, flags);
-		if (copy_to_user(arg, &val, 1))
-			return -EFAULT;
-		return 0;
+		return put_user(val, (unsigned long __user *)arg);
 
-	case ATA_IOC_SET_IO32:
+	case HDIO_SET_32BIT:
 		val = (unsigned long) arg;
 		rc = 0;
 		spin_lock_irqsave(ap->lock, flags);
diff --git a/drivers/ata/libata-sff.c b/drivers/ata/libata-sff.c
index cdf6215a9a22..051b6158d1b7 100644
--- a/drivers/ata/libata-sff.c
+++ b/drivers/ata/libata-sff.c
@@ -997,12 +997,9 @@ static inline int ata_hsm_ok_in_wq(struct ata_port *ap,
 static void ata_hsm_qc_complete(struct ata_queued_cmd *qc, int in_wq)
 {
 	struct ata_port *ap = qc->ap;
-	unsigned long flags;
 
 	if (ap->ops->error_handler) {
 		if (in_wq) {
-			spin_lock_irqsave(ap->lock, flags);
-
 			/* EH might have kicked in while host lock is
 			 * released.
 			 */
@@ -1014,8 +1011,6 @@ static void ata_hsm_qc_complete(struct ata_queued_cmd *qc, int in_wq)
 				} else
 					ata_port_freeze(ap);
 			}
-
-			spin_unlock_irqrestore(ap->lock, flags);
 		} else {
 			if (likely(!(qc->err_mask & AC_ERR_HSM)))
 				ata_qc_complete(qc);
@@ -1024,10 +1019,8 @@ static void ata_hsm_qc_complete(struct ata_queued_cmd *qc, int in_wq)
 		}
 	} else {
 		if (in_wq) {
-			spin_lock_irqsave(ap->lock, flags);
 			ata_sff_irq_on(ap);
 			ata_qc_complete(qc);
-			spin_unlock_irqrestore(ap->lock, flags);
 		} else
 			ata_qc_complete(qc);
 	}
@@ -1048,9 +1041,10 @@ int ata_sff_hsm_move(struct ata_port *ap, struct ata_queued_cmd *qc,
 {
 	struct ata_link *link = qc->dev->link;
 	struct ata_eh_info *ehi = &link->eh_info;
-	unsigned long flags = 0;
 	int poll_next;
 
+	lockdep_assert_held(ap->lock);
+
 	WARN_ON_ONCE((qc->flags & ATA_QCFLAG_ACTIVE) == 0);
 
 	/* Make sure ata_sff_qc_issue() does not throw things
@@ -1112,14 +1106,6 @@ fsm_start:
 			}
 		}
 
-		/* Send the CDB (atapi) or the first data block (ata pio out).
-		 * During the state transition, interrupt handler shouldn't
-		 * be invoked before the data transfer is complete and
-		 * hsm_task_state is changed. Hence, the following locking.
-		 */
-		if (in_wq)
-			spin_lock_irqsave(ap->lock, flags);
-
 		if (qc->tf.protocol == ATA_PROT_PIO) {
 			/* PIO data out protocol.
 			 * send first data block.
@@ -1135,9 +1121,6 @@ fsm_start:
 			/* send CDB */
 			atapi_send_cdb(ap, qc);
 
-		if (in_wq)
-			spin_unlock_irqrestore(ap->lock, flags);
-
 		/* if polling, ata_sff_pio_task() handles the rest.
 		 * otherwise, interrupt handler takes over from here.
 		 */
@@ -1296,7 +1279,8 @@ fsm_start:
 		break;
 	default:
 		poll_next = 0;
-		BUG();
+		WARN(true, "ata%d: SFF host state machine in invalid state %d",
+		     ap->print_id, ap->hsm_task_state);
 	}
 
 	return poll_next;
@@ -1361,12 +1345,14 @@ static void ata_sff_pio_task(struct work_struct *work)
 	u8 status;
 	int poll_next;
 
+	spin_lock_irq(ap->lock);
+
 	BUG_ON(ap->sff_pio_task_link == NULL);
 	/* qc can be NULL if timeout occurred */
 	qc = ata_qc_from_tag(ap, link->active_tag);
 	if (!qc) {
 		ap->sff_pio_task_link = NULL;
-		return;
+		goto out_unlock;
 	}
 
 fsm_start:
@@ -1381,11 +1367,14 @@ fsm_start:
 	 */
 	status = ata_sff_busy_wait(ap, ATA_BUSY, 5);
 	if (status & ATA_BUSY) {
+		spin_unlock_irq(ap->lock);
 		ata_msleep(ap, 2);
+		spin_lock_irq(ap->lock);
+
 		status = ata_sff_busy_wait(ap, ATA_BUSY, 10);
 		if (status & ATA_BUSY) {
 			ata_sff_queue_pio_task(link, ATA_SHORT_PAUSE);
-			return;
+			goto out_unlock;
 		}
 	}
 
@@ -1402,6 +1391,8 @@ fsm_start:
 	 */
 	if (poll_next)
 		goto fsm_start;
+out_unlock:
+	spin_unlock_irq(ap->lock);
 }
 
 /**
diff --git a/drivers/ata/pata_rb532_cf.c b/drivers/ata/pata_rb532_cf.c
index 12fe0f3bb7e9..c8b6a780a290 100644
--- a/drivers/ata/pata_rb532_cf.c
+++ b/drivers/ata/pata_rb532_cf.c
@@ -32,6 +32,8 @@
 #include <linux/libata.h>
 #include <scsi/scsi_host.h>
 
+#include <asm/mach-rc32434/rb.h>
+
 #define DRV_NAME	"pata-rb532-cf"
 #define DRV_VERSION	"0.1.0"
 #define DRV_DESC	"PATA driver for RouterBOARD 532 Compact Flash"
@@ -107,6 +109,7 @@ static int rb532_pata_driver_probe(struct platform_device *pdev)
 	int gpio;
 	struct resource *res;
 	struct ata_host *ah;
+	struct cf_device *pdata;
 	struct rb532_cf_info *info;
 	int ret;
 
@@ -122,7 +125,13 @@ static int rb532_pata_driver_probe(struct platform_device *pdev)
 		return -ENOENT;
 	}
 
-	gpio = irq_to_gpio(irq);
+	pdata = dev_get_platdata(&pdev->dev);
+	if (!pdata) {
+		dev_err(&pdev->dev, "no platform data specified\n");
+		return -EINVAL;
+	}
+
+	gpio = pdata->gpio_pin;
 	if (gpio < 0) {
 		dev_err(&pdev->dev, "no GPIO found for irq%d\n", irq);
 		return -ENOENT;
diff --git a/drivers/base/component.c b/drivers/base/component.c
index 89f5cf68d80a..04a1582e80bb 100644
--- a/drivers/base/component.c
+++ b/drivers/base/component.c
@@ -206,6 +206,8 @@ static void component_match_release(struct device *master,
 		if (mc->release)
 			mc->release(master, mc->data);
 	}
+
+	kfree(match->compare);
 }
 
 static void devm_component_match_release(struct device *dev, void *res)
@@ -221,14 +223,14 @@ static int component_match_realloc(struct device *dev,
 	if (match->alloc == num)
 		return 0;
 
-	new = devm_kmalloc_array(dev, num, sizeof(*new), GFP_KERNEL);
+	new = kmalloc_array(num, sizeof(*new), GFP_KERNEL);
 	if (!new)
 		return -ENOMEM;
 
 	if (match->compare) {
 		memcpy(new, match->compare, sizeof(*new) *
 					    min(match->num, num));
-		devm_kfree(dev, match->compare);
+		kfree(match->compare);
 	}
 	match->compare = new;
 	match->alloc = num;
@@ -283,6 +285,24 @@ void component_match_add_release(struct device *master,
 }
 EXPORT_SYMBOL(component_match_add_release);
 
+static void free_master(struct master *master)
+{
+	struct component_match *match = master->match;
+	int i;
+
+	list_del(&master->node);
+
+	if (match) {
+		for (i = 0; i < match->num; i++) {
+			struct component *c = match->compare[i].component;
+			if (c)
+				c->master = NULL;
+		}
+	}
+
+	kfree(master);
+}
+
 int component_master_add_with_match(struct device *dev,
 	const struct component_master_ops *ops,
 	struct component_match *match)
@@ -309,11 +329,9 @@ int component_master_add_with_match(struct device *dev,
 
 	ret = try_to_bring_up_master(master, NULL);
 
-	if (ret < 0) {
-		/* Delete off the list if we weren't successful */
-		list_del(&master->node);
-		kfree(master);
-	}
+	if (ret < 0)
+		free_master(master);
+
 	mutex_unlock(&component_mutex);
 
 	return ret < 0 ? ret : 0;
@@ -324,25 +342,12 @@ void component_master_del(struct device *dev,
 	const struct component_master_ops *ops)
 {
 	struct master *master;
-	int i;
 
 	mutex_lock(&component_mutex);
 	master = __master_find(dev, ops);
 	if (master) {
-		struct component_match *match = master->match;
-
 		take_down_master(master);
-
-		list_del(&master->node);
-
-		if (match) {
-			for (i = 0; i < match->num; i++) {
-				struct component *c = match->compare[i].component;
-				if (c)
-					c->master = NULL;
-			}
-		}
-		kfree(master);
+		free_master(master);
 	}
 	mutex_unlock(&component_mutex);
 }
@@ -486,6 +491,8 @@ int component_add(struct device *dev, const struct component_ops *ops)
 
 	ret = try_to_bring_up_masters(component);
 	if (ret < 0) {
+		if (component->master)
+			remove_component(component->master, component);
 		list_del(&component->node);
 
 		kfree(component);
diff --git a/drivers/base/property.c b/drivers/base/property.c
index c359351d50f1..a163f2c59aa3 100644
--- a/drivers/base/property.c
+++ b/drivers/base/property.c
@@ -218,7 +218,7 @@ bool fwnode_property_present(struct fwnode_handle *fwnode, const char *propname)
 	bool ret;
 
 	ret = __fwnode_property_present(fwnode, propname);
-	if (ret == false && fwnode && fwnode->secondary)
+	if (ret == false && fwnode && !IS_ERR_OR_NULL(fwnode->secondary))
 		ret = __fwnode_property_present(fwnode->secondary, propname);
 	return ret;
 }
@@ -423,7 +423,7 @@ EXPORT_SYMBOL_GPL(device_property_match_string);
 	int _ret_;									\
 	_ret_ = FWNODE_PROP_READ(_fwnode_, _propname_, _type_, _proptype_,		\
 				 _val_, _nval_);					\
-	if (_ret_ == -EINVAL && _fwnode_ && _fwnode_->secondary)			\
+	if (_ret_ == -EINVAL && _fwnode_ && !IS_ERR_OR_NULL(_fwnode_->secondary))	\
 		_ret_ = FWNODE_PROP_READ(_fwnode_->secondary, _propname_, _type_,	\
 				_proptype_, _val_, _nval_);				\
 	_ret_;										\
@@ -593,7 +593,7 @@ int fwnode_property_read_string_array(struct fwnode_handle *fwnode,
 	int ret;
 
 	ret = __fwnode_property_read_string_array(fwnode, propname, val, nval);
-	if (ret == -EINVAL && fwnode && fwnode->secondary)
+	if (ret == -EINVAL && fwnode && !IS_ERR_OR_NULL(fwnode->secondary))
 		ret = __fwnode_property_read_string_array(fwnode->secondary,
 							  propname, val, nval);
 	return ret;
@@ -621,7 +621,7 @@ int fwnode_property_read_string(struct fwnode_handle *fwnode,
 	int ret;
 
 	ret = __fwnode_property_read_string(fwnode, propname, val);
-	if (ret == -EINVAL && fwnode && fwnode->secondary)
+	if (ret == -EINVAL && fwnode && !IS_ERR_OR_NULL(fwnode->secondary))
 		ret = __fwnode_property_read_string(fwnode->secondary,
 						    propname, val);
 	return ret;
diff --git a/drivers/base/regmap/regmap-mmio.c b/drivers/base/regmap/regmap-mmio.c
index 8812bfb9e3b8..eea51569f0eb 100644
--- a/drivers/base/regmap/regmap-mmio.c
+++ b/drivers/base/regmap/regmap-mmio.c
@@ -133,17 +133,17 @@ static int regmap_mmio_gather_write(void *context,
 	while (val_size) {
 		switch (ctx->val_bytes) {
 		case 1:
-			__raw_writeb(*(u8 *)val, ctx->regs + offset);
+			writeb(*(u8 *)val, ctx->regs + offset);
 			break;
 		case 2:
-			__raw_writew(*(u16 *)val, ctx->regs + offset);
+			writew(*(u16 *)val, ctx->regs + offset);
 			break;
 		case 4:
-			__raw_writel(*(u32 *)val, ctx->regs + offset);
+			writel(*(u32 *)val, ctx->regs + offset);
 			break;
 #ifdef CONFIG_64BIT
 		case 8:
-			__raw_writeq(*(u64 *)val, ctx->regs + offset);
+			writeq(*(u64 *)val, ctx->regs + offset);
 			break;
 #endif
 		default:
@@ -193,17 +193,17 @@ static int regmap_mmio_read(void *context,
 	while (val_size) {
 		switch (ctx->val_bytes) {
 		case 1:
-			*(u8 *)val = __raw_readb(ctx->regs + offset);
+			*(u8 *)val = readb(ctx->regs + offset);
 			break;
 		case 2:
-			*(u16 *)val = __raw_readw(ctx->regs + offset);
+			*(u16 *)val = readw(ctx->regs + offset);
 			break;
 		case 4:
-			*(u32 *)val = __raw_readl(ctx->regs + offset);
+			*(u32 *)val = readl(ctx->regs + offset);
 			break;
 #ifdef CONFIG_64BIT
 		case 8:
-			*(u64 *)val = __raw_readq(ctx->regs + offset);
+			*(u64 *)val = readq(ctx->regs + offset);
 			break;
 #endif
 		default:
diff --git a/drivers/block/floppy.c b/drivers/block/floppy.c
index 9e251201dd48..84708a5f8c52 100644
--- a/drivers/block/floppy.c
+++ b/drivers/block/floppy.c
@@ -866,7 +866,7 @@ static void set_fdc(int drive)
 }
 
 /* locks the driver */
-static int lock_fdc(int drive, bool interruptible)
+static int lock_fdc(int drive)
 {
 	if (WARN(atomic_read(&usage_count) == 0,
 		 "Trying to lock fdc while usage count=0\n"))
@@ -2173,7 +2173,7 @@ static int do_format(int drive, struct format_descr *tmp_format_req)
 {
 	int ret;
 
-	if (lock_fdc(drive, true))
+	if (lock_fdc(drive))
 		return -EINTR;
 
 	set_floppy(drive);
@@ -2960,7 +2960,7 @@ static int user_reset_fdc(int drive, int arg, bool interruptible)
 {
 	int ret;
 
-	if (lock_fdc(drive, interruptible))
+	if (lock_fdc(drive))
 		return -EINTR;
 
 	if (arg == FD_RESET_ALWAYS)
@@ -3243,7 +3243,7 @@ static int set_geometry(unsigned int cmd, struct floppy_struct *g,
 		if (!capable(CAP_SYS_ADMIN))
 			return -EPERM;
 		mutex_lock(&open_lock);
-		if (lock_fdc(drive, true)) {
+		if (lock_fdc(drive)) {
 			mutex_unlock(&open_lock);
 			return -EINTR;
 		}
@@ -3263,7 +3263,7 @@ static int set_geometry(unsigned int cmd, struct floppy_struct *g,
 	} else {
 		int oldStretch;
 
-		if (lock_fdc(drive, true))
+		if (lock_fdc(drive))
 			return -EINTR;
 		if (cmd != FDDEFPRM) {
 			/* notice a disk change immediately, else
@@ -3349,7 +3349,7 @@ static int get_floppy_geometry(int drive, int type, struct floppy_struct **g)
 	if (type)
 		*g = &floppy_type[type];
 	else {
-		if (lock_fdc(drive, false))
+		if (lock_fdc(drive))
 			return -EINTR;
 		if (poll_drive(false, 0) == -EINTR)
 			return -EINTR;
@@ -3433,7 +3433,7 @@ static int fd_locked_ioctl(struct block_device *bdev, fmode_t mode, unsigned int
 		if (UDRS->fd_ref != 1)
 			/* somebody else has this drive open */
 			return -EBUSY;
-		if (lock_fdc(drive, true))
+		if (lock_fdc(drive))
 			return -EINTR;
 
 		/* do the actual eject. Fails on
@@ -3445,7 +3445,7 @@ static int fd_locked_ioctl(struct block_device *bdev, fmode_t mode, unsigned int
 		process_fd_request();
 		return ret;
 	case FDCLRPRM:
-		if (lock_fdc(drive, true))
+		if (lock_fdc(drive))
 			return -EINTR;
 		current_type[drive] = NULL;
 		floppy_sizes[drive] = MAX_DISK_SIZE << 1;
@@ -3467,7 +3467,7 @@ static int fd_locked_ioctl(struct block_device *bdev, fmode_t mode, unsigned int
 		UDP->flags &= ~FTD_MSG;
 		return 0;
 	case FDFMTBEG:
-		if (lock_fdc(drive, true))
+		if (lock_fdc(drive))
 			return -EINTR;
 		if (poll_drive(true, FD_RAW_NEED_DISK) == -EINTR)
 			return -EINTR;
@@ -3484,7 +3484,7 @@ static int fd_locked_ioctl(struct block_device *bdev, fmode_t mode, unsigned int
 		return do_format(drive, &inparam.f);
 	case FDFMTEND:
 	case FDFLUSH:
-		if (lock_fdc(drive, true))
+		if (lock_fdc(drive))
 			return -EINTR;
 		return invalidate_drive(bdev);
 	case FDSETEMSGTRESH:
@@ -3507,7 +3507,7 @@ static int fd_locked_ioctl(struct block_device *bdev, fmode_t mode, unsigned int
 		outparam = UDP;
 		break;
 	case FDPOLLDRVSTAT:
-		if (lock_fdc(drive, true))
+		if (lock_fdc(drive))
 			return -EINTR;
 		if (poll_drive(true, FD_RAW_NEED_DISK) == -EINTR)
 			return -EINTR;
@@ -3530,7 +3530,7 @@ static int fd_locked_ioctl(struct block_device *bdev, fmode_t mode, unsigned int
 	case FDRAWCMD:
 		if (type)
 			return -EINVAL;
-		if (lock_fdc(drive, true))
+		if (lock_fdc(drive))
 			return -EINTR;
 		set_floppy(drive);
 		i = raw_cmd_ioctl(cmd, (void __user *)param);
@@ -3539,7 +3539,7 @@ static int fd_locked_ioctl(struct block_device *bdev, fmode_t mode, unsigned int
 		process_fd_request();
 		return i;
 	case FDTWADDLE:
-		if (lock_fdc(drive, true))
+		if (lock_fdc(drive))
 			return -EINTR;
 		twaddle();
 		process_fd_request();
@@ -3663,6 +3663,11 @@ static int floppy_open(struct block_device *bdev, fmode_t mode)
 
 	opened_bdev[drive] = bdev;
 
+	if (!(mode & (FMODE_READ|FMODE_WRITE))) {
+		res = -EINVAL;
+		goto out;
+	}
+
 	res = -ENXIO;
 
 	if (!floppy_track_buffer) {
@@ -3706,21 +3711,20 @@ static int floppy_open(struct block_device *bdev, fmode_t mode)
 	if (UFDCS->rawcmd == 1)
 		UFDCS->rawcmd = 2;
 
-	if (!(mode & FMODE_NDELAY)) {
-		if (mode & (FMODE_READ|FMODE_WRITE)) {
-			UDRS->last_checked = 0;
-			clear_bit(FD_OPEN_SHOULD_FAIL_BIT, &UDRS->flags);
-			check_disk_change(bdev);
-			if (test_bit(FD_DISK_CHANGED_BIT, &UDRS->flags))
-				goto out;
-			if (test_bit(FD_OPEN_SHOULD_FAIL_BIT, &UDRS->flags))
-				goto out;
-		}
-		res = -EROFS;
-		if ((mode & FMODE_WRITE) &&
-		    !test_bit(FD_DISK_WRITABLE_BIT, &UDRS->flags))
-			goto out;
-	}
+	UDRS->last_checked = 0;
+	clear_bit(FD_OPEN_SHOULD_FAIL_BIT, &UDRS->flags);
+	check_disk_change(bdev);
+	if (test_bit(FD_DISK_CHANGED_BIT, &UDRS->flags))
+		goto out;
+	if (test_bit(FD_OPEN_SHOULD_FAIL_BIT, &UDRS->flags))
+		goto out;
+
+	res = -EROFS;
+
+	if ((mode & FMODE_WRITE) &&
+			!test_bit(FD_DISK_WRITABLE_BIT, &UDRS->flags))
+		goto out;
+
 	mutex_unlock(&open_lock);
 	mutex_unlock(&floppy_mutex);
 	return 0;
@@ -3748,7 +3752,8 @@ static unsigned int floppy_check_events(struct gendisk *disk,
 		return DISK_EVENT_MEDIA_CHANGE;
 
 	if (time_after(jiffies, UDRS->last_checked + UDP->checkfreq)) {
-		lock_fdc(drive, false);
+		if (lock_fdc(drive))
+			return -EINTR;
 		poll_drive(false, 0);
 		process_fd_request();
 	}
@@ -3847,7 +3852,9 @@ static int floppy_revalidate(struct gendisk *disk)
 			 "VFS: revalidate called on non-open device.\n"))
 			return -EFAULT;
 
-		lock_fdc(drive, false);
+		res = lock_fdc(drive);
+		if (res)
+			return res;
 		cf = (test_bit(FD_DISK_CHANGED_BIT, &UDRS->flags) ||
 		      test_bit(FD_VERIFY_BIT, &UDRS->flags));
 		if (!(cf || test_bit(drive, &fake_change) || drive_no_geom(drive))) {
diff --git a/drivers/block/null_blk.c b/drivers/block/null_blk.c
index 8ba1e97d573c..64a7b5971b57 100644
--- a/drivers/block/null_blk.c
+++ b/drivers/block/null_blk.c
@@ -478,7 +478,7 @@ static int null_lnvm_id(struct nvm_dev *dev, struct nvm_id *id)
 	id->ver_id = 0x1;
 	id->vmnt = 0;
 	id->cgrps = 1;
-	id->cap = 0x3;
+	id->cap = 0x2;
 	id->dom = 0x1;
 
 	id->ppaf.blk_offset = 0;
@@ -707,9 +707,7 @@ static int null_add_dev(void)
 	queue_flag_set_unlocked(QUEUE_FLAG_NONROT, nullb->q);
 	queue_flag_clear_unlocked(QUEUE_FLAG_ADD_RANDOM, nullb->q);
 
-
 	mutex_lock(&lock);
-	list_add_tail(&nullb->list, &nullb_list);
 	nullb->index = nullb_indexes++;
 	mutex_unlock(&lock);
 
@@ -743,6 +741,10 @@ static int null_add_dev(void)
 	strncpy(disk->disk_name, nullb->disk_name, DISK_NAME_LEN);
 
 	add_disk(disk);
+
+	mutex_lock(&lock);
+	list_add_tail(&nullb->list, &nullb_list);
+	mutex_unlock(&lock);
 done:
 	return 0;
 
diff --git a/drivers/block/xen-blkfront.c b/drivers/block/xen-blkfront.c
index 8a8dc91c39f7..83eb9e6bf8b0 100644
--- a/drivers/block/xen-blkfront.c
+++ b/drivers/block/xen-blkfront.c
@@ -1873,6 +1873,43 @@ again:
 	return err;
 }
 
+static int negotiate_mq(struct blkfront_info *info)
+{
+	unsigned int backend_max_queues = 0;
+	int err;
+	unsigned int i;
+
+	BUG_ON(info->nr_rings);
+
+	/* Check if backend supports multiple queues. */
+	err = xenbus_scanf(XBT_NIL, info->xbdev->otherend,
+			   "multi-queue-max-queues", "%u", &backend_max_queues);
+	if (err < 0)
+		backend_max_queues = 1;
+
+	info->nr_rings = min(backend_max_queues, xen_blkif_max_queues);
+	/* We need at least one ring. */
+	if (!info->nr_rings)
+		info->nr_rings = 1;
+
+	info->rinfo = kzalloc(sizeof(struct blkfront_ring_info) * info->nr_rings, GFP_KERNEL);
+	if (!info->rinfo) {
+		xenbus_dev_fatal(info->xbdev, -ENOMEM, "allocating ring_info structure");
+		return -ENOMEM;
+	}
+
+	for (i = 0; i < info->nr_rings; i++) {
+		struct blkfront_ring_info *rinfo;
+
+		rinfo = &info->rinfo[i];
+		INIT_LIST_HEAD(&rinfo->indirect_pages);
+		INIT_LIST_HEAD(&rinfo->grants);
+		rinfo->dev_info = info;
+		INIT_WORK(&rinfo->work, blkif_restart_queue);
+		spin_lock_init(&rinfo->ring_lock);
+	}
+	return 0;
+}
 /**
  * Entry point to this code when a new device is created.  Allocate the basic
  * structures and the ring buffer for communication with the backend, and
@@ -1883,9 +1920,7 @@ static int blkfront_probe(struct xenbus_device *dev,
 			  const struct xenbus_device_id *id)
 {
 	int err, vdevice;
-	unsigned int r_index;
 	struct blkfront_info *info;
-	unsigned int backend_max_queues = 0;
 
 	/* FIXME: Use dynamic device id if this is not set. */
 	err = xenbus_scanf(XBT_NIL, dev->nodename,
@@ -1936,33 +1971,10 @@ static int blkfront_probe(struct xenbus_device *dev,
 	}
 
 	info->xbdev = dev;
-	/* Check if backend supports multiple queues. */
-	err = xenbus_scanf(XBT_NIL, info->xbdev->otherend,
-			   "multi-queue-max-queues", "%u", &backend_max_queues);
-	if (err < 0)
-		backend_max_queues = 1;
-
-	info->nr_rings = min(backend_max_queues, xen_blkif_max_queues);
-	/* We need at least one ring. */
-	if (!info->nr_rings)
-		info->nr_rings = 1;
-
-	info->rinfo = kzalloc(sizeof(struct blkfront_ring_info) * info->nr_rings, GFP_KERNEL);
-	if (!info->rinfo) {
-		xenbus_dev_fatal(dev, -ENOMEM, "allocating ring_info structure");
+	err = negotiate_mq(info);
+	if (err) {
 		kfree(info);
-		return -ENOMEM;
-	}
-
-	for (r_index = 0; r_index < info->nr_rings; r_index++) {
-		struct blkfront_ring_info *rinfo;
-
-		rinfo = &info->rinfo[r_index];
-		INIT_LIST_HEAD(&rinfo->indirect_pages);
-		INIT_LIST_HEAD(&rinfo->grants);
-		rinfo->dev_info = info;
-		INIT_WORK(&rinfo->work, blkif_restart_queue);
-		spin_lock_init(&rinfo->ring_lock);
+		return err;
 	}
 
 	mutex_init(&info->mutex);
@@ -2123,12 +2135,16 @@ static int blkif_recover(struct blkfront_info *info)
 static int blkfront_resume(struct xenbus_device *dev)
 {
 	struct blkfront_info *info = dev_get_drvdata(&dev->dev);
-	int err;
+	int err = 0;
 
 	dev_dbg(&dev->dev, "blkfront_resume: %s\n", dev->nodename);
 
 	blkif_free(info, info->connected == BLKIF_STATE_CONNECTED);
 
+	err = negotiate_mq(info);
+	if (err)
+		return err;
+
 	err = talk_to_blkback(dev, info);
 
 	/*
diff --git a/drivers/char/agp/intel-gtt.c b/drivers/char/agp/intel-gtt.c
index 1341a94cc779..aef87fdbd187 100644
--- a/drivers/char/agp/intel-gtt.c
+++ b/drivers/char/agp/intel-gtt.c
@@ -555,8 +555,10 @@ static unsigned int intel_gtt_mappable_entries(void)
 static void intel_gtt_teardown_scratch_page(void)
 {
 	set_pages_wb(intel_private.scratch_page, 1);
-	pci_unmap_page(intel_private.pcidev, intel_private.scratch_page_dma,
-		       PAGE_SIZE, PCI_DMA_BIDIRECTIONAL);
+	if (intel_private.needs_dmar)
+		pci_unmap_page(intel_private.pcidev,
+			       intel_private.scratch_page_dma,
+			       PAGE_SIZE, PCI_DMA_BIDIRECTIONAL);
 	__free_page(intel_private.scratch_page);
 }
 
@@ -1346,16 +1348,6 @@ int intel_gmch_probe(struct pci_dev *bridge_pdev, struct pci_dev *gpu_pdev,
 {
 	int i, mask;
 
-	/*
-	 * Can be called from the fake agp driver but also directly from
-	 * drm/i915.ko. Hence we need to check whether everything is set up
-	 * already.
-	 */
-	if (intel_private.driver) {
-		intel_private.refcount++;
-		return 1;
-	}
-
 	for (i = 0; intel_gtt_chipsets[i].name != NULL; i++) {
 		if (gpu_pdev) {
 			if (gpu_pdev->device ==
@@ -1376,16 +1368,26 @@ int intel_gmch_probe(struct pci_dev *bridge_pdev, struct pci_dev *gpu_pdev,
 	if (!intel_private.driver)
 		return 0;
 
-	intel_private.refcount++;
-
 #if IS_ENABLED(CONFIG_AGP_INTEL)
 	if (bridge) {
+		if (INTEL_GTT_GEN > 1)
+			return 0;
+
 		bridge->driver = &intel_fake_agp_driver;
 		bridge->dev_private_data = &intel_private;
 		bridge->dev = bridge_pdev;
 	}
 #endif
 
+
+	/*
+	 * Can be called from the fake agp driver but also directly from
+	 * drm/i915.ko. Hence we need to check whether everything is set up
+	 * already.
+	 */
+	if (intel_private.refcount++)
+		return 1;
+
 	intel_private.bridge_dev = pci_dev_get(bridge_pdev);
 
 	dev_info(&bridge_pdev->dev, "Intel %s Chipset\n", intel_gtt_chipsets[i].name);
@@ -1430,6 +1432,8 @@ void intel_gmch_remove(void)
 	if (--intel_private.refcount)
 		return;
 
+	if (intel_private.scratch_page)
+		intel_gtt_teardown_scratch_page();
 	if (intel_private.pcidev)
 		pci_dev_put(intel_private.pcidev);
 	if (intel_private.bridge_dev)
diff --git a/drivers/char/hpet.c b/drivers/char/hpet.c
index 240b6cf1d97c..be54e5331a45 100644
--- a/drivers/char/hpet.c
+++ b/drivers/char/hpet.c
@@ -42,7 +42,7 @@
 /*
  * The High Precision Event Timer driver.
  * This driver is closely modelled after the rtc.c driver.
- * http://www.intel.com/hardwaredesign/hpetspec_1.pdf
+ * See HPET spec revision 1.
  */
 #define	HPET_USER_FREQ	(64)
 #define	HPET_DRIFT	(500)
diff --git a/drivers/char/random.c b/drivers/char/random.c
index d0da5d852d41..b583e5336630 100644
--- a/drivers/char/random.c
+++ b/drivers/char/random.c
@@ -1819,6 +1819,28 @@ unsigned int get_random_int(void)
 EXPORT_SYMBOL(get_random_int);
 
 /*
+ * Same as get_random_int(), but returns unsigned long.
+ */
+unsigned long get_random_long(void)
+{
+	__u32 *hash;
+	unsigned long ret;
+
+	if (arch_get_random_long(&ret))
+		return ret;
+
+	hash = get_cpu_var(get_random_int_hash);
+
+	hash[0] += current->pid + jiffies + random_get_entropy();
+	md5_transform(hash, random_int_secret);
+	ret = *(unsigned long *)hash;
+	put_cpu_var(get_random_int_hash);
+
+	return ret;
+}
+EXPORT_SYMBOL(get_random_long);
+
+/*
  * randomize_range() returns a start address such that
  *
  *    [...... <range> .....]
diff --git a/drivers/clk/Makefile b/drivers/clk/Makefile
index b038e3666058..bae4be6501df 100644
--- a/drivers/clk/Makefile
+++ b/drivers/clk/Makefile
@@ -43,7 +43,7 @@ obj-$(CONFIG_COMMON_CLK_SI514)		+= clk-si514.o
 obj-$(CONFIG_COMMON_CLK_SI570)		+= clk-si570.o
 obj-$(CONFIG_COMMON_CLK_CDCE925)	+= clk-cdce925.o
 obj-$(CONFIG_ARCH_STM32)		+= clk-stm32f4.o
-obj-$(CONFIG_ARCH_TANGOX)		+= clk-tango4.o
+obj-$(CONFIG_ARCH_TANGO)		+= clk-tango4.o
 obj-$(CONFIG_CLK_TWL6040)		+= clk-twl6040.o
 obj-$(CONFIG_ARCH_U300)			+= clk-u300.o
 obj-$(CONFIG_ARCH_VT8500)		+= clk-vt8500.o
diff --git a/drivers/clk/clk-gpio.c b/drivers/clk/clk-gpio.c
index 19fed65587e8..7b09a265d79f 100644
--- a/drivers/clk/clk-gpio.c
+++ b/drivers/clk/clk-gpio.c
@@ -289,7 +289,7 @@ static void __init of_gpio_clk_setup(struct device_node *node,
 
 	num_parents = of_clk_get_parent_count(node);
 	if (num_parents < 0)
-		return;
+		num_parents = 0;
 
 	data = kzalloc(sizeof(*data), GFP_KERNEL);
 	if (!data)
diff --git a/drivers/clk/clk-scpi.c b/drivers/clk/clk-scpi.c
index cd0f2726f5e0..89e9ca78bb94 100644
--- a/drivers/clk/clk-scpi.c
+++ b/drivers/clk/clk-scpi.c
@@ -299,7 +299,7 @@ static int scpi_clocks_probe(struct platform_device *pdev)
 	/* Add the virtual cpufreq device */
 	cpufreq_dev = platform_device_register_simple("scpi-cpufreq",
 						      -1, NULL, 0);
-	if (!cpufreq_dev)
+	if (IS_ERR(cpufreq_dev))
 		pr_warn("unable to register cpufreq device");
 
 	return 0;
diff --git a/drivers/clk/mvebu/dove-divider.c b/drivers/clk/mvebu/dove-divider.c
index d5c5bfa35a5a..3e0b52daa35f 100644
--- a/drivers/clk/mvebu/dove-divider.c
+++ b/drivers/clk/mvebu/dove-divider.c
@@ -247,7 +247,7 @@ static struct clk_onecell_data dove_divider_data = {
 
 void __init dove_divider_clk_init(struct device_node *np)
 {
-	void *base;
+	void __iomem *base;
 
 	base = of_iomap(np, 0);
 	if (WARN_ON(!base))
diff --git a/drivers/clk/qcom/gcc-apq8084.c b/drivers/clk/qcom/gcc-apq8084.c
index cf73e539e9f6..070037a29ea5 100644
--- a/drivers/clk/qcom/gcc-apq8084.c
+++ b/drivers/clk/qcom/gcc-apq8084.c
@@ -3587,7 +3587,6 @@ static const struct regmap_config gcc_apq8084_regmap_config = {
 	.val_bits	= 32,
 	.max_register	= 0x1fc0,
 	.fast_io	= true,
-	.val_format_endian = REGMAP_ENDIAN_LITTLE,
 };
 
 static const struct qcom_cc_desc gcc_apq8084_desc = {
diff --git a/drivers/clk/qcom/gcc-ipq806x.c b/drivers/clk/qcom/gcc-ipq806x.c
index b692ae881d6a..dd5402bac620 100644
--- a/drivers/clk/qcom/gcc-ipq806x.c
+++ b/drivers/clk/qcom/gcc-ipq806x.c
@@ -3005,7 +3005,6 @@ static const struct regmap_config gcc_ipq806x_regmap_config = {
 	.val_bits	= 32,
 	.max_register	= 0x3e40,
 	.fast_io	= true,
-	.val_format_endian = REGMAP_ENDIAN_LITTLE,
 };
 
 static const struct qcom_cc_desc gcc_ipq806x_desc = {
diff --git a/drivers/clk/qcom/gcc-msm8660.c b/drivers/clk/qcom/gcc-msm8660.c
index f6a2b14dfec4..ad413036f7c7 100644
--- a/drivers/clk/qcom/gcc-msm8660.c
+++ b/drivers/clk/qcom/gcc-msm8660.c
@@ -2702,7 +2702,6 @@ static const struct regmap_config gcc_msm8660_regmap_config = {
 	.val_bits	= 32,
 	.max_register	= 0x363c,
 	.fast_io	= true,
-	.val_format_endian = REGMAP_ENDIAN_LITTLE,
 };
 
 static const struct qcom_cc_desc gcc_msm8660_desc = {
diff --git a/drivers/clk/qcom/gcc-msm8916.c b/drivers/clk/qcom/gcc-msm8916.c
index e3bf09d7d0ef..8cc9b2868b41 100644
--- a/drivers/clk/qcom/gcc-msm8916.c
+++ b/drivers/clk/qcom/gcc-msm8916.c
@@ -3336,7 +3336,6 @@ static const struct regmap_config gcc_msm8916_regmap_config = {
 	.val_bits	= 32,
 	.max_register	= 0x80000,
 	.fast_io	= true,
-	.val_format_endian = REGMAP_ENDIAN_LITTLE,
 };
 
 static const struct qcom_cc_desc gcc_msm8916_desc = {
diff --git a/drivers/clk/qcom/gcc-msm8960.c b/drivers/clk/qcom/gcc-msm8960.c
index f31111e32d44..983dd7dc89a7 100644
--- a/drivers/clk/qcom/gcc-msm8960.c
+++ b/drivers/clk/qcom/gcc-msm8960.c
@@ -3468,7 +3468,6 @@ static const struct regmap_config gcc_msm8960_regmap_config = {
 	.val_bits	= 32,
 	.max_register	= 0x3660,
 	.fast_io	= true,
-	.val_format_endian = REGMAP_ENDIAN_LITTLE,
 };
 
 static const struct regmap_config gcc_apq8064_regmap_config = {
@@ -3477,7 +3476,6 @@ static const struct regmap_config gcc_apq8064_regmap_config = {
 	.val_bits	= 32,
 	.max_register	= 0x3880,
 	.fast_io	= true,
-	.val_format_endian = REGMAP_ENDIAN_LITTLE,
 };
 
 static const struct qcom_cc_desc gcc_msm8960_desc = {
diff --git a/drivers/clk/qcom/gcc-msm8974.c b/drivers/clk/qcom/gcc-msm8974.c
index df164d618e34..335952db309b 100644
--- a/drivers/clk/qcom/gcc-msm8974.c
+++ b/drivers/clk/qcom/gcc-msm8974.c
@@ -2680,7 +2680,6 @@ static const struct regmap_config gcc_msm8974_regmap_config = {
 	.val_bits	= 32,
 	.max_register	= 0x1fc0,
 	.fast_io	= true,
-	.val_format_endian = REGMAP_ENDIAN_LITTLE,
 };
 
 static const struct qcom_cc_desc gcc_msm8974_desc = {
diff --git a/drivers/clk/qcom/lcc-ipq806x.c b/drivers/clk/qcom/lcc-ipq806x.c
index 62e79fadd5f7..db3998e5e2d8 100644
--- a/drivers/clk/qcom/lcc-ipq806x.c
+++ b/drivers/clk/qcom/lcc-ipq806x.c
@@ -419,7 +419,6 @@ static const struct regmap_config lcc_ipq806x_regmap_config = {
 	.val_bits	= 32,
 	.max_register	= 0xfc,
 	.fast_io	= true,
-	.val_format_endian = REGMAP_ENDIAN_LITTLE,
 };
 
 static const struct qcom_cc_desc lcc_ipq806x_desc = {
diff --git a/drivers/clk/qcom/lcc-msm8960.c b/drivers/clk/qcom/lcc-msm8960.c
index bf95bb0ea1b8..4fcf9d1d233c 100644
--- a/drivers/clk/qcom/lcc-msm8960.c
+++ b/drivers/clk/qcom/lcc-msm8960.c
@@ -524,7 +524,6 @@ static const struct regmap_config lcc_msm8960_regmap_config = {
 	.val_bits	= 32,
 	.max_register	= 0xfc,
 	.fast_io	= true,
-	.val_format_endian = REGMAP_ENDIAN_LITTLE,
 };
 
 static const struct qcom_cc_desc lcc_msm8960_desc = {
diff --git a/drivers/clk/qcom/mmcc-apq8084.c b/drivers/clk/qcom/mmcc-apq8084.c
index 1e703fda8a0f..30777f9f1a43 100644
--- a/drivers/clk/qcom/mmcc-apq8084.c
+++ b/drivers/clk/qcom/mmcc-apq8084.c
@@ -3368,7 +3368,6 @@ static const struct regmap_config mmcc_apq8084_regmap_config = {
 	.val_bits	= 32,
 	.max_register	= 0x5104,
 	.fast_io	= true,
-	.val_format_endian = REGMAP_ENDIAN_LITTLE,
 };
 
 static const struct qcom_cc_desc mmcc_apq8084_desc = {
diff --git a/drivers/clk/qcom/mmcc-msm8960.c b/drivers/clk/qcom/mmcc-msm8960.c
index d73a048d3b9d..00e36192a1de 100644
--- a/drivers/clk/qcom/mmcc-msm8960.c
+++ b/drivers/clk/qcom/mmcc-msm8960.c
@@ -3029,7 +3029,6 @@ static const struct regmap_config mmcc_msm8960_regmap_config = {
 	.val_bits	= 32,
 	.max_register	= 0x334,
 	.fast_io	= true,
-	.val_format_endian = REGMAP_ENDIAN_LITTLE,
 };
 
 static const struct regmap_config mmcc_apq8064_regmap_config = {
@@ -3038,7 +3037,6 @@ static const struct regmap_config mmcc_apq8064_regmap_config = {
 	.val_bits	= 32,
 	.max_register	= 0x350,
 	.fast_io	= true,
-	.val_format_endian = REGMAP_ENDIAN_LITTLE,
 };
 
 static const struct qcom_cc_desc mmcc_msm8960_desc = {
diff --git a/drivers/clk/qcom/mmcc-msm8974.c b/drivers/clk/qcom/mmcc-msm8974.c
index bbe28ed93669..9d790bcadf25 100644
--- a/drivers/clk/qcom/mmcc-msm8974.c
+++ b/drivers/clk/qcom/mmcc-msm8974.c
@@ -2594,7 +2594,6 @@ static const struct regmap_config mmcc_msm8974_regmap_config = {
 	.val_bits	= 32,
 	.max_register	= 0x5104,
 	.fast_io	= true,
-	.val_format_endian = REGMAP_ENDIAN_LITTLE,
 };
 
 static const struct qcom_cc_desc mmcc_msm8974_desc = {
diff --git a/drivers/clk/rockchip/clk-rk3036.c b/drivers/clk/rockchip/clk-rk3036.c
index ebce98033fbb..bc7fbac83ab7 100644
--- a/drivers/clk/rockchip/clk-rk3036.c
+++ b/drivers/clk/rockchip/clk-rk3036.c
@@ -133,7 +133,7 @@ PNAME(mux_spdif_p)	= { "spdif_src", "spdif_frac", "xin12m" };
 PNAME(mux_uart0_p)	= { "uart0_src", "uart0_frac", "xin24m" };
 PNAME(mux_uart1_p)	= { "uart1_src", "uart1_frac", "xin24m" };
 PNAME(mux_uart2_p)	= { "uart2_src", "uart2_frac", "xin24m" };
-PNAME(mux_mac_p)	= { "mac_pll_src", "ext_gmac" };
+PNAME(mux_mac_p)	= { "mac_pll_src", "rmii_clkin" };
 PNAME(mux_dclk_p)	= { "dclk_lcdc", "dclk_cru" };
 
 static struct rockchip_pll_clock rk3036_pll_clks[] __initdata = {
@@ -224,16 +224,16 @@ static struct rockchip_clk_branch rk3036_clk_branches[] __initdata = {
 			RK2928_CLKGATE_CON(2), 2, GFLAGS),
 
 	COMPOSITE_NODIV(SCLK_TIMER0, "sclk_timer0", mux_timer_p, CLK_IGNORE_UNUSED,
-			RK2928_CLKSEL_CON(2), 4, 1, DFLAGS,
+			RK2928_CLKSEL_CON(2), 4, 1, MFLAGS,
 			RK2928_CLKGATE_CON(1), 0, GFLAGS),
 	COMPOSITE_NODIV(SCLK_TIMER1, "sclk_timer1", mux_timer_p, CLK_IGNORE_UNUSED,
-			RK2928_CLKSEL_CON(2), 5, 1, DFLAGS,
+			RK2928_CLKSEL_CON(2), 5, 1, MFLAGS,
 			RK2928_CLKGATE_CON(1), 1, GFLAGS),
 	COMPOSITE_NODIV(SCLK_TIMER2, "sclk_timer2", mux_timer_p, CLK_IGNORE_UNUSED,
-			RK2928_CLKSEL_CON(2), 6, 1, DFLAGS,
+			RK2928_CLKSEL_CON(2), 6, 1, MFLAGS,
 			RK2928_CLKGATE_CON(2), 4, GFLAGS),
 	COMPOSITE_NODIV(SCLK_TIMER3, "sclk_timer3", mux_timer_p, CLK_IGNORE_UNUSED,
-			RK2928_CLKSEL_CON(2), 7, 1, DFLAGS,
+			RK2928_CLKSEL_CON(2), 7, 1, MFLAGS,
 			RK2928_CLKGATE_CON(2), 5, GFLAGS),
 
 	MUX(0, "uart_pll_clk", mux_pll_src_apll_dpll_gpll_usb480m_p, 0,
@@ -242,11 +242,11 @@ static struct rockchip_clk_branch rk3036_clk_branches[] __initdata = {
 			RK2928_CLKSEL_CON(13), 0, 7, DFLAGS,
 			RK2928_CLKGATE_CON(1), 8, GFLAGS),
 	COMPOSITE_NOMUX(0, "uart1_src", "uart_pll_clk", 0,
-			RK2928_CLKSEL_CON(13), 0, 7, DFLAGS,
-			RK2928_CLKGATE_CON(1), 8, GFLAGS),
+			RK2928_CLKSEL_CON(14), 0, 7, DFLAGS,
+			RK2928_CLKGATE_CON(1), 10, GFLAGS),
 	COMPOSITE_NOMUX(0, "uart2_src", "uart_pll_clk", 0,
-			RK2928_CLKSEL_CON(13), 0, 7, DFLAGS,
-			RK2928_CLKGATE_CON(1), 8, GFLAGS),
+			RK2928_CLKSEL_CON(15), 0, 7, DFLAGS,
+			RK2928_CLKGATE_CON(1), 12, GFLAGS),
 	COMPOSITE_FRACMUX(0, "uart0_frac", "uart0_src", CLK_SET_RATE_PARENT,
 			RK2928_CLKSEL_CON(17), 0,
 			RK2928_CLKGATE_CON(1), 9, GFLAGS,
@@ -279,13 +279,13 @@ static struct rockchip_clk_branch rk3036_clk_branches[] __initdata = {
 			RK2928_CLKGATE_CON(3), 2, GFLAGS),
 
 	COMPOSITE_NODIV(0, "sclk_sdmmc_src", mux_mmc_src_p, 0,
-			RK2928_CLKSEL_CON(12), 8, 2, DFLAGS,
+			RK2928_CLKSEL_CON(12), 8, 2, MFLAGS,
 			RK2928_CLKGATE_CON(2), 11, GFLAGS),
 	DIV(SCLK_SDMMC, "sclk_sdmmc", "sclk_sdmmc_src", 0,
 			RK2928_CLKSEL_CON(11), 0, 7, DFLAGS),
 
 	COMPOSITE_NODIV(0, "sclk_sdio_src", mux_mmc_src_p, 0,
-			RK2928_CLKSEL_CON(12), 10, 2, DFLAGS,
+			RK2928_CLKSEL_CON(12), 10, 2, MFLAGS,
 			RK2928_CLKGATE_CON(2), 13, GFLAGS),
 	DIV(SCLK_SDIO, "sclk_sdio", "sclk_sdio_src", 0,
 			RK2928_CLKSEL_CON(11), 8, 7, DFLAGS),
@@ -344,12 +344,12 @@ static struct rockchip_clk_branch rk3036_clk_branches[] __initdata = {
 			RK2928_CLKGATE_CON(10), 5, GFLAGS),
 
 	COMPOSITE_NOGATE(0, "mac_pll_src", mux_pll_src_3plls_p, 0,
-			RK2928_CLKSEL_CON(21), 0, 2, MFLAGS, 4, 5, DFLAGS),
+			RK2928_CLKSEL_CON(21), 0, 2, MFLAGS, 9, 5, DFLAGS),
 	MUX(SCLK_MACREF, "mac_clk_ref", mux_mac_p, CLK_SET_RATE_PARENT,
 			RK2928_CLKSEL_CON(21), 3, 1, MFLAGS),
 
 	COMPOSITE_NOMUX(SCLK_MAC, "mac_clk", "mac_clk_ref", 0,
-			RK2928_CLKSEL_CON(21), 9, 5, DFLAGS,
+			RK2928_CLKSEL_CON(21), 4, 5, DFLAGS,
 			RK2928_CLKGATE_CON(2), 6, GFLAGS),
 
 	MUX(SCLK_HDMI, "dclk_hdmi", mux_dclk_p, 0,
diff --git a/drivers/clk/rockchip/clk-rk3368.c b/drivers/clk/rockchip/clk-rk3368.c
index be0ede522269..21f3ea909fab 100644
--- a/drivers/clk/rockchip/clk-rk3368.c
+++ b/drivers/clk/rockchip/clk-rk3368.c
@@ -780,13 +780,13 @@ static struct rockchip_clk_branch rk3368_clk_branches[] __initdata = {
 	GATE(PCLK_TSADC, "pclk_tsadc", "pclk_peri", 0, RK3368_CLKGATE_CON(20), 0, GFLAGS),
 
 	/* pclk_pd_alive gates */
-	GATE(PCLK_TIMER1, "pclk_timer1", "pclk_pd_alive", 0, RK3368_CLKGATE_CON(14), 8, GFLAGS),
-	GATE(PCLK_TIMER0, "pclk_timer0", "pclk_pd_alive", 0, RK3368_CLKGATE_CON(14), 7, GFLAGS),
-	GATE(0, "pclk_alive_niu", "pclk_pd_alive", CLK_IGNORE_UNUSED, RK3368_CLKGATE_CON(14), 12, GFLAGS),
-	GATE(PCLK_GRF, "pclk_grf", "pclk_pd_alive", CLK_IGNORE_UNUSED, RK3368_CLKGATE_CON(14), 11, GFLAGS),
-	GATE(PCLK_GPIO3, "pclk_gpio3", "pclk_pd_alive", 0, RK3368_CLKGATE_CON(14), 3, GFLAGS),
-	GATE(PCLK_GPIO2, "pclk_gpio2", "pclk_pd_alive", 0, RK3368_CLKGATE_CON(14), 2, GFLAGS),
-	GATE(PCLK_GPIO1, "pclk_gpio1", "pclk_pd_alive", 0, RK3368_CLKGATE_CON(14), 1, GFLAGS),
+	GATE(PCLK_TIMER1, "pclk_timer1", "pclk_pd_alive", 0, RK3368_CLKGATE_CON(22), 13, GFLAGS),
+	GATE(PCLK_TIMER0, "pclk_timer0", "pclk_pd_alive", 0, RK3368_CLKGATE_CON(22), 12, GFLAGS),
+	GATE(0, "pclk_alive_niu", "pclk_pd_alive", CLK_IGNORE_UNUSED, RK3368_CLKGATE_CON(22), 9, GFLAGS),
+	GATE(PCLK_GRF, "pclk_grf", "pclk_pd_alive", CLK_IGNORE_UNUSED, RK3368_CLKGATE_CON(22), 8, GFLAGS),
+	GATE(PCLK_GPIO3, "pclk_gpio3", "pclk_pd_alive", 0, RK3368_CLKGATE_CON(22), 3, GFLAGS),
+	GATE(PCLK_GPIO2, "pclk_gpio2", "pclk_pd_alive", 0, RK3368_CLKGATE_CON(22), 2, GFLAGS),
+	GATE(PCLK_GPIO1, "pclk_gpio1", "pclk_pd_alive", 0, RK3368_CLKGATE_CON(22), 1, GFLAGS),
 
 	/*
 	 * pclk_vio gates
@@ -796,12 +796,12 @@ static struct rockchip_clk_branch rk3368_clk_branches[] __initdata = {
 	GATE(0, "pclk_dphytx", "hclk_vio", CLK_IGNORE_UNUSED, RK3368_CLKGATE_CON(14), 8, GFLAGS),
 
 	/* pclk_pd_pmu gates */
-	GATE(PCLK_PMUGRF, "pclk_pmugrf", "pclk_pd_pmu", CLK_IGNORE_UNUSED, RK3368_CLKGATE_CON(17), 0, GFLAGS),
-	GATE(PCLK_GPIO0, "pclk_gpio0", "pclk_pd_pmu", 0, RK3368_CLKGATE_CON(17), 4, GFLAGS),
-	GATE(PCLK_SGRF, "pclk_sgrf", "pclk_pd_pmu", CLK_IGNORE_UNUSED, RK3368_CLKGATE_CON(17), 3, GFLAGS),
-	GATE(0, "pclk_pmu_noc", "pclk_pd_pmu", CLK_IGNORE_UNUSED, RK3368_CLKGATE_CON(17), 2, GFLAGS),
-	GATE(0, "pclk_intmem1", "pclk_pd_pmu", CLK_IGNORE_UNUSED, RK3368_CLKGATE_CON(17), 1, GFLAGS),
-	GATE(PCLK_PMU, "pclk_pmu", "pclk_pd_pmu", CLK_IGNORE_UNUSED, RK3368_CLKGATE_CON(17), 2, GFLAGS),
+	GATE(PCLK_PMUGRF, "pclk_pmugrf", "pclk_pd_pmu", CLK_IGNORE_UNUSED, RK3368_CLKGATE_CON(23), 5, GFLAGS),
+	GATE(PCLK_GPIO0, "pclk_gpio0", "pclk_pd_pmu", 0, RK3368_CLKGATE_CON(23), 4, GFLAGS),
+	GATE(PCLK_SGRF, "pclk_sgrf", "pclk_pd_pmu", CLK_IGNORE_UNUSED, RK3368_CLKGATE_CON(23), 3, GFLAGS),
+	GATE(0, "pclk_pmu_noc", "pclk_pd_pmu", CLK_IGNORE_UNUSED, RK3368_CLKGATE_CON(23), 2, GFLAGS),
+	GATE(0, "pclk_intmem1", "pclk_pd_pmu", CLK_IGNORE_UNUSED, RK3368_CLKGATE_CON(23), 1, GFLAGS),
+	GATE(PCLK_PMU, "pclk_pmu", "pclk_pd_pmu", CLK_IGNORE_UNUSED, RK3368_CLKGATE_CON(23), 0, GFLAGS),
 
 	/* timer gates */
 	GATE(0, "sclk_timer15", "xin24m", CLK_IGNORE_UNUSED, RK3368_CLKGATE_CON(24), 11, GFLAGS),
diff --git a/drivers/clk/tegra/clk-emc.c b/drivers/clk/tegra/clk-emc.c
index e1fe8f35d45c..74e7544f861b 100644
--- a/drivers/clk/tegra/clk-emc.c
+++ b/drivers/clk/tegra/clk-emc.c
@@ -450,8 +450,10 @@ static int load_timings_from_dt(struct tegra_clk_emc *tegra,
 		struct emc_timing *timing = tegra->timings + (i++);
 
 		err = load_one_timing_from_dt(tegra, timing, child);
-		if (err)
+		if (err) {
+			of_node_put(child);
 			return err;
+		}
 
 		timing->ram_code = ram_code;
 	}
@@ -499,9 +501,9 @@ struct clk *tegra_clk_register_emc(void __iomem *base, struct device_node *np,
 		 * fuses until the apbmisc driver is loaded.
 		 */
 		err = load_timings_from_dt(tegra, node, node_ram_code);
+		of_node_put(node);
 		if (err)
 			return ERR_PTR(err);
-		of_node_put(node);
 		break;
 	}
 
diff --git a/drivers/clk/tegra/clk-id.h b/drivers/clk/tegra/clk-id.h
index 19ce0738ee76..62ea38187b71 100644
--- a/drivers/clk/tegra/clk-id.h
+++ b/drivers/clk/tegra/clk-id.h
@@ -11,6 +11,7 @@ enum clk_id {
 	tegra_clk_afi,
 	tegra_clk_amx,
 	tegra_clk_amx1,
+	tegra_clk_apb2ape,
 	tegra_clk_apbdma,
 	tegra_clk_apbif,
 	tegra_clk_ape,
diff --git a/drivers/clk/tegra/clk-pll.c b/drivers/clk/tegra/clk-pll.c
index a534bfab30b3..6ac3f843e7ca 100644
--- a/drivers/clk/tegra/clk-pll.c
+++ b/drivers/clk/tegra/clk-pll.c
@@ -86,15 +86,21 @@
 #define PLLE_SS_DISABLE (PLLE_SS_CNTL_BYPASS_SS | PLLE_SS_CNTL_INTERP_RESET |\
 				PLLE_SS_CNTL_SSC_BYP)
 #define PLLE_SS_MAX_MASK 0x1ff
-#define PLLE_SS_MAX_VAL 0x25
+#define PLLE_SS_MAX_VAL_TEGRA114 0x25
+#define PLLE_SS_MAX_VAL_TEGRA210 0x21
 #define PLLE_SS_INC_MASK (0xff << 16)
 #define PLLE_SS_INC_VAL (0x1 << 16)
 #define PLLE_SS_INCINTRV_MASK (0x3f << 24)
-#define PLLE_SS_INCINTRV_VAL (0x20 << 24)
+#define PLLE_SS_INCINTRV_VAL_TEGRA114 (0x20 << 24)
+#define PLLE_SS_INCINTRV_VAL_TEGRA210 (0x23 << 24)
 #define PLLE_SS_COEFFICIENTS_MASK \
 	(PLLE_SS_MAX_MASK | PLLE_SS_INC_MASK | PLLE_SS_INCINTRV_MASK)
-#define PLLE_SS_COEFFICIENTS_VAL \
-	(PLLE_SS_MAX_VAL | PLLE_SS_INC_VAL | PLLE_SS_INCINTRV_VAL)
+#define PLLE_SS_COEFFICIENTS_VAL_TEGRA114 \
+	(PLLE_SS_MAX_VAL_TEGRA114 | PLLE_SS_INC_VAL |\
+	 PLLE_SS_INCINTRV_VAL_TEGRA114)
+#define PLLE_SS_COEFFICIENTS_VAL_TEGRA210 \
+	(PLLE_SS_MAX_VAL_TEGRA210 | PLLE_SS_INC_VAL |\
+	 PLLE_SS_INCINTRV_VAL_TEGRA210)
 
 #define PLLE_AUX_PLLP_SEL	BIT(2)
 #define PLLE_AUX_USE_LOCKDET	BIT(3)
@@ -880,7 +886,7 @@ static int clk_plle_training(struct tegra_clk_pll *pll)
 static int clk_plle_enable(struct clk_hw *hw)
 {
 	struct tegra_clk_pll *pll = to_clk_pll(hw);
-	unsigned long input_rate = clk_get_rate(clk_get_parent(hw->clk));
+	unsigned long input_rate = clk_hw_get_rate(clk_hw_get_parent(hw));
 	struct tegra_clk_pll_freq_table sel;
 	u32 val;
 	int err;
@@ -1378,7 +1384,7 @@ static int clk_plle_tegra114_enable(struct clk_hw *hw)
 	u32 val;
 	int ret;
 	unsigned long flags = 0;
-	unsigned long input_rate = clk_get_rate(clk_get_parent(hw->clk));
+	unsigned long input_rate = clk_hw_get_rate(clk_hw_get_parent(hw));
 
 	if (_get_table_rate(hw, &sel, pll->params->fixed_rate, input_rate))
 		return -EINVAL;
@@ -1401,7 +1407,7 @@ static int clk_plle_tegra114_enable(struct clk_hw *hw)
 	val |= PLLE_MISC_IDDQ_SW_CTRL;
 	val &= ~PLLE_MISC_IDDQ_SW_VALUE;
 	val |= PLLE_MISC_PLLE_PTS;
-	val |= PLLE_MISC_VREG_BG_CTRL_MASK | PLLE_MISC_VREG_CTRL_MASK;
+	val &= ~(PLLE_MISC_VREG_BG_CTRL_MASK | PLLE_MISC_VREG_CTRL_MASK);
 	pll_writel_misc(val, pll);
 	udelay(5);
 
@@ -1428,7 +1434,7 @@ static int clk_plle_tegra114_enable(struct clk_hw *hw)
 	val = pll_readl(PLLE_SS_CTRL, pll);
 	val &= ~(PLLE_SS_CNTL_CENTER | PLLE_SS_CNTL_INVERT);
 	val &= ~PLLE_SS_COEFFICIENTS_MASK;
-	val |= PLLE_SS_COEFFICIENTS_VAL;
+	val |= PLLE_SS_COEFFICIENTS_VAL_TEGRA114;
 	pll_writel(val, PLLE_SS_CTRL, pll);
 	val &= ~(PLLE_SS_CNTL_SSC_BYP | PLLE_SS_CNTL_BYPASS_SS);
 	pll_writel(val, PLLE_SS_CTRL, pll);
@@ -2012,9 +2018,9 @@ static int clk_plle_tegra210_enable(struct clk_hw *hw)
 	struct tegra_clk_pll *pll = to_clk_pll(hw);
 	struct tegra_clk_pll_freq_table sel;
 	u32 val;
-	int ret;
+	int ret = 0;
 	unsigned long flags = 0;
-	unsigned long input_rate = clk_get_rate(clk_get_parent(hw->clk));
+	unsigned long input_rate = clk_hw_get_rate(clk_hw_get_parent(hw));
 
 	if (_get_table_rate(hw, &sel, pll->params->fixed_rate, input_rate))
 		return -EINVAL;
@@ -2022,22 +2028,20 @@ static int clk_plle_tegra210_enable(struct clk_hw *hw)
 	if (pll->lock)
 		spin_lock_irqsave(pll->lock, flags);
 
+	val = pll_readl(pll->params->aux_reg, pll);
+	if (val & PLLE_AUX_SEQ_ENABLE)
+		goto out;
+
 	val = pll_readl_base(pll);
 	val &= ~BIT(30); /* Disable lock override */
 	pll_writel_base(val, pll);
 
-	val = pll_readl(pll->params->aux_reg, pll);
-	val |= PLLE_AUX_ENABLE_SWCTL;
-	val &= ~PLLE_AUX_SEQ_ENABLE;
-	pll_writel(val, pll->params->aux_reg, pll);
-	udelay(1);
-
 	val = pll_readl_misc(pll);
 	val |= PLLE_MISC_LOCK_ENABLE;
 	val |= PLLE_MISC_IDDQ_SW_CTRL;
 	val &= ~PLLE_MISC_IDDQ_SW_VALUE;
 	val |= PLLE_MISC_PLLE_PTS;
-	val |= PLLE_MISC_VREG_BG_CTRL_MASK | PLLE_MISC_VREG_CTRL_MASK;
+	val &= ~(PLLE_MISC_VREG_BG_CTRL_MASK | PLLE_MISC_VREG_CTRL_MASK);
 	pll_writel_misc(val, pll);
 	udelay(5);
 
@@ -2067,7 +2071,7 @@ static int clk_plle_tegra210_enable(struct clk_hw *hw)
 	val = pll_readl(PLLE_SS_CTRL, pll);
 	val &= ~(PLLE_SS_CNTL_CENTER | PLLE_SS_CNTL_INVERT);
 	val &= ~PLLE_SS_COEFFICIENTS_MASK;
-	val |= PLLE_SS_COEFFICIENTS_VAL;
+	val |= PLLE_SS_COEFFICIENTS_VAL_TEGRA210;
 	pll_writel(val, PLLE_SS_CTRL, pll);
 	val &= ~(PLLE_SS_CNTL_SSC_BYP | PLLE_SS_CNTL_BYPASS_SS);
 	pll_writel(val, PLLE_SS_CTRL, pll);
@@ -2104,15 +2108,25 @@ static void clk_plle_tegra210_disable(struct clk_hw *hw)
 	if (pll->lock)
 		spin_lock_irqsave(pll->lock, flags);
 
+	/* If PLLE HW sequencer is enabled, SW should not disable PLLE */
+	val = pll_readl(pll->params->aux_reg, pll);
+	if (val & PLLE_AUX_SEQ_ENABLE)
+		goto out;
+
 	val = pll_readl_base(pll);
 	val &= ~PLLE_BASE_ENABLE;
 	pll_writel_base(val, pll);
 
+	val = pll_readl(pll->params->aux_reg, pll);
+	val |= PLLE_AUX_ENABLE_SWCTL | PLLE_AUX_SS_SWCTL;
+	pll_writel(val, pll->params->aux_reg, pll);
+
 	val = pll_readl_misc(pll);
 	val |= PLLE_MISC_IDDQ_SW_CTRL | PLLE_MISC_IDDQ_SW_VALUE;
 	pll_writel_misc(val, pll);
 	udelay(1);
 
+out:
 	if (pll->lock)
 		spin_unlock_irqrestore(pll->lock, flags);
 }
diff --git a/drivers/clk/tegra/clk-tegra-periph.c b/drivers/clk/tegra/clk-tegra-periph.c
index 6ad381a888a6..ea2b9cbf9e70 100644
--- a/drivers/clk/tegra/clk-tegra-periph.c
+++ b/drivers/clk/tegra/clk-tegra-periph.c
@@ -773,7 +773,7 @@ static struct tegra_periph_init_data periph_clks[] = {
 	XUSB("xusb_dev_src", mux_clkm_pllp_pllc_pllre, CLK_SOURCE_XUSB_DEV_SRC, 95, TEGRA_PERIPH_ON_APB | TEGRA_PERIPH_NO_RESET, tegra_clk_xusb_dev_src),
 	XUSB("xusb_dev_src", mux_clkm_pllp_pllre, CLK_SOURCE_XUSB_DEV_SRC, 95, TEGRA_PERIPH_ON_APB | TEGRA_PERIPH_NO_RESET, tegra_clk_xusb_dev_src_8),
 	MUX8("dbgapb", mux_pllp_clkm_2, CLK_SOURCE_DBGAPB, 185, TEGRA_PERIPH_NO_RESET, tegra_clk_dbgapb),
-	MUX8("msenc", mux_pllc2_c_c3_pllp_plla1_clkm, CLK_SOURCE_NVENC, 219, 0, tegra_clk_nvenc),
+	MUX8("nvenc", mux_pllc2_c_c3_pllp_plla1_clkm, CLK_SOURCE_NVENC, 219, 0, tegra_clk_nvenc),
 	MUX8("nvdec", mux_pllc2_c_c3_pllp_plla1_clkm, CLK_SOURCE_NVDEC, 194, 0, tegra_clk_nvdec),
 	MUX8("nvjpg", mux_pllc2_c_c3_pllp_plla1_clkm, CLK_SOURCE_NVJPG, 195, 0, tegra_clk_nvjpg),
 	MUX8("ape", mux_plla_pllc4_out0_pllc_pllc4_out1_pllp_pllc4_out2_clkm, CLK_SOURCE_APE, 198, TEGRA_PERIPH_ON_APB, tegra_clk_ape),
@@ -782,7 +782,7 @@ static struct tegra_periph_init_data periph_clks[] = {
 	NODIV("sor1", mux_clkm_sor1_brick_sor1_src, CLK_SOURCE_SOR1, 15, MASK(1), 183, 0, tegra_clk_sor1, &sor1_lock),
 	MUX8("sdmmc_legacy", mux_pllp_out3_clkm_pllp_pllc4, CLK_SOURCE_SDMMC_LEGACY, 193, TEGRA_PERIPH_ON_APB | TEGRA_PERIPH_NO_RESET, tegra_clk_sdmmc_legacy),
 	MUX8("qspi", mux_pllp_pllc_pllc_out1_pllc4_out2_pllc4_out1_clkm_pllc4_out0, CLK_SOURCE_QSPI, 211, TEGRA_PERIPH_ON_APB, tegra_clk_qspi),
-	MUX("vii2c", mux_pllp_pllc_clkm, CLK_SOURCE_VI_I2C, 208, TEGRA_PERIPH_ON_APB, tegra_clk_vi_i2c),
+	I2C("vii2c", mux_pllp_pllc_clkm, CLK_SOURCE_VI_I2C, 208, tegra_clk_vi_i2c),
 	MUX("mipibif", mux_pllp_clkm, CLK_SOURCE_MIPIBIF, 173, TEGRA_PERIPH_ON_APB, tegra_clk_mipibif),
 	MUX("uartape", mux_pllp_pllc_clkm, CLK_SOURCE_UARTAPE, 212, TEGRA_PERIPH_ON_APB | TEGRA_PERIPH_NO_RESET, tegra_clk_uartape),
 	MUX8("tsecb", mux_pllp_pllc2_c_c3_clkm, CLK_SOURCE_TSECB, 206, 0, tegra_clk_tsecb),
@@ -829,6 +829,7 @@ static struct tegra_periph_init_data gate_clks[] = {
 	GATE("xusb_gate", "osc", 143, 0, tegra_clk_xusb_gate, 0),
 	GATE("pll_p_out_cpu", "pll_p", 223, 0, tegra_clk_pll_p_out_cpu, 0),
 	GATE("pll_p_out_adsp", "pll_p", 187, 0, tegra_clk_pll_p_out_adsp, 0),
+	GATE("apb2ape", "clk_m", 107, 0, tegra_clk_apb2ape, 0),
 };
 
 static struct tegra_periph_init_data div_clks[] = {
diff --git a/drivers/clk/tegra/clk-tegra-super-gen4.c b/drivers/clk/tegra/clk-tegra-super-gen4.c
index 4559a20e3af6..474de0f0c26d 100644
--- a/drivers/clk/tegra/clk-tegra-super-gen4.c
+++ b/drivers/clk/tegra/clk-tegra-super-gen4.c
@@ -67,7 +67,7 @@ static const char *cclk_lp_parents[] = { "clk_m", "pll_c", "clk_32k", "pll_m",
 					 "pll_p", "pll_p_out4", "unused",
 					 "unused", "pll_x", "pll_x_out0" };
 
-const struct tegra_super_gen_info tegra_super_gen_info_gen4 = {
+static const struct tegra_super_gen_info tegra_super_gen_info_gen4 = {
 	.gen = gen4,
 	.sclk_parents = sclk_parents,
 	.cclk_g_parents = cclk_g_parents,
@@ -93,7 +93,7 @@ static const char *cclk_lp_parents_gen5[] = { "clk_m", "unused", "clk_32k", "unu
 					"unused", "unused", "unused", "unused",
 					"dfllCPU_out" };
 
-const struct tegra_super_gen_info tegra_super_gen_info_gen5 = {
+static const struct tegra_super_gen_info tegra_super_gen_info_gen5 = {
 	.gen = gen5,
 	.sclk_parents = sclk_parents_gen5,
 	.cclk_g_parents = cclk_g_parents_gen5,
@@ -171,7 +171,7 @@ static void __init tegra_sclk_init(void __iomem *clk_base,
 	*dt_clk = clk;
 }
 
-void __init tegra_super_clk_init(void __iomem *clk_base,
+static void __init tegra_super_clk_init(void __iomem *clk_base,
 				void __iomem *pmc_base,
 				struct tegra_clk *tegra_clks,
 				struct tegra_clk_pll_params *params,
diff --git a/drivers/clk/tegra/clk-tegra210.c b/drivers/clk/tegra/clk-tegra210.c
index 58514c44ea83..637041fd53ad 100644
--- a/drivers/clk/tegra/clk-tegra210.c
+++ b/drivers/clk/tegra/clk-tegra210.c
@@ -59,8 +59,8 @@
 #define PLLC3_MISC3 0x50c
 
 #define PLLM_BASE 0x90
-#define PLLM_MISC0 0x9c
 #define PLLM_MISC1 0x98
+#define PLLM_MISC2 0x9c
 #define PLLP_BASE 0xa0
 #define PLLP_MISC0 0xac
 #define PLLP_MISC1 0x680
@@ -99,7 +99,7 @@
 #define PLLC4_MISC0 0x5a8
 #define PLLC4_OUT 0x5e4
 #define PLLMB_BASE 0x5e8
-#define PLLMB_MISC0 0x5ec
+#define PLLMB_MISC1 0x5ec
 #define PLLA1_BASE 0x6a4
 #define PLLA1_MISC0 0x6a8
 #define PLLA1_MISC1 0x6ac
@@ -243,7 +243,8 @@ static unsigned long tegra210_input_freq[] = {
 };
 
 static const char *mux_pllmcp_clkm[] = {
-	"pll_m", "pll_c", "pll_p", "clk_m", "pll_m_ud", "pll_c2", "pll_c3",
+	"pll_m", "pll_c", "pll_p", "clk_m", "pll_m_ud", "pll_mb", "pll_mb",
+	"pll_p",
 };
 #define mux_pllmcp_clkm_idx NULL
 
@@ -367,12 +368,12 @@ static const char *mux_pllmcp_clkm[] = {
 /* PLLMB */
 #define PLLMB_BASE_LOCK			(1 << 27)
 
-#define PLLMB_MISC0_LOCK_OVERRIDE	(1 << 18)
-#define PLLMB_MISC0_IDDQ		(1 << 17)
-#define PLLMB_MISC0_LOCK_ENABLE		(1 << 16)
+#define PLLMB_MISC1_LOCK_OVERRIDE	(1 << 18)
+#define PLLMB_MISC1_IDDQ		(1 << 17)
+#define PLLMB_MISC1_LOCK_ENABLE		(1 << 16)
 
-#define PLLMB_MISC0_DEFAULT_VALUE	0x00030000
-#define PLLMB_MISC0_WRITE_MASK		0x0007ffff
+#define PLLMB_MISC1_DEFAULT_VALUE	0x00030000
+#define PLLMB_MISC1_WRITE_MASK		0x0007ffff
 
 /* PLLP */
 #define PLLP_BASE_OVERRIDE		(1 << 28)
@@ -457,7 +458,8 @@ static void pllcx_check_defaults(struct tegra_clk_pll_params *params)
 			PLLCX_MISC3_WRITE_MASK);
 }
 
-void tegra210_pllcx_set_defaults(const char *name, struct tegra_clk_pll *pllcx)
+static void tegra210_pllcx_set_defaults(const char *name,
+					struct tegra_clk_pll *pllcx)
 {
 	pllcx->params->defaults_set = true;
 
@@ -482,22 +484,22 @@ void tegra210_pllcx_set_defaults(const char *name, struct tegra_clk_pll *pllcx)
 	udelay(1);
 }
 
-void _pllc_set_defaults(struct tegra_clk_pll *pllcx)
+static void _pllc_set_defaults(struct tegra_clk_pll *pllcx)
 {
 	tegra210_pllcx_set_defaults("PLL_C", pllcx);
 }
 
-void _pllc2_set_defaults(struct tegra_clk_pll *pllcx)
+static void _pllc2_set_defaults(struct tegra_clk_pll *pllcx)
 {
 	tegra210_pllcx_set_defaults("PLL_C2", pllcx);
 }
 
-void _pllc3_set_defaults(struct tegra_clk_pll *pllcx)
+static void _pllc3_set_defaults(struct tegra_clk_pll *pllcx)
 {
 	tegra210_pllcx_set_defaults("PLL_C3", pllcx);
 }
 
-void _plla1_set_defaults(struct tegra_clk_pll *pllcx)
+static void _plla1_set_defaults(struct tegra_clk_pll *pllcx)
 {
 	tegra210_pllcx_set_defaults("PLL_A1", pllcx);
 }
@@ -507,7 +509,7 @@ void _plla1_set_defaults(struct tegra_clk_pll *pllcx)
  * PLL with dynamic ramp and fractional SDM. Dynamic ramp is not used.
  * Fractional SDM is allowed to provide exact audio rates.
  */
-void tegra210_plla_set_defaults(struct tegra_clk_pll *plla)
+static void tegra210_plla_set_defaults(struct tegra_clk_pll *plla)
 {
 	u32 mask;
 	u32 val = readl_relaxed(clk_base + plla->params->base_reg);
@@ -559,7 +561,7 @@ void tegra210_plla_set_defaults(struct tegra_clk_pll *plla)
  * PLLD
  * PLL with fractional SDM.
  */
-void tegra210_plld_set_defaults(struct tegra_clk_pll *plld)
+static void tegra210_plld_set_defaults(struct tegra_clk_pll *plld)
 {
 	u32 val;
 	u32 mask = 0xffff;
@@ -698,7 +700,7 @@ static void plldss_defaults(const char *pll_name, struct tegra_clk_pll *plldss,
 	udelay(1);
 }
 
-void tegra210_plld2_set_defaults(struct tegra_clk_pll *plld2)
+static void tegra210_plld2_set_defaults(struct tegra_clk_pll *plld2)
 {
 	plldss_defaults("PLL_D2", plld2, PLLD2_MISC0_DEFAULT_VALUE,
 			PLLD2_MISC1_CFG_DEFAULT_VALUE,
@@ -706,7 +708,7 @@ void tegra210_plld2_set_defaults(struct tegra_clk_pll *plld2)
 			PLLD2_MISC3_CTRL2_DEFAULT_VALUE);
 }
 
-void tegra210_plldp_set_defaults(struct tegra_clk_pll *plldp)
+static void tegra210_plldp_set_defaults(struct tegra_clk_pll *plldp)
 {
 	plldss_defaults("PLL_DP", plldp, PLLDP_MISC0_DEFAULT_VALUE,
 			PLLDP_MISC1_CFG_DEFAULT_VALUE,
@@ -719,7 +721,7 @@ void tegra210_plldp_set_defaults(struct tegra_clk_pll *plldp)
  * Base and misc0 layout is the same as PLLD2/PLLDP, but no SDM/SSC support.
  * VCO is exposed to the clock tree via fixed 1/3 and 1/5 dividers.
  */
-void tegra210_pllc4_set_defaults(struct tegra_clk_pll *pllc4)
+static void tegra210_pllc4_set_defaults(struct tegra_clk_pll *pllc4)
 {
 	plldss_defaults("PLL_C4", pllc4, PLLC4_MISC0_DEFAULT_VALUE, 0, 0, 0);
 }
@@ -728,7 +730,7 @@ void tegra210_pllc4_set_defaults(struct tegra_clk_pll *pllc4)
  * PLLRE
  * VCO is exposed to the clock tree directly along with post-divider output
  */
-void tegra210_pllre_set_defaults(struct tegra_clk_pll *pllre)
+static void tegra210_pllre_set_defaults(struct tegra_clk_pll *pllre)
 {
 	u32 mask;
 	u32 val = readl_relaxed(clk_base + pllre->params->base_reg);
@@ -780,13 +782,13 @@ static void pllx_get_dyn_steps(struct clk_hw *hw, u32 *step_a, u32 *step_b)
 {
 	unsigned long input_rate;
 
-	if (!IS_ERR_OR_NULL(hw->clk)) {
+	/* cf rate */
+	if (!IS_ERR_OR_NULL(hw->clk))
 		input_rate = clk_hw_get_rate(clk_hw_get_parent(hw));
-		/* cf rate */
-		input_rate /= tegra_pll_get_fixed_mdiv(hw, input_rate);
-	} else {
+	else
 		input_rate = 38400000;
-	}
+
+	input_rate /= tegra_pll_get_fixed_mdiv(hw, input_rate);
 
 	switch (input_rate) {
 	case 12000000:
@@ -841,7 +843,7 @@ static void pllx_check_defaults(struct tegra_clk_pll *pll)
 			PLLX_MISC5_WRITE_MASK);
 }
 
-void tegra210_pllx_set_defaults(struct tegra_clk_pll *pllx)
+static void tegra210_pllx_set_defaults(struct tegra_clk_pll *pllx)
 {
 	u32 val;
 	u32 step_a, step_b;
@@ -901,7 +903,7 @@ void tegra210_pllx_set_defaults(struct tegra_clk_pll *pllx)
 }
 
 /* PLLMB */
-void tegra210_pllmb_set_defaults(struct tegra_clk_pll *pllmb)
+static void tegra210_pllmb_set_defaults(struct tegra_clk_pll *pllmb)
 {
 	u32 mask, val = readl_relaxed(clk_base + pllmb->params->base_reg);
 
@@ -914,15 +916,15 @@ void tegra210_pllmb_set_defaults(struct tegra_clk_pll *pllmb)
 		 * PLL is ON: check if defaults already set, then set those
 		 * that can be updated in flight.
 		 */
-		val = PLLMB_MISC0_DEFAULT_VALUE & (~PLLMB_MISC0_IDDQ);
-		mask = PLLMB_MISC0_LOCK_ENABLE | PLLMB_MISC0_LOCK_OVERRIDE;
+		val = PLLMB_MISC1_DEFAULT_VALUE & (~PLLMB_MISC1_IDDQ);
+		mask = PLLMB_MISC1_LOCK_ENABLE | PLLMB_MISC1_LOCK_OVERRIDE;
 		_pll_misc_chk_default(clk_base, pllmb->params, 0, val,
-				~mask & PLLMB_MISC0_WRITE_MASK);
+				~mask & PLLMB_MISC1_WRITE_MASK);
 
 		/* Enable lock detect */
 		val = readl_relaxed(clk_base + pllmb->params->ext_misc_reg[0]);
 		val &= ~mask;
-		val |= PLLMB_MISC0_DEFAULT_VALUE & mask;
+		val |= PLLMB_MISC1_DEFAULT_VALUE & mask;
 		writel_relaxed(val, clk_base + pllmb->params->ext_misc_reg[0]);
 		udelay(1);
 
@@ -930,7 +932,7 @@ void tegra210_pllmb_set_defaults(struct tegra_clk_pll *pllmb)
 	}
 
 	/* set IDDQ, enable lock detect */
-	writel_relaxed(PLLMB_MISC0_DEFAULT_VALUE,
+	writel_relaxed(PLLMB_MISC1_DEFAULT_VALUE,
 			clk_base + pllmb->params->ext_misc_reg[0]);
 	udelay(1);
 }
@@ -960,7 +962,7 @@ static void pllp_check_defaults(struct tegra_clk_pll *pll, bool enabled)
 			~mask & PLLP_MISC1_WRITE_MASK);
 }
 
-void tegra210_pllp_set_defaults(struct tegra_clk_pll *pllp)
+static void tegra210_pllp_set_defaults(struct tegra_clk_pll *pllp)
 {
 	u32 mask;
 	u32 val = readl_relaxed(clk_base + pllp->params->base_reg);
@@ -1022,7 +1024,7 @@ static void pllu_check_defaults(struct tegra_clk_pll *pll, bool hw_control)
 			~mask & PLLU_MISC1_WRITE_MASK);
 }
 
-void tegra210_pllu_set_defaults(struct tegra_clk_pll *pllu)
+static void tegra210_pllu_set_defaults(struct tegra_clk_pll *pllu)
 {
 	u32 val = readl_relaxed(clk_base + pllu->params->base_reg);
 
@@ -1212,8 +1214,9 @@ static void tegra210_clk_pll_set_gain(struct tegra_clk_pll_freq_table *cfg)
 	cfg->m *= PLL_SDM_COEFF;
 }
 
-unsigned long tegra210_clk_adjust_vco_min(struct tegra_clk_pll_params *params,
-					  unsigned long parent_rate)
+static unsigned long
+tegra210_clk_adjust_vco_min(struct tegra_clk_pll_params *params,
+			    unsigned long parent_rate)
 {
 	unsigned long vco_min = params->vco_min;
 
@@ -1386,7 +1389,7 @@ static struct tegra_clk_pll_params pll_c_params = {
 	.mdiv_default = 3,
 	.div_nmp = &pllc_nmp,
 	.freq_table = pll_cx_freq_table,
-	.flags = TEGRA_PLL_USE_LOCK | TEGRA_PLL_HAS_LOCK_ENABLE,
+	.flags = TEGRA_PLL_USE_LOCK,
 	.set_defaults = _pllc_set_defaults,
 	.calc_rate = tegra210_pll_fixed_mdiv_cfg,
 };
@@ -1425,7 +1428,7 @@ static struct tegra_clk_pll_params pll_c2_params = {
 	.ext_misc_reg[2] = PLLC2_MISC2,
 	.ext_misc_reg[3] = PLLC2_MISC3,
 	.freq_table = pll_cx_freq_table,
-	.flags = TEGRA_PLL_USE_LOCK | TEGRA_PLL_HAS_LOCK_ENABLE,
+	.flags = TEGRA_PLL_USE_LOCK,
 	.set_defaults = _pllc2_set_defaults,
 	.calc_rate = tegra210_pll_fixed_mdiv_cfg,
 };
@@ -1455,7 +1458,7 @@ static struct tegra_clk_pll_params pll_c3_params = {
 	.ext_misc_reg[2] = PLLC3_MISC2,
 	.ext_misc_reg[3] = PLLC3_MISC3,
 	.freq_table = pll_cx_freq_table,
-	.flags = TEGRA_PLL_USE_LOCK | TEGRA_PLL_HAS_LOCK_ENABLE,
+	.flags = TEGRA_PLL_USE_LOCK,
 	.set_defaults = _pllc3_set_defaults,
 	.calc_rate = tegra210_pll_fixed_mdiv_cfg,
 };
@@ -1505,7 +1508,6 @@ static struct tegra_clk_pll_params pll_c4_vco_params = {
 	.base_reg = PLLC4_BASE,
 	.misc_reg = PLLC4_MISC0,
 	.lock_mask = PLL_BASE_LOCK,
-	.lock_enable_bit_idx = PLLSS_MISC_LOCK_ENABLE,
 	.lock_delay = 300,
 	.max_p = PLL_QLIN_PDIV_MAX,
 	.ext_misc_reg[0] = PLLC4_MISC0,
@@ -1517,8 +1519,7 @@ static struct tegra_clk_pll_params pll_c4_vco_params = {
 	.div_nmp = &pllss_nmp,
 	.freq_table = pll_c4_vco_freq_table,
 	.set_defaults = tegra210_pllc4_set_defaults,
-	.flags = TEGRA_PLL_USE_LOCK | TEGRA_PLL_HAS_LOCK_ENABLE |
-		 TEGRA_PLL_VCO_OUT,
+	.flags = TEGRA_PLL_USE_LOCK | TEGRA_PLL_VCO_OUT,
 	.calc_rate = tegra210_pll_fixed_mdiv_cfg,
 };
 
@@ -1559,15 +1560,15 @@ static struct tegra_clk_pll_params pll_m_params = {
 	.vco_min = 800000000,
 	.vco_max = 1866000000,
 	.base_reg = PLLM_BASE,
-	.misc_reg = PLLM_MISC1,
+	.misc_reg = PLLM_MISC2,
 	.lock_mask = PLL_BASE_LOCK,
 	.lock_enable_bit_idx = PLLM_MISC_LOCK_ENABLE,
 	.lock_delay = 300,
-	.iddq_reg = PLLM_MISC0,
+	.iddq_reg = PLLM_MISC2,
 	.iddq_bit_idx = PLLM_IDDQ_BIT,
 	.max_p = PLL_QLIN_PDIV_MAX,
-	.ext_misc_reg[0] = PLLM_MISC0,
-	.ext_misc_reg[0] = PLLM_MISC1,
+	.ext_misc_reg[0] = PLLM_MISC2,
+	.ext_misc_reg[1] = PLLM_MISC1,
 	.round_p_to_pdiv = pll_qlin_p_to_pdiv,
 	.pdiv_tohw = pll_qlin_pdiv_to_hw,
 	.div_nmp = &pllm_nmp,
@@ -1586,19 +1587,18 @@ static struct tegra_clk_pll_params pll_mb_params = {
 	.vco_min = 800000000,
 	.vco_max = 1866000000,
 	.base_reg = PLLMB_BASE,
-	.misc_reg = PLLMB_MISC0,
+	.misc_reg = PLLMB_MISC1,
 	.lock_mask = PLL_BASE_LOCK,
-	.lock_enable_bit_idx = PLLMB_MISC_LOCK_ENABLE,
 	.lock_delay = 300,
-	.iddq_reg = PLLMB_MISC0,
+	.iddq_reg = PLLMB_MISC1,
 	.iddq_bit_idx = PLLMB_IDDQ_BIT,
 	.max_p = PLL_QLIN_PDIV_MAX,
-	.ext_misc_reg[0] = PLLMB_MISC0,
+	.ext_misc_reg[0] = PLLMB_MISC1,
 	.round_p_to_pdiv = pll_qlin_p_to_pdiv,
 	.pdiv_tohw = pll_qlin_pdiv_to_hw,
 	.div_nmp = &pllm_nmp,
 	.freq_table = pll_m_freq_table,
-	.flags = TEGRA_PLL_USE_LOCK | TEGRA_PLL_HAS_LOCK_ENABLE,
+	.flags = TEGRA_PLL_USE_LOCK,
 	.set_defaults = tegra210_pllmb_set_defaults,
 	.calc_rate = tegra210_pll_fixed_mdiv_cfg,
 };
@@ -1671,7 +1671,6 @@ static struct tegra_clk_pll_params pll_re_vco_params = {
 	.base_reg = PLLRE_BASE,
 	.misc_reg = PLLRE_MISC0,
 	.lock_mask = PLLRE_MISC_LOCK,
-	.lock_enable_bit_idx = PLLRE_MISC_LOCK_ENABLE,
 	.lock_delay = 300,
 	.max_p = PLL_QLIN_PDIV_MAX,
 	.ext_misc_reg[0] = PLLRE_MISC0,
@@ -1681,8 +1680,7 @@ static struct tegra_clk_pll_params pll_re_vco_params = {
 	.pdiv_tohw = pll_qlin_pdiv_to_hw,
 	.div_nmp = &pllre_nmp,
 	.freq_table = pll_re_vco_freq_table,
-	.flags = TEGRA_PLL_USE_LOCK | TEGRA_PLL_LOCK_MISC |
-		 TEGRA_PLL_HAS_LOCK_ENABLE | TEGRA_PLL_VCO_OUT,
+	.flags = TEGRA_PLL_USE_LOCK | TEGRA_PLL_LOCK_MISC | TEGRA_PLL_VCO_OUT,
 	.set_defaults = tegra210_pllre_set_defaults,
 	.calc_rate = tegra210_pll_fixed_mdiv_cfg,
 };
@@ -1712,7 +1710,6 @@ static struct tegra_clk_pll_params pll_p_params = {
 	.base_reg = PLLP_BASE,
 	.misc_reg = PLLP_MISC0,
 	.lock_mask = PLL_BASE_LOCK,
-	.lock_enable_bit_idx = PLLP_MISC_LOCK_ENABLE,
 	.lock_delay = 300,
 	.iddq_reg = PLLP_MISC0,
 	.iddq_bit_idx = PLLXP_IDDQ_BIT,
@@ -1721,8 +1718,7 @@ static struct tegra_clk_pll_params pll_p_params = {
 	.div_nmp = &pllp_nmp,
 	.freq_table = pll_p_freq_table,
 	.fixed_rate = 408000000,
-	.flags = TEGRA_PLL_FIXED | TEGRA_PLL_USE_LOCK |
-		 TEGRA_PLL_HAS_LOCK_ENABLE | TEGRA_PLL_VCO_OUT,
+	.flags = TEGRA_PLL_FIXED | TEGRA_PLL_USE_LOCK | TEGRA_PLL_VCO_OUT,
 	.set_defaults = tegra210_pllp_set_defaults,
 	.calc_rate = tegra210_pll_fixed_mdiv_cfg,
 };
@@ -1750,7 +1746,7 @@ static struct tegra_clk_pll_params pll_a1_params = {
 	.ext_misc_reg[2] = PLLA1_MISC2,
 	.ext_misc_reg[3] = PLLA1_MISC3,
 	.freq_table = pll_cx_freq_table,
-	.flags = TEGRA_PLL_USE_LOCK | TEGRA_PLL_HAS_LOCK_ENABLE,
+	.flags = TEGRA_PLL_USE_LOCK,
 	.set_defaults = _plla1_set_defaults,
 	.calc_rate = tegra210_pll_fixed_mdiv_cfg,
 };
@@ -1787,7 +1783,6 @@ static struct tegra_clk_pll_params pll_a_params = {
 	.base_reg = PLLA_BASE,
 	.misc_reg = PLLA_MISC0,
 	.lock_mask = PLL_BASE_LOCK,
-	.lock_enable_bit_idx = PLLA_MISC_LOCK_ENABLE,
 	.lock_delay = 300,
 	.round_p_to_pdiv = pll_qlin_p_to_pdiv,
 	.pdiv_tohw = pll_qlin_pdiv_to_hw,
@@ -1802,8 +1797,7 @@ static struct tegra_clk_pll_params pll_a_params = {
 	.ext_misc_reg[1] = PLLA_MISC1,
 	.ext_misc_reg[2] = PLLA_MISC2,
 	.freq_table = pll_a_freq_table,
-	.flags = TEGRA_PLL_USE_LOCK | TEGRA_MDIV_NEW |
-		 TEGRA_PLL_HAS_LOCK_ENABLE,
+	.flags = TEGRA_PLL_USE_LOCK | TEGRA_MDIV_NEW,
 	.set_defaults = tegra210_plla_set_defaults,
 	.calc_rate = tegra210_pll_fixed_mdiv_cfg,
 	.set_gain = tegra210_clk_pll_set_gain,
@@ -1836,7 +1830,6 @@ static struct tegra_clk_pll_params pll_d_params = {
 	.base_reg = PLLD_BASE,
 	.misc_reg = PLLD_MISC0,
 	.lock_mask = PLL_BASE_LOCK,
-	.lock_enable_bit_idx = PLLD_MISC_LOCK_ENABLE,
 	.lock_delay = 1000,
 	.iddq_reg = PLLD_MISC0,
 	.iddq_bit_idx = PLLD_IDDQ_BIT,
@@ -1850,7 +1843,7 @@ static struct tegra_clk_pll_params pll_d_params = {
 	.ext_misc_reg[0] = PLLD_MISC0,
 	.ext_misc_reg[1] = PLLD_MISC1,
 	.freq_table = pll_d_freq_table,
-	.flags = TEGRA_PLL_USE_LOCK | TEGRA_PLL_HAS_LOCK_ENABLE,
+	.flags = TEGRA_PLL_USE_LOCK,
 	.mdiv_default = 1,
 	.set_defaults = tegra210_plld_set_defaults,
 	.calc_rate = tegra210_pll_fixed_mdiv_cfg,
@@ -1876,7 +1869,6 @@ static struct tegra_clk_pll_params pll_d2_params = {
 	.base_reg = PLLD2_BASE,
 	.misc_reg = PLLD2_MISC0,
 	.lock_mask = PLL_BASE_LOCK,
-	.lock_enable_bit_idx = PLLSS_MISC_LOCK_ENABLE,
 	.lock_delay = 300,
 	.iddq_reg = PLLD2_BASE,
 	.iddq_bit_idx = PLLSS_IDDQ_BIT,
@@ -1897,7 +1889,7 @@ static struct tegra_clk_pll_params pll_d2_params = {
 	.mdiv_default = 1,
 	.freq_table = tegra210_pll_d2_freq_table,
 	.set_defaults = tegra210_plld2_set_defaults,
-	.flags = TEGRA_PLL_USE_LOCK | TEGRA_PLL_HAS_LOCK_ENABLE,
+	.flags = TEGRA_PLL_USE_LOCK,
 	.calc_rate = tegra210_pll_fixed_mdiv_cfg,
 	.set_gain = tegra210_clk_pll_set_gain,
 	.adjust_vco = tegra210_clk_adjust_vco_min,
@@ -1920,7 +1912,6 @@ static struct tegra_clk_pll_params pll_dp_params = {
 	.base_reg = PLLDP_BASE,
 	.misc_reg = PLLDP_MISC,
 	.lock_mask = PLL_BASE_LOCK,
-	.lock_enable_bit_idx = PLLSS_MISC_LOCK_ENABLE,
 	.lock_delay = 300,
 	.iddq_reg = PLLDP_BASE,
 	.iddq_bit_idx = PLLSS_IDDQ_BIT,
@@ -1941,7 +1932,7 @@ static struct tegra_clk_pll_params pll_dp_params = {
 	.mdiv_default = 1,
 	.freq_table = pll_dp_freq_table,
 	.set_defaults = tegra210_plldp_set_defaults,
-	.flags = TEGRA_PLL_USE_LOCK | TEGRA_PLL_HAS_LOCK_ENABLE,
+	.flags = TEGRA_PLL_USE_LOCK,
 	.calc_rate = tegra210_pll_fixed_mdiv_cfg,
 	.set_gain = tegra210_clk_pll_set_gain,
 	.adjust_vco = tegra210_clk_adjust_vco_min,
@@ -1973,7 +1964,6 @@ static struct tegra_clk_pll_params pll_u_vco_params = {
 	.base_reg = PLLU_BASE,
 	.misc_reg = PLLU_MISC0,
 	.lock_mask = PLL_BASE_LOCK,
-	.lock_enable_bit_idx = PLLU_MISC_LOCK_ENABLE,
 	.lock_delay = 1000,
 	.iddq_reg = PLLU_MISC0,
 	.iddq_bit_idx = PLLU_IDDQ_BIT,
@@ -1983,8 +1973,7 @@ static struct tegra_clk_pll_params pll_u_vco_params = {
 	.pdiv_tohw = pll_qlin_pdiv_to_hw,
 	.div_nmp = &pllu_nmp,
 	.freq_table = pll_u_freq_table,
-	.flags = TEGRA_PLLU | TEGRA_PLL_USE_LOCK | TEGRA_PLL_HAS_LOCK_ENABLE |
-		 TEGRA_PLL_VCO_OUT,
+	.flags = TEGRA_PLLU | TEGRA_PLL_USE_LOCK | TEGRA_PLL_VCO_OUT,
 	.set_defaults = tegra210_pllu_set_defaults,
 	.calc_rate = tegra210_pll_fixed_mdiv_cfg,
 };
@@ -2218,6 +2207,7 @@ static struct tegra_clk tegra210_clks[tegra_clk_max] __initdata = {
 	[tegra_clk_pll_c4_out1] = { .dt_id = TEGRA210_CLK_PLL_C4_OUT1, .present = true },
 	[tegra_clk_pll_c4_out2] = { .dt_id = TEGRA210_CLK_PLL_C4_OUT2, .present = true },
 	[tegra_clk_pll_c4_out3] = { .dt_id = TEGRA210_CLK_PLL_C4_OUT3, .present = true },
+	[tegra_clk_apb2ape] = { .dt_id = TEGRA210_CLK_APB2APE, .present = true },
 };
 
 static struct tegra_devclk devclks[] __initdata = {
@@ -2519,7 +2509,7 @@ static void __init tegra210_pll_init(void __iomem *clk_base,
 
 	/* PLLU_VCO */
 	val = readl(clk_base + pll_u_vco_params.base_reg);
-	val &= ~BIT(24); /* disable PLLU_OVERRIDE */
+	val &= ~PLLU_BASE_OVERRIDE; /* disable PLLU_OVERRIDE */
 	writel(val, clk_base + pll_u_vco_params.base_reg);
 
 	clk = tegra_clk_register_pllre("pll_u_vco", "pll_ref", clk_base, pmc,
@@ -2738,8 +2728,6 @@ static struct tegra_clk_init_table init_table[] __initdata = {
 	{ TEGRA210_CLK_DFLL_REF, TEGRA210_CLK_PLL_P, 51000000, 1 },
 	{ TEGRA210_CLK_SBC4, TEGRA210_CLK_PLL_P, 12000000, 1 },
 	{ TEGRA210_CLK_PLL_RE_VCO, TEGRA210_CLK_CLK_MAX, 672000000, 1 },
-	{ TEGRA210_CLK_PLL_U_OUT1, TEGRA210_CLK_CLK_MAX, 48000000, 1 },
-	{ TEGRA210_CLK_PLL_U_OUT2, TEGRA210_CLK_CLK_MAX, 60000000, 1 },
 	{ TEGRA210_CLK_XUSB_GATE, TEGRA210_CLK_CLK_MAX, 0, 1 },
 	{ TEGRA210_CLK_XUSB_SS_SRC, TEGRA210_CLK_PLL_U_480M, 120000000, 0 },
 	{ TEGRA210_CLK_XUSB_FS_SRC, TEGRA210_CLK_PLL_U_48M, 48000000, 0 },
diff --git a/drivers/clk/ti/dpll3xxx.c b/drivers/clk/ti/dpll3xxx.c
index 1c300388782b..cc739291a3ce 100644
--- a/drivers/clk/ti/dpll3xxx.c
+++ b/drivers/clk/ti/dpll3xxx.c
@@ -460,7 +460,8 @@ int omap3_noncore_dpll_enable(struct clk_hw *hw)
 
 	parent = clk_hw_get_parent(hw);
 
-	if (clk_hw_get_rate(hw) == clk_get_rate(dd->clk_bypass)) {
+	if (clk_hw_get_rate(hw) ==
+	    clk_hw_get_rate(__clk_get_hw(dd->clk_bypass))) {
 		WARN_ON(parent != __clk_get_hw(dd->clk_bypass));
 		r = _omap3_noncore_dpll_bypass(clk);
 	} else {
diff --git a/drivers/clk/versatile/clk-icst.c b/drivers/clk/versatile/clk-icst.c
index e62f8cb2c9b5..3bca438ecd19 100644
--- a/drivers/clk/versatile/clk-icst.c
+++ b/drivers/clk/versatile/clk-icst.c
@@ -78,6 +78,9 @@ static int vco_set(struct clk_icst *icst, struct icst_vco vco)
 	ret = regmap_read(icst->map, icst->vcoreg_off, &val);
 	if (ret)
 		return ret;
+
+	/* Mask the 18 bits used by the VCO */
+	val &= ~0x7ffff;
 	val |= vco.v | (vco.r << 9) | (vco.s << 16);
 
 	/* This magic unlocks the VCO so it can be controlled */
diff --git a/drivers/cpufreq/Kconfig b/drivers/cpufreq/Kconfig
index 659879a56dba..f93511031177 100644
--- a/drivers/cpufreq/Kconfig
+++ b/drivers/cpufreq/Kconfig
@@ -296,6 +296,7 @@ endif
 config QORIQ_CPUFREQ
 	tristate "CPU frequency scaling driver for Freescale QorIQ SoCs"
 	depends on OF && COMMON_CLK && (PPC_E500MC || ARM)
+	depends on !CPU_THERMAL || THERMAL
 	select CLK_QORIQ
 	help
 	  This adds the CPUFreq driver support for Freescale QorIQ SoCs
diff --git a/drivers/cpufreq/Kconfig.arm b/drivers/cpufreq/Kconfig.arm
index 0031069b64c9..14b1f9393b05 100644
--- a/drivers/cpufreq/Kconfig.arm
+++ b/drivers/cpufreq/Kconfig.arm
@@ -84,10 +84,10 @@ config ARM_KIRKWOOD_CPUFREQ
 	  SoCs.
 
 config ARM_MT8173_CPUFREQ
-	bool "Mediatek MT8173 CPUFreq support"
+	tristate "Mediatek MT8173 CPUFreq support"
 	depends on ARCH_MEDIATEK && REGULATOR
 	depends on ARM64 || (ARM_CPU_TOPOLOGY && COMPILE_TEST)
-	depends on !CPU_THERMAL || THERMAL=y
+	depends on !CPU_THERMAL || THERMAL
 	select PM_OPP
 	help
 	  This adds the CPUFreq driver support for Mediatek MT8173 SoC.
diff --git a/drivers/cpufreq/mt8173-cpufreq.c b/drivers/cpufreq/mt8173-cpufreq.c
index 1efba340456d..2058e6d292ce 100644
--- a/drivers/cpufreq/mt8173-cpufreq.c
+++ b/drivers/cpufreq/mt8173-cpufreq.c
@@ -17,6 +17,7 @@
 #include <linux/cpu_cooling.h>
 #include <linux/cpufreq.h>
 #include <linux/cpumask.h>
+#include <linux/module.h>
 #include <linux/of.h>
 #include <linux/platform_device.h>
 #include <linux/pm_opp.h>
diff --git a/drivers/crypto/atmel-sha.c b/drivers/crypto/atmel-sha.c
index 20de861aa0ea..8bf9914d4d15 100644
--- a/drivers/crypto/atmel-sha.c
+++ b/drivers/crypto/atmel-sha.c
@@ -782,7 +782,7 @@ static void atmel_sha_finish_req(struct ahash_request *req, int err)
 	dd->flags &= ~(SHA_FLAGS_BUSY | SHA_FLAGS_FINAL | SHA_FLAGS_CPU |
 			SHA_FLAGS_DMA_READY | SHA_FLAGS_OUTPUT_READY);
 
-	clk_disable_unprepare(dd->iclk);
+	clk_disable(dd->iclk);
 
 	if (req->base.complete)
 		req->base.complete(&req->base, err);
@@ -795,7 +795,7 @@ static int atmel_sha_hw_init(struct atmel_sha_dev *dd)
 {
 	int err;
 
-	err = clk_prepare_enable(dd->iclk);
+	err = clk_enable(dd->iclk);
 	if (err)
 		return err;
 
@@ -822,7 +822,7 @@ static void atmel_sha_hw_version_init(struct atmel_sha_dev *dd)
 	dev_info(dd->dev,
 			"version: 0x%x\n", dd->hw_version);
 
-	clk_disable_unprepare(dd->iclk);
+	clk_disable(dd->iclk);
 }
 
 static int atmel_sha_handle_queue(struct atmel_sha_dev *dd,
@@ -1410,6 +1410,10 @@ static int atmel_sha_probe(struct platform_device *pdev)
 		goto res_err;
 	}
 
+	err = clk_prepare(sha_dd->iclk);
+	if (err)
+		goto res_err;
+
 	atmel_sha_hw_version_init(sha_dd);
 
 	atmel_sha_get_cap(sha_dd);
@@ -1421,12 +1425,12 @@ static int atmel_sha_probe(struct platform_device *pdev)
 			if (IS_ERR(pdata)) {
 				dev_err(&pdev->dev, "platform data not available\n");
 				err = PTR_ERR(pdata);
-				goto res_err;
+				goto iclk_unprepare;
 			}
 		}
 		if (!pdata->dma_slave) {
 			err = -ENXIO;
-			goto res_err;
+			goto iclk_unprepare;
 		}
 		err = atmel_sha_dma_init(sha_dd, pdata);
 		if (err)
@@ -1457,6 +1461,8 @@ err_algs:
 	if (sha_dd->caps.has_dma)
 		atmel_sha_dma_cleanup(sha_dd);
 err_sha_dma:
+iclk_unprepare:
+	clk_unprepare(sha_dd->iclk);
 res_err:
 	tasklet_kill(&sha_dd->done_task);
 sha_dd_err:
@@ -1483,12 +1489,7 @@ static int atmel_sha_remove(struct platform_device *pdev)
 	if (sha_dd->caps.has_dma)
 		atmel_sha_dma_cleanup(sha_dd);
 
-	iounmap(sha_dd->io_base);
-
-	clk_put(sha_dd->iclk);
-
-	if (sha_dd->irq >= 0)
-		free_irq(sha_dd->irq, sha_dd);
+	clk_unprepare(sha_dd->iclk);
 
 	return 0;
 }
diff --git a/drivers/crypto/marvell/cesa.c b/drivers/crypto/marvell/cesa.c
index 0643e3366e33..c0656e7f37b5 100644
--- a/drivers/crypto/marvell/cesa.c
+++ b/drivers/crypto/marvell/cesa.c
@@ -306,7 +306,7 @@ static int mv_cesa_dev_dma_init(struct mv_cesa_dev *cesa)
 		return -ENOMEM;
 
 	dma->padding_pool = dmam_pool_create("cesa_padding", dev, 72, 1, 0);
-	if (!dma->cache_pool)
+	if (!dma->padding_pool)
 		return -ENOMEM;
 
 	cesa->dma = dma;
diff --git a/drivers/devfreq/tegra-devfreq.c b/drivers/devfreq/tegra-devfreq.c
index 848b93ee930f..fe9dce0245bf 100644
--- a/drivers/devfreq/tegra-devfreq.c
+++ b/drivers/devfreq/tegra-devfreq.c
@@ -500,6 +500,8 @@ static int tegra_devfreq_target(struct device *dev, unsigned long *freq,
 	clk_set_min_rate(tegra->emc_clock, rate);
 	clk_set_rate(tegra->emc_clock, 0);
 
+	*freq = rate;
+
 	return 0;
 }
 
diff --git a/drivers/dma/at_xdmac.c b/drivers/dma/at_xdmac.c
index 64f5d1bdbb48..8e304b1befc5 100644
--- a/drivers/dma/at_xdmac.c
+++ b/drivers/dma/at_xdmac.c
@@ -176,6 +176,7 @@
 #define AT_XDMAC_MAX_CHAN	0x20
 #define AT_XDMAC_MAX_CSIZE	16	/* 16 data */
 #define AT_XDMAC_MAX_DWIDTH	8	/* 64 bits */
+#define AT_XDMAC_RESIDUE_MAX_RETRIES	5
 
 #define AT_XDMAC_DMA_BUSWIDTHS\
 	(BIT(DMA_SLAVE_BUSWIDTH_UNDEFINED) |\
@@ -1395,8 +1396,8 @@ at_xdmac_tx_status(struct dma_chan *chan, dma_cookie_t cookie,
 	struct at_xdmac_desc	*desc, *_desc;
 	struct list_head	*descs_list;
 	enum dma_status		ret;
-	int			residue;
-	u32			cur_nda, mask, value;
+	int			residue, retry;
+	u32			cur_nda, check_nda, cur_ubc, mask, value;
 	u8			dwidth = 0;
 	unsigned long		flags;
 
@@ -1433,7 +1434,42 @@ at_xdmac_tx_status(struct dma_chan *chan, dma_cookie_t cookie,
 			cpu_relax();
 	}
 
+	/*
+	 * When processing the residue, we need to read two registers but we
+	 * can't do it in an atomic way. AT_XDMAC_CNDA is used to find where
+	 * we stand in the descriptor list and AT_XDMAC_CUBC is used
+	 * to know how many data are remaining for the current descriptor.
+	 * Since the dma channel is not paused to not loose data, between the
+	 * AT_XDMAC_CNDA and AT_XDMAC_CUBC read, we may have change of
+	 * descriptor.
+	 * For that reason, after reading AT_XDMAC_CUBC, we check if we are
+	 * still using the same descriptor by reading a second time
+	 * AT_XDMAC_CNDA. If AT_XDMAC_CNDA has changed, it means we have to
+	 * read again AT_XDMAC_CUBC.
+	 * Memory barriers are used to ensure the read order of the registers.
+	 * A max number of retries is set because unlikely it can never ends if
+	 * we are transferring a lot of data with small buffers.
+	 */
 	cur_nda = at_xdmac_chan_read(atchan, AT_XDMAC_CNDA) & 0xfffffffc;
+	rmb();
+	cur_ubc = at_xdmac_chan_read(atchan, AT_XDMAC_CUBC);
+	for (retry = 0; retry < AT_XDMAC_RESIDUE_MAX_RETRIES; retry++) {
+		rmb();
+		check_nda = at_xdmac_chan_read(atchan, AT_XDMAC_CNDA) & 0xfffffffc;
+
+		if (likely(cur_nda == check_nda))
+			break;
+
+		cur_nda = check_nda;
+		rmb();
+		cur_ubc = at_xdmac_chan_read(atchan, AT_XDMAC_CUBC);
+	}
+
+	if (unlikely(retry >= AT_XDMAC_RESIDUE_MAX_RETRIES)) {
+		ret = DMA_ERROR;
+		goto spin_unlock;
+	}
+
 	/*
 	 * Remove size of all microblocks already transferred and the current
 	 * one. Then add the remaining size to transfer of the current
@@ -1446,7 +1482,7 @@ at_xdmac_tx_status(struct dma_chan *chan, dma_cookie_t cookie,
 		if ((desc->lld.mbr_nda & 0xfffffffc) == cur_nda)
 			break;
 	}
-	residue += at_xdmac_chan_read(atchan, AT_XDMAC_CUBC) << dwidth;
+	residue += cur_ubc << dwidth;
 
 	dma_set_residue(txstate, residue);
 
diff --git a/drivers/dma/dw/core.c b/drivers/dma/dw/core.c
index e893318560db..5ad0ec1f0e29 100644
--- a/drivers/dma/dw/core.c
+++ b/drivers/dma/dw/core.c
@@ -156,7 +156,6 @@ static void dwc_initialize(struct dw_dma_chan *dwc)
 
 	/* Enable interrupts */
 	channel_set_bit(dw, MASK.XFER, dwc->mask);
-	channel_set_bit(dw, MASK.BLOCK, dwc->mask);
 	channel_set_bit(dw, MASK.ERROR, dwc->mask);
 
 	dwc->initialized = true;
@@ -588,6 +587,9 @@ static void dwc_handle_cyclic(struct dw_dma *dw, struct dw_dma_chan *dwc,
 
 		spin_unlock_irqrestore(&dwc->lock, flags);
 	}
+
+	/* Re-enable interrupts */
+	channel_set_bit(dw, MASK.BLOCK, dwc->mask);
 }
 
 /* ------------------------------------------------------------------------- */
@@ -618,11 +620,8 @@ static void dw_dma_tasklet(unsigned long data)
 			dwc_scan_descriptors(dw, dwc);
 	}
 
-	/*
-	 * Re-enable interrupts.
-	 */
+	/* Re-enable interrupts */
 	channel_set_bit(dw, MASK.XFER, dw->all_chan_mask);
-	channel_set_bit(dw, MASK.BLOCK, dw->all_chan_mask);
 	channel_set_bit(dw, MASK.ERROR, dw->all_chan_mask);
 }
 
@@ -1261,6 +1260,7 @@ static void dwc_free_chan_resources(struct dma_chan *chan)
 int dw_dma_cyclic_start(struct dma_chan *chan)
 {
 	struct dw_dma_chan	*dwc = to_dw_dma_chan(chan);
+	struct dw_dma		*dw = to_dw_dma(chan->device);
 	unsigned long		flags;
 
 	if (!test_bit(DW_DMA_IS_CYCLIC, &dwc->flags)) {
@@ -1269,7 +1269,12 @@ int dw_dma_cyclic_start(struct dma_chan *chan)
 	}
 
 	spin_lock_irqsave(&dwc->lock, flags);
+
+	/* Enable interrupts to perform cyclic transfer */
+	channel_set_bit(dw, MASK.BLOCK, dwc->mask);
+
 	dwc_dostart(dwc, dwc->cdesc->desc[0]);
+
 	spin_unlock_irqrestore(&dwc->lock, flags);
 
 	return 0;
diff --git a/drivers/dma/dw/pci.c b/drivers/dma/dw/pci.c
index 4c30fdd092b3..358f9689a3f5 100644
--- a/drivers/dma/dw/pci.c
+++ b/drivers/dma/dw/pci.c
@@ -108,6 +108,10 @@ static const struct pci_device_id dw_pci_id_table[] = {
 
 	/* Haswell */
 	{ PCI_VDEVICE(INTEL, 0x9c60) },
+
+	/* Broadwell */
+	{ PCI_VDEVICE(INTEL, 0x9ce0) },
+
 	{ }
 };
 MODULE_DEVICE_TABLE(pci, dw_pci_id_table);
diff --git a/drivers/dma/edma.c b/drivers/dma/edma.c
index d92d65549406..e3d7fcb69b4c 100644
--- a/drivers/dma/edma.c
+++ b/drivers/dma/edma.c
@@ -113,6 +113,9 @@
 #define GET_NUM_REGN(x)		((x & 0x300000) >> 20) /* bits 20-21 */
 #define CHMAP_EXIST		BIT(24)
 
+/* CCSTAT register */
+#define EDMA_CCSTAT_ACTV	BIT(4)
+
 /*
  * Max of 20 segments per channel to conserve PaRAM slots
  * Also note that MAX_NR_SG should be atleast the no.of periods
@@ -1680,9 +1683,20 @@ static void edma_issue_pending(struct dma_chan *chan)
 	spin_unlock_irqrestore(&echan->vchan.lock, flags);
 }
 
+/*
+ * This limit exists to avoid a possible infinite loop when waiting for proof
+ * that a particular transfer is completed. This limit can be hit if there
+ * are large bursts to/from slow devices or the CPU is never able to catch
+ * the DMA hardware idle. On an AM335x transfering 48 bytes from the UART
+ * RX-FIFO, as many as 55 loops have been seen.
+ */
+#define EDMA_MAX_TR_WAIT_LOOPS 1000
+
 static u32 edma_residue(struct edma_desc *edesc)
 {
 	bool dst = edesc->direction == DMA_DEV_TO_MEM;
+	int loop_count = EDMA_MAX_TR_WAIT_LOOPS;
+	struct edma_chan *echan = edesc->echan;
 	struct edma_pset *pset = edesc->pset;
 	dma_addr_t done, pos;
 	int i;
@@ -1691,7 +1705,32 @@ static u32 edma_residue(struct edma_desc *edesc)
 	 * We always read the dst/src position from the first RamPar
 	 * pset. That's the one which is active now.
 	 */
-	pos = edma_get_position(edesc->echan->ecc, edesc->echan->slot[0], dst);
+	pos = edma_get_position(echan->ecc, echan->slot[0], dst);
+
+	/*
+	 * "pos" may represent a transfer request that is still being
+	 * processed by the EDMACC or EDMATC. We will busy wait until
+	 * any one of the situations occurs:
+	 *   1. the DMA hardware is idle
+	 *   2. a new transfer request is setup
+	 *   3. we hit the loop limit
+	 */
+	while (edma_read(echan->ecc, EDMA_CCSTAT) & EDMA_CCSTAT_ACTV) {
+		/* check if a new transfer request is setup */
+		if (edma_get_position(echan->ecc,
+				      echan->slot[0], dst) != pos) {
+			break;
+		}
+
+		if (!--loop_count) {
+			dev_dbg_ratelimited(echan->vchan.chan.device->dev,
+				"%s: timeout waiting for PaRAM update\n",
+				__func__);
+			break;
+		}
+
+		cpu_relax();
+	}
 
 	/*
 	 * Cyclic is simple. Just subtract pset[0].addr from pos.
diff --git a/drivers/dma/fsldma.c b/drivers/dma/fsldma.c
index 2209f75fdf05..aac85c30c2cf 100644
--- a/drivers/dma/fsldma.c
+++ b/drivers/dma/fsldma.c
@@ -522,6 +522,8 @@ static dma_cookie_t fsldma_run_tx_complete_actions(struct fsldma_chan *chan,
 			chan_dbg(chan, "LD %p callback\n", desc);
 			txd->callback(txd->callback_param);
 		}
+
+		dma_descriptor_unmap(txd);
 	}
 
 	/* Run any dependencies */
diff --git a/drivers/dma/ioat/dma.c b/drivers/dma/ioat/dma.c
index 1d5df2ef148b..21539d5c54c3 100644
--- a/drivers/dma/ioat/dma.c
+++ b/drivers/dma/ioat/dma.c
@@ -861,32 +861,42 @@ void ioat_timer_event(unsigned long data)
 			return;
 	}
 
+	spin_lock_bh(&ioat_chan->cleanup_lock);
+
+	/* handle the no-actives case */
+	if (!ioat_ring_active(ioat_chan)) {
+		spin_lock_bh(&ioat_chan->prep_lock);
+		check_active(ioat_chan);
+		spin_unlock_bh(&ioat_chan->prep_lock);
+		spin_unlock_bh(&ioat_chan->cleanup_lock);
+		return;
+	}
+
 	/* if we haven't made progress and we have already
 	 * acknowledged a pending completion once, then be more
 	 * forceful with a restart
 	 */
-	spin_lock_bh(&ioat_chan->cleanup_lock);
 	if (ioat_cleanup_preamble(ioat_chan, &phys_complete))
 		__cleanup(ioat_chan, phys_complete);
 	else if (test_bit(IOAT_COMPLETION_ACK, &ioat_chan->state)) {
+		u32 chanerr;
+
+		chanerr = readl(ioat_chan->reg_base + IOAT_CHANERR_OFFSET);
+		dev_warn(to_dev(ioat_chan), "Restarting channel...\n");
+		dev_warn(to_dev(ioat_chan), "CHANSTS: %#Lx CHANERR: %#x\n",
+			 status, chanerr);
+		dev_warn(to_dev(ioat_chan), "Active descriptors: %d\n",
+			 ioat_ring_active(ioat_chan));
+
 		spin_lock_bh(&ioat_chan->prep_lock);
 		ioat_restart_channel(ioat_chan);
 		spin_unlock_bh(&ioat_chan->prep_lock);
 		spin_unlock_bh(&ioat_chan->cleanup_lock);
 		return;
-	} else {
+	} else
 		set_bit(IOAT_COMPLETION_ACK, &ioat_chan->state);
-		mod_timer(&ioat_chan->timer, jiffies + COMPLETION_TIMEOUT);
-	}
-
 
-	if (ioat_ring_active(ioat_chan))
-		mod_timer(&ioat_chan->timer, jiffies + COMPLETION_TIMEOUT);
-	else {
-		spin_lock_bh(&ioat_chan->prep_lock);
-		check_active(ioat_chan);
-		spin_unlock_bh(&ioat_chan->prep_lock);
-	}
+	mod_timer(&ioat_chan->timer, jiffies + COMPLETION_TIMEOUT);
 	spin_unlock_bh(&ioat_chan->cleanup_lock);
 }
 
diff --git a/drivers/dma/pxa_dma.c b/drivers/dma/pxa_dma.c
index f2a0310ae771..debca824bed6 100644
--- a/drivers/dma/pxa_dma.c
+++ b/drivers/dma/pxa_dma.c
@@ -583,6 +583,8 @@ static void set_updater_desc(struct pxad_desc_sw *sw_desc,
 		(PXA_DCMD_LENGTH & sizeof(u32));
 	if (flags & DMA_PREP_INTERRUPT)
 		updater->dcmd |= PXA_DCMD_ENDIRQEN;
+	if (sw_desc->cyclic)
+		sw_desc->hw_desc[sw_desc->nb_desc - 2]->ddadr = sw_desc->first;
 }
 
 static bool is_desc_completed(struct virt_dma_desc *vd)
@@ -673,6 +675,10 @@ static irqreturn_t pxad_chan_handler(int irq, void *dev_id)
 		dev_dbg(&chan->vc.chan.dev->device,
 			"%s(): checking txd %p[%x]: completed=%d\n",
 			__func__, vd, vd->tx.cookie, is_desc_completed(vd));
+		if (to_pxad_sw_desc(vd)->cyclic) {
+			vchan_cyclic_callback(vd);
+			break;
+		}
 		if (is_desc_completed(vd)) {
 			list_del(&vd->node);
 			vchan_cookie_complete(vd);
@@ -1080,7 +1086,7 @@ pxad_prep_dma_cyclic(struct dma_chan *dchan,
 		return NULL;
 
 	pxad_get_config(chan, dir, &dcmd, &dsadr, &dtadr);
-	dcmd |= PXA_DCMD_ENDIRQEN | (PXA_DCMD_LENGTH | period_len);
+	dcmd |= PXA_DCMD_ENDIRQEN | (PXA_DCMD_LENGTH & period_len);
 	dev_dbg(&chan->vc.chan.dev->device,
 		"%s(): buf_addr=0x%lx len=%zu period=%zu dir=%d flags=%lx\n",
 		__func__, (unsigned long)buf_addr, len, period_len, dir, flags);
diff --git a/drivers/edac/sb_edac.c b/drivers/edac/sb_edac.c
index e438ee5b433f..f5c6b97c8958 100644
--- a/drivers/edac/sb_edac.c
+++ b/drivers/edac/sb_edac.c
@@ -1574,7 +1574,7 @@ static int knl_get_dimm_capacity(struct sbridge_pvt *pvt, u64 *mc_sizes)
 				for (cha = 0; cha < KNL_MAX_CHAS; cha++) {
 					if (knl_get_mc_route(target,
 						mc_route_reg[cha]) == channel
-						&& participants[channel]) {
+						&& !participants[channel]) {
 						participant_count++;
 						participants[channel] = 1;
 						break;
diff --git a/drivers/firmware/efi/efivars.c b/drivers/firmware/efi/efivars.c
index 756eca8c4cf8..10e6774ab2a2 100644
--- a/drivers/firmware/efi/efivars.c
+++ b/drivers/firmware/efi/efivars.c
@@ -221,7 +221,7 @@ sanity_check(struct efi_variable *var, efi_char16_t *name, efi_guid_t vendor,
 	}
 
 	if ((attributes & ~EFI_VARIABLE_MASK) != 0 ||
-	    efivar_validate(name, data, size) == false) {
+	    efivar_validate(vendor, name, data, size) == false) {
 		printk(KERN_ERR "efivars: Malformed variable content\n");
 		return -EINVAL;
 	}
@@ -447,7 +447,8 @@ static ssize_t efivar_create(struct file *filp, struct kobject *kobj,
 	}
 
 	if ((attributes & ~EFI_VARIABLE_MASK) != 0 ||
-	    efivar_validate(name, data, size) == false) {
+	    efivar_validate(new_var->VendorGuid, name, data,
+			    size) == false) {
 		printk(KERN_ERR "efivars: Malformed variable content\n");
 		return -EINVAL;
 	}
@@ -540,38 +541,30 @@ static ssize_t efivar_delete(struct file *filp, struct kobject *kobj,
 static int
 efivar_create_sysfs_entry(struct efivar_entry *new_var)
 {
-	int i, short_name_size;
+	int short_name_size;
 	char *short_name;
-	unsigned long variable_name_size;
-	efi_char16_t *variable_name;
+	unsigned long utf8_name_size;
+	efi_char16_t *variable_name = new_var->var.VariableName;
 	int ret;
 
-	variable_name = new_var->var.VariableName;
-	variable_name_size = ucs2_strlen(variable_name) * sizeof(efi_char16_t);
-
 	/*
-	 * Length of the variable bytes in ASCII, plus the '-' separator,
+	 * Length of the variable bytes in UTF8, plus the '-' separator,
 	 * plus the GUID, plus trailing NUL
 	 */
-	short_name_size = variable_name_size / sizeof(efi_char16_t)
-				+ 1 + EFI_VARIABLE_GUID_LEN + 1;
-
-	short_name = kzalloc(short_name_size, GFP_KERNEL);
+	utf8_name_size = ucs2_utf8size(variable_name);
+	short_name_size = utf8_name_size + 1 + EFI_VARIABLE_GUID_LEN + 1;
 
+	short_name = kmalloc(short_name_size, GFP_KERNEL);
 	if (!short_name)
 		return -ENOMEM;
 
-	/* Convert Unicode to normal chars (assume top bits are 0),
-	   ala UTF-8 */
-	for (i=0; i < (int)(variable_name_size / sizeof(efi_char16_t)); i++) {
-		short_name[i] = variable_name[i] & 0xFF;
-	}
+	ucs2_as_utf8(short_name, variable_name, short_name_size);
+
 	/* This is ugly, but necessary to separate one vendor's
 	   private variables from another's.         */
-
-	*(short_name + strlen(short_name)) = '-';
+	short_name[utf8_name_size] = '-';
 	efi_guid_to_str(&new_var->var.VendorGuid,
-			 short_name + strlen(short_name));
+			 short_name + utf8_name_size + 1);
 
 	new_var->kobj.kset = efivars_kset;
 
diff --git a/drivers/firmware/efi/vars.c b/drivers/firmware/efi/vars.c
index 70a0fb10517f..7f2ea21c730d 100644
--- a/drivers/firmware/efi/vars.c
+++ b/drivers/firmware/efi/vars.c
@@ -165,67 +165,133 @@ validate_ascii_string(efi_char16_t *var_name, int match, u8 *buffer,
 }
 
 struct variable_validate {
+	efi_guid_t vendor;
 	char *name;
 	bool (*validate)(efi_char16_t *var_name, int match, u8 *data,
 			 unsigned long len);
 };
 
+/*
+ * This is the list of variables we need to validate, as well as the
+ * whitelist for what we think is safe not to default to immutable.
+ *
+ * If it has a validate() method that's not NULL, it'll go into the
+ * validation routine.  If not, it is assumed valid, but still used for
+ * whitelisting.
+ *
+ * Note that it's sorted by {vendor,name}, but globbed names must come after
+ * any other name with the same prefix.
+ */
 static const struct variable_validate variable_validate[] = {
-	{ "BootNext", validate_uint16 },
-	{ "BootOrder", validate_boot_order },
-	{ "DriverOrder", validate_boot_order },
-	{ "Boot*", validate_load_option },
-	{ "Driver*", validate_load_option },
-	{ "ConIn", validate_device_path },
-	{ "ConInDev", validate_device_path },
-	{ "ConOut", validate_device_path },
-	{ "ConOutDev", validate_device_path },
-	{ "ErrOut", validate_device_path },
-	{ "ErrOutDev", validate_device_path },
-	{ "Timeout", validate_uint16 },
-	{ "Lang", validate_ascii_string },
-	{ "PlatformLang", validate_ascii_string },
-	{ "", NULL },
+	{ EFI_GLOBAL_VARIABLE_GUID, "BootNext", validate_uint16 },
+	{ EFI_GLOBAL_VARIABLE_GUID, "BootOrder", validate_boot_order },
+	{ EFI_GLOBAL_VARIABLE_GUID, "Boot*", validate_load_option },
+	{ EFI_GLOBAL_VARIABLE_GUID, "DriverOrder", validate_boot_order },
+	{ EFI_GLOBAL_VARIABLE_GUID, "Driver*", validate_load_option },
+	{ EFI_GLOBAL_VARIABLE_GUID, "ConIn", validate_device_path },
+	{ EFI_GLOBAL_VARIABLE_GUID, "ConInDev", validate_device_path },
+	{ EFI_GLOBAL_VARIABLE_GUID, "ConOut", validate_device_path },
+	{ EFI_GLOBAL_VARIABLE_GUID, "ConOutDev", validate_device_path },
+	{ EFI_GLOBAL_VARIABLE_GUID, "ErrOut", validate_device_path },
+	{ EFI_GLOBAL_VARIABLE_GUID, "ErrOutDev", validate_device_path },
+	{ EFI_GLOBAL_VARIABLE_GUID, "Lang", validate_ascii_string },
+	{ EFI_GLOBAL_VARIABLE_GUID, "OsIndications", NULL },
+	{ EFI_GLOBAL_VARIABLE_GUID, "PlatformLang", validate_ascii_string },
+	{ EFI_GLOBAL_VARIABLE_GUID, "Timeout", validate_uint16 },
+	{ LINUX_EFI_CRASH_GUID, "*", NULL },
+	{ NULL_GUID, "", NULL },
 };
 
+static bool
+variable_matches(const char *var_name, size_t len, const char *match_name,
+		 int *match)
+{
+	for (*match = 0; ; (*match)++) {
+		char c = match_name[*match];
+		char u = var_name[*match];
+
+		/* Wildcard in the matching name means we've matched */
+		if (c == '*')
+			return true;
+
+		/* Case sensitive match */
+		if (!c && *match == len)
+			return true;
+
+		if (c != u)
+			return false;
+
+		if (!c)
+			return true;
+	}
+	return true;
+}
+
 bool
-efivar_validate(efi_char16_t *var_name, u8 *data, unsigned long len)
+efivar_validate(efi_guid_t vendor, efi_char16_t *var_name, u8 *data,
+		unsigned long data_size)
 {
 	int i;
-	u16 *unicode_name = var_name;
+	unsigned long utf8_size;
+	u8 *utf8_name;
 
-	for (i = 0; variable_validate[i].validate != NULL; i++) {
-		const char *name = variable_validate[i].name;
-		int match;
+	utf8_size = ucs2_utf8size(var_name);
+	utf8_name = kmalloc(utf8_size + 1, GFP_KERNEL);
+	if (!utf8_name)
+		return false;
 
-		for (match = 0; ; match++) {
-			char c = name[match];
-			u16 u = unicode_name[match];
+	ucs2_as_utf8(utf8_name, var_name, utf8_size);
+	utf8_name[utf8_size] = '\0';
 
-			/* All special variables are plain ascii */
-			if (u > 127)
-				return true;
+	for (i = 0; variable_validate[i].name[0] != '\0'; i++) {
+		const char *name = variable_validate[i].name;
+		int match = 0;
 
-			/* Wildcard in the matching name means we've matched */
-			if (c == '*')
-				return variable_validate[i].validate(var_name,
-							     match, data, len);
+		if (efi_guidcmp(vendor, variable_validate[i].vendor))
+			continue;
 
-			/* Case sensitive match */
-			if (c != u)
+		if (variable_matches(utf8_name, utf8_size+1, name, &match)) {
+			if (variable_validate[i].validate == NULL)
 				break;
-
-			/* Reached the end of the string while matching */
-			if (!c)
-				return variable_validate[i].validate(var_name,
-							     match, data, len);
+			kfree(utf8_name);
+			return variable_validate[i].validate(var_name, match,
+							     data, data_size);
 		}
 	}
-
+	kfree(utf8_name);
 	return true;
 }
 EXPORT_SYMBOL_GPL(efivar_validate);
 
+bool
+efivar_variable_is_removable(efi_guid_t vendor, const char *var_name,
+			     size_t len)
+{
+	int i;
+	bool found = false;
+	int match = 0;
+
+	/*
+	 * Check if our variable is in the validated variables list
+	 */
+	for (i = 0; variable_validate[i].name[0] != '\0'; i++) {
+		if (efi_guidcmp(variable_validate[i].vendor, vendor))
+			continue;
+
+		if (variable_matches(var_name, len,
+				     variable_validate[i].name, &match)) {
+			found = true;
+			break;
+		}
+	}
+
+	/*
+	 * If it's in our list, it is removable.
+	 */
+	return found;
+}
+EXPORT_SYMBOL_GPL(efivar_variable_is_removable);
+
 static efi_status_t
 check_var_size(u32 attributes, unsigned long size)
 {
@@ -852,7 +918,7 @@ int efivar_entry_set_get_size(struct efivar_entry *entry, u32 attributes,
 
 	*set = false;
 
-	if (efivar_validate(name, data, *size) == false)
+	if (efivar_validate(*vendor, name, data, *size) == false)
 		return -EINVAL;
 
 	/*
diff --git a/drivers/gpio/gpio-altera.c b/drivers/gpio/gpio-altera.c
index 2aeaebd1c6e7..3f87a03abc22 100644
--- a/drivers/gpio/gpio-altera.c
+++ b/drivers/gpio/gpio-altera.c
@@ -312,8 +312,8 @@ static int altera_gpio_probe(struct platform_device *pdev)
 		handle_simple_irq, IRQ_TYPE_NONE);
 
 	if (ret) {
-		dev_info(&pdev->dev, "could not add irqchip\n");
-		return ret;
+		dev_err(&pdev->dev, "could not add irqchip\n");
+		goto teardown;
 	}
 
 	gpiochip_set_chained_irqchip(&altera_gc->mmchip.gc,
@@ -326,6 +326,7 @@ static int altera_gpio_probe(struct platform_device *pdev)
 skip_irq:
 	return 0;
 teardown:
+	of_mm_gpiochip_remove(&altera_gc->mmchip);
 	pr_err("%s: registration failed with status %d\n",
 		node->full_name, ret);
 
diff --git a/drivers/gpio/gpio-davinci.c b/drivers/gpio/gpio-davinci.c
index ec58f4288649..cd007a67b302 100644
--- a/drivers/gpio/gpio-davinci.c
+++ b/drivers/gpio/gpio-davinci.c
@@ -195,7 +195,7 @@ static int davinci_gpio_of_xlate(struct gpio_chip *gc,
 static int davinci_gpio_probe(struct platform_device *pdev)
 {
 	int i, base;
-	unsigned ngpio;
+	unsigned ngpio, nbank;
 	struct davinci_gpio_controller *chips;
 	struct davinci_gpio_platform_data *pdata;
 	struct davinci_gpio_regs __iomem *regs;
@@ -224,8 +224,9 @@ static int davinci_gpio_probe(struct platform_device *pdev)
 	if (WARN_ON(ARCH_NR_GPIOS < ngpio))
 		ngpio = ARCH_NR_GPIOS;
 
+	nbank = DIV_ROUND_UP(ngpio, 32);
 	chips = devm_kzalloc(dev,
-			     ngpio * sizeof(struct davinci_gpio_controller),
+			     nbank * sizeof(struct davinci_gpio_controller),
 			     GFP_KERNEL);
 	if (!chips)
 		return -ENOMEM;
@@ -511,7 +512,7 @@ static int davinci_gpio_irq_setup(struct platform_device *pdev)
 			return irq;
 		}
 
-		irq_domain = irq_domain_add_legacy(NULL, ngpio, irq, 0,
+		irq_domain = irq_domain_add_legacy(dev->of_node, ngpio, irq, 0,
 							&davinci_gpio_irq_ops,
 							chips);
 		if (!irq_domain) {
diff --git a/drivers/gpio/gpio-rcar.c b/drivers/gpio/gpio-rcar.c
index cf41440aff91..d9ab0cd1d205 100644
--- a/drivers/gpio/gpio-rcar.c
+++ b/drivers/gpio/gpio-rcar.c
@@ -196,6 +196,44 @@ static int gpio_rcar_irq_set_wake(struct irq_data *d, unsigned int on)
 	return 0;
 }
 
+static void gpio_rcar_irq_bus_lock(struct irq_data *d)
+{
+	struct gpio_chip *gc = irq_data_get_irq_chip_data(d);
+	struct gpio_rcar_priv *p = gpiochip_get_data(gc);
+
+	pm_runtime_get_sync(&p->pdev->dev);
+}
+
+static void gpio_rcar_irq_bus_sync_unlock(struct irq_data *d)
+{
+	struct gpio_chip *gc = irq_data_get_irq_chip_data(d);
+	struct gpio_rcar_priv *p = gpiochip_get_data(gc);
+
+	pm_runtime_put(&p->pdev->dev);
+}
+
+
+static int gpio_rcar_irq_request_resources(struct irq_data *d)
+{
+	struct gpio_chip *gc = irq_data_get_irq_chip_data(d);
+	struct gpio_rcar_priv *p = gpiochip_get_data(gc);
+	int error;
+
+	error = pm_runtime_get_sync(&p->pdev->dev);
+	if (error < 0)
+		return error;
+
+	return 0;
+}
+
+static void gpio_rcar_irq_release_resources(struct irq_data *d)
+{
+	struct gpio_chip *gc = irq_data_get_irq_chip_data(d);
+	struct gpio_rcar_priv *p = gpiochip_get_data(gc);
+
+	pm_runtime_put(&p->pdev->dev);
+}
+
 static irqreturn_t gpio_rcar_irq_handler(int irq, void *dev_id)
 {
 	struct gpio_rcar_priv *p = dev_id;
@@ -450,6 +488,10 @@ static int gpio_rcar_probe(struct platform_device *pdev)
 	irq_chip->irq_unmask = gpio_rcar_irq_enable;
 	irq_chip->irq_set_type = gpio_rcar_irq_set_type;
 	irq_chip->irq_set_wake = gpio_rcar_irq_set_wake;
+	irq_chip->irq_bus_lock = gpio_rcar_irq_bus_lock;
+	irq_chip->irq_bus_sync_unlock = gpio_rcar_irq_bus_sync_unlock;
+	irq_chip->irq_request_resources = gpio_rcar_irq_request_resources;
+	irq_chip->irq_release_resources = gpio_rcar_irq_release_resources;
 	irq_chip->flags	= IRQCHIP_SET_TYPE_MASKED | IRQCHIP_MASK_ON_SUSPEND;
 
 	ret = gpiochip_add_data(gpio_chip, p);
diff --git a/drivers/gpu/drm/amd/acp/Kconfig b/drivers/gpu/drm/amd/acp/Kconfig
index 2b07813bceed..0f734ee05274 100644
--- a/drivers/gpu/drm/amd/acp/Kconfig
+++ b/drivers/gpu/drm/amd/acp/Kconfig
@@ -2,7 +2,6 @@ menu "ACP Configuration"
 
 config DRM_AMD_ACP
        bool "Enable ACP IP support"
-       default y
        select MFD_CORE
        select PM_GENERIC_DOMAINS if PM
        help
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu.h b/drivers/gpu/drm/amd/amdgpu/amdgpu.h
index f5bac97a438b..c4a21c6428f5 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu.h
@@ -85,6 +85,8 @@ extern int amdgpu_vm_debug;
 extern int amdgpu_sched_jobs;
 extern int amdgpu_sched_hw_submission;
 extern int amdgpu_powerplay;
+extern unsigned amdgpu_pcie_gen_cap;
+extern unsigned amdgpu_pcie_lane_cap;
 
 #define AMDGPU_WAIT_IDLE_TIMEOUT_IN_MS	        3000
 #define AMDGPU_MAX_USEC_TIMEOUT			100000	/* 100 ms */
@@ -127,47 +129,6 @@ extern int amdgpu_powerplay;
 #define AMDGPU_RESET_VCE			(1 << 13)
 #define AMDGPU_RESET_VCE1			(1 << 14)
 
-/* CG block flags */
-#define AMDGPU_CG_BLOCK_GFX			(1 << 0)
-#define AMDGPU_CG_BLOCK_MC			(1 << 1)
-#define AMDGPU_CG_BLOCK_SDMA			(1 << 2)
-#define AMDGPU_CG_BLOCK_UVD			(1 << 3)
-#define AMDGPU_CG_BLOCK_VCE			(1 << 4)
-#define AMDGPU_CG_BLOCK_HDP			(1 << 5)
-#define AMDGPU_CG_BLOCK_BIF			(1 << 6)
-
-/* CG flags */
-#define AMDGPU_CG_SUPPORT_GFX_MGCG		(1 << 0)
-#define AMDGPU_CG_SUPPORT_GFX_MGLS		(1 << 1)
-#define AMDGPU_CG_SUPPORT_GFX_CGCG		(1 << 2)
-#define AMDGPU_CG_SUPPORT_GFX_CGLS		(1 << 3)
-#define AMDGPU_CG_SUPPORT_GFX_CGTS		(1 << 4)
-#define AMDGPU_CG_SUPPORT_GFX_CGTS_LS		(1 << 5)
-#define AMDGPU_CG_SUPPORT_GFX_CP_LS		(1 << 6)
-#define AMDGPU_CG_SUPPORT_GFX_RLC_LS		(1 << 7)
-#define AMDGPU_CG_SUPPORT_MC_LS			(1 << 8)
-#define AMDGPU_CG_SUPPORT_MC_MGCG		(1 << 9)
-#define AMDGPU_CG_SUPPORT_SDMA_LS		(1 << 10)
-#define AMDGPU_CG_SUPPORT_SDMA_MGCG		(1 << 11)
-#define AMDGPU_CG_SUPPORT_BIF_LS		(1 << 12)
-#define AMDGPU_CG_SUPPORT_UVD_MGCG		(1 << 13)
-#define AMDGPU_CG_SUPPORT_VCE_MGCG		(1 << 14)
-#define AMDGPU_CG_SUPPORT_HDP_LS		(1 << 15)
-#define AMDGPU_CG_SUPPORT_HDP_MGCG		(1 << 16)
-
-/* PG flags */
-#define AMDGPU_PG_SUPPORT_GFX_PG		(1 << 0)
-#define AMDGPU_PG_SUPPORT_GFX_SMG		(1 << 1)
-#define AMDGPU_PG_SUPPORT_GFX_DMG		(1 << 2)
-#define AMDGPU_PG_SUPPORT_UVD			(1 << 3)
-#define AMDGPU_PG_SUPPORT_VCE			(1 << 4)
-#define AMDGPU_PG_SUPPORT_CP			(1 << 5)
-#define AMDGPU_PG_SUPPORT_GDS			(1 << 6)
-#define AMDGPU_PG_SUPPORT_RLC_SMU_HS		(1 << 7)
-#define AMDGPU_PG_SUPPORT_SDMA			(1 << 8)
-#define AMDGPU_PG_SUPPORT_ACP			(1 << 9)
-#define AMDGPU_PG_SUPPORT_SAMU			(1 << 10)
-
 /* GFX current status */
 #define AMDGPU_GFX_NORMAL_MODE			0x00000000L
 #define AMDGPU_GFX_SAFE_MODE			0x00000001L
@@ -180,7 +141,6 @@ extern int amdgpu_powerplay;
 #define CIK_CURSOR_HEIGHT 128
 
 struct amdgpu_device;
-struct amdgpu_fence;
 struct amdgpu_ib;
 struct amdgpu_vm;
 struct amdgpu_ring;
@@ -326,9 +286,11 @@ struct amdgpu_ring_funcs {
 			struct amdgpu_ib *ib);
 	void (*emit_fence)(struct amdgpu_ring *ring, uint64_t addr,
 			   uint64_t seq, unsigned flags);
+	void (*emit_pipeline_sync)(struct amdgpu_ring *ring);
 	void (*emit_vm_flush)(struct amdgpu_ring *ring, unsigned vm_id,
 			      uint64_t pd_addr);
 	void (*emit_hdp_flush)(struct amdgpu_ring *ring);
+	void (*emit_hdp_invalidate)(struct amdgpu_ring *ring);
 	void (*emit_gds_switch)(struct amdgpu_ring *ring, uint32_t vmid,
 				uint32_t gds_base, uint32_t gds_size,
 				uint32_t gws_base, uint32_t gws_size,
@@ -385,13 +347,15 @@ struct amdgpu_fence_driver {
 	uint64_t			gpu_addr;
 	volatile uint32_t		*cpu_addr;
 	/* sync_seq is protected by ring emission lock */
-	uint64_t			sync_seq;
-	atomic64_t			last_seq;
+	uint32_t			sync_seq;
+	atomic_t			last_seq;
 	bool				initialized;
 	struct amdgpu_irq_src		*irq_src;
 	unsigned			irq_type;
 	struct timer_list		fallback_timer;
-	wait_queue_head_t		fence_queue;
+	unsigned			num_fences_mask;
+	spinlock_t			lock;
+	struct fence			**fences;
 };
 
 /* some special values for the owner field */
@@ -401,19 +365,6 @@ struct amdgpu_fence_driver {
 #define AMDGPU_FENCE_FLAG_64BIT         (1 << 0)
 #define AMDGPU_FENCE_FLAG_INT           (1 << 1)
 
-struct amdgpu_fence {
-	struct fence base;
-
-	/* RB, DMA, etc. */
-	struct amdgpu_ring		*ring;
-	uint64_t			seq;
-
-	/* filp or special value for fence creator */
-	void				*owner;
-
-	wait_queue_t			fence_wake;
-};
-
 struct amdgpu_user_fence {
 	/* write-back bo */
 	struct amdgpu_bo 	*bo;
@@ -425,16 +376,15 @@ int amdgpu_fence_driver_init(struct amdgpu_device *adev);
 void amdgpu_fence_driver_fini(struct amdgpu_device *adev);
 void amdgpu_fence_driver_force_completion(struct amdgpu_device *adev);
 
-int amdgpu_fence_driver_init_ring(struct amdgpu_ring *ring);
+int amdgpu_fence_driver_init_ring(struct amdgpu_ring *ring,
+				  unsigned num_hw_submission);
 int amdgpu_fence_driver_start_ring(struct amdgpu_ring *ring,
 				   struct amdgpu_irq_src *irq_src,
 				   unsigned irq_type);
 void amdgpu_fence_driver_suspend(struct amdgpu_device *adev);
 void amdgpu_fence_driver_resume(struct amdgpu_device *adev);
-int amdgpu_fence_emit(struct amdgpu_ring *ring, void *owner,
-		      struct amdgpu_fence **fence);
+int amdgpu_fence_emit(struct amdgpu_ring *ring, struct fence **fence);
 void amdgpu_fence_process(struct amdgpu_ring *ring);
-int amdgpu_fence_wait_next(struct amdgpu_ring *ring);
 int amdgpu_fence_wait_empty(struct amdgpu_ring *ring);
 unsigned amdgpu_fence_count_emitted(struct amdgpu_ring *ring);
 
@@ -473,6 +423,8 @@ struct amdgpu_bo_list_entry {
 	struct ttm_validate_buffer	tv;
 	struct amdgpu_bo_va		*bo_va;
 	uint32_t			priority;
+	struct page			**user_pages;
+	int				user_invalidated;
 };
 
 struct amdgpu_bo_va_mapping {
@@ -484,7 +436,6 @@ struct amdgpu_bo_va_mapping {
 
 /* bo virtual addresses in a specific vm */
 struct amdgpu_bo_va {
-	struct mutex		        mutex;
 	/* protected by bo being reserved */
 	struct list_head		bo_list;
 	struct fence		        *last_pt_update;
@@ -579,11 +530,14 @@ int amdgpu_gem_debugfs_init(struct amdgpu_device *adev);
  * Assumption is that there won't be hole (all object on same
  * alignment).
  */
+
+#define AMDGPU_SA_NUM_FENCE_LISTS	32
+
 struct amdgpu_sa_manager {
 	wait_queue_head_t	wq;
 	struct amdgpu_bo	*bo;
 	struct list_head	*hole;
-	struct list_head	flist[AMDGPU_MAX_RINGS];
+	struct list_head	flist[AMDGPU_SA_NUM_FENCE_LISTS];
 	struct list_head	olist;
 	unsigned		size;
 	uint64_t		gpu_addr;
@@ -592,8 +546,6 @@ struct amdgpu_sa_manager {
 	uint32_t		align;
 };
 
-struct amdgpu_sa_bo;
-
 /* sub-allocation buffer */
 struct amdgpu_sa_bo {
 	struct list_head		olist;
@@ -637,6 +589,8 @@ int amdgpu_sync_resv(struct amdgpu_device *adev,
 struct fence *amdgpu_sync_get_fence(struct amdgpu_sync *sync);
 int amdgpu_sync_wait(struct amdgpu_sync *sync);
 void amdgpu_sync_free(struct amdgpu_sync *sync);
+int amdgpu_sync_init(void);
+void amdgpu_sync_fini(void);
 
 /*
  * GART structures, functions & helpers
@@ -767,10 +721,10 @@ struct amdgpu_ib {
 	uint32_t			length_dw;
 	uint64_t			gpu_addr;
 	uint32_t			*ptr;
-	struct amdgpu_fence		*fence;
 	struct amdgpu_user_fence        *user;
-	bool				grabbed_vmid;
 	struct amdgpu_vm		*vm;
+	unsigned			vm_id;
+	uint64_t			vm_pd_addr;
 	struct amdgpu_ctx		*ctx;
 	uint32_t			gds_base, gds_size;
 	uint32_t			gws_base, gws_size;
@@ -877,15 +831,14 @@ struct amdgpu_vm_pt {
 };
 
 struct amdgpu_vm_id {
-	unsigned		id;
-	uint64_t		pd_gpu_addr;
+	struct amdgpu_vm_manager_id	*mgr_id;
+	uint64_t			pd_gpu_addr;
 	/* last flushed PD/PT update */
-	struct fence	        *flushed_updates;
+	struct fence			*flushed_updates;
 };
 
 struct amdgpu_vm {
 	/* tree of virtual addresses mapped */
-	spinlock_t		it_lock;
 	struct rb_root		va;
 
 	/* protecting invalidated */
@@ -922,6 +875,13 @@ struct amdgpu_vm_manager_id {
 	struct list_head	list;
 	struct fence		*active;
 	atomic_long_t		owner;
+
+	uint32_t		gds_base;
+	uint32_t		gds_size;
+	uint32_t		gws_base;
+	uint32_t		gws_size;
+	uint32_t		oa_base;
+	uint32_t		oa_size;
 };
 
 struct amdgpu_vm_manager {
@@ -954,10 +914,14 @@ void amdgpu_vm_get_pt_bos(struct amdgpu_vm *vm, struct list_head *duplicates);
 void amdgpu_vm_move_pt_bos_in_lru(struct amdgpu_device *adev,
 				  struct amdgpu_vm *vm);
 int amdgpu_vm_grab_id(struct amdgpu_vm *vm, struct amdgpu_ring *ring,
-		      struct amdgpu_sync *sync, struct fence *fence);
+		      struct amdgpu_sync *sync, struct fence *fence,
+		      unsigned *vm_id, uint64_t *vm_pd_addr);
 void amdgpu_vm_flush(struct amdgpu_ring *ring,
-		     struct amdgpu_vm *vm,
-		     struct fence *updates);
+		     unsigned vm_id, uint64_t pd_addr,
+		     uint32_t gds_base, uint32_t gds_size,
+		     uint32_t gws_base, uint32_t gws_size,
+		     uint32_t oa_base, uint32_t oa_size);
+void amdgpu_vm_reset_id(struct amdgpu_device *adev, unsigned vm_id);
 uint64_t amdgpu_vm_map_gart(const dma_addr_t *pages_addr, uint64_t addr);
 int amdgpu_vm_update_page_directory(struct amdgpu_device *adev,
 				    struct amdgpu_vm *vm);
@@ -1045,7 +1009,7 @@ struct amdgpu_bo_list {
 	struct amdgpu_bo *gds_obj;
 	struct amdgpu_bo *gws_obj;
 	struct amdgpu_bo *oa_obj;
-	bool has_userptr;
+	unsigned first_userptr;
 	unsigned num_entries;
 	struct amdgpu_bo_list_entry *array;
 };
@@ -1172,10 +1136,9 @@ struct amdgpu_gfx {
 
 int amdgpu_ib_get(struct amdgpu_device *adev, struct amdgpu_vm *vm,
 		  unsigned size, struct amdgpu_ib *ib);
-void amdgpu_ib_free(struct amdgpu_device *adev, struct amdgpu_ib *ib);
+void amdgpu_ib_free(struct amdgpu_device *adev, struct amdgpu_ib *ib, struct fence *f);
 int amdgpu_ib_schedule(struct amdgpu_ring *ring, unsigned num_ibs,
-		       struct amdgpu_ib *ib, void *owner,
-		       struct fence *last_vm_update,
+		       struct amdgpu_ib *ib, struct fence *last_vm_update,
 		       struct fence **f);
 int amdgpu_ib_pool_init(struct amdgpu_device *adev);
 void amdgpu_ib_pool_fini(struct amdgpu_device *adev);
@@ -1194,7 +1157,6 @@ int amdgpu_ring_init(struct amdgpu_device *adev, struct amdgpu_ring *ring,
 		     struct amdgpu_irq_src *irq_src, unsigned irq_type,
 		     enum amdgpu_ring_type ring_type);
 void amdgpu_ring_fini(struct amdgpu_ring *ring);
-struct amdgpu_ring *amdgpu_ring_from_fence(struct fence *f);
 
 /*
  * CS.
@@ -1236,6 +1198,7 @@ struct amdgpu_job {
 	struct amdgpu_ring	*ring;
 	struct amdgpu_sync	sync;
 	struct amdgpu_ib	*ibs;
+	struct fence		*fence; /* the hw fence */
 	uint32_t		num_ibs;
 	void			*owner;
 	struct amdgpu_user_fence uf;
@@ -2051,7 +2014,6 @@ struct amdgpu_device {
 	struct amdgpu_sdma		sdma;
 
 	/* uvd */
-	bool				has_uvd;
 	struct amdgpu_uvd		uvd;
 
 	/* vce */
@@ -2098,20 +2060,6 @@ u32 amdgpu_mm_rdoorbell(struct amdgpu_device *adev, u32 index);
 void amdgpu_mm_wdoorbell(struct amdgpu_device *adev, u32 index, u32 v);
 
 /*
- * Cast helper
- */
-extern const struct fence_ops amdgpu_fence_ops;
-static inline struct amdgpu_fence *to_amdgpu_fence(struct fence *f)
-{
-	struct amdgpu_fence *__f = container_of(f, struct amdgpu_fence, base);
-
-	if (__f->base.ops == &amdgpu_fence_ops)
-		return __f;
-
-	return NULL;
-}
-
-/*
  * Registers read & write functions.
  */
 #define RREG32(reg) amdgpu_mm_rreg(adev, (reg), false)
@@ -2225,10 +2173,12 @@ amdgpu_get_sdma_instance(struct amdgpu_ring *ring)
 #define amdgpu_ring_get_wptr(r) (r)->funcs->get_wptr((r))
 #define amdgpu_ring_set_wptr(r) (r)->funcs->set_wptr((r))
 #define amdgpu_ring_emit_ib(r, ib) (r)->funcs->emit_ib((r), (ib))
+#define amdgpu_ring_emit_pipeline_sync(r) (r)->funcs->emit_pipeline_sync((r))
 #define amdgpu_ring_emit_vm_flush(r, vmid, addr) (r)->funcs->emit_vm_flush((r), (vmid), (addr))
 #define amdgpu_ring_emit_fence(r, addr, seq, flags) (r)->funcs->emit_fence((r), (addr), (seq), (flags))
 #define amdgpu_ring_emit_gds_switch(r, v, db, ds, wb, ws, ab, as) (r)->funcs->emit_gds_switch((r), (v), (db), (ds), (wb), (ws), (ab), (as))
 #define amdgpu_ring_emit_hdp_flush(r) (r)->funcs->emit_hdp_flush((r))
+#define amdgpu_ring_emit_hdp_invalidate(r) (r)->funcs->emit_hdp_invalidate((r))
 #define amdgpu_ring_pad_ib(r, ib) ((r)->funcs->pad_ib((r), (ib)))
 #define amdgpu_ih_get_wptr(adev) (adev)->irq.ih_funcs->get_wptr((adev))
 #define amdgpu_ih_decode_iv(adev, iv) (adev)->irq.ih_funcs->decode_iv((adev), (iv))
@@ -2353,11 +2303,15 @@ int amdgpu_cs_get_ring(struct amdgpu_device *adev, u32 ip_type,
 		       struct amdgpu_ring **out_ring);
 void amdgpu_ttm_placement_from_domain(struct amdgpu_bo *rbo, u32 domain);
 bool amdgpu_ttm_bo_is_amdgpu_bo(struct ttm_buffer_object *bo);
+int amdgpu_ttm_tt_get_user_pages(struct ttm_tt *ttm, struct page **pages);
 int amdgpu_ttm_tt_set_userptr(struct ttm_tt *ttm, uint64_t addr,
 				     uint32_t flags);
+bool amdgpu_ttm_tt_has_userptr(struct ttm_tt *ttm);
 struct mm_struct *amdgpu_ttm_tt_get_usermm(struct ttm_tt *ttm);
 bool amdgpu_ttm_tt_affect_userptr(struct ttm_tt *ttm, unsigned long start,
 				  unsigned long end);
+bool amdgpu_ttm_tt_userptr_invalidated(struct ttm_tt *ttm,
+				       int *last_invalidated);
 bool amdgpu_ttm_tt_is_readonly(struct ttm_tt *ttm);
 uint32_t amdgpu_ttm_tt_pte_flags(struct amdgpu_device *adev, struct ttm_tt *ttm,
 				 struct ttm_mem_reg *mem);
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_acp.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_acp.c
index 9f8cfaab3004..d6b0bff510aa 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_acp.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_acp.c
@@ -240,12 +240,10 @@ static int acp_poweron(struct generic_pm_domain *genpd)
 static struct device *get_mfd_cell_dev(const char *device_name, int r)
 {
 	char auto_dev_name[25];
-	char buf[8];
 	struct device *dev;
 
-	sprintf(buf, ".%d.auto", r);
-	strcpy(auto_dev_name, device_name);
-	strcat(auto_dev_name, buf);
+	snprintf(auto_dev_name, sizeof(auto_dev_name),
+		 "%s.%d.auto", device_name, r);
 	dev = bus_find_device_by_name(&platform_bus_type, NULL, auto_dev_name);
 	dev_info(dev, "device %s added to pm domain\n", auto_dev_name);
 
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.c
index 84d68d658f8a..32809f749903 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.c
@@ -30,25 +30,38 @@ const struct kfd2kgd_calls *kfd2kgd;
 const struct kgd2kfd_calls *kgd2kfd;
 bool (*kgd2kfd_init_p)(unsigned, const struct kgd2kfd_calls**);
 
-bool amdgpu_amdkfd_init(void)
+int amdgpu_amdkfd_init(void)
 {
+	int ret;
+
 #if defined(CONFIG_HSA_AMD_MODULE)
-	bool (*kgd2kfd_init_p)(unsigned, const struct kgd2kfd_calls**);
+	int (*kgd2kfd_init_p)(unsigned, const struct kgd2kfd_calls**);
 
 	kgd2kfd_init_p = symbol_request(kgd2kfd_init);
 
 	if (kgd2kfd_init_p == NULL)
-		return false;
+		return -ENOENT;
+
+	ret = kgd2kfd_init_p(KFD_INTERFACE_VERSION, &kgd2kfd);
+	if (ret) {
+		symbol_put(kgd2kfd_init);
+		kgd2kfd = NULL;
+	}
+
+#elif defined(CONFIG_HSA_AMD)
+	ret = kgd2kfd_init(KFD_INTERFACE_VERSION, &kgd2kfd);
+	if (ret)
+		kgd2kfd = NULL;
+
+#else
+	ret = -ENOENT;
 #endif
-	return true;
+
+	return ret;
 }
 
 bool amdgpu_amdkfd_load_interface(struct amdgpu_device *rdev)
 {
-#if defined(CONFIG_HSA_AMD_MODULE)
-	bool (*kgd2kfd_init_p)(unsigned, const struct kgd2kfd_calls**);
-#endif
-
 	switch (rdev->asic_type) {
 #ifdef CONFIG_DRM_AMDGPU_CIK
 	case CHIP_KAVERI:
@@ -62,35 +75,7 @@ bool amdgpu_amdkfd_load_interface(struct amdgpu_device *rdev)
 		return false;
 	}
 
-#if defined(CONFIG_HSA_AMD_MODULE)
-	kgd2kfd_init_p = symbol_request(kgd2kfd_init);
-
-	if (kgd2kfd_init_p == NULL) {
-		kfd2kgd = NULL;
-		return false;
-	}
-
-	if (!kgd2kfd_init_p(KFD_INTERFACE_VERSION, &kgd2kfd)) {
-		symbol_put(kgd2kfd_init);
-		kfd2kgd = NULL;
-		kgd2kfd = NULL;
-
-		return false;
-	}
-
 	return true;
-#elif defined(CONFIG_HSA_AMD)
-	if (!kgd2kfd_init(KFD_INTERFACE_VERSION, &kgd2kfd)) {
-		kfd2kgd = NULL;
-		kgd2kfd = NULL;
-		return false;
-	}
-
-	return true;
-#else
-	kfd2kgd = NULL;
-	return false;
-#endif
 }
 
 void amdgpu_amdkfd_fini(void)
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.h
index a8be765542e6..de530f68d4e3 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd.h
@@ -36,7 +36,7 @@ struct kgd_mem {
 	void *cpu_ptr;
 };
 
-bool amdgpu_amdkfd_init(void);
+int amdgpu_amdkfd_init(void);
 void amdgpu_amdkfd_fini(void);
 
 bool amdgpu_amdkfd_load_interface(struct amdgpu_device *rdev);
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_atpx_handler.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_atpx_handler.c
index fa948dcbdd5d..0020a0ea43ff 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_atpx_handler.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_atpx_handler.c
@@ -63,6 +63,10 @@ bool amdgpu_has_atpx(void) {
 	return amdgpu_atpx_priv.atpx_detected;
 }
 
+bool amdgpu_has_atpx_dgpu_power_cntl(void) {
+	return amdgpu_atpx_priv.atpx.functions.power_cntl;
+}
+
 /**
  * amdgpu_atpx_call - call an ATPX method
  *
@@ -142,10 +146,6 @@ static void amdgpu_atpx_parse_functions(struct amdgpu_atpx_functions *f, u32 mas
  */
 static int amdgpu_atpx_validate(struct amdgpu_atpx *atpx)
 {
-	/* make sure required functions are enabled */
-	/* dGPU power control is required */
-	atpx->functions.power_cntl = true;
-
 	if (atpx->functions.px_params) {
 		union acpi_object *info;
 		struct atpx_px_params output;
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_bo_list.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_bo_list.c
index 90d6fc1618aa..eacd810fc09b 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_bo_list.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_bo_list.c
@@ -91,7 +91,7 @@ static int amdgpu_bo_list_set(struct amdgpu_device *adev,
 	struct amdgpu_bo *gws_obj = adev->gds.gws_gfx_bo;
 	struct amdgpu_bo *oa_obj = adev->gds.oa_gfx_bo;
 
-	bool has_userptr = false;
+	unsigned last_entry = 0, first_userptr = num_entries;
 	unsigned i;
 	int r;
 
@@ -101,8 +101,9 @@ static int amdgpu_bo_list_set(struct amdgpu_device *adev,
 	memset(array, 0, num_entries * sizeof(struct amdgpu_bo_list_entry));
 
 	for (i = 0; i < num_entries; ++i) {
-		struct amdgpu_bo_list_entry *entry = &array[i];
+		struct amdgpu_bo_list_entry *entry;
 		struct drm_gem_object *gobj;
+		struct amdgpu_bo *bo;
 		struct mm_struct *usermm;
 
 		gobj = drm_gem_object_lookup(adev->ddev, filp, info[i].bo_handle);
@@ -111,18 +112,24 @@ static int amdgpu_bo_list_set(struct amdgpu_device *adev,
 			goto error_free;
 		}
 
-		entry->robj = amdgpu_bo_ref(gem_to_amdgpu_bo(gobj));
+		bo = amdgpu_bo_ref(gem_to_amdgpu_bo(gobj));
 		drm_gem_object_unreference_unlocked(gobj);
-		entry->priority = min(info[i].bo_priority,
-				      AMDGPU_BO_LIST_MAX_PRIORITY);
-		usermm = amdgpu_ttm_tt_get_usermm(entry->robj->tbo.ttm);
+
+		usermm = amdgpu_ttm_tt_get_usermm(bo->tbo.ttm);
 		if (usermm) {
 			if (usermm != current->mm) {
+				amdgpu_bo_unref(&bo);
 				r = -EPERM;
 				goto error_free;
 			}
-			has_userptr = true;
+			entry = &array[--first_userptr];
+		} else {
+			entry = &array[last_entry++];
 		}
+
+		entry->robj = bo;
+		entry->priority = min(info[i].bo_priority,
+				      AMDGPU_BO_LIST_MAX_PRIORITY);
 		entry->tv.bo = &entry->robj->tbo;
 		entry->tv.shared = true;
 
@@ -144,13 +151,15 @@ static int amdgpu_bo_list_set(struct amdgpu_device *adev,
 	list->gds_obj = gds_obj;
 	list->gws_obj = gws_obj;
 	list->oa_obj = oa_obj;
-	list->has_userptr = has_userptr;
+	list->first_userptr = first_userptr;
 	list->array = array;
 	list->num_entries = num_entries;
 
 	return 0;
 
 error_free:
+	while (i--)
+		amdgpu_bo_unref(&array[i].robj);
 	drm_free_large(array);
 	return r;
 }
@@ -191,6 +200,7 @@ void amdgpu_bo_list_get_list(struct amdgpu_bo_list *list,
 
 		list_add_tail(&list->array[i].tv.head,
 			      &bucket[priority]);
+		list->array[i].user_pages = NULL;
 	}
 
 	/* Connect the sorted buckets in the output list. */
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_cgs.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_cgs.c
index a081dda9fa2f..7a4b101e10c6 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_cgs.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_cgs.c
@@ -795,6 +795,12 @@ static int amdgpu_cgs_query_system_info(void *cgs_device,
 	case CGS_SYSTEM_INFO_PCIE_MLW:
 		sys_info->value = adev->pm.pcie_mlw_mask;
 		break;
+	case CGS_SYSTEM_INFO_CG_FLAGS:
+		sys_info->value = adev->cg_flags;
+		break;
+	case CGS_SYSTEM_INFO_PG_FLAGS:
+		sys_info->value = adev->pg_flags;
+		break;
 	default:
 		return -ENODEV;
 	}
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_connectors.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_connectors.c
index 89c3dd62ba21..119cdc2c43e7 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_connectors.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_connectors.c
@@ -77,7 +77,7 @@ void amdgpu_connector_hotplug(struct drm_connector *connector)
 			} else if (amdgpu_atombios_dp_needs_link_train(amdgpu_connector)) {
 				/* Don't try to start link training before we
 				 * have the dpcd */
-				if (!amdgpu_atombios_dp_get_dpcd(amdgpu_connector))
+				if (amdgpu_atombios_dp_get_dpcd(amdgpu_connector))
 					return;
 
 				/* set it to OFF so that drm_helper_connector_dpms()
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c
index 52c3eb96b199..9392e50a7ba4 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c
@@ -25,6 +25,7 @@
  *    Jerome Glisse <glisse@freedesktop.org>
  */
 #include <linux/list_sort.h>
+#include <linux/pagemap.h>
 #include <drm/drmP.h>
 #include <drm/amdgpu_drm.h>
 #include "amdgpu.h"
@@ -111,6 +112,7 @@ static int amdgpu_cs_user_fence_chunk(struct amdgpu_cs_parser *p,
 	p->uf_entry.priority = 0;
 	p->uf_entry.tv.bo = &p->uf_entry.robj->tbo;
 	p->uf_entry.tv.shared = true;
+	p->uf_entry.user_pages = NULL;
 
 	drm_gem_object_unreference_unlocked(gobj);
 	return 0;
@@ -297,6 +299,7 @@ int amdgpu_cs_list_validate(struct amdgpu_cs_parser *p,
 
 	list_for_each_entry(lobj, validated, tv.head) {
 		struct amdgpu_bo *bo = lobj->robj;
+		bool binding_userptr = false;
 		struct mm_struct *usermm;
 		uint32_t domain;
 
@@ -304,6 +307,15 @@ int amdgpu_cs_list_validate(struct amdgpu_cs_parser *p,
 		if (usermm && usermm != current->mm)
 			return -EPERM;
 
+		/* Check if we have user pages and nobody bound the BO already */
+		if (lobj->user_pages && bo->tbo.ttm->state != tt_bound) {
+			size_t size = sizeof(struct page *);
+
+			size *= bo->tbo.ttm->num_pages;
+			memcpy(bo->tbo.ttm->pages, lobj->user_pages, size);
+			binding_userptr = true;
+		}
+
 		if (bo->pin_count)
 			continue;
 
@@ -334,6 +346,11 @@ int amdgpu_cs_list_validate(struct amdgpu_cs_parser *p,
 			}
 			return r;
 		}
+
+		if (binding_userptr) {
+			drm_free_large(lobj->user_pages);
+			lobj->user_pages = NULL;
+		}
 	}
 	return 0;
 }
@@ -342,15 +359,18 @@ static int amdgpu_cs_parser_bos(struct amdgpu_cs_parser *p,
 				union drm_amdgpu_cs *cs)
 {
 	struct amdgpu_fpriv *fpriv = p->filp->driver_priv;
+	struct amdgpu_bo_list_entry *e;
 	struct list_head duplicates;
 	bool need_mmap_lock = false;
+	unsigned i, tries = 10;
 	int r;
 
 	INIT_LIST_HEAD(&p->validated);
 
 	p->bo_list = amdgpu_bo_list_get(fpriv, cs->in.bo_list_handle);
 	if (p->bo_list) {
-		need_mmap_lock = p->bo_list->has_userptr;
+		need_mmap_lock = p->bo_list->first_userptr !=
+			p->bo_list->num_entries;
 		amdgpu_bo_list_get_list(p->bo_list, &p->validated);
 	}
 
@@ -363,9 +383,81 @@ static int amdgpu_cs_parser_bos(struct amdgpu_cs_parser *p,
 	if (need_mmap_lock)
 		down_read(&current->mm->mmap_sem);
 
-	r = ttm_eu_reserve_buffers(&p->ticket, &p->validated, true, &duplicates);
-	if (unlikely(r != 0))
-		goto error_reserve;
+	while (1) {
+		struct list_head need_pages;
+		unsigned i;
+
+		r = ttm_eu_reserve_buffers(&p->ticket, &p->validated, true,
+					   &duplicates);
+		if (unlikely(r != 0))
+			goto error_free_pages;
+
+		/* Without a BO list we don't have userptr BOs */
+		if (!p->bo_list)
+			break;
+
+		INIT_LIST_HEAD(&need_pages);
+		for (i = p->bo_list->first_userptr;
+		     i < p->bo_list->num_entries; ++i) {
+
+			e = &p->bo_list->array[i];
+
+			if (amdgpu_ttm_tt_userptr_invalidated(e->robj->tbo.ttm,
+				 &e->user_invalidated) && e->user_pages) {
+
+				/* We acquired a page array, but somebody
+				 * invalidated it. Free it an try again
+				 */
+				release_pages(e->user_pages,
+					      e->robj->tbo.ttm->num_pages,
+					      false);
+				drm_free_large(e->user_pages);
+				e->user_pages = NULL;
+			}
+
+			if (e->robj->tbo.ttm->state != tt_bound &&
+			    !e->user_pages) {
+				list_del(&e->tv.head);
+				list_add(&e->tv.head, &need_pages);
+
+				amdgpu_bo_unreserve(e->robj);
+			}
+		}
+
+		if (list_empty(&need_pages))
+			break;
+
+		/* Unreserve everything again. */
+		ttm_eu_backoff_reservation(&p->ticket, &p->validated);
+
+		/* We tried to often, just abort */
+		if (!--tries) {
+			r = -EDEADLK;
+			goto error_free_pages;
+		}
+
+		/* Fill the page arrays for all useptrs. */
+		list_for_each_entry(e, &need_pages, tv.head) {
+			struct ttm_tt *ttm = e->robj->tbo.ttm;
+
+			e->user_pages = drm_calloc_large(ttm->num_pages,
+							 sizeof(struct page*));
+			if (!e->user_pages) {
+				r = -ENOMEM;
+				goto error_free_pages;
+			}
+
+			r = amdgpu_ttm_tt_get_user_pages(ttm, e->user_pages);
+			if (r) {
+				drm_free_large(e->user_pages);
+				e->user_pages = NULL;
+				goto error_free_pages;
+			}
+		}
+
+		/* And try again. */
+		list_splice(&need_pages, &p->validated);
+	}
 
 	amdgpu_vm_get_pt_bos(&fpriv->vm, &duplicates);
 
@@ -397,10 +489,26 @@ error_validate:
 		ttm_eu_backoff_reservation(&p->ticket, &p->validated);
 	}
 
-error_reserve:
+error_free_pages:
+
 	if (need_mmap_lock)
 		up_read(&current->mm->mmap_sem);
 
+	if (p->bo_list) {
+		for (i = p->bo_list->first_userptr;
+		     i < p->bo_list->num_entries; ++i) {
+			e = &p->bo_list->array[i];
+
+			if (!e->user_pages)
+				continue;
+
+			release_pages(e->user_pages,
+				      e->robj->tbo.ttm->num_pages,
+				      false);
+			drm_free_large(e->user_pages);
+		}
+	}
+
 	return r;
 }
 
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
index db20d2783def..612117478b57 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
@@ -62,6 +62,12 @@ static const char *amdgpu_asic_name[] = {
 	"LAST",
 };
 
+#if defined(CONFIG_VGA_SWITCHEROO)
+bool amdgpu_has_atpx_dgpu_power_cntl(void);
+#else
+static inline bool amdgpu_has_atpx_dgpu_power_cntl(void) { return false; }
+#endif
+
 bool amdgpu_device_is_px(struct drm_device *dev)
 {
 	struct amdgpu_device *adev = dev->dev_private;
@@ -1479,7 +1485,7 @@ int amdgpu_device_init(struct amdgpu_device *adev,
 
 	if (amdgpu_runtime_pm == 1)
 		runtime = true;
-	if (amdgpu_device_is_px(ddev))
+	if (amdgpu_device_is_px(ddev) && amdgpu_has_atpx_dgpu_power_cntl())
 		runtime = true;
 	vga_switcheroo_register_client(adev->pdev, &amdgpu_switcheroo_ops, runtime);
 	if (runtime)
@@ -1762,15 +1768,20 @@ int amdgpu_resume_kms(struct drm_device *dev, bool resume, bool fbcon)
 	}
 
 	/* post card */
-	amdgpu_atom_asic_init(adev->mode_info.atom_context);
+	if (!amdgpu_card_posted(adev))
+		amdgpu_atom_asic_init(adev->mode_info.atom_context);
 
 	r = amdgpu_resume(adev);
+	if (r)
+		DRM_ERROR("amdgpu_resume failed (%d).\n", r);
 
 	amdgpu_fence_driver_resume(adev);
 
-	r = amdgpu_ib_ring_tests(adev);
-	if (r)
-		DRM_ERROR("ib ring test failed (%d).\n", r);
+	if (resume) {
+		r = amdgpu_ib_ring_tests(adev);
+		if (r)
+			DRM_ERROR("ib ring test failed (%d).\n", r);
+	}
 
 	r = amdgpu_late_init(adev);
 	if (r)
@@ -1903,80 +1914,97 @@ retry:
 	return r;
 }
 
+#define AMDGPU_DEFAULT_PCIE_GEN_MASK 0x30007  /* gen: chipset 1/2, asic 1/2/3 */
+#define AMDGPU_DEFAULT_PCIE_MLW_MASK 0x2f0000 /* 1/2/4/8/16 lanes */
+
 void amdgpu_get_pcie_info(struct amdgpu_device *adev)
 {
 	u32 mask;
 	int ret;
 
-	if (pci_is_root_bus(adev->pdev->bus))
-		return;
+	if (amdgpu_pcie_gen_cap)
+		adev->pm.pcie_gen_mask = amdgpu_pcie_gen_cap;
 
-	if (amdgpu_pcie_gen2 == 0)
-		return;
+	if (amdgpu_pcie_lane_cap)
+		adev->pm.pcie_mlw_mask = amdgpu_pcie_lane_cap;
 
-	if (adev->flags & AMD_IS_APU)
+	/* covers APUs as well */
+	if (pci_is_root_bus(adev->pdev->bus)) {
+		if (adev->pm.pcie_gen_mask == 0)
+			adev->pm.pcie_gen_mask = AMDGPU_DEFAULT_PCIE_GEN_MASK;
+		if (adev->pm.pcie_mlw_mask == 0)
+			adev->pm.pcie_mlw_mask = AMDGPU_DEFAULT_PCIE_MLW_MASK;
 		return;
+	}
 
-	ret = drm_pcie_get_speed_cap_mask(adev->ddev, &mask);
-	if (!ret) {
-		adev->pm.pcie_gen_mask = (CAIL_ASIC_PCIE_LINK_SPEED_SUPPORT_GEN1 |
-					  CAIL_ASIC_PCIE_LINK_SPEED_SUPPORT_GEN2 |
-					  CAIL_ASIC_PCIE_LINK_SPEED_SUPPORT_GEN3);
-
-		if (mask & DRM_PCIE_SPEED_25)
-			adev->pm.pcie_gen_mask |= CAIL_PCIE_LINK_SPEED_SUPPORT_GEN1;
-		if (mask & DRM_PCIE_SPEED_50)
-			adev->pm.pcie_gen_mask |= CAIL_PCIE_LINK_SPEED_SUPPORT_GEN2;
-		if (mask & DRM_PCIE_SPEED_80)
-			adev->pm.pcie_gen_mask |= CAIL_PCIE_LINK_SPEED_SUPPORT_GEN3;
-	}
-	ret = drm_pcie_get_max_link_width(adev->ddev, &mask);
-	if (!ret) {
-		switch (mask) {
-		case 32:
-			adev->pm.pcie_mlw_mask = (CAIL_PCIE_LINK_WIDTH_SUPPORT_X32 |
-						  CAIL_PCIE_LINK_WIDTH_SUPPORT_X16 |
-						  CAIL_PCIE_LINK_WIDTH_SUPPORT_X12 |
-						  CAIL_PCIE_LINK_WIDTH_SUPPORT_X8 |
-						  CAIL_PCIE_LINK_WIDTH_SUPPORT_X4 |
-						  CAIL_PCIE_LINK_WIDTH_SUPPORT_X2 |
-						  CAIL_PCIE_LINK_WIDTH_SUPPORT_X1);
-			break;
-		case 16:
-			adev->pm.pcie_mlw_mask = (CAIL_PCIE_LINK_WIDTH_SUPPORT_X16 |
-						  CAIL_PCIE_LINK_WIDTH_SUPPORT_X12 |
-						  CAIL_PCIE_LINK_WIDTH_SUPPORT_X8 |
-						  CAIL_PCIE_LINK_WIDTH_SUPPORT_X4 |
-						  CAIL_PCIE_LINK_WIDTH_SUPPORT_X2 |
-						  CAIL_PCIE_LINK_WIDTH_SUPPORT_X1);
-			break;
-		case 12:
-			adev->pm.pcie_mlw_mask = (CAIL_PCIE_LINK_WIDTH_SUPPORT_X12 |
-						  CAIL_PCIE_LINK_WIDTH_SUPPORT_X8 |
-						  CAIL_PCIE_LINK_WIDTH_SUPPORT_X4 |
-						  CAIL_PCIE_LINK_WIDTH_SUPPORT_X2 |
-						  CAIL_PCIE_LINK_WIDTH_SUPPORT_X1);
-			break;
-		case 8:
-			adev->pm.pcie_mlw_mask = (CAIL_PCIE_LINK_WIDTH_SUPPORT_X8 |
-						  CAIL_PCIE_LINK_WIDTH_SUPPORT_X4 |
-						  CAIL_PCIE_LINK_WIDTH_SUPPORT_X2 |
-						  CAIL_PCIE_LINK_WIDTH_SUPPORT_X1);
-			break;
-		case 4:
-			adev->pm.pcie_mlw_mask = (CAIL_PCIE_LINK_WIDTH_SUPPORT_X4 |
-						  CAIL_PCIE_LINK_WIDTH_SUPPORT_X2 |
-						  CAIL_PCIE_LINK_WIDTH_SUPPORT_X1);
-			break;
-		case 2:
-			adev->pm.pcie_mlw_mask = (CAIL_PCIE_LINK_WIDTH_SUPPORT_X2 |
-						  CAIL_PCIE_LINK_WIDTH_SUPPORT_X1);
-			break;
-		case 1:
-			adev->pm.pcie_mlw_mask = CAIL_PCIE_LINK_WIDTH_SUPPORT_X1;
-			break;
-		default:
-			break;
+	if (adev->pm.pcie_gen_mask == 0) {
+		ret = drm_pcie_get_speed_cap_mask(adev->ddev, &mask);
+		if (!ret) {
+			adev->pm.pcie_gen_mask = (CAIL_ASIC_PCIE_LINK_SPEED_SUPPORT_GEN1 |
+						  CAIL_ASIC_PCIE_LINK_SPEED_SUPPORT_GEN2 |
+						  CAIL_ASIC_PCIE_LINK_SPEED_SUPPORT_GEN3);
+
+			if (mask & DRM_PCIE_SPEED_25)
+				adev->pm.pcie_gen_mask |= CAIL_PCIE_LINK_SPEED_SUPPORT_GEN1;
+			if (mask & DRM_PCIE_SPEED_50)
+				adev->pm.pcie_gen_mask |= CAIL_PCIE_LINK_SPEED_SUPPORT_GEN2;
+			if (mask & DRM_PCIE_SPEED_80)
+				adev->pm.pcie_gen_mask |= CAIL_PCIE_LINK_SPEED_SUPPORT_GEN3;
+		} else {
+			adev->pm.pcie_gen_mask = AMDGPU_DEFAULT_PCIE_GEN_MASK;
+		}
+	}
+	if (adev->pm.pcie_mlw_mask == 0) {
+		ret = drm_pcie_get_max_link_width(adev->ddev, &mask);
+		if (!ret) {
+			switch (mask) {
+			case 32:
+				adev->pm.pcie_mlw_mask = (CAIL_PCIE_LINK_WIDTH_SUPPORT_X32 |
+							  CAIL_PCIE_LINK_WIDTH_SUPPORT_X16 |
+							  CAIL_PCIE_LINK_WIDTH_SUPPORT_X12 |
+							  CAIL_PCIE_LINK_WIDTH_SUPPORT_X8 |
+							  CAIL_PCIE_LINK_WIDTH_SUPPORT_X4 |
+							  CAIL_PCIE_LINK_WIDTH_SUPPORT_X2 |
+							  CAIL_PCIE_LINK_WIDTH_SUPPORT_X1);
+				break;
+			case 16:
+				adev->pm.pcie_mlw_mask = (CAIL_PCIE_LINK_WIDTH_SUPPORT_X16 |
+							  CAIL_PCIE_LINK_WIDTH_SUPPORT_X12 |
+							  CAIL_PCIE_LINK_WIDTH_SUPPORT_X8 |
+							  CAIL_PCIE_LINK_WIDTH_SUPPORT_X4 |
+							  CAIL_PCIE_LINK_WIDTH_SUPPORT_X2 |
+							  CAIL_PCIE_LINK_WIDTH_SUPPORT_X1);
+				break;
+			case 12:
+				adev->pm.pcie_mlw_mask = (CAIL_PCIE_LINK_WIDTH_SUPPORT_X12 |
+							  CAIL_PCIE_LINK_WIDTH_SUPPORT_X8 |
+							  CAIL_PCIE_LINK_WIDTH_SUPPORT_X4 |
+							  CAIL_PCIE_LINK_WIDTH_SUPPORT_X2 |
+							  CAIL_PCIE_LINK_WIDTH_SUPPORT_X1);
+				break;
+			case 8:
+				adev->pm.pcie_mlw_mask = (CAIL_PCIE_LINK_WIDTH_SUPPORT_X8 |
+							  CAIL_PCIE_LINK_WIDTH_SUPPORT_X4 |
+							  CAIL_PCIE_LINK_WIDTH_SUPPORT_X2 |
+							  CAIL_PCIE_LINK_WIDTH_SUPPORT_X1);
+				break;
+			case 4:
+				adev->pm.pcie_mlw_mask = (CAIL_PCIE_LINK_WIDTH_SUPPORT_X4 |
+							  CAIL_PCIE_LINK_WIDTH_SUPPORT_X2 |
+							  CAIL_PCIE_LINK_WIDTH_SUPPORT_X1);
+				break;
+			case 2:
+				adev->pm.pcie_mlw_mask = (CAIL_PCIE_LINK_WIDTH_SUPPORT_X2 |
+							  CAIL_PCIE_LINK_WIDTH_SUPPORT_X1);
+				break;
+			case 1:
+				adev->pm.pcie_mlw_mask = CAIL_PCIE_LINK_WIDTH_SUPPORT_X1;
+				break;
+			default:
+				break;
+			}
+		} else {
+			adev->pm.pcie_mlw_mask = AMDGPU_DEFAULT_PCIE_MLW_MASK;
 		}
 	}
 }
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_display.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_display.c
index 2cb53c24dec0..f0ed974bd4e0 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_display.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_display.c
@@ -70,8 +70,8 @@ static void amdgpu_flip_work_func(struct work_struct *__work)
 
 	struct drm_crtc *crtc = &amdgpuCrtc->base;
 	unsigned long flags;
-	unsigned i;
-	int vpos, hpos, stat, min_udelay;
+	unsigned i, repcnt = 4;
+	int vpos, hpos, stat, min_udelay = 0;
 	struct drm_vblank_crtc *vblank = &crtc->dev->vblank[work->crtc_id];
 
 	if (amdgpu_flip_handle_fence(work, &work->excl))
@@ -97,7 +97,7 @@ static void amdgpu_flip_work_func(struct work_struct *__work)
 	 * In practice this won't execute very often unless on very fast
 	 * machines because the time window for this to happen is very small.
 	 */
-	for (;;) {
+	while (amdgpuCrtc->enabled && --repcnt) {
 		/* GET_DISTANCE_TO_VBLANKSTART returns distance to real vblank
 		 * start in hpos, and to the "fudged earlier" vblank start in
 		 * vpos.
@@ -113,12 +113,24 @@ static void amdgpu_flip_work_func(struct work_struct *__work)
 			break;
 
 		/* Sleep at least until estimated real start of hw vblank */
-		spin_unlock_irqrestore(&crtc->dev->event_lock, flags);
 		min_udelay = (-hpos + 1) * max(vblank->linedur_ns / 1000, 5);
+		if (min_udelay > vblank->framedur_ns / 2000) {
+			/* Don't wait ridiculously long - something is wrong */
+			repcnt = 0;
+			break;
+		}
+		spin_unlock_irqrestore(&crtc->dev->event_lock, flags);
 		usleep_range(min_udelay, 2 * min_udelay);
 		spin_lock_irqsave(&crtc->dev->event_lock, flags);
 	};
 
+	if (!repcnt)
+		DRM_DEBUG_DRIVER("Delay problem on crtc %d: min_udelay %d, "
+				 "framedur %d, linedur %d, stat %d, vpos %d, "
+				 "hpos %d\n", work->crtc_id, min_udelay,
+				 vblank->framedur_ns / 1000,
+				 vblank->linedur_ns / 1000, stat, vpos, hpos);
+
 	/* set the flip status */
 	amdgpuCrtc->pflip_status = AMDGPU_FLIP_SUBMITTED;
 	spin_unlock_irqrestore(&crtc->dev->event_lock, flags);
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c
index ce79a8b605a0..f1e17d60055a 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c
@@ -80,6 +80,8 @@ int amdgpu_exp_hw_support = 0;
 int amdgpu_sched_jobs = 32;
 int amdgpu_sched_hw_submission = 2;
 int amdgpu_powerplay = -1;
+unsigned amdgpu_pcie_gen_cap = 0;
+unsigned amdgpu_pcie_lane_cap = 0;
 
 MODULE_PARM_DESC(vramlimit, "Restrict VRAM for testing, in megabytes");
 module_param_named(vramlimit, amdgpu_vram_limit, int, 0600);
@@ -158,6 +160,12 @@ MODULE_PARM_DESC(powerplay, "Powerplay component (1 = enable, 0 = disable, -1 =
 module_param_named(powerplay, amdgpu_powerplay, int, 0444);
 #endif
 
+MODULE_PARM_DESC(pcie_gen_cap, "PCIE Gen Caps (0: autodetect (default))");
+module_param_named(pcie_gen_cap, amdgpu_pcie_gen_cap, uint, 0444);
+
+MODULE_PARM_DESC(pcie_lane_cap, "PCIE Lane Caps (0: autodetect (default))");
+module_param_named(pcie_lane_cap, amdgpu_pcie_lane_cap, uint, 0444);
+
 static struct pci_device_id pciidlist[] = {
 #ifdef CONFIG_DRM_AMDGPU_CIK
 	/* Kaveri */
@@ -310,6 +318,14 @@ static int amdgpu_pci_probe(struct pci_dev *pdev,
 		return -ENODEV;
 	}
 
+	/*
+	 * Initialize amdkfd before starting radeon. If it was not loaded yet,
+	 * defer radeon probing
+	 */
+	ret = amdgpu_amdkfd_init();
+	if (ret == -EPROBE_DEFER)
+		return ret;
+
 	/* Get rid of things like offb */
 	ret = amdgpu_kick_out_firmware_fb(pdev);
 	if (ret)
@@ -539,6 +555,7 @@ static struct pci_driver amdgpu_kms_pci_driver = {
 
 static int __init amdgpu_init(void)
 {
+	amdgpu_sync_init();
 #ifdef CONFIG_VGA_CONSOLE
 	if (vgacon_text_force()) {
 		DRM_ERROR("VGACON disables amdgpu kernel modesetting.\n");
@@ -552,8 +569,6 @@ static int __init amdgpu_init(void)
 	driver->num_ioctls = amdgpu_max_kms_ioctl;
 	amdgpu_register_atpx_handler();
 
-	amdgpu_amdkfd_init();
-
 	/* let modprobe override vga console setting */
 	return drm_pci_init(driver, pdriver);
 }
@@ -563,6 +578,7 @@ static void __exit amdgpu_exit(void)
 	amdgpu_amdkfd_fini();
 	drm_pci_exit(driver, pdriver);
 	amdgpu_unregister_atpx_handler();
+	amdgpu_sync_fini();
 }
 
 module_init(amdgpu_init);
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_fence.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_fence.c
index 97db196dc6f8..4303b447efe8 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_fence.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_fence.c
@@ -47,9 +47,30 @@
  * that the the relevant GPU caches have been flushed.
  */
 
+struct amdgpu_fence {
+	struct fence base;
+
+	/* RB, DMA, etc. */
+	struct amdgpu_ring		*ring;
+};
+
 static struct kmem_cache *amdgpu_fence_slab;
 static atomic_t amdgpu_fence_slab_ref = ATOMIC_INIT(0);
 
+/*
+ * Cast helper
+ */
+static const struct fence_ops amdgpu_fence_ops;
+static inline struct amdgpu_fence *to_amdgpu_fence(struct fence *f)
+{
+	struct amdgpu_fence *__f = container_of(f, struct amdgpu_fence, base);
+
+	if (__f->base.ops == &amdgpu_fence_ops)
+		return __f;
+
+	return NULL;
+}
+
 /**
  * amdgpu_fence_write - write a fence value
  *
@@ -82,7 +103,7 @@ static u32 amdgpu_fence_read(struct amdgpu_ring *ring)
 	if (drv->cpu_addr)
 		seq = le32_to_cpu(*drv->cpu_addr);
 	else
-		seq = lower_32_bits(atomic64_read(&drv->last_seq));
+		seq = atomic_read(&drv->last_seq);
 
 	return seq;
 }
@@ -91,32 +112,41 @@ static u32 amdgpu_fence_read(struct amdgpu_ring *ring)
  * amdgpu_fence_emit - emit a fence on the requested ring
  *
  * @ring: ring the fence is associated with
- * @owner: creator of the fence
- * @fence: amdgpu fence object
+ * @f: resulting fence object
  *
  * Emits a fence command on the requested ring (all asics).
  * Returns 0 on success, -ENOMEM on failure.
  */
-int amdgpu_fence_emit(struct amdgpu_ring *ring, void *owner,
-		      struct amdgpu_fence **fence)
+int amdgpu_fence_emit(struct amdgpu_ring *ring, struct fence **f)
 {
 	struct amdgpu_device *adev = ring->adev;
+	struct amdgpu_fence *fence;
+	struct fence **ptr;
+	uint32_t seq;
 
-	/* we are protected by the ring emission mutex */
-	*fence = kmem_cache_alloc(amdgpu_fence_slab, GFP_KERNEL);
-	if ((*fence) == NULL) {
+	fence = kmem_cache_alloc(amdgpu_fence_slab, GFP_KERNEL);
+	if (fence == NULL)
 		return -ENOMEM;
-	}
-	(*fence)->seq = ++ring->fence_drv.sync_seq;
-	(*fence)->ring = ring;
-	(*fence)->owner = owner;
-	fence_init(&(*fence)->base, &amdgpu_fence_ops,
-		&ring->fence_drv.fence_queue.lock,
-		adev->fence_context + ring->idx,
-		(*fence)->seq);
+
+	seq = ++ring->fence_drv.sync_seq;
+	fence->ring = ring;
+	fence_init(&fence->base, &amdgpu_fence_ops,
+		   &ring->fence_drv.lock,
+		   adev->fence_context + ring->idx,
+		   seq);
 	amdgpu_ring_emit_fence(ring, ring->fence_drv.gpu_addr,
-			       (*fence)->seq,
-			       AMDGPU_FENCE_FLAG_INT);
+			       seq, AMDGPU_FENCE_FLAG_INT);
+
+	ptr = &ring->fence_drv.fences[seq & ring->fence_drv.num_fences_mask];
+	/* This function can't be called concurrently anyway, otherwise
+	 * emitting the fence would mess up the hardware ring buffer.
+	 */
+	BUG_ON(rcu_dereference_protected(*ptr, 1));
+
+	rcu_assign_pointer(*ptr, fence_get(&fence->base));
+
+	*f = &fence->base;
+
 	return 0;
 }
 
@@ -134,89 +164,48 @@ static void amdgpu_fence_schedule_fallback(struct amdgpu_ring *ring)
 }
 
 /**
- * amdgpu_fence_activity - check for fence activity
+ * amdgpu_fence_process - check for fence activity
  *
  * @ring: pointer to struct amdgpu_ring
  *
  * Checks the current fence value and calculates the last
- * signalled fence value. Returns true if activity occured
- * on the ring, and the fence_queue should be waken up.
+ * signalled fence value. Wakes the fence queue if the
+ * sequence number has increased.
  */
-static bool amdgpu_fence_activity(struct amdgpu_ring *ring)
+void amdgpu_fence_process(struct amdgpu_ring *ring)
 {
-	uint64_t seq, last_seq, last_emitted;
-	unsigned count_loop = 0;
-	bool wake = false;
-
-	/* Note there is a scenario here for an infinite loop but it's
-	 * very unlikely to happen. For it to happen, the current polling
-	 * process need to be interrupted by another process and another
-	 * process needs to update the last_seq btw the atomic read and
-	 * xchg of the current process.
-	 *
-	 * More over for this to go in infinite loop there need to be
-	 * continuously new fence signaled ie amdgpu_fence_read needs
-	 * to return a different value each time for both the currently
-	 * polling process and the other process that xchg the last_seq
-	 * btw atomic read and xchg of the current process. And the
-	 * value the other process set as last seq must be higher than
-	 * the seq value we just read. Which means that current process
-	 * need to be interrupted after amdgpu_fence_read and before
-	 * atomic xchg.
-	 *
-	 * To be even more safe we count the number of time we loop and
-	 * we bail after 10 loop just accepting the fact that we might
-	 * have temporarly set the last_seq not to the true real last
-	 * seq but to an older one.
-	 */
-	last_seq = atomic64_read(&ring->fence_drv.last_seq);
+	struct amdgpu_fence_driver *drv = &ring->fence_drv;
+	uint32_t seq, last_seq;
+	int r;
+
 	do {
-		last_emitted = ring->fence_drv.sync_seq;
+		last_seq = atomic_read(&ring->fence_drv.last_seq);
 		seq = amdgpu_fence_read(ring);
-		seq |= last_seq & 0xffffffff00000000LL;
-		if (seq < last_seq) {
-			seq &= 0xffffffff;
-			seq |= last_emitted & 0xffffffff00000000LL;
-		}
 
-		if (seq <= last_seq || seq > last_emitted) {
-			break;
-		}
-		/* If we loop over we don't want to return without
-		 * checking if a fence is signaled as it means that the
-		 * seq we just read is different from the previous on.
-		 */
-		wake = true;
-		last_seq = seq;
-		if ((count_loop++) > 10) {
-			/* We looped over too many time leave with the
-			 * fact that we might have set an older fence
-			 * seq then the current real last seq as signaled
-			 * by the hw.
-			 */
-			break;
-		}
-	} while (atomic64_xchg(&ring->fence_drv.last_seq, seq) > seq);
+	} while (atomic_cmpxchg(&drv->last_seq, last_seq, seq) != last_seq);
 
-	if (seq < last_emitted)
+	if (seq != ring->fence_drv.sync_seq)
 		amdgpu_fence_schedule_fallback(ring);
 
-	return wake;
-}
+	while (last_seq != seq) {
+		struct fence *fence, **ptr;
 
-/**
- * amdgpu_fence_process - process a fence
- *
- * @adev: amdgpu_device pointer
- * @ring: ring index the fence is associated with
- *
- * Checks the current fence value and wakes the fence queue
- * if the sequence number has increased (all asics).
- */
-void amdgpu_fence_process(struct amdgpu_ring *ring)
-{
-	if (amdgpu_fence_activity(ring))
-		wake_up_all(&ring->fence_drv.fence_queue);
+		ptr = &drv->fences[++last_seq & drv->num_fences_mask];
+
+		/* There is always exactly one thread signaling this fence slot */
+		fence = rcu_dereference_protected(*ptr, 1);
+		rcu_assign_pointer(*ptr, NULL);
+
+		BUG_ON(!fence);
+
+		r = fence_signal(fence);
+		if (!r)
+			FENCE_TRACE(fence, "signaled from irq context\n");
+		else
+			BUG();
+
+		fence_put(fence);
+	}
 }
 
 /**
@@ -234,77 +223,6 @@ static void amdgpu_fence_fallback(unsigned long arg)
 }
 
 /**
- * amdgpu_fence_seq_signaled - check if a fence sequence number has signaled
- *
- * @ring: ring the fence is associated with
- * @seq: sequence number
- *
- * Check if the last signaled fence sequnce number is >= the requested
- * sequence number (all asics).
- * Returns true if the fence has signaled (current fence value
- * is >= requested value) or false if it has not (current fence
- * value is < the requested value.  Helper function for
- * amdgpu_fence_signaled().
- */
-static bool amdgpu_fence_seq_signaled(struct amdgpu_ring *ring, u64 seq)
-{
-	if (atomic64_read(&ring->fence_drv.last_seq) >= seq)
-		return true;
-
-	/* poll new last sequence at least once */
-	amdgpu_fence_process(ring);
-	if (atomic64_read(&ring->fence_drv.last_seq) >= seq)
-		return true;
-
-	return false;
-}
-
-/*
- * amdgpu_ring_wait_seq - wait for seq of the specific ring to signal
- * @ring: ring to wait on for the seq number
- * @seq: seq number wait for
- *
- * return value:
- * 0: seq signaled, and gpu not hang
- * -EINVAL: some paramter is not valid
- */
-static int amdgpu_fence_ring_wait_seq(struct amdgpu_ring *ring, uint64_t seq)
-{
-	BUG_ON(!ring);
-	if (seq > ring->fence_drv.sync_seq)
-		return -EINVAL;
-
-	if (atomic64_read(&ring->fence_drv.last_seq) >= seq)
-		return 0;
-
-	amdgpu_fence_schedule_fallback(ring);
-	wait_event(ring->fence_drv.fence_queue,
-		   amdgpu_fence_seq_signaled(ring, seq));
-
-	return 0;
-}
-
-/**
- * amdgpu_fence_wait_next - wait for the next fence to signal
- *
- * @adev: amdgpu device pointer
- * @ring: ring index the fence is associated with
- *
- * Wait for the next fence on the requested ring to signal (all asics).
- * Returns 0 if the next fence has passed, error for all other cases.
- * Caller must hold ring lock.
- */
-int amdgpu_fence_wait_next(struct amdgpu_ring *ring)
-{
-	uint64_t seq = atomic64_read(&ring->fence_drv.last_seq) + 1ULL;
-
-	if (seq >= ring->fence_drv.sync_seq)
-		return -ENOENT;
-
-	return amdgpu_fence_ring_wait_seq(ring, seq);
-}
-
-/**
  * amdgpu_fence_wait_empty - wait for all fences to signal
  *
  * @adev: amdgpu device pointer
@@ -312,16 +230,28 @@ int amdgpu_fence_wait_next(struct amdgpu_ring *ring)
  *
  * Wait for all fences on the requested ring to signal (all asics).
  * Returns 0 if the fences have passed, error for all other cases.
- * Caller must hold ring lock.
  */
 int amdgpu_fence_wait_empty(struct amdgpu_ring *ring)
 {
-	uint64_t seq = ring->fence_drv.sync_seq;
+	uint64_t seq = ACCESS_ONCE(ring->fence_drv.sync_seq);
+	struct fence *fence, **ptr;
+	int r;
 
 	if (!seq)
 		return 0;
 
-	return amdgpu_fence_ring_wait_seq(ring, seq);
+	ptr = &ring->fence_drv.fences[seq & ring->fence_drv.num_fences_mask];
+	rcu_read_lock();
+	fence = rcu_dereference(*ptr);
+	if (!fence || !fence_get_rcu(fence)) {
+		rcu_read_unlock();
+		return 0;
+	}
+	rcu_read_unlock();
+
+	r = fence_wait(fence, false);
+	fence_put(fence);
+	return r;
 }
 
 /**
@@ -341,13 +271,10 @@ unsigned amdgpu_fence_count_emitted(struct amdgpu_ring *ring)
 	 * but it's ok to report slightly wrong fence count here.
 	 */
 	amdgpu_fence_process(ring);
-	emitted = ring->fence_drv.sync_seq
-		- atomic64_read(&ring->fence_drv.last_seq);
-	/* to avoid 32bits warp around */
-	if (emitted > 0x10000000)
-		emitted = 0x10000000;
-
-	return (unsigned)emitted;
+	emitted = 0x100000000ull;
+	emitted -= atomic_read(&ring->fence_drv.last_seq);
+	emitted += ACCESS_ONCE(ring->fence_drv.sync_seq);
+	return lower_32_bits(emitted);
 }
 
 /**
@@ -379,7 +306,7 @@ int amdgpu_fence_driver_start_ring(struct amdgpu_ring *ring,
 		ring->fence_drv.cpu_addr = adev->uvd.cpu_addr + index;
 		ring->fence_drv.gpu_addr = adev->uvd.gpu_addr + index;
 	}
-	amdgpu_fence_write(ring, atomic64_read(&ring->fence_drv.last_seq));
+	amdgpu_fence_write(ring, atomic_read(&ring->fence_drv.last_seq));
 	amdgpu_irq_get(adev, irq_src, irq_type);
 
 	ring->fence_drv.irq_src = irq_src;
@@ -397,25 +324,36 @@ int amdgpu_fence_driver_start_ring(struct amdgpu_ring *ring,
  * for the requested ring.
  *
  * @ring: ring to init the fence driver on
+ * @num_hw_submission: number of entries on the hardware queue
  *
  * Init the fence driver for the requested ring (all asics).
  * Helper function for amdgpu_fence_driver_init().
  */
-int amdgpu_fence_driver_init_ring(struct amdgpu_ring *ring)
+int amdgpu_fence_driver_init_ring(struct amdgpu_ring *ring,
+				  unsigned num_hw_submission)
 {
 	long timeout;
 	int r;
 
+	/* Check that num_hw_submission is a power of two */
+	if ((num_hw_submission & (num_hw_submission - 1)) != 0)
+		return -EINVAL;
+
 	ring->fence_drv.cpu_addr = NULL;
 	ring->fence_drv.gpu_addr = 0;
 	ring->fence_drv.sync_seq = 0;
-	atomic64_set(&ring->fence_drv.last_seq, 0);
+	atomic_set(&ring->fence_drv.last_seq, 0);
 	ring->fence_drv.initialized = false;
 
 	setup_timer(&ring->fence_drv.fallback_timer, amdgpu_fence_fallback,
 		    (unsigned long)ring);
 
-	init_waitqueue_head(&ring->fence_drv.fence_queue);
+	ring->fence_drv.num_fences_mask = num_hw_submission - 1;
+	spin_lock_init(&ring->fence_drv.lock);
+	ring->fence_drv.fences = kcalloc(num_hw_submission, sizeof(void *),
+					 GFP_KERNEL);
+	if (!ring->fence_drv.fences)
+		return -ENOMEM;
 
 	timeout = msecs_to_jiffies(amdgpu_lockup_timeout);
 	if (timeout == 0) {
@@ -429,7 +367,7 @@ int amdgpu_fence_driver_init_ring(struct amdgpu_ring *ring)
 		timeout = MAX_SCHEDULE_TIMEOUT;
 	}
 	r = amd_sched_init(&ring->sched, &amdgpu_sched_ops,
-			   amdgpu_sched_hw_submission,
+			   num_hw_submission,
 			   timeout, ring->name);
 	if (r) {
 		DRM_ERROR("Failed to create scheduler on ring %s.\n",
@@ -477,10 +415,9 @@ int amdgpu_fence_driver_init(struct amdgpu_device *adev)
  */
 void amdgpu_fence_driver_fini(struct amdgpu_device *adev)
 {
-	int i, r;
+	unsigned i, j;
+	int r;
 
-	if (atomic_dec_and_test(&amdgpu_fence_slab_ref))
-		kmem_cache_destroy(amdgpu_fence_slab);
 	for (i = 0; i < AMDGPU_MAX_RINGS; i++) {
 		struct amdgpu_ring *ring = adev->rings[i];
 
@@ -491,13 +428,18 @@ void amdgpu_fence_driver_fini(struct amdgpu_device *adev)
 			/* no need to trigger GPU reset as we are unloading */
 			amdgpu_fence_driver_force_completion(adev);
 		}
-		wake_up_all(&ring->fence_drv.fence_queue);
 		amdgpu_irq_put(adev, ring->fence_drv.irq_src,
 			       ring->fence_drv.irq_type);
 		amd_sched_fini(&ring->sched);
 		del_timer_sync(&ring->fence_drv.fallback_timer);
+		for (j = 0; j <= ring->fence_drv.num_fences_mask; ++j)
+			fence_put(ring->fence_drv.fences[i]);
+		kfree(ring->fence_drv.fences);
 		ring->fence_drv.initialized = false;
 	}
+
+	if (atomic_dec_and_test(&amdgpu_fence_slab_ref))
+		kmem_cache_destroy(amdgpu_fence_slab);
 }
 
 /**
@@ -594,103 +536,57 @@ static const char *amdgpu_fence_get_timeline_name(struct fence *f)
 }
 
 /**
- * amdgpu_fence_is_signaled - test if fence is signaled
- *
- * @f: fence to test
+ * amdgpu_fence_enable_signaling - enable signalling on fence
+ * @fence: fence
  *
- * Test the fence sequence number if it is already signaled. If it isn't
- * signaled start fence processing. Returns True if the fence is signaled.
+ * This function is called with fence_queue lock held, and adds a callback
+ * to fence_queue that checks if this fence is signaled, and if so it
+ * signals the fence and removes itself.
  */
-static bool amdgpu_fence_is_signaled(struct fence *f)
+static bool amdgpu_fence_enable_signaling(struct fence *f)
 {
 	struct amdgpu_fence *fence = to_amdgpu_fence(f);
 	struct amdgpu_ring *ring = fence->ring;
 
-	if (atomic64_read(&ring->fence_drv.last_seq) >= fence->seq)
-		return true;
-
-	amdgpu_fence_process(ring);
+	if (!timer_pending(&ring->fence_drv.fallback_timer))
+		amdgpu_fence_schedule_fallback(ring);
 
-	if (atomic64_read(&ring->fence_drv.last_seq) >= fence->seq)
-		return true;
+	FENCE_TRACE(&fence->base, "armed on ring %i!\n", ring->idx);
 
-	return false;
+	return true;
 }
 
 /**
- * amdgpu_fence_check_signaled - callback from fence_queue
+ * amdgpu_fence_free - free up the fence memory
+ *
+ * @rcu: RCU callback head
  *
- * this function is called with fence_queue lock held, which is also used
- * for the fence locking itself, so unlocked variants are used for
- * fence_signal, and remove_wait_queue.
+ * Free up the fence memory after the RCU grace period.
  */
-static int amdgpu_fence_check_signaled(wait_queue_t *wait, unsigned mode, int flags, void *key)
+static void amdgpu_fence_free(struct rcu_head *rcu)
 {
-	struct amdgpu_fence *fence;
-	struct amdgpu_device *adev;
-	u64 seq;
-	int ret;
-
-	fence = container_of(wait, struct amdgpu_fence, fence_wake);
-	adev = fence->ring->adev;
-
-	/*
-	 * We cannot use amdgpu_fence_process here because we're already
-	 * in the waitqueue, in a call from wake_up_all.
-	 */
-	seq = atomic64_read(&fence->ring->fence_drv.last_seq);
-	if (seq >= fence->seq) {
-		ret = fence_signal_locked(&fence->base);
-		if (!ret)
-			FENCE_TRACE(&fence->base, "signaled from irq context\n");
-		else
-			FENCE_TRACE(&fence->base, "was already signaled\n");
-
-		__remove_wait_queue(&fence->ring->fence_drv.fence_queue, &fence->fence_wake);
-		fence_put(&fence->base);
-	} else
-		FENCE_TRACE(&fence->base, "pending\n");
-	return 0;
+	struct fence *f = container_of(rcu, struct fence, rcu);
+	struct amdgpu_fence *fence = to_amdgpu_fence(f);
+	kmem_cache_free(amdgpu_fence_slab, fence);
 }
 
 /**
- * amdgpu_fence_enable_signaling - enable signalling on fence
+ * amdgpu_fence_release - callback that fence can be freed
+ *
  * @fence: fence
  *
- * This function is called with fence_queue lock held, and adds a callback
- * to fence_queue that checks if this fence is signaled, and if so it
- * signals the fence and removes itself.
+ * This function is called when the reference count becomes zero.
+ * It just RCU schedules freeing up the fence.
  */
-static bool amdgpu_fence_enable_signaling(struct fence *f)
-{
-	struct amdgpu_fence *fence = to_amdgpu_fence(f);
-	struct amdgpu_ring *ring = fence->ring;
-
-	if (atomic64_read(&ring->fence_drv.last_seq) >= fence->seq)
-		return false;
-
-	fence->fence_wake.flags = 0;
-	fence->fence_wake.private = NULL;
-	fence->fence_wake.func = amdgpu_fence_check_signaled;
-	__add_wait_queue(&ring->fence_drv.fence_queue, &fence->fence_wake);
-	fence_get(f);
-	if (!timer_pending(&ring->fence_drv.fallback_timer))
-		amdgpu_fence_schedule_fallback(ring);
-	FENCE_TRACE(&fence->base, "armed on ring %i!\n", ring->idx);
-	return true;
-}
-
 static void amdgpu_fence_release(struct fence *f)
 {
-	struct amdgpu_fence *fence = to_amdgpu_fence(f);
-	kmem_cache_free(amdgpu_fence_slab, fence);
+	call_rcu(&f->rcu, amdgpu_fence_free);
 }
 
-const struct fence_ops amdgpu_fence_ops = {
+static const struct fence_ops amdgpu_fence_ops = {
 	.get_driver_name = amdgpu_fence_get_driver_name,
 	.get_timeline_name = amdgpu_fence_get_timeline_name,
 	.enable_signaling = amdgpu_fence_enable_signaling,
-	.signaled = amdgpu_fence_is_signaled,
 	.wait = fence_default_wait,
 	.release = amdgpu_fence_release,
 };
@@ -714,9 +610,9 @@ static int amdgpu_debugfs_fence_info(struct seq_file *m, void *data)
 		amdgpu_fence_process(ring);
 
 		seq_printf(m, "--- ring %d (%s) ---\n", i, ring->name);
-		seq_printf(m, "Last signaled fence 0x%016llx\n",
-			   (unsigned long long)atomic64_read(&ring->fence_drv.last_seq));
-		seq_printf(m, "Last emitted        0x%016llx\n",
+		seq_printf(m, "Last signaled fence 0x%08x\n",
+			   atomic_read(&ring->fence_drv.last_seq));
+		seq_printf(m, "Last emitted        0x%08x\n",
 			   ring->fence_drv.sync_seq);
 	}
 	return 0;
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c
index 2e26a517f2d6..fa6a27bff298 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c
@@ -26,6 +26,7 @@
  *          Jerome Glisse
  */
 #include <linux/ktime.h>
+#include <linux/pagemap.h>
 #include <drm/drmP.h>
 #include <drm/amdgpu_drm.h>
 #include "amdgpu.h"
@@ -140,25 +141,40 @@ int amdgpu_gem_object_open(struct drm_gem_object *obj, struct drm_file *file_pri
 void amdgpu_gem_object_close(struct drm_gem_object *obj,
 			     struct drm_file *file_priv)
 {
-	struct amdgpu_bo *rbo = gem_to_amdgpu_bo(obj);
-	struct amdgpu_device *adev = rbo->adev;
+	struct amdgpu_bo *bo = gem_to_amdgpu_bo(obj);
+	struct amdgpu_device *adev = bo->adev;
 	struct amdgpu_fpriv *fpriv = file_priv->driver_priv;
 	struct amdgpu_vm *vm = &fpriv->vm;
+
+	struct amdgpu_bo_list_entry vm_pd;
+	struct list_head list, duplicates;
+	struct ttm_validate_buffer tv;
+	struct ww_acquire_ctx ticket;
 	struct amdgpu_bo_va *bo_va;
 	int r;
-	r = amdgpu_bo_reserve(rbo, true);
+
+	INIT_LIST_HEAD(&list);
+	INIT_LIST_HEAD(&duplicates);
+
+	tv.bo = &bo->tbo;
+	tv.shared = true;
+	list_add(&tv.head, &list);
+
+	amdgpu_vm_get_pd_bo(vm, &list, &vm_pd);
+
+	r = ttm_eu_reserve_buffers(&ticket, &list, false, &duplicates);
 	if (r) {
 		dev_err(adev->dev, "leaking bo va because "
 			"we fail to reserve bo (%d)\n", r);
 		return;
 	}
-	bo_va = amdgpu_vm_bo_find(vm, rbo);
+	bo_va = amdgpu_vm_bo_find(vm, bo);
 	if (bo_va) {
 		if (--bo_va->ref_count == 0) {
 			amdgpu_vm_bo_rmv(adev, bo_va);
 		}
 	}
-	amdgpu_bo_unreserve(rbo);
+	ttm_eu_backoff_reservation(&ticket, &list);
 }
 
 static int amdgpu_gem_handle_lockup(struct amdgpu_device *adev, int r)
@@ -243,12 +259,10 @@ int amdgpu_gem_userptr_ioctl(struct drm_device *dev, void *data,
 	    AMDGPU_GEM_USERPTR_REGISTER))
 		return -EINVAL;
 
-	if (!(args->flags & AMDGPU_GEM_USERPTR_READONLY) && (
-	     !(args->flags & AMDGPU_GEM_USERPTR_ANONONLY) ||
-	     !(args->flags & AMDGPU_GEM_USERPTR_REGISTER))) {
+	if (!(args->flags & AMDGPU_GEM_USERPTR_READONLY) &&
+	     !(args->flags & AMDGPU_GEM_USERPTR_REGISTER)) {
 
-		/* if we want to write to it we must require anonymous
-		   memory and install a MMU notifier */
+		/* if we want to write to it we must install a MMU notifier */
 		return -EACCES;
 	}
 
@@ -274,18 +288,23 @@ int amdgpu_gem_userptr_ioctl(struct drm_device *dev, void *data,
 
 	if (args->flags & AMDGPU_GEM_USERPTR_VALIDATE) {
 		down_read(&current->mm->mmap_sem);
+
+		r = amdgpu_ttm_tt_get_user_pages(bo->tbo.ttm,
+						 bo->tbo.ttm->pages);
+		if (r)
+			goto unlock_mmap_sem;
+
 		r = amdgpu_bo_reserve(bo, true);
-		if (r) {
-			up_read(&current->mm->mmap_sem);
-			goto release_object;
-		}
+		if (r)
+			goto free_pages;
 
 		amdgpu_ttm_placement_from_domain(bo, AMDGPU_GEM_DOMAIN_GTT);
 		r = ttm_bo_validate(&bo->tbo, &bo->placement, true, false);
 		amdgpu_bo_unreserve(bo);
-		up_read(&current->mm->mmap_sem);
 		if (r)
-			goto release_object;
+			goto free_pages;
+
+		up_read(&current->mm->mmap_sem);
 	}
 
 	r = drm_gem_handle_create(filp, gobj, &handle);
@@ -297,6 +316,12 @@ int amdgpu_gem_userptr_ioctl(struct drm_device *dev, void *data,
 	args->handle = handle;
 	return 0;
 
+free_pages:
+	release_pages(bo->tbo.ttm->pages, bo->tbo.ttm->num_pages, false);
+
+unlock_mmap_sem:
+	up_read(&current->mm->mmap_sem);
+
 release_object:
 	drm_gem_object_unreference_unlocked(gobj);
 
@@ -569,11 +594,10 @@ int amdgpu_gem_va_ioctl(struct drm_device *dev, void *data,
 	tv.shared = true;
 	list_add(&tv.head, &list);
 
-	if (args->operation == AMDGPU_VA_OP_MAP) {
-		tv_pd.bo = &fpriv->vm.page_directory->tbo;
-		tv_pd.shared = true;
-		list_add(&tv_pd.head, &list);
-	}
+	tv_pd.bo = &fpriv->vm.page_directory->tbo;
+	tv_pd.shared = true;
+	list_add(&tv_pd.head, &list);
+
 	r = ttm_eu_reserve_buffers(&ticket, &list, true, &duplicates);
 	if (r) {
 		drm_gem_object_unreference_unlocked(gobj);
@@ -606,7 +630,8 @@ int amdgpu_gem_va_ioctl(struct drm_device *dev, void *data,
 		break;
 	}
 	ttm_eu_backoff_reservation(&ticket, &list);
-	if (!r && !(args->flags & AMDGPU_VM_DELAY_UPDATE))
+	if (!r && !(args->flags & AMDGPU_VM_DELAY_UPDATE) &&
+	    !amdgpu_vm_debug)
 		amdgpu_gem_va_update_vm(adev, bo_va, args->operation);
 
 	drm_gem_object_unreference_unlocked(gobj);
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ib.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ib.c
index b5bdd5d59b58..8443cea6821a 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ib.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ib.c
@@ -75,6 +75,7 @@ int amdgpu_ib_get(struct amdgpu_device *adev, struct amdgpu_vm *vm,
 	}
 
 	ib->vm = vm;
+	ib->vm_id = 0;
 
 	return 0;
 }
@@ -84,14 +85,13 @@ int amdgpu_ib_get(struct amdgpu_device *adev, struct amdgpu_vm *vm,
  *
  * @adev: amdgpu_device pointer
  * @ib: IB object to free
+ * @f: the fence SA bo need wait on for the ib alloation
  *
  * Free an IB (all asics).
  */
-void amdgpu_ib_free(struct amdgpu_device *adev, struct amdgpu_ib *ib)
+void amdgpu_ib_free(struct amdgpu_device *adev, struct amdgpu_ib *ib, struct fence *f)
 {
-	amdgpu_sa_bo_free(adev, &ib->sa_bo, &ib->fence->base);
-	if (ib->fence)
-		fence_put(&ib->fence->base);
+	amdgpu_sa_bo_free(adev, &ib->sa_bo, f);
 }
 
 /**
@@ -100,7 +100,6 @@ void amdgpu_ib_free(struct amdgpu_device *adev, struct amdgpu_ib *ib)
  * @adev: amdgpu_device pointer
  * @num_ibs: number of IBs to schedule
  * @ibs: IB objects to schedule
- * @owner: owner for creating the fences
  * @f: fence created during this submission
  *
  * Schedule an IB on the associated ring (all asics).
@@ -117,14 +116,14 @@ void amdgpu_ib_free(struct amdgpu_device *adev, struct amdgpu_ib *ib)
  * to SI there was just a DE IB.
  */
 int amdgpu_ib_schedule(struct amdgpu_ring *ring, unsigned num_ibs,
-		       struct amdgpu_ib *ibs, void *owner,
-		       struct fence *last_vm_update,
+		       struct amdgpu_ib *ibs, struct fence *last_vm_update,
 		       struct fence **f)
 {
 	struct amdgpu_device *adev = ring->adev;
 	struct amdgpu_ib *ib = &ibs[0];
 	struct amdgpu_ctx *ctx, *old_ctx;
 	struct amdgpu_vm *vm;
+	struct fence *hwf;
 	unsigned i;
 	int r = 0;
 
@@ -139,7 +138,7 @@ int amdgpu_ib_schedule(struct amdgpu_ring *ring, unsigned num_ibs,
 		return -EINVAL;
 	}
 
-	if (vm && !ibs->grabbed_vmid) {
+	if (vm && !ibs->vm_id) {
 		dev_err(adev->dev, "VM IB without ID\n");
 		return -EINVAL;
 	}
@@ -152,13 +151,10 @@ int amdgpu_ib_schedule(struct amdgpu_ring *ring, unsigned num_ibs,
 
 	if (vm) {
 		/* do context switch */
-		amdgpu_vm_flush(ring, vm, last_vm_update);
-
-		if (ring->funcs->emit_gds_switch)
-			amdgpu_ring_emit_gds_switch(ring, ib->vm->ids[ring->idx].id,
-						    ib->gds_base, ib->gds_size,
-						    ib->gws_base, ib->gws_size,
-						    ib->oa_base, ib->oa_size);
+		amdgpu_vm_flush(ring, ib->vm_id, ib->vm_pd_addr,
+				ib->gds_base, ib->gds_size,
+				ib->gws_base, ib->gws_size,
+				ib->oa_base, ib->oa_size);
 
 		if (ring->funcs->emit_hdp_flush)
 			amdgpu_ring_emit_hdp_flush(ring);
@@ -170,6 +166,8 @@ int amdgpu_ib_schedule(struct amdgpu_ring *ring, unsigned num_ibs,
 
 		if (ib->ctx != ctx || ib->vm != vm) {
 			ring->current_ctx = old_ctx;
+			if (ib->vm_id)
+				amdgpu_vm_reset_id(adev, ib->vm_id);
 			amdgpu_ring_undo(ring);
 			return -EINVAL;
 		}
@@ -177,10 +175,17 @@ int amdgpu_ib_schedule(struct amdgpu_ring *ring, unsigned num_ibs,
 		ring->current_ctx = ctx;
 	}
 
-	r = amdgpu_fence_emit(ring, owner, &ib->fence);
+	if (vm) {
+		if (ring->funcs->emit_hdp_invalidate)
+			amdgpu_ring_emit_hdp_invalidate(ring);
+	}
+
+	r = amdgpu_fence_emit(ring, &hwf);
 	if (r) {
 		dev_err(adev->dev, "failed to emit fence (%d)\n", r);
 		ring->current_ctx = old_ctx;
+		if (ib->vm_id)
+			amdgpu_vm_reset_id(adev, ib->vm_id);
 		amdgpu_ring_undo(ring);
 		return r;
 	}
@@ -194,7 +199,7 @@ int amdgpu_ib_schedule(struct amdgpu_ring *ring, unsigned num_ibs,
 	}
 
 	if (f)
-		*f = fence_get(&ib->fence->base);
+		*f = fence_get(hwf);
 
 	amdgpu_ring_commit(ring);
 	return 0;
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_job.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_job.c
index f29bbb96a881..9c9b19e2f353 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_job.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_job.c
@@ -70,9 +70,13 @@ int amdgpu_job_alloc_with_ib(struct amdgpu_device *adev, unsigned size,
 void amdgpu_job_free(struct amdgpu_job *job)
 {
 	unsigned i;
+	struct fence *f;
+	/* use sched fence if available */
+	f = (job->base.s_fence)? &job->base.s_fence->base : job->fence;
 
 	for (i = 0; i < job->num_ibs; ++i)
-		amdgpu_ib_free(job->adev, &job->ibs[i]);
+		amdgpu_sa_bo_free(job->adev, &job->ibs[i].sa_bo, f);
+	fence_put(job->fence);
 
 	amdgpu_bo_unref(&job->uf.bo);
 	amdgpu_sync_free(&job->sync);
@@ -105,16 +109,23 @@ static struct fence *amdgpu_job_dependency(struct amd_sched_job *sched_job)
 
 	struct fence *fence = amdgpu_sync_get_fence(&job->sync);
 
-	if (fence == NULL && vm && !job->ibs->grabbed_vmid) {
+	if (fence == NULL && vm && !job->ibs->vm_id) {
 		struct amdgpu_ring *ring = job->ring;
+		unsigned i, vm_id;
+		uint64_t vm_pd_addr;
 		int r;
 
 		r = amdgpu_vm_grab_id(vm, ring, &job->sync,
-				      &job->base.s_fence->base);
+				      &job->base.s_fence->base,
+				      &vm_id, &vm_pd_addr);
 		if (r)
 			DRM_ERROR("Error getting VM ID (%d)\n", r);
-		else
-			job->ibs->grabbed_vmid = true;
+		else {
+			for (i = 0; i < job->num_ibs; ++i) {
+				job->ibs[i].vm_id = vm_id;
+				job->ibs[i].vm_pd_addr = vm_pd_addr;
+			}
+		}
 
 		fence = amdgpu_sync_get_fence(&job->sync);
 	}
@@ -141,7 +152,7 @@ static struct fence *amdgpu_job_run(struct amd_sched_job *sched_job)
 	}
 
 	trace_amdgpu_sched_run_job(job);
-	r = amdgpu_ib_schedule(job->ring, job->num_ibs, job->ibs, job->owner,
+	r = amdgpu_ib_schedule(job->ring, job->num_ibs, job->ibs,
 			       job->sync.last_vm_update, &fence);
 	if (r) {
 		DRM_ERROR("Error scheduling IBs (%d)\n", r);
@@ -149,6 +160,7 @@ static struct fence *amdgpu_job_run(struct amd_sched_job *sched_job)
 	}
 
 err:
+	job->fence = fence;
 	amdgpu_job_free(job);
 	return fence;
 }
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_object.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_object.c
index 9a025a77958d..151a2d42c639 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_object.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_object.c
@@ -308,7 +308,7 @@ int amdgpu_bo_create(struct amdgpu_device *adev,
 int amdgpu_bo_kmap(struct amdgpu_bo *bo, void **ptr)
 {
 	bool is_iomem;
-	int r;
+	long r;
 
 	if (bo->flags & AMDGPU_GEM_CREATE_NO_CPU_ACCESS)
 		return -EPERM;
@@ -319,14 +319,20 @@ int amdgpu_bo_kmap(struct amdgpu_bo *bo, void **ptr)
 		}
 		return 0;
 	}
+
+	r = reservation_object_wait_timeout_rcu(bo->tbo.resv, false, false,
+						MAX_SCHEDULE_TIMEOUT);
+	if (r < 0)
+		return r;
+
 	r = ttm_bo_kmap(&bo->tbo, 0, bo->tbo.num_pages, &bo->kmap);
-	if (r) {
+	if (r)
 		return r;
-	}
+
 	bo->kptr = ttm_kmap_obj_virtual(&bo->kmap, &is_iomem);
-	if (ptr) {
+	if (ptr)
 		*ptr = bo->kptr;
-	}
+
 	return 0;
 }
 
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_pm.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_pm.c
index d77b2bdbe800..ff9597ce268c 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_pm.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_pm.c
@@ -113,6 +113,10 @@ static ssize_t amdgpu_get_dpm_forced_performance_level(struct device *dev,
 	struct drm_device *ddev = dev_get_drvdata(dev);
 	struct amdgpu_device *adev = ddev->dev_private;
 
+	if  ((adev->flags & AMD_IS_PX) &&
+	     (ddev->switch_power_state != DRM_SWITCH_POWER_ON))
+		return snprintf(buf, PAGE_SIZE, "off\n");
+
 	if (adev->pp_enabled) {
 		enum amd_dpm_forced_level level;
 
@@ -142,6 +146,11 @@ static ssize_t amdgpu_set_dpm_forced_performance_level(struct device *dev,
 	enum amdgpu_dpm_forced_level level;
 	int ret = 0;
 
+	/* Can't force performance level when the card is off */
+	if  ((adev->flags & AMD_IS_PX) &&
+	     (ddev->switch_power_state != DRM_SWITCH_POWER_ON))
+		return -EINVAL;
+
 	if (strncmp("low", buf, strlen("low")) == 0) {
 		level = AMDGPU_DPM_FORCED_LEVEL_LOW;
 	} else if (strncmp("high", buf, strlen("high")) == 0) {
@@ -161,6 +170,7 @@ static ssize_t amdgpu_set_dpm_forced_performance_level(struct device *dev,
 		mutex_lock(&adev->pm.mutex);
 		if (adev->pm.dpm.thermal_active) {
 			count = -EINVAL;
+			mutex_unlock(&adev->pm.mutex);
 			goto fail;
 		}
 		ret = amdgpu_dpm_force_performance_level(adev, level);
@@ -171,8 +181,6 @@ static ssize_t amdgpu_set_dpm_forced_performance_level(struct device *dev,
 		mutex_unlock(&adev->pm.mutex);
 	}
 fail:
-	mutex_unlock(&adev->pm.mutex);
-
 	return count;
 }
 
@@ -469,8 +477,14 @@ static ssize_t amdgpu_hwmon_show_temp(struct device *dev,
 				      char *buf)
 {
 	struct amdgpu_device *adev = dev_get_drvdata(dev);
+	struct drm_device *ddev = adev->ddev;
 	int temp;
 
+	/* Can't get temperature when the card is off */
+	if  ((adev->flags & AMD_IS_PX) &&
+	     (ddev->switch_power_state != DRM_SWITCH_POWER_ON))
+		return -EINVAL;
+
 	if (!adev->pp_enabled && !adev->pm.funcs->get_temperature)
 		temp = 0;
 	else
@@ -919,11 +933,6 @@ force:
 
 	/* update display watermarks based on new power state */
 	amdgpu_display_bandwidth_update(adev);
-	/* update displays */
-	amdgpu_dpm_display_configuration_changed(adev);
-
-	adev->pm.dpm.current_active_crtcs = adev->pm.dpm.new_active_crtcs;
-	adev->pm.dpm.current_active_crtc_count = adev->pm.dpm.new_active_crtc_count;
 
 	/* wait for the rings to drain */
 	for (i = 0; i < AMDGPU_MAX_RINGS; i++) {
@@ -940,6 +949,12 @@ force:
 
 	amdgpu_dpm_post_set_power_state(adev);
 
+	/* update displays */
+	amdgpu_dpm_display_configuration_changed(adev);
+
+	adev->pm.dpm.current_active_crtcs = adev->pm.dpm.new_active_crtcs;
+	adev->pm.dpm.current_active_crtc_count = adev->pm.dpm.new_active_crtc_count;
+
 	if (adev->pm.funcs->force_performance_level) {
 		if (adev->pm.dpm.thermal_active) {
 			enum amdgpu_dpm_forced_level level = adev->pm.dpm.forced_level;
@@ -1174,12 +1189,16 @@ static int amdgpu_debugfs_pm_info(struct seq_file *m, void *data)
 	struct drm_info_node *node = (struct drm_info_node *) m->private;
 	struct drm_device *dev = node->minor->dev;
 	struct amdgpu_device *adev = dev->dev_private;
+	struct drm_device *ddev = adev->ddev;
 
 	if (!adev->pm.dpm_enabled) {
 		seq_printf(m, "dpm not enabled\n");
 		return 0;
 	}
-	if (adev->pp_enabled) {
+	if  ((adev->flags & AMD_IS_PX) &&
+	     (ddev->switch_power_state != DRM_SWITCH_POWER_ON)) {
+		seq_printf(m, "PX asic powered off\n");
+	} else if (adev->pp_enabled) {
 		amdgpu_dpm_debugfs_print_current_performance_level(adev, m);
 	} else {
 		mutex_lock(&adev->pm.mutex);
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_powerplay.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_powerplay.c
index b9d0d55f6b47..3cb6d6c413c7 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_powerplay.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_powerplay.c
@@ -143,8 +143,10 @@ static int amdgpu_pp_late_init(void *handle)
 					adev->powerplay.pp_handle);
 
 #ifdef CONFIG_DRM_AMD_POWERPLAY
-	if (adev->pp_enabled)
+	if (adev->pp_enabled) {
 		amdgpu_pm_sysfs_init(adev);
+		amdgpu_dpm_dispatch_task(adev, AMD_PP_EVENT_COMPLETE_INIT, NULL, NULL);
+	}
 #endif
 	return ret;
 }
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.c
index 56c07e3fdb33..972eed2ef787 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ring.c
@@ -236,7 +236,8 @@ int amdgpu_ring_init(struct amdgpu_device *adev, struct amdgpu_ring *ring,
 		ring->adev = adev;
 		ring->idx = adev->num_rings++;
 		adev->rings[ring->idx] = ring;
-		r = amdgpu_fence_driver_init_ring(ring);
+		r = amdgpu_fence_driver_init_ring(ring,
+			amdgpu_sched_hw_submission);
 		if (r)
 			return r;
 	}
@@ -352,30 +353,6 @@ void amdgpu_ring_fini(struct amdgpu_ring *ring)
 	}
 }
 
-/**
- * amdgpu_ring_from_fence - get ring from fence
- *
- * @f: fence structure
- *
- * Extract the ring a fence belongs to. Handles both scheduler as
- * well as hardware fences.
- */
-struct amdgpu_ring *amdgpu_ring_from_fence(struct fence *f)
-{
-	struct amdgpu_fence *a_fence;
-	struct amd_sched_fence *s_fence;
-
-	s_fence = to_amd_sched_fence(f);
-	if (s_fence)
-		return container_of(s_fence->sched, struct amdgpu_ring, sched);
-
-	a_fence = to_amdgpu_fence(f);
-	if (a_fence)
-		return a_fence->ring;
-
-	return NULL;
-}
-
 /*
  * Debugfs info
  */
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_sa.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_sa.c
index 7d8f8f1e3f7f..8bf84efafb04 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_sa.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_sa.c
@@ -60,9 +60,8 @@ int amdgpu_sa_bo_manager_init(struct amdgpu_device *adev,
 	sa_manager->align = align;
 	sa_manager->hole = &sa_manager->olist;
 	INIT_LIST_HEAD(&sa_manager->olist);
-	for (i = 0; i < AMDGPU_MAX_RINGS; ++i) {
+	for (i = 0; i < AMDGPU_SA_NUM_FENCE_LISTS; ++i)
 		INIT_LIST_HEAD(&sa_manager->flist[i]);
-	}
 
 	r = amdgpu_bo_create(adev, size, align, true, domain,
 			     0, NULL, NULL, &sa_manager->bo);
@@ -228,11 +227,9 @@ static bool amdgpu_sa_event(struct amdgpu_sa_manager *sa_manager,
 	unsigned soffset, eoffset, wasted;
 	int i;
 
-	for (i = 0; i < AMDGPU_MAX_RINGS; ++i) {
-		if (!list_empty(&sa_manager->flist[i])) {
+	for (i = 0; i < AMDGPU_SA_NUM_FENCE_LISTS; ++i)
+		if (!list_empty(&sa_manager->flist[i]))
 			return true;
-		}
-	}
 
 	soffset = amdgpu_sa_bo_hole_soffset(sa_manager);
 	eoffset = amdgpu_sa_bo_hole_eoffset(sa_manager);
@@ -265,12 +262,11 @@ static bool amdgpu_sa_bo_next_hole(struct amdgpu_sa_manager *sa_manager,
 	/* go over all fence list and try to find the closest sa_bo
 	 * of the current last
 	 */
-	for (i = 0; i < AMDGPU_MAX_RINGS; ++i) {
+	for (i = 0; i < AMDGPU_SA_NUM_FENCE_LISTS; ++i) {
 		struct amdgpu_sa_bo *sa_bo;
 
-		if (list_empty(&sa_manager->flist[i])) {
+		if (list_empty(&sa_manager->flist[i]))
 			continue;
-		}
 
 		sa_bo = list_first_entry(&sa_manager->flist[i],
 					 struct amdgpu_sa_bo, flist);
@@ -299,7 +295,9 @@ static bool amdgpu_sa_bo_next_hole(struct amdgpu_sa_manager *sa_manager,
 	}
 
 	if (best_bo) {
-		uint32_t idx = amdgpu_ring_from_fence(best_bo->fence)->idx;
+		uint32_t idx = best_bo->fence->context;
+
+		idx %= AMDGPU_SA_NUM_FENCE_LISTS;
 		++tries[idx];
 		sa_manager->hole = best_bo->olist.prev;
 
@@ -315,8 +313,8 @@ int amdgpu_sa_bo_new(struct amdgpu_sa_manager *sa_manager,
 		     struct amdgpu_sa_bo **sa_bo,
 		     unsigned size, unsigned align)
 {
-	struct fence *fences[AMDGPU_MAX_RINGS];
-	unsigned tries[AMDGPU_MAX_RINGS];
+	struct fence *fences[AMDGPU_SA_NUM_FENCE_LISTS];
+	unsigned tries[AMDGPU_SA_NUM_FENCE_LISTS];
 	unsigned count;
 	int i, r;
 	signed long t;
@@ -338,7 +336,7 @@ int amdgpu_sa_bo_new(struct amdgpu_sa_manager *sa_manager,
 
 	spin_lock(&sa_manager->wq.lock);
 	do {
-		for (i = 0; i < AMDGPU_MAX_RINGS; ++i) {
+		for (i = 0; i < AMDGPU_SA_NUM_FENCE_LISTS; ++i) {
 			fences[i] = NULL;
 			tries[i] = 0;
 		}
@@ -355,14 +353,17 @@ int amdgpu_sa_bo_new(struct amdgpu_sa_manager *sa_manager,
 			/* see if we can skip over some allocations */
 		} while (amdgpu_sa_bo_next_hole(sa_manager, fences, tries));
 
-		for (i = 0, count = 0; i < AMDGPU_MAX_RINGS; ++i)
+		for (i = 0, count = 0; i < AMDGPU_SA_NUM_FENCE_LISTS; ++i)
 			if (fences[i])
-				fences[count++] = fences[i];
+				fences[count++] = fence_get(fences[i]);
 
 		if (count) {
 			spin_unlock(&sa_manager->wq.lock);
 			t = fence_wait_any_timeout(fences, count, false,
 						   MAX_SCHEDULE_TIMEOUT);
+			for (i = 0; i < count; ++i)
+				fence_put(fences[i]);
+
 			r = (t > 0) ? 0 : t;
 			spin_lock(&sa_manager->wq.lock);
 		} else {
@@ -394,8 +395,9 @@ void amdgpu_sa_bo_free(struct amdgpu_device *adev, struct amdgpu_sa_bo **sa_bo,
 	spin_lock(&sa_manager->wq.lock);
 	if (fence && !fence_is_signaled(fence)) {
 		uint32_t idx;
+
 		(*sa_bo)->fence = fence_get(fence);
-		idx = amdgpu_ring_from_fence(fence)->idx;
+		idx = fence->context % AMDGPU_SA_NUM_FENCE_LISTS;
 		list_add_tail(&(*sa_bo)->flist, &sa_manager->flist[idx]);
 	} else {
 		amdgpu_sa_bo_remove_locked(*sa_bo);
@@ -407,25 +409,6 @@ void amdgpu_sa_bo_free(struct amdgpu_device *adev, struct amdgpu_sa_bo **sa_bo,
 
 #if defined(CONFIG_DEBUG_FS)
 
-static void amdgpu_sa_bo_dump_fence(struct fence *fence, struct seq_file *m)
-{
-	struct amdgpu_fence *a_fence = to_amdgpu_fence(fence);
-	struct amd_sched_fence *s_fence = to_amd_sched_fence(fence);
-
-	if (a_fence)
-		seq_printf(m, " protected by 0x%016llx on ring %d",
-			   a_fence->seq, a_fence->ring->idx);
-
-	if (s_fence) {
-		struct amdgpu_ring *ring;
-
-
-		ring = container_of(s_fence->sched, struct amdgpu_ring, sched);
-		seq_printf(m, " protected by 0x%016x on ring %d",
-			   s_fence->base.seqno, ring->idx);
-	}
-}
-
 void amdgpu_sa_bo_dump_debug_info(struct amdgpu_sa_manager *sa_manager,
 				  struct seq_file *m)
 {
@@ -442,8 +425,11 @@ void amdgpu_sa_bo_dump_debug_info(struct amdgpu_sa_manager *sa_manager,
 		}
 		seq_printf(m, "[0x%010llx 0x%010llx] size %8lld",
 			   soffset, eoffset, eoffset - soffset);
+
 		if (i->fence)
-			amdgpu_sa_bo_dump_fence(i->fence, m);
+			seq_printf(m, " protected by 0x%08x on context %d",
+				   i->fence->seqno, i->fence->context);
+
 		seq_printf(m, "\n");
 	}
 	spin_unlock(&sa_manager->wq.lock);
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_sync.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_sync.c
index c15be00de904..c48b4fce5e57 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_sync.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_sync.c
@@ -37,6 +37,8 @@ struct amdgpu_sync_entry {
 	struct fence		*fence;
 };
 
+static struct kmem_cache *amdgpu_sync_slab;
+
 /**
  * amdgpu_sync_create - zero init sync object
  *
@@ -50,14 +52,18 @@ void amdgpu_sync_create(struct amdgpu_sync *sync)
 	sync->last_vm_update = NULL;
 }
 
+/**
+ * amdgpu_sync_same_dev - test if fence belong to us
+ *
+ * @adev: amdgpu device to use for the test
+ * @f: fence to test
+ *
+ * Test if the fence was issued by us.
+ */
 static bool amdgpu_sync_same_dev(struct amdgpu_device *adev, struct fence *f)
 {
-	struct amdgpu_fence *a_fence = to_amdgpu_fence(f);
 	struct amd_sched_fence *s_fence = to_amd_sched_fence(f);
 
-	if (a_fence)
-		return a_fence->ring->adev == adev;
-
 	if (s_fence) {
 		struct amdgpu_ring *ring;
 
@@ -68,17 +74,31 @@ static bool amdgpu_sync_same_dev(struct amdgpu_device *adev, struct fence *f)
 	return false;
 }
 
-static bool amdgpu_sync_test_owner(struct fence *f, void *owner)
+/**
+ * amdgpu_sync_get_owner - extract the owner of a fence
+ *
+ * @fence: fence get the owner from
+ *
+ * Extract who originally created the fence.
+ */
+static void *amdgpu_sync_get_owner(struct fence *f)
 {
-	struct amdgpu_fence *a_fence = to_amdgpu_fence(f);
 	struct amd_sched_fence *s_fence = to_amd_sched_fence(f);
+
 	if (s_fence)
-		return s_fence->owner == owner;
-	if (a_fence)
-		return a_fence->owner == owner;
-	return false;
+		return s_fence->owner;
+
+	return AMDGPU_FENCE_OWNER_UNDEFINED;
 }
 
+/**
+ * amdgpu_sync_keep_later - Keep the later fence
+ *
+ * @keep: existing fence to test
+ * @fence: new fence
+ *
+ * Either keep the existing fence or the new one, depending which one is later.
+ */
 static void amdgpu_sync_keep_later(struct fence **keep, struct fence *fence)
 {
 	if (*keep && fence_is_later(*keep, fence))
@@ -104,7 +124,7 @@ int amdgpu_sync_fence(struct amdgpu_device *adev, struct amdgpu_sync *sync,
 		return 0;
 
 	if (amdgpu_sync_same_dev(adev, f) &&
-	    amdgpu_sync_test_owner(f, AMDGPU_FENCE_OWNER_VM))
+	    amdgpu_sync_get_owner(f) == AMDGPU_FENCE_OWNER_VM)
 		amdgpu_sync_keep_later(&sync->last_vm_update, f);
 
 	hash_for_each_possible(sync->fences, e, node, f->context) {
@@ -115,7 +135,7 @@ int amdgpu_sync_fence(struct amdgpu_device *adev, struct amdgpu_sync *sync,
 		return 0;
 	}
 
-	e = kmalloc(sizeof(struct amdgpu_sync_entry), GFP_KERNEL);
+	e = kmem_cache_alloc(amdgpu_sync_slab, GFP_KERNEL);
 	if (!e)
 		return -ENOMEM;
 
@@ -124,18 +144,6 @@ int amdgpu_sync_fence(struct amdgpu_device *adev, struct amdgpu_sync *sync,
 	return 0;
 }
 
-static void *amdgpu_sync_get_owner(struct fence *f)
-{
-	struct amdgpu_fence *a_fence = to_amdgpu_fence(f);
-	struct amd_sched_fence *s_fence = to_amd_sched_fence(f);
-
-	if (s_fence)
-		return s_fence->owner;
-	else if (a_fence)
-		return a_fence->owner;
-	return AMDGPU_FENCE_OWNER_UNDEFINED;
-}
-
 /**
  * amdgpu_sync_resv - sync to a reservation object
  *
@@ -208,7 +216,7 @@ struct fence *amdgpu_sync_get_fence(struct amdgpu_sync *sync)
 		f = e->fence;
 
 		hash_del(&e->node);
-		kfree(e);
+		kmem_cache_free(amdgpu_sync_slab, e);
 
 		if (!fence_is_signaled(f))
 			return f;
@@ -231,7 +239,7 @@ int amdgpu_sync_wait(struct amdgpu_sync *sync)
 
 		hash_del(&e->node);
 		fence_put(e->fence);
-		kfree(e);
+		kmem_cache_free(amdgpu_sync_slab, e);
 	}
 
 	return 0;
@@ -253,8 +261,34 @@ void amdgpu_sync_free(struct amdgpu_sync *sync)
 	hash_for_each_safe(sync->fences, i, tmp, e, node) {
 		hash_del(&e->node);
 		fence_put(e->fence);
-		kfree(e);
+		kmem_cache_free(amdgpu_sync_slab, e);
 	}
 
 	fence_put(sync->last_vm_update);
 }
+
+/**
+ * amdgpu_sync_init - init sync object subsystem
+ *
+ * Allocate the slab allocator.
+ */
+int amdgpu_sync_init(void)
+{
+	amdgpu_sync_slab = kmem_cache_create(
+		"amdgpu_sync", sizeof(struct amdgpu_sync_entry), 0,
+		SLAB_HWCACHE_ALIGN, NULL);
+	if (!amdgpu_sync_slab)
+		return -ENOMEM;
+
+	return 0;
+}
+
+/**
+ * amdgpu_sync_fini - fini sync object subsystem
+ *
+ * Free the slab allocator.
+ */
+void amdgpu_sync_fini(void)
+{
+	kmem_cache_destroy(amdgpu_sync_slab);
+}
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_trace.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_trace.h
index 9ca3735c563c..26a5f4acf584 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_trace.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_trace.h
@@ -100,21 +100,24 @@ TRACE_EVENT(amdgpu_sched_run_job,
 
 
 TRACE_EVENT(amdgpu_vm_grab_id,
-	    TP_PROTO(struct amdgpu_vm *vm, unsigned vmid, int ring),
-	    TP_ARGS(vm, vmid, ring),
+	    TP_PROTO(struct amdgpu_vm *vm, int ring, unsigned vmid,
+		     uint64_t pd_addr),
+	    TP_ARGS(vm, ring, vmid, pd_addr),
 	    TP_STRUCT__entry(
 			     __field(struct amdgpu_vm *, vm)
-			     __field(u32, vmid)
 			     __field(u32, ring)
+			     __field(u32, vmid)
+			     __field(u64, pd_addr)
 			     ),
 
 	    TP_fast_assign(
 			   __entry->vm = vm;
-			   __entry->vmid = vmid;
 			   __entry->ring = ring;
+			   __entry->vmid = vmid;
+			   __entry->pd_addr = pd_addr;
 			   ),
-	    TP_printk("vm=%p, id=%u, ring=%u", __entry->vm, __entry->vmid,
-		      __entry->ring)
+	    TP_printk("vm=%p, ring=%u, id=%u, pd_addr=%010Lx", __entry->vm,
+		      __entry->ring, __entry->vmid, __entry->pd_addr)
 );
 
 TRACE_EVENT(amdgpu_vm_bo_map,
@@ -231,8 +234,8 @@ TRACE_EVENT(amdgpu_vm_flush,
 			   __entry->ring = ring;
 			   __entry->id = id;
 			   ),
-	    TP_printk("pd_addr=%010Lx, ring=%u, id=%u",
-		      __entry->pd_addr, __entry->ring, __entry->id)
+	    TP_printk("ring=%u, id=%u, pd_addr=%010Lx",
+		      __entry->ring, __entry->id, __entry->pd_addr)
 );
 
 TRACE_EVENT(amdgpu_bo_list_set,
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c
index e52fc641edfb..0f42b1a24446 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c
@@ -494,29 +494,32 @@ static void amdgpu_ttm_io_mem_free(struct ttm_bo_device *bdev, struct ttm_mem_re
 /*
  * TTM backend functions.
  */
+struct amdgpu_ttm_gup_task_list {
+	struct list_head	list;
+	struct task_struct	*task;
+};
+
 struct amdgpu_ttm_tt {
-	struct ttm_dma_tt		ttm;
-	struct amdgpu_device		*adev;
-	u64				offset;
-	uint64_t			userptr;
-	struct mm_struct		*usermm;
-	uint32_t			userflags;
+	struct ttm_dma_tt	ttm;
+	struct amdgpu_device	*adev;
+	u64			offset;
+	uint64_t		userptr;
+	struct mm_struct	*usermm;
+	uint32_t		userflags;
+	spinlock_t              guptasklock;
+	struct list_head        guptasks;
+	atomic_t		mmu_invalidations;
 };
 
-/* prepare the sg table with the user pages */
-static int amdgpu_ttm_tt_pin_userptr(struct ttm_tt *ttm)
+int amdgpu_ttm_tt_get_user_pages(struct ttm_tt *ttm, struct page **pages)
 {
-	struct amdgpu_device *adev = amdgpu_get_adev(ttm->bdev);
 	struct amdgpu_ttm_tt *gtt = (void *)ttm;
-	unsigned pinned = 0, nents;
-	int r;
-
 	int write = !(gtt->userflags & AMDGPU_GEM_USERPTR_READONLY);
-	enum dma_data_direction direction = write ?
-		DMA_BIDIRECTIONAL : DMA_TO_DEVICE;
+	unsigned pinned = 0;
+	int r;
 
 	if (gtt->userflags & AMDGPU_GEM_USERPTR_ANONONLY) {
-		/* check that we only pin down anonymous memory
+		/* check that we only use anonymous memory
 		   to prevent problems with writeback */
 		unsigned long end = gtt->userptr + ttm->num_pages * PAGE_SIZE;
 		struct vm_area_struct *vma;
@@ -529,10 +532,21 @@ static int amdgpu_ttm_tt_pin_userptr(struct ttm_tt *ttm)
 	do {
 		unsigned num_pages = ttm->num_pages - pinned;
 		uint64_t userptr = gtt->userptr + pinned * PAGE_SIZE;
-		struct page **pages = ttm->pages + pinned;
+		struct page **p = pages + pinned;
+		struct amdgpu_ttm_gup_task_list guptask;
+
+		guptask.task = current;
+		spin_lock(&gtt->guptasklock);
+		list_add(&guptask.list, &gtt->guptasks);
+		spin_unlock(&gtt->guptasklock);
 
 		r = get_user_pages(current, current->mm, userptr, num_pages,
-				   write, 0, pages, NULL);
+				   write, 0, p, NULL);
+
+		spin_lock(&gtt->guptasklock);
+		list_del(&guptask.list);
+		spin_unlock(&gtt->guptasklock);
+
 		if (r < 0)
 			goto release_pages;
 
@@ -540,6 +554,25 @@ static int amdgpu_ttm_tt_pin_userptr(struct ttm_tt *ttm)
 
 	} while (pinned < ttm->num_pages);
 
+	return 0;
+
+release_pages:
+	release_pages(pages, pinned, 0);
+	return r;
+}
+
+/* prepare the sg table with the user pages */
+static int amdgpu_ttm_tt_pin_userptr(struct ttm_tt *ttm)
+{
+	struct amdgpu_device *adev = amdgpu_get_adev(ttm->bdev);
+	struct amdgpu_ttm_tt *gtt = (void *)ttm;
+	unsigned nents;
+	int r;
+
+	int write = !(gtt->userflags & AMDGPU_GEM_USERPTR_READONLY);
+	enum dma_data_direction direction = write ?
+		DMA_BIDIRECTIONAL : DMA_TO_DEVICE;
+
 	r = sg_alloc_table_from_pages(ttm->sg, ttm->pages, ttm->num_pages, 0,
 				      ttm->num_pages << PAGE_SHIFT,
 				      GFP_KERNEL);
@@ -558,9 +591,6 @@ static int amdgpu_ttm_tt_pin_userptr(struct ttm_tt *ttm)
 
 release_sg:
 	kfree(ttm->sg);
-
-release_pages:
-	release_pages(ttm->pages, pinned, 0);
 	return r;
 }
 
@@ -725,7 +755,7 @@ static int amdgpu_ttm_tt_populate(struct ttm_tt *ttm)
 						       0, PAGE_SIZE,
 						       PCI_DMA_BIDIRECTIONAL);
 		if (pci_dma_mapping_error(adev->pdev, gtt->ttm.dma_address[i])) {
-			while (--i) {
+			while (i--) {
 				pci_unmap_page(adev->pdev, gtt->ttm.dma_address[i],
 					       PAGE_SIZE, PCI_DMA_BIDIRECTIONAL);
 				gtt->ttm.dma_address[i] = 0;
@@ -783,6 +813,10 @@ int amdgpu_ttm_tt_set_userptr(struct ttm_tt *ttm, uint64_t addr,
 	gtt->userptr = addr;
 	gtt->usermm = current->mm;
 	gtt->userflags = flags;
+	spin_lock_init(&gtt->guptasklock);
+	INIT_LIST_HEAD(&gtt->guptasks);
+	atomic_set(&gtt->mmu_invalidations, 0);
+
 	return 0;
 }
 
@@ -800,21 +834,40 @@ bool amdgpu_ttm_tt_affect_userptr(struct ttm_tt *ttm, unsigned long start,
 				  unsigned long end)
 {
 	struct amdgpu_ttm_tt *gtt = (void *)ttm;
+	struct amdgpu_ttm_gup_task_list *entry;
 	unsigned long size;
 
-	if (gtt == NULL)
-		return false;
-
-	if (gtt->ttm.ttm.state != tt_bound || !gtt->userptr)
+	if (gtt == NULL || !gtt->userptr)
 		return false;
 
 	size = (unsigned long)gtt->ttm.ttm.num_pages * PAGE_SIZE;
 	if (gtt->userptr > end || gtt->userptr + size <= start)
 		return false;
 
+	spin_lock(&gtt->guptasklock);
+	list_for_each_entry(entry, &gtt->guptasks, list) {
+		if (entry->task == current) {
+			spin_unlock(&gtt->guptasklock);
+			return false;
+		}
+	}
+	spin_unlock(&gtt->guptasklock);
+
+	atomic_inc(&gtt->mmu_invalidations);
+
 	return true;
 }
 
+bool amdgpu_ttm_tt_userptr_invalidated(struct ttm_tt *ttm,
+				       int *last_invalidated)
+{
+	struct amdgpu_ttm_tt *gtt = (void *)ttm;
+	int prev_invalidated = *last_invalidated;
+
+	*last_invalidated = atomic_read(&gtt->mmu_invalidations);
+	return prev_invalidated != *last_invalidated;
+}
+
 bool amdgpu_ttm_tt_is_readonly(struct ttm_tt *ttm)
 {
 	struct amdgpu_ttm_tt *gtt = (void *)ttm;
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_uvd.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_uvd.c
index 1de82bf4fc79..c1a581044417 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_uvd.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_uvd.c
@@ -539,13 +539,6 @@ static int amdgpu_uvd_cs_msg(struct amdgpu_uvd_cs_ctx *ctx,
 		return -EINVAL;
 	}
 
-	r = reservation_object_wait_timeout_rcu(bo->tbo.resv, true, false,
-						MAX_SCHEDULE_TIMEOUT);
-	if (r < 0) {
-		DRM_ERROR("Failed waiting for UVD message (%ld)!\n", r);
-		return r;
-	}
-
 	r = amdgpu_bo_kmap(bo, &ptr);
 	if (r) {
 		DRM_ERROR("Failed mapping the UVD message (%ld)!\n", r);
@@ -886,8 +879,8 @@ static int amdgpu_uvd_send_msg(struct amdgpu_ring *ring, struct amdgpu_bo *bo,
 	ib->length_dw = 16;
 
 	if (direct) {
-		r = amdgpu_ib_schedule(ring, 1, ib,
-				       AMDGPU_FENCE_OWNER_UNDEFINED, NULL, &f);
+		r = amdgpu_ib_schedule(ring, 1, ib, NULL, &f);
+		job->fence = f;
 		if (r)
 			goto err_free;
 
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vce.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_vce.c
index 39c3aa60381a..4bec0c108cea 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vce.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vce.c
@@ -425,8 +425,8 @@ int amdgpu_vce_get_create_msg(struct amdgpu_ring *ring, uint32_t handle,
 	for (i = ib->length_dw; i < ib_size_dw; ++i)
 		ib->ptr[i] = 0x0;
 
-	r = amdgpu_ib_schedule(ring, 1, ib, AMDGPU_FENCE_OWNER_UNDEFINED,
-			       NULL, &f);
+	r = amdgpu_ib_schedule(ring, 1, ib, NULL, &f);
+	job->fence = f;
 	if (r)
 		goto err;
 
@@ -487,9 +487,8 @@ int amdgpu_vce_get_destroy_msg(struct amdgpu_ring *ring, uint32_t handle,
 		ib->ptr[i] = 0x0;
 
 	if (direct) {
-		r = amdgpu_ib_schedule(ring, 1, ib,
-				       AMDGPU_FENCE_OWNER_UNDEFINED,
-				       NULL, &f);
+		r = amdgpu_ib_schedule(ring, 1, ib, NULL, &f);
+		job->fence = f;
 		if (r)
 			goto err;
 
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c
index 264c5968a1d3..b6c011b83641 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c
@@ -50,6 +50,9 @@
  * SI supports 16.
  */
 
+/* Special value that no flush is necessary */
+#define AMDGPU_VM_NO_FLUSH (~0ll)
+
 /**
  * amdgpu_vm_num_pde - return the number of page directory entries
  *
@@ -92,6 +95,7 @@ void amdgpu_vm_get_pd_bo(struct amdgpu_vm *vm,
 	entry->priority = 0;
 	entry->tv.bo = &vm->page_directory->tbo;
 	entry->tv.shared = true;
+	entry->user_pages = NULL;
 	list_add(&entry->tv.head, validated);
 }
 
@@ -157,50 +161,77 @@ void amdgpu_vm_move_pt_bos_in_lru(struct amdgpu_device *adev,
  * Allocate an id for the vm, adding fences to the sync obj as necessary.
  */
 int amdgpu_vm_grab_id(struct amdgpu_vm *vm, struct amdgpu_ring *ring,
-		      struct amdgpu_sync *sync, struct fence *fence)
+		      struct amdgpu_sync *sync, struct fence *fence,
+		      unsigned *vm_id, uint64_t *vm_pd_addr)
 {
-	struct amdgpu_vm_id *vm_id = &vm->ids[ring->idx];
+	uint64_t pd_addr = amdgpu_bo_gpu_offset(vm->page_directory);
 	struct amdgpu_device *adev = ring->adev;
-	struct amdgpu_vm_manager_id *id;
+	struct amdgpu_vm_id *id = &vm->ids[ring->idx];
+	struct fence *updates = sync->last_vm_update;
 	int r;
 
 	mutex_lock(&adev->vm_manager.lock);
 
 	/* check if the id is still valid */
-	if (vm_id->id) {
+	if (id->mgr_id) {
+		struct fence *flushed = id->flushed_updates;
+		bool is_later;
 		long owner;
 
-		id = &adev->vm_manager.ids[vm_id->id];
-		owner = atomic_long_read(&id->owner);
-		if (owner == (long)vm) {
-			list_move_tail(&id->list, &adev->vm_manager.ids_lru);
-			trace_amdgpu_vm_grab_id(vm, vm_id->id, ring->idx);
+		if (!flushed)
+			is_later = true;
+		else if (!updates)
+			is_later = false;
+		else
+			is_later = fence_is_later(updates, flushed);
+
+		owner = atomic_long_read(&id->mgr_id->owner);
+		if (!is_later && owner == (long)id &&
+		    pd_addr == id->pd_gpu_addr) {
+
+			r = amdgpu_sync_fence(ring->adev, sync,
+					      id->mgr_id->active);
+			if (r) {
+				mutex_unlock(&adev->vm_manager.lock);
+				return r;
+			}
+
+			fence_put(id->mgr_id->active);
+			id->mgr_id->active = fence_get(fence);
+
+			list_move_tail(&id->mgr_id->list,
+				       &adev->vm_manager.ids_lru);
 
-			fence_put(id->active);
-			id->active = fence_get(fence);
+			*vm_id = id->mgr_id - adev->vm_manager.ids;
+			*vm_pd_addr = AMDGPU_VM_NO_FLUSH;
+			trace_amdgpu_vm_grab_id(vm, ring->idx, *vm_id,
+						*vm_pd_addr);
 
 			mutex_unlock(&adev->vm_manager.lock);
 			return 0;
 		}
 	}
 
-	/* we definately need to flush */
-	vm_id->pd_gpu_addr = ~0ll;
+	id->mgr_id = list_first_entry(&adev->vm_manager.ids_lru,
+				      struct amdgpu_vm_manager_id,
+				      list);
 
-	id = list_first_entry(&adev->vm_manager.ids_lru,
-			      struct amdgpu_vm_manager_id,
-			      list);
-	list_move_tail(&id->list, &adev->vm_manager.ids_lru);
-	atomic_long_set(&id->owner, (long)vm);
+	r = amdgpu_sync_fence(ring->adev, sync, id->mgr_id->active);
+	if (!r) {
+		fence_put(id->mgr_id->active);
+		id->mgr_id->active = fence_get(fence);
 
-	vm_id->id = id - adev->vm_manager.ids;
-	trace_amdgpu_vm_grab_id(vm, vm_id->id, ring->idx);
+		fence_put(id->flushed_updates);
+		id->flushed_updates = fence_get(updates);
 
-	r = amdgpu_sync_fence(ring->adev, sync, id->active);
+		id->pd_gpu_addr = pd_addr;
 
-	if (!r) {
-		fence_put(id->active);
-		id->active = fence_get(fence);
+		list_move_tail(&id->mgr_id->list, &adev->vm_manager.ids_lru);
+		atomic_long_set(&id->mgr_id->owner, (long)id);
+
+		*vm_id = id->mgr_id - adev->vm_manager.ids;
+		*vm_pd_addr = pd_addr;
+		trace_amdgpu_vm_grab_id(vm, ring->idx, *vm_id, *vm_pd_addr);
 	}
 
 	mutex_unlock(&adev->vm_manager.lock);
@@ -211,39 +242,71 @@ int amdgpu_vm_grab_id(struct amdgpu_vm *vm, struct amdgpu_ring *ring,
  * amdgpu_vm_flush - hardware flush the vm
  *
  * @ring: ring to use for flush
- * @vm: vm we want to flush
- * @updates: last vm update that we waited for
+ * @vm_id: vmid number to use
+ * @pd_addr: address of the page directory
  *
- * Flush the vm.
+ * Emit a VM flush when it is necessary.
  */
 void amdgpu_vm_flush(struct amdgpu_ring *ring,
-		     struct amdgpu_vm *vm,
-		     struct fence *updates)
+		     unsigned vm_id, uint64_t pd_addr,
+		     uint32_t gds_base, uint32_t gds_size,
+		     uint32_t gws_base, uint32_t gws_size,
+		     uint32_t oa_base, uint32_t oa_size)
 {
-	uint64_t pd_addr = amdgpu_bo_gpu_offset(vm->page_directory);
-	struct amdgpu_vm_id *vm_id = &vm->ids[ring->idx];
-	struct fence *flushed_updates = vm_id->flushed_updates;
-	bool is_later;
-
-	if (!flushed_updates)
-		is_later = true;
-	else if (!updates)
-		is_later = false;
-	else
-		is_later = fence_is_later(updates, flushed_updates);
+	struct amdgpu_device *adev = ring->adev;
+	struct amdgpu_vm_manager_id *mgr_id = &adev->vm_manager.ids[vm_id];
+	bool gds_switch_needed = ring->funcs->emit_gds_switch && (
+		mgr_id->gds_base != gds_base ||
+		mgr_id->gds_size != gds_size ||
+		mgr_id->gws_base != gws_base ||
+		mgr_id->gws_size != gws_size ||
+		mgr_id->oa_base != oa_base ||
+		mgr_id->oa_size != oa_size);
+
+	if (ring->funcs->emit_pipeline_sync && (
+	    pd_addr != AMDGPU_VM_NO_FLUSH || gds_switch_needed))
+		amdgpu_ring_emit_pipeline_sync(ring);
+
+	if (pd_addr != AMDGPU_VM_NO_FLUSH) {
+		trace_amdgpu_vm_flush(pd_addr, ring->idx, vm_id);
+		amdgpu_ring_emit_vm_flush(ring, vm_id, pd_addr);
+	}
 
-	if (pd_addr != vm_id->pd_gpu_addr || is_later) {
-		trace_amdgpu_vm_flush(pd_addr, ring->idx, vm_id->id);
-		if (is_later) {
-			vm_id->flushed_updates = fence_get(updates);
-			fence_put(flushed_updates);
-		}
-		vm_id->pd_gpu_addr = pd_addr;
-		amdgpu_ring_emit_vm_flush(ring, vm_id->id, vm_id->pd_gpu_addr);
+	if (gds_switch_needed) {
+		mgr_id->gds_base = gds_base;
+		mgr_id->gds_size = gds_size;
+		mgr_id->gws_base = gws_base;
+		mgr_id->gws_size = gws_size;
+		mgr_id->oa_base = oa_base;
+		mgr_id->oa_size = oa_size;
+		amdgpu_ring_emit_gds_switch(ring, vm_id,
+					    gds_base, gds_size,
+					    gws_base, gws_size,
+					    oa_base, oa_size);
 	}
 }
 
 /**
+ * amdgpu_vm_reset_id - reset VMID to zero
+ *
+ * @adev: amdgpu device structure
+ * @vm_id: vmid number to use
+ *
+ * Reset saved GDW, GWS and OA to force switch on next flush.
+ */
+void amdgpu_vm_reset_id(struct amdgpu_device *adev, unsigned vm_id)
+{
+	struct amdgpu_vm_manager_id *mgr_id = &adev->vm_manager.ids[vm_id];
+
+	mgr_id->gds_base = 0;
+	mgr_id->gds_size = 0;
+	mgr_id->gws_base = 0;
+	mgr_id->gws_size = 0;
+	mgr_id->oa_base = 0;
+	mgr_id->oa_size = 0;
+}
+
+/**
  * amdgpu_vm_bo_find - find the bo_va for a specific vm & bo
  *
  * @vm: requested vm
@@ -804,7 +867,7 @@ static int amdgpu_vm_bo_split_mapping(struct amdgpu_device *adev,
 	while (start != mapping->it.last + 1) {
 		uint64_t last;
 
-		last = min((uint64_t)mapping->it.last, start + max_size);
+		last = min((uint64_t)mapping->it.last, start + max_size - 1);
 		r = amdgpu_vm_bo_update_mapping(adev, gtt, gtt_flags, vm,
 						start, last, flags, addr,
 						fence);
@@ -812,7 +875,7 @@ static int amdgpu_vm_bo_split_mapping(struct amdgpu_device *adev,
 			return r;
 
 		start = last + 1;
-		addr += max_size;
+		addr += max_size * AMDGPU_GPU_PAGE_SIZE;
 	}
 
 	return 0;
@@ -908,22 +971,18 @@ int amdgpu_vm_clear_freed(struct amdgpu_device *adev,
 	struct amdgpu_bo_va_mapping *mapping;
 	int r;
 
-	spin_lock(&vm->freed_lock);
 	while (!list_empty(&vm->freed)) {
 		mapping = list_first_entry(&vm->freed,
 			struct amdgpu_bo_va_mapping, list);
 		list_del(&mapping->list);
-		spin_unlock(&vm->freed_lock);
+
 		r = amdgpu_vm_bo_split_mapping(adev, NULL, 0, vm, mapping,
 					       0, NULL);
 		kfree(mapping);
 		if (r)
 			return r;
 
-		spin_lock(&vm->freed_lock);
 	}
-	spin_unlock(&vm->freed_lock);
-
 	return 0;
 
 }
@@ -950,9 +1009,8 @@ int amdgpu_vm_clear_invalids(struct amdgpu_device *adev,
 		bo_va = list_first_entry(&vm->invalidated,
 			struct amdgpu_bo_va, vm_status);
 		spin_unlock(&vm->status_lock);
-		mutex_lock(&bo_va->mutex);
+
 		r = amdgpu_vm_bo_update(adev, bo_va, NULL);
-		mutex_unlock(&bo_va->mutex);
 		if (r)
 			return r;
 
@@ -996,7 +1054,7 @@ struct amdgpu_bo_va *amdgpu_vm_bo_add(struct amdgpu_device *adev,
 	INIT_LIST_HEAD(&bo_va->valids);
 	INIT_LIST_HEAD(&bo_va->invalids);
 	INIT_LIST_HEAD(&bo_va->vm_status);
-	mutex_init(&bo_va->mutex);
+
 	list_add_tail(&bo_va->bo_list, &bo->va);
 
 	return bo_va;
@@ -1048,9 +1106,7 @@ int amdgpu_vm_bo_map(struct amdgpu_device *adev,
 	saddr /= AMDGPU_GPU_PAGE_SIZE;
 	eaddr /= AMDGPU_GPU_PAGE_SIZE;
 
-	spin_lock(&vm->it_lock);
 	it = interval_tree_iter_first(&vm->va, saddr, eaddr);
-	spin_unlock(&vm->it_lock);
 	if (it) {
 		struct amdgpu_bo_va_mapping *tmp;
 		tmp = container_of(it, struct amdgpu_bo_va_mapping, it);
@@ -1074,13 +1130,8 @@ int amdgpu_vm_bo_map(struct amdgpu_device *adev,
 	mapping->offset = offset;
 	mapping->flags = flags;
 
-	mutex_lock(&bo_va->mutex);
 	list_add(&mapping->list, &bo_va->invalids);
-	mutex_unlock(&bo_va->mutex);
-	spin_lock(&vm->it_lock);
 	interval_tree_insert(&mapping->it, &vm->va);
-	spin_unlock(&vm->it_lock);
-	trace_amdgpu_vm_bo_map(bo_va, mapping);
 
 	/* Make sure the page tables are allocated */
 	saddr >>= amdgpu_vm_block_size;
@@ -1124,6 +1175,7 @@ int amdgpu_vm_bo_map(struct amdgpu_device *adev,
 		entry->priority = 0;
 		entry->tv.bo = &entry->robj->tbo;
 		entry->tv.shared = true;
+		entry->user_pages = NULL;
 		vm->page_tables[pt_idx].addr = 0;
 	}
 
@@ -1131,9 +1183,7 @@ int amdgpu_vm_bo_map(struct amdgpu_device *adev,
 
 error_free:
 	list_del(&mapping->list);
-	spin_lock(&vm->it_lock);
 	interval_tree_remove(&mapping->it, &vm->va);
-	spin_unlock(&vm->it_lock);
 	trace_amdgpu_vm_bo_unmap(bo_va, mapping);
 	kfree(mapping);
 
@@ -1162,7 +1212,7 @@ int amdgpu_vm_bo_unmap(struct amdgpu_device *adev,
 	bool valid = true;
 
 	saddr /= AMDGPU_GPU_PAGE_SIZE;
-	mutex_lock(&bo_va->mutex);
+
 	list_for_each_entry(mapping, &bo_va->valids, list) {
 		if (mapping->it.start == saddr)
 			break;
@@ -1176,25 +1226,18 @@ int amdgpu_vm_bo_unmap(struct amdgpu_device *adev,
 				break;
 		}
 
-		if (&mapping->list == &bo_va->invalids) {
-			mutex_unlock(&bo_va->mutex);
+		if (&mapping->list == &bo_va->invalids)
 			return -ENOENT;
-		}
 	}
-	mutex_unlock(&bo_va->mutex);
+
 	list_del(&mapping->list);
-	spin_lock(&vm->it_lock);
 	interval_tree_remove(&mapping->it, &vm->va);
-	spin_unlock(&vm->it_lock);
 	trace_amdgpu_vm_bo_unmap(bo_va, mapping);
 
-	if (valid) {
-		spin_lock(&vm->freed_lock);
+	if (valid)
 		list_add(&mapping->list, &vm->freed);
-		spin_unlock(&vm->freed_lock);
-	} else {
+	else
 		kfree(mapping);
-	}
 
 	return 0;
 }
@@ -1223,23 +1266,17 @@ void amdgpu_vm_bo_rmv(struct amdgpu_device *adev,
 
 	list_for_each_entry_safe(mapping, next, &bo_va->valids, list) {
 		list_del(&mapping->list);
-		spin_lock(&vm->it_lock);
 		interval_tree_remove(&mapping->it, &vm->va);
-		spin_unlock(&vm->it_lock);
 		trace_amdgpu_vm_bo_unmap(bo_va, mapping);
-		spin_lock(&vm->freed_lock);
 		list_add(&mapping->list, &vm->freed);
-		spin_unlock(&vm->freed_lock);
 	}
 	list_for_each_entry_safe(mapping, next, &bo_va->invalids, list) {
 		list_del(&mapping->list);
-		spin_lock(&vm->it_lock);
 		interval_tree_remove(&mapping->it, &vm->va);
-		spin_unlock(&vm->it_lock);
 		kfree(mapping);
 	}
+
 	fence_put(bo_va->last_pt_update);
-	mutex_destroy(&bo_va->mutex);
 	kfree(bo_va);
 }
 
@@ -1284,7 +1321,7 @@ int amdgpu_vm_init(struct amdgpu_device *adev, struct amdgpu_vm *vm)
 	int i, r;
 
 	for (i = 0; i < AMDGPU_MAX_RINGS; ++i) {
-		vm->ids[i].id = 0;
+		vm->ids[i].mgr_id = NULL;
 		vm->ids[i].flushed_updates = NULL;
 	}
 	vm->va = RB_ROOT;
@@ -1292,8 +1329,7 @@ int amdgpu_vm_init(struct amdgpu_device *adev, struct amdgpu_vm *vm)
 	INIT_LIST_HEAD(&vm->invalidated);
 	INIT_LIST_HEAD(&vm->cleared);
 	INIT_LIST_HEAD(&vm->freed);
-	spin_lock_init(&vm->it_lock);
-	spin_lock_init(&vm->freed_lock);
+
 	pd_size = amdgpu_vm_directory_size(adev);
 	pd_entries = amdgpu_vm_num_pdes(adev);
 
@@ -1380,14 +1416,15 @@ void amdgpu_vm_fini(struct amdgpu_device *adev, struct amdgpu_vm *vm)
 
 	amdgpu_bo_unref(&vm->page_directory);
 	fence_put(vm->page_directory_fence);
+
 	for (i = 0; i < AMDGPU_MAX_RINGS; ++i) {
-		unsigned id = vm->ids[i].id;
+		struct amdgpu_vm_id *id = &vm->ids[i];
 
-		atomic_long_cmpxchg(&adev->vm_manager.ids[id].owner,
-				    (long)vm, 0);
-		fence_put(vm->ids[i].flushed_updates);
+		if (id->mgr_id)
+			atomic_long_cmpxchg(&id->mgr_id->owner,
+					    (long)id, 0);
+		fence_put(id->flushed_updates);
 	}
-
 }
 
 /**
@@ -1404,9 +1441,11 @@ void amdgpu_vm_manager_init(struct amdgpu_device *adev)
 	INIT_LIST_HEAD(&adev->vm_manager.ids_lru);
 
 	/* skip over VMID 0, since it is the system VM */
-	for (i = 1; i < adev->vm_manager.num_ids; ++i)
+	for (i = 1; i < adev->vm_manager.num_ids; ++i) {
+		amdgpu_vm_reset_id(adev, i);
 		list_add_tail(&adev->vm_manager.ids[i].list,
 			      &adev->vm_manager.ids_lru);
+	}
 
 	atomic_set(&adev->vm_manager.vm_pte_next_ring, 0);
 }
diff --git a/drivers/gpu/drm/amd/amdgpu/atombios_dp.c b/drivers/gpu/drm/amd/amdgpu/atombios_dp.c
index 21aacc1f45c1..bf731e9f643e 100644
--- a/drivers/gpu/drm/amd/amdgpu/atombios_dp.c
+++ b/drivers/gpu/drm/amd/amdgpu/atombios_dp.c
@@ -265,15 +265,27 @@ static int amdgpu_atombios_dp_get_dp_link_config(struct drm_connector *connector
 	unsigned max_lane_num = drm_dp_max_lane_count(dpcd);
 	unsigned lane_num, i, max_pix_clock;
 
-	for (lane_num = 1; lane_num <= max_lane_num; lane_num <<= 1) {
-		for (i = 0; i < ARRAY_SIZE(link_rates) && link_rates[i] <= max_link_rate; i++) {
-			max_pix_clock = (lane_num * link_rates[i] * 8) / bpp;
+	if (amdgpu_connector_encoder_get_dp_bridge_encoder_id(connector) ==
+	    ENCODER_OBJECT_ID_NUTMEG) {
+		for (lane_num = 1; lane_num <= max_lane_num; lane_num <<= 1) {
+			max_pix_clock = (lane_num * 270000 * 8) / bpp;
 			if (max_pix_clock >= pix_clock) {
 				*dp_lanes = lane_num;
-				*dp_rate = link_rates[i];
+				*dp_rate = 270000;
 				return 0;
 			}
 		}
+	} else {
+		for (lane_num = 1; lane_num <= max_lane_num; lane_num <<= 1) {
+			for (i = 0; i < ARRAY_SIZE(link_rates) && link_rates[i] <= max_link_rate; i++) {
+				max_pix_clock = (lane_num * link_rates[i] * 8) / bpp;
+				if (max_pix_clock >= pix_clock) {
+					*dp_lanes = lane_num;
+					*dp_rate = link_rates[i];
+					return 0;
+				}
+			}
+		}
 	}
 
 	return -EINVAL;
diff --git a/drivers/gpu/drm/amd/amdgpu/ci_dpm.c b/drivers/gpu/drm/amd/amdgpu/ci_dpm.c
index 8b4731d4e10e..1f9109d3348b 100644
--- a/drivers/gpu/drm/amd/amdgpu/ci_dpm.c
+++ b/drivers/gpu/drm/amd/amdgpu/ci_dpm.c
@@ -31,6 +31,7 @@
 #include "ci_dpm.h"
 #include "gfx_v7_0.h"
 #include "atom.h"
+#include "amd_pcie.h"
 #include <linux/seq_file.h>
 
 #include "smu/smu_7_0_1_d.h"
@@ -3016,7 +3017,6 @@ static int ci_populate_single_memory_level(struct amdgpu_device *adev,
 						      &memory_level->MinVddcPhases);
 
 	memory_level->EnabledForThrottle = 1;
-	memory_level->EnabledForActivity = 1;
 	memory_level->UpH = 0;
 	memory_level->DownH = 100;
 	memory_level->VoltageDownH = 0;
@@ -3375,7 +3375,6 @@ static int ci_populate_single_graphic_level(struct amdgpu_device *adev,
 	graphic_level->SpllSpreadSpectrum2 = cpu_to_be32(graphic_level->SpllSpreadSpectrum2);
 	graphic_level->CcPwrDynRm = cpu_to_be32(graphic_level->CcPwrDynRm);
 	graphic_level->CcPwrDynRm1 = cpu_to_be32(graphic_level->CcPwrDynRm1);
-	graphic_level->EnabledForActivity = 1;
 
 	return 0;
 }
@@ -3406,6 +3405,7 @@ static int ci_populate_all_graphic_levels(struct amdgpu_device *adev)
 			pi->smc_state_table.GraphicsLevel[i].DisplayWatermark =
 				PPSMC_DISPLAY_WATERMARK_HIGH;
 	}
+	pi->smc_state_table.GraphicsLevel[0].EnabledForActivity = 1;
 
 	pi->smc_state_table.GraphicsDpmLevelCount = (u8)dpm_table->sclk_table.count;
 	pi->dpm_level_enable_mask.sclk_dpm_enable_mask =
@@ -3449,6 +3449,8 @@ static int ci_populate_all_memory_levels(struct amdgpu_device *adev)
 			return ret;
 	}
 
+	pi->smc_state_table.MemoryLevel[0].EnabledForActivity = 1;
+
 	if ((dpm_table->mclk_table.count >= 2) &&
 	    ((adev->pdev->device == 0x67B0) || (adev->pdev->device == 0x67B1))) {
 		pi->smc_state_table.MemoryLevel[1].MinVddc =
@@ -4380,26 +4382,6 @@ static int ci_dpm_force_performance_level(struct amdgpu_device *adev,
 				}
 			}
 		}
-		if ((!pi->pcie_dpm_key_disabled) &&
-		    pi->dpm_level_enable_mask.pcie_dpm_enable_mask) {
-			levels = 0;
-			tmp = pi->dpm_level_enable_mask.pcie_dpm_enable_mask;
-			while (tmp >>= 1)
-				levels++;
-			if (levels) {
-				ret = ci_dpm_force_state_pcie(adev, level);
-				if (ret)
-					return ret;
-				for (i = 0; i < adev->usec_timeout; i++) {
-					tmp = (RREG32_SMC(ixTARGET_AND_CURRENT_PROFILE_INDEX_1) &
-					TARGET_AND_CURRENT_PROFILE_INDEX_1__CURR_PCIE_INDEX_MASK) >>
-					TARGET_AND_CURRENT_PROFILE_INDEX_1__CURR_PCIE_INDEX__SHIFT;
-					if (tmp == levels)
-						break;
-					udelay(1);
-				}
-			}
-		}
 	} else if (level == AMDGPU_DPM_FORCED_LEVEL_LOW) {
 		if ((!pi->sclk_dpm_key_disabled) &&
 		    pi->dpm_level_enable_mask.sclk_dpm_enable_mask) {
@@ -5394,30 +5376,6 @@ static int ci_dpm_enable(struct amdgpu_device *adev)
 
 	ci_update_current_ps(adev, boot_ps);
 
-	if (adev->irq.installed &&
-	    amdgpu_is_internal_thermal_sensor(adev->pm.int_thermal_type)) {
-#if 0
-		PPSMC_Result result;
-#endif
-		ret = ci_thermal_set_temperature_range(adev, CISLANDS_TEMP_RANGE_MIN,
-						       CISLANDS_TEMP_RANGE_MAX);
-		if (ret) {
-			DRM_ERROR("ci_thermal_set_temperature_range failed\n");
-			return ret;
-		}
-		amdgpu_irq_get(adev, &adev->pm.dpm.thermal.irq,
-			       AMDGPU_THERMAL_IRQ_LOW_TO_HIGH);
-		amdgpu_irq_get(adev, &adev->pm.dpm.thermal.irq,
-			       AMDGPU_THERMAL_IRQ_HIGH_TO_LOW);
-
-#if 0
-		result = amdgpu_ci_send_msg_to_smc(adev, PPSMC_MSG_EnableThermalInterrupt);
-
-		if (result != PPSMC_Result_OK)
-			DRM_DEBUG_KMS("Could not enable thermal interrupts.\n");
-#endif
-	}
-
 	return 0;
 }
 
@@ -5835,18 +5793,16 @@ static int ci_dpm_init(struct amdgpu_device *adev)
 	u8 frev, crev;
 	struct ci_power_info *pi;
 	int ret;
-	u32 mask;
 
 	pi = kzalloc(sizeof(struct ci_power_info), GFP_KERNEL);
 	if (pi == NULL)
 		return -ENOMEM;
 	adev->pm.dpm.priv = pi;
 
-	ret = drm_pcie_get_speed_cap_mask(adev->ddev, &mask);
-	if (ret)
-		pi->sys_pcie_mask = 0;
-	else
-		pi->sys_pcie_mask = mask;
+	pi->sys_pcie_mask =
+		(adev->pm.pcie_gen_mask & CAIL_PCIE_LINK_SPEED_SUPPORT_MASK) >>
+		CAIL_PCIE_LINK_SPEED_SUPPORT_SHIFT;
+
 	pi->force_pcie_gen = AMDGPU_PCIE_GEN_INVALID;
 
 	pi->pcie_gen_performance.max = AMDGPU_PCIE_GEN1;
diff --git a/drivers/gpu/drm/amd/amdgpu/cik.c b/drivers/gpu/drm/amd/amdgpu/cik.c
index 6b1f0539ce9d..bddc9ba11495 100644
--- a/drivers/gpu/drm/amd/amdgpu/cik.c
+++ b/drivers/gpu/drm/amd/amdgpu/cik.c
@@ -1462,6 +1462,9 @@ static void cik_program_aspm(struct amdgpu_device *adev)
 	if (amdgpu_aspm == 0)
 		return;
 
+	if (pci_is_root_bus(adev->pdev->bus))
+		return;
+
 	/* XXX double check APUs */
 	if (adev->flags & AMD_IS_APU)
 		return;
@@ -2025,79 +2028,77 @@ static int cik_common_early_init(void *handle)
 
 	adev->asic_funcs = &cik_asic_funcs;
 
-	adev->has_uvd = true;
-
 	adev->rev_id = cik_get_rev_id(adev);
 	adev->external_rev_id = 0xFF;
 	switch (adev->asic_type) {
 	case CHIP_BONAIRE:
 		adev->cg_flags =
-			AMDGPU_CG_SUPPORT_GFX_MGCG |
-			AMDGPU_CG_SUPPORT_GFX_MGLS |
-			/*AMDGPU_CG_SUPPORT_GFX_CGCG |*/
-			AMDGPU_CG_SUPPORT_GFX_CGLS |
-			AMDGPU_CG_SUPPORT_GFX_CGTS |
-			AMDGPU_CG_SUPPORT_GFX_CGTS_LS |
-			AMDGPU_CG_SUPPORT_GFX_CP_LS |
-			AMDGPU_CG_SUPPORT_MC_LS |
-			AMDGPU_CG_SUPPORT_MC_MGCG |
-			AMDGPU_CG_SUPPORT_SDMA_MGCG |
-			AMDGPU_CG_SUPPORT_SDMA_LS |
-			AMDGPU_CG_SUPPORT_BIF_LS |
-			AMDGPU_CG_SUPPORT_VCE_MGCG |
-			AMDGPU_CG_SUPPORT_UVD_MGCG |
-			AMDGPU_CG_SUPPORT_HDP_LS |
-			AMDGPU_CG_SUPPORT_HDP_MGCG;
+			AMD_CG_SUPPORT_GFX_MGCG |
+			AMD_CG_SUPPORT_GFX_MGLS |
+			/*AMD_CG_SUPPORT_GFX_CGCG |*/
+			AMD_CG_SUPPORT_GFX_CGLS |
+			AMD_CG_SUPPORT_GFX_CGTS |
+			AMD_CG_SUPPORT_GFX_CGTS_LS |
+			AMD_CG_SUPPORT_GFX_CP_LS |
+			AMD_CG_SUPPORT_MC_LS |
+			AMD_CG_SUPPORT_MC_MGCG |
+			AMD_CG_SUPPORT_SDMA_MGCG |
+			AMD_CG_SUPPORT_SDMA_LS |
+			AMD_CG_SUPPORT_BIF_LS |
+			AMD_CG_SUPPORT_VCE_MGCG |
+			AMD_CG_SUPPORT_UVD_MGCG |
+			AMD_CG_SUPPORT_HDP_LS |
+			AMD_CG_SUPPORT_HDP_MGCG;
 		adev->pg_flags = 0;
 		adev->external_rev_id = adev->rev_id + 0x14;
 		break;
 	case CHIP_HAWAII:
 		adev->cg_flags =
-			AMDGPU_CG_SUPPORT_GFX_MGCG |
-			AMDGPU_CG_SUPPORT_GFX_MGLS |
-			/*AMDGPU_CG_SUPPORT_GFX_CGCG |*/
-			AMDGPU_CG_SUPPORT_GFX_CGLS |
-			AMDGPU_CG_SUPPORT_GFX_CGTS |
-			AMDGPU_CG_SUPPORT_GFX_CP_LS |
-			AMDGPU_CG_SUPPORT_MC_LS |
-			AMDGPU_CG_SUPPORT_MC_MGCG |
-			AMDGPU_CG_SUPPORT_SDMA_MGCG |
-			AMDGPU_CG_SUPPORT_SDMA_LS |
-			AMDGPU_CG_SUPPORT_BIF_LS |
-			AMDGPU_CG_SUPPORT_VCE_MGCG |
-			AMDGPU_CG_SUPPORT_UVD_MGCG |
-			AMDGPU_CG_SUPPORT_HDP_LS |
-			AMDGPU_CG_SUPPORT_HDP_MGCG;
+			AMD_CG_SUPPORT_GFX_MGCG |
+			AMD_CG_SUPPORT_GFX_MGLS |
+			/*AMD_CG_SUPPORT_GFX_CGCG |*/
+			AMD_CG_SUPPORT_GFX_CGLS |
+			AMD_CG_SUPPORT_GFX_CGTS |
+			AMD_CG_SUPPORT_GFX_CP_LS |
+			AMD_CG_SUPPORT_MC_LS |
+			AMD_CG_SUPPORT_MC_MGCG |
+			AMD_CG_SUPPORT_SDMA_MGCG |
+			AMD_CG_SUPPORT_SDMA_LS |
+			AMD_CG_SUPPORT_BIF_LS |
+			AMD_CG_SUPPORT_VCE_MGCG |
+			AMD_CG_SUPPORT_UVD_MGCG |
+			AMD_CG_SUPPORT_HDP_LS |
+			AMD_CG_SUPPORT_HDP_MGCG;
 		adev->pg_flags = 0;
 		adev->external_rev_id = 0x28;
 		break;
 	case CHIP_KAVERI:
 		adev->cg_flags =
-			AMDGPU_CG_SUPPORT_GFX_MGCG |
-			AMDGPU_CG_SUPPORT_GFX_MGLS |
-			/*AMDGPU_CG_SUPPORT_GFX_CGCG |*/
-			AMDGPU_CG_SUPPORT_GFX_CGLS |
-			AMDGPU_CG_SUPPORT_GFX_CGTS |
-			AMDGPU_CG_SUPPORT_GFX_CGTS_LS |
-			AMDGPU_CG_SUPPORT_GFX_CP_LS |
-			AMDGPU_CG_SUPPORT_SDMA_MGCG |
-			AMDGPU_CG_SUPPORT_SDMA_LS |
-			AMDGPU_CG_SUPPORT_BIF_LS |
-			AMDGPU_CG_SUPPORT_VCE_MGCG |
-			AMDGPU_CG_SUPPORT_UVD_MGCG |
-			AMDGPU_CG_SUPPORT_HDP_LS |
-			AMDGPU_CG_SUPPORT_HDP_MGCG;
+			AMD_CG_SUPPORT_GFX_MGCG |
+			AMD_CG_SUPPORT_GFX_MGLS |
+			/*AMD_CG_SUPPORT_GFX_CGCG |*/
+			AMD_CG_SUPPORT_GFX_CGLS |
+			AMD_CG_SUPPORT_GFX_CGTS |
+			AMD_CG_SUPPORT_GFX_CGTS_LS |
+			AMD_CG_SUPPORT_GFX_CP_LS |
+			AMD_CG_SUPPORT_SDMA_MGCG |
+			AMD_CG_SUPPORT_SDMA_LS |
+			AMD_CG_SUPPORT_BIF_LS |
+			AMD_CG_SUPPORT_VCE_MGCG |
+			AMD_CG_SUPPORT_UVD_MGCG |
+			AMD_CG_SUPPORT_HDP_LS |
+			AMD_CG_SUPPORT_HDP_MGCG;
 		adev->pg_flags =
-			/*AMDGPU_PG_SUPPORT_GFX_PG |
-			  AMDGPU_PG_SUPPORT_GFX_SMG |
-			  AMDGPU_PG_SUPPORT_GFX_DMG |*/
-			AMDGPU_PG_SUPPORT_UVD |
-			/*AMDGPU_PG_SUPPORT_VCE |
-			  AMDGPU_PG_SUPPORT_CP |
-			  AMDGPU_PG_SUPPORT_GDS |
-			  AMDGPU_PG_SUPPORT_RLC_SMU_HS |
-			  AMDGPU_PG_SUPPORT_ACP |
-			  AMDGPU_PG_SUPPORT_SAMU |*/
+			/*AMD_PG_SUPPORT_GFX_PG |
+			  AMD_PG_SUPPORT_GFX_SMG |
+			  AMD_PG_SUPPORT_GFX_DMG |*/
+			AMD_PG_SUPPORT_UVD |
+			/*AMD_PG_SUPPORT_VCE |
+			  AMD_PG_SUPPORT_CP |
+			  AMD_PG_SUPPORT_GDS |
+			  AMD_PG_SUPPORT_RLC_SMU_HS |
+			  AMD_PG_SUPPORT_ACP |
+			  AMD_PG_SUPPORT_SAMU |*/
 			0;
 		if (adev->pdev->device == 0x1312 ||
 			adev->pdev->device == 0x1316 ||
@@ -2109,29 +2110,29 @@ static int cik_common_early_init(void *handle)
 	case CHIP_KABINI:
 	case CHIP_MULLINS:
 		adev->cg_flags =
-			AMDGPU_CG_SUPPORT_GFX_MGCG |
-			AMDGPU_CG_SUPPORT_GFX_MGLS |
-			/*AMDGPU_CG_SUPPORT_GFX_CGCG |*/
-			AMDGPU_CG_SUPPORT_GFX_CGLS |
-			AMDGPU_CG_SUPPORT_GFX_CGTS |
-			AMDGPU_CG_SUPPORT_GFX_CGTS_LS |
-			AMDGPU_CG_SUPPORT_GFX_CP_LS |
-			AMDGPU_CG_SUPPORT_SDMA_MGCG |
-			AMDGPU_CG_SUPPORT_SDMA_LS |
-			AMDGPU_CG_SUPPORT_BIF_LS |
-			AMDGPU_CG_SUPPORT_VCE_MGCG |
-			AMDGPU_CG_SUPPORT_UVD_MGCG |
-			AMDGPU_CG_SUPPORT_HDP_LS |
-			AMDGPU_CG_SUPPORT_HDP_MGCG;
+			AMD_CG_SUPPORT_GFX_MGCG |
+			AMD_CG_SUPPORT_GFX_MGLS |
+			/*AMD_CG_SUPPORT_GFX_CGCG |*/
+			AMD_CG_SUPPORT_GFX_CGLS |
+			AMD_CG_SUPPORT_GFX_CGTS |
+			AMD_CG_SUPPORT_GFX_CGTS_LS |
+			AMD_CG_SUPPORT_GFX_CP_LS |
+			AMD_CG_SUPPORT_SDMA_MGCG |
+			AMD_CG_SUPPORT_SDMA_LS |
+			AMD_CG_SUPPORT_BIF_LS |
+			AMD_CG_SUPPORT_VCE_MGCG |
+			AMD_CG_SUPPORT_UVD_MGCG |
+			AMD_CG_SUPPORT_HDP_LS |
+			AMD_CG_SUPPORT_HDP_MGCG;
 		adev->pg_flags =
-			/*AMDGPU_PG_SUPPORT_GFX_PG |
-			  AMDGPU_PG_SUPPORT_GFX_SMG | */
-			AMDGPU_PG_SUPPORT_UVD |
-			/*AMDGPU_PG_SUPPORT_VCE |
-			  AMDGPU_PG_SUPPORT_CP |
-			  AMDGPU_PG_SUPPORT_GDS |
-			  AMDGPU_PG_SUPPORT_RLC_SMU_HS |
-			  AMDGPU_PG_SUPPORT_SAMU |*/
+			/*AMD_PG_SUPPORT_GFX_PG |
+			  AMD_PG_SUPPORT_GFX_SMG | */
+			AMD_PG_SUPPORT_UVD |
+			/*AMD_PG_SUPPORT_VCE |
+			  AMD_PG_SUPPORT_CP |
+			  AMD_PG_SUPPORT_GDS |
+			  AMD_PG_SUPPORT_RLC_SMU_HS |
+			  AMD_PG_SUPPORT_SAMU |*/
 			0;
 		if (adev->asic_type == CHIP_KABINI) {
 			if (adev->rev_id == 0)
diff --git a/drivers/gpu/drm/amd/amdgpu/cik_sdma.c b/drivers/gpu/drm/amd/amdgpu/cik_sdma.c
index 675f34916aab..d3ac3298fba8 100644
--- a/drivers/gpu/drm/amd/amdgpu/cik_sdma.c
+++ b/drivers/gpu/drm/amd/amdgpu/cik_sdma.c
@@ -212,7 +212,7 @@ static void cik_sdma_ring_insert_nop(struct amdgpu_ring *ring, uint32_t count)
 static void cik_sdma_ring_emit_ib(struct amdgpu_ring *ring,
 			   struct amdgpu_ib *ib)
 {
-	u32 extra_bits = (ib->vm ? ib->vm->ids[ring->idx].id : 0) & 0xf;
+	u32 extra_bits = ib->vm_id & 0xf;
 	u32 next_rptr = ring->wptr + 5;
 
 	while ((next_rptr & 7) != 4)
@@ -261,6 +261,13 @@ static void cik_sdma_ring_emit_hdp_flush(struct amdgpu_ring *ring)
 	amdgpu_ring_write(ring, (0xfff << 16) | 10); /* retry count, poll interval */
 }
 
+static void cik_sdma_ring_emit_hdp_invalidate(struct amdgpu_ring *ring)
+{
+	amdgpu_ring_write(ring, SDMA_PACKET(SDMA_OPCODE_SRBM_WRITE, 0, 0xf000));
+	amdgpu_ring_write(ring, mmHDP_DEBUG0);
+	amdgpu_ring_write(ring, 1);
+}
+
 /**
  * cik_sdma_ring_emit_fence - emit a fence on the DMA ring
  *
@@ -636,8 +643,7 @@ static int cik_sdma_ring_test_ib(struct amdgpu_ring *ring)
 	ib.ptr[3] = 1;
 	ib.ptr[4] = 0xDEADBEEF;
 	ib.length_dw = 5;
-	r = amdgpu_ib_schedule(ring, 1, &ib, AMDGPU_FENCE_OWNER_UNDEFINED,
-			       NULL, &f);
+	r = amdgpu_ib_schedule(ring, 1, &ib, NULL, &f);
 	if (r)
 		goto err1;
 
@@ -663,7 +669,8 @@ static int cik_sdma_ring_test_ib(struct amdgpu_ring *ring)
 
 err1:
 	fence_put(f);
-	amdgpu_ib_free(adev, &ib);
+	amdgpu_ib_free(adev, &ib, NULL);
+	fence_put(f);
 err0:
 	amdgpu_wb_free(adev, index);
 	return r;
@@ -816,6 +823,30 @@ static void cik_sdma_ring_pad_ib(struct amdgpu_ring *ring, struct amdgpu_ib *ib)
 }
 
 /**
+ * cik_sdma_ring_emit_pipeline_sync - sync the pipeline
+ *
+ * @ring: amdgpu_ring pointer
+ *
+ * Make sure all previous operations are completed (CIK).
+ */
+static void cik_sdma_ring_emit_pipeline_sync(struct amdgpu_ring *ring)
+{
+	uint32_t seq = ring->fence_drv.sync_seq;
+	uint64_t addr = ring->fence_drv.gpu_addr;
+
+	/* wait for idle */
+	amdgpu_ring_write(ring, SDMA_PACKET(SDMA_OPCODE_POLL_REG_MEM, 0,
+					    SDMA_POLL_REG_MEM_EXTRA_OP(0) |
+					    SDMA_POLL_REG_MEM_EXTRA_FUNC(3) | /* equal */
+					    SDMA_POLL_REG_MEM_EXTRA_M));
+	amdgpu_ring_write(ring, addr & 0xfffffffc);
+	amdgpu_ring_write(ring, upper_32_bits(addr) & 0xffffffff);
+	amdgpu_ring_write(ring, seq); /* reference */
+	amdgpu_ring_write(ring, 0xfffffff); /* mask */
+	amdgpu_ring_write(ring, (0xfff << 16) | 4); /* retry count, poll interval */
+}
+
+/**
  * cik_sdma_ring_emit_vm_flush - cik vm flush using sDMA
  *
  * @ring: amdgpu_ring pointer
@@ -856,7 +887,7 @@ static void cik_enable_sdma_mgcg(struct amdgpu_device *adev,
 {
 	u32 orig, data;
 
-	if (enable && (adev->cg_flags & AMDGPU_CG_SUPPORT_SDMA_MGCG)) {
+	if (enable && (adev->cg_flags & AMD_CG_SUPPORT_SDMA_MGCG)) {
 		WREG32(mmSDMA0_CLK_CTRL + SDMA0_REGISTER_OFFSET, 0x00000100);
 		WREG32(mmSDMA0_CLK_CTRL + SDMA1_REGISTER_OFFSET, 0x00000100);
 	} else {
@@ -877,7 +908,7 @@ static void cik_enable_sdma_mgls(struct amdgpu_device *adev,
 {
 	u32 orig, data;
 
-	if (enable && (adev->cg_flags & AMDGPU_CG_SUPPORT_SDMA_LS)) {
+	if (enable && (adev->cg_flags & AMD_CG_SUPPORT_SDMA_LS)) {
 		orig = data = RREG32(mmSDMA0_POWER_CNTL + SDMA0_REGISTER_OFFSET);
 		data |= 0x100;
 		if (orig != data)
@@ -1270,8 +1301,10 @@ static const struct amdgpu_ring_funcs cik_sdma_ring_funcs = {
 	.parse_cs = NULL,
 	.emit_ib = cik_sdma_ring_emit_ib,
 	.emit_fence = cik_sdma_ring_emit_fence,
+	.emit_pipeline_sync = cik_sdma_ring_emit_pipeline_sync,
 	.emit_vm_flush = cik_sdma_ring_emit_vm_flush,
 	.emit_hdp_flush = cik_sdma_ring_emit_hdp_flush,
+	.emit_hdp_invalidate = cik_sdma_ring_emit_hdp_invalidate,
 	.test_ring = cik_sdma_ring_test_ring,
 	.test_ib = cik_sdma_ring_test_ib,
 	.insert_nop = cik_sdma_ring_insert_nop,
diff --git a/drivers/gpu/drm/amd/amdgpu/cikd.h b/drivers/gpu/drm/amd/amdgpu/cikd.h
index 7f6d457f250a..60d4493206dd 100644
--- a/drivers/gpu/drm/amd/amdgpu/cikd.h
+++ b/drivers/gpu/drm/amd/amdgpu/cikd.h
@@ -46,9 +46,6 @@
 #define BONAIRE_GB_ADDR_CONFIG_GOLDEN        0x12010001
 #define HAWAII_GB_ADDR_CONFIG_GOLDEN         0x12011003
 
-#define CIK_RB_BITMAP_WIDTH_PER_SH     2
-#define HAWAII_RB_BITMAP_WIDTH_PER_SH  4
-
 #define AMDGPU_NUM_OF_VMIDS	8
 
 #define		PIPEID(x)					((x) << 0)
diff --git a/drivers/gpu/drm/amd/amdgpu/cz_dpm.c b/drivers/gpu/drm/amd/amdgpu/cz_dpm.c
index 4dd17f2dd905..e7ef2261ff4a 100644
--- a/drivers/gpu/drm/amd/amdgpu/cz_dpm.c
+++ b/drivers/gpu/drm/amd/amdgpu/cz_dpm.c
@@ -445,13 +445,13 @@ static int cz_dpm_init(struct amdgpu_device *adev)
 	pi->gfx_pg_threshold = 500;
 	pi->caps_fps = true;
 	/* uvd */
-	pi->caps_uvd_pg = (adev->pg_flags & AMDGPU_PG_SUPPORT_UVD) ? true : false;
+	pi->caps_uvd_pg = (adev->pg_flags & AMD_PG_SUPPORT_UVD) ? true : false;
 	pi->caps_uvd_dpm = true;
 	/* vce */
-	pi->caps_vce_pg = (adev->pg_flags & AMDGPU_PG_SUPPORT_VCE) ? true : false;
+	pi->caps_vce_pg = (adev->pg_flags & AMD_PG_SUPPORT_VCE) ? true : false;
 	pi->caps_vce_dpm = true;
 	/* acp */
-	pi->caps_acp_pg = (adev->pg_flags & AMDGPU_PG_SUPPORT_ACP) ? true : false;
+	pi->caps_acp_pg = (adev->pg_flags & AMD_PG_SUPPORT_ACP) ? true : false;
 	pi->caps_acp_dpm = true;
 
 	pi->caps_stable_power_state = false;
@@ -2202,8 +2202,7 @@ static void cz_dpm_powergate_vce(struct amdgpu_device *adev, bool gate)
 							    AMD_PG_STATE_GATE);
 
 				cz_enable_vce_dpm(adev, false);
-				/* TODO: to figure out why vce can't be poweroff. */
-				/* cz_send_msg_to_smc(adev, PPSMC_MSG_VCEPowerOFF); */
+				cz_send_msg_to_smc(adev, PPSMC_MSG_VCEPowerOFF);
 				pi->vce_power_gated = true;
 			} else {
 				cz_send_msg_to_smc(adev, PPSMC_MSG_VCEPowerON);
@@ -2226,10 +2225,8 @@ static void cz_dpm_powergate_vce(struct amdgpu_device *adev, bool gate)
 		}
 	} else { /*pi->caps_vce_pg*/
 		cz_update_vce_dpm(adev);
-		cz_enable_vce_dpm(adev, true);
+		cz_enable_vce_dpm(adev, !gate);
 	}
-
-	return;
 }
 
 const struct amd_ip_funcs cz_dpm_ip_funcs = {
diff --git a/drivers/gpu/drm/amd/amdgpu/dce_v10_0.c b/drivers/gpu/drm/amd/amdgpu/dce_v10_0.c
index e3ff809a0cae..6de2ce535e37 100644
--- a/drivers/gpu/drm/amd/amdgpu/dce_v10_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/dce_v10_0.c
@@ -1668,6 +1668,9 @@ static void dce_v10_0_audio_fini(struct amdgpu_device *adev)
 {
 	int i;
 
+	if (!amdgpu_audio)
+		return;
+
 	if (!adev->mode_info.audio.enabled)
 		return;
 
@@ -1973,7 +1976,7 @@ static void dce_v10_0_afmt_enable(struct drm_encoder *encoder, bool enable)
 		  enable ? "En" : "Dis", dig->afmt->offset, amdgpu_encoder->encoder_id);
 }
 
-static void dce_v10_0_afmt_init(struct amdgpu_device *adev)
+static int dce_v10_0_afmt_init(struct amdgpu_device *adev)
 {
 	int i;
 
@@ -1986,8 +1989,16 @@ static void dce_v10_0_afmt_init(struct amdgpu_device *adev)
 		if (adev->mode_info.afmt[i]) {
 			adev->mode_info.afmt[i]->offset = dig_offsets[i];
 			adev->mode_info.afmt[i]->id = i;
+		} else {
+			int j;
+			for (j = 0; j < i; j++) {
+				kfree(adev->mode_info.afmt[j]);
+				adev->mode_info.afmt[j] = NULL;
+			}
+			return -ENOMEM;
 		}
 	}
+	return 0;
 }
 
 static void dce_v10_0_afmt_fini(struct amdgpu_device *adev)
@@ -2064,8 +2075,7 @@ static int dce_v10_0_crtc_do_set_base(struct drm_crtc *crtc,
 	if (atomic) {
 		amdgpu_fb = to_amdgpu_framebuffer(fb);
 		target_fb = fb;
-	}
-	else {
+	} else {
 		amdgpu_fb = to_amdgpu_framebuffer(crtc->primary->fb);
 		target_fb = crtc->primary->fb;
 	}
@@ -2079,9 +2089,9 @@ static int dce_v10_0_crtc_do_set_base(struct drm_crtc *crtc,
 	if (unlikely(r != 0))
 		return r;
 
-	if (atomic)
+	if (atomic) {
 		fb_location = amdgpu_bo_gpu_offset(rbo);
-	else {
+	} else {
 		r = amdgpu_bo_pin(rbo, AMDGPU_GEM_DOMAIN_VRAM, &fb_location);
 		if (unlikely(r != 0)) {
 			amdgpu_bo_unreserve(rbo);
@@ -2700,13 +2710,13 @@ static void dce_v10_0_crtc_dpms(struct drm_crtc *crtc, int mode)
 		type = amdgpu_crtc_idx_to_irq_type(adev, amdgpu_crtc->crtc_id);
 		amdgpu_irq_update(adev, &adev->crtc_irq, type);
 		amdgpu_irq_update(adev, &adev->pageflip_irq, type);
-		drm_vblank_post_modeset(dev, amdgpu_crtc->crtc_id);
+		drm_vblank_on(dev, amdgpu_crtc->crtc_id);
 		dce_v10_0_crtc_load_lut(crtc);
 		break;
 	case DRM_MODE_DPMS_STANDBY:
 	case DRM_MODE_DPMS_SUSPEND:
 	case DRM_MODE_DPMS_OFF:
-		drm_vblank_pre_modeset(dev, amdgpu_crtc->crtc_id);
+		drm_vblank_off(dev, amdgpu_crtc->crtc_id);
 		if (amdgpu_crtc->enabled) {
 			dce_v10_0_vga_enable(crtc, true);
 			amdgpu_atombios_crtc_blank(crtc, ATOM_ENABLE);
@@ -2980,8 +2990,6 @@ static int dce_v10_0_sw_init(void *handle)
 	if (r)
 		return r;
 
-	adev->mode_info.mode_config_initialized = true;
-
 	adev->ddev->mode_config.funcs = &amdgpu_mode_funcs;
 
 	adev->ddev->mode_config.max_width = 16384;
@@ -3012,7 +3020,9 @@ static int dce_v10_0_sw_init(void *handle)
 		return -EINVAL;
 
 	/* setup afmt */
-	dce_v10_0_afmt_init(adev);
+	r = dce_v10_0_afmt_init(adev);
+	if (r)
+		return r;
 
 	r = dce_v10_0_audio_init(adev);
 	if (r)
@@ -3020,7 +3030,8 @@ static int dce_v10_0_sw_init(void *handle)
 
 	drm_kms_helper_poll_init(adev->ddev);
 
-	return r;
+	adev->mode_info.mode_config_initialized = true;
+	return 0;
 }
 
 static int dce_v10_0_sw_fini(void *handle)
diff --git a/drivers/gpu/drm/amd/amdgpu/dce_v11_0.c b/drivers/gpu/drm/amd/amdgpu/dce_v11_0.c
index 6b6c9b6879ae..e9ccc6b787f3 100644
--- a/drivers/gpu/drm/amd/amdgpu/dce_v11_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/dce_v11_0.c
@@ -1658,6 +1658,9 @@ static void dce_v11_0_audio_fini(struct amdgpu_device *adev)
 {
 	int i;
 
+	if (!amdgpu_audio)
+		return;
+
 	if (!adev->mode_info.audio.enabled)
 		return;
 
@@ -1963,7 +1966,7 @@ static void dce_v11_0_afmt_enable(struct drm_encoder *encoder, bool enable)
 		  enable ? "En" : "Dis", dig->afmt->offset, amdgpu_encoder->encoder_id);
 }
 
-static void dce_v11_0_afmt_init(struct amdgpu_device *adev)
+static int dce_v11_0_afmt_init(struct amdgpu_device *adev)
 {
 	int i;
 
@@ -1976,8 +1979,16 @@ static void dce_v11_0_afmt_init(struct amdgpu_device *adev)
 		if (adev->mode_info.afmt[i]) {
 			adev->mode_info.afmt[i]->offset = dig_offsets[i];
 			adev->mode_info.afmt[i]->id = i;
+		} else {
+			int j;
+			for (j = 0; j < i; j++) {
+				kfree(adev->mode_info.afmt[j]);
+				adev->mode_info.afmt[j] = NULL;
+			}
+			return -ENOMEM;
 		}
 	}
+	return 0;
 }
 
 static void dce_v11_0_afmt_fini(struct amdgpu_device *adev)
@@ -2054,8 +2065,7 @@ static int dce_v11_0_crtc_do_set_base(struct drm_crtc *crtc,
 	if (atomic) {
 		amdgpu_fb = to_amdgpu_framebuffer(fb);
 		target_fb = fb;
-	}
-	else {
+	} else {
 		amdgpu_fb = to_amdgpu_framebuffer(crtc->primary->fb);
 		target_fb = crtc->primary->fb;
 	}
@@ -2069,9 +2079,9 @@ static int dce_v11_0_crtc_do_set_base(struct drm_crtc *crtc,
 	if (unlikely(r != 0))
 		return r;
 
-	if (atomic)
+	if (atomic) {
 		fb_location = amdgpu_bo_gpu_offset(rbo);
-	else {
+	} else {
 		r = amdgpu_bo_pin(rbo, AMDGPU_GEM_DOMAIN_VRAM, &fb_location);
 		if (unlikely(r != 0)) {
 			amdgpu_bo_unreserve(rbo);
@@ -2691,13 +2701,13 @@ static void dce_v11_0_crtc_dpms(struct drm_crtc *crtc, int mode)
 		type = amdgpu_crtc_idx_to_irq_type(adev, amdgpu_crtc->crtc_id);
 		amdgpu_irq_update(adev, &adev->crtc_irq, type);
 		amdgpu_irq_update(adev, &adev->pageflip_irq, type);
-		drm_vblank_post_modeset(dev, amdgpu_crtc->crtc_id);
+		drm_vblank_on(dev, amdgpu_crtc->crtc_id);
 		dce_v11_0_crtc_load_lut(crtc);
 		break;
 	case DRM_MODE_DPMS_STANDBY:
 	case DRM_MODE_DPMS_SUSPEND:
 	case DRM_MODE_DPMS_OFF:
-		drm_vblank_pre_modeset(dev, amdgpu_crtc->crtc_id);
+		drm_vblank_off(dev, amdgpu_crtc->crtc_id);
 		if (amdgpu_crtc->enabled) {
 			dce_v11_0_vga_enable(crtc, true);
 			amdgpu_atombios_crtc_blank(crtc, ATOM_ENABLE);
@@ -2961,7 +2971,7 @@ static int dce_v11_0_sw_init(void *handle)
 	for (i = 0; i < adev->mode_info.num_crtc; i++) {
 		r = amdgpu_irq_add_id(adev, i + 1, &adev->crtc_irq);
 		if (r)
-		return r;
+			return r;
 	}
 
 	for (i = 8; i < 20; i += 2) {
@@ -2973,9 +2983,7 @@ static int dce_v11_0_sw_init(void *handle)
 	/* HPD hotplug */
 	r = amdgpu_irq_add_id(adev, 42, &adev->hpd_irq);
 	if (r)
-	return r;
-
-	adev->mode_info.mode_config_initialized = true;
+		return r;
 
 	adev->ddev->mode_config.funcs = &amdgpu_mode_funcs;
 
@@ -2994,6 +3002,7 @@ static int dce_v11_0_sw_init(void *handle)
 	adev->ddev->mode_config.max_width = 16384;
 	adev->ddev->mode_config.max_height = 16384;
 
+
 	/* allocate crtcs */
 	for (i = 0; i < adev->mode_info.num_crtc; i++) {
 		r = dce_v11_0_crtc_init(adev, i);
@@ -3007,7 +3016,9 @@ static int dce_v11_0_sw_init(void *handle)
 		return -EINVAL;
 
 	/* setup afmt */
-	dce_v11_0_afmt_init(adev);
+	r = dce_v11_0_afmt_init(adev);
+	if (r)
+		return r;
 
 	r = dce_v11_0_audio_init(adev);
 	if (r)
@@ -3015,7 +3026,8 @@ static int dce_v11_0_sw_init(void *handle)
 
 	drm_kms_helper_poll_init(adev->ddev);
 
-	return r;
+	adev->mode_info.mode_config_initialized = true;
+	return 0;
 }
 
 static int dce_v11_0_sw_fini(void *handle)
diff --git a/drivers/gpu/drm/amd/amdgpu/dce_v8_0.c b/drivers/gpu/drm/amd/amdgpu/dce_v8_0.c
index 56bea36a6b18..e56b55d8c280 100644
--- a/drivers/gpu/drm/amd/amdgpu/dce_v8_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/dce_v8_0.c
@@ -1639,6 +1639,9 @@ static void dce_v8_0_audio_fini(struct amdgpu_device *adev)
 {
 	int i;
 
+	if (!amdgpu_audio)
+		return;
+
 	if (!adev->mode_info.audio.enabled)
 		return;
 
@@ -1910,7 +1913,7 @@ static void dce_v8_0_afmt_enable(struct drm_encoder *encoder, bool enable)
 		  enable ? "En" : "Dis", dig->afmt->offset, amdgpu_encoder->encoder_id);
 }
 
-static void dce_v8_0_afmt_init(struct amdgpu_device *adev)
+static int dce_v8_0_afmt_init(struct amdgpu_device *adev)
 {
 	int i;
 
@@ -1923,8 +1926,16 @@ static void dce_v8_0_afmt_init(struct amdgpu_device *adev)
 		if (adev->mode_info.afmt[i]) {
 			adev->mode_info.afmt[i]->offset = dig_offsets[i];
 			adev->mode_info.afmt[i]->id = i;
+		} else {
+			int j;
+			for (j = 0; j < i; j++) {
+				kfree(adev->mode_info.afmt[j]);
+				adev->mode_info.afmt[j] = NULL;
+			}
+			return -ENOMEM;
 		}
 	}
+	return 0;
 }
 
 static void dce_v8_0_afmt_fini(struct amdgpu_device *adev)
@@ -2001,8 +2012,7 @@ static int dce_v8_0_crtc_do_set_base(struct drm_crtc *crtc,
 	if (atomic) {
 		amdgpu_fb = to_amdgpu_framebuffer(fb);
 		target_fb = fb;
-	}
-	else {
+	} else {
 		amdgpu_fb = to_amdgpu_framebuffer(crtc->primary->fb);
 		target_fb = crtc->primary->fb;
 	}
@@ -2016,9 +2026,9 @@ static int dce_v8_0_crtc_do_set_base(struct drm_crtc *crtc,
 	if (unlikely(r != 0))
 		return r;
 
-	if (atomic)
+	if (atomic) {
 		fb_location = amdgpu_bo_gpu_offset(rbo);
-	else {
+	} else {
 		r = amdgpu_bo_pin(rbo, AMDGPU_GEM_DOMAIN_VRAM, &fb_location);
 		if (unlikely(r != 0)) {
 			amdgpu_bo_unreserve(rbo);
@@ -2612,13 +2622,13 @@ static void dce_v8_0_crtc_dpms(struct drm_crtc *crtc, int mode)
 		type = amdgpu_crtc_idx_to_irq_type(adev, amdgpu_crtc->crtc_id);
 		amdgpu_irq_update(adev, &adev->crtc_irq, type);
 		amdgpu_irq_update(adev, &adev->pageflip_irq, type);
-		drm_vblank_post_modeset(dev, amdgpu_crtc->crtc_id);
+		drm_vblank_on(dev, amdgpu_crtc->crtc_id);
 		dce_v8_0_crtc_load_lut(crtc);
 		break;
 	case DRM_MODE_DPMS_STANDBY:
 	case DRM_MODE_DPMS_SUSPEND:
 	case DRM_MODE_DPMS_OFF:
-		drm_vblank_pre_modeset(dev, amdgpu_crtc->crtc_id);
+		drm_vblank_off(dev, amdgpu_crtc->crtc_id);
 		if (amdgpu_crtc->enabled) {
 			dce_v8_0_vga_enable(crtc, true);
 			amdgpu_atombios_crtc_blank(crtc, ATOM_ENABLE);
@@ -2890,8 +2900,6 @@ static int dce_v8_0_sw_init(void *handle)
 	if (r)
 		return r;
 
-	adev->mode_info.mode_config_initialized = true;
-
 	adev->ddev->mode_config.funcs = &amdgpu_mode_funcs;
 
 	adev->ddev->mode_config.max_width = 16384;
@@ -2922,7 +2930,9 @@ static int dce_v8_0_sw_init(void *handle)
 		return -EINVAL;
 
 	/* setup afmt */
-	dce_v8_0_afmt_init(adev);
+	r = dce_v8_0_afmt_init(adev);
+	if (r)
+		return r;
 
 	r = dce_v8_0_audio_init(adev);
 	if (r)
@@ -2930,7 +2940,8 @@ static int dce_v8_0_sw_init(void *handle)
 
 	drm_kms_helper_poll_init(adev->ddev);
 
-	return r;
+	adev->mode_info.mode_config_initialized = true;
+	return 0;
 }
 
 static int dce_v8_0_sw_fini(void *handle)
diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v7_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v7_0.c
index 250bcbce7fdc..bb8709066fd8 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfx_v7_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfx_v7_0.c
@@ -1635,30 +1635,25 @@ static u32 gfx_v7_0_get_rb_active_bitmap(struct amdgpu_device *adev)
 static void gfx_v7_0_setup_rb(struct amdgpu_device *adev)
 {
 	int i, j;
-	u32 data, tmp, num_rbs = 0;
+	u32 data;
 	u32 active_rbs = 0;
+	u32 rb_bitmap_width_per_sh = adev->gfx.config.max_backends_per_se /
+					adev->gfx.config.max_sh_per_se;
 
 	mutex_lock(&adev->grbm_idx_mutex);
 	for (i = 0; i < adev->gfx.config.max_shader_engines; i++) {
 		for (j = 0; j < adev->gfx.config.max_sh_per_se; j++) {
 			gfx_v7_0_select_se_sh(adev, i, j);
 			data = gfx_v7_0_get_rb_active_bitmap(adev);
-			if (adev->asic_type == CHIP_HAWAII)
-				active_rbs |= data << ((i * adev->gfx.config.max_sh_per_se + j) *
-						       HAWAII_RB_BITMAP_WIDTH_PER_SH);
-			else
-				active_rbs |= data << ((i * adev->gfx.config.max_sh_per_se + j) *
-						       CIK_RB_BITMAP_WIDTH_PER_SH);
+			active_rbs |= data << ((i * adev->gfx.config.max_sh_per_se + j) *
+					       rb_bitmap_width_per_sh);
 		}
 	}
 	gfx_v7_0_select_se_sh(adev, 0xffffffff, 0xffffffff);
 	mutex_unlock(&adev->grbm_idx_mutex);
 
 	adev->gfx.config.backend_enable_mask = active_rbs;
-	tmp = active_rbs;
-	while (tmp >>= 1)
-		num_rbs++;
-	adev->gfx.config.num_rbs = num_rbs;
+	adev->gfx.config.num_rbs = hweight32(active_rbs);
 }
 
 /**
@@ -1930,6 +1925,25 @@ static void gfx_v7_0_ring_emit_hdp_flush(struct amdgpu_ring *ring)
 }
 
 /**
+ * gfx_v7_0_ring_emit_hdp_invalidate - emit an hdp invalidate on the cp
+ *
+ * @adev: amdgpu_device pointer
+ * @ridx: amdgpu ring index
+ *
+ * Emits an hdp invalidate on the cp.
+ */
+static void gfx_v7_0_ring_emit_hdp_invalidate(struct amdgpu_ring *ring)
+{
+	amdgpu_ring_write(ring, PACKET3(PACKET3_WRITE_DATA, 3));
+	amdgpu_ring_write(ring, (WRITE_DATA_ENGINE_SEL(0) |
+				 WRITE_DATA_DST_SEL(0) |
+				 WR_CONFIRM));
+	amdgpu_ring_write(ring, mmHDP_DEBUG0);
+	amdgpu_ring_write(ring, 0);
+	amdgpu_ring_write(ring, 1);
+}
+
+/**
  * gfx_v7_0_ring_emit_fence_gfx - emit a fence on the gfx ring
  *
  * @adev: amdgpu_device pointer
@@ -2046,8 +2060,7 @@ static void gfx_v7_0_ring_emit_ib_gfx(struct amdgpu_ring *ring,
 	else
 		header = PACKET3(PACKET3_INDIRECT_BUFFER, 2);
 
-	control |= ib->length_dw |
-		(ib->vm ? (ib->vm->ids[ring->idx].id << 24) : 0);
+	control |= ib->length_dw | (ib->vm_id << 24);
 
 	amdgpu_ring_write(ring, header);
 	amdgpu_ring_write(ring,
@@ -2075,8 +2088,7 @@ static void gfx_v7_0_ring_emit_ib_compute(struct amdgpu_ring *ring,
 
 	header = PACKET3(PACKET3_INDIRECT_BUFFER, 2);
 
-	control |= ib->length_dw |
-			   (ib->vm ? (ib->vm->ids[ring->idx].id << 24) : 0);
+	control |= ib->length_dw | (ib->vm_id << 24);
 
 	amdgpu_ring_write(ring, header);
 	amdgpu_ring_write(ring,
@@ -2124,8 +2136,7 @@ static int gfx_v7_0_ring_test_ib(struct amdgpu_ring *ring)
 	ib.ptr[2] = 0xDEADBEEF;
 	ib.length_dw = 3;
 
-	r = amdgpu_ib_schedule(ring, 1, &ib, AMDGPU_FENCE_OWNER_UNDEFINED,
-			       NULL, &f);
+	r = amdgpu_ib_schedule(ring, 1, &ib, NULL, &f);
 	if (r)
 		goto err2;
 
@@ -2152,7 +2163,8 @@ static int gfx_v7_0_ring_test_ib(struct amdgpu_ring *ring)
 
 err2:
 	fence_put(f);
-	amdgpu_ib_free(adev, &ib);
+	amdgpu_ib_free(adev, &ib, NULL);
+	fence_put(f);
 err1:
 	amdgpu_gfx_scratch_free(adev, scratch);
 	return r;
@@ -3030,6 +3042,26 @@ static int gfx_v7_0_cp_resume(struct amdgpu_device *adev)
 	return 0;
 }
 
+/**
+ * gfx_v7_0_ring_emit_vm_flush - cik vm flush using the CP
+ *
+ * @ring: the ring to emmit the commands to
+ *
+ * Sync the command pipeline with the PFP. E.g. wait for everything
+ * to be completed.
+ */
+static void gfx_v7_0_ring_emit_pipeline_sync(struct amdgpu_ring *ring)
+{
+	int usepfp = (ring->type == AMDGPU_RING_TYPE_GFX);
+	if (usepfp) {
+		/* synce CE with ME to prevent CE fetch CEIB before context switch done */
+		amdgpu_ring_write(ring, PACKET3(PACKET3_SWITCH_BUFFER, 0));
+		amdgpu_ring_write(ring, 0);
+		amdgpu_ring_write(ring, PACKET3(PACKET3_SWITCH_BUFFER, 0));
+		amdgpu_ring_write(ring, 0);
+	}
+}
+
 /*
  * vm
  * VMID 0 is the physical GPU addresses as used by the kernel.
@@ -3048,13 +3080,18 @@ static void gfx_v7_0_ring_emit_vm_flush(struct amdgpu_ring *ring,
 					unsigned vm_id, uint64_t pd_addr)
 {
 	int usepfp = (ring->type == AMDGPU_RING_TYPE_GFX);
-	if (usepfp) {
-		/* synce CE with ME to prevent CE fetch CEIB before context switch done */
-		amdgpu_ring_write(ring, PACKET3(PACKET3_SWITCH_BUFFER, 0));
-		amdgpu_ring_write(ring, 0);
-		amdgpu_ring_write(ring, PACKET3(PACKET3_SWITCH_BUFFER, 0));
-		amdgpu_ring_write(ring, 0);
-	}
+	uint32_t seq = ring->fence_drv.sync_seq;
+	uint64_t addr = ring->fence_drv.gpu_addr;
+
+	amdgpu_ring_write(ring, PACKET3(PACKET3_WAIT_REG_MEM, 5));
+	amdgpu_ring_write(ring, (WAIT_REG_MEM_MEM_SPACE(1) | /* memory */
+				 WAIT_REG_MEM_FUNCTION(3) | /* equal */
+				 WAIT_REG_MEM_ENGINE(usepfp)));   /* pfp or me */
+	amdgpu_ring_write(ring, addr & 0xfffffffc);
+	amdgpu_ring_write(ring, upper_32_bits(addr) & 0xffffffff);
+	amdgpu_ring_write(ring, seq);
+	amdgpu_ring_write(ring, 0xffffffff);
+	amdgpu_ring_write(ring, 4); /* poll interval */
 
 	amdgpu_ring_write(ring, PACKET3(PACKET3_WRITE_DATA, 3));
 	amdgpu_ring_write(ring, (WRITE_DATA_ENGINE_SEL(usepfp) |
@@ -3529,7 +3566,7 @@ static void gfx_v7_0_enable_cgcg(struct amdgpu_device *adev, bool enable)
 
 	orig = data = RREG32(mmRLC_CGCG_CGLS_CTRL);
 
-	if (enable && (adev->cg_flags & AMDGPU_CG_SUPPORT_GFX_CGCG)) {
+	if (enable && (adev->cg_flags & AMD_CG_SUPPORT_GFX_CGCG)) {
 		gfx_v7_0_enable_gui_idle_interrupt(adev, true);
 
 		tmp = gfx_v7_0_halt_rlc(adev);
@@ -3567,9 +3604,9 @@ static void gfx_v7_0_enable_mgcg(struct amdgpu_device *adev, bool enable)
 {
 	u32 data, orig, tmp = 0;
 
-	if (enable && (adev->cg_flags & AMDGPU_CG_SUPPORT_GFX_MGCG)) {
-		if (adev->cg_flags & AMDGPU_CG_SUPPORT_GFX_MGLS) {
-			if (adev->cg_flags & AMDGPU_CG_SUPPORT_GFX_CP_LS) {
+	if (enable && (adev->cg_flags & AMD_CG_SUPPORT_GFX_MGCG)) {
+		if (adev->cg_flags & AMD_CG_SUPPORT_GFX_MGLS) {
+			if (adev->cg_flags & AMD_CG_SUPPORT_GFX_CP_LS) {
 				orig = data = RREG32(mmCP_MEM_SLP_CNTL);
 				data |= CP_MEM_SLP_CNTL__CP_MEM_LS_EN_MASK;
 				if (orig != data)
@@ -3596,14 +3633,14 @@ static void gfx_v7_0_enable_mgcg(struct amdgpu_device *adev, bool enable)
 
 		gfx_v7_0_update_rlc(adev, tmp);
 
-		if (adev->cg_flags & AMDGPU_CG_SUPPORT_GFX_CGTS) {
+		if (adev->cg_flags & AMD_CG_SUPPORT_GFX_CGTS) {
 			orig = data = RREG32(mmCGTS_SM_CTRL_REG);
 			data &= ~CGTS_SM_CTRL_REG__SM_MODE_MASK;
 			data |= (0x2 << CGTS_SM_CTRL_REG__SM_MODE__SHIFT);
 			data |= CGTS_SM_CTRL_REG__SM_MODE_ENABLE_MASK;
 			data &= ~CGTS_SM_CTRL_REG__OVERRIDE_MASK;
-			if ((adev->cg_flags & AMDGPU_CG_SUPPORT_GFX_MGLS) &&
-			    (adev->cg_flags & AMDGPU_CG_SUPPORT_GFX_CGTS_LS))
+			if ((adev->cg_flags & AMD_CG_SUPPORT_GFX_MGLS) &&
+			    (adev->cg_flags & AMD_CG_SUPPORT_GFX_CGTS_LS))
 				data &= ~CGTS_SM_CTRL_REG__LS_OVERRIDE_MASK;
 			data &= ~CGTS_SM_CTRL_REG__ON_MONITOR_ADD_MASK;
 			data |= CGTS_SM_CTRL_REG__ON_MONITOR_ADD_EN_MASK;
@@ -3669,7 +3706,7 @@ static void gfx_v7_0_enable_sclk_slowdown_on_pu(struct amdgpu_device *adev,
 	u32 data, orig;
 
 	orig = data = RREG32(mmRLC_PG_CNTL);
-	if (enable && (adev->pg_flags & AMDGPU_PG_SUPPORT_RLC_SMU_HS))
+	if (enable && (adev->pg_flags & AMD_PG_SUPPORT_RLC_SMU_HS))
 		data |= RLC_PG_CNTL__SMU_CLK_SLOWDOWN_ON_PU_ENABLE_MASK;
 	else
 		data &= ~RLC_PG_CNTL__SMU_CLK_SLOWDOWN_ON_PU_ENABLE_MASK;
@@ -3683,7 +3720,7 @@ static void gfx_v7_0_enable_sclk_slowdown_on_pd(struct amdgpu_device *adev,
 	u32 data, orig;
 
 	orig = data = RREG32(mmRLC_PG_CNTL);
-	if (enable && (adev->pg_flags & AMDGPU_PG_SUPPORT_RLC_SMU_HS))
+	if (enable && (adev->pg_flags & AMD_PG_SUPPORT_RLC_SMU_HS))
 		data |= RLC_PG_CNTL__SMU_CLK_SLOWDOWN_ON_PD_ENABLE_MASK;
 	else
 		data &= ~RLC_PG_CNTL__SMU_CLK_SLOWDOWN_ON_PD_ENABLE_MASK;
@@ -3696,7 +3733,7 @@ static void gfx_v7_0_enable_cp_pg(struct amdgpu_device *adev, bool enable)
 	u32 data, orig;
 
 	orig = data = RREG32(mmRLC_PG_CNTL);
-	if (enable && (adev->pg_flags & AMDGPU_PG_SUPPORT_CP))
+	if (enable && (adev->pg_flags & AMD_PG_SUPPORT_CP))
 		data &= ~0x8000;
 	else
 		data |= 0x8000;
@@ -3709,7 +3746,7 @@ static void gfx_v7_0_enable_gds_pg(struct amdgpu_device *adev, bool enable)
 	u32 data, orig;
 
 	orig = data = RREG32(mmRLC_PG_CNTL);
-	if (enable && (adev->pg_flags & AMDGPU_PG_SUPPORT_GDS))
+	if (enable && (adev->pg_flags & AMD_PG_SUPPORT_GDS))
 		data &= ~0x2000;
 	else
 		data |= 0x2000;
@@ -3790,7 +3827,7 @@ static void gfx_v7_0_enable_gfx_cgpg(struct amdgpu_device *adev,
 {
 	u32 data, orig;
 
-	if (enable && (adev->pg_flags & AMDGPU_PG_SUPPORT_GFX_PG)) {
+	if (enable && (adev->pg_flags & AMD_PG_SUPPORT_GFX_PG)) {
 		orig = data = RREG32(mmRLC_PG_CNTL);
 		data |= RLC_PG_CNTL__GFX_POWER_GATING_ENABLE_MASK;
 		if (orig != data)
@@ -3825,8 +3862,7 @@ static u32 gfx_v7_0_get_cu_active_bitmap(struct amdgpu_device *adev)
 	data &= CC_GC_SHADER_ARRAY_CONFIG__INACTIVE_CUS_MASK;
 	data >>= CC_GC_SHADER_ARRAY_CONFIG__INACTIVE_CUS__SHIFT;
 
-	mask = gfx_v7_0_create_bitmask(adev->gfx.config.max_backends_per_se /
-				       adev->gfx.config.max_sh_per_se);
+	mask = gfx_v7_0_create_bitmask(adev->gfx.config.max_cu_per_sh);
 
 	return (~data) & mask;
 }
@@ -3854,7 +3890,7 @@ static void gfx_v7_0_enable_gfx_static_mgpg(struct amdgpu_device *adev,
 	u32 data, orig;
 
 	orig = data = RREG32(mmRLC_PG_CNTL);
-	if (enable && (adev->pg_flags & AMDGPU_PG_SUPPORT_GFX_SMG))
+	if (enable && (adev->pg_flags & AMD_PG_SUPPORT_GFX_SMG))
 		data |= RLC_PG_CNTL__STATIC_PER_CU_PG_ENABLE_MASK;
 	else
 		data &= ~RLC_PG_CNTL__STATIC_PER_CU_PG_ENABLE_MASK;
@@ -3868,7 +3904,7 @@ static void gfx_v7_0_enable_gfx_dynamic_mgpg(struct amdgpu_device *adev,
 	u32 data, orig;
 
 	orig = data = RREG32(mmRLC_PG_CNTL);
-	if (enable && (adev->pg_flags & AMDGPU_PG_SUPPORT_GFX_DMG))
+	if (enable && (adev->pg_flags & AMD_PG_SUPPORT_GFX_DMG))
 		data |= RLC_PG_CNTL__DYN_PER_CU_PG_ENABLE_MASK;
 	else
 		data &= ~RLC_PG_CNTL__DYN_PER_CU_PG_ENABLE_MASK;
@@ -4035,15 +4071,15 @@ static void gfx_v7_0_get_csb_buffer(struct amdgpu_device *adev,
 
 static void gfx_v7_0_init_pg(struct amdgpu_device *adev)
 {
-	if (adev->pg_flags & (AMDGPU_PG_SUPPORT_GFX_PG |
-			      AMDGPU_PG_SUPPORT_GFX_SMG |
-			      AMDGPU_PG_SUPPORT_GFX_DMG |
-			      AMDGPU_PG_SUPPORT_CP |
-			      AMDGPU_PG_SUPPORT_GDS |
-			      AMDGPU_PG_SUPPORT_RLC_SMU_HS)) {
+	if (adev->pg_flags & (AMD_PG_SUPPORT_GFX_PG |
+			      AMD_PG_SUPPORT_GFX_SMG |
+			      AMD_PG_SUPPORT_GFX_DMG |
+			      AMD_PG_SUPPORT_CP |
+			      AMD_PG_SUPPORT_GDS |
+			      AMD_PG_SUPPORT_RLC_SMU_HS)) {
 		gfx_v7_0_enable_sclk_slowdown_on_pu(adev, true);
 		gfx_v7_0_enable_sclk_slowdown_on_pd(adev, true);
-		if (adev->pg_flags & AMDGPU_PG_SUPPORT_GFX_PG) {
+		if (adev->pg_flags & AMD_PG_SUPPORT_GFX_PG) {
 			gfx_v7_0_init_gfx_cgpg(adev);
 			gfx_v7_0_enable_cp_pg(adev, true);
 			gfx_v7_0_enable_gds_pg(adev, true);
@@ -4055,14 +4091,14 @@ static void gfx_v7_0_init_pg(struct amdgpu_device *adev)
 
 static void gfx_v7_0_fini_pg(struct amdgpu_device *adev)
 {
-	if (adev->pg_flags & (AMDGPU_PG_SUPPORT_GFX_PG |
-			      AMDGPU_PG_SUPPORT_GFX_SMG |
-			      AMDGPU_PG_SUPPORT_GFX_DMG |
-			      AMDGPU_PG_SUPPORT_CP |
-			      AMDGPU_PG_SUPPORT_GDS |
-			      AMDGPU_PG_SUPPORT_RLC_SMU_HS)) {
+	if (adev->pg_flags & (AMD_PG_SUPPORT_GFX_PG |
+			      AMD_PG_SUPPORT_GFX_SMG |
+			      AMD_PG_SUPPORT_GFX_DMG |
+			      AMD_PG_SUPPORT_CP |
+			      AMD_PG_SUPPORT_GDS |
+			      AMD_PG_SUPPORT_RLC_SMU_HS)) {
 		gfx_v7_0_update_gfx_pg(adev, false);
-		if (adev->pg_flags & AMDGPU_PG_SUPPORT_GFX_PG) {
+		if (adev->pg_flags & AMD_PG_SUPPORT_GFX_PG) {
 			gfx_v7_0_enable_cp_pg(adev, false);
 			gfx_v7_0_enable_gds_pg(adev, false);
 		}
@@ -5097,14 +5133,14 @@ static int gfx_v7_0_set_powergating_state(void *handle,
 	if (state == AMD_PG_STATE_GATE)
 		gate = true;
 
-	if (adev->pg_flags & (AMDGPU_PG_SUPPORT_GFX_PG |
-			      AMDGPU_PG_SUPPORT_GFX_SMG |
-			      AMDGPU_PG_SUPPORT_GFX_DMG |
-			      AMDGPU_PG_SUPPORT_CP |
-			      AMDGPU_PG_SUPPORT_GDS |
-			      AMDGPU_PG_SUPPORT_RLC_SMU_HS)) {
+	if (adev->pg_flags & (AMD_PG_SUPPORT_GFX_PG |
+			      AMD_PG_SUPPORT_GFX_SMG |
+			      AMD_PG_SUPPORT_GFX_DMG |
+			      AMD_PG_SUPPORT_CP |
+			      AMD_PG_SUPPORT_GDS |
+			      AMD_PG_SUPPORT_RLC_SMU_HS)) {
 		gfx_v7_0_update_gfx_pg(adev, gate);
-		if (adev->pg_flags & AMDGPU_PG_SUPPORT_GFX_PG) {
+		if (adev->pg_flags & AMD_PG_SUPPORT_GFX_PG) {
 			gfx_v7_0_enable_cp_pg(adev, gate);
 			gfx_v7_0_enable_gds_pg(adev, gate);
 		}
@@ -5137,9 +5173,11 @@ static const struct amdgpu_ring_funcs gfx_v7_0_ring_funcs_gfx = {
 	.parse_cs = NULL,
 	.emit_ib = gfx_v7_0_ring_emit_ib_gfx,
 	.emit_fence = gfx_v7_0_ring_emit_fence_gfx,
+	.emit_pipeline_sync = gfx_v7_0_ring_emit_pipeline_sync,
 	.emit_vm_flush = gfx_v7_0_ring_emit_vm_flush,
 	.emit_gds_switch = gfx_v7_0_ring_emit_gds_switch,
 	.emit_hdp_flush = gfx_v7_0_ring_emit_hdp_flush,
+	.emit_hdp_invalidate = gfx_v7_0_ring_emit_hdp_invalidate,
 	.test_ring = gfx_v7_0_ring_test_ring,
 	.test_ib = gfx_v7_0_ring_test_ib,
 	.insert_nop = amdgpu_ring_insert_nop,
@@ -5153,9 +5191,11 @@ static const struct amdgpu_ring_funcs gfx_v7_0_ring_funcs_compute = {
 	.parse_cs = NULL,
 	.emit_ib = gfx_v7_0_ring_emit_ib_compute,
 	.emit_fence = gfx_v7_0_ring_emit_fence_compute,
+	.emit_pipeline_sync = gfx_v7_0_ring_emit_pipeline_sync,
 	.emit_vm_flush = gfx_v7_0_ring_emit_vm_flush,
 	.emit_gds_switch = gfx_v7_0_ring_emit_gds_switch,
 	.emit_hdp_flush = gfx_v7_0_ring_emit_hdp_flush,
+	.emit_hdp_invalidate = gfx_v7_0_ring_emit_hdp_invalidate,
 	.test_ring = gfx_v7_0_ring_test_ring,
 	.test_ib = gfx_v7_0_ring_test_ib,
 	.insert_nop = amdgpu_ring_insert_nop,
@@ -5237,6 +5277,8 @@ int gfx_v7_0_get_cu_info(struct amdgpu_device *adev,
 	if (!adev || !cu_info)
 		return -EINVAL;
 
+	memset(cu_info, 0, sizeof(*cu_info));
+
 	mutex_lock(&adev->grbm_idx_mutex);
 	for (i = 0; i < adev->gfx.config.max_shader_engines; i++) {
 		for (j = 0; j < adev->gfx.config.max_sh_per_se; j++) {
diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c
index 10c865087d0a..f0c7b3596480 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c
@@ -706,8 +706,7 @@ static int gfx_v8_0_ring_test_ib(struct amdgpu_ring *ring)
 	ib.ptr[2] = 0xDEADBEEF;
 	ib.length_dw = 3;
 
-	r = amdgpu_ib_schedule(ring, 1, &ib, AMDGPU_FENCE_OWNER_UNDEFINED,
-			       NULL, &f);
+	r = amdgpu_ib_schedule(ring, 1, &ib, NULL, &f);
 	if (r)
 		goto err2;
 
@@ -733,7 +732,8 @@ static int gfx_v8_0_ring_test_ib(struct amdgpu_ring *ring)
 	}
 err2:
 	fence_put(f);
-	amdgpu_ib_free(adev, &ib);
+	amdgpu_ib_free(adev, &ib, NULL);
+	fence_put(f);
 err1:
 	amdgpu_gfx_scratch_free(adev, scratch);
 	return r;
@@ -1262,8 +1262,7 @@ static int gfx_v8_0_do_edc_gpr_workarounds(struct amdgpu_device *adev)
 	ib.ptr[ib.length_dw++] = EVENT_TYPE(7) | EVENT_INDEX(4);
 
 	/* shedule the ib on the ring */
-	r = amdgpu_ib_schedule(ring, 1, &ib, AMDGPU_FENCE_OWNER_UNDEFINED,
-			       NULL, &f);
+	r = amdgpu_ib_schedule(ring, 1, &ib, NULL, &f);
 	if (r) {
 		DRM_ERROR("amdgpu: ib submit failed (%d).\n", r);
 		goto fail;
@@ -1291,7 +1290,8 @@ static int gfx_v8_0_do_edc_gpr_workarounds(struct amdgpu_device *adev)
 
 fail:
 	fence_put(f);
-	amdgpu_ib_free(adev, &ib);
+	amdgpu_ib_free(adev, &ib, NULL);
+	fence_put(f);
 
 	return r;
 }
@@ -2613,8 +2613,10 @@ static u32 gfx_v8_0_get_rb_active_bitmap(struct amdgpu_device *adev)
 static void gfx_v8_0_setup_rb(struct amdgpu_device *adev)
 {
 	int i, j;
-	u32 data, tmp, num_rbs = 0;
+	u32 data;
 	u32 active_rbs = 0;
+	u32 rb_bitmap_width_per_sh = adev->gfx.config.max_backends_per_se /
+					adev->gfx.config.max_sh_per_se;
 
 	mutex_lock(&adev->grbm_idx_mutex);
 	for (i = 0; i < adev->gfx.config.max_shader_engines; i++) {
@@ -2622,17 +2624,14 @@ static void gfx_v8_0_setup_rb(struct amdgpu_device *adev)
 			gfx_v8_0_select_se_sh(adev, i, j);
 			data = gfx_v8_0_get_rb_active_bitmap(adev);
 			active_rbs |= data << ((i * adev->gfx.config.max_sh_per_se + j) *
-					       RB_BITMAP_WIDTH_PER_SH);
+					       rb_bitmap_width_per_sh);
 		}
 	}
 	gfx_v8_0_select_se_sh(adev, 0xffffffff, 0xffffffff);
 	mutex_unlock(&adev->grbm_idx_mutex);
 
 	adev->gfx.config.backend_enable_mask = active_rbs;
-	tmp = active_rbs;
-	while (tmp >>= 1)
-		num_rbs++;
-	adev->gfx.config.num_rbs = num_rbs;
+	adev->gfx.config.num_rbs = hweight32(active_rbs);
 }
 
 /**
@@ -4590,6 +4589,18 @@ static void gfx_v8_0_ring_emit_hdp_flush(struct amdgpu_ring *ring)
 	amdgpu_ring_write(ring, 0x20); /* poll interval */
 }
 
+static void gfx_v8_0_ring_emit_hdp_invalidate(struct amdgpu_ring *ring)
+{
+	amdgpu_ring_write(ring, PACKET3(PACKET3_WRITE_DATA, 3));
+	amdgpu_ring_write(ring, (WRITE_DATA_ENGINE_SEL(0) |
+				 WRITE_DATA_DST_SEL(0) |
+				 WR_CONFIRM));
+	amdgpu_ring_write(ring, mmHDP_DEBUG0);
+	amdgpu_ring_write(ring, 0);
+	amdgpu_ring_write(ring, 1);
+
+}
+
 static void gfx_v8_0_ring_emit_ib_gfx(struct amdgpu_ring *ring,
 				  struct amdgpu_ib *ib)
 {
@@ -4622,8 +4633,7 @@ static void gfx_v8_0_ring_emit_ib_gfx(struct amdgpu_ring *ring,
 	else
 		header = PACKET3(PACKET3_INDIRECT_BUFFER, 2);
 
-	control |= ib->length_dw |
-		(ib->vm ? (ib->vm->ids[ring->idx].id << 24) : 0);
+	control |= ib->length_dw | (ib->vm_id << 24);
 
 	amdgpu_ring_write(ring, header);
 	amdgpu_ring_write(ring,
@@ -4652,8 +4662,7 @@ static void gfx_v8_0_ring_emit_ib_compute(struct amdgpu_ring *ring,
 
 	header = PACKET3(PACKET3_INDIRECT_BUFFER, 2);
 
-	control |= ib->length_dw |
-			   (ib->vm ? (ib->vm->ids[ring->idx].id << 24) : 0);
+	control |= ib->length_dw | (ib->vm_id << 24);
 
 	amdgpu_ring_write(ring, header);
 	amdgpu_ring_write(ring,
@@ -4685,8 +4694,7 @@ static void gfx_v8_0_ring_emit_fence_gfx(struct amdgpu_ring *ring, u64 addr,
 
 }
 
-static void gfx_v8_0_ring_emit_vm_flush(struct amdgpu_ring *ring,
-					unsigned vm_id, uint64_t pd_addr)
+static void gfx_v8_0_ring_emit_pipeline_sync(struct amdgpu_ring *ring)
 {
 	int usepfp = (ring->type == AMDGPU_RING_TYPE_GFX);
 	uint32_t seq = ring->fence_drv.sync_seq;
@@ -4694,7 +4702,8 @@ static void gfx_v8_0_ring_emit_vm_flush(struct amdgpu_ring *ring,
 
 	amdgpu_ring_write(ring, PACKET3(PACKET3_WAIT_REG_MEM, 5));
 	amdgpu_ring_write(ring, (WAIT_REG_MEM_MEM_SPACE(1) | /* memory */
-		 WAIT_REG_MEM_FUNCTION(3))); /* equal */
+				 WAIT_REG_MEM_FUNCTION(3) | /* equal */
+				 WAIT_REG_MEM_ENGINE(usepfp))); /* pfp or me */
 	amdgpu_ring_write(ring, addr & 0xfffffffc);
 	amdgpu_ring_write(ring, upper_32_bits(addr) & 0xffffffff);
 	amdgpu_ring_write(ring, seq);
@@ -4708,6 +4717,12 @@ static void gfx_v8_0_ring_emit_vm_flush(struct amdgpu_ring *ring,
 		amdgpu_ring_write(ring, PACKET3(PACKET3_SWITCH_BUFFER, 0));
 		amdgpu_ring_write(ring, 0);
 	}
+}
+
+static void gfx_v8_0_ring_emit_vm_flush(struct amdgpu_ring *ring,
+					unsigned vm_id, uint64_t pd_addr)
+{
+	int usepfp = (ring->type == AMDGPU_RING_TYPE_GFX);
 
 	amdgpu_ring_write(ring, PACKET3(PACKET3_WRITE_DATA, 3));
 	amdgpu_ring_write(ring, (WRITE_DATA_ENGINE_SEL(usepfp) |
@@ -4880,7 +4895,7 @@ static int gfx_v8_0_set_priv_reg_fault_state(struct amdgpu_device *adev,
 	case AMDGPU_IRQ_STATE_ENABLE:
 		cp_int_cntl = RREG32(mmCP_INT_CNTL_RING0);
 		cp_int_cntl = REG_SET_FIELD(cp_int_cntl, CP_INT_CNTL_RING0,
-					    PRIV_REG_INT_ENABLE, 0);
+					    PRIV_REG_INT_ENABLE, 1);
 		WREG32(mmCP_INT_CNTL_RING0, cp_int_cntl);
 		break;
 	default:
@@ -5030,9 +5045,11 @@ static const struct amdgpu_ring_funcs gfx_v8_0_ring_funcs_gfx = {
 	.parse_cs = NULL,
 	.emit_ib = gfx_v8_0_ring_emit_ib_gfx,
 	.emit_fence = gfx_v8_0_ring_emit_fence_gfx,
+	.emit_pipeline_sync = gfx_v8_0_ring_emit_pipeline_sync,
 	.emit_vm_flush = gfx_v8_0_ring_emit_vm_flush,
 	.emit_gds_switch = gfx_v8_0_ring_emit_gds_switch,
 	.emit_hdp_flush = gfx_v8_0_ring_emit_hdp_flush,
+	.emit_hdp_invalidate = gfx_v8_0_ring_emit_hdp_invalidate,
 	.test_ring = gfx_v8_0_ring_test_ring,
 	.test_ib = gfx_v8_0_ring_test_ib,
 	.insert_nop = amdgpu_ring_insert_nop,
@@ -5046,9 +5063,11 @@ static const struct amdgpu_ring_funcs gfx_v8_0_ring_funcs_compute = {
 	.parse_cs = NULL,
 	.emit_ib = gfx_v8_0_ring_emit_ib_compute,
 	.emit_fence = gfx_v8_0_ring_emit_fence_compute,
+	.emit_pipeline_sync = gfx_v8_0_ring_emit_pipeline_sync,
 	.emit_vm_flush = gfx_v8_0_ring_emit_vm_flush,
 	.emit_gds_switch = gfx_v8_0_ring_emit_gds_switch,
 	.emit_hdp_flush = gfx_v8_0_ring_emit_hdp_flush,
+	.emit_hdp_invalidate = gfx_v8_0_ring_emit_hdp_invalidate,
 	.test_ring = gfx_v8_0_ring_test_ring,
 	.test_ib = gfx_v8_0_ring_test_ib,
 	.insert_nop = amdgpu_ring_insert_nop,
@@ -5131,8 +5150,7 @@ static u32 gfx_v8_0_get_cu_active_bitmap(struct amdgpu_device *adev)
 	data &= CC_GC_SHADER_ARRAY_CONFIG__INACTIVE_CUS_MASK;
 	data >>= CC_GC_SHADER_ARRAY_CONFIG__INACTIVE_CUS__SHIFT;
 
-	mask = gfx_v8_0_create_bitmask(adev->gfx.config.max_backends_per_se /
-				       adev->gfx.config.max_sh_per_se);
+	mask = gfx_v8_0_create_bitmask(adev->gfx.config.max_cu_per_sh);
 
 	return (~data) & mask;
 }
@@ -5146,6 +5164,8 @@ int gfx_v8_0_get_cu_info(struct amdgpu_device *adev,
 	if (!adev || !cu_info)
 		return -EINVAL;
 
+	memset(cu_info, 0, sizeof(*cu_info));
+
 	mutex_lock(&adev->grbm_idx_mutex);
 	for (i = 0; i < adev->gfx.config.max_shader_engines; i++) {
 		for (j = 0; j < adev->gfx.config.max_sh_per_se; j++) {
diff --git a/drivers/gpu/drm/amd/amdgpu/gmc_v7_0.c b/drivers/gpu/drm/amd/amdgpu/gmc_v7_0.c
index 68ee66b38e5c..82ce7d943884 100644
--- a/drivers/gpu/drm/amd/amdgpu/gmc_v7_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gmc_v7_0.c
@@ -339,7 +339,7 @@ static void gmc_v7_0_mc_program(struct amdgpu_device *adev)
 	WREG32(mmBIF_FB_EN, BIF_FB_EN__FB_READ_EN_MASK | BIF_FB_EN__FB_WRITE_EN_MASK);
 
 	tmp = RREG32(mmHDP_MISC_CNTL);
-	tmp = REG_SET_FIELD(tmp, HDP_MISC_CNTL, FLUSH_INVALIDATE_CACHE, 1);
+	tmp = REG_SET_FIELD(tmp, HDP_MISC_CNTL, FLUSH_INVALIDATE_CACHE, 0);
 	WREG32(mmHDP_MISC_CNTL, tmp);
 
 	tmp = RREG32(mmHDP_HOST_PATH_CNTL);
@@ -793,7 +793,7 @@ static void gmc_v7_0_enable_mc_ls(struct amdgpu_device *adev,
 
 	for (i = 0; i < ARRAY_SIZE(mc_cg_registers); i++) {
 		orig = data = RREG32(mc_cg_registers[i]);
-		if (enable && (adev->cg_flags & AMDGPU_CG_SUPPORT_MC_LS))
+		if (enable && (adev->cg_flags & AMD_CG_SUPPORT_MC_LS))
 			data |= mc_cg_ls_en[i];
 		else
 			data &= ~mc_cg_ls_en[i];
@@ -810,7 +810,7 @@ static void gmc_v7_0_enable_mc_mgcg(struct amdgpu_device *adev,
 
 	for (i = 0; i < ARRAY_SIZE(mc_cg_registers); i++) {
 		orig = data = RREG32(mc_cg_registers[i]);
-		if (enable && (adev->cg_flags & AMDGPU_CG_SUPPORT_MC_MGCG))
+		if (enable && (adev->cg_flags & AMD_CG_SUPPORT_MC_MGCG))
 			data |= mc_cg_en[i];
 		else
 			data &= ~mc_cg_en[i];
@@ -826,7 +826,7 @@ static void gmc_v7_0_enable_bif_mgls(struct amdgpu_device *adev,
 
 	orig = data = RREG32_PCIE(ixPCIE_CNTL2);
 
-	if (enable && (adev->cg_flags & AMDGPU_CG_SUPPORT_BIF_LS)) {
+	if (enable && (adev->cg_flags & AMD_CG_SUPPORT_BIF_LS)) {
 		data = REG_SET_FIELD(data, PCIE_CNTL2, SLV_MEM_LS_EN, 1);
 		data = REG_SET_FIELD(data, PCIE_CNTL2, MST_MEM_LS_EN, 1);
 		data = REG_SET_FIELD(data, PCIE_CNTL2, REPLAY_MEM_LS_EN, 1);
@@ -849,7 +849,7 @@ static void gmc_v7_0_enable_hdp_mgcg(struct amdgpu_device *adev,
 
 	orig = data = RREG32(mmHDP_HOST_PATH_CNTL);
 
-	if (enable && (adev->cg_flags & AMDGPU_CG_SUPPORT_HDP_MGCG))
+	if (enable && (adev->cg_flags & AMD_CG_SUPPORT_HDP_MGCG))
 		data = REG_SET_FIELD(data, HDP_HOST_PATH_CNTL, CLOCK_GATING_DIS, 0);
 	else
 		data = REG_SET_FIELD(data, HDP_HOST_PATH_CNTL, CLOCK_GATING_DIS, 1);
@@ -865,7 +865,7 @@ static void gmc_v7_0_enable_hdp_ls(struct amdgpu_device *adev,
 
 	orig = data = RREG32(mmHDP_MEM_POWER_LS);
 
-	if (enable && (adev->cg_flags & AMDGPU_CG_SUPPORT_HDP_LS))
+	if (enable && (adev->cg_flags & AMD_CG_SUPPORT_HDP_LS))
 		data = REG_SET_FIELD(data, HDP_MEM_POWER_LS, LS_ENABLE, 1);
 	else
 		data = REG_SET_FIELD(data, HDP_MEM_POWER_LS, LS_ENABLE, 0);
diff --git a/drivers/gpu/drm/amd/amdgpu/gmc_v8_0.c b/drivers/gpu/drm/amd/amdgpu/gmc_v8_0.c
index 757803ae7c4a..29bd7b57dc91 100644
--- a/drivers/gpu/drm/amd/amdgpu/gmc_v8_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gmc_v8_0.c
@@ -386,7 +386,7 @@ static void gmc_v8_0_mc_program(struct amdgpu_device *adev)
 	WREG32(mmBIF_FB_EN, BIF_FB_EN__FB_READ_EN_MASK | BIF_FB_EN__FB_WRITE_EN_MASK);
 
 	tmp = RREG32(mmHDP_MISC_CNTL);
-	tmp = REG_SET_FIELD(tmp, HDP_MISC_CNTL, FLUSH_INVALIDATE_CACHE, 1);
+	tmp = REG_SET_FIELD(tmp, HDP_MISC_CNTL, FLUSH_INVALIDATE_CACHE, 0);
 	WREG32(mmHDP_MISC_CNTL, tmp);
 
 	tmp = RREG32(mmHDP_HOST_PATH_CNTL);
diff --git a/drivers/gpu/drm/amd/amdgpu/kv_dpm.c b/drivers/gpu/drm/amd/amdgpu/kv_dpm.c
index 7e9154c7f1db..654d76723bc3 100644
--- a/drivers/gpu/drm/amd/amdgpu/kv_dpm.c
+++ b/drivers/gpu/drm/amd/amdgpu/kv_dpm.c
@@ -2859,11 +2859,11 @@ static int kv_dpm_init(struct amdgpu_device *adev)
 	pi->voltage_drop_t = 0;
 	pi->caps_sclk_throttle_low_notification = false;
 	pi->caps_fps = false; /* true? */
-	pi->caps_uvd_pg = (adev->pg_flags & AMDGPU_PG_SUPPORT_UVD) ? true : false;
+	pi->caps_uvd_pg = (adev->pg_flags & AMD_PG_SUPPORT_UVD) ? true : false;
 	pi->caps_uvd_dpm = true;
-	pi->caps_vce_pg = (adev->pg_flags & AMDGPU_PG_SUPPORT_VCE) ? true : false;
-	pi->caps_samu_pg = (adev->pg_flags & AMDGPU_PG_SUPPORT_SAMU) ? true : false;
-	pi->caps_acp_pg = (adev->pg_flags & AMDGPU_PG_SUPPORT_ACP) ? true : false;
+	pi->caps_vce_pg = (adev->pg_flags & AMD_PG_SUPPORT_VCE) ? true : false;
+	pi->caps_samu_pg = (adev->pg_flags & AMD_PG_SUPPORT_SAMU) ? true : false;
+	pi->caps_acp_pg = (adev->pg_flags & AMD_PG_SUPPORT_ACP) ? true : false;
 	pi->caps_stable_p_state = false;
 
 	ret = kv_parse_sys_info_table(adev);
diff --git a/drivers/gpu/drm/amd/amdgpu/sdma_v2_4.c b/drivers/gpu/drm/amd/amdgpu/sdma_v2_4.c
index 29ec986dd6fc..6e0a86a563f3 100644
--- a/drivers/gpu/drm/amd/amdgpu/sdma_v2_4.c
+++ b/drivers/gpu/drm/amd/amdgpu/sdma_v2_4.c
@@ -32,8 +32,8 @@
 #include "oss/oss_2_4_d.h"
 #include "oss/oss_2_4_sh_mask.h"
 
-#include "gmc/gmc_8_1_d.h"
-#include "gmc/gmc_8_1_sh_mask.h"
+#include "gmc/gmc_7_1_d.h"
+#include "gmc/gmc_7_1_sh_mask.h"
 
 #include "gca/gfx_8_0_d.h"
 #include "gca/gfx_8_0_enum.h"
@@ -244,7 +244,7 @@ static void sdma_v2_4_ring_insert_nop(struct amdgpu_ring *ring, uint32_t count)
 static void sdma_v2_4_ring_emit_ib(struct amdgpu_ring *ring,
 				   struct amdgpu_ib *ib)
 {
-	u32 vmid = (ib->vm ? ib->vm->ids[ring->idx].id : 0) & 0xf;
+	u32 vmid = ib->vm_id & 0xf;
 	u32 next_rptr = ring->wptr + 5;
 
 	while ((next_rptr & 7) != 2)
@@ -300,6 +300,13 @@ static void sdma_v2_4_ring_emit_hdp_flush(struct amdgpu_ring *ring)
 			  SDMA_PKT_POLL_REGMEM_DW5_INTERVAL(10)); /* retry count, poll interval */
 }
 
+static void sdma_v2_4_ring_emit_hdp_invalidate(struct amdgpu_ring *ring)
+{
+	amdgpu_ring_write(ring, SDMA_PKT_HEADER_OP(SDMA_OP_SRBM_WRITE) |
+			  SDMA_PKT_SRBM_WRITE_HEADER_BYTE_EN(0xf));
+	amdgpu_ring_write(ring, mmHDP_DEBUG0);
+	amdgpu_ring_write(ring, 1);
+}
 /**
  * sdma_v2_4_ring_emit_fence - emit a fence on the DMA ring
  *
@@ -694,8 +701,7 @@ static int sdma_v2_4_ring_test_ib(struct amdgpu_ring *ring)
 	ib.ptr[7] = SDMA_PKT_HEADER_OP(SDMA_OP_NOP);
 	ib.length_dw = 8;
 
-	r = amdgpu_ib_schedule(ring, 1, &ib, AMDGPU_FENCE_OWNER_UNDEFINED,
-			       NULL, &f);
+	r = amdgpu_ib_schedule(ring, 1, &ib, NULL, &f);
 	if (r)
 		goto err1;
 
@@ -721,7 +727,8 @@ static int sdma_v2_4_ring_test_ib(struct amdgpu_ring *ring)
 
 err1:
 	fence_put(f);
-	amdgpu_ib_free(adev, &ib);
+	amdgpu_ib_free(adev, &ib, NULL);
+	fence_put(f);
 err0:
 	amdgpu_wb_free(adev, index);
 	return r;
@@ -874,6 +881,31 @@ static void sdma_v2_4_ring_pad_ib(struct amdgpu_ring *ring, struct amdgpu_ib *ib
 }
 
 /**
+ * sdma_v2_4_ring_emit_pipeline_sync - sync the pipeline
+ *
+ * @ring: amdgpu_ring pointer
+ *
+ * Make sure all previous operations are completed (CIK).
+ */
+static void sdma_v2_4_ring_emit_pipeline_sync(struct amdgpu_ring *ring)
+{
+	uint32_t seq = ring->fence_drv.sync_seq;
+	uint64_t addr = ring->fence_drv.gpu_addr;
+
+	/* wait for idle */
+	amdgpu_ring_write(ring, SDMA_PKT_HEADER_OP(SDMA_OP_POLL_REGMEM) |
+			  SDMA_PKT_POLL_REGMEM_HEADER_HDP_FLUSH(0) |
+			  SDMA_PKT_POLL_REGMEM_HEADER_FUNC(3) | /* equal */
+			  SDMA_PKT_POLL_REGMEM_HEADER_MEM_POLL(1));
+	amdgpu_ring_write(ring, addr & 0xfffffffc);
+	amdgpu_ring_write(ring, upper_32_bits(addr) & 0xffffffff);
+	amdgpu_ring_write(ring, seq); /* reference */
+	amdgpu_ring_write(ring, 0xfffffff); /* mask */
+	amdgpu_ring_write(ring, SDMA_PKT_POLL_REGMEM_DW5_RETRY_COUNT(0xfff) |
+			  SDMA_PKT_POLL_REGMEM_DW5_INTERVAL(4)); /* retry count, poll interval */
+}
+
+/**
  * sdma_v2_4_ring_emit_vm_flush - cik vm flush using sDMA
  *
  * @ring: amdgpu_ring pointer
@@ -1274,8 +1306,10 @@ static const struct amdgpu_ring_funcs sdma_v2_4_ring_funcs = {
 	.parse_cs = NULL,
 	.emit_ib = sdma_v2_4_ring_emit_ib,
 	.emit_fence = sdma_v2_4_ring_emit_fence,
+	.emit_pipeline_sync = sdma_v2_4_ring_emit_pipeline_sync,
 	.emit_vm_flush = sdma_v2_4_ring_emit_vm_flush,
 	.emit_hdp_flush = sdma_v2_4_ring_emit_hdp_flush,
+	.emit_hdp_invalidate = sdma_v2_4_ring_emit_hdp_invalidate,
 	.test_ring = sdma_v2_4_ring_test_ring,
 	.test_ib = sdma_v2_4_ring_test_ib,
 	.insert_nop = sdma_v2_4_ring_insert_nop,
diff --git a/drivers/gpu/drm/amd/amdgpu/sdma_v3_0.c b/drivers/gpu/drm/amd/amdgpu/sdma_v3_0.c
index 6f064d7076e6..8c8ca98dd129 100644
--- a/drivers/gpu/drm/amd/amdgpu/sdma_v3_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/sdma_v3_0.c
@@ -355,7 +355,7 @@ static void sdma_v3_0_ring_insert_nop(struct amdgpu_ring *ring, uint32_t count)
 static void sdma_v3_0_ring_emit_ib(struct amdgpu_ring *ring,
 				   struct amdgpu_ib *ib)
 {
-	u32 vmid = (ib->vm ? ib->vm->ids[ring->idx].id : 0) & 0xf;
+	u32 vmid = ib->vm_id & 0xf;
 	u32 next_rptr = ring->wptr + 5;
 
 	while ((next_rptr & 7) != 2)
@@ -410,6 +410,14 @@ static void sdma_v3_0_ring_emit_hdp_flush(struct amdgpu_ring *ring)
 			  SDMA_PKT_POLL_REGMEM_DW5_INTERVAL(10)); /* retry count, poll interval */
 }
 
+static void sdma_v3_0_ring_emit_hdp_invalidate(struct amdgpu_ring *ring)
+{
+	amdgpu_ring_write(ring, SDMA_PKT_HEADER_OP(SDMA_OP_SRBM_WRITE) |
+			  SDMA_PKT_SRBM_WRITE_HEADER_BYTE_EN(0xf));
+	amdgpu_ring_write(ring, mmHDP_DEBUG0);
+	amdgpu_ring_write(ring, 1);
+}
+
 /**
  * sdma_v3_0_ring_emit_fence - emit a fence on the DMA ring
  *
@@ -845,8 +853,7 @@ static int sdma_v3_0_ring_test_ib(struct amdgpu_ring *ring)
 	ib.ptr[7] = SDMA_PKT_NOP_HEADER_OP(SDMA_OP_NOP);
 	ib.length_dw = 8;
 
-	r = amdgpu_ib_schedule(ring, 1, &ib, AMDGPU_FENCE_OWNER_UNDEFINED,
-			       NULL, &f);
+	r = amdgpu_ib_schedule(ring, 1, &ib, NULL, &f);
 	if (r)
 		goto err1;
 
@@ -871,7 +878,8 @@ static int sdma_v3_0_ring_test_ib(struct amdgpu_ring *ring)
 	}
 err1:
 	fence_put(f);
-	amdgpu_ib_free(adev, &ib);
+	amdgpu_ib_free(adev, &ib, NULL);
+	fence_put(f);
 err0:
 	amdgpu_wb_free(adev, index);
 	return r;
@@ -1024,6 +1032,31 @@ static void sdma_v3_0_ring_pad_ib(struct amdgpu_ring *ring, struct amdgpu_ib *ib
 }
 
 /**
+ * sdma_v3_0_ring_emit_pipeline_sync - sync the pipeline
+ *
+ * @ring: amdgpu_ring pointer
+ *
+ * Make sure all previous operations are completed (CIK).
+ */
+static void sdma_v3_0_ring_emit_pipeline_sync(struct amdgpu_ring *ring)
+{
+	uint32_t seq = ring->fence_drv.sync_seq;
+	uint64_t addr = ring->fence_drv.gpu_addr;
+
+	/* wait for idle */
+	amdgpu_ring_write(ring, SDMA_PKT_HEADER_OP(SDMA_OP_POLL_REGMEM) |
+			  SDMA_PKT_POLL_REGMEM_HEADER_HDP_FLUSH(0) |
+			  SDMA_PKT_POLL_REGMEM_HEADER_FUNC(3) | /* equal */
+			  SDMA_PKT_POLL_REGMEM_HEADER_MEM_POLL(1));
+	amdgpu_ring_write(ring, addr & 0xfffffffc);
+	amdgpu_ring_write(ring, upper_32_bits(addr) & 0xffffffff);
+	amdgpu_ring_write(ring, seq); /* reference */
+	amdgpu_ring_write(ring, 0xfffffff); /* mask */
+	amdgpu_ring_write(ring, SDMA_PKT_POLL_REGMEM_DW5_RETRY_COUNT(0xfff) |
+			  SDMA_PKT_POLL_REGMEM_DW5_INTERVAL(4)); /* retry count, poll interval */
+}
+
+/**
  * sdma_v3_0_ring_emit_vm_flush - cik vm flush using sDMA
  *
  * @ring: amdgpu_ring pointer
@@ -1541,8 +1574,10 @@ static const struct amdgpu_ring_funcs sdma_v3_0_ring_funcs = {
 	.parse_cs = NULL,
 	.emit_ib = sdma_v3_0_ring_emit_ib,
 	.emit_fence = sdma_v3_0_ring_emit_fence,
+	.emit_pipeline_sync = sdma_v3_0_ring_emit_pipeline_sync,
 	.emit_vm_flush = sdma_v3_0_ring_emit_vm_flush,
 	.emit_hdp_flush = sdma_v3_0_ring_emit_hdp_flush,
+	.emit_hdp_invalidate = sdma_v3_0_ring_emit_hdp_invalidate,
 	.test_ring = sdma_v3_0_ring_test_ring,
 	.test_ib = sdma_v3_0_ring_test_ib,
 	.insert_nop = sdma_v3_0_ring_insert_nop,
diff --git a/drivers/gpu/drm/amd/amdgpu/uvd_v4_2.c b/drivers/gpu/drm/amd/amdgpu/uvd_v4_2.c
index 70ed73fa5156..c606ccb38d8b 100644
--- a/drivers/gpu/drm/amd/amdgpu/uvd_v4_2.c
+++ b/drivers/gpu/drm/amd/amdgpu/uvd_v4_2.c
@@ -588,7 +588,7 @@ static void uvd_v4_2_enable_mgcg(struct amdgpu_device *adev,
 {
 	u32 orig, data;
 
-	if (enable && (adev->cg_flags & AMDGPU_CG_SUPPORT_UVD_MGCG)) {
+	if (enable && (adev->cg_flags & AMD_CG_SUPPORT_UVD_MGCG)) {
 		data = RREG32_UVD_CTX(ixUVD_CGC_MEM_CTRL);
 		data = 0xfff;
 		WREG32_UVD_CTX(ixUVD_CGC_MEM_CTRL, data);
@@ -814,6 +814,9 @@ static int uvd_v4_2_set_clockgating_state(void *handle,
 	bool gate = false;
 	struct amdgpu_device *adev = (struct amdgpu_device *)handle;
 
+	if (!(adev->cg_flags & AMD_CG_SUPPORT_UVD_MGCG))
+		return 0;
+
 	if (state == AMD_CG_STATE_GATE)
 		gate = true;
 
@@ -832,7 +835,10 @@ static int uvd_v4_2_set_powergating_state(void *handle,
 	 * revisit this when there is a cleaner line between
 	 * the smc and the hw blocks
 	 */
-	 struct amdgpu_device *adev = (struct amdgpu_device *)handle;
+	struct amdgpu_device *adev = (struct amdgpu_device *)handle;
+
+	if (!(adev->pg_flags & AMD_PG_SUPPORT_UVD))
+		return 0;
 
 	if (state == AMD_PG_STATE_GATE) {
 		uvd_v4_2_stop(adev);
diff --git a/drivers/gpu/drm/amd/amdgpu/uvd_v5_0.c b/drivers/gpu/drm/amd/amdgpu/uvd_v5_0.c
index 578ffb62fdb2..e3c852d9d79a 100644
--- a/drivers/gpu/drm/amd/amdgpu/uvd_v5_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/uvd_v5_0.c
@@ -757,6 +757,11 @@ static int uvd_v5_0_process_interrupt(struct amdgpu_device *adev,
 static int uvd_v5_0_set_clockgating_state(void *handle,
 					  enum amd_clockgating_state state)
 {
+	struct amdgpu_device *adev = (struct amdgpu_device *)handle;
+
+	if (!(adev->cg_flags & AMD_CG_SUPPORT_UVD_MGCG))
+		return 0;
+
 	return 0;
 }
 
@@ -772,6 +777,9 @@ static int uvd_v5_0_set_powergating_state(void *handle,
 	 */
 	struct amdgpu_device *adev = (struct amdgpu_device *)handle;
 
+	if (!(adev->pg_flags & AMD_PG_SUPPORT_UVD))
+		return 0;
+
 	if (state == AMD_PG_STATE_GATE) {
 		uvd_v5_0_stop(adev);
 		return 0;
diff --git a/drivers/gpu/drm/amd/amdgpu/uvd_v6_0.c b/drivers/gpu/drm/amd/amdgpu/uvd_v6_0.c
index d4da1f04378c..3375e614ac67 100644
--- a/drivers/gpu/drm/amd/amdgpu/uvd_v6_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/uvd_v6_0.c
@@ -536,7 +536,7 @@ static int uvd_v6_0_start(struct amdgpu_device *adev)
 	uvd_v6_0_mc_resume(adev);
 
 	/* Set dynamic clock gating in S/W control mode */
-	if (adev->cg_flags & AMDGPU_CG_SUPPORT_UVD_MGCG) {
+	if (adev->cg_flags & AMD_CG_SUPPORT_UVD_MGCG) {
 		if (adev->flags & AMD_IS_APU)
 			cz_set_uvd_clock_gating_branches(adev, false);
 		else
@@ -983,7 +983,7 @@ static int uvd_v6_0_set_clockgating_state(void *handle,
 	struct amdgpu_device *adev = (struct amdgpu_device *)handle;
 	bool enable = (state == AMD_CG_STATE_GATE) ? true : false;
 
-	if (!(adev->cg_flags & AMDGPU_CG_SUPPORT_UVD_MGCG))
+	if (!(adev->cg_flags & AMD_CG_SUPPORT_UVD_MGCG))
 		return 0;
 
 	if (enable) {
@@ -1013,6 +1013,9 @@ static int uvd_v6_0_set_powergating_state(void *handle,
 	 */
 	struct amdgpu_device *adev = (struct amdgpu_device *)handle;
 
+	if (!(adev->pg_flags & AMD_PG_SUPPORT_UVD))
+		return 0;
+
 	if (state == AMD_PG_STATE_GATE) {
 		uvd_v6_0_stop(adev);
 		return 0;
diff --git a/drivers/gpu/drm/amd/amdgpu/vce_v2_0.c b/drivers/gpu/drm/amd/amdgpu/vce_v2_0.c
index 9c804f436974..c7e885bcfd41 100644
--- a/drivers/gpu/drm/amd/amdgpu/vce_v2_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/vce_v2_0.c
@@ -373,7 +373,7 @@ static void vce_v2_0_enable_mgcg(struct amdgpu_device *adev, bool enable)
 {
 	bool sw_cg = false;
 
-	if (enable && (adev->cg_flags & AMDGPU_CG_SUPPORT_VCE_MGCG)) {
+	if (enable && (adev->cg_flags & AMD_CG_SUPPORT_VCE_MGCG)) {
 		if (sw_cg)
 			vce_v2_0_set_sw_cg(adev, true);
 		else
@@ -608,6 +608,9 @@ static int vce_v2_0_set_powergating_state(void *handle,
 	 */
 	struct amdgpu_device *adev = (struct amdgpu_device *)handle;
 
+	if (!(adev->pg_flags & AMD_PG_SUPPORT_VCE))
+		return 0;
+
 	if (state == AMD_PG_STATE_GATE)
 		/* XXX do we need a vce_v2_0_stop()? */
 		return 0;
diff --git a/drivers/gpu/drm/amd/amdgpu/vce_v3_0.c b/drivers/gpu/drm/amd/amdgpu/vce_v3_0.c
index 8f8d479061f8..ce468ee5da2a 100644
--- a/drivers/gpu/drm/amd/amdgpu/vce_v3_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/vce_v3_0.c
@@ -277,7 +277,7 @@ static int vce_v3_0_start(struct amdgpu_device *adev)
 		WREG32_P(mmVCE_STATUS, 0, ~1);
 
 		/* Set Clock-Gating off */
-		if (adev->cg_flags & AMDGPU_CG_SUPPORT_VCE_MGCG)
+		if (adev->cg_flags & AMD_CG_SUPPORT_VCE_MGCG)
 			vce_v3_0_set_vce_sw_clock_gating(adev, false);
 
 		if (r) {
@@ -676,7 +676,7 @@ static int vce_v3_0_set_clockgating_state(void *handle,
 	bool enable = (state == AMD_CG_STATE_GATE) ? true : false;
 	int i;
 
-	if (!(adev->cg_flags & AMDGPU_CG_SUPPORT_VCE_MGCG))
+	if (!(adev->cg_flags & AMD_CG_SUPPORT_VCE_MGCG))
 		return 0;
 
 	mutex_lock(&adev->grbm_idx_mutex);
@@ -728,6 +728,9 @@ static int vce_v3_0_set_powergating_state(void *handle,
 	 */
 	struct amdgpu_device *adev = (struct amdgpu_device *)handle;
 
+	if (!(adev->pg_flags & AMD_PG_SUPPORT_VCE))
+		return 0;
+
 	if (state == AMD_PG_STATE_GATE)
 		/* XXX do we need a vce_v3_0_stop()? */
 		return 0;
diff --git a/drivers/gpu/drm/amd/amdgpu/vi.c b/drivers/gpu/drm/amd/amdgpu/vi.c
index 125003517544..1c120efa292c 100644
--- a/drivers/gpu/drm/amd/amdgpu/vi.c
+++ b/drivers/gpu/drm/amd/amdgpu/vi.c
@@ -1071,29 +1071,24 @@ static int vi_common_early_init(void *handle)
 	adev->external_rev_id = 0xFF;
 	switch (adev->asic_type) {
 	case CHIP_TOPAZ:
-		adev->has_uvd = false;
 		adev->cg_flags = 0;
 		adev->pg_flags = 0;
 		adev->external_rev_id = 0x1;
 		break;
 	case CHIP_FIJI:
-		adev->has_uvd = true;
 		adev->cg_flags = 0;
 		adev->pg_flags = 0;
 		adev->external_rev_id = adev->rev_id + 0x3c;
 		break;
 	case CHIP_TONGA:
-		adev->has_uvd = true;
 		adev->cg_flags = 0;
 		adev->pg_flags = 0;
 		adev->external_rev_id = adev->rev_id + 0x14;
 		break;
 	case CHIP_CARRIZO:
 	case CHIP_STONEY:
-		adev->has_uvd = true;
 		adev->cg_flags = 0;
-		/* Disable UVD pg */
-		adev->pg_flags = /* AMDGPU_PG_SUPPORT_UVD | */AMDGPU_PG_SUPPORT_VCE;
+		adev->pg_flags = 0;
 		adev->external_rev_id = adev->rev_id + 0x1;
 		break;
 	default:
diff --git a/drivers/gpu/drm/amd/amdgpu/vid.h b/drivers/gpu/drm/amd/amdgpu/vid.h
index d98aa9d82fa1..ace49976f7be 100644
--- a/drivers/gpu/drm/amd/amdgpu/vid.h
+++ b/drivers/gpu/drm/amd/amdgpu/vid.h
@@ -71,8 +71,6 @@
 #define		VMID(x)						((x) << 4)
 #define		QUEUEID(x)					((x) << 8)
 
-#define RB_BITMAP_WIDTH_PER_SH     2
-
 #define MC_SEQ_MISC0__MT__MASK	0xf0000000
 #define MC_SEQ_MISC0__MT__GDDR1  0x10000000
 #define MC_SEQ_MISC0__MT__DDR2   0x20000000
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_dbgdev.c b/drivers/gpu/drm/amd/amdkfd/kfd_dbgdev.c
index c34c393e9aea..d5e19b5fbbfb 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_dbgdev.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_dbgdev.c
@@ -513,7 +513,7 @@ static int dbgdev_wave_control_set_registers(
 				union SQ_CMD_BITS *in_reg_sq_cmd,
 				union GRBM_GFX_INDEX_BITS *in_reg_gfx_index)
 {
-	int status;
+	int status = 0;
 	union SQ_CMD_BITS reg_sq_cmd;
 	union GRBM_GFX_INDEX_BITS reg_gfx_index;
 	struct HsaDbgWaveMsgAMDGen2 *pMsg;
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_module.c b/drivers/gpu/drm/amd/amdkfd/kfd_module.c
index ca8410e8683d..850a5623661f 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_module.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_module.c
@@ -59,18 +59,23 @@ module_param(send_sigterm, int, 0444);
 MODULE_PARM_DESC(send_sigterm,
 	"Send sigterm to HSA process on unhandled exception (0 = disable, 1 = enable)");
 
-bool kgd2kfd_init(unsigned interface_version, const struct kgd2kfd_calls **g2f)
+static int amdkfd_init_completed;
+
+int kgd2kfd_init(unsigned interface_version, const struct kgd2kfd_calls **g2f)
 {
+	if (!amdkfd_init_completed)
+		return -EPROBE_DEFER;
+
 	/*
 	 * Only one interface version is supported,
 	 * no kfd/kgd version skew allowed.
 	 */
 	if (interface_version != KFD_INTERFACE_VERSION)
-		return false;
+		return -EINVAL;
 
 	*g2f = &kgd2kfd;
 
-	return true;
+	return 0;
 }
 EXPORT_SYMBOL(kgd2kfd_init);
 
@@ -111,6 +116,8 @@ static int __init kfd_module_init(void)
 
 	kfd_process_create_wq();
 
+	amdkfd_init_completed = 1;
+
 	dev_info(kfd_device, "Initialized module\n");
 
 	return 0;
@@ -125,6 +132,8 @@ err_pasid:
 
 static void __exit kfd_module_exit(void)
 {
+	amdkfd_init_completed = 0;
+
 	kfd_process_destroy_wq();
 	kfd_topology_shutdown();
 	kfd_chardev_exit();
diff --git a/drivers/gpu/drm/amd/include/amd_acpi.h b/drivers/gpu/drm/amd/include/amd_acpi.h
index 496360eb3fba..50e893325141 100644
--- a/drivers/gpu/drm/amd/include/amd_acpi.h
+++ b/drivers/gpu/drm/amd/include/amd_acpi.h
@@ -340,6 +340,8 @@ struct atcs_pref_req_output {
 #       define ATPX_FIXED_NOT_SUPPORTED                            (1 << 9)
 #       define ATPX_DYNAMIC_DGPU_POWER_OFF_SUPPORTED               (1 << 10)
 #       define ATPX_DGPU_REQ_POWER_FOR_DISPLAYS                    (1 << 11)
+#       define ATPX_DGPU_CAN_DRIVE_DISPLAYS                        (1 << 12)
+#       define ATPX_MS_HYBRID_GFX_SUPPORTED                        (1 << 14)
 #define ATPX_FUNCTION_POWER_CONTROL                                0x2
 /* ARG0: ATPX_FUNCTION_POWER_CONTROL
  * ARG1:
diff --git a/drivers/gpu/drm/amd/include/amd_shared.h b/drivers/gpu/drm/amd/include/amd_shared.h
index 15ff8b2c26e7..04e4090666fb 100644
--- a/drivers/gpu/drm/amd/include/amd_shared.h
+++ b/drivers/gpu/drm/amd/include/amd_shared.h
@@ -86,6 +86,38 @@ enum amd_powergating_state {
 	AMD_PG_STATE_UNGATE,
 };
 
+/* CG flags */
+#define AMD_CG_SUPPORT_GFX_MGCG			(1 << 0)
+#define AMD_CG_SUPPORT_GFX_MGLS			(1 << 1)
+#define AMD_CG_SUPPORT_GFX_CGCG			(1 << 2)
+#define AMD_CG_SUPPORT_GFX_CGLS			(1 << 3)
+#define AMD_CG_SUPPORT_GFX_CGTS			(1 << 4)
+#define AMD_CG_SUPPORT_GFX_CGTS_LS		(1 << 5)
+#define AMD_CG_SUPPORT_GFX_CP_LS		(1 << 6)
+#define AMD_CG_SUPPORT_GFX_RLC_LS		(1 << 7)
+#define AMD_CG_SUPPORT_MC_LS			(1 << 8)
+#define AMD_CG_SUPPORT_MC_MGCG			(1 << 9)
+#define AMD_CG_SUPPORT_SDMA_LS			(1 << 10)
+#define AMD_CG_SUPPORT_SDMA_MGCG		(1 << 11)
+#define AMD_CG_SUPPORT_BIF_LS			(1 << 12)
+#define AMD_CG_SUPPORT_UVD_MGCG			(1 << 13)
+#define AMD_CG_SUPPORT_VCE_MGCG			(1 << 14)
+#define AMD_CG_SUPPORT_HDP_LS			(1 << 15)
+#define AMD_CG_SUPPORT_HDP_MGCG			(1 << 16)
+
+/* PG flags */
+#define AMD_PG_SUPPORT_GFX_PG			(1 << 0)
+#define AMD_PG_SUPPORT_GFX_SMG			(1 << 1)
+#define AMD_PG_SUPPORT_GFX_DMG			(1 << 2)
+#define AMD_PG_SUPPORT_UVD			(1 << 3)
+#define AMD_PG_SUPPORT_VCE			(1 << 4)
+#define AMD_PG_SUPPORT_CP			(1 << 5)
+#define AMD_PG_SUPPORT_GDS			(1 << 6)
+#define AMD_PG_SUPPORT_RLC_SMU_HS		(1 << 7)
+#define AMD_PG_SUPPORT_SDMA			(1 << 8)
+#define AMD_PG_SUPPORT_ACP			(1 << 9)
+#define AMD_PG_SUPPORT_SAMU			(1 << 10)
+
 enum amd_pm_state_type {
 	/* not used for dpm */
 	POWER_STATE_TYPE_DEFAULT,
diff --git a/drivers/gpu/drm/amd/include/asic_reg/gca/gfx_7_2_enum.h b/drivers/gpu/drm/amd/include/asic_reg/gca/gfx_7_2_enum.h
index 9d4347dd6125..dfe78799100d 100644
--- a/drivers/gpu/drm/amd/include/asic_reg/gca/gfx_7_2_enum.h
+++ b/drivers/gpu/drm/amd/include/asic_reg/gca/gfx_7_2_enum.h
@@ -6225,6 +6225,12 @@ typedef enum TCC_CACHE_POLICIES {
 	TCC_CACHE_POLICY_STREAM                          = 0x1,
 	TCC_CACHE_POLICY_BYPASS                          = 0x2,
 } TCC_CACHE_POLICIES;
+typedef enum MTYPE {
+	MTYPE_NC_NV                                      = 0x0,
+	MTYPE_NC                                         = 0x1,
+	MTYPE_CC                                         = 0x2,
+	MTYPE_UC                                         = 0x3,
+} MTYPE;
 typedef enum PERFMON_COUNTER_MODE {
 	PERFMON_COUNTER_MODE_ACCUM                       = 0x0,
 	PERFMON_COUNTER_MODE_ACTIVE_CYCLES               = 0x1,
diff --git a/drivers/gpu/drm/amd/include/cgs_common.h b/drivers/gpu/drm/amd/include/cgs_common.h
index 713aec954692..aec38fc3834f 100644
--- a/drivers/gpu/drm/amd/include/cgs_common.h
+++ b/drivers/gpu/drm/amd/include/cgs_common.h
@@ -109,6 +109,8 @@ enum cgs_system_info_id {
 	CGS_SYSTEM_INFO_ADAPTER_BDF_ID = 1,
 	CGS_SYSTEM_INFO_PCIE_GEN_INFO,
 	CGS_SYSTEM_INFO_PCIE_MLW,
+	CGS_SYSTEM_INFO_CG_FLAGS,
+	CGS_SYSTEM_INFO_PG_FLAGS,
 	CGS_SYSTEM_INFO_ID_MAXIMUM,
 };
 
diff --git a/drivers/gpu/drm/amd/include/kgd_kfd_interface.h b/drivers/gpu/drm/amd/include/kgd_kfd_interface.h
index 888250b33ea8..a09d9f352871 100644
--- a/drivers/gpu/drm/amd/include/kgd_kfd_interface.h
+++ b/drivers/gpu/drm/amd/include/kgd_kfd_interface.h
@@ -221,7 +221,7 @@ struct kgd2kfd_calls {
 	int (*resume)(struct kfd_dev *kfd);
 };
 
-bool kgd2kfd_init(unsigned interface_version,
+int kgd2kfd_init(unsigned interface_version,
 		const struct kgd2kfd_calls **g2f);
 
 #endif	/* KGD_KFD_INTERFACE_H_INCLUDED */
diff --git a/drivers/gpu/drm/amd/powerplay/amd_powerplay.c b/drivers/gpu/drm/amd/powerplay/amd_powerplay.c
index 2ee4190f8c89..9d2290044708 100644
--- a/drivers/gpu/drm/amd/powerplay/amd_powerplay.c
+++ b/drivers/gpu/drm/amd/powerplay/amd_powerplay.c
@@ -403,8 +403,11 @@ int pp_dpm_dispatch_tasks(void *handle, enum amd_pp_event event_id, void *input,
 
 		data.requested_ui_label = power_state_convert(ps);
 		ret = pem_handle_event(pp_handle->eventmgr, event_id, &data);
+		break;
 	}
-	break;
+	case AMD_PP_EVENT_COMPLETE_INIT:
+		ret = pem_handle_event(pp_handle->eventmgr, event_id, &data);
+		break;
 	default:
 		break;
 	}
@@ -606,7 +609,7 @@ static int pp_dpm_set_pp_table(void *handle, const char *buf, size_t size)
 
 	if (hwmgr == NULL || hwmgr->hwmgr_func == NULL ||
 		hwmgr->hwmgr_func->set_pp_table == NULL)
-			return -EINVAL;
+		return -EINVAL;
 
 	return hwmgr->hwmgr_func->set_pp_table(hwmgr, buf, size);
 }
@@ -623,7 +626,7 @@ static int pp_dpm_force_clock_level(void *handle,
 
 	if (hwmgr == NULL || hwmgr->hwmgr_func == NULL ||
 			hwmgr->hwmgr_func->force_clock_level == NULL)
-				return -EINVAL;
+		return -EINVAL;
 
 	return hwmgr->hwmgr_func->force_clock_level(hwmgr, type, level);
 }
diff --git a/drivers/gpu/drm/amd/powerplay/eventmgr/eventactionchains.c b/drivers/gpu/drm/amd/powerplay/eventmgr/eventactionchains.c
index 83be3cf210e0..6b52c78cb404 100644
--- a/drivers/gpu/drm/amd/powerplay/eventmgr/eventactionchains.c
+++ b/drivers/gpu/drm/amd/powerplay/eventmgr/eventactionchains.c
@@ -165,6 +165,7 @@ const struct action_chain resume_action_chain = {
 };
 
 static const pem_event_action *complete_init_event[] = {
+	unblock_adjust_power_state_tasks,
 	adjust_power_state_tasks,
 	enable_gfx_clock_gating_tasks,
 	enable_gfx_voltage_island_power_gating_tasks,
diff --git a/drivers/gpu/drm/amd/powerplay/eventmgr/eventmgr.c b/drivers/gpu/drm/amd/powerplay/eventmgr/eventmgr.c
index 52a3efc97f05..46410e3c7349 100644
--- a/drivers/gpu/drm/amd/powerplay/eventmgr/eventmgr.c
+++ b/drivers/gpu/drm/amd/powerplay/eventmgr/eventmgr.c
@@ -31,7 +31,7 @@
 static int pem_init(struct pp_eventmgr *eventmgr)
 {
 	int result = 0;
-	struct pem_event_data event_data;
+	struct pem_event_data event_data = { {0} };
 
 	/* Initialize PowerPlay feature info */
 	pem_init_feature_info(eventmgr);
@@ -52,7 +52,7 @@ static int pem_init(struct pp_eventmgr *eventmgr)
 
 static void pem_fini(struct pp_eventmgr *eventmgr)
 {
-	struct pem_event_data event_data;
+	struct pem_event_data event_data = { {0} };
 
 	pem_uninit_featureInfo(eventmgr);
 	pem_unregister_interrupts(eventmgr);
diff --git a/drivers/gpu/drm/amd/powerplay/hwmgr/cz_clockpowergating.c b/drivers/gpu/drm/amd/powerplay/hwmgr/cz_clockpowergating.c
index ad7700822a1c..ff08ce41bde9 100644
--- a/drivers/gpu/drm/amd/powerplay/hwmgr/cz_clockpowergating.c
+++ b/drivers/gpu/drm/amd/powerplay/hwmgr/cz_clockpowergating.c
@@ -226,7 +226,7 @@ int cz_dpm_powergate_vce(struct pp_hwmgr *hwmgr, bool bgate)
 		}
 	} else {
 		cz_dpm_update_vce_dpm(hwmgr);
-		cz_enable_disable_vce_dpm(hwmgr, true);
+		cz_enable_disable_vce_dpm(hwmgr, !bgate);
 		return 0;
 	}
 
diff --git a/drivers/gpu/drm/amd/powerplay/hwmgr/cz_hwmgr.c b/drivers/gpu/drm/amd/powerplay/hwmgr/cz_hwmgr.c
index ef1daf1251c7..5682490337e3 100644
--- a/drivers/gpu/drm/amd/powerplay/hwmgr/cz_hwmgr.c
+++ b/drivers/gpu/drm/amd/powerplay/hwmgr/cz_hwmgr.c
@@ -174,6 +174,8 @@ static int cz_initialize_dpm_defaults(struct pp_hwmgr *hwmgr)
 {
 	struct cz_hwmgr *cz_hwmgr = (struct cz_hwmgr *)(hwmgr->backend);
 	uint32_t i;
+	struct cgs_system_info sys_info = {0};
+	int result;
 
 	cz_hwmgr->gfx_ramp_step = 256*25/100;
 
@@ -239,6 +241,11 @@ static int cz_initialize_dpm_defaults(struct pp_hwmgr *hwmgr)
 	phm_cap_set(hwmgr->platform_descriptor.platformCaps,
 					PHM_PlatformCaps_DynamicUVDState);
 
+	phm_cap_set(hwmgr->platform_descriptor.platformCaps,
+			PHM_PlatformCaps_UVDDPM);
+	phm_cap_set(hwmgr->platform_descriptor.platformCaps,
+			PHM_PlatformCaps_VCEDPM);
+
 	cz_hwmgr->cc6_settings.cpu_cc6_disable = false;
 	cz_hwmgr->cc6_settings.cpu_pstate_disable = false;
 	cz_hwmgr->cc6_settings.nb_pstate_switch_disable = false;
@@ -247,6 +254,22 @@ static int cz_initialize_dpm_defaults(struct pp_hwmgr *hwmgr)
 	phm_cap_set(hwmgr->platform_descriptor.platformCaps,
 				   PHM_PlatformCaps_DisableVoltageIsland);
 
+	phm_cap_unset(hwmgr->platform_descriptor.platformCaps,
+		      PHM_PlatformCaps_UVDPowerGating);
+	phm_cap_unset(hwmgr->platform_descriptor.platformCaps,
+		      PHM_PlatformCaps_VCEPowerGating);
+	sys_info.size = sizeof(struct cgs_system_info);
+	sys_info.info_id = CGS_SYSTEM_INFO_PG_FLAGS;
+	result = cgs_query_system_info(hwmgr->device, &sys_info);
+	if (!result) {
+		if (sys_info.value & AMD_PG_SUPPORT_UVD)
+			phm_cap_set(hwmgr->platform_descriptor.platformCaps,
+				      PHM_PlatformCaps_UVDPowerGating);
+		if (sys_info.value & AMD_PG_SUPPORT_VCE)
+			phm_cap_set(hwmgr->platform_descriptor.platformCaps,
+				      PHM_PlatformCaps_VCEPowerGating);
+	}
+
 	return 0;
 }
 
@@ -726,8 +749,9 @@ static int cz_tf_update_sclk_limit(struct pp_hwmgr *hwmgr,
 		cz_hwmgr->sclk_dpm.soft_max_clk  = table->entries[table->count - 1].clk;
 
 	clock = hwmgr->display_config.min_core_set_clock;
+;
 	if (clock == 0)
-		printk(KERN_ERR "[ powerplay ] min_core_set_clock not set\n");
+		printk(KERN_INFO "[ powerplay ] min_core_set_clock not set\n");
 
 	if (cz_hwmgr->sclk_dpm.hard_min_clk != clock) {
 		cz_hwmgr->sclk_dpm.hard_min_clk = clock;
diff --git a/drivers/gpu/drm/amd/powerplay/hwmgr/fiji_hwmgr.c b/drivers/gpu/drm/amd/powerplay/hwmgr/fiji_hwmgr.c
index 5cca2ecc6bea..51dedf84623c 100644
--- a/drivers/gpu/drm/amd/powerplay/hwmgr/fiji_hwmgr.c
+++ b/drivers/gpu/drm/amd/powerplay/hwmgr/fiji_hwmgr.c
@@ -4275,7 +4275,6 @@ static int fiji_populate_and_upload_sclk_mclk_dpm_levels(
 	if (data->need_update_smu7_dpm_table & DPMTABLE_OD_UPDATE_MCLK) {
 		dpm_table->mclk_table.dpm_levels
 			[dpm_table->mclk_table.count - 1].value = mclk;
-
 		if (phm_cap_enabled(hwmgr->platform_descriptor.platformCaps,
 				PHM_PlatformCaps_OD6PlusinACSupport) ||
 			phm_cap_enabled(hwmgr->platform_descriptor.platformCaps,
@@ -4886,6 +4885,10 @@ static void fiji_print_current_perforce_level(
 	activity_percent >>= 8;
 
 	seq_printf(m, "\n [GPU load]: %u%%\n\n", activity_percent > 100 ? 100 : activity_percent);
+
+	seq_printf(m, "uvd    %sabled\n", data->uvd_power_gated ? "dis" : "en");
+
+	seq_printf(m, "vce    %sabled\n", data->vce_power_gated ? "dis" : "en");
 }
 
 static int fiji_program_display_gap(struct pp_hwmgr *hwmgr)
diff --git a/drivers/gpu/drm/amd/powerplay/hwmgr/fiji_hwmgr.h b/drivers/gpu/drm/amd/powerplay/hwmgr/fiji_hwmgr.h
index 22e273b1c1c5..a16f7cd4c238 100644
--- a/drivers/gpu/drm/amd/powerplay/hwmgr/fiji_hwmgr.h
+++ b/drivers/gpu/drm/amd/powerplay/hwmgr/fiji_hwmgr.h
@@ -29,6 +29,7 @@
 #include "smu73_discrete.h"
 #include "ppatomctrl.h"
 #include "fiji_ppsmc.h"
+#include "pp_endian.h"
 
 #define FIJI_MAX_HARDWARE_POWERLEVELS	2
 #define FIJI_AT_DFLT	30
@@ -347,15 +348,4 @@ int fiji_update_samu_dpm(struct pp_hwmgr *hwmgr, bool bgate);
 int fiji_update_acp_dpm(struct pp_hwmgr *hwmgr, bool bgate);
 int fiji_enable_disable_vce_dpm(struct pp_hwmgr *hwmgr, bool enable);
 
-#define PP_HOST_TO_SMC_UL(X) cpu_to_be32(X)
-#define PP_SMC_TO_HOST_UL(X) be32_to_cpu(X)
-
-#define PP_HOST_TO_SMC_US(X) cpu_to_be16(X)
-#define PP_SMC_TO_HOST_US(X) be16_to_cpu(X)
-
-#define CONVERT_FROM_HOST_TO_SMC_UL(X) ((X) = PP_HOST_TO_SMC_UL(X))
-#define CONVERT_FROM_SMC_TO_HOST_UL(X) ((X) = PP_SMC_TO_HOST_UL(X))
-
-#define CONVERT_FROM_HOST_TO_SMC_US(X) ((X) = PP_HOST_TO_SMC_US(X))
-
 #endif /* _FIJI_HWMGR_H_ */
diff --git a/drivers/gpu/drm/amd/powerplay/hwmgr/functiontables.c b/drivers/gpu/drm/amd/powerplay/hwmgr/functiontables.c
index 9deadabbc81c..72cfecc4f9f7 100644
--- a/drivers/gpu/drm/amd/powerplay/hwmgr/functiontables.c
+++ b/drivers/gpu/drm/amd/powerplay/hwmgr/functiontables.c
@@ -34,6 +34,11 @@ static int phm_run_table(struct pp_hwmgr *hwmgr,
 	int result = 0;
 	phm_table_function *function;
 
+	if (rt_table->function_list == NULL) {
+		printk(KERN_INFO "[ powerplay ] this function not implement!\n");
+		return 0;
+	}
+
 	for (function = rt_table->function_list; NULL != *function; function++) {
 		int tmp = (*function)(hwmgr, input, output, temp_storage, result);
 
@@ -57,9 +62,9 @@ int phm_dispatch_table(struct pp_hwmgr *hwmgr,
 	int result = 0;
 	void *temp_storage = NULL;
 
-	if (hwmgr == NULL || rt_table == NULL || rt_table->function_list == NULL) {
+	if (hwmgr == NULL || rt_table == NULL) {
 		printk(KERN_ERR "[ powerplay ] Invalid Parameter!\n");
-		return 0; /*temp return ture because some function not implement on some asic */
+		return -EINVAL;
 	}
 
 	if (0 != rt_table->storage_size) {
diff --git a/drivers/gpu/drm/amd/powerplay/hwmgr/tonga_hwmgr.c b/drivers/gpu/drm/amd/powerplay/hwmgr/tonga_hwmgr.c
index bc83fa35ec46..0d5d8372953e 100644
--- a/drivers/gpu/drm/amd/powerplay/hwmgr/tonga_hwmgr.c
+++ b/drivers/gpu/drm/amd/powerplay/hwmgr/tonga_hwmgr.c
@@ -4451,6 +4451,7 @@ int tonga_hwmgr_backend_init(struct pp_hwmgr *hwmgr)
 	pp_atomctrl_gpio_pin_assignment gpio_pin_assignment;
 	struct phm_ppt_v1_information *pptable_info = (struct phm_ppt_v1_information *)(hwmgr->pptable);
 	phw_tonga_ulv_parm *ulv;
+	struct cgs_system_info sys_info = {0};
 
 	PP_ASSERT_WITH_CODE((NULL != hwmgr),
 		"Invalid Parameter!", return -1;);
@@ -4615,9 +4616,23 @@ int tonga_hwmgr_backend_init(struct pp_hwmgr *hwmgr)
 
 	data->vddc_phase_shed_control = 0;
 
-	if (0 == result) {
-		struct cgs_system_info sys_info = {0};
+	phm_cap_unset(hwmgr->platform_descriptor.platformCaps,
+		      PHM_PlatformCaps_UVDPowerGating);
+	phm_cap_unset(hwmgr->platform_descriptor.platformCaps,
+		      PHM_PlatformCaps_VCEPowerGating);
+	sys_info.size = sizeof(struct cgs_system_info);
+	sys_info.info_id = CGS_SYSTEM_INFO_PG_FLAGS;
+	result = cgs_query_system_info(hwmgr->device, &sys_info);
+	if (!result) {
+		if (sys_info.value & AMD_PG_SUPPORT_UVD)
+			phm_cap_set(hwmgr->platform_descriptor.platformCaps,
+				      PHM_PlatformCaps_UVDPowerGating);
+		if (sys_info.value & AMD_PG_SUPPORT_VCE)
+			phm_cap_set(hwmgr->platform_descriptor.platformCaps,
+				      PHM_PlatformCaps_VCEPowerGating);
+	}
 
+	if (0 == result) {
 		data->is_tlu_enabled = 0;
 		hwmgr->platform_descriptor.hardwareActivityPerformanceLevels =
 			TONGA_MAX_HARDWARE_POWERLEVELS;
@@ -5170,7 +5185,6 @@ tonga_print_current_perforce_level(struct pp_hwmgr *hwmgr, struct seq_file *m)
 	mclk = cgs_read_register(hwmgr->device, mmSMC_MSG_ARG_0);
 	seq_printf(m, "\n [  mclk  ]: %u MHz\n\n [  sclk  ]: %u MHz\n", mclk/100, sclk/100);
 
-
 	offset = data->soft_regs_start + offsetof(SMU72_SoftRegisters, AverageGraphicsActivity);
 	activity_percent = cgs_read_ind_register(hwmgr->device, CGS_IND_REG__SMC, offset);
 	activity_percent += 0x80;
@@ -5178,6 +5192,9 @@ tonga_print_current_perforce_level(struct pp_hwmgr *hwmgr, struct seq_file *m)
 
 	seq_printf(m, "\n [GPU load]: %u%%\n\n", activity_percent > 100 ? 100 : activity_percent);
 
+	seq_printf(m, "uvd    %sabled\n", data->uvd_power_gated ? "dis" : "en");
+
+	seq_printf(m, "vce    %sabled\n", data->vce_power_gated ? "dis" : "en");
 }
 
 static int tonga_find_dpm_states_clocks_in_dpm_table(struct pp_hwmgr *hwmgr, const void *input)
diff --git a/drivers/gpu/drm/amd/powerplay/hwmgr/tonga_hwmgr.h b/drivers/gpu/drm/amd/powerplay/hwmgr/tonga_hwmgr.h
index 49168d262ccc..f88d3bbe6671 100644
--- a/drivers/gpu/drm/amd/powerplay/hwmgr/tonga_hwmgr.h
+++ b/drivers/gpu/drm/amd/powerplay/hwmgr/tonga_hwmgr.h
@@ -28,6 +28,7 @@
 #include "ppatomctrl.h"
 #include "ppinterrupt.h"
 #include "tonga_powertune.h"
+#include "pp_endian.h"
 
 #define TONGA_MAX_HARDWARE_POWERLEVELS 2
 #define TONGA_DYNCLK_NUMBER_OF_TREND_COEFFICIENTS 15
@@ -386,17 +387,6 @@ typedef struct tonga_hwmgr tonga_hwmgr;
 
 #define TONGA_UNUSED_GPIO_PIN                        0x7F
 
-#define PP_HOST_TO_SMC_UL(X) cpu_to_be32(X)
-#define PP_SMC_TO_HOST_UL(X) be32_to_cpu(X)
-
-#define PP_HOST_TO_SMC_US(X) cpu_to_be16(X)
-#define PP_SMC_TO_HOST_US(X) be16_to_cpu(X)
-
-#define CONVERT_FROM_HOST_TO_SMC_UL(X) ((X) = PP_HOST_TO_SMC_UL(X))
-#define CONVERT_FROM_SMC_TO_HOST_UL(X) ((X) = PP_SMC_TO_HOST_UL(X))
-
-#define CONVERT_FROM_HOST_TO_SMC_US(X) ((X) = PP_HOST_TO_SMC_US(X))
-
 int tonga_hwmgr_init(struct pp_hwmgr *hwmgr);
 int tonga_update_vce_dpm(struct pp_hwmgr *hwmgr, const void *input);
 int tonga_update_uvd_dpm(struct pp_hwmgr *hwmgr, bool bgate);
diff --git a/drivers/gpu/drm/amd/powerplay/inc/pp_endian.h b/drivers/gpu/drm/amd/powerplay/inc/pp_endian.h
new file mode 100644
index 000000000000..f49d1963fe85
--- /dev/null
+++ b/drivers/gpu/drm/amd/powerplay/inc/pp_endian.h
@@ -0,0 +1,38 @@
+/*
+ * Copyright 2016 Advanced Micro Devices, Inc.
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
+ * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR
+ * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
+ * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
+ * OTHER DEALINGS IN THE SOFTWARE.
+ *
+ */
+
+#ifndef _PP_ENDIAN_H_
+#define _PP_ENDIAN_H_
+
+#define PP_HOST_TO_SMC_UL(X) cpu_to_be32(X)
+#define PP_SMC_TO_HOST_UL(X) be32_to_cpu(X)
+
+#define PP_HOST_TO_SMC_US(X) cpu_to_be16(X)
+#define PP_SMC_TO_HOST_US(X) be16_to_cpu(X)
+
+#define CONVERT_FROM_HOST_TO_SMC_UL(X) ((X) = PP_HOST_TO_SMC_UL(X))
+#define CONVERT_FROM_SMC_TO_HOST_UL(X) ((X) = PP_SMC_TO_HOST_UL(X))
+
+#define CONVERT_FROM_HOST_TO_SMC_US(X) ((X) = PP_HOST_TO_SMC_US(X))
+
+#endif /* _PP_ENDIAN_H_ */
diff --git a/drivers/gpu/drm/amd/powerplay/inc/smumgr.h b/drivers/gpu/drm/amd/powerplay/inc/smumgr.h
index 504f035d1843..fc9e3d1dd409 100644
--- a/drivers/gpu/drm/amd/powerplay/inc/smumgr.h
+++ b/drivers/gpu/drm/amd/powerplay/inc/smumgr.h
@@ -32,6 +32,27 @@ struct pp_instance;
 #define smu_lower_32_bits(n) ((uint32_t)(n))
 #define smu_upper_32_bits(n) ((uint32_t)(((n)>>16)>>16))
 
+enum AVFS_BTC_STATUS {
+	AVFS_BTC_BOOT = 0,
+	AVFS_BTC_BOOT_STARTEDSMU,
+	AVFS_LOAD_VIRUS,
+	AVFS_BTC_VIRUS_LOADED,
+	AVFS_BTC_VIRUS_FAIL,
+	AVFS_BTC_COMPLETED_PREVIOUSLY,
+	AVFS_BTC_ENABLEAVFS,
+	AVFS_BTC_STARTED,
+	AVFS_BTC_FAILED,
+	AVFS_BTC_RESTOREVFT_FAILED,
+	AVFS_BTC_SAVEVFT_FAILED,
+	AVFS_BTC_DPMTABLESETUP_FAILED,
+	AVFS_BTC_COMPLETED_UNSAVED,
+	AVFS_BTC_COMPLETED_SAVED,
+	AVFS_BTC_COMPLETED_RESTORED,
+	AVFS_BTC_DISABLED,
+	AVFS_BTC_NOTSUPPORTED,
+	AVFS_BTC_SMUMSG_ERROR
+};
+
 struct pp_smumgr_func {
 	int (*smu_init)(struct pp_smumgr *smumgr);
 	int (*smu_fini)(struct pp_smumgr *smumgr);
diff --git a/drivers/gpu/drm/amd/powerplay/smumgr/fiji_smumgr.h b/drivers/gpu/drm/amd/powerplay/smumgr/fiji_smumgr.h
index 8cd22d9c9140..b4eb483215b1 100644
--- a/drivers/gpu/drm/amd/powerplay/smumgr/fiji_smumgr.h
+++ b/drivers/gpu/drm/amd/powerplay/smumgr/fiji_smumgr.h
@@ -23,24 +23,6 @@
 #ifndef _FIJI_SMUMANAGER_H_
 #define _FIJI_SMUMANAGER_H_
 
-enum AVFS_BTC_STATUS {
-	AVFS_BTC_BOOT = 0,
-	AVFS_BTC_BOOT_STARTEDSMU,
-	AVFS_LOAD_VIRUS,
-	AVFS_BTC_VIRUS_LOADED,
-	AVFS_BTC_VIRUS_FAIL,
-	AVFS_BTC_STARTED,
-	AVFS_BTC_FAILED,
-	AVFS_BTC_RESTOREVFT_FAILED,
-	AVFS_BTC_SAVEVFT_FAILED,
-	AVFS_BTC_DPMTABLESETUP_FAILED,
-	AVFS_BTC_COMPLETED_UNSAVED,
-	AVFS_BTC_COMPLETED_SAVED,
-	AVFS_BTC_COMPLETED_RESTORED,
-	AVFS_BTC_DISABLED,
-	AVFS_BTC_NOTSUPPORTED,
-	AVFS_BTC_SMUMSG_ERROR
-};
 
 struct fiji_smu_avfs {
 	enum AVFS_BTC_STATUS AvfsBtcStatus;
diff --git a/drivers/gpu/drm/amd/scheduler/sched_fence.c b/drivers/gpu/drm/amd/scheduler/sched_fence.c
index 87c78eecea64..dc115aea352b 100644
--- a/drivers/gpu/drm/amd/scheduler/sched_fence.c
+++ b/drivers/gpu/drm/amd/scheduler/sched_fence.c
@@ -84,12 +84,33 @@ static bool amd_sched_fence_enable_signaling(struct fence *f)
 	return true;
 }
 
-static void amd_sched_fence_release(struct fence *f)
+/**
+ * amd_sched_fence_free - free up the fence memory
+ *
+ * @rcu: RCU callback head
+ *
+ * Free up the fence memory after the RCU grace period.
+ */
+static void amd_sched_fence_free(struct rcu_head *rcu)
 {
+	struct fence *f = container_of(rcu, struct fence, rcu);
 	struct amd_sched_fence *fence = to_amd_sched_fence(f);
 	kmem_cache_free(sched_fence_slab, fence);
 }
 
+/**
+ * amd_sched_fence_release - callback that fence can be freed
+ *
+ * @fence: fence
+ *
+ * This function is called when the reference count becomes zero.
+ * It just RCU schedules freeing up the fence.
+ */
+static void amd_sched_fence_release(struct fence *f)
+{
+	call_rcu(&f->rcu, amd_sched_fence_free);
+}
+
 const struct fence_ops amd_sched_fence_ops = {
 	.get_driver_name = amd_sched_fence_get_driver_name,
 	.get_timeline_name = amd_sched_fence_get_timeline_name,
diff --git a/drivers/gpu/drm/ast/ast_main.c b/drivers/gpu/drm/ast/ast_main.c
index 9759009d1da3..b1480acbb3c3 100644
--- a/drivers/gpu/drm/ast/ast_main.c
+++ b/drivers/gpu/drm/ast/ast_main.c
@@ -227,7 +227,7 @@ static int ast_get_dram_info(struct drm_device *dev)
 	} while (ast_read32(ast, 0x10000) != 0x01);
 	data = ast_read32(ast, 0x10004);
 
-	if (data & 0x400)
+	if (data & 0x40)
 		ast->dram_bus_width = 16;
 	else
 		ast->dram_bus_width = 32;
diff --git a/drivers/gpu/drm/drm_atomic.c b/drivers/gpu/drm/drm_atomic.c
index d78f8ea5167e..8ee1db866e80 100644
--- a/drivers/gpu/drm/drm_atomic.c
+++ b/drivers/gpu/drm/drm_atomic.c
@@ -66,8 +66,6 @@ drm_atomic_state_init(struct drm_device *dev, struct drm_atomic_state *state)
 	 */
 	state->allow_modeset = true;
 
-	state->num_connector = ACCESS_ONCE(dev->mode_config.num_connector);
-
 	state->crtcs = kcalloc(dev->mode_config.num_crtc,
 			       sizeof(*state->crtcs), GFP_KERNEL);
 	if (!state->crtcs)
@@ -84,16 +82,6 @@ drm_atomic_state_init(struct drm_device *dev, struct drm_atomic_state *state)
 				      sizeof(*state->plane_states), GFP_KERNEL);
 	if (!state->plane_states)
 		goto fail;
-	state->connectors = kcalloc(state->num_connector,
-				    sizeof(*state->connectors),
-				    GFP_KERNEL);
-	if (!state->connectors)
-		goto fail;
-	state->connector_states = kcalloc(state->num_connector,
-					  sizeof(*state->connector_states),
-					  GFP_KERNEL);
-	if (!state->connector_states)
-		goto fail;
 
 	state->dev = dev;
 
@@ -906,19 +894,27 @@ drm_atomic_get_connector_state(struct drm_atomic_state *state,
 
 	index = drm_connector_index(connector);
 
-	/*
-	 * Construction of atomic state updates can race with a connector
-	 * hot-add which might overflow. In this case flip the table and just
-	 * restart the entire ioctl - no one is fast enough to livelock a cpu
-	 * with physical hotplug events anyway.
-	 *
-	 * Note that we only grab the indexes once we have the right lock to
-	 * prevent hotplug/unplugging of connectors. So removal is no problem,
-	 * at most the array is a bit too large.
-	 */
 	if (index >= state->num_connector) {
-		DRM_DEBUG_ATOMIC("Hot-added connector would overflow state array, restarting\n");
-		return ERR_PTR(-EAGAIN);
+		struct drm_connector **c;
+		struct drm_connector_state **cs;
+		int alloc = max(index + 1, config->num_connector);
+
+		c = krealloc(state->connectors, alloc * sizeof(*state->connectors), GFP_KERNEL);
+		if (!c)
+			return ERR_PTR(-ENOMEM);
+
+		state->connectors = c;
+		memset(&state->connectors[state->num_connector], 0,
+		       sizeof(*state->connectors) * (alloc - state->num_connector));
+
+		cs = krealloc(state->connector_states, alloc * sizeof(*state->connector_states), GFP_KERNEL);
+		if (!cs)
+			return ERR_PTR(-ENOMEM);
+
+		state->connector_states = cs;
+		memset(&state->connector_states[state->num_connector], 0,
+		       sizeof(*state->connector_states) * (alloc - state->num_connector));
+		state->num_connector = alloc;
 	}
 
 	if (state->connector_states[index])
diff --git a/drivers/gpu/drm/drm_atomic_helper.c b/drivers/gpu/drm/drm_atomic_helper.c
index 91351111d0f9..4befe25c81c7 100644
--- a/drivers/gpu/drm/drm_atomic_helper.c
+++ b/drivers/gpu/drm/drm_atomic_helper.c
@@ -1550,7 +1550,7 @@ void drm_atomic_helper_swap_state(struct drm_device *dev,
 {
 	int i;
 
-	for (i = 0; i < dev->mode_config.num_connector; i++) {
+	for (i = 0; i < state->num_connector; i++) {
 		struct drm_connector *connector = state->connectors[i];
 
 		if (!connector)
diff --git a/drivers/gpu/drm/drm_bridge.c b/drivers/gpu/drm/drm_bridge.c
index bd93453afa61..b3654404abd0 100644
--- a/drivers/gpu/drm/drm_bridge.c
+++ b/drivers/gpu/drm/drm_bridge.c
@@ -186,7 +186,8 @@ void drm_bridge_disable(struct drm_bridge *bridge)
 
 	drm_bridge_disable(bridge->next);
 
-	bridge->funcs->disable(bridge);
+	if (bridge->funcs->disable)
+		bridge->funcs->disable(bridge);
 }
 EXPORT_SYMBOL(drm_bridge_disable);
 
@@ -206,7 +207,8 @@ void drm_bridge_post_disable(struct drm_bridge *bridge)
 	if (!bridge)
 		return;
 
-	bridge->funcs->post_disable(bridge);
+	if (bridge->funcs->post_disable)
+		bridge->funcs->post_disable(bridge);
 
 	drm_bridge_post_disable(bridge->next);
 }
@@ -256,7 +258,8 @@ void drm_bridge_pre_enable(struct drm_bridge *bridge)
 
 	drm_bridge_pre_enable(bridge->next);
 
-	bridge->funcs->pre_enable(bridge);
+	if (bridge->funcs->pre_enable)
+		bridge->funcs->pre_enable(bridge);
 }
 EXPORT_SYMBOL(drm_bridge_pre_enable);
 
@@ -276,7 +279,8 @@ void drm_bridge_enable(struct drm_bridge *bridge)
 	if (!bridge)
 		return;
 
-	bridge->funcs->enable(bridge);
+	if (bridge->funcs->enable)
+		bridge->funcs->enable(bridge);
 
 	drm_bridge_enable(bridge->next);
 }
diff --git a/drivers/gpu/drm/drm_crtc.c b/drivers/gpu/drm/drm_crtc.c
index 49781a9f8a94..e08f962288d9 100644
--- a/drivers/gpu/drm/drm_crtc.c
+++ b/drivers/gpu/drm/drm_crtc.c
@@ -916,12 +916,19 @@ int drm_connector_init(struct drm_device *dev,
 	connector->base.properties = &connector->properties;
 	connector->dev = dev;
 	connector->funcs = funcs;
+
+	connector->connector_id = ida_simple_get(&config->connector_ida, 0, 0, GFP_KERNEL);
+	if (connector->connector_id < 0) {
+		ret = connector->connector_id;
+		goto out_put;
+	}
+
 	connector->connector_type = connector_type;
 	connector->connector_type_id =
 		ida_simple_get(connector_ida, 1, 0, GFP_KERNEL);
 	if (connector->connector_type_id < 0) {
 		ret = connector->connector_type_id;
-		goto out_put;
+		goto out_put_id;
 	}
 	connector->name =
 		kasprintf(GFP_KERNEL, "%s-%d",
@@ -929,7 +936,7 @@ int drm_connector_init(struct drm_device *dev,
 			  connector->connector_type_id);
 	if (!connector->name) {
 		ret = -ENOMEM;
-		goto out_put;
+		goto out_put_type_id;
 	}
 
 	INIT_LIST_HEAD(&connector->probed_modes);
@@ -957,7 +964,12 @@ int drm_connector_init(struct drm_device *dev,
 	}
 
 	connector->debugfs_entry = NULL;
-
+out_put_type_id:
+	if (ret)
+		ida_remove(connector_ida, connector->connector_type_id);
+out_put_id:
+	if (ret)
+		ida_remove(&config->connector_ida, connector->connector_id);
 out_put:
 	if (ret)
 		drm_mode_object_put(dev, &connector->base);
@@ -994,6 +1006,9 @@ void drm_connector_cleanup(struct drm_connector *connector)
 	ida_remove(&drm_connector_enum_list[connector->connector_type].ida,
 		   connector->connector_type_id);
 
+	ida_remove(&dev->mode_config.connector_ida,
+		   connector->connector_id);
+
 	kfree(connector->display_info.bus_formats);
 	drm_mode_object_put(dev, &connector->base);
 	kfree(connector->name);
@@ -1011,32 +1026,6 @@ void drm_connector_cleanup(struct drm_connector *connector)
 EXPORT_SYMBOL(drm_connector_cleanup);
 
 /**
- * drm_connector_index - find the index of a registered connector
- * @connector: connector to find index for
- *
- * Given a registered connector, return the index of that connector within a DRM
- * device's list of connectors.
- */
-unsigned int drm_connector_index(struct drm_connector *connector)
-{
-	unsigned int index = 0;
-	struct drm_connector *tmp;
-	struct drm_mode_config *config = &connector->dev->mode_config;
-
-	WARN_ON(!drm_modeset_is_locked(&config->connection_mutex));
-
-	drm_for_each_connector(tmp, connector->dev) {
-		if (tmp == connector)
-			return index;
-
-		index++;
-	}
-
-	BUG();
-}
-EXPORT_SYMBOL(drm_connector_index);
-
-/**
  * drm_connector_register - register a connector
  * @connector: the connector to register
  *
@@ -5871,6 +5860,7 @@ void drm_mode_config_init(struct drm_device *dev)
 	INIT_LIST_HEAD(&dev->mode_config.plane_list);
 	idr_init(&dev->mode_config.crtc_idr);
 	idr_init(&dev->mode_config.tile_idr);
+	ida_init(&dev->mode_config.connector_ida);
 
 	drm_modeset_lock_all(dev);
 	drm_mode_create_standard_properties(dev);
@@ -5951,6 +5941,7 @@ void drm_mode_config_cleanup(struct drm_device *dev)
 		crtc->funcs->destroy(crtc);
 	}
 
+	ida_destroy(&dev->mode_config.connector_ida);
 	idr_destroy(&dev->mode_config.tile_idr);
 	idr_destroy(&dev->mode_config.crtc_idr);
 	drm_modeset_lock_fini(&dev->mode_config.connection_mutex);
diff --git a/drivers/gpu/drm/drm_dp_mst_topology.c b/drivers/gpu/drm/drm_dp_mst_topology.c
index 8ae13de272c4..27fbd79d0daf 100644
--- a/drivers/gpu/drm/drm_dp_mst_topology.c
+++ b/drivers/gpu/drm/drm_dp_mst_topology.c
@@ -1159,11 +1159,13 @@ static void drm_dp_add_port(struct drm_dp_mst_branch *mstb,
 			drm_dp_put_port(port);
 			goto out;
 		}
-
-		drm_mode_connector_set_tile_property(port->connector);
-
+		if (port->port_num >= DP_MST_LOGICAL_PORT_0) {
+			port->cached_edid = drm_get_edid(port->connector, &port->aux.ddc);
+			drm_mode_connector_set_tile_property(port->connector);
+		}
 		(*mstb->mgr->cbs->register_connector)(port->connector);
 	}
+
 out:
 	/* put reference to this port */
 	drm_dp_put_port(port);
@@ -1188,8 +1190,8 @@ static void drm_dp_update_port(struct drm_dp_mst_branch *mstb,
 	port->ddps = conn_stat->displayport_device_plug_status;
 
 	if (old_ddps != port->ddps) {
-		dowork = true;
 		if (port->ddps) {
+			dowork = true;
 		} else {
 			port->available_pbn = 0;
 		}
@@ -1294,13 +1296,8 @@ static void drm_dp_check_and_send_link_address(struct drm_dp_mst_topology_mgr *m
 		if (port->input)
 			continue;
 
-		if (!port->ddps) {
-			if (port->cached_edid) {
-				kfree(port->cached_edid);
-				port->cached_edid = NULL;
-			}
+		if (!port->ddps)
 			continue;
-		}
 
 		if (!port->available_pbn)
 			drm_dp_send_enum_path_resources(mgr, mstb, port);
@@ -1311,12 +1308,6 @@ static void drm_dp_check_and_send_link_address(struct drm_dp_mst_topology_mgr *m
 				drm_dp_check_and_send_link_address(mgr, mstb_child);
 				drm_dp_put_mst_branch_device(mstb_child);
 			}
-		} else if (port->pdt == DP_PEER_DEVICE_SST_SINK ||
-			port->pdt == DP_PEER_DEVICE_DP_LEGACY_CONV) {
-			if (!port->cached_edid) {
-				port->cached_edid =
-					drm_get_edid(port->connector, &port->aux.ddc);
-			}
 		}
 	}
 }
@@ -1336,8 +1327,6 @@ static void drm_dp_mst_link_probe_work(struct work_struct *work)
 		drm_dp_check_and_send_link_address(mgr, mstb);
 		drm_dp_put_mst_branch_device(mstb);
 	}
-
-	(*mgr->cbs->hotplug)(mgr);
 }
 
 static bool drm_dp_validate_guid(struct drm_dp_mst_topology_mgr *mgr,
@@ -1597,6 +1586,7 @@ static void drm_dp_send_link_address(struct drm_dp_mst_topology_mgr *mgr,
 			for (i = 0; i < txmsg->reply.u.link_addr.nports; i++) {
 				drm_dp_add_port(mstb, mgr->dev, &txmsg->reply.u.link_addr.ports[i]);
 			}
+			(*mgr->cbs->hotplug)(mgr);
 		}
 	} else {
 		mstb->link_address_sent = false;
@@ -2293,6 +2283,8 @@ static int drm_dp_mst_handle_up_req(struct drm_dp_mst_topology_mgr *mgr)
 			drm_dp_update_port(mstb, &msg.u.conn_stat);
 
 			DRM_DEBUG_KMS("Got CSN: pn: %d ldps:%d ddps: %d mcs: %d ip: %d pdt: %d\n", msg.u.conn_stat.port_number, msg.u.conn_stat.legacy_device_plug_status, msg.u.conn_stat.displayport_device_plug_status, msg.u.conn_stat.message_capability_status, msg.u.conn_stat.input_port, msg.u.conn_stat.peer_device_type);
+			(*mgr->cbs->hotplug)(mgr);
+
 		} else if (msg.req_type == DP_RESOURCE_STATUS_NOTIFY) {
 			drm_dp_send_up_ack_reply(mgr, mgr->mst_primary, msg.req_type, seqno, false);
 			if (!mstb)
@@ -2379,6 +2371,10 @@ enum drm_connector_status drm_dp_mst_detect_port(struct drm_connector *connector
 
 	case DP_PEER_DEVICE_SST_SINK:
 		status = connector_status_connected;
+		/* for logical ports - cache the EDID */
+		if (port->port_num >= 8 && !port->cached_edid) {
+			port->cached_edid = drm_get_edid(connector, &port->aux.ddc);
+		}
 		break;
 	case DP_PEER_DEVICE_DP_LEGACY_CONV:
 		if (port->ldps)
@@ -2433,7 +2429,10 @@ struct edid *drm_dp_mst_get_edid(struct drm_connector *connector, struct drm_dp_
 
 	if (port->cached_edid)
 		edid = drm_edid_duplicate(port->cached_edid);
-
+	else {
+		edid = drm_get_edid(connector, &port->aux.ddc);
+		drm_mode_connector_set_tile_property(connector);
+	}
 	port->has_audio = drm_detect_monitor_audio(edid);
 	drm_dp_put_port(port);
 	return edid;
diff --git a/drivers/gpu/drm/drm_irq.c b/drivers/gpu/drm/drm_irq.c
index 96d03ac38ef7..881c5a6c180c 100644
--- a/drivers/gpu/drm/drm_irq.c
+++ b/drivers/gpu/drm/drm_irq.c
@@ -224,6 +224,64 @@ static void drm_update_vblank_count(struct drm_device *dev, unsigned int pipe,
 		diff = (flags & DRM_CALLED_FROM_VBLIRQ) != 0;
 	}
 
+	/*
+	 * Within a drm_vblank_pre_modeset - drm_vblank_post_modeset
+	 * interval? If so then vblank irqs keep running and it will likely
+	 * happen that the hardware vblank counter is not trustworthy as it
+	 * might reset at some point in that interval and vblank timestamps
+	 * are not trustworthy either in that interval. Iow. this can result
+	 * in a bogus diff >> 1 which must be avoided as it would cause
+	 * random large forward jumps of the software vblank counter.
+	 */
+	if (diff > 1 && (vblank->inmodeset & 0x2)) {
+		DRM_DEBUG_VBL("clamping vblank bump to 1 on crtc %u: diffr=%u"
+			      " due to pre-modeset.\n", pipe, diff);
+		diff = 1;
+	}
+
+	/*
+	 * FIMXE: Need to replace this hack with proper seqlocks.
+	 *
+	 * Restrict the bump of the software vblank counter to a safe maximum
+	 * value of +1 whenever there is the possibility that concurrent readers
+	 * of vblank timestamps could be active at the moment, as the current
+	 * implementation of the timestamp caching and updating is not safe
+	 * against concurrent readers for calls to store_vblank() with a bump
+	 * of anything but +1. A bump != 1 would very likely return corrupted
+	 * timestamps to userspace, because the same slot in the cache could
+	 * be concurrently written by store_vblank() and read by one of those
+	 * readers without the read-retry logic detecting the collision.
+	 *
+	 * Concurrent readers can exist when we are called from the
+	 * drm_vblank_off() or drm_vblank_on() functions and other non-vblank-
+	 * irq callers. However, all those calls to us are happening with the
+	 * vbl_lock locked to prevent drm_vblank_get(), so the vblank refcount
+	 * can't increase while we are executing. Therefore a zero refcount at
+	 * this point is safe for arbitrary counter bumps if we are called
+	 * outside vblank irq, a non-zero count is not 100% safe. Unfortunately
+	 * we must also accept a refcount of 1, as whenever we are called from
+	 * drm_vblank_get() -> drm_vblank_enable() the refcount will be 1 and
+	 * we must let that one pass through in order to not lose vblank counts
+	 * during vblank irq off - which would completely defeat the whole
+	 * point of this routine.
+	 *
+	 * Whenever we are called from vblank irq, we have to assume concurrent
+	 * readers exist or can show up any time during our execution, even if
+	 * the refcount is currently zero, as vblank irqs are usually only
+	 * enabled due to the presence of readers, and because when we are called
+	 * from vblank irq we can't hold the vbl_lock to protect us from sudden
+	 * bumps in vblank refcount. Therefore also restrict bumps to +1 when
+	 * called from vblank irq.
+	 */
+	if ((diff > 1) && (atomic_read(&vblank->refcount) > 1 ||
+	    (flags & DRM_CALLED_FROM_VBLIRQ))) {
+		DRM_DEBUG_VBL("clamping vblank bump to 1 on crtc %u: diffr=%u "
+			      "refcount %u, vblirq %u\n", pipe, diff,
+			      atomic_read(&vblank->refcount),
+			      (flags & DRM_CALLED_FROM_VBLIRQ) != 0);
+		diff = 1;
+	}
+
 	DRM_DEBUG_VBL("updating vblank count on crtc %u:"
 		      " current=%u, diff=%u, hw=%u hw_last=%u\n",
 		      pipe, vblank->count, diff, cur_vblank, vblank->last);
@@ -1313,7 +1371,13 @@ void drm_vblank_off(struct drm_device *dev, unsigned int pipe)
 	spin_lock_irqsave(&dev->event_lock, irqflags);
 
 	spin_lock(&dev->vbl_lock);
-	vblank_disable_and_save(dev, pipe);
+	DRM_DEBUG_VBL("crtc %d, vblank enabled %d, inmodeset %d\n",
+		      pipe, vblank->enabled, vblank->inmodeset);
+
+	/* Avoid redundant vblank disables without previous drm_vblank_on(). */
+	if (drm_core_check_feature(dev, DRIVER_ATOMIC) || !vblank->inmodeset)
+		vblank_disable_and_save(dev, pipe);
+
 	wake_up(&vblank->queue);
 
 	/*
@@ -1415,6 +1479,9 @@ void drm_vblank_on(struct drm_device *dev, unsigned int pipe)
 		return;
 
 	spin_lock_irqsave(&dev->vbl_lock, irqflags);
+	DRM_DEBUG_VBL("crtc %d, vblank enabled %d, inmodeset %d\n",
+		      pipe, vblank->enabled, vblank->inmodeset);
+
 	/* Drop our private "prevent drm_vblank_get" refcount */
 	if (vblank->inmodeset) {
 		atomic_dec(&vblank->refcount);
@@ -1427,8 +1494,7 @@ void drm_vblank_on(struct drm_device *dev, unsigned int pipe)
 	 * re-enable interrupts if there are users left, or the
 	 * user wishes vblank interrupts to be enabled all the time.
 	 */
-	if (atomic_read(&vblank->refcount) != 0 ||
-	    (!dev->vblank_disable_immediate && drm_vblank_offdelay == 0))
+	if (atomic_read(&vblank->refcount) != 0 || drm_vblank_offdelay == 0)
 		WARN_ON(drm_vblank_enable(dev, pipe));
 	spin_unlock_irqrestore(&dev->vbl_lock, irqflags);
 }
@@ -1523,6 +1589,7 @@ void drm_vblank_post_modeset(struct drm_device *dev, unsigned int pipe)
 	if (vblank->inmodeset) {
 		spin_lock_irqsave(&dev->vbl_lock, irqflags);
 		dev->vblank_disable_allowed = true;
+		drm_reset_vblank_timestamp(dev, pipe);
 		spin_unlock_irqrestore(&dev->vbl_lock, irqflags);
 
 		if (vblank->inmodeset & 0x2)
diff --git a/drivers/gpu/drm/drm_mipi_dsi.c b/drivers/gpu/drm/drm_mipi_dsi.c
index 6e6a9c58d404..f5d80839a90c 100644
--- a/drivers/gpu/drm/drm_mipi_dsi.c
+++ b/drivers/gpu/drm/drm_mipi_dsi.c
@@ -47,7 +47,17 @@
 
 static int mipi_dsi_device_match(struct device *dev, struct device_driver *drv)
 {
-	return of_driver_match_device(dev, drv);
+	struct mipi_dsi_device *dsi = to_mipi_dsi_device(dev);
+
+	/* attempt OF style match */
+	if (of_driver_match_device(dev, drv))
+		return 1;
+
+	/* compare DSI device and driver names */
+	if (!strcmp(dsi->name, drv->name))
+		return 1;
+
+	return 0;
 }
 
 static const struct dev_pm_ops mipi_dsi_device_pm_ops = {
@@ -129,14 +139,20 @@ static int mipi_dsi_device_add(struct mipi_dsi_device *dsi)
 	return device_add(&dsi->dev);
 }
 
+#if IS_ENABLED(CONFIG_OF)
 static struct mipi_dsi_device *
 of_mipi_dsi_device_add(struct mipi_dsi_host *host, struct device_node *node)
 {
-	struct mipi_dsi_device *dsi;
 	struct device *dev = host->dev;
+	struct mipi_dsi_device_info info = { };
 	int ret;
 	u32 reg;
 
+	if (of_modalias_node(node, info.type, sizeof(info.type)) < 0) {
+		dev_err(dev, "modalias failure on %s\n", node->full_name);
+		return ERR_PTR(-EINVAL);
+	}
+
 	ret = of_property_read_u32(node, "reg", &reg);
 	if (ret) {
 		dev_err(dev, "device node %s has no valid reg property: %d\n",
@@ -144,32 +160,111 @@ of_mipi_dsi_device_add(struct mipi_dsi_host *host, struct device_node *node)
 		return ERR_PTR(-EINVAL);
 	}
 
-	if (reg > 3) {
-		dev_err(dev, "device node %s has invalid reg property: %u\n",
-			node->full_name, reg);
+	info.channel = reg;
+	info.node = of_node_get(node);
+
+	return mipi_dsi_device_register_full(host, &info);
+}
+#else
+static struct mipi_dsi_device *
+of_mipi_dsi_device_add(struct mipi_dsi_host *host, struct device_node *node)
+{
+	return ERR_PTR(-ENODEV);
+}
+#endif
+
+/**
+ * mipi_dsi_device_register_full - create a MIPI DSI device
+ * @host: DSI host to which this device is connected
+ * @info: pointer to template containing DSI device information
+ *
+ * Create a MIPI DSI device by using the device information provided by
+ * mipi_dsi_device_info template
+ *
+ * Returns:
+ * A pointer to the newly created MIPI DSI device, or, a pointer encoded
+ * with an error
+ */
+struct mipi_dsi_device *
+mipi_dsi_device_register_full(struct mipi_dsi_host *host,
+			      const struct mipi_dsi_device_info *info)
+{
+	struct mipi_dsi_device *dsi;
+	struct device *dev = host->dev;
+	int ret;
+
+	if (!info) {
+		dev_err(dev, "invalid mipi_dsi_device_info pointer\n");
+		return ERR_PTR(-EINVAL);
+	}
+
+	if (info->channel > 3) {
+		dev_err(dev, "invalid virtual channel: %u\n", info->channel);
 		return ERR_PTR(-EINVAL);
 	}
 
 	dsi = mipi_dsi_device_alloc(host);
 	if (IS_ERR(dsi)) {
-		dev_err(dev, "failed to allocate DSI device %s: %ld\n",
-			node->full_name, PTR_ERR(dsi));
+		dev_err(dev, "failed to allocate DSI device %ld\n",
+			PTR_ERR(dsi));
 		return dsi;
 	}
 
-	dsi->dev.of_node = of_node_get(node);
-	dsi->channel = reg;
+	dsi->dev.of_node = info->node;
+	dsi->channel = info->channel;
+	strlcpy(dsi->name, info->type, sizeof(dsi->name));
 
 	ret = mipi_dsi_device_add(dsi);
 	if (ret) {
-		dev_err(dev, "failed to add DSI device %s: %d\n",
-			node->full_name, ret);
+		dev_err(dev, "failed to add DSI device %d\n", ret);
 		kfree(dsi);
 		return ERR_PTR(ret);
 	}
 
 	return dsi;
 }
+EXPORT_SYMBOL(mipi_dsi_device_register_full);
+
+/**
+ * mipi_dsi_device_unregister - unregister MIPI DSI device
+ * @dsi: DSI peripheral device
+ */
+void mipi_dsi_device_unregister(struct mipi_dsi_device *dsi)
+{
+	device_unregister(&dsi->dev);
+}
+EXPORT_SYMBOL(mipi_dsi_device_unregister);
+
+static DEFINE_MUTEX(host_lock);
+static LIST_HEAD(host_list);
+
+/**
+ * of_find_mipi_dsi_host_by_node() - find the MIPI DSI host matching a
+ *				     device tree node
+ * @node: device tree node
+ *
+ * Returns:
+ * A pointer to the MIPI DSI host corresponding to @node or NULL if no
+ * such device exists (or has not been registered yet).
+ */
+struct mipi_dsi_host *of_find_mipi_dsi_host_by_node(struct device_node *node)
+{
+	struct mipi_dsi_host *host;
+
+	mutex_lock(&host_lock);
+
+	list_for_each_entry(host, &host_list, list) {
+		if (host->dev->of_node == node) {
+			mutex_unlock(&host_lock);
+			return host;
+		}
+	}
+
+	mutex_unlock(&host_lock);
+
+	return NULL;
+}
+EXPORT_SYMBOL(of_find_mipi_dsi_host_by_node);
 
 int mipi_dsi_host_register(struct mipi_dsi_host *host)
 {
@@ -182,6 +277,10 @@ int mipi_dsi_host_register(struct mipi_dsi_host *host)
 		of_mipi_dsi_device_add(host, node);
 	}
 
+	mutex_lock(&host_lock);
+	list_add_tail(&host->list, &host_list);
+	mutex_unlock(&host_lock);
+
 	return 0;
 }
 EXPORT_SYMBOL(mipi_dsi_host_register);
@@ -190,7 +289,7 @@ static int mipi_dsi_remove_device_fn(struct device *dev, void *priv)
 {
 	struct mipi_dsi_device *dsi = to_mipi_dsi_device(dev);
 
-	device_unregister(&dsi->dev);
+	mipi_dsi_device_unregister(dsi);
 
 	return 0;
 }
@@ -198,6 +297,10 @@ static int mipi_dsi_remove_device_fn(struct device *dev, void *priv)
 void mipi_dsi_host_unregister(struct mipi_dsi_host *host)
 {
 	device_for_each_child(host->dev, NULL, mipi_dsi_remove_device_fn);
+
+	mutex_lock(&host_lock);
+	list_del_init(&host->list);
+	mutex_unlock(&host_lock);
 }
 EXPORT_SYMBOL(mipi_dsi_host_unregister);
 
diff --git a/drivers/gpu/drm/drm_of.c b/drivers/gpu/drm/drm_of.c
index 493c05c9ce4f..bc98bb94264d 100644
--- a/drivers/gpu/drm/drm_of.c
+++ b/drivers/gpu/drm/drm_of.c
@@ -149,3 +149,37 @@ int drm_of_component_probe(struct device *dev,
 	return component_master_add_with_match(dev, m_ops, match);
 }
 EXPORT_SYMBOL(drm_of_component_probe);
+
+/*
+ * drm_of_encoder_active_endpoint - return the active encoder endpoint
+ * @node: device tree node containing encoder input ports
+ * @encoder: drm_encoder
+ *
+ * Given an encoder device node and a drm_encoder with a connected crtc,
+ * parse the encoder endpoint connecting to the crtc port.
+ */
+int drm_of_encoder_active_endpoint(struct device_node *node,
+				   struct drm_encoder *encoder,
+				   struct of_endpoint *endpoint)
+{
+	struct device_node *ep;
+	struct drm_crtc *crtc = encoder->crtc;
+	struct device_node *port;
+	int ret;
+
+	if (!node || !crtc)
+		return -EINVAL;
+
+	for_each_endpoint_of_node(node, ep) {
+		port = of_graph_get_remote_port(ep);
+		of_node_put(port);
+		if (port == crtc->port) {
+			ret = of_graph_parse_endpoint(ep, endpoint);
+			of_node_put(ep);
+			return ret;
+		}
+	}
+
+	return -EINVAL;
+}
+EXPORT_SYMBOL_GPL(drm_of_encoder_active_endpoint);
diff --git a/drivers/gpu/drm/etnaviv/etnaviv_buffer.c b/drivers/gpu/drm/etnaviv/etnaviv_buffer.c
index 332c55ebba6d..d8d556457427 100644
--- a/drivers/gpu/drm/etnaviv/etnaviv_buffer.c
+++ b/drivers/gpu/drm/etnaviv/etnaviv_buffer.c
@@ -21,6 +21,7 @@
 
 #include "common.xml.h"
 #include "state.xml.h"
+#include "state_3d.xml.h"
 #include "cmdstream.xml.h"
 
 /*
@@ -85,10 +86,17 @@ static inline void CMD_STALL(struct etnaviv_cmdbuf *buffer,
 	OUT(buffer, VIV_FE_STALL_TOKEN_FROM(from) | VIV_FE_STALL_TOKEN_TO(to));
 }
 
-static void etnaviv_cmd_select_pipe(struct etnaviv_cmdbuf *buffer, u8 pipe)
+static inline void CMD_SEM(struct etnaviv_cmdbuf *buffer, u32 from, u32 to)
 {
-	u32 flush;
-	u32 stall;
+	CMD_LOAD_STATE(buffer, VIVS_GL_SEMAPHORE_TOKEN,
+		       VIVS_GL_SEMAPHORE_TOKEN_FROM(from) |
+		       VIVS_GL_SEMAPHORE_TOKEN_TO(to));
+}
+
+static void etnaviv_cmd_select_pipe(struct etnaviv_gpu *gpu,
+	struct etnaviv_cmdbuf *buffer, u8 pipe)
+{
+	u32 flush = 0;
 
 	/*
 	 * This assumes that if we're switching to 2D, we're switching
@@ -96,17 +104,13 @@ static void etnaviv_cmd_select_pipe(struct etnaviv_cmdbuf *buffer, u8 pipe)
 	 * the 2D core, we need to flush the 3D depth and color caches,
 	 * otherwise we need to flush the 2D pixel engine cache.
 	 */
-	if (pipe == ETNA_PIPE_2D)
-		flush = VIVS_GL_FLUSH_CACHE_DEPTH | VIVS_GL_FLUSH_CACHE_COLOR;
-	else
+	if (gpu->exec_state == ETNA_PIPE_2D)
 		flush = VIVS_GL_FLUSH_CACHE_PE2D;
-
-	stall = VIVS_GL_SEMAPHORE_TOKEN_FROM(SYNC_RECIPIENT_FE) |
-		VIVS_GL_SEMAPHORE_TOKEN_TO(SYNC_RECIPIENT_PE);
+	else if (gpu->exec_state == ETNA_PIPE_3D)
+		flush = VIVS_GL_FLUSH_CACHE_DEPTH | VIVS_GL_FLUSH_CACHE_COLOR;
 
 	CMD_LOAD_STATE(buffer, VIVS_GL_FLUSH_CACHE, flush);
-	CMD_LOAD_STATE(buffer, VIVS_GL_SEMAPHORE_TOKEN, stall);
-
+	CMD_SEM(buffer, SYNC_RECIPIENT_FE, SYNC_RECIPIENT_PE);
 	CMD_STALL(buffer, SYNC_RECIPIENT_FE, SYNC_RECIPIENT_PE);
 
 	CMD_LOAD_STATE(buffer, VIVS_GL_PIPE_SELECT,
@@ -131,6 +135,36 @@ static void etnaviv_buffer_dump(struct etnaviv_gpu *gpu,
 			ptr, len * 4, 0);
 }
 
+/*
+ * Safely replace the WAIT of a waitlink with a new command and argument.
+ * The GPU may be executing this WAIT while we're modifying it, so we have
+ * to write it in a specific order to avoid the GPU branching to somewhere
+ * else.  'wl_offset' is the offset to the first byte of the WAIT command.
+ */
+static void etnaviv_buffer_replace_wait(struct etnaviv_cmdbuf *buffer,
+	unsigned int wl_offset, u32 cmd, u32 arg)
+{
+	u32 *lw = buffer->vaddr + wl_offset;
+
+	lw[1] = arg;
+	mb();
+	lw[0] = cmd;
+	mb();
+}
+
+/*
+ * Ensure that there is space in the command buffer to contiguously write
+ * 'cmd_dwords' 64-bit words into the buffer, wrapping if necessary.
+ */
+static u32 etnaviv_buffer_reserve(struct etnaviv_gpu *gpu,
+	struct etnaviv_cmdbuf *buffer, unsigned int cmd_dwords)
+{
+	if (buffer->user_size + cmd_dwords * sizeof(u64) > buffer->size)
+		buffer->user_size = 0;
+
+	return gpu_va(gpu, buffer) + buffer->user_size;
+}
+
 u16 etnaviv_buffer_init(struct etnaviv_gpu *gpu)
 {
 	struct etnaviv_cmdbuf *buffer = gpu->buffer;
@@ -147,81 +181,79 @@ u16 etnaviv_buffer_init(struct etnaviv_gpu *gpu)
 void etnaviv_buffer_end(struct etnaviv_gpu *gpu)
 {
 	struct etnaviv_cmdbuf *buffer = gpu->buffer;
+	unsigned int waitlink_offset = buffer->user_size - 16;
+	u32 link_target, flush = 0;
 
-	/* Replace the last WAIT with an END */
-	buffer->user_size -= 16;
-
-	CMD_END(buffer);
-	mb();
+	if (gpu->exec_state == ETNA_PIPE_2D)
+		flush = VIVS_GL_FLUSH_CACHE_PE2D;
+	else if (gpu->exec_state == ETNA_PIPE_3D)
+		flush = VIVS_GL_FLUSH_CACHE_DEPTH |
+			VIVS_GL_FLUSH_CACHE_COLOR |
+			VIVS_GL_FLUSH_CACHE_TEXTURE |
+			VIVS_GL_FLUSH_CACHE_TEXTUREVS |
+			VIVS_GL_FLUSH_CACHE_SHADER_L2;
+
+	if (flush) {
+		unsigned int dwords = 7;
+
+		link_target = etnaviv_buffer_reserve(gpu, buffer, dwords);
+
+		CMD_SEM(buffer, SYNC_RECIPIENT_FE, SYNC_RECIPIENT_PE);
+		CMD_STALL(buffer, SYNC_RECIPIENT_FE, SYNC_RECIPIENT_PE);
+		CMD_LOAD_STATE(buffer, VIVS_GL_FLUSH_CACHE, flush);
+		if (gpu->exec_state == ETNA_PIPE_3D)
+			CMD_LOAD_STATE(buffer, VIVS_TS_FLUSH_CACHE,
+				       VIVS_TS_FLUSH_CACHE_FLUSH);
+		CMD_SEM(buffer, SYNC_RECIPIENT_FE, SYNC_RECIPIENT_PE);
+		CMD_STALL(buffer, SYNC_RECIPIENT_FE, SYNC_RECIPIENT_PE);
+		CMD_END(buffer);
+
+		etnaviv_buffer_replace_wait(buffer, waitlink_offset,
+					    VIV_FE_LINK_HEADER_OP_LINK |
+					    VIV_FE_LINK_HEADER_PREFETCH(dwords),
+					    link_target);
+	} else {
+		/* Replace the last link-wait with an "END" command */
+		etnaviv_buffer_replace_wait(buffer, waitlink_offset,
+					    VIV_FE_END_HEADER_OP_END, 0);
+	}
 }
 
+/* Append a command buffer to the ring buffer. */
 void etnaviv_buffer_queue(struct etnaviv_gpu *gpu, unsigned int event,
 	struct etnaviv_cmdbuf *cmdbuf)
 {
 	struct etnaviv_cmdbuf *buffer = gpu->buffer;
-	u32 *lw = buffer->vaddr + buffer->user_size - 16;
-	u32 back, link_target, link_size, reserve_size, extra_size = 0;
+	unsigned int waitlink_offset = buffer->user_size - 16;
+	u32 return_target, return_dwords;
+	u32 link_target, link_dwords;
 
 	if (drm_debug & DRM_UT_DRIVER)
 		etnaviv_buffer_dump(gpu, buffer, 0, 0x50);
 
+	link_target = gpu_va(gpu, cmdbuf);
+	link_dwords = cmdbuf->size / 8;
+
 	/*
-	 * If we need to flush the MMU prior to submitting this buffer, we
-	 * will need to append a mmu flush load state, followed by a new
+	 * If we need maintanence prior to submitting this buffer, we will
+	 * need to append a mmu flush load state, followed by a new
 	 * link to this buffer - a total of four additional words.
 	 */
 	if (gpu->mmu->need_flush || gpu->switch_context) {
+		u32 target, extra_dwords;
+
 		/* link command */
-		extra_size += 2;
+		extra_dwords = 1;
+
 		/* flush command */
 		if (gpu->mmu->need_flush)
-			extra_size += 2;
+			extra_dwords += 1;
+
 		/* pipe switch commands */
 		if (gpu->switch_context)
-			extra_size += 8;
-	}
+			extra_dwords += 4;
 
-	reserve_size = (6 + extra_size) * 4;
-
-	/*
-	 * if we are going to completely overflow the buffer, we need to wrap.
-	 */
-	if (buffer->user_size + reserve_size > buffer->size)
-		buffer->user_size = 0;
-
-	/* save offset back into main buffer */
-	back = buffer->user_size + reserve_size - 6 * 4;
-	link_target = gpu_va(gpu, buffer) + buffer->user_size;
-	link_size = 6;
-
-	/* Skip over any extra instructions */
-	link_target += extra_size * sizeof(u32);
-
-	if (drm_debug & DRM_UT_DRIVER)
-		pr_info("stream link to 0x%08x @ 0x%08x %p\n",
-			link_target, gpu_va(gpu, cmdbuf), cmdbuf->vaddr);
-
-	/* jump back from cmd to main buffer */
-	CMD_LINK(cmdbuf, link_size, link_target);
-
-	link_target = gpu_va(gpu, cmdbuf);
-	link_size = cmdbuf->size / 8;
-
-
-
-	if (drm_debug & DRM_UT_DRIVER) {
-		print_hex_dump(KERN_INFO, "cmd ", DUMP_PREFIX_OFFSET, 16, 4,
-			       cmdbuf->vaddr, cmdbuf->size, 0);
-
-		pr_info("link op: %p\n", lw);
-		pr_info("link addr: %p\n", lw + 1);
-		pr_info("addr: 0x%08x\n", link_target);
-		pr_info("back: 0x%08x\n", gpu_va(gpu, buffer) + back);
-		pr_info("event: %d\n", event);
-	}
-
-	if (gpu->mmu->need_flush || gpu->switch_context) {
-		u32 new_target = gpu_va(gpu, buffer) + buffer->user_size;
+		target = etnaviv_buffer_reserve(gpu, buffer, extra_dwords);
 
 		if (gpu->mmu->need_flush) {
 			/* Add the MMU flush */
@@ -236,32 +268,59 @@ void etnaviv_buffer_queue(struct etnaviv_gpu *gpu, unsigned int event,
 		}
 
 		if (gpu->switch_context) {
-			etnaviv_cmd_select_pipe(buffer, cmdbuf->exec_state);
+			etnaviv_cmd_select_pipe(gpu, buffer, cmdbuf->exec_state);
+			gpu->exec_state = cmdbuf->exec_state;
 			gpu->switch_context = false;
 		}
 
-		/* And the link to the first buffer */
-		CMD_LINK(buffer, link_size, link_target);
+		/* And the link to the submitted buffer */
+		CMD_LINK(buffer, link_dwords, link_target);
 
 		/* Update the link target to point to above instructions */
-		link_target = new_target;
-		link_size = extra_size;
+		link_target = target;
+		link_dwords = extra_dwords;
 	}
 
-	/* trigger event */
+	/*
+	 * Append a LINK to the submitted command buffer to return to
+	 * the ring buffer.  return_target is the ring target address.
+	 * We need three dwords: event, wait, link.
+	 */
+	return_dwords = 3;
+	return_target = etnaviv_buffer_reserve(gpu, buffer, return_dwords);
+	CMD_LINK(cmdbuf, return_dwords, return_target);
+
+	/*
+	 * Append event, wait and link pointing back to the wait
+	 * command to the ring buffer.
+	 */
 	CMD_LOAD_STATE(buffer, VIVS_GL_EVENT, VIVS_GL_EVENT_EVENT_ID(event) |
 		       VIVS_GL_EVENT_FROM_PE);
-
-	/* append WAIT/LINK to main buffer */
 	CMD_WAIT(buffer);
-	CMD_LINK(buffer, 2, gpu_va(gpu, buffer) + (buffer->user_size - 4));
+	CMD_LINK(buffer, 2, return_target + 8);
 
-	/* Change WAIT into a LINK command; write the address first. */
-	*(lw + 1) = link_target;
-	mb();
-	*(lw) = VIV_FE_LINK_HEADER_OP_LINK |
-		VIV_FE_LINK_HEADER_PREFETCH(link_size);
-	mb();
+	if (drm_debug & DRM_UT_DRIVER)
+		pr_info("stream link to 0x%08x @ 0x%08x %p\n",
+			return_target, gpu_va(gpu, cmdbuf), cmdbuf->vaddr);
+
+	if (drm_debug & DRM_UT_DRIVER) {
+		print_hex_dump(KERN_INFO, "cmd ", DUMP_PREFIX_OFFSET, 16, 4,
+			       cmdbuf->vaddr, cmdbuf->size, 0);
+
+		pr_info("link op: %p\n", buffer->vaddr + waitlink_offset);
+		pr_info("addr: 0x%08x\n", link_target);
+		pr_info("back: 0x%08x\n", return_target);
+		pr_info("event: %d\n", event);
+	}
+
+	/*
+	 * Kick off the submitted command by replacing the previous
+	 * WAIT with a link to the address in the ring buffer.
+	 */
+	etnaviv_buffer_replace_wait(buffer, waitlink_offset,
+				    VIV_FE_LINK_HEADER_OP_LINK |
+				    VIV_FE_LINK_HEADER_PREFETCH(link_dwords),
+				    link_target);
 
 	if (drm_debug & DRM_UT_DRIVER)
 		etnaviv_buffer_dump(gpu, buffer, 0, 0x50);
diff --git a/drivers/gpu/drm/etnaviv/etnaviv_drv.h b/drivers/gpu/drm/etnaviv/etnaviv_drv.h
index 1cd6046e76b1..115c5bc6d7c8 100644
--- a/drivers/gpu/drm/etnaviv/etnaviv_drv.h
+++ b/drivers/gpu/drm/etnaviv/etnaviv_drv.h
@@ -75,9 +75,6 @@ int etnaviv_ioctl_gem_submit(struct drm_device *dev, void *data,
 int etnaviv_gem_mmap(struct file *filp, struct vm_area_struct *vma);
 int etnaviv_gem_fault(struct vm_area_struct *vma, struct vm_fault *vmf);
 int etnaviv_gem_mmap_offset(struct drm_gem_object *obj, u64 *offset);
-int etnaviv_gem_get_iova(struct etnaviv_gpu *gpu,
-	struct drm_gem_object *obj, u32 *iova);
-void etnaviv_gem_put_iova(struct etnaviv_gpu *gpu, struct drm_gem_object *obj);
 struct sg_table *etnaviv_gem_prime_get_sg_table(struct drm_gem_object *obj);
 void *etnaviv_gem_prime_vmap(struct drm_gem_object *obj);
 void etnaviv_gem_prime_vunmap(struct drm_gem_object *obj, void *vaddr);
diff --git a/drivers/gpu/drm/etnaviv/etnaviv_gem.c b/drivers/gpu/drm/etnaviv/etnaviv_gem.c
index 4b519e4309b2..937a77520f58 100644
--- a/drivers/gpu/drm/etnaviv/etnaviv_gem.c
+++ b/drivers/gpu/drm/etnaviv/etnaviv_gem.c
@@ -260,8 +260,32 @@ etnaviv_gem_get_vram_mapping(struct etnaviv_gem_object *obj,
 	return NULL;
 }
 
-int etnaviv_gem_get_iova(struct etnaviv_gpu *gpu,
-	struct drm_gem_object *obj, u32 *iova)
+void etnaviv_gem_mapping_reference(struct etnaviv_vram_mapping *mapping)
+{
+	struct etnaviv_gem_object *etnaviv_obj = mapping->object;
+
+	drm_gem_object_reference(&etnaviv_obj->base);
+
+	mutex_lock(&etnaviv_obj->lock);
+	WARN_ON(mapping->use == 0);
+	mapping->use += 1;
+	mutex_unlock(&etnaviv_obj->lock);
+}
+
+void etnaviv_gem_mapping_unreference(struct etnaviv_vram_mapping *mapping)
+{
+	struct etnaviv_gem_object *etnaviv_obj = mapping->object;
+
+	mutex_lock(&etnaviv_obj->lock);
+	WARN_ON(mapping->use == 0);
+	mapping->use -= 1;
+	mutex_unlock(&etnaviv_obj->lock);
+
+	drm_gem_object_unreference_unlocked(&etnaviv_obj->base);
+}
+
+struct etnaviv_vram_mapping *etnaviv_gem_mapping_get(
+	struct drm_gem_object *obj, struct etnaviv_gpu *gpu)
 {
 	struct etnaviv_gem_object *etnaviv_obj = to_etnaviv_bo(obj);
 	struct etnaviv_vram_mapping *mapping;
@@ -329,28 +353,12 @@ int etnaviv_gem_get_iova(struct etnaviv_gpu *gpu,
 out:
 	mutex_unlock(&etnaviv_obj->lock);
 
-	if (!ret) {
-		/* Take a reference on the object */
-		drm_gem_object_reference(obj);
-		*iova = mapping->iova;
-	}
-
-	return ret;
-}
-
-void etnaviv_gem_put_iova(struct etnaviv_gpu *gpu, struct drm_gem_object *obj)
-{
-	struct etnaviv_gem_object *etnaviv_obj = to_etnaviv_bo(obj);
-	struct etnaviv_vram_mapping *mapping;
-
-	mutex_lock(&etnaviv_obj->lock);
-	mapping = etnaviv_gem_get_vram_mapping(etnaviv_obj, gpu->mmu);
-
-	WARN_ON(mapping->use == 0);
-	mapping->use -= 1;
-	mutex_unlock(&etnaviv_obj->lock);
+	if (ret)
+		return ERR_PTR(ret);
 
-	drm_gem_object_unreference_unlocked(obj);
+	/* Take a reference on the object */
+	drm_gem_object_reference(obj);
+	return mapping;
 }
 
 void *etnaviv_gem_vmap(struct drm_gem_object *obj)
diff --git a/drivers/gpu/drm/etnaviv/etnaviv_gem.h b/drivers/gpu/drm/etnaviv/etnaviv_gem.h
index ab5df8147a5f..02665d8c10ee 100644
--- a/drivers/gpu/drm/etnaviv/etnaviv_gem.h
+++ b/drivers/gpu/drm/etnaviv/etnaviv_gem.h
@@ -88,6 +88,12 @@ static inline bool is_active(struct etnaviv_gem_object *etnaviv_obj)
 
 #define MAX_CMDS 4
 
+struct etnaviv_gem_submit_bo {
+	u32 flags;
+	struct etnaviv_gem_object *obj;
+	struct etnaviv_vram_mapping *mapping;
+};
+
 /* Created per submit-ioctl, to track bo's and cmdstream bufs, etc,
  * associated with the cmdstream submission for synchronization (and
  * make it easier to unwind when things go wrong, etc).  This only
@@ -99,11 +105,7 @@ struct etnaviv_gem_submit {
 	struct ww_acquire_ctx ticket;
 	u32 fence;
 	unsigned int nr_bos;
-	struct {
-		u32 flags;
-		struct etnaviv_gem_object *obj;
-		u32 iova;
-	} bos[0];
+	struct etnaviv_gem_submit_bo bos[0];
 };
 
 int etnaviv_gem_wait_bo(struct etnaviv_gpu *gpu, struct drm_gem_object *obj,
@@ -115,4 +117,9 @@ int etnaviv_gem_obj_add(struct drm_device *dev, struct drm_gem_object *obj);
 struct page **etnaviv_gem_get_pages(struct etnaviv_gem_object *obj);
 void etnaviv_gem_put_pages(struct etnaviv_gem_object *obj);
 
+struct etnaviv_vram_mapping *etnaviv_gem_mapping_get(
+	struct drm_gem_object *obj, struct etnaviv_gpu *gpu);
+void etnaviv_gem_mapping_reference(struct etnaviv_vram_mapping *mapping);
+void etnaviv_gem_mapping_unreference(struct etnaviv_vram_mapping *mapping);
+
 #endif /* __ETNAVIV_GEM_H__ */
diff --git a/drivers/gpu/drm/etnaviv/etnaviv_gem_submit.c b/drivers/gpu/drm/etnaviv/etnaviv_gem_submit.c
index 1aba01a999df..236ada93df53 100644
--- a/drivers/gpu/drm/etnaviv/etnaviv_gem_submit.c
+++ b/drivers/gpu/drm/etnaviv/etnaviv_gem_submit.c
@@ -187,12 +187,10 @@ static void submit_unpin_objects(struct etnaviv_gem_submit *submit)
 	int i;
 
 	for (i = 0; i < submit->nr_bos; i++) {
-		struct etnaviv_gem_object *etnaviv_obj = submit->bos[i].obj;
-
 		if (submit->bos[i].flags & BO_PINNED)
-			etnaviv_gem_put_iova(submit->gpu, &etnaviv_obj->base);
+			etnaviv_gem_mapping_unreference(submit->bos[i].mapping);
 
-		submit->bos[i].iova = 0;
+		submit->bos[i].mapping = NULL;
 		submit->bos[i].flags &= ~BO_PINNED;
 	}
 }
@@ -203,22 +201,24 @@ static int submit_pin_objects(struct etnaviv_gem_submit *submit)
 
 	for (i = 0; i < submit->nr_bos; i++) {
 		struct etnaviv_gem_object *etnaviv_obj = submit->bos[i].obj;
-		u32 iova;
+		struct etnaviv_vram_mapping *mapping;
 
-		ret = etnaviv_gem_get_iova(submit->gpu, &etnaviv_obj->base,
-					   &iova);
-		if (ret)
+		mapping = etnaviv_gem_mapping_get(&etnaviv_obj->base,
+						  submit->gpu);
+		if (IS_ERR(mapping)) {
+			ret = PTR_ERR(mapping);
 			break;
+		}
 
 		submit->bos[i].flags |= BO_PINNED;
-		submit->bos[i].iova = iova;
+		submit->bos[i].mapping = mapping;
 	}
 
 	return ret;
 }
 
 static int submit_bo(struct etnaviv_gem_submit *submit, u32 idx,
-		struct etnaviv_gem_object **obj, u32 *iova)
+	struct etnaviv_gem_submit_bo **bo)
 {
 	if (idx >= submit->nr_bos) {
 		DRM_ERROR("invalid buffer index: %u (out of %u)\n",
@@ -226,10 +226,7 @@ static int submit_bo(struct etnaviv_gem_submit *submit, u32 idx,
 		return -EINVAL;
 	}
 
-	if (obj)
-		*obj = submit->bos[idx].obj;
-	if (iova)
-		*iova = submit->bos[idx].iova;
+	*bo = &submit->bos[idx];
 
 	return 0;
 }
@@ -245,8 +242,8 @@ static int submit_reloc(struct etnaviv_gem_submit *submit, void *stream,
 
 	for (i = 0; i < nr_relocs; i++) {
 		const struct drm_etnaviv_gem_submit_reloc *r = relocs + i;
-		struct etnaviv_gem_object *bobj;
-		u32 iova, off;
+		struct etnaviv_gem_submit_bo *bo;
+		u32 off;
 
 		if (unlikely(r->flags)) {
 			DRM_ERROR("invalid reloc flags\n");
@@ -268,17 +265,16 @@ static int submit_reloc(struct etnaviv_gem_submit *submit, void *stream,
 			return -EINVAL;
 		}
 
-		ret = submit_bo(submit, r->reloc_idx, &bobj, &iova);
+		ret = submit_bo(submit, r->reloc_idx, &bo);
 		if (ret)
 			return ret;
 
-		if (r->reloc_offset >=
-		    bobj->base.size - sizeof(*ptr)) {
+		if (r->reloc_offset >= bo->obj->base.size - sizeof(*ptr)) {
 			DRM_ERROR("relocation %u outside object", i);
 			return -EINVAL;
 		}
 
-		ptr[off] = iova + r->reloc_offset;
+		ptr[off] = bo->mapping->iova + r->reloc_offset;
 
 		last_offset = off;
 	}
diff --git a/drivers/gpu/drm/etnaviv/etnaviv_gpu.c b/drivers/gpu/drm/etnaviv/etnaviv_gpu.c
index a33162cf4f4c..d13303ce530d 100644
--- a/drivers/gpu/drm/etnaviv/etnaviv_gpu.c
+++ b/drivers/gpu/drm/etnaviv/etnaviv_gpu.c
@@ -628,6 +628,7 @@ int etnaviv_gpu_init(struct etnaviv_gpu *gpu)
 	/* Now program the hardware */
 	mutex_lock(&gpu->lock);
 	etnaviv_gpu_hw_init(gpu);
+	gpu->exec_state = -1;
 	mutex_unlock(&gpu->lock);
 
 	pm_runtime_mark_last_busy(gpu->dev);
@@ -871,17 +872,13 @@ static void recover_worker(struct work_struct *work)
 		gpu->event[i].fence = NULL;
 		gpu->event[i].used = false;
 		complete(&gpu->event_free);
-		/*
-		 * Decrement the PM count for each stuck event. This is safe
-		 * even in atomic context as we use ASYNC RPM here.
-		 */
-		pm_runtime_put_autosuspend(gpu->dev);
 	}
 	spin_unlock_irqrestore(&gpu->event_spinlock, flags);
 	gpu->completed_fence = gpu->active_fence;
 
 	etnaviv_gpu_hw_init(gpu);
 	gpu->switch_context = true;
+	gpu->exec_state = -1;
 
 	mutex_unlock(&gpu->lock);
 	pm_runtime_mark_last_busy(gpu->dev);
@@ -1106,7 +1103,7 @@ struct etnaviv_cmdbuf *etnaviv_gpu_cmdbuf_new(struct etnaviv_gpu *gpu, u32 size,
 	size_t nr_bos)
 {
 	struct etnaviv_cmdbuf *cmdbuf;
-	size_t sz = size_vstruct(nr_bos, sizeof(cmdbuf->bo[0]),
+	size_t sz = size_vstruct(nr_bos, sizeof(cmdbuf->bo_map[0]),
 				 sizeof(*cmdbuf));
 
 	cmdbuf = kzalloc(sz, GFP_KERNEL);
@@ -1150,14 +1147,23 @@ static void retire_worker(struct work_struct *work)
 		fence_put(cmdbuf->fence);
 
 		for (i = 0; i < cmdbuf->nr_bos; i++) {
-			struct etnaviv_gem_object *etnaviv_obj = cmdbuf->bo[i];
+			struct etnaviv_vram_mapping *mapping = cmdbuf->bo_map[i];
+			struct etnaviv_gem_object *etnaviv_obj = mapping->object;
 
 			atomic_dec(&etnaviv_obj->gpu_active);
 			/* drop the refcount taken in etnaviv_gpu_submit */
-			etnaviv_gem_put_iova(gpu, &etnaviv_obj->base);
+			etnaviv_gem_mapping_unreference(mapping);
 		}
 
 		etnaviv_gpu_cmdbuf_free(cmdbuf);
+		/*
+		 * We need to balance the runtime PM count caused by
+		 * each submission.  Upon submission, we increment
+		 * the runtime PM counter, and allocate one event.
+		 * So here, we put the runtime PM count for each
+		 * completed event.
+		 */
+		pm_runtime_put_autosuspend(gpu->dev);
 	}
 
 	gpu->retired_fence = fence;
@@ -1304,11 +1310,10 @@ int etnaviv_gpu_submit(struct etnaviv_gpu *gpu,
 
 	for (i = 0; i < submit->nr_bos; i++) {
 		struct etnaviv_gem_object *etnaviv_obj = submit->bos[i].obj;
-		u32 iova;
 
-		/* Each cmdbuf takes a refcount on the iova */
-		etnaviv_gem_get_iova(gpu, &etnaviv_obj->base, &iova);
-		cmdbuf->bo[i] = etnaviv_obj;
+		/* Each cmdbuf takes a refcount on the mapping */
+		etnaviv_gem_mapping_reference(submit->bos[i].mapping);
+		cmdbuf->bo_map[i] = submit->bos[i].mapping;
 		atomic_inc(&etnaviv_obj->gpu_active);
 
 		if (submit->bos[i].flags & ETNA_SUBMIT_BO_WRITE)
@@ -1378,15 +1383,6 @@ static irqreturn_t irq_handler(int irq, void *data)
 				gpu->completed_fence = fence->seqno;
 
 			event_free(gpu, event);
-
-			/*
-			 * We need to balance the runtime PM count caused by
-			 * each submission.  Upon submission, we increment
-			 * the runtime PM counter, and allocate one event.
-			 * So here, we put the runtime PM count for each
-			 * completed event.
-			 */
-			pm_runtime_put_autosuspend(gpu->dev);
 		}
 
 		/* Retire the buffer objects in a work */
@@ -1481,6 +1477,7 @@ static int etnaviv_gpu_hw_resume(struct etnaviv_gpu *gpu)
 	etnaviv_gpu_hw_init(gpu);
 
 	gpu->switch_context = true;
+	gpu->exec_state = -1;
 
 	mutex_unlock(&gpu->lock);
 
@@ -1569,6 +1566,7 @@ static int etnaviv_gpu_platform_probe(struct platform_device *pdev)
 {
 	struct device *dev = &pdev->dev;
 	struct etnaviv_gpu *gpu;
+	u32 dma_mask;
 	int err = 0;
 
 	gpu = devm_kzalloc(dev, sizeof(*gpu), GFP_KERNEL);
@@ -1579,12 +1577,16 @@ static int etnaviv_gpu_platform_probe(struct platform_device *pdev)
 	mutex_init(&gpu->lock);
 
 	/*
-	 * Set the GPU base address to the start of physical memory.  This
-	 * ensures that if we have up to 2GB, the v1 MMU can address the
-	 * highest memory.  This is important as command buffers may be
-	 * allocated outside of this limit.
+	 * Set the GPU linear window to be at the end of the DMA window, where
+	 * the CMA area is likely to reside. This ensures that we are able to
+	 * map the command buffers while having the linear window overlap as
+	 * much RAM as possible, so we can optimize mappings for other buffers.
 	 */
-	gpu->memory_base = PHYS_OFFSET;
+	dma_mask = (u32)dma_get_required_mask(dev);
+	if (dma_mask < PHYS_OFFSET + SZ_2G)
+		gpu->memory_base = PHYS_OFFSET;
+	else
+		gpu->memory_base = dma_mask - SZ_2G + 1;
 
 	/* Map registers: */
 	gpu->mmio = etnaviv_ioremap(pdev, NULL, dev_name(gpu->dev));
diff --git a/drivers/gpu/drm/etnaviv/etnaviv_gpu.h b/drivers/gpu/drm/etnaviv/etnaviv_gpu.h
index f233ac4c7c1c..f5321e2f25ff 100644
--- a/drivers/gpu/drm/etnaviv/etnaviv_gpu.h
+++ b/drivers/gpu/drm/etnaviv/etnaviv_gpu.h
@@ -23,6 +23,7 @@
 #include "etnaviv_drv.h"
 
 struct etnaviv_gem_submit;
+struct etnaviv_vram_mapping;
 
 struct etnaviv_chip_identity {
 	/* Chip model. */
@@ -103,6 +104,7 @@ struct etnaviv_gpu {
 
 	/* 'ring'-buffer: */
 	struct etnaviv_cmdbuf *buffer;
+	int exec_state;
 
 	/* bus base address of memory  */
 	u32 memory_base;
@@ -166,7 +168,7 @@ struct etnaviv_cmdbuf {
 	struct list_head node;
 	/* BOs attached to this command buffer */
 	unsigned int nr_bos;
-	struct etnaviv_gem_object *bo[0];
+	struct etnaviv_vram_mapping *bo_map[0];
 };
 
 static inline void gpu_write(struct etnaviv_gpu *gpu, u32 reg, u32 data)
diff --git a/drivers/gpu/drm/etnaviv/etnaviv_mmu.c b/drivers/gpu/drm/etnaviv/etnaviv_mmu.c
index 6743bc648dc8..29a723fabc17 100644
--- a/drivers/gpu/drm/etnaviv/etnaviv_mmu.c
+++ b/drivers/gpu/drm/etnaviv/etnaviv_mmu.c
@@ -193,7 +193,7 @@ int etnaviv_iommu_map_gem(struct etnaviv_iommu *mmu,
 
 		/*
 		 * Unmap the blocks which need to be reaped from the MMU.
-		 * Clear the mmu pointer to prevent the get_iova finding
+		 * Clear the mmu pointer to prevent the mapping_get finding
 		 * this mapping.
 		 */
 		list_for_each_entry_safe(m, n, &list, scan_node) {
diff --git a/drivers/gpu/drm/etnaviv/state_3d.xml.h b/drivers/gpu/drm/etnaviv/state_3d.xml.h
new file mode 100644
index 000000000000..d7146fd13943
--- /dev/null
+++ b/drivers/gpu/drm/etnaviv/state_3d.xml.h
@@ -0,0 +1,9 @@
+#ifndef STATE_3D_XML
+#define STATE_3D_XML
+
+/* This is a cut-down version of the state_3d.xml.h file */
+
+#define VIVS_TS_FLUSH_CACHE					0x00001650
+#define VIVS_TS_FLUSH_CACHE_FLUSH				0x00000001
+
+#endif /* STATE_3D_XML */
diff --git a/drivers/gpu/drm/exynos/Kconfig b/drivers/gpu/drm/exynos/Kconfig
index 83efca941388..f17d39279596 100644
--- a/drivers/gpu/drm/exynos/Kconfig
+++ b/drivers/gpu/drm/exynos/Kconfig
@@ -1,6 +1,6 @@
 config DRM_EXYNOS
 	tristate "DRM Support for Samsung SoC EXYNOS Series"
-	depends on OF && DRM && (PLAT_SAMSUNG || ARCH_MULTIPLATFORM)
+	depends on OF && DRM && (ARCH_S3C64XX || ARCH_EXYNOS || ARCH_MULTIPLATFORM)
 	select DRM_KMS_HELPER
 	select DRM_KMS_FB_HELPER
 	select FB_CFB_FILLRECT
diff --git a/drivers/gpu/drm/exynos/Makefile b/drivers/gpu/drm/exynos/Makefile
index 6496532aaa91..968b31c522b2 100644
--- a/drivers/gpu/drm/exynos/Makefile
+++ b/drivers/gpu/drm/exynos/Makefile
@@ -2,7 +2,6 @@
 # Makefile for the drm device driver.  This driver provides support for the
 # Direct Rendering Infrastructure (DRI) in XFree86 4.1.0 and higher.
 
-ccflags-y := -Iinclude/drm -Idrivers/gpu/drm/exynos
 exynosdrm-y := exynos_drm_drv.o exynos_drm_crtc.o exynos_drm_fbdev.o \
 		exynos_drm_fb.o exynos_drm_gem.o exynos_drm_core.o \
 		exynos_drm_plane.o
diff --git a/drivers/gpu/drm/exynos/exynos5433_drm_decon.c b/drivers/gpu/drm/exynos/exynos5433_drm_decon.c
index 1bf6a21130c7..5245bc5e82e9 100644
--- a/drivers/gpu/drm/exynos/exynos5433_drm_decon.c
+++ b/drivers/gpu/drm/exynos/exynos5433_drm_decon.c
@@ -93,7 +93,7 @@ static int decon_enable_vblank(struct exynos_drm_crtc *crtc)
 	if (test_bit(BIT_SUSPENDED, &ctx->flags))
 		return -EPERM;
 
-	if (test_and_set_bit(BIT_IRQS_ENABLED, &ctx->flags)) {
+	if (!test_and_set_bit(BIT_IRQS_ENABLED, &ctx->flags)) {
 		val = VIDINTCON0_INTEN;
 		if (ctx->out_type == IFTYPE_I80)
 			val |= VIDINTCON0_FRAMEDONE;
@@ -402,8 +402,6 @@ static void decon_enable(struct exynos_drm_crtc *crtc)
 		decon_enable_vblank(ctx->crtc);
 
 	decon_commit(ctx->crtc);
-
-	set_bit(BIT_SUSPENDED, &ctx->flags);
 }
 
 static void decon_disable(struct exynos_drm_crtc *crtc)
@@ -431,7 +429,7 @@ static void decon_disable(struct exynos_drm_crtc *crtc)
 	set_bit(BIT_SUSPENDED, &ctx->flags);
 }
 
-void decon_te_irq_handler(struct exynos_drm_crtc *crtc)
+static void decon_te_irq_handler(struct exynos_drm_crtc *crtc)
 {
 	struct decon_context *ctx = crtc->ctx;
 
@@ -582,9 +580,9 @@ out:
 static int exynos5433_decon_suspend(struct device *dev)
 {
 	struct decon_context *ctx = dev_get_drvdata(dev);
-	int i;
+	int i = ARRAY_SIZE(decon_clks_name);
 
-	for (i = 0; i < ARRAY_SIZE(decon_clks_name); i++)
+	while (--i >= 0)
 		clk_disable_unprepare(ctx->clks[i]);
 
 	return 0;
diff --git a/drivers/gpu/drm/exynos/exynos7_drm_decon.c b/drivers/gpu/drm/exynos/exynos7_drm_decon.c
index 52bda3b42fe0..93361073af9a 100644
--- a/drivers/gpu/drm/exynos/exynos7_drm_decon.c
+++ b/drivers/gpu/drm/exynos/exynos7_drm_decon.c
@@ -60,7 +60,6 @@ struct decon_context {
 	wait_queue_head_t		wait_vsync_queue;
 	atomic_t			wait_vsync_event;
 
-	struct exynos_drm_panel_info panel;
 	struct drm_encoder *encoder;
 };
 
diff --git a/drivers/gpu/drm/exynos/exynos_dp_core.c b/drivers/gpu/drm/exynos/exynos_dp_core.c
index 9fd12c621270..cff8dc788820 100644
--- a/drivers/gpu/drm/exynos/exynos_dp_core.c
+++ b/drivers/gpu/drm/exynos/exynos_dp_core.c
@@ -977,9 +977,7 @@ static int exynos_dp_get_modes(struct drm_connector *connector)
 		return 0;
 	}
 
-	drm_display_mode_from_videomode(&dp->priv.vm, mode);
-	mode->width_mm = dp->priv.width_mm;
-	mode->height_mm = dp->priv.height_mm;
+	drm_display_mode_from_videomode(&dp->vm, mode);
 	connector->display_info.width_mm = mode->width_mm;
 	connector->display_info.height_mm = mode->height_mm;
 
@@ -1241,8 +1239,7 @@ static int exynos_dp_dt_parse_panel(struct exynos_dp_device *dp)
 {
 	int ret;
 
-	ret = of_get_videomode(dp->dev->of_node, &dp->priv.vm,
-			OF_USE_NATIVE_MODE);
+	ret = of_get_videomode(dp->dev->of_node, &dp->vm, OF_USE_NATIVE_MODE);
 	if (ret) {
 		DRM_ERROR("failed: of_get_videomode() : %d\n", ret);
 		return ret;
diff --git a/drivers/gpu/drm/exynos/exynos_dp_core.h b/drivers/gpu/drm/exynos/exynos_dp_core.h
index 66eec4b2d5c6..b5c2d8f47f9c 100644
--- a/drivers/gpu/drm/exynos/exynos_dp_core.h
+++ b/drivers/gpu/drm/exynos/exynos_dp_core.h
@@ -16,6 +16,7 @@
 #include <drm/drm_crtc.h>
 #include <drm/drm_dp_helper.h>
 #include <drm/exynos_drm.h>
+#include <video/videomode.h>
 
 #include "exynos_drm_drv.h"
 
@@ -164,8 +165,7 @@ struct exynos_dp_device {
 	struct phy		*phy;
 	int			dpms_mode;
 	int			hpd_gpio;
-
-	struct exynos_drm_panel_info priv;
+	struct videomode	vm;
 };
 
 /* exynos_dp_reg.c */
diff --git a/drivers/gpu/drm/exynos/exynos_drm_drv.c b/drivers/gpu/drm/exynos/exynos_drm_drv.c
index 1e535f981240..5344940c8a07 100644
--- a/drivers/gpu/drm/exynos/exynos_drm_drv.c
+++ b/drivers/gpu/drm/exynos/exynos_drm_drv.c
@@ -130,6 +130,8 @@ static void exynos_drm_atomic_work(struct work_struct *work)
 	exynos_atomic_commit_complete(commit);
 }
 
+static struct device *exynos_drm_get_dma_device(void);
+
 static int exynos_drm_load(struct drm_device *dev, unsigned long flags)
 {
 	struct exynos_drm_private *private;
@@ -147,6 +149,16 @@ static int exynos_drm_load(struct drm_device *dev, unsigned long flags)
 	dev_set_drvdata(dev->dev, dev);
 	dev->dev_private = (void *)private;
 
+	/* the first real CRTC device is used for all dma mapping operations */
+	private->dma_dev = exynos_drm_get_dma_device();
+	if (!private->dma_dev) {
+		DRM_ERROR("no device found for DMA mapping operations.\n");
+		ret = -ENODEV;
+		goto err_free_private;
+	}
+	DRM_INFO("Exynos DRM: using %s device for DMA mapping operations\n",
+		 dev_name(private->dma_dev));
+
 	/*
 	 * create mapping to manage iommu table and set a pointer to iommu
 	 * mapping structure to iommu_mapping of private data.
@@ -358,6 +370,8 @@ static const struct vm_operations_struct exynos_drm_gem_vm_ops = {
 static const struct drm_ioctl_desc exynos_ioctls[] = {
 	DRM_IOCTL_DEF_DRV(EXYNOS_GEM_CREATE, exynos_drm_gem_create_ioctl,
 			DRM_AUTH | DRM_RENDER_ALLOW),
+	DRM_IOCTL_DEF_DRV(EXYNOS_GEM_MAP, exynos_drm_gem_map_ioctl,
+			DRM_AUTH | DRM_RENDER_ALLOW),
 	DRM_IOCTL_DEF_DRV(EXYNOS_GEM_GET, exynos_drm_gem_get_ioctl,
 			DRM_RENDER_ALLOW),
 	DRM_IOCTL_DEF_DRV(EXYNOS_VIDI_CONNECTION, vidi_connection_ioctl,
@@ -481,69 +495,65 @@ static const struct dev_pm_ops exynos_drm_pm_ops = {
 /* forward declaration */
 static struct platform_driver exynos_drm_platform_driver;
 
+struct exynos_drm_driver_info {
+	struct platform_driver *driver;
+	unsigned int flags;
+};
+
+#define DRM_COMPONENT_DRIVER	BIT(0)	/* supports component framework */
+#define DRM_VIRTUAL_DEVICE	BIT(1)	/* create virtual platform device */
+#define DRM_DMA_DEVICE		BIT(2)	/* can be used for dma allocations */
+
+#define DRV_PTR(drv, cond) (IS_ENABLED(cond) ? &drv : NULL)
+
 /*
  * Connector drivers should not be placed before associated crtc drivers,
  * because connector requires pipe number of its crtc during initialization.
  */
-static struct platform_driver *const exynos_drm_kms_drivers[] = {
-#ifdef CONFIG_DRM_EXYNOS_FIMD
-	&fimd_driver,
-#endif
-#ifdef CONFIG_DRM_EXYNOS5433_DECON
-	&exynos5433_decon_driver,
-#endif
-#ifdef CONFIG_DRM_EXYNOS7_DECON
-	&decon_driver,
-#endif
-#ifdef CONFIG_DRM_EXYNOS_MIC
-	&mic_driver,
-#endif
-#ifdef CONFIG_DRM_EXYNOS_DP
-	&dp_driver,
-#endif
-#ifdef CONFIG_DRM_EXYNOS_DSI
-	&dsi_driver,
-#endif
-#ifdef CONFIG_DRM_EXYNOS_MIXER
-	&mixer_driver,
-#endif
-#ifdef CONFIG_DRM_EXYNOS_HDMI
-	&hdmi_driver,
-#endif
-#ifdef CONFIG_DRM_EXYNOS_VIDI
-	&vidi_driver,
-#endif
-};
-
-static struct platform_driver *const exynos_drm_non_kms_drivers[] = {
-#ifdef CONFIG_DRM_EXYNOS_G2D
-	&g2d_driver,
-#endif
-#ifdef CONFIG_DRM_EXYNOS_FIMC
-	&fimc_driver,
-#endif
-#ifdef CONFIG_DRM_EXYNOS_ROTATOR
-	&rotator_driver,
-#endif
-#ifdef CONFIG_DRM_EXYNOS_GSC
-	&gsc_driver,
-#endif
-#ifdef CONFIG_DRM_EXYNOS_IPP
-	&ipp_driver,
-#endif
-	&exynos_drm_platform_driver,
-};
-
-static struct platform_driver *const exynos_drm_drv_with_simple_dev[] = {
-#ifdef CONFIG_DRM_EXYNOS_VIDI
-	&vidi_driver,
-#endif
-#ifdef CONFIG_DRM_EXYNOS_IPP
-	&ipp_driver,
-#endif
-	&exynos_drm_platform_driver,
+static struct exynos_drm_driver_info exynos_drm_drivers[] = {
+	{
+		DRV_PTR(fimd_driver, CONFIG_DRM_EXYNOS_FIMD),
+		DRM_COMPONENT_DRIVER | DRM_DMA_DEVICE
+	}, {
+		DRV_PTR(exynos5433_decon_driver, CONFIG_DRM_EXYNOS5433_DECON),
+		DRM_COMPONENT_DRIVER | DRM_DMA_DEVICE
+	}, {
+		DRV_PTR(decon_driver, CONFIG_DRM_EXYNOS7_DECON),
+		DRM_COMPONENT_DRIVER | DRM_DMA_DEVICE
+	}, {
+		DRV_PTR(mixer_driver, CONFIG_DRM_EXYNOS_MIXER),
+		DRM_COMPONENT_DRIVER | DRM_DMA_DEVICE
+	}, {
+		DRV_PTR(mic_driver, CONFIG_DRM_EXYNOS_MIC),
+		DRM_COMPONENT_DRIVER
+	}, {
+		DRV_PTR(dp_driver, CONFIG_DRM_EXYNOS_DP),
+		DRM_COMPONENT_DRIVER
+	}, {
+		DRV_PTR(dsi_driver, CONFIG_DRM_EXYNOS_DSI),
+		DRM_COMPONENT_DRIVER
+	}, {
+		DRV_PTR(hdmi_driver, CONFIG_DRM_EXYNOS_HDMI),
+		DRM_COMPONENT_DRIVER
+	}, {
+		DRV_PTR(vidi_driver, CONFIG_DRM_EXYNOS_VIDI),
+		DRM_COMPONENT_DRIVER | DRM_VIRTUAL_DEVICE
+	}, {
+		DRV_PTR(g2d_driver, CONFIG_DRM_EXYNOS_G2D),
+	}, {
+		DRV_PTR(fimc_driver, CONFIG_DRM_EXYNOS_FIMC),
+	}, {
+		DRV_PTR(rotator_driver, CONFIG_DRM_EXYNOS_ROTATOR),
+	}, {
+		DRV_PTR(gsc_driver, CONFIG_DRM_EXYNOS_GSC),
+	}, {
+		DRV_PTR(ipp_driver, CONFIG_DRM_EXYNOS_IPP),
+		DRM_VIRTUAL_DEVICE
+	}, {
+		&exynos_drm_platform_driver,
+		DRM_VIRTUAL_DEVICE
+	}
 };
-#define PDEV_COUNT ARRAY_SIZE(exynos_drm_drv_with_simple_dev)
 
 static int compare_dev(struct device *dev, void *data)
 {
@@ -555,11 +565,15 @@ static struct component_match *exynos_drm_match_add(struct device *dev)
 	struct component_match *match = NULL;
 	int i;
 
-	for (i = 0; i < ARRAY_SIZE(exynos_drm_kms_drivers); ++i) {
-		struct device_driver *drv = &exynos_drm_kms_drivers[i]->driver;
+	for (i = 0; i < ARRAY_SIZE(exynos_drm_drivers); ++i) {
+		struct exynos_drm_driver_info *info = &exynos_drm_drivers[i];
 		struct device *p = NULL, *d;
 
-		while ((d = bus_find_device(&platform_bus_type, p, drv,
+		if (!info->driver || !(info->flags & DRM_COMPONENT_DRIVER))
+			continue;
+
+		while ((d = bus_find_device(&platform_bus_type, p,
+					    &info->driver->driver,
 					    (void *)platform_bus_type.match))) {
 			put_device(p);
 			component_match_add(dev, &match, compare_dev, d);
@@ -616,91 +630,102 @@ static struct platform_driver exynos_drm_platform_driver = {
 	},
 };
 
-static struct platform_device *exynos_drm_pdevs[PDEV_COUNT];
+static struct device *exynos_drm_get_dma_device(void)
+{
+	int i;
+
+	for (i = 0; i < ARRAY_SIZE(exynos_drm_drivers); ++i) {
+		struct exynos_drm_driver_info *info = &exynos_drm_drivers[i];
+		struct device *dev;
+
+		if (!info->driver || !(info->flags & DRM_DMA_DEVICE))
+			continue;
+
+		while ((dev = bus_find_device(&platform_bus_type, NULL,
+					    &info->driver->driver,
+					    (void *)platform_bus_type.match))) {
+			put_device(dev);
+			return dev;
+		}
+	}
+	return NULL;
+}
 
 static void exynos_drm_unregister_devices(void)
 {
-	int i = PDEV_COUNT;
+	int i;
+
+	for (i = ARRAY_SIZE(exynos_drm_drivers) - 1; i >= 0; --i) {
+		struct exynos_drm_driver_info *info = &exynos_drm_drivers[i];
+		struct device *dev;
 
-	while (--i >= 0) {
-		platform_device_unregister(exynos_drm_pdevs[i]);
-		exynos_drm_pdevs[i] = NULL;
+		if (!info->driver || !(info->flags & DRM_VIRTUAL_DEVICE))
+			continue;
+
+		while ((dev = bus_find_device(&platform_bus_type, NULL,
+					    &info->driver->driver,
+					    (void *)platform_bus_type.match))) {
+			put_device(dev);
+			platform_device_unregister(to_platform_device(dev));
+		}
 	}
 }
 
 static int exynos_drm_register_devices(void)
 {
+	struct platform_device *pdev;
 	int i;
 
-	for (i = 0; i < PDEV_COUNT; ++i) {
-		struct platform_driver *d = exynos_drm_drv_with_simple_dev[i];
-		struct platform_device *pdev =
-			platform_device_register_simple(d->driver.name, -1,
-							NULL, 0);
+	for (i = 0; i < ARRAY_SIZE(exynos_drm_drivers); ++i) {
+		struct exynos_drm_driver_info *info = &exynos_drm_drivers[i];
 
-		if (!IS_ERR(pdev)) {
-			exynos_drm_pdevs[i] = pdev;
+		if (!info->driver || !(info->flags & DRM_VIRTUAL_DEVICE))
 			continue;
-		}
-		while (--i >= 0) {
-			platform_device_unregister(exynos_drm_pdevs[i]);
-			exynos_drm_pdevs[i] = NULL;
-		}
 
-		return PTR_ERR(pdev);
+		pdev = platform_device_register_simple(
+					info->driver->driver.name, -1, NULL, 0);
+		if (IS_ERR(pdev))
+			goto fail;
 	}
 
 	return 0;
+fail:
+	exynos_drm_unregister_devices();
+	return PTR_ERR(pdev);
 }
 
-static void exynos_drm_unregister_drivers(struct platform_driver * const *drv,
-					  int count)
+static void exynos_drm_unregister_drivers(void)
 {
-	while (--count >= 0)
-		platform_driver_unregister(drv[count]);
-}
+	int i;
 
-static int exynos_drm_register_drivers(struct platform_driver * const *drv,
-				       int count)
-{
-	int i, ret;
+	for (i = ARRAY_SIZE(exynos_drm_drivers) - 1; i >= 0; --i) {
+		struct exynos_drm_driver_info *info = &exynos_drm_drivers[i];
 
-	for (i = 0; i < count; ++i) {
-		ret = platform_driver_register(drv[i]);
-		if (!ret)
+		if (!info->driver)
 			continue;
 
-		while (--i >= 0)
-			platform_driver_unregister(drv[i]);
-
-		return ret;
+		platform_driver_unregister(info->driver);
 	}
-
-	return 0;
 }
 
-static inline int exynos_drm_register_kms_drivers(void)
+static int exynos_drm_register_drivers(void)
 {
-	return exynos_drm_register_drivers(exynos_drm_kms_drivers,
-					ARRAY_SIZE(exynos_drm_kms_drivers));
-}
+	int i, ret;
 
-static inline int exynos_drm_register_non_kms_drivers(void)
-{
-	return exynos_drm_register_drivers(exynos_drm_non_kms_drivers,
-					ARRAY_SIZE(exynos_drm_non_kms_drivers));
-}
+	for (i = 0; i < ARRAY_SIZE(exynos_drm_drivers); ++i) {
+		struct exynos_drm_driver_info *info = &exynos_drm_drivers[i];
 
-static inline void exynos_drm_unregister_kms_drivers(void)
-{
-	exynos_drm_unregister_drivers(exynos_drm_kms_drivers,
-					ARRAY_SIZE(exynos_drm_kms_drivers));
-}
+		if (!info->driver)
+			continue;
 
-static inline void exynos_drm_unregister_non_kms_drivers(void)
-{
-	exynos_drm_unregister_drivers(exynos_drm_non_kms_drivers,
-					ARRAY_SIZE(exynos_drm_non_kms_drivers));
+		ret = platform_driver_register(info->driver);
+		if (ret)
+			goto fail;
+	}
+	return 0;
+fail:
+	exynos_drm_unregister_drivers();
+	return ret;
 }
 
 static int exynos_drm_init(void)
@@ -711,19 +736,12 @@ static int exynos_drm_init(void)
 	if (ret)
 		return ret;
 
-	ret = exynos_drm_register_kms_drivers();
+	ret = exynos_drm_register_drivers();
 	if (ret)
 		goto err_unregister_pdevs;
 
-	ret = exynos_drm_register_non_kms_drivers();
-	if (ret)
-		goto err_unregister_kms_drivers;
-
 	return 0;
 
-err_unregister_kms_drivers:
-	exynos_drm_unregister_kms_drivers();
-
 err_unregister_pdevs:
 	exynos_drm_unregister_devices();
 
@@ -732,8 +750,7 @@ err_unregister_pdevs:
 
 static void exynos_drm_exit(void)
 {
-	exynos_drm_unregister_non_kms_drivers();
-	exynos_drm_unregister_kms_drivers();
+	exynos_drm_unregister_drivers();
 	exynos_drm_unregister_devices();
 }
 
diff --git a/drivers/gpu/drm/exynos/exynos_drm_drv.h b/drivers/gpu/drm/exynos/exynos_drm_drv.h
index 17b5ded72ff1..502f750bad2a 100644
--- a/drivers/gpu/drm/exynos/exynos_drm_drv.h
+++ b/drivers/gpu/drm/exynos/exynos_drm_drv.h
@@ -219,8 +219,10 @@ struct exynos_drm_private {
 	struct drm_crtc *crtc[MAX_CRTC];
 	struct drm_property *plane_zpos_property;
 
+	struct device *dma_dev;
 	unsigned long da_start;
 	unsigned long da_space_size;
+	void *mapping;
 
 	unsigned int pipe;
 
@@ -230,6 +232,13 @@ struct exynos_drm_private {
 	wait_queue_head_t	wait;
 };
 
+static inline struct device *to_dma_dev(struct drm_device *dev)
+{
+	struct exynos_drm_private *priv = dev->dev_private;
+
+	return priv->dma_dev;
+}
+
 /*
  * Exynos drm sub driver structure.
  *
@@ -297,7 +306,6 @@ extern struct platform_driver dp_driver;
 extern struct platform_driver dsi_driver;
 extern struct platform_driver mixer_driver;
 extern struct platform_driver hdmi_driver;
-extern struct platform_driver exynos_drm_common_hdmi_driver;
 extern struct platform_driver vidi_driver;
 extern struct platform_driver g2d_driver;
 extern struct platform_driver fimc_driver;
diff --git a/drivers/gpu/drm/exynos/exynos_drm_dsi.c b/drivers/gpu/drm/exynos/exynos_drm_dsi.c
index 736115c580fc..63c84a106c0b 100644
--- a/drivers/gpu/drm/exynos/exynos_drm_dsi.c
+++ b/drivers/gpu/drm/exynos/exynos_drm_dsi.c
@@ -10,6 +10,8 @@
  * published by the Free Software Foundation.
 */
 
+#include <asm/unaligned.h>
+
 #include <drm/drmP.h>
 #include <drm/drm_crtc_helper.h>
 #include <drm/drm_mipi_dsi.h>
@@ -209,12 +211,6 @@
 
 #define OLD_SCLK_MIPI_CLK_NAME "pll_clk"
 
-#define REG_ADDR(dsi, reg_idx)		((dsi)->reg_base + \
-					dsi->driver_data->reg_ofs[(reg_idx)])
-#define DSI_WRITE(dsi, reg_idx, val)	writel((val), \
-					REG_ADDR((dsi), (reg_idx)))
-#define DSI_READ(dsi, reg_idx)		readl(REG_ADDR((dsi), (reg_idx)))
-
 static char *clk_names[5] = { "bus_clk", "sclk_mipi",
 	"phyclk_mipidphy0_bitclkdiv8", "phyclk_mipidphy0_rxclkesc0",
 	"sclk_rgb_vclk_to_dsim0" };
@@ -228,12 +224,8 @@ struct exynos_dsi_transfer {
 	struct list_head list;
 	struct completion completed;
 	int result;
-	u8 data_id;
-	u8 data[2];
+	struct mipi_dsi_packet packet;
 	u16 flags;
-
-	const u8 *tx_payload;
-	u16 tx_len;
 	u16 tx_done;
 
 	u8 *rx_payload;
@@ -247,7 +239,7 @@ struct exynos_dsi_transfer {
 #define DSIM_STATE_VIDOUT_AVAILABLE	BIT(3)
 
 struct exynos_dsi_driver_data {
-	unsigned int *reg_ofs;
+	const unsigned int *reg_ofs;
 	unsigned int plltmr_reg;
 	unsigned int has_freqband:1;
 	unsigned int has_clklane_stop:1;
@@ -255,7 +247,7 @@ struct exynos_dsi_driver_data {
 	unsigned int max_freq;
 	unsigned int wait_for_reset;
 	unsigned int num_bits_resol;
-	unsigned int *reg_values;
+	const unsigned int *reg_values;
 };
 
 struct exynos_dsi {
@@ -324,7 +316,20 @@ enum reg_idx {
 	DSIM_PHYTIMING2_REG,
 	NUM_REGS
 };
-static unsigned int exynos_reg_ofs[] = {
+
+static inline void exynos_dsi_write(struct exynos_dsi *dsi, enum reg_idx idx,
+				    u32 val)
+{
+
+	writel(val, dsi->reg_base + dsi->driver_data->reg_ofs[idx]);
+}
+
+static inline u32 exynos_dsi_read(struct exynos_dsi *dsi, enum reg_idx idx)
+{
+	return readl(dsi->reg_base + dsi->driver_data->reg_ofs[idx]);
+}
+
+static const unsigned int exynos_reg_ofs[] = {
 	[DSIM_STATUS_REG] =  0x00,
 	[DSIM_SWRST_REG] =  0x04,
 	[DSIM_CLKCTRL_REG] =  0x08,
@@ -348,7 +353,7 @@ static unsigned int exynos_reg_ofs[] = {
 	[DSIM_PHYTIMING2_REG] =  0x6c,
 };
 
-static unsigned int exynos5433_reg_ofs[] = {
+static const unsigned int exynos5433_reg_ofs[] = {
 	[DSIM_STATUS_REG] = 0x04,
 	[DSIM_SWRST_REG] = 0x0C,
 	[DSIM_CLKCTRL_REG] = 0x10,
@@ -390,7 +395,7 @@ enum reg_value_idx {
 	PHYTIMING_HS_TRAIL
 };
 
-static unsigned int reg_values[] = {
+static const unsigned int reg_values[] = {
 	[RESET_TYPE] = DSIM_SWRST,
 	[PLL_TIMER] = 500,
 	[STOP_STATE_CNT] = 0xf,
@@ -408,7 +413,25 @@ static unsigned int reg_values[] = {
 	[PHYTIMING_HS_TRAIL] = DSIM_PHYTIMING2_HS_TRAIL(0x0b),
 };
 
-static unsigned int exynos5433_reg_values[] = {
+static const unsigned int exynos5422_reg_values[] = {
+	[RESET_TYPE] = DSIM_SWRST,
+	[PLL_TIMER] = 500,
+	[STOP_STATE_CNT] = 0xf,
+	[PHYCTRL_ULPS_EXIT] = DSIM_PHYCTRL_ULPS_EXIT(0xaf),
+	[PHYCTRL_VREG_LP] = 0,
+	[PHYCTRL_SLEW_UP] = 0,
+	[PHYTIMING_LPX] = DSIM_PHYTIMING_LPX(0x08),
+	[PHYTIMING_HS_EXIT] = DSIM_PHYTIMING_HS_EXIT(0x0d),
+	[PHYTIMING_CLK_PREPARE] = DSIM_PHYTIMING1_CLK_PREPARE(0x09),
+	[PHYTIMING_CLK_ZERO] = DSIM_PHYTIMING1_CLK_ZERO(0x30),
+	[PHYTIMING_CLK_POST] = DSIM_PHYTIMING1_CLK_POST(0x0e),
+	[PHYTIMING_CLK_TRAIL] = DSIM_PHYTIMING1_CLK_TRAIL(0x0a),
+	[PHYTIMING_HS_PREPARE] = DSIM_PHYTIMING2_HS_PREPARE(0x0c),
+	[PHYTIMING_HS_ZERO] = DSIM_PHYTIMING2_HS_ZERO(0x11),
+	[PHYTIMING_HS_TRAIL] = DSIM_PHYTIMING2_HS_TRAIL(0x0d),
+};
+
+static const unsigned int exynos5433_reg_values[] = {
 	[RESET_TYPE] = DSIM_FUNCRST,
 	[PLL_TIMER] = 22200,
 	[STOP_STATE_CNT] = 0xa,
@@ -426,7 +449,7 @@ static unsigned int exynos5433_reg_values[] = {
 	[PHYTIMING_HS_TRAIL] = DSIM_PHYTIMING2_HS_TRAIL(0x0c),
 };
 
-static struct exynos_dsi_driver_data exynos3_dsi_driver_data = {
+static const struct exynos_dsi_driver_data exynos3_dsi_driver_data = {
 	.reg_ofs = exynos_reg_ofs,
 	.plltmr_reg = 0x50,
 	.has_freqband = 1,
@@ -438,7 +461,7 @@ static struct exynos_dsi_driver_data exynos3_dsi_driver_data = {
 	.reg_values = reg_values,
 };
 
-static struct exynos_dsi_driver_data exynos4_dsi_driver_data = {
+static const struct exynos_dsi_driver_data exynos4_dsi_driver_data = {
 	.reg_ofs = exynos_reg_ofs,
 	.plltmr_reg = 0x50,
 	.has_freqband = 1,
@@ -450,7 +473,7 @@ static struct exynos_dsi_driver_data exynos4_dsi_driver_data = {
 	.reg_values = reg_values,
 };
 
-static struct exynos_dsi_driver_data exynos4415_dsi_driver_data = {
+static const struct exynos_dsi_driver_data exynos4415_dsi_driver_data = {
 	.reg_ofs = exynos_reg_ofs,
 	.plltmr_reg = 0x58,
 	.has_clklane_stop = 1,
@@ -461,7 +484,7 @@ static struct exynos_dsi_driver_data exynos4415_dsi_driver_data = {
 	.reg_values = reg_values,
 };
 
-static struct exynos_dsi_driver_data exynos5_dsi_driver_data = {
+static const struct exynos_dsi_driver_data exynos5_dsi_driver_data = {
 	.reg_ofs = exynos_reg_ofs,
 	.plltmr_reg = 0x58,
 	.num_clks = 2,
@@ -471,7 +494,7 @@ static struct exynos_dsi_driver_data exynos5_dsi_driver_data = {
 	.reg_values = reg_values,
 };
 
-static struct exynos_dsi_driver_data exynos5433_dsi_driver_data = {
+static const struct exynos_dsi_driver_data exynos5433_dsi_driver_data = {
 	.reg_ofs = exynos5433_reg_ofs,
 	.plltmr_reg = 0xa0,
 	.has_clklane_stop = 1,
@@ -482,7 +505,18 @@ static struct exynos_dsi_driver_data exynos5433_dsi_driver_data = {
 	.reg_values = exynos5433_reg_values,
 };
 
-static struct of_device_id exynos_dsi_of_match[] = {
+static const struct exynos_dsi_driver_data exynos5422_dsi_driver_data = {
+	.reg_ofs = exynos5433_reg_ofs,
+	.plltmr_reg = 0xa0,
+	.has_clklane_stop = 1,
+	.num_clks = 2,
+	.max_freq = 1500,
+	.wait_for_reset = 1,
+	.num_bits_resol = 12,
+	.reg_values = exynos5422_reg_values,
+};
+
+static const struct of_device_id exynos_dsi_of_match[] = {
 	{ .compatible = "samsung,exynos3250-mipi-dsi",
 	  .data = &exynos3_dsi_driver_data },
 	{ .compatible = "samsung,exynos4210-mipi-dsi",
@@ -491,6 +525,8 @@ static struct of_device_id exynos_dsi_of_match[] = {
 	  .data = &exynos4415_dsi_driver_data },
 	{ .compatible = "samsung,exynos5410-mipi-dsi",
 	  .data = &exynos5_dsi_driver_data },
+	{ .compatible = "samsung,exynos5422-mipi-dsi",
+	  .data = &exynos5422_dsi_driver_data },
 	{ .compatible = "samsung,exynos5433-mipi-dsi",
 	  .data = &exynos5433_dsi_driver_data },
 	{ }
@@ -515,10 +551,10 @@ static void exynos_dsi_wait_for_reset(struct exynos_dsi *dsi)
 
 static void exynos_dsi_reset(struct exynos_dsi *dsi)
 {
-	struct exynos_dsi_driver_data *driver_data = dsi->driver_data;
+	u32 reset_val = dsi->driver_data->reg_values[RESET_TYPE];
 
 	reinit_completion(&dsi->completed);
-	DSI_WRITE(dsi, DSIM_SWRST_REG, driver_data->reg_values[RESET_TYPE]);
+	exynos_dsi_write(dsi, DSIM_SWRST_REG, reset_val);
 }
 
 #ifndef MHZ
@@ -621,7 +657,7 @@ static unsigned long exynos_dsi_set_pll(struct exynos_dsi *dsi,
 		reg |= DSIM_FREQ_BAND(band);
 	}
 
-	DSI_WRITE(dsi, DSIM_PLLCTRL_REG, reg);
+	exynos_dsi_write(dsi, DSIM_PLLCTRL_REG, reg);
 
 	timeout = 1000;
 	do {
@@ -629,7 +665,7 @@ static unsigned long exynos_dsi_set_pll(struct exynos_dsi *dsi,
 			dev_err(dsi->dev, "PLL failed to stabilize\n");
 			return 0;
 		}
-		reg = DSI_READ(dsi, DSIM_STATUS_REG);
+		reg = exynos_dsi_read(dsi, DSIM_STATUS_REG);
 	} while ((reg & DSIM_PLL_STABLE) == 0);
 
 	return fout;
@@ -659,7 +695,7 @@ static int exynos_dsi_enable_clock(struct exynos_dsi *dsi)
 	dev_dbg(dsi->dev, "hs_clk = %lu, byte_clk = %lu, esc_clk = %lu\n",
 		hs_clk, byte_clk, esc_clk);
 
-	reg = DSI_READ(dsi, DSIM_CLKCTRL_REG);
+	reg = exynos_dsi_read(dsi, DSIM_CLKCTRL_REG);
 	reg &= ~(DSIM_ESC_PRESCALER_MASK | DSIM_LANE_ESC_CLK_EN_CLK
 			| DSIM_LANE_ESC_CLK_EN_DATA_MASK | DSIM_PLL_BYPASS
 			| DSIM_BYTE_CLK_SRC_MASK);
@@ -669,7 +705,7 @@ static int exynos_dsi_enable_clock(struct exynos_dsi *dsi)
 			| DSIM_LANE_ESC_CLK_EN_DATA(BIT(dsi->lanes) - 1)
 			| DSIM_BYTE_CLK_SRC(0)
 			| DSIM_TX_REQUEST_HSCLK;
-	DSI_WRITE(dsi, DSIM_CLKCTRL_REG, reg);
+	exynos_dsi_write(dsi, DSIM_CLKCTRL_REG, reg);
 
 	return 0;
 }
@@ -677,7 +713,7 @@ static int exynos_dsi_enable_clock(struct exynos_dsi *dsi)
 static void exynos_dsi_set_phy_ctrl(struct exynos_dsi *dsi)
 {
 	struct exynos_dsi_driver_data *driver_data = dsi->driver_data;
-	unsigned int *reg_values = driver_data->reg_values;
+	const unsigned int *reg_values = driver_data->reg_values;
 	u32 reg;
 
 	if (driver_data->has_freqband)
@@ -686,7 +722,7 @@ static void exynos_dsi_set_phy_ctrl(struct exynos_dsi *dsi)
 	/* B D-PHY: D-PHY Master & Slave Analog Block control */
 	reg = reg_values[PHYCTRL_ULPS_EXIT] | reg_values[PHYCTRL_VREG_LP] |
 		reg_values[PHYCTRL_SLEW_UP];
-	DSI_WRITE(dsi, DSIM_PHYCTRL_REG, reg);
+	exynos_dsi_write(dsi, DSIM_PHYCTRL_REG, reg);
 
 	/*
 	 * T LPX: Transmitted length of any Low-Power state period
@@ -694,7 +730,7 @@ static void exynos_dsi_set_phy_ctrl(struct exynos_dsi *dsi)
 	 *	burst
 	 */
 	reg = reg_values[PHYTIMING_LPX] | reg_values[PHYTIMING_HS_EXIT];
-	DSI_WRITE(dsi, DSIM_PHYTIMING_REG, reg);
+	exynos_dsi_write(dsi, DSIM_PHYTIMING_REG, reg);
 
 	/*
 	 * T CLK-PREPARE: Time that the transmitter drives the Clock Lane LP-00
@@ -714,7 +750,7 @@ static void exynos_dsi_set_phy_ctrl(struct exynos_dsi *dsi)
 		reg_values[PHYTIMING_CLK_POST] |
 		reg_values[PHYTIMING_CLK_TRAIL];
 
-	DSI_WRITE(dsi, DSIM_PHYTIMING1_REG, reg);
+	exynos_dsi_write(dsi, DSIM_PHYTIMING1_REG, reg);
 
 	/*
 	 * T HS-PREPARE: Time that the transmitter drives the Data Lane LP-00
@@ -727,29 +763,29 @@ static void exynos_dsi_set_phy_ctrl(struct exynos_dsi *dsi)
 	 */
 	reg = reg_values[PHYTIMING_HS_PREPARE] | reg_values[PHYTIMING_HS_ZERO] |
 		reg_values[PHYTIMING_HS_TRAIL];
-	DSI_WRITE(dsi, DSIM_PHYTIMING2_REG, reg);
+	exynos_dsi_write(dsi, DSIM_PHYTIMING2_REG, reg);
 }
 
 static void exynos_dsi_disable_clock(struct exynos_dsi *dsi)
 {
 	u32 reg;
 
-	reg = DSI_READ(dsi, DSIM_CLKCTRL_REG);
+	reg = exynos_dsi_read(dsi, DSIM_CLKCTRL_REG);
 	reg &= ~(DSIM_LANE_ESC_CLK_EN_CLK | DSIM_LANE_ESC_CLK_EN_DATA_MASK
 			| DSIM_ESC_CLKEN | DSIM_BYTE_CLKEN);
-	DSI_WRITE(dsi, DSIM_CLKCTRL_REG, reg);
+	exynos_dsi_write(dsi, DSIM_CLKCTRL_REG, reg);
 
-	reg = DSI_READ(dsi, DSIM_PLLCTRL_REG);
+	reg = exynos_dsi_read(dsi, DSIM_PLLCTRL_REG);
 	reg &= ~DSIM_PLL_EN;
-	DSI_WRITE(dsi, DSIM_PLLCTRL_REG, reg);
+	exynos_dsi_write(dsi, DSIM_PLLCTRL_REG, reg);
 }
 
 static void exynos_dsi_enable_lane(struct exynos_dsi *dsi, u32 lane)
 {
-	u32 reg = DSI_READ(dsi, DSIM_CONFIG_REG);
+	u32 reg = exynos_dsi_read(dsi, DSIM_CONFIG_REG);
 	reg |= (DSIM_NUM_OF_DATA_LANE(dsi->lanes - 1) | DSIM_LANE_EN_CLK |
 			DSIM_LANE_EN(lane));
-	DSI_WRITE(dsi, DSIM_CONFIG_REG, reg);
+	exynos_dsi_write(dsi, DSIM_CONFIG_REG, reg);
 }
 
 static int exynos_dsi_init_link(struct exynos_dsi *dsi)
@@ -760,14 +796,14 @@ static int exynos_dsi_init_link(struct exynos_dsi *dsi)
 	u32 lanes_mask;
 
 	/* Initialize FIFO pointers */
-	reg = DSI_READ(dsi, DSIM_FIFOCTRL_REG);
+	reg = exynos_dsi_read(dsi, DSIM_FIFOCTRL_REG);
 	reg &= ~0x1f;
-	DSI_WRITE(dsi, DSIM_FIFOCTRL_REG, reg);
+	exynos_dsi_write(dsi, DSIM_FIFOCTRL_REG, reg);
 
 	usleep_range(9000, 11000);
 
 	reg |= 0x1f;
-	DSI_WRITE(dsi, DSIM_FIFOCTRL_REG, reg);
+	exynos_dsi_write(dsi, DSIM_FIFOCTRL_REG, reg);
 	usleep_range(9000, 11000);
 
 	/* DSI configuration */
@@ -836,7 +872,7 @@ static int exynos_dsi_init_link(struct exynos_dsi *dsi)
 			dsi->mode_flags & MIPI_DSI_CLOCK_NON_CONTINUOUS) {
 		reg |= DSIM_CLKLANE_STOP;
 	}
-	DSI_WRITE(dsi, DSIM_CONFIG_REG, reg);
+	exynos_dsi_write(dsi, DSIM_CONFIG_REG, reg);
 
 	lanes_mask = BIT(dsi->lanes) - 1;
 	exynos_dsi_enable_lane(dsi, lanes_mask);
@@ -849,19 +885,19 @@ static int exynos_dsi_init_link(struct exynos_dsi *dsi)
 			return -EFAULT;
 		}
 
-		reg = DSI_READ(dsi, DSIM_STATUS_REG);
+		reg = exynos_dsi_read(dsi, DSIM_STATUS_REG);
 		if ((reg & DSIM_STOP_STATE_DAT(lanes_mask))
 		    != DSIM_STOP_STATE_DAT(lanes_mask))
 			continue;
 	} while (!(reg & (DSIM_STOP_STATE_CLK | DSIM_TX_READY_HS_CLK)));
 
-	reg = DSI_READ(dsi, DSIM_ESCMODE_REG);
+	reg = exynos_dsi_read(dsi, DSIM_ESCMODE_REG);
 	reg &= ~DSIM_STOP_STATE_CNT_MASK;
 	reg |= DSIM_STOP_STATE_CNT(driver_data->reg_values[STOP_STATE_CNT]);
-	DSI_WRITE(dsi, DSIM_ESCMODE_REG, reg);
+	exynos_dsi_write(dsi, DSIM_ESCMODE_REG, reg);
 
 	reg = DSIM_BTA_TIMEOUT(0xff) | DSIM_LPDR_TIMEOUT(0xffff);
-	DSI_WRITE(dsi, DSIM_TIMEOUT_REG, reg);
+	exynos_dsi_write(dsi, DSIM_TIMEOUT_REG, reg);
 
 	return 0;
 }
@@ -876,20 +912,20 @@ static void exynos_dsi_set_display_mode(struct exynos_dsi *dsi)
 		reg = DSIM_CMD_ALLOW(0xf)
 			| DSIM_STABLE_VFP(vm->vfront_porch)
 			| DSIM_MAIN_VBP(vm->vback_porch);
-		DSI_WRITE(dsi, DSIM_MVPORCH_REG, reg);
+		exynos_dsi_write(dsi, DSIM_MVPORCH_REG, reg);
 
 		reg = DSIM_MAIN_HFP(vm->hfront_porch)
 			| DSIM_MAIN_HBP(vm->hback_porch);
-		DSI_WRITE(dsi, DSIM_MHPORCH_REG, reg);
+		exynos_dsi_write(dsi, DSIM_MHPORCH_REG, reg);
 
 		reg = DSIM_MAIN_VSA(vm->vsync_len)
 			| DSIM_MAIN_HSA(vm->hsync_len);
-		DSI_WRITE(dsi, DSIM_MSYNC_REG, reg);
+		exynos_dsi_write(dsi, DSIM_MSYNC_REG, reg);
 	}
 	reg =  DSIM_MAIN_HRESOL(vm->hactive, num_bits_resol) |
 		DSIM_MAIN_VRESOL(vm->vactive, num_bits_resol);
 
-	DSI_WRITE(dsi, DSIM_MDRESOL_REG, reg);
+	exynos_dsi_write(dsi, DSIM_MDRESOL_REG, reg);
 
 	dev_dbg(dsi->dev, "LCD size = %dx%d\n", vm->hactive, vm->vactive);
 }
@@ -898,12 +934,12 @@ static void exynos_dsi_set_display_enable(struct exynos_dsi *dsi, bool enable)
 {
 	u32 reg;
 
-	reg = DSI_READ(dsi, DSIM_MDRESOL_REG);
+	reg = exynos_dsi_read(dsi, DSIM_MDRESOL_REG);
 	if (enable)
 		reg |= DSIM_MAIN_STAND_BY;
 	else
 		reg &= ~DSIM_MAIN_STAND_BY;
-	DSI_WRITE(dsi, DSIM_MDRESOL_REG, reg);
+	exynos_dsi_write(dsi, DSIM_MDRESOL_REG, reg);
 }
 
 static int exynos_dsi_wait_for_hdr_fifo(struct exynos_dsi *dsi)
@@ -911,7 +947,7 @@ static int exynos_dsi_wait_for_hdr_fifo(struct exynos_dsi *dsi)
 	int timeout = 2000;
 
 	do {
-		u32 reg = DSI_READ(dsi, DSIM_FIFOCTRL_REG);
+		u32 reg = exynos_dsi_read(dsi, DSIM_FIFOCTRL_REG);
 
 		if (!(reg & DSIM_SFR_HEADER_FULL))
 			return 0;
@@ -925,34 +961,35 @@ static int exynos_dsi_wait_for_hdr_fifo(struct exynos_dsi *dsi)
 
 static void exynos_dsi_set_cmd_lpm(struct exynos_dsi *dsi, bool lpm)
 {
-	u32 v = DSI_READ(dsi, DSIM_ESCMODE_REG);
+	u32 v = exynos_dsi_read(dsi, DSIM_ESCMODE_REG);
 
 	if (lpm)
 		v |= DSIM_CMD_LPDT_LP;
 	else
 		v &= ~DSIM_CMD_LPDT_LP;
 
-	DSI_WRITE(dsi, DSIM_ESCMODE_REG, v);
+	exynos_dsi_write(dsi, DSIM_ESCMODE_REG, v);
 }
 
 static void exynos_dsi_force_bta(struct exynos_dsi *dsi)
 {
-	u32 v = DSI_READ(dsi, DSIM_ESCMODE_REG);
+	u32 v = exynos_dsi_read(dsi, DSIM_ESCMODE_REG);
 	v |= DSIM_FORCE_BTA;
-	DSI_WRITE(dsi, DSIM_ESCMODE_REG, v);
+	exynos_dsi_write(dsi, DSIM_ESCMODE_REG, v);
 }
 
 static void exynos_dsi_send_to_fifo(struct exynos_dsi *dsi,
 					struct exynos_dsi_transfer *xfer)
 {
 	struct device *dev = dsi->dev;
-	const u8 *payload = xfer->tx_payload + xfer->tx_done;
-	u16 length = xfer->tx_len - xfer->tx_done;
+	struct mipi_dsi_packet *pkt = &xfer->packet;
+	const u8 *payload = pkt->payload + xfer->tx_done;
+	u16 length = pkt->payload_length - xfer->tx_done;
 	bool first = !xfer->tx_done;
 	u32 reg;
 
 	dev_dbg(dev, "< xfer %p: tx len %u, done %u, rx len %u, done %u\n",
-		xfer, xfer->tx_len, xfer->tx_done, xfer->rx_len, xfer->rx_done);
+		xfer, length, xfer->tx_done, xfer->rx_len, xfer->rx_done);
 
 	if (length > DSI_TX_FIFO_SIZE)
 		length = DSI_TX_FIFO_SIZE;
@@ -961,9 +998,8 @@ static void exynos_dsi_send_to_fifo(struct exynos_dsi *dsi,
 
 	/* Send payload */
 	while (length >= 4) {
-		reg = (payload[3] << 24) | (payload[2] << 16)
-					| (payload[1] << 8) | payload[0];
-		DSI_WRITE(dsi, DSIM_PAYLOAD_REG, reg);
+		reg = get_unaligned_le32(payload);
+		exynos_dsi_write(dsi, DSIM_PAYLOAD_REG, reg);
 		payload += 4;
 		length -= 4;
 	}
@@ -978,10 +1014,7 @@ static void exynos_dsi_send_to_fifo(struct exynos_dsi *dsi,
 		/* Fall through */
 	case 1:
 		reg |= payload[0];
-		DSI_WRITE(dsi, DSIM_PAYLOAD_REG, reg);
-		break;
-	case 0:
-		/* Do nothing */
+		exynos_dsi_write(dsi, DSIM_PAYLOAD_REG, reg);
 		break;
 	}
 
@@ -989,7 +1022,7 @@ static void exynos_dsi_send_to_fifo(struct exynos_dsi *dsi,
 	if (!first)
 		return;
 
-	reg = (xfer->data[1] << 16) | (xfer->data[0] << 8) | xfer->data_id;
+	reg = get_unaligned_le32(pkt->header);
 	if (exynos_dsi_wait_for_hdr_fifo(dsi)) {
 		dev_err(dev, "waiting for header FIFO timed out\n");
 		return;
@@ -1001,7 +1034,7 @@ static void exynos_dsi_send_to_fifo(struct exynos_dsi *dsi,
 		dsi->state ^= DSIM_STATE_CMD_LPM;
 	}
 
-	DSI_WRITE(dsi, DSIM_PKTHDR_REG, reg);
+	exynos_dsi_write(dsi, DSIM_PKTHDR_REG, reg);
 
 	if (xfer->flags & MIPI_DSI_MSG_REQ_ACK)
 		exynos_dsi_force_bta(dsi);
@@ -1017,7 +1050,7 @@ static void exynos_dsi_read_from_fifo(struct exynos_dsi *dsi,
 	u32 reg;
 
 	if (first) {
-		reg = DSI_READ(dsi, DSIM_RXFIFO_REG);
+		reg = exynos_dsi_read(dsi, DSIM_RXFIFO_REG);
 
 		switch (reg & 0x3f) {
 		case MIPI_DSI_RX_GENERIC_SHORT_READ_RESPONSE_2BYTE:
@@ -1056,7 +1089,7 @@ static void exynos_dsi_read_from_fifo(struct exynos_dsi *dsi,
 
 	/* Receive payload */
 	while (length >= 4) {
-		reg = DSI_READ(dsi, DSIM_RXFIFO_REG);
+		reg = exynos_dsi_read(dsi, DSIM_RXFIFO_REG);
 		payload[0] = (reg >>  0) & 0xff;
 		payload[1] = (reg >>  8) & 0xff;
 		payload[2] = (reg >> 16) & 0xff;
@@ -1066,7 +1099,7 @@ static void exynos_dsi_read_from_fifo(struct exynos_dsi *dsi,
 	}
 
 	if (length) {
-		reg = DSI_READ(dsi, DSIM_RXFIFO_REG);
+		reg = exynos_dsi_read(dsi, DSIM_RXFIFO_REG);
 		switch (length) {
 		case 3:
 			payload[2] = (reg >> 16) & 0xff;
@@ -1085,7 +1118,7 @@ static void exynos_dsi_read_from_fifo(struct exynos_dsi *dsi,
 clear_fifo:
 	length = DSI_RX_FIFO_SIZE / 4;
 	do {
-		reg = DSI_READ(dsi, DSIM_RXFIFO_REG);
+		reg = exynos_dsi_read(dsi, DSIM_RXFIFO_REG);
 		if (reg == DSI_RX_FIFO_EMPTY)
 			break;
 	} while (--length);
@@ -1110,13 +1143,14 @@ again:
 
 	spin_unlock_irqrestore(&dsi->transfer_lock, flags);
 
-	if (xfer->tx_len && xfer->tx_done == xfer->tx_len)
+	if (xfer->packet.payload_length &&
+	    xfer->tx_done == xfer->packet.payload_length)
 		/* waiting for RX */
 		return;
 
 	exynos_dsi_send_to_fifo(dsi, xfer);
 
-	if (xfer->tx_len || xfer->rx_len)
+	if (xfer->packet.payload_length || xfer->rx_len)
 		return;
 
 	xfer->result = 0;
@@ -1152,10 +1186,11 @@ static bool exynos_dsi_transfer_finish(struct exynos_dsi *dsi)
 	spin_unlock_irqrestore(&dsi->transfer_lock, flags);
 
 	dev_dbg(dsi->dev,
-		"> xfer %p, tx_len %u, tx_done %u, rx_len %u, rx_done %u\n",
-		xfer, xfer->tx_len, xfer->tx_done, xfer->rx_len, xfer->rx_done);
+		"> xfer %p, tx_len %zu, tx_done %u, rx_len %u, rx_done %u\n",
+		xfer, xfer->packet.payload_length, xfer->tx_done, xfer->rx_len,
+		xfer->rx_done);
 
-	if (xfer->tx_done != xfer->tx_len)
+	if (xfer->tx_done != xfer->packet.payload_length)
 		return true;
 
 	if (xfer->rx_done != xfer->rx_len)
@@ -1226,9 +1261,10 @@ static int exynos_dsi_transfer(struct exynos_dsi *dsi,
 	wait_for_completion_timeout(&xfer->completed,
 				    msecs_to_jiffies(DSI_XFER_TIMEOUT_MS));
 	if (xfer->result == -ETIMEDOUT) {
+		struct mipi_dsi_packet *pkt = &xfer->packet;
 		exynos_dsi_remove_transfer(dsi, xfer);
-		dev_err(dsi->dev, "xfer timed out: %*ph %*ph\n", 2, xfer->data,
-			xfer->tx_len, xfer->tx_payload);
+		dev_err(dsi->dev, "xfer timed out: %*ph %*ph\n", 4, pkt->header,
+			(int)pkt->payload_length, pkt->payload);
 		return -ETIMEDOUT;
 	}
 
@@ -1241,20 +1277,20 @@ static irqreturn_t exynos_dsi_irq(int irq, void *dev_id)
 	struct exynos_dsi *dsi = dev_id;
 	u32 status;
 
-	status = DSI_READ(dsi, DSIM_INTSRC_REG);
+	status = exynos_dsi_read(dsi, DSIM_INTSRC_REG);
 	if (!status) {
 		static unsigned long int j;
 		if (printk_timed_ratelimit(&j, 500))
 			dev_warn(dsi->dev, "spurious interrupt\n");
 		return IRQ_HANDLED;
 	}
-	DSI_WRITE(dsi, DSIM_INTSRC_REG, status);
+	exynos_dsi_write(dsi, DSIM_INTSRC_REG, status);
 
 	if (status & DSIM_INT_SW_RST_RELEASE) {
 		u32 mask = ~(DSIM_INT_RX_DONE | DSIM_INT_SFR_FIFO_EMPTY |
 			DSIM_INT_SFR_HDR_FIFO_EMPTY | DSIM_INT_FRAME_DONE |
 			DSIM_INT_RX_ECC_ERR | DSIM_INT_SW_RST_RELEASE);
-		DSI_WRITE(dsi, DSIM_INTMSK_REG, mask);
+		exynos_dsi_write(dsi, DSIM_INTMSK_REG, mask);
 		complete(&dsi->completed);
 		return IRQ_HANDLED;
 	}
@@ -1401,12 +1437,6 @@ static int exynos_dsi_host_detach(struct mipi_dsi_host *host,
 	return 0;
 }
 
-/* distinguish between short and long DSI packet types */
-static bool exynos_dsi_is_short_dsi_type(u8 type)
-{
-	return (type & 0x0f) <= 8;
-}
-
 static ssize_t exynos_dsi_host_transfer(struct mipi_dsi_host *host,
 				        const struct mipi_dsi_msg *msg)
 {
@@ -1424,25 +1454,9 @@ static ssize_t exynos_dsi_host_transfer(struct mipi_dsi_host *host,
 		dsi->state |= DSIM_STATE_INITIALIZED;
 	}
 
-	if (msg->tx_len == 0)
-		return -EINVAL;
-
-	xfer.data_id = msg->type | (msg->channel << 6);
-
-	if (exynos_dsi_is_short_dsi_type(msg->type)) {
-		const char *tx_buf = msg->tx_buf;
-
-		if (msg->tx_len > 2)
-			return -EINVAL;
-		xfer.tx_len = 0;
-		xfer.data[0] = tx_buf[0];
-		xfer.data[1] = (msg->tx_len == 2) ? tx_buf[1] : 0;
-	} else {
-		xfer.tx_len = msg->tx_len;
-		xfer.data[0] = msg->tx_len & 0xff;
-		xfer.data[1] = msg->tx_len >> 8;
-		xfer.tx_payload = msg->tx_buf;
-	}
+	ret = mipi_dsi_create_packet(&xfer.packet, msg);
+	if (ret < 0)
+		return ret;
 
 	xfer.rx_len = msg->rx_len;
 	xfer.rx_payload = msg->rx_buf;
@@ -1774,6 +1788,7 @@ static int exynos_dsi_bind(struct device *dev, struct device *master,
 
 	bridge = of_drm_find_bridge(dsi->bridge_node);
 	if (bridge) {
+		encoder->bridge = bridge;
 		drm_bridge_attach(drm_dev, bridge);
 	}
 
diff --git a/drivers/gpu/drm/exynos/exynos_drm_fbdev.c b/drivers/gpu/drm/exynos/exynos_drm_fbdev.c
index f6118baa8e3e..4ae860c44f1d 100644
--- a/drivers/gpu/drm/exynos/exynos_drm_fbdev.c
+++ b/drivers/gpu/drm/exynos/exynos_drm_fbdev.c
@@ -50,7 +50,7 @@ static int exynos_drm_fb_mmap(struct fb_info *info,
 	if (vm_size > exynos_gem->size)
 		return -EINVAL;
 
-	ret = dma_mmap_attrs(helper->dev->dev, vma, exynos_gem->pages,
+	ret = dma_mmap_attrs(to_dma_dev(helper->dev), vma, exynos_gem->cookie,
 			     exynos_gem->dma_addr, exynos_gem->size,
 			     &exynos_gem->dma_attrs);
 	if (ret < 0) {
diff --git a/drivers/gpu/drm/exynos/exynos_drm_fimc.c b/drivers/gpu/drm/exynos/exynos_drm_fimc.c
index c747824f3c98..0525c56145db 100644
--- a/drivers/gpu/drm/exynos/exynos_drm_fimc.c
+++ b/drivers/gpu/drm/exynos/exynos_drm_fimc.c
@@ -163,7 +163,6 @@ struct fimc_context {
 	u32		clk_frequency;
 	struct regmap	*sysreg;
 	struct fimc_scaler	sc;
-	struct exynos_drm_ipp_pol	pol;
 	int	id;
 	int	irq;
 	bool	suspended;
@@ -260,32 +259,6 @@ static void fimc_set_type_ctrl(struct fimc_context *ctx, enum fimc_wb wb)
 	fimc_write(ctx, cfg, EXYNOS_CIGCTRL);
 }
 
-static void fimc_set_polarity(struct fimc_context *ctx,
-		struct exynos_drm_ipp_pol *pol)
-{
-	u32 cfg;
-
-	DRM_DEBUG_KMS("inv_pclk[%d]inv_vsync[%d]\n",
-		pol->inv_pclk, pol->inv_vsync);
-	DRM_DEBUG_KMS("inv_href[%d]inv_hsync[%d]\n",
-		pol->inv_href, pol->inv_hsync);
-
-	cfg = fimc_read(ctx, EXYNOS_CIGCTRL);
-	cfg &= ~(EXYNOS_CIGCTRL_INVPOLPCLK | EXYNOS_CIGCTRL_INVPOLVSYNC |
-		 EXYNOS_CIGCTRL_INVPOLHREF | EXYNOS_CIGCTRL_INVPOLHSYNC);
-
-	if (pol->inv_pclk)
-		cfg |= EXYNOS_CIGCTRL_INVPOLPCLK;
-	if (pol->inv_vsync)
-		cfg |= EXYNOS_CIGCTRL_INVPOLVSYNC;
-	if (pol->inv_href)
-		cfg |= EXYNOS_CIGCTRL_INVPOLHREF;
-	if (pol->inv_hsync)
-		cfg |= EXYNOS_CIGCTRL_INVPOLHSYNC;
-
-	fimc_write(ctx, cfg, EXYNOS_CIGCTRL);
-}
-
 static void fimc_handle_jpeg(struct fimc_context *ctx, bool enable)
 {
 	u32 cfg;
@@ -1467,7 +1440,6 @@ static int fimc_ippdrv_start(struct device *dev, enum drm_exynos_ipp_cmd cmd)
 	/* If set ture, we can save jpeg about screen */
 	fimc_handle_jpeg(ctx, false);
 	fimc_set_scaler(ctx, &ctx->sc);
-	fimc_set_polarity(ctx, &ctx->pol);
 
 	switch (cmd) {
 	case IPP_CMD_M2M:
@@ -1723,7 +1695,7 @@ static int fimc_probe(struct platform_device *pdev)
 		goto err_put_clk;
 	}
 
-	DRM_DEBUG_KMS("id[%d]ippdrv[0x%x]\n", ctx->id, (int)ippdrv);
+	DRM_DEBUG_KMS("id[%d]ippdrv[%p]\n", ctx->id, ippdrv);
 
 	spin_lock_init(&ctx->lock);
 	platform_set_drvdata(pdev, ctx);
diff --git a/drivers/gpu/drm/exynos/exynos_drm_fimd.c b/drivers/gpu/drm/exynos/exynos_drm_fimd.c
index 70194d0e4fe4..51d484ae9f49 100644
--- a/drivers/gpu/drm/exynos/exynos_drm_fimd.c
+++ b/drivers/gpu/drm/exynos/exynos_drm_fimd.c
@@ -94,12 +94,14 @@ struct fimd_driver_data {
 	unsigned int lcdblk_offset;
 	unsigned int lcdblk_vt_shift;
 	unsigned int lcdblk_bypass_shift;
+	unsigned int lcdblk_mic_bypass_shift;
 
 	unsigned int has_shadowcon:1;
 	unsigned int has_clksel:1;
 	unsigned int has_limited_fmt:1;
 	unsigned int has_vidoutcon:1;
 	unsigned int has_vtsel:1;
+	unsigned int has_mic_bypass:1;
 };
 
 static struct fimd_driver_data s3c64xx_fimd_driver_data = {
@@ -145,6 +147,18 @@ static struct fimd_driver_data exynos5_fimd_driver_data = {
 	.has_vtsel = 1,
 };
 
+static struct fimd_driver_data exynos5420_fimd_driver_data = {
+	.timing_base = 0x20000,
+	.lcdblk_offset = 0x214,
+	.lcdblk_vt_shift = 24,
+	.lcdblk_bypass_shift = 15,
+	.lcdblk_mic_bypass_shift = 11,
+	.has_shadowcon = 1,
+	.has_vidoutcon = 1,
+	.has_vtsel = 1,
+	.has_mic_bypass = 1,
+};
+
 struct fimd_context {
 	struct device			*dev;
 	struct drm_device		*drm_dev;
@@ -168,7 +182,6 @@ struct fimd_context {
 	atomic_t			win_updated;
 	atomic_t			triggering;
 
-	struct exynos_drm_panel_info panel;
 	struct fimd_driver_data *driver_data;
 	struct drm_encoder *encoder;
 };
@@ -184,6 +197,8 @@ static const struct of_device_id fimd_driver_dt_match[] = {
 	  .data = &exynos4415_fimd_driver_data },
 	{ .compatible = "samsung,exynos5250-fimd",
 	  .data = &exynos5_fimd_driver_data },
+	{ .compatible = "samsung,exynos5420-fimd",
+	  .data = &exynos5420_fimd_driver_data },
 	{},
 };
 MODULE_DEVICE_TABLE(of, fimd_driver_dt_match);
@@ -380,7 +395,7 @@ static u32 fimd_calc_clkdiv(struct fimd_context *ctx,
 	}
 
 	/* Find the clock divider value that gets us closest to ideal_clk */
-	clkdiv = DIV_ROUND_UP(clk_get_rate(ctx->lcd_clk), ideal_clk);
+	clkdiv = DIV_ROUND_CLOSEST(clk_get_rate(ctx->lcd_clk), ideal_clk);
 
 	return (clkdiv < 0x100) ? clkdiv : 0xff;
 }
@@ -461,6 +476,18 @@ static void fimd_commit(struct exynos_drm_crtc *crtc)
 		return;
 	}
 
+	/* TODO: When MIC is enabled for display path, the lcdblk_mic_bypass
+	 * bit should be cleared.
+	 */
+	if (driver_data->has_mic_bypass && ctx->sysreg &&
+	    regmap_update_bits(ctx->sysreg,
+				driver_data->lcdblk_offset,
+				0x1 << driver_data->lcdblk_mic_bypass_shift,
+				0x1 << driver_data->lcdblk_mic_bypass_shift)) {
+		DRM_ERROR("Failed to update sysreg for bypass mic.\n");
+		return;
+	}
+
 	/* setup horizontal and vertical display size. */
 	val = VIDTCON2_LINEVAL(mode->vdisplay - 1) |
 	       VIDTCON2_HOZVAL(mode->hdisplay - 1) |
@@ -861,7 +888,8 @@ static void fimd_dp_clock_enable(struct exynos_drm_crtc *crtc, bool enable)
 	 * clock. On these SoCs the bootloader may enable it but any
 	 * power domain off/on will reset it to disable state.
 	 */
-	if (ctx->driver_data != &exynos5_fimd_driver_data)
+	if (ctx->driver_data != &exynos5_fimd_driver_data ||
+	    ctx->driver_data != &exynos5420_fimd_driver_data)
 		return;
 
 	val = enable ? DP_MIE_CLK_DP_ENABLE : DP_MIE_CLK_DISABLE;
diff --git a/drivers/gpu/drm/exynos/exynos_drm_g2d.c b/drivers/gpu/drm/exynos/exynos_drm_g2d.c
index 99369816ff97..193d3602dffb 100644
--- a/drivers/gpu/drm/exynos/exynos_drm_g2d.c
+++ b/drivers/gpu/drm/exynos/exynos_drm_g2d.c
@@ -259,7 +259,7 @@ static int g2d_init_cmdlist(struct g2d_data *g2d)
 	init_dma_attrs(&g2d->cmdlist_dma_attrs);
 	dma_set_attr(DMA_ATTR_WRITE_COMBINE, &g2d->cmdlist_dma_attrs);
 
-	g2d->cmdlist_pool_virt = dma_alloc_attrs(subdrv->drm_dev->dev,
+	g2d->cmdlist_pool_virt = dma_alloc_attrs(to_dma_dev(subdrv->drm_dev),
 						G2D_CMDLIST_POOL_SIZE,
 						&g2d->cmdlist_pool, GFP_KERNEL,
 						&g2d->cmdlist_dma_attrs);
@@ -293,7 +293,7 @@ static int g2d_init_cmdlist(struct g2d_data *g2d)
 	return 0;
 
 err:
-	dma_free_attrs(subdrv->drm_dev->dev, G2D_CMDLIST_POOL_SIZE,
+	dma_free_attrs(to_dma_dev(subdrv->drm_dev), G2D_CMDLIST_POOL_SIZE,
 			g2d->cmdlist_pool_virt,
 			g2d->cmdlist_pool, &g2d->cmdlist_dma_attrs);
 	return ret;
@@ -306,7 +306,8 @@ static void g2d_fini_cmdlist(struct g2d_data *g2d)
 	kfree(g2d->cmdlist_node);
 
 	if (g2d->cmdlist_pool_virt && g2d->cmdlist_pool) {
-		dma_free_attrs(subdrv->drm_dev->dev, G2D_CMDLIST_POOL_SIZE,
+		dma_free_attrs(to_dma_dev(subdrv->drm_dev),
+				G2D_CMDLIST_POOL_SIZE,
 				g2d->cmdlist_pool_virt,
 				g2d->cmdlist_pool, &g2d->cmdlist_dma_attrs);
 	}
@@ -1151,7 +1152,7 @@ int exynos_g2d_set_cmdlist_ioctl(struct drm_device *drm_dev, void *data,
 		goto err_free_event;
 	}
 
-	cmd = (struct drm_exynos_g2d_cmd *)(uint32_t)req->cmd;
+	cmd = (struct drm_exynos_g2d_cmd *)(unsigned long)req->cmd;
 
 	if (copy_from_user(cmdlist->data + cmdlist->last,
 				(void __user *)cmd,
@@ -1169,7 +1170,8 @@ int exynos_g2d_set_cmdlist_ioctl(struct drm_device *drm_dev, void *data,
 	if (req->cmd_buf_nr) {
 		struct drm_exynos_g2d_cmd *cmd_buf;
 
-		cmd_buf = (struct drm_exynos_g2d_cmd *)(uint32_t)req->cmd_buf;
+		cmd_buf = (struct drm_exynos_g2d_cmd *)
+				(unsigned long)req->cmd_buf;
 
 		if (copy_from_user(cmdlist->data + cmdlist->last,
 					(void __user *)cmd_buf,
diff --git a/drivers/gpu/drm/exynos/exynos_drm_gem.c b/drivers/gpu/drm/exynos/exynos_drm_gem.c
index 32358c5e3db4..2914d62d0d80 100644
--- a/drivers/gpu/drm/exynos/exynos_drm_gem.c
+++ b/drivers/gpu/drm/exynos/exynos_drm_gem.c
@@ -65,7 +65,7 @@ static int exynos_drm_alloc_buf(struct exynos_drm_gem *exynos_gem)
 		return -ENOMEM;
 	}
 
-	exynos_gem->cookie = dma_alloc_attrs(dev->dev, exynos_gem->size,
+	exynos_gem->cookie = dma_alloc_attrs(to_dma_dev(dev), exynos_gem->size,
 					     &exynos_gem->dma_addr, GFP_KERNEL,
 					     &exynos_gem->dma_attrs);
 	if (!exynos_gem->cookie) {
@@ -73,7 +73,7 @@ static int exynos_drm_alloc_buf(struct exynos_drm_gem *exynos_gem)
 		goto err_free;
 	}
 
-	ret = dma_get_sgtable_attrs(dev->dev, &sgt, exynos_gem->cookie,
+	ret = dma_get_sgtable_attrs(to_dma_dev(dev), &sgt, exynos_gem->cookie,
 				    exynos_gem->dma_addr, exynos_gem->size,
 				    &exynos_gem->dma_attrs);
 	if (ret < 0) {
@@ -98,7 +98,7 @@ static int exynos_drm_alloc_buf(struct exynos_drm_gem *exynos_gem)
 err_sgt_free:
 	sg_free_table(&sgt);
 err_dma_free:
-	dma_free_attrs(dev->dev, exynos_gem->size, exynos_gem->cookie,
+	dma_free_attrs(to_dma_dev(dev), exynos_gem->size, exynos_gem->cookie,
 		       exynos_gem->dma_addr, &exynos_gem->dma_attrs);
 err_free:
 	drm_free_large(exynos_gem->pages);
@@ -118,7 +118,7 @@ static void exynos_drm_free_buf(struct exynos_drm_gem *exynos_gem)
 	DRM_DEBUG_KMS("dma_addr(0x%lx), size(0x%lx)\n",
 			(unsigned long)exynos_gem->dma_addr, exynos_gem->size);
 
-	dma_free_attrs(dev->dev, exynos_gem->size, exynos_gem->cookie,
+	dma_free_attrs(to_dma_dev(dev), exynos_gem->size, exynos_gem->cookie,
 			(dma_addr_t)exynos_gem->dma_addr,
 			&exynos_gem->dma_attrs);
 
@@ -218,7 +218,7 @@ static struct exynos_drm_gem *exynos_drm_gem_init(struct drm_device *dev,
 		return ERR_PTR(ret);
 	}
 
-	DRM_DEBUG_KMS("created file object = 0x%x\n", (unsigned int)obj->filp);
+	DRM_DEBUG_KMS("created file object = %p\n", obj->filp);
 
 	return exynos_gem;
 }
@@ -280,6 +280,15 @@ int exynos_drm_gem_create_ioctl(struct drm_device *dev, void *data,
 	return 0;
 }
 
+int exynos_drm_gem_map_ioctl(struct drm_device *dev, void *data,
+			     struct drm_file *file_priv)
+{
+	struct drm_exynos_gem_map *args = data;
+
+	return exynos_drm_gem_dumb_map_offset(file_priv, dev, args->handle,
+					      &args->offset);
+}
+
 dma_addr_t *exynos_drm_gem_get_dma_addr(struct drm_device *dev,
 					unsigned int gem_handle,
 					struct drm_file *filp)
@@ -335,7 +344,7 @@ static int exynos_drm_gem_mmap_buffer(struct exynos_drm_gem *exynos_gem,
 	if (vm_size > exynos_gem->size)
 		return -EINVAL;
 
-	ret = dma_mmap_attrs(drm_dev->dev, vma, exynos_gem->pages,
+	ret = dma_mmap_attrs(to_dma_dev(drm_dev), vma, exynos_gem->cookie,
 			     exynos_gem->dma_addr, exynos_gem->size,
 			     &exynos_gem->dma_attrs);
 	if (ret < 0) {
@@ -381,7 +390,7 @@ int exynos_gem_map_sgt_with_dma(struct drm_device *drm_dev,
 
 	mutex_lock(&drm_dev->struct_mutex);
 
-	nents = dma_map_sg(drm_dev->dev, sgt->sgl, sgt->nents, dir);
+	nents = dma_map_sg(to_dma_dev(drm_dev), sgt->sgl, sgt->nents, dir);
 	if (!nents) {
 		DRM_ERROR("failed to map sgl with dma.\n");
 		mutex_unlock(&drm_dev->struct_mutex);
@@ -396,7 +405,7 @@ void exynos_gem_unmap_sgt_from_dma(struct drm_device *drm_dev,
 				struct sg_table *sgt,
 				enum dma_data_direction dir)
 {
-	dma_unmap_sg(drm_dev->dev, sgt->sgl, sgt->nents, dir);
+	dma_unmap_sg(to_dma_dev(drm_dev), sgt->sgl, sgt->nents, dir);
 }
 
 void exynos_drm_gem_free_object(struct drm_gem_object *obj)
diff --git a/drivers/gpu/drm/exynos/exynos_drm_gem.h b/drivers/gpu/drm/exynos/exynos_drm_gem.h
index 9ca5047959ec..00223052b87b 100644
--- a/drivers/gpu/drm/exynos/exynos_drm_gem.h
+++ b/drivers/gpu/drm/exynos/exynos_drm_gem.h
@@ -71,6 +71,10 @@ struct exynos_drm_gem *exynos_drm_gem_create(struct drm_device *dev,
 int exynos_drm_gem_create_ioctl(struct drm_device *dev, void *data,
 				struct drm_file *file_priv);
 
+/* get fake-offset of gem object that can be used with mmap. */
+int exynos_drm_gem_map_ioctl(struct drm_device *dev, void *data,
+			     struct drm_file *file_priv);
+
 /*
  * get dma address from gem handle and this function could be used for
  * other drivers such as 2d/3d acceleration drivers.
diff --git a/drivers/gpu/drm/exynos/exynos_drm_gsc.c b/drivers/gpu/drm/exynos/exynos_drm_gsc.c
index 7aecd23cfa11..5d20da8f957e 100644
--- a/drivers/gpu/drm/exynos/exynos_drm_gsc.c
+++ b/drivers/gpu/drm/exynos/exynos_drm_gsc.c
@@ -1723,7 +1723,7 @@ static int gsc_probe(struct platform_device *pdev)
 		return ret;
 	}
 
-	DRM_DEBUG_KMS("id[%d]ippdrv[0x%x]\n", ctx->id, (int)ippdrv);
+	DRM_DEBUG_KMS("id[%d]ippdrv[%p]\n", ctx->id, ippdrv);
 
 	mutex_init(&ctx->lock);
 	platform_set_drvdata(pdev, ctx);
diff --git a/drivers/gpu/drm/exynos/exynos_drm_iommu.c b/drivers/gpu/drm/exynos/exynos_drm_iommu.c
index d73b9ad35b7a..7ca09ee19656 100644
--- a/drivers/gpu/drm/exynos/exynos_drm_iommu.c
+++ b/drivers/gpu/drm/exynos/exynos_drm_iommu.c
@@ -9,7 +9,7 @@
  * option) any later version.
  */
 
-#include <drmP.h>
+#include <drm/drmP.h>
 #include <drm/exynos_drm.h>
 
 #include <linux/dma-mapping.h>
@@ -30,7 +30,6 @@ int drm_create_iommu_mapping(struct drm_device *drm_dev)
 {
 	struct dma_iommu_mapping *mapping = NULL;
 	struct exynos_drm_private *priv = drm_dev->dev_private;
-	struct device *dev = drm_dev->dev;
 
 	if (!priv->da_start)
 		priv->da_start = EXYNOS_DEV_ADDR_START;
@@ -43,18 +42,9 @@ int drm_create_iommu_mapping(struct drm_device *drm_dev)
 	if (IS_ERR(mapping))
 		return PTR_ERR(mapping);
 
-	dev->dma_parms = devm_kzalloc(dev, sizeof(*dev->dma_parms),
-					GFP_KERNEL);
-	if (!dev->dma_parms)
-		goto error;
-
-	dma_set_max_seg_size(dev, 0xffffffffu);
-	dev->archdata.mapping = mapping;
+	priv->mapping = mapping;
 
 	return 0;
-error:
-	arm_iommu_release_mapping(mapping);
-	return -ENOMEM;
 }
 
 /*
@@ -67,9 +57,9 @@ error:
  */
 void drm_release_iommu_mapping(struct drm_device *drm_dev)
 {
-	struct device *dev = drm_dev->dev;
+	struct exynos_drm_private *priv = drm_dev->dev_private;
 
-	arm_iommu_release_mapping(dev->archdata.mapping);
+	arm_iommu_release_mapping(priv->mapping);
 }
 
 /*
@@ -84,10 +74,10 @@ void drm_release_iommu_mapping(struct drm_device *drm_dev)
 int drm_iommu_attach_device(struct drm_device *drm_dev,
 				struct device *subdrv_dev)
 {
-	struct device *dev = drm_dev->dev;
+	struct exynos_drm_private *priv = drm_dev->dev_private;
 	int ret;
 
-	if (!dev->archdata.mapping)
+	if (!priv->mapping)
 		return 0;
 
 	subdrv_dev->dma_parms = devm_kzalloc(subdrv_dev,
@@ -101,23 +91,12 @@ int drm_iommu_attach_device(struct drm_device *drm_dev,
 	if (subdrv_dev->archdata.mapping)
 		arm_iommu_detach_device(subdrv_dev);
 
-	ret = arm_iommu_attach_device(subdrv_dev, dev->archdata.mapping);
+	ret = arm_iommu_attach_device(subdrv_dev, priv->mapping);
 	if (ret < 0) {
 		DRM_DEBUG_KMS("failed iommu attach.\n");
 		return ret;
 	}
 
-	/*
-	 * Set dma_ops to drm_device just one time.
-	 *
-	 * The dma mapping api needs device object and the api is used
-	 * to allocate physial memory and map it with iommu table.
-	 * If iommu attach succeeded, the sub driver would have dma_ops
-	 * for iommu and also all sub drivers have same dma_ops.
-	 */
-	if (get_dma_ops(dev) == get_dma_ops(NULL))
-		set_dma_ops(dev, get_dma_ops(subdrv_dev));
-
 	return 0;
 }
 
@@ -133,8 +112,8 @@ int drm_iommu_attach_device(struct drm_device *drm_dev,
 void drm_iommu_detach_device(struct drm_device *drm_dev,
 				struct device *subdrv_dev)
 {
-	struct device *dev = drm_dev->dev;
-	struct dma_iommu_mapping *mapping = dev->archdata.mapping;
+	struct exynos_drm_private *priv = drm_dev->dev_private;
+	struct dma_iommu_mapping *mapping = priv->mapping;
 
 	if (!mapping || !mapping->domain)
 		return;
diff --git a/drivers/gpu/drm/exynos/exynos_drm_iommu.h b/drivers/gpu/drm/exynos/exynos_drm_iommu.h
index dc1b5441f491..5ffebe02ee4d 100644
--- a/drivers/gpu/drm/exynos/exynos_drm_iommu.h
+++ b/drivers/gpu/drm/exynos/exynos_drm_iommu.h
@@ -29,9 +29,9 @@ void drm_iommu_detach_device(struct drm_device *dev_dev,
 
 static inline bool is_drm_iommu_supported(struct drm_device *drm_dev)
 {
-	struct device *dev = drm_dev->dev;
+	struct exynos_drm_private *priv = drm_dev->dev_private;
 
-	return dev->archdata.mapping ? true : false;
+	return priv->mapping ? true : false;
 }
 
 #else
diff --git a/drivers/gpu/drm/exynos/exynos_drm_ipp.c b/drivers/gpu/drm/exynos/exynos_drm_ipp.c
index 3eab0d15f0b4..9c84ee76f18a 100644
--- a/drivers/gpu/drm/exynos/exynos_drm_ipp.c
+++ b/drivers/gpu/drm/exynos/exynos_drm_ipp.c
@@ -208,7 +208,7 @@ static struct exynos_drm_ippdrv *ipp_find_drv_by_handle(u32 prop_id)
 	 * e.g PAUSE state, queue buf, command control.
 	 */
 	list_for_each_entry(ippdrv, &exynos_drm_ippdrv_list, drv_list) {
-		DRM_DEBUG_KMS("count[%d]ippdrv[0x%x]\n", count++, (int)ippdrv);
+		DRM_DEBUG_KMS("count[%d]ippdrv[%p]\n", count++, ippdrv);
 
 		mutex_lock(&ippdrv->cmd_lock);
 		list_for_each_entry(c_node, &ippdrv->cmd_list, list) {
@@ -388,8 +388,8 @@ int exynos_drm_ipp_set_property(struct drm_device *drm_dev, void *data,
 	}
 	property->prop_id = ret;
 
-	DRM_DEBUG_KMS("created prop_id[%d]cmd[%d]ippdrv[0x%x]\n",
-		property->prop_id, property->cmd, (int)ippdrv);
+	DRM_DEBUG_KMS("created prop_id[%d]cmd[%d]ippdrv[%p]\n",
+		property->prop_id, property->cmd, ippdrv);
 
 	/* stored property information and ippdrv in private data */
 	c_node->property = *property;
@@ -518,7 +518,7 @@ static int ipp_put_mem_node(struct drm_device *drm_dev,
 {
 	int i;
 
-	DRM_DEBUG_KMS("node[0x%x]\n", (int)m_node);
+	DRM_DEBUG_KMS("node[%p]\n", m_node);
 
 	if (!m_node) {
 		DRM_ERROR("invalid dequeue node.\n");
@@ -562,7 +562,7 @@ static struct drm_exynos_ipp_mem_node
 	m_node->buf_id = qbuf->buf_id;
 	INIT_LIST_HEAD(&m_node->list);
 
-	DRM_DEBUG_KMS("m_node[0x%x]ops_id[%d]\n", (int)m_node, qbuf->ops_id);
+	DRM_DEBUG_KMS("m_node[%p]ops_id[%d]\n", m_node, qbuf->ops_id);
 	DRM_DEBUG_KMS("prop_id[%d]buf_id[%d]\n", qbuf->prop_id, m_node->buf_id);
 
 	for_each_ipp_planar(i) {
@@ -582,8 +582,8 @@ static struct drm_exynos_ipp_mem_node
 
 			buf_info->handles[i] = qbuf->handle[i];
 			buf_info->base[i] = *addr;
-			DRM_DEBUG_KMS("i[%d]base[0x%x]hd[0x%lx]\n", i,
-				      buf_info->base[i], buf_info->handles[i]);
+			DRM_DEBUG_KMS("i[%d]base[%pad]hd[0x%lx]\n", i,
+				      &buf_info->base[i], buf_info->handles[i]);
 		}
 	}
 
@@ -659,7 +659,7 @@ static void ipp_put_event(struct drm_exynos_ipp_cmd_node *c_node,
 
 	mutex_lock(&c_node->event_lock);
 	list_for_each_entry_safe(e, te, &c_node->event_list, base.link) {
-		DRM_DEBUG_KMS("count[%d]e[0x%x]\n", count++, (int)e);
+		DRM_DEBUG_KMS("count[%d]e[%p]\n", count++, e);
 
 		/*
 		 * qbuf == NULL condition means all event deletion.
@@ -750,7 +750,7 @@ static struct drm_exynos_ipp_mem_node
 
 	/* find memory node from memory list */
 	list_for_each_entry(m_node, head, list) {
-		DRM_DEBUG_KMS("count[%d]m_node[0x%x]\n", count++, (int)m_node);
+		DRM_DEBUG_KMS("count[%d]m_node[%p]\n", count++, m_node);
 
 		/* compare buffer id */
 		if (m_node->buf_id == qbuf->buf_id)
@@ -767,7 +767,7 @@ static int ipp_set_mem_node(struct exynos_drm_ippdrv *ippdrv,
 	struct exynos_drm_ipp_ops *ops = NULL;
 	int ret = 0;
 
-	DRM_DEBUG_KMS("node[0x%x]\n", (int)m_node);
+	DRM_DEBUG_KMS("node[%p]\n", m_node);
 
 	if (!m_node) {
 		DRM_ERROR("invalid queue node.\n");
@@ -1232,7 +1232,7 @@ static int ipp_start_property(struct exynos_drm_ippdrv *ippdrv,
 			m_node = list_first_entry(head,
 				struct drm_exynos_ipp_mem_node, list);
 
-			DRM_DEBUG_KMS("m_node[0x%x]\n", (int)m_node);
+			DRM_DEBUG_KMS("m_node[%p]\n", m_node);
 
 			ret = ipp_set_mem_node(ippdrv, c_node, m_node);
 			if (ret) {
@@ -1601,8 +1601,8 @@ static int ipp_subdrv_probe(struct drm_device *drm_dev, struct device *dev)
 		}
 		ippdrv->prop_list.ipp_id = ret;
 
-		DRM_DEBUG_KMS("count[%d]ippdrv[0x%x]ipp_id[%d]\n",
-			count++, (int)ippdrv, ret);
+		DRM_DEBUG_KMS("count[%d]ippdrv[%p]ipp_id[%d]\n",
+			count++, ippdrv, ret);
 
 		/* store parent device for node */
 		ippdrv->parent_dev = dev;
@@ -1659,7 +1659,7 @@ static int ipp_subdrv_open(struct drm_device *drm_dev, struct device *dev,
 
 	file_priv->ipp_dev = dev;
 
-	DRM_DEBUG_KMS("done priv[0x%x]\n", (int)dev);
+	DRM_DEBUG_KMS("done priv[%p]\n", dev);
 
 	return 0;
 }
@@ -1676,8 +1676,8 @@ static void ipp_subdrv_close(struct drm_device *drm_dev, struct device *dev,
 		mutex_lock(&ippdrv->cmd_lock);
 		list_for_each_entry_safe(c_node, tc_node,
 			&ippdrv->cmd_list, list) {
-			DRM_DEBUG_KMS("count[%d]ippdrv[0x%x]\n",
-				count++, (int)ippdrv);
+			DRM_DEBUG_KMS("count[%d]ippdrv[%p]\n",
+				count++, ippdrv);
 
 			if (c_node->filp == file) {
 				/*
diff --git a/drivers/gpu/drm/exynos/exynos_drm_mic.c b/drivers/gpu/drm/exynos/exynos_drm_mic.c
index 4eaef36aec5a..9869d70e9e54 100644
--- a/drivers/gpu/drm/exynos/exynos_drm_mic.c
+++ b/drivers/gpu/drm/exynos/exynos_drm_mic.c
@@ -18,6 +18,7 @@
 #include <linux/of.h>
 #include <linux/of_graph.h>
 #include <linux/clk.h>
+#include <linux/component.h>
 #include <drm/drmP.h>
 #include <linux/mfd/syscon.h>
 #include <linux/regmap.h>
@@ -306,9 +307,9 @@ exit:
 	return ret;
 }
 
-void mic_disable(struct drm_bridge *bridge) { }
+static void mic_disable(struct drm_bridge *bridge) { }
 
-void mic_post_disable(struct drm_bridge *bridge)
+static void mic_post_disable(struct drm_bridge *bridge)
 {
 	struct exynos_mic *mic = bridge->driver_private;
 	int i;
@@ -328,7 +329,7 @@ already_disabled:
 	mutex_unlock(&mic_mutex);
 }
 
-void mic_pre_enable(struct drm_bridge *bridge)
+static void mic_pre_enable(struct drm_bridge *bridge)
 {
 	struct exynos_mic *mic = bridge->driver_private;
 	int ret, i;
@@ -371,11 +372,35 @@ already_enabled:
 	mutex_unlock(&mic_mutex);
 }
 
-void mic_enable(struct drm_bridge *bridge) { }
+static void mic_enable(struct drm_bridge *bridge) { }
 
-void mic_destroy(struct drm_bridge *bridge)
+static const struct drm_bridge_funcs mic_bridge_funcs = {
+	.disable = mic_disable,
+	.post_disable = mic_post_disable,
+	.pre_enable = mic_pre_enable,
+	.enable = mic_enable,
+};
+
+static int exynos_mic_bind(struct device *dev, struct device *master,
+			   void *data)
 {
-	struct exynos_mic *mic = bridge->driver_private;
+	struct exynos_mic *mic = dev_get_drvdata(dev);
+	int ret;
+
+	mic->bridge.funcs = &mic_bridge_funcs;
+	mic->bridge.of_node = dev->of_node;
+	mic->bridge.driver_private = mic;
+	ret = drm_bridge_add(&mic->bridge);
+	if (ret)
+		DRM_ERROR("mic: Failed to add MIC to the global bridge list\n");
+
+	return ret;
+}
+
+static void exynos_mic_unbind(struct device *dev, struct device *master,
+			      void *data)
+{
+	struct exynos_mic *mic = dev_get_drvdata(dev);
 	int i;
 
 	mutex_lock(&mic_mutex);
@@ -387,16 +412,16 @@ void mic_destroy(struct drm_bridge *bridge)
 
 already_disabled:
 	mutex_unlock(&mic_mutex);
+
+	drm_bridge_remove(&mic->bridge);
 }
 
-static const struct drm_bridge_funcs mic_bridge_funcs = {
-	.disable = mic_disable,
-	.post_disable = mic_post_disable,
-	.pre_enable = mic_pre_enable,
-	.enable = mic_enable,
+static const struct component_ops exynos_mic_component_ops = {
+	.bind	= exynos_mic_bind,
+	.unbind	= exynos_mic_unbind,
 };
 
-int exynos_mic_probe(struct platform_device *pdev)
+static int exynos_mic_probe(struct platform_device *pdev)
 {
 	struct device *dev = &pdev->dev;
 	struct exynos_mic *mic;
@@ -435,17 +460,8 @@ int exynos_mic_probe(struct platform_device *pdev)
 		goto err;
 	}
 
-	mic->bridge.funcs = &mic_bridge_funcs;
-	mic->bridge.of_node = dev->of_node;
-	mic->bridge.driver_private = mic;
-	ret = drm_bridge_add(&mic->bridge);
-	if (ret) {
-		DRM_ERROR("mic: Failed to add MIC to the global bridge list\n");
-		goto err;
-	}
-
 	for (i = 0; i < NUM_CLKS; i++) {
-		mic->clks[i] = of_clk_get_by_name(dev->of_node, clk_names[i]);
+		mic->clks[i] = devm_clk_get(dev, clk_names[i]);
 		if (IS_ERR(mic->clks[i])) {
 			DRM_ERROR("mic: Failed to get clock (%s)\n",
 								clk_names[i]);
@@ -454,7 +470,10 @@ int exynos_mic_probe(struct platform_device *pdev)
 		}
 	}
 
+	platform_set_drvdata(pdev, mic);
+
 	DRM_DEBUG_KMS("MIC has been probed\n");
+	return component_add(dev, &exynos_mic_component_ops);
 
 err:
 	return ret;
@@ -462,14 +481,7 @@ err:
 
 static int exynos_mic_remove(struct platform_device *pdev)
 {
-	struct exynos_mic *mic = platform_get_drvdata(pdev);
-	int i;
-
-	drm_bridge_remove(&mic->bridge);
-
-	for (i = NUM_CLKS - 1; i > -1; i--)
-		clk_put(mic->clks[i]);
-
+	component_del(&pdev->dev, &exynos_mic_component_ops);
 	return 0;
 }
 
diff --git a/drivers/gpu/drm/exynos/exynos_drm_rotator.c b/drivers/gpu/drm/exynos/exynos_drm_rotator.c
index bea0f7826d30..f18fbe43f55f 100644
--- a/drivers/gpu/drm/exynos/exynos_drm_rotator.c
+++ b/drivers/gpu/drm/exynos/exynos_drm_rotator.c
@@ -20,7 +20,6 @@
 #include <drm/drmP.h>
 #include <drm/exynos_drm.h>
 #include "regs-rotator.h"
-#include "exynos_drm.h"
 #include "exynos_drm_drv.h"
 #include "exynos_drm_ipp.h"
 
@@ -754,7 +753,7 @@ static int rotator_probe(struct platform_device *pdev)
 		goto err_ippdrv_register;
 	}
 
-	DRM_DEBUG_KMS("ippdrv[0x%x]\n", (int)ippdrv);
+	DRM_DEBUG_KMS("ippdrv[%p]\n", ippdrv);
 
 	platform_set_drvdata(pdev, rot);
 
diff --git a/drivers/gpu/drm/exynos/exynos_drm_vidi.c b/drivers/gpu/drm/exynos/exynos_drm_vidi.c
index 65108cbd79d4..608b0afa337f 100644
--- a/drivers/gpu/drm/exynos/exynos_drm_vidi.c
+++ b/drivers/gpu/drm/exynos/exynos_drm_vidi.c
@@ -223,7 +223,7 @@ static void vidi_fake_vblank_handler(struct work_struct *work)
 	}
 }
 
-static int vidi_show_connection(struct device *dev,
+static ssize_t vidi_show_connection(struct device *dev,
 				struct device_attribute *attr, char *buf)
 {
 	struct vidi_context *ctx = dev_get_drvdata(dev);
@@ -238,7 +238,7 @@ static int vidi_show_connection(struct device *dev,
 	return rc;
 }
 
-static int vidi_store_connection(struct device *dev,
+static ssize_t vidi_store_connection(struct device *dev,
 				struct device_attribute *attr,
 				const char *buf, size_t len)
 {
@@ -294,7 +294,9 @@ int vidi_connection_ioctl(struct drm_device *drm_dev, void *data,
 	}
 
 	if (vidi->connection) {
-		struct edid *raw_edid  = (struct edid *)(uint32_t)vidi->edid;
+		struct edid *raw_edid;
+
+		raw_edid = (struct edid *)(unsigned long)vidi->edid;
 		if (!drm_edid_is_valid(raw_edid)) {
 			DRM_DEBUG_KMS("edid data is invalid.\n");
 			return -EINVAL;
diff --git a/drivers/gpu/drm/exynos/exynos_hdmi.c b/drivers/gpu/drm/exynos/exynos_hdmi.c
index 21a29dbce18c..e148d728e28c 100644
--- a/drivers/gpu/drm/exynos/exynos_hdmi.c
+++ b/drivers/gpu/drm/exynos/exynos_hdmi.c
@@ -867,10 +867,8 @@ static void hdmi_reg_infoframe(struct hdmi_context *hdata,
 {
 	u32 hdr_sum;
 	u8 chksum;
-	u32 mod;
 	u8 ar;
 
-	mod = hdmi_reg_read(hdata, HDMI_MODE_SEL);
 	if (hdata->dvi_mode) {
 		hdmi_reg_writeb(hdata, HDMI_VSI_CON,
 				HDMI_VSI_CON_DO_NOT_TRANSMIT);
diff --git a/drivers/gpu/drm/i2c/tda998x_drv.c b/drivers/gpu/drm/i2c/tda998x_drv.c
index b61282d89aa3..f4315bc8d471 100644
--- a/drivers/gpu/drm/i2c/tda998x_drv.c
+++ b/drivers/gpu/drm/i2c/tda998x_drv.c
@@ -1373,8 +1373,16 @@ static void tda998x_connector_destroy(struct drm_connector *connector)
 	drm_connector_cleanup(connector);
 }
 
+static int tda998x_connector_dpms(struct drm_connector *connector, int mode)
+{
+	if (drm_core_check_feature(connector->dev, DRIVER_ATOMIC))
+		return drm_atomic_helper_connector_dpms(connector, mode);
+	else
+		return drm_helper_connector_dpms(connector, mode);
+}
+
 static const struct drm_connector_funcs tda998x_connector_funcs = {
-	.dpms = drm_atomic_helper_connector_dpms,
+	.dpms = tda998x_connector_dpms,
 	.reset = drm_atomic_helper_connector_reset,
 	.fill_modes = drm_helper_probe_single_connector_modes,
 	.detect = tda998x_connector_detect,
diff --git a/drivers/gpu/drm/i915/Kconfig b/drivers/gpu/drm/i915/Kconfig
index 051eab33e4c7..20a5d0455e19 100644
--- a/drivers/gpu/drm/i915/Kconfig
+++ b/drivers/gpu/drm/i915/Kconfig
@@ -2,9 +2,7 @@ config DRM_I915
 	tristate "Intel 8xx/9xx/G3x/G4x/HD Graphics"
 	depends on DRM
 	depends on X86 && PCI
-	depends on (AGP || AGP=n)
 	select INTEL_GTT
-	select AGP_INTEL if AGP
 	select INTERVAL_TREE
 	# we need shmfs for the swappable backing store, and in particular
 	# the shmem_readpage() which depends upon tmpfs
@@ -47,3 +45,14 @@ config DRM_I915_PRELIMINARY_HW_SUPPORT
 	  option changes the default for that module option.
 
 	  If in doubt, say "N".
+
+config DRM_I915_USERPTR
+	bool "Always enable userptr support"
+	depends on DRM_I915
+	select MMU_NOTIFIER
+	default y
+	help
+	  This option selects CONFIG_MMU_NOTIFIER if it isn't already
+	  selected to enabled full userptr support.
+
+	  If in doubt, say "Y".
diff --git a/drivers/gpu/drm/i915/i915_debugfs.c b/drivers/gpu/drm/i915/i915_debugfs.c
index c5db23511184..a0f1bd711b53 100644
--- a/drivers/gpu/drm/i915/i915_debugfs.c
+++ b/drivers/gpu/drm/i915/i915_debugfs.c
@@ -117,9 +117,8 @@ static u64 i915_gem_obj_total_ggtt_size(struct drm_i915_gem_object *obj)
 	u64 size = 0;
 	struct i915_vma *vma;
 
-	list_for_each_entry(vma, &obj->vma_list, vma_link) {
-		if (i915_is_ggtt(vma->vm) &&
-		    drm_mm_node_allocated(&vma->node))
+	list_for_each_entry(vma, &obj->vma_list, obj_link) {
+		if (vma->is_ggtt && drm_mm_node_allocated(&vma->node))
 			size += vma->node.size;
 	}
 
@@ -155,7 +154,7 @@ describe_obj(struct seq_file *m, struct drm_i915_gem_object *obj)
 		   obj->madv == I915_MADV_DONTNEED ? " purgeable" : "");
 	if (obj->base.name)
 		seq_printf(m, " (name: %d)", obj->base.name);
-	list_for_each_entry(vma, &obj->vma_list, vma_link) {
+	list_for_each_entry(vma, &obj->vma_list, obj_link) {
 		if (vma->pin_count > 0)
 			pin_count++;
 	}
@@ -164,14 +163,13 @@ describe_obj(struct seq_file *m, struct drm_i915_gem_object *obj)
 		seq_printf(m, " (display)");
 	if (obj->fence_reg != I915_FENCE_REG_NONE)
 		seq_printf(m, " (fence: %d)", obj->fence_reg);
-	list_for_each_entry(vma, &obj->vma_list, vma_link) {
+	list_for_each_entry(vma, &obj->vma_list, obj_link) {
 		seq_printf(m, " (%sgtt offset: %08llx, size: %08llx",
-			   i915_is_ggtt(vma->vm) ? "g" : "pp",
+			   vma->is_ggtt ? "g" : "pp",
 			   vma->node.start, vma->node.size);
-		if (i915_is_ggtt(vma->vm))
-			seq_printf(m, ", type: %u)", vma->ggtt_view.type);
-		else
-			seq_puts(m, ")");
+		if (vma->is_ggtt)
+			seq_printf(m, ", type: %u", vma->ggtt_view.type);
+		seq_puts(m, ")");
 	}
 	if (obj->stolen)
 		seq_printf(m, " (stolen: %08llx)", obj->stolen->start);
@@ -230,7 +228,7 @@ static int i915_gem_object_list_info(struct seq_file *m, void *data)
 	}
 
 	total_obj_size = total_gtt_size = count = 0;
-	list_for_each_entry(vma, head, mm_list) {
+	list_for_each_entry(vma, head, vm_link) {
 		seq_printf(m, "   ");
 		describe_obj(m, vma->obj);
 		seq_printf(m, "\n");
@@ -342,13 +340,13 @@ static int per_file_stats(int id, void *ptr, void *data)
 		stats->shared += obj->base.size;
 
 	if (USES_FULL_PPGTT(obj->base.dev)) {
-		list_for_each_entry(vma, &obj->vma_list, vma_link) {
+		list_for_each_entry(vma, &obj->vma_list, obj_link) {
 			struct i915_hw_ppgtt *ppgtt;
 
 			if (!drm_mm_node_allocated(&vma->node))
 				continue;
 
-			if (i915_is_ggtt(vma->vm)) {
+			if (vma->is_ggtt) {
 				stats->global += obj->base.size;
 				continue;
 			}
@@ -454,12 +452,12 @@ static int i915_gem_object_info(struct seq_file *m, void* data)
 		   count, mappable_count, size, mappable_size);
 
 	size = count = mappable_size = mappable_count = 0;
-	count_vmas(&vm->active_list, mm_list);
+	count_vmas(&vm->active_list, vm_link);
 	seq_printf(m, "  %u [%u] active objects, %llu [%llu] bytes\n",
 		   count, mappable_count, size, mappable_size);
 
 	size = count = mappable_size = mappable_count = 0;
-	count_vmas(&vm->inactive_list, mm_list);
+	count_vmas(&vm->inactive_list, vm_link);
 	seq_printf(m, "  %u [%u] inactive objects, %llu [%llu] bytes\n",
 		   count, mappable_count, size, mappable_size);
 
@@ -825,8 +823,11 @@ static int i915_interrupt_info(struct seq_file *m, void *data)
 		}
 
 		for_each_pipe(dev_priv, pipe) {
-			if (!intel_display_power_is_enabled(dev_priv,
-						POWER_DOMAIN_PIPE(pipe))) {
+			enum intel_display_power_domain power_domain;
+
+			power_domain = POWER_DOMAIN_PIPE(pipe);
+			if (!intel_display_power_get_if_enabled(dev_priv,
+								power_domain)) {
 				seq_printf(m, "Pipe %c power disabled\n",
 					   pipe_name(pipe));
 				continue;
@@ -840,6 +841,8 @@ static int i915_interrupt_info(struct seq_file *m, void *data)
 			seq_printf(m, "Pipe %c IER:\t%08x\n",
 				   pipe_name(pipe),
 				   I915_READ(GEN8_DE_PIPE_IER(pipe)));
+
+			intel_display_power_put(dev_priv, power_domain);
 		}
 
 		seq_printf(m, "Display Engine port interrupt mask:\t%08x\n",
@@ -2463,9 +2466,9 @@ static void i915_guc_client_info(struct seq_file *m,
 
 	for_each_ring(ring, dev_priv, i) {
 		seq_printf(m, "\tSubmissions: %llu %s\n",
-				client->submissions[i],
+				client->submissions[ring->guc_id],
 				ring->name);
-		tot += client->submissions[i];
+		tot += client->submissions[ring->guc_id];
 	}
 	seq_printf(m, "\tTotal: %llu\n", tot);
 }
@@ -2502,10 +2505,10 @@ static int i915_guc_info(struct seq_file *m, void *data)
 
 	seq_printf(m, "\nGuC submissions:\n");
 	for_each_ring(ring, dev_priv, i) {
-		seq_printf(m, "\t%-24s: %10llu, last seqno 0x%08x %9d\n",
-			ring->name, guc.submissions[i],
-			guc.last_seqno[i], guc.last_seqno[i]);
-		total += guc.submissions[i];
+		seq_printf(m, "\t%-24s: %10llu, last seqno 0x%08x\n",
+			ring->name, guc.submissions[ring->guc_id],
+			guc.last_seqno[ring->guc_id]);
+		total += guc.submissions[ring->guc_id];
 	}
 	seq_printf(m, "\t%s: %llu\n", "Total", total);
 
@@ -2583,6 +2586,10 @@ static int i915_edp_psr_status(struct seq_file *m, void *data)
 				enabled = true;
 		}
 	}
+
+	seq_printf(m, "Main link in standby mode: %s\n",
+		   yesno(dev_priv->psr.link_standby));
+
 	seq_printf(m, "HW Enabled & Active bit: %s", yesno(enabled));
 
 	if (!HAS_DDI(dev))
@@ -3221,9 +3228,11 @@ static int i915_wa_registers(struct seq_file *m, void *unused)
 {
 	int i;
 	int ret;
+	struct intel_engine_cs *ring;
 	struct drm_info_node *node = (struct drm_info_node *) m->private;
 	struct drm_device *dev = node->minor->dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
+	struct i915_workarounds *workarounds = &dev_priv->workarounds;
 
 	ret = mutex_lock_interruptible(&dev->struct_mutex);
 	if (ret)
@@ -3231,15 +3240,18 @@ static int i915_wa_registers(struct seq_file *m, void *unused)
 
 	intel_runtime_pm_get(dev_priv);
 
-	seq_printf(m, "Workarounds applied: %d\n", dev_priv->workarounds.count);
-	for (i = 0; i < dev_priv->workarounds.count; ++i) {
+	seq_printf(m, "Workarounds applied: %d\n", workarounds->count);
+	for_each_ring(ring, dev_priv, i)
+		seq_printf(m, "HW whitelist count for %s: %d\n",
+			   ring->name, workarounds->hw_whitelist_count[i]);
+	for (i = 0; i < workarounds->count; ++i) {
 		i915_reg_t addr;
 		u32 mask, value, read;
 		bool ok;
 
-		addr = dev_priv->workarounds.reg[i].addr;
-		mask = dev_priv->workarounds.reg[i].mask;
-		value = dev_priv->workarounds.reg[i].value;
+		addr = workarounds->reg[i].addr;
+		mask = workarounds->reg[i].mask;
+		value = workarounds->reg[i].value;
 		read = I915_READ(addr);
 		ok = (value & mask) == (read & mask);
 		seq_printf(m, "0x%X: 0x%08X, mask: 0x%08X, read: 0x%08x, status: %s\n",
@@ -3995,6 +4007,7 @@ static int pipe_crc_set_source(struct drm_device *dev, enum pipe pipe,
 	struct intel_pipe_crc *pipe_crc = &dev_priv->pipe_crc[pipe];
 	struct intel_crtc *crtc = to_intel_crtc(intel_get_crtc_for_pipe(dev,
 									pipe));
+	enum intel_display_power_domain power_domain;
 	u32 val = 0; /* shut up gcc */
 	int ret;
 
@@ -4005,7 +4018,8 @@ static int pipe_crc_set_source(struct drm_device *dev, enum pipe pipe,
 	if (pipe_crc->source && source)
 		return -EINVAL;
 
-	if (!intel_display_power_is_enabled(dev_priv, POWER_DOMAIN_PIPE(pipe))) {
+	power_domain = POWER_DOMAIN_PIPE(pipe);
+	if (!intel_display_power_get_if_enabled(dev_priv, power_domain)) {
 		DRM_DEBUG_KMS("Trying to capture CRC while pipe is off\n");
 		return -EIO;
 	}
@@ -4022,7 +4036,7 @@ static int pipe_crc_set_source(struct drm_device *dev, enum pipe pipe,
 		ret = ivb_pipe_crc_ctl_reg(dev, pipe, &source, &val);
 
 	if (ret != 0)
-		return ret;
+		goto out;
 
 	/* none -> real source transition */
 	if (source) {
@@ -4034,8 +4048,10 @@ static int pipe_crc_set_source(struct drm_device *dev, enum pipe pipe,
 		entries = kcalloc(INTEL_PIPE_CRC_ENTRIES_NR,
 				  sizeof(pipe_crc->entries[0]),
 				  GFP_KERNEL);
-		if (!entries)
-			return -ENOMEM;
+		if (!entries) {
+			ret = -ENOMEM;
+			goto out;
+		}
 
 		/*
 		 * When IPS gets enabled, the pipe CRC changes. Since IPS gets
@@ -4091,7 +4107,12 @@ static int pipe_crc_set_source(struct drm_device *dev, enum pipe pipe,
 		hsw_enable_ips(crtc);
 	}
 
-	return 0;
+	ret = 0;
+
+out:
+	intel_display_power_put(dev_priv, power_domain);
+
+	return ret;
 }
 
 /*
diff --git a/drivers/gpu/drm/i915/i915_dma.c b/drivers/gpu/drm/i915/i915_dma.c
index 1c3d2544fec4..1c6d227aae7c 100644
--- a/drivers/gpu/drm/i915/i915_dma.c
+++ b/drivers/gpu/drm/i915/i915_dma.c
@@ -391,20 +391,13 @@ static int i915_load_modeset_init(struct drm_device *dev)
 	if (ret)
 		goto cleanup_vga_client;
 
-	/* Initialise stolen first so that we may reserve preallocated
-	 * objects for the BIOS to KMS transition.
-	 */
-	ret = i915_gem_init_stolen(dev);
-	if (ret)
-		goto cleanup_vga_switcheroo;
-
 	intel_power_domains_init_hw(dev_priv, false);
 
 	intel_csr_ucode_init(dev_priv);
 
 	ret = intel_irq_install(dev_priv);
 	if (ret)
-		goto cleanup_gem_stolen;
+		goto cleanup_csr;
 
 	intel_setup_gmbus(dev);
 
@@ -458,9 +451,8 @@ cleanup_irq:
 	intel_guc_ucode_fini(dev);
 	drm_irq_uninstall(dev);
 	intel_teardown_gmbus(dev);
-cleanup_gem_stolen:
-	i915_gem_cleanup_stolen(dev);
-cleanup_vga_switcheroo:
+cleanup_csr:
+	intel_csr_ucode_fini(dev_priv);
 	vga_switcheroo_unregister_client(dev->pdev);
 cleanup_vga_client:
 	vga_client_register(dev->pdev, NULL, NULL, NULL);
@@ -816,7 +808,41 @@ static void intel_device_info_runtime_init(struct drm_device *dev)
 		     !(sfuse_strap & SFUSE_STRAP_FUSE_LOCK))) {
 			DRM_INFO("Display fused off, disabling\n");
 			info->num_pipes = 0;
+		} else if (fuse_strap & IVB_PIPE_C_DISABLE) {
+			DRM_INFO("PipeC fused off\n");
+			info->num_pipes -= 1;
 		}
+	} else if (info->num_pipes > 0 && INTEL_INFO(dev)->gen == 9) {
+		u32 dfsm = I915_READ(SKL_DFSM);
+		u8 disabled_mask = 0;
+		bool invalid;
+		int num_bits;
+
+		if (dfsm & SKL_DFSM_PIPE_A_DISABLE)
+			disabled_mask |= BIT(PIPE_A);
+		if (dfsm & SKL_DFSM_PIPE_B_DISABLE)
+			disabled_mask |= BIT(PIPE_B);
+		if (dfsm & SKL_DFSM_PIPE_C_DISABLE)
+			disabled_mask |= BIT(PIPE_C);
+
+		num_bits = hweight8(disabled_mask);
+
+		switch (disabled_mask) {
+		case BIT(PIPE_A):
+		case BIT(PIPE_B):
+		case BIT(PIPE_A) | BIT(PIPE_B):
+		case BIT(PIPE_A) | BIT(PIPE_C):
+			invalid = true;
+			break;
+		default:
+			invalid = false;
+		}
+
+		if (num_bits > info->num_pipes || invalid)
+			DRM_ERROR("invalid pipe fuse configuration: 0x%x\n",
+				  disabled_mask);
+		else
+			info->num_pipes -= num_bits;
 	}
 
 	/* Initialize slice/subslice/EU info */
@@ -855,6 +881,94 @@ static void intel_init_dpio(struct drm_i915_private *dev_priv)
 	}
 }
 
+static int i915_workqueues_init(struct drm_i915_private *dev_priv)
+{
+	/*
+	 * The i915 workqueue is primarily used for batched retirement of
+	 * requests (and thus managing bo) once the task has been completed
+	 * by the GPU. i915_gem_retire_requests() is called directly when we
+	 * need high-priority retirement, such as waiting for an explicit
+	 * bo.
+	 *
+	 * It is also used for periodic low-priority events, such as
+	 * idle-timers and recording error state.
+	 *
+	 * All tasks on the workqueue are expected to acquire the dev mutex
+	 * so there is no point in running more than one instance of the
+	 * workqueue at any time.  Use an ordered one.
+	 */
+	dev_priv->wq = alloc_ordered_workqueue("i915", 0);
+	if (dev_priv->wq == NULL)
+		goto out_err;
+
+	dev_priv->hotplug.dp_wq = alloc_ordered_workqueue("i915-dp", 0);
+	if (dev_priv->hotplug.dp_wq == NULL)
+		goto out_free_wq;
+
+	dev_priv->gpu_error.hangcheck_wq =
+		alloc_ordered_workqueue("i915-hangcheck", 0);
+	if (dev_priv->gpu_error.hangcheck_wq == NULL)
+		goto out_free_dp_wq;
+
+	return 0;
+
+out_free_dp_wq:
+	destroy_workqueue(dev_priv->hotplug.dp_wq);
+out_free_wq:
+	destroy_workqueue(dev_priv->wq);
+out_err:
+	DRM_ERROR("Failed to allocate workqueues.\n");
+
+	return -ENOMEM;
+}
+
+static void i915_workqueues_cleanup(struct drm_i915_private *dev_priv)
+{
+	destroy_workqueue(dev_priv->gpu_error.hangcheck_wq);
+	destroy_workqueue(dev_priv->hotplug.dp_wq);
+	destroy_workqueue(dev_priv->wq);
+}
+
+static int i915_mmio_setup(struct drm_device *dev)
+{
+	struct drm_i915_private *dev_priv = to_i915(dev);
+	int mmio_bar;
+	int mmio_size;
+
+	mmio_bar = IS_GEN2(dev) ? 1 : 0;
+	/*
+	 * Before gen4, the registers and the GTT are behind different BARs.
+	 * However, from gen4 onwards, the registers and the GTT are shared
+	 * in the same BAR, so we want to restrict this ioremap from
+	 * clobbering the GTT which we want ioremap_wc instead. Fortunately,
+	 * the register BAR remains the same size for all the earlier
+	 * generations up to Ironlake.
+	 */
+	if (INTEL_INFO(dev)->gen < 5)
+		mmio_size = 512 * 1024;
+	else
+		mmio_size = 2 * 1024 * 1024;
+	dev_priv->regs = pci_iomap(dev->pdev, mmio_bar, mmio_size);
+	if (dev_priv->regs == NULL) {
+		DRM_ERROR("failed to map registers\n");
+
+		return -EIO;
+	}
+
+	/* Try to make sure MCHBAR is enabled before poking at it */
+	intel_setup_mchbar(dev);
+
+	return 0;
+}
+
+static void i915_mmio_cleanup(struct drm_device *dev)
+{
+	struct drm_i915_private *dev_priv = to_i915(dev);
+
+	intel_teardown_mchbar(dev);
+	pci_iounmap(dev->pdev, dev_priv->regs);
+}
+
 /**
  * i915_driver_load - setup chip and create an initial config
  * @dev: DRM device
@@ -870,7 +984,7 @@ int i915_driver_load(struct drm_device *dev, unsigned long flags)
 {
 	struct drm_i915_private *dev_priv;
 	struct intel_device_info *info, *device_info;
-	int ret = 0, mmio_bar, mmio_size;
+	int ret = 0;
 	uint32_t aperture_size;
 
 	info = (struct intel_device_info *) flags;
@@ -897,6 +1011,10 @@ int i915_driver_load(struct drm_device *dev, unsigned long flags)
 	mutex_init(&dev_priv->modeset_restore_lock);
 	mutex_init(&dev_priv->av_mutex);
 
+	ret = i915_workqueues_init(dev_priv);
+	if (ret < 0)
+		goto out_free_priv;
+
 	intel_pm_setup(dev);
 
 	intel_runtime_pm_get(dev_priv);
@@ -915,28 +1033,12 @@ int i915_driver_load(struct drm_device *dev, unsigned long flags)
 
 	if (i915_get_bridge_dev(dev)) {
 		ret = -EIO;
-		goto free_priv;
+		goto out_runtime_pm_put;
 	}
 
-	mmio_bar = IS_GEN2(dev) ? 1 : 0;
-	/* Before gen4, the registers and the GTT are behind different BARs.
-	 * However, from gen4 onwards, the registers and the GTT are shared
-	 * in the same BAR, so we want to restrict this ioremap from
-	 * clobbering the GTT which we want ioremap_wc instead. Fortunately,
-	 * the register BAR remains the same size for all the earlier
-	 * generations up to Ironlake.
-	 */
-	if (info->gen < 5)
-		mmio_size = 512*1024;
-	else
-		mmio_size = 2*1024*1024;
-
-	dev_priv->regs = pci_iomap(dev->pdev, mmio_bar, mmio_size);
-	if (!dev_priv->regs) {
-		DRM_ERROR("failed to map registers\n");
-		ret = -EIO;
+	ret = i915_mmio_setup(dev);
+	if (ret < 0)
 		goto put_bridge;
-	}
 
 	/* This must be called before any calls to HAS_PCH_* */
 	intel_detect_pch(dev);
@@ -945,7 +1047,7 @@ int i915_driver_load(struct drm_device *dev, unsigned long flags)
 
 	ret = i915_gem_gtt_init(dev);
 	if (ret)
-		goto out_freecsr;
+		goto out_uncore_fini;
 
 	/* WARNING: Apparently we must kick fbdev drivers before vgacon,
 	 * otherwise the vga fbdev driver falls over. */
@@ -991,49 +1093,13 @@ int i915_driver_load(struct drm_device *dev, unsigned long flags)
 	dev_priv->gtt.mtrr = arch_phys_wc_add(dev_priv->gtt.mappable_base,
 					      aperture_size);
 
-	/* The i915 workqueue is primarily used for batched retirement of
-	 * requests (and thus managing bo) once the task has been completed
-	 * by the GPU. i915_gem_retire_requests() is called directly when we
-	 * need high-priority retirement, such as waiting for an explicit
-	 * bo.
-	 *
-	 * It is also used for periodic low-priority events, such as
-	 * idle-timers and recording error state.
-	 *
-	 * All tasks on the workqueue are expected to acquire the dev mutex
-	 * so there is no point in running more than one instance of the
-	 * workqueue at any time.  Use an ordered one.
-	 */
-	dev_priv->wq = alloc_ordered_workqueue("i915", 0);
-	if (dev_priv->wq == NULL) {
-		DRM_ERROR("Failed to create our workqueue.\n");
-		ret = -ENOMEM;
-		goto out_mtrrfree;
-	}
-
-	dev_priv->hotplug.dp_wq = alloc_ordered_workqueue("i915-dp", 0);
-	if (dev_priv->hotplug.dp_wq == NULL) {
-		DRM_ERROR("Failed to create our dp workqueue.\n");
-		ret = -ENOMEM;
-		goto out_freewq;
-	}
-
-	dev_priv->gpu_error.hangcheck_wq =
-		alloc_ordered_workqueue("i915-hangcheck", 0);
-	if (dev_priv->gpu_error.hangcheck_wq == NULL) {
-		DRM_ERROR("Failed to create our hangcheck workqueue.\n");
-		ret = -ENOMEM;
-		goto out_freedpwq;
-	}
-
 	intel_irq_init(dev_priv);
 	intel_uncore_sanitize(dev);
 
-	/* Try to make sure MCHBAR is enabled before poking at it */
-	intel_setup_mchbar(dev);
 	intel_opregion_setup(dev);
 
-	i915_gem_load(dev);
+	i915_gem_load_init(dev);
+	i915_gem_shrinker_init(dev_priv);
 
 	/* On the 945G/GM, the chipset reports the MSI capability on the
 	 * integrated graphics even though the support isn't actually there
@@ -1046,8 +1112,10 @@ int i915_driver_load(struct drm_device *dev, unsigned long flags)
 	 * be lost or delayed, but we use them anyways to avoid
 	 * stuck interrupts on some machines.
 	 */
-	if (!IS_I945G(dev) && !IS_I945GM(dev))
-		pci_enable_msi(dev->pdev);
+	if (!IS_I945G(dev) && !IS_I945GM(dev)) {
+		if (pci_enable_msi(dev->pdev) < 0)
+			DRM_DEBUG_DRIVER("can't enable MSI");
+	}
 
 	intel_device_info_runtime_init(dev);
 
@@ -1097,38 +1165,29 @@ out_power_well:
 	intel_power_domains_fini(dev_priv);
 	drm_vblank_cleanup(dev);
 out_gem_unload:
-	WARN_ON(unregister_oom_notifier(&dev_priv->mm.oom_notifier));
-	unregister_shrinker(&dev_priv->mm.shrinker);
+	i915_gem_shrinker_cleanup(dev_priv);
 
 	if (dev->pdev->msi_enabled)
 		pci_disable_msi(dev->pdev);
 
 	intel_teardown_mchbar(dev);
 	pm_qos_remove_request(&dev_priv->pm_qos);
-	destroy_workqueue(dev_priv->gpu_error.hangcheck_wq);
-out_freedpwq:
-	destroy_workqueue(dev_priv->hotplug.dp_wq);
-out_freewq:
-	destroy_workqueue(dev_priv->wq);
-out_mtrrfree:
 	arch_phys_wc_del(dev_priv->gtt.mtrr);
 	io_mapping_free(dev_priv->gtt.mappable);
 out_gtt:
 	i915_global_gtt_cleanup(dev);
-out_freecsr:
-	intel_csr_ucode_fini(dev_priv);
+out_uncore_fini:
 	intel_uncore_fini(dev);
-	pci_iounmap(dev->pdev, dev_priv->regs);
+	i915_mmio_cleanup(dev);
 put_bridge:
 	pci_dev_put(dev_priv->bridge_dev);
-free_priv:
-	kmem_cache_destroy(dev_priv->requests);
-	kmem_cache_destroy(dev_priv->vmas);
-	kmem_cache_destroy(dev_priv->objects);
-
+	i915_gem_load_cleanup(dev);
+out_runtime_pm_put:
 	intel_runtime_pm_put(dev_priv);
-
+	i915_workqueues_cleanup(dev_priv);
+out_free_priv:
 	kfree(dev_priv);
+
 	return ret;
 }
 
@@ -1153,8 +1212,7 @@ int i915_driver_unload(struct drm_device *dev)
 
 	i915_teardown_sysfs(dev);
 
-	WARN_ON(unregister_oom_notifier(&dev_priv->mm.oom_notifier));
-	unregister_shrinker(&dev_priv->mm.shrinker);
+	i915_gem_shrinker_cleanup(dev_priv);
 
 	io_mapping_free(dev_priv->gtt.mappable);
 	arch_phys_wc_del(dev_priv->gtt.mtrr);
@@ -1182,6 +1240,8 @@ int i915_driver_unload(struct drm_device *dev)
 	vga_switcheroo_unregister_client(dev->pdev);
 	vga_client_register(dev->pdev, NULL, NULL, NULL);
 
+	intel_csr_ucode_fini(dev_priv);
+
 	/* Free error state after interrupts are fully disabled. */
 	cancel_delayed_work_sync(&dev_priv->gpu_error.hangcheck_work);
 	i915_destroy_error_state(dev);
@@ -1200,27 +1260,17 @@ int i915_driver_unload(struct drm_device *dev)
 	i915_gem_context_fini(dev);
 	mutex_unlock(&dev->struct_mutex);
 	intel_fbc_cleanup_cfb(dev_priv);
-	i915_gem_cleanup_stolen(dev);
 
-	intel_csr_ucode_fini(dev_priv);
-
-	intel_teardown_mchbar(dev);
-
-	destroy_workqueue(dev_priv->hotplug.dp_wq);
-	destroy_workqueue(dev_priv->wq);
-	destroy_workqueue(dev_priv->gpu_error.hangcheck_wq);
 	pm_qos_remove_request(&dev_priv->pm_qos);
 
 	i915_global_gtt_cleanup(dev);
 
 	intel_uncore_fini(dev);
-	if (dev_priv->regs != NULL)
-		pci_iounmap(dev->pdev, dev_priv->regs);
+	i915_mmio_cleanup(dev);
 
-	kmem_cache_destroy(dev_priv->requests);
-	kmem_cache_destroy(dev_priv->vmas);
-	kmem_cache_destroy(dev_priv->objects);
+	i915_gem_load_cleanup(dev);
 	pci_dev_put(dev_priv->bridge_dev);
+	i915_workqueues_cleanup(dev_priv);
 	kfree(dev_priv);
 
 	return 0;
diff --git a/drivers/gpu/drm/i915/i915_drv.c b/drivers/gpu/drm/i915/i915_drv.c
index 44912ecebc1a..20e82008b8b6 100644
--- a/drivers/gpu/drm/i915/i915_drv.c
+++ b/drivers/gpu/drm/i915/i915_drv.c
@@ -603,13 +603,7 @@ static int i915_drm_suspend(struct drm_device *dev)
 
 	intel_suspend_gt_powersave(dev);
 
-	/*
-	 * Disable CRTCs directly since we want to preserve sw state
-	 * for _thaw. Also, power gate the CRTC power wells.
-	 */
-	drm_modeset_lock_all(dev);
 	intel_display_suspend(dev);
-	drm_modeset_unlock_all(dev);
 
 	intel_dp_mst_suspend(dev);
 
@@ -764,9 +758,7 @@ static int i915_drm_resume(struct drm_device *dev)
 		dev_priv->display.hpd_irq_setup(dev);
 	spin_unlock_irq(&dev_priv->irq_lock);
 
-	drm_modeset_lock_all(dev);
 	intel_display_resume(dev);
-	drm_modeset_unlock_all(dev);
 
 	intel_dp_mst_resume(dev);
 
diff --git a/drivers/gpu/drm/i915/i915_drv.h b/drivers/gpu/drm/i915/i915_drv.h
index afb0beee9975..10480939159c 100644
--- a/drivers/gpu/drm/i915/i915_drv.h
+++ b/drivers/gpu/drm/i915/i915_drv.h
@@ -59,7 +59,7 @@
 
 #define DRIVER_NAME		"i915"
 #define DRIVER_DESC		"Intel Graphics"
-#define DRIVER_DATE		"20160124"
+#define DRIVER_DATE		"20160229"
 
 #undef WARN_ON
 /* Many gcc seem to no see through this and fall over :( */
@@ -261,6 +261,9 @@ struct i915_hotplug {
 
 #define for_each_pipe(__dev_priv, __p) \
 	for ((__p) = 0; (__p) < INTEL_INFO(__dev_priv)->num_pipes; (__p)++)
+#define for_each_pipe_masked(__dev_priv, __p, __mask) \
+	for ((__p) = 0; (__p) < INTEL_INFO(__dev_priv)->num_pipes; (__p)++) \
+		for_each_if ((__mask) & (1 << (__p)))
 #define for_each_plane(__dev_priv, __pipe, __p)				\
 	for ((__p) = 0;							\
 	     (__p) < INTEL_INFO(__dev_priv)->num_sprites[(__pipe)] + 1;	\
@@ -746,6 +749,7 @@ struct intel_csr {
 	uint32_t mmio_count;
 	i915_reg_t mmioaddr[8];
 	uint32_t mmiodata[8];
+	uint32_t dc_state;
 };
 
 #define DEV_INFO_FOR_EACH_FLAG(func, sep) \
@@ -900,16 +904,15 @@ enum fb_op_origin {
 	ORIGIN_DIRTYFB,
 };
 
-struct i915_fbc {
+struct intel_fbc {
 	/* This is always the inner lock when overlapping with struct_mutex and
 	 * it's the outer lock when overlapping with stolen_lock. */
 	struct mutex lock;
 	unsigned threshold;
-	unsigned int fb_id;
 	unsigned int possible_framebuffer_bits;
 	unsigned int busy_bits;
+	unsigned int visible_pipes_mask;
 	struct intel_crtc *crtc;
-	int y;
 
 	struct drm_mm_node compressed_fb;
 	struct drm_mm_node *compressed_llb;
@@ -919,18 +922,52 @@ struct i915_fbc {
 	bool enabled;
 	bool active;
 
+	struct intel_fbc_state_cache {
+		struct {
+			unsigned int mode_flags;
+			uint32_t hsw_bdw_pixel_rate;
+		} crtc;
+
+		struct {
+			unsigned int rotation;
+			int src_w;
+			int src_h;
+			bool visible;
+		} plane;
+
+		struct {
+			u64 ilk_ggtt_offset;
+			uint32_t pixel_format;
+			unsigned int stride;
+			int fence_reg;
+			unsigned int tiling_mode;
+		} fb;
+	} state_cache;
+
+	struct intel_fbc_reg_params {
+		struct {
+			enum pipe pipe;
+			enum plane plane;
+			unsigned int fence_y_offset;
+		} crtc;
+
+		struct {
+			u64 ggtt_offset;
+			uint32_t pixel_format;
+			unsigned int stride;
+			int fence_reg;
+		} fb;
+
+		int cfb_size;
+	} params;
+
 	struct intel_fbc_work {
 		bool scheduled;
+		u32 scheduled_vblank;
 		struct work_struct work;
-		struct drm_framebuffer *fb;
-		unsigned long enable_jiffies;
 	} work;
 
 	const char *no_fbc_reason;
-
-	bool (*is_active)(struct drm_i915_private *dev_priv);
-	void (*activate)(struct intel_crtc *crtc);
-	void (*deactivate)(struct drm_i915_private *dev_priv);
 };
 
 /**
@@ -970,6 +1007,7 @@ struct i915_psr {
 	unsigned busy_frontbuffer_bits;
 	bool psr2_support;
 	bool aux_frame_sync;
+	bool link_standby;
 };
 
 enum intel_pch {
@@ -1657,11 +1695,18 @@ struct i915_wa_reg {
 	u32 mask;
 };
 
-#define I915_MAX_WA_REGS 16
+/*
+ * RING_MAX_NONPRIV_SLOTS is per-engine but at this point we are only
+ * allowing it for RCS as we don't foresee any requirement of having
+ * a whitelist for other engines. When it is really required for
+ * other engines then the limit need to be increased.
+ */
+#define I915_MAX_WA_REGS (16 + RING_MAX_NONPRIV_SLOTS)
 
 struct i915_workarounds {
 	struct i915_wa_reg reg[I915_MAX_WA_REGS];
 	u32 count;
+	u32 hw_whitelist_count[I915_NUM_RINGS];
 };
 
 struct i915_virtual_gpu {
@@ -1758,7 +1803,7 @@ struct drm_i915_private {
 	u32 pipestat_irq_mask[I915_MAX_PIPES];
 
 	struct i915_hotplug hotplug;
-	struct i915_fbc fbc;
+	struct intel_fbc fbc;
 	struct i915_drrs drrs;
 	struct intel_opregion opregion;
 	struct intel_vbt_data vbt;
@@ -1807,6 +1852,7 @@ struct drm_i915_private {
 
 	enum modeset_restore modeset_restore;
 	struct mutex modeset_restore_lock;
+	struct drm_atomic_state *modeset_restore_state;
 
 	struct list_head vm_list; /* Global list of all address spaces */
 	struct i915_gtt gtt; /* VM representing the global address space */
@@ -1993,6 +2039,9 @@ enum hdmi_force_audio {
 #define I915_GTT_OFFSET_NONE ((u32)-1)
 
 struct drm_i915_gem_object_ops {
+	unsigned int flags;
+#define I915_GEM_OBJECT_HAS_STRUCT_PAGE 0x1
+
 	/* Interface between the GEM object and its backing storage.
 	 * get_pages() is called once prior to the use of the associated set
 	 * of pages before to binding them into the GTT, and put_pages() is
@@ -2008,6 +2057,7 @@ struct drm_i915_gem_object_ops {
 	 */
 	int (*get_pages)(struct drm_i915_gem_object *);
 	void (*put_pages)(struct drm_i915_gem_object *);
+
 	int (*dmabuf_export)(struct drm_i915_gem_object *);
 	void (*release)(struct drm_i915_gem_object *);
 };
@@ -2841,7 +2891,8 @@ int i915_gem_get_aperture_ioctl(struct drm_device *dev, void *data,
 				struct drm_file *file_priv);
 int i915_gem_wait_ioctl(struct drm_device *dev, void *data,
 			struct drm_file *file_priv);
-void i915_gem_load(struct drm_device *dev);
+void i915_gem_load_init(struct drm_device *dev);
+void i915_gem_load_cleanup(struct drm_device *dev);
 void *i915_gem_object_alloc(struct drm_device *dev);
 void i915_gem_object_free(struct drm_i915_gem_object *obj);
 void i915_gem_object_init(struct drm_i915_gem_object *obj,
@@ -3105,18 +3156,11 @@ bool i915_gem_obj_is_pinned(struct drm_i915_gem_object *obj);
 /* Some GGTT VM helpers */
 #define i915_obj_to_ggtt(obj) \
 	(&((struct drm_i915_private *)(obj)->base.dev->dev_private)->gtt.base)
-static inline bool i915_is_ggtt(struct i915_address_space *vm)
-{
-	struct i915_address_space *ggtt =
-		&((struct drm_i915_private *)(vm)->dev->dev_private)->gtt.base;
-	return vm == ggtt;
-}
 
 static inline struct i915_hw_ppgtt *
 i915_vm_to_ppgtt(struct i915_address_space *vm)
 {
 	WARN_ON(i915_is_ggtt(vm));
-
 	return container_of(vm, struct i915_hw_ppgtt, base);
 }
 
@@ -3254,6 +3298,7 @@ unsigned long i915_gem_shrink(struct drm_i915_private *dev_priv,
 #define I915_SHRINK_ACTIVE 0x8
 unsigned long i915_gem_shrink_all(struct drm_i915_private *dev_priv);
 void i915_gem_shrinker_init(struct drm_i915_private *dev_priv);
+void i915_gem_shrinker_cleanup(struct drm_i915_private *dev_priv);
 
 
 /* i915_gem_tiling.c */
@@ -3424,16 +3469,14 @@ int sandybridge_pcode_write(struct drm_i915_private *dev_priv, u32 mbox, u32 val
 u32 vlv_punit_read(struct drm_i915_private *dev_priv, u32 addr);
 void vlv_punit_write(struct drm_i915_private *dev_priv, u32 addr, u32 val);
 u32 vlv_nc_read(struct drm_i915_private *dev_priv, u8 addr);
-u32 vlv_gpio_nc_read(struct drm_i915_private *dev_priv, u32 reg);
-void vlv_gpio_nc_write(struct drm_i915_private *dev_priv, u32 reg, u32 val);
+u32 vlv_iosf_sb_read(struct drm_i915_private *dev_priv, u8 port, u32 reg);
+void vlv_iosf_sb_write(struct drm_i915_private *dev_priv, u8 port, u32 reg, u32 val);
 u32 vlv_cck_read(struct drm_i915_private *dev_priv, u32 reg);
 void vlv_cck_write(struct drm_i915_private *dev_priv, u32 reg, u32 val);
 u32 vlv_ccu_read(struct drm_i915_private *dev_priv, u32 reg);
 void vlv_ccu_write(struct drm_i915_private *dev_priv, u32 reg, u32 val);
 u32 vlv_bunit_read(struct drm_i915_private *dev_priv, u32 reg);
 void vlv_bunit_write(struct drm_i915_private *dev_priv, u32 reg, u32 val);
-u32 vlv_gps_core_read(struct drm_i915_private *dev_priv, u32 reg);
-void vlv_gps_core_write(struct drm_i915_private *dev_priv, u32 reg, u32 val);
 u32 vlv_dpio_read(struct drm_i915_private *dev_priv, enum pipe pipe, int reg);
 void vlv_dpio_write(struct drm_i915_private *dev_priv, enum pipe pipe, int reg, u32 val);
 u32 intel_sbi_read(struct drm_i915_private *dev_priv, u16 reg,
diff --git a/drivers/gpu/drm/i915/i915_gem.c b/drivers/gpu/drm/i915/i915_gem.c
index 371bbb28c471..3d31d3ac589e 100644
--- a/drivers/gpu/drm/i915/i915_gem.c
+++ b/drivers/gpu/drm/i915/i915_gem.c
@@ -138,10 +138,10 @@ i915_gem_get_aperture_ioctl(struct drm_device *dev, void *data,
 
 	pinned = 0;
 	mutex_lock(&dev->struct_mutex);
-	list_for_each_entry(vma, &ggtt->base.active_list, mm_list)
+	list_for_each_entry(vma, &ggtt->base.active_list, vm_link)
 		if (vma->pin_count)
 			pinned += vma->node.size;
-	list_for_each_entry(vma, &ggtt->base.inactive_list, mm_list)
+	list_for_each_entry(vma, &ggtt->base.inactive_list, vm_link)
 		if (vma->pin_count)
 			pinned += vma->node.size;
 	mutex_unlock(&dev->struct_mutex);
@@ -272,7 +272,7 @@ drop_pages(struct drm_i915_gem_object *obj)
 	int ret;
 
 	drm_gem_object_reference(&obj->base);
-	list_for_each_entry_safe(vma, next, &obj->vma_list, vma_link)
+	list_for_each_entry_safe(vma, next, &obj->vma_list, obj_link)
 		if (i915_vma_unbind(vma))
 			break;
 
@@ -489,7 +489,7 @@ int i915_gem_obj_prepare_shmem_read(struct drm_i915_gem_object *obj,
 
 	*needs_clflush = 0;
 
-	if (!obj->base.filp)
+	if (WARN_ON((obj->ops->flags & I915_GEM_OBJECT_HAS_STRUCT_PAGE) == 0))
 		return -EINVAL;
 
 	if (!(obj->base.read_domains & I915_GEM_DOMAIN_CPU)) {
@@ -2416,7 +2416,7 @@ void i915_vma_move_to_active(struct i915_vma *vma,
 	list_move_tail(&obj->ring_list[ring->id], &ring->active_list);
 	i915_gem_request_assign(&obj->last_read_req[ring->id], req);
 
-	list_move_tail(&vma->mm_list, &vma->vm->active_list);
+	list_move_tail(&vma->vm_link, &vma->vm->active_list);
 }
 
 static void
@@ -2454,9 +2454,9 @@ i915_gem_object_retire__read(struct drm_i915_gem_object *obj, int ring)
 	list_move_tail(&obj->global_list,
 		       &to_i915(obj->base.dev)->mm.bound_list);
 
-	list_for_each_entry(vma, &obj->vma_list, vma_link) {
-		if (!list_empty(&vma->mm_list))
-			list_move_tail(&vma->mm_list, &vma->vm->inactive_list);
+	list_for_each_entry(vma, &obj->vma_list, obj_link) {
+		if (!list_empty(&vma->vm_link))
+			list_move_tail(&vma->vm_link, &vma->vm->inactive_list);
 	}
 
 	i915_gem_request_assign(&obj->last_fenced_req, NULL);
@@ -2680,7 +2680,7 @@ void i915_gem_request_free(struct kref *req_ref)
 
 	if (ctx) {
 		if (i915.enable_execlists && ctx != req->i915->kernel_context)
-			intel_lr_context_unpin(req);
+			intel_lr_context_unpin(ctx, req->ring);
 
 		i915_gem_context_unreference(ctx);
 	}
@@ -2970,11 +2970,9 @@ i915_gem_retire_requests(struct drm_device *dev)
 		i915_gem_retire_requests_ring(ring);
 		idle &= list_empty(&ring->request_list);
 		if (i915.enable_execlists) {
-			unsigned long flags;
-
-			spin_lock_irqsave(&ring->execlist_lock, flags);
+			spin_lock_irq(&ring->execlist_lock);
 			idle &= list_empty(&ring->execlist_queue);
-			spin_unlock_irqrestore(&ring->execlist_lock, flags);
+			spin_unlock_irq(&ring->execlist_lock);
 
 			intel_execlists_retire_requests(ring);
 		}
@@ -3319,7 +3317,7 @@ static int __i915_vma_unbind(struct i915_vma *vma, bool wait)
 	struct drm_i915_private *dev_priv = obj->base.dev->dev_private;
 	int ret;
 
-	if (list_empty(&vma->vma_link))
+	if (list_empty(&vma->obj_link))
 		return 0;
 
 	if (!drm_mm_node_allocated(&vma->node)) {
@@ -3338,8 +3336,7 @@ static int __i915_vma_unbind(struct i915_vma *vma, bool wait)
 			return ret;
 	}
 
-	if (i915_is_ggtt(vma->vm) &&
-	    vma->ggtt_view.type == I915_GGTT_VIEW_NORMAL) {
+	if (vma->is_ggtt && vma->ggtt_view.type == I915_GGTT_VIEW_NORMAL) {
 		i915_gem_object_finish_gtt(obj);
 
 		/* release the fence reg _after_ flushing */
@@ -3353,8 +3350,8 @@ static int __i915_vma_unbind(struct i915_vma *vma, bool wait)
 	vma->vm->unbind_vma(vma);
 	vma->bound = 0;
 
-	list_del_init(&vma->mm_list);
-	if (i915_is_ggtt(vma->vm)) {
+	list_del_init(&vma->vm_link);
+	if (vma->is_ggtt) {
 		if (vma->ggtt_view.type == I915_GGTT_VIEW_NORMAL) {
 			obj->map_and_fenceable = false;
 		} else if (vma->ggtt_view.pages) {
@@ -3611,7 +3608,7 @@ search_free:
 		goto err_remove_node;
 
 	list_move_tail(&obj->global_list, &dev_priv->mm.bound_list);
-	list_add_tail(&vma->mm_list, &vm->inactive_list);
+	list_add_tail(&vma->vm_link, &vm->inactive_list);
 
 	return vma;
 
@@ -3776,7 +3773,7 @@ i915_gem_object_set_to_gtt_domain(struct drm_i915_gem_object *obj, bool write)
 	/* And bump the LRU for this access */
 	vma = i915_gem_obj_to_ggtt(obj);
 	if (vma && drm_mm_node_allocated(&vma->node) && !obj->active)
-		list_move_tail(&vma->mm_list,
+		list_move_tail(&vma->vm_link,
 			       &to_i915(obj->base.dev)->gtt.base.inactive_list);
 
 	return 0;
@@ -3811,7 +3808,7 @@ int i915_gem_object_set_cache_level(struct drm_i915_gem_object *obj,
 	 * catch the issue of the CS prefetch crossing page boundaries and
 	 * reading an invalid PTE on older architectures.
 	 */
-	list_for_each_entry_safe(vma, next, &obj->vma_list, vma_link) {
+	list_for_each_entry_safe(vma, next, &obj->vma_list, obj_link) {
 		if (!drm_mm_node_allocated(&vma->node))
 			continue;
 
@@ -3874,7 +3871,7 @@ int i915_gem_object_set_cache_level(struct drm_i915_gem_object *obj,
 			 */
 		}
 
-		list_for_each_entry(vma, &obj->vma_list, vma_link) {
+		list_for_each_entry(vma, &obj->vma_list, obj_link) {
 			if (!drm_mm_node_allocated(&vma->node))
 				continue;
 
@@ -3884,7 +3881,7 @@ int i915_gem_object_set_cache_level(struct drm_i915_gem_object *obj,
 		}
 	}
 
-	list_for_each_entry(vma, &obj->vma_list, vma_link)
+	list_for_each_entry(vma, &obj->vma_list, obj_link)
 		vma->node.color = cache_level;
 	obj->cache_level = cache_level;
 
@@ -4465,6 +4462,7 @@ void i915_gem_object_init(struct drm_i915_gem_object *obj,
 }
 
 static const struct drm_i915_gem_object_ops i915_gem_object_ops = {
+	.flags = I915_GEM_OBJECT_HAS_STRUCT_PAGE,
 	.get_pages = i915_gem_object_get_pages_gtt,
 	.put_pages = i915_gem_object_put_pages_gtt,
 };
@@ -4557,7 +4555,7 @@ void i915_gem_free_object(struct drm_gem_object *gem_obj)
 
 	trace_i915_gem_object_destroy(obj);
 
-	list_for_each_entry_safe(vma, next, &obj->vma_list, vma_link) {
+	list_for_each_entry_safe(vma, next, &obj->vma_list, obj_link) {
 		int ret;
 
 		vma->pin_count = 0;
@@ -4614,7 +4612,7 @@ struct i915_vma *i915_gem_obj_to_vma(struct drm_i915_gem_object *obj,
 				     struct i915_address_space *vm)
 {
 	struct i915_vma *vma;
-	list_for_each_entry(vma, &obj->vma_list, vma_link) {
+	list_for_each_entry(vma, &obj->vma_list, obj_link) {
 		if (vma->ggtt_view.type == I915_GGTT_VIEW_NORMAL &&
 		    vma->vm == vm)
 			return vma;
@@ -4631,7 +4629,7 @@ struct i915_vma *i915_gem_obj_to_ggtt_view(struct drm_i915_gem_object *obj,
 	if (WARN_ONCE(!view, "no view specified"))
 		return ERR_PTR(-EINVAL);
 
-	list_for_each_entry(vma, &obj->vma_list, vma_link)
+	list_for_each_entry(vma, &obj->vma_list, obj_link)
 		if (vma->vm == ggtt &&
 		    i915_ggtt_view_equal(&vma->ggtt_view, view))
 			return vma;
@@ -4640,19 +4638,16 @@ struct i915_vma *i915_gem_obj_to_ggtt_view(struct drm_i915_gem_object *obj,
 
 void i915_gem_vma_destroy(struct i915_vma *vma)
 {
-	struct i915_address_space *vm = NULL;
 	WARN_ON(vma->node.allocated);
 
 	/* Keep the vma as a placeholder in the execbuffer reservation lists */
 	if (!list_empty(&vma->exec_list))
 		return;
 
-	vm = vma->vm;
-
-	if (!i915_is_ggtt(vm))
-		i915_ppgtt_put(i915_vm_to_ppgtt(vm));
+	if (!vma->is_ggtt)
+		i915_ppgtt_put(i915_vm_to_ppgtt(vma->vm));
 
-	list_del(&vma->vma_link);
+	list_del(&vma->obj_link);
 
 	kmem_cache_free(to_i915(vma->obj->base.dev)->vmas, vma);
 }
@@ -5034,7 +5029,7 @@ init_ring_lists(struct intel_engine_cs *ring)
 }
 
 void
-i915_gem_load(struct drm_device *dev)
+i915_gem_load_init(struct drm_device *dev)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	int i;
@@ -5100,11 +5095,18 @@ i915_gem_load(struct drm_device *dev)
 
 	dev_priv->mm.interruptible = true;
 
-	i915_gem_shrinker_init(dev_priv);
-
 	mutex_init(&dev_priv->fb_tracking.lock);
 }
 
+void i915_gem_load_cleanup(struct drm_device *dev)
+{
+	struct drm_i915_private *dev_priv = to_i915(dev);
+
+	kmem_cache_destroy(dev_priv->requests);
+	kmem_cache_destroy(dev_priv->vmas);
+	kmem_cache_destroy(dev_priv->objects);
+}
+
 void i915_gem_release(struct drm_device *dev, struct drm_file *file)
 {
 	struct drm_i915_file_private *file_priv = file->driver_priv;
@@ -5195,8 +5197,8 @@ u64 i915_gem_obj_offset(struct drm_i915_gem_object *o,
 
 	WARN_ON(vm == &dev_priv->mm.aliasing_ppgtt->base);
 
-	list_for_each_entry(vma, &o->vma_list, vma_link) {
-		if (i915_is_ggtt(vma->vm) &&
+	list_for_each_entry(vma, &o->vma_list, obj_link) {
+		if (vma->is_ggtt &&
 		    vma->ggtt_view.type != I915_GGTT_VIEW_NORMAL)
 			continue;
 		if (vma->vm == vm)
@@ -5214,7 +5216,7 @@ u64 i915_gem_obj_ggtt_offset_view(struct drm_i915_gem_object *o,
 	struct i915_address_space *ggtt = i915_obj_to_ggtt(o);
 	struct i915_vma *vma;
 
-	list_for_each_entry(vma, &o->vma_list, vma_link)
+	list_for_each_entry(vma, &o->vma_list, obj_link)
 		if (vma->vm == ggtt &&
 		    i915_ggtt_view_equal(&vma->ggtt_view, view))
 			return vma->node.start;
@@ -5228,8 +5230,8 @@ bool i915_gem_obj_bound(struct drm_i915_gem_object *o,
 {
 	struct i915_vma *vma;
 
-	list_for_each_entry(vma, &o->vma_list, vma_link) {
-		if (i915_is_ggtt(vma->vm) &&
+	list_for_each_entry(vma, &o->vma_list, obj_link) {
+		if (vma->is_ggtt &&
 		    vma->ggtt_view.type != I915_GGTT_VIEW_NORMAL)
 			continue;
 		if (vma->vm == vm && drm_mm_node_allocated(&vma->node))
@@ -5245,7 +5247,7 @@ bool i915_gem_obj_ggtt_bound_view(struct drm_i915_gem_object *o,
 	struct i915_address_space *ggtt = i915_obj_to_ggtt(o);
 	struct i915_vma *vma;
 
-	list_for_each_entry(vma, &o->vma_list, vma_link)
+	list_for_each_entry(vma, &o->vma_list, obj_link)
 		if (vma->vm == ggtt &&
 		    i915_ggtt_view_equal(&vma->ggtt_view, view) &&
 		    drm_mm_node_allocated(&vma->node))
@@ -5258,7 +5260,7 @@ bool i915_gem_obj_bound_any(struct drm_i915_gem_object *o)
 {
 	struct i915_vma *vma;
 
-	list_for_each_entry(vma, &o->vma_list, vma_link)
+	list_for_each_entry(vma, &o->vma_list, obj_link)
 		if (drm_mm_node_allocated(&vma->node))
 			return true;
 
@@ -5275,8 +5277,8 @@ unsigned long i915_gem_obj_size(struct drm_i915_gem_object *o,
 
 	BUG_ON(list_empty(&o->vma_list));
 
-	list_for_each_entry(vma, &o->vma_list, vma_link) {
-		if (i915_is_ggtt(vma->vm) &&
+	list_for_each_entry(vma, &o->vma_list, obj_link) {
+		if (vma->is_ggtt &&
 		    vma->ggtt_view.type != I915_GGTT_VIEW_NORMAL)
 			continue;
 		if (vma->vm == vm)
@@ -5288,7 +5290,7 @@ unsigned long i915_gem_obj_size(struct drm_i915_gem_object *o,
 bool i915_gem_obj_is_pinned(struct drm_i915_gem_object *obj)
 {
 	struct i915_vma *vma;
-	list_for_each_entry(vma, &obj->vma_list, vma_link)
+	list_for_each_entry(vma, &obj->vma_list, obj_link)
 		if (vma->pin_count > 0)
 			return true;
 
@@ -5302,7 +5304,7 @@ i915_gem_object_get_dirty_page(struct drm_i915_gem_object *obj, int n)
 	struct page *page;
 
 	/* Only default objects have per-page dirty tracking */
-	if (WARN_ON(obj->ops != &i915_gem_object_ops))
+	if (WARN_ON((obj->ops->flags & I915_GEM_OBJECT_HAS_STRUCT_PAGE) == 0))
 		return NULL;
 
 	page = i915_gem_object_get_page(obj, n);
diff --git a/drivers/gpu/drm/i915/i915_gem_context.c b/drivers/gpu/drm/i915/i915_gem_context.c
index 6a4f64b03db6..5dd84e148bba 100644
--- a/drivers/gpu/drm/i915/i915_gem_context.c
+++ b/drivers/gpu/drm/i915/i915_gem_context.c
@@ -142,7 +142,7 @@ static void i915_gem_context_clean(struct intel_context *ctx)
 		return;
 
 	list_for_each_entry_safe(vma, next, &ppgtt->base.inactive_list,
-				 mm_list) {
+				 vm_link) {
 		if (WARN_ON(__i915_vma_unbind_no_wait(vma)))
 			break;
 	}
@@ -321,6 +321,18 @@ err_destroy:
 	return ERR_PTR(ret);
 }
 
+static void i915_gem_context_unpin(struct intel_context *ctx,
+				   struct intel_engine_cs *engine)
+{
+	if (i915.enable_execlists) {
+		intel_lr_context_unpin(ctx, engine);
+	} else {
+		if (engine->id == RCS && ctx->legacy_hw_ctx.rcs_state)
+			i915_gem_object_ggtt_unpin(ctx->legacy_hw_ctx.rcs_state);
+		i915_gem_context_unreference(ctx);
+	}
+}
+
 void i915_gem_context_reset(struct drm_device *dev)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
@@ -329,22 +341,15 @@ void i915_gem_context_reset(struct drm_device *dev)
 	if (i915.enable_execlists) {
 		struct intel_context *ctx;
 
-		list_for_each_entry(ctx, &dev_priv->context_list, link) {
+		list_for_each_entry(ctx, &dev_priv->context_list, link)
 			intel_lr_context_reset(dev, ctx);
-		}
-
-		return;
 	}
 
 	for (i = 0; i < I915_NUM_RINGS; i++) {
 		struct intel_engine_cs *ring = &dev_priv->ring[i];
-		struct intel_context *lctx = ring->last_context;
-
-		if (lctx) {
-			if (lctx->legacy_hw_ctx.rcs_state && i == RCS)
-				i915_gem_object_ggtt_unpin(lctx->legacy_hw_ctx.rcs_state);
 
-			i915_gem_context_unreference(lctx);
+		if (ring->last_context) {
+			i915_gem_context_unpin(ring->last_context, ring);
 			ring->last_context = NULL;
 		}
 	}
@@ -417,13 +422,6 @@ void i915_gem_context_fini(struct drm_device *dev)
 		 * to offset the do_switch part, so that i915_gem_context_unreference()
 		 * can then free the base object correctly. */
 		WARN_ON(!dev_priv->ring[RCS].last_context);
-		if (dev_priv->ring[RCS].last_context == dctx) {
-			/* Fake switch to NULL context */
-			WARN_ON(dctx->legacy_hw_ctx.rcs_state->active);
-			i915_gem_object_ggtt_unpin(dctx->legacy_hw_ctx.rcs_state);
-			i915_gem_context_unreference(dctx);
-			dev_priv->ring[RCS].last_context = NULL;
-		}
 
 		i915_gem_object_ggtt_unpin(dctx->legacy_hw_ctx.rcs_state);
 	}
@@ -432,7 +430,7 @@ void i915_gem_context_fini(struct drm_device *dev)
 		struct intel_engine_cs *ring = &dev_priv->ring[i];
 
 		if (ring->last_context) {
-			i915_gem_context_unreference(ring->last_context);
+			i915_gem_context_unpin(ring->last_context, ring);
 			ring->last_context = NULL;
 		}
 	}
@@ -857,6 +855,9 @@ int i915_gem_context_create_ioctl(struct drm_device *dev, void *data,
 	if (!contexts_enabled(dev))
 		return -ENODEV;
 
+	if (args->pad != 0)
+		return -EINVAL;
+
 	ret = i915_mutex_lock_interruptible(dev);
 	if (ret)
 		return ret;
@@ -880,6 +881,9 @@ int i915_gem_context_destroy_ioctl(struct drm_device *dev, void *data,
 	struct intel_context *ctx;
 	int ret;
 
+	if (args->pad != 0)
+		return -EINVAL;
+
 	if (args->ctx_id == DEFAULT_CONTEXT_HANDLE)
 		return -ENOENT;
 
diff --git a/drivers/gpu/drm/i915/i915_gem_evict.c b/drivers/gpu/drm/i915/i915_gem_evict.c
index 07c6e4d320c9..ea1f8d1bd228 100644
--- a/drivers/gpu/drm/i915/i915_gem_evict.c
+++ b/drivers/gpu/drm/i915/i915_gem_evict.c
@@ -116,7 +116,7 @@ i915_gem_evict_something(struct drm_device *dev, struct i915_address_space *vm,
 
 search_again:
 	/* First see if there is a large enough contiguous idle region... */
-	list_for_each_entry(vma, &vm->inactive_list, mm_list) {
+	list_for_each_entry(vma, &vm->inactive_list, vm_link) {
 		if (mark_free(vma, &unwind_list))
 			goto found;
 	}
@@ -125,7 +125,7 @@ search_again:
 		goto none;
 
 	/* Now merge in the soon-to-be-expired objects... */
-	list_for_each_entry(vma, &vm->active_list, mm_list) {
+	list_for_each_entry(vma, &vm->active_list, vm_link) {
 		if (mark_free(vma, &unwind_list))
 			goto found;
 	}
@@ -270,7 +270,7 @@ int i915_gem_evict_vm(struct i915_address_space *vm, bool do_idle)
 		WARN_ON(!list_empty(&vm->active_list));
 	}
 
-	list_for_each_entry_safe(vma, next, &vm->inactive_list, mm_list)
+	list_for_each_entry_safe(vma, next, &vm->inactive_list, vm_link)
 		if (vma->pin_count == 0)
 			WARN_ON(i915_vma_unbind(vma));
 
diff --git a/drivers/gpu/drm/i915/i915_gem_execbuffer.c b/drivers/gpu/drm/i915/i915_gem_execbuffer.c
index 2dc08ce1079a..1328bc5021b4 100644
--- a/drivers/gpu/drm/i915/i915_gem_execbuffer.c
+++ b/drivers/gpu/drm/i915/i915_gem_execbuffer.c
@@ -668,7 +668,7 @@ need_reloc_mappable(struct i915_vma *vma)
 	if (entry->relocation_count == 0)
 		return false;
 
-	if (!i915_is_ggtt(vma->vm))
+	if (!vma->is_ggtt)
 		return false;
 
 	/* See also use_cpu_reloc() */
@@ -687,8 +687,7 @@ eb_vma_misplaced(struct i915_vma *vma)
 	struct drm_i915_gem_exec_object2 *entry = vma->exec_entry;
 	struct drm_i915_gem_object *obj = vma->obj;
 
-	WARN_ON(entry->flags & __EXEC_OBJECT_NEEDS_MAP &&
-	       !i915_is_ggtt(vma->vm));
+	WARN_ON(entry->flags & __EXEC_OBJECT_NEEDS_MAP && !vma->is_ggtt);
 
 	if (entry->alignment &&
 	    vma->node.start & (entry->alignment - 1))
@@ -1401,6 +1400,7 @@ eb_select_ring(struct drm_i915_private *dev_priv,
 			bsd_idx = gen8_dispatch_bsd_ring(dev_priv, file);
 		} else if (bsd_idx >= I915_EXEC_BSD_RING1 &&
 			   bsd_idx <= I915_EXEC_BSD_RING2) {
+			bsd_idx >>= I915_EXEC_BSD_SHIFT;
 			bsd_idx--;
 		} else {
 			DRM_DEBUG("execbuf with unknown bsd ring: %u\n",
@@ -1654,7 +1654,7 @@ err:
 	 * must be freed again. If it was submitted then it is being tracked
 	 * on the active request list and no clean up is required here.
 	 */
-	if (ret && req)
+	if (ret && !IS_ERR_OR_NULL(req))
 		i915_gem_request_cancel(req);
 
 	mutex_unlock(&dev->struct_mutex);
diff --git a/drivers/gpu/drm/i915/i915_gem_gtt.c b/drivers/gpu/drm/i915/i915_gem_gtt.c
index 7377b6725c33..49e4f26b79d8 100644
--- a/drivers/gpu/drm/i915/i915_gem_gtt.c
+++ b/drivers/gpu/drm/i915/i915_gem_gtt.c
@@ -2132,6 +2132,25 @@ static void i915_address_space_init(struct i915_address_space *vm,
 	list_add_tail(&vm->global_link, &dev_priv->vm_list);
 }
 
+static void gtt_write_workarounds(struct drm_device *dev)
+{
+	struct drm_i915_private *dev_priv = dev->dev_private;
+
+	/* This function is for gtt related workarounds. This function is
+	 * called on driver load and after a GPU reset, so you can place
+	 * workarounds here even if they get overwritten by GPU reset.
+	 */
+	/* WaIncreaseDefaultTLBEntries:chv,bdw,skl,bxt */
+	if (IS_BROADWELL(dev))
+		I915_WRITE(GEN8_L3_LRA_1_GPGPU, GEN8_L3_LRA_1_GPGPU_DEFAULT_VALUE_BDW);
+	else if (IS_CHERRYVIEW(dev))
+		I915_WRITE(GEN8_L3_LRA_1_GPGPU, GEN8_L3_LRA_1_GPGPU_DEFAULT_VALUE_CHV);
+	else if (IS_SKYLAKE(dev))
+		I915_WRITE(GEN8_L3_LRA_1_GPGPU, GEN9_L3_LRA_1_GPGPU_DEFAULT_VALUE_SKL);
+	else if (IS_BROXTON(dev))
+		I915_WRITE(GEN8_L3_LRA_1_GPGPU, GEN9_L3_LRA_1_GPGPU_DEFAULT_VALUE_BXT);
+}
+
 int i915_ppgtt_init(struct drm_device *dev, struct i915_hw_ppgtt *ppgtt)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
@@ -2148,6 +2167,8 @@ int i915_ppgtt_init(struct drm_device *dev, struct i915_hw_ppgtt *ppgtt)
 
 int i915_ppgtt_init_hw(struct drm_device *dev)
 {
+	gtt_write_workarounds(dev);
+
 	/* In the case of execlists, PPGTT is enabled by the context descriptor
 	 * and the PDPs are contained within the context itself.  We don't
 	 * need to do anything here. */
@@ -2737,7 +2758,7 @@ static int i915_gem_setup_global_gtt(struct drm_device *dev,
 		}
 		vma->bound |= GLOBAL_BIND;
 		__i915_vma_set_map_and_fenceable(vma);
-		list_add_tail(&vma->mm_list, &ggtt_vm->inactive_list);
+		list_add_tail(&vma->vm_link, &ggtt_vm->inactive_list);
 	}
 
 	/* Clear any non-preallocated blocks */
@@ -2809,6 +2830,8 @@ void i915_global_gtt_cleanup(struct drm_device *dev)
 		ppgtt->base.cleanup(&ppgtt->base);
 	}
 
+	i915_gem_cleanup_stolen(dev);
+
 	if (drm_mm_initialized(&vm->mm)) {
 		if (intel_vgpu_active(dev))
 			intel_vgt_deballoon();
@@ -3175,12 +3198,21 @@ int i915_gem_gtt_init(struct drm_device *dev)
 	}
 
 	gtt->base.dev = dev;
+	gtt->base.is_ggtt = true;
 
 	ret = gtt->gtt_probe(dev, &gtt->base.total, &gtt->stolen_size,
 			     &gtt->mappable_base, &gtt->mappable_end);
 	if (ret)
 		return ret;
 
+	/*
+	 * Initialise stolen early so that we may reserve preallocated
+	 * objects for the BIOS to KMS transition.
+	 */
+	ret = i915_gem_init_stolen(dev);
+	if (ret)
+		goto out_gtt_cleanup;
+
 	/* GMADR is the PCI mmio aperture into the global GTT. */
 	DRM_INFO("Memory usable by graphics device = %lluM\n",
 		 gtt->base.total >> 20);
@@ -3200,6 +3232,11 @@ int i915_gem_gtt_init(struct drm_device *dev)
 	DRM_DEBUG_DRIVER("ppgtt mode: %i\n", i915.enable_ppgtt);
 
 	return 0;
+
+out_gtt_cleanup:
+	gtt->base.cleanup(&dev_priv->gtt.base);
+
+	return ret;
 }
 
 void i915_gem_restore_gtt_mappings(struct drm_device *dev)
@@ -3222,7 +3259,7 @@ void i915_gem_restore_gtt_mappings(struct drm_device *dev)
 	vm = &dev_priv->gtt.base;
 	list_for_each_entry(obj, &dev_priv->mm.bound_list, global_list) {
 		flush = false;
-		list_for_each_entry(vma, &obj->vma_list, vma_link) {
+		list_for_each_entry(vma, &obj->vma_list, obj_link) {
 			if (vma->vm != vm)
 				continue;
 
@@ -3278,19 +3315,20 @@ __i915_gem_vma_create(struct drm_i915_gem_object *obj,
 	if (vma == NULL)
 		return ERR_PTR(-ENOMEM);
 
-	INIT_LIST_HEAD(&vma->vma_link);
-	INIT_LIST_HEAD(&vma->mm_list);
+	INIT_LIST_HEAD(&vma->vm_link);
+	INIT_LIST_HEAD(&vma->obj_link);
 	INIT_LIST_HEAD(&vma->exec_list);
 	vma->vm = vm;
 	vma->obj = obj;
+	vma->is_ggtt = i915_is_ggtt(vm);
 
 	if (i915_is_ggtt(vm))
 		vma->ggtt_view = *ggtt_view;
-
-	list_add_tail(&vma->vma_link, &obj->vma_list);
-	if (!i915_is_ggtt(vm))
+	else
 		i915_ppgtt_get(i915_vm_to_ppgtt(vm));
 
+	list_add_tail(&vma->obj_link, &obj->vma_list);
+
 	return vma;
 }
 
@@ -3333,6 +3371,7 @@ i915_gem_obj_lookup_or_create_ggtt_vma(struct drm_i915_gem_object *obj,
 static struct scatterlist *
 rotate_pages(const dma_addr_t *in, unsigned int offset,
 	     unsigned int width, unsigned int height,
+	     unsigned int stride,
 	     struct sg_table *st, struct scatterlist *sg)
 {
 	unsigned int column, row;
@@ -3344,7 +3383,7 @@ rotate_pages(const dma_addr_t *in, unsigned int offset,
 	}
 
 	for (column = 0; column < width; column++) {
-		src_idx = width * (height - 1) + column;
+		src_idx = stride * (height - 1) + column;
 		for (row = 0; row < height; row++) {
 			st->nents++;
 			/* We don't need the pages, but need to initialize
@@ -3355,7 +3394,7 @@ rotate_pages(const dma_addr_t *in, unsigned int offset,
 			sg_dma_address(sg) = in[offset + src_idx];
 			sg_dma_len(sg) = PAGE_SIZE;
 			sg = sg_next(sg);
-			src_idx -= width;
+			src_idx -= stride;
 		}
 	}
 
@@ -3363,10 +3402,9 @@ rotate_pages(const dma_addr_t *in, unsigned int offset,
 }
 
 static struct sg_table *
-intel_rotate_fb_obj_pages(struct i915_ggtt_view *ggtt_view,
+intel_rotate_fb_obj_pages(struct intel_rotation_info *rot_info,
 			  struct drm_i915_gem_object *obj)
 {
-	struct intel_rotation_info *rot_info = &ggtt_view->params.rotation_info;
 	unsigned int size_pages = rot_info->size >> PAGE_SHIFT;
 	unsigned int size_pages_uv;
 	struct sg_page_iter sg_iter;
@@ -3408,6 +3446,7 @@ intel_rotate_fb_obj_pages(struct i915_ggtt_view *ggtt_view,
 	/* Rotate the pages. */
 	sg = rotate_pages(page_addr_list, 0,
 		     rot_info->width_pages, rot_info->height_pages,
+		     rot_info->width_pages,
 		     st, NULL);
 
 	/* Append the UV plane if NV12. */
@@ -3423,6 +3462,7 @@ intel_rotate_fb_obj_pages(struct i915_ggtt_view *ggtt_view,
 		rotate_pages(page_addr_list, uv_start_page,
 			     rot_info->width_pages_uv,
 			     rot_info->height_pages_uv,
+			     rot_info->width_pages_uv,
 			     st, sg);
 	}
 
@@ -3504,7 +3544,7 @@ i915_get_ggtt_vma_pages(struct i915_vma *vma)
 		vma->ggtt_view.pages = vma->obj->pages;
 	else if (vma->ggtt_view.type == I915_GGTT_VIEW_ROTATED)
 		vma->ggtt_view.pages =
-			intel_rotate_fb_obj_pages(&vma->ggtt_view, vma->obj);
+			intel_rotate_fb_obj_pages(&vma->ggtt_view.params.rotated, vma->obj);
 	else if (vma->ggtt_view.type == I915_GGTT_VIEW_PARTIAL)
 		vma->ggtt_view.pages =
 			intel_partial_pages(&vma->ggtt_view, vma->obj);
@@ -3560,13 +3600,9 @@ int i915_vma_bind(struct i915_vma *vma, enum i915_cache_level cache_level,
 		return 0;
 
 	if (vma->bound == 0 && vma->vm->allocate_va_range) {
-		trace_i915_va_alloc(vma->vm,
-				    vma->node.start,
-				    vma->node.size,
-				    VM_TO_TRACE_NAME(vma->vm));
-
 		/* XXX: i915_vma_pin() will fix this +- hack */
 		vma->pin_count++;
+		trace_i915_va_alloc(vma);
 		ret = vma->vm->allocate_va_range(vma->vm,
 						 vma->node.start,
 						 vma->node.size);
@@ -3598,7 +3634,7 @@ i915_ggtt_view_size(struct drm_i915_gem_object *obj,
 	if (view->type == I915_GGTT_VIEW_NORMAL) {
 		return obj->base.size;
 	} else if (view->type == I915_GGTT_VIEW_ROTATED) {
-		return view->params.rotation_info.size;
+		return view->params.rotated.size;
 	} else if (view->type == I915_GGTT_VIEW_PARTIAL) {
 		return view->params.partial.size << PAGE_SHIFT;
 	} else {
diff --git a/drivers/gpu/drm/i915/i915_gem_gtt.h b/drivers/gpu/drm/i915/i915_gem_gtt.h
index e5737963ab79..8774f1ba46e7 100644
--- a/drivers/gpu/drm/i915/i915_gem_gtt.h
+++ b/drivers/gpu/drm/i915/i915_gem_gtt.h
@@ -155,7 +155,7 @@ struct i915_ggtt_view {
 			u64 offset;
 			unsigned int size;
 		} partial;
-		struct intel_rotation_info rotation_info;
+		struct intel_rotation_info rotated;
 	} params;
 
 	struct sg_table *pages;
@@ -183,6 +183,7 @@ struct i915_vma {
 #define GLOBAL_BIND	(1<<0)
 #define LOCAL_BIND	(1<<1)
 	unsigned int bound : 4;
+	bool is_ggtt : 1;
 
 	/**
 	 * Support different GGTT views into the same object.
@@ -194,9 +195,9 @@ struct i915_vma {
 	struct i915_ggtt_view ggtt_view;
 
 	/** This object's place on the active/inactive lists */
-	struct list_head mm_list;
+	struct list_head vm_link;
 
-	struct list_head vma_link; /* Link in the object's VMA list */
+	struct list_head obj_link; /* Link in the object's VMA list */
 
 	/** This vma's place in the batchbuffer or on the eviction list */
 	struct list_head exec_list;
@@ -275,6 +276,8 @@ struct i915_address_space {
 	u64 start;		/* Start offset always 0 for dri2 */
 	u64 total;		/* size addr space maps (ex. 2GB for ggtt) */
 
+	bool is_ggtt;
+
 	struct i915_page_scratch *scratch_page;
 	struct i915_page_table *scratch_pt;
 	struct i915_page_directory *scratch_pd;
@@ -330,6 +333,8 @@ struct i915_address_space {
 			u32 flags);
 };
 
+#define i915_is_ggtt(V) ((V)->is_ggtt)
+
 /* The Graphics Translation Table is the way in which GEN hardware translates a
  * Graphics Virtual Address into a Physical Address. In addition to the normal
  * collateral associated with any va->pa translations GEN hardware also has a
@@ -342,6 +347,8 @@ struct i915_gtt {
 
 	size_t stolen_size;		/* Total size of stolen memory */
 	size_t stolen_usable_size;	/* Total size minus BIOS reserved */
+	size_t stolen_reserved_base;
+	size_t stolen_reserved_size;
 	u64 mappable_end;		/* End offset that we can CPU map */
 	struct io_mapping *mappable;	/* Mapping to our CPU mappable region */
 	phys_addr_t mappable_base;	/* PA of our GMADR */
@@ -416,7 +423,7 @@ static inline uint32_t i915_pte_index(uint64_t address, uint32_t pde_shift)
 static inline uint32_t i915_pte_count(uint64_t addr, size_t length,
 				      uint32_t pde_shift)
 {
-	const uint64_t mask = ~((1 << pde_shift) - 1);
+	const uint64_t mask = ~((1ULL << pde_shift) - 1);
 	uint64_t end;
 
 	WARN_ON(length == 0);
diff --git a/drivers/gpu/drm/i915/i915_gem_shrinker.c b/drivers/gpu/drm/i915/i915_gem_shrinker.c
index 16da9c1422cc..d3c473ffb90a 100644
--- a/drivers/gpu/drm/i915/i915_gem_shrinker.c
+++ b/drivers/gpu/drm/i915/i915_gem_shrinker.c
@@ -52,7 +52,7 @@ static int num_vma_bound(struct drm_i915_gem_object *obj)
 	struct i915_vma *vma;
 	int count = 0;
 
-	list_for_each_entry(vma, &obj->vma_list, vma_link) {
+	list_for_each_entry(vma, &obj->vma_list, obj_link) {
 		if (drm_mm_node_allocated(&vma->node))
 			count++;
 		if (vma->pin_count)
@@ -176,7 +176,7 @@ i915_gem_shrink(struct drm_i915_private *dev_priv,
 
 			/* For the unbound phase, this should be a no-op! */
 			list_for_each_entry_safe(vma, v,
-						 &obj->vma_list, vma_link)
+						 &obj->vma_list, obj_link)
 				if (i915_vma_unbind(vma))
 					break;
 
@@ -367,8 +367,20 @@ void i915_gem_shrinker_init(struct drm_i915_private *dev_priv)
 	dev_priv->mm.shrinker.scan_objects = i915_gem_shrinker_scan;
 	dev_priv->mm.shrinker.count_objects = i915_gem_shrinker_count;
 	dev_priv->mm.shrinker.seeks = DEFAULT_SEEKS;
-	register_shrinker(&dev_priv->mm.shrinker);
+	WARN_ON(register_shrinker(&dev_priv->mm.shrinker));
 
 	dev_priv->mm.oom_notifier.notifier_call = i915_gem_shrinker_oom;
-	register_oom_notifier(&dev_priv->mm.oom_notifier);
+	WARN_ON(register_oom_notifier(&dev_priv->mm.oom_notifier));
+}
+
+/**
+ * i915_gem_shrinker_cleanup - Clean up i915 shrinker
+ * @dev_priv: i915 device
+ *
+ * This function unregisters the i915 shrinker and OOM handler.
+ */
+void i915_gem_shrinker_cleanup(struct drm_i915_private *dev_priv)
+{
+	WARN_ON(unregister_oom_notifier(&dev_priv->mm.oom_notifier));
+	unregister_shrinker(&dev_priv->mm.shrinker);
 }
diff --git a/drivers/gpu/drm/i915/i915_gem_stolen.c b/drivers/gpu/drm/i915/i915_gem_stolen.c
index c384dc9c8a63..2e6e9fb6f80d 100644
--- a/drivers/gpu/drm/i915/i915_gem_stolen.c
+++ b/drivers/gpu/drm/i915/i915_gem_stolen.c
@@ -458,6 +458,9 @@ int i915_gem_init_stolen(struct drm_device *dev)
 		return 0;
 	}
 
+	dev_priv->gtt.stolen_reserved_base = reserved_base;
+	dev_priv->gtt.stolen_reserved_size = reserved_size;
+
 	/* It is possible for the reserved area to end before the end of stolen
 	 * memory, so just consider the start. */
 	reserved_total = stolen_top - reserved_base;
@@ -635,6 +638,8 @@ i915_gem_object_create_stolen_for_preallocated(struct drm_device *dev,
 	if (!drm_mm_initialized(&dev_priv->mm.stolen))
 		return NULL;
 
+	lockdep_assert_held(&dev->struct_mutex);
+
 	DRM_DEBUG_KMS("creating preallocated stolen object: stolen_offset=%x, gtt_offset=%x, size=%x\n",
 			stolen_offset, gtt_offset, size);
 
@@ -692,7 +697,7 @@ i915_gem_object_create_stolen_for_preallocated(struct drm_device *dev,
 
 		vma->bound |= GLOBAL_BIND;
 		__i915_vma_set_map_and_fenceable(vma);
-		list_add_tail(&vma->mm_list, &ggtt->inactive_list);
+		list_add_tail(&vma->vm_link, &ggtt->inactive_list);
 	}
 
 	list_add_tail(&obj->global_list, &dev_priv->mm.bound_list);
diff --git a/drivers/gpu/drm/i915/i915_gem_userptr.c b/drivers/gpu/drm/i915/i915_gem_userptr.c
index 19fb0bddc1cd..4b09c840d493 100644
--- a/drivers/gpu/drm/i915/i915_gem_userptr.c
+++ b/drivers/gpu/drm/i915/i915_gem_userptr.c
@@ -49,21 +49,18 @@ struct i915_mmu_notifier {
 	struct hlist_node node;
 	struct mmu_notifier mn;
 	struct rb_root objects;
-	struct list_head linear;
-	bool has_linear;
 };
 
 struct i915_mmu_object {
 	struct i915_mmu_notifier *mn;
+	struct drm_i915_gem_object *obj;
 	struct interval_tree_node it;
 	struct list_head link;
-	struct drm_i915_gem_object *obj;
 	struct work_struct work;
-	bool active;
-	bool is_linear;
+	bool attached;
 };
 
-static void __cancel_userptr__worker(struct work_struct *work)
+static void cancel_userptr(struct work_struct *work)
 {
 	struct i915_mmu_object *mo = container_of(work, typeof(*mo), work);
 	struct drm_i915_gem_object *obj = mo->obj;
@@ -81,7 +78,7 @@ static void __cancel_userptr__worker(struct work_struct *work)
 		was_interruptible = dev_priv->mm.interruptible;
 		dev_priv->mm.interruptible = false;
 
-		list_for_each_entry_safe(vma, tmp, &obj->vma_list, vma_link) {
+		list_for_each_entry_safe(vma, tmp, &obj->vma_list, obj_link) {
 			int ret = i915_vma_unbind(vma);
 			WARN_ON(ret && ret != -EIO);
 		}
@@ -94,24 +91,22 @@ static void __cancel_userptr__worker(struct work_struct *work)
 	mutex_unlock(&dev->struct_mutex);
 }
 
-static unsigned long cancel_userptr(struct i915_mmu_object *mo)
+static void add_object(struct i915_mmu_object *mo)
 {
-	unsigned long end = mo->obj->userptr.ptr + mo->obj->base.size;
-
-	/* The mmu_object is released late when destroying the
-	 * GEM object so it is entirely possible to gain a
-	 * reference on an object in the process of being freed
-	 * since our serialisation is via the spinlock and not
-	 * the struct_mutex - and consequently use it after it
-	 * is freed and then double free it.
-	 */
-	if (mo->active && kref_get_unless_zero(&mo->obj->base.refcount)) {
-		schedule_work(&mo->work);
-		/* only schedule one work packet to avoid the refleak */
-		mo->active = false;
-	}
+	if (mo->attached)
+		return;
+
+	interval_tree_insert(&mo->it, &mo->mn->objects);
+	mo->attached = true;
+}
 
-	return end;
+static void del_object(struct i915_mmu_object *mo)
+{
+	if (!mo->attached)
+		return;
+
+	interval_tree_remove(&mo->it, &mo->mn->objects);
+	mo->attached = false;
 }
 
 static void i915_gem_userptr_mn_invalidate_range_start(struct mmu_notifier *_mn,
@@ -122,28 +117,36 @@ static void i915_gem_userptr_mn_invalidate_range_start(struct mmu_notifier *_mn,
 	struct i915_mmu_notifier *mn =
 		container_of(_mn, struct i915_mmu_notifier, mn);
 	struct i915_mmu_object *mo;
+	struct interval_tree_node *it;
+	LIST_HEAD(cancelled);
+
+	if (RB_EMPTY_ROOT(&mn->objects))
+		return;
 
 	/* interval ranges are inclusive, but invalidate range is exclusive */
 	end--;
 
 	spin_lock(&mn->lock);
-	if (mn->has_linear) {
-		list_for_each_entry(mo, &mn->linear, link) {
-			if (mo->it.last < start || mo->it.start > end)
-				continue;
-
-			cancel_userptr(mo);
-		}
-	} else {
-		struct interval_tree_node *it;
+	it = interval_tree_iter_first(&mn->objects, start, end);
+	while (it) {
+		/* The mmu_object is released late when destroying the
+		 * GEM object so it is entirely possible to gain a
+		 * reference on an object in the process of being freed
+		 * since our serialisation is via the spinlock and not
+		 * the struct_mutex - and consequently use it after it
+		 * is freed and then double free it. To prevent that
+		 * use-after-free we only acquire a reference on the
+		 * object if it is not in the process of being destroyed.
+		 */
+		mo = container_of(it, struct i915_mmu_object, it);
+		if (kref_get_unless_zero(&mo->obj->base.refcount))
+			schedule_work(&mo->work);
 
-		it = interval_tree_iter_first(&mn->objects, start, end);
-		while (it) {
-			mo = container_of(it, struct i915_mmu_object, it);
-			start = cancel_userptr(mo);
-			it = interval_tree_iter_next(it, start, end);
-		}
+		list_add(&mo->link, &cancelled);
+		it = interval_tree_iter_next(it, start, end);
 	}
+	list_for_each_entry(mo, &cancelled, link)
+		del_object(mo);
 	spin_unlock(&mn->lock);
 }
 
@@ -164,8 +167,6 @@ i915_mmu_notifier_create(struct mm_struct *mm)
 	spin_lock_init(&mn->lock);
 	mn->mn.ops = &i915_gem_userptr_notifier;
 	mn->objects = RB_ROOT;
-	INIT_LIST_HEAD(&mn->linear);
-	mn->has_linear = false;
 
 	 /* Protected by mmap_sem (write-lock) */
 	ret = __mmu_notifier_register(&mn->mn, mm);
@@ -177,85 +178,6 @@ i915_mmu_notifier_create(struct mm_struct *mm)
 	return mn;
 }
 
-static int
-i915_mmu_notifier_add(struct drm_device *dev,
-		      struct i915_mmu_notifier *mn,
-		      struct i915_mmu_object *mo)
-{
-	struct interval_tree_node *it;
-	int ret = 0;
-
-	/* By this point we have already done a lot of expensive setup that
-	 * we do not want to repeat just because the caller (e.g. X) has a
-	 * signal pending (and partly because of that expensive setup, X
-	 * using an interrupt timer is likely to get stuck in an EINTR loop).
-	 */
-	mutex_lock(&dev->struct_mutex);
-
-	/* Make sure we drop the final active reference (and thereby
-	 * remove the objects from the interval tree) before we do
-	 * the check for overlapping objects.
-	 */
-	i915_gem_retire_requests(dev);
-
-	spin_lock(&mn->lock);
-	it = interval_tree_iter_first(&mn->objects,
-				      mo->it.start, mo->it.last);
-	if (it) {
-		struct drm_i915_gem_object *obj;
-
-		/* We only need to check the first object in the range as it
-		 * either has cancelled gup work queued and we need to
-		 * return back to the user to give time for the gup-workers
-		 * to flush their object references upon which the object will
-		 * be removed from the interval-tree, or the the range is
-		 * still in use by another client and the overlap is invalid.
-		 *
-		 * If we do have an overlap, we cannot use the interval tree
-		 * for fast range invalidation.
-		 */
-
-		obj = container_of(it, struct i915_mmu_object, it)->obj;
-		if (!obj->userptr.workers)
-			mn->has_linear = mo->is_linear = true;
-		else
-			ret = -EAGAIN;
-	} else
-		interval_tree_insert(&mo->it, &mn->objects);
-
-	if (ret == 0)
-		list_add(&mo->link, &mn->linear);
-
-	spin_unlock(&mn->lock);
-	mutex_unlock(&dev->struct_mutex);
-
-	return ret;
-}
-
-static bool i915_mmu_notifier_has_linear(struct i915_mmu_notifier *mn)
-{
-	struct i915_mmu_object *mo;
-
-	list_for_each_entry(mo, &mn->linear, link)
-		if (mo->is_linear)
-			return true;
-
-	return false;
-}
-
-static void
-i915_mmu_notifier_del(struct i915_mmu_notifier *mn,
-		      struct i915_mmu_object *mo)
-{
-	spin_lock(&mn->lock);
-	list_del(&mo->link);
-	if (mo->is_linear)
-		mn->has_linear = i915_mmu_notifier_has_linear(mn);
-	else
-		interval_tree_remove(&mo->it, &mn->objects);
-	spin_unlock(&mn->lock);
-}
-
 static void
 i915_gem_userptr_release__mmu_notifier(struct drm_i915_gem_object *obj)
 {
@@ -265,7 +187,9 @@ i915_gem_userptr_release__mmu_notifier(struct drm_i915_gem_object *obj)
 	if (mo == NULL)
 		return;
 
-	i915_mmu_notifier_del(mo->mn, mo);
+	spin_lock(&mo->mn->lock);
+	del_object(mo);
+	spin_unlock(&mo->mn->lock);
 	kfree(mo);
 
 	obj->userptr.mmu_object = NULL;
@@ -299,7 +223,6 @@ i915_gem_userptr_init__mmu_notifier(struct drm_i915_gem_object *obj,
 {
 	struct i915_mmu_notifier *mn;
 	struct i915_mmu_object *mo;
-	int ret;
 
 	if (flags & I915_USERPTR_UNSYNCHRONIZED)
 		return capable(CAP_SYS_ADMIN) ? 0 : -EPERM;
@@ -316,16 +239,10 @@ i915_gem_userptr_init__mmu_notifier(struct drm_i915_gem_object *obj,
 		return -ENOMEM;
 
 	mo->mn = mn;
-	mo->it.start = obj->userptr.ptr;
-	mo->it.last = mo->it.start + obj->base.size - 1;
 	mo->obj = obj;
-	INIT_WORK(&mo->work, __cancel_userptr__worker);
-
-	ret = i915_mmu_notifier_add(obj->base.dev, mn, mo);
-	if (ret) {
-		kfree(mo);
-		return ret;
-	}
+	mo->it.start = obj->userptr.ptr;
+	mo->it.last = obj->userptr.ptr + obj->base.size - 1;
+	INIT_WORK(&mo->work, cancel_userptr);
 
 	obj->userptr.mmu_object = mo;
 	return 0;
@@ -552,8 +469,10 @@ __i915_gem_userptr_set_active(struct drm_i915_gem_object *obj,
 	/* In order to serialise get_pages with an outstanding
 	 * cancel_userptr, we must drop the struct_mutex and try again.
 	 */
-	if (!value || !work_pending(&obj->userptr.mmu_object->work))
-		obj->userptr.mmu_object->active = value;
+	if (!value)
+		del_object(obj->userptr.mmu_object);
+	else if (!work_pending(&obj->userptr.mmu_object->work))
+		add_object(obj->userptr.mmu_object);
 	else
 		ret = -EAGAIN;
 	spin_unlock(&obj->userptr.mmu_object->mn->lock);
@@ -789,9 +708,10 @@ i915_gem_userptr_dmabuf_export(struct drm_i915_gem_object *obj)
 }
 
 static const struct drm_i915_gem_object_ops i915_gem_userptr_ops = {
-	.dmabuf_export = i915_gem_userptr_dmabuf_export,
+	.flags = I915_GEM_OBJECT_HAS_STRUCT_PAGE,
 	.get_pages = i915_gem_userptr_get_pages,
 	.put_pages = i915_gem_userptr_put_pages,
+	.dmabuf_export = i915_gem_userptr_dmabuf_export,
 	.release = i915_gem_userptr_release,
 };
 
diff --git a/drivers/gpu/drm/i915/i915_gpu_error.c b/drivers/gpu/drm/i915/i915_gpu_error.c
index 7eeb24427785..831895b8cb75 100644
--- a/drivers/gpu/drm/i915/i915_gpu_error.c
+++ b/drivers/gpu/drm/i915/i915_gpu_error.c
@@ -365,6 +365,10 @@ int i915_error_state_to_str(struct drm_i915_error_state_buf *m,
 	err_printf(m, "Reset count: %u\n", error->reset_count);
 	err_printf(m, "Suspend count: %u\n", error->suspend_count);
 	err_printf(m, "PCI ID: 0x%04x\n", dev->pdev->device);
+	err_printf(m, "PCI Revision: 0x%02x\n", dev->pdev->revision);
+	err_printf(m, "PCI Subsystem: %04x:%04x\n",
+		   dev->pdev->subsystem_vendor,
+		   dev->pdev->subsystem_device);
 	err_printf(m, "IOMMU enabled?: %d\n", error->iommu);
 
 	if (HAS_CSR(dev)) {
@@ -732,7 +736,7 @@ static u32 capture_active_bo(struct drm_i915_error_buffer *err,
 	struct i915_vma *vma;
 	int i = 0;
 
-	list_for_each_entry(vma, head, mm_list) {
+	list_for_each_entry(vma, head, vm_link) {
 		capture_bo(err++, vma);
 		if (++i == count)
 			break;
@@ -755,7 +759,7 @@ static u32 capture_pinned_bo(struct drm_i915_error_buffer *err,
 		if (err == last)
 			break;
 
-		list_for_each_entry(vma, &obj->vma_list, vma_link)
+		list_for_each_entry(vma, &obj->vma_list, obj_link)
 			if (vma->vm == vm && vma->pin_count > 0)
 				capture_bo(err++, vma);
 	}
@@ -1123,12 +1127,12 @@ static void i915_gem_capture_vm(struct drm_i915_private *dev_priv,
 	int i;
 
 	i = 0;
-	list_for_each_entry(vma, &vm->active_list, mm_list)
+	list_for_each_entry(vma, &vm->active_list, vm_link)
 		i++;
 	error->active_bo_count[ndx] = i;
 
 	list_for_each_entry(obj, &dev_priv->mm.bound_list, global_list) {
-		list_for_each_entry(vma, &obj->vma_list, vma_link)
+		list_for_each_entry(vma, &obj->vma_list, obj_link)
 			if (vma->vm == vm && vma->pin_count > 0)
 				i++;
 	}
diff --git a/drivers/gpu/drm/i915/i915_guc_submission.c b/drivers/gpu/drm/i915/i915_guc_submission.c
index 51ae5c1f806d..d7543efc8a5e 100644
--- a/drivers/gpu/drm/i915/i915_guc_submission.c
+++ b/drivers/gpu/drm/i915/i915_guc_submission.c
@@ -376,6 +376,8 @@ static void guc_init_proc_desc(struct intel_guc *guc,
 static void guc_init_ctx_desc(struct intel_guc *guc,
 			      struct i915_guc_client *client)
 {
+	struct drm_i915_private *dev_priv = guc_to_i915(guc);
+	struct intel_engine_cs *ring;
 	struct intel_context *ctx = client->owner;
 	struct guc_context_desc desc;
 	struct sg_table *sg;
@@ -388,10 +390,8 @@ static void guc_init_ctx_desc(struct intel_guc *guc,
 	desc.priority = client->priority;
 	desc.db_id = client->doorbell_id;
 
-	for (i = 0; i < I915_NUM_RINGS; i++) {
-		struct guc_execlist_context *lrc = &desc.lrc[i];
-		struct intel_ringbuffer *ringbuf = ctx->engine[i].ringbuf;
-		struct intel_engine_cs *ring;
+	for_each_ring(ring, dev_priv, i) {
+		struct guc_execlist_context *lrc = &desc.lrc[ring->guc_id];
 		struct drm_i915_gem_object *obj;
 		uint64_t ctx_desc;
 
@@ -406,7 +406,6 @@ static void guc_init_ctx_desc(struct intel_guc *guc,
 		if (!obj)
 			break;	/* XXX: continue? */
 
-		ring = ringbuf->ring;
 		ctx_desc = intel_lr_context_descriptor(ctx, ring);
 		lrc->context_desc = (u32)ctx_desc;
 
@@ -414,16 +413,16 @@ static void guc_init_ctx_desc(struct intel_guc *guc,
 		lrc->ring_lcra = i915_gem_obj_ggtt_offset(obj) +
 				LRC_STATE_PN * PAGE_SIZE;
 		lrc->context_id = (client->ctx_index << GUC_ELC_CTXID_OFFSET) |
-				(ring->id << GUC_ELC_ENGINE_OFFSET);
+				(ring->guc_id << GUC_ELC_ENGINE_OFFSET);
 
-		obj = ringbuf->obj;
+		obj = ctx->engine[i].ringbuf->obj;
 
 		lrc->ring_begin = i915_gem_obj_ggtt_offset(obj);
 		lrc->ring_end = lrc->ring_begin + obj->base.size - 1;
 		lrc->ring_next_free_location = lrc->ring_begin;
 		lrc->ring_current_tail_pointer_value = 0;
 
-		desc.engines_used |= (1 << ring->id);
+		desc.engines_used |= (1 << ring->guc_id);
 	}
 
 	WARN_ON(desc.engines_used == 0);
@@ -510,7 +509,6 @@ int i915_guc_wq_check_space(struct i915_guc_client *gc)
 static int guc_add_workqueue_item(struct i915_guc_client *gc,
 				  struct drm_i915_gem_request *rq)
 {
-	enum intel_ring_id ring_id = rq->ring->id;
 	struct guc_wq_item *wqi;
 	void *base;
 	u32 tail, wq_len, wq_off, space;
@@ -544,7 +542,7 @@ static int guc_add_workqueue_item(struct i915_guc_client *gc,
 	wq_len = sizeof(struct guc_wq_item) / sizeof(u32) - 1;
 	wqi->header = WQ_TYPE_INORDER |
 			(wq_len << WQ_LEN_SHIFT) |
-			(ring_id << WQ_TARGET_SHIFT) |
+			(rq->ring->guc_id << WQ_TARGET_SHIFT) |
 			WQ_NO_WCFLUSH_WAIT;
 
 	/* The GuC wants only the low-order word of the context descriptor */
@@ -560,29 +558,6 @@ static int guc_add_workqueue_item(struct i915_guc_client *gc,
 	return 0;
 }
 
-#define CTX_RING_BUFFER_START		0x08
-
-/* Update the ringbuffer pointer in a saved context image */
-static void lr_context_update(struct drm_i915_gem_request *rq)
-{
-	enum intel_ring_id ring_id = rq->ring->id;
-	struct drm_i915_gem_object *ctx_obj = rq->ctx->engine[ring_id].state;
-	struct drm_i915_gem_object *rb_obj = rq->ringbuf->obj;
-	struct page *page;
-	uint32_t *reg_state;
-
-	BUG_ON(!ctx_obj);
-	WARN_ON(!i915_gem_obj_is_pinned(ctx_obj));
-	WARN_ON(!i915_gem_obj_is_pinned(rb_obj));
-
-	page = i915_gem_object_get_dirty_page(ctx_obj, LRC_STATE_PN);
-	reg_state = kmap_atomic(page);
-
-	reg_state[CTX_RING_BUFFER_START+1] = i915_gem_obj_ggtt_offset(rb_obj);
-
-	kunmap_atomic(reg_state);
-}
-
 /**
  * i915_guc_submit() - Submit commands through GuC
  * @client:	the guc client where commands will go through
@@ -594,18 +569,14 @@ int i915_guc_submit(struct i915_guc_client *client,
 		    struct drm_i915_gem_request *rq)
 {
 	struct intel_guc *guc = client->guc;
-	enum intel_ring_id ring_id = rq->ring->id;
+	unsigned int engine_id = rq->ring->guc_id;
 	int q_ret, b_ret;
 
-	/* Need this because of the deferred pin ctx and ring */
-	/* Shall we move this right after ring is pinned? */
-	lr_context_update(rq);
-
 	q_ret = guc_add_workqueue_item(client, rq);
 	if (q_ret == 0)
 		b_ret = guc_ring_doorbell(client);
 
-	client->submissions[ring_id] += 1;
+	client->submissions[engine_id] += 1;
 	if (q_ret) {
 		client->q_fail += 1;
 		client->retcode = q_ret;
@@ -615,8 +586,8 @@ int i915_guc_submit(struct i915_guc_client *client,
 	} else {
 		client->retcode = 0;
 	}
-	guc->submissions[ring_id] += 1;
-	guc->last_seqno[ring_id] = rq->seqno;
+	guc->submissions[engine_id] += 1;
+	guc->last_seqno[engine_id] = rq->seqno;
 
 	return q_ret;
 }
@@ -848,7 +819,7 @@ static void init_guc_policies(struct guc_policies *policies)
 	policies->max_num_work_items = POLICY_MAX_NUM_WI;
 
 	for (p = 0; p < GUC_CTX_PRIORITY_NUM; p++) {
-		for (i = 0; i < I915_NUM_RINGS; i++) {
+		for (i = GUC_RENDER_ENGINE; i < GUC_MAX_ENGINES_NUM; i++) {
 			policy = &policies->policy[p][i];
 
 			policy->execution_quantum = 1000000;
@@ -900,7 +871,7 @@ static void guc_create_ads(struct intel_guc *guc)
 	ads->golden_context_lrca = ring->status_page.gfx_addr;
 
 	for_each_ring(ring, dev_priv, i)
-		ads->eng_state_size[i] = intel_lr_context_size(ring);
+		ads->eng_state_size[ring->guc_id] = intel_lr_context_size(ring);
 
 	/* GuC scheduling policies */
 	policies = (void *)ads + sizeof(struct guc_ads);
@@ -912,12 +883,12 @@ static void guc_create_ads(struct intel_guc *guc)
 	/* MMIO reg state */
 	reg_state = (void *)policies + sizeof(struct guc_policies);
 
-	for (i = 0; i < I915_NUM_RINGS; i++) {
-		reg_state->mmio_white_list[i].mmio_start =
-			dev_priv->ring[i].mmio_base + GUC_MMIO_WHITE_LIST_START;
+	for_each_ring(ring, dev_priv, i) {
+		reg_state->mmio_white_list[ring->guc_id].mmio_start =
+			ring->mmio_base + GUC_MMIO_WHITE_LIST_START;
 
 		/* Nothing to be saved or restored for now. */
-		reg_state->mmio_white_list[i].count = 0;
+		reg_state->mmio_white_list[ring->guc_id].count = 0;
 	}
 
 	ads->reg_state_addr = ads->scheduler_policies +
diff --git a/drivers/gpu/drm/i915/i915_irq.c b/drivers/gpu/drm/i915/i915_irq.c
index 25a89373df63..d1a46ef5ab3f 100644
--- a/drivers/gpu/drm/i915/i915_irq.c
+++ b/drivers/gpu/drm/i915/i915_irq.c
@@ -1651,6 +1651,12 @@ static void valleyview_pipestat_irq_handler(struct drm_device *dev, u32 iir)
 	int pipe;
 
 	spin_lock(&dev_priv->irq_lock);
+
+	if (!dev_priv->display_irqs_enabled) {
+		spin_unlock(&dev_priv->irq_lock);
+		return;
+	}
+
 	for_each_pipe(dev_priv, pipe) {
 		i915_reg_t reg;
 		u32 mask, iir_bit = 0;
@@ -3343,21 +3349,28 @@ void gen8_irq_power_well_post_enable(struct drm_i915_private *dev_priv,
 				     unsigned int pipe_mask)
 {
 	uint32_t extra_ier = GEN8_PIPE_VBLANK | GEN8_PIPE_FIFO_UNDERRUN;
+	enum pipe pipe;
+
+	spin_lock_irq(&dev_priv->irq_lock);
+	for_each_pipe_masked(dev_priv, pipe, pipe_mask)
+		GEN8_IRQ_INIT_NDX(DE_PIPE, pipe,
+				  dev_priv->de_irq_mask[pipe],
+				  ~dev_priv->de_irq_mask[pipe] | extra_ier);
+	spin_unlock_irq(&dev_priv->irq_lock);
+}
+
+void gen8_irq_power_well_pre_disable(struct drm_i915_private *dev_priv,
+				     unsigned int pipe_mask)
+{
+	enum pipe pipe;
 
 	spin_lock_irq(&dev_priv->irq_lock);
-	if (pipe_mask & 1 << PIPE_A)
-		GEN8_IRQ_INIT_NDX(DE_PIPE, PIPE_A,
-				  dev_priv->de_irq_mask[PIPE_A],
-				  ~dev_priv->de_irq_mask[PIPE_A] | extra_ier);
-	if (pipe_mask & 1 << PIPE_B)
-		GEN8_IRQ_INIT_NDX(DE_PIPE, PIPE_B,
-				  dev_priv->de_irq_mask[PIPE_B],
-				  ~dev_priv->de_irq_mask[PIPE_B] | extra_ier);
-	if (pipe_mask & 1 << PIPE_C)
-		GEN8_IRQ_INIT_NDX(DE_PIPE, PIPE_C,
-				  dev_priv->de_irq_mask[PIPE_C],
-				  ~dev_priv->de_irq_mask[PIPE_C] | extra_ier);
+	for_each_pipe_masked(dev_priv, pipe, pipe_mask)
+		GEN8_IRQ_RESET_NDX(DE_PIPE, pipe);
 	spin_unlock_irq(&dev_priv->irq_lock);
+
+	/* make sure we're done processing display irqs */
+	synchronize_irq(dev_priv->dev->irq);
 }
 
 static void cherryview_irq_preinstall(struct drm_device *dev)
diff --git a/drivers/gpu/drm/i915/i915_params.c b/drivers/gpu/drm/i915/i915_params.c
index 8d90c256520a..278c9c40c2e0 100644
--- a/drivers/gpu/drm/i915/i915_params.c
+++ b/drivers/gpu/drm/i915/i915_params.c
@@ -38,7 +38,7 @@ struct i915_params i915 __read_mostly = {
 	.enable_execlists = -1,
 	.enable_hangcheck = true,
 	.enable_ppgtt = -1,
-	.enable_psr = 0,
+	.enable_psr = -1,
 	.preliminary_hw_support = IS_ENABLED(CONFIG_DRM_I915_PRELIMINARY_HW_SUPPORT),
 	.disable_power_well = -1,
 	.enable_ips = 1,
@@ -49,7 +49,6 @@ struct i915_params i915 __read_mostly = {
 	.invert_brightness = 0,
 	.disable_display = 0,
 	.enable_cmd_parser = 1,
-	.disable_vtd_wa = 0,
 	.use_mmio_flip = 0,
 	.mmio_debug = 0,
 	.verbose_state_checks = 1,
@@ -92,7 +91,7 @@ MODULE_PARM_DESC(enable_fbc,
 	"Enable frame buffer compression for power savings "
 	"(default: -1 (use per-chip default))");
 
-module_param_named_unsafe(lvds_channel_mode, i915.lvds_channel_mode, int, 0600);
+module_param_named_unsafe(lvds_channel_mode, i915.lvds_channel_mode, int, 0400);
 MODULE_PARM_DESC(lvds_channel_mode,
 	 "Specify LVDS channel mode "
 	 "(0=probe BIOS [default], 1=single-channel, 2=dual-channel)");
@@ -102,7 +101,7 @@ MODULE_PARM_DESC(lvds_use_ssc,
 	"Use Spread Spectrum Clock with panels [LVDS/eDP] "
 	"(default: auto from VBT)");
 
-module_param_named_unsafe(vbt_sdvo_panel_type, i915.vbt_sdvo_panel_type, int, 0600);
+module_param_named_unsafe(vbt_sdvo_panel_type, i915.vbt_sdvo_panel_type, int, 0400);
 MODULE_PARM_DESC(vbt_sdvo_panel_type,
 	"Override/Ignore selection of SDVO panel mode in the VBT "
 	"(-2=ignore, -1=auto [default], index in VBT BIOS table)");
@@ -127,9 +126,11 @@ MODULE_PARM_DESC(enable_execlists,
 	"(-1=auto [default], 0=disabled, 1=enabled)");
 
 module_param_named_unsafe(enable_psr, i915.enable_psr, int, 0600);
-MODULE_PARM_DESC(enable_psr, "Enable PSR (default: false)");
+MODULE_PARM_DESC(enable_psr, "Enable PSR "
+		 "(0=disabled, 1=enabled - link mode chosen per-platform, 2=force link-standby mode, 3=force link-off mode) "
+		 "Default: -1 (use per-chip default)");
 
-module_param_named_unsafe(preliminary_hw_support, i915.preliminary_hw_support, int, 0600);
+module_param_named_unsafe(preliminary_hw_support, i915.preliminary_hw_support, int, 0400);
 MODULE_PARM_DESC(preliminary_hw_support,
 	"Enable preliminary hardware support.");
 
@@ -163,12 +164,9 @@ MODULE_PARM_DESC(invert_brightness,
 	"to dri-devel@lists.freedesktop.org, if your machine needs it. "
 	"It will then be included in an upcoming module version.");
 
-module_param_named(disable_display, i915.disable_display, bool, 0600);
+module_param_named(disable_display, i915.disable_display, bool, 0400);
 MODULE_PARM_DESC(disable_display, "Disable display (default: false)");
 
-module_param_named_unsafe(disable_vtd_wa, i915.disable_vtd_wa, bool, 0600);
-MODULE_PARM_DESC(disable_vtd_wa, "Disable all VT-d workarounds (default: false)");
-
 module_param_named_unsafe(enable_cmd_parser, i915.enable_cmd_parser, int, 0600);
 MODULE_PARM_DESC(enable_cmd_parser,
 		 "Enable command parsing (1=enabled [default], 0=disabled)");
diff --git a/drivers/gpu/drm/i915/i915_params.h b/drivers/gpu/drm/i915/i915_params.h
index 529929073120..bd5026b15d3e 100644
--- a/drivers/gpu/drm/i915/i915_params.h
+++ b/drivers/gpu/drm/i915/i915_params.h
@@ -56,7 +56,6 @@ struct i915_params {
 	bool load_detect_test;
 	bool reset;
 	bool disable_display;
-	bool disable_vtd_wa;
 	bool enable_guc_submission;
 	bool verbose_state_checks;
 	bool nuclear_pageflip;
diff --git a/drivers/gpu/drm/i915/i915_reg.h b/drivers/gpu/drm/i915/i915_reg.h
index 0a988895165f..f76cbf3e5d1e 100644
--- a/drivers/gpu/drm/i915/i915_reg.h
+++ b/drivers/gpu/drm/i915/i915_reg.h
@@ -610,16 +610,17 @@ static inline bool i915_mmio_reg_valid(i915_reg_t reg)
 #define   IOSF_BYTE_ENABLES_SHIFT		4
 #define   IOSF_BAR_SHIFT			1
 #define   IOSF_SB_BUSY				(1<<0)
-#define   IOSF_PORT_BUNIT			0x3
-#define   IOSF_PORT_PUNIT			0x4
+#define   IOSF_PORT_BUNIT			0x03
+#define   IOSF_PORT_PUNIT			0x04
 #define   IOSF_PORT_NC				0x11
 #define   IOSF_PORT_DPIO			0x12
-#define   IOSF_PORT_DPIO_2			0x1a
 #define   IOSF_PORT_GPIO_NC			0x13
 #define   IOSF_PORT_CCK				0x14
-#define   IOSF_PORT_CCU				0xA9
-#define   IOSF_PORT_GPS_CORE			0x48
-#define   IOSF_PORT_FLISDSI			0x1B
+#define   IOSF_PORT_DPIO_2			0x1a
+#define   IOSF_PORT_FLISDSI			0x1b
+#define   IOSF_PORT_GPIO_SC			0x48
+#define   IOSF_PORT_GPIO_SUS			0xa8
+#define   IOSF_PORT_CCU				0xa9
 #define VLV_IOSF_DATA				_MMIO(VLV_DISPLAY_BASE + 0x2104)
 #define VLV_IOSF_ADDR				_MMIO(VLV_DISPLAY_BASE + 0x2108)
 
@@ -1635,6 +1636,9 @@ enum skl_disp_power_wells {
 #define   RING_WAIT		(1<<11) /* gen3+, PRBx_CTL */
 #define   RING_WAIT_SEMAPHORE	(1<<10) /* gen6+ */
 
+#define RING_FORCE_TO_NONPRIV(base, i) _MMIO(((base)+0x4D0) + (i)*4)
+#define   RING_MAX_NONPRIV_SLOTS  12
+
 #define GEN7_TLB_RD_ADDR	_MMIO(0x4700)
 
 #if 0
@@ -3292,19 +3296,20 @@ enum skl_disp_power_wells {
 
 #define PORT_HOTPLUG_STAT	_MMIO(dev_priv->info.display_mmio_offset + 0x61114)
 /*
- * HDMI/DP bits are gen4+
+ * HDMI/DP bits are g4x+
  *
  * WARNING: Bspec for hpd status bits on gen4 seems to be completely confused.
  * Please check the detailed lore in the commit message for for experimental
  * evidence.
  */
-#define   PORTD_HOTPLUG_LIVE_STATUS_G4X		(1 << 29)
+/* Bspec says GM45 should match G4X/VLV/CHV, but reality disagrees */
+#define   PORTD_HOTPLUG_LIVE_STATUS_GM45	(1 << 29)
+#define   PORTC_HOTPLUG_LIVE_STATUS_GM45	(1 << 28)
+#define   PORTB_HOTPLUG_LIVE_STATUS_GM45	(1 << 27)
+/* G4X/VLV/CHV DP/HDMI bits again match Bspec */
+#define   PORTD_HOTPLUG_LIVE_STATUS_G4X		(1 << 27)
 #define   PORTC_HOTPLUG_LIVE_STATUS_G4X		(1 << 28)
-#define   PORTB_HOTPLUG_LIVE_STATUS_G4X		(1 << 27)
-/* VLV DP/HDMI bits again match Bspec */
-#define   PORTD_HOTPLUG_LIVE_STATUS_VLV		(1 << 27)
-#define   PORTC_HOTPLUG_LIVE_STATUS_VLV		(1 << 28)
-#define   PORTB_HOTPLUG_LIVE_STATUS_VLV		(1 << 29)
+#define   PORTB_HOTPLUG_LIVE_STATUS_G4X		(1 << 29)
 #define   PORTD_HOTPLUG_INT_STATUS		(3 << 21)
 #define   PORTD_HOTPLUG_INT_LONG_PULSE		(2 << 21)
 #define   PORTD_HOTPLUG_INT_SHORT_PULSE		(1 << 21)
@@ -5945,6 +5950,7 @@ enum skl_disp_power_wells {
 #define  ILK_INTERNAL_GRAPHICS_DISABLE	(1 << 31)
 #define  ILK_INTERNAL_DISPLAY_DISABLE	(1 << 30)
 #define  ILK_DISPLAY_DEBUG_DISABLE	(1 << 29)
+#define  IVB_PIPE_C_DISABLE		(1 << 28)
 #define  ILK_HDCP_DISABLE		(1 << 25)
 #define  ILK_eDP_A_DISABLE		(1 << 24)
 #define  HSW_CDCLK_LIMIT		(1 << 24)
@@ -5991,10 +5997,19 @@ enum skl_disp_power_wells {
 #define SKL_DFSM_CDCLK_LIMIT_540	(1 << 23)
 #define SKL_DFSM_CDCLK_LIMIT_450	(2 << 23)
 #define SKL_DFSM_CDCLK_LIMIT_337_5	(3 << 23)
+#define SKL_DFSM_PIPE_A_DISABLE		(1 << 30)
+#define SKL_DFSM_PIPE_B_DISABLE		(1 << 21)
+#define SKL_DFSM_PIPE_C_DISABLE		(1 << 28)
+
+#define GEN7_FF_SLICE_CS_CHICKEN1	_MMIO(0x20e0)
+#define   GEN9_FFSC_PERCTX_PREEMPT_CTRL	(1<<14)
 
 #define FF_SLICE_CS_CHICKEN2			_MMIO(0x20e4)
 #define  GEN9_TSG_BARRIER_ACK_DISABLE		(1<<8)
 
+#define GEN9_CS_DEBUG_MODE1		_MMIO(0x20ec)
+#define GEN8_CS_CHICKEN1		_MMIO(0x2580)
+
 /* GEN7 chicken */
 #define GEN7_COMMON_SLICE_CHICKEN1		_MMIO(0x7010)
 # define GEN7_CSC1_RHWO_OPT_DISABLE_IN_RCC	((1<<10) | (1<<26))
@@ -6040,6 +6055,8 @@ enum skl_disp_power_wells {
 #define  HDC_FORCE_NON_COHERENT			(1<<4)
 #define  HDC_BARRIER_PERFORMANCE_DISABLE	(1<<10)
 
+#define GEN8_HDC_CHICKEN1			_MMIO(0x7304)
+
 /* GEN9 chicken */
 #define SLICE_ECO_CHICKEN0			_MMIO(0x7308)
 #define   PIXEL_MASK_CAMMING_DISABLE		(1 << 14)
@@ -6770,6 +6787,16 @@ enum skl_disp_power_wells {
 
 #define  VLV_PMWGICZ				_MMIO(0x1300a4)
 
+#define  RC6_LOCATION				_MMIO(0xD40)
+#define	   RC6_CTX_IN_DRAM			(1 << 0)
+#define  RC6_CTX_BASE				_MMIO(0xD48)
+#define    RC6_CTX_BASE_MASK			0xFFFFFFF0
+#define  PWRCTX_MAXCNT_RCSUNIT			_MMIO(0x2054)
+#define  PWRCTX_MAXCNT_VCSUNIT0			_MMIO(0x12054)
+#define  PWRCTX_MAXCNT_BCSUNIT			_MMIO(0x22054)
+#define  PWRCTX_MAXCNT_VECSUNIT			_MMIO(0x1A054)
+#define  PWRCTX_MAXCNT_VCSUNIT1			_MMIO(0x1C054)
+#define    IDLE_TIME_MASK			0xFFFFF
 #define  FORCEWAKE				_MMIO(0xA18C)
 #define  FORCEWAKE_VLV				_MMIO(0x1300b0)
 #define  FORCEWAKE_ACK_VLV			_MMIO(0x1300b4)
@@ -6908,6 +6935,7 @@ enum skl_disp_power_wells {
 #define GEN6_RPDEUC				_MMIO(0xA084)
 #define GEN6_RPDEUCSW				_MMIO(0xA088)
 #define GEN6_RC_STATE				_MMIO(0xA094)
+#define   RC6_STATE				(1 << 18)
 #define GEN6_RC1_WAKE_RATE_LIMIT		_MMIO(0xA098)
 #define GEN6_RC6_WAKE_RATE_LIMIT		_MMIO(0xA09C)
 #define GEN6_RC6pp_WAKE_RATE_LIMIT		_MMIO(0xA0A0)
@@ -7519,7 +7547,7 @@ enum skl_disp_power_wells {
 #define  DPLL_CFGCR2_PDIV_7 (4<<2)
 #define  DPLL_CFGCR2_CENTRAL_FREQ_MASK	(3)
 
-#define DPLL_CFGCR1(id)	_MMIO_PIPE((id) - SKL_DPLL1, _DPLL1_CFGCR1, _DPLL2_CFGCR2)
+#define DPLL_CFGCR1(id)	_MMIO_PIPE((id) - SKL_DPLL1, _DPLL1_CFGCR1, _DPLL2_CFGCR1)
 #define DPLL_CFGCR2(id)	_MMIO_PIPE((id) - SKL_DPLL1, _DPLL1_CFGCR2, _DPLL2_CFGCR2)
 
 /* BXT display engine PLL */
@@ -7540,6 +7568,7 @@ enum skl_disp_power_wells {
 #define  DC_STATE_EN_UPTO_DC5_DC6_MASK   0x3
 
 #define  DC_STATE_DEBUG                  _MMIO(0x45520)
+#define  DC_STATE_DEBUG_MASK_CORES	(1<<0)
 #define  DC_STATE_DEBUG_MASK_MEMORY_UP	(1<<1)
 
 /* Please see hsw_read_dcomp() and hsw_write_dcomp() before using this register,
@@ -8159,4 +8188,11 @@ enum skl_disp_power_wells {
 #define GEN9_VEBOX_MOCS(i)	_MMIO(0xcb00 + (i) * 4)	/* Video MOCS registers */
 #define GEN9_BLT_MOCS(i)	_MMIO(0xcc00 + (i) * 4)	/* Blitter MOCS registers */
 
+/* gamt regs */
+#define GEN8_L3_LRA_1_GPGPU _MMIO(0x4dd4)
+#define   GEN8_L3_LRA_1_GPGPU_DEFAULT_VALUE_BDW  0x67F1427F /* max/min for LRA1/2 */
+#define   GEN8_L3_LRA_1_GPGPU_DEFAULT_VALUE_CHV  0x5FF101FF /* max/min for LRA1/2 */
+#define   GEN9_L3_LRA_1_GPGPU_DEFAULT_VALUE_SKL  0x67F1427F /*    "        " */
+#define   GEN9_L3_LRA_1_GPGPU_DEFAULT_VALUE_BXT  0x5FF101FF /*    "        " */
+
 #endif /* _I915_REG_H_ */
diff --git a/drivers/gpu/drm/i915/i915_suspend.c b/drivers/gpu/drm/i915/i915_suspend.c
index a2aa09ce3202..34e061a9ef06 100644
--- a/drivers/gpu/drm/i915/i915_suspend.c
+++ b/drivers/gpu/drm/i915/i915_suspend.c
@@ -49,7 +49,7 @@ static void i915_save_display(struct drm_device *dev)
 		dev_priv->regfile.savePP_ON_DELAYS = I915_READ(PCH_PP_ON_DELAYS);
 		dev_priv->regfile.savePP_OFF_DELAYS = I915_READ(PCH_PP_OFF_DELAYS);
 		dev_priv->regfile.savePP_DIVISOR = I915_READ(PCH_PP_DIVISOR);
-	} else if (!IS_VALLEYVIEW(dev) && !IS_CHERRYVIEW(dev)) {
+	} else if (INTEL_INFO(dev)->gen <= 4) {
 		dev_priv->regfile.savePP_CONTROL = I915_READ(PP_CONTROL);
 		dev_priv->regfile.savePP_ON_DELAYS = I915_READ(PP_ON_DELAYS);
 		dev_priv->regfile.savePP_OFF_DELAYS = I915_READ(PP_OFF_DELAYS);
@@ -84,7 +84,7 @@ static void i915_restore_display(struct drm_device *dev)
 		I915_WRITE(PCH_PP_OFF_DELAYS, dev_priv->regfile.savePP_OFF_DELAYS);
 		I915_WRITE(PCH_PP_DIVISOR, dev_priv->regfile.savePP_DIVISOR);
 		I915_WRITE(PCH_PP_CONTROL, dev_priv->regfile.savePP_CONTROL);
-	} else if (!IS_VALLEYVIEW(dev) && !IS_CHERRYVIEW(dev)) {
+	} else if (INTEL_INFO(dev)->gen <= 4) {
 		I915_WRITE(PP_ON_DELAYS, dev_priv->regfile.savePP_ON_DELAYS);
 		I915_WRITE(PP_OFF_DELAYS, dev_priv->regfile.savePP_OFF_DELAYS);
 		I915_WRITE(PP_DIVISOR, dev_priv->regfile.savePP_DIVISOR);
@@ -92,7 +92,7 @@ static void i915_restore_display(struct drm_device *dev)
 	}
 
 	/* only restore FBC info on the platform that supports FBC*/
-	intel_fbc_disable(dev_priv);
+	intel_fbc_global_disable(dev_priv);
 
 	/* restore FBC interval */
 	if (HAS_FBC(dev) && INTEL_INFO(dev)->gen <= 4 && !IS_G4X(dev))
diff --git a/drivers/gpu/drm/i915/i915_trace.h b/drivers/gpu/drm/i915/i915_trace.h
index 52b2d409945d..fa09e5581137 100644
--- a/drivers/gpu/drm/i915/i915_trace.h
+++ b/drivers/gpu/drm/i915/i915_trace.h
@@ -175,35 +175,24 @@ TRACE_EVENT(i915_vma_unbind,
 		      __entry->obj, __entry->offset, __entry->size, __entry->vm)
 );
 
-#define VM_TO_TRACE_NAME(vm) \
-	(i915_is_ggtt(vm) ? "G" : \
-		      "P")
-
-DECLARE_EVENT_CLASS(i915_va,
-	TP_PROTO(struct i915_address_space *vm, u64 start, u64 length, const char *name),
-	TP_ARGS(vm, start, length, name),
+TRACE_EVENT(i915_va_alloc,
+	TP_PROTO(struct i915_vma *vma),
+	TP_ARGS(vma),
 
 	TP_STRUCT__entry(
 		__field(struct i915_address_space *, vm)
 		__field(u64, start)
 		__field(u64, end)
-		__string(name, name)
 	),
 
 	TP_fast_assign(
-		__entry->vm = vm;
-		__entry->start = start;
-		__entry->end = start + length - 1;
-		__assign_str(name, name);
+		__entry->vm = vma->vm;
+		__entry->start = vma->node.start;
+		__entry->end = vma->node.start + vma->node.size - 1;
 	),
 
-	TP_printk("vm=%p (%s), 0x%llx-0x%llx",
-		  __entry->vm, __get_str(name),  __entry->start, __entry->end)
-);
-
-DEFINE_EVENT(i915_va, i915_va_alloc,
-	     TP_PROTO(struct i915_address_space *vm, u64 start, u64 length, const char *name),
-	     TP_ARGS(vm, start, length, name)
+	TP_printk("vm=%p (%c), 0x%llx-0x%llx",
+		  __entry->vm, i915_is_ggtt(__entry->vm) ? 'G' : 'P',  __entry->start, __entry->end)
 );
 
 DECLARE_EVENT_CLASS(i915_px_entry,
diff --git a/drivers/gpu/drm/i915/intel_atomic.c b/drivers/gpu/drm/i915/intel_atomic.c
index 4625f8a9ba12..8e579a8505ac 100644
--- a/drivers/gpu/drm/i915/intel_atomic.c
+++ b/drivers/gpu/drm/i915/intel_atomic.c
@@ -97,6 +97,7 @@ intel_crtc_duplicate_state(struct drm_crtc *crtc)
 	crtc_state->disable_lp_wm = false;
 	crtc_state->disable_cxsr = false;
 	crtc_state->wm_changed = false;
+	crtc_state->fb_changed = false;
 
 	return &crtc_state->base;
 }
diff --git a/drivers/gpu/drm/i915/intel_audio.c b/drivers/gpu/drm/i915/intel_audio.c
index 31f6d212fb1b..30f921421b0c 100644
--- a/drivers/gpu/drm/i915/intel_audio.c
+++ b/drivers/gpu/drm/i915/intel_audio.c
@@ -527,6 +527,8 @@ void intel_audio_codec_enable(struct intel_encoder *intel_encoder)
 
 	mutex_lock(&dev_priv->av_mutex);
 	intel_dig_port->audio_connector = connector;
+	/* referred in audio callbacks */
+	dev_priv->dig_port_map[port] = intel_encoder;
 	mutex_unlock(&dev_priv->av_mutex);
 
 	if (acomp && acomp->audio_ops && acomp->audio_ops->pin_eld_notify)
@@ -554,6 +556,7 @@ void intel_audio_codec_disable(struct intel_encoder *intel_encoder)
 
 	mutex_lock(&dev_priv->av_mutex);
 	intel_dig_port->audio_connector = NULL;
+	dev_priv->dig_port_map[port] = NULL;
 	mutex_unlock(&dev_priv->av_mutex);
 
 	if (acomp && acomp->audio_ops && acomp->audio_ops->pin_eld_notify)
diff --git a/drivers/gpu/drm/i915/intel_crt.c b/drivers/gpu/drm/i915/intel_crt.c
index 9c89df1af036..505fc5cf26f8 100644
--- a/drivers/gpu/drm/i915/intel_crt.c
+++ b/drivers/gpu/drm/i915/intel_crt.c
@@ -71,22 +71,29 @@ static bool intel_crt_get_hw_state(struct intel_encoder *encoder,
 	struct intel_crt *crt = intel_encoder_to_crt(encoder);
 	enum intel_display_power_domain power_domain;
 	u32 tmp;
+	bool ret;
 
 	power_domain = intel_display_port_power_domain(encoder);
-	if (!intel_display_power_is_enabled(dev_priv, power_domain))
+	if (!intel_display_power_get_if_enabled(dev_priv, power_domain))
 		return false;
 
+	ret = false;
+
 	tmp = I915_READ(crt->adpa_reg);
 
 	if (!(tmp & ADPA_DAC_ENABLE))
-		return false;
+		goto out;
 
 	if (HAS_PCH_CPT(dev))
 		*pipe = PORT_TO_PIPE_CPT(tmp);
 	else
 		*pipe = PORT_TO_PIPE(tmp);
 
-	return true;
+	ret = true;
+out:
+	intel_display_power_put(dev_priv, power_domain);
+
+	return ret;
 }
 
 static unsigned int intel_crt_get_flags(struct intel_encoder *encoder)
@@ -206,9 +213,7 @@ static void pch_post_disable_crt(struct intel_encoder *encoder)
 
 static void intel_enable_crt(struct intel_encoder *encoder)
 {
-	struct intel_crt *crt = intel_encoder_to_crt(encoder);
-
-	intel_crt_set_dpms(encoder, crt->connector->base.dpms);
+	intel_crt_set_dpms(encoder, DRM_MODE_DPMS_ON);
 }
 
 static enum drm_mode_status
@@ -216,6 +221,7 @@ intel_crt_mode_valid(struct drm_connector *connector,
 		     struct drm_display_mode *mode)
 {
 	struct drm_device *dev = connector->dev;
+	int max_dotclk = to_i915(dev)->max_dotclk_freq;
 
 	int max_clock = 0;
 	if (mode->flags & DRM_MODE_FLAG_DBLSCAN)
@@ -231,6 +237,9 @@ intel_crt_mode_valid(struct drm_connector *connector,
 	if (mode->clock > max_clock)
 		return MODE_CLOCK_HIGH;
 
+	if (mode->clock > max_dotclk)
+		return MODE_CLOCK_HIGH;
+
 	/* The FDI receiver on LPT only supports 8bpc and only has 2 lanes. */
 	if (HAS_PCH_LPT(dev) &&
 	    (ironlake_get_lanes_required(mode->clock, 270000, 24) > 2))
@@ -469,11 +478,10 @@ static bool intel_crt_detect_ddc(struct drm_connector *connector)
 }
 
 static enum drm_connector_status
-intel_crt_load_detect(struct intel_crt *crt)
+intel_crt_load_detect(struct intel_crt *crt, uint32_t pipe)
 {
 	struct drm_device *dev = crt->base.base.dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
-	uint32_t pipe = to_intel_crtc(crt->base.base.crtc)->pipe;
 	uint32_t save_bclrpat;
 	uint32_t save_vtotal;
 	uint32_t vtotal, vactive;
@@ -642,7 +650,8 @@ intel_crt_detect(struct drm_connector *connector, bool force)
 		if (intel_crt_detect_ddc(connector))
 			status = connector_status_connected;
 		else if (INTEL_INFO(dev)->gen < 4)
-			status = intel_crt_load_detect(crt);
+			status = intel_crt_load_detect(crt,
+				to_intel_crtc(connector->state->crtc)->pipe);
 		else
 			status = connector_status_unknown;
 		intel_release_load_detect_pipe(connector, &tmp, &ctx);
diff --git a/drivers/gpu/drm/i915/intel_csr.c b/drivers/gpu/drm/i915/intel_csr.c
index 5c2f9a40c81b..902054efb902 100644
--- a/drivers/gpu/drm/i915/intel_csr.c
+++ b/drivers/gpu/drm/i915/intel_csr.c
@@ -179,7 +179,8 @@ static const struct stepping_info kbl_stepping_info[] = {
 static const struct stepping_info skl_stepping_info[] = {
 	{'A', '0'}, {'B', '0'}, {'C', '0'},
 	{'D', '0'}, {'E', '0'}, {'F', '0'},
-	{'G', '0'}, {'H', '0'}, {'I', '0'}
+	{'G', '0'}, {'H', '0'}, {'I', '0'},
+	{'J', '0'}, {'K', '0'}
 };
 
 static const struct stepping_info bxt_stepping_info[] = {
@@ -219,19 +220,19 @@ static const struct stepping_info *intel_get_stepping_info(struct drm_device *de
  * Everytime display comes back from low power state this function is called to
  * copy the firmware from internal memory to registers.
  */
-void intel_csr_load_program(struct drm_i915_private *dev_priv)
+bool intel_csr_load_program(struct drm_i915_private *dev_priv)
 {
 	u32 *payload = dev_priv->csr.dmc_payload;
 	uint32_t i, fw_size;
 
 	if (!IS_GEN9(dev_priv)) {
 		DRM_ERROR("No CSR support available for this platform\n");
-		return;
+		return false;
 	}
 
 	if (!dev_priv->csr.dmc_payload) {
 		DRM_ERROR("Tried to program CSR with empty payload\n");
-		return;
+		return false;
 	}
 
 	fw_size = dev_priv->csr.dmc_fw_size;
@@ -242,6 +243,10 @@ void intel_csr_load_program(struct drm_i915_private *dev_priv)
 		I915_WRITE(dev_priv->csr.mmioaddr[i],
 			   dev_priv->csr.mmiodata[i]);
 	}
+
+	dev_priv->csr.dc_state = 0;
+
+	return true;
 }
 
 static uint32_t *parse_csr_fw(struct drm_i915_private *dev_priv,
diff --git a/drivers/gpu/drm/i915/intel_ddi.c b/drivers/gpu/drm/i915/intel_ddi.c
index 1f9a3687b540..62de9f4bce09 100644
--- a/drivers/gpu/drm/i915/intel_ddi.c
+++ b/drivers/gpu/drm/i915/intel_ddi.c
@@ -1531,7 +1531,8 @@ skl_ddi_pll_select(struct intel_crtc *intel_crtc,
 			 DPLL_CFGCR2_KDIV(wrpll_params.kdiv) |
 			 DPLL_CFGCR2_PDIV(wrpll_params.pdiv) |
 			 wrpll_params.central_freq;
-	} else if (intel_encoder->type == INTEL_OUTPUT_DISPLAYPORT) {
+	} else if (intel_encoder->type == INTEL_OUTPUT_DISPLAYPORT ||
+		   intel_encoder->type == INTEL_OUTPUT_DP_MST) {
 		switch (crtc_state->port_clock / 2) {
 		case 81000:
 			ctrl1 |= DPLL_CTRL1_LINK_RATE(DPLL_CTRL1_LINK_RATE_810, 0);
@@ -1545,8 +1546,10 @@ skl_ddi_pll_select(struct intel_crtc *intel_crtc,
 		}
 
 		cfgcr1 = cfgcr2 = 0;
-	} else /* eDP */
+	} else if (intel_encoder->type == INTEL_OUTPUT_EDP) {
 		return true;
+	} else
+		return false;
 
 	memset(&crtc_state->dpll_hw_state, 0,
 	       sizeof(crtc_state->dpll_hw_state));
@@ -1910,13 +1913,16 @@ bool intel_ddi_connector_get_hw_state(struct intel_connector *intel_connector)
 	enum transcoder cpu_transcoder;
 	enum intel_display_power_domain power_domain;
 	uint32_t tmp;
+	bool ret;
 
 	power_domain = intel_display_port_power_domain(intel_encoder);
-	if (!intel_display_power_is_enabled(dev_priv, power_domain))
+	if (!intel_display_power_get_if_enabled(dev_priv, power_domain))
 		return false;
 
-	if (!intel_encoder->get_hw_state(intel_encoder, &pipe))
-		return false;
+	if (!intel_encoder->get_hw_state(intel_encoder, &pipe)) {
+		ret = false;
+		goto out;
+	}
 
 	if (port == PORT_A)
 		cpu_transcoder = TRANSCODER_EDP;
@@ -1928,23 +1934,33 @@ bool intel_ddi_connector_get_hw_state(struct intel_connector *intel_connector)
 	switch (tmp & TRANS_DDI_MODE_SELECT_MASK) {
 	case TRANS_DDI_MODE_SELECT_HDMI:
 	case TRANS_DDI_MODE_SELECT_DVI:
-		return (type == DRM_MODE_CONNECTOR_HDMIA);
+		ret = type == DRM_MODE_CONNECTOR_HDMIA;
+		break;
 
 	case TRANS_DDI_MODE_SELECT_DP_SST:
-		if (type == DRM_MODE_CONNECTOR_eDP)
-			return true;
-		return (type == DRM_MODE_CONNECTOR_DisplayPort);
+		ret = type == DRM_MODE_CONNECTOR_eDP ||
+		      type == DRM_MODE_CONNECTOR_DisplayPort;
+		break;
+
 	case TRANS_DDI_MODE_SELECT_DP_MST:
 		/* if the transcoder is in MST state then
 		 * connector isn't connected */
-		return false;
+		ret = false;
+		break;
 
 	case TRANS_DDI_MODE_SELECT_FDI:
-		return (type == DRM_MODE_CONNECTOR_VGA);
+		ret = type == DRM_MODE_CONNECTOR_VGA;
+		break;
 
 	default:
-		return false;
+		ret = false;
+		break;
 	}
+
+out:
+	intel_display_power_put(dev_priv, power_domain);
+
+	return ret;
 }
 
 bool intel_ddi_get_hw_state(struct intel_encoder *encoder,
@@ -1956,15 +1972,18 @@ bool intel_ddi_get_hw_state(struct intel_encoder *encoder,
 	enum intel_display_power_domain power_domain;
 	u32 tmp;
 	int i;
+	bool ret;
 
 	power_domain = intel_display_port_power_domain(encoder);
-	if (!intel_display_power_is_enabled(dev_priv, power_domain))
+	if (!intel_display_power_get_if_enabled(dev_priv, power_domain))
 		return false;
 
+	ret = false;
+
 	tmp = I915_READ(DDI_BUF_CTL(port));
 
 	if (!(tmp & DDI_BUF_CTL_ENABLE))
-		return false;
+		goto out;
 
 	if (port == PORT_A) {
 		tmp = I915_READ(TRANS_DDI_FUNC_CTL(TRANSCODER_EDP));
@@ -1982,25 +2001,32 @@ bool intel_ddi_get_hw_state(struct intel_encoder *encoder,
 			break;
 		}
 
-		return true;
-	} else {
-		for (i = TRANSCODER_A; i <= TRANSCODER_C; i++) {
-			tmp = I915_READ(TRANS_DDI_FUNC_CTL(i));
+		ret = true;
 
-			if ((tmp & TRANS_DDI_PORT_MASK)
-			    == TRANS_DDI_SELECT_PORT(port)) {
-				if ((tmp & TRANS_DDI_MODE_SELECT_MASK) == TRANS_DDI_MODE_SELECT_DP_MST)
-					return false;
+		goto out;
+	}
 
-				*pipe = i;
-				return true;
-			}
+	for (i = TRANSCODER_A; i <= TRANSCODER_C; i++) {
+		tmp = I915_READ(TRANS_DDI_FUNC_CTL(i));
+
+		if ((tmp & TRANS_DDI_PORT_MASK) == TRANS_DDI_SELECT_PORT(port)) {
+			if ((tmp & TRANS_DDI_MODE_SELECT_MASK) ==
+			    TRANS_DDI_MODE_SELECT_DP_MST)
+				goto out;
+
+			*pipe = i;
+			ret = true;
+
+			goto out;
 		}
 	}
 
 	DRM_DEBUG_KMS("No pipe for ddi port %c found\n", port_name(port));
 
-	return false;
+out:
+	intel_display_power_put(dev_priv, power_domain);
+
+	return ret;
 }
 
 void intel_ddi_enable_pipe_clock(struct intel_crtc *intel_crtc)
@@ -2446,12 +2472,14 @@ static bool hsw_ddi_wrpll_get_hw_state(struct drm_i915_private *dev_priv,
 {
 	uint32_t val;
 
-	if (!intel_display_power_is_enabled(dev_priv, POWER_DOMAIN_PLLS))
+	if (!intel_display_power_get_if_enabled(dev_priv, POWER_DOMAIN_PLLS))
 		return false;
 
 	val = I915_READ(WRPLL_CTL(pll->id));
 	hw_state->wrpll = val;
 
+	intel_display_power_put(dev_priv, POWER_DOMAIN_PLLS);
+
 	return val & WRPLL_PLL_ENABLE;
 }
 
@@ -2461,12 +2489,14 @@ static bool hsw_ddi_spll_get_hw_state(struct drm_i915_private *dev_priv,
 {
 	uint32_t val;
 
-	if (!intel_display_power_is_enabled(dev_priv, POWER_DOMAIN_PLLS))
+	if (!intel_display_power_get_if_enabled(dev_priv, POWER_DOMAIN_PLLS))
 		return false;
 
 	val = I915_READ(SPLL_CTL);
 	hw_state->spll = val;
 
+	intel_display_power_put(dev_priv, POWER_DOMAIN_PLLS);
+
 	return val & SPLL_PLL_ENABLE;
 }
 
@@ -2583,16 +2613,19 @@ static bool skl_ddi_pll_get_hw_state(struct drm_i915_private *dev_priv,
 	uint32_t val;
 	unsigned int dpll;
 	const struct skl_dpll_regs *regs = skl_dpll_regs;
+	bool ret;
 
-	if (!intel_display_power_is_enabled(dev_priv, POWER_DOMAIN_PLLS))
+	if (!intel_display_power_get_if_enabled(dev_priv, POWER_DOMAIN_PLLS))
 		return false;
 
+	ret = false;
+
 	/* DPLL0 is not part of the shared DPLLs, so pll->id is 0 for DPLL1 */
 	dpll = pll->id + 1;
 
 	val = I915_READ(regs[pll->id].ctl);
 	if (!(val & LCPLL_PLL_ENABLE))
-		return false;
+		goto out;
 
 	val = I915_READ(DPLL_CTRL1);
 	hw_state->ctrl1 = (val >> (dpll * 6)) & 0x3f;
@@ -2602,8 +2635,12 @@ static bool skl_ddi_pll_get_hw_state(struct drm_i915_private *dev_priv,
 		hw_state->cfgcr1 = I915_READ(regs[pll->id].cfgcr1);
 		hw_state->cfgcr2 = I915_READ(regs[pll->id].cfgcr2);
 	}
+	ret = true;
 
-	return true;
+out:
+	intel_display_power_put(dev_priv, POWER_DOMAIN_PLLS);
+
+	return ret;
 }
 
 static void skl_shared_dplls_init(struct drm_i915_private *dev_priv)
@@ -2870,13 +2907,16 @@ static bool bxt_ddi_pll_get_hw_state(struct drm_i915_private *dev_priv,
 {
 	enum port port = (enum port)pll->id;	/* 1:1 port->PLL mapping */
 	uint32_t val;
+	bool ret;
 
-	if (!intel_display_power_is_enabled(dev_priv, POWER_DOMAIN_PLLS))
+	if (!intel_display_power_get_if_enabled(dev_priv, POWER_DOMAIN_PLLS))
 		return false;
 
+	ret = false;
+
 	val = I915_READ(BXT_PORT_PLL_ENABLE(port));
 	if (!(val & PORT_PLL_ENABLE))
-		return false;
+		goto out;
 
 	hw_state->ebb0 = I915_READ(BXT_PORT_PLL_EBB_0(port));
 	hw_state->ebb0 &= PORT_PLL_P1_MASK | PORT_PLL_P2_MASK;
@@ -2923,7 +2963,12 @@ static bool bxt_ddi_pll_get_hw_state(struct drm_i915_private *dev_priv,
 				 I915_READ(BXT_PORT_PCS_DW12_LN23(port)));
 	hw_state->pcsdw12 &= LANE_STAGGER_MASK | LANESTAGGER_STRAP_OVRD;
 
-	return true;
+	ret = true;
+
+out:
+	intel_display_power_put(dev_priv, POWER_DOMAIN_PLLS);
+
+	return ret;
 }
 
 static void bxt_shared_dplls_init(struct drm_i915_private *dev_priv)
@@ -3058,11 +3103,15 @@ bool intel_ddi_is_audio_enabled(struct drm_i915_private *dev_priv,
 {
 	u32 temp;
 
-	if (intel_display_power_is_enabled(dev_priv, POWER_DOMAIN_AUDIO)) {
+	if (intel_display_power_get_if_enabled(dev_priv, POWER_DOMAIN_AUDIO)) {
 		temp = I915_READ(HSW_AUD_PIN_ELD_CP_VLD);
+
+		intel_display_power_put(dev_priv, POWER_DOMAIN_AUDIO);
+
 		if (temp & AUDIO_OUTPUT_ENABLE(intel_crtc->pipe))
 			return true;
 	}
+
 	return false;
 }
 
@@ -3277,11 +3326,9 @@ void intel_ddi_init(struct drm_device *dev, enum port port)
 	intel_encoder->get_config = intel_ddi_get_config;
 
 	intel_dig_port->port = port;
-	dev_priv->dig_port_map[port] = intel_encoder;
 	intel_dig_port->saved_port_bits = I915_READ(DDI_BUF_CTL(port)) &
 					  (DDI_BUF_PORT_REVERSAL |
 					   DDI_A_4_LANES);
-	intel_dig_port->max_lanes = max_lanes;
 
 	/*
 	 * Bspec says that DDI_A_4_LANES is the only supported configuration
@@ -3294,9 +3341,12 @@ void intel_ddi_init(struct drm_device *dev, enum port port)
 		if (!(intel_dig_port->saved_port_bits & DDI_A_4_LANES)) {
 			DRM_DEBUG_KMS("BXT BIOS forgot to set DDI_A_4_LANES for port A; fixing\n");
 			intel_dig_port->saved_port_bits |= DDI_A_4_LANES;
+			max_lanes = 4;
 		}
 	}
 
+	intel_dig_port->max_lanes = max_lanes;
+
 	intel_encoder->type = INTEL_OUTPUT_UNKNOWN;
 	intel_encoder->crtc_mask = (1 << 0) | (1 << 1) | (1 << 2);
 	intel_encoder->cloneable = 0;
diff --git a/drivers/gpu/drm/i915/intel_display.c b/drivers/gpu/drm/i915/intel_display.c
index e1e7cdee3bbd..6e0d8283daa6 100644
--- a/drivers/gpu/drm/i915/intel_display.c
+++ b/drivers/gpu/drm/i915/intel_display.c
@@ -1343,18 +1343,21 @@ void assert_pipe(struct drm_i915_private *dev_priv,
 	bool cur_state;
 	enum transcoder cpu_transcoder = intel_pipe_to_cpu_transcoder(dev_priv,
 								      pipe);
+	enum intel_display_power_domain power_domain;
 
 	/* if we need the pipe quirk it must be always on */
 	if ((pipe == PIPE_A && dev_priv->quirks & QUIRK_PIPEA_FORCE) ||
 	    (pipe == PIPE_B && dev_priv->quirks & QUIRK_PIPEB_FORCE))
 		state = true;
 
-	if (!intel_display_power_is_enabled(dev_priv,
-				POWER_DOMAIN_TRANSCODER(cpu_transcoder))) {
-		cur_state = false;
-	} else {
+	power_domain = POWER_DOMAIN_TRANSCODER(cpu_transcoder);
+	if (intel_display_power_get_if_enabled(dev_priv, power_domain)) {
 		u32 val = I915_READ(PIPECONF(cpu_transcoder));
 		cur_state = !!(val & PIPECONF_ENABLE);
+
+		intel_display_power_put(dev_priv, power_domain);
+	} else {
+		cur_state = false;
 	}
 
 	I915_STATE_WARN(cur_state != state,
@@ -2284,7 +2287,7 @@ intel_fill_fb_ggtt_view(struct i915_ggtt_view *view, struct drm_framebuffer *fb,
 			const struct drm_plane_state *plane_state)
 {
 	struct drm_i915_private *dev_priv = to_i915(fb->dev);
-	struct intel_rotation_info *info = &view->params.rotation_info;
+	struct intel_rotation_info *info = &view->params.rotated;
 	unsigned int tile_size, tile_width, tile_height, cpp;
 
 	*view = i915_ggtt_view_normal;
@@ -2306,7 +2309,7 @@ intel_fill_fb_ggtt_view(struct i915_ggtt_view *view, struct drm_framebuffer *fb,
 	tile_size = intel_tile_size(dev_priv);
 
 	cpp = drm_format_plane_cpp(fb->pixel_format, 0);
-	tile_width = intel_tile_width(dev_priv, cpp, fb->modifier[0]);
+	tile_width = intel_tile_width(dev_priv, fb->modifier[0], cpp);
 	tile_height = tile_size / tile_width;
 
 	info->width_pages = DIV_ROUND_UP(fb->pitches[0], tile_width);
@@ -2448,11 +2451,11 @@ static void intel_unpin_fb_obj(struct drm_framebuffer *fb,
 
 /* Computes the linear offset to the base tile and adjusts x, y. bytes per pixel
  * is assumed to be a power-of-two. */
-unsigned long intel_compute_tile_offset(struct drm_i915_private *dev_priv,
-					int *x, int *y,
-					uint64_t fb_modifier,
-					unsigned int cpp,
-					unsigned int pitch)
+u32 intel_compute_tile_offset(struct drm_i915_private *dev_priv,
+			      int *x, int *y,
+			      uint64_t fb_modifier,
+			      unsigned int cpp,
+			      unsigned int pitch)
 {
 	if (fb_modifier != DRM_FORMAT_MOD_NONE) {
 		unsigned int tile_size, tile_width, tile_height;
@@ -2551,12 +2554,16 @@ intel_alloc_initial_plane_obj(struct intel_crtc *crtc,
 	if (size_aligned * 2 > dev_priv->gtt.stolen_usable_size)
 		return false;
 
+	mutex_lock(&dev->struct_mutex);
+
 	obj = i915_gem_object_create_stolen_for_preallocated(dev,
 							     base_aligned,
 							     base_aligned,
 							     size_aligned);
-	if (!obj)
+	if (!obj) {
+		mutex_unlock(&dev->struct_mutex);
 		return false;
+	}
 
 	obj->tiling_mode = plane_config->tiling;
 	if (obj->tiling_mode == I915_TILING_X)
@@ -2569,12 +2576,12 @@ intel_alloc_initial_plane_obj(struct intel_crtc *crtc,
 	mode_cmd.modifier[0] = fb->modifier[0];
 	mode_cmd.flags = DRM_MODE_FB_MODIFIERS;
 
-	mutex_lock(&dev->struct_mutex);
 	if (intel_framebuffer_init(dev, to_intel_framebuffer(fb),
 				   &mode_cmd, obj)) {
 		DRM_DEBUG_KMS("intel fb init failed\n");
 		goto out_unref_obj;
 	}
+
 	mutex_unlock(&dev->struct_mutex);
 
 	DRM_DEBUG_KMS("initial plane fb obj %p\n", obj);
@@ -2706,14 +2713,12 @@ static void i9xx_update_primary_plane(struct drm_plane *primary,
 	struct drm_framebuffer *fb = plane_state->base.fb;
 	struct drm_i915_gem_object *obj = intel_fb_obj(fb);
 	int plane = intel_crtc->plane;
-	unsigned long linear_offset;
-	int x = plane_state->src.x1 >> 16;
-	int y = plane_state->src.y1 >> 16;
+	u32 linear_offset;
 	u32 dspcntr;
 	i915_reg_t reg = DSPCNTR(plane);
-	int pixel_size;
-
-	pixel_size = drm_format_plane_cpp(fb->pixel_format, 0);
+	int cpp = drm_format_plane_cpp(fb->pixel_format, 0);
+	int x = plane_state->src.x1 >> 16;
+	int y = plane_state->src.y1 >> 16;
 
 	dspcntr = DISPPLANE_GAMMA_ENABLE;
 
@@ -2771,13 +2776,12 @@ static void i9xx_update_primary_plane(struct drm_plane *primary,
 	if (IS_G4X(dev))
 		dspcntr |= DISPPLANE_TRICKLE_FEED_DISABLE;
 
-	linear_offset = y * fb->pitches[0] + x * pixel_size;
+	linear_offset = y * fb->pitches[0] + x * cpp;
 
 	if (INTEL_INFO(dev)->gen >= 4) {
 		intel_crtc->dspaddr_offset =
 			intel_compute_tile_offset(dev_priv, &x, &y,
-						  fb->modifier[0],
-						  pixel_size,
+						  fb->modifier[0], cpp,
 						  fb->pitches[0]);
 		linear_offset -= intel_crtc->dspaddr_offset;
 	} else {
@@ -2794,7 +2798,7 @@ static void i9xx_update_primary_plane(struct drm_plane *primary,
 		data and adding to linear_offset*/
 		linear_offset +=
 			(crtc_state->pipe_src_h - 1) * fb->pitches[0] +
-			(crtc_state->pipe_src_w - 1) * pixel_size;
+			(crtc_state->pipe_src_w - 1) * cpp;
 	}
 
 	intel_crtc->adjusted_x = x;
@@ -2839,10 +2843,10 @@ static void ironlake_update_primary_plane(struct drm_plane *primary,
 	struct drm_framebuffer *fb = plane_state->base.fb;
 	struct drm_i915_gem_object *obj = intel_fb_obj(fb);
 	int plane = intel_crtc->plane;
-	unsigned long linear_offset;
+	u32 linear_offset;
 	u32 dspcntr;
 	i915_reg_t reg = DSPCNTR(plane);
-	int pixel_size = drm_format_plane_cpp(fb->pixel_format, 0);
+	int cpp = drm_format_plane_cpp(fb->pixel_format, 0);
 	int x = plane_state->src.x1 >> 16;
 	int y = plane_state->src.y1 >> 16;
 
@@ -2881,11 +2885,10 @@ static void ironlake_update_primary_plane(struct drm_plane *primary,
 	if (!IS_HASWELL(dev) && !IS_BROADWELL(dev))
 		dspcntr |= DISPPLANE_TRICKLE_FEED_DISABLE;
 
-	linear_offset = y * fb->pitches[0] + x * pixel_size;
+	linear_offset = y * fb->pitches[0] + x * cpp;
 	intel_crtc->dspaddr_offset =
 		intel_compute_tile_offset(dev_priv, &x, &y,
-					  fb->modifier[0],
-					  pixel_size,
+					  fb->modifier[0], cpp,
 					  fb->pitches[0]);
 	linear_offset -= intel_crtc->dspaddr_offset;
 	if (plane_state->base.rotation == BIT(DRM_ROTATE_180)) {
@@ -2899,7 +2902,7 @@ static void ironlake_update_primary_plane(struct drm_plane *primary,
 			data and adding to linear_offset*/
 			linear_offset +=
 				(crtc_state->pipe_src_h - 1) * fb->pitches[0] +
-				(crtc_state->pipe_src_w - 1) * pixel_size;
+				(crtc_state->pipe_src_w - 1) * cpp;
 		}
 	}
 
@@ -2951,7 +2954,7 @@ u32 intel_plane_obj_offset(struct intel_plane *intel_plane,
 	offset = vma->node.start;
 
 	if (plane == 1) {
-		offset += vma->ggtt_view.params.rotation_info.uv_start_page *
+		offset += vma->ggtt_view.params.rotated.uv_start_page *
 			  PAGE_SIZE;
 	}
 
@@ -3160,9 +3163,6 @@ static void skylake_disable_primary_plane(struct drm_plane *primary,
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	int pipe = to_intel_crtc(crtc)->pipe;
 
-	if (dev_priv->fbc.deactivate)
-		dev_priv->fbc.deactivate(dev_priv);
-
 	I915_WRITE(PLANE_CTL(pipe, 0), 0);
 	I915_WRITE(PLANE_SURF(pipe, 0), 0);
 	POSTING_READ(PLANE_SURF(pipe, 0));
@@ -4792,9 +4792,6 @@ static void intel_post_plane_update(struct intel_crtc *crtc)
 		to_intel_crtc_state(crtc->base.state);
 	struct drm_device *dev = crtc->base.dev;
 
-	if (atomic->wait_vblank)
-		intel_wait_for_vblank(dev, crtc->pipe);
-
 	intel_frontbuffer_flip(dev, atomic->fb_bits);
 
 	crtc->wm.cxsr_allowed = true;
@@ -4803,7 +4800,7 @@ static void intel_post_plane_update(struct intel_crtc *crtc)
 		intel_update_watermarks(&crtc->base);
 
 	if (atomic->update_fbc)
-		intel_fbc_update(crtc);
+		intel_fbc_post_update(crtc);
 
 	if (atomic->post_enable_primary)
 		intel_post_enable_primary(&crtc->base);
@@ -4811,26 +4808,39 @@ static void intel_post_plane_update(struct intel_crtc *crtc)
 	memset(atomic, 0, sizeof(*atomic));
 }
 
-static void intel_pre_plane_update(struct intel_crtc *crtc)
+static void intel_pre_plane_update(struct intel_crtc_state *old_crtc_state)
 {
+	struct intel_crtc *crtc = to_intel_crtc(old_crtc_state->base.crtc);
 	struct drm_device *dev = crtc->base.dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	struct intel_crtc_atomic_commit *atomic = &crtc->atomic;
 	struct intel_crtc_state *pipe_config =
 		to_intel_crtc_state(crtc->base.state);
+	struct drm_atomic_state *old_state = old_crtc_state->base.state;
+	struct drm_plane *primary = crtc->base.primary;
+	struct drm_plane_state *old_pri_state =
+		drm_atomic_get_existing_plane_state(old_state, primary);
+	bool modeset = needs_modeset(&pipe_config->base);
 
-	if (atomic->disable_fbc)
-		intel_fbc_deactivate(crtc);
+	if (atomic->update_fbc)
+		intel_fbc_pre_update(crtc);
 
-	if (crtc->atomic.disable_ips)
-		hsw_disable_ips(crtc);
+	if (old_pri_state) {
+		struct intel_plane_state *primary_state =
+			to_intel_plane_state(primary->state);
+		struct intel_plane_state *old_primary_state =
+			to_intel_plane_state(old_pri_state);
 
-	if (atomic->pre_disable_primary)
-		intel_pre_disable_primary(&crtc->base);
+		if (old_primary_state->visible &&
+		    (modeset || !primary_state->visible))
+			intel_pre_disable_primary(&crtc->base);
+	}
 
 	if (pipe_config->disable_cxsr) {
 		crtc->wm.cxsr_allowed = false;
-		intel_set_memory_cxsr(dev_priv, false);
+
+		if (old_crtc_state->base.active)
+			intel_set_memory_cxsr(dev_priv, false);
 	}
 
 	if (!needs_modeset(&pipe_config->base) && pipe_config->wm_changed)
@@ -4931,8 +4941,6 @@ static void ironlake_crtc_enable(struct drm_crtc *crtc)
 	if (intel_crtc->config->has_pch_encoder)
 		intel_wait_for_vblank(dev, pipe);
 	intel_set_pch_fifo_underrun_reporting(dev_priv, pipe, true);
-
-	intel_fbc_enable(intel_crtc);
 }
 
 /* IPS only exists on ULT machines and is tied to pipe A. */
@@ -5045,8 +5053,6 @@ static void haswell_crtc_enable(struct drm_crtc *crtc)
 		intel_wait_for_vblank(dev, hsw_workaround_pipe);
 		intel_wait_for_vblank(dev, hsw_workaround_pipe);
 	}
-
-	intel_fbc_enable(intel_crtc);
 }
 
 static void ironlake_pfit_disable(struct intel_crtc *crtc, bool force)
@@ -5127,8 +5133,6 @@ static void ironlake_crtc_disable(struct drm_crtc *crtc)
 	}
 
 	intel_set_pch_fifo_underrun_reporting(dev_priv, pipe, true);
-
-	intel_fbc_disable_crtc(intel_crtc);
 }
 
 static void haswell_crtc_disable(struct drm_crtc *crtc)
@@ -5179,8 +5183,6 @@ static void haswell_crtc_disable(struct drm_crtc *crtc)
 		intel_set_pch_fifo_underrun_reporting(dev_priv, TRANSCODER_A,
 						      true);
 	}
-
-	intel_fbc_disable_crtc(intel_crtc);
 }
 
 static void i9xx_pfit_enable(struct intel_crtc *crtc)
@@ -5303,31 +5305,37 @@ intel_display_port_aux_power_domain(struct intel_encoder *intel_encoder)
 	}
 }
 
-static unsigned long get_crtc_power_domains(struct drm_crtc *crtc)
+static unsigned long get_crtc_power_domains(struct drm_crtc *crtc,
+					    struct intel_crtc_state *crtc_state)
 {
 	struct drm_device *dev = crtc->dev;
-	struct intel_encoder *intel_encoder;
+	struct drm_encoder *encoder;
 	struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
 	enum pipe pipe = intel_crtc->pipe;
 	unsigned long mask;
-	enum transcoder transcoder = intel_crtc->config->cpu_transcoder;
+	enum transcoder transcoder = crtc_state->cpu_transcoder;
 
-	if (!crtc->state->active)
+	if (!crtc_state->base.active)
 		return 0;
 
 	mask = BIT(POWER_DOMAIN_PIPE(pipe));
 	mask |= BIT(POWER_DOMAIN_TRANSCODER(transcoder));
-	if (intel_crtc->config->pch_pfit.enabled ||
-	    intel_crtc->config->pch_pfit.force_thru)
+	if (crtc_state->pch_pfit.enabled ||
+	    crtc_state->pch_pfit.force_thru)
 		mask |= BIT(POWER_DOMAIN_PIPE_PANEL_FITTER(pipe));
 
-	for_each_encoder_on_crtc(dev, crtc, intel_encoder)
+	drm_for_each_encoder_mask(encoder, dev, crtc_state->base.encoder_mask) {
+		struct intel_encoder *intel_encoder = to_intel_encoder(encoder);
+
 		mask |= BIT(intel_display_port_power_domain(intel_encoder));
+	}
 
 	return mask;
 }
 
-static unsigned long modeset_get_crtc_power_domains(struct drm_crtc *crtc)
+static unsigned long
+modeset_get_crtc_power_domains(struct drm_crtc *crtc,
+			       struct intel_crtc_state *crtc_state)
 {
 	struct drm_i915_private *dev_priv = crtc->dev->dev_private;
 	struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
@@ -5335,7 +5343,8 @@ static unsigned long modeset_get_crtc_power_domains(struct drm_crtc *crtc)
 	unsigned long domains, new_domains, old_domains;
 
 	old_domains = intel_crtc->enabled_power_domains;
-	intel_crtc->enabled_power_domains = new_domains = get_crtc_power_domains(crtc);
+	intel_crtc->enabled_power_domains = new_domains =
+		get_crtc_power_domains(crtc, crtc_state);
 
 	domains = new_domains & ~old_domains;
 
@@ -5354,31 +5363,6 @@ static void modeset_put_power_domains(struct drm_i915_private *dev_priv,
 		intel_display_power_put(dev_priv, domain);
 }
 
-static void modeset_update_crtc_power_domains(struct drm_atomic_state *state)
-{
-	struct intel_atomic_state *intel_state = to_intel_atomic_state(state);
-	struct drm_device *dev = state->dev;
-	struct drm_i915_private *dev_priv = dev->dev_private;
-	unsigned long put_domains[I915_MAX_PIPES] = {};
-	struct drm_crtc_state *crtc_state;
-	struct drm_crtc *crtc;
-	int i;
-
-	for_each_crtc_in_state(state, crtc, crtc_state, i) {
-		if (needs_modeset(crtc->state))
-			put_domains[to_intel_crtc(crtc)->pipe] =
-				modeset_get_crtc_power_domains(crtc);
-	}
-
-	if (dev_priv->display.modeset_commit_cdclk &&
-	    intel_state->dev_cdclk != dev_priv->cdclk_freq)
-		dev_priv->display.modeset_commit_cdclk(state);
-
-	for (i = 0; i < I915_MAX_PIPES; i++)
-		if (put_domains[i])
-			modeset_put_power_domains(dev_priv, put_domains[i]);
-}
-
 static int intel_compute_max_dotclk(struct drm_i915_private *dev_priv)
 {
 	int max_cdclk_freq = dev_priv->max_cdclk_freq;
@@ -6041,8 +6025,7 @@ static int broxton_calc_cdclk(struct drm_i915_private *dev_priv,
 		return 144000;
 }
 
-/* Compute the max pixel clock for new configuration. Uses atomic state if
- * that's non-NULL, look at current state otherwise. */
+/* Compute the max pixel clock for new configuration. */
 static int intel_mode_max_pixclk(struct drm_device *dev,
 				 struct drm_atomic_state *state)
 {
@@ -6065,9 +6048,6 @@ static int intel_mode_max_pixclk(struct drm_device *dev,
 		intel_state->min_pixclk[i] = pixclk;
 	}
 
-	if (!intel_state->active_crtcs)
-		return 0;
-
 	for_each_pipe(dev_priv, pipe)
 		max_pixclk = max(intel_state->min_pixclk[pipe], max_pixclk);
 
@@ -6291,8 +6271,6 @@ static void i9xx_crtc_enable(struct drm_crtc *crtc)
 
 	for_each_encoder_on_crtc(dev, crtc, encoder)
 		encoder->enable(encoder);
-
-	intel_fbc_enable(intel_crtc);
 }
 
 static void i9xx_pfit_disable(struct intel_crtc *crtc)
@@ -6355,8 +6333,6 @@ static void i9xx_crtc_disable(struct drm_crtc *crtc)
 
 	if (!IS_GEN2(dev))
 		intel_set_cpu_fifo_underrun_reporting(dev_priv, pipe, false);
-
-	intel_fbc_disable_crtc(intel_crtc);
 }
 
 static void intel_crtc_disable_noatomic(struct drm_crtc *crtc)
@@ -6380,6 +6356,7 @@ static void intel_crtc_disable_noatomic(struct drm_crtc *crtc)
 
 	dev_priv->display.crtc_disable(crtc);
 	intel_crtc->active = false;
+	intel_fbc_disable(intel_crtc);
 	intel_update_watermarks(crtc);
 	intel_disable_shared_dpll(intel_crtc);
 
@@ -6398,55 +6375,16 @@ static void intel_crtc_disable_noatomic(struct drm_crtc *crtc)
  */
 int intel_display_suspend(struct drm_device *dev)
 {
-	struct drm_mode_config *config = &dev->mode_config;
-	struct drm_modeset_acquire_ctx *ctx = config->acquire_ctx;
+	struct drm_i915_private *dev_priv = to_i915(dev);
 	struct drm_atomic_state *state;
-	struct drm_crtc *crtc;
-	unsigned crtc_mask = 0;
-	int ret = 0;
-
-	if (WARN_ON(!ctx))
-		return 0;
-
-	lockdep_assert_held(&ctx->ww_ctx);
-	state = drm_atomic_state_alloc(dev);
-	if (WARN_ON(!state))
-		return -ENOMEM;
-
-	state->acquire_ctx = ctx;
-	state->allow_modeset = true;
-
-	for_each_crtc(dev, crtc) {
-		struct drm_crtc_state *crtc_state =
-			drm_atomic_get_crtc_state(state, crtc);
-
-		ret = PTR_ERR_OR_ZERO(crtc_state);
-		if (ret)
-			goto free;
-
-		if (!crtc_state->active)
-			continue;
-
-		crtc_state->active = false;
-		crtc_mask |= 1 << drm_crtc_index(crtc);
-	}
-
-	if (crtc_mask) {
-		ret = drm_atomic_commit(state);
-
-		if (!ret) {
-			for_each_crtc(dev, crtc)
-				if (crtc_mask & (1 << drm_crtc_index(crtc)))
-					crtc->state->active = true;
-
-			return ret;
-		}
-	}
+	int ret;
 
-free:
+	state = drm_atomic_helper_suspend(dev);
+	ret = PTR_ERR_OR_ZERO(state);
 	if (ret)
 		DRM_ERROR("Suspending crtc's failed with %i\n", ret);
-	drm_atomic_state_free(state);
+	else
+		dev_priv->modeset_restore_state = state;
 	return ret;
 }
 
@@ -8186,18 +8124,22 @@ static bool i9xx_get_pipe_config(struct intel_crtc *crtc,
 {
 	struct drm_device *dev = crtc->base.dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
+	enum intel_display_power_domain power_domain;
 	uint32_t tmp;
+	bool ret;
 
-	if (!intel_display_power_is_enabled(dev_priv,
-					    POWER_DOMAIN_PIPE(crtc->pipe)))
+	power_domain = POWER_DOMAIN_PIPE(crtc->pipe);
+	if (!intel_display_power_get_if_enabled(dev_priv, power_domain))
 		return false;
 
 	pipe_config->cpu_transcoder = (enum transcoder) crtc->pipe;
 	pipe_config->shared_dpll = DPLL_ID_PRIVATE;
 
+	ret = false;
+
 	tmp = I915_READ(PIPECONF(crtc->pipe));
 	if (!(tmp & PIPECONF_ENABLE))
-		return false;
+		goto out;
 
 	if (IS_G4X(dev) || IS_VALLEYVIEW(dev) || IS_CHERRYVIEW(dev)) {
 		switch (tmp & PIPECONF_BPC_MASK) {
@@ -8277,7 +8219,12 @@ static bool i9xx_get_pipe_config(struct intel_crtc *crtc,
 	pipe_config->base.adjusted_mode.crtc_clock =
 		pipe_config->port_clock / pipe_config->pixel_multiplier;
 
-	return true;
+	ret = true;
+
+out:
+	intel_display_power_put(dev_priv, power_domain);
+
+	return ret;
 }
 
 static void ironlake_init_pch_refclk(struct drm_device *dev)
@@ -9381,18 +9328,21 @@ static bool ironlake_get_pipe_config(struct intel_crtc *crtc,
 {
 	struct drm_device *dev = crtc->base.dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
+	enum intel_display_power_domain power_domain;
 	uint32_t tmp;
+	bool ret;
 
-	if (!intel_display_power_is_enabled(dev_priv,
-					    POWER_DOMAIN_PIPE(crtc->pipe)))
+	power_domain = POWER_DOMAIN_PIPE(crtc->pipe);
+	if (!intel_display_power_get_if_enabled(dev_priv, power_domain))
 		return false;
 
 	pipe_config->cpu_transcoder = (enum transcoder) crtc->pipe;
 	pipe_config->shared_dpll = DPLL_ID_PRIVATE;
 
+	ret = false;
 	tmp = I915_READ(PIPECONF(crtc->pipe));
 	if (!(tmp & PIPECONF_ENABLE))
-		return false;
+		goto out;
 
 	switch (tmp & PIPECONF_BPC_MASK) {
 	case PIPECONF_6BPC:
@@ -9455,7 +9405,12 @@ static bool ironlake_get_pipe_config(struct intel_crtc *crtc,
 
 	ironlake_get_pfit_config(crtc, pipe_config);
 
-	return true;
+	ret = true;
+
+out:
+	intel_display_power_put(dev_priv, power_domain);
+
+	return ret;
 }
 
 static void assert_can_disable_lcpll(struct drm_i915_private *dev_priv)
@@ -9721,9 +9676,6 @@ static int ilk_max_pixel_rate(struct drm_atomic_state *state)
 		intel_state->min_pixclk[i] = pixel_rate;
 	}
 
-	if (!intel_state->active_crtcs)
-		return 0;
-
 	for_each_pipe(dev_priv, pipe)
 		max_pixel_rate = max(intel_state->min_pixclk[pipe], max_pixel_rate);
 
@@ -9853,8 +9805,13 @@ static void broadwell_modeset_commit_cdclk(struct drm_atomic_state *old_state)
 static int haswell_crtc_compute_clock(struct intel_crtc *crtc,
 				      struct intel_crtc_state *crtc_state)
 {
-	if (!intel_ddi_pll_select(crtc, crtc_state))
-		return -EINVAL;
+	struct intel_encoder *intel_encoder =
+		intel_ddi_get_crtc_new_encoder(crtc_state);
+
+	if (intel_encoder->type != INTEL_OUTPUT_DSI) {
+		if (!intel_ddi_pll_select(crtc, crtc_state))
+			return -EINVAL;
+	}
 
 	crtc->lowfreq_avail = false;
 
@@ -9982,12 +9939,17 @@ static bool haswell_get_pipe_config(struct intel_crtc *crtc,
 {
 	struct drm_device *dev = crtc->base.dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
-	enum intel_display_power_domain pfit_domain;
+	enum intel_display_power_domain power_domain;
+	unsigned long power_domain_mask;
 	uint32_t tmp;
+	bool ret;
 
-	if (!intel_display_power_is_enabled(dev_priv,
-					 POWER_DOMAIN_PIPE(crtc->pipe)))
+	power_domain = POWER_DOMAIN_PIPE(crtc->pipe);
+	if (!intel_display_power_get_if_enabled(dev_priv, power_domain))
 		return false;
+	power_domain_mask = BIT(power_domain);
+
+	ret = false;
 
 	pipe_config->cpu_transcoder = (enum transcoder) crtc->pipe;
 	pipe_config->shared_dpll = DPLL_ID_PRIVATE;
@@ -10014,13 +9976,14 @@ static bool haswell_get_pipe_config(struct intel_crtc *crtc,
 			pipe_config->cpu_transcoder = TRANSCODER_EDP;
 	}
 
-	if (!intel_display_power_is_enabled(dev_priv,
-			POWER_DOMAIN_TRANSCODER(pipe_config->cpu_transcoder)))
-		return false;
+	power_domain = POWER_DOMAIN_TRANSCODER(pipe_config->cpu_transcoder);
+	if (!intel_display_power_get_if_enabled(dev_priv, power_domain))
+		goto out;
+	power_domain_mask |= BIT(power_domain);
 
 	tmp = I915_READ(PIPECONF(pipe_config->cpu_transcoder));
 	if (!(tmp & PIPECONF_ENABLE))
-		return false;
+		goto out;
 
 	haswell_get_ddi_port_state(crtc, pipe_config);
 
@@ -10030,14 +9993,14 @@ static bool haswell_get_pipe_config(struct intel_crtc *crtc,
 		skl_init_scalers(dev, crtc, pipe_config);
 	}
 
-	pfit_domain = POWER_DOMAIN_PIPE_PANEL_FITTER(crtc->pipe);
-
 	if (INTEL_INFO(dev)->gen >= 9) {
 		pipe_config->scaler_state.scaler_id = -1;
 		pipe_config->scaler_state.scaler_users &= ~(1 << SKL_CRTC_INDEX);
 	}
 
-	if (intel_display_power_is_enabled(dev_priv, pfit_domain)) {
+	power_domain = POWER_DOMAIN_PIPE_PANEL_FITTER(crtc->pipe);
+	if (intel_display_power_get_if_enabled(dev_priv, power_domain)) {
+		power_domain_mask |= BIT(power_domain);
 		if (INTEL_INFO(dev)->gen >= 9)
 			skylake_get_pfit_config(crtc, pipe_config);
 		else
@@ -10055,7 +10018,13 @@ static bool haswell_get_pipe_config(struct intel_crtc *crtc,
 		pipe_config->pixel_multiplier = 1;
 	}
 
-	return true;
+	ret = true;
+
+out:
+	for_each_power_domain(power_domain, power_domain_mask)
+		intel_display_power_put(dev_priv, power_domain);
+
+	return ret;
 }
 
 static void i845_update_cursor(struct drm_crtc *crtc, u32 base,
@@ -10376,6 +10345,7 @@ mode_fits_in_fbdev(struct drm_device *dev,
 	if (obj->base.size < mode->vdisplay * fb->pitches[0])
 		return NULL;
 
+	drm_framebuffer_reference(fb);
 	return fb;
 #else
 	return NULL;
@@ -10431,7 +10401,7 @@ bool intel_get_load_detect_pipe(struct drm_connector *connector,
 	struct drm_device *dev = encoder->dev;
 	struct drm_framebuffer *fb;
 	struct drm_mode_config *config = &dev->mode_config;
-	struct drm_atomic_state *state = NULL;
+	struct drm_atomic_state *state = NULL, *restore_state = NULL;
 	struct drm_connector_state *connector_state;
 	struct intel_crtc_state *crtc_state;
 	int ret, i = -1;
@@ -10440,6 +10410,8 @@ bool intel_get_load_detect_pipe(struct drm_connector *connector,
 		      connector->base.id, connector->name,
 		      encoder->base.id, encoder->name);
 
+	old->restore_state = NULL;
+
 retry:
 	ret = drm_modeset_lock(&config->connection_mutex, ctx);
 	if (ret)
@@ -10456,24 +10428,15 @@ retry:
 	 */
 
 	/* See if we already have a CRTC for this connector */
-	if (encoder->crtc) {
-		crtc = encoder->crtc;
+	if (connector->state->crtc) {
+		crtc = connector->state->crtc;
 
 		ret = drm_modeset_lock(&crtc->mutex, ctx);
 		if (ret)
 			goto fail;
-		ret = drm_modeset_lock(&crtc->primary->mutex, ctx);
-		if (ret)
-			goto fail;
-
-		old->dpms_mode = connector->dpms;
-		old->load_detect_temp = false;
 
 		/* Make sure the crtc and connector are running */
-		if (connector->dpms != DRM_MODE_DPMS_ON)
-			connector->funcs->dpms(connector, DRM_MODE_DPMS_ON);
-
-		return true;
+		goto found;
 	}
 
 	/* Find an unused one (if possible) */
@@ -10481,8 +10444,15 @@ retry:
 		i++;
 		if (!(encoder->possible_crtcs & (1 << i)))
 			continue;
-		if (possible_crtc->state->enable)
+
+		ret = drm_modeset_lock(&possible_crtc->mutex, ctx);
+		if (ret)
+			goto fail;
+
+		if (possible_crtc->state->enable) {
+			drm_modeset_unlock(&possible_crtc->mutex);
 			continue;
+		}
 
 		crtc = possible_crtc;
 		break;
@@ -10496,23 +10466,22 @@ retry:
 		goto fail;
 	}
 
-	ret = drm_modeset_lock(&crtc->mutex, ctx);
-	if (ret)
-		goto fail;
+found:
+	intel_crtc = to_intel_crtc(crtc);
+
 	ret = drm_modeset_lock(&crtc->primary->mutex, ctx);
 	if (ret)
 		goto fail;
 
-	intel_crtc = to_intel_crtc(crtc);
-	old->dpms_mode = connector->dpms;
-	old->load_detect_temp = true;
-	old->release_fb = NULL;
-
 	state = drm_atomic_state_alloc(dev);
-	if (!state)
-		return false;
+	restore_state = drm_atomic_state_alloc(dev);
+	if (!state || !restore_state) {
+		ret = -ENOMEM;
+		goto fail;
+	}
 
 	state->acquire_ctx = ctx;
+	restore_state->acquire_ctx = ctx;
 
 	connector_state = drm_atomic_get_connector_state(state, connector);
 	if (IS_ERR(connector_state)) {
@@ -10520,7 +10489,9 @@ retry:
 		goto fail;
 	}
 
-	connector_state->crtc = crtc;
+	ret = drm_atomic_set_crtc_for_connector(connector_state, crtc);
+	if (ret)
+		goto fail;
 
 	crtc_state = intel_atomic_get_crtc_state(state, intel_crtc);
 	if (IS_ERR(crtc_state)) {
@@ -10544,7 +10515,6 @@ retry:
 	if (fb == NULL) {
 		DRM_DEBUG_KMS("creating tmp fb for load-detection\n");
 		fb = intel_framebuffer_create_for_mode(dev, mode, 24, 32);
-		old->release_fb = fb;
 	} else
 		DRM_DEBUG_KMS("reusing fbdev for load-detection framebuffer\n");
 	if (IS_ERR(fb)) {
@@ -10556,15 +10526,29 @@ retry:
 	if (ret)
 		goto fail;
 
-	drm_mode_copy(&crtc_state->base.mode, mode);
+	drm_framebuffer_unreference(fb);
+
+	ret = drm_atomic_set_mode_for_crtc(&crtc_state->base, mode);
+	if (ret)
+		goto fail;
+
+	ret = PTR_ERR_OR_ZERO(drm_atomic_get_connector_state(restore_state, connector));
+	if (!ret)
+		ret = PTR_ERR_OR_ZERO(drm_atomic_get_crtc_state(restore_state, crtc));
+	if (!ret)
+		ret = PTR_ERR_OR_ZERO(drm_atomic_get_plane_state(restore_state, crtc->primary));
+	if (ret) {
+		DRM_DEBUG_KMS("Failed to create a copy of old state to restore: %i\n", ret);
+		goto fail;
+	}
 
-	if (drm_atomic_commit(state)) {
+	ret = drm_atomic_commit(state);
+	if (ret) {
 		DRM_DEBUG_KMS("failed to set mode on load-detect pipe\n");
-		if (old->release_fb)
-			old->release_fb->funcs->destroy(old->release_fb);
 		goto fail;
 	}
-	crtc->primary->crtc = crtc;
+
+	old->restore_state = restore_state;
 
 	/* let the connector get through one full cycle before testing */
 	intel_wait_for_vblank(dev, intel_crtc->pipe);
@@ -10572,7 +10556,8 @@ retry:
 
 fail:
 	drm_atomic_state_free(state);
-	state = NULL;
+	drm_atomic_state_free(restore_state);
+	restore_state = state = NULL;
 
 	if (ret == -EDEADLK) {
 		drm_modeset_backoff(ctx);
@@ -10586,65 +10571,24 @@ void intel_release_load_detect_pipe(struct drm_connector *connector,
 				    struct intel_load_detect_pipe *old,
 				    struct drm_modeset_acquire_ctx *ctx)
 {
-	struct drm_device *dev = connector->dev;
 	struct intel_encoder *intel_encoder =
 		intel_attached_encoder(connector);
 	struct drm_encoder *encoder = &intel_encoder->base;
-	struct drm_crtc *crtc = encoder->crtc;
-	struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
-	struct drm_atomic_state *state;
-	struct drm_connector_state *connector_state;
-	struct intel_crtc_state *crtc_state;
+	struct drm_atomic_state *state = old->restore_state;
 	int ret;
 
 	DRM_DEBUG_KMS("[CONNECTOR:%d:%s], [ENCODER:%d:%s]\n",
 		      connector->base.id, connector->name,
 		      encoder->base.id, encoder->name);
 
-	if (old->load_detect_temp) {
-		state = drm_atomic_state_alloc(dev);
-		if (!state)
-			goto fail;
-
-		state->acquire_ctx = ctx;
-
-		connector_state = drm_atomic_get_connector_state(state, connector);
-		if (IS_ERR(connector_state))
-			goto fail;
-
-		crtc_state = intel_atomic_get_crtc_state(state, intel_crtc);
-		if (IS_ERR(crtc_state))
-			goto fail;
-
-		connector_state->crtc = NULL;
-
-		crtc_state->base.enable = crtc_state->base.active = false;
-
-		ret = intel_modeset_setup_plane_state(state, crtc, NULL, NULL,
-						      0, 0);
-		if (ret)
-			goto fail;
-
-		ret = drm_atomic_commit(state);
-		if (ret)
-			goto fail;
-
-		if (old->release_fb) {
-			drm_framebuffer_unregister_private(old->release_fb);
-			drm_framebuffer_unreference(old->release_fb);
-		}
-
+	if (!state)
 		return;
-	}
 
-	/* Switch crtc and encoder back off if necessary */
-	if (old->dpms_mode != DRM_MODE_DPMS_ON)
-		connector->funcs->dpms(connector, old->dpms_mode);
-
-	return;
-fail:
-	DRM_DEBUG_KMS("Couldn't release load detect pipe.\n");
-	drm_atomic_state_free(state);
+	ret = drm_atomic_commit(state);
+	if (ret) {
+		DRM_DEBUG_KMS("Couldn't release load detect pipe: %i\n", ret);
+		drm_atomic_state_free(state);
+	}
 }
 
 static int i9xx_pll_refclk(struct drm_device *dev,
@@ -10914,6 +10858,7 @@ static void intel_unpin_work_fn(struct work_struct *__work)
 	mutex_unlock(&dev->struct_mutex);
 
 	intel_frontbuffer_flip_complete(dev, to_intel_plane(primary)->frontbuffer_bit);
+	intel_fbc_post_update(crtc);
 	drm_framebuffer_unreference(work->old_fb);
 
 	BUG_ON(atomic_read(&crtc->unpin_work_count) == 0);
@@ -10995,6 +10940,12 @@ static bool page_flip_finished(struct intel_crtc *crtc)
 		return true;
 
 	/*
+	 * BDW signals flip done immediately if the plane
+	 * is disabled, even if the plane enable is already
+	 * armed to occur at the next vblank :(
+	 */
+
+	/*
 	 * A DSPSURFLIVE check isn't enough in case the mmio and CS flips
 	 * used the same base address. In that case the mmio flip might
 	 * have completed, but the CS hasn't even executed the flip yet.
@@ -11629,6 +11580,7 @@ static int intel_crtc_page_flip(struct drm_crtc *crtc,
 
 	crtc->primary->fb = fb;
 	update_state_fb(crtc->primary);
+	intel_fbc_pre_update(intel_crtc);
 
 	work->pending_flip_obj = obj;
 
@@ -11713,7 +11665,6 @@ static int intel_crtc_page_flip(struct drm_crtc *crtc,
 			  to_intel_plane(primary)->frontbuffer_bit);
 	mutex_unlock(&dev->struct_mutex);
 
-	intel_fbc_deactivate(intel_crtc);
 	intel_frontbuffer_flip_prepare(dev,
 				       to_intel_plane(primary)->frontbuffer_bit);
 
@@ -11724,7 +11675,7 @@ static int intel_crtc_page_flip(struct drm_crtc *crtc,
 cleanup_unpin:
 	intel_unpin_fb_obj(fb, crtc->primary->state);
 cleanup_pending:
-	if (request)
+	if (!IS_ERR_OR_NULL(request))
 		i915_gem_request_cancel(request);
 	atomic_dec(&intel_crtc->unpin_work_count);
 	mutex_unlock(&dev->struct_mutex);
@@ -11835,11 +11786,9 @@ int intel_plane_atomic_calc_changes(struct drm_crtc_state *crtc_state,
 	struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
 	struct drm_plane *plane = plane_state->plane;
 	struct drm_device *dev = crtc->dev;
-	struct drm_i915_private *dev_priv = dev->dev_private;
 	struct intel_plane_state *old_plane_state =
 		to_intel_plane_state(plane->state);
 	int idx = intel_crtc->base.base.id, ret;
-	int i = drm_plane_index(plane);
 	bool mode_changed = needs_modeset(crtc_state);
 	bool was_crtc_enabled = crtc->state->active;
 	bool is_crtc_enabled = crtc_state->active;
@@ -11872,6 +11821,9 @@ int intel_plane_atomic_calc_changes(struct drm_crtc_state *crtc_state,
 	if (!was_visible && !visible)
 		return 0;
 
+	if (fb != old_plane_state->base.fb)
+		pipe_config->fb_changed = true;
+
 	turn_off = was_visible && (!visible || mode_changed);
 	turn_on = visible && (!was_visible || mode_changed);
 
@@ -11886,11 +11838,8 @@ int intel_plane_atomic_calc_changes(struct drm_crtc_state *crtc_state,
 		pipe_config->wm_changed = true;
 
 		/* must disable cxsr around plane enable/disable */
-		if (plane->type != DRM_PLANE_TYPE_CURSOR) {
-			if (is_crtc_enabled)
-				intel_crtc->atomic.wait_vblank = true;
+		if (plane->type != DRM_PLANE_TYPE_CURSOR)
 			pipe_config->disable_cxsr = true;
-		}
 	} else if (intel_wm_need_update(plane, plane_state)) {
 		pipe_config->wm_changed = true;
 	}
@@ -11901,49 +11850,9 @@ int intel_plane_atomic_calc_changes(struct drm_crtc_state *crtc_state,
 
 	switch (plane->type) {
 	case DRM_PLANE_TYPE_PRIMARY:
-		intel_crtc->atomic.pre_disable_primary = turn_off;
 		intel_crtc->atomic.post_enable_primary = turn_on;
+		intel_crtc->atomic.update_fbc = true;
 
-		if (turn_off) {
-			/*
-			 * FIXME: Actually if we will still have any other
-			 * plane enabled on the pipe we could let IPS enabled
-			 * still, but for now lets consider that when we make
-			 * primary invisible by setting DSPCNTR to 0 on
-			 * update_primary_plane function IPS needs to be
-			 * disable.
-			 */
-			intel_crtc->atomic.disable_ips = true;
-
-			intel_crtc->atomic.disable_fbc = true;
-		}
-
-		/*
-		 * FBC does not work on some platforms for rotated
-		 * planes, so disable it when rotation is not 0 and
-		 * update it when rotation is set back to 0.
-		 *
-		 * FIXME: This is redundant with the fbc update done in
-		 * the primary plane enable function except that that
-		 * one is done too late. We eventually need to unify
-		 * this.
-		 */
-
-		if (visible &&
-		    INTEL_INFO(dev)->gen <= 4 && !IS_G4X(dev) &&
-		    dev_priv->fbc.crtc == intel_crtc &&
-		    plane_state->rotation != BIT(DRM_ROTATE_0))
-			intel_crtc->atomic.disable_fbc = true;
-
-		/*
-		 * BDW signals flip done immediately if the plane
-		 * is disabled, even if the plane enable is already
-		 * armed to occur at the next vblank :(
-		 */
-		if (turn_on && IS_BROADWELL(dev))
-			intel_crtc->atomic.wait_vblank = true;
-
-		intel_crtc->atomic.update_fbc |= visible || mode_changed;
 		break;
 	case DRM_PLANE_TYPE_CURSOR:
 		break;
@@ -11956,13 +11865,8 @@ int intel_plane_atomic_calc_changes(struct drm_crtc_state *crtc_state,
 		 */
 		if (IS_IVYBRIDGE(dev) &&
 		    needs_scaling(to_intel_plane_state(plane_state)) &&
-		    !needs_scaling(old_plane_state)) {
-			to_intel_crtc_state(crtc_state)->disable_lp_wm = true;
-		} else if (turn_off && !mode_changed) {
-			intel_crtc->atomic.wait_vblank = true;
-			intel_crtc->atomic.update_sprite_watermarks |=
-				1 << i;
-		}
+		    !needs_scaling(old_plane_state))
+			pipe_config->disable_lp_wm = true;
 
 		break;
 	}
@@ -13130,8 +13034,6 @@ static void intel_modeset_clear_plls(struct drm_atomic_state *state)
 	struct drm_device *dev = state->dev;
 	struct drm_i915_private *dev_priv = to_i915(dev);
 	struct intel_shared_dpll_config *shared_dpll = NULL;
-	struct intel_crtc *intel_crtc;
-	struct intel_crtc_state *intel_crtc_state;
 	struct drm_crtc *crtc;
 	struct drm_crtc_state *crtc_state;
 	int i;
@@ -13140,21 +13042,21 @@ static void intel_modeset_clear_plls(struct drm_atomic_state *state)
 		return;
 
 	for_each_crtc_in_state(state, crtc, crtc_state, i) {
-		int dpll;
-
-		intel_crtc = to_intel_crtc(crtc);
-		intel_crtc_state = to_intel_crtc_state(crtc_state);
-		dpll = intel_crtc_state->shared_dpll;
+		struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
+		int old_dpll = to_intel_crtc_state(crtc->state)->shared_dpll;
 
-		if (!needs_modeset(crtc_state) || dpll == DPLL_ID_PRIVATE)
+		if (!needs_modeset(crtc_state))
 			continue;
 
-		intel_crtc_state->shared_dpll = DPLL_ID_PRIVATE;
+		to_intel_crtc_state(crtc_state)->shared_dpll = DPLL_ID_PRIVATE;
+
+		if (old_dpll == DPLL_ID_PRIVATE)
+			continue;
 
 		if (!shared_dpll)
 			shared_dpll = intel_atomic_get_shared_dpll_state(state);
 
-		shared_dpll[dpll].crtc_mask &= ~(1 << intel_crtc->pipe);
+		shared_dpll[old_dpll].crtc_mask &= ~(1 << intel_crtc->pipe);
 	}
 }
 
@@ -13290,6 +13192,9 @@ static int intel_modeset_checks(struct drm_atomic_state *state)
 
 		if (ret < 0)
 			return ret;
+
+		DRM_DEBUG_KMS("New cdclk calculated to be atomic %u, actual %u\n",
+			      intel_state->cdclk, intel_state->dev_cdclk);
 	} else
 		to_intel_atomic_state(state)->cdclk = dev_priv->atomic_cdclk_freq;
 
@@ -13348,6 +13253,7 @@ static void calc_watermark_data(struct drm_atomic_state *state)
 static int intel_atomic_check(struct drm_device *dev,
 			      struct drm_atomic_state *state)
 {
+	struct drm_i915_private *dev_priv = to_i915(dev);
 	struct intel_atomic_state *intel_state = to_intel_atomic_state(state);
 	struct drm_crtc *crtc;
 	struct drm_crtc_state *crtc_state;
@@ -13390,7 +13296,7 @@ static int intel_atomic_check(struct drm_device *dev,
 			return ret;
 
 		if (i915.fastboot &&
-		    intel_pipe_config_compare(state->dev,
+		    intel_pipe_config_compare(dev,
 					to_intel_crtc_state(crtc->state),
 					pipe_config, true)) {
 			crtc_state->mode_changed = false;
@@ -13416,12 +13322,13 @@ static int intel_atomic_check(struct drm_device *dev,
 		if (ret)
 			return ret;
 	} else
-		intel_state->cdclk = to_i915(state->dev)->cdclk_freq;
+		intel_state->cdclk = dev_priv->cdclk_freq;
 
-	ret = drm_atomic_helper_check_planes(state->dev, state);
+	ret = drm_atomic_helper_check_planes(dev, state);
 	if (ret)
 		return ret;
 
+	intel_fbc_choose_crtc(dev_priv, state);
 	calc_watermark_data(state);
 
 	return 0;
@@ -13493,6 +13400,71 @@ static int intel_atomic_prepare_commit(struct drm_device *dev,
 	return ret;
 }
 
+static void intel_atomic_wait_for_vblanks(struct drm_device *dev,
+					  struct drm_i915_private *dev_priv,
+					  unsigned crtc_mask)
+{
+	unsigned last_vblank_count[I915_MAX_PIPES];
+	enum pipe pipe;
+	int ret;
+
+	if (!crtc_mask)
+		return;
+
+	for_each_pipe(dev_priv, pipe) {
+		struct drm_crtc *crtc = dev_priv->pipe_to_crtc_mapping[pipe];
+
+		if (!((1 << pipe) & crtc_mask))
+			continue;
+
+		ret = drm_crtc_vblank_get(crtc);
+		if (WARN_ON(ret != 0)) {
+			crtc_mask &= ~(1 << pipe);
+			continue;
+		}
+
+		last_vblank_count[pipe] = drm_crtc_vblank_count(crtc);
+	}
+
+	for_each_pipe(dev_priv, pipe) {
+		struct drm_crtc *crtc = dev_priv->pipe_to_crtc_mapping[pipe];
+		long lret;
+
+		if (!((1 << pipe) & crtc_mask))
+			continue;
+
+		lret = wait_event_timeout(dev->vblank[pipe].queue,
+				last_vblank_count[pipe] !=
+					drm_crtc_vblank_count(crtc),
+				msecs_to_jiffies(50));
+
+		WARN_ON(!lret);
+
+		drm_crtc_vblank_put(crtc);
+	}
+}
+
+static bool needs_vblank_wait(struct intel_crtc_state *crtc_state)
+{
+	/* fb updated, need to unpin old fb */
+	if (crtc_state->fb_changed)
+		return true;
+
+	/* wm changes, need vblank before final wm's */
+	if (crtc_state->wm_changed)
+		return true;
+
+	/*
+	 * cxsr is re-enabled after vblank.
+	 * This is already handled by crtc_state->wm_changed,
+	 * but added for clarity.
+	 */
+	if (crtc_state->disable_cxsr)
+		return true;
+
+	return false;
+}
+
 /**
  * intel_atomic_commit - commit validated state object
  * @dev: DRM device
@@ -13519,6 +13491,8 @@ static int intel_atomic_commit(struct drm_device *dev,
 	struct drm_crtc *crtc;
 	int ret = 0, i;
 	bool hw_check = intel_state->modeset;
+	unsigned long put_domains[I915_MAX_PIPES] = {};
+	unsigned crtc_vblank_mask = 0;
 
 	ret = intel_atomic_prepare_commit(dev, state, async);
 	if (ret) {
@@ -13534,20 +13508,32 @@ static int intel_atomic_commit(struct drm_device *dev,
 		       sizeof(intel_state->min_pixclk));
 		dev_priv->active_crtcs = intel_state->active_crtcs;
 		dev_priv->atomic_cdclk_freq = intel_state->cdclk;
+
+		intel_display_power_get(dev_priv, POWER_DOMAIN_MODESET);
 	}
 
 	for_each_crtc_in_state(state, crtc, crtc_state, i) {
 		struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
 
+		if (needs_modeset(crtc->state) ||
+		    to_intel_crtc_state(crtc->state)->update_pipe) {
+			hw_check = true;
+
+			put_domains[to_intel_crtc(crtc)->pipe] =
+				modeset_get_crtc_power_domains(crtc,
+					to_intel_crtc_state(crtc->state));
+		}
+
 		if (!needs_modeset(crtc->state))
 			continue;
 
-		intel_pre_plane_update(intel_crtc);
+		intel_pre_plane_update(to_intel_crtc_state(crtc_state));
 
 		if (crtc_state->active) {
 			intel_crtc_disable_planes(crtc, crtc_state->plane_mask);
 			dev_priv->display.crtc_disable(crtc);
 			intel_crtc->active = false;
+			intel_fbc_disable(intel_crtc);
 			intel_disable_shared_dpll(intel_crtc);
 
 			/*
@@ -13570,51 +13556,53 @@ static int intel_atomic_commit(struct drm_device *dev,
 		intel_shared_dpll_commit(state);
 
 		drm_atomic_helper_update_legacy_modeset_state(state->dev, state);
-		modeset_update_crtc_power_domains(state);
+
+		if (dev_priv->display.modeset_commit_cdclk &&
+		    intel_state->dev_cdclk != dev_priv->cdclk_freq)
+			dev_priv->display.modeset_commit_cdclk(state);
 	}
 
 	/* Now enable the clocks, plane, pipe, and connectors that we set up. */
 	for_each_crtc_in_state(state, crtc, crtc_state, i) {
 		struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
 		bool modeset = needs_modeset(crtc->state);
-		bool update_pipe = !modeset &&
-			to_intel_crtc_state(crtc->state)->update_pipe;
-		unsigned long put_domains = 0;
-
-		if (modeset)
-			intel_display_power_get(dev_priv, POWER_DOMAIN_MODESET);
+		struct intel_crtc_state *pipe_config =
+			to_intel_crtc_state(crtc->state);
+		bool update_pipe = !modeset && pipe_config->update_pipe;
 
 		if (modeset && crtc->state->active) {
 			update_scanline_offset(to_intel_crtc(crtc));
 			dev_priv->display.crtc_enable(crtc);
 		}
 
-		if (update_pipe) {
-			put_domains = modeset_get_crtc_power_domains(crtc);
-
-			/* make sure intel_modeset_check_state runs */
-			hw_check = true;
-		}
-
 		if (!modeset)
-			intel_pre_plane_update(intel_crtc);
+			intel_pre_plane_update(to_intel_crtc_state(crtc_state));
+
+		if (crtc->state->active && intel_crtc->atomic.update_fbc)
+			intel_fbc_enable(intel_crtc);
 
 		if (crtc->state->active &&
 		    (crtc->state->planes_changed || update_pipe))
 			drm_atomic_helper_commit_planes_on_crtc(crtc_state);
 
-		if (put_domains)
-			modeset_put_power_domains(dev_priv, put_domains);
-
-		intel_post_plane_update(intel_crtc);
-
-		if (modeset)
-			intel_display_power_put(dev_priv, POWER_DOMAIN_MODESET);
+		if (pipe_config->base.active && needs_vblank_wait(pipe_config))
+			crtc_vblank_mask |= 1 << i;
 	}
 
 	/* FIXME: add subpixel order */
 
-	drm_atomic_helper_wait_for_vblanks(dev, state);
+	if (!state->legacy_cursor_update)
+		intel_atomic_wait_for_vblanks(dev, dev_priv, crtc_vblank_mask);
+
+	for_each_crtc_in_state(state, crtc, crtc_state, i) {
+		intel_post_plane_update(to_intel_crtc(crtc));
+
+		if (put_domains[i])
+			modeset_put_power_domains(dev_priv, put_domains[i]);
+	}
+
+	if (intel_state->modeset)
+		intel_display_power_put(dev_priv, POWER_DOMAIN_MODESET);
 
 	mutex_lock(&dev->struct_mutex);
 	drm_atomic_helper_cleanup_planes(dev, state);
@@ -13696,7 +13684,7 @@ static bool ibx_pch_dpll_get_hw_state(struct drm_i915_private *dev_priv,
 {
 	uint32_t val;
 
-	if (!intel_display_power_is_enabled(dev_priv, POWER_DOMAIN_PLLS))
+	if (!intel_display_power_get_if_enabled(dev_priv, POWER_DOMAIN_PLLS))
 		return false;
 
 	val = I915_READ(PCH_DPLL(pll->id));
@@ -13704,6 +13692,8 @@ static bool ibx_pch_dpll_get_hw_state(struct drm_i915_private *dev_priv,
 	hw_state->fp0 = I915_READ(PCH_FP0(pll->id));
 	hw_state->fp1 = I915_READ(PCH_FP1(pll->id));
 
+	intel_display_power_put(dev_priv, POWER_DOMAIN_PLLS);
+
 	return val & DPLL_VCO_ENABLE;
 }
 
@@ -14682,10 +14672,12 @@ u32 intel_fb_pitch_limit(struct drm_device *dev, uint64_t fb_modifier,
 	u32 gen = INTEL_INFO(dev)->gen;
 
 	if (gen >= 9) {
+		int cpp = drm_format_plane_cpp(pixel_format, 0);
+
 		/* "The stride in bytes must not exceed the of the size of 8K
 		 *  pixels and 32K bytes."
 		 */
-		 return min(8192*drm_format_plane_cpp(pixel_format, 0), 32768);
+		return min(8192 * cpp, 32768);
 	} else if (gen >= 5 && !IS_VALLEYVIEW(dev) && !IS_CHERRYVIEW(dev)) {
 		return 32*1024;
 	} else if (gen >= 4) {
@@ -15517,6 +15509,17 @@ static bool intel_crtc_has_encoders(struct intel_crtc *crtc)
 	return false;
 }
 
+static bool intel_encoder_has_connectors(struct intel_encoder *encoder)
+{
+	struct drm_device *dev = encoder->base.dev;
+	struct intel_connector *connector;
+
+	for_each_connector_on_encoder(dev, &encoder->base, connector)
+		return true;
+
+	return false;
+}
+
 static void intel_sanitize_crtc(struct intel_crtc *crtc)
 {
 	struct drm_device *dev = crtc->base.dev;
@@ -15627,7 +15630,6 @@ static void intel_sanitize_encoder(struct intel_encoder *encoder)
 {
 	struct intel_connector *connector;
 	struct drm_device *dev = encoder->base.dev;
-	bool active = false;
 
 	/* We need to check both for a crtc link (meaning that the
 	 * encoder is active and trying to read from a pipe) and the
@@ -15635,15 +15637,7 @@ static void intel_sanitize_encoder(struct intel_encoder *encoder)
 	bool has_active_crtc = encoder->base.crtc &&
 		to_intel_crtc(encoder->base.crtc)->active;
 
-	for_each_intel_connector(dev, connector) {
-		if (connector->base.encoder != &encoder->base)
-			continue;
-
-		active = true;
-		break;
-	}
-
-	if (active && !has_active_crtc) {
+	if (intel_encoder_has_connectors(encoder) && !has_active_crtc) {
 		DRM_DEBUG_KMS("[ENCODER:%d:%s] has active connectors but no active pipe!\n",
 			      encoder->base.base.id,
 			      encoder->base.name);
@@ -15698,10 +15692,12 @@ void i915_redisable_vga(struct drm_device *dev)
 	 * level, just check if the power well is enabled instead of trying to
 	 * follow the "don't touch the power well if we don't need it" policy
 	 * the rest of the driver uses. */
-	if (!intel_display_power_is_enabled(dev_priv, POWER_DOMAIN_VGA))
+	if (!intel_display_power_get_if_enabled(dev_priv, POWER_DOMAIN_VGA))
 		return;
 
 	i915_redisable_vga_power_on(dev);
+
+	intel_display_power_put(dev_priv, POWER_DOMAIN_VGA);
 }
 
 static bool primary_get_hw_state(struct intel_plane *plane)
@@ -15929,63 +15925,76 @@ intel_modeset_setup_hw_state(struct drm_device *dev)
 	for_each_intel_crtc(dev, crtc) {
 		unsigned long put_domains;
 
-		put_domains = modeset_get_crtc_power_domains(&crtc->base);
+		put_domains = modeset_get_crtc_power_domains(&crtc->base, crtc->config);
 		if (WARN_ON(put_domains))
 			modeset_put_power_domains(dev_priv, put_domains);
 	}
 	intel_display_set_init_power(dev_priv, false);
+
+	intel_fbc_init_pipe_state(dev_priv);
 }
 
 void intel_display_resume(struct drm_device *dev)
 {
-	struct drm_atomic_state *state = drm_atomic_state_alloc(dev);
-	struct intel_connector *conn;
-	struct intel_plane *plane;
-	struct drm_crtc *crtc;
+	struct drm_i915_private *dev_priv = to_i915(dev);
+	struct drm_atomic_state *state = dev_priv->modeset_restore_state;
+	struct drm_modeset_acquire_ctx ctx;
 	int ret;
+	bool setup = false;
 
-	if (!state)
-		return;
-
-	state->acquire_ctx = dev->mode_config.acquire_ctx;
+	dev_priv->modeset_restore_state = NULL;
 
-	/* preserve complete old state, including dpll */
-	intel_atomic_get_shared_dpll_state(state);
+	/*
+	 * This is a cludge because with real atomic modeset mode_config.mutex
+	 * won't be taken. Unfortunately some probed state like
+	 * audio_codec_enable is still protected by mode_config.mutex, so lock
+	 * it here for now.
+	 */
+	mutex_lock(&dev->mode_config.mutex);
+	drm_modeset_acquire_init(&ctx, 0);
 
-	for_each_crtc(dev, crtc) {
-		struct drm_crtc_state *crtc_state =
-			drm_atomic_get_crtc_state(state, crtc);
+retry:
+	ret = drm_modeset_lock_all_ctx(dev, &ctx);
 
-		ret = PTR_ERR_OR_ZERO(crtc_state);
-		if (ret)
-			goto err;
+	if (ret == 0 && !setup) {
+		setup = true;
 
-		/* force a restore */
-		crtc_state->mode_changed = true;
+		intel_modeset_setup_hw_state(dev);
+		i915_redisable_vga(dev);
 	}
 
-	for_each_intel_plane(dev, plane) {
-		ret = PTR_ERR_OR_ZERO(drm_atomic_get_plane_state(state, &plane->base));
-		if (ret)
-			goto err;
-	}
+	if (ret == 0 && state) {
+		struct drm_crtc_state *crtc_state;
+		struct drm_crtc *crtc;
+		int i;
 
-	for_each_intel_connector(dev, conn) {
-		ret = PTR_ERR_OR_ZERO(drm_atomic_get_connector_state(state, &conn->base));
-		if (ret)
-			goto err;
+		state->acquire_ctx = &ctx;
+
+		for_each_crtc_in_state(state, crtc, crtc_state, i) {
+			/*
+			 * Force recalculation even if we restore
+			 * current state. With fast modeset this may not result
+			 * in a modeset when the state is compatible.
+			 */
+			crtc_state->mode_changed = true;
+		}
+
+		ret = drm_atomic_commit(state);
 	}
 
-	intel_modeset_setup_hw_state(dev);
+	if (ret == -EDEADLK) {
+		drm_modeset_backoff(&ctx);
+		goto retry;
+	}
 
-	i915_redisable_vga(dev);
-	ret = drm_atomic_commit(state);
-	if (!ret)
-		return;
+	drm_modeset_drop_locks(&ctx);
+	drm_modeset_acquire_fini(&ctx);
+	mutex_unlock(&dev->mode_config.mutex);
 
-err:
-	DRM_ERROR("Restoring old state failed with %i\n", ret);
-	drm_atomic_state_free(state);
+	if (ret) {
+		DRM_ERROR("Restoring old state failed with %i\n", ret);
+		drm_atomic_state_free(state);
+	}
 }
 
 void intel_modeset_gem_init(struct drm_device *dev)
@@ -15994,9 +16003,7 @@ void intel_modeset_gem_init(struct drm_device *dev)
 	struct drm_i915_gem_object *obj;
 	int ret;
 
-	mutex_lock(&dev->struct_mutex);
 	intel_init_gt_powersave(dev);
-	mutex_unlock(&dev->struct_mutex);
 
 	intel_modeset_init_hw(dev);
 
@@ -16063,7 +16070,7 @@ void intel_modeset_cleanup(struct drm_device *dev)
 
 	intel_unregister_dsm_handler();
 
-	intel_fbc_disable(dev_priv);
+	intel_fbc_global_disable(dev_priv);
 
 	/* flush any delayed tasks or pending work */
 	flush_scheduled_work();
@@ -16076,9 +16083,7 @@ void intel_modeset_cleanup(struct drm_device *dev)
 
 	intel_cleanup_overlay(dev);
 
-	mutex_lock(&dev->struct_mutex);
 	intel_cleanup_gt_powersave(dev);
-	mutex_unlock(&dev->struct_mutex);
 
 	intel_teardown_gmbus(dev);
 }
diff --git a/drivers/gpu/drm/i915/intel_dp.c b/drivers/gpu/drm/i915/intel_dp.c
index da704c6ee5cb..f069a82deb57 100644
--- a/drivers/gpu/drm/i915/intel_dp.c
+++ b/drivers/gpu/drm/i915/intel_dp.c
@@ -203,6 +203,7 @@ intel_dp_mode_valid(struct drm_connector *connector,
 	struct drm_display_mode *fixed_mode = intel_connector->panel.fixed_mode;
 	int target_clock = mode->clock;
 	int max_rate, mode_rate, max_lanes, max_link_clock;
+	int max_dotclk = to_i915(connector->dev)->max_dotclk_freq;
 
 	if (is_edp(intel_dp) && fixed_mode) {
 		if (mode->hdisplay > fixed_mode->hdisplay)
@@ -220,7 +221,7 @@ intel_dp_mode_valid(struct drm_connector *connector,
 	max_rate = intel_dp_max_data_rate(max_link_clock, max_lanes);
 	mode_rate = intel_dp_link_required(target_clock, 18);
 
-	if (mode_rate > max_rate)
+	if (mode_rate > max_rate || target_clock > max_dotclk)
 		return MODE_CLOCK_HIGH;
 
 	if (mode->clock < 10000)
@@ -979,7 +980,10 @@ intel_dp_aux_transfer(struct drm_dp_aux *aux, struct drm_dp_aux_msg *msg)
 		if (WARN_ON(txsize > 20))
 			return -E2BIG;
 
-		memcpy(txbuf + HEADER_SIZE, msg->buffer, msg->size);
+		if (msg->buffer)
+			memcpy(txbuf + HEADER_SIZE, msg->buffer, msg->size);
+		else
+			WARN_ON(msg->size);
 
 		ret = intel_dp_aux_ch(intel_dp, txbuf, txsize, rxbuf, rxsize);
 		if (ret > 0) {
@@ -1798,12 +1802,21 @@ static void wait_panel_off(struct intel_dp *intel_dp)
 
 static void wait_panel_power_cycle(struct intel_dp *intel_dp)
 {
+	ktime_t panel_power_on_time;
+	s64 panel_power_off_duration;
+
 	DRM_DEBUG_KMS("Wait for panel power cycle\n");
 
+	/* take the difference of currrent time and panel power off time
+	 * and then make panel wait for t11_t12 if needed. */
+	panel_power_on_time = ktime_get_boottime();
+	panel_power_off_duration = ktime_ms_delta(panel_power_on_time, intel_dp->panel_power_off_time);
+
 	/* When we disable the VDD override bit last we have to do the manual
 	 * wait. */
-	wait_remaining_ms_from_jiffies(intel_dp->last_power_cycle,
-				       intel_dp->panel_power_cycle_delay);
+	if (panel_power_off_duration < (s64)intel_dp->panel_power_cycle_delay)
+		wait_remaining_ms_from_jiffies(jiffies,
+				       intel_dp->panel_power_cycle_delay - panel_power_off_duration);
 
 	wait_panel_status(intel_dp, IDLE_CYCLE_MASK, IDLE_CYCLE_VALUE);
 }
@@ -1955,7 +1968,7 @@ static void edp_panel_vdd_off_sync(struct intel_dp *intel_dp)
 	I915_READ(pp_stat_reg), I915_READ(pp_ctrl_reg));
 
 	if ((pp & POWER_TARGET_ON) == 0)
-		intel_dp->last_power_cycle = jiffies;
+		intel_dp->panel_power_off_time = ktime_get_boottime();
 
 	power_domain = intel_display_port_aux_power_domain(intel_encoder);
 	intel_display_power_put(dev_priv, power_domain);
@@ -2104,7 +2117,7 @@ static void edp_panel_off(struct intel_dp *intel_dp)
 	I915_WRITE(pp_ctrl_reg, pp);
 	POSTING_READ(pp_ctrl_reg);
 
-	intel_dp->last_power_cycle = jiffies;
+	intel_dp->panel_power_off_time = ktime_get_boottime();
 	wait_panel_off(intel_dp);
 
 	/* We got a reference when we enabled the VDD. */
@@ -2343,15 +2356,18 @@ static bool intel_dp_get_hw_state(struct intel_encoder *encoder,
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	enum intel_display_power_domain power_domain;
 	u32 tmp;
+	bool ret;
 
 	power_domain = intel_display_port_power_domain(encoder);
-	if (!intel_display_power_is_enabled(dev_priv, power_domain))
+	if (!intel_display_power_get_if_enabled(dev_priv, power_domain))
 		return false;
 
+	ret = false;
+
 	tmp = I915_READ(intel_dp->output_reg);
 
 	if (!(tmp & DP_PORT_EN))
-		return false;
+		goto out;
 
 	if (IS_GEN7(dev) && port == PORT_A) {
 		*pipe = PORT_TO_PIPE_CPT(tmp);
@@ -2362,7 +2378,9 @@ static bool intel_dp_get_hw_state(struct intel_encoder *encoder,
 			u32 trans_dp = I915_READ(TRANS_DP_CTL(p));
 			if (TRANS_DP_PIPE_TO_PORT(trans_dp) == port) {
 				*pipe = p;
-				return true;
+				ret = true;
+
+				goto out;
 			}
 		}
 
@@ -2374,7 +2392,12 @@ static bool intel_dp_get_hw_state(struct intel_encoder *encoder,
 		*pipe = PORT_TO_PIPE(tmp);
 	}
 
-	return true;
+	ret = true;
+
+out:
+	intel_display_power_put(dev_priv, power_domain);
+
+	return ret;
 }
 
 static void intel_dp_get_config(struct intel_encoder *encoder,
@@ -3995,7 +4018,7 @@ static int intel_dp_sink_crc_stop(struct intel_dp *intel_dp)
 	} while (--attempts && count);
 
 	if (attempts == 0) {
-		DRM_ERROR("TIMEOUT: Sink CRC counter is not zeroed\n");
+		DRM_DEBUG_KMS("TIMEOUT: Sink CRC counter is not zeroed after calculation is stopped\n");
 		ret = -ETIMEDOUT;
 	}
 
@@ -4474,20 +4497,20 @@ static bool g4x_digital_port_connected(struct drm_i915_private *dev_priv,
 	return I915_READ(PORT_HOTPLUG_STAT) & bit;
 }
 
-static bool vlv_digital_port_connected(struct drm_i915_private *dev_priv,
-				       struct intel_digital_port *port)
+static bool gm45_digital_port_connected(struct drm_i915_private *dev_priv,
+					struct intel_digital_port *port)
 {
 	u32 bit;
 
 	switch (port->port) {
 	case PORT_B:
-		bit = PORTB_HOTPLUG_LIVE_STATUS_VLV;
+		bit = PORTB_HOTPLUG_LIVE_STATUS_GM45;
 		break;
 	case PORT_C:
-		bit = PORTC_HOTPLUG_LIVE_STATUS_VLV;
+		bit = PORTC_HOTPLUG_LIVE_STATUS_GM45;
 		break;
 	case PORT_D:
-		bit = PORTD_HOTPLUG_LIVE_STATUS_VLV;
+		bit = PORTD_HOTPLUG_LIVE_STATUS_GM45;
 		break;
 	default:
 		MISSING_CASE(port->port);
@@ -4535,12 +4558,12 @@ bool intel_digital_port_connected(struct drm_i915_private *dev_priv,
 {
 	if (HAS_PCH_IBX(dev_priv))
 		return ibx_digital_port_connected(dev_priv, port);
-	if (HAS_PCH_SPLIT(dev_priv))
+	else if (HAS_PCH_SPLIT(dev_priv))
 		return cpt_digital_port_connected(dev_priv, port);
 	else if (IS_BROXTON(dev_priv))
 		return bxt_digital_port_connected(dev_priv, port);
-	else if (IS_VALLEYVIEW(dev_priv) || IS_CHERRYVIEW(dev_priv))
-		return vlv_digital_port_connected(dev_priv, port);
+	else if (IS_GM45(dev_priv))
+		return gm45_digital_port_connected(dev_priv, port);
 	else
 		return g4x_digital_port_connected(dev_priv, port);
 }
@@ -5102,7 +5125,7 @@ intel_dp_add_properties(struct intel_dp *intel_dp, struct drm_connector *connect
 
 static void intel_dp_init_panel_power_timestamps(struct intel_dp *intel_dp)
 {
-	intel_dp->last_power_cycle = jiffies;
+	intel_dp->panel_power_off_time = ktime_get_boottime();
 	intel_dp->last_power_on = jiffies;
 	intel_dp->last_backlight_off = jiffies;
 }
@@ -6020,7 +6043,6 @@ intel_dp_init(struct drm_device *dev,
 	}
 
 	intel_dig_port->port = port;
-	dev_priv->dig_port_map[port] = intel_encoder;
 	intel_dig_port->dp.output_reg = output_reg;
 	intel_dig_port->max_lanes = 4;
 
diff --git a/drivers/gpu/drm/i915/intel_dp_link_training.c b/drivers/gpu/drm/i915/intel_dp_link_training.c
index 88887938e0bf..0b8eefc2acc5 100644
--- a/drivers/gpu/drm/i915/intel_dp_link_training.c
+++ b/drivers/gpu/drm/i915/intel_dp_link_training.c
@@ -215,27 +215,46 @@ intel_dp_link_training_clock_recovery(struct intel_dp *intel_dp)
 	}
 }
 
-static void
-intel_dp_link_training_channel_equalization(struct intel_dp *intel_dp)
+/*
+ * Pick training pattern for channel equalization. Training Pattern 3 for HBR2
+ * or 1.2 devices that support it, Training Pattern 2 otherwise.
+ */
+static u32 intel_dp_training_pattern(struct intel_dp *intel_dp)
 {
-	bool channel_eq = false;
-	int tries, cr_tries;
-	uint32_t training_pattern = DP_TRAINING_PATTERN_2;
+	u32 training_pattern = DP_TRAINING_PATTERN_2;
+	bool source_tps3, sink_tps3;
 
 	/*
-	 * Training Pattern 3 for HBR2 or 1.2 devices that support it.
-	 *
 	 * Intel platforms that support HBR2 also support TPS3. TPS3 support is
-	 * also mandatory for downstream devices that support HBR2.
+	 * also mandatory for downstream devices that support HBR2. However, not
+	 * all sinks follow the spec.
 	 *
 	 * Due to WaDisableHBR2 SKL < B0 is the only exception where TPS3 is
-	 * supported but still not enabled.
+	 * supported in source but still not enabled.
 	 */
-	if (intel_dp_source_supports_hbr2(intel_dp) &&
-	    drm_dp_tps3_supported(intel_dp->dpcd))
+	source_tps3 = intel_dp_source_supports_hbr2(intel_dp);
+	sink_tps3 = drm_dp_tps3_supported(intel_dp->dpcd);
+
+	if (source_tps3 && sink_tps3) {
 		training_pattern = DP_TRAINING_PATTERN_3;
-	else if (intel_dp->link_rate == 540000)
-		DRM_ERROR("5.4 Gbps link rate without HBR2/TPS3 support\n");
+	} else if (intel_dp->link_rate == 540000) {
+		if (!source_tps3)
+			DRM_DEBUG_KMS("5.4 Gbps link rate without source HBR2/TPS3 support\n");
+		if (!sink_tps3)
+			DRM_DEBUG_KMS("5.4 Gbps link rate without sink TPS3 support\n");
+	}
+
+	return training_pattern;
+}
+
+static void
+intel_dp_link_training_channel_equalization(struct intel_dp *intel_dp)
+{
+	bool channel_eq = false;
+	int tries, cr_tries;
+	u32 training_pattern;
+
+	training_pattern = intel_dp_training_pattern(intel_dp);
 
 	/* channel equalization */
 	if (!intel_dp_set_link_train(intel_dp,
diff --git a/drivers/gpu/drm/i915/intel_dp_mst.c b/drivers/gpu/drm/i915/intel_dp_mst.c
index 2a2ab306ad84..a2bd698fe2f7 100644
--- a/drivers/gpu/drm/i915/intel_dp_mst.c
+++ b/drivers/gpu/drm/i915/intel_dp_mst.c
@@ -371,6 +371,8 @@ static enum drm_mode_status
 intel_dp_mst_mode_valid(struct drm_connector *connector,
 			struct drm_display_mode *mode)
 {
+	int max_dotclk = to_i915(connector->dev)->max_dotclk_freq;
+
 	/* TODO - validate mode against available PBN for link */
 	if (mode->clock < 10000)
 		return MODE_CLOCK_LOW;
@@ -378,6 +380,9 @@ intel_dp_mst_mode_valid(struct drm_connector *connector,
 	if (mode->flags & DRM_MODE_FLAG_DBLCLK)
 		return MODE_H_ILLEGAL;
 
+	if (mode->clock > max_dotclk)
+		return MODE_CLOCK_HIGH;
+
 	return MODE_OK;
 }
 
diff --git a/drivers/gpu/drm/i915/intel_drv.h b/drivers/gpu/drm/i915/intel_drv.h
index bf6f98134b50..4c027d69fac9 100644
--- a/drivers/gpu/drm/i915/intel_drv.h
+++ b/drivers/gpu/drm/i915/intel_drv.h
@@ -379,6 +379,7 @@ struct intel_crtc_state {
 	bool update_pipe; /* can a fast modeset be performed? */
 	bool disable_cxsr;
 	bool wm_changed; /* watermarks are updated */
+	bool fb_changed; /* fb on any of the planes is changed */
 
 	/* Pipe source size (ie. panel fitter input size)
 	 * All planes will be positioned inside this space,
@@ -492,6 +493,8 @@ struct intel_crtc_state {
 
 	bool ips_enabled;
 
+	bool enable_fbc;
+
 	bool double_wide;
 
 	bool dp_encoder_is_mst;
@@ -542,16 +545,13 @@ struct intel_mmio_flip {
  */
 struct intel_crtc_atomic_commit {
 	/* Sleepable operations to perform before commit */
-	bool disable_fbc;
-	bool disable_ips;
-	bool pre_disable_primary;
 
 	/* Sleepable operations to perform after commit */
 	unsigned fb_bits;
-	bool wait_vblank;
-	bool update_fbc;
 	bool post_enable_primary;
-	unsigned update_sprite_watermarks;
+
+	/* Sleepable operations to perform before and after commit */
+	bool update_fbc;
 };
 
 struct intel_crtc {
@@ -575,7 +575,7 @@ struct intel_crtc {
 	/* Display surface base address adjustement for pageflips. Note that on
 	 * gen4+ this only adjusts up to a tile, offsets within a tile are
 	 * handled in the hw itself (with the TILEOFF register). */
-	unsigned long dspaddr_offset;
+	u32 dspaddr_offset;
 	int adjusted_x;
 	int adjusted_y;
 
@@ -770,9 +770,9 @@ struct intel_dp {
 	int backlight_off_delay;
 	struct delayed_work panel_vdd_work;
 	bool want_panel_vdd;
-	unsigned long last_power_cycle;
 	unsigned long last_power_on;
 	unsigned long last_backlight_off;
+	ktime_t panel_power_off_time;
 
 	struct notifier_block edp_notifier;
 
@@ -909,9 +909,7 @@ struct intel_unpin_work {
 };
 
 struct intel_load_detect_pipe {
-	struct drm_framebuffer *release_fb;
-	bool load_detect_temp;
-	int dpms_mode;
+	struct drm_atomic_state *restore_state;
 };
 
 static inline struct intel_encoder *
@@ -994,6 +992,8 @@ static inline bool intel_irqs_enabled(struct drm_i915_private *dev_priv)
 int intel_get_crtc_scanline(struct intel_crtc *crtc);
 void gen8_irq_power_well_post_enable(struct drm_i915_private *dev_priv,
 				     unsigned int pipe_mask);
+void gen8_irq_power_well_pre_disable(struct drm_i915_private *dev_priv,
+				     unsigned int pipe_mask);
 
 /* intel_crt.c */
 void intel_crt_init(struct drm_device *dev);
@@ -1172,11 +1172,11 @@ void assert_fdi_rx_pll(struct drm_i915_private *dev_priv,
 void assert_pipe(struct drm_i915_private *dev_priv, enum pipe pipe, bool state);
 #define assert_pipe_enabled(d, p) assert_pipe(d, p, true)
 #define assert_pipe_disabled(d, p) assert_pipe(d, p, false)
-unsigned long intel_compute_tile_offset(struct drm_i915_private *dev_priv,
-					int *x, int *y,
-					uint64_t fb_modifier,
-					unsigned int cpp,
-					unsigned int pitch);
+u32 intel_compute_tile_offset(struct drm_i915_private *dev_priv,
+			      int *x, int *y,
+			      uint64_t fb_modifier,
+			      unsigned int cpp,
+			      unsigned int pitch);
 void intel_prepare_reset(struct drm_device *dev);
 void intel_finish_reset(struct drm_device *dev);
 void hsw_enable_pc8(struct drm_i915_private *dev_priv);
@@ -1226,7 +1226,7 @@ u32 skl_plane_ctl_rotation(unsigned int rotation);
 
 /* intel_csr.c */
 void intel_csr_ucode_init(struct drm_i915_private *);
-void intel_csr_load_program(struct drm_i915_private *);
+bool intel_csr_load_program(struct drm_i915_private *);
 void intel_csr_ucode_fini(struct drm_i915_private *);
 
 /* intel_dp.c */
@@ -1327,13 +1327,16 @@ static inline void intel_fbdev_restore_mode(struct drm_device *dev)
 #endif
 
 /* intel_fbc.c */
+void intel_fbc_choose_crtc(struct drm_i915_private *dev_priv,
+			   struct drm_atomic_state *state);
 bool intel_fbc_is_active(struct drm_i915_private *dev_priv);
-void intel_fbc_deactivate(struct intel_crtc *crtc);
-void intel_fbc_update(struct intel_crtc *crtc);
+void intel_fbc_pre_update(struct intel_crtc *crtc);
+void intel_fbc_post_update(struct intel_crtc *crtc);
 void intel_fbc_init(struct drm_i915_private *dev_priv);
+void intel_fbc_init_pipe_state(struct drm_i915_private *dev_priv);
 void intel_fbc_enable(struct intel_crtc *crtc);
-void intel_fbc_disable(struct drm_i915_private *dev_priv);
-void intel_fbc_disable_crtc(struct intel_crtc *crtc);
+void intel_fbc_disable(struct intel_crtc *crtc);
+void intel_fbc_global_disable(struct drm_i915_private *dev_priv);
 void intel_fbc_invalidate(struct drm_i915_private *dev_priv,
 			  unsigned int frontbuffer_bits,
 			  enum fb_op_origin origin);
@@ -1432,6 +1435,8 @@ bool __intel_display_power_is_enabled(struct drm_i915_private *dev_priv,
 				      enum intel_display_power_domain domain);
 void intel_display_power_get(struct drm_i915_private *dev_priv,
 			     enum intel_display_power_domain domain);
+bool intel_display_power_get_if_enabled(struct drm_i915_private *dev_priv,
+					enum intel_display_power_domain domain);
 void intel_display_power_put(struct drm_i915_private *dev_priv,
 			     enum intel_display_power_domain domain);
 
@@ -1518,6 +1523,7 @@ enable_rpm_wakeref_asserts(struct drm_i915_private *dev_priv)
 	enable_rpm_wakeref_asserts(dev_priv)
 
 void intel_runtime_pm_get(struct drm_i915_private *dev_priv);
+bool intel_runtime_pm_get_if_in_use(struct drm_i915_private *dev_priv);
 void intel_runtime_pm_get_noresume(struct drm_i915_private *dev_priv);
 void intel_runtime_pm_put(struct drm_i915_private *dev_priv);
 
@@ -1559,6 +1565,7 @@ void skl_wm_get_hw_state(struct drm_device *dev);
 void skl_ddb_get_hw_state(struct drm_i915_private *dev_priv,
 			  struct skl_ddb_allocation *ddb /* out */);
 uint32_t ilk_pipe_pixel_rate(const struct intel_crtc_state *pipe_config);
+int sanitize_rc6_option(const struct drm_device *dev, int enable_rc6);
 
 /* intel_sdvo.c */
 bool intel_sdvo_init(struct drm_device *dev,
diff --git a/drivers/gpu/drm/i915/intel_dsi.c b/drivers/gpu/drm/i915/intel_dsi.c
index 91cef3525c93..01b8e9f4c272 100644
--- a/drivers/gpu/drm/i915/intel_dsi.c
+++ b/drivers/gpu/drm/i915/intel_dsi.c
@@ -478,8 +478,8 @@ static void intel_dsi_pre_enable(struct intel_encoder *encoder)
 
 	DRM_DEBUG_KMS("\n");
 
-	intel_dsi_prepare(encoder);
 	intel_enable_dsi_pll(encoder);
+	intel_dsi_prepare(encoder);
 
 	/* Panel Enable over CRC PMIC */
 	if (intel_dsi->gpio_panel)
@@ -634,7 +634,6 @@ static void intel_dsi_post_disable(struct intel_encoder *encoder)
 {
 	struct drm_i915_private *dev_priv = encoder->base.dev->dev_private;
 	struct intel_dsi *intel_dsi = enc_to_intel_dsi(&encoder->base);
-	u32 val;
 
 	DRM_DEBUG_KMS("\n");
 
@@ -642,9 +641,13 @@ static void intel_dsi_post_disable(struct intel_encoder *encoder)
 
 	intel_dsi_clear_device_ready(encoder);
 
-	val = I915_READ(DSPCLK_GATE_D);
-	val &= ~DPOUNIT_CLOCK_GATE_DISABLE;
-	I915_WRITE(DSPCLK_GATE_D, val);
+	if (!IS_BROXTON(dev_priv)) {
+		u32 val;
+
+		val = I915_READ(DSPCLK_GATE_D);
+		val &= ~DPOUNIT_CLOCK_GATE_DISABLE;
+		I915_WRITE(DSPCLK_GATE_D, val);
+	}
 
 	drm_panel_unprepare(intel_dsi->panel);
 
@@ -664,13 +667,16 @@ static bool intel_dsi_get_hw_state(struct intel_encoder *encoder,
 	struct drm_device *dev = encoder->base.dev;
 	enum intel_display_power_domain power_domain;
 	enum port port;
+	bool ret;
 
 	DRM_DEBUG_KMS("\n");
 
 	power_domain = intel_display_port_power_domain(encoder);
-	if (!intel_display_power_is_enabled(dev_priv, power_domain))
+	if (!intel_display_power_get_if_enabled(dev_priv, power_domain))
 		return false;
 
+	ret = false;
+
 	/* XXX: this only works for one DSI output */
 	for_each_dsi_port(port, intel_dsi->ports) {
 		i915_reg_t ctrl_reg = IS_BROXTON(dev) ?
@@ -691,12 +697,16 @@ static bool intel_dsi_get_hw_state(struct intel_encoder *encoder,
 		if (dpi_enabled || (func & CMD_MODE_DATA_WIDTH_MASK)) {
 			if (I915_READ(MIPI_DEVICE_READY(port)) & DEVICE_READY) {
 				*pipe = port == PORT_A ? PIPE_A : PIPE_B;
-				return true;
+				ret = true;
+
+				goto out;
 			}
 		}
 	}
+out:
+	intel_display_power_put(dev_priv, power_domain);
 
-	return false;
+	return ret;
 }
 
 static void intel_dsi_get_config(struct intel_encoder *encoder,
@@ -775,10 +785,9 @@ static void set_dsi_timings(struct drm_encoder *encoder,
 {
 	struct drm_device *dev = encoder->dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
-	struct intel_crtc *intel_crtc = to_intel_crtc(encoder->crtc);
 	struct intel_dsi *intel_dsi = enc_to_intel_dsi(encoder);
 	enum port port;
-	unsigned int bpp = intel_crtc->config->pipe_bpp;
+	unsigned int bpp = dsi_pixel_format_bpp(intel_dsi->pixel_format);
 	unsigned int lane_count = intel_dsi->lane_count;
 
 	u16 hactive, hfp, hsync, hbp, vfp, vsync, vbp;
@@ -849,7 +858,7 @@ static void intel_dsi_prepare(struct intel_encoder *intel_encoder)
 	struct intel_dsi *intel_dsi = enc_to_intel_dsi(encoder);
 	const struct drm_display_mode *adjusted_mode = &intel_crtc->config->base.adjusted_mode;
 	enum port port;
-	unsigned int bpp = intel_crtc->config->pipe_bpp;
+	unsigned int bpp = dsi_pixel_format_bpp(intel_dsi->pixel_format);
 	u32 val, tmp;
 	u16 mode_hdisplay;
 
diff --git a/drivers/gpu/drm/i915/intel_dsi.h b/drivers/gpu/drm/i915/intel_dsi.h
index de7be7f3fb42..92f39227b361 100644
--- a/drivers/gpu/drm/i915/intel_dsi.h
+++ b/drivers/gpu/drm/i915/intel_dsi.h
@@ -34,6 +34,8 @@
 #define DSI_DUAL_LINK_FRONT_BACK	1
 #define DSI_DUAL_LINK_PIXEL_ALT		2
 
+int dsi_pixel_format_bpp(int pixel_format);
+
 struct intel_dsi_host;
 
 struct intel_dsi {
diff --git a/drivers/gpu/drm/i915/intel_dsi_panel_vbt.c b/drivers/gpu/drm/i915/intel_dsi_panel_vbt.c
index 1d43e6f37fc1..7f145b4fec6a 100644
--- a/drivers/gpu/drm/i915/intel_dsi_panel_vbt.c
+++ b/drivers/gpu/drm/i915/intel_dsi_panel_vbt.c
@@ -204,10 +204,28 @@ static const u8 *mipi_exec_gpio(struct intel_dsi *intel_dsi, const u8 *data)
 	struct drm_device *dev = intel_dsi->base.base.dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
 
+	if (dev_priv->vbt.dsi.seq_version >= 3)
+		data++;
+
 	gpio = *data++;
 
 	/* pull up/down */
-	action = *data++;
+	action = *data++ & 1;
+
+	if (gpio >= ARRAY_SIZE(gtable)) {
+		DRM_DEBUG_KMS("unknown gpio %u\n", gpio);
+		goto out;
+	}
+
+	if (!IS_VALLEYVIEW(dev_priv)) {
+		DRM_DEBUG_KMS("GPIO element not supported on this platform\n");
+		goto out;
+	}
+
+	if (dev_priv->vbt.dsi.seq_version >= 3) {
+		DRM_DEBUG_KMS("GPIO element v3 not supported\n");
+		goto out;
+	}
 
 	function = gtable[gpio].function_reg;
 	pad = gtable[gpio].pad_reg;
@@ -216,16 +234,18 @@ static const u8 *mipi_exec_gpio(struct intel_dsi *intel_dsi, const u8 *data)
 	if (!gtable[gpio].init) {
 		/* program the function */
 		/* FIXME: remove constant below */
-		vlv_gpio_nc_write(dev_priv, function, 0x2000CC00);
+		vlv_iosf_sb_write(dev_priv, IOSF_PORT_GPIO_NC, function,
+				  0x2000CC00);
 		gtable[gpio].init = 1;
 	}
 
 	val = 0x4 | action;
 
 	/* pull up/down */
-	vlv_gpio_nc_write(dev_priv, pad, val);
+	vlv_iosf_sb_write(dev_priv, IOSF_PORT_GPIO_NC, pad, val);
 	mutex_unlock(&dev_priv->sb_lock);
 
+out:
 	return data;
 }
 
@@ -420,10 +440,7 @@ struct drm_panel *vbt_panel_init(struct intel_dsi *intel_dsi, u16 panel_id)
 	intel_dsi->dual_link = mipi_config->dual_link;
 	intel_dsi->pixel_overlap = mipi_config->pixel_overlap;
 
-	if (intel_dsi->pixel_format == VID_MODE_FORMAT_RGB666)
-		bits_per_pixel = 18;
-	else if (intel_dsi->pixel_format == VID_MODE_FORMAT_RGB565)
-		bits_per_pixel = 16;
+	bits_per_pixel = dsi_pixel_format_bpp(intel_dsi->pixel_format);
 
 	intel_dsi->operation_mode = mipi_config->is_cmd_mode;
 	intel_dsi->video_mode_format = mipi_config->video_transfer_mode;
diff --git a/drivers/gpu/drm/i915/intel_dsi_pll.c b/drivers/gpu/drm/i915/intel_dsi_pll.c
index bb5e95a1a453..70883c54cb0a 100644
--- a/drivers/gpu/drm/i915/intel_dsi_pll.c
+++ b/drivers/gpu/drm/i915/intel_dsi_pll.c
@@ -30,7 +30,7 @@
 #include "i915_drv.h"
 #include "intel_dsi.h"
 
-static int dsi_pixel_format_bpp(int pixel_format)
+int dsi_pixel_format_bpp(int pixel_format)
 {
 	int bpp;
 
diff --git a/drivers/gpu/drm/i915/intel_fbc.c b/drivers/gpu/drm/i915/intel_fbc.c
index a1988a486b92..0f0492f4a357 100644
--- a/drivers/gpu/drm/i915/intel_fbc.c
+++ b/drivers/gpu/drm/i915/intel_fbc.c
@@ -43,7 +43,7 @@
 
 static inline bool fbc_supported(struct drm_i915_private *dev_priv)
 {
-	return dev_priv->fbc.activate != NULL;
+	return HAS_FBC(dev_priv);
 }
 
 static inline bool fbc_on_pipe_a_only(struct drm_i915_private *dev_priv)
@@ -56,6 +56,11 @@ static inline bool fbc_on_plane_a_only(struct drm_i915_private *dev_priv)
 	return INTEL_INFO(dev_priv)->gen < 4;
 }
 
+static inline bool no_fbc_on_multiple_pipes(struct drm_i915_private *dev_priv)
+{
+	return INTEL_INFO(dev_priv)->gen <= 3;
+}
+
 /*
  * In some platforms where the CRTC's x:0/y:0 coordinates doesn't match the
  * frontbuffer's x:0/y:0 coordinates we lie to the hardware about the plane's
@@ -74,19 +79,17 @@ static unsigned int get_crtc_fence_y_offset(struct intel_crtc *crtc)
  * write to the PLANE_SIZE register. For BDW-, the hardware looks at the value
  * we wrote to PIPESRC.
  */
-static void intel_fbc_get_plane_source_size(struct intel_crtc *crtc,
+static void intel_fbc_get_plane_source_size(struct intel_fbc_state_cache *cache,
 					    int *width, int *height)
 {
-	struct intel_plane_state *plane_state =
-			to_intel_plane_state(crtc->base.primary->state);
 	int w, h;
 
-	if (intel_rotation_90_or_270(plane_state->base.rotation)) {
-		w = drm_rect_height(&plane_state->src) >> 16;
-		h = drm_rect_width(&plane_state->src) >> 16;
+	if (intel_rotation_90_or_270(cache->plane.rotation)) {
+		w = cache->plane.src_h;
+		h = cache->plane.src_w;
 	} else {
-		w = drm_rect_width(&plane_state->src) >> 16;
-		h = drm_rect_height(&plane_state->src) >> 16;
+		w = cache->plane.src_w;
+		h = cache->plane.src_h;
 	}
 
 	if (width)
@@ -95,26 +98,23 @@ static void intel_fbc_get_plane_source_size(struct intel_crtc *crtc,
 		*height = h;
 }
 
-static int intel_fbc_calculate_cfb_size(struct intel_crtc *crtc,
-					struct drm_framebuffer *fb)
+static int intel_fbc_calculate_cfb_size(struct drm_i915_private *dev_priv,
+					struct intel_fbc_state_cache *cache)
 {
-	struct drm_i915_private *dev_priv = crtc->base.dev->dev_private;
 	int lines;
 
-	intel_fbc_get_plane_source_size(crtc, NULL, &lines);
+	intel_fbc_get_plane_source_size(cache, NULL, &lines);
 	if (INTEL_INFO(dev_priv)->gen >= 7)
 		lines = min(lines, 2048);
 
 	/* Hardware needs the full buffer stride, not just the active area. */
-	return lines * fb->pitches[0];
+	return lines * cache->fb.stride;
 }
 
 static void i8xx_fbc_deactivate(struct drm_i915_private *dev_priv)
 {
 	u32 fbc_ctl;
 
-	dev_priv->fbc.active = false;
-
 	/* Disable compression */
 	fbc_ctl = I915_READ(FBC_CONTROL);
 	if ((fbc_ctl & FBC_CTL_EN) == 0)
@@ -130,21 +130,17 @@ static void i8xx_fbc_deactivate(struct drm_i915_private *dev_priv)
 	}
 }
 
-static void i8xx_fbc_activate(struct intel_crtc *crtc)
+static void i8xx_fbc_activate(struct drm_i915_private *dev_priv)
 {
-	struct drm_i915_private *dev_priv = crtc->base.dev->dev_private;
-	struct drm_framebuffer *fb = crtc->base.primary->fb;
-	struct drm_i915_gem_object *obj = intel_fb_obj(fb);
+	struct intel_fbc_reg_params *params = &dev_priv->fbc.params;
 	int cfb_pitch;
 	int i;
 	u32 fbc_ctl;
 
-	dev_priv->fbc.active = true;
-
 	/* Note: fbc.threshold == 1 for i8xx */
-	cfb_pitch = intel_fbc_calculate_cfb_size(crtc, fb) / FBC_LL_SIZE;
-	if (fb->pitches[0] < cfb_pitch)
-		cfb_pitch = fb->pitches[0];
+	cfb_pitch = params->cfb_size / FBC_LL_SIZE;
+	if (params->fb.stride < cfb_pitch)
+		cfb_pitch = params->fb.stride;
 
 	/* FBC_CTL wants 32B or 64B units */
 	if (IS_GEN2(dev_priv))
@@ -161,9 +157,9 @@ static void i8xx_fbc_activate(struct intel_crtc *crtc)
 
 		/* Set it up... */
 		fbc_ctl2 = FBC_CTL_FENCE_DBL | FBC_CTL_IDLE_IMM | FBC_CTL_CPU_FENCE;
-		fbc_ctl2 |= FBC_CTL_PLANE(crtc->plane);
+		fbc_ctl2 |= FBC_CTL_PLANE(params->crtc.plane);
 		I915_WRITE(FBC_CONTROL2, fbc_ctl2);
-		I915_WRITE(FBC_FENCE_OFF, get_crtc_fence_y_offset(crtc));
+		I915_WRITE(FBC_FENCE_OFF, params->crtc.fence_y_offset);
 	}
 
 	/* enable it... */
@@ -173,7 +169,7 @@ static void i8xx_fbc_activate(struct intel_crtc *crtc)
 	if (IS_I945GM(dev_priv))
 		fbc_ctl |= FBC_CTL_C3_IDLE; /* 945 needs special SR handling */
 	fbc_ctl |= (cfb_pitch & 0xff) << FBC_CTL_STRIDE_SHIFT;
-	fbc_ctl |= obj->fence_reg;
+	fbc_ctl |= params->fb.fence_reg;
 	I915_WRITE(FBC_CONTROL, fbc_ctl);
 }
 
@@ -182,23 +178,19 @@ static bool i8xx_fbc_is_active(struct drm_i915_private *dev_priv)
 	return I915_READ(FBC_CONTROL) & FBC_CTL_EN;
 }
 
-static void g4x_fbc_activate(struct intel_crtc *crtc)
+static void g4x_fbc_activate(struct drm_i915_private *dev_priv)
 {
-	struct drm_i915_private *dev_priv = crtc->base.dev->dev_private;
-	struct drm_framebuffer *fb = crtc->base.primary->fb;
-	struct drm_i915_gem_object *obj = intel_fb_obj(fb);
+	struct intel_fbc_reg_params *params = &dev_priv->fbc.params;
 	u32 dpfc_ctl;
 
-	dev_priv->fbc.active = true;
-
-	dpfc_ctl = DPFC_CTL_PLANE(crtc->plane) | DPFC_SR_EN;
-	if (drm_format_plane_cpp(fb->pixel_format, 0) == 2)
+	dpfc_ctl = DPFC_CTL_PLANE(params->crtc.plane) | DPFC_SR_EN;
+	if (drm_format_plane_cpp(params->fb.pixel_format, 0) == 2)
 		dpfc_ctl |= DPFC_CTL_LIMIT_2X;
 	else
 		dpfc_ctl |= DPFC_CTL_LIMIT_1X;
-	dpfc_ctl |= DPFC_CTL_FENCE_EN | obj->fence_reg;
+	dpfc_ctl |= DPFC_CTL_FENCE_EN | params->fb.fence_reg;
 
-	I915_WRITE(DPFC_FENCE_YOFF, get_crtc_fence_y_offset(crtc));
+	I915_WRITE(DPFC_FENCE_YOFF, params->crtc.fence_y_offset);
 
 	/* enable it... */
 	I915_WRITE(DPFC_CONTROL, dpfc_ctl | DPFC_CTL_EN);
@@ -208,8 +200,6 @@ static void g4x_fbc_deactivate(struct drm_i915_private *dev_priv)
 {
 	u32 dpfc_ctl;
 
-	dev_priv->fbc.active = false;
-
 	/* Disable compression */
 	dpfc_ctl = I915_READ(DPFC_CONTROL);
 	if (dpfc_ctl & DPFC_CTL_EN) {
@@ -230,19 +220,14 @@ static void intel_fbc_recompress(struct drm_i915_private *dev_priv)
 	POSTING_READ(MSG_FBC_REND_STATE);
 }
 
-static void ilk_fbc_activate(struct intel_crtc *crtc)
+static void ilk_fbc_activate(struct drm_i915_private *dev_priv)
 {
-	struct drm_i915_private *dev_priv = crtc->base.dev->dev_private;
-	struct drm_framebuffer *fb = crtc->base.primary->fb;
-	struct drm_i915_gem_object *obj = intel_fb_obj(fb);
+	struct intel_fbc_reg_params *params = &dev_priv->fbc.params;
 	u32 dpfc_ctl;
 	int threshold = dev_priv->fbc.threshold;
-	unsigned int y_offset;
 
-	dev_priv->fbc.active = true;
-
-	dpfc_ctl = DPFC_CTL_PLANE(crtc->plane);
-	if (drm_format_plane_cpp(fb->pixel_format, 0) == 2)
+	dpfc_ctl = DPFC_CTL_PLANE(params->crtc.plane);
+	if (drm_format_plane_cpp(params->fb.pixel_format, 0) == 2)
 		threshold++;
 
 	switch (threshold) {
@@ -259,18 +244,17 @@ static void ilk_fbc_activate(struct intel_crtc *crtc)
 	}
 	dpfc_ctl |= DPFC_CTL_FENCE_EN;
 	if (IS_GEN5(dev_priv))
-		dpfc_ctl |= obj->fence_reg;
+		dpfc_ctl |= params->fb.fence_reg;
 
-	y_offset = get_crtc_fence_y_offset(crtc);
-	I915_WRITE(ILK_DPFC_FENCE_YOFF, y_offset);
-	I915_WRITE(ILK_FBC_RT_BASE, i915_gem_obj_ggtt_offset(obj) | ILK_FBC_RT_VALID);
+	I915_WRITE(ILK_DPFC_FENCE_YOFF, params->crtc.fence_y_offset);
+	I915_WRITE(ILK_FBC_RT_BASE, params->fb.ggtt_offset | ILK_FBC_RT_VALID);
 	/* enable it... */
 	I915_WRITE(ILK_DPFC_CONTROL, dpfc_ctl | DPFC_CTL_EN);
 
 	if (IS_GEN6(dev_priv)) {
 		I915_WRITE(SNB_DPFC_CTL_SA,
-			   SNB_CPU_FENCE_ENABLE | obj->fence_reg);
-		I915_WRITE(DPFC_CPU_FENCE_OFFSET, y_offset);
+			   SNB_CPU_FENCE_ENABLE | params->fb.fence_reg);
+		I915_WRITE(DPFC_CPU_FENCE_OFFSET, params->crtc.fence_y_offset);
 	}
 
 	intel_fbc_recompress(dev_priv);
@@ -280,8 +264,6 @@ static void ilk_fbc_deactivate(struct drm_i915_private *dev_priv)
 {
 	u32 dpfc_ctl;
 
-	dev_priv->fbc.active = false;
-
 	/* Disable compression */
 	dpfc_ctl = I915_READ(ILK_DPFC_CONTROL);
 	if (dpfc_ctl & DPFC_CTL_EN) {
@@ -295,21 +277,17 @@ static bool ilk_fbc_is_active(struct drm_i915_private *dev_priv)
 	return I915_READ(ILK_DPFC_CONTROL) & DPFC_CTL_EN;
 }
 
-static void gen7_fbc_activate(struct intel_crtc *crtc)
+static void gen7_fbc_activate(struct drm_i915_private *dev_priv)
 {
-	struct drm_i915_private *dev_priv = crtc->base.dev->dev_private;
-	struct drm_framebuffer *fb = crtc->base.primary->fb;
-	struct drm_i915_gem_object *obj = intel_fb_obj(fb);
+	struct intel_fbc_reg_params *params = &dev_priv->fbc.params;
 	u32 dpfc_ctl;
 	int threshold = dev_priv->fbc.threshold;
 
-	dev_priv->fbc.active = true;
-
 	dpfc_ctl = 0;
 	if (IS_IVYBRIDGE(dev_priv))
-		dpfc_ctl |= IVB_DPFC_CTL_PLANE(crtc->plane);
+		dpfc_ctl |= IVB_DPFC_CTL_PLANE(params->crtc.plane);
 
-	if (drm_format_plane_cpp(fb->pixel_format, 0) == 2)
+	if (drm_format_plane_cpp(params->fb.pixel_format, 0) == 2)
 		threshold++;
 
 	switch (threshold) {
@@ -337,20 +315,60 @@ static void gen7_fbc_activate(struct intel_crtc *crtc)
 			   ILK_FBCQ_DIS);
 	} else if (IS_HASWELL(dev_priv) || IS_BROADWELL(dev_priv)) {
 		/* WaFbcAsynchFlipDisableFbcQueue:hsw,bdw */
-		I915_WRITE(CHICKEN_PIPESL_1(crtc->pipe),
-			   I915_READ(CHICKEN_PIPESL_1(crtc->pipe)) |
+		I915_WRITE(CHICKEN_PIPESL_1(params->crtc.pipe),
+			   I915_READ(CHICKEN_PIPESL_1(params->crtc.pipe)) |
 			   HSW_FBCQ_DIS);
 	}
 
 	I915_WRITE(ILK_DPFC_CONTROL, dpfc_ctl | DPFC_CTL_EN);
 
 	I915_WRITE(SNB_DPFC_CTL_SA,
-		   SNB_CPU_FENCE_ENABLE | obj->fence_reg);
-	I915_WRITE(DPFC_CPU_FENCE_OFFSET, get_crtc_fence_y_offset(crtc));
+		   SNB_CPU_FENCE_ENABLE | params->fb.fence_reg);
+	I915_WRITE(DPFC_CPU_FENCE_OFFSET, params->crtc.fence_y_offset);
 
 	intel_fbc_recompress(dev_priv);
 }
 
+static bool intel_fbc_hw_is_active(struct drm_i915_private *dev_priv)
+{
+	if (INTEL_INFO(dev_priv)->gen >= 5)
+		return ilk_fbc_is_active(dev_priv);
+	else if (IS_GM45(dev_priv))
+		return g4x_fbc_is_active(dev_priv);
+	else
+		return i8xx_fbc_is_active(dev_priv);
+}
+
+static void intel_fbc_hw_activate(struct drm_i915_private *dev_priv)
+{
+	struct intel_fbc *fbc = &dev_priv->fbc;
+
+	fbc->active = true;
+
+	if (INTEL_INFO(dev_priv)->gen >= 7)
+		gen7_fbc_activate(dev_priv);
+	else if (INTEL_INFO(dev_priv)->gen >= 5)
+		ilk_fbc_activate(dev_priv);
+	else if (IS_GM45(dev_priv))
+		g4x_fbc_activate(dev_priv);
+	else
+		i8xx_fbc_activate(dev_priv);
+}
+
+static void intel_fbc_hw_deactivate(struct drm_i915_private *dev_priv)
+{
+	struct intel_fbc *fbc = &dev_priv->fbc;
+
+	fbc->active = false;
+
+	if (INTEL_INFO(dev_priv)->gen >= 5)
+		ilk_fbc_deactivate(dev_priv);
+	else if (IS_GM45(dev_priv))
+		g4x_fbc_deactivate(dev_priv);
+	else
+		i8xx_fbc_deactivate(dev_priv);
+}
+
 /**
  * intel_fbc_is_active - Is FBC active?
  * @dev_priv: i915 device instance
@@ -364,24 +382,24 @@ bool intel_fbc_is_active(struct drm_i915_private *dev_priv)
 	return dev_priv->fbc.active;
 }
 
-static void intel_fbc_activate(const struct drm_framebuffer *fb)
-{
-	struct drm_i915_private *dev_priv = fb->dev->dev_private;
-	struct intel_crtc *crtc = dev_priv->fbc.crtc;
-
-	dev_priv->fbc.activate(crtc);
-
-	dev_priv->fbc.fb_id = fb->base.id;
-	dev_priv->fbc.y = crtc->base.y;
-}
-
 static void intel_fbc_work_fn(struct work_struct *__work)
 {
 	struct drm_i915_private *dev_priv =
 		container_of(__work, struct drm_i915_private, fbc.work.work);
-	struct intel_fbc_work *work = &dev_priv->fbc.work;
-	struct intel_crtc *crtc = dev_priv->fbc.crtc;
-	int delay_ms = 50;
+	struct intel_fbc *fbc = &dev_priv->fbc;
+	struct intel_fbc_work *work = &fbc->work;
+	struct intel_crtc *crtc = fbc->crtc;
+	struct drm_vblank_crtc *vblank = &dev_priv->dev->vblank[crtc->pipe];
+
+	if (drm_crtc_vblank_get(&crtc->base)) {
+		DRM_ERROR("vblank not available for FBC on pipe %c\n",
+			  pipe_name(crtc->pipe));
+
+		mutex_lock(&fbc->lock);
+		work->scheduled = false;
+		mutex_unlock(&fbc->lock);
+		return;
+	}
 
 retry:
 	/* Delay the actual enabling to let pageflipping cease and the
@@ -390,142 +408,97 @@ retry:
 	 * vblank to pass after disabling the FBC before we attempt
 	 * to modify the control registers.
 	 *
-	 * A more complicated solution would involve tracking vblanks
-	 * following the termination of the page-flipping sequence
-	 * and indeed performing the enable as a co-routine and not
-	 * waiting synchronously upon the vblank.
-	 *
 	 * WaFbcWaitForVBlankBeforeEnable:ilk,snb
+	 *
+	 * It is also worth mentioning that since work->scheduled_vblank can be
+	 * updated multiple times by the other threads, hitting the timeout is
+	 * not an error condition. We'll just end up hitting the "goto retry"
+	 * case below.
 	 */
-	wait_remaining_ms_from_jiffies(work->enable_jiffies, delay_ms);
+	wait_event_timeout(vblank->queue,
+		drm_crtc_vblank_count(&crtc->base) != work->scheduled_vblank,
+		msecs_to_jiffies(50));
 
-	mutex_lock(&dev_priv->fbc.lock);
+	mutex_lock(&fbc->lock);
 
 	/* Were we cancelled? */
 	if (!work->scheduled)
 		goto out;
 
 	/* Were we delayed again while this function was sleeping? */
-	if (time_after(work->enable_jiffies + msecs_to_jiffies(delay_ms),
-		       jiffies)) {
-		mutex_unlock(&dev_priv->fbc.lock);
+	if (drm_crtc_vblank_count(&crtc->base) == work->scheduled_vblank) {
+		mutex_unlock(&fbc->lock);
 		goto retry;
 	}
 
-	if (crtc->base.primary->fb == work->fb)
-		intel_fbc_activate(work->fb);
+	intel_fbc_hw_activate(dev_priv);
 
 	work->scheduled = false;
 
 out:
-	mutex_unlock(&dev_priv->fbc.lock);
-}
-
-static void intel_fbc_cancel_work(struct drm_i915_private *dev_priv)
-{
-	WARN_ON(!mutex_is_locked(&dev_priv->fbc.lock));
-	dev_priv->fbc.work.scheduled = false;
+	mutex_unlock(&fbc->lock);
+	drm_crtc_vblank_put(&crtc->base);
 }
 
 static void intel_fbc_schedule_activation(struct intel_crtc *crtc)
 {
 	struct drm_i915_private *dev_priv = crtc->base.dev->dev_private;
-	struct intel_fbc_work *work = &dev_priv->fbc.work;
+	struct intel_fbc *fbc = &dev_priv->fbc;
+	struct intel_fbc_work *work = &fbc->work;
+
+	WARN_ON(!mutex_is_locked(&fbc->lock));
 
-	WARN_ON(!mutex_is_locked(&dev_priv->fbc.lock));
+	if (drm_crtc_vblank_get(&crtc->base)) {
+		DRM_ERROR("vblank not available for FBC on pipe %c\n",
+			  pipe_name(crtc->pipe));
+		return;
+	}
 
-	/* It is useless to call intel_fbc_cancel_work() in this function since
-	 * we're not releasing fbc.lock, so it won't have an opportunity to grab
-	 * it to discover that it was cancelled. So we just update the expected
-	 * jiffy count. */
-	work->fb = crtc->base.primary->fb;
+	/* It is useless to call intel_fbc_cancel_work() or cancel_work() in
+	 * this function since we're not releasing fbc.lock, so it won't have an
+	 * opportunity to grab it to discover that it was cancelled. So we just
+	 * update the expected jiffy count. */
 	work->scheduled = true;
-	work->enable_jiffies = jiffies;
+	work->scheduled_vblank = drm_crtc_vblank_count(&crtc->base);
+	drm_crtc_vblank_put(&crtc->base);
 
 	schedule_work(&work->work);
 }
 
-static void __intel_fbc_deactivate(struct drm_i915_private *dev_priv)
+static void intel_fbc_deactivate(struct drm_i915_private *dev_priv)
 {
-	WARN_ON(!mutex_is_locked(&dev_priv->fbc.lock));
-
-	intel_fbc_cancel_work(dev_priv);
-
-	if (dev_priv->fbc.active)
-		dev_priv->fbc.deactivate(dev_priv);
-}
-
-/*
- * intel_fbc_deactivate - deactivate FBC if it's associated with crtc
- * @crtc: the CRTC
- *
- * This function deactivates FBC if it's associated with the provided CRTC.
- */
-void intel_fbc_deactivate(struct intel_crtc *crtc)
-{
-	struct drm_i915_private *dev_priv = crtc->base.dev->dev_private;
-
-	if (!fbc_supported(dev_priv))
-		return;
+	struct intel_fbc *fbc = &dev_priv->fbc;
 
-	mutex_lock(&dev_priv->fbc.lock);
-	if (dev_priv->fbc.crtc == crtc)
-		__intel_fbc_deactivate(dev_priv);
-	mutex_unlock(&dev_priv->fbc.lock);
-}
+	WARN_ON(!mutex_is_locked(&fbc->lock));
 
-static void set_no_fbc_reason(struct drm_i915_private *dev_priv,
-			      const char *reason)
-{
-	if (dev_priv->fbc.no_fbc_reason == reason)
-		return;
+	/* Calling cancel_work() here won't help due to the fact that the work
+	 * function grabs fbc->lock. Just set scheduled to false so the work
+	 * function can know it was cancelled. */
+	fbc->work.scheduled = false;
 
-	dev_priv->fbc.no_fbc_reason = reason;
-	DRM_DEBUG_KMS("Disabling FBC: %s\n", reason);
+	if (fbc->active)
+		intel_fbc_hw_deactivate(dev_priv);
 }
 
-static bool crtc_can_fbc(struct intel_crtc *crtc)
+static bool multiple_pipes_ok(struct intel_crtc *crtc)
 {
 	struct drm_i915_private *dev_priv = crtc->base.dev->dev_private;
+	struct drm_plane *primary = crtc->base.primary;
+	struct intel_fbc *fbc = &dev_priv->fbc;
+	enum pipe pipe = crtc->pipe;
 
-	if (fbc_on_pipe_a_only(dev_priv) && crtc->pipe != PIPE_A)
-		return false;
-
-	if (fbc_on_plane_a_only(dev_priv) && crtc->plane != PLANE_A)
-		return false;
-
-	return true;
-}
-
-static bool crtc_is_valid(struct intel_crtc *crtc)
-{
-	if (!intel_crtc_active(&crtc->base))
-		return false;
-
-	if (!to_intel_plane_state(crtc->base.primary->state)->visible)
-		return false;
-
-	return true;
-}
-
-static bool multiple_pipes_ok(struct drm_i915_private *dev_priv)
-{
-	enum pipe pipe;
-	int n_pipes = 0;
-	struct drm_crtc *crtc;
-
-	if (INTEL_INFO(dev_priv)->gen > 4)
+	/* Don't even bother tracking anything we don't need. */
+	if (!no_fbc_on_multiple_pipes(dev_priv))
 		return true;
 
-	for_each_pipe(dev_priv, pipe) {
-		crtc = dev_priv->pipe_to_crtc_mapping[pipe];
+	WARN_ON(!drm_modeset_is_locked(&primary->mutex));
 
-		if (intel_crtc_active(crtc) &&
-		    to_intel_plane_state(crtc->primary->state)->visible)
-			n_pipes++;
-	}
+	if (to_intel_plane_state(primary->state)->visible)
+		fbc->visible_pipes_mask |= (1 << pipe);
+	else
+		fbc->visible_pipes_mask &= ~(1 << pipe);
 
-	return (n_pipes < 2);
+	return (fbc->visible_pipes_mask & ~(1 << pipe)) != 0;
 }
 
 static int find_compression_threshold(struct drm_i915_private *dev_priv,
@@ -581,16 +554,16 @@ again:
 static int intel_fbc_alloc_cfb(struct intel_crtc *crtc)
 {
 	struct drm_i915_private *dev_priv = crtc->base.dev->dev_private;
-	struct drm_framebuffer *fb = crtc->base.primary->state->fb;
+	struct intel_fbc *fbc = &dev_priv->fbc;
 	struct drm_mm_node *uninitialized_var(compressed_llb);
 	int size, fb_cpp, ret;
 
-	WARN_ON(drm_mm_node_allocated(&dev_priv->fbc.compressed_fb));
+	WARN_ON(drm_mm_node_allocated(&fbc->compressed_fb));
 
-	size = intel_fbc_calculate_cfb_size(crtc, fb);
-	fb_cpp = drm_format_plane_cpp(fb->pixel_format, 0);
+	size = intel_fbc_calculate_cfb_size(dev_priv, &fbc->state_cache);
+	fb_cpp = drm_format_plane_cpp(fbc->state_cache.fb.pixel_format, 0);
 
-	ret = find_compression_threshold(dev_priv, &dev_priv->fbc.compressed_fb,
+	ret = find_compression_threshold(dev_priv, &fbc->compressed_fb,
 					 size, fb_cpp);
 	if (!ret)
 		goto err_llb;
@@ -599,12 +572,12 @@ static int intel_fbc_alloc_cfb(struct intel_crtc *crtc)
 
 	}
 
-	dev_priv->fbc.threshold = ret;
+	fbc->threshold = ret;
 
 	if (INTEL_INFO(dev_priv)->gen >= 5)
-		I915_WRITE(ILK_DPFC_CB_BASE, dev_priv->fbc.compressed_fb.start);
+		I915_WRITE(ILK_DPFC_CB_BASE, fbc->compressed_fb.start);
 	else if (IS_GM45(dev_priv)) {
-		I915_WRITE(DPFC_CB_BASE, dev_priv->fbc.compressed_fb.start);
+		I915_WRITE(DPFC_CB_BASE, fbc->compressed_fb.start);
 	} else {
 		compressed_llb = kzalloc(sizeof(*compressed_llb), GFP_KERNEL);
 		if (!compressed_llb)
@@ -615,23 +588,22 @@ static int intel_fbc_alloc_cfb(struct intel_crtc *crtc)
 		if (ret)
 			goto err_fb;
 
-		dev_priv->fbc.compressed_llb = compressed_llb;
+		fbc->compressed_llb = compressed_llb;
 
 		I915_WRITE(FBC_CFB_BASE,
-			   dev_priv->mm.stolen_base + dev_priv->fbc.compressed_fb.start);
+			   dev_priv->mm.stolen_base + fbc->compressed_fb.start);
 		I915_WRITE(FBC_LL_BASE,
 			   dev_priv->mm.stolen_base + compressed_llb->start);
 	}
 
 	DRM_DEBUG_KMS("reserved %llu bytes of contiguous stolen space for FBC, threshold: %d\n",
-		      dev_priv->fbc.compressed_fb.size,
-		      dev_priv->fbc.threshold);
+		      fbc->compressed_fb.size, fbc->threshold);
 
 	return 0;
 
 err_fb:
 	kfree(compressed_llb);
-	i915_gem_stolen_remove_node(dev_priv, &dev_priv->fbc.compressed_fb);
+	i915_gem_stolen_remove_node(dev_priv, &fbc->compressed_fb);
 err_llb:
 	pr_info_once("drm: not enough stolen space for compressed buffer (need %d more bytes), disabling. Hint: you may be able to increase stolen memory size in the BIOS to avoid this.\n", size);
 	return -ENOSPC;
@@ -639,25 +611,27 @@ err_llb:
 
 static void __intel_fbc_cleanup_cfb(struct drm_i915_private *dev_priv)
 {
-	if (drm_mm_node_allocated(&dev_priv->fbc.compressed_fb))
-		i915_gem_stolen_remove_node(dev_priv,
-					    &dev_priv->fbc.compressed_fb);
-
-	if (dev_priv->fbc.compressed_llb) {
-		i915_gem_stolen_remove_node(dev_priv,
-					    dev_priv->fbc.compressed_llb);
-		kfree(dev_priv->fbc.compressed_llb);
+	struct intel_fbc *fbc = &dev_priv->fbc;
+
+	if (drm_mm_node_allocated(&fbc->compressed_fb))
+		i915_gem_stolen_remove_node(dev_priv, &fbc->compressed_fb);
+
+	if (fbc->compressed_llb) {
+		i915_gem_stolen_remove_node(dev_priv, fbc->compressed_llb);
+		kfree(fbc->compressed_llb);
 	}
 }
 
 void intel_fbc_cleanup_cfb(struct drm_i915_private *dev_priv)
 {
+	struct intel_fbc *fbc = &dev_priv->fbc;
+
 	if (!fbc_supported(dev_priv))
 		return;
 
-	mutex_lock(&dev_priv->fbc.lock);
+	mutex_lock(&fbc->lock);
 	__intel_fbc_cleanup_cfb(dev_priv);
-	mutex_unlock(&dev_priv->fbc.lock);
+	mutex_unlock(&fbc->lock);
 }
 
 static bool stride_is_valid(struct drm_i915_private *dev_priv,
@@ -681,19 +655,17 @@ static bool stride_is_valid(struct drm_i915_private *dev_priv,
 	return true;
 }
 
-static bool pixel_format_is_valid(struct drm_framebuffer *fb)
+static bool pixel_format_is_valid(struct drm_i915_private *dev_priv,
+				  uint32_t pixel_format)
 {
-	struct drm_device *dev = fb->dev;
-	struct drm_i915_private *dev_priv = dev->dev_private;
-
-	switch (fb->pixel_format) {
+	switch (pixel_format) {
 	case DRM_FORMAT_XRGB8888:
 	case DRM_FORMAT_XBGR8888:
 		return true;
 	case DRM_FORMAT_XRGB1555:
 	case DRM_FORMAT_RGB565:
 		/* 16bpp not supported on gen2 */
-		if (IS_GEN2(dev))
+		if (IS_GEN2(dev_priv))
 			return false;
 		/* WaFbcOnly1to1Ratio:ctg */
 		if (IS_G4X(dev_priv))
@@ -713,6 +685,7 @@ static bool pixel_format_is_valid(struct drm_framebuffer *fb)
 static bool intel_fbc_hw_tracking_covers_screen(struct intel_crtc *crtc)
 {
 	struct drm_i915_private *dev_priv = crtc->base.dev->dev_private;
+	struct intel_fbc *fbc = &dev_priv->fbc;
 	unsigned int effective_w, effective_h, max_w, max_h;
 
 	if (INTEL_INFO(dev_priv)->gen >= 8 || IS_HASWELL(dev_priv)) {
@@ -726,87 +699,105 @@ static bool intel_fbc_hw_tracking_covers_screen(struct intel_crtc *crtc)
 		max_h = 1536;
 	}
 
-	intel_fbc_get_plane_source_size(crtc, &effective_w, &effective_h);
+	intel_fbc_get_plane_source_size(&fbc->state_cache, &effective_w,
+					&effective_h);
 	effective_w += crtc->adjusted_x;
 	effective_h += crtc->adjusted_y;
 
 	return effective_w <= max_w && effective_h <= max_h;
 }
 
-/**
- * __intel_fbc_update - activate/deactivate FBC as needed, unlocked
- * @crtc: the CRTC that triggered the update
- *
- * This function completely reevaluates the status of FBC, then activates,
- * deactivates or maintains it on the same state.
- */
-static void __intel_fbc_update(struct intel_crtc *crtc)
+static void intel_fbc_update_state_cache(struct intel_crtc *crtc)
 {
 	struct drm_i915_private *dev_priv = crtc->base.dev->dev_private;
-	struct drm_framebuffer *fb;
+	struct intel_fbc *fbc = &dev_priv->fbc;
+	struct intel_fbc_state_cache *cache = &fbc->state_cache;
+	struct intel_crtc_state *crtc_state =
+		to_intel_crtc_state(crtc->base.state);
+	struct intel_plane_state *plane_state =
+		to_intel_plane_state(crtc->base.primary->state);
+	struct drm_framebuffer *fb = plane_state->base.fb;
 	struct drm_i915_gem_object *obj;
-	const struct drm_display_mode *adjusted_mode;
 
-	WARN_ON(!mutex_is_locked(&dev_priv->fbc.lock));
+	WARN_ON(!drm_modeset_is_locked(&crtc->base.mutex));
+	WARN_ON(!drm_modeset_is_locked(&crtc->base.primary->mutex));
 
-	if (!multiple_pipes_ok(dev_priv)) {
-		set_no_fbc_reason(dev_priv, "more than one pipe active");
-		goto out_disable;
-	}
+	cache->crtc.mode_flags = crtc_state->base.adjusted_mode.flags;
+	if (IS_HASWELL(dev_priv) || IS_BROADWELL(dev_priv))
+		cache->crtc.hsw_bdw_pixel_rate =
+			ilk_pipe_pixel_rate(crtc_state);
 
-	if (!dev_priv->fbc.enabled || dev_priv->fbc.crtc != crtc)
-		return;
+	cache->plane.rotation = plane_state->base.rotation;
+	cache->plane.src_w = drm_rect_width(&plane_state->src) >> 16;
+	cache->plane.src_h = drm_rect_height(&plane_state->src) >> 16;
+	cache->plane.visible = plane_state->visible;
 
-	if (!crtc_is_valid(crtc)) {
-		set_no_fbc_reason(dev_priv, "no output");
-		goto out_disable;
-	}
+	if (!cache->plane.visible)
+		return;
 
-	fb = crtc->base.primary->fb;
 	obj = intel_fb_obj(fb);
-	adjusted_mode = &crtc->config->base.adjusted_mode;
 
-	if ((adjusted_mode->flags & DRM_MODE_FLAG_INTERLACE) ||
-	    (adjusted_mode->flags & DRM_MODE_FLAG_DBLSCAN)) {
-		set_no_fbc_reason(dev_priv, "incompatible mode");
-		goto out_disable;
+	/* FIXME: We lack the proper locking here, so only run this on the
+	 * platforms that need. */
+	if (INTEL_INFO(dev_priv)->gen >= 5 && INTEL_INFO(dev_priv)->gen < 7)
+		cache->fb.ilk_ggtt_offset = i915_gem_obj_ggtt_offset(obj);
+	cache->fb.pixel_format = fb->pixel_format;
+	cache->fb.stride = fb->pitches[0];
+	cache->fb.fence_reg = obj->fence_reg;
+	cache->fb.tiling_mode = obj->tiling_mode;
+}
+
+static bool intel_fbc_can_activate(struct intel_crtc *crtc)
+{
+	struct drm_i915_private *dev_priv = crtc->base.dev->dev_private;
+	struct intel_fbc *fbc = &dev_priv->fbc;
+	struct intel_fbc_state_cache *cache = &fbc->state_cache;
+
+	if (!cache->plane.visible) {
+		fbc->no_fbc_reason = "primary plane not visible";
+		return false;
+	}
+
+	if ((cache->crtc.mode_flags & DRM_MODE_FLAG_INTERLACE) ||
+	    (cache->crtc.mode_flags & DRM_MODE_FLAG_DBLSCAN)) {
+		fbc->no_fbc_reason = "incompatible mode";
+		return false;
 	}
 
 	if (!intel_fbc_hw_tracking_covers_screen(crtc)) {
-		set_no_fbc_reason(dev_priv, "mode too large for compression");
-		goto out_disable;
+		fbc->no_fbc_reason = "mode too large for compression";
+		return false;
 	}
 
 	/* The use of a CPU fence is mandatory in order to detect writes
 	 * by the CPU to the scanout and trigger updates to the FBC.
 	 */
-	if (obj->tiling_mode != I915_TILING_X ||
-	    obj->fence_reg == I915_FENCE_REG_NONE) {
-		set_no_fbc_reason(dev_priv, "framebuffer not tiled or fenced");
-		goto out_disable;
+	if (cache->fb.tiling_mode != I915_TILING_X ||
+	    cache->fb.fence_reg == I915_FENCE_REG_NONE) {
+		fbc->no_fbc_reason = "framebuffer not tiled or fenced";
+		return false;
 	}
 	if (INTEL_INFO(dev_priv)->gen <= 4 && !IS_G4X(dev_priv) &&
-	    crtc->base.primary->state->rotation != BIT(DRM_ROTATE_0)) {
-		set_no_fbc_reason(dev_priv, "rotation unsupported");
-		goto out_disable;
+	    cache->plane.rotation != BIT(DRM_ROTATE_0)) {
+		fbc->no_fbc_reason = "rotation unsupported";
+		return false;
 	}
 
-	if (!stride_is_valid(dev_priv, fb->pitches[0])) {
-		set_no_fbc_reason(dev_priv, "framebuffer stride not supported");
-		goto out_disable;
+	if (!stride_is_valid(dev_priv, cache->fb.stride)) {
+		fbc->no_fbc_reason = "framebuffer stride not supported";
+		return false;
 	}
 
-	if (!pixel_format_is_valid(fb)) {
-		set_no_fbc_reason(dev_priv, "pixel format is invalid");
-		goto out_disable;
+	if (!pixel_format_is_valid(dev_priv, cache->fb.pixel_format)) {
+		fbc->no_fbc_reason = "pixel format is invalid";
+		return false;
 	}
 
 	/* WaFbcExceedCdClockThreshold:hsw,bdw */
 	if ((IS_HASWELL(dev_priv) || IS_BROADWELL(dev_priv)) &&
-	    ilk_pipe_pixel_rate(crtc->config) >=
-	    dev_priv->cdclk_freq * 95 / 100) {
-		set_no_fbc_reason(dev_priv, "pixel rate is too big");
-		goto out_disable;
+	    cache->crtc.hsw_bdw_pixel_rate >= dev_priv->cdclk_freq * 95 / 100) {
+		fbc->no_fbc_reason = "pixel rate is too big";
+		return false;
 	}
 
 	/* It is possible for the required CFB size change without a
@@ -819,189 +810,322 @@ static void __intel_fbc_update(struct intel_crtc *crtc)
 	 * we didn't get any invalidate/deactivate calls, but this would require
 	 * a lot of tracking just for a specific case. If we conclude it's an
 	 * important case, we can implement it later. */
-	if (intel_fbc_calculate_cfb_size(crtc, fb) >
-	    dev_priv->fbc.compressed_fb.size * dev_priv->fbc.threshold) {
-		set_no_fbc_reason(dev_priv, "CFB requirements changed");
-		goto out_disable;
+	if (intel_fbc_calculate_cfb_size(dev_priv, &fbc->state_cache) >
+	    fbc->compressed_fb.size * fbc->threshold) {
+		fbc->no_fbc_reason = "CFB requirements changed";
+		return false;
 	}
 
+	return true;
+}
+
+static bool intel_fbc_can_choose(struct intel_crtc *crtc)
+{
+	struct drm_i915_private *dev_priv = crtc->base.dev->dev_private;
+	struct intel_fbc *fbc = &dev_priv->fbc;
+	bool enable_by_default = IS_HASWELL(dev_priv) ||
+				 IS_BROADWELL(dev_priv);
+
+	if (intel_vgpu_active(dev_priv->dev)) {
+		fbc->no_fbc_reason = "VGPU is active";
+		return false;
+	}
+
+	if (i915.enable_fbc < 0 && !enable_by_default) {
+		fbc->no_fbc_reason = "disabled per chip default";
+		return false;
+	}
+
+	if (!i915.enable_fbc) {
+		fbc->no_fbc_reason = "disabled per module param";
+		return false;
+	}
+
+	if (fbc_on_pipe_a_only(dev_priv) && crtc->pipe != PIPE_A) {
+		fbc->no_fbc_reason = "no enabled pipes can have FBC";
+		return false;
+	}
+
+	if (fbc_on_plane_a_only(dev_priv) && crtc->plane != PLANE_A) {
+		fbc->no_fbc_reason = "no enabled planes can have FBC";
+		return false;
+	}
+
+	return true;
+}
+
+static void intel_fbc_get_reg_params(struct intel_crtc *crtc,
+				     struct intel_fbc_reg_params *params)
+{
+	struct drm_i915_private *dev_priv = crtc->base.dev->dev_private;
+	struct intel_fbc *fbc = &dev_priv->fbc;
+	struct intel_fbc_state_cache *cache = &fbc->state_cache;
+
+	/* Since all our fields are integer types, use memset here so the
+	 * comparison function can rely on memcmp because the padding will be
+	 * zero. */
+	memset(params, 0, sizeof(*params));
+
+	params->crtc.pipe = crtc->pipe;
+	params->crtc.plane = crtc->plane;
+	params->crtc.fence_y_offset = get_crtc_fence_y_offset(crtc);
+
+	params->fb.pixel_format = cache->fb.pixel_format;
+	params->fb.stride = cache->fb.stride;
+	params->fb.fence_reg = cache->fb.fence_reg;
+
+	params->cfb_size = intel_fbc_calculate_cfb_size(dev_priv, cache);
+
+	params->fb.ggtt_offset = cache->fb.ilk_ggtt_offset;
+}
+
+static bool intel_fbc_reg_params_equal(struct intel_fbc_reg_params *params1,
+				       struct intel_fbc_reg_params *params2)
+{
+	/* We can use this since intel_fbc_get_reg_params() does a memset. */
+	return memcmp(params1, params2, sizeof(*params1)) == 0;
+}
+
+void intel_fbc_pre_update(struct intel_crtc *crtc)
+{
+	struct drm_i915_private *dev_priv = crtc->base.dev->dev_private;
+	struct intel_fbc *fbc = &dev_priv->fbc;
+
+	if (!fbc_supported(dev_priv))
+		return;
+
+	mutex_lock(&fbc->lock);
+
+	if (!multiple_pipes_ok(crtc)) {
+		fbc->no_fbc_reason = "more than one pipe active";
+		goto deactivate;
+	}
+
+	if (!fbc->enabled || fbc->crtc != crtc)
+		goto unlock;
+
+	intel_fbc_update_state_cache(crtc);
+
+deactivate:
+	intel_fbc_deactivate(dev_priv);
+unlock:
+	mutex_unlock(&fbc->lock);
+}
+
+static void __intel_fbc_post_update(struct intel_crtc *crtc)
+{
+	struct drm_i915_private *dev_priv = crtc->base.dev->dev_private;
+	struct intel_fbc *fbc = &dev_priv->fbc;
+	struct intel_fbc_reg_params old_params;
+
+	WARN_ON(!mutex_is_locked(&fbc->lock));
+
+	if (!fbc->enabled || fbc->crtc != crtc)
+		return;
+
+	if (!intel_fbc_can_activate(crtc)) {
+		WARN_ON(fbc->active);
+		return;
+	}
+
+	old_params = fbc->params;
+	intel_fbc_get_reg_params(crtc, &fbc->params);
+
 	/* If the scanout has not changed, don't modify the FBC settings.
 	 * Note that we make the fundamental assumption that the fb->obj
 	 * cannot be unpinned (and have its GTT offset and fence revoked)
 	 * without first being decoupled from the scanout and FBC disabled.
 	 */
-	if (dev_priv->fbc.crtc == crtc &&
-	    dev_priv->fbc.fb_id == fb->base.id &&
-	    dev_priv->fbc.y == crtc->base.y &&
-	    dev_priv->fbc.active)
+	if (fbc->active &&
+	    intel_fbc_reg_params_equal(&old_params, &fbc->params))
 		return;
 
-	if (intel_fbc_is_active(dev_priv)) {
-		/* We update FBC along two paths, after changing fb/crtc
-		 * configuration (modeswitching) and after page-flipping
-		 * finishes. For the latter, we know that not only did
-		 * we disable the FBC at the start of the page-flip
-		 * sequence, but also more than one vblank has passed.
-		 *
-		 * For the former case of modeswitching, it is possible
-		 * to switch between two FBC valid configurations
-		 * instantaneously so we do need to disable the FBC
-		 * before we can modify its control registers. We also
-		 * have to wait for the next vblank for that to take
-		 * effect. However, since we delay enabling FBC we can
-		 * assume that a vblank has passed since disabling and
-		 * that we can safely alter the registers in the deferred
-		 * callback.
-		 *
-		 * In the scenario that we go from a valid to invalid
-		 * and then back to valid FBC configuration we have
-		 * no strict enforcement that a vblank occurred since
-		 * disabling the FBC. However, along all current pipe
-		 * disabling paths we do need to wait for a vblank at
-		 * some point. And we wait before enabling FBC anyway.
-		 */
-		DRM_DEBUG_KMS("deactivating FBC for update\n");
-		__intel_fbc_deactivate(dev_priv);
-	}
-
+	intel_fbc_deactivate(dev_priv);
 	intel_fbc_schedule_activation(crtc);
-	dev_priv->fbc.no_fbc_reason = "FBC enabled (not necessarily active)";
-	return;
-
-out_disable:
-	/* Multiple disables should be harmless */
-	if (intel_fbc_is_active(dev_priv)) {
-		DRM_DEBUG_KMS("unsupported config, deactivating FBC\n");
-		__intel_fbc_deactivate(dev_priv);
-	}
+	fbc->no_fbc_reason = "FBC enabled (active or scheduled)";
 }
 
-/*
- * intel_fbc_update - activate/deactivate FBC as needed
- * @crtc: the CRTC that triggered the update
- *
- * This function reevaluates the overall state and activates or deactivates FBC.
- */
-void intel_fbc_update(struct intel_crtc *crtc)
+void intel_fbc_post_update(struct intel_crtc *crtc)
 {
 	struct drm_i915_private *dev_priv = crtc->base.dev->dev_private;
+	struct intel_fbc *fbc = &dev_priv->fbc;
 
 	if (!fbc_supported(dev_priv))
 		return;
 
-	mutex_lock(&dev_priv->fbc.lock);
-	__intel_fbc_update(crtc);
-	mutex_unlock(&dev_priv->fbc.lock);
+	mutex_lock(&fbc->lock);
+	__intel_fbc_post_update(crtc);
+	mutex_unlock(&fbc->lock);
+}
+
+static unsigned int intel_fbc_get_frontbuffer_bit(struct intel_fbc *fbc)
+{
+	if (fbc->enabled)
+		return to_intel_plane(fbc->crtc->base.primary)->frontbuffer_bit;
+	else
+		return fbc->possible_framebuffer_bits;
 }
 
 void intel_fbc_invalidate(struct drm_i915_private *dev_priv,
 			  unsigned int frontbuffer_bits,
 			  enum fb_op_origin origin)
 {
-	unsigned int fbc_bits;
+	struct intel_fbc *fbc = &dev_priv->fbc;
 
 	if (!fbc_supported(dev_priv))
 		return;
 
-	if (origin == ORIGIN_GTT)
+	if (origin == ORIGIN_GTT || origin == ORIGIN_FLIP)
 		return;
 
-	mutex_lock(&dev_priv->fbc.lock);
+	mutex_lock(&fbc->lock);
 
-	if (dev_priv->fbc.enabled)
-		fbc_bits = INTEL_FRONTBUFFER_PRIMARY(dev_priv->fbc.crtc->pipe);
-	else
-		fbc_bits = dev_priv->fbc.possible_framebuffer_bits;
-
-	dev_priv->fbc.busy_bits |= (fbc_bits & frontbuffer_bits);
+	fbc->busy_bits |= intel_fbc_get_frontbuffer_bit(fbc) & frontbuffer_bits;
 
-	if (dev_priv->fbc.busy_bits)
-		__intel_fbc_deactivate(dev_priv);
+	if (fbc->enabled && fbc->busy_bits)
+		intel_fbc_deactivate(dev_priv);
 
-	mutex_unlock(&dev_priv->fbc.lock);
+	mutex_unlock(&fbc->lock);
 }
 
 void intel_fbc_flush(struct drm_i915_private *dev_priv,
 		     unsigned int frontbuffer_bits, enum fb_op_origin origin)
 {
+	struct intel_fbc *fbc = &dev_priv->fbc;
+
 	if (!fbc_supported(dev_priv))
 		return;
 
-	if (origin == ORIGIN_GTT)
+	if (origin == ORIGIN_GTT || origin == ORIGIN_FLIP)
 		return;
 
-	mutex_lock(&dev_priv->fbc.lock);
+	mutex_lock(&fbc->lock);
 
-	dev_priv->fbc.busy_bits &= ~frontbuffer_bits;
+	fbc->busy_bits &= ~frontbuffer_bits;
 
-	if (!dev_priv->fbc.busy_bits && dev_priv->fbc.enabled) {
-		if (origin != ORIGIN_FLIP && dev_priv->fbc.active) {
+	if (!fbc->busy_bits && fbc->enabled &&
+	    (frontbuffer_bits & intel_fbc_get_frontbuffer_bit(fbc))) {
+		if (fbc->active)
 			intel_fbc_recompress(dev_priv);
-		} else {
-			__intel_fbc_deactivate(dev_priv);
-			__intel_fbc_update(dev_priv->fbc.crtc);
+		else
+			__intel_fbc_post_update(fbc->crtc);
+	}
+
+	mutex_unlock(&fbc->lock);
+}
+
+/**
+ * intel_fbc_choose_crtc - select a CRTC to enable FBC on
+ * @dev_priv: i915 device instance
+ * @state: the atomic state structure
+ *
+ * This function looks at the proposed state for CRTCs and planes, then chooses
+ * which pipe is going to have FBC by setting intel_crtc_state->enable_fbc to
+ * true.
+ *
+ * Later, intel_fbc_enable is going to look for state->enable_fbc and then maybe
+ * enable FBC for the chosen CRTC. If it does, it will set dev_priv->fbc.crtc.
+ */
+void intel_fbc_choose_crtc(struct drm_i915_private *dev_priv,
+			   struct drm_atomic_state *state)
+{
+	struct intel_fbc *fbc = &dev_priv->fbc;
+	struct drm_crtc *crtc;
+	struct drm_crtc_state *crtc_state;
+	struct drm_plane *plane;
+	struct drm_plane_state *plane_state;
+	bool fbc_crtc_present = false;
+	int i, j;
+
+	mutex_lock(&fbc->lock);
+
+	for_each_crtc_in_state(state, crtc, crtc_state, i) {
+		if (fbc->crtc == to_intel_crtc(crtc)) {
+			fbc_crtc_present = true;
+			break;
+		}
+	}
+	/* This atomic commit doesn't involve the CRTC currently tied to FBC. */
+	if (!fbc_crtc_present && fbc->crtc != NULL)
+		goto out;
+
+	/* Simply choose the first CRTC that is compatible and has a visible
+	 * plane. We could go for fancier schemes such as checking the plane
+	 * size, but this would just affect the few platforms that don't tie FBC
+	 * to pipe or plane A. */
+	for_each_plane_in_state(state, plane, plane_state, i) {
+		struct intel_plane_state *intel_plane_state =
+			to_intel_plane_state(plane_state);
+
+		if (!intel_plane_state->visible)
+			continue;
+
+		for_each_crtc_in_state(state, crtc, crtc_state, j) {
+			struct intel_crtc_state *intel_crtc_state =
+				to_intel_crtc_state(crtc_state);
+
+			if (plane_state->crtc != crtc)
+				continue;
+
+			if (!intel_fbc_can_choose(to_intel_crtc(crtc)))
+				break;
+
+			intel_crtc_state->enable_fbc = true;
+			goto out;
 		}
 	}
 
-	mutex_unlock(&dev_priv->fbc.lock);
+out:
+	mutex_unlock(&fbc->lock);
 }
 
 /**
  * intel_fbc_enable: tries to enable FBC on the CRTC
  * @crtc: the CRTC
  *
- * This function checks if it's possible to enable FBC on the following CRTC,
- * then enables it. Notice that it doesn't activate FBC.
+ * This function checks if the given CRTC was chosen for FBC, then enables it if
+ * possible. Notice that it doesn't activate FBC. It is valid to call
+ * intel_fbc_enable multiple times for the same pipe without an
+ * intel_fbc_disable in the middle, as long as it is deactivated.
  */
 void intel_fbc_enable(struct intel_crtc *crtc)
 {
 	struct drm_i915_private *dev_priv = crtc->base.dev->dev_private;
+	struct intel_fbc *fbc = &dev_priv->fbc;
 
 	if (!fbc_supported(dev_priv))
 		return;
 
-	mutex_lock(&dev_priv->fbc.lock);
-
-	if (dev_priv->fbc.enabled) {
-		WARN_ON(dev_priv->fbc.crtc == crtc);
-		goto out;
-	}
-
-	WARN_ON(dev_priv->fbc.active);
-	WARN_ON(dev_priv->fbc.crtc != NULL);
+	mutex_lock(&fbc->lock);
 
-	if (intel_vgpu_active(dev_priv->dev)) {
-		set_no_fbc_reason(dev_priv, "VGPU is active");
-		goto out;
-	}
-
-	if (i915.enable_fbc < 0) {
-		set_no_fbc_reason(dev_priv, "disabled per chip default");
+	if (fbc->enabled) {
+		WARN_ON(fbc->crtc == NULL);
+		if (fbc->crtc == crtc) {
+			WARN_ON(!crtc->config->enable_fbc);
+			WARN_ON(fbc->active);
+		}
 		goto out;
 	}
 
-	if (!i915.enable_fbc) {
-		set_no_fbc_reason(dev_priv, "disabled per module param");
+	if (!crtc->config->enable_fbc)
 		goto out;
-	}
 
-	if (!crtc_can_fbc(crtc)) {
-		set_no_fbc_reason(dev_priv, "no enabled pipes can have FBC");
-		goto out;
-	}
+	WARN_ON(fbc->active);
+	WARN_ON(fbc->crtc != NULL);
 
+	intel_fbc_update_state_cache(crtc);
 	if (intel_fbc_alloc_cfb(crtc)) {
-		set_no_fbc_reason(dev_priv, "not enough stolen memory");
+		fbc->no_fbc_reason = "not enough stolen memory";
 		goto out;
 	}
 
 	DRM_DEBUG_KMS("Enabling FBC on pipe %c\n", pipe_name(crtc->pipe));
-	dev_priv->fbc.no_fbc_reason = "FBC enabled but not active yet\n";
+	fbc->no_fbc_reason = "FBC enabled but not active yet\n";
 
-	dev_priv->fbc.enabled = true;
-	dev_priv->fbc.crtc = crtc;
+	fbc->enabled = true;
+	fbc->crtc = crtc;
 out:
-	mutex_unlock(&dev_priv->fbc.lock);
+	mutex_unlock(&fbc->lock);
 }
 
 /**
@@ -1013,58 +1137,88 @@ out:
  */
 static void __intel_fbc_disable(struct drm_i915_private *dev_priv)
 {
-	struct intel_crtc *crtc = dev_priv->fbc.crtc;
+	struct intel_fbc *fbc = &dev_priv->fbc;
+	struct intel_crtc *crtc = fbc->crtc;
 
-	WARN_ON(!mutex_is_locked(&dev_priv->fbc.lock));
-	WARN_ON(!dev_priv->fbc.enabled);
-	WARN_ON(dev_priv->fbc.active);
-	assert_pipe_disabled(dev_priv, crtc->pipe);
+	WARN_ON(!mutex_is_locked(&fbc->lock));
+	WARN_ON(!fbc->enabled);
+	WARN_ON(fbc->active);
+	WARN_ON(crtc->active);
 
 	DRM_DEBUG_KMS("Disabling FBC on pipe %c\n", pipe_name(crtc->pipe));
 
 	__intel_fbc_cleanup_cfb(dev_priv);
 
-	dev_priv->fbc.enabled = false;
-	dev_priv->fbc.crtc = NULL;
+	fbc->enabled = false;
+	fbc->crtc = NULL;
 }
 
 /**
- * intel_fbc_disable_crtc - disable FBC if it's associated with crtc
+ * intel_fbc_disable - disable FBC if it's associated with crtc
  * @crtc: the CRTC
  *
  * This function disables FBC if it's associated with the provided CRTC.
  */
-void intel_fbc_disable_crtc(struct intel_crtc *crtc)
+void intel_fbc_disable(struct intel_crtc *crtc)
 {
 	struct drm_i915_private *dev_priv = crtc->base.dev->dev_private;
+	struct intel_fbc *fbc = &dev_priv->fbc;
 
 	if (!fbc_supported(dev_priv))
 		return;
 
-	mutex_lock(&dev_priv->fbc.lock);
-	if (dev_priv->fbc.crtc == crtc) {
-		WARN_ON(!dev_priv->fbc.enabled);
-		WARN_ON(dev_priv->fbc.active);
+	mutex_lock(&fbc->lock);
+	if (fbc->crtc == crtc) {
+		WARN_ON(!fbc->enabled);
+		WARN_ON(fbc->active);
 		__intel_fbc_disable(dev_priv);
 	}
-	mutex_unlock(&dev_priv->fbc.lock);
+	mutex_unlock(&fbc->lock);
+
+	cancel_work_sync(&fbc->work.work);
 }
 
 /**
- * intel_fbc_disable - globally disable FBC
+ * intel_fbc_global_disable - globally disable FBC
  * @dev_priv: i915 device instance
  *
  * This function disables FBC regardless of which CRTC is associated with it.
  */
-void intel_fbc_disable(struct drm_i915_private *dev_priv)
+void intel_fbc_global_disable(struct drm_i915_private *dev_priv)
 {
+	struct intel_fbc *fbc = &dev_priv->fbc;
+
 	if (!fbc_supported(dev_priv))
 		return;
 
-	mutex_lock(&dev_priv->fbc.lock);
-	if (dev_priv->fbc.enabled)
+	mutex_lock(&fbc->lock);
+	if (fbc->enabled)
 		__intel_fbc_disable(dev_priv);
-	mutex_unlock(&dev_priv->fbc.lock);
+	mutex_unlock(&fbc->lock);
+
+	cancel_work_sync(&fbc->work.work);
+}
+
+/**
+ * intel_fbc_init_pipe_state - initialize FBC's CRTC visibility tracking
+ * @dev_priv: i915 device instance
+ *
+ * The FBC code needs to track CRTC visibility since the older platforms can't
+ * have FBC enabled while multiple pipes are used. This function does the
+ * initial setup at driver load to make sure FBC is matching the real hardware.
+ */
+void intel_fbc_init_pipe_state(struct drm_i915_private *dev_priv)
+{
+	struct intel_crtc *crtc;
+
+	/* Don't even bother tracking anything if we don't need. */
+	if (!no_fbc_on_multiple_pipes(dev_priv))
+		return;
+
+	for_each_intel_crtc(dev_priv->dev, crtc)
+		if (intel_crtc_active(&crtc->base) &&
+		    to_intel_plane_state(crtc->base.primary->state)->visible)
+			dev_priv->fbc.visible_pipes_mask |= (1 << crtc->pipe);
 }
 
 /**
@@ -1075,51 +1229,35 @@ void intel_fbc_disable(struct drm_i915_private *dev_priv)
  */
 void intel_fbc_init(struct drm_i915_private *dev_priv)
 {
+	struct intel_fbc *fbc = &dev_priv->fbc;
 	enum pipe pipe;
 
-	INIT_WORK(&dev_priv->fbc.work.work, intel_fbc_work_fn);
-	mutex_init(&dev_priv->fbc.lock);
-	dev_priv->fbc.enabled = false;
-	dev_priv->fbc.active = false;
-	dev_priv->fbc.work.scheduled = false;
+	INIT_WORK(&fbc->work.work, intel_fbc_work_fn);
+	mutex_init(&fbc->lock);
+	fbc->enabled = false;
+	fbc->active = false;
+	fbc->work.scheduled = false;
 
 	if (!HAS_FBC(dev_priv)) {
-		dev_priv->fbc.no_fbc_reason = "unsupported by this chipset";
+		fbc->no_fbc_reason = "unsupported by this chipset";
 		return;
 	}
 
 	for_each_pipe(dev_priv, pipe) {
-		dev_priv->fbc.possible_framebuffer_bits |=
+		fbc->possible_framebuffer_bits |=
 				INTEL_FRONTBUFFER_PRIMARY(pipe);
 
 		if (fbc_on_pipe_a_only(dev_priv))
 			break;
 	}
 
-	if (INTEL_INFO(dev_priv)->gen >= 7) {
-		dev_priv->fbc.is_active = ilk_fbc_is_active;
-		dev_priv->fbc.activate = gen7_fbc_activate;
-		dev_priv->fbc.deactivate = ilk_fbc_deactivate;
-	} else if (INTEL_INFO(dev_priv)->gen >= 5) {
-		dev_priv->fbc.is_active = ilk_fbc_is_active;
-		dev_priv->fbc.activate = ilk_fbc_activate;
-		dev_priv->fbc.deactivate = ilk_fbc_deactivate;
-	} else if (IS_GM45(dev_priv)) {
-		dev_priv->fbc.is_active = g4x_fbc_is_active;
-		dev_priv->fbc.activate = g4x_fbc_activate;
-		dev_priv->fbc.deactivate = g4x_fbc_deactivate;
-	} else {
-		dev_priv->fbc.is_active = i8xx_fbc_is_active;
-		dev_priv->fbc.activate = i8xx_fbc_activate;
-		dev_priv->fbc.deactivate = i8xx_fbc_deactivate;
-
-		/* This value was pulled out of someone's hat */
+	/* This value was pulled out of someone's hat */
+	if (INTEL_INFO(dev_priv)->gen <= 4 && !IS_GM45(dev_priv))
 		I915_WRITE(FBC_CONTROL, 500 << FBC_CTL_INTERVAL_SHIFT);
-	}
 
 	/* We still don't have any sort of hardware state readout for FBC, so
 	 * deactivate it in case the BIOS activated it to make sure software
 	 * matches the hardware state. */
-	if (dev_priv->fbc.is_active(dev_priv))
-		dev_priv->fbc.deactivate(dev_priv);
+	if (intel_fbc_hw_is_active(dev_priv))
+		intel_fbc_hw_deactivate(dev_priv);
 }
diff --git a/drivers/gpu/drm/i915/intel_fbdev.c b/drivers/gpu/drm/i915/intel_fbdev.c
index 09840f4380f9..97a91e631915 100644
--- a/drivers/gpu/drm/i915/intel_fbdev.c
+++ b/drivers/gpu/drm/i915/intel_fbdev.c
@@ -406,8 +406,8 @@ retry:
 			continue;
 		}
 
-		encoder = connector->encoder;
-		if (!encoder || WARN_ON(!encoder->crtc)) {
+		encoder = connector->state->best_encoder;
+		if (!encoder || WARN_ON(!connector->state->crtc)) {
 			if (connector->force > DRM_FORCE_OFF)
 				goto bail;
 
@@ -420,7 +420,7 @@ retry:
 
 		num_connectors_enabled++;
 
-		new_crtc = intel_fb_helper_crtc(fb_helper, encoder->crtc);
+		new_crtc = intel_fb_helper_crtc(fb_helper, connector->state->crtc);
 
 		/*
 		 * Make sure we're not trying to drive multiple connectors
@@ -466,17 +466,22 @@ retry:
 			 * usually contains. But since our current
 			 * code puts a mode derived from the post-pfit timings
 			 * into crtc->mode this works out correctly.
+			 *
+			 * This is crtc->mode and not crtc->state->mode for the
+			 * fastboot check to work correctly. crtc_state->mode has
+			 * I915_MODE_FLAG_INHERITED, which we clear to force check
+			 * state.
 			 */
 			DRM_DEBUG_KMS("looking for current mode on connector %s\n",
 				      connector->name);
-			modes[i] = &encoder->crtc->mode;
+			modes[i] = &connector->state->crtc->mode;
 		}
 		crtcs[i] = new_crtc;
 
 		DRM_DEBUG_KMS("connector %s on pipe %c [CRTC:%d]: %dx%d%s\n",
 			      connector->name,
-			      pipe_name(to_intel_crtc(encoder->crtc)->pipe),
-			      encoder->crtc->base.id,
+			      pipe_name(to_intel_crtc(connector->state->crtc)->pipe),
+			      connector->state->crtc->base.id,
 			      modes[i]->hdisplay, modes[i]->vdisplay,
 			      modes[i]->flags & DRM_MODE_FLAG_INTERLACE ? "i" :"");
 
diff --git a/drivers/gpu/drm/i915/intel_guc.h b/drivers/gpu/drm/i915/intel_guc.h
index 045b1491ff7a..73002e901ff2 100644
--- a/drivers/gpu/drm/i915/intel_guc.h
+++ b/drivers/gpu/drm/i915/intel_guc.h
@@ -46,7 +46,7 @@ struct i915_guc_client {
 	uint32_t wq_head;
 
 	/* GuC submission statistics & status */
-	uint64_t submissions[I915_NUM_RINGS];
+	uint64_t submissions[GUC_MAX_ENGINES_NUM];
 	uint32_t q_fail;
 	uint32_t b_fail;
 	int retcode;
@@ -106,8 +106,8 @@ struct intel_guc {
 	uint32_t action_fail;		/* Total number of failures	*/
 	int32_t action_err;		/* Last error code		*/
 
-	uint64_t submissions[I915_NUM_RINGS];
-	uint32_t last_seqno[I915_NUM_RINGS];
+	uint64_t submissions[GUC_MAX_ENGINES_NUM];
+	uint32_t last_seqno[GUC_MAX_ENGINES_NUM];
 };
 
 /* intel_guc_loader.c */
diff --git a/drivers/gpu/drm/i915/intel_guc_fwif.h b/drivers/gpu/drm/i915/intel_guc_fwif.h
index 1856a4740b83..2de57ffe5e18 100644
--- a/drivers/gpu/drm/i915/intel_guc_fwif.h
+++ b/drivers/gpu/drm/i915/intel_guc_fwif.h
@@ -44,6 +44,13 @@
 #define GUC_MAX_GPU_CONTEXTS		1024
 #define	GUC_INVALID_CTX_ID		GUC_MAX_GPU_CONTEXTS
 
+#define GUC_RENDER_ENGINE		0
+#define GUC_VIDEO_ENGINE		1
+#define GUC_BLITTER_ENGINE		2
+#define GUC_VIDEOENHANCE_ENGINE		3
+#define GUC_VIDEO_ENGINE2		4
+#define GUC_MAX_ENGINES_NUM		(GUC_VIDEO_ENGINE2 + 1)
+
 /* Work queue item header definitions */
 #define WQ_STATUS_ACTIVE		1
 #define WQ_STATUS_SUSPENDED		2
@@ -285,7 +292,7 @@ struct guc_context_desc {
 	u64 db_trigger_phy;
 	u16 db_id;
 
-	struct guc_execlist_context lrc[I915_NUM_RINGS];
+	struct guc_execlist_context lrc[GUC_MAX_ENGINES_NUM];
 
 	u8 attribute;
 
@@ -344,7 +351,7 @@ struct guc_policy {
 } __packed;
 
 struct guc_policies {
-	struct guc_policy policy[GUC_CTX_PRIORITY_NUM][I915_NUM_RINGS];
+	struct guc_policy policy[GUC_CTX_PRIORITY_NUM][GUC_MAX_ENGINES_NUM];
 
 	/* In micro seconds. How much time to allow before DPC processing is
 	 * called back via interrupt (to prevent DPC queue drain starving).
@@ -388,14 +395,14 @@ struct guc_mmio_regset {
 
 struct guc_mmio_reg_state {
 	struct guc_mmio_regset global_reg;
-	struct guc_mmio_regset engine_reg[I915_NUM_RINGS];
+	struct guc_mmio_regset engine_reg[GUC_MAX_ENGINES_NUM];
 
 	/* MMIO registers that are set as non privileged */
 	struct __packed {
 		u32 mmio_start;
 		u32 offsets[GUC_MMIO_WHITE_LIST_MAX];
 		u32 count;
-	} mmio_white_list[I915_NUM_RINGS];
+	} mmio_white_list[GUC_MAX_ENGINES_NUM];
 } __packed;
 
 /* GuC Additional Data Struct */
@@ -406,7 +413,7 @@ struct guc_ads {
 	u32 golden_context_lrca;
 	u32 scheduler_policies;
 	u32 reserved0[3];
-	u32 eng_state_size[I915_NUM_RINGS];
+	u32 eng_state_size[GUC_MAX_ENGINES_NUM];
 	u32 reserved2[4];
 } __packed;
 
diff --git a/drivers/gpu/drm/i915/intel_guc_loader.c b/drivers/gpu/drm/i915/intel_guc_loader.c
index 3accd914490f..82a3c03fbc0e 100644
--- a/drivers/gpu/drm/i915/intel_guc_loader.c
+++ b/drivers/gpu/drm/i915/intel_guc_loader.c
@@ -199,7 +199,7 @@ static void set_guc_init_params(struct drm_i915_private *dev_priv)
  * the value matches either of two values representing completion
  * of the GuC boot process.
  *
- * This is used for polling the GuC status in a wait_for_atomic()
+ * This is used for polling the GuC status in a wait_for()
  * loop below.
  */
 static inline bool guc_ucode_response(struct drm_i915_private *dev_priv,
@@ -259,14 +259,14 @@ static int guc_ucode_xfer_dma(struct drm_i915_private *dev_priv)
 	I915_WRITE(DMA_CTRL, _MASKED_BIT_ENABLE(UOS_MOVE | START_DMA));
 
 	/*
-	 * Spin-wait for the DMA to complete & the GuC to start up.
+	 * Wait for the DMA to complete & the GuC to start up.
 	 * NB: Docs recommend not using the interrupt for completion.
 	 * Measurements indicate this should take no more than 20ms, so a
 	 * timeout here indicates that the GuC has failed and is unusable.
 	 * (Higher levels of the driver will attempt to fall back to
 	 * execlist mode if this happens.)
 	 */
-	ret = wait_for_atomic(guc_ucode_response(dev_priv, &status), 100);
+	ret = wait_for(guc_ucode_response(dev_priv, &status), 100);
 
 	DRM_DEBUG_DRIVER("DMA status 0x%x, GuC status 0x%x\n",
 			I915_READ(DMA_CTRL), status);
diff --git a/drivers/gpu/drm/i915/intel_hdmi.c b/drivers/gpu/drm/i915/intel_hdmi.c
index 8698a643d027..a0d8daed2470 100755..100644
--- a/drivers/gpu/drm/i915/intel_hdmi.c
+++ b/drivers/gpu/drm/i915/intel_hdmi.c
@@ -880,15 +880,18 @@ static bool intel_hdmi_get_hw_state(struct intel_encoder *encoder,
 	struct intel_hdmi *intel_hdmi = enc_to_intel_hdmi(&encoder->base);
 	enum intel_display_power_domain power_domain;
 	u32 tmp;
+	bool ret;
 
 	power_domain = intel_display_port_power_domain(encoder);
-	if (!intel_display_power_is_enabled(dev_priv, power_domain))
+	if (!intel_display_power_get_if_enabled(dev_priv, power_domain))
 		return false;
 
+	ret = false;
+
 	tmp = I915_READ(intel_hdmi->hdmi_reg);
 
 	if (!(tmp & SDVO_ENABLE))
-		return false;
+		goto out;
 
 	if (HAS_PCH_CPT(dev))
 		*pipe = PORT_TO_PIPE_CPT(tmp);
@@ -897,7 +900,12 @@ static bool intel_hdmi_get_hw_state(struct intel_encoder *encoder,
 	else
 		*pipe = PORT_TO_PIPE(tmp);
 
-	return true;
+	ret = true;
+
+out:
+	intel_display_power_put(dev_priv, power_domain);
+
+	return ret;
 }
 
 static void intel_hdmi_get_config(struct intel_encoder *encoder,
@@ -1202,11 +1210,19 @@ intel_hdmi_mode_valid(struct drm_connector *connector,
 	struct drm_device *dev = intel_hdmi_to_dev(hdmi);
 	enum drm_mode_status status;
 	int clock;
+	int max_dotclk = to_i915(connector->dev)->max_dotclk_freq;
 
 	if (mode->flags & DRM_MODE_FLAG_DBLSCAN)
 		return MODE_NO_DBLESCAN;
 
 	clock = mode->clock;
+
+	if ((mode->flags & DRM_MODE_FLAG_3D_MASK) == DRM_MODE_FLAG_3D_FRAME_PACKING)
+		clock *= 2;
+
+	if (clock > max_dotclk)
+		return MODE_CLOCK_HIGH;
+
 	if (mode->flags & DRM_MODE_FLAG_DBLCLK)
 		clock *= 2;
 
@@ -2151,7 +2167,6 @@ void intel_hdmi_init_connector(struct intel_digital_port *intel_dig_port,
 void intel_hdmi_init(struct drm_device *dev,
 		     i915_reg_t hdmi_reg, enum port port)
 {
-	struct drm_i915_private *dev_priv = dev->dev_private;
 	struct intel_digital_port *intel_dig_port;
 	struct intel_encoder *intel_encoder;
 	struct intel_connector *intel_connector;
@@ -2220,7 +2235,6 @@ void intel_hdmi_init(struct drm_device *dev,
 		intel_encoder->cloneable |= 1 << INTEL_OUTPUT_HDMI;
 
 	intel_dig_port->port = port;
-	dev_priv->dig_port_map[port] = intel_encoder;
 	intel_dig_port->hdmi.hdmi_reg = hdmi_reg;
 	intel_dig_port->dp.output_reg = INVALID_MMIO_REG;
 	intel_dig_port->max_lanes = 4;
diff --git a/drivers/gpu/drm/i915/intel_i2c.c b/drivers/gpu/drm/i915/intel_i2c.c
index 25254b5c1ac5..52fbe530fc9e 100644
--- a/drivers/gpu/drm/i915/intel_i2c.c
+++ b/drivers/gpu/drm/i915/intel_i2c.c
@@ -664,6 +664,12 @@ int intel_setup_gmbus(struct drm_device *dev)
 
 		bus->adapter.algo = &gmbus_algorithm;
 
+		/*
+		 * We wish to retry with bit banging
+		 * after a timed out GMBUS attempt.
+		 */
+		bus->adapter.retries = 1;
+
 		/* By default use a conservative clock rate */
 		bus->reg0 = pin | GMBUS_RATE_100KHZ;
 
@@ -683,7 +689,7 @@ int intel_setup_gmbus(struct drm_device *dev)
 	return 0;
 
 err:
-	while (--pin) {
+	while (pin--) {
 		if (!intel_gmbus_is_valid_pin(dev_priv, pin))
 			continue;
 
diff --git a/drivers/gpu/drm/i915/intel_lrc.c b/drivers/gpu/drm/i915/intel_lrc.c
index 73d4347429df..6a978ce80244 100644
--- a/drivers/gpu/drm/i915/intel_lrc.c
+++ b/drivers/gpu/drm/i915/intel_lrc.c
@@ -223,9 +223,11 @@ enum {
 	FAULT_AND_CONTINUE /* Unsupported */
 };
 #define GEN8_CTX_ID_SHIFT 32
-#define CTX_RCS_INDIRECT_CTX_OFFSET_DEFAULT  0x17
+#define GEN8_CTX_RCS_INDIRECT_CTX_OFFSET_DEFAULT	0x17
+#define GEN9_CTX_RCS_INDIRECT_CTX_OFFSET_DEFAULT	0x26
 
-static int intel_lr_context_pin(struct drm_i915_gem_request *rq);
+static int intel_lr_context_pin(struct intel_context *ctx,
+				struct intel_engine_cs *engine);
 static void lrc_setup_hardware_status_page(struct intel_engine_cs *ring,
 		struct drm_i915_gem_object *default_ctx_obj);
 
@@ -393,7 +395,6 @@ static int execlists_update_context(struct drm_i915_gem_request *rq)
 	uint32_t *reg_state = rq->ctx->engine[ring->id].lrc_reg_state;
 
 	reg_state[CTX_RING_TAIL+1] = rq->tail;
-	reg_state[CTX_RING_BUFFER_START+1] = rq->ringbuf->vma->node.start;
 
 	if (ppgtt && !USES_FULL_48BIT_PPGTT(ppgtt->base.dev)) {
 		/* True 32b PPGTT with dynamic page allocation: update PDP
@@ -599,7 +600,7 @@ static int execlists_context_queue(struct drm_i915_gem_request *request)
 	int num_elements = 0;
 
 	if (request->ctx != request->i915->kernel_context)
-		intel_lr_context_pin(request);
+		intel_lr_context_pin(request->ctx, ring);
 
 	i915_gem_request_reference(request);
 
@@ -704,7 +705,7 @@ int intel_logical_ring_alloc_request_extras(struct drm_i915_gem_request *request
 	}
 
 	if (request->ctx != request->i915->kernel_context)
-		ret = intel_lr_context_pin(request);
+		ret = intel_lr_context_pin(request->ctx, request->ring);
 
 	return ret;
 }
@@ -765,6 +766,7 @@ intel_logical_ring_advance_and_submit(struct drm_i915_gem_request *request)
 {
 	struct intel_ringbuffer *ringbuf = request->ringbuf;
 	struct drm_i915_private *dev_priv = request->i915;
+	struct intel_engine_cs *engine = request->ring;
 
 	intel_logical_ring_advance(ringbuf);
 	request->tail = ringbuf->tail;
@@ -779,9 +781,20 @@ intel_logical_ring_advance_and_submit(struct drm_i915_gem_request *request)
 	intel_logical_ring_emit(ringbuf, MI_NOOP);
 	intel_logical_ring_advance(ringbuf);
 
-	if (intel_ring_stopped(request->ring))
+	if (intel_ring_stopped(engine))
 		return 0;
 
+	if (engine->last_context != request->ctx) {
+		if (engine->last_context)
+			intel_lr_context_unpin(engine->last_context, engine);
+		if (request->ctx != request->i915->kernel_context) {
+			intel_lr_context_pin(request->ctx, engine);
+			engine->last_context = request->ctx;
+		} else {
+			engine->last_context = NULL;
+		}
+	}
+
 	if (dev_priv->guc.execbuf_client)
 		i915_guc_submit(dev_priv->guc.execbuf_client, request);
 	else
@@ -1015,7 +1028,8 @@ void intel_execlists_retire_requests(struct intel_engine_cs *ring)
 				ctx->engine[ring->id].state;
 
 		if (ctx_obj && (ctx != req->i915->kernel_context))
-			intel_lr_context_unpin(req);
+			intel_lr_context_unpin(ctx, ring);
+
 		list_del(&req->execlist_link);
 		i915_gem_request_unreference(req);
 	}
@@ -1059,14 +1073,15 @@ int logical_ring_flush_all_caches(struct drm_i915_gem_request *req)
 	return 0;
 }
 
-static int intel_lr_context_do_pin(struct intel_engine_cs *ring,
-				   struct intel_context *ctx)
+static int intel_lr_context_do_pin(struct intel_context *ctx,
+				   struct intel_engine_cs *ring)
 {
 	struct drm_device *dev = ring->dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	struct drm_i915_gem_object *ctx_obj = ctx->engine[ring->id].state;
 	struct intel_ringbuffer *ringbuf = ctx->engine[ring->id].ringbuf;
 	struct page *lrc_state_page;
+	uint32_t *lrc_reg_state;
 	int ret;
 
 	WARN_ON(!mutex_is_locked(&ring->dev->struct_mutex));
@@ -1088,7 +1103,9 @@ static int intel_lr_context_do_pin(struct intel_engine_cs *ring,
 
 	ctx->engine[ring->id].lrc_vma = i915_gem_obj_to_ggtt(ctx_obj);
 	intel_lr_context_descriptor_update(ctx, ring);
-	ctx->engine[ring->id].lrc_reg_state = kmap(lrc_state_page);
+	lrc_reg_state = kmap(lrc_state_page);
+	lrc_reg_state[CTX_RING_BUFFER_START+1] = ringbuf->vma->node.start;
+	ctx->engine[ring->id].lrc_reg_state = lrc_reg_state;
 	ctx_obj->dirty = true;
 
 	/* Invalidate GuC TLB. */
@@ -1103,41 +1120,40 @@ unpin_ctx_obj:
 	return ret;
 }
 
-static int intel_lr_context_pin(struct drm_i915_gem_request *rq)
+static int intel_lr_context_pin(struct intel_context *ctx,
+				struct intel_engine_cs *engine)
 {
 	int ret = 0;
-	struct intel_engine_cs *ring = rq->ring;
 
-	if (rq->ctx->engine[ring->id].pin_count++ == 0) {
-		ret = intel_lr_context_do_pin(ring, rq->ctx);
+	if (ctx->engine[engine->id].pin_count++ == 0) {
+		ret = intel_lr_context_do_pin(ctx, engine);
 		if (ret)
 			goto reset_pin_count;
+
+		i915_gem_context_reference(ctx);
 	}
 	return ret;
 
 reset_pin_count:
-	rq->ctx->engine[ring->id].pin_count = 0;
+	ctx->engine[engine->id].pin_count = 0;
 	return ret;
 }
 
-void intel_lr_context_unpin(struct drm_i915_gem_request *rq)
+void intel_lr_context_unpin(struct intel_context *ctx,
+			    struct intel_engine_cs *engine)
 {
-	struct intel_engine_cs *ring = rq->ring;
-	struct drm_i915_gem_object *ctx_obj = rq->ctx->engine[ring->id].state;
-	struct intel_ringbuffer *ringbuf = rq->ringbuf;
+	struct drm_i915_gem_object *ctx_obj = ctx->engine[engine->id].state;
 
-	WARN_ON(!mutex_is_locked(&ring->dev->struct_mutex));
-
-	if (!ctx_obj)
-		return;
-
-	if (--rq->ctx->engine[ring->id].pin_count == 0) {
-		kunmap(kmap_to_page(rq->ctx->engine[ring->id].lrc_reg_state));
-		intel_unpin_ringbuffer_obj(ringbuf);
+	WARN_ON(!mutex_is_locked(&ctx->i915->dev->struct_mutex));
+	if (--ctx->engine[engine->id].pin_count == 0) {
+		kunmap(kmap_to_page(ctx->engine[engine->id].lrc_reg_state));
+		intel_unpin_ringbuffer_obj(ctx->engine[engine->id].ringbuf);
 		i915_gem_object_ggtt_unpin(ctx_obj);
-		rq->ctx->engine[ring->id].lrc_vma = NULL;
-		rq->ctx->engine[ring->id].lrc_desc = 0;
-		rq->ctx->engine[ring->id].lrc_reg_state = NULL;
+		ctx->engine[engine->id].lrc_vma = NULL;
+		ctx->engine[engine->id].lrc_desc = 0;
+		ctx->engine[engine->id].lrc_reg_state = NULL;
+
+		i915_gem_context_unreference(ctx);
 	}
 }
 
@@ -2062,7 +2078,7 @@ logical_ring_init(struct drm_device *dev, struct intel_engine_cs *ring)
 		goto error;
 
 	/* As this is the default context, always pin it */
-	ret = intel_lr_context_do_pin(ring, dctx);
+	ret = intel_lr_context_do_pin(dctx, ring);
 	if (ret) {
 		DRM_ERROR(
 			"Failed to pin and map ringbuffer %s: %d\n",
@@ -2086,6 +2102,7 @@ static int logical_render_ring_init(struct drm_device *dev)
 	ring->name = "render ring";
 	ring->id = RCS;
 	ring->exec_id = I915_EXEC_RENDER;
+	ring->guc_id = GUC_RENDER_ENGINE;
 	ring->mmio_base = RENDER_RING_BASE;
 
 	logical_ring_default_irqs(ring, GEN8_RCS_IRQ_SHIFT);
@@ -2137,6 +2154,7 @@ static int logical_bsd_ring_init(struct drm_device *dev)
 	ring->name = "bsd ring";
 	ring->id = VCS;
 	ring->exec_id = I915_EXEC_BSD;
+	ring->guc_id = GUC_VIDEO_ENGINE;
 	ring->mmio_base = GEN6_BSD_RING_BASE;
 
 	logical_ring_default_irqs(ring, GEN8_VCS1_IRQ_SHIFT);
@@ -2153,6 +2171,7 @@ static int logical_bsd2_ring_init(struct drm_device *dev)
 	ring->name = "bsd2 ring";
 	ring->id = VCS2;
 	ring->exec_id = I915_EXEC_BSD;
+	ring->guc_id = GUC_VIDEO_ENGINE2;
 	ring->mmio_base = GEN8_BSD2_RING_BASE;
 
 	logical_ring_default_irqs(ring, GEN8_VCS2_IRQ_SHIFT);
@@ -2169,6 +2188,7 @@ static int logical_blt_ring_init(struct drm_device *dev)
 	ring->name = "blitter ring";
 	ring->id = BCS;
 	ring->exec_id = I915_EXEC_BLT;
+	ring->guc_id = GUC_BLITTER_ENGINE;
 	ring->mmio_base = BLT_RING_BASE;
 
 	logical_ring_default_irqs(ring, GEN8_BCS_IRQ_SHIFT);
@@ -2185,6 +2205,7 @@ static int logical_vebox_ring_init(struct drm_device *dev)
 	ring->name = "video enhancement ring";
 	ring->id = VECS;
 	ring->exec_id = I915_EXEC_VEBOX;
+	ring->guc_id = GUC_VIDEOENHANCE_ENGINE;
 	ring->mmio_base = VEBOX_RING_BASE;
 
 	logical_ring_default_irqs(ring, GEN8_VECS_IRQ_SHIFT);
@@ -2293,6 +2314,27 @@ make_rpcs(struct drm_device *dev)
 	return rpcs;
 }
 
+static u32 intel_lr_indirect_ctx_offset(struct intel_engine_cs *ring)
+{
+	u32 indirect_ctx_offset;
+
+	switch (INTEL_INFO(ring->dev)->gen) {
+	default:
+		MISSING_CASE(INTEL_INFO(ring->dev)->gen);
+		/* fall through */
+	case 9:
+		indirect_ctx_offset =
+			GEN9_CTX_RCS_INDIRECT_CTX_OFFSET_DEFAULT;
+		break;
+	case 8:
+		indirect_ctx_offset =
+			GEN8_CTX_RCS_INDIRECT_CTX_OFFSET_DEFAULT;
+		break;
+	}
+
+	return indirect_ctx_offset;
+}
+
 static int
 populate_lr_context(struct intel_context *ctx, struct drm_i915_gem_object *ctx_obj,
 		    struct intel_engine_cs *ring, struct intel_ringbuffer *ringbuf)
@@ -2336,7 +2378,8 @@ populate_lr_context(struct intel_context *ctx, struct drm_i915_gem_object *ctx_o
 	ASSIGN_CTX_REG(reg_state, CTX_CONTEXT_CONTROL, RING_CONTEXT_CONTROL(ring),
 		       _MASKED_BIT_ENABLE(CTX_CTRL_INHIBIT_SYN_CTX_SWITCH |
 					  CTX_CTRL_ENGINE_CTX_RESTORE_INHIBIT |
-					  CTX_CTRL_RS_CTX_ENABLE));
+					  (HAS_RESOURCE_STREAMER(dev) ?
+					    CTX_CTRL_RS_CTX_ENABLE : 0)));
 	ASSIGN_CTX_REG(reg_state, CTX_RING_HEAD, RING_HEAD(ring->mmio_base), 0);
 	ASSIGN_CTX_REG(reg_state, CTX_RING_TAIL, RING_TAIL(ring->mmio_base), 0);
 	/* Ring buffer start address is not known until the buffer is pinned.
@@ -2365,7 +2408,7 @@ populate_lr_context(struct intel_context *ctx, struct drm_i915_gem_object *ctx_o
 				(wa_ctx->indirect_ctx.size / CACHELINE_DWORDS);
 
 			reg_state[CTX_RCS_INDIRECT_CTX_OFFSET+1] =
-				CTX_RCS_INDIRECT_CTX_OFFSET_DEFAULT << 6;
+				intel_lr_indirect_ctx_offset(ring) << 6;
 
 			reg_state[CTX_BB_PER_CTX_PTR+1] =
 				(ggtt_offset + wa_ctx->per_ctx.offset * sizeof(uint32_t)) |
diff --git a/drivers/gpu/drm/i915/intel_lrc.h b/drivers/gpu/drm/i915/intel_lrc.h
index 49af638f6213..e6cda3e225d0 100644
--- a/drivers/gpu/drm/i915/intel_lrc.h
+++ b/drivers/gpu/drm/i915/intel_lrc.h
@@ -101,7 +101,8 @@ void intel_lr_context_free(struct intel_context *ctx);
 uint32_t intel_lr_context_size(struct intel_engine_cs *ring);
 int intel_lr_context_deferred_alloc(struct intel_context *ctx,
 				    struct intel_engine_cs *ring);
-void intel_lr_context_unpin(struct drm_i915_gem_request *req);
+void intel_lr_context_unpin(struct intel_context *ctx,
+			    struct intel_engine_cs *engine);
 void intel_lr_context_reset(struct drm_device *dev,
 			struct intel_context *ctx);
 uint64_t intel_lr_context_descriptor(struct intel_context *ctx,
diff --git a/drivers/gpu/drm/i915/intel_lvds.c b/drivers/gpu/drm/i915/intel_lvds.c
index 811ddf7799f0..30a8403a8f4f 100644
--- a/drivers/gpu/drm/i915/intel_lvds.c
+++ b/drivers/gpu/drm/i915/intel_lvds.c
@@ -76,22 +76,30 @@ static bool intel_lvds_get_hw_state(struct intel_encoder *encoder,
 	struct intel_lvds_encoder *lvds_encoder = to_lvds_encoder(&encoder->base);
 	enum intel_display_power_domain power_domain;
 	u32 tmp;
+	bool ret;
 
 	power_domain = intel_display_port_power_domain(encoder);
-	if (!intel_display_power_is_enabled(dev_priv, power_domain))
+	if (!intel_display_power_get_if_enabled(dev_priv, power_domain))
 		return false;
 
+	ret = false;
+
 	tmp = I915_READ(lvds_encoder->reg);
 
 	if (!(tmp & LVDS_PORT_EN))
-		return false;
+		goto out;
 
 	if (HAS_PCH_CPT(dev))
 		*pipe = PORT_TO_PIPE_CPT(tmp);
 	else
 		*pipe = PORT_TO_PIPE(tmp);
 
-	return true;
+	ret = true;
+
+out:
+	intel_display_power_put(dev_priv, power_domain);
+
+	return ret;
 }
 
 static void intel_lvds_get_config(struct intel_encoder *encoder,
diff --git a/drivers/gpu/drm/i915/intel_pm.c b/drivers/gpu/drm/i915/intel_pm.c
index 20bf854eae8c..347d4df49a9b 100644
--- a/drivers/gpu/drm/i915/intel_pm.c
+++ b/drivers/gpu/drm/i915/intel_pm.c
@@ -548,7 +548,7 @@ static const struct intel_watermark_params i845_wm_info = {
  * intel_calculate_wm - calculate watermark level
  * @clock_in_khz: pixel clock
  * @wm: chip FIFO params
- * @pixel_size: display pixel size
+ * @cpp: bytes per pixel
  * @latency_ns: memory latency for the platform
  *
  * Calculate the watermark level (the level at which the display plane will
@@ -564,8 +564,7 @@ static const struct intel_watermark_params i845_wm_info = {
  */
 static unsigned long intel_calculate_wm(unsigned long clock_in_khz,
 					const struct intel_watermark_params *wm,
-					int fifo_size,
-					int pixel_size,
+					int fifo_size, int cpp,
 					unsigned long latency_ns)
 {
 	long entries_required, wm_size;
@@ -576,7 +575,7 @@ static unsigned long intel_calculate_wm(unsigned long clock_in_khz,
 	 * clocks go from a few thousand to several hundred thousand.
 	 * latency is usually a few thousand
 	 */
-	entries_required = ((clock_in_khz / 1000) * pixel_size * latency_ns) /
+	entries_required = ((clock_in_khz / 1000) * cpp * latency_ns) /
 		1000;
 	entries_required = DIV_ROUND_UP(entries_required, wm->cacheline_size);
 
@@ -640,13 +639,13 @@ static void pineview_update_wm(struct drm_crtc *unused_crtc)
 	crtc = single_enabled_crtc(dev);
 	if (crtc) {
 		const struct drm_display_mode *adjusted_mode = &to_intel_crtc(crtc)->config->base.adjusted_mode;
-		int pixel_size = crtc->primary->state->fb->bits_per_pixel / 8;
+		int cpp = drm_format_plane_cpp(crtc->primary->state->fb->pixel_format, 0);
 		int clock = adjusted_mode->crtc_clock;
 
 		/* Display SR */
 		wm = intel_calculate_wm(clock, &pineview_display_wm,
 					pineview_display_wm.fifo_size,
-					pixel_size, latency->display_sr);
+					cpp, latency->display_sr);
 		reg = I915_READ(DSPFW1);
 		reg &= ~DSPFW_SR_MASK;
 		reg |= FW_WM(wm, SR);
@@ -656,7 +655,7 @@ static void pineview_update_wm(struct drm_crtc *unused_crtc)
 		/* cursor SR */
 		wm = intel_calculate_wm(clock, &pineview_cursor_wm,
 					pineview_display_wm.fifo_size,
-					pixel_size, latency->cursor_sr);
+					cpp, latency->cursor_sr);
 		reg = I915_READ(DSPFW3);
 		reg &= ~DSPFW_CURSOR_SR_MASK;
 		reg |= FW_WM(wm, CURSOR_SR);
@@ -665,7 +664,7 @@ static void pineview_update_wm(struct drm_crtc *unused_crtc)
 		/* Display HPLL off SR */
 		wm = intel_calculate_wm(clock, &pineview_display_hplloff_wm,
 					pineview_display_hplloff_wm.fifo_size,
-					pixel_size, latency->display_hpll_disable);
+					cpp, latency->display_hpll_disable);
 		reg = I915_READ(DSPFW3);
 		reg &= ~DSPFW_HPLL_SR_MASK;
 		reg |= FW_WM(wm, HPLL_SR);
@@ -674,7 +673,7 @@ static void pineview_update_wm(struct drm_crtc *unused_crtc)
 		/* cursor HPLL off SR */
 		wm = intel_calculate_wm(clock, &pineview_cursor_hplloff_wm,
 					pineview_display_hplloff_wm.fifo_size,
-					pixel_size, latency->cursor_hpll_disable);
+					cpp, latency->cursor_hpll_disable);
 		reg = I915_READ(DSPFW3);
 		reg &= ~DSPFW_HPLL_CURSOR_MASK;
 		reg |= FW_WM(wm, HPLL_CURSOR);
@@ -698,7 +697,7 @@ static bool g4x_compute_wm0(struct drm_device *dev,
 {
 	struct drm_crtc *crtc;
 	const struct drm_display_mode *adjusted_mode;
-	int htotal, hdisplay, clock, pixel_size;
+	int htotal, hdisplay, clock, cpp;
 	int line_time_us, line_count;
 	int entries, tlb_miss;
 
@@ -713,10 +712,10 @@ static bool g4x_compute_wm0(struct drm_device *dev,
 	clock = adjusted_mode->crtc_clock;
 	htotal = adjusted_mode->crtc_htotal;
 	hdisplay = to_intel_crtc(crtc)->config->pipe_src_w;
-	pixel_size = crtc->primary->state->fb->bits_per_pixel / 8;
+	cpp = drm_format_plane_cpp(crtc->primary->state->fb->pixel_format, 0);
 
 	/* Use the small buffer method to calculate plane watermark */
-	entries = ((clock * pixel_size / 1000) * display_latency_ns) / 1000;
+	entries = ((clock * cpp / 1000) * display_latency_ns) / 1000;
 	tlb_miss = display->fifo_size*display->cacheline_size - hdisplay * 8;
 	if (tlb_miss > 0)
 		entries += tlb_miss;
@@ -728,7 +727,7 @@ static bool g4x_compute_wm0(struct drm_device *dev,
 	/* Use the large buffer method to calculate cursor watermark */
 	line_time_us = max(htotal * 1000 / clock, 1);
 	line_count = (cursor_latency_ns / line_time_us + 1000) / 1000;
-	entries = line_count * crtc->cursor->state->crtc_w * pixel_size;
+	entries = line_count * crtc->cursor->state->crtc_w * cpp;
 	tlb_miss = cursor->fifo_size*cursor->cacheline_size - hdisplay * 8;
 	if (tlb_miss > 0)
 		entries += tlb_miss;
@@ -784,7 +783,7 @@ static bool g4x_compute_srwm(struct drm_device *dev,
 {
 	struct drm_crtc *crtc;
 	const struct drm_display_mode *adjusted_mode;
-	int hdisplay, htotal, pixel_size, clock;
+	int hdisplay, htotal, cpp, clock;
 	unsigned long line_time_us;
 	int line_count, line_size;
 	int small, large;
@@ -800,21 +799,21 @@ static bool g4x_compute_srwm(struct drm_device *dev,
 	clock = adjusted_mode->crtc_clock;
 	htotal = adjusted_mode->crtc_htotal;
 	hdisplay = to_intel_crtc(crtc)->config->pipe_src_w;
-	pixel_size = crtc->primary->state->fb->bits_per_pixel / 8;
+	cpp = drm_format_plane_cpp(crtc->primary->state->fb->pixel_format, 0);
 
 	line_time_us = max(htotal * 1000 / clock, 1);
 	line_count = (latency_ns / line_time_us + 1000) / 1000;
-	line_size = hdisplay * pixel_size;
+	line_size = hdisplay * cpp;
 
 	/* Use the minimum of the small and large buffer method for primary */
-	small = ((clock * pixel_size / 1000) * latency_ns) / 1000;
+	small = ((clock * cpp / 1000) * latency_ns) / 1000;
 	large = line_count * line_size;
 
 	entries = DIV_ROUND_UP(min(small, large), display->cacheline_size);
 	*display_wm = entries + display->guard_size;
 
 	/* calculate the self-refresh watermark for display cursor */
-	entries = line_count * pixel_size * crtc->cursor->state->crtc_w;
+	entries = line_count * cpp * crtc->cursor->state->crtc_w;
 	entries = DIV_ROUND_UP(entries, cursor->cacheline_size);
 	*cursor_wm = entries + cursor->guard_size;
 
@@ -906,13 +905,13 @@ enum vlv_wm_level {
 static unsigned int vlv_wm_method2(unsigned int pixel_rate,
 				   unsigned int pipe_htotal,
 				   unsigned int horiz_pixels,
-				   unsigned int bytes_per_pixel,
+				   unsigned int cpp,
 				   unsigned int latency)
 {
 	unsigned int ret;
 
 	ret = (latency * pixel_rate) / (pipe_htotal * 10000);
-	ret = (ret + 1) * horiz_pixels * bytes_per_pixel;
+	ret = (ret + 1) * horiz_pixels * cpp;
 	ret = DIV_ROUND_UP(ret, 64);
 
 	return ret;
@@ -941,7 +940,7 @@ static uint16_t vlv_compute_wm_level(struct intel_plane *plane,
 				     int level)
 {
 	struct drm_i915_private *dev_priv = to_i915(plane->base.dev);
-	int clock, htotal, pixel_size, width, wm;
+	int clock, htotal, cpp, width, wm;
 
 	if (dev_priv->wm.pri_latency[level] == 0)
 		return USHRT_MAX;
@@ -949,7 +948,7 @@ static uint16_t vlv_compute_wm_level(struct intel_plane *plane,
 	if (!state->visible)
 		return 0;
 
-	pixel_size = drm_format_plane_cpp(state->base.fb->pixel_format, 0);
+	cpp = drm_format_plane_cpp(state->base.fb->pixel_format, 0);
 	clock = crtc->config->base.adjusted_mode.crtc_clock;
 	htotal = crtc->config->base.adjusted_mode.crtc_htotal;
 	width = crtc->config->pipe_src_w;
@@ -965,7 +964,7 @@ static uint16_t vlv_compute_wm_level(struct intel_plane *plane,
 		 */
 		wm = 63;
 	} else {
-		wm = vlv_wm_method2(clock, htotal, width, pixel_size,
+		wm = vlv_wm_method2(clock, htotal, width, cpp,
 				    dev_priv->wm.pri_latency[level] * 10);
 	}
 
@@ -1439,7 +1438,7 @@ static void i965_update_wm(struct drm_crtc *unused_crtc)
 		int clock = adjusted_mode->crtc_clock;
 		int htotal = adjusted_mode->crtc_htotal;
 		int hdisplay = to_intel_crtc(crtc)->config->pipe_src_w;
-		int pixel_size = crtc->primary->state->fb->bits_per_pixel / 8;
+		int cpp = drm_format_plane_cpp(crtc->primary->state->fb->pixel_format, 0);
 		unsigned long line_time_us;
 		int entries;
 
@@ -1447,7 +1446,7 @@ static void i965_update_wm(struct drm_crtc *unused_crtc)
 
 		/* Use ns/us then divide to preserve precision */
 		entries = (((sr_latency_ns / line_time_us) + 1000) / 1000) *
-			pixel_size * hdisplay;
+			cpp * hdisplay;
 		entries = DIV_ROUND_UP(entries, I915_FIFO_LINE_SIZE);
 		srwm = I965_FIFO_SIZE - entries;
 		if (srwm < 0)
@@ -1457,7 +1456,7 @@ static void i965_update_wm(struct drm_crtc *unused_crtc)
 			      entries, srwm);
 
 		entries = (((sr_latency_ns / line_time_us) + 1000) / 1000) *
-			pixel_size * crtc->cursor->state->crtc_w;
+			cpp * crtc->cursor->state->crtc_w;
 		entries = DIV_ROUND_UP(entries,
 					  i965_cursor_wm_info.cacheline_size);
 		cursor_sr = i965_cursor_wm_info.fifo_size -
@@ -1518,7 +1517,7 @@ static void i9xx_update_wm(struct drm_crtc *unused_crtc)
 	crtc = intel_get_crtc_for_plane(dev, 0);
 	if (intel_crtc_active(crtc)) {
 		const struct drm_display_mode *adjusted_mode;
-		int cpp = crtc->primary->state->fb->bits_per_pixel / 8;
+		int cpp = drm_format_plane_cpp(crtc->primary->state->fb->pixel_format, 0);
 		if (IS_GEN2(dev))
 			cpp = 4;
 
@@ -1540,7 +1539,7 @@ static void i9xx_update_wm(struct drm_crtc *unused_crtc)
 	crtc = intel_get_crtc_for_plane(dev, 1);
 	if (intel_crtc_active(crtc)) {
 		const struct drm_display_mode *adjusted_mode;
-		int cpp = crtc->primary->state->fb->bits_per_pixel / 8;
+		int cpp = drm_format_plane_cpp(crtc->primary->state->fb->pixel_format, 0);
 		if (IS_GEN2(dev))
 			cpp = 4;
 
@@ -1586,7 +1585,7 @@ static void i9xx_update_wm(struct drm_crtc *unused_crtc)
 		int clock = adjusted_mode->crtc_clock;
 		int htotal = adjusted_mode->crtc_htotal;
 		int hdisplay = to_intel_crtc(enabled)->config->pipe_src_w;
-		int pixel_size = enabled->primary->state->fb->bits_per_pixel / 8;
+		int cpp = drm_format_plane_cpp(enabled->primary->state->fb->pixel_format, 0);
 		unsigned long line_time_us;
 		int entries;
 
@@ -1594,7 +1593,7 @@ static void i9xx_update_wm(struct drm_crtc *unused_crtc)
 
 		/* Use ns/us then divide to preserve precision */
 		entries = (((sr_latency_ns / line_time_us) + 1000) / 1000) *
-			pixel_size * hdisplay;
+			cpp * hdisplay;
 		entries = DIV_ROUND_UP(entries, wm_info->cacheline_size);
 		DRM_DEBUG_KMS("self-refresh entries: %d\n", entries);
 		srwm = wm_info->fifo_size - entries;
@@ -1685,15 +1684,14 @@ uint32_t ilk_pipe_pixel_rate(const struct intel_crtc_state *pipe_config)
 }
 
 /* latency must be in 0.1us units. */
-static uint32_t ilk_wm_method1(uint32_t pixel_rate, uint8_t bytes_per_pixel,
-			       uint32_t latency)
+static uint32_t ilk_wm_method1(uint32_t pixel_rate, uint8_t cpp, uint32_t latency)
 {
 	uint64_t ret;
 
 	if (WARN(latency == 0, "Latency value missing\n"))
 		return UINT_MAX;
 
-	ret = (uint64_t) pixel_rate * bytes_per_pixel * latency;
+	ret = (uint64_t) pixel_rate * cpp * latency;
 	ret = DIV_ROUND_UP_ULL(ret, 64 * 10000) + 2;
 
 	return ret;
@@ -1701,7 +1699,7 @@ static uint32_t ilk_wm_method1(uint32_t pixel_rate, uint8_t bytes_per_pixel,
 
 /* latency must be in 0.1us units. */
 static uint32_t ilk_wm_method2(uint32_t pixel_rate, uint32_t pipe_htotal,
-			       uint32_t horiz_pixels, uint8_t bytes_per_pixel,
+			       uint32_t horiz_pixels, uint8_t cpp,
 			       uint32_t latency)
 {
 	uint32_t ret;
@@ -1712,13 +1710,13 @@ static uint32_t ilk_wm_method2(uint32_t pixel_rate, uint32_t pipe_htotal,
 		return UINT_MAX;
 
 	ret = (latency * pixel_rate) / (pipe_htotal * 10000);
-	ret = (ret + 1) * horiz_pixels * bytes_per_pixel;
+	ret = (ret + 1) * horiz_pixels * cpp;
 	ret = DIV_ROUND_UP(ret, 64) + 2;
 	return ret;
 }
 
 static uint32_t ilk_wm_fbc(uint32_t pri_val, uint32_t horiz_pixels,
-			   uint8_t bytes_per_pixel)
+			   uint8_t cpp)
 {
 	/*
 	 * Neither of these should be possible since this function shouldn't be
@@ -1726,12 +1724,12 @@ static uint32_t ilk_wm_fbc(uint32_t pri_val, uint32_t horiz_pixels,
 	 * extra paranoid to avoid a potential divide-by-zero if we screw up
 	 * elsewhere in the driver.
 	 */
-	if (WARN_ON(!bytes_per_pixel))
+	if (WARN_ON(!cpp))
 		return 0;
 	if (WARN_ON(!horiz_pixels))
 		return 0;
 
-	return DIV_ROUND_UP(pri_val * 64, horiz_pixels * bytes_per_pixel) + 2;
+	return DIV_ROUND_UP(pri_val * 64, horiz_pixels * cpp) + 2;
 }
 
 struct ilk_wm_maximums {
@@ -1750,13 +1748,14 @@ static uint32_t ilk_compute_pri_wm(const struct intel_crtc_state *cstate,
 				   uint32_t mem_value,
 				   bool is_lp)
 {
-	int bpp = pstate->base.fb ? pstate->base.fb->bits_per_pixel / 8 : 0;
+	int cpp = pstate->base.fb ?
+		drm_format_plane_cpp(pstate->base.fb->pixel_format, 0) : 0;
 	uint32_t method1, method2;
 
 	if (!cstate->base.active || !pstate->visible)
 		return 0;
 
-	method1 = ilk_wm_method1(ilk_pipe_pixel_rate(cstate), bpp, mem_value);
+	method1 = ilk_wm_method1(ilk_pipe_pixel_rate(cstate), cpp, mem_value);
 
 	if (!is_lp)
 		return method1;
@@ -1764,8 +1763,7 @@ static uint32_t ilk_compute_pri_wm(const struct intel_crtc_state *cstate,
 	method2 = ilk_wm_method2(ilk_pipe_pixel_rate(cstate),
 				 cstate->base.adjusted_mode.crtc_htotal,
 				 drm_rect_width(&pstate->dst),
-				 bpp,
-				 mem_value);
+				 cpp, mem_value);
 
 	return min(method1, method2);
 }
@@ -1778,18 +1776,18 @@ static uint32_t ilk_compute_spr_wm(const struct intel_crtc_state *cstate,
 				   const struct intel_plane_state *pstate,
 				   uint32_t mem_value)
 {
-	int bpp = pstate->base.fb ? pstate->base.fb->bits_per_pixel / 8 : 0;
+	int cpp = pstate->base.fb ?
+		drm_format_plane_cpp(pstate->base.fb->pixel_format, 0) : 0;
 	uint32_t method1, method2;
 
 	if (!cstate->base.active || !pstate->visible)
 		return 0;
 
-	method1 = ilk_wm_method1(ilk_pipe_pixel_rate(cstate), bpp, mem_value);
+	method1 = ilk_wm_method1(ilk_pipe_pixel_rate(cstate), cpp, mem_value);
 	method2 = ilk_wm_method2(ilk_pipe_pixel_rate(cstate),
 				 cstate->base.adjusted_mode.crtc_htotal,
 				 drm_rect_width(&pstate->dst),
-				 bpp,
-				 mem_value);
+				 cpp, mem_value);
 	return min(method1, method2);
 }
 
@@ -1801,16 +1799,20 @@ static uint32_t ilk_compute_cur_wm(const struct intel_crtc_state *cstate,
 				   const struct intel_plane_state *pstate,
 				   uint32_t mem_value)
 {
-	int bpp = pstate->base.fb ? pstate->base.fb->bits_per_pixel / 8 : 0;
+	/*
+	 * We treat the cursor plane as always-on for the purposes of watermark
+	 * calculation.  Until we have two-stage watermark programming merged,
+	 * this is necessary to avoid flickering.
+	 */
+	int cpp = 4;
+	int width = pstate->visible ? pstate->base.crtc_w : 64;
 
-	if (!cstate->base.active || !pstate->visible)
+	if (!cstate->base.active)
 		return 0;
 
 	return ilk_wm_method2(ilk_pipe_pixel_rate(cstate),
 			      cstate->base.adjusted_mode.crtc_htotal,
-			      drm_rect_width(&pstate->dst),
-			      bpp,
-			      mem_value);
+			      width, cpp, mem_value);
 }
 
 /* Only for WM_LP. */
@@ -1818,12 +1820,13 @@ static uint32_t ilk_compute_fbc_wm(const struct intel_crtc_state *cstate,
 				   const struct intel_plane_state *pstate,
 				   uint32_t pri_val)
 {
-	int bpp = pstate->base.fb ? pstate->base.fb->bits_per_pixel / 8 : 0;
+	int cpp = pstate->base.fb ?
+		drm_format_plane_cpp(pstate->base.fb->pixel_format, 0) : 0;
 
 	if (!cstate->base.active || !pstate->visible)
 		return 0;
 
-	return ilk_wm_fbc(pri_val, drm_rect_width(&pstate->dst), bpp);
+	return ilk_wm_fbc(pri_val, drm_rect_width(&pstate->dst), cpp);
 }
 
 static unsigned int ilk_display_fifo_size(const struct drm_device *dev)
@@ -2848,7 +2851,10 @@ void skl_ddb_get_hw_state(struct drm_i915_private *dev_priv,
 	memset(ddb, 0, sizeof(*ddb));
 
 	for_each_pipe(dev_priv, pipe) {
-		if (!intel_display_power_is_enabled(dev_priv, POWER_DOMAIN_PIPE(pipe)))
+		enum intel_display_power_domain power_domain;
+
+		power_domain = POWER_DOMAIN_PIPE(pipe);
+		if (!intel_display_power_get_if_enabled(dev_priv, power_domain))
 			continue;
 
 		for_each_plane(dev_priv, pipe, plane) {
@@ -2860,6 +2866,8 @@ void skl_ddb_get_hw_state(struct drm_i915_private *dev_priv,
 		val = I915_READ(CUR_BUF_CFG(pipe));
 		skl_ddb_entry_init_from_hw(&ddb->plane[pipe][PLANE_CURSOR],
 					   val);
+
+		intel_display_power_put(dev_priv, power_domain);
 	}
 }
 
@@ -3042,26 +3050,25 @@ static uint32_t skl_pipe_pixel_rate(const struct intel_crtc_state *config)
 
 /*
  * The max latency should be 257 (max the punit can code is 255 and we add 2us
- * for the read latency) and bytes_per_pixel should always be <= 8, so that
+ * for the read latency) and cpp should always be <= 8, so that
  * should allow pixel_rate up to ~2 GHz which seems sufficient since max
  * 2xcdclk is 1350 MHz and the pixel rate should never exceed that.
 */
-static uint32_t skl_wm_method1(uint32_t pixel_rate, uint8_t bytes_per_pixel,
-			       uint32_t latency)
+static uint32_t skl_wm_method1(uint32_t pixel_rate, uint8_t cpp, uint32_t latency)
 {
 	uint32_t wm_intermediate_val, ret;
 
 	if (latency == 0)
 		return UINT_MAX;
 
-	wm_intermediate_val = latency * pixel_rate * bytes_per_pixel / 512;
+	wm_intermediate_val = latency * pixel_rate * cpp / 512;
 	ret = DIV_ROUND_UP(wm_intermediate_val, 1000);
 
 	return ret;
 }
 
 static uint32_t skl_wm_method2(uint32_t pixel_rate, uint32_t pipe_htotal,
-			       uint32_t horiz_pixels, uint8_t bytes_per_pixel,
+			       uint32_t horiz_pixels, uint8_t cpp,
 			       uint64_t tiling, uint32_t latency)
 {
 	uint32_t ret;
@@ -3071,7 +3078,7 @@ static uint32_t skl_wm_method2(uint32_t pixel_rate, uint32_t pipe_htotal,
 	if (latency == 0)
 		return UINT_MAX;
 
-	plane_bytes_per_line = horiz_pixels * bytes_per_pixel;
+	plane_bytes_per_line = horiz_pixels * cpp;
 
 	if (tiling == I915_FORMAT_MOD_Y_TILED ||
 	    tiling == I915_FORMAT_MOD_Yf_TILED) {
@@ -3121,23 +3128,21 @@ static bool skl_compute_plane_wm(const struct drm_i915_private *dev_priv,
 	uint32_t plane_bytes_per_line, plane_blocks_per_line;
 	uint32_t res_blocks, res_lines;
 	uint32_t selected_result;
-	uint8_t bytes_per_pixel;
+	uint8_t cpp;
 
 	if (latency == 0 || !cstate->base.active || !fb)
 		return false;
 
-	bytes_per_pixel = drm_format_plane_cpp(fb->pixel_format, 0);
+	cpp = drm_format_plane_cpp(fb->pixel_format, 0);
 	method1 = skl_wm_method1(skl_pipe_pixel_rate(cstate),
-				 bytes_per_pixel,
-				 latency);
+				 cpp, latency);
 	method2 = skl_wm_method2(skl_pipe_pixel_rate(cstate),
 				 cstate->base.adjusted_mode.crtc_htotal,
 				 cstate->pipe_src_w,
-				 bytes_per_pixel,
-				 fb->modifier[0],
+				 cpp, fb->modifier[0],
 				 latency);
 
-	plane_bytes_per_line = cstate->pipe_src_w * bytes_per_pixel;
+	plane_bytes_per_line = cstate->pipe_src_w * cpp;
 	plane_blocks_per_line = DIV_ROUND_UP(plane_bytes_per_line, 512);
 
 	if (fb->modifier[0] == I915_FORMAT_MOD_Y_TILED ||
@@ -3145,11 +3150,11 @@ static bool skl_compute_plane_wm(const struct drm_i915_private *dev_priv,
 		uint32_t min_scanlines = 4;
 		uint32_t y_tile_minimum;
 		if (intel_rotation_90_or_270(plane->state->rotation)) {
-			int bpp = (fb->pixel_format == DRM_FORMAT_NV12) ?
+			int cpp = (fb->pixel_format == DRM_FORMAT_NV12) ?
 				drm_format_plane_cpp(fb->pixel_format, 1) :
 				drm_format_plane_cpp(fb->pixel_format, 0);
 
-			switch (bpp) {
+			switch (cpp) {
 			case 1:
 				min_scanlines = 16;
 				break;
@@ -4116,11 +4121,13 @@ bool ironlake_set_drps(struct drm_device *dev, u8 val)
 static void ironlake_enable_drps(struct drm_device *dev)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
-	u32 rgvmodectl = I915_READ(MEMMODECTL);
+	u32 rgvmodectl;
 	u8 fmax, fmin, fstart, vstart;
 
 	spin_lock_irq(&mchdev_lock);
 
+	rgvmodectl = I915_READ(MEMMODECTL);
+
 	/* Enable temp reporting */
 	I915_WRITE16(PMMISC, I915_READ(PMMISC) | MCPPCE_EN);
 	I915_WRITE16(TSC1, I915_READ(TSC1) | TSE);
@@ -4562,12 +4569,62 @@ static void intel_print_rc6_info(struct drm_device *dev, u32 mode)
 			      onoff(mode & GEN6_RC_CTL_RC6_ENABLE));
 }
 
-static int sanitize_rc6_option(const struct drm_device *dev, int enable_rc6)
+static bool bxt_check_bios_rc6_setup(const struct drm_device *dev)
+{
+	struct drm_i915_private *dev_priv = dev->dev_private;
+	bool enable_rc6 = true;
+	unsigned long rc6_ctx_base;
+
+	if (!(I915_READ(RC6_LOCATION) & RC6_CTX_IN_DRAM)) {
+		DRM_DEBUG_KMS("RC6 Base location not set properly.\n");
+		enable_rc6 = false;
+	}
+
+	/*
+	 * The exact context size is not known for BXT, so assume a page size
+	 * for this check.
+	 */
+	rc6_ctx_base = I915_READ(RC6_CTX_BASE) & RC6_CTX_BASE_MASK;
+	if (!((rc6_ctx_base >= dev_priv->gtt.stolen_reserved_base) &&
+	      (rc6_ctx_base + PAGE_SIZE <= dev_priv->gtt.stolen_reserved_base +
+					dev_priv->gtt.stolen_reserved_size))) {
+		DRM_DEBUG_KMS("RC6 Base address not as expected.\n");
+		enable_rc6 = false;
+	}
+
+	if (!(((I915_READ(PWRCTX_MAXCNT_RCSUNIT) & IDLE_TIME_MASK) > 1) &&
+	      ((I915_READ(PWRCTX_MAXCNT_VCSUNIT0) & IDLE_TIME_MASK) > 1) &&
+	      ((I915_READ(PWRCTX_MAXCNT_BCSUNIT) & IDLE_TIME_MASK) > 1) &&
+	      ((I915_READ(PWRCTX_MAXCNT_VECSUNIT) & IDLE_TIME_MASK) > 1))) {
+		DRM_DEBUG_KMS("Engine Idle wait time not set properly.\n");
+		enable_rc6 = false;
+	}
+
+	if (!(I915_READ(GEN6_RC_CONTROL) & (GEN6_RC_CTL_RC6_ENABLE |
+					    GEN6_RC_CTL_HW_ENABLE)) &&
+	    ((I915_READ(GEN6_RC_CONTROL) & GEN6_RC_CTL_HW_ENABLE) ||
+	     !(I915_READ(GEN6_RC_STATE) & RC6_STATE))) {
+		DRM_DEBUG_KMS("HW/SW RC6 is not enabled by BIOS.\n");
+		enable_rc6 = false;
+	}
+
+	return enable_rc6;
+}
+
+int sanitize_rc6_option(const struct drm_device *dev, int enable_rc6)
 {
 	/* No RC6 before Ironlake and code is gone for ilk. */
 	if (INTEL_INFO(dev)->gen < 6)
 		return 0;
 
+	if (!enable_rc6)
+		return 0;
+
+	if (IS_BROXTON(dev) && !bxt_check_bios_rc6_setup(dev)) {
+		DRM_INFO("RC6 disabled by BIOS\n");
+		return 0;
+	}
+
 	/* Respect the kernel parameter if it is set */
 	if (enable_rc6 >= 0) {
 		int mask;
@@ -5179,8 +5236,6 @@ static void cherryview_setup_pctx(struct drm_device *dev)
 	u32 pcbr;
 	int pctx_size = 32*1024;
 
-	WARN_ON(!mutex_is_locked(&dev->struct_mutex));
-
 	pcbr = I915_READ(VLV_PCBR);
 	if ((pcbr >> VLV_PCBR_ADDR_SHIFT) == 0) {
 		DRM_DEBUG_DRIVER("BIOS didn't set up PCBR, fixing up\n");
@@ -5202,7 +5257,7 @@ static void valleyview_setup_pctx(struct drm_device *dev)
 	u32 pcbr;
 	int pctx_size = 24*1024;
 
-	WARN_ON(!mutex_is_locked(&dev->struct_mutex));
+	mutex_lock(&dev->struct_mutex);
 
 	pcbr = I915_READ(VLV_PCBR);
 	if (pcbr) {
@@ -5230,7 +5285,7 @@ static void valleyview_setup_pctx(struct drm_device *dev)
 	pctx = i915_gem_object_create_stolen(dev, pctx_size);
 	if (!pctx) {
 		DRM_DEBUG("not enough stolen space for PCTX, disabling\n");
-		return;
+		goto out;
 	}
 
 	pctx_paddr = dev_priv->mm.stolen_base + pctx->stolen->start;
@@ -5239,6 +5294,7 @@ static void valleyview_setup_pctx(struct drm_device *dev)
 out:
 	DRM_DEBUG_DRIVER("PCBR: 0x%08x\n", I915_READ(VLV_PCBR));
 	dev_priv->vlv_pctx = pctx;
+	mutex_unlock(&dev->struct_mutex);
 }
 
 static void valleyview_cleanup_pctx(struct drm_device *dev)
@@ -5248,7 +5304,7 @@ static void valleyview_cleanup_pctx(struct drm_device *dev)
 	if (WARN_ON(!dev_priv->vlv_pctx))
 		return;
 
-	drm_gem_object_unreference(&dev_priv->vlv_pctx->base);
+	drm_gem_object_unreference_unlocked(&dev_priv->vlv_pctx->base);
 	dev_priv->vlv_pctx = NULL;
 }
 
@@ -6057,7 +6113,6 @@ void intel_init_gt_powersave(struct drm_device *dev)
 {
 	struct drm_i915_private *dev_priv = dev->dev_private;
 
-	i915.enable_rc6 = sanitize_rc6_option(dev, i915.enable_rc6);
 	/*
 	 * RPM depends on RC6 to save restore the GT HW context, so make RC6 a
 	 * requirement.
@@ -6192,8 +6247,8 @@ void intel_enable_gt_powersave(struct drm_device *dev)
 		return;
 
 	if (IS_IRONLAKE_M(dev)) {
-		mutex_lock(&dev->struct_mutex);
 		ironlake_enable_drps(dev);
+		mutex_lock(&dev->struct_mutex);
 		intel_init_emon(dev);
 		mutex_unlock(&dev->struct_mutex);
 	} else if (INTEL_INFO(dev)->gen >= 6) {
@@ -7189,9 +7244,10 @@ static int chv_gpu_freq(struct drm_i915_private *dev_priv, int val)
 {
 	int div, czclk_freq = DIV_ROUND_CLOSEST(dev_priv->czclk_freq, 1000);
 
-	div = vlv_gpu_freq_div(czclk_freq) / 2;
+	div = vlv_gpu_freq_div(czclk_freq);
 	if (div < 0)
 		return div;
+	div /= 2;
 
 	return DIV_ROUND_CLOSEST(czclk_freq * val, 2 * div) / 2;
 }
@@ -7200,9 +7256,10 @@ static int chv_freq_opcode(struct drm_i915_private *dev_priv, int val)
 {
 	int mul, czclk_freq = DIV_ROUND_CLOSEST(dev_priv->czclk_freq, 1000);
 
-	mul = vlv_gpu_freq_div(czclk_freq) / 2;
+	mul = vlv_gpu_freq_div(czclk_freq);
 	if (mul < 0)
 		return mul;
+	mul /= 2;
 
 	/* CHV needs even values */
 	return DIV_ROUND_CLOSEST(val * 2 * mul, czclk_freq) * 2;
diff --git a/drivers/gpu/drm/i915/intel_psr.c b/drivers/gpu/drm/i915/intel_psr.c
index 9ccff3011523..0b42ada338c8 100644
--- a/drivers/gpu/drm/i915/intel_psr.c
+++ b/drivers/gpu/drm/i915/intel_psr.c
@@ -225,7 +225,12 @@ static void hsw_psr_enable_sink(struct intel_dp *intel_dp)
 		   (aux_clock_divider << DP_AUX_CH_CTL_BIT_CLOCK_2X_SHIFT));
 	}
 
-	drm_dp_dpcd_writeb(&intel_dp->aux, DP_PSR_EN_CFG, DP_PSR_ENABLE);
+	if (dev_priv->psr.link_standby)
+		drm_dp_dpcd_writeb(&intel_dp->aux, DP_PSR_EN_CFG,
+				   DP_PSR_ENABLE | DP_PSR_MAIN_LINK_ACTIVE);
+	else
+		drm_dp_dpcd_writeb(&intel_dp->aux, DP_PSR_EN_CFG,
+				   DP_PSR_ENABLE);
 }
 
 static void vlv_psr_enable_source(struct intel_dp *intel_dp)
@@ -280,6 +285,9 @@ static void hsw_psr_enable_source(struct intel_dp *intel_dp)
 	if (IS_HASWELL(dev))
 		val |= EDP_PSR_MIN_LINK_ENTRY_TIME_8_LINES;
 
+	if (dev_priv->psr.link_standby)
+		val |= EDP_PSR_LINK_STANDBY;
+
 	I915_WRITE(EDP_PSR_CTL, val |
 		   max_sleep_time << EDP_PSR_MAX_SLEEP_TIME_SHIFT |
 		   idle_frames << EDP_PSR_IDLE_FRAME_SHIFT |
@@ -304,8 +312,15 @@ static bool intel_psr_match_conditions(struct intel_dp *intel_dp)
 
 	dev_priv->psr.source_ok = false;
 
-	if (IS_HASWELL(dev) && dig_port->port != PORT_A) {
-		DRM_DEBUG_KMS("HSW ties PSR to DDI A (eDP)\n");
+	/*
+	 * HSW spec explicitly says PSR is tied to port A.
+	 * BDW+ platforms with DDI implementation of PSR have different
+	 * PSR registers per transcoder and we only implement transcoder EDP
+	 * ones. Since by Display design transcoder EDP is tied to port A
+	 * we can safely escape based on the port A.
+	 */
+	if (HAS_DDI(dev) && dig_port->port != PORT_A) {
+		DRM_DEBUG_KMS("PSR condition failed: Port not supported\n");
 		return false;
 	}
 
@@ -314,6 +329,12 @@ static bool intel_psr_match_conditions(struct intel_dp *intel_dp)
 		return false;
 	}
 
+	if ((IS_VALLEYVIEW(dev) || IS_CHERRYVIEW(dev)) &&
+	    !dev_priv->psr.link_standby) {
+		DRM_ERROR("PSR condition failed: Link off requested but not supported on this platform\n");
+		return false;
+	}
+
 	if (IS_HASWELL(dev) &&
 	    I915_READ(HSW_STEREO_3D_CTL(intel_crtc->config->cpu_transcoder)) &
 		      S3D_ENABLE) {
@@ -327,12 +348,6 @@ static bool intel_psr_match_conditions(struct intel_dp *intel_dp)
 		return false;
 	}
 
-	if (!IS_VALLEYVIEW(dev) && !IS_CHERRYVIEW(dev) &&
-	    ((dev_priv->vbt.psr.full_link) || (dig_port->port != PORT_A))) {
-		DRM_DEBUG_KMS("PSR condition failed: Link Standby requested/needed but not supported on this platform\n");
-		return false;
-	}
-
 	dev_priv->psr.source_ok = true;
 	return true;
 }
@@ -763,6 +778,36 @@ void intel_psr_init(struct drm_device *dev)
 	dev_priv->psr_mmio_base = IS_HASWELL(dev_priv) ?
 		HSW_EDP_PSR_BASE : BDW_EDP_PSR_BASE;
 
+	/* Per platform default */
+	if (i915.enable_psr == -1) {
+		if (IS_HASWELL(dev) || IS_BROADWELL(dev) ||
+		    IS_VALLEYVIEW(dev) || IS_CHERRYVIEW(dev))
+			i915.enable_psr = 1;
+		else
+			i915.enable_psr = 0;
+	}
+
+	/* Set link_standby x link_off defaults */
+	if (IS_HASWELL(dev) || IS_BROADWELL(dev))
+		/* HSW and BDW require workarounds that we don't implement. */
+		dev_priv->psr.link_standby = false;
+	else if (IS_VALLEYVIEW(dev) || IS_CHERRYVIEW(dev))
+		/* On VLV and CHV only standby mode is supported. */
+		dev_priv->psr.link_standby = true;
+	else
+		/* For new platforms let's respect VBT back again */
+		dev_priv->psr.link_standby = dev_priv->vbt.psr.full_link;
+
+	/* Override link_standby x link_off defaults */
+	if (i915.enable_psr == 2 && !dev_priv->psr.link_standby) {
+		DRM_DEBUG_KMS("PSR: Forcing link standby\n");
+		dev_priv->psr.link_standby = true;
+	}
+	if (i915.enable_psr == 3 && dev_priv->psr.link_standby) {
+		DRM_DEBUG_KMS("PSR: Forcing main link off\n");
+		dev_priv->psr.link_standby = false;
+	}
+
 	INIT_DELAYED_WORK(&dev_priv->psr.work, intel_psr_work);
 	mutex_init(&dev_priv->psr.lock);
 }
diff --git a/drivers/gpu/drm/i915/intel_ringbuffer.c b/drivers/gpu/drm/i915/intel_ringbuffer.c
index 9030e2bca0c0..45ce45a5e122 100644
--- a/drivers/gpu/drm/i915/intel_ringbuffer.c
+++ b/drivers/gpu/drm/i915/intel_ringbuffer.c
@@ -746,9 +746,9 @@ static int intel_rcs_ctx_init(struct drm_i915_gem_request *req)
 
 	ret = i915_gem_render_state_init(req);
 	if (ret)
-		DRM_ERROR("init render state: %d\n", ret);
+		return ret;
 
-	return ret;
+	return 0;
 }
 
 static int wa_add(struct drm_i915_private *dev_priv,
@@ -789,6 +789,22 @@ static int wa_add(struct drm_i915_private *dev_priv,
 
 #define WA_WRITE(addr, val) WA_REG(addr, 0xffffffff, val)
 
+static int wa_ring_whitelist_reg(struct intel_engine_cs *ring, i915_reg_t reg)
+{
+	struct drm_i915_private *dev_priv = ring->dev->dev_private;
+	struct i915_workarounds *wa = &dev_priv->workarounds;
+	const uint32_t index = wa->hw_whitelist_count[ring->id];
+
+	if (WARN_ON(index >= RING_MAX_NONPRIV_SLOTS))
+		return -EINVAL;
+
+	WA_WRITE(RING_FORCE_TO_NONPRIV(ring->mmio_base, index),
+		 i915_mmio_reg_offset(reg));
+	wa->hw_whitelist_count[ring->id]++;
+
+	return 0;
+}
+
 static int gen8_init_workarounds(struct intel_engine_cs *ring)
 {
 	struct drm_device *dev = ring->dev;
@@ -894,6 +910,7 @@ static int gen9_init_workarounds(struct intel_engine_cs *ring)
 	struct drm_device *dev = ring->dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	uint32_t tmp;
+	int ret;
 
 	/* WaEnableLbsSlaRetryTimerDecrement:skl */
 	I915_WRITE(BDW_SCRATCH1, I915_READ(BDW_SCRATCH1) |
@@ -964,6 +981,20 @@ static int gen9_init_workarounds(struct intel_engine_cs *ring)
 	/* WaDisableSTUnitPowerOptimization:skl,bxt */
 	WA_SET_BIT_MASKED(HALF_SLICE_CHICKEN2, GEN8_ST_PO_DISABLE);
 
+	/* WaOCLCoherentLineFlush:skl,bxt */
+	I915_WRITE(GEN8_L3SQCREG4, (I915_READ(GEN8_L3SQCREG4) |
+				    GEN8_LQSC_FLUSH_COHERENT_LINES));
+
+	/* WaEnablePreemptionGranularityControlByUMD:skl,bxt */
+	ret= wa_ring_whitelist_reg(ring, GEN8_CS_CHICKEN1);
+	if (ret)
+		return ret;
+
+	/* WaAllowUMDToModifyHDCChicken1:skl,bxt */
+	ret = wa_ring_whitelist_reg(ring, GEN8_HDC_CHICKEN1);
+	if (ret)
+		return ret;
+
 	return 0;
 }
 
@@ -1019,6 +1050,16 @@ static int skl_init_workarounds(struct intel_engine_cs *ring)
 	if (ret)
 		return ret;
 
+	/*
+	 * Actual WA is to disable percontext preemption granularity control
+	 * until D0 which is the default case so this is equivalent to
+	 * !WaDisablePerCtxtPreemptionGranularityControl:skl
+	 */
+	if (IS_SKL_REVID(dev, SKL_REVID_E0, REVID_FOREVER)) {
+		I915_WRITE(GEN7_FF_SLICE_CS_CHICKEN1,
+			   _MASKED_BIT_ENABLE(GEN9_FFSC_PERCTX_PREEMPT_CTRL));
+	}
+
 	if (IS_SKL_REVID(dev, 0, SKL_REVID_D0)) {
 		/* WaDisableChickenBitTSGBarrierAckForFFSliceCS:skl */
 		I915_WRITE(FF_SLICE_CS_CHICKEN2,
@@ -1071,6 +1112,11 @@ static int skl_init_workarounds(struct intel_engine_cs *ring)
 			GEN7_HALF_SLICE_CHICKEN1,
 			GEN7_SBE_SS_CACHE_DISPATCH_PORT_SHARING_DISABLE);
 
+	/* WaDisableLSQCROPERFforOCL:skl */
+	ret = wa_ring_whitelist_reg(ring, GEN8_L3SQCREG4);
+	if (ret)
+		return ret;
+
 	return skl_tune_iz_hashing(ring);
 }
 
@@ -1106,6 +1152,20 @@ static int bxt_init_workarounds(struct intel_engine_cs *ring)
 			GEN7_SBE_SS_CACHE_DISPATCH_PORT_SHARING_DISABLE);
 	}
 
+	/* WaDisableObjectLevelPreemptionForTrifanOrPolygon:bxt */
+	/* WaDisableObjectLevelPreemptionForInstancedDraw:bxt */
+	/* WaDisableObjectLevelPreemtionForInstanceId:bxt */
+	/* WaDisableLSQCROPERFforOCL:bxt */
+	if (IS_BXT_REVID(dev, 0, BXT_REVID_A1)) {
+		ret = wa_ring_whitelist_reg(ring, GEN9_CS_DEBUG_MODE1);
+		if (ret)
+			return ret;
+
+		ret = wa_ring_whitelist_reg(ring, GEN8_L3SQCREG4);
+		if (ret)
+			return ret;
+	}
+
 	return 0;
 }
 
@@ -1117,6 +1177,7 @@ int init_workarounds_ring(struct intel_engine_cs *ring)
 	WARN_ON(ring->id != RCS);
 
 	dev_priv->workarounds.count = 0;
+	dev_priv->workarounds.hw_whitelist_count[RCS] = 0;
 
 	if (IS_BROADWELL(dev))
 		return bdw_init_workarounds(ring);
@@ -2058,6 +2119,9 @@ int intel_pin_and_map_ringbuffer_obj(struct drm_device *dev,
 			return ret;
 		}
 
+		/* Access through the GTT requires the device to be awake. */
+		assert_rpm_wakelock_held(dev_priv);
+
 		ringbuf->virtual_start = ioremap_wc(dev_priv->gtt.mappable_base +
 						    i915_gem_obj_ggtt_offset(obj), ringbuf->size);
 		if (ringbuf->virtual_start == NULL) {
diff --git a/drivers/gpu/drm/i915/intel_ringbuffer.h b/drivers/gpu/drm/i915/intel_ringbuffer.h
index b12f2aabd104..566b0ae10ce0 100644
--- a/drivers/gpu/drm/i915/intel_ringbuffer.h
+++ b/drivers/gpu/drm/i915/intel_ringbuffer.h
@@ -158,6 +158,7 @@ struct  intel_engine_cs {
 #define I915_NUM_RINGS 5
 #define _VCS(n) (VCS + (n))
 	unsigned int exec_id;
+	unsigned int guc_id;
 	u32		mmio_base;
 	struct		drm_device *dev;
 	struct intel_ringbuffer *buffer;
diff --git a/drivers/gpu/drm/i915/intel_runtime_pm.c b/drivers/gpu/drm/i915/intel_runtime_pm.c
index bbca527184d0..6e54d978d9d4 100644
--- a/drivers/gpu/drm/i915/intel_runtime_pm.c
+++ b/drivers/gpu/drm/i915/intel_runtime_pm.c
@@ -284,6 +284,13 @@ static void hsw_power_well_post_enable(struct drm_i915_private *dev_priv)
 						1 << PIPE_C | 1 << PIPE_B);
 }
 
+static void hsw_power_well_pre_disable(struct drm_i915_private *dev_priv)
+{
+	if (IS_BROADWELL(dev_priv))
+		gen8_irq_power_well_pre_disable(dev_priv,
+						1 << PIPE_C | 1 << PIPE_B);
+}
+
 static void skl_power_well_post_enable(struct drm_i915_private *dev_priv,
 				       struct i915_power_well *power_well)
 {
@@ -309,6 +316,14 @@ static void skl_power_well_post_enable(struct drm_i915_private *dev_priv,
 	}
 }
 
+static void skl_power_well_pre_disable(struct drm_i915_private *dev_priv,
+				       struct i915_power_well *power_well)
+{
+	if (power_well->data == SKL_DISP_PW_2)
+		gen8_irq_power_well_pre_disable(dev_priv,
+						1 << PIPE_C | 1 << PIPE_B);
+}
+
 static void hsw_set_power_well(struct drm_i915_private *dev_priv,
 			       struct i915_power_well *power_well, bool enable)
 {
@@ -334,6 +349,7 @@ static void hsw_set_power_well(struct drm_i915_private *dev_priv,
 
 	} else {
 		if (enable_requested) {
+			hsw_power_well_pre_disable(dev_priv);
 			I915_WRITE(HSW_PWR_WELL_DRIVER, 0);
 			POSTING_READ(HSW_PWR_WELL_DRIVER);
 			DRM_DEBUG_KMS("Requesting to disable the power well\n");
@@ -456,20 +472,61 @@ static void assert_can_disable_dc9(struct drm_i915_private *dev_priv)
 	  */
 }
 
-static void gen9_set_dc_state_debugmask_memory_up(
-			struct drm_i915_private *dev_priv)
+static void gen9_set_dc_state_debugmask(struct drm_i915_private *dev_priv)
 {
-	uint32_t val;
+	uint32_t val, mask;
+
+	mask = DC_STATE_DEBUG_MASK_MEMORY_UP;
+
+	if (IS_BROXTON(dev_priv))
+		mask |= DC_STATE_DEBUG_MASK_CORES;
 
 	/* The below bit doesn't need to be cleared ever afterwards */
 	val = I915_READ(DC_STATE_DEBUG);
-	if (!(val & DC_STATE_DEBUG_MASK_MEMORY_UP)) {
-		val |= DC_STATE_DEBUG_MASK_MEMORY_UP;
+	if ((val & mask) != mask) {
+		val |= mask;
 		I915_WRITE(DC_STATE_DEBUG, val);
 		POSTING_READ(DC_STATE_DEBUG);
 	}
 }
 
+static void gen9_write_dc_state(struct drm_i915_private *dev_priv,
+				u32 state)
+{
+	int rewrites = 0;
+	int rereads = 0;
+	u32 v;
+
+	I915_WRITE(DC_STATE_EN, state);
+
+	/* It has been observed that disabling the dc6 state sometimes
+	 * doesn't stick and dmc keeps returning old value. Make sure
+	 * the write really sticks enough times and also force rewrite until
+	 * we are confident that state is exactly what we want.
+	 */
+	do  {
+		v = I915_READ(DC_STATE_EN);
+
+		if (v != state) {
+			I915_WRITE(DC_STATE_EN, state);
+			rewrites++;
+			rereads = 0;
+		} else if (rereads++ > 5) {
+			break;
+		}
+
+	} while (rewrites < 100);
+
+	if (v != state)
+		DRM_ERROR("Writing dc state to 0x%x failed, now 0x%x\n",
+			  state, v);
+
+	/* Most of the times we need one retry, avoid spam */
+	if (rewrites > 1)
+		DRM_DEBUG_KMS("Rewrote dc state to 0x%x %d times\n",
+			      state, rewrites);
+}
+
 static void gen9_set_dc_state(struct drm_i915_private *dev_priv, uint32_t state)
 {
 	uint32_t val;
@@ -488,16 +545,21 @@ static void gen9_set_dc_state(struct drm_i915_private *dev_priv, uint32_t state)
 	else if (i915.enable_dc == 1 && state > DC_STATE_EN_UPTO_DC5)
 		state = DC_STATE_EN_UPTO_DC5;
 
-	if (state & DC_STATE_EN_UPTO_DC5_DC6_MASK)
-		gen9_set_dc_state_debugmask_memory_up(dev_priv);
-
 	val = I915_READ(DC_STATE_EN);
 	DRM_DEBUG_KMS("Setting DC state from %02x to %02x\n",
 		      val & mask, state);
+
+	/* Check if DMC is ignoring our DC state requests */
+	if ((val & mask) != dev_priv->csr.dc_state)
+		DRM_ERROR("DC state mismatch (0x%x -> 0x%x)\n",
+			  dev_priv->csr.dc_state, val & mask);
+
 	val &= ~mask;
 	val |= state;
-	I915_WRITE(DC_STATE_EN, val);
-	POSTING_READ(DC_STATE_EN);
+
+	gen9_write_dc_state(dev_priv, val);
+
+	dev_priv->csr.dc_state = val & mask;
 }
 
 void bxt_enable_dc9(struct drm_i915_private *dev_priv)
@@ -663,6 +725,9 @@ static void skl_set_power_well(struct drm_i915_private *dev_priv,
 	state_mask = SKL_POWER_WELL_STATE(power_well->data);
 	is_enabled = tmp & state_mask;
 
+	if (!enable && enable_requested)
+		skl_power_well_pre_disable(dev_priv, power_well);
+
 	if (enable) {
 		if (!enable_requested) {
 			WARN((tmp & state_mask) &&
@@ -941,6 +1006,9 @@ static void vlv_display_power_well_deinit(struct drm_i915_private *dev_priv)
 	valleyview_disable_display_irqs(dev_priv);
 	spin_unlock_irq(&dev_priv->irq_lock);
 
+	/* make sure we're done processing display irqs */
+	synchronize_irq(dev_priv->dev->irq);
+
 	vlv_power_sequencer_reset(dev_priv);
 }
 
@@ -1435,6 +1503,22 @@ static void chv_pipe_power_well_disable(struct drm_i915_private *dev_priv,
 	chv_set_pipe_power_well(dev_priv, power_well, false);
 }
 
+static void
+__intel_display_power_get_domain(struct drm_i915_private *dev_priv,
+				 enum intel_display_power_domain domain)
+{
+	struct i915_power_domains *power_domains = &dev_priv->power_domains;
+	struct i915_power_well *power_well;
+	int i;
+
+	for_each_power_well(i, power_well, BIT(domain), power_domains) {
+		if (!power_well->count++)
+			intel_power_well_enable(dev_priv, power_well);
+	}
+
+	power_domains->domain_use_count[domain]++;
+}
+
 /**
  * intel_display_power_get - grab a power domain reference
  * @dev_priv: i915 device instance
@@ -1450,24 +1534,53 @@ static void chv_pipe_power_well_disable(struct drm_i915_private *dev_priv,
 void intel_display_power_get(struct drm_i915_private *dev_priv,
 			     enum intel_display_power_domain domain)
 {
-	struct i915_power_domains *power_domains;
-	struct i915_power_well *power_well;
-	int i;
+	struct i915_power_domains *power_domains = &dev_priv->power_domains;
 
 	intel_runtime_pm_get(dev_priv);
 
-	power_domains = &dev_priv->power_domains;
+	mutex_lock(&power_domains->lock);
+
+	__intel_display_power_get_domain(dev_priv, domain);
+
+	mutex_unlock(&power_domains->lock);
+}
+
+/**
+ * intel_display_power_get_if_enabled - grab a reference for an enabled display power domain
+ * @dev_priv: i915 device instance
+ * @domain: power domain to reference
+ *
+ * This function grabs a power domain reference for @domain and ensures that the
+ * power domain and all its parents are powered up. Therefore users should only
+ * grab a reference to the innermost power domain they need.
+ *
+ * Any power domain reference obtained by this function must have a symmetric
+ * call to intel_display_power_put() to release the reference again.
+ */
+bool intel_display_power_get_if_enabled(struct drm_i915_private *dev_priv,
+					enum intel_display_power_domain domain)
+{
+	struct i915_power_domains *power_domains = &dev_priv->power_domains;
+	bool is_enabled;
+
+	if (!intel_runtime_pm_get_if_in_use(dev_priv))
+		return false;
 
 	mutex_lock(&power_domains->lock);
 
-	for_each_power_well(i, power_well, BIT(domain), power_domains) {
-		if (!power_well->count++)
-			intel_power_well_enable(dev_priv, power_well);
+	if (__intel_display_power_is_enabled(dev_priv, domain)) {
+		__intel_display_power_get_domain(dev_priv, domain);
+		is_enabled = true;
+	} else {
+		is_enabled = false;
 	}
 
-	power_domains->domain_use_count[domain]++;
-
 	mutex_unlock(&power_domains->lock);
+
+	if (!is_enabled)
+		intel_runtime_pm_put(dev_priv);
+
+	return is_enabled;
 }
 
 /**
@@ -2028,8 +2141,8 @@ static void skl_display_core_init(struct drm_i915_private *dev_priv,
 
 	skl_init_cdclk(dev_priv);
 
-	if (dev_priv->csr.dmc_payload)
-		intel_csr_load_program(dev_priv);
+	if (dev_priv->csr.dmc_payload && intel_csr_load_program(dev_priv))
+		gen9_set_dc_state_debugmask(dev_priv);
 }
 
 static void skl_display_core_uninit(struct drm_i915_private *dev_priv)
@@ -2206,15 +2319,15 @@ void intel_power_domains_init_hw(struct drm_i915_private *dev_priv, bool resume)
  */
 void intel_power_domains_suspend(struct drm_i915_private *dev_priv)
 {
-	if (IS_SKYLAKE(dev_priv) || IS_KABYLAKE(dev_priv))
-		skl_display_core_uninit(dev_priv);
-
 	/*
 	 * Even if power well support was disabled we still want to disable
 	 * power wells while we are system suspended.
 	 */
 	if (!i915.disable_power_well)
 		intel_display_power_put(dev_priv, POWER_DOMAIN_INIT);
+
+	if (IS_SKYLAKE(dev_priv) || IS_KABYLAKE(dev_priv))
+		skl_display_core_uninit(dev_priv);
 }
 
 /**
@@ -2239,6 +2352,41 @@ void intel_runtime_pm_get(struct drm_i915_private *dev_priv)
 }
 
 /**
+ * intel_runtime_pm_get_if_in_use - grab a runtime pm reference if device in use
+ * @dev_priv: i915 device instance
+ *
+ * This function grabs a device-level runtime pm reference if the device is
+ * already in use and ensures that it is powered up.
+ *
+ * Any runtime pm reference obtained by this function must have a symmetric
+ * call to intel_runtime_pm_put() to release the reference again.
+ */
+bool intel_runtime_pm_get_if_in_use(struct drm_i915_private *dev_priv)
+{
+	struct drm_device *dev = dev_priv->dev;
+	struct device *device = &dev->pdev->dev;
+
+	if (IS_ENABLED(CONFIG_PM)) {
+		int ret = pm_runtime_get_if_in_use(device);
+
+		/*
+		 * In cases runtime PM is disabled by the RPM core and we get
+		 * an -EINVAL return value we are not supposed to call this
+		 * function, since the power state is undefined. This applies
+		 * atm to the late/early system suspend/resume handlers.
+		 */
+		WARN_ON_ONCE(ret < 0);
+		if (ret <= 0)
+			return false;
+	}
+
+	atomic_inc(&dev_priv->pm.wakeref_count);
+	assert_rpm_wakelock_held(dev_priv);
+
+	return true;
+}
+
+/**
  * intel_runtime_pm_get_noresume - grab a runtime pm reference
  * @dev_priv: i915 device instance
  *
diff --git a/drivers/gpu/drm/i915/intel_sdvo.c b/drivers/gpu/drm/i915/intel_sdvo.c
index 2e1da060b0e1..4ecc076c4041 100644
--- a/drivers/gpu/drm/i915/intel_sdvo.c
+++ b/drivers/gpu/drm/i915/intel_sdvo.c
@@ -1527,6 +1527,7 @@ intel_sdvo_mode_valid(struct drm_connector *connector,
 		      struct drm_display_mode *mode)
 {
 	struct intel_sdvo *intel_sdvo = intel_attached_sdvo(connector);
+	int max_dotclk = to_i915(connector->dev)->max_dotclk_freq;
 
 	if (mode->flags & DRM_MODE_FLAG_DBLSCAN)
 		return MODE_NO_DBLESCAN;
@@ -1537,6 +1538,9 @@ intel_sdvo_mode_valid(struct drm_connector *connector,
 	if (intel_sdvo->pixel_clock_max < mode->clock)
 		return MODE_CLOCK_HIGH;
 
+	if (mode->clock > max_dotclk)
+		return MODE_CLOCK_HIGH;
+
 	if (intel_sdvo->is_lvds) {
 		if (mode->hdisplay > intel_sdvo->sdvo_lvds_fixed_mode->hdisplay)
 			return MODE_PANEL;
diff --git a/drivers/gpu/drm/i915/intel_sideband.c b/drivers/gpu/drm/i915/intel_sideband.c
index 8831fc579ade..c3998188cf35 100644
--- a/drivers/gpu/drm/i915/intel_sideband.c
+++ b/drivers/gpu/drm/i915/intel_sideband.c
@@ -129,17 +129,18 @@ u32 vlv_nc_read(struct drm_i915_private *dev_priv, u8 addr)
 	return val;
 }
 
-u32 vlv_gpio_nc_read(struct drm_i915_private *dev_priv, u32 reg)
+u32 vlv_iosf_sb_read(struct drm_i915_private *dev_priv, u8 port, u32 reg)
 {
 	u32 val = 0;
-	vlv_sideband_rw(dev_priv, PCI_DEVFN(0, 0), IOSF_PORT_GPIO_NC,
+	vlv_sideband_rw(dev_priv, PCI_DEVFN(0, 0), port,
 			SB_CRRDDA_NP, reg, &val);
 	return val;
 }
 
-void vlv_gpio_nc_write(struct drm_i915_private *dev_priv, u32 reg, u32 val)
+void vlv_iosf_sb_write(struct drm_i915_private *dev_priv,
+		       u8 port, u32 reg, u32 val)
 {
-	vlv_sideband_rw(dev_priv, PCI_DEVFN(0, 0), IOSF_PORT_GPIO_NC,
+	vlv_sideband_rw(dev_priv, PCI_DEVFN(0, 0), port,
 			SB_CRWRDA_NP, reg, &val);
 }
 
@@ -171,20 +172,6 @@ void vlv_ccu_write(struct drm_i915_private *dev_priv, u32 reg, u32 val)
 			SB_CRWRDA_NP, reg, &val);
 }
 
-u32 vlv_gps_core_read(struct drm_i915_private *dev_priv, u32 reg)
-{
-	u32 val = 0;
-	vlv_sideband_rw(dev_priv, PCI_DEVFN(0, 0), IOSF_PORT_GPS_CORE,
-			SB_CRRDDA_NP, reg, &val);
-	return val;
-}
-
-void vlv_gps_core_write(struct drm_i915_private *dev_priv, u32 reg, u32 val)
-{
-	vlv_sideband_rw(dev_priv, PCI_DEVFN(0, 0), IOSF_PORT_GPS_CORE,
-			SB_CRWRDA_NP, reg, &val);
-}
-
 u32 vlv_dpio_read(struct drm_i915_private *dev_priv, enum pipe pipe, int reg)
 {
 	u32 val = 0;
diff --git a/drivers/gpu/drm/i915/intel_sprite.c b/drivers/gpu/drm/i915/intel_sprite.c
index 0875c8e0ec0a..a2582c455b36 100644
--- a/drivers/gpu/drm/i915/intel_sprite.c
+++ b/drivers/gpu/drm/i915/intel_sprite.c
@@ -350,8 +350,8 @@ vlv_update_plane(struct drm_plane *dplane,
 	int pipe = intel_plane->pipe;
 	int plane = intel_plane->plane;
 	u32 sprctl;
-	unsigned long sprsurf_offset, linear_offset;
-	int pixel_size = drm_format_plane_cpp(fb->pixel_format, 0);
+	u32 sprsurf_offset, linear_offset;
+	int cpp = drm_format_plane_cpp(fb->pixel_format, 0);
 	const struct drm_intel_sprite_colorkey *key = &plane_state->ckey;
 	int crtc_x = plane_state->dst.x1;
 	int crtc_y = plane_state->dst.y1;
@@ -422,10 +422,9 @@ vlv_update_plane(struct drm_plane *dplane,
 	crtc_w--;
 	crtc_h--;
 
-	linear_offset = y * fb->pitches[0] + x * pixel_size;
+	linear_offset = y * fb->pitches[0] + x * cpp;
 	sprsurf_offset = intel_compute_tile_offset(dev_priv, &x, &y,
-						   fb->modifier[0],
-						   pixel_size,
+						   fb->modifier[0], cpp,
 						   fb->pitches[0]);
 	linear_offset -= sprsurf_offset;
 
@@ -434,7 +433,7 @@ vlv_update_plane(struct drm_plane *dplane,
 
 		x += src_w;
 		y += src_h;
-		linear_offset += src_h * fb->pitches[0] + src_w * pixel_size;
+		linear_offset += src_h * fb->pitches[0] + src_w * cpp;
 	}
 
 	if (key->flags) {
@@ -493,8 +492,8 @@ ivb_update_plane(struct drm_plane *plane,
 	struct drm_i915_gem_object *obj = intel_fb_obj(fb);
 	enum pipe pipe = intel_plane->pipe;
 	u32 sprctl, sprscale = 0;
-	unsigned long sprsurf_offset, linear_offset;
-	int pixel_size = drm_format_plane_cpp(fb->pixel_format, 0);
+	u32 sprsurf_offset, linear_offset;
+	int cpp = drm_format_plane_cpp(fb->pixel_format, 0);
 	const struct drm_intel_sprite_colorkey *key = &plane_state->ckey;
 	int crtc_x = plane_state->dst.x1;
 	int crtc_y = plane_state->dst.y1;
@@ -556,10 +555,9 @@ ivb_update_plane(struct drm_plane *plane,
 	if (crtc_w != src_w || crtc_h != src_h)
 		sprscale = SPRITE_SCALE_ENABLE | (src_w << 16) | src_h;
 
-	linear_offset = y * fb->pitches[0] + x * pixel_size;
+	linear_offset = y * fb->pitches[0] + x * cpp;
 	sprsurf_offset = intel_compute_tile_offset(dev_priv, &x, &y,
-						   fb->modifier[0],
-						   pixel_size,
+						   fb->modifier[0], cpp,
 						   fb->pitches[0]);
 	linear_offset -= sprsurf_offset;
 
@@ -570,8 +568,7 @@ ivb_update_plane(struct drm_plane *plane,
 		if (!IS_HASWELL(dev) && !IS_BROADWELL(dev)) {
 			x += src_w;
 			y += src_h;
-			linear_offset += src_h * fb->pitches[0] +
-				src_w * pixel_size;
+			linear_offset += src_h * fb->pitches[0] + src_w * cpp;
 		}
 	}
 
@@ -635,9 +632,9 @@ ilk_update_plane(struct drm_plane *plane,
 	struct drm_framebuffer *fb = plane_state->base.fb;
 	struct drm_i915_gem_object *obj = intel_fb_obj(fb);
 	int pipe = intel_plane->pipe;
-	unsigned long dvssurf_offset, linear_offset;
 	u32 dvscntr, dvsscale;
-	int pixel_size = drm_format_plane_cpp(fb->pixel_format, 0);
+	u32 dvssurf_offset, linear_offset;
+	int cpp = drm_format_plane_cpp(fb->pixel_format, 0);
 	const struct drm_intel_sprite_colorkey *key = &plane_state->ckey;
 	int crtc_x = plane_state->dst.x1;
 	int crtc_y = plane_state->dst.y1;
@@ -695,10 +692,9 @@ ilk_update_plane(struct drm_plane *plane,
 	if (crtc_w != src_w || crtc_h != src_h)
 		dvsscale = DVS_SCALE_ENABLE | (src_w << 16) | src_h;
 
-	linear_offset = y * fb->pitches[0] + x * pixel_size;
+	linear_offset = y * fb->pitches[0] + x * cpp;
 	dvssurf_offset = intel_compute_tile_offset(dev_priv, &x, &y,
-						   fb->modifier[0],
-						   pixel_size,
+						   fb->modifier[0], cpp,
 						   fb->pitches[0]);
 	linear_offset -= dvssurf_offset;
 
@@ -707,7 +703,7 @@ ilk_update_plane(struct drm_plane *plane,
 
 		x += src_w;
 		y += src_h;
-		linear_offset += src_h * fb->pitches[0] + src_w * pixel_size;
+		linear_offset += src_h * fb->pitches[0] + src_w * cpp;
 	}
 
 	if (key->flags) {
@@ -772,7 +768,6 @@ intel_check_sprite_plane(struct drm_plane *plane,
 	int hscale, vscale;
 	int max_scale, min_scale;
 	bool can_scale;
-	int pixel_size;
 
 	if (!fb) {
 		state->visible = false;
@@ -894,6 +889,7 @@ intel_check_sprite_plane(struct drm_plane *plane,
 	/* Check size restrictions when scaling */
 	if (state->visible && (src_w != crtc_w || src_h != crtc_h)) {
 		unsigned int width_bytes;
+		int cpp = drm_format_plane_cpp(fb->pixel_format, 0);
 
 		WARN_ON(!can_scale);
 
@@ -905,9 +901,7 @@ intel_check_sprite_plane(struct drm_plane *plane,
 		if (src_w < 3 || src_h < 3)
 			state->visible = false;
 
-		pixel_size = drm_format_plane_cpp(fb->pixel_format, 0);
-		width_bytes = ((src_x * pixel_size) & 63) +
-					src_w * pixel_size;
+		width_bytes = ((src_x * cpp) & 63) + src_w * cpp;
 
 		if (INTEL_INFO(dev)->gen < 9 && (src_w > 2048 || src_h > 2048 ||
 		    width_bytes > 4096 || fb->pitches[0] > 4096)) {
diff --git a/drivers/gpu/drm/i915/intel_tv.c b/drivers/gpu/drm/i915/intel_tv.c
index 948cbff6c62e..6745bad5bff0 100644
--- a/drivers/gpu/drm/i915/intel_tv.c
+++ b/drivers/gpu/drm/i915/intel_tv.c
@@ -897,6 +897,10 @@ intel_tv_mode_valid(struct drm_connector *connector,
 {
 	struct intel_tv *intel_tv = intel_attached_tv(connector);
 	const struct tv_mode *tv_mode = intel_tv_mode_find(intel_tv);
+	int max_dotclk = to_i915(connector->dev)->max_dotclk_freq;
+
+	if (mode->clock > max_dotclk)
+		return MODE_CLOCK_HIGH;
 
 	/* Ensure TV refresh is close to desired refresh */
 	if (tv_mode && abs(tv_mode->refresh - drm_mode_vrefresh(mode) * 1000)
@@ -1178,10 +1182,9 @@ static int
 intel_tv_detect_type(struct intel_tv *intel_tv,
 		      struct drm_connector *connector)
 {
-	struct drm_encoder *encoder = &intel_tv->base.base;
-	struct drm_crtc *crtc = encoder->crtc;
+	struct drm_crtc *crtc = connector->state->crtc;
 	struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
-	struct drm_device *dev = encoder->dev;
+	struct drm_device *dev = connector->dev;
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	u32 tv_ctl, save_tv_ctl;
 	u32 tv_dac, save_tv_dac;
@@ -1230,8 +1233,7 @@ intel_tv_detect_type(struct intel_tv *intel_tv,
 	I915_WRITE(TV_DAC, tv_dac);
 	POSTING_READ(TV_DAC);
 
-	intel_wait_for_vblank(intel_tv->base.base.dev,
-			      to_intel_crtc(intel_tv->base.base.crtc)->pipe);
+	intel_wait_for_vblank(dev, intel_crtc->pipe);
 
 	type = -1;
 	tv_dac = I915_READ(TV_DAC);
@@ -1261,8 +1263,7 @@ intel_tv_detect_type(struct intel_tv *intel_tv,
 	POSTING_READ(TV_CTL);
 
 	/* For unknown reasons the hw barfs if we don't do this vblank wait. */
-	intel_wait_for_vblank(intel_tv->base.base.dev,
-			      to_intel_crtc(intel_tv->base.base.crtc)->pipe);
+	intel_wait_for_vblank(dev, intel_crtc->pipe);
 
 	/* Restore interrupt config */
 	if (connector->polled & DRM_CONNECTOR_POLL_HPD) {
@@ -1420,6 +1421,7 @@ intel_tv_get_modes(struct drm_connector *connector)
 		if (!mode_ptr)
 			continue;
 		strncpy(mode_ptr->name, input->name, DRM_DISPLAY_MODE_LEN);
+		mode_ptr->name[DRM_DISPLAY_MODE_LEN - 1] = '\0';
 
 		mode_ptr->hdisplay = hactive_s;
 		mode_ptr->hsync_start = hactive_s + 1;
diff --git a/drivers/gpu/drm/i915/intel_uncore.c b/drivers/gpu/drm/i915/intel_uncore.c
index bfa79e5c214e..436d8f2b8682 100644
--- a/drivers/gpu/drm/i915/intel_uncore.c
+++ b/drivers/gpu/drm/i915/intel_uncore.c
@@ -400,6 +400,8 @@ void intel_uncore_early_sanitize(struct drm_device *dev, bool restore_forcewake)
 
 void intel_uncore_sanitize(struct drm_device *dev)
 {
+	i915.enable_rc6 = sanitize_rc6_option(dev, i915.enable_rc6);
+
 	/* BIOS often leaves RC6 enabled, but disable it for hw init */
 	intel_disable_gt_powersave(dev);
 }
diff --git a/drivers/gpu/drm/imx/dw_hdmi-imx.c b/drivers/gpu/drm/imx/dw_hdmi-imx.c
index 21d615827837..2a95d10e9d92 100644
--- a/drivers/gpu/drm/imx/dw_hdmi-imx.c
+++ b/drivers/gpu/drm/imx/dw_hdmi-imx.c
@@ -118,7 +118,7 @@ static void dw_hdmi_imx_encoder_mode_set(struct drm_encoder *encoder,
 static void dw_hdmi_imx_encoder_commit(struct drm_encoder *encoder)
 {
 	struct imx_hdmi *hdmi = container_of(encoder, struct imx_hdmi, encoder);
-	int mux = imx_drm_encoder_get_mux_id(hdmi->dev->of_node, encoder);
+	int mux = drm_of_encoder_active_port_id(hdmi->dev->of_node, encoder);
 
 	regmap_update_bits(hdmi->regmap, IOMUXC_GPR3,
 			   IMX6Q_GPR3_HDMI_MUX_CTL_MASK,
diff --git a/drivers/gpu/drm/imx/imx-drm-core.c b/drivers/gpu/drm/imx/imx-drm-core.c
index 7c4d1250e071..9876e0f0c3e1 100644
--- a/drivers/gpu/drm/imx/imx-drm-core.c
+++ b/drivers/gpu/drm/imx/imx-drm-core.c
@@ -17,7 +17,6 @@
 #include <linux/device.h>
 #include <linux/fb.h>
 #include <linux/module.h>
-#include <linux/of_graph.h>
 #include <linux/platform_device.h>
 #include <drm/drmP.h>
 #include <drm/drm_fb_helper.h>
@@ -412,36 +411,6 @@ int imx_drm_encoder_parse_of(struct drm_device *drm,
 }
 EXPORT_SYMBOL_GPL(imx_drm_encoder_parse_of);
 
-/*
- * @node: device tree node containing encoder input ports
- * @encoder: drm_encoder
- */
-int imx_drm_encoder_get_mux_id(struct device_node *node,
-			       struct drm_encoder *encoder)
-{
-	struct imx_drm_crtc *imx_crtc = imx_drm_find_crtc(encoder->crtc);
-	struct device_node *ep;
-	struct of_endpoint endpoint;
-	struct device_node *port;
-	int ret;
-
-	if (!node || !imx_crtc)
-		return -EINVAL;
-
-	for_each_endpoint_of_node(node, ep) {
-		port = of_graph_get_remote_port(ep);
-		of_node_put(port);
-		if (port == imx_crtc->crtc->port) {
-			ret = of_graph_parse_endpoint(ep, &endpoint);
-			of_node_put(ep);
-			return ret ? ret : endpoint.port;
-		}
-	}
-
-	return -EINVAL;
-}
-EXPORT_SYMBOL_GPL(imx_drm_encoder_get_mux_id);
-
 static const struct drm_ioctl_desc imx_drm_ioctls[] = {
 	/* none so far */
 };
diff --git a/drivers/gpu/drm/imx/imx-drm.h b/drivers/gpu/drm/imx/imx-drm.h
index 71cf6d9c714f..b0241b9d1334 100644
--- a/drivers/gpu/drm/imx/imx-drm.h
+++ b/drivers/gpu/drm/imx/imx-drm.h
@@ -46,8 +46,6 @@ int imx_drm_set_bus_format_pins(struct drm_encoder *encoder,
 int imx_drm_set_bus_format(struct drm_encoder *encoder,
 		u32 bus_format);
 
-int imx_drm_encoder_get_mux_id(struct device_node *node,
-		struct drm_encoder *encoder);
 int imx_drm_encoder_parse_of(struct drm_device *drm,
 	struct drm_encoder *encoder, struct device_node *np);
 
diff --git a/drivers/gpu/drm/imx/imx-ldb.c b/drivers/gpu/drm/imx/imx-ldb.c
index 024d6134b848..a58eee59550a 100644
--- a/drivers/gpu/drm/imx/imx-ldb.c
+++ b/drivers/gpu/drm/imx/imx-ldb.c
@@ -19,6 +19,7 @@
 #include <drm/drmP.h>
 #include <drm/drm_fb_helper.h>
 #include <drm/drm_crtc_helper.h>
+#include <drm/drm_of.h>
 #include <drm/drm_panel.h>
 #include <linux/mfd/syscon.h>
 #include <linux/mfd/syscon/imx6q-iomuxc-gpr.h>
@@ -208,7 +209,7 @@ static void imx_ldb_encoder_commit(struct drm_encoder *encoder)
 	struct imx_ldb_channel *imx_ldb_ch = enc_to_imx_ldb_ch(encoder);
 	struct imx_ldb *ldb = imx_ldb_ch->ldb;
 	int dual = ldb->ldb_ctrl & LDB_SPLIT_MODE_EN;
-	int mux = imx_drm_encoder_get_mux_id(imx_ldb_ch->child, encoder);
+	int mux = drm_of_encoder_active_port_id(imx_ldb_ch->child, encoder);
 
 	drm_panel_prepare(imx_ldb_ch->panel);
 
@@ -258,7 +259,7 @@ static void imx_ldb_encoder_mode_set(struct drm_encoder *encoder,
 	int dual = ldb->ldb_ctrl & LDB_SPLIT_MODE_EN;
 	unsigned long serial_clk;
 	unsigned long di_clk = mode->clock * 1000;
-	int mux = imx_drm_encoder_get_mux_id(imx_ldb_ch->child, encoder);
+	int mux = drm_of_encoder_active_port_id(imx_ldb_ch->child, encoder);
 
 	if (mode->clock > 170000) {
 		dev_warn(ldb->dev,
diff --git a/drivers/gpu/drm/imx/ipuv3-crtc.c b/drivers/gpu/drm/imx/ipuv3-crtc.c
index 846b5f558897..dee8e8b3523b 100644
--- a/drivers/gpu/drm/imx/ipuv3-crtc.c
+++ b/drivers/gpu/drm/imx/ipuv3-crtc.c
@@ -22,6 +22,8 @@
 #include <linux/fb.h>
 #include <linux/clk.h>
 #include <linux/errno.h>
+#include <linux/reservation.h>
+#include <linux/dma-buf.h>
 #include <drm/drm_gem_cma_helper.h>
 #include <drm/drm_fb_cma_helper.h>
 
@@ -31,6 +33,23 @@
 
 #define DRIVER_DESC		"i.MX IPUv3 Graphics"
 
+enum ipu_flip_status {
+	IPU_FLIP_NONE,
+	IPU_FLIP_PENDING,
+	IPU_FLIP_SUBMITTED,
+};
+
+struct ipu_flip_work {
+	struct work_struct		unref_work;
+	struct drm_gem_object		*bo;
+	struct drm_pending_vblank_event *page_flip_event;
+	struct work_struct		fence_work;
+	struct ipu_crtc			*crtc;
+	struct fence			*excl;
+	unsigned			shared_count;
+	struct fence			**shared;
+};
+
 struct ipu_crtc {
 	struct device		*dev;
 	struct drm_crtc		base;
@@ -42,8 +61,9 @@ struct ipu_crtc {
 	struct ipu_dc		*dc;
 	struct ipu_di		*di;
 	int			enabled;
-	struct drm_pending_vblank_event *page_flip_event;
-	struct drm_framebuffer	*newfb;
+	enum ipu_flip_status	flip_state;
+	struct workqueue_struct *flip_queue;
+	struct ipu_flip_work	*flip_work;
 	int			irq;
 	u32			bus_format;
 	int			di_hsync_pin;
@@ -64,6 +84,7 @@ static void ipu_fb_enable(struct ipu_crtc *ipu_crtc)
 	/* Start DC channel and DI after IDMAC */
 	ipu_dc_enable_channel(ipu_crtc->dc);
 	ipu_di_enable(ipu_crtc->di);
+	drm_crtc_vblank_on(&ipu_crtc->base);
 
 	ipu_crtc->enabled = 1;
 }
@@ -80,6 +101,7 @@ static void ipu_fb_disable(struct ipu_crtc *ipu_crtc)
 	ipu_di_disable(ipu_crtc->di);
 	ipu_plane_disable(ipu_crtc->plane[0]);
 	ipu_dc_disable(ipu);
+	drm_crtc_vblank_off(&ipu_crtc->base);
 
 	ipu_crtc->enabled = 0;
 }
@@ -102,15 +124,45 @@ static void ipu_crtc_dpms(struct drm_crtc *crtc, int mode)
 	}
 }
 
+static void ipu_flip_unref_work_func(struct work_struct *__work)
+{
+	struct ipu_flip_work *work =
+			container_of(__work, struct ipu_flip_work, unref_work);
+
+	drm_gem_object_unreference_unlocked(work->bo);
+	kfree(work);
+}
+
+static void ipu_flip_fence_work_func(struct work_struct *__work)
+{
+	struct ipu_flip_work *work =
+			container_of(__work, struct ipu_flip_work, fence_work);
+	int i;
+
+	/* wait for all fences attached to the FB obj to signal */
+	if (work->excl) {
+		fence_wait(work->excl, false);
+		fence_put(work->excl);
+	}
+	for (i = 0; i < work->shared_count; i++) {
+		fence_wait(work->shared[i], false);
+		fence_put(work->shared[i]);
+	}
+
+	work->crtc->flip_state = IPU_FLIP_SUBMITTED;
+}
+
 static int ipu_page_flip(struct drm_crtc *crtc,
 		struct drm_framebuffer *fb,
 		struct drm_pending_vblank_event *event,
 		uint32_t page_flip_flags)
 {
+	struct drm_gem_cma_object *cma_obj = drm_fb_cma_get_gem_obj(fb, 0);
 	struct ipu_crtc *ipu_crtc = to_ipu_crtc(crtc);
+	struct ipu_flip_work *flip_work;
 	int ret;
 
-	if (ipu_crtc->newfb)
+	if (ipu_crtc->flip_state != IPU_FLIP_NONE)
 		return -EBUSY;
 
 	ret = imx_drm_crtc_vblank_get(ipu_crtc->imx_crtc);
@@ -121,11 +173,58 @@ static int ipu_page_flip(struct drm_crtc *crtc,
 		return ret;
 	}
 
-	ipu_crtc->newfb = fb;
-	ipu_crtc->page_flip_event = event;
-	crtc->primary->fb = fb;
+	flip_work = kzalloc(sizeof *flip_work, GFP_KERNEL);
+	if (!flip_work) {
+		ret = -ENOMEM;
+		goto put_vblank;
+	}
+	INIT_WORK(&flip_work->unref_work, ipu_flip_unref_work_func);
+	flip_work->page_flip_event = event;
+
+	/* get BO backing the old framebuffer and take a reference */
+	flip_work->bo = &drm_fb_cma_get_gem_obj(crtc->primary->fb, 0)->base;
+	drm_gem_object_reference(flip_work->bo);
+
+	ipu_crtc->flip_work = flip_work;
+	/*
+	 * If the object has a DMABUF attached, we need to wait on its fences
+	 * if there are any.
+	 */
+	if (cma_obj->base.dma_buf) {
+		INIT_WORK(&flip_work->fence_work, ipu_flip_fence_work_func);
+		flip_work->crtc = ipu_crtc;
+
+		ret = reservation_object_get_fences_rcu(
+				cma_obj->base.dma_buf->resv, &flip_work->excl,
+				&flip_work->shared_count, &flip_work->shared);
+
+		if (unlikely(ret)) {
+			DRM_ERROR("failed to get fences for buffer\n");
+			goto free_flip_work;
+		}
+
+		/* No need to queue the worker if the are no fences */
+		if (!flip_work->excl && !flip_work->shared_count) {
+			ipu_crtc->flip_state = IPU_FLIP_SUBMITTED;
+		} else {
+			ipu_crtc->flip_state = IPU_FLIP_PENDING;
+			queue_work(ipu_crtc->flip_queue,
+				   &flip_work->fence_work);
+		}
+	} else {
+		ipu_crtc->flip_state = IPU_FLIP_SUBMITTED;
+	}
 
 	return 0;
+
+free_flip_work:
+	drm_gem_object_unreference_unlocked(flip_work->bo);
+	kfree(flip_work);
+	ipu_crtc->flip_work = NULL;
+put_vblank:
+	imx_drm_crtc_vblank_put(ipu_crtc->imx_crtc);
+
+	return ret;
 }
 
 static const struct drm_crtc_funcs ipu_crtc_funcs = {
@@ -209,12 +308,12 @@ static void ipu_crtc_handle_pageflip(struct ipu_crtc *ipu_crtc)
 {
 	unsigned long flags;
 	struct drm_device *drm = ipu_crtc->base.dev;
+	struct ipu_flip_work *work = ipu_crtc->flip_work;
 
 	spin_lock_irqsave(&drm->event_lock, flags);
-	if (ipu_crtc->page_flip_event)
+	if (work->page_flip_event)
 		drm_crtc_send_vblank_event(&ipu_crtc->base,
-					   ipu_crtc->page_flip_event);
-	ipu_crtc->page_flip_event = NULL;
+					   work->page_flip_event);
 	imx_drm_crtc_vblank_put(ipu_crtc->imx_crtc);
 	spin_unlock_irqrestore(&drm->event_lock, flags);
 }
@@ -225,13 +324,15 @@ static irqreturn_t ipu_irq_handler(int irq, void *dev_id)
 
 	imx_drm_handle_vblank(ipu_crtc->imx_crtc);
 
-	if (ipu_crtc->newfb) {
+	if (ipu_crtc->flip_state == IPU_FLIP_SUBMITTED) {
 		struct ipu_plane *plane = ipu_crtc->plane[0];
 
-		ipu_crtc->newfb = NULL;
 		ipu_plane_set_base(plane, ipu_crtc->base.primary->fb,
 				   plane->x, plane->y);
 		ipu_crtc_handle_pageflip(ipu_crtc);
+		queue_work(ipu_crtc->flip_queue,
+			   &ipu_crtc->flip_work->unref_work);
+		ipu_crtc->flip_state = IPU_FLIP_NONE;
 	}
 
 	return IRQ_HANDLED;
@@ -280,11 +381,18 @@ static const struct drm_crtc_helper_funcs ipu_helper_funcs = {
 
 static int ipu_enable_vblank(struct drm_crtc *crtc)
 {
+	struct ipu_crtc *ipu_crtc = to_ipu_crtc(crtc);
+
+	enable_irq(ipu_crtc->irq);
+
 	return 0;
 }
 
 static void ipu_disable_vblank(struct drm_crtc *crtc)
 {
+	struct ipu_crtc *ipu_crtc = to_ipu_crtc(crtc);
+
+	disable_irq_nosync(ipu_crtc->irq);
 }
 
 static int ipu_set_interface_pix_fmt(struct drm_crtc *crtc,
@@ -395,6 +503,10 @@ static int ipu_crtc_init(struct ipu_crtc *ipu_crtc,
 		dev_err(ipu_crtc->dev, "irq request failed with %d.\n", ret);
 		goto err_put_plane_res;
 	}
+	/* Only enable IRQ when we actually need it to trigger work. */
+	disable_irq(ipu_crtc->irq);
+
+	ipu_crtc->flip_queue = create_singlethread_workqueue("ipu-crtc-flip");
 
 	return 0;
 
@@ -437,6 +549,7 @@ static void ipu_drm_unbind(struct device *dev, struct device *master,
 
 	imx_drm_remove_crtc(ipu_crtc->imx_crtc);
 
+	destroy_workqueue(ipu_crtc->flip_queue);
 	ipu_plane_put_resources(ipu_crtc->plane[0]);
 	ipu_put_resources(ipu_crtc);
 }
diff --git a/drivers/gpu/drm/imx/ipuv3-plane.c b/drivers/gpu/drm/imx/ipuv3-plane.c
index 591ba2f1ae03..588827844f30 100644
--- a/drivers/gpu/drm/imx/ipuv3-plane.c
+++ b/drivers/gpu/drm/imx/ipuv3-plane.c
@@ -42,6 +42,7 @@ static const uint32_t ipu_plane_formats[] = {
 	DRM_FORMAT_YVYU,
 	DRM_FORMAT_YUV420,
 	DRM_FORMAT_YVU420,
+	DRM_FORMAT_RGB565,
 };
 
 int ipu_plane_irq(struct ipu_plane *ipu_plane)
@@ -338,7 +339,7 @@ static int ipu_update_plane(struct drm_plane *plane, struct drm_crtc *crtc,
 	}
 
 	if (crtc != plane->crtc)
-		dev_info(plane->dev->dev, "crtc change: %p -> %p\n",
+		dev_dbg(plane->dev->dev, "crtc change: %p -> %p\n",
 				plane->crtc, crtc);
 	plane->crtc = crtc;
 
diff --git a/drivers/gpu/drm/msm/Makefile b/drivers/gpu/drm/msm/Makefile
index 065ad4138799..ddb4c9d097e4 100644
--- a/drivers/gpu/drm/msm/Makefile
+++ b/drivers/gpu/drm/msm/Makefile
@@ -12,6 +12,7 @@ msm-y := \
 	hdmi/hdmi_connector.o \
 	hdmi/hdmi_hdcp.o \
 	hdmi/hdmi_i2c.o \
+	hdmi/hdmi_phy.o \
 	hdmi/hdmi_phy_8960.o \
 	hdmi/hdmi_phy_8x60.o \
 	hdmi/hdmi_phy_8x74.o \
@@ -52,6 +53,8 @@ msm-y := \
 
 msm-$(CONFIG_DRM_FBDEV_EMULATION) += msm_fbdev.o
 msm-$(CONFIG_COMMON_CLK) += mdp/mdp4/mdp4_lvds_pll.o
+msm-$(CONFIG_COMMON_CLK) += hdmi/hdmi_pll_8960.o
+msm-$(CONFIG_COMMON_CLK) += hdmi/hdmi_phy_8996.o
 
 msm-$(CONFIG_DRM_MSM_DSI) += dsi/dsi.o \
 			mdp/mdp4/mdp4_dsi_encoder.o \
diff --git a/drivers/gpu/drm/msm/adreno/a2xx.xml.h b/drivers/gpu/drm/msm/adreno/a2xx.xml.h
index 9e2aceb4ffe6..fee24297fb92 100644
--- a/drivers/gpu/drm/msm/adreno/a2xx.xml.h
+++ b/drivers/gpu/drm/msm/adreno/a2xx.xml.h
@@ -9,16 +9,17 @@ git clone https://github.com/freedreno/envytools.git
 
 The rules-ng-ng source files this header was generated from are:
 - /home/robclark/src/freedreno/envytools/rnndb/adreno.xml               (    398 bytes, from 2015-09-24 17:25:31)
-- /home/robclark/src/freedreno/envytools/rnndb/freedreno_copyright.xml  (   1453 bytes, from 2015-05-20 20:03:07)
+- /home/robclark/src/freedreno/envytools/rnndb/freedreno_copyright.xml  (   1572 bytes, from 2016-02-10 17:07:21)
 - /home/robclark/src/freedreno/envytools/rnndb/adreno/a2xx.xml          (  32901 bytes, from 2015-05-20 20:03:14)
-- /home/robclark/src/freedreno/envytools/rnndb/adreno/adreno_common.xml (  10755 bytes, from 2015-09-14 20:46:55)
-- /home/robclark/src/freedreno/envytools/rnndb/adreno/adreno_pm4.xml    (  14968 bytes, from 2015-05-20 20:12:27)
-- /home/robclark/src/freedreno/envytools/rnndb/adreno/a3xx.xml          (  67771 bytes, from 2015-09-14 20:46:55)
-- /home/robclark/src/freedreno/envytools/rnndb/adreno/a4xx.xml          (  63970 bytes, from 2015-09-14 20:50:12)
+- /home/robclark/src/freedreno/envytools/rnndb/adreno/adreno_common.xml (  11518 bytes, from 2016-02-10 21:03:25)
+- /home/robclark/src/freedreno/envytools/rnndb/adreno/adreno_pm4.xml    (  16166 bytes, from 2016-02-11 21:20:31)
+- /home/robclark/src/freedreno/envytools/rnndb/adreno/a3xx.xml          (  83967 bytes, from 2016-02-10 17:07:21)
+- /home/robclark/src/freedreno/envytools/rnndb/adreno/a4xx.xml          ( 109916 bytes, from 2016-02-20 18:44:48)
 - /home/robclark/src/freedreno/envytools/rnndb/adreno/ocmem.xml         (   1773 bytes, from 2015-09-24 17:30:00)
 
 Copyright (C) 2013-2015 by the following authors:
 - Rob Clark <robdclark@gmail.com> (robclark)
+- Ilia Mirkin <imirkin@alum.mit.edu> (imirkin)
 
 Permission is hereby granted, free of charge, to any person obtaining
 a copy of this software and associated documentation files (the
diff --git a/drivers/gpu/drm/msm/adreno/a3xx.xml.h b/drivers/gpu/drm/msm/adreno/a3xx.xml.h
index 97dc1c6ec107..27dabd5e57fb 100644
--- a/drivers/gpu/drm/msm/adreno/a3xx.xml.h
+++ b/drivers/gpu/drm/msm/adreno/a3xx.xml.h
@@ -9,16 +9,17 @@ git clone https://github.com/freedreno/envytools.git
 
 The rules-ng-ng source files this header was generated from are:
 - /home/robclark/src/freedreno/envytools/rnndb/adreno.xml               (    398 bytes, from 2015-09-24 17:25:31)
-- /home/robclark/src/freedreno/envytools/rnndb/freedreno_copyright.xml  (   1453 bytes, from 2015-05-20 20:03:07)
+- /home/robclark/src/freedreno/envytools/rnndb/freedreno_copyright.xml  (   1572 bytes, from 2016-02-10 17:07:21)
 - /home/robclark/src/freedreno/envytools/rnndb/adreno/a2xx.xml          (  32901 bytes, from 2015-05-20 20:03:14)
-- /home/robclark/src/freedreno/envytools/rnndb/adreno/adreno_common.xml (  10755 bytes, from 2015-09-14 20:46:55)
-- /home/robclark/src/freedreno/envytools/rnndb/adreno/adreno_pm4.xml    (  14968 bytes, from 2015-05-20 20:12:27)
-- /home/robclark/src/freedreno/envytools/rnndb/adreno/a3xx.xml          (  67771 bytes, from 2015-09-14 20:46:55)
-- /home/robclark/src/freedreno/envytools/rnndb/adreno/a4xx.xml          (  63970 bytes, from 2015-09-14 20:50:12)
+- /home/robclark/src/freedreno/envytools/rnndb/adreno/adreno_common.xml (  11518 bytes, from 2016-02-10 21:03:25)
+- /home/robclark/src/freedreno/envytools/rnndb/adreno/adreno_pm4.xml    (  16166 bytes, from 2016-02-11 21:20:31)
+- /home/robclark/src/freedreno/envytools/rnndb/adreno/a3xx.xml          (  83967 bytes, from 2016-02-10 17:07:21)
+- /home/robclark/src/freedreno/envytools/rnndb/adreno/a4xx.xml          ( 109916 bytes, from 2016-02-20 18:44:48)
 - /home/robclark/src/freedreno/envytools/rnndb/adreno/ocmem.xml         (   1773 bytes, from 2015-09-24 17:30:00)
 
-Copyright (C) 2013-2015 by the following authors:
+Copyright (C) 2013-2016 by the following authors:
 - Rob Clark <robdclark@gmail.com> (robclark)
+- Ilia Mirkin <imirkin@alum.mit.edu> (imirkin)
 
 Permission is hereby granted, free of charge, to any person obtaining
 a copy of this software and associated documentation files (the
@@ -111,10 +112,14 @@ enum a3xx_vtx_fmt {
 	VFMT_8_8_SNORM = 53,
 	VFMT_8_8_8_SNORM = 54,
 	VFMT_8_8_8_8_SNORM = 55,
-	VFMT_10_10_10_2_UINT = 60,
-	VFMT_10_10_10_2_UNORM = 61,
-	VFMT_10_10_10_2_SINT = 62,
-	VFMT_10_10_10_2_SNORM = 63,
+	VFMT_10_10_10_2_UINT = 56,
+	VFMT_10_10_10_2_UNORM = 57,
+	VFMT_10_10_10_2_SINT = 58,
+	VFMT_10_10_10_2_SNORM = 59,
+	VFMT_2_10_10_10_UINT = 60,
+	VFMT_2_10_10_10_UNORM = 61,
+	VFMT_2_10_10_10_SINT = 62,
+	VFMT_2_10_10_10_SNORM = 63,
 };
 
 enum a3xx_tex_fmt {
@@ -138,10 +143,12 @@ enum a3xx_tex_fmt {
 	TFMT_DXT1 = 36,
 	TFMT_DXT3 = 37,
 	TFMT_DXT5 = 38,
+	TFMT_2_10_10_10_UNORM = 40,
 	TFMT_10_10_10_2_UNORM = 41,
 	TFMT_9_9_9_E5_FLOAT = 42,
 	TFMT_11_11_10_FLOAT = 43,
 	TFMT_A8_UNORM = 44,
+	TFMT_L8_UNORM = 45,
 	TFMT_L8_A8_UNORM = 47,
 	TFMT_8_UNORM = 48,
 	TFMT_8_8_UNORM = 49,
@@ -183,6 +190,8 @@ enum a3xx_tex_fmt {
 	TFMT_32_SINT = 92,
 	TFMT_32_32_SINT = 93,
 	TFMT_32_32_32_32_SINT = 95,
+	TFMT_2_10_10_10_UINT = 96,
+	TFMT_10_10_10_2_UINT = 97,
 	TFMT_ETC2_RG11_SNORM = 112,
 	TFMT_ETC2_RG11_UNORM = 113,
 	TFMT_ETC2_R11_SNORM = 114,
@@ -215,6 +224,9 @@ enum a3xx_color_fmt {
 	RB_R8_UINT = 14,
 	RB_R8_SINT = 15,
 	RB_R10G10B10A2_UNORM = 16,
+	RB_A2R10G10B10_UNORM = 17,
+	RB_R10G10B10A2_UINT = 18,
+	RB_A2R10G10B10_UINT = 19,
 	RB_A8_UNORM = 20,
 	RB_R8_UNORM = 21,
 	RB_R16_FLOAT = 24,
@@ -244,30 +256,273 @@ enum a3xx_color_fmt {
 	RB_R32G32B32A32_UINT = 59,
 };
 
+enum a3xx_cp_perfcounter_select {
+	CP_ALWAYS_COUNT = 0,
+	CP_AHB_PFPTRANS_WAIT = 3,
+	CP_AHB_NRTTRANS_WAIT = 6,
+	CP_CSF_NRT_READ_WAIT = 8,
+	CP_CSF_I1_FIFO_FULL = 9,
+	CP_CSF_I2_FIFO_FULL = 10,
+	CP_CSF_ST_FIFO_FULL = 11,
+	CP_RESERVED_12 = 12,
+	CP_CSF_RING_ROQ_FULL = 13,
+	CP_CSF_I1_ROQ_FULL = 14,
+	CP_CSF_I2_ROQ_FULL = 15,
+	CP_CSF_ST_ROQ_FULL = 16,
+	CP_RESERVED_17 = 17,
+	CP_MIU_TAG_MEM_FULL = 18,
+	CP_MIU_NRT_WRITE_STALLED = 22,
+	CP_MIU_NRT_READ_STALLED = 23,
+	CP_ME_REGS_RB_DONE_FIFO_FULL = 26,
+	CP_ME_REGS_VS_EVENT_FIFO_FULL = 27,
+	CP_ME_REGS_PS_EVENT_FIFO_FULL = 28,
+	CP_ME_REGS_CF_EVENT_FIFO_FULL = 29,
+	CP_ME_MICRO_RB_STARVED = 30,
+	CP_AHB_RBBM_DWORD_SENT = 40,
+	CP_ME_BUSY_CLOCKS = 41,
+	CP_ME_WAIT_CONTEXT_AVAIL = 42,
+	CP_PFP_TYPE0_PACKET = 43,
+	CP_PFP_TYPE3_PACKET = 44,
+	CP_CSF_RB_WPTR_NEQ_RPTR = 45,
+	CP_CSF_I1_SIZE_NEQ_ZERO = 46,
+	CP_CSF_I2_SIZE_NEQ_ZERO = 47,
+	CP_CSF_RBI1I2_FETCHING = 48,
+};
+
+enum a3xx_gras_tse_perfcounter_select {
+	GRAS_TSEPERF_INPUT_PRIM = 0,
+	GRAS_TSEPERF_INPUT_NULL_PRIM = 1,
+	GRAS_TSEPERF_TRIVAL_REJ_PRIM = 2,
+	GRAS_TSEPERF_CLIPPED_PRIM = 3,
+	GRAS_TSEPERF_NEW_PRIM = 4,
+	GRAS_TSEPERF_ZERO_AREA_PRIM = 5,
+	GRAS_TSEPERF_FACENESS_CULLED_PRIM = 6,
+	GRAS_TSEPERF_ZERO_PIXEL_PRIM = 7,
+	GRAS_TSEPERF_OUTPUT_NULL_PRIM = 8,
+	GRAS_TSEPERF_OUTPUT_VISIBLE_PRIM = 9,
+	GRAS_TSEPERF_PRE_CLIP_PRIM = 10,
+	GRAS_TSEPERF_POST_CLIP_PRIM = 11,
+	GRAS_TSEPERF_WORKING_CYCLES = 12,
+	GRAS_TSEPERF_PC_STARVE = 13,
+	GRAS_TSERASPERF_STALL = 14,
+};
+
+enum a3xx_gras_ras_perfcounter_select {
+	GRAS_RASPERF_16X16_TILES = 0,
+	GRAS_RASPERF_8X8_TILES = 1,
+	GRAS_RASPERF_4X4_TILES = 2,
+	GRAS_RASPERF_WORKING_CYCLES = 3,
+	GRAS_RASPERF_STALL_CYCLES_BY_RB = 4,
+	GRAS_RASPERF_STALL_CYCLES_BY_VSC = 5,
+	GRAS_RASPERF_STARVE_CYCLES_BY_TSE = 6,
+};
+
+enum a3xx_hlsq_perfcounter_select {
+	HLSQ_PERF_SP_VS_CONSTANT = 0,
+	HLSQ_PERF_SP_VS_INSTRUCTIONS = 1,
+	HLSQ_PERF_SP_FS_CONSTANT = 2,
+	HLSQ_PERF_SP_FS_INSTRUCTIONS = 3,
+	HLSQ_PERF_TP_STATE = 4,
+	HLSQ_PERF_QUADS = 5,
+	HLSQ_PERF_PIXELS = 6,
+	HLSQ_PERF_VERTICES = 7,
+	HLSQ_PERF_FS8_THREADS = 8,
+	HLSQ_PERF_FS16_THREADS = 9,
+	HLSQ_PERF_FS32_THREADS = 10,
+	HLSQ_PERF_VS8_THREADS = 11,
+	HLSQ_PERF_VS16_THREADS = 12,
+	HLSQ_PERF_SP_VS_DATA_BYTES = 13,
+	HLSQ_PERF_SP_FS_DATA_BYTES = 14,
+	HLSQ_PERF_ACTIVE_CYCLES = 15,
+	HLSQ_PERF_STALL_CYCLES_SP_STATE = 16,
+	HLSQ_PERF_STALL_CYCLES_SP_VS = 17,
+	HLSQ_PERF_STALL_CYCLES_SP_FS = 18,
+	HLSQ_PERF_STALL_CYCLES_UCHE = 19,
+	HLSQ_PERF_RBBM_LOAD_CYCLES = 20,
+	HLSQ_PERF_DI_TO_VS_START_SP0 = 21,
+	HLSQ_PERF_DI_TO_FS_START_SP0 = 22,
+	HLSQ_PERF_VS_START_TO_DONE_SP0 = 23,
+	HLSQ_PERF_FS_START_TO_DONE_SP0 = 24,
+	HLSQ_PERF_SP_STATE_COPY_CYCLES_VS = 25,
+	HLSQ_PERF_SP_STATE_COPY_CYCLES_FS = 26,
+	HLSQ_PERF_UCHE_LATENCY_CYCLES = 27,
+	HLSQ_PERF_UCHE_LATENCY_COUNT = 28,
+};
+
+enum a3xx_pc_perfcounter_select {
+	PC_PCPERF_VISIBILITY_STREAMS = 0,
+	PC_PCPERF_TOTAL_INSTANCES = 1,
+	PC_PCPERF_PRIMITIVES_PC_VPC = 2,
+	PC_PCPERF_PRIMITIVES_KILLED_BY_VS = 3,
+	PC_PCPERF_PRIMITIVES_VISIBLE_BY_VS = 4,
+	PC_PCPERF_DRAWCALLS_KILLED_BY_VS = 5,
+	PC_PCPERF_DRAWCALLS_VISIBLE_BY_VS = 6,
+	PC_PCPERF_VERTICES_TO_VFD = 7,
+	PC_PCPERF_REUSED_VERTICES = 8,
+	PC_PCPERF_CYCLES_STALLED_BY_VFD = 9,
+	PC_PCPERF_CYCLES_STALLED_BY_TSE = 10,
+	PC_PCPERF_CYCLES_STALLED_BY_VBIF = 11,
+	PC_PCPERF_CYCLES_IS_WORKING = 12,
+};
+
+enum a3xx_rb_perfcounter_select {
+	RB_RBPERF_ACTIVE_CYCLES_ANY = 0,
+	RB_RBPERF_ACTIVE_CYCLES_ALL = 1,
+	RB_RBPERF_STARVE_CYCLES_BY_SP = 2,
+	RB_RBPERF_STARVE_CYCLES_BY_RAS = 3,
+	RB_RBPERF_STARVE_CYCLES_BY_MARB = 4,
+	RB_RBPERF_STALL_CYCLES_BY_MARB = 5,
+	RB_RBPERF_STALL_CYCLES_BY_HLSQ = 6,
+	RB_RBPERF_RB_MARB_DATA = 7,
+	RB_RBPERF_SP_RB_QUAD = 8,
+	RB_RBPERF_RAS_EARLY_Z_QUADS = 9,
+	RB_RBPERF_GMEM_CH0_READ = 10,
+	RB_RBPERF_GMEM_CH1_READ = 11,
+	RB_RBPERF_GMEM_CH0_WRITE = 12,
+	RB_RBPERF_GMEM_CH1_WRITE = 13,
+	RB_RBPERF_CP_CONTEXT_DONE = 14,
+	RB_RBPERF_CP_CACHE_FLUSH = 15,
+	RB_RBPERF_CP_ZPASS_DONE = 16,
+};
+
+enum a3xx_rbbm_perfcounter_select {
+	RBBM_ALAWYS_ON = 0,
+	RBBM_VBIF_BUSY = 1,
+	RBBM_TSE_BUSY = 2,
+	RBBM_RAS_BUSY = 3,
+	RBBM_PC_DCALL_BUSY = 4,
+	RBBM_PC_VSD_BUSY = 5,
+	RBBM_VFD_BUSY = 6,
+	RBBM_VPC_BUSY = 7,
+	RBBM_UCHE_BUSY = 8,
+	RBBM_VSC_BUSY = 9,
+	RBBM_HLSQ_BUSY = 10,
+	RBBM_ANY_RB_BUSY = 11,
+	RBBM_ANY_TEX_BUSY = 12,
+	RBBM_ANY_USP_BUSY = 13,
+	RBBM_ANY_MARB_BUSY = 14,
+	RBBM_ANY_ARB_BUSY = 15,
+	RBBM_AHB_STATUS_BUSY = 16,
+	RBBM_AHB_STATUS_STALLED = 17,
+	RBBM_AHB_STATUS_TXFR = 18,
+	RBBM_AHB_STATUS_TXFR_SPLIT = 19,
+	RBBM_AHB_STATUS_TXFR_ERROR = 20,
+	RBBM_AHB_STATUS_LONG_STALL = 21,
+	RBBM_RBBM_STATUS_MASKED = 22,
+};
+
 enum a3xx_sp_perfcounter_select {
+	SP_LM_LOAD_INSTRUCTIONS = 0,
+	SP_LM_STORE_INSTRUCTIONS = 1,
+	SP_LM_ATOMICS = 2,
+	SP_UCHE_LOAD_INSTRUCTIONS = 3,
+	SP_UCHE_STORE_INSTRUCTIONS = 4,
+	SP_UCHE_ATOMICS = 5,
+	SP_VS_TEX_INSTRUCTIONS = 6,
+	SP_VS_CFLOW_INSTRUCTIONS = 7,
+	SP_VS_EFU_INSTRUCTIONS = 8,
+	SP_VS_FULL_ALU_INSTRUCTIONS = 9,
+	SP_VS_HALF_ALU_INSTRUCTIONS = 10,
+	SP_FS_TEX_INSTRUCTIONS = 11,
 	SP_FS_CFLOW_INSTRUCTIONS = 12,
+	SP_FS_EFU_INSTRUCTIONS = 13,
 	SP_FS_FULL_ALU_INSTRUCTIONS = 14,
-	SP0_ICL1_MISSES = 26,
+	SP_FS_HALF_ALU_INSTRUCTIONS = 15,
+	SP_FS_BARY_INSTRUCTIONS = 16,
+	SP_VS_INSTRUCTIONS = 17,
+	SP_FS_INSTRUCTIONS = 18,
+	SP_ADDR_LOCK_COUNT = 19,
+	SP_UCHE_READ_TRANS = 20,
+	SP_UCHE_WRITE_TRANS = 21,
+	SP_EXPORT_VPC_TRANS = 22,
+	SP_EXPORT_RB_TRANS = 23,
+	SP_PIXELS_KILLED = 24,
+	SP_ICL1_REQUESTS = 25,
+	SP_ICL1_MISSES = 26,
+	SP_ICL0_REQUESTS = 27,
+	SP_ICL0_MISSES = 28,
 	SP_ALU_ACTIVE_CYCLES = 29,
+	SP_EFU_ACTIVE_CYCLES = 30,
+	SP_STALL_CYCLES_BY_VPC = 31,
+	SP_STALL_CYCLES_BY_TP = 32,
+	SP_STALL_CYCLES_BY_UCHE = 33,
+	SP_STALL_CYCLES_BY_RB = 34,
+	SP_ACTIVE_CYCLES_ANY = 35,
+	SP_ACTIVE_CYCLES_ALL = 36,
+};
+
+enum a3xx_tp_perfcounter_select {
+	TPL1_TPPERF_L1_REQUESTS = 0,
+	TPL1_TPPERF_TP0_L1_REQUESTS = 1,
+	TPL1_TPPERF_TP0_L1_MISSES = 2,
+	TPL1_TPPERF_TP1_L1_REQUESTS = 3,
+	TPL1_TPPERF_TP1_L1_MISSES = 4,
+	TPL1_TPPERF_TP2_L1_REQUESTS = 5,
+	TPL1_TPPERF_TP2_L1_MISSES = 6,
+	TPL1_TPPERF_TP3_L1_REQUESTS = 7,
+	TPL1_TPPERF_TP3_L1_MISSES = 8,
+	TPL1_TPPERF_OUTPUT_TEXELS_POINT = 9,
+	TPL1_TPPERF_OUTPUT_TEXELS_BILINEAR = 10,
+	TPL1_TPPERF_OUTPUT_TEXELS_MIP = 11,
+	TPL1_TPPERF_OUTPUT_TEXELS_ANISO = 12,
+	TPL1_TPPERF_BILINEAR_OPS = 13,
+	TPL1_TPPERF_QUADSQUADS_OFFSET = 14,
+	TPL1_TPPERF_QUADQUADS_SHADOW = 15,
+	TPL1_TPPERF_QUADS_ARRAY = 16,
+	TPL1_TPPERF_QUADS_PROJECTION = 17,
+	TPL1_TPPERF_QUADS_GRADIENT = 18,
+	TPL1_TPPERF_QUADS_1D2D = 19,
+	TPL1_TPPERF_QUADS_3DCUBE = 20,
+	TPL1_TPPERF_ZERO_LOD = 21,
+	TPL1_TPPERF_OUTPUT_TEXELS = 22,
+	TPL1_TPPERF_ACTIVE_CYCLES_ANY = 23,
+	TPL1_TPPERF_ACTIVE_CYCLES_ALL = 24,
+	TPL1_TPPERF_STALL_CYCLES_BY_ARB = 25,
+	TPL1_TPPERF_LATENCY = 26,
+	TPL1_TPPERF_LATENCY_TRANS = 27,
 };
 
-enum a3xx_rop_code {
-	ROP_CLEAR = 0,
-	ROP_NOR = 1,
-	ROP_AND_INVERTED = 2,
-	ROP_COPY_INVERTED = 3,
-	ROP_AND_REVERSE = 4,
-	ROP_INVERT = 5,
-	ROP_XOR = 6,
-	ROP_NAND = 7,
-	ROP_AND = 8,
-	ROP_EQUIV = 9,
-	ROP_NOOP = 10,
-	ROP_OR_INVERTED = 11,
-	ROP_COPY = 12,
-	ROP_OR_REVERSE = 13,
-	ROP_OR = 14,
-	ROP_SET = 15,
+enum a3xx_vfd_perfcounter_select {
+	VFD_PERF_UCHE_BYTE_FETCHED = 0,
+	VFD_PERF_UCHE_TRANS = 1,
+	VFD_PERF_VPC_BYPASS_COMPONENTS = 2,
+	VFD_PERF_FETCH_INSTRUCTIONS = 3,
+	VFD_PERF_DECODE_INSTRUCTIONS = 4,
+	VFD_PERF_ACTIVE_CYCLES = 5,
+	VFD_PERF_STALL_CYCLES_UCHE = 6,
+	VFD_PERF_STALL_CYCLES_HLSQ = 7,
+	VFD_PERF_STALL_CYCLES_VPC_BYPASS = 8,
+	VFD_PERF_STALL_CYCLES_VPC_ALLOC = 9,
+};
+
+enum a3xx_vpc_perfcounter_select {
+	VPC_PERF_SP_LM_PRIMITIVES = 0,
+	VPC_PERF_COMPONENTS_FROM_SP = 1,
+	VPC_PERF_SP_LM_COMPONENTS = 2,
+	VPC_PERF_ACTIVE_CYCLES = 3,
+	VPC_PERF_STALL_CYCLES_LM = 4,
+	VPC_PERF_STALL_CYCLES_RAS = 5,
+};
+
+enum a3xx_uche_perfcounter_select {
+	UCHE_UCHEPERF_VBIF_READ_BEATS_TP = 0,
+	UCHE_UCHEPERF_VBIF_READ_BEATS_VFD = 1,
+	UCHE_UCHEPERF_VBIF_READ_BEATS_HLSQ = 2,
+	UCHE_UCHEPERF_VBIF_READ_BEATS_MARB = 3,
+	UCHE_UCHEPERF_VBIF_READ_BEATS_SP = 4,
+	UCHE_UCHEPERF_READ_REQUESTS_TP = 8,
+	UCHE_UCHEPERF_READ_REQUESTS_VFD = 9,
+	UCHE_UCHEPERF_READ_REQUESTS_HLSQ = 10,
+	UCHE_UCHEPERF_READ_REQUESTS_MARB = 11,
+	UCHE_UCHEPERF_READ_REQUESTS_SP = 12,
+	UCHE_UCHEPERF_WRITE_REQUESTS_MARB = 13,
+	UCHE_UCHEPERF_WRITE_REQUESTS_SP = 14,
+	UCHE_UCHEPERF_TAG_CHECK_FAILS = 15,
+	UCHE_UCHEPERF_EVICTS = 16,
+	UCHE_UCHEPERF_FLUSHES = 17,
+	UCHE_UCHEPERF_VBIF_LATENCY_CYCLES = 18,
+	UCHE_UCHEPERF_VBIF_LATENCY_SAMPLES = 19,
+	UCHE_UCHEPERF_ACTIVE_CYCLES = 20,
 };
 
 enum a3xx_rb_blend_opcode {
@@ -1429,15 +1684,23 @@ static inline uint32_t A3XX_PC_PRIM_VTX_CNTL_POLYMODE_BACK_PTYPE(enum adreno_pa_
 #define REG_A3XX_PC_RESTART_INDEX				0x000021ed
 
 #define REG_A3XX_HLSQ_CONTROL_0_REG				0x00002200
-#define A3XX_HLSQ_CONTROL_0_REG_FSTHREADSIZE__MASK		0x00000010
+#define A3XX_HLSQ_CONTROL_0_REG_FSTHREADSIZE__MASK		0x00000030
 #define A3XX_HLSQ_CONTROL_0_REG_FSTHREADSIZE__SHIFT		4
 static inline uint32_t A3XX_HLSQ_CONTROL_0_REG_FSTHREADSIZE(enum a3xx_threadsize val)
 {
 	return ((val) << A3XX_HLSQ_CONTROL_0_REG_FSTHREADSIZE__SHIFT) & A3XX_HLSQ_CONTROL_0_REG_FSTHREADSIZE__MASK;
 }
 #define A3XX_HLSQ_CONTROL_0_REG_FSSUPERTHREADENABLE		0x00000040
+#define A3XX_HLSQ_CONTROL_0_REG_COMPUTEMODE			0x00000100
 #define A3XX_HLSQ_CONTROL_0_REG_SPSHADERRESTART			0x00000200
 #define A3XX_HLSQ_CONTROL_0_REG_RESERVED2			0x00000400
+#define A3XX_HLSQ_CONTROL_0_REG_CYCLETIMEOUTLIMITVPC__MASK	0x00fff000
+#define A3XX_HLSQ_CONTROL_0_REG_CYCLETIMEOUTLIMITVPC__SHIFT	12
+static inline uint32_t A3XX_HLSQ_CONTROL_0_REG_CYCLETIMEOUTLIMITVPC(uint32_t val)
+{
+	return ((val) << A3XX_HLSQ_CONTROL_0_REG_CYCLETIMEOUTLIMITVPC__SHIFT) & A3XX_HLSQ_CONTROL_0_REG_CYCLETIMEOUTLIMITVPC__MASK;
+}
+#define A3XX_HLSQ_CONTROL_0_REG_FSONLYTEX			0x02000000
 #define A3XX_HLSQ_CONTROL_0_REG_CHUNKDISABLE			0x04000000
 #define A3XX_HLSQ_CONTROL_0_REG_CONSTMODE__MASK			0x08000000
 #define A3XX_HLSQ_CONTROL_0_REG_CONSTMODE__SHIFT		27
@@ -1451,17 +1714,39 @@ static inline uint32_t A3XX_HLSQ_CONTROL_0_REG_CONSTMODE(uint32_t val)
 #define A3XX_HLSQ_CONTROL_0_REG_SINGLECONTEXT			0x80000000
 
 #define REG_A3XX_HLSQ_CONTROL_1_REG				0x00002201
-#define A3XX_HLSQ_CONTROL_1_REG_VSTHREADSIZE__MASK		0x00000040
+#define A3XX_HLSQ_CONTROL_1_REG_VSTHREADSIZE__MASK		0x000000c0
 #define A3XX_HLSQ_CONTROL_1_REG_VSTHREADSIZE__SHIFT		6
 static inline uint32_t A3XX_HLSQ_CONTROL_1_REG_VSTHREADSIZE(enum a3xx_threadsize val)
 {
 	return ((val) << A3XX_HLSQ_CONTROL_1_REG_VSTHREADSIZE__SHIFT) & A3XX_HLSQ_CONTROL_1_REG_VSTHREADSIZE__MASK;
 }
 #define A3XX_HLSQ_CONTROL_1_REG_VSSUPERTHREADENABLE		0x00000100
-#define A3XX_HLSQ_CONTROL_1_REG_RESERVED1			0x00000200
-#define A3XX_HLSQ_CONTROL_1_REG_ZWCOORD				0x02000000
+#define A3XX_HLSQ_CONTROL_1_REG_FRAGCOORDXYREGID__MASK		0x00ff0000
+#define A3XX_HLSQ_CONTROL_1_REG_FRAGCOORDXYREGID__SHIFT		16
+static inline uint32_t A3XX_HLSQ_CONTROL_1_REG_FRAGCOORDXYREGID(uint32_t val)
+{
+	return ((val) << A3XX_HLSQ_CONTROL_1_REG_FRAGCOORDXYREGID__SHIFT) & A3XX_HLSQ_CONTROL_1_REG_FRAGCOORDXYREGID__MASK;
+}
+#define A3XX_HLSQ_CONTROL_1_REG_FRAGCOORDZWREGID__MASK		0xff000000
+#define A3XX_HLSQ_CONTROL_1_REG_FRAGCOORDZWREGID__SHIFT		24
+static inline uint32_t A3XX_HLSQ_CONTROL_1_REG_FRAGCOORDZWREGID(uint32_t val)
+{
+	return ((val) << A3XX_HLSQ_CONTROL_1_REG_FRAGCOORDZWREGID__SHIFT) & A3XX_HLSQ_CONTROL_1_REG_FRAGCOORDZWREGID__MASK;
+}
 
 #define REG_A3XX_HLSQ_CONTROL_2_REG				0x00002202
+#define A3XX_HLSQ_CONTROL_2_REG_FACENESSREGID__MASK		0x000003fc
+#define A3XX_HLSQ_CONTROL_2_REG_FACENESSREGID__SHIFT		2
+static inline uint32_t A3XX_HLSQ_CONTROL_2_REG_FACENESSREGID(uint32_t val)
+{
+	return ((val) << A3XX_HLSQ_CONTROL_2_REG_FACENESSREGID__SHIFT) & A3XX_HLSQ_CONTROL_2_REG_FACENESSREGID__MASK;
+}
+#define A3XX_HLSQ_CONTROL_2_REG_COVVALUEREGID__MASK		0x03fc0000
+#define A3XX_HLSQ_CONTROL_2_REG_COVVALUEREGID__SHIFT		18
+static inline uint32_t A3XX_HLSQ_CONTROL_2_REG_COVVALUEREGID(uint32_t val)
+{
+	return ((val) << A3XX_HLSQ_CONTROL_2_REG_COVVALUEREGID__SHIFT) & A3XX_HLSQ_CONTROL_2_REG_COVVALUEREGID__MASK;
+}
 #define A3XX_HLSQ_CONTROL_2_REG_PRIMALLOCTHRESHOLD__MASK	0xfc000000
 #define A3XX_HLSQ_CONTROL_2_REG_PRIMALLOCTHRESHOLD__SHIFT	26
 static inline uint32_t A3XX_HLSQ_CONTROL_2_REG_PRIMALLOCTHRESHOLD(uint32_t val)
@@ -1478,13 +1763,13 @@ static inline uint32_t A3XX_HLSQ_CONTROL_3_REG_REGID(uint32_t val)
 }
 
 #define REG_A3XX_HLSQ_VS_CONTROL_REG				0x00002204
-#define A3XX_HLSQ_VS_CONTROL_REG_CONSTLENGTH__MASK		0x00000fff
+#define A3XX_HLSQ_VS_CONTROL_REG_CONSTLENGTH__MASK		0x000003ff
 #define A3XX_HLSQ_VS_CONTROL_REG_CONSTLENGTH__SHIFT		0
 static inline uint32_t A3XX_HLSQ_VS_CONTROL_REG_CONSTLENGTH(uint32_t val)
 {
 	return ((val) << A3XX_HLSQ_VS_CONTROL_REG_CONSTLENGTH__SHIFT) & A3XX_HLSQ_VS_CONTROL_REG_CONSTLENGTH__MASK;
 }
-#define A3XX_HLSQ_VS_CONTROL_REG_CONSTSTARTOFFSET__MASK		0x00fff000
+#define A3XX_HLSQ_VS_CONTROL_REG_CONSTSTARTOFFSET__MASK		0x001ff000
 #define A3XX_HLSQ_VS_CONTROL_REG_CONSTSTARTOFFSET__SHIFT	12
 static inline uint32_t A3XX_HLSQ_VS_CONTROL_REG_CONSTSTARTOFFSET(uint32_t val)
 {
@@ -1498,13 +1783,13 @@ static inline uint32_t A3XX_HLSQ_VS_CONTROL_REG_INSTRLENGTH(uint32_t val)
 }
 
 #define REG_A3XX_HLSQ_FS_CONTROL_REG				0x00002205
-#define A3XX_HLSQ_FS_CONTROL_REG_CONSTLENGTH__MASK		0x00000fff
+#define A3XX_HLSQ_FS_CONTROL_REG_CONSTLENGTH__MASK		0x000003ff
 #define A3XX_HLSQ_FS_CONTROL_REG_CONSTLENGTH__SHIFT		0
 static inline uint32_t A3XX_HLSQ_FS_CONTROL_REG_CONSTLENGTH(uint32_t val)
 {
 	return ((val) << A3XX_HLSQ_FS_CONTROL_REG_CONSTLENGTH__SHIFT) & A3XX_HLSQ_FS_CONTROL_REG_CONSTLENGTH__MASK;
 }
-#define A3XX_HLSQ_FS_CONTROL_REG_CONSTSTARTOFFSET__MASK		0x00fff000
+#define A3XX_HLSQ_FS_CONTROL_REG_CONSTSTARTOFFSET__MASK		0x001ff000
 #define A3XX_HLSQ_FS_CONTROL_REG_CONSTSTARTOFFSET__SHIFT	12
 static inline uint32_t A3XX_HLSQ_FS_CONTROL_REG_CONSTSTARTOFFSET(uint32_t val)
 {
@@ -1518,13 +1803,13 @@ static inline uint32_t A3XX_HLSQ_FS_CONTROL_REG_INSTRLENGTH(uint32_t val)
 }
 
 #define REG_A3XX_HLSQ_CONST_VSPRESV_RANGE_REG			0x00002206
-#define A3XX_HLSQ_CONST_VSPRESV_RANGE_REG_STARTENTRY__MASK	0x0000ffff
+#define A3XX_HLSQ_CONST_VSPRESV_RANGE_REG_STARTENTRY__MASK	0x000001ff
 #define A3XX_HLSQ_CONST_VSPRESV_RANGE_REG_STARTENTRY__SHIFT	0
 static inline uint32_t A3XX_HLSQ_CONST_VSPRESV_RANGE_REG_STARTENTRY(uint32_t val)
 {
 	return ((val) << A3XX_HLSQ_CONST_VSPRESV_RANGE_REG_STARTENTRY__SHIFT) & A3XX_HLSQ_CONST_VSPRESV_RANGE_REG_STARTENTRY__MASK;
 }
-#define A3XX_HLSQ_CONST_VSPRESV_RANGE_REG_ENDENTRY__MASK	0xffff0000
+#define A3XX_HLSQ_CONST_VSPRESV_RANGE_REG_ENDENTRY__MASK	0x01ff0000
 #define A3XX_HLSQ_CONST_VSPRESV_RANGE_REG_ENDENTRY__SHIFT	16
 static inline uint32_t A3XX_HLSQ_CONST_VSPRESV_RANGE_REG_ENDENTRY(uint32_t val)
 {
@@ -1532,13 +1817,13 @@ static inline uint32_t A3XX_HLSQ_CONST_VSPRESV_RANGE_REG_ENDENTRY(uint32_t val)
 }
 
 #define REG_A3XX_HLSQ_CONST_FSPRESV_RANGE_REG			0x00002207
-#define A3XX_HLSQ_CONST_FSPRESV_RANGE_REG_STARTENTRY__MASK	0x0000ffff
+#define A3XX_HLSQ_CONST_FSPRESV_RANGE_REG_STARTENTRY__MASK	0x000001ff
 #define A3XX_HLSQ_CONST_FSPRESV_RANGE_REG_STARTENTRY__SHIFT	0
 static inline uint32_t A3XX_HLSQ_CONST_FSPRESV_RANGE_REG_STARTENTRY(uint32_t val)
 {
 	return ((val) << A3XX_HLSQ_CONST_FSPRESV_RANGE_REG_STARTENTRY__SHIFT) & A3XX_HLSQ_CONST_FSPRESV_RANGE_REG_STARTENTRY__MASK;
 }
-#define A3XX_HLSQ_CONST_FSPRESV_RANGE_REG_ENDENTRY__MASK	0xffff0000
+#define A3XX_HLSQ_CONST_FSPRESV_RANGE_REG_ENDENTRY__MASK	0x01ff0000
 #define A3XX_HLSQ_CONST_FSPRESV_RANGE_REG_ENDENTRY__SHIFT	16
 static inline uint32_t A3XX_HLSQ_CONST_FSPRESV_RANGE_REG_ENDENTRY(uint32_t val)
 {
@@ -1620,12 +1905,24 @@ static inline uint32_t A3XX_VFD_CONTROL_0_STRMFETCHINSTRCNT(uint32_t val)
 }
 
 #define REG_A3XX_VFD_CONTROL_1					0x00002241
-#define A3XX_VFD_CONTROL_1_MAXSTORAGE__MASK			0x0000ffff
+#define A3XX_VFD_CONTROL_1_MAXSTORAGE__MASK			0x0000000f
 #define A3XX_VFD_CONTROL_1_MAXSTORAGE__SHIFT			0
 static inline uint32_t A3XX_VFD_CONTROL_1_MAXSTORAGE(uint32_t val)
 {
 	return ((val) << A3XX_VFD_CONTROL_1_MAXSTORAGE__SHIFT) & A3XX_VFD_CONTROL_1_MAXSTORAGE__MASK;
 }
+#define A3XX_VFD_CONTROL_1_MAXTHRESHOLD__MASK			0x000000f0
+#define A3XX_VFD_CONTROL_1_MAXTHRESHOLD__SHIFT			4
+static inline uint32_t A3XX_VFD_CONTROL_1_MAXTHRESHOLD(uint32_t val)
+{
+	return ((val) << A3XX_VFD_CONTROL_1_MAXTHRESHOLD__SHIFT) & A3XX_VFD_CONTROL_1_MAXTHRESHOLD__MASK;
+}
+#define A3XX_VFD_CONTROL_1_MINTHRESHOLD__MASK			0x00000f00
+#define A3XX_VFD_CONTROL_1_MINTHRESHOLD__SHIFT			8
+static inline uint32_t A3XX_VFD_CONTROL_1_MINTHRESHOLD(uint32_t val)
+{
+	return ((val) << A3XX_VFD_CONTROL_1_MINTHRESHOLD__SHIFT) & A3XX_VFD_CONTROL_1_MINTHRESHOLD__MASK;
+}
 #define A3XX_VFD_CONTROL_1_REGID4VTX__MASK			0x00ff0000
 #define A3XX_VFD_CONTROL_1_REGID4VTX__SHIFT			16
 static inline uint32_t A3XX_VFD_CONTROL_1_REGID4VTX(uint32_t val)
@@ -2008,24 +2305,19 @@ static inline uint32_t A3XX_SP_VS_CTRL_REG0_INSTRBUFFERMODE(enum a3xx_instrbuffe
 	return ((val) << A3XX_SP_VS_CTRL_REG0_INSTRBUFFERMODE__SHIFT) & A3XX_SP_VS_CTRL_REG0_INSTRBUFFERMODE__MASK;
 }
 #define A3XX_SP_VS_CTRL_REG0_CACHEINVALID			0x00000004
+#define A3XX_SP_VS_CTRL_REG0_ALUSCHMODE				0x00000008
 #define A3XX_SP_VS_CTRL_REG0_HALFREGFOOTPRINT__MASK		0x000003f0
 #define A3XX_SP_VS_CTRL_REG0_HALFREGFOOTPRINT__SHIFT		4
 static inline uint32_t A3XX_SP_VS_CTRL_REG0_HALFREGFOOTPRINT(uint32_t val)
 {
 	return ((val) << A3XX_SP_VS_CTRL_REG0_HALFREGFOOTPRINT__SHIFT) & A3XX_SP_VS_CTRL_REG0_HALFREGFOOTPRINT__MASK;
 }
-#define A3XX_SP_VS_CTRL_REG0_FULLREGFOOTPRINT__MASK		0x0003fc00
+#define A3XX_SP_VS_CTRL_REG0_FULLREGFOOTPRINT__MASK		0x0000fc00
 #define A3XX_SP_VS_CTRL_REG0_FULLREGFOOTPRINT__SHIFT		10
 static inline uint32_t A3XX_SP_VS_CTRL_REG0_FULLREGFOOTPRINT(uint32_t val)
 {
 	return ((val) << A3XX_SP_VS_CTRL_REG0_FULLREGFOOTPRINT__SHIFT) & A3XX_SP_VS_CTRL_REG0_FULLREGFOOTPRINT__MASK;
 }
-#define A3XX_SP_VS_CTRL_REG0_INOUTREGOVERLAP__MASK		0x000c0000
-#define A3XX_SP_VS_CTRL_REG0_INOUTREGOVERLAP__SHIFT		18
-static inline uint32_t A3XX_SP_VS_CTRL_REG0_INOUTREGOVERLAP(uint32_t val)
-{
-	return ((val) << A3XX_SP_VS_CTRL_REG0_INOUTREGOVERLAP__SHIFT) & A3XX_SP_VS_CTRL_REG0_INOUTREGOVERLAP__MASK;
-}
 #define A3XX_SP_VS_CTRL_REG0_THREADSIZE__MASK			0x00100000
 #define A3XX_SP_VS_CTRL_REG0_THREADSIZE__SHIFT			20
 static inline uint32_t A3XX_SP_VS_CTRL_REG0_THREADSIZE(enum a3xx_threadsize val)
@@ -2033,8 +2325,6 @@ static inline uint32_t A3XX_SP_VS_CTRL_REG0_THREADSIZE(enum a3xx_threadsize val)
 	return ((val) << A3XX_SP_VS_CTRL_REG0_THREADSIZE__SHIFT) & A3XX_SP_VS_CTRL_REG0_THREADSIZE__MASK;
 }
 #define A3XX_SP_VS_CTRL_REG0_SUPERTHREADMODE			0x00200000
-#define A3XX_SP_VS_CTRL_REG0_PIXLODENABLE			0x00400000
-#define A3XX_SP_VS_CTRL_REG0_COMPUTEMODE			0x00800000
 #define A3XX_SP_VS_CTRL_REG0_LENGTH__MASK			0xff000000
 #define A3XX_SP_VS_CTRL_REG0_LENGTH__SHIFT			24
 static inline uint32_t A3XX_SP_VS_CTRL_REG0_LENGTH(uint32_t val)
@@ -2075,7 +2365,8 @@ static inline uint32_t A3XX_SP_VS_PARAM_REG_PSIZEREGID(uint32_t val)
 {
 	return ((val) << A3XX_SP_VS_PARAM_REG_PSIZEREGID__SHIFT) & A3XX_SP_VS_PARAM_REG_PSIZEREGID__MASK;
 }
-#define A3XX_SP_VS_PARAM_REG_TOTALVSOUTVAR__MASK		0xfff00000
+#define A3XX_SP_VS_PARAM_REG_POS2DMODE				0x00010000
+#define A3XX_SP_VS_PARAM_REG_TOTALVSOUTVAR__MASK		0x01f00000
 #define A3XX_SP_VS_PARAM_REG_TOTALVSOUTVAR__SHIFT		20
 static inline uint32_t A3XX_SP_VS_PARAM_REG_TOTALVSOUTVAR(uint32_t val)
 {
@@ -2085,24 +2376,26 @@ static inline uint32_t A3XX_SP_VS_PARAM_REG_TOTALVSOUTVAR(uint32_t val)
 static inline uint32_t REG_A3XX_SP_VS_OUT(uint32_t i0) { return 0x000022c7 + 0x1*i0; }
 
 static inline uint32_t REG_A3XX_SP_VS_OUT_REG(uint32_t i0) { return 0x000022c7 + 0x1*i0; }
-#define A3XX_SP_VS_OUT_REG_A_REGID__MASK			0x000001ff
+#define A3XX_SP_VS_OUT_REG_A_REGID__MASK			0x000000ff
 #define A3XX_SP_VS_OUT_REG_A_REGID__SHIFT			0
 static inline uint32_t A3XX_SP_VS_OUT_REG_A_REGID(uint32_t val)
 {
 	return ((val) << A3XX_SP_VS_OUT_REG_A_REGID__SHIFT) & A3XX_SP_VS_OUT_REG_A_REGID__MASK;
 }
+#define A3XX_SP_VS_OUT_REG_A_HALF				0x00000100
 #define A3XX_SP_VS_OUT_REG_A_COMPMASK__MASK			0x00001e00
 #define A3XX_SP_VS_OUT_REG_A_COMPMASK__SHIFT			9
 static inline uint32_t A3XX_SP_VS_OUT_REG_A_COMPMASK(uint32_t val)
 {
 	return ((val) << A3XX_SP_VS_OUT_REG_A_COMPMASK__SHIFT) & A3XX_SP_VS_OUT_REG_A_COMPMASK__MASK;
 }
-#define A3XX_SP_VS_OUT_REG_B_REGID__MASK			0x01ff0000
+#define A3XX_SP_VS_OUT_REG_B_REGID__MASK			0x00ff0000
 #define A3XX_SP_VS_OUT_REG_B_REGID__SHIFT			16
 static inline uint32_t A3XX_SP_VS_OUT_REG_B_REGID(uint32_t val)
 {
 	return ((val) << A3XX_SP_VS_OUT_REG_B_REGID__SHIFT) & A3XX_SP_VS_OUT_REG_B_REGID__MASK;
 }
+#define A3XX_SP_VS_OUT_REG_B_HALF				0x01000000
 #define A3XX_SP_VS_OUT_REG_B_COMPMASK__MASK			0x1e000000
 #define A3XX_SP_VS_OUT_REG_B_COMPMASK__SHIFT			25
 static inline uint32_t A3XX_SP_VS_OUT_REG_B_COMPMASK(uint32_t val)
@@ -2113,25 +2406,25 @@ static inline uint32_t A3XX_SP_VS_OUT_REG_B_COMPMASK(uint32_t val)
 static inline uint32_t REG_A3XX_SP_VS_VPC_DST(uint32_t i0) { return 0x000022d0 + 0x1*i0; }
 
 static inline uint32_t REG_A3XX_SP_VS_VPC_DST_REG(uint32_t i0) { return 0x000022d0 + 0x1*i0; }
-#define A3XX_SP_VS_VPC_DST_REG_OUTLOC0__MASK			0x000000ff
+#define A3XX_SP_VS_VPC_DST_REG_OUTLOC0__MASK			0x0000007f
 #define A3XX_SP_VS_VPC_DST_REG_OUTLOC0__SHIFT			0
 static inline uint32_t A3XX_SP_VS_VPC_DST_REG_OUTLOC0(uint32_t val)
 {
 	return ((val) << A3XX_SP_VS_VPC_DST_REG_OUTLOC0__SHIFT) & A3XX_SP_VS_VPC_DST_REG_OUTLOC0__MASK;
 }
-#define A3XX_SP_VS_VPC_DST_REG_OUTLOC1__MASK			0x0000ff00
+#define A3XX_SP_VS_VPC_DST_REG_OUTLOC1__MASK			0x00007f00
 #define A3XX_SP_VS_VPC_DST_REG_OUTLOC1__SHIFT			8
 static inline uint32_t A3XX_SP_VS_VPC_DST_REG_OUTLOC1(uint32_t val)
 {
 	return ((val) << A3XX_SP_VS_VPC_DST_REG_OUTLOC1__SHIFT) & A3XX_SP_VS_VPC_DST_REG_OUTLOC1__MASK;
 }
-#define A3XX_SP_VS_VPC_DST_REG_OUTLOC2__MASK			0x00ff0000
+#define A3XX_SP_VS_VPC_DST_REG_OUTLOC2__MASK			0x007f0000
 #define A3XX_SP_VS_VPC_DST_REG_OUTLOC2__SHIFT			16
 static inline uint32_t A3XX_SP_VS_VPC_DST_REG_OUTLOC2(uint32_t val)
 {
 	return ((val) << A3XX_SP_VS_VPC_DST_REG_OUTLOC2__SHIFT) & A3XX_SP_VS_VPC_DST_REG_OUTLOC2__MASK;
 }
-#define A3XX_SP_VS_VPC_DST_REG_OUTLOC3__MASK			0xff000000
+#define A3XX_SP_VS_VPC_DST_REG_OUTLOC3__MASK			0x7f000000
 #define A3XX_SP_VS_VPC_DST_REG_OUTLOC3__SHIFT			24
 static inline uint32_t A3XX_SP_VS_VPC_DST_REG_OUTLOC3(uint32_t val)
 {
@@ -2139,6 +2432,12 @@ static inline uint32_t A3XX_SP_VS_VPC_DST_REG_OUTLOC3(uint32_t val)
 }
 
 #define REG_A3XX_SP_VS_OBJ_OFFSET_REG				0x000022d4
+#define A3XX_SP_VS_OBJ_OFFSET_REG_FIRSTEXECINSTROFFSET__MASK	0x0000ffff
+#define A3XX_SP_VS_OBJ_OFFSET_REG_FIRSTEXECINSTROFFSET__SHIFT	0
+static inline uint32_t A3XX_SP_VS_OBJ_OFFSET_REG_FIRSTEXECINSTROFFSET(uint32_t val)
+{
+	return ((val) << A3XX_SP_VS_OBJ_OFFSET_REG_FIRSTEXECINSTROFFSET__SHIFT) & A3XX_SP_VS_OBJ_OFFSET_REG_FIRSTEXECINSTROFFSET__MASK;
+}
 #define A3XX_SP_VS_OBJ_OFFSET_REG_CONSTOBJECTOFFSET__MASK	0x01ff0000
 #define A3XX_SP_VS_OBJ_OFFSET_REG_CONSTOBJECTOFFSET__SHIFT	16
 static inline uint32_t A3XX_SP_VS_OBJ_OFFSET_REG_CONSTOBJECTOFFSET(uint32_t val)
@@ -2155,8 +2454,38 @@ static inline uint32_t A3XX_SP_VS_OBJ_OFFSET_REG_SHADEROBJOFFSET(uint32_t val)
 #define REG_A3XX_SP_VS_OBJ_START_REG				0x000022d5
 
 #define REG_A3XX_SP_VS_PVT_MEM_PARAM_REG			0x000022d6
+#define A3XX_SP_VS_PVT_MEM_PARAM_REG_MEMSIZEPERITEM__MASK	0x000000ff
+#define A3XX_SP_VS_PVT_MEM_PARAM_REG_MEMSIZEPERITEM__SHIFT	0
+static inline uint32_t A3XX_SP_VS_PVT_MEM_PARAM_REG_MEMSIZEPERITEM(uint32_t val)
+{
+	return ((val) << A3XX_SP_VS_PVT_MEM_PARAM_REG_MEMSIZEPERITEM__SHIFT) & A3XX_SP_VS_PVT_MEM_PARAM_REG_MEMSIZEPERITEM__MASK;
+}
+#define A3XX_SP_VS_PVT_MEM_PARAM_REG_HWSTACKOFFSET__MASK	0x00ffff00
+#define A3XX_SP_VS_PVT_MEM_PARAM_REG_HWSTACKOFFSET__SHIFT	8
+static inline uint32_t A3XX_SP_VS_PVT_MEM_PARAM_REG_HWSTACKOFFSET(uint32_t val)
+{
+	return ((val) << A3XX_SP_VS_PVT_MEM_PARAM_REG_HWSTACKOFFSET__SHIFT) & A3XX_SP_VS_PVT_MEM_PARAM_REG_HWSTACKOFFSET__MASK;
+}
+#define A3XX_SP_VS_PVT_MEM_PARAM_REG_HWSTACKSIZEPERTHREAD__MASK	0xff000000
+#define A3XX_SP_VS_PVT_MEM_PARAM_REG_HWSTACKSIZEPERTHREAD__SHIFT	24
+static inline uint32_t A3XX_SP_VS_PVT_MEM_PARAM_REG_HWSTACKSIZEPERTHREAD(uint32_t val)
+{
+	return ((val) << A3XX_SP_VS_PVT_MEM_PARAM_REG_HWSTACKSIZEPERTHREAD__SHIFT) & A3XX_SP_VS_PVT_MEM_PARAM_REG_HWSTACKSIZEPERTHREAD__MASK;
+}
 
 #define REG_A3XX_SP_VS_PVT_MEM_ADDR_REG				0x000022d7
+#define A3XX_SP_VS_PVT_MEM_ADDR_REG_BURSTLEN__MASK		0x0000001f
+#define A3XX_SP_VS_PVT_MEM_ADDR_REG_BURSTLEN__SHIFT		0
+static inline uint32_t A3XX_SP_VS_PVT_MEM_ADDR_REG_BURSTLEN(uint32_t val)
+{
+	return ((val) << A3XX_SP_VS_PVT_MEM_ADDR_REG_BURSTLEN__SHIFT) & A3XX_SP_VS_PVT_MEM_ADDR_REG_BURSTLEN__MASK;
+}
+#define A3XX_SP_VS_PVT_MEM_ADDR_REG_SHADERSTARTADDRESS__MASK	0xffffffe0
+#define A3XX_SP_VS_PVT_MEM_ADDR_REG_SHADERSTARTADDRESS__SHIFT	5
+static inline uint32_t A3XX_SP_VS_PVT_MEM_ADDR_REG_SHADERSTARTADDRESS(uint32_t val)
+{
+	return ((val >> 5) << A3XX_SP_VS_PVT_MEM_ADDR_REG_SHADERSTARTADDRESS__SHIFT) & A3XX_SP_VS_PVT_MEM_ADDR_REG_SHADERSTARTADDRESS__MASK;
+}
 
 #define REG_A3XX_SP_VS_PVT_MEM_SIZE_REG				0x000022d8
 
@@ -2182,24 +2511,22 @@ static inline uint32_t A3XX_SP_FS_CTRL_REG0_INSTRBUFFERMODE(enum a3xx_instrbuffe
 	return ((val) << A3XX_SP_FS_CTRL_REG0_INSTRBUFFERMODE__SHIFT) & A3XX_SP_FS_CTRL_REG0_INSTRBUFFERMODE__MASK;
 }
 #define A3XX_SP_FS_CTRL_REG0_CACHEINVALID			0x00000004
+#define A3XX_SP_FS_CTRL_REG0_ALUSCHMODE				0x00000008
 #define A3XX_SP_FS_CTRL_REG0_HALFREGFOOTPRINT__MASK		0x000003f0
 #define A3XX_SP_FS_CTRL_REG0_HALFREGFOOTPRINT__SHIFT		4
 static inline uint32_t A3XX_SP_FS_CTRL_REG0_HALFREGFOOTPRINT(uint32_t val)
 {
 	return ((val) << A3XX_SP_FS_CTRL_REG0_HALFREGFOOTPRINT__SHIFT) & A3XX_SP_FS_CTRL_REG0_HALFREGFOOTPRINT__MASK;
 }
-#define A3XX_SP_FS_CTRL_REG0_FULLREGFOOTPRINT__MASK		0x0003fc00
+#define A3XX_SP_FS_CTRL_REG0_FULLREGFOOTPRINT__MASK		0x0000fc00
 #define A3XX_SP_FS_CTRL_REG0_FULLREGFOOTPRINT__SHIFT		10
 static inline uint32_t A3XX_SP_FS_CTRL_REG0_FULLREGFOOTPRINT(uint32_t val)
 {
 	return ((val) << A3XX_SP_FS_CTRL_REG0_FULLREGFOOTPRINT__SHIFT) & A3XX_SP_FS_CTRL_REG0_FULLREGFOOTPRINT__MASK;
 }
-#define A3XX_SP_FS_CTRL_REG0_INOUTREGOVERLAP__MASK		0x000c0000
-#define A3XX_SP_FS_CTRL_REG0_INOUTREGOVERLAP__SHIFT		18
-static inline uint32_t A3XX_SP_FS_CTRL_REG0_INOUTREGOVERLAP(uint32_t val)
-{
-	return ((val) << A3XX_SP_FS_CTRL_REG0_INOUTREGOVERLAP__SHIFT) & A3XX_SP_FS_CTRL_REG0_INOUTREGOVERLAP__MASK;
-}
+#define A3XX_SP_FS_CTRL_REG0_FSBYPASSENABLE			0x00020000
+#define A3XX_SP_FS_CTRL_REG0_INOUTREGOVERLAP			0x00040000
+#define A3XX_SP_FS_CTRL_REG0_OUTORDERED				0x00080000
 #define A3XX_SP_FS_CTRL_REG0_THREADSIZE__MASK			0x00100000
 #define A3XX_SP_FS_CTRL_REG0_THREADSIZE__SHIFT			20
 static inline uint32_t A3XX_SP_FS_CTRL_REG0_THREADSIZE(enum a3xx_threadsize val)
@@ -2235,7 +2562,7 @@ static inline uint32_t A3XX_SP_FS_CTRL_REG1_INITIALOUTSTANDING(uint32_t val)
 {
 	return ((val) << A3XX_SP_FS_CTRL_REG1_INITIALOUTSTANDING__SHIFT) & A3XX_SP_FS_CTRL_REG1_INITIALOUTSTANDING__MASK;
 }
-#define A3XX_SP_FS_CTRL_REG1_HALFPRECVAROFFSET__MASK		0x3f000000
+#define A3XX_SP_FS_CTRL_REG1_HALFPRECVAROFFSET__MASK		0x7f000000
 #define A3XX_SP_FS_CTRL_REG1_HALFPRECVAROFFSET__SHIFT		24
 static inline uint32_t A3XX_SP_FS_CTRL_REG1_HALFPRECVAROFFSET(uint32_t val)
 {
@@ -2243,6 +2570,12 @@ static inline uint32_t A3XX_SP_FS_CTRL_REG1_HALFPRECVAROFFSET(uint32_t val)
 }
 
 #define REG_A3XX_SP_FS_OBJ_OFFSET_REG				0x000022e2
+#define A3XX_SP_FS_OBJ_OFFSET_REG_FIRSTEXECINSTROFFSET__MASK	0x0000ffff
+#define A3XX_SP_FS_OBJ_OFFSET_REG_FIRSTEXECINSTROFFSET__SHIFT	0
+static inline uint32_t A3XX_SP_FS_OBJ_OFFSET_REG_FIRSTEXECINSTROFFSET(uint32_t val)
+{
+	return ((val) << A3XX_SP_FS_OBJ_OFFSET_REG_FIRSTEXECINSTROFFSET__SHIFT) & A3XX_SP_FS_OBJ_OFFSET_REG_FIRSTEXECINSTROFFSET__MASK;
+}
 #define A3XX_SP_FS_OBJ_OFFSET_REG_CONSTOBJECTOFFSET__MASK	0x01ff0000
 #define A3XX_SP_FS_OBJ_OFFSET_REG_CONSTOBJECTOFFSET__SHIFT	16
 static inline uint32_t A3XX_SP_FS_OBJ_OFFSET_REG_CONSTOBJECTOFFSET(uint32_t val)
@@ -2259,8 +2592,38 @@ static inline uint32_t A3XX_SP_FS_OBJ_OFFSET_REG_SHADEROBJOFFSET(uint32_t val)
 #define REG_A3XX_SP_FS_OBJ_START_REG				0x000022e3
 
 #define REG_A3XX_SP_FS_PVT_MEM_PARAM_REG			0x000022e4
+#define A3XX_SP_FS_PVT_MEM_PARAM_REG_MEMSIZEPERITEM__MASK	0x000000ff
+#define A3XX_SP_FS_PVT_MEM_PARAM_REG_MEMSIZEPERITEM__SHIFT	0
+static inline uint32_t A3XX_SP_FS_PVT_MEM_PARAM_REG_MEMSIZEPERITEM(uint32_t val)
+{
+	return ((val) << A3XX_SP_FS_PVT_MEM_PARAM_REG_MEMSIZEPERITEM__SHIFT) & A3XX_SP_FS_PVT_MEM_PARAM_REG_MEMSIZEPERITEM__MASK;
+}
+#define A3XX_SP_FS_PVT_MEM_PARAM_REG_HWSTACKOFFSET__MASK	0x00ffff00
+#define A3XX_SP_FS_PVT_MEM_PARAM_REG_HWSTACKOFFSET__SHIFT	8
+static inline uint32_t A3XX_SP_FS_PVT_MEM_PARAM_REG_HWSTACKOFFSET(uint32_t val)
+{
+	return ((val) << A3XX_SP_FS_PVT_MEM_PARAM_REG_HWSTACKOFFSET__SHIFT) & A3XX_SP_FS_PVT_MEM_PARAM_REG_HWSTACKOFFSET__MASK;
+}
+#define A3XX_SP_FS_PVT_MEM_PARAM_REG_HWSTACKSIZEPERTHREAD__MASK	0xff000000
+#define A3XX_SP_FS_PVT_MEM_PARAM_REG_HWSTACKSIZEPERTHREAD__SHIFT	24
+static inline uint32_t A3XX_SP_FS_PVT_MEM_PARAM_REG_HWSTACKSIZEPERTHREAD(uint32_t val)
+{
+	return ((val) << A3XX_SP_FS_PVT_MEM_PARAM_REG_HWSTACKSIZEPERTHREAD__SHIFT) & A3XX_SP_FS_PVT_MEM_PARAM_REG_HWSTACKSIZEPERTHREAD__MASK;
+}
 
 #define REG_A3XX_SP_FS_PVT_MEM_ADDR_REG				0x000022e5
+#define A3XX_SP_FS_PVT_MEM_ADDR_REG_BURSTLEN__MASK		0x0000001f
+#define A3XX_SP_FS_PVT_MEM_ADDR_REG_BURSTLEN__SHIFT		0
+static inline uint32_t A3XX_SP_FS_PVT_MEM_ADDR_REG_BURSTLEN(uint32_t val)
+{
+	return ((val) << A3XX_SP_FS_PVT_MEM_ADDR_REG_BURSTLEN__SHIFT) & A3XX_SP_FS_PVT_MEM_ADDR_REG_BURSTLEN__MASK;
+}
+#define A3XX_SP_FS_PVT_MEM_ADDR_REG_SHADERSTARTADDRESS__MASK	0xffffffe0
+#define A3XX_SP_FS_PVT_MEM_ADDR_REG_SHADERSTARTADDRESS__SHIFT	5
+static inline uint32_t A3XX_SP_FS_PVT_MEM_ADDR_REG_SHADERSTARTADDRESS(uint32_t val)
+{
+	return ((val >> 5) << A3XX_SP_FS_PVT_MEM_ADDR_REG_SHADERSTARTADDRESS__SHIFT) & A3XX_SP_FS_PVT_MEM_ADDR_REG_SHADERSTARTADDRESS__MASK;
+}
 
 #define REG_A3XX_SP_FS_PVT_MEM_SIZE_REG				0x000022e6
 
diff --git a/drivers/gpu/drm/msm/adreno/a4xx.xml.h b/drivers/gpu/drm/msm/adreno/a4xx.xml.h
index 99de8271dba8..3220b91f559a 100644
--- a/drivers/gpu/drm/msm/adreno/a4xx.xml.h
+++ b/drivers/gpu/drm/msm/adreno/a4xx.xml.h
@@ -9,16 +9,17 @@ git clone https://github.com/freedreno/envytools.git
 
 The rules-ng-ng source files this header was generated from are:
 - /home/robclark/src/freedreno/envytools/rnndb/adreno.xml               (    398 bytes, from 2015-09-24 17:25:31)
-- /home/robclark/src/freedreno/envytools/rnndb/freedreno_copyright.xml  (   1453 bytes, from 2015-05-20 20:03:07)
+- /home/robclark/src/freedreno/envytools/rnndb/freedreno_copyright.xml  (   1572 bytes, from 2016-02-10 17:07:21)
 - /home/robclark/src/freedreno/envytools/rnndb/adreno/a2xx.xml          (  32901 bytes, from 2015-05-20 20:03:14)
-- /home/robclark/src/freedreno/envytools/rnndb/adreno/adreno_common.xml (  10755 bytes, from 2015-09-14 20:46:55)
-- /home/robclark/src/freedreno/envytools/rnndb/adreno/adreno_pm4.xml    (  14968 bytes, from 2015-05-20 20:12:27)
-- /home/robclark/src/freedreno/envytools/rnndb/adreno/a3xx.xml          (  67771 bytes, from 2015-09-14 20:46:55)
-- /home/robclark/src/freedreno/envytools/rnndb/adreno/a4xx.xml          (  63970 bytes, from 2015-09-14 20:50:12)
+- /home/robclark/src/freedreno/envytools/rnndb/adreno/adreno_common.xml (  11518 bytes, from 2016-02-10 21:03:25)
+- /home/robclark/src/freedreno/envytools/rnndb/adreno/adreno_pm4.xml    (  16166 bytes, from 2016-02-11 21:20:31)
+- /home/robclark/src/freedreno/envytools/rnndb/adreno/a3xx.xml          (  83967 bytes, from 2016-02-10 17:07:21)
+- /home/robclark/src/freedreno/envytools/rnndb/adreno/a4xx.xml          ( 109916 bytes, from 2016-02-20 18:44:48)
 - /home/robclark/src/freedreno/envytools/rnndb/adreno/ocmem.xml         (   1773 bytes, from 2015-09-24 17:30:00)
 
-Copyright (C) 2013-2015 by the following authors:
+Copyright (C) 2013-2016 by the following authors:
 - Rob Clark <robdclark@gmail.com> (robclark)
+- Ilia Mirkin <imirkin@alum.mit.edu> (imirkin)
 
 Permission is hereby granted, free of charge, to any person obtaining
 a copy of this software and associated documentation files (the
@@ -47,11 +48,13 @@ enum a4xx_color_fmt {
 	RB4_R8_UNORM = 2,
 	RB4_R4G4B4A4_UNORM = 8,
 	RB4_R5G5B5A1_UNORM = 10,
-	RB4_R5G6R5_UNORM = 14,
+	RB4_R5G6B5_UNORM = 14,
 	RB4_R8G8_UNORM = 15,
 	RB4_R8G8_SNORM = 16,
 	RB4_R8G8_UINT = 17,
 	RB4_R8G8_SINT = 18,
+	RB4_R16_UNORM = 19,
+	RB4_R16_SNORM = 20,
 	RB4_R16_FLOAT = 21,
 	RB4_R16_UINT = 22,
 	RB4_R16_SINT = 23,
@@ -63,12 +66,16 @@ enum a4xx_color_fmt {
 	RB4_R10G10B10A2_UNORM = 31,
 	RB4_R10G10B10A2_UINT = 34,
 	RB4_R11G11B10_FLOAT = 39,
+	RB4_R16G16_UNORM = 40,
+	RB4_R16G16_SNORM = 41,
 	RB4_R16G16_FLOAT = 42,
 	RB4_R16G16_UINT = 43,
 	RB4_R16G16_SINT = 44,
 	RB4_R32_FLOAT = 45,
 	RB4_R32_UINT = 46,
 	RB4_R32_SINT = 47,
+	RB4_R16G16B16A16_UNORM = 52,
+	RB4_R16G16B16A16_SNORM = 53,
 	RB4_R16G16B16A16_FLOAT = 54,
 	RB4_R16G16B16A16_UINT = 55,
 	RB4_R16G16B16A16_SINT = 56,
@@ -106,6 +113,7 @@ enum a4xx_vtx_fmt {
 	VFMT4_32_32_FIXED = 10,
 	VFMT4_32_32_32_FIXED = 11,
 	VFMT4_32_32_32_32_FIXED = 12,
+	VFMT4_11_11_10_FLOAT = 13,
 	VFMT4_16_SINT = 16,
 	VFMT4_16_16_SINT = 17,
 	VFMT4_16_16_16_SINT = 18,
@@ -146,52 +154,76 @@ enum a4xx_vtx_fmt {
 	VFMT4_8_8_SNORM = 53,
 	VFMT4_8_8_8_SNORM = 54,
 	VFMT4_8_8_8_8_SNORM = 55,
-	VFMT4_10_10_10_2_UINT = 60,
-	VFMT4_10_10_10_2_UNORM = 61,
-	VFMT4_10_10_10_2_SINT = 62,
-	VFMT4_10_10_10_2_SNORM = 63,
+	VFMT4_10_10_10_2_UINT = 56,
+	VFMT4_10_10_10_2_UNORM = 57,
+	VFMT4_10_10_10_2_SINT = 58,
+	VFMT4_10_10_10_2_SNORM = 59,
+	VFMT4_2_10_10_10_UINT = 60,
+	VFMT4_2_10_10_10_UNORM = 61,
+	VFMT4_2_10_10_10_SINT = 62,
+	VFMT4_2_10_10_10_SNORM = 63,
 };
 
 enum a4xx_tex_fmt {
-	TFMT4_5_6_5_UNORM = 11,
-	TFMT4_5_5_5_1_UNORM = 10,
-	TFMT4_4_4_4_4_UNORM = 8,
-	TFMT4_X8Z24_UNORM = 71,
-	TFMT4_10_10_10_2_UNORM = 33,
 	TFMT4_A8_UNORM = 3,
-	TFMT4_L8_A8_UNORM = 13,
 	TFMT4_8_UNORM = 4,
-	TFMT4_8_8_UNORM = 14,
-	TFMT4_8_8_8_8_UNORM = 28,
 	TFMT4_8_SNORM = 5,
-	TFMT4_8_8_SNORM = 15,
-	TFMT4_8_8_8_8_SNORM = 29,
 	TFMT4_8_UINT = 6,
-	TFMT4_8_8_UINT = 16,
-	TFMT4_8_8_8_8_UINT = 30,
 	TFMT4_8_SINT = 7,
+	TFMT4_4_4_4_4_UNORM = 8,
+	TFMT4_5_5_5_1_UNORM = 9,
+	TFMT4_5_6_5_UNORM = 11,
+	TFMT4_L8_A8_UNORM = 13,
+	TFMT4_8_8_UNORM = 14,
+	TFMT4_8_8_SNORM = 15,
+	TFMT4_8_8_UINT = 16,
 	TFMT4_8_8_SINT = 17,
-	TFMT4_8_8_8_8_SINT = 31,
+	TFMT4_16_UNORM = 18,
+	TFMT4_16_SNORM = 19,
+	TFMT4_16_FLOAT = 20,
 	TFMT4_16_UINT = 21,
-	TFMT4_16_16_UINT = 41,
-	TFMT4_16_16_16_16_UINT = 54,
 	TFMT4_16_SINT = 22,
+	TFMT4_8_8_8_8_UNORM = 28,
+	TFMT4_8_8_8_8_SNORM = 29,
+	TFMT4_8_8_8_8_UINT = 30,
+	TFMT4_8_8_8_8_SINT = 31,
+	TFMT4_9_9_9_E5_FLOAT = 32,
+	TFMT4_10_10_10_2_UNORM = 33,
+	TFMT4_10_10_10_2_UINT = 34,
+	TFMT4_11_11_10_FLOAT = 37,
+	TFMT4_16_16_UNORM = 38,
+	TFMT4_16_16_SNORM = 39,
+	TFMT4_16_16_FLOAT = 40,
+	TFMT4_16_16_UINT = 41,
 	TFMT4_16_16_SINT = 42,
-	TFMT4_16_16_16_16_SINT = 55,
+	TFMT4_32_FLOAT = 43,
 	TFMT4_32_UINT = 44,
-	TFMT4_32_32_UINT = 57,
-	TFMT4_32_32_32_32_UINT = 64,
 	TFMT4_32_SINT = 45,
-	TFMT4_32_32_SINT = 58,
-	TFMT4_32_32_32_32_SINT = 65,
-	TFMT4_16_FLOAT = 20,
-	TFMT4_16_16_FLOAT = 40,
+	TFMT4_16_16_16_16_UNORM = 51,
+	TFMT4_16_16_16_16_SNORM = 52,
 	TFMT4_16_16_16_16_FLOAT = 53,
-	TFMT4_32_FLOAT = 43,
+	TFMT4_16_16_16_16_UINT = 54,
+	TFMT4_16_16_16_16_SINT = 55,
 	TFMT4_32_32_FLOAT = 56,
+	TFMT4_32_32_UINT = 57,
+	TFMT4_32_32_SINT = 58,
+	TFMT4_32_32_32_FLOAT = 59,
+	TFMT4_32_32_32_UINT = 60,
+	TFMT4_32_32_32_SINT = 61,
 	TFMT4_32_32_32_32_FLOAT = 63,
-	TFMT4_9_9_9_E5_FLOAT = 32,
-	TFMT4_11_11_10_FLOAT = 37,
+	TFMT4_32_32_32_32_UINT = 64,
+	TFMT4_32_32_32_32_SINT = 65,
+	TFMT4_X8Z24_UNORM = 71,
+	TFMT4_DXT1 = 86,
+	TFMT4_DXT3 = 87,
+	TFMT4_DXT5 = 88,
+	TFMT4_RGTC1_UNORM = 90,
+	TFMT4_RGTC1_SNORM = 91,
+	TFMT4_RGTC2_UNORM = 94,
+	TFMT4_RGTC2_SNORM = 95,
+	TFMT4_BPTC_UFLOAT = 97,
+	TFMT4_BPTC_FLOAT = 98,
+	TFMT4_BPTC = 99,
 	TFMT4_ATC_RGB = 100,
 	TFMT4_ATC_RGBA_EXPLICIT = 101,
 	TFMT4_ATC_RGBA_INTERPOLATED = 102,
@@ -240,6 +272,545 @@ enum a4xx_tess_spacing {
 	EVEN_SPACING = 3,
 };
 
+enum a4xx_ccu_perfcounter_select {
+	CCU_BUSY_CYCLES = 0,
+	CCU_RB_DEPTH_RETURN_STALL = 2,
+	CCU_RB_COLOR_RETURN_STALL = 3,
+	CCU_DEPTH_BLOCKS = 6,
+	CCU_COLOR_BLOCKS = 7,
+	CCU_DEPTH_BLOCK_HIT = 8,
+	CCU_COLOR_BLOCK_HIT = 9,
+	CCU_DEPTH_FLAG1_COUNT = 10,
+	CCU_DEPTH_FLAG2_COUNT = 11,
+	CCU_DEPTH_FLAG3_COUNT = 12,
+	CCU_DEPTH_FLAG4_COUNT = 13,
+	CCU_COLOR_FLAG1_COUNT = 14,
+	CCU_COLOR_FLAG2_COUNT = 15,
+	CCU_COLOR_FLAG3_COUNT = 16,
+	CCU_COLOR_FLAG4_COUNT = 17,
+	CCU_PARTIAL_BLOCK_READ = 18,
+};
+
+enum a4xx_cp_perfcounter_select {
+	CP_ALWAYS_COUNT = 0,
+	CP_BUSY = 1,
+	CP_PFP_IDLE = 2,
+	CP_PFP_BUSY_WORKING = 3,
+	CP_PFP_STALL_CYCLES_ANY = 4,
+	CP_PFP_STARVE_CYCLES_ANY = 5,
+	CP_PFP_STARVED_PER_LOAD_ADDR = 6,
+	CP_PFP_STALLED_PER_STORE_ADDR = 7,
+	CP_PFP_PC_PROFILE = 8,
+	CP_PFP_MATCH_PM4_PKT_PROFILE = 9,
+	CP_PFP_COND_INDIRECT_DISCARDED = 10,
+	CP_LONG_RESUMPTIONS = 11,
+	CP_RESUME_CYCLES = 12,
+	CP_RESUME_TO_BOUNDARY_CYCLES = 13,
+	CP_LONG_PREEMPTIONS = 14,
+	CP_PREEMPT_CYCLES = 15,
+	CP_PREEMPT_TO_BOUNDARY_CYCLES = 16,
+	CP_ME_FIFO_EMPTY_PFP_IDLE = 17,
+	CP_ME_FIFO_EMPTY_PFP_BUSY = 18,
+	CP_ME_FIFO_NOT_EMPTY_NOT_FULL = 19,
+	CP_ME_FIFO_FULL_ME_BUSY = 20,
+	CP_ME_FIFO_FULL_ME_NON_WORKING = 21,
+	CP_ME_WAITING_FOR_PACKETS = 22,
+	CP_ME_BUSY_WORKING = 23,
+	CP_ME_STARVE_CYCLES_ANY = 24,
+	CP_ME_STARVE_CYCLES_PER_PROFILE = 25,
+	CP_ME_STALL_CYCLES_PER_PROFILE = 26,
+	CP_ME_PC_PROFILE = 27,
+	CP_RCIU_FIFO_EMPTY = 28,
+	CP_RCIU_FIFO_NOT_EMPTY_NOT_FULL = 29,
+	CP_RCIU_FIFO_FULL = 30,
+	CP_RCIU_FIFO_FULL_NO_CONTEXT = 31,
+	CP_RCIU_FIFO_FULL_AHB_MASTER = 32,
+	CP_RCIU_FIFO_FULL_OTHER = 33,
+	CP_AHB_IDLE = 34,
+	CP_AHB_STALL_ON_GRANT_NO_SPLIT = 35,
+	CP_AHB_STALL_ON_GRANT_SPLIT = 36,
+	CP_AHB_STALL_ON_GRANT_SPLIT_PROFILE = 37,
+	CP_AHB_BUSY_WORKING = 38,
+	CP_AHB_BUSY_STALL_ON_HRDY = 39,
+	CP_AHB_BUSY_STALL_ON_HRDY_PROFILE = 40,
+};
+
+enum a4xx_gras_ras_perfcounter_select {
+	RAS_SUPER_TILES = 0,
+	RAS_8X8_TILES = 1,
+	RAS_4X4_TILES = 2,
+	RAS_BUSY_CYCLES = 3,
+	RAS_STALL_CYCLES_BY_RB = 4,
+	RAS_STALL_CYCLES_BY_VSC = 5,
+	RAS_STARVE_CYCLES_BY_TSE = 6,
+	RAS_SUPERTILE_CYCLES = 7,
+	RAS_TILE_CYCLES = 8,
+	RAS_FULLY_COVERED_SUPER_TILES = 9,
+	RAS_FULLY_COVERED_8X8_TILES = 10,
+	RAS_4X4_PRIM = 11,
+	RAS_8X4_4X8_PRIM = 12,
+	RAS_8X8_PRIM = 13,
+};
+
+enum a4xx_gras_tse_perfcounter_select {
+	TSE_INPUT_PRIM = 0,
+	TSE_INPUT_NULL_PRIM = 1,
+	TSE_TRIVAL_REJ_PRIM = 2,
+	TSE_CLIPPED_PRIM = 3,
+	TSE_NEW_PRIM = 4,
+	TSE_ZERO_AREA_PRIM = 5,
+	TSE_FACENESS_CULLED_PRIM = 6,
+	TSE_ZERO_PIXEL_PRIM = 7,
+	TSE_OUTPUT_NULL_PRIM = 8,
+	TSE_OUTPUT_VISIBLE_PRIM = 9,
+	TSE_PRE_CLIP_PRIM = 10,
+	TSE_POST_CLIP_PRIM = 11,
+	TSE_BUSY_CYCLES = 12,
+	TSE_PC_STARVE = 13,
+	TSE_RAS_STALL = 14,
+	TSE_STALL_BARYPLANE_FIFO_FULL = 15,
+	TSE_STALL_ZPLANE_FIFO_FULL = 16,
+};
+
+enum a4xx_hlsq_perfcounter_select {
+	HLSQ_SP_VS_STAGE_CONSTANT = 0,
+	HLSQ_SP_VS_STAGE_INSTRUCTIONS = 1,
+	HLSQ_SP_FS_STAGE_CONSTANT = 2,
+	HLSQ_SP_FS_STAGE_INSTRUCTIONS = 3,
+	HLSQ_TP_STATE = 4,
+	HLSQ_QUADS = 5,
+	HLSQ_PIXELS = 6,
+	HLSQ_VERTICES = 7,
+	HLSQ_SP_VS_STAGE_DATA_BYTES = 13,
+	HLSQ_SP_FS_STAGE_DATA_BYTES = 14,
+	HLSQ_BUSY_CYCLES = 15,
+	HLSQ_STALL_CYCLES_SP_STATE = 16,
+	HLSQ_STALL_CYCLES_SP_VS_STAGE = 17,
+	HLSQ_STALL_CYCLES_SP_FS_STAGE = 18,
+	HLSQ_STALL_CYCLES_UCHE = 19,
+	HLSQ_RBBM_LOAD_CYCLES = 20,
+	HLSQ_DI_TO_VS_START_SP = 21,
+	HLSQ_DI_TO_FS_START_SP = 22,
+	HLSQ_VS_STAGE_START_TO_DONE_SP = 23,
+	HLSQ_FS_STAGE_START_TO_DONE_SP = 24,
+	HLSQ_SP_STATE_COPY_CYCLES_VS_STAGE = 25,
+	HLSQ_SP_STATE_COPY_CYCLES_FS_STAGE = 26,
+	HLSQ_UCHE_LATENCY_CYCLES = 27,
+	HLSQ_UCHE_LATENCY_COUNT = 28,
+	HLSQ_STARVE_CYCLES_VFD = 29,
+};
+
+enum a4xx_pc_perfcounter_select {
+	PC_VIS_STREAMS_LOADED = 0,
+	PC_VPC_PRIMITIVES = 2,
+	PC_DEAD_PRIM = 3,
+	PC_LIVE_PRIM = 4,
+	PC_DEAD_DRAWCALLS = 5,
+	PC_LIVE_DRAWCALLS = 6,
+	PC_VERTEX_MISSES = 7,
+	PC_STALL_CYCLES_VFD = 9,
+	PC_STALL_CYCLES_TSE = 10,
+	PC_STALL_CYCLES_UCHE = 11,
+	PC_WORKING_CYCLES = 12,
+	PC_IA_VERTICES = 13,
+	PC_GS_PRIMITIVES = 14,
+	PC_HS_INVOCATIONS = 15,
+	PC_DS_INVOCATIONS = 16,
+	PC_DS_PRIMITIVES = 17,
+	PC_STARVE_CYCLES_FOR_INDEX = 20,
+	PC_STARVE_CYCLES_FOR_TESS_FACTOR = 21,
+	PC_STARVE_CYCLES_FOR_VIZ_STREAM = 22,
+	PC_STALL_CYCLES_TESS = 23,
+	PC_STARVE_CYCLES_FOR_POSITION = 24,
+	PC_MODE0_DRAWCALL = 25,
+	PC_MODE1_DRAWCALL = 26,
+	PC_MODE2_DRAWCALL = 27,
+	PC_MODE3_DRAWCALL = 28,
+	PC_MODE4_DRAWCALL = 29,
+	PC_PREDICATED_DEAD_DRAWCALL = 30,
+	PC_STALL_CYCLES_BY_TSE_ONLY = 31,
+	PC_STALL_CYCLES_BY_VPC_ONLY = 32,
+	PC_VPC_POS_DATA_TRANSACTION = 33,
+	PC_BUSY_CYCLES = 34,
+	PC_STARVE_CYCLES_DI = 35,
+	PC_STALL_CYCLES_VPC = 36,
+	TESS_WORKING_CYCLES = 37,
+	TESS_NUM_CYCLES_SETUP_WORKING = 38,
+	TESS_NUM_CYCLES_PTGEN_WORKING = 39,
+	TESS_NUM_CYCLES_CONNGEN_WORKING = 40,
+	TESS_BUSY_CYCLES = 41,
+	TESS_STARVE_CYCLES_PC = 42,
+	TESS_STALL_CYCLES_PC = 43,
+};
+
+enum a4xx_pwr_perfcounter_select {
+	PWR_CORE_CLOCK_CYCLES = 0,
+	PWR_BUSY_CLOCK_CYCLES = 1,
+};
+
+enum a4xx_rb_perfcounter_select {
+	RB_BUSY_CYCLES = 0,
+	RB_BUSY_CYCLES_BINNING = 1,
+	RB_BUSY_CYCLES_RENDERING = 2,
+	RB_BUSY_CYCLES_RESOLVE = 3,
+	RB_STARVE_CYCLES_BY_SP = 4,
+	RB_STARVE_CYCLES_BY_RAS = 5,
+	RB_STARVE_CYCLES_BY_MARB = 6,
+	RB_STALL_CYCLES_BY_MARB = 7,
+	RB_STALL_CYCLES_BY_HLSQ = 8,
+	RB_RB_RB_MARB_DATA = 9,
+	RB_SP_RB_QUAD = 10,
+	RB_RAS_RB_Z_QUADS = 11,
+	RB_GMEM_CH0_READ = 12,
+	RB_GMEM_CH1_READ = 13,
+	RB_GMEM_CH0_WRITE = 14,
+	RB_GMEM_CH1_WRITE = 15,
+	RB_CP_CONTEXT_DONE = 16,
+	RB_CP_CACHE_FLUSH = 17,
+	RB_CP_ZPASS_DONE = 18,
+	RB_STALL_FIFO0_FULL = 19,
+	RB_STALL_FIFO1_FULL = 20,
+	RB_STALL_FIFO2_FULL = 21,
+	RB_STALL_FIFO3_FULL = 22,
+	RB_RB_HLSQ_TRANSACTIONS = 23,
+	RB_Z_READ = 24,
+	RB_Z_WRITE = 25,
+	RB_C_READ = 26,
+	RB_C_WRITE = 27,
+	RB_C_READ_LATENCY = 28,
+	RB_Z_READ_LATENCY = 29,
+	RB_STALL_BY_UCHE = 30,
+	RB_MARB_UCHE_TRANSACTIONS = 31,
+	RB_CACHE_STALL_MISS = 32,
+	RB_CACHE_STALL_FIFO_FULL = 33,
+	RB_8BIT_BLENDER_UNITS_ACTIVE = 34,
+	RB_16BIT_BLENDER_UNITS_ACTIVE = 35,
+	RB_SAMPLER_UNITS_ACTIVE = 36,
+	RB_TOTAL_PASS = 38,
+	RB_Z_PASS = 39,
+	RB_Z_FAIL = 40,
+	RB_S_FAIL = 41,
+	RB_POWER0 = 42,
+	RB_POWER1 = 43,
+	RB_POWER2 = 44,
+	RB_POWER3 = 45,
+	RB_POWER4 = 46,
+	RB_POWER5 = 47,
+	RB_POWER6 = 48,
+	RB_POWER7 = 49,
+};
+
+enum a4xx_rbbm_perfcounter_select {
+	RBBM_ALWAYS_ON = 0,
+	RBBM_VBIF_BUSY = 1,
+	RBBM_TSE_BUSY = 2,
+	RBBM_RAS_BUSY = 3,
+	RBBM_PC_DCALL_BUSY = 4,
+	RBBM_PC_VSD_BUSY = 5,
+	RBBM_VFD_BUSY = 6,
+	RBBM_VPC_BUSY = 7,
+	RBBM_UCHE_BUSY = 8,
+	RBBM_VSC_BUSY = 9,
+	RBBM_HLSQ_BUSY = 10,
+	RBBM_ANY_RB_BUSY = 11,
+	RBBM_ANY_TPL1_BUSY = 12,
+	RBBM_ANY_SP_BUSY = 13,
+	RBBM_ANY_MARB_BUSY = 14,
+	RBBM_ANY_ARB_BUSY = 15,
+	RBBM_AHB_STATUS_BUSY = 16,
+	RBBM_AHB_STATUS_STALLED = 17,
+	RBBM_AHB_STATUS_TXFR = 18,
+	RBBM_AHB_STATUS_TXFR_SPLIT = 19,
+	RBBM_AHB_STATUS_TXFR_ERROR = 20,
+	RBBM_AHB_STATUS_LONG_STALL = 21,
+	RBBM_STATUS_MASKED = 22,
+	RBBM_CP_BUSY_GFX_CORE_IDLE = 23,
+	RBBM_TESS_BUSY = 24,
+	RBBM_COM_BUSY = 25,
+	RBBM_DCOM_BUSY = 32,
+	RBBM_ANY_CCU_BUSY = 33,
+	RBBM_DPM_BUSY = 34,
+};
+
+enum a4xx_sp_perfcounter_select {
+	SP_LM_LOAD_INSTRUCTIONS = 0,
+	SP_LM_STORE_INSTRUCTIONS = 1,
+	SP_LM_ATOMICS = 2,
+	SP_GM_LOAD_INSTRUCTIONS = 3,
+	SP_GM_STORE_INSTRUCTIONS = 4,
+	SP_GM_ATOMICS = 5,
+	SP_VS_STAGE_TEX_INSTRUCTIONS = 6,
+	SP_VS_STAGE_CFLOW_INSTRUCTIONS = 7,
+	SP_VS_STAGE_EFU_INSTRUCTIONS = 8,
+	SP_VS_STAGE_FULL_ALU_INSTRUCTIONS = 9,
+	SP_VS_STAGE_HALF_ALU_INSTRUCTIONS = 10,
+	SP_FS_STAGE_TEX_INSTRUCTIONS = 11,
+	SP_FS_STAGE_CFLOW_INSTRUCTIONS = 12,
+	SP_FS_STAGE_EFU_INSTRUCTIONS = 13,
+	SP_FS_STAGE_FULL_ALU_INSTRUCTIONS = 14,
+	SP_FS_STAGE_HALF_ALU_INSTRUCTIONS = 15,
+	SP_VS_INSTRUCTIONS = 17,
+	SP_FS_INSTRUCTIONS = 18,
+	SP_ADDR_LOCK_COUNT = 19,
+	SP_UCHE_READ_TRANS = 20,
+	SP_UCHE_WRITE_TRANS = 21,
+	SP_EXPORT_VPC_TRANS = 22,
+	SP_EXPORT_RB_TRANS = 23,
+	SP_PIXELS_KILLED = 24,
+	SP_ICL1_REQUESTS = 25,
+	SP_ICL1_MISSES = 26,
+	SP_ICL0_REQUESTS = 27,
+	SP_ICL0_MISSES = 28,
+	SP_ALU_WORKING_CYCLES = 29,
+	SP_EFU_WORKING_CYCLES = 30,
+	SP_STALL_CYCLES_BY_VPC = 31,
+	SP_STALL_CYCLES_BY_TP = 32,
+	SP_STALL_CYCLES_BY_UCHE = 33,
+	SP_STALL_CYCLES_BY_RB = 34,
+	SP_BUSY_CYCLES = 35,
+	SP_HS_INSTRUCTIONS = 36,
+	SP_DS_INSTRUCTIONS = 37,
+	SP_GS_INSTRUCTIONS = 38,
+	SP_CS_INSTRUCTIONS = 39,
+	SP_SCHEDULER_NON_WORKING = 40,
+	SP_WAVE_CONTEXTS = 41,
+	SP_WAVE_CONTEXT_CYCLES = 42,
+	SP_POWER0 = 43,
+	SP_POWER1 = 44,
+	SP_POWER2 = 45,
+	SP_POWER3 = 46,
+	SP_POWER4 = 47,
+	SP_POWER5 = 48,
+	SP_POWER6 = 49,
+	SP_POWER7 = 50,
+	SP_POWER8 = 51,
+	SP_POWER9 = 52,
+	SP_POWER10 = 53,
+	SP_POWER11 = 54,
+	SP_POWER12 = 55,
+	SP_POWER13 = 56,
+	SP_POWER14 = 57,
+	SP_POWER15 = 58,
+};
+
+enum a4xx_tp_perfcounter_select {
+	TP_L1_REQUESTS = 0,
+	TP_L1_MISSES = 1,
+	TP_QUADS_OFFSET = 8,
+	TP_QUAD_SHADOW = 9,
+	TP_QUADS_ARRAY = 10,
+	TP_QUADS_GRADIENT = 11,
+	TP_QUADS_1D2D = 12,
+	TP_QUADS_3DCUBE = 13,
+	TP_BUSY_CYCLES = 16,
+	TP_STALL_CYCLES_BY_ARB = 17,
+	TP_STATE_CACHE_REQUESTS = 20,
+	TP_STATE_CACHE_MISSES = 21,
+	TP_POWER0 = 22,
+	TP_POWER1 = 23,
+	TP_POWER2 = 24,
+	TP_POWER3 = 25,
+	TP_POWER4 = 26,
+	TP_POWER5 = 27,
+	TP_POWER6 = 28,
+	TP_POWER7 = 29,
+};
+
+enum a4xx_uche_perfcounter_select {
+	UCHE_VBIF_READ_BEATS_TP = 0,
+	UCHE_VBIF_READ_BEATS_VFD = 1,
+	UCHE_VBIF_READ_BEATS_HLSQ = 2,
+	UCHE_VBIF_READ_BEATS_MARB = 3,
+	UCHE_VBIF_READ_BEATS_SP = 4,
+	UCHE_READ_REQUESTS_TP = 5,
+	UCHE_READ_REQUESTS_VFD = 6,
+	UCHE_READ_REQUESTS_HLSQ = 7,
+	UCHE_READ_REQUESTS_MARB = 8,
+	UCHE_READ_REQUESTS_SP = 9,
+	UCHE_WRITE_REQUESTS_MARB = 10,
+	UCHE_WRITE_REQUESTS_SP = 11,
+	UCHE_TAG_CHECK_FAILS = 12,
+	UCHE_EVICTS = 13,
+	UCHE_FLUSHES = 14,
+	UCHE_VBIF_LATENCY_CYCLES = 15,
+	UCHE_VBIF_LATENCY_SAMPLES = 16,
+	UCHE_BUSY_CYCLES = 17,
+	UCHE_VBIF_READ_BEATS_PC = 18,
+	UCHE_READ_REQUESTS_PC = 19,
+	UCHE_WRITE_REQUESTS_VPC = 20,
+	UCHE_STALL_BY_VBIF = 21,
+	UCHE_WRITE_REQUESTS_VSC = 22,
+	UCHE_POWER0 = 23,
+	UCHE_POWER1 = 24,
+	UCHE_POWER2 = 25,
+	UCHE_POWER3 = 26,
+	UCHE_POWER4 = 27,
+	UCHE_POWER5 = 28,
+	UCHE_POWER6 = 29,
+	UCHE_POWER7 = 30,
+};
+
+enum a4xx_vbif_perfcounter_select {
+	AXI_READ_REQUESTS_ID_0 = 0,
+	AXI_READ_REQUESTS_ID_1 = 1,
+	AXI_READ_REQUESTS_ID_2 = 2,
+	AXI_READ_REQUESTS_ID_3 = 3,
+	AXI_READ_REQUESTS_ID_4 = 4,
+	AXI_READ_REQUESTS_ID_5 = 5,
+	AXI_READ_REQUESTS_ID_6 = 6,
+	AXI_READ_REQUESTS_ID_7 = 7,
+	AXI_READ_REQUESTS_ID_8 = 8,
+	AXI_READ_REQUESTS_ID_9 = 9,
+	AXI_READ_REQUESTS_ID_10 = 10,
+	AXI_READ_REQUESTS_ID_11 = 11,
+	AXI_READ_REQUESTS_ID_12 = 12,
+	AXI_READ_REQUESTS_ID_13 = 13,
+	AXI_READ_REQUESTS_ID_14 = 14,
+	AXI_READ_REQUESTS_ID_15 = 15,
+	AXI0_READ_REQUESTS_TOTAL = 16,
+	AXI1_READ_REQUESTS_TOTAL = 17,
+	AXI2_READ_REQUESTS_TOTAL = 18,
+	AXI3_READ_REQUESTS_TOTAL = 19,
+	AXI_READ_REQUESTS_TOTAL = 20,
+	AXI_WRITE_REQUESTS_ID_0 = 21,
+	AXI_WRITE_REQUESTS_ID_1 = 22,
+	AXI_WRITE_REQUESTS_ID_2 = 23,
+	AXI_WRITE_REQUESTS_ID_3 = 24,
+	AXI_WRITE_REQUESTS_ID_4 = 25,
+	AXI_WRITE_REQUESTS_ID_5 = 26,
+	AXI_WRITE_REQUESTS_ID_6 = 27,
+	AXI_WRITE_REQUESTS_ID_7 = 28,
+	AXI_WRITE_REQUESTS_ID_8 = 29,
+	AXI_WRITE_REQUESTS_ID_9 = 30,
+	AXI_WRITE_REQUESTS_ID_10 = 31,
+	AXI_WRITE_REQUESTS_ID_11 = 32,
+	AXI_WRITE_REQUESTS_ID_12 = 33,
+	AXI_WRITE_REQUESTS_ID_13 = 34,
+	AXI_WRITE_REQUESTS_ID_14 = 35,
+	AXI_WRITE_REQUESTS_ID_15 = 36,
+	AXI0_WRITE_REQUESTS_TOTAL = 37,
+	AXI1_WRITE_REQUESTS_TOTAL = 38,
+	AXI2_WRITE_REQUESTS_TOTAL = 39,
+	AXI3_WRITE_REQUESTS_TOTAL = 40,
+	AXI_WRITE_REQUESTS_TOTAL = 41,
+	AXI_TOTAL_REQUESTS = 42,
+	AXI_READ_DATA_BEATS_ID_0 = 43,
+	AXI_READ_DATA_BEATS_ID_1 = 44,
+	AXI_READ_DATA_BEATS_ID_2 = 45,
+	AXI_READ_DATA_BEATS_ID_3 = 46,
+	AXI_READ_DATA_BEATS_ID_4 = 47,
+	AXI_READ_DATA_BEATS_ID_5 = 48,
+	AXI_READ_DATA_BEATS_ID_6 = 49,
+	AXI_READ_DATA_BEATS_ID_7 = 50,
+	AXI_READ_DATA_BEATS_ID_8 = 51,
+	AXI_READ_DATA_BEATS_ID_9 = 52,
+	AXI_READ_DATA_BEATS_ID_10 = 53,
+	AXI_READ_DATA_BEATS_ID_11 = 54,
+	AXI_READ_DATA_BEATS_ID_12 = 55,
+	AXI_READ_DATA_BEATS_ID_13 = 56,
+	AXI_READ_DATA_BEATS_ID_14 = 57,
+	AXI_READ_DATA_BEATS_ID_15 = 58,
+	AXI0_READ_DATA_BEATS_TOTAL = 59,
+	AXI1_READ_DATA_BEATS_TOTAL = 60,
+	AXI2_READ_DATA_BEATS_TOTAL = 61,
+	AXI3_READ_DATA_BEATS_TOTAL = 62,
+	AXI_READ_DATA_BEATS_TOTAL = 63,
+	AXI_WRITE_DATA_BEATS_ID_0 = 64,
+	AXI_WRITE_DATA_BEATS_ID_1 = 65,
+	AXI_WRITE_DATA_BEATS_ID_2 = 66,
+	AXI_WRITE_DATA_BEATS_ID_3 = 67,
+	AXI_WRITE_DATA_BEATS_ID_4 = 68,
+	AXI_WRITE_DATA_BEATS_ID_5 = 69,
+	AXI_WRITE_DATA_BEATS_ID_6 = 70,
+	AXI_WRITE_DATA_BEATS_ID_7 = 71,
+	AXI_WRITE_DATA_BEATS_ID_8 = 72,
+	AXI_WRITE_DATA_BEATS_ID_9 = 73,
+	AXI_WRITE_DATA_BEATS_ID_10 = 74,
+	AXI_WRITE_DATA_BEATS_ID_11 = 75,
+	AXI_WRITE_DATA_BEATS_ID_12 = 76,
+	AXI_WRITE_DATA_BEATS_ID_13 = 77,
+	AXI_WRITE_DATA_BEATS_ID_14 = 78,
+	AXI_WRITE_DATA_BEATS_ID_15 = 79,
+	AXI0_WRITE_DATA_BEATS_TOTAL = 80,
+	AXI1_WRITE_DATA_BEATS_TOTAL = 81,
+	AXI2_WRITE_DATA_BEATS_TOTAL = 82,
+	AXI3_WRITE_DATA_BEATS_TOTAL = 83,
+	AXI_WRITE_DATA_BEATS_TOTAL = 84,
+	AXI_DATA_BEATS_TOTAL = 85,
+	CYCLES_HELD_OFF_ID_0 = 86,
+	CYCLES_HELD_OFF_ID_1 = 87,
+	CYCLES_HELD_OFF_ID_2 = 88,
+	CYCLES_HELD_OFF_ID_3 = 89,
+	CYCLES_HELD_OFF_ID_4 = 90,
+	CYCLES_HELD_OFF_ID_5 = 91,
+	CYCLES_HELD_OFF_ID_6 = 92,
+	CYCLES_HELD_OFF_ID_7 = 93,
+	CYCLES_HELD_OFF_ID_8 = 94,
+	CYCLES_HELD_OFF_ID_9 = 95,
+	CYCLES_HELD_OFF_ID_10 = 96,
+	CYCLES_HELD_OFF_ID_11 = 97,
+	CYCLES_HELD_OFF_ID_12 = 98,
+	CYCLES_HELD_OFF_ID_13 = 99,
+	CYCLES_HELD_OFF_ID_14 = 100,
+	CYCLES_HELD_OFF_ID_15 = 101,
+	AXI_READ_REQUEST_HELD_OFF = 102,
+	AXI_WRITE_REQUEST_HELD_OFF = 103,
+	AXI_REQUEST_HELD_OFF = 104,
+	AXI_WRITE_DATA_HELD_OFF = 105,
+	OCMEM_AXI_READ_REQUEST_HELD_OFF = 106,
+	OCMEM_AXI_WRITE_REQUEST_HELD_OFF = 107,
+	OCMEM_AXI_REQUEST_HELD_OFF = 108,
+	OCMEM_AXI_WRITE_DATA_HELD_OFF = 109,
+	ELAPSED_CYCLES_DDR = 110,
+	ELAPSED_CYCLES_OCMEM = 111,
+};
+
+enum a4xx_vfd_perfcounter_select {
+	VFD_UCHE_BYTE_FETCHED = 0,
+	VFD_UCHE_TRANS = 1,
+	VFD_FETCH_INSTRUCTIONS = 3,
+	VFD_BUSY_CYCLES = 5,
+	VFD_STALL_CYCLES_UCHE = 6,
+	VFD_STALL_CYCLES_HLSQ = 7,
+	VFD_STALL_CYCLES_VPC_BYPASS = 8,
+	VFD_STALL_CYCLES_VPC_ALLOC = 9,
+	VFD_MODE_0_FIBERS = 13,
+	VFD_MODE_1_FIBERS = 14,
+	VFD_MODE_2_FIBERS = 15,
+	VFD_MODE_3_FIBERS = 16,
+	VFD_MODE_4_FIBERS = 17,
+	VFD_BFIFO_STALL = 18,
+	VFD_NUM_VERTICES_TOTAL = 19,
+	VFD_PACKER_FULL = 20,
+	VFD_UCHE_REQUEST_FIFO_FULL = 21,
+	VFD_STARVE_CYCLES_PC = 22,
+	VFD_STARVE_CYCLES_UCHE = 23,
+};
+
+enum a4xx_vpc_perfcounter_select {
+	VPC_SP_LM_COMPONENTS = 2,
+	VPC_SP0_LM_BYTES = 3,
+	VPC_SP1_LM_BYTES = 4,
+	VPC_SP2_LM_BYTES = 5,
+	VPC_SP3_LM_BYTES = 6,
+	VPC_WORKING_CYCLES = 7,
+	VPC_STALL_CYCLES_LM = 8,
+	VPC_STARVE_CYCLES_RAS = 9,
+	VPC_STREAMOUT_CYCLES = 10,
+	VPC_UCHE_TRANSACTIONS = 12,
+	VPC_STALL_CYCLES_UCHE = 13,
+	VPC_BUSY_CYCLES = 14,
+	VPC_STARVE_CYCLES_SP = 15,
+};
+
+enum a4xx_vsc_perfcounter_select {
+	VSC_BUSY_CYCLES = 0,
+	VSC_WORKING_CYCLES = 1,
+	VSC_STALL_CYCLES_UCHE = 2,
+	VSC_STARVE_CYCLES_RAS = 3,
+	VSC_EOT_NUM = 4,
+};
+
 enum a4xx_tex_filter {
 	A4XX_TEX_NEAREST = 0,
 	A4XX_TEX_LINEAR = 1,
@@ -326,6 +897,12 @@ static inline uint32_t A4XX_CGC_HLSQ_EARLY_CYC(uint32_t val)
 
 #define REG_A4XX_RB_PERFCTR_RB_SEL_7				0x00000cce
 
+#define REG_A4XX_RB_PERFCTR_CCU_SEL_0				0x00000ccf
+
+#define REG_A4XX_RB_PERFCTR_CCU_SEL_1				0x00000cd0
+
+#define REG_A4XX_RB_PERFCTR_CCU_SEL_2				0x00000cd1
+
 #define REG_A4XX_RB_PERFCTR_CCU_SEL_3				0x00000cd2
 
 #define REG_A4XX_RB_FRAME_BUFFER_DIMENSION			0x00000ce0
@@ -400,8 +977,13 @@ static inline uint32_t REG_A4XX_RB_MRT_CONTROL(uint32_t i0) { return 0x000020a4
 #define A4XX_RB_MRT_CONTROL_READ_DEST_ENABLE			0x00000008
 #define A4XX_RB_MRT_CONTROL_BLEND				0x00000010
 #define A4XX_RB_MRT_CONTROL_BLEND2				0x00000020
-#define A4XX_RB_MRT_CONTROL_FASTCLEAR				0x00000400
-#define A4XX_RB_MRT_CONTROL_B11					0x00000800
+#define A4XX_RB_MRT_CONTROL_ROP_ENABLE				0x00000040
+#define A4XX_RB_MRT_CONTROL_ROP_CODE__MASK			0x00000f00
+#define A4XX_RB_MRT_CONTROL_ROP_CODE__SHIFT			8
+static inline uint32_t A4XX_RB_MRT_CONTROL_ROP_CODE(enum a3xx_rop_code val)
+{
+	return ((val) << A4XX_RB_MRT_CONTROL_ROP_CODE__SHIFT) & A4XX_RB_MRT_CONTROL_ROP_CODE__MASK;
+}
 #define A4XX_RB_MRT_CONTROL_COMPONENT_ENABLE__MASK		0x0f000000
 #define A4XX_RB_MRT_CONTROL_COMPONENT_ENABLE__SHIFT		24
 static inline uint32_t A4XX_RB_MRT_CONTROL_COMPONENT_ENABLE(uint32_t val)
@@ -490,8 +1072,8 @@ static inline uint32_t A4XX_RB_MRT_BLEND_CONTROL_ALPHA_DEST_FACTOR(enum adreno_r
 	return ((val) << A4XX_RB_MRT_BLEND_CONTROL_ALPHA_DEST_FACTOR__SHIFT) & A4XX_RB_MRT_BLEND_CONTROL_ALPHA_DEST_FACTOR__MASK;
 }
 
-#define REG_A4XX_RB_BLEND_RED					0x000020f3
-#define A4XX_RB_BLEND_RED_UINT__MASK				0x00007fff
+#define REG_A4XX_RB_BLEND_RED					0x000020f0
+#define A4XX_RB_BLEND_RED_UINT__MASK				0x0000ffff
 #define A4XX_RB_BLEND_RED_UINT__SHIFT				0
 static inline uint32_t A4XX_RB_BLEND_RED_UINT(uint32_t val)
 {
@@ -504,8 +1086,16 @@ static inline uint32_t A4XX_RB_BLEND_RED_FLOAT(float val)
 	return ((util_float_to_half(val)) << A4XX_RB_BLEND_RED_FLOAT__SHIFT) & A4XX_RB_BLEND_RED_FLOAT__MASK;
 }
 
-#define REG_A4XX_RB_BLEND_GREEN					0x000020f4
-#define A4XX_RB_BLEND_GREEN_UINT__MASK				0x00007fff
+#define REG_A4XX_RB_BLEND_RED_F32				0x000020f1
+#define A4XX_RB_BLEND_RED_F32__MASK				0xffffffff
+#define A4XX_RB_BLEND_RED_F32__SHIFT				0
+static inline uint32_t A4XX_RB_BLEND_RED_F32(float val)
+{
+	return ((fui(val)) << A4XX_RB_BLEND_RED_F32__SHIFT) & A4XX_RB_BLEND_RED_F32__MASK;
+}
+
+#define REG_A4XX_RB_BLEND_GREEN					0x000020f2
+#define A4XX_RB_BLEND_GREEN_UINT__MASK				0x0000ffff
 #define A4XX_RB_BLEND_GREEN_UINT__SHIFT				0
 static inline uint32_t A4XX_RB_BLEND_GREEN_UINT(uint32_t val)
 {
@@ -518,8 +1108,16 @@ static inline uint32_t A4XX_RB_BLEND_GREEN_FLOAT(float val)
 	return ((util_float_to_half(val)) << A4XX_RB_BLEND_GREEN_FLOAT__SHIFT) & A4XX_RB_BLEND_GREEN_FLOAT__MASK;
 }
 
-#define REG_A4XX_RB_BLEND_BLUE					0x000020f5
-#define A4XX_RB_BLEND_BLUE_UINT__MASK				0x00007fff
+#define REG_A4XX_RB_BLEND_GREEN_F32				0x000020f3
+#define A4XX_RB_BLEND_GREEN_F32__MASK				0xffffffff
+#define A4XX_RB_BLEND_GREEN_F32__SHIFT				0
+static inline uint32_t A4XX_RB_BLEND_GREEN_F32(float val)
+{
+	return ((fui(val)) << A4XX_RB_BLEND_GREEN_F32__SHIFT) & A4XX_RB_BLEND_GREEN_F32__MASK;
+}
+
+#define REG_A4XX_RB_BLEND_BLUE					0x000020f4
+#define A4XX_RB_BLEND_BLUE_UINT__MASK				0x0000ffff
 #define A4XX_RB_BLEND_BLUE_UINT__SHIFT				0
 static inline uint32_t A4XX_RB_BLEND_BLUE_UINT(uint32_t val)
 {
@@ -532,8 +1130,16 @@ static inline uint32_t A4XX_RB_BLEND_BLUE_FLOAT(float val)
 	return ((util_float_to_half(val)) << A4XX_RB_BLEND_BLUE_FLOAT__SHIFT) & A4XX_RB_BLEND_BLUE_FLOAT__MASK;
 }
 
+#define REG_A4XX_RB_BLEND_BLUE_F32				0x000020f5
+#define A4XX_RB_BLEND_BLUE_F32__MASK				0xffffffff
+#define A4XX_RB_BLEND_BLUE_F32__SHIFT				0
+static inline uint32_t A4XX_RB_BLEND_BLUE_F32(float val)
+{
+	return ((fui(val)) << A4XX_RB_BLEND_BLUE_F32__SHIFT) & A4XX_RB_BLEND_BLUE_F32__MASK;
+}
+
 #define REG_A4XX_RB_BLEND_ALPHA					0x000020f6
-#define A4XX_RB_BLEND_ALPHA_UINT__MASK				0x00007fff
+#define A4XX_RB_BLEND_ALPHA_UINT__MASK				0x0000ffff
 #define A4XX_RB_BLEND_ALPHA_UINT__SHIFT				0
 static inline uint32_t A4XX_RB_BLEND_ALPHA_UINT(uint32_t val)
 {
@@ -546,6 +1152,14 @@ static inline uint32_t A4XX_RB_BLEND_ALPHA_FLOAT(float val)
 	return ((util_float_to_half(val)) << A4XX_RB_BLEND_ALPHA_FLOAT__SHIFT) & A4XX_RB_BLEND_ALPHA_FLOAT__MASK;
 }
 
+#define REG_A4XX_RB_BLEND_ALPHA_F32				0x000020f7
+#define A4XX_RB_BLEND_ALPHA_F32__MASK				0xffffffff
+#define A4XX_RB_BLEND_ALPHA_F32__SHIFT				0
+static inline uint32_t A4XX_RB_BLEND_ALPHA_F32(float val)
+{
+	return ((fui(val)) << A4XX_RB_BLEND_ALPHA_F32__SHIFT) & A4XX_RB_BLEND_ALPHA_F32__MASK;
+}
+
 #define REG_A4XX_RB_ALPHA_CONTROL				0x000020f8
 #define A4XX_RB_ALPHA_CONTROL_ALPHA_REF__MASK			0x000000ff
 #define A4XX_RB_ALPHA_CONTROL_ALPHA_REF__SHIFT			0
@@ -568,7 +1182,7 @@ static inline uint32_t A4XX_RB_FS_OUTPUT_ENABLE_BLEND(uint32_t val)
 {
 	return ((val) << A4XX_RB_FS_OUTPUT_ENABLE_BLEND__SHIFT) & A4XX_RB_FS_OUTPUT_ENABLE_BLEND__MASK;
 }
-#define A4XX_RB_FS_OUTPUT_FAST_CLEAR				0x00000100
+#define A4XX_RB_FS_OUTPUT_INDEPENDENT_BLEND			0x00000100
 #define A4XX_RB_FS_OUTPUT_SAMPLE_MASK__MASK			0xffff0000
 #define A4XX_RB_FS_OUTPUT_SAMPLE_MASK__SHIFT			16
 static inline uint32_t A4XX_RB_FS_OUTPUT_SAMPLE_MASK(uint32_t val)
@@ -736,6 +1350,7 @@ static inline uint32_t A4XX_RB_DEPTH_CONTROL_ZFUNC(enum adreno_compare_func val)
 }
 #define A4XX_RB_DEPTH_CONTROL_BF_ENABLE				0x00000080
 #define A4XX_RB_DEPTH_CONTROL_EARLY_Z_DISABLE			0x00010000
+#define A4XX_RB_DEPTH_CONTROL_FORCE_FRAGZ_TO_FS			0x00020000
 #define A4XX_RB_DEPTH_CONTROL_Z_TEST_ENABLE			0x80000000
 
 #define REG_A4XX_RB_DEPTH_CLEAR					0x00002102
@@ -996,8 +1611,386 @@ static inline uint32_t REG_A4XX_RBBM_CLOCK_DELAY_TP_REG(uint32_t i0) { return 0x
 
 #define REG_A4XX_RBBM_CFG_DEBBUS_SEL_D				0x0000004d
 
+#define REG_A4XX_RBBM_POWER_CNTL_IP				0x00000098
+#define A4XX_RBBM_POWER_CNTL_IP_SW_COLLAPSE			0x00000001
+#define A4XX_RBBM_POWER_CNTL_IP_SP_TP_PWR_ON			0x00100000
+
 #define REG_A4XX_RBBM_PERFCTR_CP_0_LO				0x0000009c
 
+#define REG_A4XX_RBBM_PERFCTR_CP_0_HI				0x0000009d
+
+#define REG_A4XX_RBBM_PERFCTR_CP_1_LO				0x0000009e
+
+#define REG_A4XX_RBBM_PERFCTR_CP_1_HI				0x0000009f
+
+#define REG_A4XX_RBBM_PERFCTR_CP_2_LO				0x000000a0
+
+#define REG_A4XX_RBBM_PERFCTR_CP_2_HI				0x000000a1
+
+#define REG_A4XX_RBBM_PERFCTR_CP_3_LO				0x000000a2
+
+#define REG_A4XX_RBBM_PERFCTR_CP_3_HI				0x000000a3
+
+#define REG_A4XX_RBBM_PERFCTR_CP_4_LO				0x000000a4
+
+#define REG_A4XX_RBBM_PERFCTR_CP_4_HI				0x000000a5
+
+#define REG_A4XX_RBBM_PERFCTR_CP_5_LO				0x000000a6
+
+#define REG_A4XX_RBBM_PERFCTR_CP_5_HI				0x000000a7
+
+#define REG_A4XX_RBBM_PERFCTR_CP_6_LO				0x000000a8
+
+#define REG_A4XX_RBBM_PERFCTR_CP_6_HI				0x000000a9
+
+#define REG_A4XX_RBBM_PERFCTR_CP_7_LO				0x000000aa
+
+#define REG_A4XX_RBBM_PERFCTR_CP_7_HI				0x000000ab
+
+#define REG_A4XX_RBBM_PERFCTR_RBBM_0_LO				0x000000ac
+
+#define REG_A4XX_RBBM_PERFCTR_RBBM_0_HI				0x000000ad
+
+#define REG_A4XX_RBBM_PERFCTR_RBBM_1_LO				0x000000ae
+
+#define REG_A4XX_RBBM_PERFCTR_RBBM_1_HI				0x000000af
+
+#define REG_A4XX_RBBM_PERFCTR_RBBM_2_LO				0x000000b0
+
+#define REG_A4XX_RBBM_PERFCTR_RBBM_2_HI				0x000000b1
+
+#define REG_A4XX_RBBM_PERFCTR_RBBM_3_LO				0x000000b2
+
+#define REG_A4XX_RBBM_PERFCTR_RBBM_3_HI				0x000000b3
+
+#define REG_A4XX_RBBM_PERFCTR_PC_0_LO				0x000000b4
+
+#define REG_A4XX_RBBM_PERFCTR_PC_0_HI				0x000000b5
+
+#define REG_A4XX_RBBM_PERFCTR_PC_1_LO				0x000000b6
+
+#define REG_A4XX_RBBM_PERFCTR_PC_1_HI				0x000000b7
+
+#define REG_A4XX_RBBM_PERFCTR_PC_2_LO				0x000000b8
+
+#define REG_A4XX_RBBM_PERFCTR_PC_2_HI				0x000000b9
+
+#define REG_A4XX_RBBM_PERFCTR_PC_3_LO				0x000000ba
+
+#define REG_A4XX_RBBM_PERFCTR_PC_3_HI				0x000000bb
+
+#define REG_A4XX_RBBM_PERFCTR_PC_4_LO				0x000000bc
+
+#define REG_A4XX_RBBM_PERFCTR_PC_4_HI				0x000000bd
+
+#define REG_A4XX_RBBM_PERFCTR_PC_5_LO				0x000000be
+
+#define REG_A4XX_RBBM_PERFCTR_PC_5_HI				0x000000bf
+
+#define REG_A4XX_RBBM_PERFCTR_PC_6_LO				0x000000c0
+
+#define REG_A4XX_RBBM_PERFCTR_PC_6_HI				0x000000c1
+
+#define REG_A4XX_RBBM_PERFCTR_PC_7_LO				0x000000c2
+
+#define REG_A4XX_RBBM_PERFCTR_PC_7_HI				0x000000c3
+
+#define REG_A4XX_RBBM_PERFCTR_VFD_0_LO				0x000000c4
+
+#define REG_A4XX_RBBM_PERFCTR_VFD_0_HI				0x000000c5
+
+#define REG_A4XX_RBBM_PERFCTR_VFD_1_LO				0x000000c6
+
+#define REG_A4XX_RBBM_PERFCTR_VFD_1_HI				0x000000c7
+
+#define REG_A4XX_RBBM_PERFCTR_VFD_2_LO				0x000000c8
+
+#define REG_A4XX_RBBM_PERFCTR_VFD_2_HI				0x000000c9
+
+#define REG_A4XX_RBBM_PERFCTR_VFD_3_LO				0x000000ca
+
+#define REG_A4XX_RBBM_PERFCTR_VFD_3_HI				0x000000cb
+
+#define REG_A4XX_RBBM_PERFCTR_VFD_4_LO				0x000000cc
+
+#define REG_A4XX_RBBM_PERFCTR_VFD_4_HI				0x000000cd
+
+#define REG_A4XX_RBBM_PERFCTR_VFD_5_LO				0x000000ce
+
+#define REG_A4XX_RBBM_PERFCTR_VFD_5_HI				0x000000cf
+
+#define REG_A4XX_RBBM_PERFCTR_VFD_6_LO				0x000000d0
+
+#define REG_A4XX_RBBM_PERFCTR_VFD_6_HI				0x000000d1
+
+#define REG_A4XX_RBBM_PERFCTR_VFD_7_LO				0x000000d2
+
+#define REG_A4XX_RBBM_PERFCTR_VFD_7_HI				0x000000d3
+
+#define REG_A4XX_RBBM_PERFCTR_HLSQ_0_LO				0x000000d4
+
+#define REG_A4XX_RBBM_PERFCTR_HLSQ_0_HI				0x000000d5
+
+#define REG_A4XX_RBBM_PERFCTR_HLSQ_1_LO				0x000000d6
+
+#define REG_A4XX_RBBM_PERFCTR_HLSQ_1_HI				0x000000d7
+
+#define REG_A4XX_RBBM_PERFCTR_HLSQ_2_LO				0x000000d8
+
+#define REG_A4XX_RBBM_PERFCTR_HLSQ_2_HI				0x000000d9
+
+#define REG_A4XX_RBBM_PERFCTR_HLSQ_3_LO				0x000000da
+
+#define REG_A4XX_RBBM_PERFCTR_HLSQ_3_HI				0x000000db
+
+#define REG_A4XX_RBBM_PERFCTR_HLSQ_4_LO				0x000000dc
+
+#define REG_A4XX_RBBM_PERFCTR_HLSQ_4_HI				0x000000dd
+
+#define REG_A4XX_RBBM_PERFCTR_HLSQ_5_LO				0x000000de
+
+#define REG_A4XX_RBBM_PERFCTR_HLSQ_5_HI				0x000000df
+
+#define REG_A4XX_RBBM_PERFCTR_HLSQ_6_LO				0x000000e0
+
+#define REG_A4XX_RBBM_PERFCTR_HLSQ_6_HI				0x000000e1
+
+#define REG_A4XX_RBBM_PERFCTR_HLSQ_7_LO				0x000000e2
+
+#define REG_A4XX_RBBM_PERFCTR_HLSQ_7_HI				0x000000e3
+
+#define REG_A4XX_RBBM_PERFCTR_VPC_0_LO				0x000000e4
+
+#define REG_A4XX_RBBM_PERFCTR_VPC_0_HI				0x000000e5
+
+#define REG_A4XX_RBBM_PERFCTR_VPC_1_LO				0x000000e6
+
+#define REG_A4XX_RBBM_PERFCTR_VPC_1_HI				0x000000e7
+
+#define REG_A4XX_RBBM_PERFCTR_VPC_2_LO				0x000000e8
+
+#define REG_A4XX_RBBM_PERFCTR_VPC_2_HI				0x000000e9
+
+#define REG_A4XX_RBBM_PERFCTR_VPC_3_LO				0x000000ea
+
+#define REG_A4XX_RBBM_PERFCTR_VPC_3_HI				0x000000eb
+
+#define REG_A4XX_RBBM_PERFCTR_CCU_0_LO				0x000000ec
+
+#define REG_A4XX_RBBM_PERFCTR_CCU_0_HI				0x000000ed
+
+#define REG_A4XX_RBBM_PERFCTR_CCU_1_LO				0x000000ee
+
+#define REG_A4XX_RBBM_PERFCTR_CCU_1_HI				0x000000ef
+
+#define REG_A4XX_RBBM_PERFCTR_CCU_2_LO				0x000000f0
+
+#define REG_A4XX_RBBM_PERFCTR_CCU_2_HI				0x000000f1
+
+#define REG_A4XX_RBBM_PERFCTR_CCU_3_LO				0x000000f2
+
+#define REG_A4XX_RBBM_PERFCTR_CCU_3_HI				0x000000f3
+
+#define REG_A4XX_RBBM_PERFCTR_TSE_0_LO				0x000000f4
+
+#define REG_A4XX_RBBM_PERFCTR_TSE_0_HI				0x000000f5
+
+#define REG_A4XX_RBBM_PERFCTR_TSE_1_LO				0x000000f6
+
+#define REG_A4XX_RBBM_PERFCTR_TSE_1_HI				0x000000f7
+
+#define REG_A4XX_RBBM_PERFCTR_TSE_2_LO				0x000000f8
+
+#define REG_A4XX_RBBM_PERFCTR_TSE_2_HI				0x000000f9
+
+#define REG_A4XX_RBBM_PERFCTR_TSE_3_LO				0x000000fa
+
+#define REG_A4XX_RBBM_PERFCTR_TSE_3_HI				0x000000fb
+
+#define REG_A4XX_RBBM_PERFCTR_RAS_0_LO				0x000000fc
+
+#define REG_A4XX_RBBM_PERFCTR_RAS_0_HI				0x000000fd
+
+#define REG_A4XX_RBBM_PERFCTR_RAS_1_LO				0x000000fe
+
+#define REG_A4XX_RBBM_PERFCTR_RAS_1_HI				0x000000ff
+
+#define REG_A4XX_RBBM_PERFCTR_RAS_2_LO				0x00000100
+
+#define REG_A4XX_RBBM_PERFCTR_RAS_2_HI				0x00000101
+
+#define REG_A4XX_RBBM_PERFCTR_RAS_3_LO				0x00000102
+
+#define REG_A4XX_RBBM_PERFCTR_RAS_3_HI				0x00000103
+
+#define REG_A4XX_RBBM_PERFCTR_UCHE_0_LO				0x00000104
+
+#define REG_A4XX_RBBM_PERFCTR_UCHE_0_HI				0x00000105
+
+#define REG_A4XX_RBBM_PERFCTR_UCHE_1_LO				0x00000106
+
+#define REG_A4XX_RBBM_PERFCTR_UCHE_1_HI				0x00000107
+
+#define REG_A4XX_RBBM_PERFCTR_UCHE_2_LO				0x00000108
+
+#define REG_A4XX_RBBM_PERFCTR_UCHE_2_HI				0x00000109
+
+#define REG_A4XX_RBBM_PERFCTR_UCHE_3_LO				0x0000010a
+
+#define REG_A4XX_RBBM_PERFCTR_UCHE_3_HI				0x0000010b
+
+#define REG_A4XX_RBBM_PERFCTR_UCHE_4_LO				0x0000010c
+
+#define REG_A4XX_RBBM_PERFCTR_UCHE_4_HI				0x0000010d
+
+#define REG_A4XX_RBBM_PERFCTR_UCHE_5_LO				0x0000010e
+
+#define REG_A4XX_RBBM_PERFCTR_UCHE_5_HI				0x0000010f
+
+#define REG_A4XX_RBBM_PERFCTR_UCHE_6_LO				0x00000110
+
+#define REG_A4XX_RBBM_PERFCTR_UCHE_6_HI				0x00000111
+
+#define REG_A4XX_RBBM_PERFCTR_UCHE_7_LO				0x00000112
+
+#define REG_A4XX_RBBM_PERFCTR_UCHE_7_HI				0x00000113
+
+#define REG_A4XX_RBBM_PERFCTR_TP_0_LO				0x00000114
+
+#define REG_A4XX_RBBM_PERFCTR_TP_0_HI				0x00000115
+
+#define REG_A4XX_RBBM_PERFCTR_TP_0_LO				0x00000114
+
+#define REG_A4XX_RBBM_PERFCTR_TP_0_HI				0x00000115
+
+#define REG_A4XX_RBBM_PERFCTR_TP_1_LO				0x00000116
+
+#define REG_A4XX_RBBM_PERFCTR_TP_1_HI				0x00000117
+
+#define REG_A4XX_RBBM_PERFCTR_TP_2_LO				0x00000118
+
+#define REG_A4XX_RBBM_PERFCTR_TP_2_HI				0x00000119
+
+#define REG_A4XX_RBBM_PERFCTR_TP_3_LO				0x0000011a
+
+#define REG_A4XX_RBBM_PERFCTR_TP_3_HI				0x0000011b
+
+#define REG_A4XX_RBBM_PERFCTR_TP_4_LO				0x0000011c
+
+#define REG_A4XX_RBBM_PERFCTR_TP_4_HI				0x0000011d
+
+#define REG_A4XX_RBBM_PERFCTR_TP_5_LO				0x0000011e
+
+#define REG_A4XX_RBBM_PERFCTR_TP_5_HI				0x0000011f
+
+#define REG_A4XX_RBBM_PERFCTR_TP_6_LO				0x00000120
+
+#define REG_A4XX_RBBM_PERFCTR_TP_6_HI				0x00000121
+
+#define REG_A4XX_RBBM_PERFCTR_TP_7_LO				0x00000122
+
+#define REG_A4XX_RBBM_PERFCTR_TP_7_HI				0x00000123
+
+#define REG_A4XX_RBBM_PERFCTR_SP_0_LO				0x00000124
+
+#define REG_A4XX_RBBM_PERFCTR_SP_0_HI				0x00000125
+
+#define REG_A4XX_RBBM_PERFCTR_SP_1_LO				0x00000126
+
+#define REG_A4XX_RBBM_PERFCTR_SP_1_HI				0x00000127
+
+#define REG_A4XX_RBBM_PERFCTR_SP_2_LO				0x00000128
+
+#define REG_A4XX_RBBM_PERFCTR_SP_2_HI				0x00000129
+
+#define REG_A4XX_RBBM_PERFCTR_SP_3_LO				0x0000012a
+
+#define REG_A4XX_RBBM_PERFCTR_SP_3_HI				0x0000012b
+
+#define REG_A4XX_RBBM_PERFCTR_SP_4_LO				0x0000012c
+
+#define REG_A4XX_RBBM_PERFCTR_SP_4_HI				0x0000012d
+
+#define REG_A4XX_RBBM_PERFCTR_SP_5_LO				0x0000012e
+
+#define REG_A4XX_RBBM_PERFCTR_SP_5_HI				0x0000012f
+
+#define REG_A4XX_RBBM_PERFCTR_SP_6_LO				0x00000130
+
+#define REG_A4XX_RBBM_PERFCTR_SP_6_HI				0x00000131
+
+#define REG_A4XX_RBBM_PERFCTR_SP_7_LO				0x00000132
+
+#define REG_A4XX_RBBM_PERFCTR_SP_7_HI				0x00000133
+
+#define REG_A4XX_RBBM_PERFCTR_SP_8_LO				0x00000134
+
+#define REG_A4XX_RBBM_PERFCTR_SP_8_HI				0x00000135
+
+#define REG_A4XX_RBBM_PERFCTR_SP_9_LO				0x00000136
+
+#define REG_A4XX_RBBM_PERFCTR_SP_9_HI				0x00000137
+
+#define REG_A4XX_RBBM_PERFCTR_SP_10_LO				0x00000138
+
+#define REG_A4XX_RBBM_PERFCTR_SP_10_HI				0x00000139
+
+#define REG_A4XX_RBBM_PERFCTR_SP_11_LO				0x0000013a
+
+#define REG_A4XX_RBBM_PERFCTR_SP_11_HI				0x0000013b
+
+#define REG_A4XX_RBBM_PERFCTR_RB_0_LO				0x0000013c
+
+#define REG_A4XX_RBBM_PERFCTR_RB_0_HI				0x0000013d
+
+#define REG_A4XX_RBBM_PERFCTR_RB_1_LO				0x0000013e
+
+#define REG_A4XX_RBBM_PERFCTR_RB_1_HI				0x0000013f
+
+#define REG_A4XX_RBBM_PERFCTR_RB_2_LO				0x00000140
+
+#define REG_A4XX_RBBM_PERFCTR_RB_2_HI				0x00000141
+
+#define REG_A4XX_RBBM_PERFCTR_RB_3_LO				0x00000142
+
+#define REG_A4XX_RBBM_PERFCTR_RB_3_HI				0x00000143
+
+#define REG_A4XX_RBBM_PERFCTR_RB_4_LO				0x00000144
+
+#define REG_A4XX_RBBM_PERFCTR_RB_4_HI				0x00000145
+
+#define REG_A4XX_RBBM_PERFCTR_RB_5_LO				0x00000146
+
+#define REG_A4XX_RBBM_PERFCTR_RB_5_HI				0x00000147
+
+#define REG_A4XX_RBBM_PERFCTR_RB_6_LO				0x00000148
+
+#define REG_A4XX_RBBM_PERFCTR_RB_6_HI				0x00000149
+
+#define REG_A4XX_RBBM_PERFCTR_RB_7_LO				0x0000014a
+
+#define REG_A4XX_RBBM_PERFCTR_RB_7_HI				0x0000014b
+
+#define REG_A4XX_RBBM_PERFCTR_VSC_0_LO				0x0000014c
+
+#define REG_A4XX_RBBM_PERFCTR_VSC_0_HI				0x0000014d
+
+#define REG_A4XX_RBBM_PERFCTR_VSC_1_LO				0x0000014e
+
+#define REG_A4XX_RBBM_PERFCTR_VSC_1_HI				0x0000014f
+
+#define REG_A4XX_RBBM_PERFCTR_PWR_0_LO				0x00000166
+
+#define REG_A4XX_RBBM_PERFCTR_PWR_0_HI				0x00000167
+
+#define REG_A4XX_RBBM_PERFCTR_PWR_1_LO				0x00000168
+
+#define REG_A4XX_RBBM_PERFCTR_PWR_1_HI				0x00000169
+
+#define REG_A4XX_RBBM_ALWAYSON_COUNTER_LO			0x0000016e
+
+#define REG_A4XX_RBBM_ALWAYSON_COUNTER_HI			0x0000016f
+
 static inline uint32_t REG_A4XX_RBBM_CLOCK_CTL_SP(uint32_t i0) { return 0x00000068 + 0x1*i0; }
 
 static inline uint32_t REG_A4XX_RBBM_CLOCK_CTL_SP_REG(uint32_t i0) { return 0x00000068 + 0x1*i0; }
@@ -1046,6 +2039,10 @@ static inline uint32_t REG_A4XX_RBBM_CLOCK_DELAY_RB_MARB_CCU_L1(uint32_t i0) { r
 
 static inline uint32_t REG_A4XX_RBBM_CLOCK_DELAY_RB_MARB_CCU_L1_REG(uint32_t i0) { return 0x0000008e + 0x1*i0; }
 
+#define REG_A4XX_RBBM_SP_REGFILE_SLEEP_CNTL_0			0x00000099
+
+#define REG_A4XX_RBBM_SP_REGFILE_SLEEP_CNTL_1			0x0000009a
+
 #define REG_A4XX_RBBM_PERFCTR_PWR_1_LO				0x00000168
 
 #define REG_A4XX_RBBM_PERFCTR_CTL				0x00000170
@@ -1060,6 +2057,14 @@ static inline uint32_t REG_A4XX_RBBM_CLOCK_DELAY_RB_MARB_CCU_L1_REG(uint32_t i0)
 
 #define REG_A4XX_RBBM_PERFCTR_LOAD_VALUE_HI			0x00000175
 
+#define REG_A4XX_RBBM_PERFCTR_RBBM_SEL_0			0x00000176
+
+#define REG_A4XX_RBBM_PERFCTR_RBBM_SEL_1			0x00000177
+
+#define REG_A4XX_RBBM_PERFCTR_RBBM_SEL_2			0x00000178
+
+#define REG_A4XX_RBBM_PERFCTR_RBBM_SEL_3			0x00000179
+
 #define REG_A4XX_RBBM_GPU_BUSY_MASKED				0x0000017a
 
 #define REG_A4XX_RBBM_INT_0_STATUS				0x0000017d
@@ -1099,6 +2104,11 @@ static inline uint32_t REG_A4XX_RBBM_CLOCK_DELAY_RB_MARB_CCU_L1_REG(uint32_t i0)
 
 #define REG_A4XX_RBBM_INTERFACE_RRDY_STATUS5			0x0000019f
 
+#define REG_A4XX_RBBM_POWER_STATUS				0x000001b0
+#define A4XX_RBBM_POWER_STATUS_SP_TP_PWR_ON			0x00100000
+
+#define REG_A4XX_RBBM_WAIT_IDLE_CLOCKS_CTL2			0x000001b8
+
 #define REG_A4XX_CP_SCRATCH_UMASK				0x00000228
 
 #define REG_A4XX_CP_SCRATCH_ADDR				0x00000229
@@ -1191,6 +2201,20 @@ static inline uint32_t REG_A4XX_CP_PROTECT_REG(uint32_t i0) { return 0x00000240
 
 #define REG_A4XX_CP_PERFCTR_CP_SEL_0				0x00000500
 
+#define REG_A4XX_CP_PERFCTR_CP_SEL_1				0x00000501
+
+#define REG_A4XX_CP_PERFCTR_CP_SEL_2				0x00000502
+
+#define REG_A4XX_CP_PERFCTR_CP_SEL_3				0x00000503
+
+#define REG_A4XX_CP_PERFCTR_CP_SEL_4				0x00000504
+
+#define REG_A4XX_CP_PERFCTR_CP_SEL_5				0x00000505
+
+#define REG_A4XX_CP_PERFCTR_CP_SEL_6				0x00000506
+
+#define REG_A4XX_CP_PERFCTR_CP_SEL_7				0x00000507
+
 #define REG_A4XX_CP_PERFCOMBINER_SELECT				0x0000050b
 
 static inline uint32_t REG_A4XX_CP_SCRATCH(uint32_t i0) { return 0x00000578 + 0x1*i0; }
@@ -1201,6 +2225,28 @@ static inline uint32_t REG_A4XX_CP_SCRATCH_REG(uint32_t i0) { return 0x00000578
 
 #define REG_A4XX_SP_MODE_CONTROL				0x00000ec3
 
+#define REG_A4XX_SP_PERFCTR_SP_SEL_0				0x00000ec4
+
+#define REG_A4XX_SP_PERFCTR_SP_SEL_1				0x00000ec5
+
+#define REG_A4XX_SP_PERFCTR_SP_SEL_2				0x00000ec6
+
+#define REG_A4XX_SP_PERFCTR_SP_SEL_3				0x00000ec7
+
+#define REG_A4XX_SP_PERFCTR_SP_SEL_4				0x00000ec8
+
+#define REG_A4XX_SP_PERFCTR_SP_SEL_5				0x00000ec9
+
+#define REG_A4XX_SP_PERFCTR_SP_SEL_6				0x00000eca
+
+#define REG_A4XX_SP_PERFCTR_SP_SEL_7				0x00000ecb
+
+#define REG_A4XX_SP_PERFCTR_SP_SEL_8				0x00000ecc
+
+#define REG_A4XX_SP_PERFCTR_SP_SEL_9				0x00000ecd
+
+#define REG_A4XX_SP_PERFCTR_SP_SEL_10				0x00000ece
+
 #define REG_A4XX_SP_PERFCTR_SP_SEL_11				0x00000ecf
 
 #define REG_A4XX_SP_SP_CTRL_REG					0x000022c0
@@ -1699,6 +2745,12 @@ static inline uint32_t A4XX_SP_GS_OBJ_OFFSET_REG_SHADEROBJOFFSET(uint32_t val)
 
 #define REG_A4XX_VPC_DEBUG_ECO_CONTROL				0x00000e64
 
+#define REG_A4XX_VPC_PERFCTR_VPC_SEL_0				0x00000e65
+
+#define REG_A4XX_VPC_PERFCTR_VPC_SEL_1				0x00000e66
+
+#define REG_A4XX_VPC_PERFCTR_VPC_SEL_2				0x00000e67
+
 #define REG_A4XX_VPC_PERFCTR_VPC_SEL_3				0x00000e68
 
 #define REG_A4XX_VPC_ATTR					0x00002140
@@ -1811,6 +2863,20 @@ static inline uint32_t REG_A4XX_VSC_PIPE_DATA_LENGTH_REG(uint32_t i0) { return 0
 
 #define REG_A4XX_VFD_DEBUG_CONTROL				0x00000e40
 
+#define REG_A4XX_VFD_PERFCTR_VFD_SEL_0				0x00000e43
+
+#define REG_A4XX_VFD_PERFCTR_VFD_SEL_1				0x00000e44
+
+#define REG_A4XX_VFD_PERFCTR_VFD_SEL_2				0x00000e45
+
+#define REG_A4XX_VFD_PERFCTR_VFD_SEL_3				0x00000e46
+
+#define REG_A4XX_VFD_PERFCTR_VFD_SEL_4				0x00000e47
+
+#define REG_A4XX_VFD_PERFCTR_VFD_SEL_5				0x00000e48
+
+#define REG_A4XX_VFD_PERFCTR_VFD_SEL_6				0x00000e49
+
 #define REG_A4XX_VFD_PERFCTR_VFD_SEL_7				0x00000e4a
 
 #define REG_A4XX_VGT_CL_INITIATOR				0x000021d0
@@ -1967,6 +3033,20 @@ static inline uint32_t A4XX_VFD_DECODE_INSTR_SHIFTCNT(uint32_t val)
 
 #define REG_A4XX_TPL1_TP_MODE_CONTROL				0x00000f03
 
+#define REG_A4XX_TPL1_PERFCTR_TP_SEL_0				0x00000f04
+
+#define REG_A4XX_TPL1_PERFCTR_TP_SEL_1				0x00000f05
+
+#define REG_A4XX_TPL1_PERFCTR_TP_SEL_2				0x00000f06
+
+#define REG_A4XX_TPL1_PERFCTR_TP_SEL_3				0x00000f07
+
+#define REG_A4XX_TPL1_PERFCTR_TP_SEL_4				0x00000f08
+
+#define REG_A4XX_TPL1_PERFCTR_TP_SEL_5				0x00000f09
+
+#define REG_A4XX_TPL1_PERFCTR_TP_SEL_6				0x00000f0a
+
 #define REG_A4XX_TPL1_PERFCTR_TP_SEL_7				0x00000f0b
 
 #define REG_A4XX_TPL1_TP_TEX_OFFSET				0x00002380
@@ -2021,9 +3101,23 @@ static inline uint32_t A4XX_TPL1_TP_TEX_COUNT_GS(uint32_t val)
 
 #define REG_A4XX_GRAS_PERFCTR_TSE_SEL_0				0x00000c88
 
+#define REG_A4XX_GRAS_PERFCTR_TSE_SEL_1				0x00000c89
+
+#define REG_A4XX_GRAS_PERFCTR_TSE_SEL_2				0x00000c8a
+
 #define REG_A4XX_GRAS_PERFCTR_TSE_SEL_3				0x00000c8b
 
+#define REG_A4XX_GRAS_PERFCTR_RAS_SEL_0				0x00000c8c
+
+#define REG_A4XX_GRAS_PERFCTR_RAS_SEL_1				0x00000c8d
+
+#define REG_A4XX_GRAS_PERFCTR_RAS_SEL_2				0x00000c8e
+
+#define REG_A4XX_GRAS_PERFCTR_RAS_SEL_3				0x00000c8f
+
 #define REG_A4XX_GRAS_CL_CLIP_CNTL				0x00002000
+#define A4XX_GRAS_CL_CLIP_CNTL_CLIP_DISABLE			0x00008000
+#define A4XX_GRAS_CL_CLIP_CNTL_ZERO_GB_SCALE_Z			0x00400000
 
 #define REG_A4XX_GRAS_CLEAR_CNTL				0x00002003
 #define A4XX_GRAS_CLEAR_CNTL_NOT_FASTCLEAR			0x00000001
@@ -2114,6 +3208,7 @@ static inline uint32_t A4XX_GRAS_SU_POINT_SIZE(float val)
 
 #define REG_A4XX_GRAS_ALPHA_CONTROL				0x00002073
 #define A4XX_GRAS_ALPHA_CONTROL_ALPHA_TEST_ENABLE		0x00000004
+#define A4XX_GRAS_ALPHA_CONTROL_FORCE_FRAGZ_TO_FS		0x00000008
 
 #define REG_A4XX_GRAS_SU_POLY_OFFSET_SCALE			0x00002074
 #define A4XX_GRAS_SU_POLY_OFFSET_SCALE__MASK			0xffffffff
@@ -2285,6 +3380,20 @@ static inline uint32_t A4XX_GRAS_SC_EXTENT_WINDOW_TL_Y(uint32_t val)
 
 #define REG_A4XX_UCHE_CACHE_WAYS_VFD				0x00000e8c
 
+#define REG_A4XX_UCHE_PERFCTR_UCHE_SEL_0			0x00000e8e
+
+#define REG_A4XX_UCHE_PERFCTR_UCHE_SEL_1			0x00000e8f
+
+#define REG_A4XX_UCHE_PERFCTR_UCHE_SEL_2			0x00000e90
+
+#define REG_A4XX_UCHE_PERFCTR_UCHE_SEL_3			0x00000e91
+
+#define REG_A4XX_UCHE_PERFCTR_UCHE_SEL_4			0x00000e92
+
+#define REG_A4XX_UCHE_PERFCTR_UCHE_SEL_5			0x00000e93
+
+#define REG_A4XX_UCHE_PERFCTR_UCHE_SEL_6			0x00000e94
+
 #define REG_A4XX_UCHE_PERFCTR_UCHE_SEL_7			0x00000e95
 
 #define REG_A4XX_HLSQ_TIMEOUT_THRESHOLD				0x00000e00
@@ -2295,6 +3404,22 @@ static inline uint32_t A4XX_GRAS_SC_EXTENT_WINDOW_TL_Y(uint32_t val)
 
 #define REG_A4XX_HLSQ_PERF_PIPE_MASK				0x00000e0e
 
+#define REG_A4XX_HLSQ_PERFCTR_HLSQ_SEL_0			0x00000e06
+
+#define REG_A4XX_HLSQ_PERFCTR_HLSQ_SEL_1			0x00000e07
+
+#define REG_A4XX_HLSQ_PERFCTR_HLSQ_SEL_2			0x00000e08
+
+#define REG_A4XX_HLSQ_PERFCTR_HLSQ_SEL_3			0x00000e09
+
+#define REG_A4XX_HLSQ_PERFCTR_HLSQ_SEL_4			0x00000e0a
+
+#define REG_A4XX_HLSQ_PERFCTR_HLSQ_SEL_5			0x00000e0b
+
+#define REG_A4XX_HLSQ_PERFCTR_HLSQ_SEL_6			0x00000e0c
+
+#define REG_A4XX_HLSQ_PERFCTR_HLSQ_SEL_7			0x00000e0d
+
 #define REG_A4XX_HLSQ_CONTROL_0_REG				0x000023c0
 #define A4XX_HLSQ_CONTROL_0_REG_FSTHREADSIZE__MASK		0x00000010
 #define A4XX_HLSQ_CONTROL_0_REG_FSTHREADSIZE__SHIFT		4
@@ -2549,6 +3674,18 @@ static inline uint32_t A4XX_HLSQ_GS_CONTROL_REG_INSTRLENGTH(uint32_t val)
 
 #define REG_A4XX_PC_PERFCTR_PC_SEL_0				0x00000d10
 
+#define REG_A4XX_PC_PERFCTR_PC_SEL_1				0x00000d11
+
+#define REG_A4XX_PC_PERFCTR_PC_SEL_2				0x00000d12
+
+#define REG_A4XX_PC_PERFCTR_PC_SEL_3				0x00000d13
+
+#define REG_A4XX_PC_PERFCTR_PC_SEL_4				0x00000d14
+
+#define REG_A4XX_PC_PERFCTR_PC_SEL_5				0x00000d15
+
+#define REG_A4XX_PC_PERFCTR_PC_SEL_6				0x00000d16
+
 #define REG_A4XX_PC_PERFCTR_PC_SEL_7				0x00000d17
 
 #define REG_A4XX_PC_BIN_BASE					0x000021c0
@@ -2564,7 +3701,20 @@ static inline uint32_t A4XX_PC_PRIM_VTX_CNTL_VAROUT(uint32_t val)
 #define A4XX_PC_PRIM_VTX_CNTL_PROVOKING_VTX_LAST		0x02000000
 #define A4XX_PC_PRIM_VTX_CNTL_PSIZE				0x04000000
 
-#define REG_A4XX_UNKNOWN_21C5					0x000021c5
+#define REG_A4XX_PC_PRIM_VTX_CNTL2				0x000021c5
+#define A4XX_PC_PRIM_VTX_CNTL2_POLYMODE_FRONT_PTYPE__MASK	0x00000007
+#define A4XX_PC_PRIM_VTX_CNTL2_POLYMODE_FRONT_PTYPE__SHIFT	0
+static inline uint32_t A4XX_PC_PRIM_VTX_CNTL2_POLYMODE_FRONT_PTYPE(enum adreno_pa_su_sc_draw val)
+{
+	return ((val) << A4XX_PC_PRIM_VTX_CNTL2_POLYMODE_FRONT_PTYPE__SHIFT) & A4XX_PC_PRIM_VTX_CNTL2_POLYMODE_FRONT_PTYPE__MASK;
+}
+#define A4XX_PC_PRIM_VTX_CNTL2_POLYMODE_BACK_PTYPE__MASK	0x00000038
+#define A4XX_PC_PRIM_VTX_CNTL2_POLYMODE_BACK_PTYPE__SHIFT	3
+static inline uint32_t A4XX_PC_PRIM_VTX_CNTL2_POLYMODE_BACK_PTYPE(enum adreno_pa_su_sc_draw val)
+{
+	return ((val) << A4XX_PC_PRIM_VTX_CNTL2_POLYMODE_BACK_PTYPE__SHIFT) & A4XX_PC_PRIM_VTX_CNTL2_POLYMODE_BACK_PTYPE__MASK;
+}
+#define A4XX_PC_PRIM_VTX_CNTL2_POLYMODE_ENABLE			0x00000040
 
 #define REG_A4XX_PC_RESTART_INDEX				0x000021c6
 
@@ -2646,20 +3796,6 @@ static inline uint32_t A4XX_PC_HS_PARAM_PRIMTYPE(enum adreno_pa_su_sc_draw val)
 
 #define REG_A4XX_UNKNOWN_20EF					0x000020ef
 
-#define REG_A4XX_UNKNOWN_20F0					0x000020f0
-
-#define REG_A4XX_UNKNOWN_20F1					0x000020f1
-
-#define REG_A4XX_UNKNOWN_20F2					0x000020f2
-
-#define REG_A4XX_UNKNOWN_20F7					0x000020f7
-#define A4XX_UNKNOWN_20F7__MASK					0xffffffff
-#define A4XX_UNKNOWN_20F7__SHIFT				0
-static inline uint32_t A4XX_UNKNOWN_20F7(float val)
-{
-	return ((fui(val)) << A4XX_UNKNOWN_20F7__SHIFT) & A4XX_UNKNOWN_20F7__MASK;
-}
-
 #define REG_A4XX_UNKNOWN_2152					0x00002152
 
 #define REG_A4XX_UNKNOWN_2153					0x00002153
@@ -2720,6 +3856,12 @@ static inline uint32_t A4XX_TEX_SAMP_0_ANISO(enum a4xx_tex_aniso val)
 {
 	return ((val) << A4XX_TEX_SAMP_0_ANISO__SHIFT) & A4XX_TEX_SAMP_0_ANISO__MASK;
 }
+#define A4XX_TEX_SAMP_0_LOD_BIAS__MASK				0xfff80000
+#define A4XX_TEX_SAMP_0_LOD_BIAS__SHIFT				19
+static inline uint32_t A4XX_TEX_SAMP_0_LOD_BIAS(float val)
+{
+	return ((((int32_t)(val * 256.0))) << A4XX_TEX_SAMP_0_LOD_BIAS__SHIFT) & A4XX_TEX_SAMP_0_LOD_BIAS__MASK;
+}
 
 #define REG_A4XX_TEX_SAMP_1					0x00000001
 #define A4XX_TEX_SAMP_1_COMPARE_FUNC__MASK			0x0000000e
@@ -2728,6 +3870,7 @@ static inline uint32_t A4XX_TEX_SAMP_1_COMPARE_FUNC(enum adreno_compare_func val
 {
 	return ((val) << A4XX_TEX_SAMP_1_COMPARE_FUNC__SHIFT) & A4XX_TEX_SAMP_1_COMPARE_FUNC__MASK;
 }
+#define A4XX_TEX_SAMP_1_CUBEMAPSEAMLESSFILTOFF			0x00000010
 #define A4XX_TEX_SAMP_1_UNNORM_COORDS				0x00000020
 #define A4XX_TEX_SAMP_1_MIPFILTER_LINEAR_FAR			0x00000040
 #define A4XX_TEX_SAMP_1_MAX_LOD__MASK				0x000fff00
@@ -2796,7 +3939,7 @@ static inline uint32_t A4XX_TEX_CONST_1_HEIGHT(uint32_t val)
 {
 	return ((val) << A4XX_TEX_CONST_1_HEIGHT__SHIFT) & A4XX_TEX_CONST_1_HEIGHT__MASK;
 }
-#define A4XX_TEX_CONST_1_WIDTH__MASK				0x1fff8000
+#define A4XX_TEX_CONST_1_WIDTH__MASK				0x3fff8000
 #define A4XX_TEX_CONST_1_WIDTH__SHIFT				15
 static inline uint32_t A4XX_TEX_CONST_1_WIDTH(uint32_t val)
 {
diff --git a/drivers/gpu/drm/msm/adreno/a4xx_gpu.c b/drivers/gpu/drm/msm/adreno/a4xx_gpu.c
index a53f1be05f75..d0d3c7baa8fe 100644
--- a/drivers/gpu/drm/msm/adreno/a4xx_gpu.c
+++ b/drivers/gpu/drm/msm/adreno/a4xx_gpu.c
@@ -102,11 +102,17 @@ static void a4xx_enable_hwcg(struct msm_gpu *gpu)
 	gpu_write(gpu, REG_A4XX_RBBM_CLOCK_DELAY_TSE_RAS_RBBM, 0x00000222);
 	gpu_write(gpu, REG_A4XX_RBBM_CLOCK_CTL_HLSQ , 0x00000000);
 	gpu_write(gpu, REG_A4XX_RBBM_CLOCK_HYST_HLSQ, 0x00000000);
-	gpu_write(gpu, REG_A4XX_RBBM_CLOCK_DELAY_HLSQ, 0x00020000);
-	gpu_write(gpu, REG_A4XX_RBBM_CLOCK_CTL, 0xAAAAAAAA);
+	gpu_write(gpu, REG_A4XX_RBBM_CLOCK_DELAY_HLSQ, 0x00220000);
+	/* Early A430's have a timing issue with SP/TP power collapse;
+	   disabling HW clock gating prevents it. */
+	if (adreno_is_a430(adreno_gpu) && adreno_gpu->rev.patchid < 2)
+		gpu_write(gpu, REG_A4XX_RBBM_CLOCK_CTL, 0);
+	else
+		gpu_write(gpu, REG_A4XX_RBBM_CLOCK_CTL, 0xAAAAAAAA);
 	gpu_write(gpu, REG_A4XX_RBBM_CLOCK_CTL2, 0);
 }
 
+
 static void a4xx_me_init(struct msm_gpu *gpu)
 {
 	struct msm_ringbuffer *ring = gpu->rb;
@@ -141,7 +147,7 @@ static int a4xx_hw_init(struct msm_gpu *gpu)
 	uint32_t *ptr, len;
 	int i, ret;
 
-	if (adreno_is_a4xx(adreno_gpu)) {
+	if (adreno_is_a420(adreno_gpu)) {
 		gpu_write(gpu, REG_A4XX_VBIF_ABIT_SORT, 0x0001001F);
 		gpu_write(gpu, REG_A4XX_VBIF_ABIT_SORT_CONF, 0x000000A4);
 		gpu_write(gpu, REG_A4XX_VBIF_GATE_OFF_WRREQ_EN, 0x00000001);
@@ -150,6 +156,13 @@ static int a4xx_hw_init(struct msm_gpu *gpu)
 		gpu_write(gpu, REG_A4XX_VBIF_IN_WR_LIM_CONF0, 0x18181818);
 		gpu_write(gpu, REG_A4XX_VBIF_IN_WR_LIM_CONF1, 0x00000018);
 		gpu_write(gpu, REG_A4XX_VBIF_ROUND_ROBIN_QOS_ARB, 0x00000003);
+	} else if (adreno_is_a430(adreno_gpu)) {
+		gpu_write(gpu, REG_A4XX_VBIF_GATE_OFF_WRREQ_EN, 0x00000001);
+		gpu_write(gpu, REG_A4XX_VBIF_IN_RD_LIM_CONF0, 0x18181818);
+		gpu_write(gpu, REG_A4XX_VBIF_IN_RD_LIM_CONF1, 0x00000018);
+		gpu_write(gpu, REG_A4XX_VBIF_IN_WR_LIM_CONF0, 0x18181818);
+		gpu_write(gpu, REG_A4XX_VBIF_IN_WR_LIM_CONF1, 0x00000018);
+		gpu_write(gpu, REG_A4XX_VBIF_ROUND_ROBIN_QOS_ARB, 0x00000003);
 	} else {
 		BUG();
 	}
@@ -161,6 +174,10 @@ static int a4xx_hw_init(struct msm_gpu *gpu)
 	gpu_write(gpu, REG_A4XX_RBBM_SP_HYST_CNT, 0x10);
 	gpu_write(gpu, REG_A4XX_RBBM_WAIT_IDLE_CLOCKS_CTL, 0x10);
 
+	if (adreno_is_a430(adreno_gpu)) {
+		gpu_write(gpu, REG_A4XX_RBBM_WAIT_IDLE_CLOCKS_CTL2, 0x30);
+	}
+
 	 /* Enable the RBBM error reporting bits */
 	gpu_write(gpu, REG_A4XX_RBBM_AHB_CTL0, 0x00000001);
 
@@ -183,6 +200,14 @@ static int a4xx_hw_init(struct msm_gpu *gpu)
 	/* Turn on performance counters: */
 	gpu_write(gpu, REG_A4XX_RBBM_PERFCTR_CTL, 0x01);
 
+	/* use the first CP counter for timestamp queries.. userspace may set
+	 * this as well but it selects the same counter/countable:
+	 */
+	gpu_write(gpu, REG_A4XX_CP_PERFCTR_CP_SEL_0, CP_ALWAYS_COUNT);
+
+	if (adreno_is_a430(adreno_gpu))
+		gpu_write(gpu, REG_A4XX_UCHE_CACHE_WAYS_VFD, 0x07);
+
 	/* Disable L2 bypass to avoid UCHE out of bounds errors */
 	gpu_write(gpu, REG_A4XX_UCHE_TRAP_BASE_LO, 0xffff0000);
 	gpu_write(gpu, REG_A4XX_UCHE_TRAP_BASE_HI, 0xffff0000);
@@ -190,6 +215,15 @@ static int a4xx_hw_init(struct msm_gpu *gpu)
 	gpu_write(gpu, REG_A4XX_CP_DEBUG, (1 << 25) |
 			(adreno_is_a420(adreno_gpu) ? (1 << 29) : 0));
 
+	/* On A430 enable SP regfile sleep for power savings */
+	/* TODO downstream does this for !420, so maybe applies for 405 too? */
+	if (!adreno_is_a420(adreno_gpu)) {
+		gpu_write(gpu, REG_A4XX_RBBM_SP_REGFILE_SLEEP_CNTL_0,
+			0x00000441);
+		gpu_write(gpu, REG_A4XX_RBBM_SP_REGFILE_SLEEP_CNTL_1,
+			0x00000441);
+	}
+
 	a4xx_enable_hwcg(gpu);
 
 	/*
@@ -204,10 +238,6 @@ static int a4xx_hw_init(struct msm_gpu *gpu)
 		gpu_write(gpu, REG_A4XX_RBBM_CLOCK_DELAY_HLSQ, val);
 	}
 
-	ret = adreno_hw_init(gpu);
-	if (ret)
-		return ret;
-
 	/* setup access protection: */
 	gpu_write(gpu, REG_A4XX_CP_PROTECT_CTRL, 0x00000007);
 
@@ -263,6 +293,7 @@ static int a4xx_hw_init(struct msm_gpu *gpu)
 	gpu_write(gpu, REG_A4XX_CP_ME_CNTL, 0);
 
 	a4xx_me_init(gpu);
+
 	return 0;
 }
 
@@ -317,6 +348,13 @@ static irqreturn_t a4xx_irq(struct msm_gpu *gpu)
 	status = gpu_read(gpu, REG_A4XX_RBBM_INT_0_STATUS);
 	DBG("%s: Int status %08x", gpu->name, status);
 
+	if (status & A4XX_INT0_CP_REG_PROTECT_FAULT) {
+		uint32_t reg = gpu_read(gpu, REG_A4XX_CP_PROTECT_STATUS);
+		printk("CP | Protected mode error| %s | addr=%x\n",
+			reg & (1 << 24) ? "WRITE" : "READ",
+			(reg & 0xFFFFF) >> 2);
+	}
+
 	gpu_write(gpu, REG_A4XX_RBBM_INT_CLEAR_CMD, status);
 
 	msm_gpu_retire(gpu);
@@ -512,12 +550,63 @@ static void a4xx_dump(struct msm_gpu *gpu)
 	adreno_dump(gpu);
 }
 
+static int a4xx_pm_resume(struct msm_gpu *gpu) {
+	struct adreno_gpu *adreno_gpu = to_adreno_gpu(gpu);
+	int ret;
+
+	ret = msm_gpu_pm_resume(gpu);
+	if (ret)
+		return ret;
+
+	if (adreno_is_a430(adreno_gpu)) {
+		unsigned int reg;
+		/* Set the default register values; set SW_COLLAPSE to 0 */
+		gpu_write(gpu, REG_A4XX_RBBM_POWER_CNTL_IP, 0x778000);
+		do {
+			udelay(5);
+			reg = gpu_read(gpu, REG_A4XX_RBBM_POWER_STATUS);
+		} while (!(reg & A4XX_RBBM_POWER_CNTL_IP_SP_TP_PWR_ON));
+	}
+	return 0;
+}
+
+static int a4xx_pm_suspend(struct msm_gpu *gpu) {
+	struct adreno_gpu *adreno_gpu = to_adreno_gpu(gpu);
+	int ret;
+
+	ret = msm_gpu_pm_suspend(gpu);
+	if (ret)
+		return ret;
+
+	if (adreno_is_a430(adreno_gpu)) {
+		/* Set the default register values; set SW_COLLAPSE to 1 */
+		gpu_write(gpu, REG_A4XX_RBBM_POWER_CNTL_IP, 0x778001);
+	}
+	return 0;
+}
+
+static int a4xx_get_timestamp(struct msm_gpu *gpu, uint64_t *value)
+{
+	uint32_t hi, lo, tmp;
+
+	tmp = gpu_read(gpu, REG_A4XX_RBBM_PERFCTR_CP_0_HI);
+	do {
+		hi = tmp;
+		lo = gpu_read(gpu, REG_A4XX_RBBM_PERFCTR_CP_0_LO);
+		tmp = gpu_read(gpu, REG_A4XX_RBBM_PERFCTR_CP_0_HI);
+	} while (tmp != hi);
+
+	*value = (((uint64_t)hi) << 32) | lo;
+
+	return 0;
+}
+
 static const struct adreno_gpu_funcs funcs = {
 	.base = {
 		.get_param = adreno_get_param,
 		.hw_init = a4xx_hw_init,
-		.pm_suspend = msm_gpu_pm_suspend,
-		.pm_resume = msm_gpu_pm_resume,
+		.pm_suspend = a4xx_pm_suspend,
+		.pm_resume = a4xx_pm_resume,
 		.recover = a4xx_recover,
 		.last_fence = adreno_last_fence,
 		.submit = adreno_submit,
@@ -529,6 +618,7 @@ static const struct adreno_gpu_funcs funcs = {
 		.show = a4xx_show,
 #endif
 	},
+	.get_timestamp = a4xx_get_timestamp,
 };
 
 struct msm_gpu *a4xx_gpu_init(struct drm_device *dev)
diff --git a/drivers/gpu/drm/msm/adreno/adreno_common.xml.h b/drivers/gpu/drm/msm/adreno/adreno_common.xml.h
index c304468cf2bd..e81481d1b7df 100644
--- a/drivers/gpu/drm/msm/adreno/adreno_common.xml.h
+++ b/drivers/gpu/drm/msm/adreno/adreno_common.xml.h
@@ -9,16 +9,17 @@ git clone https://github.com/freedreno/envytools.git
 
 The rules-ng-ng source files this header was generated from are:
 - /home/robclark/src/freedreno/envytools/rnndb/adreno.xml               (    398 bytes, from 2015-09-24 17:25:31)
-- /home/robclark/src/freedreno/envytools/rnndb/freedreno_copyright.xml  (   1453 bytes, from 2015-05-20 20:03:07)
+- /home/robclark/src/freedreno/envytools/rnndb/freedreno_copyright.xml  (   1572 bytes, from 2016-02-10 17:07:21)
 - /home/robclark/src/freedreno/envytools/rnndb/adreno/a2xx.xml          (  32901 bytes, from 2015-05-20 20:03:14)
-- /home/robclark/src/freedreno/envytools/rnndb/adreno/adreno_common.xml (  10755 bytes, from 2015-09-14 20:46:55)
-- /home/robclark/src/freedreno/envytools/rnndb/adreno/adreno_pm4.xml    (  14968 bytes, from 2015-05-20 20:12:27)
-- /home/robclark/src/freedreno/envytools/rnndb/adreno/a3xx.xml          (  67771 bytes, from 2015-09-14 20:46:55)
-- /home/robclark/src/freedreno/envytools/rnndb/adreno/a4xx.xml          (  63970 bytes, from 2015-09-14 20:50:12)
+- /home/robclark/src/freedreno/envytools/rnndb/adreno/adreno_common.xml (  11518 bytes, from 2016-02-10 21:03:25)
+- /home/robclark/src/freedreno/envytools/rnndb/adreno/adreno_pm4.xml    (  16166 bytes, from 2016-02-11 21:20:31)
+- /home/robclark/src/freedreno/envytools/rnndb/adreno/a3xx.xml          (  83967 bytes, from 2016-02-10 17:07:21)
+- /home/robclark/src/freedreno/envytools/rnndb/adreno/a4xx.xml          ( 109916 bytes, from 2016-02-20 18:44:48)
 - /home/robclark/src/freedreno/envytools/rnndb/adreno/ocmem.xml         (   1773 bytes, from 2015-09-24 17:30:00)
 
-Copyright (C) 2013-2015 by the following authors:
+Copyright (C) 2013-2016 by the following authors:
 - Rob Clark <robdclark@gmail.com> (robclark)
+- Ilia Mirkin <imirkin@alum.mit.edu> (imirkin)
 
 Permission is hereby granted, free of charge, to any person obtaining
 a copy of this software and associated documentation files (the
@@ -119,6 +120,23 @@ enum adreno_rb_copy_control_mode {
 	RB_COPY_DEPTH_STENCIL = 5,
 };
 
+enum a3xx_rop_code {
+	ROP_CLEAR = 0,
+	ROP_NOR = 1,
+	ROP_AND_INVERTED = 2,
+	ROP_COPY_INVERTED = 3,
+	ROP_AND_REVERSE = 4,
+	ROP_INVERT = 5,
+	ROP_NAND = 7,
+	ROP_AND = 8,
+	ROP_EQUIV = 9,
+	ROP_NOOP = 10,
+	ROP_OR_INVERTED = 11,
+	ROP_OR_REVERSE = 13,
+	ROP_OR = 14,
+	ROP_SET = 15,
+};
+
 enum a3xx_render_mode {
 	RB_RENDERING_PASS = 0,
 	RB_TILING_PASS = 1,
diff --git a/drivers/gpu/drm/msm/adreno/adreno_device.c b/drivers/gpu/drm/msm/adreno/adreno_device.c
index 950d27d26b30..5127b75dbf40 100644
--- a/drivers/gpu/drm/msm/adreno/adreno_device.c
+++ b/drivers/gpu/drm/msm/adreno/adreno_device.c
@@ -69,6 +69,14 @@ static const struct adreno_info gpulist[] = {
 		.pfpfw = "a420_pfp.fw",
 		.gmem  = (SZ_1M + SZ_512K),
 		.init  = a4xx_gpu_init,
+	}, {
+		.rev   = ADRENO_REV(4, 3, 0, ANY_ID),
+		.revn  = 430,
+		.name  = "A430",
+		.pm4fw = "a420_pm4.fw",
+		.pfpfw = "a420_pfp.fw",
+		.gmem  = (SZ_1M + SZ_512K),
+		.init  = a4xx_gpu_init,
 	},
 };
 
diff --git a/drivers/gpu/drm/msm/adreno/adreno_gpu.c b/drivers/gpu/drm/msm/adreno/adreno_gpu.c
index 0e1d0c57cd3d..4951172ede06 100644
--- a/drivers/gpu/drm/msm/adreno/adreno_gpu.c
+++ b/drivers/gpu/drm/msm/adreno/adreno_gpu.c
@@ -44,6 +44,10 @@ int adreno_get_param(struct msm_gpu *gpu, uint32_t param, uint64_t *value)
 	case MSM_PARAM_MAX_FREQ:
 		*value = adreno_gpu->base.fast_rate;
 		return 0;
+	case MSM_PARAM_TIMESTAMP:
+		if (adreno_gpu->funcs->get_timestamp)
+			return adreno_gpu->funcs->get_timestamp(gpu, value);
+		return -EINVAL;
 	default:
 		DBG("%s: invalid param: %u", gpu->name, param);
 		return -EINVAL;
@@ -71,18 +75,15 @@ int adreno_hw_init(struct msm_gpu *gpu)
 	adreno_gpu_write(adreno_gpu, REG_ADRENO_CP_RB_CNTL,
 			/* size is log2(quad-words): */
 			AXXX_CP_RB_CNTL_BUFSZ(ilog2(gpu->rb->size / 8)) |
-			AXXX_CP_RB_CNTL_BLKSZ(ilog2(RB_BLKSIZE / 8)));
+			AXXX_CP_RB_CNTL_BLKSZ(ilog2(RB_BLKSIZE / 8)) |
+			(adreno_is_a430(adreno_gpu) ? AXXX_CP_RB_CNTL_NO_UPDATE : 0));
 
 	/* Setup ringbuffer address: */
 	adreno_gpu_write(adreno_gpu, REG_ADRENO_CP_RB_BASE, gpu->rb_iova);
-	adreno_gpu_write(adreno_gpu, REG_ADRENO_CP_RB_RPTR_ADDR,
-			rbmemptr(adreno_gpu, rptr));
-
-	/* Setup scratch/timestamp: */
-	adreno_gpu_write(adreno_gpu, REG_ADRENO_SCRATCH_ADDR,
-			rbmemptr(adreno_gpu, fence));
 
-	adreno_gpu_write(adreno_gpu, REG_ADRENO_SCRATCH_UMSK, 0x1);
+	if (!adreno_is_a430(adreno_gpu))
+		adreno_gpu_write(adreno_gpu, REG_ADRENO_CP_RB_RPTR_ADDR,
+						rbmemptr(adreno_gpu, rptr));
 
 	return 0;
 }
@@ -92,6 +93,16 @@ static uint32_t get_wptr(struct msm_ringbuffer *ring)
 	return ring->cur - ring->start;
 }
 
+/* Use this helper to read rptr, since a430 doesn't update rptr in memory */
+static uint32_t get_rptr(struct adreno_gpu *adreno_gpu)
+{
+	if (adreno_is_a430(adreno_gpu))
+		return adreno_gpu->memptrs->rptr = adreno_gpu_read(
+			adreno_gpu, REG_ADRENO_CP_RB_RPTR);
+	else
+		return adreno_gpu->memptrs->rptr;
+}
+
 uint32_t adreno_last_fence(struct msm_gpu *gpu)
 {
 	struct adreno_gpu *adreno_gpu = to_adreno_gpu(gpu);
@@ -140,7 +151,8 @@ int adreno_submit(struct msm_gpu *gpu, struct msm_gem_submit *submit,
 			if (priv->lastctx == ctx)
 				break;
 		case MSM_SUBMIT_CMD_BUF:
-			OUT_PKT3(ring, CP_INDIRECT_BUFFER_PFD, 2);
+			OUT_PKT3(ring, adreno_is_a430(adreno_gpu) ?
+				CP_INDIRECT_BUFFER_PFE : CP_INDIRECT_BUFFER_PFD, 2);
 			OUT_RING(ring, submit->cmd[i].iova);
 			OUT_RING(ring, submit->cmd[i].size);
 			ibs++;
@@ -219,9 +231,12 @@ void adreno_idle(struct msm_gpu *gpu)
 {
 	struct adreno_gpu *adreno_gpu = to_adreno_gpu(gpu);
 	uint32_t wptr = get_wptr(gpu->rb);
+	int ret;
 
 	/* wait for CP to drain ringbuffer: */
-	if (spin_until(adreno_gpu->memptrs->rptr == wptr))
+	ret = spin_until(get_rptr(adreno_gpu) == wptr);
+
+	if (ret)
 		DRM_ERROR("%s: timeout waiting to drain ringbuffer!\n", gpu->name);
 
 	/* TODO maybe we need to reset GPU here to recover from hang? */
@@ -240,7 +255,7 @@ void adreno_show(struct msm_gpu *gpu, struct seq_file *m)
 
 	seq_printf(m, "fence:    %d/%d\n", adreno_gpu->memptrs->fence,
 			gpu->submitted_fence);
-	seq_printf(m, "rptr:     %d\n", adreno_gpu->memptrs->rptr);
+	seq_printf(m, "rptr:     %d\n", get_rptr(adreno_gpu));
 	seq_printf(m, "wptr:     %d\n", adreno_gpu->memptrs->wptr);
 	seq_printf(m, "rb wptr:  %d\n", get_wptr(gpu->rb));
 
@@ -281,7 +296,7 @@ void adreno_dump_info(struct msm_gpu *gpu)
 
 	printk("fence:    %d/%d\n", adreno_gpu->memptrs->fence,
 			gpu->submitted_fence);
-	printk("rptr:     %d\n", adreno_gpu->memptrs->rptr);
+	printk("rptr:     %d\n", get_rptr(adreno_gpu));
 	printk("wptr:     %d\n", adreno_gpu->memptrs->wptr);
 	printk("rb wptr:  %d\n", get_wptr(gpu->rb));
 
@@ -316,7 +331,7 @@ static uint32_t ring_freewords(struct msm_gpu *gpu)
 	struct adreno_gpu *adreno_gpu = to_adreno_gpu(gpu);
 	uint32_t size = gpu->rb->size / 4;
 	uint32_t wptr = get_wptr(gpu->rb);
-	uint32_t rptr = adreno_gpu->memptrs->rptr;
+	uint32_t rptr = get_rptr(adreno_gpu);
 	return (rptr + (size - 1) - wptr) % size;
 }
 
diff --git a/drivers/gpu/drm/msm/adreno/adreno_gpu.h b/drivers/gpu/drm/msm/adreno/adreno_gpu.h
index 0a312e9d3afd..1d07511f4d22 100644
--- a/drivers/gpu/drm/msm/adreno/adreno_gpu.h
+++ b/drivers/gpu/drm/msm/adreno/adreno_gpu.h
@@ -114,6 +114,7 @@ struct adreno_rev {
 
 struct adreno_gpu_funcs {
 	struct msm_gpu_funcs base;
+	int (*get_timestamp)(struct msm_gpu *gpu, uint64_t *value);
 };
 
 struct adreno_info {
@@ -228,6 +229,11 @@ static inline int adreno_is_a420(struct adreno_gpu *gpu)
 	return gpu->revn == 420;
 }
 
+static inline int adreno_is_a430(struct adreno_gpu *gpu)
+{
+       return gpu->revn == 430;
+}
+
 int adreno_get_param(struct msm_gpu *gpu, uint32_t param, uint64_t *value);
 int adreno_hw_init(struct msm_gpu *gpu);
 uint32_t adreno_last_fence(struct msm_gpu *gpu);
diff --git a/drivers/gpu/drm/msm/adreno/adreno_pm4.xml.h b/drivers/gpu/drm/msm/adreno/adreno_pm4.xml.h
index a22fef569499..d7477ff867c9 100644
--- a/drivers/gpu/drm/msm/adreno/adreno_pm4.xml.h
+++ b/drivers/gpu/drm/msm/adreno/adreno_pm4.xml.h
@@ -9,16 +9,17 @@ git clone https://github.com/freedreno/envytools.git
 
 The rules-ng-ng source files this header was generated from are:
 - /home/robclark/src/freedreno/envytools/rnndb/adreno.xml               (    398 bytes, from 2015-09-24 17:25:31)
-- /home/robclark/src/freedreno/envytools/rnndb/freedreno_copyright.xml  (   1453 bytes, from 2015-05-20 20:03:07)
+- /home/robclark/src/freedreno/envytools/rnndb/freedreno_copyright.xml  (   1572 bytes, from 2016-02-10 17:07:21)
 - /home/robclark/src/freedreno/envytools/rnndb/adreno/a2xx.xml          (  32901 bytes, from 2015-05-20 20:03:14)
-- /home/robclark/src/freedreno/envytools/rnndb/adreno/adreno_common.xml (  10755 bytes, from 2015-09-14 20:46:55)
-- /home/robclark/src/freedreno/envytools/rnndb/adreno/adreno_pm4.xml    (  14968 bytes, from 2015-05-20 20:12:27)
-- /home/robclark/src/freedreno/envytools/rnndb/adreno/a3xx.xml          (  67771 bytes, from 2015-09-14 20:46:55)
-- /home/robclark/src/freedreno/envytools/rnndb/adreno/a4xx.xml          (  63970 bytes, from 2015-09-14 20:50:12)
+- /home/robclark/src/freedreno/envytools/rnndb/adreno/adreno_common.xml (  11518 bytes, from 2016-02-10 21:03:25)
+- /home/robclark/src/freedreno/envytools/rnndb/adreno/adreno_pm4.xml    (  16166 bytes, from 2016-02-11 21:20:31)
+- /home/robclark/src/freedreno/envytools/rnndb/adreno/a3xx.xml          (  83967 bytes, from 2016-02-10 17:07:21)
+- /home/robclark/src/freedreno/envytools/rnndb/adreno/a4xx.xml          ( 109916 bytes, from 2016-02-20 18:44:48)
 - /home/robclark/src/freedreno/envytools/rnndb/adreno/ocmem.xml         (   1773 bytes, from 2015-09-24 17:30:00)
 
-Copyright (C) 2013-2015 by the following authors:
+Copyright (C) 2013-2016 by the following authors:
 - Rob Clark <robdclark@gmail.com> (robclark)
+- Ilia Mirkin <imirkin@alum.mit.edu> (imirkin)
 
 Permission is hereby granted, free of charge, to any person obtaining
 a copy of this software and associated documentation files (the
@@ -172,6 +173,11 @@ enum adreno_pm4_type3_packets {
 	CP_UNKNOWN_1A = 26,
 	CP_UNKNOWN_4E = 78,
 	CP_WIDE_REG_WRITE = 116,
+	CP_SCRATCH_TO_REG = 77,
+	CP_REG_TO_SCRATCH = 74,
+	CP_WAIT_MEM_WRITES = 18,
+	CP_COND_REG_EXEC = 71,
+	CP_MEM_TO_REG = 66,
 	IN_IB_PREFETCH_END = 23,
 	IN_SUBBLK_PREFETCH = 31,
 	IN_INSTR_PREFETCH = 32,
@@ -199,7 +205,11 @@ enum adreno_state_type {
 
 enum adreno_state_src {
 	SS_DIRECT = 0,
+	SS_INVALID_ALL_IC = 2,
+	SS_INVALID_PART_IC = 3,
 	SS_INDIRECT = 4,
+	SS_INDIRECT_TCM = 5,
+	SS_INDIRECT_STM = 6,
 };
 
 enum a4xx_index_size {
@@ -227,7 +237,7 @@ static inline uint32_t CP_LOAD_STATE_0_STATE_BLOCK(enum adreno_state_block val)
 {
 	return ((val) << CP_LOAD_STATE_0_STATE_BLOCK__SHIFT) & CP_LOAD_STATE_0_STATE_BLOCK__MASK;
 }
-#define CP_LOAD_STATE_0_NUM_UNIT__MASK				0x7fc00000
+#define CP_LOAD_STATE_0_NUM_UNIT__MASK				0xffc00000
 #define CP_LOAD_STATE_0_NUM_UNIT__SHIFT				22
 static inline uint32_t CP_LOAD_STATE_0_NUM_UNIT(uint32_t val)
 {
@@ -499,5 +509,29 @@ static inline uint32_t CP_SET_BIN_DATA_1_BIN_SIZE_ADDRESS(uint32_t val)
 	return ((val) << CP_SET_BIN_DATA_1_BIN_SIZE_ADDRESS__SHIFT) & CP_SET_BIN_DATA_1_BIN_SIZE_ADDRESS__MASK;
 }
 
+#define REG_CP_REG_TO_MEM_0					0x00000000
+#define CP_REG_TO_MEM_0_REG__MASK				0x0000ffff
+#define CP_REG_TO_MEM_0_REG__SHIFT				0
+static inline uint32_t CP_REG_TO_MEM_0_REG(uint32_t val)
+{
+	return ((val) << CP_REG_TO_MEM_0_REG__SHIFT) & CP_REG_TO_MEM_0_REG__MASK;
+}
+#define CP_REG_TO_MEM_0_CNT__MASK				0x3ff80000
+#define CP_REG_TO_MEM_0_CNT__SHIFT				19
+static inline uint32_t CP_REG_TO_MEM_0_CNT(uint32_t val)
+{
+	return ((val) << CP_REG_TO_MEM_0_CNT__SHIFT) & CP_REG_TO_MEM_0_CNT__MASK;
+}
+#define CP_REG_TO_MEM_0_64B					0x40000000
+#define CP_REG_TO_MEM_0_ACCUMULATE				0x80000000
+
+#define REG_CP_REG_TO_MEM_1					0x00000001
+#define CP_REG_TO_MEM_1_DEST__MASK				0xffffffff
+#define CP_REG_TO_MEM_1_DEST__SHIFT				0
+static inline uint32_t CP_REG_TO_MEM_1_DEST(uint32_t val)
+{
+	return ((val) << CP_REG_TO_MEM_1_DEST__SHIFT) & CP_REG_TO_MEM_1_DEST__MASK;
+}
+
 
 #endif /* ADRENO_PM4_XML */
diff --git a/drivers/gpu/drm/msm/dsi/dsi.xml.h b/drivers/gpu/drm/msm/dsi/dsi.xml.h
index b2b5f3dd1b4c..4958594d5266 100644
--- a/drivers/gpu/drm/msm/dsi/dsi.xml.h
+++ b/drivers/gpu/drm/msm/dsi/dsi.xml.h
@@ -9,7 +9,7 @@ git clone https://github.com/freedreno/envytools.git
 
 The rules-ng-ng source files this header was generated from are:
 - /home/robclark/src/freedreno/envytools/rnndb/msm.xml                 (    676 bytes, from 2015-05-20 20:03:14)
-- /home/robclark/src/freedreno/envytools/rnndb/freedreno_copyright.xml (   1453 bytes, from 2015-05-20 20:03:07)
+- /home/robclark/src/freedreno/envytools/rnndb/freedreno_copyright.xml (   1572 bytes, from 2016-02-10 17:07:21)
 - /home/robclark/src/freedreno/envytools/rnndb/mdp/mdp4.xml            (  20915 bytes, from 2015-05-20 20:03:14)
 - /home/robclark/src/freedreno/envytools/rnndb/mdp/mdp_common.xml      (   2849 bytes, from 2015-09-18 12:07:28)
 - /home/robclark/src/freedreno/envytools/rnndb/mdp/mdp5.xml            (  37194 bytes, from 2015-09-18 12:07:28)
@@ -17,11 +17,12 @@ The rules-ng-ng source files this header was generated from are:
 - /home/robclark/src/freedreno/envytools/rnndb/dsi/sfpb.xml            (    602 bytes, from 2015-10-22 16:35:02)
 - /home/robclark/src/freedreno/envytools/rnndb/dsi/mmss_cc.xml         (   1686 bytes, from 2015-05-20 20:03:14)
 - /home/robclark/src/freedreno/envytools/rnndb/hdmi/qfprom.xml         (    600 bytes, from 2015-05-20 20:03:07)
-- /home/robclark/src/freedreno/envytools/rnndb/hdmi/hdmi.xml           (  29154 bytes, from 2015-08-10 21:25:43)
+- /home/robclark/src/freedreno/envytools/rnndb/hdmi/hdmi.xml           (  41472 bytes, from 2016-01-22 18:18:18)
 - /home/robclark/src/freedreno/envytools/rnndb/edp/edp.xml             (  10416 bytes, from 2015-05-20 20:03:14)
 
 Copyright (C) 2013-2015 by the following authors:
 - Rob Clark <robdclark@gmail.com> (robclark)
+- Ilia Mirkin <imirkin@alum.mit.edu> (imirkin)
 
 Permission is hereby granted, free of charge, to any person obtaining
 a copy of this software and associated documentation files (the
diff --git a/drivers/gpu/drm/msm/dsi/dsi_cfg.c b/drivers/gpu/drm/msm/dsi/dsi_cfg.c
index 2a827d8093a2..e58e9b91b34d 100644
--- a/drivers/gpu/drm/msm/dsi/dsi_cfg.c
+++ b/drivers/gpu/drm/msm/dsi/dsi_cfg.c
@@ -57,10 +57,9 @@ static const char * const dsi_8916_bus_clk_names[] = {
 static const struct msm_dsi_config msm8916_dsi_cfg = {
 	.io_offset = DSI_6G_REG_SHIFT,
 	.reg_cfg = {
-		.num = 4,
+		.num = 3,
 		.regs = {
 			{"gdsc", -1, -1, -1, -1},
-			{"vdd", 2850000, 2850000, 100000, 100},
 			{"vdda", 1200000, 1200000, 100000, 100},
 			{"vddio", 1800000, 1800000, 100000, 100},
 		},
diff --git a/drivers/gpu/drm/msm/dsi/dsi_host.c b/drivers/gpu/drm/msm/dsi/dsi_host.c
index 48f9967b4a1b..4282ec6bbaaf 100644
--- a/drivers/gpu/drm/msm/dsi/dsi_host.c
+++ b/drivers/gpu/drm/msm/dsi/dsi_host.c
@@ -163,6 +163,10 @@ struct msm_dsi_host {
 	enum mipi_dsi_pixel_format format;
 	unsigned long mode_flags;
 
+	/* lane data parsed via DT */
+	int dlane_swap;
+	int num_data_lanes;
+
 	u32 dma_cmd_ctrl_restore;
 
 	bool registered;
@@ -845,19 +849,10 @@ static void dsi_ctrl_config(struct msm_dsi_host *msm_host, bool enable,
 	data = DSI_CTRL_CLK_EN;
 
 	DBG("lane number=%d", msm_host->lanes);
-	if (msm_host->lanes == 2) {
-		data |= DSI_CTRL_LANE1 | DSI_CTRL_LANE2;
-		/* swap lanes for 2-lane panel for better performance */
-		dsi_write(msm_host, REG_DSI_LANE_SWAP_CTRL,
-			DSI_LANE_SWAP_CTRL_DLN_SWAP_SEL(LANE_SWAP_1230));
-	} else {
-		/* Take 4 lanes as default */
-		data |= DSI_CTRL_LANE0 | DSI_CTRL_LANE1 | DSI_CTRL_LANE2 |
-			DSI_CTRL_LANE3;
-		/* Do not swap lanes for 4-lane panel */
-		dsi_write(msm_host, REG_DSI_LANE_SWAP_CTRL,
-			DSI_LANE_SWAP_CTRL_DLN_SWAP_SEL(LANE_SWAP_0123));
-	}
+	data |= ((DSI_CTRL_LANE0 << msm_host->lanes) - DSI_CTRL_LANE0);
+
+	dsi_write(msm_host, REG_DSI_LANE_SWAP_CTRL,
+		  DSI_LANE_SWAP_CTRL_DLN_SWAP_SEL(msm_host->dlane_swap));
 
 	if (!(flags & MIPI_DSI_CLOCK_NON_CONTINUOUS))
 		dsi_write(msm_host, REG_DSI_LANE_CTRL,
@@ -1479,13 +1474,14 @@ static int dsi_host_attach(struct mipi_dsi_host *host,
 	struct msm_dsi_host *msm_host = to_msm_dsi_host(host);
 	int ret;
 
+	if (dsi->lanes > msm_host->num_data_lanes)
+		return -EINVAL;
+
 	msm_host->channel = dsi->channel;
 	msm_host->lanes = dsi->lanes;
 	msm_host->format = dsi->format;
 	msm_host->mode_flags = dsi->mode_flags;
 
-	WARN_ON(dsi->dev.of_node != msm_host->device_node);
-
 	/* Some gpios defined in panel DT need to be controlled by host */
 	ret = dsi_host_init_panel_gpios(msm_host, &dsi->dev);
 	if (ret)
@@ -1534,6 +1530,75 @@ static struct mipi_dsi_host_ops dsi_host_ops = {
 	.transfer = dsi_host_transfer,
 };
 
+/*
+ * List of supported physical to logical lane mappings.
+ * For example, the 2nd entry represents the following mapping:
+ *
+ * "3012": Logic 3->Phys 0; Logic 0->Phys 1; Logic 1->Phys 2; Logic 2->Phys 3;
+ */
+static const int supported_data_lane_swaps[][4] = {
+	{ 0, 1, 2, 3 },
+	{ 3, 0, 1, 2 },
+	{ 2, 3, 0, 1 },
+	{ 1, 2, 3, 0 },
+	{ 0, 3, 2, 1 },
+	{ 1, 0, 3, 2 },
+	{ 2, 1, 0, 3 },
+	{ 3, 2, 1, 0 },
+};
+
+static int dsi_host_parse_lane_data(struct msm_dsi_host *msm_host,
+				    struct device_node *ep)
+{
+	struct device *dev = &msm_host->pdev->dev;
+	struct property *prop;
+	u32 lane_map[4];
+	int ret, i, len, num_lanes;
+
+	prop = of_find_property(ep, "qcom,data-lane-map", &len);
+	if (!prop) {
+		dev_dbg(dev, "failed to find data lane mapping\n");
+		return -EINVAL;
+	}
+
+	num_lanes = len / sizeof(u32);
+
+	if (num_lanes < 1 || num_lanes > 4) {
+		dev_err(dev, "bad number of data lanes\n");
+		return -EINVAL;
+	}
+
+	msm_host->num_data_lanes = num_lanes;
+
+	ret = of_property_read_u32_array(ep, "qcom,data-lane-map", lane_map,
+					 num_lanes);
+	if (ret) {
+		dev_err(dev, "failed to read lane data\n");
+		return ret;
+	}
+
+	/*
+	 * compare DT specified physical-logical lane mappings with the ones
+	 * supported by hardware
+	 */
+	for (i = 0; i < ARRAY_SIZE(supported_data_lane_swaps); i++) {
+		const int *swap = supported_data_lane_swaps[i];
+		int j;
+
+		for (j = 0; j < num_lanes; j++) {
+			if (swap[j] != lane_map[j])
+				break;
+		}
+
+		if (j == num_lanes) {
+			msm_host->dlane_swap = i;
+			return 0;
+		}
+	}
+
+	return -EINVAL;
+}
+
 static int dsi_host_parse_dt(struct msm_dsi_host *msm_host)
 {
 	struct device *dev = &msm_host->pdev->dev;
@@ -1560,17 +1625,21 @@ static int dsi_host_parse_dt(struct msm_dsi_host *msm_host)
 		return 0;
 	}
 
+	ret = dsi_host_parse_lane_data(msm_host, endpoint);
+	if (ret) {
+		dev_err(dev, "%s: invalid lane configuration %d\n",
+			__func__, ret);
+		goto err;
+	}
+
 	/* Get panel node from the output port's endpoint data */
 	device_node = of_graph_get_remote_port_parent(endpoint);
 	if (!device_node) {
 		dev_err(dev, "%s: no valid device\n", __func__);
-		of_node_put(endpoint);
-		return -ENODEV;
+		ret = -ENODEV;
+		goto err;
 	}
 
-	of_node_put(endpoint);
-	of_node_put(device_node);
-
 	msm_host->device_node = device_node;
 
 	if (of_property_read_bool(np, "syscon-sfpb")) {
@@ -1579,11 +1648,16 @@ static int dsi_host_parse_dt(struct msm_dsi_host *msm_host)
 		if (IS_ERR(msm_host->sfpb)) {
 			dev_err(dev, "%s: failed to get sfpb regmap\n",
 				__func__);
-			return PTR_ERR(msm_host->sfpb);
+			ret = PTR_ERR(msm_host->sfpb);
 		}
 	}
 
-	return 0;
+	of_node_put(device_node);
+
+err:
+	of_node_put(endpoint);
+
+	return ret;
 }
 
 int msm_dsi_host_init(struct msm_dsi *msm_dsi)
diff --git a/drivers/gpu/drm/msm/dsi/mmss_cc.xml.h b/drivers/gpu/drm/msm/dsi/mmss_cc.xml.h
index 80ec65e47468..2d999494cdea 100644
--- a/drivers/gpu/drm/msm/dsi/mmss_cc.xml.h
+++ b/drivers/gpu/drm/msm/dsi/mmss_cc.xml.h
@@ -9,7 +9,7 @@ git clone https://github.com/freedreno/envytools.git
 
 The rules-ng-ng source files this header was generated from are:
 - /home/robclark/src/freedreno/envytools/rnndb/msm.xml                 (    676 bytes, from 2015-05-20 20:03:14)
-- /home/robclark/src/freedreno/envytools/rnndb/freedreno_copyright.xml (   1453 bytes, from 2015-05-20 20:03:07)
+- /home/robclark/src/freedreno/envytools/rnndb/freedreno_copyright.xml (   1572 bytes, from 2016-02-10 17:07:21)
 - /home/robclark/src/freedreno/envytools/rnndb/mdp/mdp4.xml            (  20915 bytes, from 2015-05-20 20:03:14)
 - /home/robclark/src/freedreno/envytools/rnndb/mdp/mdp_common.xml      (   2849 bytes, from 2015-09-18 12:07:28)
 - /home/robclark/src/freedreno/envytools/rnndb/mdp/mdp5.xml            (  37194 bytes, from 2015-09-18 12:07:28)
@@ -17,11 +17,12 @@ The rules-ng-ng source files this header was generated from are:
 - /home/robclark/src/freedreno/envytools/rnndb/dsi/sfpb.xml            (    602 bytes, from 2015-10-22 16:35:02)
 - /home/robclark/src/freedreno/envytools/rnndb/dsi/mmss_cc.xml         (   1686 bytes, from 2015-05-20 20:03:14)
 - /home/robclark/src/freedreno/envytools/rnndb/hdmi/qfprom.xml         (    600 bytes, from 2015-05-20 20:03:07)
-- /home/robclark/src/freedreno/envytools/rnndb/hdmi/hdmi.xml           (  29154 bytes, from 2015-08-10 21:25:43)
+- /home/robclark/src/freedreno/envytools/rnndb/hdmi/hdmi.xml           (  41472 bytes, from 2016-01-22 18:18:18)
 - /home/robclark/src/freedreno/envytools/rnndb/edp/edp.xml             (  10416 bytes, from 2015-05-20 20:03:14)
 
 Copyright (C) 2013-2015 by the following authors:
 - Rob Clark <robdclark@gmail.com> (robclark)
+- Ilia Mirkin <imirkin@alum.mit.edu> (imirkin)
 
 Permission is hereby granted, free of charge, to any person obtaining
 a copy of this software and associated documentation files (the
diff --git a/drivers/gpu/drm/msm/dsi/pll/dsi_pll.h b/drivers/gpu/drm/msm/dsi/pll/dsi_pll.h
index 80b6038334a6..2cf1664723e8 100644
--- a/drivers/gpu/drm/msm/dsi/pll/dsi_pll.h
+++ b/drivers/gpu/drm/msm/dsi/pll/dsi_pll.h
@@ -97,8 +97,8 @@ static inline struct msm_dsi_pll *msm_dsi_pll_28nm_init(
 struct msm_dsi_pll *msm_dsi_pll_28nm_8960_init(struct platform_device *pdev,
 					       int id);
 #else
-struct msm_dsi_pll *msm_dsi_pll_28nm_8960_init(struct platform_device *pdev,
-					       int id)
+static inline struct msm_dsi_pll *msm_dsi_pll_28nm_8960_init(
+	struct platform_device *pdev, int id)
 {
 	return ERR_PTR(-ENODEV);
 }
diff --git a/drivers/gpu/drm/msm/dsi/sfpb.xml.h b/drivers/gpu/drm/msm/dsi/sfpb.xml.h
index 7d7662e69e11..506434fac993 100644
--- a/drivers/gpu/drm/msm/dsi/sfpb.xml.h
+++ b/drivers/gpu/drm/msm/dsi/sfpb.xml.h
@@ -9,7 +9,7 @@ git clone https://github.com/freedreno/envytools.git
 
 The rules-ng-ng source files this header was generated from are:
 - /home/robclark/src/freedreno/envytools/rnndb/msm.xml                 (    676 bytes, from 2015-05-20 20:03:14)
-- /home/robclark/src/freedreno/envytools/rnndb/freedreno_copyright.xml (   1453 bytes, from 2015-05-20 20:03:07)
+- /home/robclark/src/freedreno/envytools/rnndb/freedreno_copyright.xml (   1572 bytes, from 2016-02-10 17:07:21)
 - /home/robclark/src/freedreno/envytools/rnndb/mdp/mdp4.xml            (  20915 bytes, from 2015-05-20 20:03:14)
 - /home/robclark/src/freedreno/envytools/rnndb/mdp/mdp_common.xml      (   2849 bytes, from 2015-09-18 12:07:28)
 - /home/robclark/src/freedreno/envytools/rnndb/mdp/mdp5.xml            (  37194 bytes, from 2015-09-18 12:07:28)
@@ -17,11 +17,12 @@ The rules-ng-ng source files this header was generated from are:
 - /home/robclark/src/freedreno/envytools/rnndb/dsi/sfpb.xml            (    602 bytes, from 2015-10-22 16:35:02)
 - /home/robclark/src/freedreno/envytools/rnndb/dsi/mmss_cc.xml         (   1686 bytes, from 2015-05-20 20:03:14)
 - /home/robclark/src/freedreno/envytools/rnndb/hdmi/qfprom.xml         (    600 bytes, from 2015-05-20 20:03:07)
-- /home/robclark/src/freedreno/envytools/rnndb/hdmi/hdmi.xml           (  29154 bytes, from 2015-08-10 21:25:43)
+- /home/robclark/src/freedreno/envytools/rnndb/hdmi/hdmi.xml           (  41472 bytes, from 2016-01-22 18:18:18)
 - /home/robclark/src/freedreno/envytools/rnndb/edp/edp.xml             (  10416 bytes, from 2015-05-20 20:03:14)
 
 Copyright (C) 2013-2015 by the following authors:
 - Rob Clark <robdclark@gmail.com> (robclark)
+- Ilia Mirkin <imirkin@alum.mit.edu> (imirkin)
 
 Permission is hereby granted, free of charge, to any person obtaining
 a copy of this software and associated documentation files (the
diff --git a/drivers/gpu/drm/msm/edp/edp.xml.h b/drivers/gpu/drm/msm/edp/edp.xml.h
index 90bf5ed46746..f1072c18c81e 100644
--- a/drivers/gpu/drm/msm/edp/edp.xml.h
+++ b/drivers/gpu/drm/msm/edp/edp.xml.h
@@ -9,7 +9,7 @@ git clone https://github.com/freedreno/envytools.git
 
 The rules-ng-ng source files this header was generated from are:
 - /home/robclark/src/freedreno/envytools/rnndb/msm.xml                 (    676 bytes, from 2015-05-20 20:03:14)
-- /home/robclark/src/freedreno/envytools/rnndb/freedreno_copyright.xml (   1453 bytes, from 2015-05-20 20:03:07)
+- /home/robclark/src/freedreno/envytools/rnndb/freedreno_copyright.xml (   1572 bytes, from 2016-02-10 17:07:21)
 - /home/robclark/src/freedreno/envytools/rnndb/mdp/mdp4.xml            (  20915 bytes, from 2015-05-20 20:03:14)
 - /home/robclark/src/freedreno/envytools/rnndb/mdp/mdp_common.xml      (   2849 bytes, from 2015-09-18 12:07:28)
 - /home/robclark/src/freedreno/envytools/rnndb/mdp/mdp5.xml            (  37194 bytes, from 2015-09-18 12:07:28)
@@ -17,11 +17,12 @@ The rules-ng-ng source files this header was generated from are:
 - /home/robclark/src/freedreno/envytools/rnndb/dsi/sfpb.xml            (    602 bytes, from 2015-10-22 16:35:02)
 - /home/robclark/src/freedreno/envytools/rnndb/dsi/mmss_cc.xml         (   1686 bytes, from 2015-05-20 20:03:14)
 - /home/robclark/src/freedreno/envytools/rnndb/hdmi/qfprom.xml         (    600 bytes, from 2015-05-20 20:03:07)
-- /home/robclark/src/freedreno/envytools/rnndb/hdmi/hdmi.xml           (  29154 bytes, from 2015-08-10 21:25:43)
+- /home/robclark/src/freedreno/envytools/rnndb/hdmi/hdmi.xml           (  41472 bytes, from 2016-01-22 18:18:18)
 - /home/robclark/src/freedreno/envytools/rnndb/edp/edp.xml             (  10416 bytes, from 2015-05-20 20:03:14)
 
 Copyright (C) 2013-2015 by the following authors:
 - Rob Clark <robdclark@gmail.com> (robclark)
+- Ilia Mirkin <imirkin@alum.mit.edu> (imirkin)
 
 Permission is hereby granted, free of charge, to any person obtaining
 a copy of this software and associated documentation files (the
diff --git a/drivers/gpu/drm/msm/hdmi/hdmi.c b/drivers/gpu/drm/msm/hdmi/hdmi.c
index 9a0989c0b4de..51b9ea552f97 100644
--- a/drivers/gpu/drm/msm/hdmi/hdmi.c
+++ b/drivers/gpu/drm/msm/hdmi/hdmi.c
@@ -21,7 +21,7 @@
 
 #include "hdmi.h"
 
-void hdmi_set_mode(struct hdmi *hdmi, bool power_on)
+void msm_hdmi_set_mode(struct hdmi *hdmi, bool power_on)
 {
 	uint32_t ctrl = 0;
 	unsigned long flags;
@@ -46,29 +46,27 @@ void hdmi_set_mode(struct hdmi *hdmi, bool power_on)
 			power_on ? "Enable" : "Disable", ctrl);
 }
 
-static irqreturn_t hdmi_irq(int irq, void *dev_id)
+static irqreturn_t msm_hdmi_irq(int irq, void *dev_id)
 {
 	struct hdmi *hdmi = dev_id;
 
 	/* Process HPD: */
-	hdmi_connector_irq(hdmi->connector);
+	msm_hdmi_connector_irq(hdmi->connector);
 
 	/* Process DDC: */
-	hdmi_i2c_irq(hdmi->i2c);
+	msm_hdmi_i2c_irq(hdmi->i2c);
 
 	/* Process HDCP: */
 	if (hdmi->hdcp_ctrl)
-		hdmi_hdcp_irq(hdmi->hdcp_ctrl);
+		msm_hdmi_hdcp_irq(hdmi->hdcp_ctrl);
 
 	/* TODO audio.. */
 
 	return IRQ_HANDLED;
 }
 
-static void hdmi_destroy(struct hdmi *hdmi)
+static void msm_hdmi_destroy(struct hdmi *hdmi)
 {
-	struct hdmi_phy *phy = hdmi->phy;
-
 	/*
 	 * at this point, hpd has been disabled,
 	 * after flush workq, it's safe to deinit hdcp
@@ -77,21 +75,53 @@ static void hdmi_destroy(struct hdmi *hdmi)
 		flush_workqueue(hdmi->workq);
 		destroy_workqueue(hdmi->workq);
 	}
-	hdmi_hdcp_destroy(hdmi);
-	if (phy)
-		phy->funcs->destroy(phy);
+	msm_hdmi_hdcp_destroy(hdmi);
+
+	if (hdmi->phy_dev) {
+		put_device(hdmi->phy_dev);
+		hdmi->phy = NULL;
+		hdmi->phy_dev = NULL;
+	}
 
 	if (hdmi->i2c)
-		hdmi_i2c_destroy(hdmi->i2c);
+		msm_hdmi_i2c_destroy(hdmi->i2c);
 
 	platform_set_drvdata(hdmi->pdev, NULL);
 }
 
+static int msm_hdmi_get_phy(struct hdmi *hdmi)
+{
+	struct platform_device *pdev = hdmi->pdev;
+	struct platform_device *phy_pdev;
+	struct device_node *phy_node;
+
+	phy_node = of_parse_phandle(pdev->dev.of_node, "phys", 0);
+	if (!phy_node) {
+		dev_err(&pdev->dev, "cannot find phy device\n");
+		return -ENXIO;
+	}
+
+	phy_pdev = of_find_device_by_node(phy_node);
+	if (phy_pdev)
+		hdmi->phy = platform_get_drvdata(phy_pdev);
+
+	of_node_put(phy_node);
+
+	if (!phy_pdev || !hdmi->phy) {
+		dev_err(&pdev->dev, "phy driver is not ready\n");
+		return -EPROBE_DEFER;
+	}
+
+	hdmi->phy_dev = get_device(&phy_pdev->dev);
+
+	return 0;
+}
+
 /* construct hdmi at bind/probe time, grab all the resources.  If
  * we are to EPROBE_DEFER we want to do it here, rather than later
  * at modeset_init() time
  */
-static struct hdmi *hdmi_init(struct platform_device *pdev)
+static struct hdmi *msm_hdmi_init(struct platform_device *pdev)
 {
 	struct hdmi_platform_config *config = pdev->dev.platform_data;
 	struct hdmi *hdmi = NULL;
@@ -108,18 +138,6 @@ static struct hdmi *hdmi_init(struct platform_device *pdev)
 	hdmi->config = config;
 	spin_lock_init(&hdmi->reg_lock);
 
-	/* not sure about which phy maps to which msm.. probably I miss some */
-	if (config->phy_init) {
-		hdmi->phy = config->phy_init(hdmi);
-
-		if (IS_ERR(hdmi->phy)) {
-			ret = PTR_ERR(hdmi->phy);
-			dev_err(&pdev->dev, "failed to load phy: %d\n", ret);
-			hdmi->phy = NULL;
-			goto fail;
-		}
-	}
-
 	hdmi->mmio = msm_ioremap(pdev, config->mmio_name, "HDMI");
 	if (IS_ERR(hdmi->mmio)) {
 		ret = PTR_ERR(hdmi->mmio);
@@ -222,7 +240,7 @@ static struct hdmi *hdmi_init(struct platform_device *pdev)
 
 	hdmi->workq = alloc_ordered_workqueue("msm_hdmi", 0);
 
-	hdmi->i2c = hdmi_i2c_init(hdmi);
+	hdmi->i2c = msm_hdmi_i2c_init(hdmi);
 	if (IS_ERR(hdmi->i2c)) {
 		ret = PTR_ERR(hdmi->i2c);
 		dev_err(&pdev->dev, "failed to get i2c: %d\n", ret);
@@ -230,7 +248,13 @@ static struct hdmi *hdmi_init(struct platform_device *pdev)
 		goto fail;
 	}
 
-	hdmi->hdcp_ctrl = hdmi_hdcp_init(hdmi);
+	ret = msm_hdmi_get_phy(hdmi);
+	if (ret) {
+		dev_err(&pdev->dev, "failed to get phy\n");
+		goto fail;
+	}
+
+	hdmi->hdcp_ctrl = msm_hdmi_hdcp_init(hdmi);
 	if (IS_ERR(hdmi->hdcp_ctrl)) {
 		dev_warn(&pdev->dev, "failed to init hdcp: disabled\n");
 		hdmi->hdcp_ctrl = NULL;
@@ -240,7 +264,7 @@ static struct hdmi *hdmi_init(struct platform_device *pdev)
 
 fail:
 	if (hdmi)
-		hdmi_destroy(hdmi);
+		msm_hdmi_destroy(hdmi);
 
 	return ERR_PTR(ret);
 }
@@ -250,10 +274,10 @@ fail:
  * driver (not hdmi sub-device's probe/bind!)
  *
  * Any resource (regulator/clk/etc) which could be missing at boot
- * should be handled in hdmi_init() so that failure happens from
+ * should be handled in msm_hdmi_init() so that failure happens from
  * hdmi sub-device's probe.
  */
-int hdmi_modeset_init(struct hdmi *hdmi,
+int msm_hdmi_modeset_init(struct hdmi *hdmi,
 		struct drm_device *dev, struct drm_encoder *encoder)
 {
 	struct msm_drm_private *priv = dev->dev_private;
@@ -265,7 +289,7 @@ int hdmi_modeset_init(struct hdmi *hdmi,
 
 	hdmi_audio_infoframe_init(&hdmi->audio.infoframe);
 
-	hdmi->bridge = hdmi_bridge_init(hdmi);
+	hdmi->bridge = msm_hdmi_bridge_init(hdmi);
 	if (IS_ERR(hdmi->bridge)) {
 		ret = PTR_ERR(hdmi->bridge);
 		dev_err(dev->dev, "failed to create HDMI bridge: %d\n", ret);
@@ -273,7 +297,7 @@ int hdmi_modeset_init(struct hdmi *hdmi,
 		goto fail;
 	}
 
-	hdmi->connector = hdmi_connector_init(hdmi);
+	hdmi->connector = msm_hdmi_connector_init(hdmi);
 	if (IS_ERR(hdmi->connector)) {
 		ret = PTR_ERR(hdmi->connector);
 		dev_err(dev->dev, "failed to create HDMI connector: %d\n", ret);
@@ -289,7 +313,7 @@ int hdmi_modeset_init(struct hdmi *hdmi,
 	}
 
 	ret = devm_request_irq(&pdev->dev, hdmi->irq,
-			hdmi_irq, IRQF_TRIGGER_HIGH | IRQF_ONESHOT,
+			msm_hdmi_irq, IRQF_TRIGGER_HIGH | IRQF_ONESHOT,
 			"hdmi_isr", hdmi);
 	if (ret < 0) {
 		dev_err(dev->dev, "failed to request IRQ%u: %d\n",
@@ -309,7 +333,7 @@ int hdmi_modeset_init(struct hdmi *hdmi,
 fail:
 	/* bridge is normally destroyed by drm: */
 	if (hdmi->bridge) {
-		hdmi_bridge_destroy(hdmi->bridge);
+		msm_hdmi_bridge_destroy(hdmi->bridge);
 		hdmi->bridge = NULL;
 	}
 	if (hdmi->connector) {
@@ -331,15 +355,12 @@ fail:
 static const char *pwr_reg_names_none[] = {};
 static const char *hpd_reg_names_none[] = {};
 
-static struct hdmi_platform_config hdmi_tx_8660_config = {
-		.phy_init = hdmi_phy_8x60_init,
-};
+static struct hdmi_platform_config hdmi_tx_8660_config;
 
 static const char *hpd_reg_names_8960[] = {"core-vdda", "hdmi-mux"};
 static const char *hpd_clk_names_8960[] = {"core_clk", "master_iface_clk", "slave_iface_clk"};
 
 static struct hdmi_platform_config hdmi_tx_8960_config = {
-		.phy_init = hdmi_phy_8960_init,
 		HDMI_CFG(hpd_reg, 8960),
 		HDMI_CFG(hpd_clk, 8960),
 };
@@ -351,7 +372,6 @@ static const char *hpd_clk_names_8x74[] = {"iface_clk", "core_clk", "mdp_core_cl
 static unsigned long hpd_clk_freq_8x74[] = {0, 19200000, 0};
 
 static struct hdmi_platform_config hdmi_tx_8974_config = {
-		.phy_init = hdmi_phy_8x74_init,
 		HDMI_CFG(pwr_reg, 8x74),
 		HDMI_CFG(hpd_reg, 8x74),
 		HDMI_CFG(pwr_clk, 8x74),
@@ -362,7 +382,6 @@ static struct hdmi_platform_config hdmi_tx_8974_config = {
 static const char *hpd_reg_names_8084[] = {"hpd-gdsc", "hpd-5v", "hpd-5v-en"};
 
 static struct hdmi_platform_config hdmi_tx_8084_config = {
-		.phy_init = hdmi_phy_8x74_init,
 		HDMI_CFG(pwr_reg, 8x74),
 		HDMI_CFG(hpd_reg, 8084),
 		HDMI_CFG(pwr_clk, 8x74),
@@ -371,7 +390,6 @@ static struct hdmi_platform_config hdmi_tx_8084_config = {
 };
 
 static struct hdmi_platform_config hdmi_tx_8994_config = {
-		.phy_init = NULL, /* nothing to do for this HDMI PHY 20nm */
 		HDMI_CFG(pwr_reg, 8x74),
 		HDMI_CFG(hpd_reg, none),
 		HDMI_CFG(pwr_clk, 8x74),
@@ -380,7 +398,6 @@ static struct hdmi_platform_config hdmi_tx_8994_config = {
 };
 
 static struct hdmi_platform_config hdmi_tx_8996_config = {
-		.phy_init = NULL,
 		HDMI_CFG(pwr_reg, none),
 		HDMI_CFG(hpd_reg, none),
 		HDMI_CFG(pwr_clk, 8x74),
@@ -388,7 +405,21 @@ static struct hdmi_platform_config hdmi_tx_8996_config = {
 		.hpd_freq      = hpd_clk_freq_8x74,
 };
 
-static int get_gpio(struct device *dev, struct device_node *of_node, const char *name)
+static const struct {
+	const char *name;
+	const bool output;
+	const int value;
+	const char *label;
+} msm_hdmi_gpio_pdata[] = {
+	{ "qcom,hdmi-tx-ddc-clk", true, 1, "HDMI_DDC_CLK" },
+	{ "qcom,hdmi-tx-ddc-data", true, 1, "HDMI_DDC_DATA" },
+	{ "qcom,hdmi-tx-hpd", false, 1, "HDMI_HPD" },
+	{ "qcom,hdmi-tx-mux-en", true, 1, "HDMI_MUX_EN" },
+	{ "qcom,hdmi-tx-mux-sel", true, 0, "HDMI_MUX_SEL" },
+	{ "qcom,hdmi-tx-mux-lpm", true, 1, "HDMI_MUX_LPM" },
+};
+
+static int msm_hdmi_get_gpio(struct device_node *of_node, const char *name)
 {
 	int gpio = of_get_named_gpio(of_node, name, 0);
 	if (gpio < 0) {
@@ -403,13 +434,14 @@ static int get_gpio(struct device *dev, struct device_node *of_node, const char
 	return gpio;
 }
 
-static int hdmi_bind(struct device *dev, struct device *master, void *data)
+static int msm_hdmi_bind(struct device *dev, struct device *master, void *data)
 {
 	struct drm_device *drm = dev_get_drvdata(master);
 	struct msm_drm_private *priv = drm->dev_private;
 	static struct hdmi_platform_config *hdmi_cfg;
 	struct hdmi *hdmi;
 	struct device_node *of_node = dev->of_node;
+	int i;
 
 	hdmi_cfg = (struct hdmi_platform_config *)
 			of_device_get_match_data(dev);
@@ -420,16 +452,18 @@ static int hdmi_bind(struct device *dev, struct device *master, void *data)
 
 	hdmi_cfg->mmio_name     = "core_physical";
 	hdmi_cfg->qfprom_mmio_name = "qfprom_physical";
-	hdmi_cfg->ddc_clk_gpio  = get_gpio(dev, of_node, "qcom,hdmi-tx-ddc-clk");
-	hdmi_cfg->ddc_data_gpio = get_gpio(dev, of_node, "qcom,hdmi-tx-ddc-data");
-	hdmi_cfg->hpd_gpio      = get_gpio(dev, of_node, "qcom,hdmi-tx-hpd");
-	hdmi_cfg->mux_en_gpio   = get_gpio(dev, of_node, "qcom,hdmi-tx-mux-en");
-	hdmi_cfg->mux_sel_gpio  = get_gpio(dev, of_node, "qcom,hdmi-tx-mux-sel");
-	hdmi_cfg->mux_lpm_gpio  = get_gpio(dev, of_node, "qcom,hdmi-tx-mux-lpm");
+
+	for (i = 0; i < HDMI_MAX_NUM_GPIO; i++) {
+		hdmi_cfg->gpios[i].num = msm_hdmi_get_gpio(of_node,
+						msm_hdmi_gpio_pdata[i].name);
+		hdmi_cfg->gpios[i].output = msm_hdmi_gpio_pdata[i].output;
+		hdmi_cfg->gpios[i].value = msm_hdmi_gpio_pdata[i].value;
+		hdmi_cfg->gpios[i].label = msm_hdmi_gpio_pdata[i].label;
+	}
 
 	dev->platform_data = hdmi_cfg;
 
-	hdmi = hdmi_init(to_platform_device(dev));
+	hdmi = msm_hdmi_init(to_platform_device(dev));
 	if (IS_ERR(hdmi))
 		return PTR_ERR(hdmi);
 	priv->hdmi = hdmi;
@@ -437,34 +471,34 @@ static int hdmi_bind(struct device *dev, struct device *master, void *data)
 	return 0;
 }
 
-static void hdmi_unbind(struct device *dev, struct device *master,
+static void msm_hdmi_unbind(struct device *dev, struct device *master,
 		void *data)
 {
 	struct drm_device *drm = dev_get_drvdata(master);
 	struct msm_drm_private *priv = drm->dev_private;
 	if (priv->hdmi) {
-		hdmi_destroy(priv->hdmi);
+		msm_hdmi_destroy(priv->hdmi);
 		priv->hdmi = NULL;
 	}
 }
 
-static const struct component_ops hdmi_ops = {
-		.bind   = hdmi_bind,
-		.unbind = hdmi_unbind,
+static const struct component_ops msm_hdmi_ops = {
+		.bind   = msm_hdmi_bind,
+		.unbind = msm_hdmi_unbind,
 };
 
-static int hdmi_dev_probe(struct platform_device *pdev)
+static int msm_hdmi_dev_probe(struct platform_device *pdev)
 {
-	return component_add(&pdev->dev, &hdmi_ops);
+	return component_add(&pdev->dev, &msm_hdmi_ops);
 }
 
-static int hdmi_dev_remove(struct platform_device *pdev)
+static int msm_hdmi_dev_remove(struct platform_device *pdev)
 {
-	component_del(&pdev->dev, &hdmi_ops);
+	component_del(&pdev->dev, &msm_hdmi_ops);
 	return 0;
 }
 
-static const struct of_device_id dt_match[] = {
+static const struct of_device_id msm_hdmi_dt_match[] = {
 	{ .compatible = "qcom,hdmi-tx-8996", .data = &hdmi_tx_8996_config },
 	{ .compatible = "qcom,hdmi-tx-8994", .data = &hdmi_tx_8994_config },
 	{ .compatible = "qcom,hdmi-tx-8084", .data = &hdmi_tx_8084_config },
@@ -474,21 +508,23 @@ static const struct of_device_id dt_match[] = {
 	{}
 };
 
-static struct platform_driver hdmi_driver = {
-	.probe = hdmi_dev_probe,
-	.remove = hdmi_dev_remove,
+static struct platform_driver msm_hdmi_driver = {
+	.probe = msm_hdmi_dev_probe,
+	.remove = msm_hdmi_dev_remove,
 	.driver = {
 		.name = "hdmi_msm",
-		.of_match_table = dt_match,
+		.of_match_table = msm_hdmi_dt_match,
 	},
 };
 
-void __init hdmi_register(void)
+void __init msm_hdmi_register(void)
 {
-	platform_driver_register(&hdmi_driver);
+	msm_hdmi_phy_driver_register();
+	platform_driver_register(&msm_hdmi_driver);
 }
 
-void __exit hdmi_unregister(void)
+void __exit msm_hdmi_unregister(void)
 {
-	platform_driver_unregister(&hdmi_driver);
+	platform_driver_unregister(&msm_hdmi_driver);
+	msm_hdmi_phy_driver_unregister();
 }
diff --git a/drivers/gpu/drm/msm/hdmi/hdmi.h b/drivers/gpu/drm/msm/hdmi/hdmi.h
index d0e663192d01..b04a64664673 100644
--- a/drivers/gpu/drm/msm/hdmi/hdmi.h
+++ b/drivers/gpu/drm/msm/hdmi/hdmi.h
@@ -27,10 +27,18 @@
 #include "msm_drv.h"
 #include "hdmi.xml.h"
 
+#define HDMI_MAX_NUM_GPIO	6
 
 struct hdmi_phy;
 struct hdmi_platform_config;
 
+struct hdmi_gpio_data {
+	int num;
+	bool output;
+	int value;
+	const char *label;
+};
+
 struct hdmi_audio {
 	bool enabled;
 	struct hdmi_audio_infoframe infoframe;
@@ -62,6 +70,8 @@ struct hdmi {
 	struct clk **pwr_clks;
 
 	struct hdmi_phy *phy;
+	struct device *phy_dev;
+
 	struct i2c_adapter *i2c;
 	struct drm_connector *connector;
 	struct drm_bridge *bridge;
@@ -88,7 +98,6 @@ struct hdmi {
 
 /* platform config data (ie. from DT, or pdata) */
 struct hdmi_platform_config {
-	struct hdmi_phy *(*phy_init)(struct hdmi *hdmi);
 	const char *mmio_name;
 	const char *qfprom_mmio_name;
 
@@ -110,11 +119,10 @@ struct hdmi_platform_config {
 	int pwr_clk_cnt;
 
 	/* gpio's: */
-	int ddc_clk_gpio, ddc_data_gpio, hpd_gpio, mux_en_gpio, mux_sel_gpio;
-	int mux_lpm_gpio;
+	struct hdmi_gpio_data gpios[HDMI_MAX_NUM_GPIO];
 };
 
-void hdmi_set_mode(struct hdmi *hdmi, bool power_on);
+void msm_hdmi_set_mode(struct hdmi *hdmi, bool power_on);
 
 static inline void hdmi_write(struct hdmi *hdmi, u32 reg, u32 data)
 {
@@ -132,65 +140,113 @@ static inline u32 hdmi_qfprom_read(struct hdmi *hdmi, u32 reg)
 }
 
 /*
- * The phy appears to be different, for example between 8960 and 8x60,
- * so split the phy related functions out and load the correct one at
- * runtime:
+ * hdmi phy:
  */
 
-struct hdmi_phy_funcs {
-	void (*destroy)(struct hdmi_phy *phy);
+enum hdmi_phy_type {
+	MSM_HDMI_PHY_8x60,
+	MSM_HDMI_PHY_8960,
+	MSM_HDMI_PHY_8x74,
+	MSM_HDMI_PHY_8996,
+	MSM_HDMI_PHY_MAX,
+};
+
+struct hdmi_phy_cfg {
+	enum hdmi_phy_type type;
 	void (*powerup)(struct hdmi_phy *phy, unsigned long int pixclock);
 	void (*powerdown)(struct hdmi_phy *phy);
+	const char * const *reg_names;
+	int num_regs;
+	const char * const *clk_names;
+	int num_clks;
 };
 
+extern const struct hdmi_phy_cfg msm_hdmi_phy_8x60_cfg;
+extern const struct hdmi_phy_cfg msm_hdmi_phy_8960_cfg;
+extern const struct hdmi_phy_cfg msm_hdmi_phy_8x74_cfg;
+extern const struct hdmi_phy_cfg msm_hdmi_phy_8996_cfg;
+
 struct hdmi_phy {
+	struct platform_device *pdev;
+	void __iomem *mmio;
+	struct hdmi_phy_cfg *cfg;
 	const struct hdmi_phy_funcs *funcs;
+	struct regulator **regs;
+	struct clk **clks;
 };
 
-struct hdmi_phy *hdmi_phy_8960_init(struct hdmi *hdmi);
-struct hdmi_phy *hdmi_phy_8x60_init(struct hdmi *hdmi);
-struct hdmi_phy *hdmi_phy_8x74_init(struct hdmi *hdmi);
+static inline void hdmi_phy_write(struct hdmi_phy *phy, u32 reg, u32 data)
+{
+	msm_writel(data, phy->mmio + reg);
+}
+
+static inline u32 hdmi_phy_read(struct hdmi_phy *phy, u32 reg)
+{
+	return msm_readl(phy->mmio + reg);
+}
+
+int msm_hdmi_phy_resource_enable(struct hdmi_phy *phy);
+void msm_hdmi_phy_resource_disable(struct hdmi_phy *phy);
+void msm_hdmi_phy_powerup(struct hdmi_phy *phy, unsigned long int pixclock);
+void msm_hdmi_phy_powerdown(struct hdmi_phy *phy);
+void __init msm_hdmi_phy_driver_register(void);
+void __exit msm_hdmi_phy_driver_unregister(void);
+
+#ifdef CONFIG_COMMON_CLK
+int msm_hdmi_pll_8960_init(struct platform_device *pdev);
+int msm_hdmi_pll_8996_init(struct platform_device *pdev);
+#else
+static inline int msm_hdmi_pll_8960_init(struct platform_device *pdev);
+{
+	return -ENODEV;
+}
+
+static inline int msm_hdmi_pll_8996_init(struct platform_device *pdev)
+{
+	return -ENODEV;
+}
+#endif
 
 /*
  * audio:
  */
 
-int hdmi_audio_update(struct hdmi *hdmi);
-int hdmi_audio_info_setup(struct hdmi *hdmi, bool enabled,
+int msm_hdmi_audio_update(struct hdmi *hdmi);
+int msm_hdmi_audio_info_setup(struct hdmi *hdmi, bool enabled,
 	uint32_t num_of_channels, uint32_t channel_allocation,
 	uint32_t level_shift, bool down_mix);
-void hdmi_audio_set_sample_rate(struct hdmi *hdmi, int rate);
+void msm_hdmi_audio_set_sample_rate(struct hdmi *hdmi, int rate);
 
 
 /*
  * hdmi bridge:
  */
 
-struct drm_bridge *hdmi_bridge_init(struct hdmi *hdmi);
-void hdmi_bridge_destroy(struct drm_bridge *bridge);
+struct drm_bridge *msm_hdmi_bridge_init(struct hdmi *hdmi);
+void msm_hdmi_bridge_destroy(struct drm_bridge *bridge);
 
 /*
  * hdmi connector:
  */
 
-void hdmi_connector_irq(struct drm_connector *connector);
-struct drm_connector *hdmi_connector_init(struct hdmi *hdmi);
+void msm_hdmi_connector_irq(struct drm_connector *connector);
+struct drm_connector *msm_hdmi_connector_init(struct hdmi *hdmi);
 
 /*
  * i2c adapter for ddc:
  */
 
-void hdmi_i2c_irq(struct i2c_adapter *i2c);
-void hdmi_i2c_destroy(struct i2c_adapter *i2c);
-struct i2c_adapter *hdmi_i2c_init(struct hdmi *hdmi);
+void msm_hdmi_i2c_irq(struct i2c_adapter *i2c);
+void msm_hdmi_i2c_destroy(struct i2c_adapter *i2c);
+struct i2c_adapter *msm_hdmi_i2c_init(struct hdmi *hdmi);
 
 /*
  * hdcp
  */
-struct hdmi_hdcp_ctrl *hdmi_hdcp_init(struct hdmi *hdmi);
-void hdmi_hdcp_destroy(struct hdmi *hdmi);
-void hdmi_hdcp_on(struct hdmi_hdcp_ctrl *hdcp_ctrl);
-void hdmi_hdcp_off(struct hdmi_hdcp_ctrl *hdcp_ctrl);
-void hdmi_hdcp_irq(struct hdmi_hdcp_ctrl *hdcp_ctrl);
+struct hdmi_hdcp_ctrl *msm_hdmi_hdcp_init(struct hdmi *hdmi);
+void msm_hdmi_hdcp_destroy(struct hdmi *hdmi);
+void msm_hdmi_hdcp_on(struct hdmi_hdcp_ctrl *hdcp_ctrl);
+void msm_hdmi_hdcp_off(struct hdmi_hdcp_ctrl *hdcp_ctrl);
+void msm_hdmi_hdcp_irq(struct hdmi_hdcp_ctrl *hdcp_ctrl);
 
 #endif /* __HDMI_CONNECTOR_H__ */
diff --git a/drivers/gpu/drm/msm/hdmi/hdmi.xml.h b/drivers/gpu/drm/msm/hdmi/hdmi.xml.h
index 10c45700aefe..34c7df6549c1 100644
--- a/drivers/gpu/drm/msm/hdmi/hdmi.xml.h
+++ b/drivers/gpu/drm/msm/hdmi/hdmi.xml.h
@@ -9,7 +9,7 @@ git clone https://github.com/freedreno/envytools.git
 
 The rules-ng-ng source files this header was generated from are:
 - /home/robclark/src/freedreno/envytools/rnndb/msm.xml                 (    676 bytes, from 2015-05-20 20:03:14)
-- /home/robclark/src/freedreno/envytools/rnndb/freedreno_copyright.xml (   1453 bytes, from 2015-05-20 20:03:07)
+- /home/robclark/src/freedreno/envytools/rnndb/freedreno_copyright.xml (   1572 bytes, from 2016-02-10 17:07:21)
 - /home/robclark/src/freedreno/envytools/rnndb/mdp/mdp4.xml            (  20915 bytes, from 2015-05-20 20:03:14)
 - /home/robclark/src/freedreno/envytools/rnndb/mdp/mdp_common.xml      (   2849 bytes, from 2015-09-18 12:07:28)
 - /home/robclark/src/freedreno/envytools/rnndb/mdp/mdp5.xml            (  37194 bytes, from 2015-09-18 12:07:28)
@@ -17,11 +17,12 @@ The rules-ng-ng source files this header was generated from are:
 - /home/robclark/src/freedreno/envytools/rnndb/dsi/sfpb.xml            (    602 bytes, from 2015-10-22 16:35:02)
 - /home/robclark/src/freedreno/envytools/rnndb/dsi/mmss_cc.xml         (   1686 bytes, from 2015-05-20 20:03:14)
 - /home/robclark/src/freedreno/envytools/rnndb/hdmi/qfprom.xml         (    600 bytes, from 2015-05-20 20:03:07)
-- /home/robclark/src/freedreno/envytools/rnndb/hdmi/hdmi.xml           (  29154 bytes, from 2015-08-10 21:25:43)
+- /home/robclark/src/freedreno/envytools/rnndb/hdmi/hdmi.xml           (  41472 bytes, from 2016-01-22 18:18:18)
 - /home/robclark/src/freedreno/envytools/rnndb/edp/edp.xml             (  10416 bytes, from 2015-05-20 20:03:14)
 
-Copyright (C) 2013-2015 by the following authors:
+Copyright (C) 2013-2016 by the following authors:
 - Rob Clark <robdclark@gmail.com> (robclark)
+- Ilia Mirkin <imirkin@alum.mit.edu> (imirkin)
 
 Permission is hereby granted, free of charge, to any person obtaining
 a copy of this software and associated documentation files (the
@@ -559,7 +560,7 @@ static inline uint32_t HDMI_VSYNC_TOTAL_F2_V_TOTAL(uint32_t val)
 
 #define REG_HDMI_CEC_WR_CHECK_CONFIG				0x00000370
 
-#define REG_HDMI_8x60_PHY_REG0					0x00000300
+#define REG_HDMI_8x60_PHY_REG0					0x00000000
 #define HDMI_8x60_PHY_REG0_DESER_DEL_CTRL__MASK			0x0000001c
 #define HDMI_8x60_PHY_REG0_DESER_DEL_CTRL__SHIFT		2
 static inline uint32_t HDMI_8x60_PHY_REG0_DESER_DEL_CTRL(uint32_t val)
@@ -567,7 +568,7 @@ static inline uint32_t HDMI_8x60_PHY_REG0_DESER_DEL_CTRL(uint32_t val)
 	return ((val) << HDMI_8x60_PHY_REG0_DESER_DEL_CTRL__SHIFT) & HDMI_8x60_PHY_REG0_DESER_DEL_CTRL__MASK;
 }
 
-#define REG_HDMI_8x60_PHY_REG1					0x00000304
+#define REG_HDMI_8x60_PHY_REG1					0x00000004
 #define HDMI_8x60_PHY_REG1_DTEST_MUX_SEL__MASK			0x000000f0
 #define HDMI_8x60_PHY_REG1_DTEST_MUX_SEL__SHIFT			4
 static inline uint32_t HDMI_8x60_PHY_REG1_DTEST_MUX_SEL(uint32_t val)
@@ -581,7 +582,7 @@ static inline uint32_t HDMI_8x60_PHY_REG1_OUTVOL_SWING_CTRL(uint32_t val)
 	return ((val) << HDMI_8x60_PHY_REG1_OUTVOL_SWING_CTRL__SHIFT) & HDMI_8x60_PHY_REG1_OUTVOL_SWING_CTRL__MASK;
 }
 
-#define REG_HDMI_8x60_PHY_REG2					0x00000308
+#define REG_HDMI_8x60_PHY_REG2					0x00000008
 #define HDMI_8x60_PHY_REG2_PD_DESER				0x00000001
 #define HDMI_8x60_PHY_REG2_PD_DRIVE_1				0x00000002
 #define HDMI_8x60_PHY_REG2_PD_DRIVE_2				0x00000004
@@ -591,152 +592,152 @@ static inline uint32_t HDMI_8x60_PHY_REG1_OUTVOL_SWING_CTRL(uint32_t val)
 #define HDMI_8x60_PHY_REG2_PD_PWRGEN				0x00000040
 #define HDMI_8x60_PHY_REG2_RCV_SENSE_EN				0x00000080
 
-#define REG_HDMI_8x60_PHY_REG3					0x0000030c
+#define REG_HDMI_8x60_PHY_REG3					0x0000000c
 #define HDMI_8x60_PHY_REG3_PLL_ENABLE				0x00000001
 
-#define REG_HDMI_8x60_PHY_REG4					0x00000310
+#define REG_HDMI_8x60_PHY_REG4					0x00000010
 
-#define REG_HDMI_8x60_PHY_REG5					0x00000314
+#define REG_HDMI_8x60_PHY_REG5					0x00000014
 
-#define REG_HDMI_8x60_PHY_REG6					0x00000318
+#define REG_HDMI_8x60_PHY_REG6					0x00000018
 
-#define REG_HDMI_8x60_PHY_REG7					0x0000031c
+#define REG_HDMI_8x60_PHY_REG7					0x0000001c
 
-#define REG_HDMI_8x60_PHY_REG8					0x00000320
+#define REG_HDMI_8x60_PHY_REG8					0x00000020
 
-#define REG_HDMI_8x60_PHY_REG9					0x00000324
+#define REG_HDMI_8x60_PHY_REG9					0x00000024
 
-#define REG_HDMI_8x60_PHY_REG10					0x00000328
+#define REG_HDMI_8x60_PHY_REG10					0x00000028
 
-#define REG_HDMI_8x60_PHY_REG11					0x0000032c
+#define REG_HDMI_8x60_PHY_REG11					0x0000002c
 
-#define REG_HDMI_8x60_PHY_REG12					0x00000330
+#define REG_HDMI_8x60_PHY_REG12					0x00000030
 #define HDMI_8x60_PHY_REG12_RETIMING_EN				0x00000001
 #define HDMI_8x60_PHY_REG12_PLL_LOCK_DETECT_EN			0x00000002
 #define HDMI_8x60_PHY_REG12_FORCE_LOCK				0x00000010
 
-#define REG_HDMI_8960_PHY_REG0					0x00000400
+#define REG_HDMI_8960_PHY_REG0					0x00000000
 
-#define REG_HDMI_8960_PHY_REG1					0x00000404
+#define REG_HDMI_8960_PHY_REG1					0x00000004
 
-#define REG_HDMI_8960_PHY_REG2					0x00000408
+#define REG_HDMI_8960_PHY_REG2					0x00000008
 
-#define REG_HDMI_8960_PHY_REG3					0x0000040c
+#define REG_HDMI_8960_PHY_REG3					0x0000000c
 
-#define REG_HDMI_8960_PHY_REG4					0x00000410
+#define REG_HDMI_8960_PHY_REG4					0x00000010
 
-#define REG_HDMI_8960_PHY_REG5					0x00000414
+#define REG_HDMI_8960_PHY_REG5					0x00000014
 
-#define REG_HDMI_8960_PHY_REG6					0x00000418
+#define REG_HDMI_8960_PHY_REG6					0x00000018
 
-#define REG_HDMI_8960_PHY_REG7					0x0000041c
+#define REG_HDMI_8960_PHY_REG7					0x0000001c
 
-#define REG_HDMI_8960_PHY_REG8					0x00000420
+#define REG_HDMI_8960_PHY_REG8					0x00000020
 
-#define REG_HDMI_8960_PHY_REG9					0x00000424
+#define REG_HDMI_8960_PHY_REG9					0x00000024
 
-#define REG_HDMI_8960_PHY_REG10					0x00000428
+#define REG_HDMI_8960_PHY_REG10					0x00000028
 
-#define REG_HDMI_8960_PHY_REG11					0x0000042c
+#define REG_HDMI_8960_PHY_REG11					0x0000002c
 
-#define REG_HDMI_8960_PHY_REG12					0x00000430
+#define REG_HDMI_8960_PHY_REG12					0x00000030
 #define HDMI_8960_PHY_REG12_SW_RESET				0x00000020
 #define HDMI_8960_PHY_REG12_PWRDN_B				0x00000080
 
-#define REG_HDMI_8960_PHY_REG_BIST_CFG				0x00000434
+#define REG_HDMI_8960_PHY_REG_BIST_CFG				0x00000034
 
-#define REG_HDMI_8960_PHY_DEBUG_BUS_SEL				0x00000438
+#define REG_HDMI_8960_PHY_DEBUG_BUS_SEL				0x00000038
 
-#define REG_HDMI_8960_PHY_REG_MISC0				0x0000043c
+#define REG_HDMI_8960_PHY_REG_MISC0				0x0000003c
 
-#define REG_HDMI_8960_PHY_REG13					0x00000440
+#define REG_HDMI_8960_PHY_REG13					0x00000040
 
-#define REG_HDMI_8960_PHY_REG14					0x00000444
+#define REG_HDMI_8960_PHY_REG14					0x00000044
 
-#define REG_HDMI_8960_PHY_REG15					0x00000448
+#define REG_HDMI_8960_PHY_REG15					0x00000048
 
-#define REG_HDMI_8960_PHY_PLL_REFCLK_CFG			0x00000500
+#define REG_HDMI_8960_PHY_PLL_REFCLK_CFG			0x00000000
 
-#define REG_HDMI_8960_PHY_PLL_CHRG_PUMP_CFG			0x00000504
+#define REG_HDMI_8960_PHY_PLL_CHRG_PUMP_CFG			0x00000004
 
-#define REG_HDMI_8960_PHY_PLL_LOOP_FLT_CFG0			0x00000508
+#define REG_HDMI_8960_PHY_PLL_LOOP_FLT_CFG0			0x00000008
 
-#define REG_HDMI_8960_PHY_PLL_LOOP_FLT_CFG1			0x0000050c
+#define REG_HDMI_8960_PHY_PLL_LOOP_FLT_CFG1			0x0000000c
 
-#define REG_HDMI_8960_PHY_PLL_IDAC_ADJ_CFG			0x00000510
+#define REG_HDMI_8960_PHY_PLL_IDAC_ADJ_CFG			0x00000010
 
-#define REG_HDMI_8960_PHY_PLL_I_VI_KVCO_CFG			0x00000514
+#define REG_HDMI_8960_PHY_PLL_I_VI_KVCO_CFG			0x00000014
 
-#define REG_HDMI_8960_PHY_PLL_PWRDN_B				0x00000518
+#define REG_HDMI_8960_PHY_PLL_PWRDN_B				0x00000018
 #define HDMI_8960_PHY_PLL_PWRDN_B_PD_PLL			0x00000002
 #define HDMI_8960_PHY_PLL_PWRDN_B_PLL_PWRDN_B			0x00000008
 
-#define REG_HDMI_8960_PHY_PLL_SDM_CFG0				0x0000051c
+#define REG_HDMI_8960_PHY_PLL_SDM_CFG0				0x0000001c
 
-#define REG_HDMI_8960_PHY_PLL_SDM_CFG1				0x00000520
+#define REG_HDMI_8960_PHY_PLL_SDM_CFG1				0x00000020
 
-#define REG_HDMI_8960_PHY_PLL_SDM_CFG2				0x00000524
+#define REG_HDMI_8960_PHY_PLL_SDM_CFG2				0x00000024
 
-#define REG_HDMI_8960_PHY_PLL_SDM_CFG3				0x00000528
+#define REG_HDMI_8960_PHY_PLL_SDM_CFG3				0x00000028
 
-#define REG_HDMI_8960_PHY_PLL_SDM_CFG4				0x0000052c
+#define REG_HDMI_8960_PHY_PLL_SDM_CFG4				0x0000002c
 
-#define REG_HDMI_8960_PHY_PLL_SSC_CFG0				0x00000530
+#define REG_HDMI_8960_PHY_PLL_SSC_CFG0				0x00000030
 
-#define REG_HDMI_8960_PHY_PLL_SSC_CFG1				0x00000534
+#define REG_HDMI_8960_PHY_PLL_SSC_CFG1				0x00000034
 
-#define REG_HDMI_8960_PHY_PLL_SSC_CFG2				0x00000538
+#define REG_HDMI_8960_PHY_PLL_SSC_CFG2				0x00000038
 
-#define REG_HDMI_8960_PHY_PLL_SSC_CFG3				0x0000053c
+#define REG_HDMI_8960_PHY_PLL_SSC_CFG3				0x0000003c
 
-#define REG_HDMI_8960_PHY_PLL_LOCKDET_CFG0			0x00000540
+#define REG_HDMI_8960_PHY_PLL_LOCKDET_CFG0			0x00000040
 
-#define REG_HDMI_8960_PHY_PLL_LOCKDET_CFG1			0x00000544
+#define REG_HDMI_8960_PHY_PLL_LOCKDET_CFG1			0x00000044
 
-#define REG_HDMI_8960_PHY_PLL_LOCKDET_CFG2			0x00000548
+#define REG_HDMI_8960_PHY_PLL_LOCKDET_CFG2			0x00000048
 
-#define REG_HDMI_8960_PHY_PLL_VCOCAL_CFG0			0x0000054c
+#define REG_HDMI_8960_PHY_PLL_VCOCAL_CFG0			0x0000004c
 
-#define REG_HDMI_8960_PHY_PLL_VCOCAL_CFG1			0x00000550
+#define REG_HDMI_8960_PHY_PLL_VCOCAL_CFG1			0x00000050
 
-#define REG_HDMI_8960_PHY_PLL_VCOCAL_CFG2			0x00000554
+#define REG_HDMI_8960_PHY_PLL_VCOCAL_CFG2			0x00000054
 
-#define REG_HDMI_8960_PHY_PLL_VCOCAL_CFG3			0x00000558
+#define REG_HDMI_8960_PHY_PLL_VCOCAL_CFG3			0x00000058
 
-#define REG_HDMI_8960_PHY_PLL_VCOCAL_CFG4			0x0000055c
+#define REG_HDMI_8960_PHY_PLL_VCOCAL_CFG4			0x0000005c
 
-#define REG_HDMI_8960_PHY_PLL_VCOCAL_CFG5			0x00000560
+#define REG_HDMI_8960_PHY_PLL_VCOCAL_CFG5			0x00000060
 
-#define REG_HDMI_8960_PHY_PLL_VCOCAL_CFG6			0x00000564
+#define REG_HDMI_8960_PHY_PLL_VCOCAL_CFG6			0x00000064
 
-#define REG_HDMI_8960_PHY_PLL_VCOCAL_CFG7			0x00000568
+#define REG_HDMI_8960_PHY_PLL_VCOCAL_CFG7			0x00000068
 
-#define REG_HDMI_8960_PHY_PLL_DEBUG_SEL				0x0000056c
+#define REG_HDMI_8960_PHY_PLL_DEBUG_SEL				0x0000006c
 
-#define REG_HDMI_8960_PHY_PLL_MISC0				0x00000570
+#define REG_HDMI_8960_PHY_PLL_MISC0				0x00000070
 
-#define REG_HDMI_8960_PHY_PLL_MISC1				0x00000574
+#define REG_HDMI_8960_PHY_PLL_MISC1				0x00000074
 
-#define REG_HDMI_8960_PHY_PLL_MISC2				0x00000578
+#define REG_HDMI_8960_PHY_PLL_MISC2				0x00000078
 
-#define REG_HDMI_8960_PHY_PLL_MISC3				0x0000057c
+#define REG_HDMI_8960_PHY_PLL_MISC3				0x0000007c
 
-#define REG_HDMI_8960_PHY_PLL_MISC4				0x00000580
+#define REG_HDMI_8960_PHY_PLL_MISC4				0x00000080
 
-#define REG_HDMI_8960_PHY_PLL_MISC5				0x00000584
+#define REG_HDMI_8960_PHY_PLL_MISC5				0x00000084
 
-#define REG_HDMI_8960_PHY_PLL_MISC6				0x00000588
+#define REG_HDMI_8960_PHY_PLL_MISC6				0x00000088
 
-#define REG_HDMI_8960_PHY_PLL_DEBUG_BUS0			0x0000058c
+#define REG_HDMI_8960_PHY_PLL_DEBUG_BUS0			0x0000008c
 
-#define REG_HDMI_8960_PHY_PLL_DEBUG_BUS1			0x00000590
+#define REG_HDMI_8960_PHY_PLL_DEBUG_BUS1			0x00000090
 
-#define REG_HDMI_8960_PHY_PLL_DEBUG_BUS2			0x00000594
+#define REG_HDMI_8960_PHY_PLL_DEBUG_BUS2			0x00000094
 
-#define REG_HDMI_8960_PHY_PLL_STATUS0				0x00000598
+#define REG_HDMI_8960_PHY_PLL_STATUS0				0x00000098
 #define HDMI_8960_PHY_PLL_STATUS0_PLL_LOCK			0x00000001
 
-#define REG_HDMI_8960_PHY_PLL_STATUS1				0x0000059c
+#define REG_HDMI_8960_PHY_PLL_STATUS1				0x0000009c
 
 #define REG_HDMI_8x74_ANA_CFG0					0x00000000
 
@@ -843,5 +844,501 @@ static inline uint32_t HDMI_8x60_PHY_REG1_OUTVOL_SWING_CTRL(uint32_t val)
 
 #define REG_HDMI_28nm_PHY_PLL_DEBUG_BUS_SEL			0x000000a0
 
+#define REG_HDMI_8996_PHY_CFG					0x00000000
+
+#define REG_HDMI_8996_PHY_PD_CTL				0x00000004
+
+#define REG_HDMI_8996_PHY_MODE					0x00000008
+
+#define REG_HDMI_8996_PHY_MISR_CLEAR				0x0000000c
+
+#define REG_HDMI_8996_PHY_TX0_TX1_BIST_CFG0			0x00000010
+
+#define REG_HDMI_8996_PHY_TX0_TX1_BIST_CFG1			0x00000014
+
+#define REG_HDMI_8996_PHY_TX0_TX1_PRBS_SEED_BYTE0		0x00000018
+
+#define REG_HDMI_8996_PHY_TX0_TX1_PRBS_SEED_BYTE1		0x0000001c
+
+#define REG_HDMI_8996_PHY_TX0_TX1_BIST_PATTERN0			0x00000020
+
+#define REG_HDMI_8996_PHY_TX0_TX1_BIST_PATTERN1			0x00000024
+
+#define REG_HDMI_8996_PHY_TX2_TX3_BIST_CFG0			0x00000028
+
+#define REG_HDMI_8996_PHY_TX2_TX3_BIST_CFG1			0x0000002c
+
+#define REG_HDMI_8996_PHY_TX2_TX3_PRBS_SEED_BYTE0		0x00000030
+
+#define REG_HDMI_8996_PHY_TX2_TX3_PRBS_SEED_BYTE1		0x00000034
+
+#define REG_HDMI_8996_PHY_TX2_TX3_BIST_PATTERN0			0x00000038
+
+#define REG_HDMI_8996_PHY_TX2_TX3_BIST_PATTERN1			0x0000003c
+
+#define REG_HDMI_8996_PHY_DEBUG_BUS_SEL				0x00000040
+
+#define REG_HDMI_8996_PHY_TXCAL_CFG0				0x00000044
+
+#define REG_HDMI_8996_PHY_TXCAL_CFG1				0x00000048
+
+#define REG_HDMI_8996_PHY_TX0_TX1_LANE_CTL			0x0000004c
+
+#define REG_HDMI_8996_PHY_TX2_TX3_LANE_CTL			0x00000050
+
+#define REG_HDMI_8996_PHY_LANE_BIST_CONFIG			0x00000054
+
+#define REG_HDMI_8996_PHY_CLOCK					0x00000058
+
+#define REG_HDMI_8996_PHY_MISC1					0x0000005c
+
+#define REG_HDMI_8996_PHY_MISC2					0x00000060
+
+#define REG_HDMI_8996_PHY_TX0_TX1_BIST_STATUS0			0x00000064
+
+#define REG_HDMI_8996_PHY_TX0_TX1_BIST_STATUS1			0x00000068
+
+#define REG_HDMI_8996_PHY_TX0_TX1_BIST_STATUS2			0x0000006c
+
+#define REG_HDMI_8996_PHY_TX2_TX3_BIST_STATUS0			0x00000070
+
+#define REG_HDMI_8996_PHY_TX2_TX3_BIST_STATUS1			0x00000074
+
+#define REG_HDMI_8996_PHY_TX2_TX3_BIST_STATUS2			0x00000078
+
+#define REG_HDMI_8996_PHY_PRE_MISR_STATUS0			0x0000007c
+
+#define REG_HDMI_8996_PHY_PRE_MISR_STATUS1			0x00000080
+
+#define REG_HDMI_8996_PHY_PRE_MISR_STATUS2			0x00000084
+
+#define REG_HDMI_8996_PHY_PRE_MISR_STATUS3			0x00000088
+
+#define REG_HDMI_8996_PHY_POST_MISR_STATUS0			0x0000008c
+
+#define REG_HDMI_8996_PHY_POST_MISR_STATUS1			0x00000090
+
+#define REG_HDMI_8996_PHY_POST_MISR_STATUS2			0x00000094
+
+#define REG_HDMI_8996_PHY_POST_MISR_STATUS3			0x00000098
+
+#define REG_HDMI_8996_PHY_STATUS				0x0000009c
+
+#define REG_HDMI_8996_PHY_MISC3_STATUS				0x000000a0
+
+#define REG_HDMI_8996_PHY_MISC4_STATUS				0x000000a4
+
+#define REG_HDMI_8996_PHY_DEBUG_BUS0				0x000000a8
+
+#define REG_HDMI_8996_PHY_DEBUG_BUS1				0x000000ac
+
+#define REG_HDMI_8996_PHY_DEBUG_BUS2				0x000000b0
+
+#define REG_HDMI_8996_PHY_DEBUG_BUS3				0x000000b4
+
+#define REG_HDMI_8996_PHY_PHY_REVISION_ID0			0x000000b8
+
+#define REG_HDMI_8996_PHY_PHY_REVISION_ID1			0x000000bc
+
+#define REG_HDMI_8996_PHY_PHY_REVISION_ID2			0x000000c0
+
+#define REG_HDMI_8996_PHY_PHY_REVISION_ID3			0x000000c4
+
+#define REG_HDMI_PHY_QSERDES_COM_ATB_SEL1			0x00000000
+
+#define REG_HDMI_PHY_QSERDES_COM_ATB_SEL2			0x00000004
+
+#define REG_HDMI_PHY_QSERDES_COM_FREQ_UPDATE			0x00000008
+
+#define REG_HDMI_PHY_QSERDES_COM_BG_TIMER			0x0000000c
+
+#define REG_HDMI_PHY_QSERDES_COM_SSC_EN_CENTER			0x00000010
+
+#define REG_HDMI_PHY_QSERDES_COM_SSC_ADJ_PER1			0x00000014
+
+#define REG_HDMI_PHY_QSERDES_COM_SSC_ADJ_PER2			0x00000018
+
+#define REG_HDMI_PHY_QSERDES_COM_SSC_PER1			0x0000001c
+
+#define REG_HDMI_PHY_QSERDES_COM_SSC_PER2			0x00000020
+
+#define REG_HDMI_PHY_QSERDES_COM_SSC_STEP_SIZE1			0x00000024
+
+#define REG_HDMI_PHY_QSERDES_COM_SSC_STEP_SIZE2			0x00000028
+
+#define REG_HDMI_PHY_QSERDES_COM_POST_DIV			0x0000002c
+
+#define REG_HDMI_PHY_QSERDES_COM_POST_DIV_MUX			0x00000030
+
+#define REG_HDMI_PHY_QSERDES_COM_BIAS_EN_CLKBUFLR_EN		0x00000034
+
+#define REG_HDMI_PHY_QSERDES_COM_CLK_ENABLE1			0x00000038
+
+#define REG_HDMI_PHY_QSERDES_COM_SYS_CLK_CTRL			0x0000003c
+
+#define REG_HDMI_PHY_QSERDES_COM_SYSCLK_BUF_ENABLE		0x00000040
+
+#define REG_HDMI_PHY_QSERDES_COM_PLL_EN				0x00000044
+
+#define REG_HDMI_PHY_QSERDES_COM_PLL_IVCO			0x00000048
+
+#define REG_HDMI_PHY_QSERDES_COM_LOCK_CMP1_MODE0		0x0000004c
+
+#define REG_HDMI_PHY_QSERDES_COM_LOCK_CMP2_MODE0		0x00000050
+
+#define REG_HDMI_PHY_QSERDES_COM_LOCK_CMP3_MODE0		0x00000054
+
+#define REG_HDMI_PHY_QSERDES_COM_LOCK_CMP1_MODE1		0x00000058
+
+#define REG_HDMI_PHY_QSERDES_COM_LOCK_CMP2_MODE1		0x0000005c
+
+#define REG_HDMI_PHY_QSERDES_COM_LOCK_CMP3_MODE1		0x00000060
+
+#define REG_HDMI_PHY_QSERDES_COM_LOCK_CMP1_MODE2		0x00000064
+
+#define REG_HDMI_PHY_QSERDES_COM_CMN_RSVD0			0x00000064
+
+#define REG_HDMI_PHY_QSERDES_COM_LOCK_CMP2_MODE2		0x00000068
+
+#define REG_HDMI_PHY_QSERDES_COM_EP_CLOCK_DETECT_CTRL		0x00000068
+
+#define REG_HDMI_PHY_QSERDES_COM_LOCK_CMP3_MODE2		0x0000006c
+
+#define REG_HDMI_PHY_QSERDES_COM_SYSCLK_DET_COMP_STATUS		0x0000006c
+
+#define REG_HDMI_PHY_QSERDES_COM_BG_TRIM			0x00000070
+
+#define REG_HDMI_PHY_QSERDES_COM_CLK_EP_DIV			0x00000074
+
+#define REG_HDMI_PHY_QSERDES_COM_CP_CTRL_MODE0			0x00000078
+
+#define REG_HDMI_PHY_QSERDES_COM_CP_CTRL_MODE1			0x0000007c
+
+#define REG_HDMI_PHY_QSERDES_COM_CP_CTRL_MODE2			0x00000080
+
+#define REG_HDMI_PHY_QSERDES_COM_CMN_RSVD1			0x00000080
+
+#define REG_HDMI_PHY_QSERDES_COM_PLL_RCTRL_MODE0		0x00000084
+
+#define REG_HDMI_PHY_QSERDES_COM_PLL_RCTRL_MODE1		0x00000088
+
+#define REG_HDMI_PHY_QSERDES_COM_PLL_RCTRL_MODE2		0x0000008c
+
+#define REG_HDMI_PHY_QSERDES_COM_CMN_RSVD2			0x0000008c
+
+#define REG_HDMI_PHY_QSERDES_COM_PLL_CCTRL_MODE0		0x00000090
+
+#define REG_HDMI_PHY_QSERDES_COM_PLL_CCTRL_MODE1		0x00000094
+
+#define REG_HDMI_PHY_QSERDES_COM_PLL_CCTRL_MODE2		0x00000098
+
+#define REG_HDMI_PHY_QSERDES_COM_CMN_RSVD3			0x00000098
+
+#define REG_HDMI_PHY_QSERDES_COM_PLL_CNTRL			0x0000009c
+
+#define REG_HDMI_PHY_QSERDES_COM_PHASE_SEL_CTRL			0x000000a0
+
+#define REG_HDMI_PHY_QSERDES_COM_PHASE_SEL_DC			0x000000a4
+
+#define REG_HDMI_PHY_QSERDES_COM_CORE_CLK_IN_SYNC_SEL		0x000000a8
+
+#define REG_HDMI_PHY_QSERDES_COM_BIAS_EN_CTRL_BY_PSM		0x000000a8
+
+#define REG_HDMI_PHY_QSERDES_COM_SYSCLK_EN_SEL			0x000000ac
+
+#define REG_HDMI_PHY_QSERDES_COM_CML_SYSCLK_SEL			0x000000b0
+
+#define REG_HDMI_PHY_QSERDES_COM_RESETSM_CNTRL			0x000000b4
+
+#define REG_HDMI_PHY_QSERDES_COM_RESETSM_CNTRL2			0x000000b8
+
+#define REG_HDMI_PHY_QSERDES_COM_RESTRIM_CTRL			0x000000bc
+
+#define REG_HDMI_PHY_QSERDES_COM_RESTRIM_CTRL2			0x000000c0
+
+#define REG_HDMI_PHY_QSERDES_COM_RESCODE_DIV_NUM		0x000000c4
+
+#define REG_HDMI_PHY_QSERDES_COM_LOCK_CMP_EN			0x000000c8
+
+#define REG_HDMI_PHY_QSERDES_COM_LOCK_CMP_CFG			0x000000cc
+
+#define REG_HDMI_PHY_QSERDES_COM_DEC_START_MODE0		0x000000d0
+
+#define REG_HDMI_PHY_QSERDES_COM_DEC_START_MODE1		0x000000d4
+
+#define REG_HDMI_PHY_QSERDES_COM_DEC_START_MODE2		0x000000d8
+
+#define REG_HDMI_PHY_QSERDES_COM_VCOCAL_DEADMAN_CTRL		0x000000d8
+
+#define REG_HDMI_PHY_QSERDES_COM_DIV_FRAC_START1_MODE0		0x000000dc
+
+#define REG_HDMI_PHY_QSERDES_COM_DIV_FRAC_START2_MODE0		0x000000e0
+
+#define REG_HDMI_PHY_QSERDES_COM_DIV_FRAC_START3_MODE0		0x000000e4
+
+#define REG_HDMI_PHY_QSERDES_COM_DIV_FRAC_START1_MODE1		0x000000e8
+
+#define REG_HDMI_PHY_QSERDES_COM_DIV_FRAC_START2_MODE1		0x000000ec
+
+#define REG_HDMI_PHY_QSERDES_COM_DIV_FRAC_START3_MODE1		0x000000f0
+
+#define REG_HDMI_PHY_QSERDES_COM_DIV_FRAC_START1_MODE2		0x000000f4
+
+#define REG_HDMI_PHY_QSERDES_COM_VCO_TUNE_MINVAL1		0x000000f4
+
+#define REG_HDMI_PHY_QSERDES_COM_DIV_FRAC_START2_MODE2		0x000000f8
+
+#define REG_HDMI_PHY_QSERDES_COM_VCO_TUNE_MINVAL2		0x000000f8
+
+#define REG_HDMI_PHY_QSERDES_COM_DIV_FRAC_START3_MODE2		0x000000fc
+
+#define REG_HDMI_PHY_QSERDES_COM_CMN_RSVD4			0x000000fc
+
+#define REG_HDMI_PHY_QSERDES_COM_INTEGLOOP_INITVAL		0x00000100
+
+#define REG_HDMI_PHY_QSERDES_COM_INTEGLOOP_EN			0x00000104
+
+#define REG_HDMI_PHY_QSERDES_COM_INTEGLOOP_GAIN0_MODE0		0x00000108
+
+#define REG_HDMI_PHY_QSERDES_COM_INTEGLOOP_GAIN1_MODE0		0x0000010c
+
+#define REG_HDMI_PHY_QSERDES_COM_INTEGLOOP_GAIN0_MODE1		0x00000110
+
+#define REG_HDMI_PHY_QSERDES_COM_INTEGLOOP_GAIN1_MODE1		0x00000114
+
+#define REG_HDMI_PHY_QSERDES_COM_INTEGLOOP_GAIN0_MODE2		0x00000118
+
+#define REG_HDMI_PHY_QSERDES_COM_VCO_TUNE_MAXVAL1		0x00000118
+
+#define REG_HDMI_PHY_QSERDES_COM_INTEGLOOP_GAIN1_MODE2		0x0000011c
+
+#define REG_HDMI_PHY_QSERDES_COM_VCO_TUNE_MAXVAL2		0x0000011c
+
+#define REG_HDMI_PHY_QSERDES_COM_RES_TRIM_CONTROL2		0x00000120
+
+#define REG_HDMI_PHY_QSERDES_COM_VCO_TUNE_CTRL			0x00000124
+
+#define REG_HDMI_PHY_QSERDES_COM_VCO_TUNE_MAP			0x00000128
+
+#define REG_HDMI_PHY_QSERDES_COM_VCO_TUNE1_MODE0		0x0000012c
+
+#define REG_HDMI_PHY_QSERDES_COM_VCO_TUNE2_MODE0		0x00000130
+
+#define REG_HDMI_PHY_QSERDES_COM_VCO_TUNE1_MODE1		0x00000134
+
+#define REG_HDMI_PHY_QSERDES_COM_VCO_TUNE2_MODE1		0x00000138
+
+#define REG_HDMI_PHY_QSERDES_COM_VCO_TUNE1_MODE2		0x0000013c
+
+#define REG_HDMI_PHY_QSERDES_COM_VCO_TUNE_INITVAL1		0x0000013c
+
+#define REG_HDMI_PHY_QSERDES_COM_VCO_TUNE2_MODE2		0x00000140
+
+#define REG_HDMI_PHY_QSERDES_COM_VCO_TUNE_INITVAL2		0x00000140
+
+#define REG_HDMI_PHY_QSERDES_COM_VCO_TUNE_TIMER1		0x00000144
+
+#define REG_HDMI_PHY_QSERDES_COM_VCO_TUNE_TIMER2		0x00000148
+
+#define REG_HDMI_PHY_QSERDES_COM_SAR				0x0000014c
+
+#define REG_HDMI_PHY_QSERDES_COM_SAR_CLK			0x00000150
+
+#define REG_HDMI_PHY_QSERDES_COM_SAR_CODE_OUT_STATUS		0x00000154
+
+#define REG_HDMI_PHY_QSERDES_COM_SAR_CODE_READY_STATUS		0x00000158
+
+#define REG_HDMI_PHY_QSERDES_COM_CMN_STATUS			0x0000015c
+
+#define REG_HDMI_PHY_QSERDES_COM_RESET_SM_STATUS		0x00000160
+
+#define REG_HDMI_PHY_QSERDES_COM_RESTRIM_CODE_STATUS		0x00000164
+
+#define REG_HDMI_PHY_QSERDES_COM_PLLCAL_CODE1_STATUS		0x00000168
+
+#define REG_HDMI_PHY_QSERDES_COM_PLLCAL_CODE2_STATUS		0x0000016c
+
+#define REG_HDMI_PHY_QSERDES_COM_BG_CTRL			0x00000170
+
+#define REG_HDMI_PHY_QSERDES_COM_CLK_SELECT			0x00000174
+
+#define REG_HDMI_PHY_QSERDES_COM_HSCLK_SEL			0x00000178
+
+#define REG_HDMI_PHY_QSERDES_COM_INTEGLOOP_BINCODE_STATUS	0x0000017c
+
+#define REG_HDMI_PHY_QSERDES_COM_PLL_ANALOG			0x00000180
+
+#define REG_HDMI_PHY_QSERDES_COM_CORECLK_DIV			0x00000184
+
+#define REG_HDMI_PHY_QSERDES_COM_SW_RESET			0x00000188
+
+#define REG_HDMI_PHY_QSERDES_COM_CORE_CLK_EN			0x0000018c
+
+#define REG_HDMI_PHY_QSERDES_COM_C_READY_STATUS			0x00000190
+
+#define REG_HDMI_PHY_QSERDES_COM_CMN_CONFIG			0x00000194
+
+#define REG_HDMI_PHY_QSERDES_COM_CMN_RATE_OVERRIDE		0x00000198
+
+#define REG_HDMI_PHY_QSERDES_COM_SVS_MODE_CLK_SEL		0x0000019c
+
+#define REG_HDMI_PHY_QSERDES_COM_DEBUG_BUS0			0x000001a0
+
+#define REG_HDMI_PHY_QSERDES_COM_DEBUG_BUS1			0x000001a4
+
+#define REG_HDMI_PHY_QSERDES_COM_DEBUG_BUS2			0x000001a8
+
+#define REG_HDMI_PHY_QSERDES_COM_DEBUG_BUS3			0x000001ac
+
+#define REG_HDMI_PHY_QSERDES_COM_DEBUG_BUS_SEL			0x000001b0
+
+#define REG_HDMI_PHY_QSERDES_COM_CMN_MISC1			0x000001b4
+
+#define REG_HDMI_PHY_QSERDES_COM_CMN_MISC2			0x000001b8
+
+#define REG_HDMI_PHY_QSERDES_COM_CORECLK_DIV_MODE1		0x000001bc
+
+#define REG_HDMI_PHY_QSERDES_COM_CORECLK_DIV_MODE2		0x000001c0
+
+#define REG_HDMI_PHY_QSERDES_COM_CMN_RSVD5			0x000001c4
+
+#define REG_HDMI_PHY_QSERDES_TX_LX_BIST_MODE_LANENO		0x00000000
+
+#define REG_HDMI_PHY_QSERDES_TX_LX_BIST_INVERT			0x00000004
+
+#define REG_HDMI_PHY_QSERDES_TX_LX_CLKBUF_ENABLE		0x00000008
+
+#define REG_HDMI_PHY_QSERDES_TX_LX_CMN_CONTROL_ONE		0x0000000c
+
+#define REG_HDMI_PHY_QSERDES_TX_LX_CMN_CONTROL_TWO		0x00000010
+
+#define REG_HDMI_PHY_QSERDES_TX_LX_CMN_CONTROL_THREE		0x00000014
+
+#define REG_HDMI_PHY_QSERDES_TX_LX_TX_EMP_POST1_LVL		0x00000018
+
+#define REG_HDMI_PHY_QSERDES_TX_LX_TX_POST2_EMPH		0x0000001c
+
+#define REG_HDMI_PHY_QSERDES_TX_LX_TX_BOOST_LVL_UP_DN		0x00000020
+
+#define REG_HDMI_PHY_QSERDES_TX_LX_HP_PD_ENABLES		0x00000024
+
+#define REG_HDMI_PHY_QSERDES_TX_LX_TX_IDLE_LVL_LARGE_AMP	0x00000028
+
+#define REG_HDMI_PHY_QSERDES_TX_LX_TX_DRV_LVL			0x0000002c
+
+#define REG_HDMI_PHY_QSERDES_TX_LX_TX_DRV_LVL_OFFSET		0x00000030
+
+#define REG_HDMI_PHY_QSERDES_TX_LX_RESET_TSYNC_EN		0x00000034
+
+#define REG_HDMI_PHY_QSERDES_TX_LX_PRE_STALL_LDO_BOOST_EN	0x00000038
+
+#define REG_HDMI_PHY_QSERDES_TX_LX_TX_BAND			0x0000003c
+
+#define REG_HDMI_PHY_QSERDES_TX_LX_SLEW_CNTL			0x00000040
+
+#define REG_HDMI_PHY_QSERDES_TX_LX_INTERFACE_SELECT		0x00000044
+
+#define REG_HDMI_PHY_QSERDES_TX_LX_LPB_EN			0x00000048
+
+#define REG_HDMI_PHY_QSERDES_TX_LX_RES_CODE_LANE_TX		0x0000004c
+
+#define REG_HDMI_PHY_QSERDES_TX_LX_RES_CODE_LANE_RX		0x00000050
+
+#define REG_HDMI_PHY_QSERDES_TX_LX_RES_CODE_LANE_OFFSET		0x00000054
+
+#define REG_HDMI_PHY_QSERDES_TX_LX_PERL_LENGTH1			0x00000058
+
+#define REG_HDMI_PHY_QSERDES_TX_LX_PERL_LENGTH2			0x0000005c
+
+#define REG_HDMI_PHY_QSERDES_TX_LX_SERDES_BYP_EN_OUT		0x00000060
+
+#define REG_HDMI_PHY_QSERDES_TX_LX_DEBUG_BUS_SEL		0x00000064
+
+#define REG_HDMI_PHY_QSERDES_TX_LX_HIGHZ_TRANSCEIVEREN_BIAS_DRVR_EN	0x00000068
+
+#define REG_HDMI_PHY_QSERDES_TX_LX_TX_POL_INV			0x0000006c
+
+#define REG_HDMI_PHY_QSERDES_TX_LX_PARRATE_REC_DETECT_IDLE_EN	0x00000070
+
+#define REG_HDMI_PHY_QSERDES_TX_LX_BIST_PATTERN1		0x00000074
+
+#define REG_HDMI_PHY_QSERDES_TX_LX_BIST_PATTERN2		0x00000078
+
+#define REG_HDMI_PHY_QSERDES_TX_LX_BIST_PATTERN3		0x0000007c
+
+#define REG_HDMI_PHY_QSERDES_TX_LX_BIST_PATTERN4		0x00000080
+
+#define REG_HDMI_PHY_QSERDES_TX_LX_BIST_PATTERN5		0x00000084
+
+#define REG_HDMI_PHY_QSERDES_TX_LX_BIST_PATTERN6		0x00000088
+
+#define REG_HDMI_PHY_QSERDES_TX_LX_BIST_PATTERN7		0x0000008c
+
+#define REG_HDMI_PHY_QSERDES_TX_LX_BIST_PATTERN8		0x00000090
+
+#define REG_HDMI_PHY_QSERDES_TX_LX_LANE_MODE			0x00000094
+
+#define REG_HDMI_PHY_QSERDES_TX_LX_IDAC_CAL_LANE_MODE		0x00000098
+
+#define REG_HDMI_PHY_QSERDES_TX_LX_IDAC_CAL_LANE_MODE_CONFIGURATION	0x0000009c
+
+#define REG_HDMI_PHY_QSERDES_TX_LX_ATB_SEL1			0x000000a0
+
+#define REG_HDMI_PHY_QSERDES_TX_LX_ATB_SEL2			0x000000a4
+
+#define REG_HDMI_PHY_QSERDES_TX_LX_RCV_DETECT_LVL		0x000000a8
+
+#define REG_HDMI_PHY_QSERDES_TX_LX_RCV_DETECT_LVL_2		0x000000ac
+
+#define REG_HDMI_PHY_QSERDES_TX_LX_PRBS_SEED1			0x000000b0
+
+#define REG_HDMI_PHY_QSERDES_TX_LX_PRBS_SEED2			0x000000b4
+
+#define REG_HDMI_PHY_QSERDES_TX_LX_PRBS_SEED3			0x000000b8
+
+#define REG_HDMI_PHY_QSERDES_TX_LX_PRBS_SEED4			0x000000bc
+
+#define REG_HDMI_PHY_QSERDES_TX_LX_RESET_GEN			0x000000c0
+
+#define REG_HDMI_PHY_QSERDES_TX_LX_RESET_GEN_MUXES		0x000000c4
+
+#define REG_HDMI_PHY_QSERDES_TX_LX_TRAN_DRVR_EMP_EN		0x000000c8
+
+#define REG_HDMI_PHY_QSERDES_TX_LX_TX_INTERFACE_MODE		0x000000cc
+
+#define REG_HDMI_PHY_QSERDES_TX_LX_PWM_CTRL			0x000000d0
+
+#define REG_HDMI_PHY_QSERDES_TX_LX_PWM_ENCODED_OR_DATA		0x000000d4
+
+#define REG_HDMI_PHY_QSERDES_TX_LX_PWM_GEAR_1_DIVIDER_BAND2	0x000000d8
+
+#define REG_HDMI_PHY_QSERDES_TX_LX_PWM_GEAR_2_DIVIDER_BAND2	0x000000dc
+
+#define REG_HDMI_PHY_QSERDES_TX_LX_PWM_GEAR_3_DIVIDER_BAND2	0x000000e0
+
+#define REG_HDMI_PHY_QSERDES_TX_LX_PWM_GEAR_4_DIVIDER_BAND2	0x000000e4
+
+#define REG_HDMI_PHY_QSERDES_TX_LX_PWM_GEAR_1_DIVIDER_BAND0_1	0x000000e8
+
+#define REG_HDMI_PHY_QSERDES_TX_LX_PWM_GEAR_2_DIVIDER_BAND0_1	0x000000ec
+
+#define REG_HDMI_PHY_QSERDES_TX_LX_PWM_GEAR_3_DIVIDER_BAND0_1	0x000000f0
+
+#define REG_HDMI_PHY_QSERDES_TX_LX_PWM_GEAR_4_DIVIDER_BAND0_1	0x000000f4
+
+#define REG_HDMI_PHY_QSERDES_TX_LX_VMODE_CTRL1			0x000000f8
+
+#define REG_HDMI_PHY_QSERDES_TX_LX_VMODE_CTRL2			0x000000fc
+
+#define REG_HDMI_PHY_QSERDES_TX_LX_TX_ALOG_INTF_OBSV_CNTL	0x00000100
+
+#define REG_HDMI_PHY_QSERDES_TX_LX_BIST_STATUS			0x00000104
+
+#define REG_HDMI_PHY_QSERDES_TX_LX_BIST_ERROR_COUNT1		0x00000108
+
+#define REG_HDMI_PHY_QSERDES_TX_LX_BIST_ERROR_COUNT2		0x0000010c
+
+#define REG_HDMI_PHY_QSERDES_TX_LX_TX_ALOG_INTF_OBSV		0x00000110
+
 
 #endif /* HDMI_XML */
diff --git a/drivers/gpu/drm/msm/hdmi/hdmi_audio.c b/drivers/gpu/drm/msm/hdmi/hdmi_audio.c
index df232e20c13e..a54d3bb5baad 100644
--- a/drivers/gpu/drm/msm/hdmi/hdmi_audio.c
+++ b/drivers/gpu/drm/msm/hdmi/hdmi_audio.c
@@ -89,7 +89,7 @@ static const struct hdmi_msm_audio_arcs *get_arcs(unsigned long int pixclock)
 	return NULL;
 }
 
-int hdmi_audio_update(struct hdmi *hdmi)
+int msm_hdmi_audio_update(struct hdmi *hdmi)
 {
 	struct hdmi_audio *audio = &hdmi->audio;
 	struct hdmi_audio_infoframe *info = &audio->infoframe;
@@ -232,7 +232,7 @@ int hdmi_audio_update(struct hdmi *hdmi)
 	return 0;
 }
 
-int hdmi_audio_info_setup(struct hdmi *hdmi, bool enabled,
+int msm_hdmi_audio_info_setup(struct hdmi *hdmi, bool enabled,
 	uint32_t num_of_channels, uint32_t channel_allocation,
 	uint32_t level_shift, bool down_mix)
 {
@@ -252,10 +252,10 @@ int hdmi_audio_info_setup(struct hdmi *hdmi, bool enabled,
 	audio->infoframe.level_shift_value = level_shift;
 	audio->infoframe.downmix_inhibit = down_mix;
 
-	return hdmi_audio_update(hdmi);
+	return msm_hdmi_audio_update(hdmi);
 }
 
-void hdmi_audio_set_sample_rate(struct hdmi *hdmi, int rate)
+void msm_hdmi_audio_set_sample_rate(struct hdmi *hdmi, int rate)
 {
 	struct hdmi_audio *audio;
 
@@ -268,5 +268,5 @@ void hdmi_audio_set_sample_rate(struct hdmi *hdmi, int rate)
 		return;
 
 	audio->rate = rate;
-	hdmi_audio_update(hdmi);
+	msm_hdmi_audio_update(hdmi);
 }
diff --git a/drivers/gpu/drm/msm/hdmi/hdmi_bridge.c b/drivers/gpu/drm/msm/hdmi/hdmi_bridge.c
index 92b69ae8caf9..bacbd5d8df0e 100644
--- a/drivers/gpu/drm/msm/hdmi/hdmi_bridge.c
+++ b/drivers/gpu/drm/msm/hdmi/hdmi_bridge.c
@@ -23,11 +23,11 @@ struct hdmi_bridge {
 };
 #define to_hdmi_bridge(x) container_of(x, struct hdmi_bridge, base)
 
-void hdmi_bridge_destroy(struct drm_bridge *bridge)
+void msm_hdmi_bridge_destroy(struct drm_bridge *bridge)
 {
 }
 
-static void power_on(struct drm_bridge *bridge)
+static void msm_hdmi_power_on(struct drm_bridge *bridge)
 {
 	struct drm_device *dev = bridge->dev;
 	struct hdmi_bridge *hdmi_bridge = to_hdmi_bridge(bridge);
@@ -86,7 +86,7 @@ static void power_off(struct drm_bridge *bridge)
 	}
 }
 
-static void hdmi_bridge_pre_enable(struct drm_bridge *bridge)
+static void msm_hdmi_bridge_pre_enable(struct drm_bridge *bridge)
 {
 	struct hdmi_bridge *hdmi_bridge = to_hdmi_bridge(bridge);
 	struct hdmi *hdmi = hdmi_bridge->hdmi;
@@ -95,51 +95,51 @@ static void hdmi_bridge_pre_enable(struct drm_bridge *bridge)
 	DBG("power up");
 
 	if (!hdmi->power_on) {
-		power_on(bridge);
+		msm_hdmi_phy_resource_enable(phy);
+		msm_hdmi_power_on(bridge);
 		hdmi->power_on = true;
-		hdmi_audio_update(hdmi);
+		msm_hdmi_audio_update(hdmi);
 	}
 
-	if (phy)
-		phy->funcs->powerup(phy, hdmi->pixclock);
+	msm_hdmi_phy_powerup(phy, hdmi->pixclock);
 
-	hdmi_set_mode(hdmi, true);
+	msm_hdmi_set_mode(hdmi, true);
 
 	if (hdmi->hdcp_ctrl)
-		hdmi_hdcp_on(hdmi->hdcp_ctrl);
+		msm_hdmi_hdcp_on(hdmi->hdcp_ctrl);
 }
 
-static void hdmi_bridge_enable(struct drm_bridge *bridge)
+static void msm_hdmi_bridge_enable(struct drm_bridge *bridge)
 {
 }
 
-static void hdmi_bridge_disable(struct drm_bridge *bridge)
+static void msm_hdmi_bridge_disable(struct drm_bridge *bridge)
 {
 }
 
-static void hdmi_bridge_post_disable(struct drm_bridge *bridge)
+static void msm_hdmi_bridge_post_disable(struct drm_bridge *bridge)
 {
 	struct hdmi_bridge *hdmi_bridge = to_hdmi_bridge(bridge);
 	struct hdmi *hdmi = hdmi_bridge->hdmi;
 	struct hdmi_phy *phy = hdmi->phy;
 
 	if (hdmi->hdcp_ctrl)
-		hdmi_hdcp_off(hdmi->hdcp_ctrl);
+		msm_hdmi_hdcp_off(hdmi->hdcp_ctrl);
 
 	DBG("power down");
-	hdmi_set_mode(hdmi, false);
+	msm_hdmi_set_mode(hdmi, false);
 
-	if (phy)
-		phy->funcs->powerdown(phy);
+	msm_hdmi_phy_powerdown(phy);
 
 	if (hdmi->power_on) {
 		power_off(bridge);
 		hdmi->power_on = false;
-		hdmi_audio_update(hdmi);
+		msm_hdmi_audio_update(hdmi);
+		msm_hdmi_phy_resource_disable(phy);
 	}
 }
 
-static void hdmi_bridge_mode_set(struct drm_bridge *bridge,
+static void msm_hdmi_bridge_mode_set(struct drm_bridge *bridge,
 		 struct drm_display_mode *mode,
 		 struct drm_display_mode *adjusted_mode)
 {
@@ -196,20 +196,20 @@ static void hdmi_bridge_mode_set(struct drm_bridge *bridge,
 	DBG("frame_ctrl=%08x", frame_ctrl);
 	hdmi_write(hdmi, REG_HDMI_FRAME_CTRL, frame_ctrl);
 
-	hdmi_audio_update(hdmi);
+	msm_hdmi_audio_update(hdmi);
 }
 
-static const struct drm_bridge_funcs hdmi_bridge_funcs = {
-		.pre_enable = hdmi_bridge_pre_enable,
-		.enable = hdmi_bridge_enable,
-		.disable = hdmi_bridge_disable,
-		.post_disable = hdmi_bridge_post_disable,
-		.mode_set = hdmi_bridge_mode_set,
+static const struct drm_bridge_funcs msm_hdmi_bridge_funcs = {
+		.pre_enable = msm_hdmi_bridge_pre_enable,
+		.enable = msm_hdmi_bridge_enable,
+		.disable = msm_hdmi_bridge_disable,
+		.post_disable = msm_hdmi_bridge_post_disable,
+		.mode_set = msm_hdmi_bridge_mode_set,
 };
 
 
 /* initialize bridge */
-struct drm_bridge *hdmi_bridge_init(struct hdmi *hdmi)
+struct drm_bridge *msm_hdmi_bridge_init(struct hdmi *hdmi)
 {
 	struct drm_bridge *bridge = NULL;
 	struct hdmi_bridge *hdmi_bridge;
@@ -225,7 +225,7 @@ struct drm_bridge *hdmi_bridge_init(struct hdmi *hdmi)
 	hdmi_bridge->hdmi = hdmi;
 
 	bridge = &hdmi_bridge->base;
-	bridge->funcs = &hdmi_bridge_funcs;
+	bridge->funcs = &msm_hdmi_bridge_funcs;
 
 	ret = drm_bridge_attach(hdmi->dev, bridge);
 	if (ret)
@@ -235,7 +235,7 @@ struct drm_bridge *hdmi_bridge_init(struct hdmi *hdmi)
 
 fail:
 	if (bridge)
-		hdmi_bridge_destroy(bridge);
+		msm_hdmi_bridge_destroy(bridge);
 
 	return ERR_PTR(ret);
 }
diff --git a/drivers/gpu/drm/msm/hdmi/hdmi_connector.c b/drivers/gpu/drm/msm/hdmi/hdmi_connector.c
index a3b05ae52dae..26129bff2dd6 100644
--- a/drivers/gpu/drm/msm/hdmi/hdmi_connector.c
+++ b/drivers/gpu/drm/msm/hdmi/hdmi_connector.c
@@ -28,7 +28,7 @@ struct hdmi_connector {
 };
 #define to_hdmi_connector(x) container_of(x, struct hdmi_connector, base)
 
-static void hdmi_phy_reset(struct hdmi *hdmi)
+static void msm_hdmi_phy_reset(struct hdmi *hdmi)
 {
 	unsigned int val;
 
@@ -81,114 +81,54 @@ static int gpio_config(struct hdmi *hdmi, bool on)
 {
 	struct device *dev = &hdmi->pdev->dev;
 	const struct hdmi_platform_config *config = hdmi->config;
-	int ret;
+	int ret, i;
 
 	if (on) {
-		if (config->ddc_clk_gpio != -1) {
-			ret = gpio_request(config->ddc_clk_gpio, "HDMI_DDC_CLK");
-			if (ret) {
-				dev_err(dev, "'%s'(%d) gpio_request failed: %d\n",
-					"HDMI_DDC_CLK", config->ddc_clk_gpio, ret);
-				goto error1;
-			}
-			gpio_set_value_cansleep(config->ddc_clk_gpio, 1);
-		}
-
-		if (config->ddc_data_gpio != -1) {
-			ret = gpio_request(config->ddc_data_gpio, "HDMI_DDC_DATA");
-			if (ret) {
-				dev_err(dev, "'%s'(%d) gpio_request failed: %d\n",
-					"HDMI_DDC_DATA", config->ddc_data_gpio, ret);
-				goto error2;
-			}
-			gpio_set_value_cansleep(config->ddc_data_gpio, 1);
-		}
-
-		ret = gpio_request(config->hpd_gpio, "HDMI_HPD");
-		if (ret) {
-			dev_err(dev, "'%s'(%d) gpio_request failed: %d\n",
-				"HDMI_HPD", config->hpd_gpio, ret);
-			goto error3;
-		}
-		gpio_direction_input(config->hpd_gpio);
-		gpio_set_value_cansleep(config->hpd_gpio, 1);
-
-		if (config->mux_en_gpio != -1) {
-			ret = gpio_request(config->mux_en_gpio, "HDMI_MUX_EN");
-			if (ret) {
-				dev_err(dev, "'%s'(%d) gpio_request failed: %d\n",
-					"HDMI_MUX_EN", config->mux_en_gpio, ret);
-				goto error4;
-			}
-			gpio_set_value_cansleep(config->mux_en_gpio, 1);
-		}
-
-		if (config->mux_sel_gpio != -1) {
-			ret = gpio_request(config->mux_sel_gpio, "HDMI_MUX_SEL");
-			if (ret) {
-				dev_err(dev, "'%s'(%d) gpio_request failed: %d\n",
-					"HDMI_MUX_SEL", config->mux_sel_gpio, ret);
-				goto error5;
+		for (i = 0; i < HDMI_MAX_NUM_GPIO; i++) {
+			struct hdmi_gpio_data gpio = config->gpios[i];
+
+			if (gpio.num != -1) {
+				ret = gpio_request(gpio.num, gpio.label);
+				if (ret) {
+					dev_err(dev,
+						"'%s'(%d) gpio_request failed: %d\n",
+						gpio.label, gpio.num, ret);
+					goto err;
+				}
+
+				if (gpio.output) {
+					gpio_direction_output(gpio.num,
+							      gpio.value);
+				} else {
+					gpio_direction_input(gpio.num);
+					gpio_set_value_cansleep(gpio.num,
+								gpio.value);
+				}
 			}
-			gpio_set_value_cansleep(config->mux_sel_gpio, 0);
 		}
 
-		if (config->mux_lpm_gpio != -1) {
-			ret = gpio_request(config->mux_lpm_gpio,
-					"HDMI_MUX_LPM");
-			if (ret) {
-				dev_err(dev,
-					"'%s'(%d) gpio_request failed: %d\n",
-					"HDMI_MUX_LPM",
-					config->mux_lpm_gpio, ret);
-				goto error6;
-			}
-			gpio_set_value_cansleep(config->mux_lpm_gpio, 1);
-		}
 		DBG("gpio on");
 	} else {
-		if (config->ddc_clk_gpio != -1)
-			gpio_free(config->ddc_clk_gpio);
-
-		if (config->ddc_data_gpio != -1)
-			gpio_free(config->ddc_data_gpio);
+		for (i = 0; i < HDMI_MAX_NUM_GPIO; i++) {
+			struct hdmi_gpio_data gpio = config->gpios[i];
 
-		gpio_free(config->hpd_gpio);
+			if (gpio.output) {
+				int value = gpio.value ? 0 : 1;
 
-		if (config->mux_en_gpio != -1) {
-			gpio_set_value_cansleep(config->mux_en_gpio, 0);
-			gpio_free(config->mux_en_gpio);
-		}
+				gpio_set_value_cansleep(gpio.num, value);
+			}
 
-		if (config->mux_sel_gpio != -1) {
-			gpio_set_value_cansleep(config->mux_sel_gpio, 1);
-			gpio_free(config->mux_sel_gpio);
-		}
+			gpio_free(gpio.num);
+		};
 
-		if (config->mux_lpm_gpio != -1) {
-			gpio_set_value_cansleep(config->mux_lpm_gpio, 0);
-			gpio_free(config->mux_lpm_gpio);
-		}
 		DBG("gpio off");
 	}
 
 	return 0;
+err:
+	while (i--)
+		gpio_free(config->gpios[i].num);
 
-error6:
-	if (config->mux_sel_gpio != -1)
-		gpio_free(config->mux_sel_gpio);
-error5:
-	if (config->mux_en_gpio != -1)
-		gpio_free(config->mux_en_gpio);
-error4:
-	gpio_free(config->hpd_gpio);
-error3:
-	if (config->ddc_data_gpio != -1)
-		gpio_free(config->ddc_data_gpio);
-error2:
-	if (config->ddc_clk_gpio != -1)
-		gpio_free(config->ddc_clk_gpio);
-error1:
 	return ret;
 }
 
@@ -239,9 +179,9 @@ static int hpd_enable(struct hdmi_connector *hdmi_connector)
 		}
 	}
 
-	hdmi_set_mode(hdmi, false);
-	hdmi_phy_reset(hdmi);
-	hdmi_set_mode(hdmi, true);
+	msm_hdmi_set_mode(hdmi, false);
+	msm_hdmi_phy_reset(hdmi);
+	msm_hdmi_set_mode(hdmi, true);
 
 	hdmi_write(hdmi, REG_HDMI_USEC_REFTIMER, 0x0001001b);
 
@@ -278,7 +218,7 @@ static void hdp_disable(struct hdmi_connector *hdmi_connector)
 	/* Disable HPD interrupt */
 	hdmi_write(hdmi, REG_HDMI_HPD_INT_CTRL, 0);
 
-	hdmi_set_mode(hdmi, false);
+	msm_hdmi_set_mode(hdmi, false);
 
 	for (i = 0; i < config->hpd_clk_cnt; i++)
 		clk_disable_unprepare(hdmi->hpd_clks[i]);
@@ -300,7 +240,7 @@ static void hdp_disable(struct hdmi_connector *hdmi_connector)
 }
 
 static void
-hotplug_work(struct work_struct *work)
+msm_hdmi_hotplug_work(struct work_struct *work)
 {
 	struct hdmi_connector *hdmi_connector =
 		container_of(work, struct hdmi_connector, hpd_work);
@@ -308,7 +248,7 @@ hotplug_work(struct work_struct *work)
 	drm_helper_hpd_irq_event(connector->dev);
 }
 
-void hdmi_connector_irq(struct drm_connector *connector)
+void msm_hdmi_connector_irq(struct drm_connector *connector)
 {
 	struct hdmi_connector *hdmi_connector = to_hdmi_connector(connector);
 	struct hdmi *hdmi = hdmi_connector->hdmi;
@@ -345,10 +285,13 @@ static enum drm_connector_status detect_reg(struct hdmi *hdmi)
 			connector_status_connected : connector_status_disconnected;
 }
 
+#define HPD_GPIO_INDEX	2
 static enum drm_connector_status detect_gpio(struct hdmi *hdmi)
 {
 	const struct hdmi_platform_config *config = hdmi->config;
-	return gpio_get_value(config->hpd_gpio) ?
+	struct hdmi_gpio_data hpd_gpio = config->gpios[HPD_GPIO_INDEX];
+
+	return gpio_get_value(hpd_gpio.num) ?
 			connector_status_connected :
 			connector_status_disconnected;
 }
@@ -358,9 +301,18 @@ static enum drm_connector_status hdmi_connector_detect(
 {
 	struct hdmi_connector *hdmi_connector = to_hdmi_connector(connector);
 	struct hdmi *hdmi = hdmi_connector->hdmi;
+	const struct hdmi_platform_config *config = hdmi->config;
+	struct hdmi_gpio_data hpd_gpio = config->gpios[HPD_GPIO_INDEX];
 	enum drm_connector_status stat_gpio, stat_reg;
 	int retry = 20;
 
+	/*
+	 * some platforms may not have hpd gpio. Rely only on the status
+	 * provided by REG_HDMI_HPD_INT_STATUS in this case.
+	 */
+	if (hpd_gpio.num == -1)
+		return detect_reg(hdmi);
+
 	do {
 		stat_gpio = detect_gpio(hdmi);
 		stat_reg  = detect_reg(hdmi);
@@ -395,7 +347,7 @@ static void hdmi_connector_destroy(struct drm_connector *connector)
 	kfree(hdmi_connector);
 }
 
-static int hdmi_connector_get_modes(struct drm_connector *connector)
+static int msm_hdmi_connector_get_modes(struct drm_connector *connector)
 {
 	struct hdmi_connector *hdmi_connector = to_hdmi_connector(connector);
 	struct hdmi *hdmi = hdmi_connector->hdmi;
@@ -421,7 +373,7 @@ static int hdmi_connector_get_modes(struct drm_connector *connector)
 	return ret;
 }
 
-static int hdmi_connector_mode_valid(struct drm_connector *connector,
+static int msm_hdmi_connector_mode_valid(struct drm_connector *connector,
 				 struct drm_display_mode *mode)
 {
 	struct hdmi_connector *hdmi_connector = to_hdmi_connector(connector);
@@ -451,7 +403,7 @@ static int hdmi_connector_mode_valid(struct drm_connector *connector,
 }
 
 static struct drm_encoder *
-hdmi_connector_best_encoder(struct drm_connector *connector)
+msm_hdmi_connector_best_encoder(struct drm_connector *connector)
 {
 	struct hdmi_connector *hdmi_connector = to_hdmi_connector(connector);
 	return hdmi_connector->hdmi->encoder;
@@ -467,14 +419,14 @@ static const struct drm_connector_funcs hdmi_connector_funcs = {
 	.atomic_destroy_state = drm_atomic_helper_connector_destroy_state,
 };
 
-static const struct drm_connector_helper_funcs hdmi_connector_helper_funcs = {
-	.get_modes = hdmi_connector_get_modes,
-	.mode_valid = hdmi_connector_mode_valid,
-	.best_encoder = hdmi_connector_best_encoder,
+static const struct drm_connector_helper_funcs msm_hdmi_connector_helper_funcs = {
+	.get_modes = msm_hdmi_connector_get_modes,
+	.mode_valid = msm_hdmi_connector_mode_valid,
+	.best_encoder = msm_hdmi_connector_best_encoder,
 };
 
 /* initialize connector */
-struct drm_connector *hdmi_connector_init(struct hdmi *hdmi)
+struct drm_connector *msm_hdmi_connector_init(struct hdmi *hdmi)
 {
 	struct drm_connector *connector = NULL;
 	struct hdmi_connector *hdmi_connector;
@@ -487,13 +439,13 @@ struct drm_connector *hdmi_connector_init(struct hdmi *hdmi)
 	}
 
 	hdmi_connector->hdmi = hdmi;
-	INIT_WORK(&hdmi_connector->hpd_work, hotplug_work);
+	INIT_WORK(&hdmi_connector->hpd_work, msm_hdmi_hotplug_work);
 
 	connector = &hdmi_connector->base;
 
 	drm_connector_init(hdmi->dev, connector, &hdmi_connector_funcs,
 			DRM_MODE_CONNECTOR_HDMIA);
-	drm_connector_helper_add(connector, &hdmi_connector_helper_funcs);
+	drm_connector_helper_add(connector, &msm_hdmi_connector_helper_funcs);
 
 	connector->polled = DRM_CONNECTOR_POLL_CONNECT |
 			DRM_CONNECTOR_POLL_DISCONNECT;
diff --git a/drivers/gpu/drm/msm/hdmi/hdmi_hdcp.c b/drivers/gpu/drm/msm/hdmi/hdmi_hdcp.c
index 1dc9c34eb0df..0baaaaabd002 100644
--- a/drivers/gpu/drm/msm/hdmi/hdmi_hdcp.c
+++ b/drivers/gpu/drm/msm/hdmi/hdmi_hdcp.c
@@ -84,7 +84,7 @@ struct hdmi_hdcp_ctrl {
 	bool max_dev_exceeded;
 };
 
-static int hdmi_ddc_read(struct hdmi *hdmi, u16 addr, u8 offset,
+static int msm_hdmi_ddc_read(struct hdmi *hdmi, u16 addr, u8 offset,
 	u8 *data, u16 data_len)
 {
 	int rc;
@@ -122,7 +122,7 @@ retry:
 
 #define HDCP_DDC_WRITE_MAX_BYTE_NUM 32
 
-static int hdmi_ddc_write(struct hdmi *hdmi, u16 addr, u8 offset,
+static int msm_hdmi_ddc_write(struct hdmi *hdmi, u16 addr, u8 offset,
 	u8 *data, u16 data_len)
 {
 	int rc;
@@ -162,7 +162,7 @@ retry:
 	return rc;
 }
 
-static int hdmi_hdcp_scm_wr(struct hdmi_hdcp_ctrl *hdcp_ctrl, u32 *preg,
+static int msm_hdmi_hdcp_scm_wr(struct hdmi_hdcp_ctrl *hdcp_ctrl, u32 *preg,
 	u32 *pdata, u32 count)
 {
 	struct hdmi *hdmi = hdcp_ctrl->hdmi;
@@ -202,7 +202,7 @@ static int hdmi_hdcp_scm_wr(struct hdmi_hdcp_ctrl *hdcp_ctrl, u32 *preg,
 	return ret;
 }
 
-void hdmi_hdcp_irq(struct hdmi_hdcp_ctrl *hdcp_ctrl)
+void msm_hdmi_hdcp_irq(struct hdmi_hdcp_ctrl *hdcp_ctrl)
 {
 	struct hdmi *hdmi = hdcp_ctrl->hdmi;
 	u32 reg_val, hdcp_int_status;
@@ -247,7 +247,7 @@ void hdmi_hdcp_irq(struct hdmi_hdcp_ctrl *hdcp_ctrl)
 	}
 }
 
-static int hdmi_hdcp_msleep(struct hdmi_hdcp_ctrl *hdcp_ctrl, u32 ms, u32 ev)
+static int msm_hdmi_hdcp_msleep(struct hdmi_hdcp_ctrl *hdcp_ctrl, u32 ms, u32 ev)
 {
 	int rc;
 
@@ -264,7 +264,7 @@ static int hdmi_hdcp_msleep(struct hdmi_hdcp_ctrl *hdcp_ctrl, u32 ms, u32 ev)
 	return 0;
 }
 
-static int hdmi_hdcp_read_validate_aksv(struct hdmi_hdcp_ctrl *hdcp_ctrl)
+static int msm_hdmi_hdcp_read_validate_aksv(struct hdmi_hdcp_ctrl *hdcp_ctrl)
 {
 	struct hdmi *hdmi = hdcp_ctrl->hdmi;
 
@@ -287,7 +287,7 @@ static int hdmi_hdcp_read_validate_aksv(struct hdmi_hdcp_ctrl *hdcp_ctrl)
 	return 0;
 }
 
-static int reset_hdcp_ddc_failures(struct hdmi_hdcp_ctrl *hdcp_ctrl)
+static int msm_reset_hdcp_ddc_failures(struct hdmi_hdcp_ctrl *hdcp_ctrl)
 {
 	struct hdmi *hdmi = hdcp_ctrl->hdmi;
 	u32 reg_val, failure, nack0;
@@ -337,7 +337,7 @@ static int reset_hdcp_ddc_failures(struct hdmi_hdcp_ctrl *hdcp_ctrl)
 		reg_val |= HDMI_DDC_CTRL_SW_STATUS_RESET;
 		hdmi_write(hdmi, REG_HDMI_DDC_CTRL, reg_val);
 
-		rc = hdmi_hdcp_msleep(hdcp_ctrl, 20, AUTH_ABORT_EV);
+		rc = msm_hdmi_hdcp_msleep(hdcp_ctrl, 20, AUTH_ABORT_EV);
 
 		reg_val = hdmi_read(hdmi, REG_HDMI_DDC_CTRL);
 		reg_val &= ~HDMI_DDC_CTRL_SW_STATUS_RESET;
@@ -350,7 +350,7 @@ static int reset_hdcp_ddc_failures(struct hdmi_hdcp_ctrl *hdcp_ctrl)
 
 		/* If previous msleep is aborted, skip this msleep */
 		if (!rc)
-			rc = hdmi_hdcp_msleep(hdcp_ctrl, 20, AUTH_ABORT_EV);
+			rc = msm_hdmi_hdcp_msleep(hdcp_ctrl, 20, AUTH_ABORT_EV);
 
 		reg_val = hdmi_read(hdmi, REG_HDMI_DDC_CTRL);
 		reg_val &= ~HDMI_DDC_CTRL_SOFT_RESET;
@@ -362,7 +362,7 @@ static int reset_hdcp_ddc_failures(struct hdmi_hdcp_ctrl *hdcp_ctrl)
 	return rc;
 }
 
-static int hdmi_hdcp_hw_ddc_clean(struct hdmi_hdcp_ctrl *hdcp_ctrl)
+static int msm_hdmi_hdcp_hw_ddc_clean(struct hdmi_hdcp_ctrl *hdcp_ctrl)
 {
 	int rc;
 	u32 hdcp_ddc_status, ddc_hw_status;
@@ -394,7 +394,7 @@ static int hdmi_hdcp_hw_ddc_clean(struct hdmi_hdcp_ctrl *hdcp_ctrl)
 			return -ETIMEDOUT;
 		}
 
-		rc = hdmi_hdcp_msleep(hdcp_ctrl, 20, AUTH_ABORT_EV);
+		rc = msm_hdmi_hdcp_msleep(hdcp_ctrl, 20, AUTH_ABORT_EV);
 		if (rc)
 			return rc;
 	} while (1);
@@ -402,7 +402,7 @@ static int hdmi_hdcp_hw_ddc_clean(struct hdmi_hdcp_ctrl *hdcp_ctrl)
 	return 0;
 }
 
-static void hdmi_hdcp_reauth_work(struct work_struct *work)
+static void msm_hdmi_hdcp_reauth_work(struct work_struct *work)
 {
 	struct hdmi_hdcp_ctrl *hdcp_ctrl = container_of(work,
 		struct hdmi_hdcp_ctrl, hdcp_reauth_work);
@@ -430,7 +430,7 @@ static void hdmi_hdcp_reauth_work(struct work_struct *work)
 		HDMI_HDCP_RESET_LINK0_DEAUTHENTICATE);
 
 	/* Wait to be clean on DDC HW engine */
-	if (hdmi_hdcp_hw_ddc_clean(hdcp_ctrl)) {
+	if (msm_hdmi_hdcp_hw_ddc_clean(hdcp_ctrl)) {
 		pr_info("%s: reauth work aborted\n", __func__);
 		return;
 	}
@@ -461,7 +461,7 @@ static void hdmi_hdcp_reauth_work(struct work_struct *work)
 	queue_work(hdmi->workq, &hdcp_ctrl->hdcp_auth_work);
 }
 
-static int hdmi_hdcp_auth_prepare(struct hdmi_hdcp_ctrl *hdcp_ctrl)
+static int msm_hdmi_hdcp_auth_prepare(struct hdmi_hdcp_ctrl *hdcp_ctrl)
 {
 	struct hdmi *hdmi = hdcp_ctrl->hdmi;
 	u32 link0_status;
@@ -470,7 +470,7 @@ static int hdmi_hdcp_auth_prepare(struct hdmi_hdcp_ctrl *hdcp_ctrl)
 	int rc;
 
 	if (!hdcp_ctrl->aksv_valid) {
-		rc = hdmi_hdcp_read_validate_aksv(hdcp_ctrl);
+		rc = msm_hdmi_hdcp_read_validate_aksv(hdcp_ctrl);
 		if (rc) {
 			pr_err("%s: ASKV validation failed\n", __func__);
 			hdcp_ctrl->hdcp_state = HDCP_STATE_NO_AKSV;
@@ -538,12 +538,12 @@ static int hdmi_hdcp_auth_prepare(struct hdmi_hdcp_ctrl *hdcp_ctrl)
 		DBG("An not ready after enabling HDCP");
 
 	/* Clear any DDC failures from previous tries before enable HDCP*/
-	rc = reset_hdcp_ddc_failures(hdcp_ctrl);
+	rc = msm_reset_hdcp_ddc_failures(hdcp_ctrl);
 
 	return rc;
 }
 
-static void hdmi_hdcp_auth_fail(struct hdmi_hdcp_ctrl *hdcp_ctrl)
+static void msm_hdmi_hdcp_auth_fail(struct hdmi_hdcp_ctrl *hdcp_ctrl)
 {
 	struct hdmi *hdmi = hdcp_ctrl->hdmi;
 	u32 reg_val;
@@ -561,7 +561,7 @@ static void hdmi_hdcp_auth_fail(struct hdmi_hdcp_ctrl *hdcp_ctrl)
 	queue_work(hdmi->workq, &hdcp_ctrl->hdcp_reauth_work);
 }
 
-static void hdmi_hdcp_auth_done(struct hdmi_hdcp_ctrl *hdcp_ctrl)
+static void msm_hdmi_hdcp_auth_done(struct hdmi_hdcp_ctrl *hdcp_ctrl)
 {
 	struct hdmi *hdmi = hdcp_ctrl->hdmi;
 	u32 reg_val;
@@ -596,7 +596,7 @@ static void hdmi_hdcp_auth_done(struct hdmi_hdcp_ctrl *hdcp_ctrl)
  * Write An and AKSV to sink
  * Read BKSV from sink and write into HDCP engine
  */
-static int hdmi_hdcp_wait_key_an_ready(struct hdmi_hdcp_ctrl *hdcp_ctrl)
+static int msm_hdmi_hdcp_wait_key_an_ready(struct hdmi_hdcp_ctrl *hdcp_ctrl)
 {
 	int rc;
 	struct hdmi *hdmi = hdcp_ctrl->hdmi;
@@ -621,7 +621,7 @@ static int hdmi_hdcp_wait_key_an_ready(struct hdmi_hdcp_ctrl *hdcp_ctrl)
 			return -ETIMEDOUT;
 		}
 
-		rc = hdmi_hdcp_msleep(hdcp_ctrl, 20, AUTH_ABORT_EV);
+		rc = msm_hdmi_hdcp_msleep(hdcp_ctrl, 20, AUTH_ABORT_EV);
 		if (rc)
 			return rc;
 	} while (1);
@@ -643,7 +643,7 @@ static int hdmi_hdcp_wait_key_an_ready(struct hdmi_hdcp_ctrl *hdcp_ctrl)
 			return -ETIMEDOUT;
 		}
 
-		rc = hdmi_hdcp_msleep(hdcp_ctrl, 20, AUTH_ABORT_EV);
+		rc = msm_hdmi_hdcp_msleep(hdcp_ctrl, 20, AUTH_ABORT_EV);
 		if (rc)
 			return rc;
 	} while (1);
@@ -651,7 +651,7 @@ static int hdmi_hdcp_wait_key_an_ready(struct hdmi_hdcp_ctrl *hdcp_ctrl)
 	return 0;
 }
 
-static int hdmi_hdcp_send_aksv_an(struct hdmi_hdcp_ctrl *hdcp_ctrl)
+static int msm_hdmi_hdcp_send_aksv_an(struct hdmi_hdcp_ctrl *hdcp_ctrl)
 {
 	int rc = 0;
 	struct hdmi *hdmi = hdcp_ctrl->hdmi;
@@ -676,7 +676,7 @@ static int hdmi_hdcp_send_aksv_an(struct hdmi_hdcp_ctrl *hdcp_ctrl)
 	aksv[4] =  link0_aksv_1        & 0xFF;
 
 	/* Write An to offset 0x18 */
-	rc = hdmi_ddc_write(hdmi, HDCP_PORT_ADDR, 0x18, (u8 *)link0_an,
+	rc = msm_hdmi_ddc_write(hdmi, HDCP_PORT_ADDR, 0x18, (u8 *)link0_an,
 		(u16)sizeof(link0_an));
 	if (rc) {
 		pr_err("%s:An write failed\n", __func__);
@@ -685,7 +685,7 @@ static int hdmi_hdcp_send_aksv_an(struct hdmi_hdcp_ctrl *hdcp_ctrl)
 	DBG("Link0-An=%08x%08x", link0_an[0], link0_an[1]);
 
 	/* Write AKSV to offset 0x10 */
-	rc = hdmi_ddc_write(hdmi, HDCP_PORT_ADDR, 0x10, aksv, 5);
+	rc = msm_hdmi_ddc_write(hdmi, HDCP_PORT_ADDR, 0x10, aksv, 5);
 	if (rc) {
 		pr_err("%s:AKSV write failed\n", __func__);
 		return rc;
@@ -695,7 +695,7 @@ static int hdmi_hdcp_send_aksv_an(struct hdmi_hdcp_ctrl *hdcp_ctrl)
 	return 0;
 }
 
-static int hdmi_hdcp_recv_bksv(struct hdmi_hdcp_ctrl *hdcp_ctrl)
+static int msm_hdmi_hdcp_recv_bksv(struct hdmi_hdcp_ctrl *hdcp_ctrl)
 {
 	int rc = 0;
 	struct hdmi *hdmi = hdcp_ctrl->hdmi;
@@ -703,7 +703,7 @@ static int hdmi_hdcp_recv_bksv(struct hdmi_hdcp_ctrl *hdcp_ctrl)
 	u32 reg[2], data[2];
 
 	/* Read BKSV at offset 0x00 */
-	rc = hdmi_ddc_read(hdmi, HDCP_PORT_ADDR, 0x00, bksv, 5);
+	rc = msm_hdmi_ddc_read(hdmi, HDCP_PORT_ADDR, 0x00, bksv, 5);
 	if (rc) {
 		pr_err("%s:BKSV read failed\n", __func__);
 		return rc;
@@ -728,19 +728,19 @@ static int hdmi_hdcp_recv_bksv(struct hdmi_hdcp_ctrl *hdcp_ctrl)
 	data[0] = hdcp_ctrl->bksv_lsb;
 	reg[1] = REG_HDMI_HDCP_RCVPORT_DATA1;
 	data[1] = hdcp_ctrl->bksv_msb;
-	rc = hdmi_hdcp_scm_wr(hdcp_ctrl, reg, data, 2);
+	rc = msm_hdmi_hdcp_scm_wr(hdcp_ctrl, reg, data, 2);
 
 	return rc;
 }
 
-static int hdmi_hdcp_recv_bcaps(struct hdmi_hdcp_ctrl *hdcp_ctrl)
+static int msm_hdmi_hdcp_recv_bcaps(struct hdmi_hdcp_ctrl *hdcp_ctrl)
 {
 	int rc = 0;
 	struct hdmi *hdmi = hdcp_ctrl->hdmi;
 	u32 reg, data;
 	u8 bcaps;
 
-	rc = hdmi_ddc_read(hdmi, HDCP_PORT_ADDR, 0x40, &bcaps, 1);
+	rc = msm_hdmi_ddc_read(hdmi, HDCP_PORT_ADDR, 0x40, &bcaps, 1);
 	if (rc) {
 		pr_err("%s:BCAPS read failed\n", __func__);
 		return rc;
@@ -753,26 +753,26 @@ static int hdmi_hdcp_recv_bcaps(struct hdmi_hdcp_ctrl *hdcp_ctrl)
 	/* Write BCAPS to the hardware */
 	reg = REG_HDMI_HDCP_RCVPORT_DATA12;
 	data = (u32)bcaps;
-	rc = hdmi_hdcp_scm_wr(hdcp_ctrl, &reg, &data, 1);
+	rc = msm_hdmi_hdcp_scm_wr(hdcp_ctrl, &reg, &data, 1);
 
 	return rc;
 }
 
-static int hdmi_hdcp_auth_part1_key_exchange(struct hdmi_hdcp_ctrl *hdcp_ctrl)
+static int msm_hdmi_hdcp_auth_part1_key_exchange(struct hdmi_hdcp_ctrl *hdcp_ctrl)
 {
 	struct hdmi *hdmi = hdcp_ctrl->hdmi;
 	unsigned long flags;
 	int rc;
 
 	/* Wait for AKSV key and An ready */
-	rc = hdmi_hdcp_wait_key_an_ready(hdcp_ctrl);
+	rc = msm_hdmi_hdcp_wait_key_an_ready(hdcp_ctrl);
 	if (rc) {
 		pr_err("%s: wait key and an ready failed\n", __func__);
 		return rc;
 	};
 
 	/* Read BCAPS and send to HDCP engine */
-	rc = hdmi_hdcp_recv_bcaps(hdcp_ctrl);
+	rc = msm_hdmi_hdcp_recv_bcaps(hdcp_ctrl);
 	if (rc) {
 		pr_err("%s: read bcaps error, abort\n", __func__);
 		return rc;
@@ -785,14 +785,14 @@ static int hdmi_hdcp_auth_part1_key_exchange(struct hdmi_hdcp_ctrl *hdcp_ctrl)
 	hdmi_write(hdmi, REG_HDMI_HDCP_RCVPORT_DATA4, 0);
 
 	/* Send AKSV and An to sink */
-	rc = hdmi_hdcp_send_aksv_an(hdcp_ctrl);
+	rc = msm_hdmi_hdcp_send_aksv_an(hdcp_ctrl);
 	if (rc) {
 		pr_err("%s:An/Aksv write failed\n", __func__);
 		return rc;
 	}
 
 	/* Read BKSV and send to HDCP engine*/
-	rc = hdmi_hdcp_recv_bksv(hdcp_ctrl);
+	rc = msm_hdmi_hdcp_recv_bksv(hdcp_ctrl);
 	if (rc) {
 		pr_err("%s:BKSV Process failed\n", __func__);
 		return rc;
@@ -812,7 +812,7 @@ static int hdmi_hdcp_auth_part1_key_exchange(struct hdmi_hdcp_ctrl *hdcp_ctrl)
 }
 
 /* read R0' from sink and pass it to HDCP engine */
-static int hdmi_hdcp_auth_part1_recv_r0(struct hdmi_hdcp_ctrl *hdcp_ctrl)
+static int msm_hdmi_hdcp_auth_part1_recv_r0(struct hdmi_hdcp_ctrl *hdcp_ctrl)
 {
 	struct hdmi *hdmi = hdcp_ctrl->hdmi;
 	int rc = 0;
@@ -822,12 +822,12 @@ static int hdmi_hdcp_auth_part1_recv_r0(struct hdmi_hdcp_ctrl *hdcp_ctrl)
 	 * HDCP Compliance Test case 1A-01:
 	 * Wait here at least 100ms before reading R0'
 	 */
-	rc = hdmi_hdcp_msleep(hdcp_ctrl, 125, AUTH_ABORT_EV);
+	rc = msm_hdmi_hdcp_msleep(hdcp_ctrl, 125, AUTH_ABORT_EV);
 	if (rc)
 		return rc;
 
 	/* Read R0' at offset 0x08 */
-	rc = hdmi_ddc_read(hdmi, HDCP_PORT_ADDR, 0x08, buf, 2);
+	rc = msm_hdmi_ddc_read(hdmi, HDCP_PORT_ADDR, 0x08, buf, 2);
 	if (rc) {
 		pr_err("%s:R0' read failed\n", __func__);
 		return rc;
@@ -842,14 +842,14 @@ static int hdmi_hdcp_auth_part1_recv_r0(struct hdmi_hdcp_ctrl *hdcp_ctrl)
 }
 
 /* Wait for authenticating result: R0/R0' are matched or not */
-static int hdmi_hdcp_auth_part1_verify_r0(struct hdmi_hdcp_ctrl *hdcp_ctrl)
+static int msm_hdmi_hdcp_auth_part1_verify_r0(struct hdmi_hdcp_ctrl *hdcp_ctrl)
 {
 	struct hdmi *hdmi = hdcp_ctrl->hdmi;
 	u32 link0_status;
 	int rc;
 
 	/* wait for hdcp irq, 10 sec should be long enough */
-	rc = hdmi_hdcp_msleep(hdcp_ctrl, 10000, AUTH_RESULT_RDY_EV);
+	rc = msm_hdmi_hdcp_msleep(hdcp_ctrl, 10000, AUTH_RESULT_RDY_EV);
 	if (!rc) {
 		pr_err("%s: Wait Auth IRQ timeout\n", __func__);
 		return -ETIMEDOUT;
@@ -869,7 +869,7 @@ static int hdmi_hdcp_auth_part1_verify_r0(struct hdmi_hdcp_ctrl *hdcp_ctrl)
 	return 0;
 }
 
-static int hdmi_hdcp_recv_check_bstatus(struct hdmi_hdcp_ctrl *hdcp_ctrl,
+static int msm_hdmi_hdcp_recv_check_bstatus(struct hdmi_hdcp_ctrl *hdcp_ctrl,
 	u16 *pbstatus)
 {
 	int rc;
@@ -880,7 +880,7 @@ static int hdmi_hdcp_recv_check_bstatus(struct hdmi_hdcp_ctrl *hdcp_ctrl,
 	u8 buf[2];
 
 	/* Read BSTATUS at offset 0x41 */
-	rc = hdmi_ddc_read(hdmi, HDCP_PORT_ADDR, 0x41, buf, 2);
+	rc = msm_hdmi_ddc_read(hdmi, HDCP_PORT_ADDR, 0x41, buf, 2);
 	if (rc) {
 		pr_err("%s: BSTATUS read failed\n", __func__);
 		goto error;
@@ -936,7 +936,7 @@ error:
 	return rc;
 }
 
-static int hdmi_hdcp_auth_part2_wait_ksv_fifo_ready(
+static int msm_hdmi_hdcp_auth_part2_wait_ksv_fifo_ready(
 	struct hdmi_hdcp_ctrl *hdcp_ctrl)
 {
 	int rc;
@@ -953,7 +953,7 @@ static int hdmi_hdcp_auth_part2_wait_ksv_fifo_ready(
 	timeout_count = 100;
 	do {
 		/* Read BCAPS at offset 0x40 */
-		rc = hdmi_ddc_read(hdmi, HDCP_PORT_ADDR, 0x40, &bcaps, 1);
+		rc = msm_hdmi_ddc_read(hdmi, HDCP_PORT_ADDR, 0x40, &bcaps, 1);
 		if (rc) {
 			pr_err("%s: BCAPS read failed\n", __func__);
 			return rc;
@@ -968,12 +968,12 @@ static int hdmi_hdcp_auth_part2_wait_ksv_fifo_ready(
 			return -ETIMEDOUT;
 		}
 
-		rc = hdmi_hdcp_msleep(hdcp_ctrl, 20, AUTH_ABORT_EV);
+		rc = msm_hdmi_hdcp_msleep(hdcp_ctrl, 20, AUTH_ABORT_EV);
 		if (rc)
 			return rc;
 	} while (1);
 
-	rc = hdmi_hdcp_recv_check_bstatus(hdcp_ctrl, &bstatus);
+	rc = msm_hdmi_hdcp_recv_check_bstatus(hdcp_ctrl, &bstatus);
 	if (rc) {
 		pr_err("%s: bstatus error\n", __func__);
 		return rc;
@@ -982,7 +982,7 @@ static int hdmi_hdcp_auth_part2_wait_ksv_fifo_ready(
 	/* Write BSTATUS and BCAPS to HDCP registers */
 	reg = REG_HDMI_HDCP_RCVPORT_DATA12;
 	data = bcaps | (bstatus << 8);
-	rc = hdmi_hdcp_scm_wr(hdcp_ctrl, &reg, &data, 1);
+	rc = msm_hdmi_hdcp_scm_wr(hdcp_ctrl, &reg, &data, 1);
 	if (rc) {
 		pr_err("%s: BSTATUS write failed\n", __func__);
 		return rc;
@@ -997,7 +997,7 @@ static int hdmi_hdcp_auth_part2_wait_ksv_fifo_ready(
  * transfer V' from sink to HDCP engine
  * reset SHA engine
  */
-static int hdmi_hdcp_transfer_v_h(struct hdmi_hdcp_ctrl *hdcp_ctrl)
+static int msm_hdmi_hdcp_transfer_v_h(struct hdmi_hdcp_ctrl *hdcp_ctrl)
 {
 	struct hdmi *hdmi = hdcp_ctrl->hdmi;
 	int rc = 0;
@@ -1016,7 +1016,7 @@ static int hdmi_hdcp_transfer_v_h(struct hdmi_hdcp_ctrl *hdcp_ctrl)
 
 	for (i = 0; i < size; i++) {
 		rd = &reg_data[i];
-		rc = hdmi_ddc_read(hdmi, HDCP_PORT_ADDR,
+		rc = msm_hdmi_ddc_read(hdmi, HDCP_PORT_ADDR,
 			rd->off, (u8 *)&data[i], (u16)sizeof(data[i]));
 		if (rc) {
 			pr_err("%s: Read %s failed\n", __func__, rd->name);
@@ -1027,13 +1027,13 @@ static int hdmi_hdcp_transfer_v_h(struct hdmi_hdcp_ctrl *hdcp_ctrl)
 		reg[i] = reg_data[i].reg_id;
 	}
 
-	rc = hdmi_hdcp_scm_wr(hdcp_ctrl, reg, data, size);
+	rc = msm_hdmi_hdcp_scm_wr(hdcp_ctrl, reg, data, size);
 
 error:
 	return rc;
 }
 
-static int hdmi_hdcp_recv_ksv_fifo(struct hdmi_hdcp_ctrl *hdcp_ctrl)
+static int msm_hdmi_hdcp_recv_ksv_fifo(struct hdmi_hdcp_ctrl *hdcp_ctrl)
 {
 	int rc;
 	struct hdmi *hdmi = hdcp_ctrl->hdmi;
@@ -1041,7 +1041,7 @@ static int hdmi_hdcp_recv_ksv_fifo(struct hdmi_hdcp_ctrl *hdcp_ctrl)
 
 	ksv_bytes = 5 * hdcp_ctrl->dev_count;
 
-	rc = hdmi_ddc_read(hdmi, HDCP_PORT_ADDR, 0x43,
+	rc = msm_hdmi_ddc_read(hdmi, HDCP_PORT_ADDR, 0x43,
 		hdcp_ctrl->ksv_list, ksv_bytes);
 	if (rc)
 		pr_err("%s: KSV FIFO read failed\n", __func__);
@@ -1049,7 +1049,7 @@ static int hdmi_hdcp_recv_ksv_fifo(struct hdmi_hdcp_ctrl *hdcp_ctrl)
 	return rc;
 }
 
-static int hdmi_hdcp_reset_sha_engine(struct hdmi_hdcp_ctrl *hdcp_ctrl)
+static int msm_hdmi_hdcp_reset_sha_engine(struct hdmi_hdcp_ctrl *hdcp_ctrl)
 {
 	u32 reg[2], data[2];
 	u32 rc  = 0;
@@ -1059,12 +1059,12 @@ static int hdmi_hdcp_reset_sha_engine(struct hdmi_hdcp_ctrl *hdcp_ctrl)
 	reg[1] = REG_HDMI_HDCP_SHA_CTRL;
 	data[1] = HDCP_REG_DISABLE;
 
-	rc = hdmi_hdcp_scm_wr(hdcp_ctrl, reg, data, 2);
+	rc = msm_hdmi_hdcp_scm_wr(hdcp_ctrl, reg, data, 2);
 
 	return rc;
 }
 
-static int hdmi_hdcp_auth_part2_recv_ksv_fifo(
+static int msm_hdmi_hdcp_auth_part2_recv_ksv_fifo(
 	struct hdmi_hdcp_ctrl *hdcp_ctrl)
 {
 	int rc;
@@ -1081,7 +1081,7 @@ static int hdmi_hdcp_auth_part2_recv_ksv_fifo(
 	 */
 	timeout_count = 100;
 	do {
-		rc = hdmi_hdcp_recv_ksv_fifo(hdcp_ctrl);
+		rc = msm_hdmi_hdcp_recv_ksv_fifo(hdcp_ctrl);
 		if (!rc)
 			break;
 
@@ -1091,19 +1091,19 @@ static int hdmi_hdcp_auth_part2_recv_ksv_fifo(
 			return -ETIMEDOUT;
 		}
 
-		rc = hdmi_hdcp_msleep(hdcp_ctrl, 25, AUTH_ABORT_EV);
+		rc = msm_hdmi_hdcp_msleep(hdcp_ctrl, 25, AUTH_ABORT_EV);
 		if (rc)
 			return rc;
 	} while (1);
 
-	rc = hdmi_hdcp_transfer_v_h(hdcp_ctrl);
+	rc = msm_hdmi_hdcp_transfer_v_h(hdcp_ctrl);
 	if (rc) {
 		pr_err("%s: transfer V failed\n", __func__);
 		return rc;
 	}
 
 	/* reset SHA engine before write ksv fifo */
-	rc = hdmi_hdcp_reset_sha_engine(hdcp_ctrl);
+	rc = msm_hdmi_hdcp_reset_sha_engine(hdcp_ctrl);
 	if (rc) {
 		pr_err("%s: fail to reset sha engine\n", __func__);
 		return rc;
@@ -1120,7 +1120,7 @@ static int hdmi_hdcp_auth_part2_recv_ksv_fifo(
  * If the last byte is written, we need to poll for
  * HDCP_SHA_COMP_DONE to wait until HW finish
  */
-static int hdmi_hdcp_write_ksv_fifo(struct hdmi_hdcp_ctrl *hdcp_ctrl)
+static int msm_hdmi_hdcp_write_ksv_fifo(struct hdmi_hdcp_ctrl *hdcp_ctrl)
 {
 	int i;
 	struct hdmi *hdmi = hdcp_ctrl->hdmi;
@@ -1169,7 +1169,7 @@ static int hdmi_hdcp_write_ksv_fifo(struct hdmi_hdcp_ctrl *hdcp_ctrl)
 
 		reg = REG_HDMI_HDCP_SHA_DATA;
 		data = reg_val;
-		rc = hdmi_hdcp_scm_wr(hdcp_ctrl, &reg, &data, 1);
+		rc = msm_hdmi_hdcp_scm_wr(hdcp_ctrl, &reg, &data, 1);
 
 		if (rc)
 			return rc;
@@ -1184,7 +1184,7 @@ static int hdmi_hdcp_write_ksv_fifo(struct hdmi_hdcp_ctrl *hdcp_ctrl)
 }
 
 /* write ksv fifo into HDCP engine */
-static int hdmi_hdcp_auth_part2_write_ksv_fifo(
+static int msm_hdmi_hdcp_auth_part2_write_ksv_fifo(
 	struct hdmi_hdcp_ctrl *hdcp_ctrl)
 {
 	int rc;
@@ -1193,7 +1193,7 @@ static int hdmi_hdcp_auth_part2_write_ksv_fifo(
 	hdcp_ctrl->ksv_fifo_w_index = 0;
 	timeout_count = 100;
 	do {
-		rc = hdmi_hdcp_write_ksv_fifo(hdcp_ctrl);
+		rc = msm_hdmi_hdcp_write_ksv_fifo(hdcp_ctrl);
 		if (!rc)
 			break;
 
@@ -1206,7 +1206,7 @@ static int hdmi_hdcp_auth_part2_write_ksv_fifo(
 			return -ETIMEDOUT;
 		}
 
-		rc = hdmi_hdcp_msleep(hdcp_ctrl, 20, AUTH_ABORT_EV);
+		rc = msm_hdmi_hdcp_msleep(hdcp_ctrl, 20, AUTH_ABORT_EV);
 		if (rc)
 			return rc;
 	} while (1);
@@ -1214,7 +1214,7 @@ static int hdmi_hdcp_auth_part2_write_ksv_fifo(
 	return 0;
 }
 
-static int hdmi_hdcp_auth_part2_check_v_match(struct hdmi_hdcp_ctrl *hdcp_ctrl)
+static int msm_hdmi_hdcp_auth_part2_check_v_match(struct hdmi_hdcp_ctrl *hdcp_ctrl)
 {
 	int rc = 0;
 	struct hdmi *hdmi = hdcp_ctrl->hdmi;
@@ -1232,7 +1232,7 @@ static int hdmi_hdcp_auth_part2_check_v_match(struct hdmi_hdcp_ctrl *hdcp_ctrl)
 				return -ETIMEDOUT;
 		}
 
-		rc = hdmi_hdcp_msleep(hdcp_ctrl, 20, AUTH_ABORT_EV);
+		rc = msm_hdmi_hdcp_msleep(hdcp_ctrl, 20, AUTH_ABORT_EV);
 		if (rc)
 			return rc;
 	} while (1);
@@ -1240,32 +1240,32 @@ static int hdmi_hdcp_auth_part2_check_v_match(struct hdmi_hdcp_ctrl *hdcp_ctrl)
 	return 0;
 }
 
-static void hdmi_hdcp_auth_work(struct work_struct *work)
+static void msm_hdmi_hdcp_auth_work(struct work_struct *work)
 {
 	struct hdmi_hdcp_ctrl *hdcp_ctrl = container_of(work,
 		struct hdmi_hdcp_ctrl, hdcp_auth_work);
 	int rc;
 
-	rc = hdmi_hdcp_auth_prepare(hdcp_ctrl);
+	rc = msm_hdmi_hdcp_auth_prepare(hdcp_ctrl);
 	if (rc) {
 		pr_err("%s: auth prepare failed %d\n", __func__, rc);
 		goto end;
 	}
 
 	/* HDCP PartI */
-	rc = hdmi_hdcp_auth_part1_key_exchange(hdcp_ctrl);
+	rc = msm_hdmi_hdcp_auth_part1_key_exchange(hdcp_ctrl);
 	if (rc) {
 		pr_err("%s: key exchange failed %d\n", __func__, rc);
 		goto end;
 	}
 
-	rc = hdmi_hdcp_auth_part1_recv_r0(hdcp_ctrl);
+	rc = msm_hdmi_hdcp_auth_part1_recv_r0(hdcp_ctrl);
 	if (rc) {
 		pr_err("%s: receive r0 failed %d\n", __func__, rc);
 		goto end;
 	}
 
-	rc = hdmi_hdcp_auth_part1_verify_r0(hdcp_ctrl);
+	rc = msm_hdmi_hdcp_auth_part1_verify_r0(hdcp_ctrl);
 	if (rc) {
 		pr_err("%s: verify r0 failed %d\n", __func__, rc);
 		goto end;
@@ -1275,25 +1275,25 @@ static void hdmi_hdcp_auth_work(struct work_struct *work)
 		goto end;
 
 	/* HDCP PartII */
-	rc = hdmi_hdcp_auth_part2_wait_ksv_fifo_ready(hdcp_ctrl);
+	rc = msm_hdmi_hdcp_auth_part2_wait_ksv_fifo_ready(hdcp_ctrl);
 	if (rc) {
 		pr_err("%s: wait ksv fifo ready failed %d\n", __func__, rc);
 		goto end;
 	}
 
-	rc = hdmi_hdcp_auth_part2_recv_ksv_fifo(hdcp_ctrl);
+	rc = msm_hdmi_hdcp_auth_part2_recv_ksv_fifo(hdcp_ctrl);
 	if (rc) {
 		pr_err("%s: recv ksv fifo failed %d\n", __func__, rc);
 		goto end;
 	}
 
-	rc = hdmi_hdcp_auth_part2_write_ksv_fifo(hdcp_ctrl);
+	rc = msm_hdmi_hdcp_auth_part2_write_ksv_fifo(hdcp_ctrl);
 	if (rc) {
 		pr_err("%s: write ksv fifo failed %d\n", __func__, rc);
 		goto end;
 	}
 
-	rc = hdmi_hdcp_auth_part2_check_v_match(hdcp_ctrl);
+	rc = msm_hdmi_hdcp_auth_part2_check_v_match(hdcp_ctrl);
 	if (rc)
 		pr_err("%s: check v match failed %d\n", __func__, rc);
 
@@ -1304,13 +1304,13 @@ end:
 		pr_info("%s: hdcp is not supported\n", __func__);
 	} else if (rc) {
 		pr_err("%s: hdcp authentication failed\n", __func__);
-		hdmi_hdcp_auth_fail(hdcp_ctrl);
+		msm_hdmi_hdcp_auth_fail(hdcp_ctrl);
 	} else {
-		hdmi_hdcp_auth_done(hdcp_ctrl);
+		msm_hdmi_hdcp_auth_done(hdcp_ctrl);
 	}
 }
 
-void hdmi_hdcp_on(struct hdmi_hdcp_ctrl *hdcp_ctrl)
+void msm_hdmi_hdcp_on(struct hdmi_hdcp_ctrl *hdcp_ctrl)
 {
 	struct hdmi *hdmi = hdcp_ctrl->hdmi;
 	u32 reg_val;
@@ -1335,7 +1335,7 @@ void hdmi_hdcp_on(struct hdmi_hdcp_ctrl *hdcp_ctrl)
 	queue_work(hdmi->workq, &hdcp_ctrl->hdcp_auth_work);
 }
 
-void hdmi_hdcp_off(struct hdmi_hdcp_ctrl *hdcp_ctrl)
+void msm_hdmi_hdcp_off(struct hdmi_hdcp_ctrl *hdcp_ctrl)
 {
 	struct hdmi *hdmi = hdcp_ctrl->hdmi;
 	unsigned long flags;
@@ -1399,7 +1399,7 @@ void hdmi_hdcp_off(struct hdmi_hdcp_ctrl *hdcp_ctrl)
 	DBG("HDCP: Off");
 }
 
-struct hdmi_hdcp_ctrl *hdmi_hdcp_init(struct hdmi *hdmi)
+struct hdmi_hdcp_ctrl *msm_hdmi_hdcp_init(struct hdmi *hdmi)
 {
 	struct hdmi_hdcp_ctrl *hdcp_ctrl = NULL;
 
@@ -1413,8 +1413,8 @@ struct hdmi_hdcp_ctrl *hdmi_hdcp_init(struct hdmi *hdmi)
 	if (!hdcp_ctrl)
 		return ERR_PTR(-ENOMEM);
 
-	INIT_WORK(&hdcp_ctrl->hdcp_auth_work, hdmi_hdcp_auth_work);
-	INIT_WORK(&hdcp_ctrl->hdcp_reauth_work, hdmi_hdcp_reauth_work);
+	INIT_WORK(&hdcp_ctrl->hdcp_auth_work, msm_hdmi_hdcp_auth_work);
+	INIT_WORK(&hdcp_ctrl->hdcp_reauth_work, msm_hdmi_hdcp_reauth_work);
 	init_waitqueue_head(&hdcp_ctrl->auth_event_queue);
 	hdcp_ctrl->hdmi = hdmi;
 	hdcp_ctrl->hdcp_state = HDCP_STATE_INACTIVE;
@@ -1428,7 +1428,7 @@ struct hdmi_hdcp_ctrl *hdmi_hdcp_init(struct hdmi *hdmi)
 	return hdcp_ctrl;
 }
 
-void hdmi_hdcp_destroy(struct hdmi *hdmi)
+void msm_hdmi_hdcp_destroy(struct hdmi *hdmi)
 {
 	if (hdmi && hdmi->hdcp_ctrl) {
 		kfree(hdmi->hdcp_ctrl);
diff --git a/drivers/gpu/drm/msm/hdmi/hdmi_i2c.c b/drivers/gpu/drm/msm/hdmi/hdmi_i2c.c
index f4ab7f70fed1..de9007e72f4e 100644
--- a/drivers/gpu/drm/msm/hdmi/hdmi_i2c.c
+++ b/drivers/gpu/drm/msm/hdmi/hdmi_i2c.c
@@ -97,7 +97,7 @@ static bool sw_done(struct hdmi_i2c_adapter *hdmi_i2c)
 	return hdmi_i2c->sw_done;
 }
 
-static int hdmi_i2c_xfer(struct i2c_adapter *i2c,
+static int msm_hdmi_i2c_xfer(struct i2c_adapter *i2c,
 		struct i2c_msg *msgs, int num)
 {
 	struct hdmi_i2c_adapter *hdmi_i2c = to_hdmi_i2c_adapter(i2c);
@@ -216,17 +216,17 @@ static int hdmi_i2c_xfer(struct i2c_adapter *i2c,
 	return i;
 }
 
-static u32 hdmi_i2c_func(struct i2c_adapter *adapter)
+static u32 msm_hdmi_i2c_func(struct i2c_adapter *adapter)
 {
 	return I2C_FUNC_I2C | I2C_FUNC_SMBUS_EMUL;
 }
 
-static const struct i2c_algorithm hdmi_i2c_algorithm = {
-	.master_xfer	= hdmi_i2c_xfer,
-	.functionality	= hdmi_i2c_func,
+static const struct i2c_algorithm msm_hdmi_i2c_algorithm = {
+	.master_xfer	= msm_hdmi_i2c_xfer,
+	.functionality	= msm_hdmi_i2c_func,
 };
 
-void hdmi_i2c_irq(struct i2c_adapter *i2c)
+void msm_hdmi_i2c_irq(struct i2c_adapter *i2c)
 {
 	struct hdmi_i2c_adapter *hdmi_i2c = to_hdmi_i2c_adapter(i2c);
 
@@ -234,14 +234,14 @@ void hdmi_i2c_irq(struct i2c_adapter *i2c)
 		wake_up_all(&hdmi_i2c->ddc_event);
 }
 
-void hdmi_i2c_destroy(struct i2c_adapter *i2c)
+void msm_hdmi_i2c_destroy(struct i2c_adapter *i2c)
 {
 	struct hdmi_i2c_adapter *hdmi_i2c = to_hdmi_i2c_adapter(i2c);
 	i2c_del_adapter(i2c);
 	kfree(hdmi_i2c);
 }
 
-struct i2c_adapter *hdmi_i2c_init(struct hdmi *hdmi)
+struct i2c_adapter *msm_hdmi_i2c_init(struct hdmi *hdmi)
 {
 	struct drm_device *dev = hdmi->dev;
 	struct hdmi_i2c_adapter *hdmi_i2c;
@@ -264,7 +264,7 @@ struct i2c_adapter *hdmi_i2c_init(struct hdmi *hdmi)
 	i2c->class = I2C_CLASS_DDC;
 	snprintf(i2c->name, sizeof(i2c->name), "msm hdmi i2c");
 	i2c->dev.parent = &hdmi->pdev->dev;
-	i2c->algo = &hdmi_i2c_algorithm;
+	i2c->algo = &msm_hdmi_i2c_algorithm;
 
 	ret = i2c_add_adapter(i2c);
 	if (ret) {
@@ -276,6 +276,6 @@ struct i2c_adapter *hdmi_i2c_init(struct hdmi *hdmi)
 
 fail:
 	if (i2c)
-		hdmi_i2c_destroy(i2c);
+		msm_hdmi_i2c_destroy(i2c);
 	return ERR_PTR(ret);
 }
diff --git a/drivers/gpu/drm/msm/hdmi/hdmi_phy.c b/drivers/gpu/drm/msm/hdmi/hdmi_phy.c
new file mode 100644
index 000000000000..534ce5b49781
--- /dev/null
+++ b/drivers/gpu/drm/msm/hdmi/hdmi_phy.c
@@ -0,0 +1,230 @@
+/*
+ * Copyright (c) 2016, The Linux Foundation. All rights reserved.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 and
+ * only version 2 as published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ */
+
+#include <linux/of_device.h>
+
+#include "hdmi.h"
+
+static int msm_hdmi_phy_resource_init(struct hdmi_phy *phy)
+{
+	struct hdmi_phy_cfg *cfg = phy->cfg;
+	struct device *dev = &phy->pdev->dev;
+	int i, ret;
+
+	phy->regs = devm_kzalloc(dev, sizeof(phy->regs[0]) * cfg->num_regs,
+				 GFP_KERNEL);
+	if (!phy->regs)
+		return -ENOMEM;
+
+	phy->clks = devm_kzalloc(dev, sizeof(phy->clks[0]) * cfg->num_clks,
+				 GFP_KERNEL);
+	if (!phy->clks)
+		return -ENOMEM;
+
+	for (i = 0; i < cfg->num_regs; i++) {
+		struct regulator *reg;
+
+		reg = devm_regulator_get(dev, cfg->reg_names[i]);
+		if (IS_ERR(reg)) {
+			ret = PTR_ERR(reg);
+			dev_err(dev, "failed to get phy regulator: %s (%d)\n",
+				cfg->reg_names[i], ret);
+			return ret;
+		}
+
+		phy->regs[i] = reg;
+	}
+
+	for (i = 0; i < cfg->num_clks; i++) {
+		struct clk *clk;
+
+		clk = devm_clk_get(dev, cfg->clk_names[i]);
+		if (IS_ERR(clk)) {
+			ret = PTR_ERR(clk);
+			dev_err(dev, "failed to get phy clock: %s (%d)\n",
+				cfg->clk_names[i], ret);
+			return ret;
+		}
+
+		phy->clks[i] = clk;
+	}
+
+	return 0;
+}
+
+int msm_hdmi_phy_resource_enable(struct hdmi_phy *phy)
+{
+	struct hdmi_phy_cfg *cfg = phy->cfg;
+	struct device *dev = &phy->pdev->dev;
+	int i, ret = 0;
+
+	pm_runtime_get_sync(dev);
+
+	for (i = 0; i < cfg->num_regs; i++) {
+		ret = regulator_enable(phy->regs[i]);
+		if (ret)
+			dev_err(dev, "failed to enable regulator: %s (%d)\n",
+				cfg->reg_names[i], ret);
+	}
+
+	for (i = 0; i < cfg->num_clks; i++) {
+		ret = clk_prepare_enable(phy->clks[i]);
+		if (ret)
+			dev_err(dev, "failed to enable clock: %s (%d)\n",
+				cfg->clk_names[i], ret);
+	}
+
+	return ret;
+}
+
+void msm_hdmi_phy_resource_disable(struct hdmi_phy *phy)
+{
+	struct hdmi_phy_cfg *cfg = phy->cfg;
+	struct device *dev = &phy->pdev->dev;
+	int i;
+
+	for (i = cfg->num_clks - 1; i >= 0; i--)
+		clk_disable_unprepare(phy->clks[i]);
+
+	for (i = cfg->num_regs - 1; i >= 0; i--)
+		regulator_disable(phy->regs[i]);
+
+	pm_runtime_put_sync(dev);
+}
+
+void msm_hdmi_phy_powerup(struct hdmi_phy *phy, unsigned long int pixclock)
+{
+	if (!phy || !phy->cfg->powerup)
+		return;
+
+	phy->cfg->powerup(phy, pixclock);
+}
+
+void msm_hdmi_phy_powerdown(struct hdmi_phy *phy)
+{
+	if (!phy || !phy->cfg->powerdown)
+		return;
+
+	phy->cfg->powerdown(phy);
+}
+
+static int msm_hdmi_phy_pll_init(struct platform_device *pdev,
+			     enum hdmi_phy_type type)
+{
+	int ret;
+
+	switch (type) {
+	case MSM_HDMI_PHY_8960:
+		ret = msm_hdmi_pll_8960_init(pdev);
+		break;
+	case MSM_HDMI_PHY_8996:
+		ret = msm_hdmi_pll_8996_init(pdev);
+		break;
+	/*
+	 * we don't have PLL support for these, don't report an error for now
+	 */
+	case MSM_HDMI_PHY_8x60:
+	case MSM_HDMI_PHY_8x74:
+	default:
+		ret = 0;
+		break;
+	}
+
+	return ret;
+}
+
+static int msm_hdmi_phy_probe(struct platform_device *pdev)
+{
+	struct device *dev = &pdev->dev;
+	struct hdmi_phy *phy;
+	int ret;
+
+	phy = devm_kzalloc(dev, sizeof(*phy), GFP_KERNEL);
+	if (!phy)
+		return -ENODEV;
+
+	phy->cfg = (struct hdmi_phy_cfg *)of_device_get_match_data(dev);
+	if (!phy->cfg)
+		return -ENODEV;
+
+	phy->mmio = msm_ioremap(pdev, "hdmi_phy", "HDMI_PHY");
+	if (IS_ERR(phy->mmio)) {
+		dev_err(dev, "%s: failed to map phy base\n", __func__);
+		return -ENOMEM;
+	}
+
+	phy->pdev = pdev;
+
+	ret = msm_hdmi_phy_resource_init(phy);
+	if (ret)
+		return ret;
+
+	pm_runtime_enable(&pdev->dev);
+
+	ret = msm_hdmi_phy_resource_enable(phy);
+	if (ret)
+		return ret;
+
+	ret = msm_hdmi_phy_pll_init(pdev, phy->cfg->type);
+	if (ret) {
+		dev_err(dev, "couldn't init PLL\n");
+		msm_hdmi_phy_resource_disable(phy);
+		return ret;
+	}
+
+	msm_hdmi_phy_resource_disable(phy);
+
+	platform_set_drvdata(pdev, phy);
+
+	return 0;
+}
+
+static int msm_hdmi_phy_remove(struct platform_device *pdev)
+{
+	pm_runtime_disable(&pdev->dev);
+
+	return 0;
+}
+
+static const struct of_device_id msm_hdmi_phy_dt_match[] = {
+	{ .compatible = "qcom,hdmi-phy-8660",
+	  .data = &msm_hdmi_phy_8x60_cfg },
+	{ .compatible = "qcom,hdmi-phy-8960",
+	  .data = &msm_hdmi_phy_8960_cfg },
+	{ .compatible = "qcom,hdmi-phy-8974",
+	  .data = &msm_hdmi_phy_8x74_cfg },
+	{ .compatible = "qcom,hdmi-phy-8084",
+	  .data = &msm_hdmi_phy_8x74_cfg },
+	{ .compatible = "qcom,hdmi-phy-8996",
+	  .data = &msm_hdmi_phy_8996_cfg },
+	{}
+};
+
+static struct platform_driver msm_hdmi_phy_platform_driver = {
+	.probe      = msm_hdmi_phy_probe,
+	.remove     = msm_hdmi_phy_remove,
+	.driver     = {
+		.name   = "msm_hdmi_phy",
+		.of_match_table = msm_hdmi_phy_dt_match,
+	},
+};
+
+void __init msm_hdmi_phy_driver_register(void)
+{
+	platform_driver_register(&msm_hdmi_phy_platform_driver);
+}
+
+void __exit msm_hdmi_phy_driver_unregister(void)
+{
+	platform_driver_unregister(&msm_hdmi_phy_platform_driver);
+}
diff --git a/drivers/gpu/drm/msm/hdmi/hdmi_phy_8960.c b/drivers/gpu/drm/msm/hdmi/hdmi_phy_8960.c
index 3a01cb5051e2..e6ee6b745ab7 100644
--- a/drivers/gpu/drm/msm/hdmi/hdmi_phy_8960.c
+++ b/drivers/gpu/drm/msm/hdmi/hdmi_phy_8960.c
@@ -15,495 +15,48 @@
  * this program.  If not, see <http://www.gnu.org/licenses/>.
  */
 
-#ifdef CONFIG_COMMON_CLK
-#include <linux/clk.h>
-#include <linux/clk-provider.h>
-#endif
-
 #include "hdmi.h"
 
-struct hdmi_phy_8960 {
-	struct hdmi_phy base;
-	struct hdmi *hdmi;
-#ifdef CONFIG_COMMON_CLK
-	struct clk_hw pll_hw;
-	struct clk *pll;
-	unsigned long pixclk;
-#endif
-};
-#define to_hdmi_phy_8960(x) container_of(x, struct hdmi_phy_8960, base)
-
-#ifdef CONFIG_COMMON_CLK
-#define clk_to_phy(x) container_of(x, struct hdmi_phy_8960, pll_hw)
-
-/*
- * HDMI PLL:
- *
- * To get the parent clock setup properly, we need to plug in hdmi pll
- * configuration into common-clock-framework.
- */
-
-struct pll_rate {
-	unsigned long rate;
-	struct {
-		uint32_t val;
-		uint32_t reg;
-	} conf[32];
-};
-
-/* NOTE: keep sorted highest freq to lowest: */
-static const struct pll_rate freqtbl[] = {
-	{ 154000000, {
-		{ 0x08, REG_HDMI_8960_PHY_PLL_REFCLK_CFG    },
-		{ 0x20, REG_HDMI_8960_PHY_PLL_LOOP_FLT_CFG0 },
-		{ 0xf9, REG_HDMI_8960_PHY_PLL_LOOP_FLT_CFG1 },
-		{ 0x02, REG_HDMI_8960_PHY_PLL_VCOCAL_CFG0   },
-		{ 0x03, REG_HDMI_8960_PHY_PLL_VCOCAL_CFG1   },
-		{ 0x3b, REG_HDMI_8960_PHY_PLL_VCOCAL_CFG2   },
-		{ 0x00, REG_HDMI_8960_PHY_PLL_VCOCAL_CFG3   },
-		{ 0x86, REG_HDMI_8960_PHY_PLL_VCOCAL_CFG4   },
-		{ 0x00, REG_HDMI_8960_PHY_PLL_VCOCAL_CFG5   },
-		{ 0x0d, REG_HDMI_8960_PHY_PLL_SDM_CFG0      },
-		{ 0x4d, REG_HDMI_8960_PHY_PLL_SDM_CFG1      },
-		{ 0x5e, REG_HDMI_8960_PHY_PLL_SDM_CFG2      },
-		{ 0x42, REG_HDMI_8960_PHY_PLL_SDM_CFG3      },
-		{ 0x00, REG_HDMI_8960_PHY_PLL_SDM_CFG4      },
-		{ 0, 0 } }
-	},
-	/* 1080p60/1080p50 case */
-	{ 148500000, {
-		{ 0x02, REG_HDMI_8960_PHY_PLL_REFCLK_CFG    },
-		{ 0x02, REG_HDMI_8960_PHY_PLL_CHRG_PUMP_CFG },
-		{ 0x01, REG_HDMI_8960_PHY_PLL_LOOP_FLT_CFG0 },
-		{ 0x33, REG_HDMI_8960_PHY_PLL_LOOP_FLT_CFG1 },
-		{ 0x2c, REG_HDMI_8960_PHY_PLL_IDAC_ADJ_CFG  },
-		{ 0x06, REG_HDMI_8960_PHY_PLL_I_VI_KVCO_CFG },
-		{ 0x0a, REG_HDMI_8960_PHY_PLL_PWRDN_B       },
-		{ 0x76, REG_HDMI_8960_PHY_PLL_SDM_CFG0      },
-		{ 0x01, REG_HDMI_8960_PHY_PLL_SDM_CFG1      },
-		{ 0x4c, REG_HDMI_8960_PHY_PLL_SDM_CFG2      },
-		{ 0xc0, REG_HDMI_8960_PHY_PLL_SDM_CFG3      },
-		{ 0x00, REG_HDMI_8960_PHY_PLL_SDM_CFG4      },
-		{ 0x9a, REG_HDMI_8960_PHY_PLL_SSC_CFG0      },
-		{ 0x00, REG_HDMI_8960_PHY_PLL_SSC_CFG1      },
-		{ 0x00, REG_HDMI_8960_PHY_PLL_SSC_CFG2      },
-		{ 0x00, REG_HDMI_8960_PHY_PLL_SSC_CFG3      },
-		{ 0x10, REG_HDMI_8960_PHY_PLL_LOCKDET_CFG0  },
-		{ 0x1a, REG_HDMI_8960_PHY_PLL_LOCKDET_CFG1  },
-		{ 0x0d, REG_HDMI_8960_PHY_PLL_LOCKDET_CFG2  },
-		{ 0xe6, REG_HDMI_8960_PHY_PLL_VCOCAL_CFG0   },
-		{ 0x02, REG_HDMI_8960_PHY_PLL_VCOCAL_CFG1   },
-		{ 0x3b, REG_HDMI_8960_PHY_PLL_VCOCAL_CFG2   },
-		{ 0x00, REG_HDMI_8960_PHY_PLL_VCOCAL_CFG3   },
-		{ 0x86, REG_HDMI_8960_PHY_PLL_VCOCAL_CFG4   },
-		{ 0x00, REG_HDMI_8960_PHY_PLL_VCOCAL_CFG5   },
-		{ 0x33, REG_HDMI_8960_PHY_PLL_VCOCAL_CFG6   },
-		{ 0x00, REG_HDMI_8960_PHY_PLL_VCOCAL_CFG7   },
-		{ 0, 0 } }
-	},
-	{ 108000000, {
-		{ 0x08, REG_HDMI_8960_PHY_PLL_REFCLK_CFG    },
-		{ 0x21, REG_HDMI_8960_PHY_PLL_LOOP_FLT_CFG0 },
-		{ 0xf9, REG_HDMI_8960_PHY_PLL_LOOP_FLT_CFG1 },
-		{ 0x1c, REG_HDMI_8960_PHY_PLL_VCOCAL_CFG0   },
-		{ 0x02, REG_HDMI_8960_PHY_PLL_VCOCAL_CFG1   },
-		{ 0x3b, REG_HDMI_8960_PHY_PLL_VCOCAL_CFG2   },
-		{ 0x86, REG_HDMI_8960_PHY_PLL_VCOCAL_CFG4   },
-		{ 0x00, REG_HDMI_8960_PHY_PLL_VCOCAL_CFG5   },
-		{ 0x49, REG_HDMI_8960_PHY_PLL_SDM_CFG0      },
-		{ 0x49, REG_HDMI_8960_PHY_PLL_SDM_CFG1      },
-		{ 0x00, REG_HDMI_8960_PHY_PLL_SDM_CFG2      },
-		{ 0x00, REG_HDMI_8960_PHY_PLL_SDM_CFG3      },
-		{ 0x00, REG_HDMI_8960_PHY_PLL_SDM_CFG4      },
-		{ 0, 0 } }
-	},
-	/* 720p60/720p50/1080i60/1080i50/1080p24/1080p30/1080p25 */
-	{ 74250000, {
-		{ 0x0a, REG_HDMI_8960_PHY_PLL_PWRDN_B       },
-		{ 0x12, REG_HDMI_8960_PHY_PLL_REFCLK_CFG    },
-		{ 0x01, REG_HDMI_8960_PHY_PLL_LOOP_FLT_CFG0 },
-		{ 0x33, REG_HDMI_8960_PHY_PLL_LOOP_FLT_CFG1 },
-		{ 0x76, REG_HDMI_8960_PHY_PLL_SDM_CFG0      },
-		{ 0xe6, REG_HDMI_8960_PHY_PLL_VCOCAL_CFG0   },
-		{ 0x02, REG_HDMI_8960_PHY_PLL_VCOCAL_CFG1   },
-		{ 0x3b, REG_HDMI_8960_PHY_PLL_VCOCAL_CFG2   },
-		{ 0, 0 } }
-	},
-	{ 74176000, {
-		{ 0x18, REG_HDMI_8960_PHY_PLL_REFCLK_CFG    },
-		{ 0x20, REG_HDMI_8960_PHY_PLL_LOOP_FLT_CFG0 },
-		{ 0xf9, REG_HDMI_8960_PHY_PLL_LOOP_FLT_CFG1 },
-		{ 0xe5, REG_HDMI_8960_PHY_PLL_VCOCAL_CFG0   },
-		{ 0x02, REG_HDMI_8960_PHY_PLL_VCOCAL_CFG1   },
-		{ 0x3b, REG_HDMI_8960_PHY_PLL_VCOCAL_CFG2   },
-		{ 0x00, REG_HDMI_8960_PHY_PLL_VCOCAL_CFG3   },
-		{ 0x86, REG_HDMI_8960_PHY_PLL_VCOCAL_CFG4   },
-		{ 0x00, REG_HDMI_8960_PHY_PLL_VCOCAL_CFG5   },
-		{ 0x0c, REG_HDMI_8960_PHY_PLL_SDM_CFG0      },
-		{ 0x4c, REG_HDMI_8960_PHY_PLL_SDM_CFG1      },
-		{ 0x7d, REG_HDMI_8960_PHY_PLL_SDM_CFG2      },
-		{ 0xbc, REG_HDMI_8960_PHY_PLL_SDM_CFG3      },
-		{ 0x00, REG_HDMI_8960_PHY_PLL_SDM_CFG4      },
-		{ 0, 0 } }
-	},
-	{ 65000000, {
-		{ 0x18, REG_HDMI_8960_PHY_PLL_REFCLK_CFG    },
-		{ 0x20, REG_HDMI_8960_PHY_PLL_LOOP_FLT_CFG0 },
-		{ 0xf9, REG_HDMI_8960_PHY_PLL_LOOP_FLT_CFG1 },
-		{ 0x8a, REG_HDMI_8960_PHY_PLL_VCOCAL_CFG0   },
-		{ 0x02, REG_HDMI_8960_PHY_PLL_VCOCAL_CFG1   },
-		{ 0x3b, REG_HDMI_8960_PHY_PLL_VCOCAL_CFG2   },
-		{ 0x00, REG_HDMI_8960_PHY_PLL_VCOCAL_CFG3   },
-		{ 0x86, REG_HDMI_8960_PHY_PLL_VCOCAL_CFG4   },
-		{ 0x00, REG_HDMI_8960_PHY_PLL_VCOCAL_CFG5   },
-		{ 0x0b, REG_HDMI_8960_PHY_PLL_SDM_CFG0      },
-		{ 0x4b, REG_HDMI_8960_PHY_PLL_SDM_CFG1      },
-		{ 0x7b, REG_HDMI_8960_PHY_PLL_SDM_CFG2      },
-		{ 0x09, REG_HDMI_8960_PHY_PLL_SDM_CFG3      },
-		{ 0x00, REG_HDMI_8960_PHY_PLL_SDM_CFG4      },
-		{ 0, 0 } }
-	},
-	/* 480p60/480i60 */
-	{ 27030000, {
-		{ 0x0a, REG_HDMI_8960_PHY_PLL_PWRDN_B       },
-		{ 0x38, REG_HDMI_8960_PHY_PLL_REFCLK_CFG    },
-		{ 0x02, REG_HDMI_8960_PHY_PLL_CHRG_PUMP_CFG },
-		{ 0x20, REG_HDMI_8960_PHY_PLL_LOOP_FLT_CFG0 },
-		{ 0xff, REG_HDMI_8960_PHY_PLL_LOOP_FLT_CFG1 },
-		{ 0x00, REG_HDMI_8960_PHY_PLL_SDM_CFG0      },
-		{ 0x4e, REG_HDMI_8960_PHY_PLL_SDM_CFG1      },
-		{ 0xd7, REG_HDMI_8960_PHY_PLL_SDM_CFG2      },
-		{ 0x03, REG_HDMI_8960_PHY_PLL_SDM_CFG3      },
-		{ 0x00, REG_HDMI_8960_PHY_PLL_SDM_CFG4      },
-		{ 0x2a, REG_HDMI_8960_PHY_PLL_VCOCAL_CFG0   },
-		{ 0x03, REG_HDMI_8960_PHY_PLL_VCOCAL_CFG1   },
-		{ 0x3b, REG_HDMI_8960_PHY_PLL_VCOCAL_CFG2   },
-		{ 0x00, REG_HDMI_8960_PHY_PLL_VCOCAL_CFG3   },
-		{ 0x86, REG_HDMI_8960_PHY_PLL_VCOCAL_CFG4   },
-		{ 0x00, REG_HDMI_8960_PHY_PLL_VCOCAL_CFG5   },
-		{ 0x33, REG_HDMI_8960_PHY_PLL_VCOCAL_CFG6   },
-		{ 0x00, REG_HDMI_8960_PHY_PLL_VCOCAL_CFG7   },
-		{ 0, 0 } }
-	},
-	/* 576p50/576i50 */
-	{ 27000000, {
-		{ 0x32, REG_HDMI_8960_PHY_PLL_REFCLK_CFG    },
-		{ 0x02, REG_HDMI_8960_PHY_PLL_CHRG_PUMP_CFG },
-		{ 0x01, REG_HDMI_8960_PHY_PLL_LOOP_FLT_CFG0 },
-		{ 0x33, REG_HDMI_8960_PHY_PLL_LOOP_FLT_CFG1 },
-		{ 0x2c, REG_HDMI_8960_PHY_PLL_IDAC_ADJ_CFG  },
-		{ 0x06, REG_HDMI_8960_PHY_PLL_I_VI_KVCO_CFG },
-		{ 0x0a, REG_HDMI_8960_PHY_PLL_PWRDN_B       },
-		{ 0x7b, REG_HDMI_8960_PHY_PLL_SDM_CFG0      },
-		{ 0x01, REG_HDMI_8960_PHY_PLL_SDM_CFG1      },
-		{ 0x4c, REG_HDMI_8960_PHY_PLL_SDM_CFG2      },
-		{ 0xc0, REG_HDMI_8960_PHY_PLL_SDM_CFG3      },
-		{ 0x00, REG_HDMI_8960_PHY_PLL_SDM_CFG4      },
-		{ 0x9a, REG_HDMI_8960_PHY_PLL_SSC_CFG0      },
-		{ 0x00, REG_HDMI_8960_PHY_PLL_SSC_CFG1      },
-		{ 0x00, REG_HDMI_8960_PHY_PLL_SSC_CFG2      },
-		{ 0x00, REG_HDMI_8960_PHY_PLL_SSC_CFG3      },
-		{ 0x10, REG_HDMI_8960_PHY_PLL_LOCKDET_CFG0  },
-		{ 0x1a, REG_HDMI_8960_PHY_PLL_LOCKDET_CFG1  },
-		{ 0x0d, REG_HDMI_8960_PHY_PLL_LOCKDET_CFG2  },
-		{ 0x2a, REG_HDMI_8960_PHY_PLL_VCOCAL_CFG0   },
-		{ 0x03, REG_HDMI_8960_PHY_PLL_VCOCAL_CFG1   },
-		{ 0x3b, REG_HDMI_8960_PHY_PLL_VCOCAL_CFG2   },
-		{ 0x00, REG_HDMI_8960_PHY_PLL_VCOCAL_CFG3   },
-		{ 0x86, REG_HDMI_8960_PHY_PLL_VCOCAL_CFG4   },
-		{ 0x00, REG_HDMI_8960_PHY_PLL_VCOCAL_CFG5   },
-		{ 0x33, REG_HDMI_8960_PHY_PLL_VCOCAL_CFG6   },
-		{ 0x00, REG_HDMI_8960_PHY_PLL_VCOCAL_CFG7   },
-		{ 0, 0 } }
-	},
-	/* 640x480p60 */
-	{ 25200000, {
-		{ 0x32, REG_HDMI_8960_PHY_PLL_REFCLK_CFG    },
-		{ 0x02, REG_HDMI_8960_PHY_PLL_CHRG_PUMP_CFG },
-		{ 0x01, REG_HDMI_8960_PHY_PLL_LOOP_FLT_CFG0 },
-		{ 0x33, REG_HDMI_8960_PHY_PLL_LOOP_FLT_CFG1 },
-		{ 0x2c, REG_HDMI_8960_PHY_PLL_IDAC_ADJ_CFG  },
-		{ 0x06, REG_HDMI_8960_PHY_PLL_I_VI_KVCO_CFG },
-		{ 0x0a, REG_HDMI_8960_PHY_PLL_PWRDN_B       },
-		{ 0x77, REG_HDMI_8960_PHY_PLL_SDM_CFG0      },
-		{ 0x4c, REG_HDMI_8960_PHY_PLL_SDM_CFG1      },
-		{ 0x00, REG_HDMI_8960_PHY_PLL_SDM_CFG2      },
-		{ 0xc0, REG_HDMI_8960_PHY_PLL_SDM_CFG3      },
-		{ 0x00, REG_HDMI_8960_PHY_PLL_SDM_CFG4      },
-		{ 0x9a, REG_HDMI_8960_PHY_PLL_SSC_CFG0      },
-		{ 0x00, REG_HDMI_8960_PHY_PLL_SSC_CFG1      },
-		{ 0x00, REG_HDMI_8960_PHY_PLL_SSC_CFG2      },
-		{ 0x20, REG_HDMI_8960_PHY_PLL_SSC_CFG3      },
-		{ 0x10, REG_HDMI_8960_PHY_PLL_LOCKDET_CFG0  },
-		{ 0x1a, REG_HDMI_8960_PHY_PLL_LOCKDET_CFG1  },
-		{ 0x0d, REG_HDMI_8960_PHY_PLL_LOCKDET_CFG2  },
-		{ 0xf4, REG_HDMI_8960_PHY_PLL_VCOCAL_CFG0   },
-		{ 0x02, REG_HDMI_8960_PHY_PLL_VCOCAL_CFG1   },
-		{ 0x3b, REG_HDMI_8960_PHY_PLL_VCOCAL_CFG2   },
-		{ 0x00, REG_HDMI_8960_PHY_PLL_VCOCAL_CFG3   },
-		{ 0x86, REG_HDMI_8960_PHY_PLL_VCOCAL_CFG4   },
-		{ 0x00, REG_HDMI_8960_PHY_PLL_VCOCAL_CFG5   },
-		{ 0x33, REG_HDMI_8960_PHY_PLL_VCOCAL_CFG6   },
-		{ 0x00, REG_HDMI_8960_PHY_PLL_VCOCAL_CFG7   },
-		{ 0, 0 } }
-	},
-};
-
-static int hdmi_pll_enable(struct clk_hw *hw)
-{
-	struct hdmi_phy_8960 *phy_8960 = clk_to_phy(hw);
-	struct hdmi *hdmi = phy_8960->hdmi;
-	int timeout_count, pll_lock_retry = 10;
-	unsigned int val;
-
-	DBG("");
-
-	/* Assert PLL S/W reset */
-	hdmi_write(hdmi, REG_HDMI_8960_PHY_PLL_LOCKDET_CFG2, 0x8d);
-	hdmi_write(hdmi, REG_HDMI_8960_PHY_PLL_LOCKDET_CFG0, 0x10);
-	hdmi_write(hdmi, REG_HDMI_8960_PHY_PLL_LOCKDET_CFG1, 0x1a);
-
-	/* Wait for a short time before de-asserting
-	 * to allow the hardware to complete its job.
-	 * This much of delay should be fine for hardware
-	 * to assert and de-assert.
-	 */
-	udelay(10);
-
-	/* De-assert PLL S/W reset */
-	hdmi_write(hdmi, REG_HDMI_8960_PHY_PLL_LOCKDET_CFG2, 0x0d);
-
-	val = hdmi_read(hdmi, REG_HDMI_8960_PHY_REG12);
-	val |= HDMI_8960_PHY_REG12_SW_RESET;
-	/* Assert PHY S/W reset */
-	hdmi_write(hdmi, REG_HDMI_8960_PHY_REG12, val);
-	val &= ~HDMI_8960_PHY_REG12_SW_RESET;
-	/* Wait for a short time before de-asserting
-	   to allow the hardware to complete its job.
-	   This much of delay should be fine for hardware
-	   to assert and de-assert. */
-	udelay(10);
-	/* De-assert PHY S/W reset */
-	hdmi_write(hdmi, REG_HDMI_8960_PHY_REG12, val);
-	hdmi_write(hdmi, REG_HDMI_8960_PHY_REG2,  0x3f);
-
-	val = hdmi_read(hdmi, REG_HDMI_8960_PHY_REG12);
-	val |= HDMI_8960_PHY_REG12_PWRDN_B;
-	hdmi_write(hdmi, REG_HDMI_8960_PHY_REG12, val);
-	/* Wait 10 us for enabling global power for PHY */
-	mb();
-	udelay(10);
-
-	val = hdmi_read(hdmi, REG_HDMI_8960_PHY_PLL_PWRDN_B);
-	val |= HDMI_8960_PHY_PLL_PWRDN_B_PLL_PWRDN_B;
-	val &= ~HDMI_8960_PHY_PLL_PWRDN_B_PD_PLL;
-	hdmi_write(hdmi, REG_HDMI_8960_PHY_PLL_PWRDN_B, val);
-	hdmi_write(hdmi, REG_HDMI_8960_PHY_REG2, 0x80);
-
-	timeout_count = 1000;
-	while (--pll_lock_retry > 0) {
-
-		/* are we there yet? */
-		val = hdmi_read(hdmi, REG_HDMI_8960_PHY_PLL_STATUS0);
-		if (val & HDMI_8960_PHY_PLL_STATUS0_PLL_LOCK)
-			break;
-
-		udelay(1);
-
-		if (--timeout_count > 0)
-			continue;
-
-		/*
-		 * PLL has still not locked.
-		 * Do a software reset and try again
-		 * Assert PLL S/W reset first
-		 */
-		hdmi_write(hdmi, REG_HDMI_8960_PHY_PLL_LOCKDET_CFG2, 0x8d);
-		udelay(10);
-		hdmi_write(hdmi, REG_HDMI_8960_PHY_PLL_LOCKDET_CFG2, 0x0d);
-
-		/*
-		 * Wait for a short duration for the PLL calibration
-		 * before checking if the PLL gets locked
-		 */
-		udelay(350);
-
-		timeout_count = 1000;
-	}
-
-	return 0;
-}
-
-static void hdmi_pll_disable(struct clk_hw *hw)
-{
-	struct hdmi_phy_8960 *phy_8960 = clk_to_phy(hw);
-	struct hdmi *hdmi = phy_8960->hdmi;
-	unsigned int val;
-
-	DBG("");
-
-	val = hdmi_read(hdmi, REG_HDMI_8960_PHY_REG12);
-	val &= ~HDMI_8960_PHY_REG12_PWRDN_B;
-	hdmi_write(hdmi, REG_HDMI_8960_PHY_REG12, val);
-
-	val = hdmi_read(hdmi, REG_HDMI_8960_PHY_PLL_PWRDN_B);
-	val |= HDMI_8960_PHY_REG12_SW_RESET;
-	val &= ~HDMI_8960_PHY_REG12_PWRDN_B;
-	hdmi_write(hdmi, REG_HDMI_8960_PHY_PLL_PWRDN_B, val);
-	/* Make sure HDMI PHY/PLL are powered down */
-	mb();
-}
-
-static const struct pll_rate *find_rate(unsigned long rate)
-{
-	int i;
-	for (i = 1; i < ARRAY_SIZE(freqtbl); i++)
-		if (rate > freqtbl[i].rate)
-			return &freqtbl[i-1];
-	return &freqtbl[i-1];
-}
-
-static unsigned long hdmi_pll_recalc_rate(struct clk_hw *hw,
-				unsigned long parent_rate)
-{
-	struct hdmi_phy_8960 *phy_8960 = clk_to_phy(hw);
-	return phy_8960->pixclk;
-}
-
-static long hdmi_pll_round_rate(struct clk_hw *hw, unsigned long rate,
-		unsigned long *parent_rate)
-{
-	const struct pll_rate *pll_rate = find_rate(rate);
-	return pll_rate->rate;
-}
-
-static int hdmi_pll_set_rate(struct clk_hw *hw, unsigned long rate,
-		unsigned long parent_rate)
-{
-	struct hdmi_phy_8960 *phy_8960 = clk_to_phy(hw);
-	struct hdmi *hdmi = phy_8960->hdmi;
-	const struct pll_rate *pll_rate = find_rate(rate);
-	int i;
-
-	DBG("rate=%lu", rate);
-
-	for (i = 0; pll_rate->conf[i].reg; i++)
-		hdmi_write(hdmi, pll_rate->conf[i].reg, pll_rate->conf[i].val);
-
-	phy_8960->pixclk = rate;
-
-	return 0;
-}
-
-
-static const struct clk_ops hdmi_pll_ops = {
-	.enable = hdmi_pll_enable,
-	.disable = hdmi_pll_disable,
-	.recalc_rate = hdmi_pll_recalc_rate,
-	.round_rate = hdmi_pll_round_rate,
-	.set_rate = hdmi_pll_set_rate,
-};
-
-static const char *hdmi_pll_parents[] = {
-	"pxo",
-};
-
-static struct clk_init_data pll_init = {
-	.name = "hdmi_pll",
-	.ops = &hdmi_pll_ops,
-	.parent_names = hdmi_pll_parents,
-	.num_parents = ARRAY_SIZE(hdmi_pll_parents),
-};
-#endif
-
-/*
- * HDMI Phy:
- */
-
-static void hdmi_phy_8960_destroy(struct hdmi_phy *phy)
-{
-	struct hdmi_phy_8960 *phy_8960 = to_hdmi_phy_8960(phy);
-	kfree(phy_8960);
-}
-
 static void hdmi_phy_8960_powerup(struct hdmi_phy *phy,
-		unsigned long int pixclock)
+				  unsigned long int pixclock)
 {
-	struct hdmi_phy_8960 *phy_8960 = to_hdmi_phy_8960(phy);
-	struct hdmi *hdmi = phy_8960->hdmi;
-
 	DBG("pixclock: %lu", pixclock);
 
-	hdmi_write(hdmi, REG_HDMI_8960_PHY_REG2, 0x00);
-	hdmi_write(hdmi, REG_HDMI_8960_PHY_REG0, 0x1b);
-	hdmi_write(hdmi, REG_HDMI_8960_PHY_REG1, 0xf2);
-	hdmi_write(hdmi, REG_HDMI_8960_PHY_REG4, 0x00);
-	hdmi_write(hdmi, REG_HDMI_8960_PHY_REG5, 0x00);
-	hdmi_write(hdmi, REG_HDMI_8960_PHY_REG6, 0x00);
-	hdmi_write(hdmi, REG_HDMI_8960_PHY_REG7, 0x00);
-	hdmi_write(hdmi, REG_HDMI_8960_PHY_REG8, 0x00);
-	hdmi_write(hdmi, REG_HDMI_8960_PHY_REG9, 0x00);
-	hdmi_write(hdmi, REG_HDMI_8960_PHY_REG10, 0x00);
-	hdmi_write(hdmi, REG_HDMI_8960_PHY_REG11, 0x00);
-	hdmi_write(hdmi, REG_HDMI_8960_PHY_REG3, 0x20);
+	hdmi_phy_write(phy, REG_HDMI_8960_PHY_REG2, 0x00);
+	hdmi_phy_write(phy, REG_HDMI_8960_PHY_REG0, 0x1b);
+	hdmi_phy_write(phy, REG_HDMI_8960_PHY_REG1, 0xf2);
+	hdmi_phy_write(phy, REG_HDMI_8960_PHY_REG4, 0x00);
+	hdmi_phy_write(phy, REG_HDMI_8960_PHY_REG5, 0x00);
+	hdmi_phy_write(phy, REG_HDMI_8960_PHY_REG6, 0x00);
+	hdmi_phy_write(phy, REG_HDMI_8960_PHY_REG7, 0x00);
+	hdmi_phy_write(phy, REG_HDMI_8960_PHY_REG8, 0x00);
+	hdmi_phy_write(phy, REG_HDMI_8960_PHY_REG9, 0x00);
+	hdmi_phy_write(phy, REG_HDMI_8960_PHY_REG10, 0x00);
+	hdmi_phy_write(phy, REG_HDMI_8960_PHY_REG11, 0x00);
+	hdmi_phy_write(phy, REG_HDMI_8960_PHY_REG3, 0x20);
 }
 
 static void hdmi_phy_8960_powerdown(struct hdmi_phy *phy)
 {
-	struct hdmi_phy_8960 *phy_8960 = to_hdmi_phy_8960(phy);
-	struct hdmi *hdmi = phy_8960->hdmi;
-
 	DBG("");
 
-	hdmi_write(hdmi, REG_HDMI_8960_PHY_REG2, 0x7f);
+	hdmi_phy_write(phy, REG_HDMI_8960_PHY_REG2, 0x7f);
 }
 
-static const struct hdmi_phy_funcs hdmi_phy_8960_funcs = {
-		.destroy = hdmi_phy_8960_destroy,
-		.powerup = hdmi_phy_8960_powerup,
-		.powerdown = hdmi_phy_8960_powerdown,
+static const char * const hdmi_phy_8960_reg_names[] = {
+	"core-vdda",
 };
 
-struct hdmi_phy *hdmi_phy_8960_init(struct hdmi *hdmi)
-{
-	struct hdmi_phy_8960 *phy_8960;
-	struct hdmi_phy *phy = NULL;
-	int ret;
-#ifdef CONFIG_COMMON_CLK
-	int i;
-
-	/* sanity check: */
-	for (i = 0; i < (ARRAY_SIZE(freqtbl) - 1); i++)
-		if (WARN_ON(freqtbl[i].rate < freqtbl[i+1].rate))
-			return ERR_PTR(-EINVAL);
-#endif
-
-	phy_8960 = kzalloc(sizeof(*phy_8960), GFP_KERNEL);
-	if (!phy_8960) {
-		ret = -ENOMEM;
-		goto fail;
-	}
-
-	phy = &phy_8960->base;
-
-	phy->funcs = &hdmi_phy_8960_funcs;
-
-	phy_8960->hdmi = hdmi;
-
-#ifdef CONFIG_COMMON_CLK
-	phy_8960->pll_hw.init = &pll_init;
-	phy_8960->pll = devm_clk_register(&hdmi->pdev->dev, &phy_8960->pll_hw);
-	if (IS_ERR(phy_8960->pll)) {
-		ret = PTR_ERR(phy_8960->pll);
-		phy_8960->pll = NULL;
-		goto fail;
-	}
-#endif
-
-	return phy;
+static const char * const hdmi_phy_8960_clk_names[] = {
+	"slave_iface_clk",
+};
 
-fail:
-	if (phy)
-		hdmi_phy_8960_destroy(phy);
-	return ERR_PTR(ret);
-}
+const struct hdmi_phy_cfg msm_hdmi_phy_8960_cfg = {
+	.type = MSM_HDMI_PHY_8960,
+	.powerup = hdmi_phy_8960_powerup,
+	.powerdown = hdmi_phy_8960_powerdown,
+	.reg_names = hdmi_phy_8960_reg_names,
+	.num_regs = ARRAY_SIZE(hdmi_phy_8960_reg_names),
+	.clk_names = hdmi_phy_8960_clk_names,
+	.num_clks = ARRAY_SIZE(hdmi_phy_8960_clk_names),
+};
diff --git a/drivers/gpu/drm/msm/hdmi/hdmi_phy_8996.c b/drivers/gpu/drm/msm/hdmi/hdmi_phy_8996.c
new file mode 100644
index 000000000000..aa94a553794f
--- /dev/null
+++ b/drivers/gpu/drm/msm/hdmi/hdmi_phy_8996.c
@@ -0,0 +1,766 @@
+/*
+ * Copyright (c) 2016, The Linux Foundation. All rights reserved.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 and
+ * only version 2 as published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ */
+
+#include <linux/clk-provider.h>
+
+#include "hdmi.h"
+
+#define HDMI_VCO_MAX_FREQ			12000000000UL
+#define HDMI_VCO_MIN_FREQ			8000000000UL
+
+#define HDMI_PCLK_MAX_FREQ			600000000
+#define HDMI_PCLK_MIN_FREQ			25000000
+
+#define HDMI_HIGH_FREQ_BIT_CLK_THRESHOLD	3400000000UL
+#define HDMI_DIG_FREQ_BIT_CLK_THRESHOLD		1500000000UL
+#define HDMI_MID_FREQ_BIT_CLK_THRESHOLD		750000000UL
+#define HDMI_CORECLK_DIV			5
+#define HDMI_DEFAULT_REF_CLOCK			19200000
+#define HDMI_PLL_CMP_CNT			1024
+
+#define HDMI_PLL_POLL_MAX_READS			100
+#define HDMI_PLL_POLL_TIMEOUT_US		150
+
+#define HDMI_NUM_TX_CHANNEL			4
+
+struct hdmi_pll_8996 {
+	struct platform_device *pdev;
+	struct clk_hw clk_hw;
+
+	/* pll mmio base */
+	void __iomem *mmio_qserdes_com;
+	/* tx channel base */
+	void __iomem *mmio_qserdes_tx[HDMI_NUM_TX_CHANNEL];
+};
+
+#define hw_clk_to_pll(x) container_of(x, struct hdmi_pll_8996, clk_hw)
+
+struct hdmi_8996_phy_pll_reg_cfg {
+	u32 tx_lx_lane_mode[HDMI_NUM_TX_CHANNEL];
+	u32 tx_lx_tx_band[HDMI_NUM_TX_CHANNEL];
+	u32 com_svs_mode_clk_sel;
+	u32 com_hsclk_sel;
+	u32 com_pll_cctrl_mode0;
+	u32 com_pll_rctrl_mode0;
+	u32 com_cp_ctrl_mode0;
+	u32 com_dec_start_mode0;
+	u32 com_div_frac_start1_mode0;
+	u32 com_div_frac_start2_mode0;
+	u32 com_div_frac_start3_mode0;
+	u32 com_integloop_gain0_mode0;
+	u32 com_integloop_gain1_mode0;
+	u32 com_lock_cmp_en;
+	u32 com_lock_cmp1_mode0;
+	u32 com_lock_cmp2_mode0;
+	u32 com_lock_cmp3_mode0;
+	u32 com_core_clk_en;
+	u32 com_coreclk_div;
+	u32 com_vco_tune_ctrl;
+
+	u32 tx_lx_tx_drv_lvl[HDMI_NUM_TX_CHANNEL];
+	u32 tx_lx_tx_emp_post1_lvl[HDMI_NUM_TX_CHANNEL];
+	u32 tx_lx_vmode_ctrl1[HDMI_NUM_TX_CHANNEL];
+	u32 tx_lx_vmode_ctrl2[HDMI_NUM_TX_CHANNEL];
+	u32 tx_lx_res_code_lane_tx[HDMI_NUM_TX_CHANNEL];
+	u32 tx_lx_hp_pd_enables[HDMI_NUM_TX_CHANNEL];
+
+	u32 phy_mode;
+};
+
+struct hdmi_8996_post_divider {
+	u64 vco_freq;
+	int hsclk_divsel;
+	int vco_ratio;
+	int tx_band_sel;
+	int half_rate_mode;
+};
+
+static inline struct hdmi_phy *pll_get_phy(struct hdmi_pll_8996 *pll)
+{
+	return platform_get_drvdata(pll->pdev);
+}
+
+static inline void hdmi_pll_write(struct hdmi_pll_8996 *pll, int offset,
+				  u32 data)
+{
+	msm_writel(data, pll->mmio_qserdes_com + offset);
+}
+
+static inline u32 hdmi_pll_read(struct hdmi_pll_8996 *pll, int offset)
+{
+	return msm_readl(pll->mmio_qserdes_com + offset);
+}
+
+static inline void hdmi_tx_chan_write(struct hdmi_pll_8996 *pll, int channel,
+				      int offset, int data)
+{
+	 msm_writel(data, pll->mmio_qserdes_tx[channel] + offset);
+}
+
+static inline u32 pll_get_cpctrl(u64 frac_start, unsigned long ref_clk,
+				 bool gen_ssc)
+{
+	if ((frac_start != 0) || gen_ssc)
+		return (11000000 / (ref_clk / 20));
+
+	return 0x23;
+}
+
+static inline u32 pll_get_rctrl(u64 frac_start, bool gen_ssc)
+{
+	if ((frac_start != 0) || gen_ssc)
+		return 0x16;
+
+	return 0x10;
+}
+
+static inline u32 pll_get_cctrl(u64 frac_start, bool gen_ssc)
+{
+	if ((frac_start != 0) || gen_ssc)
+		return 0x28;
+
+	return 0x1;
+}
+
+static inline u32 pll_get_integloop_gain(u64 frac_start, u64 bclk, u32 ref_clk,
+					 bool gen_ssc)
+{
+	int digclk_divsel = bclk >= HDMI_DIG_FREQ_BIT_CLK_THRESHOLD ? 1 : 2;
+	u64 base;
+
+	if ((frac_start != 0) || gen_ssc)
+		base = (64 * ref_clk) / HDMI_DEFAULT_REF_CLOCK;
+	else
+		base = (1022 * ref_clk) / 100;
+
+	base <<= digclk_divsel;
+
+	return (base <= 2046 ? base : 2046);
+}
+
+static inline u32 pll_get_pll_cmp(u64 fdata, unsigned long ref_clk)
+{
+	u64 dividend = HDMI_PLL_CMP_CNT * fdata;
+	u32 divisor = ref_clk * 10;
+	u32 rem;
+
+	rem = do_div(dividend, divisor);
+	if (rem > (divisor >> 1))
+		dividend++;
+
+	return dividend - 1;
+}
+
+static inline u64 pll_cmp_to_fdata(u32 pll_cmp, unsigned long ref_clk)
+{
+	u64 fdata = ((u64)pll_cmp) * ref_clk * 10;
+
+	do_div(fdata, HDMI_PLL_CMP_CNT);
+
+	return fdata;
+}
+
+static int pll_get_post_div(struct hdmi_8996_post_divider *pd, u64 bclk)
+{
+	int ratio[] = { 2, 3, 4, 5, 6, 9, 10, 12, 14, 15, 20, 21, 25, 28, 35 };
+	int hs_divsel[] = { 0, 4, 8, 12, 1, 5, 2, 9, 3, 13, 10, 7, 14, 11, 15 };
+	int tx_band_sel[] = { 0, 1, 2, 3 };
+	u64 vco_freq[60];
+	u64 vco, vco_optimal;
+	int half_rate_mode = 0;
+	int vco_optimal_index, vco_freq_index;
+	int i, j;
+
+retry:
+	vco_optimal = HDMI_VCO_MAX_FREQ;
+	vco_optimal_index = -1;
+	vco_freq_index = 0;
+	for (i = 0; i < 15; i++) {
+		for (j = 0; j < 4; j++) {
+			u32 ratio_mult = ratio[i] << tx_band_sel[j];
+
+			vco = bclk >> half_rate_mode;
+			vco *= ratio_mult;
+			vco_freq[vco_freq_index++] = vco;
+		}
+	}
+
+	for (i = 0; i < 60; i++) {
+		u64 vco_tmp = vco_freq[i];
+
+		if ((vco_tmp >= HDMI_VCO_MIN_FREQ) &&
+		    (vco_tmp <= vco_optimal)) {
+			vco_optimal = vco_tmp;
+			vco_optimal_index = i;
+		}
+	}
+
+	if (vco_optimal_index == -1) {
+		if (!half_rate_mode) {
+			half_rate_mode = 1;
+			goto retry;
+		}
+	} else {
+		pd->vco_freq = vco_optimal;
+		pd->tx_band_sel = tx_band_sel[vco_optimal_index % 4];
+		pd->vco_ratio = ratio[vco_optimal_index / 4];
+		pd->hsclk_divsel = hs_divsel[vco_optimal_index / 4];
+
+		return 0;
+	}
+
+	return -EINVAL;
+}
+
+static int pll_calculate(unsigned long pix_clk, unsigned long ref_clk,
+			 struct hdmi_8996_phy_pll_reg_cfg *cfg)
+{
+	struct hdmi_8996_post_divider pd;
+	u64 bclk;
+	u64 tmds_clk;
+	u64 dec_start;
+	u64 frac_start;
+	u64 fdata;
+	u32 pll_divisor;
+	u32 rem;
+	u32 cpctrl;
+	u32 rctrl;
+	u32 cctrl;
+	u32 integloop_gain;
+	u32 pll_cmp;
+	int i, ret;
+
+	/* bit clk = 10 * pix_clk */
+	bclk = ((u64)pix_clk) * 10;
+
+	if (bclk > HDMI_HIGH_FREQ_BIT_CLK_THRESHOLD)
+		tmds_clk = pix_clk >> 2;
+	else
+		tmds_clk = pix_clk;
+
+	ret = pll_get_post_div(&pd, bclk);
+	if (ret)
+		return ret;
+
+	dec_start = pd.vco_freq;
+	pll_divisor = 4 * ref_clk;
+	do_div(dec_start, pll_divisor);
+
+	frac_start = pd.vco_freq * (1 << 20);
+
+	rem = do_div(frac_start, pll_divisor);
+	frac_start -= dec_start * (1 << 20);
+	if (rem > (pll_divisor >> 1))
+		frac_start++;
+
+	cpctrl = pll_get_cpctrl(frac_start, ref_clk, false);
+	rctrl = pll_get_rctrl(frac_start, false);
+	cctrl = pll_get_cctrl(frac_start, false);
+	integloop_gain = pll_get_integloop_gain(frac_start, bclk,
+						ref_clk, false);
+
+	fdata = pd.vco_freq;
+	do_div(fdata, pd.vco_ratio);
+
+	pll_cmp = pll_get_pll_cmp(fdata, ref_clk);
+
+	DBG("VCO freq: %llu", pd.vco_freq);
+	DBG("fdata: %llu", fdata);
+	DBG("pix_clk: %lu", pix_clk);
+	DBG("tmds clk: %llu", tmds_clk);
+	DBG("HSCLK_SEL: %d", pd.hsclk_divsel);
+	DBG("DEC_START: %llu", dec_start);
+	DBG("DIV_FRAC_START: %llu", frac_start);
+	DBG("PLL_CPCTRL: %u", cpctrl);
+	DBG("PLL_RCTRL: %u", rctrl);
+	DBG("PLL_CCTRL: %u", cctrl);
+	DBG("INTEGLOOP_GAIN: %u", integloop_gain);
+	DBG("TX_BAND: %d", pd.tx_band_sel);
+	DBG("PLL_CMP: %u", pll_cmp);
+
+	/* Convert these values to register specific values */
+	if (bclk > HDMI_DIG_FREQ_BIT_CLK_THRESHOLD)
+		cfg->com_svs_mode_clk_sel = 1;
+	else
+		cfg->com_svs_mode_clk_sel = 2;
+
+	cfg->com_hsclk_sel = (0x20 | pd.hsclk_divsel);
+	cfg->com_pll_cctrl_mode0 = cctrl;
+	cfg->com_pll_rctrl_mode0 = rctrl;
+	cfg->com_cp_ctrl_mode0 = cpctrl;
+	cfg->com_dec_start_mode0 = dec_start;
+	cfg->com_div_frac_start1_mode0 = (frac_start & 0xff);
+	cfg->com_div_frac_start2_mode0 = ((frac_start & 0xff00) >> 8);
+	cfg->com_div_frac_start3_mode0 = ((frac_start & 0xf0000) >> 16);
+	cfg->com_integloop_gain0_mode0 = (integloop_gain & 0xff);
+	cfg->com_integloop_gain1_mode0 = ((integloop_gain & 0xf00) >> 8);
+	cfg->com_lock_cmp1_mode0 = (pll_cmp & 0xff);
+	cfg->com_lock_cmp2_mode0 = ((pll_cmp & 0xff00) >> 8);
+	cfg->com_lock_cmp3_mode0 = ((pll_cmp & 0x30000) >> 16);
+	cfg->com_lock_cmp_en = 0x0;
+	cfg->com_core_clk_en = 0x2c;
+	cfg->com_coreclk_div = HDMI_CORECLK_DIV;
+	cfg->phy_mode = (bclk > HDMI_HIGH_FREQ_BIT_CLK_THRESHOLD) ? 0x10 : 0x0;
+	cfg->com_vco_tune_ctrl = 0x0;
+
+	cfg->tx_lx_lane_mode[0] =
+		cfg->tx_lx_lane_mode[2] = 0x43;
+
+	cfg->tx_lx_hp_pd_enables[0] =
+		cfg->tx_lx_hp_pd_enables[1] =
+		cfg->tx_lx_hp_pd_enables[2] = 0x0c;
+	cfg->tx_lx_hp_pd_enables[3] = 0x3;
+
+	for (i = 0; i < HDMI_NUM_TX_CHANNEL; i++)
+		cfg->tx_lx_tx_band[i] = pd.tx_band_sel + 4;
+
+	if (bclk > HDMI_HIGH_FREQ_BIT_CLK_THRESHOLD) {
+		cfg->tx_lx_tx_drv_lvl[0] =
+			cfg->tx_lx_tx_drv_lvl[1] =
+			cfg->tx_lx_tx_drv_lvl[2] = 0x25;
+		cfg->tx_lx_tx_drv_lvl[3] = 0x22;
+
+		cfg->tx_lx_tx_emp_post1_lvl[0] =
+			cfg->tx_lx_tx_emp_post1_lvl[1] =
+			cfg->tx_lx_tx_emp_post1_lvl[2] = 0x23;
+		cfg->tx_lx_tx_emp_post1_lvl[3] = 0x27;
+
+		cfg->tx_lx_vmode_ctrl1[0] =
+			cfg->tx_lx_vmode_ctrl1[1] =
+			cfg->tx_lx_vmode_ctrl1[2] =
+			cfg->tx_lx_vmode_ctrl1[3] = 0x00;
+
+		cfg->tx_lx_vmode_ctrl2[0] =
+			cfg->tx_lx_vmode_ctrl2[1] =
+			cfg->tx_lx_vmode_ctrl2[2] = 0x0D;
+
+		cfg->tx_lx_vmode_ctrl2[3] = 0x00;
+	} else if (bclk > HDMI_MID_FREQ_BIT_CLK_THRESHOLD) {
+		for (i = 0; i < HDMI_NUM_TX_CHANNEL; i++) {
+			cfg->tx_lx_tx_drv_lvl[i] = 0x25;
+			cfg->tx_lx_tx_emp_post1_lvl[i] = 0x23;
+			cfg->tx_lx_vmode_ctrl1[i] = 0x00;
+		}
+
+		cfg->tx_lx_vmode_ctrl2[0] =
+			cfg->tx_lx_vmode_ctrl2[1] =
+			cfg->tx_lx_vmode_ctrl2[2] = 0x0D;
+		cfg->tx_lx_vmode_ctrl2[3] = 0x00;
+	} else {
+		for (i = 0; i < HDMI_NUM_TX_CHANNEL; i++) {
+			cfg->tx_lx_tx_drv_lvl[i] = 0x20;
+			cfg->tx_lx_tx_emp_post1_lvl[i] = 0x20;
+			cfg->tx_lx_vmode_ctrl1[i] = 0x00;
+			cfg->tx_lx_vmode_ctrl2[i] = 0x0E;
+		}
+	}
+
+	DBG("com_svs_mode_clk_sel = 0x%x", cfg->com_svs_mode_clk_sel);
+	DBG("com_hsclk_sel = 0x%x", cfg->com_hsclk_sel);
+	DBG("com_lock_cmp_en = 0x%x", cfg->com_lock_cmp_en);
+	DBG("com_pll_cctrl_mode0 = 0x%x", cfg->com_pll_cctrl_mode0);
+	DBG("com_pll_rctrl_mode0 = 0x%x", cfg->com_pll_rctrl_mode0);
+	DBG("com_cp_ctrl_mode0 = 0x%x", cfg->com_cp_ctrl_mode0);
+	DBG("com_dec_start_mode0 = 0x%x", cfg->com_dec_start_mode0);
+	DBG("com_div_frac_start1_mode0 = 0x%x", cfg->com_div_frac_start1_mode0);
+	DBG("com_div_frac_start2_mode0 = 0x%x", cfg->com_div_frac_start2_mode0);
+	DBG("com_div_frac_start3_mode0 = 0x%x", cfg->com_div_frac_start3_mode0);
+	DBG("com_integloop_gain0_mode0 = 0x%x", cfg->com_integloop_gain0_mode0);
+	DBG("com_integloop_gain1_mode0 = 0x%x", cfg->com_integloop_gain1_mode0);
+	DBG("com_lock_cmp1_mode0 = 0x%x", cfg->com_lock_cmp1_mode0);
+	DBG("com_lock_cmp2_mode0 = 0x%x", cfg->com_lock_cmp2_mode0);
+	DBG("com_lock_cmp3_mode0 = 0x%x", cfg->com_lock_cmp3_mode0);
+	DBG("com_core_clk_en = 0x%x", cfg->com_core_clk_en);
+	DBG("com_coreclk_div = 0x%x", cfg->com_coreclk_div);
+	DBG("phy_mode = 0x%x", cfg->phy_mode);
+
+	DBG("tx_l0_lane_mode = 0x%x", cfg->tx_lx_lane_mode[0]);
+	DBG("tx_l2_lane_mode = 0x%x", cfg->tx_lx_lane_mode[2]);
+
+	for (i = 0; i < HDMI_NUM_TX_CHANNEL; i++) {
+		DBG("tx_l%d_tx_band = 0x%x", i, cfg->tx_lx_tx_band[i]);
+		DBG("tx_l%d_tx_drv_lvl = 0x%x", i, cfg->tx_lx_tx_drv_lvl[i]);
+		DBG("tx_l%d_tx_emp_post1_lvl = 0x%x", i,
+		    cfg->tx_lx_tx_emp_post1_lvl[i]);
+		DBG("tx_l%d_vmode_ctrl1 = 0x%x", i, cfg->tx_lx_vmode_ctrl1[i]);
+		DBG("tx_l%d_vmode_ctrl2 = 0x%x", i, cfg->tx_lx_vmode_ctrl2[i]);
+	}
+
+	return 0;
+}
+
+static int hdmi_8996_pll_set_clk_rate(struct clk_hw *hw, unsigned long rate,
+				      unsigned long parent_rate)
+{
+	struct hdmi_pll_8996 *pll = hw_clk_to_pll(hw);
+	struct hdmi_phy *phy = pll_get_phy(pll);
+	struct hdmi_8996_phy_pll_reg_cfg cfg;
+	int i, ret;
+
+	memset(&cfg, 0x00, sizeof(cfg));
+
+	ret = pll_calculate(rate, parent_rate, &cfg);
+	if (ret) {
+		DRM_ERROR("PLL calculation failed\n");
+		return ret;
+	}
+
+	/* Initially shut down PHY */
+	DBG("Disabling PHY");
+	hdmi_phy_write(phy, REG_HDMI_8996_PHY_PD_CTL, 0x0);
+	udelay(500);
+
+	/* Power up sequence */
+	hdmi_pll_write(pll, REG_HDMI_PHY_QSERDES_COM_BG_CTRL, 0x04);
+
+	hdmi_phy_write(phy, REG_HDMI_8996_PHY_PD_CTL, 0x1);
+	hdmi_pll_write(pll, REG_HDMI_PHY_QSERDES_COM_RESETSM_CNTRL, 0x20);
+	hdmi_phy_write(phy, REG_HDMI_8996_PHY_TX0_TX1_LANE_CTL, 0x0F);
+	hdmi_phy_write(phy, REG_HDMI_8996_PHY_TX2_TX3_LANE_CTL, 0x0F);
+
+	for (i = 0; i < HDMI_NUM_TX_CHANNEL; i++) {
+		hdmi_tx_chan_write(pll, i,
+				   REG_HDMI_PHY_QSERDES_TX_LX_CLKBUF_ENABLE,
+				   0x03);
+		hdmi_tx_chan_write(pll, i,
+				   REG_HDMI_PHY_QSERDES_TX_LX_TX_BAND,
+				   cfg.tx_lx_tx_band[i]);
+		hdmi_tx_chan_write(pll, i,
+				   REG_HDMI_PHY_QSERDES_TX_LX_RESET_TSYNC_EN,
+				   0x03);
+	}
+
+	hdmi_tx_chan_write(pll, 0, REG_HDMI_PHY_QSERDES_TX_LX_LANE_MODE,
+			   cfg.tx_lx_lane_mode[0]);
+	hdmi_tx_chan_write(pll, 2, REG_HDMI_PHY_QSERDES_TX_LX_LANE_MODE,
+			   cfg.tx_lx_lane_mode[2]);
+
+	hdmi_pll_write(pll, REG_HDMI_PHY_QSERDES_COM_SYSCLK_BUF_ENABLE, 0x1E);
+	hdmi_pll_write(pll, REG_HDMI_PHY_QSERDES_COM_BIAS_EN_CLKBUFLR_EN, 0x07);
+	hdmi_pll_write(pll, REG_HDMI_PHY_QSERDES_COM_SYSCLK_EN_SEL, 0x37);
+	hdmi_pll_write(pll, REG_HDMI_PHY_QSERDES_COM_SYS_CLK_CTRL, 0x02);
+	hdmi_pll_write(pll, REG_HDMI_PHY_QSERDES_COM_CLK_ENABLE1, 0x0E);
+
+	/* Bypass VCO calibration */
+	hdmi_pll_write(pll, REG_HDMI_PHY_QSERDES_COM_SVS_MODE_CLK_SEL,
+		       cfg.com_svs_mode_clk_sel);
+
+	hdmi_pll_write(pll, REG_HDMI_PHY_QSERDES_COM_BG_TRIM, 0x0F);
+	hdmi_pll_write(pll, REG_HDMI_PHY_QSERDES_COM_PLL_IVCO, 0x0F);
+	hdmi_pll_write(pll, REG_HDMI_PHY_QSERDES_COM_VCO_TUNE_CTRL,
+		       cfg.com_vco_tune_ctrl);
+
+	hdmi_pll_write(pll, REG_HDMI_PHY_QSERDES_COM_BG_CTRL, 0x06);
+
+	hdmi_pll_write(pll, REG_HDMI_PHY_QSERDES_COM_CLK_SELECT, 0x30);
+	hdmi_pll_write(pll, REG_HDMI_PHY_QSERDES_COM_HSCLK_SEL,
+		       cfg.com_hsclk_sel);
+	hdmi_pll_write(pll, REG_HDMI_PHY_QSERDES_COM_LOCK_CMP_EN,
+		       cfg.com_lock_cmp_en);
+
+	hdmi_pll_write(pll, REG_HDMI_PHY_QSERDES_COM_PLL_CCTRL_MODE0,
+		       cfg.com_pll_cctrl_mode0);
+	hdmi_pll_write(pll, REG_HDMI_PHY_QSERDES_COM_PLL_RCTRL_MODE0,
+		       cfg.com_pll_rctrl_mode0);
+	hdmi_pll_write(pll, REG_HDMI_PHY_QSERDES_COM_CP_CTRL_MODE0,
+		       cfg.com_cp_ctrl_mode0);
+	hdmi_pll_write(pll, REG_HDMI_PHY_QSERDES_COM_DEC_START_MODE0,
+		       cfg.com_dec_start_mode0);
+	hdmi_pll_write(pll, REG_HDMI_PHY_QSERDES_COM_DIV_FRAC_START1_MODE0,
+		       cfg.com_div_frac_start1_mode0);
+	hdmi_pll_write(pll, REG_HDMI_PHY_QSERDES_COM_DIV_FRAC_START2_MODE0,
+		       cfg.com_div_frac_start2_mode0);
+	hdmi_pll_write(pll, REG_HDMI_PHY_QSERDES_COM_DIV_FRAC_START3_MODE0,
+		       cfg.com_div_frac_start3_mode0);
+
+	hdmi_pll_write(pll, REG_HDMI_PHY_QSERDES_COM_INTEGLOOP_GAIN0_MODE0,
+		       cfg.com_integloop_gain0_mode0);
+	hdmi_pll_write(pll, REG_HDMI_PHY_QSERDES_COM_INTEGLOOP_GAIN1_MODE0,
+		       cfg.com_integloop_gain1_mode0);
+
+	hdmi_pll_write(pll, REG_HDMI_PHY_QSERDES_COM_LOCK_CMP1_MODE0,
+		       cfg.com_lock_cmp1_mode0);
+	hdmi_pll_write(pll, REG_HDMI_PHY_QSERDES_COM_LOCK_CMP2_MODE0,
+		       cfg.com_lock_cmp2_mode0);
+	hdmi_pll_write(pll, REG_HDMI_PHY_QSERDES_COM_LOCK_CMP3_MODE0,
+		       cfg.com_lock_cmp3_mode0);
+
+	hdmi_pll_write(pll, REG_HDMI_PHY_QSERDES_COM_VCO_TUNE_MAP, 0x00);
+	hdmi_pll_write(pll, REG_HDMI_PHY_QSERDES_COM_CORE_CLK_EN,
+		       cfg.com_core_clk_en);
+	hdmi_pll_write(pll, REG_HDMI_PHY_QSERDES_COM_CORECLK_DIV,
+		       cfg.com_coreclk_div);
+	hdmi_pll_write(pll, REG_HDMI_PHY_QSERDES_COM_CMN_CONFIG, 0x02);
+
+	hdmi_pll_write(pll, REG_HDMI_PHY_QSERDES_COM_RESCODE_DIV_NUM, 0x15);
+
+	/* TX lanes setup (TX 0/1/2/3) */
+	for (i = 0; i < HDMI_NUM_TX_CHANNEL; i++) {
+		hdmi_tx_chan_write(pll, i,
+				   REG_HDMI_PHY_QSERDES_TX_LX_TX_DRV_LVL,
+				   cfg.tx_lx_tx_drv_lvl[i]);
+		hdmi_tx_chan_write(pll, i,
+				   REG_HDMI_PHY_QSERDES_TX_LX_TX_EMP_POST1_LVL,
+				   cfg.tx_lx_tx_emp_post1_lvl[i]);
+		hdmi_tx_chan_write(pll, i,
+				   REG_HDMI_PHY_QSERDES_TX_LX_VMODE_CTRL1,
+				   cfg.tx_lx_vmode_ctrl1[i]);
+		hdmi_tx_chan_write(pll, i,
+				   REG_HDMI_PHY_QSERDES_TX_LX_VMODE_CTRL2,
+				   cfg.tx_lx_vmode_ctrl2[i]);
+		hdmi_tx_chan_write(pll, i,
+				   REG_HDMI_PHY_QSERDES_TX_LX_TX_DRV_LVL_OFFSET,
+				   0x00);
+		hdmi_tx_chan_write(pll, i,
+			REG_HDMI_PHY_QSERDES_TX_LX_RES_CODE_LANE_OFFSET,
+			0x00);
+		hdmi_tx_chan_write(pll, i,
+			REG_HDMI_PHY_QSERDES_TX_LX_TRAN_DRVR_EMP_EN,
+			0x03);
+		hdmi_tx_chan_write(pll, i,
+			REG_HDMI_PHY_QSERDES_TX_LX_PARRATE_REC_DETECT_IDLE_EN,
+			0x40);
+		hdmi_tx_chan_write(pll, i,
+				   REG_HDMI_PHY_QSERDES_TX_LX_HP_PD_ENABLES,
+				   cfg.tx_lx_hp_pd_enables[i]);
+	}
+
+	hdmi_phy_write(phy, REG_HDMI_8996_PHY_MODE, cfg.phy_mode);
+	hdmi_phy_write(phy, REG_HDMI_8996_PHY_PD_CTL, 0x1F);
+
+	/*
+	 * Ensure that vco configuration gets flushed to hardware before
+	 * enabling the PLL
+	 */
+	wmb();
+
+	return 0;
+}
+
+static int hdmi_8996_phy_ready_status(struct hdmi_phy *phy)
+{
+	u32 nb_tries = HDMI_PLL_POLL_MAX_READS;
+	unsigned long timeout = HDMI_PLL_POLL_TIMEOUT_US;
+	u32 status;
+	int phy_ready = 0;
+
+	DBG("Waiting for PHY ready");
+
+	while (nb_tries--) {
+		status = hdmi_phy_read(phy, REG_HDMI_8996_PHY_STATUS);
+		phy_ready = status & BIT(0);
+
+		if (phy_ready)
+			break;
+
+		udelay(timeout);
+	}
+
+	DBG("PHY is %sready", phy_ready ? "" : "*not* ");
+
+	return phy_ready;
+}
+
+static int hdmi_8996_pll_lock_status(struct hdmi_pll_8996 *pll)
+{
+	u32 status;
+	int nb_tries = HDMI_PLL_POLL_MAX_READS;
+	unsigned long timeout = HDMI_PLL_POLL_TIMEOUT_US;
+	int pll_locked = 0;
+
+	DBG("Waiting for PLL lock");
+
+	while (nb_tries--) {
+		status = hdmi_pll_read(pll,
+				       REG_HDMI_PHY_QSERDES_COM_C_READY_STATUS);
+		pll_locked = status & BIT(0);
+
+		if (pll_locked)
+			break;
+
+		udelay(timeout);
+	}
+
+	DBG("HDMI PLL is %slocked", pll_locked ? "" : "*not* ");
+
+	return pll_locked;
+}
+
+static int hdmi_8996_pll_prepare(struct clk_hw *hw)
+{
+	struct hdmi_pll_8996 *pll = hw_clk_to_pll(hw);
+	struct hdmi_phy *phy = pll_get_phy(pll);
+	int i, ret = 0;
+
+	hdmi_phy_write(phy, REG_HDMI_8996_PHY_CFG, 0x1);
+	udelay(100);
+
+	hdmi_phy_write(phy, REG_HDMI_8996_PHY_CFG, 0x19);
+	udelay(100);
+
+	ret = hdmi_8996_pll_lock_status(pll);
+	if (!ret)
+		return ret;
+
+	for (i = 0; i < HDMI_NUM_TX_CHANNEL; i++)
+		hdmi_tx_chan_write(pll, i,
+			REG_HDMI_PHY_QSERDES_TX_LX_HIGHZ_TRANSCEIVEREN_BIAS_DRVR_EN,
+			0x6F);
+
+	/* Disable SSC */
+	hdmi_pll_write(pll, REG_HDMI_PHY_QSERDES_COM_SSC_PER1, 0x0);
+	hdmi_pll_write(pll, REG_HDMI_PHY_QSERDES_COM_SSC_PER2, 0x0);
+	hdmi_pll_write(pll, REG_HDMI_PHY_QSERDES_COM_SSC_STEP_SIZE1, 0x0);
+	hdmi_pll_write(pll, REG_HDMI_PHY_QSERDES_COM_SSC_STEP_SIZE2, 0x0);
+	hdmi_pll_write(pll, REG_HDMI_PHY_QSERDES_COM_SSC_EN_CENTER, 0x2);
+
+	ret = hdmi_8996_phy_ready_status(phy);
+	if (!ret)
+		return ret;
+
+	/* Restart the retiming buffer */
+	hdmi_phy_write(phy, REG_HDMI_8996_PHY_CFG, 0x18);
+	udelay(1);
+	hdmi_phy_write(phy, REG_HDMI_8996_PHY_CFG, 0x19);
+
+	return 0;
+}
+
+static long hdmi_8996_pll_round_rate(struct clk_hw *hw,
+				     unsigned long rate,
+				     unsigned long *parent_rate)
+{
+	if (rate < HDMI_PCLK_MIN_FREQ)
+		return HDMI_PCLK_MIN_FREQ;
+	else if (rate > HDMI_PCLK_MAX_FREQ)
+		return HDMI_PCLK_MAX_FREQ;
+	else
+		return rate;
+}
+
+static unsigned long hdmi_8996_pll_recalc_rate(struct clk_hw *hw,
+					       unsigned long parent_rate)
+{
+	struct hdmi_pll_8996 *pll = hw_clk_to_pll(hw);
+	u64 fdata;
+	u32 cmp1, cmp2, cmp3, pll_cmp;
+
+	cmp1 = hdmi_pll_read(pll, REG_HDMI_PHY_QSERDES_COM_LOCK_CMP1_MODE0);
+	cmp2 = hdmi_pll_read(pll, REG_HDMI_PHY_QSERDES_COM_LOCK_CMP2_MODE0);
+	cmp3 = hdmi_pll_read(pll, REG_HDMI_PHY_QSERDES_COM_LOCK_CMP3_MODE0);
+
+	pll_cmp = cmp1 | (cmp2 << 8) | (cmp3 << 16);
+
+	fdata = pll_cmp_to_fdata(pll_cmp + 1, parent_rate);
+
+	do_div(fdata, 10);
+
+	return fdata;
+}
+
+static void hdmi_8996_pll_unprepare(struct clk_hw *hw)
+{
+}
+
+static int hdmi_8996_pll_is_enabled(struct clk_hw *hw)
+{
+	struct hdmi_pll_8996 *pll = hw_clk_to_pll(hw);
+	u32 status;
+	int pll_locked;
+
+	status = hdmi_pll_read(pll, REG_HDMI_PHY_QSERDES_COM_C_READY_STATUS);
+	pll_locked = status & BIT(0);
+
+	return pll_locked;
+}
+
+static struct clk_ops hdmi_8996_pll_ops = {
+	.set_rate = hdmi_8996_pll_set_clk_rate,
+	.round_rate = hdmi_8996_pll_round_rate,
+	.recalc_rate = hdmi_8996_pll_recalc_rate,
+	.prepare = hdmi_8996_pll_prepare,
+	.unprepare = hdmi_8996_pll_unprepare,
+	.is_enabled = hdmi_8996_pll_is_enabled,
+};
+
+static const char * const hdmi_pll_parents[] = {
+	"xo",
+};
+
+static struct clk_init_data pll_init = {
+	.name = "hdmipll",
+	.ops = &hdmi_8996_pll_ops,
+	.parent_names = hdmi_pll_parents,
+	.num_parents = ARRAY_SIZE(hdmi_pll_parents),
+};
+
+int msm_hdmi_pll_8996_init(struct platform_device *pdev)
+{
+	struct device *dev = &pdev->dev;
+	struct hdmi_pll_8996 *pll;
+	struct clk *clk;
+	int i;
+
+	pll = devm_kzalloc(dev, sizeof(*pll), GFP_KERNEL);
+	if (!pll)
+		return -ENOMEM;
+
+	pll->pdev = pdev;
+
+	pll->mmio_qserdes_com = msm_ioremap(pdev, "hdmi_pll", "HDMI_PLL");
+	if (IS_ERR(pll->mmio_qserdes_com)) {
+		dev_err(dev, "failed to map pll base\n");
+		return -ENOMEM;
+	}
+
+	for (i = 0; i < HDMI_NUM_TX_CHANNEL; i++) {
+		char name[32], label[32];
+
+		snprintf(name, sizeof(name), "hdmi_tx_l%d", i);
+		snprintf(label, sizeof(label), "HDMI_TX_L%d", i);
+
+		pll->mmio_qserdes_tx[i] = msm_ioremap(pdev, name, label);
+		if (IS_ERR(pll->mmio_qserdes_tx[i])) {
+			dev_err(dev, "failed to map pll base\n");
+			return -ENOMEM;
+		}
+	}
+	pll->clk_hw.init = &pll_init;
+
+	clk = devm_clk_register(dev, &pll->clk_hw);
+	if (IS_ERR(clk)) {
+		dev_err(dev, "failed to register pll clock\n");
+		return -EINVAL;
+	}
+
+	return 0;
+}
+
+static const char * const hdmi_phy_8996_reg_names[] = {
+	"vddio",
+	"vcca",
+};
+
+static const char * const hdmi_phy_8996_clk_names[] = {
+	"mmagic_iface_clk",
+	"iface_clk",
+	"ref_clk",
+};
+
+const struct hdmi_phy_cfg msm_hdmi_phy_8996_cfg = {
+	.type = MSM_HDMI_PHY_8996,
+	.reg_names = hdmi_phy_8996_reg_names,
+	.num_regs = ARRAY_SIZE(hdmi_phy_8996_reg_names),
+	.clk_names = hdmi_phy_8996_clk_names,
+	.num_clks = ARRAY_SIZE(hdmi_phy_8996_clk_names),
+};
diff --git a/drivers/gpu/drm/msm/hdmi/hdmi_phy_8x60.c b/drivers/gpu/drm/msm/hdmi/hdmi_phy_8x60.c
index cb01421ae1e4..a68eea4153fc 100644
--- a/drivers/gpu/drm/msm/hdmi/hdmi_phy_8x60.c
+++ b/drivers/gpu/drm/msm/hdmi/hdmi_phy_8x60.c
@@ -17,166 +17,122 @@
 
 #include "hdmi.h"
 
-struct hdmi_phy_8x60 {
-	struct hdmi_phy base;
-	struct hdmi *hdmi;
-};
-#define to_hdmi_phy_8x60(x) container_of(x, struct hdmi_phy_8x60, base)
-
-static void hdmi_phy_8x60_destroy(struct hdmi_phy *phy)
-{
-	struct hdmi_phy_8x60 *phy_8x60 = to_hdmi_phy_8x60(phy);
-	kfree(phy_8x60);
-}
-
 static void hdmi_phy_8x60_powerup(struct hdmi_phy *phy,
 		unsigned long int pixclock)
 {
-	struct hdmi_phy_8x60 *phy_8x60 = to_hdmi_phy_8x60(phy);
-	struct hdmi *hdmi = phy_8x60->hdmi;
-
 	/* De-serializer delay D/C for non-lbk mode: */
-	hdmi_write(hdmi, REG_HDMI_8x60_PHY_REG0,
-			HDMI_8x60_PHY_REG0_DESER_DEL_CTRL(3));
+	hdmi_phy_write(phy, REG_HDMI_8x60_PHY_REG0,
+		       HDMI_8x60_PHY_REG0_DESER_DEL_CTRL(3));
 
 	if (pixclock == 27000000) {
 		/* video_format == HDMI_VFRMT_720x480p60_16_9 */
-		hdmi_write(hdmi, REG_HDMI_8x60_PHY_REG1,
-				HDMI_8x60_PHY_REG1_DTEST_MUX_SEL(5) |
-				HDMI_8x60_PHY_REG1_OUTVOL_SWING_CTRL(3));
+		hdmi_phy_write(phy, REG_HDMI_8x60_PHY_REG1,
+			       HDMI_8x60_PHY_REG1_DTEST_MUX_SEL(5) |
+			       HDMI_8x60_PHY_REG1_OUTVOL_SWING_CTRL(3));
 	} else {
-		hdmi_write(hdmi, REG_HDMI_8x60_PHY_REG1,
-				HDMI_8x60_PHY_REG1_DTEST_MUX_SEL(5) |
-				HDMI_8x60_PHY_REG1_OUTVOL_SWING_CTRL(4));
+		hdmi_phy_write(phy, REG_HDMI_8x60_PHY_REG1,
+			       HDMI_8x60_PHY_REG1_DTEST_MUX_SEL(5) |
+			       HDMI_8x60_PHY_REG1_OUTVOL_SWING_CTRL(4));
 	}
 
 	/* No matter what, start from the power down mode: */
-	hdmi_write(hdmi, REG_HDMI_8x60_PHY_REG2,
-			HDMI_8x60_PHY_REG2_PD_PWRGEN |
-			HDMI_8x60_PHY_REG2_PD_PLL |
-			HDMI_8x60_PHY_REG2_PD_DRIVE_4 |
-			HDMI_8x60_PHY_REG2_PD_DRIVE_3 |
-			HDMI_8x60_PHY_REG2_PD_DRIVE_2 |
-			HDMI_8x60_PHY_REG2_PD_DRIVE_1 |
-			HDMI_8x60_PHY_REG2_PD_DESER);
+	hdmi_phy_write(phy, REG_HDMI_8x60_PHY_REG2,
+		       HDMI_8x60_PHY_REG2_PD_PWRGEN |
+		       HDMI_8x60_PHY_REG2_PD_PLL |
+		       HDMI_8x60_PHY_REG2_PD_DRIVE_4 |
+		       HDMI_8x60_PHY_REG2_PD_DRIVE_3 |
+		       HDMI_8x60_PHY_REG2_PD_DRIVE_2 |
+		       HDMI_8x60_PHY_REG2_PD_DRIVE_1 |
+		       HDMI_8x60_PHY_REG2_PD_DESER);
 
 	/* Turn PowerGen on: */
-	hdmi_write(hdmi, REG_HDMI_8x60_PHY_REG2,
-			HDMI_8x60_PHY_REG2_PD_PLL |
-			HDMI_8x60_PHY_REG2_PD_DRIVE_4 |
-			HDMI_8x60_PHY_REG2_PD_DRIVE_3 |
-			HDMI_8x60_PHY_REG2_PD_DRIVE_2 |
-			HDMI_8x60_PHY_REG2_PD_DRIVE_1 |
-			HDMI_8x60_PHY_REG2_PD_DESER);
+	hdmi_phy_write(phy, REG_HDMI_8x60_PHY_REG2,
+		       HDMI_8x60_PHY_REG2_PD_PLL |
+		       HDMI_8x60_PHY_REG2_PD_DRIVE_4 |
+		       HDMI_8x60_PHY_REG2_PD_DRIVE_3 |
+		       HDMI_8x60_PHY_REG2_PD_DRIVE_2 |
+		       HDMI_8x60_PHY_REG2_PD_DRIVE_1 |
+		       HDMI_8x60_PHY_REG2_PD_DESER);
 
 	/* Turn PLL power on: */
-	hdmi_write(hdmi, REG_HDMI_8x60_PHY_REG2,
-			HDMI_8x60_PHY_REG2_PD_DRIVE_4 |
-			HDMI_8x60_PHY_REG2_PD_DRIVE_3 |
-			HDMI_8x60_PHY_REG2_PD_DRIVE_2 |
-			HDMI_8x60_PHY_REG2_PD_DRIVE_1 |
-			HDMI_8x60_PHY_REG2_PD_DESER);
+	hdmi_phy_write(phy, REG_HDMI_8x60_PHY_REG2,
+		       HDMI_8x60_PHY_REG2_PD_DRIVE_4 |
+		       HDMI_8x60_PHY_REG2_PD_DRIVE_3 |
+		       HDMI_8x60_PHY_REG2_PD_DRIVE_2 |
+		       HDMI_8x60_PHY_REG2_PD_DRIVE_1 |
+		       HDMI_8x60_PHY_REG2_PD_DESER);
 
 	/* Write to HIGH after PLL power down de-assert: */
-	hdmi_write(hdmi, REG_HDMI_8x60_PHY_REG3,
-			HDMI_8x60_PHY_REG3_PLL_ENABLE);
+	hdmi_phy_write(phy, REG_HDMI_8x60_PHY_REG3,
+		       HDMI_8x60_PHY_REG3_PLL_ENABLE);
 
 	/* ASIC power on; PHY REG9 = 0 */
-	hdmi_write(hdmi, REG_HDMI_8x60_PHY_REG9, 0);
+	hdmi_phy_write(phy, REG_HDMI_8x60_PHY_REG9, 0);
 
 	/* Enable PLL lock detect, PLL lock det will go high after lock
 	 * Enable the re-time logic
 	 */
-	hdmi_write(hdmi, REG_HDMI_8x60_PHY_REG12,
-			HDMI_8x60_PHY_REG12_RETIMING_EN |
-			HDMI_8x60_PHY_REG12_PLL_LOCK_DETECT_EN);
+	hdmi_phy_write(phy, REG_HDMI_8x60_PHY_REG12,
+		       HDMI_8x60_PHY_REG12_RETIMING_EN |
+		       HDMI_8x60_PHY_REG12_PLL_LOCK_DETECT_EN);
 
 	/* Drivers are on: */
-	hdmi_write(hdmi, REG_HDMI_8x60_PHY_REG2,
-			HDMI_8x60_PHY_REG2_PD_DESER);
+	hdmi_phy_write(phy, REG_HDMI_8x60_PHY_REG2,
+		       HDMI_8x60_PHY_REG2_PD_DESER);
 
 	/* If the RX detector is needed: */
-	hdmi_write(hdmi, REG_HDMI_8x60_PHY_REG2,
-			HDMI_8x60_PHY_REG2_RCV_SENSE_EN |
-			HDMI_8x60_PHY_REG2_PD_DESER);
-
-	hdmi_write(hdmi, REG_HDMI_8x60_PHY_REG4, 0);
-	hdmi_write(hdmi, REG_HDMI_8x60_PHY_REG5, 0);
-	hdmi_write(hdmi, REG_HDMI_8x60_PHY_REG6, 0);
-	hdmi_write(hdmi, REG_HDMI_8x60_PHY_REG7, 0);
-	hdmi_write(hdmi, REG_HDMI_8x60_PHY_REG8, 0);
-	hdmi_write(hdmi, REG_HDMI_8x60_PHY_REG9, 0);
-	hdmi_write(hdmi, REG_HDMI_8x60_PHY_REG10, 0);
-	hdmi_write(hdmi, REG_HDMI_8x60_PHY_REG11, 0);
+	hdmi_phy_write(phy, REG_HDMI_8x60_PHY_REG2,
+		       HDMI_8x60_PHY_REG2_RCV_SENSE_EN |
+		       HDMI_8x60_PHY_REG2_PD_DESER);
+
+	hdmi_phy_write(phy, REG_HDMI_8x60_PHY_REG4, 0);
+	hdmi_phy_write(phy, REG_HDMI_8x60_PHY_REG5, 0);
+	hdmi_phy_write(phy, REG_HDMI_8x60_PHY_REG6, 0);
+	hdmi_phy_write(phy, REG_HDMI_8x60_PHY_REG7, 0);
+	hdmi_phy_write(phy, REG_HDMI_8x60_PHY_REG8, 0);
+	hdmi_phy_write(phy, REG_HDMI_8x60_PHY_REG9, 0);
+	hdmi_phy_write(phy, REG_HDMI_8x60_PHY_REG10, 0);
+	hdmi_phy_write(phy, REG_HDMI_8x60_PHY_REG11, 0);
 
 	/* If we want to use lock enable based on counting: */
-	hdmi_write(hdmi, REG_HDMI_8x60_PHY_REG12,
-			HDMI_8x60_PHY_REG12_RETIMING_EN |
-			HDMI_8x60_PHY_REG12_PLL_LOCK_DETECT_EN |
-			HDMI_8x60_PHY_REG12_FORCE_LOCK);
+	hdmi_phy_write(phy, REG_HDMI_8x60_PHY_REG12,
+		       HDMI_8x60_PHY_REG12_RETIMING_EN |
+		       HDMI_8x60_PHY_REG12_PLL_LOCK_DETECT_EN |
+		       HDMI_8x60_PHY_REG12_FORCE_LOCK);
 }
 
 static void hdmi_phy_8x60_powerdown(struct hdmi_phy *phy)
 {
-	struct hdmi_phy_8x60 *phy_8x60 = to_hdmi_phy_8x60(phy);
-	struct hdmi *hdmi = phy_8x60->hdmi;
-
 	/* Assert RESET PHY from controller */
-	hdmi_write(hdmi, REG_HDMI_PHY_CTRL,
-			HDMI_PHY_CTRL_SW_RESET);
+	hdmi_phy_write(phy, REG_HDMI_PHY_CTRL,
+		       HDMI_PHY_CTRL_SW_RESET);
 	udelay(10);
 	/* De-assert RESET PHY from controller */
-	hdmi_write(hdmi, REG_HDMI_PHY_CTRL, 0);
+	hdmi_phy_write(phy, REG_HDMI_PHY_CTRL, 0);
 	/* Turn off Driver */
-	hdmi_write(hdmi, REG_HDMI_8x60_PHY_REG2,
-			HDMI_8x60_PHY_REG2_PD_DRIVE_4 |
-			HDMI_8x60_PHY_REG2_PD_DRIVE_3 |
-			HDMI_8x60_PHY_REG2_PD_DRIVE_2 |
-			HDMI_8x60_PHY_REG2_PD_DRIVE_1 |
-			HDMI_8x60_PHY_REG2_PD_DESER);
+	hdmi_phy_write(phy, REG_HDMI_8x60_PHY_REG2,
+		       HDMI_8x60_PHY_REG2_PD_DRIVE_4 |
+		       HDMI_8x60_PHY_REG2_PD_DRIVE_3 |
+		       HDMI_8x60_PHY_REG2_PD_DRIVE_2 |
+		       HDMI_8x60_PHY_REG2_PD_DRIVE_1 |
+		       HDMI_8x60_PHY_REG2_PD_DESER);
 	udelay(10);
 	/* Disable PLL */
-	hdmi_write(hdmi, REG_HDMI_8x60_PHY_REG3, 0);
+	hdmi_phy_write(phy, REG_HDMI_8x60_PHY_REG3, 0);
 	/* Power down PHY, but keep RX-sense: */
-	hdmi_write(hdmi, REG_HDMI_8x60_PHY_REG2,
-			HDMI_8x60_PHY_REG2_RCV_SENSE_EN |
-			HDMI_8x60_PHY_REG2_PD_PWRGEN |
-			HDMI_8x60_PHY_REG2_PD_PLL |
-			HDMI_8x60_PHY_REG2_PD_DRIVE_4 |
-			HDMI_8x60_PHY_REG2_PD_DRIVE_3 |
-			HDMI_8x60_PHY_REG2_PD_DRIVE_2 |
-			HDMI_8x60_PHY_REG2_PD_DRIVE_1 |
-			HDMI_8x60_PHY_REG2_PD_DESER);
+	hdmi_phy_write(phy, REG_HDMI_8x60_PHY_REG2,
+		       HDMI_8x60_PHY_REG2_RCV_SENSE_EN |
+		       HDMI_8x60_PHY_REG2_PD_PWRGEN |
+		       HDMI_8x60_PHY_REG2_PD_PLL |
+		       HDMI_8x60_PHY_REG2_PD_DRIVE_4 |
+		       HDMI_8x60_PHY_REG2_PD_DRIVE_3 |
+		       HDMI_8x60_PHY_REG2_PD_DRIVE_2 |
+		       HDMI_8x60_PHY_REG2_PD_DRIVE_1 |
+		       HDMI_8x60_PHY_REG2_PD_DESER);
 }
 
-static const struct hdmi_phy_funcs hdmi_phy_8x60_funcs = {
-		.destroy = hdmi_phy_8x60_destroy,
-		.powerup = hdmi_phy_8x60_powerup,
-		.powerdown = hdmi_phy_8x60_powerdown,
+const struct hdmi_phy_cfg msm_hdmi_phy_8x60_cfg = {
+	.type = MSM_HDMI_PHY_8x60,
+	.powerup = hdmi_phy_8x60_powerup,
+	.powerdown = hdmi_phy_8x60_powerdown,
 };
-
-struct hdmi_phy *hdmi_phy_8x60_init(struct hdmi *hdmi)
-{
-	struct hdmi_phy_8x60 *phy_8x60;
-	struct hdmi_phy *phy = NULL;
-	int ret;
-
-	phy_8x60 = kzalloc(sizeof(*phy_8x60), GFP_KERNEL);
-	if (!phy_8x60) {
-		ret = -ENOMEM;
-		goto fail;
-	}
-
-	phy = &phy_8x60->base;
-
-	phy->funcs = &hdmi_phy_8x60_funcs;
-
-	phy_8x60->hdmi = hdmi;
-
-	return phy;
-
-fail:
-	if (phy)
-		hdmi_phy_8x60_destroy(phy);
-	return ERR_PTR(ret);
-}
diff --git a/drivers/gpu/drm/msm/hdmi/hdmi_phy_8x74.c b/drivers/gpu/drm/msm/hdmi/hdmi_phy_8x74.c
index 56ab8917ee9a..c4a61e537851 100644
--- a/drivers/gpu/drm/msm/hdmi/hdmi_phy_8x74.c
+++ b/drivers/gpu/drm/msm/hdmi/hdmi_phy_8x74.c
@@ -17,84 +17,40 @@
 
 #include "hdmi.h"
 
-struct hdmi_phy_8x74 {
-	struct hdmi_phy base;
-	void __iomem *mmio;
-};
-#define to_hdmi_phy_8x74(x) container_of(x, struct hdmi_phy_8x74, base)
-
-
-static void phy_write(struct hdmi_phy_8x74 *phy, u32 reg, u32 data)
-{
-	msm_writel(data, phy->mmio + reg);
-}
-
-//static u32 phy_read(struct hdmi_phy_8x74 *phy, u32 reg)
-//{
-//	return msm_readl(phy->mmio + reg);
-//}
-
-static void hdmi_phy_8x74_destroy(struct hdmi_phy *phy)
-{
-	struct hdmi_phy_8x74 *phy_8x74 = to_hdmi_phy_8x74(phy);
-	kfree(phy_8x74);
-}
-
 static void hdmi_phy_8x74_powerup(struct hdmi_phy *phy,
 		unsigned long int pixclock)
 {
-	struct hdmi_phy_8x74 *phy_8x74 = to_hdmi_phy_8x74(phy);
-
-	phy_write(phy_8x74, REG_HDMI_8x74_ANA_CFG0,   0x1b);
-	phy_write(phy_8x74, REG_HDMI_8x74_ANA_CFG1,   0xf2);
-	phy_write(phy_8x74, REG_HDMI_8x74_BIST_CFG0,  0x0);
-	phy_write(phy_8x74, REG_HDMI_8x74_BIST_PATN0, 0x0);
-	phy_write(phy_8x74, REG_HDMI_8x74_BIST_PATN1, 0x0);
-	phy_write(phy_8x74, REG_HDMI_8x74_BIST_PATN2, 0x0);
-	phy_write(phy_8x74, REG_HDMI_8x74_BIST_PATN3, 0x0);
-	phy_write(phy_8x74, REG_HDMI_8x74_PD_CTRL1,   0x20);
+	hdmi_phy_write(phy, REG_HDMI_8x74_ANA_CFG0,   0x1b);
+	hdmi_phy_write(phy, REG_HDMI_8x74_ANA_CFG1,   0xf2);
+	hdmi_phy_write(phy, REG_HDMI_8x74_BIST_CFG0,  0x0);
+	hdmi_phy_write(phy, REG_HDMI_8x74_BIST_PATN0, 0x0);
+	hdmi_phy_write(phy, REG_HDMI_8x74_BIST_PATN1, 0x0);
+	hdmi_phy_write(phy, REG_HDMI_8x74_BIST_PATN2, 0x0);
+	hdmi_phy_write(phy, REG_HDMI_8x74_BIST_PATN3, 0x0);
+	hdmi_phy_write(phy, REG_HDMI_8x74_PD_CTRL1,   0x20);
 }
 
 static void hdmi_phy_8x74_powerdown(struct hdmi_phy *phy)
 {
-	struct hdmi_phy_8x74 *phy_8x74 = to_hdmi_phy_8x74(phy);
-	phy_write(phy_8x74, REG_HDMI_8x74_PD_CTRL0, 0x7f);
+	hdmi_phy_write(phy, REG_HDMI_8x74_PD_CTRL0, 0x7f);
 }
 
-static const struct hdmi_phy_funcs hdmi_phy_8x74_funcs = {
-		.destroy = hdmi_phy_8x74_destroy,
-		.powerup = hdmi_phy_8x74_powerup,
-		.powerdown = hdmi_phy_8x74_powerdown,
+static const char * const hdmi_phy_8x74_reg_names[] = {
+	"core-vdda",
+	"vddio",
 };
 
-struct hdmi_phy *hdmi_phy_8x74_init(struct hdmi *hdmi)
-{
-	struct hdmi_phy_8x74 *phy_8x74;
-	struct hdmi_phy *phy = NULL;
-	int ret;
-
-	phy_8x74 = kzalloc(sizeof(*phy_8x74), GFP_KERNEL);
-	if (!phy_8x74) {
-		ret = -ENOMEM;
-		goto fail;
-	}
-
-	phy = &phy_8x74->base;
-
-	phy->funcs = &hdmi_phy_8x74_funcs;
-
-	/* for 8x74, the phy mmio is mapped separately: */
-	phy_8x74->mmio = msm_ioremap(hdmi->pdev,
-			"phy_physical", "HDMI_8x74");
-	if (IS_ERR(phy_8x74->mmio)) {
-		ret = PTR_ERR(phy_8x74->mmio);
-		goto fail;
-	}
-
-	return phy;
+static const char * const hdmi_phy_8x74_clk_names[] = {
+	"iface_clk",
+	"alt_iface_clk"
+};
 
-fail:
-	if (phy)
-		hdmi_phy_8x74_destroy(phy);
-	return ERR_PTR(ret);
-}
+const struct hdmi_phy_cfg msm_hdmi_phy_8x74_cfg = {
+	.type = MSM_HDMI_PHY_8x74,
+	.powerup = hdmi_phy_8x74_powerup,
+	.powerdown = hdmi_phy_8x74_powerdown,
+	.reg_names = hdmi_phy_8x74_reg_names,
+	.num_regs = ARRAY_SIZE(hdmi_phy_8x74_reg_names),
+	.clk_names = hdmi_phy_8x74_clk_names,
+	.num_clks = ARRAY_SIZE(hdmi_phy_8x74_clk_names),
+};
diff --git a/drivers/gpu/drm/msm/hdmi/hdmi_pll_8960.c b/drivers/gpu/drm/msm/hdmi/hdmi_pll_8960.c
new file mode 100644
index 000000000000..92da69aa6187
--- /dev/null
+++ b/drivers/gpu/drm/msm/hdmi/hdmi_pll_8960.c
@@ -0,0 +1,461 @@
+/*
+ * Copyright (c) 2016, The Linux Foundation. All rights reserved.
+ * Copyright (C) 2013 Red Hat
+ * Author: Rob Clark <robdclark@gmail.com>
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms of the GNU General Public License version 2 as published by
+ * the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful, but WITHOUT
+ * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
+ * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
+ * more details.
+ *
+ * You should have received a copy of the GNU General Public License along with
+ * this program.  If not, see <http://www.gnu.org/licenses/>.
+ */
+
+#include <linux/clk-provider.h>
+#include "hdmi.h"
+
+struct hdmi_pll_8960 {
+	struct platform_device *pdev;
+	struct clk_hw clk_hw;
+	void __iomem *mmio;
+
+	unsigned long pixclk;
+};
+
+#define hw_clk_to_pll(x) container_of(x, struct hdmi_pll_8960, clk_hw)
+
+/*
+ * HDMI PLL:
+ *
+ * To get the parent clock setup properly, we need to plug in hdmi pll
+ * configuration into common-clock-framework.
+ */
+
+struct pll_rate {
+	unsigned long rate;
+	int num_reg;
+	struct {
+		u32 val;
+		u32 reg;
+	} conf[32];
+};
+
+/* NOTE: keep sorted highest freq to lowest: */
+static const struct pll_rate freqtbl[] = {
+	{ 154000000, 14, {
+		{ 0x08, REG_HDMI_8960_PHY_PLL_REFCLK_CFG    },
+		{ 0x20, REG_HDMI_8960_PHY_PLL_LOOP_FLT_CFG0 },
+		{ 0xf9, REG_HDMI_8960_PHY_PLL_LOOP_FLT_CFG1 },
+		{ 0x02, REG_HDMI_8960_PHY_PLL_VCOCAL_CFG0   },
+		{ 0x03, REG_HDMI_8960_PHY_PLL_VCOCAL_CFG1   },
+		{ 0x3b, REG_HDMI_8960_PHY_PLL_VCOCAL_CFG2   },
+		{ 0x00, REG_HDMI_8960_PHY_PLL_VCOCAL_CFG3   },
+		{ 0x86, REG_HDMI_8960_PHY_PLL_VCOCAL_CFG4   },
+		{ 0x00, REG_HDMI_8960_PHY_PLL_VCOCAL_CFG5   },
+		{ 0x0d, REG_HDMI_8960_PHY_PLL_SDM_CFG0      },
+		{ 0x4d, REG_HDMI_8960_PHY_PLL_SDM_CFG1      },
+		{ 0x5e, REG_HDMI_8960_PHY_PLL_SDM_CFG2      },
+		{ 0x42, REG_HDMI_8960_PHY_PLL_SDM_CFG3      },
+		{ 0x00, REG_HDMI_8960_PHY_PLL_SDM_CFG4      },
+			}
+	},
+	/* 1080p60/1080p50 case */
+	{ 148500000, 27, {
+		{ 0x02, REG_HDMI_8960_PHY_PLL_REFCLK_CFG    },
+		{ 0x02, REG_HDMI_8960_PHY_PLL_CHRG_PUMP_CFG },
+		{ 0x01, REG_HDMI_8960_PHY_PLL_LOOP_FLT_CFG0 },
+		{ 0x33, REG_HDMI_8960_PHY_PLL_LOOP_FLT_CFG1 },
+		{ 0x2c, REG_HDMI_8960_PHY_PLL_IDAC_ADJ_CFG  },
+		{ 0x06, REG_HDMI_8960_PHY_PLL_I_VI_KVCO_CFG },
+		{ 0x0a, REG_HDMI_8960_PHY_PLL_PWRDN_B       },
+		{ 0x76, REG_HDMI_8960_PHY_PLL_SDM_CFG0      },
+		{ 0x01, REG_HDMI_8960_PHY_PLL_SDM_CFG1      },
+		{ 0x4c, REG_HDMI_8960_PHY_PLL_SDM_CFG2      },
+		{ 0xc0, REG_HDMI_8960_PHY_PLL_SDM_CFG3      },
+		{ 0x00, REG_HDMI_8960_PHY_PLL_SDM_CFG4      },
+		{ 0x9a, REG_HDMI_8960_PHY_PLL_SSC_CFG0      },
+		{ 0x00, REG_HDMI_8960_PHY_PLL_SSC_CFG1      },
+		{ 0x00, REG_HDMI_8960_PHY_PLL_SSC_CFG2      },
+		{ 0x00, REG_HDMI_8960_PHY_PLL_SSC_CFG3      },
+		{ 0x10, REG_HDMI_8960_PHY_PLL_LOCKDET_CFG0  },
+		{ 0x1a, REG_HDMI_8960_PHY_PLL_LOCKDET_CFG1  },
+		{ 0x0d, REG_HDMI_8960_PHY_PLL_LOCKDET_CFG2  },
+		{ 0xe6, REG_HDMI_8960_PHY_PLL_VCOCAL_CFG0   },
+		{ 0x02, REG_HDMI_8960_PHY_PLL_VCOCAL_CFG1   },
+		{ 0x3b, REG_HDMI_8960_PHY_PLL_VCOCAL_CFG2   },
+		{ 0x00, REG_HDMI_8960_PHY_PLL_VCOCAL_CFG3   },
+		{ 0x86, REG_HDMI_8960_PHY_PLL_VCOCAL_CFG4   },
+		{ 0x00, REG_HDMI_8960_PHY_PLL_VCOCAL_CFG5   },
+		{ 0x33, REG_HDMI_8960_PHY_PLL_VCOCAL_CFG6   },
+		{ 0x00, REG_HDMI_8960_PHY_PLL_VCOCAL_CFG7   },
+			}
+	},
+	{ 108000000, 13, {
+		{ 0x08, REG_HDMI_8960_PHY_PLL_REFCLK_CFG    },
+		{ 0x21, REG_HDMI_8960_PHY_PLL_LOOP_FLT_CFG0 },
+		{ 0xf9, REG_HDMI_8960_PHY_PLL_LOOP_FLT_CFG1 },
+		{ 0x1c, REG_HDMI_8960_PHY_PLL_VCOCAL_CFG0   },
+		{ 0x02, REG_HDMI_8960_PHY_PLL_VCOCAL_CFG1   },
+		{ 0x3b, REG_HDMI_8960_PHY_PLL_VCOCAL_CFG2   },
+		{ 0x86, REG_HDMI_8960_PHY_PLL_VCOCAL_CFG4   },
+		{ 0x00, REG_HDMI_8960_PHY_PLL_VCOCAL_CFG5   },
+		{ 0x49, REG_HDMI_8960_PHY_PLL_SDM_CFG0      },
+		{ 0x49, REG_HDMI_8960_PHY_PLL_SDM_CFG1      },
+		{ 0x00, REG_HDMI_8960_PHY_PLL_SDM_CFG2      },
+		{ 0x00, REG_HDMI_8960_PHY_PLL_SDM_CFG3      },
+		{ 0x00, REG_HDMI_8960_PHY_PLL_SDM_CFG4      },
+			}
+	},
+	/* 720p60/720p50/1080i60/1080i50/1080p24/1080p30/1080p25 */
+	{ 74250000, 8, {
+		{ 0x0a, REG_HDMI_8960_PHY_PLL_PWRDN_B       },
+		{ 0x12, REG_HDMI_8960_PHY_PLL_REFCLK_CFG    },
+		{ 0x01, REG_HDMI_8960_PHY_PLL_LOOP_FLT_CFG0 },
+		{ 0x33, REG_HDMI_8960_PHY_PLL_LOOP_FLT_CFG1 },
+		{ 0x76, REG_HDMI_8960_PHY_PLL_SDM_CFG0      },
+		{ 0xe6, REG_HDMI_8960_PHY_PLL_VCOCAL_CFG0   },
+		{ 0x02, REG_HDMI_8960_PHY_PLL_VCOCAL_CFG1   },
+		{ 0x3b, REG_HDMI_8960_PHY_PLL_VCOCAL_CFG2   },
+			}
+	},
+	{ 74176000, 14, {
+		{ 0x18, REG_HDMI_8960_PHY_PLL_REFCLK_CFG    },
+		{ 0x20, REG_HDMI_8960_PHY_PLL_LOOP_FLT_CFG0 },
+		{ 0xf9, REG_HDMI_8960_PHY_PLL_LOOP_FLT_CFG1 },
+		{ 0xe5, REG_HDMI_8960_PHY_PLL_VCOCAL_CFG0   },
+		{ 0x02, REG_HDMI_8960_PHY_PLL_VCOCAL_CFG1   },
+		{ 0x3b, REG_HDMI_8960_PHY_PLL_VCOCAL_CFG2   },
+		{ 0x00, REG_HDMI_8960_PHY_PLL_VCOCAL_CFG3   },
+		{ 0x86, REG_HDMI_8960_PHY_PLL_VCOCAL_CFG4   },
+		{ 0x00, REG_HDMI_8960_PHY_PLL_VCOCAL_CFG5   },
+		{ 0x0c, REG_HDMI_8960_PHY_PLL_SDM_CFG0      },
+		{ 0x4c, REG_HDMI_8960_PHY_PLL_SDM_CFG1      },
+		{ 0x7d, REG_HDMI_8960_PHY_PLL_SDM_CFG2      },
+		{ 0xbc, REG_HDMI_8960_PHY_PLL_SDM_CFG3      },
+		{ 0x00, REG_HDMI_8960_PHY_PLL_SDM_CFG4      },
+			}
+	},
+	{ 65000000, 14, {
+		{ 0x18, REG_HDMI_8960_PHY_PLL_REFCLK_CFG    },
+		{ 0x20, REG_HDMI_8960_PHY_PLL_LOOP_FLT_CFG0 },
+		{ 0xf9, REG_HDMI_8960_PHY_PLL_LOOP_FLT_CFG1 },
+		{ 0x8a, REG_HDMI_8960_PHY_PLL_VCOCAL_CFG0   },
+		{ 0x02, REG_HDMI_8960_PHY_PLL_VCOCAL_CFG1   },
+		{ 0x3b, REG_HDMI_8960_PHY_PLL_VCOCAL_CFG2   },
+		{ 0x00, REG_HDMI_8960_PHY_PLL_VCOCAL_CFG3   },
+		{ 0x86, REG_HDMI_8960_PHY_PLL_VCOCAL_CFG4   },
+		{ 0x00, REG_HDMI_8960_PHY_PLL_VCOCAL_CFG5   },
+		{ 0x0b, REG_HDMI_8960_PHY_PLL_SDM_CFG0      },
+		{ 0x4b, REG_HDMI_8960_PHY_PLL_SDM_CFG1      },
+		{ 0x7b, REG_HDMI_8960_PHY_PLL_SDM_CFG2      },
+		{ 0x09, REG_HDMI_8960_PHY_PLL_SDM_CFG3      },
+		{ 0x00, REG_HDMI_8960_PHY_PLL_SDM_CFG4      },
+			}
+	},
+	/* 480p60/480i60 */
+	{ 27030000, 18, {
+		{ 0x0a, REG_HDMI_8960_PHY_PLL_PWRDN_B       },
+		{ 0x38, REG_HDMI_8960_PHY_PLL_REFCLK_CFG    },
+		{ 0x02, REG_HDMI_8960_PHY_PLL_CHRG_PUMP_CFG },
+		{ 0x20, REG_HDMI_8960_PHY_PLL_LOOP_FLT_CFG0 },
+		{ 0xff, REG_HDMI_8960_PHY_PLL_LOOP_FLT_CFG1 },
+		{ 0x00, REG_HDMI_8960_PHY_PLL_SDM_CFG0      },
+		{ 0x4e, REG_HDMI_8960_PHY_PLL_SDM_CFG1      },
+		{ 0xd7, REG_HDMI_8960_PHY_PLL_SDM_CFG2      },
+		{ 0x03, REG_HDMI_8960_PHY_PLL_SDM_CFG3      },
+		{ 0x00, REG_HDMI_8960_PHY_PLL_SDM_CFG4      },
+		{ 0x2a, REG_HDMI_8960_PHY_PLL_VCOCAL_CFG0   },
+		{ 0x03, REG_HDMI_8960_PHY_PLL_VCOCAL_CFG1   },
+		{ 0x3b, REG_HDMI_8960_PHY_PLL_VCOCAL_CFG2   },
+		{ 0x00, REG_HDMI_8960_PHY_PLL_VCOCAL_CFG3   },
+		{ 0x86, REG_HDMI_8960_PHY_PLL_VCOCAL_CFG4   },
+		{ 0x00, REG_HDMI_8960_PHY_PLL_VCOCAL_CFG5   },
+		{ 0x33, REG_HDMI_8960_PHY_PLL_VCOCAL_CFG6   },
+		{ 0x00, REG_HDMI_8960_PHY_PLL_VCOCAL_CFG7   },
+			}
+	},
+	/* 576p50/576i50 */
+	{ 27000000, 27, {
+		{ 0x32, REG_HDMI_8960_PHY_PLL_REFCLK_CFG    },
+		{ 0x02, REG_HDMI_8960_PHY_PLL_CHRG_PUMP_CFG },
+		{ 0x01, REG_HDMI_8960_PHY_PLL_LOOP_FLT_CFG0 },
+		{ 0x33, REG_HDMI_8960_PHY_PLL_LOOP_FLT_CFG1 },
+		{ 0x2c, REG_HDMI_8960_PHY_PLL_IDAC_ADJ_CFG  },
+		{ 0x06, REG_HDMI_8960_PHY_PLL_I_VI_KVCO_CFG },
+		{ 0x0a, REG_HDMI_8960_PHY_PLL_PWRDN_B       },
+		{ 0x7b, REG_HDMI_8960_PHY_PLL_SDM_CFG0      },
+		{ 0x01, REG_HDMI_8960_PHY_PLL_SDM_CFG1      },
+		{ 0x4c, REG_HDMI_8960_PHY_PLL_SDM_CFG2      },
+		{ 0xc0, REG_HDMI_8960_PHY_PLL_SDM_CFG3      },
+		{ 0x00, REG_HDMI_8960_PHY_PLL_SDM_CFG4      },
+		{ 0x9a, REG_HDMI_8960_PHY_PLL_SSC_CFG0      },
+		{ 0x00, REG_HDMI_8960_PHY_PLL_SSC_CFG1      },
+		{ 0x00, REG_HDMI_8960_PHY_PLL_SSC_CFG2      },
+		{ 0x00, REG_HDMI_8960_PHY_PLL_SSC_CFG3      },
+		{ 0x10, REG_HDMI_8960_PHY_PLL_LOCKDET_CFG0  },
+		{ 0x1a, REG_HDMI_8960_PHY_PLL_LOCKDET_CFG1  },
+		{ 0x0d, REG_HDMI_8960_PHY_PLL_LOCKDET_CFG2  },
+		{ 0x2a, REG_HDMI_8960_PHY_PLL_VCOCAL_CFG0   },
+		{ 0x03, REG_HDMI_8960_PHY_PLL_VCOCAL_CFG1   },
+		{ 0x3b, REG_HDMI_8960_PHY_PLL_VCOCAL_CFG2   },
+		{ 0x00, REG_HDMI_8960_PHY_PLL_VCOCAL_CFG3   },
+		{ 0x86, REG_HDMI_8960_PHY_PLL_VCOCAL_CFG4   },
+		{ 0x00, REG_HDMI_8960_PHY_PLL_VCOCAL_CFG5   },
+		{ 0x33, REG_HDMI_8960_PHY_PLL_VCOCAL_CFG6   },
+		{ 0x00, REG_HDMI_8960_PHY_PLL_VCOCAL_CFG7   },
+			}
+	},
+	/* 640x480p60 */
+	{ 25200000, 27, {
+		{ 0x32, REG_HDMI_8960_PHY_PLL_REFCLK_CFG    },
+		{ 0x02, REG_HDMI_8960_PHY_PLL_CHRG_PUMP_CFG },
+		{ 0x01, REG_HDMI_8960_PHY_PLL_LOOP_FLT_CFG0 },
+		{ 0x33, REG_HDMI_8960_PHY_PLL_LOOP_FLT_CFG1 },
+		{ 0x2c, REG_HDMI_8960_PHY_PLL_IDAC_ADJ_CFG  },
+		{ 0x06, REG_HDMI_8960_PHY_PLL_I_VI_KVCO_CFG },
+		{ 0x0a, REG_HDMI_8960_PHY_PLL_PWRDN_B       },
+		{ 0x77, REG_HDMI_8960_PHY_PLL_SDM_CFG0      },
+		{ 0x4c, REG_HDMI_8960_PHY_PLL_SDM_CFG1      },
+		{ 0x00, REG_HDMI_8960_PHY_PLL_SDM_CFG2      },
+		{ 0xc0, REG_HDMI_8960_PHY_PLL_SDM_CFG3      },
+		{ 0x00, REG_HDMI_8960_PHY_PLL_SDM_CFG4      },
+		{ 0x9a, REG_HDMI_8960_PHY_PLL_SSC_CFG0      },
+		{ 0x00, REG_HDMI_8960_PHY_PLL_SSC_CFG1      },
+		{ 0x00, REG_HDMI_8960_PHY_PLL_SSC_CFG2      },
+		{ 0x20, REG_HDMI_8960_PHY_PLL_SSC_CFG3      },
+		{ 0x10, REG_HDMI_8960_PHY_PLL_LOCKDET_CFG0  },
+		{ 0x1a, REG_HDMI_8960_PHY_PLL_LOCKDET_CFG1  },
+		{ 0x0d, REG_HDMI_8960_PHY_PLL_LOCKDET_CFG2  },
+		{ 0xf4, REG_HDMI_8960_PHY_PLL_VCOCAL_CFG0   },
+		{ 0x02, REG_HDMI_8960_PHY_PLL_VCOCAL_CFG1   },
+		{ 0x3b, REG_HDMI_8960_PHY_PLL_VCOCAL_CFG2   },
+		{ 0x00, REG_HDMI_8960_PHY_PLL_VCOCAL_CFG3   },
+		{ 0x86, REG_HDMI_8960_PHY_PLL_VCOCAL_CFG4   },
+		{ 0x00, REG_HDMI_8960_PHY_PLL_VCOCAL_CFG5   },
+		{ 0x33, REG_HDMI_8960_PHY_PLL_VCOCAL_CFG6   },
+		{ 0x00, REG_HDMI_8960_PHY_PLL_VCOCAL_CFG7   },
+			}
+	},
+};
+
+static inline void pll_write(struct hdmi_pll_8960 *pll, u32 reg, u32 data)
+{
+	msm_writel(data, pll->mmio + reg);
+}
+
+static inline u32 pll_read(struct hdmi_pll_8960 *pll, u32 reg)
+{
+	return msm_readl(pll->mmio + reg);
+}
+
+static inline struct hdmi_phy *pll_get_phy(struct hdmi_pll_8960 *pll)
+{
+	return platform_get_drvdata(pll->pdev);
+}
+
+static int hdmi_pll_enable(struct clk_hw *hw)
+{
+	struct hdmi_pll_8960 *pll = hw_clk_to_pll(hw);
+	struct hdmi_phy *phy = pll_get_phy(pll);
+	int timeout_count, pll_lock_retry = 10;
+	unsigned int val;
+
+	DBG("");
+
+	/* Assert PLL S/W reset */
+	pll_write(pll, REG_HDMI_8960_PHY_PLL_LOCKDET_CFG2, 0x8d);
+	pll_write(pll, REG_HDMI_8960_PHY_PLL_LOCKDET_CFG0, 0x10);
+	pll_write(pll, REG_HDMI_8960_PHY_PLL_LOCKDET_CFG1, 0x1a);
+
+	/* Wait for a short time before de-asserting
+	 * to allow the hardware to complete its job.
+	 * This much of delay should be fine for hardware
+	 * to assert and de-assert.
+	 */
+	udelay(10);
+
+	/* De-assert PLL S/W reset */
+	pll_write(pll, REG_HDMI_8960_PHY_PLL_LOCKDET_CFG2, 0x0d);
+
+	val = hdmi_phy_read(phy, REG_HDMI_8960_PHY_REG12);
+	val |= HDMI_8960_PHY_REG12_SW_RESET;
+	/* Assert PHY S/W reset */
+	hdmi_phy_write(phy, REG_HDMI_8960_PHY_REG12, val);
+	val &= ~HDMI_8960_PHY_REG12_SW_RESET;
+	/*
+	 * Wait for a short time before de-asserting to allow the hardware to
+	 * complete its job. This much of delay should be fine for hardware to
+	 * assert and de-assert.
+	 */
+	udelay(10);
+	/* De-assert PHY S/W reset */
+	hdmi_phy_write(phy, REG_HDMI_8960_PHY_REG12, val);
+	hdmi_phy_write(phy, REG_HDMI_8960_PHY_REG2,  0x3f);
+
+	val = hdmi_phy_read(phy, REG_HDMI_8960_PHY_REG12);
+	val |= HDMI_8960_PHY_REG12_PWRDN_B;
+	hdmi_phy_write(phy, REG_HDMI_8960_PHY_REG12, val);
+	/* Wait 10 us for enabling global power for PHY */
+	mb();
+	udelay(10);
+
+	val = pll_read(pll, REG_HDMI_8960_PHY_PLL_PWRDN_B);
+	val |= HDMI_8960_PHY_PLL_PWRDN_B_PLL_PWRDN_B;
+	val &= ~HDMI_8960_PHY_PLL_PWRDN_B_PD_PLL;
+	pll_write(pll, REG_HDMI_8960_PHY_PLL_PWRDN_B, val);
+	hdmi_phy_write(phy, REG_HDMI_8960_PHY_REG2, 0x80);
+
+	timeout_count = 1000;
+	while (--pll_lock_retry > 0) {
+		/* are we there yet? */
+		val = pll_read(pll, REG_HDMI_8960_PHY_PLL_STATUS0);
+		if (val & HDMI_8960_PHY_PLL_STATUS0_PLL_LOCK)
+			break;
+
+		udelay(1);
+
+		if (--timeout_count > 0)
+			continue;
+
+		/*
+		 * PLL has still not locked.
+		 * Do a software reset and try again
+		 * Assert PLL S/W reset first
+		 */
+		pll_write(pll, REG_HDMI_8960_PHY_PLL_LOCKDET_CFG2, 0x8d);
+		udelay(10);
+		pll_write(pll, REG_HDMI_8960_PHY_PLL_LOCKDET_CFG2, 0x0d);
+
+		/*
+		 * Wait for a short duration for the PLL calibration
+		 * before checking if the PLL gets locked
+		 */
+		udelay(350);
+
+		timeout_count = 1000;
+	}
+
+	return 0;
+}
+
+static void hdmi_pll_disable(struct clk_hw *hw)
+{
+	struct hdmi_pll_8960 *pll = hw_clk_to_pll(hw);
+	struct hdmi_phy *phy = pll_get_phy(pll);
+	unsigned int val;
+
+	DBG("");
+
+	val = hdmi_phy_read(phy, REG_HDMI_8960_PHY_REG12);
+	val &= ~HDMI_8960_PHY_REG12_PWRDN_B;
+	hdmi_phy_write(phy, REG_HDMI_8960_PHY_REG12, val);
+
+	val = pll_read(pll, REG_HDMI_8960_PHY_PLL_PWRDN_B);
+	val |= HDMI_8960_PHY_REG12_SW_RESET;
+	val &= ~HDMI_8960_PHY_REG12_PWRDN_B;
+	pll_write(pll, REG_HDMI_8960_PHY_PLL_PWRDN_B, val);
+	/* Make sure HDMI PHY/PLL are powered down */
+	mb();
+}
+
+static const struct pll_rate *find_rate(unsigned long rate)
+{
+	int i;
+
+	for (i = 1; i < ARRAY_SIZE(freqtbl); i++)
+		if (rate > freqtbl[i].rate)
+			return &freqtbl[i - 1];
+
+	return &freqtbl[i - 1];
+}
+
+static unsigned long hdmi_pll_recalc_rate(struct clk_hw *hw,
+					  unsigned long parent_rate)
+{
+	struct hdmi_pll_8960 *pll = hw_clk_to_pll(hw);
+
+	return pll->pixclk;
+}
+
+static long hdmi_pll_round_rate(struct clk_hw *hw, unsigned long rate,
+				unsigned long *parent_rate)
+{
+	const struct pll_rate *pll_rate = find_rate(rate);
+
+	return pll_rate->rate;
+}
+
+static int hdmi_pll_set_rate(struct clk_hw *hw, unsigned long rate,
+			     unsigned long parent_rate)
+{
+	struct hdmi_pll_8960 *pll = hw_clk_to_pll(hw);
+	const struct pll_rate *pll_rate = find_rate(rate);
+	int i;
+
+	DBG("rate=%lu", rate);
+
+	for (i = 0; i < pll_rate->num_reg; i++)
+		pll_write(pll, pll_rate->conf[i].reg, pll_rate->conf[i].val);
+
+	pll->pixclk = rate;
+
+	return 0;
+}
+
+static const struct clk_ops hdmi_pll_ops = {
+	.enable = hdmi_pll_enable,
+	.disable = hdmi_pll_disable,
+	.recalc_rate = hdmi_pll_recalc_rate,
+	.round_rate = hdmi_pll_round_rate,
+	.set_rate = hdmi_pll_set_rate,
+};
+
+static const char * const hdmi_pll_parents[] = {
+	"pxo",
+};
+
+static struct clk_init_data pll_init = {
+	.name = "hdmi_pll",
+	.ops = &hdmi_pll_ops,
+	.parent_names = hdmi_pll_parents,
+	.num_parents = ARRAY_SIZE(hdmi_pll_parents),
+};
+
+int msm_hdmi_pll_8960_init(struct platform_device *pdev)
+{
+	struct device *dev = &pdev->dev;
+	struct hdmi_pll_8960 *pll;
+	struct clk *clk;
+	int i;
+
+	/* sanity check: */
+	for (i = 0; i < (ARRAY_SIZE(freqtbl) - 1); i++)
+		if (WARN_ON(freqtbl[i].rate < freqtbl[i + 1].rate))
+			return -EINVAL;
+
+	pll = devm_kzalloc(dev, sizeof(*pll), GFP_KERNEL);
+	if (!pll)
+		return -ENOMEM;
+
+	pll->mmio = msm_ioremap(pdev, "hdmi_pll", "HDMI_PLL");
+	if (IS_ERR(pll->mmio)) {
+		dev_err(dev, "failed to map pll base\n");
+		return -ENOMEM;
+	}
+
+	pll->pdev = pdev;
+	pll->clk_hw.init = &pll_init;
+
+	clk = devm_clk_register(dev, &pll->clk_hw);
+	if (IS_ERR(clk)) {
+		dev_err(dev, "failed to register pll clock\n");
+		return -EINVAL;
+	}
+
+	return 0;
+}
diff --git a/drivers/gpu/drm/msm/hdmi/qfprom.xml.h b/drivers/gpu/drm/msm/hdmi/qfprom.xml.h
index dbd9cc4daf2e..6eab7d0cf6b5 100644
--- a/drivers/gpu/drm/msm/hdmi/qfprom.xml.h
+++ b/drivers/gpu/drm/msm/hdmi/qfprom.xml.h
@@ -9,7 +9,7 @@ git clone https://github.com/freedreno/envytools.git
 
 The rules-ng-ng source files this header was generated from are:
 - /home/robclark/src/freedreno/envytools/rnndb/msm.xml                 (    676 bytes, from 2015-05-20 20:03:14)
-- /home/robclark/src/freedreno/envytools/rnndb/freedreno_copyright.xml (   1453 bytes, from 2015-05-20 20:03:07)
+- /home/robclark/src/freedreno/envytools/rnndb/freedreno_copyright.xml (   1572 bytes, from 2016-02-10 17:07:21)
 - /home/robclark/src/freedreno/envytools/rnndb/mdp/mdp4.xml            (  20915 bytes, from 2015-05-20 20:03:14)
 - /home/robclark/src/freedreno/envytools/rnndb/mdp/mdp_common.xml      (   2849 bytes, from 2015-09-18 12:07:28)
 - /home/robclark/src/freedreno/envytools/rnndb/mdp/mdp5.xml            (  37194 bytes, from 2015-09-18 12:07:28)
@@ -17,11 +17,12 @@ The rules-ng-ng source files this header was generated from are:
 - /home/robclark/src/freedreno/envytools/rnndb/dsi/sfpb.xml            (    602 bytes, from 2015-10-22 16:35:02)
 - /home/robclark/src/freedreno/envytools/rnndb/dsi/mmss_cc.xml         (   1686 bytes, from 2015-05-20 20:03:14)
 - /home/robclark/src/freedreno/envytools/rnndb/hdmi/qfprom.xml         (    600 bytes, from 2015-05-20 20:03:07)
-- /home/robclark/src/freedreno/envytools/rnndb/hdmi/hdmi.xml           (  29154 bytes, from 2015-08-10 21:25:43)
+- /home/robclark/src/freedreno/envytools/rnndb/hdmi/hdmi.xml           (  41472 bytes, from 2016-01-22 18:18:18)
 - /home/robclark/src/freedreno/envytools/rnndb/edp/edp.xml             (  10416 bytes, from 2015-05-20 20:03:14)
 
 Copyright (C) 2013-2015 by the following authors:
 - Rob Clark <robdclark@gmail.com> (robclark)
+- Ilia Mirkin <imirkin@alum.mit.edu> (imirkin)
 
 Permission is hereby granted, free of charge, to any person obtaining
 a copy of this software and associated documentation files (the
diff --git a/drivers/gpu/drm/msm/mdp/mdp4/mdp4.xml.h b/drivers/gpu/drm/msm/mdp/mdp4/mdp4.xml.h
index d5d94575fa1b..6688e79cc88e 100644
--- a/drivers/gpu/drm/msm/mdp/mdp4/mdp4.xml.h
+++ b/drivers/gpu/drm/msm/mdp/mdp4/mdp4.xml.h
@@ -9,7 +9,7 @@ git clone https://github.com/freedreno/envytools.git
 
 The rules-ng-ng source files this header was generated from are:
 - /home/robclark/src/freedreno/envytools/rnndb/msm.xml                 (    676 bytes, from 2015-05-20 20:03:14)
-- /home/robclark/src/freedreno/envytools/rnndb/freedreno_copyright.xml (   1453 bytes, from 2015-05-20 20:03:07)
+- /home/robclark/src/freedreno/envytools/rnndb/freedreno_copyright.xml (   1572 bytes, from 2016-02-10 17:07:21)
 - /home/robclark/src/freedreno/envytools/rnndb/mdp/mdp4.xml            (  20915 bytes, from 2015-05-20 20:03:14)
 - /home/robclark/src/freedreno/envytools/rnndb/mdp/mdp_common.xml      (   2849 bytes, from 2015-09-18 12:07:28)
 - /home/robclark/src/freedreno/envytools/rnndb/mdp/mdp5.xml            (  37194 bytes, from 2015-09-18 12:07:28)
@@ -17,11 +17,12 @@ The rules-ng-ng source files this header was generated from are:
 - /home/robclark/src/freedreno/envytools/rnndb/dsi/sfpb.xml            (    602 bytes, from 2015-10-22 16:35:02)
 - /home/robclark/src/freedreno/envytools/rnndb/dsi/mmss_cc.xml         (   1686 bytes, from 2015-05-20 20:03:14)
 - /home/robclark/src/freedreno/envytools/rnndb/hdmi/qfprom.xml         (    600 bytes, from 2015-05-20 20:03:07)
-- /home/robclark/src/freedreno/envytools/rnndb/hdmi/hdmi.xml           (  29154 bytes, from 2015-08-10 21:25:43)
+- /home/robclark/src/freedreno/envytools/rnndb/hdmi/hdmi.xml           (  41472 bytes, from 2016-01-22 18:18:18)
 - /home/robclark/src/freedreno/envytools/rnndb/edp/edp.xml             (  10416 bytes, from 2015-05-20 20:03:14)
 
 Copyright (C) 2013-2015 by the following authors:
 - Rob Clark <robdclark@gmail.com> (robclark)
+- Ilia Mirkin <imirkin@alum.mit.edu> (imirkin)
 
 Permission is hereby granted, free of charge, to any person obtaining
 a copy of this software and associated documentation files (the
diff --git a/drivers/gpu/drm/msm/mdp/mdp4/mdp4_crtc.c b/drivers/gpu/drm/msm/mdp/mdp4/mdp4_crtc.c
index 38329a617da0..e233acf52334 100644
--- a/drivers/gpu/drm/msm/mdp/mdp4/mdp4_crtc.c
+++ b/drivers/gpu/drm/msm/mdp/mdp4/mdp4_crtc.c
@@ -354,13 +354,6 @@ static void mdp4_crtc_atomic_flush(struct drm_crtc *crtc,
 	request_pending(crtc, PENDING_FLIP);
 }
 
-static int mdp4_crtc_set_property(struct drm_crtc *crtc,
-		struct drm_property *property, uint64_t val)
-{
-	// XXX
-	return -EINVAL;
-}
-
 #define CURSOR_WIDTH 64
 #define CURSOR_HEIGHT 64
 
@@ -492,7 +485,7 @@ static const struct drm_crtc_funcs mdp4_crtc_funcs = {
 	.set_config = drm_atomic_helper_set_config,
 	.destroy = mdp4_crtc_destroy,
 	.page_flip = drm_atomic_helper_page_flip,
-	.set_property = mdp4_crtc_set_property,
+	.set_property = drm_atomic_helper_crtc_set_property,
 	.cursor_set = mdp4_crtc_cursor_set,
 	.cursor_move = mdp4_crtc_cursor_move,
 	.reset = drm_atomic_helper_crtc_reset,
diff --git a/drivers/gpu/drm/msm/mdp/mdp4/mdp4_kms.c b/drivers/gpu/drm/msm/mdp/mdp4/mdp4_kms.c
index 1c8e330f8d98..76e1dfb5d25e 100644
--- a/drivers/gpu/drm/msm/mdp/mdp4/mdp4_kms.c
+++ b/drivers/gpu/drm/msm/mdp/mdp4/mdp4_kms.c
@@ -179,9 +179,20 @@ static long mdp4_round_pixclk(struct msm_kms *kms, unsigned long rate,
 	}
 }
 
+static const char * const iommu_ports[] = {
+	"mdp_port0_cb0", "mdp_port1_cb0",
+};
+
 static void mdp4_destroy(struct msm_kms *kms)
 {
 	struct mdp4_kms *mdp4_kms = to_mdp4_kms(to_mdp_kms(kms));
+	struct msm_mmu *mmu = mdp4_kms->mmu;
+
+	if (mmu) {
+		mmu->funcs->detach(mmu, iommu_ports, ARRAY_SIZE(iommu_ports));
+		mmu->funcs->destroy(mmu);
+	}
+
 	if (mdp4_kms->blank_cursor_iova)
 		msm_gem_put_iova(mdp4_kms->blank_cursor_bo, mdp4_kms->id);
 	if (mdp4_kms->blank_cursor_bo)
@@ -315,7 +326,7 @@ static int mdp4_modeset_init_intf(struct mdp4_kms *mdp4_kms,
 
 		if (priv->hdmi) {
 			/* Construct bridge/connector for HDMI: */
-			ret = hdmi_modeset_init(priv->hdmi, dev, encoder);
+			ret = msm_hdmi_modeset_init(priv->hdmi, dev, encoder);
 			if (ret) {
 				dev_err(dev->dev, "failed to initialize HDMI: %d\n", ret);
 				return ret;
@@ -446,10 +457,6 @@ fail:
 	return ret;
 }
 
-static const char *iommu_ports[] = {
-		"mdp_port0_cb0", "mdp_port1_cb0",
-};
-
 struct msm_kms *mdp4_kms_init(struct drm_device *dev)
 {
 	struct platform_device *pdev = dev->platformdev;
@@ -554,6 +561,8 @@ struct msm_kms *mdp4_kms_init(struct drm_device *dev)
 				ARRAY_SIZE(iommu_ports));
 		if (ret)
 			goto fail;
+
+		mdp4_kms->mmu = mmu;
 	} else {
 		dev_info(dev->dev, "no iommu, fallback to phys "
 				"contig buffers for scanout\n");
diff --git a/drivers/gpu/drm/msm/mdp/mdp4/mdp4_kms.h b/drivers/gpu/drm/msm/mdp/mdp4/mdp4_kms.h
index 9ec53b464662..b2828717be2a 100644
--- a/drivers/gpu/drm/msm/mdp/mdp4/mdp4_kms.h
+++ b/drivers/gpu/drm/msm/mdp/mdp4/mdp4_kms.h
@@ -45,6 +45,7 @@ struct mdp4_kms {
 	struct clk *pclk;
 	struct clk *lut_clk;
 	struct clk *axi_clk;
+	struct msm_mmu *mmu;
 
 	struct mdp_irq error_handler;
 
diff --git a/drivers/gpu/drm/msm/mdp/mdp5/mdp5.xml.h b/drivers/gpu/drm/msm/mdp/mdp5/mdp5.xml.h
index c37da9c61e29..b275ce11b24b 100644
--- a/drivers/gpu/drm/msm/mdp/mdp5/mdp5.xml.h
+++ b/drivers/gpu/drm/msm/mdp/mdp5/mdp5.xml.h
@@ -9,7 +9,7 @@ git clone https://github.com/freedreno/envytools.git
 
 The rules-ng-ng source files this header was generated from are:
 - /home/robclark/src/freedreno/envytools/rnndb/msm.xml                 (    676 bytes, from 2015-05-20 20:03:14)
-- /home/robclark/src/freedreno/envytools/rnndb/freedreno_copyright.xml (   1453 bytes, from 2015-05-20 20:03:07)
+- /home/robclark/src/freedreno/envytools/rnndb/freedreno_copyright.xml (   1572 bytes, from 2016-02-10 17:07:21)
 - /home/robclark/src/freedreno/envytools/rnndb/mdp/mdp4.xml            (  20915 bytes, from 2015-05-20 20:03:14)
 - /home/robclark/src/freedreno/envytools/rnndb/mdp/mdp_common.xml      (   2849 bytes, from 2015-09-18 12:07:28)
 - /home/robclark/src/freedreno/envytools/rnndb/mdp/mdp5.xml            (  37194 bytes, from 2015-09-18 12:07:28)
@@ -17,11 +17,12 @@ The rules-ng-ng source files this header was generated from are:
 - /home/robclark/src/freedreno/envytools/rnndb/dsi/sfpb.xml            (    602 bytes, from 2015-10-22 16:35:02)
 - /home/robclark/src/freedreno/envytools/rnndb/dsi/mmss_cc.xml         (   1686 bytes, from 2015-05-20 20:03:14)
 - /home/robclark/src/freedreno/envytools/rnndb/hdmi/qfprom.xml         (    600 bytes, from 2015-05-20 20:03:07)
-- /home/robclark/src/freedreno/envytools/rnndb/hdmi/hdmi.xml           (  29154 bytes, from 2015-08-10 21:25:43)
+- /home/robclark/src/freedreno/envytools/rnndb/hdmi/hdmi.xml           (  41472 bytes, from 2016-01-22 18:18:18)
 - /home/robclark/src/freedreno/envytools/rnndb/edp/edp.xml             (  10416 bytes, from 2015-05-20 20:03:14)
 
 Copyright (C) 2013-2015 by the following authors:
 - Rob Clark <robdclark@gmail.com> (robclark)
+- Ilia Mirkin <imirkin@alum.mit.edu> (imirkin)
 
 Permission is hereby granted, free of charge, to any person obtaining
 a copy of this software and associated documentation files (the
diff --git a/drivers/gpu/drm/msm/mdp/mdp5/mdp5_crtc.c b/drivers/gpu/drm/msm/mdp/mdp5/mdp5_crtc.c
index d6b45eb57d4f..9673b9520b6a 100644
--- a/drivers/gpu/drm/msm/mdp/mdp5/mdp5_crtc.c
+++ b/drivers/gpu/drm/msm/mdp/mdp5/mdp5_crtc.c
@@ -461,13 +461,6 @@ static void mdp5_crtc_atomic_flush(struct drm_crtc *crtc,
 	request_pending(crtc, PENDING_FLIP);
 }
 
-static int mdp5_crtc_set_property(struct drm_crtc *crtc,
-		struct drm_property *property, uint64_t val)
-{
-	// XXX
-	return -EINVAL;
-}
-
 static void get_roi(struct drm_crtc *crtc, uint32_t *roi_w, uint32_t *roi_h)
 {
 	struct mdp5_crtc *mdp5_crtc = to_mdp5_crtc(crtc);
@@ -618,7 +611,7 @@ static const struct drm_crtc_funcs mdp5_crtc_funcs = {
 	.set_config = drm_atomic_helper_set_config,
 	.destroy = mdp5_crtc_destroy,
 	.page_flip = drm_atomic_helper_page_flip,
-	.set_property = mdp5_crtc_set_property,
+	.set_property = drm_atomic_helper_crtc_set_property,
 	.reset = drm_atomic_helper_crtc_reset,
 	.atomic_duplicate_state = drm_atomic_helper_crtc_duplicate_state,
 	.atomic_destroy_state = drm_atomic_helper_crtc_destroy_state,
diff --git a/drivers/gpu/drm/msm/mdp/mdp5/mdp5_kms.c b/drivers/gpu/drm/msm/mdp/mdp5/mdp5_kms.c
index 5e4d16b399c7..484b4d15e71d 100644
--- a/drivers/gpu/drm/msm/mdp/mdp5/mdp5_kms.c
+++ b/drivers/gpu/drm/msm/mdp/mdp5/mdp5_kms.c
@@ -284,7 +284,7 @@ static int modeset_init_intf(struct mdp5_kms *mdp5_kms, int intf_num)
 			break;
 		}
 
-		ret = hdmi_modeset_init(priv->hdmi, dev, encoder);
+		ret = msm_hdmi_modeset_init(priv->hdmi, dev, encoder);
 		break;
 	case INTF_DSI:
 	{
diff --git a/drivers/gpu/drm/msm/mdp/mdp_common.xml.h b/drivers/gpu/drm/msm/mdp/mdp_common.xml.h
index 0aec1ac1f6d0..452e3518f98b 100644
--- a/drivers/gpu/drm/msm/mdp/mdp_common.xml.h
+++ b/drivers/gpu/drm/msm/mdp/mdp_common.xml.h
@@ -9,7 +9,7 @@ git clone https://github.com/freedreno/envytools.git
 
 The rules-ng-ng source files this header was generated from are:
 - /home/robclark/src/freedreno/envytools/rnndb/msm.xml                 (    676 bytes, from 2015-05-20 20:03:14)
-- /home/robclark/src/freedreno/envytools/rnndb/freedreno_copyright.xml (   1453 bytes, from 2015-05-20 20:03:07)
+- /home/robclark/src/freedreno/envytools/rnndb/freedreno_copyright.xml (   1572 bytes, from 2016-02-10 17:07:21)
 - /home/robclark/src/freedreno/envytools/rnndb/mdp/mdp4.xml            (  20915 bytes, from 2015-05-20 20:03:14)
 - /home/robclark/src/freedreno/envytools/rnndb/mdp/mdp_common.xml      (   2849 bytes, from 2015-09-18 12:07:28)
 - /home/robclark/src/freedreno/envytools/rnndb/mdp/mdp5.xml            (  37194 bytes, from 2015-09-18 12:07:28)
@@ -17,11 +17,12 @@ The rules-ng-ng source files this header was generated from are:
 - /home/robclark/src/freedreno/envytools/rnndb/dsi/sfpb.xml            (    602 bytes, from 2015-10-22 16:35:02)
 - /home/robclark/src/freedreno/envytools/rnndb/dsi/mmss_cc.xml         (   1686 bytes, from 2015-05-20 20:03:14)
 - /home/robclark/src/freedreno/envytools/rnndb/hdmi/qfprom.xml         (    600 bytes, from 2015-05-20 20:03:07)
-- /home/robclark/src/freedreno/envytools/rnndb/hdmi/hdmi.xml           (  29154 bytes, from 2015-08-10 21:25:43)
+- /home/robclark/src/freedreno/envytools/rnndb/hdmi/hdmi.xml           (  41472 bytes, from 2016-01-22 18:18:18)
 - /home/robclark/src/freedreno/envytools/rnndb/edp/edp.xml             (  10416 bytes, from 2015-05-20 20:03:14)
 
 Copyright (C) 2013-2015 by the following authors:
 - Rob Clark <robdclark@gmail.com> (robclark)
+- Ilia Mirkin <imirkin@alum.mit.edu> (imirkin)
 
 Permission is hereby granted, free of charge, to any person obtaining
 a copy of this software and associated documentation files (the
diff --git a/drivers/gpu/drm/msm/msm_drv.c b/drivers/gpu/drm/msm/msm_drv.c
index 9a30807b900b..d52910e2c26c 100644
--- a/drivers/gpu/drm/msm/msm_drv.c
+++ b/drivers/gpu/drm/msm/msm_drv.c
@@ -61,7 +61,7 @@ module_param(fbdev, bool, 0600);
 #endif
 
 static char *vram = "16m";
-MODULE_PARM_DESC(vram, "Configure VRAM size (for devices without IOMMU/GPUMMU");
+MODULE_PARM_DESC(vram, "Configure VRAM size (for devices without IOMMU/GPUMMU)");
 module_param(vram, charp, 0);
 
 /*
@@ -196,6 +196,11 @@ static int msm_unload(struct drm_device *dev)
 	}
 
 	drm_kms_helper_poll_fini(dev);
+
+#ifdef CONFIG_DRM_FBDEV_EMULATION
+	if (fbdev && priv->fbdev)
+		msm_fbdev_free(dev);
+#endif
 	drm_mode_config_cleanup(dev);
 	drm_vblank_cleanup(dev);
 
@@ -1116,7 +1121,7 @@ static int __init msm_drm_register(void)
 	DBG("init");
 	msm_dsi_register();
 	msm_edp_register();
-	hdmi_register();
+	msm_hdmi_register();
 	adreno_register();
 	return platform_driver_register(&msm_platform_driver);
 }
@@ -1125,7 +1130,7 @@ static void __exit msm_drm_unregister(void)
 {
 	DBG("fini");
 	platform_driver_unregister(&msm_platform_driver);
-	hdmi_unregister();
+	msm_hdmi_unregister();
 	adreno_unregister();
 	msm_edp_unregister();
 	msm_dsi_unregister();
diff --git a/drivers/gpu/drm/msm/msm_drv.h b/drivers/gpu/drm/msm/msm_drv.h
index c1e7bba2fdb7..870dbe58c259 100644
--- a/drivers/gpu/drm/msm/msm_drv.h
+++ b/drivers/gpu/drm/msm/msm_drv.h
@@ -240,12 +240,13 @@ struct drm_framebuffer *msm_framebuffer_create(struct drm_device *dev,
 		struct drm_file *file, const struct drm_mode_fb_cmd2 *mode_cmd);
 
 struct drm_fb_helper *msm_fbdev_init(struct drm_device *dev);
+void msm_fbdev_free(struct drm_device *dev);
 
 struct hdmi;
-int hdmi_modeset_init(struct hdmi *hdmi, struct drm_device *dev,
+int msm_hdmi_modeset_init(struct hdmi *hdmi, struct drm_device *dev,
 		struct drm_encoder *encoder);
-void __init hdmi_register(void);
-void __exit hdmi_unregister(void);
+void __init msm_hdmi_register(void);
+void __exit msm_hdmi_unregister(void);
 
 struct msm_edp;
 void __init msm_edp_register(void);
diff --git a/drivers/gpu/drm/msm/msm_gem_submit.c b/drivers/gpu/drm/msm/msm_gem_submit.c
index 6d7cd3fe21e7..43d2181231c0 100644
--- a/drivers/gpu/drm/msm/msm_gem_submit.c
+++ b/drivers/gpu/drm/msm/msm_gem_submit.c
@@ -323,28 +323,27 @@ int msm_ioctl_gem_submit(struct drm_device *dev, void *data,
 	struct drm_msm_gem_submit *args = data;
 	struct msm_file_private *ctx = file->driver_priv;
 	struct msm_gem_submit *submit;
-	struct msm_gpu *gpu;
+	struct msm_gpu *gpu = priv->gpu;
 	unsigned i;
 	int ret;
 
+	if (!gpu)
+		return -ENXIO;
+
 	/* for now, we just have 3d pipe.. eventually this would need to
 	 * be more clever to dispatch to appropriate gpu module:
 	 */
 	if (args->pipe != MSM_PIPE_3D0)
 		return -EINVAL;
 
-	gpu = priv->gpu;
-
 	if (args->nr_cmds > MAX_CMDS)
 		return -EINVAL;
 
-	mutex_lock(&dev->struct_mutex);
-
 	submit = submit_create(dev, gpu, args->nr_bos);
-	if (!submit) {
-		ret = -ENOMEM;
-		goto out;
-	}
+	if (!submit)
+		return -ENOMEM;
+
+	mutex_lock(&dev->struct_mutex);
 
 	ret = submit_lookup_objects(submit, args, file);
 	if (ret)
@@ -419,8 +418,7 @@ int msm_ioctl_gem_submit(struct drm_device *dev, void *data,
 	args->fence = submit->fence;
 
 out:
-	if (submit)
-		submit_cleanup(submit, !!ret);
+	submit_cleanup(submit, !!ret);
 	mutex_unlock(&dev->struct_mutex);
 	return ret;
 }
diff --git a/drivers/gpu/drm/msm/msm_iommu.c b/drivers/gpu/drm/msm/msm_iommu.c
index 7ac2f1997e4a..a7a0b6d9b057 100644
--- a/drivers/gpu/drm/msm/msm_iommu.c
+++ b/drivers/gpu/drm/msm/msm_iommu.c
@@ -31,13 +31,15 @@ static int msm_fault_handler(struct iommu_domain *iommu, struct device *dev,
 	return 0;
 }
 
-static int msm_iommu_attach(struct msm_mmu *mmu, const char **names, int cnt)
+static int msm_iommu_attach(struct msm_mmu *mmu, const char * const *names,
+			    int cnt)
 {
 	struct msm_iommu *iommu = to_msm_iommu(mmu);
 	return iommu_attach_device(iommu->domain, mmu->dev);
 }
 
-static void msm_iommu_detach(struct msm_mmu *mmu, const char **names, int cnt)
+static void msm_iommu_detach(struct msm_mmu *mmu, const char * const *names,
+			     int cnt)
 {
 	struct msm_iommu *iommu = to_msm_iommu(mmu);
 	iommu_detach_device(iommu->domain, mmu->dev);
diff --git a/drivers/gpu/drm/msm/msm_mmu.h b/drivers/gpu/drm/msm/msm_mmu.h
index 7cd88d9dc155..b8ca9a0e9170 100644
--- a/drivers/gpu/drm/msm/msm_mmu.h
+++ b/drivers/gpu/drm/msm/msm_mmu.h
@@ -21,8 +21,8 @@
 #include <linux/iommu.h>
 
 struct msm_mmu_funcs {
-	int (*attach)(struct msm_mmu *mmu, const char **names, int cnt);
-	void (*detach)(struct msm_mmu *mmu, const char **names, int cnt);
+	int (*attach)(struct msm_mmu *mmu, const char * const *names, int cnt);
+	void (*detach)(struct msm_mmu *mmu, const char * const *names, int cnt);
 	int (*map)(struct msm_mmu *mmu, uint32_t iova, struct sg_table *sgt,
 			unsigned len, int prot);
 	int (*unmap)(struct msm_mmu *mmu, uint32_t iova, struct sg_table *sgt,
diff --git a/drivers/gpu/drm/nouveau/include/nvif/cla06f.h b/drivers/gpu/drm/nouveau/include/nvif/cla06f.h
index 85b7827eb782..46301ec018ce 100644
--- a/drivers/gpu/drm/nouveau/include/nvif/cla06f.h
+++ b/drivers/gpu/drm/nouveau/include/nvif/cla06f.h
@@ -3,19 +3,27 @@
 
 struct kepler_channel_gpfifo_a_v0 {
 	__u8  version;
-#define KEPLER_CHANNEL_GPFIFO_A_V0_ENGINE_GR                               0x01
-#define KEPLER_CHANNEL_GPFIFO_A_V0_ENGINE_MSPDEC                           0x02
-#define KEPLER_CHANNEL_GPFIFO_A_V0_ENGINE_MSPPP                            0x04
-#define KEPLER_CHANNEL_GPFIFO_A_V0_ENGINE_MSVLD                            0x08
-#define KEPLER_CHANNEL_GPFIFO_A_V0_ENGINE_CE0                              0x10
-#define KEPLER_CHANNEL_GPFIFO_A_V0_ENGINE_CE1                              0x20
-#define KEPLER_CHANNEL_GPFIFO_A_V0_ENGINE_ENC                              0x40
-	__u8  engine;
+	__u8  pad01[5];
 	__u16 chid;
+#define NVA06F_V0_ENGINE_SW                                          0x00000001
+#define NVA06F_V0_ENGINE_GR                                          0x00000002
+#define NVA06F_V0_ENGINE_SEC                                         0x00000004
+#define NVA06F_V0_ENGINE_MSVLD                                       0x00000010
+#define NVA06F_V0_ENGINE_MSPDEC                                      0x00000020
+#define NVA06F_V0_ENGINE_MSPPP                                       0x00000040
+#define NVA06F_V0_ENGINE_MSENC                                       0x00000080
+#define NVA06F_V0_ENGINE_VIC                                         0x00000100
+#define NVA06F_V0_ENGINE_NVDEC                                       0x00000200
+#define NVA06F_V0_ENGINE_NVENC0                                      0x00000400
+#define NVA06F_V0_ENGINE_NVENC1                                      0x00000800
+#define NVA06F_V0_ENGINE_CE0                                         0x00010000
+#define NVA06F_V0_ENGINE_CE1                                         0x00020000
+#define NVA06F_V0_ENGINE_CE2                                         0x00040000
+	__u32 engines;
 	__u32 ilength;
 	__u64 ioffset;
 	__u64 vm;
 };
 
-#define KEPLER_CHANNEL_GPFIFO_A_V0_NTFY_UEVENT                             0x00
+#define NVA06F_V0_NTFY_UEVENT                                              0x00
 #endif
diff --git a/drivers/gpu/drm/nouveau/include/nvif/class.h b/drivers/gpu/drm/nouveau/include/nvif/class.h
index 4179cd65ac0a..982aad8fa645 100644
--- a/drivers/gpu/drm/nouveau/include/nvif/class.h
+++ b/drivers/gpu/drm/nouveau/include/nvif/class.h
@@ -37,6 +37,7 @@
 #define G82_CHANNEL_GPFIFO                            /* cl826f.h */ 0x0000826f
 #define FERMI_CHANNEL_GPFIFO                          /* cl906f.h */ 0x0000906f
 #define KEPLER_CHANNEL_GPFIFO_A                       /* cla06f.h */ 0x0000a06f
+#define KEPLER_CHANNEL_GPFIFO_B                       /* cla06f.h */ 0x0000a16f
 #define MAXWELL_CHANNEL_GPFIFO_A                      /* cla06f.h */ 0x0000b06f
 
 #define NV50_DISP                                     /* cl5070.h */ 0x00005070
@@ -48,7 +49,7 @@
 #define GK104_DISP                                    /* cl5070.h */ 0x00009170
 #define GK110_DISP                                    /* cl5070.h */ 0x00009270
 #define GM107_DISP                                    /* cl5070.h */ 0x00009470
-#define GM204_DISP                                    /* cl5070.h */ 0x00009570
+#define GM200_DISP                                    /* cl5070.h */ 0x00009570
 
 #define NV31_MPEG                                                    0x00003174
 #define G82_MPEG                                                     0x00008274
@@ -84,7 +85,7 @@
 #define GK104_DISP_CORE_CHANNEL_DMA                   /* cl507d.h */ 0x0000917d
 #define GK110_DISP_CORE_CHANNEL_DMA                   /* cl507d.h */ 0x0000927d
 #define GM107_DISP_CORE_CHANNEL_DMA                   /* cl507d.h */ 0x0000947d
-#define GM204_DISP_CORE_CHANNEL_DMA                   /* cl507d.h */ 0x0000957d
+#define GM200_DISP_CORE_CHANNEL_DMA                   /* cl507d.h */ 0x0000957d
 
 #define NV50_DISP_OVERLAY_CHANNEL_DMA                 /* cl507e.h */ 0x0000507e
 #define G82_DISP_OVERLAY_CHANNEL_DMA                  /* cl507e.h */ 0x0000827e
diff --git a/drivers/gpu/drm/nouveau/include/nvif/device.h b/drivers/gpu/drm/nouveau/include/nvif/device.h
index e0ed2f4b2f43..bcb981711617 100644
--- a/drivers/gpu/drm/nouveau/include/nvif/device.h
+++ b/drivers/gpu/drm/nouveau/include/nvif/device.h
@@ -62,6 +62,7 @@ u64  nvif_device_time(struct nvif_device *);
 #define nvxx_gpio(a) nvxx_device(a)->gpio
 #define nvxx_clk(a) nvxx_device(a)->clk
 #define nvxx_i2c(a) nvxx_device(a)->i2c
+#define nvxx_iccsense(a) nvxx_device(a)->iccsense
 #define nvxx_therm(a) nvxx_device(a)->therm
 #define nvxx_volt(a) nvxx_device(a)->volt
 
diff --git a/drivers/gpu/drm/nouveau/include/nvkm/core/device.h b/drivers/gpu/drm/nouveau/include/nvkm/core/device.h
index 913192c94876..4993a863adb9 100644
--- a/drivers/gpu/drm/nouveau/include/nvkm/core/device.h
+++ b/drivers/gpu/drm/nouveau/include/nvkm/core/device.h
@@ -22,30 +22,41 @@ enum nvkm_devidx {
 	NVKM_SUBDEV_BAR,
 	NVKM_SUBDEV_PMU,
 	NVKM_SUBDEV_VOLT,
+	NVKM_SUBDEV_ICCSENSE,
 	NVKM_SUBDEV_THERM,
 	NVKM_SUBDEV_CLK,
+	NVKM_SUBDEV_SECBOOT,
 
-	NVKM_ENGINE_DMAOBJ,
-	NVKM_ENGINE_IFB,
-	NVKM_ENGINE_FIFO,
-	NVKM_ENGINE_SW,
-	NVKM_ENGINE_GR,
-	NVKM_ENGINE_MPEG,
-	NVKM_ENGINE_ME,
-	NVKM_ENGINE_VP,
-	NVKM_ENGINE_CIPHER,
 	NVKM_ENGINE_BSP,
-	NVKM_ENGINE_MSPPP,
+
 	NVKM_ENGINE_CE0,
 	NVKM_ENGINE_CE1,
 	NVKM_ENGINE_CE2,
-	NVKM_ENGINE_VIC,
-	NVKM_ENGINE_MSENC,
+	NVKM_ENGINE_CE_LAST = NVKM_ENGINE_CE2,
+
+	NVKM_ENGINE_CIPHER,
 	NVKM_ENGINE_DISP,
-	NVKM_ENGINE_PM,
+	NVKM_ENGINE_DMAOBJ,
+	NVKM_ENGINE_FIFO,
+	NVKM_ENGINE_GR,
+	NVKM_ENGINE_IFB,
+	NVKM_ENGINE_ME,
+	NVKM_ENGINE_MPEG,
+	NVKM_ENGINE_MSENC,
+	NVKM_ENGINE_MSPDEC,
+	NVKM_ENGINE_MSPPP,
 	NVKM_ENGINE_MSVLD,
+
+	NVKM_ENGINE_NVENC0,
+	NVKM_ENGINE_NVENC1,
+	NVKM_ENGINE_NVENC_LAST = NVKM_ENGINE_NVENC1,
+
+	NVKM_ENGINE_NVDEC,
+	NVKM_ENGINE_PM,
 	NVKM_ENGINE_SEC,
-	NVKM_ENGINE_MSPDEC,
+	NVKM_ENGINE_SW,
+	NVKM_ENGINE_VIC,
+	NVKM_ENGINE_VP,
 
 	NVKM_SUBDEV_NR
 };
@@ -109,6 +120,7 @@ struct nvkm_device {
 	struct nvkm_gpio *gpio;
 	struct nvkm_i2c *i2c;
 	struct nvkm_subdev *ibus;
+	struct nvkm_iccsense *iccsense;
 	struct nvkm_instmem *imem;
 	struct nvkm_ltc *ltc;
 	struct nvkm_mc *mc;
@@ -116,6 +128,7 @@ struct nvkm_device {
 	struct nvkm_subdev *mxm;
 	struct nvkm_pci *pci;
 	struct nvkm_pmu *pmu;
+	struct nvkm_secboot *secboot;
 	struct nvkm_therm *therm;
 	struct nvkm_timer *timer;
 	struct nvkm_volt *volt;
@@ -134,6 +147,8 @@ struct nvkm_device {
 	struct nvkm_engine *mspdec;
 	struct nvkm_engine *msppp;
 	struct nvkm_engine *msvld;
+	struct nvkm_engine *nvenc[2];
+	struct nvkm_engine *nvdec;
 	struct nvkm_pm *pm;
 	struct nvkm_engine *sec;
 	struct nvkm_sw *sw;
@@ -164,46 +179,50 @@ struct nvkm_device_quirk {
 struct nvkm_device_chip {
 	const char *name;
 
-	int (*bar    )(struct nvkm_device *, int idx, struct nvkm_bar **);
-	int (*bios   )(struct nvkm_device *, int idx, struct nvkm_bios **);
-	int (*bus    )(struct nvkm_device *, int idx, struct nvkm_bus **);
-	int (*clk    )(struct nvkm_device *, int idx, struct nvkm_clk **);
-	int (*devinit)(struct nvkm_device *, int idx, struct nvkm_devinit **);
-	int (*fb     )(struct nvkm_device *, int idx, struct nvkm_fb **);
-	int (*fuse   )(struct nvkm_device *, int idx, struct nvkm_fuse **);
-	int (*gpio   )(struct nvkm_device *, int idx, struct nvkm_gpio **);
-	int (*i2c    )(struct nvkm_device *, int idx, struct nvkm_i2c **);
-	int (*ibus   )(struct nvkm_device *, int idx, struct nvkm_subdev **);
-	int (*imem   )(struct nvkm_device *, int idx, struct nvkm_instmem **);
-	int (*ltc    )(struct nvkm_device *, int idx, struct nvkm_ltc **);
-	int (*mc     )(struct nvkm_device *, int idx, struct nvkm_mc **);
-	int (*mmu    )(struct nvkm_device *, int idx, struct nvkm_mmu **);
-	int (*mxm    )(struct nvkm_device *, int idx, struct nvkm_subdev **);
-	int (*pci    )(struct nvkm_device *, int idx, struct nvkm_pci **);
-	int (*pmu    )(struct nvkm_device *, int idx, struct nvkm_pmu **);
-	int (*therm  )(struct nvkm_device *, int idx, struct nvkm_therm **);
-	int (*timer  )(struct nvkm_device *, int idx, struct nvkm_timer **);
-	int (*volt   )(struct nvkm_device *, int idx, struct nvkm_volt **);
-
-	int (*bsp    )(struct nvkm_device *, int idx, struct nvkm_engine **);
-	int (*ce[3]  )(struct nvkm_device *, int idx, struct nvkm_engine **);
-	int (*cipher )(struct nvkm_device *, int idx, struct nvkm_engine **);
-	int (*disp   )(struct nvkm_device *, int idx, struct nvkm_disp **);
-	int (*dma    )(struct nvkm_device *, int idx, struct nvkm_dma **);
-	int (*fifo   )(struct nvkm_device *, int idx, struct nvkm_fifo **);
-	int (*gr     )(struct nvkm_device *, int idx, struct nvkm_gr **);
-	int (*ifb    )(struct nvkm_device *, int idx, struct nvkm_engine **);
-	int (*me     )(struct nvkm_device *, int idx, struct nvkm_engine **);
-	int (*mpeg   )(struct nvkm_device *, int idx, struct nvkm_engine **);
-	int (*msenc  )(struct nvkm_device *, int idx, struct nvkm_engine **);
-	int (*mspdec )(struct nvkm_device *, int idx, struct nvkm_engine **);
-	int (*msppp  )(struct nvkm_device *, int idx, struct nvkm_engine **);
-	int (*msvld  )(struct nvkm_device *, int idx, struct nvkm_engine **);
-	int (*pm     )(struct nvkm_device *, int idx, struct nvkm_pm **);
-	int (*sec    )(struct nvkm_device *, int idx, struct nvkm_engine **);
-	int (*sw     )(struct nvkm_device *, int idx, struct nvkm_sw **);
-	int (*vic    )(struct nvkm_device *, int idx, struct nvkm_engine **);
-	int (*vp     )(struct nvkm_device *, int idx, struct nvkm_engine **);
+	int (*bar     )(struct nvkm_device *, int idx, struct nvkm_bar **);
+	int (*bios    )(struct nvkm_device *, int idx, struct nvkm_bios **);
+	int (*bus     )(struct nvkm_device *, int idx, struct nvkm_bus **);
+	int (*clk     )(struct nvkm_device *, int idx, struct nvkm_clk **);
+	int (*devinit )(struct nvkm_device *, int idx, struct nvkm_devinit **);
+	int (*fb      )(struct nvkm_device *, int idx, struct nvkm_fb **);
+	int (*fuse    )(struct nvkm_device *, int idx, struct nvkm_fuse **);
+	int (*gpio    )(struct nvkm_device *, int idx, struct nvkm_gpio **);
+	int (*i2c     )(struct nvkm_device *, int idx, struct nvkm_i2c **);
+	int (*ibus    )(struct nvkm_device *, int idx, struct nvkm_subdev **);
+	int (*iccsense)(struct nvkm_device *, int idx, struct nvkm_iccsense **);
+	int (*imem    )(struct nvkm_device *, int idx, struct nvkm_instmem **);
+	int (*ltc     )(struct nvkm_device *, int idx, struct nvkm_ltc **);
+	int (*mc      )(struct nvkm_device *, int idx, struct nvkm_mc **);
+	int (*mmu     )(struct nvkm_device *, int idx, struct nvkm_mmu **);
+	int (*mxm     )(struct nvkm_device *, int idx, struct nvkm_subdev **);
+	int (*pci     )(struct nvkm_device *, int idx, struct nvkm_pci **);
+	int (*pmu     )(struct nvkm_device *, int idx, struct nvkm_pmu **);
+	int (*secboot )(struct nvkm_device *, int idx, struct nvkm_secboot **);
+	int (*therm   )(struct nvkm_device *, int idx, struct nvkm_therm **);
+	int (*timer   )(struct nvkm_device *, int idx, struct nvkm_timer **);
+	int (*volt    )(struct nvkm_device *, int idx, struct nvkm_volt **);
+
+	int (*bsp     )(struct nvkm_device *, int idx, struct nvkm_engine **);
+	int (*ce[3]   )(struct nvkm_device *, int idx, struct nvkm_engine **);
+	int (*cipher  )(struct nvkm_device *, int idx, struct nvkm_engine **);
+	int (*disp    )(struct nvkm_device *, int idx, struct nvkm_disp **);
+	int (*dma     )(struct nvkm_device *, int idx, struct nvkm_dma **);
+	int (*fifo    )(struct nvkm_device *, int idx, struct nvkm_fifo **);
+	int (*gr      )(struct nvkm_device *, int idx, struct nvkm_gr **);
+	int (*ifb     )(struct nvkm_device *, int idx, struct nvkm_engine **);
+	int (*me      )(struct nvkm_device *, int idx, struct nvkm_engine **);
+	int (*mpeg    )(struct nvkm_device *, int idx, struct nvkm_engine **);
+	int (*msenc   )(struct nvkm_device *, int idx, struct nvkm_engine **);
+	int (*mspdec  )(struct nvkm_device *, int idx, struct nvkm_engine **);
+	int (*msppp   )(struct nvkm_device *, int idx, struct nvkm_engine **);
+	int (*msvld   )(struct nvkm_device *, int idx, struct nvkm_engine **);
+	int (*nvenc[2])(struct nvkm_device *, int idx, struct nvkm_engine **);
+	int (*nvdec   )(struct nvkm_device *, int idx, struct nvkm_engine **);
+	int (*pm      )(struct nvkm_device *, int idx, struct nvkm_pm **);
+	int (*sec     )(struct nvkm_device *, int idx, struct nvkm_engine **);
+	int (*sw      )(struct nvkm_device *, int idx, struct nvkm_sw **);
+	int (*vic     )(struct nvkm_device *, int idx, struct nvkm_engine **);
+	int (*vp      )(struct nvkm_device *, int idx, struct nvkm_engine **);
 };
 
 struct nvkm_device *nvkm_device_find(u64 name);
diff --git a/drivers/gpu/drm/nouveau/include/nvkm/core/firmware.h b/drivers/gpu/drm/nouveau/include/nvkm/core/firmware.h
new file mode 100644
index 000000000000..a626ce378f04
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/include/nvkm/core/firmware.h
@@ -0,0 +1,11 @@
+#ifndef __NVKM_FIRMWARE_H__
+#define __NVKM_FIRMWARE_H__
+
+#include <core/device.h>
+
+int nvkm_firmware_get(struct nvkm_device *device, const char *fwname,
+		      const struct firmware **fw);
+
+void nvkm_firmware_put(const struct firmware *fw);
+
+#endif
diff --git a/drivers/gpu/drm/nouveau/include/nvkm/core/gpuobj.h b/drivers/gpu/drm/nouveau/include/nvkm/core/gpuobj.h
index d4f56eafb073..c23da4f05929 100644
--- a/drivers/gpu/drm/nouveau/include/nvkm/core/gpuobj.h
+++ b/drivers/gpu/drm/nouveau/include/nvkm/core/gpuobj.h
@@ -37,4 +37,8 @@ int nvkm_gpuobj_wrap(struct nvkm_memory *, struct nvkm_gpuobj **);
 int nvkm_gpuobj_map(struct nvkm_gpuobj *, struct nvkm_vm *, u32 access,
 		    struct nvkm_vma *);
 void nvkm_gpuobj_unmap(struct nvkm_vma *);
+void nvkm_gpuobj_memcpy_to(struct nvkm_gpuobj *dst, u32 dstoffset, void *src,
+			   u32 length);
+void nvkm_gpuobj_memcpy_from(void *dst, struct nvkm_gpuobj *src, u32 srcoffset,
+			     u32 length);
 #endif
diff --git a/drivers/gpu/drm/nouveau/include/nvkm/engine/ce.h b/drivers/gpu/drm/nouveau/include/nvkm/engine/ce.h
index e2e22cd5305b..594d719ba41e 100644
--- a/drivers/gpu/drm/nouveau/include/nvkm/engine/ce.h
+++ b/drivers/gpu/drm/nouveau/include/nvkm/engine/ce.h
@@ -5,5 +5,6 @@
 int gt215_ce_new(struct nvkm_device *, int, struct nvkm_engine **);
 int gf100_ce_new(struct nvkm_device *, int, struct nvkm_engine **);
 int gk104_ce_new(struct nvkm_device *, int, struct nvkm_engine **);
-int gm204_ce_new(struct nvkm_device *, int, struct nvkm_engine **);
+int gm107_ce_new(struct nvkm_device *, int, struct nvkm_engine **);
+int gm200_ce_new(struct nvkm_device *, int, struct nvkm_engine **);
 #endif
diff --git a/drivers/gpu/drm/nouveau/include/nvkm/engine/disp.h b/drivers/gpu/drm/nouveau/include/nvkm/engine/disp.h
index efc74d03346b..d4fdce27b297 100644
--- a/drivers/gpu/drm/nouveau/include/nvkm/engine/disp.h
+++ b/drivers/gpu/drm/nouveau/include/nvkm/engine/disp.h
@@ -31,5 +31,5 @@ int gf119_disp_new(struct nvkm_device *, int, struct nvkm_disp **);
 int gk104_disp_new(struct nvkm_device *, int, struct nvkm_disp **);
 int gk110_disp_new(struct nvkm_device *, int, struct nvkm_disp **);
 int gm107_disp_new(struct nvkm_device *, int, struct nvkm_disp **);
-int gm204_disp_new(struct nvkm_device *, int, struct nvkm_disp **);
+int gm200_disp_new(struct nvkm_device *, int, struct nvkm_disp **);
 #endif
diff --git a/drivers/gpu/drm/nouveau/include/nvkm/engine/fifo.h b/drivers/gpu/drm/nouveau/include/nvkm/engine/fifo.h
index 9e6644955d19..15ddfcf5e8db 100644
--- a/drivers/gpu/drm/nouveau/include/nvkm/engine/fifo.h
+++ b/drivers/gpu/drm/nouveau/include/nvkm/engine/fifo.h
@@ -60,8 +60,10 @@ int nv50_fifo_new(struct nvkm_device *, int, struct nvkm_fifo **);
 int g84_fifo_new(struct nvkm_device *, int, struct nvkm_fifo **);
 int gf100_fifo_new(struct nvkm_device *, int, struct nvkm_fifo **);
 int gk104_fifo_new(struct nvkm_device *, int, struct nvkm_fifo **);
+int gk110_fifo_new(struct nvkm_device *, int, struct nvkm_fifo **);
 int gk208_fifo_new(struct nvkm_device *, int, struct nvkm_fifo **);
 int gk20a_fifo_new(struct nvkm_device *, int, struct nvkm_fifo **);
-int gm204_fifo_new(struct nvkm_device *, int, struct nvkm_fifo **);
+int gm107_fifo_new(struct nvkm_device *, int, struct nvkm_fifo **);
+int gm200_fifo_new(struct nvkm_device *, int, struct nvkm_fifo **);
 int gm20b_fifo_new(struct nvkm_device *, int, struct nvkm_fifo **);
 #endif
diff --git a/drivers/gpu/drm/nouveau/include/nvkm/engine/gr.h b/drivers/gpu/drm/nouveau/include/nvkm/engine/gr.h
index f126e54d2e30..6515f5810a26 100644
--- a/drivers/gpu/drm/nouveau/include/nvkm/engine/gr.h
+++ b/drivers/gpu/drm/nouveau/include/nvkm/engine/gr.h
@@ -40,7 +40,6 @@ int gk110b_gr_new(struct nvkm_device *, int, struct nvkm_gr **);
 int gk208_gr_new(struct nvkm_device *, int, struct nvkm_gr **);
 int gk20a_gr_new(struct nvkm_device *, int, struct nvkm_gr **);
 int gm107_gr_new(struct nvkm_device *, int, struct nvkm_gr **);
-int gm204_gr_new(struct nvkm_device *, int, struct nvkm_gr **);
-int gm206_gr_new(struct nvkm_device *, int, struct nvkm_gr **);
+int gm200_gr_new(struct nvkm_device *, int, struct nvkm_gr **);
 int gm20b_gr_new(struct nvkm_device *, int, struct nvkm_gr **);
 #endif
diff --git a/drivers/gpu/drm/nouveau/include/nvkm/engine/msenc.h b/drivers/gpu/drm/nouveau/include/nvkm/engine/msenc.h
new file mode 100644
index 000000000000..748ea9b7e559
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/include/nvkm/engine/msenc.h
@@ -0,0 +1,4 @@
+#ifndef __NVKM_MSENC_H__
+#define __NVKM_MSENC_H__
+#include <core/engine.h>
+#endif
diff --git a/drivers/gpu/drm/nouveau/include/nvkm/engine/nvdec.h b/drivers/gpu/drm/nouveau/include/nvkm/engine/nvdec.h
new file mode 100644
index 000000000000..30b76d13fdcb
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/include/nvkm/engine/nvdec.h
@@ -0,0 +1,4 @@
+#ifndef __NVKM_NVDEC_H__
+#define __NVKM_NVDEC_H__
+#include <core/engine.h>
+#endif
diff --git a/drivers/gpu/drm/nouveau/include/nvkm/engine/nvenc.h b/drivers/gpu/drm/nouveau/include/nvkm/engine/nvenc.h
new file mode 100644
index 000000000000..8a819328059b
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/include/nvkm/engine/nvenc.h
@@ -0,0 +1,4 @@
+#ifndef __NVKM_NVENC_H__
+#define __NVKM_NVENC_H__
+#include <core/engine.h>
+#endif
diff --git a/drivers/gpu/drm/nouveau/include/nvkm/engine/vic.h b/drivers/gpu/drm/nouveau/include/nvkm/engine/vic.h
new file mode 100644
index 000000000000..2b0dc4c695c2
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/include/nvkm/engine/vic.h
@@ -0,0 +1,4 @@
+#ifndef __NVKM_VIC_H__
+#define __NVKM_VIC_H__
+#include <core/engine.h>
+#endif
diff --git a/drivers/gpu/drm/nouveau/include/nvkm/subdev/bios/extdev.h b/drivers/gpu/drm/nouveau/include/nvkm/subdev/bios/extdev.h
index 6d3bedc633b3..bb49bd5f879e 100644
--- a/drivers/gpu/drm/nouveau/include/nvkm/subdev/bios/extdev.h
+++ b/drivers/gpu/drm/nouveau/include/nvkm/subdev/bios/extdev.h
@@ -5,6 +5,9 @@ enum nvbios_extdev_type {
 	NVBIOS_EXTDEV_VT1103M		= 0x40,
 	NVBIOS_EXTDEV_PX3540		= 0x41,
 	NVBIOS_EXTDEV_VT1105M		= 0x42, /* or close enough... */
+	NVBIOS_EXTDEV_INA219		= 0x4c,
+	NVBIOS_EXTDEV_INA209		= 0x4d,
+	NVBIOS_EXTDEV_INA3221		= 0x4e,
 	NVBIOS_EXTDEV_ADT7473		= 0x70, /* can also be a LM64 */
 	NVBIOS_EXTDEV_HDCP_EEPROM	= 0x90,
 	NVBIOS_EXTDEV_NONE		= 0xff,
diff --git a/drivers/gpu/drm/nouveau/include/nvkm/subdev/bios/iccsense.h b/drivers/gpu/drm/nouveau/include/nvkm/subdev/bios/iccsense.h
new file mode 100644
index 000000000000..9cb97477248b
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/include/nvkm/subdev/bios/iccsense.h
@@ -0,0 +1,16 @@
+#ifndef __NVBIOS_ICCSENSE_H__
+#define __NVBIOS_ICCSENSE_H__
+struct pwr_rail_t {
+	u8 mode;
+	u8 extdev_id;
+	u8 resistor_mohm;
+	u8 rail;
+};
+
+struct nvbios_iccsense {
+	int nr_entry;
+	struct pwr_rail_t *rail;
+};
+
+int nvbios_iccsense_parse(struct nvkm_bios *, struct nvbios_iccsense *);
+#endif
diff --git a/drivers/gpu/drm/nouveau/include/nvkm/subdev/clk.h b/drivers/gpu/drm/nouveau/include/nvkm/subdev/clk.h
index 6b33bc058924..fb54417bc458 100644
--- a/drivers/gpu/drm/nouveau/include/nvkm/subdev/clk.h
+++ b/drivers/gpu/drm/nouveau/include/nvkm/subdev/clk.h
@@ -121,4 +121,5 @@ int gt215_clk_new(struct nvkm_device *, int, struct nvkm_clk **);
 int gf100_clk_new(struct nvkm_device *, int, struct nvkm_clk **);
 int gk104_clk_new(struct nvkm_device *, int, struct nvkm_clk **);
 int gk20a_clk_new(struct nvkm_device *, int, struct nvkm_clk **);
+int gm20b_clk_new(struct nvkm_device *, int, struct nvkm_clk **);
 #endif
diff --git a/drivers/gpu/drm/nouveau/include/nvkm/subdev/devinit.h b/drivers/gpu/drm/nouveau/include/nvkm/subdev/devinit.h
index 6c1407fd317b..193626c69517 100644
--- a/drivers/gpu/drm/nouveau/include/nvkm/subdev/devinit.h
+++ b/drivers/gpu/drm/nouveau/include/nvkm/subdev/devinit.h
@@ -27,5 +27,5 @@ int gt215_devinit_new(struct nvkm_device *, int, struct nvkm_devinit **);
 int mcp89_devinit_new(struct nvkm_device *, int, struct nvkm_devinit **);
 int gf100_devinit_new(struct nvkm_device *, int, struct nvkm_devinit **);
 int gm107_devinit_new(struct nvkm_device *, int, struct nvkm_devinit **);
-int gm204_devinit_new(struct nvkm_device *, int, struct nvkm_devinit **);
+int gm200_devinit_new(struct nvkm_device *, int, struct nvkm_devinit **);
 #endif
diff --git a/drivers/gpu/drm/nouveau/include/nvkm/subdev/i2c.h b/drivers/gpu/drm/nouveau/include/nvkm/subdev/i2c.h
index 6b6224dbd5bb..a63c5ac69f66 100644
--- a/drivers/gpu/drm/nouveau/include/nvkm/subdev/i2c.h
+++ b/drivers/gpu/drm/nouveau/include/nvkm/subdev/i2c.h
@@ -89,7 +89,7 @@ int g94_i2c_new(struct nvkm_device *, int, struct nvkm_i2c **);
 int gf117_i2c_new(struct nvkm_device *, int, struct nvkm_i2c **);
 int gf119_i2c_new(struct nvkm_device *, int, struct nvkm_i2c **);
 int gk104_i2c_new(struct nvkm_device *, int, struct nvkm_i2c **);
-int gm204_i2c_new(struct nvkm_device *, int, struct nvkm_i2c **);
+int gm200_i2c_new(struct nvkm_device *, int, struct nvkm_i2c **);
 
 static inline int
 nvkm_rdi2cr(struct i2c_adapter *adap, u8 addr, u8 reg)
@@ -108,6 +108,22 @@ nvkm_rdi2cr(struct i2c_adapter *adap, u8 addr, u8 reg)
 }
 
 static inline int
+nv_rd16i2cr(struct i2c_adapter *adap, u8 addr, u8 reg)
+{
+	u8 val[2];
+	struct i2c_msg msgs[] = {
+		{ .addr = addr, .flags = 0, .len = 1, .buf = &reg },
+		{ .addr = addr, .flags = I2C_M_RD, .len = 2, .buf = val },
+	};
+
+	int ret = i2c_transfer(adap, msgs, ARRAY_SIZE(msgs));
+	if (ret != 2)
+		return -EIO;
+
+	return val[0] << 8 | val[1];
+}
+
+static inline int
 nvkm_wri2cr(struct i2c_adapter *adap, u8 addr, u8 reg, u8 val)
 {
 	u8 buf[2] = { reg, val };
@@ -122,6 +138,21 @@ nvkm_wri2cr(struct i2c_adapter *adap, u8 addr, u8 reg, u8 val)
 	return 0;
 }
 
+static inline int
+nv_wr16i2cr(struct i2c_adapter *adap, u8 addr, u8 reg, u16 val)
+{
+	u8 buf[3] = { reg, val >> 8, val & 0xff};
+	struct i2c_msg msgs[] = {
+		{ .addr = addr, .flags = 0, .len = 3, .buf = buf },
+	};
+
+	int ret = i2c_transfer(adap, msgs, ARRAY_SIZE(msgs));
+	if (ret != 1)
+		return -EIO;
+
+	return 0;
+}
+
 static inline bool
 nvkm_probe_i2c(struct i2c_adapter *adap, u8 addr)
 {
diff --git a/drivers/gpu/drm/nouveau/include/nvkm/subdev/ibus.h b/drivers/gpu/drm/nouveau/include/nvkm/subdev/ibus.h
index ea23e24a246c..c4ecf255ff39 100644
--- a/drivers/gpu/drm/nouveau/include/nvkm/subdev/ibus.h
+++ b/drivers/gpu/drm/nouveau/include/nvkm/subdev/ibus.h
@@ -6,5 +6,5 @@ int gf100_ibus_new(struct nvkm_device *, int, struct nvkm_subdev **);
 int gf117_ibus_new(struct nvkm_device *, int, struct nvkm_subdev **);
 int gk104_ibus_new(struct nvkm_device *, int, struct nvkm_subdev **);
 int gk20a_ibus_new(struct nvkm_device *, int, struct nvkm_subdev **);
-int gm204_ibus_new(struct nvkm_device *, int, struct nvkm_subdev **);
+int gm200_ibus_new(struct nvkm_device *, int, struct nvkm_subdev **);
 #endif
diff --git a/drivers/gpu/drm/nouveau/include/nvkm/subdev/iccsense.h b/drivers/gpu/drm/nouveau/include/nvkm/subdev/iccsense.h
new file mode 100644
index 000000000000..530c6215fe4f
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/include/nvkm/subdev/iccsense.h
@@ -0,0 +1,17 @@
+#ifndef __NVKM_ICCSENSE_H__
+#define __NVKM_ICCSENSE_H__
+
+#include <core/subdev.h>
+
+struct nkvm_iccsense_rail;
+struct nvkm_iccsense {
+	struct nvkm_subdev subdev;
+	u8 rail_count;
+	bool data_valid;
+	struct nvkm_iccsense_rail *rails;
+};
+
+int gf100_iccsense_new(struct nvkm_device *, int index, struct nvkm_iccsense **);
+int nvkm_iccsense_read(struct nvkm_iccsense *iccsense, u8 idx);
+int nvkm_iccsense_read_all(struct nvkm_iccsense *iccsense);
+#endif
diff --git a/drivers/gpu/drm/nouveau/include/nvkm/subdev/ltc.h b/drivers/gpu/drm/nouveau/include/nvkm/subdev/ltc.h
index 0ffa2ec106d6..c6b90b6543b3 100644
--- a/drivers/gpu/drm/nouveau/include/nvkm/subdev/ltc.h
+++ b/drivers/gpu/drm/nouveau/include/nvkm/subdev/ltc.h
@@ -37,5 +37,5 @@ int gf100_ltc_new(struct nvkm_device *, int, struct nvkm_ltc **);
 int gk104_ltc_new(struct nvkm_device *, int, struct nvkm_ltc **);
 int gk20a_ltc_new(struct nvkm_device *, int, struct nvkm_ltc **);
 int gm107_ltc_new(struct nvkm_device *, int, struct nvkm_ltc **);
-int gm204_ltc_new(struct nvkm_device *, int, struct nvkm_ltc **);
+int gm200_ltc_new(struct nvkm_device *, int, struct nvkm_ltc **);
 #endif
diff --git a/drivers/gpu/drm/nouveau/include/nvkm/subdev/secboot.h b/drivers/gpu/drm/nouveau/include/nvkm/subdev/secboot.h
new file mode 100644
index 000000000000..c6edd95a5b69
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/include/nvkm/subdev/secboot.h
@@ -0,0 +1,59 @@
+/*
+ * Copyright (c) 2016, NVIDIA CORPORATION. All rights reserved.
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
+ * THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+ * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
+ * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER
+ * DEALINGS IN THE SOFTWARE.
+ */
+
+#ifndef __NVKM_SECURE_BOOT_H__
+#define __NVKM_SECURE_BOOT_H__
+
+#include <core/subdev.h>
+
+enum nvkm_secboot_falcon {
+	NVKM_SECBOOT_FALCON_PMU	= 0,
+	NVKM_SECBOOT_FALCON_RESERVED = 1,
+	NVKM_SECBOOT_FALCON_FECS = 2,
+	NVKM_SECBOOT_FALCON_GPCCS = 3,
+	NVKM_SECBOOT_FALCON_END = 4,
+	NVKM_SECBOOT_FALCON_INVALID = 0xffffffff,
+};
+
+/**
+ * @base:		base IO address of the falcon performing secure boot
+ * @irq_mask:		IRQ mask of the falcon performing secure boot
+ * @enable_mask:	enable mask of the falcon performing secure boot
+*/
+struct nvkm_secboot {
+	const struct nvkm_secboot_func *func;
+	struct nvkm_subdev subdev;
+
+	u32 base;
+	u32 irq_mask;
+	u32 enable_mask;
+};
+#define nvkm_secboot(p) container_of((p), struct nvkm_secboot, subdev)
+
+bool nvkm_secboot_is_managed(struct nvkm_secboot *, enum nvkm_secboot_falcon);
+int nvkm_secboot_reset(struct nvkm_secboot *, u32 falcon);
+int nvkm_secboot_start(struct nvkm_secboot *, u32 falcon);
+
+int gm200_secboot_new(struct nvkm_device *, int, struct nvkm_secboot **);
+int gm20b_secboot_new(struct nvkm_device *, int, struct nvkm_secboot **);
+
+#endif
diff --git a/drivers/gpu/drm/nouveau/include/nvkm/subdev/volt.h b/drivers/gpu/drm/nouveau/include/nvkm/subdev/volt.h
index b458d046dba7..feff55cff05b 100644
--- a/drivers/gpu/drm/nouveau/include/nvkm/subdev/volt.h
+++ b/drivers/gpu/drm/nouveau/include/nvkm/subdev/volt.h
@@ -20,4 +20,5 @@ int nvkm_volt_set_id(struct nvkm_volt *, u8 id, int condition);
 int nv40_volt_new(struct nvkm_device *, int, struct nvkm_volt **);
 int gk104_volt_new(struct nvkm_device *, int, struct nvkm_volt **);
 int gk20a_volt_new(struct nvkm_device *, int, struct nvkm_volt **);
+int gm20b_volt_new(struct nvkm_device *, int, struct nvkm_volt **);
 #endif
diff --git a/drivers/gpu/drm/nouveau/nouveau_abi16.c b/drivers/gpu/drm/nouveau/nouveau_abi16.c
index 50f52ffe5b0c..a59e524c028c 100644
--- a/drivers/gpu/drm/nouveau/nouveau_abi16.c
+++ b/drivers/gpu/drm/nouveau/nouveau_abi16.c
@@ -263,13 +263,23 @@ nouveau_abi16_ioctl_channel_alloc(ABI16_IOCTL_ARGS)
 	/* hack to allow channel engine type specification on kepler */
 	if (device->info.family >= NV_DEVICE_INFO_V0_KEPLER) {
 		if (init->fb_ctxdma_handle != ~0)
-			init->fb_ctxdma_handle = KEPLER_CHANNEL_GPFIFO_A_V0_ENGINE_GR;
-		else
-			init->fb_ctxdma_handle = init->tt_ctxdma_handle;
+			init->fb_ctxdma_handle = NVA06F_V0_ENGINE_GR;
+		else {
+			init->fb_ctxdma_handle = 0;
+#define _(A,B) if (init->tt_ctxdma_handle & (A)) init->fb_ctxdma_handle |= (B)
+			_(0x01, NVA06F_V0_ENGINE_GR);
+			_(0x02, NVA06F_V0_ENGINE_MSPDEC);
+			_(0x04, NVA06F_V0_ENGINE_MSPPP);
+			_(0x08, NVA06F_V0_ENGINE_MSVLD);
+			_(0x10, NVA06F_V0_ENGINE_CE0);
+			_(0x20, NVA06F_V0_ENGINE_CE1);
+			_(0x40, NVA06F_V0_ENGINE_MSENC);
+#undef _
+		}
 
 		/* allow flips to be executed if this is a graphics channel */
 		init->tt_ctxdma_handle = 0;
-		if (init->fb_ctxdma_handle == KEPLER_CHANNEL_GPFIFO_A_V0_ENGINE_GR)
+		if (init->fb_ctxdma_handle == NVA06F_V0_ENGINE_GR)
 			init->tt_ctxdma_handle = 1;
 	}
 
diff --git a/drivers/gpu/drm/nouveau/nouveau_bo.c b/drivers/gpu/drm/nouveau/nouveau_bo.c
index 78f520d05de9..2cdaea58678d 100644
--- a/drivers/gpu/drm/nouveau/nouveau_bo.c
+++ b/drivers/gpu/drm/nouveau/nouveau_bo.c
@@ -1502,7 +1502,7 @@ nouveau_ttm_tt_populate(struct ttm_tt *ttm)
 	}
 #endif
 
-#ifdef CONFIG_SWIOTLB
+#if IS_ENABLED(CONFIG_SWIOTLB) && IS_ENABLED(CONFIG_X86)
 	if (swiotlb_nr_tbl()) {
 		return ttm_dma_populate((void *)ttm, dev->dev);
 	}
@@ -1520,7 +1520,7 @@ nouveau_ttm_tt_populate(struct ttm_tt *ttm)
 				    DMA_BIDIRECTIONAL);
 
 		if (dma_mapping_error(pdev, addr)) {
-			while (--i) {
+			while (i--) {
 				dma_unmap_page(pdev, ttm_dma->dma_address[i],
 					       PAGE_SIZE, DMA_BIDIRECTIONAL);
 				ttm_dma->dma_address[i] = 0;
@@ -1570,7 +1570,7 @@ nouveau_ttm_tt_unpopulate(struct ttm_tt *ttm)
 	}
 #endif
 
-#ifdef CONFIG_SWIOTLB
+#if IS_ENABLED(CONFIG_SWIOTLB) && IS_ENABLED(CONFIG_X86)
 	if (swiotlb_nr_tbl()) {
 		ttm_dma_unpopulate((void *)ttm, dev->dev);
 		return;
diff --git a/drivers/gpu/drm/nouveau/nouveau_chan.c b/drivers/gpu/drm/nouveau/nouveau_chan.c
index 3f804a8c590c..879655c03ae9 100644
--- a/drivers/gpu/drm/nouveau/nouveau_chan.c
+++ b/drivers/gpu/drm/nouveau/nouveau_chan.c
@@ -192,6 +192,7 @@ nouveau_channel_ind(struct nouveau_drm *drm, struct nvif_device *device,
 		    u32 engine, struct nouveau_channel **pchan)
 {
 	static const u16 oclasses[] = { MAXWELL_CHANNEL_GPFIFO_A,
+					KEPLER_CHANNEL_GPFIFO_B,
 					KEPLER_CHANNEL_GPFIFO_A,
 					FERMI_CHANNEL_GPFIFO,
 					G82_CHANNEL_GPFIFO,
@@ -217,7 +218,7 @@ nouveau_channel_ind(struct nouveau_drm *drm, struct nvif_device *device,
 	do {
 		if (oclass[0] >= KEPLER_CHANNEL_GPFIFO_A) {
 			args.kepler.version = 0;
-			args.kepler.engine  = engine;
+			args.kepler.engines = engine;
 			args.kepler.ilength = 0x02000;
 			args.kepler.ioffset = 0x10000 + chan->push.vma.offset;
 			args.kepler.vm = 0;
diff --git a/drivers/gpu/drm/nouveau/nouveau_display.c b/drivers/gpu/drm/nouveau/nouveau_display.c
index 24be27d3cd18..7ce7fa5cb5e6 100644
--- a/drivers/gpu/drm/nouveau/nouveau_display.c
+++ b/drivers/gpu/drm/nouveau/nouveau_display.c
@@ -495,7 +495,7 @@ nouveau_display_create(struct drm_device *dev)
 
 	if (nouveau_modeset != 2 && drm->vbios.dcb.entries) {
 		static const u16 oclass[] = {
-			GM204_DISP,
+			GM200_DISP,
 			GM107_DISP,
 			GK110_DISP,
 			GK104_DISP,
@@ -635,10 +635,6 @@ nouveau_display_resume(struct drm_device *dev, bool runtime)
 		nv_crtc->lut.depth = 0;
 	}
 
-	/* Make sure that drm and hw vblank irqs get resumed if needed. */
-	for (head = 0; head < dev->mode_config.num_crtc; head++)
-		drm_vblank_on(dev, head);
-
 	/* This should ensure we don't hit a locking problem when someone
 	 * wakes us up via a connector.  We should never go into suspend
 	 * while the display is on anyways.
@@ -648,6 +644,10 @@ nouveau_display_resume(struct drm_device *dev, bool runtime)
 
 	drm_helper_resume_force_mode(dev);
 
+	/* Make sure that drm and hw vblank irqs get resumed if needed. */
+	for (head = 0; head < dev->mode_config.num_crtc; head++)
+		drm_vblank_on(dev, head);
+
 	list_for_each_entry(crtc, &dev->mode_config.crtc_list, head) {
 		struct nouveau_crtc *nv_crtc = nouveau_crtc(crtc);
 
diff --git a/drivers/gpu/drm/nouveau/nouveau_drm.c b/drivers/gpu/drm/nouveau/nouveau_drm.c
index bb8498c9b13e..d06877d9c1ed 100644
--- a/drivers/gpu/drm/nouveau/nouveau_drm.c
+++ b/drivers/gpu/drm/nouveau/nouveau_drm.c
@@ -198,6 +198,7 @@ nouveau_accel_init(struct nouveau_drm *drm)
 			break;
 		case FERMI_CHANNEL_GPFIFO:
 		case KEPLER_CHANNEL_GPFIFO_A:
+		case KEPLER_CHANNEL_GPFIFO_B:
 		case MAXWELL_CHANNEL_GPFIFO_A:
 			ret = nvc0_fence_create(drm);
 			break;
@@ -215,13 +216,13 @@ nouveau_accel_init(struct nouveau_drm *drm)
 
 	if (device->info.family >= NV_DEVICE_INFO_V0_KEPLER) {
 		ret = nouveau_channel_new(drm, &drm->device,
-					  KEPLER_CHANNEL_GPFIFO_A_V0_ENGINE_CE0|
-					  KEPLER_CHANNEL_GPFIFO_A_V0_ENGINE_CE1,
+					  NVA06F_V0_ENGINE_CE0 |
+					  NVA06F_V0_ENGINE_CE1,
 					  0, &drm->cechan);
 		if (ret)
 			NV_ERROR(drm, "failed to create ce channel, %d\n", ret);
 
-		arg0 = KEPLER_CHANNEL_GPFIFO_A_V0_ENGINE_GR;
+		arg0 = NVA06F_V0_ENGINE_GR;
 		arg1 = 1;
 	} else
 	if (device->info.chipset >= 0xa3 &&
@@ -375,7 +376,7 @@ nouveau_get_hdmi_dev(struct nouveau_drm *drm)
 	struct pci_dev *pdev = drm->dev->pdev;
 
 	if (!pdev) {
-		DRM_INFO("not a PCI device; no HDMI\n");
+		NV_DEBUG(drm, "not a PCI device; no HDMI\n");
 		drm->hdmi_device = NULL;
 		return;
 	}
diff --git a/drivers/gpu/drm/nouveau/nouveau_hwmon.c b/drivers/gpu/drm/nouveau/nouveau_hwmon.c
index 8e13467d0ddb..67edd2f5b71a 100644
--- a/drivers/gpu/drm/nouveau/nouveau_hwmon.c
+++ b/drivers/gpu/drm/nouveau/nouveau_hwmon.c
@@ -34,6 +34,7 @@
 #include "nouveau_drm.h"
 #include "nouveau_hwmon.h"
 
+#include <nvkm/subdev/iccsense.h>
 #include <nvkm/subdev/volt.h>
 
 #if defined(CONFIG_HWMON) || (defined(MODULE) && defined(CONFIG_HWMON_MODULE))
@@ -543,6 +544,24 @@ nouveau_hwmon_get_in0_label(struct device *d,
 static SENSOR_DEVICE_ATTR(in0_label, S_IRUGO,
 			  nouveau_hwmon_get_in0_label, NULL, 0);
 
+static ssize_t
+nouveau_hwmon_get_power1_input(struct device *d, struct device_attribute *a,
+			       char *buf)
+{
+	struct drm_device *dev = dev_get_drvdata(d);
+	struct nouveau_drm *drm = nouveau_drm(dev);
+	struct nvkm_iccsense *iccsense = nvxx_iccsense(&drm->device);
+	int result = nvkm_iccsense_read_all(iccsense);
+
+	if (result < 0)
+		return result;
+
+	return sprintf(buf, "%i\n", result);
+}
+
+static SENSOR_DEVICE_ATTR(power1_input, S_IRUGO,
+			  nouveau_hwmon_get_power1_input, NULL, 0);
+
 static struct attribute *hwmon_default_attributes[] = {
 	&sensor_dev_attr_name.dev_attr.attr,
 	&sensor_dev_attr_update_rate.dev_attr.attr,
@@ -579,6 +598,11 @@ static struct attribute *hwmon_in0_attributes[] = {
 	NULL
 };
 
+static struct attribute *hwmon_power_attributes[] = {
+	&sensor_dev_attr_power1_input.dev_attr.attr,
+	NULL
+};
+
 static const struct attribute_group hwmon_default_attrgroup = {
 	.attrs = hwmon_default_attributes,
 };
@@ -594,6 +618,9 @@ static const struct attribute_group hwmon_pwm_fan_attrgroup = {
 static const struct attribute_group hwmon_in0_attrgroup = {
 	.attrs = hwmon_in0_attributes,
 };
+static const struct attribute_group hwmon_power_attrgroup = {
+	.attrs = hwmon_power_attributes,
+};
 #endif
 
 int
@@ -603,6 +630,7 @@ nouveau_hwmon_init(struct drm_device *dev)
 	struct nouveau_drm *drm = nouveau_drm(dev);
 	struct nvkm_therm *therm = nvxx_therm(&drm->device);
 	struct nvkm_volt *volt = nvxx_volt(&drm->device);
+	struct nvkm_iccsense *iccsense = nvxx_iccsense(&drm->device);
 	struct nouveau_hwmon *hwmon;
 	struct device *hwmon_dev;
 	int ret = 0;
@@ -612,10 +640,7 @@ nouveau_hwmon_init(struct drm_device *dev)
 		return -ENOMEM;
 	hwmon->dev = dev;
 
-	if (!therm || !therm->attr_get || !therm->attr_set)
-		return -ENODEV;
-
-	hwmon_dev = hwmon_device_register(&dev->pdev->dev);
+	hwmon_dev = hwmon_device_register(dev->dev);
 	if (IS_ERR(hwmon_dev)) {
 		ret = PTR_ERR(hwmon_dev);
 		NV_ERROR(drm, "Unable to register hwmon device: %d\n", ret);
@@ -628,26 +653,28 @@ nouveau_hwmon_init(struct drm_device *dev)
 	if (ret)
 		goto error;
 
-	/* if the card has a working thermal sensor */
-	if (nvkm_therm_temp_get(therm) >= 0) {
-		ret = sysfs_create_group(&hwmon_dev->kobj, &hwmon_temp_attrgroup);
-		if (ret)
-			goto error;
-	}
-
-	/* if the card has a pwm fan */
-	/*XXX: incorrect, need better detection for this, some boards have
-	 *     the gpio entries for pwm fan control even when there's no
-	 *     actual fan connected to it... therm table? */
-	if (therm->fan_get && therm->fan_get(therm) >= 0) {
-		ret = sysfs_create_group(&hwmon_dev->kobj,
-					 &hwmon_pwm_fan_attrgroup);
-		if (ret)
-			goto error;
+	if (therm && therm->attr_get && therm->attr_set) {
+		/* if the card has a working thermal sensor */
+		if (nvkm_therm_temp_get(therm) >= 0) {
+			ret = sysfs_create_group(&hwmon_dev->kobj, &hwmon_temp_attrgroup);
+			if (ret)
+				goto error;
+		}
+
+		/* if the card has a pwm fan */
+		/*XXX: incorrect, need better detection for this, some boards have
+		 *     the gpio entries for pwm fan control even when there's no
+		 *     actual fan connected to it... therm table? */
+		if (therm->fan_get && therm->fan_get(therm) >= 0) {
+			ret = sysfs_create_group(&hwmon_dev->kobj,
+						 &hwmon_pwm_fan_attrgroup);
+			if (ret)
+				goto error;
+		}
 	}
 
 	/* if the card can read the fan rpm */
-	if (nvkm_therm_fan_sense(therm) >= 0) {
+	if (therm && nvkm_therm_fan_sense(therm) >= 0) {
 		ret = sysfs_create_group(&hwmon_dev->kobj,
 					 &hwmon_fan_rpm_attrgroup);
 		if (ret)
@@ -662,6 +689,13 @@ nouveau_hwmon_init(struct drm_device *dev)
 			goto error;
 	}
 
+	if (iccsense && iccsense->data_valid && iccsense->rail_count) {
+		ret = sysfs_create_group(&hwmon_dev->kobj,
+					 &hwmon_power_attrgroup);
+		if (ret)
+			goto error;
+	}
+
 	hwmon->hwmon = hwmon_dev;
 
 	return 0;
@@ -688,6 +722,7 @@ nouveau_hwmon_fini(struct drm_device *dev)
 		sysfs_remove_group(&hwmon->hwmon->kobj, &hwmon_pwm_fan_attrgroup);
 		sysfs_remove_group(&hwmon->hwmon->kobj, &hwmon_fan_rpm_attrgroup);
 		sysfs_remove_group(&hwmon->hwmon->kobj, &hwmon_in0_attrgroup);
+		sysfs_remove_group(&hwmon->hwmon->kobj, &hwmon_power_attrgroup);
 
 		hwmon_device_unregister(hwmon->hwmon);
 	}
diff --git a/drivers/gpu/drm/nouveau/nouveau_platform.c b/drivers/gpu/drm/nouveau/nouveau_platform.c
index 8a70cec59bcd..2dfe58af12e4 100644
--- a/drivers/gpu/drm/nouveau/nouveau_platform.c
+++ b/drivers/gpu/drm/nouveau/nouveau_platform.c
@@ -24,7 +24,7 @@
 static int nouveau_platform_probe(struct platform_device *pdev)
 {
 	const struct nvkm_device_tegra_func *func;
-	struct nvkm_device *device;
+	struct nvkm_device *device = NULL;
 	struct drm_device *drm;
 	int ret;
 
diff --git a/drivers/gpu/drm/nouveau/nv50_display.c b/drivers/gpu/drm/nouveau/nv50_display.c
index ea3921652449..a43445caae60 100644
--- a/drivers/gpu/drm/nouveau/nv50_display.c
+++ b/drivers/gpu/drm/nouveau/nv50_display.c
@@ -297,7 +297,7 @@ nv50_core_create(struct nvif_device *device, struct nvif_object *disp,
 		.pushbuf = 0xb0007d00,
 	};
 	static const s32 oclass[] = {
-		GM204_DISP_CORE_CHANNEL_DMA,
+		GM200_DISP_CORE_CHANNEL_DMA,
 		GM107_DISP_CORE_CHANNEL_DMA,
 		GK110_DISP_CORE_CHANNEL_DMA,
 		GK104_DISP_CORE_CHANNEL_DMA,
diff --git a/drivers/gpu/drm/nouveau/nvkm/core/Kbuild b/drivers/gpu/drm/nouveau/nvkm/core/Kbuild
index 7f66963f305c..86a31a8e1e51 100644
--- a/drivers/gpu/drm/nouveau/nvkm/core/Kbuild
+++ b/drivers/gpu/drm/nouveau/nvkm/core/Kbuild
@@ -2,6 +2,7 @@ nvkm-y := nvkm/core/client.o
 nvkm-y += nvkm/core/engine.o
 nvkm-y += nvkm/core/enum.o
 nvkm-y += nvkm/core/event.o
+nvkm-y += nvkm/core/firmware.o
 nvkm-y += nvkm/core/gpuobj.o
 nvkm-y += nvkm/core/ioctl.o
 nvkm-y += nvkm/core/memory.o
diff --git a/drivers/gpu/drm/nouveau/nvkm/core/firmware.c b/drivers/gpu/drm/nouveau/nvkm/core/firmware.c
new file mode 100644
index 000000000000..34ecd4a7e0c1
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/nvkm/core/firmware.c
@@ -0,0 +1,61 @@
+/*
+ * Copyright (c) 2016, NVIDIA CORPORATION. All rights reserved.
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
+ * THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+ * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
+ * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER
+ * DEALINGS IN THE SOFTWARE.
+ */
+#include <core/device.h>
+
+/**
+ * nvkm_firmware_get - load firmware from the official nvidia/chip/ directory
+ * @device	device that will use that firmware
+ * @fwname	name of firmware file to load
+ * @fw		firmware structure to load to
+ *
+ * Use this function to load firmware files in the form nvidia/chip/fwname.bin.
+ * Firmware files released by NVIDIA will always follow this format.
+ */
+int
+nvkm_firmware_get(struct nvkm_device *device, const char *fwname,
+		  const struct firmware **fw)
+{
+	char f[64];
+	char cname[16];
+	int i;
+
+	/* Convert device name to lowercase */
+	strncpy(cname, device->chip->name, sizeof(cname));
+	cname[sizeof(cname) - 1] = '\0';
+	i = strlen(cname);
+	while (i) {
+		--i;
+		cname[i] = tolower(cname[i]);
+	}
+
+	snprintf(f, sizeof(f), "nvidia/%s/%s.bin", cname, fwname);
+	return request_firmware(fw, f, device->dev);
+}
+
+/**
+ * nvkm_firmware_put - release firmware loaded with nvkm_firmware_get
+ */
+void
+nvkm_firmware_put(const struct firmware *fw)
+{
+	release_firmware(fw);
+}
diff --git a/drivers/gpu/drm/nouveau/nvkm/core/gpuobj.c b/drivers/gpu/drm/nouveau/nvkm/core/gpuobj.c
index c3a790eb8d6a..a7bd22706b2a 100644
--- a/drivers/gpu/drm/nouveau/nvkm/core/gpuobj.c
+++ b/drivers/gpu/drm/nouveau/nvkm/core/gpuobj.c
@@ -253,3 +253,23 @@ nvkm_gpuobj_wrap(struct nvkm_memory *memory, struct nvkm_gpuobj **pgpuobj)
 	(*pgpuobj)->size = nvkm_memory_size(memory);
 	return 0;
 }
+
+void
+nvkm_gpuobj_memcpy_to(struct nvkm_gpuobj *dst, u32 dstoffset, void *src,
+		      u32 length)
+{
+	int i;
+
+	for (i = 0; i < length; i += 4)
+		nvkm_wo32(dst, dstoffset + i, *(u32 *)(src + i));
+}
+
+void
+nvkm_gpuobj_memcpy_from(void *dst, struct nvkm_gpuobj *src, u32 srcoffset,
+			u32 length)
+{
+	int i;
+
+	for (i = 0; i < length; i += 4)
+		((u32 *)src)[i / 4] = nvkm_ro32(src, srcoffset + i);
+}
diff --git a/drivers/gpu/drm/nouveau/nvkm/core/ramht.c b/drivers/gpu/drm/nouveau/nvkm/core/ramht.c
index 3216e157a8a0..89da47234016 100644
--- a/drivers/gpu/drm/nouveau/nvkm/core/ramht.c
+++ b/drivers/gpu/drm/nouveau/nvkm/core/ramht.c
@@ -131,7 +131,7 @@ nvkm_ramht_del(struct nvkm_ramht **pramht)
 	struct nvkm_ramht *ramht = *pramht;
 	if (ramht) {
 		nvkm_gpuobj_del(&ramht->gpuobj);
-		kfree(*pramht);
+		vfree(*pramht);
 		*pramht = NULL;
 	}
 }
@@ -143,8 +143,8 @@ nvkm_ramht_new(struct nvkm_device *device, u32 size, u32 align,
 	struct nvkm_ramht *ramht;
 	int ret, i;
 
-	if (!(ramht = *pramht = kzalloc(sizeof(*ramht) + (size >> 3) *
-					sizeof(*ramht->data), GFP_KERNEL)))
+	if (!(ramht = *pramht = vzalloc(sizeof(*ramht) +
+					(size >> 3) * sizeof(*ramht->data))))
 		return -ENOMEM;
 
 	ramht->device = device;
diff --git a/drivers/gpu/drm/nouveau/nvkm/core/subdev.c b/drivers/gpu/drm/nouveau/nvkm/core/subdev.c
index 7de98470a2a0..3bf08cb1a289 100644
--- a/drivers/gpu/drm/nouveau/nvkm/core/subdev.c
+++ b/drivers/gpu/drm/nouveau/nvkm/core/subdev.c
@@ -29,47 +29,52 @@ static struct lock_class_key nvkm_subdev_lock_class[NVKM_SUBDEV_NR];
 
 const char *
 nvkm_subdev_name[NVKM_SUBDEV_NR] = {
-	[NVKM_SUBDEV_BAR    ] = "bar",
-	[NVKM_SUBDEV_VBIOS  ] = "bios",
-	[NVKM_SUBDEV_BUS    ] = "bus",
-	[NVKM_SUBDEV_CLK    ] = "clk",
-	[NVKM_SUBDEV_DEVINIT] = "devinit",
-	[NVKM_SUBDEV_FB     ] = "fb",
-	[NVKM_SUBDEV_FUSE   ] = "fuse",
-	[NVKM_SUBDEV_GPIO   ] = "gpio",
-	[NVKM_SUBDEV_I2C    ] = "i2c",
-	[NVKM_SUBDEV_IBUS   ] = "priv",
-	[NVKM_SUBDEV_INSTMEM] = "imem",
-	[NVKM_SUBDEV_LTC    ] = "ltc",
-	[NVKM_SUBDEV_MC     ] = "mc",
-	[NVKM_SUBDEV_MMU    ] = "mmu",
-	[NVKM_SUBDEV_MXM    ] = "mxm",
-	[NVKM_SUBDEV_PCI    ] = "pci",
-	[NVKM_SUBDEV_PMU    ] = "pmu",
-	[NVKM_SUBDEV_THERM  ] = "therm",
-	[NVKM_SUBDEV_TIMER  ] = "tmr",
-	[NVKM_SUBDEV_VOLT   ] = "volt",
-	[NVKM_ENGINE_BSP    ] = "bsp",
-	[NVKM_ENGINE_CE0    ] = "ce0",
-	[NVKM_ENGINE_CE1    ] = "ce1",
-	[NVKM_ENGINE_CE2    ] = "ce2",
-	[NVKM_ENGINE_CIPHER ] = "cipher",
-	[NVKM_ENGINE_DISP   ] = "disp",
-	[NVKM_ENGINE_DMAOBJ ] = "dma",
-	[NVKM_ENGINE_FIFO   ] = "fifo",
-	[NVKM_ENGINE_GR     ] = "gr",
-	[NVKM_ENGINE_IFB    ] = "ifb",
-	[NVKM_ENGINE_ME     ] = "me",
-	[NVKM_ENGINE_MPEG   ] = "mpeg",
-	[NVKM_ENGINE_MSENC  ] = "msenc",
-	[NVKM_ENGINE_MSPDEC ] = "mspdec",
-	[NVKM_ENGINE_MSPPP  ] = "msppp",
-	[NVKM_ENGINE_MSVLD  ] = "msvld",
-	[NVKM_ENGINE_PM     ] = "pm",
-	[NVKM_ENGINE_SEC    ] = "sec",
-	[NVKM_ENGINE_SW     ] = "sw",
-	[NVKM_ENGINE_VIC    ] = "vic",
-	[NVKM_ENGINE_VP     ] = "vp",
+	[NVKM_SUBDEV_BAR     ] = "bar",
+	[NVKM_SUBDEV_VBIOS   ] = "bios",
+	[NVKM_SUBDEV_BUS     ] = "bus",
+	[NVKM_SUBDEV_CLK     ] = "clk",
+	[NVKM_SUBDEV_DEVINIT ] = "devinit",
+	[NVKM_SUBDEV_FB      ] = "fb",
+	[NVKM_SUBDEV_FUSE    ] = "fuse",
+	[NVKM_SUBDEV_GPIO    ] = "gpio",
+	[NVKM_SUBDEV_I2C     ] = "i2c",
+	[NVKM_SUBDEV_IBUS    ] = "priv",
+	[NVKM_SUBDEV_ICCSENSE] = "iccsense",
+	[NVKM_SUBDEV_INSTMEM ] = "imem",
+	[NVKM_SUBDEV_LTC     ] = "ltc",
+	[NVKM_SUBDEV_MC      ] = "mc",
+	[NVKM_SUBDEV_MMU     ] = "mmu",
+	[NVKM_SUBDEV_MXM     ] = "mxm",
+	[NVKM_SUBDEV_PCI     ] = "pci",
+	[NVKM_SUBDEV_PMU     ] = "pmu",
+	[NVKM_SUBDEV_SECBOOT ] = "secboot",
+	[NVKM_SUBDEV_THERM   ] = "therm",
+	[NVKM_SUBDEV_TIMER   ] = "tmr",
+	[NVKM_SUBDEV_VOLT    ] = "volt",
+	[NVKM_ENGINE_BSP     ] = "bsp",
+	[NVKM_ENGINE_CE0     ] = "ce0",
+	[NVKM_ENGINE_CE1     ] = "ce1",
+	[NVKM_ENGINE_CE2     ] = "ce2",
+	[NVKM_ENGINE_CIPHER  ] = "cipher",
+	[NVKM_ENGINE_DISP    ] = "disp",
+	[NVKM_ENGINE_DMAOBJ  ] = "dma",
+	[NVKM_ENGINE_FIFO    ] = "fifo",
+	[NVKM_ENGINE_GR      ] = "gr",
+	[NVKM_ENGINE_IFB     ] = "ifb",
+	[NVKM_ENGINE_ME      ] = "me",
+	[NVKM_ENGINE_MPEG    ] = "mpeg",
+	[NVKM_ENGINE_MSENC   ] = "msenc",
+	[NVKM_ENGINE_MSPDEC  ] = "mspdec",
+	[NVKM_ENGINE_MSPPP   ] = "msppp",
+	[NVKM_ENGINE_MSVLD   ] = "msvld",
+	[NVKM_ENGINE_NVENC0  ] = "nvenc0",
+	[NVKM_ENGINE_NVENC1  ] = "nvenc1",
+	[NVKM_ENGINE_NVDEC   ] = "nvdec",
+	[NVKM_ENGINE_PM      ] = "pm",
+	[NVKM_ENGINE_SEC     ] = "sec",
+	[NVKM_ENGINE_SW      ] = "sw",
+	[NVKM_ENGINE_VIC     ] = "vic",
+	[NVKM_ENGINE_VP      ] = "vp",
 };
 
 void
diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/Kbuild b/drivers/gpu/drm/nouveau/nvkm/engine/Kbuild
index 36f724763fde..c2c8d2ac01b8 100644
--- a/drivers/gpu/drm/nouveau/nvkm/engine/Kbuild
+++ b/drivers/gpu/drm/nouveau/nvkm/engine/Kbuild
@@ -10,10 +10,14 @@ include $(src)/nvkm/engine/dma/Kbuild
 include $(src)/nvkm/engine/fifo/Kbuild
 include $(src)/nvkm/engine/gr/Kbuild
 include $(src)/nvkm/engine/mpeg/Kbuild
+include $(src)/nvkm/engine/msenc/Kbuild
 include $(src)/nvkm/engine/mspdec/Kbuild
 include $(src)/nvkm/engine/msppp/Kbuild
 include $(src)/nvkm/engine/msvld/Kbuild
+include $(src)/nvkm/engine/nvenc/Kbuild
+include $(src)/nvkm/engine/nvdec/Kbuild
 include $(src)/nvkm/engine/pm/Kbuild
 include $(src)/nvkm/engine/sec/Kbuild
 include $(src)/nvkm/engine/sw/Kbuild
+include $(src)/nvkm/engine/vic/Kbuild
 include $(src)/nvkm/engine/vp/Kbuild
diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/ce/Kbuild b/drivers/gpu/drm/nouveau/nvkm/engine/ce/Kbuild
index fa8cda7058cd..9c19d59b47df 100644
--- a/drivers/gpu/drm/nouveau/nvkm/engine/ce/Kbuild
+++ b/drivers/gpu/drm/nouveau/nvkm/engine/ce/Kbuild
@@ -1,4 +1,5 @@
 nvkm-y += nvkm/engine/ce/gt215.o
 nvkm-y += nvkm/engine/ce/gf100.o
 nvkm-y += nvkm/engine/ce/gk104.o
-nvkm-y += nvkm/engine/ce/gm204.o
+nvkm-y += nvkm/engine/ce/gm107.o
+nvkm-y += nvkm/engine/ce/gm200.o
diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/gr/gm206.c b/drivers/gpu/drm/nouveau/nvkm/engine/ce/gm107.c
index 341dc560acbb..4c2f42919c1f 100644
--- a/drivers/gpu/drm/nouveau/nvkm/engine/gr/gm206.c
+++ b/drivers/gpu/drm/nouveau/nvkm/engine/ce/gm107.c
@@ -1,5 +1,5 @@
 /*
- * Copyright 2015 Red Hat Inc.
+ * Copyright 2016 Red Hat Inc.
  *
  * Permission is hereby granted, free of charge, to any person obtaining a
  * copy of this software and associated documentation files (the "Software"),
@@ -19,30 +19,37 @@
  * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
  * OTHER DEALINGS IN THE SOFTWARE.
  *
- * Authors: Ben Skeggs <bskeggs@redhat.com>
+ * Authors: Ben Skeggs
  */
-#include "gf100.h"
-#include "ctxgf100.h"
+#include "priv.h"
 
 #include <nvif/class.h>
 
-static const struct gf100_gr_func
-gm206_gr = {
-	.init = gm204_gr_init,
-	.mmio = gm204_gr_pack_mmio,
-	.ppc_nr = 2,
-	.grctx = &gm206_grctx,
+static const struct nvkm_engine_func
+gm107_ce = {
+	.intr = gk104_ce_intr,
 	.sclass = {
-		{ -1, -1, FERMI_TWOD_A },
-		{ -1, -1, KEPLER_INLINE_TO_MEMORY_B },
-		{ -1, -1, MAXWELL_B, &gf100_fermi },
-		{ -1, -1, MAXWELL_COMPUTE_B },
+		{ -1, -1, KEPLER_DMA_COPY_A },
+		{ -1, -1, MAXWELL_DMA_COPY_A },
 		{}
 	}
 };
 
 int
-gm206_gr_new(struct nvkm_device *device, int index, struct nvkm_gr **pgr)
+gm107_ce_new(struct nvkm_device *device, int index,
+	     struct nvkm_engine **pengine)
 {
-	return gf100_gr_new_(&gm206_gr, device, index, pgr);
+	if (index == NVKM_ENGINE_CE0) {
+		return nvkm_engine_new_(&gm107_ce, device, index,
+					0x00000040, true, pengine);
+	} else
+	if (index == NVKM_ENGINE_CE1) {
+		return nvkm_engine_new_(&gm107_ce, device, index,
+					0x00000080, true, pengine);
+	} else
+	if (index == NVKM_ENGINE_CE2) {
+		return nvkm_engine_new_(&gm107_ce, device, index,
+					0x00200000, true, pengine);
+	}
+	return -ENODEV;
 }
diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/ce/gm204.c b/drivers/gpu/drm/nouveau/nvkm/engine/ce/gm200.c
index 8eaa72a59f40..13f07b32cd9c 100644
--- a/drivers/gpu/drm/nouveau/nvkm/engine/ce/gm204.c
+++ b/drivers/gpu/drm/nouveau/nvkm/engine/ce/gm200.c
@@ -26,7 +26,7 @@
 #include <nvif/class.h>
 
 static const struct nvkm_engine_func
-gm204_ce = {
+gm200_ce = {
 	.intr = gk104_ce_intr,
 	.sclass = {
 		{ -1, -1, MAXWELL_DMA_COPY_A },
@@ -35,19 +35,19 @@ gm204_ce = {
 };
 
 int
-gm204_ce_new(struct nvkm_device *device, int index,
+gm200_ce_new(struct nvkm_device *device, int index,
 	     struct nvkm_engine **pengine)
 {
 	if (index == NVKM_ENGINE_CE0) {
-		return nvkm_engine_new_(&gm204_ce, device, index,
+		return nvkm_engine_new_(&gm200_ce, device, index,
 					0x00000040, true, pengine);
 	} else
 	if (index == NVKM_ENGINE_CE1) {
-		return nvkm_engine_new_(&gm204_ce, device, index,
+		return nvkm_engine_new_(&gm200_ce, device, index,
 					0x00000080, true, pengine);
 	} else
 	if (index == NVKM_ENGINE_CE2) {
-		return nvkm_engine_new_(&gm204_ce, device, index,
+		return nvkm_engine_new_(&gm200_ce, device, index,
 					0x00200000, true, pengine);
 	}
 	return -ENODEV;
diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/device/base.c b/drivers/gpu/drm/nouveau/nvkm/engine/device/base.c
index b1ba1c782a2b..9f32c8739254 100644
--- a/drivers/gpu/drm/nouveau/nvkm/engine/device/base.c
+++ b/drivers/gpu/drm/nouveau/nvkm/engine/device/base.c
@@ -1347,6 +1347,7 @@ nvc0_chipset = {
 	.gpio = g94_gpio_new,
 	.i2c = g94_i2c_new,
 	.ibus = gf100_ibus_new,
+	.iccsense = gf100_iccsense_new,
 	.imem = nv50_instmem_new,
 	.ltc = gf100_ltc_new,
 	.mc = gf100_mc_new,
@@ -1383,6 +1384,7 @@ nvc1_chipset = {
 	.gpio = g94_gpio_new,
 	.i2c = g94_i2c_new,
 	.ibus = gf100_ibus_new,
+	.iccsense = gf100_iccsense_new,
 	.imem = nv50_instmem_new,
 	.ltc = gf100_ltc_new,
 	.mc = gf100_mc_new,
@@ -1418,6 +1420,7 @@ nvc3_chipset = {
 	.gpio = g94_gpio_new,
 	.i2c = g94_i2c_new,
 	.ibus = gf100_ibus_new,
+	.iccsense = gf100_iccsense_new,
 	.imem = nv50_instmem_new,
 	.ltc = gf100_ltc_new,
 	.mc = gf100_mc_new,
@@ -1453,6 +1456,7 @@ nvc4_chipset = {
 	.gpio = g94_gpio_new,
 	.i2c = g94_i2c_new,
 	.ibus = gf100_ibus_new,
+	.iccsense = gf100_iccsense_new,
 	.imem = nv50_instmem_new,
 	.ltc = gf100_ltc_new,
 	.mc = gf100_mc_new,
@@ -1489,6 +1493,7 @@ nvc8_chipset = {
 	.gpio = g94_gpio_new,
 	.i2c = g94_i2c_new,
 	.ibus = gf100_ibus_new,
+	.iccsense = gf100_iccsense_new,
 	.imem = nv50_instmem_new,
 	.ltc = gf100_ltc_new,
 	.mc = gf100_mc_new,
@@ -1525,6 +1530,7 @@ nvce_chipset = {
 	.gpio = g94_gpio_new,
 	.i2c = g94_i2c_new,
 	.ibus = gf100_ibus_new,
+	.iccsense = gf100_iccsense_new,
 	.imem = nv50_instmem_new,
 	.ltc = gf100_ltc_new,
 	.mc = gf100_mc_new,
@@ -1561,6 +1567,7 @@ nvcf_chipset = {
 	.gpio = g94_gpio_new,
 	.i2c = g94_i2c_new,
 	.ibus = gf100_ibus_new,
+	.iccsense = gf100_iccsense_new,
 	.imem = nv50_instmem_new,
 	.ltc = gf100_ltc_new,
 	.mc = gf100_mc_new,
@@ -1596,6 +1603,7 @@ nvd7_chipset = {
 	.gpio = gf119_gpio_new,
 	.i2c = gf117_i2c_new,
 	.ibus = gf117_ibus_new,
+	.iccsense = gf100_iccsense_new,
 	.imem = nv50_instmem_new,
 	.ltc = gf100_ltc_new,
 	.mc = gf100_mc_new,
@@ -1629,6 +1637,7 @@ nvd9_chipset = {
 	.gpio = gf119_gpio_new,
 	.i2c = gf119_i2c_new,
 	.ibus = gf117_ibus_new,
+	.iccsense = gf100_iccsense_new,
 	.imem = nv50_instmem_new,
 	.ltc = gf100_ltc_new,
 	.mc = gf100_mc_new,
@@ -1664,6 +1673,7 @@ nve4_chipset = {
 	.gpio = gk104_gpio_new,
 	.i2c = gk104_i2c_new,
 	.ibus = gk104_ibus_new,
+	.iccsense = gf100_iccsense_new,
 	.imem = nv50_instmem_new,
 	.ltc = gk104_ltc_new,
 	.mc = gf100_mc_new,
@@ -1701,6 +1711,7 @@ nve6_chipset = {
 	.gpio = gk104_gpio_new,
 	.i2c = gk104_i2c_new,
 	.ibus = gk104_ibus_new,
+	.iccsense = gf100_iccsense_new,
 	.imem = nv50_instmem_new,
 	.ltc = gk104_ltc_new,
 	.mc = gf100_mc_new,
@@ -1738,6 +1749,7 @@ nve7_chipset = {
 	.gpio = gk104_gpio_new,
 	.i2c = gk104_i2c_new,
 	.ibus = gk104_ibus_new,
+	.iccsense = gf100_iccsense_new,
 	.imem = nv50_instmem_new,
 	.ltc = gk104_ltc_new,
 	.mc = gf100_mc_new,
@@ -1799,6 +1811,7 @@ nvf0_chipset = {
 	.gpio = gk104_gpio_new,
 	.i2c = gk104_i2c_new,
 	.ibus = gk104_ibus_new,
+	.iccsense = gf100_iccsense_new,
 	.imem = nv50_instmem_new,
 	.ltc = gk104_ltc_new,
 	.mc = gf100_mc_new,
@@ -1814,7 +1827,7 @@ nvf0_chipset = {
 	.ce[2] = gk104_ce_new,
 	.disp = gk110_disp_new,
 	.dma = gf119_dma_new,
-	.fifo = gk104_fifo_new,
+	.fifo = gk110_fifo_new,
 	.gr = gk110_gr_new,
 	.mspdec = gk104_mspdec_new,
 	.msppp = gf100_msppp_new,
@@ -1835,6 +1848,7 @@ nvf1_chipset = {
 	.gpio = gk104_gpio_new,
 	.i2c = gf119_i2c_new,
 	.ibus = gk104_ibus_new,
+	.iccsense = gf100_iccsense_new,
 	.imem = nv50_instmem_new,
 	.ltc = gk104_ltc_new,
 	.mc = gf100_mc_new,
@@ -1850,7 +1864,7 @@ nvf1_chipset = {
 	.ce[2] = gk104_ce_new,
 	.disp = gk110_disp_new,
 	.dma = gf119_dma_new,
-	.fifo = gk104_fifo_new,
+	.fifo = gk110_fifo_new,
 	.gr = gk110b_gr_new,
 	.mspdec = gk104_mspdec_new,
 	.msppp = gf100_msppp_new,
@@ -1871,6 +1885,7 @@ nv106_chipset = {
 	.gpio = gk104_gpio_new,
 	.i2c = gk104_i2c_new,
 	.ibus = gk104_ibus_new,
+	.iccsense = gf100_iccsense_new,
 	.imem = nv50_instmem_new,
 	.ltc = gk104_ltc_new,
 	.mc = gk20a_mc_new,
@@ -1907,6 +1922,7 @@ nv108_chipset = {
 	.gpio = gk104_gpio_new,
 	.i2c = gk104_i2c_new,
 	.ibus = gk104_ibus_new,
+	.iccsense = gf100_iccsense_new,
 	.imem = nv50_instmem_new,
 	.ltc = gk104_ltc_new,
 	.mc = gk20a_mc_new,
@@ -1943,6 +1959,7 @@ nv117_chipset = {
 	.gpio = gk104_gpio_new,
 	.i2c = gf119_i2c_new,
 	.ibus = gk104_ibus_new,
+	.iccsense = gf100_iccsense_new,
 	.imem = nv50_instmem_new,
 	.ltc = gm107_ltc_new,
 	.mc = gk20a_mc_new,
@@ -1953,43 +1970,78 @@ nv117_chipset = {
 	.therm = gm107_therm_new,
 	.timer = gk20a_timer_new,
 	.volt = gk104_volt_new,
-	.ce[0] = gk104_ce_new,
-	.ce[2] = gk104_ce_new,
+	.ce[0] = gm107_ce_new,
+	.ce[2] = gm107_ce_new,
 	.disp = gm107_disp_new,
 	.dma = gf119_dma_new,
-	.fifo = gk208_fifo_new,
+	.fifo = gm107_fifo_new,
 	.gr = gm107_gr_new,
 	.sw = gf100_sw_new,
 };
 
 static const struct nvkm_device_chip
+nv120_chipset = {
+	.name = "GM200",
+	.bar = gf100_bar_new,
+	.bios = nvkm_bios_new,
+	.bus = gf100_bus_new,
+	.devinit = gm200_devinit_new,
+	.fb = gm107_fb_new,
+	.fuse = gm107_fuse_new,
+	.gpio = gk104_gpio_new,
+	.i2c = gm200_i2c_new,
+	.ibus = gm200_ibus_new,
+	.iccsense = gf100_iccsense_new,
+	.imem = nv50_instmem_new,
+	.ltc = gm200_ltc_new,
+	.mc = gk20a_mc_new,
+	.mmu = gf100_mmu_new,
+	.mxm = nv50_mxm_new,
+	.pci = gk104_pci_new,
+	.pmu = gm107_pmu_new,
+	.secboot = gm200_secboot_new,
+	.timer = gk20a_timer_new,
+	.volt = gk104_volt_new,
+	.ce[0] = gm200_ce_new,
+	.ce[1] = gm200_ce_new,
+	.ce[2] = gm200_ce_new,
+	.disp = gm200_disp_new,
+	.dma = gf119_dma_new,
+	.fifo = gm200_fifo_new,
+	.gr = gm200_gr_new,
+	.sw = gf100_sw_new,
+};
+
+static const struct nvkm_device_chip
 nv124_chipset = {
 	.name = "GM204",
 	.bar = gf100_bar_new,
 	.bios = nvkm_bios_new,
 	.bus = gf100_bus_new,
-	.devinit = gm204_devinit_new,
+	.devinit = gm200_devinit_new,
 	.fb = gm107_fb_new,
 	.fuse = gm107_fuse_new,
 	.gpio = gk104_gpio_new,
-	.i2c = gm204_i2c_new,
-	.ibus = gm204_ibus_new,
+	.i2c = gm200_i2c_new,
+	.ibus = gm200_ibus_new,
+	.iccsense = gf100_iccsense_new,
 	.imem = nv50_instmem_new,
-	.ltc = gm204_ltc_new,
+	.ltc = gm200_ltc_new,
 	.mc = gk20a_mc_new,
 	.mmu = gf100_mmu_new,
 	.mxm = nv50_mxm_new,
 	.pci = gk104_pci_new,
 	.pmu = gm107_pmu_new,
+	.secboot = gm200_secboot_new,
 	.timer = gk20a_timer_new,
 	.volt = gk104_volt_new,
-	.ce[0] = gm204_ce_new,
-	.ce[1] = gm204_ce_new,
-	.ce[2] = gm204_ce_new,
-	.disp = gm204_disp_new,
+	.ce[0] = gm200_ce_new,
+	.ce[1] = gm200_ce_new,
+	.ce[2] = gm200_ce_new,
+	.disp = gm200_disp_new,
 	.dma = gf119_dma_new,
-	.fifo = gm204_fifo_new,
-	.gr = gm204_gr_new,
+	.fifo = gm200_fifo_new,
+	.gr = gm200_gr_new,
 	.sw = gf100_sw_new,
 };
 
@@ -1999,28 +2051,30 @@ nv126_chipset = {
 	.bar = gf100_bar_new,
 	.bios = nvkm_bios_new,
 	.bus = gf100_bus_new,
-	.devinit = gm204_devinit_new,
+	.devinit = gm200_devinit_new,
 	.fb = gm107_fb_new,
 	.fuse = gm107_fuse_new,
 	.gpio = gk104_gpio_new,
-	.i2c = gm204_i2c_new,
-	.ibus = gm204_ibus_new,
+	.i2c = gm200_i2c_new,
+	.ibus = gm200_ibus_new,
+	.iccsense = gf100_iccsense_new,
 	.imem = nv50_instmem_new,
-	.ltc = gm204_ltc_new,
+	.ltc = gm200_ltc_new,
 	.mc = gk20a_mc_new,
 	.mmu = gf100_mmu_new,
 	.mxm = nv50_mxm_new,
 	.pci = gk104_pci_new,
 	.pmu = gm107_pmu_new,
+	.secboot = gm200_secboot_new,
 	.timer = gk20a_timer_new,
 	.volt = gk104_volt_new,
-	.ce[0] = gm204_ce_new,
-	.ce[1] = gm204_ce_new,
-	.ce[2] = gm204_ce_new,
-	.disp = gm204_disp_new,
+	.ce[0] = gm200_ce_new,
+	.ce[1] = gm200_ce_new,
+	.ce[2] = gm200_ce_new,
+	.disp = gm200_disp_new,
 	.dma = gf119_dma_new,
-	.fifo = gm204_fifo_new,
-	.gr = gm206_gr_new,
+	.fifo = gm200_fifo_new,
+	.gr = gm200_gr_new,
 	.sw = gf100_sw_new,
 };
 
@@ -2029,15 +2083,18 @@ nv12b_chipset = {
 	.name = "GM20B",
 	.bar = gk20a_bar_new,
 	.bus = gf100_bus_new,
+	.clk = gm20b_clk_new,
 	.fb = gk20a_fb_new,
 	.fuse = gm107_fuse_new,
 	.ibus = gk20a_ibus_new,
 	.imem = gk20a_instmem_new,
-	.ltc = gm204_ltc_new,
+	.ltc = gm200_ltc_new,
 	.mc = gk20a_mc_new,
 	.mmu = gf100_mmu_new,
+	.secboot = gm20b_secboot_new,
 	.timer = gk20a_timer_new,
-	.ce[2] = gm204_ce_new,
+	.ce[2] = gm200_ce_new,
+	.volt = gm20b_volt_new,
 	.dma = gf119_dma_new,
 	.fifo = gm20b_fifo_new,
 	.gr = gm20b_gr_new,
@@ -2072,26 +2129,28 @@ nvkm_device_subdev(struct nvkm_device *device, int index)
 
 	switch (index) {
 #define _(n,p,m) case NVKM_SUBDEV_##n: if (p) return (m); break
-	_(BAR    , device->bar    , &device->bar->subdev);
-	_(VBIOS  , device->bios   , &device->bios->subdev);
-	_(BUS    , device->bus    , &device->bus->subdev);
-	_(CLK    , device->clk    , &device->clk->subdev);
-	_(DEVINIT, device->devinit, &device->devinit->subdev);
-	_(FB     , device->fb     , &device->fb->subdev);
-	_(FUSE   , device->fuse   , &device->fuse->subdev);
-	_(GPIO   , device->gpio   , &device->gpio->subdev);
-	_(I2C    , device->i2c    , &device->i2c->subdev);
-	_(IBUS   , device->ibus   ,  device->ibus);
-	_(INSTMEM, device->imem   , &device->imem->subdev);
-	_(LTC    , device->ltc    , &device->ltc->subdev);
-	_(MC     , device->mc     , &device->mc->subdev);
-	_(MMU    , device->mmu    , &device->mmu->subdev);
-	_(MXM    , device->mxm    ,  device->mxm);
-	_(PCI    , device->pci    , &device->pci->subdev);
-	_(PMU    , device->pmu    , &device->pmu->subdev);
-	_(THERM  , device->therm  , &device->therm->subdev);
-	_(TIMER  , device->timer  , &device->timer->subdev);
-	_(VOLT   , device->volt   , &device->volt->subdev);
+	_(BAR     , device->bar     , &device->bar->subdev);
+	_(VBIOS   , device->bios    , &device->bios->subdev);
+	_(BUS     , device->bus     , &device->bus->subdev);
+	_(CLK     , device->clk     , &device->clk->subdev);
+	_(DEVINIT , device->devinit , &device->devinit->subdev);
+	_(FB      , device->fb      , &device->fb->subdev);
+	_(FUSE    , device->fuse    , &device->fuse->subdev);
+	_(GPIO    , device->gpio    , &device->gpio->subdev);
+	_(I2C     , device->i2c     , &device->i2c->subdev);
+	_(IBUS    , device->ibus    ,  device->ibus);
+	_(ICCSENSE, device->iccsense, &device->iccsense->subdev);
+	_(INSTMEM , device->imem    , &device->imem->subdev);
+	_(LTC     , device->ltc     , &device->ltc->subdev);
+	_(MC      , device->mc      , &device->mc->subdev);
+	_(MMU     , device->mmu     , &device->mmu->subdev);
+	_(MXM     , device->mxm     ,  device->mxm);
+	_(PCI     , device->pci     , &device->pci->subdev);
+	_(PMU     , device->pmu     , &device->pmu->subdev);
+	_(SECBOOT , device->secboot , &device->secboot->subdev);
+	_(THERM   , device->therm   , &device->therm->subdev);
+	_(TIMER   , device->timer   , &device->timer->subdev);
+	_(VOLT    , device->volt    , &device->volt->subdev);
 #undef _
 	default:
 		engine = nvkm_device_engine(device, index);
@@ -2110,27 +2169,30 @@ nvkm_device_engine(struct nvkm_device *device, int index)
 
 	switch (index) {
 #define _(n,p,m) case NVKM_ENGINE_##n: if (p) return (m); break
-	_(BSP    , device->bsp    ,  device->bsp);
-	_(CE0    , device->ce[0]  ,  device->ce[0]);
-	_(CE1    , device->ce[1]  ,  device->ce[1]);
-	_(CE2    , device->ce[2]  ,  device->ce[2]);
-	_(CIPHER , device->cipher ,  device->cipher);
-	_(DISP   , device->disp   , &device->disp->engine);
-	_(DMAOBJ , device->dma    , &device->dma->engine);
-	_(FIFO   , device->fifo   , &device->fifo->engine);
-	_(GR     , device->gr     , &device->gr->engine);
-	_(IFB    , device->ifb    ,  device->ifb);
-	_(ME     , device->me     ,  device->me);
-	_(MPEG   , device->mpeg   ,  device->mpeg);
-	_(MSENC  , device->msenc  ,  device->msenc);
-	_(MSPDEC , device->mspdec ,  device->mspdec);
-	_(MSPPP  , device->msppp  ,  device->msppp);
-	_(MSVLD  , device->msvld  ,  device->msvld);
-	_(PM     , device->pm     , &device->pm->engine);
-	_(SEC    , device->sec    ,  device->sec);
-	_(SW     , device->sw     , &device->sw->engine);
-	_(VIC    , device->vic    ,  device->vic);
-	_(VP     , device->vp     ,  device->vp);
+	_(BSP    , device->bsp     ,  device->bsp);
+	_(CE0    , device->ce[0]   ,  device->ce[0]);
+	_(CE1    , device->ce[1]   ,  device->ce[1]);
+	_(CE2    , device->ce[2]   ,  device->ce[2]);
+	_(CIPHER , device->cipher  ,  device->cipher);
+	_(DISP   , device->disp    , &device->disp->engine);
+	_(DMAOBJ , device->dma     , &device->dma->engine);
+	_(FIFO   , device->fifo    , &device->fifo->engine);
+	_(GR     , device->gr      , &device->gr->engine);
+	_(IFB    , device->ifb     ,  device->ifb);
+	_(ME     , device->me      ,  device->me);
+	_(MPEG   , device->mpeg    ,  device->mpeg);
+	_(MSENC  , device->msenc   ,  device->msenc);
+	_(MSPDEC , device->mspdec  ,  device->mspdec);
+	_(MSPPP  , device->msppp   ,  device->msppp);
+	_(MSVLD  , device->msvld   ,  device->msvld);
+	_(NVENC0 , device->nvenc[0],  device->nvenc[0]);
+	_(NVENC1 , device->nvenc[1],  device->nvenc[1]);
+	_(NVDEC  , device->nvdec   ,  device->nvdec);
+	_(PM     , device->pm      , &device->pm->engine);
+	_(SEC    , device->sec     ,  device->sec);
+	_(SW     , device->sw      , &device->sw->engine);
+	_(VIC    , device->vic     ,  device->vic);
+	_(VP     , device->vp      ,  device->vp);
 #undef _
 	default:
 		WARN_ON(1);
@@ -2261,6 +2323,8 @@ fail_subdev:
 	} while (--i >= 0);
 
 fail:
+	nvkm_device_fini(device, false);
+
 	nvdev_error(device, "init failed with %d\n", ret);
 	return ret;
 }
@@ -2459,6 +2523,7 @@ nvkm_device_ctor(const struct nvkm_device_func *func,
 		case 0x106: device->chip = &nv106_chipset; break;
 		case 0x108: device->chip = &nv108_chipset; break;
 		case 0x117: device->chip = &nv117_chipset; break;
+		case 0x120: device->chip = &nv120_chipset; break;
 		case 0x124: device->chip = &nv124_chipset; break;
 		case 0x126: device->chip = &nv126_chipset; break;
 		case 0x12b: device->chip = &nv12b_chipset; break;
@@ -2518,47 +2583,52 @@ nvkm_device_ctor(const struct nvkm_device_func *func,
 	}                                                                      \
 	break
 		switch (i) {
-		_(NVKM_SUBDEV_BAR    ,     bar);
-		_(NVKM_SUBDEV_VBIOS  ,    bios);
-		_(NVKM_SUBDEV_BUS    ,     bus);
-		_(NVKM_SUBDEV_CLK    ,     clk);
-		_(NVKM_SUBDEV_DEVINIT, devinit);
-		_(NVKM_SUBDEV_FB     ,      fb);
-		_(NVKM_SUBDEV_FUSE   ,    fuse);
-		_(NVKM_SUBDEV_GPIO   ,    gpio);
-		_(NVKM_SUBDEV_I2C    ,     i2c);
-		_(NVKM_SUBDEV_IBUS   ,    ibus);
-		_(NVKM_SUBDEV_INSTMEM,    imem);
-		_(NVKM_SUBDEV_LTC    ,     ltc);
-		_(NVKM_SUBDEV_MC     ,      mc);
-		_(NVKM_SUBDEV_MMU    ,     mmu);
-		_(NVKM_SUBDEV_MXM    ,     mxm);
-		_(NVKM_SUBDEV_PCI    ,     pci);
-		_(NVKM_SUBDEV_PMU    ,     pmu);
-		_(NVKM_SUBDEV_THERM  ,   therm);
-		_(NVKM_SUBDEV_TIMER  ,   timer);
-		_(NVKM_SUBDEV_VOLT   ,    volt);
-		_(NVKM_ENGINE_BSP    ,     bsp);
-		_(NVKM_ENGINE_CE0    ,   ce[0]);
-		_(NVKM_ENGINE_CE1    ,   ce[1]);
-		_(NVKM_ENGINE_CE2    ,   ce[2]);
-		_(NVKM_ENGINE_CIPHER ,  cipher);
-		_(NVKM_ENGINE_DISP   ,    disp);
-		_(NVKM_ENGINE_DMAOBJ ,     dma);
-		_(NVKM_ENGINE_FIFO   ,    fifo);
-		_(NVKM_ENGINE_GR     ,      gr);
-		_(NVKM_ENGINE_IFB    ,     ifb);
-		_(NVKM_ENGINE_ME     ,      me);
-		_(NVKM_ENGINE_MPEG   ,    mpeg);
-		_(NVKM_ENGINE_MSENC  ,   msenc);
-		_(NVKM_ENGINE_MSPDEC ,  mspdec);
-		_(NVKM_ENGINE_MSPPP  ,   msppp);
-		_(NVKM_ENGINE_MSVLD  ,   msvld);
-		_(NVKM_ENGINE_PM     ,      pm);
-		_(NVKM_ENGINE_SEC    ,     sec);
-		_(NVKM_ENGINE_SW     ,      sw);
-		_(NVKM_ENGINE_VIC    ,     vic);
-		_(NVKM_ENGINE_VP     ,      vp);
+		_(NVKM_SUBDEV_BAR     ,      bar);
+		_(NVKM_SUBDEV_VBIOS   ,     bios);
+		_(NVKM_SUBDEV_BUS     ,      bus);
+		_(NVKM_SUBDEV_CLK     ,      clk);
+		_(NVKM_SUBDEV_DEVINIT ,  devinit);
+		_(NVKM_SUBDEV_FB      ,       fb);
+		_(NVKM_SUBDEV_FUSE    ,     fuse);
+		_(NVKM_SUBDEV_GPIO    ,     gpio);
+		_(NVKM_SUBDEV_I2C     ,      i2c);
+		_(NVKM_SUBDEV_IBUS    ,     ibus);
+		_(NVKM_SUBDEV_ICCSENSE, iccsense);
+		_(NVKM_SUBDEV_INSTMEM ,     imem);
+		_(NVKM_SUBDEV_LTC     ,      ltc);
+		_(NVKM_SUBDEV_MC      ,       mc);
+		_(NVKM_SUBDEV_MMU     ,      mmu);
+		_(NVKM_SUBDEV_MXM     ,      mxm);
+		_(NVKM_SUBDEV_PCI     ,      pci);
+		_(NVKM_SUBDEV_PMU     ,      pmu);
+		_(NVKM_SUBDEV_SECBOOT ,  secboot);
+		_(NVKM_SUBDEV_THERM   ,    therm);
+		_(NVKM_SUBDEV_TIMER   ,    timer);
+		_(NVKM_SUBDEV_VOLT    ,     volt);
+		_(NVKM_ENGINE_BSP     ,      bsp);
+		_(NVKM_ENGINE_CE0     ,    ce[0]);
+		_(NVKM_ENGINE_CE1     ,    ce[1]);
+		_(NVKM_ENGINE_CE2     ,    ce[2]);
+		_(NVKM_ENGINE_CIPHER  ,   cipher);
+		_(NVKM_ENGINE_DISP    ,     disp);
+		_(NVKM_ENGINE_DMAOBJ  ,      dma);
+		_(NVKM_ENGINE_FIFO    ,     fifo);
+		_(NVKM_ENGINE_GR      ,       gr);
+		_(NVKM_ENGINE_IFB     ,      ifb);
+		_(NVKM_ENGINE_ME      ,       me);
+		_(NVKM_ENGINE_MPEG    ,     mpeg);
+		_(NVKM_ENGINE_MSENC   ,    msenc);
+		_(NVKM_ENGINE_MSPDEC  ,   mspdec);
+		_(NVKM_ENGINE_MSPPP   ,    msppp);
+		_(NVKM_ENGINE_MSVLD   ,    msvld);
+		_(NVKM_ENGINE_NVENC0  , nvenc[0]);
+		_(NVKM_ENGINE_NVENC1  , nvenc[1]);
+		_(NVKM_ENGINE_NVDEC   ,    nvdec);
+		_(NVKM_ENGINE_PM      ,       pm);
+		_(NVKM_ENGINE_SEC     ,      sec);
+		_(NVKM_ENGINE_SW      ,       sw);
+		_(NVKM_ENGINE_VIC     ,      vic);
+		_(NVKM_ENGINE_VP      ,       vp);
 		default:
 			WARN_ON(1);
 			continue;
diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/device/pci.c b/drivers/gpu/drm/nouveau/nvkm/engine/device/pci.c
index 62ad0300cfa5..18fab3973ce5 100644
--- a/drivers/gpu/drm/nouveau/nvkm/engine/device/pci.c
+++ b/drivers/gpu/drm/nouveau/nvkm/engine/device/pci.c
@@ -1614,7 +1614,7 @@ nvkm_device_pci_func = {
 	.fini = nvkm_device_pci_fini,
 	.resource_addr = nvkm_device_pci_resource_addr,
 	.resource_size = nvkm_device_pci_resource_size,
-	.cpu_coherent = !IS_ENABLED(CONFIG_ARM),
+	.cpu_coherent = !IS_ENABLED(CONFIG_ARM) && !IS_ENABLED(CONFIG_ARM64),
 };
 
 int
diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/device/priv.h b/drivers/gpu/drm/nouveau/nvkm/engine/device/priv.h
index ed3ad2c30e17..e80f6ab1c415 100644
--- a/drivers/gpu/drm/nouveau/nvkm/engine/device/priv.h
+++ b/drivers/gpu/drm/nouveau/nvkm/engine/device/priv.h
@@ -12,6 +12,7 @@
 #include <subdev/gpio.h>
 #include <subdev/i2c.h>
 #include <subdev/ibus.h>
+#include <subdev/iccsense.h>
 #include <subdev/instmem.h>
 #include <subdev/ltc.h>
 #include <subdev/mc.h>
@@ -22,6 +23,7 @@
 #include <subdev/therm.h>
 #include <subdev/timer.h>
 #include <subdev/volt.h>
+#include <subdev/secboot.h>
 
 #include <engine/bsp.h>
 #include <engine/ce.h>
@@ -34,9 +36,12 @@
 #include <engine/mspdec.h>
 #include <engine/msppp.h>
 #include <engine/msvld.h>
+#include <engine/nvenc.h>
+#include <engine/nvdec.h>
 #include <engine/pm.h>
 #include <engine/sec.h>
 #include <engine/sw.h>
+#include <engine/vic.h>
 #include <engine/vp.h>
 
 int  nvkm_device_ctor(const struct nvkm_device_func *,
diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/device/tegra.c b/drivers/gpu/drm/nouveau/nvkm/engine/device/tegra.c
index 7f8a42721eb2..9afa5f3e3c1c 100644
--- a/drivers/gpu/drm/nouveau/nvkm/engine/device/tegra.c
+++ b/drivers/gpu/drm/nouveau/nvkm/engine/device/tegra.c
@@ -252,32 +252,48 @@ nvkm_device_tegra_new(const struct nvkm_device_tegra_func *func,
 
 	if (!(tdev = kzalloc(sizeof(*tdev), GFP_KERNEL)))
 		return -ENOMEM;
-	*pdevice = &tdev->device;
+
 	tdev->func = func;
 	tdev->pdev = pdev;
-	tdev->irq = -1;
 
 	tdev->vdd = devm_regulator_get(&pdev->dev, "vdd");
-	if (IS_ERR(tdev->vdd))
-		return PTR_ERR(tdev->vdd);
+	if (IS_ERR(tdev->vdd)) {
+		ret = PTR_ERR(tdev->vdd);
+		goto free;
+	}
 
 	tdev->rst = devm_reset_control_get(&pdev->dev, "gpu");
-	if (IS_ERR(tdev->rst))
-		return PTR_ERR(tdev->rst);
+	if (IS_ERR(tdev->rst)) {
+		ret = PTR_ERR(tdev->rst);
+		goto free;
+	}
 
 	tdev->clk = devm_clk_get(&pdev->dev, "gpu");
-	if (IS_ERR(tdev->clk))
-		return PTR_ERR(tdev->clk);
+	if (IS_ERR(tdev->clk)) {
+		ret = PTR_ERR(tdev->clk);
+		goto free;
+	}
 
 	tdev->clk_pwr = devm_clk_get(&pdev->dev, "pwr");
-	if (IS_ERR(tdev->clk_pwr))
-		return PTR_ERR(tdev->clk_pwr);
+	if (IS_ERR(tdev->clk_pwr)) {
+		ret = PTR_ERR(tdev->clk_pwr);
+		goto free;
+	}
+
+	/**
+	 * The IOMMU bit defines the upper limit of the GPU-addressable space.
+	 * This will be refined in nouveau_ttm_init but we need to do it early
+	 * for instmem to behave properly
+	 */
+	ret = dma_set_mask(&pdev->dev, DMA_BIT_MASK(tdev->func->iommu_bit));
+	if (ret)
+		goto free;
 
 	nvkm_device_tegra_probe_iommu(tdev);
 
 	ret = nvkm_device_tegra_power_up(tdev);
 	if (ret)
-		return ret;
+		goto remove;
 
 	tdev->gpu_speedo = tegra_sku_info.gpu_speedo_value;
 	ret = nvkm_device_ctor(&nvkm_device_tegra_func, NULL, &pdev->dev,
@@ -285,9 +301,19 @@ nvkm_device_tegra_new(const struct nvkm_device_tegra_func *func,
 			       cfg, dbg, detect, mmio, subdev_mask,
 			       &tdev->device);
 	if (ret)
-		return ret;
+		goto powerdown;
+
+	*pdevice = &tdev->device;
 
 	return 0;
+
+powerdown:
+	nvkm_device_tegra_power_down(tdev);
+remove:
+	nvkm_device_tegra_remove_iommu(tdev);
+free:
+	kfree(tdev);
+	return ret;
 }
 #else
 int
diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/disp/Kbuild b/drivers/gpu/drm/nouveau/nvkm/engine/disp/Kbuild
index 04f60452011e..a74c5dd27dc0 100644
--- a/drivers/gpu/drm/nouveau/nvkm/engine/disp/Kbuild
+++ b/drivers/gpu/drm/nouveau/nvkm/engine/disp/Kbuild
@@ -9,7 +9,7 @@ nvkm-y += nvkm/engine/disp/gf119.o
 nvkm-y += nvkm/engine/disp/gk104.o
 nvkm-y += nvkm/engine/disp/gk110.o
 nvkm-y += nvkm/engine/disp/gm107.o
-nvkm-y += nvkm/engine/disp/gm204.o
+nvkm-y += nvkm/engine/disp/gm200.o
 
 nvkm-y += nvkm/engine/disp/outp.o
 nvkm-y += nvkm/engine/disp/outpdp.o
@@ -18,7 +18,7 @@ nvkm-y += nvkm/engine/disp/piornv50.o
 nvkm-y += nvkm/engine/disp/sornv50.o
 nvkm-y += nvkm/engine/disp/sorg94.o
 nvkm-y += nvkm/engine/disp/sorgf119.o
-nvkm-y += nvkm/engine/disp/sorgm204.o
+nvkm-y += nvkm/engine/disp/sorgm200.o
 nvkm-y += nvkm/engine/disp/dport.o
 
 nvkm-y += nvkm/engine/disp/conn.o
@@ -43,7 +43,7 @@ nvkm-y += nvkm/engine/disp/rootgf119.o
 nvkm-y += nvkm/engine/disp/rootgk104.o
 nvkm-y += nvkm/engine/disp/rootgk110.o
 nvkm-y += nvkm/engine/disp/rootgm107.o
-nvkm-y += nvkm/engine/disp/rootgm204.o
+nvkm-y += nvkm/engine/disp/rootgm200.o
 
 nvkm-y += nvkm/engine/disp/channv50.o
 nvkm-y += nvkm/engine/disp/changf119.o
@@ -68,7 +68,7 @@ nvkm-y += nvkm/engine/disp/coregf119.o
 nvkm-y += nvkm/engine/disp/coregk104.o
 nvkm-y += nvkm/engine/disp/coregk110.o
 nvkm-y += nvkm/engine/disp/coregm107.o
-nvkm-y += nvkm/engine/disp/coregm204.o
+nvkm-y += nvkm/engine/disp/coregm200.o
 
 nvkm-y += nvkm/engine/disp/ovlynv50.o
 nvkm-y += nvkm/engine/disp/ovlyg84.o
diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/disp/coregm204.c b/drivers/gpu/drm/nouveau/nvkm/engine/disp/coregm200.c
index 222f4a822f4d..bb23a8658ac0 100644
--- a/drivers/gpu/drm/nouveau/nvkm/engine/disp/coregm204.c
+++ b/drivers/gpu/drm/nouveau/nvkm/engine/disp/coregm200.c
@@ -27,8 +27,8 @@
 #include <nvif/class.h>
 
 const struct nv50_disp_dmac_oclass
-gm204_disp_core_oclass = {
-	.base.oclass = GM204_DISP_CORE_CHANNEL_DMA,
+gm200_disp_core_oclass = {
+	.base.oclass = GM200_DISP_CORE_CHANNEL_DMA,
 	.base.minver = 0,
 	.base.maxver = 0,
 	.ctor = nv50_disp_core_new,
diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/disp/dmacnv50.h b/drivers/gpu/drm/nouveau/nvkm/engine/disp/dmacnv50.h
index c748ca23ab70..fc84eb8b5c45 100644
--- a/drivers/gpu/drm/nouveau/nvkm/engine/disp/dmacnv50.h
+++ b/drivers/gpu/drm/nouveau/nvkm/engine/disp/dmacnv50.h
@@ -87,5 +87,5 @@ extern const struct nv50_disp_dmac_oclass gk110_disp_base_oclass;
 
 extern const struct nv50_disp_dmac_oclass gm107_disp_core_oclass;
 
-extern const struct nv50_disp_dmac_oclass gm204_disp_core_oclass;
+extern const struct nv50_disp_dmac_oclass gm200_disp_core_oclass;
 #endif
diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/disp/dport.c b/drivers/gpu/drm/nouveau/nvkm/engine/disp/dport.c
index 74e2f7c6c07e..9688970eca47 100644
--- a/drivers/gpu/drm/nouveau/nvkm/engine/disp/dport.c
+++ b/drivers/gpu/drm/nouveau/nvkm/engine/disp/dport.c
@@ -328,6 +328,7 @@ nvkm_dp_train(struct work_struct *w)
 		.outp = outp,
 	}, *dp = &_dp;
 	u32 datarate = 0;
+	u8  pwr;
 	int ret;
 
 	if (!outp->base.info.location && disp->func->sor.magic)
@@ -355,6 +356,15 @@ nvkm_dp_train(struct work_struct *w)
 	/* disable link interrupt handling during link training */
 	nvkm_notify_put(&outp->irq);
 
+	/* ensure sink is not in a low-power state */
+	if (!nvkm_rdaux(outp->aux, DPCD_SC00, &pwr, 1)) {
+		if ((pwr & DPCD_SC00_SET_POWER) != DPCD_SC00_SET_POWER_D0) {
+			pwr &= ~DPCD_SC00_SET_POWER;
+			pwr |=  DPCD_SC00_SET_POWER_D0;
+			nvkm_wraux(outp->aux, DPCD_SC00, &pwr, 1);
+		}
+	}
+
 	/* enable down-spreading and execute pre-train script from vbios */
 	dp_link_train_init(dp, outp->dpcd[3] & 0x01);
 
diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/disp/dport.h b/drivers/gpu/drm/nouveau/nvkm/engine/disp/dport.h
index 9596290329c7..6e10c5e0ef11 100644
--- a/drivers/gpu/drm/nouveau/nvkm/engine/disp/dport.h
+++ b/drivers/gpu/drm/nouveau/nvkm/engine/disp/dport.h
@@ -71,5 +71,11 @@
 #define DPCD_LS0C_LANE1_POST_CURSOR2                                       0x0c
 #define DPCD_LS0C_LANE0_POST_CURSOR2                                       0x03
 
+/* DPCD Sink Control */
+#define DPCD_SC00                                                       0x00600
+#define DPCD_SC00_SET_POWER                                                0x03
+#define DPCD_SC00_SET_POWER_D0                                             0x01
+#define DPCD_SC00_SET_POWER_D3                                             0x03
+
 void nvkm_dp_train(struct work_struct *);
 #endif
diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/disp/gm204.c b/drivers/gpu/drm/nouveau/nvkm/engine/disp/gm200.c
index 30f1987b5b40..67eec8620719 100644
--- a/drivers/gpu/drm/nouveau/nvkm/engine/disp/gm204.c
+++ b/drivers/gpu/drm/nouveau/nvkm/engine/disp/gm200.c
@@ -25,18 +25,18 @@
 #include "rootnv50.h"
 
 static const struct nv50_disp_func
-gm204_disp = {
+gm200_disp = {
 	.intr = gf119_disp_intr,
 	.uevent = &gf119_disp_chan_uevent,
 	.super = gf119_disp_intr_supervisor,
-	.root = &gm204_disp_root_oclass,
+	.root = &gm200_disp_root_oclass,
 	.head.vblank_init = gf119_disp_vblank_init,
 	.head.vblank_fini = gf119_disp_vblank_fini,
 	.head.scanoutpos = gf119_disp_root_scanoutpos,
 	.outp.internal.crt = nv50_dac_output_new,
 	.outp.internal.tmds = nv50_sor_output_new,
 	.outp.internal.lvds = nv50_sor_output_new,
-	.outp.internal.dp = gm204_sor_dp_new,
+	.outp.internal.dp = gm200_sor_dp_new,
 	.dac.nr = 3,
 	.dac.power = nv50_dac_power,
 	.dac.sense = nv50_dac_sense,
@@ -44,11 +44,11 @@ gm204_disp = {
 	.sor.power = nv50_sor_power,
 	.sor.hda_eld = gf119_hda_eld,
 	.sor.hdmi = gk104_hdmi_ctrl,
-	.sor.magic = gm204_sor_magic,
+	.sor.magic = gm200_sor_magic,
 };
 
 int
-gm204_disp_new(struct nvkm_device *device, int index, struct nvkm_disp **pdisp)
+gm200_disp_new(struct nvkm_device *device, int index, struct nvkm_disp **pdisp)
 {
-	return gf119_disp_new_(&gm204_disp, device, index, pdisp);
+	return gf119_disp_new_(&gm200_disp, device, index, pdisp);
 }
diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/disp/outp.h b/drivers/gpu/drm/nouveau/nvkm/engine/disp/outp.h
index 2590fec67ca9..07727198d7ce 100644
--- a/drivers/gpu/drm/nouveau/nvkm/engine/disp/outp.h
+++ b/drivers/gpu/drm/nouveau/nvkm/engine/disp/outp.h
@@ -42,7 +42,7 @@ int nv50_pior_output_new(struct nvkm_disp *, int, struct dcb_output *,
 
 u32 g94_sor_dp_lane_map(struct nvkm_device *, u8 lane);
 
-void gm204_sor_magic(struct nvkm_output *outp);
+void gm200_sor_magic(struct nvkm_output *outp);
 
 #define OUTP_MSG(o,l,f,a...) do {                                              \
 	struct nvkm_output *_outp = (o);                                       \
diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/disp/outpdp.h b/drivers/gpu/drm/nouveau/nvkm/engine/disp/outpdp.h
index 731136d660b7..e9067ba4e179 100644
--- a/drivers/gpu/drm/nouveau/nvkm/engine/disp/outpdp.h
+++ b/drivers/gpu/drm/nouveau/nvkm/engine/disp/outpdp.h
@@ -63,6 +63,6 @@ int gf119_sor_dp_new(struct nvkm_disp *, int, struct dcb_output *,
 		     struct nvkm_output **);
 int gf119_sor_dp_lnk_ctl(struct nvkm_output_dp *, int, int, bool);
 
-int  gm204_sor_dp_new(struct nvkm_disp *, int, struct dcb_output *,
+int  gm200_sor_dp_new(struct nvkm_disp *, int, struct dcb_output *,
 		      struct nvkm_output **);
 #endif
diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/disp/rootgm204.c b/drivers/gpu/drm/nouveau/nvkm/engine/disp/rootgm200.c
index 168bffe0643c..38f5ee1dfc58 100644
--- a/drivers/gpu/drm/nouveau/nvkm/engine/disp/rootgm204.c
+++ b/drivers/gpu/drm/nouveau/nvkm/engine/disp/rootgm200.c
@@ -27,11 +27,11 @@
 #include <nvif/class.h>
 
 static const struct nv50_disp_root_func
-gm204_disp_root = {
+gm200_disp_root = {
 	.init = gf119_disp_root_init,
 	.fini = gf119_disp_root_fini,
 	.dmac = {
-		&gm204_disp_core_oclass,
+		&gm200_disp_core_oclass,
 		&gk110_disp_base_oclass,
 		&gk104_disp_ovly_oclass,
 	},
@@ -42,17 +42,17 @@ gm204_disp_root = {
 };
 
 static int
-gm204_disp_root_new(struct nvkm_disp *disp, const struct nvkm_oclass *oclass,
+gm200_disp_root_new(struct nvkm_disp *disp, const struct nvkm_oclass *oclass,
 		    void *data, u32 size, struct nvkm_object **pobject)
 {
-	return nv50_disp_root_new_(&gm204_disp_root, disp, oclass,
+	return nv50_disp_root_new_(&gm200_disp_root, disp, oclass,
 				   data, size, pobject);
 }
 
 const struct nvkm_disp_oclass
-gm204_disp_root_oclass = {
-	.base.oclass = GM204_DISP,
+gm200_disp_root_oclass = {
+	.base.oclass = GM200_DISP,
 	.base.minver = -1,
 	.base.maxver = -1,
-	.ctor = gm204_disp_root_new,
+	.ctor = gm200_disp_root_new,
 };
diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/disp/rootnv50.h b/drivers/gpu/drm/nouveau/nvkm/engine/disp/rootnv50.h
index 5b2c903ce9ee..cb449ed8d92c 100644
--- a/drivers/gpu/drm/nouveau/nvkm/engine/disp/rootnv50.h
+++ b/drivers/gpu/drm/nouveau/nvkm/engine/disp/rootnv50.h
@@ -39,5 +39,5 @@ extern const struct nvkm_disp_oclass gf119_disp_root_oclass;
 extern const struct nvkm_disp_oclass gk104_disp_root_oclass;
 extern const struct nvkm_disp_oclass gk110_disp_root_oclass;
 extern const struct nvkm_disp_oclass gm107_disp_root_oclass;
-extern const struct nvkm_disp_oclass gm204_disp_root_oclass;
+extern const struct nvkm_disp_oclass gm200_disp_root_oclass;
 #endif
diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/disp/sorgm204.c b/drivers/gpu/drm/nouveau/nvkm/engine/disp/sorgm200.c
index 029e5f16c2a8..2cfbef9c344f 100644
--- a/drivers/gpu/drm/nouveau/nvkm/engine/disp/sorgm204.c
+++ b/drivers/gpu/drm/nouveau/nvkm/engine/disp/sorgm200.c
@@ -27,19 +27,19 @@
 #include <subdev/timer.h>
 
 static inline u32
-gm204_sor_soff(struct nvkm_output_dp *outp)
+gm200_sor_soff(struct nvkm_output_dp *outp)
 {
 	return (ffs(outp->base.info.or) - 1) * 0x800;
 }
 
 static inline u32
-gm204_sor_loff(struct nvkm_output_dp *outp)
+gm200_sor_loff(struct nvkm_output_dp *outp)
 {
-	return gm204_sor_soff(outp) + !(outp->base.info.sorconf.link & 1) * 0x80;
+	return gm200_sor_soff(outp) + !(outp->base.info.sorconf.link & 1) * 0x80;
 }
 
 void
-gm204_sor_magic(struct nvkm_output *outp)
+gm200_sor_magic(struct nvkm_output *outp)
 {
 	struct nvkm_device *device = outp->disp->engine.subdev.device;
 	const u32 soff = outp->or * 0x100;
@@ -51,16 +51,16 @@ gm204_sor_magic(struct nvkm_output *outp)
 }
 
 static inline u32
-gm204_sor_dp_lane_map(struct nvkm_device *device, u8 lane)
+gm200_sor_dp_lane_map(struct nvkm_device *device, u8 lane)
 {
 	return lane * 0x08;
 }
 
 static int
-gm204_sor_dp_pattern(struct nvkm_output_dp *outp, int pattern)
+gm200_sor_dp_pattern(struct nvkm_output_dp *outp, int pattern)
 {
 	struct nvkm_device *device = outp->base.disp->engine.subdev.device;
-	const u32 soff = gm204_sor_soff(outp);
+	const u32 soff = gm200_sor_soff(outp);
 	const u32 data = 0x01010101 * pattern;
 	if (outp->base.info.sorconf.link & 1)
 		nvkm_mask(device, 0x61c110 + soff, 0x0f0f0f0f, data);
@@ -70,15 +70,15 @@ gm204_sor_dp_pattern(struct nvkm_output_dp *outp, int pattern)
 }
 
 static int
-gm204_sor_dp_lnk_pwr(struct nvkm_output_dp *outp, int nr)
+gm200_sor_dp_lnk_pwr(struct nvkm_output_dp *outp, int nr)
 {
 	struct nvkm_device *device = outp->base.disp->engine.subdev.device;
-	const u32 soff = gm204_sor_soff(outp);
-	const u32 loff = gm204_sor_loff(outp);
+	const u32 soff = gm200_sor_soff(outp);
+	const u32 loff = gm200_sor_loff(outp);
 	u32 mask = 0, i;
 
 	for (i = 0; i < nr; i++)
-		mask |= 1 << (gm204_sor_dp_lane_map(device, i) >> 3);
+		mask |= 1 << (gm200_sor_dp_lane_map(device, i) >> 3);
 
 	nvkm_mask(device, 0x61c130 + loff, 0x0000000f, mask);
 	nvkm_mask(device, 0x61c034 + soff, 0x80000000, 0x80000000);
@@ -90,13 +90,13 @@ gm204_sor_dp_lnk_pwr(struct nvkm_output_dp *outp, int nr)
 }
 
 static int
-gm204_sor_dp_drv_ctl(struct nvkm_output_dp *outp,
+gm200_sor_dp_drv_ctl(struct nvkm_output_dp *outp,
 		     int ln, int vs, int pe, int pc)
 {
 	struct nvkm_device *device = outp->base.disp->engine.subdev.device;
 	struct nvkm_bios *bios = device->bios;
-	const u32 shift = gm204_sor_dp_lane_map(device, ln);
-	const u32 loff = gm204_sor_loff(outp);
+	const u32 shift = gm200_sor_dp_lane_map(device, ln);
+	const u32 loff = gm200_sor_loff(outp);
 	u32 addr, data[4];
 	u8  ver, hdr, cnt, len;
 	struct nvbios_dpout info;
@@ -128,16 +128,16 @@ gm204_sor_dp_drv_ctl(struct nvkm_output_dp *outp,
 }
 
 static const struct nvkm_output_dp_func
-gm204_sor_dp_func = {
-	.pattern = gm204_sor_dp_pattern,
-	.lnk_pwr = gm204_sor_dp_lnk_pwr,
+gm200_sor_dp_func = {
+	.pattern = gm200_sor_dp_pattern,
+	.lnk_pwr = gm200_sor_dp_lnk_pwr,
 	.lnk_ctl = gf119_sor_dp_lnk_ctl,
-	.drv_ctl = gm204_sor_dp_drv_ctl,
+	.drv_ctl = gm200_sor_dp_drv_ctl,
 };
 
 int
-gm204_sor_dp_new(struct nvkm_disp *disp, int index, struct dcb_output *dcbE,
+gm200_sor_dp_new(struct nvkm_disp *disp, int index, struct dcb_output *dcbE,
 		 struct nvkm_output **poutp)
 {
-	return nvkm_output_dp_new_(&gm204_sor_dp_func, disp, index, dcbE, poutp);
+	return nvkm_output_dp_new_(&gm200_sor_dp_func, disp, index, dcbE, poutp);
 }
diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/fifo/Kbuild b/drivers/gpu/drm/nouveau/nvkm/engine/fifo/Kbuild
index 74993c144a84..65e5d291ecda 100644
--- a/drivers/gpu/drm/nouveau/nvkm/engine/fifo/Kbuild
+++ b/drivers/gpu/drm/nouveau/nvkm/engine/fifo/Kbuild
@@ -7,9 +7,11 @@ nvkm-y += nvkm/engine/fifo/nv50.o
 nvkm-y += nvkm/engine/fifo/g84.o
 nvkm-y += nvkm/engine/fifo/gf100.o
 nvkm-y += nvkm/engine/fifo/gk104.o
+nvkm-y += nvkm/engine/fifo/gk110.o
 nvkm-y += nvkm/engine/fifo/gk208.o
 nvkm-y += nvkm/engine/fifo/gk20a.o
-nvkm-y += nvkm/engine/fifo/gm204.o
+nvkm-y += nvkm/engine/fifo/gm107.o
+nvkm-y += nvkm/engine/fifo/gm200.o
 nvkm-y += nvkm/engine/fifo/gm20b.o
 
 nvkm-y += nvkm/engine/fifo/chan.o
@@ -27,4 +29,5 @@ nvkm-y += nvkm/engine/fifo/gpfifonv50.o
 nvkm-y += nvkm/engine/fifo/gpfifog84.o
 nvkm-y += nvkm/engine/fifo/gpfifogf100.o
 nvkm-y += nvkm/engine/fifo/gpfifogk104.o
-nvkm-y += nvkm/engine/fifo/gpfifogm204.o
+nvkm-y += nvkm/engine/fifo/gpfifogk110.o
+nvkm-y += nvkm/engine/fifo/gpfifogm200.o
diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/fifo/changk104.h b/drivers/gpu/drm/nouveau/nvkm/engine/fifo/changk104.h
index 97bdddb7644a..e06f4d46f802 100644
--- a/drivers/gpu/drm/nouveau/nvkm/engine/fifo/changk104.h
+++ b/drivers/gpu/drm/nouveau/nvkm/engine/fifo/changk104.h
@@ -7,7 +7,7 @@
 struct gk104_fifo_chan {
 	struct nvkm_fifo_chan base;
 	struct gk104_fifo *fifo;
-	int engine;
+	int runl;
 
 	struct list_head head;
 	bool killed;
@@ -25,5 +25,6 @@ int gk104_fifo_gpfifo_new(struct nvkm_fifo *, const struct nvkm_oclass *,
 			  void *data, u32 size, struct nvkm_object **);
 
 extern const struct nvkm_fifo_chan_oclass gk104_fifo_gpfifo_oclass;
-extern const struct nvkm_fifo_chan_oclass gm204_fifo_gpfifo_oclass;
+extern const struct nvkm_fifo_chan_oclass gk110_fifo_gpfifo_oclass;
+extern const struct nvkm_fifo_chan_oclass gm200_fifo_gpfifo_oclass;
 #endif
diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/fifo/gf100.c b/drivers/gpu/drm/nouveau/nvkm/engine/fifo/gf100.c
index 36a39c7fd8d2..352a0baec84d 100644
--- a/drivers/gpu/drm/nouveau/nvkm/engine/fifo/gf100.c
+++ b/drivers/gpu/drm/nouveau/nvkm/engine/fifo/gf100.c
@@ -54,6 +54,7 @@ gf100_fifo_runlist_commit(struct gf100_fifo *fifo)
 	struct nvkm_device *device = subdev->device;
 	struct nvkm_memory *cur;
 	int nr = 0;
+	int target;
 
 	mutex_lock(&subdev->mutex);
 	cur = fifo->runlist.mem[fifo->runlist.active];
@@ -67,7 +68,10 @@ gf100_fifo_runlist_commit(struct gf100_fifo *fifo)
 	}
 	nvkm_done(cur);
 
-	nvkm_wr32(device, 0x002270, nvkm_memory_addr(cur) >> 12);
+	target = (nvkm_memory_target(cur) == NVKM_MEM_TARGET_HOST) ? 0x3 : 0x0;
+
+	nvkm_wr32(device, 0x002270, (nvkm_memory_addr(cur) >> 12) |
+				    (target << 28));
 	nvkm_wr32(device, 0x002274, 0x01f00000 | nr);
 
 	if (wait_event_timeout(fifo->runlist.wait,
@@ -130,9 +134,9 @@ gf100_fifo_engine(struct gf100_fifo *fifo, u32 engn)
 }
 
 static void
-gf100_fifo_recover_work(struct work_struct *work)
+gf100_fifo_recover_work(struct work_struct *w)
 {
-	struct gf100_fifo *fifo = container_of(work, typeof(*fifo), fault);
+	struct gf100_fifo *fifo = container_of(w, typeof(*fifo), recover.work);
 	struct nvkm_device *device = fifo->base.engine.subdev.device;
 	struct nvkm_engine *engine;
 	unsigned long flags;
@@ -140,15 +144,15 @@ gf100_fifo_recover_work(struct work_struct *work)
 	u64 mask, todo;
 
 	spin_lock_irqsave(&fifo->base.lock, flags);
-	mask = fifo->mask;
-	fifo->mask = 0ULL;
+	mask = fifo->recover.mask;
+	fifo->recover.mask = 0ULL;
 	spin_unlock_irqrestore(&fifo->base.lock, flags);
 
-	for (todo = mask; engn = __ffs64(todo), todo; todo &= ~(1 << engn))
+	for (todo = mask; engn = __ffs64(todo), todo; todo &= ~BIT_ULL(engn))
 		engm |= 1 << gf100_fifo_engidx(fifo, engn);
 	nvkm_mask(device, 0x002630, engm, engm);
 
-	for (todo = mask; engn = __ffs64(todo), todo; todo &= ~(1 << engn)) {
+	for (todo = mask; engn = __ffs64(todo), todo; todo &= ~BIT_ULL(engn)) {
 		if ((engine = nvkm_device_engine(device, engn))) {
 			nvkm_subdev_fini(&engine->subdev, false);
 			WARN_ON(nvkm_subdev_init(&engine->subdev));
@@ -176,8 +180,8 @@ gf100_fifo_recover(struct gf100_fifo *fifo, struct nvkm_engine *engine,
 	list_del_init(&chan->head);
 	chan->killed = true;
 
-	fifo->mask |= 1ULL << engine->subdev.index;
-	schedule_work(&fifo->fault);
+	fifo->recover.mask |= 1ULL << engine->subdev.index;
+	schedule_work(&fifo->recover.work);
 }
 
 static const struct nvkm_enum
@@ -330,7 +334,7 @@ gf100_fifo_intr_fault(struct gf100_fifo *fifo, int unit)
 		snprintf(gpcid, sizeof(gpcid), "GPC%d/", gpc);
 	}
 
-	if (eu) {
+	if (eu && eu->data2) {
 		switch (eu->data2) {
 		case NVKM_SUBDEV_BAR:
 			nvkm_mask(device, 0x001704, 0x00000000, 0x00000000);
@@ -544,9 +548,16 @@ static int
 gf100_fifo_oneinit(struct nvkm_fifo *base)
 {
 	struct gf100_fifo *fifo = gf100_fifo(base);
-	struct nvkm_device *device = fifo->base.engine.subdev.device;
+	struct nvkm_subdev *subdev = &fifo->base.engine.subdev;
+	struct nvkm_device *device = subdev->device;
 	int ret;
 
+	/* Determine number of PBDMAs by checking valid enable bits. */
+	nvkm_wr32(device, 0x002204, 0xffffffff);
+	fifo->pbdma_nr = hweight32(nvkm_rd32(device, 0x002204));
+	nvkm_debug(subdev, "%d PBDMA(s)\n", fifo->pbdma_nr);
+
+
 	ret = nvkm_memory_new(device, NVKM_MEM_TARGET_INST, 0x1000, 0x1000,
 			      false, &fifo->runlist.mem[0]);
 	if (ret)
@@ -576,25 +587,22 @@ static void
 gf100_fifo_fini(struct nvkm_fifo *base)
 {
 	struct gf100_fifo *fifo = gf100_fifo(base);
-	flush_work(&fifo->fault);
+	flush_work(&fifo->recover.work);
 }
 
 static void
 gf100_fifo_init(struct nvkm_fifo *base)
 {
 	struct gf100_fifo *fifo = gf100_fifo(base);
-	struct nvkm_subdev *subdev = &fifo->base.engine.subdev;
-	struct nvkm_device *device = subdev->device;
+	struct nvkm_device *device = fifo->base.engine.subdev.device;
 	int i;
 
-	nvkm_wr32(device, 0x000204, 0xffffffff);
-	nvkm_wr32(device, 0x002204, 0xffffffff);
-
-	fifo->spoon_nr = hweight32(nvkm_rd32(device, 0x002204));
-	nvkm_debug(subdev, "%d PBDMA unit(s)\n", fifo->spoon_nr);
+	/* Enable PBDMAs. */
+	nvkm_wr32(device, 0x000204, (1 << fifo->pbdma_nr) - 1);
+	nvkm_wr32(device, 0x002204, (1 << fifo->pbdma_nr) - 1);
 
-	/* assign engines to PBDMAs */
-	if (fifo->spoon_nr >= 3) {
+	/* Assign engines to PBDMAs. */
+	if (fifo->pbdma_nr >= 3) {
 		nvkm_wr32(device, 0x002208, ~(1 << 0)); /* PGRAPH */
 		nvkm_wr32(device, 0x00220c, ~(1 << 1)); /* PVP */
 		nvkm_wr32(device, 0x002210, ~(1 << 1)); /* PMSPP */
@@ -604,7 +612,7 @@ gf100_fifo_init(struct nvkm_fifo *base)
 	}
 
 	/* PBDMA[n] */
-	for (i = 0; i < fifo->spoon_nr; i++) {
+	for (i = 0; i < fifo->pbdma_nr; i++) {
 		nvkm_mask(device, 0x04013c + (i * 0x2000), 0x10000100, 0x00000000);
 		nvkm_wr32(device, 0x040108 + (i * 0x2000), 0xffffffff); /* INTR */
 		nvkm_wr32(device, 0x04010c + (i * 0x2000), 0xfffffeff); /* INTREN */
@@ -652,7 +660,7 @@ gf100_fifo_new(struct nvkm_device *device, int index, struct nvkm_fifo **pfifo)
 	if (!(fifo = kzalloc(sizeof(*fifo), GFP_KERNEL)))
 		return -ENOMEM;
 	INIT_LIST_HEAD(&fifo->chan);
-	INIT_WORK(&fifo->fault, gf100_fifo_recover_work);
+	INIT_WORK(&fifo->recover.work, gf100_fifo_recover_work);
 	*pfifo = &fifo->base;
 
 	return nvkm_fifo_ctor(&gf100_fifo, device, index, 128, &fifo->base);
diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/fifo/gf100.h b/drivers/gpu/drm/nouveau/nvkm/engine/fifo/gf100.h
index 08c33c3ceaf7..70db58eab9c3 100644
--- a/drivers/gpu/drm/nouveau/nvkm/engine/fifo/gf100.h
+++ b/drivers/gpu/drm/nouveau/nvkm/engine/fifo/gf100.h
@@ -11,8 +11,12 @@ struct gf100_fifo {
 
 	struct list_head chan;
 
-	struct work_struct fault;
-	u64 mask;
+	struct {
+		struct work_struct work;
+		u64 mask;
+	} recover;
+
+	int pbdma_nr;
 
 	struct {
 		struct nvkm_memory *mem[2];
@@ -24,7 +28,6 @@ struct gf100_fifo {
 		struct nvkm_memory *mem;
 		struct nvkm_vma bar;
 	} user;
-	int spoon_nr;
 };
 
 void gf100_fifo_intr_engine(struct gf100_fifo *);
diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/fifo/gk104.c b/drivers/gpu/drm/nouveau/nvkm/engine/fifo/gk104.c
index 4fcd147d43c8..68acb36b3e6d 100644
--- a/drivers/gpu/drm/nouveau/nvkm/engine/fifo/gk104.c
+++ b/drivers/gpu/drm/nouveau/nvkm/engine/fifo/gk104.c
@@ -47,34 +47,41 @@ gk104_fifo_uevent_init(struct nvkm_fifo *fifo)
 }
 
 void
-gk104_fifo_runlist_commit(struct gk104_fifo *fifo, u32 engine)
+gk104_fifo_runlist_commit(struct gk104_fifo *fifo, int runl)
 {
-	struct gk104_fifo_engn *engn = &fifo->engine[engine];
 	struct gk104_fifo_chan *chan;
 	struct nvkm_subdev *subdev = &fifo->base.engine.subdev;
 	struct nvkm_device *device = subdev->device;
-	struct nvkm_memory *cur;
+	struct nvkm_memory *mem;
 	int nr = 0;
+	int target;
 
 	mutex_lock(&subdev->mutex);
-	cur = engn->runlist[engn->cur_runlist];
-	engn->cur_runlist = !engn->cur_runlist;
+	mem = fifo->runlist[runl].mem[fifo->runlist[runl].next];
+	fifo->runlist[runl].next = !fifo->runlist[runl].next;
 
-	nvkm_kmap(cur);
-	list_for_each_entry(chan, &engn->chan, head) {
-		nvkm_wo32(cur, (nr * 8) + 0, chan->base.chid);
-		nvkm_wo32(cur, (nr * 8) + 4, 0x00000000);
+	nvkm_kmap(mem);
+	list_for_each_entry(chan, &fifo->runlist[runl].chan, head) {
+		nvkm_wo32(mem, (nr * 8) + 0, chan->base.chid);
+		nvkm_wo32(mem, (nr * 8) + 4, 0x00000000);
 		nr++;
 	}
-	nvkm_done(cur);
-
-	nvkm_wr32(device, 0x002270, nvkm_memory_addr(cur) >> 12);
-	nvkm_wr32(device, 0x002274, (engine << 20) | nr);
-
-	if (wait_event_timeout(engn->wait, !(nvkm_rd32(device, 0x002284 +
-			       (engine * 0x08)) & 0x00100000),
-				msecs_to_jiffies(2000)) == 0)
-		nvkm_error(subdev, "runlist %d update timeout\n", engine);
+	nvkm_done(mem);
+
+	if (nvkm_memory_target(mem) == NVKM_MEM_TARGET_VRAM)
+		target = 0;
+	else
+		target = 3;
+
+	nvkm_wr32(device, 0x002270, (nvkm_memory_addr(mem) >> 12) |
+				    (target << 28));
+	nvkm_wr32(device, 0x002274, (runl << 20) | nr);
+
+	if (wait_event_timeout(fifo->runlist[runl].wait,
+			       !(nvkm_rd32(device, 0x002284 + (runl * 0x08))
+				       & 0x00100000),
+			       msecs_to_jiffies(2000)) == 0)
+		nvkm_error(subdev, "runlist %d update timeout\n", runl);
 	mutex_unlock(&subdev->mutex);
 }
 
@@ -90,58 +97,51 @@ void
 gk104_fifo_runlist_insert(struct gk104_fifo *fifo, struct gk104_fifo_chan *chan)
 {
 	mutex_lock(&fifo->base.engine.subdev.mutex);
-	list_add_tail(&chan->head, &fifo->engine[chan->engine].chan);
+	list_add_tail(&chan->head, &fifo->runlist[chan->runl].chan);
 	mutex_unlock(&fifo->base.engine.subdev.mutex);
 }
 
-static inline struct nvkm_engine *
-gk104_fifo_engine(struct gk104_fifo *fifo, u32 engn)
-{
-	struct nvkm_device *device = fifo->base.engine.subdev.device;
-	u64 subdevs = gk104_fifo_engine_subdev(engn);
-	if (subdevs)
-		return nvkm_device_engine(device, __ffs(subdevs));
-	return NULL;
-}
-
 static void
-gk104_fifo_recover_work(struct work_struct *work)
+gk104_fifo_recover_work(struct work_struct *w)
 {
-	struct gk104_fifo *fifo = container_of(work, typeof(*fifo), fault);
+	struct gk104_fifo *fifo = container_of(w, typeof(*fifo), recover.work);
 	struct nvkm_device *device = fifo->base.engine.subdev.device;
 	struct nvkm_engine *engine;
 	unsigned long flags;
-	u32 engn, engm = 0;
-	u64 mask, todo;
+	u32 engm, runm, todo;
+	int engn, runl;
 
 	spin_lock_irqsave(&fifo->base.lock, flags);
-	mask = fifo->mask;
-	fifo->mask = 0ULL;
+	runm = fifo->recover.runm;
+	engm = fifo->recover.engm;
+	fifo->recover.engm = 0;
+	fifo->recover.runm = 0;
 	spin_unlock_irqrestore(&fifo->base.lock, flags);
 
-	for (todo = mask; engn = __ffs64(todo), todo; todo &= ~(1 << engn))
-		engm |= 1 << gk104_fifo_subdev_engine(engn);
-	nvkm_mask(device, 0x002630, engm, engm);
+	nvkm_mask(device, 0x002630, runm, runm);
 
-	for (todo = mask; engn = __ffs64(todo), todo; todo &= ~(1 << engn)) {
-		if ((engine = nvkm_device_engine(device, engn))) {
+	for (todo = engm; engn = __ffs(todo), todo; todo &= ~BIT(engn)) {
+		if ((engine = fifo->engine[engn].engine)) {
 			nvkm_subdev_fini(&engine->subdev, false);
 			WARN_ON(nvkm_subdev_init(&engine->subdev));
 		}
-		gk104_fifo_runlist_commit(fifo, gk104_fifo_subdev_engine(engn));
 	}
 
-	nvkm_wr32(device, 0x00262c, engm);
-	nvkm_mask(device, 0x002630, engm, 0x00000000);
+	for (todo = runm; runl = __ffs(todo), todo; todo &= ~BIT(runl))
+		gk104_fifo_runlist_commit(fifo, runl);
+
+	nvkm_wr32(device, 0x00262c, runm);
+	nvkm_mask(device, 0x002630, runm, 0x00000000);
 }
 
 static void
 gk104_fifo_recover(struct gk104_fifo *fifo, struct nvkm_engine *engine,
-		  struct gk104_fifo_chan *chan)
+		   struct gk104_fifo_chan *chan)
 {
 	struct nvkm_subdev *subdev = &fifo->base.engine.subdev;
 	struct nvkm_device *device = subdev->device;
 	u32 chid = chan->base.chid;
+	int engn;
 
 	nvkm_error(subdev, "%s engine fault on channel %d, recovering...\n",
 		   nvkm_subdev_name[engine->subdev.index], chid);
@@ -151,8 +151,15 @@ gk104_fifo_recover(struct gk104_fifo *fifo, struct nvkm_engine *engine,
 	list_del_init(&chan->head);
 	chan->killed = true;
 
-	fifo->mask |= 1ULL << engine->subdev.index;
-	schedule_work(&fifo->fault);
+	for (engn = 0; engn < fifo->engine_nr; engn++) {
+		if (fifo->engine[engn].engine == engine) {
+			fifo->recover.engm |= BIT(engn);
+			break;
+		}
+	}
+
+	fifo->recover.runm |= BIT(chan->runl);
+	schedule_work(&fifo->recover.work);
 }
 
 static const struct nvkm_enum
@@ -189,32 +196,31 @@ static void
 gk104_fifo_intr_sched_ctxsw(struct gk104_fifo *fifo)
 {
 	struct nvkm_device *device = fifo->base.engine.subdev.device;
-	struct nvkm_engine *engine;
 	struct gk104_fifo_chan *chan;
 	unsigned long flags;
 	u32 engn;
 
 	spin_lock_irqsave(&fifo->base.lock, flags);
-	for (engn = 0; engn < ARRAY_SIZE(fifo->engine); engn++) {
+	for (engn = 0; engn < fifo->engine_nr; engn++) {
+		struct nvkm_engine *engine = fifo->engine[engn].engine;
+		int runl = fifo->engine[engn].runl;
 		u32 stat = nvkm_rd32(device, 0x002640 + (engn * 0x08));
 		u32 busy = (stat & 0x80000000);
-		u32 next = (stat & 0x07ff0000) >> 16;
+		u32 next = (stat & 0x0fff0000) >> 16;
 		u32 chsw = (stat & 0x00008000);
 		u32 save = (stat & 0x00004000);
 		u32 load = (stat & 0x00002000);
-		u32 prev = (stat & 0x000007ff);
+		u32 prev = (stat & 0x00000fff);
 		u32 chid = load ? next : prev;
 		(void)save;
 
-		if (busy && chsw) {
-			list_for_each_entry(chan, &fifo->engine[engn].chan, head) {
-				if (chan->base.chid == chid) {
-					engine = gk104_fifo_engine(fifo, engn);
-					if (!engine)
-						break;
-					gk104_fifo_recover(fifo, engine, chan);
-					break;
-				}
+		if (!busy || !chsw)
+			continue;
+
+		list_for_each_entry(chan, &fifo->runlist[runl].chan, head) {
+			if (chan->base.chid == chid && engine) {
+				gk104_fifo_recover(fifo, engine, chan);
+				break;
 			}
 		}
 	}
@@ -395,7 +401,7 @@ gk104_fifo_intr_fault(struct gk104_fifo *fifo, int unit)
 		snprintf(gpcid, sizeof(gpcid), "GPC%d/", gpc);
 	}
 
-	if (eu) {
+	if (eu && eu->data2) {
 		switch (eu->data2) {
 		case NVKM_SUBDEV_BAR:
 			nvkm_mask(device, 0x001704, 0x00000000, 0x00000000);
@@ -484,9 +490,10 @@ gk104_fifo_intr_pbdma_0(struct gk104_fifo *fifo, int unit)
 			if (nvkm_sw_mthd(device->sw, chid, subc, mthd, data))
 				show &= ~0x00800000;
 		}
-		nvkm_wr32(device, 0x0400c0 + (unit * 0x2000), 0x80600008);
 	}
 
+	nvkm_wr32(device, 0x0400c0 + (unit * 0x2000), 0x80600008);
+
 	if (show) {
 		nvkm_snprintbf(msg, sizeof(msg), gk104_fifo_pbdma_intr_0, show);
 		chan = nvkm_fifo_chan_chid(&fifo->base, chid, &flags);
@@ -537,10 +544,10 @@ gk104_fifo_intr_runlist(struct gk104_fifo *fifo)
 	struct nvkm_device *device = fifo->base.engine.subdev.device;
 	u32 mask = nvkm_rd32(device, 0x002a00);
 	while (mask) {
-		u32 engn = __ffs(mask);
-		wake_up(&fifo->engine[engn].wait);
-		nvkm_wr32(device, 0x002a00, 1 << engn);
-		mask &= ~(1 << engn);
+		int runl = __ffs(mask);
+		wake_up(&fifo->runlist[runl].wait);
+		nvkm_wr32(device, 0x002a00, 1 << runl);
+		mask &= ~(1 << runl);
 	}
 }
 
@@ -647,7 +654,7 @@ gk104_fifo_fini(struct nvkm_fifo *base)
 {
 	struct gk104_fifo *fifo = gk104_fifo(base);
 	struct nvkm_device *device = fifo->base.engine.subdev.device;
-	flush_work(&fifo->fault);
+	flush_work(&fifo->recover.work);
 	/* allow mmu fault interrupts, even when we're not using fifo */
 	nvkm_mask(device, 0x002140, 0x10000000, 0x10000000);
 }
@@ -656,24 +663,122 @@ int
 gk104_fifo_oneinit(struct nvkm_fifo *base)
 {
 	struct gk104_fifo *fifo = gk104_fifo(base);
-	struct nvkm_device *device = fifo->base.engine.subdev.device;
+	struct nvkm_subdev *subdev = &fifo->base.engine.subdev;
+	struct nvkm_device *device = subdev->device;
 	int ret, i;
+	u32 *map;
+
+	/* Determine number of PBDMAs by checking valid enable bits. */
+	nvkm_wr32(device, 0x000204, 0xffffffff);
+	fifo->pbdma_nr = hweight32(nvkm_rd32(device, 0x000204));
+	nvkm_debug(subdev, "%d PBDMA(s)\n", fifo->pbdma_nr);
+
+	/* Read PBDMA->runlist(s) mapping from HW. */
+	if (!(map = kzalloc(sizeof(*map) * fifo->pbdma_nr, GFP_KERNEL)))
+		return -ENOMEM;
+
+	for (i = 0; i < fifo->pbdma_nr; i++)
+		map[i] = nvkm_rd32(device, 0x002390 + (i * 0x04));
+
+	/* Read device topology from HW. */
+	for (i = 0; i < 64; i++) {
+		int type = -1, pbid = -1, engidx = -1;
+		int engn = -1, runl = -1, intr = -1, mcen = -1;
+		int fault = -1, j;
+		u32 data, addr = 0;
+
+		do {
+			data = nvkm_rd32(device, 0x022700 + (i * 0x04));
+			nvkm_trace(subdev, "%02x: %08x\n", i, data);
+			switch (data & 0x00000003) {
+			case 0x00000000: /* NOT_VALID */
+				continue;
+			case 0x00000001: /* DATA */
+				addr  = (data & 0x00fff000);
+				fault = (data & 0x000000f8) >> 3;
+				break;
+			case 0x00000002: /* ENUM */
+				if (data & 0x00000020)
+					engn = (data & 0x3c000000) >> 26;
+				if (data & 0x00000010)
+					runl = (data & 0x01e00000) >> 21;
+				if (data & 0x00000008)
+					intr = (data & 0x000f8000) >> 15;
+				if (data & 0x00000004)
+					mcen = (data & 0x00003e00) >> 9;
+				break;
+			case 0x00000003: /* ENGINE_TYPE */
+				type = (data & 0x7ffffffc) >> 2;
+				break;
+			}
+		} while ((data & 0x80000000) && ++i < 64);
+
+		if (!data)
+			continue;
+
+		/* Determine which PBDMA handles requests for this engine. */
+		for (j = 0; runl >= 0 && j < fifo->pbdma_nr; j++) {
+			if (map[j] & (1 << runl)) {
+				pbid = j;
+				break;
+			}
+		}
+
+		/* Translate engine type to NVKM engine identifier. */
+		switch (type) {
+		case 0x00000000: engidx = NVKM_ENGINE_GR; break;
+		case 0x00000001: engidx = NVKM_ENGINE_CE0; break;
+		case 0x00000002: engidx = NVKM_ENGINE_CE1; break;
+		case 0x00000003: engidx = NVKM_ENGINE_CE2; break;
+		case 0x00000008: engidx = NVKM_ENGINE_MSPDEC; break;
+		case 0x00000009: engidx = NVKM_ENGINE_MSPPP; break;
+		case 0x0000000a: engidx = NVKM_ENGINE_MSVLD; break;
+		case 0x0000000b: engidx = NVKM_ENGINE_MSENC; break;
+		case 0x0000000c: engidx = NVKM_ENGINE_VIC; break;
+		case 0x0000000d: engidx = NVKM_ENGINE_SEC; break;
+		case 0x0000000e: engidx = NVKM_ENGINE_NVENC0; break;
+		case 0x0000000f: engidx = NVKM_ENGINE_NVENC1; break;
+		case 0x00000010: engidx = NVKM_ENGINE_NVDEC; break;
+			break;
+		default:
+			break;
+		}
+
+		nvkm_debug(subdev, "%02x (%8s): engine %2d runlist %2d "
+				   "pbdma %2d intr %2d reset %2d "
+				   "fault %2d addr %06x\n", type,
+			   engidx < 0 ? NULL : nvkm_subdev_name[engidx],
+			   engn, runl, pbid, intr, mcen, fault, addr);
+
+		/* Mark the engine as supported if everything checks out. */
+		if (engn >= 0 && runl >= 0) {
+			fifo->engine[engn].engine = engidx < 0 ? NULL :
+				nvkm_device_engine(device, engidx);
+			fifo->engine[engn].runl = runl;
+			fifo->engine[engn].pbid = pbid;
+			fifo->engine_nr = max(fifo->engine_nr, engn + 1);
+			fifo->runlist[runl].engm |= 1 << engn;
+			fifo->runlist_nr = max(fifo->runlist_nr, runl + 1);
+		}
+	}
 
-	for (i = 0; i < ARRAY_SIZE(fifo->engine); i++) {
+	kfree(map);
+
+	for (i = 0; i < fifo->runlist_nr; i++) {
 		ret = nvkm_memory_new(device, NVKM_MEM_TARGET_INST,
 				      0x8000, 0x1000, false,
-				      &fifo->engine[i].runlist[0]);
+				      &fifo->runlist[i].mem[0]);
 		if (ret)
 			return ret;
 
 		ret = nvkm_memory_new(device, NVKM_MEM_TARGET_INST,
 				      0x8000, 0x1000, false,
-				      &fifo->engine[i].runlist[1]);
+				      &fifo->runlist[i].mem[1]);
 		if (ret)
 			return ret;
 
-		init_waitqueue_head(&fifo->engine[i].wait);
-		INIT_LIST_HEAD(&fifo->engine[i].chan);
+		init_waitqueue_head(&fifo->runlist[i].wait);
+		INIT_LIST_HEAD(&fifo->runlist[i].chan);
 	}
 
 	ret = nvkm_memory_new(device, NVKM_MEM_TARGET_INST,
@@ -695,24 +800,21 @@ void
 gk104_fifo_init(struct nvkm_fifo *base)
 {
 	struct gk104_fifo *fifo = gk104_fifo(base);
-	struct nvkm_subdev *subdev = &fifo->base.engine.subdev;
-	struct nvkm_device *device = subdev->device;
+	struct nvkm_device *device = fifo->base.engine.subdev.device;
 	int i;
 
-	/* enable all available PBDMA units */
-	nvkm_wr32(device, 0x000204, 0xffffffff);
-	fifo->spoon_nr = hweight32(nvkm_rd32(device, 0x000204));
-	nvkm_debug(subdev, "%d PBDMA unit(s)\n", fifo->spoon_nr);
+	/* Enable PBDMAs. */
+	nvkm_wr32(device, 0x000204, (1 << fifo->pbdma_nr) - 1);
 
 	/* PBDMA[n] */
-	for (i = 0; i < fifo->spoon_nr; i++) {
+	for (i = 0; i < fifo->pbdma_nr; i++) {
 		nvkm_mask(device, 0x04013c + (i * 0x2000), 0x10000100, 0x00000000);
 		nvkm_wr32(device, 0x040108 + (i * 0x2000), 0xffffffff); /* INTR */
 		nvkm_wr32(device, 0x04010c + (i * 0x2000), 0xfffffeff); /* INTREN */
 	}
 
 	/* PBDMA[n].HCE */
-	for (i = 0; i < fifo->spoon_nr; i++) {
+	for (i = 0; i < fifo->pbdma_nr; i++) {
 		nvkm_wr32(device, 0x040148 + (i * 0x2000), 0xffffffff); /* INTR */
 		nvkm_wr32(device, 0x04014c + (i * 0x2000), 0xffffffff); /* INTREN */
 	}
@@ -732,9 +834,9 @@ gk104_fifo_dtor(struct nvkm_fifo *base)
 	nvkm_vm_put(&fifo->user.bar);
 	nvkm_memory_del(&fifo->user.mem);
 
-	for (i = 0; i < ARRAY_SIZE(fifo->engine); i++) {
-		nvkm_memory_del(&fifo->engine[i].runlist[1]);
-		nvkm_memory_del(&fifo->engine[i].runlist[0]);
+	for (i = 0; i < fifo->runlist_nr; i++) {
+		nvkm_memory_del(&fifo->runlist[i].mem[1]);
+		nvkm_memory_del(&fifo->runlist[i].mem[0]);
 	}
 
 	return fifo;
@@ -748,7 +850,7 @@ gk104_fifo_new_(const struct nvkm_fifo_func *func, struct nvkm_device *device,
 
 	if (!(fifo = kzalloc(sizeof(*fifo), GFP_KERNEL)))
 		return -ENOMEM;
-	INIT_WORK(&fifo->fault, gk104_fifo_recover_work);
+	INIT_WORK(&fifo->recover.work, gk104_fifo_recover_work);
 	*pfifo = &fifo->base;
 
 	return nvkm_fifo_ctor(func, device, index, nr, &fifo->base);
diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/fifo/gk104.h b/drivers/gpu/drm/nouveau/nvkm/engine/fifo/gk104.h
index bec519d8f91e..9e5d00ba34a2 100644
--- a/drivers/gpu/drm/nouveau/nvkm/engine/fifo/gk104.h
+++ b/drivers/gpu/drm/nouveau/nvkm/engine/fifo/gk104.h
@@ -6,25 +6,37 @@
 #include <subdev/mmu.h>
 
 struct gk104_fifo_chan;
-struct gk104_fifo_engn {
-	struct nvkm_memory *runlist[2];
-	int cur_runlist;
-	wait_queue_head_t wait;
-	struct list_head chan;
-};
-
 struct gk104_fifo {
 	struct nvkm_fifo base;
 
-	struct work_struct fault;
-	u64 mask;
+	struct {
+		struct work_struct work;
+		u32 engm;
+		u32 runm;
+	} recover;
+
+	int pbdma_nr;
+
+	struct {
+		struct nvkm_engine *engine;
+		int runl;
+		int pbid;
+	} engine[16];
+	int engine_nr;
+
+	struct {
+		struct nvkm_memory *mem[2];
+		int next;
+		wait_queue_head_t wait;
+		struct list_head chan;
+		u32 engm;
+	} runlist[16];
+	int runlist_nr;
 
-	struct gk104_fifo_engn engine[7];
 	struct {
 		struct nvkm_memory *mem;
 		struct nvkm_vma bar;
 	} user;
-	int spoon_nr;
 };
 
 int gk104_fifo_new_(const struct nvkm_fifo_func *, struct nvkm_device *,
@@ -38,7 +50,7 @@ void gk104_fifo_uevent_init(struct nvkm_fifo *);
 void gk104_fifo_uevent_fini(struct nvkm_fifo *);
 void gk104_fifo_runlist_insert(struct gk104_fifo *, struct gk104_fifo_chan *);
 void gk104_fifo_runlist_remove(struct gk104_fifo *, struct gk104_fifo_chan *);
-void gk104_fifo_runlist_commit(struct gk104_fifo *, u32 engine);
+void gk104_fifo_runlist_commit(struct gk104_fifo *, int runl);
 
 static inline u64
 gk104_fifo_engine_subdev(int engine)
@@ -58,23 +70,4 @@ gk104_fifo_engine_subdev(int engine)
 		return 0;
 	}
 }
-
-static inline int
-gk104_fifo_subdev_engine(int subdev)
-{
-	switch (subdev) {
-	case NVKM_ENGINE_GR:
-	case NVKM_ENGINE_SW:
-	case NVKM_ENGINE_CE2   : return 0;
-	case NVKM_ENGINE_MSPDEC: return 1;
-	case NVKM_ENGINE_MSPPP : return 2;
-	case NVKM_ENGINE_MSVLD : return 3;
-	case NVKM_ENGINE_CE0   : return 4;
-	case NVKM_ENGINE_CE1   : return 5;
-	case NVKM_ENGINE_MSENC : return 6;
-	default:
-		WARN_ON(1);
-		return 0;
-	}
-}
 #endif
diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/fifo/gk110.c b/drivers/gpu/drm/nouveau/nvkm/engine/fifo/gk110.c
new file mode 100644
index 000000000000..41307fcd4bb3
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/nvkm/engine/fifo/gk110.c
@@ -0,0 +1,46 @@
+/*
+ * Copyright 2016 Red Hat Inc.
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
+ * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR
+ * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
+ * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
+ * OTHER DEALINGS IN THE SOFTWARE.
+ *
+ * Authors: Ben Skeggs
+ */
+#include "gk104.h"
+#include "changk104.h"
+
+static const struct nvkm_fifo_func
+gk110_fifo = {
+	.dtor = gk104_fifo_dtor,
+	.oneinit = gk104_fifo_oneinit,
+	.init = gk104_fifo_init,
+	.fini = gk104_fifo_fini,
+	.intr = gk104_fifo_intr,
+	.uevent_init = gk104_fifo_uevent_init,
+	.uevent_fini = gk104_fifo_uevent_fini,
+	.chan = {
+		&gk110_fifo_gpfifo_oclass,
+		NULL
+	},
+};
+
+int
+gk110_fifo_new(struct nvkm_device *device, int index, struct nvkm_fifo **pfifo)
+{
+	return gk104_fifo_new_(&gk110_fifo, device, index, 4096, pfifo);
+}
diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/fifo/gm107.c b/drivers/gpu/drm/nouveau/nvkm/engine/fifo/gm107.c
new file mode 100644
index 000000000000..6d59d65794a1
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/nvkm/engine/fifo/gm107.c
@@ -0,0 +1,46 @@
+/*
+ * Copyright 2016 Red Hat Inc.
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
+ * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR
+ * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
+ * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
+ * OTHER DEALINGS IN THE SOFTWARE.
+ *
+ * Authors: Ben Skeggs
+ */
+#include "gk104.h"
+#include "changk104.h"
+
+static const struct nvkm_fifo_func
+gm107_fifo = {
+	.dtor = gk104_fifo_dtor,
+	.oneinit = gk104_fifo_oneinit,
+	.init = gk104_fifo_init,
+	.fini = gk104_fifo_fini,
+	.intr = gk104_fifo_intr,
+	.uevent_init = gk104_fifo_uevent_init,
+	.uevent_fini = gk104_fifo_uevent_fini,
+	.chan = {
+		&gk110_fifo_gpfifo_oclass,
+		NULL
+	},
+};
+
+int
+gm107_fifo_new(struct nvkm_device *device, int index, struct nvkm_fifo **pfifo)
+{
+	return gk104_fifo_new_(&gm107_fifo, device, index, 2048, pfifo);
+}
diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/fifo/gm204.c b/drivers/gpu/drm/nouveau/nvkm/engine/fifo/gm200.c
index 2db629f1bf7e..4bdd43078df9 100644
--- a/drivers/gpu/drm/nouveau/nvkm/engine/fifo/gm204.c
+++ b/drivers/gpu/drm/nouveau/nvkm/engine/fifo/gm200.c
@@ -25,7 +25,7 @@
 #include "changk104.h"
 
 static const struct nvkm_fifo_func
-gm204_fifo = {
+gm200_fifo = {
 	.dtor = gk104_fifo_dtor,
 	.oneinit = gk104_fifo_oneinit,
 	.init = gk104_fifo_init,
@@ -34,13 +34,13 @@ gm204_fifo = {
 	.uevent_init = gk104_fifo_uevent_init,
 	.uevent_fini = gk104_fifo_uevent_fini,
 	.chan = {
-		&gm204_fifo_gpfifo_oclass,
+		&gm200_fifo_gpfifo_oclass,
 		NULL
 	},
 };
 
 int
-gm204_fifo_new(struct nvkm_device *device, int index, struct nvkm_fifo **pfifo)
+gm200_fifo_new(struct nvkm_device *device, int index, struct nvkm_fifo **pfifo)
 {
-	return gk104_fifo_new_(&gm204_fifo, device, index, 4096, pfifo);
+	return gk104_fifo_new_(&gm200_fifo, device, index, 4096, pfifo);
 }
diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/fifo/gm20b.c b/drivers/gpu/drm/nouveau/nvkm/engine/fifo/gm20b.c
index ae6375d9760f..4c91d4aa1e9e 100644
--- a/drivers/gpu/drm/nouveau/nvkm/engine/fifo/gm20b.c
+++ b/drivers/gpu/drm/nouveau/nvkm/engine/fifo/gm20b.c
@@ -32,7 +32,7 @@ gm20b_fifo = {
 	.uevent_init = gk104_fifo_uevent_init,
 	.uevent_fini = gk104_fifo_uevent_fini,
 	.chan = {
-		&gm204_fifo_gpfifo_oclass,
+		&gm200_fifo_gpfifo_oclass,
 		NULL
 	},
 };
diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/fifo/gpfifogk104.c b/drivers/gpu/drm/nouveau/nvkm/engine/fifo/gpfifogk104.c
index 2e1df01bd928..ed4351032ed6 100644
--- a/drivers/gpu/drm/nouveau/nvkm/engine/fifo/gpfifogk104.c
+++ b/drivers/gpu/drm/nouveau/nvkm/engine/fifo/gpfifogk104.c
@@ -63,9 +63,15 @@ gk104_fifo_gpfifo_engine_addr(struct nvkm_engine *engine)
 	case NVKM_ENGINE_CE1   :
 	case NVKM_ENGINE_CE2   : return 0x0000;
 	case NVKM_ENGINE_GR    : return 0x0210;
+	case NVKM_ENGINE_SEC   : return 0x0220;
 	case NVKM_ENGINE_MSPDEC: return 0x0250;
 	case NVKM_ENGINE_MSPPP : return 0x0260;
 	case NVKM_ENGINE_MSVLD : return 0x0270;
+	case NVKM_ENGINE_VIC   : return 0x0280;
+	case NVKM_ENGINE_MSENC : return 0x0290;
+	case NVKM_ENGINE_NVDEC : return 0x02100270;
+	case NVKM_ENGINE_NVENC0: return 0x02100290;
+	case NVKM_ENGINE_NVENC1: return 0x0210;
 	default:
 		WARN_ON(1);
 		return 0;
@@ -76,9 +82,9 @@ static int
 gk104_fifo_gpfifo_engine_fini(struct nvkm_fifo_chan *base,
 			      struct nvkm_engine *engine, bool suspend)
 {
-	const u32 offset = gk104_fifo_gpfifo_engine_addr(engine);
 	struct gk104_fifo_chan *chan = gk104_fifo_chan(base);
 	struct nvkm_gpuobj *inst = chan->base.inst;
+	u32 offset = gk104_fifo_gpfifo_engine_addr(engine);
 	int ret;
 
 	ret = gk104_fifo_gpfifo_kick(chan);
@@ -87,8 +93,12 @@ gk104_fifo_gpfifo_engine_fini(struct nvkm_fifo_chan *base,
 
 	if (offset) {
 		nvkm_kmap(inst);
-		nvkm_wo32(inst, offset + 0x00, 0x00000000);
-		nvkm_wo32(inst, offset + 0x04, 0x00000000);
+		nvkm_wo32(inst, (offset & 0xffff) + 0x00, 0x00000000);
+		nvkm_wo32(inst, (offset & 0xffff) + 0x04, 0x00000000);
+		if ((offset >>= 16)) {
+			nvkm_wo32(inst, offset + 0x00, 0x00000000);
+			nvkm_wo32(inst, offset + 0x04, 0x00000000);
+		}
 		nvkm_done(inst);
 	}
 
@@ -99,15 +109,21 @@ static int
 gk104_fifo_gpfifo_engine_init(struct nvkm_fifo_chan *base,
 			      struct nvkm_engine *engine)
 {
-	const u32 offset = gk104_fifo_gpfifo_engine_addr(engine);
 	struct gk104_fifo_chan *chan = gk104_fifo_chan(base);
 	struct nvkm_gpuobj *inst = chan->base.inst;
+	u32 offset = gk104_fifo_gpfifo_engine_addr(engine);
 
 	if (offset) {
-		u64 addr = chan->engn[engine->subdev.index].vma.offset;
+		u64   addr = chan->engn[engine->subdev.index].vma.offset;
+		u32 datalo = lower_32_bits(addr) | 0x00000004;
+		u32 datahi = upper_32_bits(addr);
 		nvkm_kmap(inst);
-		nvkm_wo32(inst, offset + 0x00, lower_32_bits(addr) | 4);
-		nvkm_wo32(inst, offset + 0x04, upper_32_bits(addr));
+		nvkm_wo32(inst, (offset & 0xffff) + 0x00, datalo);
+		nvkm_wo32(inst, (offset & 0xffff) + 0x04, datahi);
+		if ((offset >>= 16)) {
+			nvkm_wo32(inst, offset + 0x00, datalo);
+			nvkm_wo32(inst, offset + 0x04, datahi);
+		}
 		nvkm_done(inst);
 	}
 
@@ -154,7 +170,8 @@ gk104_fifo_gpfifo_fini(struct nvkm_fifo_chan *base)
 	if (!list_empty(&chan->head)) {
 		gk104_fifo_runlist_remove(fifo, chan);
 		nvkm_mask(device, 0x800004 + coff, 0x00000800, 0x00000800);
-		gk104_fifo_runlist_commit(fifo, chan->engine);
+		gk104_fifo_gpfifo_kick(chan);
+		gk104_fifo_runlist_commit(fifo, chan->runl);
 	}
 
 	nvkm_wr32(device, 0x800000 + coff, 0x00000000);
@@ -169,13 +186,13 @@ gk104_fifo_gpfifo_init(struct nvkm_fifo_chan *base)
 	u32 addr = chan->base.inst->addr >> 12;
 	u32 coff = chan->base.chid * 8;
 
-	nvkm_mask(device, 0x800004 + coff, 0x000f0000, chan->engine << 16);
+	nvkm_mask(device, 0x800004 + coff, 0x000f0000, chan->runl << 16);
 	nvkm_wr32(device, 0x800000 + coff, 0x80000000 | addr);
 
 	if (list_empty(&chan->head) && !chan->killed) {
 		gk104_fifo_runlist_insert(fifo, chan);
 		nvkm_mask(device, 0x800004 + coff, 0x00000400, 0x00000400);
-		gk104_fifo_runlist_commit(fifo, chan->engine);
+		gk104_fifo_runlist_commit(fifo, chan->runl);
 		nvkm_mask(device, 0x800004 + coff, 0x00000400, 0x00000400);
 	}
 }
@@ -201,73 +218,79 @@ gk104_fifo_gpfifo_func = {
 	.engine_fini = gk104_fifo_gpfifo_engine_fini,
 };
 
-int
-gk104_fifo_gpfifo_new(struct nvkm_fifo *base, const struct nvkm_oclass *oclass,
-		      void *data, u32 size, struct nvkm_object **pobject)
+struct gk104_fifo_chan_func {
+	u32 engine;
+	u64 subdev;
+};
+
+static int
+gk104_fifo_gpfifo_new_(const struct gk104_fifo_chan_func *func,
+		       struct gk104_fifo *fifo, u32 *engmask, u16 *chid,
+		       u64 vm, u64 ioffset, u64 ilength,
+		       const struct nvkm_oclass *oclass,
+		       struct nvkm_object **pobject)
 {
-	union {
-		struct kepler_channel_gpfifo_a_v0 v0;
-	} *args = data;
-	struct gk104_fifo *fifo = gk104_fifo(base);
 	struct nvkm_device *device = fifo->base.engine.subdev.device;
-	struct nvkm_object *parent = oclass->parent;
 	struct gk104_fifo_chan *chan;
-	u64 usermem, ioffset, ilength;
-	u32 engines;
-	int ret = -ENOSYS, i;
-
-	nvif_ioctl(parent, "create channel gpfifo size %d\n", size);
-	if (!(ret = nvif_unpack(ret, &data, &size, args->v0, 0, 0, false))) {
-		nvif_ioctl(parent, "create channel gpfifo vers %d vm %llx "
-				   "ioffset %016llx ilength %08x engine %08x\n",
-			   args->v0.version, args->v0.vm, args->v0.ioffset,
-			   args->v0.ilength, args->v0.engine);
-	} else
-		return ret;
-
-	/* determine which downstream engines are present */
-	for (i = 0, engines = 0; i < ARRAY_SIZE(fifo->engine); i++) {
-		u64 subdevs = gk104_fifo_engine_subdev(i);
-		if (!nvkm_device_engine(device, __ffs64(subdevs)))
-			continue;
-		engines |= (1 << i);
+	int runlist = -1, ret = -ENOSYS, i, j;
+	u32 engines = 0, present = 0;
+	u64 subdevs = 0;
+	u64 usermem;
+
+	/* Determine which downstream engines are present */
+	for (i = 0; i < fifo->engine_nr; i++) {
+		struct nvkm_engine *engine = fifo->engine[i].engine;
+		if (engine) {
+			u64 submask = BIT_ULL(engine->subdev.index);
+			for (j = 0; func[j].subdev; j++) {
+				if (func[j].subdev & submask) {
+					present |= func[j].engine;
+					break;
+				}
+			}
+
+			if (!func[j].subdev)
+				continue;
+
+			if (runlist < 0 && (*engmask & present))
+				runlist = fifo->engine[i].runl;
+			if (runlist == fifo->engine[i].runl) {
+				engines |= func[j].engine;
+				subdevs |= func[j].subdev;
+			}
+		}
 	}
 
-	/* if this is an engine mask query, we're done */
-	if (!args->v0.engine) {
-		args->v0.engine = engines;
+	/* Just an engine mask query?  All done here! */
+	if (!*engmask) {
+		*engmask = present;
 		return nvkm_object_new(oclass, NULL, 0, pobject);
 	}
 
-	/* check that we support a requested engine - note that the user
-	 * argument is a mask in order to allow the user to request (for
-	 * example) *any* copy engine, but doesn't matter which.
-	 */
-	args->v0.engine &= engines;
-	if (!args->v0.engine) {
-		nvif_ioctl(parent, "no supported engine\n");
+	/* No runlist?  No supported engines. */
+	*engmask = present;
+	if (runlist < 0)
 		return -ENODEV;
-	}
+	*engmask = engines;
 
-	/* allocate the channel */
+	/* Allocate the channel. */
 	if (!(chan = kzalloc(sizeof(*chan), GFP_KERNEL)))
 		return -ENOMEM;
 	*pobject = &chan->base.object;
 	chan->fifo = fifo;
-	chan->engine = __ffs(args->v0.engine);
+	chan->runl = runlist;
 	INIT_LIST_HEAD(&chan->head);
 
 	ret = nvkm_fifo_chan_ctor(&gk104_fifo_gpfifo_func, &fifo->base,
-				  0x1000, 0x1000, true, args->v0.vm, 0,
-				  gk104_fifo_engine_subdev(chan->engine),
+				  0x1000, 0x1000, true, vm, 0, subdevs,
 				  1, fifo->user.bar.offset, 0x200,
 				  oclass, &chan->base);
 	if (ret)
 		return ret;
 
-	args->v0.chid = chan->base.chid;
+	*chid = chan->base.chid;
 
-	/* page directory */
+	/* Page directory. */
 	ret = nvkm_gpuobj_new(device, 0x10000, 0x1000, false, NULL, &chan->pgd);
 	if (ret)
 		return ret;
@@ -283,10 +306,9 @@ gk104_fifo_gpfifo_new(struct nvkm_fifo *base, const struct nvkm_oclass *oclass,
 	if (ret)
 		return ret;
 
-	/* clear channel control registers */
+	/* Clear channel control registers. */
 	usermem = chan->base.chid * 0x200;
-	ioffset = args->v0.ioffset;
-	ilength = order_base_2(args->v0.ilength / 8);
+	ilength = order_base_2(ilength / 8);
 
 	nvkm_kmap(fifo->user.mem);
 	for (i = 0; i < 0x200; i += 4)
@@ -315,6 +337,56 @@ gk104_fifo_gpfifo_new(struct nvkm_fifo *base, const struct nvkm_oclass *oclass,
 	return 0;
 }
 
+static const struct gk104_fifo_chan_func
+gk104_fifo_gpfifo[] = {
+	{ NVA06F_V0_ENGINE_SW | NVA06F_V0_ENGINE_GR,
+		BIT_ULL(NVKM_ENGINE_SW) | BIT_ULL(NVKM_ENGINE_GR)
+	},
+	{ NVA06F_V0_ENGINE_SEC   , BIT_ULL(NVKM_ENGINE_SEC   ) },
+	{ NVA06F_V0_ENGINE_MSVLD , BIT_ULL(NVKM_ENGINE_MSVLD ) },
+	{ NVA06F_V0_ENGINE_MSPDEC, BIT_ULL(NVKM_ENGINE_MSPDEC) },
+	{ NVA06F_V0_ENGINE_MSPPP , BIT_ULL(NVKM_ENGINE_MSPPP ) },
+	{ NVA06F_V0_ENGINE_MSENC , BIT_ULL(NVKM_ENGINE_MSENC ) },
+	{ NVA06F_V0_ENGINE_VIC   , BIT_ULL(NVKM_ENGINE_VIC   ) },
+	{ NVA06F_V0_ENGINE_NVDEC , BIT_ULL(NVKM_ENGINE_NVDEC ) },
+	{ NVA06F_V0_ENGINE_NVENC0, BIT_ULL(NVKM_ENGINE_NVENC0) },
+	{ NVA06F_V0_ENGINE_NVENC1, BIT_ULL(NVKM_ENGINE_NVENC1) },
+	{ NVA06F_V0_ENGINE_CE0   , BIT_ULL(NVKM_ENGINE_CE0   ) },
+	{ NVA06F_V0_ENGINE_CE1   , BIT_ULL(NVKM_ENGINE_CE1   ) },
+	{ NVA06F_V0_ENGINE_CE2   , BIT_ULL(NVKM_ENGINE_CE2   ) },
+	{}
+};
+
+int
+gk104_fifo_gpfifo_new(struct nvkm_fifo *base, const struct nvkm_oclass *oclass,
+		      void *data, u32 size, struct nvkm_object **pobject)
+{
+	struct nvkm_object *parent = oclass->parent;
+	union {
+		struct kepler_channel_gpfifo_a_v0 v0;
+	} *args = data;
+	struct gk104_fifo *fifo = gk104_fifo(base);
+	int ret = -ENOSYS;
+
+	nvif_ioctl(parent, "create channel gpfifo size %d\n", size);
+	if (!(ret = nvif_unpack(ret, &data, &size, args->v0, 0, 0, false))) {
+		nvif_ioctl(parent, "create channel gpfifo vers %d vm %llx "
+				   "ioffset %016llx ilength %08x engine %08x\n",
+			   args->v0.version, args->v0.vm, args->v0.ioffset,
+			   args->v0.ilength, args->v0.engines);
+		return gk104_fifo_gpfifo_new_(gk104_fifo_gpfifo, fifo,
+					      &args->v0.engines,
+					      &args->v0.chid,
+					       args->v0.vm,
+					       args->v0.ioffset,
+					       args->v0.ilength,
+					      oclass, pobject);
+
+	}
+
+	return ret;
+}
+
 const struct nvkm_fifo_chan_oclass
 gk104_fifo_gpfifo_oclass = {
 	.base.oclass = KEPLER_CHANNEL_GPFIFO_A,
diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/fifo/gpfifogk110.c b/drivers/gpu/drm/nouveau/nvkm/engine/fifo/gpfifogk110.c
new file mode 100644
index 000000000000..a9aa69c82e8e
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/nvkm/engine/fifo/gpfifogk110.c
@@ -0,0 +1,34 @@
+/*
+ * Copyright 2016 Red Hat Inc.
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
+ * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR
+ * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
+ * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
+ * OTHER DEALINGS IN THE SOFTWARE.
+ *
+ * Authors: Ben Skeggs
+ */
+#include "changk104.h"
+
+#include <nvif/class.h>
+
+const struct nvkm_fifo_chan_oclass
+gk110_fifo_gpfifo_oclass = {
+	.base.oclass = KEPLER_CHANNEL_GPFIFO_B,
+	.base.minver = 0,
+	.base.maxver = 0,
+	.ctor = gk104_fifo_gpfifo_new,
+};
diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/fifo/gpfifogm204.c b/drivers/gpu/drm/nouveau/nvkm/engine/fifo/gpfifogm200.c
index 6511d6e21ecc..a13315147391 100644
--- a/drivers/gpu/drm/nouveau/nvkm/engine/fifo/gpfifogm204.c
+++ b/drivers/gpu/drm/nouveau/nvkm/engine/fifo/gpfifogm200.c
@@ -26,7 +26,7 @@
 #include <nvif/class.h>
 
 const struct nvkm_fifo_chan_oclass
-gm204_fifo_gpfifo_oclass = {
+gm200_fifo_gpfifo_oclass = {
 	.base.oclass = MAXWELL_CHANNEL_GPFIFO_A,
 	.base.minver = 0,
 	.base.maxver = 0,
diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/gr/Kbuild b/drivers/gpu/drm/nouveau/nvkm/engine/gr/Kbuild
index 9ad0d0e78a96..290ed0db8047 100644
--- a/drivers/gpu/drm/nouveau/nvkm/engine/gr/Kbuild
+++ b/drivers/gpu/drm/nouveau/nvkm/engine/gr/Kbuild
@@ -29,8 +29,7 @@ nvkm-y += nvkm/engine/gr/gk110b.o
 nvkm-y += nvkm/engine/gr/gk208.o
 nvkm-y += nvkm/engine/gr/gk20a.o
 nvkm-y += nvkm/engine/gr/gm107.o
-nvkm-y += nvkm/engine/gr/gm204.o
-nvkm-y += nvkm/engine/gr/gm206.o
+nvkm-y += nvkm/engine/gr/gm200.o
 nvkm-y += nvkm/engine/gr/gm20b.o
 
 nvkm-y += nvkm/engine/gr/ctxnv40.o
@@ -47,6 +46,5 @@ nvkm-y += nvkm/engine/gr/ctxgk110b.o
 nvkm-y += nvkm/engine/gr/ctxgk208.o
 nvkm-y += nvkm/engine/gr/ctxgk20a.o
 nvkm-y += nvkm/engine/gr/ctxgm107.o
-nvkm-y += nvkm/engine/gr/ctxgm204.o
-nvkm-y += nvkm/engine/gr/ctxgm206.o
+nvkm-y += nvkm/engine/gr/ctxgm200.o
 nvkm-y += nvkm/engine/gr/ctxgm20b.o
diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/gr/ctxgf100.h b/drivers/gpu/drm/nouveau/nvkm/engine/gr/ctxgf100.h
index 3c64040ec5a2..3c8673958f22 100644
--- a/drivers/gpu/drm/nouveau/nvkm/engine/gr/ctxgf100.h
+++ b/drivers/gpu/drm/nouveau/nvkm/engine/gr/ctxgf100.h
@@ -97,12 +97,11 @@ void gm107_grctx_generate_bundle(struct gf100_grctx *);
 void gm107_grctx_generate_pagepool(struct gf100_grctx *);
 void gm107_grctx_generate_attrib(struct gf100_grctx *);
 
-extern const struct gf100_grctx_func gm204_grctx;
-void gm204_grctx_generate_main(struct gf100_gr *, struct gf100_grctx *);
-void gm204_grctx_generate_tpcid(struct gf100_gr *);
-void gm204_grctx_generate_405b60(struct gf100_gr *);
+extern const struct gf100_grctx_func gm200_grctx;
+void gm200_grctx_generate_main(struct gf100_gr *, struct gf100_grctx *);
+void gm200_grctx_generate_tpcid(struct gf100_gr *);
+void gm200_grctx_generate_405b60(struct gf100_gr *);
 
-extern const struct gf100_grctx_func gm206_grctx;
 extern const struct gf100_grctx_func gm20b_grctx;
 
 /* context init value lists */
@@ -210,19 +209,4 @@ extern const struct gf100_gr_init gk208_grctx_init_crstr_0[];
 
 extern const struct gf100_gr_init gm107_grctx_init_gpc_unk_0[];
 extern const struct gf100_gr_init gm107_grctx_init_wwdx_0[];
-
-extern const struct gf100_gr_pack gm204_grctx_pack_icmd[];
-
-extern const struct gf100_gr_pack gm204_grctx_pack_mthd[];
-
-extern const struct gf100_gr_pack gm204_grctx_pack_hub[];
-
-extern const struct gf100_gr_init gm204_grctx_init_prop_0[];
-extern const struct gf100_gr_init gm204_grctx_init_setup_0[];
-extern const struct gf100_gr_init gm204_grctx_init_gpm_0[];
-extern const struct gf100_gr_init gm204_grctx_init_gpc_unk_2[];
-
-extern const struct gf100_gr_pack gm204_grctx_pack_tpc[];
-
-extern const struct gf100_gr_pack gm204_grctx_pack_ppc[];
 #endif
diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/gr/ctxgm200.c b/drivers/gpu/drm/nouveau/nvkm/engine/gr/ctxgm200.c
new file mode 100644
index 000000000000..e586699fc43f
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/nvkm/engine/gr/ctxgm200.c
@@ -0,0 +1,147 @@
+/*
+ * Copyright 2015 Red Hat Inc.
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
+ * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR
+ * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
+ * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
+ * OTHER DEALINGS IN THE SOFTWARE.
+ *
+ * Authors: Ben Skeggs <bskeggs@redhat.com>
+ */
+#include "ctxgf100.h"
+
+/*******************************************************************************
+ * PGRAPH context implementation
+ ******************************************************************************/
+
+void
+gm200_grctx_generate_tpcid(struct gf100_gr *gr)
+{
+	struct nvkm_device *device = gr->base.engine.subdev.device;
+	int gpc, tpc, id;
+
+	for (tpc = 0, id = 0; tpc < 4; tpc++) {
+		for (gpc = 0; gpc < gr->gpc_nr; gpc++) {
+			if (tpc < gr->tpc_nr[gpc]) {
+				nvkm_wr32(device, TPC_UNIT(gpc, tpc, 0x698), id);
+				nvkm_wr32(device, GPC_UNIT(gpc, 0x0c10 + tpc * 4), id);
+				nvkm_wr32(device, TPC_UNIT(gpc, tpc, 0x088), id);
+				id++;
+			}
+		}
+	}
+}
+
+static void
+gm200_grctx_generate_rop_active_fbps(struct gf100_gr *gr)
+{
+	struct nvkm_device *device = gr->base.engine.subdev.device;
+	const u32 fbp_count = nvkm_rd32(device, 0x12006c);
+	nvkm_mask(device, 0x408850, 0x0000000f, fbp_count); /* zrop */
+	nvkm_mask(device, 0x408958, 0x0000000f, fbp_count); /* crop */
+}
+
+void
+gm200_grctx_generate_405b60(struct gf100_gr *gr)
+{
+	struct nvkm_device *device = gr->base.engine.subdev.device;
+	const u32 dist_nr = DIV_ROUND_UP(gr->tpc_total, 4);
+	u32 dist[TPC_MAX / 4] = {};
+	u32 gpcs[GPC_MAX] = {};
+	u8  tpcnr[GPC_MAX];
+	int tpc, gpc, i;
+
+	memcpy(tpcnr, gr->tpc_nr, sizeof(gr->tpc_nr));
+
+	/* won't result in the same distribution as the binary driver where
+	 * some of the gpcs have more tpcs than others, but this shall do
+	 * for the moment.  the code for earlier gpus has this issue too.
+	 */
+	for (gpc = -1, i = 0; i < gr->tpc_total; i++) {
+		do {
+			gpc = (gpc + 1) % gr->gpc_nr;
+		} while(!tpcnr[gpc]);
+		tpc = gr->tpc_nr[gpc] - tpcnr[gpc]--;
+
+		dist[i / 4] |= ((gpc << 4) | tpc) << ((i % 4) * 8);
+		gpcs[gpc] |= i << (tpc * 8);
+	}
+
+	for (i = 0; i < dist_nr; i++)
+		nvkm_wr32(device, 0x405b60 + (i * 4), dist[i]);
+	for (i = 0; i < gr->gpc_nr; i++)
+		nvkm_wr32(device, 0x405ba0 + (i * 4), gpcs[i]);
+}
+
+void
+gm200_grctx_generate_main(struct gf100_gr *gr, struct gf100_grctx *info)
+{
+	struct nvkm_device *device = gr->base.engine.subdev.device;
+	const struct gf100_grctx_func *grctx = gr->func->grctx;
+	u32 tmp;
+	int i;
+
+	gf100_gr_mmio(gr, gr->fuc_sw_ctx);
+
+	nvkm_wr32(device, 0x404154, 0x00000000);
+
+	grctx->bundle(info);
+	grctx->pagepool(info);
+	grctx->attrib(info);
+	grctx->unkn(gr);
+
+	gm200_grctx_generate_tpcid(gr);
+	gf100_grctx_generate_r406028(gr);
+	gk104_grctx_generate_r418bb8(gr);
+
+	for (i = 0; i < 8; i++)
+		nvkm_wr32(device, 0x4064d0 + (i * 0x04), 0x00000000);
+	nvkm_wr32(device, 0x406500, 0x00000000);
+
+	nvkm_wr32(device, 0x405b00, (gr->tpc_total << 8) | gr->gpc_nr);
+
+	gm200_grctx_generate_rop_active_fbps(gr);
+
+	for (tmp = 0, i = 0; i < gr->gpc_nr; i++)
+		tmp |= ((1 << gr->tpc_nr[i]) - 1) << (i * 4);
+	nvkm_wr32(device, 0x4041c4, tmp);
+
+	gm200_grctx_generate_405b60(gr);
+
+	gf100_gr_icmd(gr, gr->fuc_bundle);
+	nvkm_wr32(device, 0x404154, 0x00000800);
+	gf100_gr_mthd(gr, gr->fuc_method);
+
+	nvkm_mask(device, 0x418e94, 0xffffffff, 0xc4230000);
+	nvkm_mask(device, 0x418e4c, 0xffffffff, 0x70000000);
+}
+
+const struct gf100_grctx_func
+gm200_grctx = {
+	.main  = gm200_grctx_generate_main,
+	.unkn  = gk104_grctx_generate_unkn,
+	.bundle = gm107_grctx_generate_bundle,
+	.bundle_size = 0x3000,
+	.bundle_min_gpm_fifo_depth = 0x180,
+	.bundle_token_limit = 0x780,
+	.pagepool = gm107_grctx_generate_pagepool,
+	.pagepool_size = 0x20000,
+	.attrib = gm107_grctx_generate_attrib,
+	.attrib_nr_max = 0x600,
+	.attrib_nr = 0x400,
+	.alpha_nr_max = 0x1800,
+	.alpha_nr = 0x1000,
+};
diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/gr/ctxgm204.c b/drivers/gpu/drm/nouveau/nvkm/engine/gr/ctxgm204.c
deleted file mode 100644
index 170cbfdbe1ae..000000000000
--- a/drivers/gpu/drm/nouveau/nvkm/engine/gr/ctxgm204.c
+++ /dev/null
@@ -1,1049 +0,0 @@
-/*
- * Copyright 2015 Red Hat Inc.
- *
- * Permission is hereby granted, free of charge, to any person obtaining a
- * copy of this software and associated documentation files (the "Software"),
- * to deal in the Software without restriction, including without limitation
- * the rights to use, copy, modify, merge, publish, distribute, sublicense,
- * and/or sell copies of the Software, and to permit persons to whom the
- * Software is furnished to do so, subject to the following conditions:
- *
- * The above copyright notice and this permission notice shall be included in
- * all copies or substantial portions of the Software.
- *
- * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
- * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
- * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
- * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR
- * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
- * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
- * OTHER DEALINGS IN THE SOFTWARE.
- *
- * Authors: Ben Skeggs <bskeggs@redhat.com>
- */
-#include "ctxgf100.h"
-
-/*******************************************************************************
- * PGRAPH context register lists
- ******************************************************************************/
-
-static const struct gf100_gr_init
-gm204_grctx_init_icmd_0[] = {
-	{ 0x001000,   1, 0x01, 0x00000002 },
-	{ 0x0006aa,   1, 0x01, 0x00000001 },
-	{ 0x0006ad,   2, 0x01, 0x00000100 },
-	{ 0x0006b1,   1, 0x01, 0x00000011 },
-	{ 0x00078c,   1, 0x01, 0x00000008 },
-	{ 0x000792,   1, 0x01, 0x00000001 },
-	{ 0x000794,   3, 0x01, 0x00000001 },
-	{ 0x000797,   1, 0x01, 0x000000cf },
-	{ 0x00079a,   1, 0x01, 0x00000002 },
-	{ 0x0007a1,   1, 0x01, 0x00000001 },
-	{ 0x0007a3,   3, 0x01, 0x00000001 },
-	{ 0x000831,   1, 0x01, 0x00000004 },
-	{ 0x01e100,   1, 0x01, 0x00000001 },
-	{ 0x001000,   1, 0x01, 0x00000008 },
-	{ 0x000039,   3, 0x01, 0x00000000 },
-	{ 0x000380,   1, 0x01, 0x00000001 },
-	{ 0x000366,   2, 0x01, 0x00000000 },
-	{ 0x000368,   1, 0x01, 0x00000fff },
-	{ 0x000370,   2, 0x01, 0x00000000 },
-	{ 0x000372,   1, 0x01, 0x000fffff },
-	{ 0x000374,   1, 0x01, 0x00000100 },
-	{ 0x000818,   8, 0x01, 0x00000000 },
-	{ 0x000848,  16, 0x01, 0x00000000 },
-	{ 0x000738,   1, 0x01, 0x00000000 },
-	{ 0x000b07,   1, 0x01, 0x00000002 },
-	{ 0x000b08,   2, 0x01, 0x00000100 },
-	{ 0x000b0a,   1, 0x01, 0x00000001 },
-	{ 0x000a04,   1, 0x01, 0x000000ff },
-	{ 0x000a0b,   1, 0x01, 0x00000040 },
-	{ 0x00097f,   1, 0x01, 0x00000100 },
-	{ 0x000a02,   1, 0x01, 0x00000001 },
-	{ 0x000809,   1, 0x01, 0x00000007 },
-	{ 0x00c221,   1, 0x01, 0x00000040 },
-	{ 0x00c401,   1, 0x01, 0x00000001 },
-	{ 0x00c402,   1, 0x01, 0x00010001 },
-	{ 0x00c403,   2, 0x01, 0x00000001 },
-	{ 0x00c40e,   1, 0x01, 0x00000020 },
-	{ 0x01e100,   1, 0x01, 0x00000001 },
-	{ 0x001000,   1, 0x01, 0x00000001 },
-	{ 0x000b07,   1, 0x01, 0x00000002 },
-	{ 0x000b08,   2, 0x01, 0x00000100 },
-	{ 0x000b0a,   1, 0x01, 0x00000001 },
-	{ 0x01e100,   1, 0x01, 0x00000001 },
-	{ 0x001000,   1, 0x01, 0x00000004 },
-	{ 0x000039,   3, 0x01, 0x00000000 },
-	{ 0x0000a9,   1, 0x01, 0x0000ffff },
-	{ 0x000038,   1, 0x01, 0x0fac6881 },
-	{ 0x00003d,   1, 0x01, 0x00000001 },
-	{ 0x0000e8,   8, 0x01, 0x00000400 },
-	{ 0x000078,   8, 0x01, 0x00000300 },
-	{ 0x000050,   1, 0x01, 0x00000011 },
-	{ 0x000058,   8, 0x01, 0x00000008 },
-	{ 0x000208,   8, 0x01, 0x00000001 },
-	{ 0x000081,   1, 0x01, 0x00000001 },
-	{ 0x000085,   1, 0x01, 0x00000004 },
-	{ 0x000088,   1, 0x01, 0x00000400 },
-	{ 0x000090,   1, 0x01, 0x00000300 },
-	{ 0x000098,   1, 0x01, 0x00001001 },
-	{ 0x0000e3,   1, 0x01, 0x00000001 },
-	{ 0x0000da,   1, 0x01, 0x00000001 },
-	{ 0x0000b4,   4, 0x01, 0x88888888 },
-	{ 0x0000f8,   1, 0x01, 0x00000003 },
-	{ 0x0000fa,   1, 0x01, 0x00000001 },
-	{ 0x0000b1,   2, 0x01, 0x00000001 },
-	{ 0x00009f,   4, 0x01, 0x0000ffff },
-	{ 0x0000a8,   1, 0x01, 0x0000ffff },
-	{ 0x0000ad,   1, 0x01, 0x0000013e },
-	{ 0x0000e1,   1, 0x01, 0x00000010 },
-	{ 0x000290,  16, 0x01, 0x00000000 },
-	{ 0x0003b0,  16, 0x01, 0x00000000 },
-	{ 0x0002a0,  16, 0x01, 0x00000000 },
-	{ 0x000420,  16, 0x01, 0x00000000 },
-	{ 0x0002b0,  16, 0x01, 0x00000000 },
-	{ 0x000430,  16, 0x01, 0x00000000 },
-	{ 0x0002c0,  16, 0x01, 0x00000000 },
-	{ 0x0004d0,  16, 0x01, 0x00000000 },
-	{ 0x000720,  16, 0x01, 0x00000000 },
-	{ 0x0008c0,  16, 0x01, 0x00000000 },
-	{ 0x000890,  16, 0x01, 0x00000000 },
-	{ 0x0008e0,  16, 0x01, 0x00000000 },
-	{ 0x0008a0,  16, 0x01, 0x00000000 },
-	{ 0x0008f0,  16, 0x01, 0x00000000 },
-	{ 0x00094c,   1, 0x01, 0x000000ff },
-	{ 0x00094d,   1, 0x01, 0xffffffff },
-	{ 0x00094e,   1, 0x01, 0x00000002 },
-	{ 0x0002f2,   2, 0x01, 0x00000001 },
-	{ 0x0002f5,   1, 0x01, 0x00000001 },
-	{ 0x0002f7,   1, 0x01, 0x00000001 },
-	{ 0x000303,   1, 0x01, 0x00000001 },
-	{ 0x0002e6,   1, 0x01, 0x00000001 },
-	{ 0x000466,   1, 0x01, 0x00000052 },
-	{ 0x000301,   1, 0x01, 0x3f800000 },
-	{ 0x000304,   1, 0x01, 0x30201000 },
-	{ 0x000305,   1, 0x01, 0x70605040 },
-	{ 0x000306,   1, 0x01, 0xb8a89888 },
-	{ 0x000307,   1, 0x01, 0xf8e8d8c8 },
-	{ 0x00030a,   1, 0x01, 0x00ffff00 },
-	{ 0x00030b,   1, 0x01, 0x0000001a },
-	{ 0x00030c,   1, 0x01, 0x00000001 },
-	{ 0x000318,   1, 0x01, 0x00000001 },
-	{ 0x000340,   1, 0x01, 0x00000000 },
-	{ 0x00037d,   1, 0x01, 0x00000006 },
-	{ 0x0003a0,   1, 0x01, 0x00000002 },
-	{ 0x0003aa,   1, 0x01, 0x00000001 },
-	{ 0x0003a9,   1, 0x01, 0x00000001 },
-	{ 0x000380,   1, 0x01, 0x00000001 },
-	{ 0x000383,   1, 0x01, 0x00000011 },
-	{ 0x000360,   1, 0x01, 0x00000040 },
-	{ 0x000366,   2, 0x01, 0x00000000 },
-	{ 0x000368,   1, 0x01, 0x00000fff },
-	{ 0x000370,   2, 0x01, 0x00000000 },
-	{ 0x000372,   1, 0x01, 0x000fffff },
-	{ 0x000374,   1, 0x01, 0x00000100 },
-	{ 0x00037a,   1, 0x01, 0x00000012 },
-	{ 0x000619,   1, 0x01, 0x00000003 },
-	{ 0x000811,   1, 0x01, 0x00000003 },
-	{ 0x000812,   1, 0x01, 0x00000004 },
-	{ 0x000813,   1, 0x01, 0x00000006 },
-	{ 0x000814,   1, 0x01, 0x00000008 },
-	{ 0x000815,   1, 0x01, 0x0000000b },
-	{ 0x000800,   6, 0x01, 0x00000001 },
-	{ 0x000632,   1, 0x01, 0x00000001 },
-	{ 0x000633,   1, 0x01, 0x00000002 },
-	{ 0x000634,   1, 0x01, 0x00000003 },
-	{ 0x000635,   1, 0x01, 0x00000004 },
-	{ 0x000654,   1, 0x01, 0x3f800000 },
-	{ 0x000657,   1, 0x01, 0x3f800000 },
-	{ 0x000655,   2, 0x01, 0x3f800000 },
-	{ 0x0006cd,   1, 0x01, 0x3f800000 },
-	{ 0x0007f5,   1, 0x01, 0x3f800000 },
-	{ 0x0007dc,   1, 0x01, 0x39291909 },
-	{ 0x0007dd,   1, 0x01, 0x79695949 },
-	{ 0x0007de,   1, 0x01, 0xb9a99989 },
-	{ 0x0007df,   1, 0x01, 0xf9e9d9c9 },
-	{ 0x0007e8,   1, 0x01, 0x00003210 },
-	{ 0x0007e9,   1, 0x01, 0x00007654 },
-	{ 0x0007ea,   1, 0x01, 0x00000098 },
-	{ 0x0007ec,   1, 0x01, 0x39291909 },
-	{ 0x0007ed,   1, 0x01, 0x79695949 },
-	{ 0x0007ee,   1, 0x01, 0xb9a99989 },
-	{ 0x0007ef,   1, 0x01, 0xf9e9d9c9 },
-	{ 0x0007f0,   1, 0x01, 0x00003210 },
-	{ 0x0007f1,   1, 0x01, 0x00007654 },
-	{ 0x0007f2,   1, 0x01, 0x00000098 },
-	{ 0x0005a5,   1, 0x01, 0x00000001 },
-	{ 0x0005aa,   1, 0x01, 0x00000002 },
-	{ 0x0005cb,   1, 0x01, 0x00000004 },
-	{ 0x0005d0,   1, 0x01, 0x20181008 },
-	{ 0x0005d1,   1, 0x01, 0x40383028 },
-	{ 0x0005d2,   1, 0x01, 0x60585048 },
-	{ 0x0005d3,   1, 0x01, 0x80787068 },
-	{ 0x000980, 128, 0x01, 0x00000000 },
-	{ 0x000468,   1, 0x01, 0x00000004 },
-	{ 0x00046c,   1, 0x01, 0x00000001 },
-	{ 0x000470,  96, 0x01, 0x00000000 },
-	{ 0x0005e0,  16, 0x01, 0x00000d10 },
-	{ 0x000510,  16, 0x01, 0x3f800000 },
-	{ 0x000520,   1, 0x01, 0x000002b6 },
-	{ 0x000529,   1, 0x01, 0x00000001 },
-	{ 0x000530,  16, 0x01, 0xffff0000 },
-	{ 0x000550,  32, 0x01, 0xffff0000 },
-	{ 0x000585,   1, 0x01, 0x0000003f },
-	{ 0x000576,   1, 0x01, 0x00000003 },
-	{ 0x00057b,   1, 0x01, 0x00000059 },
-	{ 0x000586,   1, 0x01, 0x00000040 },
-	{ 0x000582,   2, 0x01, 0x00000080 },
-	{ 0x000595,   1, 0x01, 0x00400040 },
-	{ 0x000596,   1, 0x01, 0x00000492 },
-	{ 0x000597,   1, 0x01, 0x08080203 },
-	{ 0x0005ad,   1, 0x01, 0x00000008 },
-	{ 0x000598,   1, 0x01, 0x00020001 },
-	{ 0x0005d4,   1, 0x01, 0x00000001 },
-	{ 0x0005c2,   1, 0x01, 0x00000001 },
-	{ 0x000638,   2, 0x01, 0x00000001 },
-	{ 0x00063a,   1, 0x01, 0x00000002 },
-	{ 0x00063b,   2, 0x01, 0x00000001 },
-	{ 0x00063d,   1, 0x01, 0x00000002 },
-	{ 0x00063e,   1, 0x01, 0x00000001 },
-	{ 0x0008b8,   8, 0x01, 0x00000001 },
-	{ 0x000900,   8, 0x01, 0x00000001 },
-	{ 0x000908,   8, 0x01, 0x00000002 },
-	{ 0x000910,  16, 0x01, 0x00000001 },
-	{ 0x000920,   8, 0x01, 0x00000002 },
-	{ 0x000928,   8, 0x01, 0x00000001 },
-	{ 0x000662,   1, 0x01, 0x00000001 },
-	{ 0x000648,   9, 0x01, 0x00000001 },
-	{ 0x000674,   1, 0x01, 0x00000001 },
-	{ 0x000658,   1, 0x01, 0x0000000f },
-	{ 0x0007ff,   1, 0x01, 0x0000000a },
-	{ 0x00066a,   1, 0x01, 0x40000000 },
-	{ 0x00066b,   1, 0x01, 0x10000000 },
-	{ 0x00066c,   2, 0x01, 0xffff0000 },
-	{ 0x0007af,   2, 0x01, 0x00000008 },
-	{ 0x0007f6,   1, 0x01, 0x00000001 },
-	{ 0x0006b2,   1, 0x01, 0x00000055 },
-	{ 0x0007ad,   1, 0x01, 0x00000003 },
-	{ 0x000971,   1, 0x01, 0x00000008 },
-	{ 0x000972,   1, 0x01, 0x00000040 },
-	{ 0x000973,   1, 0x01, 0x0000012c },
-	{ 0x00097c,   1, 0x01, 0x00000040 },
-	{ 0x000975,   1, 0x01, 0x00000020 },
-	{ 0x000976,   1, 0x01, 0x00000001 },
-	{ 0x000977,   1, 0x01, 0x00000020 },
-	{ 0x000978,   1, 0x01, 0x00000001 },
-	{ 0x000957,   1, 0x01, 0x00000003 },
-	{ 0x00095e,   1, 0x01, 0x20164010 },
-	{ 0x00095f,   1, 0x01, 0x00000020 },
-	{ 0x000a0d,   1, 0x01, 0x00000006 },
-	{ 0x00097d,   1, 0x01, 0x0000000c },
-	{ 0x000683,   1, 0x01, 0x00000006 },
-	{ 0x000687,   1, 0x01, 0x003fffff },
-	{ 0x0006a0,   1, 0x01, 0x00000005 },
-	{ 0x000840,   1, 0x01, 0x00400008 },
-	{ 0x000841,   1, 0x01, 0x08000080 },
-	{ 0x000842,   1, 0x01, 0x00400008 },
-	{ 0x000843,   1, 0x01, 0x08000080 },
-	{ 0x000818,   8, 0x01, 0x00000000 },
-	{ 0x000848,  16, 0x01, 0x00000000 },
-	{ 0x000738,   1, 0x01, 0x00000000 },
-	{ 0x0006aa,   1, 0x01, 0x00000001 },
-	{ 0x0006ab,   1, 0x01, 0x00000002 },
-	{ 0x0006ac,   1, 0x01, 0x00000080 },
-	{ 0x0006ad,   2, 0x01, 0x00000100 },
-	{ 0x0006b1,   1, 0x01, 0x00000011 },
-	{ 0x0006bb,   1, 0x01, 0x000000cf },
-	{ 0x0006ce,   1, 0x01, 0x2a712488 },
-	{ 0x000739,   1, 0x01, 0x4085c000 },
-	{ 0x00073a,   1, 0x01, 0x00000080 },
-	{ 0x000786,   1, 0x01, 0x80000100 },
-	{ 0x00073c,   1, 0x01, 0x00010100 },
-	{ 0x00073d,   1, 0x01, 0x02800000 },
-	{ 0x000787,   1, 0x01, 0x000000cf },
-	{ 0x00078c,   1, 0x01, 0x00000008 },
-	{ 0x000792,   1, 0x01, 0x00000001 },
-	{ 0x000794,   3, 0x01, 0x00000001 },
-	{ 0x000797,   1, 0x01, 0x000000cf },
-	{ 0x000836,   1, 0x01, 0x00000001 },
-	{ 0x00079a,   1, 0x01, 0x00000002 },
-	{ 0x000833,   1, 0x01, 0x04444480 },
-	{ 0x0007a1,   1, 0x01, 0x00000001 },
-	{ 0x0007a3,   3, 0x01, 0x00000001 },
-	{ 0x000831,   1, 0x01, 0x00000004 },
-	{ 0x000b07,   1, 0x01, 0x00000002 },
-	{ 0x000b08,   2, 0x01, 0x00000100 },
-	{ 0x000b0a,   1, 0x01, 0x00000001 },
-	{ 0x000a04,   1, 0x01, 0x000000ff },
-	{ 0x000a0b,   1, 0x01, 0x00000040 },
-	{ 0x00097f,   1, 0x01, 0x00000100 },
-	{ 0x000a02,   1, 0x01, 0x00000001 },
-	{ 0x000809,   1, 0x01, 0x00000007 },
-	{ 0x00c221,   1, 0x01, 0x00000040 },
-	{ 0x00c1b0,   8, 0x01, 0x0000000f },
-	{ 0x00c1b8,   1, 0x01, 0x0fac6881 },
-	{ 0x00c1b9,   1, 0x01, 0x00fac688 },
-	{ 0x00c401,   1, 0x01, 0x00000001 },
-	{ 0x00c402,   1, 0x01, 0x00010001 },
-	{ 0x00c403,   2, 0x01, 0x00000001 },
-	{ 0x00c40e,   1, 0x01, 0x00000020 },
-	{ 0x00c413,   4, 0x01, 0x88888888 },
-	{ 0x00c423,   1, 0x01, 0x0000ff00 },
-	{ 0x00c420,   1, 0x01, 0x00880101 },
-	{ 0x01e100,   1, 0x01, 0x00000001 },
-	{}
-};
-
-const struct gf100_gr_pack
-gm204_grctx_pack_icmd[] = {
-	{ gm204_grctx_init_icmd_0 },
-	{}
-};
-
-static const struct gf100_gr_init
-gm204_grctx_init_b197_0[] = {
-	{ 0x000800,   8, 0x40, 0x00000000 },
-	{ 0x000804,   8, 0x40, 0x00000000 },
-	{ 0x000808,   8, 0x40, 0x00000400 },
-	{ 0x00080c,   8, 0x40, 0x00000300 },
-	{ 0x000810,   1, 0x04, 0x000000cf },
-	{ 0x000850,   7, 0x40, 0x00000000 },
-	{ 0x000814,   8, 0x40, 0x00000040 },
-	{ 0x000818,   8, 0x40, 0x00000001 },
-	{ 0x00081c,   8, 0x40, 0x00000000 },
-	{ 0x000820,   8, 0x40, 0x00000000 },
-	{ 0x001c00,  16, 0x10, 0x00000000 },
-	{ 0x001c04,  16, 0x10, 0x00000000 },
-	{ 0x001c08,  16, 0x10, 0x00000000 },
-	{ 0x001c0c,  16, 0x10, 0x00000000 },
-	{ 0x001d00,  16, 0x10, 0x00000000 },
-	{ 0x001d04,  16, 0x10, 0x00000000 },
-	{ 0x001d08,  16, 0x10, 0x00000000 },
-	{ 0x001d0c,  16, 0x10, 0x00000000 },
-	{ 0x001f00,  16, 0x08, 0x00000000 },
-	{ 0x001f04,  16, 0x08, 0x00000000 },
-	{ 0x001f80,  16, 0x08, 0x00000000 },
-	{ 0x001f84,  16, 0x08, 0x00000000 },
-	{ 0x002000,   1, 0x04, 0x00000000 },
-	{ 0x002040,   1, 0x04, 0x00000011 },
-	{ 0x002080,   1, 0x04, 0x00000020 },
-	{ 0x0020c0,   1, 0x04, 0x00000030 },
-	{ 0x002100,   1, 0x04, 0x00000040 },
-	{ 0x002140,   1, 0x04, 0x00000051 },
-	{ 0x00200c,   6, 0x40, 0x00000001 },
-	{ 0x002010,   1, 0x04, 0x00000000 },
-	{ 0x002050,   1, 0x04, 0x00000000 },
-	{ 0x002090,   1, 0x04, 0x00000001 },
-	{ 0x0020d0,   1, 0x04, 0x00000002 },
-	{ 0x002110,   1, 0x04, 0x00000003 },
-	{ 0x002150,   1, 0x04, 0x00000004 },
-	{ 0x000380,   4, 0x20, 0x00000000 },
-	{ 0x000384,   4, 0x20, 0x00000000 },
-	{ 0x000388,   4, 0x20, 0x00000000 },
-	{ 0x00038c,   4, 0x20, 0x00000000 },
-	{ 0x000700,   4, 0x10, 0x00000000 },
-	{ 0x000704,   4, 0x10, 0x00000000 },
-	{ 0x000708,   4, 0x10, 0x00000000 },
-	{ 0x002800, 128, 0x04, 0x00000000 },
-	{ 0x000a00,  16, 0x20, 0x00000000 },
-	{ 0x000a04,  16, 0x20, 0x00000000 },
-	{ 0x000a08,  16, 0x20, 0x00000000 },
-	{ 0x000a0c,  16, 0x20, 0x00000000 },
-	{ 0x000a10,  16, 0x20, 0x00000000 },
-	{ 0x000a14,  16, 0x20, 0x00000000 },
-	{ 0x000a18,  16, 0x20, 0x00006420 },
-	{ 0x000a1c,  16, 0x20, 0x00000000 },
-	{ 0x000c00,  16, 0x10, 0x00000000 },
-	{ 0x000c04,  16, 0x10, 0x00000000 },
-	{ 0x000c08,  16, 0x10, 0x00000000 },
-	{ 0x000c0c,  16, 0x10, 0x3f800000 },
-	{ 0x000d00,   8, 0x08, 0xffff0000 },
-	{ 0x000d04,   8, 0x08, 0xffff0000 },
-	{ 0x000e00,  16, 0x10, 0x00000000 },
-	{ 0x000e04,  16, 0x10, 0xffff0000 },
-	{ 0x000e08,  16, 0x10, 0xffff0000 },
-	{ 0x000d40,   4, 0x08, 0x00000000 },
-	{ 0x000d44,   4, 0x08, 0x00000000 },
-	{ 0x001e00,   8, 0x20, 0x00000001 },
-	{ 0x001e04,   8, 0x20, 0x00000001 },
-	{ 0x001e08,   8, 0x20, 0x00000002 },
-	{ 0x001e0c,   8, 0x20, 0x00000001 },
-	{ 0x001e10,   8, 0x20, 0x00000001 },
-	{ 0x001e14,   8, 0x20, 0x00000002 },
-	{ 0x001e18,   8, 0x20, 0x00000001 },
-	{ 0x001480,   8, 0x10, 0x00000000 },
-	{ 0x001484,   8, 0x10, 0x00000000 },
-	{ 0x001488,   8, 0x10, 0x00000000 },
-	{ 0x003400, 128, 0x04, 0x00000000 },
-	{ 0x00030c,   1, 0x04, 0x00000001 },
-	{ 0x001944,   1, 0x04, 0x00000000 },
-	{ 0x001514,   1, 0x04, 0x00000000 },
-	{ 0x000d68,   1, 0x04, 0x0000ffff },
-	{ 0x00121c,   1, 0x04, 0x0fac6881 },
-	{ 0x000fac,   1, 0x04, 0x00000001 },
-	{ 0x001538,   1, 0x04, 0x00000001 },
-	{ 0x000fe0,   2, 0x04, 0x00000000 },
-	{ 0x000fe8,   1, 0x04, 0x00000014 },
-	{ 0x000fec,   1, 0x04, 0x00000040 },
-	{ 0x000ff0,   1, 0x04, 0x00000000 },
-	{ 0x00179c,   1, 0x04, 0x00000000 },
-	{ 0x001228,   1, 0x04, 0x00000400 },
-	{ 0x00122c,   1, 0x04, 0x00000300 },
-	{ 0x001230,   1, 0x04, 0x00010001 },
-	{ 0x0007f8,   1, 0x04, 0x00000000 },
-	{ 0x001208,   1, 0x04, 0x00000000 },
-	{ 0x0015b4,   1, 0x04, 0x00000001 },
-	{ 0x0015cc,   1, 0x04, 0x00000000 },
-	{ 0x001534,   1, 0x04, 0x00000000 },
-	{ 0x000754,   1, 0x04, 0x00000001 },
-	{ 0x000fb0,   1, 0x04, 0x00000000 },
-	{ 0x0015d0,   1, 0x04, 0x00000000 },
-	{ 0x0011e0,   4, 0x04, 0x88888888 },
-	{ 0x00153c,   1, 0x04, 0x00000000 },
-	{ 0x0016b4,   1, 0x04, 0x00000003 },
-	{ 0x000fa4,   1, 0x04, 0x00000001 },
-	{ 0x000fbc,   4, 0x04, 0x0000ffff },
-	{ 0x000fa8,   1, 0x04, 0x0000ffff },
-	{ 0x000df8,   2, 0x04, 0x00000000 },
-	{ 0x001948,   1, 0x04, 0x00000000 },
-	{ 0x001970,   1, 0x04, 0x00000001 },
-	{ 0x00161c,   1, 0x04, 0x000009f0 },
-	{ 0x000dcc,   1, 0x04, 0x00000010 },
-	{ 0x0015e4,   1, 0x04, 0x00000000 },
-	{ 0x001160,  32, 0x04, 0x25e00040 },
-	{ 0x001880,  32, 0x04, 0x00000000 },
-	{ 0x000f84,   2, 0x04, 0x00000000 },
-	{ 0x0017c8,   2, 0x04, 0x00000000 },
-	{ 0x0017d0,   1, 0x04, 0x000000ff },
-	{ 0x0017d4,   1, 0x04, 0xffffffff },
-	{ 0x0017d8,   1, 0x04, 0x00000002 },
-	{ 0x0017dc,   1, 0x04, 0x00000000 },
-	{ 0x0015f4,   2, 0x04, 0x00000000 },
-	{ 0x001434,   2, 0x04, 0x00000000 },
-	{ 0x000d74,   1, 0x04, 0x00000000 },
-	{ 0x0013a4,   1, 0x04, 0x00000000 },
-	{ 0x001318,   1, 0x04, 0x00000001 },
-	{ 0x001080,   2, 0x04, 0x00000000 },
-	{ 0x001088,   2, 0x04, 0x00000001 },
-	{ 0x001090,   1, 0x04, 0x00000000 },
-	{ 0x001094,   1, 0x04, 0x00000001 },
-	{ 0x001098,   1, 0x04, 0x00000000 },
-	{ 0x00109c,   1, 0x04, 0x00000001 },
-	{ 0x0010a0,   2, 0x04, 0x00000000 },
-	{ 0x001644,   1, 0x04, 0x00000000 },
-	{ 0x000748,   1, 0x04, 0x00000000 },
-	{ 0x000de8,   1, 0x04, 0x00000000 },
-	{ 0x001648,   1, 0x04, 0x00000000 },
-	{ 0x0012a4,   1, 0x04, 0x00000000 },
-	{ 0x001120,   4, 0x04, 0x00000000 },
-	{ 0x001118,   1, 0x04, 0x00000000 },
-	{ 0x00164c,   1, 0x04, 0x00000000 },
-	{ 0x001658,   1, 0x04, 0x00000000 },
-	{ 0x001910,   1, 0x04, 0x00000290 },
-	{ 0x001518,   1, 0x04, 0x00000000 },
-	{ 0x00165c,   1, 0x04, 0x00000001 },
-	{ 0x001520,   1, 0x04, 0x00000000 },
-	{ 0x001604,   1, 0x04, 0x00000000 },
-	{ 0x001570,   1, 0x04, 0x00000000 },
-	{ 0x0013b0,   2, 0x04, 0x3f800000 },
-	{ 0x00020c,   1, 0x04, 0x00000000 },
-	{ 0x001670,   1, 0x04, 0x30201000 },
-	{ 0x001674,   1, 0x04, 0x70605040 },
-	{ 0x001678,   1, 0x04, 0xb8a89888 },
-	{ 0x00167c,   1, 0x04, 0xf8e8d8c8 },
-	{ 0x00166c,   1, 0x04, 0x00000000 },
-	{ 0x001680,   1, 0x04, 0x00ffff00 },
-	{ 0x0012d0,   1, 0x04, 0x00000003 },
-	{ 0x00113c,   1, 0x04, 0x00000000 },
-	{ 0x0012d4,   1, 0x04, 0x00000002 },
-	{ 0x001684,   2, 0x04, 0x00000000 },
-	{ 0x000dac,   2, 0x04, 0x00001b02 },
-	{ 0x000db4,   1, 0x04, 0x00000000 },
-	{ 0x00168c,   1, 0x04, 0x00000000 },
-	{ 0x0015bc,   1, 0x04, 0x00000000 },
-	{ 0x00156c,   1, 0x04, 0x00000000 },
-	{ 0x00187c,   1, 0x04, 0x00000000 },
-	{ 0x001110,   1, 0x04, 0x00000001 },
-	{ 0x000dc0,   3, 0x04, 0x00000000 },
-	{ 0x000f40,   5, 0x04, 0x00000000 },
-	{ 0x001234,   1, 0x04, 0x00000000 },
-	{ 0x001690,   1, 0x04, 0x00000000 },
-	{ 0x000790,   5, 0x04, 0x00000000 },
-	{ 0x00077c,   1, 0x04, 0x00000000 },
-	{ 0x001000,   1, 0x04, 0x00000010 },
-	{ 0x0010fc,   1, 0x04, 0x00000000 },
-	{ 0x001290,   1, 0x04, 0x00000000 },
-	{ 0x000218,   1, 0x04, 0x00000010 },
-	{ 0x0012d8,   1, 0x04, 0x00000000 },
-	{ 0x0012dc,   1, 0x04, 0x00000010 },
-	{ 0x000d94,   1, 0x04, 0x00000001 },
-	{ 0x00155c,   2, 0x04, 0x00000000 },
-	{ 0x001564,   1, 0x04, 0x00000fff },
-	{ 0x001574,   2, 0x04, 0x00000000 },
-	{ 0x00157c,   1, 0x04, 0x000fffff },
-	{ 0x001354,   1, 0x04, 0x00000000 },
-	{ 0x001610,   1, 0x04, 0x00000012 },
-	{ 0x001608,   2, 0x04, 0x00000000 },
-	{ 0x00260c,   1, 0x04, 0x00000000 },
-	{ 0x0007ac,   1, 0x04, 0x00000000 },
-	{ 0x00162c,   1, 0x04, 0x00000003 },
-	{ 0x000210,   1, 0x04, 0x00000000 },
-	{ 0x000320,   1, 0x04, 0x00000000 },
-	{ 0x000324,   6, 0x04, 0x3f800000 },
-	{ 0x000750,   1, 0x04, 0x00000000 },
-	{ 0x000760,   1, 0x04, 0x39291909 },
-	{ 0x000764,   1, 0x04, 0x79695949 },
-	{ 0x000768,   1, 0x04, 0xb9a99989 },
-	{ 0x00076c,   1, 0x04, 0xf9e9d9c9 },
-	{ 0x000770,   1, 0x04, 0x30201000 },
-	{ 0x000774,   1, 0x04, 0x70605040 },
-	{ 0x000778,   1, 0x04, 0x00009080 },
-	{ 0x000780,   1, 0x04, 0x39291909 },
-	{ 0x000784,   1, 0x04, 0x79695949 },
-	{ 0x000788,   1, 0x04, 0xb9a99989 },
-	{ 0x00078c,   1, 0x04, 0xf9e9d9c9 },
-	{ 0x0007d0,   1, 0x04, 0x30201000 },
-	{ 0x0007d4,   1, 0x04, 0x70605040 },
-	{ 0x0007d8,   1, 0x04, 0x00009080 },
-	{ 0x001004,   1, 0x04, 0x00000000 },
-	{ 0x001240,   8, 0x04, 0x00000000 },
-	{ 0x00037c,   1, 0x04, 0x00000001 },
-	{ 0x000740,   1, 0x04, 0x00000000 },
-	{ 0x001148,   1, 0x04, 0x00000000 },
-	{ 0x000fb4,   1, 0x04, 0x00000000 },
-	{ 0x000fb8,   1, 0x04, 0x00000002 },
-	{ 0x001130,   1, 0x04, 0x00000002 },
-	{ 0x000fd4,   2, 0x04, 0x00000000 },
-	{ 0x001030,   1, 0x04, 0x20181008 },
-	{ 0x001034,   1, 0x04, 0x40383028 },
-	{ 0x001038,   1, 0x04, 0x60585048 },
-	{ 0x00103c,   1, 0x04, 0x80787068 },
-	{ 0x000744,   1, 0x04, 0x00000000 },
-	{ 0x002600,   1, 0x04, 0x00000000 },
-	{ 0x001918,   1, 0x04, 0x00000000 },
-	{ 0x00191c,   1, 0x04, 0x00000900 },
-	{ 0x001920,   1, 0x04, 0x00000405 },
-	{ 0x001308,   1, 0x04, 0x00000001 },
-	{ 0x001924,   1, 0x04, 0x00000000 },
-	{ 0x0013ac,   1, 0x04, 0x00000000 },
-	{ 0x00192c,   1, 0x04, 0x00000001 },
-	{ 0x00193c,   1, 0x04, 0x00002c1c },
-	{ 0x000d7c,   1, 0x04, 0x00000000 },
-	{ 0x000f8c,   1, 0x04, 0x00000000 },
-	{ 0x0002c0,   1, 0x04, 0x00000001 },
-	{ 0x001510,   1, 0x04, 0x00000000 },
-	{ 0x001940,   1, 0x04, 0x00000000 },
-	{ 0x000ff4,   2, 0x04, 0x00000000 },
-	{ 0x00194c,   2, 0x04, 0x00000000 },
-	{ 0x001968,   1, 0x04, 0x00000000 },
-	{ 0x001590,   1, 0x04, 0x0000003f },
-	{ 0x0007e8,   4, 0x04, 0x00000000 },
-	{ 0x00196c,   1, 0x04, 0x00000011 },
-	{ 0x0002e4,   1, 0x04, 0x0000b001 },
-	{ 0x00036c,   2, 0x04, 0x00000000 },
-	{ 0x00197c,   1, 0x04, 0x00000000 },
-	{ 0x000fcc,   2, 0x04, 0x00000000 },
-	{ 0x0002d8,   1, 0x04, 0x00000040 },
-	{ 0x001980,   1, 0x04, 0x00000080 },
-	{ 0x001504,   1, 0x04, 0x00000080 },
-	{ 0x001984,   1, 0x04, 0x00000000 },
-	{ 0x000f60,   1, 0x04, 0x00000000 },
-	{ 0x000f64,   1, 0x04, 0x00400040 },
-	{ 0x000f68,   1, 0x04, 0x00002212 },
-	{ 0x000f6c,   1, 0x04, 0x08080203 },
-	{ 0x001108,   1, 0x04, 0x00000008 },
-	{ 0x000f70,   1, 0x04, 0x00080001 },
-	{ 0x000ffc,   1, 0x04, 0x00000000 },
-	{ 0x001134,   1, 0x04, 0x00000000 },
-	{ 0x000f1c,   1, 0x04, 0x00000000 },
-	{ 0x0011f8,   1, 0x04, 0x00000000 },
-	{ 0x001138,   1, 0x04, 0x00000001 },
-	{ 0x000300,   1, 0x04, 0x00000001 },
-	{ 0x0013a8,   1, 0x04, 0x00000000 },
-	{ 0x001224,   1, 0x04, 0x00000000 },
-	{ 0x0012ec,   1, 0x04, 0x00000000 },
-	{ 0x001310,   1, 0x04, 0x00000000 },
-	{ 0x001314,   1, 0x04, 0x00000001 },
-	{ 0x001380,   1, 0x04, 0x00000000 },
-	{ 0x001384,   4, 0x04, 0x00000001 },
-	{ 0x001394,   1, 0x04, 0x00000000 },
-	{ 0x00139c,   1, 0x04, 0x00000000 },
-	{ 0x001398,   1, 0x04, 0x00000000 },
-	{ 0x001594,   1, 0x04, 0x00000000 },
-	{ 0x001598,   4, 0x04, 0x00000001 },
-	{ 0x000f54,   3, 0x04, 0x00000000 },
-	{ 0x0019bc,   1, 0x04, 0x00000000 },
-	{ 0x000f9c,   2, 0x04, 0x00000000 },
-	{ 0x0012cc,   1, 0x04, 0x00000000 },
-	{ 0x0012e8,   1, 0x04, 0x00000000 },
-	{ 0x00130c,   1, 0x04, 0x00000001 },
-	{ 0x001360,   8, 0x04, 0x00000000 },
-	{ 0x00133c,   2, 0x04, 0x00000001 },
-	{ 0x001344,   1, 0x04, 0x00000002 },
-	{ 0x001348,   2, 0x04, 0x00000001 },
-	{ 0x001350,   1, 0x04, 0x00000002 },
-	{ 0x001358,   1, 0x04, 0x00000001 },
-	{ 0x0012e4,   1, 0x04, 0x00000000 },
-	{ 0x00131c,   4, 0x04, 0x00000000 },
-	{ 0x0019c0,   1, 0x04, 0x00000000 },
-	{ 0x001140,   1, 0x04, 0x00000000 },
-	{ 0x000dd0,   1, 0x04, 0x00000000 },
-	{ 0x000dd4,   1, 0x04, 0x00000001 },
-	{ 0x0002f4,   1, 0x04, 0x00000000 },
-	{ 0x0019c4,   1, 0x04, 0x00000000 },
-	{ 0x0019c8,   1, 0x04, 0x00001500 },
-	{ 0x00135c,   1, 0x04, 0x00000000 },
-	{ 0x000f90,   1, 0x04, 0x00000000 },
-	{ 0x0019e0,   8, 0x04, 0x00000001 },
-	{ 0x0019cc,   1, 0x04, 0x00000001 },
-	{ 0x00111c,   1, 0x04, 0x00000001 },
-	{ 0x0015b8,   1, 0x04, 0x00000000 },
-	{ 0x001a00,   1, 0x04, 0x00001111 },
-	{ 0x001a04,   7, 0x04, 0x00000000 },
-	{ 0x000d6c,   2, 0x04, 0xffff0000 },
-	{ 0x0010f8,   1, 0x04, 0x00001010 },
-	{ 0x000d80,   5, 0x04, 0x00000000 },
-	{ 0x000da0,   1, 0x04, 0x00000000 },
-	{ 0x0007a4,   2, 0x04, 0x00000000 },
-	{ 0x001508,   1, 0x04, 0x80000000 },
-	{ 0x00150c,   1, 0x04, 0x40000000 },
-	{ 0x001668,   1, 0x04, 0x00000000 },
-	{ 0x000318,   2, 0x04, 0x00000008 },
-	{ 0x000d9c,   1, 0x04, 0x00000001 },
-	{ 0x000f14,   1, 0x04, 0x00000000 },
-	{ 0x000374,   1, 0x04, 0x00000000 },
-	{ 0x000378,   1, 0x04, 0x0000000c },
-	{ 0x0007dc,   1, 0x04, 0x00000000 },
-	{ 0x00074c,   1, 0x04, 0x00000055 },
-	{ 0x001420,   1, 0x04, 0x00000003 },
-	{ 0x001008,   1, 0x04, 0x00000008 },
-	{ 0x00100c,   1, 0x04, 0x00000040 },
-	{ 0x001010,   1, 0x04, 0x0000012c },
-	{ 0x000d60,   1, 0x04, 0x00000040 },
-	{ 0x001018,   1, 0x04, 0x00000020 },
-	{ 0x00101c,   1, 0x04, 0x00000001 },
-	{ 0x001020,   1, 0x04, 0x00000020 },
-	{ 0x001024,   1, 0x04, 0x00000001 },
-	{ 0x001444,   3, 0x04, 0x00000000 },
-	{ 0x000360,   1, 0x04, 0x20164010 },
-	{ 0x000364,   1, 0x04, 0x00000020 },
-	{ 0x000368,   1, 0x04, 0x00000000 },
-	{ 0x000da8,   1, 0x04, 0x00000030 },
-	{ 0x000de4,   1, 0x04, 0x00000000 },
-	{ 0x000204,   1, 0x04, 0x00000006 },
-	{ 0x0002d0,   1, 0x04, 0x003fffff },
-	{ 0x001220,   1, 0x04, 0x00000005 },
-	{ 0x000fdc,   1, 0x04, 0x00000000 },
-	{ 0x000f98,   1, 0x04, 0x00400008 },
-	{ 0x001284,   1, 0x04, 0x08000080 },
-	{ 0x001450,   1, 0x04, 0x00400008 },
-	{ 0x001454,   1, 0x04, 0x08000080 },
-	{ 0x000214,   1, 0x04, 0x00000000 },
-	{}
-};
-
-const struct gf100_gr_pack
-gm204_grctx_pack_mthd[] = {
-	{ gm204_grctx_init_b197_0, 0xb197 },
-	{ gf100_grctx_init_902d_0, 0x902d },
-	{}
-};
-
-static const struct gf100_gr_init
-gm204_grctx_init_fe_0[] = {
-	{ 0x404004,   8, 0x04, 0x00000000 },
-	{ 0x404024,   1, 0x04, 0x0000e000 },
-	{ 0x404028,   8, 0x04, 0x00000000 },
-	{ 0x4040a8,   8, 0x04, 0x00000000 },
-	{ 0x4040c8,   1, 0x04, 0xf801008f },
-	{ 0x4040d0,   6, 0x04, 0x00000000 },
-	{ 0x4040f8,   1, 0x04, 0x00000000 },
-	{ 0x404100,  10, 0x04, 0x00000000 },
-	{ 0x404130,   2, 0x04, 0x00000000 },
-	{ 0x404150,   1, 0x04, 0x0000002e },
-	{ 0x404154,   2, 0x04, 0x00000800 },
-	{ 0x404164,   1, 0x04, 0x00000045 },
-	{ 0x40417c,   2, 0x04, 0x00000000 },
-	{ 0x404194,   1, 0x04, 0x33000700 },
-	{ 0x4041a0,   4, 0x04, 0x00000000 },
-	{ 0x4041c4,   2, 0x04, 0x00000000 },
-	{}
-};
-
-static const struct gf100_gr_init
-gm204_grctx_init_ds_0[] = {
-	{ 0x405800,   1, 0x04, 0x8f8001bf },
-	{ 0x405830,   1, 0x04, 0x04001000 },
-	{ 0x405834,   1, 0x04, 0x08000000 },
-	{ 0x405838,   1, 0x04, 0x00010000 },
-	{ 0x405854,   1, 0x04, 0x00000000 },
-	{ 0x405870,   4, 0x04, 0x00000001 },
-	{ 0x405a00,   2, 0x04, 0x00000000 },
-	{ 0x405a18,   1, 0x04, 0x00000000 },
-	{ 0x405a1c,   1, 0x04, 0x000000ff },
-	{}
-};
-
-static const struct gf100_gr_init
-gm204_grctx_init_cwd_0[] = {
-	{ 0x405b00,   1, 0x04, 0x00000000 },
-	{ 0x405b10,   1, 0x04, 0x00001000 },
-	{ 0x405b20,   1, 0x04, 0x04000000 },
-	{ 0x405b60,   6, 0x04, 0x00000000 },
-	{ 0x405ba0,   6, 0x04, 0x00000000 },
-	{}
-};
-
-static const struct gf100_gr_init
-gm204_grctx_init_pd_0[] = {
-	{ 0x406020,   1, 0x04, 0x17410001 },
-	{ 0x406028,   4, 0x04, 0x00000001 },
-	{ 0x4064a8,   1, 0x04, 0x00000000 },
-	{ 0x4064ac,   1, 0x04, 0x00003fff },
-	{ 0x4064b0,   3, 0x04, 0x00000000 },
-	{ 0x4064c0,   1, 0x04, 0x80400280 },
-	{ 0x4064c4,   1, 0x04, 0x0400ffff },
-	{ 0x4064c8,   1, 0x04, 0x01800780 },
-	{ 0x4064cc,   9, 0x04, 0x00000000 },
-	{ 0x4064fc,   1, 0x04, 0x0000022a },
-	{ 0x406500,   1, 0x04, 0x00000000 },
-	{}
-};
-
-static const struct gf100_gr_init
-gm204_grctx_init_be_0[] = {
-	{ 0x408800,   1, 0x04, 0x32882a3c },
-	{ 0x408804,   1, 0x04, 0x00000040 },
-	{ 0x408808,   1, 0x04, 0x1003e005 },
-	{ 0x408840,   1, 0x04, 0x00000e0b },
-	{ 0x408900,   1, 0x04, 0xb080b801 },
-	{ 0x408904,   1, 0x04, 0x63038001 },
-	{ 0x408908,   1, 0x04, 0x12c8502f },
-	{ 0x408980,   1, 0x04, 0x0000011d },
-	{}
-};
-
-const struct gf100_gr_pack
-gm204_grctx_pack_hub[] = {
-	{ gf100_grctx_init_main_0 },
-	{ gm204_grctx_init_fe_0 },
-	{ gk110_grctx_init_pri_0 },
-	{ gk104_grctx_init_memfmt_0 },
-	{ gm204_grctx_init_ds_0 },
-	{ gm204_grctx_init_cwd_0 },
-	{ gm204_grctx_init_pd_0 },
-	{ gk208_grctx_init_rstr2d_0 },
-	{ gk104_grctx_init_scc_0 },
-	{ gm204_grctx_init_be_0 },
-	{}
-};
-
-const struct gf100_gr_init
-gm204_grctx_init_prop_0[] = {
-	{ 0x418400,   1, 0x04, 0x38e01e00 },
-	{ 0x418404,   1, 0x04, 0x70001fff },
-	{ 0x41840c,   1, 0x04, 0x20001008 },
-	{ 0x418410,   2, 0x04, 0x0fff0fff },
-	{ 0x418418,   1, 0x04, 0x07ff07ff },
-	{ 0x41841c,   1, 0x04, 0x3feffbff },
-	{ 0x418450,   6, 0x04, 0x00000000 },
-	{ 0x418468,   1, 0x04, 0x00000001 },
-	{ 0x41846c,   2, 0x04, 0x00000000 },
-	{}
-};
-
-static const struct gf100_gr_init
-gm204_grctx_init_gpc_unk_1[] = {
-	{ 0x418600,   1, 0x04, 0x0000007f },
-	{ 0x418684,   1, 0x04, 0x0000001f },
-	{ 0x418700,   1, 0x04, 0x00000002 },
-	{ 0x418704,   1, 0x04, 0x00000080 },
-	{ 0x418708,   1, 0x04, 0x40000000 },
-	{ 0x41870c,   2, 0x04, 0x00000000 },
-	{ 0x418728,   1, 0x04, 0x00010000 },
-	{}
-};
-
-const struct gf100_gr_init
-gm204_grctx_init_setup_0[] = {
-	{ 0x418800,   1, 0x04, 0x7006863a },
-	{ 0x418808,   1, 0x04, 0x00000000 },
-	{ 0x418810,   1, 0x04, 0x00000000 },
-	{ 0x418828,   1, 0x04, 0x00000044 },
-	{ 0x418830,   1, 0x04, 0x10000001 },
-	{ 0x4188d8,   1, 0x04, 0x00000008 },
-	{ 0x4188e0,   1, 0x04, 0x01000000 },
-	{ 0x4188e8,   5, 0x04, 0x00000000 },
-	{ 0x4188fc,   1, 0x04, 0x20100058 },
-	{}
-};
-
-const struct gf100_gr_init
-gm204_grctx_init_gpm_0[] = {
-	{ 0x418c10,   8, 0x04, 0x00000000 },
-	{ 0x418c40,   1, 0x04, 0xffffffff },
-	{ 0x418c6c,   1, 0x04, 0x00000001 },
-	{ 0x418c80,   1, 0x04, 0x20200000 },
-	{}
-};
-
-const struct gf100_gr_init
-gm204_grctx_init_gpc_unk_2[] = {
-	{ 0x418e00,   1, 0x04, 0x90040000 },
-	{ 0x418e24,   1, 0x04, 0x00000000 },
-	{ 0x418e28,   1, 0x04, 0x00000030 },
-	{ 0x418e2c,   1, 0x04, 0x00000100 },
-	{ 0x418e30,   3, 0x04, 0x00000000 },
-	{ 0x418e40,  22, 0x04, 0x00000000 },
-	{ 0x418ea0,  12, 0x04, 0x00000000 },
-	{}
-};
-
-static const struct gf100_gr_pack
-gm204_grctx_pack_gpc[] = {
-	{ gm107_grctx_init_gpc_unk_0 },
-	{ gm204_grctx_init_prop_0 },
-	{ gm204_grctx_init_gpc_unk_1 },
-	{ gm204_grctx_init_setup_0 },
-	{ gf100_grctx_init_zcull_0 },
-	{ gk208_grctx_init_crstr_0 },
-	{ gm204_grctx_init_gpm_0 },
-	{ gm204_grctx_init_gpc_unk_2 },
-	{ gf100_grctx_init_gcc_0 },
-	{}
-};
-
-static const struct gf100_gr_init
-gm204_grctx_init_pe_0[] = {
-	{ 0x419848,   1, 0x04, 0x00000000 },
-	{ 0x419864,   1, 0x04, 0x00000029 },
-	{ 0x419888,   1, 0x04, 0x00000000 },
-	{}
-};
-
-static const struct gf100_gr_init
-gm204_grctx_init_tex_0[] = {
-	{ 0x419a00,   1, 0x04, 0x000100f0 },
-	{ 0x419a04,   1, 0x04, 0x00000005 },
-	{ 0x419a08,   1, 0x04, 0x00000621 },
-	{ 0x419a0c,   1, 0x04, 0x00320000 },
-	{ 0x419a10,   1, 0x04, 0x00000000 },
-	{ 0x419a14,   1, 0x04, 0x00000200 },
-	{ 0x419a1c,   1, 0x04, 0x0010c000 },
-	{ 0x419a20,   1, 0x04, 0x20008a00 },
-	{ 0x419a30,   1, 0x04, 0x00000001 },
-	{ 0x419a3c,   1, 0x04, 0x0000181e },
-	{ 0x419ac4,   1, 0x04, 0x00000000 },
-	{}
-};
-
-static const struct gf100_gr_init
-gm204_grctx_init_mpc_0[] = {
-	{ 0x419c00,   1, 0x04, 0x0000009a },
-	{ 0x419c04,   1, 0x04, 0x80000bd6 },
-	{ 0x419c08,   1, 0x04, 0x00000002 },
-	{ 0x419c20,   1, 0x04, 0x00000000 },
-	{ 0x419c24,   1, 0x04, 0x00084210 },
-	{ 0x419c28,   1, 0x04, 0x3efbefbe },
-	{ 0x419c2c,   1, 0x04, 0x00000000 },
-	{ 0x419c34,   1, 0x04, 0x71ff1ff3 },
-	{ 0x419c3c,   1, 0x04, 0x00001919 },
-	{ 0x419c50,   1, 0x04, 0x00000005 },
-	{}
-};
-
-static const struct gf100_gr_init
-gm204_grctx_init_l1c_0[] = {
-	{ 0x419c84,   1, 0x04, 0x0000003e },
-	{ 0x419c90,   1, 0x04, 0x0000000a },
-	{}
-};
-
-static const struct gf100_gr_init
-gm204_grctx_init_sm_0[] = {
-	{ 0x419e04,   3, 0x04, 0x00000000 },
-	{ 0x419e10,   1, 0x04, 0x00001c02 },
-	{ 0x419e44,   1, 0x04, 0x00d3eff2 },
-	{ 0x419e48,   1, 0x04, 0x00000000 },
-	{ 0x419e4c,   1, 0x04, 0x0000007f },
-	{ 0x419e50,   1, 0x04, 0x00000000 },
-	{ 0x419e58,   6, 0x04, 0x00000000 },
-	{ 0x419e74,  10, 0x04, 0x00000000 },
-	{ 0x419eac,   1, 0x04, 0x0001cf8b },
-	{ 0x419eb0,   1, 0x04, 0x00030300 },
-	{ 0x419eb8,   1, 0x04, 0x40000000 },
-	{ 0x419ef0,  24, 0x04, 0x00000000 },
-	{ 0x419f68,   2, 0x04, 0x00000000 },
-	{ 0x419f70,   1, 0x04, 0x00000020 },
-	{ 0x419f78,   1, 0x04, 0x00010beb },
-	{ 0x419f7c,   1, 0x04, 0x00000000 },
-	{}
-};
-
-const struct gf100_gr_pack
-gm204_grctx_pack_tpc[] = {
-	{ gm204_grctx_init_pe_0 },
-	{ gm204_grctx_init_tex_0 },
-	{ gm204_grctx_init_mpc_0 },
-	{ gm204_grctx_init_l1c_0 },
-	{ gm204_grctx_init_sm_0 },
-	{}
-};
-
-static const struct gf100_gr_init
-gm204_grctx_init_pes_0[] = {
-	{ 0x41be24,   1, 0x04, 0x0000000e },
-	{}
-};
-
-static const struct gf100_gr_init
-gm204_grctx_init_cbm_0[] = {
-	{ 0x41bec0,   1, 0x04, 0x00000000 },
-	{ 0x41bec4,   1, 0x04, 0x01030000 },
-	{ 0x41bee4,   1, 0x04, 0x00000000 },
-	{ 0x41bef0,   1, 0x04, 0x000003ff },
-	{ 0x41bef4,   2, 0x04, 0x00000000 },
-	{}
-};
-
-const struct gf100_gr_pack
-gm204_grctx_pack_ppc[] = {
-	{ gm204_grctx_init_pes_0 },
-	{ gm204_grctx_init_cbm_0 },
-	{ gm107_grctx_init_wwdx_0 },
-	{}
-};
-
-/*******************************************************************************
- * PGRAPH context implementation
- ******************************************************************************/
-
-void
-gm204_grctx_generate_tpcid(struct gf100_gr *gr)
-{
-	struct nvkm_device *device = gr->base.engine.subdev.device;
-	int gpc, tpc, id;
-
-	for (tpc = 0, id = 0; tpc < 4; tpc++) {
-		for (gpc = 0; gpc < gr->gpc_nr; gpc++) {
-			if (tpc < gr->tpc_nr[gpc]) {
-				nvkm_wr32(device, TPC_UNIT(gpc, tpc, 0x698), id);
-				nvkm_wr32(device, GPC_UNIT(gpc, 0x0c10 + tpc * 4), id);
-				nvkm_wr32(device, TPC_UNIT(gpc, tpc, 0x088), id);
-				id++;
-			}
-		}
-	}
-}
-
-static void
-gm204_grctx_generate_rop_active_fbps(struct gf100_gr *gr)
-{
-	struct nvkm_device *device = gr->base.engine.subdev.device;
-	const u32 fbp_count = nvkm_rd32(device, 0x12006c);
-	nvkm_mask(device, 0x408850, 0x0000000f, fbp_count); /* zrop */
-	nvkm_mask(device, 0x408958, 0x0000000f, fbp_count); /* crop */
-}
-
-void
-gm204_grctx_generate_405b60(struct gf100_gr *gr)
-{
-	struct nvkm_device *device = gr->base.engine.subdev.device;
-	const u32 dist_nr = DIV_ROUND_UP(gr->tpc_total, 4);
-	u32 dist[TPC_MAX / 4] = {};
-	u32 gpcs[GPC_MAX] = {};
-	u8  tpcnr[GPC_MAX];
-	int tpc, gpc, i;
-
-	memcpy(tpcnr, gr->tpc_nr, sizeof(gr->tpc_nr));
-
-	/* won't result in the same distribution as the binary driver where
-	 * some of the gpcs have more tpcs than others, but this shall do
-	 * for the moment.  the code for earlier gpus has this issue too.
-	 */
-	for (gpc = -1, i = 0; i < gr->tpc_total; i++) {
-		do {
-			gpc = (gpc + 1) % gr->gpc_nr;
-		} while(!tpcnr[gpc]);
-		tpc = gr->tpc_nr[gpc] - tpcnr[gpc]--;
-
-		dist[i / 4] |= ((gpc << 4) | tpc) << ((i % 4) * 8);
-		gpcs[gpc] |= i << (tpc * 8);
-	}
-
-	for (i = 0; i < dist_nr; i++)
-		nvkm_wr32(device, 0x405b60 + (i * 4), dist[i]);
-	for (i = 0; i < gr->gpc_nr; i++)
-		nvkm_wr32(device, 0x405ba0 + (i * 4), gpcs[i]);
-}
-
-void
-gm204_grctx_generate_main(struct gf100_gr *gr, struct gf100_grctx *info)
-{
-	struct nvkm_device *device = gr->base.engine.subdev.device;
-	const struct gf100_grctx_func *grctx = gr->func->grctx;
-	u32 tmp;
-	int i;
-
-	gf100_gr_mmio(gr, grctx->hub);
-	gf100_gr_mmio(gr, grctx->gpc);
-	gf100_gr_mmio(gr, grctx->zcull);
-	gf100_gr_mmio(gr, grctx->tpc);
-	gf100_gr_mmio(gr, grctx->ppc);
-
-	nvkm_wr32(device, 0x404154, 0x00000000);
-
-	grctx->bundle(info);
-	grctx->pagepool(info);
-	grctx->attrib(info);
-	grctx->unkn(gr);
-
-	gm204_grctx_generate_tpcid(gr);
-	gf100_grctx_generate_r406028(gr);
-	gk104_grctx_generate_r418bb8(gr);
-
-	for (i = 0; i < 8; i++)
-		nvkm_wr32(device, 0x4064d0 + (i * 0x04), 0x00000000);
-	nvkm_wr32(device, 0x406500, 0x00000000);
-
-	nvkm_wr32(device, 0x405b00, (gr->tpc_total << 8) | gr->gpc_nr);
-
-	gm204_grctx_generate_rop_active_fbps(gr);
-
-	for (tmp = 0, i = 0; i < gr->gpc_nr; i++)
-		tmp |= ((1 << gr->tpc_nr[i]) - 1) << (i * 4);
-	nvkm_wr32(device, 0x4041c4, tmp);
-
-	gm204_grctx_generate_405b60(gr);
-
-	gf100_gr_icmd(gr, grctx->icmd);
-	nvkm_wr32(device, 0x404154, 0x00000800);
-	gf100_gr_mthd(gr, grctx->mthd);
-
-	nvkm_mask(device, 0x418e94, 0xffffffff, 0xc4230000);
-	nvkm_mask(device, 0x418e4c, 0xffffffff, 0x70000000);
-}
-
-const struct gf100_grctx_func
-gm204_grctx = {
-	.main  = gm204_grctx_generate_main,
-	.unkn  = gk104_grctx_generate_unkn,
-	.hub   = gm204_grctx_pack_hub,
-	.gpc   = gm204_grctx_pack_gpc,
-	.zcull = gf100_grctx_pack_zcull,
-	.tpc   = gm204_grctx_pack_tpc,
-	.ppc   = gm204_grctx_pack_ppc,
-	.icmd  = gm204_grctx_pack_icmd,
-	.mthd  = gm204_grctx_pack_mthd,
-	.bundle = gm107_grctx_generate_bundle,
-	.bundle_size = 0x3000,
-	.bundle_min_gpm_fifo_depth = 0x180,
-	.bundle_token_limit = 0x780,
-	.pagepool = gm107_grctx_generate_pagepool,
-	.pagepool_size = 0x20000,
-	.attrib = gm107_grctx_generate_attrib,
-	.attrib_nr_max = 0x600,
-	.attrib_nr = 0x400,
-	.alpha_nr_max = 0x1800,
-	.alpha_nr = 0x1000,
-};
diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/gr/ctxgm206.c b/drivers/gpu/drm/nouveau/nvkm/engine/gr/ctxgm206.c
deleted file mode 100644
index d6be6034c2c2..000000000000
--- a/drivers/gpu/drm/nouveau/nvkm/engine/gr/ctxgm206.c
+++ /dev/null
@@ -1,74 +0,0 @@
-/*
- * Copyright 2015 Red Hat Inc.
- *
- * Permission is hereby granted, free of charge, to any person obtaining a
- * copy of this software and associated documentation files (the "Software"),
- * to deal in the Software without restriction, including without limitation
- * the rights to use, copy, modify, merge, publish, distribute, sublicense,
- * and/or sell copies of the Software, and to permit persons to whom the
- * Software is furnished to do so, subject to the following conditions:
- *
- * The above copyright notice and this permission notice shall be included in
- * all copies or substantial portions of the Software.
- *
- * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
- * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
- * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
- * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR
- * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
- * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
- * OTHER DEALINGS IN THE SOFTWARE.
- *
- * Authors: Ben Skeggs <bskeggs@redhat.com>
- */
-#include "ctxgf100.h"
-
-static const struct gf100_gr_init
-gm206_grctx_init_gpc_unk_1[] = {
-	{ 0x418600,   1, 0x04, 0x0000007f },
-	{ 0x418684,   1, 0x04, 0x0000001f },
-	{ 0x418700,   1, 0x04, 0x00000002 },
-	{ 0x418704,   1, 0x04, 0x00000080 },
-	{ 0x418708,   1, 0x04, 0x40000000 },
-	{ 0x41870c,   2, 0x04, 0x00000000 },
-	{ 0x418728,   1, 0x04, 0x00300020 },
-	{}
-};
-
-static const struct gf100_gr_pack
-gm206_grctx_pack_gpc[] = {
-	{ gm107_grctx_init_gpc_unk_0 },
-	{ gm204_grctx_init_prop_0 },
-	{ gm206_grctx_init_gpc_unk_1 },
-	{ gm204_grctx_init_setup_0 },
-	{ gf100_grctx_init_zcull_0 },
-	{ gk208_grctx_init_crstr_0 },
-	{ gm204_grctx_init_gpm_0 },
-	{ gm204_grctx_init_gpc_unk_2 },
-	{ gf100_grctx_init_gcc_0 },
-	{}
-};
-
-const struct gf100_grctx_func
-gm206_grctx = {
-	.main  = gm204_grctx_generate_main,
-	.unkn  = gk104_grctx_generate_unkn,
-	.hub   = gm204_grctx_pack_hub,
-	.gpc   = gm206_grctx_pack_gpc,
-	.zcull = gf100_grctx_pack_zcull,
-	.tpc   = gm204_grctx_pack_tpc,
-	.ppc   = gm204_grctx_pack_ppc,
-	.icmd  = gm204_grctx_pack_icmd,
-	.mthd  = gm204_grctx_pack_mthd,
-	.bundle = gm107_grctx_generate_bundle,
-	.bundle_size = 0x3000,
-	.bundle_min_gpm_fifo_depth = 0x180,
-	.bundle_token_limit = 0x780,
-	.pagepool = gm107_grctx_generate_pagepool,
-	.pagepool_size = 0x20000,
-	.attrib = gm107_grctx_generate_attrib,
-	.attrib_nr_max = 0x600,
-	.attrib_nr = 0x400,
-	.alpha_nr_max = 0x1800,
-	.alpha_nr = 0x1000,
-};
diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/gr/ctxgm20b.c b/drivers/gpu/drm/nouveau/nvkm/engine/gr/ctxgm20b.c
index 670260402538..a8827efa90ae 100644
--- a/drivers/gpu/drm/nouveau/nvkm/engine/gr/ctxgm20b.c
+++ b/drivers/gpu/drm/nouveau/nvkm/engine/gr/ctxgm20b.c
@@ -54,7 +54,7 @@ gm20b_grctx_generate_main(struct gf100_gr *gr, struct gf100_grctx *info)
 
 	grctx->unkn(gr);
 
-	gm204_grctx_generate_tpcid(gr);
+	gm200_grctx_generate_tpcid(gr);
 	gm20b_grctx_generate_r406028(gr);
 	gk104_grctx_generate_r418bb8(gr);
 
@@ -70,7 +70,7 @@ gm20b_grctx_generate_main(struct gf100_gr *gr, struct gf100_grctx *info)
 		tmp |= ((1 << gr->tpc_nr[i]) - 1) << (i * 4);
 	nvkm_wr32(device, 0x4041c4, tmp);
 
-	gm204_grctx_generate_405b60(gr);
+	gm200_grctx_generate_405b60(gr);
 
 	gf100_gr_wait_idle(gr);
 
diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/gr/fuc/gpc.fuc b/drivers/gpu/drm/nouveau/nvkm/engine/gr/fuc/gpc.fuc
index e168b83a10c9..dc60509f76f7 100644
--- a/drivers/gpu/drm/nouveau/nvkm/engine/gr/fuc/gpc.fuc
+++ b/drivers/gpu/drm/nouveau/nvkm/engine/gr/fuc/gpc.fuc
@@ -322,6 +322,7 @@ main:
 
 // interrupt handler
 ih:
+	push $r0
 	push $r8
 	mov $r8 $flags
 	push $r8
@@ -358,6 +359,7 @@ ih:
 	pop $r8
 	mov $flags $r8
 	pop $r8
+	pop $r0
 	bclr $flags $p0
 	iret
 
diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/gr/fuc/gpcgf100.fuc3.h b/drivers/gpu/drm/nouveau/nvkm/engine/gr/fuc/gpcgf100.fuc3.h
index 231f696d1e0a..5f4ddfee48a2 100644
--- a/drivers/gpu/drm/nouveau/nvkm/engine/gr/fuc/gpcgf100.fuc3.h
+++ b/drivers/gpu/drm/nouveau/nvkm/engine/gr/fuc/gpcgf100.fuc3.h
@@ -382,56 +382,57 @@ uint32_t gf100_grgpc_code[] = {
 	0xb60412fd,
 	0x1efd01e4,
 	0x0018fe05,
-	0x05b021f5,
+	0x05b421f5,
 /* 0x04eb: main_not_ctx_xfer */
 	0x94d30ef4,
 	0xf5f010ef,
 	0x7e21f501,
 	0xc60ef403,
 /* 0x04f8: ih */
-	0x88fe80f9,
-	0xf980f901,
-	0xf9a0f990,
-	0xf9d0f9b0,
-	0xbdf0f9e0,
-	0x00a7f104,
-	0x00a3f002,
-	0xc400aacf,
-	0x0bf404ab,
-	0x1cd7f02c,
-	0x1a00e7f1,
-	0xcf00e3f0,
-	0xf7f100ee,
-	0xf3f01900,
-	0x00ffcf00,
-	0xf00421f4,
-	0x07f101e7,
-	0x03f01d00,
-	0x000ed000,
-/* 0x0546: ih_no_fifo */
-	0x07f104bd,
-	0x03f00100,
-	0x000ad000,
-	0xf0fc04bd,
-	0xd0fce0fc,
-	0xa0fcb0fc,
-	0x80fc90fc,
-	0xfc0088fe,
-	0x0032f480,
-/* 0x056a: hub_barrier_done */
+	0x80f900f9,
+	0xf90188fe,
+	0xf990f980,
+	0xf9b0f9a0,
+	0xf9e0f9d0,
+	0xf104bdf0,
+	0xf00200a7,
+	0xaacf00a3,
+	0x04abc400,
+	0xf02c0bf4,
+	0xe7f11cd7,
+	0xe3f01a00,
+	0x00eecf00,
+	0x1900f7f1,
+	0xcf00f3f0,
+	0x21f400ff,
+	0x01e7f004,
+	0x1d0007f1,
+	0xd00003f0,
+	0x04bd000e,
+/* 0x0548: ih_no_fifo */
+	0x010007f1,
+	0xd00003f0,
+	0x04bd000a,
+	0xe0fcf0fc,
+	0xb0fcd0fc,
+	0x90fca0fc,
+	0x88fe80fc,
+	0xfc80fc00,
+	0x0032f400,
+/* 0x056e: hub_barrier_done */
 	0xf7f001f8,
 	0x040e9801,
 	0xb904febb,
 	0xe7f102ff,
 	0xe3f09418,
 	0x9d21f440,
-/* 0x0582: ctx_redswitch */
+/* 0x0586: ctx_redswitch */
 	0xf7f000f8,
 	0x0007f120,
 	0x0103f085,
 	0xbd000fd0,
 	0x08e7f004,
-/* 0x0594: ctx_redswitch_delay */
+/* 0x0598: ctx_redswitch_delay */
 	0xf401e2b6,
 	0xf5f1fd1b,
 	0xf5f10800,
@@ -439,13 +440,13 @@ uint32_t gf100_grgpc_code[] = {
 	0x03f08500,
 	0x000fd001,
 	0x00f804bd,
-/* 0x05b0: ctx_xfer */
+/* 0x05b4: ctx_xfer */
 	0x810007f1,
 	0xd00203f0,
 	0x04bd000f,
 	0xf50711f4,
-/* 0x05c3: ctx_xfer_not_load */
-	0xf5058221,
+/* 0x05c7: ctx_xfer_not_load */
+	0xf5058621,
 	0xbd026a21,
 	0xfc07f124,
 	0x0203f047,
@@ -475,12 +476,11 @@ uint32_t gf100_grgpc_code[] = {
 	0x6f21f508,
 	0x5e21f501,
 	0x0601f402,
-/* 0x063b: ctx_xfer_post */
+/* 0x063f: ctx_xfer_post */
 	0xf50712f4,
-/* 0x063f: ctx_xfer_done */
+/* 0x0643: ctx_xfer_done */
 	0xf5027f21,
-	0xf8056a21,
-	0x00000000,
+	0xf8056e21,
 	0x00000000,
 	0x00000000,
 	0x00000000,
diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/gr/fuc/gpcgf117.fuc3.h b/drivers/gpu/drm/nouveau/nvkm/engine/gr/fuc/gpcgf117.fuc3.h
index bb820ff28621..03381b163cfc 100644
--- a/drivers/gpu/drm/nouveau/nvkm/engine/gr/fuc/gpcgf117.fuc3.h
+++ b/drivers/gpu/drm/nouveau/nvkm/engine/gr/fuc/gpcgf117.fuc3.h
@@ -408,56 +408,57 @@ uint32_t gf117_grgpc_code[] = {
 	0x0412fd20,
 	0xfd01e4b6,
 	0x18fe051e,
-	0xfd21f500,
-	0xd30ef405,
+	0x0121f500,
+	0xd30ef406,
 /* 0x0538: main_not_ctx_xfer */
 	0xf010ef94,
 	0x21f501f5,
 	0x0ef4037e,
 /* 0x0545: ih */
-	0xfe80f9c6,
-	0x80f90188,
-	0xa0f990f9,
-	0xd0f9b0f9,
-	0xf0f9e0f9,
-	0xa7f104bd,
-	0xa3f00200,
-	0x00aacf00,
-	0xf404abc4,
-	0xd7f02c0b,
-	0x00e7f124,
-	0x00e3f01a,
-	0xf100eecf,
-	0xf01900f7,
-	0xffcf00f3,
-	0x0421f400,
-	0xf101e7f0,
-	0xf01d0007,
-	0x0ed00003,
-/* 0x0593: ih_no_fifo */
-	0xf104bd00,
-	0xf0010007,
-	0x0ad00003,
-	0xfc04bd00,
-	0xfce0fcf0,
-	0xfcb0fcd0,
-	0xfc90fca0,
-	0x0088fe80,
-	0x32f480fc,
-/* 0x05b7: hub_barrier_done */
+	0xf900f9c6,
+	0x0188fe80,
+	0x90f980f9,
+	0xb0f9a0f9,
+	0xe0f9d0f9,
+	0x04bdf0f9,
+	0x0200a7f1,
+	0xcf00a3f0,
+	0xabc400aa,
+	0x2c0bf404,
+	0xf124d7f0,
+	0xf01a00e7,
+	0xeecf00e3,
+	0x00f7f100,
+	0x00f3f019,
+	0xf400ffcf,
+	0xe7f00421,
+	0x0007f101,
+	0x0003f01d,
+	0xbd000ed0,
+/* 0x0595: ih_no_fifo */
+	0x0007f104,
+	0x0003f001,
+	0xbd000ad0,
+	0xfcf0fc04,
+	0xfcd0fce0,
+	0xfca0fcb0,
+	0xfe80fc90,
+	0x80fc0088,
+	0x32f400fc,
+/* 0x05bb: hub_barrier_done */
 	0xf001f800,
 	0x0e9801f7,
 	0x04febb04,
 	0xf102ffb9,
 	0xf09418e7,
 	0x21f440e3,
-/* 0x05cf: ctx_redswitch */
+/* 0x05d3: ctx_redswitch */
 	0xf000f89d,
 	0x07f120f7,
 	0x03f08500,
 	0x000fd001,
 	0xe7f004bd,
-/* 0x05e1: ctx_redswitch_delay */
+/* 0x05e5: ctx_redswitch_delay */
 	0x01e2b608,
 	0xf1fd1bf4,
 	0xf10800f5,
@@ -465,13 +466,13 @@ uint32_t gf117_grgpc_code[] = {
 	0xf0850007,
 	0x0fd00103,
 	0xf804bd00,
-/* 0x05fd: ctx_xfer */
+/* 0x0601: ctx_xfer */
 	0x0007f100,
 	0x0203f081,
 	0xbd000fd0,
 	0x0711f404,
-	0x05cf21f5,
-/* 0x0610: ctx_xfer_not_load */
+	0x05d321f5,
+/* 0x0614: ctx_xfer_not_load */
 	0x026a21f5,
 	0x07f124bd,
 	0x03f047fc,
@@ -511,10 +512,10 @@ uint32_t gf117_grgpc_code[] = {
 	0x21f5016f,
 	0x01f4025e,
 	0x0712f406,
-/* 0x06ac: ctx_xfer_post */
+/* 0x06b0: ctx_xfer_post */
 	0x027f21f5,
-/* 0x06b0: ctx_xfer_done */
-	0x05b721f5,
+/* 0x06b4: ctx_xfer_done */
+	0x05bb21f5,
 	0x000000f8,
 	0x00000000,
 	0x00000000,
@@ -533,5 +534,4 @@ uint32_t gf117_grgpc_code[] = {
 	0x00000000,
 	0x00000000,
 	0x00000000,
-	0x00000000,
 };
diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/gr/fuc/gpcgk104.fuc3.h b/drivers/gpu/drm/nouveau/nvkm/engine/gr/fuc/gpcgk104.fuc3.h
index 911976d20940..99d9b48a3b50 100644
--- a/drivers/gpu/drm/nouveau/nvkm/engine/gr/fuc/gpcgk104.fuc3.h
+++ b/drivers/gpu/drm/nouveau/nvkm/engine/gr/fuc/gpcgk104.fuc3.h
@@ -408,56 +408,57 @@ uint32_t gk104_grgpc_code[] = {
 	0x0412fd20,
 	0xfd01e4b6,
 	0x18fe051e,
-	0xfd21f500,
-	0xd30ef405,
+	0x0121f500,
+	0xd30ef406,
 /* 0x0538: main_not_ctx_xfer */
 	0xf010ef94,
 	0x21f501f5,
 	0x0ef4037e,
 /* 0x0545: ih */
-	0xfe80f9c6,
-	0x80f90188,
-	0xa0f990f9,
-	0xd0f9b0f9,
-	0xf0f9e0f9,
-	0xa7f104bd,
-	0xa3f00200,
-	0x00aacf00,
-	0xf404abc4,
-	0xd7f02c0b,
-	0x00e7f124,
-	0x00e3f01a,
-	0xf100eecf,
-	0xf01900f7,
-	0xffcf00f3,
-	0x0421f400,
-	0xf101e7f0,
-	0xf01d0007,
-	0x0ed00003,
-/* 0x0593: ih_no_fifo */
-	0xf104bd00,
-	0xf0010007,
-	0x0ad00003,
-	0xfc04bd00,
-	0xfce0fcf0,
-	0xfcb0fcd0,
-	0xfc90fca0,
-	0x0088fe80,
-	0x32f480fc,
-/* 0x05b7: hub_barrier_done */
+	0xf900f9c6,
+	0x0188fe80,
+	0x90f980f9,
+	0xb0f9a0f9,
+	0xe0f9d0f9,
+	0x04bdf0f9,
+	0x0200a7f1,
+	0xcf00a3f0,
+	0xabc400aa,
+	0x2c0bf404,
+	0xf124d7f0,
+	0xf01a00e7,
+	0xeecf00e3,
+	0x00f7f100,
+	0x00f3f019,
+	0xf400ffcf,
+	0xe7f00421,
+	0x0007f101,
+	0x0003f01d,
+	0xbd000ed0,
+/* 0x0595: ih_no_fifo */
+	0x0007f104,
+	0x0003f001,
+	0xbd000ad0,
+	0xfcf0fc04,
+	0xfcd0fce0,
+	0xfca0fcb0,
+	0xfe80fc90,
+	0x80fc0088,
+	0x32f400fc,
+/* 0x05bb: hub_barrier_done */
 	0xf001f800,
 	0x0e9801f7,
 	0x04febb04,
 	0xf102ffb9,
 	0xf09418e7,
 	0x21f440e3,
-/* 0x05cf: ctx_redswitch */
+/* 0x05d3: ctx_redswitch */
 	0xf000f89d,
 	0x07f120f7,
 	0x03f08500,
 	0x000fd001,
 	0xe7f004bd,
-/* 0x05e1: ctx_redswitch_delay */
+/* 0x05e5: ctx_redswitch_delay */
 	0x01e2b608,
 	0xf1fd1bf4,
 	0xf10800f5,
@@ -465,13 +466,13 @@ uint32_t gk104_grgpc_code[] = {
 	0xf0850007,
 	0x0fd00103,
 	0xf804bd00,
-/* 0x05fd: ctx_xfer */
+/* 0x0601: ctx_xfer */
 	0x0007f100,
 	0x0203f081,
 	0xbd000fd0,
 	0x0711f404,
-	0x05cf21f5,
-/* 0x0610: ctx_xfer_not_load */
+	0x05d321f5,
+/* 0x0614: ctx_xfer_not_load */
 	0x026a21f5,
 	0x07f124bd,
 	0x03f047fc,
@@ -511,10 +512,10 @@ uint32_t gk104_grgpc_code[] = {
 	0x21f5016f,
 	0x01f4025e,
 	0x0712f406,
-/* 0x06ac: ctx_xfer_post */
+/* 0x06b0: ctx_xfer_post */
 	0x027f21f5,
-/* 0x06b0: ctx_xfer_done */
-	0x05b721f5,
+/* 0x06b4: ctx_xfer_done */
+	0x05bb21f5,
 	0x000000f8,
 	0x00000000,
 	0x00000000,
@@ -533,5 +534,4 @@ uint32_t gk104_grgpc_code[] = {
 	0x00000000,
 	0x00000000,
 	0x00000000,
-	0x00000000,
 };
diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/gr/fuc/gpcgk110.fuc3.h b/drivers/gpu/drm/nouveau/nvkm/engine/gr/fuc/gpcgk110.fuc3.h
index 1c6e11b05df2..f7267696cbfd 100644
--- a/drivers/gpu/drm/nouveau/nvkm/engine/gr/fuc/gpcgk110.fuc3.h
+++ b/drivers/gpu/drm/nouveau/nvkm/engine/gr/fuc/gpcgk110.fuc3.h
@@ -408,56 +408,57 @@ uint32_t gk110_grgpc_code[] = {
 	0x0412fd20,
 	0xfd01e4b6,
 	0x18fe051e,
-	0xfd21f500,
-	0xd30ef405,
+	0x0121f500,
+	0xd30ef406,
 /* 0x0538: main_not_ctx_xfer */
 	0xf010ef94,
 	0x21f501f5,
 	0x0ef4037e,
 /* 0x0545: ih */
-	0xfe80f9c6,
-	0x80f90188,
-	0xa0f990f9,
-	0xd0f9b0f9,
-	0xf0f9e0f9,
-	0xa7f104bd,
-	0xa3f00200,
-	0x00aacf00,
-	0xf404abc4,
-	0xd7f02c0b,
-	0x00e7f124,
-	0x00e3f01a,
-	0xf100eecf,
-	0xf01900f7,
-	0xffcf00f3,
-	0x0421f400,
-	0xf101e7f0,
-	0xf01d0007,
-	0x0ed00003,
-/* 0x0593: ih_no_fifo */
-	0xf104bd00,
-	0xf0010007,
-	0x0ad00003,
-	0xfc04bd00,
-	0xfce0fcf0,
-	0xfcb0fcd0,
-	0xfc90fca0,
-	0x0088fe80,
-	0x32f480fc,
-/* 0x05b7: hub_barrier_done */
+	0xf900f9c6,
+	0x0188fe80,
+	0x90f980f9,
+	0xb0f9a0f9,
+	0xe0f9d0f9,
+	0x04bdf0f9,
+	0x0200a7f1,
+	0xcf00a3f0,
+	0xabc400aa,
+	0x2c0bf404,
+	0xf124d7f0,
+	0xf01a00e7,
+	0xeecf00e3,
+	0x00f7f100,
+	0x00f3f019,
+	0xf400ffcf,
+	0xe7f00421,
+	0x0007f101,
+	0x0003f01d,
+	0xbd000ed0,
+/* 0x0595: ih_no_fifo */
+	0x0007f104,
+	0x0003f001,
+	0xbd000ad0,
+	0xfcf0fc04,
+	0xfcd0fce0,
+	0xfca0fcb0,
+	0xfe80fc90,
+	0x80fc0088,
+	0x32f400fc,
+/* 0x05bb: hub_barrier_done */
 	0xf001f800,
 	0x0e9801f7,
 	0x04febb04,
 	0xf102ffb9,
 	0xf09418e7,
 	0x21f440e3,
-/* 0x05cf: ctx_redswitch */
+/* 0x05d3: ctx_redswitch */
 	0xf000f89d,
 	0x07f120f7,
 	0x03f08500,
 	0x000fd001,
 	0xe7f004bd,
-/* 0x05e1: ctx_redswitch_delay */
+/* 0x05e5: ctx_redswitch_delay */
 	0x01e2b608,
 	0xf1fd1bf4,
 	0xf10800f5,
@@ -465,13 +466,13 @@ uint32_t gk110_grgpc_code[] = {
 	0xf0850007,
 	0x0fd00103,
 	0xf804bd00,
-/* 0x05fd: ctx_xfer */
+/* 0x0601: ctx_xfer */
 	0x0007f100,
 	0x0203f081,
 	0xbd000fd0,
 	0x0711f404,
-	0x05cf21f5,
-/* 0x0610: ctx_xfer_not_load */
+	0x05d321f5,
+/* 0x0614: ctx_xfer_not_load */
 	0x026a21f5,
 	0x07f124bd,
 	0x03f047fc,
@@ -511,10 +512,10 @@ uint32_t gk110_grgpc_code[] = {
 	0x21f5016f,
 	0x01f4025e,
 	0x0712f406,
-/* 0x06ac: ctx_xfer_post */
+/* 0x06b0: ctx_xfer_post */
 	0x027f21f5,
-/* 0x06b0: ctx_xfer_done */
-	0x05b721f5,
+/* 0x06b4: ctx_xfer_done */
+	0x05bb21f5,
 	0x000000f8,
 	0x00000000,
 	0x00000000,
@@ -533,5 +534,4 @@ uint32_t gk110_grgpc_code[] = {
 	0x00000000,
 	0x00000000,
 	0x00000000,
-	0x00000000,
 };
diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/gr/fuc/gpcgk208.fuc5.h b/drivers/gpu/drm/nouveau/nvkm/engine/gr/fuc/gpcgk208.fuc5.h
index 84af7ec6a78e..387d1fa3e231 100644
--- a/drivers/gpu/drm/nouveau/nvkm/engine/gr/fuc/gpcgk208.fuc5.h
+++ b/drivers/gpu/drm/nouveau/nvkm/engine/gr/fuc/gpcgk208.fuc5.h
@@ -360,61 +360,62 @@ uint32_t gk208_grgpc_code[] = {
 	0xb60412fd,
 	0x1efd01e4,
 	0x0018fe05,
-	0x00051b7e,
+	0x00051f7e,
 /* 0x0477: main_not_ctx_xfer */
 	0x94d40ef4,
 	0xf5f010ef,
 	0x02f87e01,
 	0xc70ef400,
 /* 0x0484: ih */
-	0x88fe80f9,
-	0xf980f901,
-	0xf9a0f990,
-	0xf9d0f9b0,
-	0xbdf0f9e0,
-	0x02004a04,
-	0xc400aacf,
-	0x0bf404ab,
-	0x4e240d1f,
-	0xeecf1a00,
-	0x19004f00,
-	0x7e00ffcf,
-	0x0e000004,
-	0x1d004001,
-	0xbd000ef6,
-/* 0x04c1: ih_no_fifo */
-	0x01004004,
-	0xbd000af6,
-	0xfcf0fc04,
-	0xfcd0fce0,
-	0xfca0fcb0,
-	0xfe80fc90,
-	0x80fc0088,
+	0x80f900f9,
+	0xf90188fe,
+	0xf990f980,
+	0xf9b0f9a0,
+	0xf9e0f9d0,
+	0x4a04bdf0,
+	0xaacf0200,
+	0x04abc400,
+	0x0d1f0bf4,
+	0x1a004e24,
+	0x4f00eecf,
+	0xffcf1900,
+	0x00047e00,
+	0x40010e00,
+	0x0ef61d00,
+/* 0x04c3: ih_no_fifo */
+	0x4004bd00,
+	0x0af60100,
+	0xfc04bd00,
+	0xfce0fcf0,
+	0xfcb0fcd0,
+	0xfc90fca0,
+	0x0088fe80,
+	0x00fc80fc,
 	0xf80032f4,
-/* 0x04e1: hub_barrier_done */
+/* 0x04e5: hub_barrier_done */
 	0x98010f01,
 	0xfebb040e,
 	0x8effb204,
 	0x7e409418,
 	0xf800008f,
-/* 0x04f5: ctx_redswitch */
+/* 0x04f9: ctx_redswitch */
 	0x80200f00,
 	0xf6018500,
 	0x04bd000f,
-/* 0x0502: ctx_redswitch_delay */
+/* 0x0506: ctx_redswitch_delay */
 	0xe2b6080e,
 	0xfd1bf401,
 	0x0800f5f1,
 	0x0200f5f1,
 	0x01850080,
 	0xbd000ff6,
-/* 0x051b: ctx_xfer */
+/* 0x051f: ctx_xfer */
 	0x8000f804,
 	0xf6028100,
 	0x04bd000f,
 	0x7e0711f4,
-/* 0x052b: ctx_xfer_not_load */
-	0x7e0004f5,
+/* 0x052f: ctx_xfer_not_load */
+	0x7e0004f9,
 	0xbd000216,
 	0x47fc8024,
 	0x0002f602,
@@ -449,10 +450,10 @@ uint32_t gk208_grgpc_code[] = {
 	0x7e00013d,
 	0xf400020a,
 	0x12f40601,
-/* 0x05b5: ctx_xfer_post */
+/* 0x05b9: ctx_xfer_post */
 	0x02277e07,
-/* 0x05b9: ctx_xfer_done */
-	0x04e17e00,
+/* 0x05bd: ctx_xfer_done */
+	0x04e57e00,
 	0x0000f800,
 	0x00000000,
 	0x00000000,
@@ -469,5 +470,4 @@ uint32_t gk208_grgpc_code[] = {
 	0x00000000,
 	0x00000000,
 	0x00000000,
-	0x00000000,
 };
diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/gr/fuc/gpcgm107.fuc5.h b/drivers/gpu/drm/nouveau/nvkm/engine/gr/fuc/gpcgm107.fuc5.h
index 5136f9161706..fa9f3c0c5994 100644
--- a/drivers/gpu/drm/nouveau/nvkm/engine/gr/fuc/gpcgm107.fuc5.h
+++ b/drivers/gpu/drm/nouveau/nvkm/engine/gr/fuc/gpcgm107.fuc5.h
@@ -438,48 +438,49 @@ uint32_t gm107_grgpc_code[] = {
 	0x0412fd20,
 	0xfd01e4b6,
 	0x18fe051e,
-	0x06447e00,
+	0x06487e00,
 	0xd40ef400,
 /* 0x05a0: main_not_ctx_xfer */
 	0xf010ef94,
 	0xf87e01f5,
 	0x0ef40002,
 /* 0x05ad: ih */
-	0xfe80f9c7,
-	0x80f90188,
-	0xa0f990f9,
-	0xd0f9b0f9,
-	0xf0f9e0f9,
-	0x004a04bd,
-	0x00aacf02,
-	0xf404abc4,
-	0x240d1f0b,
-	0xcf1a004e,
-	0x004f00ee,
-	0x00ffcf19,
-	0x0000047e,
-	0x0040010e,
-	0x000ef61d,
-/* 0x05ea: ih_no_fifo */
-	0x004004bd,
-	0x000af601,
-	0xf0fc04bd,
-	0xd0fce0fc,
-	0xa0fcb0fc,
-	0x80fc90fc,
-	0xfc0088fe,
-	0x0032f480,
-/* 0x060a: hub_barrier_done */
+	0xf900f9c7,
+	0x0188fe80,
+	0x90f980f9,
+	0xb0f9a0f9,
+	0xe0f9d0f9,
+	0x04bdf0f9,
+	0xcf02004a,
+	0xabc400aa,
+	0x1f0bf404,
+	0x004e240d,
+	0x00eecf1a,
+	0xcf19004f,
+	0x047e00ff,
+	0x010e0000,
+	0xf61d0040,
+	0x04bd000e,
+/* 0x05ec: ih_no_fifo */
+	0xf6010040,
+	0x04bd000a,
+	0xe0fcf0fc,
+	0xb0fcd0fc,
+	0x90fca0fc,
+	0x88fe80fc,
+	0xfc80fc00,
+	0x0032f400,
+/* 0x060e: hub_barrier_done */
 	0x010f01f8,
 	0xbb040e98,
 	0xffb204fe,
 	0x4094188e,
 	0x00008f7e,
-/* 0x061e: ctx_redswitch */
+/* 0x0622: ctx_redswitch */
 	0x200f00f8,
 	0x01850080,
 	0xbd000ff6,
-/* 0x062b: ctx_redswitch_delay */
+/* 0x062f: ctx_redswitch_delay */
 	0xb6080e04,
 	0x1bf401e2,
 	0x00f5f1fd,
@@ -487,15 +488,15 @@ uint32_t gm107_grgpc_code[] = {
 	0x85008002,
 	0x000ff601,
 	0x00f804bd,
-/* 0x0644: ctx_xfer */
+/* 0x0648: ctx_xfer */
 	0x02810080,
 	0xbd000ff6,
 	0x1dc48e04,
 	0x01e5f050,
 	0x8f7effb2,
 	0x11f40000,
-	0x061e7e07,
-/* 0x0661: ctx_xfer_not_load */
+	0x06227e07,
+/* 0x0665: ctx_xfer_not_load */
 	0x02167e00,
 	0x8024bd00,
 	0xf60247fc,
@@ -550,15 +551,15 @@ uint32_t gm107_grgpc_code[] = {
 	0x7e00020a,
 	0xf4000314,
 	0x12f40601,
-/* 0x0739: ctx_xfer_post */
+/* 0x073d: ctx_xfer_post */
 	0x02277e1a,
 	0x8e0d0f00,
 	0xf0501da8,
 	0xffb201e5,
 	0x00008f7e,
 	0x0003147e,
-/* 0x0750: ctx_xfer_done */
-	0x00060a7e,
+/* 0x0754: ctx_xfer_done */
+	0x00060e7e,
 	0x000000f8,
 	0x00000000,
 	0x00000000,
@@ -601,5 +602,4 @@ uint32_t gm107_grgpc_code[] = {
 	0x00000000,
 	0x00000000,
 	0x00000000,
-	0x00000000,
 };
diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/gr/fuc/hub.fuc b/drivers/gpu/drm/nouveau/nvkm/engine/gr/fuc/hub.fuc
index 87f99e38acbf..e3a2fb308271 100644
--- a/drivers/gpu/drm/nouveau/nvkm/engine/gr/fuc/hub.fuc
+++ b/drivers/gpu/drm/nouveau/nvkm/engine/gr/fuc/hub.fuc
@@ -306,6 +306,7 @@ main:
 
 // interrupt handler
 ih:
+	push $r0
 	push $r8
 	mov $r8 $flags
 	push $r8
@@ -380,6 +381,7 @@ ih:
 	pop $r8
 	mov $flags $r8
 	pop $r8
+	pop $r0
 	bclr $flags $p0
 	iret
 
diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/gr/fuc/hubgf100.fuc3.h b/drivers/gpu/drm/nouveau/nvkm/engine/gr/fuc/hubgf100.fuc3.h
index f6acda505677..397921a9a46c 100644
--- a/drivers/gpu/drm/nouveau/nvkm/engine/gr/fuc/hubgf100.fuc3.h
+++ b/drivers/gpu/drm/nouveau/nvkm/engine/gr/fuc/hubgf100.fuc3.h
@@ -528,10 +528,10 @@ uint32_t gf100_grhub_code[] = {
 	0x0001d001,
 	0x17f104bd,
 	0xf7f00100,
-	0x0d21f502,
-	0x1f21f508,
+	0x1121f502,
+	0x2321f508,
 	0x10f7f008,
-	0x086c21f5,
+	0x087021f5,
 	0x98000e98,
 	0x21f5010f,
 	0x14950150,
@@ -574,9 +574,9 @@ uint32_t gf100_grhub_code[] = {
 	0xb6800040,
 	0x1bf40132,
 	0x00f7f0be,
-	0x086c21f5,
+	0x087021f5,
 	0xf500f7f0,
-	0xf1080d21,
+	0xf1081121,
 	0xf0010007,
 	0x01d00203,
 	0xbd04bd00,
@@ -610,7 +610,7 @@ uint32_t gf100_grhub_code[] = {
 	0x09d00203,
 	0xf404bd00,
 	0x31f40132,
-	0x4021f502,
+	0x4421f502,
 	0xf094bd0a,
 	0x07f10799,
 	0x03f01700,
@@ -621,7 +621,7 @@ uint32_t gf100_grhub_code[] = {
 	0x0203f00f,
 	0xbd0009d0,
 	0x0131f404,
-	0x0a4021f5,
+	0x0a4421f5,
 	0x99f094bd,
 	0x0007f106,
 	0x0203f017,
@@ -631,7 +631,7 @@ uint32_t gf100_grhub_code[] = {
 	0x12b920f9,
 	0x0132f402,
 	0xf50232f4,
-	0xfc0a4021,
+	0xfc0a4421,
 	0x0007f120,
 	0x0203f0c0,
 	0xbd0002d0,
@@ -640,7 +640,7 @@ uint32_t gf100_grhub_code[] = {
 	0xf41f23c8,
 	0x31f40d0b,
 	0x0232f401,
-	0x0a4021f5,
+	0x0a4421f5,
 /* 0x063c: chsw_done */
 	0xf10127f0,
 	0xf0c30007,
@@ -654,7 +654,7 @@ uint32_t gf100_grhub_code[] = {
 /* 0x0660: main_not_ctx_switch */
 	0xf401e4b0,
 	0xf2b90d1b,
-	0xd021f502,
+	0xd421f502,
 	0x460ef409,
 /* 0x0670: main_not_ctx_chan */
 	0xf402e4b0,
@@ -664,7 +664,7 @@ uint32_t gf100_grhub_code[] = {
 	0x09d00203,
 	0xf404bd00,
 	0x32f40132,
-	0x4021f502,
+	0x4421f502,
 	0xf094bd0a,
 	0x07f10799,
 	0x03f01700,
@@ -682,107 +682,108 @@ uint32_t gf100_grhub_code[] = {
 	0x04bd0002,
 	0xfea00ef5,
 /* 0x06c8: ih */
-	0x88fe80f9,
-	0xf980f901,
-	0xf9a0f990,
-	0xf9d0f9b0,
-	0xbdf0f9e0,
-	0x00a7f104,
-	0x00a3f002,
-	0xc400aacf,
-	0x0bf404ab,
-	0x10d7f030,
-	0x1a00e7f1,
-	0xcf00e3f0,
-	0xf7f100ee,
-	0xf3f01900,
-	0x00ffcf00,
-	0xb70421f4,
-	0xf00400b0,
-	0x07f101e7,
-	0x03f01d00,
-	0x000ed000,
-/* 0x071a: ih_no_fifo */
-	0xabe404bd,
-	0x0bf40100,
-	0x10d7f00d,
-	0x4001e7f1,
-/* 0x072b: ih_no_ctxsw */
-	0xe40421f4,
-	0xf40400ab,
-	0xe7f16c0b,
-	0xe3f00708,
-	0x6821f440,
-	0xf102ffb9,
-	0xf0040007,
-	0x0fd00203,
-	0xf104bd00,
-	0xf00704e7,
+	0x80f900f9,
+	0xf90188fe,
+	0xf990f980,
+	0xf9b0f9a0,
+	0xf9e0f9d0,
+	0xf104bdf0,
+	0xf00200a7,
+	0xaacf00a3,
+	0x04abc400,
+	0xf0300bf4,
+	0xe7f110d7,
+	0xe3f01a00,
+	0x00eecf00,
+	0x1900f7f1,
+	0xcf00f3f0,
+	0x21f400ff,
+	0x00b0b704,
+	0x01e7f004,
+	0x1d0007f1,
+	0xd00003f0,
+	0x04bd000e,
+/* 0x071c: ih_no_fifo */
+	0x0100abe4,
+	0xf00d0bf4,
+	0xe7f110d7,
+	0x21f44001,
+/* 0x072d: ih_no_ctxsw */
+	0x00abe404,
+	0x6c0bf404,
+	0x0708e7f1,
+	0xf440e3f0,
+	0xffb96821,
+	0x0007f102,
+	0x0203f004,
+	0xbd000fd0,
+	0x04e7f104,
+	0x40e3f007,
+	0xb96821f4,
+	0x07f102ff,
+	0x03f00300,
+	0x000fd002,
+	0xfec704bd,
+	0x02ee9450,
+	0x0700f7f1,
+	0xbb40f3f0,
+	0x21f400ef,
+	0x0007f168,
+	0x0203f002,
+	0xbd000fd0,
+	0x03f7f004,
+	0x037e21f5,
+	0x0100b7f1,
+	0xf102bfb9,
+	0xf00144e7,
 	0x21f440e3,
-	0x02ffb968,
-	0x030007f1,
-	0xd00203f0,
-	0x04bd000f,
-	0x9450fec7,
-	0xf7f102ee,
-	0xf3f00700,
-	0x00efbb40,
-	0xf16821f4,
-	0xf0020007,
-	0x0fd00203,
-	0xf004bd00,
-	0x21f503f7,
-	0xb7f1037e,
-	0xbfb90100,
-	0x44e7f102,
-	0x40e3f001,
-/* 0x079b: ih_no_fwmthd */
-	0xf19d21f4,
-	0xbd0504b7,
-	0xb4abffb0,
-	0xf10f0bf4,
-	0xf0070007,
-	0x0bd00303,
-/* 0x07b3: ih_no_other */
-	0xf104bd00,
-	0xf0010007,
-	0x0ad00003,
-	0xfc04bd00,
-	0xfce0fcf0,
-	0xfcb0fcd0,
-	0xfc90fca0,
-	0x0088fe80,
-	0x32f480fc,
-/* 0x07d7: ctx_4160s */
+/* 0x079d: ih_no_fwmthd */
+	0x04b7f19d,
+	0xffb0bd05,
+	0x0bf4b4ab,
+	0x0007f10f,
+	0x0303f007,
+	0xbd000bd0,
+/* 0x07b5: ih_no_other */
+	0x0007f104,
+	0x0003f001,
+	0xbd000ad0,
+	0xfcf0fc04,
+	0xfcd0fce0,
+	0xfca0fcb0,
+	0xfe80fc90,
+	0x80fc0088,
+	0x32f400fc,
+/* 0x07db: ctx_4160s */
 	0xf001f800,
 	0xffb901f7,
 	0x60e7f102,
 	0x40e3f041,
-/* 0x07e7: ctx_4160s_wait */
+/* 0x07eb: ctx_4160s_wait */
 	0xf19d21f4,
 	0xf04160e7,
 	0x21f440e3,
 	0x02ffb968,
 	0xf404ffc8,
 	0x00f8f00b,
-/* 0x07fc: ctx_4160c */
+/* 0x0800: ctx_4160c */
 	0xffb9f4bd,
 	0x60e7f102,
 	0x40e3f041,
 	0xf89d21f4,
-/* 0x080d: ctx_4170s */
+/* 0x0811: ctx_4170s */
 	0x10f5f000,
 	0xf102ffb9,
 	0xf04170e7,
 	0x21f440e3,
-/* 0x081f: ctx_4170w */
+/* 0x0823: ctx_4170w */
 	0xf100f89d,
 	0xf04170e7,
 	0x21f440e3,
 	0x02ffb968,
 	0xf410f4f0,
 	0x00f8f01b,
-/* 0x0834: ctx_redswitch */
+/* 0x0838: ctx_redswitch */
 	0x0200e7f1,
 	0xf040e5f0,
 	0xe5f020e5,
@@ -790,7 +791,7 @@ uint32_t gf100_grhub_code[] = {
 	0x0103f085,
 	0xbd000ed0,
 	0x08f7f004,
-/* 0x0850: ctx_redswitch_delay */
+/* 0x0854: ctx_redswitch_delay */
 	0xf401f2b6,
 	0xe5f1fd1b,
 	0xe5f10400,
@@ -798,7 +799,7 @@ uint32_t gf100_grhub_code[] = {
 	0x03f08500,
 	0x000ed001,
 	0x00f804bd,
-/* 0x086c: ctx_86c */
+/* 0x0870: ctx_86c */
 	0x1b0007f1,
 	0xd00203f0,
 	0x04bd000f,
@@ -809,16 +810,16 @@ uint32_t gf100_grhub_code[] = {
 	0xa86ce7f1,
 	0xf441e3f0,
 	0x00f89d21,
-/* 0x0894: ctx_mem */
+/* 0x0898: ctx_mem */
 	0x840007f1,
 	0xd00203f0,
 	0x04bd000f,
-/* 0x08a0: ctx_mem_wait */
+/* 0x08a4: ctx_mem_wait */
 	0x8400f7f1,
 	0xcf02f3f0,
 	0xfffd00ff,
 	0xf31bf405,
-/* 0x08b2: ctx_load */
+/* 0x08b6: ctx_load */
 	0x94bd00f8,
 	0xf10599f0,
 	0xf00f0007,
@@ -836,7 +837,7 @@ uint32_t gf100_grhub_code[] = {
 	0x02d00203,
 	0xf004bd00,
 	0x21f507f7,
-	0x07f10894,
+	0x07f10898,
 	0x03f0c000,
 	0x0002d002,
 	0x0bfe04bd,
@@ -891,31 +892,31 @@ uint32_t gf100_grhub_code[] = {
 	0x03f01700,
 	0x0009d002,
 	0x00f804bd,
-/* 0x09d0: ctx_chan */
-	0x07d721f5,
-	0x08b221f5,
+/* 0x09d4: ctx_chan */
+	0x07db21f5,
+	0x08b621f5,
 	0xf40ca7f0,
 	0xf7f0d021,
-	0x9421f505,
-	0xfc21f508,
-/* 0x09eb: ctx_mmio_exec */
-	0x9800f807,
+	0x9821f505,
+	0x0021f508,
+/* 0x09ef: ctx_mmio_exec */
+	0x9800f808,
 	0x07f14103,
 	0x03f08100,
 	0x0003d002,
 	0x34bd04bd,
-/* 0x09fc: ctx_mmio_loop */
+/* 0x0a00: ctx_mmio_loop */
 	0xf4ff34c4,
 	0x57f10f1b,
 	0x53f00200,
 	0x0535fa06,
-/* 0x0a0e: ctx_mmio_pull */
+/* 0x0a12: ctx_mmio_pull */
 	0x4e9803f8,
 	0x814f9880,
 	0xb69d21f4,
 	0x12b60830,
 	0xdf1bf401,
-/* 0x0a20: ctx_mmio_done */
+/* 0x0a24: ctx_mmio_done */
 	0xf1160398,
 	0xf0810007,
 	0x03d00203,
@@ -924,30 +925,30 @@ uint32_t gf100_grhub_code[] = {
 	0x13f00100,
 	0x0601fa06,
 	0x00f803f8,
-/* 0x0a40: ctx_xfer */
+/* 0x0a44: ctx_xfer */
 	0xf104e7f0,
 	0xf0020007,
 	0x0ed00303,
-/* 0x0a4f: ctx_xfer_idle */
+/* 0x0a53: ctx_xfer_idle */
 	0xf104bd00,
 	0xf00000e7,
 	0xeecf03e3,
 	0x00e4f100,
 	0xf21bf420,
 	0xf40611f4,
-/* 0x0a66: ctx_xfer_pre */
+/* 0x0a6a: ctx_xfer_pre */
 	0xf7f01102,
-	0x6c21f510,
-	0xd721f508,
+	0x7021f510,
+	0xdb21f508,
 	0x1c11f407,
-/* 0x0a74: ctx_xfer_pre_load */
+/* 0x0a78: ctx_xfer_pre_load */
 	0xf502f7f0,
-	0xf5080d21,
-	0xf5081f21,
-	0xbd083421,
-	0x0d21f5f4,
-	0xb221f508,
-/* 0x0a8d: ctx_xfer_exec */
+	0xf5081121,
+	0xf5082321,
+	0xbd083821,
+	0x1121f5f4,
+	0xb621f508,
+/* 0x0a91: ctx_xfer_exec */
 	0x16019808,
 	0x07f124bd,
 	0x03f00500,
@@ -982,24 +983,23 @@ uint32_t gf100_grhub_code[] = {
 	0x1301f402,
 	0xf40ca7f0,
 	0xf7f0d021,
-	0x9421f505,
+	0x9821f505,
 	0x3202f408,
-/* 0x0b1c: ctx_xfer_post */
+/* 0x0b20: ctx_xfer_post */
 	0xf502f7f0,
-	0xbd080d21,
-	0x6c21f5f4,
+	0xbd081121,
+	0x7021f5f4,
 	0x7f21f508,
-	0x1f21f502,
+	0x2321f502,
 	0xf5f4bd08,
-	0xf4080d21,
+	0xf4081121,
 	0x01981011,
 	0x0511fd40,
 	0xf5070bf4,
-/* 0x0b47: ctx_xfer_no_post_mmio */
-	0xf509eb21,
-/* 0x0b4b: ctx_xfer_done */
-	0xf807fc21,
-	0x00000000,
+/* 0x0b4b: ctx_xfer_no_post_mmio */
+	0xf509ef21,
+/* 0x0b4f: ctx_xfer_done */
+	0xf8080021,
 	0x00000000,
 	0x00000000,
 	0x00000000,
diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/gr/fuc/hubgf117.fuc3.h b/drivers/gpu/drm/nouveau/nvkm/engine/gr/fuc/hubgf117.fuc3.h
index 7cb14e59dea1..50c97163dcdb 100644
--- a/drivers/gpu/drm/nouveau/nvkm/engine/gr/fuc/hubgf117.fuc3.h
+++ b/drivers/gpu/drm/nouveau/nvkm/engine/gr/fuc/hubgf117.fuc3.h
@@ -528,10 +528,10 @@ uint32_t gf117_grhub_code[] = {
 	0x0001d001,
 	0x17f104bd,
 	0xf7f00100,
-	0x0d21f502,
-	0x1f21f508,
+	0x1121f502,
+	0x2321f508,
 	0x10f7f008,
-	0x086c21f5,
+	0x087021f5,
 	0x98000e98,
 	0x21f5010f,
 	0x14950150,
@@ -574,9 +574,9 @@ uint32_t gf117_grhub_code[] = {
 	0xb6800040,
 	0x1bf40132,
 	0x00f7f0be,
-	0x086c21f5,
+	0x087021f5,
 	0xf500f7f0,
-	0xf1080d21,
+	0xf1081121,
 	0xf0010007,
 	0x01d00203,
 	0xbd04bd00,
@@ -610,7 +610,7 @@ uint32_t gf117_grhub_code[] = {
 	0x09d00203,
 	0xf404bd00,
 	0x31f40132,
-	0x4021f502,
+	0x4421f502,
 	0xf094bd0a,
 	0x07f10799,
 	0x03f01700,
@@ -621,7 +621,7 @@ uint32_t gf117_grhub_code[] = {
 	0x0203f00f,
 	0xbd0009d0,
 	0x0131f404,
-	0x0a4021f5,
+	0x0a4421f5,
 	0x99f094bd,
 	0x0007f106,
 	0x0203f017,
@@ -631,7 +631,7 @@ uint32_t gf117_grhub_code[] = {
 	0x12b920f9,
 	0x0132f402,
 	0xf50232f4,
-	0xfc0a4021,
+	0xfc0a4421,
 	0x0007f120,
 	0x0203f0c0,
 	0xbd0002d0,
@@ -640,7 +640,7 @@ uint32_t gf117_grhub_code[] = {
 	0xf41f23c8,
 	0x31f40d0b,
 	0x0232f401,
-	0x0a4021f5,
+	0x0a4421f5,
 /* 0x063c: chsw_done */
 	0xf10127f0,
 	0xf0c30007,
@@ -654,7 +654,7 @@ uint32_t gf117_grhub_code[] = {
 /* 0x0660: main_not_ctx_switch */
 	0xf401e4b0,
 	0xf2b90d1b,
-	0xd021f502,
+	0xd421f502,
 	0x460ef409,
 /* 0x0670: main_not_ctx_chan */
 	0xf402e4b0,
@@ -664,7 +664,7 @@ uint32_t gf117_grhub_code[] = {
 	0x09d00203,
 	0xf404bd00,
 	0x32f40132,
-	0x4021f502,
+	0x4421f502,
 	0xf094bd0a,
 	0x07f10799,
 	0x03f01700,
@@ -682,107 +682,108 @@ uint32_t gf117_grhub_code[] = {
 	0x04bd0002,
 	0xfea00ef5,
 /* 0x06c8: ih */
-	0x88fe80f9,
-	0xf980f901,
-	0xf9a0f990,
-	0xf9d0f9b0,
-	0xbdf0f9e0,
-	0x00a7f104,
-	0x00a3f002,
-	0xc400aacf,
-	0x0bf404ab,
-	0x10d7f030,
-	0x1a00e7f1,
-	0xcf00e3f0,
-	0xf7f100ee,
-	0xf3f01900,
-	0x00ffcf00,
-	0xb70421f4,
-	0xf00400b0,
-	0x07f101e7,
-	0x03f01d00,
-	0x000ed000,
-/* 0x071a: ih_no_fifo */
-	0xabe404bd,
-	0x0bf40100,
-	0x10d7f00d,
-	0x4001e7f1,
-/* 0x072b: ih_no_ctxsw */
-	0xe40421f4,
-	0xf40400ab,
-	0xe7f16c0b,
-	0xe3f00708,
-	0x6821f440,
-	0xf102ffb9,
-	0xf0040007,
-	0x0fd00203,
-	0xf104bd00,
-	0xf00704e7,
+	0x80f900f9,
+	0xf90188fe,
+	0xf990f980,
+	0xf9b0f9a0,
+	0xf9e0f9d0,
+	0xf104bdf0,
+	0xf00200a7,
+	0xaacf00a3,
+	0x04abc400,
+	0xf0300bf4,
+	0xe7f110d7,
+	0xe3f01a00,
+	0x00eecf00,
+	0x1900f7f1,
+	0xcf00f3f0,
+	0x21f400ff,
+	0x00b0b704,
+	0x01e7f004,
+	0x1d0007f1,
+	0xd00003f0,
+	0x04bd000e,
+/* 0x071c: ih_no_fifo */
+	0x0100abe4,
+	0xf00d0bf4,
+	0xe7f110d7,
+	0x21f44001,
+/* 0x072d: ih_no_ctxsw */
+	0x00abe404,
+	0x6c0bf404,
+	0x0708e7f1,
+	0xf440e3f0,
+	0xffb96821,
+	0x0007f102,
+	0x0203f004,
+	0xbd000fd0,
+	0x04e7f104,
+	0x40e3f007,
+	0xb96821f4,
+	0x07f102ff,
+	0x03f00300,
+	0x000fd002,
+	0xfec704bd,
+	0x02ee9450,
+	0x0700f7f1,
+	0xbb40f3f0,
+	0x21f400ef,
+	0x0007f168,
+	0x0203f002,
+	0xbd000fd0,
+	0x03f7f004,
+	0x037e21f5,
+	0x0100b7f1,
+	0xf102bfb9,
+	0xf00144e7,
 	0x21f440e3,
-	0x02ffb968,
-	0x030007f1,
-	0xd00203f0,
-	0x04bd000f,
-	0x9450fec7,
-	0xf7f102ee,
-	0xf3f00700,
-	0x00efbb40,
-	0xf16821f4,
-	0xf0020007,
-	0x0fd00203,
-	0xf004bd00,
-	0x21f503f7,
-	0xb7f1037e,
-	0xbfb90100,
-	0x44e7f102,
-	0x40e3f001,
-/* 0x079b: ih_no_fwmthd */
-	0xf19d21f4,
-	0xbd0504b7,
-	0xb4abffb0,
-	0xf10f0bf4,
-	0xf0070007,
-	0x0bd00303,
-/* 0x07b3: ih_no_other */
-	0xf104bd00,
-	0xf0010007,
-	0x0ad00003,
-	0xfc04bd00,
-	0xfce0fcf0,
-	0xfcb0fcd0,
-	0xfc90fca0,
-	0x0088fe80,
-	0x32f480fc,
-/* 0x07d7: ctx_4160s */
+/* 0x079d: ih_no_fwmthd */
+	0x04b7f19d,
+	0xffb0bd05,
+	0x0bf4b4ab,
+	0x0007f10f,
+	0x0303f007,
+	0xbd000bd0,
+/* 0x07b5: ih_no_other */
+	0x0007f104,
+	0x0003f001,
+	0xbd000ad0,
+	0xfcf0fc04,
+	0xfcd0fce0,
+	0xfca0fcb0,
+	0xfe80fc90,
+	0x80fc0088,
+	0x32f400fc,
+/* 0x07db: ctx_4160s */
 	0xf001f800,
 	0xffb901f7,
 	0x60e7f102,
 	0x40e3f041,
-/* 0x07e7: ctx_4160s_wait */
+/* 0x07eb: ctx_4160s_wait */
 	0xf19d21f4,
 	0xf04160e7,
 	0x21f440e3,
 	0x02ffb968,
 	0xf404ffc8,
 	0x00f8f00b,
-/* 0x07fc: ctx_4160c */
+/* 0x0800: ctx_4160c */
 	0xffb9f4bd,
 	0x60e7f102,
 	0x40e3f041,
 	0xf89d21f4,
-/* 0x080d: ctx_4170s */
+/* 0x0811: ctx_4170s */
 	0x10f5f000,
 	0xf102ffb9,
 	0xf04170e7,
 	0x21f440e3,
-/* 0x081f: ctx_4170w */
+/* 0x0823: ctx_4170w */
 	0xf100f89d,
 	0xf04170e7,
 	0x21f440e3,
 	0x02ffb968,
 	0xf410f4f0,
 	0x00f8f01b,
-/* 0x0834: ctx_redswitch */
+/* 0x0838: ctx_redswitch */
 	0x0200e7f1,
 	0xf040e5f0,
 	0xe5f020e5,
@@ -790,7 +791,7 @@ uint32_t gf117_grhub_code[] = {
 	0x0103f085,
 	0xbd000ed0,
 	0x08f7f004,
-/* 0x0850: ctx_redswitch_delay */
+/* 0x0854: ctx_redswitch_delay */
 	0xf401f2b6,
 	0xe5f1fd1b,
 	0xe5f10400,
@@ -798,7 +799,7 @@ uint32_t gf117_grhub_code[] = {
 	0x03f08500,
 	0x000ed001,
 	0x00f804bd,
-/* 0x086c: ctx_86c */
+/* 0x0870: ctx_86c */
 	0x1b0007f1,
 	0xd00203f0,
 	0x04bd000f,
@@ -809,16 +810,16 @@ uint32_t gf117_grhub_code[] = {
 	0xa86ce7f1,
 	0xf441e3f0,
 	0x00f89d21,
-/* 0x0894: ctx_mem */
+/* 0x0898: ctx_mem */
 	0x840007f1,
 	0xd00203f0,
 	0x04bd000f,
-/* 0x08a0: ctx_mem_wait */
+/* 0x08a4: ctx_mem_wait */
 	0x8400f7f1,
 	0xcf02f3f0,
 	0xfffd00ff,
 	0xf31bf405,
-/* 0x08b2: ctx_load */
+/* 0x08b6: ctx_load */
 	0x94bd00f8,
 	0xf10599f0,
 	0xf00f0007,
@@ -836,7 +837,7 @@ uint32_t gf117_grhub_code[] = {
 	0x02d00203,
 	0xf004bd00,
 	0x21f507f7,
-	0x07f10894,
+	0x07f10898,
 	0x03f0c000,
 	0x0002d002,
 	0x0bfe04bd,
@@ -891,31 +892,31 @@ uint32_t gf117_grhub_code[] = {
 	0x03f01700,
 	0x0009d002,
 	0x00f804bd,
-/* 0x09d0: ctx_chan */
-	0x07d721f5,
-	0x08b221f5,
+/* 0x09d4: ctx_chan */
+	0x07db21f5,
+	0x08b621f5,
 	0xf40ca7f0,
 	0xf7f0d021,
-	0x9421f505,
-	0xfc21f508,
-/* 0x09eb: ctx_mmio_exec */
-	0x9800f807,
+	0x9821f505,
+	0x0021f508,
+/* 0x09ef: ctx_mmio_exec */
+	0x9800f808,
 	0x07f14103,
 	0x03f08100,
 	0x0003d002,
 	0x34bd04bd,
-/* 0x09fc: ctx_mmio_loop */
+/* 0x0a00: ctx_mmio_loop */
 	0xf4ff34c4,
 	0x57f10f1b,
 	0x53f00200,
 	0x0535fa06,
-/* 0x0a0e: ctx_mmio_pull */
+/* 0x0a12: ctx_mmio_pull */
 	0x4e9803f8,
 	0x814f9880,
 	0xb69d21f4,
 	0x12b60830,
 	0xdf1bf401,
-/* 0x0a20: ctx_mmio_done */
+/* 0x0a24: ctx_mmio_done */
 	0xf1160398,
 	0xf0810007,
 	0x03d00203,
@@ -924,30 +925,30 @@ uint32_t gf117_grhub_code[] = {
 	0x13f00100,
 	0x0601fa06,
 	0x00f803f8,
-/* 0x0a40: ctx_xfer */
+/* 0x0a44: ctx_xfer */
 	0xf104e7f0,
 	0xf0020007,
 	0x0ed00303,
-/* 0x0a4f: ctx_xfer_idle */
+/* 0x0a53: ctx_xfer_idle */
 	0xf104bd00,
 	0xf00000e7,
 	0xeecf03e3,
 	0x00e4f100,
 	0xf21bf420,
 	0xf40611f4,
-/* 0x0a66: ctx_xfer_pre */
+/* 0x0a6a: ctx_xfer_pre */
 	0xf7f01102,
-	0x6c21f510,
-	0xd721f508,
+	0x7021f510,
+	0xdb21f508,
 	0x1c11f407,
-/* 0x0a74: ctx_xfer_pre_load */
+/* 0x0a78: ctx_xfer_pre_load */
 	0xf502f7f0,
-	0xf5080d21,
-	0xf5081f21,
-	0xbd083421,
-	0x0d21f5f4,
-	0xb221f508,
-/* 0x0a8d: ctx_xfer_exec */
+	0xf5081121,
+	0xf5082321,
+	0xbd083821,
+	0x1121f5f4,
+	0xb621f508,
+/* 0x0a91: ctx_xfer_exec */
 	0x16019808,
 	0x07f124bd,
 	0x03f00500,
@@ -982,24 +983,23 @@ uint32_t gf117_grhub_code[] = {
 	0x1301f402,
 	0xf40ca7f0,
 	0xf7f0d021,
-	0x9421f505,
+	0x9821f505,
 	0x3202f408,
-/* 0x0b1c: ctx_xfer_post */
+/* 0x0b20: ctx_xfer_post */
 	0xf502f7f0,
-	0xbd080d21,
-	0x6c21f5f4,
+	0xbd081121,
+	0x7021f5f4,
 	0x7f21f508,
-	0x1f21f502,
+	0x2321f502,
 	0xf5f4bd08,
-	0xf4080d21,
+	0xf4081121,
 	0x01981011,
 	0x0511fd40,
 	0xf5070bf4,
-/* 0x0b47: ctx_xfer_no_post_mmio */
-	0xf509eb21,
-/* 0x0b4b: ctx_xfer_done */
-	0xf807fc21,
-	0x00000000,
+/* 0x0b4b: ctx_xfer_no_post_mmio */
+	0xf509ef21,
+/* 0x0b4f: ctx_xfer_done */
+	0xf8080021,
 	0x00000000,
 	0x00000000,
 	0x00000000,
diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/gr/fuc/hubgk104.fuc3.h b/drivers/gpu/drm/nouveau/nvkm/engine/gr/fuc/hubgk104.fuc3.h
index 95ac15110049..125824b394bb 100644
--- a/drivers/gpu/drm/nouveau/nvkm/engine/gr/fuc/hubgk104.fuc3.h
+++ b/drivers/gpu/drm/nouveau/nvkm/engine/gr/fuc/hubgk104.fuc3.h
@@ -528,10 +528,10 @@ uint32_t gk104_grhub_code[] = {
 	0x0001d001,
 	0x17f104bd,
 	0xf7f00100,
-	0xd721f502,
-	0xe921f507,
+	0xdb21f502,
+	0xed21f507,
 	0x10f7f007,
-	0x083621f5,
+	0x083a21f5,
 	0x98000e98,
 	0x21f5010f,
 	0x14950150,
@@ -574,9 +574,9 @@ uint32_t gk104_grhub_code[] = {
 	0xb6800040,
 	0x1bf40132,
 	0x00f7f0be,
-	0x083621f5,
+	0x083a21f5,
 	0xf500f7f0,
-	0xf107d721,
+	0xf107db21,
 	0xf0010007,
 	0x01d00203,
 	0xbd04bd00,
@@ -610,7 +610,7 @@ uint32_t gk104_grhub_code[] = {
 	0x09d00203,
 	0xf404bd00,
 	0x31f40132,
-	0x0221f502,
+	0x0621f502,
 	0xf094bd0a,
 	0x07f10799,
 	0x03f01700,
@@ -621,7 +621,7 @@ uint32_t gk104_grhub_code[] = {
 	0x0203f00f,
 	0xbd0009d0,
 	0x0131f404,
-	0x0a0221f5,
+	0x0a0621f5,
 	0x99f094bd,
 	0x0007f106,
 	0x0203f017,
@@ -631,7 +631,7 @@ uint32_t gk104_grhub_code[] = {
 	0x12b920f9,
 	0x0132f402,
 	0xf50232f4,
-	0xfc0a0221,
+	0xfc0a0621,
 	0x0007f120,
 	0x0203f0c0,
 	0xbd0002d0,
@@ -640,7 +640,7 @@ uint32_t gk104_grhub_code[] = {
 	0xf41f23c8,
 	0x31f40d0b,
 	0x0232f401,
-	0x0a0221f5,
+	0x0a0621f5,
 /* 0x063c: chsw_done */
 	0xf10127f0,
 	0xf0c30007,
@@ -654,7 +654,7 @@ uint32_t gk104_grhub_code[] = {
 /* 0x0660: main_not_ctx_switch */
 	0xf401e4b0,
 	0xf2b90d1b,
-	0x9a21f502,
+	0x9e21f502,
 	0x460ef409,
 /* 0x0670: main_not_ctx_chan */
 	0xf402e4b0,
@@ -664,7 +664,7 @@ uint32_t gk104_grhub_code[] = {
 	0x09d00203,
 	0xf404bd00,
 	0x32f40132,
-	0x0221f502,
+	0x0621f502,
 	0xf094bd0a,
 	0x07f10799,
 	0x03f01700,
@@ -682,90 +682,91 @@ uint32_t gk104_grhub_code[] = {
 	0x04bd0002,
 	0xfea00ef5,
 /* 0x06c8: ih */
-	0x88fe80f9,
-	0xf980f901,
-	0xf9a0f990,
-	0xf9d0f9b0,
-	0xbdf0f9e0,
-	0x00a7f104,
-	0x00a3f002,
-	0xc400aacf,
-	0x0bf404ab,
-	0x10d7f030,
-	0x1a00e7f1,
-	0xcf00e3f0,
-	0xf7f100ee,
-	0xf3f01900,
-	0x00ffcf00,
-	0xb70421f4,
-	0xf00400b0,
-	0x07f101e7,
-	0x03f01d00,
-	0x000ed000,
-/* 0x071a: ih_no_fifo */
-	0xabe404bd,
-	0x0bf40100,
-	0x10d7f00d,
-	0x4001e7f1,
-/* 0x072b: ih_no_ctxsw */
-	0xe40421f4,
-	0xf40400ab,
-	0xe7f16c0b,
-	0xe3f00708,
-	0x6821f440,
-	0xf102ffb9,
-	0xf0040007,
-	0x0fd00203,
-	0xf104bd00,
-	0xf00704e7,
+	0x80f900f9,
+	0xf90188fe,
+	0xf990f980,
+	0xf9b0f9a0,
+	0xf9e0f9d0,
+	0xf104bdf0,
+	0xf00200a7,
+	0xaacf00a3,
+	0x04abc400,
+	0xf0300bf4,
+	0xe7f110d7,
+	0xe3f01a00,
+	0x00eecf00,
+	0x1900f7f1,
+	0xcf00f3f0,
+	0x21f400ff,
+	0x00b0b704,
+	0x01e7f004,
+	0x1d0007f1,
+	0xd00003f0,
+	0x04bd000e,
+/* 0x071c: ih_no_fifo */
+	0x0100abe4,
+	0xf00d0bf4,
+	0xe7f110d7,
+	0x21f44001,
+/* 0x072d: ih_no_ctxsw */
+	0x00abe404,
+	0x6c0bf404,
+	0x0708e7f1,
+	0xf440e3f0,
+	0xffb96821,
+	0x0007f102,
+	0x0203f004,
+	0xbd000fd0,
+	0x04e7f104,
+	0x40e3f007,
+	0xb96821f4,
+	0x07f102ff,
+	0x03f00300,
+	0x000fd002,
+	0xfec704bd,
+	0x02ee9450,
+	0x0700f7f1,
+	0xbb40f3f0,
+	0x21f400ef,
+	0x0007f168,
+	0x0203f002,
+	0xbd000fd0,
+	0x03f7f004,
+	0x037e21f5,
+	0x0100b7f1,
+	0xf102bfb9,
+	0xf00144e7,
 	0x21f440e3,
-	0x02ffb968,
-	0x030007f1,
-	0xd00203f0,
-	0x04bd000f,
-	0x9450fec7,
-	0xf7f102ee,
-	0xf3f00700,
-	0x00efbb40,
-	0xf16821f4,
-	0xf0020007,
-	0x0fd00203,
-	0xf004bd00,
-	0x21f503f7,
-	0xb7f1037e,
-	0xbfb90100,
-	0x44e7f102,
-	0x40e3f001,
-/* 0x079b: ih_no_fwmthd */
-	0xf19d21f4,
-	0xbd0504b7,
-	0xb4abffb0,
-	0xf10f0bf4,
-	0xf0070007,
-	0x0bd00303,
-/* 0x07b3: ih_no_other */
-	0xf104bd00,
-	0xf0010007,
-	0x0ad00003,
-	0xfc04bd00,
-	0xfce0fcf0,
-	0xfcb0fcd0,
-	0xfc90fca0,
-	0x0088fe80,
-	0x32f480fc,
-/* 0x07d7: ctx_4170s */
+/* 0x079d: ih_no_fwmthd */
+	0x04b7f19d,
+	0xffb0bd05,
+	0x0bf4b4ab,
+	0x0007f10f,
+	0x0303f007,
+	0xbd000bd0,
+/* 0x07b5: ih_no_other */
+	0x0007f104,
+	0x0003f001,
+	0xbd000ad0,
+	0xfcf0fc04,
+	0xfcd0fce0,
+	0xfca0fcb0,
+	0xfe80fc90,
+	0x80fc0088,
+	0x32f400fc,
+/* 0x07db: ctx_4170s */
 	0xf001f800,
 	0xffb910f5,
 	0x70e7f102,
 	0x40e3f041,
 	0xf89d21f4,
-/* 0x07e9: ctx_4170w */
+/* 0x07ed: ctx_4170w */
 	0x70e7f100,
 	0x40e3f041,
 	0xb96821f4,
 	0xf4f002ff,
 	0xf01bf410,
-/* 0x07fe: ctx_redswitch */
+/* 0x0802: ctx_redswitch */
 	0xe7f100f8,
 	0xe5f00200,
 	0x20e5f040,
@@ -773,7 +774,7 @@ uint32_t gk104_grhub_code[] = {
 	0xf0850007,
 	0x0ed00103,
 	0xf004bd00,
-/* 0x081a: ctx_redswitch_delay */
+/* 0x081e: ctx_redswitch_delay */
 	0xf2b608f7,
 	0xfd1bf401,
 	0x0400e5f1,
@@ -781,7 +782,7 @@ uint32_t gk104_grhub_code[] = {
 	0x850007f1,
 	0xd00103f0,
 	0x04bd000e,
-/* 0x0836: ctx_86c */
+/* 0x083a: ctx_86c */
 	0x07f100f8,
 	0x03f01b00,
 	0x000fd002,
@@ -792,17 +793,17 @@ uint32_t gk104_grhub_code[] = {
 	0xe7f102ff,
 	0xe3f0a86c,
 	0x9d21f441,
-/* 0x085e: ctx_mem */
+/* 0x0862: ctx_mem */
 	0x07f100f8,
 	0x03f08400,
 	0x000fd002,
-/* 0x086a: ctx_mem_wait */
+/* 0x086e: ctx_mem_wait */
 	0xf7f104bd,
 	0xf3f08400,
 	0x00ffcf02,
 	0xf405fffd,
 	0x00f8f31b,
-/* 0x087c: ctx_load */
+/* 0x0880: ctx_load */
 	0x99f094bd,
 	0x0007f105,
 	0x0203f00f,
@@ -819,7 +820,7 @@ uint32_t gk104_grhub_code[] = {
 	0x0203f083,
 	0xbd0002d0,
 	0x07f7f004,
-	0x085e21f5,
+	0x086221f5,
 	0xc00007f1,
 	0xd00203f0,
 	0x04bd0002,
@@ -874,29 +875,29 @@ uint32_t gk104_grhub_code[] = {
 	0x170007f1,
 	0xd00203f0,
 	0x04bd0009,
-/* 0x099a: ctx_chan */
+/* 0x099e: ctx_chan */
 	0x21f500f8,
-	0xa7f0087c,
+	0xa7f00880,
 	0xd021f40c,
 	0xf505f7f0,
-	0xf8085e21,
-/* 0x09ad: ctx_mmio_exec */
+	0xf8086221,
+/* 0x09b1: ctx_mmio_exec */
 	0x41039800,
 	0x810007f1,
 	0xd00203f0,
 	0x04bd0003,
-/* 0x09be: ctx_mmio_loop */
+/* 0x09c2: ctx_mmio_loop */
 	0x34c434bd,
 	0x0f1bf4ff,
 	0x020057f1,
 	0xfa0653f0,
 	0x03f80535,
-/* 0x09d0: ctx_mmio_pull */
+/* 0x09d4: ctx_mmio_pull */
 	0x98804e98,
 	0x21f4814f,
 	0x0830b69d,
 	0xf40112b6,
-/* 0x09e2: ctx_mmio_done */
+/* 0x09e6: ctx_mmio_done */
 	0x0398df1b,
 	0x0007f116,
 	0x0203f081,
@@ -905,30 +906,30 @@ uint32_t gk104_grhub_code[] = {
 	0x010017f1,
 	0xfa0613f0,
 	0x03f80601,
-/* 0x0a02: ctx_xfer */
+/* 0x0a06: ctx_xfer */
 	0xe7f000f8,
 	0x0007f104,
 	0x0303f002,
 	0xbd000ed0,
-/* 0x0a11: ctx_xfer_idle */
+/* 0x0a15: ctx_xfer_idle */
 	0x00e7f104,
 	0x03e3f000,
 	0xf100eecf,
 	0xf42000e4,
 	0x11f4f21b,
 	0x0d02f406,
-/* 0x0a28: ctx_xfer_pre */
+/* 0x0a2c: ctx_xfer_pre */
 	0xf510f7f0,
-	0xf4083621,
-/* 0x0a32: ctx_xfer_pre_load */
+	0xf4083a21,
+/* 0x0a36: ctx_xfer_pre_load */
 	0xf7f01c11,
-	0xd721f502,
-	0xe921f507,
-	0xfe21f507,
-	0xf5f4bd07,
-	0xf507d721,
-/* 0x0a4b: ctx_xfer_exec */
-	0x98087c21,
+	0xdb21f502,
+	0xed21f507,
+	0x0221f507,
+	0xf5f4bd08,
+	0xf507db21,
+/* 0x0a4f: ctx_xfer_exec */
+	0x98088021,
 	0x24bd1601,
 	0x050007f1,
 	0xd00103f0,
@@ -963,21 +964,21 @@ uint32_t gk104_grhub_code[] = {
 	0xa7f01301,
 	0xd021f40c,
 	0xf505f7f0,
-	0xf4085e21,
-/* 0x0ada: ctx_xfer_post */
+	0xf4086221,
+/* 0x0ade: ctx_xfer_post */
 	0xf7f02e02,
-	0xd721f502,
+	0xdb21f502,
 	0xf5f4bd07,
-	0xf5083621,
+	0xf5083a21,
 	0xf5027f21,
-	0xbd07e921,
-	0xd721f5f4,
+	0xbd07ed21,
+	0xdb21f5f4,
 	0x1011f407,
 	0xfd400198,
 	0x0bf40511,
-	0xad21f507,
-/* 0x0b05: ctx_xfer_no_post_mmio */
-/* 0x0b05: ctx_xfer_done */
+	0xb121f507,
+/* 0x0b09: ctx_xfer_no_post_mmio */
+/* 0x0b09: ctx_xfer_done */
 	0x0000f809,
 	0x00000000,
 	0x00000000,
@@ -1040,5 +1041,4 @@ uint32_t gk104_grhub_code[] = {
 	0x00000000,
 	0x00000000,
 	0x00000000,
-	0x00000000,
 };
diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/gr/fuc/hubgk110.fuc3.h b/drivers/gpu/drm/nouveau/nvkm/engine/gr/fuc/hubgk110.fuc3.h
index 89986878480f..0a1b8c0b8b82 100644
--- a/drivers/gpu/drm/nouveau/nvkm/engine/gr/fuc/hubgk110.fuc3.h
+++ b/drivers/gpu/drm/nouveau/nvkm/engine/gr/fuc/hubgk110.fuc3.h
@@ -528,10 +528,10 @@ uint32_t gk110_grhub_code[] = {
 	0x0001d001,
 	0x17f104bd,
 	0xf7f00100,
-	0xd721f502,
-	0xe921f507,
+	0xdb21f502,
+	0xed21f507,
 	0x10f7f007,
-	0x083621f5,
+	0x083a21f5,
 	0x98000e98,
 	0x21f5010f,
 	0x14950150,
@@ -574,9 +574,9 @@ uint32_t gk110_grhub_code[] = {
 	0xb6800040,
 	0x1bf40132,
 	0x00f7f0be,
-	0x083621f5,
+	0x083a21f5,
 	0xf500f7f0,
-	0xf107d721,
+	0xf107db21,
 	0xf0010007,
 	0x01d00203,
 	0xbd04bd00,
@@ -610,7 +610,7 @@ uint32_t gk110_grhub_code[] = {
 	0x09d00203,
 	0xf404bd00,
 	0x31f40132,
-	0x0221f502,
+	0x0621f502,
 	0xf094bd0a,
 	0x07f10799,
 	0x03f01700,
@@ -621,7 +621,7 @@ uint32_t gk110_grhub_code[] = {
 	0x0203f037,
 	0xbd0009d0,
 	0x0131f404,
-	0x0a0221f5,
+	0x0a0621f5,
 	0x99f094bd,
 	0x0007f106,
 	0x0203f017,
@@ -631,7 +631,7 @@ uint32_t gk110_grhub_code[] = {
 	0x12b920f9,
 	0x0132f402,
 	0xf50232f4,
-	0xfc0a0221,
+	0xfc0a0621,
 	0x0007f120,
 	0x0203f0c0,
 	0xbd0002d0,
@@ -640,7 +640,7 @@ uint32_t gk110_grhub_code[] = {
 	0xf41f23c8,
 	0x31f40d0b,
 	0x0232f401,
-	0x0a0221f5,
+	0x0a0621f5,
 /* 0x063c: chsw_done */
 	0xf10127f0,
 	0xf0c30007,
@@ -654,7 +654,7 @@ uint32_t gk110_grhub_code[] = {
 /* 0x0660: main_not_ctx_switch */
 	0xf401e4b0,
 	0xf2b90d1b,
-	0x9a21f502,
+	0x9e21f502,
 	0x460ef409,
 /* 0x0670: main_not_ctx_chan */
 	0xf402e4b0,
@@ -664,7 +664,7 @@ uint32_t gk110_grhub_code[] = {
 	0x09d00203,
 	0xf404bd00,
 	0x32f40132,
-	0x0221f502,
+	0x0621f502,
 	0xf094bd0a,
 	0x07f10799,
 	0x03f01700,
@@ -682,90 +682,91 @@ uint32_t gk110_grhub_code[] = {
 	0x04bd0002,
 	0xfea00ef5,
 /* 0x06c8: ih */
-	0x88fe80f9,
-	0xf980f901,
-	0xf9a0f990,
-	0xf9d0f9b0,
-	0xbdf0f9e0,
-	0x00a7f104,
-	0x00a3f002,
-	0xc400aacf,
-	0x0bf404ab,
-	0x10d7f030,
-	0x1a00e7f1,
-	0xcf00e3f0,
-	0xf7f100ee,
-	0xf3f01900,
-	0x00ffcf00,
-	0xb70421f4,
-	0xf00400b0,
-	0x07f101e7,
-	0x03f01d00,
-	0x000ed000,
-/* 0x071a: ih_no_fifo */
-	0xabe404bd,
-	0x0bf40100,
-	0x10d7f00d,
-	0x4001e7f1,
-/* 0x072b: ih_no_ctxsw */
-	0xe40421f4,
-	0xf40400ab,
-	0xe7f16c0b,
-	0xe3f00708,
-	0x6821f440,
-	0xf102ffb9,
-	0xf0040007,
-	0x0fd00203,
-	0xf104bd00,
-	0xf00704e7,
+	0x80f900f9,
+	0xf90188fe,
+	0xf990f980,
+	0xf9b0f9a0,
+	0xf9e0f9d0,
+	0xf104bdf0,
+	0xf00200a7,
+	0xaacf00a3,
+	0x04abc400,
+	0xf0300bf4,
+	0xe7f110d7,
+	0xe3f01a00,
+	0x00eecf00,
+	0x1900f7f1,
+	0xcf00f3f0,
+	0x21f400ff,
+	0x00b0b704,
+	0x01e7f004,
+	0x1d0007f1,
+	0xd00003f0,
+	0x04bd000e,
+/* 0x071c: ih_no_fifo */
+	0x0100abe4,
+	0xf00d0bf4,
+	0xe7f110d7,
+	0x21f44001,
+/* 0x072d: ih_no_ctxsw */
+	0x00abe404,
+	0x6c0bf404,
+	0x0708e7f1,
+	0xf440e3f0,
+	0xffb96821,
+	0x0007f102,
+	0x0203f004,
+	0xbd000fd0,
+	0x04e7f104,
+	0x40e3f007,
+	0xb96821f4,
+	0x07f102ff,
+	0x03f00300,
+	0x000fd002,
+	0xfec704bd,
+	0x02ee9450,
+	0x0700f7f1,
+	0xbb40f3f0,
+	0x21f400ef,
+	0x0007f168,
+	0x0203f002,
+	0xbd000fd0,
+	0x03f7f004,
+	0x037e21f5,
+	0x0100b7f1,
+	0xf102bfb9,
+	0xf00144e7,
 	0x21f440e3,
-	0x02ffb968,
-	0x030007f1,
-	0xd00203f0,
-	0x04bd000f,
-	0x9450fec7,
-	0xf7f102ee,
-	0xf3f00700,
-	0x00efbb40,
-	0xf16821f4,
-	0xf0020007,
-	0x0fd00203,
-	0xf004bd00,
-	0x21f503f7,
-	0xb7f1037e,
-	0xbfb90100,
-	0x44e7f102,
-	0x40e3f001,
-/* 0x079b: ih_no_fwmthd */
-	0xf19d21f4,
-	0xbd0504b7,
-	0xb4abffb0,
-	0xf10f0bf4,
-	0xf0070007,
-	0x0bd00303,
-/* 0x07b3: ih_no_other */
-	0xf104bd00,
-	0xf0010007,
-	0x0ad00003,
-	0xfc04bd00,
-	0xfce0fcf0,
-	0xfcb0fcd0,
-	0xfc90fca0,
-	0x0088fe80,
-	0x32f480fc,
-/* 0x07d7: ctx_4170s */
+/* 0x079d: ih_no_fwmthd */
+	0x04b7f19d,
+	0xffb0bd05,
+	0x0bf4b4ab,
+	0x0007f10f,
+	0x0303f007,
+	0xbd000bd0,
+/* 0x07b5: ih_no_other */
+	0x0007f104,
+	0x0003f001,
+	0xbd000ad0,
+	0xfcf0fc04,
+	0xfcd0fce0,
+	0xfca0fcb0,
+	0xfe80fc90,
+	0x80fc0088,
+	0x32f400fc,
+/* 0x07db: ctx_4170s */
 	0xf001f800,
 	0xffb910f5,
 	0x70e7f102,
 	0x40e3f041,
 	0xf89d21f4,
-/* 0x07e9: ctx_4170w */
+/* 0x07ed: ctx_4170w */
 	0x70e7f100,
 	0x40e3f041,
 	0xb96821f4,
 	0xf4f002ff,
 	0xf01bf410,
-/* 0x07fe: ctx_redswitch */
+/* 0x0802: ctx_redswitch */
 	0xe7f100f8,
 	0xe5f00200,
 	0x20e5f040,
@@ -773,7 +774,7 @@ uint32_t gk110_grhub_code[] = {
 	0xf0850007,
 	0x0ed00103,
 	0xf004bd00,
-/* 0x081a: ctx_redswitch_delay */
+/* 0x081e: ctx_redswitch_delay */
 	0xf2b608f7,
 	0xfd1bf401,
 	0x0400e5f1,
@@ -781,7 +782,7 @@ uint32_t gk110_grhub_code[] = {
 	0x850007f1,
 	0xd00103f0,
 	0x04bd000e,
-/* 0x0836: ctx_86c */
+/* 0x083a: ctx_86c */
 	0x07f100f8,
 	0x03f02300,
 	0x000fd002,
@@ -792,17 +793,17 @@ uint32_t gk110_grhub_code[] = {
 	0xe7f102ff,
 	0xe3f0a88c,
 	0x9d21f441,
-/* 0x085e: ctx_mem */
+/* 0x0862: ctx_mem */
 	0x07f100f8,
 	0x03f08400,
 	0x000fd002,
-/* 0x086a: ctx_mem_wait */
+/* 0x086e: ctx_mem_wait */
 	0xf7f104bd,
 	0xf3f08400,
 	0x00ffcf02,
 	0xf405fffd,
 	0x00f8f31b,
-/* 0x087c: ctx_load */
+/* 0x0880: ctx_load */
 	0x99f094bd,
 	0x0007f105,
 	0x0203f037,
@@ -819,7 +820,7 @@ uint32_t gk110_grhub_code[] = {
 	0x0203f083,
 	0xbd0002d0,
 	0x07f7f004,
-	0x085e21f5,
+	0x086221f5,
 	0xc00007f1,
 	0xd00203f0,
 	0x04bd0002,
@@ -874,29 +875,29 @@ uint32_t gk110_grhub_code[] = {
 	0x170007f1,
 	0xd00203f0,
 	0x04bd0009,
-/* 0x099a: ctx_chan */
+/* 0x099e: ctx_chan */
 	0x21f500f8,
-	0xa7f0087c,
+	0xa7f00880,
 	0xd021f40c,
 	0xf505f7f0,
-	0xf8085e21,
-/* 0x09ad: ctx_mmio_exec */
+	0xf8086221,
+/* 0x09b1: ctx_mmio_exec */
 	0x41039800,
 	0x810007f1,
 	0xd00203f0,
 	0x04bd0003,
-/* 0x09be: ctx_mmio_loop */
+/* 0x09c2: ctx_mmio_loop */
 	0x34c434bd,
 	0x0f1bf4ff,
 	0x020057f1,
 	0xfa0653f0,
 	0x03f80535,
-/* 0x09d0: ctx_mmio_pull */
+/* 0x09d4: ctx_mmio_pull */
 	0x98804e98,
 	0x21f4814f,
 	0x0830b69d,
 	0xf40112b6,
-/* 0x09e2: ctx_mmio_done */
+/* 0x09e6: ctx_mmio_done */
 	0x0398df1b,
 	0x0007f116,
 	0x0203f081,
@@ -905,30 +906,30 @@ uint32_t gk110_grhub_code[] = {
 	0x010017f1,
 	0xfa0613f0,
 	0x03f80601,
-/* 0x0a02: ctx_xfer */
+/* 0x0a06: ctx_xfer */
 	0xe7f000f8,
 	0x0007f104,
 	0x0303f002,
 	0xbd000ed0,
-/* 0x0a11: ctx_xfer_idle */
+/* 0x0a15: ctx_xfer_idle */
 	0x00e7f104,
 	0x03e3f000,
 	0xf100eecf,
 	0xf42000e4,
 	0x11f4f21b,
 	0x0d02f406,
-/* 0x0a28: ctx_xfer_pre */
+/* 0x0a2c: ctx_xfer_pre */
 	0xf510f7f0,
-	0xf4083621,
-/* 0x0a32: ctx_xfer_pre_load */
+	0xf4083a21,
+/* 0x0a36: ctx_xfer_pre_load */
 	0xf7f01c11,
-	0xd721f502,
-	0xe921f507,
-	0xfe21f507,
-	0xf5f4bd07,
-	0xf507d721,
-/* 0x0a4b: ctx_xfer_exec */
-	0x98087c21,
+	0xdb21f502,
+	0xed21f507,
+	0x0221f507,
+	0xf5f4bd08,
+	0xf507db21,
+/* 0x0a4f: ctx_xfer_exec */
+	0x98088021,
 	0x24bd1601,
 	0x050007f1,
 	0xd00103f0,
@@ -963,21 +964,21 @@ uint32_t gk110_grhub_code[] = {
 	0xa7f01301,
 	0xd021f40c,
 	0xf505f7f0,
-	0xf4085e21,
-/* 0x0ada: ctx_xfer_post */
+	0xf4086221,
+/* 0x0ade: ctx_xfer_post */
 	0xf7f02e02,
-	0xd721f502,
+	0xdb21f502,
 	0xf5f4bd07,
-	0xf5083621,
+	0xf5083a21,
 	0xf5027f21,
-	0xbd07e921,
-	0xd721f5f4,
+	0xbd07ed21,
+	0xdb21f5f4,
 	0x1011f407,
 	0xfd400198,
 	0x0bf40511,
-	0xad21f507,
-/* 0x0b05: ctx_xfer_no_post_mmio */
-/* 0x0b05: ctx_xfer_done */
+	0xb121f507,
+/* 0x0b09: ctx_xfer_no_post_mmio */
+/* 0x0b09: ctx_xfer_done */
 	0x0000f809,
 	0x00000000,
 	0x00000000,
@@ -1040,5 +1041,4 @@ uint32_t gk110_grhub_code[] = {
 	0x00000000,
 	0x00000000,
 	0x00000000,
-	0x00000000,
 };
diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/gr/fuc/hubgk208.fuc5.h b/drivers/gpu/drm/nouveau/nvkm/engine/gr/fuc/hubgk208.fuc5.h
index 0e98fa4a386e..16869d0b109b 100644
--- a/drivers/gpu/drm/nouveau/nvkm/engine/gr/fuc/hubgk208.fuc5.h
+++ b/drivers/gpu/drm/nouveau/nvkm/engine/gr/fuc/hubgk208.fuc5.h
@@ -478,10 +478,10 @@ uint32_t gk208_grhub_code[] = {
 	0x01040080,
 	0xbd0001f6,
 	0x01004104,
-	0xa87e020f,
-	0xb77e0006,
+	0xac7e020f,
+	0xbb7e0006,
 	0x100f0006,
-	0x0006f97e,
+	0x0006fd7e,
 	0x98000e98,
 	0x207e010f,
 	0x14950001,
@@ -523,8 +523,8 @@ uint32_t gk208_grhub_code[] = {
 	0x800040b7,
 	0xf40132b6,
 	0x000fb41b,
-	0x0006f97e,
-	0xa87e000f,
+	0x0006fd7e,
+	0xac7e000f,
 	0x00800006,
 	0x01f60201,
 	0xbd04bd00,
@@ -554,7 +554,7 @@ uint32_t gk208_grhub_code[] = {
 	0x0009f602,
 	0x32f404bd,
 	0x0231f401,
-	0x00087c7e,
+	0x0008807e,
 	0x99f094bd,
 	0x17008007,
 	0x0009f602,
@@ -563,7 +563,7 @@ uint32_t gk208_grhub_code[] = {
 	0x37008006,
 	0x0009f602,
 	0x31f404bd,
-	0x087c7e01,
+	0x08807e01,
 	0xf094bd00,
 	0x00800699,
 	0x09f60217,
@@ -572,7 +572,7 @@ uint32_t gk208_grhub_code[] = {
 	0x20f92f0e,
 	0x32f412b2,
 	0x0232f401,
-	0x00087c7e,
+	0x0008807e,
 	0x008020fc,
 	0x02f602c0,
 	0xf404bd00,
@@ -580,7 +580,7 @@ uint32_t gk208_grhub_code[] = {
 	0x23c8130e,
 	0x0d0bf41f,
 	0xf40131f4,
-	0x7c7e0232,
+	0x807e0232,
 /* 0x054e: chsw_done */
 	0x01020008,
 	0x02c30080,
@@ -593,7 +593,7 @@ uint32_t gk208_grhub_code[] = {
 	0xb0ff2a0e,
 	0x1bf401e4,
 	0x7ef2b20c,
-	0xf400081c,
+	0xf4000820,
 /* 0x057a: main_not_ctx_chan */
 	0xe4b0400e,
 	0x2c1bf402,
@@ -602,7 +602,7 @@ uint32_t gk208_grhub_code[] = {
 	0x0009f602,
 	0x32f404bd,
 	0x0232f401,
-	0x00087c7e,
+	0x0008807e,
 	0x99f094bd,
 	0x17008007,
 	0x0009f602,
@@ -618,91 +618,92 @@ uint32_t gk208_grhub_code[] = {
 	0xbd0002f6,
 	0xcc0ef504,
 /* 0x05c9: ih */
-	0xfe80f9fe,
-	0x80f90188,
-	0xa0f990f9,
-	0xd0f9b0f9,
-	0xf0f9e0f9,
-	0x004a04bd,
-	0x00aacf02,
-	0xf404abc4,
-	0x100d230b,
-	0xcf1a004e,
-	0x004f00ee,
-	0x00ffcf19,
+	0xf900f9fe,
+	0x0188fe80,
+	0x90f980f9,
+	0xb0f9a0f9,
+	0xe0f9d0f9,
+	0x04bdf0f9,
+	0xcf02004a,
+	0xabc400aa,
+	0x230bf404,
+	0x004e100d,
+	0x00eecf1a,
+	0xcf19004f,
+	0x047e00ff,
+	0xb0b70000,
+	0x010e0400,
+	0xf61d0040,
+	0x04bd000e,
+/* 0x060c: ih_no_fifo */
+	0x0100abe4,
+	0x0d0c0bf4,
+	0x40014e10,
 	0x0000047e,
-	0x0400b0b7,
-	0x0040010e,
-	0x000ef61d,
-/* 0x060a: ih_no_fifo */
-	0xabe404bd,
-	0x0bf40100,
-	0x4e100d0c,
-	0x047e4001,
-/* 0x061a: ih_no_ctxsw */
-	0xabe40000,
-	0x0bf40400,
-	0x07088e56,
-	0x00657e40,
-	0x80ffb200,
-	0xf6020400,
-	0x04bd000f,
-	0x4007048e,
-	0x0000657e,
-	0x0080ffb2,
-	0x0ff60203,
-	0xc704bd00,
-	0xee9450fe,
-	0x07008f02,
-	0x00efbb40,
-	0x0000657e,
-	0x02020080,
+/* 0x061c: ih_no_ctxsw */
+	0x0400abe4,
+	0x8e560bf4,
+	0x7e400708,
+	0xb2000065,
+	0x040080ff,
+	0x000ff602,
+	0x048e04bd,
+	0x657e4007,
+	0xffb20000,
+	0x02030080,
 	0xbd000ff6,
-	0x7e030f04,
-	0x4b0002f8,
-	0xbfb20100,
-	0x4001448e,
-	0x00008f7e,
-/* 0x0674: ih_no_fwmthd */
-	0xbd05044b,
-	0xb4abffb0,
-	0x800c0bf4,
-	0xf6030700,
-	0x04bd000b,
-/* 0x0688: ih_no_other */
-	0xf6010040,
-	0x04bd000a,
-	0xe0fcf0fc,
-	0xb0fcd0fc,
-	0x90fca0fc,
-	0x88fe80fc,
-	0xf480fc00,
+	0x50fec704,
+	0x8f02ee94,
+	0xbb400700,
+	0x657e00ef,
+	0x00800000,
+	0x0ff60202,
+	0x0f04bd00,
+	0x02f87e03,
+	0x01004b00,
+	0x448ebfb2,
+	0x8f7e4001,
+/* 0x0676: ih_no_fwmthd */
+	0x044b0000,
+	0xffb0bd05,
+	0x0bf4b4ab,
+	0x0700800c,
+	0x000bf603,
+/* 0x068a: ih_no_other */
+	0x004004bd,
+	0x000af601,
+	0xf0fc04bd,
+	0xd0fce0fc,
+	0xa0fcb0fc,
+	0x80fc90fc,
+	0xfc0088fe,
+	0xf400fc80,
 	0x01f80032,
-/* 0x06a8: ctx_4170s */
+/* 0x06ac: ctx_4170s */
 	0xb210f5f0,
 	0x41708eff,
 	0x008f7e40,
-/* 0x06b7: ctx_4170w */
+/* 0x06bb: ctx_4170w */
 	0x8e00f800,
 	0x7e404170,
 	0xb2000065,
 	0x10f4f0ff,
 	0xf8f31bf4,
-/* 0x06c9: ctx_redswitch */
+/* 0x06cd: ctx_redswitch */
 	0x02004e00,
 	0xf040e5f0,
 	0xe5f020e5,
 	0x85008010,
 	0x000ef601,
 	0x080f04bd,
-/* 0x06e0: ctx_redswitch_delay */
+/* 0x06e4: ctx_redswitch_delay */
 	0xf401f2b6,
 	0xe5f1fd1b,
 	0xe5f10400,
 	0x00800100,
 	0x0ef60185,
 	0xf804bd00,
-/* 0x06f9: ctx_86c */
+/* 0x06fd: ctx_86c */
 	0x23008000,
 	0x000ff602,
 	0xffb204bd,
@@ -711,15 +712,15 @@ uint32_t gk208_grhub_code[] = {
 	0x8c8effb2,
 	0x8f7e41a8,
 	0x00f80000,
-/* 0x0718: ctx_mem */
+/* 0x071c: ctx_mem */
 	0x02840080,
 	0xbd000ff6,
-/* 0x0721: ctx_mem_wait */
+/* 0x0725: ctx_mem_wait */
 	0x84008f04,
 	0x00ffcf02,
 	0xf405fffd,
 	0x00f8f61b,
-/* 0x0730: ctx_load */
+/* 0x0734: ctx_load */
 	0x99f094bd,
 	0x37008005,
 	0x0009f602,
@@ -733,7 +734,7 @@ uint32_t gk208_grhub_code[] = {
 	0x02830080,
 	0xbd0002f6,
 	0x7e070f04,
-	0x80000718,
+	0x8000071c,
 	0xf602c000,
 	0x04bd0002,
 	0xf0000bfe,
@@ -779,28 +780,28 @@ uint32_t gk208_grhub_code[] = {
 	0x17008005,
 	0x0009f602,
 	0x00f804bd,
-/* 0x081c: ctx_chan */
-	0x0007307e,
+/* 0x0820: ctx_chan */
+	0x0007347e,
 	0xb87e0c0a,
 	0x050f0000,
-	0x0007187e,
-/* 0x082e: ctx_mmio_exec */
+	0x00071c7e,
+/* 0x0832: ctx_mmio_exec */
 	0x039800f8,
 	0x81008041,
 	0x0003f602,
 	0x34bd04bd,
-/* 0x083c: ctx_mmio_loop */
+/* 0x0840: ctx_mmio_loop */
 	0xf4ff34c4,
 	0x00450e1b,
 	0x0653f002,
 	0xf80535fa,
-/* 0x084d: ctx_mmio_pull */
+/* 0x0851: ctx_mmio_pull */
 	0x804e9803,
 	0x7e814f98,
 	0xb600008f,
 	0x12b60830,
 	0xdf1bf401,
-/* 0x0860: ctx_mmio_done */
+/* 0x0864: ctx_mmio_done */
 	0x80160398,
 	0xf6028100,
 	0x04bd0003,
@@ -808,27 +809,27 @@ uint32_t gk208_grhub_code[] = {
 	0x13f00100,
 	0x0601fa06,
 	0x00f803f8,
-/* 0x087c: ctx_xfer */
+/* 0x0880: ctx_xfer */
 	0x0080040e,
 	0x0ef60302,
-/* 0x0887: ctx_xfer_idle */
+/* 0x088b: ctx_xfer_idle */
 	0x8e04bd00,
 	0xcf030000,
 	0xe4f100ee,
 	0x1bf42000,
 	0x0611f4f5,
-/* 0x089b: ctx_xfer_pre */
+/* 0x089f: ctx_xfer_pre */
 	0x0f0c02f4,
-	0x06f97e10,
+	0x06fd7e10,
 	0x1b11f400,
-/* 0x08a4: ctx_xfer_pre_load */
-	0xa87e020f,
-	0xb77e0006,
-	0xc97e0006,
+/* 0x08a8: ctx_xfer_pre_load */
+	0xac7e020f,
+	0xbb7e0006,
+	0xcd7e0006,
 	0xf4bd0006,
-	0x0006a87e,
-	0x0007307e,
-/* 0x08bc: ctx_xfer_exec */
+	0x0006ac7e,
+	0x0007347e,
+/* 0x08c0: ctx_xfer_exec */
 	0xbd160198,
 	0x05008024,
 	0x0002f601,
@@ -858,21 +859,21 @@ uint32_t gk208_grhub_code[] = {
 	0x01f40002,
 	0x7e0c0a12,
 	0x0f0000b8,
-	0x07187e05,
+	0x071c7e05,
 	0x2d02f400,
-/* 0x0938: ctx_xfer_post */
-	0xa87e020f,
+/* 0x093c: ctx_xfer_post */
+	0xac7e020f,
 	0xf4bd0006,
-	0x0006f97e,
+	0x0006fd7e,
 	0x0002277e,
-	0x0006b77e,
-	0xa87ef4bd,
+	0x0006bb7e,
+	0xac7ef4bd,
 	0x11f40006,
 	0x40019810,
 	0xf40511fd,
-	0x2e7e070b,
-/* 0x0962: ctx_xfer_no_post_mmio */
-/* 0x0962: ctx_xfer_done */
+	0x327e070b,
+/* 0x0966: ctx_xfer_no_post_mmio */
+/* 0x0966: ctx_xfer_done */
 	0x00f80008,
 	0x00000000,
 	0x00000000,
@@ -912,5 +913,4 @@ uint32_t gk208_grhub_code[] = {
 	0x00000000,
 	0x00000000,
 	0x00000000,
-	0x00000000,
 };
diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/gr/fuc/hubgm107.fuc5.h b/drivers/gpu/drm/nouveau/nvkm/engine/gr/fuc/hubgm107.fuc5.h
index 5f953c5c20b7..d6343d2a614c 100644
--- a/drivers/gpu/drm/nouveau/nvkm/engine/gr/fuc/hubgm107.fuc5.h
+++ b/drivers/gpu/drm/nouveau/nvkm/engine/gr/fuc/hubgm107.fuc5.h
@@ -478,10 +478,10 @@ uint32_t gm107_grhub_code[] = {
 	0x01040080,
 	0xbd0001f6,
 	0x01004104,
-	0xa87e020f,
-	0xb77e0006,
+	0xac7e020f,
+	0xbb7e0006,
 	0x100f0006,
-	0x0006f97e,
+	0x0006fd7e,
 	0x98000e98,
 	0x207e010f,
 	0x14950001,
@@ -523,8 +523,8 @@ uint32_t gm107_grhub_code[] = {
 	0x800040b7,
 	0xf40132b6,
 	0x000fb41b,
-	0x0006f97e,
-	0xa87e000f,
+	0x0006fd7e,
+	0xac7e000f,
 	0x00800006,
 	0x01f60201,
 	0xbd04bd00,
@@ -554,7 +554,7 @@ uint32_t gm107_grhub_code[] = {
 	0x0009f602,
 	0x32f404bd,
 	0x0231f401,
-	0x00087c7e,
+	0x0008807e,
 	0x99f094bd,
 	0x17008007,
 	0x0009f602,
@@ -563,7 +563,7 @@ uint32_t gm107_grhub_code[] = {
 	0x37008006,
 	0x0009f602,
 	0x31f404bd,
-	0x087c7e01,
+	0x08807e01,
 	0xf094bd00,
 	0x00800699,
 	0x09f60217,
@@ -572,7 +572,7 @@ uint32_t gm107_grhub_code[] = {
 	0x20f92f0e,
 	0x32f412b2,
 	0x0232f401,
-	0x00087c7e,
+	0x0008807e,
 	0x008020fc,
 	0x02f602c0,
 	0xf404bd00,
@@ -580,7 +580,7 @@ uint32_t gm107_grhub_code[] = {
 	0x23c8130e,
 	0x0d0bf41f,
 	0xf40131f4,
-	0x7c7e0232,
+	0x807e0232,
 /* 0x054e: chsw_done */
 	0x01020008,
 	0x02c30080,
@@ -593,7 +593,7 @@ uint32_t gm107_grhub_code[] = {
 	0xb0ff2a0e,
 	0x1bf401e4,
 	0x7ef2b20c,
-	0xf400081c,
+	0xf4000820,
 /* 0x057a: main_not_ctx_chan */
 	0xe4b0400e,
 	0x2c1bf402,
@@ -602,7 +602,7 @@ uint32_t gm107_grhub_code[] = {
 	0x0009f602,
 	0x32f404bd,
 	0x0232f401,
-	0x00087c7e,
+	0x0008807e,
 	0x99f094bd,
 	0x17008007,
 	0x0009f602,
@@ -618,91 +618,92 @@ uint32_t gm107_grhub_code[] = {
 	0xbd0002f6,
 	0xcc0ef504,
 /* 0x05c9: ih */
-	0xfe80f9fe,
-	0x80f90188,
-	0xa0f990f9,
-	0xd0f9b0f9,
-	0xf0f9e0f9,
-	0x004a04bd,
-	0x00aacf02,
-	0xf404abc4,
-	0x100d230b,
-	0xcf1a004e,
-	0x004f00ee,
-	0x00ffcf19,
+	0xf900f9fe,
+	0x0188fe80,
+	0x90f980f9,
+	0xb0f9a0f9,
+	0xe0f9d0f9,
+	0x04bdf0f9,
+	0xcf02004a,
+	0xabc400aa,
+	0x230bf404,
+	0x004e100d,
+	0x00eecf1a,
+	0xcf19004f,
+	0x047e00ff,
+	0xb0b70000,
+	0x010e0400,
+	0xf61d0040,
+	0x04bd000e,
+/* 0x060c: ih_no_fifo */
+	0x0100abe4,
+	0x0d0c0bf4,
+	0x40014e10,
 	0x0000047e,
-	0x0400b0b7,
-	0x0040010e,
-	0x000ef61d,
-/* 0x060a: ih_no_fifo */
-	0xabe404bd,
-	0x0bf40100,
-	0x4e100d0c,
-	0x047e4001,
-/* 0x061a: ih_no_ctxsw */
-	0xabe40000,
-	0x0bf40400,
-	0x07088e56,
-	0x00657e40,
-	0x80ffb200,
-	0xf6020400,
-	0x04bd000f,
-	0x4007048e,
-	0x0000657e,
-	0x0080ffb2,
-	0x0ff60203,
-	0xc704bd00,
-	0xee9450fe,
-	0x07008f02,
-	0x00efbb40,
-	0x0000657e,
-	0x02020080,
+/* 0x061c: ih_no_ctxsw */
+	0x0400abe4,
+	0x8e560bf4,
+	0x7e400708,
+	0xb2000065,
+	0x040080ff,
+	0x000ff602,
+	0x048e04bd,
+	0x657e4007,
+	0xffb20000,
+	0x02030080,
 	0xbd000ff6,
-	0x7e030f04,
-	0x4b0002f8,
-	0xbfb20100,
-	0x4001448e,
-	0x00008f7e,
-/* 0x0674: ih_no_fwmthd */
-	0xbd05044b,
-	0xb4abffb0,
-	0x800c0bf4,
-	0xf6030700,
-	0x04bd000b,
-/* 0x0688: ih_no_other */
-	0xf6010040,
-	0x04bd000a,
-	0xe0fcf0fc,
-	0xb0fcd0fc,
-	0x90fca0fc,
-	0x88fe80fc,
-	0xf480fc00,
+	0x50fec704,
+	0x8f02ee94,
+	0xbb400700,
+	0x657e00ef,
+	0x00800000,
+	0x0ff60202,
+	0x0f04bd00,
+	0x02f87e03,
+	0x01004b00,
+	0x448ebfb2,
+	0x8f7e4001,
+/* 0x0676: ih_no_fwmthd */
+	0x044b0000,
+	0xffb0bd05,
+	0x0bf4b4ab,
+	0x0700800c,
+	0x000bf603,
+/* 0x068a: ih_no_other */
+	0x004004bd,
+	0x000af601,
+	0xf0fc04bd,
+	0xd0fce0fc,
+	0xa0fcb0fc,
+	0x80fc90fc,
+	0xfc0088fe,
+	0xf400fc80,
 	0x01f80032,
-/* 0x06a8: ctx_4170s */
+/* 0x06ac: ctx_4170s */
 	0xb210f5f0,
 	0x41708eff,
 	0x008f7e40,
-/* 0x06b7: ctx_4170w */
+/* 0x06bb: ctx_4170w */
 	0x8e00f800,
 	0x7e404170,
 	0xb2000065,
 	0x10f4f0ff,
 	0xf8f31bf4,
-/* 0x06c9: ctx_redswitch */
+/* 0x06cd: ctx_redswitch */
 	0x02004e00,
 	0xf040e5f0,
 	0xe5f020e5,
 	0x85008010,
 	0x000ef601,
 	0x080f04bd,
-/* 0x06e0: ctx_redswitch_delay */
+/* 0x06e4: ctx_redswitch_delay */
 	0xf401f2b6,
 	0xe5f1fd1b,
 	0xe5f10400,
 	0x00800100,
 	0x0ef60185,
 	0xf804bd00,
-/* 0x06f9: ctx_86c */
+/* 0x06fd: ctx_86c */
 	0x23008000,
 	0x000ff602,
 	0xffb204bd,
@@ -711,15 +712,15 @@ uint32_t gm107_grhub_code[] = {
 	0x8c8effb2,
 	0x8f7e41a8,
 	0x00f80000,
-/* 0x0718: ctx_mem */
+/* 0x071c: ctx_mem */
 	0x02840080,
 	0xbd000ff6,
-/* 0x0721: ctx_mem_wait */
+/* 0x0725: ctx_mem_wait */
 	0x84008f04,
 	0x00ffcf02,
 	0xf405fffd,
 	0x00f8f61b,
-/* 0x0730: ctx_load */
+/* 0x0734: ctx_load */
 	0x99f094bd,
 	0x37008005,
 	0x0009f602,
@@ -733,7 +734,7 @@ uint32_t gm107_grhub_code[] = {
 	0x02830080,
 	0xbd0002f6,
 	0x7e070f04,
-	0x80000718,
+	0x8000071c,
 	0xf602c000,
 	0x04bd0002,
 	0xf0000bfe,
@@ -779,28 +780,28 @@ uint32_t gm107_grhub_code[] = {
 	0x17008005,
 	0x0009f602,
 	0x00f804bd,
-/* 0x081c: ctx_chan */
-	0x0007307e,
+/* 0x0820: ctx_chan */
+	0x0007347e,
 	0xb87e0c0a,
 	0x050f0000,
-	0x0007187e,
-/* 0x082e: ctx_mmio_exec */
+	0x00071c7e,
+/* 0x0832: ctx_mmio_exec */
 	0x039800f8,
 	0x81008041,
 	0x0003f602,
 	0x34bd04bd,
-/* 0x083c: ctx_mmio_loop */
+/* 0x0840: ctx_mmio_loop */
 	0xf4ff34c4,
 	0x00450e1b,
 	0x0653f002,
 	0xf80535fa,
-/* 0x084d: ctx_mmio_pull */
+/* 0x0851: ctx_mmio_pull */
 	0x804e9803,
 	0x7e814f98,
 	0xb600008f,
 	0x12b60830,
 	0xdf1bf401,
-/* 0x0860: ctx_mmio_done */
+/* 0x0864: ctx_mmio_done */
 	0x80160398,
 	0xf6028100,
 	0x04bd0003,
@@ -808,27 +809,27 @@ uint32_t gm107_grhub_code[] = {
 	0x13f00100,
 	0x0601fa06,
 	0x00f803f8,
-/* 0x087c: ctx_xfer */
+/* 0x0880: ctx_xfer */
 	0x0080040e,
 	0x0ef60302,
-/* 0x0887: ctx_xfer_idle */
+/* 0x088b: ctx_xfer_idle */
 	0x8e04bd00,
 	0xcf030000,
 	0xe4f100ee,
 	0x1bf42000,
 	0x0611f4f5,
-/* 0x089b: ctx_xfer_pre */
+/* 0x089f: ctx_xfer_pre */
 	0x0f0c02f4,
-	0x06f97e10,
+	0x06fd7e10,
 	0x1b11f400,
-/* 0x08a4: ctx_xfer_pre_load */
-	0xa87e020f,
-	0xb77e0006,
-	0xc97e0006,
+/* 0x08a8: ctx_xfer_pre_load */
+	0xac7e020f,
+	0xbb7e0006,
+	0xcd7e0006,
 	0xf4bd0006,
-	0x0006a87e,
-	0x0007307e,
-/* 0x08bc: ctx_xfer_exec */
+	0x0006ac7e,
+	0x0007347e,
+/* 0x08c0: ctx_xfer_exec */
 	0xbd160198,
 	0x05008024,
 	0x0002f601,
@@ -858,21 +859,21 @@ uint32_t gm107_grhub_code[] = {
 	0x01f40002,
 	0x7e0c0a12,
 	0x0f0000b8,
-	0x07187e05,
+	0x071c7e05,
 	0x2d02f400,
-/* 0x0938: ctx_xfer_post */
-	0xa87e020f,
+/* 0x093c: ctx_xfer_post */
+	0xac7e020f,
 	0xf4bd0006,
-	0x0006f97e,
+	0x0006fd7e,
 	0x0002277e,
-	0x0006b77e,
-	0xa87ef4bd,
+	0x0006bb7e,
+	0xac7ef4bd,
 	0x11f40006,
 	0x40019810,
 	0xf40511fd,
-	0x2e7e070b,
-/* 0x0962: ctx_xfer_no_post_mmio */
-/* 0x0962: ctx_xfer_done */
+	0x327e070b,
+/* 0x0966: ctx_xfer_no_post_mmio */
+/* 0x0966: ctx_xfer_done */
 	0x00f80008,
 	0x00000000,
 	0x00000000,
@@ -912,5 +913,4 @@ uint32_t gm107_grhub_code[] = {
 	0x00000000,
 	0x00000000,
 	0x00000000,
-	0x00000000,
 };
diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/gr/gf100.c b/drivers/gpu/drm/nouveau/nvkm/engine/gr/gf100.c
index 1f81069edc58..c56a886229f1 100644
--- a/drivers/gpu/drm/nouveau/nvkm/engine/gr/gf100.c
+++ b/drivers/gpu/drm/nouveau/nvkm/engine/gr/gf100.c
@@ -27,6 +27,8 @@
 
 #include <core/client.h>
 #include <core/option.h>
+#include <core/firmware.h>
+#include <subdev/secboot.h>
 #include <subdev/fb.h>
 #include <subdev/mc.h>
 #include <subdev/pmu.h>
@@ -1427,21 +1429,40 @@ gf100_gr_init_ctxctl(struct gf100_gr *gr)
 	const struct gf100_grctx_func *grctx = gr->func->grctx;
 	struct nvkm_subdev *subdev = &gr->base.engine.subdev;
 	struct nvkm_device *device = subdev->device;
+	struct nvkm_secboot *sb = device->secboot;
 	int i;
 
 	if (gr->firmware) {
 		/* load fuc microcode */
 		nvkm_mc_unk260(device->mc, 0);
-		gf100_gr_init_fw(gr, 0x409000, &gr->fuc409c, &gr->fuc409d);
-		gf100_gr_init_fw(gr, 0x41a000, &gr->fuc41ac, &gr->fuc41ad);
+
+		/* securely-managed falcons must be reset using secure boot */
+		if (nvkm_secboot_is_managed(sb, NVKM_SECBOOT_FALCON_FECS))
+			nvkm_secboot_reset(sb, NVKM_SECBOOT_FALCON_FECS);
+		else
+			gf100_gr_init_fw(gr, 0x409000, &gr->fuc409c,
+					 &gr->fuc409d);
+		if (nvkm_secboot_is_managed(sb, NVKM_SECBOOT_FALCON_GPCCS))
+			nvkm_secboot_reset(sb, NVKM_SECBOOT_FALCON_GPCCS);
+		else
+			gf100_gr_init_fw(gr, 0x41a000, &gr->fuc41ac,
+					 &gr->fuc41ad);
+
 		nvkm_mc_unk260(device->mc, 1);
 
 		/* start both of them running */
 		nvkm_wr32(device, 0x409840, 0xffffffff);
 		nvkm_wr32(device, 0x41a10c, 0x00000000);
 		nvkm_wr32(device, 0x40910c, 0x00000000);
-		nvkm_wr32(device, 0x41a100, 0x00000002);
-		nvkm_wr32(device, 0x409100, 0x00000002);
+
+		if (nvkm_secboot_is_managed(sb, NVKM_SECBOOT_FALCON_GPCCS))
+			nvkm_secboot_start(sb, NVKM_SECBOOT_FALCON_GPCCS);
+		else
+			nvkm_wr32(device, 0x41a100, 0x00000002);
+		if (nvkm_secboot_is_managed(sb, NVKM_SECBOOT_FALCON_FECS))
+			nvkm_secboot_start(sb, NVKM_SECBOOT_FALCON_FECS);
+		else
+			nvkm_wr32(device, 0x409100, 0x00000002);
 		if (nvkm_msec(device, 2000,
 			if (nvkm_rd32(device, 0x409800) & 0x00000001)
 				break;
@@ -1683,6 +1704,12 @@ gf100_gr_dtor_fw(struct gf100_gr_fuc *fuc)
 	fuc->data = NULL;
 }
 
+static void
+gf100_gr_dtor_init(struct gf100_gr_pack *pack)
+{
+	vfree(pack);
+}
+
 void *
 gf100_gr_dtor(struct nvkm_gr *base)
 {
@@ -1697,6 +1724,11 @@ gf100_gr_dtor(struct nvkm_gr *base)
 	gf100_gr_dtor_fw(&gr->fuc41ac);
 	gf100_gr_dtor_fw(&gr->fuc41ad);
 
+	gf100_gr_dtor_init(gr->fuc_bundle);
+	gf100_gr_dtor_init(gr->fuc_method);
+	gf100_gr_dtor_init(gr->fuc_sw_ctx);
+	gf100_gr_dtor_init(gr->fuc_sw_nonctx);
+
 	nvkm_memory_del(&gr->unk4188b8);
 	nvkm_memory_del(&gr->unk4188b4);
 	return gr;
@@ -1720,22 +1752,9 @@ gf100_gr_ctor_fw(struct gf100_gr *gr, const char *fwname,
 	struct nvkm_subdev *subdev = &gr->base.engine.subdev;
 	struct nvkm_device *device = subdev->device;
 	const struct firmware *fw;
-	char f[64];
-	char cname[16];
 	int ret;
-	int i;
 
-	/* Convert device name to lowercase */
-	strncpy(cname, device->chip->name, sizeof(cname));
-	cname[sizeof(cname) - 1] = '\0';
-	i = strlen(cname);
-	while (i) {
-		--i;
-		cname[i] = tolower(cname[i]);
-	}
-
-	snprintf(f, sizeof(f), "nvidia/%s/%s.bin", cname, fwname);
-	ret = request_firmware(&fw, f, device->dev);
+	ret = nvkm_firmware_get(device, fwname, &fw);
 	if (ret) {
 		nvkm_error(subdev, "failed to load %s\n", fwname);
 		return ret;
@@ -1743,7 +1762,7 @@ gf100_gr_ctor_fw(struct gf100_gr *gr, const char *fwname,
 
 	fuc->size = fw->size;
 	fuc->data = kmemdup(fw->data, fuc->size, GFP_KERNEL);
-	release_firmware(fw);
+	nvkm_firmware_put(fw);
 	return (fuc->data != NULL) ? 0 : -ENOMEM;
 }
 
@@ -1763,15 +1782,6 @@ gf100_gr_ctor(const struct gf100_gr_func *func, struct nvkm_device *device,
 	if (ret)
 		return ret;
 
-	if (gr->firmware) {
-		nvkm_info(&gr->base.engine.subdev, "using external firmware\n");
-		if (gf100_gr_ctor_fw(gr, "fecs_inst", &gr->fuc409c) ||
-		    gf100_gr_ctor_fw(gr, "fecs_data", &gr->fuc409d) ||
-		    gf100_gr_ctor_fw(gr, "gpccs_inst", &gr->fuc41ac) ||
-		    gf100_gr_ctor_fw(gr, "gpccs_data", &gr->fuc41ad))
-			return -ENODEV;
-	}
-
 	return 0;
 }
 
@@ -1780,10 +1790,25 @@ gf100_gr_new_(const struct gf100_gr_func *func, struct nvkm_device *device,
 	      int index, struct nvkm_gr **pgr)
 {
 	struct gf100_gr *gr;
+	int ret;
+
 	if (!(gr = kzalloc(sizeof(*gr), GFP_KERNEL)))
 		return -ENOMEM;
 	*pgr = &gr->base;
-	return gf100_gr_ctor(func, device, index, gr);
+
+	ret = gf100_gr_ctor(func, device, index, gr);
+	if (ret)
+		return ret;
+
+	if (gr->firmware) {
+		if (gf100_gr_ctor_fw(gr, "fecs_inst", &gr->fuc409c) ||
+		    gf100_gr_ctor_fw(gr, "fecs_data", &gr->fuc409d) ||
+		    gf100_gr_ctor_fw(gr, "gpccs_inst", &gr->fuc41ac) ||
+		    gf100_gr_ctor_fw(gr, "gpccs_data", &gr->fuc41ad))
+			return -ENODEV;
+	}
+
+	return 0;
 }
 
 int
diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/gr/gf100.h b/drivers/gpu/drm/nouveau/nvkm/engine/gr/gf100.h
index 02e78b8d93f6..f0c6acb0f8fd 100644
--- a/drivers/gpu/drm/nouveau/nvkm/engine/gr/gf100.h
+++ b/drivers/gpu/drm/nouveau/nvkm/engine/gr/gf100.h
@@ -82,7 +82,7 @@ struct gf100_gr {
 
 	/*
 	 * Used if the register packs are loaded from NVIDIA fw instead of
-	 * using hardcoded arrays.
+	 * using hardcoded arrays. To be allocated with vzalloc().
 	 */
 	struct gf100_gr_pack *fuc_sw_nonctx;
 	struct gf100_gr_pack *fuc_sw_ctx;
@@ -138,12 +138,9 @@ int gf100_gr_init(struct gf100_gr *);
 
 int gk104_gr_init(struct gf100_gr *);
 
-int gk20a_gr_new_(const struct gf100_gr_func *, struct nvkm_device *,
-		  int, struct nvkm_gr **);
-void gk20a_gr_dtor(struct gf100_gr *);
 int gk20a_gr_init(struct gf100_gr *);
 
-int gm204_gr_init(struct gf100_gr *);
+int gm200_gr_init(struct gf100_gr *);
 
 #define gf100_gr_chan(p) container_of((p), struct gf100_gr_chan, object)
 
@@ -204,6 +201,17 @@ void gf100_gr_icmd(struct gf100_gr *, const struct gf100_gr_pack *);
 void gf100_gr_mthd(struct gf100_gr *, const struct gf100_gr_pack *);
 int  gf100_gr_init_ctxctl(struct gf100_gr *);
 
+/* external bundles loading functions */
+int gk20a_gr_av_to_init(struct gf100_gr *, const char *,
+			struct gf100_gr_pack **);
+int gk20a_gr_aiv_to_init(struct gf100_gr *, const char *,
+			 struct gf100_gr_pack **);
+int gk20a_gr_av_to_method(struct gf100_gr *, const char *,
+			  struct gf100_gr_pack **);
+
+int gm200_gr_new_(const struct gf100_gr_func *, struct nvkm_device *, int,
+		  struct nvkm_gr **);
+
 /* register init value lists */
 
 extern const struct gf100_gr_init gf100_gr_init_main_0[];
@@ -279,6 +287,4 @@ extern const struct gf100_gr_init gm107_gr_init_l1c_0[];
 extern const struct gf100_gr_init gm107_gr_init_wwdx_0[];
 extern const struct gf100_gr_init gm107_gr_init_cbm_0[];
 void gm107_gr_init_bios(struct gf100_gr *);
-
-extern const struct gf100_gr_pack gm204_gr_pack_mmio[];
 #endif
diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/gr/gk20a.c b/drivers/gpu/drm/nouveau/nvkm/engine/gr/gk20a.c
index b8758d3b8b51..7ffb8a626196 100644
--- a/drivers/gpu/drm/nouveau/nvkm/engine/gr/gk20a.c
+++ b/drivers/gpu/drm/nouveau/nvkm/engine/gr/gk20a.c
@@ -26,37 +26,40 @@
 
 #include <nvif/class.h>
 
-static void
-gk20a_gr_init_dtor(struct gf100_gr_pack *pack)
-{
-	vfree(pack);
-}
-
 struct gk20a_fw_av
 {
 	u32 addr;
 	u32 data;
 };
 
-static struct gf100_gr_pack *
-gk20a_gr_av_to_init(struct gf100_gr_fuc *fuc)
+int
+gk20a_gr_av_to_init(struct gf100_gr *gr, const char *fw_name,
+		    struct gf100_gr_pack **ppack)
 {
+	struct gf100_gr_fuc fuc;
 	struct gf100_gr_init *init;
 	struct gf100_gr_pack *pack;
-	const int nent = (fuc->size / sizeof(struct gk20a_fw_av));
+	int nent;
+	int ret;
 	int i;
 
+	ret = gf100_gr_ctor_fw(gr, fw_name, &fuc);
+	if (ret)
+		return ret;
+
+	nent = (fuc.size / sizeof(struct gk20a_fw_av));
 	pack = vzalloc((sizeof(*pack) * 2) + (sizeof(*init) * (nent + 1)));
-	if (!pack)
-		return ERR_PTR(-ENOMEM);
+	if (!pack) {
+		ret = -ENOMEM;
+		goto end;
+	}
 
 	init = (void *)(pack + 2);
-
 	pack[0].init = init;
 
 	for (i = 0; i < nent; i++) {
 		struct gf100_gr_init *ent = &init[i];
-		struct gk20a_fw_av *av = &((struct gk20a_fw_av *)fuc->data)[i];
+		struct gk20a_fw_av *av = &((struct gk20a_fw_av *)fuc.data)[i];
 
 		ent->addr = av->addr;
 		ent->data = av->data;
@@ -64,7 +67,11 @@ gk20a_gr_av_to_init(struct gf100_gr_fuc *fuc)
 		ent->pitch = 1;
 	}
 
-	return pack;
+	*ppack = pack;
+
+end:
+	gf100_gr_dtor_fw(&fuc);
+	return ret;
 }
 
 struct gk20a_fw_aiv
@@ -74,25 +81,34 @@ struct gk20a_fw_aiv
 	u32 data;
 };
 
-static struct gf100_gr_pack *
-gk20a_gr_aiv_to_init(struct gf100_gr_fuc *fuc)
+int
+gk20a_gr_aiv_to_init(struct gf100_gr *gr, const char *fw_name,
+		     struct gf100_gr_pack **ppack)
 {
+	struct gf100_gr_fuc fuc;
 	struct gf100_gr_init *init;
 	struct gf100_gr_pack *pack;
-	const int nent = (fuc->size / sizeof(struct gk20a_fw_aiv));
+	int nent;
+	int ret;
 	int i;
 
+	ret = gf100_gr_ctor_fw(gr, fw_name, &fuc);
+	if (ret)
+		return ret;
+
+	nent = (fuc.size / sizeof(struct gk20a_fw_aiv));
 	pack = vzalloc((sizeof(*pack) * 2) + (sizeof(*init) * (nent + 1)));
-	if (!pack)
-		return ERR_PTR(-ENOMEM);
+	if (!pack) {
+		ret = -ENOMEM;
+		goto end;
+	}
 
 	init = (void *)(pack + 2);
-
 	pack[0].init = init;
 
 	for (i = 0; i < nent; i++) {
 		struct gf100_gr_init *ent = &init[i];
-		struct gk20a_fw_aiv *av = &((struct gk20a_fw_aiv *)fuc->data)[i];
+		struct gk20a_fw_aiv *av = &((struct gk20a_fw_aiv *)fuc.data)[i];
 
 		ent->addr = av->addr;
 		ent->data = av->data;
@@ -100,30 +116,45 @@ gk20a_gr_aiv_to_init(struct gf100_gr_fuc *fuc)
 		ent->pitch = 1;
 	}
 
-	return pack;
+	*ppack = pack;
+
+end:
+	gf100_gr_dtor_fw(&fuc);
+	return ret;
 }
 
-static struct gf100_gr_pack *
-gk20a_gr_av_to_method(struct gf100_gr_fuc *fuc)
+int
+gk20a_gr_av_to_method(struct gf100_gr *gr, const char *fw_name,
+		      struct gf100_gr_pack **ppack)
 {
+	struct gf100_gr_fuc fuc;
 	struct gf100_gr_init *init;
 	struct gf100_gr_pack *pack;
 	/* We don't suppose we will initialize more than 16 classes here... */
 	static const unsigned int max_classes = 16;
-	const int nent = (fuc->size / sizeof(struct gk20a_fw_av));
-	int i, classidx = 0;
-	u32 prevclass = 0;
+	u32 classidx = 0, prevclass = 0;
+	int nent;
+	int ret;
+	int i;
+
+	ret = gf100_gr_ctor_fw(gr, fw_name, &fuc);
+	if (ret)
+		return ret;
+
+	nent = (fuc.size / sizeof(struct gk20a_fw_av));
 
 	pack = vzalloc((sizeof(*pack) * max_classes) +
 		       (sizeof(*init) * (nent + 1)));
-	if (!pack)
-		return ERR_PTR(-ENOMEM);
+	if (!pack) {
+		ret = -ENOMEM;
+		goto end;
+	}
 
 	init = (void *)(pack + max_classes);
 
 	for (i = 0; i < nent; i++) {
 		struct gf100_gr_init *ent = &init[i];
-		struct gk20a_fw_av *av = &((struct gk20a_fw_av *)fuc->data)[i];
+		struct gk20a_fw_av *av = &((struct gk20a_fw_av *)fuc.data)[i];
 		u32 class = av->addr & 0xffff;
 		u32 addr = (av->addr & 0xffff0000) >> 14;
 
@@ -133,7 +164,8 @@ gk20a_gr_av_to_method(struct gf100_gr_fuc *fuc)
 			prevclass = class;
 			if (++classidx >= max_classes) {
 				vfree(pack);
-				return ERR_PTR(-ENOSPC);
+				ret = -ENOSPC;
+				goto end;
 			}
 		}
 
@@ -143,7 +175,11 @@ gk20a_gr_av_to_method(struct gf100_gr_fuc *fuc)
 		ent->pitch = 1;
 	}
 
-	return pack;
+	*ppack = pack;
+
+end:
+	gf100_gr_dtor_fw(&fuc);
+	return ret;
 }
 
 static int
@@ -273,20 +309,24 @@ gk20a_gr_init(struct gf100_gr *gr)
 	return gf100_gr_init_ctxctl(gr);
 }
 
-void
-gk20a_gr_dtor(struct gf100_gr *gr)
-{
-	gk20a_gr_init_dtor(gr->fuc_method);
-	gk20a_gr_init_dtor(gr->fuc_bundle);
-	gk20a_gr_init_dtor(gr->fuc_sw_ctx);
-	gk20a_gr_init_dtor(gr->fuc_sw_nonctx);
-}
+static const struct gf100_gr_func
+gk20a_gr = {
+	.init = gk20a_gr_init,
+	.set_hww_esr_report_mask = gk20a_gr_set_hww_esr_report_mask,
+	.ppc_nr = 1,
+	.grctx = &gk20a_grctx,
+	.sclass = {
+		{ -1, -1, FERMI_TWOD_A },
+		{ -1, -1, KEPLER_INLINE_TO_MEMORY_A },
+		{ -1, -1, KEPLER_C, &gf100_fermi },
+		{ -1, -1, KEPLER_COMPUTE_A },
+		{}
+	}
+};
 
 int
-gk20a_gr_new_(const struct gf100_gr_func *func, struct nvkm_device *device,
-	      int index, struct nvkm_gr **pgr)
+gk20a_gr_new(struct nvkm_device *device, int index, struct nvkm_gr **pgr)
 {
-	struct gf100_gr_fuc fuc;
 	struct gf100_gr *gr;
 	int ret;
 
@@ -294,63 +334,32 @@ gk20a_gr_new_(const struct gf100_gr_func *func, struct nvkm_device *device,
 		return -ENOMEM;
 	*pgr = &gr->base;
 
-	ret = gf100_gr_ctor(func, device, index, gr);
+	ret = gf100_gr_ctor(&gk20a_gr, device, index, gr);
 	if (ret)
 		return ret;
 
-	ret = gf100_gr_ctor_fw(gr, "sw_nonctx", &fuc);
+	if (gf100_gr_ctor_fw(gr, "fecs_inst", &gr->fuc409c) ||
+	    gf100_gr_ctor_fw(gr, "fecs_data", &gr->fuc409d) ||
+	    gf100_gr_ctor_fw(gr, "gpccs_inst", &gr->fuc41ac) ||
+	    gf100_gr_ctor_fw(gr, "gpccs_data", &gr->fuc41ad))
+		return -ENODEV;
+
+	ret = gk20a_gr_av_to_init(gr, "sw_nonctx", &gr->fuc_sw_nonctx);
 	if (ret)
 		return ret;
-	gr->fuc_sw_nonctx = gk20a_gr_av_to_init(&fuc);
-	gf100_gr_dtor_fw(&fuc);
-	if (IS_ERR(gr->fuc_sw_nonctx))
-		return PTR_ERR(gr->fuc_sw_nonctx);
 
-	ret = gf100_gr_ctor_fw(gr, "sw_ctx", &fuc);
+	ret = gk20a_gr_aiv_to_init(gr, "sw_ctx", &gr->fuc_sw_ctx);
 	if (ret)
 		return ret;
-	gr->fuc_sw_ctx = gk20a_gr_aiv_to_init(&fuc);
-	gf100_gr_dtor_fw(&fuc);
-	if (IS_ERR(gr->fuc_sw_ctx))
-		return PTR_ERR(gr->fuc_sw_ctx);
 
-	ret = gf100_gr_ctor_fw(gr, "sw_bundle_init", &fuc);
+	ret = gk20a_gr_av_to_init(gr, "sw_bundle_init", &gr->fuc_bundle);
 	if (ret)
 		return ret;
-	gr->fuc_bundle = gk20a_gr_av_to_init(&fuc);
-	gf100_gr_dtor_fw(&fuc);
-	if (IS_ERR(gr->fuc_bundle))
-		return PTR_ERR(gr->fuc_bundle);
 
-	ret = gf100_gr_ctor_fw(gr, "sw_method_init", &fuc);
+	ret = gk20a_gr_av_to_method(gr, "sw_method_init", &gr->fuc_method);
 	if (ret)
 		return ret;
-	gr->fuc_method = gk20a_gr_av_to_method(&fuc);
-	gf100_gr_dtor_fw(&fuc);
-	if (IS_ERR(gr->fuc_method))
-		return PTR_ERR(gr->fuc_method);
 
-	return 0;
-}
 
-static const struct gf100_gr_func
-gk20a_gr = {
-	.dtor = gk20a_gr_dtor,
-	.init = gk20a_gr_init,
-	.set_hww_esr_report_mask = gk20a_gr_set_hww_esr_report_mask,
-	.ppc_nr = 1,
-	.grctx = &gk20a_grctx,
-	.sclass = {
-		{ -1, -1, FERMI_TWOD_A },
-		{ -1, -1, KEPLER_INLINE_TO_MEMORY_A },
-		{ -1, -1, KEPLER_C, &gf100_fermi },
-		{ -1, -1, KEPLER_COMPUTE_A },
-		{}
-	}
-};
-
-int
-gk20a_gr_new(struct nvkm_device *device, int index, struct nvkm_gr **pgr)
-{
-	return gk20a_gr_new_(&gk20a_gr, device, index, pgr);
+	return 0;
 }
diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/gr/gm200.c b/drivers/gpu/drm/nouveau/nvkm/engine/gr/gm200.c
new file mode 100644
index 000000000000..058fc1d22c09
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/nvkm/engine/gr/gm200.c
@@ -0,0 +1,207 @@
+/*
+ * Copyright 2015 Red Hat Inc.
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
+ * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR
+ * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
+ * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
+ * OTHER DEALINGS IN THE SOFTWARE.
+ *
+ * Authors: Ben Skeggs <bskeggs@redhat.com>
+ */
+#include "gf100.h"
+#include "ctxgf100.h"
+
+#include <subdev/secboot.h>
+
+#include <nvif/class.h>
+
+/*******************************************************************************
+ * PGRAPH engine/subdev functions
+ ******************************************************************************/
+
+int
+gm200_gr_init(struct gf100_gr *gr)
+{
+	struct nvkm_device *device = gr->base.engine.subdev.device;
+	const u32 magicgpc918 = DIV_ROUND_UP(0x00800000, gr->tpc_total);
+	u32 data[TPC_MAX / 8] = {}, tmp;
+	u8  tpcnr[GPC_MAX];
+	int gpc, tpc, ppc, rop;
+	int i;
+
+	tmp = nvkm_rd32(device, 0x100c80); /*XXX: mask? */
+	nvkm_wr32(device, 0x418880, 0x00001000 | (tmp & 0x00000fff));
+	nvkm_wr32(device, 0x418890, 0x00000000);
+	nvkm_wr32(device, 0x418894, 0x00000000);
+	nvkm_wr32(device, 0x4188b4, nvkm_memory_addr(gr->unk4188b4) >> 8);
+	nvkm_wr32(device, 0x4188b8, nvkm_memory_addr(gr->unk4188b8) >> 8);
+	nvkm_mask(device, 0x4188b0, 0x00040000, 0x00040000);
+
+	/*XXX: belongs in fb */
+	nvkm_wr32(device, 0x100cc8, nvkm_memory_addr(gr->unk4188b4) >> 8);
+	nvkm_wr32(device, 0x100ccc, nvkm_memory_addr(gr->unk4188b8) >> 8);
+	nvkm_mask(device, 0x100cc4, 0x00040000, 0x00040000);
+
+	gf100_gr_mmio(gr, gr->fuc_sw_nonctx);
+
+	gm107_gr_init_bios(gr);
+
+	nvkm_wr32(device, GPC_UNIT(0, 0x3018), 0x00000001);
+
+	memset(data, 0x00, sizeof(data));
+	memcpy(tpcnr, gr->tpc_nr, sizeof(gr->tpc_nr));
+	for (i = 0, gpc = -1; i < gr->tpc_total; i++) {
+		do {
+			gpc = (gpc + 1) % gr->gpc_nr;
+		} while (!tpcnr[gpc]);
+		tpc = gr->tpc_nr[gpc] - tpcnr[gpc]--;
+
+		data[i / 8] |= tpc << ((i % 8) * 4);
+	}
+
+	nvkm_wr32(device, GPC_BCAST(0x0980), data[0]);
+	nvkm_wr32(device, GPC_BCAST(0x0984), data[1]);
+	nvkm_wr32(device, GPC_BCAST(0x0988), data[2]);
+	nvkm_wr32(device, GPC_BCAST(0x098c), data[3]);
+
+	for (gpc = 0; gpc < gr->gpc_nr; gpc++) {
+		nvkm_wr32(device, GPC_UNIT(gpc, 0x0914),
+			gr->magic_not_rop_nr << 8 | gr->tpc_nr[gpc]);
+		nvkm_wr32(device, GPC_UNIT(gpc, 0x0910), 0x00040000 |
+			gr->tpc_total);
+		nvkm_wr32(device, GPC_UNIT(gpc, 0x0918), magicgpc918);
+	}
+
+	nvkm_wr32(device, GPC_BCAST(0x3fd4), magicgpc918);
+	nvkm_wr32(device, GPC_BCAST(0x08ac), nvkm_rd32(device, 0x100800));
+	nvkm_wr32(device, GPC_BCAST(0x033c), nvkm_rd32(device, 0x100804));
+
+	nvkm_wr32(device, 0x400500, 0x00010001);
+	nvkm_wr32(device, 0x400100, 0xffffffff);
+	nvkm_wr32(device, 0x40013c, 0xffffffff);
+	nvkm_wr32(device, 0x400124, 0x00000002);
+	nvkm_wr32(device, 0x409c24, 0x000e0000);
+	nvkm_wr32(device, 0x405848, 0xc0000000);
+	nvkm_wr32(device, 0x40584c, 0x00000001);
+	nvkm_wr32(device, 0x404000, 0xc0000000);
+	nvkm_wr32(device, 0x404600, 0xc0000000);
+	nvkm_wr32(device, 0x408030, 0xc0000000);
+	nvkm_wr32(device, 0x404490, 0xc0000000);
+	nvkm_wr32(device, 0x406018, 0xc0000000);
+	nvkm_wr32(device, 0x407020, 0x40000000);
+	nvkm_wr32(device, 0x405840, 0xc0000000);
+	nvkm_wr32(device, 0x405844, 0x00ffffff);
+	nvkm_mask(device, 0x419cc0, 0x00000008, 0x00000008);
+
+	for (gpc = 0; gpc < gr->gpc_nr; gpc++) {
+		for (ppc = 0; ppc < gr->ppc_nr[gpc]; ppc++)
+			nvkm_wr32(device, PPC_UNIT(gpc, ppc, 0x038), 0xc0000000);
+		nvkm_wr32(device, GPC_UNIT(gpc, 0x0420), 0xc0000000);
+		nvkm_wr32(device, GPC_UNIT(gpc, 0x0900), 0xc0000000);
+		nvkm_wr32(device, GPC_UNIT(gpc, 0x1028), 0xc0000000);
+		nvkm_wr32(device, GPC_UNIT(gpc, 0x0824), 0xc0000000);
+		for (tpc = 0; tpc < gr->tpc_nr[gpc]; tpc++) {
+			nvkm_wr32(device, TPC_UNIT(gpc, tpc, 0x508), 0xffffffff);
+			nvkm_wr32(device, TPC_UNIT(gpc, tpc, 0x50c), 0xffffffff);
+			nvkm_wr32(device, TPC_UNIT(gpc, tpc, 0x224), 0xc0000000);
+			nvkm_wr32(device, TPC_UNIT(gpc, tpc, 0x48c), 0xc0000000);
+			nvkm_wr32(device, TPC_UNIT(gpc, tpc, 0x084), 0xc0000000);
+			nvkm_wr32(device, TPC_UNIT(gpc, tpc, 0x430), 0xc0000000);
+			nvkm_wr32(device, TPC_UNIT(gpc, tpc, 0x644), 0x00dffffe);
+			nvkm_wr32(device, TPC_UNIT(gpc, tpc, 0x64c), 0x00000005);
+		}
+		nvkm_wr32(device, GPC_UNIT(gpc, 0x2c90), 0xffffffff);
+		nvkm_wr32(device, GPC_UNIT(gpc, 0x2c94), 0xffffffff);
+	}
+
+	for (rop = 0; rop < gr->rop_nr; rop++) {
+		nvkm_wr32(device, ROP_UNIT(rop, 0x144), 0x40000000);
+		nvkm_wr32(device, ROP_UNIT(rop, 0x070), 0x40000000);
+		nvkm_wr32(device, ROP_UNIT(rop, 0x204), 0xffffffff);
+		nvkm_wr32(device, ROP_UNIT(rop, 0x208), 0xffffffff);
+	}
+
+	nvkm_wr32(device, 0x400108, 0xffffffff);
+	nvkm_wr32(device, 0x400138, 0xffffffff);
+	nvkm_wr32(device, 0x400118, 0xffffffff);
+	nvkm_wr32(device, 0x400130, 0xffffffff);
+	nvkm_wr32(device, 0x40011c, 0xffffffff);
+	nvkm_wr32(device, 0x400134, 0xffffffff);
+
+	nvkm_wr32(device, 0x400054, 0x2c350f63);
+
+	gf100_gr_zbc_init(gr);
+
+	return gf100_gr_init_ctxctl(gr);
+}
+
+int
+gm200_gr_new_(const struct gf100_gr_func *func, struct nvkm_device *device,
+	      int index, struct nvkm_gr **pgr)
+{
+	struct gf100_gr *gr;
+	int ret;
+
+	if (!(gr = kzalloc(sizeof(*gr), GFP_KERNEL)))
+		return -ENOMEM;
+	*pgr = &gr->base;
+
+	ret = gf100_gr_ctor(func, device, index, gr);
+	if (ret)
+		return ret;
+
+	/* Load firmwares for non-secure falcons */
+	if (!nvkm_secboot_is_managed(device->secboot,
+				     NVKM_SECBOOT_FALCON_FECS)) {
+		if ((ret = gf100_gr_ctor_fw(gr, "gr/fecs_inst", &gr->fuc409c)) ||
+		    (ret = gf100_gr_ctor_fw(gr, "gr/fecs_data", &gr->fuc409d)))
+			return ret;
+	}
+	if (!nvkm_secboot_is_managed(device->secboot,
+				     NVKM_SECBOOT_FALCON_GPCCS)) {
+		if ((ret = gf100_gr_ctor_fw(gr, "gr/gpccs_inst", &gr->fuc41ac)) ||
+		    (ret = gf100_gr_ctor_fw(gr, "gr/gpccs_data", &gr->fuc41ad)))
+			return ret;
+	}
+
+	if ((ret = gk20a_gr_av_to_init(gr, "gr/sw_nonctx", &gr->fuc_sw_nonctx)) ||
+	    (ret = gk20a_gr_aiv_to_init(gr, "gr/sw_ctx", &gr->fuc_sw_ctx)) ||
+	    (ret = gk20a_gr_av_to_init(gr, "gr/sw_bundle_init", &gr->fuc_bundle)) ||
+	    (ret = gk20a_gr_av_to_method(gr, "gr/sw_method_init", &gr->fuc_method)))
+		return ret;
+
+	return 0;
+}
+
+static const struct gf100_gr_func
+gm200_gr = {
+	.init = gm200_gr_init,
+	.ppc_nr = 2,
+	.grctx = &gm200_grctx,
+	.sclass = {
+		{ -1, -1, FERMI_TWOD_A },
+		{ -1, -1, KEPLER_INLINE_TO_MEMORY_B },
+		{ -1, -1, MAXWELL_B, &gf100_fermi },
+		{ -1, -1, MAXWELL_COMPUTE_B },
+		{}
+	}
+};
+
+int
+gm200_gr_new(struct nvkm_device *device, int index, struct nvkm_gr **pgr)
+{
+	return gm200_gr_new_(&gm200_gr, device, index, pgr);
+}
diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/gr/gm204.c b/drivers/gpu/drm/nouveau/nvkm/engine/gr/gm204.c
deleted file mode 100644
index 90381dde451a..000000000000
--- a/drivers/gpu/drm/nouveau/nvkm/engine/gr/gm204.c
+++ /dev/null
@@ -1,373 +0,0 @@
-/*
- * Copyright 2015 Red Hat Inc.
- *
- * Permission is hereby granted, free of charge, to any person obtaining a
- * copy of this software and associated documentation files (the "Software"),
- * to deal in the Software without restriction, including without limitation
- * the rights to use, copy, modify, merge, publish, distribute, sublicense,
- * and/or sell copies of the Software, and to permit persons to whom the
- * Software is furnished to do so, subject to the following conditions:
- *
- * The above copyright notice and this permission notice shall be included in
- * all copies or substantial portions of the Software.
- *
- * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
- * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
- * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
- * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR
- * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
- * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
- * OTHER DEALINGS IN THE SOFTWARE.
- *
- * Authors: Ben Skeggs <bskeggs@redhat.com>
- */
-#include "gf100.h"
-#include "ctxgf100.h"
-
-#include <nvif/class.h>
-
-/*******************************************************************************
- * PGRAPH register lists
- ******************************************************************************/
-
-static const struct gf100_gr_init
-gm204_gr_init_main_0[] = {
-	{ 0x400080,   1, 0x04, 0x003003e2 },
-	{ 0x400088,   1, 0x04, 0xe007bfe7 },
-	{ 0x40008c,   1, 0x04, 0x00060000 },
-	{ 0x400090,   1, 0x04, 0x00000030 },
-	{ 0x40013c,   1, 0x04, 0x003901f3 },
-	{ 0x400140,   1, 0x04, 0x00000100 },
-	{ 0x400144,   1, 0x04, 0x00000000 },
-	{ 0x400148,   1, 0x04, 0x00000110 },
-	{ 0x400138,   1, 0x04, 0x00000000 },
-	{ 0x400130,   2, 0x04, 0x00000000 },
-	{ 0x400124,   1, 0x04, 0x00000002 },
-	{}
-};
-
-static const struct gf100_gr_init
-gm204_gr_init_fe_0[] = {
-	{ 0x40415c,   1, 0x04, 0x00000000 },
-	{ 0x404170,   1, 0x04, 0x00000000 },
-	{ 0x4041b4,   1, 0x04, 0x00000000 },
-	{ 0x4041b8,   1, 0x04, 0x00000010 },
-	{}
-};
-
-static const struct gf100_gr_init
-gm204_gr_init_ds_0[] = {
-	{ 0x40583c,   1, 0x04, 0x00000000 },
-	{ 0x405844,   1, 0x04, 0x00ffffff },
-	{ 0x40584c,   1, 0x04, 0x00000001 },
-	{ 0x405850,   1, 0x04, 0x00000000 },
-	{ 0x405900,   1, 0x04, 0x00000000 },
-	{ 0x405908,   1, 0x04, 0x00000000 },
-	{}
-};
-
-static const struct gf100_gr_init
-gm204_gr_init_sked_0[] = {
-	{ 0x407010,   1, 0x04, 0x00000000 },
-	{ 0x407040,   1, 0x04, 0x80440434 },
-	{ 0x407048,   1, 0x04, 0x00000008 },
-	{}
-};
-
-static const struct gf100_gr_init
-gm204_gr_init_tpccs_0[] = {
-	{ 0x419d60,   1, 0x04, 0x0000003f },
-	{ 0x419d88,   3, 0x04, 0x00000000 },
-	{ 0x419dc4,   1, 0x04, 0x00000000 },
-	{ 0x419dc8,   1, 0x04, 0x00000501 },
-	{ 0x419dd0,   1, 0x04, 0x00000000 },
-	{ 0x419dd4,   1, 0x04, 0x00000100 },
-	{ 0x419dd8,   1, 0x04, 0x00000001 },
-	{ 0x419ddc,   1, 0x04, 0x00000002 },
-	{ 0x419de0,   1, 0x04, 0x00000001 },
-	{ 0x419de8,   1, 0x04, 0x000000cc },
-	{ 0x419dec,   1, 0x04, 0x00000000 },
-	{ 0x419df0,   1, 0x04, 0x000000cc },
-	{ 0x419df4,   1, 0x04, 0x00000000 },
-	{ 0x419d0c,   1, 0x04, 0x00000000 },
-	{ 0x419d10,   1, 0x04, 0x00000014 },
-	{}
-};
-
-static const struct gf100_gr_init
-gm204_gr_init_pe_0[] = {
-	{ 0x419900,   1, 0x04, 0x000000ff },
-	{ 0x419810,   1, 0x04, 0x00000000 },
-	{ 0x41980c,   1, 0x04, 0x00000010 },
-	{ 0x419844,   1, 0x04, 0x00000000 },
-	{ 0x419838,   1, 0x04, 0x000000ff },
-	{ 0x419850,   1, 0x04, 0x00000004 },
-	{ 0x419854,   2, 0x04, 0x00000000 },
-	{ 0x419894,   3, 0x04, 0x00100401 },
-	{}
-};
-
-static const struct gf100_gr_init
-gm204_gr_init_sm_0[] = {
-	{ 0x419e30,   1, 0x04, 0x000000ff },
-	{ 0x419e00,   1, 0x04, 0x00000000 },
-	{ 0x419ea0,   1, 0x04, 0x00000000 },
-	{ 0x419ee4,   1, 0x04, 0x00000000 },
-	{ 0x419ea4,   1, 0x04, 0x00000100 },
-	{ 0x419ea8,   1, 0x04, 0x00000000 },
-	{ 0x419ee8,   1, 0x04, 0x00000091 },
-	{ 0x419eb4,   1, 0x04, 0x00000000 },
-	{ 0x419ebc,   2, 0x04, 0x00000000 },
-	{ 0x419edc,   1, 0x04, 0x000c1810 },
-	{ 0x419ed8,   1, 0x04, 0x00000000 },
-	{ 0x419ee0,   1, 0x04, 0x00000000 },
-	{}
-};
-
-static const struct gf100_gr_init
-gm204_gr_init_l1c_1[] = {
-	{ 0x419cf8,   2, 0x04, 0x00000000 },
-	{}
-};
-
-static const struct gf100_gr_init
-gm204_gr_init_sm_1[] = {
-	{ 0x419f74,   1, 0x04, 0x00055155 },
-	{ 0x419f80,   4, 0x04, 0x00000000 },
-	{}
-};
-
-static const struct gf100_gr_init
-gm204_gr_init_l1c_2[] = {
-	{ 0x419ccc,   2, 0x04, 0x00000000 },
-	{ 0x419c80,   1, 0x04, 0x3f006022 },
-	{ 0x419c88,   1, 0x04, 0x00210000 },
-	{}
-};
-
-static const struct gf100_gr_init
-gm204_gr_init_pes_0[] = {
-	{ 0x41be50,   1, 0x04, 0x000000ff },
-	{ 0x41be04,   1, 0x04, 0x00000000 },
-	{ 0x41be08,   1, 0x04, 0x00000004 },
-	{ 0x41be0c,   1, 0x04, 0x00000008 },
-	{ 0x41be10,   1, 0x04, 0x2e3b8bc7 },
-	{ 0x41be14,   2, 0x04, 0x00000000 },
-	{ 0x41be3c,   5, 0x04, 0x00100401 },
-	{}
-};
-
-static const struct gf100_gr_init
-gm204_gr_init_be_0[] = {
-	{ 0x408890,   1, 0x04, 0x000000ff },
-	{ 0x40880c,   1, 0x04, 0x00000000 },
-	{ 0x408850,   1, 0x04, 0x00000004 },
-	{ 0x408878,   1, 0x04, 0x01b4201c },
-	{ 0x40887c,   1, 0x04, 0x80004c55 },
-	{ 0x408880,   1, 0x04, 0x0018c258 },
-	{ 0x408884,   1, 0x04, 0x0000160f },
-	{ 0x408974,   1, 0x04, 0x000000ff },
-	{ 0x408910,   9, 0x04, 0x00000000 },
-	{ 0x408950,   1, 0x04, 0x00000000 },
-	{ 0x408954,   1, 0x04, 0x0000ffff },
-	{ 0x408958,   1, 0x04, 0x00000034 },
-	{ 0x40895c,   1, 0x04, 0x84b17403 },
-	{ 0x408960,   1, 0x04, 0x04c1884f },
-	{ 0x408964,   1, 0x04, 0x04714445 },
-	{ 0x408968,   1, 0x04, 0x0280802f },
-	{ 0x40896c,   1, 0x04, 0x04304856 },
-	{ 0x408970,   1, 0x04, 0x00012800 },
-	{ 0x408984,   1, 0x04, 0x00000000 },
-	{ 0x408988,   1, 0x04, 0x08040201 },
-	{ 0x40898c,   1, 0x04, 0x80402010 },
-	{}
-};
-
-const struct gf100_gr_pack
-gm204_gr_pack_mmio[] = {
-	{ gm204_gr_init_main_0 },
-	{ gm204_gr_init_fe_0 },
-	{ gf100_gr_init_pri_0 },
-	{ gf100_gr_init_rstr2d_0 },
-	{ gf100_gr_init_pd_0 },
-	{ gm204_gr_init_ds_0 },
-	{ gm107_gr_init_scc_0 },
-	{ gm204_gr_init_sked_0 },
-	{ gk110_gr_init_cwd_0 },
-	{ gm107_gr_init_prop_0 },
-	{ gk208_gr_init_gpc_unk_0 },
-	{ gf100_gr_init_setup_0 },
-	{ gf100_gr_init_crstr_0 },
-	{ gm107_gr_init_setup_1 },
-	{ gm107_gr_init_zcull_0 },
-	{ gf100_gr_init_gpm_0 },
-	{ gm107_gr_init_gpc_unk_1 },
-	{ gf100_gr_init_gcc_0 },
-	{ gm204_gr_init_tpccs_0 },
-	{ gm107_gr_init_tex_0 },
-	{ gm204_gr_init_pe_0 },
-	{ gm107_gr_init_l1c_0 },
-	{ gf100_gr_init_mpc_0 },
-	{ gm204_gr_init_sm_0 },
-	{ gm204_gr_init_l1c_1 },
-	{ gm204_gr_init_sm_1 },
-	{ gm204_gr_init_l1c_2 },
-	{ gm204_gr_init_pes_0 },
-	{ gm107_gr_init_wwdx_0 },
-	{ gm107_gr_init_cbm_0 },
-	{ gm204_gr_init_be_0 },
-	{}
-};
-
-const struct gf100_gr_pack *
-gm204_gr_data[] = {
-	gm204_gr_pack_mmio,
-	NULL
-};
-
-/*******************************************************************************
- * PGRAPH engine/subdev functions
- ******************************************************************************/
-
-static int
-gm204_gr_init_ctxctl(struct gf100_gr *gr)
-{
-	return 0;
-}
-
-int
-gm204_gr_init(struct gf100_gr *gr)
-{
-	struct nvkm_device *device = gr->base.engine.subdev.device;
-	const u32 magicgpc918 = DIV_ROUND_UP(0x00800000, gr->tpc_total);
-	u32 data[TPC_MAX / 8] = {}, tmp;
-	u8  tpcnr[GPC_MAX];
-	int gpc, tpc, ppc, rop;
-	int i;
-
-	tmp = nvkm_rd32(device, 0x100c80); /*XXX: mask? */
-	nvkm_wr32(device, 0x418880, 0x00001000 | (tmp & 0x00000fff));
-	nvkm_wr32(device, 0x418890, 0x00000000);
-	nvkm_wr32(device, 0x418894, 0x00000000);
-	nvkm_wr32(device, 0x4188b4, nvkm_memory_addr(gr->unk4188b4) >> 8);
-	nvkm_wr32(device, 0x4188b8, nvkm_memory_addr(gr->unk4188b8) >> 8);
-	nvkm_mask(device, 0x4188b0, 0x00040000, 0x00040000);
-
-	/*XXX: belongs in fb */
-	nvkm_wr32(device, 0x100cc8, nvkm_memory_addr(gr->unk4188b4) >> 8);
-	nvkm_wr32(device, 0x100ccc, nvkm_memory_addr(gr->unk4188b8) >> 8);
-	nvkm_mask(device, 0x100cc4, 0x00040000, 0x00040000);
-
-	gf100_gr_mmio(gr, gr->func->mmio);
-
-	gm107_gr_init_bios(gr);
-
-	nvkm_wr32(device, GPC_UNIT(0, 0x3018), 0x00000001);
-
-	memset(data, 0x00, sizeof(data));
-	memcpy(tpcnr, gr->tpc_nr, sizeof(gr->tpc_nr));
-	for (i = 0, gpc = -1; i < gr->tpc_total; i++) {
-		do {
-			gpc = (gpc + 1) % gr->gpc_nr;
-		} while (!tpcnr[gpc]);
-		tpc = gr->tpc_nr[gpc] - tpcnr[gpc]--;
-
-		data[i / 8] |= tpc << ((i % 8) * 4);
-	}
-
-	nvkm_wr32(device, GPC_BCAST(0x0980), data[0]);
-	nvkm_wr32(device, GPC_BCAST(0x0984), data[1]);
-	nvkm_wr32(device, GPC_BCAST(0x0988), data[2]);
-	nvkm_wr32(device, GPC_BCAST(0x098c), data[3]);
-
-	for (gpc = 0; gpc < gr->gpc_nr; gpc++) {
-		nvkm_wr32(device, GPC_UNIT(gpc, 0x0914),
-			gr->magic_not_rop_nr << 8 | gr->tpc_nr[gpc]);
-		nvkm_wr32(device, GPC_UNIT(gpc, 0x0910), 0x00040000 |
-			gr->tpc_total);
-		nvkm_wr32(device, GPC_UNIT(gpc, 0x0918), magicgpc918);
-	}
-
-	nvkm_wr32(device, GPC_BCAST(0x3fd4), magicgpc918);
-	nvkm_wr32(device, GPC_BCAST(0x08ac), nvkm_rd32(device, 0x100800));
-	nvkm_wr32(device, GPC_BCAST(0x033c), nvkm_rd32(device, 0x100804));
-
-	nvkm_wr32(device, 0x400500, 0x00010001);
-	nvkm_wr32(device, 0x400100, 0xffffffff);
-	nvkm_wr32(device, 0x40013c, 0xffffffff);
-	nvkm_wr32(device, 0x400124, 0x00000002);
-	nvkm_wr32(device, 0x409c24, 0x000e0000);
-	nvkm_wr32(device, 0x405848, 0xc0000000);
-	nvkm_wr32(device, 0x40584c, 0x00000001);
-	nvkm_wr32(device, 0x404000, 0xc0000000);
-	nvkm_wr32(device, 0x404600, 0xc0000000);
-	nvkm_wr32(device, 0x408030, 0xc0000000);
-	nvkm_wr32(device, 0x404490, 0xc0000000);
-	nvkm_wr32(device, 0x406018, 0xc0000000);
-	nvkm_wr32(device, 0x407020, 0x40000000);
-	nvkm_wr32(device, 0x405840, 0xc0000000);
-	nvkm_wr32(device, 0x405844, 0x00ffffff);
-	nvkm_mask(device, 0x419cc0, 0x00000008, 0x00000008);
-
-	for (gpc = 0; gpc < gr->gpc_nr; gpc++) {
-		for (ppc = 0; ppc < gr->ppc_nr[gpc]; ppc++)
-			nvkm_wr32(device, PPC_UNIT(gpc, ppc, 0x038), 0xc0000000);
-		nvkm_wr32(device, GPC_UNIT(gpc, 0x0420), 0xc0000000);
-		nvkm_wr32(device, GPC_UNIT(gpc, 0x0900), 0xc0000000);
-		nvkm_wr32(device, GPC_UNIT(gpc, 0x1028), 0xc0000000);
-		nvkm_wr32(device, GPC_UNIT(gpc, 0x0824), 0xc0000000);
-		for (tpc = 0; tpc < gr->tpc_nr[gpc]; tpc++) {
-			nvkm_wr32(device, TPC_UNIT(gpc, tpc, 0x508), 0xffffffff);
-			nvkm_wr32(device, TPC_UNIT(gpc, tpc, 0x50c), 0xffffffff);
-			nvkm_wr32(device, TPC_UNIT(gpc, tpc, 0x224), 0xc0000000);
-			nvkm_wr32(device, TPC_UNIT(gpc, tpc, 0x48c), 0xc0000000);
-			nvkm_wr32(device, TPC_UNIT(gpc, tpc, 0x084), 0xc0000000);
-			nvkm_wr32(device, TPC_UNIT(gpc, tpc, 0x430), 0xc0000000);
-			nvkm_wr32(device, TPC_UNIT(gpc, tpc, 0x644), 0x00dffffe);
-			nvkm_wr32(device, TPC_UNIT(gpc, tpc, 0x64c), 0x00000005);
-		}
-		nvkm_wr32(device, GPC_UNIT(gpc, 0x2c90), 0xffffffff);
-		nvkm_wr32(device, GPC_UNIT(gpc, 0x2c94), 0xffffffff);
-	}
-
-	for (rop = 0; rop < gr->rop_nr; rop++) {
-		nvkm_wr32(device, ROP_UNIT(rop, 0x144), 0x40000000);
-		nvkm_wr32(device, ROP_UNIT(rop, 0x070), 0x40000000);
-		nvkm_wr32(device, ROP_UNIT(rop, 0x204), 0xffffffff);
-		nvkm_wr32(device, ROP_UNIT(rop, 0x208), 0xffffffff);
-	}
-
-	nvkm_wr32(device, 0x400108, 0xffffffff);
-	nvkm_wr32(device, 0x400138, 0xffffffff);
-	nvkm_wr32(device, 0x400118, 0xffffffff);
-	nvkm_wr32(device, 0x400130, 0xffffffff);
-	nvkm_wr32(device, 0x40011c, 0xffffffff);
-	nvkm_wr32(device, 0x400134, 0xffffffff);
-
-	nvkm_wr32(device, 0x400054, 0x2c350f63);
-
-	gf100_gr_zbc_init(gr);
-
-	return gm204_gr_init_ctxctl(gr);
-}
-
-static const struct gf100_gr_func
-gm204_gr = {
-	.init = gm204_gr_init,
-	.mmio = gm204_gr_pack_mmio,
-	.ppc_nr = 2,
-	.grctx = &gm204_grctx,
-	.sclass = {
-		{ -1, -1, FERMI_TWOD_A },
-		{ -1, -1, KEPLER_INLINE_TO_MEMORY_B },
-		{ -1, -1, MAXWELL_B, &gf100_fermi },
-		{ -1, -1, MAXWELL_COMPUTE_B },
-		{}
-	}
-};
-
-int
-gm204_gr_new(struct nvkm_device *device, int index, struct nvkm_gr **pgr)
-{
-	return gf100_gr_new_(&gm204_gr, device, index, pgr);
-}
diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/gr/gm20b.c b/drivers/gpu/drm/nouveau/nvkm/engine/gr/gm20b.c
index 65b6e3d1e90d..29732bc14415 100644
--- a/drivers/gpu/drm/nouveau/nvkm/engine/gr/gm20b.c
+++ b/drivers/gpu/drm/nouveau/nvkm/engine/gr/gm20b.c
@@ -32,12 +32,15 @@ gm20b_gr_init_gpc_mmu(struct gf100_gr *gr)
 	struct nvkm_device *device = gr->base.engine.subdev.device;
 	u32 val;
 
-	/* TODO this needs to be removed once secure boot works */
-	if (1) {
+	/* Bypass MMU check for non-secure boot */
+	if (!device->secboot) {
 		nvkm_wr32(device, 0x100ce4, 0xffffffff);
+
+		if (nvkm_rd32(device, 0x100ce4) != 0xffffffff)
+			nvdev_warn(device,
+			  "cannot bypass secure boot - expect failure soon!\n");
 	}
 
-	/* TODO update once secure boot works */
 	val = nvkm_rd32(device, 0x100c80);
 	val &= 0xf000087f;
 	nvkm_wr32(device, 0x418880, val);
@@ -61,7 +64,6 @@ gm20b_gr_set_hww_esr_report_mask(struct gf100_gr *gr)
 
 static const struct gf100_gr_func
 gm20b_gr = {
-	.dtor = gk20a_gr_dtor,
 	.init = gk20a_gr_init,
 	.init_gpc_mmu = gm20b_gr_init_gpc_mmu,
 	.set_hww_esr_report_mask = gm20b_gr_set_hww_esr_report_mask,
@@ -79,5 +81,5 @@ gm20b_gr = {
 int
 gm20b_gr_new(struct nvkm_device *device, int index, struct nvkm_gr **pgr)
 {
-	return gk20a_gr_new_(&gm20b_gr, device, index, pgr);
+	return gm200_gr_new_(&gm20b_gr, device, index, pgr);
 }
diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/msenc/Kbuild b/drivers/gpu/drm/nouveau/nvkm/engine/msenc/Kbuild
new file mode 100644
index 000000000000..b5119564f608
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/nvkm/engine/msenc/Kbuild
@@ -0,0 +1 @@
+#nvkm-y += nvkm/engine/msenc/base.o
diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/nvdec/Kbuild b/drivers/gpu/drm/nouveau/nvkm/engine/nvdec/Kbuild
new file mode 100644
index 000000000000..13b7c71ff900
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/nvkm/engine/nvdec/Kbuild
@@ -0,0 +1 @@
+#nvkm-y += nvkm/engine/nvdec/base.o
diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/nvenc/Kbuild b/drivers/gpu/drm/nouveau/nvkm/engine/nvenc/Kbuild
new file mode 100644
index 000000000000..ad8f1820fa53
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/nvkm/engine/nvenc/Kbuild
@@ -0,0 +1 @@
+#nvkm-y += nvkm/engine/nvenc/base.o
diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/pm/nv40.c b/drivers/gpu/drm/nouveau/nvkm/engine/pm/nv40.c
index 4bef72a9d106..3fda594700e0 100644
--- a/drivers/gpu/drm/nouveau/nvkm/engine/pm/nv40.c
+++ b/drivers/gpu/drm/nouveau/nvkm/engine/pm/nv40.c
@@ -59,9 +59,11 @@ static void
 nv40_perfctr_next(struct nvkm_pm *pm, struct nvkm_perfdom *dom)
 {
 	struct nvkm_device *device = pm->engine.subdev.device;
-	if (pm->sequence != pm->sequence) {
+	struct nv40_pm *nv40pm = container_of(pm, struct nv40_pm, base);
+
+	if (nv40pm->sequence != pm->sequence) {
 		nvkm_wr32(device, 0x400084, 0x00000020);
-		pm->sequence = pm->sequence;
+		nv40pm->sequence = pm->sequence;
 	}
 }
 
diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/vic/Kbuild b/drivers/gpu/drm/nouveau/nvkm/engine/vic/Kbuild
new file mode 100644
index 000000000000..ed4fb6488013
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/nvkm/engine/vic/Kbuild
@@ -0,0 +1 @@
+#nvkm-y += nvkm/engine/vic/base.o
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/Kbuild b/drivers/gpu/drm/nouveau/nvkm/subdev/Kbuild
index ee2c38f50ef5..642d27dc99a3 100644
--- a/drivers/gpu/drm/nouveau/nvkm/subdev/Kbuild
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/Kbuild
@@ -8,6 +8,7 @@ include $(src)/nvkm/subdev/fuse/Kbuild
 include $(src)/nvkm/subdev/gpio/Kbuild
 include $(src)/nvkm/subdev/i2c/Kbuild
 include $(src)/nvkm/subdev/ibus/Kbuild
+include $(src)/nvkm/subdev/iccsense/Kbuild
 include $(src)/nvkm/subdev/instmem/Kbuild
 include $(src)/nvkm/subdev/ltc/Kbuild
 include $(src)/nvkm/subdev/mc/Kbuild
@@ -15,6 +16,7 @@ include $(src)/nvkm/subdev/mmu/Kbuild
 include $(src)/nvkm/subdev/mxm/Kbuild
 include $(src)/nvkm/subdev/pci/Kbuild
 include $(src)/nvkm/subdev/pmu/Kbuild
+include $(src)/nvkm/subdev/secboot/Kbuild
 include $(src)/nvkm/subdev/therm/Kbuild
 include $(src)/nvkm/subdev/timer/Kbuild
 include $(src)/nvkm/subdev/volt/Kbuild
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/bios/Kbuild b/drivers/gpu/drm/nouveau/nvkm/subdev/bios/Kbuild
index 64730d5e9351..dbcb0ef21587 100644
--- a/drivers/gpu/drm/nouveau/nvkm/subdev/bios/Kbuild
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/bios/Kbuild
@@ -10,6 +10,7 @@ nvkm-y += nvkm/subdev/bios/extdev.o
 nvkm-y += nvkm/subdev/bios/fan.o
 nvkm-y += nvkm/subdev/bios/gpio.o
 nvkm-y += nvkm/subdev/bios/i2c.o
+nvkm-y += nvkm/subdev/bios/iccsense.o
 nvkm-y += nvkm/subdev/bios/image.o
 nvkm-y += nvkm/subdev/bios/init.o
 nvkm-y += nvkm/subdev/bios/mxm.o
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/bios/extdev.c b/drivers/gpu/drm/nouveau/nvkm/subdev/bios/extdev.c
index c9e6f6ff7c50..b8578359e61b 100644
--- a/drivers/gpu/drm/nouveau/nvkm/subdev/bios/extdev.c
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/bios/extdev.c
@@ -32,7 +32,7 @@ extdev_table(struct nvkm_bios *bios, u8 *ver, u8 *hdr, u8 *len, u8 *cnt)
 	u16 dcb, extdev = 0;
 
 	dcb = dcb_table(bios, &dcb_ver, &dcb_hdr, &dcb_cnt, &dcb_len);
-	if (!dcb || (dcb_ver != 0x30 && dcb_ver != 0x40))
+	if (!dcb || (dcb_ver != 0x30 && dcb_ver != 0x40 && dcb_ver != 0x41))
 		return 0x0000;
 
 	extdev = nvbios_rd16(bios, dcb + 18);
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/bios/iccsense.c b/drivers/gpu/drm/nouveau/nvkm/subdev/bios/iccsense.c
new file mode 100644
index 000000000000..084328028af1
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/bios/iccsense.c
@@ -0,0 +1,100 @@
+/*
+ * Copyright 2015 Martin Peres
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
+ * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR
+ * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
+ * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
+ * OTHER DEALINGS IN THE SOFTWARE.
+ *
+ * Authors: Martin Peres
+ */
+#include <subdev/bios.h>
+#include <subdev/bios/bit.h>
+#include <subdev/bios/iccsense.h>
+
+static u16
+nvbios_iccsense_table(struct nvkm_bios *bios, u8 *ver, u8 *hdr, u8 *cnt,
+		      u8 *len)
+{
+	struct bit_entry bit_P;
+	u16 iccsense;
+
+	if (bit_entry(bios, 'P', &bit_P) || bit_P.version != 2 ||
+	    bit_P.length < 0x2c)
+		return 0;
+
+	iccsense = nvbios_rd16(bios, bit_P.offset + 0x28);
+	if (!iccsense)
+		return 0;
+
+	*ver = nvbios_rd08(bios, iccsense + 0);
+	switch (*ver) {
+	case 0x10:
+	case 0x20:
+		*hdr = nvbios_rd08(bios, iccsense + 1);
+		*len = nvbios_rd08(bios, iccsense + 2);
+		*cnt = nvbios_rd08(bios, iccsense + 3);
+		return iccsense;
+	default:
+		break;
+	}
+
+	return 0;
+}
+
+int
+nvbios_iccsense_parse(struct nvkm_bios *bios, struct nvbios_iccsense *iccsense)
+{
+	struct nvkm_subdev *subdev = &bios->subdev;
+	u8 ver, hdr, cnt, len, i;
+	u16 table, entry;
+
+	table = nvbios_iccsense_table(bios, &ver, &hdr, &cnt, &len);
+	if (!table || !cnt)
+		return -EINVAL;
+
+	if (ver != 0x10 && ver != 0x20) {
+		nvkm_error(subdev, "ICCSENSE version 0x%02x unknown\n", ver);
+		return -EINVAL;
+	}
+
+	iccsense->nr_entry = cnt;
+	iccsense->rail = kmalloc(sizeof(struct pwr_rail_t) * cnt, GFP_KERNEL);
+	if (!iccsense->rail)
+		return -ENOMEM;
+
+	for (i = 0; i < cnt; ++i) {
+		struct pwr_rail_t *rail = &iccsense->rail[i];
+		entry = table + hdr + i * len;
+
+		switch(ver) {
+		case 0x10:
+			rail->mode = nvbios_rd08(bios, entry + 0x1);
+			rail->extdev_id = nvbios_rd08(bios, entry + 0x2);
+			rail->resistor_mohm = nvbios_rd08(bios, entry + 0x3);
+			rail->rail = nvbios_rd08(bios, entry + 0x4);
+			break;
+		case 0x20:
+			rail->mode = nvbios_rd08(bios, entry);
+			rail->extdev_id = nvbios_rd08(bios, entry + 0x1);
+			rail->resistor_mohm = nvbios_rd08(bios, entry + 0x5);
+			rail->rail = nvbios_rd08(bios, entry + 0x6);
+			break;
+		};
+	}
+
+	return 0;
+}
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/bios/init.c b/drivers/gpu/drm/nouveau/nvkm/subdev/bios/init.c
index a7d69ce7abc1..38ed09fd3d2f 100644
--- a/drivers/gpu/drm/nouveau/nvkm/subdev/bios/init.c
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/bios/init.c
@@ -786,20 +786,20 @@ init_io_flag_condition(struct nvbios_init *init)
 }
 
 /**
- * INIT_DP_CONDITION - opcode 0x3a
+ * INIT_GENERIC_CONDITION - opcode 0x3a
  *
  */
 static void
-init_dp_condition(struct nvbios_init *init)
+init_generic_condition(struct nvbios_init *init)
 {
 	struct nvkm_bios *bios = init->bios;
 	struct nvbios_dpout info;
 	u8  cond = nvbios_rd08(bios, init->offset + 1);
-	u8  unkn = nvbios_rd08(bios, init->offset + 2);
+	u8  size = nvbios_rd08(bios, init->offset + 2);
 	u8  ver, hdr, cnt, len;
 	u16 data;
 
-	trace("DP_CONDITION\t0x%02x 0x%02x\n", cond, unkn);
+	trace("GENERIC_CONDITION\t0x%02x 0x%02x\n", cond, size);
 	init->offset += 3;
 
 	switch (cond) {
@@ -828,7 +828,8 @@ init_dp_condition(struct nvbios_init *init)
 			init_exec_set(init, false);
 		break;
 	default:
-		warn("unknown dp condition 0x%02x\n", cond);
+		warn("INIT_GENERIC_CONDITON: unknown 0x%02x\n", cond);
+		init->offset += size;
 		break;
 	}
 }
@@ -2205,7 +2206,7 @@ static struct nvbios_init_opcode {
 	[0x37] = { init_copy },
 	[0x38] = { init_not },
 	[0x39] = { init_io_flag_condition },
-	[0x3a] = { init_dp_condition },
+	[0x3a] = { init_generic_condition },
 	[0x3b] = { init_io_mask_or },
 	[0x3c] = { init_io_or },
 	[0x47] = { init_andn_reg },
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/clk/Kbuild b/drivers/gpu/drm/nouveau/nvkm/subdev/clk/Kbuild
index ed7717bcc3a1..87d94883f790 100644
--- a/drivers/gpu/drm/nouveau/nvkm/subdev/clk/Kbuild
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/clk/Kbuild
@@ -8,6 +8,7 @@ nvkm-y += nvkm/subdev/clk/mcp77.o
 nvkm-y += nvkm/subdev/clk/gf100.o
 nvkm-y += nvkm/subdev/clk/gk104.o
 nvkm-y += nvkm/subdev/clk/gk20a.o
+nvkm-y += nvkm/subdev/clk/gm20b.o
 
 nvkm-y += nvkm/subdev/clk/pllnv04.o
 nvkm-y += nvkm/subdev/clk/pllgt215.o
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/clk/gk20a.c b/drivers/gpu/drm/nouveau/nvkm/subdev/clk/gk20a.c
index 5da2aa8cc333..5f0ee24e31b8 100644
--- a/drivers/gpu/drm/nouveau/nvkm/subdev/clk/gk20a.c
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/clk/gk20a.c
@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2014, NVIDIA CORPORATION. All rights reserved.
+ * Copyright (c) 2014-2016, NVIDIA CORPORATION. All rights reserved.
  *
  * Permission is hereby granted, free of charge, to any person obtaining a
  * copy of this software and associated documentation files (the "Software"),
@@ -22,19 +22,17 @@
  * Shamelessly ripped off from ChromeOS's gk20a/clk_pllg.c
  *
  */
-#define gk20a_clk(p) container_of((p), struct gk20a_clk, base)
 #include "priv.h"
+#include "gk20a.h"
 
 #include <core/tegra.h>
 #include <subdev/timer.h>
 
-#define MHZ (1000 * 1000)
+#define KHZ (1000)
+#define MHZ (KHZ * 1000)
 
 #define MASK(w)	((1 << w) - 1)
 
-#define SYS_GPCPLL_CFG_BASE			0x00137000
-#define GPC_BCASE_GPCPLL_CFG_BASE		0x00132800
-
 #define GPCPLL_CFG		(SYS_GPCPLL_CFG_BASE + 0)
 #define GPCPLL_CFG_ENABLE	BIT(0)
 #define GPCPLL_CFG_IDDQ		BIT(1)
@@ -56,6 +54,7 @@
 #define GPCPLL_CFG3			(SYS_GPCPLL_CFG_BASE + 0x18)
 #define GPCPLL_CFG3_PLL_STEPB_SHIFT	16
 
+#define GPC_BCASE_GPCPLL_CFG_BASE		0x00132800
 #define GPCPLL_NDIV_SLOWDOWN			(SYS_GPCPLL_CFG_BASE + 0x1c)
 #define GPCPLL_NDIV_SLOWDOWN_NDIV_LO_SHIFT	0
 #define GPCPLL_NDIV_SLOWDOWN_NDIV_MID_SHIFT	8
@@ -75,7 +74,7 @@
 #define GPC2CLK_OUT_VCODIV1		0
 #define GPC2CLK_OUT_VCODIV_MASK		(MASK(GPC2CLK_OUT_VCODIV_WIDTH) << \
 					GPC2CLK_OUT_VCODIV_SHIFT)
-#define	GPC2CLK_OUT_BYPDIV_WIDTH	6
+#define GPC2CLK_OUT_BYPDIV_WIDTH	6
 #define GPC2CLK_OUT_BYPDIV_SHIFT	0
 #define GPC2CLK_OUT_BYPDIV31		0x3c
 #define GPC2CLK_OUT_INIT_MASK	((MASK(GPC2CLK_OUT_SDIV14_INDIV4_WIDTH) << \
@@ -92,45 +91,49 @@
 #define GPC_BCAST_NDIV_SLOWDOWN_DEBUG_PLL_DYNRAMP_DONE_SYNCED_MASK \
 	    (0x1 << GPC_BCAST_NDIV_SLOWDOWN_DEBUG_PLL_DYNRAMP_DONE_SYNCED_SHIFT)
 
-static const u8 pl_to_div[] = {
+static const u8 _pl_to_div[] = {
 /* PL:   0, 1, 2, 3, 4, 5, 6,  7,  8,  9, 10, 11, 12, 13, 14 */
 /* p: */ 1, 2, 3, 4, 5, 6, 8, 10, 12, 16, 12, 16, 20, 24, 32,
 };
 
-/* All frequencies in Mhz */
-struct gk20a_clk_pllg_params {
-	u32 min_vco, max_vco;
-	u32 min_u, max_u;
-	u32 min_m, max_m;
-	u32 min_n, max_n;
-	u32 min_pl, max_pl;
-};
+static u32 pl_to_div(u32 pl)
+{
+	if (pl >= ARRAY_SIZE(_pl_to_div))
+		return 1;
+
+	return _pl_to_div[pl];
+}
+
+static u32 div_to_pl(u32 div)
+{
+	u32 pl;
+
+	for (pl = 0; pl < ARRAY_SIZE(_pl_to_div) - 1; pl++) {
+		if (_pl_to_div[pl] >= div)
+			return pl;
+	}
+
+	return ARRAY_SIZE(_pl_to_div) - 1;
+}
 
 static const struct gk20a_clk_pllg_params gk20a_pllg_params = {
-	.min_vco = 1000, .max_vco = 2064,
-	.min_u = 12, .max_u = 38,
+	.min_vco = 1000000, .max_vco = 2064000,
+	.min_u = 12000, .max_u = 38000,
 	.min_m = 1, .max_m = 255,
 	.min_n = 8, .max_n = 255,
 	.min_pl = 1, .max_pl = 32,
 };
 
-struct gk20a_clk {
-	struct nvkm_clk base;
-	const struct gk20a_clk_pllg_params *params;
-	u32 m, n, pl;
-	u32 parent_rate;
-};
-
 static void
-gk20a_pllg_read_mnp(struct gk20a_clk *clk)
+gk20a_pllg_read_mnp(struct gk20a_clk *clk, struct gk20a_pll *pll)
 {
 	struct nvkm_device *device = clk->base.subdev.device;
 	u32 val;
 
 	val = nvkm_rd32(device, GPCPLL_COEFF);
-	clk->m = (val >> GPCPLL_COEFF_M_SHIFT) & MASK(GPCPLL_COEFF_M_WIDTH);
-	clk->n = (val >> GPCPLL_COEFF_N_SHIFT) & MASK(GPCPLL_COEFF_N_WIDTH);
-	clk->pl = (val >> GPCPLL_COEFF_P_SHIFT) & MASK(GPCPLL_COEFF_P_WIDTH);
+	pll->m = (val >> GPCPLL_COEFF_M_SHIFT) & MASK(GPCPLL_COEFF_M_WIDTH);
+	pll->n = (val >> GPCPLL_COEFF_N_SHIFT) & MASK(GPCPLL_COEFF_N_WIDTH);
+	pll->pl = (val >> GPCPLL_COEFF_P_SHIFT) & MASK(GPCPLL_COEFF_P_WIDTH);
 }
 
 static u32
@@ -139,8 +142,8 @@ gk20a_pllg_calc_rate(struct gk20a_clk *clk)
 	u32 rate;
 	u32 divider;
 
-	rate = clk->parent_rate * clk->n;
-	divider = clk->m * pl_to_div[clk->pl];
+	rate = clk->parent_rate * clk->pll.n;
+	divider = clk->pll.m * clk->pl_to_div(clk->pll.pl);
 
 	return rate / divider / 2;
 }
@@ -152,15 +155,13 @@ gk20a_pllg_calc_mnp(struct gk20a_clk *clk, unsigned long rate)
 	u32 target_clk_f, ref_clk_f, target_freq;
 	u32 min_vco_f, max_vco_f;
 	u32 low_pl, high_pl, best_pl;
-	u32 target_vco_f, vco_f;
+	u32 target_vco_f;
 	u32 best_m, best_n;
-	u32 u_f;
-	u32 m, n, n2;
-	u32 delta, lwv, best_delta = ~0;
+	u32 best_delta = ~0;
 	u32 pl;
 
-	target_clk_f = rate * 2 / MHZ;
-	ref_clk_f = clk->parent_rate / MHZ;
+	target_clk_f = rate * 2 / KHZ;
+	ref_clk_f = clk->parent_rate / KHZ;
 
 	max_vco_f = clk->params->max_vco;
 	min_vco_f = clk->params->min_vco;
@@ -176,33 +177,26 @@ gk20a_pllg_calc_mnp(struct gk20a_clk *clk, unsigned long rate)
 	high_pl = (max_vco_f + target_vco_f - 1) / target_vco_f;
 	high_pl = min(high_pl, clk->params->max_pl);
 	high_pl = max(high_pl, clk->params->min_pl);
+	high_pl = clk->div_to_pl(high_pl);
 
 	/* min_pl <= low_pl <= max_pl */
 	low_pl = min_vco_f / target_vco_f;
 	low_pl = min(low_pl, clk->params->max_pl);
 	low_pl = max(low_pl, clk->params->min_pl);
-
-	/* Find Indices of high_pl and low_pl */
-	for (pl = 0; pl < ARRAY_SIZE(pl_to_div) - 1; pl++) {
-		if (pl_to_div[pl] >= low_pl) {
-			low_pl = pl;
-			break;
-		}
-	}
-	for (pl = 0; pl < ARRAY_SIZE(pl_to_div) - 1; pl++) {
-		if (pl_to_div[pl] >= high_pl) {
-			high_pl = pl;
-			break;
-		}
-	}
+	low_pl = clk->div_to_pl(low_pl);
 
 	nvkm_debug(subdev, "low_PL %d(div%d), high_PL %d(div%d)", low_pl,
-		   pl_to_div[low_pl], high_pl, pl_to_div[high_pl]);
+		   clk->pl_to_div(low_pl), high_pl, clk->pl_to_div(high_pl));
 
 	/* Select lowest possible VCO */
 	for (pl = low_pl; pl <= high_pl; pl++) {
-		target_vco_f = target_clk_f * pl_to_div[pl];
+		u32 m, n, n2;
+
+		target_vco_f = target_clk_f * clk->pl_to_div(pl);
+
 		for (m = clk->params->min_m; m <= clk->params->max_m; m++) {
+			u32 u_f, vco_f;
+
 			u_f = ref_clk_f / m;
 
 			if (u_f < clk->params->min_u)
@@ -225,8 +219,10 @@ gk20a_pllg_calc_mnp(struct gk20a_clk *clk, unsigned long rate)
 				vco_f = ref_clk_f * n / m;
 
 				if (vco_f >= min_vco_f && vco_f <= max_vco_f) {
-					lwv = (vco_f + (pl_to_div[pl] / 2))
-						/ pl_to_div[pl];
+					u32 delta, lwv;
+
+					lwv = (vco_f + (clk->pl_to_div(pl) / 2))
+						/ clk->pl_to_div(pl);
 					delta = abs(lwv - target_clk_f);
 
 					if (delta < best_delta) {
@@ -249,17 +245,18 @@ found_match:
 	if (best_delta != 0)
 		nvkm_debug(subdev,
 			   "no best match for target @ %dMHz on gpc_pll",
-			   target_clk_f);
+			   target_clk_f / KHZ);
 
-	clk->m = best_m;
-	clk->n = best_n;
-	clk->pl = best_pl;
+	clk->pll.m = best_m;
+	clk->pll.n = best_n;
+	clk->pll.pl = best_pl;
 
-	target_freq = gk20a_pllg_calc_rate(clk) / MHZ;
+	target_freq = gk20a_pllg_calc_rate(clk);
 
 	nvkm_debug(subdev,
 		   "actual target freq %d MHz, M %d, N %d, PL %d(div%d)\n",
-		   target_freq, clk->m, clk->n, clk->pl, pl_to_div[clk->pl]);
+		   target_freq / MHZ, clk->pll.m, clk->pll.n, clk->pll.pl,
+		   clk->pl_to_div(clk->pll.pl));
 	return 0;
 }
 
@@ -323,17 +320,19 @@ gk20a_pllg_slide(struct gk20a_clk *clk, u32 n)
 }
 
 static void
-_gk20a_pllg_enable(struct gk20a_clk *clk)
+gk20a_pllg_enable(struct gk20a_clk *clk)
 {
 	struct nvkm_device *device = clk->base.subdev.device;
+
 	nvkm_mask(device, GPCPLL_CFG, GPCPLL_CFG_ENABLE, GPCPLL_CFG_ENABLE);
 	nvkm_rd32(device, GPCPLL_CFG);
 }
 
 static void
-_gk20a_pllg_disable(struct gk20a_clk *clk)
+gk20a_pllg_disable(struct gk20a_clk *clk)
 {
 	struct nvkm_device *device = clk->base.subdev.device;
+
 	nvkm_mask(device, GPCPLL_CFG, GPCPLL_CFG_ENABLE, 0);
 	nvkm_rd32(device, GPCPLL_CFG);
 }
@@ -344,25 +343,26 @@ _gk20a_pllg_program_mnp(struct gk20a_clk *clk, bool allow_slide)
 	struct nvkm_subdev *subdev = &clk->base.subdev;
 	struct nvkm_device *device = subdev->device;
 	u32 val, cfg;
-	u32 m_old, pl_old, n_lo;
+	struct gk20a_pll old_pll;
+	u32 n_lo;
 
 	/* get old coefficients */
-	val = nvkm_rd32(device, GPCPLL_COEFF);
-	m_old = (val >> GPCPLL_COEFF_M_SHIFT) & MASK(GPCPLL_COEFF_M_WIDTH);
-	pl_old = (val >> GPCPLL_COEFF_P_SHIFT) & MASK(GPCPLL_COEFF_P_WIDTH);
+	gk20a_pllg_read_mnp(clk, &old_pll);
 
 	/* do NDIV slide if there is no change in M and PL */
 	cfg = nvkm_rd32(device, GPCPLL_CFG);
-	if (allow_slide && clk->m == m_old && clk->pl == pl_old &&
-	    (cfg & GPCPLL_CFG_ENABLE)) {
-		return gk20a_pllg_slide(clk, clk->n);
+	if (allow_slide && clk->pll.m == old_pll.m &&
+	    clk->pll.pl == old_pll.pl && (cfg & GPCPLL_CFG_ENABLE)) {
+		return gk20a_pllg_slide(clk, clk->pll.n);
 	}
 
 	/* slide down to NDIV_LO */
-	n_lo = DIV_ROUND_UP(m_old * clk->params->min_vco,
-			    clk->parent_rate / MHZ);
 	if (allow_slide && (cfg & GPCPLL_CFG_ENABLE)) {
-		int ret = gk20a_pllg_slide(clk, n_lo);
+		int ret;
+
+		n_lo = DIV_ROUND_UP(old_pll.m * clk->params->min_vco,
+				    clk->parent_rate / KHZ);
+		ret = gk20a_pllg_slide(clk, n_lo);
 
 		if (ret)
 			return ret;
@@ -387,19 +387,19 @@ _gk20a_pllg_program_mnp(struct gk20a_clk *clk, bool allow_slide)
 		udelay(2);
 	}
 
-	_gk20a_pllg_disable(clk);
+	gk20a_pllg_disable(clk);
 
 	nvkm_debug(subdev, "%s: m=%d n=%d pl=%d\n", __func__,
-		   clk->m, clk->n, clk->pl);
+		   clk->pll.m, clk->pll.n, clk->pll.pl);
 
-	n_lo = DIV_ROUND_UP(clk->m * clk->params->min_vco,
-			    clk->parent_rate / MHZ);
-	val = clk->m << GPCPLL_COEFF_M_SHIFT;
-	val |= (allow_slide ? n_lo : clk->n) << GPCPLL_COEFF_N_SHIFT;
-	val |= clk->pl << GPCPLL_COEFF_P_SHIFT;
+	n_lo = DIV_ROUND_UP(clk->pll.m * clk->params->min_vco,
+			    clk->parent_rate / KHZ);
+	val = clk->pll.m << GPCPLL_COEFF_M_SHIFT;
+	val |= (allow_slide ? n_lo : clk->pll.n) << GPCPLL_COEFF_N_SHIFT;
+	val |= clk->pll.pl << GPCPLL_COEFF_P_SHIFT;
 	nvkm_wr32(device, GPCPLL_COEFF, val);
 
-	_gk20a_pllg_enable(clk);
+	gk20a_pllg_enable(clk);
 
 	val = nvkm_rd32(device, GPCPLL_CFG);
 	if (val & GPCPLL_CFG_LOCK_DET_OFF) {
@@ -414,16 +414,24 @@ _gk20a_pllg_program_mnp(struct gk20a_clk *clk, bool allow_slide)
 		return -ETIMEDOUT;
 
 	/* switch to VCO mode */
-	nvkm_mask(device, SEL_VCO, 0, BIT(SEL_VCO_GPC2CLK_OUT_SHIFT));
+	nvkm_mask(device, SEL_VCO, BIT(SEL_VCO_GPC2CLK_OUT_SHIFT),
+		  BIT(SEL_VCO_GPC2CLK_OUT_SHIFT));
 
 	/* restore out divider 1:1 */
 	val = nvkm_rd32(device, GPC2CLK_OUT);
-	val &= ~GPC2CLK_OUT_VCODIV_MASK;
-	udelay(2);
-	nvkm_wr32(device, GPC2CLK_OUT, val);
+	if ((val & GPC2CLK_OUT_VCODIV_MASK) !=
+	    (GPC2CLK_OUT_VCODIV1 << GPC2CLK_OUT_VCODIV_SHIFT)) {
+		val &= ~GPC2CLK_OUT_VCODIV_MASK;
+		val |= GPC2CLK_OUT_VCODIV1 << GPC2CLK_OUT_VCODIV_SHIFT;
+		udelay(2);
+		nvkm_wr32(device, GPC2CLK_OUT, val);
+		/* Intentional 2nd write to assure linear divider operation */
+		nvkm_wr32(device, GPC2CLK_OUT, val);
+		nvkm_rd32(device, GPC2CLK_OUT);
+	}
 
 	/* slide up to new NDIV */
-	return allow_slide ? gk20a_pllg_slide(clk, clk->n) : 0;
+	return allow_slide ? gk20a_pllg_slide(clk, clk->pll.n) : 0;
 }
 
 static int
@@ -438,32 +446,6 @@ gk20a_pllg_program_mnp(struct gk20a_clk *clk)
 	return err;
 }
 
-static void
-gk20a_pllg_disable(struct gk20a_clk *clk)
-{
-	struct nvkm_device *device = clk->base.subdev.device;
-	u32 val;
-
-	/* slide to VCO min */
-	val = nvkm_rd32(device, GPCPLL_CFG);
-	if (val & GPCPLL_CFG_ENABLE) {
-		u32 coeff, m, n_lo;
-
-		coeff = nvkm_rd32(device, GPCPLL_COEFF);
-		m = (coeff >> GPCPLL_COEFF_M_SHIFT) & MASK(GPCPLL_COEFF_M_WIDTH);
-		n_lo = DIV_ROUND_UP(m * clk->params->min_vco,
-				    clk->parent_rate / MHZ);
-		gk20a_pllg_slide(clk, n_lo);
-	}
-
-	/* put PLL in bypass before disabling it */
-	nvkm_mask(device, SEL_VCO, BIT(SEL_VCO_GPC2CLK_OUT_SHIFT), 0);
-
-	_gk20a_pllg_disable(clk);
-}
-
-#define GK20A_CLK_GPC_MDIV 1000
-
 static struct nvkm_pstate
 gk20a_pstates[] = {
 	{
@@ -558,7 +540,7 @@ gk20a_pstates[] = {
 	},
 };
 
-static int
+int
 gk20a_clk_read(struct nvkm_clk *base, enum nv_clk_src src)
 {
 	struct gk20a_clk *clk = gk20a_clk(base);
@@ -569,7 +551,7 @@ gk20a_clk_read(struct nvkm_clk *base, enum nv_clk_src src)
 	case nv_clk_src_crystal:
 		return device->crystal;
 	case nv_clk_src_gpc:
-		gk20a_pllg_read_mnp(clk);
+		gk20a_pllg_read_mnp(clk, &clk->pll);
 		return gk20a_pllg_calc_rate(clk) / GK20A_CLK_GPC_MDIV;
 	default:
 		nvkm_error(subdev, "invalid clock source %d\n", src);
@@ -577,7 +559,7 @@ gk20a_clk_read(struct nvkm_clk *base, enum nv_clk_src src)
 	}
 }
 
-static int
+int
 gk20a_clk_calc(struct nvkm_clk *base, struct nvkm_cstate *cstate)
 {
 	struct gk20a_clk *clk = gk20a_clk(base);
@@ -586,7 +568,7 @@ gk20a_clk_calc(struct nvkm_clk *base, struct nvkm_cstate *cstate)
 					 GK20A_CLK_GPC_MDIV);
 }
 
-static int
+int
 gk20a_clk_prog(struct nvkm_clk *base)
 {
 	struct gk20a_clk *clk = gk20a_clk(base);
@@ -594,15 +576,33 @@ gk20a_clk_prog(struct nvkm_clk *base)
 	return gk20a_pllg_program_mnp(clk);
 }
 
-static void
+void
 gk20a_clk_tidy(struct nvkm_clk *base)
 {
 }
 
-static void
+void
 gk20a_clk_fini(struct nvkm_clk *base)
 {
+	struct nvkm_device *device = base->subdev.device;
 	struct gk20a_clk *clk = gk20a_clk(base);
+	u32 val;
+
+	/* slide to VCO min */
+	val = nvkm_rd32(device, GPCPLL_CFG);
+	if (val & GPCPLL_CFG_ENABLE) {
+		struct gk20a_pll pll;
+		u32 n_lo;
+
+		gk20a_pllg_read_mnp(clk, &pll);
+		n_lo = DIV_ROUND_UP(pll.m * clk->params->min_vco,
+				    clk->parent_rate / KHZ);
+		gk20a_pllg_slide(clk, n_lo);
+	}
+
+	/* put PLL in bypass before disabling it */
+	nvkm_mask(device, SEL_VCO, BIT(SEL_VCO_GPC2CLK_OUT_SHIFT), 0);
+
 	gk20a_pllg_disable(clk);
 }
 
@@ -614,9 +614,12 @@ gk20a_clk_init(struct nvkm_clk *base)
 	struct nvkm_device *device = subdev->device;
 	int ret;
 
-	nvkm_mask(device, GPC2CLK_OUT, GPC2CLK_OUT_INIT_MASK, GPC2CLK_OUT_INIT_VAL);
+	nvkm_mask(device, GPC2CLK_OUT, GPC2CLK_OUT_INIT_MASK,
+		  GPC2CLK_OUT_INIT_VAL);
 
-	ret = gk20a_clk_prog(&clk->base);
+	/* Start with lowest frequency */
+	base->func->calc(base, &base->func->pstates[0].base);
+	ret = base->func->prog(&clk->base);
 	if (ret) {
 		nvkm_error(subdev, "cannot initialize clock\n");
 		return ret;
@@ -643,27 +646,50 @@ gk20a_clk = {
 };
 
 int
-gk20a_clk_new(struct nvkm_device *device, int index, struct nvkm_clk **pclk)
+_gk20a_clk_ctor(struct nvkm_device *device, int index,
+		const struct nvkm_clk_func *func,
+		const struct gk20a_clk_pllg_params *params,
+		struct gk20a_clk *clk)
 {
 	struct nvkm_device_tegra *tdev = device->func->tegra(device);
+	int ret;
+	int i;
+
+	/* Finish initializing the pstates */
+	for (i = 0; i < func->nr_pstates; i++) {
+		INIT_LIST_HEAD(&func->pstates[i].list);
+		func->pstates[i].pstate = i + 1;
+	}
+
+	clk->params = params;
+	clk->parent_rate = clk_get_rate(tdev->clk);
+
+	ret = nvkm_clk_ctor(func, device, index, true, &clk->base);
+	if (ret)
+		return ret;
+
+	nvkm_debug(&clk->base.subdev, "parent clock rate: %d Khz\n",
+		   clk->parent_rate / KHZ);
+
+	return 0;
+}
+
+int
+gk20a_clk_new(struct nvkm_device *device, int index, struct nvkm_clk **pclk)
+{
 	struct gk20a_clk *clk;
-	int ret, i;
+	int ret;
 
-	if (!(clk = kzalloc(sizeof(*clk), GFP_KERNEL)))
+	clk = kzalloc(sizeof(*clk), GFP_KERNEL);
+	if (!clk)
 		return -ENOMEM;
 	*pclk = &clk->base;
 
-	/* Finish initializing the pstates */
-	for (i = 0; i < ARRAY_SIZE(gk20a_pstates); i++) {
-		INIT_LIST_HEAD(&gk20a_pstates[i].list);
-		gk20a_pstates[i].pstate = i + 1;
-	}
+	ret = _gk20a_clk_ctor(device, index, &gk20a_clk, &gk20a_pllg_params,
+			      clk);
 
-	clk->params = &gk20a_pllg_params;
-	clk->parent_rate = clk_get_rate(tdev->clk);
+	clk->pl_to_div = pl_to_div;
+	clk->div_to_pl = div_to_pl;
 
-	ret = nvkm_clk_ctor(&gk20a_clk, device, index, true, &clk->base);
-	nvkm_info(&clk->base.subdev, "parent clock rate: %d Mhz\n",
-		  clk->parent_rate / MHZ);
 	return ret;
 }
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/clk/gk20a.h b/drivers/gpu/drm/nouveau/nvkm/subdev/clk/gk20a.h
new file mode 100644
index 000000000000..13c46740197d
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/clk/gk20a.h
@@ -0,0 +1,65 @@
+/*
+ * Copyright (c) 2016, NVIDIA CORPORATION. All rights reserved.
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
+ * THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+ * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
+ * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER
+ * DEALINGS IN THE SOFTWARE.
+ *
+ */
+
+#ifndef __NVKM_CLK_GK20A_H__
+#define __NVKM_CLK_GK20A_H__
+
+#define GK20A_CLK_GPC_MDIV 1000
+
+#define SYS_GPCPLL_CFG_BASE	0x00137000
+
+/* All frequencies in Khz */
+struct gk20a_clk_pllg_params {
+	u32 min_vco, max_vco;
+	u32 min_u, max_u;
+	u32 min_m, max_m;
+	u32 min_n, max_n;
+	u32 min_pl, max_pl;
+};
+
+struct gk20a_pll {
+	u32 m;
+	u32 n;
+	u32 pl;
+};
+
+struct gk20a_clk {
+	struct nvkm_clk base;
+	const struct gk20a_clk_pllg_params *params;
+	struct gk20a_pll pll;
+	u32 parent_rate;
+
+	u32 (*div_to_pl)(u32);
+	u32 (*pl_to_div)(u32);
+};
+#define gk20a_clk(p) container_of((p), struct gk20a_clk, base)
+
+int _gk20a_clk_ctor(struct nvkm_device *, int, const struct nvkm_clk_func *,
+		    const struct gk20a_clk_pllg_params *, struct gk20a_clk *);
+void gk20a_clk_fini(struct nvkm_clk *);
+int gk20a_clk_read(struct nvkm_clk *, enum nv_clk_src);
+int gk20a_clk_calc(struct nvkm_clk *, struct nvkm_cstate *);
+int gk20a_clk_prog(struct nvkm_clk *);
+void gk20a_clk_tidy(struct nvkm_clk *);
+
+#endif
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/clk/gm20b.c b/drivers/gpu/drm/nouveau/nvkm/subdev/clk/gm20b.c
new file mode 100644
index 000000000000..71b2bbb61973
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/clk/gm20b.c
@@ -0,0 +1,198 @@
+/*
+ * Copyright (c) 2016, NVIDIA CORPORATION. All rights reserved.
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
+ * THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+ * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
+ * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER
+ * DEALINGS IN THE SOFTWARE.
+ */
+
+#include <subdev/clk.h>
+#include <core/device.h>
+
+#include "priv.h"
+#include "gk20a.h"
+
+#define KHZ (1000)
+#define MHZ (KHZ * 1000)
+
+#define MASK(w)	((1 << w) - 1)
+
+#define BYPASSCTRL_SYS	(SYS_GPCPLL_CFG_BASE + 0x340)
+#define BYPASSCTRL_SYS_GPCPLL_SHIFT	0
+#define BYPASSCTRL_SYS_GPCPLL_WIDTH	1
+
+static u32 pl_to_div(u32 pl)
+{
+	return pl;
+}
+
+static u32 div_to_pl(u32 div)
+{
+	return div;
+}
+
+static const struct gk20a_clk_pllg_params gm20b_pllg_params = {
+	.min_vco = 1300000, .max_vco = 2600000,
+	.min_u = 12000, .max_u = 38400,
+	.min_m = 1, .max_m = 255,
+	.min_n = 8, .max_n = 255,
+	.min_pl = 1, .max_pl = 31,
+};
+
+static struct nvkm_pstate
+gm20b_pstates[] = {
+	{
+		.base = {
+			.domain[nv_clk_src_gpc] = 76800,
+			.voltage = 0,
+		},
+	},
+	{
+		.base = {
+			.domain[nv_clk_src_gpc] = 153600,
+			.voltage = 1,
+		},
+	},
+	{
+		.base = {
+			.domain[nv_clk_src_gpc] = 230400,
+			.voltage = 2,
+		},
+	},
+	{
+		.base = {
+			.domain[nv_clk_src_gpc] = 307200,
+			.voltage = 3,
+		},
+	},
+	{
+		.base = {
+			.domain[nv_clk_src_gpc] = 384000,
+			.voltage = 4,
+		},
+	},
+	{
+		.base = {
+			.domain[nv_clk_src_gpc] = 460800,
+			.voltage = 5,
+		},
+	},
+	{
+		.base = {
+			.domain[nv_clk_src_gpc] = 537600,
+			.voltage = 6,
+		},
+	},
+	{
+		.base = {
+			.domain[nv_clk_src_gpc] = 614400,
+			.voltage = 7,
+		},
+	},
+	{
+		.base = {
+			.domain[nv_clk_src_gpc] = 691200,
+			.voltage = 8,
+		},
+	},
+	{
+		.base = {
+			.domain[nv_clk_src_gpc] = 768000,
+			.voltage = 9,
+		},
+	},
+	{
+		.base = {
+			.domain[nv_clk_src_gpc] = 844800,
+			.voltage = 10,
+		},
+	},
+	{
+		.base = {
+			.domain[nv_clk_src_gpc] = 921600,
+			.voltage = 11,
+		},
+	},
+	{
+		.base = {
+			.domain[nv_clk_src_gpc] = 998400,
+			.voltage = 12,
+		},
+	},
+
+};
+
+static int
+gm20b_clk_init(struct nvkm_clk *base)
+{
+	struct gk20a_clk *clk = gk20a_clk(base);
+	struct nvkm_subdev *subdev = &clk->base.subdev;
+	struct nvkm_device *device = subdev->device;
+	int ret;
+
+	/* Set the global bypass control to VCO */
+	nvkm_mask(device, BYPASSCTRL_SYS,
+	       MASK(BYPASSCTRL_SYS_GPCPLL_WIDTH) << BYPASSCTRL_SYS_GPCPLL_SHIFT,
+	       0);
+
+	/* Start with lowest frequency */
+	base->func->calc(base, &base->func->pstates[0].base);
+	ret = base->func->prog(&clk->base);
+	if (ret) {
+		nvkm_error(subdev, "cannot initialize clock\n");
+		return ret;
+	}
+
+	return 0;
+}
+
+static const struct nvkm_clk_func
+gm20b_clk_speedo0 = {
+	.init = gm20b_clk_init,
+	.fini = gk20a_clk_fini,
+	.read = gk20a_clk_read,
+	.calc = gk20a_clk_calc,
+	.prog = gk20a_clk_prog,
+	.tidy = gk20a_clk_tidy,
+	.pstates = gm20b_pstates,
+	.nr_pstates = ARRAY_SIZE(gm20b_pstates) - 1,
+	.domains = {
+		{ nv_clk_src_crystal, 0xff },
+		{ nv_clk_src_gpc, 0xff, 0, "core", GK20A_CLK_GPC_MDIV },
+		{ nv_clk_src_max },
+	},
+};
+
+int
+gm20b_clk_new(struct nvkm_device *device, int index, struct nvkm_clk **pclk)
+{
+	struct gk20a_clk *clk;
+	int ret;
+
+	clk = kzalloc(sizeof(*clk), GFP_KERNEL);
+	if (!clk)
+		return -ENOMEM;
+	*pclk = &clk->base;
+
+	ret = _gk20a_clk_ctor(device, index, &gm20b_clk_speedo0,
+			      &gm20b_pllg_params, clk);
+
+	clk->pl_to_div = pl_to_div;
+	clk->div_to_pl = div_to_pl;
+
+	return ret;
+}
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/devinit/Kbuild b/drivers/gpu/drm/nouveau/nvkm/subdev/devinit/Kbuild
index 793e73d16dac..eac88e3dc6e5 100644
--- a/drivers/gpu/drm/nouveau/nvkm/subdev/devinit/Kbuild
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/devinit/Kbuild
@@ -11,4 +11,4 @@ nvkm-y += nvkm/subdev/devinit/gt215.o
 nvkm-y += nvkm/subdev/devinit/mcp89.o
 nvkm-y += nvkm/subdev/devinit/gf100.o
 nvkm-y += nvkm/subdev/devinit/gm107.o
-nvkm-y += nvkm/subdev/devinit/gm204.o
+nvkm-y += nvkm/subdev/devinit/gm200.o
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/devinit/gf100.c b/drivers/gpu/drm/nouveau/nvkm/subdev/devinit/gf100.c
index 22b0140e28c6..2923598b5fe9 100644
--- a/drivers/gpu/drm/nouveau/nvkm/subdev/devinit/gf100.c
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/devinit/gf100.c
@@ -90,9 +90,21 @@ gf100_devinit_disable(struct nvkm_devinit *init)
 	return disable;
 }
 
+void
+gf100_devinit_preinit(struct nvkm_devinit *base)
+{
+	struct nv50_devinit *init = nv50_devinit(base);
+	struct nvkm_subdev *subdev = &init->base.subdev;
+	struct nvkm_device *device = subdev->device;
+
+	/* This bit is set by devinit, and flips back to 0 on suspend */
+	if (!base->post)
+		base->post = ((nvkm_rd32(device, 0x2240c) & BIT(1)) == 0);
+}
+
 static const struct nvkm_devinit_func
 gf100_devinit = {
-	.preinit = nv50_devinit_preinit,
+	.preinit = gf100_devinit_preinit,
 	.init = nv50_devinit_init,
 	.post = nv04_devinit_post,
 	.pll_set = gf100_devinit_pll_set,
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/devinit/gm107.c b/drivers/gpu/drm/nouveau/nvkm/subdev/devinit/gm107.c
index 2be98bd78214..28ca01be3d38 100644
--- a/drivers/gpu/drm/nouveau/nvkm/subdev/devinit/gm107.c
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/devinit/gm107.c
@@ -46,7 +46,7 @@ gm107_devinit_disable(struct nvkm_devinit *init)
 
 static const struct nvkm_devinit_func
 gm107_devinit = {
-	.preinit = nv50_devinit_preinit,
+	.preinit = gf100_devinit_preinit,
 	.init = nv50_devinit_init,
 	.post = nv04_devinit_post,
 	.pll_set = gf100_devinit_pll_set,
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/devinit/gm204.c b/drivers/gpu/drm/nouveau/nvkm/subdev/devinit/gm200.c
index 2b9c3f11b7a8..a410c0db8a08 100644
--- a/drivers/gpu/drm/nouveau/nvkm/subdev/devinit/gm204.c
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/devinit/gm200.c
@@ -107,7 +107,7 @@ pmu_load(struct nv50_devinit *init, u8 type, bool post,
 }
 
 static int
-gm204_devinit_post(struct nvkm_devinit *base, bool post)
+gm200_devinit_post(struct nvkm_devinit *base, bool post)
 {
 	struct nv50_devinit *init = nv50_devinit(base);
 	struct nvkm_subdev *subdev = &init->base.subdev;
@@ -165,17 +165,17 @@ gm204_devinit_post(struct nvkm_devinit *base, bool post)
 }
 
 static const struct nvkm_devinit_func
-gm204_devinit = {
-	.preinit = nv50_devinit_preinit,
+gm200_devinit = {
+	.preinit = gf100_devinit_preinit,
 	.init = nv50_devinit_init,
-	.post = gm204_devinit_post,
+	.post = gm200_devinit_post,
 	.pll_set = gf100_devinit_pll_set,
 	.disable = gm107_devinit_disable,
 };
 
 int
-gm204_devinit_new(struct nvkm_device *device, int index,
+gm200_devinit_new(struct nvkm_device *device, int index,
 		struct nvkm_devinit **pinit)
 {
-	return nv50_devinit_new_(&gm204_devinit, device, index, pinit);
+	return nv50_devinit_new_(&gm200_devinit, device, index, pinit);
 }
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/devinit/nv50.c b/drivers/gpu/drm/nouveau/nvkm/subdev/devinit/nv50.c
index 337c2c692dc7..c714b097719c 100644
--- a/drivers/gpu/drm/nouveau/nvkm/subdev/devinit/nv50.c
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/devinit/nv50.c
@@ -93,28 +93,27 @@ nv50_devinit_disable(struct nvkm_devinit *init)
 void
 nv50_devinit_preinit(struct nvkm_devinit *base)
 {
-	struct nv50_devinit *init = nv50_devinit(base);
-	struct nvkm_subdev *subdev = &init->base.subdev;
+	struct nvkm_subdev *subdev = &base->subdev;
 	struct nvkm_device *device = subdev->device;
 
 	/* our heuristics can't detect whether the board has had its
 	 * devinit scripts executed or not if the display engine is
 	 * missing, assume it's a secondary gpu which requires post
 	 */
-	if (!init->base.post) {
-		u64 disable = nvkm_devinit_disable(&init->base);
+	if (!base->post) {
+		u64 disable = nvkm_devinit_disable(base);
 		if (disable & (1ULL << NVKM_ENGINE_DISP))
-			init->base.post = true;
+			base->post = true;
 	}
 
 	/* magic to detect whether or not x86 vbios code has executed
 	 * the devinit scripts to initialise the board
 	 */
-	if (!init->base.post) {
+	if (!base->post) {
 		if (!nvkm_rdvgac(device, 0, 0x00) &&
 		    !nvkm_rdvgac(device, 0, 0x1a)) {
 			nvkm_debug(subdev, "adaptor not initialised\n");
-			init->base.post = true;
+			base->post = true;
 		}
 	}
 }
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/devinit/nv50.h b/drivers/gpu/drm/nouveau/nvkm/subdev/devinit/nv50.h
index 5de70a8486b4..25d2ae3af1c6 100644
--- a/drivers/gpu/drm/nouveau/nvkm/subdev/devinit/nv50.h
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/devinit/nv50.h
@@ -20,6 +20,7 @@ int  gf100_devinit_ctor(struct nvkm_object *, struct nvkm_object *,
 			struct nvkm_oclass *, void *, u32,
 			struct nvkm_object **);
 int  gf100_devinit_pll_set(struct nvkm_devinit *, u32, u32);
+void gf100_devinit_preinit(struct nvkm_devinit *);
 
 u64  gm107_devinit_disable(struct nvkm_devinit *);
 #endif
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/i2c/Kbuild b/drivers/gpu/drm/nouveau/nvkm/subdev/i2c/Kbuild
index 1f730613c237..48f01e40b8fc 100644
--- a/drivers/gpu/drm/nouveau/nvkm/subdev/i2c/Kbuild
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/i2c/Kbuild
@@ -6,7 +6,7 @@ nvkm-y += nvkm/subdev/i2c/g94.o
 nvkm-y += nvkm/subdev/i2c/gf117.o
 nvkm-y += nvkm/subdev/i2c/gf119.o
 nvkm-y += nvkm/subdev/i2c/gk104.o
-nvkm-y += nvkm/subdev/i2c/gm204.o
+nvkm-y += nvkm/subdev/i2c/gm200.o
 
 nvkm-y += nvkm/subdev/i2c/pad.o
 nvkm-y += nvkm/subdev/i2c/padnv04.o
@@ -14,7 +14,7 @@ nvkm-y += nvkm/subdev/i2c/padnv4e.o
 nvkm-y += nvkm/subdev/i2c/padnv50.o
 nvkm-y += nvkm/subdev/i2c/padg94.o
 nvkm-y += nvkm/subdev/i2c/padgf119.o
-nvkm-y += nvkm/subdev/i2c/padgm204.o
+nvkm-y += nvkm/subdev/i2c/padgm200.o
 
 nvkm-y += nvkm/subdev/i2c/bus.o
 nvkm-y += nvkm/subdev/i2c/busnv04.o
@@ -25,6 +25,6 @@ nvkm-y += nvkm/subdev/i2c/bit.o
 
 nvkm-y += nvkm/subdev/i2c/aux.o
 nvkm-y += nvkm/subdev/i2c/auxg94.o
-nvkm-y += nvkm/subdev/i2c/auxgm204.o
+nvkm-y += nvkm/subdev/i2c/auxgm200.o
 
 nvkm-y += nvkm/subdev/i2c/anx9805.o
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/i2c/aux.h b/drivers/gpu/drm/nouveau/nvkm/subdev/i2c/aux.h
index 35a892e4a4c3..fc6b162fa0b1 100644
--- a/drivers/gpu/drm/nouveau/nvkm/subdev/i2c/aux.h
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/i2c/aux.h
@@ -18,7 +18,7 @@ int nvkm_i2c_aux_xfer(struct nvkm_i2c_aux *, bool retry, u8 type,
 		      u32 addr, u8 *data, u8 size);
 
 int g94_i2c_aux_new(struct nvkm_i2c_pad *, int, u8, struct nvkm_i2c_aux **);
-int gm204_i2c_aux_new(struct nvkm_i2c_pad *, int, u8, struct nvkm_i2c_aux **);
+int gm200_i2c_aux_new(struct nvkm_i2c_pad *, int, u8, struct nvkm_i2c_aux **);
 
 #define AUX_MSG(b,l,f,a...) do {                                               \
 	struct nvkm_i2c_aux *_aux = (b);                                       \
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/i2c/auxgm204.c b/drivers/gpu/drm/nouveau/nvkm/subdev/i2c/auxgm200.c
index bed231b56dbd..61d729b82c69 100644
--- a/drivers/gpu/drm/nouveau/nvkm/subdev/i2c/auxgm204.c
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/i2c/auxgm200.c
@@ -21,23 +21,23 @@
  *
  * Authors: Ben Skeggs <bskeggs@redhat.com>
  */
-#define gm204_i2c_aux(p) container_of((p), struct gm204_i2c_aux, base)
+#define gm200_i2c_aux(p) container_of((p), struct gm200_i2c_aux, base)
 #include "aux.h"
 
-struct gm204_i2c_aux {
+struct gm200_i2c_aux {
 	struct nvkm_i2c_aux base;
 	int ch;
 };
 
 static void
-gm204_i2c_aux_fini(struct gm204_i2c_aux *aux)
+gm200_i2c_aux_fini(struct gm200_i2c_aux *aux)
 {
 	struct nvkm_device *device = aux->base.pad->i2c->subdev.device;
 	nvkm_mask(device, 0x00d954 + (aux->ch * 0x50), 0x00310000, 0x00000000);
 }
 
 static int
-gm204_i2c_aux_init(struct gm204_i2c_aux *aux)
+gm200_i2c_aux_init(struct gm200_i2c_aux *aux)
 {
 	struct nvkm_device *device = aux->base.pad->i2c->subdev.device;
 	const u32 unksel = 1; /* nfi which to use, or if it matters.. */
@@ -64,7 +64,7 @@ gm204_i2c_aux_init(struct gm204_i2c_aux *aux)
 		udelay(1);
 		if (!timeout--) {
 			AUX_ERR(&aux->base, "magic wait %08x", ctrl);
-			gm204_i2c_aux_fini(aux);
+			gm200_i2c_aux_fini(aux);
 			return -EBUSY;
 		}
 	} while ((ctrl & 0x03000000) != urep);
@@ -73,10 +73,10 @@ gm204_i2c_aux_init(struct gm204_i2c_aux *aux)
 }
 
 static int
-gm204_i2c_aux_xfer(struct nvkm_i2c_aux *obj, bool retry,
+gm200_i2c_aux_xfer(struct nvkm_i2c_aux *obj, bool retry,
 		   u8 type, u32 addr, u8 *data, u8 size)
 {
-	struct gm204_i2c_aux *aux = gm204_i2c_aux(obj);
+	struct gm200_i2c_aux *aux = gm200_i2c_aux(obj);
 	struct nvkm_device *device = aux->base.pad->i2c->subdev.device;
 	const u32 base = aux->ch * 0x50;
 	u32 ctrl, stat, timeout, retries;
@@ -85,7 +85,7 @@ gm204_i2c_aux_xfer(struct nvkm_i2c_aux *obj, bool retry,
 
 	AUX_TRACE(&aux->base, "%d: %08x %d", type, addr, size);
 
-	ret = gm204_i2c_aux_init(aux);
+	ret = gm200_i2c_aux_init(aux);
 	if (ret < 0)
 		goto out;
 
@@ -155,26 +155,26 @@ gm204_i2c_aux_xfer(struct nvkm_i2c_aux *obj, bool retry,
 	}
 
 out:
-	gm204_i2c_aux_fini(aux);
+	gm200_i2c_aux_fini(aux);
 	return ret < 0 ? ret : (stat & 0x000f0000) >> 16;
 }
 
 static const struct nvkm_i2c_aux_func
-gm204_i2c_aux_func = {
-	.xfer = gm204_i2c_aux_xfer,
+gm200_i2c_aux_func = {
+	.xfer = gm200_i2c_aux_xfer,
 };
 
 int
-gm204_i2c_aux_new(struct nvkm_i2c_pad *pad, int index, u8 drive,
+gm200_i2c_aux_new(struct nvkm_i2c_pad *pad, int index, u8 drive,
 		struct nvkm_i2c_aux **paux)
 {
-	struct gm204_i2c_aux *aux;
+	struct gm200_i2c_aux *aux;
 
 	if (!(aux = kzalloc(sizeof(*aux), GFP_KERNEL)))
 		return -ENOMEM;
 	*paux = &aux->base;
 
-	nvkm_i2c_aux_ctor(&gm204_i2c_aux_func, pad, index, &aux->base);
+	nvkm_i2c_aux_ctor(&gm200_i2c_aux_func, pad, index, &aux->base);
 	aux->ch = drive;
 	aux->base.intr = 1 << aux->ch;
 	return 0;
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/i2c/gm204.c b/drivers/gpu/drm/nouveau/nvkm/subdev/i2c/gm200.c
index ff9f7d62f6be..a23c5f315221 100644
--- a/drivers/gpu/drm/nouveau/nvkm/subdev/i2c/gm204.c
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/i2c/gm200.c
@@ -25,16 +25,16 @@
 #include "pad.h"
 
 static const struct nvkm_i2c_func
-gm204_i2c = {
+gm200_i2c = {
 	.pad_x_new = gf119_i2c_pad_x_new,
-	.pad_s_new = gm204_i2c_pad_s_new,
+	.pad_s_new = gm200_i2c_pad_s_new,
 	.aux = 8,
 	.aux_stat = gk104_aux_stat,
 	.aux_mask = gk104_aux_mask,
 };
 
 int
-gm204_i2c_new(struct nvkm_device *device, int index, struct nvkm_i2c **pi2c)
+gm200_i2c_new(struct nvkm_device *device, int index, struct nvkm_i2c **pi2c)
 {
-	return nvkm_i2c_new_(&gm204_i2c, device, index, pi2c);
+	return nvkm_i2c_new_(&gm200_i2c, device, index, pi2c);
 }
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/i2c/pad.h b/drivers/gpu/drm/nouveau/nvkm/subdev/i2c/pad.h
index 9eeb992944c6..316c4536f29a 100644
--- a/drivers/gpu/drm/nouveau/nvkm/subdev/i2c/pad.h
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/i2c/pad.h
@@ -49,11 +49,11 @@ int nv4e_i2c_pad_new(struct nvkm_i2c *, int, struct nvkm_i2c_pad **);
 int nv50_i2c_pad_new(struct nvkm_i2c *, int, struct nvkm_i2c_pad **);
 int g94_i2c_pad_x_new(struct nvkm_i2c *, int, struct nvkm_i2c_pad **);
 int gf119_i2c_pad_x_new(struct nvkm_i2c *, int, struct nvkm_i2c_pad **);
-int gm204_i2c_pad_x_new(struct nvkm_i2c *, int, struct nvkm_i2c_pad **);
+int gm200_i2c_pad_x_new(struct nvkm_i2c *, int, struct nvkm_i2c_pad **);
 
 int g94_i2c_pad_s_new(struct nvkm_i2c *, int, struct nvkm_i2c_pad **);
 int gf119_i2c_pad_s_new(struct nvkm_i2c *, int, struct nvkm_i2c_pad **);
-int gm204_i2c_pad_s_new(struct nvkm_i2c *, int, struct nvkm_i2c_pad **);
+int gm200_i2c_pad_s_new(struct nvkm_i2c *, int, struct nvkm_i2c_pad **);
 
 int anx9805_pad_new(struct nvkm_i2c_bus *, int, u8, struct nvkm_i2c_pad **);
 
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/i2c/padgm204.c b/drivers/gpu/drm/nouveau/nvkm/subdev/i2c/padgm200.c
index 24a4d760c67b..7d417f6a816e 100644
--- a/drivers/gpu/drm/nouveau/nvkm/subdev/i2c/padgm204.c
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/i2c/padgm200.c
@@ -26,7 +26,7 @@
 #include "bus.h"
 
 static void
-gm204_i2c_pad_mode(struct nvkm_i2c_pad *pad, enum nvkm_i2c_pad_mode mode)
+gm200_i2c_pad_mode(struct nvkm_i2c_pad *pad, enum nvkm_i2c_pad_mode mode)
 {
 	struct nvkm_subdev *subdev = &pad->i2c->subdev;
 	struct nvkm_device *device = subdev->device;
@@ -51,26 +51,26 @@ gm204_i2c_pad_mode(struct nvkm_i2c_pad *pad, enum nvkm_i2c_pad_mode mode)
 }
 
 static const struct nvkm_i2c_pad_func
-gm204_i2c_pad_s_func = {
+gm200_i2c_pad_s_func = {
 	.bus_new_4 = gf119_i2c_bus_new,
-	.aux_new_6 = gm204_i2c_aux_new,
-	.mode = gm204_i2c_pad_mode,
+	.aux_new_6 = gm200_i2c_aux_new,
+	.mode = gm200_i2c_pad_mode,
 };
 
 int
-gm204_i2c_pad_s_new(struct nvkm_i2c *i2c, int id, struct nvkm_i2c_pad **ppad)
+gm200_i2c_pad_s_new(struct nvkm_i2c *i2c, int id, struct nvkm_i2c_pad **ppad)
 {
-	return nvkm_i2c_pad_new_(&gm204_i2c_pad_s_func, i2c, id, ppad);
+	return nvkm_i2c_pad_new_(&gm200_i2c_pad_s_func, i2c, id, ppad);
 }
 
 static const struct nvkm_i2c_pad_func
-gm204_i2c_pad_x_func = {
+gm200_i2c_pad_x_func = {
 	.bus_new_4 = gf119_i2c_bus_new,
-	.aux_new_6 = gm204_i2c_aux_new,
+	.aux_new_6 = gm200_i2c_aux_new,
 };
 
 int
-gm204_i2c_pad_x_new(struct nvkm_i2c *i2c, int id, struct nvkm_i2c_pad **ppad)
+gm200_i2c_pad_x_new(struct nvkm_i2c *i2c, int id, struct nvkm_i2c_pad **ppad)
 {
-	return nvkm_i2c_pad_new_(&gm204_i2c_pad_x_func, i2c, id, ppad);
+	return nvkm_i2c_pad_new_(&gm200_i2c_pad_x_func, i2c, id, ppad);
 }
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/ibus/Kbuild b/drivers/gpu/drm/nouveau/nvkm/subdev/ibus/Kbuild
index 7e77a7466992..ad572d3b5466 100644
--- a/drivers/gpu/drm/nouveau/nvkm/subdev/ibus/Kbuild
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/ibus/Kbuild
@@ -2,4 +2,4 @@ nvkm-y += nvkm/subdev/ibus/gf100.o
 nvkm-y += nvkm/subdev/ibus/gf117.o
 nvkm-y += nvkm/subdev/ibus/gk104.o
 nvkm-y += nvkm/subdev/ibus/gk20a.o
-nvkm-y += nvkm/subdev/ibus/gm204.o
+nvkm-y += nvkm/subdev/ibus/gm200.o
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/ibus/gm204.c b/drivers/gpu/drm/nouveau/nvkm/subdev/ibus/gm200.c
index b3839dc254ee..ef0b7f3b1128 100644
--- a/drivers/gpu/drm/nouveau/nvkm/subdev/ibus/gm204.c
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/ibus/gm200.c
@@ -24,17 +24,17 @@
 #include "priv.h"
 
 static const struct nvkm_subdev_func
-gm204_ibus = {
+gm200_ibus = {
 	.intr = gk104_ibus_intr,
 };
 
 int
-gm204_ibus_new(struct nvkm_device *device, int index,
+gm200_ibus_new(struct nvkm_device *device, int index,
 	       struct nvkm_subdev **pibus)
 {
 	struct nvkm_subdev *ibus;
 	if (!(ibus = *pibus = kzalloc(sizeof(*ibus), GFP_KERNEL)))
 		return -ENOMEM;
-	nvkm_subdev_ctor(&gm204_ibus, device, index, 0, ibus);
+	nvkm_subdev_ctor(&gm200_ibus, device, index, 0, ibus);
 	return 0;
 }
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/iccsense/Kbuild b/drivers/gpu/drm/nouveau/nvkm/subdev/iccsense/Kbuild
new file mode 100644
index 000000000000..98a4bd3e98ed
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/iccsense/Kbuild
@@ -0,0 +1,2 @@
+nvkm-y += nvkm/subdev/iccsense/base.o
+nvkm-y += nvkm/subdev/iccsense/gf100.o
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/iccsense/base.c b/drivers/gpu/drm/nouveau/nvkm/subdev/iccsense/base.c
new file mode 100644
index 000000000000..c44a85228074
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/iccsense/base.c
@@ -0,0 +1,232 @@
+/*
+ * Copyright 2015 Martin Peres
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
+ * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR
+ * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
+ * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
+ * OTHER DEALINGS IN THE SOFTWARE.
+ *
+ * Authors: Martin Peres
+ */
+#include "priv.h"
+
+#include <subdev/bios.h>
+#include <subdev/bios/extdev.h>
+#include <subdev/bios/iccsense.h>
+#include <subdev/i2c.h>
+
+static bool
+nvkm_iccsense_validate_device(struct i2c_adapter *i2c, u8 addr,
+			      enum nvbios_extdev_type type, u8 rail)
+{
+	switch (type) {
+	case NVBIOS_EXTDEV_INA209:
+	case NVBIOS_EXTDEV_INA219:
+		return rail == 0 && nv_rd16i2cr(i2c, addr, 0x0) >= 0;
+	case NVBIOS_EXTDEV_INA3221:
+		return rail <= 3 &&
+		       nv_rd16i2cr(i2c, addr, 0xff) == 0x3220 &&
+		       nv_rd16i2cr(i2c, addr, 0xfe) == 0x5449;
+	default:
+		return false;
+	}
+}
+
+static int
+nvkm_iccsense_poll_lane(struct i2c_adapter *i2c, u8 addr, u8 shunt_reg,
+			u8 shunt_shift, u8 bus_reg, u8 bus_shift, u8 shunt,
+			u16 lsb)
+{
+	int vshunt = nv_rd16i2cr(i2c, addr, shunt_reg);
+	int vbus = nv_rd16i2cr(i2c, addr, bus_reg);
+
+	if (vshunt < 0 || vbus < 0)
+		return -EINVAL;
+
+	vshunt >>= shunt_shift;
+	vbus >>= bus_shift;
+
+	return vbus * vshunt * lsb / shunt;
+}
+
+static int
+nvkm_iccsense_ina2x9_read(struct nvkm_iccsense *iccsense,
+                          struct nvkm_iccsense_rail *rail,
+			  u8 shunt_reg, u8 bus_reg)
+{
+	return nvkm_iccsense_poll_lane(rail->i2c, rail->addr, shunt_reg, 0,
+				       bus_reg, 3, rail->mohm, 10 * 4);
+}
+
+static int
+nvkm_iccsense_ina209_read(struct nvkm_iccsense *iccsense,
+			  struct nvkm_iccsense_rail *rail)
+{
+	return nvkm_iccsense_ina2x9_read(iccsense, rail, 3, 4);
+}
+
+static int
+nvkm_iccsense_ina219_read(struct nvkm_iccsense *iccsense,
+			  struct nvkm_iccsense_rail *rail)
+{
+	return nvkm_iccsense_ina2x9_read(iccsense, rail, 1, 2);
+}
+
+static int
+nvkm_iccsense_ina3221_read(struct nvkm_iccsense *iccsense,
+			   struct nvkm_iccsense_rail *rail)
+{
+	return nvkm_iccsense_poll_lane(rail->i2c, rail->addr,
+				       1 + (rail->rail * 2), 3,
+				       2 + (rail->rail * 2), 3, rail->mohm,
+				       40 * 8);
+}
+
+int
+nvkm_iccsense_read(struct nvkm_iccsense *iccsense, u8 idx)
+{
+	struct nvkm_iccsense_rail *rail;
+
+	if (!iccsense || idx >= iccsense->rail_count)
+		return -EINVAL;
+
+	rail = &iccsense->rails[idx];
+	if (!rail->read)
+		return -ENODEV;
+
+	return rail->read(iccsense, rail);
+}
+
+int
+nvkm_iccsense_read_all(struct nvkm_iccsense *iccsense)
+{
+	int result = 0, i;
+	for (i = 0; i < iccsense->rail_count; ++i) {
+		int res = nvkm_iccsense_read(iccsense, i);
+		if (res >= 0)
+			result += res;
+		else
+			return res;
+	}
+	return result;
+}
+
+static void *
+nvkm_iccsense_dtor(struct nvkm_subdev *subdev)
+{
+	struct nvkm_iccsense *iccsense = nvkm_iccsense(subdev);
+
+	if (iccsense->rails)
+		kfree(iccsense->rails);
+
+	return iccsense;
+}
+
+static int
+nvkm_iccsense_oneinit(struct nvkm_subdev *subdev)
+{
+	struct nvkm_iccsense *iccsense = nvkm_iccsense(subdev);
+	struct nvkm_bios *bios = subdev->device->bios;
+	struct nvkm_i2c *i2c = subdev->device->i2c;
+	struct nvbios_iccsense stbl;
+	int i;
+
+	if (!i2c || !bios || nvbios_iccsense_parse(bios, &stbl)
+	    || !stbl.nr_entry)
+		return 0;
+
+	iccsense->rails = kmalloc(sizeof(*iccsense->rails) * stbl.nr_entry,
+	                          GFP_KERNEL);
+	if (!iccsense->rails)
+		return -ENOMEM;
+
+	iccsense->data_valid = true;
+	for (i = 0; i < stbl.nr_entry; ++i) {
+		struct pwr_rail_t *r = &stbl.rail[i];
+		struct nvbios_extdev_func extdev;
+		struct nvkm_iccsense_rail *rail;
+		struct nvkm_i2c_bus *i2c_bus;
+		u8 addr;
+
+		if (!r->mode || r->resistor_mohm == 0)
+			continue;
+
+		if (nvbios_extdev_parse(bios, r->extdev_id, &extdev))
+			continue;
+
+		if (extdev.type == 0xff)
+			continue;
+
+		if (extdev.bus)
+			i2c_bus = nvkm_i2c_bus_find(i2c, NVKM_I2C_BUS_SEC);
+		else
+			i2c_bus = nvkm_i2c_bus_find(i2c, NVKM_I2C_BUS_PRI);
+		if (!i2c_bus)
+			continue;
+
+		addr = extdev.addr >> 1;
+		if (!nvkm_iccsense_validate_device(&i2c_bus->i2c, addr,
+						   extdev.type, r->rail)) {
+			iccsense->data_valid = false;
+			nvkm_warn(subdev, "found unknown or invalid rail entry"
+				  " type 0x%x rail %i, power reading might be"
+				  " invalid\n", extdev.type, r->rail);
+			continue;
+		}
+
+		rail = &iccsense->rails[iccsense->rail_count];
+		switch (extdev.type) {
+		case NVBIOS_EXTDEV_INA209:
+			rail->read = nvkm_iccsense_ina209_read;
+			break;
+		case NVBIOS_EXTDEV_INA219:
+			rail->read = nvkm_iccsense_ina219_read;
+			break;
+		case NVBIOS_EXTDEV_INA3221:
+			rail->read = nvkm_iccsense_ina3221_read;
+			break;
+		}
+
+		rail->addr = addr;
+		rail->rail = r->rail;
+		rail->mohm = r->resistor_mohm;
+		rail->i2c = &i2c_bus->i2c;
+		++iccsense->rail_count;
+	}
+	return 0;
+}
+
+struct nvkm_subdev_func iccsense_func = {
+	.oneinit = nvkm_iccsense_oneinit,
+	.dtor = nvkm_iccsense_dtor,
+};
+
+void
+nvkm_iccsense_ctor(struct nvkm_device *device, int index,
+		   struct nvkm_iccsense *iccsense)
+{
+	nvkm_subdev_ctor(&iccsense_func, device, index, 0, &iccsense->subdev);
+}
+
+int
+nvkm_iccsense_new_(struct nvkm_device *device, int index,
+		   struct nvkm_iccsense **iccsense)
+{
+	if (!(*iccsense = kzalloc(sizeof(**iccsense), GFP_KERNEL)))
+		return -ENOMEM;
+	nvkm_iccsense_ctor(device, index, *iccsense);
+	return 0;
+}
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/iccsense/gf100.c b/drivers/gpu/drm/nouveau/nvkm/subdev/iccsense/gf100.c
new file mode 100644
index 000000000000..cccff1c8a409
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/iccsense/gf100.c
@@ -0,0 +1,31 @@
+/*
+ * Copyright 2015 Karol Herbst
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
+ * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR
+ * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
+ * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
+ * OTHER DEALINGS IN THE SOFTWARE.
+ *
+ * Authors: Karol Herbst
+ */
+#include "priv.h"
+
+int
+gf100_iccsense_new(struct nvkm_device *device, int index,
+		   struct nvkm_iccsense **piccsense)
+{
+	return nvkm_iccsense_new_(device, index, piccsense);
+}
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/iccsense/priv.h b/drivers/gpu/drm/nouveau/nvkm/subdev/iccsense/priv.h
new file mode 100644
index 000000000000..ed398b81e86e
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/iccsense/priv.h
@@ -0,0 +1,16 @@
+#ifndef __NVKM_ICCSENSE_PRIV_H__
+#define __NVKM_ICCSENSE_PRIV_H__
+#define nvkm_iccsense(p) container_of((p), struct nvkm_iccsense, subdev)
+#include <subdev/iccsense.h>
+
+struct nvkm_iccsense_rail {
+	int (*read)(struct nvkm_iccsense *, struct nvkm_iccsense_rail *);
+	struct i2c_adapter *i2c;
+	u8 addr;
+	u8 rail;
+	u8 mohm;
+};
+
+void nvkm_iccsense_ctor(struct nvkm_device *, int, struct nvkm_iccsense *);
+int nvkm_iccsense_new_(struct nvkm_device *, int, struct nvkm_iccsense **);
+#endif
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/instmem/gk20a.c b/drivers/gpu/drm/nouveau/nvkm/subdev/instmem/gk20a.c
index 4c20fec64d96..6b8f2a19b2d9 100644
--- a/drivers/gpu/drm/nouveau/nvkm/subdev/instmem/gk20a.c
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/instmem/gk20a.c
@@ -228,6 +228,8 @@ gk20a_instobj_release_dma(struct nvkm_memory *memory)
 	struct gk20a_instmem *imem = node->imem;
 	struct nvkm_ltc *ltc = imem->base.subdev.device->ltc;
 
+	/* in case we got a write-combined mapping */
+	wmb();
 	nvkm_ltc_invalidate(ltc);
 }
 
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/ltc/Kbuild b/drivers/gpu/drm/nouveau/nvkm/subdev/ltc/Kbuild
index f8108df3cb38..932b366598aa 100644
--- a/drivers/gpu/drm/nouveau/nvkm/subdev/ltc/Kbuild
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/ltc/Kbuild
@@ -2,4 +2,4 @@ nvkm-y += nvkm/subdev/ltc/base.o
 nvkm-y += nvkm/subdev/ltc/gf100.o
 nvkm-y += nvkm/subdev/ltc/gk104.o
 nvkm-y += nvkm/subdev/ltc/gm107.o
-nvkm-y += nvkm/subdev/ltc/gm204.o
+nvkm-y += nvkm/subdev/ltc/gm200.o
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/ltc/gf100.c b/drivers/gpu/drm/nouveau/nvkm/subdev/ltc/gf100.c
index fb0de83da13c..c9eb677967a8 100644
--- a/drivers/gpu/drm/nouveau/nvkm/subdev/ltc/gf100.c
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/ltc/gf100.c
@@ -129,9 +129,7 @@ gf100_ltc_invalidate(struct nvkm_ltc *ltc)
 	s64 taken;
 
 	nvkm_wr32(device, 0x70004, 0x00000001);
-	taken = nvkm_wait_msec(device, 2, 0x70004, 0x00000003, 0x00000000);
-	if (taken < 0)
-		nvkm_warn(&ltc->subdev, "LTC invalidate timeout\n");
+	taken = nvkm_wait_msec(device, 2000, 0x70004, 0x00000003, 0x00000000);
 
 	if (taken > 0)
 		nvkm_debug(&ltc->subdev, "LTC invalidate took %lld ns\n", taken);
@@ -144,9 +142,7 @@ gf100_ltc_flush(struct nvkm_ltc *ltc)
 	s64 taken;
 
 	nvkm_wr32(device, 0x70010, 0x00000001);
-	taken = nvkm_wait_msec(device, 2, 0x70010, 0x00000003, 0x00000000);
-	if (taken < 0)
-		nvkm_warn(&ltc->subdev, "LTC flush timeout\n");
+	taken = nvkm_wait_msec(device, 2000, 0x70010, 0x00000003, 0x00000000);
 
 	if (taken > 0)
 		nvkm_debug(&ltc->subdev, "LTC flush took %lld ns\n", taken);
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/ltc/gm107.c b/drivers/gpu/drm/nouveau/nvkm/subdev/ltc/gm107.c
index 2af1f9e100fc..e292f5679418 100644
--- a/drivers/gpu/drm/nouveau/nvkm/subdev/ltc/gm107.c
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/ltc/gm107.c
@@ -43,10 +43,8 @@ gm107_ltc_cbc_wait(struct nvkm_ltc *ltc)
 	for (c = 0; c < ltc->ltc_nr; c++) {
 		for (s = 0; s < ltc->lts_nr; s++) {
 			const u32 addr = 0x14046c + (c * 0x2000) + (s * 0x200);
-			nvkm_msec(device, 2000,
-				if (!nvkm_rd32(device, addr))
-					break;
-			);
+			nvkm_wait_msec(device, 2000, addr,
+				       0x00000004, 0x00000000);
 		}
 	}
 }
@@ -75,7 +73,7 @@ gm107_ltc_lts_isr(struct nvkm_ltc *ltc, int c, int s)
 {
 	struct nvkm_subdev *subdev = &ltc->subdev;
 	struct nvkm_device *device = subdev->device;
-	u32 base = 0x140000 + (c * 0x2000) + (s * 0x400);
+	u32 base = 0x140000 + (c * 0x2000) + (s * 0x200);
 	u32 stat = nvkm_rd32(device, base + 0x00c);
 
 	if (stat) {
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/ltc/gm204.c b/drivers/gpu/drm/nouveau/nvkm/subdev/ltc/gm200.c
index 5ad6fb9d022d..2a29bfd5125a 100644
--- a/drivers/gpu/drm/nouveau/nvkm/subdev/ltc/gm204.c
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/ltc/gm200.c
@@ -27,7 +27,7 @@
 #include <subdev/timer.h>
 
 static int
-gm204_ltc_oneinit(struct nvkm_ltc *ltc)
+gm200_ltc_oneinit(struct nvkm_ltc *ltc)
 {
 	struct nvkm_device *device = ltc->subdev.device;
 
@@ -37,15 +37,15 @@ gm204_ltc_oneinit(struct nvkm_ltc *ltc)
 	return gf100_ltc_oneinit_tag_ram(ltc);
 }
 static void
-gm204_ltc_init(struct nvkm_ltc *ltc)
+gm200_ltc_init(struct nvkm_ltc *ltc)
 {
 	nvkm_wr32(ltc->subdev.device, 0x17e278, ltc->tag_base);
 }
 
 static const struct nvkm_ltc_func
-gm204_ltc = {
-	.oneinit = gm204_ltc_oneinit,
-	.init = gm204_ltc_init,
+gm200_ltc = {
+	.oneinit = gm200_ltc_oneinit,
+	.init = gm200_ltc_init,
 	.intr = gm107_ltc_intr, /*XXX: not validated */
 	.cbc_clear = gm107_ltc_cbc_clear,
 	.cbc_wait = gm107_ltc_cbc_wait,
@@ -57,7 +57,7 @@ gm204_ltc = {
 };
 
 int
-gm204_ltc_new(struct nvkm_device *device, int index, struct nvkm_ltc **pltc)
+gm200_ltc_new(struct nvkm_device *device, int index, struct nvkm_ltc **pltc)
 {
-	return nvkm_ltc_new_(&gm204_ltc, device, index, pltc);
+	return nvkm_ltc_new_(&gm200_ltc, device, index, pltc);
 }
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/pmu/fuc/gf100.fuc3.h b/drivers/gpu/drm/nouveau/nvkm/subdev/pmu/fuc/gf100.fuc3.h
index 770294457274..e2faccffee6f 100644
--- a/drivers/gpu/drm/nouveau/nvkm/subdev/pmu/fuc/gf100.fuc3.h
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/pmu/fuc/gf100.fuc3.h
@@ -24,8 +24,8 @@ uint32_t gf100_pmu_data[] = {
 	0x00000000,
 /* 0x0058: proc_list_head */
 	0x54534f48,
-	0x00000507,
-	0x000004a4,
+	0x0000050a,
+	0x000004a7,
 	0x00000000,
 	0x00000000,
 	0x00000000,
@@ -46,8 +46,8 @@ uint32_t gf100_pmu_data[] = {
 	0x00000000,
 	0x00000000,
 	0x584d454d,
-	0x00000753,
-	0x00000745,
+	0x00000756,
+	0x00000748,
 	0x00000000,
 	0x00000000,
 	0x00000000,
@@ -68,8 +68,8 @@ uint32_t gf100_pmu_data[] = {
 	0x00000000,
 	0x00000000,
 	0x46524550,
-	0x00000757,
-	0x00000755,
+	0x0000075a,
+	0x00000758,
 	0x00000000,
 	0x00000000,
 	0x00000000,
@@ -90,8 +90,8 @@ uint32_t gf100_pmu_data[] = {
 	0x00000000,
 	0x00000000,
 	0x5f433249,
-	0x00000b87,
-	0x00000a2a,
+	0x00000b8a,
+	0x00000a2d,
 	0x00000000,
 	0x00000000,
 	0x00000000,
@@ -112,8 +112,8 @@ uint32_t gf100_pmu_data[] = {
 	0x00000000,
 	0x00000000,
 	0x54534554,
-	0x00000bb0,
-	0x00000b89,
+	0x00000bb3,
+	0x00000b8c,
 	0x00000000,
 	0x00000000,
 	0x00000000,
@@ -134,8 +134,8 @@ uint32_t gf100_pmu_data[] = {
 	0x00000000,
 	0x00000000,
 	0x454c4449,
-	0x00000bbc,
-	0x00000bba,
+	0x00000bbf,
+	0x00000bbd,
 	0x00000000,
 	0x00000000,
 	0x00000000,
@@ -229,26 +229,26 @@ uint32_t gf100_pmu_data[] = {
 /* 0x0370: memx_func_head */
 	0x00000001,
 	0x00000000,
-	0x00000546,
+	0x00000549,
 /* 0x037c: memx_func_next */
 	0x00000002,
 	0x00000000,
-	0x000005d0,
+	0x000005d3,
 	0x00000003,
 	0x00000002,
-	0x0000069a,
+	0x0000069d,
 	0x00040004,
 	0x00000000,
-	0x000006b6,
+	0x000006b9,
 	0x00010005,
 	0x00000000,
-	0x000006d3,
+	0x000006d6,
 	0x00010006,
 	0x00000000,
-	0x00000658,
+	0x0000065b,
 	0x00000007,
 	0x00000000,
-	0x000006de,
+	0x000006e1,
 /* 0x03c4: memx_func_tail */
 /* 0x03c4: memx_ts_start */
 	0x00000000,
@@ -917,887 +917,887 @@ uint32_t gf100_pmu_data[] = {
 };
 
 uint32_t gf100_pmu_code[] = {
-	0x03930ef5,
+	0x03920ef5,
 /* 0x0004: rd32 */
 	0x07a007f1,
 	0xd00604b6,
 	0x04bd000e,
-	0xf001d7f0,
-	0x07f101d3,
-	0x04b607ac,
-	0x000dd006,
-/* 0x0022: rd32_wait */
-	0xd7f104bd,
-	0xd4b607ac,
-	0x00ddcf06,
-	0x7000d4f1,
-	0xf1f21bf4,
-	0xb607a4d7,
-	0xddcf06d4,
-/* 0x003f: wr32 */
-	0xf100f800,
-	0xb607a007,
-	0x0ed00604,
-	0xf104bd00,
-	0xb607a407,
+	0x0001d7f1,
+	0xf101d3f0,
+	0xb607ac07,
 	0x0dd00604,
-	0xf004bd00,
-	0xd5f002d7,
-	0x01d3f0f0,
-	0x07ac07f1,
+/* 0x0023: rd32_wait */
+	0xf104bd00,
+	0xb607acd7,
+	0xddcf06d4,
+	0x00d4f100,
+	0xf21bf470,
+	0x07a4d7f1,
+	0xcf06d4b6,
+	0x00f800dd,
+/* 0x0040: wr32 */
+	0x07a007f1,
+	0xd00604b6,
+	0x04bd000e,
+	0x07a407f1,
 	0xd00604b6,
 	0x04bd000d,
-/* 0x006c: wr32_wait */
-	0x07acd7f1,
-	0xcf06d4b6,
-	0xd4f100dd,
-	0x1bf47000,
-/* 0x007f: nsec */
-	0xf900f8f2,
-	0xf080f990,
-	0x84b62c87,
-	0x0088cf06,
-/* 0x008c: nsec_loop */
-	0xb62c97f0,
-	0x99cf0694,
-	0x0298bb00,
-	0xf4069eb8,
-	0x80fcf11e,
-	0x00f890fc,
-/* 0x00a4: wait */
-	0x80f990f9,
-	0xb62c87f0,
-	0x88cf0684,
-/* 0x00b1: wait_loop */
-	0x02eeb900,
-	0xb90421f4,
-	0xadfd02da,
-	0x06acb804,
-	0xf0150bf4,
+	0x00f2d7f1,
+	0xf101d3f0,
+	0xb607ac07,
+	0x0dd00604,
+/* 0x006b: wr32_wait */
+	0xf104bd00,
+	0xb607acd7,
+	0xddcf06d4,
+	0x00d4f100,
+	0xf21bf470,
+/* 0x007e: nsec */
+	0x90f900f8,
+	0x87f080f9,
+	0x0684b62c,
+/* 0x008b: nsec_loop */
+	0xf00088cf,
 	0x94b62c97,
 	0x0099cf06,
 	0xb80298bb,
-	0x1ef4069b,
-/* 0x00d5: wait_done */
-	0xfc80fcdf,
-/* 0x00db: intr_watchdog */
-	0x9800f890,
-	0x96b003e9,
-	0x2a0bf400,
-	0xbb9a0a98,
-	0x1cf4029a,
-	0x01d7f00f,
-	0x02d221f5,
-	0x0ef494bd,
-/* 0x00f9: intr_watchdog_next_time */
-	0x9b0a9815,
-	0xf400a6b0,
-	0x9ab8090b,
-	0x061cf406,
-/* 0x0108: intr_watchdog_next_time_set */
-/* 0x010b: intr_watchdog_next_proc */
-	0x809b0980,
-	0xe0b603e9,
-	0x68e6b158,
-	0xc61bf402,
-/* 0x011a: intr */
-	0x00f900f8,
-	0x80f904bd,
-	0xa0f990f9,
-	0xc0f9b0f9,
-	0xe0f9d0f9,
-	0xf7f0f0f9,
-	0x0188fe00,
-	0x87f180f9,
-	0x84b605d0,
+	0x1ef4069e,
+	0xfc80fcf1,
+/* 0x00a3: wait */
+	0xf900f890,
+	0xf080f990,
+	0x84b62c87,
 	0x0088cf06,
-	0xf10180b6,
-	0xb605d007,
+/* 0x00b0: wait_loop */
+	0xf402eeb9,
+	0xdab90421,
+	0x04adfd02,
+	0xf406acb8,
+	0x97f0150b,
+	0x0694b62c,
+	0xbb0099cf,
+	0x9bb80298,
+	0xdf1ef406,
+/* 0x00d4: wait_done */
+	0x90fc80fc,
+/* 0x00da: intr_watchdog */
+	0xe99800f8,
+	0x0096b003,
+	0x982a0bf4,
+	0x9abb9a0a,
+	0x0f1cf402,
+	0xf501d7f0,
+	0xbd02d121,
+	0x150ef494,
+/* 0x00f8: intr_watchdog_next_time */
+	0xb09b0a98,
+	0x0bf400a6,
+	0x069ab809,
+/* 0x0107: intr_watchdog_next_time_set */
+	0x80061cf4,
+/* 0x010a: intr_watchdog_next_proc */
+	0xe9809b09,
+	0x58e0b603,
+	0x0268e6b1,
+	0xf8c61bf4,
+/* 0x0119: intr */
+	0xbd00f900,
+	0xf980f904,
+	0xf9a0f990,
+	0xf9c0f9b0,
+	0xf9e0f9d0,
+	0x00f7f0f0,
+	0xf90188fe,
+	0xd087f180,
+	0x0684b605,
+	0xb60088cf,
+	0x07f10180,
+	0x04b605d0,
+	0x0008d006,
+	0x87f004bd,
+	0x0684b608,
+	0xc40088cf,
+	0x0bf40289,
+	0x9b008023,
+	0xf458e7f0,
+	0x0998da21,
+	0x0096b09b,
+	0xf0110bf4,
+	0x04b63407,
+	0x0009d006,
+	0x098004bd,
+/* 0x017d: intr_skip_watchdog */
+	0x0089e49a,
+	0x480bf408,
+	0x068897f1,
+	0xcf0694b6,
+	0x9ac40099,
+	0x2c0bf402,
+	0x04c0c7f1,
+	0xcf06c4b6,
+	0xc0f900cc,
+	0x4f48e7f1,
+	0x5453e3f1,
+	0xf500d7f0,
+	0xfc033621,
+	0xc007f1c0,
+	0x0604b604,
+	0xbd000cd0,
+/* 0x01bd: intr_subintr_skip_fifo */
+	0x8807f104,
+	0x0604b606,
+	0xbd0009d0,
+/* 0x01c9: intr_skip_subintr */
+	0xe097f104,
+	0xfd90bd00,
+	0x07f00489,
+	0x0604b604,
+	0xbd0008d0,
+	0xfe80fc04,
+	0xf0fc0088,
+	0xd0fce0fc,
+	0xb0fcc0fc,
+	0x90fca0fc,
+	0x00fc80fc,
+	0xf80032f4,
+/* 0x01f9: ticks_from_ns */
+	0xf9c0f901,
+	0xcbd7f1b0,
+	0x00d3f000,
+	0x040b21f5,
+	0x03e8ccec,
+	0xf400b4b0,
+	0xeeec120b,
+	0xd7f103e8,
+	0xd3f000cb,
+	0x0b21f500,
+/* 0x0221: ticks_from_ns_quit */
+	0x02ceb904,
+	0xc0fcb0fc,
+/* 0x022a: ticks_from_us */
+	0xc0f900f8,
+	0xd7f1b0f9,
+	0xd3f000cb,
+	0x0b21f500,
+	0x02ceb904,
+	0xf400b4b0,
+	0xe4bd050b,
+/* 0x0244: ticks_from_us_quit */
+	0xc0fcb0fc,
+/* 0x024a: ticks_to_us */
+	0xd7f100f8,
+	0xd3f000cb,
+	0xecedff00,
+/* 0x0256: timer */
+	0x90f900f8,
+	0x32f480f9,
+	0x03f89810,
+	0xf40086b0,
+	0x84bd651c,
+	0xb63807f0,
 	0x08d00604,
 	0xf004bd00,
-	0x84b60887,
+	0x84b63487,
 	0x0088cf06,
-	0xf40289c4,
-	0x0080230b,
-	0x58e7f09b,
-	0x98db21f4,
-	0x96b09b09,
-	0x110bf400,
+	0xbb9a0998,
+	0xe9bb0298,
+	0x03fe8000,
+	0xb60887f0,
+	0x88cf0684,
+	0x0284f000,
+	0xf0261bf4,
+	0x84b63487,
+	0x0088cf06,
+	0xf406e0b8,
+	0xe8b8090b,
+	0x111cf406,
+/* 0x02ac: timer_reset */
 	0xb63407f0,
-	0x09d00604,
+	0x0ed00604,
 	0x8004bd00,
-/* 0x017e: intr_skip_watchdog */
-	0x89e49a09,
-	0x0bf40800,
-	0x8897f148,
-	0x0694b606,
-	0xc40099cf,
-	0x0bf4029a,
-	0xc0c7f12c,
-	0x06c4b604,
-	0xf900cccf,
-	0x48e7f1c0,
-	0x53e3f14f,
-	0x00d7f054,
-	0x033721f5,
-	0x07f1c0fc,
-	0x04b604c0,
-	0x000cd006,
-/* 0x01be: intr_subintr_skip_fifo */
-	0x07f104bd,
-	0x04b60688,
-	0x0009d006,
-/* 0x01ca: intr_skip_subintr */
-	0x97f104bd,
-	0x90bd00e0,
-	0xf00489fd,
-	0x04b60407,
-	0x0008d006,
-	0x80fc04bd,
-	0xfc0088fe,
-	0xfce0fcf0,
-	0xfcc0fcd0,
-	0xfca0fcb0,
-	0xfc80fc90,
-	0x0032f400,
-/* 0x01fa: ticks_from_ns */
-	0xc0f901f8,
-	0xd7f1b0f9,
-	0xd3f000cb,
-	0x0821f500,
-	0xe8ccec04,
-	0x00b4b003,
-	0xec120bf4,
-	0xf103e8ee,
-	0xf000cbd7,
-	0x21f500d3,
-/* 0x0222: ticks_from_ns_quit */
-	0xceb90408,
-	0xfcb0fc02,
-/* 0x022b: ticks_from_us */
-	0xf900f8c0,
-	0xf1b0f9c0,
-	0xf000cbd7,
-	0x21f500d3,
-	0xceb90408,
-	0x00b4b002,
-	0xbd050bf4,
-/* 0x0245: ticks_from_us_quit */
-	0xfcb0fce4,
-/* 0x024b: ticks_to_us */
-	0xf100f8c0,
-	0xf000cbd7,
-	0xedff00d3,
-/* 0x0257: timer */
-	0xf900f8ec,
-	0xf480f990,
-	0xf8981032,
-	0x0086b003,
-	0xbd651cf4,
-	0x3807f084,
+/* 0x02ba: timer_enable */
+	0x87f09a0e,
+	0x3807f001,
 	0xd00604b6,
 	0x04bd0008,
-	0xb63487f0,
-	0x88cf0684,
-	0x9a099800,
-	0xbb0298bb,
-	0xfe8000e9,
-	0x0887f003,
-	0xcf0684b6,
-	0x84f00088,
-	0x261bf402,
-	0xb63487f0,
-	0x88cf0684,
-	0x06e0b800,
-	0xb8090bf4,
-	0x1cf406e8,
-/* 0x02ad: timer_reset */
-	0x3407f011,
-	0xd00604b6,
-	0x04bd000e,
-/* 0x02bb: timer_enable */
-	0xf09a0e80,
-	0x07f00187,
-	0x0604b638,
-	0xbd0008d0,
-/* 0x02c9: timer_done */
-	0x1031f404,
+/* 0x02c8: timer_done */
+	0xfc1031f4,
+	0xf890fc80,
+/* 0x02d1: send_proc */
+	0xf980f900,
+	0x05e89890,
+	0xf004e998,
+	0x89b80486,
+	0x2a0bf406,
+	0x940398c4,
+	0x80b60488,
+	0x008ebb18,
+	0x8000fa98,
+	0x8d80008a,
+	0x028c8001,
+	0xb6038b80,
+	0x94f00190,
+	0x04e98007,
+/* 0x030b: send_done */
+	0xfc0231f4,
+	0xf880fc90,
+/* 0x0311: find */
+	0xf080f900,
+	0x31f45887,
+/* 0x0319: find_loop */
+	0x008a9801,
+	0xf406aeb8,
+	0x80b6100b,
+	0x6886b158,
+	0xf01bf402,
+/* 0x032f: find_done */
+	0xb90132f4,
+	0x80fc028e,
+/* 0x0336: send */
+	0x21f500f8,
+	0x01f40311,
+/* 0x033f: recv */
+	0xf900f897,
+	0x9880f990,
+	0xe99805e8,
+	0x0132f404,
+	0xf40689b8,
+	0x89c43d0b,
+	0x0180b603,
+	0x800784f0,
+	0xea9805e8,
+	0xfef0f902,
+	0xf0f9018f,
+	0x9402efb9,
+	0xe9bb0499,
+	0x18e0b600,
+	0x9803eb98,
+	0xed9802ec,
+	0x00ee9801,
+	0xf0fca5f9,
+	0xf400f8fe,
+	0xf0fc0131,
+/* 0x038c: recv_done */
 	0x90fc80fc,
-/* 0x02d2: send_proc */
-	0x80f900f8,
-	0xe89890f9,
-	0x04e99805,
-	0xb80486f0,
-	0x0bf40689,
-	0x0398c42a,
-	0xb6048894,
-	0x8ebb1880,
-	0x00fa9800,
-	0x80008a80,
-	0x8c80018d,
-	0x038b8002,
-	0xf00190b6,
-	0xe9800794,
-	0x0231f404,
-/* 0x030c: send_done */
-	0x80fc90fc,
-/* 0x0312: find */
-	0x80f900f8,
-	0xf45887f0,
-/* 0x031a: find_loop */
-	0x8a980131,
-	0x06aeb800,
-	0xb6100bf4,
-	0x86b15880,
-	0x1bf40268,
-	0x0132f4f0,
-/* 0x0330: find_done */
-	0xfc028eb9,
-/* 0x0337: send */
-	0xf500f880,
-	0xf4031221,
-	0x00f89701,
-/* 0x0340: recv */
-	0x80f990f9,
-	0x9805e898,
-	0x32f404e9,
-	0x0689b801,
-	0xc43d0bf4,
-	0x80b60389,
-	0x0784f001,
-	0x9805e880,
-	0xf0f902ea,
-	0xf9018ffe,
-	0x02efb9f0,
-	0xbb049994,
-	0xe0b600e9,
-	0x03eb9818,
-	0x9802ec98,
-	0xee9801ed,
-	0xfca5f900,
-	0x00f8fef0,
-	0xfc0131f4,
-/* 0x038d: recv_done */
-	0xfc80fcf0,
-/* 0x0393: init */
-	0xf100f890,
-	0xb6010817,
-	0x11cf0614,
-	0x0911e700,
-	0x0814b601,
-	0xf10014fe,
-	0xf000e017,
-	0x07f00013,
-	0x0604b61c,
-	0xbd0001d0,
-	0xff17f004,
-	0xb61407f0,
-	0x01d00604,
-	0xf004bd00,
-	0x15f10217,
-	0x07f00800,
-	0x0604b610,
-	0xbd0001d0,
-	0x1a17f104,
-	0x0013f001,
-	0xf40010fe,
-	0x17f01031,
-	0x3807f001,
+/* 0x0392: init */
+	0x17f100f8,
+	0x14b60108,
+	0x0011cf06,
+	0x010911e7,
+	0xfe0814b6,
+	0x17f10014,
+	0x13f000e0,
+	0x1c07f000,
+	0xd00604b6,
+	0x04bd0001,
+	0xf0ff17f0,
+	0x04b61407,
+	0x0001d006,
+	0x17f004bd,
+	0x0015f102,
+	0x1007f008,
 	0xd00604b6,
 	0x04bd0001,
-/* 0x03f7: init_proc */
-	0x9858f7f0,
-	0x16b001f1,
-	0xfa0bf400,
-	0xf0b615f9,
-	0xf20ef458,
-/* 0x0408: mulu32_32_64 */
-	0x20f910f9,
-	0x40f930f9,
-	0x9510e195,
-	0xc4bd10d2,
-	0xedffb4bd,
-	0x301dffc0,
-	0xf10234b9,
-	0xb6ffff34,
-	0x45b61034,
-	0x00c3bb10,
-	0xff01b4bb,
-	0x34b930e2,
-	0xff34f102,
-	0x1034b6ff,
-	0xbb1045b6,
-	0xb4bb00c3,
-	0x3012ff01,
-	0xfc00b3bb,
-	0xfc30fc40,
-	0xf810fc20,
-/* 0x0459: host_send */
-	0xb017f100,
+	0x011917f1,
+	0xf10013f0,
+	0xfeffff14,
+	0x31f40010,
+	0x0117f010,
+	0xb63807f0,
+	0x01d00604,
+	0xf004bd00,
+/* 0x03fa: init_proc */
+	0xf19858f7,
+	0x0016b001,
+	0xf9fa0bf4,
+	0x58f0b615,
+/* 0x040b: mulu32_32_64 */
+	0xf9f20ef4,
+	0xf920f910,
+	0x9540f930,
+	0xd29510e1,
+	0xbdc4bd10,
+	0xc0edffb4,
+	0xb9301dff,
+	0x34f10234,
+	0x34b6ffff,
+	0x1045b610,
+	0xbb00c3bb,
+	0xe2ff01b4,
+	0x0234b930,
+	0xffff34f1,
+	0xb61034b6,
+	0xc3bb1045,
+	0x01b4bb00,
+	0xbb3012ff,
+	0x40fc00b3,
+	0x20fc30fc,
+	0x00f810fc,
+/* 0x045c: host_send */
+	0x04b017f1,
+	0xcf0614b6,
+	0x27f10011,
+	0x24b604a0,
+	0x0022cf06,
+	0xf40612b8,
+	0x1ec4320b,
+	0x04ee9407,
+	0x0270e0b7,
+	0x9803eb98,
+	0xed9802ec,
+	0x00ee9801,
+	0x033621f5,
+	0xc40110b6,
+	0x07f10f1e,
+	0x04b604b0,
+	0x000ed006,
+	0x0ef404bd,
+/* 0x04a5: host_send_done */
+/* 0x04a7: host_recv */
+	0xf100f8ba,
+	0xf14e4917,
+	0xb8525413,
+	0x0bf406e1,
+/* 0x04b5: host_recv_wait */
+	0xcc17f1aa,
 	0x0614b604,
 	0xf10011cf,
-	0xb604a027,
+	0xb604c827,
 	0x22cf0624,
-	0x0612b800,
-	0xc4320bf4,
-	0xee94071e,
-	0x70e0b704,
-	0x03eb9802,
-	0x9802ec98,
-	0xee9801ed,
-	0x3721f500,
-	0x0110b603,
-	0xf10f1ec4,
-	0xb604b007,
-	0x0ed00604,
-	0xf404bd00,
-/* 0x04a2: host_send_done */
-	0x00f8ba0e,
-/* 0x04a4: host_recv */
-	0x4e4917f1,
-	0x525413f1,
-	0xf406e1b8,
-/* 0x04b2: host_recv_wait */
-	0x17f1aa0b,
-	0x14b604cc,
-	0x0011cf06,
-	0x04c827f1,
-	0xcf0624b6,
-	0x16f00022,
-	0x0612b808,
-	0xc4e60bf4,
-	0x34b60723,
-	0xf030b704,
-	0x033b8002,
-	0x80023c80,
-	0x3e80013d,
-	0x0120b600,
-	0xf10f24f0,
-	0xb604c807,
-	0x02d00604,
-	0xf004bd00,
-	0x07f04027,
-	0x0604b600,
-	0xbd0002d0,
-/* 0x0507: host_init */
-	0xf100f804,
-	0xb6008017,
-	0x15f11014,
-	0x07f10270,
-	0x04b604d0,
-	0x0001d006,
-	0x17f104bd,
+	0x0816f000,
+	0xf40612b8,
+	0x23c4e60b,
+	0x0434b607,
+	0x02f030b7,
+	0x80033b80,
+	0x3d80023c,
+	0x003e8001,
+	0xf00120b6,
+	0x07f10f24,
+	0x04b604c8,
+	0x0002d006,
+	0x27f004bd,
+	0x0007f040,
+	0xd00604b6,
+	0x04bd0002,
+/* 0x050a: host_init */
+	0x17f100f8,
 	0x14b60080,
-	0xf015f110,
-	0xdc07f102,
+	0x7015f110,
+	0xd007f102,
 	0x0604b604,
 	0xbd0001d0,
-	0x0117f004,
-	0x04c407f1,
+	0x8017f104,
+	0x1014b600,
+	0x02f015f1,
+	0x04dc07f1,
 	0xd00604b6,
 	0x04bd0001,
-/* 0x0546: memx_func_enter */
-	0x67f100f8,
-	0x77f11620,
-	0x73f1f55d,
-	0x6eb9ffff,
-	0x0421f402,
-	0xfd02d8b9,
-	0x60f90487,
-	0xd0fc80f9,
-	0x21f4e0fc,
-	0xfe77f13f,
-	0xff73f1ff,
+	0xf10117f0,
+	0xb604c407,
+	0x01d00604,
+	0xf804bd00,
+/* 0x0549: memx_func_enter */
+	0x2067f100,
+	0x5d77f116,
+	0xff73f1f5,
 	0x026eb9ff,
 	0xb90421f4,
 	0x87fd02d8,
 	0xf960f904,
 	0xfcd0fc80,
-	0x3f21f4e0,
-	0x26f067f1,
+	0x4021f4e0,
+	0xfffe77f1,
+	0xffff73f1,
 	0xf4026eb9,
 	0xd8b90421,
 	0x0487fd02,
 	0x80f960f9,
 	0xe0fcd0fc,
-	0xf03f21f4,
+	0xf14021f4,
+	0xb926f067,
+	0x21f4026e,
+	0x02d8b904,
+	0xf90487fd,
+	0xfc80f960,
+	0xf4e0fcd0,
+	0x67f04021,
+	0xe007f104,
+	0x0604b607,
+	0xbd0006d0,
+/* 0x05b5: memx_func_enter_wait */
+	0xc067f104,
+	0x0664b607,
+	0xf00066cf,
+	0x0bf40464,
+	0x2c67f0f3,
+	0xcf0664b6,
+	0x06800066,
+/* 0x05d3: memx_func_leave */
+	0xf000f8f1,
+	0x64b62c67,
+	0x0066cf06,
+	0xf0f20680,
 	0x07f10467,
-	0x04b607e0,
+	0x04b607e4,
 	0x0006d006,
-/* 0x05b2: memx_func_enter_wait */
+/* 0x05ee: memx_func_leave_wait */
 	0x67f104bd,
 	0x64b607c0,
 	0x0066cf06,
 	0xf40464f0,
-	0x67f0f30b,
-	0x0664b62c,
-	0x800066cf,
-	0x00f8f106,
-/* 0x05d0: memx_func_leave */
-	0xb62c67f0,
-	0x66cf0664,
-	0xf2068000,
-	0xf10467f0,
-	0xb607e407,
-	0x06d00604,
-/* 0x05eb: memx_func_leave_wait */
-	0xf104bd00,
-	0xb607c067,
-	0x66cf0664,
-	0x0464f000,
-	0xf1f31bf4,
-	0xf126f067,
-	0xf0000177,
+	0x67f1f31b,
+	0x77f126f0,
+	0x73f00001,
+	0x026eb900,
+	0xb90421f4,
+	0x87fd02d8,
+	0xf960f905,
+	0xfcd0fc80,
+	0x4021f4e0,
+	0x162067f1,
+	0xf4026eb9,
+	0xd8b90421,
+	0x0587fd02,
+	0x80f960f9,
+	0xe0fcd0fc,
+	0xf14021f4,
+	0xf00aa277,
 	0x6eb90073,
 	0x0421f402,
 	0xfd02d8b9,
 	0x60f90587,
 	0xd0fc80f9,
 	0x21f4e0fc,
-	0x2067f13f,
-	0x026eb916,
-	0xb90421f4,
-	0x87fd02d8,
-	0xf960f905,
-	0xfcd0fc80,
-	0x3f21f4e0,
-	0x0aa277f1,
-	0xb90073f0,
-	0x21f4026e,
-	0x02d8b904,
-	0xf90587fd,
-	0xfc80f960,
-	0xf4e0fcd0,
-	0x00f83f21,
-/* 0x0658: memx_func_wait_vblank */
-	0xb0001698,
-	0x0bf40066,
-	0x0166b013,
-	0xf4060bf4,
-/* 0x066a: memx_func_wait_vblank_head1 */
-	0x77f12e0e,
-	0x0ef40020,
-/* 0x0671: memx_func_wait_vblank_head0 */
-	0x0877f107,
-/* 0x0675: memx_func_wait_vblank_0 */
-	0xc467f100,
-	0x0664b607,
-	0xfd0066cf,
-	0x1bf40467,
-/* 0x0685: memx_func_wait_vblank_1 */
-	0xc467f1f3,
-	0x0664b607,
-	0xfd0066cf,
-	0x0bf40467,
-/* 0x0695: memx_func_wait_vblank_fini */
-	0x0410b6f3,
-/* 0x069a: memx_func_wr32 */
-	0x169800f8,
-	0x01159800,
-	0xf90810b6,
-	0xfc50f960,
-	0xf4e0fcd0,
-	0x42b63f21,
-	0xe91bf402,
-/* 0x06b6: memx_func_wait */
-	0x87f000f8,
-	0x0684b62c,
-	0x980088cf,
-	0x1d98001e,
-	0x021c9801,
-	0xb6031b98,
-	0x21f41010,
-/* 0x06d3: memx_func_delay */
-	0x9800f8a4,
-	0x10b6001e,
-	0x7f21f404,
-/* 0x06de: memx_func_train */
-	0x00f800f8,
-/* 0x06e0: memx_exec */
-	0xd0f9e0f9,
-	0xb902c1b9,
-/* 0x06ea: memx_exec_next */
-	0x139802b2,
+/* 0x065b: memx_func_wait_vblank */
+	0x9800f840,
+	0x66b00016,
+	0x130bf400,
+	0xf40166b0,
+	0x0ef4060b,
+/* 0x066d: memx_func_wait_vblank_head1 */
+	0x2077f12e,
+	0x070ef400,
+/* 0x0674: memx_func_wait_vblank_head0 */
+	0x000877f1,
+/* 0x0678: memx_func_wait_vblank_0 */
+	0x07c467f1,
+	0xcf0664b6,
+	0x67fd0066,
+	0xf31bf404,
+/* 0x0688: memx_func_wait_vblank_1 */
+	0x07c467f1,
+	0xcf0664b6,
+	0x67fd0066,
+	0xf30bf404,
+/* 0x0698: memx_func_wait_vblank_fini */
+	0xf80410b6,
+/* 0x069d: memx_func_wr32 */
+	0x00169800,
+	0xb6011598,
+	0x60f90810,
+	0xd0fc50f9,
+	0x21f4e0fc,
+	0x0242b640,
+	0xf8e91bf4,
+/* 0x06b9: memx_func_wait */
+	0x2c87f000,
+	0xcf0684b6,
+	0x1e980088,
+	0x011d9800,
+	0x98021c98,
+	0x10b6031b,
+	0xa321f410,
+/* 0x06d6: memx_func_delay */
+	0x1e9800f8,
 	0x0410b600,
-	0x01f034e7,
-	0x01e033e7,
-	0xf00132b6,
-	0x35980c30,
-	0xb855f9de,
-	0x1ef40612,
-	0xf10b98e4,
-	0xbbf20c98,
-	0xb7f102cb,
-	0xb4b607c4,
-	0x00bbcf06,
-	0xe0fcd0fc,
-	0x033721f5,
-/* 0x0726: memx_info */
-	0xc67000f8,
-	0x0e0bf401,
-/* 0x072c: memx_info_data */
-	0x03ccc7f1,
-	0x0800b7f1,
-/* 0x0737: memx_info_train */
-	0xf10b0ef4,
-	0xf10bccc7,
-/* 0x073f: memx_info_send */
-	0xf50100b7,
-	0xf8033721,
-/* 0x0745: memx_recv */
-	0x01d6b000,
-	0xb0980bf4,
-	0x0bf400d6,
-/* 0x0753: memx_init */
-	0xf800f8d8,
-/* 0x0755: perf_recv */
-/* 0x0757: perf_init */
-	0xf800f800,
-/* 0x0759: i2c_drive_scl */
-	0x0036b000,
-	0xf1110bf4,
-	0xb607e007,
-	0x01d00604,
-	0xf804bd00,
-/* 0x076d: i2c_drive_scl_lo */
-	0xe407f100,
-	0x0604b607,
-	0xbd0001d0,
-/* 0x077b: i2c_drive_sda */
-	0xb000f804,
-	0x0bf40036,
-	0xe007f111,
-	0x0604b607,
-	0xbd0002d0,
-/* 0x078f: i2c_drive_sda_lo */
-	0xf100f804,
-	0xb607e407,
-	0x02d00604,
-	0xf804bd00,
-/* 0x079d: i2c_sense_scl */
-	0x0132f400,
-	0x07c437f1,
-	0xcf0634b6,
-	0x31fd0033,
-	0x060bf404,
-/* 0x07b3: i2c_sense_scl_done */
-	0xf80131f4,
-/* 0x07b5: i2c_sense_sda */
-	0x0132f400,
-	0x07c437f1,
-	0xcf0634b6,
-	0x32fd0033,
-	0x060bf404,
-/* 0x07cb: i2c_sense_sda_done */
-	0xf80131f4,
-/* 0x07cd: i2c_raise_scl */
-	0xf140f900,
-	0xf0089847,
-	0x21f50137,
-/* 0x07da: i2c_raise_scl_wait */
-	0xe7f10759,
-	0x21f403e8,
-	0x9d21f57f,
-	0x0901f407,
-	0xf40142b6,
-/* 0x07ee: i2c_raise_scl_done */
-	0x40fcef1b,
-/* 0x07f2: i2c_start */
-	0x21f500f8,
-	0x11f4079d,
-	0xb521f50d,
-	0x0611f407,
-/* 0x0803: i2c_start_rep */
-	0xf0300ef4,
-	0x21f50037,
-	0x37f00759,
-	0x7b21f501,
-	0x0076bb07,
-	0xf90465b6,
-	0x04659450,
-	0xbd0256bb,
-	0x0475fd50,
-	0x21f550fc,
-	0x64b607cd,
-	0x1f11f404,
-/* 0x0830: i2c_start_send */
-	0xf50037f0,
-	0xf1077b21,
-	0xf41388e7,
-	0x37f07f21,
-	0x5921f500,
-	0x88e7f107,
-	0x7f21f413,
-/* 0x084c: i2c_start_out */
-/* 0x084e: i2c_stop */
-	0x37f000f8,
-	0x5921f500,
-	0x0037f007,
-	0x077b21f5,
-	0x03e8e7f1,
-	0xf07f21f4,
-	0x21f50137,
-	0xe7f10759,
-	0x21f41388,
-	0x0137f07f,
-	0x077b21f5,
-	0x1388e7f1,
-	0xf87f21f4,
-/* 0x0881: i2c_bitw */
-	0x7b21f500,
-	0xe8e7f107,
-	0x7f21f403,
-	0xb60076bb,
-	0x50f90465,
-	0xbb046594,
-	0x50bd0256,
-	0xfc0475fd,
-	0xcd21f550,
-	0x0464b607,
-	0xf11811f4,
-	0xf41388e7,
-	0x37f07f21,
-	0x5921f500,
-	0x88e7f107,
-	0x7f21f413,
-/* 0x08c0: i2c_bitw_out */
-/* 0x08c2: i2c_bitr */
-	0x37f000f8,
-	0x7b21f501,
+	0xf87e21f4,
+/* 0x06e1: memx_func_train */
+/* 0x06e3: memx_exec */
+	0xf900f800,
+	0xb9d0f9e0,
+	0xb2b902c1,
+/* 0x06ed: memx_exec_next */
+	0x00139802,
+	0xe70410b6,
+	0xe701f034,
+	0xb601e033,
+	0x30f00132,
+	0xde35980c,
+	0x12b855f9,
+	0xe41ef406,
+	0x98f10b98,
+	0xcbbbf20c,
+	0xc4b7f102,
+	0x06b4b607,
+	0xfc00bbcf,
+	0xf5e0fcd0,
+	0xf8033621,
+/* 0x0729: memx_info */
+	0x01c67000,
+/* 0x072f: memx_info_data */
+	0xf10e0bf4,
+	0xf103ccc7,
+	0xf40800b7,
+/* 0x073a: memx_info_train */
+	0xc7f10b0e,
+	0xb7f10bcc,
+/* 0x0742: memx_info_send */
+	0x21f50100,
+	0x00f80336,
+/* 0x0748: memx_recv */
+	0xf401d6b0,
+	0xd6b0980b,
+	0xd80bf400,
+/* 0x0756: memx_init */
+	0x00f800f8,
+/* 0x0758: perf_recv */
+/* 0x075a: perf_init */
+	0x00f800f8,
+/* 0x075c: i2c_drive_scl */
+	0xf40036b0,
+	0x07f1110b,
+	0x04b607e0,
+	0x0001d006,
+	0x00f804bd,
+/* 0x0770: i2c_drive_scl_lo */
+	0x07e407f1,
+	0xd00604b6,
+	0x04bd0001,
+/* 0x077e: i2c_drive_sda */
+	0x36b000f8,
+	0x110bf400,
+	0x07e007f1,
+	0xd00604b6,
+	0x04bd0002,
+/* 0x0792: i2c_drive_sda_lo */
+	0x07f100f8,
+	0x04b607e4,
+	0x0002d006,
+	0x00f804bd,
+/* 0x07a0: i2c_sense_scl */
+	0xf10132f4,
+	0xb607c437,
+	0x33cf0634,
+	0x0431fd00,
+	0xf4060bf4,
+/* 0x07b6: i2c_sense_scl_done */
+	0x00f80131,
+/* 0x07b8: i2c_sense_sda */
+	0xf10132f4,
+	0xb607c437,
+	0x33cf0634,
+	0x0432fd00,
+	0xf4060bf4,
+/* 0x07ce: i2c_sense_sda_done */
+	0x00f80131,
+/* 0x07d0: i2c_raise_scl */
+	0x47f140f9,
+	0x37f00898,
+	0x5c21f501,
+/* 0x07dd: i2c_raise_scl_wait */
 	0xe8e7f107,
-	0x7f21f403,
+	0x7e21f403,
+	0x07a021f5,
+	0xb60901f4,
+	0x1bf40142,
+/* 0x07f1: i2c_raise_scl_done */
+	0xf840fcef,
+/* 0x07f5: i2c_start */
+	0xa021f500,
+	0x0d11f407,
+	0x07b821f5,
+	0xf40611f4,
+/* 0x0806: i2c_start_rep */
+	0x37f0300e,
+	0x5c21f500,
+	0x0137f007,
+	0x077e21f5,
 	0xb60076bb,
 	0x50f90465,
 	0xbb046594,
 	0x50bd0256,
 	0xfc0475fd,
-	0xcd21f550,
+	0xd021f550,
 	0x0464b607,
-	0xf51b11f4,
-	0xf007b521,
+/* 0x0833: i2c_start_send */
+	0xf01f11f4,
 	0x21f50037,
-	0xe7f10759,
+	0xe7f1077e,
 	0x21f41388,
-	0x013cf07f,
-/* 0x0907: i2c_bitr_done */
-	0xf80131f4,
-/* 0x0909: i2c_get_byte */
-	0x0057f000,
-/* 0x090f: i2c_get_byte_next */
-	0xb60847f0,
-	0x76bb0154,
-	0x0465b600,
-	0x659450f9,
-	0x0256bb04,
-	0x75fd50bd,
-	0xf550fc04,
-	0xb608c221,
-	0x11f40464,
-	0x0553fd2b,
-	0xf40142b6,
-	0x37f0d81b,
+	0x0037f07e,
+	0x075c21f5,
+	0x1388e7f1,
+/* 0x084f: i2c_start_out */
+	0xf87e21f4,
+/* 0x0851: i2c_stop */
+	0x0037f000,
+	0x075c21f5,
+	0xf50037f0,
+	0xf1077e21,
+	0xf403e8e7,
+	0x37f07e21,
+	0x5c21f501,
+	0x88e7f107,
+	0x7e21f413,
+	0xf50137f0,
+	0xf1077e21,
+	0xf41388e7,
+	0x00f87e21,
+/* 0x0884: i2c_bitw */
+	0x077e21f5,
+	0x03e8e7f1,
+	0xbb7e21f4,
+	0x65b60076,
+	0x9450f904,
+	0x56bb0465,
+	0xfd50bd02,
+	0x50fc0475,
+	0x07d021f5,
+	0xf40464b6,
+	0xe7f11811,
+	0x21f41388,
+	0x0037f07e,
+	0x075c21f5,
+	0x1388e7f1,
+/* 0x08c3: i2c_bitw_out */
+	0xf87e21f4,
+/* 0x08c5: i2c_bitr */
+	0x0137f000,
+	0x077e21f5,
+	0x03e8e7f1,
+	0xbb7e21f4,
+	0x65b60076,
+	0x9450f904,
+	0x56bb0465,
+	0xfd50bd02,
+	0x50fc0475,
+	0x07d021f5,
+	0xf40464b6,
+	0x21f51b11,
+	0x37f007b8,
+	0x5c21f500,
+	0x88e7f107,
+	0x7e21f413,
+	0xf4013cf0,
+/* 0x090a: i2c_bitr_done */
+	0x00f80131,
+/* 0x090c: i2c_get_byte */
+	0xf00057f0,
+/* 0x0912: i2c_get_byte_next */
+	0x54b60847,
 	0x0076bb01,
 	0xf90465b6,
 	0x04659450,
 	0xbd0256bb,
 	0x0475fd50,
 	0x21f550fc,
-	0x64b60881,
-/* 0x0959: i2c_get_byte_done */
-/* 0x095b: i2c_put_byte */
-	0xf000f804,
-/* 0x095e: i2c_put_byte_next */
-	0x42b60847,
-	0x3854ff01,
+	0x64b608c5,
+	0x2b11f404,
+	0xb60553fd,
+	0x1bf40142,
+	0x0137f0d8,
+	0xb60076bb,
+	0x50f90465,
+	0xbb046594,
+	0x50bd0256,
+	0xfc0475fd,
+	0x8421f550,
+	0x0464b608,
+/* 0x095c: i2c_get_byte_done */
+/* 0x095e: i2c_put_byte */
+	0x47f000f8,
+/* 0x0961: i2c_put_byte_next */
+	0x0142b608,
+	0xbb3854ff,
+	0x65b60076,
+	0x9450f904,
+	0x56bb0465,
+	0xfd50bd02,
+	0x50fc0475,
+	0x088421f5,
+	0xf40464b6,
+	0x46b03411,
+	0xd81bf400,
 	0xb60076bb,
 	0x50f90465,
 	0xbb046594,
 	0x50bd0256,
 	0xfc0475fd,
-	0x8121f550,
+	0xc521f550,
 	0x0464b608,
-	0xb03411f4,
-	0x1bf40046,
-	0x0076bbd8,
+	0xbb0f11f4,
+	0x36b00076,
+	0x061bf401,
+/* 0x09b7: i2c_put_byte_done */
+	0xf80132f4,
+/* 0x09b9: i2c_addr */
+	0x0076bb00,
 	0xf90465b6,
 	0x04659450,
 	0xbd0256bb,
 	0x0475fd50,
 	0x21f550fc,
-	0x64b608c2,
-	0x0f11f404,
-	0xb00076bb,
-	0x1bf40136,
-	0x0132f406,
-/* 0x09b4: i2c_put_byte_done */
-/* 0x09b6: i2c_addr */
-	0x76bb00f8,
+	0x64b607f5,
+	0x2911f404,
+	0x012ec3e7,
+	0xfd0134b6,
+	0x76bb0553,
 	0x0465b600,
 	0x659450f9,
 	0x0256bb04,
 	0x75fd50bd,
 	0xf550fc04,
-	0xb607f221,
-	0x11f40464,
-	0x2ec3e729,
-	0x0134b601,
-	0xbb0553fd,
+	0xb6095e21,
+/* 0x09fe: i2c_addr_done */
+	0x00f80464,
+/* 0x0a00: i2c_acquire_addr */
+	0xb6f8cec7,
+	0xe0b702e4,
+	0xee980d1c,
+/* 0x0a0f: i2c_acquire */
+	0xf500f800,
+	0xf40a0021,
+	0xd9f00421,
+	0x4021f403,
+/* 0x0a1e: i2c_release */
+	0x21f500f8,
+	0x21f40a00,
+	0x03daf004,
+	0xf84021f4,
+/* 0x0a2d: i2c_recv */
+	0x0132f400,
+	0xb6f8c1c7,
+	0x16b00214,
+	0x3a1ff528,
+	0xf413a001,
+	0x0032980c,
+	0x0ccc13a0,
+	0xf4003198,
+	0xd0f90231,
+	0xd0f9e0f9,
+	0x000067f1,
+	0x100063f1,
+	0xbb016792,
 	0x65b60076,
 	0x9450f904,
 	0x56bb0465,
 	0xfd50bd02,
 	0x50fc0475,
-	0x095b21f5,
-/* 0x09fb: i2c_addr_done */
-	0xf80464b6,
-/* 0x09fd: i2c_acquire_addr */
-	0xf8cec700,
-	0xb702e4b6,
-	0x980d1ce0,
-	0x00f800ee,
-/* 0x0a0c: i2c_acquire */
-	0x09fd21f5,
-	0xf00421f4,
-	0x21f403d9,
-/* 0x0a1b: i2c_release */
-	0xf500f83f,
-	0xf409fd21,
-	0xdaf00421,
-	0x3f21f403,
-/* 0x0a2a: i2c_recv */
-	0x32f400f8,
-	0xf8c1c701,
-	0xb00214b6,
-	0x1ff52816,
-	0x13a0013a,
-	0x32980cf4,
-	0xcc13a000,
-	0x0031980c,
-	0xf90231f4,
-	0xf9e0f9d0,
-	0x0067f1d0,
-	0x0063f100,
-	0x01679210,
-	0xb60076bb,
-	0x50f90465,
-	0xbb046594,
-	0x50bd0256,
-	0xfc0475fd,
-	0x0c21f550,
-	0x0464b60a,
-	0xd6b0d0fc,
-	0xb31bf500,
-	0x0057f000,
-	0xb60076bb,
-	0x50f90465,
-	0xbb046594,
-	0x50bd0256,
-	0xfc0475fd,
-	0xb621f550,
-	0x0464b609,
-	0x00d011f5,
-	0xbbe0c5c7,
+	0x0a0f21f5,
+	0xfc0464b6,
+	0x00d6b0d0,
+	0x00b31bf5,
+	0xbb0057f0,
 	0x65b60076,
 	0x9450f904,
 	0x56bb0465,
 	0xfd50bd02,
 	0x50fc0475,
-	0x095b21f5,
+	0x09b921f5,
 	0xf50464b6,
-	0xf000ad11,
-	0x76bb0157,
+	0xc700d011,
+	0x76bbe0c5,
 	0x0465b600,
 	0x659450f9,
 	0x0256bb04,
 	0x75fd50bd,
 	0xf550fc04,
-	0xb609b621,
+	0xb6095e21,
 	0x11f50464,
-	0x76bb008a,
-	0x0465b600,
-	0x659450f9,
-	0x0256bb04,
-	0x75fd50bd,
-	0xf550fc04,
-	0xb6090921,
-	0x11f40464,
-	0xe05bcb6a,
-	0xb60076bb,
-	0x50f90465,
-	0xbb046594,
-	0x50bd0256,
-	0xfc0475fd,
-	0x4e21f550,
-	0x0464b608,
-	0xbd025bb9,
-	0x430ef474,
-/* 0x0b30: i2c_recv_not_rd08 */
-	0xf401d6b0,
-	0x57f03d1b,
-	0xb621f500,
-	0x3311f409,
-	0xf5e0c5c7,
-	0xf4095b21,
-	0x57f02911,
-	0xb621f500,
-	0x1f11f409,
-	0xf5e0b5c7,
-	0xf4095b21,
-	0x21f51511,
-	0x74bd084e,
-	0xf408c5c7,
-	0x32f4091b,
-	0x030ef402,
-/* 0x0b70: i2c_recv_not_wr08 */
-/* 0x0b70: i2c_recv_done */
-	0xf5f8cec7,
-	0xfc0a1b21,
-	0xf4d0fce0,
-	0x7cb90a12,
-	0x3721f502,
-/* 0x0b85: i2c_recv_exit */
-/* 0x0b87: i2c_init */
-	0xf800f803,
-/* 0x0b89: test_recv */
-	0xd817f100,
-	0x0614b605,
-	0xb60011cf,
-	0x07f10110,
-	0x04b605d8,
-	0x0001d006,
-	0xe7f104bd,
-	0xe3f1d900,
-	0x21f5134f,
-	0x00f80257,
-/* 0x0bb0: test_init */
-	0x0800e7f1,
-	0x025721f5,
-/* 0x0bba: idle_recv */
+	0x57f000ad,
+	0x0076bb01,
+	0xf90465b6,
+	0x04659450,
+	0xbd0256bb,
+	0x0475fd50,
+	0x21f550fc,
+	0x64b609b9,
+	0x8a11f504,
+	0x0076bb00,
+	0xf90465b6,
+	0x04659450,
+	0xbd0256bb,
+	0x0475fd50,
+	0x21f550fc,
+	0x64b6090c,
+	0x6a11f404,
+	0xbbe05bcb,
+	0x65b60076,
+	0x9450f904,
+	0x56bb0465,
+	0xfd50bd02,
+	0x50fc0475,
+	0x085121f5,
+	0xb90464b6,
+	0x74bd025b,
+/* 0x0b33: i2c_recv_not_rd08 */
+	0xb0430ef4,
+	0x1bf401d6,
+	0x0057f03d,
+	0x09b921f5,
+	0xc73311f4,
+	0x21f5e0c5,
+	0x11f4095e,
+	0x0057f029,
+	0x09b921f5,
+	0xc71f11f4,
+	0x21f5e0b5,
+	0x11f4095e,
+	0x5121f515,
+	0xc774bd08,
+	0x1bf408c5,
+	0x0232f409,
+/* 0x0b73: i2c_recv_not_wr08 */
+/* 0x0b73: i2c_recv_done */
+	0xc7030ef4,
+	0x21f5f8ce,
+	0xe0fc0a1e,
+	0x12f4d0fc,
+	0x027cb90a,
+	0x033621f5,
+/* 0x0b88: i2c_recv_exit */
+/* 0x0b8a: i2c_init */
 	0x00f800f8,
-/* 0x0bbc: idle */
-	0xf10031f4,
-	0xb605d417,
-	0x11cf0614,
-	0x0110b600,
-	0x05d407f1,
-	0xd00604b6,
-	0x04bd0001,
-/* 0x0bd8: idle_loop */
-	0xf45817f0,
-/* 0x0bde: idle_proc */
-/* 0x0bde: idle_proc_exec */
-	0x10f90232,
-	0xf5021eb9,
-	0xfc034021,
-	0x0911f410,
-	0xf40231f4,
-/* 0x0bf2: idle_proc_next */
-	0x10b6ef0e,
-	0x061fb858,
-	0xf4e61bf4,
-	0x28f4dd02,
-	0xbb0ef400,
-	0x00000000,
+/* 0x0b8c: test_recv */
+	0x05d817f1,
+	0xcf0614b6,
+	0x10b60011,
+	0xd807f101,
+	0x0604b605,
+	0xbd0001d0,
+	0x00e7f104,
+	0x4fe3f1d9,
+	0x5621f513,
+/* 0x0bb3: test_init */
+	0xf100f802,
+	0xf50800e7,
+	0xf8025621,
+/* 0x0bbd: idle_recv */
+/* 0x0bbf: idle */
+	0xf400f800,
+	0x17f10031,
+	0x14b605d4,
+	0x0011cf06,
+	0xf10110b6,
+	0xb605d407,
+	0x01d00604,
+/* 0x0bdb: idle_loop */
+	0xf004bd00,
+	0x32f45817,
+/* 0x0be1: idle_proc */
+/* 0x0be1: idle_proc_exec */
+	0xb910f902,
+	0x21f5021e,
+	0x10fc033f,
+	0xf40911f4,
+	0x0ef40231,
+/* 0x0bf5: idle_proc_next */
+	0x5810b6ef,
+	0xf4061fb8,
+	0x02f4e61b,
+	0x0028f4dd,
+	0x00bb0ef4,
 	0x00000000,
 	0x00000000,
 	0x00000000,
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/pmu/fuc/gf119.fuc4.h b/drivers/gpu/drm/nouveau/nvkm/subdev/pmu/fuc/gf119.fuc4.h
index 7bf6b39ed205..2d5bdc539697 100644
--- a/drivers/gpu/drm/nouveau/nvkm/subdev/pmu/fuc/gf119.fuc4.h
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/pmu/fuc/gf119.fuc4.h
@@ -24,8 +24,8 @@ uint32_t gf119_pmu_data[] = {
 	0x00000000,
 /* 0x0058: proc_list_head */
 	0x54534f48,
-	0x00000492,
-	0x0000043b,
+	0x00000495,
+	0x0000043e,
 	0x00000000,
 	0x00000000,
 	0x00000000,
@@ -46,8 +46,8 @@ uint32_t gf119_pmu_data[] = {
 	0x00000000,
 	0x00000000,
 	0x584d454d,
-	0x00000680,
-	0x00000672,
+	0x00000683,
+	0x00000675,
 	0x00000000,
 	0x00000000,
 	0x00000000,
@@ -68,8 +68,8 @@ uint32_t gf119_pmu_data[] = {
 	0x00000000,
 	0x00000000,
 	0x46524550,
-	0x00000684,
-	0x00000682,
+	0x00000687,
+	0x00000685,
 	0x00000000,
 	0x00000000,
 	0x00000000,
@@ -90,8 +90,8 @@ uint32_t gf119_pmu_data[] = {
 	0x00000000,
 	0x00000000,
 	0x5f433249,
-	0x00000a9f,
-	0x00000942,
+	0x00000aa2,
+	0x00000945,
 	0x00000000,
 	0x00000000,
 	0x00000000,
@@ -112,8 +112,8 @@ uint32_t gf119_pmu_data[] = {
 	0x00000000,
 	0x00000000,
 	0x54534554,
-	0x00000ac2,
-	0x00000aa1,
+	0x00000ac5,
+	0x00000aa4,
 	0x00000000,
 	0x00000000,
 	0x00000000,
@@ -134,8 +134,8 @@ uint32_t gf119_pmu_data[] = {
 	0x00000000,
 	0x00000000,
 	0x454c4449,
-	0x00000ace,
-	0x00000acc,
+	0x00000ad1,
+	0x00000acf,
 	0x00000000,
 	0x00000000,
 	0x00000000,
@@ -229,26 +229,26 @@ uint32_t gf119_pmu_data[] = {
 /* 0x0370: memx_func_head */
 	0x00000001,
 	0x00000000,
-	0x000004c8,
+	0x000004cb,
 /* 0x037c: memx_func_next */
 	0x00000002,
 	0x00000000,
-	0x00000549,
+	0x0000054c,
 	0x00000003,
 	0x00000002,
-	0x000005cd,
+	0x000005d0,
 	0x00040004,
 	0x00000000,
-	0x000005e9,
+	0x000005ec,
 	0x00010005,
 	0x00000000,
-	0x00000603,
+	0x00000606,
 	0x00010006,
 	0x00000000,
-	0x000005c8,
+	0x000005cb,
 	0x00000007,
 	0x00000000,
-	0x0000060e,
+	0x00000611,
 /* 0x03c4: memx_func_tail */
 /* 0x03c4: memx_ts_start */
 	0x00000000,
@@ -916,821 +916,821 @@ uint32_t gf119_pmu_data[] = {
 };
 
 uint32_t gf119_pmu_code[] = {
-	0x03420ef5,
+	0x03410ef5,
 /* 0x0004: rd32 */
 	0x07a007f1,
 	0xbd000ed0,
-	0x01d7f004,
-	0xf101d3f0,
-	0xd007ac07,
-	0x04bd000d,
-/* 0x001c: rd32_wait */
-	0x07acd7f1,
-	0xf100ddcf,
-	0xf47000d4,
-	0xd7f1f51b,
-	0xddcf07a4,
-/* 0x0033: wr32 */
-	0xf100f800,
-	0xd007a007,
-	0x04bd000e,
-	0x07a407f1,
+	0x01d7f104,
+	0x01d3f000,
+	0x07ac07f1,
 	0xbd000dd0,
-	0x02d7f004,
-	0xf0f0d5f0,
-	0x07f101d3,
-	0x0dd007ac,
-/* 0x0057: wr32_wait */
-	0xf104bd00,
-	0xcf07acd7,
-	0xd4f100dd,
-	0x1bf47000,
-/* 0x0067: nsec */
-	0xf900f8f5,
-	0xf080f990,
-	0x88cf2c87,
-/* 0x0071: nsec_loop */
-	0x2c97f000,
-	0xbb0099cf,
-	0x9eb80298,
-	0xf41ef406,
-	0x90fc80fc,
-/* 0x0086: wait */
+/* 0x001d: rd32_wait */
+	0xacd7f104,
+	0x00ddcf07,
+	0x7000d4f1,
+	0xf1f51bf4,
+	0xcf07a4d7,
+	0x00f800dd,
+/* 0x0034: wr32 */
+	0x07a007f1,
+	0xbd000ed0,
+	0xa407f104,
+	0x000dd007,
+	0xd7f104bd,
+	0xd3f000f2,
+	0xac07f101,
+	0x000dd007,
+/* 0x0056: wr32_wait */
+	0xd7f104bd,
+	0xddcf07ac,
+	0x00d4f100,
+	0xf51bf470,
+/* 0x0066: nsec */
 	0x90f900f8,
 	0x87f080f9,
 	0x0088cf2c,
-/* 0x0090: wait_loop */
-	0xf402eeb9,
-	0xdab90421,
-	0x04adfd02,
-	0xf406acb8,
-	0x97f0120b,
-	0x0099cf2c,
-	0xb80298bb,
-	0x1ef4069b,
-/* 0x00b1: wait_done */
-	0xfc80fce2,
-/* 0x00b7: intr_watchdog */
-	0x9800f890,
-	0x96b003e9,
-	0x2a0bf400,
-	0xbb9a0a98,
-	0x1cf4029a,
-	0x01d7f00f,
-	0x028121f5,
-	0x0ef494bd,
-/* 0x00d5: intr_watchdog_next_time */
-	0x9b0a9815,
-	0xf400a6b0,
-	0x9ab8090b,
-	0x061cf406,
-/* 0x00e4: intr_watchdog_next_time_set */
-/* 0x00e7: intr_watchdog_next_proc */
-	0x809b0980,
-	0xe0b603e9,
-	0x68e6b158,
-	0xc61bf402,
-/* 0x00f6: intr */
-	0x00f900f8,
-	0x80f904bd,
-	0xa0f990f9,
-	0xc0f9b0f9,
-	0xe0f9d0f9,
-	0xf7f0f0f9,
-	0x0188fe00,
-	0x87f180f9,
-	0x88cf05d0,
-	0x0180b600,
-	0x05d007f1,
-	0xbd0008d0,
-	0x0887f004,
-	0xc40088cf,
-	0x0bf40289,
-	0x9b008020,
-	0xf458e7f0,
-	0x0998b721,
-	0x0096b09b,
-	0xf00e0bf4,
-	0x09d03407,
-	0x8004bd00,
-/* 0x014e: intr_skip_watchdog */
-	0x89e49a09,
-	0x0bf40800,
-	0x8897f13c,
-	0x0099cf06,
-	0xf4029ac4,
-	0xc7f1260b,
-	0xcccf04c0,
-	0xf1c0f900,
-	0xf14f48e7,
-	0xf05453e3,
-	0x21f500d7,
-	0xc0fc02e6,
-	0x04c007f1,
-	0xbd000cd0,
-/* 0x0185: intr_subintr_skip_fifo */
-	0x8807f104,
-	0x0009d006,
-/* 0x018e: intr_skip_subintr */
-	0x97f104bd,
-	0x90bd00e0,
-	0xf00489fd,
-	0x08d00407,
-	0xfc04bd00,
-	0x0088fe80,
-	0xe0fcf0fc,
-	0xc0fcd0fc,
-	0xa0fcb0fc,
-	0x80fc90fc,
-	0x32f400fc,
-/* 0x01bb: ticks_from_ns */
-	0xf901f800,
+/* 0x0070: nsec_loop */
+	0xcf2c97f0,
+	0x98bb0099,
+	0x069eb802,
+	0xfcf41ef4,
+	0xf890fc80,
+/* 0x0085: wait */
+	0xf990f900,
+	0x2c87f080,
+/* 0x008f: wait_loop */
+	0xb90088cf,
+	0x21f402ee,
+	0x02dab904,
+	0xb804adfd,
+	0x0bf406ac,
+	0x2c97f012,
+	0xbb0099cf,
+	0x9bb80298,
+	0xe21ef406,
+/* 0x00b0: wait_done */
+	0x90fc80fc,
+/* 0x00b6: intr_watchdog */
+	0xe99800f8,
+	0x0096b003,
+	0x982a0bf4,
+	0x9abb9a0a,
+	0x0f1cf402,
+	0xf501d7f0,
+	0xbd028021,
+	0x150ef494,
+/* 0x00d4: intr_watchdog_next_time */
+	0xb09b0a98,
+	0x0bf400a6,
+	0x069ab809,
+/* 0x00e3: intr_watchdog_next_time_set */
+	0x80061cf4,
+/* 0x00e6: intr_watchdog_next_proc */
+	0xe9809b09,
+	0x58e0b603,
+	0x0268e6b1,
+	0xf8c61bf4,
+/* 0x00f5: intr */
+	0xbd00f900,
+	0xf980f904,
+	0xf9a0f990,
+	0xf9c0f9b0,
+	0xf9e0f9d0,
+	0x00f7f0f0,
+	0xf90188fe,
+	0xd087f180,
+	0x0088cf05,
+	0xf10180b6,
+	0xd005d007,
+	0x04bd0008,
+	0xcf0887f0,
+	0x89c40088,
+	0x200bf402,
+	0xf09b0080,
+	0x21f458e7,
+	0x9b0998b6,
+	0xf40096b0,
+	0x07f00e0b,
+	0x0009d034,
+	0x098004bd,
+/* 0x014d: intr_skip_watchdog */
+	0x0089e49a,
+	0x3c0bf408,
+	0x068897f1,
+	0xc40099cf,
+	0x0bf4029a,
+	0xc0c7f126,
+	0x00cccf04,
+	0xe7f1c0f9,
+	0xe3f14f48,
+	0xd7f05453,
+	0xe521f500,
+	0xf1c0fc02,
+	0xd004c007,
+	0x04bd000c,
+/* 0x0184: intr_subintr_skip_fifo */
+	0x068807f1,
+	0xbd0009d0,
+/* 0x018d: intr_skip_subintr */
+	0xe097f104,
+	0xfd90bd00,
+	0x07f00489,
+	0x0008d004,
+	0x80fc04bd,
+	0xfc0088fe,
+	0xfce0fcf0,
+	0xfcc0fcd0,
+	0xfca0fcb0,
+	0xfc80fc90,
+	0x0032f400,
+/* 0x01ba: ticks_from_ns */
+	0xc0f901f8,
+	0xd7f1b0f9,
+	0xd3f00144,
+	0xab21f500,
+	0xe8ccec03,
+	0x00b4b003,
+	0xec120bf4,
+	0xf103e8ee,
+	0xf00144d7,
+	0x21f500d3,
+/* 0x01e2: ticks_from_ns_quit */
+	0xceb903ab,
+	0xfcb0fc02,
+/* 0x01eb: ticks_from_us */
+	0xf900f8c0,
 	0xf1b0f9c0,
 	0xf00144d7,
 	0x21f500d3,
-	0xccec03a8,
-	0xb4b003e8,
-	0x120bf400,
-	0x03e8eeec,
-	0x0144d7f1,
-	0xf500d3f0,
-/* 0x01e3: ticks_from_ns_quit */
-	0xb903a821,
-	0xb0fc02ce,
-	0x00f8c0fc,
-/* 0x01ec: ticks_from_us */
-	0xb0f9c0f9,
-	0x0144d7f1,
-	0xf500d3f0,
-	0xb903a821,
-	0xb4b002ce,
-	0x050bf400,
-/* 0x0206: ticks_from_us_quit */
-	0xb0fce4bd,
-	0x00f8c0fc,
-/* 0x020c: ticks_to_us */
-	0x0144d7f1,
-	0xff00d3f0,
-	0x00f8eced,
-/* 0x0218: timer */
-	0x80f990f9,
-	0x981032f4,
-	0x86b003f8,
-	0x531cf400,
-	0x07f084bd,
-	0x0008d038,
-	0x87f004bd,
-	0x0088cf34,
-	0xbb9a0998,
-	0xe9bb0298,
-	0x03fe8000,
-	0xcf0887f0,
-	0x84f00088,
-	0x201bf402,
-	0xcf3487f0,
-	0xe0b80088,
-	0x090bf406,
-	0xf406e8b8,
-/* 0x0262: timer_reset */
-	0x07f00e1c,
-	0x000ed034,
-	0x0e8004bd,
-/* 0x026d: timer_enable */
-	0x0187f09a,
-	0xd03807f0,
-	0x04bd0008,
-/* 0x0278: timer_done */
-	0xfc1031f4,
+	0xceb903ab,
+	0x00b4b002,
+	0xbd050bf4,
+/* 0x0205: ticks_from_us_quit */
+	0xfcb0fce4,
+/* 0x020b: ticks_to_us */
+	0xf100f8c0,
+	0xf00144d7,
+	0xedff00d3,
+/* 0x0217: timer */
+	0xf900f8ec,
+	0xf480f990,
+	0xf8981032,
+	0x0086b003,
+	0xbd531cf4,
+	0x3807f084,
+	0xbd0008d0,
+	0x3487f004,
+	0x980088cf,
+	0x98bb9a09,
+	0x00e9bb02,
+	0xf003fe80,
+	0x88cf0887,
+	0x0284f000,
+	0xf0201bf4,
+	0x88cf3487,
+	0x06e0b800,
+	0xb8090bf4,
+	0x1cf406e8,
+/* 0x0261: timer_reset */
+	0x3407f00e,
+	0xbd000ed0,
+	0x9a0e8004,
+/* 0x026c: timer_enable */
+	0xf00187f0,
+	0x08d03807,
+/* 0x0277: timer_done */
+	0xf404bd00,
+	0x80fc1031,
+	0x00f890fc,
+/* 0x0280: send_proc */
+	0x90f980f9,
+	0x9805e898,
+	0x86f004e9,
+	0x0689b804,
+	0xc42a0bf4,
+	0x88940398,
+	0x1880b604,
+	0x98008ebb,
+	0x8a8000fa,
+	0x018d8000,
+	0x80028c80,
+	0x90b6038b,
+	0x0794f001,
+	0xf404e980,
+/* 0x02ba: send_done */
+	0x90fc0231,
+	0x00f880fc,
+/* 0x02c0: find */
+	0x87f080f9,
+	0x0131f458,
+/* 0x02c8: find_loop */
+	0xb8008a98,
+	0x0bf406ae,
+	0x5880b610,
+	0x026886b1,
+	0xf4f01bf4,
+/* 0x02de: find_done */
+	0x8eb90132,
+	0xf880fc02,
+/* 0x02e5: send */
+	0xc021f500,
+	0x9701f402,
+/* 0x02ee: recv */
+	0x90f900f8,
+	0xe89880f9,
+	0x04e99805,
+	0xb80132f4,
+	0x0bf40689,
+	0x0389c43d,
+	0xf00180b6,
+	0xe8800784,
+	0x02ea9805,
+	0x8ffef0f9,
+	0xb9f0f901,
+	0x999402ef,
+	0x00e9bb04,
+	0x9818e0b6,
+	0xec9803eb,
+	0x01ed9802,
+	0xf900ee98,
+	0xfef0fca5,
+	0x31f400f8,
+/* 0x033b: recv_done */
+	0xfcf0fc01,
 	0xf890fc80,
-/* 0x0281: send_proc */
-	0xf980f900,
-	0x05e89890,
-	0xf004e998,
-	0x89b80486,
-	0x2a0bf406,
-	0x940398c4,
-	0x80b60488,
-	0x008ebb18,
-	0x8000fa98,
-	0x8d80008a,
-	0x028c8001,
-	0xb6038b80,
-	0x94f00190,
-	0x04e98007,
-/* 0x02bb: send_done */
-	0xfc0231f4,
-	0xf880fc90,
-/* 0x02c1: find */
-	0xf080f900,
-	0x31f45887,
-/* 0x02c9: find_loop */
-	0x008a9801,
-	0xf406aeb8,
-	0x80b6100b,
-	0x6886b158,
-	0xf01bf402,
-/* 0x02df: find_done */
-	0xb90132f4,
-	0x80fc028e,
-/* 0x02e6: send */
-	0x21f500f8,
-	0x01f402c1,
-/* 0x02ef: recv */
-	0xf900f897,
-	0x9880f990,
-	0xe99805e8,
-	0x0132f404,
-	0xf40689b8,
-	0x89c43d0b,
-	0x0180b603,
-	0x800784f0,
-	0xea9805e8,
-	0xfef0f902,
-	0xf0f9018f,
-	0x9402efb9,
-	0xe9bb0499,
-	0x18e0b600,
-	0x9803eb98,
-	0xed9802ec,
-	0x00ee9801,
-	0xf0fca5f9,
-	0xf400f8fe,
-	0xf0fc0131,
-/* 0x033c: recv_done */
-	0x90fc80fc,
-/* 0x0342: init */
-	0x17f100f8,
-	0x11cf0108,
-	0x0911e700,
-	0x0814b601,
-	0xf10014fe,
-	0xf000e017,
-	0x07f00013,
-	0x0001d01c,
-	0x17f004bd,
-	0x1407f0ff,
+/* 0x0341: init */
+	0x0817f100,
+	0x0011cf01,
+	0x010911e7,
+	0xfe0814b6,
+	0x17f10014,
+	0x13f000e0,
+	0x1c07f000,
 	0xbd0001d0,
-	0x0217f004,
-	0x080015f1,
-	0xd01007f0,
-	0x04bd0001,
-	0x00f617f1,
-	0xfe0013f0,
-	0x31f40010,
-	0x0117f010,
-	0xd03807f0,
+	0xff17f004,
+	0xd01407f0,
 	0x04bd0001,
-/* 0x0397: init_proc */
-	0x9858f7f0,
-	0x16b001f1,
-	0xfa0bf400,
-	0xf0b615f9,
-	0xf20ef458,
-/* 0x03a8: mulu32_32_64 */
-	0x20f910f9,
-	0x40f930f9,
-	0x9510e195,
-	0xc4bd10d2,
-	0xedffb4bd,
-	0x301dffc0,
-	0xf10234b9,
-	0xb6ffff34,
-	0x45b61034,
-	0x00c3bb10,
-	0xff01b4bb,
-	0x34b930e2,
-	0xff34f102,
-	0x1034b6ff,
-	0xbb1045b6,
-	0xb4bb00c3,
-	0x3012ff01,
-	0xfc00b3bb,
-	0xfc30fc40,
-	0xf810fc20,
-/* 0x03f9: host_send */
-	0xb017f100,
-	0x0011cf04,
-	0x04a027f1,
-	0xb80022cf,
-	0x0bf40612,
-	0x071ec42f,
-	0xb704ee94,
-	0x980270e0,
-	0xec9803eb,
-	0x01ed9802,
-	0xf500ee98,
-	0xb602e621,
-	0x1ec40110,
-	0xb007f10f,
-	0x000ed004,
-	0x0ef404bd,
-/* 0x0439: host_send_done */
-/* 0x043b: host_recv */
-	0xf100f8c3,
-	0xf14e4917,
-	0xb8525413,
-	0x0bf406e1,
-/* 0x0449: host_recv_wait */
-	0xcc17f1b3,
-	0x0011cf04,
-	0x04c827f1,
-	0xf00022cf,
-	0x12b80816,
-	0xec0bf406,
-	0xb60723c4,
-	0x30b70434,
-	0x3b8002f0,
-	0x023c8003,
-	0x80013d80,
-	0x20b6003e,
-	0x0f24f001,
-	0x04c807f1,
-	0xbd0002d0,
-	0x4027f004,
-	0xd00007f0,
-	0x04bd0002,
-/* 0x0492: host_init */
+	0xf10217f0,
+	0xf0080015,
+	0x01d01007,
+	0xf104bd00,
+	0xf000f517,
+	0x14f10013,
+	0x10feffff,
+	0x1031f400,
+	0xf00117f0,
+	0x01d03807,
+	0xf004bd00,
+/* 0x039a: init_proc */
+	0xf19858f7,
+	0x0016b001,
+	0xf9fa0bf4,
+	0x58f0b615,
+/* 0x03ab: mulu32_32_64 */
+	0xf9f20ef4,
+	0xf920f910,
+	0x9540f930,
+	0xd29510e1,
+	0xbdc4bd10,
+	0xc0edffb4,
+	0xb9301dff,
+	0x34f10234,
+	0x34b6ffff,
+	0x1045b610,
+	0xbb00c3bb,
+	0xe2ff01b4,
+	0x0234b930,
+	0xffff34f1,
+	0xb61034b6,
+	0xc3bb1045,
+	0x01b4bb00,
+	0xbb3012ff,
+	0x40fc00b3,
+	0x20fc30fc,
+	0x00f810fc,
+/* 0x03fc: host_send */
+	0x04b017f1,
+	0xf10011cf,
+	0xcf04a027,
+	0x12b80022,
+	0x2f0bf406,
+	0x94071ec4,
+	0xe0b704ee,
+	0xeb980270,
+	0x02ec9803,
+	0x9801ed98,
+	0x21f500ee,
+	0x10b602e5,
+	0x0f1ec401,
+	0x04b007f1,
+	0xbd000ed0,
+	0xc30ef404,
+/* 0x043c: host_send_done */
+/* 0x043e: host_recv */
 	0x17f100f8,
-	0x14b60080,
-	0x7015f110,
-	0xd007f102,
-	0x0001d004,
-	0x17f104bd,
-	0x14b60080,
-	0xf015f110,
-	0xdc07f102,
-	0x0001d004,
-	0x17f004bd,
-	0xc407f101,
-	0x0001d004,
-	0x00f804bd,
-/* 0x04c8: memx_func_enter */
-	0x162067f1,
-	0xf55d77f1,
-	0xffff73f1,
-	0xf4026eb9,
-	0xd8b90421,
-	0x0487fd02,
-	0x80f960f9,
-	0xe0fcd0fc,
-	0xf13321f4,
-	0xf1fffe77,
+	0x13f14e49,
+	0xe1b85254,
+	0xb30bf406,
+/* 0x044c: host_recv_wait */
+	0x04cc17f1,
+	0xf10011cf,
+	0xcf04c827,
+	0x16f00022,
+	0x0612b808,
+	0xc4ec0bf4,
+	0x34b60723,
+	0xf030b704,
+	0x033b8002,
+	0x80023c80,
+	0x3e80013d,
+	0x0120b600,
+	0xf10f24f0,
+	0xd004c807,
+	0x04bd0002,
+	0xf04027f0,
+	0x02d00007,
+	0xf804bd00,
+/* 0x0495: host_init */
+	0x8017f100,
+	0x1014b600,
+	0x027015f1,
+	0x04d007f1,
+	0xbd0001d0,
+	0x8017f104,
+	0x1014b600,
+	0x02f015f1,
+	0x04dc07f1,
+	0xbd0001d0,
+	0x0117f004,
+	0x04c407f1,
+	0xbd0001d0,
+/* 0x04cb: memx_func_enter */
+	0xf100f804,
+	0xf1162067,
+	0xf1f55d77,
 	0xb9ffff73,
 	0x21f4026e,
 	0x02d8b904,
 	0xf90487fd,
 	0xfc80f960,
 	0xf4e0fcd0,
-	0x67f13321,
-	0x6eb926f0,
+	0x77f13421,
+	0x73f1fffe,
+	0x6eb9ffff,
 	0x0421f402,
 	0xfd02d8b9,
 	0x60f90487,
 	0xd0fc80f9,
 	0x21f4e0fc,
-	0x0467f033,
-	0x07e007f1,
+	0xf067f134,
+	0x026eb926,
+	0xb90421f4,
+	0x87fd02d8,
+	0xf960f904,
+	0xfcd0fc80,
+	0x3421f4e0,
+	0xf10467f0,
+	0xd007e007,
+	0x04bd0006,
+/* 0x0534: memx_func_enter_wait */
+	0x07c067f1,
+	0xf00066cf,
+	0x0bf40464,
+	0x2c67f0f6,
+	0x800066cf,
+	0x00f8f106,
+/* 0x054c: memx_func_leave */
+	0xcf2c67f0,
+	0x06800066,
+	0x0467f0f2,
+	0x07e407f1,
 	0xbd0006d0,
-/* 0x0531: memx_func_enter_wait */
+/* 0x0561: memx_func_leave_wait */
 	0xc067f104,
 	0x0066cf07,
 	0xf40464f0,
-	0x67f0f60b,
-	0x0066cf2c,
-	0xf8f10680,
-/* 0x0549: memx_func_leave */
-	0x2c67f000,
-	0x800066cf,
-	0x67f0f206,
-	0xe407f104,
-	0x0006d007,
-/* 0x055e: memx_func_leave_wait */
-	0x67f104bd,
-	0x66cf07c0,
-	0x0464f000,
-	0xf1f61bf4,
-	0xf126f067,
-	0xf0000177,
+	0x67f1f61b,
+	0x77f126f0,
+	0x73f00001,
+	0x026eb900,
+	0xb90421f4,
+	0x87fd02d8,
+	0xf960f905,
+	0xfcd0fc80,
+	0x3421f4e0,
+	0x162067f1,
+	0xf4026eb9,
+	0xd8b90421,
+	0x0587fd02,
+	0x80f960f9,
+	0xe0fcd0fc,
+	0xf13421f4,
+	0xf00aa277,
 	0x6eb90073,
 	0x0421f402,
 	0xfd02d8b9,
 	0x60f90587,
 	0xd0fc80f9,
 	0x21f4e0fc,
-	0x2067f133,
-	0x026eb916,
-	0xb90421f4,
-	0x87fd02d8,
-	0xf960f905,
-	0xfcd0fc80,
-	0x3321f4e0,
-	0x0aa277f1,
-	0xb90073f0,
-	0x21f4026e,
-	0x02d8b904,
-	0xf90587fd,
-	0xfc80f960,
-	0xf4e0fcd0,
-	0x00f83321,
-/* 0x05c8: memx_func_wait_vblank */
-	0xf80410b6,
-/* 0x05cd: memx_func_wr32 */
-	0x00169800,
-	0xb6011598,
-	0x60f90810,
-	0xd0fc50f9,
-	0x21f4e0fc,
-	0x0242b633,
-	0xf8e91bf4,
-/* 0x05e9: memx_func_wait */
-	0x2c87f000,
-	0x980088cf,
-	0x1d98001e,
-	0x021c9801,
-	0xb6031b98,
-	0x21f41010,
-/* 0x0603: memx_func_delay */
-	0x9800f886,
-	0x10b6001e,
-	0x6721f404,
-/* 0x060e: memx_func_train */
-	0x00f800f8,
-/* 0x0610: memx_exec */
-	0xd0f9e0f9,
-	0xb902c1b9,
-/* 0x061a: memx_exec_next */
-	0x139802b2,
+/* 0x05cb: memx_func_wait_vblank */
+	0xb600f834,
+	0x00f80410,
+/* 0x05d0: memx_func_wr32 */
+	0x98001698,
+	0x10b60115,
+	0xf960f908,
+	0xfcd0fc50,
+	0x3421f4e0,
+	0xf40242b6,
+	0x00f8e91b,
+/* 0x05ec: memx_func_wait */
+	0xcf2c87f0,
+	0x1e980088,
+	0x011d9800,
+	0x98021c98,
+	0x10b6031b,
+	0x8521f410,
+/* 0x0606: memx_func_delay */
+	0x1e9800f8,
 	0x0410b600,
-	0x01f034e7,
-	0x01e033e7,
-	0xf00132b6,
-	0x35980c30,
-	0xb855f9de,
-	0x1ef40612,
-	0xf10b98e4,
-	0xbbf20c98,
-	0xb7f102cb,
-	0xbbcf07c4,
-	0xfcd0fc00,
-	0xe621f5e0,
-/* 0x0653: memx_info */
-	0x7000f802,
-	0x0bf401c6,
-/* 0x0659: memx_info_data */
-	0xccc7f10e,
-	0x00b7f103,
-	0x0b0ef408,
-/* 0x0664: memx_info_train */
-	0x0bccc7f1,
-	0x0100b7f1,
-/* 0x066c: memx_info_send */
-	0x02e621f5,
-/* 0x0672: memx_recv */
-	0xd6b000f8,
-	0x9b0bf401,
-	0xf400d6b0,
-	0x00f8d80b,
-/* 0x0680: memx_init */
-/* 0x0682: perf_recv */
-	0x00f800f8,
-/* 0x0684: perf_init */
-/* 0x0686: i2c_drive_scl */
-	0x36b000f8,
-	0x0e0bf400,
-	0x07e007f1,
-	0xbd0001d0,
-/* 0x0697: i2c_drive_scl_lo */
-	0xf100f804,
-	0xd007e407,
+	0xf86621f4,
+/* 0x0611: memx_func_train */
+/* 0x0613: memx_exec */
+	0xf900f800,
+	0xb9d0f9e0,
+	0xb2b902c1,
+/* 0x061d: memx_exec_next */
+	0x00139802,
+	0xe70410b6,
+	0xe701f034,
+	0xb601e033,
+	0x30f00132,
+	0xde35980c,
+	0x12b855f9,
+	0xe41ef406,
+	0x98f10b98,
+	0xcbbbf20c,
+	0xc4b7f102,
+	0x00bbcf07,
+	0xe0fcd0fc,
+	0x02e521f5,
+/* 0x0656: memx_info */
+	0xc67000f8,
+	0x0e0bf401,
+/* 0x065c: memx_info_data */
+	0x03ccc7f1,
+	0x0800b7f1,
+/* 0x0667: memx_info_train */
+	0xf10b0ef4,
+	0xf10bccc7,
+/* 0x066f: memx_info_send */
+	0xf50100b7,
+	0xf802e521,
+/* 0x0675: memx_recv */
+	0x01d6b000,
+	0xb09b0bf4,
+	0x0bf400d6,
+/* 0x0683: memx_init */
+	0xf800f8d8,
+/* 0x0685: perf_recv */
+/* 0x0687: perf_init */
+	0xf800f800,
+/* 0x0689: i2c_drive_scl */
+	0x0036b000,
+	0xf10e0bf4,
+	0xd007e007,
 	0x04bd0001,
-/* 0x06a2: i2c_drive_sda */
-	0x36b000f8,
-	0x0e0bf400,
-	0x07e007f1,
-	0xbd0002d0,
-/* 0x06b3: i2c_drive_sda_lo */
-	0xf100f804,
-	0xd007e407,
+/* 0x069a: i2c_drive_scl_lo */
+	0x07f100f8,
+	0x01d007e4,
+	0xf804bd00,
+/* 0x06a5: i2c_drive_sda */
+	0x0036b000,
+	0xf10e0bf4,
+	0xd007e007,
 	0x04bd0002,
-/* 0x06be: i2c_sense_scl */
+/* 0x06b6: i2c_drive_sda_lo */
+	0x07f100f8,
+	0x02d007e4,
+	0xf804bd00,
+/* 0x06c1: i2c_sense_scl */
+	0x0132f400,
+	0x07c437f1,
+	0xfd0033cf,
+	0x0bf40431,
+	0x0131f406,
+/* 0x06d4: i2c_sense_scl_done */
+/* 0x06d6: i2c_sense_sda */
 	0x32f400f8,
 	0xc437f101,
 	0x0033cf07,
-	0xf40431fd,
+	0xf40432fd,
 	0x31f4060b,
-/* 0x06d1: i2c_sense_scl_done */
-/* 0x06d3: i2c_sense_sda */
-	0xf400f801,
-	0x37f10132,
-	0x33cf07c4,
-	0x0432fd00,
-	0xf4060bf4,
-/* 0x06e6: i2c_sense_sda_done */
-	0x00f80131,
-/* 0x06e8: i2c_raise_scl */
-	0x47f140f9,
-	0x37f00898,
-	0x8621f501,
-/* 0x06f5: i2c_raise_scl_wait */
-	0xe8e7f106,
-	0x6721f403,
-	0x06be21f5,
-	0xb60901f4,
-	0x1bf40142,
-/* 0x0709: i2c_raise_scl_done */
-	0xf840fcef,
-/* 0x070d: i2c_start */
-	0xbe21f500,
-	0x0d11f406,
-	0x06d321f5,
-	0xf40611f4,
-/* 0x071e: i2c_start_rep */
-	0x37f0300e,
-	0x8621f500,
-	0x0137f006,
-	0x06a221f5,
-	0xb60076bb,
-	0x50f90465,
-	0xbb046594,
-	0x50bd0256,
-	0xfc0475fd,
-	0xe821f550,
-	0x0464b606,
-/* 0x074b: i2c_start_send */
-	0xf01f11f4,
-	0x21f50037,
-	0xe7f106a2,
-	0x21f41388,
-	0x0037f067,
-	0x068621f5,
-	0x1388e7f1,
-/* 0x0767: i2c_start_out */
-	0xf86721f4,
-/* 0x0769: i2c_stop */
-	0x0037f000,
-	0x068621f5,
-	0xf50037f0,
-	0xf106a221,
-	0xf403e8e7,
-	0x37f06721,
-	0x8621f501,
-	0x88e7f106,
-	0x6721f413,
-	0xf50137f0,
-	0xf106a221,
-	0xf41388e7,
-	0x00f86721,
-/* 0x079c: i2c_bitw */
-	0x06a221f5,
+/* 0x06e9: i2c_sense_sda_done */
+/* 0x06eb: i2c_raise_scl */
+	0xf900f801,
+	0x9847f140,
+	0x0137f008,
+	0x068921f5,
+/* 0x06f8: i2c_raise_scl_wait */
 	0x03e8e7f1,
-	0xbb6721f4,
+	0xf56621f4,
+	0xf406c121,
+	0x42b60901,
+	0xef1bf401,
+/* 0x070c: i2c_raise_scl_done */
+	0x00f840fc,
+/* 0x0710: i2c_start */
+	0x06c121f5,
+	0xf50d11f4,
+	0xf406d621,
+	0x0ef40611,
+/* 0x0721: i2c_start_rep */
+	0x0037f030,
+	0x068921f5,
+	0xf50137f0,
+	0xbb06a521,
 	0x65b60076,
 	0x9450f904,
 	0x56bb0465,
 	0xfd50bd02,
 	0x50fc0475,
-	0x06e821f5,
+	0x06eb21f5,
 	0xf40464b6,
-	0xe7f11811,
+/* 0x074e: i2c_start_send */
+	0x37f01f11,
+	0xa521f500,
+	0x88e7f106,
+	0x6621f413,
+	0xf50037f0,
+	0xf1068921,
+	0xf41388e7,
+/* 0x076a: i2c_start_out */
+	0x00f86621,
+/* 0x076c: i2c_stop */
+	0xf50037f0,
+	0xf0068921,
+	0x21f50037,
+	0xe7f106a5,
+	0x21f403e8,
+	0x0137f066,
+	0x068921f5,
+	0x1388e7f1,
+	0xf06621f4,
+	0x21f50137,
+	0xe7f106a5,
 	0x21f41388,
-	0x0037f067,
-	0x068621f5,
+/* 0x079f: i2c_bitw */
+	0xf500f866,
+	0xf106a521,
+	0xf403e8e7,
+	0x76bb6621,
+	0x0465b600,
+	0x659450f9,
+	0x0256bb04,
+	0x75fd50bd,
+	0xf550fc04,
+	0xb606eb21,
+	0x11f40464,
+	0x88e7f118,
+	0x6621f413,
+	0xf50037f0,
+	0xf1068921,
+	0xf41388e7,
+/* 0x07de: i2c_bitw_out */
+	0x00f86621,
+/* 0x07e0: i2c_bitr */
+	0xf50137f0,
+	0xf106a521,
+	0xf403e8e7,
+	0x76bb6621,
+	0x0465b600,
+	0x659450f9,
+	0x0256bb04,
+	0x75fd50bd,
+	0xf550fc04,
+	0xb606eb21,
+	0x11f40464,
+	0xd621f51b,
+	0x0037f006,
+	0x068921f5,
 	0x1388e7f1,
-/* 0x07db: i2c_bitw_out */
-	0xf86721f4,
-/* 0x07dd: i2c_bitr */
-	0x0137f000,
-	0x06a221f5,
-	0x03e8e7f1,
-	0xbb6721f4,
-	0x65b60076,
-	0x9450f904,
-	0x56bb0465,
-	0xfd50bd02,
-	0x50fc0475,
-	0x06e821f5,
-	0xf40464b6,
-	0x21f51b11,
-	0x37f006d3,
-	0x8621f500,
-	0x88e7f106,
-	0x6721f413,
-	0xf4013cf0,
-/* 0x0822: i2c_bitr_done */
-	0x00f80131,
-/* 0x0824: i2c_get_byte */
-	0xf00057f0,
-/* 0x082a: i2c_get_byte_next */
-	0x54b60847,
-	0x0076bb01,
-	0xf90465b6,
-	0x04659450,
-	0xbd0256bb,
-	0x0475fd50,
-	0x21f550fc,
-	0x64b607dd,
-	0x2b11f404,
-	0xb60553fd,
-	0x1bf40142,
-	0x0137f0d8,
+	0xf06621f4,
+	0x31f4013c,
+/* 0x0825: i2c_bitr_done */
+/* 0x0827: i2c_get_byte */
+	0xf000f801,
+	0x47f00057,
+/* 0x082d: i2c_get_byte_next */
+	0x0154b608,
 	0xb60076bb,
 	0x50f90465,
 	0xbb046594,
 	0x50bd0256,
 	0xfc0475fd,
-	0x9c21f550,
+	0xe021f550,
 	0x0464b607,
-/* 0x0874: i2c_get_byte_done */
-/* 0x0876: i2c_put_byte */
-	0x47f000f8,
-/* 0x0879: i2c_put_byte_next */
-	0x0142b608,
-	0xbb3854ff,
+	0xfd2b11f4,
+	0x42b60553,
+	0xd81bf401,
+	0xbb0137f0,
+	0x65b60076,
+	0x9450f904,
+	0x56bb0465,
+	0xfd50bd02,
+	0x50fc0475,
+	0x079f21f5,
+/* 0x0877: i2c_get_byte_done */
+	0xf80464b6,
+/* 0x0879: i2c_put_byte */
+	0x0847f000,
+/* 0x087c: i2c_put_byte_next */
+	0xff0142b6,
+	0x76bb3854,
+	0x0465b600,
+	0x659450f9,
+	0x0256bb04,
+	0x75fd50bd,
+	0xf550fc04,
+	0xb6079f21,
+	0x11f40464,
+	0x0046b034,
+	0xbbd81bf4,
 	0x65b60076,
 	0x9450f904,
 	0x56bb0465,
 	0xfd50bd02,
 	0x50fc0475,
-	0x079c21f5,
+	0x07e021f5,
 	0xf40464b6,
-	0x46b03411,
-	0xd81bf400,
+	0x76bb0f11,
+	0x0136b000,
+	0xf4061bf4,
+/* 0x08d2: i2c_put_byte_done */
+	0x00f80132,
+/* 0x08d4: i2c_addr */
 	0xb60076bb,
 	0x50f90465,
 	0xbb046594,
 	0x50bd0256,
 	0xfc0475fd,
-	0xdd21f550,
+	0x1021f550,
 	0x0464b607,
-	0xbb0f11f4,
-	0x36b00076,
-	0x061bf401,
-/* 0x08cf: i2c_put_byte_done */
-	0xf80132f4,
-/* 0x08d1: i2c_addr */
-	0x0076bb00,
+	0xe72911f4,
+	0xb6012ec3,
+	0x53fd0134,
+	0x0076bb05,
 	0xf90465b6,
 	0x04659450,
 	0xbd0256bb,
 	0x0475fd50,
 	0x21f550fc,
-	0x64b6070d,
-	0x2911f404,
-	0x012ec3e7,
-	0xfd0134b6,
-	0x76bb0553,
-	0x0465b600,
-	0x659450f9,
-	0x0256bb04,
-	0x75fd50bd,
-	0xf550fc04,
-	0xb6087621,
-/* 0x0916: i2c_addr_done */
-	0x00f80464,
-/* 0x0918: i2c_acquire_addr */
-	0xb6f8cec7,
-	0xe0b705e4,
-	0x00f8d014,
-/* 0x0924: i2c_acquire */
-	0x091821f5,
-	0xf00421f4,
-	0x21f403d9,
-/* 0x0933: i2c_release */
-	0xf500f833,
-	0xf4091821,
-	0xdaf00421,
-	0x3321f403,
-/* 0x0942: i2c_recv */
-	0x32f400f8,
-	0xf8c1c701,
-	0xb00214b6,
-	0x1ff52816,
-	0x13a0013a,
-	0x32980cf4,
-	0xcc13a000,
-	0x0031980c,
-	0xf90231f4,
-	0xf9e0f9d0,
-	0x0067f1d0,
-	0x0063f100,
-	0x01679210,
-	0xb60076bb,
-	0x50f90465,
-	0xbb046594,
-	0x50bd0256,
-	0xfc0475fd,
-	0x2421f550,
-	0x0464b609,
-	0xd6b0d0fc,
-	0xb31bf500,
-	0x0057f000,
-	0xb60076bb,
-	0x50f90465,
-	0xbb046594,
-	0x50bd0256,
-	0xfc0475fd,
-	0xd121f550,
-	0x0464b608,
-	0x00d011f5,
-	0xbbe0c5c7,
+	0x64b60879,
+/* 0x0919: i2c_addr_done */
+/* 0x091b: i2c_acquire_addr */
+	0xc700f804,
+	0xe4b6f8ce,
+	0x14e0b705,
+/* 0x0927: i2c_acquire */
+	0xf500f8d0,
+	0xf4091b21,
+	0xd9f00421,
+	0x3421f403,
+/* 0x0936: i2c_release */
+	0x21f500f8,
+	0x21f4091b,
+	0x03daf004,
+	0xf83421f4,
+/* 0x0945: i2c_recv */
+	0x0132f400,
+	0xb6f8c1c7,
+	0x16b00214,
+	0x3a1ff528,
+	0xf413a001,
+	0x0032980c,
+	0x0ccc13a0,
+	0xf4003198,
+	0xd0f90231,
+	0xd0f9e0f9,
+	0x000067f1,
+	0x100063f1,
+	0xbb016792,
 	0x65b60076,
 	0x9450f904,
 	0x56bb0465,
 	0xfd50bd02,
 	0x50fc0475,
-	0x087621f5,
+	0x092721f5,
+	0xfc0464b6,
+	0x00d6b0d0,
+	0x00b31bf5,
+	0xbb0057f0,
+	0x65b60076,
+	0x9450f904,
+	0x56bb0465,
+	0xfd50bd02,
+	0x50fc0475,
+	0x08d421f5,
 	0xf50464b6,
-	0xf000ad11,
-	0x76bb0157,
+	0xc700d011,
+	0x76bbe0c5,
 	0x0465b600,
 	0x659450f9,
 	0x0256bb04,
 	0x75fd50bd,
 	0xf550fc04,
-	0xb608d121,
+	0xb6087921,
 	0x11f50464,
-	0x76bb008a,
-	0x0465b600,
-	0x659450f9,
-	0x0256bb04,
-	0x75fd50bd,
-	0xf550fc04,
-	0xb6082421,
-	0x11f40464,
-	0xe05bcb6a,
-	0xb60076bb,
-	0x50f90465,
-	0xbb046594,
-	0x50bd0256,
-	0xfc0475fd,
-	0x6921f550,
-	0x0464b607,
-	0xbd025bb9,
-	0x430ef474,
-/* 0x0a48: i2c_recv_not_rd08 */
-	0xf401d6b0,
-	0x57f03d1b,
-	0xd121f500,
-	0x3311f408,
-	0xf5e0c5c7,
-	0xf4087621,
-	0x57f02911,
-	0xd121f500,
-	0x1f11f408,
-	0xf5e0b5c7,
-	0xf4087621,
-	0x21f51511,
-	0x74bd0769,
-	0xf408c5c7,
-	0x32f4091b,
-	0x030ef402,
-/* 0x0a88: i2c_recv_not_wr08 */
-/* 0x0a88: i2c_recv_done */
-	0xf5f8cec7,
-	0xfc093321,
-	0xf4d0fce0,
-	0x7cb90a12,
-	0xe621f502,
-/* 0x0a9d: i2c_recv_exit */
-/* 0x0a9f: i2c_init */
+	0x57f000ad,
+	0x0076bb01,
+	0xf90465b6,
+	0x04659450,
+	0xbd0256bb,
+	0x0475fd50,
+	0x21f550fc,
+	0x64b608d4,
+	0x8a11f504,
+	0x0076bb00,
+	0xf90465b6,
+	0x04659450,
+	0xbd0256bb,
+	0x0475fd50,
+	0x21f550fc,
+	0x64b60827,
+	0x6a11f404,
+	0xbbe05bcb,
+	0x65b60076,
+	0x9450f904,
+	0x56bb0465,
+	0xfd50bd02,
+	0x50fc0475,
+	0x076c21f5,
+	0xb90464b6,
+	0x74bd025b,
+/* 0x0a4b: i2c_recv_not_rd08 */
+	0xb0430ef4,
+	0x1bf401d6,
+	0x0057f03d,
+	0x08d421f5,
+	0xc73311f4,
+	0x21f5e0c5,
+	0x11f40879,
+	0x0057f029,
+	0x08d421f5,
+	0xc71f11f4,
+	0x21f5e0b5,
+	0x11f40879,
+	0x6c21f515,
+	0xc774bd07,
+	0x1bf408c5,
+	0x0232f409,
+/* 0x0a8b: i2c_recv_not_wr08 */
+/* 0x0a8b: i2c_recv_done */
+	0xc7030ef4,
+	0x21f5f8ce,
+	0xe0fc0936,
+	0x12f4d0fc,
+	0x027cb90a,
+	0x02e521f5,
+/* 0x0aa0: i2c_recv_exit */
+/* 0x0aa2: i2c_init */
+	0x00f800f8,
+/* 0x0aa4: test_recv */
+	0x05d817f1,
+	0xb60011cf,
+	0x07f10110,
+	0x01d005d8,
+	0xf104bd00,
+	0xf1d900e7,
+	0xf5134fe3,
+	0xf8021721,
+/* 0x0ac5: test_init */
+	0x00e7f100,
+	0x1721f508,
+/* 0x0acf: idle_recv */
 	0xf800f802,
-/* 0x0aa1: test_recv */
-	0xd817f100,
-	0x0011cf05,
-	0xf10110b6,
-	0xd005d807,
-	0x04bd0001,
-	0xd900e7f1,
-	0x134fe3f1,
-	0x021821f5,
-/* 0x0ac2: test_init */
-	0xe7f100f8,
-	0x21f50800,
-	0x00f80218,
-/* 0x0acc: idle_recv */
-/* 0x0ace: idle */
-	0x31f400f8,
-	0xd417f100,
-	0x0011cf05,
-	0xf10110b6,
-	0xd005d407,
-	0x04bd0001,
-/* 0x0ae4: idle_loop */
-	0xf45817f0,
-/* 0x0aea: idle_proc */
-/* 0x0aea: idle_proc_exec */
-	0x10f90232,
-	0xf5021eb9,
-	0xfc02ef21,
-	0x0911f410,
-	0xf40231f4,
-/* 0x0afe: idle_proc_next */
-	0x10b6ef0e,
-	0x061fb858,
-	0xf4e61bf4,
-	0x28f4dd02,
-	0xc10ef400,
-	0x00000000,
+/* 0x0ad1: idle */
+	0x0031f400,
+	0x05d417f1,
+	0xb60011cf,
+	0x07f10110,
+	0x01d005d4,
+/* 0x0ae7: idle_loop */
+	0xf004bd00,
+	0x32f45817,
+/* 0x0aed: idle_proc */
+/* 0x0aed: idle_proc_exec */
+	0xb910f902,
+	0x21f5021e,
+	0x10fc02ee,
+	0xf40911f4,
+	0x0ef40231,
+/* 0x0b01: idle_proc_next */
+	0x5810b6ef,
+	0xf4061fb8,
+	0x02f4e61b,
+	0x0028f4dd,
+	0x00c10ef4,
 	0x00000000,
 	0x00000000,
 	0x00000000,
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/pmu/fuc/gk208.fuc5.h b/drivers/gpu/drm/nouveau/nvkm/subdev/pmu/fuc/gk208.fuc5.h
index 8a2b628642ac..3c731ff12871 100644
--- a/drivers/gpu/drm/nouveau/nvkm/subdev/pmu/fuc/gk208.fuc5.h
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/pmu/fuc/gk208.fuc5.h
@@ -24,8 +24,8 @@ uint32_t gk208_pmu_data[] = {
 	0x00000000,
 /* 0x0058: proc_list_head */
 	0x54534f48,
-	0x00000447,
-	0x000003f8,
+	0x0000042c,
+	0x000003df,
 	0x00000000,
 	0x00000000,
 	0x00000000,
@@ -46,8 +46,8 @@ uint32_t gk208_pmu_data[] = {
 	0x00000000,
 	0x00000000,
 	0x584d454d,
-	0x00000621,
-	0x00000613,
+	0x000005f3,
+	0x000005e5,
 	0x00000000,
 	0x00000000,
 	0x00000000,
@@ -68,8 +68,8 @@ uint32_t gk208_pmu_data[] = {
 	0x00000000,
 	0x00000000,
 	0x46524550,
-	0x00000625,
-	0x00000623,
+	0x000005f7,
+	0x000005f5,
 	0x00000000,
 	0x00000000,
 	0x00000000,
@@ -90,8 +90,8 @@ uint32_t gk208_pmu_data[] = {
 	0x00000000,
 	0x00000000,
 	0x5f433249,
-	0x00000a29,
-	0x000008d0,
+	0x000009f8,
+	0x000008a2,
 	0x00000000,
 	0x00000000,
 	0x00000000,
@@ -112,8 +112,8 @@ uint32_t gk208_pmu_data[] = {
 	0x00000000,
 	0x00000000,
 	0x54534554,
-	0x00000a4a,
-	0x00000a2b,
+	0x00000a16,
+	0x000009fa,
 	0x00000000,
 	0x00000000,
 	0x00000000,
@@ -134,8 +134,8 @@ uint32_t gk208_pmu_data[] = {
 	0x00000000,
 	0x00000000,
 	0x454c4449,
-	0x00000a55,
-	0x00000a53,
+	0x00000a21,
+	0x00000a1f,
 	0x00000000,
 	0x00000000,
 	0x00000000,
@@ -229,26 +229,26 @@ uint32_t gk208_pmu_data[] = {
 /* 0x0370: memx_func_head */
 	0x00000001,
 	0x00000000,
-	0x00000477,
+	0x0000045c,
 /* 0x037c: memx_func_next */
 	0x00000002,
 	0x00000000,
-	0x000004f4,
+	0x000004cf,
 	0x00000003,
 	0x00000002,
-	0x00000574,
+	0x00000546,
 	0x00040004,
 	0x00000000,
-	0x00000591,
+	0x00000563,
 	0x00010005,
 	0x00000000,
-	0x000005ab,
+	0x0000057d,
 	0x00010006,
 	0x00000000,
-	0x0000056f,
+	0x00000541,
 	0x00000007,
 	0x00000000,
-	0x000005b7,
+	0x00000589,
 /* 0x03c4: memx_func_tail */
 /* 0x03c4: memx_ts_start */
 	0x00000000,
@@ -916,784 +916,771 @@ uint32_t gk208_pmu_data[] = {
 };
 
 uint32_t gk208_pmu_code[] = {
-	0x03100ef5,
+	0x02f90ef5,
 /* 0x0004: rd32 */
 	0xf607a040,
 	0x04bd000e,
-	0xd3f0010d,
+	0x0100018d,
+	0xf607ac40,
+	0x04bd000d,
+/* 0x0018: rd32_wait */
+	0xcf07ac4d,
+	0xd4f100dd,
+	0x1bf47000,
+	0x07a44df6,
+	0xf800ddcf,
+/* 0x002d: wr32 */
+	0x07a04000,
+	0xbd000ef6,
+	0x07a44004,
+	0xbd000df6,
+	0x00f28d04,
 	0x07ac4001,
 	0xbd000df6,
-/* 0x0019: rd32_wait */
+/* 0x0049: wr32_wait */
 	0x07ac4d04,
 	0xf100ddcf,
 	0xf47000d4,
-	0xa44df61b,
-	0x00ddcf07,
-/* 0x002e: wr32 */
-	0xa04000f8,
-	0x000ef607,
-	0xa44004bd,
-	0x000df607,
-	0x020d04bd,
-	0xf0f0d5f0,
-	0xac4001d3,
-	0x000df607,
-/* 0x004e: wr32_wait */
-	0xac4d04bd,
-	0x00ddcf07,
-	0x7000d4f1,
-	0xf8f61bf4,
-/* 0x005d: nsec */
-	0xf990f900,
-	0xcf2c0880,
-/* 0x0066: nsec_loop */
-	0x2c090088,
-	0xbb0099cf,
-	0x9ea60298,
-	0xfcf61ef4,
-	0xf890fc80,
-/* 0x0079: wait */
-	0xf990f900,
-	0xcf2c0880,
-/* 0x0082: wait_loop */
-	0xeeb20088,
-	0x0000047e,
-	0xadfddab2,
-	0xf4aca604,
-	0x2c09100b,
-	0xbb0099cf,
-	0x9ba60298,
-/* 0x009f: wait_done */
-	0xfce61ef4,
-	0xf890fc80,
-/* 0x00a5: intr_watchdog */
-	0x03e99800,
+	0x00f8f61b,
+/* 0x0058: nsec */
+	0x80f990f9,
+	0x88cf2c08,
+/* 0x0061: nsec_loop */
+	0xcf2c0900,
+	0x98bb0099,
+	0xf49ea602,
+	0x80fcf61e,
+	0x00f890fc,
+/* 0x0074: wait */
+	0x80f990f9,
+	0x88cf2c08,
+/* 0x007d: wait_loop */
+	0x7eeeb200,
+	0xb2000004,
+	0x04adfdda,
+	0x0bf4aca6,
+	0xcf2c0910,
+	0x98bb0099,
+	0xf49ba602,
+/* 0x009a: wait_done */
+	0x80fce61e,
+	0x00f890fc,
+/* 0x00a0: intr_watchdog */
+	0xb003e998,
+	0x0bf40096,
+	0x9a0a9828,
+	0xf4029abb,
+	0x010d0e1c,
+	0x00023e7e,
+	0x0ef494bd,
+/* 0x00bd: intr_watchdog_next_time */
+	0x9b0a9814,
+	0xf400a6b0,
+	0x9aa6080b,
+/* 0x00cb: intr_watchdog_next_time_set */
+	0xb5061cf4,
+/* 0x00ce: intr_watchdog_next_proc */
+	0xe9b59b09,
+	0x58e0b603,
+	0x0268e6b1,
+	0xf8c81bf4,
+/* 0x00dd: intr */
+	0xbd00f900,
+	0xf980f904,
+	0xf9a0f990,
+	0xf9c0f9b0,
+	0xf9e0f9d0,
+	0xfe000ff0,
+	0x80f90188,
+	0xcf045048,
+	0x80b60088,
+	0x04504001,
+	0xbd0008f6,
+	0xcf080804,
+	0x89c40088,
+	0x1f0bf402,
+	0x0e9b00b5,
+	0x00a07e58,
+	0x9b099800,
 	0xf40096b0,
-	0x0a98280b,
-	0x029abb9a,
-	0x0d0e1cf4,
-	0x02557e01,
-	0xf494bd00,
-/* 0x00c2: intr_watchdog_next_time */
-	0x0a98140e,
-	0x00a6b09b,
-	0xa6080bf4,
-	0x061cf49a,
-/* 0x00d0: intr_watchdog_next_time_set */
-/* 0x00d3: intr_watchdog_next_proc */
-	0xb59b09b5,
-	0xe0b603e9,
-	0x68e6b158,
-	0xc81bf402,
-/* 0x00e2: intr */
-	0x00f900f8,
-	0x80f904bd,
-	0xa0f990f9,
-	0xc0f9b0f9,
-	0xe0f9d0f9,
-	0x000ff0f9,
-	0xf90188fe,
-	0x04504880,
-	0xb60088cf,
-	0x50400180,
-	0x0008f604,
-	0x080804bd,
-	0xc40088cf,
-	0x0bf40289,
-	0x9b00b51f,
-	0xa57e580e,
-	0x09980000,
-	0x0096b09b,
-	0x000d0bf4,
-	0x0009f634,
-	0x09b504bd,
-/* 0x0135: intr_skip_watchdog */
-	0x0089e49a,
-	0x360bf408,
-	0xcf068849,
-	0x9ac40099,
-	0x220bf402,
-	0xcf04c04c,
-	0xc0f900cc,
-	0xf14f484e,
-	0x0d5453e3,
-	0x02b67e00,
-	0x40c0fc00,
-	0x0cf604c0,
-/* 0x0167: intr_subintr_skip_fifo */
-	0x4004bd00,
-	0x09f60688,
-/* 0x016f: intr_skip_subintr */
-	0x4904bd00,
-	0x90bd00e0,
-	0x000489fd,
-	0x0008f604,
-	0x80fc04bd,
-	0xfc0088fe,
-	0xfce0fcf0,
-	0xfcc0fcd0,
-	0xfca0fcb0,
-	0xfc80fc90,
-	0x0032f400,
-/* 0x019a: ticks_from_ns */
-	0xc0f901f8,
-	0xd7f1b0f9,
-	0xd3f00144,
-	0x6b21f500,
-	0xe8ccec03,
-	0x00b4b003,
-	0xec120bf4,
-	0xf103e8ee,
-	0xf00144d7,
-	0x21f500d3,
-/* 0x01c2: ticks_from_ns_quit */
-	0xceb2036b,
+	0x34000d0b,
+	0xbd0009f6,
+	0x9a09b504,
+/* 0x0130: intr_skip_watchdog */
+	0x080089e4,
+	0x49340bf4,
+	0x99cf0688,
+	0x029ac400,
+	0x4c200bf4,
+	0xcccf04c0,
+	0xdec0f900,
+	0x54534f48,
+	0x9f7e000d,
+	0xc0fc0002,
+	0xf604c040,
+	0x04bd000c,
+/* 0x0160: intr_subintr_skip_fifo */
+	0xf6068840,
+	0x04bd0009,
+/* 0x0168: intr_skip_subintr */
+	0xbd00e049,
+	0x0489fd90,
+	0x08f60400,
+	0xfc04bd00,
+	0x0088fe80,
+	0xe0fcf0fc,
+	0xc0fcd0fc,
+	0xa0fcb0fc,
+	0x80fc90fc,
+	0x32f400fc,
+/* 0x0193: ticks_from_ns */
+	0xf901f800,
+	0x4db0f9c0,
+	0x527e0144,
+	0xccec0003,
+	0xb4b003e8,
+	0x0e0bf400,
+	0x03e8eeec,
+	0x7e01444d,
+/* 0x01b3: ticks_from_ns_quit */
+	0xb2000352,
+	0xfcb0fcce,
+/* 0x01bb: ticks_from_us */
+	0xf900f8c0,
+	0x4db0f9c0,
+	0x527e0144,
+	0xceb20003,
+	0xf400b4b0,
+	0xe4bd050b,
+/* 0x01d0: ticks_from_us_quit */
 	0xc0fcb0fc,
-/* 0x01ca: ticks_from_us */
-	0xc0f900f8,
-	0xd7f1b0f9,
-	0xd3f00144,
-	0x6b21f500,
-	0xb0ceb203,
-	0x0bf400b4,
-/* 0x01e3: ticks_from_us_quit */
-	0xfce4bd05,
-	0xf8c0fcb0,
-/* 0x01e9: ticks_to_us */
-	0x44d7f100,
-	0x00d3f001,
-	0xf8ecedff,
-/* 0x01f5: timer */
-	0xf990f900,
-	0x1032f480,
-	0xb003f898,
-	0x1cf40086,
-	0x0084bd4a,
-	0x0008f638,
-	0x340804bd,
-	0x980088cf,
-	0x98bb9a09,
-	0x00e9bb02,
-	0x0803feb5,
-	0x0088cf08,
-	0xf40284f0,
-	0x34081c1b,
-	0xa60088cf,
-	0x080bf4e0,
-	0x1cf4e8a6,
-/* 0x0239: timer_reset */
-	0xf634000d,
-	0x04bd000e,
-/* 0x0243: timer_enable */
-	0x089a0eb5,
-	0xf6380001,
-	0x04bd0008,
-/* 0x024c: timer_done */
-	0xfc1031f4,
+/* 0x01d6: ticks_to_us */
+	0x444d00f8,
+	0xecedff01,
+/* 0x01de: timer */
+	0x90f900f8,
+	0x32f480f9,
+	0x03f89810,
+	0xf40086b0,
+	0x84bd4a1c,
+	0x08f63800,
+	0x0804bd00,
+	0x0088cf34,
+	0xbb9a0998,
+	0xe9bb0298,
+	0x03feb500,
+	0x88cf0808,
+	0x0284f000,
+	0x081c1bf4,
+	0x0088cf34,
+	0x0bf4e0a6,
+	0xf4e8a608,
+/* 0x0222: timer_reset */
+	0x34000d1c,
+	0xbd000ef6,
+	0x9a0eb504,
+/* 0x022c: timer_enable */
+	0x38000108,
+	0xbd0008f6,
+/* 0x0235: timer_done */
+	0x1031f404,
+	0x90fc80fc,
+/* 0x023e: send_proc */
+	0x80f900f8,
+	0xe89890f9,
+	0x04e99805,
+	0xa60486f0,
+	0x2a0bf489,
+	0x940398c4,
+	0x80b60488,
+	0x008ebb18,
+	0xb500fa98,
+	0x8db5008a,
+	0x028cb501,
+	0xb6038bb5,
+	0x94f00190,
+	0x04e9b507,
+/* 0x0277: send_done */
+	0xfc0231f4,
+	0xf880fc90,
+/* 0x027d: find */
+	0x0880f900,
+	0x0131f458,
+/* 0x0284: find_loop */
+	0xa6008a98,
+	0x100bf4ae,
+	0xb15880b6,
+	0xf4026886,
+	0x32f4f11b,
+/* 0x0299: find_done */
+	0xfc8eb201,
+/* 0x029f: send */
+	0x7e00f880,
+	0xf400027d,
+	0x00f89b01,
+/* 0x02a8: recv */
+	0x80f990f9,
+	0x9805e898,
+	0x32f404e9,
+	0xf489a601,
+	0x89c43c0b,
+	0x0180b603,
+	0xb50784f0,
+	0xea9805e8,
+	0xfef0f902,
+	0xf0f9018f,
+	0x9994efb2,
+	0x00e9bb04,
+	0x9818e0b6,
+	0xec9803eb,
+	0x01ed9802,
+	0xf900ee98,
+	0xfef0fca5,
+	0x31f400f8,
+/* 0x02f3: recv_done */
+	0xfcf0fc01,
 	0xf890fc80,
-/* 0x0255: send_proc */
-	0xf980f900,
-	0x05e89890,
-	0xf004e998,
-	0x89a60486,
-	0xc42a0bf4,
-	0x88940398,
-	0x1880b604,
-	0x98008ebb,
-	0x8ab500fa,
-	0x018db500,
-	0xb5028cb5,
-	0x90b6038b,
-	0x0794f001,
-	0xf404e9b5,
-/* 0x028e: send_done */
-	0x90fc0231,
-	0x00f880fc,
-/* 0x0294: find */
-	0x580880f9,
-/* 0x029b: find_loop */
-	0x980131f4,
-	0xaea6008a,
-	0xb6100bf4,
-	0x86b15880,
-	0x1bf40268,
-	0x0132f4f1,
-/* 0x02b0: find_done */
-	0x80fc8eb2,
-/* 0x02b6: send */
-	0x947e00f8,
-	0x01f40002,
-/* 0x02bf: recv */
-	0xf900f89b,
-	0x9880f990,
-	0xe99805e8,
-	0x0132f404,
-	0x0bf489a6,
-	0x0389c43c,
-	0xf00180b6,
-	0xe8b50784,
-	0x02ea9805,
-	0x8ffef0f9,
-	0xb2f0f901,
-	0x049994ef,
-	0xb600e9bb,
-	0xeb9818e0,
-	0x02ec9803,
-	0x9801ed98,
-	0xa5f900ee,
-	0xf8fef0fc,
-	0x0131f400,
-/* 0x030a: recv_done */
-	0x80fcf0fc,
-	0x00f890fc,
-/* 0x0310: init */
-	0xcf010841,
-	0x11e70011,
-	0x14b60109,
-	0x0014fe08,
-	0xf000e041,
-	0x1c000013,
-	0xbd0001f6,
-	0x00ff0104,
-	0x0001f614,
-	0x020104bd,
-	0x080015f1,
-	0x01f61000,
-	0x4104bd00,
-	0x13f000e2,
-	0x0010fe00,
-	0x011031f4,
-	0xf6380001,
+/* 0x02f9: init */
+	0x01084100,
+	0xe70011cf,
+	0xb6010911,
+	0x14fe0814,
+	0x00e04100,
+	0x01f61c00,
+	0x0104bd00,
+	0xf61400ff,
 	0x04bd0001,
-/* 0x035a: init_proc */
-	0xf198580f,
-	0x0016b001,
-	0xf9fa0bf4,
-	0x58f0b615,
-/* 0x036b: mulu32_32_64 */
-	0xf9f20ef4,
-	0xf920f910,
-	0x9540f930,
-	0xd29510e1,
-	0xbdc4bd10,
-	0xc0edffb4,
-	0xb2301dff,
+	0x15f10201,
+	0x10000800,
+	0xbd0001f6,
+	0x00dd4104,
+	0xffff14f1,
+	0xf40010fe,
+	0x01011031,
+	0x01f63800,
+	0x0f04bd00,
+/* 0x0341: init_proc */
+	0x01f19858,
+	0xf40016b0,
+	0x15f9fa0b,
+	0xf458f0b6,
+/* 0x0352: mulu32_32_64 */
+	0x10f9f20e,
+	0x30f920f9,
+	0xe19540f9,
+	0x10d29510,
+	0xb4bdc4bd,
+	0xffc0edff,
+	0x34b2301d,
+	0xffff34f1,
+	0xb61034b6,
+	0xc3bb1045,
+	0x01b4bb00,
+	0xb230e2ff,
 	0xff34f134,
 	0x1034b6ff,
 	0xbb1045b6,
 	0xb4bb00c3,
-	0x30e2ff01,
-	0x34f134b2,
-	0x34b6ffff,
-	0x1045b610,
-	0xbb00c3bb,
-	0x12ff01b4,
-	0x00b3bb30,
-	0x30fc40fc,
-	0x10fc20fc,
-/* 0x03ba: host_send */
-	0xb04100f8,
-	0x0011cf04,
-	0xcf04a042,
-	0x12a60022,
-	0xc42e0bf4,
-	0xee94071e,
-	0x70e0b704,
-	0x03eb9802,
-	0x9802ec98,
-	0xee9801ed,
-	0x02b67e00,
-	0x0110b600,
-	0x400f1ec4,
-	0x0ef604b0,
-	0xf404bd00,
-/* 0x03f6: host_send_done */
-	0x00f8c70e,
-/* 0x03f8: host_recv */
-	0xf14e4941,
-	0xa6525413,
-	0xb90bf4e1,
-/* 0x0404: host_recv_wait */
-	0xcf04cc41,
-	0xc8420011,
-	0x0022cf04,
-	0xa60816f0,
-	0xef0bf412,
-	0xb60723c4,
-	0x30b70434,
-	0x3bb502f0,
-	0x023cb503,
-	0xb5013db5,
-	0x20b6003e,
-	0x0f24f001,
-	0xf604c840,
-	0x04bd0002,
-	0x00004002,
+	0x3012ff01,
+	0xfc00b3bb,
+	0xfc30fc40,
+	0xf810fc20,
+/* 0x03a1: host_send */
+	0x04b04100,
+	0x420011cf,
+	0x22cf04a0,
+	0xf412a600,
+	0x1ec42e0b,
+	0x04ee9407,
+	0x0270e0b7,
+	0x9803eb98,
+	0xed9802ec,
+	0x00ee9801,
+	0x00029f7e,
+	0xc40110b6,
+	0xb0400f1e,
+	0x000ef604,
+	0x0ef404bd,
+/* 0x03dd: host_send_done */
+/* 0x03df: host_recv */
+	0xd100f8c7,
+	0x52544e49,
+	0x0bf4e1a6,
+/* 0x03e9: host_recv_wait */
+	0x04cc41bb,
+	0x420011cf,
+	0x22cf04c8,
+	0x0816f000,
+	0x0bf412a6,
+	0x0723c4ef,
+	0xb70434b6,
+	0xb502f030,
+	0x3cb5033b,
+	0x013db502,
+	0xb6003eb5,
+	0x24f00120,
+	0x04c8400f,
 	0xbd0002f6,
-/* 0x0447: host_init */
-	0x4100f804,
-	0x14b60080,
-	0x7015f110,
-	0x04d04002,
-	0xbd0001f6,
-	0x00804104,
-	0xf11014b6,
-	0x4002f015,
-	0x01f604dc,
-	0x0104bd00,
-	0x04c44001,
-	0xbd0001f6,
-/* 0x0477: memx_func_enter */
-	0xf100f804,
-	0xf1162067,
-	0xf1f55d77,
-	0xb2ffff73,
-	0x00047e6e,
-	0xfdd8b200,
-	0x60f90487,
-	0xd0fc80f9,
-	0x2e7ee0fc,
-	0x77f10000,
-	0x73f1fffe,
-	0x6eb2ffff,
-	0x0000047e,
-	0x87fdd8b2,
-	0xf960f904,
-	0xfcd0fc80,
-	0x002e7ee0,
-	0xf067f100,
-	0x7e6eb226,
+	0x00400204,
+	0x0002f600,
+	0x00f804bd,
+/* 0x042c: host_init */
+	0xb6008041,
+	0x15f11014,
+	0xd0400270,
+	0x0001f604,
+	0x804104bd,
+	0x1014b600,
+	0x02f015f1,
+	0xf604dc40,
+	0x04bd0001,
+	0xc4400101,
+	0x0001f604,
+	0x00f804bd,
+/* 0x045c: memx_func_enter */
+	0x162067f1,
+	0xf55d77f1,
+	0x047e6eb2,
+	0xd8b20000,
+	0xf90487fd,
+	0xfc80f960,
+	0x7ee0fcd0,
+	0x0700002d,
+	0x7e6eb2fe,
 	0xb2000004,
 	0x0487fdd8,
 	0x80f960f9,
 	0xe0fcd0fc,
-	0x00002e7e,
-	0xe0400406,
-	0x0006f607,
-/* 0x04de: memx_func_enter_wait */
-	0xc04604bd,
-	0x0066cf07,
-	0xf40464f0,
-	0x2c06f70b,
-	0xb50066cf,
-	0x00f8f106,
-/* 0x04f4: memx_func_leave */
-	0x66cf2c06,
-	0xf206b500,
-	0xe4400406,
-	0x0006f607,
-/* 0x0506: memx_func_leave_wait */
-	0xc04604bd,
-	0x0066cf07,
-	0xf40464f0,
-	0x67f1f71b,
-	0x77f126f0,
-	0x73f00001,
-	0x7e6eb200,
-	0xb2000004,
-	0x0587fdd8,
-	0x80f960f9,
-	0xe0fcd0fc,
-	0x00002e7e,
-	0x162067f1,
+	0x00002d7e,
+	0x26f067f1,
 	0x047e6eb2,
 	0xd8b20000,
-	0xf90587fd,
+	0xf90487fd,
 	0xfc80f960,
 	0x7ee0fcd0,
-	0xf100002e,
-	0xf00aa277,
-	0x6eb20073,
+	0x0600002d,
+	0x07e04004,
+	0xbd0006f6,
+/* 0x04b9: memx_func_enter_wait */
+	0x07c04604,
+	0xf00066cf,
+	0x0bf40464,
+	0xcf2c06f7,
+	0x06b50066,
+/* 0x04cf: memx_func_leave */
+	0x0600f8f1,
+	0x0066cf2c,
+	0x06f206b5,
+	0x07e44004,
+	0xbd0006f6,
+/* 0x04e1: memx_func_leave_wait */
+	0x07c04604,
+	0xf00066cf,
+	0x1bf40464,
+	0xf067f1f7,
+	0xb2010726,
+	0x00047e6e,
+	0xfdd8b200,
+	0x60f90587,
+	0xd0fc80f9,
+	0x2d7ee0fc,
+	0x67f10000,
+	0x6eb21620,
 	0x0000047e,
 	0x87fdd8b2,
 	0xf960f905,
 	0xfcd0fc80,
-	0x002e7ee0,
-/* 0x056f: memx_func_wait_vblank */
-	0xb600f800,
-	0x00f80410,
-/* 0x0574: memx_func_wr32 */
-	0x98001698,
-	0x10b60115,
-	0xf960f908,
-	0xfcd0fc50,
-	0x002e7ee0,
-	0x0242b600,
-	0xf8e81bf4,
-/* 0x0591: memx_func_wait */
-	0xcf2c0800,
-	0x1e980088,
-	0x011d9800,
-	0x98021c98,
-	0x10b6031b,
-	0x00797e10,
-/* 0x05ab: memx_func_delay */
-	0x9800f800,
-	0x10b6001e,
-	0x005d7e04,
-/* 0x05b7: memx_func_train */
-	0xf800f800,
-/* 0x05b9: memx_exec */
-	0xf9e0f900,
-	0xb2c1b2d0,
-/* 0x05c1: memx_exec_next */
-	0x001398b2,
-	0xe70410b6,
-	0xe701f034,
-	0xb601e033,
-	0x30f00132,
-	0xde35980c,
-	0x12a655f9,
-	0x98e51ef4,
-	0x0c98f10b,
-	0x02cbbbf2,
-	0xcf07c44b,
-	0xd0fc00bb,
-	0xb67ee0fc,
-	0x00f80002,
-/* 0x05f8: memx_info */
-	0xf401c670,
-/* 0x05fe: memx_info_data */
-	0xcc4c0c0b,
-	0x08004b03,
-/* 0x0607: memx_info_train */
-	0x4c090ef4,
-	0x004b0bcc,
-/* 0x060d: memx_info_send */
-	0x02b67e01,
-/* 0x0613: memx_recv */
-	0xb000f800,
-	0x0bf401d6,
-	0x00d6b0a3,
-	0xf8dc0bf4,
-/* 0x0621: memx_init */
-/* 0x0623: perf_recv */
+	0x002d7ee0,
+	0x0aa24700,
+	0x047e6eb2,
+	0xd8b20000,
+	0xf90587fd,
+	0xfc80f960,
+	0x7ee0fcd0,
+	0xf800002d,
+/* 0x0541: memx_func_wait_vblank */
+	0x0410b600,
+/* 0x0546: memx_func_wr32 */
+	0x169800f8,
+	0x01159800,
+	0xf90810b6,
+	0xfc50f960,
+	0x7ee0fcd0,
+	0xb600002d,
+	0x1bf40242,
+/* 0x0563: memx_func_wait */
+	0x0800f8e8,
+	0x0088cf2c,
+	0x98001e98,
+	0x1c98011d,
+	0x031b9802,
+	0x7e1010b6,
+	0xf8000074,
+/* 0x057d: memx_func_delay */
+	0x001e9800,
+	0x7e0410b6,
+	0xf8000058,
+/* 0x0589: memx_func_train */
+/* 0x058b: memx_exec */
+	0xf900f800,
+	0xb2d0f9e0,
+/* 0x0593: memx_exec_next */
+	0x98b2b2c1,
+	0x10b60013,
+	0xf034e704,
+	0xe033e701,
+	0x0132b601,
+	0x980c30f0,
+	0x55f9de35,
+	0x1ef412a6,
+	0xf10b98e5,
+	0xbbf20c98,
+	0xc44b02cb,
+	0x00bbcf07,
+	0xe0fcd0fc,
+	0x00029f7e,
+/* 0x05ca: memx_info */
+	0xc67000f8,
+	0x0c0bf401,
+/* 0x05d0: memx_info_data */
+	0x4b03cc4c,
+	0x0ef40800,
+/* 0x05d9: memx_info_train */
+	0x0bcc4c09,
+/* 0x05df: memx_info_send */
+	0x7e01004b,
+	0xf800029f,
+/* 0x05e5: memx_recv */
+	0x01d6b000,
+	0xb0a30bf4,
+	0x0bf400d6,
+/* 0x05f3: memx_init */
+	0xf800f8dc,
+/* 0x05f5: perf_recv */
+/* 0x05f7: perf_init */
 	0xf800f800,
-/* 0x0625: perf_init */
-/* 0x0627: i2c_drive_scl */
-	0xb000f800,
-	0x0bf40036,
-	0x07e0400d,
-	0xbd0001f6,
-/* 0x0637: i2c_drive_scl_lo */
-	0x4000f804,
-	0x01f607e4,
-	0xf804bd00,
-/* 0x0641: i2c_drive_sda */
+/* 0x05f9: i2c_drive_scl */
 	0x0036b000,
 	0x400d0bf4,
-	0x02f607e0,
+	0x01f607e0,
 	0xf804bd00,
-/* 0x0651: i2c_drive_sda_lo */
+/* 0x0609: i2c_drive_scl_lo */
 	0x07e44000,
+	0xbd0001f6,
+/* 0x0613: i2c_drive_sda */
+	0xb000f804,
+	0x0bf40036,
+	0x07e0400d,
 	0xbd0002f6,
-/* 0x065b: i2c_sense_scl */
-	0xf400f804,
-	0xc4430132,
-	0x0033cf07,
-	0xf40431fd,
-	0x31f4060b,
-/* 0x066d: i2c_sense_scl_done */
-/* 0x066f: i2c_sense_sda */
-	0xf400f801,
-	0xc4430132,
-	0x0033cf07,
-	0xf40432fd,
-	0x31f4060b,
-/* 0x0681: i2c_sense_sda_done */
-/* 0x0683: i2c_raise_scl */
-	0xf900f801,
-	0x08984440,
-	0x277e0103,
-/* 0x068e: i2c_raise_scl_wait */
-	0xe84e0006,
-	0x005d7e03,
-	0x065b7e00,
-	0x0901f400,
-	0xf40142b6,
-/* 0x06a2: i2c_raise_scl_done */
-	0x40fcef1b,
-/* 0x06a6: i2c_start */
-	0x5b7e00f8,
-	0x11f40006,
-	0x066f7e0d,
-	0x0611f400,
-/* 0x06b7: i2c_start_rep */
-	0x032e0ef4,
-	0x06277e00,
-	0x7e010300,
-	0xbb000641,
-	0x65b60076,
-	0x9450f904,
-	0x56bb0465,
-	0xfd50bd02,
-	0x50fc0475,
-	0x0006837e,
-	0xf40464b6,
-/* 0x06e2: i2c_start_send */
-	0x00031d11,
-	0x0006417e,
-	0x7e13884e,
-	0x0300005d,
-	0x06277e00,
+/* 0x0623: i2c_drive_sda_lo */
+	0x4000f804,
+	0x02f607e4,
+	0xf804bd00,
+/* 0x062d: i2c_sense_scl */
+	0x0132f400,
+	0xcf07c443,
+	0x31fd0033,
+	0x060bf404,
+/* 0x063f: i2c_sense_scl_done */
+	0xf80131f4,
+/* 0x0641: i2c_sense_sda */
+	0x0132f400,
+	0xcf07c443,
+	0x32fd0033,
+	0x060bf404,
+/* 0x0653: i2c_sense_sda_done */
+	0xf80131f4,
+/* 0x0655: i2c_raise_scl */
+	0x4440f900,
+	0x01030898,
+	0x0005f97e,
+/* 0x0660: i2c_raise_scl_wait */
+	0x7e03e84e,
+	0x7e000058,
+	0xf400062d,
+	0x42b60901,
+	0xef1bf401,
+/* 0x0674: i2c_raise_scl_done */
+	0x00f840fc,
+/* 0x0678: i2c_start */
+	0x00062d7e,
+	0x7e0d11f4,
+	0xf4000641,
+	0x0ef40611,
+/* 0x0689: i2c_start_rep */
+	0x7e00032e,
+	0x030005f9,
+	0x06137e01,
+	0x0076bb00,
+	0xf90465b6,
+	0x04659450,
+	0xbd0256bb,
+	0x0475fd50,
+	0x557e50fc,
+	0x64b60006,
+	0x1d11f404,
+/* 0x06b4: i2c_start_send */
+	0x137e0003,
+	0x884e0006,
+	0x00587e13,
+	0x7e000300,
+	0x4e0005f9,
+	0x587e1388,
+/* 0x06ce: i2c_start_out */
+	0x00f80000,
+/* 0x06d0: i2c_stop */
+	0xf97e0003,
+	0x00030005,
+	0x0006137e,
+	0x7e03e84e,
+	0x03000058,
+	0x05f97e01,
 	0x13884e00,
-	0x00005d7e,
-/* 0x06fc: i2c_start_out */
-/* 0x06fe: i2c_stop */
-	0x000300f8,
-	0x0006277e,
-	0x417e0003,
-	0xe84e0006,
-	0x005d7e03,
-	0x7e010300,
-	0x4e000627,
-	0x5d7e1388,
-	0x01030000,
-	0x0006417e,
-	0x7e13884e,
-	0xf800005d,
-/* 0x072d: i2c_bitw */
-	0x06417e00,
-	0x03e84e00,
-	0x00005d7e,
-	0xb60076bb,
-	0x50f90465,
-	0xbb046594,
-	0x50bd0256,
-	0xfc0475fd,
-	0x06837e50,
-	0x0464b600,
-	0x4e1711f4,
-	0x5d7e1388,
-	0x00030000,
-	0x0006277e,
-	0x7e13884e,
-/* 0x076b: i2c_bitw_out */
-	0xf800005d,
-/* 0x076d: i2c_bitr */
-	0x7e010300,
-	0x4e000641,
-	0x5d7e03e8,
+	0x0000587e,
+	0x137e0103,
+	0x884e0006,
+	0x00587e13,
+/* 0x06ff: i2c_bitw */
+	0x7e00f800,
+	0x4e000613,
+	0x587e03e8,
 	0x76bb0000,
 	0x0465b600,
 	0x659450f9,
 	0x0256bb04,
 	0x75fd50bd,
 	0x7e50fc04,
-	0xb6000683,
+	0xb6000655,
 	0x11f40464,
-	0x066f7e1a,
-	0x7e000300,
-	0x4e000627,
-	0x5d7e1388,
-	0x3cf00000,
-	0x0131f401,
-/* 0x07b0: i2c_bitr_done */
-/* 0x07b2: i2c_get_byte */
-	0x000500f8,
-/* 0x07b6: i2c_get_byte_next */
-	0x54b60804,
-	0x0076bb01,
-	0xf90465b6,
-	0x04659450,
-	0xbd0256bb,
-	0x0475fd50,
-	0x6d7e50fc,
-	0x64b60007,
-	0x2a11f404,
-	0xb60553fd,
-	0x1bf40142,
-	0xbb0103d8,
+	0x13884e17,
+	0x0000587e,
+	0xf97e0003,
+	0x884e0005,
+	0x00587e13,
+/* 0x073d: i2c_bitw_out */
+/* 0x073f: i2c_bitr */
+	0x0300f800,
+	0x06137e01,
+	0x03e84e00,
+	0x0000587e,
+	0xb60076bb,
+	0x50f90465,
+	0xbb046594,
+	0x50bd0256,
+	0xfc0475fd,
+	0x06557e50,
+	0x0464b600,
+	0x7e1a11f4,
+	0x03000641,
+	0x05f97e00,
+	0x13884e00,
+	0x0000587e,
+	0xf4013cf0,
+/* 0x0782: i2c_bitr_done */
+	0x00f80131,
+/* 0x0784: i2c_get_byte */
+	0x08040005,
+/* 0x0788: i2c_get_byte_next */
+	0xbb0154b6,
 	0x65b60076,
 	0x9450f904,
 	0x56bb0465,
 	0xfd50bd02,
 	0x50fc0475,
-	0x00072d7e,
-/* 0x07ff: i2c_get_byte_done */
-	0xf80464b6,
-/* 0x0801: i2c_put_byte */
-/* 0x0803: i2c_put_byte_next */
-	0xb6080400,
-	0x54ff0142,
-	0x0076bb38,
+	0x00073f7e,
+	0xf40464b6,
+	0x53fd2a11,
+	0x0142b605,
+	0x03d81bf4,
+	0x0076bb01,
 	0xf90465b6,
 	0x04659450,
 	0xbd0256bb,
 	0x0475fd50,
-	0x2d7e50fc,
-	0x64b60007,
-	0x3411f404,
-	0xf40046b0,
-	0x76bbd81b,
-	0x0465b600,
-	0x659450f9,
-	0x0256bb04,
-	0x75fd50bd,
-	0x7e50fc04,
-	0xb600076d,
-	0x11f40464,
-	0x0076bb0f,
-	0xf40136b0,
-	0x32f4061b,
-/* 0x0859: i2c_put_byte_done */
-/* 0x085b: i2c_addr */
-	0xbb00f801,
+	0xff7e50fc,
+	0x64b60006,
+/* 0x07d1: i2c_get_byte_done */
+/* 0x07d3: i2c_put_byte */
+	0x0400f804,
+/* 0x07d5: i2c_put_byte_next */
+	0x0142b608,
+	0xbb3854ff,
 	0x65b60076,
 	0x9450f904,
 	0x56bb0465,
 	0xfd50bd02,
 	0x50fc0475,
-	0x0006a67e,
+	0x0006ff7e,
 	0xf40464b6,
-	0xc3e72911,
-	0x34b6012e,
-	0x0553fd01,
-	0xb60076bb,
-	0x50f90465,
-	0xbb046594,
-	0x50bd0256,
-	0xfc0475fd,
-	0x08017e50,
-	0x0464b600,
-/* 0x08a0: i2c_addr_done */
-/* 0x08a2: i2c_acquire_addr */
-	0xcec700f8,
-	0x05e4b6f8,
-	0xd014e0b7,
-/* 0x08ae: i2c_acquire */
-	0xa27e00f8,
-	0x047e0008,
-	0xd9f00000,
-	0x002e7e03,
-/* 0x08bf: i2c_release */
-	0x7e00f800,
-	0x7e0008a2,
-	0xf0000004,
-	0x2e7e03da,
-	0x00f80000,
-/* 0x08d0: i2c_recv */
-	0xc70132f4,
-	0x14b6f8c1,
-	0x2816b002,
-	0x01371ff5,
-	0x0cf413b8,
-	0x00329800,
-	0x0ccc13b8,
-	0x00319800,
-	0xf90231f4,
-	0xf9e0f9d0,
-	0x0067f1d0,
-	0x0063f100,
-	0x01679210,
+	0x46b03411,
+	0xd81bf400,
 	0xb60076bb,
 	0x50f90465,
 	0xbb046594,
 	0x50bd0256,
 	0xfc0475fd,
-	0x08ae7e50,
+	0x073f7e50,
 	0x0464b600,
-	0xd6b0d0fc,
-	0xb01bf500,
-	0xbb000500,
+	0xbb0f11f4,
+	0x36b00076,
+	0x061bf401,
+/* 0x082b: i2c_put_byte_done */
+	0xf80132f4,
+/* 0x082d: i2c_addr */
+	0x0076bb00,
+	0xf90465b6,
+	0x04659450,
+	0xbd0256bb,
+	0x0475fd50,
+	0x787e50fc,
+	0x64b60006,
+	0x2911f404,
+	0x012ec3e7,
+	0xfd0134b6,
+	0x76bb0553,
+	0x0465b600,
+	0x659450f9,
+	0x0256bb04,
+	0x75fd50bd,
+	0x7e50fc04,
+	0xb60007d3,
+/* 0x0872: i2c_addr_done */
+	0x00f80464,
+/* 0x0874: i2c_acquire_addr */
+	0xb6f8cec7,
+	0xe0b705e4,
+	0x00f8d014,
+/* 0x0880: i2c_acquire */
+	0x0008747e,
+	0x0000047e,
+	0x7e03d9f0,
+	0xf800002d,
+/* 0x0891: i2c_release */
+	0x08747e00,
+	0x00047e00,
+	0x03daf000,
+	0x00002d7e,
+/* 0x08a2: i2c_recv */
+	0x32f400f8,
+	0xf8c1c701,
+	0xb00214b6,
+	0x1ff52816,
+	0x13b80134,
+	0x98000cf4,
+	0x13b80032,
+	0x98000ccc,
+	0x31f40031,
+	0xf9d0f902,
+	0xd6d0f9e0,
+	0x10000000,
+	0xbb016792,
 	0x65b60076,
 	0x9450f904,
 	0x56bb0465,
 	0xfd50bd02,
 	0x50fc0475,
-	0x00085b7e,
-	0xf50464b6,
-	0xc700cc11,
-	0x76bbe0c5,
+	0x0008807e,
+	0xfc0464b6,
+	0x00d6b0d0,
+	0x00b01bf5,
+	0x76bb0005,
 	0x0465b600,
 	0x659450f9,
 	0x0256bb04,
 	0x75fd50bd,
 	0x7e50fc04,
-	0xb6000801,
+	0xb600082d,
 	0x11f50464,
-	0x010500a9,
-	0xb60076bb,
-	0x50f90465,
-	0xbb046594,
-	0x50bd0256,
-	0xfc0475fd,
-	0x085b7e50,
-	0x0464b600,
-	0x008711f5,
-	0xb60076bb,
-	0x50f90465,
-	0xbb046594,
-	0x50bd0256,
-	0xfc0475fd,
-	0x07b27e50,
-	0x0464b600,
-	0xcb6711f4,
-	0x76bbe05b,
-	0x0465b600,
-	0x659450f9,
-	0x0256bb04,
-	0x75fd50bd,
-	0x7e50fc04,
-	0xb60006fe,
-	0x5bb20464,
-	0x0ef474bd,
-/* 0x09d5: i2c_recv_not_rd08 */
-	0x01d6b041,
-	0x053b1bf4,
-	0x085b7e00,
-	0x3211f400,
-	0x7ee0c5c7,
-	0xf4000801,
-	0x00052811,
-	0x00085b7e,
-	0xc71f11f4,
-	0x017ee0b5,
-	0x11f40008,
-	0x06fe7e15,
-	0xc774bd00,
-	0x1bf408c5,
-	0x0232f409,
-/* 0x0a13: i2c_recv_not_wr08 */
-/* 0x0a13: i2c_recv_done */
-	0xc7030ef4,
-	0xbf7ef8ce,
-	0xe0fc0008,
-	0x12f4d0fc,
-	0x7e7cb209,
-/* 0x0a27: i2c_recv_exit */
-	0xf80002b6,
-/* 0x0a29: i2c_init */
-/* 0x0a2b: test_recv */
-	0x4100f800,
-	0x11cf0458,
-	0x0110b600,
-	0xf6045840,
-	0x04bd0001,
-	0xd900e7f1,
-	0x134fe3f1,
-	0x0001f57e,
-/* 0x0a4a: test_init */
+	0xc5c700cc,
+	0x0076bbe0,
+	0xf90465b6,
+	0x04659450,
+	0xbd0256bb,
+	0x0475fd50,
+	0xd37e50fc,
+	0x64b60007,
+	0xa911f504,
+	0xbb010500,
+	0x65b60076,
+	0x9450f904,
+	0x56bb0465,
+	0xfd50bd02,
+	0x50fc0475,
+	0x00082d7e,
+	0xf50464b6,
+	0xbb008711,
+	0x65b60076,
+	0x9450f904,
+	0x56bb0465,
+	0xfd50bd02,
+	0x50fc0475,
+	0x0007847e,
+	0xf40464b6,
+	0x5bcb6711,
+	0x0076bbe0,
+	0xf90465b6,
+	0x04659450,
+	0xbd0256bb,
+	0x0475fd50,
+	0xd07e50fc,
+	0x64b60006,
+	0xbd5bb204,
+	0x410ef474,
+/* 0x09a4: i2c_recv_not_rd08 */
+	0xf401d6b0,
+	0x00053b1b,
+	0x00082d7e,
+	0xc73211f4,
+	0xd37ee0c5,
+	0x11f40007,
+	0x7e000528,
+	0xf400082d,
+	0xb5c71f11,
+	0x07d37ee0,
+	0x1511f400,
+	0x0006d07e,
+	0xc5c774bd,
+	0x091bf408,
+	0xf40232f4,
+/* 0x09e2: i2c_recv_not_wr08 */
+/* 0x09e2: i2c_recv_done */
+	0xcec7030e,
+	0x08917ef8,
+	0xfce0fc00,
+	0x0912f4d0,
+	0x9f7e7cb2,
+/* 0x09f6: i2c_recv_exit */
+	0x00f80002,
+/* 0x09f8: i2c_init */
+/* 0x09fa: test_recv */
+	0x584100f8,
+	0x0011cf04,
+	0x400110b6,
+	0x01f60458,
+	0xde04bd00,
+	0x134fd900,
+	0x0001de7e,
+/* 0x0a16: test_init */
 	0x004e00f8,
-	0x01f57e08,
-/* 0x0a53: idle_recv */
+	0x01de7e08,
+/* 0x0a1f: idle_recv */
 	0xf800f800,
-/* 0x0a55: idle */
+/* 0x0a21: idle */
 	0x0031f400,
 	0xcf045441,
 	0x10b60011,
 	0x04544001,
 	0xbd0001f6,
-/* 0x0a69: idle_loop */
+/* 0x0a35: idle_loop */
 	0xf4580104,
-/* 0x0a6e: idle_proc */
-/* 0x0a6e: idle_proc_exec */
+/* 0x0a3a: idle_proc */
+/* 0x0a3a: idle_proc_exec */
 	0x10f90232,
-	0xbf7e1eb2,
+	0xa87e1eb2,
 	0x10fc0002,
 	0xf40911f4,
 	0x0ef40231,
-/* 0x0a81: idle_proc_next */
+/* 0x0a4d: idle_proc_next */
 	0x5810b6f0,
 	0x1bf41fa6,
 	0xe002f4e8,
@@ -1726,4 +1713,17 @@ uint32_t gk208_pmu_code[] = {
 	0x00000000,
 	0x00000000,
 	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
+	0x00000000,
 };
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/pmu/fuc/gt215.fuc3.h b/drivers/gpu/drm/nouveau/nvkm/subdev/pmu/fuc/gt215.fuc3.h
index 516569270bac..e83341815ec6 100644
--- a/drivers/gpu/drm/nouveau/nvkm/subdev/pmu/fuc/gt215.fuc3.h
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/pmu/fuc/gt215.fuc3.h
@@ -24,8 +24,8 @@ uint32_t gt215_pmu_data[] = {
 	0x00000000,
 /* 0x0058: proc_list_head */
 	0x54534f48,
-	0x00000507,
-	0x000004a4,
+	0x0000050a,
+	0x000004a7,
 	0x00000000,
 	0x00000000,
 	0x00000000,
@@ -46,8 +46,8 @@ uint32_t gt215_pmu_data[] = {
 	0x00000000,
 	0x00000000,
 	0x584d454d,
-	0x00000837,
-	0x00000829,
+	0x0000083a,
+	0x0000082c,
 	0x00000000,
 	0x00000000,
 	0x00000000,
@@ -68,8 +68,8 @@ uint32_t gt215_pmu_data[] = {
 	0x00000000,
 	0x00000000,
 	0x46524550,
-	0x0000083b,
-	0x00000839,
+	0x0000083e,
+	0x0000083c,
 	0x00000000,
 	0x00000000,
 	0x00000000,
@@ -90,8 +90,8 @@ uint32_t gt215_pmu_data[] = {
 	0x00000000,
 	0x00000000,
 	0x5f433249,
-	0x00000c6b,
-	0x00000b0e,
+	0x00000c6e,
+	0x00000b11,
 	0x00000000,
 	0x00000000,
 	0x00000000,
@@ -112,8 +112,8 @@ uint32_t gt215_pmu_data[] = {
 	0x00000000,
 	0x00000000,
 	0x54534554,
-	0x00000c94,
-	0x00000c6d,
+	0x00000c97,
+	0x00000c70,
 	0x00000000,
 	0x00000000,
 	0x00000000,
@@ -134,8 +134,8 @@ uint32_t gt215_pmu_data[] = {
 	0x00000000,
 	0x00000000,
 	0x454c4449,
-	0x00000ca0,
-	0x00000c9e,
+	0x00000ca3,
+	0x00000ca1,
 	0x00000000,
 	0x00000000,
 	0x00000000,
@@ -229,26 +229,26 @@ uint32_t gt215_pmu_data[] = {
 /* 0x0370: memx_func_head */
 	0x00000001,
 	0x00000000,
-	0x00000546,
+	0x00000549,
 /* 0x037c: memx_func_next */
 	0x00000002,
 	0x00000000,
-	0x0000059d,
+	0x000005a0,
 	0x00000003,
 	0x00000002,
-	0x0000062f,
+	0x00000632,
 	0x00040004,
 	0x00000000,
-	0x0000064b,
+	0x0000064e,
 	0x00010005,
 	0x00000000,
-	0x00000668,
+	0x0000066b,
 	0x00010006,
 	0x00000000,
-	0x000005ed,
+	0x000005f0,
 	0x00000007,
 	0x00000000,
-	0x00000673,
+	0x00000676,
 /* 0x03c4: memx_func_tail */
 /* 0x03c4: memx_ts_start */
 	0x00000000,
@@ -917,947 +917,947 @@ uint32_t gt215_pmu_data[] = {
 };
 
 uint32_t gt215_pmu_code[] = {
-	0x03930ef5,
+	0x03920ef5,
 /* 0x0004: rd32 */
 	0x07a007f1,
 	0xd00604b6,
 	0x04bd000e,
-	0xf001d7f0,
-	0x07f101d3,
-	0x04b607ac,
-	0x000dd006,
-/* 0x0022: rd32_wait */
-	0xd7f104bd,
-	0xd4b607ac,
-	0x00ddcf06,
-	0x7000d4f1,
-	0xf1f21bf4,
-	0xb607a4d7,
-	0xddcf06d4,
-/* 0x003f: wr32 */
-	0xf100f800,
-	0xb607a007,
-	0x0ed00604,
-	0xf104bd00,
-	0xb607a407,
+	0x0001d7f1,
+	0xf101d3f0,
+	0xb607ac07,
 	0x0dd00604,
-	0xf004bd00,
-	0xd5f002d7,
-	0x01d3f0f0,
-	0x07ac07f1,
+/* 0x0023: rd32_wait */
+	0xf104bd00,
+	0xb607acd7,
+	0xddcf06d4,
+	0x00d4f100,
+	0xf21bf470,
+	0x07a4d7f1,
+	0xcf06d4b6,
+	0x00f800dd,
+/* 0x0040: wr32 */
+	0x07a007f1,
+	0xd00604b6,
+	0x04bd000e,
+	0x07a407f1,
 	0xd00604b6,
 	0x04bd000d,
-/* 0x006c: wr32_wait */
-	0x07acd7f1,
-	0xcf06d4b6,
-	0xd4f100dd,
-	0x1bf47000,
-/* 0x007f: nsec */
-	0xf900f8f2,
-	0xf080f990,
-	0x84b62c87,
-	0x0088cf06,
-/* 0x008c: nsec_loop */
-	0xb62c97f0,
-	0x99cf0694,
-	0x0298bb00,
-	0xf4069eb8,
-	0x80fcf11e,
-	0x00f890fc,
-/* 0x00a4: wait */
-	0x80f990f9,
-	0xb62c87f0,
-	0x88cf0684,
-/* 0x00b1: wait_loop */
-	0x02eeb900,
-	0xb90421f4,
-	0xadfd02da,
-	0x06acb804,
-	0xf0150bf4,
+	0x00f2d7f1,
+	0xf101d3f0,
+	0xb607ac07,
+	0x0dd00604,
+/* 0x006b: wr32_wait */
+	0xf104bd00,
+	0xb607acd7,
+	0xddcf06d4,
+	0x00d4f100,
+	0xf21bf470,
+/* 0x007e: nsec */
+	0x90f900f8,
+	0x87f080f9,
+	0x0684b62c,
+/* 0x008b: nsec_loop */
+	0xf00088cf,
 	0x94b62c97,
 	0x0099cf06,
 	0xb80298bb,
-	0x1ef4069b,
-/* 0x00d5: wait_done */
-	0xfc80fcdf,
-/* 0x00db: intr_watchdog */
-	0x9800f890,
-	0x96b003e9,
-	0x2a0bf400,
-	0xbb9a0a98,
-	0x1cf4029a,
-	0x01d7f00f,
-	0x02d221f5,
-	0x0ef494bd,
-/* 0x00f9: intr_watchdog_next_time */
-	0x9b0a9815,
-	0xf400a6b0,
-	0x9ab8090b,
-	0x061cf406,
-/* 0x0108: intr_watchdog_next_time_set */
-/* 0x010b: intr_watchdog_next_proc */
-	0x809b0980,
-	0xe0b603e9,
-	0x68e6b158,
-	0xc61bf402,
-/* 0x011a: intr */
-	0x00f900f8,
-	0x80f904bd,
-	0xa0f990f9,
-	0xc0f9b0f9,
-	0xe0f9d0f9,
-	0xf7f0f0f9,
-	0x0188fe00,
-	0x87f180f9,
-	0x84b605d0,
+	0x1ef4069e,
+	0xfc80fcf1,
+/* 0x00a3: wait */
+	0xf900f890,
+	0xf080f990,
+	0x84b62c87,
 	0x0088cf06,
-	0xf10180b6,
-	0xb605d007,
+/* 0x00b0: wait_loop */
+	0xf402eeb9,
+	0xdab90421,
+	0x04adfd02,
+	0xf406acb8,
+	0x97f0150b,
+	0x0694b62c,
+	0xbb0099cf,
+	0x9bb80298,
+	0xdf1ef406,
+/* 0x00d4: wait_done */
+	0x90fc80fc,
+/* 0x00da: intr_watchdog */
+	0xe99800f8,
+	0x0096b003,
+	0x982a0bf4,
+	0x9abb9a0a,
+	0x0f1cf402,
+	0xf501d7f0,
+	0xbd02d121,
+	0x150ef494,
+/* 0x00f8: intr_watchdog_next_time */
+	0xb09b0a98,
+	0x0bf400a6,
+	0x069ab809,
+/* 0x0107: intr_watchdog_next_time_set */
+	0x80061cf4,
+/* 0x010a: intr_watchdog_next_proc */
+	0xe9809b09,
+	0x58e0b603,
+	0x0268e6b1,
+	0xf8c61bf4,
+/* 0x0119: intr */
+	0xbd00f900,
+	0xf980f904,
+	0xf9a0f990,
+	0xf9c0f9b0,
+	0xf9e0f9d0,
+	0x00f7f0f0,
+	0xf90188fe,
+	0xd087f180,
+	0x0684b605,
+	0xb60088cf,
+	0x07f10180,
+	0x04b605d0,
+	0x0008d006,
+	0x87f004bd,
+	0x0684b608,
+	0xc40088cf,
+	0x0bf40289,
+	0x9b008023,
+	0xf458e7f0,
+	0x0998da21,
+	0x0096b09b,
+	0xf0110bf4,
+	0x04b63407,
+	0x0009d006,
+	0x098004bd,
+/* 0x017d: intr_skip_watchdog */
+	0x0089e49a,
+	0x480bf408,
+	0x068897f1,
+	0xcf0694b6,
+	0x9ac40099,
+	0x2c0bf402,
+	0x04c0c7f1,
+	0xcf06c4b6,
+	0xc0f900cc,
+	0x4f48e7f1,
+	0x5453e3f1,
+	0xf500d7f0,
+	0xfc033621,
+	0xc007f1c0,
+	0x0604b604,
+	0xbd000cd0,
+/* 0x01bd: intr_subintr_skip_fifo */
+	0x8807f104,
+	0x0604b606,
+	0xbd0009d0,
+/* 0x01c9: intr_skip_subintr */
+	0xe097f104,
+	0xfd90bd00,
+	0x07f00489,
+	0x0604b604,
+	0xbd0008d0,
+	0xfe80fc04,
+	0xf0fc0088,
+	0xd0fce0fc,
+	0xb0fcc0fc,
+	0x90fca0fc,
+	0x00fc80fc,
+	0xf80032f4,
+/* 0x01f9: ticks_from_ns */
+	0xf9c0f901,
+	0xcbd7f1b0,
+	0x00d3f000,
+	0x040b21f5,
+	0x03e8ccec,
+	0xf400b4b0,
+	0xeeec120b,
+	0xd7f103e8,
+	0xd3f000cb,
+	0x0b21f500,
+/* 0x0221: ticks_from_ns_quit */
+	0x02ceb904,
+	0xc0fcb0fc,
+/* 0x022a: ticks_from_us */
+	0xc0f900f8,
+	0xd7f1b0f9,
+	0xd3f000cb,
+	0x0b21f500,
+	0x02ceb904,
+	0xf400b4b0,
+	0xe4bd050b,
+/* 0x0244: ticks_from_us_quit */
+	0xc0fcb0fc,
+/* 0x024a: ticks_to_us */
+	0xd7f100f8,
+	0xd3f000cb,
+	0xecedff00,
+/* 0x0256: timer */
+	0x90f900f8,
+	0x32f480f9,
+	0x03f89810,
+	0xf40086b0,
+	0x84bd651c,
+	0xb63807f0,
 	0x08d00604,
 	0xf004bd00,
-	0x84b60887,
+	0x84b63487,
 	0x0088cf06,
-	0xf40289c4,
-	0x0080230b,
-	0x58e7f09b,
-	0x98db21f4,
-	0x96b09b09,
-	0x110bf400,
+	0xbb9a0998,
+	0xe9bb0298,
+	0x03fe8000,
+	0xb60887f0,
+	0x88cf0684,
+	0x0284f000,
+	0xf0261bf4,
+	0x84b63487,
+	0x0088cf06,
+	0xf406e0b8,
+	0xe8b8090b,
+	0x111cf406,
+/* 0x02ac: timer_reset */
 	0xb63407f0,
-	0x09d00604,
+	0x0ed00604,
 	0x8004bd00,
-/* 0x017e: intr_skip_watchdog */
-	0x89e49a09,
-	0x0bf40800,
-	0x8897f148,
-	0x0694b606,
-	0xc40099cf,
-	0x0bf4029a,
-	0xc0c7f12c,
-	0x06c4b604,
-	0xf900cccf,
-	0x48e7f1c0,
-	0x53e3f14f,
-	0x00d7f054,
-	0x033721f5,
-	0x07f1c0fc,
-	0x04b604c0,
-	0x000cd006,
-/* 0x01be: intr_subintr_skip_fifo */
-	0x07f104bd,
-	0x04b60688,
-	0x0009d006,
-/* 0x01ca: intr_skip_subintr */
-	0x97f104bd,
-	0x90bd00e0,
-	0xf00489fd,
-	0x04b60407,
-	0x0008d006,
-	0x80fc04bd,
-	0xfc0088fe,
-	0xfce0fcf0,
-	0xfcc0fcd0,
-	0xfca0fcb0,
-	0xfc80fc90,
-	0x0032f400,
-/* 0x01fa: ticks_from_ns */
-	0xc0f901f8,
-	0xd7f1b0f9,
-	0xd3f000cb,
-	0x0821f500,
-	0xe8ccec04,
-	0x00b4b003,
-	0xec120bf4,
-	0xf103e8ee,
-	0xf000cbd7,
-	0x21f500d3,
-/* 0x0222: ticks_from_ns_quit */
-	0xceb90408,
-	0xfcb0fc02,
-/* 0x022b: ticks_from_us */
-	0xf900f8c0,
-	0xf1b0f9c0,
-	0xf000cbd7,
-	0x21f500d3,
-	0xceb90408,
-	0x00b4b002,
-	0xbd050bf4,
-/* 0x0245: ticks_from_us_quit */
-	0xfcb0fce4,
-/* 0x024b: ticks_to_us */
-	0xf100f8c0,
-	0xf000cbd7,
-	0xedff00d3,
-/* 0x0257: timer */
-	0xf900f8ec,
-	0xf480f990,
-	0xf8981032,
-	0x0086b003,
-	0xbd651cf4,
-	0x3807f084,
+/* 0x02ba: timer_enable */
+	0x87f09a0e,
+	0x3807f001,
 	0xd00604b6,
 	0x04bd0008,
-	0xb63487f0,
-	0x88cf0684,
-	0x9a099800,
-	0xbb0298bb,
-	0xfe8000e9,
-	0x0887f003,
-	0xcf0684b6,
-	0x84f00088,
-	0x261bf402,
-	0xb63487f0,
-	0x88cf0684,
-	0x06e0b800,
-	0xb8090bf4,
-	0x1cf406e8,
-/* 0x02ad: timer_reset */
-	0x3407f011,
-	0xd00604b6,
-	0x04bd000e,
-/* 0x02bb: timer_enable */
-	0xf09a0e80,
-	0x07f00187,
-	0x0604b638,
-	0xbd0008d0,
-/* 0x02c9: timer_done */
-	0x1031f404,
+/* 0x02c8: timer_done */
+	0xfc1031f4,
+	0xf890fc80,
+/* 0x02d1: send_proc */
+	0xf980f900,
+	0x05e89890,
+	0xf004e998,
+	0x89b80486,
+	0x2a0bf406,
+	0x940398c4,
+	0x80b60488,
+	0x008ebb18,
+	0x8000fa98,
+	0x8d80008a,
+	0x028c8001,
+	0xb6038b80,
+	0x94f00190,
+	0x04e98007,
+/* 0x030b: send_done */
+	0xfc0231f4,
+	0xf880fc90,
+/* 0x0311: find */
+	0xf080f900,
+	0x31f45887,
+/* 0x0319: find_loop */
+	0x008a9801,
+	0xf406aeb8,
+	0x80b6100b,
+	0x6886b158,
+	0xf01bf402,
+/* 0x032f: find_done */
+	0xb90132f4,
+	0x80fc028e,
+/* 0x0336: send */
+	0x21f500f8,
+	0x01f40311,
+/* 0x033f: recv */
+	0xf900f897,
+	0x9880f990,
+	0xe99805e8,
+	0x0132f404,
+	0xf40689b8,
+	0x89c43d0b,
+	0x0180b603,
+	0x800784f0,
+	0xea9805e8,
+	0xfef0f902,
+	0xf0f9018f,
+	0x9402efb9,
+	0xe9bb0499,
+	0x18e0b600,
+	0x9803eb98,
+	0xed9802ec,
+	0x00ee9801,
+	0xf0fca5f9,
+	0xf400f8fe,
+	0xf0fc0131,
+/* 0x038c: recv_done */
 	0x90fc80fc,
-/* 0x02d2: send_proc */
-	0x80f900f8,
-	0xe89890f9,
-	0x04e99805,
-	0xb80486f0,
-	0x0bf40689,
-	0x0398c42a,
-	0xb6048894,
-	0x8ebb1880,
-	0x00fa9800,
-	0x80008a80,
-	0x8c80018d,
-	0x038b8002,
-	0xf00190b6,
-	0xe9800794,
-	0x0231f404,
-/* 0x030c: send_done */
-	0x80fc90fc,
-/* 0x0312: find */
-	0x80f900f8,
-	0xf45887f0,
-/* 0x031a: find_loop */
-	0x8a980131,
-	0x06aeb800,
-	0xb6100bf4,
-	0x86b15880,
-	0x1bf40268,
-	0x0132f4f0,
-/* 0x0330: find_done */
-	0xfc028eb9,
-/* 0x0337: send */
-	0xf500f880,
-	0xf4031221,
-	0x00f89701,
-/* 0x0340: recv */
-	0x80f990f9,
-	0x9805e898,
-	0x32f404e9,
-	0x0689b801,
-	0xc43d0bf4,
-	0x80b60389,
-	0x0784f001,
-	0x9805e880,
-	0xf0f902ea,
-	0xf9018ffe,
-	0x02efb9f0,
-	0xbb049994,
-	0xe0b600e9,
-	0x03eb9818,
-	0x9802ec98,
-	0xee9801ed,
-	0xfca5f900,
-	0x00f8fef0,
-	0xfc0131f4,
-/* 0x038d: recv_done */
-	0xfc80fcf0,
-/* 0x0393: init */
-	0xf100f890,
-	0xb6010817,
-	0x11cf0614,
-	0x0911e700,
-	0x0814b601,
-	0xf10014fe,
-	0xf000e017,
-	0x07f00013,
-	0x0604b61c,
-	0xbd0001d0,
-	0xff17f004,
-	0xb61407f0,
-	0x01d00604,
-	0xf004bd00,
-	0x15f10217,
-	0x07f00800,
-	0x0604b610,
-	0xbd0001d0,
-	0x1a17f104,
-	0x0013f001,
-	0xf40010fe,
-	0x17f01031,
-	0x3807f001,
+/* 0x0392: init */
+	0x17f100f8,
+	0x14b60108,
+	0x0011cf06,
+	0x010911e7,
+	0xfe0814b6,
+	0x17f10014,
+	0x13f000e0,
+	0x1c07f000,
+	0xd00604b6,
+	0x04bd0001,
+	0xf0ff17f0,
+	0x04b61407,
+	0x0001d006,
+	0x17f004bd,
+	0x0015f102,
+	0x1007f008,
 	0xd00604b6,
 	0x04bd0001,
-/* 0x03f7: init_proc */
-	0x9858f7f0,
-	0x16b001f1,
-	0xfa0bf400,
-	0xf0b615f9,
-	0xf20ef458,
-/* 0x0408: mulu32_32_64 */
-	0x20f910f9,
-	0x40f930f9,
-	0x9510e195,
-	0xc4bd10d2,
-	0xedffb4bd,
-	0x301dffc0,
-	0xf10234b9,
-	0xb6ffff34,
-	0x45b61034,
-	0x00c3bb10,
-	0xff01b4bb,
-	0x34b930e2,
-	0xff34f102,
-	0x1034b6ff,
-	0xbb1045b6,
-	0xb4bb00c3,
-	0x3012ff01,
-	0xfc00b3bb,
-	0xfc30fc40,
-	0xf810fc20,
-/* 0x0459: host_send */
-	0xb017f100,
+	0x011917f1,
+	0xf10013f0,
+	0xfeffff14,
+	0x31f40010,
+	0x0117f010,
+	0xb63807f0,
+	0x01d00604,
+	0xf004bd00,
+/* 0x03fa: init_proc */
+	0xf19858f7,
+	0x0016b001,
+	0xf9fa0bf4,
+	0x58f0b615,
+/* 0x040b: mulu32_32_64 */
+	0xf9f20ef4,
+	0xf920f910,
+	0x9540f930,
+	0xd29510e1,
+	0xbdc4bd10,
+	0xc0edffb4,
+	0xb9301dff,
+	0x34f10234,
+	0x34b6ffff,
+	0x1045b610,
+	0xbb00c3bb,
+	0xe2ff01b4,
+	0x0234b930,
+	0xffff34f1,
+	0xb61034b6,
+	0xc3bb1045,
+	0x01b4bb00,
+	0xbb3012ff,
+	0x40fc00b3,
+	0x20fc30fc,
+	0x00f810fc,
+/* 0x045c: host_send */
+	0x04b017f1,
+	0xcf0614b6,
+	0x27f10011,
+	0x24b604a0,
+	0x0022cf06,
+	0xf40612b8,
+	0x1ec4320b,
+	0x04ee9407,
+	0x0270e0b7,
+	0x9803eb98,
+	0xed9802ec,
+	0x00ee9801,
+	0x033621f5,
+	0xc40110b6,
+	0x07f10f1e,
+	0x04b604b0,
+	0x000ed006,
+	0x0ef404bd,
+/* 0x04a5: host_send_done */
+/* 0x04a7: host_recv */
+	0xf100f8ba,
+	0xf14e4917,
+	0xb8525413,
+	0x0bf406e1,
+/* 0x04b5: host_recv_wait */
+	0xcc17f1aa,
 	0x0614b604,
 	0xf10011cf,
-	0xb604a027,
+	0xb604c827,
 	0x22cf0624,
-	0x0612b800,
-	0xc4320bf4,
-	0xee94071e,
-	0x70e0b704,
-	0x03eb9802,
-	0x9802ec98,
-	0xee9801ed,
-	0x3721f500,
-	0x0110b603,
-	0xf10f1ec4,
-	0xb604b007,
-	0x0ed00604,
-	0xf404bd00,
-/* 0x04a2: host_send_done */
-	0x00f8ba0e,
-/* 0x04a4: host_recv */
-	0x4e4917f1,
-	0x525413f1,
-	0xf406e1b8,
-/* 0x04b2: host_recv_wait */
-	0x17f1aa0b,
-	0x14b604cc,
-	0x0011cf06,
-	0x04c827f1,
-	0xcf0624b6,
-	0x16f00022,
-	0x0612b808,
-	0xc4e60bf4,
-	0x34b60723,
-	0xf030b704,
-	0x033b8002,
-	0x80023c80,
-	0x3e80013d,
-	0x0120b600,
-	0xf10f24f0,
-	0xb604c807,
-	0x02d00604,
-	0xf004bd00,
-	0x07f04027,
-	0x0604b600,
-	0xbd0002d0,
-/* 0x0507: host_init */
-	0xf100f804,
-	0xb6008017,
-	0x15f11014,
-	0x07f10270,
-	0x04b604d0,
-	0x0001d006,
-	0x17f104bd,
+	0x0816f000,
+	0xf40612b8,
+	0x23c4e60b,
+	0x0434b607,
+	0x02f030b7,
+	0x80033b80,
+	0x3d80023c,
+	0x003e8001,
+	0xf00120b6,
+	0x07f10f24,
+	0x04b604c8,
+	0x0002d006,
+	0x27f004bd,
+	0x0007f040,
+	0xd00604b6,
+	0x04bd0002,
+/* 0x050a: host_init */
+	0x17f100f8,
 	0x14b60080,
-	0xf015f110,
-	0xdc07f102,
+	0x7015f110,
+	0xd007f102,
 	0x0604b604,
 	0xbd0001d0,
-	0x0117f004,
-	0x04c407f1,
+	0x8017f104,
+	0x1014b600,
+	0x02f015f1,
+	0x04dc07f1,
 	0xd00604b6,
 	0x04bd0001,
-/* 0x0546: memx_func_enter */
-	0x87f100f8,
-	0x8eb91610,
-	0x0421f402,
-	0xf102d7b9,
-	0xf1fffc67,
-	0xfdffff63,
-	0x67f10476,
-	0x76fd0002,
-	0xf980f905,
-	0xfcd0fc70,
-	0x3f21f4e0,
+	0xf10117f0,
+	0xb604c407,
+	0x01d00604,
+	0xf804bd00,
+/* 0x0549: memx_func_enter */
+	0x1087f100,
+	0x028eb916,
+	0xb90421f4,
+	0x67f102d7,
+	0x63f1fffc,
+	0x76fdffff,
+	0x0267f104,
+	0x0576fd00,
+	0x70f980f9,
+	0xe0fcd0fc,
+	0xf04021f4,
+	0x07f10467,
+	0x04b607e0,
+	0x0006d006,
+/* 0x0582: memx_func_enter_wait */
+	0x67f104bd,
+	0x64b607c0,
+	0x0066cf06,
+	0xf40464f0,
+	0x67f0f30b,
+	0x0664b62c,
+	0x800066cf,
+	0x00f8f106,
+/* 0x05a0: memx_func_leave */
+	0xb62c67f0,
+	0x66cf0664,
+	0xf2068000,
 	0xf10467f0,
-	0xb607e007,
+	0xb607e407,
 	0x06d00604,
-/* 0x057f: memx_func_enter_wait */
+/* 0x05bb: memx_func_leave_wait */
 	0xf104bd00,
 	0xb607c067,
 	0x66cf0664,
 	0x0464f000,
-	0xf0f30bf4,
-	0x64b62c67,
-	0x0066cf06,
-	0xf8f10680,
-/* 0x059d: memx_func_leave */
-	0x2c67f000,
-	0xcf0664b6,
-	0x06800066,
-	0x0467f0f2,
-	0x07e407f1,
-	0xd00604b6,
-	0x04bd0006,
-/* 0x05b8: memx_func_leave_wait */
-	0x07c067f1,
-	0xcf0664b6,
-	0x64f00066,
-	0xf31bf404,
-	0x161087f1,
-	0xf4028eb9,
-	0xd7b90421,
-	0xcc67f102,
-	0xff63f1ff,
-	0x0476fdff,
-	0x70f980f9,
-	0xe0fcd0fc,
-	0xf83f21f4,
-/* 0x05ed: memx_func_wait_vblank */
-	0x00169800,
-	0xf40066b0,
-	0x66b0130b,
-	0x060bf401,
-/* 0x05ff: memx_func_wait_vblank_head1 */
-	0xf12e0ef4,
-	0xf4002077,
-/* 0x0606: memx_func_wait_vblank_head0 */
-	0x77f1070e,
-/* 0x060a: memx_func_wait_vblank_0 */
-	0x67f10008,
-	0x64b607c4,
-	0x0066cf06,
-	0xf40467fd,
-/* 0x061a: memx_func_wait_vblank_1 */
-	0x67f1f31b,
-	0x64b607c4,
-	0x0066cf06,
-	0xf40467fd,
-/* 0x062a: memx_func_wait_vblank_fini */
-	0x10b6f30b,
-/* 0x062f: memx_func_wr32 */
-	0x9800f804,
-	0x15980016,
-	0x0810b601,
-	0x50f960f9,
-	0xe0fcd0fc,
-	0xb63f21f4,
-	0x1bf40242,
-/* 0x064b: memx_func_wait */
-	0xf000f8e9,
-	0x84b62c87,
-	0x0088cf06,
-	0x98001e98,
-	0x1c98011d,
-	0x031b9802,
-	0xf41010b6,
-	0x00f8a421,
-/* 0x0668: memx_func_delay */
-	0xb6001e98,
-	0x21f40410,
-/* 0x0673: memx_func_train */
-	0xf100f87f,
-	0xf1000357,
-	0xf1000077,
-	0xf0000097,
-	0x9eb97093,
-	0x0421f402,
-	0xf102d8b9,
-	0xf42710e7,
-/* 0x0692: memx_func_train_loop_outer */
-	0x58e07f21,
-	0x83f10101,
-	0x97f10200,
-	0x93f011e0,
-	0xf990f911,
-	0xfcd0fc80,
-	0x3f21f4e0,
-	0x67f150f9,
-/* 0x06b2: memx_func_train_loop_inner */
-	0x87f10000,
-	0x68ff1111,
-	0x10989490,
-	0xf10589fd,
-	0xf0072097,
-	0x90f91093,
-	0xd0fc80f9,
-	0x21f4e0fc,
-	0x8097f13f,
-	0x1093f000,
-	0xf4029eb9,
-	0xd8b90421,
-	0x2088c502,
+	0xf1f31bf4,
+	0xb9161087,
+	0x21f4028e,
+	0x02d7b904,
+	0xffcc67f1,
+	0xffff63f1,
+	0xf90476fd,
+	0xfc70f980,
+	0xf4e0fcd0,
+	0x00f84021,
+/* 0x05f0: memx_func_wait_vblank */
+	0xb0001698,
+	0x0bf40066,
+	0x0166b013,
+	0xf4060bf4,
+/* 0x0602: memx_func_wait_vblank_head1 */
+	0x77f12e0e,
+	0x0ef40020,
+/* 0x0609: memx_func_wait_vblank_head0 */
+	0x0877f107,
+/* 0x060d: memx_func_wait_vblank_0 */
+	0xc467f100,
+	0x0664b607,
+	0xfd0066cf,
+	0x1bf40467,
+/* 0x061d: memx_func_wait_vblank_1 */
+	0xc467f1f3,
+	0x0664b607,
+	0xfd0066cf,
+	0x0bf40467,
+/* 0x062d: memx_func_wait_vblank_fini */
+	0x0410b6f3,
+/* 0x0632: memx_func_wr32 */
+	0x169800f8,
+	0x01159800,
+	0xf90810b6,
+	0xfc50f960,
+	0xf4e0fcd0,
+	0x42b64021,
+	0xe91bf402,
+/* 0x064e: memx_func_wait */
+	0x87f000f8,
+	0x0684b62c,
+	0x980088cf,
+	0x1d98001e,
+	0x021c9801,
+	0xb6031b98,
+	0x21f41010,
+/* 0x066b: memx_func_delay */
+	0x9800f8a3,
+	0x10b6001e,
+	0x7e21f404,
+/* 0x0676: memx_func_train */
+	0x57f100f8,
+	0x77f10003,
+	0x97f10000,
+	0x93f00000,
+	0x029eb970,
+	0xb90421f4,
+	0xe7f102d8,
+	0x21f42710,
+/* 0x0695: memx_func_train_loop_outer */
+	0x0158e07e,
+	0x0083f101,
+	0xe097f102,
+	0x1193f011,
 	0x80f990f9,
 	0xe0fcd0fc,
-	0xf13f21f4,
-	0xf0053c97,
-	0x87f11093,
-	0x83f13002,
-	0x90f98000,
-	0xd0fc80f9,
-	0x21f4e0fc,
-	0x60e7f13f,
-	0x10e3f005,
-	0x0000d7f1,
-	0x8000d3f1,
-	0xf100dc90,
-	0xf08480b7,
-	0x21f41eb3,
-	0x0057f1a4,
-	0xff97f100,
-	0x0093f1ff,
-/* 0x0731: memx_func_train_loop_4x */
-	0x80a7f183,
-	0x10a3f000,
-	0xf402aeb9,
-	0xd8b90421,
-	0xdfb7f102,
-	0xffb3f1ff,
-	0x048bfdff,
-	0x80f9a0f9,
-	0xe0fcd0fc,
-	0xf13f21f4,
-	0xf0053ca7,
-	0x87f110a3,
-	0x83f13002,
-	0xa0f98000,
-	0xd0fc80f9,
-	0x21f4e0fc,
-	0x60e7f13f,
-	0x10e3f005,
-	0x0000d7f1,
-	0x8000d3f1,
-	0xf102dcb9,
-	0xf02710b7,
-	0x21f400b3,
-	0x02eeb9a4,
-	0xb90421f4,
-	0x9dff02dd,
-	0x0150b694,
-	0xf4045670,
-	0x7aa0921e,
-	0xa9800bcc,
-	0x0160b600,
-	0x700470b6,
-	0x1ef51066,
-	0x50fcff00,
+	0xf94021f4,
+	0x0067f150,
+/* 0x06b5: memx_func_train_loop_inner */
+	0x1187f100,
+	0x9068ff11,
+	0xfd109894,
+	0x97f10589,
+	0x93f00720,
+	0xf990f910,
+	0xfcd0fc80,
+	0x4021f4e0,
+	0x008097f1,
+	0xb91093f0,
+	0x21f4029e,
+	0x02d8b904,
+	0xf92088c5,
+	0xfc80f990,
+	0xf4e0fcd0,
+	0x97f14021,
+	0x93f0053c,
+	0x0287f110,
+	0x0083f130,
+	0xf990f980,
+	0xfcd0fc80,
+	0x4021f4e0,
+	0x0560e7f1,
+	0xf110e3f0,
+	0xf10000d7,
+	0x908000d3,
+	0xb7f100dc,
+	0xb3f08480,
+	0xa321f41e,
+	0x000057f1,
+	0xffff97f1,
+	0x830093f1,
+/* 0x0734: memx_func_train_loop_4x */
+	0x0080a7f1,
+	0xb910a3f0,
+	0x21f402ae,
+	0x02d8b904,
+	0xffdfb7f1,
+	0xffffb3f1,
+	0xf9048bfd,
+	0xfc80f9a0,
+	0xf4e0fcd0,
+	0xa7f14021,
+	0xa3f0053c,
+	0x0287f110,
+	0x0083f130,
+	0xf9a0f980,
+	0xfcd0fc80,
+	0x4021f4e0,
+	0x0560e7f1,
+	0xf110e3f0,
+	0xf10000d7,
+	0xb98000d3,
+	0xb7f102dc,
+	0xb3f02710,
+	0xa321f400,
+	0xf402eeb9,
+	0xddb90421,
+	0x949dff02,
 	0x700150b6,
-	0x1ef50756,
-	0x00f8fed4,
-/* 0x07c4: memx_exec */
-	0xd0f9e0f9,
-	0xb902c1b9,
-/* 0x07ce: memx_exec_next */
-	0x139802b2,
-	0x0410b600,
-	0x01f034e7,
-	0x01e033e7,
-	0xf00132b6,
-	0x35980c30,
-	0xb855f9de,
-	0x1ef40612,
-	0xf10b98e4,
-	0xbbf20c98,
-	0xb7f102cb,
-	0xb4b607c4,
-	0x00bbcf06,
-	0xe0fcd0fc,
-	0x033721f5,
-/* 0x080a: memx_info */
-	0xc67000f8,
-	0x0e0bf401,
-/* 0x0810: memx_info_data */
-	0x03ccc7f1,
-	0x0800b7f1,
-/* 0x081b: memx_info_train */
-	0xf10b0ef4,
-	0xf10bccc7,
-/* 0x0823: memx_info_send */
-	0xf50100b7,
-	0xf8033721,
-/* 0x0829: memx_recv */
-	0x01d6b000,
-	0xb0980bf4,
-	0x0bf400d6,
-/* 0x0837: memx_init */
-	0xf800f8d8,
-/* 0x0839: perf_recv */
-/* 0x083b: perf_init */
-	0xf800f800,
-/* 0x083d: i2c_drive_scl */
-	0x0036b000,
-	0xf1110bf4,
-	0xb607e007,
-	0x01d00604,
-	0xf804bd00,
-/* 0x0851: i2c_drive_scl_lo */
-	0xe407f100,
-	0x0604b607,
-	0xbd0001d0,
-/* 0x085f: i2c_drive_sda */
-	0xb000f804,
-	0x0bf40036,
-	0xe007f111,
-	0x0604b607,
-	0xbd0002d0,
-/* 0x0873: i2c_drive_sda_lo */
-	0xf100f804,
-	0xb607e407,
-	0x02d00604,
-	0xf804bd00,
-/* 0x0881: i2c_sense_scl */
-	0x0132f400,
-	0x07c437f1,
-	0xcf0634b6,
-	0x31fd0033,
-	0x060bf404,
-/* 0x0897: i2c_sense_scl_done */
-	0xf80131f4,
-/* 0x0899: i2c_sense_sda */
-	0x0132f400,
-	0x07c437f1,
-	0xcf0634b6,
-	0x32fd0033,
-	0x060bf404,
-/* 0x08af: i2c_sense_sda_done */
-	0xf80131f4,
-/* 0x08b1: i2c_raise_scl */
-	0xf140f900,
-	0xf0089847,
-	0x21f50137,
-/* 0x08be: i2c_raise_scl_wait */
-	0xe7f1083d,
-	0x21f403e8,
-	0x8121f57f,
-	0x0901f408,
-	0xf40142b6,
-/* 0x08d2: i2c_raise_scl_done */
-	0x40fcef1b,
-/* 0x08d6: i2c_start */
-	0x21f500f8,
-	0x11f40881,
-	0x9921f50d,
-	0x0611f408,
-/* 0x08e7: i2c_start_rep */
-	0xf0300ef4,
-	0x21f50037,
-	0x37f0083d,
-	0x5f21f501,
-	0x0076bb08,
-	0xf90465b6,
-	0x04659450,
-	0xbd0256bb,
-	0x0475fd50,
-	0x21f550fc,
-	0x64b608b1,
-	0x1f11f404,
-/* 0x0914: i2c_start_send */
-	0xf50037f0,
-	0xf1085f21,
-	0xf41388e7,
-	0x37f07f21,
-	0x3d21f500,
-	0x88e7f108,
-	0x7f21f413,
-/* 0x0930: i2c_start_out */
-/* 0x0932: i2c_stop */
-	0x37f000f8,
-	0x3d21f500,
-	0x0037f008,
-	0x085f21f5,
-	0x03e8e7f1,
-	0xf07f21f4,
-	0x21f50137,
-	0xe7f1083d,
-	0x21f41388,
-	0x0137f07f,
-	0x085f21f5,
-	0x1388e7f1,
-	0xf87f21f4,
-/* 0x0965: i2c_bitw */
-	0x5f21f500,
-	0xe8e7f108,
-	0x7f21f403,
-	0xb60076bb,
-	0x50f90465,
-	0xbb046594,
-	0x50bd0256,
-	0xfc0475fd,
-	0xb121f550,
-	0x0464b608,
-	0xf11811f4,
-	0xf41388e7,
-	0x37f07f21,
-	0x3d21f500,
-	0x88e7f108,
-	0x7f21f413,
-/* 0x09a4: i2c_bitw_out */
-/* 0x09a6: i2c_bitr */
-	0x37f000f8,
-	0x5f21f501,
+	0x1ef40456,
+	0xcc7aa092,
+	0x00a9800b,
+	0xb60160b6,
+	0x66700470,
+	0x001ef510,
+	0xb650fcff,
+	0x56700150,
+	0xd41ef507,
+/* 0x07c7: memx_exec */
+	0xf900f8fe,
+	0xb9d0f9e0,
+	0xb2b902c1,
+/* 0x07d1: memx_exec_next */
+	0x00139802,
+	0xe70410b6,
+	0xe701f034,
+	0xb601e033,
+	0x30f00132,
+	0xde35980c,
+	0x12b855f9,
+	0xe41ef406,
+	0x98f10b98,
+	0xcbbbf20c,
+	0xc4b7f102,
+	0x06b4b607,
+	0xfc00bbcf,
+	0xf5e0fcd0,
+	0xf8033621,
+/* 0x080d: memx_info */
+	0x01c67000,
+/* 0x0813: memx_info_data */
+	0xf10e0bf4,
+	0xf103ccc7,
+	0xf40800b7,
+/* 0x081e: memx_info_train */
+	0xc7f10b0e,
+	0xb7f10bcc,
+/* 0x0826: memx_info_send */
+	0x21f50100,
+	0x00f80336,
+/* 0x082c: memx_recv */
+	0xf401d6b0,
+	0xd6b0980b,
+	0xd80bf400,
+/* 0x083a: memx_init */
+	0x00f800f8,
+/* 0x083c: perf_recv */
+/* 0x083e: perf_init */
+	0x00f800f8,
+/* 0x0840: i2c_drive_scl */
+	0xf40036b0,
+	0x07f1110b,
+	0x04b607e0,
+	0x0001d006,
+	0x00f804bd,
+/* 0x0854: i2c_drive_scl_lo */
+	0x07e407f1,
+	0xd00604b6,
+	0x04bd0001,
+/* 0x0862: i2c_drive_sda */
+	0x36b000f8,
+	0x110bf400,
+	0x07e007f1,
+	0xd00604b6,
+	0x04bd0002,
+/* 0x0876: i2c_drive_sda_lo */
+	0x07f100f8,
+	0x04b607e4,
+	0x0002d006,
+	0x00f804bd,
+/* 0x0884: i2c_sense_scl */
+	0xf10132f4,
+	0xb607c437,
+	0x33cf0634,
+	0x0431fd00,
+	0xf4060bf4,
+/* 0x089a: i2c_sense_scl_done */
+	0x00f80131,
+/* 0x089c: i2c_sense_sda */
+	0xf10132f4,
+	0xb607c437,
+	0x33cf0634,
+	0x0432fd00,
+	0xf4060bf4,
+/* 0x08b2: i2c_sense_sda_done */
+	0x00f80131,
+/* 0x08b4: i2c_raise_scl */
+	0x47f140f9,
+	0x37f00898,
+	0x4021f501,
+/* 0x08c1: i2c_raise_scl_wait */
 	0xe8e7f108,
-	0x7f21f403,
+	0x7e21f403,
+	0x088421f5,
+	0xb60901f4,
+	0x1bf40142,
+/* 0x08d5: i2c_raise_scl_done */
+	0xf840fcef,
+/* 0x08d9: i2c_start */
+	0x8421f500,
+	0x0d11f408,
+	0x089c21f5,
+	0xf40611f4,
+/* 0x08ea: i2c_start_rep */
+	0x37f0300e,
+	0x4021f500,
+	0x0137f008,
+	0x086221f5,
 	0xb60076bb,
 	0x50f90465,
 	0xbb046594,
 	0x50bd0256,
 	0xfc0475fd,
-	0xb121f550,
+	0xb421f550,
 	0x0464b608,
-	0xf51b11f4,
-	0xf0089921,
+/* 0x0917: i2c_start_send */
+	0xf01f11f4,
 	0x21f50037,
-	0xe7f1083d,
+	0xe7f10862,
 	0x21f41388,
-	0x013cf07f,
-/* 0x09eb: i2c_bitr_done */
-	0xf80131f4,
-/* 0x09ed: i2c_get_byte */
-	0x0057f000,
-/* 0x09f3: i2c_get_byte_next */
-	0xb60847f0,
-	0x76bb0154,
-	0x0465b600,
-	0x659450f9,
-	0x0256bb04,
-	0x75fd50bd,
-	0xf550fc04,
-	0xb609a621,
-	0x11f40464,
-	0x0553fd2b,
-	0xf40142b6,
-	0x37f0d81b,
+	0x0037f07e,
+	0x084021f5,
+	0x1388e7f1,
+/* 0x0933: i2c_start_out */
+	0xf87e21f4,
+/* 0x0935: i2c_stop */
+	0x0037f000,
+	0x084021f5,
+	0xf50037f0,
+	0xf1086221,
+	0xf403e8e7,
+	0x37f07e21,
+	0x4021f501,
+	0x88e7f108,
+	0x7e21f413,
+	0xf50137f0,
+	0xf1086221,
+	0xf41388e7,
+	0x00f87e21,
+/* 0x0968: i2c_bitw */
+	0x086221f5,
+	0x03e8e7f1,
+	0xbb7e21f4,
+	0x65b60076,
+	0x9450f904,
+	0x56bb0465,
+	0xfd50bd02,
+	0x50fc0475,
+	0x08b421f5,
+	0xf40464b6,
+	0xe7f11811,
+	0x21f41388,
+	0x0037f07e,
+	0x084021f5,
+	0x1388e7f1,
+/* 0x09a7: i2c_bitw_out */
+	0xf87e21f4,
+/* 0x09a9: i2c_bitr */
+	0x0137f000,
+	0x086221f5,
+	0x03e8e7f1,
+	0xbb7e21f4,
+	0x65b60076,
+	0x9450f904,
+	0x56bb0465,
+	0xfd50bd02,
+	0x50fc0475,
+	0x08b421f5,
+	0xf40464b6,
+	0x21f51b11,
+	0x37f0089c,
+	0x4021f500,
+	0x88e7f108,
+	0x7e21f413,
+	0xf4013cf0,
+/* 0x09ee: i2c_bitr_done */
+	0x00f80131,
+/* 0x09f0: i2c_get_byte */
+	0xf00057f0,
+/* 0x09f6: i2c_get_byte_next */
+	0x54b60847,
 	0x0076bb01,
 	0xf90465b6,
 	0x04659450,
 	0xbd0256bb,
 	0x0475fd50,
 	0x21f550fc,
-	0x64b60965,
-/* 0x0a3d: i2c_get_byte_done */
-/* 0x0a3f: i2c_put_byte */
-	0xf000f804,
-/* 0x0a42: i2c_put_byte_next */
-	0x42b60847,
-	0x3854ff01,
+	0x64b609a9,
+	0x2b11f404,
+	0xb60553fd,
+	0x1bf40142,
+	0x0137f0d8,
 	0xb60076bb,
 	0x50f90465,
 	0xbb046594,
 	0x50bd0256,
 	0xfc0475fd,
-	0x6521f550,
+	0x6821f550,
 	0x0464b609,
-	0xb03411f4,
-	0x1bf40046,
-	0x0076bbd8,
+/* 0x0a40: i2c_get_byte_done */
+/* 0x0a42: i2c_put_byte */
+	0x47f000f8,
+/* 0x0a45: i2c_put_byte_next */
+	0x0142b608,
+	0xbb3854ff,
+	0x65b60076,
+	0x9450f904,
+	0x56bb0465,
+	0xfd50bd02,
+	0x50fc0475,
+	0x096821f5,
+	0xf40464b6,
+	0x46b03411,
+	0xd81bf400,
+	0xb60076bb,
+	0x50f90465,
+	0xbb046594,
+	0x50bd0256,
+	0xfc0475fd,
+	0xa921f550,
+	0x0464b609,
+	0xbb0f11f4,
+	0x36b00076,
+	0x061bf401,
+/* 0x0a9b: i2c_put_byte_done */
+	0xf80132f4,
+/* 0x0a9d: i2c_addr */
+	0x0076bb00,
 	0xf90465b6,
 	0x04659450,
 	0xbd0256bb,
 	0x0475fd50,
 	0x21f550fc,
-	0x64b609a6,
-	0x0f11f404,
-	0xb00076bb,
-	0x1bf40136,
-	0x0132f406,
-/* 0x0a98: i2c_put_byte_done */
-/* 0x0a9a: i2c_addr */
-	0x76bb00f8,
+	0x64b608d9,
+	0x2911f404,
+	0x012ec3e7,
+	0xfd0134b6,
+	0x76bb0553,
 	0x0465b600,
 	0x659450f9,
 	0x0256bb04,
 	0x75fd50bd,
 	0xf550fc04,
-	0xb608d621,
-	0x11f40464,
-	0x2ec3e729,
-	0x0134b601,
-	0xbb0553fd,
+	0xb60a4221,
+/* 0x0ae2: i2c_addr_done */
+	0x00f80464,
+/* 0x0ae4: i2c_acquire_addr */
+	0xb6f8cec7,
+	0xe0b702e4,
+	0xee980d1c,
+/* 0x0af3: i2c_acquire */
+	0xf500f800,
+	0xf40ae421,
+	0xd9f00421,
+	0x4021f403,
+/* 0x0b02: i2c_release */
+	0x21f500f8,
+	0x21f40ae4,
+	0x03daf004,
+	0xf84021f4,
+/* 0x0b11: i2c_recv */
+	0x0132f400,
+	0xb6f8c1c7,
+	0x16b00214,
+	0x3a1ff528,
+	0xf413a001,
+	0x0032980c,
+	0x0ccc13a0,
+	0xf4003198,
+	0xd0f90231,
+	0xd0f9e0f9,
+	0x000067f1,
+	0x100063f1,
+	0xbb016792,
 	0x65b60076,
 	0x9450f904,
 	0x56bb0465,
 	0xfd50bd02,
 	0x50fc0475,
-	0x0a3f21f5,
-/* 0x0adf: i2c_addr_done */
-	0xf80464b6,
-/* 0x0ae1: i2c_acquire_addr */
-	0xf8cec700,
-	0xb702e4b6,
-	0x980d1ce0,
-	0x00f800ee,
-/* 0x0af0: i2c_acquire */
-	0x0ae121f5,
-	0xf00421f4,
-	0x21f403d9,
-/* 0x0aff: i2c_release */
-	0xf500f83f,
-	0xf40ae121,
-	0xdaf00421,
-	0x3f21f403,
-/* 0x0b0e: i2c_recv */
-	0x32f400f8,
-	0xf8c1c701,
-	0xb00214b6,
-	0x1ff52816,
-	0x13a0013a,
-	0x32980cf4,
-	0xcc13a000,
-	0x0031980c,
-	0xf90231f4,
-	0xf9e0f9d0,
-	0x0067f1d0,
-	0x0063f100,
-	0x01679210,
-	0xb60076bb,
-	0x50f90465,
-	0xbb046594,
-	0x50bd0256,
-	0xfc0475fd,
-	0xf021f550,
-	0x0464b60a,
-	0xd6b0d0fc,
-	0xb31bf500,
-	0x0057f000,
-	0xb60076bb,
-	0x50f90465,
-	0xbb046594,
-	0x50bd0256,
-	0xfc0475fd,
-	0x9a21f550,
-	0x0464b60a,
-	0x00d011f5,
-	0xbbe0c5c7,
+	0x0af321f5,
+	0xfc0464b6,
+	0x00d6b0d0,
+	0x00b31bf5,
+	0xbb0057f0,
 	0x65b60076,
 	0x9450f904,
 	0x56bb0465,
 	0xfd50bd02,
 	0x50fc0475,
-	0x0a3f21f5,
+	0x0a9d21f5,
 	0xf50464b6,
-	0xf000ad11,
-	0x76bb0157,
+	0xc700d011,
+	0x76bbe0c5,
 	0x0465b600,
 	0x659450f9,
 	0x0256bb04,
 	0x75fd50bd,
 	0xf550fc04,
-	0xb60a9a21,
+	0xb60a4221,
 	0x11f50464,
-	0x76bb008a,
-	0x0465b600,
-	0x659450f9,
-	0x0256bb04,
-	0x75fd50bd,
-	0xf550fc04,
-	0xb609ed21,
-	0x11f40464,
-	0xe05bcb6a,
-	0xb60076bb,
-	0x50f90465,
-	0xbb046594,
-	0x50bd0256,
-	0xfc0475fd,
-	0x3221f550,
-	0x0464b609,
-	0xbd025bb9,
-	0x430ef474,
-/* 0x0c14: i2c_recv_not_rd08 */
-	0xf401d6b0,
-	0x57f03d1b,
-	0x9a21f500,
-	0x3311f40a,
-	0xf5e0c5c7,
-	0xf40a3f21,
-	0x57f02911,
-	0x9a21f500,
-	0x1f11f40a,
-	0xf5e0b5c7,
-	0xf40a3f21,
-	0x21f51511,
-	0x74bd0932,
-	0xf408c5c7,
-	0x32f4091b,
-	0x030ef402,
-/* 0x0c54: i2c_recv_not_wr08 */
-/* 0x0c54: i2c_recv_done */
-	0xf5f8cec7,
-	0xfc0aff21,
-	0xf4d0fce0,
-	0x7cb90a12,
-	0x3721f502,
-/* 0x0c69: i2c_recv_exit */
-/* 0x0c6b: i2c_init */
-	0xf800f803,
-/* 0x0c6d: test_recv */
-	0xd817f100,
-	0x0614b605,
-	0xb60011cf,
-	0x07f10110,
-	0x04b605d8,
-	0x0001d006,
-	0xe7f104bd,
-	0xe3f1d900,
-	0x21f5134f,
-	0x00f80257,
-/* 0x0c94: test_init */
-	0x0800e7f1,
-	0x025721f5,
-/* 0x0c9e: idle_recv */
+	0x57f000ad,
+	0x0076bb01,
+	0xf90465b6,
+	0x04659450,
+	0xbd0256bb,
+	0x0475fd50,
+	0x21f550fc,
+	0x64b60a9d,
+	0x8a11f504,
+	0x0076bb00,
+	0xf90465b6,
+	0x04659450,
+	0xbd0256bb,
+	0x0475fd50,
+	0x21f550fc,
+	0x64b609f0,
+	0x6a11f404,
+	0xbbe05bcb,
+	0x65b60076,
+	0x9450f904,
+	0x56bb0465,
+	0xfd50bd02,
+	0x50fc0475,
+	0x093521f5,
+	0xb90464b6,
+	0x74bd025b,
+/* 0x0c17: i2c_recv_not_rd08 */
+	0xb0430ef4,
+	0x1bf401d6,
+	0x0057f03d,
+	0x0a9d21f5,
+	0xc73311f4,
+	0x21f5e0c5,
+	0x11f40a42,
+	0x0057f029,
+	0x0a9d21f5,
+	0xc71f11f4,
+	0x21f5e0b5,
+	0x11f40a42,
+	0x3521f515,
+	0xc774bd09,
+	0x1bf408c5,
+	0x0232f409,
+/* 0x0c57: i2c_recv_not_wr08 */
+/* 0x0c57: i2c_recv_done */
+	0xc7030ef4,
+	0x21f5f8ce,
+	0xe0fc0b02,
+	0x12f4d0fc,
+	0x027cb90a,
+	0x033621f5,
+/* 0x0c6c: i2c_recv_exit */
+/* 0x0c6e: i2c_init */
 	0x00f800f8,
-/* 0x0ca0: idle */
-	0xf10031f4,
-	0xb605d417,
-	0x11cf0614,
-	0x0110b600,
-	0x05d407f1,
-	0xd00604b6,
-	0x04bd0001,
-/* 0x0cbc: idle_loop */
-	0xf45817f0,
-/* 0x0cc2: idle_proc */
-/* 0x0cc2: idle_proc_exec */
-	0x10f90232,
-	0xf5021eb9,
-	0xfc034021,
-	0x0911f410,
-	0xf40231f4,
-/* 0x0cd6: idle_proc_next */
-	0x10b6ef0e,
-	0x061fb858,
-	0xf4e61bf4,
-	0x28f4dd02,
-	0xbb0ef400,
-	0x00000000,
+/* 0x0c70: test_recv */
+	0x05d817f1,
+	0xcf0614b6,
+	0x10b60011,
+	0xd807f101,
+	0x0604b605,
+	0xbd0001d0,
+	0x00e7f104,
+	0x4fe3f1d9,
+	0x5621f513,
+/* 0x0c97: test_init */
+	0xf100f802,
+	0xf50800e7,
+	0xf8025621,
+/* 0x0ca1: idle_recv */
+/* 0x0ca3: idle */
+	0xf400f800,
+	0x17f10031,
+	0x14b605d4,
+	0x0011cf06,
+	0xf10110b6,
+	0xb605d407,
+	0x01d00604,
+/* 0x0cbf: idle_loop */
+	0xf004bd00,
+	0x32f45817,
+/* 0x0cc5: idle_proc */
+/* 0x0cc5: idle_proc_exec */
+	0xb910f902,
+	0x21f5021e,
+	0x10fc033f,
+	0xf40911f4,
+	0x0ef40231,
+/* 0x0cd9: idle_proc_next */
+	0x5810b6ef,
+	0xf4061fb8,
+	0x02f4e61b,
+	0x0028f4dd,
+	0x00bb0ef4,
 	0x00000000,
 	0x00000000,
 	0x00000000,
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/pmu/fuc/host.fuc b/drivers/gpu/drm/nouveau/nvkm/subdev/pmu/fuc/host.fuc
index c2bb616a8da5..f2420a37f45b 100644
--- a/drivers/gpu/drm/nouveau/nvkm/subdev/pmu/fuc/host.fuc
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/pmu/fuc/host.fuc
@@ -98,8 +98,7 @@ host_send:
 // $r0  - zero
 host_recv:
 	// message from intr handler == HOST->PWR comms pending
-	mov $r1 (PROC_KERN & 0x0000ffff)
-	sethi $r1 (PROC_KERN & 0xffff0000)
+	imm32($r1, PROC_KERN)
 	cmp b32 $r14 $r1
 	bra e #host_send
 
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/pmu/fuc/kernel.fuc b/drivers/gpu/drm/nouveau/nvkm/subdev/pmu/fuc/kernel.fuc
index ad35fa57be94..c20a3bd33775 100644
--- a/drivers/gpu/drm/nouveau/nvkm/subdev/pmu/fuc/kernel.fuc
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/pmu/fuc/kernel.fuc
@@ -51,8 +51,7 @@ time_next: .b32 0
 // $r0  - zero
 rd32:
 	nv_iowr(NV_PPWR_MMIO_ADDR, $r14)
-	mov $r13 NV_PPWR_MMIO_CTRL_OP_RD
-	sethi $r13 NV_PPWR_MMIO_CTRL_TRIGGER
+	imm32($r13, NV_PPWR_MMIO_CTRL_OP_RD | NV_PPWR_MMIO_CTRL_TRIGGER)
 	nv_iowr(NV_PPWR_MMIO_CTRL, $r13)
 	rd32_wait:
 		nv_iord($r13, NV_PPWR_MMIO_CTRL)
@@ -70,9 +69,7 @@ rd32:
 wr32:
 	nv_iowr(NV_PPWR_MMIO_ADDR, $r14)
 	nv_iowr(NV_PPWR_MMIO_DATA, $r13)
-	mov $r13 NV_PPWR_MMIO_CTRL_OP_WR
-	or $r13 NV_PPWR_MMIO_CTRL_MASK_B32_0
-	sethi $r13 NV_PPWR_MMIO_CTRL_TRIGGER
+	imm32($r13, NV_PPWR_MMIO_CTRL_OP_WR | NV_PPWR_MMIO_CTRL_MASK_B32_0 | NV_PPWR_MMIO_CTRL_TRIGGER)
 
 #ifdef NVKM_FALCON_MMIO_TRAP
 	push $r13
@@ -215,8 +212,7 @@ intr:
 		bra z #intr_subintr_skip_fifo
 			nv_iord($r12, NV_PPWR_FIFO_INTR)
 			push $r12
-			mov $r14 (PROC_HOST & 0x0000ffff)
-			sethi $r14 (PROC_HOST & 0xffff0000)
+			imm32($r14, PROC_HOST)
 			mov $r13 KMSG_FIFO
 			call(send)
 			pop $r12
@@ -256,7 +252,7 @@ ticks_from_ns:
 
 	/* try not losing precision (multiply then divide) */
 	imm32($r13, HW_TICKS_PER_US)
-	call #mulu32_32_64
+	call(mulu32_32_64)
 
 	/* use an immeditate, it's ok because HW_TICKS_PER_US < 16 bits */
 	div $r12 $r12 1000
@@ -268,7 +264,7 @@ ticks_from_ns:
 	/* let's divide then multiply, too bad for the precision! */
 	div $r14 $r14 1000
 	imm32($r13, HW_TICKS_PER_US)
-	call #mulu32_32_64
+	call(mulu32_32_64)
 
 	/* this cannot overflow as long as HW_TICKS_PER_US < 1000 */
 
@@ -290,7 +286,7 @@ ticks_from_us:
 
 	/* simply multiply $us by HW_TICKS_PER_US */
 	imm32($r13, HW_TICKS_PER_US)
-	call #mulu32_32_64
+	call(mulu32_32_64)
 	mov b32 $r14 $r12
 
 	/* check if there wasn't any overflow */
@@ -511,14 +507,12 @@ init:
 #ifdef NVKM_FALCON_MMIO_UAS
 	// somehow allows the magic "access mmio via D[]" stuff that's
 	// used by the nv_rd32/nv_wr32 macros to work
-	mov $r1 0x0010
-	sethi $r1 NV_PPWR_UAS_CONFIG_ENABLE
+	imm32($r1, 0x10 | NV_PPWR_UAS_CONFIG_ENABLE)
 	nv_iowrs(NV_PPWR_UAS_CONFIG, $r1)
 #endif
 
 	// route all interrupts except user0/1 and pause to fuc
-	mov $r1 0x00e0
-	sethi $r1 0x00000000
+	imm32($r1, 0xe0)
 	nv_iowr(NV_PPWR_INTR_ROUTE, $r1)
 
 	// enable watchdog and subintr intrs
@@ -529,8 +523,8 @@ init:
 	nv_iowr(NV_PPWR_INTR_EN_SET, $r1)
 
 	// enable interrupts globally
-	mov $r1 #intr
-	sethi $r1 0x00000000
+	imm32($r1, #intr)
+	and $r1 0xffff
 	mov $iv0 $r1
 	bset $flags ie0
 
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/pmu/fuc/macros.fuc b/drivers/gpu/drm/nouveau/nvkm/subdev/pmu/fuc/macros.fuc
index 96fc984dafdc..3737bd27f74e 100644
--- a/drivers/gpu/drm/nouveau/nvkm/subdev/pmu/fuc/macros.fuc
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/pmu/fuc/macros.fuc
@@ -169,7 +169,7 @@
 */	.b32 0 /*
 */	.skip 64
 
-#if NV_PPWR_CHIPSET < GK208
+#if NVKM_PPWR_CHIPSET < GK208
 #define imm32(reg,val) /*
 */	movw reg  ((val) & 0x0000ffff) /*
 */	sethi reg ((val) & 0xffff0000)
@@ -252,12 +252,12 @@
 #endif
 
 #define st(size, addr, reg) /*
-*/	movw $r0 addr /*
+*/	imm32($r0, addr) /*
 */	st size D[$r0] reg /*
 */	clear b32 $r0
 
 #define ld(size, reg, addr) /*
-*/	movw $r0 addr /*
+*/	imm32($r0, addr)  /*
 */	ld size reg D[$r0] /*
 */	clear b32 $r0
 
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/pmu/fuc/test.fuc b/drivers/gpu/drm/nouveau/nvkm/subdev/pmu/fuc/test.fuc
index 0c3a71bf5459..9e3f4e690dd1 100644
--- a/drivers/gpu/drm/nouveau/nvkm/subdev/pmu/fuc/test.fuc
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/pmu/fuc/test.fuc
@@ -48,8 +48,7 @@ test_recv:
 	nv_iord($r1, NV_PPWR_DSCRATCH(2))
 	add b32 $r1 1
 	nv_iowr(NV_PPWR_DSCRATCH(2), $r1)
-	mov $r14 -0x2700 /* 0xd900, envyas grrr! */
-	sethi $r14 0x134f0000
+	imm32($r14, 0x134fd900)
 	call(timer)
 	ret
 
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/secboot/Kbuild b/drivers/gpu/drm/nouveau/nvkm/subdev/secboot/Kbuild
new file mode 100644
index 000000000000..b02b868a6589
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/secboot/Kbuild
@@ -0,0 +1,3 @@
+nvkm-y += nvkm/subdev/secboot/base.o
+nvkm-y += nvkm/subdev/secboot/gm200.o
+nvkm-y += nvkm/subdev/secboot/gm20b.o
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/secboot/base.c b/drivers/gpu/drm/nouveau/nvkm/subdev/secboot/base.c
new file mode 100644
index 000000000000..520facf9bc07
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/secboot/base.c
@@ -0,0 +1,288 @@
+/*
+ * Copyright (c) 2016, NVIDIA CORPORATION. All rights reserved.
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
+ * THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+ * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
+ * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER
+ * DEALINGS IN THE SOFTWARE.
+ */
+
+#include "priv.h"
+#include <subdev/timer.h>
+
+static const char *
+managed_falcons_names[] = {
+	[NVKM_SECBOOT_FALCON_PMU] = "PMU",
+	[NVKM_SECBOOT_FALCON_RESERVED] = "<reserved>",
+	[NVKM_SECBOOT_FALCON_FECS] = "FECS",
+	[NVKM_SECBOOT_FALCON_GPCCS] = "GPCCS",
+	[NVKM_SECBOOT_FALCON_END] = "<invalid>",
+};
+
+/*
+ * Helper falcon functions
+ */
+
+static int
+falcon_clear_halt_interrupt(struct nvkm_device *device, u32 base)
+{
+	int ret;
+
+	/* clear halt interrupt */
+	nvkm_mask(device, base + 0x004, 0x10, 0x10);
+	/* wait until halt interrupt is cleared */
+	ret = nvkm_wait_msec(device, 10, base + 0x008, 0x10, 0x0);
+	if (ret < 0)
+		return ret;
+
+	return 0;
+}
+
+static int
+falcon_wait_idle(struct nvkm_device *device, u32 base)
+{
+	int ret;
+
+	ret = nvkm_wait_msec(device, 10, base + 0x04c, 0xffff, 0x0);
+	if (ret < 0)
+		return ret;
+
+	return 0;
+}
+
+static int
+nvkm_secboot_falcon_enable(struct nvkm_secboot *sb)
+{
+	struct nvkm_device *device = sb->subdev.device;
+	int ret;
+
+	/* enable engine */
+	nvkm_mask(device, 0x200, sb->enable_mask, sb->enable_mask);
+	nvkm_rd32(device, 0x200);
+	ret = nvkm_wait_msec(device, 10, sb->base + 0x10c, 0x6, 0x0);
+	if (ret < 0) {
+		nvkm_mask(device, 0x200, sb->enable_mask, 0x0);
+		nvkm_error(&sb->subdev, "Falcon mem scrubbing timeout\n");
+		return ret;
+	}
+
+	ret = falcon_wait_idle(device, sb->base);
+	if (ret)
+		return ret;
+
+	/* enable IRQs */
+	nvkm_wr32(device, sb->base + 0x010, 0xff);
+	nvkm_mask(device, 0x640, sb->irq_mask, sb->irq_mask);
+	nvkm_mask(device, 0x644, sb->irq_mask, sb->irq_mask);
+
+	return 0;
+}
+
+static int
+nvkm_secboot_falcon_disable(struct nvkm_secboot *sb)
+{
+	struct nvkm_device *device = sb->subdev.device;
+
+	/* disable IRQs and wait for any previous code to complete */
+	nvkm_mask(device, 0x644, sb->irq_mask, 0x0);
+	nvkm_mask(device, 0x640, sb->irq_mask, 0x0);
+	nvkm_wr32(device, sb->base + 0x014, 0xff);
+
+	falcon_wait_idle(device, sb->base);
+
+	/* disable engine */
+	nvkm_mask(device, 0x200, sb->enable_mask, 0x0);
+
+	return 0;
+}
+
+int
+nvkm_secboot_falcon_reset(struct nvkm_secboot *sb)
+{
+	int ret;
+
+	ret = nvkm_secboot_falcon_disable(sb);
+	if (ret)
+		return ret;
+
+	ret = nvkm_secboot_falcon_enable(sb);
+	if (ret)
+		return ret;
+
+	return 0;
+}
+
+/**
+ * nvkm_secboot_falcon_run - run the falcon that will perform secure boot
+ *
+ * This function is to be called after all chip-specific preparations have
+ * been completed. It will start the falcon to perform secure boot, wait for
+ * it to halt, and report if an error occurred.
+ */
+int
+nvkm_secboot_falcon_run(struct nvkm_secboot *sb)
+{
+	struct nvkm_device *device = sb->subdev.device;
+	int ret;
+
+	/* Start falcon */
+	nvkm_wr32(device, sb->base + 0x100, 0x2);
+
+	/* Wait for falcon halt */
+	ret = nvkm_wait_msec(device, 100, sb->base + 0x100, 0x10, 0x10);
+	if (ret < 0)
+		return ret;
+
+	/* If mailbox register contains an error code, then ACR has failed */
+	ret = nvkm_rd32(device, sb->base + 0x040);
+	if (ret) {
+		nvkm_error(&sb->subdev, "ACR boot failed, ret 0x%08x", ret);
+		falcon_clear_halt_interrupt(device, sb->base);
+		return -EINVAL;
+	}
+
+	return 0;
+}
+
+
+/**
+ * nvkm_secboot_reset() - reset specified falcon
+ */
+int
+nvkm_secboot_reset(struct nvkm_secboot *sb, u32 falcon)
+{
+	/* Unmanaged falcon? */
+	if (!(BIT(falcon) & sb->func->managed_falcons)) {
+		nvkm_error(&sb->subdev, "cannot reset unmanaged falcon!\n");
+		return -EINVAL;
+	}
+
+	return sb->func->reset(sb, falcon);
+}
+
+/**
+ * nvkm_secboot_start() - start specified falcon
+ */
+int
+nvkm_secboot_start(struct nvkm_secboot *sb, u32 falcon)
+{
+	/* Unmanaged falcon? */
+	if (!(BIT(falcon) & sb->func->managed_falcons)) {
+		nvkm_error(&sb->subdev, "cannot start unmanaged falcon!\n");
+		return -EINVAL;
+	}
+
+	return sb->func->start(sb, falcon);
+}
+
+/**
+ * nvkm_secboot_is_managed() - check whether a given falcon is securely-managed
+ */
+bool
+nvkm_secboot_is_managed(struct nvkm_secboot *secboot,
+			enum nvkm_secboot_falcon fid)
+{
+	if (!secboot)
+		return false;
+
+	return secboot->func->managed_falcons & BIT(fid);
+}
+
+static int
+nvkm_secboot_oneinit(struct nvkm_subdev *subdev)
+{
+	struct nvkm_secboot *sb = nvkm_secboot(subdev);
+	int ret = 0;
+
+	/* Call chip-specific init function */
+	if (sb->func->init)
+		ret = sb->func->init(sb);
+	if (ret) {
+		nvkm_error(subdev, "Secure Boot initialization failed: %d\n",
+			   ret);
+		return ret;
+	}
+
+	/*
+	 * Build all blobs - the same blobs can be used to perform secure boot
+	 * multiple times
+	 */
+	if (sb->func->prepare_blobs)
+		ret = sb->func->prepare_blobs(sb);
+
+	return ret;
+}
+
+static int
+nvkm_secboot_fini(struct nvkm_subdev *subdev, bool suspend)
+{
+	struct nvkm_secboot *sb = nvkm_secboot(subdev);
+	int ret = 0;
+
+	if (sb->func->fini)
+		ret = sb->func->fini(sb, suspend);
+
+	return ret;
+}
+
+static void *
+nvkm_secboot_dtor(struct nvkm_subdev *subdev)
+{
+	struct nvkm_secboot *sb = nvkm_secboot(subdev);
+	void *ret = NULL;
+
+	if (sb->func->dtor)
+		ret = sb->func->dtor(sb);
+
+	return ret;
+}
+
+static const struct nvkm_subdev_func
+nvkm_secboot = {
+	.oneinit = nvkm_secboot_oneinit,
+	.fini = nvkm_secboot_fini,
+	.dtor = nvkm_secboot_dtor,
+};
+
+int
+nvkm_secboot_ctor(const struct nvkm_secboot_func *func,
+		  struct nvkm_device *device, int index,
+		  struct nvkm_secboot *sb)
+{
+	unsigned long fid;
+
+	nvkm_subdev_ctor(&nvkm_secboot, device, index, 0, &sb->subdev);
+	sb->func = func;
+
+	/* setup the performing falcon's base address and masks */
+	switch (func->boot_falcon) {
+	case NVKM_SECBOOT_FALCON_PMU:
+		sb->base = 0x10a000;
+		sb->irq_mask = 0x1000000;
+		sb->enable_mask = 0x2000;
+		break;
+	default:
+		nvkm_error(&sb->subdev, "invalid secure boot falcon\n");
+		return -EINVAL;
+	};
+
+	nvkm_debug(&sb->subdev, "securely managed falcons:\n");
+	for_each_set_bit(fid, &sb->func->managed_falcons,
+			 NVKM_SECBOOT_FALCON_END)
+		nvkm_debug(&sb->subdev, "- %s\n", managed_falcons_names[fid]);
+
+	return 0;
+}
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/secboot/gm200.c b/drivers/gpu/drm/nouveau/nvkm/subdev/secboot/gm200.c
new file mode 100644
index 000000000000..cc100dc940ea
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/secboot/gm200.c
@@ -0,0 +1,1489 @@
+/*
+ * Copyright (c) 2016, NVIDIA CORPORATION. All rights reserved.
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
+ * THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+ * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
+ * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER
+ * DEALINGS IN THE SOFTWARE.
+ */
+
+/*
+ * Secure boot is the process by which NVIDIA-signed firmware is loaded into
+ * some of the falcons of a GPU. For production devices this is the only way
+ * for the firmware to access useful (but sensitive) registers.
+ *
+ * A Falcon microprocessor supporting advanced security modes can run in one of
+ * three modes:
+ *
+ * - Non-secure (NS). In this mode, functionality is similar to Falcon
+ *   architectures before security modes were introduced (pre-Maxwell), but
+ *   capability is restricted. In particular, certain registers may be
+ *   inaccessible for reads and/or writes, and physical memory access may be
+ *   disabled (on certain Falcon instances). This is the only possible mode that
+ *   can be used if you don't have microcode cryptographically signed by NVIDIA.
+ *
+ * - Heavy Secure (HS). In this mode, the microprocessor is a black box - it's
+ *   not possible to read or write any Falcon internal state or Falcon registers
+ *   from outside the Falcon (for example, from the host system). The only way
+ *   to enable this mode is by loading microcode that has been signed by NVIDIA.
+ *   (The loading process involves tagging the IMEM block as secure, writing the
+ *   signature into a Falcon register, and starting execution. The hardware will
+ *   validate the signature, and if valid, grant HS privileges.)
+ *
+ * - Light Secure (LS). In this mode, the microprocessor has more privileges
+ *   than NS but fewer than HS. Some of the microprocessor state is visible to
+ *   host software to ease debugging. The only way to enable this mode is by HS
+ *   microcode enabling LS mode. Some privileges available to HS mode are not
+ *   available here. LS mode is introduced in GM20x.
+ *
+ * Secure boot consists in temporarily switching a HS-capable falcon (typically
+ * PMU) into HS mode in order to validate the LS firmwares of managed falcons,
+ * load them, and switch managed falcons into LS mode. Once secure boot
+ * completes, no falcon remains in HS mode.
+ *
+ * Secure boot requires a write-protected memory region (WPR) which can only be
+ * written by the secure falcon. On dGPU, the driver sets up the WPR region in
+ * video memory. On Tegra, it is set up by the bootloader and its location and
+ * size written into memory controller registers.
+ *
+ * The secure boot process takes place as follows:
+ *
+ * 1) A LS blob is constructed that contains all the LS firmwares we want to
+ *    load, along with their signatures and bootloaders.
+ *
+ * 2) A HS blob (also called ACR) is created that contains the signed HS
+ *    firmware in charge of loading the LS firmwares into their respective
+ *    falcons.
+ *
+ * 3) The HS blob is loaded (via its own bootloader) and executed on the
+ *    HS-capable falcon. It authenticates itself, switches the secure falcon to
+ *    HS mode and setup the WPR region around the LS blob (dGPU) or copies the
+ *    LS blob into the WPR region (Tegra).
+ *
+ * 4) The LS blob is now secure from all external tampering. The HS falcon
+ *    checks the signatures of the LS firmwares and, if valid, switches the
+ *    managed falcons to LS mode and makes them ready to run the LS firmware.
+ *
+ * 5) The managed falcons remain in LS mode and can be started.
+ *
+ */
+
+#include "priv.h"
+
+#include <core/gpuobj.h>
+#include <core/firmware.h>
+#include <subdev/fb.h>
+
+enum {
+	FALCON_DMAIDX_UCODE		= 0,
+	FALCON_DMAIDX_VIRT		= 1,
+	FALCON_DMAIDX_PHYS_VID		= 2,
+	FALCON_DMAIDX_PHYS_SYS_COH	= 3,
+	FALCON_DMAIDX_PHYS_SYS_NCOH	= 4,
+};
+
+/**
+ * struct fw_bin_header - header of firmware files
+ * @bin_magic:		always 0x3b1d14f0
+ * @bin_ver:		version of the bin format
+ * @bin_size:		entire image size including this header
+ * @header_offset:	offset of the firmware/bootloader header in the file
+ * @data_offset:	offset of the firmware/bootloader payload in the file
+ * @data_size:		size of the payload
+ *
+ * This header is located at the beginning of the HS firmware and HS bootloader
+ * files, to describe where the headers and data can be found.
+ */
+struct fw_bin_header {
+	u32 bin_magic;
+	u32 bin_ver;
+	u32 bin_size;
+	u32 header_offset;
+	u32 data_offset;
+	u32 data_size;
+};
+
+/**
+ * struct fw_bl_desc - firmware bootloader descriptor
+ * @start_tag:		starting tag of bootloader
+ * @desc_dmem_load_off:	DMEM offset of flcn_bl_dmem_desc
+ * @code_off:		offset of code section
+ * @code_size:		size of code section
+ * @data_off:		offset of data section
+ * @data_size:		size of data section
+ *
+ * This structure is embedded in bootloader firmware files at to describe the
+ * IMEM and DMEM layout expected by the bootloader.
+ */
+struct fw_bl_desc {
+	u32 start_tag;
+	u32 dmem_load_off;
+	u32 code_off;
+	u32 code_size;
+	u32 data_off;
+	u32 data_size;
+};
+
+
+/*
+ *
+ * LS blob structures
+ *
+ */
+
+/**
+ * struct lsf_ucode_desc - LS falcon signatures
+ * @prd_keys:		signature to use when the GPU is in production mode
+ * @dgb_keys:		signature to use when the GPU is in debug mode
+ * @b_prd_present:	whether the production key is present
+ * @b_dgb_present:	whether the debug key is present
+ * @falcon_id:		ID of the falcon the ucode applies to
+ *
+ * Directly loaded from a signature file.
+ */
+struct lsf_ucode_desc {
+	u8  prd_keys[2][16];
+	u8  dbg_keys[2][16];
+	u32 b_prd_present;
+	u32 b_dbg_present;
+	u32 falcon_id;
+};
+
+/**
+ * struct lsf_lsb_header - LS firmware header
+ * @signature:		signature to verify the firmware against
+ * @ucode_off:		offset of the ucode blob in the WPR region. The ucode
+ *                      blob contains the bootloader, code and data of the
+ *                      LS falcon
+ * @ucode_size:		size of the ucode blob, including bootloader
+ * @data_size:		size of the ucode blob data
+ * @bl_code_size:	size of the bootloader code
+ * @bl_imem_off:	offset in imem of the bootloader
+ * @bl_data_off:	offset of the bootloader data in WPR region
+ * @bl_data_size:	size of the bootloader data
+ * @app_code_off:	offset of the app code relative to ucode_off
+ * @app_code_size:	size of the app code
+ * @app_data_off:	offset of the app data relative to ucode_off
+ * @app_data_size:	size of the app data
+ * @flags:		flags for the secure bootloader
+ *
+ * This structure is written into the WPR region for each managed falcon. Each
+ * instance is referenced by the lsb_offset member of the corresponding
+ * lsf_wpr_header.
+ */
+struct lsf_lsb_header {
+	struct lsf_ucode_desc signature;
+	u32 ucode_off;
+	u32 ucode_size;
+	u32 data_size;
+	u32 bl_code_size;
+	u32 bl_imem_off;
+	u32 bl_data_off;
+	u32 bl_data_size;
+	u32 app_code_off;
+	u32 app_code_size;
+	u32 app_data_off;
+	u32 app_data_size;
+	u32 flags;
+#define LSF_FLAG_LOAD_CODE_AT_0		1
+#define LSF_FLAG_DMACTL_REQ_CTX		4
+#define LSF_FLAG_FORCE_PRIV_LOAD	8
+};
+
+/**
+ * struct lsf_wpr_header - LS blob WPR Header
+ * @falcon_id:		LS falcon ID
+ * @lsb_offset:		offset of the lsb_lsf_header in the WPR region
+ * @bootstrap_owner:	secure falcon reponsible for bootstrapping the LS falcon
+ * @lazy_bootstrap:	skip bootstrapping by ACR
+ * @status:		bootstrapping status
+ *
+ * An array of these is written at the beginning of the WPR region, one for
+ * each managed falcon. The array is terminated by an instance which falcon_id
+ * is LSF_FALCON_ID_INVALID.
+ */
+struct lsf_wpr_header {
+	u32  falcon_id;
+	u32  lsb_offset;
+	u32  bootstrap_owner;
+	u32  lazy_bootstrap;
+	u32  status;
+#define LSF_IMAGE_STATUS_NONE				0
+#define LSF_IMAGE_STATUS_COPY				1
+#define LSF_IMAGE_STATUS_VALIDATION_CODE_FAILED		2
+#define LSF_IMAGE_STATUS_VALIDATION_DATA_FAILED		3
+#define LSF_IMAGE_STATUS_VALIDATION_DONE		4
+#define LSF_IMAGE_STATUS_VALIDATION_SKIPPED		5
+#define LSF_IMAGE_STATUS_BOOTSTRAP_READY		6
+};
+
+
+/**
+ * struct ls_ucode_img_desc - descriptor of firmware image
+ * @descriptor_size:		size of this descriptor
+ * @image_size:			size of the whole image
+ * @bootloader_start_offset:	start offset of the bootloader in ucode image
+ * @bootloader_size:		size of the bootloader
+ * @bootloader_imem_offset:	start off set of the bootloader in IMEM
+ * @bootloader_entry_point:	entry point of the bootloader in IMEM
+ * @app_start_offset:		start offset of the LS firmware
+ * @app_size:			size of the LS firmware's code and data
+ * @app_imem_offset:		offset of the app in IMEM
+ * @app_imem_entry:		entry point of the app in IMEM
+ * @app_dmem_offset:		offset of the data in DMEM
+ * @app_resident_code_offset:	offset of app code from app_start_offset
+ * @app_resident_code_size:	size of the code
+ * @app_resident_data_offset:	offset of data from app_start_offset
+ * @app_resident_data_size:	size of data
+ *
+ * A firmware image contains the code, data, and bootloader of a given LS
+ * falcon in a single blob. This structure describes where everything is.
+ *
+ * This can be generated from a (bootloader, code, data) set if they have
+ * been loaded separately, or come directly from a file.
+ */
+struct ls_ucode_img_desc {
+	u32 descriptor_size;
+	u32 image_size;
+	u32 tools_version;
+	u32 app_version;
+	char date[64];
+	u32 bootloader_start_offset;
+	u32 bootloader_size;
+	u32 bootloader_imem_offset;
+	u32 bootloader_entry_point;
+	u32 app_start_offset;
+	u32 app_size;
+	u32 app_imem_offset;
+	u32 app_imem_entry;
+	u32 app_dmem_offset;
+	u32 app_resident_code_offset;
+	u32 app_resident_code_size;
+	u32 app_resident_data_offset;
+	u32 app_resident_data_size;
+	u32 nb_overlays;
+	struct {u32 start; u32 size; } load_ovl[64];
+	u32 compressed;
+};
+
+/**
+ * struct ls_ucode_img - temporary storage for loaded LS firmwares
+ * @node:		to link within lsf_ucode_mgr
+ * @falcon_id:		ID of the falcon this LS firmware is for
+ * @ucode_desc:		loaded or generated map of ucode_data
+ * @ucode_header:	header of the firmware
+ * @ucode_data:		firmware payload (code and data)
+ * @ucode_size:		size in bytes of data in ucode_data
+ * @wpr_header:		WPR header to be written to the LS blob
+ * @lsb_header:		LSB header to be written to the LS blob
+ *
+ * Preparing the WPR LS blob requires information about all the LS firmwares
+ * (size, etc) to be known. This structure contains all the data of one LS
+ * firmware.
+ */
+struct ls_ucode_img {
+	struct list_head node;
+	enum nvkm_secboot_falcon falcon_id;
+
+	struct ls_ucode_img_desc ucode_desc;
+	u32 *ucode_header;
+	u8 *ucode_data;
+	u32 ucode_size;
+
+	struct lsf_wpr_header wpr_header;
+	struct lsf_lsb_header lsb_header;
+};
+
+/**
+ * struct ls_ucode_mgr - manager for all LS falcon firmwares
+ * @count:	number of managed LS falcons
+ * @wpr_size:	size of the required WPR region in bytes
+ * @img_list:	linked list of lsf_ucode_img
+ */
+struct ls_ucode_mgr {
+	u16 count;
+	u32 wpr_size;
+	struct list_head img_list;
+};
+
+
+/*
+ *
+ * HS blob structures
+ *
+ */
+
+/**
+ * struct hsf_fw_header - HS firmware descriptor
+ * @sig_dbg_offset:	offset of the debug signature
+ * @sig_dbg_size:	size of the debug signature
+ * @sig_prod_offset:	offset of the production signature
+ * @sig_prod_size:	size of the production signature
+ * @patch_loc:		offset of the offset (sic) of where the signature is
+ * @patch_sig:		offset of the offset (sic) to add to sig_*_offset
+ * @hdr_offset:		offset of the load header (see struct hs_load_header)
+ * @hdr_size:		size of above header
+ *
+ * This structure is embedded in the HS firmware image at
+ * hs_bin_hdr.header_offset.
+ */
+struct hsf_fw_header {
+	u32 sig_dbg_offset;
+	u32 sig_dbg_size;
+	u32 sig_prod_offset;
+	u32 sig_prod_size;
+	u32 patch_loc;
+	u32 patch_sig;
+	u32 hdr_offset;
+	u32 hdr_size;
+};
+
+/**
+ * struct hsf_load_header - HS firmware load header
+ */
+struct hsf_load_header {
+	u32 non_sec_code_off;
+	u32 non_sec_code_size;
+	u32 data_dma_base;
+	u32 data_size;
+	u32 num_apps;
+	struct {
+		u32 sec_code_off;
+		u32 sec_code_size;
+	} app[0];
+};
+
+/**
+ * Convenience function to duplicate a firmware file in memory and check that
+ * it has the required minimum size.
+ */
+static void *
+gm200_secboot_load_firmware(struct nvkm_subdev *subdev, const char *name,
+		    size_t min_size)
+{
+	const struct firmware *fw;
+	void *blob;
+	int ret;
+
+	ret = nvkm_firmware_get(subdev->device, name, &fw);
+	if (ret)
+		return ERR_PTR(ret);
+	if (fw->size < min_size) {
+		nvkm_error(subdev, "%s is smaller than expected size %zu\n",
+			   name, min_size);
+		nvkm_firmware_put(fw);
+		return ERR_PTR(-EINVAL);
+	}
+	blob = kmemdup(fw->data, fw->size, GFP_KERNEL);
+	nvkm_firmware_put(fw);
+	if (!blob)
+		return ERR_PTR(-ENOMEM);
+
+	return blob;
+}
+
+
+/*
+ * Low-secure blob creation
+ */
+
+#define BL_DESC_BLK_SIZE 256
+/**
+ * Build a ucode image and descriptor from provided bootloader, code and data.
+ *
+ * @bl:		bootloader image, including 16-bytes descriptor
+ * @code:	LS firmware code segment
+ * @data:	LS firmware data segment
+ * @desc:	ucode descriptor to be written
+ *
+ * Return: allocated ucode image with corresponding descriptor information. desc
+ *         is also updated to contain the right offsets within returned image.
+ */
+static void *
+ls_ucode_img_build(const struct firmware *bl, const struct firmware *code,
+		   const struct firmware *data, struct ls_ucode_img_desc *desc)
+{
+	struct fw_bin_header *bin_hdr = (void *)bl->data;
+	struct fw_bl_desc *bl_desc = (void *)bl->data + bin_hdr->header_offset;
+	void *bl_data = (void *)bl->data + bin_hdr->data_offset;
+	u32 pos = 0;
+	void *image;
+
+	desc->bootloader_start_offset = pos;
+	desc->bootloader_size = ALIGN(bl_desc->code_size, sizeof(u32));
+	desc->bootloader_imem_offset = bl_desc->start_tag * 256;
+	desc->bootloader_entry_point = bl_desc->start_tag * 256;
+
+	pos = ALIGN(pos + desc->bootloader_size, BL_DESC_BLK_SIZE);
+	desc->app_start_offset = pos;
+	desc->app_size = ALIGN(code->size, BL_DESC_BLK_SIZE) +
+			 ALIGN(data->size, BL_DESC_BLK_SIZE);
+	desc->app_imem_offset = 0;
+	desc->app_imem_entry = 0;
+	desc->app_dmem_offset = 0;
+	desc->app_resident_code_offset = 0;
+	desc->app_resident_code_size = ALIGN(code->size, BL_DESC_BLK_SIZE);
+
+	pos = ALIGN(pos + desc->app_resident_code_size, BL_DESC_BLK_SIZE);
+	desc->app_resident_data_offset = pos - desc->app_start_offset;
+	desc->app_resident_data_size = ALIGN(data->size, BL_DESC_BLK_SIZE);
+
+	desc->image_size = ALIGN(bl_desc->code_size, BL_DESC_BLK_SIZE) +
+			   desc->app_size;
+
+	image = kzalloc(desc->image_size, GFP_KERNEL);
+	if (!image)
+		return ERR_PTR(-ENOMEM);
+
+	memcpy(image + desc->bootloader_start_offset, bl_data,
+	       bl_desc->code_size);
+	memcpy(image + desc->app_start_offset, code->data, code->size);
+	memcpy(image + desc->app_start_offset + desc->app_resident_data_offset,
+	       data->data, data->size);
+
+	return image;
+}
+
+/**
+ * ls_ucode_img_load_generic() - load and prepare a LS ucode image
+ *
+ * Load the LS microcode, bootloader and signature and pack them into a single
+ * blob. Also generate the corresponding ucode descriptor.
+ */
+static int
+ls_ucode_img_load_generic(struct nvkm_subdev *subdev,
+			  struct ls_ucode_img *img, const char *falcon_name,
+			  const u32 falcon_id)
+{
+	const struct firmware *bl, *code, *data;
+	struct lsf_ucode_desc *lsf_desc;
+	char f[64];
+	int ret;
+
+	img->ucode_header = NULL;
+
+	snprintf(f, sizeof(f), "gr/%s_bl", falcon_name);
+	ret = nvkm_firmware_get(subdev->device, f, &bl);
+	if (ret)
+		goto error;
+
+	snprintf(f, sizeof(f), "gr/%s_inst", falcon_name);
+	ret = nvkm_firmware_get(subdev->device, f, &code);
+	if (ret)
+		goto free_bl;
+
+	snprintf(f, sizeof(f), "gr/%s_data", falcon_name);
+	ret = nvkm_firmware_get(subdev->device, f, &data);
+	if (ret)
+		goto free_inst;
+
+	img->ucode_data = ls_ucode_img_build(bl, code, data,
+					     &img->ucode_desc);
+	if (IS_ERR(img->ucode_data)) {
+		ret = PTR_ERR(img->ucode_data);
+		goto free_data;
+	}
+	img->ucode_size = img->ucode_desc.image_size;
+
+	snprintf(f, sizeof(f), "gr/%s_sig", falcon_name);
+	lsf_desc = gm200_secboot_load_firmware(subdev, f, sizeof(*lsf_desc));
+	if (IS_ERR(lsf_desc)) {
+		ret = PTR_ERR(lsf_desc);
+		goto free_image;
+	}
+	/* not needed? the signature should already have the right value */
+	lsf_desc->falcon_id = falcon_id;
+	memcpy(&img->lsb_header.signature, lsf_desc, sizeof(*lsf_desc));
+	img->falcon_id = lsf_desc->falcon_id;
+	kfree(lsf_desc);
+
+	/* success path - only free requested firmware files */
+	goto free_data;
+
+free_image:
+	kfree(img->ucode_data);
+free_data:
+	nvkm_firmware_put(data);
+free_inst:
+	nvkm_firmware_put(code);
+free_bl:
+	nvkm_firmware_put(bl);
+error:
+	return ret;
+}
+
+typedef int (*lsf_load_func)(struct nvkm_subdev *, struct ls_ucode_img *);
+
+static int
+ls_ucode_img_load_fecs(struct nvkm_subdev *subdev, struct ls_ucode_img *img)
+{
+	return ls_ucode_img_load_generic(subdev, img, "fecs",
+					 NVKM_SECBOOT_FALCON_FECS);
+}
+
+static int
+ls_ucode_img_load_gpccs(struct nvkm_subdev *subdev, struct ls_ucode_img *img)
+{
+	return ls_ucode_img_load_generic(subdev, img, "gpccs",
+					 NVKM_SECBOOT_FALCON_GPCCS);
+}
+
+/**
+ * ls_ucode_img_load() - create a lsf_ucode_img and load it
+ */
+static struct ls_ucode_img *
+ls_ucode_img_load(struct nvkm_subdev *subdev, lsf_load_func load_func)
+{
+	struct ls_ucode_img *img;
+	int ret;
+
+	img = kzalloc(sizeof(*img), GFP_KERNEL);
+	if (!img)
+		return ERR_PTR(-ENOMEM);
+
+	ret = load_func(subdev, img);
+	if (ret) {
+		kfree(img);
+		return ERR_PTR(ret);
+	}
+
+	return img;
+}
+
+static const lsf_load_func lsf_load_funcs[] = {
+	[NVKM_SECBOOT_FALCON_END] = NULL, /* reserve enough space */
+	[NVKM_SECBOOT_FALCON_FECS] = ls_ucode_img_load_fecs,
+	[NVKM_SECBOOT_FALCON_GPCCS] = ls_ucode_img_load_gpccs,
+};
+
+/**
+ * ls_ucode_img_populate_bl_desc() - populate a DMEM BL descriptor for LS image
+ * @img:	ucode image to generate against
+ * @desc:	descriptor to populate
+ * @sb:		secure boot state to use for base addresses
+ *
+ * Populate the DMEM BL descriptor with the information contained in a
+ * ls_ucode_desc.
+ *
+ */
+static void
+ls_ucode_img_populate_bl_desc(struct ls_ucode_img *img, u64 wpr_addr,
+			      struct gm200_flcn_bl_desc *desc)
+{
+	struct ls_ucode_img_desc *pdesc = &img->ucode_desc;
+	u64 addr_base;
+
+	addr_base = wpr_addr + img->lsb_header.ucode_off +
+		    pdesc->app_start_offset;
+
+	memset(desc, 0, sizeof(*desc));
+	desc->ctx_dma = FALCON_DMAIDX_UCODE;
+	desc->code_dma_base.lo = lower_32_bits(
+		(addr_base + pdesc->app_resident_code_offset));
+	desc->code_dma_base.hi = upper_32_bits(
+		(addr_base + pdesc->app_resident_code_offset));
+	desc->non_sec_code_size = pdesc->app_resident_code_size;
+	desc->data_dma_base.lo = lower_32_bits(
+		(addr_base + pdesc->app_resident_data_offset));
+	desc->data_dma_base.hi = upper_32_bits(
+		(addr_base + pdesc->app_resident_data_offset));
+	desc->data_size = pdesc->app_resident_data_size;
+	desc->code_entry_point = pdesc->app_imem_entry;
+}
+
+#define LSF_LSB_HEADER_ALIGN 256
+#define LSF_BL_DATA_ALIGN 256
+#define LSF_BL_DATA_SIZE_ALIGN 256
+#define LSF_BL_CODE_SIZE_ALIGN 256
+#define LSF_UCODE_DATA_ALIGN 4096
+
+/**
+ * ls_ucode_img_fill_headers - fill the WPR and LSB headers of an image
+ * @gsb:	secure boot device used
+ * @img:	image to generate for
+ * @offset:	offset in the WPR region where this image starts
+ *
+ * Allocate space in the WPR area from offset and write the WPR and LSB headers
+ * accordingly.
+ *
+ * Return: offset at the end of this image.
+ */
+static u32
+ls_ucode_img_fill_headers(struct gm200_secboot *gsb, struct ls_ucode_img *img,
+			  u32 offset)
+{
+	struct lsf_wpr_header *whdr = &img->wpr_header;
+	struct lsf_lsb_header *lhdr = &img->lsb_header;
+	struct ls_ucode_img_desc *desc = &img->ucode_desc;
+
+	if (img->ucode_header) {
+		nvkm_fatal(&gsb->base.subdev,
+			    "images withough loader are not supported yet!\n");
+		return offset;
+	}
+
+	/* Fill WPR header */
+	whdr->falcon_id = img->falcon_id;
+	whdr->bootstrap_owner = gsb->base.func->boot_falcon;
+	whdr->status = LSF_IMAGE_STATUS_COPY;
+
+	/* Align, save off, and include an LSB header size */
+	offset = ALIGN(offset, LSF_LSB_HEADER_ALIGN);
+	whdr->lsb_offset = offset;
+	offset += sizeof(struct lsf_lsb_header);
+
+	/*
+	 * Align, save off, and include the original (static) ucode
+	 * image size
+	 */
+	offset = ALIGN(offset, LSF_UCODE_DATA_ALIGN);
+	lhdr->ucode_off = offset;
+	offset += img->ucode_size;
+
+	/*
+	 * For falcons that use a boot loader (BL), we append a loader
+	 * desc structure on the end of the ucode image and consider
+	 * this the boot loader data. The host will then copy the loader
+	 * desc args to this space within the WPR region (before locking
+	 * down) and the HS bin will then copy them to DMEM 0 for the
+	 * loader.
+	 */
+	lhdr->bl_code_size = ALIGN(desc->bootloader_size,
+				   LSF_BL_CODE_SIZE_ALIGN);
+	lhdr->ucode_size = ALIGN(desc->app_resident_data_offset,
+				 LSF_BL_CODE_SIZE_ALIGN) + lhdr->bl_code_size;
+	lhdr->data_size = ALIGN(desc->app_size, LSF_BL_CODE_SIZE_ALIGN) +
+				lhdr->bl_code_size - lhdr->ucode_size;
+	/*
+	 * Though the BL is located at 0th offset of the image, the VA
+	 * is different to make sure that it doesn't collide the actual
+	 * OS VA range
+	 */
+	lhdr->bl_imem_off = desc->bootloader_imem_offset;
+	lhdr->app_code_off = desc->app_start_offset +
+			     desc->app_resident_code_offset;
+	lhdr->app_code_size = desc->app_resident_code_size;
+	lhdr->app_data_off = desc->app_start_offset +
+			     desc->app_resident_data_offset;
+	lhdr->app_data_size = desc->app_resident_data_size;
+
+	lhdr->flags = 0;
+	if (img->falcon_id == gsb->base.func->boot_falcon)
+		lhdr->flags = LSF_FLAG_DMACTL_REQ_CTX;
+
+	/* GPCCS will be loaded using PRI */
+	if (img->falcon_id == NVKM_SECBOOT_FALCON_GPCCS)
+		lhdr->flags |= LSF_FLAG_FORCE_PRIV_LOAD;
+
+	/* Align (size bloat) and save off BL descriptor size */
+	lhdr->bl_data_size = ALIGN(sizeof(struct gm200_flcn_bl_desc),
+				   LSF_BL_DATA_SIZE_ALIGN);
+	/*
+	 * Align, save off, and include the additional BL data
+	 */
+	offset = ALIGN(offset, LSF_BL_DATA_ALIGN);
+	lhdr->bl_data_off = offset;
+	offset += lhdr->bl_data_size;
+
+	return offset;
+}
+
+static void
+ls_ucode_mgr_init(struct ls_ucode_mgr *mgr)
+{
+	memset(mgr, 0, sizeof(*mgr));
+	INIT_LIST_HEAD(&mgr->img_list);
+}
+
+static void
+ls_ucode_mgr_cleanup(struct ls_ucode_mgr *mgr)
+{
+	struct ls_ucode_img *img, *t;
+
+	list_for_each_entry_safe(img, t, &mgr->img_list, node) {
+		kfree(img->ucode_data);
+		kfree(img->ucode_header);
+		kfree(img);
+	}
+}
+
+static void
+ls_ucode_mgr_add_img(struct ls_ucode_mgr *mgr, struct ls_ucode_img *img)
+{
+	mgr->count++;
+	list_add_tail(&img->node, &mgr->img_list);
+}
+
+/**
+ * ls_ucode_mgr_fill_headers - fill WPR and LSB headers of all managed images
+ */
+static void
+ls_ucode_mgr_fill_headers(struct gm200_secboot *gsb, struct ls_ucode_mgr *mgr)
+{
+	struct ls_ucode_img *img;
+	u32 offset;
+
+	/*
+	 * Start with an array of WPR headers at the base of the WPR.
+	 * The expectation here is that the secure falcon will do a single DMA
+	 * read of this array and cache it internally so it's ok to pack these.
+	 * Also, we add 1 to the falcon count to indicate the end of the array.
+	 */
+	offset = sizeof(struct lsf_wpr_header) * (mgr->count + 1);
+
+	/*
+	 * Walk the managed falcons, accounting for the LSB structs
+	 * as well as the ucode images.
+	 */
+	list_for_each_entry(img, &mgr->img_list, node) {
+		offset = ls_ucode_img_fill_headers(gsb, img, offset);
+	}
+
+	mgr->wpr_size = offset;
+}
+
+/**
+ * ls_ucode_mgr_write_wpr - write the WPR blob contents
+ */
+static int
+ls_ucode_mgr_write_wpr(struct gm200_secboot *gsb, struct ls_ucode_mgr *mgr,
+		       struct nvkm_gpuobj *wpr_blob)
+{
+	struct ls_ucode_img *img;
+	u32 pos = 0;
+
+	nvkm_kmap(wpr_blob);
+
+	list_for_each_entry(img, &mgr->img_list, node) {
+		nvkm_gpuobj_memcpy_to(wpr_blob, pos, &img->wpr_header,
+				      sizeof(img->wpr_header));
+
+		nvkm_gpuobj_memcpy_to(wpr_blob, img->wpr_header.lsb_offset,
+				     &img->lsb_header, sizeof(img->lsb_header));
+
+		/* Generate and write BL descriptor */
+		if (!img->ucode_header) {
+			u8 desc[gsb->func->bl_desc_size];
+			struct gm200_flcn_bl_desc gdesc;
+
+			ls_ucode_img_populate_bl_desc(img, gsb->wpr_addr,
+						      &gdesc);
+			gsb->func->fixup_bl_desc(&gdesc, &desc);
+			nvkm_gpuobj_memcpy_to(wpr_blob,
+					      img->lsb_header.bl_data_off,
+					      &desc, gsb->func->bl_desc_size);
+		}
+
+		/* Copy ucode */
+		nvkm_gpuobj_memcpy_to(wpr_blob, img->lsb_header.ucode_off,
+				      img->ucode_data, img->ucode_size);
+
+		pos += sizeof(img->wpr_header);
+	}
+
+	nvkm_wo32(wpr_blob, pos, NVKM_SECBOOT_FALCON_INVALID);
+
+	nvkm_done(wpr_blob);
+
+	return 0;
+}
+
+/* Both size and address of WPR need to be 128K-aligned */
+#define WPR_ALIGNMENT	0x20000
+/**
+ * gm200_secboot_prepare_ls_blob() - prepare the LS blob
+ *
+ * For each securely managed falcon, load the FW, signatures and bootloaders and
+ * prepare a ucode blob. Then, compute the offsets in the WPR region for each
+ * blob, and finally write the headers and ucode blobs into a GPU object that
+ * will be copied into the WPR region by the HS firmware.
+ */
+static int
+gm200_secboot_prepare_ls_blob(struct gm200_secboot *gsb)
+{
+	struct nvkm_secboot *sb = &gsb->base;
+	struct nvkm_device *device = sb->subdev.device;
+	struct ls_ucode_mgr mgr;
+	int falcon_id;
+	int ret;
+
+	ls_ucode_mgr_init(&mgr);
+
+	/* Load all LS blobs */
+	for_each_set_bit(falcon_id, &gsb->base.func->managed_falcons,
+			 NVKM_SECBOOT_FALCON_END) {
+		struct ls_ucode_img *img;
+
+		img = ls_ucode_img_load(&sb->subdev, lsf_load_funcs[falcon_id]);
+
+		if (IS_ERR(img)) {
+			ret = PTR_ERR(img);
+			goto cleanup;
+		}
+		ls_ucode_mgr_add_img(&mgr, img);
+	}
+
+	/*
+	 * Fill the WPR and LSF headers with the right offsets and compute
+	 * required WPR size
+	 */
+	ls_ucode_mgr_fill_headers(gsb, &mgr);
+	mgr.wpr_size = ALIGN(mgr.wpr_size, WPR_ALIGNMENT);
+
+	/* Allocate GPU object that will contain the WPR region */
+	ret = nvkm_gpuobj_new(device, mgr.wpr_size, WPR_ALIGNMENT, false, NULL,
+			      &gsb->ls_blob);
+	if (ret)
+		goto cleanup;
+
+	nvkm_debug(&sb->subdev, "%d managed LS falcons, WPR size is %d bytes\n",
+		    mgr.count, mgr.wpr_size);
+
+	/* If WPR address and size are not fixed, set them to fit the LS blob */
+	if (!gsb->wpr_size) {
+		gsb->wpr_addr = gsb->ls_blob->addr;
+		gsb->wpr_size = gsb->ls_blob->size;
+	}
+
+	/* Write LS blob */
+	ret = ls_ucode_mgr_write_wpr(gsb, &mgr, gsb->ls_blob);
+
+cleanup:
+	ls_ucode_mgr_cleanup(&mgr);
+
+	return ret;
+}
+
+/*
+ * High-secure blob creation
+ */
+
+/**
+ * gm200_secboot_hsf_patch_signature() - patch HS blob with correct signature
+ */
+static void
+gm200_secboot_hsf_patch_signature(struct gm200_secboot *gsb, void *acr_image)
+{
+	struct nvkm_secboot *sb = &gsb->base;
+	struct fw_bin_header *hsbin_hdr = acr_image;
+	struct hsf_fw_header *fw_hdr = acr_image + hsbin_hdr->header_offset;
+	void *hs_data = acr_image + hsbin_hdr->data_offset;
+	void *sig;
+	u32 sig_size;
+
+	/* Falcon in debug or production mode? */
+	if ((nvkm_rd32(sb->subdev.device, sb->base + 0xc08) >> 20) & 0x1) {
+		sig = acr_image + fw_hdr->sig_dbg_offset;
+		sig_size = fw_hdr->sig_dbg_size;
+	} else {
+		sig = acr_image + fw_hdr->sig_prod_offset;
+		sig_size = fw_hdr->sig_prod_size;
+	}
+
+	/* Patch signature */
+	memcpy(hs_data + fw_hdr->patch_loc, sig + fw_hdr->patch_sig, sig_size);
+}
+
+/**
+ * gm200_secboot_populate_hsf_bl_desc() - populate BL descriptor for HS image
+ */
+static void
+gm200_secboot_populate_hsf_bl_desc(void *acr_image,
+				   struct gm200_flcn_bl_desc *bl_desc)
+{
+	struct fw_bin_header *hsbin_hdr = acr_image;
+	struct hsf_fw_header *fw_hdr = acr_image + hsbin_hdr->header_offset;
+	struct hsf_load_header *load_hdr = acr_image + fw_hdr->hdr_offset;
+
+	/*
+	 * Descriptor for the bootloader that will load the ACR image into
+	 * IMEM/DMEM memory.
+	 */
+	fw_hdr = acr_image + hsbin_hdr->header_offset;
+	load_hdr = acr_image + fw_hdr->hdr_offset;
+	memset(bl_desc, 0, sizeof(*bl_desc));
+	bl_desc->ctx_dma = FALCON_DMAIDX_VIRT;
+	bl_desc->non_sec_code_off = load_hdr->non_sec_code_off;
+	bl_desc->non_sec_code_size = load_hdr->non_sec_code_size;
+	bl_desc->sec_code_off = load_hdr->app[0].sec_code_off;
+	bl_desc->sec_code_size = load_hdr->app[0].sec_code_size;
+	bl_desc->code_entry_point = 0;
+	/*
+	 * We need to set code_dma_base to the virtual address of the acr_blob,
+	 * and add this address to data_dma_base before writing it into DMEM
+	 */
+	bl_desc->code_dma_base.lo = 0;
+	bl_desc->data_dma_base.lo = load_hdr->data_dma_base;
+	bl_desc->data_size = load_hdr->data_size;
+}
+
+/**
+ * gm200_secboot_prepare_hs_blob - load and prepare a HS blob and BL descriptor
+ *
+ * @gsb secure boot instance to prepare for
+ * @fw name of the HS firmware to load
+ * @blob pointer to gpuobj that will be allocated to receive the HS FW payload
+ * @bl_desc pointer to the BL descriptor to write for this firmware
+ * @patch whether we should patch the HS descriptor (only for HS loaders)
+ */
+static int
+gm200_secboot_prepare_hs_blob(struct gm200_secboot *gsb, const char *fw,
+			      struct nvkm_gpuobj **blob,
+			      struct gm200_flcn_bl_desc *bl_desc, bool patch)
+{
+	struct nvkm_subdev *subdev = &gsb->base.subdev;
+	void *acr_image;
+	struct fw_bin_header *hsbin_hdr;
+	struct hsf_fw_header *fw_hdr;
+	void *acr_data;
+	struct hsf_load_header *load_hdr;
+	struct hsflcn_acr_desc *desc;
+	int ret;
+
+	acr_image = gm200_secboot_load_firmware(subdev, fw, 0);
+	if (IS_ERR(acr_image))
+		return PTR_ERR(acr_image);
+	hsbin_hdr = acr_image;
+
+	/* Patch signature */
+	gm200_secboot_hsf_patch_signature(gsb, acr_image);
+
+	acr_data = acr_image + hsbin_hdr->data_offset;
+
+	/* Patch descriptor? */
+	if (patch) {
+		fw_hdr = acr_image + hsbin_hdr->header_offset;
+		load_hdr = acr_image + fw_hdr->hdr_offset;
+		desc = acr_data + load_hdr->data_dma_base;
+		gsb->func->fixup_hs_desc(gsb, desc);
+	}
+
+	/* Generate HS BL descriptor */
+	gm200_secboot_populate_hsf_bl_desc(acr_image, bl_desc);
+
+	/* Create ACR blob and copy HS data to it */
+	ret = nvkm_gpuobj_new(subdev->device, ALIGN(hsbin_hdr->data_size, 256),
+			      0x1000, false, NULL, blob);
+	if (ret)
+		goto cleanup;
+
+	nvkm_kmap(*blob);
+	nvkm_gpuobj_memcpy_to(*blob, 0, acr_data, hsbin_hdr->data_size);
+	nvkm_done(*blob);
+
+cleanup:
+	kfree(acr_image);
+
+	return ret;
+}
+
+/*
+ * High-secure bootloader blob creation
+ */
+
+static int
+gm200_secboot_prepare_hsbl_blob(struct gm200_secboot *gsb)
+{
+	struct nvkm_subdev *subdev = &gsb->base.subdev;
+
+	gsb->hsbl_blob = gm200_secboot_load_firmware(subdev, "acr/bl", 0);
+	if (IS_ERR(gsb->hsbl_blob)) {
+		int ret = PTR_ERR(gsb->hsbl_blob);
+
+		gsb->hsbl_blob = NULL;
+		return ret;
+	}
+
+	return 0;
+}
+
+/**
+ * gm20x_secboot_prepare_blobs - load blobs common to all GM20X GPUs.
+ *
+ * This includes the LS blob, HS ucode loading blob, and HS bootloader.
+ *
+ * The HS ucode unload blob is only used on dGPU.
+ */
+int
+gm20x_secboot_prepare_blobs(struct gm200_secboot *gsb)
+{
+	int ret;
+
+	/* Load and prepare the managed falcon's firmwares */
+	ret = gm200_secboot_prepare_ls_blob(gsb);
+	if (ret)
+		return ret;
+
+	/* Load the HS firmware that will load the LS firmwares */
+	ret = gm200_secboot_prepare_hs_blob(gsb, "acr/ucode_load",
+					    &gsb->acr_load_blob,
+					    &gsb->acr_load_bl_desc, true);
+	if (ret)
+		return ret;
+
+	/* Load the HS firmware bootloader */
+	ret = gm200_secboot_prepare_hsbl_blob(gsb);
+	if (ret)
+		return ret;
+
+	return 0;
+}
+
+static int
+gm200_secboot_prepare_blobs(struct nvkm_secboot *sb)
+{
+	struct gm200_secboot *gsb = gm200_secboot(sb);
+	int ret;
+
+	ret = gm20x_secboot_prepare_blobs(gsb);
+	if (ret)
+		return ret;
+
+	/* dGPU only: load the HS firmware that unprotects the WPR region */
+	ret = gm200_secboot_prepare_hs_blob(gsb, "acr/ucode_unload",
+					    &gsb->acr_unload_blob,
+					    &gsb->acr_unload_bl_desc, false);
+	if (ret)
+		return ret;
+
+	return 0;
+}
+
+
+
+/*
+ * Secure Boot Execution
+ */
+
+/**
+ * gm200_secboot_load_hs_bl() - load HS bootloader into DMEM and IMEM
+ */
+static void
+gm200_secboot_load_hs_bl(struct gm200_secboot *gsb, void *data, u32 data_size)
+{
+	struct nvkm_device *device = gsb->base.subdev.device;
+	struct fw_bin_header *hdr = gsb->hsbl_blob;
+	struct fw_bl_desc *hsbl_desc = gsb->hsbl_blob + hdr->header_offset;
+	void *blob_data = gsb->hsbl_blob + hdr->data_offset;
+	void *hsbl_code = blob_data + hsbl_desc->code_off;
+	void *hsbl_data = blob_data + hsbl_desc->data_off;
+	u32 code_size = ALIGN(hsbl_desc->code_size, 256);
+	const u32 base = gsb->base.base;
+	u32 blk;
+	u32 tag;
+	int i;
+
+	/*
+	 * Copy HS bootloader data
+	 */
+	nvkm_wr32(device, base + 0x1c0, (0x00000000 | (0x1 << 24)));
+	for (i = 0; i < hsbl_desc->data_size / 4; i++)
+		nvkm_wr32(device, base + 0x1c4, ((u32 *)hsbl_data)[i]);
+
+	/*
+	 * Copy HS bootloader interface structure where the HS descriptor
+	 * expects it to be
+	 */
+	nvkm_wr32(device, base + 0x1c0,
+		  (hsbl_desc->dmem_load_off | (0x1 << 24)));
+	for (i = 0; i < data_size / 4; i++)
+		nvkm_wr32(device, base + 0x1c4, ((u32 *)data)[i]);
+
+	/* Copy HS bootloader code to end of IMEM */
+	blk = (nvkm_rd32(device, base + 0x108) & 0x1ff) - (code_size >> 8);
+	tag = hsbl_desc->start_tag;
+	nvkm_wr32(device, base + 0x180, ((blk & 0xff) << 8) | (0x1 << 24));
+	for (i = 0; i < code_size / 4; i++) {
+		/* write new tag every 256B */
+		if ((i & 0x3f) == 0) {
+			nvkm_wr32(device, base + 0x188, tag & 0xffff);
+			tag++;
+		}
+		nvkm_wr32(device, base + 0x184, ((u32 *)hsbl_code)[i]);
+	}
+	nvkm_wr32(device, base + 0x188, 0);
+}
+
+/**
+ * gm200_secboot_setup_falcon() - set up the secure falcon for secure boot
+ */
+static int
+gm200_secboot_setup_falcon(struct gm200_secboot *gsb)
+{
+	struct nvkm_device *device = gsb->base.subdev.device;
+	struct fw_bin_header *hdr = gsb->hsbl_blob;
+	struct fw_bl_desc *hsbl_desc = gsb->hsbl_blob + hdr->header_offset;
+	/* virtual start address for boot vector */
+	u32 virt_addr = hsbl_desc->start_tag << 8;
+	const u32 base = gsb->base.base;
+	const u32 reg_base = base + 0xe00;
+	u32 inst_loc;
+	int ret;
+
+	ret = nvkm_secboot_falcon_reset(&gsb->base);
+	if (ret)
+		return ret;
+
+	/* setup apertures - virtual */
+	nvkm_wr32(device, reg_base + 4 * (FALCON_DMAIDX_UCODE), 0x4);
+	nvkm_wr32(device, reg_base + 4 * (FALCON_DMAIDX_VIRT), 0x0);
+	/* setup apertures - physical */
+	nvkm_wr32(device, reg_base + 4 * (FALCON_DMAIDX_PHYS_VID), 0x4);
+	nvkm_wr32(device, reg_base + 4 * (FALCON_DMAIDX_PHYS_SYS_COH),
+		  0x4 | 0x1);
+	nvkm_wr32(device, reg_base + 4 * (FALCON_DMAIDX_PHYS_SYS_NCOH),
+		  0x4 | 0x2);
+
+	/* Set context */
+	if (nvkm_memory_target(gsb->inst->memory) == NVKM_MEM_TARGET_VRAM)
+		inst_loc = 0x0; /* FB */
+	else
+		inst_loc = 0x3; /* Non-coherent sysmem */
+
+	nvkm_mask(device, base + 0x048, 0x1, 0x1);
+	nvkm_wr32(device, base + 0x480,
+		  ((gsb->inst->addr >> 12) & 0xfffffff) |
+		  (inst_loc << 28) | (1 << 30));
+
+	/* Set boot vector to code's starting virtual address */
+	nvkm_wr32(device, base + 0x104, virt_addr);
+
+	return 0;
+}
+
+/**
+ * gm200_secboot_run_hs_blob() - run the given high-secure blob
+ */
+static int
+gm200_secboot_run_hs_blob(struct gm200_secboot *gsb, struct nvkm_gpuobj *blob,
+			  struct gm200_flcn_bl_desc *desc)
+{
+	struct nvkm_vma vma;
+	u64 vma_addr;
+	const u32 bl_desc_size = gsb->func->bl_desc_size;
+	u8 bl_desc[bl_desc_size];
+	int ret;
+
+	/* Map the HS firmware so the HS bootloader can see it */
+	ret = nvkm_gpuobj_map(blob, gsb->vm, NV_MEM_ACCESS_RW, &vma);
+	if (ret)
+		return ret;
+
+	/* Add the mapping address to the DMA bases */
+	vma_addr = flcn64_to_u64(desc->code_dma_base) + vma.offset;
+	desc->code_dma_base.lo = lower_32_bits(vma_addr);
+	desc->code_dma_base.hi = upper_32_bits(vma_addr);
+	vma_addr = flcn64_to_u64(desc->data_dma_base) + vma.offset;
+	desc->data_dma_base.lo = lower_32_bits(vma_addr);
+	desc->data_dma_base.hi = upper_32_bits(vma_addr);
+
+	/* Fixup the BL header */
+	gsb->func->fixup_bl_desc(desc, &bl_desc);
+
+	/* Reset the falcon and make it ready to run the HS bootloader */
+	ret = gm200_secboot_setup_falcon(gsb);
+	if (ret)
+		goto done;
+
+	/* Load the HS bootloader into the falcon's IMEM/DMEM */
+	gm200_secboot_load_hs_bl(gsb, &bl_desc, bl_desc_size);
+
+	/* Start the HS bootloader */
+	ret = nvkm_secboot_falcon_run(&gsb->base);
+	if (ret)
+		goto done;
+
+done:
+	/* Restore the original DMA addresses */
+	vma_addr = flcn64_to_u64(desc->code_dma_base) - vma.offset;
+	desc->code_dma_base.lo = lower_32_bits(vma_addr);
+	desc->code_dma_base.hi = upper_32_bits(vma_addr);
+	vma_addr = flcn64_to_u64(desc->data_dma_base) - vma.offset;
+	desc->data_dma_base.lo = lower_32_bits(vma_addr);
+	desc->data_dma_base.hi = upper_32_bits(vma_addr);
+
+	/* We don't need the ACR firmware anymore */
+	nvkm_gpuobj_unmap(&vma);
+
+	return ret;
+}
+
+/*
+ * gm200_secboot_reset() - execute secure boot from the prepared state
+ *
+ * Load the HS bootloader and ask the falcon to run it. This will in turn
+ * load the HS firmware and run it, so once the falcon stops all the managed
+ * falcons should have their LS firmware loaded and be ready to run.
+ */
+int
+gm200_secboot_reset(struct nvkm_secboot *sb, enum nvkm_secboot_falcon falcon)
+{
+	struct gm200_secboot *gsb = gm200_secboot(sb);
+	int ret;
+
+	/*
+	 * Dummy GM200 implementation: perform secure boot each time we are
+	 * called on FECS. Since only FECS and GPCCS are managed and started
+	 * together, this ought to be safe.
+	 *
+	 * Once we have proper PMU firmware and support, this will be changed
+	 * to a proper call to the PMU method.
+	 */
+	if (falcon != NVKM_SECBOOT_FALCON_FECS)
+		goto end;
+
+	/* If WPR is set and we have an unload blob, run it to unlock WPR */
+	if (gsb->acr_unload_blob &&
+	    gsb->falcon_state[NVKM_SECBOOT_FALCON_FECS] != NON_SECURE) {
+		ret = gm200_secboot_run_hs_blob(gsb, gsb->acr_unload_blob,
+						&gsb->acr_unload_bl_desc);
+		if (ret)
+			return ret;
+	}
+
+	/* Reload all managed falcons */
+	ret = gm200_secboot_run_hs_blob(gsb, gsb->acr_load_blob,
+					&gsb->acr_load_bl_desc);
+	if (ret)
+		return ret;
+
+end:
+	gsb->falcon_state[falcon] = RESET;
+	return 0;
+}
+
+int
+gm200_secboot_start(struct nvkm_secboot *sb, enum nvkm_secboot_falcon falcon)
+{
+	struct gm200_secboot *gsb = gm200_secboot(sb);
+	int base;
+
+	switch (falcon) {
+	case NVKM_SECBOOT_FALCON_FECS:
+		base = 0x409000;
+		break;
+	case NVKM_SECBOOT_FALCON_GPCCS:
+		base = 0x41a000;
+		break;
+	default:
+		nvkm_error(&sb->subdev, "cannot start unhandled falcon!\n");
+		return -EINVAL;
+	}
+
+	nvkm_wr32(sb->subdev.device, base + 0x130, 0x00000002);
+	gsb->falcon_state[falcon] = RUNNING;
+
+	return 0;
+}
+
+
+
+int
+gm200_secboot_init(struct nvkm_secboot *sb)
+{
+	struct gm200_secboot *gsb = gm200_secboot(sb);
+	struct nvkm_device *device = sb->subdev.device;
+	struct nvkm_vm *vm;
+	const u64 vm_area_len = 600 * 1024;
+	int ret;
+
+	/* Allocate instance block and VM */
+	ret = nvkm_gpuobj_new(device, 0x1000, 0, true, NULL, &gsb->inst);
+	if (ret)
+		return ret;
+
+	ret = nvkm_gpuobj_new(device, 0x8000, 0, true, NULL, &gsb->pgd);
+	if (ret)
+		return ret;
+
+	ret = nvkm_vm_new(device, 0, vm_area_len, 0, NULL, &vm);
+	if (ret)
+		return ret;
+
+	atomic_inc(&vm->engref[NVKM_SUBDEV_PMU]);
+
+	ret = nvkm_vm_ref(vm, &gsb->vm, gsb->pgd);
+	nvkm_vm_ref(NULL, &vm, NULL);
+	if (ret)
+		return ret;
+
+	nvkm_kmap(gsb->inst);
+	nvkm_wo32(gsb->inst, 0x200, lower_32_bits(gsb->pgd->addr));
+	nvkm_wo32(gsb->inst, 0x204, upper_32_bits(gsb->pgd->addr));
+	nvkm_wo32(gsb->inst, 0x208, lower_32_bits(vm_area_len - 1));
+	nvkm_wo32(gsb->inst, 0x20c, upper_32_bits(vm_area_len - 1));
+	nvkm_done(gsb->inst);
+
+	return 0;
+}
+
+int
+gm200_secboot_fini(struct nvkm_secboot *sb, bool suspend)
+{
+	struct gm200_secboot *gsb = gm200_secboot(sb);
+	int ret = 0;
+	int i;
+
+	/* Run the unload blob to unprotect the WPR region */
+	if (gsb->acr_unload_blob &&
+	    gsb->falcon_state[NVKM_SECBOOT_FALCON_FECS] != NON_SECURE)
+		ret = gm200_secboot_run_hs_blob(gsb, gsb->acr_unload_blob,
+						&gsb->acr_unload_bl_desc);
+
+	for (i = 0; i < NVKM_SECBOOT_FALCON_END; i++)
+		gsb->falcon_state[i] = NON_SECURE;
+
+	return ret;
+}
+
+void *
+gm200_secboot_dtor(struct nvkm_secboot *sb)
+{
+	struct gm200_secboot *gsb = gm200_secboot(sb);
+
+	nvkm_gpuobj_del(&gsb->acr_unload_blob);
+
+	kfree(gsb->hsbl_blob);
+	nvkm_gpuobj_del(&gsb->acr_load_blob);
+	nvkm_gpuobj_del(&gsb->ls_blob);
+
+	nvkm_vm_ref(NULL, &gsb->vm, gsb->pgd);
+	nvkm_gpuobj_del(&gsb->pgd);
+	nvkm_gpuobj_del(&gsb->inst);
+
+	return gsb;
+}
+
+
+static const struct nvkm_secboot_func
+gm200_secboot = {
+	.dtor = gm200_secboot_dtor,
+	.init = gm200_secboot_init,
+	.fini = gm200_secboot_fini,
+	.prepare_blobs = gm200_secboot_prepare_blobs,
+	.reset = gm200_secboot_reset,
+	.start = gm200_secboot_start,
+	.managed_falcons = BIT(NVKM_SECBOOT_FALCON_FECS) |
+			   BIT(NVKM_SECBOOT_FALCON_GPCCS),
+	.boot_falcon = NVKM_SECBOOT_FALCON_PMU,
+};
+
+/**
+ * gm200_fixup_bl_desc - just copy the BL descriptor
+ *
+ * Use the GM200 descriptor format by default.
+ */
+static void
+gm200_secboot_fixup_bl_desc(const struct gm200_flcn_bl_desc *desc, void *ret)
+{
+	memcpy(ret, desc, sizeof(*desc));
+}
+
+static void
+gm200_secboot_fixup_hs_desc(struct gm200_secboot *gsb,
+			    struct hsflcn_acr_desc *desc)
+{
+	desc->ucode_blob_base = gsb->ls_blob->addr;
+	desc->ucode_blob_size = gsb->ls_blob->size;
+
+	desc->wpr_offset = 0;
+
+	/* WPR region information for the HS binary to set up */
+	desc->wpr_region_id = 1;
+	desc->regions.no_regions = 1;
+	desc->regions.region_props[0].region_id = 1;
+	desc->regions.region_props[0].start_addr = gsb->wpr_addr >> 8;
+	desc->regions.region_props[0].end_addr =
+		(gsb->wpr_addr + gsb->wpr_size) >> 8;
+}
+
+static const struct gm200_secboot_func
+gm200_secboot_func = {
+	.bl_desc_size = sizeof(struct gm200_flcn_bl_desc),
+	.fixup_bl_desc = gm200_secboot_fixup_bl_desc,
+	.fixup_hs_desc = gm200_secboot_fixup_hs_desc,
+};
+
+int
+gm200_secboot_new(struct nvkm_device *device, int index,
+		  struct nvkm_secboot **psb)
+{
+	int ret;
+	struct gm200_secboot *gsb;
+
+	gsb = kzalloc(sizeof(*gsb), GFP_KERNEL);
+	if (!gsb) {
+		psb = NULL;
+		return -ENOMEM;
+	}
+	*psb = &gsb->base;
+
+	ret = nvkm_secboot_ctor(&gm200_secboot, device, index, &gsb->base);
+	if (ret)
+		return ret;
+
+	gsb->func = &gm200_secboot_func;
+
+	return 0;
+}
+
+MODULE_FIRMWARE("nvidia/gm200/acr/bl.bin");
+MODULE_FIRMWARE("nvidia/gm200/acr/ucode_load.bin");
+MODULE_FIRMWARE("nvidia/gm200/acr/ucode_unload.bin");
+MODULE_FIRMWARE("nvidia/gm200/gr/fecs_bl.bin");
+MODULE_FIRMWARE("nvidia/gm200/gr/fecs_inst.bin");
+MODULE_FIRMWARE("nvidia/gm200/gr/fecs_data.bin");
+MODULE_FIRMWARE("nvidia/gm200/gr/fecs_sig.bin");
+MODULE_FIRMWARE("nvidia/gm200/gr/gpccs_bl.bin");
+MODULE_FIRMWARE("nvidia/gm200/gr/gpccs_inst.bin");
+MODULE_FIRMWARE("nvidia/gm200/gr/gpccs_data.bin");
+MODULE_FIRMWARE("nvidia/gm200/gr/gpccs_sig.bin");
+MODULE_FIRMWARE("nvidia/gm200/gr/sw_ctx.bin");
+MODULE_FIRMWARE("nvidia/gm200/gr/sw_nonctx.bin");
+MODULE_FIRMWARE("nvidia/gm200/gr/sw_bundle_init.bin");
+MODULE_FIRMWARE("nvidia/gm200/gr/sw_method_init.bin");
+
+MODULE_FIRMWARE("nvidia/gm204/acr/bl.bin");
+MODULE_FIRMWARE("nvidia/gm204/acr/ucode_load.bin");
+MODULE_FIRMWARE("nvidia/gm204/acr/ucode_unload.bin");
+MODULE_FIRMWARE("nvidia/gm204/gr/fecs_bl.bin");
+MODULE_FIRMWARE("nvidia/gm204/gr/fecs_inst.bin");
+MODULE_FIRMWARE("nvidia/gm204/gr/fecs_data.bin");
+MODULE_FIRMWARE("nvidia/gm204/gr/fecs_sig.bin");
+MODULE_FIRMWARE("nvidia/gm204/gr/gpccs_bl.bin");
+MODULE_FIRMWARE("nvidia/gm204/gr/gpccs_inst.bin");
+MODULE_FIRMWARE("nvidia/gm204/gr/gpccs_data.bin");
+MODULE_FIRMWARE("nvidia/gm204/gr/gpccs_sig.bin");
+MODULE_FIRMWARE("nvidia/gm204/gr/sw_ctx.bin");
+MODULE_FIRMWARE("nvidia/gm204/gr/sw_nonctx.bin");
+MODULE_FIRMWARE("nvidia/gm204/gr/sw_bundle_init.bin");
+MODULE_FIRMWARE("nvidia/gm204/gr/sw_method_init.bin");
+
+MODULE_FIRMWARE("nvidia/gm206/acr/bl.bin");
+MODULE_FIRMWARE("nvidia/gm206/acr/ucode_load.bin");
+MODULE_FIRMWARE("nvidia/gm206/acr/ucode_unload.bin");
+MODULE_FIRMWARE("nvidia/gm206/gr/fecs_bl.bin");
+MODULE_FIRMWARE("nvidia/gm206/gr/fecs_inst.bin");
+MODULE_FIRMWARE("nvidia/gm206/gr/fecs_data.bin");
+MODULE_FIRMWARE("nvidia/gm206/gr/fecs_sig.bin");
+MODULE_FIRMWARE("nvidia/gm206/gr/gpccs_bl.bin");
+MODULE_FIRMWARE("nvidia/gm206/gr/gpccs_inst.bin");
+MODULE_FIRMWARE("nvidia/gm206/gr/gpccs_data.bin");
+MODULE_FIRMWARE("nvidia/gm206/gr/gpccs_sig.bin");
+MODULE_FIRMWARE("nvidia/gm206/gr/sw_ctx.bin");
+MODULE_FIRMWARE("nvidia/gm206/gr/sw_nonctx.bin");
+MODULE_FIRMWARE("nvidia/gm206/gr/sw_bundle_init.bin");
+MODULE_FIRMWARE("nvidia/gm206/gr/sw_method_init.bin");
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/secboot/gm20b.c b/drivers/gpu/drm/nouveau/nvkm/subdev/secboot/gm20b.c
new file mode 100644
index 000000000000..684320484b70
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/secboot/gm20b.c
@@ -0,0 +1,233 @@
+/*
+ * Copyright (c) 2016, NVIDIA CORPORATION. All rights reserved.
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
+ * THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+ * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
+ * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER
+ * DEALINGS IN THE SOFTWARE.
+ */
+
+#include "priv.h"
+
+#include <core/gpuobj.h>
+
+/*
+ * The BL header format used by GM20B's firmware is slightly different
+ * from the one of GM200. Fix the differences here.
+ */
+struct gm20b_flcn_bl_desc {
+	u32 reserved[4];
+	u32 signature[4];
+	u32 ctx_dma;
+	u32 code_dma_base;
+	u32 non_sec_code_off;
+	u32 non_sec_code_size;
+	u32 sec_code_off;
+	u32 sec_code_size;
+	u32 code_entry_point;
+	u32 data_dma_base;
+	u32 data_size;
+};
+
+/**
+ * gm20b_secboot_fixup_bl_desc - adapt BL descriptor to format used by GM20B FW
+ *
+ * There is only a slight format difference (DMA addresses being 32-bits and
+ * 256B-aligned) to address.
+ */
+static void
+gm20b_secboot_fixup_bl_desc(const struct gm200_flcn_bl_desc *desc, void *ret)
+{
+	struct gm20b_flcn_bl_desc *gdesc = ret;
+	u64 addr;
+
+	memcpy(gdesc->reserved, desc->reserved, sizeof(gdesc->reserved));
+	memcpy(gdesc->signature, desc->signature, sizeof(gdesc->signature));
+	gdesc->ctx_dma = desc->ctx_dma;
+	addr = desc->code_dma_base.hi;
+	addr <<= 32;
+	addr |= desc->code_dma_base.lo;
+	gdesc->code_dma_base = lower_32_bits(addr >> 8);
+	gdesc->non_sec_code_off = desc->non_sec_code_off;
+	gdesc->non_sec_code_size = desc->non_sec_code_size;
+	gdesc->sec_code_off = desc->sec_code_off;
+	gdesc->sec_code_size = desc->sec_code_size;
+	gdesc->code_entry_point = desc->code_entry_point;
+	addr = desc->data_dma_base.hi;
+	addr <<= 32;
+	addr |= desc->data_dma_base.lo;
+	gdesc->data_dma_base = lower_32_bits(addr >> 8);
+	gdesc->data_size = desc->data_size;
+}
+
+static void
+gm20b_secboot_fixup_hs_desc(struct gm200_secboot *gsb,
+			    struct hsflcn_acr_desc *desc)
+{
+	desc->ucode_blob_base = gsb->ls_blob->addr;
+	desc->ucode_blob_size = gsb->ls_blob->size;
+
+	desc->wpr_offset = 0;
+}
+
+static const struct gm200_secboot_func
+gm20b_secboot_func = {
+	.bl_desc_size = sizeof(struct gm20b_flcn_bl_desc),
+	.fixup_bl_desc = gm20b_secboot_fixup_bl_desc,
+	.fixup_hs_desc = gm20b_secboot_fixup_hs_desc,
+};
+
+
+#ifdef CONFIG_ARCH_TEGRA
+#define TEGRA_MC_BASE				0x70019000
+#define MC_SECURITY_CARVEOUT2_CFG0		0xc58
+#define MC_SECURITY_CARVEOUT2_BOM_0		0xc5c
+#define MC_SECURITY_CARVEOUT2_BOM_HI_0		0xc60
+#define MC_SECURITY_CARVEOUT2_SIZE_128K		0xc64
+#define TEGRA_MC_SECURITY_CARVEOUT_CFG_LOCKED	(1 << 1)
+/**
+ * sb_tegra_read_wpr() - read the WPR registers on Tegra
+ *
+ * On dGPU, we can manage the WPR region ourselves, but on Tegra the WPR region
+ * is reserved from system memory by the bootloader and irreversibly locked.
+ * This function reads the address and size of the pre-configured WPR region.
+ */
+static int
+gm20b_tegra_read_wpr(struct gm200_secboot *gsb)
+{
+	struct nvkm_secboot *sb = &gsb->base;
+	void __iomem *mc;
+	u32 cfg;
+
+	mc = ioremap(TEGRA_MC_BASE, 0xd00);
+	if (!mc) {
+		nvkm_error(&sb->subdev, "Cannot map Tegra MC registers\n");
+		return PTR_ERR(mc);
+	}
+	gsb->wpr_addr = ioread32_native(mc + MC_SECURITY_CARVEOUT2_BOM_0) |
+	      ((u64)ioread32_native(mc + MC_SECURITY_CARVEOUT2_BOM_HI_0) << 32);
+	gsb->wpr_size = ioread32_native(mc + MC_SECURITY_CARVEOUT2_SIZE_128K)
+		<< 17;
+	cfg = ioread32_native(mc + MC_SECURITY_CARVEOUT2_CFG0);
+	iounmap(mc);
+
+	/* Check that WPR settings are valid */
+	if (gsb->wpr_size == 0) {
+		nvkm_error(&sb->subdev, "WPR region is empty\n");
+		return -EINVAL;
+	}
+
+	if (!(cfg & TEGRA_MC_SECURITY_CARVEOUT_CFG_LOCKED)) {
+		nvkm_error(&sb->subdev, "WPR region not locked\n");
+		return -EINVAL;
+	}
+
+	return 0;
+}
+#else
+static int
+gm20b_tegra_read_wpr(struct gm200_secboot *gsb)
+{
+	nvkm_error(&gsb->base.subdev, "Tegra support not compiled in\n");
+	return -EINVAL;
+}
+#endif
+
+static int
+gm20b_secboot_prepare_blobs(struct nvkm_secboot *sb)
+{
+	struct gm200_secboot *gsb = gm200_secboot(sb);
+	int acr_size;
+	int ret;
+
+	ret = gm20x_secboot_prepare_blobs(gsb);
+	if (ret)
+		return ret;
+
+	acr_size = gsb->acr_load_blob->size;
+	/*
+	 * On Tegra the WPR region is set by the bootloader. It is illegal for
+	 * the HS blob to be larger than this region.
+	 */
+	if (acr_size > gsb->wpr_size) {
+		nvkm_error(&sb->subdev, "WPR region too small for FW blob!\n");
+		nvkm_error(&sb->subdev, "required: %dB\n", acr_size);
+		nvkm_error(&sb->subdev, "WPR size: %dB\n", gsb->wpr_size);
+		return -ENOSPC;
+	}
+
+	return 0;
+}
+
+static int
+gm20b_secboot_init(struct nvkm_secboot *sb)
+{
+	struct gm200_secboot *gsb = gm200_secboot(sb);
+	int ret;
+
+	ret = gm20b_tegra_read_wpr(gsb);
+	if (ret)
+		return ret;
+
+	return gm200_secboot_init(sb);
+}
+
+static const struct nvkm_secboot_func
+gm20b_secboot = {
+	.dtor = gm200_secboot_dtor,
+	.init = gm20b_secboot_init,
+	.prepare_blobs = gm20b_secboot_prepare_blobs,
+	.reset = gm200_secboot_reset,
+	.start = gm200_secboot_start,
+	.managed_falcons = BIT(NVKM_SECBOOT_FALCON_FECS),
+	.boot_falcon = NVKM_SECBOOT_FALCON_PMU,
+};
+
+int
+gm20b_secboot_new(struct nvkm_device *device, int index,
+		  struct nvkm_secboot **psb)
+{
+	int ret;
+	struct gm200_secboot *gsb;
+
+	gsb = kzalloc(sizeof(*gsb), GFP_KERNEL);
+	if (!gsb) {
+		psb = NULL;
+		return -ENOMEM;
+	}
+	*psb = &gsb->base;
+
+	ret = nvkm_secboot_ctor(&gm20b_secboot, device, index, &gsb->base);
+	if (ret)
+		return ret;
+
+	gsb->func = &gm20b_secboot_func;
+
+	return 0;
+}
+
+MODULE_FIRMWARE("nvidia/gm20b/acr/bl.bin");
+MODULE_FIRMWARE("nvidia/gm20b/acr/ucode_load.bin");
+MODULE_FIRMWARE("nvidia/gm20b/gr/fecs_bl.bin");
+MODULE_FIRMWARE("nvidia/gm20b/gr/fecs_inst.bin");
+MODULE_FIRMWARE("nvidia/gm20b/gr/fecs_data.bin");
+MODULE_FIRMWARE("nvidia/gm20b/gr/fecs_sig.bin");
+MODULE_FIRMWARE("nvidia/gm20b/gr/gpccs_inst.bin");
+MODULE_FIRMWARE("nvidia/gm20b/gr/gpccs_data.bin");
+MODULE_FIRMWARE("nvidia/gm20b/gr/sw_ctx.bin");
+MODULE_FIRMWARE("nvidia/gm20b/gr/sw_nonctx.bin");
+MODULE_FIRMWARE("nvidia/gm20b/gr/sw_bundle_init.bin");
+MODULE_FIRMWARE("nvidia/gm20b/gr/sw_method_init.bin");
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/secboot/priv.h b/drivers/gpu/drm/nouveau/nvkm/subdev/secboot/priv.h
new file mode 100644
index 000000000000..f2b09dee7c5d
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/secboot/priv.h
@@ -0,0 +1,226 @@
+/*
+ * Copyright (c) 2015, NVIDIA CORPORATION. All rights reserved.
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
+ * THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+ * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
+ * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER
+ * DEALINGS IN THE SOFTWARE.
+ */
+
+#ifndef __NVKM_SECBOOT_PRIV_H__
+#define __NVKM_SECBOOT_PRIV_H__
+
+#include <subdev/secboot.h>
+#include <subdev/mmu.h>
+
+struct nvkm_secboot_func {
+	int (*init)(struct nvkm_secboot *);
+	int (*fini)(struct nvkm_secboot *, bool suspend);
+	void *(*dtor)(struct nvkm_secboot *);
+	int (*prepare_blobs)(struct nvkm_secboot *);
+	int (*reset)(struct nvkm_secboot *, enum nvkm_secboot_falcon);
+	int (*start)(struct nvkm_secboot *, enum nvkm_secboot_falcon);
+
+	/* ID of the falcon that will perform secure boot */
+	enum nvkm_secboot_falcon boot_falcon;
+	/* Bit-mask of IDs of managed falcons */
+	unsigned long managed_falcons;
+};
+
+int nvkm_secboot_ctor(const struct nvkm_secboot_func *, struct nvkm_device *,
+		      int index, struct nvkm_secboot *);
+int nvkm_secboot_falcon_reset(struct nvkm_secboot *);
+int nvkm_secboot_falcon_run(struct nvkm_secboot *);
+
+struct flcn_u64 {
+	u32 lo;
+	u32 hi;
+};
+static inline u64 flcn64_to_u64(const struct flcn_u64 f)
+{
+	return ((u64)f.hi) << 32 | f.lo;
+}
+
+/**
+ * struct gm200_flcn_bl_desc - DMEM bootloader descriptor
+ * @signature:		16B signature for secure code. 0s if no secure code
+ * @ctx_dma:		DMA context to be used by BL while loading code/data
+ * @code_dma_base:	256B-aligned Physical FB Address where code is located
+ *			(falcon's $xcbase register)
+ * @non_sec_code_off:	offset from code_dma_base where the non-secure code is
+ *                      located. The offset must be multiple of 256 to help perf
+ * @non_sec_code_size:	the size of the nonSecure code part.
+ * @sec_code_off:	offset from code_dma_base where the secure code is
+ *                      located. The offset must be multiple of 256 to help perf
+ * @sec_code_size:	offset from code_dma_base where the secure code is
+ *                      located. The offset must be multiple of 256 to help perf
+ * @code_entry_point:	code entry point which will be invoked by BL after
+ *                      code is loaded.
+ * @data_dma_base:	256B aligned Physical FB Address where data is located.
+ *			(falcon's $xdbase register)
+ * @data_size:		size of data block. Should be multiple of 256B
+ *
+ * Structure used by the bootloader to load the rest of the code. This has
+ * to be filled by host and copied into DMEM at offset provided in the
+ * hsflcn_bl_desc.bl_desc_dmem_load_off.
+ */
+struct gm200_flcn_bl_desc {
+	u32 reserved[4];
+	u32 signature[4];
+	u32 ctx_dma;
+	struct flcn_u64 code_dma_base;
+	u32 non_sec_code_off;
+	u32 non_sec_code_size;
+	u32 sec_code_off;
+	u32 sec_code_size;
+	u32 code_entry_point;
+	struct flcn_u64 data_dma_base;
+	u32 data_size;
+};
+
+/**
+ * struct hsflcn_acr_desc - data section of the HS firmware
+ *
+ * This header is to be copied at the beginning of DMEM by the HS bootloader.
+ *
+ * @signature:		signature of ACR ucode
+ * @wpr_region_id:	region ID holding the WPR header and its details
+ * @wpr_offset:		offset from the WPR region holding the wpr header
+ * @regions:		region descriptors
+ * @nonwpr_ucode_blob_size:	size of LS blob
+ * @nonwpr_ucode_blob_start:	FB location of LS blob is
+ */
+struct hsflcn_acr_desc {
+	union {
+		u8 reserved_dmem[0x200];
+		u32 signatures[4];
+	} ucode_reserved_space;
+	u32 wpr_region_id;
+	u32 wpr_offset;
+	u32 mmu_mem_range;
+#define FLCN_ACR_MAX_REGIONS 2
+	struct {
+		u32 no_regions;
+		struct {
+			u32 start_addr;
+			u32 end_addr;
+			u32 region_id;
+			u32 read_mask;
+			u32 write_mask;
+			u32 client_mask;
+		} region_props[FLCN_ACR_MAX_REGIONS];
+	} regions;
+	u32 ucode_blob_size;
+	u64 ucode_blob_base __aligned(8);
+	struct {
+		u32 vpr_enabled;
+		u32 vpr_start;
+		u32 vpr_end;
+		u32 hdcp_policies;
+	} vpr_desc;
+};
+
+/**
+ * Contains the whole secure boot state, allowing it to be performed as needed
+ * @wpr_addr:		physical address of the WPR region
+ * @wpr_size:		size in bytes of the WPR region
+ * @ls_blob:		LS blob of all the LS firmwares, signatures, bootloaders
+ * @ls_blob_size:	size of the LS blob
+ * @ls_blob_nb_regions:	number of LS firmwares that will be loaded
+ * @acr_blob:		HS blob
+ * @acr_blob_vma:	mapping of the HS blob into the secure falcon's VM
+ * @acr_bl_desc:	bootloader descriptor of the HS blob
+ * @hsbl_blob:		HS blob bootloader
+ * @inst:		instance block for HS falcon
+ * @pgd:		page directory for the HS falcon
+ * @vm:			address space used by the HS falcon
+ * @bl_desc_size:	size of the BL descriptor used by this chip.
+ * @fixup_bl_desc:	hook that generates the proper BL descriptor format from
+ *			the generic GM200 format into a data array of size
+ *			bl_desc_size
+ */
+struct gm200_secboot {
+	struct nvkm_secboot base;
+	const struct gm200_secboot_func *func;
+
+	/*
+	 * Address and size of the WPR region. On dGPU this will be the
+	 * address of the LS blob. On Tegra this is a fixed region set by the
+	 * bootloader
+	 */
+	u64 wpr_addr;
+	u32 wpr_size;
+
+	/*
+	 * HS FW - lock WPR region (dGPU only) and load LS FWs
+	 * on Tegra the HS FW copies the LS blob into the fixed WPR instead
+	 */
+	struct nvkm_gpuobj *acr_load_blob;
+	struct gm200_flcn_bl_desc acr_load_bl_desc;
+
+	/* HS FW - unlock WPR region (dGPU only) */
+	struct nvkm_gpuobj *acr_unload_blob;
+	struct gm200_flcn_bl_desc acr_unload_bl_desc;
+
+	/* HS bootloader */
+	void *hsbl_blob;
+
+	/* LS FWs, to be loaded by the HS ACR */
+	struct nvkm_gpuobj *ls_blob;
+
+	/* Instance block & address space used for HS FW execution */
+	struct nvkm_gpuobj *inst;
+	struct nvkm_gpuobj *pgd;
+	struct nvkm_vm *vm;
+
+	/* To keep track of the state of all managed falcons */
+	enum {
+		/* In non-secure state, no firmware loaded, no privileges*/
+		NON_SECURE = 0,
+		/* In low-secure mode and ready to be started */
+		RESET,
+		/* In low-secure mode and running */
+		RUNNING,
+	} falcon_state[NVKM_SECBOOT_FALCON_END];
+
+};
+#define gm200_secboot(sb) container_of(sb, struct gm200_secboot, base)
+
+struct gm200_secboot_func {
+	/*
+	 * Size of the bootloader descriptor for this chip. A block of this
+	 * size is allocated before booting a falcon and the fixup_bl_desc
+	 * callback is called on it
+	 */
+	u32 bl_desc_size;
+	void (*fixup_bl_desc)(const struct gm200_flcn_bl_desc *, void *);
+
+	/*
+	 * Chip-specific modifications of the HS descriptor can be done here.
+	 * On dGPU this is used to fill the information about the WPR region
+	 * we want the HS FW to set up.
+	 */
+	void (*fixup_hs_desc)(struct gm200_secboot *, struct hsflcn_acr_desc *);
+};
+
+int gm200_secboot_init(struct nvkm_secboot *);
+void *gm200_secboot_dtor(struct nvkm_secboot *);
+int gm200_secboot_reset(struct nvkm_secboot *, u32);
+int gm200_secboot_start(struct nvkm_secboot *, u32);
+
+int gm20x_secboot_prepare_blobs(struct gm200_secboot *);
+
+#endif
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/volt/Kbuild b/drivers/gpu/drm/nouveau/nvkm/subdev/volt/Kbuild
index b035c6e28be8..c34076223b7b 100644
--- a/drivers/gpu/drm/nouveau/nvkm/subdev/volt/Kbuild
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/volt/Kbuild
@@ -3,3 +3,4 @@ nvkm-y += nvkm/subdev/volt/gpio.o
 nvkm-y += nvkm/subdev/volt/nv40.o
 nvkm-y += nvkm/subdev/volt/gk104.o
 nvkm-y += nvkm/subdev/volt/gk20a.o
+nvkm-y += nvkm/subdev/volt/gm20b.o
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/volt/gk20a.c b/drivers/gpu/drm/nouveau/nvkm/subdev/volt/gk20a.c
index fd56c6476064..d554455326da 100644
--- a/drivers/gpu/drm/nouveau/nvkm/subdev/volt/gk20a.c
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/volt/gk20a.c
@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2014, NVIDIA CORPORATION. All rights reserved.
+ * Copyright (c) 2014-2016, NVIDIA CORPORATION. All rights reserved.
  *
  * Permission is hereby granted, free of charge, to any person obtaining a
  * copy of this software and associated documentation files (the "Software"),
@@ -24,21 +24,9 @@
 
 #include <core/tegra.h>
 
-struct cvb_coef {
-	int c0;
-	int c1;
-	int c2;
-	int c3;
-	int c4;
-	int c5;
-};
-
-struct gk20a_volt {
-	struct nvkm_volt base;
-	struct regulator *vdd;
-};
+#include "gk20a.h"
 
-const struct cvb_coef gk20a_cvb_coef[] = {
+static const struct cvb_coef gk20a_cvb_coef[] = {
 	/* MHz,        c0,     c1,   c2,    c3,     c4,   c5 */
 	/*  72 */ { 1209886, -36468,  515,   417, -13123,  203},
 	/* 108 */ { 1130804, -27659,  296,   298, -10834,  221},
@@ -89,7 +77,7 @@ gk20a_volt_get_cvb_t_voltage(int speedo, int temp, int s_scale, int t_scale,
 	return mv;
 }
 
-static int
+int
 gk20a_volt_calc_voltage(const struct cvb_coef *coef, int speedo)
 {
 	int mv;
@@ -100,7 +88,7 @@ gk20a_volt_calc_voltage(const struct cvb_coef *coef, int speedo)
 	return mv * 1000;
 }
 
-static int
+int
 gk20a_volt_vid_get(struct nvkm_volt *base)
 {
 	struct gk20a_volt *volt = gk20a_volt(base);
@@ -115,7 +103,7 @@ gk20a_volt_vid_get(struct nvkm_volt *base)
 	return -EINVAL;
 }
 
-static int
+int
 gk20a_volt_vid_set(struct nvkm_volt *base, u8 vid)
 {
 	struct gk20a_volt *volt = gk20a_volt(base);
@@ -125,7 +113,7 @@ gk20a_volt_vid_set(struct nvkm_volt *base, u8 vid)
 	return regulator_set_voltage(volt->vdd, volt->base.vid[vid].uv, 1200000);
 }
 
-static int
+int
 gk20a_volt_set_id(struct nvkm_volt *base, u8 id, int condition)
 {
 	struct gk20a_volt *volt = gk20a_volt(base);
@@ -155,30 +143,25 @@ gk20a_volt = {
 };
 
 int
-gk20a_volt_new(struct nvkm_device *device, int index, struct nvkm_volt **pvolt)
+_gk20a_volt_ctor(struct nvkm_device *device, int index,
+		 const struct cvb_coef *coefs, int nb_coefs,
+		 struct gk20a_volt *volt)
 {
 	struct nvkm_device_tegra *tdev = device->func->tegra(device);
-	struct gk20a_volt *volt;
 	int i, uv;
 
-	if (!(volt = kzalloc(sizeof(*volt), GFP_KERNEL)))
-		return -ENOMEM;
-
 	nvkm_volt_ctor(&gk20a_volt, device, index, &volt->base);
-	*pvolt = &volt->base;
 
 	uv = regulator_get_voltage(tdev->vdd);
-	nvkm_info(&volt->base.subdev, "The default voltage is %duV\n", uv);
+	nvkm_debug(&volt->base.subdev, "the default voltage is %duV\n", uv);
 
 	volt->vdd = tdev->vdd;
 
-	volt->base.vid_nr = ARRAY_SIZE(gk20a_cvb_coef);
-	nvkm_debug(&volt->base.subdev, "%s - vid_nr = %d\n", __func__,
-		   volt->base.vid_nr);
+	volt->base.vid_nr = nb_coefs;
 	for (i = 0; i < volt->base.vid_nr; i++) {
 		volt->base.vid[i].vid = i;
 		volt->base.vid[i].uv =
-			gk20a_volt_calc_voltage(&gk20a_cvb_coef[i],
+			gk20a_volt_calc_voltage(&coefs[i],
 						tdev->gpu_speedo);
 		nvkm_debug(&volt->base.subdev, "%2d: vid=%d, uv=%d\n", i,
 			   volt->base.vid[i].vid, volt->base.vid[i].uv);
@@ -186,3 +169,17 @@ gk20a_volt_new(struct nvkm_device *device, int index, struct nvkm_volt **pvolt)
 
 	return 0;
 }
+
+int
+gk20a_volt_new(struct nvkm_device *device, int index, struct nvkm_volt **pvolt)
+{
+	struct gk20a_volt *volt;
+
+	volt = kzalloc(sizeof(*volt), GFP_KERNEL);
+	if (!volt)
+		return -ENOMEM;
+	*pvolt = &volt->base;
+
+	return _gk20a_volt_ctor(device, index, gk20a_cvb_coef,
+				ARRAY_SIZE(gk20a_cvb_coef), volt);
+}
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/volt/gk20a.h b/drivers/gpu/drm/nouveau/nvkm/subdev/volt/gk20a.h
new file mode 100644
index 000000000000..0fa3b502bcf8
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/volt/gk20a.h
@@ -0,0 +1,49 @@
+/*
+ * Copyright (c) 2016, NVIDIA CORPORATION. All rights reserved.
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
+ * THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+ * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
+ * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER
+ * DEALINGS IN THE SOFTWARE.
+ */
+
+#ifndef __GK20A_VOLT_H__
+#define __GK20A_VOLT_H__
+
+struct cvb_coef {
+	int c0;
+	int c1;
+	int c2;
+	int c3;
+	int c4;
+	int c5;
+};
+
+struct gk20a_volt {
+	struct nvkm_volt base;
+	struct regulator *vdd;
+};
+
+int _gk20a_volt_ctor(struct nvkm_device *device, int index,
+		     const struct cvb_coef *coefs, int nb_coefs,
+		     struct gk20a_volt *volt);
+
+int gk20a_volt_calc_voltage(const struct cvb_coef *coef, int speedo);
+int gk20a_volt_vid_get(struct nvkm_volt *volt);
+int gk20a_volt_vid_set(struct nvkm_volt *volt, u8 vid);
+int gk20a_volt_set_id(struct nvkm_volt *volt, u8 id, int condition);
+
+#endif
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/volt/gm20b.c b/drivers/gpu/drm/nouveau/nvkm/subdev/volt/gm20b.c
new file mode 100644
index 000000000000..49b5ecb701e4
--- /dev/null
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/volt/gm20b.c
@@ -0,0 +1,56 @@
+/*
+ * Copyright (c) 2016, NVIDIA CORPORATION. All rights reserved.
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.  IN NO EVENT SHALL
+ * THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+ * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
+ * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER
+ * DEALINGS IN THE SOFTWARE.
+ */
+
+#include "priv.h"
+#include "gk20a.h"
+
+#include <core/tegra.h>
+
+const struct cvb_coef gm20b_cvb_coef[] = {
+	/* KHz,             c0,      c1,   c2 */
+	/*  76800 */ { 1786666,  -85625, 1632 },
+	/* 153600 */ { 1846729,  -87525, 1632 },
+	/* 230400 */ { 1910480,  -89425, 1632 },
+	/* 307200 */ { 1977920,  -91325, 1632 },
+	/* 384000 */ { 2049049,  -93215, 1632 },
+	/* 460800 */ { 2122872,  -95095, 1632 },
+	/* 537600 */ { 2201331,  -96985, 1632 },
+	/* 614400 */ { 2283479,  -98885, 1632 },
+	/* 691200 */ { 2369315, -100785, 1632 },
+	/* 768000 */ { 2458841, -102685, 1632 },
+	/* 844800 */ { 2550821, -104555, 1632 },
+	/* 921600 */ { 2647676, -106455, 1632 },
+};
+
+int
+gm20b_volt_new(struct nvkm_device *device, int index, struct nvkm_volt **pvolt)
+{
+	struct gk20a_volt *volt;
+
+	volt = kzalloc(sizeof(*volt), GFP_KERNEL);
+	if (!volt)
+		return -ENOMEM;
+	*pvolt = &volt->base;
+
+	return _gk20a_volt_ctor(device, index, gm20b_cvb_coef,
+				ARRAY_SIZE(gm20b_cvb_coef), volt);
+}
diff --git a/drivers/gpu/drm/omapdrm/Kconfig b/drivers/gpu/drm/omapdrm/Kconfig
index 336ad4de9981..73241c4eb7aa 100644
--- a/drivers/gpu/drm/omapdrm/Kconfig
+++ b/drivers/gpu/drm/omapdrm/Kconfig
@@ -2,7 +2,6 @@ config DRM_OMAP
 	tristate "OMAP DRM"
 	depends on DRM
 	depends on ARCH_OMAP2PLUS || ARCH_MULTIPLATFORM
-	select OMAP2_DSS
 	select DRM_KMS_HELPER
 	select DRM_KMS_FB_HELPER
 	select FB_SYS_FILLRECT
diff --git a/drivers/gpu/drm/omapdrm/Makefile b/drivers/gpu/drm/omapdrm/Makefile
index fe4c2228bc18..48b7b750c05c 100644
--- a/drivers/gpu/drm/omapdrm/Makefile
+++ b/drivers/gpu/drm/omapdrm/Makefile
@@ -6,7 +6,7 @@
 obj-y += dss/
 obj-y += displays/
 
-ccflags-y := -Iinclude/drm -Werror
+ccflags-y := -Iinclude/drm
 omapdrm-y := omap_drv.o \
 	omap_irq.o \
 	omap_debugfs.o \
diff --git a/drivers/gpu/drm/omapdrm/displays/connector-dvi.c b/drivers/gpu/drm/omapdrm/displays/connector-dvi.c
index d811e6dcaef7..747f26a55e43 100644
--- a/drivers/gpu/drm/omapdrm/displays/connector-dvi.c
+++ b/drivers/gpu/drm/omapdrm/displays/connector-dvi.c
@@ -236,46 +236,6 @@ static struct omap_dss_driver dvic_driver = {
 	.detect		= dvic_detect,
 };
 
-static int dvic_probe_pdata(struct platform_device *pdev)
-{
-	struct panel_drv_data *ddata = platform_get_drvdata(pdev);
-	struct connector_dvi_platform_data *pdata;
-	struct omap_dss_device *in, *dssdev;
-	int i2c_bus_num;
-
-	pdata = dev_get_platdata(&pdev->dev);
-	i2c_bus_num = pdata->i2c_bus_num;
-
-	if (i2c_bus_num != -1) {
-		struct i2c_adapter *adapter;
-
-		adapter = i2c_get_adapter(i2c_bus_num);
-		if (!adapter) {
-			dev_err(&pdev->dev,
-					"Failed to get I2C adapter, bus %d\n",
-					i2c_bus_num);
-			return -EPROBE_DEFER;
-		}
-
-		ddata->i2c_adapter = adapter;
-	}
-
-	in = omap_dss_find_output(pdata->source);
-	if (in == NULL) {
-		i2c_put_adapter(ddata->i2c_adapter);
-
-		dev_err(&pdev->dev, "Failed to find video source\n");
-		return -EPROBE_DEFER;
-	}
-
-	ddata->in = in;
-
-	dssdev = &ddata->dssdev;
-	dssdev->name = pdata->name;
-
-	return 0;
-}
-
 static int dvic_probe_of(struct platform_device *pdev)
 {
 	struct panel_drv_data *ddata = platform_get_drvdata(pdev);
@@ -319,17 +279,12 @@ static int dvic_probe(struct platform_device *pdev)
 
 	platform_set_drvdata(pdev, ddata);
 
-	if (dev_get_platdata(&pdev->dev)) {
-		r = dvic_probe_pdata(pdev);
-		if (r)
-			return r;
-	} else if (pdev->dev.of_node) {
-		r = dvic_probe_of(pdev);
-		if (r)
-			return r;
-	} else {
+	if (!pdev->dev.of_node)
 		return -ENODEV;
-	}
+
+	r = dvic_probe_of(pdev);
+	if (r)
+		return r;
 
 	ddata->timings = dvic_default_timings;
 
diff --git a/drivers/gpu/drm/omapdrm/displays/connector-hdmi.c b/drivers/gpu/drm/omapdrm/displays/connector-hdmi.c
index 6ee4129bc0c0..225fd8d6ab31 100644
--- a/drivers/gpu/drm/omapdrm/displays/connector-hdmi.c
+++ b/drivers/gpu/drm/omapdrm/displays/connector-hdmi.c
@@ -206,30 +206,6 @@ static struct omap_dss_driver hdmic_driver = {
 	.set_hdmi_infoframe	= hdmic_set_infoframe,
 };
 
-static int hdmic_probe_pdata(struct platform_device *pdev)
-{
-	struct panel_drv_data *ddata = platform_get_drvdata(pdev);
-	struct connector_hdmi_platform_data *pdata;
-	struct omap_dss_device *in, *dssdev;
-
-	pdata = dev_get_platdata(&pdev->dev);
-
-	ddata->hpd_gpio = -ENODEV;
-
-	in = omap_dss_find_output(pdata->source);
-	if (in == NULL) {
-		dev_err(&pdev->dev, "Failed to find video source\n");
-		return -EPROBE_DEFER;
-	}
-
-	ddata->in = in;
-
-	dssdev = &ddata->dssdev;
-	dssdev->name = pdata->name;
-
-	return 0;
-}
-
 static int hdmic_probe_of(struct platform_device *pdev)
 {
 	struct panel_drv_data *ddata = platform_get_drvdata(pdev);
@@ -268,17 +244,12 @@ static int hdmic_probe(struct platform_device *pdev)
 	platform_set_drvdata(pdev, ddata);
 	ddata->dev = &pdev->dev;
 
-	if (dev_get_platdata(&pdev->dev)) {
-		r = hdmic_probe_pdata(pdev);
-		if (r)
-			return r;
-	} else if (pdev->dev.of_node) {
-		r = hdmic_probe_of(pdev);
-		if (r)
-			return r;
-	} else {
+	if (!pdev->dev.of_node)
 		return -ENODEV;
-	}
+
+	r = hdmic_probe_of(pdev);
+	if (r)
+		return r;
 
 	if (gpio_is_valid(ddata->hpd_gpio)) {
 		r = devm_gpio_request_one(&pdev->dev, ddata->hpd_gpio,
diff --git a/drivers/gpu/drm/omapdrm/displays/encoder-tfp410.c b/drivers/gpu/drm/omapdrm/displays/encoder-tfp410.c
index d9048b3df495..2fd5602880a7 100644
--- a/drivers/gpu/drm/omapdrm/displays/encoder-tfp410.c
+++ b/drivers/gpu/drm/omapdrm/displays/encoder-tfp410.c
@@ -166,32 +166,6 @@ static const struct omapdss_dvi_ops tfp410_dvi_ops = {
 	.get_timings	= tfp410_get_timings,
 };
 
-static int tfp410_probe_pdata(struct platform_device *pdev)
-{
-	struct panel_drv_data *ddata = platform_get_drvdata(pdev);
-	struct encoder_tfp410_platform_data *pdata;
-	struct omap_dss_device *dssdev, *in;
-
-	pdata = dev_get_platdata(&pdev->dev);
-
-	ddata->pd_gpio = pdata->power_down_gpio;
-
-	ddata->data_lines = pdata->data_lines;
-
-	in = omap_dss_find_output(pdata->source);
-	if (in == NULL) {
-		dev_err(&pdev->dev, "Failed to find video source\n");
-		return -ENODEV;
-	}
-
-	ddata->in = in;
-
-	dssdev = &ddata->dssdev;
-	dssdev->name = pdata->name;
-
-	return 0;
-}
-
 static int tfp410_probe_of(struct platform_device *pdev)
 {
 	struct panel_drv_data *ddata = platform_get_drvdata(pdev);
@@ -231,17 +205,12 @@ static int tfp410_probe(struct platform_device *pdev)
 
 	platform_set_drvdata(pdev, ddata);
 
-	if (dev_get_platdata(&pdev->dev)) {
-		r = tfp410_probe_pdata(pdev);
-		if (r)
-			return r;
-	} else if (pdev->dev.of_node) {
-		r = tfp410_probe_of(pdev);
-		if (r)
-			return r;
-	} else {
+	if (!pdev->dev.of_node)
 		return -ENODEV;
-	}
+
+	r = tfp410_probe_of(pdev);
+	if (r)
+		return r;
 
 	if (gpio_is_valid(ddata->pd_gpio)) {
 		r = devm_gpio_request_one(&pdev->dev, ddata->pd_gpio,
diff --git a/drivers/gpu/drm/omapdrm/displays/encoder-tpd12s015.c b/drivers/gpu/drm/omapdrm/displays/encoder-tpd12s015.c
index 990af6baeb0f..916a89978387 100644
--- a/drivers/gpu/drm/omapdrm/displays/encoder-tpd12s015.c
+++ b/drivers/gpu/drm/omapdrm/displays/encoder-tpd12s015.c
@@ -13,9 +13,8 @@
 #include <linux/delay.h>
 #include <linux/module.h>
 #include <linux/slab.h>
-#include <linux/gpio.h>
 #include <linux/platform_device.h>
-#include <linux/of_gpio.h>
+#include <linux/gpio/consumer.h>
 
 #include <video/omapdss.h>
 #include <video/omap-panel-data.h>
@@ -24,9 +23,9 @@ struct panel_drv_data {
 	struct omap_dss_device dssdev;
 	struct omap_dss_device *in;
 
-	int ct_cp_hpd_gpio;
-	int ls_oe_gpio;
-	int hpd_gpio;
+	struct gpio_desc *ct_cp_hpd_gpio;
+	struct gpio_desc *ls_oe_gpio;
+	struct gpio_desc *hpd_gpio;
 
 	struct omap_video_timings timings;
 };
@@ -47,7 +46,7 @@ static int tpd_connect(struct omap_dss_device *dssdev,
 	dst->src = dssdev;
 	dssdev->dst = dst;
 
-	gpio_set_value_cansleep(ddata->ct_cp_hpd_gpio, 1);
+	gpiod_set_value_cansleep(ddata->ct_cp_hpd_gpio, 1);
 	/* DC-DC converter needs at max 300us to get to 90% of 5V */
 	udelay(300);
 
@@ -65,7 +64,7 @@ static void tpd_disconnect(struct omap_dss_device *dssdev,
 	if (dst != dssdev->dst)
 		return;
 
-	gpio_set_value_cansleep(ddata->ct_cp_hpd_gpio, 0);
+	gpiod_set_value_cansleep(ddata->ct_cp_hpd_gpio, 0);
 
 	dst->src = NULL;
 	dssdev->dst = NULL;
@@ -145,16 +144,14 @@ static int tpd_read_edid(struct omap_dss_device *dssdev,
 	struct omap_dss_device *in = ddata->in;
 	int r;
 
-	if (!gpio_get_value_cansleep(ddata->hpd_gpio))
+	if (!gpiod_get_value_cansleep(ddata->hpd_gpio))
 		return -ENODEV;
 
-	if (gpio_is_valid(ddata->ls_oe_gpio))
-		gpio_set_value_cansleep(ddata->ls_oe_gpio, 1);
+	gpiod_set_value_cansleep(ddata->ls_oe_gpio, 1);
 
 	r = in->ops.hdmi->read_edid(in, edid, len);
 
-	if (gpio_is_valid(ddata->ls_oe_gpio))
-		gpio_set_value_cansleep(ddata->ls_oe_gpio, 0);
+	gpiod_set_value_cansleep(ddata->ls_oe_gpio, 0);
 
 	return r;
 }
@@ -163,7 +160,7 @@ static bool tpd_detect(struct omap_dss_device *dssdev)
 {
 	struct panel_drv_data *ddata = to_panel_data(dssdev);
 
-	return gpio_get_value_cansleep(ddata->hpd_gpio);
+	return gpiod_get_value_cansleep(ddata->hpd_gpio);
 }
 
 static int tpd_set_infoframe(struct omap_dss_device *dssdev,
@@ -201,63 +198,11 @@ static const struct omapdss_hdmi_ops tpd_hdmi_ops = {
 	.set_hdmi_mode		= tpd_set_hdmi_mode,
 };
 
-static int tpd_probe_pdata(struct platform_device *pdev)
-{
-	struct panel_drv_data *ddata = platform_get_drvdata(pdev);
-	struct encoder_tpd12s015_platform_data *pdata;
-	struct omap_dss_device *dssdev, *in;
-
-	pdata = dev_get_platdata(&pdev->dev);
-
-	ddata->ct_cp_hpd_gpio = pdata->ct_cp_hpd_gpio;
-	ddata->ls_oe_gpio = pdata->ls_oe_gpio;
-	ddata->hpd_gpio = pdata->hpd_gpio;
-
-	in = omap_dss_find_output(pdata->source);
-	if (in == NULL) {
-		dev_err(&pdev->dev, "Failed to find video source\n");
-		return -ENODEV;
-	}
-
-	ddata->in = in;
-
-	dssdev = &ddata->dssdev;
-	dssdev->name = pdata->name;
-
-	return 0;
-}
-
 static int tpd_probe_of(struct platform_device *pdev)
 {
 	struct panel_drv_data *ddata = platform_get_drvdata(pdev);
 	struct device_node *node = pdev->dev.of_node;
 	struct omap_dss_device *in;
-	int gpio;
-
-	/* CT CP HPD GPIO */
-	gpio = of_get_gpio(node, 0);
-	if (!gpio_is_valid(gpio)) {
-		dev_err(&pdev->dev, "failed to parse CT CP HPD gpio\n");
-		return gpio;
-	}
-	ddata->ct_cp_hpd_gpio = gpio;
-
-	/* LS OE GPIO */
-	gpio = of_get_gpio(node, 1);
-	if (gpio_is_valid(gpio) || gpio == -ENOENT) {
-		ddata->ls_oe_gpio = gpio;
-	} else {
-		dev_err(&pdev->dev, "failed to parse LS OE gpio\n");
-		return gpio;
-	}
-
-	/* HPD GPIO */
-	gpio = of_get_gpio(node, 2);
-	if (!gpio_is_valid(gpio)) {
-		dev_err(&pdev->dev, "failed to parse HPD gpio\n");
-		return gpio;
-	}
-	ddata->hpd_gpio = gpio;
 
 	in = omapdss_of_find_source_for_first_ep(node);
 	if (IS_ERR(in)) {
@@ -275,6 +220,7 @@ static int tpd_probe(struct platform_device *pdev)
 	struct omap_dss_device *in, *dssdev;
 	struct panel_drv_data *ddata;
 	int r;
+	struct gpio_desc *gpio;
 
 	ddata = devm_kzalloc(&pdev->dev, sizeof(*ddata), GFP_KERNEL);
 	if (!ddata)
@@ -282,35 +228,35 @@ static int tpd_probe(struct platform_device *pdev)
 
 	platform_set_drvdata(pdev, ddata);
 
-	if (dev_get_platdata(&pdev->dev)) {
-		r = tpd_probe_pdata(pdev);
-		if (r)
-			return r;
-	} else if (pdev->dev.of_node) {
-		r = tpd_probe_of(pdev);
-		if (r)
-			return r;
-	} else {
+	if (!pdev->dev.of_node)
 		return -ENODEV;
-	}
 
-	r = devm_gpio_request_one(&pdev->dev, ddata->ct_cp_hpd_gpio,
-			GPIOF_OUT_INIT_LOW, "hdmi_ct_cp_hpd");
+	r = tpd_probe_of(pdev);
 	if (r)
+		return r;
+
+
+	gpio = devm_gpiod_get_index_optional(&pdev->dev, NULL, 0,
+		 GPIOD_OUT_LOW);
+	if (IS_ERR(gpio))
 		goto err_gpio;
 
-	if (gpio_is_valid(ddata->ls_oe_gpio)) {
-		r = devm_gpio_request_one(&pdev->dev, ddata->ls_oe_gpio,
-				GPIOF_OUT_INIT_LOW, "hdmi_ls_oe");
-		if (r)
-			goto err_gpio;
-	}
+	ddata->ct_cp_hpd_gpio = gpio;
 
-	r = devm_gpio_request_one(&pdev->dev, ddata->hpd_gpio,
-			GPIOF_DIR_IN, "hdmi_hpd");
-	if (r)
+	gpio = devm_gpiod_get_index_optional(&pdev->dev, NULL, 1,
+		 GPIOD_OUT_LOW);
+	if (IS_ERR(gpio))
+		goto err_gpio;
+
+	ddata->ls_oe_gpio = gpio;
+
+	gpio = devm_gpiod_get_index(&pdev->dev, NULL, 2,
+		GPIOD_IN);
+	if (IS_ERR(gpio))
 		goto err_gpio;
 
+	ddata->hpd_gpio = gpio;
+
 	dssdev = &ddata->dssdev;
 	dssdev->ops.hdmi = &tpd_hdmi_ops;
 	dssdev->dev = &pdev->dev;
diff --git a/drivers/gpu/drm/omapdrm/displays/panel-dsi-cm.c b/drivers/gpu/drm/omapdrm/displays/panel-dsi-cm.c
index 3414c2609320..36485c2137ce 100644
--- a/drivers/gpu/drm/omapdrm/displays/panel-dsi-cm.c
+++ b/drivers/gpu/drm/omapdrm/displays/panel-dsi-cm.c
@@ -1127,40 +1127,6 @@ static struct omap_dss_driver dsicm_ops = {
 	.memory_read	= dsicm_memory_read,
 };
 
-static int dsicm_probe_pdata(struct platform_device *pdev)
-{
-	const struct panel_dsicm_platform_data *pdata;
-	struct panel_drv_data *ddata = platform_get_drvdata(pdev);
-	struct omap_dss_device *dssdev, *in;
-
-	pdata = dev_get_platdata(&pdev->dev);
-
-	in = omap_dss_find_output(pdata->source);
-	if (in == NULL) {
-		dev_err(&pdev->dev, "failed to find video source\n");
-		return -EPROBE_DEFER;
-	}
-	ddata->in = in;
-
-	ddata->reset_gpio = pdata->reset_gpio;
-
-	if (pdata->use_ext_te)
-		ddata->ext_te_gpio = pdata->ext_te_gpio;
-	else
-		ddata->ext_te_gpio = -1;
-
-	ddata->ulps_timeout = pdata->ulps_timeout;
-
-	ddata->use_dsi_backlight = pdata->use_dsi_backlight;
-
-	ddata->pin_config = pdata->pin_config;
-
-	dssdev = &ddata->dssdev;
-	dssdev->name = pdata->name;
-
-	return 0;
-}
-
 static int dsicm_probe_of(struct platform_device *pdev)
 {
 	struct device_node *node = pdev->dev.of_node;
@@ -1214,17 +1180,12 @@ static int dsicm_probe(struct platform_device *pdev)
 	platform_set_drvdata(pdev, ddata);
 	ddata->pdev = pdev;
 
-	if (dev_get_platdata(dev)) {
-		r = dsicm_probe_pdata(pdev);
-		if (r)
-			return r;
-	} else if (pdev->dev.of_node) {
-		r = dsicm_probe_of(pdev);
-		if (r)
-			return r;
-	} else {
+	if (!pdev->dev.of_node)
 		return -ENODEV;
-	}
+
+	r = dsicm_probe_of(pdev);
+	if (r)
+		return r;
 
 	ddata->timings.x_res = 864;
 	ddata->timings.y_res = 480;
diff --git a/drivers/gpu/drm/omapdrm/displays/panel-lgphilips-lb035q02.c b/drivers/gpu/drm/omapdrm/displays/panel-lgphilips-lb035q02.c
index 18eb60e9c9ec..458f77bc473d 100644
--- a/drivers/gpu/drm/omapdrm/displays/panel-lgphilips-lb035q02.c
+++ b/drivers/gpu/drm/omapdrm/displays/panel-lgphilips-lb035q02.c
@@ -240,44 +240,6 @@ static struct omap_dss_driver lb035q02_ops = {
 	.get_resolution	= omapdss_default_get_resolution,
 };
 
-static int lb035q02_probe_pdata(struct spi_device *spi)
-{
-	const struct panel_lb035q02_platform_data *pdata;
-	struct panel_drv_data *ddata = dev_get_drvdata(&spi->dev);
-	struct omap_dss_device *dssdev, *in;
-	int r;
-
-	pdata = dev_get_platdata(&spi->dev);
-
-	in = omap_dss_find_output(pdata->source);
-	if (in == NULL) {
-		dev_err(&spi->dev, "failed to find video source '%s'\n",
-				pdata->source);
-		return -EPROBE_DEFER;
-	}
-
-	ddata->in = in;
-
-	ddata->data_lines = pdata->data_lines;
-
-	dssdev = &ddata->dssdev;
-	dssdev->name = pdata->name;
-
-	r = devm_gpio_request_one(&spi->dev, pdata->enable_gpio,
-					GPIOF_OUT_INIT_LOW, "panel enable");
-	if (r)
-		goto err_gpio;
-
-	ddata->enable_gpio = gpio_to_desc(pdata->enable_gpio);
-
-	ddata->backlight_gpio = pdata->backlight_gpio;
-
-	return 0;
-err_gpio:
-	omap_dss_put_device(ddata->in);
-	return r;
-}
-
 static int lb035q02_probe_of(struct spi_device *spi)
 {
 	struct device_node *node = spi->dev.of_node;
@@ -320,17 +282,12 @@ static int lb035q02_panel_spi_probe(struct spi_device *spi)
 
 	ddata->spi = spi;
 
-	if (dev_get_platdata(&spi->dev)) {
-		r = lb035q02_probe_pdata(spi);
-		if (r)
-			return r;
-	} else if (spi->dev.of_node) {
-		r = lb035q02_probe_of(spi);
-		if (r)
-			return r;
-	} else {
+	if (!spi->dev.of_node)
 		return -ENODEV;
-	}
+
+	r = lb035q02_probe_of(spi);
+	if (r)
+		return r;
 
 	if (gpio_is_valid(ddata->backlight_gpio)) {
 		r = devm_gpio_request_one(&spi->dev, ddata->backlight_gpio,
diff --git a/drivers/gpu/drm/omapdrm/displays/panel-nec-nl8048hl11.c b/drivers/gpu/drm/omapdrm/displays/panel-nec-nl8048hl11.c
index 8a928c9a2fc9..780cb263a318 100644
--- a/drivers/gpu/drm/omapdrm/displays/panel-nec-nl8048hl11.c
+++ b/drivers/gpu/drm/omapdrm/displays/panel-nec-nl8048hl11.c
@@ -19,7 +19,6 @@
 #include <linux/of_gpio.h>
 
 #include <video/omapdss.h>
-#include <video/omap-panel-data.h>
 
 struct panel_drv_data {
 	struct omap_dss_device	dssdev;
@@ -232,34 +231,6 @@ static struct omap_dss_driver nec_8048_ops = {
 	.get_resolution	= omapdss_default_get_resolution,
 };
 
-
-static int nec_8048_probe_pdata(struct spi_device *spi)
-{
-	const struct panel_nec_nl8048hl11_platform_data *pdata;
-	struct panel_drv_data *ddata = dev_get_drvdata(&spi->dev);
-	struct omap_dss_device *dssdev, *in;
-
-	pdata = dev_get_platdata(&spi->dev);
-
-	ddata->qvga_gpio = pdata->qvga_gpio;
-	ddata->res_gpio = pdata->res_gpio;
-
-	in = omap_dss_find_output(pdata->source);
-	if (in == NULL) {
-		dev_err(&spi->dev, "failed to find video source '%s'\n",
-				pdata->source);
-		return -EPROBE_DEFER;
-	}
-	ddata->in = in;
-
-	ddata->data_lines = pdata->data_lines;
-
-	dssdev = &ddata->dssdev;
-	dssdev->name = pdata->name;
-
-	return 0;
-}
-
 static int nec_8048_probe_of(struct spi_device *spi)
 {
 	struct device_node *node = spi->dev.of_node;
@@ -315,17 +286,12 @@ static int nec_8048_probe(struct spi_device *spi)
 
 	ddata->spi = spi;
 
-	if (dev_get_platdata(&spi->dev)) {
-		r = nec_8048_probe_pdata(spi);
-		if (r)
-			return r;
-	} else if (spi->dev.of_node) {
-		r = nec_8048_probe_of(spi);
-		if (r)
-			return r;
-	} else {
+	if (!spi->dev.of_node)
 		return -ENODEV;
-	}
+
+	r = nec_8048_probe_of(spi);
+	if (r)
+		return r;
 
 	if (gpio_is_valid(ddata->qvga_gpio)) {
 		r = devm_gpio_request_one(&spi->dev, ddata->qvga_gpio,
diff --git a/drivers/gpu/drm/omapdrm/displays/panel-sharp-ls037v7dw01.c b/drivers/gpu/drm/omapdrm/displays/panel-sharp-ls037v7dw01.c
index abfd1f6e3327..529a017602e4 100644
--- a/drivers/gpu/drm/omapdrm/displays/panel-sharp-ls037v7dw01.c
+++ b/drivers/gpu/drm/omapdrm/displays/panel-sharp-ls037v7dw01.c
@@ -18,7 +18,6 @@
 #include <linux/slab.h>
 #include <linux/regulator/consumer.h>
 #include <video/omapdss.h>
-#include <video/omap-panel-data.h>
 
 struct panel_drv_data {
 	struct omap_dss_device dssdev;
@@ -197,73 +196,6 @@ static struct omap_dss_driver sharp_ls_ops = {
 	.get_resolution	= omapdss_default_get_resolution,
 };
 
-static int sharp_ls_get_gpio(struct device *dev, int gpio, unsigned long flags,
-		  char *desc, struct gpio_desc **gpiod)
-{
-	struct gpio_desc *gd;
-	int r;
-
-	*gpiod = NULL;
-
-	r = devm_gpio_request_one(dev, gpio, flags, desc);
-	if (r)
-		return r == -ENOENT ? 0 : r;
-
-	gd = gpio_to_desc(gpio);
-	if (IS_ERR(gd))
-		return PTR_ERR(gd) == -ENOENT ? 0 : PTR_ERR(gd);
-
-	*gpiod = gd;
-	return 0;
-}
-
-static int sharp_ls_probe_pdata(struct platform_device *pdev)
-{
-	const struct panel_sharp_ls037v7dw01_platform_data *pdata;
-	struct panel_drv_data *ddata = platform_get_drvdata(pdev);
-	struct omap_dss_device *dssdev, *in;
-	int r;
-
-	pdata = dev_get_platdata(&pdev->dev);
-
-	in = omap_dss_find_output(pdata->source);
-	if (in == NULL) {
-		dev_err(&pdev->dev, "failed to find video source '%s'\n",
-				pdata->source);
-		return -EPROBE_DEFER;
-	}
-
-	ddata->in = in;
-
-	ddata->data_lines = pdata->data_lines;
-
-	dssdev = &ddata->dssdev;
-	dssdev->name = pdata->name;
-
-	r = sharp_ls_get_gpio(&pdev->dev, pdata->mo_gpio, GPIOF_OUT_INIT_LOW,
-		"lcd MO", &ddata->mo_gpio);
-	if (r)
-		return r;
-	r = sharp_ls_get_gpio(&pdev->dev, pdata->lr_gpio, GPIOF_OUT_INIT_HIGH,
-		"lcd LR", &ddata->lr_gpio);
-	if (r)
-		return r;
-	r = sharp_ls_get_gpio(&pdev->dev, pdata->ud_gpio, GPIOF_OUT_INIT_HIGH,
-		"lcd UD", &ddata->ud_gpio);
-	if (r)
-		return r;
-	r = sharp_ls_get_gpio(&pdev->dev, pdata->resb_gpio, GPIOF_OUT_INIT_LOW,
-		"lcd RESB", &ddata->resb_gpio);
-	if (r)
-		return r;
-	r = sharp_ls_get_gpio(&pdev->dev, pdata->ini_gpio, GPIOF_OUT_INIT_LOW,
-		"lcd INI", &ddata->ini_gpio);
-	if (r)
-		return r;
-
-	return 0;
-}
-
 static  int sharp_ls_get_gpio_of(struct device *dev, int index, int val,
 	const char *desc, struct gpio_desc **gpiod)
 {
@@ -340,17 +272,12 @@ static int sharp_ls_probe(struct platform_device *pdev)
 
 	platform_set_drvdata(pdev, ddata);
 
-	if (dev_get_platdata(&pdev->dev)) {
-		r = sharp_ls_probe_pdata(pdev);
-		if (r)
-			return r;
-	} else if (pdev->dev.of_node) {
-		r = sharp_ls_probe_of(pdev);
-		if (r)
-			return r;
-	} else {
+	if (!pdev->dev.of_node)
 		return -ENODEV;
-	}
+
+	r = sharp_ls_probe_of(pdev);
+	if (r)
+		return r;
 
 	ddata->videomode = sharp_ls_timings;
 
diff --git a/drivers/gpu/drm/omapdrm/displays/panel-tpo-td028ttec1.c b/drivers/gpu/drm/omapdrm/displays/panel-tpo-td028ttec1.c
index 4d657f3ab679..bd8d85041926 100644
--- a/drivers/gpu/drm/omapdrm/displays/panel-tpo-td028ttec1.c
+++ b/drivers/gpu/drm/omapdrm/displays/panel-tpo-td028ttec1.c
@@ -29,7 +29,6 @@
 #include <linux/spi/spi.h>
 #include <linux/gpio.h>
 #include <video/omapdss.h>
-#include <video/omap-panel-data.h>
 
 struct panel_drv_data {
 	struct omap_dss_device dssdev;
@@ -365,31 +364,6 @@ static struct omap_dss_driver td028ttec1_ops = {
 	.check_timings	= td028ttec1_panel_check_timings,
 };
 
-static int td028ttec1_panel_probe_pdata(struct spi_device *spi)
-{
-	const struct panel_tpo_td028ttec1_platform_data *pdata;
-	struct panel_drv_data *ddata = dev_get_drvdata(&spi->dev);
-	struct omap_dss_device *dssdev, *in;
-
-	pdata = dev_get_platdata(&spi->dev);
-
-	in = omap_dss_find_output(pdata->source);
-	if (in == NULL) {
-		dev_err(&spi->dev, "failed to find video source '%s'\n",
-				pdata->source);
-		return -EPROBE_DEFER;
-	}
-
-	ddata->in = in;
-
-	ddata->data_lines = pdata->data_lines;
-
-	dssdev = &ddata->dssdev;
-	dssdev->name = pdata->name;
-
-	return 0;
-}
-
 static int td028ttec1_probe_of(struct spi_device *spi)
 {
 	struct device_node *node = spi->dev.of_node;
@@ -432,17 +406,12 @@ static int td028ttec1_panel_probe(struct spi_device *spi)
 
 	ddata->spi_dev = spi;
 
-	if (dev_get_platdata(&spi->dev)) {
-		r = td028ttec1_panel_probe_pdata(spi);
-		if (r)
-			return r;
-	} else if (spi->dev.of_node) {
-		r = td028ttec1_probe_of(spi);
-		if (r)
-			return r;
-	} else {
+	if (!spi->dev.of_node)
 		return -ENODEV;
-	}
+
+	r = td028ttec1_probe_of(spi);
+	if (r)
+		return r;
 
 	ddata->videomode = td028ttec1_panel_timings;
 
diff --git a/drivers/gpu/drm/omapdrm/displays/panel-tpo-td043mtea1.c b/drivers/gpu/drm/omapdrm/displays/panel-tpo-td043mtea1.c
index 68e3b68a2920..03e2beb7b4f0 100644
--- a/drivers/gpu/drm/omapdrm/displays/panel-tpo-td043mtea1.c
+++ b/drivers/gpu/drm/omapdrm/displays/panel-tpo-td043mtea1.c
@@ -20,7 +20,6 @@
 #include <linux/of_gpio.h>
 
 #include <video/omapdss.h>
-#include <video/omap-panel-data.h>
 
 #define TPO_R02_MODE(x)		((x) & 7)
 #define TPO_R02_MODE_800x480	7
@@ -464,33 +463,6 @@ static struct omap_dss_driver tpo_td043_ops = {
 	.get_resolution	= omapdss_default_get_resolution,
 };
 
-
-static int tpo_td043_probe_pdata(struct spi_device *spi)
-{
-	const struct panel_tpo_td043mtea1_platform_data *pdata;
-	struct panel_drv_data *ddata = dev_get_drvdata(&spi->dev);
-	struct omap_dss_device *dssdev, *in;
-
-	pdata = dev_get_platdata(&spi->dev);
-
-	ddata->nreset_gpio = pdata->nreset_gpio;
-
-	in = omap_dss_find_output(pdata->source);
-	if (in == NULL) {
-		dev_err(&spi->dev, "failed to find video source '%s'\n",
-				pdata->source);
-		return -EPROBE_DEFER;
-	}
-	ddata->in = in;
-
-	ddata->data_lines = pdata->data_lines;
-
-	dssdev = &ddata->dssdev;
-	dssdev->name = pdata->name;
-
-	return 0;
-}
-
 static int tpo_td043_probe_of(struct spi_device *spi)
 {
 	struct device_node *node = spi->dev.of_node;
@@ -541,17 +513,12 @@ static int tpo_td043_probe(struct spi_device *spi)
 
 	ddata->spi = spi;
 
-	if (dev_get_platdata(&spi->dev)) {
-		r = tpo_td043_probe_pdata(spi);
-		if (r)
-			return r;
-	} else if (spi->dev.of_node) {
-		r = tpo_td043_probe_of(spi);
-		if (r)
-			return r;
-	} else {
+	if (!spi->dev.of_node)
 		return -ENODEV;
-	}
+
+	r = tpo_td043_probe_of(spi);
+	if (r)
+		return r;
 
 	ddata->mode = TPO_R02_MODE_800x480;
 	memcpy(ddata->gamma, tpo_td043_def_gamma, sizeof(ddata->gamma));
diff --git a/drivers/gpu/drm/omapdrm/dss/Makefile b/drivers/gpu/drm/omapdrm/dss/Makefile
index b5136d3d4b77..b651ec9751e6 100644
--- a/drivers/gpu/drm/omapdrm/dss/Makefile
+++ b/drivers/gpu/drm/omapdrm/dss/Makefile
@@ -3,9 +3,6 @@ obj-$(CONFIG_OMAP2_DSS) += omapdss.o
 # Core DSS files
 omapdss-y := core.o dss.o dss_features.o dispc.o dispc_coefs.o display.o \
 	output.o dss-of.o pll.o video-pll.o
-# DSS compat layer files
-omapdss-y += manager.o manager-sysfs.o overlay.o overlay-sysfs.o apply.o \
-	dispc-compat.o display-sysfs.o
 omapdss-$(CONFIG_OMAP2_DSS_DPI) += dpi.o
 omapdss-$(CONFIG_OMAP2_DSS_RFBI) += rfbi.o
 omapdss-$(CONFIG_OMAP2_DSS_VENC) += venc.o
diff --git a/drivers/gpu/drm/omapdrm/dss/apply.c b/drivers/gpu/drm/omapdrm/dss/apply.c
deleted file mode 100644
index 663ccc3bf4e5..000000000000
--- a/drivers/gpu/drm/omapdrm/dss/apply.c
+++ /dev/null
@@ -1,1702 +0,0 @@
-/*
- * Copyright (C) 2011 Texas Instruments
- * Author: Tomi Valkeinen <tomi.valkeinen@ti.com>
- *
- * This program is free software; you can redistribute it and/or modify it
- * under the terms of the GNU General Public License version 2 as published by
- * the Free Software Foundation.
- *
- * This program is distributed in the hope that it will be useful, but WITHOUT
- * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
- * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
- * more details.
- *
- * You should have received a copy of the GNU General Public License along with
- * this program.  If not, see <http://www.gnu.org/licenses/>.
- */
-
-#define DSS_SUBSYS_NAME "APPLY"
-
-#include <linux/kernel.h>
-#include <linux/module.h>
-#include <linux/slab.h>
-#include <linux/spinlock.h>
-#include <linux/jiffies.h>
-
-#include <video/omapdss.h>
-
-#include "dss.h"
-#include "dss_features.h"
-#include "dispc-compat.h"
-
-/*
- * We have 4 levels of cache for the dispc settings. First two are in SW and
- * the latter two in HW.
- *
- *       set_info()
- *          v
- * +--------------------+
- * |     user_info      |
- * +--------------------+
- *          v
- *        apply()
- *          v
- * +--------------------+
- * |       info         |
- * +--------------------+
- *          v
- *      write_regs()
- *          v
- * +--------------------+
- * |  shadow registers  |
- * +--------------------+
- *          v
- * VFP or lcd/digit_enable
- *          v
- * +--------------------+
- * |      registers     |
- * +--------------------+
- */
-
-struct ovl_priv_data {
-
-	bool user_info_dirty;
-	struct omap_overlay_info user_info;
-
-	bool info_dirty;
-	struct omap_overlay_info info;
-
-	bool shadow_info_dirty;
-
-	bool extra_info_dirty;
-	bool shadow_extra_info_dirty;
-
-	bool enabled;
-	u32 fifo_low, fifo_high;
-
-	/*
-	 * True if overlay is to be enabled. Used to check and calculate configs
-	 * for the overlay before it is enabled in the HW.
-	 */
-	bool enabling;
-};
-
-struct mgr_priv_data {
-
-	bool user_info_dirty;
-	struct omap_overlay_manager_info user_info;
-
-	bool info_dirty;
-	struct omap_overlay_manager_info info;
-
-	bool shadow_info_dirty;
-
-	/* If true, GO bit is up and shadow registers cannot be written.
-	 * Never true for manual update displays */
-	bool busy;
-
-	/* If true, dispc output is enabled */
-	bool updating;
-
-	/* If true, a display is enabled using this manager */
-	bool enabled;
-
-	bool extra_info_dirty;
-	bool shadow_extra_info_dirty;
-
-	struct omap_video_timings timings;
-	struct dss_lcd_mgr_config lcd_config;
-
-	void (*framedone_handler)(void *);
-	void *framedone_handler_data;
-};
-
-static struct {
-	struct ovl_priv_data ovl_priv_data_array[MAX_DSS_OVERLAYS];
-	struct mgr_priv_data mgr_priv_data_array[MAX_DSS_MANAGERS];
-
-	bool irq_enabled;
-} dss_data;
-
-/* protects dss_data */
-static spinlock_t data_lock;
-/* lock for blocking functions */
-static DEFINE_MUTEX(apply_lock);
-static DECLARE_COMPLETION(extra_updated_completion);
-
-static void dss_register_vsync_isr(void);
-
-static struct ovl_priv_data *get_ovl_priv(struct omap_overlay *ovl)
-{
-	return &dss_data.ovl_priv_data_array[ovl->id];
-}
-
-static struct mgr_priv_data *get_mgr_priv(struct omap_overlay_manager *mgr)
-{
-	return &dss_data.mgr_priv_data_array[mgr->id];
-}
-
-static void apply_init_priv(void)
-{
-	const int num_ovls = dss_feat_get_num_ovls();
-	struct mgr_priv_data *mp;
-	int i;
-
-	spin_lock_init(&data_lock);
-
-	for (i = 0; i < num_ovls; ++i) {
-		struct ovl_priv_data *op;
-
-		op = &dss_data.ovl_priv_data_array[i];
-
-		op->info.color_mode = OMAP_DSS_COLOR_RGB16;
-		op->info.rotation_type = OMAP_DSS_ROT_DMA;
-
-		op->info.global_alpha = 255;
-
-		switch (i) {
-		case 0:
-			op->info.zorder = 0;
-			break;
-		case 1:
-			op->info.zorder =
-				dss_has_feature(FEAT_ALPHA_FREE_ZORDER) ? 3 : 0;
-			break;
-		case 2:
-			op->info.zorder =
-				dss_has_feature(FEAT_ALPHA_FREE_ZORDER) ? 2 : 0;
-			break;
-		case 3:
-			op->info.zorder =
-				dss_has_feature(FEAT_ALPHA_FREE_ZORDER) ? 1 : 0;
-			break;
-		}
-
-		op->user_info = op->info;
-	}
-
-	/*
-	 * Initialize some of the lcd_config fields for TV manager, this lets
-	 * us prevent checking if the manager is LCD or TV at some places
-	 */
-	mp = &dss_data.mgr_priv_data_array[OMAP_DSS_CHANNEL_DIGIT];
-
-	mp->lcd_config.video_port_width = 24;
-	mp->lcd_config.clock_info.lck_div = 1;
-	mp->lcd_config.clock_info.pck_div = 1;
-}
-
-/*
- * A LCD manager's stallmode decides whether it is in manual or auto update. TV
- * manager is always auto update, stallmode field for TV manager is false by
- * default
- */
-static bool ovl_manual_update(struct omap_overlay *ovl)
-{
-	struct mgr_priv_data *mp = get_mgr_priv(ovl->manager);
-
-	return mp->lcd_config.stallmode;
-}
-
-static bool mgr_manual_update(struct omap_overlay_manager *mgr)
-{
-	struct mgr_priv_data *mp = get_mgr_priv(mgr);
-
-	return mp->lcd_config.stallmode;
-}
-
-static int dss_check_settings_low(struct omap_overlay_manager *mgr,
-		bool applying)
-{
-	struct omap_overlay_info *oi;
-	struct omap_overlay_manager_info *mi;
-	struct omap_overlay *ovl;
-	struct omap_overlay_info *ois[MAX_DSS_OVERLAYS];
-	struct ovl_priv_data *op;
-	struct mgr_priv_data *mp;
-
-	mp = get_mgr_priv(mgr);
-
-	if (!mp->enabled)
-		return 0;
-
-	if (applying && mp->user_info_dirty)
-		mi = &mp->user_info;
-	else
-		mi = &mp->info;
-
-	/* collect the infos to be tested into the array */
-	list_for_each_entry(ovl, &mgr->overlays, list) {
-		op = get_ovl_priv(ovl);
-
-		if (!op->enabled && !op->enabling)
-			oi = NULL;
-		else if (applying && op->user_info_dirty)
-			oi = &op->user_info;
-		else
-			oi = &op->info;
-
-		ois[ovl->id] = oi;
-	}
-
-	return dss_mgr_check(mgr, mi, &mp->timings, &mp->lcd_config, ois);
-}
-
-/*
- * check manager and overlay settings using overlay_info from data->info
- */
-static int dss_check_settings(struct omap_overlay_manager *mgr)
-{
-	return dss_check_settings_low(mgr, false);
-}
-
-/*
- * check manager and overlay settings using overlay_info from ovl->info if
- * dirty and from data->info otherwise
- */
-static int dss_check_settings_apply(struct omap_overlay_manager *mgr)
-{
-	return dss_check_settings_low(mgr, true);
-}
-
-static bool need_isr(void)
-{
-	const int num_mgrs = dss_feat_get_num_mgrs();
-	int i;
-
-	for (i = 0; i < num_mgrs; ++i) {
-		struct omap_overlay_manager *mgr;
-		struct mgr_priv_data *mp;
-		struct omap_overlay *ovl;
-
-		mgr = omap_dss_get_overlay_manager(i);
-		mp = get_mgr_priv(mgr);
-
-		if (!mp->enabled)
-			continue;
-
-		if (mgr_manual_update(mgr)) {
-			/* to catch FRAMEDONE */
-			if (mp->updating)
-				return true;
-		} else {
-			/* to catch GO bit going down */
-			if (mp->busy)
-				return true;
-
-			/* to write new values to registers */
-			if (mp->info_dirty)
-				return true;
-
-			/* to set GO bit */
-			if (mp->shadow_info_dirty)
-				return true;
-
-			/*
-			 * NOTE: we don't check extra_info flags for disabled
-			 * managers, once the manager is enabled, the extra_info
-			 * related manager changes will be taken in by HW.
-			 */
-
-			/* to write new values to registers */
-			if (mp->extra_info_dirty)
-				return true;
-
-			/* to set GO bit */
-			if (mp->shadow_extra_info_dirty)
-				return true;
-
-			list_for_each_entry(ovl, &mgr->overlays, list) {
-				struct ovl_priv_data *op;
-
-				op = get_ovl_priv(ovl);
-
-				/*
-				 * NOTE: we check extra_info flags even for
-				 * disabled overlays, as extra_infos need to be
-				 * always written.
-				 */
-
-				/* to write new values to registers */
-				if (op->extra_info_dirty)
-					return true;
-
-				/* to set GO bit */
-				if (op->shadow_extra_info_dirty)
-					return true;
-
-				if (!op->enabled)
-					continue;
-
-				/* to write new values to registers */
-				if (op->info_dirty)
-					return true;
-
-				/* to set GO bit */
-				if (op->shadow_info_dirty)
-					return true;
-			}
-		}
-	}
-
-	return false;
-}
-
-static bool need_go(struct omap_overlay_manager *mgr)
-{
-	struct omap_overlay *ovl;
-	struct mgr_priv_data *mp;
-	struct ovl_priv_data *op;
-
-	mp = get_mgr_priv(mgr);
-
-	if (mp->shadow_info_dirty || mp->shadow_extra_info_dirty)
-		return true;
-
-	list_for_each_entry(ovl, &mgr->overlays, list) {
-		op = get_ovl_priv(ovl);
-		if (op->shadow_info_dirty || op->shadow_extra_info_dirty)
-			return true;
-	}
-
-	return false;
-}
-
-/* returns true if an extra_info field is currently being updated */
-static bool extra_info_update_ongoing(void)
-{
-	const int num_mgrs = dss_feat_get_num_mgrs();
-	int i;
-
-	for (i = 0; i < num_mgrs; ++i) {
-		struct omap_overlay_manager *mgr;
-		struct omap_overlay *ovl;
-		struct mgr_priv_data *mp;
-
-		mgr = omap_dss_get_overlay_manager(i);
-		mp = get_mgr_priv(mgr);
-
-		if (!mp->enabled)
-			continue;
-
-		if (!mp->updating)
-			continue;
-
-		if (mp->extra_info_dirty || mp->shadow_extra_info_dirty)
-			return true;
-
-		list_for_each_entry(ovl, &mgr->overlays, list) {
-			struct ovl_priv_data *op = get_ovl_priv(ovl);
-
-			if (op->extra_info_dirty || op->shadow_extra_info_dirty)
-				return true;
-		}
-	}
-
-	return false;
-}
-
-/* wait until no extra_info updates are pending */
-static void wait_pending_extra_info_updates(void)
-{
-	bool updating;
-	unsigned long flags;
-	unsigned long t;
-	int r;
-
-	spin_lock_irqsave(&data_lock, flags);
-
-	updating = extra_info_update_ongoing();
-
-	if (!updating) {
-		spin_unlock_irqrestore(&data_lock, flags);
-		return;
-	}
-
-	init_completion(&extra_updated_completion);
-
-	spin_unlock_irqrestore(&data_lock, flags);
-
-	t = msecs_to_jiffies(500);
-	r = wait_for_completion_timeout(&extra_updated_completion, t);
-	if (r == 0)
-		DSSWARN("timeout in wait_pending_extra_info_updates\n");
-}
-
-static struct omap_dss_device *dss_mgr_get_device(struct omap_overlay_manager *mgr)
-{
-	struct omap_dss_device *dssdev;
-
-	dssdev = mgr->output;
-	if (dssdev == NULL)
-		return NULL;
-
-	while (dssdev->dst)
-		dssdev = dssdev->dst;
-
-	if (dssdev->driver)
-		return dssdev;
-	else
-		return NULL;
-}
-
-static struct omap_dss_device *dss_ovl_get_device(struct omap_overlay *ovl)
-{
-	return ovl->manager ? dss_mgr_get_device(ovl->manager) : NULL;
-}
-
-static int dss_mgr_wait_for_vsync(struct omap_overlay_manager *mgr)
-{
-	unsigned long timeout = msecs_to_jiffies(500);
-	u32 irq;
-	int r;
-
-	if (mgr->output == NULL)
-		return -ENODEV;
-
-	r = dispc_runtime_get();
-	if (r)
-		return r;
-
-	switch (mgr->output->id) {
-	case OMAP_DSS_OUTPUT_VENC:
-		irq = DISPC_IRQ_EVSYNC_ODD;
-		break;
-	case OMAP_DSS_OUTPUT_HDMI:
-		irq = DISPC_IRQ_EVSYNC_EVEN;
-		break;
-	default:
-		irq = dispc_mgr_get_vsync_irq(mgr->id);
-		break;
-	}
-
-	r = omap_dispc_wait_for_irq_interruptible_timeout(irq, timeout);
-
-	dispc_runtime_put();
-
-	return r;
-}
-
-static int dss_mgr_wait_for_go(struct omap_overlay_manager *mgr)
-{
-	unsigned long timeout = msecs_to_jiffies(500);
-	struct mgr_priv_data *mp = get_mgr_priv(mgr);
-	u32 irq;
-	unsigned long flags;
-	int r;
-	int i;
-
-	spin_lock_irqsave(&data_lock, flags);
-
-	if (mgr_manual_update(mgr)) {
-		spin_unlock_irqrestore(&data_lock, flags);
-		return 0;
-	}
-
-	if (!mp->enabled) {
-		spin_unlock_irqrestore(&data_lock, flags);
-		return 0;
-	}
-
-	spin_unlock_irqrestore(&data_lock, flags);
-
-	r = dispc_runtime_get();
-	if (r)
-		return r;
-
-	irq = dispc_mgr_get_vsync_irq(mgr->id);
-
-	i = 0;
-	while (1) {
-		bool shadow_dirty, dirty;
-
-		spin_lock_irqsave(&data_lock, flags);
-		dirty = mp->info_dirty;
-		shadow_dirty = mp->shadow_info_dirty;
-		spin_unlock_irqrestore(&data_lock, flags);
-
-		if (!dirty && !shadow_dirty) {
-			r = 0;
-			break;
-		}
-
-		/* 4 iterations is the worst case:
-		 * 1 - initial iteration, dirty = true (between VFP and VSYNC)
-		 * 2 - first VSYNC, dirty = true
-		 * 3 - dirty = false, shadow_dirty = true
-		 * 4 - shadow_dirty = false */
-		if (i++ == 3) {
-			DSSERR("mgr(%d)->wait_for_go() not finishing\n",
-					mgr->id);
-			r = 0;
-			break;
-		}
-
-		r = omap_dispc_wait_for_irq_interruptible_timeout(irq, timeout);
-		if (r == -ERESTARTSYS)
-			break;
-
-		if (r) {
-			DSSERR("mgr(%d)->wait_for_go() timeout\n", mgr->id);
-			break;
-		}
-	}
-
-	dispc_runtime_put();
-
-	return r;
-}
-
-static int dss_mgr_wait_for_go_ovl(struct omap_overlay *ovl)
-{
-	unsigned long timeout = msecs_to_jiffies(500);
-	struct ovl_priv_data *op;
-	struct mgr_priv_data *mp;
-	u32 irq;
-	unsigned long flags;
-	int r;
-	int i;
-
-	if (!ovl->manager)
-		return 0;
-
-	mp = get_mgr_priv(ovl->manager);
-
-	spin_lock_irqsave(&data_lock, flags);
-
-	if (ovl_manual_update(ovl)) {
-		spin_unlock_irqrestore(&data_lock, flags);
-		return 0;
-	}
-
-	if (!mp->enabled) {
-		spin_unlock_irqrestore(&data_lock, flags);
-		return 0;
-	}
-
-	spin_unlock_irqrestore(&data_lock, flags);
-
-	r = dispc_runtime_get();
-	if (r)
-		return r;
-
-	irq = dispc_mgr_get_vsync_irq(ovl->manager->id);
-
-	op = get_ovl_priv(ovl);
-	i = 0;
-	while (1) {
-		bool shadow_dirty, dirty;
-
-		spin_lock_irqsave(&data_lock, flags);
-		dirty = op->info_dirty;
-		shadow_dirty = op->shadow_info_dirty;
-		spin_unlock_irqrestore(&data_lock, flags);
-
-		if (!dirty && !shadow_dirty) {
-			r = 0;
-			break;
-		}
-
-		/* 4 iterations is the worst case:
-		 * 1 - initial iteration, dirty = true (between VFP and VSYNC)
-		 * 2 - first VSYNC, dirty = true
-		 * 3 - dirty = false, shadow_dirty = true
-		 * 4 - shadow_dirty = false */
-		if (i++ == 3) {
-			DSSERR("ovl(%d)->wait_for_go() not finishing\n",
-					ovl->id);
-			r = 0;
-			break;
-		}
-
-		r = omap_dispc_wait_for_irq_interruptible_timeout(irq, timeout);
-		if (r == -ERESTARTSYS)
-			break;
-
-		if (r) {
-			DSSERR("ovl(%d)->wait_for_go() timeout\n", ovl->id);
-			break;
-		}
-	}
-
-	dispc_runtime_put();
-
-	return r;
-}
-
-static void dss_ovl_write_regs(struct omap_overlay *ovl)
-{
-	struct ovl_priv_data *op = get_ovl_priv(ovl);
-	struct omap_overlay_info *oi;
-	bool replication;
-	struct mgr_priv_data *mp;
-	int r;
-
-	DSSDBG("writing ovl %d regs\n", ovl->id);
-
-	if (!op->enabled || !op->info_dirty)
-		return;
-
-	oi = &op->info;
-
-	mp = get_mgr_priv(ovl->manager);
-
-	replication = dss_ovl_use_replication(mp->lcd_config, oi->color_mode);
-
-	r = dispc_ovl_setup(ovl->id, oi, replication, &mp->timings, false);
-	if (r) {
-		/*
-		 * We can't do much here, as this function can be called from
-		 * vsync interrupt.
-		 */
-		DSSERR("dispc_ovl_setup failed for ovl %d\n", ovl->id);
-
-		/* This will leave fifo configurations in a nonoptimal state */
-		op->enabled = false;
-		dispc_ovl_enable(ovl->id, false);
-		return;
-	}
-
-	op->info_dirty = false;
-	if (mp->updating)
-		op->shadow_info_dirty = true;
-}
-
-static void dss_ovl_write_regs_extra(struct omap_overlay *ovl)
-{
-	struct ovl_priv_data *op = get_ovl_priv(ovl);
-	struct mgr_priv_data *mp;
-
-	DSSDBG("writing ovl %d regs extra\n", ovl->id);
-
-	if (!op->extra_info_dirty)
-		return;
-
-	/* note: write also when op->enabled == false, so that the ovl gets
-	 * disabled */
-
-	dispc_ovl_enable(ovl->id, op->enabled);
-	dispc_ovl_set_fifo_threshold(ovl->id, op->fifo_low, op->fifo_high);
-
-	mp = get_mgr_priv(ovl->manager);
-
-	op->extra_info_dirty = false;
-	if (mp->updating)
-		op->shadow_extra_info_dirty = true;
-}
-
-static void dss_mgr_write_regs(struct omap_overlay_manager *mgr)
-{
-	struct mgr_priv_data *mp = get_mgr_priv(mgr);
-	struct omap_overlay *ovl;
-
-	DSSDBG("writing mgr %d regs\n", mgr->id);
-
-	if (!mp->enabled)
-		return;
-
-	WARN_ON(mp->busy);
-
-	/* Commit overlay settings */
-	list_for_each_entry(ovl, &mgr->overlays, list) {
-		dss_ovl_write_regs(ovl);
-		dss_ovl_write_regs_extra(ovl);
-	}
-
-	if (mp->info_dirty) {
-		dispc_mgr_setup(mgr->id, &mp->info);
-
-		mp->info_dirty = false;
-		if (mp->updating)
-			mp->shadow_info_dirty = true;
-	}
-}
-
-static void dss_mgr_write_regs_extra(struct omap_overlay_manager *mgr)
-{
-	struct mgr_priv_data *mp = get_mgr_priv(mgr);
-
-	DSSDBG("writing mgr %d regs extra\n", mgr->id);
-
-	if (!mp->extra_info_dirty)
-		return;
-
-	dispc_mgr_set_timings(mgr->id, &mp->timings);
-
-	/* lcd_config parameters */
-	if (dss_mgr_is_lcd(mgr->id))
-		dispc_mgr_set_lcd_config(mgr->id, &mp->lcd_config);
-
-	mp->extra_info_dirty = false;
-	if (mp->updating)
-		mp->shadow_extra_info_dirty = true;
-}
-
-static void dss_write_regs(void)
-{
-	const int num_mgrs = omap_dss_get_num_overlay_managers();
-	int i;
-
-	for (i = 0; i < num_mgrs; ++i) {
-		struct omap_overlay_manager *mgr;
-		struct mgr_priv_data *mp;
-		int r;
-
-		mgr = omap_dss_get_overlay_manager(i);
-		mp = get_mgr_priv(mgr);
-
-		if (!mp->enabled || mgr_manual_update(mgr) || mp->busy)
-			continue;
-
-		r = dss_check_settings(mgr);
-		if (r) {
-			DSSERR("cannot write registers for manager %s: "
-					"illegal configuration\n", mgr->name);
-			continue;
-		}
-
-		dss_mgr_write_regs(mgr);
-		dss_mgr_write_regs_extra(mgr);
-	}
-}
-
-static void dss_set_go_bits(void)
-{
-	const int num_mgrs = omap_dss_get_num_overlay_managers();
-	int i;
-
-	for (i = 0; i < num_mgrs; ++i) {
-		struct omap_overlay_manager *mgr;
-		struct mgr_priv_data *mp;
-
-		mgr = omap_dss_get_overlay_manager(i);
-		mp = get_mgr_priv(mgr);
-
-		if (!mp->enabled || mgr_manual_update(mgr) || mp->busy)
-			continue;
-
-		if (!need_go(mgr))
-			continue;
-
-		mp->busy = true;
-
-		if (!dss_data.irq_enabled && need_isr())
-			dss_register_vsync_isr();
-
-		dispc_mgr_go(mgr->id);
-	}
-
-}
-
-static void mgr_clear_shadow_dirty(struct omap_overlay_manager *mgr)
-{
-	struct omap_overlay *ovl;
-	struct mgr_priv_data *mp;
-	struct ovl_priv_data *op;
-
-	mp = get_mgr_priv(mgr);
-	mp->shadow_info_dirty = false;
-	mp->shadow_extra_info_dirty = false;
-
-	list_for_each_entry(ovl, &mgr->overlays, list) {
-		op = get_ovl_priv(ovl);
-		op->shadow_info_dirty = false;
-		op->shadow_extra_info_dirty = false;
-	}
-}
-
-static int dss_mgr_connect_compat(struct omap_overlay_manager *mgr,
-		struct omap_dss_device *dst)
-{
-	return mgr->set_output(mgr, dst);
-}
-
-static void dss_mgr_disconnect_compat(struct omap_overlay_manager *mgr,
-		struct omap_dss_device *dst)
-{
-	mgr->unset_output(mgr);
-}
-
-static void dss_mgr_start_update_compat(struct omap_overlay_manager *mgr)
-{
-	struct mgr_priv_data *mp = get_mgr_priv(mgr);
-	unsigned long flags;
-	int r;
-
-	spin_lock_irqsave(&data_lock, flags);
-
-	WARN_ON(mp->updating);
-
-	r = dss_check_settings(mgr);
-	if (r) {
-		DSSERR("cannot start manual update: illegal configuration\n");
-		spin_unlock_irqrestore(&data_lock, flags);
-		return;
-	}
-
-	dss_mgr_write_regs(mgr);
-	dss_mgr_write_regs_extra(mgr);
-
-	mp->updating = true;
-
-	if (!dss_data.irq_enabled && need_isr())
-		dss_register_vsync_isr();
-
-	dispc_mgr_enable_sync(mgr->id);
-
-	spin_unlock_irqrestore(&data_lock, flags);
-}
-
-static void dss_apply_irq_handler(void *data, u32 mask);
-
-static void dss_register_vsync_isr(void)
-{
-	const int num_mgrs = dss_feat_get_num_mgrs();
-	u32 mask;
-	int r, i;
-
-	mask = 0;
-	for (i = 0; i < num_mgrs; ++i)
-		mask |= dispc_mgr_get_vsync_irq(i);
-
-	for (i = 0; i < num_mgrs; ++i)
-		mask |= dispc_mgr_get_framedone_irq(i);
-
-	r = omap_dispc_register_isr(dss_apply_irq_handler, NULL, mask);
-	WARN_ON(r);
-
-	dss_data.irq_enabled = true;
-}
-
-static void dss_unregister_vsync_isr(void)
-{
-	const int num_mgrs = dss_feat_get_num_mgrs();
-	u32 mask;
-	int r, i;
-
-	mask = 0;
-	for (i = 0; i < num_mgrs; ++i)
-		mask |= dispc_mgr_get_vsync_irq(i);
-
-	for (i = 0; i < num_mgrs; ++i)
-		mask |= dispc_mgr_get_framedone_irq(i);
-
-	r = omap_dispc_unregister_isr(dss_apply_irq_handler, NULL, mask);
-	WARN_ON(r);
-
-	dss_data.irq_enabled = false;
-}
-
-static void dss_apply_irq_handler(void *data, u32 mask)
-{
-	const int num_mgrs = dss_feat_get_num_mgrs();
-	int i;
-	bool extra_updating;
-
-	spin_lock(&data_lock);
-
-	/* clear busy, updating flags, shadow_dirty flags */
-	for (i = 0; i < num_mgrs; i++) {
-		struct omap_overlay_manager *mgr;
-		struct mgr_priv_data *mp;
-
-		mgr = omap_dss_get_overlay_manager(i);
-		mp = get_mgr_priv(mgr);
-
-		if (!mp->enabled)
-			continue;
-
-		mp->updating = dispc_mgr_is_enabled(i);
-
-		if (!mgr_manual_update(mgr)) {
-			bool was_busy = mp->busy;
-			mp->busy = dispc_mgr_go_busy(i);
-
-			if (was_busy && !mp->busy)
-				mgr_clear_shadow_dirty(mgr);
-		}
-	}
-
-	dss_write_regs();
-	dss_set_go_bits();
-
-	extra_updating = extra_info_update_ongoing();
-	if (!extra_updating)
-		complete_all(&extra_updated_completion);
-
-	/* call framedone handlers for manual update displays */
-	for (i = 0; i < num_mgrs; i++) {
-		struct omap_overlay_manager *mgr;
-		struct mgr_priv_data *mp;
-
-		mgr = omap_dss_get_overlay_manager(i);
-		mp = get_mgr_priv(mgr);
-
-		if (!mgr_manual_update(mgr) || !mp->framedone_handler)
-			continue;
-
-		if (mask & dispc_mgr_get_framedone_irq(i))
-			mp->framedone_handler(mp->framedone_handler_data);
-	}
-
-	if (!need_isr())
-		dss_unregister_vsync_isr();
-
-	spin_unlock(&data_lock);
-}
-
-static void omap_dss_mgr_apply_ovl(struct omap_overlay *ovl)
-{
-	struct ovl_priv_data *op;
-
-	op = get_ovl_priv(ovl);
-
-	if (!op->user_info_dirty)
-		return;
-
-	op->user_info_dirty = false;
-	op->info_dirty = true;
-	op->info = op->user_info;
-}
-
-static void omap_dss_mgr_apply_mgr(struct omap_overlay_manager *mgr)
-{
-	struct mgr_priv_data *mp;
-
-	mp = get_mgr_priv(mgr);
-
-	if (!mp->user_info_dirty)
-		return;
-
-	mp->user_info_dirty = false;
-	mp->info_dirty = true;
-	mp->info = mp->user_info;
-}
-
-static int omap_dss_mgr_apply(struct omap_overlay_manager *mgr)
-{
-	unsigned long flags;
-	struct omap_overlay *ovl;
-	int r;
-
-	DSSDBG("omap_dss_mgr_apply(%s)\n", mgr->name);
-
-	spin_lock_irqsave(&data_lock, flags);
-
-	r = dss_check_settings_apply(mgr);
-	if (r) {
-		spin_unlock_irqrestore(&data_lock, flags);
-		DSSERR("failed to apply settings: illegal configuration.\n");
-		return r;
-	}
-
-	/* Configure overlays */
-	list_for_each_entry(ovl, &mgr->overlays, list)
-		omap_dss_mgr_apply_ovl(ovl);
-
-	/* Configure manager */
-	omap_dss_mgr_apply_mgr(mgr);
-
-	dss_write_regs();
-	dss_set_go_bits();
-
-	spin_unlock_irqrestore(&data_lock, flags);
-
-	return 0;
-}
-
-static void dss_apply_ovl_enable(struct omap_overlay *ovl, bool enable)
-{
-	struct ovl_priv_data *op;
-
-	op = get_ovl_priv(ovl);
-
-	if (op->enabled == enable)
-		return;
-
-	op->enabled = enable;
-	op->extra_info_dirty = true;
-}
-
-static void dss_apply_ovl_fifo_thresholds(struct omap_overlay *ovl,
-		u32 fifo_low, u32 fifo_high)
-{
-	struct ovl_priv_data *op = get_ovl_priv(ovl);
-
-	if (op->fifo_low == fifo_low && op->fifo_high == fifo_high)
-		return;
-
-	op->fifo_low = fifo_low;
-	op->fifo_high = fifo_high;
-	op->extra_info_dirty = true;
-}
-
-static void dss_ovl_setup_fifo(struct omap_overlay *ovl)
-{
-	struct ovl_priv_data *op = get_ovl_priv(ovl);
-	u32 fifo_low, fifo_high;
-	bool use_fifo_merge = false;
-
-	if (!op->enabled && !op->enabling)
-		return;
-
-	dispc_ovl_compute_fifo_thresholds(ovl->id, &fifo_low, &fifo_high,
-			use_fifo_merge, ovl_manual_update(ovl));
-
-	dss_apply_ovl_fifo_thresholds(ovl, fifo_low, fifo_high);
-}
-
-static void dss_mgr_setup_fifos(struct omap_overlay_manager *mgr)
-{
-	struct omap_overlay *ovl;
-	struct mgr_priv_data *mp;
-
-	mp = get_mgr_priv(mgr);
-
-	if (!mp->enabled)
-		return;
-
-	list_for_each_entry(ovl, &mgr->overlays, list)
-		dss_ovl_setup_fifo(ovl);
-}
-
-static void dss_setup_fifos(void)
-{
-	const int num_mgrs = omap_dss_get_num_overlay_managers();
-	struct omap_overlay_manager *mgr;
-	int i;
-
-	for (i = 0; i < num_mgrs; ++i) {
-		mgr = omap_dss_get_overlay_manager(i);
-		dss_mgr_setup_fifos(mgr);
-	}
-}
-
-static int dss_mgr_enable_compat(struct omap_overlay_manager *mgr)
-{
-	struct mgr_priv_data *mp = get_mgr_priv(mgr);
-	unsigned long flags;
-	int r;
-
-	mutex_lock(&apply_lock);
-
-	if (mp->enabled)
-		goto out;
-
-	spin_lock_irqsave(&data_lock, flags);
-
-	mp->enabled = true;
-
-	r = dss_check_settings(mgr);
-	if (r) {
-		DSSERR("failed to enable manager %d: check_settings failed\n",
-				mgr->id);
-		goto err;
-	}
-
-	dss_setup_fifos();
-
-	dss_write_regs();
-	dss_set_go_bits();
-
-	if (!mgr_manual_update(mgr))
-		mp->updating = true;
-
-	if (!dss_data.irq_enabled && need_isr())
-		dss_register_vsync_isr();
-
-	spin_unlock_irqrestore(&data_lock, flags);
-
-	if (!mgr_manual_update(mgr))
-		dispc_mgr_enable_sync(mgr->id);
-
-out:
-	mutex_unlock(&apply_lock);
-
-	return 0;
-
-err:
-	mp->enabled = false;
-	spin_unlock_irqrestore(&data_lock, flags);
-	mutex_unlock(&apply_lock);
-	return r;
-}
-
-static void dss_mgr_disable_compat(struct omap_overlay_manager *mgr)
-{
-	struct mgr_priv_data *mp = get_mgr_priv(mgr);
-	unsigned long flags;
-
-	mutex_lock(&apply_lock);
-
-	if (!mp->enabled)
-		goto out;
-
-	wait_pending_extra_info_updates();
-
-	if (!mgr_manual_update(mgr))
-		dispc_mgr_disable_sync(mgr->id);
-
-	spin_lock_irqsave(&data_lock, flags);
-
-	mp->updating = false;
-	mp->enabled = false;
-
-	spin_unlock_irqrestore(&data_lock, flags);
-
-out:
-	mutex_unlock(&apply_lock);
-}
-
-static int dss_mgr_set_info(struct omap_overlay_manager *mgr,
-		struct omap_overlay_manager_info *info)
-{
-	struct mgr_priv_data *mp = get_mgr_priv(mgr);
-	unsigned long flags;
-	int r;
-
-	r = dss_mgr_simple_check(mgr, info);
-	if (r)
-		return r;
-
-	spin_lock_irqsave(&data_lock, flags);
-
-	mp->user_info = *info;
-	mp->user_info_dirty = true;
-
-	spin_unlock_irqrestore(&data_lock, flags);
-
-	return 0;
-}
-
-static void dss_mgr_get_info(struct omap_overlay_manager *mgr,
-		struct omap_overlay_manager_info *info)
-{
-	struct mgr_priv_data *mp = get_mgr_priv(mgr);
-	unsigned long flags;
-
-	spin_lock_irqsave(&data_lock, flags);
-
-	*info = mp->user_info;
-
-	spin_unlock_irqrestore(&data_lock, flags);
-}
-
-static int dss_mgr_set_output(struct omap_overlay_manager *mgr,
-		struct omap_dss_device *output)
-{
-	int r;
-
-	mutex_lock(&apply_lock);
-
-	if (mgr->output) {
-		DSSERR("manager %s is already connected to an output\n",
-			mgr->name);
-		r = -EINVAL;
-		goto err;
-	}
-
-	if ((mgr->supported_outputs & output->id) == 0) {
-		DSSERR("output does not support manager %s\n",
-			mgr->name);
-		r = -EINVAL;
-		goto err;
-	}
-
-	output->manager = mgr;
-	mgr->output = output;
-
-	mutex_unlock(&apply_lock);
-
-	return 0;
-err:
-	mutex_unlock(&apply_lock);
-	return r;
-}
-
-static int dss_mgr_unset_output(struct omap_overlay_manager *mgr)
-{
-	int r;
-	struct mgr_priv_data *mp = get_mgr_priv(mgr);
-	unsigned long flags;
-
-	mutex_lock(&apply_lock);
-
-	if (!mgr->output) {
-		DSSERR("failed to unset output, output not set\n");
-		r = -EINVAL;
-		goto err;
-	}
-
-	spin_lock_irqsave(&data_lock, flags);
-
-	if (mp->enabled) {
-		DSSERR("output can't be unset when manager is enabled\n");
-		r = -EINVAL;
-		goto err1;
-	}
-
-	spin_unlock_irqrestore(&data_lock, flags);
-
-	mgr->output->manager = NULL;
-	mgr->output = NULL;
-
-	mutex_unlock(&apply_lock);
-
-	return 0;
-err1:
-	spin_unlock_irqrestore(&data_lock, flags);
-err:
-	mutex_unlock(&apply_lock);
-
-	return r;
-}
-
-static void dss_apply_mgr_timings(struct omap_overlay_manager *mgr,
-		const struct omap_video_timings *timings)
-{
-	struct mgr_priv_data *mp = get_mgr_priv(mgr);
-
-	mp->timings = *timings;
-	mp->extra_info_dirty = true;
-}
-
-static void dss_mgr_set_timings_compat(struct omap_overlay_manager *mgr,
-		const struct omap_video_timings *timings)
-{
-	unsigned long flags;
-	struct mgr_priv_data *mp = get_mgr_priv(mgr);
-
-	spin_lock_irqsave(&data_lock, flags);
-
-	if (mp->updating) {
-		DSSERR("cannot set timings for %s: manager needs to be disabled\n",
-			mgr->name);
-		goto out;
-	}
-
-	dss_apply_mgr_timings(mgr, timings);
-out:
-	spin_unlock_irqrestore(&data_lock, flags);
-}
-
-static void dss_apply_mgr_lcd_config(struct omap_overlay_manager *mgr,
-		const struct dss_lcd_mgr_config *config)
-{
-	struct mgr_priv_data *mp = get_mgr_priv(mgr);
-
-	mp->lcd_config = *config;
-	mp->extra_info_dirty = true;
-}
-
-static void dss_mgr_set_lcd_config_compat(struct omap_overlay_manager *mgr,
-		const struct dss_lcd_mgr_config *config)
-{
-	unsigned long flags;
-	struct mgr_priv_data *mp = get_mgr_priv(mgr);
-
-	spin_lock_irqsave(&data_lock, flags);
-
-	if (mp->enabled) {
-		DSSERR("cannot apply lcd config for %s: manager needs to be disabled\n",
-			mgr->name);
-		goto out;
-	}
-
-	dss_apply_mgr_lcd_config(mgr, config);
-out:
-	spin_unlock_irqrestore(&data_lock, flags);
-}
-
-static int dss_ovl_set_info(struct omap_overlay *ovl,
-		struct omap_overlay_info *info)
-{
-	struct ovl_priv_data *op = get_ovl_priv(ovl);
-	unsigned long flags;
-	int r;
-
-	r = dss_ovl_simple_check(ovl, info);
-	if (r)
-		return r;
-
-	spin_lock_irqsave(&data_lock, flags);
-
-	op->user_info = *info;
-	op->user_info_dirty = true;
-
-	spin_unlock_irqrestore(&data_lock, flags);
-
-	return 0;
-}
-
-static void dss_ovl_get_info(struct omap_overlay *ovl,
-		struct omap_overlay_info *info)
-{
-	struct ovl_priv_data *op = get_ovl_priv(ovl);
-	unsigned long flags;
-
-	spin_lock_irqsave(&data_lock, flags);
-
-	*info = op->user_info;
-
-	spin_unlock_irqrestore(&data_lock, flags);
-}
-
-static int dss_ovl_set_manager(struct omap_overlay *ovl,
-		struct omap_overlay_manager *mgr)
-{
-	struct ovl_priv_data *op = get_ovl_priv(ovl);
-	unsigned long flags;
-	int r;
-
-	if (!mgr)
-		return -EINVAL;
-
-	mutex_lock(&apply_lock);
-
-	if (ovl->manager) {
-		DSSERR("overlay '%s' already has a manager '%s'\n",
-				ovl->name, ovl->manager->name);
-		r = -EINVAL;
-		goto err;
-	}
-
-	r = dispc_runtime_get();
-	if (r)
-		goto err;
-
-	spin_lock_irqsave(&data_lock, flags);
-
-	if (op->enabled) {
-		spin_unlock_irqrestore(&data_lock, flags);
-		DSSERR("overlay has to be disabled to change the manager\n");
-		r = -EINVAL;
-		goto err1;
-	}
-
-	dispc_ovl_set_channel_out(ovl->id, mgr->id);
-
-	ovl->manager = mgr;
-	list_add_tail(&ovl->list, &mgr->overlays);
-
-	spin_unlock_irqrestore(&data_lock, flags);
-
-	dispc_runtime_put();
-
-	mutex_unlock(&apply_lock);
-
-	return 0;
-
-err1:
-	dispc_runtime_put();
-err:
-	mutex_unlock(&apply_lock);
-	return r;
-}
-
-static int dss_ovl_unset_manager(struct omap_overlay *ovl)
-{
-	struct ovl_priv_data *op = get_ovl_priv(ovl);
-	unsigned long flags;
-	int r;
-
-	mutex_lock(&apply_lock);
-
-	if (!ovl->manager) {
-		DSSERR("failed to detach overlay: manager not set\n");
-		r = -EINVAL;
-		goto err;
-	}
-
-	spin_lock_irqsave(&data_lock, flags);
-
-	if (op->enabled) {
-		spin_unlock_irqrestore(&data_lock, flags);
-		DSSERR("overlay has to be disabled to unset the manager\n");
-		r = -EINVAL;
-		goto err;
-	}
-
-	spin_unlock_irqrestore(&data_lock, flags);
-
-	/* wait for pending extra_info updates to ensure the ovl is disabled */
-	wait_pending_extra_info_updates();
-
-	/*
-	 * For a manual update display, there is no guarantee that the overlay
-	 * is really disabled in HW, we may need an extra update from this
-	 * manager before the configurations can go in. Return an error if the
-	 * overlay needed an update from the manager.
-	 *
-	 * TODO: Instead of returning an error, try to do a dummy manager update
-	 * here to disable the overlay in hardware. Use the *GATED fields in
-	 * the DISPC_CONFIG registers to do a dummy update.
-	 */
-	spin_lock_irqsave(&data_lock, flags);
-
-	if (ovl_manual_update(ovl) && op->extra_info_dirty) {
-		spin_unlock_irqrestore(&data_lock, flags);
-		DSSERR("need an update to change the manager\n");
-		r = -EINVAL;
-		goto err;
-	}
-
-	ovl->manager = NULL;
-	list_del(&ovl->list);
-
-	spin_unlock_irqrestore(&data_lock, flags);
-
-	mutex_unlock(&apply_lock);
-
-	return 0;
-err:
-	mutex_unlock(&apply_lock);
-	return r;
-}
-
-static bool dss_ovl_is_enabled(struct omap_overlay *ovl)
-{
-	struct ovl_priv_data *op = get_ovl_priv(ovl);
-	unsigned long flags;
-	bool e;
-
-	spin_lock_irqsave(&data_lock, flags);
-
-	e = op->enabled;
-
-	spin_unlock_irqrestore(&data_lock, flags);
-
-	return e;
-}
-
-static int dss_ovl_enable(struct omap_overlay *ovl)
-{
-	struct ovl_priv_data *op = get_ovl_priv(ovl);
-	unsigned long flags;
-	int r;
-
-	mutex_lock(&apply_lock);
-
-	if (op->enabled) {
-		r = 0;
-		goto err1;
-	}
-
-	if (ovl->manager == NULL || ovl->manager->output == NULL) {
-		r = -EINVAL;
-		goto err1;
-	}
-
-	spin_lock_irqsave(&data_lock, flags);
-
-	op->enabling = true;
-
-	r = dss_check_settings(ovl->manager);
-	if (r) {
-		DSSERR("failed to enable overlay %d: check_settings failed\n",
-				ovl->id);
-		goto err2;
-	}
-
-	dss_setup_fifos();
-
-	op->enabling = false;
-	dss_apply_ovl_enable(ovl, true);
-
-	dss_write_regs();
-	dss_set_go_bits();
-
-	spin_unlock_irqrestore(&data_lock, flags);
-
-	mutex_unlock(&apply_lock);
-
-	return 0;
-err2:
-	op->enabling = false;
-	spin_unlock_irqrestore(&data_lock, flags);
-err1:
-	mutex_unlock(&apply_lock);
-	return r;
-}
-
-static int dss_ovl_disable(struct omap_overlay *ovl)
-{
-	struct ovl_priv_data *op = get_ovl_priv(ovl);
-	unsigned long flags;
-	int r;
-
-	mutex_lock(&apply_lock);
-
-	if (!op->enabled) {
-		r = 0;
-		goto err;
-	}
-
-	if (ovl->manager == NULL || ovl->manager->output == NULL) {
-		r = -EINVAL;
-		goto err;
-	}
-
-	spin_lock_irqsave(&data_lock, flags);
-
-	dss_apply_ovl_enable(ovl, false);
-	dss_write_regs();
-	dss_set_go_bits();
-
-	spin_unlock_irqrestore(&data_lock, flags);
-
-	mutex_unlock(&apply_lock);
-
-	return 0;
-
-err:
-	mutex_unlock(&apply_lock);
-	return r;
-}
-
-static int dss_mgr_register_framedone_handler_compat(struct omap_overlay_manager *mgr,
-		void (*handler)(void *), void *data)
-{
-	struct mgr_priv_data *mp = get_mgr_priv(mgr);
-
-	if (mp->framedone_handler)
-		return -EBUSY;
-
-	mp->framedone_handler = handler;
-	mp->framedone_handler_data = data;
-
-	return 0;
-}
-
-static void dss_mgr_unregister_framedone_handler_compat(struct omap_overlay_manager *mgr,
-		void (*handler)(void *), void *data)
-{
-	struct mgr_priv_data *mp = get_mgr_priv(mgr);
-
-	WARN_ON(mp->framedone_handler != handler ||
-			mp->framedone_handler_data != data);
-
-	mp->framedone_handler = NULL;
-	mp->framedone_handler_data = NULL;
-}
-
-static const struct dss_mgr_ops apply_mgr_ops = {
-	.connect = dss_mgr_connect_compat,
-	.disconnect = dss_mgr_disconnect_compat,
-	.start_update = dss_mgr_start_update_compat,
-	.enable = dss_mgr_enable_compat,
-	.disable = dss_mgr_disable_compat,
-	.set_timings = dss_mgr_set_timings_compat,
-	.set_lcd_config = dss_mgr_set_lcd_config_compat,
-	.register_framedone_handler = dss_mgr_register_framedone_handler_compat,
-	.unregister_framedone_handler = dss_mgr_unregister_framedone_handler_compat,
-};
-
-static int compat_refcnt;
-static DEFINE_MUTEX(compat_init_lock);
-
-int omapdss_compat_init(void)
-{
-	struct platform_device *pdev = dss_get_core_pdev();
-	int i, r;
-
-	mutex_lock(&compat_init_lock);
-
-	if (compat_refcnt++ > 0)
-		goto out;
-
-	apply_init_priv();
-
-	dss_init_overlay_managers_sysfs(pdev);
-	dss_init_overlays(pdev);
-
-	for (i = 0; i < omap_dss_get_num_overlay_managers(); i++) {
-		struct omap_overlay_manager *mgr;
-
-		mgr = omap_dss_get_overlay_manager(i);
-
-		mgr->set_output = &dss_mgr_set_output;
-		mgr->unset_output = &dss_mgr_unset_output;
-		mgr->apply = &omap_dss_mgr_apply;
-		mgr->set_manager_info = &dss_mgr_set_info;
-		mgr->get_manager_info = &dss_mgr_get_info;
-		mgr->wait_for_go = &dss_mgr_wait_for_go;
-		mgr->wait_for_vsync = &dss_mgr_wait_for_vsync;
-		mgr->get_device = &dss_mgr_get_device;
-	}
-
-	for (i = 0; i < omap_dss_get_num_overlays(); i++) {
-		struct omap_overlay *ovl = omap_dss_get_overlay(i);
-
-		ovl->is_enabled = &dss_ovl_is_enabled;
-		ovl->enable = &dss_ovl_enable;
-		ovl->disable = &dss_ovl_disable;
-		ovl->set_manager = &dss_ovl_set_manager;
-		ovl->unset_manager = &dss_ovl_unset_manager;
-		ovl->set_overlay_info = &dss_ovl_set_info;
-		ovl->get_overlay_info = &dss_ovl_get_info;
-		ovl->wait_for_go = &dss_mgr_wait_for_go_ovl;
-		ovl->get_device = &dss_ovl_get_device;
-	}
-
-	r = dss_install_mgr_ops(&apply_mgr_ops);
-	if (r)
-		goto err_mgr_ops;
-
-	r = display_init_sysfs(pdev);
-	if (r)
-		goto err_disp_sysfs;
-
-	dispc_runtime_get();
-
-	r = dss_dispc_initialize_irq();
-	if (r)
-		goto err_init_irq;
-
-	dispc_runtime_put();
-
-out:
-	mutex_unlock(&compat_init_lock);
-
-	return 0;
-
-err_init_irq:
-	dispc_runtime_put();
-	display_uninit_sysfs(pdev);
-
-err_disp_sysfs:
-	dss_uninstall_mgr_ops();
-
-err_mgr_ops:
-	dss_uninit_overlay_managers_sysfs(pdev);
-	dss_uninit_overlays(pdev);
-
-	compat_refcnt--;
-
-	mutex_unlock(&compat_init_lock);
-
-	return r;
-}
-EXPORT_SYMBOL(omapdss_compat_init);
-
-void omapdss_compat_uninit(void)
-{
-	struct platform_device *pdev = dss_get_core_pdev();
-
-	mutex_lock(&compat_init_lock);
-
-	if (--compat_refcnt > 0)
-		goto out;
-
-	dss_dispc_uninitialize_irq();
-
-	display_uninit_sysfs(pdev);
-
-	dss_uninstall_mgr_ops();
-
-	dss_uninit_overlay_managers_sysfs(pdev);
-	dss_uninit_overlays(pdev);
-out:
-	mutex_unlock(&compat_init_lock);
-}
-EXPORT_SYMBOL(omapdss_compat_uninit);
diff --git a/drivers/gpu/drm/omapdrm/dss/core.c b/drivers/gpu/drm/omapdrm/dss/core.c
index 54eeb507f9b3..7e4e5bebabbe 100644
--- a/drivers/gpu/drm/omapdrm/dss/core.c
+++ b/drivers/gpu/drm/omapdrm/dss/core.c
@@ -165,32 +165,20 @@ int dss_debugfs_create_file(const char *name, void (*write)(struct seq_file *))
 #endif /* CONFIG_OMAP2_DSS_DEBUGFS */
 
 /* PLATFORM DEVICE */
-static int omap_dss_pm_notif(struct notifier_block *b, unsigned long v, void *d)
+
+static void dss_disable_all_devices(void)
 {
-	DSSDBG("pm notif %lu\n", v);
-
-	switch (v) {
-	case PM_SUSPEND_PREPARE:
-	case PM_HIBERNATION_PREPARE:
-	case PM_RESTORE_PREPARE:
-		DSSDBG("suspending displays\n");
-		return dss_suspend_all_devices();
-
-	case PM_POST_SUSPEND:
-	case PM_POST_HIBERNATION:
-	case PM_POST_RESTORE:
-		DSSDBG("resuming displays\n");
-		return dss_resume_all_devices();
-
-	default:
-		return 0;
+	struct omap_dss_device *dssdev = NULL;
+
+	for_each_dss_dev(dssdev) {
+		if (!dssdev->driver)
+			continue;
+
+		if (dssdev->state == OMAP_DSS_DISPLAY_ACTIVE)
+			dssdev->driver->disable(dssdev);
 	}
 }
 
-static struct notifier_block omap_dss_pm_notif_block = {
-	.notifier_call = omap_dss_pm_notif,
-};
-
 static int __init omap_dss_probe(struct platform_device *pdev)
 {
 	struct omap_dss_board_info *pdata = pdev->dev.platform_data;
@@ -211,8 +199,6 @@ static int __init omap_dss_probe(struct platform_device *pdev)
 	else if (pdata->default_device)
 		core.default_display_name = pdata->default_device->name;
 
-	register_pm_notifier(&omap_dss_pm_notif_block);
-
 	return 0;
 
 err_debugfs:
@@ -222,8 +208,6 @@ err_debugfs:
 
 static int omap_dss_remove(struct platform_device *pdev)
 {
-	unregister_pm_notifier(&omap_dss_pm_notif_block);
-
 	dss_uninitialize_debugfs();
 
 	return 0;
diff --git a/drivers/gpu/drm/omapdrm/dss/dispc-compat.c b/drivers/gpu/drm/omapdrm/dss/dispc-compat.c
deleted file mode 100644
index 0918b3bfe82a..000000000000
--- a/drivers/gpu/drm/omapdrm/dss/dispc-compat.c
+++ /dev/null
@@ -1,667 +0,0 @@
-/*
- * Copyright (C) 2012 Texas Instruments
- * Author: Tomi Valkeinen <tomi.valkeinen@ti.com>
- *
- * This program is free software; you can redistribute it and/or modify it
- * under the terms of the GNU General Public License version 2 as published by
- * the Free Software Foundation.
- *
- * This program is distributed in the hope that it will be useful, but WITHOUT
- * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
- * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
- * more details.
- *
- * You should have received a copy of the GNU General Public License along with
- * this program.  If not, see <http://www.gnu.org/licenses/>.
- */
-
-#define DSS_SUBSYS_NAME "APPLY"
-
-#include <linux/kernel.h>
-#include <linux/module.h>
-#include <linux/slab.h>
-#include <linux/spinlock.h>
-#include <linux/jiffies.h>
-#include <linux/delay.h>
-#include <linux/interrupt.h>
-#include <linux/seq_file.h>
-
-#include <video/omapdss.h>
-
-#include "dss.h"
-#include "dss_features.h"
-#include "dispc-compat.h"
-
-#define DISPC_IRQ_MASK_ERROR            (DISPC_IRQ_GFX_FIFO_UNDERFLOW | \
-					 DISPC_IRQ_OCP_ERR | \
-					 DISPC_IRQ_VID1_FIFO_UNDERFLOW | \
-					 DISPC_IRQ_VID2_FIFO_UNDERFLOW | \
-					 DISPC_IRQ_SYNC_LOST | \
-					 DISPC_IRQ_SYNC_LOST_DIGIT)
-
-#define DISPC_MAX_NR_ISRS		8
-
-struct omap_dispc_isr_data {
-	omap_dispc_isr_t	isr;
-	void			*arg;
-	u32			mask;
-};
-
-struct dispc_irq_stats {
-	unsigned long last_reset;
-	unsigned irq_count;
-	unsigned irqs[32];
-};
-
-static struct {
-	spinlock_t irq_lock;
-	u32 irq_error_mask;
-	struct omap_dispc_isr_data registered_isr[DISPC_MAX_NR_ISRS];
-	u32 error_irqs;
-	struct work_struct error_work;
-
-#ifdef CONFIG_OMAP2_DSS_COLLECT_IRQ_STATS
-	spinlock_t irq_stats_lock;
-	struct dispc_irq_stats irq_stats;
-#endif
-} dispc_compat;
-
-
-#ifdef CONFIG_OMAP2_DSS_COLLECT_IRQ_STATS
-static void dispc_dump_irqs(struct seq_file *s)
-{
-	unsigned long flags;
-	struct dispc_irq_stats stats;
-
-	spin_lock_irqsave(&dispc_compat.irq_stats_lock, flags);
-
-	stats = dispc_compat.irq_stats;
-	memset(&dispc_compat.irq_stats, 0, sizeof(dispc_compat.irq_stats));
-	dispc_compat.irq_stats.last_reset = jiffies;
-
-	spin_unlock_irqrestore(&dispc_compat.irq_stats_lock, flags);
-
-	seq_printf(s, "period %u ms\n",
-			jiffies_to_msecs(jiffies - stats.last_reset));
-
-	seq_printf(s, "irqs %d\n", stats.irq_count);
-#define PIS(x) \
-	seq_printf(s, "%-20s %10d\n", #x, stats.irqs[ffs(DISPC_IRQ_##x)-1]);
-
-	PIS(FRAMEDONE);
-	PIS(VSYNC);
-	PIS(EVSYNC_EVEN);
-	PIS(EVSYNC_ODD);
-	PIS(ACBIAS_COUNT_STAT);
-	PIS(PROG_LINE_NUM);
-	PIS(GFX_FIFO_UNDERFLOW);
-	PIS(GFX_END_WIN);
-	PIS(PAL_GAMMA_MASK);
-	PIS(OCP_ERR);
-	PIS(VID1_FIFO_UNDERFLOW);
-	PIS(VID1_END_WIN);
-	PIS(VID2_FIFO_UNDERFLOW);
-	PIS(VID2_END_WIN);
-	if (dss_feat_get_num_ovls() > 3) {
-		PIS(VID3_FIFO_UNDERFLOW);
-		PIS(VID3_END_WIN);
-	}
-	PIS(SYNC_LOST);
-	PIS(SYNC_LOST_DIGIT);
-	PIS(WAKEUP);
-	if (dss_has_feature(FEAT_MGR_LCD2)) {
-		PIS(FRAMEDONE2);
-		PIS(VSYNC2);
-		PIS(ACBIAS_COUNT_STAT2);
-		PIS(SYNC_LOST2);
-	}
-	if (dss_has_feature(FEAT_MGR_LCD3)) {
-		PIS(FRAMEDONE3);
-		PIS(VSYNC3);
-		PIS(ACBIAS_COUNT_STAT3);
-		PIS(SYNC_LOST3);
-	}
-#undef PIS
-}
-#endif
-
-/* dispc.irq_lock has to be locked by the caller */
-static void _omap_dispc_set_irqs(void)
-{
-	u32 mask;
-	int i;
-	struct omap_dispc_isr_data *isr_data;
-
-	mask = dispc_compat.irq_error_mask;
-
-	for (i = 0; i < DISPC_MAX_NR_ISRS; i++) {
-		isr_data = &dispc_compat.registered_isr[i];
-
-		if (isr_data->isr == NULL)
-			continue;
-
-		mask |= isr_data->mask;
-	}
-
-	dispc_write_irqenable(mask);
-}
-
-int omap_dispc_register_isr(omap_dispc_isr_t isr, void *arg, u32 mask)
-{
-	int i;
-	int ret;
-	unsigned long flags;
-	struct omap_dispc_isr_data *isr_data;
-
-	if (isr == NULL)
-		return -EINVAL;
-
-	spin_lock_irqsave(&dispc_compat.irq_lock, flags);
-
-	/* check for duplicate entry */
-	for (i = 0; i < DISPC_MAX_NR_ISRS; i++) {
-		isr_data = &dispc_compat.registered_isr[i];
-		if (isr_data->isr == isr && isr_data->arg == arg &&
-				isr_data->mask == mask) {
-			ret = -EINVAL;
-			goto err;
-		}
-	}
-
-	isr_data = NULL;
-	ret = -EBUSY;
-
-	for (i = 0; i < DISPC_MAX_NR_ISRS; i++) {
-		isr_data = &dispc_compat.registered_isr[i];
-
-		if (isr_data->isr != NULL)
-			continue;
-
-		isr_data->isr = isr;
-		isr_data->arg = arg;
-		isr_data->mask = mask;
-		ret = 0;
-
-		break;
-	}
-
-	if (ret)
-		goto err;
-
-	_omap_dispc_set_irqs();
-
-	spin_unlock_irqrestore(&dispc_compat.irq_lock, flags);
-
-	return 0;
-err:
-	spin_unlock_irqrestore(&dispc_compat.irq_lock, flags);
-
-	return ret;
-}
-EXPORT_SYMBOL(omap_dispc_register_isr);
-
-int omap_dispc_unregister_isr(omap_dispc_isr_t isr, void *arg, u32 mask)
-{
-	int i;
-	unsigned long flags;
-	int ret = -EINVAL;
-	struct omap_dispc_isr_data *isr_data;
-
-	spin_lock_irqsave(&dispc_compat.irq_lock, flags);
-
-	for (i = 0; i < DISPC_MAX_NR_ISRS; i++) {
-		isr_data = &dispc_compat.registered_isr[i];
-		if (isr_data->isr != isr || isr_data->arg != arg ||
-				isr_data->mask != mask)
-			continue;
-
-		/* found the correct isr */
-
-		isr_data->isr = NULL;
-		isr_data->arg = NULL;
-		isr_data->mask = 0;
-
-		ret = 0;
-		break;
-	}
-
-	if (ret == 0)
-		_omap_dispc_set_irqs();
-
-	spin_unlock_irqrestore(&dispc_compat.irq_lock, flags);
-
-	return ret;
-}
-EXPORT_SYMBOL(omap_dispc_unregister_isr);
-
-static void print_irq_status(u32 status)
-{
-	if ((status & dispc_compat.irq_error_mask) == 0)
-		return;
-
-#define PIS(x) (status & DISPC_IRQ_##x) ? (#x " ") : ""
-
-	pr_debug("DISPC IRQ: 0x%x: %s%s%s%s%s%s%s%s%s\n",
-		status,
-		PIS(OCP_ERR),
-		PIS(GFX_FIFO_UNDERFLOW),
-		PIS(VID1_FIFO_UNDERFLOW),
-		PIS(VID2_FIFO_UNDERFLOW),
-		dss_feat_get_num_ovls() > 3 ? PIS(VID3_FIFO_UNDERFLOW) : "",
-		PIS(SYNC_LOST),
-		PIS(SYNC_LOST_DIGIT),
-		dss_has_feature(FEAT_MGR_LCD2) ? PIS(SYNC_LOST2) : "",
-		dss_has_feature(FEAT_MGR_LCD3) ? PIS(SYNC_LOST3) : "");
-#undef PIS
-}
-
-/* Called from dss.c. Note that we don't touch clocks here,
- * but we presume they are on because we got an IRQ. However,
- * an irq handler may turn the clocks off, so we may not have
- * clock later in the function. */
-static irqreturn_t omap_dispc_irq_handler(int irq, void *arg)
-{
-	int i;
-	u32 irqstatus, irqenable;
-	u32 handledirqs = 0;
-	u32 unhandled_errors;
-	struct omap_dispc_isr_data *isr_data;
-	struct omap_dispc_isr_data registered_isr[DISPC_MAX_NR_ISRS];
-
-	spin_lock(&dispc_compat.irq_lock);
-
-	irqstatus = dispc_read_irqstatus();
-	irqenable = dispc_read_irqenable();
-
-	/* IRQ is not for us */
-	if (!(irqstatus & irqenable)) {
-		spin_unlock(&dispc_compat.irq_lock);
-		return IRQ_NONE;
-	}
-
-#ifdef CONFIG_OMAP2_DSS_COLLECT_IRQ_STATS
-	spin_lock(&dispc_compat.irq_stats_lock);
-	dispc_compat.irq_stats.irq_count++;
-	dss_collect_irq_stats(irqstatus, dispc_compat.irq_stats.irqs);
-	spin_unlock(&dispc_compat.irq_stats_lock);
-#endif
-
-	print_irq_status(irqstatus);
-
-	/* Ack the interrupt. Do it here before clocks are possibly turned
-	 * off */
-	dispc_clear_irqstatus(irqstatus);
-	/* flush posted write */
-	dispc_read_irqstatus();
-
-	/* make a copy and unlock, so that isrs can unregister
-	 * themselves */
-	memcpy(registered_isr, dispc_compat.registered_isr,
-			sizeof(registered_isr));
-
-	spin_unlock(&dispc_compat.irq_lock);
-
-	for (i = 0; i < DISPC_MAX_NR_ISRS; i++) {
-		isr_data = &registered_isr[i];
-
-		if (!isr_data->isr)
-			continue;
-
-		if (isr_data->mask & irqstatus) {
-			isr_data->isr(isr_data->arg, irqstatus);
-			handledirqs |= isr_data->mask;
-		}
-	}
-
-	spin_lock(&dispc_compat.irq_lock);
-
-	unhandled_errors = irqstatus & ~handledirqs & dispc_compat.irq_error_mask;
-
-	if (unhandled_errors) {
-		dispc_compat.error_irqs |= unhandled_errors;
-
-		dispc_compat.irq_error_mask &= ~unhandled_errors;
-		_omap_dispc_set_irqs();
-
-		schedule_work(&dispc_compat.error_work);
-	}
-
-	spin_unlock(&dispc_compat.irq_lock);
-
-	return IRQ_HANDLED;
-}
-
-static void dispc_error_worker(struct work_struct *work)
-{
-	int i;
-	u32 errors;
-	unsigned long flags;
-	static const unsigned fifo_underflow_bits[] = {
-		DISPC_IRQ_GFX_FIFO_UNDERFLOW,
-		DISPC_IRQ_VID1_FIFO_UNDERFLOW,
-		DISPC_IRQ_VID2_FIFO_UNDERFLOW,
-		DISPC_IRQ_VID3_FIFO_UNDERFLOW,
-	};
-
-	spin_lock_irqsave(&dispc_compat.irq_lock, flags);
-	errors = dispc_compat.error_irqs;
-	dispc_compat.error_irqs = 0;
-	spin_unlock_irqrestore(&dispc_compat.irq_lock, flags);
-
-	dispc_runtime_get();
-
-	for (i = 0; i < omap_dss_get_num_overlays(); ++i) {
-		struct omap_overlay *ovl;
-		unsigned bit;
-
-		ovl = omap_dss_get_overlay(i);
-		bit = fifo_underflow_bits[i];
-
-		if (bit & errors) {
-			DSSERR("FIFO UNDERFLOW on %s, disabling the overlay\n",
-					ovl->name);
-			ovl->disable(ovl);
-			msleep(50);
-		}
-	}
-
-	for (i = 0; i < omap_dss_get_num_overlay_managers(); ++i) {
-		struct omap_overlay_manager *mgr;
-		unsigned bit;
-
-		mgr = omap_dss_get_overlay_manager(i);
-		bit = dispc_mgr_get_sync_lost_irq(i);
-
-		if (bit & errors) {
-			int j;
-
-			DSSERR("SYNC_LOST on channel %s, restarting the output "
-					"with video overlays disabled\n",
-					mgr->name);
-
-			dss_mgr_disable(mgr);
-
-			for (j = 0; j < omap_dss_get_num_overlays(); ++j) {
-				struct omap_overlay *ovl;
-				ovl = omap_dss_get_overlay(j);
-
-				if (ovl->id != OMAP_DSS_GFX &&
-						ovl->manager == mgr)
-					ovl->disable(ovl);
-			}
-
-			dss_mgr_enable(mgr);
-		}
-	}
-
-	if (errors & DISPC_IRQ_OCP_ERR) {
-		DSSERR("OCP_ERR\n");
-		for (i = 0; i < omap_dss_get_num_overlay_managers(); ++i) {
-			struct omap_overlay_manager *mgr;
-
-			mgr = omap_dss_get_overlay_manager(i);
-			dss_mgr_disable(mgr);
-		}
-	}
-
-	spin_lock_irqsave(&dispc_compat.irq_lock, flags);
-	dispc_compat.irq_error_mask |= errors;
-	_omap_dispc_set_irqs();
-	spin_unlock_irqrestore(&dispc_compat.irq_lock, flags);
-
-	dispc_runtime_put();
-}
-
-int dss_dispc_initialize_irq(void)
-{
-	int r;
-
-#ifdef CONFIG_OMAP2_DSS_COLLECT_IRQ_STATS
-	spin_lock_init(&dispc_compat.irq_stats_lock);
-	dispc_compat.irq_stats.last_reset = jiffies;
-	dss_debugfs_create_file("dispc_irq", dispc_dump_irqs);
-#endif
-
-	spin_lock_init(&dispc_compat.irq_lock);
-
-	memset(dispc_compat.registered_isr, 0,
-			sizeof(dispc_compat.registered_isr));
-
-	dispc_compat.irq_error_mask = DISPC_IRQ_MASK_ERROR;
-	if (dss_has_feature(FEAT_MGR_LCD2))
-		dispc_compat.irq_error_mask |= DISPC_IRQ_SYNC_LOST2;
-	if (dss_has_feature(FEAT_MGR_LCD3))
-		dispc_compat.irq_error_mask |= DISPC_IRQ_SYNC_LOST3;
-	if (dss_feat_get_num_ovls() > 3)
-		dispc_compat.irq_error_mask |= DISPC_IRQ_VID3_FIFO_UNDERFLOW;
-
-	/*
-	 * there's SYNC_LOST_DIGIT waiting after enabling the DSS,
-	 * so clear it
-	 */
-	dispc_clear_irqstatus(dispc_read_irqstatus());
-
-	INIT_WORK(&dispc_compat.error_work, dispc_error_worker);
-
-	_omap_dispc_set_irqs();
-
-	r = dispc_request_irq(omap_dispc_irq_handler, &dispc_compat);
-	if (r) {
-		DSSERR("dispc_request_irq failed\n");
-		return r;
-	}
-
-	return 0;
-}
-
-void dss_dispc_uninitialize_irq(void)
-{
-	dispc_free_irq(&dispc_compat);
-}
-
-static void dispc_mgr_disable_isr(void *data, u32 mask)
-{
-	struct completion *compl = data;
-	complete(compl);
-}
-
-static void dispc_mgr_enable_lcd_out(enum omap_channel channel)
-{
-	dispc_mgr_enable(channel, true);
-}
-
-static void dispc_mgr_disable_lcd_out(enum omap_channel channel)
-{
-	DECLARE_COMPLETION_ONSTACK(framedone_compl);
-	int r;
-	u32 irq;
-
-	if (!dispc_mgr_is_enabled(channel))
-		return;
-
-	/*
-	 * When we disable LCD output, we need to wait for FRAMEDONE to know
-	 * that DISPC has finished with the LCD output.
-	 */
-
-	irq = dispc_mgr_get_framedone_irq(channel);
-
-	r = omap_dispc_register_isr(dispc_mgr_disable_isr, &framedone_compl,
-			irq);
-	if (r)
-		DSSERR("failed to register FRAMEDONE isr\n");
-
-	dispc_mgr_enable(channel, false);
-
-	/* if we couldn't register for framedone, just sleep and exit */
-	if (r) {
-		msleep(100);
-		return;
-	}
-
-	if (!wait_for_completion_timeout(&framedone_compl,
-				msecs_to_jiffies(100)))
-		DSSERR("timeout waiting for FRAME DONE\n");
-
-	r = omap_dispc_unregister_isr(dispc_mgr_disable_isr, &framedone_compl,
-			irq);
-	if (r)
-		DSSERR("failed to unregister FRAMEDONE isr\n");
-}
-
-static void dispc_digit_out_enable_isr(void *data, u32 mask)
-{
-	struct completion *compl = data;
-
-	/* ignore any sync lost interrupts */
-	if (mask & (DISPC_IRQ_EVSYNC_EVEN | DISPC_IRQ_EVSYNC_ODD))
-		complete(compl);
-}
-
-static void dispc_mgr_enable_digit_out(void)
-{
-	DECLARE_COMPLETION_ONSTACK(vsync_compl);
-	int r;
-	u32 irq_mask;
-
-	if (dispc_mgr_is_enabled(OMAP_DSS_CHANNEL_DIGIT))
-		return;
-
-	/*
-	 * Digit output produces some sync lost interrupts during the first
-	 * frame when enabling. Those need to be ignored, so we register for the
-	 * sync lost irq to prevent the error handler from triggering.
-	 */
-
-	irq_mask = dispc_mgr_get_vsync_irq(OMAP_DSS_CHANNEL_DIGIT) |
-		dispc_mgr_get_sync_lost_irq(OMAP_DSS_CHANNEL_DIGIT);
-
-	r = omap_dispc_register_isr(dispc_digit_out_enable_isr, &vsync_compl,
-			irq_mask);
-	if (r) {
-		DSSERR("failed to register %x isr\n", irq_mask);
-		return;
-	}
-
-	dispc_mgr_enable(OMAP_DSS_CHANNEL_DIGIT, true);
-
-	/* wait for the first evsync */
-	if (!wait_for_completion_timeout(&vsync_compl, msecs_to_jiffies(100)))
-		DSSERR("timeout waiting for digit out to start\n");
-
-	r = omap_dispc_unregister_isr(dispc_digit_out_enable_isr, &vsync_compl,
-			irq_mask);
-	if (r)
-		DSSERR("failed to unregister %x isr\n", irq_mask);
-}
-
-static void dispc_mgr_disable_digit_out(void)
-{
-	DECLARE_COMPLETION_ONSTACK(framedone_compl);
-	int r, i;
-	u32 irq_mask;
-	int num_irqs;
-
-	if (!dispc_mgr_is_enabled(OMAP_DSS_CHANNEL_DIGIT))
-		return;
-
-	/*
-	 * When we disable the digit output, we need to wait for FRAMEDONE to
-	 * know that DISPC has finished with the output.
-	 */
-
-	irq_mask = dispc_mgr_get_framedone_irq(OMAP_DSS_CHANNEL_DIGIT);
-	num_irqs = 1;
-
-	if (!irq_mask) {
-		/*
-		 * omap 2/3 don't have framedone irq for TV, so we need to use
-		 * vsyncs for this.
-		 */
-
-		irq_mask = dispc_mgr_get_vsync_irq(OMAP_DSS_CHANNEL_DIGIT);
-		/*
-		 * We need to wait for both even and odd vsyncs. Note that this
-		 * is not totally reliable, as we could get a vsync interrupt
-		 * before we disable the output, which leads to timeout in the
-		 * wait_for_completion.
-		 */
-		num_irqs = 2;
-	}
-
-	r = omap_dispc_register_isr(dispc_mgr_disable_isr, &framedone_compl,
-			irq_mask);
-	if (r)
-		DSSERR("failed to register %x isr\n", irq_mask);
-
-	dispc_mgr_enable(OMAP_DSS_CHANNEL_DIGIT, false);
-
-	/* if we couldn't register the irq, just sleep and exit */
-	if (r) {
-		msleep(100);
-		return;
-	}
-
-	for (i = 0; i < num_irqs; ++i) {
-		if (!wait_for_completion_timeout(&framedone_compl,
-					msecs_to_jiffies(100)))
-			DSSERR("timeout waiting for digit out to stop\n");
-	}
-
-	r = omap_dispc_unregister_isr(dispc_mgr_disable_isr, &framedone_compl,
-			irq_mask);
-	if (r)
-		DSSERR("failed to unregister %x isr\n", irq_mask);
-}
-
-void dispc_mgr_enable_sync(enum omap_channel channel)
-{
-	if (dss_mgr_is_lcd(channel))
-		dispc_mgr_enable_lcd_out(channel);
-	else if (channel == OMAP_DSS_CHANNEL_DIGIT)
-		dispc_mgr_enable_digit_out();
-	else
-		WARN_ON(1);
-}
-
-void dispc_mgr_disable_sync(enum omap_channel channel)
-{
-	if (dss_mgr_is_lcd(channel))
-		dispc_mgr_disable_lcd_out(channel);
-	else if (channel == OMAP_DSS_CHANNEL_DIGIT)
-		dispc_mgr_disable_digit_out();
-	else
-		WARN_ON(1);
-}
-
-static inline void dispc_irq_wait_handler(void *data, u32 mask)
-{
-	complete((struct completion *)data);
-}
-
-int omap_dispc_wait_for_irq_interruptible_timeout(u32 irqmask,
-		unsigned long timeout)
-{
-
-	int r;
-	DECLARE_COMPLETION_ONSTACK(completion);
-
-	r = omap_dispc_register_isr(dispc_irq_wait_handler, &completion,
-			irqmask);
-
-	if (r)
-		return r;
-
-	timeout = wait_for_completion_interruptible_timeout(&completion,
-			timeout);
-
-	omap_dispc_unregister_isr(dispc_irq_wait_handler, &completion, irqmask);
-
-	if (timeout == 0)
-		return -ETIMEDOUT;
-
-	if (timeout == -ERESTARTSYS)
-		return -ERESTARTSYS;
-
-	return 0;
-}
diff --git a/drivers/gpu/drm/omapdrm/dss/dispc-compat.h b/drivers/gpu/drm/omapdrm/dss/dispc-compat.h
deleted file mode 100644
index 14a69b3d4fb0..000000000000
--- a/drivers/gpu/drm/omapdrm/dss/dispc-compat.h
+++ /dev/null
@@ -1,30 +0,0 @@
-/*
- * Copyright (C) 2012 Texas Instruments
- * Author: Tomi Valkeinen <tomi.valkeinen@ti.com>
- *
- * This program is free software; you can redistribute it and/or modify it
- * under the terms of the GNU General Public License version 2 as published by
- * the Free Software Foundation.
- *
- * This program is distributed in the hope that it will be useful, but WITHOUT
- * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
- * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
- * more details.
- *
- * You should have received a copy of the GNU General Public License along with
- * this program.  If not, see <http://www.gnu.org/licenses/>.
- */
-
-#ifndef __OMAP2_DSS_DISPC_COMPAT_H
-#define __OMAP2_DSS_DISPC_COMPAT_H
-
-void dispc_mgr_enable_sync(enum omap_channel channel);
-void dispc_mgr_disable_sync(enum omap_channel channel);
-
-int omap_dispc_wait_for_irq_interruptible_timeout(u32 irqmask,
-		unsigned long timeout);
-
-int dss_dispc_initialize_irq(void);
-void dss_dispc_uninitialize_irq(void);
-
-#endif
diff --git a/drivers/gpu/drm/omapdrm/dss/dispc.c b/drivers/gpu/drm/omapdrm/dss/dispc.c
index 6b50476ec669..f83608b69e68 100644
--- a/drivers/gpu/drm/omapdrm/dss/dispc.c
+++ b/drivers/gpu/drm/omapdrm/dss/dispc.c
@@ -104,6 +104,15 @@ struct dispc_features {
 	bool supports_sync_align:1;
 
 	bool has_writeback:1;
+
+	bool supports_double_pixel:1;
+
+	/*
+	 * Field order for VENC is different than HDMI. We should handle this in
+	 * some intelligent manner, but as the SoCs have either HDMI or VENC,
+	 * never both, we can just use this flag for now.
+	 */
+	bool reverse_ilace_field_order:1;
 };
 
 #define DISPC_MAX_NR_FIFOS 5
@@ -2552,47 +2561,6 @@ static int dispc_ovl_calc_scaling(unsigned long pclk, unsigned long lclk,
 	return 0;
 }
 
-int dispc_ovl_check(enum omap_plane plane, enum omap_channel channel,
-		const struct omap_overlay_info *oi,
-		const struct omap_video_timings *timings,
-		int *x_predecim, int *y_predecim)
-{
-	enum omap_overlay_caps caps = dss_feat_get_overlay_caps(plane);
-	bool five_taps = true;
-	bool fieldmode = false;
-	u16 in_height = oi->height;
-	u16 in_width = oi->width;
-	bool ilace = timings->interlace;
-	u16 out_width, out_height;
-	int pos_x = oi->pos_x;
-	unsigned long pclk = dispc_mgr_pclk_rate(channel);
-	unsigned long lclk = dispc_mgr_lclk_rate(channel);
-
-	out_width = oi->out_width == 0 ? oi->width : oi->out_width;
-	out_height = oi->out_height == 0 ? oi->height : oi->out_height;
-
-	if (ilace && oi->height == out_height)
-		fieldmode = true;
-
-	if (ilace) {
-		if (fieldmode)
-			in_height /= 2;
-		out_height /= 2;
-
-		DSSDBG("adjusting for ilace: height %d, out_height %d\n",
-				in_height, out_height);
-	}
-
-	if (!dss_feat_color_mode_supported(plane, oi->color_mode))
-		return -EINVAL;
-
-	return dispc_ovl_calc_scaling(pclk, lclk, caps, timings, in_width,
-			in_height, out_width, out_height, oi->color_mode,
-			&five_taps, x_predecim, y_predecim, pos_x,
-			oi->rotation_type, false);
-}
-EXPORT_SYMBOL(dispc_ovl_check);
-
 static int dispc_ovl_setup_common(enum omap_plane plane,
 		enum omap_overlay_caps caps, u32 paddr, u32 p_uv_addr,
 		u16 screen_width, int pos_x, int pos_y, u16 width, u16 height,
@@ -2747,6 +2715,9 @@ static int dispc_ovl_setup_common(enum omap_plane plane,
 
 	dispc_ovl_configure_burst_type(plane, rotation_type);
 
+	if (dispc.feat->reverse_ilace_field_order)
+		swap(offset0, offset1);
+
 	dispc_ovl_set_ba0(plane, paddr + offset0);
 	dispc_ovl_set_ba1(plane, paddr + offset1);
 
@@ -2898,6 +2869,12 @@ bool dispc_ovl_enabled(enum omap_plane plane)
 }
 EXPORT_SYMBOL(dispc_ovl_enabled);
 
+enum omap_dss_output_id dispc_mgr_get_supported_outputs(enum omap_channel channel)
+{
+	return dss_feat_get_supported_outputs(channel);
+}
+EXPORT_SYMBOL(dispc_mgr_get_supported_outputs);
+
 void dispc_mgr_enable(enum omap_channel channel, bool enable)
 {
 	mgr_fld_write(channel, DISPC_MGR_FLD_ENABLE, enable);
@@ -3287,6 +3264,10 @@ void dispc_mgr_set_timings(enum omap_channel channel,
 	} else {
 		if (t.interlace)
 			t.y_res /= 2;
+
+		if (dispc.feat->supports_double_pixel)
+			REG_FLD_MOD(DISPC_CONTROL, t.double_pixel ? 1 : 0,
+				19, 17);
 	}
 
 	dispc_mgr_set_size(channel, t.x_res, t.y_res);
@@ -3951,6 +3932,8 @@ static const struct dispc_features omap44xx_dispc_feats = {
 	.set_max_preload	=	true,
 	.supports_sync_align	=	true,
 	.has_writeback		=	true,
+	.supports_double_pixel	=	true,
+	.reverse_ilace_field_order =	true,
 };
 
 static const struct dispc_features omap54xx_dispc_feats = {
@@ -3974,6 +3957,8 @@ static const struct dispc_features omap54xx_dispc_feats = {
 	.set_max_preload	=	true,
 	.supports_sync_align	=	true,
 	.has_writeback		=	true,
+	.supports_double_pixel	=	true,
+	.reverse_ilace_field_order =	true,
 };
 
 static int dispc_init_features(struct platform_device *pdev)
@@ -4129,8 +4114,6 @@ static int dispc_bind(struct device *dev, struct device *master, void *data)
 
 	dispc_runtime_put();
 
-	dss_init_overlay_managers();
-
 	dss_debugfs_create_file("dispc", dispc_dump_regs);
 
 	return 0;
@@ -4144,8 +4127,6 @@ static void dispc_unbind(struct device *dev, struct device *master,
 			       void *data)
 {
 	pm_runtime_disable(dev);
-
-	dss_uninit_overlay_managers();
 }
 
 static const struct component_ops dispc_component_ops = {
diff --git a/drivers/gpu/drm/omapdrm/dss/display-sysfs.c b/drivers/gpu/drm/omapdrm/dss/display-sysfs.c
deleted file mode 100644
index 6ad0991f8259..000000000000
--- a/drivers/gpu/drm/omapdrm/dss/display-sysfs.c
+++ /dev/null
@@ -1,356 +0,0 @@
-/*
- * Copyright (C) 2009 Nokia Corporation
- * Author: Tomi Valkeinen <tomi.valkeinen@nokia.com>
- *
- * Some code and ideas taken from drivers/video/omap/ driver
- * by Imre Deak.
- *
- * This program is free software; you can redistribute it and/or modify it
- * under the terms of the GNU General Public License version 2 as published by
- * the Free Software Foundation.
- *
- * This program is distributed in the hope that it will be useful, but WITHOUT
- * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
- * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
- * more details.
- *
- * You should have received a copy of the GNU General Public License along with
- * this program.  If not, see <http://www.gnu.org/licenses/>.
- */
-
-#define DSS_SUBSYS_NAME "DISPLAY"
-
-#include <linux/kernel.h>
-#include <linux/module.h>
-#include <linux/platform_device.h>
-#include <linux/sysfs.h>
-
-#include <video/omapdss.h>
-#include "dss.h"
-
-static ssize_t display_name_show(struct omap_dss_device *dssdev, char *buf)
-{
-	return snprintf(buf, PAGE_SIZE, "%s\n",
-			dssdev->name ?
-			dssdev->name : "");
-}
-
-static ssize_t display_enabled_show(struct omap_dss_device *dssdev, char *buf)
-{
-	return snprintf(buf, PAGE_SIZE, "%d\n",
-			omapdss_device_is_enabled(dssdev));
-}
-
-static ssize_t display_enabled_store(struct omap_dss_device *dssdev,
-		const char *buf, size_t size)
-{
-	int r;
-	bool enable;
-
-	r = strtobool(buf, &enable);
-	if (r)
-		return r;
-
-	if (enable == omapdss_device_is_enabled(dssdev))
-		return size;
-
-	if (omapdss_device_is_connected(dssdev) == false)
-		return -ENODEV;
-
-	if (enable) {
-		r = dssdev->driver->enable(dssdev);
-		if (r)
-			return r;
-	} else {
-		dssdev->driver->disable(dssdev);
-	}
-
-	return size;
-}
-
-static ssize_t display_tear_show(struct omap_dss_device *dssdev, char *buf)
-{
-	return snprintf(buf, PAGE_SIZE, "%d\n",
-			dssdev->driver->get_te ?
-			dssdev->driver->get_te(dssdev) : 0);
-}
-
-static ssize_t display_tear_store(struct omap_dss_device *dssdev,
-	const char *buf, size_t size)
-{
-	int r;
-	bool te;
-
-	if (!dssdev->driver->enable_te || !dssdev->driver->get_te)
-		return -ENOENT;
-
-	r = strtobool(buf, &te);
-	if (r)
-		return r;
-
-	r = dssdev->driver->enable_te(dssdev, te);
-	if (r)
-		return r;
-
-	return size;
-}
-
-static ssize_t display_timings_show(struct omap_dss_device *dssdev, char *buf)
-{
-	struct omap_video_timings t;
-
-	if (!dssdev->driver->get_timings)
-		return -ENOENT;
-
-	dssdev->driver->get_timings(dssdev, &t);
-
-	return snprintf(buf, PAGE_SIZE, "%u,%u/%u/%u/%u,%u/%u/%u/%u\n",
-			t.pixelclock,
-			t.x_res, t.hfp, t.hbp, t.hsw,
-			t.y_res, t.vfp, t.vbp, t.vsw);
-}
-
-static ssize_t display_timings_store(struct omap_dss_device *dssdev,
-	const char *buf, size_t size)
-{
-	struct omap_video_timings t = dssdev->panel.timings;
-	int r, found;
-
-	if (!dssdev->driver->set_timings || !dssdev->driver->check_timings)
-		return -ENOENT;
-
-	found = 0;
-#ifdef CONFIG_OMAP2_DSS_VENC
-	if (strncmp("pal", buf, 3) == 0) {
-		t = omap_dss_pal_timings;
-		found = 1;
-	} else if (strncmp("ntsc", buf, 4) == 0) {
-		t = omap_dss_ntsc_timings;
-		found = 1;
-	}
-#endif
-	if (!found && sscanf(buf, "%u,%hu/%hu/%hu/%hu,%hu/%hu/%hu/%hu",
-				&t.pixelclock,
-				&t.x_res, &t.hfp, &t.hbp, &t.hsw,
-				&t.y_res, &t.vfp, &t.vbp, &t.vsw) != 9)
-		return -EINVAL;
-
-	r = dssdev->driver->check_timings(dssdev, &t);
-	if (r)
-		return r;
-
-	dssdev->driver->disable(dssdev);
-	dssdev->driver->set_timings(dssdev, &t);
-	r = dssdev->driver->enable(dssdev);
-	if (r)
-		return r;
-
-	return size;
-}
-
-static ssize_t display_rotate_show(struct omap_dss_device *dssdev, char *buf)
-{
-	int rotate;
-	if (!dssdev->driver->get_rotate)
-		return -ENOENT;
-	rotate = dssdev->driver->get_rotate(dssdev);
-	return snprintf(buf, PAGE_SIZE, "%u\n", rotate);
-}
-
-static ssize_t display_rotate_store(struct omap_dss_device *dssdev,
-	const char *buf, size_t size)
-{
-	int rot, r;
-
-	if (!dssdev->driver->set_rotate || !dssdev->driver->get_rotate)
-		return -ENOENT;
-
-	r = kstrtoint(buf, 0, &rot);
-	if (r)
-		return r;
-
-	r = dssdev->driver->set_rotate(dssdev, rot);
-	if (r)
-		return r;
-
-	return size;
-}
-
-static ssize_t display_mirror_show(struct omap_dss_device *dssdev, char *buf)
-{
-	int mirror;
-	if (!dssdev->driver->get_mirror)
-		return -ENOENT;
-	mirror = dssdev->driver->get_mirror(dssdev);
-	return snprintf(buf, PAGE_SIZE, "%u\n", mirror);
-}
-
-static ssize_t display_mirror_store(struct omap_dss_device *dssdev,
-	const char *buf, size_t size)
-{
-	int r;
-	bool mirror;
-
-	if (!dssdev->driver->set_mirror || !dssdev->driver->get_mirror)
-		return -ENOENT;
-
-	r = strtobool(buf, &mirror);
-	if (r)
-		return r;
-
-	r = dssdev->driver->set_mirror(dssdev, mirror);
-	if (r)
-		return r;
-
-	return size;
-}
-
-static ssize_t display_wss_show(struct omap_dss_device *dssdev, char *buf)
-{
-	unsigned int wss;
-
-	if (!dssdev->driver->get_wss)
-		return -ENOENT;
-
-	wss = dssdev->driver->get_wss(dssdev);
-
-	return snprintf(buf, PAGE_SIZE, "0x%05x\n", wss);
-}
-
-static ssize_t display_wss_store(struct omap_dss_device *dssdev,
-	const char *buf, size_t size)
-{
-	u32 wss;
-	int r;
-
-	if (!dssdev->driver->get_wss || !dssdev->driver->set_wss)
-		return -ENOENT;
-
-	r = kstrtou32(buf, 0, &wss);
-	if (r)
-		return r;
-
-	if (wss > 0xfffff)
-		return -EINVAL;
-
-	r = dssdev->driver->set_wss(dssdev, wss);
-	if (r)
-		return r;
-
-	return size;
-}
-
-struct display_attribute {
-	struct attribute attr;
-	ssize_t (*show)(struct omap_dss_device *, char *);
-	ssize_t	(*store)(struct omap_dss_device *, const char *, size_t);
-};
-
-#define DISPLAY_ATTR(_name, _mode, _show, _store) \
-	struct display_attribute display_attr_##_name = \
-	__ATTR(_name, _mode, _show, _store)
-
-static DISPLAY_ATTR(name, S_IRUGO, display_name_show, NULL);
-static DISPLAY_ATTR(display_name, S_IRUGO, display_name_show, NULL);
-static DISPLAY_ATTR(enabled, S_IRUGO|S_IWUSR,
-		display_enabled_show, display_enabled_store);
-static DISPLAY_ATTR(tear_elim, S_IRUGO|S_IWUSR,
-		display_tear_show, display_tear_store);
-static DISPLAY_ATTR(timings, S_IRUGO|S_IWUSR,
-		display_timings_show, display_timings_store);
-static DISPLAY_ATTR(rotate, S_IRUGO|S_IWUSR,
-		display_rotate_show, display_rotate_store);
-static DISPLAY_ATTR(mirror, S_IRUGO|S_IWUSR,
-		display_mirror_show, display_mirror_store);
-static DISPLAY_ATTR(wss, S_IRUGO|S_IWUSR,
-		display_wss_show, display_wss_store);
-
-static struct attribute *display_sysfs_attrs[] = {
-	&display_attr_name.attr,
-	&display_attr_display_name.attr,
-	&display_attr_enabled.attr,
-	&display_attr_tear_elim.attr,
-	&display_attr_timings.attr,
-	&display_attr_rotate.attr,
-	&display_attr_mirror.attr,
-	&display_attr_wss.attr,
-	NULL
-};
-
-static ssize_t display_attr_show(struct kobject *kobj, struct attribute *attr,
-		char *buf)
-{
-	struct omap_dss_device *dssdev;
-	struct display_attribute *display_attr;
-
-	dssdev = container_of(kobj, struct omap_dss_device, kobj);
-	display_attr = container_of(attr, struct display_attribute, attr);
-
-	if (!display_attr->show)
-		return -ENOENT;
-
-	return display_attr->show(dssdev, buf);
-}
-
-static ssize_t display_attr_store(struct kobject *kobj, struct attribute *attr,
-		const char *buf, size_t size)
-{
-	struct omap_dss_device *dssdev;
-	struct display_attribute *display_attr;
-
-	dssdev = container_of(kobj, struct omap_dss_device, kobj);
-	display_attr = container_of(attr, struct display_attribute, attr);
-
-	if (!display_attr->store)
-		return -ENOENT;
-
-	return display_attr->store(dssdev, buf, size);
-}
-
-static const struct sysfs_ops display_sysfs_ops = {
-	.show = display_attr_show,
-	.store = display_attr_store,
-};
-
-static struct kobj_type display_ktype = {
-	.sysfs_ops = &display_sysfs_ops,
-	.default_attrs = display_sysfs_attrs,
-};
-
-int display_init_sysfs(struct platform_device *pdev)
-{
-	struct omap_dss_device *dssdev = NULL;
-	int r;
-
-	for_each_dss_dev(dssdev) {
-		r = kobject_init_and_add(&dssdev->kobj, &display_ktype,
-			&pdev->dev.kobj, "%s", dssdev->alias);
-		if (r) {
-			DSSERR("failed to create sysfs files\n");
-			omap_dss_put_device(dssdev);
-			goto err;
-		}
-	}
-
-	return 0;
-
-err:
-	display_uninit_sysfs(pdev);
-
-	return r;
-}
-
-void display_uninit_sysfs(struct platform_device *pdev)
-{
-	struct omap_dss_device *dssdev = NULL;
-
-	for_each_dss_dev(dssdev) {
-		if (kobject_name(&dssdev->kobj) == NULL)
-			continue;
-
-		kobject_del(&dssdev->kobj);
-		kobject_put(&dssdev->kobj);
-
-		memset(&dssdev->kobj, 0, sizeof(dssdev->kobj));
-	}
-}
diff --git a/drivers/gpu/drm/omapdrm/dss/display.c b/drivers/gpu/drm/omapdrm/dss/display.c
index ef5b9027985d..9f3dd09b0a6c 100644
--- a/drivers/gpu/drm/omapdrm/dss/display.c
+++ b/drivers/gpu/drm/omapdrm/dss/display.c
@@ -78,55 +78,6 @@ void omapdss_default_get_timings(struct omap_dss_device *dssdev,
 }
 EXPORT_SYMBOL(omapdss_default_get_timings);
 
-int dss_suspend_all_devices(void)
-{
-	struct omap_dss_device *dssdev = NULL;
-
-	for_each_dss_dev(dssdev) {
-		if (!dssdev->driver)
-			continue;
-
-		if (dssdev->state == OMAP_DSS_DISPLAY_ACTIVE) {
-			dssdev->driver->disable(dssdev);
-			dssdev->activate_after_resume = true;
-		} else {
-			dssdev->activate_after_resume = false;
-		}
-	}
-
-	return 0;
-}
-
-int dss_resume_all_devices(void)
-{
-	struct omap_dss_device *dssdev = NULL;
-
-	for_each_dss_dev(dssdev) {
-		if (!dssdev->driver)
-			continue;
-
-		if (dssdev->activate_after_resume) {
-			dssdev->driver->enable(dssdev);
-			dssdev->activate_after_resume = false;
-		}
-	}
-
-	return 0;
-}
-
-void dss_disable_all_devices(void)
-{
-	struct omap_dss_device *dssdev = NULL;
-
-	for_each_dss_dev(dssdev) {
-		if (!dssdev->driver)
-			continue;
-
-		if (dssdev->state == OMAP_DSS_DISPLAY_ACTIVE)
-			dssdev->driver->disable(dssdev);
-	}
-}
-
 static LIST_HEAD(panel_list);
 static DEFINE_MUTEX(panel_list_mutex);
 static int disp_num_counter;
diff --git a/drivers/gpu/drm/omapdrm/dss/dpi.c b/drivers/gpu/drm/omapdrm/dss/dpi.c
index 7953e6a52346..97ea60257884 100644
--- a/drivers/gpu/drm/omapdrm/dss/dpi.c
+++ b/drivers/gpu/drm/omapdrm/dss/dpi.c
@@ -334,7 +334,7 @@ static int dpi_set_dispc_clk(struct dpi_data *dpi, unsigned long pck_req,
 static int dpi_set_mode(struct dpi_data *dpi)
 {
 	struct omap_dss_device *out = &dpi->output;
-	struct omap_overlay_manager *mgr = out->manager;
+	enum omap_channel channel = out->dispc_channel;
 	struct omap_video_timings *t = &dpi->timings;
 	int lck_div = 0, pck_div = 0;
 	unsigned long fck = 0;
@@ -342,7 +342,7 @@ static int dpi_set_mode(struct dpi_data *dpi)
 	int r = 0;
 
 	if (dpi->pll)
-		r = dpi_set_dsi_clk(dpi, mgr->id, t->pixelclock, &fck,
+		r = dpi_set_dsi_clk(dpi, channel, t->pixelclock, &fck,
 				&lck_div, &pck_div);
 	else
 		r = dpi_set_dispc_clk(dpi, t->pixelclock, &fck,
@@ -359,7 +359,7 @@ static int dpi_set_mode(struct dpi_data *dpi)
 		t->pixelclock = pck;
 	}
 
-	dss_mgr_set_timings(mgr, t);
+	dss_mgr_set_timings(channel, t);
 
 	return 0;
 }
@@ -367,7 +367,7 @@ static int dpi_set_mode(struct dpi_data *dpi)
 static void dpi_config_lcd_manager(struct dpi_data *dpi)
 {
 	struct omap_dss_device *out = &dpi->output;
-	struct omap_overlay_manager *mgr = out->manager;
+	enum omap_channel channel = out->dispc_channel;
 
 	dpi->mgr_config.io_pad_mode = DSS_IO_PAD_MODE_BYPASS;
 
@@ -378,13 +378,14 @@ static void dpi_config_lcd_manager(struct dpi_data *dpi)
 
 	dpi->mgr_config.lcden_sig_polarity = 0;
 
-	dss_mgr_set_lcd_config(mgr, &dpi->mgr_config);
+	dss_mgr_set_lcd_config(channel, &dpi->mgr_config);
 }
 
 static int dpi_display_enable(struct omap_dss_device *dssdev)
 {
 	struct dpi_data *dpi = dpi_get_data_from_dssdev(dssdev);
 	struct omap_dss_device *out = &dpi->output;
+	enum omap_channel channel = out->dispc_channel;
 	int r;
 
 	mutex_lock(&dpi->lock);
@@ -395,7 +396,7 @@ static int dpi_display_enable(struct omap_dss_device *dssdev)
 		goto err_no_reg;
 	}
 
-	if (out->manager == NULL) {
+	if (!out->dispc_channel_connected) {
 		DSSERR("failed to enable display: no output/manager\n");
 		r = -ENODEV;
 		goto err_no_out_mgr;
@@ -411,7 +412,7 @@ static int dpi_display_enable(struct omap_dss_device *dssdev)
 	if (r)
 		goto err_get_dispc;
 
-	r = dss_dpi_select_source(out->port_num, out->manager->id);
+	r = dss_dpi_select_source(out->port_num, channel);
 	if (r)
 		goto err_src_sel;
 
@@ -429,7 +430,7 @@ static int dpi_display_enable(struct omap_dss_device *dssdev)
 
 	mdelay(2);
 
-	r = dss_mgr_enable(out->manager);
+	r = dss_mgr_enable(channel);
 	if (r)
 		goto err_mgr_enable;
 
@@ -457,14 +458,14 @@ err_no_reg:
 static void dpi_display_disable(struct omap_dss_device *dssdev)
 {
 	struct dpi_data *dpi = dpi_get_data_from_dssdev(dssdev);
-	struct omap_overlay_manager *mgr = dpi->output.manager;
+	enum omap_channel channel = dpi->output.dispc_channel;
 
 	mutex_lock(&dpi->lock);
 
-	dss_mgr_disable(mgr);
+	dss_mgr_disable(channel);
 
 	if (dpi->pll) {
-		dss_select_lcd_clk_source(mgr->id, OMAP_DSS_CLK_SRC_FCK);
+		dss_select_lcd_clk_source(channel, OMAP_DSS_CLK_SRC_FCK);
 		dss_pll_disable(dpi->pll);
 	}
 
@@ -506,14 +507,17 @@ static int dpi_check_timings(struct omap_dss_device *dssdev,
 			struct omap_video_timings *timings)
 {
 	struct dpi_data *dpi = dpi_get_data_from_dssdev(dssdev);
-	struct omap_overlay_manager *mgr = dpi->output.manager;
+	enum omap_channel channel = dpi->output.dispc_channel;
 	int lck_div, pck_div;
 	unsigned long fck;
 	unsigned long pck;
 	struct dpi_clk_calc_ctx ctx;
 	bool ok;
 
-	if (mgr && !dispc_mgr_timings_ok(mgr->id, timings))
+	if (timings->x_res % 8 != 0)
+		return -EINVAL;
+
+	if (!dispc_mgr_timings_ok(channel, timings))
 		return -EINVAL;
 
 	if (timings->pixelclock == 0)
@@ -660,7 +664,7 @@ static int dpi_connect(struct omap_dss_device *dssdev,
 		struct omap_dss_device *dst)
 {
 	struct dpi_data *dpi = dpi_get_data_from_dssdev(dssdev);
-	struct omap_overlay_manager *mgr;
+	enum omap_channel channel = dpi->output.dispc_channel;
 	int r;
 
 	r = dpi_init_regulator(dpi);
@@ -669,11 +673,7 @@ static int dpi_connect(struct omap_dss_device *dssdev,
 
 	dpi_init_pll(dpi);
 
-	mgr = omap_dss_get_overlay_manager(dssdev->dispc_channel);
-	if (!mgr)
-		return -ENODEV;
-
-	r = dss_mgr_connect(mgr, dssdev);
+	r = dss_mgr_connect(channel, dssdev);
 	if (r)
 		return r;
 
@@ -681,7 +681,7 @@ static int dpi_connect(struct omap_dss_device *dssdev,
 	if (r) {
 		DSSERR("failed to connect output to new device: %s\n",
 				dst->name);
-		dss_mgr_disconnect(mgr, dssdev);
+		dss_mgr_disconnect(channel, dssdev);
 		return r;
 	}
 
@@ -691,6 +691,9 @@ static int dpi_connect(struct omap_dss_device *dssdev,
 static void dpi_disconnect(struct omap_dss_device *dssdev,
 		struct omap_dss_device *dst)
 {
+	struct dpi_data *dpi = dpi_get_data_from_dssdev(dssdev);
+	enum omap_channel channel = dpi->output.dispc_channel;
+
 	WARN_ON(dst != dssdev->dst);
 
 	if (dst != dssdev->dst)
@@ -698,8 +701,7 @@ static void dpi_disconnect(struct omap_dss_device *dssdev,
 
 	omapdss_output_unset_device(dssdev);
 
-	if (dssdev->manager)
-		dss_mgr_disconnect(dssdev->manager, dssdev);
+	dss_mgr_disconnect(channel, dssdev);
 }
 
 static const struct omapdss_dpi_ops dpi_ops = {
diff --git a/drivers/gpu/drm/omapdrm/dss/dsi.c b/drivers/gpu/drm/omapdrm/dss/dsi.c
index 43be4b2a7b05..8730646a0cbb 100644
--- a/drivers/gpu/drm/omapdrm/dss/dsi.c
+++ b/drivers/gpu/drm/omapdrm/dss/dsi.c
@@ -214,9 +214,9 @@ struct dsi_reg { u16 module; u16 idx; };
 typedef void (*omap_dsi_isr_t) (void *arg, u32 mask);
 
 static int dsi_display_init_dispc(struct platform_device *dsidev,
-	struct omap_overlay_manager *mgr);
+	enum omap_channel channel);
 static void dsi_display_uninit_dispc(struct platform_device *dsidev,
-	struct omap_overlay_manager *mgr);
+	enum omap_channel channel);
 
 static int dsi_vc_send_null(struct omap_dss_device *dssdev, int channel);
 
@@ -3826,19 +3826,19 @@ static int dsi_enable_video_output(struct omap_dss_device *dssdev, int channel)
 {
 	struct platform_device *dsidev = dsi_get_dsidev_from_dssdev(dssdev);
 	struct dsi_data *dsi = dsi_get_dsidrv_data(dsidev);
-	struct omap_overlay_manager *mgr = dsi->output.manager;
+	enum omap_channel dispc_channel = dssdev->dispc_channel;
 	int bpp = dsi_get_pixel_size(dsi->pix_fmt);
 	struct omap_dss_device *out = &dsi->output;
 	u8 data_type;
 	u16 word_count;
 	int r;
 
-	if (out->manager == NULL) {
+	if (!out->dispc_channel_connected) {
 		DSSERR("failed to enable display: no output/manager\n");
 		return -ENODEV;
 	}
 
-	r = dsi_display_init_dispc(dsidev, mgr);
+	r = dsi_display_init_dispc(dsidev, dispc_channel);
 	if (r)
 		goto err_init_dispc;
 
@@ -3876,7 +3876,7 @@ static int dsi_enable_video_output(struct omap_dss_device *dssdev, int channel)
 		dsi_if_enable(dsidev, true);
 	}
 
-	r = dss_mgr_enable(mgr);
+	r = dss_mgr_enable(dispc_channel);
 	if (r)
 		goto err_mgr_enable;
 
@@ -3888,7 +3888,7 @@ err_mgr_enable:
 		dsi_vc_enable(dsidev, channel, false);
 	}
 err_pix_fmt:
-	dsi_display_uninit_dispc(dsidev, mgr);
+	dsi_display_uninit_dispc(dsidev, dispc_channel);
 err_init_dispc:
 	return r;
 }
@@ -3897,7 +3897,7 @@ static void dsi_disable_video_output(struct omap_dss_device *dssdev, int channel
 {
 	struct platform_device *dsidev = dsi_get_dsidev_from_dssdev(dssdev);
 	struct dsi_data *dsi = dsi_get_dsidrv_data(dsidev);
-	struct omap_overlay_manager *mgr = dsi->output.manager;
+	enum omap_channel dispc_channel = dssdev->dispc_channel;
 
 	if (dsi->mode == OMAP_DSS_DSI_VIDEO_MODE) {
 		dsi_if_enable(dsidev, false);
@@ -3910,15 +3910,15 @@ static void dsi_disable_video_output(struct omap_dss_device *dssdev, int channel
 		dsi_if_enable(dsidev, true);
 	}
 
-	dss_mgr_disable(mgr);
+	dss_mgr_disable(dispc_channel);
 
-	dsi_display_uninit_dispc(dsidev, mgr);
+	dsi_display_uninit_dispc(dsidev, dispc_channel);
 }
 
 static void dsi_update_screen_dispc(struct platform_device *dsidev)
 {
 	struct dsi_data *dsi = dsi_get_dsidrv_data(dsidev);
-	struct omap_overlay_manager *mgr = dsi->output.manager;
+	enum omap_channel dispc_channel = dsi->output.dispc_channel;
 	unsigned bytespp;
 	unsigned bytespl;
 	unsigned bytespf;
@@ -3980,9 +3980,9 @@ static void dsi_update_screen_dispc(struct platform_device *dsidev)
 		msecs_to_jiffies(250));
 	BUG_ON(r == 0);
 
-	dss_mgr_set_timings(mgr, &dsi->timings);
+	dss_mgr_set_timings(dispc_channel, &dsi->timings);
 
-	dss_mgr_start_update(mgr);
+	dss_mgr_start_update(dispc_channel);
 
 	if (dsi->te_enabled) {
 		/* disable LP_RX_TO, so that we can receive TE.  Time to wait
@@ -4105,17 +4105,17 @@ static int dsi_configure_dispc_clocks(struct platform_device *dsidev)
 }
 
 static int dsi_display_init_dispc(struct platform_device *dsidev,
-		struct omap_overlay_manager *mgr)
+		enum omap_channel channel)
 {
 	struct dsi_data *dsi = dsi_get_dsidrv_data(dsidev);
 	int r;
 
-	dss_select_lcd_clk_source(mgr->id, dsi->module_id == 0 ?
+	dss_select_lcd_clk_source(channel, dsi->module_id == 0 ?
 			OMAP_DSS_CLK_SRC_DSI_PLL_HSDIV_DISPC :
 			OMAP_DSS_CLK_SRC_DSI2_PLL_HSDIV_DISPC);
 
 	if (dsi->mode == OMAP_DSS_DSI_CMD_MODE) {
-		r = dss_mgr_register_framedone_handler(mgr,
+		r = dss_mgr_register_framedone_handler(channel,
 				dsi_framedone_irq_callback, dsidev);
 		if (r) {
 			DSSERR("can't register FRAMEDONE handler\n");
@@ -4140,7 +4140,7 @@ static int dsi_display_init_dispc(struct platform_device *dsidev,
 	dsi->timings.de_level = OMAPDSS_SIG_ACTIVE_HIGH;
 	dsi->timings.sync_pclk_edge = OMAPDSS_DRIVE_SIG_FALLING_EDGE;
 
-	dss_mgr_set_timings(mgr, &dsi->timings);
+	dss_mgr_set_timings(channel, &dsi->timings);
 
 	r = dsi_configure_dispc_clocks(dsidev);
 	if (r)
@@ -4151,28 +4151,28 @@ static int dsi_display_init_dispc(struct platform_device *dsidev,
 			dsi_get_pixel_size(dsi->pix_fmt);
 	dsi->mgr_config.lcden_sig_polarity = 0;
 
-	dss_mgr_set_lcd_config(mgr, &dsi->mgr_config);
+	dss_mgr_set_lcd_config(channel, &dsi->mgr_config);
 
 	return 0;
 err1:
 	if (dsi->mode == OMAP_DSS_DSI_CMD_MODE)
-		dss_mgr_unregister_framedone_handler(mgr,
+		dss_mgr_unregister_framedone_handler(channel,
 				dsi_framedone_irq_callback, dsidev);
 err:
-	dss_select_lcd_clk_source(mgr->id, OMAP_DSS_CLK_SRC_FCK);
+	dss_select_lcd_clk_source(channel, OMAP_DSS_CLK_SRC_FCK);
 	return r;
 }
 
 static void dsi_display_uninit_dispc(struct platform_device *dsidev,
-		struct omap_overlay_manager *mgr)
+		enum omap_channel channel)
 {
 	struct dsi_data *dsi = dsi_get_dsidrv_data(dsidev);
 
 	if (dsi->mode == OMAP_DSS_DSI_CMD_MODE)
-		dss_mgr_unregister_framedone_handler(mgr,
+		dss_mgr_unregister_framedone_handler(channel,
 				dsi_framedone_irq_callback, dsidev);
 
-	dss_select_lcd_clk_source(mgr->id, OMAP_DSS_CLK_SRC_FCK);
+	dss_select_lcd_clk_source(channel, OMAP_DSS_CLK_SRC_FCK);
 }
 
 static int dsi_configure_dsi_clocks(struct platform_device *dsidev)
@@ -4983,18 +4983,14 @@ static int dsi_connect(struct omap_dss_device *dssdev,
 		struct omap_dss_device *dst)
 {
 	struct platform_device *dsidev = dsi_get_dsidev_from_dssdev(dssdev);
-	struct omap_overlay_manager *mgr;
+	enum omap_channel dispc_channel = dssdev->dispc_channel;
 	int r;
 
 	r = dsi_regulator_init(dsidev);
 	if (r)
 		return r;
 
-	mgr = omap_dss_get_overlay_manager(dssdev->dispc_channel);
-	if (!mgr)
-		return -ENODEV;
-
-	r = dss_mgr_connect(mgr, dssdev);
+	r = dss_mgr_connect(dispc_channel, dssdev);
 	if (r)
 		return r;
 
@@ -5002,7 +4998,7 @@ static int dsi_connect(struct omap_dss_device *dssdev,
 	if (r) {
 		DSSERR("failed to connect output to new device: %s\n",
 				dssdev->name);
-		dss_mgr_disconnect(mgr, dssdev);
+		dss_mgr_disconnect(dispc_channel, dssdev);
 		return r;
 	}
 
@@ -5012,6 +5008,8 @@ static int dsi_connect(struct omap_dss_device *dssdev,
 static void dsi_disconnect(struct omap_dss_device *dssdev,
 		struct omap_dss_device *dst)
 {
+	enum omap_channel dispc_channel = dssdev->dispc_channel;
+
 	WARN_ON(dst != dssdev->dst);
 
 	if (dst != dssdev->dst)
@@ -5019,8 +5017,7 @@ static void dsi_disconnect(struct omap_dss_device *dssdev,
 
 	omapdss_output_unset_device(dssdev);
 
-	if (dssdev->manager)
-		dss_mgr_disconnect(dssdev->manager, dssdev);
+	dss_mgr_disconnect(dispc_channel, dssdev);
 }
 
 static const struct omapdss_dsi_ops dsi_ops = {
diff --git a/drivers/gpu/drm/omapdrm/dss/dss.h b/drivers/gpu/drm/omapdrm/dss/dss.h
index 9a6453235585..38e6ab50142d 100644
--- a/drivers/gpu/drm/omapdrm/dss/dss.h
+++ b/drivers/gpu/drm/omapdrm/dss/dss.h
@@ -25,6 +25,8 @@
 
 #include <linux/interrupt.h>
 
+#include "omapdss.h"
+
 #ifdef pr_fmt
 #undef pr_fmt
 #endif
@@ -205,29 +207,6 @@ void dss_dsi_disable_pads(int dsi_id, unsigned lane_mask);
 int dss_set_min_bus_tput(struct device *dev, unsigned long tput);
 int dss_debugfs_create_file(const char *name, void (*write)(struct seq_file *));
 
-/* display */
-int dss_suspend_all_devices(void);
-int dss_resume_all_devices(void);
-void dss_disable_all_devices(void);
-
-int display_init_sysfs(struct platform_device *pdev);
-void display_uninit_sysfs(struct platform_device *pdev);
-
-/* manager */
-int dss_init_overlay_managers(void);
-void dss_uninit_overlay_managers(void);
-int dss_init_overlay_managers_sysfs(struct platform_device *pdev);
-void dss_uninit_overlay_managers_sysfs(struct platform_device *pdev);
-int dss_mgr_simple_check(struct omap_overlay_manager *mgr,
-		const struct omap_overlay_manager_info *info);
-int dss_mgr_check_timings(struct omap_overlay_manager *mgr,
-		const struct omap_video_timings *timings);
-int dss_mgr_check(struct omap_overlay_manager *mgr,
-		struct omap_overlay_manager_info *info,
-		const struct omap_video_timings *mgr_timings,
-		const struct dss_lcd_mgr_config *config,
-		struct omap_overlay_info **overlay_infos);
-
 static inline bool dss_mgr_is_lcd(enum omap_channel id)
 {
 	if (id == OMAP_DSS_CHANNEL_LCD || id == OMAP_DSS_CHANNEL_LCD2 ||
@@ -237,24 +216,6 @@ static inline bool dss_mgr_is_lcd(enum omap_channel id)
 		return false;
 }
 
-int dss_manager_kobj_init(struct omap_overlay_manager *mgr,
-		struct platform_device *pdev);
-void dss_manager_kobj_uninit(struct omap_overlay_manager *mgr);
-
-/* overlay */
-void dss_init_overlays(struct platform_device *pdev);
-void dss_uninit_overlays(struct platform_device *pdev);
-void dss_overlay_setup_dispc_manager(struct omap_overlay_manager *mgr);
-int dss_ovl_simple_check(struct omap_overlay *ovl,
-		const struct omap_overlay_info *info);
-int dss_ovl_check(struct omap_overlay *ovl, struct omap_overlay_info *info,
-		const struct omap_video_timings *mgr_timings);
-bool dss_ovl_use_replication(struct dss_lcd_mgr_config config,
-		enum omap_color_mode mode);
-int dss_overlay_kobj_init(struct omap_overlay *ovl,
-		struct platform_device *pdev);
-void dss_overlay_kobj_uninit(struct omap_overlay *ovl);
-
 /* DSS */
 int dss_init_platform_driver(void) __init;
 void dss_uninit_platform_driver(void);
diff --git a/drivers/gpu/drm/omapdrm/dss/hdmi4.c b/drivers/gpu/drm/omapdrm/dss/hdmi4.c
index 7103c659a534..f892ae157ff3 100644
--- a/drivers/gpu/drm/omapdrm/dss/hdmi4.c
+++ b/drivers/gpu/drm/omapdrm/dss/hdmi4.c
@@ -165,9 +165,10 @@ static int hdmi_power_on_full(struct omap_dss_device *dssdev)
 {
 	int r;
 	struct omap_video_timings *p;
-	struct omap_overlay_manager *mgr = hdmi.output.manager;
+	enum omap_channel channel = dssdev->dispc_channel;
 	struct hdmi_wp_data *wp = &hdmi.wp;
 	struct dss_pll_clock_info hdmi_cinfo = { 0 };
+	unsigned pc;
 
 	r = hdmi_power_on_core(dssdev);
 	if (r)
@@ -181,7 +182,11 @@ static int hdmi_power_on_full(struct omap_dss_device *dssdev)
 
 	DSSDBG("hdmi_power_on x_res= %d y_res = %d\n", p->x_res, p->y_res);
 
-	hdmi_pll_compute(&hdmi.pll, p->pixelclock, &hdmi_cinfo);
+	pc = p->pixelclock;
+	if (p->double_pixel)
+		pc *= 2;
+
+	hdmi_pll_compute(&hdmi.pll, pc, &hdmi_cinfo);
 
 	r = dss_pll_enable(&hdmi.pll.pll);
 	if (r) {
@@ -212,24 +217,24 @@ static int hdmi_power_on_full(struct omap_dss_device *dssdev)
 	dispc_enable_gamma_table(0);
 
 	/* tv size */
-	dss_mgr_set_timings(mgr, p);
+	dss_mgr_set_timings(channel, p);
 
-	r = hdmi_wp_video_start(&hdmi.wp);
+	r = dss_mgr_enable(channel);
 	if (r)
-		goto err_vid_enable;
+		goto err_mgr_enable;
 
-	r = dss_mgr_enable(mgr);
+	r = hdmi_wp_video_start(&hdmi.wp);
 	if (r)
-		goto err_mgr_enable;
+		goto err_vid_enable;
 
 	hdmi_wp_set_irqenable(wp,
 		HDMI_IRQ_LINK_CONNECT | HDMI_IRQ_LINK_DISCONNECT);
 
 	return 0;
 
-err_mgr_enable:
-	hdmi_wp_video_stop(&hdmi.wp);
 err_vid_enable:
+	dss_mgr_disable(channel);
+err_mgr_enable:
 	hdmi_wp_set_phy_pwr(&hdmi.wp, HDMI_PHYPWRCMD_OFF);
 err_phy_pwr:
 err_phy_cfg:
@@ -242,14 +247,14 @@ err_pll_enable:
 
 static void hdmi_power_off_full(struct omap_dss_device *dssdev)
 {
-	struct omap_overlay_manager *mgr = hdmi.output.manager;
+	enum omap_channel channel = dssdev->dispc_channel;
 
 	hdmi_wp_clear_irqenable(&hdmi.wp, 0xffffffff);
 
-	dss_mgr_disable(mgr);
-
 	hdmi_wp_video_stop(&hdmi.wp);
 
+	dss_mgr_disable(channel);
+
 	hdmi_wp_set_phy_pwr(&hdmi.wp, HDMI_PHYPWRCMD_OFF);
 
 	dss_pll_disable(&hdmi.pll.pll);
@@ -260,9 +265,7 @@ static void hdmi_power_off_full(struct omap_dss_device *dssdev)
 static int hdmi_display_check_timing(struct omap_dss_device *dssdev,
 					struct omap_video_timings *timings)
 {
-	struct omap_dss_device *out = &hdmi.output;
-
-	if (!dispc_mgr_timings_ok(out->dispc_channel, timings))
+	if (!dispc_mgr_timings_ok(dssdev->dispc_channel, timings))
 		return -EINVAL;
 
 	return 0;
@@ -343,7 +346,7 @@ static int hdmi_display_enable(struct omap_dss_device *dssdev)
 
 	mutex_lock(&hdmi.lock);
 
-	if (out->manager == NULL) {
+	if (!out->dispc_channel_connected) {
 		DSSERR("failed to enable display: no output/manager\n");
 		r = -ENODEV;
 		goto err0;
@@ -433,18 +436,14 @@ static void hdmi_core_disable(struct omap_dss_device *dssdev)
 static int hdmi_connect(struct omap_dss_device *dssdev,
 		struct omap_dss_device *dst)
 {
-	struct omap_overlay_manager *mgr;
+	enum omap_channel channel = dssdev->dispc_channel;
 	int r;
 
 	r = hdmi_init_regulator();
 	if (r)
 		return r;
 
-	mgr = omap_dss_get_overlay_manager(dssdev->dispc_channel);
-	if (!mgr)
-		return -ENODEV;
-
-	r = dss_mgr_connect(mgr, dssdev);
+	r = dss_mgr_connect(channel, dssdev);
 	if (r)
 		return r;
 
@@ -452,7 +451,7 @@ static int hdmi_connect(struct omap_dss_device *dssdev,
 	if (r) {
 		DSSERR("failed to connect output to new device: %s\n",
 				dst->name);
-		dss_mgr_disconnect(mgr, dssdev);
+		dss_mgr_disconnect(channel, dssdev);
 		return r;
 	}
 
@@ -462,6 +461,8 @@ static int hdmi_connect(struct omap_dss_device *dssdev,
 static void hdmi_disconnect(struct omap_dss_device *dssdev,
 		struct omap_dss_device *dst)
 {
+	enum omap_channel channel = dssdev->dispc_channel;
+
 	WARN_ON(dst != dssdev->dst);
 
 	if (dst != dssdev->dst)
@@ -469,8 +470,7 @@ static void hdmi_disconnect(struct omap_dss_device *dssdev,
 
 	omapdss_output_unset_device(dssdev);
 
-	if (dssdev->manager)
-		dss_mgr_disconnect(dssdev->manager, dssdev);
+	dss_mgr_disconnect(channel, dssdev);
 }
 
 static int hdmi_read_edid(struct omap_dss_device *dssdev,
diff --git a/drivers/gpu/drm/omapdrm/dss/hdmi5.c b/drivers/gpu/drm/omapdrm/dss/hdmi5.c
index a955a2c4c061..a43f7b10e113 100644
--- a/drivers/gpu/drm/omapdrm/dss/hdmi5.c
+++ b/drivers/gpu/drm/omapdrm/dss/hdmi5.c
@@ -182,8 +182,9 @@ static int hdmi_power_on_full(struct omap_dss_device *dssdev)
 {
 	int r;
 	struct omap_video_timings *p;
-	struct omap_overlay_manager *mgr = hdmi.output.manager;
+	enum omap_channel channel = dssdev->dispc_channel;
 	struct dss_pll_clock_info hdmi_cinfo = { 0 };
+	unsigned pc;
 
 	r = hdmi_power_on_core(dssdev);
 	if (r)
@@ -193,7 +194,11 @@ static int hdmi_power_on_full(struct omap_dss_device *dssdev)
 
 	DSSDBG("hdmi_power_on x_res= %d y_res = %d\n", p->x_res, p->y_res);
 
-	hdmi_pll_compute(&hdmi.pll, p->pixelclock, &hdmi_cinfo);
+	pc = p->pixelclock;
+	if (p->double_pixel)
+		pc *= 2;
+
+	hdmi_pll_compute(&hdmi.pll, pc, &hdmi_cinfo);
 
 	/* disable and clear irqs */
 	hdmi_wp_clear_irqenable(&hdmi.wp, 0xffffffff);
@@ -229,24 +234,24 @@ static int hdmi_power_on_full(struct omap_dss_device *dssdev)
 	dispc_enable_gamma_table(0);
 
 	/* tv size */
-	dss_mgr_set_timings(mgr, p);
+	dss_mgr_set_timings(channel, p);
 
-	r = hdmi_wp_video_start(&hdmi.wp);
+	r = dss_mgr_enable(channel);
 	if (r)
-		goto err_vid_enable;
+		goto err_mgr_enable;
 
-	r = dss_mgr_enable(mgr);
+	r = hdmi_wp_video_start(&hdmi.wp);
 	if (r)
-		goto err_mgr_enable;
+		goto err_vid_enable;
 
 	hdmi_wp_set_irqenable(&hdmi.wp,
 			HDMI_IRQ_LINK_CONNECT | HDMI_IRQ_LINK_DISCONNECT);
 
 	return 0;
 
-err_mgr_enable:
-	hdmi_wp_video_stop(&hdmi.wp);
 err_vid_enable:
+	dss_mgr_disable(channel);
+err_mgr_enable:
 	hdmi_wp_set_phy_pwr(&hdmi.wp, HDMI_PHYPWRCMD_OFF);
 err_phy_pwr:
 err_phy_cfg:
@@ -259,14 +264,14 @@ err_pll_enable:
 
 static void hdmi_power_off_full(struct omap_dss_device *dssdev)
 {
-	struct omap_overlay_manager *mgr = hdmi.output.manager;
+	enum omap_channel channel = dssdev->dispc_channel;
 
 	hdmi_wp_clear_irqenable(&hdmi.wp, 0xffffffff);
 
-	dss_mgr_disable(mgr);
-
 	hdmi_wp_video_stop(&hdmi.wp);
 
+	dss_mgr_disable(channel);
+
 	hdmi_wp_set_phy_pwr(&hdmi.wp, HDMI_PHYPWRCMD_OFF);
 
 	dss_pll_disable(&hdmi.pll.pll);
@@ -277,13 +282,7 @@ static void hdmi_power_off_full(struct omap_dss_device *dssdev)
 static int hdmi_display_check_timing(struct omap_dss_device *dssdev,
 					struct omap_video_timings *timings)
 {
-	struct omap_dss_device *out = &hdmi.output;
-
-	/* TODO: proper interlace support */
-	if (timings->interlace)
-		return -EINVAL;
-
-	if (!dispc_mgr_timings_ok(out->dispc_channel, timings))
+	if (!dispc_mgr_timings_ok(dssdev->dispc_channel, timings))
 		return -EINVAL;
 
 	return 0;
@@ -373,7 +372,7 @@ static int hdmi_display_enable(struct omap_dss_device *dssdev)
 
 	mutex_lock(&hdmi.lock);
 
-	if (out->manager == NULL) {
+	if (!out->dispc_channel_connected) {
 		DSSERR("failed to enable display: no output/manager\n");
 		r = -ENODEV;
 		goto err0;
@@ -463,18 +462,14 @@ static void hdmi_core_disable(struct omap_dss_device *dssdev)
 static int hdmi_connect(struct omap_dss_device *dssdev,
 		struct omap_dss_device *dst)
 {
-	struct omap_overlay_manager *mgr;
+	enum omap_channel channel = dssdev->dispc_channel;
 	int r;
 
 	r = hdmi_init_regulator();
 	if (r)
 		return r;
 
-	mgr = omap_dss_get_overlay_manager(dssdev->dispc_channel);
-	if (!mgr)
-		return -ENODEV;
-
-	r = dss_mgr_connect(mgr, dssdev);
+	r = dss_mgr_connect(channel, dssdev);
 	if (r)
 		return r;
 
@@ -482,7 +477,7 @@ static int hdmi_connect(struct omap_dss_device *dssdev,
 	if (r) {
 		DSSERR("failed to connect output to new device: %s\n",
 				dst->name);
-		dss_mgr_disconnect(mgr, dssdev);
+		dss_mgr_disconnect(channel, dssdev);
 		return r;
 	}
 
@@ -492,6 +487,8 @@ static int hdmi_connect(struct omap_dss_device *dssdev,
 static void hdmi_disconnect(struct omap_dss_device *dssdev,
 		struct omap_dss_device *dst)
 {
+	enum omap_channel channel = dssdev->dispc_channel;
+
 	WARN_ON(dst != dssdev->dst);
 
 	if (dst != dssdev->dst)
@@ -499,8 +496,7 @@ static void hdmi_disconnect(struct omap_dss_device *dssdev,
 
 	omapdss_output_unset_device(dssdev);
 
-	if (dssdev->manager)
-		dss_mgr_disconnect(dssdev->manager, dssdev);
+	dss_mgr_disconnect(channel, dssdev);
 }
 
 static int hdmi_read_edid(struct omap_dss_device *dssdev,
diff --git a/drivers/gpu/drm/omapdrm/dss/hdmi5_core.c b/drivers/gpu/drm/omapdrm/dss/hdmi5_core.c
index 8ea531d2652c..6a397520cae5 100644
--- a/drivers/gpu/drm/omapdrm/dss/hdmi5_core.c
+++ b/drivers/gpu/drm/omapdrm/dss/hdmi5_core.c
@@ -292,25 +292,36 @@ static void hdmi_core_init(struct hdmi_core_vid_config *video_cfg,
 {
 	DSSDBG("hdmi_core_init\n");
 
+	video_cfg->v_fc_config.timings = cfg->timings;
+
 	/* video core */
 	video_cfg->data_enable_pol = 1; /* It is always 1*/
-	video_cfg->v_fc_config.timings.hsync_level = cfg->timings.hsync_level;
-	video_cfg->v_fc_config.timings.x_res = cfg->timings.x_res;
-	video_cfg->v_fc_config.timings.hsw = cfg->timings.hsw - 1;
-	video_cfg->v_fc_config.timings.hbp = cfg->timings.hbp;
-	video_cfg->v_fc_config.timings.hfp = cfg->timings.hfp;
 	video_cfg->hblank = cfg->timings.hfp +
-				cfg->timings.hbp + cfg->timings.hsw - 1;
-	video_cfg->v_fc_config.timings.vsync_level = cfg->timings.vsync_level;
-	video_cfg->v_fc_config.timings.y_res = cfg->timings.y_res;
-	video_cfg->v_fc_config.timings.vsw = cfg->timings.vsw;
-	video_cfg->v_fc_config.timings.vfp = cfg->timings.vfp;
-	video_cfg->v_fc_config.timings.vbp = cfg->timings.vbp;
-	video_cfg->vblank_osc = 0; /* Always 0 - need to confirm */
+				cfg->timings.hbp + cfg->timings.hsw;
+	video_cfg->vblank_osc = 0;
 	video_cfg->vblank = cfg->timings.vsw +
 				cfg->timings.vfp + cfg->timings.vbp;
 	video_cfg->v_fc_config.hdmi_dvi_mode = cfg->hdmi_dvi_mode;
-	video_cfg->v_fc_config.timings.interlace = cfg->timings.interlace;
+
+	if (cfg->timings.interlace) {
+		/* set vblank_osc if vblank is fractional */
+		if (video_cfg->vblank % 2 != 0)
+			video_cfg->vblank_osc = 1;
+
+		video_cfg->v_fc_config.timings.y_res /= 2;
+		video_cfg->vblank /= 2;
+		video_cfg->v_fc_config.timings.vfp /= 2;
+		video_cfg->v_fc_config.timings.vsw /= 2;
+		video_cfg->v_fc_config.timings.vbp /= 2;
+	}
+
+	if (cfg->timings.double_pixel) {
+		video_cfg->v_fc_config.timings.x_res *= 2;
+		video_cfg->hblank *= 2;
+		video_cfg->v_fc_config.timings.hfp *= 2;
+		video_cfg->v_fc_config.timings.hsw *= 2;
+		video_cfg->v_fc_config.timings.hbp *= 2;
+	}
 }
 
 /* DSS_HDMI_CORE_VIDEO_CONFIG */
@@ -377,6 +388,11 @@ static void hdmi_core_video_config(struct hdmi_core_data *core,
 	/* select DVI mode */
 	REG_FLD_MOD(base, HDMI_CORE_FC_INVIDCONF,
 			cfg->v_fc_config.hdmi_dvi_mode, 3, 3);
+
+	if (cfg->v_fc_config.timings.double_pixel)
+		REG_FLD_MOD(base, HDMI_CORE_FC_PRCONF, 2, 7, 4);
+	else
+		REG_FLD_MOD(base, HDMI_CORE_FC_PRCONF, 1, 7, 4);
 }
 
 static void hdmi_core_config_video_packetizer(struct hdmi_core_data *core)
diff --git a/drivers/gpu/drm/omapdrm/dss/hdmi_wp.c b/drivers/gpu/drm/omapdrm/dss/hdmi_wp.c
index 7c544bc56fb5..13442b9052d1 100644
--- a/drivers/gpu/drm/omapdrm/dss/hdmi_wp.c
+++ b/drivers/gpu/drm/omapdrm/dss/hdmi_wp.c
@@ -165,12 +165,24 @@ void hdmi_wp_video_config_timing(struct hdmi_wp_data *wp,
 {
 	u32 timing_h = 0;
 	u32 timing_v = 0;
+	unsigned hsw_offset = 1;
 
 	DSSDBG("Enter hdmi_wp_video_config_timing\n");
 
+	/*
+	 * On OMAP4 and OMAP5 ES1 the HSW field is programmed as is. On OMAP5
+	 * ES2+ (including DRA7/AM5 SoCs) HSW field is programmed to hsw-1.
+	 * However, we don't support OMAP5 ES1 at all, so we can just check for
+	 * OMAP4 here.
+	 */
+	if (omapdss_get_version() == OMAPDSS_VER_OMAP4430_ES1 ||
+	    omapdss_get_version() == OMAPDSS_VER_OMAP4430_ES2 ||
+	    omapdss_get_version() == OMAPDSS_VER_OMAP4)
+		hsw_offset = 0;
+
 	timing_h |= FLD_VAL(timings->hbp, 31, 20);
 	timing_h |= FLD_VAL(timings->hfp, 19, 8);
-	timing_h |= FLD_VAL(timings->hsw, 7, 0);
+	timing_h |= FLD_VAL(timings->hsw - hsw_offset, 7, 0);
 	hdmi_write_reg(wp->base, HDMI_WP_VIDEO_TIMING_H, timing_h);
 
 	timing_v |= FLD_VAL(timings->vbp, 31, 20);
@@ -187,8 +199,6 @@ void hdmi_wp_init_vid_fmt_timings(struct hdmi_video_format *video_fmt,
 	video_fmt->packing_mode = HDMI_PACK_10b_RGB_YUV444;
 	video_fmt->y_res = param->timings.y_res;
 	video_fmt->x_res = param->timings.x_res;
-	if (param->timings.interlace)
-		video_fmt->y_res /= 2;
 
 	timings->hbp = param->timings.hbp;
 	timings->hfp = param->timings.hfp;
@@ -196,9 +206,25 @@ void hdmi_wp_init_vid_fmt_timings(struct hdmi_video_format *video_fmt,
 	timings->vbp = param->timings.vbp;
 	timings->vfp = param->timings.vfp;
 	timings->vsw = param->timings.vsw;
+
 	timings->vsync_level = param->timings.vsync_level;
 	timings->hsync_level = param->timings.hsync_level;
 	timings->interlace = param->timings.interlace;
+	timings->double_pixel = param->timings.double_pixel;
+
+	if (param->timings.interlace) {
+		video_fmt->y_res /= 2;
+		timings->vbp /= 2;
+		timings->vfp /= 2;
+		timings->vsw /= 2;
+	}
+
+	if (param->timings.double_pixel) {
+		video_fmt->x_res *= 2;
+		timings->hfp *= 2;
+		timings->hsw *= 2;
+		timings->hbp *= 2;
+	}
 }
 
 void hdmi_wp_audio_config_format(struct hdmi_wp_data *wp,
diff --git a/drivers/gpu/drm/omapdrm/dss/manager-sysfs.c b/drivers/gpu/drm/omapdrm/dss/manager-sysfs.c
deleted file mode 100644
index a7414fb12830..000000000000
--- a/drivers/gpu/drm/omapdrm/dss/manager-sysfs.c
+++ /dev/null
@@ -1,531 +0,0 @@
-/*
- * Copyright (C) 2009 Nokia Corporation
- * Author: Tomi Valkeinen <tomi.valkeinen@nokia.com>
- *
- * Some code and ideas taken from drivers/video/omap/ driver
- * by Imre Deak.
- *
- * This program is free software; you can redistribute it and/or modify it
- * under the terms of the GNU General Public License version 2 as published by
- * the Free Software Foundation.
- *
- * This program is distributed in the hope that it will be useful, but WITHOUT
- * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
- * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
- * more details.
- *
- * You should have received a copy of the GNU General Public License along with
- * this program.  If not, see <http://www.gnu.org/licenses/>.
- */
-
-#define DSS_SUBSYS_NAME "MANAGER"
-
-#include <linux/kernel.h>
-#include <linux/slab.h>
-#include <linux/module.h>
-#include <linux/platform_device.h>
-#include <linux/jiffies.h>
-
-#include <video/omapdss.h>
-
-#include "dss.h"
-#include "dss_features.h"
-
-static ssize_t manager_name_show(struct omap_overlay_manager *mgr, char *buf)
-{
-	return snprintf(buf, PAGE_SIZE, "%s\n", mgr->name);
-}
-
-static ssize_t manager_display_show(struct omap_overlay_manager *mgr, char *buf)
-{
-	struct omap_dss_device *dssdev = mgr->get_device(mgr);
-
-	return snprintf(buf, PAGE_SIZE, "%s\n", dssdev ?
-			dssdev->name : "<none>");
-}
-
-static int manager_display_match(struct omap_dss_device *dssdev, void *data)
-{
-	const char *str = data;
-
-	return sysfs_streq(dssdev->name, str);
-}
-
-static ssize_t manager_display_store(struct omap_overlay_manager *mgr,
-		const char *buf, size_t size)
-{
-	int r = 0;
-	size_t len = size;
-	struct omap_dss_device *dssdev = NULL;
-	struct omap_dss_device *old_dssdev;
-
-	if (buf[size-1] == '\n')
-		--len;
-
-	if (len > 0)
-		dssdev = omap_dss_find_device((void *)buf,
-			manager_display_match);
-
-	if (len > 0 && dssdev == NULL)
-		return -EINVAL;
-
-	if (dssdev) {
-		DSSDBG("display %s found\n", dssdev->name);
-
-		if (omapdss_device_is_connected(dssdev)) {
-			DSSERR("new display is already connected\n");
-			r = -EINVAL;
-			goto put_device;
-		}
-
-		if (omapdss_device_is_enabled(dssdev)) {
-			DSSERR("new display is not disabled\n");
-			r = -EINVAL;
-			goto put_device;
-		}
-	}
-
-	old_dssdev = mgr->get_device(mgr);
-	if (old_dssdev) {
-		if (omapdss_device_is_enabled(old_dssdev)) {
-			DSSERR("old display is not disabled\n");
-			r = -EINVAL;
-			goto put_device;
-		}
-
-		old_dssdev->driver->disconnect(old_dssdev);
-	}
-
-	if (dssdev) {
-		r = dssdev->driver->connect(dssdev);
-		if (r) {
-			DSSERR("failed to connect new device\n");
-			goto put_device;
-		}
-
-		old_dssdev = mgr->get_device(mgr);
-		if (old_dssdev != dssdev) {
-			DSSERR("failed to connect device to this manager\n");
-			dssdev->driver->disconnect(dssdev);
-			goto put_device;
-		}
-
-		r = mgr->apply(mgr);
-		if (r) {
-			DSSERR("failed to apply dispc config\n");
-			goto put_device;
-		}
-	}
-
-put_device:
-	if (dssdev)
-		omap_dss_put_device(dssdev);
-
-	return r ? r : size;
-}
-
-static ssize_t manager_default_color_show(struct omap_overlay_manager *mgr,
-					  char *buf)
-{
-	struct omap_overlay_manager_info info;
-
-	mgr->get_manager_info(mgr, &info);
-
-	return snprintf(buf, PAGE_SIZE, "%#x\n", info.default_color);
-}
-
-static ssize_t manager_default_color_store(struct omap_overlay_manager *mgr,
-					   const char *buf, size_t size)
-{
-	struct omap_overlay_manager_info info;
-	u32 color;
-	int r;
-
-	r = kstrtouint(buf, 0, &color);
-	if (r)
-		return r;
-
-	mgr->get_manager_info(mgr, &info);
-
-	info.default_color = color;
-
-	r = mgr->set_manager_info(mgr, &info);
-	if (r)
-		return r;
-
-	r = mgr->apply(mgr);
-	if (r)
-		return r;
-
-	return size;
-}
-
-static const char *trans_key_type_str[] = {
-	"gfx-destination",
-	"video-source",
-};
-
-static ssize_t manager_trans_key_type_show(struct omap_overlay_manager *mgr,
-					   char *buf)
-{
-	enum omap_dss_trans_key_type key_type;
-	struct omap_overlay_manager_info info;
-
-	mgr->get_manager_info(mgr, &info);
-
-	key_type = info.trans_key_type;
-	BUG_ON(key_type >= ARRAY_SIZE(trans_key_type_str));
-
-	return snprintf(buf, PAGE_SIZE, "%s\n", trans_key_type_str[key_type]);
-}
-
-static ssize_t manager_trans_key_type_store(struct omap_overlay_manager *mgr,
-					    const char *buf, size_t size)
-{
-	enum omap_dss_trans_key_type key_type;
-	struct omap_overlay_manager_info info;
-	int r;
-
-	for (key_type = OMAP_DSS_COLOR_KEY_GFX_DST;
-			key_type < ARRAY_SIZE(trans_key_type_str); key_type++) {
-		if (sysfs_streq(buf, trans_key_type_str[key_type]))
-			break;
-	}
-
-	if (key_type == ARRAY_SIZE(trans_key_type_str))
-		return -EINVAL;
-
-	mgr->get_manager_info(mgr, &info);
-
-	info.trans_key_type = key_type;
-
-	r = mgr->set_manager_info(mgr, &info);
-	if (r)
-		return r;
-
-	r = mgr->apply(mgr);
-	if (r)
-		return r;
-
-	return size;
-}
-
-static ssize_t manager_trans_key_value_show(struct omap_overlay_manager *mgr,
-					    char *buf)
-{
-	struct omap_overlay_manager_info info;
-
-	mgr->get_manager_info(mgr, &info);
-
-	return snprintf(buf, PAGE_SIZE, "%#x\n", info.trans_key);
-}
-
-static ssize_t manager_trans_key_value_store(struct omap_overlay_manager *mgr,
-					     const char *buf, size_t size)
-{
-	struct omap_overlay_manager_info info;
-	u32 key_value;
-	int r;
-
-	r = kstrtouint(buf, 0, &key_value);
-	if (r)
-		return r;
-
-	mgr->get_manager_info(mgr, &info);
-
-	info.trans_key = key_value;
-
-	r = mgr->set_manager_info(mgr, &info);
-	if (r)
-		return r;
-
-	r = mgr->apply(mgr);
-	if (r)
-		return r;
-
-	return size;
-}
-
-static ssize_t manager_trans_key_enabled_show(struct omap_overlay_manager *mgr,
-					      char *buf)
-{
-	struct omap_overlay_manager_info info;
-
-	mgr->get_manager_info(mgr, &info);
-
-	return snprintf(buf, PAGE_SIZE, "%d\n", info.trans_enabled);
-}
-
-static ssize_t manager_trans_key_enabled_store(struct omap_overlay_manager *mgr,
-					       const char *buf, size_t size)
-{
-	struct omap_overlay_manager_info info;
-	bool enable;
-	int r;
-
-	r = strtobool(buf, &enable);
-	if (r)
-		return r;
-
-	mgr->get_manager_info(mgr, &info);
-
-	info.trans_enabled = enable;
-
-	r = mgr->set_manager_info(mgr, &info);
-	if (r)
-		return r;
-
-	r = mgr->apply(mgr);
-	if (r)
-		return r;
-
-	return size;
-}
-
-static ssize_t manager_alpha_blending_enabled_show(
-		struct omap_overlay_manager *mgr, char *buf)
-{
-	struct omap_overlay_manager_info info;
-
-	if(!dss_has_feature(FEAT_ALPHA_FIXED_ZORDER))
-		return -ENODEV;
-
-	mgr->get_manager_info(mgr, &info);
-
-	return snprintf(buf, PAGE_SIZE, "%d\n",
-		info.partial_alpha_enabled);
-}
-
-static ssize_t manager_alpha_blending_enabled_store(
-		struct omap_overlay_manager *mgr,
-		const char *buf, size_t size)
-{
-	struct omap_overlay_manager_info info;
-	bool enable;
-	int r;
-
-	if(!dss_has_feature(FEAT_ALPHA_FIXED_ZORDER))
-		return -ENODEV;
-
-	r = strtobool(buf, &enable);
-	if (r)
-		return r;
-
-	mgr->get_manager_info(mgr, &info);
-
-	info.partial_alpha_enabled = enable;
-
-	r = mgr->set_manager_info(mgr, &info);
-	if (r)
-		return r;
-
-	r = mgr->apply(mgr);
-	if (r)
-		return r;
-
-	return size;
-}
-
-static ssize_t manager_cpr_enable_show(struct omap_overlay_manager *mgr,
-		char *buf)
-{
-	struct omap_overlay_manager_info info;
-
-	mgr->get_manager_info(mgr, &info);
-
-	return snprintf(buf, PAGE_SIZE, "%d\n", info.cpr_enable);
-}
-
-static ssize_t manager_cpr_enable_store(struct omap_overlay_manager *mgr,
-		const char *buf, size_t size)
-{
-	struct omap_overlay_manager_info info;
-	int r;
-	bool enable;
-
-	if (!dss_has_feature(FEAT_CPR))
-		return -ENODEV;
-
-	r = strtobool(buf, &enable);
-	if (r)
-		return r;
-
-	mgr->get_manager_info(mgr, &info);
-
-	if (info.cpr_enable == enable)
-		return size;
-
-	info.cpr_enable = enable;
-
-	r = mgr->set_manager_info(mgr, &info);
-	if (r)
-		return r;
-
-	r = mgr->apply(mgr);
-	if (r)
-		return r;
-
-	return size;
-}
-
-static ssize_t manager_cpr_coef_show(struct omap_overlay_manager *mgr,
-		char *buf)
-{
-	struct omap_overlay_manager_info info;
-
-	mgr->get_manager_info(mgr, &info);
-
-	return snprintf(buf, PAGE_SIZE,
-			"%d %d %d %d %d %d %d %d %d\n",
-			info.cpr_coefs.rr,
-			info.cpr_coefs.rg,
-			info.cpr_coefs.rb,
-			info.cpr_coefs.gr,
-			info.cpr_coefs.gg,
-			info.cpr_coefs.gb,
-			info.cpr_coefs.br,
-			info.cpr_coefs.bg,
-			info.cpr_coefs.bb);
-}
-
-static ssize_t manager_cpr_coef_store(struct omap_overlay_manager *mgr,
-		const char *buf, size_t size)
-{
-	struct omap_overlay_manager_info info;
-	struct omap_dss_cpr_coefs coefs;
-	int r, i;
-	s16 *arr;
-
-	if (!dss_has_feature(FEAT_CPR))
-		return -ENODEV;
-
-	if (sscanf(buf, "%hd %hd %hd %hd %hd %hd %hd %hd %hd",
-				&coefs.rr, &coefs.rg, &coefs.rb,
-				&coefs.gr, &coefs.gg, &coefs.gb,
-				&coefs.br, &coefs.bg, &coefs.bb) != 9)
-		return -EINVAL;
-
-	arr = (s16[]){ coefs.rr, coefs.rg, coefs.rb,
-		coefs.gr, coefs.gg, coefs.gb,
-		coefs.br, coefs.bg, coefs.bb };
-
-	for (i = 0; i < 9; ++i) {
-		if (arr[i] < -512 || arr[i] > 511)
-			return -EINVAL;
-	}
-
-	mgr->get_manager_info(mgr, &info);
-
-	info.cpr_coefs = coefs;
-
-	r = mgr->set_manager_info(mgr, &info);
-	if (r)
-		return r;
-
-	r = mgr->apply(mgr);
-	if (r)
-		return r;
-
-	return size;
-}
-
-struct manager_attribute {
-	struct attribute attr;
-	ssize_t (*show)(struct omap_overlay_manager *, char *);
-	ssize_t	(*store)(struct omap_overlay_manager *, const char *, size_t);
-};
-
-#define MANAGER_ATTR(_name, _mode, _show, _store) \
-	struct manager_attribute manager_attr_##_name = \
-	__ATTR(_name, _mode, _show, _store)
-
-static MANAGER_ATTR(name, S_IRUGO, manager_name_show, NULL);
-static MANAGER_ATTR(display, S_IRUGO|S_IWUSR,
-		manager_display_show, manager_display_store);
-static MANAGER_ATTR(default_color, S_IRUGO|S_IWUSR,
-		manager_default_color_show, manager_default_color_store);
-static MANAGER_ATTR(trans_key_type, S_IRUGO|S_IWUSR,
-		manager_trans_key_type_show, manager_trans_key_type_store);
-static MANAGER_ATTR(trans_key_value, S_IRUGO|S_IWUSR,
-		manager_trans_key_value_show, manager_trans_key_value_store);
-static MANAGER_ATTR(trans_key_enabled, S_IRUGO|S_IWUSR,
-		manager_trans_key_enabled_show,
-		manager_trans_key_enabled_store);
-static MANAGER_ATTR(alpha_blending_enabled, S_IRUGO|S_IWUSR,
-		manager_alpha_blending_enabled_show,
-		manager_alpha_blending_enabled_store);
-static MANAGER_ATTR(cpr_enable, S_IRUGO|S_IWUSR,
-		manager_cpr_enable_show,
-		manager_cpr_enable_store);
-static MANAGER_ATTR(cpr_coef, S_IRUGO|S_IWUSR,
-		manager_cpr_coef_show,
-		manager_cpr_coef_store);
-
-
-static struct attribute *manager_sysfs_attrs[] = {
-	&manager_attr_name.attr,
-	&manager_attr_display.attr,
-	&manager_attr_default_color.attr,
-	&manager_attr_trans_key_type.attr,
-	&manager_attr_trans_key_value.attr,
-	&manager_attr_trans_key_enabled.attr,
-	&manager_attr_alpha_blending_enabled.attr,
-	&manager_attr_cpr_enable.attr,
-	&manager_attr_cpr_coef.attr,
-	NULL
-};
-
-static ssize_t manager_attr_show(struct kobject *kobj, struct attribute *attr,
-		char *buf)
-{
-	struct omap_overlay_manager *manager;
-	struct manager_attribute *manager_attr;
-
-	manager = container_of(kobj, struct omap_overlay_manager, kobj);
-	manager_attr = container_of(attr, struct manager_attribute, attr);
-
-	if (!manager_attr->show)
-		return -ENOENT;
-
-	return manager_attr->show(manager, buf);
-}
-
-static ssize_t manager_attr_store(struct kobject *kobj, struct attribute *attr,
-		const char *buf, size_t size)
-{
-	struct omap_overlay_manager *manager;
-	struct manager_attribute *manager_attr;
-
-	manager = container_of(kobj, struct omap_overlay_manager, kobj);
-	manager_attr = container_of(attr, struct manager_attribute, attr);
-
-	if (!manager_attr->store)
-		return -ENOENT;
-
-	return manager_attr->store(manager, buf, size);
-}
-
-static const struct sysfs_ops manager_sysfs_ops = {
-	.show = manager_attr_show,
-	.store = manager_attr_store,
-};
-
-static struct kobj_type manager_ktype = {
-	.sysfs_ops = &manager_sysfs_ops,
-	.default_attrs = manager_sysfs_attrs,
-};
-
-int dss_manager_kobj_init(struct omap_overlay_manager *mgr,
-		struct platform_device *pdev)
-{
-	return kobject_init_and_add(&mgr->kobj, &manager_ktype,
-			&pdev->dev.kobj, "manager%d", mgr->id);
-}
-
-void dss_manager_kobj_uninit(struct omap_overlay_manager *mgr)
-{
-	kobject_del(&mgr->kobj);
-	kobject_put(&mgr->kobj);
-
-	memset(&mgr->kobj, 0, sizeof(mgr->kobj));
-}
diff --git a/drivers/gpu/drm/omapdrm/dss/manager.c b/drivers/gpu/drm/omapdrm/dss/manager.c
deleted file mode 100644
index 08a67f4f6a20..000000000000
--- a/drivers/gpu/drm/omapdrm/dss/manager.c
+++ /dev/null
@@ -1,263 +0,0 @@
-/*
- * linux/drivers/video/omap2/dss/manager.c
- *
- * Copyright (C) 2009 Nokia Corporation
- * Author: Tomi Valkeinen <tomi.valkeinen@nokia.com>
- *
- * Some code and ideas taken from drivers/video/omap/ driver
- * by Imre Deak.
- *
- * This program is free software; you can redistribute it and/or modify it
- * under the terms of the GNU General Public License version 2 as published by
- * the Free Software Foundation.
- *
- * This program is distributed in the hope that it will be useful, but WITHOUT
- * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
- * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
- * more details.
- *
- * You should have received a copy of the GNU General Public License along with
- * this program.  If not, see <http://www.gnu.org/licenses/>.
- */
-
-#define DSS_SUBSYS_NAME "MANAGER"
-
-#include <linux/kernel.h>
-#include <linux/slab.h>
-#include <linux/module.h>
-#include <linux/platform_device.h>
-#include <linux/jiffies.h>
-
-#include <video/omapdss.h>
-
-#include "dss.h"
-#include "dss_features.h"
-
-static int num_managers;
-static struct omap_overlay_manager *managers;
-
-int dss_init_overlay_managers(void)
-{
-	int i;
-
-	num_managers = dss_feat_get_num_mgrs();
-
-	managers = kzalloc(sizeof(struct omap_overlay_manager) * num_managers,
-			GFP_KERNEL);
-
-	BUG_ON(managers == NULL);
-
-	for (i = 0; i < num_managers; ++i) {
-		struct omap_overlay_manager *mgr = &managers[i];
-
-		switch (i) {
-		case 0:
-			mgr->name = "lcd";
-			mgr->id = OMAP_DSS_CHANNEL_LCD;
-			break;
-		case 1:
-			mgr->name = "tv";
-			mgr->id = OMAP_DSS_CHANNEL_DIGIT;
-			break;
-		case 2:
-			mgr->name = "lcd2";
-			mgr->id = OMAP_DSS_CHANNEL_LCD2;
-			break;
-		case 3:
-			mgr->name = "lcd3";
-			mgr->id = OMAP_DSS_CHANNEL_LCD3;
-			break;
-		}
-
-		mgr->caps = 0;
-		mgr->supported_displays =
-			dss_feat_get_supported_displays(mgr->id);
-		mgr->supported_outputs =
-			dss_feat_get_supported_outputs(mgr->id);
-
-		INIT_LIST_HEAD(&mgr->overlays);
-	}
-
-	return 0;
-}
-
-int dss_init_overlay_managers_sysfs(struct platform_device *pdev)
-{
-	int i, r;
-
-	for (i = 0; i < num_managers; ++i) {
-		struct omap_overlay_manager *mgr = &managers[i];
-
-		r = dss_manager_kobj_init(mgr, pdev);
-		if (r)
-			DSSERR("failed to create sysfs file\n");
-	}
-
-	return 0;
-}
-
-void dss_uninit_overlay_managers(void)
-{
-	kfree(managers);
-	managers = NULL;
-	num_managers = 0;
-}
-
-void dss_uninit_overlay_managers_sysfs(struct platform_device *pdev)
-{
-	int i;
-
-	for (i = 0; i < num_managers; ++i) {
-		struct omap_overlay_manager *mgr = &managers[i];
-
-		dss_manager_kobj_uninit(mgr);
-	}
-}
-
-int omap_dss_get_num_overlay_managers(void)
-{
-	return num_managers;
-}
-EXPORT_SYMBOL(omap_dss_get_num_overlay_managers);
-
-struct omap_overlay_manager *omap_dss_get_overlay_manager(int num)
-{
-	if (num >= num_managers)
-		return NULL;
-
-	return &managers[num];
-}
-EXPORT_SYMBOL(omap_dss_get_overlay_manager);
-
-int dss_mgr_simple_check(struct omap_overlay_manager *mgr,
-		const struct omap_overlay_manager_info *info)
-{
-	if (dss_has_feature(FEAT_ALPHA_FIXED_ZORDER)) {
-		/*
-		 * OMAP3 supports only graphics source transparency color key
-		 * and alpha blending simultaneously. See TRM 15.4.2.4.2.2
-		 * Alpha Mode.
-		 */
-		if (info->partial_alpha_enabled && info->trans_enabled
-			&& info->trans_key_type != OMAP_DSS_COLOR_KEY_GFX_DST) {
-			DSSERR("check_manager: illegal transparency key\n");
-			return -EINVAL;
-		}
-	}
-
-	return 0;
-}
-
-static int dss_mgr_check_zorder(struct omap_overlay_manager *mgr,
-		struct omap_overlay_info **overlay_infos)
-{
-	struct omap_overlay *ovl1, *ovl2;
-	struct omap_overlay_info *info1, *info2;
-
-	list_for_each_entry(ovl1, &mgr->overlays, list) {
-		info1 = overlay_infos[ovl1->id];
-
-		if (info1 == NULL)
-			continue;
-
-		list_for_each_entry(ovl2, &mgr->overlays, list) {
-			if (ovl1 == ovl2)
-				continue;
-
-			info2 = overlay_infos[ovl2->id];
-
-			if (info2 == NULL)
-				continue;
-
-			if (info1->zorder == info2->zorder) {
-				DSSERR("overlays %d and %d have the same "
-						"zorder %d\n",
-					ovl1->id, ovl2->id, info1->zorder);
-				return -EINVAL;
-			}
-		}
-	}
-
-	return 0;
-}
-
-int dss_mgr_check_timings(struct omap_overlay_manager *mgr,
-		const struct omap_video_timings *timings)
-{
-	if (!dispc_mgr_timings_ok(mgr->id, timings)) {
-		DSSERR("check_manager: invalid timings\n");
-		return -EINVAL;
-	}
-
-	return 0;
-}
-
-static int dss_mgr_check_lcd_config(struct omap_overlay_manager *mgr,
-		const struct dss_lcd_mgr_config *config)
-{
-	struct dispc_clock_info cinfo = config->clock_info;
-	int dl = config->video_port_width;
-	bool stallmode = config->stallmode;
-	bool fifohandcheck = config->fifohandcheck;
-
-	if (cinfo.lck_div < 1 || cinfo.lck_div > 255)
-		return -EINVAL;
-
-	if (cinfo.pck_div < 1 || cinfo.pck_div > 255)
-		return -EINVAL;
-
-	if (dl != 12 && dl != 16 && dl != 18 && dl != 24)
-		return -EINVAL;
-
-	/* fifohandcheck should be used only with stallmode */
-	if (!stallmode && fifohandcheck)
-		return -EINVAL;
-
-	/*
-	 * io pad mode can be only checked by using dssdev connected to the
-	 * manager. Ignore checking these for now, add checks when manager
-	 * is capable of holding information related to the connected interface
-	 */
-
-	return 0;
-}
-
-int dss_mgr_check(struct omap_overlay_manager *mgr,
-		struct omap_overlay_manager_info *info,
-		const struct omap_video_timings *mgr_timings,
-		const struct dss_lcd_mgr_config *lcd_config,
-		struct omap_overlay_info **overlay_infos)
-{
-	struct omap_overlay *ovl;
-	int r;
-
-	if (dss_has_feature(FEAT_ALPHA_FREE_ZORDER)) {
-		r = dss_mgr_check_zorder(mgr, overlay_infos);
-		if (r)
-			return r;
-	}
-
-	r = dss_mgr_check_timings(mgr, mgr_timings);
-	if (r)
-		return r;
-
-	r = dss_mgr_check_lcd_config(mgr, lcd_config);
-	if (r)
-		return r;
-
-	list_for_each_entry(ovl, &mgr->overlays, list) {
-		struct omap_overlay_info *oi;
-		int r;
-
-		oi = overlay_infos[ovl->id];
-
-		if (oi == NULL)
-			continue;
-
-		r = dss_ovl_check(ovl, oi, mgr_timings);
-		if (r)
-			return r;
-	}
-
-	return 0;
-}
diff --git a/drivers/gpu/drm/omapdrm/dss/omapdss.h b/drivers/gpu/drm/omapdrm/dss/omapdss.h
new file mode 100644
index 000000000000..d7e7c909bbc2
--- /dev/null
+++ b/drivers/gpu/drm/omapdrm/dss/omapdss.h
@@ -0,0 +1,96 @@
+/*
+ * Copyright (C) 2016 Texas Instruments
+ * Author: Tomi Valkeinen <tomi.valkeinen@ti.com>
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms of the GNU General Public License version 2 as published by
+ * the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful, but WITHOUT
+ * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
+ * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
+ * more details.
+ *
+ * You should have received a copy of the GNU General Public License along with
+ * this program.  If not, see <http://www.gnu.org/licenses/>.
+ */
+
+#ifndef __OMAP_DRM_DSS_H
+#define __OMAP_DRM_DSS_H
+
+#include <video/omapdss.h>
+
+u32 dispc_read_irqstatus(void);
+void dispc_clear_irqstatus(u32 mask);
+u32 dispc_read_irqenable(void);
+void dispc_write_irqenable(u32 mask);
+
+int dispc_request_irq(irq_handler_t handler, void *dev_id);
+void dispc_free_irq(void *dev_id);
+
+int dispc_runtime_get(void);
+void dispc_runtime_put(void);
+
+void dispc_mgr_enable(enum omap_channel channel, bool enable);
+bool dispc_mgr_is_enabled(enum omap_channel channel);
+u32 dispc_mgr_get_vsync_irq(enum omap_channel channel);
+u32 dispc_mgr_get_framedone_irq(enum omap_channel channel);
+u32 dispc_mgr_get_sync_lost_irq(enum omap_channel channel);
+bool dispc_mgr_go_busy(enum omap_channel channel);
+void dispc_mgr_go(enum omap_channel channel);
+void dispc_mgr_set_lcd_config(enum omap_channel channel,
+		const struct dss_lcd_mgr_config *config);
+void dispc_mgr_set_timings(enum omap_channel channel,
+		const struct omap_video_timings *timings);
+void dispc_mgr_setup(enum omap_channel channel,
+		const struct omap_overlay_manager_info *info);
+
+int dispc_ovl_enable(enum omap_plane plane, bool enable);
+bool dispc_ovl_enabled(enum omap_plane plane);
+void dispc_ovl_set_channel_out(enum omap_plane plane,
+		enum omap_channel channel);
+int dispc_ovl_setup(enum omap_plane plane, const struct omap_overlay_info *oi,
+		bool replication, const struct omap_video_timings *mgr_timings,
+		bool mem_to_mem);
+
+enum omap_dss_output_id dispc_mgr_get_supported_outputs(enum omap_channel channel);
+
+struct dss_mgr_ops {
+	int (*connect)(enum omap_channel channel,
+		struct omap_dss_device *dst);
+	void (*disconnect)(enum omap_channel channel,
+		struct omap_dss_device *dst);
+
+	void (*start_update)(enum omap_channel channel);
+	int (*enable)(enum omap_channel channel);
+	void (*disable)(enum omap_channel channel);
+	void (*set_timings)(enum omap_channel channel,
+			const struct omap_video_timings *timings);
+	void (*set_lcd_config)(enum omap_channel channel,
+			const struct dss_lcd_mgr_config *config);
+	int (*register_framedone_handler)(enum omap_channel channel,
+			void (*handler)(void *), void *data);
+	void (*unregister_framedone_handler)(enum omap_channel channel,
+			void (*handler)(void *), void *data);
+};
+
+int dss_install_mgr_ops(const struct dss_mgr_ops *mgr_ops);
+void dss_uninstall_mgr_ops(void);
+
+int dss_mgr_connect(enum omap_channel channel,
+		struct omap_dss_device *dst);
+void dss_mgr_disconnect(enum omap_channel channel,
+		struct omap_dss_device *dst);
+void dss_mgr_set_timings(enum omap_channel channel,
+		const struct omap_video_timings *timings);
+void dss_mgr_set_lcd_config(enum omap_channel channel,
+		const struct dss_lcd_mgr_config *config);
+int dss_mgr_enable(enum omap_channel channel);
+void dss_mgr_disable(enum omap_channel channel);
+void dss_mgr_start_update(enum omap_channel channel);
+int dss_mgr_register_framedone_handler(enum omap_channel channel,
+		void (*handler)(void *), void *data);
+void dss_mgr_unregister_framedone_handler(enum omap_channel channel,
+		void (*handler)(void *), void *data);
+
+#endif /* __OMAP_DRM_DSS_H */
diff --git a/drivers/gpu/drm/omapdrm/dss/output.c b/drivers/gpu/drm/omapdrm/dss/output.c
index 16072159bd24..829232ad8c81 100644
--- a/drivers/gpu/drm/omapdrm/dss/output.c
+++ b/drivers/gpu/drm/omapdrm/dss/output.c
@@ -169,24 +169,6 @@ struct omap_dss_device *omapdss_find_output_from_display(struct omap_dss_device
 }
 EXPORT_SYMBOL(omapdss_find_output_from_display);
 
-struct omap_overlay_manager *omapdss_find_mgr_from_display(struct omap_dss_device *dssdev)
-{
-	struct omap_dss_device *out;
-	struct omap_overlay_manager *mgr;
-
-	out = omapdss_find_output_from_display(dssdev);
-
-	if (out == NULL)
-		return NULL;
-
-	mgr = out->manager;
-
-	omap_dss_put_device(out);
-
-	return mgr;
-}
-EXPORT_SYMBOL(omapdss_find_mgr_from_display);
-
 static const struct dss_mgr_ops *dss_mgr_ops;
 
 int dss_install_mgr_ops(const struct dss_mgr_ops *mgr_ops)
@@ -206,62 +188,62 @@ void dss_uninstall_mgr_ops(void)
 }
 EXPORT_SYMBOL(dss_uninstall_mgr_ops);
 
-int dss_mgr_connect(struct omap_overlay_manager *mgr,
+int dss_mgr_connect(enum omap_channel channel,
 		struct omap_dss_device *dst)
 {
-	return dss_mgr_ops->connect(mgr, dst);
+	return dss_mgr_ops->connect(channel, dst);
 }
 EXPORT_SYMBOL(dss_mgr_connect);
 
-void dss_mgr_disconnect(struct omap_overlay_manager *mgr,
+void dss_mgr_disconnect(enum omap_channel channel,
 		struct omap_dss_device *dst)
 {
-	dss_mgr_ops->disconnect(mgr, dst);
+	dss_mgr_ops->disconnect(channel, dst);
 }
 EXPORT_SYMBOL(dss_mgr_disconnect);
 
-void dss_mgr_set_timings(struct omap_overlay_manager *mgr,
+void dss_mgr_set_timings(enum omap_channel channel,
 		const struct omap_video_timings *timings)
 {
-	dss_mgr_ops->set_timings(mgr, timings);
+	dss_mgr_ops->set_timings(channel, timings);
 }
 EXPORT_SYMBOL(dss_mgr_set_timings);
 
-void dss_mgr_set_lcd_config(struct omap_overlay_manager *mgr,
+void dss_mgr_set_lcd_config(enum omap_channel channel,
 		const struct dss_lcd_mgr_config *config)
 {
-	dss_mgr_ops->set_lcd_config(mgr, config);
+	dss_mgr_ops->set_lcd_config(channel, config);
 }
 EXPORT_SYMBOL(dss_mgr_set_lcd_config);
 
-int dss_mgr_enable(struct omap_overlay_manager *mgr)
+int dss_mgr_enable(enum omap_channel channel)
 {
-	return dss_mgr_ops->enable(mgr);
+	return dss_mgr_ops->enable(channel);
 }
 EXPORT_SYMBOL(dss_mgr_enable);
 
-void dss_mgr_disable(struct omap_overlay_manager *mgr)
+void dss_mgr_disable(enum omap_channel channel)
 {
-	dss_mgr_ops->disable(mgr);
+	dss_mgr_ops->disable(channel);
 }
 EXPORT_SYMBOL(dss_mgr_disable);
 
-void dss_mgr_start_update(struct omap_overlay_manager *mgr)
+void dss_mgr_start_update(enum omap_channel channel)
 {
-	dss_mgr_ops->start_update(mgr);
+	dss_mgr_ops->start_update(channel);
 }
 EXPORT_SYMBOL(dss_mgr_start_update);
 
-int dss_mgr_register_framedone_handler(struct omap_overlay_manager *mgr,
+int dss_mgr_register_framedone_handler(enum omap_channel channel,
 		void (*handler)(void *), void *data)
 {
-	return dss_mgr_ops->register_framedone_handler(mgr, handler, data);
+	return dss_mgr_ops->register_framedone_handler(channel, handler, data);
 }
 EXPORT_SYMBOL(dss_mgr_register_framedone_handler);
 
-void dss_mgr_unregister_framedone_handler(struct omap_overlay_manager *mgr,
+void dss_mgr_unregister_framedone_handler(enum omap_channel channel,
 		void (*handler)(void *), void *data)
 {
-	dss_mgr_ops->unregister_framedone_handler(mgr, handler, data);
+	dss_mgr_ops->unregister_framedone_handler(channel, handler, data);
 }
 EXPORT_SYMBOL(dss_mgr_unregister_framedone_handler);
diff --git a/drivers/gpu/drm/omapdrm/dss/overlay-sysfs.c b/drivers/gpu/drm/omapdrm/dss/overlay-sysfs.c
deleted file mode 100644
index 4cc5ddebfb34..000000000000
--- a/drivers/gpu/drm/omapdrm/dss/overlay-sysfs.c
+++ /dev/null
@@ -1,456 +0,0 @@
-/*
- * Copyright (C) 2009 Nokia Corporation
- * Author: Tomi Valkeinen <tomi.valkeinen@nokia.com>
- *
- * Some code and ideas taken from drivers/video/omap/ driver
- * by Imre Deak.
- *
- * This program is free software; you can redistribute it and/or modify it
- * under the terms of the GNU General Public License version 2 as published by
- * the Free Software Foundation.
- *
- * This program is distributed in the hope that it will be useful, but WITHOUT
- * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
- * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
- * more details.
- *
- * You should have received a copy of the GNU General Public License along with
- * this program.  If not, see <http://www.gnu.org/licenses/>.
- */
-
-#define DSS_SUBSYS_NAME "OVERLAY"
-
-#include <linux/module.h>
-#include <linux/err.h>
-#include <linux/sysfs.h>
-#include <linux/kobject.h>
-#include <linux/platform_device.h>
-
-#include <video/omapdss.h>
-
-#include "dss.h"
-#include "dss_features.h"
-
-static ssize_t overlay_name_show(struct omap_overlay *ovl, char *buf)
-{
-	return snprintf(buf, PAGE_SIZE, "%s\n", ovl->name);
-}
-
-static ssize_t overlay_manager_show(struct omap_overlay *ovl, char *buf)
-{
-	return snprintf(buf, PAGE_SIZE, "%s\n",
-			ovl->manager ? ovl->manager->name : "<none>");
-}
-
-static ssize_t overlay_manager_store(struct omap_overlay *ovl, const char *buf,
-		size_t size)
-{
-	int i, r;
-	struct omap_overlay_manager *mgr = NULL;
-	struct omap_overlay_manager *old_mgr;
-	int len = size;
-
-	if (buf[size-1] == '\n')
-		--len;
-
-	if (len > 0) {
-		for (i = 0; i < omap_dss_get_num_overlay_managers(); ++i) {
-			mgr = omap_dss_get_overlay_manager(i);
-
-			if (sysfs_streq(buf, mgr->name))
-				break;
-
-			mgr = NULL;
-		}
-	}
-
-	if (len > 0 && mgr == NULL)
-		return -EINVAL;
-
-	if (mgr)
-		DSSDBG("manager %s found\n", mgr->name);
-
-	if (mgr == ovl->manager)
-		return size;
-
-	old_mgr = ovl->manager;
-
-	r = dispc_runtime_get();
-	if (r)
-		return r;
-
-	/* detach old manager */
-	if (old_mgr) {
-		r = ovl->unset_manager(ovl);
-		if (r) {
-			DSSERR("detach failed\n");
-			goto err;
-		}
-
-		r = old_mgr->apply(old_mgr);
-		if (r)
-			goto err;
-	}
-
-	if (mgr) {
-		r = ovl->set_manager(ovl, mgr);
-		if (r) {
-			DSSERR("Failed to attach overlay\n");
-			goto err;
-		}
-
-		r = mgr->apply(mgr);
-		if (r)
-			goto err;
-	}
-
-	dispc_runtime_put();
-
-	return size;
-
-err:
-	dispc_runtime_put();
-	return r;
-}
-
-static ssize_t overlay_input_size_show(struct omap_overlay *ovl, char *buf)
-{
-	struct omap_overlay_info info;
-
-	ovl->get_overlay_info(ovl, &info);
-
-	return snprintf(buf, PAGE_SIZE, "%d,%d\n",
-			info.width, info.height);
-}
-
-static ssize_t overlay_screen_width_show(struct omap_overlay *ovl, char *buf)
-{
-	struct omap_overlay_info info;
-
-	ovl->get_overlay_info(ovl, &info);
-
-	return snprintf(buf, PAGE_SIZE, "%d\n", info.screen_width);
-}
-
-static ssize_t overlay_position_show(struct omap_overlay *ovl, char *buf)
-{
-	struct omap_overlay_info info;
-
-	ovl->get_overlay_info(ovl, &info);
-
-	return snprintf(buf, PAGE_SIZE, "%d,%d\n",
-			info.pos_x, info.pos_y);
-}
-
-static ssize_t overlay_position_store(struct omap_overlay *ovl,
-		const char *buf, size_t size)
-{
-	int r;
-	char *last;
-	struct omap_overlay_info info;
-
-	ovl->get_overlay_info(ovl, &info);
-
-	info.pos_x = simple_strtoul(buf, &last, 10);
-	++last;
-	if (last - buf >= size)
-		return -EINVAL;
-
-	info.pos_y = simple_strtoul(last, &last, 10);
-
-	r = ovl->set_overlay_info(ovl, &info);
-	if (r)
-		return r;
-
-	if (ovl->manager) {
-		r = ovl->manager->apply(ovl->manager);
-		if (r)
-			return r;
-	}
-
-	return size;
-}
-
-static ssize_t overlay_output_size_show(struct omap_overlay *ovl, char *buf)
-{
-	struct omap_overlay_info info;
-
-	ovl->get_overlay_info(ovl, &info);
-
-	return snprintf(buf, PAGE_SIZE, "%d,%d\n",
-			info.out_width, info.out_height);
-}
-
-static ssize_t overlay_output_size_store(struct omap_overlay *ovl,
-		const char *buf, size_t size)
-{
-	int r;
-	char *last;
-	struct omap_overlay_info info;
-
-	ovl->get_overlay_info(ovl, &info);
-
-	info.out_width = simple_strtoul(buf, &last, 10);
-	++last;
-	if (last - buf >= size)
-		return -EINVAL;
-
-	info.out_height = simple_strtoul(last, &last, 10);
-
-	r = ovl->set_overlay_info(ovl, &info);
-	if (r)
-		return r;
-
-	if (ovl->manager) {
-		r = ovl->manager->apply(ovl->manager);
-		if (r)
-			return r;
-	}
-
-	return size;
-}
-
-static ssize_t overlay_enabled_show(struct omap_overlay *ovl, char *buf)
-{
-	return snprintf(buf, PAGE_SIZE, "%d\n", ovl->is_enabled(ovl));
-}
-
-static ssize_t overlay_enabled_store(struct omap_overlay *ovl, const char *buf,
-		size_t size)
-{
-	int r;
-	bool enable;
-
-	r = strtobool(buf, &enable);
-	if (r)
-		return r;
-
-	if (enable)
-		r = ovl->enable(ovl);
-	else
-		r = ovl->disable(ovl);
-
-	if (r)
-		return r;
-
-	return size;
-}
-
-static ssize_t overlay_global_alpha_show(struct omap_overlay *ovl, char *buf)
-{
-	struct omap_overlay_info info;
-
-	ovl->get_overlay_info(ovl, &info);
-
-	return snprintf(buf, PAGE_SIZE, "%d\n",
-			info.global_alpha);
-}
-
-static ssize_t overlay_global_alpha_store(struct omap_overlay *ovl,
-		const char *buf, size_t size)
-{
-	int r;
-	u8 alpha;
-	struct omap_overlay_info info;
-
-	if ((ovl->caps & OMAP_DSS_OVL_CAP_GLOBAL_ALPHA) == 0)
-		return -ENODEV;
-
-	r = kstrtou8(buf, 0, &alpha);
-	if (r)
-		return r;
-
-	ovl->get_overlay_info(ovl, &info);
-
-	info.global_alpha = alpha;
-
-	r = ovl->set_overlay_info(ovl, &info);
-	if (r)
-		return r;
-
-	if (ovl->manager) {
-		r = ovl->manager->apply(ovl->manager);
-		if (r)
-			return r;
-	}
-
-	return size;
-}
-
-static ssize_t overlay_pre_mult_alpha_show(struct omap_overlay *ovl,
-		char *buf)
-{
-	struct omap_overlay_info info;
-
-	ovl->get_overlay_info(ovl, &info);
-
-	return snprintf(buf, PAGE_SIZE, "%d\n",
-			info.pre_mult_alpha);
-}
-
-static ssize_t overlay_pre_mult_alpha_store(struct omap_overlay *ovl,
-		const char *buf, size_t size)
-{
-	int r;
-	u8 alpha;
-	struct omap_overlay_info info;
-
-	if ((ovl->caps & OMAP_DSS_OVL_CAP_PRE_MULT_ALPHA) == 0)
-		return -ENODEV;
-
-	r = kstrtou8(buf, 0, &alpha);
-	if (r)
-		return r;
-
-	ovl->get_overlay_info(ovl, &info);
-
-	info.pre_mult_alpha = alpha;
-
-	r = ovl->set_overlay_info(ovl, &info);
-	if (r)
-		return r;
-
-	if (ovl->manager) {
-		r = ovl->manager->apply(ovl->manager);
-		if (r)
-			return r;
-	}
-
-	return size;
-}
-
-static ssize_t overlay_zorder_show(struct omap_overlay *ovl, char *buf)
-{
-	struct omap_overlay_info info;
-
-	ovl->get_overlay_info(ovl, &info);
-
-	return snprintf(buf, PAGE_SIZE, "%d\n", info.zorder);
-}
-
-static ssize_t overlay_zorder_store(struct omap_overlay *ovl,
-		const char *buf, size_t size)
-{
-	int r;
-	u8 zorder;
-	struct omap_overlay_info info;
-
-	if ((ovl->caps & OMAP_DSS_OVL_CAP_ZORDER) == 0)
-		return -ENODEV;
-
-	r = kstrtou8(buf, 0, &zorder);
-	if (r)
-		return r;
-
-	ovl->get_overlay_info(ovl, &info);
-
-	info.zorder = zorder;
-
-	r = ovl->set_overlay_info(ovl, &info);
-	if (r)
-		return r;
-
-	if (ovl->manager) {
-		r = ovl->manager->apply(ovl->manager);
-		if (r)
-			return r;
-	}
-
-	return size;
-}
-
-struct overlay_attribute {
-	struct attribute attr;
-	ssize_t (*show)(struct omap_overlay *, char *);
-	ssize_t	(*store)(struct omap_overlay *, const char *, size_t);
-};
-
-#define OVERLAY_ATTR(_name, _mode, _show, _store) \
-	struct overlay_attribute overlay_attr_##_name = \
-	__ATTR(_name, _mode, _show, _store)
-
-static OVERLAY_ATTR(name, S_IRUGO, overlay_name_show, NULL);
-static OVERLAY_ATTR(manager, S_IRUGO|S_IWUSR,
-		overlay_manager_show, overlay_manager_store);
-static OVERLAY_ATTR(input_size, S_IRUGO, overlay_input_size_show, NULL);
-static OVERLAY_ATTR(screen_width, S_IRUGO, overlay_screen_width_show, NULL);
-static OVERLAY_ATTR(position, S_IRUGO|S_IWUSR,
-		overlay_position_show, overlay_position_store);
-static OVERLAY_ATTR(output_size, S_IRUGO|S_IWUSR,
-		overlay_output_size_show, overlay_output_size_store);
-static OVERLAY_ATTR(enabled, S_IRUGO|S_IWUSR,
-		overlay_enabled_show, overlay_enabled_store);
-static OVERLAY_ATTR(global_alpha, S_IRUGO|S_IWUSR,
-		overlay_global_alpha_show, overlay_global_alpha_store);
-static OVERLAY_ATTR(pre_mult_alpha, S_IRUGO|S_IWUSR,
-		overlay_pre_mult_alpha_show,
-		overlay_pre_mult_alpha_store);
-static OVERLAY_ATTR(zorder, S_IRUGO|S_IWUSR,
-		overlay_zorder_show, overlay_zorder_store);
-
-static struct attribute *overlay_sysfs_attrs[] = {
-	&overlay_attr_name.attr,
-	&overlay_attr_manager.attr,
-	&overlay_attr_input_size.attr,
-	&overlay_attr_screen_width.attr,
-	&overlay_attr_position.attr,
-	&overlay_attr_output_size.attr,
-	&overlay_attr_enabled.attr,
-	&overlay_attr_global_alpha.attr,
-	&overlay_attr_pre_mult_alpha.attr,
-	&overlay_attr_zorder.attr,
-	NULL
-};
-
-static ssize_t overlay_attr_show(struct kobject *kobj, struct attribute *attr,
-		char *buf)
-{
-	struct omap_overlay *overlay;
-	struct overlay_attribute *overlay_attr;
-
-	overlay = container_of(kobj, struct omap_overlay, kobj);
-	overlay_attr = container_of(attr, struct overlay_attribute, attr);
-
-	if (!overlay_attr->show)
-		return -ENOENT;
-
-	return overlay_attr->show(overlay, buf);
-}
-
-static ssize_t overlay_attr_store(struct kobject *kobj, struct attribute *attr,
-		const char *buf, size_t size)
-{
-	struct omap_overlay *overlay;
-	struct overlay_attribute *overlay_attr;
-
-	overlay = container_of(kobj, struct omap_overlay, kobj);
-	overlay_attr = container_of(attr, struct overlay_attribute, attr);
-
-	if (!overlay_attr->store)
-		return -ENOENT;
-
-	return overlay_attr->store(overlay, buf, size);
-}
-
-static const struct sysfs_ops overlay_sysfs_ops = {
-	.show = overlay_attr_show,
-	.store = overlay_attr_store,
-};
-
-static struct kobj_type overlay_ktype = {
-	.sysfs_ops = &overlay_sysfs_ops,
-	.default_attrs = overlay_sysfs_attrs,
-};
-
-int dss_overlay_kobj_init(struct omap_overlay *ovl,
-		struct platform_device *pdev)
-{
-	return kobject_init_and_add(&ovl->kobj, &overlay_ktype,
-			&pdev->dev.kobj, "overlay%d", ovl->id);
-}
-
-void dss_overlay_kobj_uninit(struct omap_overlay *ovl)
-{
-	kobject_del(&ovl->kobj);
-	kobject_put(&ovl->kobj);
-}
diff --git a/drivers/gpu/drm/omapdrm/dss/overlay.c b/drivers/gpu/drm/omapdrm/dss/overlay.c
deleted file mode 100644
index 2f7cee985cdd..000000000000
--- a/drivers/gpu/drm/omapdrm/dss/overlay.c
+++ /dev/null
@@ -1,202 +0,0 @@
-/*
- * linux/drivers/video/omap2/dss/overlay.c
- *
- * Copyright (C) 2009 Nokia Corporation
- * Author: Tomi Valkeinen <tomi.valkeinen@nokia.com>
- *
- * Some code and ideas taken from drivers/video/omap/ driver
- * by Imre Deak.
- *
- * This program is free software; you can redistribute it and/or modify it
- * under the terms of the GNU General Public License version 2 as published by
- * the Free Software Foundation.
- *
- * This program is distributed in the hope that it will be useful, but WITHOUT
- * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
- * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
- * more details.
- *
- * You should have received a copy of the GNU General Public License along with
- * this program.  If not, see <http://www.gnu.org/licenses/>.
- */
-
-#define DSS_SUBSYS_NAME "OVERLAY"
-
-#include <linux/kernel.h>
-#include <linux/module.h>
-#include <linux/err.h>
-#include <linux/sysfs.h>
-#include <linux/platform_device.h>
-#include <linux/delay.h>
-#include <linux/slab.h>
-
-#include <video/omapdss.h>
-
-#include "dss.h"
-#include "dss_features.h"
-
-static int num_overlays;
-static struct omap_overlay *overlays;
-
-int omap_dss_get_num_overlays(void)
-{
-	return num_overlays;
-}
-EXPORT_SYMBOL(omap_dss_get_num_overlays);
-
-struct omap_overlay *omap_dss_get_overlay(int num)
-{
-	if (num >= num_overlays)
-		return NULL;
-
-	return &overlays[num];
-}
-EXPORT_SYMBOL(omap_dss_get_overlay);
-
-void dss_init_overlays(struct platform_device *pdev)
-{
-	int i, r;
-
-	num_overlays = dss_feat_get_num_ovls();
-
-	overlays = kzalloc(sizeof(struct omap_overlay) * num_overlays,
-			GFP_KERNEL);
-
-	BUG_ON(overlays == NULL);
-
-	for (i = 0; i < num_overlays; ++i) {
-		struct omap_overlay *ovl = &overlays[i];
-
-		switch (i) {
-		case 0:
-			ovl->name = "gfx";
-			ovl->id = OMAP_DSS_GFX;
-			break;
-		case 1:
-			ovl->name = "vid1";
-			ovl->id = OMAP_DSS_VIDEO1;
-			break;
-		case 2:
-			ovl->name = "vid2";
-			ovl->id = OMAP_DSS_VIDEO2;
-			break;
-		case 3:
-			ovl->name = "vid3";
-			ovl->id = OMAP_DSS_VIDEO3;
-			break;
-		}
-
-		ovl->caps = dss_feat_get_overlay_caps(ovl->id);
-		ovl->supported_modes =
-			dss_feat_get_supported_color_modes(ovl->id);
-
-		r = dss_overlay_kobj_init(ovl, pdev);
-		if (r)
-			DSSERR("failed to create sysfs file\n");
-	}
-}
-
-void dss_uninit_overlays(struct platform_device *pdev)
-{
-	int i;
-
-	for (i = 0; i < num_overlays; ++i) {
-		struct omap_overlay *ovl = &overlays[i];
-		dss_overlay_kobj_uninit(ovl);
-	}
-
-	kfree(overlays);
-	overlays = NULL;
-	num_overlays = 0;
-}
-
-int dss_ovl_simple_check(struct omap_overlay *ovl,
-		const struct omap_overlay_info *info)
-{
-	if ((ovl->caps & OMAP_DSS_OVL_CAP_SCALE) == 0) {
-		if (info->out_width != 0 && info->width != info->out_width) {
-			DSSERR("check_overlay: overlay %d doesn't support "
-					"scaling\n", ovl->id);
-			return -EINVAL;
-		}
-
-		if (info->out_height != 0 && info->height != info->out_height) {
-			DSSERR("check_overlay: overlay %d doesn't support "
-					"scaling\n", ovl->id);
-			return -EINVAL;
-		}
-	}
-
-	if ((ovl->supported_modes & info->color_mode) == 0) {
-		DSSERR("check_overlay: overlay %d doesn't support mode %d\n",
-				ovl->id, info->color_mode);
-		return -EINVAL;
-	}
-
-	if (info->zorder >= omap_dss_get_num_overlays()) {
-		DSSERR("check_overlay: zorder %d too high\n", info->zorder);
-		return -EINVAL;
-	}
-
-	if (dss_feat_rotation_type_supported(info->rotation_type) == 0) {
-		DSSERR("check_overlay: rotation type %d not supported\n",
-				info->rotation_type);
-		return -EINVAL;
-	}
-
-	return 0;
-}
-
-int dss_ovl_check(struct omap_overlay *ovl, struct omap_overlay_info *info,
-		const struct omap_video_timings *mgr_timings)
-{
-	u16 outw, outh;
-	u16 dw, dh;
-
-	dw = mgr_timings->x_res;
-	dh = mgr_timings->y_res;
-
-	if ((ovl->caps & OMAP_DSS_OVL_CAP_SCALE) == 0) {
-		outw = info->width;
-		outh = info->height;
-	} else {
-		if (info->out_width == 0)
-			outw = info->width;
-		else
-			outw = info->out_width;
-
-		if (info->out_height == 0)
-			outh = info->height;
-		else
-			outh = info->out_height;
-	}
-
-	if (dw < info->pos_x + outw) {
-		DSSERR("overlay %d horizontally not inside the display area "
-				"(%d + %d >= %d)\n",
-				ovl->id, info->pos_x, outw, dw);
-		return -EINVAL;
-	}
-
-	if (dh < info->pos_y + outh) {
-		DSSERR("overlay %d vertically not inside the display area "
-				"(%d + %d >= %d)\n",
-				ovl->id, info->pos_y, outh, dh);
-		return -EINVAL;
-	}
-
-	return 0;
-}
-
-/*
- * Checks if replication logic should be used. Only use when overlay is in
- * RGB12U or RGB16 mode, and video port width interface is 18bpp or 24bpp
- */
-bool dss_ovl_use_replication(struct dss_lcd_mgr_config config,
-		enum omap_color_mode mode)
-{
-	if (mode != OMAP_DSS_COLOR_RGB12U && mode != OMAP_DSS_COLOR_RGB16)
-		return false;
-
-	return config.video_port_width > 16;
-}
diff --git a/drivers/gpu/drm/omapdrm/dss/rfbi.c b/drivers/gpu/drm/omapdrm/dss/rfbi.c
index aea6a1d0fb20..3796576dfadf 100644
--- a/drivers/gpu/drm/omapdrm/dss/rfbi.c
+++ b/drivers/gpu/drm/omapdrm/dss/rfbi.c
@@ -880,7 +880,7 @@ static int rfbi_display_enable(struct omap_dss_device *dssdev)
 	struct omap_dss_device *out = &rfbi.output;
 	int r;
 
-	if (out->manager == NULL) {
+	if (!out->dispc_channel_connected) {
 		DSSERR("failed to enable display: no output/manager\n");
 		return -ENODEV;
 	}
diff --git a/drivers/gpu/drm/omapdrm/dss/sdi.c b/drivers/gpu/drm/omapdrm/dss/sdi.c
index d747cc6b59e1..cd6d3bfb041d 100644
--- a/drivers/gpu/drm/omapdrm/dss/sdi.c
+++ b/drivers/gpu/drm/omapdrm/dss/sdi.c
@@ -114,7 +114,7 @@ static int sdi_calc_clock_div(unsigned long pclk,
 
 static void sdi_config_lcd_manager(struct omap_dss_device *dssdev)
 {
-	struct omap_overlay_manager *mgr = sdi.output.manager;
+	enum omap_channel channel = dssdev->dispc_channel;
 
 	sdi.mgr_config.io_pad_mode = DSS_IO_PAD_MODE_BYPASS;
 
@@ -124,19 +124,20 @@ static void sdi_config_lcd_manager(struct omap_dss_device *dssdev)
 	sdi.mgr_config.video_port_width = 24;
 	sdi.mgr_config.lcden_sig_polarity = 1;
 
-	dss_mgr_set_lcd_config(mgr, &sdi.mgr_config);
+	dss_mgr_set_lcd_config(channel, &sdi.mgr_config);
 }
 
 static int sdi_display_enable(struct omap_dss_device *dssdev)
 {
 	struct omap_dss_device *out = &sdi.output;
+	enum omap_channel channel = dssdev->dispc_channel;
 	struct omap_video_timings *t = &sdi.timings;
 	unsigned long fck;
 	struct dispc_clock_info dispc_cinfo;
 	unsigned long pck;
 	int r;
 
-	if (out->manager == NULL) {
+	if (!out->dispc_channel_connected) {
 		DSSERR("failed to enable display: no output/manager\n");
 		return -ENODEV;
 	}
@@ -169,7 +170,7 @@ static int sdi_display_enable(struct omap_dss_device *dssdev)
 	}
 
 
-	dss_mgr_set_timings(out->manager, t);
+	dss_mgr_set_timings(channel, t);
 
 	r = dss_set_fck_rate(fck);
 	if (r)
@@ -188,7 +189,7 @@ static int sdi_display_enable(struct omap_dss_device *dssdev)
 	 * need to care about the shadow register mechanism for pck-free. The
 	 * exact reason for this is unknown.
 	 */
-	dispc_mgr_set_clock_div(out->manager->id, &sdi.mgr_config.clock_info);
+	dispc_mgr_set_clock_div(channel, &sdi.mgr_config.clock_info);
 
 	dss_sdi_init(sdi.datapairs);
 	r = dss_sdi_enable();
@@ -196,7 +197,7 @@ static int sdi_display_enable(struct omap_dss_device *dssdev)
 		goto err_sdi_enable;
 	mdelay(2);
 
-	r = dss_mgr_enable(out->manager);
+	r = dss_mgr_enable(channel);
 	if (r)
 		goto err_mgr_enable;
 
@@ -216,9 +217,9 @@ err_reg_enable:
 
 static void sdi_display_disable(struct omap_dss_device *dssdev)
 {
-	struct omap_overlay_manager *mgr = sdi.output.manager;
+	enum omap_channel channel = dssdev->dispc_channel;
 
-	dss_mgr_disable(mgr);
+	dss_mgr_disable(channel);
 
 	dss_sdi_disable();
 
@@ -242,9 +243,9 @@ static void sdi_get_timings(struct omap_dss_device *dssdev,
 static int sdi_check_timings(struct omap_dss_device *dssdev,
 			struct omap_video_timings *timings)
 {
-	struct omap_overlay_manager *mgr = sdi.output.manager;
+	enum omap_channel channel = dssdev->dispc_channel;
 
-	if (mgr && !dispc_mgr_timings_ok(mgr->id, timings))
+	if (!dispc_mgr_timings_ok(channel, timings))
 		return -EINVAL;
 
 	if (timings->pixelclock == 0)
@@ -280,18 +281,14 @@ static int sdi_init_regulator(void)
 static int sdi_connect(struct omap_dss_device *dssdev,
 		struct omap_dss_device *dst)
 {
-	struct omap_overlay_manager *mgr;
+	enum omap_channel channel = dssdev->dispc_channel;
 	int r;
 
 	r = sdi_init_regulator();
 	if (r)
 		return r;
 
-	mgr = omap_dss_get_overlay_manager(dssdev->dispc_channel);
-	if (!mgr)
-		return -ENODEV;
-
-	r = dss_mgr_connect(mgr, dssdev);
+	r = dss_mgr_connect(channel, dssdev);
 	if (r)
 		return r;
 
@@ -299,7 +296,7 @@ static int sdi_connect(struct omap_dss_device *dssdev,
 	if (r) {
 		DSSERR("failed to connect output to new device: %s\n",
 				dst->name);
-		dss_mgr_disconnect(mgr, dssdev);
+		dss_mgr_disconnect(channel, dssdev);
 		return r;
 	}
 
@@ -309,6 +306,8 @@ static int sdi_connect(struct omap_dss_device *dssdev,
 static void sdi_disconnect(struct omap_dss_device *dssdev,
 		struct omap_dss_device *dst)
 {
+	enum omap_channel channel = dssdev->dispc_channel;
+
 	WARN_ON(dst != dssdev->dst);
 
 	if (dst != dssdev->dst)
@@ -316,8 +315,7 @@ static void sdi_disconnect(struct omap_dss_device *dssdev,
 
 	omapdss_output_unset_device(dssdev);
 
-	if (dssdev->manager)
-		dss_mgr_disconnect(dssdev->manager, dssdev);
+	dss_mgr_disconnect(channel, dssdev);
 }
 
 static const struct omapdss_sdi_ops sdi_ops = {
diff --git a/drivers/gpu/drm/omapdrm/dss/venc.c b/drivers/gpu/drm/omapdrm/dss/venc.c
index 08f9def76e27..08a2cc778ba9 100644
--- a/drivers/gpu/drm/omapdrm/dss/venc.c
+++ b/drivers/gpu/drm/omapdrm/dss/venc.c
@@ -443,7 +443,7 @@ static const struct venc_config *venc_timings_to_config(
 
 static int venc_power_on(struct omap_dss_device *dssdev)
 {
-	struct omap_overlay_manager *mgr = venc.output.manager;
+	enum omap_channel channel = dssdev->dispc_channel;
 	u32 l;
 	int r;
 
@@ -469,13 +469,13 @@ static int venc_power_on(struct omap_dss_device *dssdev)
 
 	venc_write_reg(VENC_OUTPUT_CONTROL, l);
 
-	dss_mgr_set_timings(mgr, &venc.timings);
+	dss_mgr_set_timings(channel, &venc.timings);
 
 	r = regulator_enable(venc.vdda_dac_reg);
 	if (r)
 		goto err1;
 
-	r = dss_mgr_enable(mgr);
+	r = dss_mgr_enable(channel);
 	if (r)
 		goto err2;
 
@@ -494,12 +494,12 @@ err0:
 
 static void venc_power_off(struct omap_dss_device *dssdev)
 {
-	struct omap_overlay_manager *mgr = venc.output.manager;
+	enum omap_channel channel = dssdev->dispc_channel;
 
 	venc_write_reg(VENC_OUTPUT_CONTROL, 0);
 	dss_set_dac_pwrdn_bgz(0);
 
-	dss_mgr_disable(mgr);
+	dss_mgr_disable(channel);
 
 	regulator_disable(venc.vdda_dac_reg);
 
@@ -515,7 +515,7 @@ static int venc_display_enable(struct omap_dss_device *dssdev)
 
 	mutex_lock(&venc.venc_lock);
 
-	if (out->manager == NULL) {
+	if (!out->dispc_channel_connected) {
 		DSSERR("Failed to enable display: no output/manager\n");
 		r = -ENODEV;
 		goto err0;
@@ -742,18 +742,14 @@ static int venc_get_clocks(struct platform_device *pdev)
 static int venc_connect(struct omap_dss_device *dssdev,
 		struct omap_dss_device *dst)
 {
-	struct omap_overlay_manager *mgr;
+	enum omap_channel channel = dssdev->dispc_channel;
 	int r;
 
 	r = venc_init_regulator();
 	if (r)
 		return r;
 
-	mgr = omap_dss_get_overlay_manager(dssdev->dispc_channel);
-	if (!mgr)
-		return -ENODEV;
-
-	r = dss_mgr_connect(mgr, dssdev);
+	r = dss_mgr_connect(channel, dssdev);
 	if (r)
 		return r;
 
@@ -761,7 +757,7 @@ static int venc_connect(struct omap_dss_device *dssdev,
 	if (r) {
 		DSSERR("failed to connect output to new device: %s\n",
 				dst->name);
-		dss_mgr_disconnect(mgr, dssdev);
+		dss_mgr_disconnect(channel, dssdev);
 		return r;
 	}
 
@@ -771,6 +767,8 @@ static int venc_connect(struct omap_dss_device *dssdev,
 static void venc_disconnect(struct omap_dss_device *dssdev,
 		struct omap_dss_device *dst)
 {
+	enum omap_channel channel = dssdev->dispc_channel;
+
 	WARN_ON(dst != dssdev->dst);
 
 	if (dst != dssdev->dst)
@@ -778,8 +776,7 @@ static void venc_disconnect(struct omap_dss_device *dssdev,
 
 	omapdss_output_unset_device(dssdev);
 
-	if (dssdev->manager)
-		dss_mgr_disconnect(dssdev->manager, dssdev);
+	dss_mgr_disconnect(channel, dssdev);
 }
 
 static const struct omapdss_atv_ops venc_ops = {
diff --git a/drivers/gpu/drm/omapdrm/omap_connector.c b/drivers/gpu/drm/omapdrm/omap_connector.c
index 83f2a9177c14..ce2d67b6a8c7 100644
--- a/drivers/gpu/drm/omapdrm/omap_connector.c
+++ b/drivers/gpu/drm/omapdrm/omap_connector.c
@@ -63,6 +63,9 @@ void copy_timings_omap_to_drm(struct drm_display_mode *mode,
 	if (timings->interlace)
 		mode->flags |= DRM_MODE_FLAG_INTERLACE;
 
+	if (timings->double_pixel)
+		mode->flags |= DRM_MODE_FLAG_DBLCLK;
+
 	if (timings->hsync_level == OMAPDSS_SIG_ACTIVE_HIGH)
 		mode->flags |= DRM_MODE_FLAG_PHSYNC;
 	else
@@ -90,6 +93,7 @@ void copy_timings_drm_to_omap(struct omap_video_timings *timings,
 	timings->vbp = mode->vtotal - mode->vsync_end;
 
 	timings->interlace = !!(mode->flags & DRM_MODE_FLAG_INTERLACE);
+	timings->double_pixel = !!(mode->flags & DRM_MODE_FLAG_DBLCLK);
 
 	if (mode->flags & DRM_MODE_FLAG_PHSYNC)
 		timings->hsync_level = OMAPDSS_SIG_ACTIVE_HIGH;
diff --git a/drivers/gpu/drm/omapdrm/omap_crtc.c b/drivers/gpu/drm/omapdrm/omap_crtc.c
index 483acdbd664b..075f2bb44867 100644
--- a/drivers/gpu/drm/omapdrm/omap_crtc.c
+++ b/drivers/gpu/drm/omapdrm/omap_crtc.c
@@ -34,14 +34,6 @@ struct omap_crtc {
 	const char *name;
 	enum omap_channel channel;
 
-	/*
-	 * Temporary: eventually this will go away, but it is needed
-	 * for now to keep the output's happy.  (They only need
-	 * mgr->id.)  Eventually this will be replaced w/ something
-	 * more common-panel-framework-y
-	 */
-	struct omap_overlay_manager *mgr;
-
 	struct omap_video_timings timings;
 
 	struct omap_drm_irq vblank_irq;
@@ -80,9 +72,13 @@ int omap_crtc_wait_pending(struct drm_crtc *crtc)
 {
 	struct omap_crtc *omap_crtc = to_omap_crtc(crtc);
 
+	/*
+	 * Timeout is set to a "sufficiently" high value, which should cover
+	 * a single frame refresh even on slower displays.
+	 */
 	return wait_event_timeout(omap_crtc->pending_wait,
 				  !omap_crtc->pending,
-				  msecs_to_jiffies(50));
+				  msecs_to_jiffies(250));
 }
 
 /* -----------------------------------------------------------------------------
@@ -100,31 +96,32 @@ int omap_crtc_wait_pending(struct drm_crtc *crtc)
 
 /* ovl-mgr-id -> crtc */
 static struct omap_crtc *omap_crtcs[8];
+static struct omap_dss_device *omap_crtc_output[8];
 
 /* we can probably ignore these until we support command-mode panels: */
-static int omap_crtc_dss_connect(struct omap_overlay_manager *mgr,
+static int omap_crtc_dss_connect(enum omap_channel channel,
 		struct omap_dss_device *dst)
 {
-	if (mgr->output)
+	if (omap_crtc_output[channel])
 		return -EINVAL;
 
-	if ((mgr->supported_outputs & dst->id) == 0)
+	if ((dispc_mgr_get_supported_outputs(channel) & dst->id) == 0)
 		return -EINVAL;
 
-	dst->manager = mgr;
-	mgr->output = dst;
+	omap_crtc_output[channel] = dst;
+	dst->dispc_channel_connected = true;
 
 	return 0;
 }
 
-static void omap_crtc_dss_disconnect(struct omap_overlay_manager *mgr,
+static void omap_crtc_dss_disconnect(enum omap_channel channel,
 		struct omap_dss_device *dst)
 {
-	mgr->output->manager = NULL;
-	mgr->output = NULL;
+	omap_crtc_output[channel] = NULL;
+	dst->dispc_channel_connected = false;
 }
 
-static void omap_crtc_dss_start_update(struct omap_overlay_manager *mgr)
+static void omap_crtc_dss_start_update(enum omap_channel channel)
 {
 }
 
@@ -138,6 +135,11 @@ static void omap_crtc_set_enabled(struct drm_crtc *crtc, bool enable)
 	u32 framedone_irq, vsync_irq;
 	int ret;
 
+	if (omap_crtc_output[channel]->output_type == OMAP_DISPLAY_TYPE_HDMI) {
+		dispc_mgr_enable(channel, enable);
+		return;
+	}
+
 	if (dispc_mgr_is_enabled(channel) == enable)
 		return;
 
@@ -186,9 +188,9 @@ static void omap_crtc_set_enabled(struct drm_crtc *crtc, bool enable)
 }
 
 
-static int omap_crtc_dss_enable(struct omap_overlay_manager *mgr)
+static int omap_crtc_dss_enable(enum omap_channel channel)
 {
-	struct omap_crtc *omap_crtc = omap_crtcs[mgr->id];
+	struct omap_crtc *omap_crtc = omap_crtcs[channel];
 	struct omap_overlay_manager_info info;
 
 	memset(&info, 0, sizeof(info));
@@ -205,38 +207,38 @@ static int omap_crtc_dss_enable(struct omap_overlay_manager *mgr)
 	return 0;
 }
 
-static void omap_crtc_dss_disable(struct omap_overlay_manager *mgr)
+static void omap_crtc_dss_disable(enum omap_channel channel)
 {
-	struct omap_crtc *omap_crtc = omap_crtcs[mgr->id];
+	struct omap_crtc *omap_crtc = omap_crtcs[channel];
 
 	omap_crtc_set_enabled(&omap_crtc->base, false);
 }
 
-static void omap_crtc_dss_set_timings(struct omap_overlay_manager *mgr,
+static void omap_crtc_dss_set_timings(enum omap_channel channel,
 		const struct omap_video_timings *timings)
 {
-	struct omap_crtc *omap_crtc = omap_crtcs[mgr->id];
+	struct omap_crtc *omap_crtc = omap_crtcs[channel];
 	DBG("%s", omap_crtc->name);
 	omap_crtc->timings = *timings;
 }
 
-static void omap_crtc_dss_set_lcd_config(struct omap_overlay_manager *mgr,
+static void omap_crtc_dss_set_lcd_config(enum omap_channel channel,
 		const struct dss_lcd_mgr_config *config)
 {
-	struct omap_crtc *omap_crtc = omap_crtcs[mgr->id];
+	struct omap_crtc *omap_crtc = omap_crtcs[channel];
 	DBG("%s", omap_crtc->name);
 	dispc_mgr_set_lcd_config(omap_crtc->channel, config);
 }
 
 static int omap_crtc_dss_register_framedone(
-		struct omap_overlay_manager *mgr,
+		enum omap_channel channel,
 		void (*handler)(void *), void *data)
 {
 	return 0;
 }
 
 static void omap_crtc_dss_unregister_framedone(
-		struct omap_overlay_manager *mgr,
+		enum omap_channel channel,
 		void (*handler)(void *), void *data)
 {
 }
@@ -396,24 +398,40 @@ static void omap_crtc_atomic_flush(struct drm_crtc *crtc,
 	}
 }
 
+static bool omap_crtc_is_plane_prop(struct drm_device *dev,
+	struct drm_property *property)
+{
+	struct omap_drm_private *priv = dev->dev_private;
+
+	return property == priv->zorder_prop ||
+		property == dev->mode_config.rotation_property;
+}
+
 static int omap_crtc_atomic_set_property(struct drm_crtc *crtc,
 					 struct drm_crtc_state *state,
 					 struct drm_property *property,
 					 uint64_t val)
 {
-	struct drm_plane_state *plane_state;
-	struct drm_plane *plane = crtc->primary;
+	struct drm_device *dev = crtc->dev;
 
-	/*
-	 * Delegate property set to the primary plane. Get the plane state and
-	 * set the property directly.
-	 */
+	if (omap_crtc_is_plane_prop(dev, property)) {
+		struct drm_plane_state *plane_state;
+		struct drm_plane *plane = crtc->primary;
 
-	plane_state = drm_atomic_get_plane_state(state->state, plane);
-	if (!plane_state)
-		return -EINVAL;
+		/*
+		 * Delegate property set to the primary plane. Get the plane
+		 * state and set the property directly.
+		 */
+
+		plane_state = drm_atomic_get_plane_state(state->state, plane);
+		if (IS_ERR(plane_state))
+			return PTR_ERR(plane_state);
 
-	return drm_atomic_plane_set_property(plane, plane_state, property, val);
+		return drm_atomic_plane_set_property(plane, plane_state,
+				property, val);
+	}
+
+	return -EINVAL;
 }
 
 static int omap_crtc_atomic_get_property(struct drm_crtc *crtc,
@@ -421,14 +439,20 @@ static int omap_crtc_atomic_get_property(struct drm_crtc *crtc,
 					 struct drm_property *property,
 					 uint64_t *val)
 {
-	/*
-	 * Delegate property get to the primary plane. The
-	 * drm_atomic_plane_get_property() function isn't exported, but can be
-	 * called through drm_object_property_get_value() as that will call
-	 * drm_atomic_get_property() for atomic drivers.
-	 */
-	return drm_object_property_get_value(&crtc->primary->base, property,
-					     val);
+	struct drm_device *dev = crtc->dev;
+
+	if (omap_crtc_is_plane_prop(dev, property)) {
+		/*
+		 * Delegate property get to the primary plane. The
+		 * drm_atomic_plane_get_property() function isn't exported, but
+		 * can be called through drm_object_property_get_value() as that
+		 * will call drm_atomic_get_property() for atomic drivers.
+		 */
+		return drm_object_property_get_value(&crtc->primary->base,
+				property, val);
+	}
+
+	return -EINVAL;
 }
 
 static const struct drm_crtc_funcs omap_crtc_funcs = {
@@ -501,9 +525,6 @@ struct drm_crtc *omap_crtc_init(struct drm_device *dev,
 	omap_crtc->error_irq.irq = omap_crtc_error_irq;
 	omap_irq_register(dev, &omap_crtc->error_irq);
 
-	/* temporary: */
-	omap_crtc->mgr = omap_dss_get_overlay_manager(channel);
-
 	ret = drm_crtc_init_with_planes(dev, crtc, plane, NULL,
 					&omap_crtc_funcs, NULL);
 	if (ret < 0) {
diff --git a/drivers/gpu/drm/omapdrm/omap_dmm_tiler.c b/drivers/gpu/drm/omapdrm/omap_dmm_tiler.c
index dfebdc4aa0f2..9f94576c435d 100644
--- a/drivers/gpu/drm/omapdrm/omap_dmm_tiler.c
+++ b/drivers/gpu/drm/omapdrm/omap_dmm_tiler.c
@@ -79,6 +79,16 @@ static const uint32_t reg[][4] = {
 			DMM_PAT_DESCR__2, DMM_PAT_DESCR__3},
 };
 
+static u32 dmm_read(struct dmm *dmm, u32 reg)
+{
+	return readl(dmm->base + reg);
+}
+
+static void dmm_write(struct dmm *dmm, u32 val, u32 reg)
+{
+	writel(val, dmm->base + reg);
+}
+
 /* simple allocator to grab next 16 byte aligned memory from txn */
 static void *alloc_dma(struct dmm_txn *txn, size_t sz, dma_addr_t *pa)
 {
@@ -108,7 +118,7 @@ static int wait_status(struct refill_engine *engine, uint32_t wait_mask)
 
 	i = DMM_FIXED_RETRY_COUNT;
 	while (true) {
-		r = readl(dmm->base + reg[PAT_STATUS][engine->id]);
+		r = dmm_read(dmm, reg[PAT_STATUS][engine->id]);
 		err = r & DMM_PATSTATUS_ERR;
 		if (err)
 			return -EFAULT;
@@ -140,11 +150,11 @@ static void release_engine(struct refill_engine *engine)
 static irqreturn_t omap_dmm_irq_handler(int irq, void *arg)
 {
 	struct dmm *dmm = arg;
-	uint32_t status = readl(dmm->base + DMM_PAT_IRQSTATUS);
+	uint32_t status = dmm_read(dmm, DMM_PAT_IRQSTATUS);
 	int i;
 
 	/* ack IRQ */
-	writel(status, dmm->base + DMM_PAT_IRQSTATUS);
+	dmm_write(dmm, status, DMM_PAT_IRQSTATUS);
 
 	for (i = 0; i < dmm->num_engines; i++) {
 		if (status & DMM_IRQSTAT_LST) {
@@ -264,7 +274,7 @@ static int dmm_txn_commit(struct dmm_txn *txn, bool wait)
 	txn->last_pat->next_pa = 0;
 
 	/* write to PAT_DESCR to clear out any pending transaction */
-	writel(0x0, dmm->base + reg[PAT_DESCR][engine->id]);
+	dmm_write(dmm, 0x0, reg[PAT_DESCR][engine->id]);
 
 	/* wait for engine ready: */
 	ret = wait_status(engine, DMM_PATSTATUS_READY);
@@ -280,8 +290,7 @@ static int dmm_txn_commit(struct dmm_txn *txn, bool wait)
 	smp_mb();
 
 	/* kick reload */
-	writel(engine->refill_pa,
-		dmm->base + reg[PAT_DESCR][engine->id]);
+	dmm_write(dmm, engine->refill_pa, reg[PAT_DESCR][engine->id]);
 
 	if (wait) {
 		if (!wait_for_completion_timeout(&engine->compl,
@@ -309,6 +318,21 @@ static int fill(struct tcm_area *area, struct page **pages,
 	struct tcm_area slice, area_s;
 	struct dmm_txn *txn;
 
+	/*
+	 * FIXME
+	 *
+	 * Asynchronous fill does not work reliably, as the driver does not
+	 * handle errors in the async code paths. The fill operation may
+	 * silently fail, leading to leaking DMM engines, which may eventually
+	 * lead to deadlock if we run out of DMM engines.
+	 *
+	 * For now, always set 'wait' so that we only use sync fills. Async
+	 * fills should be fixed, or alternatively we could decide to only
+	 * support sync fills and so the whole async code path could be removed.
+	 */
+
+	wait = true;
+
 	txn = dmm_txn_init(omap_dmm, area->tcm);
 	if (IS_ERR_OR_NULL(txn))
 		return -ENOMEM;
@@ -642,7 +666,7 @@ static int omap_dmm_probe(struct platform_device *dev)
 
 	omap_dmm->dev = &dev->dev;
 
-	hwinfo = readl(omap_dmm->base + DMM_PAT_HWINFO);
+	hwinfo = dmm_read(omap_dmm, DMM_PAT_HWINFO);
 	omap_dmm->num_engines = (hwinfo >> 24) & 0x1F;
 	omap_dmm->num_lut = (hwinfo >> 16) & 0x1F;
 	omap_dmm->container_width = 256;
@@ -651,7 +675,7 @@ static int omap_dmm_probe(struct platform_device *dev)
 	atomic_set(&omap_dmm->engine_counter, omap_dmm->num_engines);
 
 	/* read out actual LUT width and height */
-	pat_geom = readl(omap_dmm->base + DMM_PAT_GEOMETRY);
+	pat_geom = dmm_read(omap_dmm, DMM_PAT_GEOMETRY);
 	omap_dmm->lut_width = ((pat_geom >> 16) & 0xF) << 5;
 	omap_dmm->lut_height = ((pat_geom >> 24) & 0xF) << 5;
 
@@ -661,12 +685,12 @@ static int omap_dmm_probe(struct platform_device *dev)
 		omap_dmm->num_lut++;
 
 	/* initialize DMM registers */
-	writel(0x88888888, omap_dmm->base + DMM_PAT_VIEW__0);
-	writel(0x88888888, omap_dmm->base + DMM_PAT_VIEW__1);
-	writel(0x80808080, omap_dmm->base + DMM_PAT_VIEW_MAP__0);
-	writel(0x80000000, omap_dmm->base + DMM_PAT_VIEW_MAP_BASE);
-	writel(0x88888888, omap_dmm->base + DMM_TILER_OR__0);
-	writel(0x88888888, omap_dmm->base + DMM_TILER_OR__1);
+	dmm_write(omap_dmm, 0x88888888, DMM_PAT_VIEW__0);
+	dmm_write(omap_dmm, 0x88888888, DMM_PAT_VIEW__1);
+	dmm_write(omap_dmm, 0x80808080, DMM_PAT_VIEW_MAP__0);
+	dmm_write(omap_dmm, 0x80000000, DMM_PAT_VIEW_MAP_BASE);
+	dmm_write(omap_dmm, 0x88888888, DMM_TILER_OR__0);
+	dmm_write(omap_dmm, 0x88888888, DMM_TILER_OR__1);
 
 	ret = request_irq(omap_dmm->irq, omap_dmm_irq_handler, IRQF_SHARED,
 				"omap_dmm_irq_handler", omap_dmm);
@@ -684,7 +708,7 @@ static int omap_dmm_probe(struct platform_device *dev)
 	 * buffers for accelerated pan/scroll) and FILL_DSC<n> which
 	 * we just generally don't care about.
 	 */
-	writel(0x7e7e7e7e, omap_dmm->base + DMM_PAT_IRQENABLE_SET);
+	dmm_write(omap_dmm, 0x7e7e7e7e, DMM_PAT_IRQENABLE_SET);
 
 	omap_dmm->dummy_page = alloc_page(GFP_KERNEL | __GFP_DMA32);
 	if (!omap_dmm->dummy_page) {
diff --git a/drivers/gpu/drm/omapdrm/omap_drv.c b/drivers/gpu/drm/omapdrm/omap_drv.c
index 33370f42e4d7..80398a684cae 100644
--- a/drivers/gpu/drm/omapdrm/omap_drv.c
+++ b/drivers/gpu/drm/omapdrm/omap_drv.c
@@ -340,7 +340,7 @@ static int omap_modeset_init(struct drm_device *dev)
 		struct drm_connector *connector;
 		struct drm_encoder *encoder;
 		enum omap_channel channel;
-		struct omap_overlay_manager *mgr;
+		struct omap_dss_device *out;
 
 		if (!omapdss_device_is_connected(dssdev))
 			continue;
@@ -387,8 +387,10 @@ static int omap_modeset_init(struct drm_device *dev)
 		 * not considered.
 		 */
 
-		mgr = omapdss_find_mgr_from_display(dssdev);
-		channel = mgr->id;
+		out = omapdss_find_output_from_display(dssdev);
+		channel = out->dispc_channel;
+		omap_dss_put_device(out);
+
 		/*
 		 * if this channel hasn't already been taken by a previously
 		 * allocated crtc, we create a new crtc for it
@@ -858,12 +860,52 @@ static int pdev_remove(struct platform_device *device)
 }
 
 #ifdef CONFIG_PM_SLEEP
+static int omap_drm_suspend_all_displays(void)
+{
+	struct omap_dss_device *dssdev = NULL;
+
+	for_each_dss_dev(dssdev) {
+		if (!dssdev->driver)
+			continue;
+
+		if (dssdev->state == OMAP_DSS_DISPLAY_ACTIVE) {
+			dssdev->driver->disable(dssdev);
+			dssdev->activate_after_resume = true;
+		} else {
+			dssdev->activate_after_resume = false;
+		}
+	}
+
+	return 0;
+}
+
+static int omap_drm_resume_all_displays(void)
+{
+	struct omap_dss_device *dssdev = NULL;
+
+	for_each_dss_dev(dssdev) {
+		if (!dssdev->driver)
+			continue;
+
+		if (dssdev->activate_after_resume) {
+			dssdev->driver->enable(dssdev);
+			dssdev->activate_after_resume = false;
+		}
+	}
+
+	return 0;
+}
+
 static int omap_drm_suspend(struct device *dev)
 {
 	struct drm_device *drm_dev = dev_get_drvdata(dev);
 
 	drm_kms_helper_poll_disable(drm_dev);
 
+	drm_modeset_lock_all(drm_dev);
+	omap_drm_suspend_all_displays();
+	drm_modeset_unlock_all(drm_dev);
+
 	return 0;
 }
 
@@ -871,6 +913,10 @@ static int omap_drm_resume(struct device *dev)
 {
 	struct drm_device *drm_dev = dev_get_drvdata(dev);
 
+	drm_modeset_lock_all(drm_dev);
+	omap_drm_resume_all_displays();
+	drm_modeset_unlock_all(drm_dev);
+
 	drm_kms_helper_poll_enable(drm_dev);
 
 	return omap_gem_resume(dev);
diff --git a/drivers/gpu/drm/omapdrm/omap_drv.h b/drivers/gpu/drm/omapdrm/omap_drv.h
index c23cbe6fe9e4..0fbe17d0ec6f 100644
--- a/drivers/gpu/drm/omapdrm/omap_drv.h
+++ b/drivers/gpu/drm/omapdrm/omap_drv.h
@@ -31,6 +31,8 @@
 #include <drm/drm_gem.h>
 #include <drm/omap_drm.h>
 
+#include "dss/omapdss.h"
+
 #define DBG(fmt, ...) DRM_DEBUG(fmt"\n", ##__VA_ARGS__)
 #define VERB(fmt, ...) if (0) DRM_DEBUG(fmt, ##__VA_ARGS__) /* verbose debug */
 
@@ -188,12 +190,15 @@ void omap_framebuffer_update_scanout(struct drm_framebuffer *fb,
 		struct omap_drm_window *win, struct omap_overlay_info *info);
 struct drm_connector *omap_framebuffer_get_next_connector(
 		struct drm_framebuffer *fb, struct drm_connector *from);
+bool omap_framebuffer_supports_rotation(struct drm_framebuffer *fb);
 
 void omap_gem_init(struct drm_device *dev);
 void omap_gem_deinit(struct drm_device *dev);
 
 struct drm_gem_object *omap_gem_new(struct drm_device *dev,
 		union omap_gem_size gsize, uint32_t flags);
+struct drm_gem_object *omap_gem_new_dmabuf(struct drm_device *dev, size_t size,
+		struct sg_table *sgt);
 int omap_gem_new_handle(struct drm_device *dev, struct drm_file *file,
 		union omap_gem_size gsize, uint32_t flags, uint32_t *handle);
 void omap_gem_free_object(struct drm_gem_object *obj);
diff --git a/drivers/gpu/drm/omapdrm/omap_encoder.c b/drivers/gpu/drm/omapdrm/omap_encoder.c
index 61714e9670ae..0bbb9c59622e 100644
--- a/drivers/gpu/drm/omapdrm/omap_encoder.c
+++ b/drivers/gpu/drm/omapdrm/omap_encoder.c
@@ -139,11 +139,16 @@ static void omap_encoder_enable(struct drm_encoder *encoder)
 	struct omap_encoder *omap_encoder = to_omap_encoder(encoder);
 	struct omap_dss_device *dssdev = omap_encoder->dssdev;
 	struct omap_dss_driver *dssdrv = dssdev->driver;
+	int r;
 
 	omap_encoder_update(encoder, omap_crtc_channel(encoder->crtc),
 			    omap_crtc_timings(encoder->crtc));
 
-	dssdrv->enable(dssdev);
+	r = dssdrv->enable(dssdev);
+	if (r)
+		dev_err(encoder->dev->dev,
+			"Failed to enable display '%s': %d\n",
+			dssdev->name, r);
 }
 
 static int omap_encoder_atomic_check(struct drm_encoder *encoder,
diff --git a/drivers/gpu/drm/omapdrm/omap_fb.c b/drivers/gpu/drm/omapdrm/omap_fb.c
index ad202dfc1a49..610962396eb0 100644
--- a/drivers/gpu/drm/omapdrm/omap_fb.c
+++ b/drivers/gpu/drm/omapdrm/omap_fb.c
@@ -145,6 +145,14 @@ static uint32_t get_linear_addr(struct plane *plane,
 	return plane->paddr + offset;
 }
 
+bool omap_framebuffer_supports_rotation(struct drm_framebuffer *fb)
+{
+	struct omap_framebuffer *omap_fb = to_omap_framebuffer(fb);
+	struct plane *plane = &omap_fb->planes[0];
+
+	return omap_gem_flags(plane->bo) & OMAP_BO_TILED;
+}
+
 /* update ovl info for scanout, handles cases of multi-planar fb's, etc.
  */
 void omap_framebuffer_update_scanout(struct drm_framebuffer *fb,
@@ -449,6 +457,14 @@ struct drm_framebuffer *omap_framebuffer_init(struct drm_device *dev,
 			goto fail;
 		}
 
+		if (i > 0 && pitch != mode_cmd->pitches[i - 1]) {
+			dev_err(dev->dev,
+				"pitches are not the same between framebuffer planes %d != %d\n",
+				pitch, mode_cmd->pitches[i - 1]);
+			ret = -EINVAL;
+			goto fail;
+		}
+
 		plane->bo     = bos[i];
 		plane->offset = mode_cmd->offsets[i];
 		plane->pitch  = pitch;
diff --git a/drivers/gpu/drm/omapdrm/omap_gem.c b/drivers/gpu/drm/omapdrm/omap_gem.c
index 8495a1a4b617..cc36a8dc9bd4 100644
--- a/drivers/gpu/drm/omapdrm/omap_gem.c
+++ b/drivers/gpu/drm/omapdrm/omap_gem.c
@@ -31,9 +31,9 @@
  */
 
 /* note: we use upper 8 bits of flags for driver-internal flags: */
-#define OMAP_BO_DMA		0x01000000	/* actually is physically contiguous */
-#define OMAP_BO_EXT_SYNC	0x02000000	/* externally allocated sync object */
-#define OMAP_BO_EXT_MEM		0x04000000	/* externally allocated memory */
+#define OMAP_BO_MEM_DMA_API	0x01000000	/* memory allocated with the dma_alloc_* API */
+#define OMAP_BO_MEM_SHMEM	0x02000000	/* memory allocated through shmem backing */
+#define OMAP_BO_MEM_DMABUF	0x08000000	/* memory imported from a dmabuf */
 
 struct omap_gem_object {
 	struct drm_gem_object base;
@@ -49,17 +49,25 @@ struct omap_gem_object {
 	uint32_t roll;
 
 	/**
-	 * If buffer is allocated physically contiguous, the OMAP_BO_DMA flag
-	 * is set and the paddr is valid.  Also if the buffer is remapped in
-	 * TILER and paddr_cnt > 0, then paddr is valid.  But if you are using
-	 * the physical address and OMAP_BO_DMA is not set, then you should
-	 * be going thru omap_gem_{get,put}_paddr() to ensure the mapping is
-	 * not removed from under your feet.
+	 * paddr contains the buffer DMA address. It is valid for
 	 *
-	 * Note that OMAP_BO_SCANOUT is a hint from userspace that DMA capable
-	 * buffer is requested, but doesn't mean that it is.  Use the
-	 * OMAP_BO_DMA flag to determine if the buffer has a DMA capable
-	 * physical address.
+	 * - buffers allocated through the DMA mapping API (with the
+	 *   OMAP_BO_MEM_DMA_API flag set)
+	 *
+	 * - buffers imported from dmabuf (with the OMAP_BO_MEM_DMABUF flag set)
+	 *   if they are physically contiguous (when sgt->orig_nents == 1)
+	 *
+	 * - buffers mapped through the TILER when paddr_cnt is not zero, in
+	 *   which case the DMA address points to the TILER aperture
+	 *
+	 * Physically contiguous buffers have their DMA address equal to the
+	 * physical address as we don't remap those buffers through the TILER.
+	 *
+	 * Buffers mapped to the TILER have their DMA address pointing to the
+	 * TILER aperture. As TILER mappings are refcounted (through paddr_cnt)
+	 * the DMA address must be accessed through omap_get_get_paddr() to
+	 * ensure that the mapping won't disappear unexpectedly. References must
+	 * be released with omap_gem_put_paddr().
 	 */
 	dma_addr_t paddr;
 
@@ -69,6 +77,12 @@ struct omap_gem_object {
 	uint32_t paddr_cnt;
 
 	/**
+	 * If the buffer has been imported from a dmabuf the OMAP_DB_DMABUF flag
+	 * is set and the sgt field is valid.
+	 */
+	struct sg_table *sgt;
+
+	/**
 	 * tiler block used when buffer is remapped in DMM/TILER.
 	 */
 	struct tiler_block *block;
@@ -91,17 +105,7 @@ struct omap_gem_object {
 	 * sync-object allocated on demand (if needed)
 	 *
 	 * Per-buffer sync-object for tracking pending and completed hw/dma
-	 * read and write operations.  The layout in memory is dictated by
-	 * the SGX firmware, which uses this information to stall the command
-	 * stream if a surface is not ready yet.
-	 *
-	 * Note that when buffer is used by SGX, the sync-object needs to be
-	 * allocated from a special heap of sync-objects.  This way many sync
-	 * objects can be packed in a page, and not waste GPU virtual address
-	 * space.  Because of this we have to have a omap_gem_set_sync_object()
-	 * API to allow replacement of the syncobj after it has (potentially)
-	 * already been allocated.  A bit ugly but I haven't thought of a
-	 * better alternative.
+	 * read and write operations.
 	 */
 	struct {
 		uint32_t write_pending;
@@ -166,16 +170,15 @@ static uint64_t mmap_offset(struct drm_gem_object *obj)
 	return drm_vma_node_offset_addr(&obj->vma_node);
 }
 
-/* GEM objects can either be allocated from contiguous memory (in which
- * case obj->filp==NULL), or w/ shmem backing (obj->filp!=NULL).  But non
- * contiguous buffers can be remapped in TILER/DMM if they need to be
- * contiguous... but we don't do this all the time to reduce pressure
- * on TILER/DMM space when we know at allocation time that the buffer
- * will need to be scanned out.
- */
-static inline bool is_shmem(struct drm_gem_object *obj)
+static bool is_contiguous(struct omap_gem_object *omap_obj)
 {
-	return obj->filp != NULL;
+	if (omap_obj->flags & OMAP_BO_MEM_DMA_API)
+		return true;
+
+	if ((omap_obj->flags & OMAP_BO_MEM_DMABUF) && omap_obj->sgt->nents == 1)
+		return true;
+
+	return false;
 }
 
 /* -----------------------------------------------------------------------------
@@ -264,6 +267,19 @@ static int omap_gem_attach_pages(struct drm_gem_object *obj)
 		for (i = 0; i < npages; i++) {
 			addrs[i] = dma_map_page(dev->dev, pages[i],
 					0, PAGE_SIZE, DMA_BIDIRECTIONAL);
+
+			if (dma_mapping_error(dev->dev, addrs[i])) {
+				dev_warn(dev->dev,
+					"%s: failed to map page\n", __func__);
+
+				for (i = i - 1; i >= 0; --i) {
+					dma_unmap_page(dev->dev, addrs[i],
+						PAGE_SIZE, DMA_BIDIRECTIONAL);
+				}
+
+				ret = -ENOMEM;
+				goto free_addrs;
+			}
 		}
 	} else {
 		addrs = kzalloc(npages * sizeof(*addrs), GFP_KERNEL);
@@ -278,6 +294,8 @@ static int omap_gem_attach_pages(struct drm_gem_object *obj)
 
 	return 0;
 
+free_addrs:
+	kfree(addrs);
 free_pages:
 	drm_gem_put_pages(obj, pages, true, false);
 
@@ -292,7 +310,7 @@ static int get_pages(struct drm_gem_object *obj, struct page ***pages)
 	struct omap_gem_object *omap_obj = to_omap_bo(obj);
 	int ret = 0;
 
-	if (is_shmem(obj) && !omap_obj->pages) {
+	if ((omap_obj->flags & OMAP_BO_MEM_SHMEM) && !omap_obj->pages) {
 		ret = omap_gem_attach_pages(obj);
 		if (ret) {
 			dev_err(obj->dev->dev, "could not attach pages\n");
@@ -396,7 +414,7 @@ static int fault_1d(struct drm_gem_object *obj,
 		omap_gem_cpu_sync(obj, pgoff);
 		pfn = page_to_pfn(omap_obj->pages[pgoff]);
 	} else {
-		BUG_ON(!(omap_obj->flags & OMAP_BO_DMA));
+		BUG_ON(!is_contiguous(omap_obj));
 		pfn = (omap_obj->paddr >> PAGE_SHIFT) + pgoff;
 	}
 
@@ -560,6 +578,11 @@ fail:
 	case 0:
 	case -ERESTARTSYS:
 	case -EINTR:
+	case -EBUSY:
+		/*
+		 * EBUSY is ok: this just means that another thread
+		 * already did the job.
+		 */
 		return VM_FAULT_NOPAGE;
 	case -ENOMEM:
 		return VM_FAULT_OOM;
@@ -728,7 +751,8 @@ fail:
 static inline bool is_cached_coherent(struct drm_gem_object *obj)
 {
 	struct omap_gem_object *omap_obj = to_omap_bo(obj);
-	return is_shmem(obj) &&
+
+	return (omap_obj->flags & OMAP_BO_MEM_SHMEM) &&
 		((omap_obj->flags & OMAP_BO_CACHE_MASK) == OMAP_BO_CACHED);
 }
 
@@ -761,9 +785,20 @@ void omap_gem_dma_sync(struct drm_gem_object *obj,
 
 		for (i = 0; i < npages; i++) {
 			if (!omap_obj->addrs[i]) {
-				omap_obj->addrs[i] = dma_map_page(dev->dev, pages[i], 0,
+				dma_addr_t addr;
+
+				addr = dma_map_page(dev->dev, pages[i], 0,
 						PAGE_SIZE, DMA_BIDIRECTIONAL);
+
+				if (dma_mapping_error(dev->dev, addr)) {
+					dev_warn(dev->dev,
+						"%s: failed to map page\n",
+						__func__);
+					break;
+				}
+
 				dirty = true;
+				omap_obj->addrs[i] = addr;
 			}
 		}
 
@@ -787,7 +822,7 @@ int omap_gem_get_paddr(struct drm_gem_object *obj,
 
 	mutex_lock(&obj->dev->struct_mutex);
 
-	if (remap && is_shmem(obj) && priv->has_dmm) {
+	if (!is_contiguous(omap_obj) && remap && priv->has_dmm) {
 		if (omap_obj->paddr_cnt == 0) {
 			struct page **pages;
 			uint32_t npages = obj->size >> PAGE_SHIFT;
@@ -834,7 +869,7 @@ int omap_gem_get_paddr(struct drm_gem_object *obj,
 		omap_obj->paddr_cnt++;
 
 		*paddr = omap_obj->paddr;
-	} else if (omap_obj->flags & OMAP_BO_DMA) {
+	} else if (is_contiguous(omap_obj)) {
 		*paddr = omap_obj->paddr;
 	} else {
 		ret = -EINVAL;
@@ -1138,20 +1173,6 @@ unlock:
 	return ret;
 }
 
-/* it is a bit lame to handle updates in this sort of polling way, but
- * in case of PVR, the GPU can directly update read/write complete
- * values, and not really tell us which ones it updated.. this also
- * means that sync_lock is not quite sufficient.  So we'll need to
- * do something a bit better when it comes time to add support for
- * separate 2d hw..
- */
-void omap_gem_op_update(void)
-{
-	spin_lock(&sync_lock);
-	sync_op_update();
-	spin_unlock(&sync_lock);
-}
-
 /* mark the start of read and/or write operation */
 int omap_gem_op_start(struct drm_gem_object *obj, enum omap_gem_op op)
 {
@@ -1219,7 +1240,7 @@ int omap_gem_op_sync(struct drm_gem_object *obj, enum omap_gem_op op)
  * is currently blocked..  fxn() can be called from any context
  *
  * (TODO for now fxn is called back from whichever context calls
- * omap_gem_op_update().. but this could be better defined later
+ * omap_gem_op_finish().. but this could be better defined later
  * if needed)
  *
  * TODO more code in common w/ _sync()..
@@ -1261,50 +1282,10 @@ int omap_gem_op_async(struct drm_gem_object *obj, enum omap_gem_op op,
 	return 0;
 }
 
-/* special API so PVR can update the buffer to use a sync-object allocated
- * from it's sync-obj heap.  Only used for a newly allocated (from PVR's
- * perspective) sync-object, so we overwrite the new syncobj w/ values
- * from the already allocated syncobj (if there is one)
- */
-int omap_gem_set_sync_object(struct drm_gem_object *obj, void *syncobj)
-{
-	struct omap_gem_object *omap_obj = to_omap_bo(obj);
-	int ret = 0;
-
-	spin_lock(&sync_lock);
-
-	if ((omap_obj->flags & OMAP_BO_EXT_SYNC) && !syncobj) {
-		/* clearing a previously set syncobj */
-		syncobj = kmemdup(omap_obj->sync, sizeof(*omap_obj->sync),
-				  GFP_ATOMIC);
-		if (!syncobj) {
-			ret = -ENOMEM;
-			goto unlock;
-		}
-		omap_obj->flags &= ~OMAP_BO_EXT_SYNC;
-		omap_obj->sync = syncobj;
-	} else if (syncobj && !(omap_obj->flags & OMAP_BO_EXT_SYNC)) {
-		/* replacing an existing syncobj */
-		if (omap_obj->sync) {
-			memcpy(syncobj, omap_obj->sync, sizeof(*omap_obj->sync));
-			kfree(omap_obj->sync);
-		}
-		omap_obj->flags |= OMAP_BO_EXT_SYNC;
-		omap_obj->sync = syncobj;
-	}
-
-unlock:
-	spin_unlock(&sync_lock);
-	return ret;
-}
-
 /* -----------------------------------------------------------------------------
  * Constructor & Destructor
  */
 
-/* don't call directly.. called from GEM core when it is time to actually
- * free the object..
- */
 void omap_gem_free_object(struct drm_gem_object *obj)
 {
 	struct drm_device *dev = obj->dev;
@@ -1324,22 +1305,23 @@ void omap_gem_free_object(struct drm_gem_object *obj)
 	 */
 	WARN_ON(omap_obj->paddr_cnt > 0);
 
-	/* don't free externally allocated backing memory */
-	if (!(omap_obj->flags & OMAP_BO_EXT_MEM)) {
-		if (omap_obj->pages)
+	if (omap_obj->pages) {
+		if (omap_obj->flags & OMAP_BO_MEM_DMABUF)
+			kfree(omap_obj->pages);
+		else
 			omap_gem_detach_pages(obj);
+	}
 
-		if (!is_shmem(obj)) {
-			dma_free_writecombine(dev->dev, obj->size,
-					omap_obj->vaddr, omap_obj->paddr);
-		} else if (omap_obj->vaddr) {
-			vunmap(omap_obj->vaddr);
-		}
+	if (omap_obj->flags & OMAP_BO_MEM_DMA_API) {
+		dma_free_writecombine(dev->dev, obj->size,
+				omap_obj->vaddr, omap_obj->paddr);
+	} else if (omap_obj->vaddr) {
+		vunmap(omap_obj->vaddr);
+	} else if (obj->import_attach) {
+		drm_prime_gem_destroy(obj, omap_obj->sgt);
 	}
 
-	/* don't free externally allocated syncobj */
-	if (!(omap_obj->flags & OMAP_BO_EXT_SYNC))
-		kfree(omap_obj->sync);
+	kfree(omap_obj->sync);
 
 	drm_gem_object_release(obj);
 
@@ -1357,84 +1339,160 @@ struct drm_gem_object *omap_gem_new(struct drm_device *dev,
 	size_t size;
 	int ret;
 
+	/* Validate the flags and compute the memory and cache flags. */
 	if (flags & OMAP_BO_TILED) {
 		if (!priv->usergart) {
 			dev_err(dev->dev, "Tiled buffers require DMM\n");
 			return NULL;
 		}
 
-		/* tiled buffers are always shmem paged backed.. when they are
-		 * scanned out, they are remapped into DMM/TILER
+		/*
+		 * Tiled buffers are always shmem paged backed. When they are
+		 * scanned out, they are remapped into DMM/TILER.
 		 */
 		flags &= ~OMAP_BO_SCANOUT;
+		flags |= OMAP_BO_MEM_SHMEM;
 
-		/* currently don't allow cached buffers.. there is some caching
-		 * stuff that needs to be handled better
+		/*
+		 * Currently don't allow cached buffers. There is some caching
+		 * stuff that needs to be handled better.
 		 */
 		flags &= ~(OMAP_BO_CACHED|OMAP_BO_WC|OMAP_BO_UNCACHED);
 		flags |= tiler_get_cpu_cache_flags();
-
-		/* align dimensions to slot boundaries... */
-		tiler_align(gem2fmt(flags),
-				&gsize.tiled.width, &gsize.tiled.height);
-
-		/* ...and calculate size based on aligned dimensions */
-		size = tiler_size(gem2fmt(flags),
-				gsize.tiled.width, gsize.tiled.height);
-	} else {
-		size = PAGE_ALIGN(gsize.bytes);
+	} else if ((flags & OMAP_BO_SCANOUT) && !priv->has_dmm) {
+		/*
+		 * OMAP_BO_SCANOUT hints that the buffer doesn't need to be
+		 * tiled. However, to lower the pressure on memory allocation,
+		 * use contiguous memory only if no TILER is available.
+		 */
+		flags |= OMAP_BO_MEM_DMA_API;
+	} else if (!(flags & OMAP_BO_MEM_DMABUF)) {
+		/*
+		 * All other buffers not backed by dma_buf are shmem-backed.
+		 */
+		flags |= OMAP_BO_MEM_SHMEM;
 	}
 
+	/* Allocate the initialize the OMAP GEM object. */
 	omap_obj = kzalloc(sizeof(*omap_obj), GFP_KERNEL);
 	if (!omap_obj)
 		return NULL;
 
 	obj = &omap_obj->base;
+	omap_obj->flags = flags;
 
-	if ((flags & OMAP_BO_SCANOUT) && !priv->has_dmm) {
-		/* attempt to allocate contiguous memory if we don't
-		 * have DMM for remappign discontiguous buffers
+	if (flags & OMAP_BO_TILED) {
+		/*
+		 * For tiled buffers align dimensions to slot boundaries and
+		 * calculate size based on aligned dimensions.
 		 */
-		omap_obj->vaddr =  dma_alloc_writecombine(dev->dev, size,
-				&omap_obj->paddr, GFP_KERNEL);
-		if (!omap_obj->vaddr) {
-			kfree(omap_obj);
+		tiler_align(gem2fmt(flags), &gsize.tiled.width,
+			    &gsize.tiled.height);
 
-			return NULL;
-		}
-
-		flags |= OMAP_BO_DMA;
-	}
+		size = tiler_size(gem2fmt(flags), gsize.tiled.width,
+				  gsize.tiled.height);
 
-	spin_lock(&priv->list_lock);
-	list_add(&omap_obj->mm_list, &priv->obj_list);
-	spin_unlock(&priv->list_lock);
-
-	omap_obj->flags = flags;
-
-	if (flags & OMAP_BO_TILED) {
 		omap_obj->width = gsize.tiled.width;
 		omap_obj->height = gsize.tiled.height;
+	} else {
+		size = PAGE_ALIGN(gsize.bytes);
 	}
 
-	if (flags & (OMAP_BO_DMA|OMAP_BO_EXT_MEM)) {
+	/* Initialize the GEM object. */
+	if (!(flags & OMAP_BO_MEM_SHMEM)) {
 		drm_gem_private_object_init(dev, obj, size);
 	} else {
 		ret = drm_gem_object_init(dev, obj, size);
 		if (ret)
-			goto fail;
+			goto err_free;
 
 		mapping = file_inode(obj->filp)->i_mapping;
 		mapping_set_gfp_mask(mapping, GFP_USER | __GFP_DMA32);
 	}
 
+	/* Allocate memory if needed. */
+	if (flags & OMAP_BO_MEM_DMA_API) {
+		omap_obj->vaddr = dma_alloc_writecombine(dev->dev, size,
+							 &omap_obj->paddr,
+							 GFP_KERNEL);
+		if (!omap_obj->vaddr)
+			goto err_release;
+	}
+
+	spin_lock(&priv->list_lock);
+	list_add(&omap_obj->mm_list, &priv->obj_list);
+	spin_unlock(&priv->list_lock);
+
 	return obj;
 
-fail:
-	omap_gem_free_object(obj);
+err_release:
+	drm_gem_object_release(obj);
+err_free:
+	kfree(omap_obj);
 	return NULL;
 }
 
+struct drm_gem_object *omap_gem_new_dmabuf(struct drm_device *dev, size_t size,
+					   struct sg_table *sgt)
+{
+	struct omap_drm_private *priv = dev->dev_private;
+	struct omap_gem_object *omap_obj;
+	struct drm_gem_object *obj;
+	union omap_gem_size gsize;
+
+	/* Without a DMM only physically contiguous buffers can be supported. */
+	if (sgt->orig_nents != 1 && !priv->has_dmm)
+		return ERR_PTR(-EINVAL);
+
+	mutex_lock(&dev->struct_mutex);
+
+	gsize.bytes = PAGE_ALIGN(size);
+	obj = omap_gem_new(dev, gsize, OMAP_BO_MEM_DMABUF | OMAP_BO_WC);
+	if (!obj) {
+		obj = ERR_PTR(-ENOMEM);
+		goto done;
+	}
+
+	omap_obj = to_omap_bo(obj);
+	omap_obj->sgt = sgt;
+
+	if (sgt->orig_nents == 1) {
+		omap_obj->paddr = sg_dma_address(sgt->sgl);
+	} else {
+		/* Create pages list from sgt */
+		struct sg_page_iter iter;
+		struct page **pages;
+		unsigned int npages;
+		unsigned int i = 0;
+
+		npages = DIV_ROUND_UP(size, PAGE_SIZE);
+		pages = kcalloc(npages, sizeof(*pages), GFP_KERNEL);
+		if (!pages) {
+			omap_gem_free_object(obj);
+			obj = ERR_PTR(-ENOMEM);
+			goto done;
+		}
+
+		omap_obj->pages = pages;
+
+		for_each_sg_page(sgt->sgl, &iter, sgt->orig_nents, 0) {
+			pages[i++] = sg_page_iter_page(&iter);
+			if (i > npages)
+				break;
+		}
+
+		if (WARN_ON(i != npages)) {
+			omap_gem_free_object(obj);
+			obj = ERR_PTR(-ENOMEM);
+			goto done;
+		}
+	}
+
+done:
+	mutex_unlock(&dev->struct_mutex);
+	return obj;
+}
+
 /* convenience method to construct a GEM buffer object, and userspace handle */
 int omap_gem_new_handle(struct drm_device *dev, struct drm_file *file,
 		union omap_gem_size gsize, uint32_t flags, uint32_t *handle)
diff --git a/drivers/gpu/drm/omapdrm/omap_gem_dmabuf.c b/drivers/gpu/drm/omapdrm/omap_gem_dmabuf.c
index c75249de2984..af267c35d813 100644
--- a/drivers/gpu/drm/omapdrm/omap_gem_dmabuf.c
+++ b/drivers/gpu/drm/omapdrm/omap_gem_dmabuf.c
@@ -21,6 +21,10 @@
 
 #include "omap_drv.h"
 
+/* -----------------------------------------------------------------------------
+ * DMABUF Export
+ */
+
 static struct sg_table *omap_gem_map_dma_buf(
 		struct dma_buf_attachment *attachment,
 		enum dma_data_direction dir)
@@ -179,15 +183,20 @@ struct dma_buf *omap_gem_prime_export(struct drm_device *dev,
 	return dma_buf_export(&exp_info);
 }
 
+/* -----------------------------------------------------------------------------
+ * DMABUF Import
+ */
+
 struct drm_gem_object *omap_gem_prime_import(struct drm_device *dev,
-		struct dma_buf *buffer)
+					     struct dma_buf *dma_buf)
 {
+	struct dma_buf_attachment *attach;
 	struct drm_gem_object *obj;
+	struct sg_table *sgt;
+	int ret;
 
-	/* is this one of own objects? */
-	if (buffer->ops == &omap_dmabuf_ops) {
-		obj = buffer->priv;
-		/* is it from our device? */
+	if (dma_buf->ops == &omap_dmabuf_ops) {
+		obj = dma_buf->priv;
 		if (obj->dev == dev) {
 			/*
 			 * Importing dmabuf exported from out own gem increases
@@ -198,9 +207,33 @@ struct drm_gem_object *omap_gem_prime_import(struct drm_device *dev,
 		}
 	}
 
-	/*
-	 * TODO add support for importing buffers from other devices..
-	 * for now we don't need this but would be nice to add eventually
-	 */
-	return ERR_PTR(-EINVAL);
+	attach = dma_buf_attach(dma_buf, dev->dev);
+	if (IS_ERR(attach))
+		return ERR_CAST(attach);
+
+	get_dma_buf(dma_buf);
+
+	sgt = dma_buf_map_attachment(attach, DMA_BIDIRECTIONAL);
+	if (IS_ERR(sgt)) {
+		ret = PTR_ERR(sgt);
+		goto fail_detach;
+	}
+
+	obj = omap_gem_new_dmabuf(dev, dma_buf->size, sgt);
+	if (IS_ERR(obj)) {
+		ret = PTR_ERR(obj);
+		goto fail_unmap;
+	}
+
+	obj->import_attach = attach;
+
+	return obj;
+
+fail_unmap:
+	dma_buf_unmap_attachment(attach, sgt, DMA_BIDIRECTIONAL);
+fail_detach:
+	dma_buf_detach(dma_buf, attach);
+	dma_buf_put(dma_buf);
+
+	return ERR_PTR(ret);
 }
diff --git a/drivers/gpu/drm/omapdrm/omap_plane.c b/drivers/gpu/drm/omapdrm/omap_plane.c
index d75b197eff46..93ee538a99f5 100644
--- a/drivers/gpu/drm/omapdrm/omap_plane.c
+++ b/drivers/gpu/drm/omapdrm/omap_plane.c
@@ -177,6 +177,12 @@ static int omap_plane_atomic_check(struct drm_plane *plane,
 	if (state->crtc_y + state->crtc_h > crtc_state->adjusted_mode.vdisplay)
 		return -EINVAL;
 
+	if (state->fb) {
+		if (state->rotation != BIT(DRM_ROTATE_0) &&
+		    !omap_framebuffer_supports_rotation(state->fb))
+			return -EINVAL;
+	}
+
 	return 0;
 }
 
diff --git a/drivers/gpu/drm/panel/panel-simple.c b/drivers/gpu/drm/panel/panel-simple.c
index 2164c999052c..ceb20486dacf 100644
--- a/drivers/gpu/drm/panel/panel-simple.c
+++ b/drivers/gpu/drm/panel/panel-simple.c
@@ -847,6 +847,7 @@ static const struct drm_display_mode innolux_g121x1_l03_mode = {
 	.vsync_end = 768 + 38 + 1,
 	.vtotal = 768 + 38 + 1 + 0,
 	.vrefresh = 60,
+	.flags = DRM_MODE_FLAG_NHSYNC | DRM_MODE_FLAG_NVSYNC,
 };
 
 static const struct panel_desc innolux_g121x1_l03 = {
@@ -982,6 +983,29 @@ static const struct panel_desc lg_lb070wv8 = {
 	.bus_format = MEDIA_BUS_FMT_RGB888_1X7X4_SPWG,
 };
 
+static const struct drm_display_mode lg_lp120up1_mode = {
+	.clock = 162300,
+	.hdisplay = 1920,
+	.hsync_start = 1920 + 40,
+	.hsync_end = 1920 + 40 + 40,
+	.htotal = 1920 + 40 + 40+ 80,
+	.vdisplay = 1280,
+	.vsync_start = 1280 + 4,
+	.vsync_end = 1280 + 4 + 4,
+	.vtotal = 1280 + 4 + 4 + 12,
+	.vrefresh = 60,
+};
+
+static const struct panel_desc lg_lp120up1 = {
+	.modes = &lg_lp120up1_mode,
+	.num_modes = 1,
+	.bpc = 8,
+	.size = {
+		.width = 267,
+		.height = 183,
+	},
+};
+
 static const struct drm_display_mode lg_lp129qe_mode = {
 	.clock = 285250,
 	.hdisplay = 2560,
@@ -1177,6 +1201,42 @@ static const struct panel_desc shelly_sca07010_bfn_lnn = {
 	.bus_format = MEDIA_BUS_FMT_RGB666_1X18,
 };
 
+static const struct display_timing urt_umsh_8596md_timing = {
+	.pixelclock = { 33260000, 33260000, 33260000 },
+	.hactive = { 800, 800, 800 },
+	.hfront_porch = { 41, 41, 41 },
+	.hback_porch = { 216 - 128, 216 - 128, 216 - 128 },
+	.hsync_len = { 71, 128, 128 },
+	.vactive = { 480, 480, 480 },
+	.vfront_porch = { 10, 10, 10 },
+	.vback_porch = { 35 - 2, 35 - 2, 35 - 2 },
+	.vsync_len = { 2, 2, 2 },
+	.flags = DISPLAY_FLAGS_DE_HIGH | DISPLAY_FLAGS_PIXDATA_NEGEDGE |
+		DISPLAY_FLAGS_HSYNC_LOW | DISPLAY_FLAGS_VSYNC_LOW,
+};
+
+static const struct panel_desc urt_umsh_8596md_lvds = {
+	.timings = &urt_umsh_8596md_timing,
+	.num_timings = 1,
+	.bpc = 6,
+	.size = {
+		.width = 152,
+		.height = 91,
+	},
+	.bus_format = MEDIA_BUS_FMT_RGB666_1X7X3_SPWG,
+};
+
+static const struct panel_desc urt_umsh_8596md_parallel = {
+	.timings = &urt_umsh_8596md_timing,
+	.num_timings = 1,
+	.bpc = 6,
+	.size = {
+		.width = 152,
+		.height = 91,
+	},
+	.bus_format = MEDIA_BUS_FMT_RGB666_1X18,
+};
+
 static const struct of_device_id platform_of_match[] = {
 	{
 		.compatible = "ampire,am800480r3tmqwa1h",
@@ -1257,6 +1317,9 @@ static const struct of_device_id platform_of_match[] = {
 		.compatible = "lg,lb070wv8",
 		.data = &lg_lb070wv8,
 	}, {
+		.compatible = "lg,lp120up1",
+		.data = &lg_lp120up1,
+	}, {
 		.compatible = "lg,lp129qe",
 		.data = &lg_lp129qe,
 	}, {
@@ -1281,6 +1344,24 @@ static const struct of_device_id platform_of_match[] = {
 		.compatible = "shelly,sca07010-bfn-lnn",
 		.data = &shelly_sca07010_bfn_lnn,
 	}, {
+		.compatible = "urt,umsh-8596md-t",
+		.data = &urt_umsh_8596md_parallel,
+	}, {
+		.compatible = "urt,umsh-8596md-1t",
+		.data = &urt_umsh_8596md_parallel,
+	}, {
+		.compatible = "urt,umsh-8596md-7t",
+		.data = &urt_umsh_8596md_parallel,
+	}, {
+		.compatible = "urt,umsh-8596md-11t",
+		.data = &urt_umsh_8596md_lvds,
+	}, {
+		.compatible = "urt,umsh-8596md-19t",
+		.data = &urt_umsh_8596md_lvds,
+	}, {
+		.compatible = "urt,umsh-8596md-20t",
+		.data = &urt_umsh_8596md_parallel,
+	}, {
 		/* sentinel */
 	}
 };
diff --git a/drivers/gpu/drm/qxl/qxl_ioctl.c b/drivers/gpu/drm/qxl/qxl_ioctl.c
index 2ae8577497ca..7c2e78201ead 100644
--- a/drivers/gpu/drm/qxl/qxl_ioctl.c
+++ b/drivers/gpu/drm/qxl/qxl_ioctl.c
@@ -168,7 +168,8 @@ static int qxl_process_single_command(struct qxl_device *qdev,
 		       cmd->command_size))
 		return -EFAULT;
 
-	reloc_info = kmalloc(sizeof(struct qxl_reloc_info) * cmd->relocs_num, GFP_KERNEL);
+	reloc_info = kmalloc_array(cmd->relocs_num,
+				   sizeof(struct qxl_reloc_info), GFP_KERNEL);
 	if (!reloc_info)
 		return -ENOMEM;
 
diff --git a/drivers/gpu/drm/qxl/qxl_prime.c b/drivers/gpu/drm/qxl/qxl_prime.c
index 3d031b50a8fd..9f029dda1f07 100644
--- a/drivers/gpu/drm/qxl/qxl_prime.c
+++ b/drivers/gpu/drm/qxl/qxl_prime.c
@@ -68,5 +68,5 @@ int qxl_gem_prime_mmap(struct drm_gem_object *obj,
 		       struct vm_area_struct *area)
 {
 	WARN_ONCE(1, "not implemented");
-	return ENOSYS;
+	return -ENOSYS;
 }
diff --git a/drivers/gpu/drm/radeon/atom.c b/drivers/gpu/drm/radeon/atom.c
index ec1593a6a561..f66c33dd21a3 100644
--- a/drivers/gpu/drm/radeon/atom.c
+++ b/drivers/gpu/drm/radeon/atom.c
@@ -66,9 +66,10 @@ int atom_debug = 0;
 static int atom_execute_table_locked(struct atom_context *ctx, int index, uint32_t * params);
 int atom_execute_table(struct atom_context *ctx, int index, uint32_t * params);
 
-static uint32_t atom_arg_mask[8] =
-    { 0xFFFFFFFF, 0xFFFF, 0xFFFF00, 0xFFFF0000, 0xFF, 0xFF00, 0xFF0000,
-0xFF000000 };
+static uint32_t atom_arg_mask[8] = {
+	0xFFFFFFFF, 0x0000FFFF, 0x00FFFF00, 0xFFFF0000,
+	0x000000FF, 0x0000FF00, 0x00FF0000, 0xFF000000
+};
 static int atom_arg_shift[8] = { 0, 0, 8, 16, 0, 8, 16, 24 };
 
 static int atom_dst_to_src[8][4] = {
diff --git a/drivers/gpu/drm/radeon/atombios_crtc.c b/drivers/gpu/drm/radeon/atombios_crtc.c
index 801dd60ac192..cf61e0856f4a 100644
--- a/drivers/gpu/drm/radeon/atombios_crtc.c
+++ b/drivers/gpu/drm/radeon/atombios_crtc.c
@@ -275,13 +275,13 @@ void atombios_crtc_dpms(struct drm_crtc *crtc, int mode)
 		if (ASIC_IS_DCE3(rdev) && !ASIC_IS_DCE6(rdev))
 			atombios_enable_crtc_memreq(crtc, ATOM_ENABLE);
 		atombios_blank_crtc(crtc, ATOM_DISABLE);
-		drm_vblank_post_modeset(dev, radeon_crtc->crtc_id);
+		drm_vblank_on(dev, radeon_crtc->crtc_id);
 		radeon_crtc_load_lut(crtc);
 		break;
 	case DRM_MODE_DPMS_STANDBY:
 	case DRM_MODE_DPMS_SUSPEND:
 	case DRM_MODE_DPMS_OFF:
-		drm_vblank_pre_modeset(dev, radeon_crtc->crtc_id);
+		drm_vblank_off(dev, radeon_crtc->crtc_id);
 		if (radeon_crtc->enabled)
 			atombios_blank_crtc(crtc, ATOM_ENABLE);
 		if (ASIC_IS_DCE3(rdev) && !ASIC_IS_DCE6(rdev))
@@ -1665,11 +1665,11 @@ int atombios_crtc_set_base(struct drm_crtc *crtc, int x, int y,
 }
 
 int atombios_crtc_set_base_atomic(struct drm_crtc *crtc,
-                                  struct drm_framebuffer *fb,
+				  struct drm_framebuffer *fb,
 				  int x, int y, enum mode_set_atomic state)
 {
-       struct drm_device *dev = crtc->dev;
-       struct radeon_device *rdev = dev->dev_private;
+	struct drm_device *dev = crtc->dev;
+	struct radeon_device *rdev = dev->dev_private;
 
 	if (ASIC_IS_DCE4(rdev))
 		return dce4_crtc_do_set_base(crtc, fb, x, y, 1);
diff --git a/drivers/gpu/drm/radeon/atombios_dp.c b/drivers/gpu/drm/radeon/atombios_dp.c
index 44ee72e04df9..afa9db1dc0e3 100644
--- a/drivers/gpu/drm/radeon/atombios_dp.c
+++ b/drivers/gpu/drm/radeon/atombios_dp.c
@@ -37,10 +37,10 @@
 #define DP_DPCD_SIZE DP_RECEIVER_CAP_SIZE
 
 static char *voltage_names[] = {
-        "0.4V", "0.6V", "0.8V", "1.2V"
+	"0.4V", "0.6V", "0.8V", "1.2V"
 };
 static char *pre_emph_names[] = {
-        "0dB", "3.5dB", "6dB", "9.5dB"
+	"0dB", "3.5dB", "6dB", "9.5dB"
 };
 
 /***** radeon AUX functions *****/
@@ -315,15 +315,27 @@ int radeon_dp_get_dp_link_config(struct drm_connector *connector,
 	unsigned max_lane_num = drm_dp_max_lane_count(dpcd);
 	unsigned lane_num, i, max_pix_clock;
 
-	for (lane_num = 1; lane_num <= max_lane_num; lane_num <<= 1) {
-		for (i = 0; i < ARRAY_SIZE(link_rates) && link_rates[i] <= max_link_rate; i++) {
-			max_pix_clock = (lane_num * link_rates[i] * 8) / bpp;
+	if (radeon_connector_encoder_get_dp_bridge_encoder_id(connector) ==
+	    ENCODER_OBJECT_ID_NUTMEG) {
+		for (lane_num = 1; lane_num <= max_lane_num; lane_num <<= 1) {
+			max_pix_clock = (lane_num * 270000 * 8) / bpp;
 			if (max_pix_clock >= pix_clock) {
 				*dp_lanes = lane_num;
-				*dp_rate = link_rates[i];
+				*dp_rate = 270000;
 				return 0;
 			}
 		}
+	} else {
+		for (lane_num = 1; lane_num <= max_lane_num; lane_num <<= 1) {
+			for (i = 0; i < ARRAY_SIZE(link_rates) && link_rates[i] <= max_link_rate; i++) {
+				max_pix_clock = (lane_num * link_rates[i] * 8) / bpp;
+				if (max_pix_clock >= pix_clock) {
+					*dp_lanes = lane_num;
+					*dp_rate = link_rates[i];
+					return 0;
+				}
+			}
+		}
 	}
 
 	return -EINVAL;
diff --git a/drivers/gpu/drm/radeon/atombios_encoders.c b/drivers/gpu/drm/radeon/atombios_encoders.c
index 1603751b1164..edd05cdb0cd8 100644
--- a/drivers/gpu/drm/radeon/atombios_encoders.c
+++ b/drivers/gpu/drm/radeon/atombios_encoders.c
@@ -892,8 +892,6 @@ atombios_dig_encoder_setup2(struct drm_encoder *encoder, int action, int panel_m
 			else
 				args.v1.ucLaneNum = 4;
 
-			if (ENCODER_MODE_IS_DP(args.v1.ucEncoderMode) && (dp_clock == 270000))
-				args.v1.ucConfig |= ATOM_ENCODER_CONFIG_DPLINKRATE_2_70GHZ;
 			switch (radeon_encoder->encoder_id) {
 			case ENCODER_OBJECT_ID_INTERNAL_UNIPHY:
 				args.v1.ucConfig = ATOM_ENCODER_CONFIG_V2_TRANSMITTER1;
@@ -910,6 +908,10 @@ atombios_dig_encoder_setup2(struct drm_encoder *encoder, int action, int panel_m
 				args.v1.ucConfig |= ATOM_ENCODER_CONFIG_LINKB;
 			else
 				args.v1.ucConfig |= ATOM_ENCODER_CONFIG_LINKA;
+
+			if (ENCODER_MODE_IS_DP(args.v1.ucEncoderMode) && (dp_clock == 270000))
+				args.v1.ucConfig |= ATOM_ENCODER_CONFIG_DPLINKRATE_2_70GHZ;
+
 			break;
 		case 2:
 		case 3:
diff --git a/drivers/gpu/drm/radeon/btc_dpm.c b/drivers/gpu/drm/radeon/btc_dpm.c
index 69556f5e247e..38e5123708e7 100644
--- a/drivers/gpu/drm/radeon/btc_dpm.c
+++ b/drivers/gpu/drm/radeon/btc_dpm.c
@@ -1163,12 +1163,11 @@ u32 btc_valid_sclk[40] =
 	155000, 160000, 165000, 170000, 175000, 180000, 185000, 190000, 195000, 200000
 };
 
-static const struct radeon_blacklist_clocks btc_blacklist_clocks[] =
-{
-        { 10000, 30000, RADEON_SCLK_UP },
-        { 15000, 30000, RADEON_SCLK_UP },
-        { 20000, 30000, RADEON_SCLK_UP },
-        { 25000, 30000, RADEON_SCLK_UP }
+static const struct radeon_blacklist_clocks btc_blacklist_clocks[] = {
+	{ 10000, 30000, RADEON_SCLK_UP },
+	{ 15000, 30000, RADEON_SCLK_UP },
+	{ 20000, 30000, RADEON_SCLK_UP },
+	{ 25000, 30000, RADEON_SCLK_UP }
 };
 
 void btc_get_max_clock_from_voltage_dependency_table(struct radeon_clock_voltage_dependency_table *table,
@@ -1637,14 +1636,14 @@ static int btc_init_smc_table(struct radeon_device *rdev,
 	cypress_populate_smc_voltage_tables(rdev, table);
 
 	switch (rdev->pm.int_thermal_type) {
-        case THERMAL_TYPE_EVERGREEN:
-        case THERMAL_TYPE_EMC2103_WITH_INTERNAL:
+	case THERMAL_TYPE_EVERGREEN:
+	case THERMAL_TYPE_EMC2103_WITH_INTERNAL:
 		table->thermalProtectType = PPSMC_THERMAL_PROTECT_TYPE_INTERNAL;
 		break;
-        case THERMAL_TYPE_NONE:
+	case THERMAL_TYPE_NONE:
 		table->thermalProtectType = PPSMC_THERMAL_PROTECT_TYPE_NONE;
 		break;
-        default:
+	default:
 		table->thermalProtectType = PPSMC_THERMAL_PROTECT_TYPE_EXTERNAL;
 		break;
 	}
@@ -1860,37 +1859,37 @@ static bool btc_check_s0_mc_reg_index(u16 in_reg, u16 *out_reg)
 	case MC_SEQ_RAS_TIMING >> 2:
 		*out_reg = MC_SEQ_RAS_TIMING_LP >> 2;
 		break;
-        case MC_SEQ_CAS_TIMING >> 2:
+	case MC_SEQ_CAS_TIMING >> 2:
 		*out_reg = MC_SEQ_CAS_TIMING_LP >> 2;
 		break;
-        case MC_SEQ_MISC_TIMING >> 2:
+	case MC_SEQ_MISC_TIMING >> 2:
 		*out_reg = MC_SEQ_MISC_TIMING_LP >> 2;
 		break;
-        case MC_SEQ_MISC_TIMING2 >> 2:
+	case MC_SEQ_MISC_TIMING2 >> 2:
 		*out_reg = MC_SEQ_MISC_TIMING2_LP >> 2;
 		break;
-        case MC_SEQ_RD_CTL_D0 >> 2:
+	case MC_SEQ_RD_CTL_D0 >> 2:
 		*out_reg = MC_SEQ_RD_CTL_D0_LP >> 2;
 		break;
-        case MC_SEQ_RD_CTL_D1 >> 2:
+	case MC_SEQ_RD_CTL_D1 >> 2:
 		*out_reg = MC_SEQ_RD_CTL_D1_LP >> 2;
 		break;
-        case MC_SEQ_WR_CTL_D0 >> 2:
+	case MC_SEQ_WR_CTL_D0 >> 2:
 		*out_reg = MC_SEQ_WR_CTL_D0_LP >> 2;
 		break;
-        case MC_SEQ_WR_CTL_D1 >> 2:
+	case MC_SEQ_WR_CTL_D1 >> 2:
 		*out_reg = MC_SEQ_WR_CTL_D1_LP >> 2;
 		break;
-        case MC_PMG_CMD_EMRS >> 2:
+	case MC_PMG_CMD_EMRS >> 2:
 		*out_reg = MC_SEQ_PMG_CMD_EMRS_LP >> 2;
 		break;
-        case MC_PMG_CMD_MRS >> 2:
+	case MC_PMG_CMD_MRS >> 2:
 		*out_reg = MC_SEQ_PMG_CMD_MRS_LP >> 2;
 		break;
-        case MC_PMG_CMD_MRS1 >> 2:
+	case MC_PMG_CMD_MRS1 >> 2:
 		*out_reg = MC_SEQ_PMG_CMD_MRS1_LP >> 2;
 		break;
-        default:
+	default:
 		result = false;
 		break;
 	}
diff --git a/drivers/gpu/drm/radeon/ci_dpm.c b/drivers/gpu/drm/radeon/ci_dpm.c
index 4a09947be244..35e0fc3ae8a7 100644
--- a/drivers/gpu/drm/radeon/ci_dpm.c
+++ b/drivers/gpu/drm/radeon/ci_dpm.c
@@ -192,9 +192,9 @@ static void ci_fan_ctrl_set_default_mode(struct radeon_device *rdev);
 
 static struct ci_power_info *ci_get_pi(struct radeon_device *rdev)
 {
-        struct ci_power_info *pi = rdev->pm.dpm.priv;
+	struct ci_power_info *pi = rdev->pm.dpm.priv;
 
-        return pi;
+	return pi;
 }
 
 static struct ci_ps *ci_get_ps(struct radeon_ps *rps)
@@ -1632,7 +1632,7 @@ static int ci_notify_hw_of_power_source(struct radeon_device *rdev,
 	else
 		power_limit = (u32)(cac_tdp_table->battery_power_limit * 256);
 
-        ci_set_power_limit(rdev, power_limit);
+	ci_set_power_limit(rdev, power_limit);
 
 	if (pi->caps_automatic_dc_transition) {
 		if (ac_power)
@@ -2017,9 +2017,9 @@ static void ci_enable_display_gap(struct radeon_device *rdev)
 {
 	u32 tmp = RREG32_SMC(CG_DISPLAY_GAP_CNTL);
 
-        tmp &= ~(DISP_GAP_MASK | DISP_GAP_MCHG_MASK);
-        tmp |= (DISP_GAP(R600_PM_DISPLAY_GAP_IGNORE) |
-                DISP_GAP_MCHG(R600_PM_DISPLAY_GAP_VBLANK));
+	tmp &= ~(DISP_GAP_MASK | DISP_GAP_MCHG_MASK);
+	tmp |= (DISP_GAP(R600_PM_DISPLAY_GAP_IGNORE) |
+		DISP_GAP_MCHG(R600_PM_DISPLAY_GAP_VBLANK));
 
 	WREG32_SMC(CG_DISPLAY_GAP_CNTL, tmp);
 }
@@ -2938,8 +2938,8 @@ static int ci_populate_single_memory_level(struct radeon_device *rdev,
 
 	memory_level->MinVddc = cpu_to_be32(memory_level->MinVddc * VOLTAGE_SCALE);
 	memory_level->MinVddcPhases = cpu_to_be32(memory_level->MinVddcPhases);
-        memory_level->MinVddci = cpu_to_be32(memory_level->MinVddci * VOLTAGE_SCALE);
-        memory_level->MinMvdd = cpu_to_be32(memory_level->MinMvdd * VOLTAGE_SCALE);
+	memory_level->MinVddci = cpu_to_be32(memory_level->MinVddci * VOLTAGE_SCALE);
+	memory_level->MinMvdd = cpu_to_be32(memory_level->MinMvdd * VOLTAGE_SCALE);
 
 	memory_level->MclkFrequency = cpu_to_be32(memory_level->MclkFrequency);
 	memory_level->ActivityLevel = cpu_to_be16(memory_level->ActivityLevel);
@@ -3152,7 +3152,7 @@ static int ci_calculate_sclk_params(struct radeon_device *rdev,
 
 	spll_func_cntl_3 &= ~SPLL_FB_DIV_MASK;
 	spll_func_cntl_3 |= SPLL_FB_DIV(fbdiv);
-        spll_func_cntl_3 |= SPLL_DITHEN;
+	spll_func_cntl_3 |= SPLL_DITHEN;
 
 	if (pi->caps_sclk_ss_support) {
 		struct radeon_atom_ss ss;
@@ -3229,7 +3229,7 @@ static int ci_populate_single_graphic_level(struct radeon_device *rdev,
 	graphic_level->DisplayWatermark = PPSMC_DISPLAY_WATERMARK_LOW;
 
 	graphic_level->Flags = cpu_to_be32(graphic_level->Flags);
-        graphic_level->MinVddc = cpu_to_be32(graphic_level->MinVddc * VOLTAGE_SCALE);
+	graphic_level->MinVddc = cpu_to_be32(graphic_level->MinVddc * VOLTAGE_SCALE);
 	graphic_level->MinVddcPhases = cpu_to_be32(graphic_level->MinVddcPhases);
 	graphic_level->SclkFrequency = cpu_to_be32(graphic_level->SclkFrequency);
 	graphic_level->ActivityLevel = cpu_to_be16(graphic_level->ActivityLevel);
@@ -4393,7 +4393,7 @@ static bool ci_check_s0_mc_reg_index(u16 in_reg, u16 *out_reg)
 		break;
 	case MC_SEQ_CAS_TIMING >> 2:
 		*out_reg = MC_SEQ_CAS_TIMING_LP >> 2;
-            break;
+		break;
 	case MC_SEQ_MISC_TIMING >> 2:
 		*out_reg = MC_SEQ_MISC_TIMING_LP >> 2;
 		break;
@@ -4625,7 +4625,7 @@ static int ci_initialize_mc_reg_table(struct radeon_device *rdev)
 	if (ret)
 		goto init_mc_done;
 
-        ret = ci_copy_vbios_mc_reg_table(table, ci_table);
+	ret = ci_copy_vbios_mc_reg_table(table, ci_table);
 	if (ret)
 		goto init_mc_done;
 
@@ -4916,7 +4916,7 @@ static int ci_set_private_data_variables_based_on_pptable(struct radeon_device *
 		allowed_mclk_vddc_table->entries[allowed_sclk_vddc_table->count - 1].clk;
 	rdev->pm.dpm.dyn_state.max_clock_voltage_on_ac.vddc =
 		allowed_sclk_vddc_table->entries[allowed_sclk_vddc_table->count - 1].v;
-        rdev->pm.dpm.dyn_state.max_clock_voltage_on_ac.vddci =
+	rdev->pm.dpm.dyn_state.max_clock_voltage_on_ac.vddci =
 		allowed_mclk_vddci_table->entries[allowed_mclk_vddci_table->count - 1].v;
 
 	return 0;
@@ -5517,7 +5517,7 @@ static int ci_parse_power_table(struct radeon_device *rdev)
 	struct _NonClockInfoArray *non_clock_info_array;
 	union power_info *power_info;
 	int index = GetIndexIntoMasterTable(DATA, PowerPlayInfo);
-        u16 data_offset;
+	u16 data_offset;
 	u8 frev, crev;
 	u8 *power_state_offset;
 	struct ci_ps *ps;
@@ -5693,8 +5693,8 @@ int ci_dpm_init(struct radeon_device *rdev)
 		return ret;
 	}
 
-        pi->dll_default_on = false;
-        pi->sram_end = SMC_RAM_END;
+	pi->dll_default_on = false;
+	pi->sram_end = SMC_RAM_END;
 
 	pi->activity_target[0] = CISLAND_TARGETACTIVITY_DFLT;
 	pi->activity_target[1] = CISLAND_TARGETACTIVITY_DFLT;
@@ -5734,9 +5734,9 @@ int ci_dpm_init(struct radeon_device *rdev)
 	pi->caps_uvd_dpm = true;
 	pi->caps_vce_dpm = true;
 
-        ci_get_leakage_voltages(rdev);
-        ci_patch_dependency_tables_with_leakage(rdev);
-        ci_set_private_data_variables_based_on_pptable(rdev);
+	ci_get_leakage_voltages(rdev);
+	ci_patch_dependency_tables_with_leakage(rdev);
+	ci_set_private_data_variables_based_on_pptable(rdev);
 
 	rdev->pm.dpm.dyn_state.vddc_dependency_on_dispclk.entries =
 		kzalloc(4 * sizeof(struct radeon_clock_voltage_dependency_entry), GFP_KERNEL);
@@ -5839,7 +5839,7 @@ int ci_dpm_init(struct radeon_device *rdev)
 			pi->vddci_control = CISLANDS_VOLTAGE_CONTROL_BY_SVID2;
 		else
 			rdev->pm.dpm.platform_caps &= ~ATOM_PP_PLATFORM_CAP_VDDCI_CONTROL;
-        }
+	}
 
 	if (rdev->pm.dpm.platform_caps & ATOM_PP_PLATFORM_CAP_MVDDCONTROL) {
 		if (radeon_atom_is_voltage_gpio(rdev, VOLTAGE_TYPE_MVDDC, VOLTAGE_OBJ_GPIO_LUT))
@@ -5860,7 +5860,7 @@ int ci_dpm_init(struct radeon_device *rdev)
 #endif
 
 	if (atom_parse_data_header(rdev->mode_info.atom_context, index, &size,
-                                   &frev, &crev, &data_offset)) {
+				   &frev, &crev, &data_offset)) {
 		pi->caps_sclk_ss_support = true;
 		pi->caps_mclk_ss_support = true;
 		pi->dynamic_ss = true;
diff --git a/drivers/gpu/drm/radeon/ci_smc.c b/drivers/gpu/drm/radeon/ci_smc.c
index 35c6f648ba04..24760ee3063e 100644
--- a/drivers/gpu/drm/radeon/ci_smc.c
+++ b/drivers/gpu/drm/radeon/ci_smc.c
@@ -194,11 +194,11 @@ PPSMC_Result ci_wait_for_smc_inactive(struct radeon_device *rdev)
 		return PPSMC_Result_OK;
 
 	for (i = 0; i < rdev->usec_timeout; i++) {
-                tmp = RREG32_SMC(SMC_SYSCON_CLOCK_CNTL_0);
-                if ((tmp & CKEN) == 0)
+		tmp = RREG32_SMC(SMC_SYSCON_CLOCK_CNTL_0);
+		if ((tmp & CKEN) == 0)
 			break;
-                udelay(1);
-        }
+		udelay(1);
+	}
 
 	return PPSMC_Result_OK;
 }
diff --git a/drivers/gpu/drm/radeon/cik.c b/drivers/gpu/drm/radeon/cik.c
index 06001400ce8b..8ac82df2efde 100644
--- a/drivers/gpu/drm/radeon/cik.c
+++ b/drivers/gpu/drm/radeon/cik.c
@@ -1712,7 +1712,7 @@ static void cik_init_golden_registers(struct radeon_device *rdev)
  */
 u32 cik_get_xclk(struct radeon_device *rdev)
 {
-        u32 reference_clock = rdev->clock.spll.reference_freq;
+	u32 reference_clock = rdev->clock.spll.reference_freq;
 
 	if (rdev->flags & RADEON_IS_IGP) {
 		if (RREG32_SMC(GENERAL_PWRMGT) & GPU_COUNTER_CLK)
@@ -2343,9 +2343,13 @@ out:
  */
 static void cik_tiling_mode_table_init(struct radeon_device *rdev)
 {
-	const u32 num_tile_mode_states = 32;
-	const u32 num_secondary_tile_mode_states = 16;
-	u32 reg_offset, gb_tile_moden, split_equal_to_row_size;
+	u32 *tile = rdev->config.cik.tile_mode_array;
+	u32 *macrotile = rdev->config.cik.macrotile_mode_array;
+	const u32 num_tile_mode_states =
+			ARRAY_SIZE(rdev->config.cik.tile_mode_array);
+	const u32 num_secondary_tile_mode_states =
+			ARRAY_SIZE(rdev->config.cik.macrotile_mode_array);
+	u32 reg_offset, split_equal_to_row_size;
 	u32 num_pipe_configs;
 	u32 num_rbs = rdev->config.cik.max_backends_per_se *
 		rdev->config.cik.max_shader_engines;
@@ -2367,1032 +2371,669 @@ static void cik_tiling_mode_table_init(struct radeon_device *rdev)
 	if (num_pipe_configs > 8)
 		num_pipe_configs = 16;
 
-	if (num_pipe_configs == 16) {
-		for (reg_offset = 0; reg_offset < num_tile_mode_states; reg_offset++) {
-			switch (reg_offset) {
-			case 0:
-				gb_tile_moden = (ARRAY_MODE(ARRAY_2D_TILED_THIN1) |
-						 MICRO_TILE_MODE_NEW(ADDR_SURF_DEPTH_MICRO_TILING) |
-						 PIPE_CONFIG(ADDR_SURF_P16_32x32_16x16) |
-						 TILE_SPLIT(ADDR_SURF_TILE_SPLIT_64B));
-				break;
-			case 1:
-				gb_tile_moden = (ARRAY_MODE(ARRAY_2D_TILED_THIN1) |
-						 MICRO_TILE_MODE_NEW(ADDR_SURF_DEPTH_MICRO_TILING) |
-						 PIPE_CONFIG(ADDR_SURF_P16_32x32_16x16) |
-						 TILE_SPLIT(ADDR_SURF_TILE_SPLIT_128B));
-				break;
-			case 2:
-				gb_tile_moden = (ARRAY_MODE(ARRAY_2D_TILED_THIN1) |
-						 MICRO_TILE_MODE_NEW(ADDR_SURF_DEPTH_MICRO_TILING) |
-						 PIPE_CONFIG(ADDR_SURF_P16_32x32_16x16) |
-						 TILE_SPLIT(ADDR_SURF_TILE_SPLIT_256B));
-				break;
-			case 3:
-				gb_tile_moden = (ARRAY_MODE(ARRAY_2D_TILED_THIN1) |
-						 MICRO_TILE_MODE_NEW(ADDR_SURF_DEPTH_MICRO_TILING) |
-						 PIPE_CONFIG(ADDR_SURF_P16_32x32_16x16) |
-						 TILE_SPLIT(ADDR_SURF_TILE_SPLIT_512B));
-				break;
-			case 4:
-				gb_tile_moden = (ARRAY_MODE(ARRAY_2D_TILED_THIN1) |
-						 MICRO_TILE_MODE_NEW(ADDR_SURF_DEPTH_MICRO_TILING) |
-						 PIPE_CONFIG(ADDR_SURF_P16_32x32_16x16) |
-						 TILE_SPLIT(split_equal_to_row_size));
-				break;
-			case 5:
-				gb_tile_moden = (ARRAY_MODE(ARRAY_1D_TILED_THIN1) |
-						 PIPE_CONFIG(ADDR_SURF_P16_32x32_16x16) |
-						 MICRO_TILE_MODE_NEW(ADDR_SURF_DEPTH_MICRO_TILING));
-				break;
-			case 6:
-				gb_tile_moden = (ARRAY_MODE(ARRAY_PRT_2D_TILED_THIN1) |
-						 MICRO_TILE_MODE_NEW(ADDR_SURF_DEPTH_MICRO_TILING) |
-						 PIPE_CONFIG(ADDR_SURF_P16_32x32_16x16) |
-						 TILE_SPLIT(ADDR_SURF_TILE_SPLIT_256B));
-				break;
-			case 7:
-				gb_tile_moden = (ARRAY_MODE(ARRAY_PRT_2D_TILED_THIN1) |
-						 MICRO_TILE_MODE_NEW(ADDR_SURF_DEPTH_MICRO_TILING) |
-						 PIPE_CONFIG(ADDR_SURF_P16_32x32_16x16) |
-						 TILE_SPLIT(split_equal_to_row_size));
-				break;
-			case 8:
-				gb_tile_moden = (ARRAY_MODE(ARRAY_LINEAR_ALIGNED) |
-						 PIPE_CONFIG(ADDR_SURF_P16_32x32_16x16));
-				break;
-			case 9:
-				gb_tile_moden = (ARRAY_MODE(ARRAY_1D_TILED_THIN1) |
-						 PIPE_CONFIG(ADDR_SURF_P16_32x32_16x16) |
-						 MICRO_TILE_MODE_NEW(ADDR_SURF_DISPLAY_MICRO_TILING));
-				break;
-			case 10:
-				gb_tile_moden = (ARRAY_MODE(ARRAY_2D_TILED_THIN1) |
-						 MICRO_TILE_MODE_NEW(ADDR_SURF_DISPLAY_MICRO_TILING) |
-						 PIPE_CONFIG(ADDR_SURF_P16_32x32_16x16) |
-						 SAMPLE_SPLIT(ADDR_SURF_SAMPLE_SPLIT_2));
-				break;
-			case 11:
-				gb_tile_moden = (ARRAY_MODE(ARRAY_PRT_TILED_THIN1) |
-						 MICRO_TILE_MODE_NEW(ADDR_SURF_DISPLAY_MICRO_TILING) |
-						 PIPE_CONFIG(ADDR_SURF_P16_32x32_8x16) |
-						 SAMPLE_SPLIT(ADDR_SURF_SAMPLE_SPLIT_2));
-				break;
-			case 12:
-				gb_tile_moden = (ARRAY_MODE(ARRAY_PRT_2D_TILED_THIN1) |
-						 MICRO_TILE_MODE_NEW(ADDR_SURF_DISPLAY_MICRO_TILING) |
-						 PIPE_CONFIG(ADDR_SURF_P16_32x32_16x16) |
-						 SAMPLE_SPLIT(ADDR_SURF_SAMPLE_SPLIT_2));
-				break;
-			case 13:
-				gb_tile_moden = (ARRAY_MODE(ARRAY_1D_TILED_THIN1) |
-						 PIPE_CONFIG(ADDR_SURF_P16_32x32_16x16) |
-						 MICRO_TILE_MODE_NEW(ADDR_SURF_THIN_MICRO_TILING));
-				break;
-			case 14:
-				gb_tile_moden = (ARRAY_MODE(ARRAY_2D_TILED_THIN1) |
-						 MICRO_TILE_MODE_NEW(ADDR_SURF_THIN_MICRO_TILING) |
-						 PIPE_CONFIG(ADDR_SURF_P16_32x32_16x16) |
-						 SAMPLE_SPLIT(ADDR_SURF_SAMPLE_SPLIT_2));
-				break;
-			case 16:
-				gb_tile_moden = (ARRAY_MODE(ARRAY_PRT_TILED_THIN1) |
-						 MICRO_TILE_MODE_NEW(ADDR_SURF_THIN_MICRO_TILING) |
-						 PIPE_CONFIG(ADDR_SURF_P16_32x32_8x16) |
-						 SAMPLE_SPLIT(ADDR_SURF_SAMPLE_SPLIT_2));
-				break;
-			case 17:
-				gb_tile_moden = (ARRAY_MODE(ARRAY_PRT_2D_TILED_THIN1) |
-						 MICRO_TILE_MODE_NEW(ADDR_SURF_THIN_MICRO_TILING) |
-						 PIPE_CONFIG(ADDR_SURF_P16_32x32_16x16) |
-						 SAMPLE_SPLIT(ADDR_SURF_SAMPLE_SPLIT_2));
-				break;
-			case 27:
-				gb_tile_moden = (ARRAY_MODE(ARRAY_1D_TILED_THIN1) |
-						 PIPE_CONFIG(ADDR_SURF_P16_32x32_16x16) |
-						 MICRO_TILE_MODE_NEW(ADDR_SURF_ROTATED_MICRO_TILING));
-				break;
-			case 28:
-				gb_tile_moden = (ARRAY_MODE(ARRAY_2D_TILED_THIN1) |
-						 MICRO_TILE_MODE_NEW(ADDR_SURF_ROTATED_MICRO_TILING) |
-						 PIPE_CONFIG(ADDR_SURF_P16_32x32_16x16) |
-						 SAMPLE_SPLIT(ADDR_SURF_SAMPLE_SPLIT_2));
-				break;
-			case 29:
-				gb_tile_moden = (ARRAY_MODE(ARRAY_PRT_TILED_THIN1) |
-						 MICRO_TILE_MODE_NEW(ADDR_SURF_ROTATED_MICRO_TILING) |
-						 PIPE_CONFIG(ADDR_SURF_P16_32x32_8x16) |
-						 SAMPLE_SPLIT(ADDR_SURF_SAMPLE_SPLIT_2));
-				break;
-			case 30:
-				gb_tile_moden = (ARRAY_MODE(ARRAY_PRT_2D_TILED_THIN1) |
-						 MICRO_TILE_MODE_NEW(ADDR_SURF_ROTATED_MICRO_TILING) |
-						 PIPE_CONFIG(ADDR_SURF_P16_32x32_16x16) |
-						 SAMPLE_SPLIT(ADDR_SURF_SAMPLE_SPLIT_2));
-				break;
-			default:
-				gb_tile_moden = 0;
-				break;
-			}
-			rdev->config.cik.tile_mode_array[reg_offset] = gb_tile_moden;
-			WREG32(GB_TILE_MODE0 + (reg_offset * 4), gb_tile_moden);
-		}
-		for (reg_offset = 0; reg_offset < num_secondary_tile_mode_states; reg_offset++) {
-			switch (reg_offset) {
-			case 0:
-				gb_tile_moden = (BANK_WIDTH(ADDR_SURF_BANK_WIDTH_1) |
-						 BANK_HEIGHT(ADDR_SURF_BANK_HEIGHT_4) |
-						 MACRO_TILE_ASPECT(ADDR_SURF_MACRO_ASPECT_2) |
-						 NUM_BANKS(ADDR_SURF_16_BANK));
-				break;
-			case 1:
-				gb_tile_moden = (BANK_WIDTH(ADDR_SURF_BANK_WIDTH_1) |
-						 BANK_HEIGHT(ADDR_SURF_BANK_HEIGHT_2) |
-						 MACRO_TILE_ASPECT(ADDR_SURF_MACRO_ASPECT_2) |
-						 NUM_BANKS(ADDR_SURF_16_BANK));
-				break;
-			case 2:
-				gb_tile_moden = (BANK_WIDTH(ADDR_SURF_BANK_WIDTH_1) |
-						 BANK_HEIGHT(ADDR_SURF_BANK_HEIGHT_1) |
-						 MACRO_TILE_ASPECT(ADDR_SURF_MACRO_ASPECT_1) |
-						 NUM_BANKS(ADDR_SURF_16_BANK));
-				break;
-			case 3:
-				gb_tile_moden = (BANK_WIDTH(ADDR_SURF_BANK_WIDTH_1) |
-						 BANK_HEIGHT(ADDR_SURF_BANK_HEIGHT_1) |
-						 MACRO_TILE_ASPECT(ADDR_SURF_MACRO_ASPECT_1) |
-						 NUM_BANKS(ADDR_SURF_16_BANK));
-				break;
-			case 4:
-				gb_tile_moden = (BANK_WIDTH(ADDR_SURF_BANK_WIDTH_1) |
-						 BANK_HEIGHT(ADDR_SURF_BANK_HEIGHT_1) |
-						 MACRO_TILE_ASPECT(ADDR_SURF_MACRO_ASPECT_1) |
-						 NUM_BANKS(ADDR_SURF_8_BANK));
-				break;
-			case 5:
-				gb_tile_moden = (BANK_WIDTH(ADDR_SURF_BANK_WIDTH_1) |
-						 BANK_HEIGHT(ADDR_SURF_BANK_HEIGHT_1) |
-						 MACRO_TILE_ASPECT(ADDR_SURF_MACRO_ASPECT_1) |
-						 NUM_BANKS(ADDR_SURF_4_BANK));
-				break;
-			case 6:
-				gb_tile_moden = (BANK_WIDTH(ADDR_SURF_BANK_WIDTH_1) |
-						 BANK_HEIGHT(ADDR_SURF_BANK_HEIGHT_1) |
-						 MACRO_TILE_ASPECT(ADDR_SURF_MACRO_ASPECT_1) |
-						 NUM_BANKS(ADDR_SURF_2_BANK));
-				break;
-			case 8:
-				gb_tile_moden = (BANK_WIDTH(ADDR_SURF_BANK_WIDTH_1) |
-						 BANK_HEIGHT(ADDR_SURF_BANK_HEIGHT_4) |
-						 MACRO_TILE_ASPECT(ADDR_SURF_MACRO_ASPECT_2) |
-						 NUM_BANKS(ADDR_SURF_16_BANK));
-				break;
-			case 9:
-				gb_tile_moden = (BANK_WIDTH(ADDR_SURF_BANK_WIDTH_1) |
-						 BANK_HEIGHT(ADDR_SURF_BANK_HEIGHT_2) |
-						 MACRO_TILE_ASPECT(ADDR_SURF_MACRO_ASPECT_2) |
-						 NUM_BANKS(ADDR_SURF_16_BANK));
-				break;
-			case 10:
-				gb_tile_moden = (BANK_WIDTH(ADDR_SURF_BANK_WIDTH_1) |
-						 BANK_HEIGHT(ADDR_SURF_BANK_HEIGHT_1) |
-						 MACRO_TILE_ASPECT(ADDR_SURF_MACRO_ASPECT_1) |
-						 NUM_BANKS(ADDR_SURF_16_BANK));
-				break;
-			case 11:
-				gb_tile_moden = (BANK_WIDTH(ADDR_SURF_BANK_WIDTH_1) |
-						 BANK_HEIGHT(ADDR_SURF_BANK_HEIGHT_1) |
-						 MACRO_TILE_ASPECT(ADDR_SURF_MACRO_ASPECT_1) |
-						 NUM_BANKS(ADDR_SURF_8_BANK));
-				break;
-			case 12:
-				gb_tile_moden = (BANK_WIDTH(ADDR_SURF_BANK_WIDTH_1) |
-						 BANK_HEIGHT(ADDR_SURF_BANK_HEIGHT_1) |
-						 MACRO_TILE_ASPECT(ADDR_SURF_MACRO_ASPECT_1) |
-						 NUM_BANKS(ADDR_SURF_4_BANK));
-				break;
-			case 13:
-				gb_tile_moden = (BANK_WIDTH(ADDR_SURF_BANK_WIDTH_1) |
-						 BANK_HEIGHT(ADDR_SURF_BANK_HEIGHT_1) |
-						 MACRO_TILE_ASPECT(ADDR_SURF_MACRO_ASPECT_1) |
-						 NUM_BANKS(ADDR_SURF_2_BANK));
-				break;
-			case 14:
-				gb_tile_moden = (BANK_WIDTH(ADDR_SURF_BANK_WIDTH_1) |
-						 BANK_HEIGHT(ADDR_SURF_BANK_HEIGHT_1) |
-						 MACRO_TILE_ASPECT(ADDR_SURF_MACRO_ASPECT_1) |
-						 NUM_BANKS(ADDR_SURF_2_BANK));
-				break;
-			default:
-				gb_tile_moden = 0;
-				break;
-			}
-			rdev->config.cik.macrotile_mode_array[reg_offset] = gb_tile_moden;
-			WREG32(GB_MACROTILE_MODE0 + (reg_offset * 4), gb_tile_moden);
-		}
-	} else if (num_pipe_configs == 8) {
-		for (reg_offset = 0; reg_offset < num_tile_mode_states; reg_offset++) {
-			switch (reg_offset) {
-			case 0:
-				gb_tile_moden = (ARRAY_MODE(ARRAY_2D_TILED_THIN1) |
-						 MICRO_TILE_MODE_NEW(ADDR_SURF_DEPTH_MICRO_TILING) |
-						 PIPE_CONFIG(ADDR_SURF_P8_32x32_16x16) |
-						 TILE_SPLIT(ADDR_SURF_TILE_SPLIT_64B));
-				break;
-			case 1:
-				gb_tile_moden = (ARRAY_MODE(ARRAY_2D_TILED_THIN1) |
-						 MICRO_TILE_MODE_NEW(ADDR_SURF_DEPTH_MICRO_TILING) |
-						 PIPE_CONFIG(ADDR_SURF_P8_32x32_16x16) |
-						 TILE_SPLIT(ADDR_SURF_TILE_SPLIT_128B));
-				break;
-			case 2:
-				gb_tile_moden = (ARRAY_MODE(ARRAY_2D_TILED_THIN1) |
-						 MICRO_TILE_MODE_NEW(ADDR_SURF_DEPTH_MICRO_TILING) |
-						 PIPE_CONFIG(ADDR_SURF_P8_32x32_16x16) |
-						 TILE_SPLIT(ADDR_SURF_TILE_SPLIT_256B));
-				break;
-			case 3:
-				gb_tile_moden = (ARRAY_MODE(ARRAY_2D_TILED_THIN1) |
-						 MICRO_TILE_MODE_NEW(ADDR_SURF_DEPTH_MICRO_TILING) |
-						 PIPE_CONFIG(ADDR_SURF_P8_32x32_16x16) |
-						 TILE_SPLIT(ADDR_SURF_TILE_SPLIT_512B));
-				break;
-			case 4:
-				gb_tile_moden = (ARRAY_MODE(ARRAY_2D_TILED_THIN1) |
-						 MICRO_TILE_MODE_NEW(ADDR_SURF_DEPTH_MICRO_TILING) |
-						 PIPE_CONFIG(ADDR_SURF_P8_32x32_16x16) |
-						 TILE_SPLIT(split_equal_to_row_size));
-				break;
-			case 5:
-				gb_tile_moden = (ARRAY_MODE(ARRAY_1D_TILED_THIN1) |
-						 PIPE_CONFIG(ADDR_SURF_P8_32x32_16x16) |
-						 MICRO_TILE_MODE_NEW(ADDR_SURF_DEPTH_MICRO_TILING));
-				break;
-			case 6:
-				gb_tile_moden = (ARRAY_MODE(ARRAY_PRT_2D_TILED_THIN1) |
-						 MICRO_TILE_MODE_NEW(ADDR_SURF_DEPTH_MICRO_TILING) |
-						 PIPE_CONFIG(ADDR_SURF_P8_32x32_16x16) |
-						 TILE_SPLIT(ADDR_SURF_TILE_SPLIT_256B));
-				break;
-			case 7:
-				gb_tile_moden = (ARRAY_MODE(ARRAY_PRT_2D_TILED_THIN1) |
-						 MICRO_TILE_MODE_NEW(ADDR_SURF_DEPTH_MICRO_TILING) |
-						 PIPE_CONFIG(ADDR_SURF_P8_32x32_16x16) |
-						 TILE_SPLIT(split_equal_to_row_size));
-				break;
-			case 8:
-				gb_tile_moden = (ARRAY_MODE(ARRAY_LINEAR_ALIGNED) |
-						 PIPE_CONFIG(ADDR_SURF_P8_32x32_16x16));
-				break;
-			case 9:
-				gb_tile_moden = (ARRAY_MODE(ARRAY_1D_TILED_THIN1) |
-						 PIPE_CONFIG(ADDR_SURF_P8_32x32_16x16) |
-						 MICRO_TILE_MODE_NEW(ADDR_SURF_DISPLAY_MICRO_TILING));
-				break;
-			case 10:
-				gb_tile_moden = (ARRAY_MODE(ARRAY_2D_TILED_THIN1) |
-						 MICRO_TILE_MODE_NEW(ADDR_SURF_DISPLAY_MICRO_TILING) |
-						 PIPE_CONFIG(ADDR_SURF_P8_32x32_16x16) |
-						 SAMPLE_SPLIT(ADDR_SURF_SAMPLE_SPLIT_2));
-				break;
-			case 11:
-				gb_tile_moden = (ARRAY_MODE(ARRAY_PRT_TILED_THIN1) |
-						 MICRO_TILE_MODE_NEW(ADDR_SURF_DISPLAY_MICRO_TILING) |
-						 PIPE_CONFIG(ADDR_SURF_P8_32x32_8x16) |
-						 SAMPLE_SPLIT(ADDR_SURF_SAMPLE_SPLIT_2));
-				break;
-			case 12:
-				gb_tile_moden = (ARRAY_MODE(ARRAY_PRT_2D_TILED_THIN1) |
-						 MICRO_TILE_MODE_NEW(ADDR_SURF_DISPLAY_MICRO_TILING) |
-						 PIPE_CONFIG(ADDR_SURF_P8_32x32_16x16) |
-						 SAMPLE_SPLIT(ADDR_SURF_SAMPLE_SPLIT_2));
-				break;
-			case 13:
-				gb_tile_moden = (ARRAY_MODE(ARRAY_1D_TILED_THIN1) |
-						 PIPE_CONFIG(ADDR_SURF_P8_32x32_16x16) |
-						 MICRO_TILE_MODE_NEW(ADDR_SURF_THIN_MICRO_TILING));
-				break;
-			case 14:
-				gb_tile_moden = (ARRAY_MODE(ARRAY_2D_TILED_THIN1) |
-						 MICRO_TILE_MODE_NEW(ADDR_SURF_THIN_MICRO_TILING) |
-						 PIPE_CONFIG(ADDR_SURF_P8_32x32_16x16) |
-						 SAMPLE_SPLIT(ADDR_SURF_SAMPLE_SPLIT_2));
-				break;
-			case 16:
-				gb_tile_moden = (ARRAY_MODE(ARRAY_PRT_TILED_THIN1) |
-						 MICRO_TILE_MODE_NEW(ADDR_SURF_THIN_MICRO_TILING) |
-						 PIPE_CONFIG(ADDR_SURF_P8_32x32_8x16) |
-						 SAMPLE_SPLIT(ADDR_SURF_SAMPLE_SPLIT_2));
-				break;
-			case 17:
-				gb_tile_moden = (ARRAY_MODE(ARRAY_PRT_2D_TILED_THIN1) |
-						 MICRO_TILE_MODE_NEW(ADDR_SURF_THIN_MICRO_TILING) |
-						 PIPE_CONFIG(ADDR_SURF_P8_32x32_16x16) |
-						 SAMPLE_SPLIT(ADDR_SURF_SAMPLE_SPLIT_2));
-				break;
-			case 27:
-				gb_tile_moden = (ARRAY_MODE(ARRAY_1D_TILED_THIN1) |
-						 PIPE_CONFIG(ADDR_SURF_P8_32x32_16x16) |
-						 MICRO_TILE_MODE_NEW(ADDR_SURF_ROTATED_MICRO_TILING));
-				break;
-			case 28:
-				gb_tile_moden = (ARRAY_MODE(ARRAY_2D_TILED_THIN1) |
-						 MICRO_TILE_MODE_NEW(ADDR_SURF_ROTATED_MICRO_TILING) |
-						 PIPE_CONFIG(ADDR_SURF_P8_32x32_16x16) |
-						 SAMPLE_SPLIT(ADDR_SURF_SAMPLE_SPLIT_2));
-				break;
-			case 29:
-				gb_tile_moden = (ARRAY_MODE(ARRAY_PRT_TILED_THIN1) |
-						 MICRO_TILE_MODE_NEW(ADDR_SURF_ROTATED_MICRO_TILING) |
-						 PIPE_CONFIG(ADDR_SURF_P8_32x32_8x16) |
-						 SAMPLE_SPLIT(ADDR_SURF_SAMPLE_SPLIT_2));
-				break;
-			case 30:
-				gb_tile_moden = (ARRAY_MODE(ARRAY_PRT_2D_TILED_THIN1) |
-						 MICRO_TILE_MODE_NEW(ADDR_SURF_ROTATED_MICRO_TILING) |
-						 PIPE_CONFIG(ADDR_SURF_P8_32x32_16x16) |
-						 SAMPLE_SPLIT(ADDR_SURF_SAMPLE_SPLIT_2));
-				break;
-			default:
-				gb_tile_moden = 0;
-				break;
-			}
-			rdev->config.cik.tile_mode_array[reg_offset] = gb_tile_moden;
-			WREG32(GB_TILE_MODE0 + (reg_offset * 4), gb_tile_moden);
-		}
-		for (reg_offset = 0; reg_offset < num_secondary_tile_mode_states; reg_offset++) {
-			switch (reg_offset) {
-			case 0:
-				gb_tile_moden = (BANK_WIDTH(ADDR_SURF_BANK_WIDTH_1) |
-						 BANK_HEIGHT(ADDR_SURF_BANK_HEIGHT_4) |
-						 MACRO_TILE_ASPECT(ADDR_SURF_MACRO_ASPECT_4) |
-						 NUM_BANKS(ADDR_SURF_16_BANK));
-				break;
-			case 1:
-				gb_tile_moden = (BANK_WIDTH(ADDR_SURF_BANK_WIDTH_1) |
-						 BANK_HEIGHT(ADDR_SURF_BANK_HEIGHT_2) |
-						 MACRO_TILE_ASPECT(ADDR_SURF_MACRO_ASPECT_2) |
-						 NUM_BANKS(ADDR_SURF_16_BANK));
-				break;
-			case 2:
-				gb_tile_moden = (BANK_WIDTH(ADDR_SURF_BANK_WIDTH_1) |
-						 BANK_HEIGHT(ADDR_SURF_BANK_HEIGHT_1) |
-						 MACRO_TILE_ASPECT(ADDR_SURF_MACRO_ASPECT_2) |
-						 NUM_BANKS(ADDR_SURF_16_BANK));
-				break;
-			case 3:
-				gb_tile_moden = (BANK_WIDTH(ADDR_SURF_BANK_WIDTH_1) |
-						 BANK_HEIGHT(ADDR_SURF_BANK_HEIGHT_1) |
-						 MACRO_TILE_ASPECT(ADDR_SURF_MACRO_ASPECT_2) |
-						 NUM_BANKS(ADDR_SURF_16_BANK));
-				break;
-			case 4:
-				gb_tile_moden = (BANK_WIDTH(ADDR_SURF_BANK_WIDTH_1) |
-						 BANK_HEIGHT(ADDR_SURF_BANK_HEIGHT_1) |
-						 MACRO_TILE_ASPECT(ADDR_SURF_MACRO_ASPECT_1) |
-						 NUM_BANKS(ADDR_SURF_8_BANK));
-				break;
-			case 5:
-				gb_tile_moden = (BANK_WIDTH(ADDR_SURF_BANK_WIDTH_1) |
-						 BANK_HEIGHT(ADDR_SURF_BANK_HEIGHT_1) |
-						 MACRO_TILE_ASPECT(ADDR_SURF_MACRO_ASPECT_1) |
-						 NUM_BANKS(ADDR_SURF_4_BANK));
-				break;
-			case 6:
-				gb_tile_moden = (BANK_WIDTH(ADDR_SURF_BANK_WIDTH_1) |
-						 BANK_HEIGHT(ADDR_SURF_BANK_HEIGHT_1) |
-						 MACRO_TILE_ASPECT(ADDR_SURF_MACRO_ASPECT_1) |
-						 NUM_BANKS(ADDR_SURF_2_BANK));
-				break;
-			case 8:
-				gb_tile_moden = (BANK_WIDTH(ADDR_SURF_BANK_WIDTH_1) |
-						 BANK_HEIGHT(ADDR_SURF_BANK_HEIGHT_8) |
-						 MACRO_TILE_ASPECT(ADDR_SURF_MACRO_ASPECT_4) |
-						 NUM_BANKS(ADDR_SURF_16_BANK));
-				break;
-			case 9:
-				gb_tile_moden = (BANK_WIDTH(ADDR_SURF_BANK_WIDTH_1) |
-						 BANK_HEIGHT(ADDR_SURF_BANK_HEIGHT_4) |
-						 MACRO_TILE_ASPECT(ADDR_SURF_MACRO_ASPECT_4) |
-						 NUM_BANKS(ADDR_SURF_16_BANK));
-				break;
-			case 10:
-				gb_tile_moden = (BANK_WIDTH(ADDR_SURF_BANK_WIDTH_1) |
-						 BANK_HEIGHT(ADDR_SURF_BANK_HEIGHT_2) |
-						 MACRO_TILE_ASPECT(ADDR_SURF_MACRO_ASPECT_2) |
-						 NUM_BANKS(ADDR_SURF_16_BANK));
-				break;
-			case 11:
-				gb_tile_moden = (BANK_WIDTH(ADDR_SURF_BANK_WIDTH_1) |
-						 BANK_HEIGHT(ADDR_SURF_BANK_HEIGHT_1) |
-						 MACRO_TILE_ASPECT(ADDR_SURF_MACRO_ASPECT_2) |
-						 NUM_BANKS(ADDR_SURF_16_BANK));
-				break;
-			case 12:
-				gb_tile_moden = (BANK_WIDTH(ADDR_SURF_BANK_WIDTH_1) |
-						 BANK_HEIGHT(ADDR_SURF_BANK_HEIGHT_1) |
-						 MACRO_TILE_ASPECT(ADDR_SURF_MACRO_ASPECT_1) |
-						 NUM_BANKS(ADDR_SURF_8_BANK));
-				break;
-			case 13:
-				gb_tile_moden = (BANK_WIDTH(ADDR_SURF_BANK_WIDTH_1) |
-						 BANK_HEIGHT(ADDR_SURF_BANK_HEIGHT_1) |
-						 MACRO_TILE_ASPECT(ADDR_SURF_MACRO_ASPECT_1) |
-						 NUM_BANKS(ADDR_SURF_4_BANK));
-				break;
-			case 14:
-				gb_tile_moden = (BANK_WIDTH(ADDR_SURF_BANK_WIDTH_1) |
-						 BANK_HEIGHT(ADDR_SURF_BANK_HEIGHT_1) |
-						 MACRO_TILE_ASPECT(ADDR_SURF_MACRO_ASPECT_1) |
-						 NUM_BANKS(ADDR_SURF_2_BANK));
-				break;
-			default:
-				gb_tile_moden = 0;
-				break;
-			}
-			rdev->config.cik.macrotile_mode_array[reg_offset] = gb_tile_moden;
-			WREG32(GB_MACROTILE_MODE0 + (reg_offset * 4), gb_tile_moden);
-		}
-	} else if (num_pipe_configs == 4) {
+	for (reg_offset = 0; reg_offset < num_tile_mode_states; reg_offset++)
+		tile[reg_offset] = 0;
+	for (reg_offset = 0; reg_offset < num_secondary_tile_mode_states; reg_offset++)
+		macrotile[reg_offset] = 0;
+
+	switch(num_pipe_configs) {
+	case 16:
+		tile[0] = (ARRAY_MODE(ARRAY_2D_TILED_THIN1) |
+			   MICRO_TILE_MODE_NEW(ADDR_SURF_DEPTH_MICRO_TILING) |
+			   PIPE_CONFIG(ADDR_SURF_P16_32x32_16x16) |
+			   TILE_SPLIT(ADDR_SURF_TILE_SPLIT_64B));
+		tile[1] = (ARRAY_MODE(ARRAY_2D_TILED_THIN1) |
+			   MICRO_TILE_MODE_NEW(ADDR_SURF_DEPTH_MICRO_TILING) |
+			   PIPE_CONFIG(ADDR_SURF_P16_32x32_16x16) |
+			   TILE_SPLIT(ADDR_SURF_TILE_SPLIT_128B));
+		tile[2] = (ARRAY_MODE(ARRAY_2D_TILED_THIN1) |
+			   MICRO_TILE_MODE_NEW(ADDR_SURF_DEPTH_MICRO_TILING) |
+			   PIPE_CONFIG(ADDR_SURF_P16_32x32_16x16) |
+			   TILE_SPLIT(ADDR_SURF_TILE_SPLIT_256B));
+		tile[3] = (ARRAY_MODE(ARRAY_2D_TILED_THIN1) |
+			   MICRO_TILE_MODE_NEW(ADDR_SURF_DEPTH_MICRO_TILING) |
+			   PIPE_CONFIG(ADDR_SURF_P16_32x32_16x16) |
+			   TILE_SPLIT(ADDR_SURF_TILE_SPLIT_512B));
+		tile[4] = (ARRAY_MODE(ARRAY_2D_TILED_THIN1) |
+			   MICRO_TILE_MODE_NEW(ADDR_SURF_DEPTH_MICRO_TILING) |
+			   PIPE_CONFIG(ADDR_SURF_P16_32x32_16x16) |
+			   TILE_SPLIT(split_equal_to_row_size));
+		tile[5] = (ARRAY_MODE(ARRAY_1D_TILED_THIN1) |
+			   PIPE_CONFIG(ADDR_SURF_P16_32x32_16x16) |
+			   MICRO_TILE_MODE_NEW(ADDR_SURF_DEPTH_MICRO_TILING));
+		tile[6] = (ARRAY_MODE(ARRAY_PRT_2D_TILED_THIN1) |
+			   MICRO_TILE_MODE_NEW(ADDR_SURF_DEPTH_MICRO_TILING) |
+			   PIPE_CONFIG(ADDR_SURF_P16_32x32_16x16) |
+			   TILE_SPLIT(ADDR_SURF_TILE_SPLIT_256B));
+		tile[7] = (ARRAY_MODE(ARRAY_PRT_2D_TILED_THIN1) |
+			   MICRO_TILE_MODE_NEW(ADDR_SURF_DEPTH_MICRO_TILING) |
+			   PIPE_CONFIG(ADDR_SURF_P16_32x32_16x16) |
+			   TILE_SPLIT(split_equal_to_row_size));
+		tile[8] = (ARRAY_MODE(ARRAY_LINEAR_ALIGNED) |
+			   PIPE_CONFIG(ADDR_SURF_P16_32x32_16x16));
+		tile[9] = (ARRAY_MODE(ARRAY_1D_TILED_THIN1) |
+			   PIPE_CONFIG(ADDR_SURF_P16_32x32_16x16) |
+			   MICRO_TILE_MODE_NEW(ADDR_SURF_DISPLAY_MICRO_TILING));
+		tile[10] = (ARRAY_MODE(ARRAY_2D_TILED_THIN1) |
+			    MICRO_TILE_MODE_NEW(ADDR_SURF_DISPLAY_MICRO_TILING) |
+			    PIPE_CONFIG(ADDR_SURF_P16_32x32_16x16) |
+			    SAMPLE_SPLIT(ADDR_SURF_SAMPLE_SPLIT_2));
+		tile[11] = (ARRAY_MODE(ARRAY_PRT_TILED_THIN1) |
+			    MICRO_TILE_MODE_NEW(ADDR_SURF_DISPLAY_MICRO_TILING) |
+			    PIPE_CONFIG(ADDR_SURF_P16_32x32_8x16) |
+			    SAMPLE_SPLIT(ADDR_SURF_SAMPLE_SPLIT_2));
+		tile[12] = (ARRAY_MODE(ARRAY_PRT_2D_TILED_THIN1) |
+			    MICRO_TILE_MODE_NEW(ADDR_SURF_DISPLAY_MICRO_TILING) |
+			    PIPE_CONFIG(ADDR_SURF_P16_32x32_16x16) |
+			    SAMPLE_SPLIT(ADDR_SURF_SAMPLE_SPLIT_2));
+		tile[13] = (ARRAY_MODE(ARRAY_1D_TILED_THIN1) |
+			    PIPE_CONFIG(ADDR_SURF_P16_32x32_16x16) |
+			    MICRO_TILE_MODE_NEW(ADDR_SURF_THIN_MICRO_TILING));
+		tile[14] = (ARRAY_MODE(ARRAY_2D_TILED_THIN1) |
+			    MICRO_TILE_MODE_NEW(ADDR_SURF_THIN_MICRO_TILING) |
+			    PIPE_CONFIG(ADDR_SURF_P16_32x32_16x16) |
+			    SAMPLE_SPLIT(ADDR_SURF_SAMPLE_SPLIT_2));
+		tile[16] = (ARRAY_MODE(ARRAY_PRT_TILED_THIN1) |
+			    MICRO_TILE_MODE_NEW(ADDR_SURF_THIN_MICRO_TILING) |
+			    PIPE_CONFIG(ADDR_SURF_P16_32x32_8x16) |
+			    SAMPLE_SPLIT(ADDR_SURF_SAMPLE_SPLIT_2));
+		tile[17] = (ARRAY_MODE(ARRAY_PRT_2D_TILED_THIN1) |
+			    MICRO_TILE_MODE_NEW(ADDR_SURF_THIN_MICRO_TILING) |
+			    PIPE_CONFIG(ADDR_SURF_P16_32x32_16x16) |
+			    SAMPLE_SPLIT(ADDR_SURF_SAMPLE_SPLIT_2));
+		tile[27] = (ARRAY_MODE(ARRAY_1D_TILED_THIN1) |
+			    PIPE_CONFIG(ADDR_SURF_P16_32x32_16x16) |
+			    MICRO_TILE_MODE_NEW(ADDR_SURF_ROTATED_MICRO_TILING));
+		tile[28] = (ARRAY_MODE(ARRAY_2D_TILED_THIN1) |
+			    MICRO_TILE_MODE_NEW(ADDR_SURF_ROTATED_MICRO_TILING) |
+			    PIPE_CONFIG(ADDR_SURF_P16_32x32_16x16) |
+			    SAMPLE_SPLIT(ADDR_SURF_SAMPLE_SPLIT_2));
+		tile[29] = (ARRAY_MODE(ARRAY_PRT_TILED_THIN1) |
+			    MICRO_TILE_MODE_NEW(ADDR_SURF_ROTATED_MICRO_TILING) |
+			    PIPE_CONFIG(ADDR_SURF_P16_32x32_8x16) |
+			    SAMPLE_SPLIT(ADDR_SURF_SAMPLE_SPLIT_2));
+		tile[30] = (ARRAY_MODE(ARRAY_PRT_2D_TILED_THIN1) |
+			    MICRO_TILE_MODE_NEW(ADDR_SURF_ROTATED_MICRO_TILING) |
+			    PIPE_CONFIG(ADDR_SURF_P16_32x32_16x16) |
+			    SAMPLE_SPLIT(ADDR_SURF_SAMPLE_SPLIT_2));
+
+		macrotile[0] = (BANK_WIDTH(ADDR_SURF_BANK_WIDTH_1) |
+			   BANK_HEIGHT(ADDR_SURF_BANK_HEIGHT_4) |
+			   MACRO_TILE_ASPECT(ADDR_SURF_MACRO_ASPECT_2) |
+			   NUM_BANKS(ADDR_SURF_16_BANK));
+		macrotile[1] = (BANK_WIDTH(ADDR_SURF_BANK_WIDTH_1) |
+			   BANK_HEIGHT(ADDR_SURF_BANK_HEIGHT_2) |
+			   MACRO_TILE_ASPECT(ADDR_SURF_MACRO_ASPECT_2) |
+			   NUM_BANKS(ADDR_SURF_16_BANK));
+		macrotile[2] = (BANK_WIDTH(ADDR_SURF_BANK_WIDTH_1) |
+			   BANK_HEIGHT(ADDR_SURF_BANK_HEIGHT_1) |
+			   MACRO_TILE_ASPECT(ADDR_SURF_MACRO_ASPECT_1) |
+			   NUM_BANKS(ADDR_SURF_16_BANK));
+		macrotile[3] = (BANK_WIDTH(ADDR_SURF_BANK_WIDTH_1) |
+			   BANK_HEIGHT(ADDR_SURF_BANK_HEIGHT_1) |
+			   MACRO_TILE_ASPECT(ADDR_SURF_MACRO_ASPECT_1) |
+			   NUM_BANKS(ADDR_SURF_16_BANK));
+		macrotile[4] = (BANK_WIDTH(ADDR_SURF_BANK_WIDTH_1) |
+			   BANK_HEIGHT(ADDR_SURF_BANK_HEIGHT_1) |
+			   MACRO_TILE_ASPECT(ADDR_SURF_MACRO_ASPECT_1) |
+			   NUM_BANKS(ADDR_SURF_8_BANK));
+		macrotile[5] = (BANK_WIDTH(ADDR_SURF_BANK_WIDTH_1) |
+			   BANK_HEIGHT(ADDR_SURF_BANK_HEIGHT_1) |
+			   MACRO_TILE_ASPECT(ADDR_SURF_MACRO_ASPECT_1) |
+			   NUM_BANKS(ADDR_SURF_4_BANK));
+		macrotile[6] = (BANK_WIDTH(ADDR_SURF_BANK_WIDTH_1) |
+			   BANK_HEIGHT(ADDR_SURF_BANK_HEIGHT_1) |
+			   MACRO_TILE_ASPECT(ADDR_SURF_MACRO_ASPECT_1) |
+			   NUM_BANKS(ADDR_SURF_2_BANK));
+		macrotile[8] = (BANK_WIDTH(ADDR_SURF_BANK_WIDTH_1) |
+			   BANK_HEIGHT(ADDR_SURF_BANK_HEIGHT_4) |
+			   MACRO_TILE_ASPECT(ADDR_SURF_MACRO_ASPECT_2) |
+			   NUM_BANKS(ADDR_SURF_16_BANK));
+		macrotile[9] = (BANK_WIDTH(ADDR_SURF_BANK_WIDTH_1) |
+			   BANK_HEIGHT(ADDR_SURF_BANK_HEIGHT_2) |
+			   MACRO_TILE_ASPECT(ADDR_SURF_MACRO_ASPECT_2) |
+			   NUM_BANKS(ADDR_SURF_16_BANK));
+		macrotile[10] = (BANK_WIDTH(ADDR_SURF_BANK_WIDTH_1) |
+			    BANK_HEIGHT(ADDR_SURF_BANK_HEIGHT_1) |
+			    MACRO_TILE_ASPECT(ADDR_SURF_MACRO_ASPECT_1) |
+			    NUM_BANKS(ADDR_SURF_16_BANK));
+		macrotile[11] = (BANK_WIDTH(ADDR_SURF_BANK_WIDTH_1) |
+			    BANK_HEIGHT(ADDR_SURF_BANK_HEIGHT_1) |
+			    MACRO_TILE_ASPECT(ADDR_SURF_MACRO_ASPECT_1) |
+			    NUM_BANKS(ADDR_SURF_8_BANK));
+		macrotile[12] = (BANK_WIDTH(ADDR_SURF_BANK_WIDTH_1) |
+			    BANK_HEIGHT(ADDR_SURF_BANK_HEIGHT_1) |
+			    MACRO_TILE_ASPECT(ADDR_SURF_MACRO_ASPECT_1) |
+			    NUM_BANKS(ADDR_SURF_4_BANK));
+		macrotile[13] = (BANK_WIDTH(ADDR_SURF_BANK_WIDTH_1) |
+			    BANK_HEIGHT(ADDR_SURF_BANK_HEIGHT_1) |
+			    MACRO_TILE_ASPECT(ADDR_SURF_MACRO_ASPECT_1) |
+			    NUM_BANKS(ADDR_SURF_2_BANK));
+		macrotile[14] = (BANK_WIDTH(ADDR_SURF_BANK_WIDTH_1) |
+			    BANK_HEIGHT(ADDR_SURF_BANK_HEIGHT_1) |
+			    MACRO_TILE_ASPECT(ADDR_SURF_MACRO_ASPECT_1) |
+			    NUM_BANKS(ADDR_SURF_2_BANK));
+
+		for (reg_offset = 0; reg_offset < num_tile_mode_states; reg_offset++)
+			WREG32(GB_TILE_MODE0 + (reg_offset * 4), tile[reg_offset]);
+		for (reg_offset = 0; reg_offset < num_secondary_tile_mode_states; reg_offset++)
+			WREG32(GB_MACROTILE_MODE0 + (reg_offset * 4), macrotile[reg_offset]);
+		break;
+
+	case 8:
+		tile[0] = (ARRAY_MODE(ARRAY_2D_TILED_THIN1) |
+			   MICRO_TILE_MODE_NEW(ADDR_SURF_DEPTH_MICRO_TILING) |
+			   PIPE_CONFIG(ADDR_SURF_P8_32x32_16x16) |
+			   TILE_SPLIT(ADDR_SURF_TILE_SPLIT_64B));
+		tile[1] = (ARRAY_MODE(ARRAY_2D_TILED_THIN1) |
+			   MICRO_TILE_MODE_NEW(ADDR_SURF_DEPTH_MICRO_TILING) |
+			   PIPE_CONFIG(ADDR_SURF_P8_32x32_16x16) |
+			   TILE_SPLIT(ADDR_SURF_TILE_SPLIT_128B));
+		tile[2] = (ARRAY_MODE(ARRAY_2D_TILED_THIN1) |
+			   MICRO_TILE_MODE_NEW(ADDR_SURF_DEPTH_MICRO_TILING) |
+			   PIPE_CONFIG(ADDR_SURF_P8_32x32_16x16) |
+			   TILE_SPLIT(ADDR_SURF_TILE_SPLIT_256B));
+		tile[3] = (ARRAY_MODE(ARRAY_2D_TILED_THIN1) |
+			   MICRO_TILE_MODE_NEW(ADDR_SURF_DEPTH_MICRO_TILING) |
+			   PIPE_CONFIG(ADDR_SURF_P8_32x32_16x16) |
+			   TILE_SPLIT(ADDR_SURF_TILE_SPLIT_512B));
+		tile[4] = (ARRAY_MODE(ARRAY_2D_TILED_THIN1) |
+			   MICRO_TILE_MODE_NEW(ADDR_SURF_DEPTH_MICRO_TILING) |
+			   PIPE_CONFIG(ADDR_SURF_P8_32x32_16x16) |
+			   TILE_SPLIT(split_equal_to_row_size));
+		tile[5] = (ARRAY_MODE(ARRAY_1D_TILED_THIN1) |
+			   PIPE_CONFIG(ADDR_SURF_P8_32x32_16x16) |
+			   MICRO_TILE_MODE_NEW(ADDR_SURF_DEPTH_MICRO_TILING));
+		tile[6] = (ARRAY_MODE(ARRAY_PRT_2D_TILED_THIN1) |
+			   MICRO_TILE_MODE_NEW(ADDR_SURF_DEPTH_MICRO_TILING) |
+			   PIPE_CONFIG(ADDR_SURF_P8_32x32_16x16) |
+			   TILE_SPLIT(ADDR_SURF_TILE_SPLIT_256B));
+		tile[7] = (ARRAY_MODE(ARRAY_PRT_2D_TILED_THIN1) |
+			   MICRO_TILE_MODE_NEW(ADDR_SURF_DEPTH_MICRO_TILING) |
+			   PIPE_CONFIG(ADDR_SURF_P8_32x32_16x16) |
+			   TILE_SPLIT(split_equal_to_row_size));
+		tile[8] = (ARRAY_MODE(ARRAY_LINEAR_ALIGNED) |
+			   PIPE_CONFIG(ADDR_SURF_P8_32x32_16x16));
+		tile[9] = (ARRAY_MODE(ARRAY_1D_TILED_THIN1) |
+			   PIPE_CONFIG(ADDR_SURF_P8_32x32_16x16) |
+			   MICRO_TILE_MODE_NEW(ADDR_SURF_DISPLAY_MICRO_TILING));
+		tile[10] = (ARRAY_MODE(ARRAY_2D_TILED_THIN1) |
+			    MICRO_TILE_MODE_NEW(ADDR_SURF_DISPLAY_MICRO_TILING) |
+			    PIPE_CONFIG(ADDR_SURF_P8_32x32_16x16) |
+			    SAMPLE_SPLIT(ADDR_SURF_SAMPLE_SPLIT_2));
+		tile[11] = (ARRAY_MODE(ARRAY_PRT_TILED_THIN1) |
+			    MICRO_TILE_MODE_NEW(ADDR_SURF_DISPLAY_MICRO_TILING) |
+			    PIPE_CONFIG(ADDR_SURF_P8_32x32_8x16) |
+			    SAMPLE_SPLIT(ADDR_SURF_SAMPLE_SPLIT_2));
+		tile[12] = (ARRAY_MODE(ARRAY_PRT_2D_TILED_THIN1) |
+			    MICRO_TILE_MODE_NEW(ADDR_SURF_DISPLAY_MICRO_TILING) |
+			    PIPE_CONFIG(ADDR_SURF_P8_32x32_16x16) |
+			    SAMPLE_SPLIT(ADDR_SURF_SAMPLE_SPLIT_2));
+		tile[13] = (ARRAY_MODE(ARRAY_1D_TILED_THIN1) |
+			    PIPE_CONFIG(ADDR_SURF_P8_32x32_16x16) |
+			    MICRO_TILE_MODE_NEW(ADDR_SURF_THIN_MICRO_TILING));
+		tile[14] = (ARRAY_MODE(ARRAY_2D_TILED_THIN1) |
+			    MICRO_TILE_MODE_NEW(ADDR_SURF_THIN_MICRO_TILING) |
+			    PIPE_CONFIG(ADDR_SURF_P8_32x32_16x16) |
+			    SAMPLE_SPLIT(ADDR_SURF_SAMPLE_SPLIT_2));
+		tile[16] = (ARRAY_MODE(ARRAY_PRT_TILED_THIN1) |
+			    MICRO_TILE_MODE_NEW(ADDR_SURF_THIN_MICRO_TILING) |
+			    PIPE_CONFIG(ADDR_SURF_P8_32x32_8x16) |
+			    SAMPLE_SPLIT(ADDR_SURF_SAMPLE_SPLIT_2));
+		tile[17] = (ARRAY_MODE(ARRAY_PRT_2D_TILED_THIN1) |
+			    MICRO_TILE_MODE_NEW(ADDR_SURF_THIN_MICRO_TILING) |
+			    PIPE_CONFIG(ADDR_SURF_P8_32x32_16x16) |
+			    SAMPLE_SPLIT(ADDR_SURF_SAMPLE_SPLIT_2));
+		tile[27] = (ARRAY_MODE(ARRAY_1D_TILED_THIN1) |
+			    PIPE_CONFIG(ADDR_SURF_P8_32x32_16x16) |
+			    MICRO_TILE_MODE_NEW(ADDR_SURF_ROTATED_MICRO_TILING));
+		tile[28] = (ARRAY_MODE(ARRAY_2D_TILED_THIN1) |
+			    MICRO_TILE_MODE_NEW(ADDR_SURF_ROTATED_MICRO_TILING) |
+			    PIPE_CONFIG(ADDR_SURF_P8_32x32_16x16) |
+			    SAMPLE_SPLIT(ADDR_SURF_SAMPLE_SPLIT_2));
+		tile[29] = (ARRAY_MODE(ARRAY_PRT_TILED_THIN1) |
+			    MICRO_TILE_MODE_NEW(ADDR_SURF_ROTATED_MICRO_TILING) |
+			    PIPE_CONFIG(ADDR_SURF_P8_32x32_8x16) |
+			    SAMPLE_SPLIT(ADDR_SURF_SAMPLE_SPLIT_2));
+		tile[30] = (ARRAY_MODE(ARRAY_PRT_2D_TILED_THIN1) |
+			    MICRO_TILE_MODE_NEW(ADDR_SURF_ROTATED_MICRO_TILING) |
+			    PIPE_CONFIG(ADDR_SURF_P8_32x32_16x16) |
+			    SAMPLE_SPLIT(ADDR_SURF_SAMPLE_SPLIT_2));
+
+		macrotile[0] = (BANK_WIDTH(ADDR_SURF_BANK_WIDTH_1) |
+				BANK_HEIGHT(ADDR_SURF_BANK_HEIGHT_4) |
+				MACRO_TILE_ASPECT(ADDR_SURF_MACRO_ASPECT_4) |
+				NUM_BANKS(ADDR_SURF_16_BANK));
+		macrotile[1] = (BANK_WIDTH(ADDR_SURF_BANK_WIDTH_1) |
+				BANK_HEIGHT(ADDR_SURF_BANK_HEIGHT_2) |
+				MACRO_TILE_ASPECT(ADDR_SURF_MACRO_ASPECT_2) |
+				NUM_BANKS(ADDR_SURF_16_BANK));
+		macrotile[2] = (BANK_WIDTH(ADDR_SURF_BANK_WIDTH_1) |
+				BANK_HEIGHT(ADDR_SURF_BANK_HEIGHT_1) |
+				MACRO_TILE_ASPECT(ADDR_SURF_MACRO_ASPECT_2) |
+				NUM_BANKS(ADDR_SURF_16_BANK));
+		macrotile[3] = (BANK_WIDTH(ADDR_SURF_BANK_WIDTH_1) |
+				BANK_HEIGHT(ADDR_SURF_BANK_HEIGHT_1) |
+				MACRO_TILE_ASPECT(ADDR_SURF_MACRO_ASPECT_2) |
+				NUM_BANKS(ADDR_SURF_16_BANK));
+		macrotile[4] = (BANK_WIDTH(ADDR_SURF_BANK_WIDTH_1) |
+				BANK_HEIGHT(ADDR_SURF_BANK_HEIGHT_1) |
+				MACRO_TILE_ASPECT(ADDR_SURF_MACRO_ASPECT_1) |
+				NUM_BANKS(ADDR_SURF_8_BANK));
+		macrotile[5] = (BANK_WIDTH(ADDR_SURF_BANK_WIDTH_1) |
+				BANK_HEIGHT(ADDR_SURF_BANK_HEIGHT_1) |
+				MACRO_TILE_ASPECT(ADDR_SURF_MACRO_ASPECT_1) |
+				NUM_BANKS(ADDR_SURF_4_BANK));
+		macrotile[6] = (BANK_WIDTH(ADDR_SURF_BANK_WIDTH_1) |
+				BANK_HEIGHT(ADDR_SURF_BANK_HEIGHT_1) |
+				MACRO_TILE_ASPECT(ADDR_SURF_MACRO_ASPECT_1) |
+				NUM_BANKS(ADDR_SURF_2_BANK));
+		macrotile[8] = (BANK_WIDTH(ADDR_SURF_BANK_WIDTH_1) |
+				BANK_HEIGHT(ADDR_SURF_BANK_HEIGHT_8) |
+				MACRO_TILE_ASPECT(ADDR_SURF_MACRO_ASPECT_4) |
+				NUM_BANKS(ADDR_SURF_16_BANK));
+		macrotile[9] = (BANK_WIDTH(ADDR_SURF_BANK_WIDTH_1) |
+				BANK_HEIGHT(ADDR_SURF_BANK_HEIGHT_4) |
+				MACRO_TILE_ASPECT(ADDR_SURF_MACRO_ASPECT_4) |
+				NUM_BANKS(ADDR_SURF_16_BANK));
+		macrotile[10] = (BANK_WIDTH(ADDR_SURF_BANK_WIDTH_1) |
+				BANK_HEIGHT(ADDR_SURF_BANK_HEIGHT_2) |
+				MACRO_TILE_ASPECT(ADDR_SURF_MACRO_ASPECT_2) |
+				NUM_BANKS(ADDR_SURF_16_BANK));
+		macrotile[11] = (BANK_WIDTH(ADDR_SURF_BANK_WIDTH_1) |
+				BANK_HEIGHT(ADDR_SURF_BANK_HEIGHT_1) |
+				MACRO_TILE_ASPECT(ADDR_SURF_MACRO_ASPECT_2) |
+				NUM_BANKS(ADDR_SURF_16_BANK));
+		macrotile[12] = (BANK_WIDTH(ADDR_SURF_BANK_WIDTH_1) |
+				BANK_HEIGHT(ADDR_SURF_BANK_HEIGHT_1) |
+				MACRO_TILE_ASPECT(ADDR_SURF_MACRO_ASPECT_1) |
+				NUM_BANKS(ADDR_SURF_8_BANK));
+		macrotile[13] = (BANK_WIDTH(ADDR_SURF_BANK_WIDTH_1) |
+				BANK_HEIGHT(ADDR_SURF_BANK_HEIGHT_1) |
+				MACRO_TILE_ASPECT(ADDR_SURF_MACRO_ASPECT_1) |
+				NUM_BANKS(ADDR_SURF_4_BANK));
+		macrotile[14] = (BANK_WIDTH(ADDR_SURF_BANK_WIDTH_1) |
+				BANK_HEIGHT(ADDR_SURF_BANK_HEIGHT_1) |
+				MACRO_TILE_ASPECT(ADDR_SURF_MACRO_ASPECT_1) |
+				NUM_BANKS(ADDR_SURF_2_BANK));
+
+		for (reg_offset = 0; reg_offset < num_tile_mode_states; reg_offset++)
+			WREG32(GB_TILE_MODE0 + (reg_offset * 4), tile[reg_offset]);
+		for (reg_offset = 0; reg_offset < num_secondary_tile_mode_states; reg_offset++)
+			WREG32(GB_MACROTILE_MODE0 + (reg_offset * 4), macrotile[reg_offset]);
+		break;
+
+	case 4:
 		if (num_rbs == 4) {
-			for (reg_offset = 0; reg_offset < num_tile_mode_states; reg_offset++) {
-				switch (reg_offset) {
-				case 0:
-					gb_tile_moden = (ARRAY_MODE(ARRAY_2D_TILED_THIN1) |
-							 MICRO_TILE_MODE_NEW(ADDR_SURF_DEPTH_MICRO_TILING) |
-							 PIPE_CONFIG(ADDR_SURF_P4_16x16) |
-							 TILE_SPLIT(ADDR_SURF_TILE_SPLIT_64B));
-					break;
-				case 1:
-					gb_tile_moden = (ARRAY_MODE(ARRAY_2D_TILED_THIN1) |
-							 MICRO_TILE_MODE_NEW(ADDR_SURF_DEPTH_MICRO_TILING) |
-							 PIPE_CONFIG(ADDR_SURF_P4_16x16) |
-							 TILE_SPLIT(ADDR_SURF_TILE_SPLIT_128B));
-					break;
-				case 2:
-					gb_tile_moden = (ARRAY_MODE(ARRAY_2D_TILED_THIN1) |
-							 MICRO_TILE_MODE_NEW(ADDR_SURF_DEPTH_MICRO_TILING) |
-							 PIPE_CONFIG(ADDR_SURF_P4_16x16) |
-							 TILE_SPLIT(ADDR_SURF_TILE_SPLIT_256B));
-					break;
-				case 3:
-					gb_tile_moden = (ARRAY_MODE(ARRAY_2D_TILED_THIN1) |
-							 MICRO_TILE_MODE_NEW(ADDR_SURF_DEPTH_MICRO_TILING) |
-							 PIPE_CONFIG(ADDR_SURF_P4_16x16) |
-							 TILE_SPLIT(ADDR_SURF_TILE_SPLIT_512B));
-					break;
-				case 4:
-					gb_tile_moden = (ARRAY_MODE(ARRAY_2D_TILED_THIN1) |
-							 MICRO_TILE_MODE_NEW(ADDR_SURF_DEPTH_MICRO_TILING) |
-							 PIPE_CONFIG(ADDR_SURF_P4_16x16) |
-							 TILE_SPLIT(split_equal_to_row_size));
-					break;
-				case 5:
-					gb_tile_moden = (ARRAY_MODE(ARRAY_1D_TILED_THIN1) |
-							 PIPE_CONFIG(ADDR_SURF_P4_16x16) |
-							 MICRO_TILE_MODE_NEW(ADDR_SURF_DEPTH_MICRO_TILING));
-					break;
-				case 6:
-					gb_tile_moden = (ARRAY_MODE(ARRAY_PRT_2D_TILED_THIN1) |
-							 MICRO_TILE_MODE_NEW(ADDR_SURF_DEPTH_MICRO_TILING) |
-							 PIPE_CONFIG(ADDR_SURF_P4_16x16) |
-							 TILE_SPLIT(ADDR_SURF_TILE_SPLIT_256B));
-					break;
-				case 7:
-					gb_tile_moden = (ARRAY_MODE(ARRAY_PRT_2D_TILED_THIN1) |
-							 MICRO_TILE_MODE_NEW(ADDR_SURF_DEPTH_MICRO_TILING) |
-							 PIPE_CONFIG(ADDR_SURF_P4_16x16) |
-							 TILE_SPLIT(split_equal_to_row_size));
-					break;
-				case 8:
-					gb_tile_moden = (ARRAY_MODE(ARRAY_LINEAR_ALIGNED) |
-							 PIPE_CONFIG(ADDR_SURF_P4_16x16));
-					break;
-				case 9:
-					gb_tile_moden = (ARRAY_MODE(ARRAY_1D_TILED_THIN1) |
-							 PIPE_CONFIG(ADDR_SURF_P4_16x16) |
-							 MICRO_TILE_MODE_NEW(ADDR_SURF_DISPLAY_MICRO_TILING));
-					break;
-				case 10:
-					gb_tile_moden = (ARRAY_MODE(ARRAY_2D_TILED_THIN1) |
-							 MICRO_TILE_MODE_NEW(ADDR_SURF_DISPLAY_MICRO_TILING) |
-							 PIPE_CONFIG(ADDR_SURF_P4_16x16) |
-							 SAMPLE_SPLIT(ADDR_SURF_SAMPLE_SPLIT_2));
-					break;
-				case 11:
-					gb_tile_moden = (ARRAY_MODE(ARRAY_PRT_TILED_THIN1) |
-							 MICRO_TILE_MODE_NEW(ADDR_SURF_DISPLAY_MICRO_TILING) |
-							 PIPE_CONFIG(ADDR_SURF_P4_8x16) |
-							 SAMPLE_SPLIT(ADDR_SURF_SAMPLE_SPLIT_2));
-					break;
-				case 12:
-					gb_tile_moden = (ARRAY_MODE(ARRAY_PRT_2D_TILED_THIN1) |
-							 MICRO_TILE_MODE_NEW(ADDR_SURF_DISPLAY_MICRO_TILING) |
-							 PIPE_CONFIG(ADDR_SURF_P4_16x16) |
-							 SAMPLE_SPLIT(ADDR_SURF_SAMPLE_SPLIT_2));
-					break;
-				case 13:
-					gb_tile_moden = (ARRAY_MODE(ARRAY_1D_TILED_THIN1) |
-							 PIPE_CONFIG(ADDR_SURF_P4_16x16) |
-							 MICRO_TILE_MODE_NEW(ADDR_SURF_THIN_MICRO_TILING));
-					break;
-				case 14:
-					gb_tile_moden = (ARRAY_MODE(ARRAY_2D_TILED_THIN1) |
-							 MICRO_TILE_MODE_NEW(ADDR_SURF_THIN_MICRO_TILING) |
-							 PIPE_CONFIG(ADDR_SURF_P4_16x16) |
-							 SAMPLE_SPLIT(ADDR_SURF_SAMPLE_SPLIT_2));
-					break;
-				case 16:
-					gb_tile_moden = (ARRAY_MODE(ARRAY_PRT_TILED_THIN1) |
-							 MICRO_TILE_MODE_NEW(ADDR_SURF_THIN_MICRO_TILING) |
-							 PIPE_CONFIG(ADDR_SURF_P4_8x16) |
-							 SAMPLE_SPLIT(ADDR_SURF_SAMPLE_SPLIT_2));
-					break;
-				case 17:
-					gb_tile_moden = (ARRAY_MODE(ARRAY_PRT_2D_TILED_THIN1) |
-							 MICRO_TILE_MODE_NEW(ADDR_SURF_THIN_MICRO_TILING) |
-							 PIPE_CONFIG(ADDR_SURF_P4_16x16) |
-							 SAMPLE_SPLIT(ADDR_SURF_SAMPLE_SPLIT_2));
-					break;
-				case 27:
-					gb_tile_moden = (ARRAY_MODE(ARRAY_1D_TILED_THIN1) |
-							 PIPE_CONFIG(ADDR_SURF_P4_16x16) |
-							 MICRO_TILE_MODE_NEW(ADDR_SURF_ROTATED_MICRO_TILING));
-					break;
-				case 28:
-					gb_tile_moden = (ARRAY_MODE(ARRAY_PRT_2D_TILED_THIN1) |
-							 MICRO_TILE_MODE_NEW(ADDR_SURF_ROTATED_MICRO_TILING) |
-							 PIPE_CONFIG(ADDR_SURF_P4_16x16) |
-							 SAMPLE_SPLIT(ADDR_SURF_SAMPLE_SPLIT_2));
-					break;
-				case 29:
-					gb_tile_moden = (ARRAY_MODE(ARRAY_PRT_TILED_THIN1) |
-							 MICRO_TILE_MODE_NEW(ADDR_SURF_ROTATED_MICRO_TILING) |
-							 PIPE_CONFIG(ADDR_SURF_P4_8x16) |
-							 SAMPLE_SPLIT(ADDR_SURF_SAMPLE_SPLIT_2));
-					break;
-				case 30:
-					gb_tile_moden = (ARRAY_MODE(ARRAY_PRT_2D_TILED_THIN1) |
-							 MICRO_TILE_MODE_NEW(ADDR_SURF_ROTATED_MICRO_TILING) |
-							 PIPE_CONFIG(ADDR_SURF_P4_16x16) |
-							 SAMPLE_SPLIT(ADDR_SURF_SAMPLE_SPLIT_2));
-					break;
-				default:
-					gb_tile_moden = 0;
-					break;
-				}
-				rdev->config.cik.tile_mode_array[reg_offset] = gb_tile_moden;
-				WREG32(GB_TILE_MODE0 + (reg_offset * 4), gb_tile_moden);
-			}
+		tile[0] = (ARRAY_MODE(ARRAY_2D_TILED_THIN1) |
+			   MICRO_TILE_MODE_NEW(ADDR_SURF_DEPTH_MICRO_TILING) |
+			   PIPE_CONFIG(ADDR_SURF_P4_16x16) |
+			   TILE_SPLIT(ADDR_SURF_TILE_SPLIT_64B));
+		tile[1] = (ARRAY_MODE(ARRAY_2D_TILED_THIN1) |
+			   MICRO_TILE_MODE_NEW(ADDR_SURF_DEPTH_MICRO_TILING) |
+			   PIPE_CONFIG(ADDR_SURF_P4_16x16) |
+			   TILE_SPLIT(ADDR_SURF_TILE_SPLIT_128B));
+		tile[2] = (ARRAY_MODE(ARRAY_2D_TILED_THIN1) |
+			   MICRO_TILE_MODE_NEW(ADDR_SURF_DEPTH_MICRO_TILING) |
+			   PIPE_CONFIG(ADDR_SURF_P4_16x16) |
+			   TILE_SPLIT(ADDR_SURF_TILE_SPLIT_256B));
+		tile[3] = (ARRAY_MODE(ARRAY_2D_TILED_THIN1) |
+			   MICRO_TILE_MODE_NEW(ADDR_SURF_DEPTH_MICRO_TILING) |
+			   PIPE_CONFIG(ADDR_SURF_P4_16x16) |
+			   TILE_SPLIT(ADDR_SURF_TILE_SPLIT_512B));
+		tile[4] = (ARRAY_MODE(ARRAY_2D_TILED_THIN1) |
+			   MICRO_TILE_MODE_NEW(ADDR_SURF_DEPTH_MICRO_TILING) |
+			   PIPE_CONFIG(ADDR_SURF_P4_16x16) |
+			   TILE_SPLIT(split_equal_to_row_size));
+		tile[5] = (ARRAY_MODE(ARRAY_1D_TILED_THIN1) |
+			   PIPE_CONFIG(ADDR_SURF_P4_16x16) |
+			   MICRO_TILE_MODE_NEW(ADDR_SURF_DEPTH_MICRO_TILING));
+		tile[6] = (ARRAY_MODE(ARRAY_PRT_2D_TILED_THIN1) |
+			   MICRO_TILE_MODE_NEW(ADDR_SURF_DEPTH_MICRO_TILING) |
+			   PIPE_CONFIG(ADDR_SURF_P4_16x16) |
+			   TILE_SPLIT(ADDR_SURF_TILE_SPLIT_256B));
+		tile[7] = (ARRAY_MODE(ARRAY_PRT_2D_TILED_THIN1) |
+			   MICRO_TILE_MODE_NEW(ADDR_SURF_DEPTH_MICRO_TILING) |
+			   PIPE_CONFIG(ADDR_SURF_P4_16x16) |
+			   TILE_SPLIT(split_equal_to_row_size));
+		tile[8] = (ARRAY_MODE(ARRAY_LINEAR_ALIGNED) |
+			   PIPE_CONFIG(ADDR_SURF_P4_16x16));
+		tile[9] = (ARRAY_MODE(ARRAY_1D_TILED_THIN1) |
+			   PIPE_CONFIG(ADDR_SURF_P4_16x16) |
+			   MICRO_TILE_MODE_NEW(ADDR_SURF_DISPLAY_MICRO_TILING));
+		tile[10] = (ARRAY_MODE(ARRAY_2D_TILED_THIN1) |
+			    MICRO_TILE_MODE_NEW(ADDR_SURF_DISPLAY_MICRO_TILING) |
+			    PIPE_CONFIG(ADDR_SURF_P4_16x16) |
+			    SAMPLE_SPLIT(ADDR_SURF_SAMPLE_SPLIT_2));
+		tile[11] = (ARRAY_MODE(ARRAY_PRT_TILED_THIN1) |
+			    MICRO_TILE_MODE_NEW(ADDR_SURF_DISPLAY_MICRO_TILING) |
+			    PIPE_CONFIG(ADDR_SURF_P4_8x16) |
+			    SAMPLE_SPLIT(ADDR_SURF_SAMPLE_SPLIT_2));
+		tile[12] = (ARRAY_MODE(ARRAY_PRT_2D_TILED_THIN1) |
+			    MICRO_TILE_MODE_NEW(ADDR_SURF_DISPLAY_MICRO_TILING) |
+			    PIPE_CONFIG(ADDR_SURF_P4_16x16) |
+			    SAMPLE_SPLIT(ADDR_SURF_SAMPLE_SPLIT_2));
+		tile[13] = (ARRAY_MODE(ARRAY_1D_TILED_THIN1) |
+			    PIPE_CONFIG(ADDR_SURF_P4_16x16) |
+			    MICRO_TILE_MODE_NEW(ADDR_SURF_THIN_MICRO_TILING));
+		tile[14] = (ARRAY_MODE(ARRAY_2D_TILED_THIN1) |
+			    MICRO_TILE_MODE_NEW(ADDR_SURF_THIN_MICRO_TILING) |
+			    PIPE_CONFIG(ADDR_SURF_P4_16x16) |
+			    SAMPLE_SPLIT(ADDR_SURF_SAMPLE_SPLIT_2));
+		tile[16] = (ARRAY_MODE(ARRAY_PRT_TILED_THIN1) |
+			    MICRO_TILE_MODE_NEW(ADDR_SURF_THIN_MICRO_TILING) |
+			    PIPE_CONFIG(ADDR_SURF_P4_8x16) |
+			    SAMPLE_SPLIT(ADDR_SURF_SAMPLE_SPLIT_2));
+		tile[17] = (ARRAY_MODE(ARRAY_PRT_2D_TILED_THIN1) |
+			    MICRO_TILE_MODE_NEW(ADDR_SURF_THIN_MICRO_TILING) |
+			    PIPE_CONFIG(ADDR_SURF_P4_16x16) |
+			    SAMPLE_SPLIT(ADDR_SURF_SAMPLE_SPLIT_2));
+		tile[27] = (ARRAY_MODE(ARRAY_1D_TILED_THIN1) |
+			    PIPE_CONFIG(ADDR_SURF_P4_16x16) |
+			    MICRO_TILE_MODE_NEW(ADDR_SURF_ROTATED_MICRO_TILING));
+		tile[28] = (ARRAY_MODE(ARRAY_PRT_2D_TILED_THIN1) |
+			    MICRO_TILE_MODE_NEW(ADDR_SURF_ROTATED_MICRO_TILING) |
+			    PIPE_CONFIG(ADDR_SURF_P4_16x16) |
+			    SAMPLE_SPLIT(ADDR_SURF_SAMPLE_SPLIT_2));
+		tile[29] = (ARRAY_MODE(ARRAY_PRT_TILED_THIN1) |
+			    MICRO_TILE_MODE_NEW(ADDR_SURF_ROTATED_MICRO_TILING) |
+			    PIPE_CONFIG(ADDR_SURF_P4_8x16) |
+			    SAMPLE_SPLIT(ADDR_SURF_SAMPLE_SPLIT_2));
+		tile[30] = (ARRAY_MODE(ARRAY_PRT_2D_TILED_THIN1) |
+			    MICRO_TILE_MODE_NEW(ADDR_SURF_ROTATED_MICRO_TILING) |
+			    PIPE_CONFIG(ADDR_SURF_P4_16x16) |
+			    SAMPLE_SPLIT(ADDR_SURF_SAMPLE_SPLIT_2));
+
 		} else if (num_rbs < 4) {
-			for (reg_offset = 0; reg_offset < num_tile_mode_states; reg_offset++) {
-				switch (reg_offset) {
-				case 0:
-					gb_tile_moden = (ARRAY_MODE(ARRAY_2D_TILED_THIN1) |
-							 MICRO_TILE_MODE_NEW(ADDR_SURF_DEPTH_MICRO_TILING) |
-							 PIPE_CONFIG(ADDR_SURF_P4_8x16) |
-							 TILE_SPLIT(ADDR_SURF_TILE_SPLIT_64B));
-					break;
-				case 1:
-					gb_tile_moden = (ARRAY_MODE(ARRAY_2D_TILED_THIN1) |
-							 MICRO_TILE_MODE_NEW(ADDR_SURF_DEPTH_MICRO_TILING) |
-							 PIPE_CONFIG(ADDR_SURF_P4_8x16) |
-							 TILE_SPLIT(ADDR_SURF_TILE_SPLIT_128B));
-					break;
-				case 2:
-					gb_tile_moden = (ARRAY_MODE(ARRAY_2D_TILED_THIN1) |
-							 MICRO_TILE_MODE_NEW(ADDR_SURF_DEPTH_MICRO_TILING) |
-							 PIPE_CONFIG(ADDR_SURF_P4_8x16) |
-							 TILE_SPLIT(ADDR_SURF_TILE_SPLIT_256B));
-					break;
-				case 3:
-					gb_tile_moden = (ARRAY_MODE(ARRAY_2D_TILED_THIN1) |
-							 MICRO_TILE_MODE_NEW(ADDR_SURF_DEPTH_MICRO_TILING) |
-							 PIPE_CONFIG(ADDR_SURF_P4_8x16) |
-							 TILE_SPLIT(ADDR_SURF_TILE_SPLIT_512B));
-					break;
-				case 4:
-					gb_tile_moden = (ARRAY_MODE(ARRAY_2D_TILED_THIN1) |
-							 MICRO_TILE_MODE_NEW(ADDR_SURF_DEPTH_MICRO_TILING) |
-							 PIPE_CONFIG(ADDR_SURF_P4_8x16) |
-							 TILE_SPLIT(split_equal_to_row_size));
-					break;
-				case 5:
-					gb_tile_moden = (ARRAY_MODE(ARRAY_1D_TILED_THIN1) |
-							 PIPE_CONFIG(ADDR_SURF_P4_8x16) |
-							 MICRO_TILE_MODE_NEW(ADDR_SURF_DEPTH_MICRO_TILING));
-					break;
-				case 6:
-					gb_tile_moden = (ARRAY_MODE(ARRAY_PRT_2D_TILED_THIN1) |
-							 MICRO_TILE_MODE_NEW(ADDR_SURF_DEPTH_MICRO_TILING) |
-							 PIPE_CONFIG(ADDR_SURF_P4_8x16) |
-							 TILE_SPLIT(ADDR_SURF_TILE_SPLIT_256B));
-					break;
-				case 7:
-					gb_tile_moden = (ARRAY_MODE(ARRAY_PRT_2D_TILED_THIN1) |
-							 MICRO_TILE_MODE_NEW(ADDR_SURF_DEPTH_MICRO_TILING) |
-							 PIPE_CONFIG(ADDR_SURF_P4_8x16) |
-							 TILE_SPLIT(split_equal_to_row_size));
-					break;
-				case 8:
-					gb_tile_moden = (ARRAY_MODE(ARRAY_LINEAR_ALIGNED) |
-						 PIPE_CONFIG(ADDR_SURF_P4_8x16));
-					break;
-				case 9:
-					gb_tile_moden = (ARRAY_MODE(ARRAY_1D_TILED_THIN1) |
-							 PIPE_CONFIG(ADDR_SURF_P4_8x16) |
-							 MICRO_TILE_MODE_NEW(ADDR_SURF_DISPLAY_MICRO_TILING));
-					break;
-				case 10:
-					gb_tile_moden = (ARRAY_MODE(ARRAY_2D_TILED_THIN1) |
-							 MICRO_TILE_MODE_NEW(ADDR_SURF_DISPLAY_MICRO_TILING) |
-							 PIPE_CONFIG(ADDR_SURF_P4_8x16) |
-							 SAMPLE_SPLIT(ADDR_SURF_SAMPLE_SPLIT_2));
-					break;
-				case 11:
-					gb_tile_moden = (ARRAY_MODE(ARRAY_PRT_TILED_THIN1) |
-							 MICRO_TILE_MODE_NEW(ADDR_SURF_DISPLAY_MICRO_TILING) |
-							 PIPE_CONFIG(ADDR_SURF_P4_8x16) |
-							 SAMPLE_SPLIT(ADDR_SURF_SAMPLE_SPLIT_2));
-					break;
-				case 12:
-					gb_tile_moden = (ARRAY_MODE(ARRAY_PRT_2D_TILED_THIN1) |
-							 MICRO_TILE_MODE_NEW(ADDR_SURF_DISPLAY_MICRO_TILING) |
-							 PIPE_CONFIG(ADDR_SURF_P4_8x16) |
-							 SAMPLE_SPLIT(ADDR_SURF_SAMPLE_SPLIT_2));
-					break;
-				case 13:
-					gb_tile_moden = (ARRAY_MODE(ARRAY_1D_TILED_THIN1) |
-							 PIPE_CONFIG(ADDR_SURF_P4_8x16) |
-							 MICRO_TILE_MODE_NEW(ADDR_SURF_THIN_MICRO_TILING));
-					break;
-				case 14:
-					gb_tile_moden = (ARRAY_MODE(ARRAY_2D_TILED_THIN1) |
-							 MICRO_TILE_MODE_NEW(ADDR_SURF_THIN_MICRO_TILING) |
-							 PIPE_CONFIG(ADDR_SURF_P4_8x16) |
-							 SAMPLE_SPLIT(ADDR_SURF_SAMPLE_SPLIT_2));
-					break;
-				case 16:
-					gb_tile_moden = (ARRAY_MODE(ARRAY_PRT_TILED_THIN1) |
-							 MICRO_TILE_MODE_NEW(ADDR_SURF_THIN_MICRO_TILING) |
-							 PIPE_CONFIG(ADDR_SURF_P4_8x16) |
-							 SAMPLE_SPLIT(ADDR_SURF_SAMPLE_SPLIT_2));
-					break;
-				case 17:
-					gb_tile_moden = (ARRAY_MODE(ARRAY_PRT_2D_TILED_THIN1) |
-							 MICRO_TILE_MODE_NEW(ADDR_SURF_THIN_MICRO_TILING) |
-							 PIPE_CONFIG(ADDR_SURF_P4_8x16) |
-							 SAMPLE_SPLIT(ADDR_SURF_SAMPLE_SPLIT_2));
-					break;
-				case 27:
-					gb_tile_moden = (ARRAY_MODE(ARRAY_1D_TILED_THIN1) |
-							 PIPE_CONFIG(ADDR_SURF_P4_8x16) |
-							 MICRO_TILE_MODE_NEW(ADDR_SURF_ROTATED_MICRO_TILING));
-					break;
-				case 28:
-					gb_tile_moden = (ARRAY_MODE(ARRAY_PRT_2D_TILED_THIN1) |
-							 MICRO_TILE_MODE_NEW(ADDR_SURF_ROTATED_MICRO_TILING) |
-							 PIPE_CONFIG(ADDR_SURF_P4_8x16) |
-							 SAMPLE_SPLIT(ADDR_SURF_SAMPLE_SPLIT_2));
-					break;
-				case 29:
-					gb_tile_moden = (ARRAY_MODE(ARRAY_PRT_TILED_THIN1) |
-							 MICRO_TILE_MODE_NEW(ADDR_SURF_ROTATED_MICRO_TILING) |
-							 PIPE_CONFIG(ADDR_SURF_P4_8x16) |
-							 SAMPLE_SPLIT(ADDR_SURF_SAMPLE_SPLIT_2));
-					break;
-				case 30:
-					gb_tile_moden = (ARRAY_MODE(ARRAY_PRT_2D_TILED_THIN1) |
-							 MICRO_TILE_MODE_NEW(ADDR_SURF_ROTATED_MICRO_TILING) |
-							 PIPE_CONFIG(ADDR_SURF_P4_8x16) |
-							 SAMPLE_SPLIT(ADDR_SURF_SAMPLE_SPLIT_2));
-					break;
-				default:
-					gb_tile_moden = 0;
-					break;
-				}
-				rdev->config.cik.tile_mode_array[reg_offset] = gb_tile_moden;
-				WREG32(GB_TILE_MODE0 + (reg_offset * 4), gb_tile_moden);
-			}
-		}
-		for (reg_offset = 0; reg_offset < num_secondary_tile_mode_states; reg_offset++) {
-			switch (reg_offset) {
-			case 0:
-				gb_tile_moden = (BANK_WIDTH(ADDR_SURF_BANK_WIDTH_1) |
-						 BANK_HEIGHT(ADDR_SURF_BANK_HEIGHT_4) |
-						 MACRO_TILE_ASPECT(ADDR_SURF_MACRO_ASPECT_4) |
-						 NUM_BANKS(ADDR_SURF_16_BANK));
-				break;
-			case 1:
-				gb_tile_moden = (BANK_WIDTH(ADDR_SURF_BANK_WIDTH_1) |
-						 BANK_HEIGHT(ADDR_SURF_BANK_HEIGHT_2) |
-						 MACRO_TILE_ASPECT(ADDR_SURF_MACRO_ASPECT_4) |
-						 NUM_BANKS(ADDR_SURF_16_BANK));
-				break;
-			case 2:
-				gb_tile_moden = (BANK_WIDTH(ADDR_SURF_BANK_WIDTH_1) |
-						 BANK_HEIGHT(ADDR_SURF_BANK_HEIGHT_1) |
-						 MACRO_TILE_ASPECT(ADDR_SURF_MACRO_ASPECT_2) |
-						 NUM_BANKS(ADDR_SURF_16_BANK));
-				break;
-			case 3:
-				gb_tile_moden = (BANK_WIDTH(ADDR_SURF_BANK_WIDTH_1) |
-						 BANK_HEIGHT(ADDR_SURF_BANK_HEIGHT_1) |
-						 MACRO_TILE_ASPECT(ADDR_SURF_MACRO_ASPECT_2) |
-						 NUM_BANKS(ADDR_SURF_16_BANK));
-				break;
-			case 4:
-				gb_tile_moden = (BANK_WIDTH(ADDR_SURF_BANK_WIDTH_1) |
-						 BANK_HEIGHT(ADDR_SURF_BANK_HEIGHT_1) |
-						 MACRO_TILE_ASPECT(ADDR_SURF_MACRO_ASPECT_2) |
-						 NUM_BANKS(ADDR_SURF_16_BANK));
-				break;
-			case 5:
-				gb_tile_moden = (BANK_WIDTH(ADDR_SURF_BANK_WIDTH_1) |
-						 BANK_HEIGHT(ADDR_SURF_BANK_HEIGHT_1) |
-						 MACRO_TILE_ASPECT(ADDR_SURF_MACRO_ASPECT_2) |
-						 NUM_BANKS(ADDR_SURF_8_BANK));
-				break;
-			case 6:
-				gb_tile_moden = (BANK_WIDTH(ADDR_SURF_BANK_WIDTH_1) |
-						 BANK_HEIGHT(ADDR_SURF_BANK_HEIGHT_1) |
-						 MACRO_TILE_ASPECT(ADDR_SURF_MACRO_ASPECT_1) |
-						 NUM_BANKS(ADDR_SURF_4_BANK));
-				break;
-			case 8:
-				gb_tile_moden = (BANK_WIDTH(ADDR_SURF_BANK_WIDTH_2) |
-						 BANK_HEIGHT(ADDR_SURF_BANK_HEIGHT_8) |
-						 MACRO_TILE_ASPECT(ADDR_SURF_MACRO_ASPECT_4) |
-						 NUM_BANKS(ADDR_SURF_16_BANK));
-				break;
-			case 9:
-				gb_tile_moden = (BANK_WIDTH(ADDR_SURF_BANK_WIDTH_2) |
-						 BANK_HEIGHT(ADDR_SURF_BANK_HEIGHT_4) |
-						 MACRO_TILE_ASPECT(ADDR_SURF_MACRO_ASPECT_4) |
-						 NUM_BANKS(ADDR_SURF_16_BANK));
-				break;
-			case 10:
-				gb_tile_moden = (BANK_WIDTH(ADDR_SURF_BANK_WIDTH_1) |
-						 BANK_HEIGHT(ADDR_SURF_BANK_HEIGHT_4) |
-						 MACRO_TILE_ASPECT(ADDR_SURF_MACRO_ASPECT_4) |
-						 NUM_BANKS(ADDR_SURF_16_BANK));
-				break;
-			case 11:
-				gb_tile_moden = (BANK_WIDTH(ADDR_SURF_BANK_WIDTH_1) |
-						 BANK_HEIGHT(ADDR_SURF_BANK_HEIGHT_2) |
-						 MACRO_TILE_ASPECT(ADDR_SURF_MACRO_ASPECT_4) |
-						 NUM_BANKS(ADDR_SURF_16_BANK));
-				break;
-			case 12:
-				gb_tile_moden = (BANK_WIDTH(ADDR_SURF_BANK_WIDTH_1) |
-						 BANK_HEIGHT(ADDR_SURF_BANK_HEIGHT_1) |
-						 MACRO_TILE_ASPECT(ADDR_SURF_MACRO_ASPECT_2) |
-						 NUM_BANKS(ADDR_SURF_16_BANK));
-				break;
-			case 13:
-				gb_tile_moden = (BANK_WIDTH(ADDR_SURF_BANK_WIDTH_1) |
-						 BANK_HEIGHT(ADDR_SURF_BANK_HEIGHT_1) |
-						 MACRO_TILE_ASPECT(ADDR_SURF_MACRO_ASPECT_2) |
-						 NUM_BANKS(ADDR_SURF_8_BANK));
-				break;
-			case 14:
-				gb_tile_moden = (BANK_WIDTH(ADDR_SURF_BANK_WIDTH_1) |
-						 BANK_HEIGHT(ADDR_SURF_BANK_HEIGHT_1) |
-						 MACRO_TILE_ASPECT(ADDR_SURF_MACRO_ASPECT_1) |
-						 NUM_BANKS(ADDR_SURF_4_BANK));
-				break;
-			default:
-				gb_tile_moden = 0;
-				break;
-			}
-			rdev->config.cik.macrotile_mode_array[reg_offset] = gb_tile_moden;
-			WREG32(GB_MACROTILE_MODE0 + (reg_offset * 4), gb_tile_moden);
-		}
-	} else if (num_pipe_configs == 2) {
-		for (reg_offset = 0; reg_offset < num_tile_mode_states; reg_offset++) {
-			switch (reg_offset) {
-			case 0:
-				gb_tile_moden = (ARRAY_MODE(ARRAY_2D_TILED_THIN1) |
-						 MICRO_TILE_MODE_NEW(ADDR_SURF_DEPTH_MICRO_TILING) |
-						 PIPE_CONFIG(ADDR_SURF_P2) |
-						 TILE_SPLIT(ADDR_SURF_TILE_SPLIT_64B));
-				break;
-			case 1:
-				gb_tile_moden = (ARRAY_MODE(ARRAY_2D_TILED_THIN1) |
-						 MICRO_TILE_MODE_NEW(ADDR_SURF_DEPTH_MICRO_TILING) |
-						 PIPE_CONFIG(ADDR_SURF_P2) |
-						 TILE_SPLIT(ADDR_SURF_TILE_SPLIT_128B));
-				break;
-			case 2:
-				gb_tile_moden = (ARRAY_MODE(ARRAY_2D_TILED_THIN1) |
-						 MICRO_TILE_MODE_NEW(ADDR_SURF_DEPTH_MICRO_TILING) |
-						 PIPE_CONFIG(ADDR_SURF_P2) |
-						 TILE_SPLIT(ADDR_SURF_TILE_SPLIT_256B));
-				break;
-			case 3:
-				gb_tile_moden = (ARRAY_MODE(ARRAY_2D_TILED_THIN1) |
-						 MICRO_TILE_MODE_NEW(ADDR_SURF_DEPTH_MICRO_TILING) |
-						 PIPE_CONFIG(ADDR_SURF_P2) |
-						 TILE_SPLIT(ADDR_SURF_TILE_SPLIT_512B));
-				break;
-			case 4:
-				gb_tile_moden = (ARRAY_MODE(ARRAY_2D_TILED_THIN1) |
-						 MICRO_TILE_MODE_NEW(ADDR_SURF_DEPTH_MICRO_TILING) |
-						 PIPE_CONFIG(ADDR_SURF_P2) |
-						 TILE_SPLIT(split_equal_to_row_size));
-				break;
-			case 5:
-				gb_tile_moden = (ARRAY_MODE(ARRAY_1D_TILED_THIN1) |
-						 PIPE_CONFIG(ADDR_SURF_P2) |
-						 MICRO_TILE_MODE_NEW(ADDR_SURF_DEPTH_MICRO_TILING));
-				break;
-			case 6:
-				gb_tile_moden = (ARRAY_MODE(ARRAY_PRT_2D_TILED_THIN1) |
-						 MICRO_TILE_MODE_NEW(ADDR_SURF_DEPTH_MICRO_TILING) |
-						 PIPE_CONFIG(ADDR_SURF_P2) |
-						 TILE_SPLIT(ADDR_SURF_TILE_SPLIT_256B));
-				break;
-			case 7:
-				gb_tile_moden = (ARRAY_MODE(ARRAY_PRT_2D_TILED_THIN1) |
-						 MICRO_TILE_MODE_NEW(ADDR_SURF_DEPTH_MICRO_TILING) |
-						 PIPE_CONFIG(ADDR_SURF_P2) |
-						 TILE_SPLIT(split_equal_to_row_size));
-				break;
-			case 8:
-				gb_tile_moden = ARRAY_MODE(ARRAY_LINEAR_ALIGNED) |
-						PIPE_CONFIG(ADDR_SURF_P2);
-				break;
-			case 9:
-				gb_tile_moden = (ARRAY_MODE(ARRAY_1D_TILED_THIN1) |
-						 MICRO_TILE_MODE_NEW(ADDR_SURF_DISPLAY_MICRO_TILING) |
-						 PIPE_CONFIG(ADDR_SURF_P2));
-				break;
-			case 10:
-				gb_tile_moden = (ARRAY_MODE(ARRAY_2D_TILED_THIN1) |
-						 MICRO_TILE_MODE_NEW(ADDR_SURF_DISPLAY_MICRO_TILING) |
-						 PIPE_CONFIG(ADDR_SURF_P2) |
-						 SAMPLE_SPLIT(ADDR_SURF_SAMPLE_SPLIT_2));
-				break;
-			case 11:
-				gb_tile_moden = (ARRAY_MODE(ARRAY_PRT_TILED_THIN1) |
-						 MICRO_TILE_MODE_NEW(ADDR_SURF_DISPLAY_MICRO_TILING) |
-						 PIPE_CONFIG(ADDR_SURF_P2) |
-						 SAMPLE_SPLIT(ADDR_SURF_SAMPLE_SPLIT_2));
-				break;
-			case 12:
-				gb_tile_moden = (ARRAY_MODE(ARRAY_PRT_2D_TILED_THIN1) |
-						 MICRO_TILE_MODE_NEW(ADDR_SURF_DISPLAY_MICRO_TILING) |
-						 PIPE_CONFIG(ADDR_SURF_P2) |
-						 SAMPLE_SPLIT(ADDR_SURF_SAMPLE_SPLIT_2));
-				break;
-			case 13:
-				gb_tile_moden = (ARRAY_MODE(ARRAY_1D_TILED_THIN1) |
-						 PIPE_CONFIG(ADDR_SURF_P2) |
-						 MICRO_TILE_MODE_NEW(ADDR_SURF_THIN_MICRO_TILING));
-				break;
-			case 14:
-				gb_tile_moden = (ARRAY_MODE(ARRAY_2D_TILED_THIN1) |
-						 MICRO_TILE_MODE_NEW(ADDR_SURF_THIN_MICRO_TILING) |
-						 PIPE_CONFIG(ADDR_SURF_P2) |
-						 SAMPLE_SPLIT(ADDR_SURF_SAMPLE_SPLIT_2));
-				break;
-			case 16:
-				gb_tile_moden = (ARRAY_MODE(ARRAY_PRT_TILED_THIN1) |
-						 MICRO_TILE_MODE_NEW(ADDR_SURF_THIN_MICRO_TILING) |
-						 PIPE_CONFIG(ADDR_SURF_P2) |
-						 SAMPLE_SPLIT(ADDR_SURF_SAMPLE_SPLIT_2));
-				break;
-			case 17:
-				gb_tile_moden = (ARRAY_MODE(ARRAY_PRT_2D_TILED_THIN1) |
-						 MICRO_TILE_MODE_NEW(ADDR_SURF_THIN_MICRO_TILING) |
-						 PIPE_CONFIG(ADDR_SURF_P2) |
-						 SAMPLE_SPLIT(ADDR_SURF_SAMPLE_SPLIT_2));
-				break;
-			case 27:
-				gb_tile_moden = (ARRAY_MODE(ARRAY_1D_TILED_THIN1) |
-						 MICRO_TILE_MODE_NEW(ADDR_SURF_ROTATED_MICRO_TILING) |
-						 PIPE_CONFIG(ADDR_SURF_P2));
-				break;
-			case 28:
-				gb_tile_moden = (ARRAY_MODE(ARRAY_PRT_2D_TILED_THIN1) |
-						 MICRO_TILE_MODE_NEW(ADDR_SURF_ROTATED_MICRO_TILING) |
-						 PIPE_CONFIG(ADDR_SURF_P2) |
-						 SAMPLE_SPLIT(ADDR_SURF_SAMPLE_SPLIT_2));
-				break;
-			case 29:
-				gb_tile_moden = (ARRAY_MODE(ARRAY_PRT_TILED_THIN1) |
-						 MICRO_TILE_MODE_NEW(ADDR_SURF_ROTATED_MICRO_TILING) |
-						 PIPE_CONFIG(ADDR_SURF_P2) |
-						 SAMPLE_SPLIT(ADDR_SURF_SAMPLE_SPLIT_2));
-				break;
-			case 30:
-				gb_tile_moden = (ARRAY_MODE(ARRAY_PRT_2D_TILED_THIN1) |
-						 MICRO_TILE_MODE_NEW(ADDR_SURF_ROTATED_MICRO_TILING) |
-						 PIPE_CONFIG(ADDR_SURF_P2) |
-						 SAMPLE_SPLIT(ADDR_SURF_SAMPLE_SPLIT_2));
-				break;
-			default:
-				gb_tile_moden = 0;
-				break;
-			}
-			rdev->config.cik.tile_mode_array[reg_offset] = gb_tile_moden;
-			WREG32(GB_TILE_MODE0 + (reg_offset * 4), gb_tile_moden);
-		}
-		for (reg_offset = 0; reg_offset < num_secondary_tile_mode_states; reg_offset++) {
-			switch (reg_offset) {
-			case 0:
-				gb_tile_moden = (BANK_WIDTH(ADDR_SURF_BANK_WIDTH_2) |
-						 BANK_HEIGHT(ADDR_SURF_BANK_HEIGHT_4) |
-						 MACRO_TILE_ASPECT(ADDR_SURF_MACRO_ASPECT_4) |
-						 NUM_BANKS(ADDR_SURF_16_BANK));
-				break;
-			case 1:
-				gb_tile_moden = (BANK_WIDTH(ADDR_SURF_BANK_WIDTH_2) |
-						 BANK_HEIGHT(ADDR_SURF_BANK_HEIGHT_2) |
-						 MACRO_TILE_ASPECT(ADDR_SURF_MACRO_ASPECT_4) |
-						 NUM_BANKS(ADDR_SURF_16_BANK));
-				break;
-			case 2:
-				gb_tile_moden = (BANK_WIDTH(ADDR_SURF_BANK_WIDTH_1) |
-						 BANK_HEIGHT(ADDR_SURF_BANK_HEIGHT_2) |
-						 MACRO_TILE_ASPECT(ADDR_SURF_MACRO_ASPECT_4) |
-						 NUM_BANKS(ADDR_SURF_16_BANK));
-				break;
-			case 3:
-				gb_tile_moden = (BANK_WIDTH(ADDR_SURF_BANK_WIDTH_1) |
-						 BANK_HEIGHT(ADDR_SURF_BANK_HEIGHT_1) |
-						 MACRO_TILE_ASPECT(ADDR_SURF_MACRO_ASPECT_4) |
-						 NUM_BANKS(ADDR_SURF_16_BANK));
-				break;
-			case 4:
-				gb_tile_moden = (BANK_WIDTH(ADDR_SURF_BANK_WIDTH_1) |
-						 BANK_HEIGHT(ADDR_SURF_BANK_HEIGHT_1) |
-						 MACRO_TILE_ASPECT(ADDR_SURF_MACRO_ASPECT_4) |
-						 NUM_BANKS(ADDR_SURF_16_BANK));
-				break;
-			case 5:
-				gb_tile_moden = (BANK_WIDTH(ADDR_SURF_BANK_WIDTH_1) |
-						 BANK_HEIGHT(ADDR_SURF_BANK_HEIGHT_1) |
-						 MACRO_TILE_ASPECT(ADDR_SURF_MACRO_ASPECT_4) |
-						 NUM_BANKS(ADDR_SURF_16_BANK));
-				break;
-			case 6:
-				gb_tile_moden = (BANK_WIDTH(ADDR_SURF_BANK_WIDTH_1) |
-						 BANK_HEIGHT(ADDR_SURF_BANK_HEIGHT_1) |
-						 MACRO_TILE_ASPECT(ADDR_SURF_MACRO_ASPECT_2) |
-						 NUM_BANKS(ADDR_SURF_8_BANK));
-				break;
-			case 8:
-				gb_tile_moden = (BANK_WIDTH(ADDR_SURF_BANK_WIDTH_4) |
-						 BANK_HEIGHT(ADDR_SURF_BANK_HEIGHT_8) |
-						 MACRO_TILE_ASPECT(ADDR_SURF_MACRO_ASPECT_4) |
-						 NUM_BANKS(ADDR_SURF_16_BANK));
-				break;
-			case 9:
-				gb_tile_moden = (BANK_WIDTH(ADDR_SURF_BANK_WIDTH_4) |
-						 BANK_HEIGHT(ADDR_SURF_BANK_HEIGHT_4) |
-						 MACRO_TILE_ASPECT(ADDR_SURF_MACRO_ASPECT_4) |
-						 NUM_BANKS(ADDR_SURF_16_BANK));
-				break;
-			case 10:
-				gb_tile_moden = (BANK_WIDTH(ADDR_SURF_BANK_WIDTH_2) |
-						 BANK_HEIGHT(ADDR_SURF_BANK_HEIGHT_4) |
-						 MACRO_TILE_ASPECT(ADDR_SURF_MACRO_ASPECT_4) |
-						 NUM_BANKS(ADDR_SURF_16_BANK));
-				break;
-			case 11:
-				gb_tile_moden = (BANK_WIDTH(ADDR_SURF_BANK_WIDTH_2) |
-						 BANK_HEIGHT(ADDR_SURF_BANK_HEIGHT_2) |
-						 MACRO_TILE_ASPECT(ADDR_SURF_MACRO_ASPECT_4) |
-						 NUM_BANKS(ADDR_SURF_16_BANK));
-				break;
-			case 12:
-				gb_tile_moden = (BANK_WIDTH(ADDR_SURF_BANK_WIDTH_1) |
-						 BANK_HEIGHT(ADDR_SURF_BANK_HEIGHT_2) |
-						 MACRO_TILE_ASPECT(ADDR_SURF_MACRO_ASPECT_4) |
-						 NUM_BANKS(ADDR_SURF_16_BANK));
-				break;
-			case 13:
-				gb_tile_moden = (BANK_WIDTH(ADDR_SURF_BANK_WIDTH_1) |
-						 BANK_HEIGHT(ADDR_SURF_BANK_HEIGHT_1) |
-						 MACRO_TILE_ASPECT(ADDR_SURF_MACRO_ASPECT_4) |
-						 NUM_BANKS(ADDR_SURF_16_BANK));
-				break;
-			case 14:
-				gb_tile_moden = (BANK_WIDTH(ADDR_SURF_BANK_WIDTH_1) |
-						 BANK_HEIGHT(ADDR_SURF_BANK_HEIGHT_1) |
-						 MACRO_TILE_ASPECT(ADDR_SURF_MACRO_ASPECT_2) |
-						 NUM_BANKS(ADDR_SURF_8_BANK));
-				break;
-			default:
-				gb_tile_moden = 0;
-				break;
-			}
-			rdev->config.cik.macrotile_mode_array[reg_offset] = gb_tile_moden;
-			WREG32(GB_MACROTILE_MODE0 + (reg_offset * 4), gb_tile_moden);
+		tile[0] = (ARRAY_MODE(ARRAY_2D_TILED_THIN1) |
+			   MICRO_TILE_MODE_NEW(ADDR_SURF_DEPTH_MICRO_TILING) |
+			   PIPE_CONFIG(ADDR_SURF_P4_8x16) |
+			   TILE_SPLIT(ADDR_SURF_TILE_SPLIT_64B));
+		tile[1] = (ARRAY_MODE(ARRAY_2D_TILED_THIN1) |
+			   MICRO_TILE_MODE_NEW(ADDR_SURF_DEPTH_MICRO_TILING) |
+			   PIPE_CONFIG(ADDR_SURF_P4_8x16) |
+			   TILE_SPLIT(ADDR_SURF_TILE_SPLIT_128B));
+		tile[2] = (ARRAY_MODE(ARRAY_2D_TILED_THIN1) |
+			   MICRO_TILE_MODE_NEW(ADDR_SURF_DEPTH_MICRO_TILING) |
+			   PIPE_CONFIG(ADDR_SURF_P4_8x16) |
+			   TILE_SPLIT(ADDR_SURF_TILE_SPLIT_256B));
+		tile[3] = (ARRAY_MODE(ARRAY_2D_TILED_THIN1) |
+			   MICRO_TILE_MODE_NEW(ADDR_SURF_DEPTH_MICRO_TILING) |
+			   PIPE_CONFIG(ADDR_SURF_P4_8x16) |
+			   TILE_SPLIT(ADDR_SURF_TILE_SPLIT_512B));
+		tile[4] = (ARRAY_MODE(ARRAY_2D_TILED_THIN1) |
+			   MICRO_TILE_MODE_NEW(ADDR_SURF_DEPTH_MICRO_TILING) |
+			   PIPE_CONFIG(ADDR_SURF_P4_8x16) |
+			   TILE_SPLIT(split_equal_to_row_size));
+		tile[5] = (ARRAY_MODE(ARRAY_1D_TILED_THIN1) |
+			   PIPE_CONFIG(ADDR_SURF_P4_8x16) |
+			   MICRO_TILE_MODE_NEW(ADDR_SURF_DEPTH_MICRO_TILING));
+		tile[6] = (ARRAY_MODE(ARRAY_PRT_2D_TILED_THIN1) |
+			   MICRO_TILE_MODE_NEW(ADDR_SURF_DEPTH_MICRO_TILING) |
+			   PIPE_CONFIG(ADDR_SURF_P4_8x16) |
+			   TILE_SPLIT(ADDR_SURF_TILE_SPLIT_256B));
+		tile[7] = (ARRAY_MODE(ARRAY_PRT_2D_TILED_THIN1) |
+			   MICRO_TILE_MODE_NEW(ADDR_SURF_DEPTH_MICRO_TILING) |
+			   PIPE_CONFIG(ADDR_SURF_P4_8x16) |
+			   TILE_SPLIT(split_equal_to_row_size));
+		tile[8] = (ARRAY_MODE(ARRAY_LINEAR_ALIGNED) |
+			   PIPE_CONFIG(ADDR_SURF_P4_8x16));
+		tile[9] = (ARRAY_MODE(ARRAY_1D_TILED_THIN1) |
+			   PIPE_CONFIG(ADDR_SURF_P4_8x16) |
+			   MICRO_TILE_MODE_NEW(ADDR_SURF_DISPLAY_MICRO_TILING));
+		tile[10] = (ARRAY_MODE(ARRAY_2D_TILED_THIN1) |
+			    MICRO_TILE_MODE_NEW(ADDR_SURF_DISPLAY_MICRO_TILING) |
+			    PIPE_CONFIG(ADDR_SURF_P4_8x16) |
+			    SAMPLE_SPLIT(ADDR_SURF_SAMPLE_SPLIT_2));
+		tile[11] = (ARRAY_MODE(ARRAY_PRT_TILED_THIN1) |
+			    MICRO_TILE_MODE_NEW(ADDR_SURF_DISPLAY_MICRO_TILING) |
+			    PIPE_CONFIG(ADDR_SURF_P4_8x16) |
+			    SAMPLE_SPLIT(ADDR_SURF_SAMPLE_SPLIT_2));
+		tile[12] = (ARRAY_MODE(ARRAY_PRT_2D_TILED_THIN1) |
+			    MICRO_TILE_MODE_NEW(ADDR_SURF_DISPLAY_MICRO_TILING) |
+			    PIPE_CONFIG(ADDR_SURF_P4_8x16) |
+			    SAMPLE_SPLIT(ADDR_SURF_SAMPLE_SPLIT_2));
+		tile[13] = (ARRAY_MODE(ARRAY_1D_TILED_THIN1) |
+			    PIPE_CONFIG(ADDR_SURF_P4_8x16) |
+			    MICRO_TILE_MODE_NEW(ADDR_SURF_THIN_MICRO_TILING));
+		tile[14] = (ARRAY_MODE(ARRAY_2D_TILED_THIN1) |
+			    MICRO_TILE_MODE_NEW(ADDR_SURF_THIN_MICRO_TILING) |
+			    PIPE_CONFIG(ADDR_SURF_P4_8x16) |
+			    SAMPLE_SPLIT(ADDR_SURF_SAMPLE_SPLIT_2));
+		tile[16] = (ARRAY_MODE(ARRAY_PRT_TILED_THIN1) |
+			    MICRO_TILE_MODE_NEW(ADDR_SURF_THIN_MICRO_TILING) |
+			    PIPE_CONFIG(ADDR_SURF_P4_8x16) |
+			    SAMPLE_SPLIT(ADDR_SURF_SAMPLE_SPLIT_2));
+		tile[17] = (ARRAY_MODE(ARRAY_PRT_2D_TILED_THIN1) |
+			    MICRO_TILE_MODE_NEW(ADDR_SURF_THIN_MICRO_TILING) |
+			    PIPE_CONFIG(ADDR_SURF_P4_8x16) |
+			    SAMPLE_SPLIT(ADDR_SURF_SAMPLE_SPLIT_2));
+		tile[27] = (ARRAY_MODE(ARRAY_1D_TILED_THIN1) |
+			    PIPE_CONFIG(ADDR_SURF_P4_8x16) |
+			    MICRO_TILE_MODE_NEW(ADDR_SURF_ROTATED_MICRO_TILING));
+		tile[28] = (ARRAY_MODE(ARRAY_PRT_2D_TILED_THIN1) |
+			    MICRO_TILE_MODE_NEW(ADDR_SURF_ROTATED_MICRO_TILING) |
+			    PIPE_CONFIG(ADDR_SURF_P4_8x16) |
+			    SAMPLE_SPLIT(ADDR_SURF_SAMPLE_SPLIT_2));
+		tile[29] = (ARRAY_MODE(ARRAY_PRT_TILED_THIN1) |
+			    MICRO_TILE_MODE_NEW(ADDR_SURF_ROTATED_MICRO_TILING) |
+			    PIPE_CONFIG(ADDR_SURF_P4_8x16) |
+			    SAMPLE_SPLIT(ADDR_SURF_SAMPLE_SPLIT_2));
+		tile[30] = (ARRAY_MODE(ARRAY_PRT_2D_TILED_THIN1) |
+			    MICRO_TILE_MODE_NEW(ADDR_SURF_ROTATED_MICRO_TILING) |
+			    PIPE_CONFIG(ADDR_SURF_P4_8x16) |
+			    SAMPLE_SPLIT(ADDR_SURF_SAMPLE_SPLIT_2));
 		}
-	} else
+
+		macrotile[0] = (BANK_WIDTH(ADDR_SURF_BANK_WIDTH_1) |
+				BANK_HEIGHT(ADDR_SURF_BANK_HEIGHT_4) |
+				MACRO_TILE_ASPECT(ADDR_SURF_MACRO_ASPECT_4) |
+				NUM_BANKS(ADDR_SURF_16_BANK));
+		macrotile[1] = (BANK_WIDTH(ADDR_SURF_BANK_WIDTH_1) |
+				BANK_HEIGHT(ADDR_SURF_BANK_HEIGHT_2) |
+				MACRO_TILE_ASPECT(ADDR_SURF_MACRO_ASPECT_4) |
+				NUM_BANKS(ADDR_SURF_16_BANK));
+		macrotile[2] = (BANK_WIDTH(ADDR_SURF_BANK_WIDTH_1) |
+				BANK_HEIGHT(ADDR_SURF_BANK_HEIGHT_1) |
+				MACRO_TILE_ASPECT(ADDR_SURF_MACRO_ASPECT_2) |
+				NUM_BANKS(ADDR_SURF_16_BANK));
+		macrotile[3] = (BANK_WIDTH(ADDR_SURF_BANK_WIDTH_1) |
+				BANK_HEIGHT(ADDR_SURF_BANK_HEIGHT_1) |
+				MACRO_TILE_ASPECT(ADDR_SURF_MACRO_ASPECT_2) |
+				NUM_BANKS(ADDR_SURF_16_BANK));
+		macrotile[4] = (BANK_WIDTH(ADDR_SURF_BANK_WIDTH_1) |
+				BANK_HEIGHT(ADDR_SURF_BANK_HEIGHT_1) |
+				MACRO_TILE_ASPECT(ADDR_SURF_MACRO_ASPECT_2) |
+				NUM_BANKS(ADDR_SURF_16_BANK));
+		macrotile[5] = (BANK_WIDTH(ADDR_SURF_BANK_WIDTH_1) |
+				BANK_HEIGHT(ADDR_SURF_BANK_HEIGHT_1) |
+				MACRO_TILE_ASPECT(ADDR_SURF_MACRO_ASPECT_2) |
+				NUM_BANKS(ADDR_SURF_8_BANK));
+		macrotile[6] = (BANK_WIDTH(ADDR_SURF_BANK_WIDTH_1) |
+				BANK_HEIGHT(ADDR_SURF_BANK_HEIGHT_1) |
+				MACRO_TILE_ASPECT(ADDR_SURF_MACRO_ASPECT_1) |
+				NUM_BANKS(ADDR_SURF_4_BANK));
+		macrotile[8] = (BANK_WIDTH(ADDR_SURF_BANK_WIDTH_2) |
+				BANK_HEIGHT(ADDR_SURF_BANK_HEIGHT_8) |
+				MACRO_TILE_ASPECT(ADDR_SURF_MACRO_ASPECT_4) |
+				NUM_BANKS(ADDR_SURF_16_BANK));
+		macrotile[9] = (BANK_WIDTH(ADDR_SURF_BANK_WIDTH_2) |
+				BANK_HEIGHT(ADDR_SURF_BANK_HEIGHT_4) |
+				MACRO_TILE_ASPECT(ADDR_SURF_MACRO_ASPECT_4) |
+				NUM_BANKS(ADDR_SURF_16_BANK));
+		macrotile[10] = (BANK_WIDTH(ADDR_SURF_BANK_WIDTH_1) |
+				BANK_HEIGHT(ADDR_SURF_BANK_HEIGHT_4) |
+				MACRO_TILE_ASPECT(ADDR_SURF_MACRO_ASPECT_4) |
+				NUM_BANKS(ADDR_SURF_16_BANK));
+		macrotile[11] = (BANK_WIDTH(ADDR_SURF_BANK_WIDTH_1) |
+				BANK_HEIGHT(ADDR_SURF_BANK_HEIGHT_2) |
+				MACRO_TILE_ASPECT(ADDR_SURF_MACRO_ASPECT_4) |
+				NUM_BANKS(ADDR_SURF_16_BANK));
+		macrotile[12] = (BANK_WIDTH(ADDR_SURF_BANK_WIDTH_1) |
+				BANK_HEIGHT(ADDR_SURF_BANK_HEIGHT_1) |
+				MACRO_TILE_ASPECT(ADDR_SURF_MACRO_ASPECT_2) |
+				NUM_BANKS(ADDR_SURF_16_BANK));
+		macrotile[13] = (BANK_WIDTH(ADDR_SURF_BANK_WIDTH_1) |
+				BANK_HEIGHT(ADDR_SURF_BANK_HEIGHT_1) |
+				MACRO_TILE_ASPECT(ADDR_SURF_MACRO_ASPECT_2) |
+				NUM_BANKS(ADDR_SURF_8_BANK));
+		macrotile[14] = (BANK_WIDTH(ADDR_SURF_BANK_WIDTH_1) |
+				BANK_HEIGHT(ADDR_SURF_BANK_HEIGHT_1) |
+				MACRO_TILE_ASPECT(ADDR_SURF_MACRO_ASPECT_1) |
+				NUM_BANKS(ADDR_SURF_4_BANK));
+
+		for (reg_offset = 0; reg_offset < num_tile_mode_states; reg_offset++)
+			WREG32(GB_TILE_MODE0 + (reg_offset * 4), tile[reg_offset]);
+		for (reg_offset = 0; reg_offset < num_secondary_tile_mode_states; reg_offset++)
+			WREG32(GB_MACROTILE_MODE0 + (reg_offset * 4), macrotile[reg_offset]);
+		break;
+
+	case 2:
+		tile[0] = (ARRAY_MODE(ARRAY_2D_TILED_THIN1) |
+			   MICRO_TILE_MODE_NEW(ADDR_SURF_DEPTH_MICRO_TILING) |
+			   PIPE_CONFIG(ADDR_SURF_P2) |
+			   TILE_SPLIT(ADDR_SURF_TILE_SPLIT_64B));
+		tile[1] = (ARRAY_MODE(ARRAY_2D_TILED_THIN1) |
+			   MICRO_TILE_MODE_NEW(ADDR_SURF_DEPTH_MICRO_TILING) |
+			   PIPE_CONFIG(ADDR_SURF_P2) |
+			   TILE_SPLIT(ADDR_SURF_TILE_SPLIT_128B));
+		tile[2] = (ARRAY_MODE(ARRAY_2D_TILED_THIN1) |
+			   MICRO_TILE_MODE_NEW(ADDR_SURF_DEPTH_MICRO_TILING) |
+			   PIPE_CONFIG(ADDR_SURF_P2) |
+			   TILE_SPLIT(ADDR_SURF_TILE_SPLIT_256B));
+		tile[3] = (ARRAY_MODE(ARRAY_2D_TILED_THIN1) |
+			   MICRO_TILE_MODE_NEW(ADDR_SURF_DEPTH_MICRO_TILING) |
+			   PIPE_CONFIG(ADDR_SURF_P2) |
+			   TILE_SPLIT(ADDR_SURF_TILE_SPLIT_512B));
+		tile[4] = (ARRAY_MODE(ARRAY_2D_TILED_THIN1) |
+			   MICRO_TILE_MODE_NEW(ADDR_SURF_DEPTH_MICRO_TILING) |
+			   PIPE_CONFIG(ADDR_SURF_P2) |
+			   TILE_SPLIT(split_equal_to_row_size));
+		tile[5] = (ARRAY_MODE(ARRAY_1D_TILED_THIN1) |
+			   PIPE_CONFIG(ADDR_SURF_P2) |
+			   MICRO_TILE_MODE_NEW(ADDR_SURF_DEPTH_MICRO_TILING));
+		tile[6] = (ARRAY_MODE(ARRAY_PRT_2D_TILED_THIN1) |
+			   MICRO_TILE_MODE_NEW(ADDR_SURF_DEPTH_MICRO_TILING) |
+			   PIPE_CONFIG(ADDR_SURF_P2) |
+			   TILE_SPLIT(ADDR_SURF_TILE_SPLIT_256B));
+		tile[7] = (ARRAY_MODE(ARRAY_PRT_2D_TILED_THIN1) |
+			   MICRO_TILE_MODE_NEW(ADDR_SURF_DEPTH_MICRO_TILING) |
+			   PIPE_CONFIG(ADDR_SURF_P2) |
+			   TILE_SPLIT(split_equal_to_row_size));
+		tile[8] = ARRAY_MODE(ARRAY_LINEAR_ALIGNED) |
+			   PIPE_CONFIG(ADDR_SURF_P2);
+		tile[9] = (ARRAY_MODE(ARRAY_1D_TILED_THIN1) |
+			   MICRO_TILE_MODE_NEW(ADDR_SURF_DISPLAY_MICRO_TILING) |
+			   PIPE_CONFIG(ADDR_SURF_P2));
+		tile[10] = (ARRAY_MODE(ARRAY_2D_TILED_THIN1) |
+			    MICRO_TILE_MODE_NEW(ADDR_SURF_DISPLAY_MICRO_TILING) |
+			    PIPE_CONFIG(ADDR_SURF_P2) |
+			    SAMPLE_SPLIT(ADDR_SURF_SAMPLE_SPLIT_2));
+		tile[11] = (ARRAY_MODE(ARRAY_PRT_TILED_THIN1) |
+			    MICRO_TILE_MODE_NEW(ADDR_SURF_DISPLAY_MICRO_TILING) |
+			    PIPE_CONFIG(ADDR_SURF_P2) |
+			    SAMPLE_SPLIT(ADDR_SURF_SAMPLE_SPLIT_2));
+		tile[12] = (ARRAY_MODE(ARRAY_PRT_2D_TILED_THIN1) |
+			    MICRO_TILE_MODE_NEW(ADDR_SURF_DISPLAY_MICRO_TILING) |
+			    PIPE_CONFIG(ADDR_SURF_P2) |
+			    SAMPLE_SPLIT(ADDR_SURF_SAMPLE_SPLIT_2));
+		tile[13] = (ARRAY_MODE(ARRAY_1D_TILED_THIN1) |
+			    PIPE_CONFIG(ADDR_SURF_P2) |
+			    MICRO_TILE_MODE_NEW(ADDR_SURF_THIN_MICRO_TILING));
+		tile[14] = (ARRAY_MODE(ARRAY_2D_TILED_THIN1) |
+			    MICRO_TILE_MODE_NEW(ADDR_SURF_THIN_MICRO_TILING) |
+			    PIPE_CONFIG(ADDR_SURF_P2) |
+			    SAMPLE_SPLIT(ADDR_SURF_SAMPLE_SPLIT_2));
+		tile[16] = (ARRAY_MODE(ARRAY_PRT_TILED_THIN1) |
+			    MICRO_TILE_MODE_NEW(ADDR_SURF_THIN_MICRO_TILING) |
+			    PIPE_CONFIG(ADDR_SURF_P2) |
+			    SAMPLE_SPLIT(ADDR_SURF_SAMPLE_SPLIT_2));
+		tile[17] = (ARRAY_MODE(ARRAY_PRT_2D_TILED_THIN1) |
+			    MICRO_TILE_MODE_NEW(ADDR_SURF_THIN_MICRO_TILING) |
+			    PIPE_CONFIG(ADDR_SURF_P2) |
+			    SAMPLE_SPLIT(ADDR_SURF_SAMPLE_SPLIT_2));
+		tile[27] = (ARRAY_MODE(ARRAY_1D_TILED_THIN1) |
+			    MICRO_TILE_MODE_NEW(ADDR_SURF_ROTATED_MICRO_TILING) |
+			    PIPE_CONFIG(ADDR_SURF_P2));
+		tile[28] = (ARRAY_MODE(ARRAY_PRT_2D_TILED_THIN1) |
+			    MICRO_TILE_MODE_NEW(ADDR_SURF_ROTATED_MICRO_TILING) |
+			    PIPE_CONFIG(ADDR_SURF_P2) |
+			    SAMPLE_SPLIT(ADDR_SURF_SAMPLE_SPLIT_2));
+		tile[29] = (ARRAY_MODE(ARRAY_PRT_TILED_THIN1) |
+			    MICRO_TILE_MODE_NEW(ADDR_SURF_ROTATED_MICRO_TILING) |
+			    PIPE_CONFIG(ADDR_SURF_P2) |
+			    SAMPLE_SPLIT(ADDR_SURF_SAMPLE_SPLIT_2));
+		tile[30] = (ARRAY_MODE(ARRAY_PRT_2D_TILED_THIN1) |
+			    MICRO_TILE_MODE_NEW(ADDR_SURF_ROTATED_MICRO_TILING) |
+			    PIPE_CONFIG(ADDR_SURF_P2) |
+			    SAMPLE_SPLIT(ADDR_SURF_SAMPLE_SPLIT_2));
+
+		macrotile[0] = (BANK_WIDTH(ADDR_SURF_BANK_WIDTH_2) |
+				BANK_HEIGHT(ADDR_SURF_BANK_HEIGHT_4) |
+				MACRO_TILE_ASPECT(ADDR_SURF_MACRO_ASPECT_4) |
+				NUM_BANKS(ADDR_SURF_16_BANK));
+		macrotile[1] = (BANK_WIDTH(ADDR_SURF_BANK_WIDTH_2) |
+				BANK_HEIGHT(ADDR_SURF_BANK_HEIGHT_2) |
+				MACRO_TILE_ASPECT(ADDR_SURF_MACRO_ASPECT_4) |
+				NUM_BANKS(ADDR_SURF_16_BANK));
+		macrotile[2] = (BANK_WIDTH(ADDR_SURF_BANK_WIDTH_1) |
+				BANK_HEIGHT(ADDR_SURF_BANK_HEIGHT_2) |
+				MACRO_TILE_ASPECT(ADDR_SURF_MACRO_ASPECT_4) |
+				NUM_BANKS(ADDR_SURF_16_BANK));
+		macrotile[3] = (BANK_WIDTH(ADDR_SURF_BANK_WIDTH_1) |
+				BANK_HEIGHT(ADDR_SURF_BANK_HEIGHT_1) |
+				MACRO_TILE_ASPECT(ADDR_SURF_MACRO_ASPECT_4) |
+				NUM_BANKS(ADDR_SURF_16_BANK));
+		macrotile[4] = (BANK_WIDTH(ADDR_SURF_BANK_WIDTH_1) |
+				BANK_HEIGHT(ADDR_SURF_BANK_HEIGHT_1) |
+				MACRO_TILE_ASPECT(ADDR_SURF_MACRO_ASPECT_4) |
+				NUM_BANKS(ADDR_SURF_16_BANK));
+		macrotile[5] = (BANK_WIDTH(ADDR_SURF_BANK_WIDTH_1) |
+				BANK_HEIGHT(ADDR_SURF_BANK_HEIGHT_1) |
+				MACRO_TILE_ASPECT(ADDR_SURF_MACRO_ASPECT_4) |
+				NUM_BANKS(ADDR_SURF_16_BANK));
+		macrotile[6] = (BANK_WIDTH(ADDR_SURF_BANK_WIDTH_1) |
+				BANK_HEIGHT(ADDR_SURF_BANK_HEIGHT_1) |
+				MACRO_TILE_ASPECT(ADDR_SURF_MACRO_ASPECT_2) |
+				NUM_BANKS(ADDR_SURF_8_BANK));
+		macrotile[8] = (BANK_WIDTH(ADDR_SURF_BANK_WIDTH_4) |
+				BANK_HEIGHT(ADDR_SURF_BANK_HEIGHT_8) |
+				MACRO_TILE_ASPECT(ADDR_SURF_MACRO_ASPECT_4) |
+				NUM_BANKS(ADDR_SURF_16_BANK));
+		macrotile[9] = (BANK_WIDTH(ADDR_SURF_BANK_WIDTH_4) |
+				BANK_HEIGHT(ADDR_SURF_BANK_HEIGHT_4) |
+				MACRO_TILE_ASPECT(ADDR_SURF_MACRO_ASPECT_4) |
+				NUM_BANKS(ADDR_SURF_16_BANK));
+		macrotile[10] = (BANK_WIDTH(ADDR_SURF_BANK_WIDTH_2) |
+				BANK_HEIGHT(ADDR_SURF_BANK_HEIGHT_4) |
+				MACRO_TILE_ASPECT(ADDR_SURF_MACRO_ASPECT_4) |
+				NUM_BANKS(ADDR_SURF_16_BANK));
+		macrotile[11] = (BANK_WIDTH(ADDR_SURF_BANK_WIDTH_2) |
+				BANK_HEIGHT(ADDR_SURF_BANK_HEIGHT_2) |
+				MACRO_TILE_ASPECT(ADDR_SURF_MACRO_ASPECT_4) |
+				NUM_BANKS(ADDR_SURF_16_BANK));
+		macrotile[12] = (BANK_WIDTH(ADDR_SURF_BANK_WIDTH_1) |
+				BANK_HEIGHT(ADDR_SURF_BANK_HEIGHT_2) |
+				MACRO_TILE_ASPECT(ADDR_SURF_MACRO_ASPECT_4) |
+				NUM_BANKS(ADDR_SURF_16_BANK));
+		macrotile[13] = (BANK_WIDTH(ADDR_SURF_BANK_WIDTH_1) |
+				BANK_HEIGHT(ADDR_SURF_BANK_HEIGHT_1) |
+				MACRO_TILE_ASPECT(ADDR_SURF_MACRO_ASPECT_4) |
+				NUM_BANKS(ADDR_SURF_16_BANK));
+		macrotile[14] = (BANK_WIDTH(ADDR_SURF_BANK_WIDTH_1) |
+				BANK_HEIGHT(ADDR_SURF_BANK_HEIGHT_1) |
+				MACRO_TILE_ASPECT(ADDR_SURF_MACRO_ASPECT_2) |
+				NUM_BANKS(ADDR_SURF_8_BANK));
+
+		for (reg_offset = 0; reg_offset < num_tile_mode_states; reg_offset++)
+			WREG32(GB_TILE_MODE0 + (reg_offset * 4), tile[reg_offset]);
+		for (reg_offset = 0; reg_offset < num_secondary_tile_mode_states; reg_offset++)
+			WREG32(GB_MACROTILE_MODE0 + (reg_offset * 4), macrotile[reg_offset]);
+		break;
+
+	default:
 		DRM_ERROR("unknown num pipe config: 0x%x\n", num_pipe_configs);
+	}
 }
 
 /**
@@ -9709,13 +9350,13 @@ uint64_t cik_get_gpu_clock_counter(struct radeon_device *rdev)
 	mutex_lock(&rdev->gpu_clock_mutex);
 	WREG32(RLC_CAPTURE_GPU_CLOCK_COUNT, 1);
 	clock = (uint64_t)RREG32(RLC_GPU_CLOCK_COUNT_LSB) |
-	        ((uint64_t)RREG32(RLC_GPU_CLOCK_COUNT_MSB) << 32ULL);
+		((uint64_t)RREG32(RLC_GPU_CLOCK_COUNT_MSB) << 32ULL);
 	mutex_unlock(&rdev->gpu_clock_mutex);
 	return clock;
 }
 
 static int cik_set_uvd_clock(struct radeon_device *rdev, u32 clock,
-                              u32 cntl_reg, u32 status_reg)
+			     u32 cntl_reg, u32 status_reg)
 {
 	int r, i;
 	struct atom_clock_dividers dividers;
diff --git a/drivers/gpu/drm/radeon/cypress_dpm.c b/drivers/gpu/drm/radeon/cypress_dpm.c
index ca058589ddef..a4edd0702718 100644
--- a/drivers/gpu/drm/radeon/cypress_dpm.c
+++ b/drivers/gpu/drm/radeon/cypress_dpm.c
@@ -1620,14 +1620,14 @@ static int cypress_init_smc_table(struct radeon_device *rdev,
 	cypress_populate_smc_voltage_tables(rdev, table);
 
 	switch (rdev->pm.int_thermal_type) {
-        case THERMAL_TYPE_EVERGREEN:
-        case THERMAL_TYPE_EMC2103_WITH_INTERNAL:
+	case THERMAL_TYPE_EVERGREEN:
+	case THERMAL_TYPE_EMC2103_WITH_INTERNAL:
 		table->thermalProtectType = PPSMC_THERMAL_PROTECT_TYPE_INTERNAL;
 		break;
-        case THERMAL_TYPE_NONE:
+	case THERMAL_TYPE_NONE:
 		table->thermalProtectType = PPSMC_THERMAL_PROTECT_TYPE_NONE;
 		break;
-        default:
+	default:
 		table->thermalProtectType = PPSMC_THERMAL_PROTECT_TYPE_EXTERNAL;
 		break;
 	}
diff --git a/drivers/gpu/drm/radeon/evergreen.c b/drivers/gpu/drm/radeon/evergreen.c
index 2ad462896896..76c4bdf21b20 100644
--- a/drivers/gpu/drm/radeon/evergreen.c
+++ b/drivers/gpu/drm/radeon/evergreen.c
@@ -1140,7 +1140,7 @@ static int sumo_set_uvd_clock(struct radeon_device *rdev, u32 clock,
 	int r, i;
 	struct atom_clock_dividers dividers;
 
-        r = radeon_atom_get_clock_dividers(rdev, COMPUTE_ENGINE_PLL_PARAM,
+	r = radeon_atom_get_clock_dividers(rdev, COMPUTE_ENGINE_PLL_PARAM,
 					   clock, false, &dividers);
 	if (r)
 		return r;
diff --git a/drivers/gpu/drm/radeon/evergreen_cs.c b/drivers/gpu/drm/radeon/evergreen_cs.c
index 46f87d4aaf31..9e93205eb9e4 100644
--- a/drivers/gpu/drm/radeon/evergreen_cs.c
+++ b/drivers/gpu/drm/radeon/evergreen_cs.c
@@ -1816,8 +1816,8 @@ static int evergreen_packet3_check(struct radeon_cs_parser *p,
 		}
 
 		offset = reloc->gpu_offset +
-		         (idx_value & 0xfffffff0) +
-		         ((u64)(tmp & 0xff) << 32);
+			 (idx_value & 0xfffffff0) +
+			 ((u64)(tmp & 0xff) << 32);
 
 		ib[idx + 0] = offset;
 		ib[idx + 1] = (tmp & 0xffffff00) | (upper_32_bits(offset) & 0xff);
@@ -1862,8 +1862,8 @@ static int evergreen_packet3_check(struct radeon_cs_parser *p,
 		}
 
 		offset = reloc->gpu_offset +
-		         idx_value +
-		         ((u64)(radeon_get_ib_value(p, idx+1) & 0xff) << 32);
+			 idx_value +
+			 ((u64)(radeon_get_ib_value(p, idx+1) & 0xff) << 32);
 
 		ib[idx+0] = offset;
 		ib[idx+1] = upper_32_bits(offset) & 0xff;
@@ -1897,8 +1897,8 @@ static int evergreen_packet3_check(struct radeon_cs_parser *p,
 		}
 
 		offset = reloc->gpu_offset +
-		         idx_value +
-		         ((u64)(radeon_get_ib_value(p, idx+1) & 0xff) << 32);
+			 idx_value +
+			 ((u64)(radeon_get_ib_value(p, idx+1) & 0xff) << 32);
 
 		ib[idx+0] = offset;
 		ib[idx+1] = upper_32_bits(offset) & 0xff;
@@ -1925,8 +1925,8 @@ static int evergreen_packet3_check(struct radeon_cs_parser *p,
 		}
 
 		offset = reloc->gpu_offset +
-		         radeon_get_ib_value(p, idx+1) +
-		         ((u64)(radeon_get_ib_value(p, idx+2) & 0xff) << 32);
+			 radeon_get_ib_value(p, idx+1) +
+			 ((u64)(radeon_get_ib_value(p, idx+2) & 0xff) << 32);
 
 		ib[idx+1] = offset;
 		ib[idx+2] = upper_32_bits(offset) & 0xff;
@@ -2098,8 +2098,8 @@ static int evergreen_packet3_check(struct radeon_cs_parser *p,
 			}
 
 			offset = reloc->gpu_offset +
-			         (radeon_get_ib_value(p, idx+1) & 0xfffffffc) +
-			         ((u64)(radeon_get_ib_value(p, idx+2) & 0xff) << 32);
+				 (radeon_get_ib_value(p, idx+1) & 0xfffffffc) +
+				 ((u64)(radeon_get_ib_value(p, idx+2) & 0xff) << 32);
 
 			ib[idx+1] = (ib[idx+1] & 0x3) | (offset & 0xfffffffc);
 			ib[idx+2] = upper_32_bits(offset) & 0xff;
@@ -2239,8 +2239,8 @@ static int evergreen_packet3_check(struct radeon_cs_parser *p,
 				return -EINVAL;
 			}
 			offset = reloc->gpu_offset +
-			         (radeon_get_ib_value(p, idx+1) & 0xfffffff8) +
-			         ((u64)(radeon_get_ib_value(p, idx+2) & 0xff) << 32);
+				 (radeon_get_ib_value(p, idx+1) & 0xfffffff8) +
+				 ((u64)(radeon_get_ib_value(p, idx+2) & 0xff) << 32);
 
 			ib[idx+1] = offset & 0xfffffff8;
 			ib[idx+2] = upper_32_bits(offset) & 0xff;
@@ -2261,8 +2261,8 @@ static int evergreen_packet3_check(struct radeon_cs_parser *p,
 		}
 
 		offset = reloc->gpu_offset +
-		         (radeon_get_ib_value(p, idx+1) & 0xfffffffc) +
-		         ((u64)(radeon_get_ib_value(p, idx+2) & 0xff) << 32);
+			 (radeon_get_ib_value(p, idx+1) & 0xfffffffc) +
+			 ((u64)(radeon_get_ib_value(p, idx+2) & 0xff) << 32);
 
 		ib[idx+1] = offset & 0xfffffffc;
 		ib[idx+2] = (ib[idx+2] & 0xffffff00) | (upper_32_bits(offset) & 0xff);
@@ -2283,8 +2283,8 @@ static int evergreen_packet3_check(struct radeon_cs_parser *p,
 		}
 
 		offset = reloc->gpu_offset +
-		         (radeon_get_ib_value(p, idx+1) & 0xfffffffc) +
-		         ((u64)(radeon_get_ib_value(p, idx+2) & 0xff) << 32);
+			 (radeon_get_ib_value(p, idx+1) & 0xfffffffc) +
+			 ((u64)(radeon_get_ib_value(p, idx+2) & 0xff) << 32);
 
 		ib[idx+1] = offset & 0xfffffffc;
 		ib[idx+2] = (ib[idx+2] & 0xffffff00) | (upper_32_bits(offset) & 0xff);
diff --git a/drivers/gpu/drm/radeon/evergreen_hdmi.c b/drivers/gpu/drm/radeon/evergreen_hdmi.c
index 3cf04a2f44bb..f766c967a284 100644
--- a/drivers/gpu/drm/radeon/evergreen_hdmi.c
+++ b/drivers/gpu/drm/radeon/evergreen_hdmi.c
@@ -206,7 +206,7 @@ void evergreen_hdmi_write_sad_regs(struct drm_encoder *encoder,
  * build a AVI Info Frame
  */
 void evergreen_set_avi_packet(struct radeon_device *rdev, u32 offset,
-    unsigned char *buffer, size_t size)
+			      unsigned char *buffer, size_t size)
 {
 	uint8_t *frame = buffer + 3;
 
diff --git a/drivers/gpu/drm/radeon/kv_dpm.c b/drivers/gpu/drm/radeon/kv_dpm.c
index 2d71da448487..d0240743a17c 100644
--- a/drivers/gpu/drm/radeon/kv_dpm.c
+++ b/drivers/gpu/drm/radeon/kv_dpm.c
@@ -2640,7 +2640,7 @@ static int kv_parse_power_table(struct radeon_device *rdev)
 	struct _NonClockInfoArray *non_clock_info_array;
 	union power_info *power_info;
 	int index = GetIndexIntoMasterTable(DATA, PowerPlayInfo);
-        u16 data_offset;
+	u16 data_offset;
 	u8 frev, crev;
 	u8 *power_state_offset;
 	struct kv_ps *ps;
@@ -2738,7 +2738,7 @@ int kv_dpm_init(struct radeon_device *rdev)
 	for (i = 0; i < SUMO_MAX_HARDWARE_POWERLEVELS; i++)
 		pi->at[i] = TRINITY_AT_DFLT;
 
-        pi->sram_end = SMC_RAM_END;
+	pi->sram_end = SMC_RAM_END;
 
 	/* Enabling nb dpm on an asrock system prevents dpm from working */
 	if (rdev->pdev->subsystem_vendor == 0x1849)
diff --git a/drivers/gpu/drm/radeon/ni.c b/drivers/gpu/drm/radeon/ni.c
index 158872eb78e4..b88d63c9be99 100644
--- a/drivers/gpu/drm/radeon/ni.c
+++ b/drivers/gpu/drm/radeon/ni.c
@@ -1257,7 +1257,7 @@ static void cayman_gpu_init(struct radeon_device *rdev)
 		tmp = RREG32_CG(CG_CGTT_LOCAL_0);
 		tmp &= ~0x00380000;
 		WREG32_CG(CG_CGTT_LOCAL_0, tmp);
-                tmp = RREG32_CG(CG_CGTT_LOCAL_1);
+		tmp = RREG32_CG(CG_CGTT_LOCAL_1);
 		tmp &= ~0x0e000000;
 		WREG32_CG(CG_CGTT_LOCAL_1, tmp);
 	}
@@ -2634,7 +2634,7 @@ int tn_set_vce_clocks(struct radeon_device *rdev, u32 evclk, u32 ecclk)
 	struct atom_clock_dividers dividers;
 	int r, i;
 
-        r = radeon_atom_get_clock_dividers(rdev, COMPUTE_ENGINE_PLL_PARAM,
+	r = radeon_atom_get_clock_dividers(rdev, COMPUTE_ENGINE_PLL_PARAM,
 					   ecclk, false, &dividers);
 	if (r)
 		return r;
diff --git a/drivers/gpu/drm/radeon/ni_dpm.c b/drivers/gpu/drm/radeon/ni_dpm.c
index c3d531a1114b..4a601f990562 100644
--- a/drivers/gpu/drm/radeon/ni_dpm.c
+++ b/drivers/gpu/drm/radeon/ni_dpm.c
@@ -725,9 +725,9 @@ extern int ni_mc_load_microcode(struct radeon_device *rdev);
 
 struct ni_power_info *ni_get_pi(struct radeon_device *rdev)
 {
-        struct ni_power_info *pi = rdev->pm.dpm.priv;
+	struct ni_power_info *pi = rdev->pm.dpm.priv;
 
-        return pi;
+	return pi;
 }
 
 struct ni_ps *ni_get_ps(struct radeon_ps *rps)
@@ -1096,9 +1096,9 @@ static void ni_stop_smc(struct radeon_device *rdev)
 
 static int ni_process_firmware_header(struct radeon_device *rdev)
 {
-        struct rv7xx_power_info *pi = rv770_get_pi(rdev);
-        struct evergreen_power_info *eg_pi = evergreen_get_pi(rdev);
-        struct ni_power_info *ni_pi = ni_get_pi(rdev);
+	struct rv7xx_power_info *pi = rv770_get_pi(rdev);
+	struct evergreen_power_info *eg_pi = evergreen_get_pi(rdev);
+	struct ni_power_info *ni_pi = ni_get_pi(rdev);
 	u32 tmp;
 	int ret;
 
@@ -1202,14 +1202,14 @@ static int ni_enter_ulp_state(struct radeon_device *rdev)
 	struct rv7xx_power_info *pi = rv770_get_pi(rdev);
 
 	if (pi->gfx_clock_gating) {
-                WREG32_P(SCLK_PWRMGT_CNTL, 0, ~DYN_GFX_CLK_OFF_EN);
+		WREG32_P(SCLK_PWRMGT_CNTL, 0, ~DYN_GFX_CLK_OFF_EN);
 		WREG32_P(SCLK_PWRMGT_CNTL, GFX_CLK_FORCE_ON, ~GFX_CLK_FORCE_ON);
-                WREG32_P(SCLK_PWRMGT_CNTL, 0, ~GFX_CLK_FORCE_ON);
+		WREG32_P(SCLK_PWRMGT_CNTL, 0, ~GFX_CLK_FORCE_ON);
 		RREG32(GB_ADDR_CONFIG);
-        }
+	}
 
 	WREG32_P(SMC_MSG, HOST_SMC_MSG(PPSMC_MSG_SwitchToMinimumPower),
-                 ~HOST_SMC_MSG_MASK);
+		 ~HOST_SMC_MSG_MASK);
 
 	udelay(25000);
 
@@ -1321,12 +1321,12 @@ static void ni_populate_mvdd_value(struct radeon_device *rdev,
 				   u32 mclk,
 				   NISLANDS_SMC_VOLTAGE_VALUE *voltage)
 {
-        struct rv7xx_power_info *pi = rv770_get_pi(rdev);
+	struct rv7xx_power_info *pi = rv770_get_pi(rdev);
 	struct evergreen_power_info *eg_pi = evergreen_get_pi(rdev);
 
 	if (!pi->mvdd_control) {
 		voltage->index = eg_pi->mvdd_high_index;
-                voltage->value = cpu_to_be16(MVDD_HIGH_VALUE);
+		voltage->value = cpu_to_be16(MVDD_HIGH_VALUE);
 		return;
 	}
 
@@ -1510,47 +1510,47 @@ int ni_copy_and_switch_arb_sets(struct radeon_device *rdev,
 	u32 mc_cg_config;
 
 	switch (arb_freq_src) {
-        case MC_CG_ARB_FREQ_F0:
+	case MC_CG_ARB_FREQ_F0:
 		mc_arb_dram_timing  = RREG32(MC_ARB_DRAM_TIMING);
 		mc_arb_dram_timing2 = RREG32(MC_ARB_DRAM_TIMING2);
 		burst_time = (RREG32(MC_ARB_BURST_TIME) & STATE0_MASK) >> STATE0_SHIFT;
 		break;
-        case MC_CG_ARB_FREQ_F1:
+	case MC_CG_ARB_FREQ_F1:
 		mc_arb_dram_timing  = RREG32(MC_ARB_DRAM_TIMING_1);
 		mc_arb_dram_timing2 = RREG32(MC_ARB_DRAM_TIMING2_1);
 		burst_time = (RREG32(MC_ARB_BURST_TIME) & STATE1_MASK) >> STATE1_SHIFT;
 		break;
-        case MC_CG_ARB_FREQ_F2:
+	case MC_CG_ARB_FREQ_F2:
 		mc_arb_dram_timing  = RREG32(MC_ARB_DRAM_TIMING_2);
 		mc_arb_dram_timing2 = RREG32(MC_ARB_DRAM_TIMING2_2);
 		burst_time = (RREG32(MC_ARB_BURST_TIME) & STATE2_MASK) >> STATE2_SHIFT;
 		break;
-        case MC_CG_ARB_FREQ_F3:
+	case MC_CG_ARB_FREQ_F3:
 		mc_arb_dram_timing  = RREG32(MC_ARB_DRAM_TIMING_3);
 		mc_arb_dram_timing2 = RREG32(MC_ARB_DRAM_TIMING2_3);
 		burst_time = (RREG32(MC_ARB_BURST_TIME) & STATE3_MASK) >> STATE3_SHIFT;
 		break;
-        default:
+	default:
 		return -EINVAL;
 	}
 
 	switch (arb_freq_dest) {
-        case MC_CG_ARB_FREQ_F0:
+	case MC_CG_ARB_FREQ_F0:
 		WREG32(MC_ARB_DRAM_TIMING, mc_arb_dram_timing);
 		WREG32(MC_ARB_DRAM_TIMING2, mc_arb_dram_timing2);
 		WREG32_P(MC_ARB_BURST_TIME, STATE0(burst_time), ~STATE0_MASK);
 		break;
-        case MC_CG_ARB_FREQ_F1:
+	case MC_CG_ARB_FREQ_F1:
 		WREG32(MC_ARB_DRAM_TIMING_1, mc_arb_dram_timing);
 		WREG32(MC_ARB_DRAM_TIMING2_1, mc_arb_dram_timing2);
 		WREG32_P(MC_ARB_BURST_TIME, STATE1(burst_time), ~STATE1_MASK);
 		break;
-        case MC_CG_ARB_FREQ_F2:
+	case MC_CG_ARB_FREQ_F2:
 		WREG32(MC_ARB_DRAM_TIMING_2, mc_arb_dram_timing);
 		WREG32(MC_ARB_DRAM_TIMING2_2, mc_arb_dram_timing2);
 		WREG32_P(MC_ARB_BURST_TIME, STATE2(burst_time), ~STATE2_MASK);
 		break;
-        case MC_CG_ARB_FREQ_F3:
+	case MC_CG_ARB_FREQ_F3:
 		WREG32(MC_ARB_DRAM_TIMING_3, mc_arb_dram_timing);
 		WREG32(MC_ARB_DRAM_TIMING2_3, mc_arb_dram_timing2);
 		WREG32_P(MC_ARB_BURST_TIME, STATE3(burst_time), ~STATE3_MASK);
@@ -1621,9 +1621,7 @@ static int ni_populate_memory_timing_parameters(struct radeon_device *rdev,
 		(u8)rv770_calculate_memory_refresh_rate(rdev, pl->sclk);
 
 
-	radeon_atom_set_engine_dram_timings(rdev,
-                                            pl->sclk,
-                                            pl->mclk);
+	radeon_atom_set_engine_dram_timings(rdev, pl->sclk, pl->mclk);
 
 	dram_timing = RREG32(MC_ARB_DRAM_TIMING);
 	dram_timing2 = RREG32(MC_ARB_DRAM_TIMING2);
@@ -1867,9 +1865,9 @@ static int ni_populate_smc_acpi_state(struct radeon_device *rdev,
 
 	mpll_ad_func_cntl_2 |= BIAS_GEN_PDNB | RESET_EN;
 
-        if (pi->mem_gddr5)
-                mpll_dq_func_cntl &= ~PDNB;
-        mpll_dq_func_cntl_2 |= BIAS_GEN_PDNB | RESET_EN | BYPASS;
+	if (pi->mem_gddr5)
+		mpll_dq_func_cntl &= ~PDNB;
+	mpll_dq_func_cntl_2 |= BIAS_GEN_PDNB | RESET_EN | BYPASS;
 
 
 	mclk_pwrmgt_cntl |= (MRDCKA0_RESET |
@@ -1891,15 +1889,15 @@ static int ni_populate_smc_acpi_state(struct radeon_device *rdev,
 			      MRDCKD1_PDNB);
 
 	dll_cntl |= (MRDCKA0_BYPASS |
-                     MRDCKA1_BYPASS |
-                     MRDCKB0_BYPASS |
-                     MRDCKB1_BYPASS |
-                     MRDCKC0_BYPASS |
-                     MRDCKC1_BYPASS |
-                     MRDCKD0_BYPASS |
-                     MRDCKD1_BYPASS);
-
-        spll_func_cntl_2 &= ~SCLK_MUX_SEL_MASK;
+		     MRDCKA1_BYPASS |
+		     MRDCKB0_BYPASS |
+		     MRDCKB1_BYPASS |
+		     MRDCKC0_BYPASS |
+		     MRDCKC1_BYPASS |
+		     MRDCKD0_BYPASS |
+		     MRDCKD1_BYPASS);
+
+	spll_func_cntl_2 &= ~SCLK_MUX_SEL_MASK;
 	spll_func_cntl_2 |= SCLK_MUX_SEL(4);
 
 	table->ACPIState.levels[0].mclk.vMPLL_AD_FUNC_CNTL = cpu_to_be32(mpll_ad_func_cntl);
@@ -2089,7 +2087,7 @@ static int ni_populate_sclk_value(struct radeon_device *rdev,
 
 static int ni_init_smc_spll_table(struct radeon_device *rdev)
 {
-        struct rv7xx_power_info *pi = rv770_get_pi(rdev);
+	struct rv7xx_power_info *pi = rv770_get_pi(rdev);
 	struct ni_power_info *ni_pi = ni_get_pi(rdev);
 	SMC_NISLANDS_SPLL_DIV_TABLE *spll_table;
 	NISLANDS_SMC_SCLK_VALUE sclk_params;
@@ -2311,8 +2309,8 @@ static int ni_convert_power_level_to_smc(struct radeon_device *rdev,
 					 NISLANDS_SMC_HW_PERFORMANCE_LEVEL *level)
 {
 	struct rv7xx_power_info *pi = rv770_get_pi(rdev);
-        struct evergreen_power_info *eg_pi = evergreen_get_pi(rdev);
-        struct ni_power_info *ni_pi = ni_get_pi(rdev);
+	struct evergreen_power_info *eg_pi = evergreen_get_pi(rdev);
+	struct ni_power_info *ni_pi = ni_get_pi(rdev);
 	int ret;
 	bool dll_state_on;
 	u16 std_vddc;
@@ -2391,8 +2389,8 @@ static int ni_populate_smc_t(struct radeon_device *rdev,
 			     struct radeon_ps *radeon_state,
 			     NISLANDS_SMC_SWSTATE *smc_state)
 {
-        struct rv7xx_power_info *pi = rv770_get_pi(rdev);
-        struct evergreen_power_info *eg_pi = evergreen_get_pi(rdev);
+	struct rv7xx_power_info *pi = rv770_get_pi(rdev);
+	struct evergreen_power_info *eg_pi = evergreen_get_pi(rdev);
 	struct ni_ps *state = ni_get_ps(radeon_state);
 	u32 a_t;
 	u32 t_l, t_h;
@@ -2451,8 +2449,8 @@ static int ni_populate_power_containment_values(struct radeon_device *rdev,
 						struct radeon_ps *radeon_state,
 						NISLANDS_SMC_SWSTATE *smc_state)
 {
-        struct rv7xx_power_info *pi = rv770_get_pi(rdev);
-        struct evergreen_power_info *eg_pi = evergreen_get_pi(rdev);
+	struct rv7xx_power_info *pi = rv770_get_pi(rdev);
+	struct evergreen_power_info *eg_pi = evergreen_get_pi(rdev);
 	struct ni_power_info *ni_pi = ni_get_pi(rdev);
 	struct ni_ps *state = ni_get_ps(radeon_state);
 	u32 prev_sclk;
@@ -2595,7 +2593,7 @@ static int ni_enable_power_containment(struct radeon_device *rdev,
 				       struct radeon_ps *radeon_new_state,
 				       bool enable)
 {
-        struct ni_power_info *ni_pi = ni_get_pi(rdev);
+	struct ni_power_info *ni_pi = ni_get_pi(rdev);
 	PPSMC_Result smc_result;
 	int ret = 0;
 
@@ -2625,7 +2623,7 @@ static int ni_convert_power_state_to_smc(struct radeon_device *rdev,
 					 struct radeon_ps *radeon_state,
 					 NISLANDS_SMC_SWSTATE *smc_state)
 {
-        struct evergreen_power_info *eg_pi = evergreen_get_pi(rdev);
+	struct evergreen_power_info *eg_pi = evergreen_get_pi(rdev);
 	struct ni_power_info *ni_pi = ni_get_pi(rdev);
 	struct ni_ps *state = ni_get_ps(radeon_state);
 	int i, ret;
@@ -2770,46 +2768,46 @@ static bool ni_check_s0_mc_reg_index(u16 in_reg, u16 *out_reg)
 	bool result = true;
 
 	switch (in_reg) {
-        case  MC_SEQ_RAS_TIMING >> 2:
+	case  MC_SEQ_RAS_TIMING >> 2:
 		*out_reg = MC_SEQ_RAS_TIMING_LP >> 2;
 		break;
-        case MC_SEQ_CAS_TIMING >> 2:
+	case MC_SEQ_CAS_TIMING >> 2:
 		*out_reg = MC_SEQ_CAS_TIMING_LP >> 2;
 		break;
-        case MC_SEQ_MISC_TIMING >> 2:
+	case MC_SEQ_MISC_TIMING >> 2:
 		*out_reg = MC_SEQ_MISC_TIMING_LP >> 2;
 		break;
-        case MC_SEQ_MISC_TIMING2 >> 2:
+	case MC_SEQ_MISC_TIMING2 >> 2:
 		*out_reg = MC_SEQ_MISC_TIMING2_LP >> 2;
 		break;
-        case MC_SEQ_RD_CTL_D0 >> 2:
+	case MC_SEQ_RD_CTL_D0 >> 2:
 		*out_reg = MC_SEQ_RD_CTL_D0_LP >> 2;
 		break;
-        case MC_SEQ_RD_CTL_D1 >> 2:
+	case MC_SEQ_RD_CTL_D1 >> 2:
 		*out_reg = MC_SEQ_RD_CTL_D1_LP >> 2;
 		break;
-        case MC_SEQ_WR_CTL_D0 >> 2:
+	case MC_SEQ_WR_CTL_D0 >> 2:
 		*out_reg = MC_SEQ_WR_CTL_D0_LP >> 2;
 		break;
-        case MC_SEQ_WR_CTL_D1 >> 2:
+	case MC_SEQ_WR_CTL_D1 >> 2:
 		*out_reg = MC_SEQ_WR_CTL_D1_LP >> 2;
 		break;
-        case MC_PMG_CMD_EMRS >> 2:
+	case MC_PMG_CMD_EMRS >> 2:
 		*out_reg = MC_SEQ_PMG_CMD_EMRS_LP >> 2;
 		break;
-        case MC_PMG_CMD_MRS >> 2:
+	case MC_PMG_CMD_MRS >> 2:
 		*out_reg = MC_SEQ_PMG_CMD_MRS_LP >> 2;
 		break;
-        case MC_PMG_CMD_MRS1 >> 2:
+	case MC_PMG_CMD_MRS1 >> 2:
 		*out_reg = MC_SEQ_PMG_CMD_MRS1_LP >> 2;
 		break;
-        case MC_SEQ_PMG_TIMING >> 2:
+	case MC_SEQ_PMG_TIMING >> 2:
 		*out_reg = MC_SEQ_PMG_TIMING_LP >> 2;
 		break;
-        case MC_PMG_CMD_MRS2 >> 2:
+	case MC_PMG_CMD_MRS2 >> 2:
 		*out_reg = MC_SEQ_PMG_CMD_MRS2_LP >> 2;
 		break;
-        default:
+	default:
 		result = false;
 		break;
 	}
@@ -2876,9 +2874,9 @@ static int ni_initialize_mc_reg_table(struct radeon_device *rdev)
 	struct ni_mc_reg_table *ni_table = &ni_pi->mc_reg_table;
 	u8 module_index = rv770_get_memory_module_index(rdev);
 
-        table = kzalloc(sizeof(struct atom_mc_reg_table), GFP_KERNEL);
-        if (!table)
-                return -ENOMEM;
+	table = kzalloc(sizeof(struct atom_mc_reg_table), GFP_KERNEL);
+	if (!table)
+		return -ENOMEM;
 
 	WREG32(MC_SEQ_RAS_TIMING_LP, RREG32(MC_SEQ_RAS_TIMING));
 	WREG32(MC_SEQ_CAS_TIMING_LP, RREG32(MC_SEQ_CAS_TIMING));
@@ -2896,25 +2894,25 @@ static int ni_initialize_mc_reg_table(struct radeon_device *rdev)
 
 	ret = radeon_atom_init_mc_reg_table(rdev, module_index, table);
 
-        if (ret)
-                goto init_mc_done;
+	if (ret)
+		goto init_mc_done;
 
 	ret = ni_copy_vbios_mc_reg_table(table, ni_table);
 
-        if (ret)
-                goto init_mc_done;
+	if (ret)
+		goto init_mc_done;
 
 	ni_set_s0_mc_reg_index(ni_table);
 
 	ret = ni_set_mc_special_registers(rdev, ni_table);
 
-        if (ret)
-                goto init_mc_done;
+	if (ret)
+		goto init_mc_done;
 
 	ni_set_valid_flag(ni_table);
 
 init_mc_done:
-        kfree(table);
+	kfree(table);
 
 	return ret;
 }
@@ -2994,7 +2992,7 @@ static int ni_populate_mc_reg_table(struct radeon_device *rdev,
 {
 	struct rv7xx_power_info *pi = rv770_get_pi(rdev);
 	struct evergreen_power_info *eg_pi = evergreen_get_pi(rdev);
-        struct ni_power_info *ni_pi = ni_get_pi(rdev);
+	struct ni_power_info *ni_pi = ni_get_pi(rdev);
 	struct ni_ps *boot_state = ni_get_ps(radeon_boot_state);
 	SMC_NIslands_MCRegisters *mc_reg_table = &ni_pi->smc_mc_reg_table;
 
@@ -3025,7 +3023,7 @@ static int ni_upload_mc_reg_table(struct radeon_device *rdev,
 {
 	struct rv7xx_power_info *pi = rv770_get_pi(rdev);
 	struct evergreen_power_info *eg_pi = evergreen_get_pi(rdev);
-        struct ni_power_info *ni_pi = ni_get_pi(rdev);
+	struct ni_power_info *ni_pi = ni_get_pi(rdev);
 	struct ni_ps *ni_new_state = ni_get_ps(radeon_new_state);
 	SMC_NIslands_MCRegisters *mc_reg_table = &ni_pi->smc_mc_reg_table;
 	u16 address;
@@ -3142,7 +3140,7 @@ static int ni_initialize_smc_cac_tables(struct radeon_device *rdev)
 	struct ni_power_info *ni_pi = ni_get_pi(rdev);
 	PP_NIslands_CACTABLES *cac_tables = NULL;
 	int i, ret;
-        u32 reg;
+	u32 reg;
 
 	if (ni_pi->enable_cac == false)
 		return 0;
@@ -3422,13 +3420,13 @@ static int ni_pcie_performance_request(struct radeon_device *rdev,
 	struct evergreen_power_info *eg_pi = evergreen_get_pi(rdev);
 
 	if ((perf_req == PCIE_PERF_REQ_PECI_GEN1) ||
-            (perf_req == PCIE_PERF_REQ_PECI_GEN2)) {
+	    (perf_req == PCIE_PERF_REQ_PECI_GEN2)) {
 		if (eg_pi->pcie_performance_request_registered == false)
 			radeon_acpi_pcie_notify_device_ready(rdev);
 		eg_pi->pcie_performance_request_registered = true;
 		return radeon_acpi_pcie_performance_request(rdev, perf_req, advertise);
 	} else if ((perf_req == PCIE_PERF_REQ_REMOVE_REGISTRY) &&
-                   eg_pi->pcie_performance_request_registered) {
+		    eg_pi->pcie_performance_request_registered) {
 		eg_pi->pcie_performance_request_registered = false;
 		return radeon_acpi_pcie_performance_request(rdev, perf_req, advertise);
 	}
@@ -3441,12 +3439,12 @@ static int ni_advertise_gen2_capability(struct radeon_device *rdev)
 	struct rv7xx_power_info *pi = rv770_get_pi(rdev);
 	u32 tmp;
 
-        tmp = RREG32_PCIE_PORT(PCIE_LC_SPEED_CNTL);
+	tmp = RREG32_PCIE_PORT(PCIE_LC_SPEED_CNTL);
 
-        if ((tmp & LC_OTHER_SIDE_EVER_SENT_GEN2) &&
-            (tmp & LC_OTHER_SIDE_SUPPORTS_GEN2))
-                pi->pcie_gen2 = true;
-        else
+	if ((tmp & LC_OTHER_SIDE_EVER_SENT_GEN2) &&
+	    (tmp & LC_OTHER_SIDE_SUPPORTS_GEN2))
+		pi->pcie_gen2 = true;
+	else
 		pi->pcie_gen2 = false;
 
 	if (!pi->pcie_gen2)
@@ -3458,8 +3456,8 @@ static int ni_advertise_gen2_capability(struct radeon_device *rdev)
 static void ni_enable_bif_dynamic_pcie_gen2(struct radeon_device *rdev,
 					    bool enable)
 {
-        struct rv7xx_power_info *pi = rv770_get_pi(rdev);
-        u32 tmp, bif;
+	struct rv7xx_power_info *pi = rv770_get_pi(rdev);
+	u32 tmp, bif;
 
 	tmp = RREG32_PCIE_PORT(PCIE_LC_SPEED_CNTL);
 
@@ -3502,7 +3500,7 @@ static void ni_enable_dynamic_pcie_gen2(struct radeon_device *rdev,
 	if (enable)
 		WREG32_P(GENERAL_PWRMGT, ENABLE_GEN2PCIE, ~ENABLE_GEN2PCIE);
 	else
-                WREG32_P(GENERAL_PWRMGT, 0, ~ENABLE_GEN2PCIE);
+		WREG32_P(GENERAL_PWRMGT, 0, ~ENABLE_GEN2PCIE);
 }
 
 void ni_set_uvd_clock_before_set_eng_clock(struct radeon_device *rdev,
@@ -3563,7 +3561,7 @@ void ni_update_current_ps(struct radeon_device *rdev,
 {
 	struct ni_ps *new_ps = ni_get_ps(rps);
 	struct evergreen_power_info *eg_pi = evergreen_get_pi(rdev);
-        struct ni_power_info *ni_pi = ni_get_pi(rdev);
+	struct ni_power_info *ni_pi = ni_get_pi(rdev);
 
 	eg_pi->current_rps = *rps;
 	ni_pi->current_ps = *new_ps;
@@ -3575,7 +3573,7 @@ void ni_update_requested_ps(struct radeon_device *rdev,
 {
 	struct ni_ps *new_ps = ni_get_ps(rps);
 	struct evergreen_power_info *eg_pi = evergreen_get_pi(rdev);
-        struct ni_power_info *ni_pi = ni_get_pi(rdev);
+	struct ni_power_info *ni_pi = ni_get_pi(rdev);
 
 	eg_pi->requested_rps = *rps;
 	ni_pi->requested_ps = *new_ps;
@@ -3591,8 +3589,8 @@ int ni_dpm_enable(struct radeon_device *rdev)
 
 	if (pi->gfx_clock_gating)
 		ni_cg_clockgating_default(rdev);
-        if (btc_dpm_enabled(rdev))
-                return -EINVAL;
+	if (btc_dpm_enabled(rdev))
+		return -EINVAL;
 	if (pi->mg_clock_gating)
 		ni_mg_clockgating_default(rdev);
 	if (eg_pi->ls_clock_gating)
@@ -3991,7 +3989,7 @@ static int ni_parse_power_table(struct radeon_device *rdev)
 	union pplib_clock_info *clock_info;
 	union power_info *power_info;
 	int index = GetIndexIntoMasterTable(DATA, PowerPlayInfo);
-        u16 data_offset;
+	u16 data_offset;
 	u8 frev, crev;
 	struct ni_ps *ps;
 
diff --git a/drivers/gpu/drm/radeon/r600.c b/drivers/gpu/drm/radeon/r600.c
index ed121042247f..f86ab695ee8f 100644
--- a/drivers/gpu/drm/radeon/r600.c
+++ b/drivers/gpu/drm/radeon/r600.c
@@ -235,8 +235,8 @@ int r600_set_uvd_clocks(struct radeon_device *rdev, u32 vclk, u32 dclk)
 		fb_div |= 1;
 
 	r = radeon_uvd_send_upll_ctlreq(rdev, CG_UPLL_FUNC_CNTL);
-        if (r)
-                return r;
+	if (r)
+		return r;
 
 	/* assert PLL_RESET */
 	WREG32_P(CG_UPLL_FUNC_CNTL, UPLL_RESET_MASK, ~UPLL_RESET_MASK);
@@ -1490,7 +1490,7 @@ static int r600_mc_init(struct radeon_device *rdev)
 					rdev->fastfb_working = true;
 				}
 			}
-  		}
+		}
 	}
 
 	radeon_update_bandwidth_info(rdev);
@@ -4574,7 +4574,7 @@ uint64_t r600_get_gpu_clock_counter(struct radeon_device *rdev)
 	mutex_lock(&rdev->gpu_clock_mutex);
 	WREG32(RLC_CAPTURE_GPU_CLOCK_COUNT, 1);
 	clock = (uint64_t)RREG32(RLC_GPU_CLOCK_COUNT_LSB) |
-	        ((uint64_t)RREG32(RLC_GPU_CLOCK_COUNT_MSB) << 32ULL);
+		((uint64_t)RREG32(RLC_GPU_CLOCK_COUNT_MSB) << 32ULL);
 	mutex_unlock(&rdev->gpu_clock_mutex);
 	return clock;
 }
diff --git a/drivers/gpu/drm/radeon/r600_cs.c b/drivers/gpu/drm/radeon/r600_cs.c
index 2f36fa1576e0..b69c8de35bd3 100644
--- a/drivers/gpu/drm/radeon/r600_cs.c
+++ b/drivers/gpu/drm/radeon/r600_cs.c
@@ -1671,8 +1671,8 @@ static int r600_packet3_check(struct radeon_cs_parser *p,
 		}
 
 		offset = reloc->gpu_offset +
-		         (idx_value & 0xfffffff0) +
-		         ((u64)(tmp & 0xff) << 32);
+			 (idx_value & 0xfffffff0) +
+			 ((u64)(tmp & 0xff) << 32);
 
 		ib[idx + 0] = offset;
 		ib[idx + 1] = (tmp & 0xffffff00) | (upper_32_bits(offset) & 0xff);
@@ -1712,8 +1712,8 @@ static int r600_packet3_check(struct radeon_cs_parser *p,
 		}
 
 		offset = reloc->gpu_offset +
-		         idx_value +
-		         ((u64)(radeon_get_ib_value(p, idx+1) & 0xff) << 32);
+			 idx_value +
+			 ((u64)(radeon_get_ib_value(p, idx+1) & 0xff) << 32);
 
 		ib[idx+0] = offset;
 		ib[idx+1] = upper_32_bits(offset) & 0xff;
@@ -1764,8 +1764,8 @@ static int r600_packet3_check(struct radeon_cs_parser *p,
 			}
 
 			offset = reloc->gpu_offset +
-			         (radeon_get_ib_value(p, idx+1) & 0xfffffff0) +
-			         ((u64)(radeon_get_ib_value(p, idx+2) & 0xff) << 32);
+				 (radeon_get_ib_value(p, idx+1) & 0xfffffff0) +
+				 ((u64)(radeon_get_ib_value(p, idx+2) & 0xff) << 32);
 
 			ib[idx+1] = (ib[idx+1] & 0x3) | (offset & 0xfffffff0);
 			ib[idx+2] = upper_32_bits(offset) & 0xff;
@@ -1876,8 +1876,8 @@ static int r600_packet3_check(struct radeon_cs_parser *p,
 				return -EINVAL;
 			}
 			offset = reloc->gpu_offset +
-			         (radeon_get_ib_value(p, idx+1) & 0xfffffff8) +
-			         ((u64)(radeon_get_ib_value(p, idx+2) & 0xff) << 32);
+				 (radeon_get_ib_value(p, idx+1) & 0xfffffff8) +
+				 ((u64)(radeon_get_ib_value(p, idx+2) & 0xff) << 32);
 
 			ib[idx+1] = offset & 0xfffffff8;
 			ib[idx+2] = upper_32_bits(offset) & 0xff;
@@ -1898,8 +1898,8 @@ static int r600_packet3_check(struct radeon_cs_parser *p,
 		}
 
 		offset = reloc->gpu_offset +
-		         (radeon_get_ib_value(p, idx+1) & 0xfffffffc) +
-		         ((u64)(radeon_get_ib_value(p, idx+2) & 0xff) << 32);
+			 (radeon_get_ib_value(p, idx+1) & 0xfffffffc) +
+			 ((u64)(radeon_get_ib_value(p, idx+2) & 0xff) << 32);
 
 		ib[idx+1] = offset & 0xfffffffc;
 		ib[idx+2] = (ib[idx+2] & 0xffffff00) | (upper_32_bits(offset) & 0xff);
diff --git a/drivers/gpu/drm/radeon/r600_dpm.c b/drivers/gpu/drm/radeon/r600_dpm.c
index fa2154493cf1..6a4b020dd0b4 100644
--- a/drivers/gpu/drm/radeon/r600_dpm.c
+++ b/drivers/gpu/drm/radeon/r600_dpm.c
@@ -844,7 +844,7 @@ int r600_get_platform_caps(struct radeon_device *rdev)
 	struct radeon_mode_info *mode_info = &rdev->mode_info;
 	union power_info *power_info;
 	int index = GetIndexIntoMasterTable(DATA, PowerPlayInfo);
-        u16 data_offset;
+	u16 data_offset;
 	u8 frev, crev;
 
 	if (!atom_parse_data_header(mode_info->atom_context, index, NULL,
@@ -874,7 +874,7 @@ int r600_parse_extended_power_table(struct radeon_device *rdev)
 	union fan_info *fan_info;
 	ATOM_PPLIB_Clock_Voltage_Dependency_Table *dep_table;
 	int index = GetIndexIntoMasterTable(DATA, PowerPlayInfo);
-        u16 data_offset;
+	u16 data_offset;
 	u8 frev, crev;
 	int ret, i;
 
@@ -1070,7 +1070,7 @@ int r600_parse_extended_power_table(struct radeon_device *rdev)
 			ext_hdr->usVCETableOffset) {
 			VCEClockInfoArray *array = (VCEClockInfoArray *)
 				(mode_info->atom_context->bios + data_offset +
-                                 le16_to_cpu(ext_hdr->usVCETableOffset) + 1);
+				 le16_to_cpu(ext_hdr->usVCETableOffset) + 1);
 			ATOM_PPLIB_VCE_Clock_Voltage_Limit_Table *limits =
 				(ATOM_PPLIB_VCE_Clock_Voltage_Limit_Table *)
 				(mode_info->atom_context->bios + data_offset +
diff --git a/drivers/gpu/drm/radeon/r600_hdmi.c b/drivers/gpu/drm/radeon/r600_hdmi.c
index e85894ade95c..e82a99cb2459 100644
--- a/drivers/gpu/drm/radeon/r600_hdmi.c
+++ b/drivers/gpu/drm/radeon/r600_hdmi.c
@@ -215,7 +215,7 @@ void r600_hdmi_update_acr(struct drm_encoder *encoder, long offset,
  * build a HDMI Video Info Frame
  */
 void r600_set_avi_packet(struct radeon_device *rdev, u32 offset,
-    unsigned char *buffer, size_t size)
+			 unsigned char *buffer, size_t size)
 {
 	uint8_t *frame = buffer + 3;
 
@@ -312,7 +312,7 @@ void r600_hdmi_audio_workaround(struct drm_encoder *encoder)
 }
 
 void r600_hdmi_audio_set_dto(struct radeon_device *rdev,
-    struct radeon_crtc *crtc, unsigned int clock)
+			     struct radeon_crtc *crtc, unsigned int clock)
 {
 	struct radeon_encoder *radeon_encoder;
 	struct radeon_encoder_atom_dig *dig;
diff --git a/drivers/gpu/drm/radeon/radeon_acpi.h b/drivers/gpu/drm/radeon/radeon_acpi.h
index be4af76f213d..cd872f7953c6 100644
--- a/drivers/gpu/drm/radeon/radeon_acpi.h
+++ b/drivers/gpu/drm/radeon/radeon_acpi.h
@@ -291,6 +291,8 @@ int radeon_atif_handler(struct radeon_device *rdev,
 #       define ATPX_FIXED_NOT_SUPPORTED                            (1 << 9)
 #       define ATPX_DYNAMIC_DGPU_POWER_OFF_SUPPORTED               (1 << 10)
 #       define ATPX_DGPU_REQ_POWER_FOR_DISPLAYS                    (1 << 11)
+#       define ATPX_DGPU_CAN_DRIVE_DISPLAYS                        (1 << 12)
+#       define ATPX_MS_HYBRID_GFX_SUPPORTED                        (1 << 14)
 #define ATPX_FUNCTION_POWER_CONTROL                                0x2
 /* ARG0: ATPX_FUNCTION_POWER_CONTROL
  * ARG1:
diff --git a/drivers/gpu/drm/radeon/radeon_atombios.c b/drivers/gpu/drm/radeon/radeon_atombios.c
index de9a2ffcf5f7..f8097a0e7a79 100644
--- a/drivers/gpu/drm/radeon/radeon_atombios.c
+++ b/drivers/gpu/drm/radeon/radeon_atombios.c
@@ -2095,7 +2095,7 @@ static int radeon_atombios_parse_power_table_1_3(struct radeon_device *rdev)
 	struct radeon_i2c_bus_rec i2c_bus;
 	union power_info *power_info;
 	int index = GetIndexIntoMasterTable(DATA, PowerPlayInfo);
-        u16 data_offset;
+	u16 data_offset;
 	u8 frev, crev;
 
 	if (!atom_parse_data_header(mode_info->atom_context, index, NULL,
@@ -2575,7 +2575,7 @@ static int radeon_atombios_parse_power_table_4_5(struct radeon_device *rdev)
 	bool valid;
 	union power_info *power_info;
 	int index = GetIndexIntoMasterTable(DATA, PowerPlayInfo);
-        u16 data_offset;
+	u16 data_offset;
 	u8 frev, crev;
 
 	if (!atom_parse_data_header(mode_info->atom_context, index, NULL,
@@ -2666,7 +2666,7 @@ static int radeon_atombios_parse_power_table_6(struct radeon_device *rdev)
 	bool valid;
 	union power_info *power_info;
 	int index = GetIndexIntoMasterTable(DATA, PowerPlayInfo);
-        u16 data_offset;
+	u16 data_offset;
 	u8 frev, crev;
 	u8 *power_state_offset;
 
diff --git a/drivers/gpu/drm/radeon/radeon_atpx_handler.c b/drivers/gpu/drm/radeon/radeon_atpx_handler.c
index 56482e35d43e..fd8c4d317e60 100644
--- a/drivers/gpu/drm/radeon/radeon_atpx_handler.c
+++ b/drivers/gpu/drm/radeon/radeon_atpx_handler.c
@@ -62,6 +62,10 @@ bool radeon_has_atpx(void) {
 	return radeon_atpx_priv.atpx_detected;
 }
 
+bool radeon_has_atpx_dgpu_power_cntl(void) {
+	return radeon_atpx_priv.atpx.functions.power_cntl;
+}
+
 /**
  * radeon_atpx_call - call an ATPX method
  *
@@ -141,10 +145,6 @@ static void radeon_atpx_parse_functions(struct radeon_atpx_functions *f, u32 mas
  */
 static int radeon_atpx_validate(struct radeon_atpx *atpx)
 {
-	/* make sure required functions are enabled */
-	/* dGPU power control is required */
-	atpx->functions.power_cntl = true;
-
 	if (atpx->functions.px_params) {
 		union acpi_object *info;
 		struct atpx_px_params output;
diff --git a/drivers/gpu/drm/radeon/radeon_device.c b/drivers/gpu/drm/radeon/radeon_device.c
index 902b59cebac5..4fd1a961012d 100644
--- a/drivers/gpu/drm/radeon/radeon_device.c
+++ b/drivers/gpu/drm/radeon/radeon_device.c
@@ -103,6 +103,12 @@ static const char radeon_family_name[][16] = {
 	"LAST",
 };
 
+#if defined(CONFIG_VGA_SWITCHEROO)
+bool radeon_has_atpx_dgpu_power_cntl(void);
+#else
+static inline bool radeon_has_atpx_dgpu_power_cntl(void) { return false; }
+#endif
+
 #define RADEON_PX_QUIRK_DISABLE_PX  (1 << 0)
 #define RADEON_PX_QUIRK_LONG_WAKEUP (1 << 1)
 
@@ -1155,9 +1161,9 @@ static void radeon_check_arguments(struct radeon_device *rdev)
 		radeon_vm_size = 4;
 	}
 
-       /*
-        * Max GPUVM size for Cayman, SI and CI are 40 bits.
-        */
+	/*
+	 * Max GPUVM size for Cayman, SI and CI are 40 bits.
+	 */
 	if (radeon_vm_size > 1024) {
 		dev_warn(rdev->dev, "VM size (%d) too large, max is 1TB\n",
 			 radeon_vm_size);
@@ -1433,7 +1439,7 @@ int radeon_device_init(struct radeon_device *rdev,
 	 * ignore it */
 	vga_client_register(rdev->pdev, rdev, NULL, radeon_vga_set_decode);
 
-	if (rdev->flags & RADEON_IS_PX)
+	if ((rdev->flags & RADEON_IS_PX) && radeon_has_atpx_dgpu_power_cntl())
 		runtime = true;
 	vga_switcheroo_register_client(rdev->pdev, &radeon_switcheroo_ops, runtime);
 	if (runtime)
@@ -1744,7 +1750,6 @@ int radeon_resume_kms(struct drm_device *dev, bool resume, bool fbcon)
 	}
 
 	drm_kms_helper_poll_enable(dev);
-	drm_helper_hpd_irq_event(dev);
 
 	/* set the power state here in case we are a PX system or headless */
 	if ((rdev->pm.pm_method == PM_METHOD_DPM) && rdev->pm.dpm_enabled)
@@ -1896,7 +1901,7 @@ int radeon_debugfs_add_files(struct radeon_device *rdev,
 	if (i > RADEON_DEBUGFS_MAX_COMPONENTS) {
 		DRM_ERROR("Reached maximum number of debugfs components.\n");
 		DRM_ERROR("Report so we increase "
-		          "RADEON_DEBUGFS_MAX_COMPONENTS.\n");
+			  "RADEON_DEBUGFS_MAX_COMPONENTS.\n");
 		return -EINVAL;
 	}
 	rdev->debugfs[rdev->debugfs_count].files = files;
diff --git a/drivers/gpu/drm/radeon/radeon_display.c b/drivers/gpu/drm/radeon/radeon_display.c
index a4674bfd979a..fcc7483d3f7b 100644
--- a/drivers/gpu/drm/radeon/radeon_display.c
+++ b/drivers/gpu/drm/radeon/radeon_display.c
@@ -403,10 +403,11 @@ static void radeon_flip_work_func(struct work_struct *__work)
 	struct drm_crtc *crtc = &radeon_crtc->base;
 	unsigned long flags;
 	int r;
-	int vpos, hpos, stat, min_udelay;
+	int vpos, hpos, stat, min_udelay = 0;
+	unsigned repcnt = 4;
 	struct drm_vblank_crtc *vblank = &crtc->dev->vblank[work->crtc_id];
 
-        down_read(&rdev->exclusive_lock);
+	down_read(&rdev->exclusive_lock);
 	if (work->fence) {
 		struct radeon_fence *fence;
 
@@ -454,7 +455,7 @@ static void radeon_flip_work_func(struct work_struct *__work)
 	 * In practice this won't execute very often unless on very fast
 	 * machines because the time window for this to happen is very small.
 	 */
-	for (;;) {
+	while (radeon_crtc->enabled && --repcnt) {
 		/* GET_DISTANCE_TO_VBLANKSTART returns distance to real vblank
 		 * start in hpos, and to the "fudged earlier" vblank start in
 		 * vpos.
@@ -470,12 +471,24 @@ static void radeon_flip_work_func(struct work_struct *__work)
 			break;
 
 		/* Sleep at least until estimated real start of hw vblank */
-		spin_unlock_irqrestore(&crtc->dev->event_lock, flags);
 		min_udelay = (-hpos + 1) * max(vblank->linedur_ns / 1000, 5);
+		if (min_udelay > vblank->framedur_ns / 2000) {
+			/* Don't wait ridiculously long - something is wrong */
+			repcnt = 0;
+			break;
+		}
+		spin_unlock_irqrestore(&crtc->dev->event_lock, flags);
 		usleep_range(min_udelay, 2 * min_udelay);
 		spin_lock_irqsave(&crtc->dev->event_lock, flags);
 	};
 
+	if (!repcnt)
+		DRM_DEBUG_DRIVER("Delay problem on crtc %d: min_udelay %d, "
+				 "framedur %d, linedur %d, stat %d, vpos %d, "
+				 "hpos %d\n", work->crtc_id, min_udelay,
+				 vblank->framedur_ns / 1000,
+				 vblank->linedur_ns / 1000, stat, vpos, hpos);
+
 	/* do the flip (mmio) */
 	radeon_page_flip(rdev, radeon_crtc->crtc_id, work->base);
 
@@ -906,7 +919,7 @@ static void avivo_reduce_ratio(unsigned *nom, unsigned *den,
 	*den /= tmp;
 
 	/* make sure nominator is large enough */
-        if (*nom < nom_min) {
+	if (*nom < nom_min) {
 		tmp = DIV_ROUND_UP(nom_min, *nom);
 		*nom *= tmp;
 		*den *= tmp;
@@ -946,7 +959,7 @@ static void avivo_get_fb_ref_div(unsigned nom, unsigned den, unsigned post_div,
 	*fb_div = DIV_ROUND_CLOSEST(nom * *ref_div * post_div, den);
 
 	/* limit fb divider to its maximum */
-        if (*fb_div > fb_div_max) {
+	if (*fb_div > fb_div_max) {
 		*ref_div = DIV_ROUND_CLOSEST(*ref_div * fb_div_max, *fb_div);
 		*fb_div = fb_div_max;
 	}
@@ -1670,10 +1683,8 @@ int radeon_modeset_init(struct radeon_device *rdev)
 	/* setup afmt */
 	radeon_afmt_init(rdev);
 
-	if (!list_empty(&rdev->ddev->mode_config.connector_list)) {
-		radeon_fbdev_init(rdev);
-		drm_kms_helper_poll_init(rdev->ddev);
-	}
+	radeon_fbdev_init(rdev);
+	drm_kms_helper_poll_init(rdev->ddev);
 
 	/* do pm late init */
 	ret = radeon_pm_late_init(rdev);
diff --git a/drivers/gpu/drm/radeon/radeon_drv.c b/drivers/gpu/drm/radeon/radeon_drv.c
index cad25557650f..ccd4ad4ee592 100644
--- a/drivers/gpu/drm/radeon/radeon_drv.c
+++ b/drivers/gpu/drm/radeon/radeon_drv.c
@@ -322,6 +322,14 @@ static int radeon_pci_probe(struct pci_dev *pdev,
 	int ret;
 
 	/*
+	 * Initialize amdkfd before starting radeon. If it was not loaded yet,
+	 * defer radeon probing
+	 */
+	ret = radeon_kfd_init();
+	if (ret == -EPROBE_DEFER)
+		return ret;
+
+	/*
 	 * apple-gmux is needed on dual GPU MacBook Pro
 	 * to probe the panel if we're the inactive GPU.
 	 */
@@ -581,8 +589,6 @@ static int __init radeon_init(void)
 		return -EINVAL;
 	}
 
-	radeon_kfd_init();
-
 	/* let modprobe override vga console setting */
 	return drm_pci_init(driver, pdriver);
 }
diff --git a/drivers/gpu/drm/radeon/radeon_fb.c b/drivers/gpu/drm/radeon/radeon_fb.c
index d2e628eea53d..0e3143acb565 100644
--- a/drivers/gpu/drm/radeon/radeon_fb.c
+++ b/drivers/gpu/drm/radeon/radeon_fb.c
@@ -38,9 +38,9 @@
 #include <linux/vga_switcheroo.h>
 
 /* object hierarchy -
-   this contains a helper + a radeon fb
-   the helper contains a pointer to radeon framebuffer baseclass.
-*/
+ * this contains a helper + a radeon fb
+ * the helper contains a pointer to radeon framebuffer baseclass.
+ */
 struct radeon_fbdev {
 	struct drm_fb_helper helper;
 	struct radeon_framebuffer rfb;
@@ -292,7 +292,8 @@ out_unref:
 
 void radeon_fb_output_poll_changed(struct radeon_device *rdev)
 {
-	drm_fb_helper_hotplug_event(&rdev->mode_info.rfbdev->helper);
+	if (rdev->mode_info.rfbdev)
+		drm_fb_helper_hotplug_event(&rdev->mode_info.rfbdev->helper);
 }
 
 static int radeon_fbdev_destroy(struct drm_device *dev, struct radeon_fbdev *rfbdev)
@@ -325,6 +326,10 @@ int radeon_fbdev_init(struct radeon_device *rdev)
 	int bpp_sel = 32;
 	int ret;
 
+	/* don't enable fbdev if no connectors */
+	if (list_empty(&rdev->ddev->mode_config.connector_list))
+		return 0;
+
 	/* select 8 bpp console on RN50 or 16MB cards */
 	if (ASIC_IS_RN50(rdev) || rdev->mc.real_vram_size <= (32*1024*1024))
 		bpp_sel = 8;
@@ -377,11 +382,15 @@ void radeon_fbdev_fini(struct radeon_device *rdev)
 
 void radeon_fbdev_set_suspend(struct radeon_device *rdev, int state)
 {
-	fb_set_suspend(rdev->mode_info.rfbdev->helper.fbdev, state);
+	if (rdev->mode_info.rfbdev)
+		fb_set_suspend(rdev->mode_info.rfbdev->helper.fbdev, state);
 }
 
 bool radeon_fbdev_robj_is_fb(struct radeon_device *rdev, struct radeon_bo *robj)
 {
+	if (!rdev->mode_info.rfbdev)
+		return false;
+
 	if (robj == gem_to_radeon_bo(rdev->mode_info.rfbdev->rfb.obj))
 		return true;
 	return false;
@@ -389,12 +398,14 @@ bool radeon_fbdev_robj_is_fb(struct radeon_device *rdev, struct radeon_bo *robj)
 
 void radeon_fb_add_connector(struct radeon_device *rdev, struct drm_connector *connector)
 {
-	drm_fb_helper_add_one_connector(&rdev->mode_info.rfbdev->helper, connector);
+	if (rdev->mode_info.rfbdev)
+		drm_fb_helper_add_one_connector(&rdev->mode_info.rfbdev->helper, connector);
 }
 
 void radeon_fb_remove_connector(struct radeon_device *rdev, struct drm_connector *connector)
 {
-	drm_fb_helper_remove_one_connector(&rdev->mode_info.rfbdev->helper, connector);
+	if (rdev->mode_info.rfbdev)
+		drm_fb_helper_remove_one_connector(&rdev->mode_info.rfbdev->helper, connector);
 }
 
 void radeon_fbdev_restore_mode(struct radeon_device *rdev)
diff --git a/drivers/gpu/drm/radeon/radeon_ib.c b/drivers/gpu/drm/radeon/radeon_ib.c
index c39ce1f05703..92ce0e533bc0 100644
--- a/drivers/gpu/drm/radeon/radeon_ib.c
+++ b/drivers/gpu/drm/radeon/radeon_ib.c
@@ -274,7 +274,7 @@ int radeon_ib_ring_tests(struct radeon_device *rdev)
 			if (i == RADEON_RING_TYPE_GFX_INDEX) {
 				/* oh, oh, that's really bad */
 				DRM_ERROR("radeon: failed testing IB on GFX ring (%d).\n", r);
-		                rdev->accel_working = false;
+				rdev->accel_working = false;
 				return r;
 
 			} else {
@@ -304,7 +304,7 @@ static int radeon_debugfs_sa_info(struct seq_file *m, void *data)
 }
 
 static struct drm_info_list radeon_debugfs_sa_list[] = {
-        {"radeon_sa_info", &radeon_debugfs_sa_info, 0, NULL},
+	{"radeon_sa_info", &radeon_debugfs_sa_info, 0, NULL},
 };
 
 #endif
diff --git a/drivers/gpu/drm/radeon/radeon_kfd.c b/drivers/gpu/drm/radeon/radeon_kfd.c
index 9a4d69e59401..87a9ebb5f58f 100644
--- a/drivers/gpu/drm/radeon/radeon_kfd.c
+++ b/drivers/gpu/drm/radeon/radeon_kfd.c
@@ -132,35 +132,34 @@ static const struct kfd2kgd_calls kfd2kgd = {
 
 static const struct kgd2kfd_calls *kgd2kfd;
 
-bool radeon_kfd_init(void)
+int radeon_kfd_init(void)
 {
+	int ret;
+
 #if defined(CONFIG_HSA_AMD_MODULE)
-	bool (*kgd2kfd_init_p)(unsigned, const struct kgd2kfd_calls**);
+	int (*kgd2kfd_init_p)(unsigned, const struct kgd2kfd_calls**);
 
 	kgd2kfd_init_p = symbol_request(kgd2kfd_init);
 
 	if (kgd2kfd_init_p == NULL)
-		return false;
+		return -ENOENT;
 
-	if (!kgd2kfd_init_p(KFD_INTERFACE_VERSION, &kgd2kfd)) {
+	ret = kgd2kfd_init_p(KFD_INTERFACE_VERSION, &kgd2kfd);
+	if (ret) {
 		symbol_put(kgd2kfd_init);
 		kgd2kfd = NULL;
-
-		return false;
 	}
 
-	return true;
 #elif defined(CONFIG_HSA_AMD)
-	if (!kgd2kfd_init(KFD_INTERFACE_VERSION, &kgd2kfd)) {
+	ret = kgd2kfd_init(KFD_INTERFACE_VERSION, &kgd2kfd);
+	if (ret)
 		kgd2kfd = NULL;
 
-		return false;
-	}
-
-	return true;
 #else
-	return false;
+	ret = -ENOENT;
 #endif
+
+	return ret;
 }
 
 void radeon_kfd_fini(void)
diff --git a/drivers/gpu/drm/radeon/radeon_kfd.h b/drivers/gpu/drm/radeon/radeon_kfd.h
index 1103f9082f6b..9df1fea8e971 100644
--- a/drivers/gpu/drm/radeon/radeon_kfd.h
+++ b/drivers/gpu/drm/radeon/radeon_kfd.h
@@ -33,7 +33,7 @@
 
 struct radeon_device;
 
-bool radeon_kfd_init(void);
+int radeon_kfd_init(void);
 void radeon_kfd_fini(void);
 
 void radeon_kfd_suspend(struct radeon_device *rdev);
diff --git a/drivers/gpu/drm/radeon/radeon_legacy_crtc.c b/drivers/gpu/drm/radeon/radeon_legacy_crtc.c
index 32b338ff436b..24152dfef199 100644
--- a/drivers/gpu/drm/radeon/radeon_legacy_crtc.c
+++ b/drivers/gpu/drm/radeon/radeon_legacy_crtc.c
@@ -331,13 +331,13 @@ static void radeon_crtc_dpms(struct drm_crtc *crtc, int mode)
 									 RADEON_CRTC_DISP_REQ_EN_B));
 			WREG32_P(RADEON_CRTC_EXT_CNTL, crtc_ext_cntl, ~(mask | crtc_ext_cntl));
 		}
-		drm_vblank_post_modeset(dev, radeon_crtc->crtc_id);
+		drm_vblank_on(dev, radeon_crtc->crtc_id);
 		radeon_crtc_load_lut(crtc);
 		break;
 	case DRM_MODE_DPMS_STANDBY:
 	case DRM_MODE_DPMS_SUSPEND:
 	case DRM_MODE_DPMS_OFF:
-		drm_vblank_pre_modeset(dev, radeon_crtc->crtc_id);
+		drm_vblank_off(dev, radeon_crtc->crtc_id);
 		if (radeon_crtc->crtc_id)
 			WREG32_P(RADEON_CRTC2_GEN_CNTL, mask, ~(RADEON_CRTC2_EN | mask));
 		else {
diff --git a/drivers/gpu/drm/radeon/radeon_legacy_encoders.c b/drivers/gpu/drm/radeon/radeon_legacy_encoders.c
index 88dc973fb209..868c3ba2efaa 100644
--- a/drivers/gpu/drm/radeon/radeon_legacy_encoders.c
+++ b/drivers/gpu/drm/radeon/radeon_legacy_encoders.c
@@ -818,52 +818,52 @@ static void radeon_legacy_tmds_int_mode_set(struct drm_encoder *encoder,
 	tmds_transmitter_cntl = RREG32(RADEON_TMDS_TRANSMITTER_CNTL) &
 		~(RADEON_TMDS_TRANSMITTER_PLLRST);
 
-    if (rdev->family == CHIP_R200 ||
-	rdev->family == CHIP_R100 ||
-	ASIC_IS_R300(rdev))
-	    tmds_transmitter_cntl &= ~(RADEON_TMDS_TRANSMITTER_PLLEN);
-    else /* RV chips got this bit reversed */
-	    tmds_transmitter_cntl |= RADEON_TMDS_TRANSMITTER_PLLEN;
-
-    fp_gen_cntl = (RREG32(RADEON_FP_GEN_CNTL) |
-		   (RADEON_FP_CRTC_DONT_SHADOW_VPAR |
-		    RADEON_FP_CRTC_DONT_SHADOW_HEND));
-
-    fp_gen_cntl &= ~(RADEON_FP_FPON | RADEON_FP_TMDS_EN);
-
-    fp_gen_cntl &= ~(RADEON_FP_RMX_HVSYNC_CONTROL_EN |
-		     RADEON_FP_DFP_SYNC_SEL |
-		     RADEON_FP_CRT_SYNC_SEL |
-		     RADEON_FP_CRTC_LOCK_8DOT |
-		     RADEON_FP_USE_SHADOW_EN |
-		     RADEON_FP_CRTC_USE_SHADOW_VEND |
-		     RADEON_FP_CRT_SYNC_ALT);
-
-    if (1) /*  FIXME rgbBits == 8 */
-	    fp_gen_cntl |= RADEON_FP_PANEL_FORMAT;  /* 24 bit format */
-    else
-	    fp_gen_cntl &= ~RADEON_FP_PANEL_FORMAT;/* 18 bit format */
-
-    if (radeon_crtc->crtc_id == 0) {
-	    if (ASIC_IS_R300(rdev) || rdev->family == CHIP_R200) {
-		    fp_gen_cntl &= ~R200_FP_SOURCE_SEL_MASK;
-		    if (radeon_encoder->rmx_type != RMX_OFF)
-			    fp_gen_cntl |= R200_FP_SOURCE_SEL_RMX;
-		    else
-			    fp_gen_cntl |= R200_FP_SOURCE_SEL_CRTC1;
-	    } else
-		    fp_gen_cntl &= ~RADEON_FP_SEL_CRTC2;
-    } else {
-	    if (ASIC_IS_R300(rdev) || rdev->family == CHIP_R200) {
-		    fp_gen_cntl &= ~R200_FP_SOURCE_SEL_MASK;
-		    fp_gen_cntl |= R200_FP_SOURCE_SEL_CRTC2;
-	    } else
-		    fp_gen_cntl |= RADEON_FP_SEL_CRTC2;
-    }
-
-    WREG32(RADEON_TMDS_PLL_CNTL, tmds_pll_cntl);
-    WREG32(RADEON_TMDS_TRANSMITTER_CNTL, tmds_transmitter_cntl);
-    WREG32(RADEON_FP_GEN_CNTL, fp_gen_cntl);
+	if (rdev->family == CHIP_R200 ||
+	    rdev->family == CHIP_R100 ||
+	    ASIC_IS_R300(rdev))
+		tmds_transmitter_cntl &= ~(RADEON_TMDS_TRANSMITTER_PLLEN);
+	else /* RV chips got this bit reversed */
+		tmds_transmitter_cntl |= RADEON_TMDS_TRANSMITTER_PLLEN;
+
+	fp_gen_cntl = (RREG32(RADEON_FP_GEN_CNTL) |
+		      (RADEON_FP_CRTC_DONT_SHADOW_VPAR |
+		       RADEON_FP_CRTC_DONT_SHADOW_HEND));
+
+	fp_gen_cntl &= ~(RADEON_FP_FPON | RADEON_FP_TMDS_EN);
+
+	fp_gen_cntl &= ~(RADEON_FP_RMX_HVSYNC_CONTROL_EN |
+			 RADEON_FP_DFP_SYNC_SEL |
+			 RADEON_FP_CRT_SYNC_SEL |
+			 RADEON_FP_CRTC_LOCK_8DOT |
+			 RADEON_FP_USE_SHADOW_EN |
+			 RADEON_FP_CRTC_USE_SHADOW_VEND |
+			 RADEON_FP_CRT_SYNC_ALT);
+
+	if (1) /*  FIXME rgbBits == 8 */
+		fp_gen_cntl |= RADEON_FP_PANEL_FORMAT;  /* 24 bit format */
+	else
+		fp_gen_cntl &= ~RADEON_FP_PANEL_FORMAT;/* 18 bit format */
+
+	if (radeon_crtc->crtc_id == 0) {
+		if (ASIC_IS_R300(rdev) || rdev->family == CHIP_R200) {
+			fp_gen_cntl &= ~R200_FP_SOURCE_SEL_MASK;
+			if (radeon_encoder->rmx_type != RMX_OFF)
+				fp_gen_cntl |= R200_FP_SOURCE_SEL_RMX;
+			else
+				fp_gen_cntl |= R200_FP_SOURCE_SEL_CRTC1;
+		} else
+			fp_gen_cntl &= ~RADEON_FP_SEL_CRTC2;
+	} else {
+		if (ASIC_IS_R300(rdev) || rdev->family == CHIP_R200) {
+			fp_gen_cntl &= ~R200_FP_SOURCE_SEL_MASK;
+			fp_gen_cntl |= R200_FP_SOURCE_SEL_CRTC2;
+		} else
+			fp_gen_cntl |= RADEON_FP_SEL_CRTC2;
+	}
+
+	WREG32(RADEON_TMDS_PLL_CNTL, tmds_pll_cntl);
+	WREG32(RADEON_TMDS_TRANSMITTER_CNTL, tmds_transmitter_cntl);
+	WREG32(RADEON_FP_GEN_CNTL, fp_gen_cntl);
 
 	if (rdev->is_atom_bios)
 		radeon_atombios_encoder_crtc_scratch_regs(encoder, radeon_crtc->crtc_id);
diff --git a/drivers/gpu/drm/radeon/radeon_object.c b/drivers/gpu/drm/radeon/radeon_object.c
index fb6ad143873f..dd46c38676db 100644
--- a/drivers/gpu/drm/radeon/radeon_object.c
+++ b/drivers/gpu/drm/radeon/radeon_object.c
@@ -214,8 +214,8 @@ int radeon_bo_create(struct radeon_device *rdev,
 	INIT_LIST_HEAD(&bo->list);
 	INIT_LIST_HEAD(&bo->va);
 	bo->initial_domain = domain & (RADEON_GEM_DOMAIN_VRAM |
-	                               RADEON_GEM_DOMAIN_GTT |
-	                               RADEON_GEM_DOMAIN_CPU);
+				       RADEON_GEM_DOMAIN_GTT |
+				       RADEON_GEM_DOMAIN_CPU);
 
 	bo->flags = flags;
 	/* PCI GART is always snooped */
@@ -848,7 +848,7 @@ int radeon_bo_wait(struct radeon_bo *bo, u32 *mem_type, bool no_wait)
  *
  */
 void radeon_bo_fence(struct radeon_bo *bo, struct radeon_fence *fence,
-                     bool shared)
+		     bool shared)
 {
 	struct reservation_object *resv = bo->tbo.resv;
 
diff --git a/drivers/gpu/drm/radeon/radeon_pm.c b/drivers/gpu/drm/radeon/radeon_pm.c
index 460c8f2989da..38226d925a5b 100644
--- a/drivers/gpu/drm/radeon/radeon_pm.c
+++ b/drivers/gpu/drm/radeon/radeon_pm.c
@@ -79,7 +79,7 @@ void radeon_pm_acpi_event_handler(struct radeon_device *rdev)
 				radeon_dpm_enable_bapm(rdev, rdev->pm.dpm.ac_power);
 		}
 		mutex_unlock(&rdev->pm.mutex);
-        } else if (rdev->pm.pm_method == PM_METHOD_PROFILE) {
+	} else if (rdev->pm.pm_method == PM_METHOD_PROFILE) {
 		if (rdev->pm.profile == PM_PROFILE_AUTO) {
 			mutex_lock(&rdev->pm.mutex);
 			radeon_pm_update_profile(rdev);
@@ -276,8 +276,12 @@ static void radeon_pm_set_clocks(struct radeon_device *rdev)
 	if (rdev->irq.installed) {
 		for (i = 0; i < rdev->num_crtc; i++) {
 			if (rdev->pm.active_crtcs & (1 << i)) {
-				rdev->pm.req_vblank |= (1 << i);
-				drm_vblank_get(rdev->ddev, i);
+				/* This can fail if a modeset is in progress */
+				if (drm_vblank_get(rdev->ddev, i) == 0)
+					rdev->pm.req_vblank |= (1 << i);
+				else
+					DRM_DEBUG_DRIVER("crtc %d no vblank, can glitch\n",
+							 i);
 			}
 		}
 	}
@@ -1078,10 +1082,6 @@ force:
 	/* update displays */
 	radeon_dpm_display_configuration_changed(rdev);
 
-	rdev->pm.dpm.current_active_crtcs = rdev->pm.dpm.new_active_crtcs;
-	rdev->pm.dpm.current_active_crtc_count = rdev->pm.dpm.new_active_crtc_count;
-	rdev->pm.dpm.single_display = single_display;
-
 	/* wait for the rings to drain */
 	for (i = 0; i < RADEON_NUM_RINGS; i++) {
 		struct radeon_ring *ring = &rdev->ring[i];
@@ -1097,6 +1097,10 @@ force:
 
 	radeon_dpm_post_set_power_state(rdev);
 
+	rdev->pm.dpm.current_active_crtcs = rdev->pm.dpm.new_active_crtcs;
+	rdev->pm.dpm.current_active_crtc_count = rdev->pm.dpm.new_active_crtc_count;
+	rdev->pm.dpm.single_display = single_display;
+
 	if (rdev->asic->dpm.force_performance_level) {
 		if (rdev->pm.dpm.thermal_active) {
 			enum radeon_dpm_forced_level level = rdev->pm.dpm.forced_level;
diff --git a/drivers/gpu/drm/radeon/radeon_sa.c b/drivers/gpu/drm/radeon/radeon_sa.c
index c507896aca45..197b157b73d0 100644
--- a/drivers/gpu/drm/radeon/radeon_sa.c
+++ b/drivers/gpu/drm/radeon/radeon_sa.c
@@ -349,8 +349,13 @@ int radeon_sa_bo_new(struct radeon_device *rdev,
 			/* see if we can skip over some allocations */
 		} while (radeon_sa_bo_next_hole(sa_manager, fences, tries));
 
+		for (i = 0; i < RADEON_NUM_RINGS; ++i)
+			radeon_fence_ref(fences[i]);
+
 		spin_unlock(&sa_manager->wq.lock);
 		r = radeon_fence_wait_any(rdev, fences, false);
+		for (i = 0; i < RADEON_NUM_RINGS; ++i)
+			radeon_fence_unref(&fences[i]);
 		spin_lock(&sa_manager->wq.lock);
 		/* if we have nothing to wait for block */
 		if (r == -ENOENT) {
diff --git a/drivers/gpu/drm/radeon/radeon_semaphore.c b/drivers/gpu/drm/radeon/radeon_semaphore.c
index e6ad54cdfa62..b0eb28e8fb73 100644
--- a/drivers/gpu/drm/radeon/radeon_semaphore.c
+++ b/drivers/gpu/drm/radeon/radeon_semaphore.c
@@ -56,7 +56,7 @@ int radeon_semaphore_create(struct radeon_device *rdev,
 }
 
 bool radeon_semaphore_emit_signal(struct radeon_device *rdev, int ridx,
-			          struct radeon_semaphore *semaphore)
+				  struct radeon_semaphore *semaphore)
 {
 	struct radeon_ring *ring = &rdev->ring[ridx];
 
@@ -73,7 +73,7 @@ bool radeon_semaphore_emit_signal(struct radeon_device *rdev, int ridx,
 }
 
 bool radeon_semaphore_emit_wait(struct radeon_device *rdev, int ridx,
-			        struct radeon_semaphore *semaphore)
+				struct radeon_semaphore *semaphore)
 {
 	struct radeon_ring *ring = &rdev->ring[ridx];
 
diff --git a/drivers/gpu/drm/radeon/radeon_ttm.c b/drivers/gpu/drm/radeon/radeon_ttm.c
index e34307459e50..e06ac546a90f 100644
--- a/drivers/gpu/drm/radeon/radeon_ttm.c
+++ b/drivers/gpu/drm/radeon/radeon_ttm.c
@@ -758,7 +758,7 @@ static int radeon_ttm_tt_populate(struct ttm_tt *ttm)
 						       0, PAGE_SIZE,
 						       PCI_DMA_BIDIRECTIONAL);
 		if (pci_dma_mapping_error(rdev->pdev, gtt->ttm.dma_address[i])) {
-			while (--i) {
+			while (i--) {
 				pci_unmap_page(rdev->pdev, gtt->ttm.dma_address[i],
 					       PAGE_SIZE, PCI_DMA_BIDIRECTIONAL);
 				gtt->ttm.dma_address[i] = 0;
diff --git a/drivers/gpu/drm/radeon/radeon_uvd.c b/drivers/gpu/drm/radeon/radeon_uvd.c
index 6edcb5485092..6fe9e4e76284 100644
--- a/drivers/gpu/drm/radeon/radeon_uvd.c
+++ b/drivers/gpu/drm/radeon/radeon_uvd.c
@@ -722,9 +722,11 @@ static int radeon_uvd_send_msg(struct radeon_device *rdev,
 	return r;
 }
 
-/* multiple fence commands without any stream commands in between can
-   crash the vcpu so just try to emmit a dummy create/destroy msg to
-   avoid this */
+/*
+ * multiple fence commands without any stream commands in between can
+ * crash the vcpu so just try to emmit a dummy create/destroy msg to
+ * avoid this
+ */
 int radeon_uvd_get_create_msg(struct radeon_device *rdev, int ring,
 			      uint32_t handle, struct radeon_fence **fence)
 {
diff --git a/drivers/gpu/drm/radeon/radeon_vce.c b/drivers/gpu/drm/radeon/radeon_vce.c
index 566a1a01f6d1..c1c619facb47 100644
--- a/drivers/gpu/drm/radeon/radeon_vce.c
+++ b/drivers/gpu/drm/radeon/radeon_vce.c
@@ -166,7 +166,7 @@ int radeon_vce_init(struct radeon_device *rdev)
 	for (i = 0; i < RADEON_MAX_VCE_HANDLES; ++i) {
 		atomic_set(&rdev->vce.handles[i], 0);
 		rdev->vce.filp[i] = NULL;
-        }
+	}
 
 	return 0;
 }
@@ -389,7 +389,7 @@ int radeon_vce_get_create_msg(struct radeon_device *rdev, int ring,
 
 	r = radeon_ib_schedule(rdev, &ib, NULL, false);
 	if (r) {
-	        DRM_ERROR("radeon: failed to schedule ib (%d).\n", r);
+		DRM_ERROR("radeon: failed to schedule ib (%d).\n", r);
 	}
 
 	if (fence)
@@ -446,7 +446,7 @@ int radeon_vce_get_destroy_msg(struct radeon_device *rdev, int ring,
 
 	r = radeon_ib_schedule(rdev, &ib, NULL, false);
 	if (r) {
-	        DRM_ERROR("radeon: failed to schedule ib (%d).\n", r);
+		DRM_ERROR("radeon: failed to schedule ib (%d).\n", r);
 	}
 
 	if (fence)
@@ -769,18 +769,18 @@ int radeon_vce_ring_test(struct radeon_device *rdev, struct radeon_ring *ring)
 	radeon_ring_unlock_commit(rdev, ring, false);
 
 	for (i = 0; i < rdev->usec_timeout; i++) {
-	        if (vce_v1_0_get_rptr(rdev, ring) != rptr)
-	                break;
-	        DRM_UDELAY(1);
+		if (vce_v1_0_get_rptr(rdev, ring) != rptr)
+			break;
+		DRM_UDELAY(1);
 	}
 
 	if (i < rdev->usec_timeout) {
-	        DRM_INFO("ring test on %d succeeded in %d usecs\n",
-	                 ring->idx, i);
+		DRM_INFO("ring test on %d succeeded in %d usecs\n",
+			 ring->idx, i);
 	} else {
-	        DRM_ERROR("radeon: ring %d test failed\n",
-	                  ring->idx);
-	        r = -ETIMEDOUT;
+		DRM_ERROR("radeon: ring %d test failed\n",
+			 ring->idx);
+		r = -ETIMEDOUT;
 	}
 
 	return r;
diff --git a/drivers/gpu/drm/radeon/radeon_vm.c b/drivers/gpu/drm/radeon/radeon_vm.c
index 3979632b9225..a1358748cea5 100644
--- a/drivers/gpu/drm/radeon/radeon_vm.c
+++ b/drivers/gpu/drm/radeon/radeon_vm.c
@@ -611,15 +611,16 @@ uint64_t radeon_vm_map_gart(struct radeon_device *rdev, uint64_t addr)
  */
 static uint32_t radeon_vm_page_flags(uint32_t flags)
 {
-        uint32_t hw_flags = 0;
-        hw_flags |= (flags & RADEON_VM_PAGE_VALID) ? R600_PTE_VALID : 0;
-        hw_flags |= (flags & RADEON_VM_PAGE_READABLE) ? R600_PTE_READABLE : 0;
-        hw_flags |= (flags & RADEON_VM_PAGE_WRITEABLE) ? R600_PTE_WRITEABLE : 0;
-        if (flags & RADEON_VM_PAGE_SYSTEM) {
-                hw_flags |= R600_PTE_SYSTEM;
-                hw_flags |= (flags & RADEON_VM_PAGE_SNOOPED) ? R600_PTE_SNOOPED : 0;
-        }
-        return hw_flags;
+	uint32_t hw_flags = 0;
+
+	hw_flags |= (flags & RADEON_VM_PAGE_VALID) ? R600_PTE_VALID : 0;
+	hw_flags |= (flags & RADEON_VM_PAGE_READABLE) ? R600_PTE_READABLE : 0;
+	hw_flags |= (flags & RADEON_VM_PAGE_WRITEABLE) ? R600_PTE_WRITEABLE : 0;
+	if (flags & RADEON_VM_PAGE_SYSTEM) {
+		hw_flags |= R600_PTE_SYSTEM;
+		hw_flags |= (flags & RADEON_VM_PAGE_SNOOPED) ? R600_PTE_SNOOPED : 0;
+	}
+	return hw_flags;
 }
 
 /**
diff --git a/drivers/gpu/drm/radeon/rs780_dpm.c b/drivers/gpu/drm/radeon/rs780_dpm.c
index cb0afe78abed..94b48fc1e266 100644
--- a/drivers/gpu/drm/radeon/rs780_dpm.c
+++ b/drivers/gpu/drm/radeon/rs780_dpm.c
@@ -795,7 +795,7 @@ static int rs780_parse_power_table(struct radeon_device *rdev)
 	union pplib_clock_info *clock_info;
 	union power_info *power_info;
 	int index = GetIndexIntoMasterTable(DATA, PowerPlayInfo);
-        u16 data_offset;
+	u16 data_offset;
 	u8 frev, crev;
 	struct igp_ps *ps;
 
diff --git a/drivers/gpu/drm/radeon/rv6xx_dpm.c b/drivers/gpu/drm/radeon/rv6xx_dpm.c
index 97e5a6f1ce58..25e29303b119 100644
--- a/drivers/gpu/drm/radeon/rv6xx_dpm.c
+++ b/drivers/gpu/drm/radeon/rv6xx_dpm.c
@@ -209,7 +209,7 @@ static struct rv6xx_sclk_stepping rv6xx_next_vco_step(struct radeon_device *rdev
 
 static bool rv6xx_can_step_post_div(struct radeon_device *rdev,
 				    struct rv6xx_sclk_stepping *cur,
-                                    struct rv6xx_sclk_stepping *target)
+				    struct rv6xx_sclk_stepping *target)
 {
 	return (cur->post_divider > target->post_divider) &&
 		((cur->vco_frequency * target->post_divider) <=
@@ -239,7 +239,7 @@ static bool rv6xx_reached_stepping_target(struct radeon_device *rdev,
 
 static void rv6xx_generate_steps(struct radeon_device *rdev,
 				 u32 low, u32 high,
-                                 u32 start_index, u8 *end_index)
+				 u32 start_index, u8 *end_index)
 {
 	struct rv6xx_sclk_stepping cur;
 	struct rv6xx_sclk_stepping target;
@@ -1356,23 +1356,23 @@ static void rv6xx_set_dpm_event_sources(struct radeon_device *rdev, u32 sources)
 	enum radeon_dpm_event_src dpm_event_src;
 
 	switch (sources) {
-        case 0:
-        default:
+	case 0:
+	default:
 		want_thermal_protection = false;
 		break;
-        case (1 << RADEON_DPM_AUTO_THROTTLE_SRC_THERMAL):
+	case (1 << RADEON_DPM_AUTO_THROTTLE_SRC_THERMAL):
 		want_thermal_protection = true;
 		dpm_event_src = RADEON_DPM_EVENT_SRC_DIGITAL;
 		break;
 
-        case (1 << RADEON_DPM_AUTO_THROTTLE_SRC_EXTERNAL):
+	case (1 << RADEON_DPM_AUTO_THROTTLE_SRC_EXTERNAL):
 		want_thermal_protection = true;
 		dpm_event_src = RADEON_DPM_EVENT_SRC_EXTERNAL;
 		break;
 
-        case ((1 << RADEON_DPM_AUTO_THROTTLE_SRC_EXTERNAL) |
+	case ((1 << RADEON_DPM_AUTO_THROTTLE_SRC_EXTERNAL) |
 	      (1 << RADEON_DPM_AUTO_THROTTLE_SRC_THERMAL)):
-		want_thermal_protection = true;
+			want_thermal_protection = true;
 		dpm_event_src = RADEON_DPM_EVENT_SRC_DIGIAL_OR_EXTERNAL;
 		break;
 	}
@@ -1879,7 +1879,7 @@ static int rv6xx_parse_power_table(struct radeon_device *rdev)
 	union pplib_clock_info *clock_info;
 	union power_info *power_info;
 	int index = GetIndexIntoMasterTable(DATA, PowerPlayInfo);
-        u16 data_offset;
+	u16 data_offset;
 	u8 frev, crev;
 	struct rv6xx_ps *ps;
 
diff --git a/drivers/gpu/drm/radeon/rv740_dpm.c b/drivers/gpu/drm/radeon/rv740_dpm.c
index c4c8da501da8..4b850824fe06 100644
--- a/drivers/gpu/drm/radeon/rv740_dpm.c
+++ b/drivers/gpu/drm/radeon/rv740_dpm.c
@@ -36,28 +36,28 @@ u32 rv740_get_decoded_reference_divider(u32 encoded_ref)
 	u32 ref = 0;
 
 	switch (encoded_ref) {
-        case 0:
+	case 0:
 		ref = 1;
 		break;
-        case 16:
+	case 16:
 		ref = 2;
 		break;
-        case 17:
+	case 17:
 		ref = 3;
 		break;
-        case 18:
+	case 18:
 		ref = 2;
 		break;
-        case 19:
+	case 19:
 		ref = 3;
 		break;
-        case 20:
+	case 20:
 		ref = 4;
 		break;
-        case 21:
+	case 21:
 		ref = 5;
 		break;
-        default:
+	default:
 		DRM_ERROR("Invalid encoded Reference Divider\n");
 		ref = 0;
 		break;
diff --git a/drivers/gpu/drm/radeon/rv770_dpm.c b/drivers/gpu/drm/radeon/rv770_dpm.c
index e830c8935db0..a010decf59af 100644
--- a/drivers/gpu/drm/radeon/rv770_dpm.c
+++ b/drivers/gpu/drm/radeon/rv770_dpm.c
@@ -345,27 +345,27 @@ static int rv770_encode_yclk_post_div(u32 postdiv, u32 *encoded_postdiv)
 	int ret = 0;
 
 	switch (postdiv) {
-        case 1:
+	case 1:
 		*encoded_postdiv = 0;
 		break;
-        case 2:
+	case 2:
 		*encoded_postdiv = 1;
 		break;
-        case 4:
+	case 4:
 		*encoded_postdiv = 2;
 		break;
-        case 8:
+	case 8:
 		*encoded_postdiv = 3;
 		break;
-        case 16:
+	case 16:
 		*encoded_postdiv = 4;
 		break;
-        default:
+	default:
 		ret = -EINVAL;
 		break;
 	}
 
-    return ret;
+	return ret;
 }
 
 u32 rv770_map_clkf_to_ibias(struct radeon_device *rdev, u32 clkf)
@@ -1175,15 +1175,15 @@ static int rv770_init_smc_table(struct radeon_device *rdev,
 	rv770_populate_smc_mvdd_table(rdev, table);
 
 	switch (rdev->pm.int_thermal_type) {
-        case THERMAL_TYPE_RV770:
-        case THERMAL_TYPE_ADT7473_WITH_INTERNAL:
+	case THERMAL_TYPE_RV770:
+	case THERMAL_TYPE_ADT7473_WITH_INTERNAL:
 		table->thermalProtectType = PPSMC_THERMAL_PROTECT_TYPE_INTERNAL;
 		break;
-        case THERMAL_TYPE_NONE:
+	case THERMAL_TYPE_NONE:
 		table->thermalProtectType = PPSMC_THERMAL_PROTECT_TYPE_NONE;
 		break;
-        case THERMAL_TYPE_EXTERNAL_GPIO:
-        default:
+	case THERMAL_TYPE_EXTERNAL_GPIO:
+	default:
 		table->thermalProtectType = PPSMC_THERMAL_PROTECT_TYPE_EXTERNAL;
 		break;
 	}
@@ -1567,18 +1567,18 @@ void rv770_reset_smio_status(struct radeon_device *rdev)
 	sw_smio_index =
 		(RREG32(GENERAL_PWRMGT) & SW_SMIO_INDEX_MASK) >> SW_SMIO_INDEX_SHIFT;
 	switch (sw_smio_index) {
-        case 3:
+	case 3:
 		vid_smio_cntl = RREG32(S3_VID_LOWER_SMIO_CNTL);
 		break;
-        case 2:
+	case 2:
 		vid_smio_cntl = RREG32(S2_VID_LOWER_SMIO_CNTL);
 		break;
-        case 1:
+	case 1:
 		vid_smio_cntl = RREG32(S1_VID_LOWER_SMIO_CNTL);
 		break;
-        case 0:
+	case 0:
 		return;
-        default:
+	default:
 		vid_smio_cntl = pi->s0_vid_lower_smio_cntl;
 		break;
 	}
@@ -1817,21 +1817,21 @@ static void rv770_set_dpm_event_sources(struct radeon_device *rdev, u32 sources)
 	enum radeon_dpm_event_src dpm_event_src;
 
 	switch (sources) {
-        case 0:
-        default:
+	case 0:
+	default:
 		want_thermal_protection = false;
 		break;
-        case (1 << RADEON_DPM_AUTO_THROTTLE_SRC_THERMAL):
+	case (1 << RADEON_DPM_AUTO_THROTTLE_SRC_THERMAL):
 		want_thermal_protection = true;
 		dpm_event_src = RADEON_DPM_EVENT_SRC_DIGITAL;
 		break;
 
-        case (1 << RADEON_DPM_AUTO_THROTTLE_SRC_EXTERNAL):
+	case (1 << RADEON_DPM_AUTO_THROTTLE_SRC_EXTERNAL):
 		want_thermal_protection = true;
 		dpm_event_src = RADEON_DPM_EVENT_SRC_EXTERNAL;
 		break;
 
-        case ((1 << RADEON_DPM_AUTO_THROTTLE_SRC_EXTERNAL) |
+	case ((1 << RADEON_DPM_AUTO_THROTTLE_SRC_EXTERNAL) |
 	      (1 << RADEON_DPM_AUTO_THROTTLE_SRC_THERMAL)):
 		want_thermal_protection = true;
 		dpm_event_src = RADEON_DPM_EVENT_SRC_DIGIAL_OR_EXTERNAL;
@@ -2273,7 +2273,7 @@ int rv7xx_parse_power_table(struct radeon_device *rdev)
 	union pplib_clock_info *clock_info;
 	union power_info *power_info;
 	int index = GetIndexIntoMasterTable(DATA, PowerPlayInfo);
-        u16 data_offset;
+	u16 data_offset;
 	u8 frev, crev;
 	struct rv7xx_ps *ps;
 
diff --git a/drivers/gpu/drm/radeon/si.c b/drivers/gpu/drm/radeon/si.c
index f878d6962da5..ae21550fe767 100644
--- a/drivers/gpu/drm/radeon/si.c
+++ b/drivers/gpu/drm/radeon/si.c
@@ -1307,7 +1307,7 @@ int si_get_allowed_info_register(struct radeon_device *rdev,
  */
 u32 si_get_xclk(struct radeon_device *rdev)
 {
-        u32 reference_clock = rdev->clock.spll.reference_freq;
+	u32 reference_clock = rdev->clock.spll.reference_freq;
 	u32 tmp;
 
 	tmp = RREG32(CG_CLKPIN_CNTL_2);
@@ -2442,8 +2442,10 @@ void dce6_bandwidth_update(struct radeon_device *rdev)
  */
 static void si_tiling_mode_table_init(struct radeon_device *rdev)
 {
-	const u32 num_tile_mode_states = 32;
-	u32 reg_offset, gb_tile_moden, split_equal_to_row_size;
+	u32 *tile = rdev->config.si.tile_mode_array;
+	const u32 num_tile_mode_states =
+			ARRAY_SIZE(rdev->config.si.tile_mode_array);
+	u32 reg_offset, split_equal_to_row_size;
 
 	switch (rdev->config.si.mem_row_size_in_kb) {
 	case 1:
@@ -2458,491 +2460,442 @@ static void si_tiling_mode_table_init(struct radeon_device *rdev)
 		break;
 	}
 
-	if ((rdev->family == CHIP_TAHITI) ||
-	    (rdev->family == CHIP_PITCAIRN)) {
-		for (reg_offset = 0; reg_offset < num_tile_mode_states; reg_offset++) {
-			switch (reg_offset) {
-			case 0:  /* non-AA compressed depth or any compressed stencil */
-				gb_tile_moden = (ARRAY_MODE(ARRAY_2D_TILED_THIN1) |
-						 MICRO_TILE_MODE(ADDR_SURF_DEPTH_MICRO_TILING) |
-						 PIPE_CONFIG(ADDR_SURF_P8_32x32_8x16) |
-						 TILE_SPLIT(ADDR_SURF_TILE_SPLIT_64B) |
-						 NUM_BANKS(ADDR_SURF_16_BANK) |
-						 BANK_WIDTH(ADDR_SURF_BANK_WIDTH_1) |
-						 BANK_HEIGHT(ADDR_SURF_BANK_HEIGHT_4) |
-						 MACRO_TILE_ASPECT(ADDR_SURF_MACRO_ASPECT_2));
-				break;
-			case 1:  /* 2xAA/4xAA compressed depth only */
-				gb_tile_moden = (ARRAY_MODE(ARRAY_2D_TILED_THIN1) |
-						 MICRO_TILE_MODE(ADDR_SURF_DEPTH_MICRO_TILING) |
-						 PIPE_CONFIG(ADDR_SURF_P8_32x32_8x16) |
-						 TILE_SPLIT(ADDR_SURF_TILE_SPLIT_128B) |
-						 NUM_BANKS(ADDR_SURF_16_BANK) |
-						 BANK_WIDTH(ADDR_SURF_BANK_WIDTH_1) |
-						 BANK_HEIGHT(ADDR_SURF_BANK_HEIGHT_4) |
-						 MACRO_TILE_ASPECT(ADDR_SURF_MACRO_ASPECT_2));
-				break;
-			case 2:  /* 8xAA compressed depth only */
-				gb_tile_moden = (ARRAY_MODE(ARRAY_2D_TILED_THIN1) |
-						 MICRO_TILE_MODE(ADDR_SURF_DEPTH_MICRO_TILING) |
-						 PIPE_CONFIG(ADDR_SURF_P8_32x32_8x16) |
-						 TILE_SPLIT(ADDR_SURF_TILE_SPLIT_256B) |
-						 NUM_BANKS(ADDR_SURF_16_BANK) |
-						 BANK_WIDTH(ADDR_SURF_BANK_WIDTH_1) |
-						 BANK_HEIGHT(ADDR_SURF_BANK_HEIGHT_4) |
-						 MACRO_TILE_ASPECT(ADDR_SURF_MACRO_ASPECT_2));
-				break;
-			case 3:  /* 2xAA/4xAA compressed depth with stencil (for depth buffer) */
-				gb_tile_moden = (ARRAY_MODE(ARRAY_2D_TILED_THIN1) |
-						 MICRO_TILE_MODE(ADDR_SURF_DEPTH_MICRO_TILING) |
-						 PIPE_CONFIG(ADDR_SURF_P8_32x32_8x16) |
-						 TILE_SPLIT(ADDR_SURF_TILE_SPLIT_128B) |
-						 NUM_BANKS(ADDR_SURF_16_BANK) |
-						 BANK_WIDTH(ADDR_SURF_BANK_WIDTH_1) |
-						 BANK_HEIGHT(ADDR_SURF_BANK_HEIGHT_4) |
-						 MACRO_TILE_ASPECT(ADDR_SURF_MACRO_ASPECT_2));
-				break;
-			case 4:  /* Maps w/ a dimension less than the 2D macro-tile dimensions (for mipmapped depth textures) */
-				gb_tile_moden = (ARRAY_MODE(ARRAY_1D_TILED_THIN1) |
-						 MICRO_TILE_MODE(ADDR_SURF_DEPTH_MICRO_TILING) |
-						 PIPE_CONFIG(ADDR_SURF_P8_32x32_8x16) |
-						 TILE_SPLIT(ADDR_SURF_TILE_SPLIT_64B) |
-						 NUM_BANKS(ADDR_SURF_16_BANK) |
-						 BANK_WIDTH(ADDR_SURF_BANK_WIDTH_1) |
-						 BANK_HEIGHT(ADDR_SURF_BANK_HEIGHT_2) |
-						 MACRO_TILE_ASPECT(ADDR_SURF_MACRO_ASPECT_2));
-				break;
-			case 5:  /* Uncompressed 16bpp depth - and stencil buffer allocated with it */
-				gb_tile_moden = (ARRAY_MODE(ARRAY_2D_TILED_THIN1) |
-						 MICRO_TILE_MODE(ADDR_SURF_DEPTH_MICRO_TILING) |
-						 PIPE_CONFIG(ADDR_SURF_P8_32x32_8x16) |
-						 TILE_SPLIT(split_equal_to_row_size) |
-						 NUM_BANKS(ADDR_SURF_16_BANK) |
-						 BANK_WIDTH(ADDR_SURF_BANK_WIDTH_1) |
-						 BANK_HEIGHT(ADDR_SURF_BANK_HEIGHT_2) |
-						 MACRO_TILE_ASPECT(ADDR_SURF_MACRO_ASPECT_2));
-				break;
-			case 6:  /* Uncompressed 32bpp depth - and stencil buffer allocated with it */
-				gb_tile_moden = (ARRAY_MODE(ARRAY_2D_TILED_THIN1) |
-						 MICRO_TILE_MODE(ADDR_SURF_DEPTH_MICRO_TILING) |
-						 PIPE_CONFIG(ADDR_SURF_P8_32x32_8x16) |
-						 TILE_SPLIT(split_equal_to_row_size) |
-						 NUM_BANKS(ADDR_SURF_16_BANK) |
-						 BANK_WIDTH(ADDR_SURF_BANK_WIDTH_1) |
-						 BANK_HEIGHT(ADDR_SURF_BANK_HEIGHT_1) |
-						 MACRO_TILE_ASPECT(ADDR_SURF_MACRO_ASPECT_1));
-				break;
-			case 7:  /* Uncompressed 8bpp stencil without depth (drivers typically do not use) */
-				gb_tile_moden = (ARRAY_MODE(ARRAY_2D_TILED_THIN1) |
-						 MICRO_TILE_MODE(ADDR_SURF_DEPTH_MICRO_TILING) |
-						 PIPE_CONFIG(ADDR_SURF_P8_32x32_8x16) |
-						 TILE_SPLIT(split_equal_to_row_size) |
-						 NUM_BANKS(ADDR_SURF_16_BANK) |
-						 BANK_WIDTH(ADDR_SURF_BANK_WIDTH_1) |
-						 BANK_HEIGHT(ADDR_SURF_BANK_HEIGHT_4) |
-						 MACRO_TILE_ASPECT(ADDR_SURF_MACRO_ASPECT_2));
-				break;
-			case 8:  /* 1D and 1D Array Surfaces */
-				gb_tile_moden = (ARRAY_MODE(ARRAY_LINEAR_ALIGNED) |
-						 MICRO_TILE_MODE(ADDR_SURF_DISPLAY_MICRO_TILING) |
-						 PIPE_CONFIG(ADDR_SURF_P8_32x32_8x16) |
-						 TILE_SPLIT(ADDR_SURF_TILE_SPLIT_64B) |
-						 NUM_BANKS(ADDR_SURF_16_BANK) |
-						 BANK_WIDTH(ADDR_SURF_BANK_WIDTH_1) |
-						 BANK_HEIGHT(ADDR_SURF_BANK_HEIGHT_2) |
-						 MACRO_TILE_ASPECT(ADDR_SURF_MACRO_ASPECT_2));
-				break;
-			case 9:  /* Displayable maps. */
-				gb_tile_moden = (ARRAY_MODE(ARRAY_1D_TILED_THIN1) |
-						 MICRO_TILE_MODE(ADDR_SURF_DISPLAY_MICRO_TILING) |
-						 PIPE_CONFIG(ADDR_SURF_P8_32x32_8x16) |
-						 TILE_SPLIT(ADDR_SURF_TILE_SPLIT_64B) |
-						 NUM_BANKS(ADDR_SURF_16_BANK) |
-						 BANK_WIDTH(ADDR_SURF_BANK_WIDTH_1) |
-						 BANK_HEIGHT(ADDR_SURF_BANK_HEIGHT_2) |
-						 MACRO_TILE_ASPECT(ADDR_SURF_MACRO_ASPECT_2));
-				break;
-			case 10:  /* Display 8bpp. */
-				gb_tile_moden = (ARRAY_MODE(ARRAY_2D_TILED_THIN1) |
-						 MICRO_TILE_MODE(ADDR_SURF_DISPLAY_MICRO_TILING) |
-						 PIPE_CONFIG(ADDR_SURF_P8_32x32_8x16) |
-						 TILE_SPLIT(ADDR_SURF_TILE_SPLIT_256B) |
-						 NUM_BANKS(ADDR_SURF_16_BANK) |
-						 BANK_WIDTH(ADDR_SURF_BANK_WIDTH_1) |
-						 BANK_HEIGHT(ADDR_SURF_BANK_HEIGHT_4) |
-						 MACRO_TILE_ASPECT(ADDR_SURF_MACRO_ASPECT_2));
-				break;
-			case 11:  /* Display 16bpp. */
-				gb_tile_moden = (ARRAY_MODE(ARRAY_2D_TILED_THIN1) |
-						 MICRO_TILE_MODE(ADDR_SURF_DISPLAY_MICRO_TILING) |
-						 PIPE_CONFIG(ADDR_SURF_P8_32x32_8x16) |
-						 TILE_SPLIT(ADDR_SURF_TILE_SPLIT_256B) |
-						 NUM_BANKS(ADDR_SURF_16_BANK) |
-						 BANK_WIDTH(ADDR_SURF_BANK_WIDTH_1) |
-						 BANK_HEIGHT(ADDR_SURF_BANK_HEIGHT_2) |
-						 MACRO_TILE_ASPECT(ADDR_SURF_MACRO_ASPECT_2));
-				break;
-			case 12:  /* Display 32bpp. */
-				gb_tile_moden = (ARRAY_MODE(ARRAY_2D_TILED_THIN1) |
-						 MICRO_TILE_MODE(ADDR_SURF_DISPLAY_MICRO_TILING) |
-						 PIPE_CONFIG(ADDR_SURF_P8_32x32_8x16) |
-						 TILE_SPLIT(ADDR_SURF_TILE_SPLIT_512B) |
-						 NUM_BANKS(ADDR_SURF_16_BANK) |
-						 BANK_WIDTH(ADDR_SURF_BANK_WIDTH_1) |
-						 BANK_HEIGHT(ADDR_SURF_BANK_HEIGHT_1) |
-						 MACRO_TILE_ASPECT(ADDR_SURF_MACRO_ASPECT_1));
-				break;
-			case 13:  /* Thin. */
-				gb_tile_moden = (ARRAY_MODE(ARRAY_1D_TILED_THIN1) |
-						 MICRO_TILE_MODE(ADDR_SURF_THIN_MICRO_TILING) |
-						 PIPE_CONFIG(ADDR_SURF_P8_32x32_8x16) |
-						 TILE_SPLIT(ADDR_SURF_TILE_SPLIT_64B) |
-						 NUM_BANKS(ADDR_SURF_16_BANK) |
-						 BANK_WIDTH(ADDR_SURF_BANK_WIDTH_1) |
-						 BANK_HEIGHT(ADDR_SURF_BANK_HEIGHT_2) |
-						 MACRO_TILE_ASPECT(ADDR_SURF_MACRO_ASPECT_2));
-				break;
-			case 14:  /* Thin 8 bpp. */
-				gb_tile_moden = (ARRAY_MODE(ARRAY_2D_TILED_THIN1) |
-						 MICRO_TILE_MODE(ADDR_SURF_THIN_MICRO_TILING) |
-						 PIPE_CONFIG(ADDR_SURF_P8_32x32_8x16) |
-						 TILE_SPLIT(ADDR_SURF_TILE_SPLIT_256B) |
-						 NUM_BANKS(ADDR_SURF_16_BANK) |
-						 BANK_WIDTH(ADDR_SURF_BANK_WIDTH_1) |
-						 BANK_HEIGHT(ADDR_SURF_BANK_HEIGHT_4) |
-						 MACRO_TILE_ASPECT(ADDR_SURF_MACRO_ASPECT_1));
-				break;
-			case 15:  /* Thin 16 bpp. */
-				gb_tile_moden = (ARRAY_MODE(ARRAY_2D_TILED_THIN1) |
-						 MICRO_TILE_MODE(ADDR_SURF_THIN_MICRO_TILING) |
-						 PIPE_CONFIG(ADDR_SURF_P8_32x32_8x16) |
-						 TILE_SPLIT(ADDR_SURF_TILE_SPLIT_256B) |
-						 NUM_BANKS(ADDR_SURF_16_BANK) |
-						 BANK_WIDTH(ADDR_SURF_BANK_WIDTH_1) |
-						 BANK_HEIGHT(ADDR_SURF_BANK_HEIGHT_2) |
-						 MACRO_TILE_ASPECT(ADDR_SURF_MACRO_ASPECT_1));
-				break;
-			case 16:  /* Thin 32 bpp. */
-				gb_tile_moden = (ARRAY_MODE(ARRAY_2D_TILED_THIN1) |
-						 MICRO_TILE_MODE(ADDR_SURF_THIN_MICRO_TILING) |
-						 PIPE_CONFIG(ADDR_SURF_P8_32x32_8x16) |
-						 TILE_SPLIT(ADDR_SURF_TILE_SPLIT_512B) |
-						 NUM_BANKS(ADDR_SURF_16_BANK) |
-						 BANK_WIDTH(ADDR_SURF_BANK_WIDTH_1) |
-						 BANK_HEIGHT(ADDR_SURF_BANK_HEIGHT_1) |
-						 MACRO_TILE_ASPECT(ADDR_SURF_MACRO_ASPECT_1));
-				break;
-			case 17:  /* Thin 64 bpp. */
-				gb_tile_moden = (ARRAY_MODE(ARRAY_2D_TILED_THIN1) |
-						 MICRO_TILE_MODE(ADDR_SURF_THIN_MICRO_TILING) |
-						 PIPE_CONFIG(ADDR_SURF_P8_32x32_8x16) |
-						 TILE_SPLIT(split_equal_to_row_size) |
-						 NUM_BANKS(ADDR_SURF_16_BANK) |
-						 BANK_WIDTH(ADDR_SURF_BANK_WIDTH_1) |
-						 BANK_HEIGHT(ADDR_SURF_BANK_HEIGHT_1) |
-						 MACRO_TILE_ASPECT(ADDR_SURF_MACRO_ASPECT_1));
-				break;
-			case 21:  /* 8 bpp PRT. */
-				gb_tile_moden = (ARRAY_MODE(ARRAY_2D_TILED_THIN1) |
-						 MICRO_TILE_MODE(ADDR_SURF_THIN_MICRO_TILING) |
-						 PIPE_CONFIG(ADDR_SURF_P8_32x32_8x16) |
-						 TILE_SPLIT(ADDR_SURF_TILE_SPLIT_256B) |
-						 NUM_BANKS(ADDR_SURF_16_BANK) |
-						 BANK_WIDTH(ADDR_SURF_BANK_WIDTH_2) |
-						 BANK_HEIGHT(ADDR_SURF_BANK_HEIGHT_4) |
-						 MACRO_TILE_ASPECT(ADDR_SURF_MACRO_ASPECT_2));
-				break;
-			case 22:  /* 16 bpp PRT */
-				gb_tile_moden = (ARRAY_MODE(ARRAY_2D_TILED_THIN1) |
-						 MICRO_TILE_MODE(ADDR_SURF_THIN_MICRO_TILING) |
-						 PIPE_CONFIG(ADDR_SURF_P8_32x32_8x16) |
-						 TILE_SPLIT(ADDR_SURF_TILE_SPLIT_256B) |
-						 NUM_BANKS(ADDR_SURF_16_BANK) |
-						 BANK_WIDTH(ADDR_SURF_BANK_WIDTH_1) |
-						 BANK_HEIGHT(ADDR_SURF_BANK_HEIGHT_4) |
-						 MACRO_TILE_ASPECT(ADDR_SURF_MACRO_ASPECT_4));
-				break;
-			case 23:  /* 32 bpp PRT */
-				gb_tile_moden = (ARRAY_MODE(ARRAY_2D_TILED_THIN1) |
-						 MICRO_TILE_MODE(ADDR_SURF_THIN_MICRO_TILING) |
-						 PIPE_CONFIG(ADDR_SURF_P8_32x32_8x16) |
-						 TILE_SPLIT(ADDR_SURF_TILE_SPLIT_256B) |
-						 NUM_BANKS(ADDR_SURF_16_BANK) |
-						 BANK_WIDTH(ADDR_SURF_BANK_WIDTH_1) |
-						 BANK_HEIGHT(ADDR_SURF_BANK_HEIGHT_2) |
-						 MACRO_TILE_ASPECT(ADDR_SURF_MACRO_ASPECT_2));
-				break;
-			case 24:  /* 64 bpp PRT */
-				gb_tile_moden = (ARRAY_MODE(ARRAY_2D_TILED_THIN1) |
-						 MICRO_TILE_MODE(ADDR_SURF_THIN_MICRO_TILING) |
-						 PIPE_CONFIG(ADDR_SURF_P8_32x32_8x16) |
-						 TILE_SPLIT(ADDR_SURF_TILE_SPLIT_512B) |
-						 NUM_BANKS(ADDR_SURF_16_BANK) |
-						 BANK_WIDTH(ADDR_SURF_BANK_WIDTH_1) |
-						 BANK_HEIGHT(ADDR_SURF_BANK_HEIGHT_1) |
-						 MACRO_TILE_ASPECT(ADDR_SURF_MACRO_ASPECT_2));
-				break;
-			case 25:  /* 128 bpp PRT */
-				gb_tile_moden = (ARRAY_MODE(ARRAY_2D_TILED_THIN1) |
-						 MICRO_TILE_MODE(ADDR_SURF_THIN_MICRO_TILING) |
-						 PIPE_CONFIG(ADDR_SURF_P8_32x32_8x16) |
-						 TILE_SPLIT(ADDR_SURF_TILE_SPLIT_1KB) |
-						 NUM_BANKS(ADDR_SURF_8_BANK) |
-						 BANK_WIDTH(ADDR_SURF_BANK_WIDTH_1) |
-						 BANK_HEIGHT(ADDR_SURF_BANK_HEIGHT_1) |
-						 MACRO_TILE_ASPECT(ADDR_SURF_MACRO_ASPECT_1));
-				break;
-			default:
-				gb_tile_moden = 0;
-				break;
-			}
-			rdev->config.si.tile_mode_array[reg_offset] = gb_tile_moden;
-			WREG32(GB_TILE_MODE0 + (reg_offset * 4), gb_tile_moden);
-		}
-	} else if ((rdev->family == CHIP_VERDE) ||
-		   (rdev->family == CHIP_OLAND) ||
-		   (rdev->family == CHIP_HAINAN)) {
-		for (reg_offset = 0; reg_offset < num_tile_mode_states; reg_offset++) {
-			switch (reg_offset) {
-			case 0:  /* non-AA compressed depth or any compressed stencil */
-				gb_tile_moden = (ARRAY_MODE(ARRAY_2D_TILED_THIN1) |
-						 MICRO_TILE_MODE(ADDR_SURF_DEPTH_MICRO_TILING) |
-						 PIPE_CONFIG(ADDR_SURF_P4_8x16) |
-						 TILE_SPLIT(ADDR_SURF_TILE_SPLIT_64B) |
-						 NUM_BANKS(ADDR_SURF_16_BANK) |
-						 BANK_WIDTH(ADDR_SURF_BANK_WIDTH_1) |
-						 BANK_HEIGHT(ADDR_SURF_BANK_HEIGHT_4) |
-						 MACRO_TILE_ASPECT(ADDR_SURF_MACRO_ASPECT_4));
-				break;
-			case 1:  /* 2xAA/4xAA compressed depth only */
-				gb_tile_moden = (ARRAY_MODE(ARRAY_2D_TILED_THIN1) |
-						 MICRO_TILE_MODE(ADDR_SURF_DEPTH_MICRO_TILING) |
-						 PIPE_CONFIG(ADDR_SURF_P4_8x16) |
-						 TILE_SPLIT(ADDR_SURF_TILE_SPLIT_128B) |
-						 NUM_BANKS(ADDR_SURF_16_BANK) |
-						 BANK_WIDTH(ADDR_SURF_BANK_WIDTH_1) |
-						 BANK_HEIGHT(ADDR_SURF_BANK_HEIGHT_4) |
-						 MACRO_TILE_ASPECT(ADDR_SURF_MACRO_ASPECT_4));
-				break;
-			case 2:  /* 8xAA compressed depth only */
-				gb_tile_moden = (ARRAY_MODE(ARRAY_2D_TILED_THIN1) |
-						 MICRO_TILE_MODE(ADDR_SURF_DEPTH_MICRO_TILING) |
-						 PIPE_CONFIG(ADDR_SURF_P4_8x16) |
-						 TILE_SPLIT(ADDR_SURF_TILE_SPLIT_256B) |
-						 NUM_BANKS(ADDR_SURF_16_BANK) |
-						 BANK_WIDTH(ADDR_SURF_BANK_WIDTH_1) |
-						 BANK_HEIGHT(ADDR_SURF_BANK_HEIGHT_4) |
-						 MACRO_TILE_ASPECT(ADDR_SURF_MACRO_ASPECT_4));
-				break;
-			case 3:  /* 2xAA/4xAA compressed depth with stencil (for depth buffer) */
-				gb_tile_moden = (ARRAY_MODE(ARRAY_2D_TILED_THIN1) |
-						 MICRO_TILE_MODE(ADDR_SURF_DEPTH_MICRO_TILING) |
-						 PIPE_CONFIG(ADDR_SURF_P4_8x16) |
-						 TILE_SPLIT(ADDR_SURF_TILE_SPLIT_128B) |
-						 NUM_BANKS(ADDR_SURF_16_BANK) |
-						 BANK_WIDTH(ADDR_SURF_BANK_WIDTH_1) |
-						 BANK_HEIGHT(ADDR_SURF_BANK_HEIGHT_4) |
-						 MACRO_TILE_ASPECT(ADDR_SURF_MACRO_ASPECT_4));
-				break;
-			case 4:  /* Maps w/ a dimension less than the 2D macro-tile dimensions (for mipmapped depth textures) */
-				gb_tile_moden = (ARRAY_MODE(ARRAY_1D_TILED_THIN1) |
-						 MICRO_TILE_MODE(ADDR_SURF_DEPTH_MICRO_TILING) |
-						 PIPE_CONFIG(ADDR_SURF_P4_8x16) |
-						 TILE_SPLIT(ADDR_SURF_TILE_SPLIT_64B) |
-						 NUM_BANKS(ADDR_SURF_16_BANK) |
-						 BANK_WIDTH(ADDR_SURF_BANK_WIDTH_1) |
-						 BANK_HEIGHT(ADDR_SURF_BANK_HEIGHT_2) |
-						 MACRO_TILE_ASPECT(ADDR_SURF_MACRO_ASPECT_2));
-				break;
-			case 5:  /* Uncompressed 16bpp depth - and stencil buffer allocated with it */
-				gb_tile_moden = (ARRAY_MODE(ARRAY_2D_TILED_THIN1) |
-						 MICRO_TILE_MODE(ADDR_SURF_DEPTH_MICRO_TILING) |
-						 PIPE_CONFIG(ADDR_SURF_P4_8x16) |
-						 TILE_SPLIT(split_equal_to_row_size) |
-						 NUM_BANKS(ADDR_SURF_16_BANK) |
-						 BANK_WIDTH(ADDR_SURF_BANK_WIDTH_1) |
-						 BANK_HEIGHT(ADDR_SURF_BANK_HEIGHT_2) |
-						 MACRO_TILE_ASPECT(ADDR_SURF_MACRO_ASPECT_2));
-				break;
-			case 6:  /* Uncompressed 32bpp depth - and stencil buffer allocated with it */
-				gb_tile_moden = (ARRAY_MODE(ARRAY_2D_TILED_THIN1) |
-						 MICRO_TILE_MODE(ADDR_SURF_DEPTH_MICRO_TILING) |
-						 PIPE_CONFIG(ADDR_SURF_P4_8x16) |
-						 TILE_SPLIT(split_equal_to_row_size) |
-						 NUM_BANKS(ADDR_SURF_16_BANK) |
-						 BANK_WIDTH(ADDR_SURF_BANK_WIDTH_1) |
-						 BANK_HEIGHT(ADDR_SURF_BANK_HEIGHT_1) |
-						 MACRO_TILE_ASPECT(ADDR_SURF_MACRO_ASPECT_2));
-				break;
-			case 7:  /* Uncompressed 8bpp stencil without depth (drivers typically do not use) */
-				gb_tile_moden = (ARRAY_MODE(ARRAY_2D_TILED_THIN1) |
-						 MICRO_TILE_MODE(ADDR_SURF_DEPTH_MICRO_TILING) |
-						 PIPE_CONFIG(ADDR_SURF_P4_8x16) |
-						 TILE_SPLIT(split_equal_to_row_size) |
-						 NUM_BANKS(ADDR_SURF_16_BANK) |
-						 BANK_WIDTH(ADDR_SURF_BANK_WIDTH_1) |
-						 BANK_HEIGHT(ADDR_SURF_BANK_HEIGHT_4) |
-						 MACRO_TILE_ASPECT(ADDR_SURF_MACRO_ASPECT_4));
-				break;
-			case 8:  /* 1D and 1D Array Surfaces */
-				gb_tile_moden = (ARRAY_MODE(ARRAY_LINEAR_ALIGNED) |
-						 MICRO_TILE_MODE(ADDR_SURF_DISPLAY_MICRO_TILING) |
-						 PIPE_CONFIG(ADDR_SURF_P4_8x16) |
-						 TILE_SPLIT(ADDR_SURF_TILE_SPLIT_64B) |
-						 NUM_BANKS(ADDR_SURF_16_BANK) |
-						 BANK_WIDTH(ADDR_SURF_BANK_WIDTH_1) |
-						 BANK_HEIGHT(ADDR_SURF_BANK_HEIGHT_2) |
-						 MACRO_TILE_ASPECT(ADDR_SURF_MACRO_ASPECT_2));
-				break;
-			case 9:  /* Displayable maps. */
-				gb_tile_moden = (ARRAY_MODE(ARRAY_1D_TILED_THIN1) |
-						 MICRO_TILE_MODE(ADDR_SURF_DISPLAY_MICRO_TILING) |
-						 PIPE_CONFIG(ADDR_SURF_P4_8x16) |
-						 TILE_SPLIT(ADDR_SURF_TILE_SPLIT_64B) |
-						 NUM_BANKS(ADDR_SURF_16_BANK) |
-						 BANK_WIDTH(ADDR_SURF_BANK_WIDTH_1) |
-						 BANK_HEIGHT(ADDR_SURF_BANK_HEIGHT_2) |
-						 MACRO_TILE_ASPECT(ADDR_SURF_MACRO_ASPECT_2));
-				break;
-			case 10:  /* Display 8bpp. */
-				gb_tile_moden = (ARRAY_MODE(ARRAY_2D_TILED_THIN1) |
-						 MICRO_TILE_MODE(ADDR_SURF_DISPLAY_MICRO_TILING) |
-						 PIPE_CONFIG(ADDR_SURF_P4_8x16) |
-						 TILE_SPLIT(ADDR_SURF_TILE_SPLIT_256B) |
-						 NUM_BANKS(ADDR_SURF_16_BANK) |
-						 BANK_WIDTH(ADDR_SURF_BANK_WIDTH_1) |
-						 BANK_HEIGHT(ADDR_SURF_BANK_HEIGHT_4) |
-						 MACRO_TILE_ASPECT(ADDR_SURF_MACRO_ASPECT_4));
-				break;
-			case 11:  /* Display 16bpp. */
-				gb_tile_moden = (ARRAY_MODE(ARRAY_2D_TILED_THIN1) |
-						 MICRO_TILE_MODE(ADDR_SURF_DISPLAY_MICRO_TILING) |
-						 PIPE_CONFIG(ADDR_SURF_P4_8x16) |
-						 TILE_SPLIT(ADDR_SURF_TILE_SPLIT_256B) |
-						 NUM_BANKS(ADDR_SURF_16_BANK) |
-						 BANK_WIDTH(ADDR_SURF_BANK_WIDTH_1) |
-						 BANK_HEIGHT(ADDR_SURF_BANK_HEIGHT_2) |
-						 MACRO_TILE_ASPECT(ADDR_SURF_MACRO_ASPECT_2));
-				break;
-			case 12:  /* Display 32bpp. */
-				gb_tile_moden = (ARRAY_MODE(ARRAY_2D_TILED_THIN1) |
-						 MICRO_TILE_MODE(ADDR_SURF_DISPLAY_MICRO_TILING) |
-						 PIPE_CONFIG(ADDR_SURF_P4_8x16) |
-						 TILE_SPLIT(ADDR_SURF_TILE_SPLIT_512B) |
-						 NUM_BANKS(ADDR_SURF_16_BANK) |
-						 BANK_WIDTH(ADDR_SURF_BANK_WIDTH_1) |
-						 BANK_HEIGHT(ADDR_SURF_BANK_HEIGHT_1) |
-						 MACRO_TILE_ASPECT(ADDR_SURF_MACRO_ASPECT_2));
-				break;
-			case 13:  /* Thin. */
-				gb_tile_moden = (ARRAY_MODE(ARRAY_1D_TILED_THIN1) |
-						 MICRO_TILE_MODE(ADDR_SURF_THIN_MICRO_TILING) |
-						 PIPE_CONFIG(ADDR_SURF_P4_8x16) |
-						 TILE_SPLIT(ADDR_SURF_TILE_SPLIT_64B) |
-						 NUM_BANKS(ADDR_SURF_16_BANK) |
-						 BANK_WIDTH(ADDR_SURF_BANK_WIDTH_1) |
-						 BANK_HEIGHT(ADDR_SURF_BANK_HEIGHT_2) |
-						 MACRO_TILE_ASPECT(ADDR_SURF_MACRO_ASPECT_2));
-				break;
-			case 14:  /* Thin 8 bpp. */
-				gb_tile_moden = (ARRAY_MODE(ARRAY_2D_TILED_THIN1) |
-						 MICRO_TILE_MODE(ADDR_SURF_THIN_MICRO_TILING) |
-						 PIPE_CONFIG(ADDR_SURF_P4_8x16) |
-						 TILE_SPLIT(ADDR_SURF_TILE_SPLIT_256B) |
-						 NUM_BANKS(ADDR_SURF_16_BANK) |
-						 BANK_WIDTH(ADDR_SURF_BANK_WIDTH_1) |
-						 BANK_HEIGHT(ADDR_SURF_BANK_HEIGHT_4) |
-						 MACRO_TILE_ASPECT(ADDR_SURF_MACRO_ASPECT_2));
-				break;
-			case 15:  /* Thin 16 bpp. */
-				gb_tile_moden = (ARRAY_MODE(ARRAY_2D_TILED_THIN1) |
-						 MICRO_TILE_MODE(ADDR_SURF_THIN_MICRO_TILING) |
-						 PIPE_CONFIG(ADDR_SURF_P4_8x16) |
-						 TILE_SPLIT(ADDR_SURF_TILE_SPLIT_256B) |
-						 NUM_BANKS(ADDR_SURF_16_BANK) |
-						 BANK_WIDTH(ADDR_SURF_BANK_WIDTH_1) |
-						 BANK_HEIGHT(ADDR_SURF_BANK_HEIGHT_2) |
-						 MACRO_TILE_ASPECT(ADDR_SURF_MACRO_ASPECT_2));
-				break;
-			case 16:  /* Thin 32 bpp. */
-				gb_tile_moden = (ARRAY_MODE(ARRAY_2D_TILED_THIN1) |
-						 MICRO_TILE_MODE(ADDR_SURF_THIN_MICRO_TILING) |
-						 PIPE_CONFIG(ADDR_SURF_P4_8x16) |
-						 TILE_SPLIT(ADDR_SURF_TILE_SPLIT_512B) |
-						 NUM_BANKS(ADDR_SURF_16_BANK) |
-						 BANK_WIDTH(ADDR_SURF_BANK_WIDTH_1) |
-						 BANK_HEIGHT(ADDR_SURF_BANK_HEIGHT_1) |
-						 MACRO_TILE_ASPECT(ADDR_SURF_MACRO_ASPECT_2));
-				break;
-			case 17:  /* Thin 64 bpp. */
-				gb_tile_moden = (ARRAY_MODE(ARRAY_2D_TILED_THIN1) |
-						 MICRO_TILE_MODE(ADDR_SURF_THIN_MICRO_TILING) |
-						 PIPE_CONFIG(ADDR_SURF_P4_8x16) |
-						 TILE_SPLIT(split_equal_to_row_size) |
-						 NUM_BANKS(ADDR_SURF_16_BANK) |
-						 BANK_WIDTH(ADDR_SURF_BANK_WIDTH_1) |
-						 BANK_HEIGHT(ADDR_SURF_BANK_HEIGHT_1) |
-						 MACRO_TILE_ASPECT(ADDR_SURF_MACRO_ASPECT_2));
-				break;
-			case 21:  /* 8 bpp PRT. */
-				gb_tile_moden = (ARRAY_MODE(ARRAY_2D_TILED_THIN1) |
-						 MICRO_TILE_MODE(ADDR_SURF_THIN_MICRO_TILING) |
-						 PIPE_CONFIG(ADDR_SURF_P8_32x32_8x16) |
-						 TILE_SPLIT(ADDR_SURF_TILE_SPLIT_256B) |
-						 NUM_BANKS(ADDR_SURF_16_BANK) |
-						 BANK_WIDTH(ADDR_SURF_BANK_WIDTH_2) |
-						 BANK_HEIGHT(ADDR_SURF_BANK_HEIGHT_4) |
-						 MACRO_TILE_ASPECT(ADDR_SURF_MACRO_ASPECT_2));
-				break;
-			case 22:  /* 16 bpp PRT */
-				gb_tile_moden = (ARRAY_MODE(ARRAY_2D_TILED_THIN1) |
-						 MICRO_TILE_MODE(ADDR_SURF_THIN_MICRO_TILING) |
-						 PIPE_CONFIG(ADDR_SURF_P8_32x32_8x16) |
-						 TILE_SPLIT(ADDR_SURF_TILE_SPLIT_256B) |
-						 NUM_BANKS(ADDR_SURF_16_BANK) |
-						 BANK_WIDTH(ADDR_SURF_BANK_WIDTH_1) |
-						 BANK_HEIGHT(ADDR_SURF_BANK_HEIGHT_4) |
-						 MACRO_TILE_ASPECT(ADDR_SURF_MACRO_ASPECT_4));
-				break;
-			case 23:  /* 32 bpp PRT */
-				gb_tile_moden = (ARRAY_MODE(ARRAY_2D_TILED_THIN1) |
-						 MICRO_TILE_MODE(ADDR_SURF_THIN_MICRO_TILING) |
-						 PIPE_CONFIG(ADDR_SURF_P8_32x32_8x16) |
-						 TILE_SPLIT(ADDR_SURF_TILE_SPLIT_256B) |
-						 NUM_BANKS(ADDR_SURF_16_BANK) |
-						 BANK_WIDTH(ADDR_SURF_BANK_WIDTH_1) |
-						 BANK_HEIGHT(ADDR_SURF_BANK_HEIGHT_2) |
-						 MACRO_TILE_ASPECT(ADDR_SURF_MACRO_ASPECT_2));
-				break;
-			case 24:  /* 64 bpp PRT */
-				gb_tile_moden = (ARRAY_MODE(ARRAY_2D_TILED_THIN1) |
-						 MICRO_TILE_MODE(ADDR_SURF_THIN_MICRO_TILING) |
-						 PIPE_CONFIG(ADDR_SURF_P8_32x32_8x16) |
-						 TILE_SPLIT(ADDR_SURF_TILE_SPLIT_512B) |
-						 NUM_BANKS(ADDR_SURF_16_BANK) |
-						 BANK_WIDTH(ADDR_SURF_BANK_WIDTH_1) |
-						 BANK_HEIGHT(ADDR_SURF_BANK_HEIGHT_1) |
-						 MACRO_TILE_ASPECT(ADDR_SURF_MACRO_ASPECT_2));
-				break;
-			case 25:  /* 128 bpp PRT */
-				gb_tile_moden = (ARRAY_MODE(ARRAY_2D_TILED_THIN1) |
-						 MICRO_TILE_MODE(ADDR_SURF_THIN_MICRO_TILING) |
-						 PIPE_CONFIG(ADDR_SURF_P8_32x32_8x16) |
-						 TILE_SPLIT(ADDR_SURF_TILE_SPLIT_1KB) |
-						 NUM_BANKS(ADDR_SURF_8_BANK) |
-						 BANK_WIDTH(ADDR_SURF_BANK_WIDTH_1) |
-						 BANK_HEIGHT(ADDR_SURF_BANK_HEIGHT_1) |
-						 MACRO_TILE_ASPECT(ADDR_SURF_MACRO_ASPECT_1));
-				break;
-			default:
-				gb_tile_moden = 0;
-				break;
-			}
-			rdev->config.si.tile_mode_array[reg_offset] = gb_tile_moden;
-			WREG32(GB_TILE_MODE0 + (reg_offset * 4), gb_tile_moden);
-		}
-	} else
+	for (reg_offset = 0; reg_offset < num_tile_mode_states; reg_offset++)
+		tile[reg_offset] = 0;
+
+	switch(rdev->family) {
+	case CHIP_TAHITI:
+	case CHIP_PITCAIRN:
+		/* non-AA compressed depth or any compressed stencil */
+		tile[0] = (ARRAY_MODE(ARRAY_2D_TILED_THIN1) |
+			   MICRO_TILE_MODE(ADDR_SURF_DEPTH_MICRO_TILING) |
+			   PIPE_CONFIG(ADDR_SURF_P8_32x32_8x16) |
+			   TILE_SPLIT(ADDR_SURF_TILE_SPLIT_64B) |
+			   NUM_BANKS(ADDR_SURF_16_BANK) |
+			   BANK_WIDTH(ADDR_SURF_BANK_WIDTH_1) |
+			   BANK_HEIGHT(ADDR_SURF_BANK_HEIGHT_4) |
+			   MACRO_TILE_ASPECT(ADDR_SURF_MACRO_ASPECT_2));
+		/* 2xAA/4xAA compressed depth only */
+		tile[1] = (ARRAY_MODE(ARRAY_2D_TILED_THIN1) |
+			   MICRO_TILE_MODE(ADDR_SURF_DEPTH_MICRO_TILING) |
+			   PIPE_CONFIG(ADDR_SURF_P8_32x32_8x16) |
+			   TILE_SPLIT(ADDR_SURF_TILE_SPLIT_128B) |
+			   NUM_BANKS(ADDR_SURF_16_BANK) |
+			   BANK_WIDTH(ADDR_SURF_BANK_WIDTH_1) |
+			   BANK_HEIGHT(ADDR_SURF_BANK_HEIGHT_4) |
+			   MACRO_TILE_ASPECT(ADDR_SURF_MACRO_ASPECT_2));
+		/* 8xAA compressed depth only */
+		tile[2] = (ARRAY_MODE(ARRAY_2D_TILED_THIN1) |
+			   MICRO_TILE_MODE(ADDR_SURF_DEPTH_MICRO_TILING) |
+			   PIPE_CONFIG(ADDR_SURF_P8_32x32_8x16) |
+			   TILE_SPLIT(ADDR_SURF_TILE_SPLIT_256B) |
+			   NUM_BANKS(ADDR_SURF_16_BANK) |
+			   BANK_WIDTH(ADDR_SURF_BANK_WIDTH_1) |
+			   BANK_HEIGHT(ADDR_SURF_BANK_HEIGHT_4) |
+			   MACRO_TILE_ASPECT(ADDR_SURF_MACRO_ASPECT_2));
+		/* 2xAA/4xAA compressed depth with stencil (for depth buffer) */
+		tile[3] = (ARRAY_MODE(ARRAY_2D_TILED_THIN1) |
+			   MICRO_TILE_MODE(ADDR_SURF_DEPTH_MICRO_TILING) |
+			   PIPE_CONFIG(ADDR_SURF_P8_32x32_8x16) |
+			   TILE_SPLIT(ADDR_SURF_TILE_SPLIT_128B) |
+			   NUM_BANKS(ADDR_SURF_16_BANK) |
+			   BANK_WIDTH(ADDR_SURF_BANK_WIDTH_1) |
+			   BANK_HEIGHT(ADDR_SURF_BANK_HEIGHT_4) |
+			   MACRO_TILE_ASPECT(ADDR_SURF_MACRO_ASPECT_2));
+		/* Maps w/ a dimension less than the 2D macro-tile dimensions (for mipmapped depth textures) */
+		tile[4] = (ARRAY_MODE(ARRAY_1D_TILED_THIN1) |
+			   MICRO_TILE_MODE(ADDR_SURF_DEPTH_MICRO_TILING) |
+			   PIPE_CONFIG(ADDR_SURF_P8_32x32_8x16) |
+			   TILE_SPLIT(ADDR_SURF_TILE_SPLIT_64B) |
+			   NUM_BANKS(ADDR_SURF_16_BANK) |
+			   BANK_WIDTH(ADDR_SURF_BANK_WIDTH_1) |
+			   BANK_HEIGHT(ADDR_SURF_BANK_HEIGHT_2) |
+			   MACRO_TILE_ASPECT(ADDR_SURF_MACRO_ASPECT_2));
+		/* Uncompressed 16bpp depth - and stencil buffer allocated with it */
+		tile[5] = (ARRAY_MODE(ARRAY_2D_TILED_THIN1) |
+			   MICRO_TILE_MODE(ADDR_SURF_DEPTH_MICRO_TILING) |
+			   PIPE_CONFIG(ADDR_SURF_P8_32x32_8x16) |
+			   TILE_SPLIT(split_equal_to_row_size) |
+			   NUM_BANKS(ADDR_SURF_16_BANK) |
+			   BANK_WIDTH(ADDR_SURF_BANK_WIDTH_1) |
+			   BANK_HEIGHT(ADDR_SURF_BANK_HEIGHT_2) |
+			   MACRO_TILE_ASPECT(ADDR_SURF_MACRO_ASPECT_2));
+		/* Uncompressed 32bpp depth - and stencil buffer allocated with it */
+		tile[6] = (ARRAY_MODE(ARRAY_2D_TILED_THIN1) |
+			   MICRO_TILE_MODE(ADDR_SURF_DEPTH_MICRO_TILING) |
+			   PIPE_CONFIG(ADDR_SURF_P8_32x32_8x16) |
+			   TILE_SPLIT(split_equal_to_row_size) |
+			   NUM_BANKS(ADDR_SURF_16_BANK) |
+			   BANK_WIDTH(ADDR_SURF_BANK_WIDTH_1) |
+			   BANK_HEIGHT(ADDR_SURF_BANK_HEIGHT_1) |
+			   MACRO_TILE_ASPECT(ADDR_SURF_MACRO_ASPECT_1));
+		/* Uncompressed 8bpp stencil without depth (drivers typically do not use) */
+		tile[7] = (ARRAY_MODE(ARRAY_2D_TILED_THIN1) |
+			   MICRO_TILE_MODE(ADDR_SURF_DEPTH_MICRO_TILING) |
+			   PIPE_CONFIG(ADDR_SURF_P8_32x32_8x16) |
+			   TILE_SPLIT(split_equal_to_row_size) |
+			   NUM_BANKS(ADDR_SURF_16_BANK) |
+			   BANK_WIDTH(ADDR_SURF_BANK_WIDTH_1) |
+			   BANK_HEIGHT(ADDR_SURF_BANK_HEIGHT_4) |
+			   MACRO_TILE_ASPECT(ADDR_SURF_MACRO_ASPECT_2));
+		/* 1D and 1D Array Surfaces */
+		tile[8] = (ARRAY_MODE(ARRAY_LINEAR_ALIGNED) |
+			   MICRO_TILE_MODE(ADDR_SURF_DISPLAY_MICRO_TILING) |
+			   PIPE_CONFIG(ADDR_SURF_P8_32x32_8x16) |
+			   TILE_SPLIT(ADDR_SURF_TILE_SPLIT_64B) |
+			   NUM_BANKS(ADDR_SURF_16_BANK) |
+			   BANK_WIDTH(ADDR_SURF_BANK_WIDTH_1) |
+			   BANK_HEIGHT(ADDR_SURF_BANK_HEIGHT_2) |
+			   MACRO_TILE_ASPECT(ADDR_SURF_MACRO_ASPECT_2));
+		/* Displayable maps. */
+		tile[9] = (ARRAY_MODE(ARRAY_1D_TILED_THIN1) |
+			   MICRO_TILE_MODE(ADDR_SURF_DISPLAY_MICRO_TILING) |
+			   PIPE_CONFIG(ADDR_SURF_P8_32x32_8x16) |
+			   TILE_SPLIT(ADDR_SURF_TILE_SPLIT_64B) |
+			   NUM_BANKS(ADDR_SURF_16_BANK) |
+			   BANK_WIDTH(ADDR_SURF_BANK_WIDTH_1) |
+			   BANK_HEIGHT(ADDR_SURF_BANK_HEIGHT_2) |
+			   MACRO_TILE_ASPECT(ADDR_SURF_MACRO_ASPECT_2));
+		/* Display 8bpp. */
+		tile[10] = (ARRAY_MODE(ARRAY_2D_TILED_THIN1) |
+			   MICRO_TILE_MODE(ADDR_SURF_DISPLAY_MICRO_TILING) |
+			   PIPE_CONFIG(ADDR_SURF_P8_32x32_8x16) |
+			   TILE_SPLIT(ADDR_SURF_TILE_SPLIT_256B) |
+			   NUM_BANKS(ADDR_SURF_16_BANK) |
+			   BANK_WIDTH(ADDR_SURF_BANK_WIDTH_1) |
+			   BANK_HEIGHT(ADDR_SURF_BANK_HEIGHT_4) |
+			   MACRO_TILE_ASPECT(ADDR_SURF_MACRO_ASPECT_2));
+		/* Display 16bpp. */
+		tile[11] = (ARRAY_MODE(ARRAY_2D_TILED_THIN1) |
+			   MICRO_TILE_MODE(ADDR_SURF_DISPLAY_MICRO_TILING) |
+			   PIPE_CONFIG(ADDR_SURF_P8_32x32_8x16) |
+			   TILE_SPLIT(ADDR_SURF_TILE_SPLIT_256B) |
+			   NUM_BANKS(ADDR_SURF_16_BANK) |
+			   BANK_WIDTH(ADDR_SURF_BANK_WIDTH_1) |
+			   BANK_HEIGHT(ADDR_SURF_BANK_HEIGHT_2) |
+			   MACRO_TILE_ASPECT(ADDR_SURF_MACRO_ASPECT_2));
+		/* Display 32bpp. */
+		tile[12] = (ARRAY_MODE(ARRAY_2D_TILED_THIN1) |
+			   MICRO_TILE_MODE(ADDR_SURF_DISPLAY_MICRO_TILING) |
+			   PIPE_CONFIG(ADDR_SURF_P8_32x32_8x16) |
+			   TILE_SPLIT(ADDR_SURF_TILE_SPLIT_512B) |
+			   NUM_BANKS(ADDR_SURF_16_BANK) |
+			   BANK_WIDTH(ADDR_SURF_BANK_WIDTH_1) |
+			   BANK_HEIGHT(ADDR_SURF_BANK_HEIGHT_1) |
+			   MACRO_TILE_ASPECT(ADDR_SURF_MACRO_ASPECT_1));
+		/* Thin. */
+		tile[13] = (ARRAY_MODE(ARRAY_1D_TILED_THIN1) |
+			   MICRO_TILE_MODE(ADDR_SURF_THIN_MICRO_TILING) |
+			   PIPE_CONFIG(ADDR_SURF_P8_32x32_8x16) |
+			   TILE_SPLIT(ADDR_SURF_TILE_SPLIT_64B) |
+			   NUM_BANKS(ADDR_SURF_16_BANK) |
+			   BANK_WIDTH(ADDR_SURF_BANK_WIDTH_1) |
+			   BANK_HEIGHT(ADDR_SURF_BANK_HEIGHT_2) |
+			   MACRO_TILE_ASPECT(ADDR_SURF_MACRO_ASPECT_2));
+		/* Thin 8 bpp. */
+		tile[14] = (ARRAY_MODE(ARRAY_2D_TILED_THIN1) |
+			   MICRO_TILE_MODE(ADDR_SURF_THIN_MICRO_TILING) |
+			   PIPE_CONFIG(ADDR_SURF_P8_32x32_8x16) |
+			   TILE_SPLIT(ADDR_SURF_TILE_SPLIT_256B) |
+			   NUM_BANKS(ADDR_SURF_16_BANK) |
+			   BANK_WIDTH(ADDR_SURF_BANK_WIDTH_1) |
+			   BANK_HEIGHT(ADDR_SURF_BANK_HEIGHT_4) |
+			   MACRO_TILE_ASPECT(ADDR_SURF_MACRO_ASPECT_1));
+		/* Thin 16 bpp. */
+		tile[15] = (ARRAY_MODE(ARRAY_2D_TILED_THIN1) |
+			   MICRO_TILE_MODE(ADDR_SURF_THIN_MICRO_TILING) |
+			   PIPE_CONFIG(ADDR_SURF_P8_32x32_8x16) |
+			   TILE_SPLIT(ADDR_SURF_TILE_SPLIT_256B) |
+			   NUM_BANKS(ADDR_SURF_16_BANK) |
+			   BANK_WIDTH(ADDR_SURF_BANK_WIDTH_1) |
+			   BANK_HEIGHT(ADDR_SURF_BANK_HEIGHT_2) |
+			   MACRO_TILE_ASPECT(ADDR_SURF_MACRO_ASPECT_1));
+		/* Thin 32 bpp. */
+		tile[16] = (ARRAY_MODE(ARRAY_2D_TILED_THIN1) |
+			   MICRO_TILE_MODE(ADDR_SURF_THIN_MICRO_TILING) |
+			   PIPE_CONFIG(ADDR_SURF_P8_32x32_8x16) |
+			   TILE_SPLIT(ADDR_SURF_TILE_SPLIT_512B) |
+			   NUM_BANKS(ADDR_SURF_16_BANK) |
+			   BANK_WIDTH(ADDR_SURF_BANK_WIDTH_1) |
+			   BANK_HEIGHT(ADDR_SURF_BANK_HEIGHT_1) |
+			   MACRO_TILE_ASPECT(ADDR_SURF_MACRO_ASPECT_1));
+		/* Thin 64 bpp. */
+		tile[17] = (ARRAY_MODE(ARRAY_2D_TILED_THIN1) |
+			   MICRO_TILE_MODE(ADDR_SURF_THIN_MICRO_TILING) |
+			   PIPE_CONFIG(ADDR_SURF_P8_32x32_8x16) |
+			   TILE_SPLIT(split_equal_to_row_size) |
+			   NUM_BANKS(ADDR_SURF_16_BANK) |
+			   BANK_WIDTH(ADDR_SURF_BANK_WIDTH_1) |
+			   BANK_HEIGHT(ADDR_SURF_BANK_HEIGHT_1) |
+			   MACRO_TILE_ASPECT(ADDR_SURF_MACRO_ASPECT_1));
+		/* 8 bpp PRT. */
+		tile[21] = (ARRAY_MODE(ARRAY_2D_TILED_THIN1) |
+			   MICRO_TILE_MODE(ADDR_SURF_THIN_MICRO_TILING) |
+			   PIPE_CONFIG(ADDR_SURF_P8_32x32_8x16) |
+			   TILE_SPLIT(ADDR_SURF_TILE_SPLIT_256B) |
+			   NUM_BANKS(ADDR_SURF_16_BANK) |
+			   BANK_WIDTH(ADDR_SURF_BANK_WIDTH_2) |
+			   BANK_HEIGHT(ADDR_SURF_BANK_HEIGHT_4) |
+			   MACRO_TILE_ASPECT(ADDR_SURF_MACRO_ASPECT_2));
+		/* 16 bpp PRT */
+		tile[22] = (ARRAY_MODE(ARRAY_2D_TILED_THIN1) |
+			   MICRO_TILE_MODE(ADDR_SURF_THIN_MICRO_TILING) |
+			   PIPE_CONFIG(ADDR_SURF_P8_32x32_8x16) |
+			   TILE_SPLIT(ADDR_SURF_TILE_SPLIT_256B) |
+			   NUM_BANKS(ADDR_SURF_16_BANK) |
+			   BANK_WIDTH(ADDR_SURF_BANK_WIDTH_1) |
+			   BANK_HEIGHT(ADDR_SURF_BANK_HEIGHT_4) |
+			   MACRO_TILE_ASPECT(ADDR_SURF_MACRO_ASPECT_4));
+		/* 32 bpp PRT */
+		tile[23] = (ARRAY_MODE(ARRAY_2D_TILED_THIN1) |
+			   MICRO_TILE_MODE(ADDR_SURF_THIN_MICRO_TILING) |
+			   PIPE_CONFIG(ADDR_SURF_P8_32x32_8x16) |
+			   TILE_SPLIT(ADDR_SURF_TILE_SPLIT_256B) |
+			   NUM_BANKS(ADDR_SURF_16_BANK) |
+			   BANK_WIDTH(ADDR_SURF_BANK_WIDTH_1) |
+			   BANK_HEIGHT(ADDR_SURF_BANK_HEIGHT_2) |
+			   MACRO_TILE_ASPECT(ADDR_SURF_MACRO_ASPECT_2));
+		/* 64 bpp PRT */
+		tile[24] = (ARRAY_MODE(ARRAY_2D_TILED_THIN1) |
+			   MICRO_TILE_MODE(ADDR_SURF_THIN_MICRO_TILING) |
+			   PIPE_CONFIG(ADDR_SURF_P8_32x32_8x16) |
+			   TILE_SPLIT(ADDR_SURF_TILE_SPLIT_512B) |
+			   NUM_BANKS(ADDR_SURF_16_BANK) |
+			   BANK_WIDTH(ADDR_SURF_BANK_WIDTH_1) |
+			   BANK_HEIGHT(ADDR_SURF_BANK_HEIGHT_1) |
+			   MACRO_TILE_ASPECT(ADDR_SURF_MACRO_ASPECT_2));
+		/* 128 bpp PRT */
+		tile[25] = (ARRAY_MODE(ARRAY_2D_TILED_THIN1) |
+			   MICRO_TILE_MODE(ADDR_SURF_THIN_MICRO_TILING) |
+			   PIPE_CONFIG(ADDR_SURF_P8_32x32_8x16) |
+			   TILE_SPLIT(ADDR_SURF_TILE_SPLIT_1KB) |
+			   NUM_BANKS(ADDR_SURF_8_BANK) |
+			   BANK_WIDTH(ADDR_SURF_BANK_WIDTH_1) |
+			   BANK_HEIGHT(ADDR_SURF_BANK_HEIGHT_1) |
+			   MACRO_TILE_ASPECT(ADDR_SURF_MACRO_ASPECT_1));
+
+		for (reg_offset = 0; reg_offset < num_tile_mode_states; reg_offset++)
+			WREG32(GB_TILE_MODE0 + (reg_offset * 4), tile[reg_offset]);
+		break;
+
+	case CHIP_VERDE:
+	case CHIP_OLAND:
+	case CHIP_HAINAN:
+		/* non-AA compressed depth or any compressed stencil */
+		tile[0] = (ARRAY_MODE(ARRAY_2D_TILED_THIN1) |
+			   MICRO_TILE_MODE(ADDR_SURF_DEPTH_MICRO_TILING) |
+			   PIPE_CONFIG(ADDR_SURF_P4_8x16) |
+			   TILE_SPLIT(ADDR_SURF_TILE_SPLIT_64B) |
+			   NUM_BANKS(ADDR_SURF_16_BANK) |
+			   BANK_WIDTH(ADDR_SURF_BANK_WIDTH_1) |
+			   BANK_HEIGHT(ADDR_SURF_BANK_HEIGHT_4) |
+			   MACRO_TILE_ASPECT(ADDR_SURF_MACRO_ASPECT_4));
+		/* 2xAA/4xAA compressed depth only */
+		tile[1] = (ARRAY_MODE(ARRAY_2D_TILED_THIN1) |
+			   MICRO_TILE_MODE(ADDR_SURF_DEPTH_MICRO_TILING) |
+			   PIPE_CONFIG(ADDR_SURF_P4_8x16) |
+			   TILE_SPLIT(ADDR_SURF_TILE_SPLIT_128B) |
+			   NUM_BANKS(ADDR_SURF_16_BANK) |
+			   BANK_WIDTH(ADDR_SURF_BANK_WIDTH_1) |
+			   BANK_HEIGHT(ADDR_SURF_BANK_HEIGHT_4) |
+			   MACRO_TILE_ASPECT(ADDR_SURF_MACRO_ASPECT_4));
+		/* 8xAA compressed depth only */
+		tile[2] = (ARRAY_MODE(ARRAY_2D_TILED_THIN1) |
+			   MICRO_TILE_MODE(ADDR_SURF_DEPTH_MICRO_TILING) |
+			   PIPE_CONFIG(ADDR_SURF_P4_8x16) |
+			   TILE_SPLIT(ADDR_SURF_TILE_SPLIT_256B) |
+			   NUM_BANKS(ADDR_SURF_16_BANK) |
+			   BANK_WIDTH(ADDR_SURF_BANK_WIDTH_1) |
+			   BANK_HEIGHT(ADDR_SURF_BANK_HEIGHT_4) |
+			   MACRO_TILE_ASPECT(ADDR_SURF_MACRO_ASPECT_4));
+		/* 2xAA/4xAA compressed depth with stencil (for depth buffer) */
+		tile[3] = (ARRAY_MODE(ARRAY_2D_TILED_THIN1) |
+			   MICRO_TILE_MODE(ADDR_SURF_DEPTH_MICRO_TILING) |
+			   PIPE_CONFIG(ADDR_SURF_P4_8x16) |
+			   TILE_SPLIT(ADDR_SURF_TILE_SPLIT_128B) |
+			   NUM_BANKS(ADDR_SURF_16_BANK) |
+			   BANK_WIDTH(ADDR_SURF_BANK_WIDTH_1) |
+			   BANK_HEIGHT(ADDR_SURF_BANK_HEIGHT_4) |
+			   MACRO_TILE_ASPECT(ADDR_SURF_MACRO_ASPECT_4));
+		/* Maps w/ a dimension less than the 2D macro-tile dimensions (for mipmapped depth textures) */
+		tile[4] = (ARRAY_MODE(ARRAY_1D_TILED_THIN1) |
+			   MICRO_TILE_MODE(ADDR_SURF_DEPTH_MICRO_TILING) |
+			   PIPE_CONFIG(ADDR_SURF_P4_8x16) |
+			   TILE_SPLIT(ADDR_SURF_TILE_SPLIT_64B) |
+			   NUM_BANKS(ADDR_SURF_16_BANK) |
+			   BANK_WIDTH(ADDR_SURF_BANK_WIDTH_1) |
+			   BANK_HEIGHT(ADDR_SURF_BANK_HEIGHT_2) |
+			   MACRO_TILE_ASPECT(ADDR_SURF_MACRO_ASPECT_2));
+		/* Uncompressed 16bpp depth - and stencil buffer allocated with it */
+		tile[5] = (ARRAY_MODE(ARRAY_2D_TILED_THIN1) |
+			   MICRO_TILE_MODE(ADDR_SURF_DEPTH_MICRO_TILING) |
+			   PIPE_CONFIG(ADDR_SURF_P4_8x16) |
+			   TILE_SPLIT(split_equal_to_row_size) |
+			   NUM_BANKS(ADDR_SURF_16_BANK) |
+			   BANK_WIDTH(ADDR_SURF_BANK_WIDTH_1) |
+			   BANK_HEIGHT(ADDR_SURF_BANK_HEIGHT_2) |
+			   MACRO_TILE_ASPECT(ADDR_SURF_MACRO_ASPECT_2));
+		/* Uncompressed 32bpp depth - and stencil buffer allocated with it */
+		tile[6] = (ARRAY_MODE(ARRAY_2D_TILED_THIN1) |
+			   MICRO_TILE_MODE(ADDR_SURF_DEPTH_MICRO_TILING) |
+			   PIPE_CONFIG(ADDR_SURF_P4_8x16) |
+			   TILE_SPLIT(split_equal_to_row_size) |
+			   NUM_BANKS(ADDR_SURF_16_BANK) |
+			   BANK_WIDTH(ADDR_SURF_BANK_WIDTH_1) |
+			   BANK_HEIGHT(ADDR_SURF_BANK_HEIGHT_1) |
+			   MACRO_TILE_ASPECT(ADDR_SURF_MACRO_ASPECT_2));
+		/* Uncompressed 8bpp stencil without depth (drivers typically do not use) */
+		tile[7] = (ARRAY_MODE(ARRAY_2D_TILED_THIN1) |
+			   MICRO_TILE_MODE(ADDR_SURF_DEPTH_MICRO_TILING) |
+			   PIPE_CONFIG(ADDR_SURF_P4_8x16) |
+			   TILE_SPLIT(split_equal_to_row_size) |
+			   NUM_BANKS(ADDR_SURF_16_BANK) |
+			   BANK_WIDTH(ADDR_SURF_BANK_WIDTH_1) |
+			   BANK_HEIGHT(ADDR_SURF_BANK_HEIGHT_4) |
+			   MACRO_TILE_ASPECT(ADDR_SURF_MACRO_ASPECT_4));
+		/* 1D and 1D Array Surfaces */
+		tile[8] = (ARRAY_MODE(ARRAY_LINEAR_ALIGNED) |
+			   MICRO_TILE_MODE(ADDR_SURF_DISPLAY_MICRO_TILING) |
+			   PIPE_CONFIG(ADDR_SURF_P4_8x16) |
+			   TILE_SPLIT(ADDR_SURF_TILE_SPLIT_64B) |
+			   NUM_BANKS(ADDR_SURF_16_BANK) |
+			   BANK_WIDTH(ADDR_SURF_BANK_WIDTH_1) |
+			   BANK_HEIGHT(ADDR_SURF_BANK_HEIGHT_2) |
+			   MACRO_TILE_ASPECT(ADDR_SURF_MACRO_ASPECT_2));
+		/* Displayable maps. */
+		tile[9] = (ARRAY_MODE(ARRAY_1D_TILED_THIN1) |
+			   MICRO_TILE_MODE(ADDR_SURF_DISPLAY_MICRO_TILING) |
+			   PIPE_CONFIG(ADDR_SURF_P4_8x16) |
+			   TILE_SPLIT(ADDR_SURF_TILE_SPLIT_64B) |
+			   NUM_BANKS(ADDR_SURF_16_BANK) |
+			   BANK_WIDTH(ADDR_SURF_BANK_WIDTH_1) |
+			   BANK_HEIGHT(ADDR_SURF_BANK_HEIGHT_2) |
+			   MACRO_TILE_ASPECT(ADDR_SURF_MACRO_ASPECT_2));
+		/* Display 8bpp. */
+		tile[10] = (ARRAY_MODE(ARRAY_2D_TILED_THIN1) |
+			   MICRO_TILE_MODE(ADDR_SURF_DISPLAY_MICRO_TILING) |
+			   PIPE_CONFIG(ADDR_SURF_P4_8x16) |
+			   TILE_SPLIT(ADDR_SURF_TILE_SPLIT_256B) |
+			   NUM_BANKS(ADDR_SURF_16_BANK) |
+			   BANK_WIDTH(ADDR_SURF_BANK_WIDTH_1) |
+			   BANK_HEIGHT(ADDR_SURF_BANK_HEIGHT_4) |
+			   MACRO_TILE_ASPECT(ADDR_SURF_MACRO_ASPECT_4));
+		/* Display 16bpp. */
+		tile[11] = (ARRAY_MODE(ARRAY_2D_TILED_THIN1) |
+			   MICRO_TILE_MODE(ADDR_SURF_DISPLAY_MICRO_TILING) |
+			   PIPE_CONFIG(ADDR_SURF_P4_8x16) |
+			   TILE_SPLIT(ADDR_SURF_TILE_SPLIT_256B) |
+			   NUM_BANKS(ADDR_SURF_16_BANK) |
+			   BANK_WIDTH(ADDR_SURF_BANK_WIDTH_1) |
+			   BANK_HEIGHT(ADDR_SURF_BANK_HEIGHT_2) |
+			   MACRO_TILE_ASPECT(ADDR_SURF_MACRO_ASPECT_2));
+		/* Display 32bpp. */
+		tile[12] = (ARRAY_MODE(ARRAY_2D_TILED_THIN1) |
+			   MICRO_TILE_MODE(ADDR_SURF_DISPLAY_MICRO_TILING) |
+			   PIPE_CONFIG(ADDR_SURF_P4_8x16) |
+			   TILE_SPLIT(ADDR_SURF_TILE_SPLIT_512B) |
+			   NUM_BANKS(ADDR_SURF_16_BANK) |
+			   BANK_WIDTH(ADDR_SURF_BANK_WIDTH_1) |
+			   BANK_HEIGHT(ADDR_SURF_BANK_HEIGHT_1) |
+			   MACRO_TILE_ASPECT(ADDR_SURF_MACRO_ASPECT_2));
+		/* Thin. */
+		tile[13] = (ARRAY_MODE(ARRAY_1D_TILED_THIN1) |
+			   MICRO_TILE_MODE(ADDR_SURF_THIN_MICRO_TILING) |
+			   PIPE_CONFIG(ADDR_SURF_P4_8x16) |
+			   TILE_SPLIT(ADDR_SURF_TILE_SPLIT_64B) |
+			   NUM_BANKS(ADDR_SURF_16_BANK) |
+			   BANK_WIDTH(ADDR_SURF_BANK_WIDTH_1) |
+			   BANK_HEIGHT(ADDR_SURF_BANK_HEIGHT_2) |
+			   MACRO_TILE_ASPECT(ADDR_SURF_MACRO_ASPECT_2));
+		/* Thin 8 bpp. */
+		tile[14] = (ARRAY_MODE(ARRAY_2D_TILED_THIN1) |
+			   MICRO_TILE_MODE(ADDR_SURF_THIN_MICRO_TILING) |
+			   PIPE_CONFIG(ADDR_SURF_P4_8x16) |
+			   TILE_SPLIT(ADDR_SURF_TILE_SPLIT_256B) |
+			   NUM_BANKS(ADDR_SURF_16_BANK) |
+			   BANK_WIDTH(ADDR_SURF_BANK_WIDTH_1) |
+			   BANK_HEIGHT(ADDR_SURF_BANK_HEIGHT_4) |
+			   MACRO_TILE_ASPECT(ADDR_SURF_MACRO_ASPECT_2));
+		/* Thin 16 bpp. */
+		tile[15] = (ARRAY_MODE(ARRAY_2D_TILED_THIN1) |
+			   MICRO_TILE_MODE(ADDR_SURF_THIN_MICRO_TILING) |
+			   PIPE_CONFIG(ADDR_SURF_P4_8x16) |
+			   TILE_SPLIT(ADDR_SURF_TILE_SPLIT_256B) |
+			   NUM_BANKS(ADDR_SURF_16_BANK) |
+			   BANK_WIDTH(ADDR_SURF_BANK_WIDTH_1) |
+			   BANK_HEIGHT(ADDR_SURF_BANK_HEIGHT_2) |
+			   MACRO_TILE_ASPECT(ADDR_SURF_MACRO_ASPECT_2));
+		/* Thin 32 bpp. */
+		tile[16] = (ARRAY_MODE(ARRAY_2D_TILED_THIN1) |
+			   MICRO_TILE_MODE(ADDR_SURF_THIN_MICRO_TILING) |
+			   PIPE_CONFIG(ADDR_SURF_P4_8x16) |
+			   TILE_SPLIT(ADDR_SURF_TILE_SPLIT_512B) |
+			   NUM_BANKS(ADDR_SURF_16_BANK) |
+			   BANK_WIDTH(ADDR_SURF_BANK_WIDTH_1) |
+			   BANK_HEIGHT(ADDR_SURF_BANK_HEIGHT_1) |
+			   MACRO_TILE_ASPECT(ADDR_SURF_MACRO_ASPECT_2));
+		/* Thin 64 bpp. */
+		tile[17] = (ARRAY_MODE(ARRAY_2D_TILED_THIN1) |
+			   MICRO_TILE_MODE(ADDR_SURF_THIN_MICRO_TILING) |
+			   PIPE_CONFIG(ADDR_SURF_P4_8x16) |
+			   TILE_SPLIT(split_equal_to_row_size) |
+			   NUM_BANKS(ADDR_SURF_16_BANK) |
+			   BANK_WIDTH(ADDR_SURF_BANK_WIDTH_1) |
+			   BANK_HEIGHT(ADDR_SURF_BANK_HEIGHT_1) |
+			   MACRO_TILE_ASPECT(ADDR_SURF_MACRO_ASPECT_2));
+		/* 8 bpp PRT. */
+		tile[21] = (ARRAY_MODE(ARRAY_2D_TILED_THIN1) |
+			   MICRO_TILE_MODE(ADDR_SURF_THIN_MICRO_TILING) |
+			   PIPE_CONFIG(ADDR_SURF_P8_32x32_8x16) |
+			   TILE_SPLIT(ADDR_SURF_TILE_SPLIT_256B) |
+			   NUM_BANKS(ADDR_SURF_16_BANK) |
+			   BANK_WIDTH(ADDR_SURF_BANK_WIDTH_2) |
+			   BANK_HEIGHT(ADDR_SURF_BANK_HEIGHT_4) |
+			   MACRO_TILE_ASPECT(ADDR_SURF_MACRO_ASPECT_2));
+		/* 16 bpp PRT */
+		tile[22] = (ARRAY_MODE(ARRAY_2D_TILED_THIN1) |
+			   MICRO_TILE_MODE(ADDR_SURF_THIN_MICRO_TILING) |
+			   PIPE_CONFIG(ADDR_SURF_P8_32x32_8x16) |
+			   TILE_SPLIT(ADDR_SURF_TILE_SPLIT_256B) |
+			   NUM_BANKS(ADDR_SURF_16_BANK) |
+			   BANK_WIDTH(ADDR_SURF_BANK_WIDTH_1) |
+			   BANK_HEIGHT(ADDR_SURF_BANK_HEIGHT_4) |
+			   MACRO_TILE_ASPECT(ADDR_SURF_MACRO_ASPECT_4));
+		/* 32 bpp PRT */
+		tile[23] = (ARRAY_MODE(ARRAY_2D_TILED_THIN1) |
+			   MICRO_TILE_MODE(ADDR_SURF_THIN_MICRO_TILING) |
+			   PIPE_CONFIG(ADDR_SURF_P8_32x32_8x16) |
+			   TILE_SPLIT(ADDR_SURF_TILE_SPLIT_256B) |
+			   NUM_BANKS(ADDR_SURF_16_BANK) |
+			   BANK_WIDTH(ADDR_SURF_BANK_WIDTH_1) |
+			   BANK_HEIGHT(ADDR_SURF_BANK_HEIGHT_2) |
+			   MACRO_TILE_ASPECT(ADDR_SURF_MACRO_ASPECT_2));
+		/* 64 bpp PRT */
+		tile[24] = (ARRAY_MODE(ARRAY_2D_TILED_THIN1) |
+			   MICRO_TILE_MODE(ADDR_SURF_THIN_MICRO_TILING) |
+			   PIPE_CONFIG(ADDR_SURF_P8_32x32_8x16) |
+			   TILE_SPLIT(ADDR_SURF_TILE_SPLIT_512B) |
+			   NUM_BANKS(ADDR_SURF_16_BANK) |
+			   BANK_WIDTH(ADDR_SURF_BANK_WIDTH_1) |
+			   BANK_HEIGHT(ADDR_SURF_BANK_HEIGHT_1) |
+			   MACRO_TILE_ASPECT(ADDR_SURF_MACRO_ASPECT_2));
+		/* 128 bpp PRT */
+		tile[25] = (ARRAY_MODE(ARRAY_2D_TILED_THIN1) |
+			   MICRO_TILE_MODE(ADDR_SURF_THIN_MICRO_TILING) |
+			   PIPE_CONFIG(ADDR_SURF_P8_32x32_8x16) |
+			   TILE_SPLIT(ADDR_SURF_TILE_SPLIT_1KB) |
+			   NUM_BANKS(ADDR_SURF_8_BANK) |
+			   BANK_WIDTH(ADDR_SURF_BANK_WIDTH_1) |
+			   BANK_HEIGHT(ADDR_SURF_BANK_HEIGHT_1) |
+			   MACRO_TILE_ASPECT(ADDR_SURF_MACRO_ASPECT_1));
+
+		for (reg_offset = 0; reg_offset < num_tile_mode_states; reg_offset++)
+			WREG32(GB_TILE_MODE0 + (reg_offset * 4), tile[reg_offset]);
+		break;
+
+	default:
 		DRM_ERROR("unknown asic: 0x%x\n", rdev->family);
+	}
 }
 
 static void si_select_se_sh(struct radeon_device *rdev,
@@ -7314,7 +7267,7 @@ uint64_t si_get_gpu_clock_counter(struct radeon_device *rdev)
 	mutex_lock(&rdev->gpu_clock_mutex);
 	WREG32(RLC_CAPTURE_GPU_CLOCK_COUNT, 1);
 	clock = (uint64_t)RREG32(RLC_GPU_CLOCK_COUNT_LSB) |
-	        ((uint64_t)RREG32(RLC_GPU_CLOCK_COUNT_MSB) << 32ULL);
+		((uint64_t)RREG32(RLC_GPU_CLOCK_COUNT_MSB) << 32ULL);
 	mutex_unlock(&rdev->gpu_clock_mutex);
 	return clock;
 }
@@ -7775,33 +7728,33 @@ static void si_program_aspm(struct radeon_device *rdev)
 
 int si_vce_send_vcepll_ctlreq(struct radeon_device *rdev)
 {
-        unsigned i;
+	unsigned i;
 
-        /* make sure VCEPLL_CTLREQ is deasserted */
-        WREG32_SMC_P(CG_VCEPLL_FUNC_CNTL, 0, ~UPLL_CTLREQ_MASK);
+	/* make sure VCEPLL_CTLREQ is deasserted */
+	WREG32_SMC_P(CG_VCEPLL_FUNC_CNTL, 0, ~UPLL_CTLREQ_MASK);
 
-        mdelay(10);
+	mdelay(10);
 
-        /* assert UPLL_CTLREQ */
-        WREG32_SMC_P(CG_VCEPLL_FUNC_CNTL, UPLL_CTLREQ_MASK, ~UPLL_CTLREQ_MASK);
+	/* assert UPLL_CTLREQ */
+	WREG32_SMC_P(CG_VCEPLL_FUNC_CNTL, UPLL_CTLREQ_MASK, ~UPLL_CTLREQ_MASK);
 
-        /* wait for CTLACK and CTLACK2 to get asserted */
-        for (i = 0; i < 100; ++i) {
-                uint32_t mask = UPLL_CTLACK_MASK | UPLL_CTLACK2_MASK;
-                if ((RREG32_SMC(CG_VCEPLL_FUNC_CNTL) & mask) == mask)
-                        break;
-                mdelay(10);
-        }
+	/* wait for CTLACK and CTLACK2 to get asserted */
+	for (i = 0; i < 100; ++i) {
+		uint32_t mask = UPLL_CTLACK_MASK | UPLL_CTLACK2_MASK;
+		if ((RREG32_SMC(CG_VCEPLL_FUNC_CNTL) & mask) == mask)
+			break;
+		mdelay(10);
+	}
 
-        /* deassert UPLL_CTLREQ */
-        WREG32_SMC_P(CG_VCEPLL_FUNC_CNTL, 0, ~UPLL_CTLREQ_MASK);
+	/* deassert UPLL_CTLREQ */
+	WREG32_SMC_P(CG_VCEPLL_FUNC_CNTL, 0, ~UPLL_CTLREQ_MASK);
 
-        if (i == 100) {
-                DRM_ERROR("Timeout setting UVD clocks!\n");
-                return -ETIMEDOUT;
-        }
+	if (i == 100) {
+		DRM_ERROR("Timeout setting UVD clocks!\n");
+		return -ETIMEDOUT;
+	}
 
-        return 0;
+	return 0;
 }
 
 int si_set_vce_clocks(struct radeon_device *rdev, u32 evclk, u32 ecclk)
diff --git a/drivers/gpu/drm/radeon/si_dpm.c b/drivers/gpu/drm/radeon/si_dpm.c
index a82b891ae1fe..cb75ab72098a 100644
--- a/drivers/gpu/drm/radeon/si_dpm.c
+++ b/drivers/gpu/drm/radeon/si_dpm.c
@@ -499,7 +499,7 @@ static const struct si_cac_config_reg lcac_pitcairn[] =
 
 static const struct si_cac_config_reg cac_override_pitcairn[] =
 {
-    { 0xFFFFFFFF }
+	{ 0xFFFFFFFF }
 };
 
 static const struct si_powertune_data powertune_data_pitcairn =
@@ -991,7 +991,7 @@ static const struct si_cac_config_reg lcac_cape_verde[] =
 
 static const struct si_cac_config_reg cac_override_cape_verde[] =
 {
-    { 0xFFFFFFFF }
+	{ 0xFFFFFFFF }
 };
 
 static const struct si_powertune_data powertune_data_cape_verde =
@@ -1762,9 +1762,9 @@ static void si_fan_ctrl_set_default_mode(struct radeon_device *rdev);
 
 static struct si_power_info *si_get_pi(struct radeon_device *rdev)
 {
-        struct si_power_info *pi = rdev->pm.dpm.priv;
+	struct si_power_info *pi = rdev->pm.dpm.priv;
 
-        return pi;
+	return pi;
 }
 
 static void si_calculate_leakage_for_v_and_t_formula(const struct ni_leakage_coeffients *coeff,
@@ -3150,9 +3150,9 @@ static void si_apply_state_adjust_rules(struct radeon_device *rdev,
 		}
 	}
 
-        for (i = 0; i < ps->performance_level_count; i++)
-                btc_adjust_clock_combinations(rdev, max_limits,
-                                              &ps->performance_levels[i]);
+	for (i = 0; i < ps->performance_level_count; i++)
+		btc_adjust_clock_combinations(rdev, max_limits,
+					      &ps->performance_levels[i]);
 
 	for (i = 0; i < ps->performance_level_count; i++) {
 		if (ps->performance_levels[i].vddc < min_vce_voltage)
@@ -3291,7 +3291,7 @@ static void si_set_dpm_event_sources(struct radeon_device *rdev, u32 sources)
 	case 0:
 	default:
 		want_thermal_protection = false;
-                break;
+		break;
 	case (1 << RADEON_DPM_AUTO_THROTTLE_SRC_THERMAL):
 		want_thermal_protection = true;
 		dpm_event_src = RADEON_DPM_EVENT_SRC_DIGITAL;
@@ -3493,7 +3493,7 @@ static int si_process_firmware_header(struct radeon_device *rdev)
 	if (ret)
 		return ret;
 
-        si_pi->state_table_start = tmp;
+	si_pi->state_table_start = tmp;
 
 	ret = si_read_smc_sram_dword(rdev,
 				     SISLANDS_SMC_FIRMWARE_HEADER_LOCATION +
@@ -3652,7 +3652,7 @@ static void si_program_response_times(struct radeon_device *rdev)
 	si_write_smc_soft_register(rdev, SI_SMC_SOFT_REGISTER_mvdd_chg_time, 1);
 
 	voltage_response_time = (u32)rdev->pm.dpm.voltage_response_time;
-        backbias_response_time = (u32)rdev->pm.dpm.backbias_response_time;
+	backbias_response_time = (u32)rdev->pm.dpm.backbias_response_time;
 
 	if (voltage_response_time == 0)
 		voltage_response_time = 1000;
@@ -3760,7 +3760,7 @@ static void si_setup_bsp(struct radeon_device *rdev)
 			       &pi->pbsu);
 
 
-        pi->dsp = BSP(pi->bsp) | BSU(pi->bsu);
+	pi->dsp = BSP(pi->bsp) | BSU(pi->bsu);
 	pi->psp = BSP(pi->pbsp) | BSU(pi->pbsu);
 
 	WREG32(CG_BSP, pi->dsp);
@@ -4308,7 +4308,7 @@ static int si_populate_memory_timing_parameters(struct radeon_device *rdev,
 
 	radeon_atom_set_engine_dram_timings(rdev,
 					    pl->sclk,
-                                            pl->mclk);
+					    pl->mclk);
 
 	dram_timing  = RREG32(MC_ARB_DRAM_TIMING);
 	dram_timing2 = RREG32(MC_ARB_DRAM_TIMING2);
@@ -4343,7 +4343,7 @@ static int si_do_program_memory_timing_parameters(struct radeon_device *rdev,
 					   si_pi->sram_end);
 		if (ret)
 			break;
-        }
+	}
 
 	return ret;
 }
@@ -4821,9 +4821,9 @@ static int si_calculate_sclk_params(struct radeon_device *rdev,
 	spll_func_cntl_2 &= ~SCLK_MUX_SEL_MASK;
 	spll_func_cntl_2 |= SCLK_MUX_SEL(2);
 
-        spll_func_cntl_3 &= ~SPLL_FB_DIV_MASK;
-        spll_func_cntl_3 |= SPLL_FB_DIV(fbdiv);
-        spll_func_cntl_3 |= SPLL_DITHEN;
+	spll_func_cntl_3 &= ~SPLL_FB_DIV_MASK;
+	spll_func_cntl_3 |= SPLL_FB_DIV(fbdiv);
+	spll_func_cntl_3 |= SPLL_DITHEN;
 
 	if (pi->sclk_ss) {
 		struct radeon_atom_ss ss;
@@ -4930,15 +4930,15 @@ static int si_populate_mclk_value(struct radeon_device *rdev,
 		tmp = freq_nom / reference_clock;
 		tmp = tmp * tmp;
 		if (radeon_atombios_get_asic_ss_info(rdev, &ss,
-                                                     ASIC_INTERNAL_MEMORY_SS, freq_nom)) {
+						     ASIC_INTERNAL_MEMORY_SS, freq_nom)) {
 			u32 clks = reference_clock * 5 / ss.rate;
 			u32 clkv = (u32)((((131 * ss.percentage * ss.rate) / 100) * tmp) / freq_nom);
 
-                        mpll_ss1 &= ~CLKV_MASK;
-                        mpll_ss1 |= CLKV(clkv);
+			mpll_ss1 &= ~CLKV_MASK;
+			mpll_ss1 |= CLKV(clkv);
 
-                        mpll_ss2 &= ~CLKS_MASK;
-                        mpll_ss2 |= CLKS(clks);
+			mpll_ss2 &= ~CLKS_MASK;
+			mpll_ss2 |= CLKS(clks);
 		}
 	}
 
@@ -5265,7 +5265,7 @@ static int si_convert_power_state_to_smc(struct radeon_device *rdev,
 		ni_pi->enable_power_containment = false;
 
 	ret = si_populate_sq_ramping_values(rdev, radeon_state, smc_state);
-        if (ret)
+	if (ret)
 		ni_pi->enable_sq_ramping = false;
 
 	return si_populate_smc_t(rdev, radeon_state, smc_state);
@@ -5436,46 +5436,46 @@ static bool si_check_s0_mc_reg_index(u16 in_reg, u16 *out_reg)
 	case  MC_SEQ_RAS_TIMING >> 2:
 		*out_reg = MC_SEQ_RAS_TIMING_LP >> 2;
 		break;
-        case MC_SEQ_CAS_TIMING >> 2:
+	case MC_SEQ_CAS_TIMING >> 2:
 		*out_reg = MC_SEQ_CAS_TIMING_LP >> 2;
 		break;
-        case MC_SEQ_MISC_TIMING >> 2:
+	case MC_SEQ_MISC_TIMING >> 2:
 		*out_reg = MC_SEQ_MISC_TIMING_LP >> 2;
 		break;
-        case MC_SEQ_MISC_TIMING2 >> 2:
+	case MC_SEQ_MISC_TIMING2 >> 2:
 		*out_reg = MC_SEQ_MISC_TIMING2_LP >> 2;
 		break;
-        case MC_SEQ_RD_CTL_D0 >> 2:
+	case MC_SEQ_RD_CTL_D0 >> 2:
 		*out_reg = MC_SEQ_RD_CTL_D0_LP >> 2;
 		break;
-        case MC_SEQ_RD_CTL_D1 >> 2:
+	case MC_SEQ_RD_CTL_D1 >> 2:
 		*out_reg = MC_SEQ_RD_CTL_D1_LP >> 2;
 		break;
-        case MC_SEQ_WR_CTL_D0 >> 2:
+	case MC_SEQ_WR_CTL_D0 >> 2:
 		*out_reg = MC_SEQ_WR_CTL_D0_LP >> 2;
 		break;
-        case MC_SEQ_WR_CTL_D1 >> 2:
+	case MC_SEQ_WR_CTL_D1 >> 2:
 		*out_reg = MC_SEQ_WR_CTL_D1_LP >> 2;
 		break;
-        case MC_PMG_CMD_EMRS >> 2:
+	case MC_PMG_CMD_EMRS >> 2:
 		*out_reg = MC_SEQ_PMG_CMD_EMRS_LP >> 2;
 		break;
-        case MC_PMG_CMD_MRS >> 2:
+	case MC_PMG_CMD_MRS >> 2:
 		*out_reg = MC_SEQ_PMG_CMD_MRS_LP >> 2;
 		break;
-        case MC_PMG_CMD_MRS1 >> 2:
+	case MC_PMG_CMD_MRS1 >> 2:
 		*out_reg = MC_SEQ_PMG_CMD_MRS1_LP >> 2;
 		break;
-        case MC_SEQ_PMG_TIMING >> 2:
+	case MC_SEQ_PMG_TIMING >> 2:
 		*out_reg = MC_SEQ_PMG_TIMING_LP >> 2;
 		break;
-        case MC_PMG_CMD_MRS2 >> 2:
+	case MC_PMG_CMD_MRS2 >> 2:
 		*out_reg = MC_SEQ_PMG_CMD_MRS2_LP >> 2;
 		break;
-        case MC_SEQ_WR_CTL_2 >> 2:
+	case MC_SEQ_WR_CTL_2 >> 2:
 		*out_reg = MC_SEQ_WR_CTL_2_LP >> 2;
 		break;
-        default:
+	default:
 		result = false;
 		break;
 	}
@@ -5562,19 +5562,19 @@ static int si_initialize_mc_reg_table(struct radeon_device *rdev)
 	WREG32(MC_SEQ_PMG_CMD_MRS2_LP, RREG32(MC_PMG_CMD_MRS2));
 	WREG32(MC_SEQ_WR_CTL_2_LP, RREG32(MC_SEQ_WR_CTL_2));
 
-        ret = radeon_atom_init_mc_reg_table(rdev, module_index, table);
-        if (ret)
-                goto init_mc_done;
+	ret = radeon_atom_init_mc_reg_table(rdev, module_index, table);
+	if (ret)
+		goto init_mc_done;
 
-        ret = si_copy_vbios_mc_reg_table(table, si_table);
-        if (ret)
-                goto init_mc_done;
+	ret = si_copy_vbios_mc_reg_table(table, si_table);
+	if (ret)
+		goto init_mc_done;
 
 	si_set_s0_mc_reg_index(si_table);
 
 	ret = si_set_mc_special_registers(rdev, si_table);
-        if (ret)
-                goto init_mc_done;
+	if (ret)
+		goto init_mc_done;
 
 	si_set_valid_flag(si_table);
 
@@ -5715,10 +5715,10 @@ static int si_upload_mc_reg_table(struct radeon_device *rdev,
 
 static void si_enable_voltage_control(struct radeon_device *rdev, bool enable)
 {
-        if (enable)
-                WREG32_P(GENERAL_PWRMGT, VOLT_PWRMGT_EN, ~VOLT_PWRMGT_EN);
-        else
-                WREG32_P(GENERAL_PWRMGT, 0, ~VOLT_PWRMGT_EN);
+	if (enable)
+		WREG32_P(GENERAL_PWRMGT, VOLT_PWRMGT_EN, ~VOLT_PWRMGT_EN);
+	else
+		WREG32_P(GENERAL_PWRMGT, 0, ~VOLT_PWRMGT_EN);
 }
 
 static enum radeon_pcie_gen si_get_maximum_link_speed(struct radeon_device *rdev,
@@ -6820,7 +6820,7 @@ static int si_parse_power_table(struct radeon_device *rdev)
 	struct _NonClockInfoArray *non_clock_info_array;
 	union power_info *power_info;
 	int index = GetIndexIntoMasterTable(DATA, PowerPlayInfo);
-        u16 data_offset;
+	u16 data_offset;
 	u8 frev, crev;
 	u8 *power_state_offset;
 	struct ni_ps *ps;
diff --git a/drivers/gpu/drm/radeon/sumo_dpm.c b/drivers/gpu/drm/radeon/sumo_dpm.c
index cd0862809adf..f0d5c1724f55 100644
--- a/drivers/gpu/drm/radeon/sumo_dpm.c
+++ b/drivers/gpu/drm/radeon/sumo_dpm.c
@@ -787,8 +787,8 @@ static void sumo_program_acpi_power_level(struct radeon_device *rdev)
 	struct atom_clock_dividers dividers;
 	int ret;
 
-        ret = radeon_atom_get_clock_dividers(rdev, COMPUTE_ENGINE_PLL_PARAM,
-                                             pi->acpi_pl.sclk,
+	ret = radeon_atom_get_clock_dividers(rdev, COMPUTE_ENGINE_PLL_PARAM,
+					     pi->acpi_pl.sclk,
 					     false, &dividers);
 	if (ret)
 		return;
@@ -1462,7 +1462,7 @@ static int sumo_parse_power_table(struct radeon_device *rdev)
 	struct _NonClockInfoArray *non_clock_info_array;
 	union power_info *power_info;
 	int index = GetIndexIntoMasterTable(DATA, PowerPlayInfo);
-        u16 data_offset;
+	u16 data_offset;
 	u8 frev, crev;
 	u8 *power_state_offset;
 	struct sumo_ps *ps;
diff --git a/drivers/gpu/drm/radeon/trinity_dpm.c b/drivers/gpu/drm/radeon/trinity_dpm.c
index d34bfcdab9be..6730367ac228 100644
--- a/drivers/gpu/drm/radeon/trinity_dpm.c
+++ b/drivers/gpu/drm/radeon/trinity_dpm.c
@@ -369,8 +369,8 @@ static void trinity_gfx_powergating_initialize(struct radeon_device *rdev)
 	int ret;
 	u32 hw_rev = (RREG32(HW_REV) & ATI_REV_ID_MASK) >> ATI_REV_ID_SHIFT;
 
-        ret = radeon_atom_get_clock_dividers(rdev, COMPUTE_ENGINE_PLL_PARAM,
-                                             25000, false, &dividers);
+	ret = radeon_atom_get_clock_dividers(rdev, COMPUTE_ENGINE_PLL_PARAM,
+					     25000, false, &dividers);
 	if (ret)
 		return;
 
@@ -587,8 +587,8 @@ static void trinity_set_divider_value(struct radeon_device *rdev,
 	u32 value;
 	u32 ix = index * TRINITY_SIZEOF_DPM_STATE_TABLE;
 
-        ret = radeon_atom_get_clock_dividers(rdev, COMPUTE_ENGINE_PLL_PARAM,
-                                             sclk, false, &dividers);
+	ret = radeon_atom_get_clock_dividers(rdev, COMPUTE_ENGINE_PLL_PARAM,
+					     sclk, false, &dividers);
 	if (ret)
 		return;
 
@@ -597,8 +597,8 @@ static void trinity_set_divider_value(struct radeon_device *rdev,
 	value |= CLK_DIVIDER(dividers.post_div);
 	WREG32_SMC(SMU_SCLK_DPM_STATE_0_CNTL_0 + ix, value);
 
-        ret = radeon_atom_get_clock_dividers(rdev, COMPUTE_ENGINE_PLL_PARAM,
-                                             sclk/2, false, &dividers);
+	ret = radeon_atom_get_clock_dividers(rdev, COMPUTE_ENGINE_PLL_PARAM,
+					     sclk/2, false, &dividers);
 	if (ret)
 		return;
 
@@ -1045,14 +1045,14 @@ static int trinity_set_thermal_temperature_range(struct radeon_device *rdev,
 	int low_temp = 0 * 1000;
 	int high_temp = 255 * 1000;
 
-        if (low_temp < min_temp)
+	if (low_temp < min_temp)
 		low_temp = min_temp;
-        if (high_temp > max_temp)
+	if (high_temp > max_temp)
 		high_temp = max_temp;
-        if (high_temp < low_temp) {
+	if (high_temp < low_temp) {
 		DRM_ERROR("invalid thermal range: %d - %d\n", low_temp, high_temp);
-                return -EINVAL;
-        }
+		return -EINVAL;
+	}
 
 	WREG32_P(CG_THERMAL_INT_CTRL, DIG_THERM_INTH(49 + (high_temp / 1000)), ~DIG_THERM_INTH_MASK);
 	WREG32_P(CG_THERMAL_INT_CTRL, DIG_THERM_INTL(49 + (low_temp / 1000)), ~DIG_THERM_INTL_MASK);
@@ -1737,7 +1737,7 @@ static int trinity_parse_power_table(struct radeon_device *rdev)
 	struct _NonClockInfoArray *non_clock_info_array;
 	union power_info *power_info;
 	int index = GetIndexIntoMasterTable(DATA, PowerPlayInfo);
-        u16 data_offset;
+	u16 data_offset;
 	u8 frev, crev;
 	u8 *power_state_offset;
 	struct sumo_ps *ps;
diff --git a/drivers/gpu/drm/radeon/vce_v2_0.c b/drivers/gpu/drm/radeon/vce_v2_0.c
index cdeaab7c7b1e..fce214482e72 100644
--- a/drivers/gpu/drm/radeon/vce_v2_0.c
+++ b/drivers/gpu/drm/radeon/vce_v2_0.c
@@ -53,7 +53,7 @@ static void vce_v2_0_set_sw_cg(struct radeon_device *rdev, bool gated)
 		WREG32(VCE_UENC_REG_CLOCK_GATING, tmp);
 
 		WREG32(VCE_CGTT_CLK_OVERRIDE, 0);
-    } else {
+	} else {
 		tmp = RREG32(VCE_CLOCK_GATING_B);
 		tmp |= 0xe7;
 		tmp &= ~0xe70000;
diff --git a/drivers/gpu/drm/rockchip/dw-mipi-dsi.c b/drivers/gpu/drm/rockchip/dw-mipi-dsi.c
index 77662068bc11..7975158064e8 100644
--- a/drivers/gpu/drm/rockchip/dw-mipi-dsi.c
+++ b/drivers/gpu/drm/rockchip/dw-mipi-dsi.c
@@ -878,7 +878,7 @@ static void dw_mipi_dsi_encoder_disable(struct drm_encoder *encoder)
 static void dw_mipi_dsi_encoder_commit(struct drm_encoder *encoder)
 {
 	struct dw_mipi_dsi *dsi = encoder_to_dsi(encoder);
-	int mux  = rockchip_drm_encoder_get_mux_id(dsi->dev->of_node, encoder);
+	int mux = drm_of_encoder_active_endpoint_id(dsi->dev->of_node, encoder);
 	u32 interface_pix_fmt;
 	u32 val;
 
diff --git a/drivers/gpu/drm/rockchip/dw_hdmi-rockchip.c b/drivers/gpu/drm/rockchip/dw_hdmi-rockchip.c
index c65ce8cb30d3..3d3cf2f8891e 100644
--- a/drivers/gpu/drm/rockchip/dw_hdmi-rockchip.c
+++ b/drivers/gpu/drm/rockchip/dw_hdmi-rockchip.c
@@ -204,7 +204,7 @@ static void dw_hdmi_rockchip_encoder_enable(struct drm_encoder *encoder)
 	rockchip_drm_crtc_mode_config(encoder->crtc, DRM_MODE_CONNECTOR_HDMIA,
 				      ROCKCHIP_OUT_MODE_AAAA);
 
-	mux = rockchip_drm_encoder_get_mux_id(hdmi->dev->of_node, encoder);
+	mux = drm_of_encoder_active_endpoint_id(hdmi->dev->of_node, encoder);
 	if (mux)
 		val = HDMI_SEL_VOP_LIT | (HDMI_SEL_VOP_LIT << 16);
 	else
diff --git a/drivers/gpu/drm/rockchip/rockchip_drm_drv.c b/drivers/gpu/drm/rockchip/rockchip_drm_drv.c
index a0d51ccb6ea4..896da09e49ee 100644
--- a/drivers/gpu/drm/rockchip/rockchip_drm_drv.c
+++ b/drivers/gpu/drm/rockchip/rockchip_drm_drv.c
@@ -384,36 +384,6 @@ static const struct dev_pm_ops rockchip_drm_pm_ops = {
 				rockchip_drm_sys_resume)
 };
 
-/*
- * @node: device tree node containing encoder input ports
- * @encoder: drm_encoder
- */
-int rockchip_drm_encoder_get_mux_id(struct device_node *node,
-				    struct drm_encoder *encoder)
-{
-	struct device_node *ep;
-	struct drm_crtc *crtc = encoder->crtc;
-	struct of_endpoint endpoint;
-	struct device_node *port;
-	int ret;
-
-	if (!node || !crtc)
-		return -EINVAL;
-
-	for_each_endpoint_of_node(node, ep) {
-		port = of_graph_get_remote_port(ep);
-		of_node_put(port);
-		if (port == crtc->port) {
-			ret = of_graph_parse_endpoint(ep, &endpoint);
-			of_node_put(ep);
-			return ret ?: endpoint.id;
-		}
-	}
-
-	return -EINVAL;
-}
-EXPORT_SYMBOL_GPL(rockchip_drm_encoder_get_mux_id);
-
 static int compare_of(struct device *dev, void *data)
 {
 	struct device_node *np = data;
diff --git a/drivers/gpu/drm/rockchip/rockchip_drm_drv.h b/drivers/gpu/drm/rockchip/rockchip_drm_drv.h
index bb8b076f1dbb..3529f692edb8 100644
--- a/drivers/gpu/drm/rockchip/rockchip_drm_drv.h
+++ b/drivers/gpu/drm/rockchip/rockchip_drm_drv.h
@@ -67,8 +67,6 @@ void rockchip_drm_atomic_work(struct work_struct *work);
 int rockchip_register_crtc_funcs(struct drm_crtc *crtc,
 				 const struct rockchip_crtc_funcs *crtc_funcs);
 void rockchip_unregister_crtc_funcs(struct drm_crtc *crtc);
-int rockchip_drm_encoder_get_mux_id(struct device_node *node,
-				    struct drm_encoder *encoder);
 int rockchip_drm_crtc_mode_config(struct drm_crtc *crtc, int connector_type,
 				  int out_mode);
 int rockchip_drm_dma_attach_device(struct drm_device *drm_dev,
diff --git a/drivers/gpu/drm/sti/sti_awg_utils.c b/drivers/gpu/drm/sti/sti_awg_utils.c
index 00d0698be9d3..a516eb869f6f 100644
--- a/drivers/gpu/drm/sti/sti_awg_utils.c
+++ b/drivers/gpu/drm/sti/sti_awg_utils.c
@@ -7,6 +7,7 @@
 #include "sti_awg_utils.h"
 
 #define AWG_OPCODE_OFFSET 10
+#define AWG_MAX_ARG       0x3ff
 
 enum opcode {
 	SET,
@@ -34,6 +35,8 @@ static int awg_generate_instr(enum opcode opcode,
 	/* skip, repeat and replay arg should not exceed 1023.
 	 * If user wants to exceed this value, the instruction should be
 	 * duplicate and arg should be adjust for each duplicated instruction.
+	 *
+	 * mux_sel is used in case of SAV/EAV synchronization.
 	 */
 
 	while (arg_tmp > 0) {
@@ -65,7 +68,7 @@ static int awg_generate_instr(enum opcode opcode,
 
 			mux = 0;
 			data_enable = 0;
-			arg &= (0x3ff);
+			arg &= AWG_MAX_ARG;
 			break;
 		case REPEAT:
 		case REPLAY:
@@ -76,13 +79,13 @@ static int awg_generate_instr(enum opcode opcode,
 
 			mux = 0;
 			data_enable = 0;
-			arg &= (0x3ff);
+			arg &= AWG_MAX_ARG;
 			break;
 		case JUMP:
 			mux = 0;
 			data_enable = 0;
 			arg |= 0x40; /* for jump instruction 7th bit is 1 */
-			arg &= 0x3ff;
+			arg &= AWG_MAX_ARG;
 			break;
 		case STOP:
 			arg = 0;
@@ -110,68 +113,75 @@ static int awg_generate_instr(enum opcode opcode,
 	return 0;
 }
 
-int sti_awg_generate_code_data_enable_mode(
+static int awg_generate_line_signal(
 		struct awg_code_generation_params *fwparams,
 		struct awg_timing *timing)
 {
 	long int val;
-	long int data_en;
 	int ret = 0;
 
-	if (timing->trailing_lines > 0) {
-		/* skip trailing lines */
-		val = timing->blanking_level;
-		data_en = 0;
-		ret |= awg_generate_instr(RPLSET, val, 0, data_en, fwparams);
-
-		val = timing->trailing_lines - 1;
-		data_en = 0;
-		ret |= awg_generate_instr(REPLAY, val, 0, data_en, fwparams);
-	}
-
 	if (timing->trailing_pixels > 0) {
 		/* skip trailing pixel */
 		val = timing->blanking_level;
-		data_en = 0;
-		ret |= awg_generate_instr(RPLSET, val, 0, data_en, fwparams);
+		ret |= awg_generate_instr(RPLSET, val, 0, 0, fwparams);
 
 		val = timing->trailing_pixels - 1;
-		data_en = 0;
-		ret |= awg_generate_instr(SKIP, val, 0, data_en, fwparams);
+		ret |= awg_generate_instr(SKIP, val, 0, 0, fwparams);
 	}
 
 	/* set DE signal high */
 	val = timing->blanking_level;
-	data_en = 1;
 	ret |= awg_generate_instr((timing->trailing_pixels > 0) ? SET : RPLSET,
-			val, 0, data_en, fwparams);
+			val, 0, 1, fwparams);
 
 	if (timing->blanking_pixels > 0) {
 		/* skip the number of active pixel */
 		val = timing->active_pixels - 1;
-		data_en = 1;
-		ret |= awg_generate_instr(SKIP, val, 0, data_en, fwparams);
+		ret |= awg_generate_instr(SKIP, val, 0, 1, fwparams);
 
 		/* set DE signal low */
 		val = timing->blanking_level;
-		data_en = 0;
-		ret |= awg_generate_instr(SET, val, 0, data_en, fwparams);
+		ret |= awg_generate_instr(SET, val, 0, 0, fwparams);
+	}
+
+	return ret;
+}
+
+int sti_awg_generate_code_data_enable_mode(
+		struct awg_code_generation_params *fwparams,
+		struct awg_timing *timing)
+{
+	long int val, tmp_val;
+	int ret = 0;
+
+	if (timing->trailing_lines > 0) {
+		/* skip trailing lines */
+		val = timing->blanking_level;
+		ret |= awg_generate_instr(RPLSET, val, 0, 0, fwparams);
+
+		val = timing->trailing_lines - 1;
+		ret |= awg_generate_instr(REPLAY, val, 0, 0, fwparams);
 	}
 
-	/* replay the sequence as many active lines defined */
-	val = timing->active_lines - 1;
-	data_en = 0;
-	ret |= awg_generate_instr(REPLAY, val, 0, data_en, fwparams);
+	tmp_val = timing->active_lines - 1;
+
+	while (tmp_val > 0) {
+		/* generate DE signal for each line */
+		ret |= awg_generate_line_signal(fwparams, timing);
+		/* replay the sequence as many active lines defined */
+		ret |= awg_generate_instr(REPLAY,
+					  min_t(int, AWG_MAX_ARG, tmp_val),
+					  0, 0, fwparams);
+		tmp_val -= AWG_MAX_ARG;
+	}
 
 	if (timing->blanking_lines > 0) {
 		/* skip blanking lines */
 		val = timing->blanking_level;
-		data_en = 0;
-		ret |= awg_generate_instr(RPLSET, val, 0, data_en, fwparams);
+		ret |= awg_generate_instr(RPLSET, val, 0, 0, fwparams);
 
 		val = timing->blanking_lines - 1;
-		data_en = 0;
-		ret |= awg_generate_instr(REPLAY, val, 0, data_en, fwparams);
+		ret |= awg_generate_instr(REPLAY, val, 0, 0, fwparams);
 	}
 
 	return ret;
diff --git a/drivers/gpu/drm/sti/sti_compositor.c b/drivers/gpu/drm/sti/sti_compositor.c
index afed2171beb9..3d2fa3ab33df 100644
--- a/drivers/gpu/drm/sti/sti_compositor.c
+++ b/drivers/gpu/drm/sti/sti_compositor.c
@@ -75,13 +75,13 @@ static int sti_compositor_bind(struct device *dev,
 		switch (desc[i].type) {
 		case STI_VID_SUBDEV:
 			compo->vid[vid_id++] =
-			    sti_vid_create(compo->dev, desc[i].id,
+			    sti_vid_create(compo->dev, drm_dev, desc[i].id,
 					   compo->regs + desc[i].offset);
 			break;
 		case STI_MIXER_MAIN_SUBDEV:
 		case STI_MIXER_AUX_SUBDEV:
 			compo->mixer[mixer_id++] =
-			    sti_mixer_create(compo->dev, desc[i].id,
+			    sti_mixer_create(compo->dev, drm_dev, desc[i].id,
 					     compo->regs + desc[i].offset);
 			break;
 		case STI_GPD_SUBDEV:
diff --git a/drivers/gpu/drm/sti/sti_crtc.c b/drivers/gpu/drm/sti/sti_crtc.c
index e04deedabd4a..505620c7c2c8 100644
--- a/drivers/gpu/drm/sti/sti_crtc.c
+++ b/drivers/gpu/drm/sti/sti_crtc.c
@@ -51,6 +51,15 @@ static void sti_crtc_disabling(struct drm_crtc *crtc)
 	mixer->status = STI_MIXER_DISABLING;
 }
 
+static bool sti_crtc_mode_fixup(struct drm_crtc *crtc,
+				const struct drm_display_mode *mode,
+				struct drm_display_mode *adjusted_mode)
+{
+	/* accept the provided drm_display_mode, do not fix it up */
+	drm_mode_set_crtcinfo(adjusted_mode, CRTC_INTERLACE_HALVE_V);
+	return true;
+}
+
 static int
 sti_crtc_mode_set(struct drm_crtc *crtc, struct drm_display_mode *mode)
 {
@@ -221,6 +230,7 @@ static void sti_crtc_atomic_flush(struct drm_crtc *crtc,
 static const struct drm_crtc_helper_funcs sti_crtc_helper_funcs = {
 	.enable = sti_crtc_enable,
 	.disable = sti_crtc_disabling,
+	.mode_fixup = sti_crtc_mode_fixup,
 	.mode_set = drm_helper_crtc_mode_set,
 	.mode_set_nofb = sti_crtc_mode_set_nofb,
 	.mode_set_base = drm_helper_crtc_mode_set_base,
diff --git a/drivers/gpu/drm/sti/sti_cursor.c b/drivers/gpu/drm/sti/sti_cursor.c
index 807863106b8d..82b5711fefef 100644
--- a/drivers/gpu/drm/sti/sti_cursor.c
+++ b/drivers/gpu/drm/sti/sti_cursor.c
@@ -5,12 +5,10 @@
  *          for STMicroelectronics.
  * License terms:  GNU General Public License (GPL), version 2
  */
-#include <drm/drmP.h>
 
-#include <drm/drm_atomic_helper.h>
+#include <drm/drm_atomic.h>
 #include <drm/drm_fb_cma_helper.h>
 #include <drm/drm_gem_cma_helper.h>
-#include <drm/drm_plane_helper.h>
 
 #include "sti_compositor.h"
 #include "sti_cursor.h"
@@ -74,6 +72,82 @@ static const uint32_t cursor_supported_formats[] = {
 
 #define to_sti_cursor(x) container_of(x, struct sti_cursor, plane)
 
+#define DBGFS_DUMP(reg) seq_printf(s, "\n  %-25s 0x%08X", #reg, \
+				   readl(cursor->regs + reg))
+
+static void cursor_dbg_vpo(struct seq_file *s, u32 val)
+{
+	seq_printf(s, "\txdo:%4d\tydo:%4d", val & 0x0FFF, (val >> 16) & 0x0FFF);
+}
+
+static void cursor_dbg_size(struct seq_file *s, u32 val)
+{
+	seq_printf(s, "\t%d x %d", val & 0x07FF, (val >> 16) & 0x07FF);
+}
+
+static void cursor_dbg_pml(struct seq_file *s,
+			   struct sti_cursor *cursor, u32 val)
+{
+	if (cursor->pixmap.paddr == val)
+		seq_printf(s, "\tVirt @: %p", cursor->pixmap.base);
+}
+
+static void cursor_dbg_cml(struct seq_file *s,
+			   struct sti_cursor *cursor, u32 val)
+{
+	if (cursor->clut_paddr == val)
+		seq_printf(s, "\tVirt @: %p", cursor->clut);
+}
+
+static int cursor_dbg_show(struct seq_file *s, void *data)
+{
+	struct drm_info_node *node = s->private;
+	struct sti_cursor *cursor = (struct sti_cursor *)node->info_ent->data;
+	struct drm_device *dev = node->minor->dev;
+	int ret;
+
+	ret = mutex_lock_interruptible(&dev->struct_mutex);
+	if (ret)
+		return ret;
+
+	seq_printf(s, "%s: (vaddr = 0x%p)",
+		   sti_plane_to_str(&cursor->plane), cursor->regs);
+
+	DBGFS_DUMP(CUR_CTL);
+	DBGFS_DUMP(CUR_VPO);
+	cursor_dbg_vpo(s, readl(cursor->regs + CUR_VPO));
+	DBGFS_DUMP(CUR_PML);
+	cursor_dbg_pml(s, cursor, readl(cursor->regs + CUR_PML));
+	DBGFS_DUMP(CUR_PMP);
+	DBGFS_DUMP(CUR_SIZE);
+	cursor_dbg_size(s, readl(cursor->regs + CUR_SIZE));
+	DBGFS_DUMP(CUR_CML);
+	cursor_dbg_cml(s, cursor, readl(cursor->regs + CUR_CML));
+	DBGFS_DUMP(CUR_AWS);
+	DBGFS_DUMP(CUR_AWE);
+	seq_puts(s, "\n");
+
+	mutex_unlock(&dev->struct_mutex);
+	return 0;
+}
+
+static struct drm_info_list cursor_debugfs_files[] = {
+	{ "cursor", cursor_dbg_show, 0, NULL },
+};
+
+static int cursor_debugfs_init(struct sti_cursor *cursor,
+			       struct drm_minor *minor)
+{
+	unsigned int i;
+
+	for (i = 0; i < ARRAY_SIZE(cursor_debugfs_files); i++)
+		cursor_debugfs_files[i].data = cursor;
+
+	return drm_debugfs_create_files(cursor_debugfs_files,
+					ARRAY_SIZE(cursor_debugfs_files),
+					minor->debugfs_root, minor);
+}
+
 static void sti_cursor_argb8888_to_clut8(struct sti_cursor *cursor, u32 *src)
 {
 	u8  *dst = cursor->pixmap.base;
@@ -110,35 +184,31 @@ static void sti_cursor_init(struct sti_cursor *cursor)
 						  (b * 5);
 }
 
-static void sti_cursor_atomic_update(struct drm_plane *drm_plane,
-				     struct drm_plane_state *oldstate)
+static int sti_cursor_atomic_check(struct drm_plane *drm_plane,
+				   struct drm_plane_state *state)
 {
-	struct drm_plane_state *state = drm_plane->state;
 	struct sti_plane *plane = to_sti_plane(drm_plane);
 	struct sti_cursor *cursor = to_sti_cursor(plane);
 	struct drm_crtc *crtc = state->crtc;
-	struct sti_mixer *mixer = to_sti_mixer(crtc);
 	struct drm_framebuffer *fb = state->fb;
-	struct drm_display_mode *mode = &crtc->mode;
-	int dst_x = state->crtc_x;
-	int dst_y = state->crtc_y;
-	int dst_w = clamp_val(state->crtc_w, 0, mode->crtc_hdisplay - dst_x);
-	int dst_h = clamp_val(state->crtc_h, 0, mode->crtc_vdisplay - dst_y);
+	struct drm_crtc_state *crtc_state;
+	struct drm_display_mode *mode;
+	int dst_x, dst_y, dst_w, dst_h;
+	int src_w, src_h;
+
+	/* no need for further checks if the plane is being disabled */
+	if (!crtc || !fb)
+		return 0;
+
+	crtc_state = drm_atomic_get_crtc_state(state->state, crtc);
+	mode = &crtc_state->mode;
+	dst_x = state->crtc_x;
+	dst_y = state->crtc_y;
+	dst_w = clamp_val(state->crtc_w, 0, mode->crtc_hdisplay - dst_x);
+	dst_h = clamp_val(state->crtc_h, 0, mode->crtc_vdisplay - dst_y);
 	/* src_x are in 16.16 format */
-	int src_w = state->src_w >> 16;
-	int src_h = state->src_h >> 16;
-	bool first_prepare = plane->status == STI_PLANE_DISABLED ? true : false;
-	struct drm_gem_cma_object *cma_obj;
-	u32 y, x;
-	u32 val;
-
-	DRM_DEBUG_KMS("CRTC:%d (%s) drm plane:%d (%s)\n",
-		      crtc->base.id, sti_mixer_to_str(mixer),
-		      drm_plane->base.id, sti_plane_to_str(plane));
-	DRM_DEBUG_KMS("(%dx%d)@(%d,%d)\n", dst_w, dst_h, dst_x, dst_y);
-
-	dev_dbg(cursor->dev, "%s %s\n", __func__,
-		sti_plane_to_str(plane));
+	src_w = state->src_w >> 16;
+	src_h = state->src_h >> 16;
 
 	if (src_w < STI_CURS_MIN_SIZE ||
 	    src_h < STI_CURS_MIN_SIZE ||
@@ -146,7 +216,7 @@ static void sti_cursor_atomic_update(struct drm_plane *drm_plane,
 	    src_h > STI_CURS_MAX_SIZE) {
 		DRM_ERROR("Invalid cursor size (%dx%d)\n",
 				src_w, src_h);
-		return;
+		return -EINVAL;
 	}
 
 	/* If the cursor size has changed, re-allocated the pixmap */
@@ -170,16 +240,46 @@ static void sti_cursor_atomic_update(struct drm_plane *drm_plane,
 							GFP_KERNEL | GFP_DMA);
 		if (!cursor->pixmap.base) {
 			DRM_ERROR("Failed to allocate memory for pixmap\n");
-			return;
+			return -EINVAL;
 		}
 	}
 
-	cma_obj = drm_fb_cma_get_gem_obj(fb, 0);
-	if (!cma_obj) {
+	if (!drm_fb_cma_get_gem_obj(fb, 0)) {
 		DRM_ERROR("Can't get CMA GEM object for fb\n");
-		return;
+		return -EINVAL;
 	}
 
+	DRM_DEBUG_KMS("CRTC:%d (%s) drm plane:%d (%s)\n",
+		      crtc->base.id, sti_mixer_to_str(to_sti_mixer(crtc)),
+		      drm_plane->base.id, sti_plane_to_str(plane));
+	DRM_DEBUG_KMS("(%dx%d)@(%d,%d)\n", dst_w, dst_h, dst_x, dst_y);
+
+	return 0;
+}
+
+static void sti_cursor_atomic_update(struct drm_plane *drm_plane,
+				     struct drm_plane_state *oldstate)
+{
+	struct drm_plane_state *state = drm_plane->state;
+	struct sti_plane *plane = to_sti_plane(drm_plane);
+	struct sti_cursor *cursor = to_sti_cursor(plane);
+	struct drm_crtc *crtc = state->crtc;
+	struct drm_framebuffer *fb = state->fb;
+	struct drm_display_mode *mode;
+	int dst_x, dst_y;
+	struct drm_gem_cma_object *cma_obj;
+	u32 y, x;
+	u32 val;
+
+	if (!crtc || !fb)
+		return;
+
+	mode = &crtc->mode;
+	dst_x = state->crtc_x;
+	dst_y = state->crtc_y;
+
+	cma_obj = drm_fb_cma_get_gem_obj(fb, 0);
+
 	/* Convert ARGB8888 to CLUT8 */
 	sti_cursor_argb8888_to_clut8(cursor, (u32 *)cma_obj->vaddr);
 
@@ -193,21 +293,21 @@ static void sti_cursor_atomic_update(struct drm_plane *drm_plane,
 	val = y << 16 | x;
 	writel(val, cursor->regs + CUR_AWE);
 
-	if (first_prepare) {
-		/* Set and fetch CLUT */
-		writel(cursor->clut_paddr, cursor->regs + CUR_CML);
-		writel(CUR_CTL_CLUT_UPDATE, cursor->regs + CUR_CTL);
-	}
-
 	/* Set memory location, size, and position */
 	writel(cursor->pixmap.paddr, cursor->regs + CUR_PML);
 	writel(cursor->width, cursor->regs + CUR_PMP);
 	writel(cursor->height << 16 | cursor->width, cursor->regs + CUR_SIZE);
 
 	y = sti_vtg_get_line_number(*mode, dst_y);
-	x = sti_vtg_get_pixel_number(*mode, dst_y);
+	x = sti_vtg_get_pixel_number(*mode, dst_x);
 	writel((y << 16) | x, cursor->regs + CUR_VPO);
 
+	/* Set and fetch CLUT */
+	writel(cursor->clut_paddr, cursor->regs + CUR_CML);
+	writel(CUR_CTL_CLUT_UPDATE, cursor->regs + CUR_CTL);
+
+	sti_plane_update_fps(plane, true, false);
+
 	plane->status = STI_PLANE_UPDATED;
 }
 
@@ -215,7 +315,6 @@ static void sti_cursor_atomic_disable(struct drm_plane *drm_plane,
 				      struct drm_plane_state *oldstate)
 {
 	struct sti_plane *plane = to_sti_plane(drm_plane);
-	struct sti_mixer *mixer = to_sti_mixer(drm_plane->crtc);
 
 	if (!drm_plane->crtc) {
 		DRM_DEBUG_DRIVER("drm plane:%d not enabled\n",
@@ -224,13 +323,15 @@ static void sti_cursor_atomic_disable(struct drm_plane *drm_plane,
 	}
 
 	DRM_DEBUG_DRIVER("CRTC:%d (%s) drm plane:%d (%s)\n",
-			 drm_plane->crtc->base.id, sti_mixer_to_str(mixer),
+			 drm_plane->crtc->base.id,
+			 sti_mixer_to_str(to_sti_mixer(drm_plane->crtc)),
 			 drm_plane->base.id, sti_plane_to_str(plane));
 
 	plane->status = STI_PLANE_DISABLING;
 }
 
 static const struct drm_plane_helper_funcs sti_cursor_helpers_funcs = {
+	.atomic_check = sti_cursor_atomic_check,
 	.atomic_update = sti_cursor_atomic_update,
 	.atomic_disable = sti_cursor_atomic_disable,
 };
@@ -283,6 +384,9 @@ struct drm_plane *sti_cursor_create(struct drm_device *drm_dev,
 
 	sti_plane_init_property(&cursor->plane, DRM_PLANE_TYPE_CURSOR);
 
+	if (cursor_debugfs_init(cursor, drm_dev->primary))
+		DRM_ERROR("CURSOR debugfs setup failed\n");
+
 	return &cursor->plane.drm_plane;
 
 err_plane:
diff --git a/drivers/gpu/drm/sti/sti_drv.c b/drivers/gpu/drm/sti/sti_drv.c
index 506b5626f3ed..6bd6abaa5a70 100644
--- a/drivers/gpu/drm/sti/sti_drv.c
+++ b/drivers/gpu/drm/sti/sti_drv.c
@@ -20,6 +20,7 @@
 
 #include "sti_crtc.h"
 #include "sti_drv.h"
+#include "sti_plane.h"
 
 #define DRIVER_NAME	"sti"
 #define DRIVER_DESC	"STMicroelectronics SoC DRM"
@@ -30,6 +31,130 @@
 #define STI_MAX_FB_HEIGHT	4096
 #define STI_MAX_FB_WIDTH	4096
 
+static int sti_drm_fps_get(void *data, u64 *val)
+{
+	struct drm_device *drm_dev = data;
+	struct drm_plane *p;
+	unsigned int i = 0;
+
+	*val = 0;
+	list_for_each_entry(p, &drm_dev->mode_config.plane_list, head) {
+		struct sti_plane *plane = to_sti_plane(p);
+
+		*val |= plane->fps_info.output << i;
+		i++;
+	}
+
+	return 0;
+}
+
+static int sti_drm_fps_set(void *data, u64 val)
+{
+	struct drm_device *drm_dev = data;
+	struct drm_plane *p;
+	unsigned int i = 0;
+
+	list_for_each_entry(p, &drm_dev->mode_config.plane_list, head) {
+		struct sti_plane *plane = to_sti_plane(p);
+
+		plane->fps_info.output = (val >> i) & 1;
+		i++;
+	}
+
+	return 0;
+}
+
+DEFINE_SIMPLE_ATTRIBUTE(sti_drm_fps_fops,
+			sti_drm_fps_get, sti_drm_fps_set, "%llu\n");
+
+static int sti_drm_fps_dbg_show(struct seq_file *s, void *data)
+{
+	struct drm_info_node *node = s->private;
+	struct drm_device *dev = node->minor->dev;
+	struct drm_plane *p;
+	int ret;
+
+	ret = mutex_lock_interruptible(&dev->struct_mutex);
+	if (ret)
+		return ret;
+
+	list_for_each_entry(p, &dev->mode_config.plane_list, head) {
+		struct sti_plane *plane = to_sti_plane(p);
+
+		seq_printf(s, "%s%s\n",
+			   plane->fps_info.fps_str,
+			   plane->fps_info.fips_str);
+	}
+
+	mutex_unlock(&dev->struct_mutex);
+	return 0;
+}
+
+static struct drm_info_list sti_drm_dbg_list[] = {
+	{"fps_get", sti_drm_fps_dbg_show, 0},
+};
+
+static int sti_drm_debugfs_create(struct dentry *root,
+				  struct drm_minor *minor,
+				  const char *name,
+				  const struct file_operations *fops)
+{
+	struct drm_device *dev = minor->dev;
+	struct drm_info_node *node;
+	struct dentry *ent;
+
+	ent = debugfs_create_file(name, S_IRUGO | S_IWUSR, root, dev, fops);
+	if (IS_ERR(ent))
+		return PTR_ERR(ent);
+
+	node = kmalloc(sizeof(*node), GFP_KERNEL);
+	if (!node) {
+		debugfs_remove(ent);
+		return -ENOMEM;
+	}
+
+	node->minor = minor;
+	node->dent = ent;
+	node->info_ent = (void *)fops;
+
+	mutex_lock(&minor->debugfs_lock);
+	list_add(&node->list, &minor->debugfs_list);
+	mutex_unlock(&minor->debugfs_lock);
+
+	return 0;
+}
+
+static int sti_drm_dbg_init(struct drm_minor *minor)
+{
+	int ret;
+
+	ret = drm_debugfs_create_files(sti_drm_dbg_list,
+				       ARRAY_SIZE(sti_drm_dbg_list),
+				       minor->debugfs_root, minor);
+	if (ret)
+		goto err;
+
+	ret = sti_drm_debugfs_create(minor->debugfs_root, minor, "fps_show",
+				     &sti_drm_fps_fops);
+	if (ret)
+		goto err;
+
+	DRM_INFO("%s: debugfs installed\n", DRIVER_NAME);
+	return 0;
+err:
+	DRM_ERROR("%s: cannot install debugfs\n", DRIVER_NAME);
+	return ret;
+}
+
+void sti_drm_dbg_cleanup(struct drm_minor *minor)
+{
+	drm_debugfs_remove_files(sti_drm_dbg_list,
+				 ARRAY_SIZE(sti_drm_dbg_list), minor);
+
+	drm_debugfs_remove_files((struct drm_info_list *)&sti_drm_fps_fops,
+				 1, minor);
+}
+
 static void sti_atomic_schedule(struct sti_private *private,
 				struct drm_atomic_state *state)
 {
@@ -181,18 +306,9 @@ static const struct file_operations sti_driver_fops = {
 	.release = drm_release,
 };
 
-static struct dma_buf *sti_gem_prime_export(struct drm_device *dev,
-					    struct drm_gem_object *obj,
-					    int flags)
-{
-	/* we want to be able to write in mmapped buffer */
-	flags |= O_RDWR;
-	return drm_gem_prime_export(dev, obj, flags);
-}
-
 static struct drm_driver sti_driver = {
 	.driver_features = DRIVER_HAVE_IRQ | DRIVER_MODESET |
-	    DRIVER_GEM | DRIVER_PRIME,
+	    DRIVER_GEM | DRIVER_PRIME | DRIVER_ATOMIC,
 	.load = sti_load,
 	.gem_free_object = drm_gem_cma_free_object,
 	.gem_vm_ops = &drm_gem_cma_vm_ops,
@@ -207,7 +323,7 @@ static struct drm_driver sti_driver = {
 
 	.prime_handle_to_fd = drm_gem_prime_handle_to_fd,
 	.prime_fd_to_handle = drm_gem_prime_fd_to_handle,
-	.gem_prime_export = sti_gem_prime_export,
+	.gem_prime_export = drm_gem_prime_export,
 	.gem_prime_import = drm_gem_prime_import,
 	.gem_prime_get_sg_table = drm_gem_cma_prime_get_sg_table,
 	.gem_prime_import_sg_table = drm_gem_cma_prime_import_sg_table,
@@ -215,6 +331,9 @@ static struct drm_driver sti_driver = {
 	.gem_prime_vunmap = drm_gem_cma_prime_vunmap,
 	.gem_prime_mmap = drm_gem_cma_prime_mmap,
 
+	.debugfs_init = sti_drm_dbg_init,
+	.debugfs_cleanup = sti_drm_dbg_cleanup,
+
 	.name = DRIVER_NAME,
 	.desc = DRIVER_DESC,
 	.date = DRIVER_DATE,
diff --git a/drivers/gpu/drm/sti/sti_dvo.c b/drivers/gpu/drm/sti/sti_dvo.c
index 45cbe2bf7dd6..25f76632002c 100644
--- a/drivers/gpu/drm/sti/sti_dvo.c
+++ b/drivers/gpu/drm/sti/sti_dvo.c
@@ -6,6 +6,7 @@
 
 #include <linux/clk.h>
 #include <linux/component.h>
+#include <linux/debugfs.h>
 #include <linux/module.h>
 #include <linux/of_gpio.h>
 #include <linux/platform_device.h>
@@ -156,6 +157,69 @@ static void dvo_awg_configure(struct sti_dvo *dvo, u32 *awg_ram_code, int nb)
 	writel(DVO_AWG_CTRL_EN, dvo->regs + DVO_AWG_DIGSYNC_CTRL);
 }
 
+#define DBGFS_DUMP(reg) seq_printf(s, "\n  %-25s 0x%08X", #reg, \
+				   readl(dvo->regs + reg))
+
+static void dvo_dbg_awg_microcode(struct seq_file *s, void __iomem *reg)
+{
+	unsigned int i;
+
+	seq_puts(s, "\n\n");
+	seq_puts(s, "  DVO AWG microcode:");
+	for (i = 0; i < AWG_MAX_INST; i++) {
+		if (i % 8 == 0)
+			seq_printf(s, "\n  %04X:", i);
+		seq_printf(s, " %04X", readl(reg + i * 4));
+	}
+}
+
+static int dvo_dbg_show(struct seq_file *s, void *data)
+{
+	struct drm_info_node *node = s->private;
+	struct sti_dvo *dvo = (struct sti_dvo *)node->info_ent->data;
+	struct drm_device *dev = node->minor->dev;
+	int ret;
+
+	ret = mutex_lock_interruptible(&dev->struct_mutex);
+	if (ret)
+		return ret;
+
+	seq_printf(s, "DVO: (vaddr = 0x%p)", dvo->regs);
+	DBGFS_DUMP(DVO_AWG_DIGSYNC_CTRL);
+	DBGFS_DUMP(DVO_DOF_CFG);
+	DBGFS_DUMP(DVO_LUT_PROG_LOW);
+	DBGFS_DUMP(DVO_LUT_PROG_MID);
+	DBGFS_DUMP(DVO_LUT_PROG_HIGH);
+	dvo_dbg_awg_microcode(s, dvo->regs + DVO_DIGSYNC_INSTR_I);
+	seq_puts(s, "\n");
+
+	mutex_unlock(&dev->struct_mutex);
+	return 0;
+}
+
+static struct drm_info_list dvo_debugfs_files[] = {
+	{ "dvo", dvo_dbg_show, 0, NULL },
+};
+
+static void dvo_debugfs_exit(struct sti_dvo *dvo, struct drm_minor *minor)
+{
+	drm_debugfs_remove_files(dvo_debugfs_files,
+				 ARRAY_SIZE(dvo_debugfs_files),
+				 minor);
+}
+
+static int dvo_debugfs_init(struct sti_dvo *dvo, struct drm_minor *minor)
+{
+	unsigned int i;
+
+	for (i = 0; i < ARRAY_SIZE(dvo_debugfs_files); i++)
+		dvo_debugfs_files[i].data = dvo;
+
+	return drm_debugfs_create_files(dvo_debugfs_files,
+					ARRAY_SIZE(dvo_debugfs_files),
+					minor->debugfs_root, minor);
+}
+
 static void sti_dvo_disable(struct drm_bridge *bridge)
 {
 	struct sti_dvo *dvo = bridge->driver_private;
@@ -345,12 +409,14 @@ sti_dvo_connector_detect(struct drm_connector *connector, bool force)
 
 	DRM_DEBUG_DRIVER("\n");
 
-	if (!dvo->panel)
+	if (!dvo->panel) {
 		dvo->panel = of_drm_find_panel(dvo->panel_node);
+		if (dvo->panel)
+			drm_panel_attach(dvo->panel, connector);
+	}
 
 	if (dvo->panel)
-		if (!drm_panel_attach(dvo->panel, connector))
-			return connector_status_connected;
+		return connector_status_connected;
 
 	return connector_status_disconnected;
 }
@@ -453,6 +519,9 @@ static int sti_dvo_bind(struct device *dev, struct device *master, void *data)
 		goto err_sysfs;
 	}
 
+	if (dvo_debugfs_init(dvo, drm_dev->primary))
+		DRM_ERROR("DVO debugfs setup failed\n");
+
 	return 0;
 
 err_sysfs:
@@ -467,6 +536,9 @@ static void sti_dvo_unbind(struct device *dev,
 			   struct device *master, void *data)
 {
 	struct sti_dvo *dvo = dev_get_drvdata(dev);
+	struct drm_device *drm_dev = data;
+
+	dvo_debugfs_exit(dvo, drm_dev->primary);
 
 	drm_bridge_remove(dvo->bridge);
 }
diff --git a/drivers/gpu/drm/sti/sti_gdp.c b/drivers/gpu/drm/sti/sti_gdp.c
index f9a1d92c9d95..67f606a41c3f 100644
--- a/drivers/gpu/drm/sti/sti_gdp.c
+++ b/drivers/gpu/drm/sti/sti_gdp.c
@@ -6,9 +6,7 @@
  * License terms:  GNU General Public License (GPL), version 2
  */
 
-#include <linux/clk.h>
-#include <linux/dma-mapping.h>
-
+#include <drm/drm_atomic.h>
 #include <drm/drm_fb_cma_helper.h>
 #include <drm/drm_gem_cma_helper.h>
 
@@ -32,10 +30,23 @@
 #define GDP_ABGR8888    (GDP_ARGB8888 | BIGNOTLITTLE | ALPHASWITCH)
 #define GDP_ARGB1555    0x06
 #define GDP_ARGB4444    0x07
-#define GDP_CLUT8       0x0B
-#define GDP_YCBR888     0x10
-#define GDP_YCBR422R    0x12
-#define GDP_AYCBR8888   0x15
+
+#define GDP2STR(fmt) { GDP_ ## fmt, #fmt }
+
+static struct gdp_format_to_str {
+	int format;
+	char name[20];
+} gdp_format_to_str[] = {
+		GDP2STR(RGB565),
+		GDP2STR(RGB888),
+		GDP2STR(RGB888_32),
+		GDP2STR(XBGR8888),
+		GDP2STR(ARGB8565),
+		GDP2STR(ARGB8888),
+		GDP2STR(ABGR8888),
+		GDP2STR(ARGB1555),
+		GDP2STR(ARGB4444)
+		};
 
 #define GAM_GDP_CTL_OFFSET      0x00
 #define GAM_GDP_AGC_OFFSET      0x04
@@ -97,6 +108,7 @@ struct sti_gdp_node_list {
  * @vtg_field_nb:       callback for VTG FIELD (top or bottom) notification
  * @is_curr_top:        true if the current node processed is the top field
  * @node_list:          array of node list
+ * @vtg:                registered vtg
  */
 struct sti_gdp {
 	struct sti_plane plane;
@@ -108,6 +120,7 @@ struct sti_gdp {
 	struct notifier_block vtg_field_nb;
 	bool is_curr_top;
 	struct sti_gdp_node_list node_list[GDP_NODE_NB_BANK];
+	struct sti_vtg *vtg;
 };
 
 #define to_sti_gdp(x) container_of(x, struct sti_gdp, plane)
@@ -121,12 +134,224 @@ static const uint32_t gdp_supported_formats[] = {
 	DRM_FORMAT_ARGB1555,
 	DRM_FORMAT_RGB565,
 	DRM_FORMAT_RGB888,
-	DRM_FORMAT_AYUV,
-	DRM_FORMAT_YUV444,
-	DRM_FORMAT_VYUY,
-	DRM_FORMAT_C8,
 };
 
+#define DBGFS_DUMP(reg) seq_printf(s, "\n  %-25s 0x%08X", #reg, \
+				   readl(gdp->regs + reg ## _OFFSET))
+
+static void gdp_dbg_ctl(struct seq_file *s, int val)
+{
+	int i;
+
+	seq_puts(s, "\tColor:");
+	for (i = 0; i < ARRAY_SIZE(gdp_format_to_str); i++) {
+		if (gdp_format_to_str[i].format == (val & 0x1F)) {
+			seq_printf(s, gdp_format_to_str[i].name);
+			break;
+		}
+	}
+	if (i == ARRAY_SIZE(gdp_format_to_str))
+		seq_puts(s, "<UNKNOWN>");
+
+	seq_printf(s, "\tWaitNextVsync:%d", val & WAIT_NEXT_VSYNC ? 1 : 0);
+}
+
+static void gdp_dbg_vpo(struct seq_file *s, int val)
+{
+	seq_printf(s, "\txdo:%4d\tydo:%4d", val & 0xFFFF, (val >> 16) & 0xFFFF);
+}
+
+static void gdp_dbg_vps(struct seq_file *s, int val)
+{
+	seq_printf(s, "\txds:%4d\tyds:%4d", val & 0xFFFF, (val >> 16) & 0xFFFF);
+}
+
+static void gdp_dbg_size(struct seq_file *s, int val)
+{
+	seq_printf(s, "\t%d x %d", val & 0xFFFF, (val >> 16) & 0xFFFF);
+}
+
+static void gdp_dbg_nvn(struct seq_file *s, struct sti_gdp *gdp, int val)
+{
+	void *base = NULL;
+	unsigned int i;
+
+	for (i = 0; i < GDP_NODE_NB_BANK; i++) {
+		if (gdp->node_list[i].top_field_paddr == val) {
+			base = gdp->node_list[i].top_field;
+			break;
+		}
+		if (gdp->node_list[i].btm_field_paddr == val) {
+			base = gdp->node_list[i].btm_field;
+			break;
+		}
+	}
+
+	if (base)
+		seq_printf(s, "\tVirt @: %p", base);
+}
+
+static void gdp_dbg_ppt(struct seq_file *s, int val)
+{
+	if (val & GAM_GDP_PPT_IGNORE)
+		seq_puts(s, "\tNot displayed on mixer!");
+}
+
+static void gdp_dbg_mst(struct seq_file *s, int val)
+{
+	if (val & 1)
+		seq_puts(s, "\tBUFFER UNDERFLOW!");
+}
+
+static int gdp_dbg_show(struct seq_file *s, void *data)
+{
+	struct drm_info_node *node = s->private;
+	struct sti_gdp *gdp = (struct sti_gdp *)node->info_ent->data;
+	struct drm_device *dev = node->minor->dev;
+	struct drm_plane *drm_plane = &gdp->plane.drm_plane;
+	struct drm_crtc *crtc = drm_plane->crtc;
+	int ret;
+
+	ret = mutex_lock_interruptible(&dev->struct_mutex);
+	if (ret)
+		return ret;
+
+	seq_printf(s, "%s: (vaddr = 0x%p)",
+		   sti_plane_to_str(&gdp->plane), gdp->regs);
+
+	DBGFS_DUMP(GAM_GDP_CTL);
+	gdp_dbg_ctl(s, readl(gdp->regs + GAM_GDP_CTL_OFFSET));
+	DBGFS_DUMP(GAM_GDP_AGC);
+	DBGFS_DUMP(GAM_GDP_VPO);
+	gdp_dbg_vpo(s, readl(gdp->regs + GAM_GDP_VPO_OFFSET));
+	DBGFS_DUMP(GAM_GDP_VPS);
+	gdp_dbg_vps(s, readl(gdp->regs + GAM_GDP_VPS_OFFSET));
+	DBGFS_DUMP(GAM_GDP_PML);
+	DBGFS_DUMP(GAM_GDP_PMP);
+	DBGFS_DUMP(GAM_GDP_SIZE);
+	gdp_dbg_size(s, readl(gdp->regs + GAM_GDP_SIZE_OFFSET));
+	DBGFS_DUMP(GAM_GDP_NVN);
+	gdp_dbg_nvn(s, gdp, readl(gdp->regs + GAM_GDP_NVN_OFFSET));
+	DBGFS_DUMP(GAM_GDP_KEY1);
+	DBGFS_DUMP(GAM_GDP_KEY2);
+	DBGFS_DUMP(GAM_GDP_PPT);
+	gdp_dbg_ppt(s, readl(gdp->regs + GAM_GDP_PPT_OFFSET));
+	DBGFS_DUMP(GAM_GDP_CML);
+	DBGFS_DUMP(GAM_GDP_MST);
+	gdp_dbg_mst(s, readl(gdp->regs + GAM_GDP_MST_OFFSET));
+
+	seq_puts(s, "\n\n");
+	if (!crtc)
+		seq_puts(s, "  Not connected to any DRM CRTC\n");
+	else
+		seq_printf(s, "  Connected to DRM CRTC #%d (%s)\n",
+			   crtc->base.id, sti_mixer_to_str(to_sti_mixer(crtc)));
+
+	mutex_unlock(&dev->struct_mutex);
+	return 0;
+}
+
+static void gdp_node_dump_node(struct seq_file *s, struct sti_gdp_node *node)
+{
+	seq_printf(s, "\t@:0x%p", node);
+	seq_printf(s, "\n\tCTL  0x%08X", node->gam_gdp_ctl);
+	gdp_dbg_ctl(s, node->gam_gdp_ctl);
+	seq_printf(s, "\n\tAGC  0x%08X", node->gam_gdp_agc);
+	seq_printf(s, "\n\tVPO  0x%08X", node->gam_gdp_vpo);
+	gdp_dbg_vpo(s, node->gam_gdp_vpo);
+	seq_printf(s, "\n\tVPS  0x%08X", node->gam_gdp_vps);
+	gdp_dbg_vps(s, node->gam_gdp_vps);
+	seq_printf(s, "\n\tPML  0x%08X", node->gam_gdp_pml);
+	seq_printf(s, "\n\tPMP  0x%08X", node->gam_gdp_pmp);
+	seq_printf(s, "\n\tSIZE 0x%08X", node->gam_gdp_size);
+	gdp_dbg_size(s, node->gam_gdp_size);
+	seq_printf(s, "\n\tNVN  0x%08X", node->gam_gdp_nvn);
+	seq_printf(s, "\n\tKEY1 0x%08X", node->gam_gdp_key1);
+	seq_printf(s, "\n\tKEY2 0x%08X", node->gam_gdp_key2);
+	seq_printf(s, "\n\tPPT  0x%08X", node->gam_gdp_ppt);
+	gdp_dbg_ppt(s, node->gam_gdp_ppt);
+	seq_printf(s, "\n\tCML  0x%08X", node->gam_gdp_cml);
+	seq_puts(s, "\n");
+}
+
+static int gdp_node_dbg_show(struct seq_file *s, void *arg)
+{
+	struct drm_info_node *node = s->private;
+	struct sti_gdp *gdp = (struct sti_gdp *)node->info_ent->data;
+	struct drm_device *dev = node->minor->dev;
+	unsigned int b;
+	int ret;
+
+	ret = mutex_lock_interruptible(&dev->struct_mutex);
+	if (ret)
+		return ret;
+
+	for (b = 0; b < GDP_NODE_NB_BANK; b++) {
+		seq_printf(s, "\n%s[%d].top", sti_plane_to_str(&gdp->plane), b);
+		gdp_node_dump_node(s, gdp->node_list[b].top_field);
+		seq_printf(s, "\n%s[%d].btm", sti_plane_to_str(&gdp->plane), b);
+		gdp_node_dump_node(s, gdp->node_list[b].btm_field);
+	}
+
+	mutex_unlock(&dev->struct_mutex);
+	return 0;
+}
+
+static struct drm_info_list gdp0_debugfs_files[] = {
+	{ "gdp0", gdp_dbg_show, 0, NULL },
+	{ "gdp0_node", gdp_node_dbg_show, 0, NULL },
+};
+
+static struct drm_info_list gdp1_debugfs_files[] = {
+	{ "gdp1", gdp_dbg_show, 0, NULL },
+	{ "gdp1_node", gdp_node_dbg_show, 0, NULL },
+};
+
+static struct drm_info_list gdp2_debugfs_files[] = {
+	{ "gdp2", gdp_dbg_show, 0, NULL },
+	{ "gdp2_node", gdp_node_dbg_show, 0, NULL },
+};
+
+static struct drm_info_list gdp3_debugfs_files[] = {
+	{ "gdp3", gdp_dbg_show, 0, NULL },
+	{ "gdp3_node", gdp_node_dbg_show, 0, NULL },
+};
+
+static int gdp_debugfs_init(struct sti_gdp *gdp, struct drm_minor *minor)
+{
+	unsigned int i;
+	struct drm_info_list *gdp_debugfs_files;
+	int nb_files;
+
+	switch (gdp->plane.desc) {
+	case STI_GDP_0:
+		gdp_debugfs_files = gdp0_debugfs_files;
+		nb_files = ARRAY_SIZE(gdp0_debugfs_files);
+		break;
+	case STI_GDP_1:
+		gdp_debugfs_files = gdp1_debugfs_files;
+		nb_files = ARRAY_SIZE(gdp1_debugfs_files);
+		break;
+	case STI_GDP_2:
+		gdp_debugfs_files = gdp2_debugfs_files;
+		nb_files = ARRAY_SIZE(gdp2_debugfs_files);
+		break;
+	case STI_GDP_3:
+		gdp_debugfs_files = gdp3_debugfs_files;
+		nb_files = ARRAY_SIZE(gdp3_debugfs_files);
+		break;
+	default:
+		return -EINVAL;
+	}
+
+	for (i = 0; i < nb_files; i++)
+		gdp_debugfs_files[i].data = gdp;
+
+	return drm_debugfs_create_files(gdp_debugfs_files,
+					nb_files,
+					minor->debugfs_root, minor);
+}
+
 static int sti_gdp_fourcc2format(int fourcc)
 {
 	switch (fourcc) {
@@ -146,14 +371,6 @@ static int sti_gdp_fourcc2format(int fourcc)
 		return GDP_RGB565;
 	case DRM_FORMAT_RGB888:
 		return GDP_RGB888;
-	case DRM_FORMAT_AYUV:
-		return GDP_AYCBR8888;
-	case DRM_FORMAT_YUV444:
-		return GDP_YCBR888;
-	case DRM_FORMAT_VYUY:
-		return GDP_YCBR422R;
-	case DRM_FORMAT_C8:
-		return GDP_CLUT8;
 	}
 	return -1;
 }
@@ -163,7 +380,6 @@ static int sti_gdp_get_alpharange(int format)
 	switch (format) {
 	case GDP_ARGB8565:
 	case GDP_ARGB8888:
-	case GDP_AYCBR8888:
 	case GDP_ABGR8888:
 		return GAM_GDP_ALPHARANGE_255;
 	}
@@ -240,9 +456,6 @@ end:
  */
 static void sti_gdp_disable(struct sti_gdp *gdp)
 {
-	struct drm_plane *drm_plane = &gdp->plane.drm_plane;
-	struct sti_mixer *mixer = to_sti_mixer(drm_plane->crtc);
-	struct sti_compositor *compo = dev_get_drvdata(gdp->dev);
 	unsigned int i;
 
 	DRM_DEBUG_DRIVER("%s\n", sti_plane_to_str(&gdp->plane));
@@ -253,8 +466,7 @@ static void sti_gdp_disable(struct sti_gdp *gdp)
 		gdp->node_list[i].btm_field->gam_gdp_ppt |= GAM_GDP_PPT_IGNORE;
 	}
 
-	if (sti_vtg_unregister_client(mixer->id == STI_MIXER_MAIN ?
-			compo->vtg_main : compo->vtg_aux, &gdp->vtg_field_nb))
+	if (sti_vtg_unregister_client(gdp->vtg, &gdp->vtg_field_nb))
 		DRM_DEBUG_DRIVER("Warning: cannot unregister VTG notifier\n");
 
 	if (gdp->clk_pix)
@@ -380,20 +592,140 @@ static void sti_gdp_init(struct sti_gdp *gdp)
 	}
 }
 
-static void sti_gdp_atomic_update(struct drm_plane *drm_plane,
-				  struct drm_plane_state *oldstate)
+/**
+ * sti_gdp_get_dst
+ * @dev: device
+ * @dst: requested destination size
+ * @src: source size
+ *
+ * Return the cropped / clamped destination size
+ *
+ * RETURNS:
+ * cropped / clamped destination size
+ */
+static int sti_gdp_get_dst(struct device *dev, int dst, int src)
+{
+	if (dst == src)
+		return dst;
+
+	if (dst < src) {
+		dev_dbg(dev, "WARNING: GDP scale not supported, will crop\n");
+		return dst;
+	}
+
+	dev_dbg(dev, "WARNING: GDP scale not supported, will clamp\n");
+	return src;
+}
+
+static int sti_gdp_atomic_check(struct drm_plane *drm_plane,
+				struct drm_plane_state *state)
 {
-	struct drm_plane_state *state = drm_plane->state;
 	struct sti_plane *plane = to_sti_plane(drm_plane);
 	struct sti_gdp *gdp = to_sti_gdp(plane);
 	struct drm_crtc *crtc = state->crtc;
 	struct sti_compositor *compo = dev_get_drvdata(gdp->dev);
 	struct drm_framebuffer *fb =  state->fb;
 	bool first_prepare = plane->status == STI_PLANE_DISABLED ? true : false;
+	struct drm_crtc_state *crtc_state;
 	struct sti_mixer *mixer;
 	struct drm_display_mode *mode;
 	int dst_x, dst_y, dst_w, dst_h;
 	int src_x, src_y, src_w, src_h;
+	int format;
+
+	/* no need for further checks if the plane is being disabled */
+	if (!crtc || !fb)
+		return 0;
+
+	mixer = to_sti_mixer(crtc);
+	crtc_state = drm_atomic_get_crtc_state(state->state, crtc);
+	mode = &crtc_state->mode;
+	dst_x = state->crtc_x;
+	dst_y = state->crtc_y;
+	dst_w = clamp_val(state->crtc_w, 0, mode->crtc_hdisplay - dst_x);
+	dst_h = clamp_val(state->crtc_h, 0, mode->crtc_vdisplay - dst_y);
+	/* src_x are in 16.16 format */
+	src_x = state->src_x >> 16;
+	src_y = state->src_y >> 16;
+	src_w = clamp_val(state->src_w >> 16, 0, GAM_GDP_SIZE_MAX);
+	src_h = clamp_val(state->src_h >> 16, 0, GAM_GDP_SIZE_MAX);
+
+	format = sti_gdp_fourcc2format(fb->pixel_format);
+	if (format == -1) {
+		DRM_ERROR("Format not supported by GDP %.4s\n",
+			  (char *)&fb->pixel_format);
+		return -EINVAL;
+	}
+
+	if (!drm_fb_cma_get_gem_obj(fb, 0)) {
+		DRM_ERROR("Can't get CMA GEM object for fb\n");
+		return -EINVAL;
+	}
+
+	if (first_prepare) {
+		/* Register gdp callback */
+		gdp->vtg = mixer->id == STI_MIXER_MAIN ?
+					compo->vtg_main : compo->vtg_aux;
+		if (sti_vtg_register_client(gdp->vtg,
+					    &gdp->vtg_field_nb, crtc)) {
+			DRM_ERROR("Cannot register VTG notifier\n");
+			return -EINVAL;
+		}
+
+		/* Set and enable gdp clock */
+		if (gdp->clk_pix) {
+			struct clk *clkp;
+			int rate = mode->clock * 1000;
+			int res;
+
+			/*
+			 * According to the mixer used, the gdp pixel clock
+			 * should have a different parent clock.
+			 */
+			if (mixer->id == STI_MIXER_MAIN)
+				clkp = gdp->clk_main_parent;
+			else
+				clkp = gdp->clk_aux_parent;
+
+			if (clkp)
+				clk_set_parent(gdp->clk_pix, clkp);
+
+			res = clk_set_rate(gdp->clk_pix, rate);
+			if (res < 0) {
+				DRM_ERROR("Cannot set rate (%dHz) for gdp\n",
+					  rate);
+				return -EINVAL;
+			}
+
+			if (clk_prepare_enable(gdp->clk_pix)) {
+				DRM_ERROR("Failed to prepare/enable gdp\n");
+				return -EINVAL;
+			}
+		}
+	}
+
+	DRM_DEBUG_KMS("CRTC:%d (%s) drm plane:%d (%s)\n",
+		      crtc->base.id, sti_mixer_to_str(mixer),
+		      drm_plane->base.id, sti_plane_to_str(plane));
+	DRM_DEBUG_KMS("%s dst=(%dx%d)@(%d,%d) - src=(%dx%d)@(%d,%d)\n",
+		      sti_plane_to_str(plane),
+		      dst_w, dst_h, dst_x, dst_y,
+		      src_w, src_h, src_x, src_y);
+
+	return 0;
+}
+
+static void sti_gdp_atomic_update(struct drm_plane *drm_plane,
+				  struct drm_plane_state *oldstate)
+{
+	struct drm_plane_state *state = drm_plane->state;
+	struct sti_plane *plane = to_sti_plane(drm_plane);
+	struct sti_gdp *gdp = to_sti_gdp(plane);
+	struct drm_crtc *crtc = state->crtc;
+	struct drm_framebuffer *fb =  state->fb;
+	struct drm_display_mode *mode;
+	int dst_x, dst_y, dst_w, dst_h;
+	int src_x, src_y, src_w, src_h;
 	struct drm_gem_cma_object *cma_obj;
 	struct sti_gdp_node_list *list;
 	struct sti_gdp_node_list *curr_list;
@@ -403,13 +735,10 @@ static void sti_gdp_atomic_update(struct drm_plane *drm_plane,
 	int format;
 	unsigned int depth, bpp;
 	u32 ydo, xdo, yds, xds;
-	int res;
 
-	/* Manage the case where crtc is null (disabled) */
-	if (!crtc)
+	if (!crtc || !fb)
 		return;
 
-	mixer = to_sti_mixer(crtc);
 	mode = &crtc->mode;
 	dst_x = state->crtc_x;
 	dst_y = state->crtc_y;
@@ -418,16 +747,8 @@ static void sti_gdp_atomic_update(struct drm_plane *drm_plane,
 	/* src_x are in 16.16 format */
 	src_x = state->src_x >> 16;
 	src_y = state->src_y >> 16;
-	src_w = state->src_w >> 16;
-	src_h = state->src_h >> 16;
-
-	DRM_DEBUG_KMS("CRTC:%d (%s) drm plane:%d (%s)\n",
-		      crtc->base.id, sti_mixer_to_str(mixer),
-		      drm_plane->base.id, sti_plane_to_str(plane));
-	DRM_DEBUG_KMS("%s dst=(%dx%d)@(%d,%d) - src=(%dx%d)@(%d,%d)\n",
-		      sti_plane_to_str(plane),
-		      dst_w, dst_h, dst_x, dst_y,
-		      src_w, src_h, src_x, src_y);
+	src_w = clamp_val(state->src_w >> 16, 0, GAM_GDP_SIZE_MAX);
+	src_h = clamp_val(state->src_h >> 16, 0, GAM_GDP_SIZE_MAX);
 
 	list = sti_gdp_get_free_nodes(gdp);
 	top_field = list->top_field;
@@ -440,20 +761,11 @@ static void sti_gdp_atomic_update(struct drm_plane *drm_plane,
 	top_field->gam_gdp_agc = GAM_GDP_AGC_FULL_RANGE;
 	top_field->gam_gdp_ctl = WAIT_NEXT_VSYNC;
 	format = sti_gdp_fourcc2format(fb->pixel_format);
-	if (format == -1) {
-		DRM_ERROR("Format not supported by GDP %.4s\n",
-			  (char *)&fb->pixel_format);
-		return;
-	}
 	top_field->gam_gdp_ctl |= format;
 	top_field->gam_gdp_ctl |= sti_gdp_get_alpharange(format);
 	top_field->gam_gdp_ppt &= ~GAM_GDP_PPT_IGNORE;
 
 	cma_obj = drm_fb_cma_get_gem_obj(fb, 0);
-	if (!cma_obj) {
-		DRM_ERROR("Can't get CMA GEM object for fb\n");
-		return;
-	}
 
 	DRM_DEBUG_DRIVER("drm FB:%d format:%.4s phys@:0x%lx\n", fb->base.id,
 			 (char *)&fb->pixel_format,
@@ -465,12 +777,9 @@ static void sti_gdp_atomic_update(struct drm_plane *drm_plane,
 	top_field->gam_gdp_pml += src_x * (bpp >> 3);
 	top_field->gam_gdp_pml += src_y * fb->pitches[0];
 
-	/* input parameters */
-	top_field->gam_gdp_pmp = fb->pitches[0];
-	top_field->gam_gdp_size = clamp_val(src_h, 0, GAM_GDP_SIZE_MAX) << 16 |
-				  clamp_val(src_w, 0, GAM_GDP_SIZE_MAX);
-
-	/* output parameters */
+	/* output parameters (clamped / cropped) */
+	dst_w = sti_gdp_get_dst(gdp->dev, dst_w, src_w);
+	dst_h = sti_gdp_get_dst(gdp->dev, dst_h, src_h);
 	ydo = sti_vtg_get_line_number(*mode, dst_y);
 	yds = sti_vtg_get_line_number(*mode, dst_y + dst_h - 1);
 	xdo = sti_vtg_get_pixel_number(*mode, dst_x);
@@ -478,6 +787,11 @@ static void sti_gdp_atomic_update(struct drm_plane *drm_plane,
 	top_field->gam_gdp_vpo = (ydo << 16) | xdo;
 	top_field->gam_gdp_vps = (yds << 16) | xds;
 
+	/* input parameters */
+	src_w = dst_w;
+	top_field->gam_gdp_pmp = fb->pitches[0];
+	top_field->gam_gdp_size = src_h << 16 | src_w;
+
 	/* Same content and chained together */
 	memcpy(btm_field, top_field, sizeof(*btm_field));
 	top_field->gam_gdp_nvn = list->btm_field_paddr;
@@ -488,44 +802,6 @@ static void sti_gdp_atomic_update(struct drm_plane *drm_plane,
 		btm_field->gam_gdp_pml = top_field->gam_gdp_pml +
 					 fb->pitches[0];
 
-	if (first_prepare) {
-		/* Register gdp callback */
-		if (sti_vtg_register_client(mixer->id == STI_MIXER_MAIN ?
-				compo->vtg_main : compo->vtg_aux,
-				&gdp->vtg_field_nb, crtc)) {
-			DRM_ERROR("Cannot register VTG notifier\n");
-			return;
-		}
-
-		/* Set and enable gdp clock */
-		if (gdp->clk_pix) {
-			struct clk *clkp;
-			int rate = mode->clock * 1000;
-
-			/* According to the mixer used, the gdp pixel clock
-			 * should have a different parent clock. */
-			if (mixer->id == STI_MIXER_MAIN)
-				clkp = gdp->clk_main_parent;
-			else
-				clkp = gdp->clk_aux_parent;
-
-			if (clkp)
-				clk_set_parent(gdp->clk_pix, clkp);
-
-			res = clk_set_rate(gdp->clk_pix, rate);
-			if (res < 0) {
-				DRM_ERROR("Cannot set rate (%dHz) for gdp\n",
-					  rate);
-				return;
-			}
-
-			if (clk_prepare_enable(gdp->clk_pix)) {
-				DRM_ERROR("Failed to prepare/enable gdp\n");
-				return;
-			}
-		}
-	}
-
 	/* Update the NVN field of the 'right' field of the current GDP node
 	 * (being used by the HW) with the address of the updated ('free') top
 	 * field GDP node.
@@ -574,6 +850,8 @@ static void sti_gdp_atomic_update(struct drm_plane *drm_plane,
 	}
 
 end:
+	sti_plane_update_fps(plane, true, false);
+
 	plane->status = STI_PLANE_UPDATED;
 }
 
@@ -581,7 +859,6 @@ static void sti_gdp_atomic_disable(struct drm_plane *drm_plane,
 				   struct drm_plane_state *oldstate)
 {
 	struct sti_plane *plane = to_sti_plane(drm_plane);
-	struct sti_mixer *mixer = to_sti_mixer(drm_plane->crtc);
 
 	if (!drm_plane->crtc) {
 		DRM_DEBUG_DRIVER("drm plane:%d not enabled\n",
@@ -590,13 +867,15 @@ static void sti_gdp_atomic_disable(struct drm_plane *drm_plane,
 	}
 
 	DRM_DEBUG_DRIVER("CRTC:%d (%s) drm plane:%d (%s)\n",
-			 drm_plane->crtc->base.id, sti_mixer_to_str(mixer),
+			 drm_plane->crtc->base.id,
+			 sti_mixer_to_str(to_sti_mixer(drm_plane->crtc)),
 			 drm_plane->base.id, sti_plane_to_str(plane));
 
 	plane->status = STI_PLANE_DISABLING;
 }
 
 static const struct drm_plane_helper_funcs sti_gdp_helpers_funcs = {
+	.atomic_check = sti_gdp_atomic_check,
 	.atomic_update = sti_gdp_atomic_update,
 	.atomic_disable = sti_gdp_atomic_disable,
 };
@@ -640,6 +919,9 @@ struct drm_plane *sti_gdp_create(struct drm_device *drm_dev,
 
 	sti_plane_init_property(&gdp->plane, type);
 
+	if (gdp_debugfs_init(gdp, drm_dev->primary))
+		DRM_ERROR("GDP debugfs setup failed\n");
+
 	return &gdp->plane.drm_plane;
 
 err:
diff --git a/drivers/gpu/drm/sti/sti_hda.c b/drivers/gpu/drm/sti/sti_hda.c
index 49cce833f2c8..ec0d017eaf1a 100644
--- a/drivers/gpu/drm/sti/sti_hda.c
+++ b/drivers/gpu/drm/sti/sti_hda.c
@@ -326,6 +326,103 @@ static void hda_enable_hd_dacs(struct sti_hda *hda, bool enable)
 	}
 }
 
+#define DBGFS_DUMP(reg) seq_printf(s, "\n  %-25s 0x%08X", #reg, \
+				   readl(hda->regs + reg))
+
+static void hda_dbg_cfg(struct seq_file *s, int val)
+{
+	seq_puts(s, "\tAWG ");
+	seq_puts(s, val & CFG_AWG_ASYNC_EN ? "enabled" : "disabled");
+}
+
+static void hda_dbg_awg_microcode(struct seq_file *s, void __iomem *reg)
+{
+	unsigned int i;
+
+	seq_puts(s, "\n\n");
+	seq_puts(s, "  HDA AWG microcode:");
+	for (i = 0; i < AWG_MAX_INST; i++) {
+		if (i % 8 == 0)
+			seq_printf(s, "\n  %04X:", i);
+		seq_printf(s, " %04X", readl(reg + i * 4));
+	}
+}
+
+static void hda_dbg_video_dacs_ctrl(struct seq_file *s, void __iomem *reg)
+{
+	u32 val = readl(reg);
+	u32 mask;
+
+	switch ((u32)reg & VIDEO_DACS_CONTROL_MASK) {
+	case VIDEO_DACS_CONTROL_SYSCFG2535:
+		mask = DAC_CFG_HD_OFF_MASK;
+		break;
+	case VIDEO_DACS_CONTROL_SYSCFG5072:
+		mask = DAC_CFG_HD_HZUVW_OFF_MASK;
+		break;
+	default:
+		DRM_DEBUG_DRIVER("Warning: DACS ctrl register not supported!");
+		return;
+	}
+
+	seq_puts(s, "\n");
+	seq_printf(s, "\n  %-25s 0x%08X", "VIDEO_DACS_CONTROL", val);
+	seq_puts(s, "\tHD DACs ");
+	seq_puts(s, val & mask ? "disabled" : "enabled");
+}
+
+static int hda_dbg_show(struct seq_file *s, void *data)
+{
+	struct drm_info_node *node = s->private;
+	struct sti_hda *hda = (struct sti_hda *)node->info_ent->data;
+	struct drm_device *dev = node->minor->dev;
+	int ret;
+
+	ret = mutex_lock_interruptible(&dev->struct_mutex);
+	if (ret)
+		return ret;
+
+	seq_printf(s, "HD Analog: (vaddr = 0x%p)", hda->regs);
+	DBGFS_DUMP(HDA_ANA_CFG);
+	hda_dbg_cfg(s, readl(hda->regs + HDA_ANA_CFG));
+	DBGFS_DUMP(HDA_ANA_SCALE_CTRL_Y);
+	DBGFS_DUMP(HDA_ANA_SCALE_CTRL_CB);
+	DBGFS_DUMP(HDA_ANA_SCALE_CTRL_CR);
+	DBGFS_DUMP(HDA_ANA_ANC_CTRL);
+	DBGFS_DUMP(HDA_ANA_SRC_Y_CFG);
+	DBGFS_DUMP(HDA_ANA_SRC_C_CFG);
+	hda_dbg_awg_microcode(s, hda->regs + HDA_SYNC_AWGI);
+	if (hda->video_dacs_ctrl)
+		hda_dbg_video_dacs_ctrl(s, hda->video_dacs_ctrl);
+	seq_puts(s, "\n");
+
+	mutex_unlock(&dev->struct_mutex);
+	return 0;
+}
+
+static struct drm_info_list hda_debugfs_files[] = {
+	{ "hda", hda_dbg_show, 0, NULL },
+};
+
+static void hda_debugfs_exit(struct sti_hda *hda, struct drm_minor *minor)
+{
+	drm_debugfs_remove_files(hda_debugfs_files,
+				 ARRAY_SIZE(hda_debugfs_files),
+				 minor);
+}
+
+static int hda_debugfs_init(struct sti_hda *hda, struct drm_minor *minor)
+{
+	unsigned int i;
+
+	for (i = 0; i < ARRAY_SIZE(hda_debugfs_files); i++)
+		hda_debugfs_files[i].data = hda;
+
+	return drm_debugfs_create_files(hda_debugfs_files,
+					ARRAY_SIZE(hda_debugfs_files),
+					minor->debugfs_root, minor);
+}
+
 /**
  * Configure AWG, writing instructions
  *
@@ -685,6 +782,12 @@ static int sti_hda_bind(struct device *dev, struct device *master, void *data)
 		goto err_sysfs;
 	}
 
+	/* force to disable hd dacs at startup */
+	hda_enable_hd_dacs(hda, false);
+
+	if (hda_debugfs_init(hda, drm_dev->primary))
+		DRM_ERROR("HDA debugfs setup failed\n");
+
 	return 0;
 
 err_sysfs:
@@ -697,7 +800,10 @@ err_connector:
 static void sti_hda_unbind(struct device *dev,
 		struct device *master, void *data)
 {
-	/* do nothing */
+	struct sti_hda *hda = dev_get_drvdata(dev);
+	struct drm_device *drm_dev = data;
+
+	hda_debugfs_exit(hda, drm_dev->primary);
 }
 
 static const struct component_ops sti_hda_ops = {
diff --git a/drivers/gpu/drm/sti/sti_hdmi.c b/drivers/gpu/drm/sti/sti_hdmi.c
index cd501563c0cc..6ef0715bd5b9 100644
--- a/drivers/gpu/drm/sti/sti_hdmi.c
+++ b/drivers/gpu/drm/sti/sti_hdmi.c
@@ -6,6 +6,7 @@
 
 #include <linux/clk.h>
 #include <linux/component.h>
+#include <linux/debugfs.h>
 #include <linux/hdmi.h>
 #include <linux/module.h>
 #include <linux/of_gpio.h>
@@ -51,9 +52,18 @@
 #define HDMI_SW_DI_2_PKT_WORD4          0x0614
 #define HDMI_SW_DI_2_PKT_WORD5          0x0618
 #define HDMI_SW_DI_2_PKT_WORD6          0x061C
+#define HDMI_SW_DI_3_HEAD_WORD          0x0620
+#define HDMI_SW_DI_3_PKT_WORD0          0x0624
+#define HDMI_SW_DI_3_PKT_WORD1          0x0628
+#define HDMI_SW_DI_3_PKT_WORD2          0x062C
+#define HDMI_SW_DI_3_PKT_WORD3          0x0630
+#define HDMI_SW_DI_3_PKT_WORD4          0x0634
+#define HDMI_SW_DI_3_PKT_WORD5          0x0638
+#define HDMI_SW_DI_3_PKT_WORD6          0x063C
 
 #define HDMI_IFRAME_SLOT_AVI            1
 #define HDMI_IFRAME_SLOT_AUDIO          2
+#define HDMI_IFRAME_SLOT_VENDOR         3
 
 #define  XCAT(prefix, x, suffix)        prefix ## x ## suffix
 #define  HDMI_SW_DI_N_HEAD_WORD(x)      XCAT(HDMI_SW_DI_, x, _HEAD_WORD)
@@ -65,6 +75,8 @@
 #define  HDMI_SW_DI_N_PKT_WORD5(x)      XCAT(HDMI_SW_DI_, x, _PKT_WORD5)
 #define  HDMI_SW_DI_N_PKT_WORD6(x)      XCAT(HDMI_SW_DI_, x, _PKT_WORD6)
 
+#define HDMI_SW_DI_MAX_WORD             7
+
 #define HDMI_IFRAME_DISABLED            0x0
 #define HDMI_IFRAME_SINGLE_SHOT         0x1
 #define HDMI_IFRAME_FIELD               0x2
@@ -117,6 +129,8 @@ struct sti_hdmi_connector {
 	struct drm_connector drm_connector;
 	struct drm_encoder *encoder;
 	struct sti_hdmi *hdmi;
+	struct drm_property *colorspace_property;
+	struct drm_property *hdmi_mode_property;
 };
 
 #define to_sti_hdmi_connector(x) \
@@ -217,8 +231,10 @@ static void hdmi_config(struct sti_hdmi *hdmi)
 	/* Clear overrun and underrun fifo */
 	conf = HDMI_CFG_FIFO_OVERRUN_CLR | HDMI_CFG_FIFO_UNDERRUN_CLR;
 
-	/* Enable HDMI mode not DVI */
-	conf |= HDMI_CFG_HDMI_NOT_DVI | HDMI_CFG_ESS_NOT_OESS;
+	/* Select encryption type and the framing mode */
+	conf |= HDMI_CFG_ESS_NOT_OESS;
+	if (hdmi->hdmi_mode == HDMI_MODE_HDMI)
+		conf |= HDMI_CFG_HDMI_NOT_DVI;
 
 	/* Enable sink term detection */
 	conf |= HDMI_CFG_SINK_TERM_DET_EN;
@@ -241,6 +257,47 @@ static void hdmi_config(struct sti_hdmi *hdmi)
 	hdmi_write(hdmi, conf, HDMI_CFG);
 }
 
+/*
+ * Helper to reset info frame
+ *
+ * @hdmi: pointer on the hdmi internal structure
+ * @slot: infoframe to reset
+ */
+static void hdmi_infoframe_reset(struct sti_hdmi *hdmi,
+				 u32 slot)
+{
+	u32 val, i;
+	u32 head_offset, pack_offset;
+
+	switch (slot) {
+	case HDMI_IFRAME_SLOT_AVI:
+		head_offset = HDMI_SW_DI_N_HEAD_WORD(HDMI_IFRAME_SLOT_AVI);
+		pack_offset = HDMI_SW_DI_N_PKT_WORD0(HDMI_IFRAME_SLOT_AVI);
+		break;
+	case HDMI_IFRAME_SLOT_AUDIO:
+		head_offset = HDMI_SW_DI_N_HEAD_WORD(HDMI_IFRAME_SLOT_AUDIO);
+		pack_offset = HDMI_SW_DI_N_PKT_WORD0(HDMI_IFRAME_SLOT_AUDIO);
+		break;
+	case HDMI_IFRAME_SLOT_VENDOR:
+		head_offset = HDMI_SW_DI_N_HEAD_WORD(HDMI_IFRAME_SLOT_VENDOR);
+		pack_offset = HDMI_SW_DI_N_PKT_WORD0(HDMI_IFRAME_SLOT_VENDOR);
+		break;
+	default:
+		DRM_ERROR("unsupported infoframe slot: %#x\n", slot);
+		return;
+	}
+
+	/* Disable transmission for the selected slot */
+	val = hdmi_read(hdmi, HDMI_SW_DI_CFG);
+	val &= ~HDMI_IFRAME_CFG_DI_N(HDMI_IFRAME_MASK, slot);
+	hdmi_write(hdmi, val, HDMI_SW_DI_CFG);
+
+	/* Reset info frame registers */
+	hdmi_write(hdmi, 0x0, head_offset);
+	for (i = 0; i < HDMI_SW_DI_MAX_WORD; i += sizeof(u32))
+		hdmi_write(hdmi, 0x0, pack_offset + i);
+}
+
 /**
  * Helper to concatenate infoframe in 32 bits word
  *
@@ -266,12 +323,13 @@ static inline unsigned int hdmi_infoframe_subpack(const u8 *ptr, size_t size)
  * @data: infoframe to write
  * @size: size to write
  */
-static void hdmi_infoframe_write_infopack(struct sti_hdmi *hdmi, const u8 *data)
+static void hdmi_infoframe_write_infopack(struct sti_hdmi *hdmi,
+					  const u8 *data,
+					  size_t size)
 {
 	const u8 *ptr = data;
 	u32 val, slot, mode, i;
 	u32 head_offset, pack_offset;
-	size_t size;
 
 	switch (*ptr) {
 	case HDMI_INFOFRAME_TYPE_AVI:
@@ -279,17 +337,19 @@ static void hdmi_infoframe_write_infopack(struct sti_hdmi *hdmi, const u8 *data)
 		mode = HDMI_IFRAME_FIELD;
 		head_offset = HDMI_SW_DI_N_HEAD_WORD(HDMI_IFRAME_SLOT_AVI);
 		pack_offset = HDMI_SW_DI_N_PKT_WORD0(HDMI_IFRAME_SLOT_AVI);
-		size = HDMI_AVI_INFOFRAME_SIZE;
 		break;
-
 	case HDMI_INFOFRAME_TYPE_AUDIO:
 		slot = HDMI_IFRAME_SLOT_AUDIO;
 		mode = HDMI_IFRAME_FRAME;
 		head_offset = HDMI_SW_DI_N_HEAD_WORD(HDMI_IFRAME_SLOT_AUDIO);
 		pack_offset = HDMI_SW_DI_N_PKT_WORD0(HDMI_IFRAME_SLOT_AUDIO);
-		size = HDMI_AUDIO_INFOFRAME_SIZE;
 		break;
-
+	case HDMI_INFOFRAME_TYPE_VENDOR:
+		slot = HDMI_IFRAME_SLOT_VENDOR;
+		mode = HDMI_IFRAME_FRAME;
+		head_offset = HDMI_SW_DI_N_HEAD_WORD(HDMI_IFRAME_SLOT_VENDOR);
+		pack_offset = HDMI_SW_DI_N_PKT_WORD0(HDMI_IFRAME_SLOT_VENDOR);
+		break;
 	default:
 		DRM_ERROR("unsupported infoframe type: %#x\n", *ptr);
 		return;
@@ -308,8 +368,9 @@ static void hdmi_infoframe_write_infopack(struct sti_hdmi *hdmi, const u8 *data)
 	/*
 	 * Each subpack contains 4 bytes
 	 * The First Bytes of the first subpacket must contain the checksum
-	 * Packet size in increase by one.
+	 * Packet size is increase by one.
 	 */
+	size = size - HDMI_INFOFRAME_HEADER_SIZE + 1;
 	for (i = 0; i < size; i += sizeof(u32)) {
 		size_t num;
 
@@ -321,7 +382,7 @@ static void hdmi_infoframe_write_infopack(struct sti_hdmi *hdmi, const u8 *data)
 
 	/* Enable transmission slot for updated infoframe */
 	val = hdmi_read(hdmi, HDMI_SW_DI_CFG);
-	val |= HDMI_IFRAME_CFG_DI_N(HDMI_IFRAME_FIELD, slot);
+	val |= HDMI_IFRAME_CFG_DI_N(mode, slot);
 	hdmi_write(hdmi, val, HDMI_SW_DI_CFG);
 }
 
@@ -352,7 +413,7 @@ static int hdmi_avi_infoframe_config(struct sti_hdmi *hdmi)
 	}
 
 	/* fixed infoframe configuration not linked to the mode */
-	infoframe.colorspace = HDMI_COLORSPACE_RGB;
+	infoframe.colorspace = hdmi->colorspace;
 	infoframe.quantization_range = HDMI_QUANTIZATION_RANGE_DEFAULT;
 	infoframe.colorimetry = HDMI_COLORIMETRY_NONE;
 
@@ -362,7 +423,7 @@ static int hdmi_avi_infoframe_config(struct sti_hdmi *hdmi)
 		return ret;
 	}
 
-	hdmi_infoframe_write_infopack(hdmi, buffer);
+	hdmi_infoframe_write_infopack(hdmi, buffer, ret);
 
 	return 0;
 }
@@ -398,7 +459,49 @@ static int hdmi_audio_infoframe_config(struct sti_hdmi *hdmi)
 		return ret;
 	}
 
-	hdmi_infoframe_write_infopack(hdmi, buffer);
+	hdmi_infoframe_write_infopack(hdmi, buffer, ret);
+
+	return 0;
+}
+
+/*
+ * Prepare and configure the VS infoframe
+ *
+ * Vendor Specific infoframe are transmitted once per frame and
+ * contains vendor specific information.
+ *
+ * @hdmi: pointer on the hdmi internal structure
+ *
+ * Return negative value if error occurs
+ */
+#define HDMI_VENDOR_INFOFRAME_MAX_SIZE 6
+static int hdmi_vendor_infoframe_config(struct sti_hdmi *hdmi)
+{
+	struct drm_display_mode *mode = &hdmi->mode;
+	struct hdmi_vendor_infoframe infoframe;
+	u8 buffer[HDMI_INFOFRAME_HEADER_SIZE + HDMI_VENDOR_INFOFRAME_MAX_SIZE];
+	int ret;
+
+	DRM_DEBUG_DRIVER("\n");
+
+	ret = drm_hdmi_vendor_infoframe_from_display_mode(&infoframe, mode);
+	if (ret < 0) {
+		/*
+		 * Going into that statement does not means vendor infoframe
+		 * fails. It just informed us that vendor infoframe is not
+		 * needed for the selected mode. Only  4k or stereoscopic 3D
+		 * mode requires vendor infoframe. So just simply return 0.
+		 */
+		return 0;
+	}
+
+	ret = hdmi_vendor_infoframe_pack(&infoframe, buffer, sizeof(buffer));
+	if (ret < 0) {
+		DRM_ERROR("failed to pack VS infoframe: %d\n", ret);
+		return ret;
+	}
+
+	hdmi_infoframe_write_infopack(hdmi, buffer, ret);
 
 	return 0;
 }
@@ -448,6 +551,172 @@ static void hdmi_swreset(struct sti_hdmi *hdmi)
 	clk_disable_unprepare(hdmi->clk_audio);
 }
 
+#define DBGFS_PRINT_STR(str1, str2) seq_printf(s, "%-24s %s\n", str1, str2)
+#define DBGFS_PRINT_INT(str1, int2) seq_printf(s, "%-24s %d\n", str1, int2)
+#define DBGFS_DUMP(str, reg) seq_printf(s, "%s  %-25s 0x%08X", str, #reg, \
+					hdmi_read(hdmi, reg))
+#define DBGFS_DUMP_DI(reg, slot) DBGFS_DUMP("\n", reg(slot))
+
+static void hdmi_dbg_cfg(struct seq_file *s, int val)
+{
+	int tmp;
+
+	seq_puts(s, "\t");
+	tmp = val & HDMI_CFG_HDMI_NOT_DVI;
+	DBGFS_PRINT_STR("mode:", tmp ? "HDMI" : "DVI");
+	seq_puts(s, "\t\t\t\t\t");
+	tmp = val & HDMI_CFG_HDCP_EN;
+	DBGFS_PRINT_STR("HDCP:", tmp ? "enable" : "disable");
+	seq_puts(s, "\t\t\t\t\t");
+	tmp = val & HDMI_CFG_ESS_NOT_OESS;
+	DBGFS_PRINT_STR("HDCP mode:", tmp ? "ESS enable" : "OESS enable");
+	seq_puts(s, "\t\t\t\t\t");
+	tmp = val & HDMI_CFG_SINK_TERM_DET_EN;
+	DBGFS_PRINT_STR("Sink term detection:", tmp ? "enable" : "disable");
+	seq_puts(s, "\t\t\t\t\t");
+	tmp = val & HDMI_CFG_H_SYNC_POL_NEG;
+	DBGFS_PRINT_STR("Hsync polarity:", tmp ? "inverted" : "normal");
+	seq_puts(s, "\t\t\t\t\t");
+	tmp = val & HDMI_CFG_V_SYNC_POL_NEG;
+	DBGFS_PRINT_STR("Vsync polarity:", tmp ? "inverted" : "normal");
+	seq_puts(s, "\t\t\t\t\t");
+	tmp = val & HDMI_CFG_422_EN;
+	DBGFS_PRINT_STR("YUV422 format:", tmp ? "enable" : "disable");
+}
+
+static void hdmi_dbg_sta(struct seq_file *s, int val)
+{
+	int tmp;
+
+	seq_puts(s, "\t");
+	tmp = (val & HDMI_STA_DLL_LCK);
+	DBGFS_PRINT_STR("pll:", tmp ? "locked" : "not locked");
+	seq_puts(s, "\t\t\t\t\t");
+	tmp = (val & HDMI_STA_HOT_PLUG);
+	DBGFS_PRINT_STR("hdmi cable:", tmp ? "connected" : "not connected");
+}
+
+static void hdmi_dbg_sw_di_cfg(struct seq_file *s, int val)
+{
+	int tmp;
+	char *const en_di[] = {"no transmission",
+			       "single transmission",
+			       "once every field",
+			       "once every frame"};
+
+	seq_puts(s, "\t");
+	tmp = (val & HDMI_IFRAME_CFG_DI_N(HDMI_IFRAME_MASK, 1));
+	DBGFS_PRINT_STR("Data island 1:", en_di[tmp]);
+	seq_puts(s, "\t\t\t\t\t");
+	tmp = (val & HDMI_IFRAME_CFG_DI_N(HDMI_IFRAME_MASK, 2)) >> 4;
+	DBGFS_PRINT_STR("Data island 2:", en_di[tmp]);
+	seq_puts(s, "\t\t\t\t\t");
+	tmp = (val & HDMI_IFRAME_CFG_DI_N(HDMI_IFRAME_MASK, 3)) >> 8;
+	DBGFS_PRINT_STR("Data island 3:", en_di[tmp]);
+	seq_puts(s, "\t\t\t\t\t");
+	tmp = (val & HDMI_IFRAME_CFG_DI_N(HDMI_IFRAME_MASK, 4)) >> 12;
+	DBGFS_PRINT_STR("Data island 4:", en_di[tmp]);
+	seq_puts(s, "\t\t\t\t\t");
+	tmp = (val & HDMI_IFRAME_CFG_DI_N(HDMI_IFRAME_MASK, 5)) >> 16;
+	DBGFS_PRINT_STR("Data island 5:", en_di[tmp]);
+	seq_puts(s, "\t\t\t\t\t");
+	tmp = (val & HDMI_IFRAME_CFG_DI_N(HDMI_IFRAME_MASK, 6)) >> 20;
+	DBGFS_PRINT_STR("Data island 6:", en_di[tmp]);
+}
+
+static int hdmi_dbg_show(struct seq_file *s, void *data)
+{
+	struct drm_info_node *node = s->private;
+	struct sti_hdmi *hdmi = (struct sti_hdmi *)node->info_ent->data;
+	struct drm_device *dev = node->minor->dev;
+	int ret;
+
+	ret = mutex_lock_interruptible(&dev->struct_mutex);
+	if (ret)
+		return ret;
+
+	seq_printf(s, "HDMI: (vaddr = 0x%p)", hdmi->regs);
+	DBGFS_DUMP("\n", HDMI_CFG);
+	hdmi_dbg_cfg(s, hdmi_read(hdmi, HDMI_CFG));
+	DBGFS_DUMP("", HDMI_INT_EN);
+	DBGFS_DUMP("\n", HDMI_STA);
+	hdmi_dbg_sta(s, hdmi_read(hdmi, HDMI_STA));
+	DBGFS_DUMP("", HDMI_ACTIVE_VID_XMIN);
+	seq_puts(s, "\t");
+	DBGFS_PRINT_INT("Xmin:", hdmi_read(hdmi, HDMI_ACTIVE_VID_XMIN));
+	DBGFS_DUMP("", HDMI_ACTIVE_VID_XMAX);
+	seq_puts(s, "\t");
+	DBGFS_PRINT_INT("Xmax:", hdmi_read(hdmi, HDMI_ACTIVE_VID_XMAX));
+	DBGFS_DUMP("", HDMI_ACTIVE_VID_YMIN);
+	seq_puts(s, "\t");
+	DBGFS_PRINT_INT("Ymin:", hdmi_read(hdmi, HDMI_ACTIVE_VID_YMIN));
+	DBGFS_DUMP("", HDMI_ACTIVE_VID_YMAX);
+	seq_puts(s, "\t");
+	DBGFS_PRINT_INT("Ymax:", hdmi_read(hdmi, HDMI_ACTIVE_VID_YMAX));
+	DBGFS_DUMP("", HDMI_SW_DI_CFG);
+	hdmi_dbg_sw_di_cfg(s, hdmi_read(hdmi, HDMI_SW_DI_CFG));
+
+	seq_printf(s, "\n AVI Infoframe (Data Island slot N=%d):",
+		   HDMI_IFRAME_SLOT_AVI);
+	DBGFS_DUMP_DI(HDMI_SW_DI_N_HEAD_WORD, HDMI_IFRAME_SLOT_AVI);
+	DBGFS_DUMP_DI(HDMI_SW_DI_N_PKT_WORD0, HDMI_IFRAME_SLOT_AVI);
+	DBGFS_DUMP_DI(HDMI_SW_DI_N_PKT_WORD1, HDMI_IFRAME_SLOT_AVI);
+	DBGFS_DUMP_DI(HDMI_SW_DI_N_PKT_WORD2, HDMI_IFRAME_SLOT_AVI);
+	DBGFS_DUMP_DI(HDMI_SW_DI_N_PKT_WORD3, HDMI_IFRAME_SLOT_AVI);
+	DBGFS_DUMP_DI(HDMI_SW_DI_N_PKT_WORD4, HDMI_IFRAME_SLOT_AVI);
+	DBGFS_DUMP_DI(HDMI_SW_DI_N_PKT_WORD5, HDMI_IFRAME_SLOT_AVI);
+	DBGFS_DUMP_DI(HDMI_SW_DI_N_PKT_WORD6, HDMI_IFRAME_SLOT_AVI);
+	seq_puts(s, "\n");
+	seq_printf(s, "\n AUDIO Infoframe (Data Island slot N=%d):",
+		   HDMI_IFRAME_SLOT_AUDIO);
+	DBGFS_DUMP_DI(HDMI_SW_DI_N_HEAD_WORD, HDMI_IFRAME_SLOT_AUDIO);
+	DBGFS_DUMP_DI(HDMI_SW_DI_N_PKT_WORD0, HDMI_IFRAME_SLOT_AUDIO);
+	DBGFS_DUMP_DI(HDMI_SW_DI_N_PKT_WORD1, HDMI_IFRAME_SLOT_AUDIO);
+	DBGFS_DUMP_DI(HDMI_SW_DI_N_PKT_WORD2, HDMI_IFRAME_SLOT_AUDIO);
+	DBGFS_DUMP_DI(HDMI_SW_DI_N_PKT_WORD3, HDMI_IFRAME_SLOT_AUDIO);
+	DBGFS_DUMP_DI(HDMI_SW_DI_N_PKT_WORD4, HDMI_IFRAME_SLOT_AUDIO);
+	DBGFS_DUMP_DI(HDMI_SW_DI_N_PKT_WORD5, HDMI_IFRAME_SLOT_AUDIO);
+	DBGFS_DUMP_DI(HDMI_SW_DI_N_PKT_WORD6, HDMI_IFRAME_SLOT_AUDIO);
+	seq_puts(s, "\n");
+	seq_printf(s, "\n VENDOR SPECIFIC Infoframe (Data Island slot N=%d):",
+		   HDMI_IFRAME_SLOT_VENDOR);
+	DBGFS_DUMP_DI(HDMI_SW_DI_N_HEAD_WORD, HDMI_IFRAME_SLOT_VENDOR);
+	DBGFS_DUMP_DI(HDMI_SW_DI_N_PKT_WORD0, HDMI_IFRAME_SLOT_VENDOR);
+	DBGFS_DUMP_DI(HDMI_SW_DI_N_PKT_WORD1, HDMI_IFRAME_SLOT_VENDOR);
+	DBGFS_DUMP_DI(HDMI_SW_DI_N_PKT_WORD2, HDMI_IFRAME_SLOT_VENDOR);
+	DBGFS_DUMP_DI(HDMI_SW_DI_N_PKT_WORD3, HDMI_IFRAME_SLOT_VENDOR);
+	DBGFS_DUMP_DI(HDMI_SW_DI_N_PKT_WORD4, HDMI_IFRAME_SLOT_VENDOR);
+	DBGFS_DUMP_DI(HDMI_SW_DI_N_PKT_WORD5, HDMI_IFRAME_SLOT_VENDOR);
+	DBGFS_DUMP_DI(HDMI_SW_DI_N_PKT_WORD6, HDMI_IFRAME_SLOT_VENDOR);
+	seq_puts(s, "\n");
+
+	mutex_unlock(&dev->struct_mutex);
+	return 0;
+}
+
+static struct drm_info_list hdmi_debugfs_files[] = {
+	{ "hdmi", hdmi_dbg_show, 0, NULL },
+};
+
+static void hdmi_debugfs_exit(struct sti_hdmi *hdmi, struct drm_minor *minor)
+{
+	drm_debugfs_remove_files(hdmi_debugfs_files,
+				 ARRAY_SIZE(hdmi_debugfs_files),
+				 minor);
+}
+
+static int hdmi_debugfs_init(struct sti_hdmi *hdmi, struct drm_minor *minor)
+{
+	unsigned int i;
+
+	for (i = 0; i < ARRAY_SIZE(hdmi_debugfs_files); i++)
+		hdmi_debugfs_files[i].data = hdmi;
+
+	return drm_debugfs_create_files(hdmi_debugfs_files,
+					ARRAY_SIZE(hdmi_debugfs_files),
+					minor->debugfs_root, minor);
+}
+
 static void sti_hdmi_disable(struct drm_bridge *bridge)
 {
 	struct sti_hdmi *hdmi = bridge->driver_private;
@@ -468,6 +737,11 @@ static void sti_hdmi_disable(struct drm_bridge *bridge)
 	/* Stop the phy */
 	hdmi->phy_ops->stop(hdmi);
 
+	/* Reset info frame transmission */
+	hdmi_infoframe_reset(hdmi, HDMI_IFRAME_SLOT_AVI);
+	hdmi_infoframe_reset(hdmi, HDMI_IFRAME_SLOT_AUDIO);
+	hdmi_infoframe_reset(hdmi, HDMI_IFRAME_SLOT_VENDOR);
+
 	/* Set the default channel data to be a dark red */
 	hdmi_write(hdmi, 0x0000, HDMI_DFLT_CHL0_DAT);
 	hdmi_write(hdmi, 0x0000, HDMI_DFLT_CHL1_DAT);
@@ -523,6 +797,10 @@ static void sti_hdmi_pre_enable(struct drm_bridge *bridge)
 	if (hdmi_audio_infoframe_config(hdmi))
 		DRM_ERROR("Unable to configure AUDIO infoframe\n");
 
+	/* Program VS infoframe */
+	if (hdmi_vendor_infoframe_config(hdmi))
+		DRM_ERROR("Unable to configure VS infoframe\n");
+
 	/* Sw reset */
 	hdmi_swreset(hdmi);
 }
@@ -664,12 +942,97 @@ static void sti_hdmi_connector_destroy(struct drm_connector *connector)
 	kfree(hdmi_connector);
 }
 
+static void sti_hdmi_connector_init_property(struct drm_device *drm_dev,
+					     struct drm_connector *connector)
+{
+	struct sti_hdmi_connector *hdmi_connector
+		= to_sti_hdmi_connector(connector);
+	struct sti_hdmi *hdmi = hdmi_connector->hdmi;
+	struct drm_property *prop;
+
+	/* colorspace property */
+	hdmi->colorspace = DEFAULT_COLORSPACE_MODE;
+	prop = drm_property_create_enum(drm_dev, 0, "colorspace",
+					colorspace_mode_names,
+					ARRAY_SIZE(colorspace_mode_names));
+	if (!prop) {
+		DRM_ERROR("fails to create colorspace property\n");
+		return;
+	}
+	hdmi_connector->colorspace_property = prop;
+	drm_object_attach_property(&connector->base, prop, hdmi->colorspace);
+
+	/* hdmi_mode property */
+	hdmi->hdmi_mode = DEFAULT_HDMI_MODE;
+	prop = drm_property_create_enum(drm_dev, 0, "hdmi_mode",
+					hdmi_mode_names,
+					ARRAY_SIZE(hdmi_mode_names));
+	if (!prop) {
+		DRM_ERROR("fails to create colorspace property\n");
+		return;
+	}
+	hdmi_connector->hdmi_mode_property = prop;
+	drm_object_attach_property(&connector->base, prop, hdmi->hdmi_mode);
+
+}
+
+static int
+sti_hdmi_connector_set_property(struct drm_connector *connector,
+				struct drm_connector_state *state,
+				struct drm_property *property,
+				uint64_t val)
+{
+	struct sti_hdmi_connector *hdmi_connector
+		= to_sti_hdmi_connector(connector);
+	struct sti_hdmi *hdmi = hdmi_connector->hdmi;
+
+	if (property == hdmi_connector->colorspace_property) {
+		hdmi->colorspace = val;
+		return 0;
+	}
+
+	if (property == hdmi_connector->hdmi_mode_property) {
+		hdmi->hdmi_mode = val;
+		return 0;
+	}
+
+	DRM_ERROR("failed to set hdmi connector property\n");
+	return -EINVAL;
+}
+
+static int
+sti_hdmi_connector_get_property(struct drm_connector *connector,
+				const struct drm_connector_state *state,
+				struct drm_property *property,
+				uint64_t *val)
+{
+	struct sti_hdmi_connector *hdmi_connector
+		= to_sti_hdmi_connector(connector);
+	struct sti_hdmi *hdmi = hdmi_connector->hdmi;
+
+	if (property == hdmi_connector->colorspace_property) {
+		*val = hdmi->colorspace;
+		return 0;
+	}
+
+	if (property == hdmi_connector->hdmi_mode_property) {
+		*val = hdmi->hdmi_mode;
+		return 0;
+	}
+
+	DRM_ERROR("failed to get hdmi connector property\n");
+	return -EINVAL;
+}
+
 static const struct drm_connector_funcs sti_hdmi_connector_funcs = {
 	.dpms = drm_atomic_helper_connector_dpms,
 	.fill_modes = drm_helper_probe_single_connector_modes,
 	.detect = sti_hdmi_connector_detect,
 	.destroy = sti_hdmi_connector_destroy,
 	.reset = drm_atomic_helper_connector_reset,
+	.set_property = drm_atomic_helper_connector_set_property,
+	.atomic_set_property = sti_hdmi_connector_set_property,
+	.atomic_get_property = sti_hdmi_connector_get_property,
 	.atomic_duplicate_state = drm_atomic_helper_connector_duplicate_state,
 	.atomic_destroy_state = drm_atomic_helper_connector_destroy_state,
 };
@@ -729,6 +1092,9 @@ static int sti_hdmi_bind(struct device *dev, struct device *master, void *data)
 	drm_connector_helper_add(drm_connector,
 			&sti_hdmi_connector_helper_funcs);
 
+	/* initialise property */
+	sti_hdmi_connector_init_property(drm_dev, drm_connector);
+
 	err = drm_connector_register(drm_connector);
 	if (err)
 		goto err_connector;
@@ -742,6 +1108,9 @@ static int sti_hdmi_bind(struct device *dev, struct device *master, void *data)
 	/* Enable default interrupts */
 	hdmi_write(hdmi, HDMI_DEFAULT_INT, HDMI_INT_EN);
 
+	if (hdmi_debugfs_init(hdmi, drm_dev->primary))
+		DRM_ERROR("HDMI debugfs setup failed\n");
+
 	return 0;
 
 err_sysfs:
@@ -755,7 +1124,10 @@ err_connector:
 static void sti_hdmi_unbind(struct device *dev,
 		struct device *master, void *data)
 {
-	/* do nothing */
+	struct sti_hdmi *hdmi = dev_get_drvdata(dev);
+	struct drm_device *drm_dev = data;
+
+	hdmi_debugfs_exit(hdmi, drm_dev->primary);
 }
 
 static const struct component_ops sti_hdmi_ops = {
diff --git a/drivers/gpu/drm/sti/sti_hdmi.h b/drivers/gpu/drm/sti/sti_hdmi.h
index 3d22390e1f3b..ef3a94583bbd 100644
--- a/drivers/gpu/drm/sti/sti_hdmi.h
+++ b/drivers/gpu/drm/sti/sti_hdmi.h
@@ -7,15 +7,14 @@
 #ifndef _STI_HDMI_H_
 #define _STI_HDMI_H_
 
+#include <linux/hdmi.h>
 #include <linux/platform_device.h>
 
 #include <drm/drmP.h>
 
 #define HDMI_STA           0x0010
 #define HDMI_STA_DLL_LCK   BIT(5)
-
-#define HDMI_STA_HOT_PLUG_SHIFT 4
-#define HDMI_STA_HOT_PLUG	(1 << HDMI_STA_HOT_PLUG_SHIFT)
+#define HDMI_STA_HOT_PLUG  BIT(4)
 
 struct sti_hdmi;
 
@@ -24,6 +23,27 @@ struct hdmi_phy_ops {
 	void (*stop)(struct sti_hdmi *hdmi);
 };
 
+/* values for the framing mode property */
+enum sti_hdmi_modes {
+	HDMI_MODE_HDMI,
+	HDMI_MODE_DVI,
+};
+
+static const struct drm_prop_enum_list hdmi_mode_names[] = {
+	{ HDMI_MODE_HDMI, "hdmi" },
+	{ HDMI_MODE_DVI, "dvi" },
+};
+
+#define DEFAULT_HDMI_MODE HDMI_MODE_HDMI
+
+static const struct drm_prop_enum_list colorspace_mode_names[] = {
+	{ HDMI_COLORSPACE_RGB, "rgb" },
+	{ HDMI_COLORSPACE_YUV422, "yuv422" },
+	{ HDMI_COLORSPACE_YUV444, "yuv444" },
+};
+
+#define DEFAULT_COLORSPACE_MODE HDMI_COLORSPACE_RGB
+
 /**
  * STI hdmi structure
  *
@@ -44,6 +64,9 @@ struct hdmi_phy_ops {
  * @wait_event: wait event
  * @event_received: wait event status
  * @reset: reset control of the hdmi phy
+ * @ddc_adapt: i2c ddc adapter
+ * @colorspace: current colorspace selected
+ * @hdmi_mode: select framing for HDMI or DVI
  */
 struct sti_hdmi {
 	struct device dev;
@@ -64,6 +87,8 @@ struct sti_hdmi {
 	bool event_received;
 	struct reset_control *reset;
 	struct i2c_adapter *ddc_adapt;
+	enum hdmi_colorspace colorspace;
+	enum sti_hdmi_modes hdmi_mode;
 };
 
 u32 hdmi_read(struct sti_hdmi *hdmi, int offset);
diff --git a/drivers/gpu/drm/sti/sti_hqvdp.c b/drivers/gpu/drm/sti/sti_hqvdp.c
index 43861b52261d..d7c1f427811d 100644
--- a/drivers/gpu/drm/sti/sti_hqvdp.c
+++ b/drivers/gpu/drm/sti/sti_hqvdp.c
@@ -4,14 +4,11 @@
  * License terms:  GNU General Public License (GPL), version 2
  */
 
-#include <linux/clk.h>
 #include <linux/component.h>
 #include <linux/firmware.h>
-#include <linux/module.h>
-#include <linux/platform_device.h>
 #include <linux/reset.h>
 
-#include <drm/drmP.h>
+#include <drm/drm_atomic.h>
 #include <drm/drm_fb_cma_helper.h>
 #include <drm/drm_gem_cma_helper.h>
 
@@ -329,8 +326,6 @@ struct sti_hqvdp_cmd {
  * @reset:             reset control
  * @vtg_nb:            notifier to handle VTG Vsync
  * @btm_field_pending: is there any bottom field (interlaced frame) to display
- * @curr_field_count:  number of field updates
- * @last_field_count:  number of field updates since last fps measure
  * @hqvdp_cmd:         buffer of commands
  * @hqvdp_cmd_paddr:   physical address of hqvdp_cmd
  * @vtg:               vtg for main data path
@@ -346,10 +341,8 @@ struct sti_hqvdp {
 	struct reset_control *reset;
 	struct notifier_block vtg_nb;
 	bool btm_field_pending;
-	unsigned int curr_field_count;
-	unsigned int last_field_count;
 	void *hqvdp_cmd;
-	dma_addr_t hqvdp_cmd_paddr;
+	u32 hqvdp_cmd_paddr;
 	struct sti_vtg *vtg;
 	bool xp70_initialized;
 };
@@ -372,8 +365,8 @@ static const uint32_t hqvdp_supported_formats[] = {
  */
 static int sti_hqvdp_get_free_cmd(struct sti_hqvdp *hqvdp)
 {
-	int curr_cmd, next_cmd;
-	dma_addr_t cmd = hqvdp->hqvdp_cmd_paddr;
+	u32 curr_cmd, next_cmd;
+	u32 cmd = hqvdp->hqvdp_cmd_paddr;
 	int i;
 
 	curr_cmd = readl(hqvdp->regs + HQVDP_MBX_CURRENT_CMD);
@@ -400,8 +393,8 @@ static int sti_hqvdp_get_free_cmd(struct sti_hqvdp *hqvdp)
  */
 static int sti_hqvdp_get_curr_cmd(struct sti_hqvdp *hqvdp)
 {
-	int curr_cmd;
-	dma_addr_t cmd = hqvdp->hqvdp_cmd_paddr;
+	u32 curr_cmd;
+	u32 cmd = hqvdp->hqvdp_cmd_paddr;
 	unsigned int i;
 
 	curr_cmd = readl(hqvdp->regs + HQVDP_MBX_CURRENT_CMD);
@@ -417,6 +410,246 @@ static int sti_hqvdp_get_curr_cmd(struct sti_hqvdp *hqvdp)
 }
 
 /**
+ * sti_hqvdp_get_next_cmd
+ * @hqvdp: hqvdp structure
+ *
+ * Look for the next hqvdp_cmd that will be used by the FW.
+ *
+ * RETURNS:
+ *  the offset of the next command that will be used.
+ * -1 in error cases
+ */
+static int sti_hqvdp_get_next_cmd(struct sti_hqvdp *hqvdp)
+{
+	int next_cmd;
+	dma_addr_t cmd = hqvdp->hqvdp_cmd_paddr;
+	unsigned int i;
+
+	next_cmd = readl(hqvdp->regs + HQVDP_MBX_NEXT_CMD);
+
+	for (i = 0; i < NB_VDP_CMD; i++) {
+		if (cmd == next_cmd)
+			return i * sizeof(struct sti_hqvdp_cmd);
+
+		cmd += sizeof(struct sti_hqvdp_cmd);
+	}
+
+	return -1;
+}
+
+#define DBGFS_DUMP(reg) seq_printf(s, "\n  %-25s 0x%08X", #reg, \
+				   readl(hqvdp->regs + reg))
+
+static const char *hqvdp_dbg_get_lut(u32 *coef)
+{
+	if (!memcmp(coef, coef_lut_a_legacy, 16))
+		return "LUT A";
+	if (!memcmp(coef, coef_lut_b, 16))
+		return "LUT B";
+	if (!memcmp(coef, coef_lut_c_y_legacy, 16))
+		return "LUT C Y";
+	if (!memcmp(coef, coef_lut_c_c_legacy, 16))
+		return "LUT C C";
+	if (!memcmp(coef, coef_lut_d_y_legacy, 16))
+		return "LUT D Y";
+	if (!memcmp(coef, coef_lut_d_c_legacy, 16))
+		return "LUT D C";
+	if (!memcmp(coef, coef_lut_e_y_legacy, 16))
+		return "LUT E Y";
+	if (!memcmp(coef, coef_lut_e_c_legacy, 16))
+		return "LUT E C";
+	if (!memcmp(coef, coef_lut_f_y_legacy, 16))
+		return "LUT F Y";
+	if (!memcmp(coef, coef_lut_f_c_legacy, 16))
+		return "LUT F C";
+	return "<UNKNOWN>";
+}
+
+static void hqvdp_dbg_dump_cmd(struct seq_file *s, struct sti_hqvdp_cmd *c)
+{
+	int src_w, src_h, dst_w, dst_h;
+
+	seq_puts(s, "\n\tTOP:");
+	seq_printf(s, "\n\t %-20s 0x%08X", "Config", c->top.config);
+	switch (c->top.config) {
+	case TOP_CONFIG_PROGRESSIVE:
+		seq_puts(s, "\tProgressive");
+		break;
+	case TOP_CONFIG_INTER_TOP:
+		seq_puts(s, "\tInterlaced, top field");
+		break;
+	case TOP_CONFIG_INTER_BTM:
+		seq_puts(s, "\tInterlaced, bottom field");
+		break;
+	default:
+		seq_puts(s, "\t<UNKNOWN>");
+		break;
+	}
+
+	seq_printf(s, "\n\t %-20s 0x%08X", "MemFormat", c->top.mem_format);
+	seq_printf(s, "\n\t %-20s 0x%08X", "CurrentY", c->top.current_luma);
+	seq_printf(s, "\n\t %-20s 0x%08X", "CurrentC", c->top.current_chroma);
+	seq_printf(s, "\n\t %-20s 0x%08X", "YSrcPitch", c->top.luma_src_pitch);
+	seq_printf(s, "\n\t %-20s 0x%08X", "CSrcPitch",
+		   c->top.chroma_src_pitch);
+	seq_printf(s, "\n\t %-20s 0x%08X", "InputFrameSize",
+		   c->top.input_frame_size);
+	seq_printf(s, "\t%dx%d",
+		   c->top.input_frame_size & 0x0000FFFF,
+		   c->top.input_frame_size >> 16);
+	seq_printf(s, "\n\t %-20s 0x%08X", "InputViewportSize",
+		   c->top.input_viewport_size);
+	src_w = c->top.input_viewport_size & 0x0000FFFF;
+	src_h = c->top.input_viewport_size >> 16;
+	seq_printf(s, "\t%dx%d", src_w, src_h);
+
+	seq_puts(s, "\n\tHVSRC:");
+	seq_printf(s, "\n\t %-20s 0x%08X", "OutputPictureSize",
+		   c->hvsrc.output_picture_size);
+	dst_w = c->hvsrc.output_picture_size & 0x0000FFFF;
+	dst_h = c->hvsrc.output_picture_size >> 16;
+	seq_printf(s, "\t%dx%d", dst_w, dst_h);
+	seq_printf(s, "\n\t %-20s 0x%08X", "ParamCtrl", c->hvsrc.param_ctrl);
+
+	seq_printf(s, "\n\t %-20s %s", "yh_coef",
+		   hqvdp_dbg_get_lut(c->hvsrc.yh_coef));
+	seq_printf(s, "\n\t %-20s %s", "ch_coef",
+		   hqvdp_dbg_get_lut(c->hvsrc.ch_coef));
+	seq_printf(s, "\n\t %-20s %s", "yv_coef",
+		   hqvdp_dbg_get_lut(c->hvsrc.yv_coef));
+	seq_printf(s, "\n\t %-20s %s", "cv_coef",
+		   hqvdp_dbg_get_lut(c->hvsrc.cv_coef));
+
+	seq_printf(s, "\n\t %-20s", "ScaleH");
+	if (dst_w > src_w)
+		seq_printf(s, " %d/1", dst_w / src_w);
+	else
+		seq_printf(s, " 1/%d", src_w / dst_w);
+
+	seq_printf(s, "\n\t %-20s", "tScaleV");
+	if (dst_h > src_h)
+		seq_printf(s, " %d/1", dst_h / src_h);
+	else
+		seq_printf(s, " 1/%d", src_h / dst_h);
+
+	seq_puts(s, "\n\tCSDI:");
+	seq_printf(s, "\n\t %-20s 0x%08X\t", "Config", c->csdi.config);
+	switch (c->csdi.config) {
+	case CSDI_CONFIG_PROG:
+		seq_puts(s, "Bypass");
+		break;
+	case CSDI_CONFIG_INTER_DIR:
+		seq_puts(s, "Deinterlace, directional");
+		break;
+	default:
+		seq_puts(s, "<UNKNOWN>");
+		break;
+	}
+
+	seq_printf(s, "\n\t %-20s 0x%08X", "Config2", c->csdi.config2);
+	seq_printf(s, "\n\t %-20s 0x%08X", "DcdiConfig", c->csdi.dcdi_config);
+}
+
+static int hqvdp_dbg_show(struct seq_file *s, void *data)
+{
+	struct drm_info_node *node = s->private;
+	struct sti_hqvdp *hqvdp = (struct sti_hqvdp *)node->info_ent->data;
+	struct drm_device *dev = node->minor->dev;
+	int cmd, cmd_offset, infoxp70;
+	void *virt;
+	int ret;
+
+	ret = mutex_lock_interruptible(&dev->struct_mutex);
+	if (ret)
+		return ret;
+
+	seq_printf(s, "%s: (vaddr = 0x%p)",
+		   sti_plane_to_str(&hqvdp->plane), hqvdp->regs);
+
+	DBGFS_DUMP(HQVDP_MBX_IRQ_TO_XP70);
+	DBGFS_DUMP(HQVDP_MBX_INFO_HOST);
+	DBGFS_DUMP(HQVDP_MBX_IRQ_TO_HOST);
+	DBGFS_DUMP(HQVDP_MBX_INFO_XP70);
+	infoxp70 = readl(hqvdp->regs + HQVDP_MBX_INFO_XP70);
+	seq_puts(s, "\tFirmware state: ");
+	if (infoxp70 & INFO_XP70_FW_READY)
+		seq_puts(s, "idle and ready");
+	else if (infoxp70 & INFO_XP70_FW_PROCESSING)
+		seq_puts(s, "processing a picture");
+	else if (infoxp70 & INFO_XP70_FW_INITQUEUES)
+		seq_puts(s, "programming queues");
+	else
+		seq_puts(s, "NOT READY");
+
+	DBGFS_DUMP(HQVDP_MBX_SW_RESET_CTRL);
+	DBGFS_DUMP(HQVDP_MBX_STARTUP_CTRL1);
+	if (readl(hqvdp->regs + HQVDP_MBX_STARTUP_CTRL1)
+					& STARTUP_CTRL1_RST_DONE)
+		seq_puts(s, "\tReset is done");
+	else
+		seq_puts(s, "\tReset is NOT done");
+	DBGFS_DUMP(HQVDP_MBX_STARTUP_CTRL2);
+	if (readl(hqvdp->regs + HQVDP_MBX_STARTUP_CTRL2)
+					& STARTUP_CTRL2_FETCH_EN)
+		seq_puts(s, "\tFetch is enabled");
+	else
+		seq_puts(s, "\tFetch is NOT enabled");
+	DBGFS_DUMP(HQVDP_MBX_GP_STATUS);
+	DBGFS_DUMP(HQVDP_MBX_NEXT_CMD);
+	DBGFS_DUMP(HQVDP_MBX_CURRENT_CMD);
+	DBGFS_DUMP(HQVDP_MBX_SOFT_VSYNC);
+	if (!(readl(hqvdp->regs + HQVDP_MBX_SOFT_VSYNC) & 3))
+		seq_puts(s, "\tHW Vsync");
+	else
+		seq_puts(s, "\tSW Vsync ?!?!");
+
+	/* Last command */
+	cmd = readl(hqvdp->regs + HQVDP_MBX_CURRENT_CMD);
+	cmd_offset = sti_hqvdp_get_curr_cmd(hqvdp);
+	if (cmd_offset == -1) {
+		seq_puts(s, "\n\n  Last command: unknown");
+	} else {
+		virt = hqvdp->hqvdp_cmd + cmd_offset;
+		seq_printf(s, "\n\n  Last command: address @ 0x%x (0x%p)",
+			   cmd, virt);
+		hqvdp_dbg_dump_cmd(s, (struct sti_hqvdp_cmd *)virt);
+	}
+
+	/* Next command */
+	cmd = readl(hqvdp->regs + HQVDP_MBX_NEXT_CMD);
+	cmd_offset = sti_hqvdp_get_next_cmd(hqvdp);
+	if (cmd_offset == -1) {
+		seq_puts(s, "\n\n  Next command: unknown");
+	} else {
+		virt = hqvdp->hqvdp_cmd + cmd_offset;
+		seq_printf(s, "\n\n  Next command address: @ 0x%x (0x%p)",
+			   cmd, virt);
+		hqvdp_dbg_dump_cmd(s, (struct sti_hqvdp_cmd *)virt);
+	}
+
+	seq_puts(s, "\n");
+
+	mutex_unlock(&dev->struct_mutex);
+	return 0;
+}
+
+static struct drm_info_list hqvdp_debugfs_files[] = {
+	{ "hqvdp", hqvdp_dbg_show, 0, NULL },
+};
+
+static int hqvdp_debugfs_init(struct sti_hqvdp *hqvdp, struct drm_minor *minor)
+{
+	unsigned int i;
+
+	for (i = 0; i < ARRAY_SIZE(hqvdp_debugfs_files); i++)
+		hqvdp_debugfs_files[i].data = hqvdp;
+
+	return drm_debugfs_create_files(hqvdp_debugfs_files,
+					ARRAY_SIZE(hqvdp_debugfs_files),
+					minor->debugfs_root, minor);
+}
+
+/**
  * sti_hqvdp_update_hvsrc
  * @orient: horizontal or vertical
  * @scale:  scaling/zoom factor
@@ -580,7 +813,7 @@ int sti_hqvdp_vtg_cb(struct notifier_block *nb, unsigned long evt, void *data)
 		btm_cmd_offset = sti_hqvdp_get_free_cmd(hqvdp);
 		top_cmd_offest = sti_hqvdp_get_curr_cmd(hqvdp);
 		if ((btm_cmd_offset == -1) || (top_cmd_offest == -1)) {
-			DRM_ERROR("Cannot get cmds, skip btm field\n");
+			DRM_DEBUG_DRIVER("Warning: no cmd, will skip field\n");
 			return -EBUSY;
 		}
 
@@ -599,11 +832,12 @@ int sti_hqvdp_vtg_cb(struct notifier_block *nb, unsigned long evt, void *data)
 		writel(hqvdp->hqvdp_cmd_paddr + btm_cmd_offset,
 				hqvdp->regs + HQVDP_MBX_NEXT_CMD);
 
-		hqvdp->curr_field_count++;
 		hqvdp->btm_field_pending = false;
 
 		dev_dbg(hqvdp->dev, "%s Posted command:0x%x\n",
 				__func__, hqvdp->hqvdp_cmd_paddr);
+
+		sti_plane_update_fps(&hqvdp->plane, false, true);
 	}
 
 	return 0;
@@ -612,19 +846,21 @@ int sti_hqvdp_vtg_cb(struct notifier_block *nb, unsigned long evt, void *data)
 static void sti_hqvdp_init(struct sti_hqvdp *hqvdp)
 {
 	int size;
+	dma_addr_t dma_addr;
 
 	hqvdp->vtg_nb.notifier_call = sti_hqvdp_vtg_cb;
 
 	/* Allocate memory for the VDP commands */
 	size = NB_VDP_CMD * sizeof(struct sti_hqvdp_cmd);
 	hqvdp->hqvdp_cmd = dma_alloc_writecombine(hqvdp->dev, size,
-					 &hqvdp->hqvdp_cmd_paddr,
+					 &dma_addr,
 					 GFP_KERNEL | GFP_DMA);
 	if (!hqvdp->hqvdp_cmd) {
 		DRM_ERROR("Failed to allocate memory for VDP cmd\n");
 		return;
 	}
 
+	hqvdp->hqvdp_cmd_paddr = (u32)dma_addr;
 	memset(hqvdp->hqvdp_cmd, 0, size);
 }
 
@@ -670,7 +906,7 @@ static void sti_hqvdp_start_xp70(struct sti_hqvdp *hqvdp)
 	DRM_DEBUG_DRIVER("\n");
 
 	if (hqvdp->xp70_initialized) {
-		DRM_INFO("HQVDP XP70 already initialized\n");
+		DRM_DEBUG_DRIVER("HQVDP XP70 already initialized\n");
 		return;
 	}
 
@@ -775,53 +1011,131 @@ out:
 	release_firmware(firmware);
 }
 
-static void sti_hqvdp_atomic_update(struct drm_plane *drm_plane,
-				    struct drm_plane_state *oldstate)
+static int sti_hqvdp_atomic_check(struct drm_plane *drm_plane,
+				  struct drm_plane_state *state)
 {
-	struct drm_plane_state *state = drm_plane->state;
 	struct sti_plane *plane = to_sti_plane(drm_plane);
 	struct sti_hqvdp *hqvdp = to_sti_hqvdp(plane);
 	struct drm_crtc *crtc = state->crtc;
-	struct sti_mixer *mixer = to_sti_mixer(crtc);
 	struct drm_framebuffer *fb = state->fb;
-	struct drm_display_mode *mode = &crtc->mode;
-	int dst_x = state->crtc_x;
-	int dst_y = state->crtc_y;
-	int dst_w = clamp_val(state->crtc_w, 0, mode->crtc_hdisplay - dst_x);
-	int dst_h = clamp_val(state->crtc_h, 0, mode->crtc_vdisplay - dst_y);
-	/* src_x are in 16.16 format */
-	int src_x = state->src_x >> 16;
-	int src_y = state->src_y >> 16;
-	int src_w = state->src_w >> 16;
-	int src_h = state->src_h >> 16;
 	bool first_prepare = plane->status == STI_PLANE_DISABLED ? true : false;
-	struct drm_gem_cma_object *cma_obj;
-	struct sti_hqvdp_cmd *cmd;
-	int scale_h, scale_v;
-	int cmd_offset;
+	struct drm_crtc_state *crtc_state;
+	struct drm_display_mode *mode;
+	int dst_x, dst_y, dst_w, dst_h;
+	int src_x, src_y, src_w, src_h;
+
+	/* no need for further checks if the plane is being disabled */
+	if (!crtc || !fb)
+		return 0;
+
+	crtc_state = drm_atomic_get_crtc_state(state->state, crtc);
+	mode = &crtc_state->mode;
+	dst_x = state->crtc_x;
+	dst_y = state->crtc_y;
+	dst_w = clamp_val(state->crtc_w, 0, mode->crtc_hdisplay - dst_x);
+	dst_h = clamp_val(state->crtc_h, 0, mode->crtc_vdisplay - dst_y);
+	/* src_x are in 16.16 format */
+	src_x = state->src_x >> 16;
+	src_y = state->src_y >> 16;
+	src_w = state->src_w >> 16;
+	src_h = state->src_h >> 16;
+
+	if (!sti_hqvdp_check_hw_scaling(hqvdp, mode,
+					src_w, src_h,
+					dst_w, dst_h)) {
+		DRM_ERROR("Scaling beyond HW capabilities\n");
+		return -EINVAL;
+	}
+
+	if (!drm_fb_cma_get_gem_obj(fb, 0)) {
+		DRM_ERROR("Can't get CMA GEM object for fb\n");
+		return -EINVAL;
+	}
+
+	/*
+	 * Input / output size
+	 * Align to upper even value
+	 */
+	dst_w = ALIGN(dst_w, 2);
+	dst_h = ALIGN(dst_h, 2);
+
+	if ((src_w > MAX_WIDTH) || (src_w < MIN_WIDTH) ||
+	    (src_h > MAX_HEIGHT) || (src_h < MIN_HEIGHT) ||
+	    (dst_w > MAX_WIDTH) || (dst_w < MIN_WIDTH) ||
+	    (dst_h > MAX_HEIGHT) || (dst_h < MIN_HEIGHT)) {
+		DRM_ERROR("Invalid in/out size %dx%d -> %dx%d\n",
+			  src_w, src_h,
+			  dst_w, dst_h);
+		return -EINVAL;
+	}
+
+	if (first_prepare) {
+		/* Start HQVDP XP70 coprocessor */
+		sti_hqvdp_start_xp70(hqvdp);
+
+		/* Prevent VTG shutdown */
+		if (clk_prepare_enable(hqvdp->clk_pix_main)) {
+			DRM_ERROR("Failed to prepare/enable pix main clk\n");
+			return -EINVAL;
+		}
+
+		/* Register VTG Vsync callback to handle bottom fields */
+		if (sti_vtg_register_client(hqvdp->vtg,
+					    &hqvdp->vtg_nb,
+					    crtc)) {
+			DRM_ERROR("Cannot register VTG notifier\n");
+			return -EINVAL;
+		}
+	}
 
 	DRM_DEBUG_KMS("CRTC:%d (%s) drm plane:%d (%s)\n",
-		      crtc->base.id, sti_mixer_to_str(mixer),
+		      crtc->base.id, sti_mixer_to_str(to_sti_mixer(crtc)),
 		      drm_plane->base.id, sti_plane_to_str(plane));
 	DRM_DEBUG_KMS("%s dst=(%dx%d)@(%d,%d) - src=(%dx%d)@(%d,%d)\n",
 		      sti_plane_to_str(plane),
 		      dst_w, dst_h, dst_x, dst_y,
 		      src_w, src_h, src_x, src_y);
 
+	return 0;
+}
+
+static void sti_hqvdp_atomic_update(struct drm_plane *drm_plane,
+				    struct drm_plane_state *oldstate)
+{
+	struct drm_plane_state *state = drm_plane->state;
+	struct sti_plane *plane = to_sti_plane(drm_plane);
+	struct sti_hqvdp *hqvdp = to_sti_hqvdp(plane);
+	struct drm_crtc *crtc = state->crtc;
+	struct drm_framebuffer *fb = state->fb;
+	struct drm_display_mode *mode;
+	int dst_x, dst_y, dst_w, dst_h;
+	int src_x, src_y, src_w, src_h;
+	struct drm_gem_cma_object *cma_obj;
+	struct sti_hqvdp_cmd *cmd;
+	int scale_h, scale_v;
+	int cmd_offset;
+
+	if (!crtc || !fb)
+		return;
+
+	mode = &crtc->mode;
+	dst_x = state->crtc_x;
+	dst_y = state->crtc_y;
+	dst_w = clamp_val(state->crtc_w, 0, mode->crtc_hdisplay - dst_x);
+	dst_h = clamp_val(state->crtc_h, 0, mode->crtc_vdisplay - dst_y);
+	/* src_x are in 16.16 format */
+	src_x = state->src_x >> 16;
+	src_y = state->src_y >> 16;
+	src_w = state->src_w >> 16;
+	src_h = state->src_h >> 16;
+
 	cmd_offset = sti_hqvdp_get_free_cmd(hqvdp);
 	if (cmd_offset == -1) {
-		DRM_ERROR("No available hqvdp_cmd now\n");
+		DRM_DEBUG_DRIVER("Warning: no cmd, will skip frame\n");
 		return;
 	}
 	cmd = hqvdp->hqvdp_cmd + cmd_offset;
 
-	if (!sti_hqvdp_check_hw_scaling(hqvdp, mode,
-					src_w, src_h,
-					dst_w, dst_h)) {
-		DRM_ERROR("Scaling beyond HW capabilities\n");
-		return;
-	}
-
 	/* Static parameters, defaulting to progressive mode */
 	cmd->top.config = TOP_CONFIG_PROGRESSIVE;
 	cmd->top.mem_format = TOP_MEM_FORMAT_DFLT;
@@ -836,10 +1150,6 @@ static void sti_hqvdp_atomic_update(struct drm_plane *drm_plane,
 	cmd->iqi.pxf_conf = IQI_PXF_CONF_DFLT;
 
 	cma_obj = drm_fb_cma_get_gem_obj(fb, 0);
-	if (!cma_obj) {
-		DRM_ERROR("Can't get CMA GEM object for fb\n");
-		return;
-	}
 
 	DRM_DEBUG_DRIVER("drm FB:%d format:%.4s phys@:0x%lx\n", fb->base.id,
 			 (char *)&fb->pixel_format,
@@ -860,16 +1170,6 @@ static void sti_hqvdp_atomic_update(struct drm_plane *drm_plane,
 	dst_w = ALIGN(dst_w, 2);
 	dst_h = ALIGN(dst_h, 2);
 
-	if ((src_w > MAX_WIDTH) || (src_w < MIN_WIDTH) ||
-	    (src_h > MAX_HEIGHT) || (src_h < MIN_HEIGHT) ||
-	    (dst_w > MAX_WIDTH) || (dst_w < MIN_WIDTH) ||
-	    (dst_h > MAX_HEIGHT) || (dst_h < MIN_HEIGHT)) {
-		DRM_ERROR("Invalid in/out size %dx%d -> %dx%d\n",
-			  src_w, src_h,
-			  dst_w, dst_h);
-		return;
-	}
-
 	cmd->top.input_viewport_size = src_h << 16 | src_w;
 	cmd->top.input_frame_size = src_h << 16 | src_w;
 	cmd->hvsrc.output_picture_size = dst_h << 16 | dst_w;
@@ -900,30 +1200,9 @@ static void sti_hqvdp_atomic_update(struct drm_plane *drm_plane,
 	scale_v = SCALE_FACTOR * dst_h / src_h;
 	sti_hqvdp_update_hvsrc(HVSRC_VERT, scale_v, &cmd->hvsrc);
 
-	if (first_prepare) {
-		/* Start HQVDP XP70 coprocessor */
-		sti_hqvdp_start_xp70(hqvdp);
-
-		/* Prevent VTG shutdown */
-		if (clk_prepare_enable(hqvdp->clk_pix_main)) {
-			DRM_ERROR("Failed to prepare/enable pix main clk\n");
-			return;
-		}
-
-		/* Register VTG Vsync callback to handle bottom fields */
-		if (sti_vtg_register_client(hqvdp->vtg,
-					    &hqvdp->vtg_nb,
-					    crtc)) {
-			DRM_ERROR("Cannot register VTG notifier\n");
-			return;
-		}
-	}
-
 	writel(hqvdp->hqvdp_cmd_paddr + cmd_offset,
 	       hqvdp->regs + HQVDP_MBX_NEXT_CMD);
 
-	hqvdp->curr_field_count++;
-
 	/* Interlaced : get ready to display the bottom field at next Vsync */
 	if (fb->flags & DRM_MODE_FB_INTERLACED)
 		hqvdp->btm_field_pending = true;
@@ -931,6 +1210,8 @@ static void sti_hqvdp_atomic_update(struct drm_plane *drm_plane,
 	dev_dbg(hqvdp->dev, "%s Posted command:0x%x\n",
 		__func__, hqvdp->hqvdp_cmd_paddr + cmd_offset);
 
+	sti_plane_update_fps(plane, true, true);
+
 	plane->status = STI_PLANE_UPDATED;
 }
 
@@ -938,7 +1219,6 @@ static void sti_hqvdp_atomic_disable(struct drm_plane *drm_plane,
 				     struct drm_plane_state *oldstate)
 {
 	struct sti_plane *plane = to_sti_plane(drm_plane);
-	struct sti_mixer *mixer = to_sti_mixer(drm_plane->crtc);
 
 	if (!drm_plane->crtc) {
 		DRM_DEBUG_DRIVER("drm plane:%d not enabled\n",
@@ -947,13 +1227,15 @@ static void sti_hqvdp_atomic_disable(struct drm_plane *drm_plane,
 	}
 
 	DRM_DEBUG_DRIVER("CRTC:%d (%s) drm plane:%d (%s)\n",
-			 drm_plane->crtc->base.id, sti_mixer_to_str(mixer),
+			 drm_plane->crtc->base.id,
+			 sti_mixer_to_str(to_sti_mixer(drm_plane->crtc)),
 			 drm_plane->base.id, sti_plane_to_str(plane));
 
 	plane->status = STI_PLANE_DISABLING;
 }
 
 static const struct drm_plane_helper_funcs sti_hqvdp_helpers_funcs = {
+	.atomic_check = sti_hqvdp_atomic_check,
 	.atomic_update = sti_hqvdp_atomic_update,
 	.atomic_disable = sti_hqvdp_atomic_disable,
 };
@@ -983,6 +1265,9 @@ static struct drm_plane *sti_hqvdp_create(struct drm_device *drm_dev,
 
 	sti_plane_init_property(&hqvdp->plane, DRM_PLANE_TYPE_OVERLAY);
 
+	if (hqvdp_debugfs_init(hqvdp, drm_dev->primary))
+		DRM_ERROR("HQVDP debugfs setup failed\n");
+
 	return &hqvdp->plane.drm_plane;
 }
 
diff --git a/drivers/gpu/drm/sti/sti_mixer.c b/drivers/gpu/drm/sti/sti_mixer.c
index 49db835dce03..e7425c38fc93 100644
--- a/drivers/gpu/drm/sti/sti_mixer.c
+++ b/drivers/gpu/drm/sti/sti_mixer.c
@@ -75,6 +75,145 @@ static inline void sti_mixer_reg_write(struct sti_mixer *mixer,
 	writel(val, mixer->regs + reg_id);
 }
 
+#define DBGFS_DUMP(reg) seq_printf(s, "\n  %-25s 0x%08X", #reg, \
+				   sti_mixer_reg_read(mixer, reg))
+
+static void mixer_dbg_ctl(struct seq_file *s, int val)
+{
+	unsigned int i;
+	int count = 0;
+	char *const disp_layer[] = {"BKG", "VID0", "VID1", "GDP0",
+				    "GDP1", "GDP2", "GDP3"};
+
+	seq_puts(s, "\tEnabled: ");
+	for (i = 0; i < 7; i++) {
+		if (val & 1) {
+			seq_printf(s, "%s ", disp_layer[i]);
+			count++;
+		}
+		val = val >> 1;
+	}
+
+	val = val >> 2;
+	if (val & 1) {
+		seq_puts(s, "CURS ");
+		count++;
+	}
+	if (!count)
+		seq_puts(s, "Nothing");
+}
+
+static void mixer_dbg_crb(struct seq_file *s, int val)
+{
+	int i;
+
+	seq_puts(s, "\tDepth: ");
+	for (i = 0; i < GAM_MIXER_NB_DEPTH_LEVEL; i++) {
+		switch (val & GAM_DEPTH_MASK_ID) {
+		case GAM_DEPTH_VID0_ID:
+			seq_puts(s, "VID0");
+			break;
+		case GAM_DEPTH_VID1_ID:
+			seq_puts(s, "VID1");
+			break;
+		case GAM_DEPTH_GDP0_ID:
+			seq_puts(s, "GDP0");
+			break;
+		case GAM_DEPTH_GDP1_ID:
+			seq_puts(s, "GDP1");
+			break;
+		case GAM_DEPTH_GDP2_ID:
+			seq_puts(s, "GDP2");
+			break;
+		case GAM_DEPTH_GDP3_ID:
+			seq_puts(s, "GDP3");
+			break;
+		default:
+			seq_puts(s, "---");
+		}
+
+		if (i < GAM_MIXER_NB_DEPTH_LEVEL - 1)
+			seq_puts(s, " < ");
+		val = val >> 3;
+	}
+}
+
+static void mixer_dbg_mxn(struct seq_file *s, void *addr)
+{
+	int i;
+
+	for (i = 1; i < 8; i++)
+		seq_printf(s, "-0x%08X", (int)readl(addr + i * 4));
+}
+
+static int mixer_dbg_show(struct seq_file *s, void *arg)
+{
+	struct drm_info_node *node = s->private;
+	struct sti_mixer *mixer = (struct sti_mixer *)node->info_ent->data;
+	struct drm_device *dev = node->minor->dev;
+	int ret;
+
+	ret = mutex_lock_interruptible(&dev->struct_mutex);
+	if (ret)
+		return ret;
+
+	seq_printf(s, "%s: (vaddr = 0x%p)",
+		   sti_mixer_to_str(mixer), mixer->regs);
+
+	DBGFS_DUMP(GAM_MIXER_CTL);
+	mixer_dbg_ctl(s, sti_mixer_reg_read(mixer, GAM_MIXER_CTL));
+	DBGFS_DUMP(GAM_MIXER_BKC);
+	DBGFS_DUMP(GAM_MIXER_BCO);
+	DBGFS_DUMP(GAM_MIXER_BCS);
+	DBGFS_DUMP(GAM_MIXER_AVO);
+	DBGFS_DUMP(GAM_MIXER_AVS);
+	DBGFS_DUMP(GAM_MIXER_CRB);
+	mixer_dbg_crb(s, sti_mixer_reg_read(mixer, GAM_MIXER_CRB));
+	DBGFS_DUMP(GAM_MIXER_ACT);
+	DBGFS_DUMP(GAM_MIXER_MBP);
+	DBGFS_DUMP(GAM_MIXER_MX0);
+	mixer_dbg_mxn(s, mixer->regs + GAM_MIXER_MX0);
+	seq_puts(s, "\n");
+
+	mutex_unlock(&dev->struct_mutex);
+	return 0;
+}
+
+static struct drm_info_list mixer0_debugfs_files[] = {
+	{ "mixer_main", mixer_dbg_show, 0, NULL },
+};
+
+static struct drm_info_list mixer1_debugfs_files[] = {
+	{ "mixer_aux", mixer_dbg_show, 0, NULL },
+};
+
+static int mixer_debugfs_init(struct sti_mixer *mixer, struct drm_minor *minor)
+{
+	unsigned int i;
+	struct drm_info_list *mixer_debugfs_files;
+	int nb_files;
+
+	switch (mixer->id) {
+	case STI_MIXER_MAIN:
+		mixer_debugfs_files = mixer0_debugfs_files;
+		nb_files = ARRAY_SIZE(mixer0_debugfs_files);
+		break;
+	case STI_MIXER_AUX:
+		mixer_debugfs_files = mixer1_debugfs_files;
+		nb_files = ARRAY_SIZE(mixer1_debugfs_files);
+		break;
+	default:
+		return -EINVAL;
+	}
+
+	for (i = 0; i < nb_files; i++)
+		mixer_debugfs_files[i].data = mixer;
+
+	return drm_debugfs_create_files(mixer_debugfs_files,
+					nb_files,
+					minor->debugfs_root, minor);
+}
+
 void sti_mixer_set_background_status(struct sti_mixer *mixer, bool enable)
 {
 	u32 val = sti_mixer_reg_read(mixer, GAM_MIXER_CTL);
@@ -237,7 +376,9 @@ void sti_mixer_set_matrix(struct sti_mixer *mixer)
 				    mixerColorSpaceMatIdentity[i]);
 }
 
-struct sti_mixer *sti_mixer_create(struct device *dev, int id,
+struct sti_mixer *sti_mixer_create(struct device *dev,
+				   struct drm_device *drm_dev,
+				   int id,
 				   void __iomem *baseaddr)
 {
 	struct sti_mixer *mixer = devm_kzalloc(dev, sizeof(*mixer), GFP_KERNEL);
@@ -258,5 +399,8 @@ struct sti_mixer *sti_mixer_create(struct device *dev, int id,
 	DRM_DEBUG_DRIVER("%s created. Regs=%p\n",
 			 sti_mixer_to_str(mixer), mixer->regs);
 
+	if (mixer_debugfs_init(mixer, drm_dev->primary))
+		DRM_ERROR("MIXER debugfs setup failed\n");
+
 	return mixer;
 }
diff --git a/drivers/gpu/drm/sti/sti_mixer.h b/drivers/gpu/drm/sti/sti_mixer.h
index efb1a9a5ba86..6f35fc086873 100644
--- a/drivers/gpu/drm/sti/sti_mixer.h
+++ b/drivers/gpu/drm/sti/sti_mixer.h
@@ -42,7 +42,9 @@ struct sti_mixer {
 
 const char *sti_mixer_to_str(struct sti_mixer *mixer);
 
-struct sti_mixer *sti_mixer_create(struct device *dev, int id,
+struct sti_mixer *sti_mixer_create(struct device *dev,
+				   struct drm_device *drm_dev,
+				   int id,
 				   void __iomem *baseaddr);
 
 int sti_mixer_set_plane_status(struct sti_mixer *mixer,
diff --git a/drivers/gpu/drm/sti/sti_plane.c b/drivers/gpu/drm/sti/sti_plane.c
index 2e5c751910c5..f10c98d3f012 100644
--- a/drivers/gpu/drm/sti/sti_plane.c
+++ b/drivers/gpu/drm/sti/sti_plane.c
@@ -43,6 +43,69 @@ const char *sti_plane_to_str(struct sti_plane *plane)
 	}
 }
 
+#define STI_FPS_INTERVAL_MS     3000
+
+static int sti_plane_timespec_ms_diff(struct timespec lhs, struct timespec rhs)
+{
+	struct timespec tmp_ts = timespec_sub(lhs, rhs);
+	u64 tmp_ns = (u64)timespec_to_ns(&tmp_ts);
+
+	do_div(tmp_ns, NSEC_PER_MSEC);
+
+	return (u32)tmp_ns;
+}
+
+void sti_plane_update_fps(struct sti_plane *plane,
+			  bool new_frame,
+			  bool new_field)
+{
+	struct timespec now;
+	struct sti_fps_info *fps;
+	int fpks, fipks, ms_since_last, num_frames, num_fields;
+
+	getrawmonotonic(&now);
+
+	/* Compute number of frame updates */
+	fps = &plane->fps_info;
+
+	if (new_field)
+		fps->curr_field_counter++;
+
+	/* do not perform fps calcul if new_frame is false */
+	if (!new_frame)
+		return;
+
+	fps->curr_frame_counter++;
+	ms_since_last = sti_plane_timespec_ms_diff(now, fps->last_timestamp);
+	num_frames = fps->curr_frame_counter - fps->last_frame_counter;
+
+	if (num_frames <= 0  || ms_since_last < STI_FPS_INTERVAL_MS)
+		return;
+
+	fps->last_timestamp = now;
+	fps->last_frame_counter = fps->curr_frame_counter;
+	fpks = (num_frames * 1000000) / ms_since_last;
+	snprintf(plane->fps_info.fps_str, FPS_LENGTH, "%-6s @ %d.%.3d fps",
+		 sti_plane_to_str(plane), fpks / 1000, fpks % 1000);
+
+	if (fps->curr_field_counter) {
+		/* Compute number of field updates */
+		num_fields = fps->curr_field_counter - fps->last_field_counter;
+		fps->last_field_counter = fps->curr_field_counter;
+		fipks = (num_fields * 1000000) / ms_since_last;
+		snprintf(plane->fps_info.fips_str,
+			 FPS_LENGTH, " - %d.%.3d field/sec",
+			 fipks / 1000, fipks % 1000);
+	} else {
+		plane->fps_info.fips_str[0] = '\0';
+	}
+
+	if (fps->output)
+		DRM_INFO("%s%s\n",
+			 plane->fps_info.fps_str,
+			 plane->fps_info.fips_str);
+}
+
 static void sti_plane_destroy(struct drm_plane *drm_plane)
 {
 	DRM_DEBUG_DRIVER("\n");
diff --git a/drivers/gpu/drm/sti/sti_plane.h b/drivers/gpu/drm/sti/sti_plane.h
index 86f1e6fc81b9..c50a3b9f5d37 100644
--- a/drivers/gpu/drm/sti/sti_plane.h
+++ b/drivers/gpu/drm/sti/sti_plane.h
@@ -50,6 +50,18 @@ enum sti_plane_status {
 	STI_PLANE_DISABLED,
 };
 
+#define FPS_LENGTH 64
+struct sti_fps_info {
+	bool output;
+	unsigned int curr_frame_counter;
+	unsigned int last_frame_counter;
+	unsigned int curr_field_counter;
+	unsigned int last_field_counter;
+	struct timespec last_timestamp;
+	char fps_str[FPS_LENGTH];
+	char fips_str[FPS_LENGTH];
+};
+
 /**
  * STI plane structure
  *
@@ -57,15 +69,20 @@ enum sti_plane_status {
  * @desc:               plane type & id
  * @status:             to know the status of the plane
  * @zorder:             plane z-order
+ * @fps_info:           frame per second info
  */
 struct sti_plane {
 	struct drm_plane drm_plane;
 	enum sti_plane_desc desc;
 	enum sti_plane_status status;
 	int zorder;
+	struct sti_fps_info fps_info;
 };
 
 const char *sti_plane_to_str(struct sti_plane *plane);
+void sti_plane_update_fps(struct sti_plane *plane,
+			  bool new_frame,
+			  bool new_field);
 void sti_plane_init_property(struct sti_plane *plane,
 			     enum drm_plane_type type);
 #endif
diff --git a/drivers/gpu/drm/sti/sti_tvout.c b/drivers/gpu/drm/sti/sti_tvout.c
index 24a3735b88fd..2c99016443e5 100644
--- a/drivers/gpu/drm/sti/sti_tvout.c
+++ b/drivers/gpu/drm/sti/sti_tvout.c
@@ -17,6 +17,7 @@
 #include <drm/drm_crtc_helper.h>
 
 #include "sti_crtc.h"
+#include "sti_vtg.h"
 
 /* glue registers */
 #define TVO_CSC_MAIN_M0                  0x000
@@ -85,19 +86,7 @@
 #define TVO_VIP_SEL_INPUT_BYPASSED       1
 
 #define TVO_SYNC_MAIN_VTG_SET_REF        0x00
-#define TVO_SYNC_MAIN_VTG_SET_1          0x01
-#define TVO_SYNC_MAIN_VTG_SET_2          0x02
-#define TVO_SYNC_MAIN_VTG_SET_3          0x03
-#define TVO_SYNC_MAIN_VTG_SET_4          0x04
-#define TVO_SYNC_MAIN_VTG_SET_5          0x05
-#define TVO_SYNC_MAIN_VTG_SET_6          0x06
 #define TVO_SYNC_AUX_VTG_SET_REF         0x10
-#define TVO_SYNC_AUX_VTG_SET_1           0x11
-#define TVO_SYNC_AUX_VTG_SET_2           0x12
-#define TVO_SYNC_AUX_VTG_SET_3           0x13
-#define TVO_SYNC_AUX_VTG_SET_4           0x14
-#define TVO_SYNC_AUX_VTG_SET_5           0x15
-#define TVO_SYNC_AUX_VTG_SET_6           0x16
 
 #define TVO_SYNC_HD_DCS_SHIFT            8
 
@@ -106,6 +95,8 @@
 
 #define ENCODER_CRTC_MASK                (BIT(0) | BIT(1))
 
+#define TVO_MIN_HD_HEIGHT                720
+
 /* enum listing the supported output data format */
 enum sti_tvout_video_out_type {
 	STI_TVOUT_VIDEO_OUT_RGB,
@@ -269,6 +260,31 @@ static void tvout_vip_set_in_vid_fmt(struct sti_tvout *tvout,
 }
 
 /**
+ * Set preformatter matrix
+ *
+ * @tvout: tvout structure
+ * @mode: display mode structure
+ */
+static void tvout_preformatter_set_matrix(struct sti_tvout *tvout,
+					  struct drm_display_mode *mode)
+{
+	unsigned int i;
+	const u32 *pf_matrix;
+
+	if (mode->vdisplay >= TVO_MIN_HD_HEIGHT)
+		pf_matrix = rgb_to_ycbcr_709;
+	else
+		pf_matrix = rgb_to_ycbcr_601;
+
+	for (i = 0; i < 8; i++) {
+		tvout_write(tvout, *(pf_matrix + i),
+			    TVO_CSC_MAIN_M0 + (i * 4));
+		tvout_write(tvout, *(pf_matrix + i),
+			    TVO_CSC_AUX_M0 + (i * 4));
+	}
+}
+
+/**
  * Start VIP block for DVO output
  *
  * @tvout: pointer on tvout structure
@@ -280,24 +296,26 @@ static void tvout_dvo_start(struct sti_tvout *tvout, bool main_path)
 	struct device_node *node = tvout->dev->of_node;
 	bool sel_input_logic_inverted = false;
 	u32 tvo_in_vid_format;
-	int val;
+	int val, tmp;
 
 	dev_dbg(tvout->dev, "%s\n", __func__);
 
 	if (main_path) {
 		DRM_DEBUG_DRIVER("main vip for DVO\n");
-		/* Select the input sync for dvo = VTG set 4 */
-		val  = TVO_SYNC_MAIN_VTG_SET_4 << TVO_SYNC_DVO_PAD_VSYNC_SHIFT;
-		val |= TVO_SYNC_MAIN_VTG_SET_4 << TVO_SYNC_DVO_PAD_HSYNC_SHIFT;
-		val |= TVO_SYNC_MAIN_VTG_SET_4;
+		/* Select the input sync for dvo */
+		tmp = TVO_SYNC_MAIN_VTG_SET_REF | VTG_SYNC_ID_DVO;
+		val  = tmp << TVO_SYNC_DVO_PAD_VSYNC_SHIFT;
+		val |= tmp << TVO_SYNC_DVO_PAD_HSYNC_SHIFT;
+		val |= tmp;
 		tvout_write(tvout, val, TVO_DVO_SYNC_SEL);
 		tvo_in_vid_format = TVO_MAIN_IN_VID_FORMAT;
 	} else {
 		DRM_DEBUG_DRIVER("aux vip for DVO\n");
-		/* Select the input sync for dvo = VTG set 4 */
-		val  = TVO_SYNC_AUX_VTG_SET_4 << TVO_SYNC_DVO_PAD_VSYNC_SHIFT;
-		val |= TVO_SYNC_AUX_VTG_SET_4 << TVO_SYNC_DVO_PAD_HSYNC_SHIFT;
-		val |= TVO_SYNC_AUX_VTG_SET_4;
+		/* Select the input sync for dvo */
+		tmp = TVO_SYNC_AUX_VTG_SET_REF | VTG_SYNC_ID_DVO;
+		val  = tmp << TVO_SYNC_DVO_PAD_VSYNC_SHIFT;
+		val |= tmp << TVO_SYNC_DVO_PAD_HSYNC_SHIFT;
+		val |= tmp;
 		tvout_write(tvout, val, TVO_DVO_SYNC_SEL);
 		tvo_in_vid_format = TVO_AUX_IN_VID_FORMAT;
 	}
@@ -308,9 +326,8 @@ static void tvout_dvo_start(struct sti_tvout *tvout, bool main_path)
 				  TVO_VIP_REORDER_Y_G_SEL,
 				  TVO_VIP_REORDER_CB_B_SEL);
 
-	/* Set clipping mode (Limited range RGB/Y) */
-	tvout_vip_set_clip_mode(tvout, TVO_VIP_DVO,
-				TVO_VIP_CLIP_LIMITED_RANGE_RGB_Y);
+	/* Set clipping mode */
+	tvout_vip_set_clip_mode(tvout, TVO_VIP_DVO, TVO_VIP_CLIP_DISABLED);
 
 	/* Set round mode (rounded to 8-bit per component) */
 	tvout_vip_set_rnd(tvout, TVO_VIP_DVO, TVO_VIP_RND_8BIT_ROUNDED);
@@ -345,13 +362,17 @@ static void tvout_hdmi_start(struct sti_tvout *tvout, bool main_path)
 
 	if (main_path) {
 		DRM_DEBUG_DRIVER("main vip for hdmi\n");
-		/* select the input sync for hdmi = VTG set 1 */
-		tvout_write(tvout, TVO_SYNC_MAIN_VTG_SET_1, TVO_HDMI_SYNC_SEL);
+		/* select the input sync for hdmi */
+		tvout_write(tvout,
+			    TVO_SYNC_MAIN_VTG_SET_REF | VTG_SYNC_ID_HDMI,
+			    TVO_HDMI_SYNC_SEL);
 		tvo_in_vid_format = TVO_MAIN_IN_VID_FORMAT;
 	} else {
 		DRM_DEBUG_DRIVER("aux vip for hdmi\n");
-		/* select the input sync for hdmi = VTG set 1 */
-		tvout_write(tvout, TVO_SYNC_AUX_VTG_SET_1, TVO_HDMI_SYNC_SEL);
+		/* select the input sync for hdmi */
+		tvout_write(tvout,
+			    TVO_SYNC_AUX_VTG_SET_REF | VTG_SYNC_ID_HDMI,
+			    TVO_HDMI_SYNC_SEL);
 		tvo_in_vid_format = TVO_AUX_IN_VID_FORMAT;
 	}
 
@@ -361,9 +382,8 @@ static void tvout_hdmi_start(struct sti_tvout *tvout, bool main_path)
 				  TVO_VIP_REORDER_Y_G_SEL,
 				  TVO_VIP_REORDER_CB_B_SEL);
 
-	/* set clipping mode (Limited range RGB/Y) */
-	tvout_vip_set_clip_mode(tvout, TVO_VIP_HDMI,
-			TVO_VIP_CLIP_LIMITED_RANGE_RGB_Y);
+	/* set clipping mode */
+	tvout_vip_set_clip_mode(tvout, TVO_VIP_HDMI, TVO_VIP_CLIP_DISABLED);
 
 	/* set round mode (rounded to 8-bit per component) */
 	tvout_vip_set_rnd(tvout, TVO_VIP_HDMI, TVO_VIP_RND_8BIT_ROUNDED);
@@ -397,13 +417,19 @@ static void tvout_hda_start(struct sti_tvout *tvout, bool main_path)
 	dev_dbg(tvout->dev, "%s\n", __func__);
 
 	if (main_path) {
-		val = TVO_SYNC_MAIN_VTG_SET_2 << TVO_SYNC_HD_DCS_SHIFT;
-		val |= TVO_SYNC_MAIN_VTG_SET_3;
+		DRM_DEBUG_DRIVER("main vip for HDF\n");
+		/* Select the input sync for HD analog and HD DCS */
+		val  = TVO_SYNC_MAIN_VTG_SET_REF | VTG_SYNC_ID_HDDCS;
+		val  = val << TVO_SYNC_HD_DCS_SHIFT;
+		val |= TVO_SYNC_MAIN_VTG_SET_REF | VTG_SYNC_ID_HDF;
 		tvout_write(tvout, val, TVO_HD_SYNC_SEL);
 		tvo_in_vid_format = TVO_MAIN_IN_VID_FORMAT;
 	} else {
-		val = TVO_SYNC_AUX_VTG_SET_2 << TVO_SYNC_HD_DCS_SHIFT;
-		val |= TVO_SYNC_AUX_VTG_SET_3;
+		DRM_DEBUG_DRIVER("aux vip for HDF\n");
+		/* Select the input sync for HD analog and HD DCS */
+		val  = TVO_SYNC_AUX_VTG_SET_REF | VTG_SYNC_ID_HDDCS;
+		val  = val << TVO_SYNC_HD_DCS_SHIFT;
+		val |= TVO_SYNC_AUX_VTG_SET_REF | VTG_SYNC_ID_HDF;
 		tvout_write(tvout, val, TVO_HD_SYNC_SEL);
 		tvo_in_vid_format = TVO_AUX_IN_VID_FORMAT;
 	}
@@ -414,8 +440,8 @@ static void tvout_hda_start(struct sti_tvout *tvout, bool main_path)
 				  TVO_VIP_REORDER_Y_G_SEL,
 				  TVO_VIP_REORDER_CB_B_SEL);
 
-	/* set clipping mode (EAV/SAV clipping) */
-	tvout_vip_set_clip_mode(tvout, TVO_VIP_HDF, TVO_VIP_CLIP_EAV_SAV);
+	/* set clipping mode */
+	tvout_vip_set_clip_mode(tvout, TVO_VIP_HDF, TVO_VIP_CLIP_DISABLED);
 
 	/* set round mode (rounded to 10-bit per component) */
 	tvout_vip_set_rnd(tvout, TVO_VIP_HDF, TVO_VIP_RND_10BIT_ROUNDED);
@@ -436,6 +462,157 @@ static void tvout_hda_start(struct sti_tvout *tvout, bool main_path)
 	tvout_write(tvout, 0, TVO_HD_DAC_CFG_OFF);
 }
 
+#define DBGFS_DUMP(reg) seq_printf(s, "\n  %-25s 0x%08X", #reg, \
+				   readl(tvout->regs + reg))
+
+static void tvout_dbg_vip(struct seq_file *s, int val)
+{
+	int r, g, b, tmp, mask;
+	char *const reorder[] = {"Y_G", "Cb_B", "Cr_R"};
+	char *const clipping[] = {"No", "EAV/SAV", "Limited range RGB/Y",
+				  "Limited range Cb/Cr", "decided by register"};
+	char *const round[] = {"8-bit", "10-bit", "12-bit"};
+	char *const input_sel[] = {"Main (color matrix enabled)",
+				   "Main (color matrix by-passed)",
+				   "", "", "", "", "", "",
+				   "Aux (color matrix enabled)",
+				   "Aux (color matrix by-passed)",
+				   "", "", "", "", "", "Force value"};
+
+	seq_puts(s, "\t");
+	mask = TVO_VIP_REORDER_MASK << TVO_VIP_REORDER_R_SHIFT;
+	r = (val & mask) >> TVO_VIP_REORDER_R_SHIFT;
+	mask = TVO_VIP_REORDER_MASK << TVO_VIP_REORDER_G_SHIFT;
+	g = (val & mask) >> TVO_VIP_REORDER_G_SHIFT;
+	mask = TVO_VIP_REORDER_MASK << TVO_VIP_REORDER_B_SHIFT;
+	b = (val & mask) >> TVO_VIP_REORDER_B_SHIFT;
+	seq_printf(s, "%-24s %s->%s %s->%s %s->%s\n", "Reorder:",
+		   reorder[r], reorder[TVO_VIP_REORDER_CR_R_SEL],
+		   reorder[g], reorder[TVO_VIP_REORDER_Y_G_SEL],
+		   reorder[b], reorder[TVO_VIP_REORDER_CB_B_SEL]);
+	seq_puts(s, "\t\t\t\t\t");
+	mask = TVO_VIP_CLIP_MASK << TVO_VIP_CLIP_SHIFT;
+	tmp = (val & mask) >> TVO_VIP_CLIP_SHIFT;
+	seq_printf(s, "%-24s %s\n", "Clipping:", clipping[tmp]);
+	seq_puts(s, "\t\t\t\t\t");
+	mask = TVO_VIP_RND_MASK << TVO_VIP_RND_SHIFT;
+	tmp = (val & mask) >> TVO_VIP_RND_SHIFT;
+	seq_printf(s, "%-24s input data rounded to %s per component\n",
+		   "Round:", round[tmp]);
+	seq_puts(s, "\t\t\t\t\t");
+	tmp = (val & TVO_VIP_SEL_INPUT_MASK);
+	seq_printf(s, "%-24s %s", "Input selection:", input_sel[tmp]);
+}
+
+static void tvout_dbg_hd_dac_cfg(struct seq_file *s, int val)
+{
+	seq_printf(s, "\t%-24s %s", "HD DAC:",
+		   val & 1 ? "disabled" : "enabled");
+}
+
+static int tvout_dbg_show(struct seq_file *s, void *data)
+{
+	struct drm_info_node *node = s->private;
+	struct sti_tvout *tvout = (struct sti_tvout *)node->info_ent->data;
+	struct drm_device *dev = node->minor->dev;
+	struct drm_crtc *crtc;
+	int ret;
+
+	ret = mutex_lock_interruptible(&dev->struct_mutex);
+	if (ret)
+		return ret;
+
+	seq_printf(s, "TVOUT: (vaddr = 0x%p)", tvout->regs);
+
+	seq_puts(s, "\n\n  HDMI encoder: ");
+	crtc = tvout->hdmi->crtc;
+	if (crtc) {
+		seq_printf(s, "connected to %s path",
+			   sti_crtc_is_main(crtc) ? "main" : "aux");
+		DBGFS_DUMP(TVO_HDMI_SYNC_SEL);
+		DBGFS_DUMP(TVO_VIP_HDMI);
+		tvout_dbg_vip(s, readl(tvout->regs + TVO_VIP_HDMI));
+	} else {
+		seq_puts(s, "disabled");
+	}
+
+	seq_puts(s, "\n\n  DVO encoder: ");
+	crtc = tvout->dvo->crtc;
+	if (crtc) {
+		seq_printf(s, "connected to %s path",
+			   sti_crtc_is_main(crtc) ? "main" : "aux");
+		DBGFS_DUMP(TVO_DVO_SYNC_SEL);
+		DBGFS_DUMP(TVO_DVO_CONFIG);
+		DBGFS_DUMP(TVO_VIP_DVO);
+		tvout_dbg_vip(s, readl(tvout->regs + TVO_VIP_DVO));
+	} else {
+		seq_puts(s, "disabled");
+	}
+
+	seq_puts(s, "\n\n  HDA encoder: ");
+	crtc = tvout->hda->crtc;
+	if (crtc) {
+		seq_printf(s, "connected to %s path",
+			   sti_crtc_is_main(crtc) ? "main" : "aux");
+		DBGFS_DUMP(TVO_HD_SYNC_SEL);
+		DBGFS_DUMP(TVO_HD_DAC_CFG_OFF);
+		tvout_dbg_hd_dac_cfg(s,
+				     readl(tvout->regs + TVO_HD_DAC_CFG_OFF));
+		DBGFS_DUMP(TVO_VIP_HDF);
+		tvout_dbg_vip(s, readl(tvout->regs + TVO_VIP_HDF));
+	} else {
+		seq_puts(s, "disabled");
+	}
+
+	seq_puts(s, "\n\n  main path configuration");
+	DBGFS_DUMP(TVO_CSC_MAIN_M0);
+	DBGFS_DUMP(TVO_CSC_MAIN_M1);
+	DBGFS_DUMP(TVO_CSC_MAIN_M2);
+	DBGFS_DUMP(TVO_CSC_MAIN_M3);
+	DBGFS_DUMP(TVO_CSC_MAIN_M4);
+	DBGFS_DUMP(TVO_CSC_MAIN_M5);
+	DBGFS_DUMP(TVO_CSC_MAIN_M6);
+	DBGFS_DUMP(TVO_CSC_MAIN_M7);
+	DBGFS_DUMP(TVO_MAIN_IN_VID_FORMAT);
+
+	seq_puts(s, "\n\n  auxiliary path configuration");
+	DBGFS_DUMP(TVO_CSC_AUX_M0);
+	DBGFS_DUMP(TVO_CSC_AUX_M2);
+	DBGFS_DUMP(TVO_CSC_AUX_M3);
+	DBGFS_DUMP(TVO_CSC_AUX_M4);
+	DBGFS_DUMP(TVO_CSC_AUX_M5);
+	DBGFS_DUMP(TVO_CSC_AUX_M6);
+	DBGFS_DUMP(TVO_CSC_AUX_M7);
+	DBGFS_DUMP(TVO_AUX_IN_VID_FORMAT);
+	seq_puts(s, "\n");
+
+	mutex_unlock(&dev->struct_mutex);
+	return 0;
+}
+
+static struct drm_info_list tvout_debugfs_files[] = {
+	{ "tvout", tvout_dbg_show, 0, NULL },
+};
+
+static void tvout_debugfs_exit(struct sti_tvout *tvout, struct drm_minor *minor)
+{
+	drm_debugfs_remove_files(tvout_debugfs_files,
+				 ARRAY_SIZE(tvout_debugfs_files),
+				 minor);
+}
+
+static int tvout_debugfs_init(struct sti_tvout *tvout, struct drm_minor *minor)
+{
+	unsigned int i;
+
+	for (i = 0; i < ARRAY_SIZE(tvout_debugfs_files); i++)
+		tvout_debugfs_files[i].data = tvout;
+
+	return drm_debugfs_create_files(tvout_debugfs_files,
+					ARRAY_SIZE(tvout_debugfs_files),
+					minor->debugfs_root, minor);
+}
+
 static void sti_tvout_encoder_dpms(struct drm_encoder *encoder, int mode)
 {
 }
@@ -446,10 +623,6 @@ static void sti_tvout_encoder_mode_set(struct drm_encoder *encoder,
 {
 }
 
-static void sti_tvout_encoder_prepare(struct drm_encoder *encoder)
-{
-}
-
 static void sti_tvout_encoder_destroy(struct drm_encoder *encoder)
 {
 	struct sti_tvout_encoder *sti_encoder = to_sti_tvout_encoder(encoder);
@@ -462,10 +635,12 @@ static const struct drm_encoder_funcs sti_tvout_encoder_funcs = {
 	.destroy = sti_tvout_encoder_destroy,
 };
 
-static void sti_dvo_encoder_commit(struct drm_encoder *encoder)
+static void sti_dvo_encoder_enable(struct drm_encoder *encoder)
 {
 	struct sti_tvout *tvout = to_sti_tvout(encoder);
 
+	tvout_preformatter_set_matrix(tvout, &encoder->crtc->mode);
+
 	tvout_dvo_start(tvout, sti_crtc_is_main(encoder->crtc));
 }
 
@@ -480,8 +655,7 @@ static void sti_dvo_encoder_disable(struct drm_encoder *encoder)
 static const struct drm_encoder_helper_funcs sti_dvo_encoder_helper_funcs = {
 	.dpms = sti_tvout_encoder_dpms,
 	.mode_set = sti_tvout_encoder_mode_set,
-	.prepare = sti_tvout_encoder_prepare,
-	.commit = sti_dvo_encoder_commit,
+	.enable = sti_dvo_encoder_enable,
 	.disable = sti_dvo_encoder_disable,
 };
 
@@ -512,10 +686,12 @@ sti_tvout_create_dvo_encoder(struct drm_device *dev,
 	return drm_encoder;
 }
 
-static void sti_hda_encoder_commit(struct drm_encoder *encoder)
+static void sti_hda_encoder_enable(struct drm_encoder *encoder)
 {
 	struct sti_tvout *tvout = to_sti_tvout(encoder);
 
+	tvout_preformatter_set_matrix(tvout, &encoder->crtc->mode);
+
 	tvout_hda_start(tvout, sti_crtc_is_main(encoder->crtc));
 }
 
@@ -533,8 +709,7 @@ static void sti_hda_encoder_disable(struct drm_encoder *encoder)
 static const struct drm_encoder_helper_funcs sti_hda_encoder_helper_funcs = {
 	.dpms = sti_tvout_encoder_dpms,
 	.mode_set = sti_tvout_encoder_mode_set,
-	.prepare = sti_tvout_encoder_prepare,
-	.commit = sti_hda_encoder_commit,
+	.commit = sti_hda_encoder_enable,
 	.disable = sti_hda_encoder_disable,
 };
 
@@ -563,10 +738,12 @@ static struct drm_encoder *sti_tvout_create_hda_encoder(struct drm_device *dev,
 	return drm_encoder;
 }
 
-static void sti_hdmi_encoder_commit(struct drm_encoder *encoder)
+static void sti_hdmi_encoder_enable(struct drm_encoder *encoder)
 {
 	struct sti_tvout *tvout = to_sti_tvout(encoder);
 
+	tvout_preformatter_set_matrix(tvout, &encoder->crtc->mode);
+
 	tvout_hdmi_start(tvout, sti_crtc_is_main(encoder->crtc));
 }
 
@@ -581,8 +758,7 @@ static void sti_hdmi_encoder_disable(struct drm_encoder *encoder)
 static const struct drm_encoder_helper_funcs sti_hdmi_encoder_helper_funcs = {
 	.dpms = sti_tvout_encoder_dpms,
 	.mode_set = sti_tvout_encoder_mode_set,
-	.prepare = sti_tvout_encoder_prepare,
-	.commit = sti_hdmi_encoder_commit,
+	.commit = sti_hdmi_encoder_enable,
 	.disable = sti_hdmi_encoder_disable,
 };
 
@@ -628,26 +804,24 @@ static void sti_tvout_destroy_encoders(struct sti_tvout *tvout)
 	if (tvout->hda)
 		drm_encoder_cleanup(tvout->hda);
 	tvout->hda = NULL;
+
+	if (tvout->dvo)
+		drm_encoder_cleanup(tvout->dvo);
+	tvout->dvo = NULL;
 }
 
 static int sti_tvout_bind(struct device *dev, struct device *master, void *data)
 {
 	struct sti_tvout *tvout = dev_get_drvdata(dev);
 	struct drm_device *drm_dev = data;
-	unsigned int i;
 
 	tvout->drm_dev = drm_dev;
 
-	/* set preformatter matrix */
-	for (i = 0; i < 8; i++) {
-		tvout_write(tvout, rgb_to_ycbcr_601[i],
-			TVO_CSC_MAIN_M0 + (i * 4));
-		tvout_write(tvout, rgb_to_ycbcr_601[i],
-			TVO_CSC_AUX_M0 + (i * 4));
-	}
-
 	sti_tvout_create_encoders(drm_dev, tvout);
 
+	if (tvout_debugfs_init(tvout, drm_dev->primary))
+		DRM_ERROR("TVOUT debugfs setup failed\n");
+
 	return 0;
 }
 
@@ -655,8 +829,11 @@ static void sti_tvout_unbind(struct device *dev, struct device *master,
 	void *data)
 {
 	struct sti_tvout *tvout = dev_get_drvdata(dev);
+	struct drm_device *drm_dev = data;
 
 	sti_tvout_destroy_encoders(tvout);
+
+	tvout_debugfs_exit(tvout, drm_dev->primary);
 }
 
 static const struct component_ops sti_tvout_ops = {
diff --git a/drivers/gpu/drm/sti/sti_vid.c b/drivers/gpu/drm/sti/sti_vid.c
index a8254cc362a1..5a2c5dc3687b 100644
--- a/drivers/gpu/drm/sti/sti_vid.c
+++ b/drivers/gpu/drm/sti/sti_vid.c
@@ -42,6 +42,104 @@
 #define VID_MPR1_BT709          0x0AC50000
 #define VID_MPR2_BT709          0x07150545
 #define VID_MPR3_BT709          0x00000AE8
+/* YCbCr to RGB BT709:
+ * R = Y+1.3711Cr
+ * G = Y-0.6992Cr-0.3359Cb
+ * B = Y+1.7344Cb
+ */
+#define VID_MPR0_BT601          0x0A800000
+#define VID_MPR1_BT601          0x0AAF0000
+#define VID_MPR2_BT601          0x094E0754
+#define VID_MPR3_BT601          0x00000ADD
+
+#define VID_MIN_HD_HEIGHT       720
+
+#define DBGFS_DUMP(reg) seq_printf(s, "\n  %-25s 0x%08X", #reg, \
+				   readl(vid->regs + reg))
+
+static void vid_dbg_ctl(struct seq_file *s, int val)
+{
+	val = val >> 30;
+	seq_puts(s, "\t");
+
+	if (!(val & 1))
+		seq_puts(s, "NOT ");
+	seq_puts(s, "ignored on main mixer - ");
+
+	if (!(val & 2))
+		seq_puts(s, "NOT ");
+	seq_puts(s, "ignored on aux mixer");
+}
+
+static void vid_dbg_vpo(struct seq_file *s, int val)
+{
+	seq_printf(s, "\txdo:%4d\tydo:%4d", val & 0x0FFF, (val >> 16) & 0x0FFF);
+}
+
+static void vid_dbg_vps(struct seq_file *s, int val)
+{
+	seq_printf(s, "\txds:%4d\tyds:%4d", val & 0x0FFF, (val >> 16) & 0x0FFF);
+}
+
+static void vid_dbg_mst(struct seq_file *s, int val)
+{
+	if (val & 1)
+		seq_puts(s, "\tBUFFER UNDERFLOW!");
+}
+
+static int vid_dbg_show(struct seq_file *s, void *arg)
+{
+	struct drm_info_node *node = s->private;
+	struct sti_vid *vid = (struct sti_vid *)node->info_ent->data;
+	struct drm_device *dev = node->minor->dev;
+	int ret;
+
+	ret = mutex_lock_interruptible(&dev->struct_mutex);
+	if (ret)
+		return ret;
+
+	seq_printf(s, "VID: (vaddr= 0x%p)", vid->regs);
+
+	DBGFS_DUMP(VID_CTL);
+	vid_dbg_ctl(s, readl(vid->regs + VID_CTL));
+	DBGFS_DUMP(VID_ALP);
+	DBGFS_DUMP(VID_CLF);
+	DBGFS_DUMP(VID_VPO);
+	vid_dbg_vpo(s, readl(vid->regs + VID_VPO));
+	DBGFS_DUMP(VID_VPS);
+	vid_dbg_vps(s, readl(vid->regs + VID_VPS));
+	DBGFS_DUMP(VID_KEY1);
+	DBGFS_DUMP(VID_KEY2);
+	DBGFS_DUMP(VID_MPR0);
+	DBGFS_DUMP(VID_MPR1);
+	DBGFS_DUMP(VID_MPR2);
+	DBGFS_DUMP(VID_MPR3);
+	DBGFS_DUMP(VID_MST);
+	vid_dbg_mst(s, readl(vid->regs + VID_MST));
+	DBGFS_DUMP(VID_BC);
+	DBGFS_DUMP(VID_TINT);
+	DBGFS_DUMP(VID_CSAT);
+	seq_puts(s, "\n");
+
+	mutex_unlock(&dev->struct_mutex);
+	return 0;
+}
+
+static struct drm_info_list vid_debugfs_files[] = {
+	{ "vid", vid_dbg_show, 0, NULL },
+};
+
+static int vid_debugfs_init(struct sti_vid *vid, struct drm_minor *minor)
+{
+	unsigned int i;
+
+	for (i = 0; i < ARRAY_SIZE(vid_debugfs_files); i++)
+		vid_debugfs_files[i].data = vid;
+
+	return drm_debugfs_create_files(vid_debugfs_files,
+					ARRAY_SIZE(vid_debugfs_files),
+					minor->debugfs_root, minor);
+}
 
 void sti_vid_commit(struct sti_vid *vid,
 		    struct drm_plane_state *state)
@@ -52,6 +150,7 @@ void sti_vid_commit(struct sti_vid *vid,
 	int dst_y = state->crtc_y;
 	int dst_w = clamp_val(state->crtc_w, 0, mode->crtc_hdisplay - dst_x);
 	int dst_h = clamp_val(state->crtc_h, 0, mode->crtc_vdisplay - dst_y);
+	int src_h = state->src_h >> 16;
 	u32 val, ydo, xdo, yds, xds;
 
 	/* Input / output size
@@ -71,6 +170,19 @@ void sti_vid_commit(struct sti_vid *vid,
 
 	writel((ydo << 16) | xdo, vid->regs + VID_VPO);
 	writel((yds << 16) | xds, vid->regs + VID_VPS);
+
+	/* Color conversion parameters */
+	if (src_h >= VID_MIN_HD_HEIGHT) {
+		writel(VID_MPR0_BT709, vid->regs + VID_MPR0);
+		writel(VID_MPR1_BT709, vid->regs + VID_MPR1);
+		writel(VID_MPR2_BT709, vid->regs + VID_MPR2);
+		writel(VID_MPR3_BT709, vid->regs + VID_MPR3);
+	} else {
+		writel(VID_MPR0_BT601, vid->regs + VID_MPR0);
+		writel(VID_MPR1_BT601, vid->regs + VID_MPR1);
+		writel(VID_MPR2_BT601, vid->regs + VID_MPR2);
+		writel(VID_MPR3_BT601, vid->regs + VID_MPR3);
+	}
 }
 
 void sti_vid_disable(struct sti_vid *vid)
@@ -91,20 +203,14 @@ static void sti_vid_init(struct sti_vid *vid)
 	/* Opaque */
 	writel(VID_ALP_OPAQUE, vid->regs + VID_ALP);
 
-	/* Color conversion parameters */
-	writel(VID_MPR0_BT709, vid->regs + VID_MPR0);
-	writel(VID_MPR1_BT709, vid->regs + VID_MPR1);
-	writel(VID_MPR2_BT709, vid->regs + VID_MPR2);
-	writel(VID_MPR3_BT709, vid->regs + VID_MPR3);
-
 	/* Brightness, contrast, tint, saturation */
 	writel(VID_BC_DFLT, vid->regs + VID_BC);
 	writel(VID_TINT_DFLT, vid->regs + VID_TINT);
 	writel(VID_CSAT_DFLT, vid->regs + VID_CSAT);
 }
 
-struct sti_vid *sti_vid_create(struct device *dev, int id,
-			       void __iomem *baseaddr)
+struct sti_vid *sti_vid_create(struct device *dev, struct drm_device *drm_dev,
+			       int id, void __iomem *baseaddr)
 {
 	struct sti_vid *vid;
 
@@ -120,5 +226,8 @@ struct sti_vid *sti_vid_create(struct device *dev, int id,
 
 	sti_vid_init(vid);
 
+	if (vid_debugfs_init(vid, drm_dev->primary))
+		DRM_ERROR("VID debugfs setup failed\n");
+
 	return vid;
 }
diff --git a/drivers/gpu/drm/sti/sti_vid.h b/drivers/gpu/drm/sti/sti_vid.h
index 5dea4791f1d6..6c842344f3d8 100644
--- a/drivers/gpu/drm/sti/sti_vid.h
+++ b/drivers/gpu/drm/sti/sti_vid.h
@@ -23,7 +23,7 @@ struct sti_vid {
 void sti_vid_commit(struct sti_vid *vid,
 		    struct drm_plane_state *state);
 void sti_vid_disable(struct sti_vid *vid);
-struct sti_vid *sti_vid_create(struct device *dev, int id,
-			       void __iomem *baseaddr);
+struct sti_vid *sti_vid_create(struct device *dev, struct drm_device *drm_dev,
+			       int id, void __iomem *baseaddr);
 
 #endif
diff --git a/drivers/gpu/drm/sti/sti_vtg.c b/drivers/gpu/drm/sti/sti_vtg.c
index d56630c60039..32c7986b63ab 100644
--- a/drivers/gpu/drm/sti/sti_vtg.c
+++ b/drivers/gpu/drm/sti/sti_vtg.c
@@ -15,8 +15,8 @@
 
 #include "sti_vtg.h"
 
-#define VTG_TYPE_MASTER         0
-#define VTG_TYPE_SLAVE_BY_EXT0  1
+#define VTG_MODE_MASTER         0
+#define VTG_MODE_SLAVE_BY_EXT0  1
 
 /* registers offset */
 #define VTG_MODE            0x0000
@@ -64,6 +64,9 @@
 /* Delay introduced by the HDMI in nb of pixel */
 #define HDMI_DELAY          (5)
 
+/* Delay introduced by the DVO in nb of pixel */
+#define DVO_DELAY           (2)
+
 /* delay introduced by the Arbitrary Waveform Generator in nb of pixels */
 #define AWG_DELAY_HD        (-9)
 #define AWG_DELAY_ED        (-8)
@@ -71,13 +74,61 @@
 
 LIST_HEAD(vtg_lookup);
 
+/*
+ * STI VTG register offset structure
+ *
+ *@h_hd:     stores the VTG_H_HD_x     register offset
+ *@top_v_vd: stores the VTG_TOP_V_VD_x register offset
+ *@bot_v_vd: stores the VTG_BOT_V_VD_x register offset
+ *@top_v_hd: stores the VTG_TOP_V_HD_x register offset
+ *@bot_v_hd: stores the VTG_BOT_V_HD_x register offset
+ */
+struct sti_vtg_regs_offs {
+	u32 h_hd;
+	u32 top_v_vd;
+	u32 bot_v_vd;
+	u32 top_v_hd;
+	u32 bot_v_hd;
+};
+
+#define VTG_MAX_SYNC_OUTPUT 4
+static const struct sti_vtg_regs_offs vtg_regs_offs[VTG_MAX_SYNC_OUTPUT] = {
+	{ VTG_H_HD_1,
+	  VTG_TOP_V_VD_1, VTG_BOT_V_VD_1, VTG_TOP_V_HD_1, VTG_BOT_V_HD_1 },
+	{ VTG_H_HD_2,
+	  VTG_TOP_V_VD_2, VTG_BOT_V_VD_2, VTG_TOP_V_HD_2, VTG_BOT_V_HD_2 },
+	{ VTG_H_HD_3,
+	  VTG_TOP_V_VD_3, VTG_BOT_V_VD_3, VTG_TOP_V_HD_3, VTG_BOT_V_HD_3 },
+	{ VTG_H_HD_4,
+	  VTG_TOP_V_VD_4, VTG_BOT_V_VD_4, VTG_TOP_V_HD_4, VTG_BOT_V_HD_4 }
+};
+
+/*
+ * STI VTG synchronisation parameters structure
+ *
+ *@hsync: sample number falling and rising edge
+ *@vsync_line_top: vertical top field line number falling and rising edge
+ *@vsync_line_bot: vertical bottom field line number falling and rising edge
+ *@vsync_off_top: vertical top field sample number rising and falling edge
+ *@vsync_off_bot: vertical bottom field sample number rising and falling edge
+ */
+struct sti_vtg_sync_params {
+	u32 hsync;
+	u32 vsync_line_top;
+	u32 vsync_line_bot;
+	u32 vsync_off_top;
+	u32 vsync_off_bot;
+};
+
 /**
  * STI VTG structure
  *
  * @dev: pointer to device driver
- * @data: data associated to the device
+ * @np: device node
+ * @regs: register mapping
+ * @sync_params: synchronisation parameters used to generate timings
  * @irq: VTG irq
- * @type: VTG type (main or aux)
+ * @irq_status: store the IRQ status value
  * @notifier_list: notifier callback
  * @crtc: the CRTC for vblank event
  * @slave: slave vtg
@@ -87,6 +138,7 @@ struct sti_vtg {
 	struct device *dev;
 	struct device_node *np;
 	void __iomem *regs;
+	struct sti_vtg_sync_params sync_params[VTG_MAX_SYNC_OUTPUT];
 	int irq;
 	u32 irq_status;
 	struct raw_notifier_head notifier_list;
@@ -146,13 +198,69 @@ static void vtg_set_output_window(void __iomem *regs,
 	writel(video_bottom_field_stop, regs + VTG_VID_BFS);
 }
 
+static void vtg_set_hsync_vsync_pos(struct sti_vtg_sync_params *sync,
+				    int delay,
+				    const struct drm_display_mode *mode)
+{
+	long clocksperline, start, stop;
+	u32 risesync_top, fallsync_top;
+	u32 risesync_offs_top, fallsync_offs_top;
+
+	clocksperline = mode->htotal;
+
+	/* Get the hsync position */
+	start = 0;
+	stop = mode->hsync_end - mode->hsync_start;
+
+	start += delay;
+	stop  += delay;
+
+	if (start < 0)
+		start += clocksperline;
+	else if (start >= clocksperline)
+		start -= clocksperline;
+
+	if (stop < 0)
+		stop += clocksperline;
+	else if (stop >= clocksperline)
+		stop -= clocksperline;
+
+	sync->hsync = (stop << 16) | start;
+
+	/* Get the vsync position */
+	if (delay >= 0) {
+		risesync_top = 1;
+		fallsync_top = risesync_top;
+		fallsync_top += mode->vsync_end - mode->vsync_start;
+
+		fallsync_offs_top = (u32)delay;
+		risesync_offs_top = (u32)delay;
+	} else {
+		risesync_top = mode->vtotal;
+		fallsync_top = mode->vsync_end - mode->vsync_start;
+
+		fallsync_offs_top = clocksperline + delay;
+		risesync_offs_top = clocksperline + delay;
+	}
+
+	sync->vsync_line_top = (fallsync_top << 16) | risesync_top;
+	sync->vsync_off_top = (fallsync_offs_top << 16) | risesync_offs_top;
+
+	/* Only progressive supported for now */
+	sync->vsync_line_bot = sync->vsync_line_top;
+	sync->vsync_off_bot = sync->vsync_off_top;
+}
+
 static void vtg_set_mode(struct sti_vtg *vtg,
-			 int type, const struct drm_display_mode *mode)
+			 int type,
+			 struct sti_vtg_sync_params *sync,
+			 const struct drm_display_mode *mode)
 {
-	u32 tmp;
+	unsigned int i;
 
 	if (vtg->slave)
-		vtg_set_mode(vtg->slave, VTG_TYPE_SLAVE_BY_EXT0, mode);
+		vtg_set_mode(vtg->slave, VTG_MODE_SLAVE_BY_EXT0,
+			     vtg->sync_params, mode);
 
 	/* Set the number of clock cycles per line */
 	writel(mode->htotal, vtg->regs + VTG_CLKLN);
@@ -163,57 +271,31 @@ static void vtg_set_mode(struct sti_vtg *vtg,
 	/* Program output window */
 	vtg_set_output_window(vtg->regs, mode);
 
-	/* prepare VTG set 1 for HDMI */
-	tmp = (mode->hsync_end - mode->hsync_start + HDMI_DELAY) << 16;
-	tmp |= HDMI_DELAY;
-	writel(tmp, vtg->regs + VTG_H_HD_1);
-
-	tmp = (mode->vsync_end - mode->vsync_start + 1) << 16;
-	tmp |= 1;
-	writel(tmp, vtg->regs + VTG_TOP_V_VD_1);
-	writel(tmp, vtg->regs + VTG_BOT_V_VD_1);
-
-	tmp = HDMI_DELAY << 16;
-	tmp |= HDMI_DELAY;
-	writel(tmp, vtg->regs + VTG_TOP_V_HD_1);
-	writel(tmp, vtg->regs + VTG_BOT_V_HD_1);
-
-	/* prepare VTG set 2 for for HD DCS */
-	tmp = (mode->hsync_end - mode->hsync_start) << 16;
-	writel(tmp, vtg->regs + VTG_H_HD_2);
-
-	tmp = (mode->vsync_end - mode->vsync_start + 1) << 16;
-	tmp |= 1;
-	writel(tmp, vtg->regs + VTG_TOP_V_VD_2);
-	writel(tmp, vtg->regs + VTG_BOT_V_VD_2);
-	writel(0, vtg->regs + VTG_TOP_V_HD_2);
-	writel(0, vtg->regs + VTG_BOT_V_HD_2);
-
-	/* prepare VTG set 3 for HD Analog in HD mode */
-	tmp = (mode->hsync_end - mode->hsync_start + AWG_DELAY_HD) << 16;
-	tmp |= mode->htotal + AWG_DELAY_HD;
-	writel(tmp, vtg->regs + VTG_H_HD_3);
-
-	tmp = (mode->vsync_end - mode->vsync_start) << 16;
-	tmp |= mode->vtotal;
-	writel(tmp, vtg->regs + VTG_TOP_V_VD_3);
-	writel(tmp, vtg->regs + VTG_BOT_V_VD_3);
-
-	tmp = (mode->htotal + AWG_DELAY_HD) << 16;
-	tmp |= mode->htotal + AWG_DELAY_HD;
-	writel(tmp, vtg->regs + VTG_TOP_V_HD_3);
-	writel(tmp, vtg->regs + VTG_BOT_V_HD_3);
-
-	/* Prepare VTG set 4 for DVO */
-	tmp = (mode->hsync_end - mode->hsync_start) << 16;
-	writel(tmp, vtg->regs + VTG_H_HD_4);
-
-	tmp = (mode->vsync_end - mode->vsync_start + 1) << 16;
-	tmp |= 1;
-	writel(tmp, vtg->regs + VTG_TOP_V_VD_4);
-	writel(tmp, vtg->regs + VTG_BOT_V_VD_4);
-	writel(0, vtg->regs + VTG_TOP_V_HD_4);
-	writel(0, vtg->regs + VTG_BOT_V_HD_4);
+	/* Set hsync and vsync position for HDMI */
+	vtg_set_hsync_vsync_pos(&sync[VTG_SYNC_ID_HDMI - 1], HDMI_DELAY, mode);
+
+	/* Set hsync and vsync position for HD DCS */
+	vtg_set_hsync_vsync_pos(&sync[VTG_SYNC_ID_HDDCS - 1], 0, mode);
+
+	/* Set hsync and vsync position for HDF */
+	vtg_set_hsync_vsync_pos(&sync[VTG_SYNC_ID_HDF - 1], AWG_DELAY_HD, mode);
+
+	/* Set hsync and vsync position for DVO */
+	vtg_set_hsync_vsync_pos(&sync[VTG_SYNC_ID_DVO - 1], DVO_DELAY, mode);
+
+	/* Progam the syncs outputs */
+	for (i = 0; i < VTG_MAX_SYNC_OUTPUT ; i++) {
+		writel(sync[i].hsync,
+		       vtg->regs + vtg_regs_offs[i].h_hd);
+		writel(sync[i].vsync_line_top,
+		       vtg->regs + vtg_regs_offs[i].top_v_vd);
+		writel(sync[i].vsync_line_bot,
+		       vtg->regs + vtg_regs_offs[i].bot_v_vd);
+		writel(sync[i].vsync_off_top,
+		       vtg->regs + vtg_regs_offs[i].top_v_hd);
+		writel(sync[i].vsync_off_bot,
+		       vtg->regs + vtg_regs_offs[i].bot_v_hd);
+	}
 
 	/* mode */
 	writel(type, vtg->regs + VTG_MODE);
@@ -231,7 +313,7 @@ void sti_vtg_set_config(struct sti_vtg *vtg,
 		const struct drm_display_mode *mode)
 {
 	/* write configuration */
-	vtg_set_mode(vtg, VTG_TYPE_MASTER, mode);
+	vtg_set_mode(vtg, VTG_MODE_MASTER, vtg->sync_params, mode);
 
 	vtg_reset(vtg);
 
diff --git a/drivers/gpu/drm/sti/sti_vtg.h b/drivers/gpu/drm/sti/sti_vtg.h
index cd2439f89d05..f1dcdf9c2342 100644
--- a/drivers/gpu/drm/sti/sti_vtg.h
+++ b/drivers/gpu/drm/sti/sti_vtg.h
@@ -10,6 +10,11 @@
 #define VTG_TOP_FIELD_EVENT     1
 #define VTG_BOTTOM_FIELD_EVENT  2
 
+#define VTG_SYNC_ID_HDMI        1
+#define VTG_SYNC_ID_HDDCS       2
+#define VTG_SYNC_ID_HDF         3
+#define VTG_SYNC_ID_DVO         4
+
 struct sti_vtg;
 struct drm_display_mode;
 struct notifier_block;
diff --git a/drivers/gpu/drm/tilcdc/tilcdc_crtc.c b/drivers/gpu/drm/tilcdc/tilcdc_crtc.c
index 4802da8e6d6f..051e5e1b7ad6 100644
--- a/drivers/gpu/drm/tilcdc/tilcdc_crtc.c
+++ b/drivers/gpu/drm/tilcdc/tilcdc_crtc.c
@@ -21,25 +21,31 @@
 #include "tilcdc_drv.h"
 #include "tilcdc_regs.h"
 
+#define TILCDC_VBLANK_SAFETY_THRESHOLD_US 1000
+
 struct tilcdc_crtc {
 	struct drm_crtc base;
 
 	const struct tilcdc_panel_info *info;
-	uint32_t dirty;
-	dma_addr_t start, end;
 	struct drm_pending_vblank_event *event;
 	int dpms;
 	wait_queue_head_t frame_done_wq;
 	bool frame_done;
+	spinlock_t irq_lock;
 
-	/* fb currently set to scanout 0/1: */
-	struct drm_framebuffer *scanout[2];
+	ktime_t last_vblank;
+
+	struct drm_framebuffer *curr_fb;
+	struct drm_framebuffer *next_fb;
 
 	/* for deferred fb unref's: */
 	struct drm_flip_work unref_work;
 
 	/* Only set if an external encoder is connected */
 	bool simulate_vesa_sync;
+
+	int sync_lost_count;
+	bool frame_intact;
 };
 #define to_tilcdc_crtc(x) container_of(x, struct tilcdc_crtc, base)
 
@@ -54,79 +60,53 @@ static void unref_worker(struct drm_flip_work *work, void *val)
 	mutex_unlock(&dev->mode_config.mutex);
 }
 
-static void set_scanout(struct drm_crtc *crtc, int n)
+static void set_scanout(struct drm_crtc *crtc, struct drm_framebuffer *fb)
 {
-	static const uint32_t base_reg[] = {
-			LCDC_DMA_FB_BASE_ADDR_0_REG,
-			LCDC_DMA_FB_BASE_ADDR_1_REG,
-	};
-	static const uint32_t ceil_reg[] = {
-			LCDC_DMA_FB_CEILING_ADDR_0_REG,
-			LCDC_DMA_FB_CEILING_ADDR_1_REG,
-	};
-	static const uint32_t stat[] = {
-			LCDC_END_OF_FRAME0, LCDC_END_OF_FRAME1,
-	};
 	struct tilcdc_crtc *tilcdc_crtc = to_tilcdc_crtc(crtc);
 	struct drm_device *dev = crtc->dev;
-	struct tilcdc_drm_private *priv = dev->dev_private;
-
-	pm_runtime_get_sync(dev->dev);
-	tilcdc_write(dev, base_reg[n], tilcdc_crtc->start);
-	tilcdc_write(dev, ceil_reg[n], tilcdc_crtc->end);
-	if (tilcdc_crtc->scanout[n]) {
-		drm_flip_work_queue(&tilcdc_crtc->unref_work, tilcdc_crtc->scanout[n]);
-		drm_flip_work_commit(&tilcdc_crtc->unref_work, priv->wq);
-	}
-	tilcdc_crtc->scanout[n] = crtc->primary->fb;
-	drm_framebuffer_reference(tilcdc_crtc->scanout[n]);
-	tilcdc_crtc->dirty &= ~stat[n];
-	pm_runtime_put_sync(dev->dev);
-}
-
-static void update_scanout(struct drm_crtc *crtc)
-{
-	struct tilcdc_crtc *tilcdc_crtc = to_tilcdc_crtc(crtc);
-	struct drm_device *dev = crtc->dev;
-	struct drm_framebuffer *fb = crtc->primary->fb;
 	struct drm_gem_cma_object *gem;
 	unsigned int depth, bpp;
+	dma_addr_t start, end;
 
 	drm_fb_get_bpp_depth(fb->pixel_format, &depth, &bpp);
 	gem = drm_fb_cma_get_gem_obj(fb, 0);
 
-	tilcdc_crtc->start = gem->paddr + fb->offsets[0] +
-			(crtc->y * fb->pitches[0]) + (crtc->x * bpp/8);
+	start = gem->paddr + fb->offsets[0] +
+		crtc->y * fb->pitches[0] +
+		crtc->x * bpp / 8;
 
-	tilcdc_crtc->end = tilcdc_crtc->start +
-			(crtc->mode.vdisplay * fb->pitches[0]);
+	end = start + (crtc->mode.vdisplay * fb->pitches[0]);
 
-	if (tilcdc_crtc->dpms == DRM_MODE_DPMS_ON) {
-		/* already enabled, so just mark the frames that need
-		 * updating and they will be updated on vblank:
-		 */
-		tilcdc_crtc->dirty |= LCDC_END_OF_FRAME0 | LCDC_END_OF_FRAME1;
-		drm_vblank_get(dev, 0);
-	} else {
-		/* not enabled yet, so update registers immediately: */
-		set_scanout(crtc, 0);
-		set_scanout(crtc, 1);
-	}
+	tilcdc_write(dev, LCDC_DMA_FB_BASE_ADDR_0_REG, start);
+	tilcdc_write(dev, LCDC_DMA_FB_CEILING_ADDR_0_REG, end);
+
+	if (tilcdc_crtc->curr_fb)
+		drm_flip_work_queue(&tilcdc_crtc->unref_work,
+			tilcdc_crtc->curr_fb);
+
+	tilcdc_crtc->curr_fb = fb;
 }
 
-static void start(struct drm_crtc *crtc)
+static void reset(struct drm_crtc *crtc)
 {
 	struct drm_device *dev = crtc->dev;
 	struct tilcdc_drm_private *priv = dev->dev_private;
 
-	if (priv->rev == 2) {
-		tilcdc_set(dev, LCDC_CLK_RESET_REG, LCDC_CLK_MAIN_RESET);
-		msleep(1);
-		tilcdc_clear(dev, LCDC_CLK_RESET_REG, LCDC_CLK_MAIN_RESET);
-		msleep(1);
-	}
+	if (priv->rev != 2)
+		return;
+
+	tilcdc_set(dev, LCDC_CLK_RESET_REG, LCDC_CLK_MAIN_RESET);
+	usleep_range(250, 1000);
+	tilcdc_clear(dev, LCDC_CLK_RESET_REG, LCDC_CLK_MAIN_RESET);
+}
 
-	tilcdc_set(dev, LCDC_DMA_CTRL_REG, LCDC_DUAL_FRAME_BUFFER_ENABLE);
+static void start(struct drm_crtc *crtc)
+{
+	struct drm_device *dev = crtc->dev;
+
+	reset(crtc);
+
+	tilcdc_clear(dev, LCDC_DMA_CTRL_REG, LCDC_DUAL_FRAME_BUFFER_ENABLE);
 	tilcdc_set(dev, LCDC_RASTER_CTRL_REG, LCDC_PALETTE_LOAD_MODE(DATA_ONLY));
 	tilcdc_set(dev, LCDC_RASTER_CTRL_REG, LCDC_RASTER_ENABLE);
 }
@@ -138,17 +118,31 @@ static void stop(struct drm_crtc *crtc)
 	tilcdc_clear(dev, LCDC_RASTER_CTRL_REG, LCDC_RASTER_ENABLE);
 }
 
-static void tilcdc_crtc_dpms(struct drm_crtc *crtc, int mode);
 static void tilcdc_crtc_destroy(struct drm_crtc *crtc)
 {
 	struct tilcdc_crtc *tilcdc_crtc = to_tilcdc_crtc(crtc);
 
 	tilcdc_crtc_dpms(crtc, DRM_MODE_DPMS_OFF);
 
+	of_node_put(crtc->port);
 	drm_crtc_cleanup(crtc);
 	drm_flip_work_cleanup(&tilcdc_crtc->unref_work);
+}
+
+static int tilcdc_verify_fb(struct drm_crtc *crtc, struct drm_framebuffer *fb)
+{
+	struct drm_device *dev = crtc->dev;
+	unsigned int depth, bpp;
+
+	drm_fb_get_bpp_depth(fb->pixel_format, &depth, &bpp);
+
+	if (fb->pitches[0] != crtc->mode.hdisplay * bpp / 8) {
+		dev_err(dev->dev,
+			"Invalid pitch: fb and crtc widths must be the same");
+		return -EINVAL;
+	}
 
-	kfree(tilcdc_crtc);
+	return 0;
 }
 
 static int tilcdc_crtc_page_flip(struct drm_crtc *crtc,
@@ -158,20 +152,48 @@ static int tilcdc_crtc_page_flip(struct drm_crtc *crtc,
 {
 	struct tilcdc_crtc *tilcdc_crtc = to_tilcdc_crtc(crtc);
 	struct drm_device *dev = crtc->dev;
+	int r;
+	unsigned long flags;
+	s64 tdiff;
+	ktime_t next_vblank;
+
+	r = tilcdc_verify_fb(crtc, fb);
+	if (r)
+		return r;
 
 	if (tilcdc_crtc->event) {
 		dev_err(dev->dev, "already pending page flip!\n");
 		return -EBUSY;
 	}
 
+	drm_framebuffer_reference(fb);
+
 	crtc->primary->fb = fb;
+
+	pm_runtime_get_sync(dev->dev);
+
+	spin_lock_irqsave(&tilcdc_crtc->irq_lock, flags);
+
+	next_vblank = ktime_add_us(tilcdc_crtc->last_vblank,
+		1000000 / crtc->hwmode.vrefresh);
+
+	tdiff = ktime_to_us(ktime_sub(next_vblank, ktime_get()));
+
+	if (tdiff >= TILCDC_VBLANK_SAFETY_THRESHOLD_US)
+		set_scanout(crtc, fb);
+	else
+		tilcdc_crtc->next_fb = fb;
+
 	tilcdc_crtc->event = event;
-	update_scanout(crtc);
+
+	spin_unlock_irqrestore(&tilcdc_crtc->irq_lock, flags);
+
+	pm_runtime_put_sync(dev->dev);
 
 	return 0;
 }
 
-static void tilcdc_crtc_dpms(struct drm_crtc *crtc, int mode)
+void tilcdc_crtc_dpms(struct drm_crtc *crtc, int mode)
 {
 	struct tilcdc_crtc *tilcdc_crtc = to_tilcdc_crtc(crtc);
 	struct drm_device *dev = crtc->dev;
@@ -186,10 +208,8 @@ static void tilcdc_crtc_dpms(struct drm_crtc *crtc, int mode)
 
 	tilcdc_crtc->dpms = mode;
 
-	pm_runtime_get_sync(dev->dev);
-
 	if (mode == DRM_MODE_DPMS_ON) {
-		pm_runtime_forbid(dev->dev);
+		pm_runtime_get_sync(dev->dev);
 		start(crtc);
 	} else {
 		tilcdc_crtc->frame_done = false;
@@ -207,10 +227,23 @@ static void tilcdc_crtc_dpms(struct drm_crtc *crtc, int mode)
 			if (ret == 0)
 				dev_err(dev->dev, "timeout waiting for framedone\n");
 		}
-		pm_runtime_allow(dev->dev);
-	}
 
-	pm_runtime_put_sync(dev->dev);
+		pm_runtime_put_sync(dev->dev);
+
+		if (tilcdc_crtc->next_fb) {
+			drm_flip_work_queue(&tilcdc_crtc->unref_work,
+					    tilcdc_crtc->next_fb);
+			tilcdc_crtc->next_fb = NULL;
+		}
+
+		if (tilcdc_crtc->curr_fb) {
+			drm_flip_work_queue(&tilcdc_crtc->unref_work,
+					    tilcdc_crtc->curr_fb);
+			tilcdc_crtc->curr_fb = NULL;
+		}
+
+		drm_flip_work_commit(&tilcdc_crtc->unref_work, priv->wq);
+	}
 }
 
 static bool tilcdc_crtc_mode_fixup(struct drm_crtc *crtc,
@@ -272,6 +305,10 @@ static int tilcdc_crtc_mode_set(struct drm_crtc *crtc,
 	if (WARN_ON(!info))
 		return -EINVAL;
 
+	ret = tilcdc_verify_fb(crtc, crtc->primary->fb);
+	if (ret)
+		return ret;
+
 	pm_runtime_get_sync(dev->dev);
 
 	/* Configure the Burst Size and fifo threshold of DMA: */
@@ -419,8 +456,10 @@ static int tilcdc_crtc_mode_set(struct drm_crtc *crtc,
 	else
 		tilcdc_clear(dev, LCDC_RASTER_CTRL_REG, LCDC_RASTER_ORDER);
 
+	drm_framebuffer_reference(crtc->primary->fb);
+
+	set_scanout(crtc, crtc->primary->fb);
 
-	update_scanout(crtc);
 	tilcdc_crtc_update_clk(crtc);
 
 	pm_runtime_put_sync(dev->dev);
@@ -431,7 +470,21 @@ static int tilcdc_crtc_mode_set(struct drm_crtc *crtc,
 static int tilcdc_crtc_mode_set_base(struct drm_crtc *crtc, int x, int y,
 		struct drm_framebuffer *old_fb)
 {
-	update_scanout(crtc);
+	struct drm_device *dev = crtc->dev;
+	int r;
+
+	r = tilcdc_verify_fb(crtc, crtc->primary->fb);
+	if (r)
+		return r;
+
+	drm_framebuffer_reference(crtc->primary->fb);
+
+	pm_runtime_get_sync(dev->dev);
+
+	set_scanout(crtc, crtc->primary->fb);
+
+	pm_runtime_put_sync(dev->dev);
+
 	return 0;
 }
 
@@ -573,7 +626,8 @@ void tilcdc_crtc_update_clk(struct drm_crtc *crtc)
 	struct drm_device *dev = crtc->dev;
 	struct tilcdc_drm_private *priv = dev->dev_private;
 	int dpms = tilcdc_crtc->dpms;
-	unsigned int lcd_clk, div;
+	unsigned long lcd_clk;
+	const unsigned clkdiv = 2; /* using a fixed divider of 2 */
 	int ret;
 
 	pm_runtime_get_sync(dev->dev);
@@ -581,22 +635,21 @@ void tilcdc_crtc_update_clk(struct drm_crtc *crtc)
 	if (dpms == DRM_MODE_DPMS_ON)
 		tilcdc_crtc_dpms(crtc, DRM_MODE_DPMS_OFF);
 
-	/* in raster mode, minimum divisor is 2: */
-	ret = clk_set_rate(priv->disp_clk, crtc->mode.clock * 1000 * 2);
-	if (ret) {
+	/* mode.clock is in KHz, set_rate wants parameter in Hz */
+	ret = clk_set_rate(priv->clk, crtc->mode.clock * 1000 * clkdiv);
+	if (ret < 0) {
 		dev_err(dev->dev, "failed to set display clock rate to: %d\n",
 				crtc->mode.clock);
 		goto out;
 	}
 
 	lcd_clk = clk_get_rate(priv->clk);
-	div = lcd_clk / (crtc->mode.clock * 1000);
 
-	DBG("lcd_clk=%u, mode clock=%d, div=%u", lcd_clk, crtc->mode.clock, div);
-	DBG("fck=%lu, dpll_disp_ck=%lu", clk_get_rate(priv->clk), clk_get_rate(priv->disp_clk));
+	DBG("lcd_clk=%lu, mode clock=%d, div=%u",
+		lcd_clk, crtc->mode.clock, clkdiv);
 
 	/* Configure the LCD clock divisor. */
-	tilcdc_write(dev, LCDC_CTRL_REG, LCDC_CLK_DIVISOR(div) |
+	tilcdc_write(dev, LCDC_CTRL_REG, LCDC_CLK_DIVISOR(clkdiv) |
 			LCDC_RASTER_MODE);
 
 	if (priv->rev == 2)
@@ -611,44 +664,58 @@ out:
 	pm_runtime_put_sync(dev->dev);
 }
 
+#define SYNC_LOST_COUNT_LIMIT 50
+
 irqreturn_t tilcdc_crtc_irq(struct drm_crtc *crtc)
 {
 	struct tilcdc_crtc *tilcdc_crtc = to_tilcdc_crtc(crtc);
 	struct drm_device *dev = crtc->dev;
 	struct tilcdc_drm_private *priv = dev->dev_private;
-	uint32_t stat = tilcdc_read_irqstatus(dev);
+	uint32_t stat;
 
-	if ((stat & LCDC_SYNC_LOST) && (stat & LCDC_FIFO_UNDERFLOW)) {
-		stop(crtc);
-		dev_err(dev->dev, "error: %08x\n", stat);
-		tilcdc_clear_irqstatus(dev, stat);
-		start(crtc);
-	} else if (stat & LCDC_PL_LOAD_DONE) {
-		tilcdc_clear_irqstatus(dev, stat);
-	} else {
-		struct drm_pending_vblank_event *event;
+	stat = tilcdc_read_irqstatus(dev);
+	tilcdc_clear_irqstatus(dev, stat);
+
+	if (stat & LCDC_END_OF_FRAME0) {
 		unsigned long flags;
-		uint32_t dirty = tilcdc_crtc->dirty & stat;
+		bool skip_event = false;
+		ktime_t now;
+
+		now = ktime_get();
 
-		tilcdc_clear_irqstatus(dev, stat);
+		drm_flip_work_commit(&tilcdc_crtc->unref_work, priv->wq);
 
-		if (dirty & LCDC_END_OF_FRAME0)
-			set_scanout(crtc, 0);
+		spin_lock_irqsave(&tilcdc_crtc->irq_lock, flags);
 
-		if (dirty & LCDC_END_OF_FRAME1)
-			set_scanout(crtc, 1);
+		tilcdc_crtc->last_vblank = now;
+
+		if (tilcdc_crtc->next_fb) {
+			set_scanout(crtc, tilcdc_crtc->next_fb);
+			tilcdc_crtc->next_fb = NULL;
+			skip_event = true;
+		}
+
+		spin_unlock_irqrestore(&tilcdc_crtc->irq_lock, flags);
 
 		drm_handle_vblank(dev, 0);
 
-		spin_lock_irqsave(&dev->event_lock, flags);
-		event = tilcdc_crtc->event;
-		tilcdc_crtc->event = NULL;
-		if (event)
-			drm_send_vblank_event(dev, 0, event);
-		spin_unlock_irqrestore(&dev->event_lock, flags);
+		if (!skip_event) {
+			struct drm_pending_vblank_event *event;
+
+			spin_lock_irqsave(&dev->event_lock, flags);
+
+			event = tilcdc_crtc->event;
+			tilcdc_crtc->event = NULL;
+			if (event)
+				drm_send_vblank_event(dev, 0, event);
+
+			spin_unlock_irqrestore(&dev->event_lock, flags);
+		}
 
-		if (dirty && !tilcdc_crtc->dirty)
-			drm_vblank_put(dev, 0);
+		if (tilcdc_crtc->frame_intact)
+			tilcdc_crtc->sync_lost_count = 0;
+		else
+			tilcdc_crtc->frame_intact = true;
 	}
 
 	if (priv->rev == 2) {
@@ -659,16 +726,34 @@ irqreturn_t tilcdc_crtc_irq(struct drm_crtc *crtc)
 		tilcdc_write(dev, LCDC_END_OF_INT_IND_REG, 0);
 	}
 
+	if (stat & LCDC_SYNC_LOST) {
+		dev_err_ratelimited(dev->dev, "%s(0x%08x): Sync lost",
+				    __func__, stat);
+		tilcdc_crtc->frame_intact = false;
+		if (tilcdc_crtc->sync_lost_count++ > SYNC_LOST_COUNT_LIMIT) {
+			dev_err(dev->dev,
+				"%s(0x%08x): Sync lost flood detected, disabling the interrupt",
+				__func__, stat);
+			tilcdc_write(dev, LCDC_INT_ENABLE_CLR_REG,
+				     LCDC_SYNC_LOST);
+		}
+	}
+
+	if (stat & LCDC_FIFO_UNDERFLOW)
+		dev_err_ratelimited(dev->dev, "%s(0x%08x): FIFO underfow",
+				    __func__, stat);
+
 	return IRQ_HANDLED;
 }
 
 struct drm_crtc *tilcdc_crtc_create(struct drm_device *dev)
 {
+	struct tilcdc_drm_private *priv = dev->dev_private;
 	struct tilcdc_crtc *tilcdc_crtc;
 	struct drm_crtc *crtc;
 	int ret;
 
-	tilcdc_crtc = kzalloc(sizeof(*tilcdc_crtc), GFP_KERNEL);
+	tilcdc_crtc = devm_kzalloc(dev->dev, sizeof(*tilcdc_crtc), GFP_KERNEL);
 	if (!tilcdc_crtc) {
 		dev_err(dev->dev, "allocation failed\n");
 		return NULL;
@@ -682,12 +767,32 @@ struct drm_crtc *tilcdc_crtc_create(struct drm_device *dev)
 	drm_flip_work_init(&tilcdc_crtc->unref_work,
 			"unref", unref_worker);
 
+	spin_lock_init(&tilcdc_crtc->irq_lock);
+
 	ret = drm_crtc_init(dev, crtc, &tilcdc_crtc_funcs);
 	if (ret < 0)
 		goto fail;
 
 	drm_crtc_helper_add(crtc, &tilcdc_crtc_helper_funcs);
 
+	if (priv->is_componentized) {
+		struct device_node *ports =
+			of_get_child_by_name(dev->dev->of_node, "ports");
+
+		if (ports) {
+			crtc->port = of_get_child_by_name(ports, "port");
+			of_node_put(ports);
+		} else {
+			crtc->port =
+				of_get_child_by_name(dev->dev->of_node, "port");
+		}
+		if (!crtc->port) { /* This should never happen */
+			dev_err(dev->dev, "Port node not found in %s\n",
+				dev->dev->of_node->full_name);
+			goto fail;
+		}
+	}
+
 	return crtc;
 
 fail:
diff --git a/drivers/gpu/drm/tilcdc/tilcdc_drv.c b/drivers/gpu/drm/tilcdc/tilcdc_drv.c
index 8190ac3b1b32..709bc903524d 100644
--- a/drivers/gpu/drm/tilcdc/tilcdc_drv.c
+++ b/drivers/gpu/drm/tilcdc/tilcdc_drv.c
@@ -18,6 +18,8 @@
 /* LCDC DRM driver, based on da8xx-fb */
 
 #include <linux/component.h>
+#include <linux/pinctrl/consumer.h>
+#include <linux/suspend.h>
 
 #include "tilcdc_drv.h"
 #include "tilcdc_regs.h"
@@ -110,6 +112,8 @@ static int tilcdc_unload(struct drm_device *dev)
 {
 	struct tilcdc_drm_private *priv = dev->dev_private;
 
+	tilcdc_crtc_dpms(priv->crtc, DRM_MODE_DPMS_OFF);
+
 	tilcdc_remove_external_encoders(dev);
 
 	drm_fbdev_cma_fini(priv->fbdev);
@@ -139,11 +143,11 @@ static int tilcdc_unload(struct drm_device *dev)
 
 	pm_runtime_disable(dev->dev);
 
-	kfree(priv);
-
 	return 0;
 }
 
+static size_t tilcdc_num_regs(void);
+
 static int tilcdc_load(struct drm_device *dev, unsigned long flags)
 {
 	struct platform_device *pdev = dev->platformdev;
@@ -154,8 +158,12 @@ static int tilcdc_load(struct drm_device *dev, unsigned long flags)
 	u32 bpp = 0;
 	int ret;
 
-	priv = kzalloc(sizeof(*priv), GFP_KERNEL);
-	if (!priv) {
+	priv = devm_kzalloc(dev->dev, sizeof(*priv), GFP_KERNEL);
+	if (priv)
+		priv->saved_register =
+			devm_kcalloc(dev->dev, tilcdc_num_regs(),
+				     sizeof(*priv->saved_register), GFP_KERNEL);
+	if (!priv || !priv->saved_register) {
 		dev_err(dev->dev, "failed to allocate private data\n");
 		return -ENOMEM;
 	}
@@ -168,7 +176,7 @@ static int tilcdc_load(struct drm_device *dev, unsigned long flags)
 	priv->wq = alloc_ordered_workqueue("tilcdc", 0);
 	if (!priv->wq) {
 		ret = -ENOMEM;
-		goto fail_free_priv;
+		goto fail_unset_priv;
 	}
 
 	res = platform_get_resource(pdev, IORESOURCE_MEM, 0);
@@ -192,13 +200,6 @@ static int tilcdc_load(struct drm_device *dev, unsigned long flags)
 		goto fail_iounmap;
 	}
 
-	priv->disp_clk = clk_get(dev->dev, "dpll_disp_ck");
-	if (IS_ERR(priv->clk)) {
-		dev_err(dev->dev, "failed to get display clock\n");
-		ret = -ENODEV;
-		goto fail_put_clk;
-	}
-
 #ifdef CONFIG_CPU_FREQ
 	priv->lcd_fck_rate = clk_get_rate(priv->clk);
 	priv->freq_transition.notifier_call = cpufreq_transition;
@@ -206,7 +207,7 @@ static int tilcdc_load(struct drm_device *dev, unsigned long flags)
 			CPUFREQ_TRANSITION_NOTIFIER);
 	if (ret) {
 		dev_err(dev->dev, "failed to register cpufreq notifier\n");
-		goto fail_put_disp_clk;
+		goto fail_put_clk;
 	}
 #endif
 
@@ -227,7 +228,6 @@ static int tilcdc_load(struct drm_device *dev, unsigned long flags)
 	DBG("Maximum Pixel Clock Value %dKHz", priv->max_pixelclock);
 
 	pm_runtime_enable(dev->dev);
-	pm_runtime_irq_safe(dev->dev);
 
 	/* Determine LCD IP Version */
 	pm_runtime_get_sync(dev->dev);
@@ -330,11 +330,9 @@ fail_cpufreq_unregister:
 #ifdef CONFIG_CPU_FREQ
 	cpufreq_unregister_notifier(&priv->freq_transition,
 			CPUFREQ_TRANSITION_NOTIFIER);
-fail_put_disp_clk:
-	clk_put(priv->disp_clk);
-#endif
 
 fail_put_clk:
+#endif
 	clk_put(priv->clk);
 
 fail_iounmap:
@@ -344,9 +342,9 @@ fail_free_wq:
 	flush_workqueue(priv->wq);
 	destroy_workqueue(priv->wq);
 
-fail_free_priv:
+fail_unset_priv:
 	dev->dev_private = NULL;
-	kfree(priv);
+
 	return ret;
 }
 
@@ -373,10 +371,14 @@ static int tilcdc_irq_postinstall(struct drm_device *dev)
 	struct tilcdc_drm_private *priv = dev->dev_private;
 
 	/* enable FIFO underflow irq: */
-	if (priv->rev == 1)
+	if (priv->rev == 1) {
 		tilcdc_set(dev, LCDC_RASTER_CTRL_REG, LCDC_V1_UNDERFLOW_INT_ENA);
-	else
-		tilcdc_set(dev, LCDC_INT_ENABLE_SET_REG, LCDC_V2_UNDERFLOW_INT_ENA);
+	} else {
+		tilcdc_write(dev, LCDC_INT_ENABLE_SET_REG,
+			   LCDC_V2_UNDERFLOW_INT_ENA |
+			   LCDC_V2_END_OF_FRAME0_INT_ENA |
+			   LCDC_FRAME_DONE | LCDC_SYNC_LOST);
+	}
 
 	return 0;
 }
@@ -391,43 +393,21 @@ static void tilcdc_irq_uninstall(struct drm_device *dev)
 				LCDC_V1_UNDERFLOW_INT_ENA | LCDC_V1_PL_INT_ENA);
 		tilcdc_clear(dev, LCDC_DMA_CTRL_REG, LCDC_V1_END_OF_FRAME_INT_ENA);
 	} else {
-		tilcdc_clear(dev, LCDC_INT_ENABLE_SET_REG,
+		tilcdc_write(dev, LCDC_INT_ENABLE_CLR_REG,
 			LCDC_V2_UNDERFLOW_INT_ENA | LCDC_V2_PL_INT_ENA |
-			LCDC_V2_END_OF_FRAME0_INT_ENA | LCDC_V2_END_OF_FRAME1_INT_ENA |
-			LCDC_FRAME_DONE);
-	}
-
-}
-
-static void enable_vblank(struct drm_device *dev, bool enable)
-{
-	struct tilcdc_drm_private *priv = dev->dev_private;
-	u32 reg, mask;
-
-	if (priv->rev == 1) {
-		reg = LCDC_DMA_CTRL_REG;
-		mask = LCDC_V1_END_OF_FRAME_INT_ENA;
-	} else {
-		reg = LCDC_INT_ENABLE_SET_REG;
-		mask = LCDC_V2_END_OF_FRAME0_INT_ENA |
-			LCDC_V2_END_OF_FRAME1_INT_ENA | LCDC_FRAME_DONE;
+			LCDC_V2_END_OF_FRAME0_INT_ENA |
+			LCDC_FRAME_DONE | LCDC_SYNC_LOST);
 	}
-
-	if (enable)
-		tilcdc_set(dev, reg, mask);
-	else
-		tilcdc_clear(dev, reg, mask);
 }
 
 static int tilcdc_enable_vblank(struct drm_device *dev, unsigned int pipe)
 {
-	enable_vblank(dev, true);
 	return 0;
 }
 
 static void tilcdc_disable_vblank(struct drm_device *dev, unsigned int pipe)
 {
-	enable_vblank(dev, false);
+	return;
 }
 
 #if defined(CONFIG_DEBUG_FS) || defined(CONFIG_PM_SLEEP)
@@ -454,13 +434,22 @@ static const struct {
 		/* new in revision 2: */
 		REG(2, false, LCDC_RAW_STAT_REG),
 		REG(2, false, LCDC_MASKED_STAT_REG),
-		REG(2, false, LCDC_INT_ENABLE_SET_REG),
+		REG(2, true, LCDC_INT_ENABLE_SET_REG),
 		REG(2, false, LCDC_INT_ENABLE_CLR_REG),
 		REG(2, false, LCDC_END_OF_INT_IND_REG),
 		REG(2, true,  LCDC_CLK_ENABLE_REG),
-		REG(2, true,  LCDC_INT_ENABLE_SET_REG),
 #undef REG
 };
+
+static size_t tilcdc_num_regs(void)
+{
+	return ARRAY_SIZE(registers);
+}
+#else
+static size_t tilcdc_num_regs(void)
+{
+	return 0;
+}
 #endif
 
 #ifdef CONFIG_DEBUG_FS
@@ -547,7 +536,8 @@ static const struct file_operations fops = {
 };
 
 static struct drm_driver tilcdc_driver = {
-	.driver_features    = DRIVER_HAVE_IRQ | DRIVER_GEM | DRIVER_MODESET,
+	.driver_features    = (DRIVER_HAVE_IRQ | DRIVER_GEM | DRIVER_MODESET |
+			       DRIVER_PRIME),
 	.load               = tilcdc_load,
 	.unload             = tilcdc_unload,
 	.lastclose          = tilcdc_lastclose,
@@ -564,6 +554,16 @@ static struct drm_driver tilcdc_driver = {
 	.dumb_create        = drm_gem_cma_dumb_create,
 	.dumb_map_offset    = drm_gem_cma_dumb_map_offset,
 	.dumb_destroy       = drm_gem_dumb_destroy,
+
+	.prime_handle_to_fd	= drm_gem_prime_handle_to_fd,
+	.prime_fd_to_handle	= drm_gem_prime_fd_to_handle,
+	.gem_prime_import	= drm_gem_prime_import,
+	.gem_prime_export	= drm_gem_prime_export,
+	.gem_prime_get_sg_table	= drm_gem_cma_prime_get_sg_table,
+	.gem_prime_import_sg_table = drm_gem_cma_prime_import_sg_table,
+	.gem_prime_vmap		= drm_gem_cma_prime_vmap,
+	.gem_prime_vunmap	= drm_gem_cma_prime_vunmap,
+	.gem_prime_mmap		= drm_gem_cma_prime_mmap,
 #ifdef CONFIG_DEBUG_FS
 	.debugfs_init       = tilcdc_debugfs_init,
 	.debugfs_cleanup    = tilcdc_debugfs_cleanup,
@@ -589,11 +589,24 @@ static int tilcdc_pm_suspend(struct device *dev)
 
 	drm_kms_helper_poll_disable(ddev);
 
+	/* Select sleep pin state */
+	pinctrl_pm_select_sleep_state(dev);
+
+	if (pm_runtime_suspended(dev)) {
+		priv->ctx_valid = false;
+		return 0;
+	}
+
+	/* Disable the LCDC controller, to avoid locking up the PRCM */
+	tilcdc_crtc_dpms(priv->crtc, DRM_MODE_DPMS_OFF);
+
 	/* Save register state: */
 	for (i = 0; i < ARRAY_SIZE(registers); i++)
 		if (registers[i].save && (priv->rev >= registers[i].rev))
 			priv->saved_register[n++] = tilcdc_read(ddev, registers[i].reg);
 
+	priv->ctx_valid = true;
+
 	return 0;
 }
 
@@ -603,10 +616,17 @@ static int tilcdc_pm_resume(struct device *dev)
 	struct tilcdc_drm_private *priv = ddev->dev_private;
 	unsigned i, n = 0;
 
-	/* Restore register state: */
-	for (i = 0; i < ARRAY_SIZE(registers); i++)
-		if (registers[i].save && (priv->rev >= registers[i].rev))
-			tilcdc_write(ddev, registers[i].reg, priv->saved_register[n++]);
+	/* Select default pin state */
+	pinctrl_pm_select_default_state(dev);
+
+	if (priv->ctx_valid == true) {
+		/* Restore register state: */
+		for (i = 0; i < ARRAY_SIZE(registers); i++)
+			if (registers[i].save &&
+			    (priv->rev >= registers[i].rev))
+				tilcdc_write(ddev, registers[i].reg,
+					     priv->saved_register[n++]);
+	}
 
 	drm_kms_helper_poll_enable(ddev);
 
diff --git a/drivers/gpu/drm/tilcdc/tilcdc_drv.h b/drivers/gpu/drm/tilcdc/tilcdc_drv.h
index 66105d8dc620..c1de18bae415 100644
--- a/drivers/gpu/drm/tilcdc/tilcdc_drv.h
+++ b/drivers/gpu/drm/tilcdc/tilcdc_drv.h
@@ -49,7 +49,6 @@
 struct tilcdc_drm_private {
 	void __iomem *mmio;
 
-	struct clk *disp_clk;    /* display dpll */
 	struct clk *clk;         /* functional clock */
 	int rev;                 /* IP revision */
 
@@ -67,7 +66,8 @@ struct tilcdc_drm_private {
 	uint32_t max_width;
 
 	/* register contents saved across suspend/resume: */
-	u32 saved_register[12];
+	u32 *saved_register;
+	bool ctx_valid;
 
 #ifdef CONFIG_CPU_FREQ
 	struct notifier_block freq_transition;
@@ -171,5 +171,6 @@ void tilcdc_crtc_set_simulate_vesa_sync(struct drm_crtc *crtc,
 					bool simulate_vesa_sync);
 int tilcdc_crtc_mode_valid(struct drm_crtc *crtc, struct drm_display_mode *mode);
 int tilcdc_crtc_max_width(struct drm_crtc *crtc);
+void tilcdc_crtc_dpms(struct drm_crtc *crtc, int mode);
 
 #endif /* __TILCDC_DRV_H__ */
diff --git a/drivers/gpu/drm/tilcdc/tilcdc_panel.c b/drivers/gpu/drm/tilcdc/tilcdc_panel.c
index 8dcf02a79b23..ff7774c17d7c 100644
--- a/drivers/gpu/drm/tilcdc/tilcdc_panel.c
+++ b/drivers/gpu/drm/tilcdc/tilcdc_panel.c
@@ -45,14 +45,6 @@ struct panel_encoder {
 };
 #define to_panel_encoder(x) container_of(x, struct panel_encoder, base)
 
-
-static void panel_encoder_destroy(struct drm_encoder *encoder)
-{
-	struct panel_encoder *panel_encoder = to_panel_encoder(encoder);
-	drm_encoder_cleanup(encoder);
-	kfree(panel_encoder);
-}
-
 static void panel_encoder_dpms(struct drm_encoder *encoder, int mode)
 {
 	struct panel_encoder *panel_encoder = to_panel_encoder(encoder);
@@ -90,7 +82,7 @@ static void panel_encoder_mode_set(struct drm_encoder *encoder,
 }
 
 static const struct drm_encoder_funcs panel_encoder_funcs = {
-		.destroy        = panel_encoder_destroy,
+		.destroy        = drm_encoder_cleanup,
 };
 
 static const struct drm_encoder_helper_funcs panel_encoder_helper_funcs = {
@@ -107,7 +99,8 @@ static struct drm_encoder *panel_encoder_create(struct drm_device *dev,
 	struct drm_encoder *encoder;
 	int ret;
 
-	panel_encoder = kzalloc(sizeof(*panel_encoder), GFP_KERNEL);
+	panel_encoder = devm_kzalloc(dev->dev, sizeof(*panel_encoder),
+				     GFP_KERNEL);
 	if (!panel_encoder) {
 		dev_err(dev->dev, "allocation failed\n");
 		return NULL;
@@ -128,7 +121,7 @@ static struct drm_encoder *panel_encoder_create(struct drm_device *dev,
 	return encoder;
 
 fail:
-	panel_encoder_destroy(encoder);
+	drm_encoder_cleanup(encoder);
 	return NULL;
 }
 
@@ -147,10 +140,8 @@ struct panel_connector {
 
 static void panel_connector_destroy(struct drm_connector *connector)
 {
-	struct panel_connector *panel_connector = to_panel_connector(connector);
 	drm_connector_unregister(connector);
 	drm_connector_cleanup(connector);
-	kfree(panel_connector);
 }
 
 static enum drm_connector_status panel_connector_detect(
@@ -223,7 +214,8 @@ static struct drm_connector *panel_connector_create(struct drm_device *dev,
 	struct drm_connector *connector;
 	int ret;
 
-	panel_connector = kzalloc(sizeof(*panel_connector), GFP_KERNEL);
+	panel_connector = devm_kzalloc(dev->dev, sizeof(*panel_connector),
+				       GFP_KERNEL);
 	if (!panel_connector) {
 		dev_err(dev->dev, "allocation failed\n");
 		return NULL;
diff --git a/drivers/gpu/drm/tilcdc/tilcdc_tfp410.c b/drivers/gpu/drm/tilcdc/tilcdc_tfp410.c
index 1c230172b402..7716f42f8aab 100644
--- a/drivers/gpu/drm/tilcdc/tilcdc_tfp410.c
+++ b/drivers/gpu/drm/tilcdc/tilcdc_tfp410.c
@@ -54,14 +54,6 @@ struct tfp410_encoder {
 };
 #define to_tfp410_encoder(x) container_of(x, struct tfp410_encoder, base)
 
-
-static void tfp410_encoder_destroy(struct drm_encoder *encoder)
-{
-	struct tfp410_encoder *tfp410_encoder = to_tfp410_encoder(encoder);
-	drm_encoder_cleanup(encoder);
-	kfree(tfp410_encoder);
-}
-
 static void tfp410_encoder_dpms(struct drm_encoder *encoder, int mode)
 {
 	struct tfp410_encoder *tfp410_encoder = to_tfp410_encoder(encoder);
@@ -99,7 +91,7 @@ static void tfp410_encoder_mode_set(struct drm_encoder *encoder,
 }
 
 static const struct drm_encoder_funcs tfp410_encoder_funcs = {
-		.destroy        = tfp410_encoder_destroy,
+		.destroy        = drm_encoder_cleanup,
 };
 
 static const struct drm_encoder_helper_funcs tfp410_encoder_helper_funcs = {
@@ -116,7 +108,8 @@ static struct drm_encoder *tfp410_encoder_create(struct drm_device *dev,
 	struct drm_encoder *encoder;
 	int ret;
 
-	tfp410_encoder = kzalloc(sizeof(*tfp410_encoder), GFP_KERNEL);
+	tfp410_encoder = devm_kzalloc(dev->dev, sizeof(*tfp410_encoder),
+				      GFP_KERNEL);
 	if (!tfp410_encoder) {
 		dev_err(dev->dev, "allocation failed\n");
 		return NULL;
@@ -138,7 +131,7 @@ static struct drm_encoder *tfp410_encoder_create(struct drm_device *dev,
 	return encoder;
 
 fail:
-	tfp410_encoder_destroy(encoder);
+	drm_encoder_cleanup(encoder);
 	return NULL;
 }
 
@@ -157,10 +150,8 @@ struct tfp410_connector {
 
 static void tfp410_connector_destroy(struct drm_connector *connector)
 {
-	struct tfp410_connector *tfp410_connector = to_tfp410_connector(connector);
 	drm_connector_unregister(connector);
 	drm_connector_cleanup(connector);
-	kfree(tfp410_connector);
 }
 
 static enum drm_connector_status tfp410_connector_detect(
@@ -228,7 +219,8 @@ static struct drm_connector *tfp410_connector_create(struct drm_device *dev,
 	struct drm_connector *connector;
 	int ret;
 
-	tfp410_connector = kzalloc(sizeof(*tfp410_connector), GFP_KERNEL);
+	tfp410_connector = devm_kzalloc(dev->dev, sizeof(*tfp410_connector),
+					GFP_KERNEL);
 	if (!tfp410_connector) {
 		dev_err(dev->dev, "allocation failed\n");
 		return NULL;
@@ -313,7 +305,7 @@ static int tfp410_probe(struct platform_device *pdev)
 		return -ENXIO;
 	}
 
-	tfp410_mod = kzalloc(sizeof(*tfp410_mod), GFP_KERNEL);
+	tfp410_mod = devm_kzalloc(&pdev->dev, sizeof(*tfp410_mod), GFP_KERNEL);
 	if (!tfp410_mod)
 		return -ENOMEM;
 
@@ -366,7 +358,6 @@ fail_adapter:
 	i2c_put_adapter(tfp410_mod->i2c);
 
 fail:
-	kfree(tfp410_mod);
 	tilcdc_module_cleanup(mod);
 	return ret;
 }
@@ -380,7 +371,6 @@ static int tfp410_remove(struct platform_device *pdev)
 	gpio_free(tfp410_mod->gpio);
 
 	tilcdc_module_cleanup(mod);
-	kfree(tfp410_mod);
 
 	return 0;
 }
diff --git a/drivers/gpu/drm/vc4/vc4_bo.c b/drivers/gpu/drm/vc4/vc4_bo.c
index 18dfe3ec9a62..ac8eafea6361 100644
--- a/drivers/gpu/drm/vc4/vc4_bo.c
+++ b/drivers/gpu/drm/vc4/vc4_bo.c
@@ -215,7 +215,7 @@ struct vc4_bo *vc4_bo_create(struct drm_device *dev, size_t unaligned_size,
 	struct drm_gem_cma_object *cma_obj;
 
 	if (size == 0)
-		return NULL;
+		return ERR_PTR(-EINVAL);
 
 	/* First, try to get a vc4_bo from the kernel BO cache. */
 	if (from_cache) {
@@ -237,7 +237,7 @@ struct vc4_bo *vc4_bo_create(struct drm_device *dev, size_t unaligned_size,
 		if (IS_ERR(cma_obj)) {
 			DRM_ERROR("Failed to allocate from CMA:\n");
 			vc4_bo_stats_dump(vc4);
-			return NULL;
+			return ERR_PTR(-ENOMEM);
 		}
 	}
 
@@ -259,8 +259,8 @@ int vc4_dumb_create(struct drm_file *file_priv,
 		args->size = args->pitch * args->height;
 
 	bo = vc4_bo_create(dev, args->size, false);
-	if (!bo)
-		return -ENOMEM;
+	if (IS_ERR(bo))
+		return PTR_ERR(bo);
 
 	ret = drm_gem_handle_create(file_priv, &bo->base.base, &args->handle);
 	drm_gem_object_unreference_unlocked(&bo->base.base);
@@ -443,8 +443,8 @@ int vc4_create_bo_ioctl(struct drm_device *dev, void *data,
 	 * get zeroed, and that might leak data between users.
 	 */
 	bo = vc4_bo_create(dev, args->size, false);
-	if (!bo)
-		return -ENOMEM;
+	if (IS_ERR(bo))
+		return PTR_ERR(bo);
 
 	ret = drm_gem_handle_create(file_priv, &bo->base.base, &args->handle);
 	drm_gem_object_unreference_unlocked(&bo->base.base);
@@ -496,14 +496,15 @@ vc4_create_shader_bo_ioctl(struct drm_device *dev, void *data,
 	}
 
 	bo = vc4_bo_create(dev, args->size, true);
-	if (!bo)
-		return -ENOMEM;
+	if (IS_ERR(bo))
+		return PTR_ERR(bo);
 
-	ret = copy_from_user(bo->base.vaddr,
+	if (copy_from_user(bo->base.vaddr,
 			     (void __user *)(uintptr_t)args->data,
-			     args->size);
-	if (ret != 0)
+			     args->size)) {
+		ret = -EFAULT;
 		goto fail;
+	}
 	/* Clear the rest of the memory from allocating from the BO
 	 * cache.
 	 */
diff --git a/drivers/gpu/drm/vc4/vc4_crtc.c b/drivers/gpu/drm/vc4/vc4_crtc.c
index 619dc781c517..355ee4b091b3 100644
--- a/drivers/gpu/drm/vc4/vc4_crtc.c
+++ b/drivers/gpu/drm/vc4/vc4_crtc.c
@@ -88,7 +88,7 @@ static const struct {
 } crtc_regs[] = {
 	CRTC_REG(PV_CONTROL),
 	CRTC_REG(PV_V_CONTROL),
-	CRTC_REG(PV_VSYNCD),
+	CRTC_REG(PV_VSYNCD_EVEN),
 	CRTC_REG(PV_HORZA),
 	CRTC_REG(PV_HORZB),
 	CRTC_REG(PV_VERTA),
@@ -188,6 +188,8 @@ static int vc4_get_clock_select(struct drm_crtc *crtc)
 
 static void vc4_crtc_mode_set_nofb(struct drm_crtc *crtc)
 {
+	struct drm_device *dev = crtc->dev;
+	struct vc4_dev *vc4 = to_vc4_dev(dev);
 	struct vc4_crtc *vc4_crtc = to_vc4_crtc(crtc);
 	struct drm_crtc_state *state = crtc->state;
 	struct drm_display_mode *mode = &state->adjusted_mode;
@@ -217,6 +219,16 @@ static void vc4_crtc_mode_set_nofb(struct drm_crtc *crtc)
 				 PV_HORZB_HFP) |
 		   VC4_SET_FIELD(mode->hdisplay, PV_HORZB_HACTIVE));
 
+	CRTC_WRITE(PV_VERTA,
+		   VC4_SET_FIELD(mode->vtotal - mode->vsync_end,
+				 PV_VERTA_VBP) |
+		   VC4_SET_FIELD(mode->vsync_end - mode->vsync_start,
+				 PV_VERTA_VSYNC));
+	CRTC_WRITE(PV_VERTB,
+		   VC4_SET_FIELD(mode->vsync_start - mode->vdisplay,
+				 PV_VERTB_VFP) |
+		   VC4_SET_FIELD(vactive, PV_VERTB_VACTIVE));
+
 	if (interlace) {
 		CRTC_WRITE(PV_VERTA_EVEN,
 			   VC4_SET_FIELD(mode->vtotal - mode->vsync_end - 1,
@@ -246,6 +258,10 @@ static void vc4_crtc_mode_set_nofb(struct drm_crtc *crtc)
 		   PV_CONTROL_FIFO_CLR |
 		   PV_CONTROL_EN);
 
+	HVS_WRITE(SCALER_DISPBKGNDX(vc4_crtc->channel),
+		  SCALER_DISPBKGND_AUTOHS |
+		  (interlace ? SCALER_DISPBKGND_INTERLACE : 0));
+
 	if (debug_dump_regs) {
 		DRM_INFO("CRTC %d regs after:\n", drm_crtc_index(crtc));
 		vc4_crtc_dump_regs(vc4_crtc);
@@ -527,6 +543,7 @@ static int vc4_async_page_flip(struct drm_crtc *crtc,
 	/* Make sure all other async modesetes have landed. */
 	ret = down_interruptible(&vc4->async_modeset);
 	if (ret) {
+		drm_framebuffer_unreference(fb);
 		kfree(flip_state);
 		return ret;
 	}
diff --git a/drivers/gpu/drm/vc4/vc4_drv.h b/drivers/gpu/drm/vc4/vc4_drv.h
index 3d1df6b1c4d3..fa2ad15d4f62 100644
--- a/drivers/gpu/drm/vc4/vc4_drv.h
+++ b/drivers/gpu/drm/vc4/vc4_drv.h
@@ -52,7 +52,7 @@ struct vc4_dev {
 	/* Protects bo_cache and the BO stats. */
 	struct mutex bo_lock;
 
-	/* Sequence number for the last job queued in job_list.
+	/* Sequence number for the last job queued in bin_job_list.
 	 * Starts at 0 (no jobs emitted).
 	 */
 	uint64_t emit_seqno;
@@ -62,11 +62,19 @@ struct vc4_dev {
 	 */
 	uint64_t finished_seqno;
 
-	/* List of all struct vc4_exec_info for jobs to be executed.
-	 * The first job in the list is the one currently programmed
-	 * into ct0ca/ct1ca for execution.
+	/* List of all struct vc4_exec_info for jobs to be executed in
+	 * the binner.  The first job in the list is the one currently
+	 * programmed into ct0ca for execution.
 	 */
-	struct list_head job_list;
+	struct list_head bin_job_list;
+
+	/* List of all struct vc4_exec_info for jobs that have
+	 * completed binning and are ready for rendering.  The first
+	 * job in the list is the one currently programmed into ct1ca
+	 * for execution.
+	 */
+	struct list_head render_job_list;
+
 	/* List of the finished vc4_exec_infos waiting to be freed by
 	 * job_done_work.
 	 */
@@ -91,8 +99,12 @@ struct vc4_dev {
 	struct vc4_bo *overflow_mem;
 	struct work_struct overflow_mem_work;
 
+	int power_refcount;
+
+	/* Mutex controlling the power refcount. */
+	struct mutex power_lock;
+
 	struct {
-		uint32_t last_ct0ca, last_ct1ca;
 		struct timer_list timer;
 		struct work_struct reset_work;
 	} hangcheck;
@@ -142,6 +154,7 @@ struct vc4_seqno_cb {
 };
 
 struct vc4_v3d {
+	struct vc4_dev *vc4;
 	struct platform_device *pdev;
 	void __iomem *regs;
 };
@@ -202,6 +215,11 @@ struct vc4_exec_info {
 	/* Sequence number for this bin/render job. */
 	uint64_t seqno;
 
+	/* Last current addresses the hardware was processing when the
+	 * hangcheck timer checked on us.
+	 */
+	uint32_t last_ct0ca, last_ct1ca;
+
 	/* Kernel-space copy of the ioctl arguments */
 	struct drm_vc4_submit_cl *args;
 
@@ -286,11 +304,20 @@ struct vc4_exec_info {
 };
 
 static inline struct vc4_exec_info *
-vc4_first_job(struct vc4_dev *vc4)
+vc4_first_bin_job(struct vc4_dev *vc4)
+{
+	if (list_empty(&vc4->bin_job_list))
+		return NULL;
+	return list_first_entry(&vc4->bin_job_list, struct vc4_exec_info, head);
+}
+
+static inline struct vc4_exec_info *
+vc4_first_render_job(struct vc4_dev *vc4)
 {
-	if (list_empty(&vc4->job_list))
+	if (list_empty(&vc4->render_job_list))
 		return NULL;
-	return list_first_entry(&vc4->job_list, struct vc4_exec_info, head);
+	return list_first_entry(&vc4->render_job_list,
+				struct vc4_exec_info, head);
 }
 
 /**
@@ -404,7 +431,9 @@ int vc4_wait_seqno_ioctl(struct drm_device *dev, void *data,
 			 struct drm_file *file_priv);
 int vc4_wait_bo_ioctl(struct drm_device *dev, void *data,
 		      struct drm_file *file_priv);
-void vc4_submit_next_job(struct drm_device *dev);
+void vc4_submit_next_bin_job(struct drm_device *dev);
+void vc4_submit_next_render_job(struct drm_device *dev);
+void vc4_move_job_to_render(struct drm_device *dev, struct vc4_exec_info *exec);
 int vc4_wait_for_seqno(struct drm_device *dev, uint64_t seqno,
 		       uint64_t timeout_ns, bool interruptible);
 void vc4_job_handle_completed(struct vc4_dev *vc4);
@@ -443,7 +472,6 @@ void vc4_plane_async_set_fb(struct drm_plane *plane,
 extern struct platform_driver vc4_v3d_driver;
 int vc4_v3d_debugfs_ident(struct seq_file *m, void *unused);
 int vc4_v3d_debugfs_regs(struct seq_file *m, void *unused);
-int vc4_v3d_set_power(struct vc4_dev *vc4, bool on);
 
 /* vc4_validate.c */
 int
diff --git a/drivers/gpu/drm/vc4/vc4_gem.c b/drivers/gpu/drm/vc4/vc4_gem.c
index 48ce30a6f4b5..8d4384f8b78d 100644
--- a/drivers/gpu/drm/vc4/vc4_gem.c
+++ b/drivers/gpu/drm/vc4/vc4_gem.c
@@ -23,6 +23,7 @@
 
 #include <linux/module.h>
 #include <linux/platform_device.h>
+#include <linux/pm_runtime.h>
 #include <linux/device.h>
 #include <linux/io.h>
 
@@ -140,10 +141,10 @@ vc4_save_hang_state(struct drm_device *dev)
 	struct vc4_dev *vc4 = to_vc4_dev(dev);
 	struct drm_vc4_get_hang_state *state;
 	struct vc4_hang_state *kernel_state;
-	struct vc4_exec_info *exec;
+	struct vc4_exec_info *exec[2];
 	struct vc4_bo *bo;
 	unsigned long irqflags;
-	unsigned int i, unref_list_count;
+	unsigned int i, j, unref_list_count, prev_idx;
 
 	kernel_state = kcalloc(1, sizeof(*kernel_state), GFP_KERNEL);
 	if (!kernel_state)
@@ -152,37 +153,55 @@ vc4_save_hang_state(struct drm_device *dev)
 	state = &kernel_state->user_state;
 
 	spin_lock_irqsave(&vc4->job_lock, irqflags);
-	exec = vc4_first_job(vc4);
-	if (!exec) {
+	exec[0] = vc4_first_bin_job(vc4);
+	exec[1] = vc4_first_render_job(vc4);
+	if (!exec[0] && !exec[1]) {
 		spin_unlock_irqrestore(&vc4->job_lock, irqflags);
 		return;
 	}
 
-	unref_list_count = 0;
-	list_for_each_entry(bo, &exec->unref_list, unref_head)
-		unref_list_count++;
+	/* Get the bos from both binner and renderer into hang state. */
+	state->bo_count = 0;
+	for (i = 0; i < 2; i++) {
+		if (!exec[i])
+			continue;
+
+		unref_list_count = 0;
+		list_for_each_entry(bo, &exec[i]->unref_list, unref_head)
+			unref_list_count++;
+		state->bo_count += exec[i]->bo_count + unref_list_count;
+	}
+
+	kernel_state->bo = kcalloc(state->bo_count,
+				   sizeof(*kernel_state->bo), GFP_ATOMIC);
 
-	state->bo_count = exec->bo_count + unref_list_count;
-	kernel_state->bo = kcalloc(state->bo_count, sizeof(*kernel_state->bo),
-				   GFP_ATOMIC);
 	if (!kernel_state->bo) {
 		spin_unlock_irqrestore(&vc4->job_lock, irqflags);
 		return;
 	}
 
-	for (i = 0; i < exec->bo_count; i++) {
-		drm_gem_object_reference(&exec->bo[i]->base);
-		kernel_state->bo[i] = &exec->bo[i]->base;
-	}
+	prev_idx = 0;
+	for (i = 0; i < 2; i++) {
+		if (!exec[i])
+			continue;
 
-	list_for_each_entry(bo, &exec->unref_list, unref_head) {
-		drm_gem_object_reference(&bo->base.base);
-		kernel_state->bo[i] = &bo->base.base;
-		i++;
+		for (j = 0; j < exec[i]->bo_count; j++) {
+			drm_gem_object_reference(&exec[i]->bo[j]->base);
+			kernel_state->bo[j + prev_idx] = &exec[i]->bo[j]->base;
+		}
+
+		list_for_each_entry(bo, &exec[i]->unref_list, unref_head) {
+			drm_gem_object_reference(&bo->base.base);
+			kernel_state->bo[j + prev_idx] = &bo->base.base;
+			j++;
+		}
+		prev_idx = j + 1;
 	}
 
-	state->start_bin = exec->ct0ca;
-	state->start_render = exec->ct1ca;
+	if (exec[0])
+		state->start_bin = exec[0]->ct0ca;
+	if (exec[1])
+		state->start_render = exec[1]->ct1ca;
 
 	spin_unlock_irqrestore(&vc4->job_lock, irqflags);
 
@@ -228,8 +247,16 @@ vc4_reset(struct drm_device *dev)
 	struct vc4_dev *vc4 = to_vc4_dev(dev);
 
 	DRM_INFO("Resetting GPU.\n");
-	vc4_v3d_set_power(vc4, false);
-	vc4_v3d_set_power(vc4, true);
+
+	mutex_lock(&vc4->power_lock);
+	if (vc4->power_refcount) {
+		/* Power the device off and back on the by dropping the
+		 * reference on runtime PM.
+		 */
+		pm_runtime_put_sync_suspend(&vc4->v3d->pdev->dev);
+		pm_runtime_get_sync(&vc4->v3d->pdev->dev);
+	}
+	mutex_unlock(&vc4->power_lock);
 
 	vc4_irq_reset(dev);
 
@@ -257,10 +284,19 @@ vc4_hangcheck_elapsed(unsigned long data)
 	struct drm_device *dev = (struct drm_device *)data;
 	struct vc4_dev *vc4 = to_vc4_dev(dev);
 	uint32_t ct0ca, ct1ca;
+	unsigned long irqflags;
+	struct vc4_exec_info *bin_exec, *render_exec;
+
+	spin_lock_irqsave(&vc4->job_lock, irqflags);
+
+	bin_exec = vc4_first_bin_job(vc4);
+	render_exec = vc4_first_render_job(vc4);
 
 	/* If idle, we can stop watching for hangs. */
-	if (list_empty(&vc4->job_list))
+	if (!bin_exec && !render_exec) {
+		spin_unlock_irqrestore(&vc4->job_lock, irqflags);
 		return;
+	}
 
 	ct0ca = V3D_READ(V3D_CTNCA(0));
 	ct1ca = V3D_READ(V3D_CTNCA(1));
@@ -268,14 +304,19 @@ vc4_hangcheck_elapsed(unsigned long data)
 	/* If we've made any progress in execution, rearm the timer
 	 * and wait.
 	 */
-	if (ct0ca != vc4->hangcheck.last_ct0ca ||
-	    ct1ca != vc4->hangcheck.last_ct1ca) {
-		vc4->hangcheck.last_ct0ca = ct0ca;
-		vc4->hangcheck.last_ct1ca = ct1ca;
+	if ((bin_exec && ct0ca != bin_exec->last_ct0ca) ||
+	    (render_exec && ct1ca != render_exec->last_ct1ca)) {
+		if (bin_exec)
+			bin_exec->last_ct0ca = ct0ca;
+		if (render_exec)
+			render_exec->last_ct1ca = ct1ca;
+		spin_unlock_irqrestore(&vc4->job_lock, irqflags);
 		vc4_queue_hangcheck(dev);
 		return;
 	}
 
+	spin_unlock_irqrestore(&vc4->job_lock, irqflags);
+
 	/* We've gone too long with no progress, reset.  This has to
 	 * be done from a work struct, since resetting can sleep and
 	 * this timer hook isn't allowed to.
@@ -340,12 +381,7 @@ vc4_wait_for_seqno(struct drm_device *dev, uint64_t seqno, uint64_t timeout_ns,
 	finish_wait(&vc4->job_wait_queue, &wait);
 	trace_vc4_wait_for_seqno_end(dev, seqno);
 
-	if (ret && ret != -ERESTARTSYS) {
-		DRM_ERROR("timeout waiting for render thread idle\n");
-		return ret;
-	}
-
-	return 0;
+	return ret;
 }
 
 static void
@@ -373,11 +409,13 @@ vc4_flush_caches(struct drm_device *dev)
  * The job_lock should be held during this.
  */
 void
-vc4_submit_next_job(struct drm_device *dev)
+vc4_submit_next_bin_job(struct drm_device *dev)
 {
 	struct vc4_dev *vc4 = to_vc4_dev(dev);
-	struct vc4_exec_info *exec = vc4_first_job(vc4);
+	struct vc4_exec_info *exec;
 
+again:
+	exec = vc4_first_bin_job(vc4);
 	if (!exec)
 		return;
 
@@ -387,11 +425,40 @@ vc4_submit_next_job(struct drm_device *dev)
 	V3D_WRITE(V3D_BPOA, 0);
 	V3D_WRITE(V3D_BPOS, 0);
 
-	if (exec->ct0ca != exec->ct0ea)
+	/* Either put the job in the binner if it uses the binner, or
+	 * immediately move it to the to-be-rendered queue.
+	 */
+	if (exec->ct0ca != exec->ct0ea) {
 		submit_cl(dev, 0, exec->ct0ca, exec->ct0ea);
+	} else {
+		vc4_move_job_to_render(dev, exec);
+		goto again;
+	}
+}
+
+void
+vc4_submit_next_render_job(struct drm_device *dev)
+{
+	struct vc4_dev *vc4 = to_vc4_dev(dev);
+	struct vc4_exec_info *exec = vc4_first_render_job(vc4);
+
+	if (!exec)
+		return;
+
 	submit_cl(dev, 1, exec->ct1ca, exec->ct1ea);
 }
 
+void
+vc4_move_job_to_render(struct drm_device *dev, struct vc4_exec_info *exec)
+{
+	struct vc4_dev *vc4 = to_vc4_dev(dev);
+	bool was_empty = list_empty(&vc4->render_job_list);
+
+	list_move_tail(&exec->head, &vc4->render_job_list);
+	if (was_empty)
+		vc4_submit_next_render_job(dev);
+}
+
 static void
 vc4_update_bo_seqnos(struct vc4_exec_info *exec, uint64_t seqno)
 {
@@ -430,14 +497,14 @@ vc4_queue_submit(struct drm_device *dev, struct vc4_exec_info *exec)
 	exec->seqno = seqno;
 	vc4_update_bo_seqnos(exec, seqno);
 
-	list_add_tail(&exec->head, &vc4->job_list);
+	list_add_tail(&exec->head, &vc4->bin_job_list);
 
 	/* If no job was executing, kick ours off.  Otherwise, it'll
-	 * get started when the previous job's frame done interrupt
+	 * get started when the previous job's flush done interrupt
 	 * occurs.
 	 */
-	if (vc4_first_job(vc4) == exec) {
-		vc4_submit_next_job(dev);
+	if (vc4_first_bin_job(vc4) == exec) {
+		vc4_submit_next_bin_job(dev);
 		vc4_queue_hangcheck(dev);
 	}
 
@@ -578,9 +645,9 @@ vc4_get_bcl(struct drm_device *dev, struct vc4_exec_info *exec)
 	}
 
 	bo = vc4_bo_create(dev, exec_size, true);
-	if (!bo) {
+	if (IS_ERR(bo)) {
 		DRM_ERROR("Couldn't allocate BO for binning\n");
-		ret = -ENOMEM;
+		ret = PTR_ERR(bo);
 		goto fail;
 	}
 	exec->exec_bo = &bo->base;
@@ -617,6 +684,7 @@ fail:
 static void
 vc4_complete_exec(struct drm_device *dev, struct vc4_exec_info *exec)
 {
+	struct vc4_dev *vc4 = to_vc4_dev(dev);
 	unsigned i;
 
 	/* Need the struct lock for drm_gem_object_unreference(). */
@@ -635,6 +703,11 @@ vc4_complete_exec(struct drm_device *dev, struct vc4_exec_info *exec)
 	}
 	mutex_unlock(&dev->struct_mutex);
 
+	mutex_lock(&vc4->power_lock);
+	if (--vc4->power_refcount == 0)
+		pm_runtime_put(&vc4->v3d->pdev->dev);
+	mutex_unlock(&vc4->power_lock);
+
 	kfree(exec);
 }
 
@@ -746,6 +819,9 @@ vc4_wait_bo_ioctl(struct drm_device *dev, void *data,
 	struct drm_gem_object *gem_obj;
 	struct vc4_bo *bo;
 
+	if (args->pad != 0)
+		return -EINVAL;
+
 	gem_obj = drm_gem_object_lookup(dev, file_priv, args->handle);
 	if (!gem_obj) {
 		DRM_ERROR("Failed to look up GEM BO %d\n", args->handle);
@@ -772,7 +848,7 @@ vc4_submit_cl_ioctl(struct drm_device *dev, void *data,
 	struct vc4_dev *vc4 = to_vc4_dev(dev);
 	struct drm_vc4_submit_cl *args = data;
 	struct vc4_exec_info *exec;
-	int ret;
+	int ret = 0;
 
 	if ((args->flags & ~VC4_SUBMIT_CL_USE_CLEAR_COLOR) != 0) {
 		DRM_ERROR("Unknown flags: 0x%02x\n", args->flags);
@@ -785,6 +861,15 @@ vc4_submit_cl_ioctl(struct drm_device *dev, void *data,
 		return -ENOMEM;
 	}
 
+	mutex_lock(&vc4->power_lock);
+	if (vc4->power_refcount++ == 0)
+		ret = pm_runtime_get_sync(&vc4->v3d->pdev->dev);
+	mutex_unlock(&vc4->power_lock);
+	if (ret < 0) {
+		kfree(exec);
+		return ret;
+	}
+
 	exec->args = args;
 	INIT_LIST_HEAD(&exec->unref_list);
 
@@ -828,7 +913,8 @@ vc4_gem_init(struct drm_device *dev)
 {
 	struct vc4_dev *vc4 = to_vc4_dev(dev);
 
-	INIT_LIST_HEAD(&vc4->job_list);
+	INIT_LIST_HEAD(&vc4->bin_job_list);
+	INIT_LIST_HEAD(&vc4->render_job_list);
 	INIT_LIST_HEAD(&vc4->job_done_list);
 	INIT_LIST_HEAD(&vc4->seqno_cb_list);
 	spin_lock_init(&vc4->job_lock);
@@ -839,6 +925,8 @@ vc4_gem_init(struct drm_device *dev)
 		    (unsigned long)dev);
 
 	INIT_WORK(&vc4->job_done_work, vc4_job_done_work);
+
+	mutex_init(&vc4->power_lock);
 }
 
 void
diff --git a/drivers/gpu/drm/vc4/vc4_hdmi.c b/drivers/gpu/drm/vc4/vc4_hdmi.c
index c69c0460196b..d8b864925fd3 100644
--- a/drivers/gpu/drm/vc4/vc4_hdmi.c
+++ b/drivers/gpu/drm/vc4/vc4_hdmi.c
@@ -47,6 +47,7 @@ struct vc4_hdmi {
 	void __iomem *hdmicore_regs;
 	void __iomem *hd_regs;
 	int hpd_gpio;
+	bool hpd_active_low;
 
 	struct clk *pixel_clock;
 	struct clk *hsm_clock;
@@ -95,6 +96,7 @@ static const struct {
 	HDMI_REG(VC4_HDMI_SW_RESET_CONTROL),
 	HDMI_REG(VC4_HDMI_HOTPLUG_INT),
 	HDMI_REG(VC4_HDMI_HOTPLUG),
+	HDMI_REG(VC4_HDMI_RAM_PACKET_CONFIG),
 	HDMI_REG(VC4_HDMI_HORZA),
 	HDMI_REG(VC4_HDMI_HORZB),
 	HDMI_REG(VC4_HDMI_FIFO_CTL),
@@ -165,7 +167,8 @@ vc4_hdmi_connector_detect(struct drm_connector *connector, bool force)
 	struct vc4_dev *vc4 = to_vc4_dev(dev);
 
 	if (vc4->hdmi->hpd_gpio) {
-		if (gpio_get_value(vc4->hdmi->hpd_gpio))
+		if (gpio_get_value_cansleep(vc4->hdmi->hpd_gpio) ^
+		    vc4->hdmi->hpd_active_low)
 			return connector_status_connected;
 		else
 			return connector_status_disconnected;
@@ -495,6 +498,16 @@ static int vc4_hdmi_bind(struct device *dev, struct device *master, void *data)
 		goto err_put_i2c;
 	}
 
+	/* This is the rate that is set by the firmware.  The number
+	 * needs to be a bit higher than the pixel clock rate
+	 * (generally 148.5Mhz).
+	 */
+	ret = clk_set_rate(hdmi->hsm_clock, 163682864);
+	if (ret) {
+		DRM_ERROR("Failed to set HSM clock rate: %d\n", ret);
+		goto err_unprepare_pix;
+	}
+
 	ret = clk_prepare_enable(hdmi->hsm_clock);
 	if (ret) {
 		DRM_ERROR("Failed to turn on HDMI state machine clock: %d\n",
@@ -506,17 +519,40 @@ static int vc4_hdmi_bind(struct device *dev, struct device *master, void *data)
 	 * we'll use the HDMI core's register.
 	 */
 	if (of_find_property(dev->of_node, "hpd-gpios", &value)) {
-		hdmi->hpd_gpio = of_get_named_gpio(dev->of_node, "hpd-gpios", 0);
+		enum of_gpio_flags hpd_gpio_flags;
+
+		hdmi->hpd_gpio = of_get_named_gpio_flags(dev->of_node,
+							 "hpd-gpios", 0,
+							 &hpd_gpio_flags);
 		if (hdmi->hpd_gpio < 0) {
 			ret = hdmi->hpd_gpio;
 			goto err_unprepare_hsm;
 		}
+
+		hdmi->hpd_active_low = hpd_gpio_flags & OF_GPIO_ACTIVE_LOW;
 	}
 
 	vc4->hdmi = hdmi;
 
 	/* HDMI core must be enabled. */
-	WARN_ON_ONCE((HD_READ(VC4_HD_M_CTL) & VC4_HD_M_ENABLE) == 0);
+	if (!(HD_READ(VC4_HD_M_CTL) & VC4_HD_M_ENABLE)) {
+		HD_WRITE(VC4_HD_M_CTL, VC4_HD_M_SW_RST);
+		udelay(1);
+		HD_WRITE(VC4_HD_M_CTL, 0);
+
+		HD_WRITE(VC4_HD_M_CTL, VC4_HD_M_ENABLE);
+
+		HDMI_WRITE(VC4_HDMI_SW_RESET_CONTROL,
+			   VC4_HDMI_SW_RESET_HDMI |
+			   VC4_HDMI_SW_RESET_FORMAT_DETECT);
+
+		HDMI_WRITE(VC4_HDMI_SW_RESET_CONTROL, 0);
+
+		/* PHY should be in reset, like
+		 * vc4_hdmi_encoder_disable() does.
+		 */
+		HDMI_WRITE(VC4_HDMI_TX_PHY_RESET_CTL, 0xf << 16);
+	}
 
 	drm_encoder_init(drm, hdmi->encoder, &vc4_hdmi_encoder_funcs,
 			 DRM_MODE_ENCODER_TMDS, NULL);
diff --git a/drivers/gpu/drm/vc4/vc4_irq.c b/drivers/gpu/drm/vc4/vc4_irq.c
index b68060e758db..b0104a346a74 100644
--- a/drivers/gpu/drm/vc4/vc4_irq.c
+++ b/drivers/gpu/drm/vc4/vc4_irq.c
@@ -30,6 +30,10 @@
  * disables that specific interrupt, and 0s written are ignored
  * (reading either one returns the set of enabled interrupts).
  *
+ * When we take a binning flush done interrupt, we need to submit the
+ * next frame for binning and move the finished frame to the render
+ * thread.
+ *
  * When we take a render frame interrupt, we need to wake the
  * processes waiting for some frame to be done, and get the next frame
  * submitted ASAP (so the hardware doesn't sit idle when there's work
@@ -44,6 +48,7 @@
 #include "vc4_regs.h"
 
 #define V3D_DRIVER_IRQS (V3D_INT_OUTOMEM | \
+			 V3D_INT_FLDONE | \
 			 V3D_INT_FRDONE)
 
 DECLARE_WAIT_QUEUE_HEAD(render_wait);
@@ -57,7 +62,7 @@ vc4_overflow_mem_work(struct work_struct *work)
 	struct vc4_bo *bo;
 
 	bo = vc4_bo_create(dev, 256 * 1024, true);
-	if (!bo) {
+	if (IS_ERR(bo)) {
 		DRM_ERROR("Couldn't allocate binner overflow mem\n");
 		return;
 	}
@@ -77,7 +82,7 @@ vc4_overflow_mem_work(struct work_struct *work)
 		unsigned long irqflags;
 
 		spin_lock_irqsave(&vc4->job_lock, irqflags);
-		current_exec = vc4_first_job(vc4);
+		current_exec = vc4_first_bin_job(vc4);
 		if (current_exec) {
 			vc4->overflow_mem->seqno = vc4->finished_seqno + 1;
 			list_add_tail(&vc4->overflow_mem->unref_head,
@@ -98,17 +103,43 @@ vc4_overflow_mem_work(struct work_struct *work)
 }
 
 static void
-vc4_irq_finish_job(struct drm_device *dev)
+vc4_irq_finish_bin_job(struct drm_device *dev)
+{
+	struct vc4_dev *vc4 = to_vc4_dev(dev);
+	struct vc4_exec_info *exec = vc4_first_bin_job(vc4);
+
+	if (!exec)
+		return;
+
+	vc4_move_job_to_render(dev, exec);
+	vc4_submit_next_bin_job(dev);
+}
+
+static void
+vc4_cancel_bin_job(struct drm_device *dev)
+{
+	struct vc4_dev *vc4 = to_vc4_dev(dev);
+	struct vc4_exec_info *exec = vc4_first_bin_job(vc4);
+
+	if (!exec)
+		return;
+
+	list_move_tail(&exec->head, &vc4->bin_job_list);
+	vc4_submit_next_bin_job(dev);
+}
+
+static void
+vc4_irq_finish_render_job(struct drm_device *dev)
 {
 	struct vc4_dev *vc4 = to_vc4_dev(dev);
-	struct vc4_exec_info *exec = vc4_first_job(vc4);
+	struct vc4_exec_info *exec = vc4_first_render_job(vc4);
 
 	if (!exec)
 		return;
 
 	vc4->finished_seqno++;
 	list_move_tail(&exec->head, &vc4->job_done_list);
-	vc4_submit_next_job(dev);
+	vc4_submit_next_render_job(dev);
 
 	wake_up_all(&vc4->job_wait_queue);
 	schedule_work(&vc4->job_done_work);
@@ -125,9 +156,10 @@ vc4_irq(int irq, void *arg)
 	barrier();
 	intctl = V3D_READ(V3D_INTCTL);
 
-	/* Acknowledge the interrupts we're handling here. The render
-	 * frame done interrupt will be cleared, while OUTOMEM will
-	 * stay high until the underlying cause is cleared.
+	/* Acknowledge the interrupts we're handling here. The binner
+	 * last flush / render frame done interrupt will be cleared,
+	 * while OUTOMEM will stay high until the underlying cause is
+	 * cleared.
 	 */
 	V3D_WRITE(V3D_INTCTL, intctl);
 
@@ -138,9 +170,16 @@ vc4_irq(int irq, void *arg)
 		status = IRQ_HANDLED;
 	}
 
+	if (intctl & V3D_INT_FLDONE) {
+		spin_lock(&vc4->job_lock);
+		vc4_irq_finish_bin_job(dev);
+		spin_unlock(&vc4->job_lock);
+		status = IRQ_HANDLED;
+	}
+
 	if (intctl & V3D_INT_FRDONE) {
 		spin_lock(&vc4->job_lock);
-		vc4_irq_finish_job(dev);
+		vc4_irq_finish_render_job(dev);
 		spin_unlock(&vc4->job_lock);
 		status = IRQ_HANDLED;
 	}
@@ -205,6 +244,7 @@ void vc4_irq_reset(struct drm_device *dev)
 	V3D_WRITE(V3D_INTENA, V3D_DRIVER_IRQS);
 
 	spin_lock_irqsave(&vc4->job_lock, irqflags);
-	vc4_irq_finish_job(dev);
+	vc4_cancel_bin_job(dev);
+	vc4_irq_finish_render_job(dev);
 	spin_unlock_irqrestore(&vc4->job_lock, irqflags);
 }
diff --git a/drivers/gpu/drm/vc4/vc4_regs.h b/drivers/gpu/drm/vc4/vc4_regs.h
index 25df20ef939c..bf42a8e87111 100644
--- a/drivers/gpu/drm/vc4/vc4_regs.h
+++ b/drivers/gpu/drm/vc4/vc4_regs.h
@@ -187,7 +187,7 @@
 # define PV_VCONTROL_CONTINUOUS			BIT(1)
 # define PV_VCONTROL_VIDEN			BIT(0)
 
-#define PV_VSYNCD				0x08
+#define PV_VSYNCD_EVEN				0x08
 
 #define PV_HORZA				0x0c
 # define PV_HORZA_HBP_MASK			VC4_MASK(31, 16)
@@ -350,6 +350,17 @@
 # define SCALER_DISPCTRLX_HEIGHT_SHIFT		0
 
 #define SCALER_DISPBKGND0                       0x00000044
+# define SCALER_DISPBKGND_AUTOHS		BIT(31)
+# define SCALER_DISPBKGND_INTERLACE		BIT(30)
+# define SCALER_DISPBKGND_GAMMA			BIT(29)
+# define SCALER_DISPBKGND_TESTMODE_MASK		VC4_MASK(28, 25)
+# define SCALER_DISPBKGND_TESTMODE_SHIFT	25
+/* Enables filling the scaler line with the RGB value in the low 24
+ * bits before compositing.  Costs cycles, so should be skipped if
+ * opaque display planes will cover everything.
+ */
+# define SCALER_DISPBKGND_FILL			BIT(24)
+
 #define SCALER_DISPSTAT0                        0x00000048
 #define SCALER_DISPBASE0                        0x0000004c
 # define SCALER_DISPSTATX_MODE_MASK		VC4_MASK(31, 30)
@@ -362,6 +373,9 @@
 # define SCALER_DISPSTATX_EMPTY			BIT(28)
 #define SCALER_DISPCTRL1                        0x00000050
 #define SCALER_DISPBKGND1                       0x00000054
+#define SCALER_DISPBKGNDX(x)			(SCALER_DISPBKGND0 +        \
+						 (x) * (SCALER_DISPBKGND1 - \
+							SCALER_DISPBKGND0))
 #define SCALER_DISPSTAT1                        0x00000058
 #define SCALER_DISPSTATX(x)			(SCALER_DISPSTAT0 +        \
 						 (x) * (SCALER_DISPSTAT1 - \
@@ -456,6 +470,8 @@
 #define VC4_HDMI_TX_PHY_RESET_CTL		0x2c0
 
 #define VC4_HD_M_CTL				0x00c
+# define VC4_HD_M_REGISTER_FILE_STANDBY		(3 << 6)
+# define VC4_HD_M_RAM_STANDBY			(3 << 4)
 # define VC4_HD_M_SW_RST			BIT(2)
 # define VC4_HD_M_ENABLE			BIT(0)
 
diff --git a/drivers/gpu/drm/vc4/vc4_render_cl.c b/drivers/gpu/drm/vc4/vc4_render_cl.c
index 8a2a312e2c1b..0f12418725e5 100644
--- a/drivers/gpu/drm/vc4/vc4_render_cl.c
+++ b/drivers/gpu/drm/vc4/vc4_render_cl.c
@@ -316,20 +316,11 @@ static int vc4_create_rcl_bo(struct drm_device *dev, struct vc4_exec_info *exec,
 	size += xtiles * ytiles * loop_body_size;
 
 	setup->rcl = &vc4_bo_create(dev, size, true)->base;
-	if (!setup->rcl)
-		return -ENOMEM;
+	if (IS_ERR(setup->rcl))
+		return PTR_ERR(setup->rcl);
 	list_add_tail(&to_vc4_bo(&setup->rcl->base)->unref_head,
 		      &exec->unref_list);
 
-	rcl_u8(setup, VC4_PACKET_TILE_RENDERING_MODE_CONFIG);
-	rcl_u32(setup,
-		(setup->color_write ? (setup->color_write->paddr +
-				       args->color_write.offset) :
-		 0));
-	rcl_u16(setup, args->width);
-	rcl_u16(setup, args->height);
-	rcl_u16(setup, args->color_write.bits);
-
 	/* The tile buffer gets cleared when the previous tile is stored.  If
 	 * the clear values changed between frames, then the tile buffer has
 	 * stale clear values in it, so we have to do a store in None mode (no
@@ -349,6 +340,15 @@ static int vc4_create_rcl_bo(struct drm_device *dev, struct vc4_exec_info *exec,
 		rcl_u32(setup, 0); /* no address, since we're in None mode */
 	}
 
+	rcl_u8(setup, VC4_PACKET_TILE_RENDERING_MODE_CONFIG);
+	rcl_u32(setup,
+		(setup->color_write ? (setup->color_write->paddr +
+				       args->color_write.offset) :
+		 0));
+	rcl_u16(setup, args->width);
+	rcl_u16(setup, args->height);
+	rcl_u16(setup, args->color_write.bits);
+
 	for (y = min_y_tile; y <= max_y_tile; y++) {
 		for (x = min_x_tile; x <= max_x_tile; x++) {
 			bool first = (x == min_x_tile && y == min_y_tile);
diff --git a/drivers/gpu/drm/vc4/vc4_v3d.c b/drivers/gpu/drm/vc4/vc4_v3d.c
index 314ff71db978..e6d3c6028341 100644
--- a/drivers/gpu/drm/vc4/vc4_v3d.c
+++ b/drivers/gpu/drm/vc4/vc4_v3d.c
@@ -17,6 +17,7 @@
  */
 
 #include "linux/component.h"
+#include "linux/pm_runtime.h"
 #include "vc4_drv.h"
 #include "vc4_regs.h"
 
@@ -144,18 +145,6 @@ int vc4_v3d_debugfs_ident(struct seq_file *m, void *unused)
 }
 #endif /* CONFIG_DEBUG_FS */
 
-int
-vc4_v3d_set_power(struct vc4_dev *vc4, bool on)
-{
-	/* XXX: This interface is needed for GPU reset, and the way to
-	 * do it is to turn our power domain off and back on.  We
-	 * can't just reset from within the driver, because the reset
-	 * bits are in the power domain's register area, and get set
-	 * during the poweron process.
-	 */
-	return 0;
-}
-
 static void vc4_v3d_init_hw(struct drm_device *dev)
 {
 	struct vc4_dev *vc4 = to_vc4_dev(dev);
@@ -167,6 +156,29 @@ static void vc4_v3d_init_hw(struct drm_device *dev)
 	V3D_WRITE(V3D_VPMBASE, 0);
 }
 
+#ifdef CONFIG_PM
+static int vc4_v3d_runtime_suspend(struct device *dev)
+{
+	struct vc4_v3d *v3d = dev_get_drvdata(dev);
+	struct vc4_dev *vc4 = v3d->vc4;
+
+	vc4_irq_uninstall(vc4->dev);
+
+	return 0;
+}
+
+static int vc4_v3d_runtime_resume(struct device *dev)
+{
+	struct vc4_v3d *v3d = dev_get_drvdata(dev);
+	struct vc4_dev *vc4 = v3d->vc4;
+
+	vc4_v3d_init_hw(vc4->dev);
+	vc4_irq_postinstall(vc4->dev);
+
+	return 0;
+}
+#endif
+
 static int vc4_v3d_bind(struct device *dev, struct device *master, void *data)
 {
 	struct platform_device *pdev = to_platform_device(dev);
@@ -179,6 +191,8 @@ static int vc4_v3d_bind(struct device *dev, struct device *master, void *data)
 	if (!v3d)
 		return -ENOMEM;
 
+	dev_set_drvdata(dev, v3d);
+
 	v3d->pdev = pdev;
 
 	v3d->regs = vc4_ioremap_regs(pdev, 0);
@@ -186,6 +200,7 @@ static int vc4_v3d_bind(struct device *dev, struct device *master, void *data)
 		return PTR_ERR(v3d->regs);
 
 	vc4->v3d = v3d;
+	v3d->vc4 = vc4;
 
 	if (V3D_READ(V3D_IDENT0) != V3D_EXPECTED_IDENT0) {
 		DRM_ERROR("V3D_IDENT0 read 0x%08x instead of 0x%08x\n",
@@ -207,6 +222,8 @@ static int vc4_v3d_bind(struct device *dev, struct device *master, void *data)
 		return ret;
 	}
 
+	pm_runtime_enable(dev);
+
 	return 0;
 }
 
@@ -216,6 +233,8 @@ static void vc4_v3d_unbind(struct device *dev, struct device *master,
 	struct drm_device *drm = dev_get_drvdata(master);
 	struct vc4_dev *vc4 = to_vc4_dev(drm);
 
+	pm_runtime_disable(dev);
+
 	drm_irq_uninstall(drm);
 
 	/* Disable the binner's overflow memory address, so the next
@@ -228,6 +247,10 @@ static void vc4_v3d_unbind(struct device *dev, struct device *master,
 	vc4->v3d = NULL;
 }
 
+static const struct dev_pm_ops vc4_v3d_pm_ops = {
+	SET_RUNTIME_PM_OPS(vc4_v3d_runtime_suspend, vc4_v3d_runtime_resume, NULL)
+};
+
 static const struct component_ops vc4_v3d_ops = {
 	.bind   = vc4_v3d_bind,
 	.unbind = vc4_v3d_unbind,
@@ -245,6 +268,7 @@ static int vc4_v3d_dev_remove(struct platform_device *pdev)
 }
 
 static const struct of_device_id vc4_v3d_dt_match[] = {
+	{ .compatible = "brcm,bcm2835-v3d" },
 	{ .compatible = "brcm,vc4-v3d" },
 	{}
 };
@@ -255,5 +279,6 @@ struct platform_driver vc4_v3d_driver = {
 	.driver = {
 		.name = "vc4_v3d",
 		.of_match_table = vc4_v3d_dt_match,
+		.pm = &vc4_v3d_pm_ops,
 	},
 };
diff --git a/drivers/gpu/drm/vc4/vc4_validate.c b/drivers/gpu/drm/vc4/vc4_validate.c
index e26d9f6face3..24c2c746e8f3 100644
--- a/drivers/gpu/drm/vc4/vc4_validate.c
+++ b/drivers/gpu/drm/vc4/vc4_validate.c
@@ -401,8 +401,8 @@ validate_tile_binning_config(VALIDATE_ARGS)
 	tile_bo = vc4_bo_create(dev, exec->tile_alloc_offset + tile_alloc_size,
 				true);
 	exec->tile_bo = &tile_bo->base;
-	if (!exec->tile_bo)
-		return -ENOMEM;
+	if (IS_ERR(exec->tile_bo))
+		return PTR_ERR(exec->tile_bo);
 	list_add_tail(&tile_bo->unref_head, &exec->unref_list);
 
 	/* tile alloc address. */
diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_drv.c b/drivers/gpu/drm/vmwgfx/vmwgfx_drv.c
index 0ee76e523a90..6cbb7d4bdd11 100644
--- a/drivers/gpu/drm/vmwgfx/vmwgfx_drv.c
+++ b/drivers/gpu/drm/vmwgfx/vmwgfx_drv.c
@@ -195,7 +195,7 @@ static const struct drm_ioctl_desc vmw_ioctls[] = {
 		      DRM_MASTER | DRM_AUTH),
 	VMW_IOCTL_DEF(VMW_UPDATE_LAYOUT,
 		      vmw_kms_update_layout_ioctl,
-		      DRM_MASTER),
+		      DRM_MASTER | DRM_CONTROL_ALLOW),
 	VMW_IOCTL_DEF(VMW_CREATE_SHADER,
 		      vmw_shader_define_ioctl,
 		      DRM_AUTH | DRM_RENDER_ALLOW),
@@ -1204,6 +1204,7 @@ static int vmw_master_set(struct drm_device *dev,
 	}
 
 	dev_priv->active_master = vmaster;
+	drm_sysfs_hotplug_event(dev);
 
 	return 0;
 }
diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_drv.h b/drivers/gpu/drm/vmwgfx/vmwgfx_drv.h
index 5cb1b1687cd4..019a6ca3e8e9 100644
--- a/drivers/gpu/drm/vmwgfx/vmwgfx_drv.h
+++ b/drivers/gpu/drm/vmwgfx/vmwgfx_drv.h
@@ -40,9 +40,9 @@
 #include <drm/ttm/ttm_module.h>
 #include "vmwgfx_fence.h"
 
-#define VMWGFX_DRIVER_DATE "20150810"
+#define VMWGFX_DRIVER_DATE "20160210"
 #define VMWGFX_DRIVER_MAJOR 2
-#define VMWGFX_DRIVER_MINOR 9
+#define VMWGFX_DRIVER_MINOR 10
 #define VMWGFX_DRIVER_PATCHLEVEL 0
 #define VMWGFX_FILE_PAGE_OFFSET 0x00100000
 #define VMWGFX_FIFO_STATIC_SIZE (1024*1024)
@@ -407,8 +407,11 @@ struct vmw_private {
 	void *fb_info;
 	enum vmw_display_unit_type active_display_unit;
 	struct vmw_legacy_display *ldu_priv;
-	struct vmw_screen_object_display *sou_priv;
 	struct vmw_overlay *overlay_priv;
+	struct drm_property *hotplug_mode_update_property;
+	struct drm_property *implicit_placement_property;
+	unsigned num_implicit;
+	struct vmw_framebuffer *implicit_fb;
 
 	/*
 	 * Context and surface management.
diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_execbuf.c b/drivers/gpu/drm/vmwgfx/vmwgfx_execbuf.c
index 5da5de0cb522..723ba16c6084 100644
--- a/drivers/gpu/drm/vmwgfx/vmwgfx_execbuf.c
+++ b/drivers/gpu/drm/vmwgfx/vmwgfx_execbuf.c
@@ -3009,6 +3009,26 @@ out_unref:
 	return ret;
 }
 
+/**
+ * vmw_cmd_dx_genmips - Validate an SVGA_3D_CMD_DX_GENMIPS command
+ *
+ * @dev_priv: Pointer to a device private struct.
+ * @sw_context: The software context being used for this batch.
+ * @header: Pointer to the command header in the command stream.
+ */
+static int vmw_cmd_dx_genmips(struct vmw_private *dev_priv,
+			      struct vmw_sw_context *sw_context,
+			      SVGA3dCmdHeader *header)
+{
+	struct {
+		SVGA3dCmdHeader header;
+		SVGA3dCmdDXGenMips body;
+	} *cmd = container_of(header, typeof(*cmd), header);
+
+	return vmw_view_id_val_add(sw_context, vmw_view_sr,
+				   cmd->body.shaderResourceViewId);
+}
+
 static int vmw_cmd_check_not_3d(struct vmw_private *dev_priv,
 				struct vmw_sw_context *sw_context,
 				void *buf, uint32_t *size)
@@ -3297,7 +3317,7 @@ static const struct vmw_cmd_entry vmw_cmd_entries[SVGA_3D_CMD_MAX] = {
 		    &vmw_cmd_dx_clear_depthstencil_view, true, false, true),
 	VMW_CMD_DEF(SVGA_3D_CMD_DX_PRED_COPY, &vmw_cmd_invalid,
 		    true, false, true),
-	VMW_CMD_DEF(SVGA_3D_CMD_DX_GENMIPS, &vmw_cmd_invalid,
+	VMW_CMD_DEF(SVGA_3D_CMD_DX_GENMIPS, &vmw_cmd_dx_genmips,
 		    true, false, true),
 	VMW_CMD_DEF(SVGA_3D_CMD_DX_UPDATE_SUBRESOURCE,
 		    &vmw_cmd_dx_check_subresource, true, false, true),
diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_kms.c b/drivers/gpu/drm/vmwgfx/vmwgfx_kms.c
index b221a8c40282..4742ec4ead27 100644
--- a/drivers/gpu/drm/vmwgfx/vmwgfx_kms.c
+++ b/drivers/gpu/drm/vmwgfx/vmwgfx_kms.c
@@ -236,8 +236,8 @@ int vmw_du_crtc_cursor_move(struct drm_crtc *crtc, int x, int y)
 	struct vmw_display_unit *du = vmw_crtc_to_du(crtc);
 	bool shown = du->cursor_surface || du->cursor_dmabuf ? true : false;
 
-	du->cursor_x = x + crtc->x;
-	du->cursor_y = y + crtc->y;
+	du->cursor_x = x + du->set_gui_x;
+	du->cursor_y = y + du->set_gui_y;
 
 	/*
 	 * FIXME: Unclear whether there's any global state touched by the
@@ -663,9 +663,8 @@ static int vmw_framebuffer_dmabuf_dirty(struct drm_framebuffer *framebuffer,
 		break;
 	case vmw_du_screen_object:
 		ret = vmw_kms_sou_do_dmabuf_dirty(dev_priv, &vfbd->base,
-						  clips, num_clips, increment,
-						  true,
-						  NULL);
+						  clips, NULL, num_clips,
+						  increment, true, NULL);
 		break;
 	case vmw_du_legacy:
 		ret = vmw_kms_ldu_do_dmabuf_dirty(dev_priv, &vfbd->base, 0, 0,
@@ -1109,6 +1108,22 @@ int vmw_kms_present(struct vmw_private *dev_priv,
 	return 0;
 }
 
+static void
+vmw_kms_create_hotplug_mode_update_property(struct vmw_private *dev_priv)
+{
+	if (dev_priv->hotplug_mode_update_property)
+		return;
+
+	dev_priv->hotplug_mode_update_property =
+		drm_property_create_range(dev_priv->dev,
+					  DRM_MODE_PROP_IMMUTABLE,
+					  "hotplug_mode_update", 0, 1);
+
+	if (!dev_priv->hotplug_mode_update_property)
+		return;
+
+}
+
 int vmw_kms_init(struct vmw_private *dev_priv)
 {
 	struct drm_device *dev = dev_priv->dev;
@@ -1121,6 +1136,9 @@ int vmw_kms_init(struct vmw_private *dev_priv)
 	dev->mode_config.max_width = dev_priv->texture_max_width;
 	dev->mode_config.max_height = dev_priv->texture_max_height;
 
+	drm_mode_create_suggested_offset_properties(dev);
+	vmw_kms_create_hotplug_mode_update_property(dev_priv);
+
 	ret = vmw_kms_stdu_init_display(dev_priv);
 	if (ret) {
 		ret = vmw_kms_sou_init_display(dev_priv);
@@ -1360,15 +1378,28 @@ static int vmw_du_update_layout(struct vmw_private *dev_priv, unsigned num,
 			du->pref_active = true;
 			du->gui_x = rects[du->unit].x;
 			du->gui_y = rects[du->unit].y;
+			drm_object_property_set_value
+			  (&con->base, dev->mode_config.suggested_x_property,
+			   du->gui_x);
+			drm_object_property_set_value
+			  (&con->base, dev->mode_config.suggested_y_property,
+			   du->gui_y);
 		} else {
 			du->pref_width = 800;
 			du->pref_height = 600;
 			du->pref_active = false;
+			drm_object_property_set_value
+			  (&con->base, dev->mode_config.suggested_x_property,
+			   0);
+			drm_object_property_set_value
+			  (&con->base, dev->mode_config.suggested_y_property,
+			   0);
 		}
 		con->status = vmw_du_connector_detect(con, true);
 	}
 
 	mutex_unlock(&dev->mode_config.mutex);
+	drm_sysfs_hotplug_event(dev);
 
 	return 0;
 }
@@ -1591,6 +1622,12 @@ int vmw_du_connector_set_property(struct drm_connector *connector,
 				  struct drm_property *property,
 				  uint64_t val)
 {
+	struct vmw_display_unit *du = vmw_connector_to_du(connector);
+	struct vmw_private *dev_priv = vmw_priv(connector->dev);
+
+	if (property == dev_priv->implicit_placement_property)
+		du->is_implicit = val;
+
 	return 0;
 }
 
@@ -2096,3 +2133,119 @@ int vmw_kms_fbdev_init_data(struct vmw_private *dev_priv,
 
 	return 0;
 }
+
+/**
+ * vmw_kms_del_active - unregister a crtc binding to the implicit framebuffer
+ *
+ * @dev_priv: Pointer to a device private struct.
+ * @du: The display unit of the crtc.
+ */
+void vmw_kms_del_active(struct vmw_private *dev_priv,
+			struct vmw_display_unit *du)
+{
+	lockdep_assert_held_once(&dev_priv->dev->mode_config.mutex);
+
+	if (du->active_implicit) {
+		if (--(dev_priv->num_implicit) == 0)
+			dev_priv->implicit_fb = NULL;
+		du->active_implicit = false;
+	}
+}
+
+/**
+ * vmw_kms_add_active - register a crtc binding to an implicit framebuffer
+ *
+ * @vmw_priv: Pointer to a device private struct.
+ * @du: The display unit of the crtc.
+ * @vfb: The implicit framebuffer
+ *
+ * Registers a binding to an implicit framebuffer.
+ */
+void vmw_kms_add_active(struct vmw_private *dev_priv,
+			struct vmw_display_unit *du,
+			struct vmw_framebuffer *vfb)
+{
+	lockdep_assert_held_once(&dev_priv->dev->mode_config.mutex);
+
+	WARN_ON_ONCE(!dev_priv->num_implicit && dev_priv->implicit_fb);
+
+	if (!du->active_implicit && du->is_implicit) {
+		dev_priv->implicit_fb = vfb;
+		du->active_implicit = true;
+		dev_priv->num_implicit++;
+	}
+}
+
+/**
+ * vmw_kms_screen_object_flippable - Check whether we can page-flip a crtc.
+ *
+ * @dev_priv: Pointer to device-private struct.
+ * @crtc: The crtc we want to flip.
+ *
+ * Returns true or false depending whether it's OK to flip this crtc
+ * based on the criterion that we must not have more than one implicit
+ * frame-buffer at any one time.
+ */
+bool vmw_kms_crtc_flippable(struct vmw_private *dev_priv,
+			    struct drm_crtc *crtc)
+{
+	struct vmw_display_unit *du = vmw_crtc_to_du(crtc);
+
+	lockdep_assert_held_once(&dev_priv->dev->mode_config.mutex);
+
+	if (!du->is_implicit)
+		return true;
+
+	if (dev_priv->num_implicit != 1)
+		return false;
+
+	return true;
+}
+
+/**
+ * vmw_kms_update_implicit_fb - Update the implicit fb.
+ *
+ * @dev_priv: Pointer to device-private struct.
+ * @crtc: The crtc the new implicit frame-buffer is bound to.
+ */
+void vmw_kms_update_implicit_fb(struct vmw_private *dev_priv,
+				struct drm_crtc *crtc)
+{
+	struct vmw_display_unit *du = vmw_crtc_to_du(crtc);
+	struct vmw_framebuffer *vfb;
+
+	lockdep_assert_held_once(&dev_priv->dev->mode_config.mutex);
+
+	if (!du->is_implicit)
+		return;
+
+	vfb = vmw_framebuffer_to_vfb(crtc->primary->fb);
+	WARN_ON_ONCE(dev_priv->num_implicit != 1 &&
+		     dev_priv->implicit_fb != vfb);
+
+	dev_priv->implicit_fb = vfb;
+}
+
+/**
+ * vmw_kms_create_implicit_placement_proparty - Set up the implicit placement
+ * property.
+ *
+ * @dev_priv: Pointer to a device private struct.
+ * @immutable: Whether the property is immutable.
+ *
+ * Sets up the implicit placement property unless it's already set up.
+ */
+void
+vmw_kms_create_implicit_placement_property(struct vmw_private *dev_priv,
+					   bool immutable)
+{
+	if (dev_priv->implicit_placement_property)
+		return;
+
+	dev_priv->implicit_placement_property =
+		drm_property_create_range(dev_priv->dev,
+					  immutable ?
+					  DRM_MODE_PROP_IMMUTABLE : 0,
+					  "implicit_placement", 0, 1);
+
+}
diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_kms.h b/drivers/gpu/drm/vmwgfx/vmwgfx_kms.h
index edd81503516d..57203212c501 100644
--- a/drivers/gpu/drm/vmwgfx/vmwgfx_kms.h
+++ b/drivers/gpu/drm/vmwgfx/vmwgfx_kms.h
@@ -178,6 +178,9 @@ struct vmw_display_unit {
 	int gui_x;
 	int gui_y;
 	bool is_implicit;
+	bool active_implicit;
+	int set_gui_x;
+	int set_gui_y;
 };
 
 #define vmw_crtc_to_du(x) \
@@ -254,6 +257,18 @@ int vmw_kms_fbdev_init_data(struct vmw_private *dev_priv,
 			    struct drm_crtc **p_crtc,
 			    struct drm_display_mode **p_mode);
 void vmw_guess_mode_timing(struct drm_display_mode *mode);
+void vmw_kms_del_active(struct vmw_private *dev_priv,
+			struct vmw_display_unit *du);
+void vmw_kms_add_active(struct vmw_private *dev_priv,
+			struct vmw_display_unit *du,
+			struct vmw_framebuffer *vfb);
+bool vmw_kms_crtc_flippable(struct vmw_private *dev_priv,
+			    struct drm_crtc *crtc);
+void vmw_kms_update_implicit_fb(struct vmw_private *dev_priv,
+				struct drm_crtc *crtc);
+void vmw_kms_create_implicit_placement_property(struct vmw_private *dev_priv,
+						bool immutable);
+
 
 /*
  * Legacy display unit functions - vmwgfx_ldu.c
@@ -287,6 +302,7 @@ int vmw_kms_sou_do_surface_dirty(struct vmw_private *dev_priv,
 int vmw_kms_sou_do_dmabuf_dirty(struct vmw_private *dev_priv,
 				struct vmw_framebuffer *framebuffer,
 				struct drm_clip_rect *clips,
+				struct drm_vmw_rect *vclips,
 				unsigned num_clips, int increment,
 				bool interruptible,
 				struct vmw_fence_obj **out_fence);
diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_ldu.c b/drivers/gpu/drm/vmwgfx/vmwgfx_ldu.c
index b6fa44fe8929..63ccd9871ec9 100644
--- a/drivers/gpu/drm/vmwgfx/vmwgfx_ldu.c
+++ b/drivers/gpu/drm/vmwgfx/vmwgfx_ldu.c
@@ -288,6 +288,8 @@ static int vmw_ldu_crtc_set_config(struct drm_mode_set *set)
 	crtc->y = set->y;
 	crtc->mode = *mode;
 	crtc->enabled = true;
+	ldu->base.set_gui_x = set->x;
+	ldu->base.set_gui_y = set->y;
 
 	vmw_ldu_add_active(dev_priv, ldu, vfb);
 
@@ -375,8 +377,19 @@ static int vmw_ldu_init(struct vmw_private *dev_priv, unsigned unit)
 	drm_mode_crtc_set_gamma_size(crtc, 256);
 
 	drm_object_attach_property(&connector->base,
-				      dev->mode_config.dirty_info_property,
-				      1);
+				   dev->mode_config.dirty_info_property,
+				   1);
+	drm_object_attach_property(&connector->base,
+				   dev_priv->hotplug_mode_update_property, 1);
+	drm_object_attach_property(&connector->base,
+				   dev->mode_config.suggested_x_property, 0);
+	drm_object_attach_property(&connector->base,
+				   dev->mode_config.suggested_y_property, 0);
+	if (dev_priv->implicit_placement_property)
+		drm_object_attach_property
+			(&connector->base,
+			 dev_priv->implicit_placement_property,
+			 1);
 
 	return 0;
 }
@@ -412,6 +425,8 @@ int vmw_kms_ldu_init_display(struct vmw_private *dev_priv)
 	if (ret != 0)
 		goto err_vblank_cleanup;
 
+	vmw_kms_create_implicit_placement_property(dev_priv, true);
+
 	if (dev_priv->capabilities & SVGA_CAP_MULTIMON)
 		for (i = 0; i < VMWGFX_NUM_DISPLAY_UNITS; ++i)
 			vmw_ldu_init(dev_priv, i);
diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_scrn.c b/drivers/gpu/drm/vmwgfx/vmwgfx_scrn.c
index db082bea8daf..0ea22fd112c9 100644
--- a/drivers/gpu/drm/vmwgfx/vmwgfx_scrn.c
+++ b/drivers/gpu/drm/vmwgfx/vmwgfx_scrn.c
@@ -74,19 +74,6 @@ struct vmw_kms_sou_dirty_cmd {
 	SVGA3dCmdBlitSurfaceToScreen body;
 };
 
-
-/*
- * Other structs.
- */
-
-struct vmw_screen_object_display {
-	unsigned num_implicit;
-
-	struct vmw_framebuffer *implicit_fb;
-	SVGAFifoCmdDefineGMRFB cur;
-	struct vmw_dma_buffer *pinned_gmrfb;
-};
-
 /**
  * Display unit using screen objects.
  */
@@ -97,7 +84,6 @@ struct vmw_screen_object_unit {
 	struct vmw_dma_buffer *buffer; /**< Backing store buffer */
 
 	bool defined;
-	bool active_implicit;
 };
 
 static void vmw_sou_destroy(struct vmw_screen_object_unit *sou)
@@ -116,33 +102,6 @@ static void vmw_sou_crtc_destroy(struct drm_crtc *crtc)
 	vmw_sou_destroy(vmw_crtc_to_sou(crtc));
 }
 
-static void vmw_sou_del_active(struct vmw_private *vmw_priv,
-			       struct vmw_screen_object_unit *sou)
-{
-	struct vmw_screen_object_display *ld = vmw_priv->sou_priv;
-
-	if (sou->active_implicit) {
-		if (--(ld->num_implicit) == 0)
-			ld->implicit_fb = NULL;
-		sou->active_implicit = false;
-	}
-}
-
-static void vmw_sou_add_active(struct vmw_private *vmw_priv,
-			       struct vmw_screen_object_unit *sou,
-			       struct vmw_framebuffer *vfb)
-{
-	struct vmw_screen_object_display *ld = vmw_priv->sou_priv;
-
-	BUG_ON(!ld->num_implicit && ld->implicit_fb);
-
-	if (!sou->active_implicit && sou->base.is_implicit) {
-		ld->implicit_fb = vfb;
-		sou->active_implicit = true;
-		ld->num_implicit++;
-	}
-}
-
 /**
  * Send the fifo command to create a screen.
  */
@@ -185,6 +144,8 @@ static int vmw_sou_fifo_create(struct vmw_private *dev_priv,
 		cmd->obj.root.x = sou->base.gui_x;
 		cmd->obj.root.y = sou->base.gui_y;
 	}
+	sou->base.set_gui_x = cmd->obj.root.x;
+	sou->base.set_gui_y = cmd->obj.root.y;
 
 	/* Ok to assume that buffer is pinned in vram */
 	vmw_bo_get_guest_ptr(&sou->buffer->base, &cmd->obj.backingStore.ptr);
@@ -323,13 +284,13 @@ static int vmw_sou_crtc_set_config(struct drm_mode_set *set)
 		return -EINVAL;
 	}
 
-	/* sou only supports one fb active at the time */
+	/* Only one active implicit frame-buffer at a time. */
 	if (sou->base.is_implicit &&
-	    dev_priv->sou_priv->implicit_fb && vfb &&
-	    !(dev_priv->sou_priv->num_implicit == 1 &&
-	      sou->active_implicit) &&
-	    dev_priv->sou_priv->implicit_fb != vfb) {
-		DRM_ERROR("Multiple framebuffers not supported\n");
+	    dev_priv->implicit_fb && vfb &&
+	    !(dev_priv->num_implicit == 1 &&
+	      sou->base.active_implicit) &&
+	    dev_priv->implicit_fb != vfb) {
+		DRM_ERROR("Multiple implicit framebuffers not supported.\n");
 		return -EINVAL;
 	}
 
@@ -351,7 +312,7 @@ static int vmw_sou_crtc_set_config(struct drm_mode_set *set)
 		crtc->y = 0;
 		crtc->enabled = false;
 
-		vmw_sou_del_active(dev_priv, sou);
+		vmw_kms_del_active(dev_priv, &sou->base);
 
 		vmw_sou_backing_free(dev_priv, sou);
 
@@ -415,7 +376,7 @@ static int vmw_sou_crtc_set_config(struct drm_mode_set *set)
 		return ret;
 	}
 
-	vmw_sou_add_active(dev_priv, sou, vfb);
+	vmw_kms_add_active(dev_priv, &sou->base, vfb);
 
 	connector->encoder = encoder;
 	encoder->crtc = crtc;
@@ -428,39 +389,6 @@ static int vmw_sou_crtc_set_config(struct drm_mode_set *set)
 	return 0;
 }
 
-/**
- * Returns if this unit can be page flipped.
- * Must be called with the mode_config mutex held.
- */
-static bool vmw_sou_screen_object_flippable(struct vmw_private *dev_priv,
-					    struct drm_crtc *crtc)
-{
-	struct vmw_screen_object_unit *sou = vmw_crtc_to_sou(crtc);
-
-	if (!sou->base.is_implicit)
-		return true;
-
-	if (dev_priv->sou_priv->num_implicit != 1)
-		return false;
-
-	return true;
-}
-
-/**
- * Update the implicit fb to the current fb of this crtc.
- * Must be called with the mode_config mutex held.
- */
-static void vmw_sou_update_implicit_fb(struct vmw_private *dev_priv,
-				       struct drm_crtc *crtc)
-{
-	struct vmw_screen_object_unit *sou = vmw_crtc_to_sou(crtc);
-
-	BUG_ON(!sou->base.is_implicit);
-
-	dev_priv->sou_priv->implicit_fb =
-		vmw_framebuffer_to_vfb(sou->base.crtc.primary->fb);
-}
-
 static int vmw_sou_crtc_page_flip(struct drm_crtc *crtc,
 				  struct drm_framebuffer *fb,
 				  struct drm_pending_vblank_event *event,
@@ -470,30 +398,27 @@ static int vmw_sou_crtc_page_flip(struct drm_crtc *crtc,
 	struct drm_framebuffer *old_fb = crtc->primary->fb;
 	struct vmw_framebuffer *vfb = vmw_framebuffer_to_vfb(fb);
 	struct vmw_fence_obj *fence = NULL;
-	struct drm_clip_rect clips;
+	struct drm_vmw_rect vclips;
 	int ret;
 
-	/* require ScreenObject support for page flipping */
-	if (!dev_priv->sou_priv)
-		return -ENOSYS;
-
-	if (!vmw_sou_screen_object_flippable(dev_priv, crtc))
+	if (!vmw_kms_crtc_flippable(dev_priv, crtc))
 		return -EINVAL;
 
 	crtc->primary->fb = fb;
 
 	/* do a full screen dirty update */
-	clips.x1 = clips.y1 = 0;
-	clips.x2 = fb->width;
-	clips.y2 = fb->height;
+	vclips.x = crtc->x;
+	vclips.y = crtc->y;
+	vclips.w = crtc->mode.hdisplay;
+	vclips.h = crtc->mode.vdisplay;
 
 	if (vfb->dmabuf)
 		ret = vmw_kms_sou_do_dmabuf_dirty(dev_priv, vfb,
-						  &clips, 1, 1,
+						  NULL, &vclips, 1, 1,
 						  true, &fence);
 	else
 		ret = vmw_kms_sou_do_surface_dirty(dev_priv, vfb,
-						   &clips, NULL, NULL,
+						   NULL, &vclips, NULL,
 						   0, 0, 1, 1, &fence);
 
 
@@ -521,7 +446,7 @@ static int vmw_sou_crtc_page_flip(struct drm_crtc *crtc,
 	vmw_fence_obj_unreference(&fence);
 
 	if (vmw_crtc_to_du(crtc)->is_implicit)
-		vmw_sou_update_implicit_fb(dev_priv, crtc);
+		vmw_kms_update_implicit_fb(dev_priv, crtc);
 
 	return ret;
 
@@ -563,6 +488,8 @@ static void vmw_sou_connector_destroy(struct drm_connector *connector)
 
 static const struct drm_connector_funcs vmw_sou_connector_funcs = {
 	.dpms = vmw_du_connector_dpms,
+	.detect = vmw_du_connector_detect,
+	.fill_modes = vmw_du_connector_fill_modes,
 	.set_property = vmw_du_connector_set_property,
 	.destroy = vmw_sou_connector_destroy,
 };
@@ -584,13 +511,12 @@ static int vmw_sou_init(struct vmw_private *dev_priv, unsigned unit)
 	encoder = &sou->base.encoder;
 	connector = &sou->base.connector;
 
-	sou->active_implicit = false;
-
+	sou->base.active_implicit = false;
 	sou->base.pref_active = (unit == 0);
 	sou->base.pref_width = dev_priv->initial_width;
 	sou->base.pref_height = dev_priv->initial_height;
 	sou->base.pref_mode = NULL;
-	sou->base.is_implicit = true;
+	sou->base.is_implicit = false;
 
 	drm_connector_init(dev, connector, &vmw_sou_connector_funcs,
 			   DRM_MODE_CONNECTOR_VIRTUAL);
@@ -609,8 +535,19 @@ static int vmw_sou_init(struct vmw_private *dev_priv, unsigned unit)
 	drm_mode_crtc_set_gamma_size(crtc, 256);
 
 	drm_object_attach_property(&connector->base,
-				      dev->mode_config.dirty_info_property,
-				      1);
+				   dev->mode_config.dirty_info_property,
+				   1);
+	drm_object_attach_property(&connector->base,
+				   dev_priv->hotplug_mode_update_property, 1);
+	drm_object_attach_property(&connector->base,
+				   dev->mode_config.suggested_x_property, 0);
+	drm_object_attach_property(&connector->base,
+				   dev->mode_config.suggested_y_property, 0);
+	if (dev_priv->implicit_placement_property)
+		drm_object_attach_property
+			(&connector->base,
+			 dev_priv->implicit_placement_property,
+			 sou->base.is_implicit);
 
 	return 0;
 }
@@ -620,11 +557,6 @@ int vmw_kms_sou_init_display(struct vmw_private *dev_priv)
 	struct drm_device *dev = dev_priv->dev;
 	int i, ret;
 
-	if (dev_priv->sou_priv) {
-		DRM_INFO("sou system already on\n");
-		return -EINVAL;
-	}
-
 	if (!(dev_priv->capabilities & SVGA_CAP_SCREEN_OBJECT_2)) {
 		DRM_INFO("Not using screen objects,"
 			 " missing cap SCREEN_OBJECT_2\n");
@@ -632,21 +564,19 @@ int vmw_kms_sou_init_display(struct vmw_private *dev_priv)
 	}
 
 	ret = -ENOMEM;
-	dev_priv->sou_priv = kmalloc(sizeof(*dev_priv->sou_priv), GFP_KERNEL);
-	if (unlikely(!dev_priv->sou_priv))
-		goto err_no_mem;
-
-	dev_priv->sou_priv->num_implicit = 0;
-	dev_priv->sou_priv->implicit_fb = NULL;
+	dev_priv->num_implicit = 0;
+	dev_priv->implicit_fb = NULL;
 
 	ret = drm_vblank_init(dev, VMWGFX_NUM_DISPLAY_UNITS);
 	if (unlikely(ret != 0))
-		goto err_free;
+		return ret;
 
 	ret = drm_mode_create_dirty_info_property(dev);
 	if (unlikely(ret != 0))
 		goto err_vblank_cleanup;
 
+	vmw_kms_create_implicit_placement_property(dev_priv, false);
+
 	for (i = 0; i < VMWGFX_NUM_DISPLAY_UNITS; ++i)
 		vmw_sou_init(dev_priv, i);
 
@@ -658,10 +588,6 @@ int vmw_kms_sou_init_display(struct vmw_private *dev_priv)
 
 err_vblank_cleanup:
 	drm_vblank_cleanup(dev);
-err_free:
-	kfree(dev_priv->sou_priv);
-	dev_priv->sou_priv = NULL;
-err_no_mem:
 	return ret;
 }
 
@@ -669,13 +595,8 @@ int vmw_kms_sou_close_display(struct vmw_private *dev_priv)
 {
 	struct drm_device *dev = dev_priv->dev;
 
-	if (!dev_priv->sou_priv)
-		return -ENOSYS;
-
 	drm_vblank_cleanup(dev);
 
-	kfree(dev_priv->sou_priv);
-
 	return 0;
 }
 
@@ -736,6 +657,11 @@ static void vmw_sou_surface_fifo_commit(struct vmw_kms_dirty *dirty)
 	SVGASignedRect *blit = (SVGASignedRect *) &cmd[1];
 	int i;
 
+	if (!dirty->num_hits) {
+		vmw_fifo_commit(dirty->dev_priv, 0);
+		return;
+	}
+
 	cmd->header.id = SVGA_3D_CMD_BLIT_SURFACE_TO_SCREEN;
 	cmd->header.size = sizeof(cmd->body) + region_size;
 
@@ -873,6 +799,11 @@ int vmw_kms_sou_do_surface_dirty(struct vmw_private *dev_priv,
  */
 static void vmw_sou_dmabuf_fifo_commit(struct vmw_kms_dirty *dirty)
 {
+	if (!dirty->num_hits) {
+		vmw_fifo_commit(dirty->dev_priv, 0);
+		return;
+	}
+
 	vmw_fifo_commit(dirty->dev_priv,
 			sizeof(struct vmw_kms_sou_dmabuf_blit) *
 			dirty->num_hits);
@@ -907,6 +838,8 @@ static void vmw_sou_dmabuf_clip(struct vmw_kms_dirty *dirty)
  * @dev_priv: Pointer to the device private structure.
  * @framebuffer: Pointer to the dma-buffer backed framebuffer.
  * @clips: Array of clip rects.
+ * @vclips: Alternate array of clip rects. Either @clips or @vclips must
+ * be NULL.
  * @num_clips: Number of clip rects in @clips.
  * @increment: Increment to use when looping over @clips.
  * @interruptible: Whether to perform waits interruptible if possible.
@@ -920,6 +853,7 @@ static void vmw_sou_dmabuf_clip(struct vmw_kms_dirty *dirty)
 int vmw_kms_sou_do_dmabuf_dirty(struct vmw_private *dev_priv,
 				struct vmw_framebuffer *framebuffer,
 				struct drm_clip_rect *clips,
+				struct drm_vmw_rect *vclips,
 				unsigned num_clips, int increment,
 				bool interruptible,
 				struct vmw_fence_obj **out_fence)
@@ -943,7 +877,7 @@ int vmw_kms_sou_do_dmabuf_dirty(struct vmw_private *dev_priv,
 	dirty.clip = vmw_sou_dmabuf_clip;
 	dirty.fifo_reserve_size = sizeof(struct vmw_kms_sou_dmabuf_blit) *
 		num_clips;
-	ret = vmw_kms_helper_dirty(dev_priv, framebuffer, clips, NULL,
+	ret = vmw_kms_helper_dirty(dev_priv, framebuffer, clips, vclips,
 				   0, 0, num_clips, increment, &dirty);
 	vmw_kms_helper_buffer_finish(dev_priv, NULL, buf, out_fence, NULL);
 
@@ -965,6 +899,11 @@ out_revert:
  */
 static void vmw_sou_readback_fifo_commit(struct vmw_kms_dirty *dirty)
 {
+	if (!dirty->num_hits) {
+		vmw_fifo_commit(dirty->dev_priv, 0);
+		return;
+	}
+
 	vmw_fifo_commit(dirty->dev_priv,
 			sizeof(struct vmw_kms_sou_readback_blit) *
 			dirty->num_hits);
diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_stdu.c b/drivers/gpu/drm/vmwgfx/vmwgfx_stdu.c
index 4ef5ffd7189d..b949102ad864 100644
--- a/drivers/gpu/drm/vmwgfx/vmwgfx_stdu.c
+++ b/drivers/gpu/drm/vmwgfx/vmwgfx_stdu.c
@@ -96,7 +96,6 @@ struct vmw_stdu_surface_copy {
  *               content_vfbs dimensions, then this is a pointer into the
  *               corresponding field in content_vfbs.  If not, then this
  *               is a separate buffer to which content_vfbs will blit to.
- * @content_fb: holds the rendered content, can be a surface or DMA buffer
  * @content_type:  content_fb type
  * @defined:  true if the current display unit has been initialized
  */
@@ -104,8 +103,6 @@ struct vmw_screen_target_display_unit {
 	struct vmw_display_unit base;
 
 	struct vmw_surface     *display_srf;
-	struct drm_framebuffer *content_fb;
-
 	enum stdu_content_type content_fb_type;
 
 	bool defined;
@@ -122,22 +119,6 @@ static void vmw_stdu_destroy(struct vmw_screen_target_display_unit *stdu);
  *****************************************************************************/
 
 /**
- * vmw_stdu_pin_display - pins the resource associated with the display surface
- *
- * @stdu: contains the display surface
- *
- * Since the display surface can either be a private surface allocated by us,
- * or it can point to the content surface, we use this function to not pin the
- * same resource twice.
- */
-static int vmw_stdu_pin_display(struct vmw_screen_target_display_unit *stdu)
-{
-	return vmw_resource_pin(&stdu->display_srf->res, false);
-}
-
-
-
-/**
  * vmw_stdu_unpin_display - unpins the resource associated with display surface
  *
  * @stdu: contains the display surface
@@ -153,13 +134,7 @@ static void vmw_stdu_unpin_display(struct vmw_screen_target_display_unit *stdu)
 		struct vmw_resource *res = &stdu->display_srf->res;
 
 		vmw_resource_unpin(res);
-
-		if (stdu->content_fb_type != SAME_AS_DISPLAY) {
-			vmw_resource_unreference(&res);
-			stdu->content_fb_type = SAME_AS_DISPLAY;
-		}
-
-		stdu->display_srf = NULL;
+		vmw_surface_unreference(&stdu->display_srf);
 	}
 }
 
@@ -185,6 +160,9 @@ static void vmw_stdu_crtc_destroy(struct drm_crtc *crtc)
  *
  * @dev_priv:  VMW DRM device
  * @stdu: display unit to create a Screen Target for
+ * @mode: The mode to set.
+ * @crtc_x: X coordinate of screen target relative to framebuffer origin.
+ * @crtc_y: Y coordinate of screen target relative to framebuffer origin.
  *
  * Creates a STDU that we can used later.  This function is called whenever the
  * framebuffer size changes.
@@ -193,7 +171,9 @@ static void vmw_stdu_crtc_destroy(struct drm_crtc *crtc)
  * 0 on success, error code on failure
  */
 static int vmw_stdu_define_st(struct vmw_private *dev_priv,
-			      struct vmw_screen_target_display_unit *stdu)
+			      struct vmw_screen_target_display_unit *stdu,
+			      struct drm_display_mode *mode,
+			      int crtc_x, int crtc_y)
 {
 	struct {
 		SVGA3dCmdHeader header;
@@ -211,17 +191,19 @@ static int vmw_stdu_define_st(struct vmw_private *dev_priv,
 	cmd->header.size = sizeof(cmd->body);
 
 	cmd->body.stid   = stdu->base.unit;
-	cmd->body.width  = stdu->display_srf->base_size.width;
-	cmd->body.height = stdu->display_srf->base_size.height;
+	cmd->body.width  = mode->hdisplay;
+	cmd->body.height = mode->vdisplay;
 	cmd->body.flags  = (0 == cmd->body.stid) ? SVGA_STFLAG_PRIMARY : 0;
 	cmd->body.dpi    = 0;
-	cmd->body.xRoot  = stdu->base.crtc.x;
-	cmd->body.yRoot  = stdu->base.crtc.y;
-
-	if (!stdu->base.is_implicit) {
+	if (stdu->base.is_implicit) {
+		cmd->body.xRoot  = crtc_x;
+		cmd->body.yRoot  = crtc_y;
+	} else {
 		cmd->body.xRoot  = stdu->base.gui_x;
 		cmd->body.yRoot  = stdu->base.gui_y;
 	}
+	stdu->base.set_gui_x = cmd->body.xRoot;
+	stdu->base.set_gui_y = cmd->body.yRoot;
 
 	vmw_fifo_commit(dev_priv, sizeof(*cmd));
 
@@ -392,126 +374,43 @@ static int vmw_stdu_destroy_st(struct vmw_private *dev_priv,
 	return ret;
 }
 
-
-
 /**
- * vmw_stdu_crtc_set_config - Sets a mode
+ * vmw_stdu_bind_fb - Bind an fb to a defined screen target
  *
- * @set:  mode parameters
- *
- * This function is the device-specific portion of the DRM CRTC mode set.
- * For the SVGA device, we do this by defining a Screen Target, binding a
- * GB Surface to that target, and finally update the screen target.
+ * @dev_priv: Pointer to a device private struct.
+ * @crtc: The crtc holding the screen target.
+ * @mode: The mode currently used by the screen target. Must be non-NULL.
+ * @new_fb: The new framebuffer to bind. Must be non-NULL.
  *
  * RETURNS:
- * 0 on success, error code otherwise
+ * 0 on success, error code on failure.
  */
-static int vmw_stdu_crtc_set_config(struct drm_mode_set *set)
+static int vmw_stdu_bind_fb(struct vmw_private *dev_priv,
+			    struct drm_crtc *crtc,
+			    struct drm_display_mode *mode,
+			    struct drm_framebuffer *new_fb)
 {
-	struct vmw_private *dev_priv;
-	struct vmw_screen_target_display_unit *stdu;
-	struct vmw_framebuffer *vfb;
+	struct vmw_screen_target_display_unit *stdu = vmw_crtc_to_stdu(crtc);
+	struct vmw_framebuffer *vfb = vmw_framebuffer_to_vfb(new_fb);
+	struct vmw_surface *new_display_srf = NULL;
+	enum stdu_content_type new_content_type;
 	struct vmw_framebuffer_surface *new_vfbs;
-	struct drm_display_mode *mode;
-	struct drm_framebuffer  *new_fb;
-	struct drm_crtc      *crtc;
-	struct drm_encoder   *encoder;
-	struct drm_connector *connector;
-	int    ret;
-
-
-	if (!set || !set->crtc)
-		return -EINVAL;
-
-	crtc     = set->crtc;
-	crtc->x  = set->x;
-	crtc->y  = set->y;
-	stdu     = vmw_crtc_to_stdu(crtc);
-	mode     = set->mode;
-	new_fb   = set->fb;
-	dev_priv = vmw_priv(crtc->dev);
-
-
-	if (set->num_connectors > 1) {
-		DRM_ERROR("Too many connectors\n");
-		return -EINVAL;
-	}
-
-	if (set->num_connectors == 1 &&
-	    set->connectors[0] != &stdu->base.connector) {
-		DRM_ERROR("Connectors don't match %p %p\n",
-			set->connectors[0], &stdu->base.connector);
-		return -EINVAL;
-	}
-
-
-	/* Since they always map one to one these are safe */
-	connector = &stdu->base.connector;
-	encoder   = &stdu->base.encoder;
-
-
-	/*
-	 * After this point the CRTC will be considered off unless a new fb
-	 * is bound
-	 */
-	if (stdu->defined) {
-		/* Unbind current surface by binding an invalid one */
-		ret = vmw_stdu_bind_st(dev_priv, stdu, NULL);
-		if (unlikely(ret != 0))
-			return ret;
-
-		/* Update Screen Target, display will now be blank */
-		if (crtc->primary->fb) {
-			vmw_stdu_update_st(dev_priv, stdu);
-			if (unlikely(ret != 0))
-				return ret;
-		}
-
-		crtc->primary->fb  = NULL;
-		crtc->enabled      = false;
-		encoder->crtc      = NULL;
-		connector->encoder = NULL;
-
-		vmw_stdu_unpin_display(stdu);
-		stdu->content_fb      = NULL;
-		stdu->content_fb_type = SAME_AS_DISPLAY;
-
-		ret = vmw_stdu_destroy_st(dev_priv, stdu);
-		/* The hardware is hung, give up */
-		if (unlikely(ret != 0))
-			return ret;
-	}
-
-
-	/* Any of these conditions means the caller wants CRTC off */
-	if (set->num_connectors == 0 || !mode || !new_fb)
-		return 0;
-
-
-	if (set->x + mode->hdisplay > new_fb->width ||
-	    set->y + mode->vdisplay > new_fb->height) {
-		DRM_ERROR("Set outside of framebuffer\n");
-		return -EINVAL;
-	}
+	int ret;
 
-	stdu->content_fb = new_fb;
-	vfb = vmw_framebuffer_to_vfb(stdu->content_fb);
+	WARN_ON_ONCE(!stdu->defined);
 
-	if (vfb->dmabuf)
-		stdu->content_fb_type = SEPARATE_DMA;
+	if (!vfb->dmabuf && new_fb->width == mode->hdisplay &&
+	    new_fb->height == mode->vdisplay)
+		new_content_type = SAME_AS_DISPLAY;
+	else if (vfb->dmabuf)
+		new_content_type = SEPARATE_DMA;
+	else
+		new_content_type = SEPARATE_SURFACE;
 
-	/*
-	 * If the requested mode is different than the width and height
-	 * of the FB or if the content buffer is a DMA buf, then allocate
-	 * a display FB that matches the dimension of the mode
-	 */
-	if (mode->hdisplay != new_fb->width  ||
-	    mode->vdisplay != new_fb->height ||
-	    stdu->content_fb_type != SAME_AS_DISPLAY) {
+	if (new_content_type != SAME_AS_DISPLAY &&
+	    !stdu->display_srf) {
 		struct vmw_surface content_srf;
 		struct drm_vmw_size display_base_size = {0};
-		struct vmw_surface *display_srf;
-
 
 		display_base_size.width  = mode->hdisplay;
 		display_base_size.height = mode->vdisplay;
@@ -521,7 +420,7 @@ static int vmw_stdu_crtc_set_config(struct drm_mode_set *set)
 		 * If content buffer is a DMA buf, then we have to construct
 		 * surface info
 		 */
-		if (stdu->content_fb_type == SEPARATE_DMA) {
+		if (new_content_type == SEPARATE_DMA) {
 
 			switch (new_fb->bits_per_pixel) {
 			case 32:
@@ -538,17 +437,13 @@ static int vmw_stdu_crtc_set_config(struct drm_mode_set *set)
 
 			default:
 				DRM_ERROR("Invalid format\n");
-				ret = -EINVAL;
-				goto err_unref_content;
+				return -EINVAL;
 			}
 
 			content_srf.flags             = 0;
 			content_srf.mip_levels[0]     = 1;
 			content_srf.multisample_count = 0;
 		} else {
-
-			stdu->content_fb_type = SEPARATE_SURFACE;
-
 			new_vfbs = vmw_framebuffer_to_vfbs(new_fb);
 			content_srf = *new_vfbs->surface;
 		}
@@ -563,26 +458,136 @@ static int vmw_stdu_crtc_set_config(struct drm_mode_set *set)
 				content_srf.multisample_count,
 				0,
 				display_base_size,
-				&display_srf);
+				&new_display_srf);
 		if (unlikely(ret != 0)) {
-			DRM_ERROR("Cannot allocate a display FB.\n");
-			goto err_unref_content;
+			DRM_ERROR("Could not allocate screen target surface.\n");
+			return ret;
 		}
-
-		stdu->display_srf = display_srf;
-	} else {
+	} else if (new_content_type == SAME_AS_DISPLAY) {
 		new_vfbs = vmw_framebuffer_to_vfbs(new_fb);
-		stdu->display_srf = new_vfbs->surface;
+		new_display_srf = vmw_surface_reference(new_vfbs->surface);
 	}
 
+	if (new_display_srf) {
+		/* Pin new surface before flipping */
+		ret = vmw_resource_pin(&new_display_srf->res, false);
+		if (ret)
+			goto out_srf_unref;
+
+		ret = vmw_stdu_bind_st(dev_priv, stdu, &new_display_srf->res);
+		if (ret)
+			goto out_srf_unpin;
+
+		/* Unpin and unreference old surface */
+		vmw_stdu_unpin_display(stdu);
 
-	ret = vmw_stdu_pin_display(stdu);
-	if (unlikely(ret != 0)) {
-		stdu->display_srf = NULL;
-		goto err_unref_content;
+		/* Transfer the reference */
+		stdu->display_srf = new_display_srf;
+		new_display_srf = NULL;
 	}
 
-	vmw_svga_enable(dev_priv);
+	crtc->primary->fb = new_fb;
+	stdu->content_fb_type = new_content_type;
+	return 0;
+
+out_srf_unpin:
+	vmw_resource_unpin(&new_display_srf->res);
+out_srf_unref:
+	vmw_surface_unreference(&new_display_srf);
+	return ret;
+}
+
+/**
+ * vmw_stdu_crtc_set_config - Sets a mode
+ *
+ * @set:  mode parameters
+ *
+ * This function is the device-specific portion of the DRM CRTC mode set.
+ * For the SVGA device, we do this by defining a Screen Target, binding a
+ * GB Surface to that target, and finally update the screen target.
+ *
+ * RETURNS:
+ * 0 on success, error code otherwise
+ */
+static int vmw_stdu_crtc_set_config(struct drm_mode_set *set)
+{
+	struct vmw_private *dev_priv;
+	struct vmw_framebuffer *vfb;
+	struct vmw_screen_target_display_unit *stdu;
+	struct drm_display_mode *mode;
+	struct drm_framebuffer  *new_fb;
+	struct drm_crtc      *crtc;
+	struct drm_encoder   *encoder;
+	struct drm_connector *connector;
+	bool turning_off;
+	int    ret;
+
+
+	if (!set || !set->crtc)
+		return -EINVAL;
+
+	crtc     = set->crtc;
+	stdu     = vmw_crtc_to_stdu(crtc);
+	mode     = set->mode;
+	new_fb   = set->fb;
+	dev_priv = vmw_priv(crtc->dev);
+	turning_off = set->num_connectors == 0 || !mode || !new_fb;
+	vfb = (new_fb) ? vmw_framebuffer_to_vfb(new_fb) : NULL;
+
+	if (set->num_connectors > 1) {
+		DRM_ERROR("Too many connectors\n");
+		return -EINVAL;
+	}
+
+	if (set->num_connectors == 1 &&
+	    set->connectors[0] != &stdu->base.connector) {
+		DRM_ERROR("Connectors don't match %p %p\n",
+			set->connectors[0], &stdu->base.connector);
+		return -EINVAL;
+	}
+
+	if (!turning_off && (set->x + mode->hdisplay > new_fb->width ||
+			     set->y + mode->vdisplay > new_fb->height)) {
+		DRM_ERROR("Set outside of framebuffer\n");
+		return -EINVAL;
+	}
+
+	/* Only one active implicit frame-buffer at a time. */
+	if (!turning_off && stdu->base.is_implicit && dev_priv->implicit_fb &&
+	    !(dev_priv->num_implicit == 1 && stdu->base.active_implicit)
+	    && dev_priv->implicit_fb != vfb) {
+		DRM_ERROR("Multiple implicit framebuffers not supported.\n");
+		return -EINVAL;
+	}
+
+	/* Since they always map one to one these are safe */
+	connector = &stdu->base.connector;
+	encoder   = &stdu->base.encoder;
+
+	if (stdu->defined) {
+		ret = vmw_stdu_bind_st(dev_priv, stdu, NULL);
+		if (ret)
+			return ret;
+
+		vmw_stdu_unpin_display(stdu);
+		(void) vmw_stdu_update_st(dev_priv, stdu);
+		vmw_kms_del_active(dev_priv, &stdu->base);
+
+		ret = vmw_stdu_destroy_st(dev_priv, stdu);
+		if (ret)
+			return ret;
+
+		crtc->primary->fb = NULL;
+		crtc->enabled = false;
+		encoder->crtc = NULL;
+		connector->encoder = NULL;
+		stdu->content_fb_type = SAME_AS_DISPLAY;
+		crtc->x = set->x;
+		crtc->y = set->y;
+	}
+
+	if (turning_off)
+		return 0;
 
 	/*
 	 * Steps to displaying a surface, assume surface is already
@@ -592,35 +597,33 @@ static int vmw_stdu_crtc_set_config(struct drm_mode_set *set)
 	 *   3.  update that screen target (this is done later by
 	 *       vmw_kms_stdu_do_surface_dirty_or_present)
 	 */
-	ret = vmw_stdu_define_st(dev_priv, stdu);
-	if (unlikely(ret != 0))
-		goto err_unpin_display_and_content;
+	/*
+	 * Note on error handling: We can't really restore the crtc to
+	 * it's original state on error, but we at least update the
+	 * current state to what's submitted to hardware to enable
+	 * future recovery.
+	 */
+	vmw_svga_enable(dev_priv);
+	ret = vmw_stdu_define_st(dev_priv, stdu, mode, set->x, set->y);
+	if (ret)
+		return ret;
 
-	ret = vmw_stdu_bind_st(dev_priv, stdu, &stdu->display_srf->res);
-	if (unlikely(ret != 0))
-		goto err_unpin_destroy_st;
+	crtc->x = set->x;
+	crtc->y = set->y;
+	crtc->mode = *mode;
 
+	ret = vmw_stdu_bind_fb(dev_priv, crtc, mode, new_fb);
+	if (ret)
+		return ret;
 
+	vmw_kms_add_active(dev_priv, &stdu->base, vfb);
+	crtc->enabled = true;
 	connector->encoder = encoder;
 	encoder->crtc      = crtc;
 
-	crtc->mode    = *mode;
-	crtc->primary->fb = new_fb;
-	crtc->enabled = true;
-
-	return ret;
-
-err_unpin_destroy_st:
-	vmw_stdu_destroy_st(dev_priv, stdu);
-err_unpin_display_and_content:
-	vmw_stdu_unpin_display(stdu);
-err_unref_content:
-	stdu->content_fb = NULL;
-	return ret;
+	return 0;
 }
 
-
-
 /**
  * vmw_stdu_crtc_page_flip - Binds a buffer to a screen target
  *
@@ -648,59 +651,34 @@ static int vmw_stdu_crtc_page_flip(struct drm_crtc *crtc,
 {
 	struct vmw_private *dev_priv = vmw_priv(crtc->dev);
 	struct vmw_screen_target_display_unit *stdu;
+	struct drm_vmw_rect vclips;
+	struct vmw_framebuffer *vfb = vmw_framebuffer_to_vfb(new_fb);
 	int ret;
 
-	if (crtc == NULL)
-		return -EINVAL;
-
 	dev_priv          = vmw_priv(crtc->dev);
 	stdu              = vmw_crtc_to_stdu(crtc);
-	crtc->primary->fb = new_fb;
-	stdu->content_fb  = new_fb;
-
-	if (stdu->display_srf) {
-		/*
-		 * If the display surface is the same as the content surface
-		 * then remove the reference
-		 */
-		if (stdu->content_fb_type == SAME_AS_DISPLAY) {
-			if (stdu->defined) {
-				/* Unbind the current surface */
-				ret = vmw_stdu_bind_st(dev_priv, stdu, NULL);
-				if (unlikely(ret != 0))
-					goto err_out;
-			}
-			vmw_stdu_unpin_display(stdu);
-			stdu->display_srf = NULL;
-		}
-	}
-
-
-	if (!new_fb) {
-		/* Blanks the display */
-		(void) vmw_stdu_update_st(dev_priv, stdu);
-
-		return 0;
-	}
 
+	if (!stdu->defined || !vmw_kms_crtc_flippable(dev_priv, crtc))
+		return -EINVAL;
 
-	if (stdu->content_fb_type == SAME_AS_DISPLAY) {
-		stdu->display_srf = vmw_framebuffer_to_vfbs(new_fb)->surface;
-		ret = vmw_stdu_pin_display(stdu);
-		if (ret) {
-			stdu->display_srf = NULL;
-			goto err_out;
-		}
+	ret = vmw_stdu_bind_fb(dev_priv, crtc, &crtc->mode, new_fb);
+	if (ret)
+		return ret;
 
-		/* Bind display surface */
-		ret = vmw_stdu_bind_st(dev_priv, stdu, &stdu->display_srf->res);
-		if (unlikely(ret != 0))
-			goto err_unpin_display_and_content;
-	}
+	if (stdu->base.is_implicit)
+		vmw_kms_update_implicit_fb(dev_priv, crtc);
 
-	/* Update display surface: after this point everything is bound */
-	ret = vmw_stdu_update_st(dev_priv, stdu);
-	if (unlikely(ret != 0))
+	vclips.x = crtc->x;
+	vclips.y = crtc->y;
+	vclips.w = crtc->mode.hdisplay;
+	vclips.h = crtc->mode.vdisplay;
+	if (vfb->dmabuf)
+		ret = vmw_kms_stdu_dma(dev_priv, NULL, vfb, NULL, NULL, &vclips,
+				       1, 1, true, false);
+	else
+		ret = vmw_kms_stdu_surface_dirty(dev_priv, vfb, NULL, &vclips,
+						 NULL, 0, 0, 1, 1, NULL);
+	if (ret)
 		return ret;
 
 	if (event) {
@@ -721,14 +699,7 @@ static int vmw_stdu_crtc_page_flip(struct drm_crtc *crtc,
 		vmw_fifo_flush(dev_priv, false);
 	}
 
-	return ret;
-
-err_unpin_display_and_content:
-	vmw_stdu_unpin_display(stdu);
-err_out:
-	crtc->primary->fb = NULL;
-	stdu->content_fb = NULL;
-	return ret;
+	return 0;
 }
 
 
@@ -1138,7 +1109,7 @@ static int vmw_stdu_init(struct vmw_private *dev_priv, unsigned unit)
 	stdu->base.pref_active = (unit == 0);
 	stdu->base.pref_width  = dev_priv->initial_width;
 	stdu->base.pref_height = dev_priv->initial_height;
-	stdu->base.is_implicit = true;
+	stdu->base.is_implicit = false;
 
 	drm_connector_init(dev, connector, &vmw_stdu_connector_funcs,
 			   DRM_MODE_CONNECTOR_VIRTUAL);
@@ -1159,7 +1130,17 @@ static int vmw_stdu_init(struct vmw_private *dev_priv, unsigned unit)
 	drm_object_attach_property(&connector->base,
 				   dev->mode_config.dirty_info_property,
 				   1);
-
+	drm_object_attach_property(&connector->base,
+				   dev_priv->hotplug_mode_update_property, 1);
+	drm_object_attach_property(&connector->base,
+				   dev->mode_config.suggested_x_property, 0);
+	drm_object_attach_property(&connector->base,
+				   dev->mode_config.suggested_y_property, 0);
+	if (dev_priv->implicit_placement_property)
+		drm_object_attach_property
+			(&connector->base,
+			 dev_priv->implicit_placement_property,
+			 stdu->base.is_implicit);
 	return 0;
 }
 
@@ -1224,6 +1205,8 @@ int vmw_kms_stdu_init_display(struct vmw_private *dev_priv)
 
 	dev_priv->active_display_unit = vmw_du_screen_target;
 
+	vmw_kms_create_implicit_placement_property(dev_priv, false);
+
 	for (i = 0; i < VMWGFX_NUM_DISPLAY_UNITS; ++i) {
 		ret = vmw_stdu_init(dev_priv, i);
 
diff --git a/drivers/gpu/host1x/bus.c b/drivers/gpu/host1x/bus.c
index da462afcb225..c27858ae0552 100644
--- a/drivers/gpu/host1x/bus.c
+++ b/drivers/gpu/host1x/bus.c
@@ -18,6 +18,7 @@
 #include <linux/host1x.h>
 #include <linux/of.h>
 #include <linux/slab.h>
+#include <linux/of_device.h>
 
 #include "bus.h"
 #include "dev.h"
@@ -82,8 +83,10 @@ static int host1x_device_parse_dt(struct host1x_device *device,
 		if (of_match_node(driver->subdevs, np) &&
 		    of_device_is_available(np)) {
 			err = host1x_subdev_add(device, np);
-			if (err < 0)
+			if (err < 0) {
+				of_node_put(np);
 				return err;
+			}
 		}
 	}
 
@@ -394,6 +397,7 @@ static int host1x_device_add(struct host1x *host1x,
 	device->dev.coherent_dma_mask = host1x->dev->coherent_dma_mask;
 	device->dev.dma_mask = &device->dev.coherent_dma_mask;
 	dev_set_name(&device->dev, "%s", driver->driver.name);
+	of_dma_configure(&device->dev, host1x->dev->of_node);
 	device->dev.release = host1x_device_release;
 	device->dev.bus = &host1x_bus_type;
 	device->dev.parent = host1x->dev;
diff --git a/drivers/gpu/host1x/dev.c b/drivers/gpu/host1x/dev.c
index 314bf3718cc7..ff348690df94 100644
--- a/drivers/gpu/host1x/dev.c
+++ b/drivers/gpu/host1x/dev.c
@@ -23,6 +23,7 @@
 #include <linux/of_device.h>
 #include <linux/clk.h>
 #include <linux/io.h>
+#include <linux/dma-mapping.h>
 
 #define CREATE_TRACE_POINTS
 #include <trace/events/host1x.h>
@@ -68,6 +69,7 @@ static const struct host1x_info host1x01_info = {
 	.nb_bases	= 8,
 	.init		= host1x01_init,
 	.sync_offset	= 0x3000,
+	.dma_mask	= DMA_BIT_MASK(32),
 };
 
 static const struct host1x_info host1x02_info = {
@@ -77,6 +79,7 @@ static const struct host1x_info host1x02_info = {
 	.nb_bases = 12,
 	.init = host1x02_init,
 	.sync_offset = 0x3000,
+	.dma_mask = DMA_BIT_MASK(32),
 };
 
 static const struct host1x_info host1x04_info = {
@@ -86,6 +89,7 @@ static const struct host1x_info host1x04_info = {
 	.nb_bases = 64,
 	.init = host1x04_init,
 	.sync_offset = 0x2100,
+	.dma_mask = DMA_BIT_MASK(34),
 };
 
 static const struct host1x_info host1x05_info = {
@@ -95,6 +99,7 @@ static const struct host1x_info host1x05_info = {
 	.nb_bases = 64,
 	.init = host1x05_init,
 	.sync_offset = 0x2100,
+	.dma_mask = DMA_BIT_MASK(34),
 };
 
 static struct of_device_id host1x_of_match[] = {
@@ -148,6 +153,8 @@ static int host1x_probe(struct platform_device *pdev)
 	if (IS_ERR(host->regs))
 		return PTR_ERR(host->regs);
 
+	dma_set_mask_and_coherent(host->dev, host->info->dma_mask);
+
 	if (host->info->init) {
 		err = host->info->init(host);
 		if (err)
diff --git a/drivers/gpu/host1x/dev.h b/drivers/gpu/host1x/dev.h
index 0b6e8e9629c5..dace124994bb 100644
--- a/drivers/gpu/host1x/dev.h
+++ b/drivers/gpu/host1x/dev.h
@@ -96,6 +96,7 @@ struct host1x_info {
 	int	nb_mlocks;		/* host1x: number of mlocks */
 	int	(*init)(struct host1x *); /* initialize per SoC ops */
 	int	sync_offset;
+	u64	dma_mask;		/* mask of addressable memory */
 };
 
 struct host1x {
diff --git a/drivers/gpu/host1x/job.c b/drivers/gpu/host1x/job.c
index 63bd63f3c7df..1919aab88c3f 100644
--- a/drivers/gpu/host1x/job.c
+++ b/drivers/gpu/host1x/job.c
@@ -225,7 +225,7 @@ unpin:
 	return 0;
 }
 
-static unsigned int do_relocs(struct host1x_job *job, struct host1x_bo *cmdbuf)
+static int do_relocs(struct host1x_job *job, struct host1x_bo *cmdbuf)
 {
 	int i = 0;
 	u32 last_page = ~0;
diff --git a/drivers/gpu/ipu-v3/ipu-common.c b/drivers/gpu/ipu-v3/ipu-common.c
index f2e13eb8339f..e00db3f510dd 100644
--- a/drivers/gpu/ipu-v3/ipu-common.c
+++ b/drivers/gpu/ipu-v3/ipu-common.c
@@ -1050,6 +1050,17 @@ static int ipu_add_client_devices(struct ipu_soc *ipu, unsigned long ipu_base)
 	for (i = 0; i < ARRAY_SIZE(client_reg); i++) {
 		const struct ipu_platform_reg *reg = &client_reg[i];
 		struct platform_device *pdev;
+		struct device_node *of_node;
+
+		/* Associate subdevice with the corresponding port node */
+		of_node = of_graph_get_port_by_id(dev->of_node, i);
+		if (!of_node) {
+			dev_info(dev,
+				 "no port@%d node in %s, not using %s%d\n",
+				 i, dev->of_node->full_name,
+				 (i / 2) ? "DI" : "CSI", i % 2);
+			continue;
+		}
 
 		pdev = platform_device_alloc(reg->name, id++);
 		if (!pdev) {
@@ -1057,17 +1068,9 @@ static int ipu_add_client_devices(struct ipu_soc *ipu, unsigned long ipu_base)
 			goto err_register;
 		}
 
+		pdev->dev.of_node = of_node;
 		pdev->dev.parent = dev;
 
-		/* Associate subdevice with the corresponding port node */
-		pdev->dev.of_node = of_graph_get_port_by_id(dev->of_node, i);
-		if (!pdev->dev.of_node) {
-			dev_err(dev, "missing port@%d node in %s\n", i,
-				dev->of_node->full_name);
-			ret = -ENODEV;
-			goto err_register;
-		}
-
 		ret = platform_device_add_data(pdev, &reg->pdata,
 					       sizeof(reg->pdata));
 		if (!ret)
@@ -1289,10 +1292,6 @@ static int ipu_probe(struct platform_device *pdev)
 	ipu->irq_sync = irq_sync;
 	ipu->irq_err = irq_err;
 
-	ret = ipu_irq_init(ipu);
-	if (ret)
-		goto out_failed_irq;
-
 	ret = device_reset(&pdev->dev);
 	if (ret) {
 		dev_err(&pdev->dev, "failed to reset: %d\n", ret);
@@ -1302,6 +1301,10 @@ static int ipu_probe(struct platform_device *pdev)
 	if (ret)
 		goto out_failed_reset;
 
+	ret = ipu_irq_init(ipu);
+	if (ret)
+		goto out_failed_irq;
+
 	/* Set MCU_T to divide MCU access window into 2 */
 	ipu_cm_write(ipu, 0x00400000L | (IPU_MCU_T_DEFAULT << 18),
 			IPU_DISP_GEN);
@@ -1324,9 +1327,9 @@ static int ipu_probe(struct platform_device *pdev)
 failed_add_clients:
 	ipu_submodules_exit(ipu);
 failed_submodules_init:
-out_failed_reset:
 	ipu_irq_exit(ipu);
 out_failed_irq:
+out_failed_reset:
 	clk_disable_unprepare(ipu->clk);
 	return ret;
 }
diff --git a/drivers/gpu/ipu-v3/ipu-dc.c b/drivers/gpu/ipu-v3/ipu-dc.c
index d3ad5347342c..2f29780e7c68 100644
--- a/drivers/gpu/ipu-v3/ipu-dc.c
+++ b/drivers/gpu/ipu-v3/ipu-dc.c
@@ -171,6 +171,7 @@ int ipu_dc_init_sync(struct ipu_dc *dc, struct ipu_di *di, bool interlaced,
 		u32 bus_format, u32 width)
 {
 	struct ipu_dc_priv *priv = dc->priv;
+	int addr, sync;
 	u32 reg = 0;
 	int map;
 
@@ -182,41 +183,39 @@ int ipu_dc_init_sync(struct ipu_dc *dc, struct ipu_di *di, bool interlaced,
 		return map;
 	}
 
-	if (interlaced) {
-		int addr;
-
-		if (dc->di)
-			addr = 1;
-		else
-			addr = 0;
+	/*
+	 * In interlaced mode we need more counters to create the asymmetric
+	 * per-field VSYNC signals. The pixel active signal synchronising DC
+	 * to DI moves to signal generator #6 (see ipu-di.c). In progressive
+	 * mode counter #5 is used.
+	 */
+	sync = interlaced ? 6 : 5;
+
+	/* Reserve 5 microcode template words for each DI */
+	if (dc->di)
+		addr = 5;
+	else
+		addr = 0;
 
+	if (interlaced) {
 		dc_link_event(dc, DC_EVT_NL, addr, 3);
 		dc_link_event(dc, DC_EVT_EOL, addr, 2);
 		dc_link_event(dc, DC_EVT_NEW_DATA, addr, 1);
 
 		/* Init template microcode */
-		dc_write_tmpl(dc, addr, WROD(0), 0, map, SYNC_WAVE, 0, 6, 1);
+		dc_write_tmpl(dc, addr, WROD(0), 0, map, SYNC_WAVE, 0, sync, 1);
 	} else {
-		if (dc->di) {
-			dc_link_event(dc, DC_EVT_NL, 2, 3);
-			dc_link_event(dc, DC_EVT_EOL, 3, 2);
-			dc_link_event(dc, DC_EVT_NEW_DATA, 1, 1);
-			/* Init template microcode */
-			dc_write_tmpl(dc, 2, WROD(0), 0, map, SYNC_WAVE, 8, 5, 1);
-			dc_write_tmpl(dc, 3, WROD(0), 0, map, SYNC_WAVE, 4, 5, 0);
-			dc_write_tmpl(dc, 4, WRG, 0, map, NULL_WAVE, 0, 0, 1);
-			dc_write_tmpl(dc, 1, WROD(0), 0, map, SYNC_WAVE, 0, 5, 1);
-		} else {
-			dc_link_event(dc, DC_EVT_NL, 5, 3);
-			dc_link_event(dc, DC_EVT_EOL, 6, 2);
-			dc_link_event(dc, DC_EVT_NEW_DATA, 8, 1);
-			/* Init template microcode */
-			dc_write_tmpl(dc, 5, WROD(0), 0, map, SYNC_WAVE, 8, 5, 1);
-			dc_write_tmpl(dc, 6, WROD(0), 0, map, SYNC_WAVE, 4, 5, 0);
-			dc_write_tmpl(dc, 7, WRG, 0, map, NULL_WAVE, 0, 0, 1);
-			dc_write_tmpl(dc, 8, WROD(0), 0, map, SYNC_WAVE, 0, 5, 1);
-		}
+		dc_link_event(dc, DC_EVT_NL, addr + 2, 3);
+		dc_link_event(dc, DC_EVT_EOL, addr + 3, 2);
+		dc_link_event(dc, DC_EVT_NEW_DATA, addr + 1, 1);
+
+		/* Init template microcode */
+		dc_write_tmpl(dc, addr + 2, WROD(0), 0, map, SYNC_WAVE, 8, sync, 1);
+		dc_write_tmpl(dc, addr + 3, WROD(0), 0, map, SYNC_WAVE, 4, sync, 0);
+		dc_write_tmpl(dc, addr + 4, WRG, 0, map, NULL_WAVE, 0, 0, 1);
+		dc_write_tmpl(dc, addr + 1, WROD(0), 0, map, SYNC_WAVE, 0, sync, 1);
 	}
+
 	dc_link_event(dc, DC_EVT_NF, 0, 0);
 	dc_link_event(dc, DC_EVT_NFIELD, 0, 0);
 	dc_link_event(dc, DC_EVT_EOF, 0, 0);
diff --git a/drivers/hwmon/ads1015.c b/drivers/hwmon/ads1015.c
index f155b8380481..2b3105c8aed3 100644
--- a/drivers/hwmon/ads1015.c
+++ b/drivers/hwmon/ads1015.c
@@ -126,7 +126,7 @@ static int ads1015_reg_to_mv(struct i2c_client *client, unsigned int channel,
 	struct ads1015_data *data = i2c_get_clientdata(client);
 	unsigned int pga = data->channel_data[channel].pga;
 	int fullscale = fullscale_table[pga];
-	const unsigned mask = data->id == ads1115 ? 0x7fff : 0x7ff0;
+	const int mask = data->id == ads1115 ? 0x7fff : 0x7ff0;
 
 	return DIV_ROUND_CLOSEST(reg * fullscale, mask);
 }
diff --git a/drivers/hwmon/gpio-fan.c b/drivers/hwmon/gpio-fan.c
index 82de3deeb18a..685568b1236d 100644
--- a/drivers/hwmon/gpio-fan.c
+++ b/drivers/hwmon/gpio-fan.c
@@ -406,16 +406,11 @@ static int gpio_fan_get_cur_state(struct thermal_cooling_device *cdev,
 				  unsigned long *state)
 {
 	struct gpio_fan_data *fan_data = cdev->devdata;
-	int r;
 
 	if (!fan_data)
 		return -EINVAL;
 
-	r = get_fan_speed_index(fan_data);
-	if (r < 0)
-		return r;
-
-	*state = r;
+	*state = fan_data->speed_index;
 	return 0;
 }
 
diff --git a/drivers/i2c/busses/i2c-brcmstb.c b/drivers/i2c/busses/i2c-brcmstb.c
index 3711df1d4526..4a45408dd820 100644
--- a/drivers/i2c/busses/i2c-brcmstb.c
+++ b/drivers/i2c/busses/i2c-brcmstb.c
@@ -586,8 +586,7 @@ static int brcmstb_i2c_probe(struct platform_device *pdev)
 	if (!dev)
 		return -ENOMEM;
 
-	dev->bsc_regmap = devm_kzalloc(&pdev->dev, sizeof(struct bsc_regs *),
-				       GFP_KERNEL);
+	dev->bsc_regmap = devm_kzalloc(&pdev->dev, sizeof(*dev->bsc_regmap), GFP_KERNEL);
 	if (!dev->bsc_regmap)
 		return -ENOMEM;
 
diff --git a/drivers/i2c/busses/i2c-i801.c b/drivers/i2c/busses/i2c-i801.c
index f62d69799a9c..27fa0cb09538 100644
--- a/drivers/i2c/busses/i2c-i801.c
+++ b/drivers/i2c/busses/i2c-i801.c
@@ -1271,6 +1271,8 @@ static int i801_probe(struct pci_dev *dev, const struct pci_device_id *id)
 	switch (dev->device) {
 	case PCI_DEVICE_ID_INTEL_SUNRISEPOINT_H_SMBUS:
 	case PCI_DEVICE_ID_INTEL_SUNRISEPOINT_LP_SMBUS:
+	case PCI_DEVICE_ID_INTEL_LEWISBURG_SMBUS:
+	case PCI_DEVICE_ID_INTEL_LEWISBURG_SSKU_SMBUS:
 	case PCI_DEVICE_ID_INTEL_DNV_SMBUS:
 		priv->features |= FEATURE_I2C_BLOCK_READ;
 		priv->features |= FEATURE_IRQ;
diff --git a/drivers/i2c/busses/i2c-omap.c b/drivers/i2c/busses/i2c-omap.c
index 08d26ba61ed3..13c45296ce5b 100644
--- a/drivers/i2c/busses/i2c-omap.c
+++ b/drivers/i2c/busses/i2c-omap.c
@@ -1450,7 +1450,8 @@ omap_i2c_probe(struct platform_device *pdev)
 
 err_unuse_clocks:
 	omap_i2c_write_reg(omap, OMAP_I2C_CON_REG, 0);
-	pm_runtime_put(omap->dev);
+	pm_runtime_dont_use_autosuspend(omap->dev);
+	pm_runtime_put_sync(omap->dev);
 	pm_runtime_disable(&pdev->dev);
 err_free_mem:
 
@@ -1468,6 +1469,7 @@ static int omap_i2c_remove(struct platform_device *pdev)
 		return ret;
 
 	omap_i2c_write_reg(omap, OMAP_I2C_CON_REG, 0);
+	pm_runtime_dont_use_autosuspend(&pdev->dev);
 	pm_runtime_put_sync(&pdev->dev);
 	pm_runtime_disable(&pdev->dev);
 	return 0;
diff --git a/drivers/i2c/busses/i2c-uniphier-f.c b/drivers/i2c/busses/i2c-uniphier-f.c
index f3e5ff8522f0..213ba55e17c3 100644
--- a/drivers/i2c/busses/i2c-uniphier-f.c
+++ b/drivers/i2c/busses/i2c-uniphier-f.c
@@ -467,7 +467,7 @@ static int uniphier_fi2c_clk_init(struct device *dev,
 		bus_speed = UNIPHIER_FI2C_DEFAULT_SPEED;
 
 	if (!bus_speed) {
-		dev_err(dev, "clock-freqyency should not be zero\n");
+		dev_err(dev, "clock-frequency should not be zero\n");
 		return -EINVAL;
 	}
 
diff --git a/drivers/i2c/busses/i2c-uniphier.c b/drivers/i2c/busses/i2c-uniphier.c
index 1f4f3f53819c..89eaa8a7e1e0 100644
--- a/drivers/i2c/busses/i2c-uniphier.c
+++ b/drivers/i2c/busses/i2c-uniphier.c
@@ -328,7 +328,7 @@ static int uniphier_i2c_clk_init(struct device *dev,
 		bus_speed = UNIPHIER_I2C_DEFAULT_SPEED;
 
 	if (!bus_speed) {
-		dev_err(dev, "clock-freqyency should not be zero\n");
+		dev_err(dev, "clock-frequency should not be zero\n");
 		return -EINVAL;
 	}
 
diff --git a/drivers/infiniband/core/device.c b/drivers/infiniband/core/device.c
index 00da80e02154..94b80a51ab68 100644
--- a/drivers/infiniband/core/device.c
+++ b/drivers/infiniband/core/device.c
@@ -358,6 +358,7 @@ int ib_register_device(struct ib_device *device,
 	ret = device->query_device(device, &device->attrs, &uhw);
 	if (ret) {
 		printk(KERN_WARNING "Couldn't query the device attributes\n");
+		ib_cache_cleanup_one(device);
 		goto out;
 	}
 
diff --git a/drivers/infiniband/core/sa_query.c b/drivers/infiniband/core/sa_query.c
index f334090bb612..1e37f3515d98 100644
--- a/drivers/infiniband/core/sa_query.c
+++ b/drivers/infiniband/core/sa_query.c
@@ -1071,7 +1071,7 @@ int ib_init_ah_from_path(struct ib_device *device, u8 port_num,
 		}
 	}
 
-	if (rec->hop_limit > 1 || use_roce) {
+	if (rec->hop_limit > 0 || use_roce) {
 		ah_attr->ah_flags = IB_AH_GRH;
 		ah_attr->grh.dgid = rec->dgid;
 
diff --git a/drivers/infiniband/core/sysfs.c b/drivers/infiniband/core/sysfs.c
index 3de93517efe4..14606afbfaa8 100644
--- a/drivers/infiniband/core/sysfs.c
+++ b/drivers/infiniband/core/sysfs.c
@@ -336,7 +336,6 @@ static ssize_t _show_port_gid_attr(struct ib_port *p,
 	union ib_gid gid;
 	struct ib_gid_attr gid_attr = {};
 	ssize_t ret;
-	va_list args;
 
 	ret = ib_query_gid(p->ibdev, p->port_num, tab_attr->index, &gid,
 			   &gid_attr);
@@ -348,7 +347,6 @@ static ssize_t _show_port_gid_attr(struct ib_port *p,
 err:
 	if (gid_attr.ndev)
 		dev_put(gid_attr.ndev);
-	va_end(args);
 	return ret;
 }
 
@@ -722,12 +720,11 @@ static struct attribute_group *get_counter_table(struct ib_device *dev,
 
 	if (get_perf_mad(dev, port_num, IB_PMA_CLASS_PORT_INFO,
 				&cpi, 40, sizeof(cpi)) >= 0) {
-
-		if (cpi.capability_mask && IB_PMA_CLASS_CAP_EXT_WIDTH)
+		if (cpi.capability_mask & IB_PMA_CLASS_CAP_EXT_WIDTH)
 			/* We have extended counters */
 			return &pma_group_ext;
 
-		if (cpi.capability_mask && IB_PMA_CLASS_CAP_EXT_WIDTH_NOIETF)
+		if (cpi.capability_mask & IB_PMA_CLASS_CAP_EXT_WIDTH_NOIETF)
 			/* But not the IETF ones */
 			return &pma_group_noietf;
 	}
diff --git a/drivers/infiniband/core/uverbs_cmd.c b/drivers/infiniband/core/uverbs_cmd.c
index 6ffc9c4e93af..6c6fbff19752 100644
--- a/drivers/infiniband/core/uverbs_cmd.c
+++ b/drivers/infiniband/core/uverbs_cmd.c
@@ -1970,7 +1970,8 @@ ssize_t ib_uverbs_create_qp(struct ib_uverbs_file *file,
 		   resp_size);
 	INIT_UDATA(&uhw, buf + sizeof(cmd),
 		   (unsigned long)cmd.response + resp_size,
-		   in_len - sizeof(cmd), out_len - resp_size);
+		   in_len - sizeof(cmd) - sizeof(struct ib_uverbs_cmd_hdr),
+		   out_len - resp_size);
 
 	memset(&cmd_ex, 0, sizeof(cmd_ex));
 	cmd_ex.user_handle = cmd.user_handle;
@@ -3413,7 +3414,8 @@ ssize_t ib_uverbs_create_srq(struct ib_uverbs_file *file,
 
 	INIT_UDATA(&udata, buf + sizeof cmd,
 		   (unsigned long) cmd.response + sizeof resp,
-		   in_len - sizeof cmd, out_len - sizeof resp);
+		   in_len - sizeof cmd - sizeof(struct ib_uverbs_cmd_hdr),
+		   out_len - sizeof resp);
 
 	ret = __uverbs_create_xsrq(file, ib_dev, &xcmd, &udata);
 	if (ret)
@@ -3439,7 +3441,8 @@ ssize_t ib_uverbs_create_xsrq(struct ib_uverbs_file *file,
 
 	INIT_UDATA(&udata, buf + sizeof cmd,
 		   (unsigned long) cmd.response + sizeof resp,
-		   in_len - sizeof cmd, out_len - sizeof resp);
+		   in_len - sizeof cmd - sizeof(struct ib_uverbs_cmd_hdr),
+		   out_len - sizeof resp);
 
 	ret = __uverbs_create_xsrq(file, ib_dev, &cmd, &udata);
 	if (ret)
diff --git a/drivers/infiniband/hw/mlx4/mad.c b/drivers/infiniband/hw/mlx4/mad.c
index 26833bfa639b..d68f506c1922 100644
--- a/drivers/infiniband/hw/mlx4/mad.c
+++ b/drivers/infiniband/hw/mlx4/mad.c
@@ -817,17 +817,48 @@ static int ib_process_mad(struct ib_device *ibdev, int mad_flags, u8 port_num,
 	return IB_MAD_RESULT_SUCCESS | IB_MAD_RESULT_REPLY;
 }
 
-static void edit_counter(struct mlx4_counter *cnt,
-					struct ib_pma_portcounters *pma_cnt)
+static void edit_counter(struct mlx4_counter *cnt, void *counters,
+			 __be16 attr_id)
 {
-	ASSIGN_32BIT_COUNTER(pma_cnt->port_xmit_data,
-			     (be64_to_cpu(cnt->tx_bytes) >> 2));
-	ASSIGN_32BIT_COUNTER(pma_cnt->port_rcv_data,
-			     (be64_to_cpu(cnt->rx_bytes) >> 2));
-	ASSIGN_32BIT_COUNTER(pma_cnt->port_xmit_packets,
-			     be64_to_cpu(cnt->tx_frames));
-	ASSIGN_32BIT_COUNTER(pma_cnt->port_rcv_packets,
-			     be64_to_cpu(cnt->rx_frames));
+	switch (attr_id) {
+	case IB_PMA_PORT_COUNTERS:
+	{
+		struct ib_pma_portcounters *pma_cnt =
+			(struct ib_pma_portcounters *)counters;
+
+		ASSIGN_32BIT_COUNTER(pma_cnt->port_xmit_data,
+				     (be64_to_cpu(cnt->tx_bytes) >> 2));
+		ASSIGN_32BIT_COUNTER(pma_cnt->port_rcv_data,
+				     (be64_to_cpu(cnt->rx_bytes) >> 2));
+		ASSIGN_32BIT_COUNTER(pma_cnt->port_xmit_packets,
+				     be64_to_cpu(cnt->tx_frames));
+		ASSIGN_32BIT_COUNTER(pma_cnt->port_rcv_packets,
+				     be64_to_cpu(cnt->rx_frames));
+		break;
+	}
+	case IB_PMA_PORT_COUNTERS_EXT:
+	{
+		struct ib_pma_portcounters_ext *pma_cnt_ext =
+			(struct ib_pma_portcounters_ext *)counters;
+
+		pma_cnt_ext->port_xmit_data =
+			cpu_to_be64(be64_to_cpu(cnt->tx_bytes) >> 2);
+		pma_cnt_ext->port_rcv_data =
+			cpu_to_be64(be64_to_cpu(cnt->rx_bytes) >> 2);
+		pma_cnt_ext->port_xmit_packets = cnt->tx_frames;
+		pma_cnt_ext->port_rcv_packets = cnt->rx_frames;
+		break;
+	}
+	}
+}
+
+static int iboe_process_mad_port_info(void *out_mad)
+{
+	struct ib_class_port_info cpi = {};
+
+	cpi.capability_mask = IB_PMA_CLASS_CAP_EXT_WIDTH;
+	memcpy(out_mad, &cpi, sizeof(cpi));
+	return IB_MAD_RESULT_SUCCESS | IB_MAD_RESULT_REPLY;
 }
 
 static int iboe_process_mad(struct ib_device *ibdev, int mad_flags, u8 port_num,
@@ -842,6 +873,9 @@ static int iboe_process_mad(struct ib_device *ibdev, int mad_flags, u8 port_num,
 	if (in_mad->mad_hdr.mgmt_class != IB_MGMT_CLASS_PERF_MGMT)
 		return -EINVAL;
 
+	if (in_mad->mad_hdr.attr_id == IB_PMA_CLASS_PORT_INFO)
+		return iboe_process_mad_port_info((void *)(out_mad->data + 40));
+
 	memset(&counter_stats, 0, sizeof(counter_stats));
 	mutex_lock(&dev->counters_table[port_num - 1].mutex);
 	list_for_each_entry(tmp_counter,
@@ -863,7 +897,8 @@ static int iboe_process_mad(struct ib_device *ibdev, int mad_flags, u8 port_num,
 		switch (counter_stats.counter_mode & 0xf) {
 		case 0:
 			edit_counter(&counter_stats,
-				     (void *)(out_mad->data + 40));
+				     (void *)(out_mad->data + 40),
+				     in_mad->mad_hdr.attr_id);
 			err = IB_MAD_RESULT_SUCCESS | IB_MAD_RESULT_REPLY;
 			break;
 		default:
@@ -894,8 +929,10 @@ int mlx4_ib_process_mad(struct ib_device *ibdev, int mad_flags, u8 port_num,
 	 */
 	if (link == IB_LINK_LAYER_INFINIBAND) {
 		if (mlx4_is_slave(dev->dev) &&
-		    in_mad->mad_hdr.mgmt_class == IB_MGMT_CLASS_PERF_MGMT &&
-		    in_mad->mad_hdr.attr_id == IB_PMA_PORT_COUNTERS)
+		    (in_mad->mad_hdr.mgmt_class == IB_MGMT_CLASS_PERF_MGMT &&
+		     (in_mad->mad_hdr.attr_id == IB_PMA_PORT_COUNTERS ||
+		      in_mad->mad_hdr.attr_id == IB_PMA_PORT_COUNTERS_EXT ||
+		      in_mad->mad_hdr.attr_id == IB_PMA_CLASS_PORT_INFO)))
 			return iboe_process_mad(ibdev, mad_flags, port_num, in_wc,
 						in_grh, in_mad, out_mad);
 
diff --git a/drivers/infiniband/hw/mlx4/qp.c b/drivers/infiniband/hw/mlx4/qp.c
index bc5536f00b6c..fd97534762b8 100644
--- a/drivers/infiniband/hw/mlx4/qp.c
+++ b/drivers/infiniband/hw/mlx4/qp.c
@@ -1681,9 +1681,12 @@ static int __mlx4_ib_modify_qp(struct ib_qp *ibqp,
 	}
 
 	if (qp->ibqp.uobject)
-		context->usr_page = cpu_to_be32(to_mucontext(ibqp->uobject->context)->uar.index);
+		context->usr_page = cpu_to_be32(
+			mlx4_to_hw_uar_index(dev->dev,
+					     to_mucontext(ibqp->uobject->context)->uar.index));
 	else
-		context->usr_page = cpu_to_be32(dev->priv_uar.index);
+		context->usr_page = cpu_to_be32(
+			mlx4_to_hw_uar_index(dev->dev, dev->priv_uar.index));
 
 	if (attr_mask & IB_QP_DEST_QPN)
 		context->remote_qpn = cpu_to_be32(attr->dest_qp_num);
diff --git a/drivers/infiniband/hw/mlx5/qp.c b/drivers/infiniband/hw/mlx5/qp.c
index 9116bc3988a6..34cb8e87c7b8 100644
--- a/drivers/infiniband/hw/mlx5/qp.c
+++ b/drivers/infiniband/hw/mlx5/qp.c
@@ -270,8 +270,10 @@ static int sq_overhead(enum ib_qp_type qp_type)
 		/* fall through */
 	case IB_QPT_RC:
 		size += sizeof(struct mlx5_wqe_ctrl_seg) +
-			sizeof(struct mlx5_wqe_atomic_seg) +
-			sizeof(struct mlx5_wqe_raddr_seg);
+			max(sizeof(struct mlx5_wqe_atomic_seg) +
+			    sizeof(struct mlx5_wqe_raddr_seg),
+			    sizeof(struct mlx5_wqe_umr_ctrl_seg) +
+			    sizeof(struct mlx5_mkey_seg));
 		break;
 
 	case IB_QPT_XRC_TGT:
@@ -279,9 +281,9 @@ static int sq_overhead(enum ib_qp_type qp_type)
 
 	case IB_QPT_UC:
 		size += sizeof(struct mlx5_wqe_ctrl_seg) +
-			sizeof(struct mlx5_wqe_raddr_seg) +
-			sizeof(struct mlx5_wqe_umr_ctrl_seg) +
-			sizeof(struct mlx5_mkey_seg);
+			max(sizeof(struct mlx5_wqe_raddr_seg),
+			    sizeof(struct mlx5_wqe_umr_ctrl_seg) +
+			    sizeof(struct mlx5_mkey_seg));
 		break;
 
 	case IB_QPT_UD:
diff --git a/drivers/infiniband/hw/mlx5/srq.c b/drivers/infiniband/hw/mlx5/srq.c
index 4659256cd95e..3b2ddd64a371 100644
--- a/drivers/infiniband/hw/mlx5/srq.c
+++ b/drivers/infiniband/hw/mlx5/srq.c
@@ -75,7 +75,8 @@ static void mlx5_ib_srq_event(struct mlx5_core_srq *srq, enum mlx5_event type)
 
 static int create_srq_user(struct ib_pd *pd, struct mlx5_ib_srq *srq,
 			   struct mlx5_create_srq_mbox_in **in,
-			   struct ib_udata *udata, int buf_size, int *inlen)
+			   struct ib_udata *udata, int buf_size, int *inlen,
+			   int is_xrc)
 {
 	struct mlx5_ib_dev *dev = to_mdev(pd->device);
 	struct mlx5_ib_create_srq ucmd = {};
@@ -87,13 +88,8 @@ static int create_srq_user(struct ib_pd *pd, struct mlx5_ib_srq *srq,
 	int ncont;
 	u32 offset;
 	u32 uidx = MLX5_IB_DEFAULT_UIDX;
-	int drv_data = udata->inlen - sizeof(struct ib_uverbs_cmd_hdr);
 
-	if (drv_data < 0)
-		return -EINVAL;
-
-	ucmdlen = (drv_data < sizeof(ucmd)) ?
-		  drv_data : sizeof(ucmd);
+	ucmdlen = min(udata->inlen, sizeof(ucmd));
 
 	if (ib_copy_from_udata(&ucmd, udata, ucmdlen)) {
 		mlx5_ib_dbg(dev, "failed copy udata\n");
@@ -103,15 +99,17 @@ static int create_srq_user(struct ib_pd *pd, struct mlx5_ib_srq *srq,
 	if (ucmd.reserved0 || ucmd.reserved1)
 		return -EINVAL;
 
-	if (drv_data > sizeof(ucmd) &&
+	if (udata->inlen > sizeof(ucmd) &&
 	    !ib_is_udata_cleared(udata, sizeof(ucmd),
-				 drv_data - sizeof(ucmd)))
+				 udata->inlen - sizeof(ucmd)))
 		return -EINVAL;
 
-	err = get_srq_user_index(to_mucontext(pd->uobject->context),
-				 &ucmd, udata->inlen, &uidx);
-	if (err)
-		return err;
+	if (is_xrc) {
+		err = get_srq_user_index(to_mucontext(pd->uobject->context),
+					 &ucmd, udata->inlen, &uidx);
+		if (err)
+			return err;
+	}
 
 	srq->wq_sig = !!(ucmd.flags & MLX5_SRQ_FLAG_SIGNATURE);
 
@@ -151,7 +149,8 @@ static int create_srq_user(struct ib_pd *pd, struct mlx5_ib_srq *srq,
 	(*in)->ctx.log_pg_sz = page_shift - MLX5_ADAPTER_PAGE_SHIFT;
 	(*in)->ctx.pgoff_cqn = cpu_to_be32(offset << 26);
 
-	if (MLX5_CAP_GEN(dev->mdev, cqe_version) == MLX5_CQE_VERSION_V1) {
+	if ((MLX5_CAP_GEN(dev->mdev, cqe_version) == MLX5_CQE_VERSION_V1) &&
+	     is_xrc){
 		xsrqc = MLX5_ADDR_OF(create_xrc_srq_in, *in,
 				     xrc_srq_context_entry);
 		MLX5_SET(xrc_srqc, xsrqc, user_index, uidx);
@@ -170,7 +169,7 @@ err_umem:
 
 static int create_srq_kernel(struct mlx5_ib_dev *dev, struct mlx5_ib_srq *srq,
 			     struct mlx5_create_srq_mbox_in **in, int buf_size,
-			     int *inlen)
+			     int *inlen, int is_xrc)
 {
 	int err;
 	int i;
@@ -224,7 +223,8 @@ static int create_srq_kernel(struct mlx5_ib_dev *dev, struct mlx5_ib_srq *srq,
 
 	(*in)->ctx.log_pg_sz = page_shift - MLX5_ADAPTER_PAGE_SHIFT;
 
-	if (MLX5_CAP_GEN(dev->mdev, cqe_version) == MLX5_CQE_VERSION_V1) {
+	if ((MLX5_CAP_GEN(dev->mdev, cqe_version) == MLX5_CQE_VERSION_V1) &&
+	     is_xrc){
 		xsrqc = MLX5_ADDR_OF(create_xrc_srq_in, *in,
 				     xrc_srq_context_entry);
 		/* 0xffffff means we ask to work with cqe version 0 */
@@ -302,10 +302,14 @@ struct ib_srq *mlx5_ib_create_srq(struct ib_pd *pd,
 		    desc_size, init_attr->attr.max_wr, srq->msrq.max, srq->msrq.max_gs,
 		    srq->msrq.max_avail_gather);
 
+	is_xrc = (init_attr->srq_type == IB_SRQT_XRC);
+
 	if (pd->uobject)
-		err = create_srq_user(pd, srq, &in, udata, buf_size, &inlen);
+		err = create_srq_user(pd, srq, &in, udata, buf_size, &inlen,
+				      is_xrc);
 	else
-		err = create_srq_kernel(dev, srq, &in, buf_size, &inlen);
+		err = create_srq_kernel(dev, srq, &in, buf_size, &inlen,
+					is_xrc);
 
 	if (err) {
 		mlx5_ib_warn(dev, "create srq %s failed, err %d\n",
@@ -313,7 +317,6 @@ struct ib_srq *mlx5_ib_create_srq(struct ib_pd *pd,
 		goto err_srq;
 	}
 
-	is_xrc = (init_attr->srq_type == IB_SRQT_XRC);
 	in->ctx.state_log_sz = ilog2(srq->msrq.max);
 	flgs = ((srq->msrq.wqe_shift - 4) | (is_xrc << 5) | (srq->wq_sig << 7)) << 24;
 	xrcdn = 0;
diff --git a/drivers/infiniband/hw/ocrdma/ocrdma.h b/drivers/infiniband/hw/ocrdma/ocrdma.h
index 040bb8b5cb15..12503f15fbd6 100644
--- a/drivers/infiniband/hw/ocrdma/ocrdma.h
+++ b/drivers/infiniband/hw/ocrdma/ocrdma.h
@@ -323,9 +323,6 @@ struct ocrdma_cq {
 			 */
 	u32 max_hw_cqe;
 	bool phase_change;
-	bool deferred_arm, deferred_sol;
-	bool first_arm;
-
 	spinlock_t cq_lock ____cacheline_aligned; /* provide synchronization
 						   * to cq polling
 						   */
diff --git a/drivers/infiniband/hw/ocrdma/ocrdma_main.c b/drivers/infiniband/hw/ocrdma/ocrdma_main.c
index 573849354cb9..f38743018cb4 100644
--- a/drivers/infiniband/hw/ocrdma/ocrdma_main.c
+++ b/drivers/infiniband/hw/ocrdma/ocrdma_main.c
@@ -228,6 +228,11 @@ static int ocrdma_alloc_resources(struct ocrdma_dev *dev)
 
 	ocrdma_alloc_pd_pool(dev);
 
+	if (!ocrdma_alloc_stats_resources(dev)) {
+		pr_err("%s: stats resource allocation failed\n", __func__);
+		goto alloc_err;
+	}
+
 	spin_lock_init(&dev->av_tbl.lock);
 	spin_lock_init(&dev->flush_q_lock);
 	return 0;
@@ -238,6 +243,7 @@ alloc_err:
 
 static void ocrdma_free_resources(struct ocrdma_dev *dev)
 {
+	ocrdma_release_stats_resources(dev);
 	kfree(dev->stag_arr);
 	kfree(dev->qp_tbl);
 	kfree(dev->cq_tbl);
diff --git a/drivers/infiniband/hw/ocrdma/ocrdma_stats.c b/drivers/infiniband/hw/ocrdma/ocrdma_stats.c
index 86c303a620c1..255f774080a4 100644
--- a/drivers/infiniband/hw/ocrdma/ocrdma_stats.c
+++ b/drivers/infiniband/hw/ocrdma/ocrdma_stats.c
@@ -64,10 +64,11 @@ static int ocrdma_add_stat(char *start, char *pcur,
 	return cpy_len;
 }
 
-static bool ocrdma_alloc_stats_mem(struct ocrdma_dev *dev)
+bool ocrdma_alloc_stats_resources(struct ocrdma_dev *dev)
 {
 	struct stats_mem *mem = &dev->stats_mem;
 
+	mutex_init(&dev->stats_lock);
 	/* Alloc mbox command mem*/
 	mem->size = max_t(u32, sizeof(struct ocrdma_rdma_stats_req),
 			sizeof(struct ocrdma_rdma_stats_resp));
@@ -91,13 +92,14 @@ static bool ocrdma_alloc_stats_mem(struct ocrdma_dev *dev)
 	return true;
 }
 
-static void ocrdma_release_stats_mem(struct ocrdma_dev *dev)
+void ocrdma_release_stats_resources(struct ocrdma_dev *dev)
 {
 	struct stats_mem *mem = &dev->stats_mem;
 
 	if (mem->va)
 		dma_free_coherent(&dev->nic_info.pdev->dev, mem->size,
 				  mem->va, mem->pa);
+	mem->va = NULL;
 	kfree(mem->debugfs_mem);
 }
 
@@ -838,15 +840,9 @@ void ocrdma_add_port_stats(struct ocrdma_dev *dev)
 				&dev->reset_stats, &ocrdma_dbg_ops))
 		goto err;
 
-	/* Now create dma_mem for stats mbx command */
-	if (!ocrdma_alloc_stats_mem(dev))
-		goto err;
-
-	mutex_init(&dev->stats_lock);
 
 	return;
 err:
-	ocrdma_release_stats_mem(dev);
 	debugfs_remove_recursive(dev->dir);
 	dev->dir = NULL;
 }
@@ -855,9 +851,7 @@ void ocrdma_rem_port_stats(struct ocrdma_dev *dev)
 {
 	if (!dev->dir)
 		return;
-	debugfs_remove(dev->dir);
-	mutex_destroy(&dev->stats_lock);
-	ocrdma_release_stats_mem(dev);
+	debugfs_remove_recursive(dev->dir);
 }
 
 void ocrdma_init_debugfs(void)
diff --git a/drivers/infiniband/hw/ocrdma/ocrdma_stats.h b/drivers/infiniband/hw/ocrdma/ocrdma_stats.h
index c9e58d04c7b8..bba1fec4f11f 100644
--- a/drivers/infiniband/hw/ocrdma/ocrdma_stats.h
+++ b/drivers/infiniband/hw/ocrdma/ocrdma_stats.h
@@ -65,6 +65,8 @@ enum OCRDMA_STATS_TYPE {
 
 void ocrdma_rem_debugfs(void);
 void ocrdma_init_debugfs(void);
+bool ocrdma_alloc_stats_resources(struct ocrdma_dev *dev);
+void ocrdma_release_stats_resources(struct ocrdma_dev *dev);
 void ocrdma_rem_port_stats(struct ocrdma_dev *dev);
 void ocrdma_add_port_stats(struct ocrdma_dev *dev);
 int ocrdma_pma_counters(struct ocrdma_dev *dev,
diff --git a/drivers/infiniband/hw/ocrdma/ocrdma_verbs.c b/drivers/infiniband/hw/ocrdma/ocrdma_verbs.c
index d4c687b548d8..12420e4ecf3d 100644
--- a/drivers/infiniband/hw/ocrdma/ocrdma_verbs.c
+++ b/drivers/infiniband/hw/ocrdma/ocrdma_verbs.c
@@ -125,8 +125,8 @@ int ocrdma_query_device(struct ib_device *ibdev, struct ib_device_attr *attr,
 					IB_DEVICE_SYS_IMAGE_GUID |
 					IB_DEVICE_LOCAL_DMA_LKEY |
 					IB_DEVICE_MEM_MGT_EXTENSIONS;
-	attr->max_sge = min(dev->attr.max_send_sge, dev->attr.max_srq_sge);
-	attr->max_sge_rd = 0;
+	attr->max_sge = dev->attr.max_send_sge;
+	attr->max_sge_rd = attr->max_sge;
 	attr->max_cq = dev->attr.max_cq;
 	attr->max_cqe = dev->attr.max_cqe;
 	attr->max_mr = dev->attr.max_mr;
@@ -1094,7 +1094,6 @@ struct ib_cq *ocrdma_create_cq(struct ib_device *ibdev,
 	spin_lock_init(&cq->comp_handler_lock);
 	INIT_LIST_HEAD(&cq->sq_head);
 	INIT_LIST_HEAD(&cq->rq_head);
-	cq->first_arm = true;
 
 	if (ib_ctx) {
 		uctx = get_ocrdma_ucontext(ib_ctx);
@@ -2726,8 +2725,7 @@ static int ocrdma_update_ud_rcqe(struct ib_wc *ibwc, struct ocrdma_cqe *cqe)
 		OCRDMA_CQE_UD_STATUS_MASK) >> OCRDMA_CQE_UD_STATUS_SHIFT;
 	ibwc->src_qp = le32_to_cpu(cqe->flags_status_srcqpn) &
 						OCRDMA_CQE_SRCQP_MASK;
-	ibwc->pkey_index = le32_to_cpu(cqe->ud.rxlen_pkey) &
-						OCRDMA_CQE_PKEY_MASK;
+	ibwc->pkey_index = 0;
 	ibwc->wc_flags = IB_WC_GRH;
 	ibwc->byte_len = (le32_to_cpu(cqe->ud.rxlen_pkey) >>
 					OCRDMA_CQE_UD_XFER_LEN_SHIFT);
@@ -2911,12 +2909,9 @@ expand_cqe:
 	}
 stop_cqe:
 	cq->getp = cur_getp;
-	if (cq->deferred_arm || polled_hw_cqes) {
-		ocrdma_ring_cq_db(dev, cq->id, cq->deferred_arm,
-				  cq->deferred_sol, polled_hw_cqes);
-		cq->deferred_arm = false;
-		cq->deferred_sol = false;
-	}
+
+	if (polled_hw_cqes)
+		ocrdma_ring_cq_db(dev, cq->id, false, false, polled_hw_cqes);
 
 	return i;
 }
@@ -3000,13 +2995,7 @@ int ocrdma_arm_cq(struct ib_cq *ibcq, enum ib_cq_notify_flags cq_flags)
 	if (cq_flags & IB_CQ_SOLICITED)
 		sol_needed = true;
 
-	if (cq->first_arm) {
-		ocrdma_ring_cq_db(dev, cq_id, arm_needed, sol_needed, 0);
-		cq->first_arm = false;
-	}
-
-	cq->deferred_arm = true;
-	cq->deferred_sol = sol_needed;
+	ocrdma_ring_cq_db(dev, cq_id, arm_needed, sol_needed, 0);
 	spin_unlock_irqrestore(&cq->cq_lock, flags);
 
 	return 0;
diff --git a/drivers/infiniband/ulp/ipoib/ipoib_ib.c b/drivers/infiniband/ulp/ipoib/ipoib_ib.c
index 5ea0c14070d1..fa9c42ff1fb0 100644
--- a/drivers/infiniband/ulp/ipoib/ipoib_ib.c
+++ b/drivers/infiniband/ulp/ipoib/ipoib_ib.c
@@ -245,8 +245,6 @@ static void ipoib_ib_handle_rx_wc(struct net_device *dev, struct ib_wc *wc)
 	skb_reset_mac_header(skb);
 	skb_pull(skb, IPOIB_ENCAP_LEN);
 
-	skb->truesize = SKB_TRUESIZE(skb->len);
-
 	++dev->stats.rx_packets;
 	dev->stats.rx_bytes += skb->len;
 
diff --git a/drivers/infiniband/ulp/ipoib/ipoib_multicast.c b/drivers/infiniband/ulp/ipoib/ipoib_multicast.c
index 050dfa175d16..25889311b1e9 100644
--- a/drivers/infiniband/ulp/ipoib/ipoib_multicast.c
+++ b/drivers/infiniband/ulp/ipoib/ipoib_multicast.c
@@ -456,7 +456,10 @@ out_locked:
 	return status;
 }
 
-static void ipoib_mcast_join(struct net_device *dev, struct ipoib_mcast *mcast)
+/*
+ * Caller must hold 'priv->lock'
+ */
+static int ipoib_mcast_join(struct net_device *dev, struct ipoib_mcast *mcast)
 {
 	struct ipoib_dev_priv *priv = netdev_priv(dev);
 	struct ib_sa_multicast *multicast;
@@ -466,6 +469,10 @@ static void ipoib_mcast_join(struct net_device *dev, struct ipoib_mcast *mcast)
 	ib_sa_comp_mask comp_mask;
 	int ret = 0;
 
+	if (!priv->broadcast ||
+	    !test_bit(IPOIB_FLAG_OPER_UP, &priv->flags))
+		return -EINVAL;
+
 	ipoib_dbg_mcast(priv, "joining MGID %pI6\n", mcast->mcmember.mgid.raw);
 
 	rec.mgid     = mcast->mcmember.mgid;
@@ -525,20 +532,23 @@ static void ipoib_mcast_join(struct net_device *dev, struct ipoib_mcast *mcast)
 			rec.join_state = 4;
 #endif
 	}
+	spin_unlock_irq(&priv->lock);
 
 	multicast = ib_sa_join_multicast(&ipoib_sa_client, priv->ca, priv->port,
 					 &rec, comp_mask, GFP_KERNEL,
 					 ipoib_mcast_join_complete, mcast);
+	spin_lock_irq(&priv->lock);
 	if (IS_ERR(multicast)) {
 		ret = PTR_ERR(multicast);
 		ipoib_warn(priv, "ib_sa_join_multicast failed, status %d\n", ret);
-		spin_lock_irq(&priv->lock);
 		/* Requeue this join task with a backoff delay */
 		__ipoib_mcast_schedule_join_thread(priv, mcast, 1);
 		clear_bit(IPOIB_MCAST_FLAG_BUSY, &mcast->flags);
 		spin_unlock_irq(&priv->lock);
 		complete(&mcast->done);
+		spin_lock_irq(&priv->lock);
 	}
+	return 0;
 }
 
 void ipoib_mcast_join_task(struct work_struct *work)
@@ -620,9 +630,10 @@ void ipoib_mcast_join_task(struct work_struct *work)
 				/* Found the next unjoined group */
 				init_completion(&mcast->done);
 				set_bit(IPOIB_MCAST_FLAG_BUSY, &mcast->flags);
-				spin_unlock_irq(&priv->lock);
-				ipoib_mcast_join(dev, mcast);
-				spin_lock_irq(&priv->lock);
+				if (ipoib_mcast_join(dev, mcast)) {
+					spin_unlock_irq(&priv->lock);
+					return;
+				}
 			} else if (!delay_until ||
 				 time_before(mcast->delay_until, delay_until))
 				delay_until = mcast->delay_until;
@@ -641,10 +652,9 @@ out:
 	if (mcast) {
 		init_completion(&mcast->done);
 		set_bit(IPOIB_MCAST_FLAG_BUSY, &mcast->flags);
+		ipoib_mcast_join(dev, mcast);
 	}
 	spin_unlock_irq(&priv->lock);
-	if (mcast)
-		ipoib_mcast_join(dev, mcast);
 }
 
 int ipoib_mcast_start_thread(struct net_device *dev)
diff --git a/drivers/input/joystick/xpad.c b/drivers/input/joystick/xpad.c
index 6727954ab74b..e8a84d12b7ff 100644
--- a/drivers/input/joystick/xpad.c
+++ b/drivers/input/joystick/xpad.c
@@ -1207,7 +1207,6 @@ static void xpad_led_disconnect(struct usb_xpad *xpad)
 #else
 static int xpad_led_probe(struct usb_xpad *xpad) { return 0; }
 static void xpad_led_disconnect(struct usb_xpad *xpad) { }
-static void xpad_identify_controller(struct usb_xpad *xpad) { }
 #endif
 
 static int xpad_start_input(struct usb_xpad *xpad)
diff --git a/drivers/input/keyboard/adp5589-keys.c b/drivers/input/keyboard/adp5589-keys.c
index 4d446d5085aa..c01a1d648f9f 100644
--- a/drivers/input/keyboard/adp5589-keys.c
+++ b/drivers/input/keyboard/adp5589-keys.c
@@ -235,7 +235,7 @@ struct adp5589_kpad {
 	unsigned short gpimapsize;
 	unsigned extend_cfg;
 	bool is_adp5585;
-	bool adp5585_support_row5;
+	bool support_row5;
 #ifdef CONFIG_GPIOLIB
 	unsigned char gpiomap[ADP5589_MAXGPIO];
 	bool export_gpio;
@@ -485,7 +485,7 @@ static int adp5589_build_gpiomap(struct adp5589_kpad *kpad,
 	if (kpad->extend_cfg & C4_EXTEND_CFG)
 		pin_used[kpad->var->c4_extend_cfg] = true;
 
-	if (!kpad->adp5585_support_row5)
+	if (!kpad->support_row5)
 		pin_used[5] = true;
 
 	for (i = 0; i < kpad->var->maxgpio; i++)
@@ -884,12 +884,13 @@ static int adp5589_probe(struct i2c_client *client,
 
 	switch (id->driver_data) {
 	case ADP5585_02:
-		kpad->adp5585_support_row5 = true;
+		kpad->support_row5 = true;
 	case ADP5585_01:
 		kpad->is_adp5585 = true;
 		kpad->var = &const_adp5585;
 		break;
 	case ADP5589:
+		kpad->support_row5 = true;
 		kpad->var = &const_adp5589;
 		break;
 	}
diff --git a/drivers/input/keyboard/cap11xx.c b/drivers/input/keyboard/cap11xx.c
index 378db10001df..4401be225d64 100644
--- a/drivers/input/keyboard/cap11xx.c
+++ b/drivers/input/keyboard/cap11xx.c
@@ -304,8 +304,10 @@ static int cap11xx_init_leds(struct device *dev,
 		led->cdev.brightness = LED_OFF;
 
 		error = of_property_read_u32(child, "reg", &reg);
-		if (error != 0 || reg >= num_leds)
+		if (error != 0 || reg >= num_leds) {
+			of_node_put(child);
 			return -EINVAL;
+		}
 
 		led->reg = reg;
 		led->priv = priv;
@@ -313,8 +315,10 @@ static int cap11xx_init_leds(struct device *dev,
 		INIT_WORK(&led->work, cap11xx_led_work);
 
 		error = devm_led_classdev_register(dev, &led->cdev);
-		if (error)
+		if (error) {
+			of_node_put(child);
 			return error;
+		}
 
 		priv->num_leds++;
 		led++;
diff --git a/drivers/input/misc/Kconfig b/drivers/input/misc/Kconfig
index d6d16fa78281..1f2337abcf2f 100644
--- a/drivers/input/misc/Kconfig
+++ b/drivers/input/misc/Kconfig
@@ -733,7 +733,7 @@ config INPUT_XEN_KBDDEV_FRONTEND
 	  module will be called xen-kbdfront.
 
 config INPUT_SIRFSOC_ONKEY
-	bool "CSR SiRFSoC power on/off/suspend key support"
+	tristate "CSR SiRFSoC power on/off/suspend key support"
 	depends on ARCH_SIRF && OF
 	default y
 	help
diff --git a/drivers/input/misc/sirfsoc-onkey.c b/drivers/input/misc/sirfsoc-onkey.c
index 9d5b89befe6f..ed7237f19539 100644
--- a/drivers/input/misc/sirfsoc-onkey.c
+++ b/drivers/input/misc/sirfsoc-onkey.c
@@ -101,7 +101,7 @@ static void sirfsoc_pwrc_close(struct input_dev *input)
 static const struct of_device_id sirfsoc_pwrc_of_match[] = {
 	{ .compatible = "sirf,prima2-pwrc" },
 	{},
-}
+};
 MODULE_DEVICE_TABLE(of, sirfsoc_pwrc_of_match);
 
 static int sirfsoc_pwrc_probe(struct platform_device *pdev)
diff --git a/drivers/input/mouse/vmmouse.c b/drivers/input/mouse/vmmouse.c
index e272f06258ce..a3f0f5a47490 100644
--- a/drivers/input/mouse/vmmouse.c
+++ b/drivers/input/mouse/vmmouse.c
@@ -458,8 +458,6 @@ int vmmouse_init(struct psmouse *psmouse)
 	priv->abs_dev = abs_dev;
 	psmouse->private = priv;
 
-	input_set_capability(rel_dev, EV_REL, REL_WHEEL);
-
 	/* Set up and register absolute device */
 	snprintf(priv->phys, sizeof(priv->phys), "%s/input1",
 		 psmouse->ps2dev.serio->phys);
@@ -475,10 +473,6 @@ int vmmouse_init(struct psmouse *psmouse)
 	abs_dev->id.version = psmouse->model;
 	abs_dev->dev.parent = &psmouse->ps2dev.serio->dev;
 
-	error = input_register_device(priv->abs_dev);
-	if (error)
-		goto init_fail;
-
 	/* Set absolute device capabilities */
 	input_set_capability(abs_dev, EV_KEY, BTN_LEFT);
 	input_set_capability(abs_dev, EV_KEY, BTN_RIGHT);
@@ -488,6 +482,13 @@ int vmmouse_init(struct psmouse *psmouse)
 	input_set_abs_params(abs_dev, ABS_X, 0, VMMOUSE_MAX_X, 0, 0);
 	input_set_abs_params(abs_dev, ABS_Y, 0, VMMOUSE_MAX_Y, 0, 0);
 
+	error = input_register_device(priv->abs_dev);
+	if (error)
+		goto init_fail;
+
+	/* Add wheel capability to the relative device */
+	input_set_capability(rel_dev, EV_REL, REL_WHEEL);
+
 	psmouse->protocol_handler = vmmouse_process_byte;
 	psmouse->disconnect = vmmouse_disconnect;
 	psmouse->reconnect = vmmouse_reconnect;
diff --git a/drivers/input/serio/serio.c b/drivers/input/serio/serio.c
index 8f828975ab10..1ca7f551e2da 100644
--- a/drivers/input/serio/serio.c
+++ b/drivers/input/serio/serio.c
@@ -134,7 +134,7 @@ static void serio_find_driver(struct serio *serio)
 	int error;
 
 	error = device_attach(&serio->dev);
-	if (error < 0)
+	if (error < 0 && error != -EPROBE_DEFER)
 		dev_warn(&serio->dev,
 			 "device_attach() failed for %s (%s), error: %d\n",
 			 serio->phys, serio->name, error);
diff --git a/drivers/input/touchscreen/colibri-vf50-ts.c b/drivers/input/touchscreen/colibri-vf50-ts.c
index 5d4903a402cc..69828d015d45 100644
--- a/drivers/input/touchscreen/colibri-vf50-ts.c
+++ b/drivers/input/touchscreen/colibri-vf50-ts.c
@@ -21,6 +21,7 @@
 #include <linux/interrupt.h>
 #include <linux/kernel.h>
 #include <linux/module.h>
+#include <linux/of.h>
 #include <linux/pinctrl/consumer.h>
 #include <linux/platform_device.h>
 #include <linux/slab.h>
diff --git a/drivers/input/touchscreen/edt-ft5x06.c b/drivers/input/touchscreen/edt-ft5x06.c
index 0b0f8c17f3f7..23fbe382da8b 100644
--- a/drivers/input/touchscreen/edt-ft5x06.c
+++ b/drivers/input/touchscreen/edt-ft5x06.c
@@ -822,16 +822,22 @@ static void edt_ft5x06_ts_get_defaults(struct device *dev,
 	int error;
 
 	error = device_property_read_u32(dev, "threshold", &val);
-	if (!error)
-		reg_addr->reg_threshold = val;
+	if (!error) {
+		edt_ft5x06_register_write(tsdata, reg_addr->reg_threshold, val);
+		tsdata->threshold = val;
+	}
 
 	error = device_property_read_u32(dev, "gain", &val);
-	if (!error)
-		reg_addr->reg_gain = val;
+	if (!error) {
+		edt_ft5x06_register_write(tsdata, reg_addr->reg_gain, val);
+		tsdata->gain = val;
+	}
 
 	error = device_property_read_u32(dev, "offset", &val);
-	if (!error)
-		reg_addr->reg_offset = val;
+	if (!error) {
+		edt_ft5x06_register_write(tsdata, reg_addr->reg_offset, val);
+		tsdata->offset = val;
+	}
 }
 
 static void
diff --git a/drivers/iommu/amd_iommu.c b/drivers/iommu/amd_iommu.c
index e5e223938eec..374c129219ef 100644
--- a/drivers/iommu/amd_iommu.c
+++ b/drivers/iommu/amd_iommu.c
@@ -114,6 +114,7 @@ struct kmem_cache *amd_iommu_irq_cache;
 
 static void update_domain(struct protection_domain *domain);
 static int protection_domain_init(struct protection_domain *domain);
+static void detach_device(struct device *dev);
 
 /*
  * For dynamic growth the aperture size is split into ranges of 128MB of
@@ -384,6 +385,9 @@ static void iommu_uninit_device(struct device *dev)
 	if (!dev_data)
 		return;
 
+	if (dev_data->domain)
+		detach_device(dev);
+
 	iommu_device_unlink(amd_iommu_rlookup_table[dev_data->devid]->iommu_dev,
 			    dev);
 
diff --git a/drivers/iommu/amd_iommu_init.c b/drivers/iommu/amd_iommu_init.c
index 013bdfff2d4d..bf4959f4225b 100644
--- a/drivers/iommu/amd_iommu_init.c
+++ b/drivers/iommu/amd_iommu_init.c
@@ -228,6 +228,10 @@ static int amd_iommu_enable_interrupts(void);
 static int __init iommu_go_to_state(enum iommu_init_state state);
 static void init_device_table_dma(void);
 
+static int iommu_pc_get_set_reg_val(struct amd_iommu *iommu,
+				    u8 bank, u8 cntr, u8 fxn,
+				    u64 *value, bool is_write);
+
 static inline void update_last_devid(u16 devid)
 {
 	if (devid > amd_iommu_last_bdf)
@@ -1016,6 +1020,34 @@ static void amd_iommu_erratum_746_workaround(struct amd_iommu *iommu)
 }
 
 /*
+ * Family15h Model 30h-3fh (IOMMU Mishandles ATS Write Permission)
+ * Workaround:
+ *     BIOS should enable ATS write permission check by setting
+ *     L2_DEBUG_3[AtsIgnoreIWDis](D0F2xF4_x47[0]) = 1b
+ */
+static void amd_iommu_ats_write_check_workaround(struct amd_iommu *iommu)
+{
+	u32 value;
+
+	if ((boot_cpu_data.x86 != 0x15) ||
+	    (boot_cpu_data.x86_model < 0x30) ||
+	    (boot_cpu_data.x86_model > 0x3f))
+		return;
+
+	/* Test L2_DEBUG_3[AtsIgnoreIWDis] == 1 */
+	value = iommu_read_l2(iommu, 0x47);
+
+	if (value & BIT(0))
+		return;
+
+	/* Set L2_DEBUG_3[AtsIgnoreIWDis] = 1 */
+	iommu_write_l2(iommu, 0x47, value | BIT(0));
+
+	pr_info("AMD-Vi: Applying ATS write check workaround for IOMMU at %s\n",
+		dev_name(&iommu->dev->dev));
+}
+
+/*
  * This function clues the initialization function for one IOMMU
  * together and also allocates the command buffer and programs the
  * hardware. It does NOT enable the IOMMU. This is done afterwards.
@@ -1142,8 +1174,8 @@ static void init_iommu_perf_ctr(struct amd_iommu *iommu)
 	amd_iommu_pc_present = true;
 
 	/* Check if the performance counters can be written to */
-	if ((0 != amd_iommu_pc_get_set_reg_val(0, 0, 0, 0, &val, true)) ||
-	    (0 != amd_iommu_pc_get_set_reg_val(0, 0, 0, 0, &val2, false)) ||
+	if ((0 != iommu_pc_get_set_reg_val(iommu, 0, 0, 0, &val, true)) ||
+	    (0 != iommu_pc_get_set_reg_val(iommu, 0, 0, 0, &val2, false)) ||
 	    (val != val2)) {
 		pr_err("AMD-Vi: Unable to write to IOMMU perf counter.\n");
 		amd_iommu_pc_present = false;
@@ -1284,6 +1316,7 @@ static int iommu_init_pci(struct amd_iommu *iommu)
 	}
 
 	amd_iommu_erratum_746_workaround(iommu);
+	amd_iommu_ats_write_check_workaround(iommu);
 
 	iommu->iommu_dev = iommu_device_create(&iommu->dev->dev, iommu,
 					       amd_iommu_groups, "ivhd%d",
@@ -2283,22 +2316,15 @@ u8 amd_iommu_pc_get_max_counters(u16 devid)
 }
 EXPORT_SYMBOL(amd_iommu_pc_get_max_counters);
 
-int amd_iommu_pc_get_set_reg_val(u16 devid, u8 bank, u8 cntr, u8 fxn,
+static int iommu_pc_get_set_reg_val(struct amd_iommu *iommu,
+				    u8 bank, u8 cntr, u8 fxn,
 				    u64 *value, bool is_write)
 {
-	struct amd_iommu *iommu;
 	u32 offset;
 	u32 max_offset_lim;
 
-	/* Make sure the IOMMU PC resource is available */
-	if (!amd_iommu_pc_present)
-		return -ENODEV;
-
-	/* Locate the iommu associated with the device ID */
-	iommu = amd_iommu_rlookup_table[devid];
-
 	/* Check for valid iommu and pc register indexing */
-	if (WARN_ON((iommu == NULL) || (fxn > 0x28) || (fxn & 7)))
+	if (WARN_ON((fxn > 0x28) || (fxn & 7)))
 		return -ENODEV;
 
 	offset = (u32)(((0x40|bank) << 12) | (cntr << 8) | fxn);
@@ -2322,3 +2348,16 @@ int amd_iommu_pc_get_set_reg_val(u16 devid, u8 bank, u8 cntr, u8 fxn,
 	return 0;
 }
 EXPORT_SYMBOL(amd_iommu_pc_get_set_reg_val);
+
+int amd_iommu_pc_get_set_reg_val(u16 devid, u8 bank, u8 cntr, u8 fxn,
+				    u64 *value, bool is_write)
+{
+	struct amd_iommu *iommu = amd_iommu_rlookup_table[devid];
+
+	/* Make sure the IOMMU PC resource is available */
+	if (!amd_iommu_pc_present || iommu == NULL)
+		return -ENODEV;
+
+	return iommu_pc_get_set_reg_val(iommu, bank, cntr, fxn,
+					value, is_write);
+}
diff --git a/drivers/iommu/dmar.c b/drivers/iommu/dmar.c
index 62a400c5ba06..8ffd7568fc91 100644
--- a/drivers/iommu/dmar.c
+++ b/drivers/iommu/dmar.c
@@ -329,7 +329,8 @@ static int dmar_pci_bus_notifier(struct notifier_block *nb,
 	/* Only care about add/remove events for physical functions */
 	if (pdev->is_virtfn)
 		return NOTIFY_DONE;
-	if (action != BUS_NOTIFY_ADD_DEVICE && action != BUS_NOTIFY_DEL_DEVICE)
+	if (action != BUS_NOTIFY_ADD_DEVICE &&
+	    action != BUS_NOTIFY_REMOVED_DEVICE)
 		return NOTIFY_DONE;
 
 	info = dmar_alloc_pci_notify_info(pdev, action);
@@ -339,7 +340,7 @@ static int dmar_pci_bus_notifier(struct notifier_block *nb,
 	down_write(&dmar_global_lock);
 	if (action == BUS_NOTIFY_ADD_DEVICE)
 		dmar_pci_bus_add_dev(info);
-	else if (action == BUS_NOTIFY_DEL_DEVICE)
+	else if (action == BUS_NOTIFY_REMOVED_DEVICE)
 		dmar_pci_bus_del_dev(info);
 	up_write(&dmar_global_lock);
 
@@ -1353,7 +1354,7 @@ void dmar_disable_qi(struct intel_iommu *iommu)
 
 	raw_spin_lock_irqsave(&iommu->register_lock, flags);
 
-	sts =  dmar_readq(iommu->reg + DMAR_GSTS_REG);
+	sts =  readl(iommu->reg + DMAR_GSTS_REG);
 	if (!(sts & DMA_GSTS_QIES))
 		goto end;
 
diff --git a/drivers/iommu/intel-iommu.c b/drivers/iommu/intel-iommu.c
index 986a53e3eb96..a2e1b7f14df2 100644
--- a/drivers/iommu/intel-iommu.c
+++ b/drivers/iommu/intel-iommu.c
@@ -4367,7 +4367,7 @@ int dmar_iommu_notify_scope_dev(struct dmar_pci_notify_info *info)
 				rmrru->devices_cnt);
 			if(ret < 0)
 				return ret;
-		} else if (info->event == BUS_NOTIFY_DEL_DEVICE) {
+		} else if (info->event == BUS_NOTIFY_REMOVED_DEVICE) {
 			dmar_remove_dev_scope(info, rmrr->segment,
 				rmrru->devices, rmrru->devices_cnt);
 		}
@@ -4387,7 +4387,7 @@ int dmar_iommu_notify_scope_dev(struct dmar_pci_notify_info *info)
 				break;
 			else if(ret < 0)
 				return ret;
-		} else if (info->event == BUS_NOTIFY_DEL_DEVICE) {
+		} else if (info->event == BUS_NOTIFY_REMOVED_DEVICE) {
 			if (dmar_remove_dev_scope(info, atsr->segment,
 					atsru->devices, atsru->devices_cnt))
 				break;
diff --git a/drivers/iommu/intel-svm.c b/drivers/iommu/intel-svm.c
index 50464833d0b8..d9939fa9b588 100644
--- a/drivers/iommu/intel-svm.c
+++ b/drivers/iommu/intel-svm.c
@@ -249,12 +249,30 @@ static void intel_flush_pasid_dev(struct intel_svm *svm, struct intel_svm_dev *s
 static void intel_mm_release(struct mmu_notifier *mn, struct mm_struct *mm)
 {
 	struct intel_svm *svm = container_of(mn, struct intel_svm, notifier);
+	struct intel_svm_dev *sdev;
 
+	/* This might end up being called from exit_mmap(), *before* the page
+	 * tables are cleared. And __mmu_notifier_release() will delete us from
+	 * the list of notifiers so that our invalidate_range() callback doesn't
+	 * get called when the page tables are cleared. So we need to protect
+	 * against hardware accessing those page tables.
+	 *
+	 * We do it by clearing the entry in the PASID table and then flushing
+	 * the IOTLB and the PASID table caches. This might upset hardware;
+	 * perhaps we'll want to point the PASID to a dummy PGD (like the zero
+	 * page) so that we end up taking a fault that the hardware really
+	 * *has* to handle gracefully without affecting other processes.
+	 */
 	svm->iommu->pasid_table[svm->pasid].val = 0;
+	wmb();
+
+	rcu_read_lock();
+	list_for_each_entry_rcu(sdev, &svm->devs, list) {
+		intel_flush_pasid_dev(svm, sdev, svm->pasid);
+		intel_flush_svm_range_dev(svm, sdev, 0, -1, 0, !svm->mm);
+	}
+	rcu_read_unlock();
 
-	/* There's no need to do any flush because we can't get here if there
-	 * are any devices left anyway. */
-	WARN_ON(!list_empty(&svm->devs));
 }
 
 static const struct mmu_notifier_ops intel_mmuops = {
@@ -379,7 +397,6 @@ int intel_svm_bind_mm(struct device *dev, int *pasid, int flags, struct svm_dev_
 				goto out;
 			}
 			iommu->pasid_table[svm->pasid].val = (u64)__pa(mm->pgd) | 1;
-			mm = NULL;
 		} else
 			iommu->pasid_table[svm->pasid].val = (u64)__pa(init_mm.pgd) | 1 | (1ULL << 11);
 		wmb();
@@ -442,11 +459,11 @@ int intel_svm_unbind_mm(struct device *dev, int pasid)
 				kfree_rcu(sdev, rcu);
 
 				if (list_empty(&svm->devs)) {
-					mmu_notifier_unregister(&svm->notifier, svm->mm);
 
 					idr_remove(&svm->iommu->pasid_idr, svm->pasid);
 					if (svm->mm)
-						mmput(svm->mm);
+						mmu_notifier_unregister(&svm->notifier, svm->mm);
+
 					/* We mandate that no page faults may be outstanding
 					 * for the PASID when intel_svm_unbind_mm() is called.
 					 * If that is not obeyed, subtle errors will happen.
@@ -507,6 +524,10 @@ static irqreturn_t prq_event_thread(int irq, void *d)
 	struct intel_svm *svm = NULL;
 	int head, tail, handled = 0;
 
+	/* Clear PPR bit before reading head/tail registers, to
+	 * ensure that we get a new interrupt if needed. */
+	writel(DMA_PRS_PPR, iommu->reg + DMAR_PRS_REG);
+
 	tail = dmar_readq(iommu->reg + DMAR_PQT_REG) & PRQ_RING_MASK;
 	head = dmar_readq(iommu->reg + DMAR_PQH_REG) & PRQ_RING_MASK;
 	while (head != tail) {
@@ -551,6 +572,9 @@ static irqreturn_t prq_event_thread(int irq, void *d)
 		 * any faults on kernel addresses. */
 		if (!svm->mm)
 			goto bad_req;
+		/* If the mm is already defunct, don't handle faults. */
+		if (!atomic_inc_not_zero(&svm->mm->mm_users))
+			goto bad_req;
 		down_read(&svm->mm->mmap_sem);
 		vma = find_extend_vma(svm->mm, address);
 		if (!vma || address < vma->vm_start)
@@ -567,6 +591,7 @@ static irqreturn_t prq_event_thread(int irq, void *d)
 		result = QI_RESP_SUCCESS;
 	invalid:
 		up_read(&svm->mm->mmap_sem);
+		mmput(svm->mm);
 	bad_req:
 		/* Accounting for major/minor faults? */
 		rcu_read_lock();
diff --git a/drivers/iommu/intel_irq_remapping.c b/drivers/iommu/intel_irq_remapping.c
index c12ba4516df2..ac596928f6b4 100644
--- a/drivers/iommu/intel_irq_remapping.c
+++ b/drivers/iommu/intel_irq_remapping.c
@@ -629,7 +629,7 @@ static void iommu_disable_irq_remapping(struct intel_iommu *iommu)
 
 	raw_spin_lock_irqsave(&iommu->register_lock, flags);
 
-	sts = dmar_readq(iommu->reg + DMAR_GSTS_REG);
+	sts = readl(iommu->reg + DMAR_GSTS_REG);
 	if (!(sts & DMA_GSTS_IRES))
 		goto end;
 
diff --git a/drivers/irqchip/irq-gic-v3-its.c b/drivers/irqchip/irq-gic-v3-its.c
index 3447549fcc93..43dfd15c1dd2 100644
--- a/drivers/irqchip/irq-gic-v3-its.c
+++ b/drivers/irqchip/irq-gic-v3-its.c
@@ -66,7 +66,10 @@ struct its_node {
 	unsigned long		phys_base;
 	struct its_cmd_block	*cmd_base;
 	struct its_cmd_block	*cmd_write;
-	void			*tables[GITS_BASER_NR_REGS];
+	struct {
+		void		*base;
+		u32		order;
+	} tables[GITS_BASER_NR_REGS];
 	struct its_collection	*collections;
 	struct list_head	its_device_list;
 	u64			flags;
@@ -75,6 +78,9 @@ struct its_node {
 
 #define ITS_ITT_ALIGN		SZ_256
 
+/* Convert page order to size in bytes */
+#define PAGE_ORDER_TO_SIZE(o)	(PAGE_SIZE << (o))
+
 struct event_lpi_map {
 	unsigned long		*lpi_map;
 	u16			*col_map;
@@ -597,11 +603,6 @@ static void its_unmask_irq(struct irq_data *d)
 	lpi_set_config(d, true);
 }
 
-static void its_eoi_irq(struct irq_data *d)
-{
-	gic_write_eoir(d->hwirq);
-}
-
 static int its_set_affinity(struct irq_data *d, const struct cpumask *mask_val,
 			    bool force)
 {
@@ -638,7 +639,7 @@ static struct irq_chip its_irq_chip = {
 	.name			= "ITS",
 	.irq_mask		= its_mask_irq,
 	.irq_unmask		= its_unmask_irq,
-	.irq_eoi		= its_eoi_irq,
+	.irq_eoi		= irq_chip_eoi_parent,
 	.irq_set_affinity	= its_set_affinity,
 	.irq_compose_msi_msg	= its_irq_compose_msi_msg,
 };
@@ -807,9 +808,10 @@ static void its_free_tables(struct its_node *its)
 	int i;
 
 	for (i = 0; i < GITS_BASER_NR_REGS; i++) {
-		if (its->tables[i]) {
-			free_page((unsigned long)its->tables[i]);
-			its->tables[i] = NULL;
+		if (its->tables[i].base) {
+			free_pages((unsigned long)its->tables[i].base,
+				   its->tables[i].order);
+			its->tables[i].base = NULL;
 		}
 	}
 }
@@ -842,7 +844,6 @@ static int its_alloc_tables(const char *node_name, struct its_node *its)
 		u64 type = GITS_BASER_TYPE(val);
 		u64 entry_size = GITS_BASER_ENTRY_SIZE(val);
 		int order = get_order(psz);
-		int alloc_size;
 		int alloc_pages;
 		u64 tmp;
 		void *base;
@@ -874,9 +875,8 @@ static int its_alloc_tables(const char *node_name, struct its_node *its)
 			}
 		}
 
-		alloc_size = (1 << order) * PAGE_SIZE;
 retry_alloc_baser:
-		alloc_pages = (alloc_size / psz);
+		alloc_pages = (PAGE_ORDER_TO_SIZE(order) / psz);
 		if (alloc_pages > GITS_BASER_PAGES_MAX) {
 			alloc_pages = GITS_BASER_PAGES_MAX;
 			order = get_order(GITS_BASER_PAGES_MAX * psz);
@@ -890,7 +890,8 @@ retry_alloc_baser:
 			goto out_free;
 		}
 
-		its->tables[i] = base;
+		its->tables[i].base = base;
+		its->tables[i].order = order;
 
 retry_baser:
 		val = (virt_to_phys(base) 				 |
@@ -928,7 +929,7 @@ retry_baser:
 			shr = tmp & GITS_BASER_SHAREABILITY_MASK;
 			if (!shr) {
 				cache = GITS_BASER_nC;
-				__flush_dcache_area(base, alloc_size);
+				__flush_dcache_area(base, PAGE_ORDER_TO_SIZE(order));
 			}
 			goto retry_baser;
 		}
@@ -940,7 +941,7 @@ retry_baser:
 			 * something is horribly wrong...
 			 */
 			free_pages((unsigned long)base, order);
-			its->tables[i] = NULL;
+			its->tables[i].base = NULL;
 
 			switch (psz) {
 			case SZ_16K:
@@ -961,7 +962,7 @@ retry_baser:
 		}
 
 		pr_info("ITS: allocated %d %s @%lx (psz %dK, shr %d)\n",
-			(int)(alloc_size / entry_size),
+			(int)(PAGE_ORDER_TO_SIZE(order) / entry_size),
 			its_base_type_string[type],
 			(unsigned long)virt_to_phys(base),
 			psz / SZ_1K, (int)shr >> GITS_BASER_SHAREABILITY_SHIFT);
diff --git a/drivers/irqchip/irq-gic.c b/drivers/irqchip/irq-gic.c
index 911758c056c1..8f9ebf714e2b 100644
--- a/drivers/irqchip/irq-gic.c
+++ b/drivers/irqchip/irq-gic.c
@@ -384,9 +384,6 @@ static struct irq_chip gic_chip = {
 	.irq_unmask		= gic_unmask_irq,
 	.irq_eoi		= gic_eoi_irq,
 	.irq_set_type		= gic_set_type,
-#ifdef CONFIG_SMP
-	.irq_set_affinity	= gic_set_affinity,
-#endif
 	.irq_get_irqchip_state	= gic_irq_get_irqchip_state,
 	.irq_set_irqchip_state	= gic_irq_set_irqchip_state,
 	.flags			= IRQCHIP_SET_TYPE_MASKED |
@@ -400,9 +397,6 @@ static struct irq_chip gic_eoimode1_chip = {
 	.irq_unmask		= gic_unmask_irq,
 	.irq_eoi		= gic_eoimode1_eoi_irq,
 	.irq_set_type		= gic_set_type,
-#ifdef CONFIG_SMP
-	.irq_set_affinity	= gic_set_affinity,
-#endif
 	.irq_get_irqchip_state	= gic_irq_get_irqchip_state,
 	.irq_set_irqchip_state	= gic_irq_set_irqchip_state,
 	.irq_set_vcpu_affinity	= gic_irq_set_vcpu_affinity,
@@ -443,7 +437,7 @@ static void gic_cpu_if_up(struct gic_chip_data *gic)
 	u32 bypass = 0;
 	u32 mode = 0;
 
-	if (static_key_true(&supports_deactivate))
+	if (gic == &gic_data[0] && static_key_true(&supports_deactivate))
 		mode = GIC_CPU_CTRL_EOImodeNS;
 
 	/*
@@ -1039,6 +1033,11 @@ static void __init __gic_init_bases(unsigned int gic_nr, int irq_start,
 		gic->chip.name = kasprintf(GFP_KERNEL, "GIC-%d", gic_nr);
 	}
 
+#ifdef CONFIG_SMP
+	if (gic_nr == 0)
+		gic->chip.irq_set_affinity = gic_set_affinity;
+#endif
+
 #ifdef CONFIG_GIC_NON_BANKED
 	if (percpu_offset) { /* Frankein-GIC without banked registers... */
 		unsigned int cpu;
diff --git a/drivers/irqchip/irq-sun4i.c b/drivers/irqchip/irq-sun4i.c
index 0704362f4c82..376b28074e0d 100644
--- a/drivers/irqchip/irq-sun4i.c
+++ b/drivers/irqchip/irq-sun4i.c
@@ -22,7 +22,6 @@
 #include <linux/of_irq.h>
 
 #include <asm/exception.h>
-#include <asm/mach/irq.h>
 
 #define SUN4I_IRQ_VECTOR_REG		0x00
 #define SUN4I_IRQ_PROTECTION_REG	0x08
diff --git a/drivers/isdn/gigaset/ser-gigaset.c b/drivers/isdn/gigaset/ser-gigaset.c
index 2a506fe0c8a4..d1f8ab915b15 100644
--- a/drivers/isdn/gigaset/ser-gigaset.c
+++ b/drivers/isdn/gigaset/ser-gigaset.c
@@ -373,13 +373,7 @@ static void gigaset_freecshw(struct cardstate *cs)
 
 static void gigaset_device_release(struct device *dev)
 {
-	struct cardstate *cs = dev_get_drvdata(dev);
-
-	if (!cs)
-		return;
-	dev_set_drvdata(dev, NULL);
-	kfree(cs->hw.ser);
-	cs->hw.ser = NULL;
+	kfree(container_of(dev, struct ser_cardstate, dev.dev));
 }
 
 /*
@@ -408,7 +402,6 @@ static int gigaset_initcshw(struct cardstate *cs)
 		cs->hw.ser = NULL;
 		return rc;
 	}
-	dev_set_drvdata(&cs->hw.ser->dev.dev, cs);
 
 	tasklet_init(&cs->write_tasklet,
 		     gigaset_modem_fill, (unsigned long) cs);
diff --git a/drivers/isdn/hardware/mISDN/netjet.c b/drivers/isdn/hardware/mISDN/netjet.c
index 8e2944784e00..afde4edef9ae 100644
--- a/drivers/isdn/hardware/mISDN/netjet.c
+++ b/drivers/isdn/hardware/mISDN/netjet.c
@@ -392,7 +392,7 @@ read_dma(struct tiger_ch *bc, u32 idx, int cnt)
 	}
 	stat = bchannel_get_rxbuf(&bc->bch, cnt);
 	/* only transparent use the count here, HDLC overun is detected later */
-	if (stat == ENOMEM) {
+	if (stat == -ENOMEM) {
 		pr_warning("%s.B%d: No memory for %d bytes\n",
 			   card->name, bc->bch.nr, cnt);
 		return;
diff --git a/drivers/lightnvm/core.c b/drivers/lightnvm/core.c
index 33224cb91c5b..9f6acd5d1d2e 100644
--- a/drivers/lightnvm/core.c
+++ b/drivers/lightnvm/core.c
@@ -572,11 +572,13 @@ int nvm_register(struct request_queue *q, char *disk_name,
 		}
 	}
 
-	ret = nvm_get_sysblock(dev, &dev->sb);
-	if (!ret)
-		pr_err("nvm: device not initialized.\n");
-	else if (ret < 0)
-		pr_err("nvm: err (%d) on device initialization\n", ret);
+	if (dev->identity.cap & NVM_ID_DCAP_BBLKMGMT) {
+		ret = nvm_get_sysblock(dev, &dev->sb);
+		if (!ret)
+			pr_err("nvm: device not initialized.\n");
+		else if (ret < 0)
+			pr_err("nvm: err (%d) on device initialization\n", ret);
+	}
 
 	/* register device with a supported media manager */
 	down_write(&nvm_lock);
@@ -1055,9 +1057,11 @@ static long __nvm_ioctl_dev_init(struct nvm_ioctl_dev_init *init)
 	strncpy(info.mmtype, init->mmtype, NVM_MMTYPE_LEN);
 	info.fs_ppa.ppa = -1;
 
-	ret = nvm_init_sysblock(dev, &info);
-	if (ret)
-		return ret;
+	if (dev->identity.cap & NVM_ID_DCAP_BBLKMGMT) {
+		ret = nvm_init_sysblock(dev, &info);
+		if (ret)
+			return ret;
+	}
 
 	memcpy(&dev->sb, &info, sizeof(struct nvm_sb_info));
 
@@ -1117,7 +1121,10 @@ static long nvm_ioctl_dev_factory(struct file *file, void __user *arg)
 		dev->mt = NULL;
 	}
 
-	return nvm_dev_factory(dev, fact.flags);
+	if (dev->identity.cap & NVM_ID_DCAP_BBLKMGMT)
+		return nvm_dev_factory(dev, fact.flags);
+
+	return 0;
 }
 
 static long nvm_ctl_ioctl(struct file *file, uint cmd, unsigned long arg)
diff --git a/drivers/lightnvm/rrpc.c b/drivers/lightnvm/rrpc.c
index d8c75958ced3..307db1ea22de 100644
--- a/drivers/lightnvm/rrpc.c
+++ b/drivers/lightnvm/rrpc.c
@@ -300,8 +300,10 @@ static int rrpc_move_valid_pages(struct rrpc *rrpc, struct rrpc_block *rblk)
 	}
 
 	page = mempool_alloc(rrpc->page_pool, GFP_NOIO);
-	if (!page)
+	if (!page) {
+		bio_put(bio);
 		return -ENOMEM;
+	}
 
 	while ((slot = find_first_zero_bit(rblk->invalid_pages,
 					    nr_pgs_per_blk)) < nr_pgs_per_blk) {
diff --git a/drivers/lightnvm/rrpc.h b/drivers/lightnvm/rrpc.h
index ef13ac7700c8..f7b37336353f 100644
--- a/drivers/lightnvm/rrpc.h
+++ b/drivers/lightnvm/rrpc.h
@@ -174,8 +174,7 @@ static inline sector_t rrpc_get_sector(sector_t laddr)
 static inline int request_intersects(struct rrpc_inflight_rq *r,
 				sector_t laddr_start, sector_t laddr_end)
 {
-	return (laddr_end >= r->l_start && laddr_end <= r->l_end) &&
-		(laddr_start >= r->l_start && laddr_start <= r->l_end);
+	return (laddr_end >= r->l_start) && (laddr_start <= r->l_end);
 }
 
 static int __rrpc_lock_laddr(struct rrpc *rrpc, sector_t laddr,
@@ -184,6 +183,8 @@ static int __rrpc_lock_laddr(struct rrpc *rrpc, sector_t laddr,
 	sector_t laddr_end = laddr + pages - 1;
 	struct rrpc_inflight_rq *rtmp;
 
+	WARN_ON(irqs_disabled());
+
 	spin_lock_irq(&rrpc->inflights.lock);
 	list_for_each_entry(rtmp, &rrpc->inflights.reqs, list) {
 		if (unlikely(request_intersects(rtmp, laddr, laddr_end))) {
diff --git a/drivers/md/dm.c b/drivers/md/dm.c
index 5df40480228b..dd834927bc66 100644
--- a/drivers/md/dm.c
+++ b/drivers/md/dm.c
@@ -1191,6 +1191,8 @@ static void dm_unprep_request(struct request *rq)
 
 	if (clone)
 		free_rq_clone(clone);
+	else if (!tio->md->queue->mq_ops)
+		free_rq_tio(tio);
 }
 
 /*
diff --git a/drivers/media/i2c/adv7604.c b/drivers/media/i2c/adv7604.c
index d35e0ba8b269..befb07df036d 100644
--- a/drivers/media/i2c/adv7604.c
+++ b/drivers/media/i2c/adv7604.c
@@ -1946,10 +1946,9 @@ static int adv76xx_isr(struct v4l2_subdev *sd, u32 status, bool *handled)
 	}
 
 	/* tx 5v detect */
-	tx_5v = io_read(sd, 0x70) & info->cable_det_mask;
+	tx_5v = irq_reg_0x70 & info->cable_det_mask;
 	if (tx_5v) {
 		v4l2_dbg(1, debug, sd, "%s: tx_5v: 0x%x\n", __func__, tx_5v);
-		io_write(sd, 0x71, tx_5v);
 		adv76xx_s_detect_tx_5v_ctrl(sd);
 		if (handled)
 			*handled = true;
diff --git a/drivers/media/media-device.c b/drivers/media/media-device.c
index 5ebb3cd31345..711c3674a5d9 100644
--- a/drivers/media/media-device.c
+++ b/drivers/media/media-device.c
@@ -20,6 +20,9 @@
  * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA  02111-1307  USA
  */
 
+/* We need to access legacy defines from linux/media.h */
+#define __NEED_MEDIA_LEGACY_API
+
 #include <linux/compat.h>
 #include <linux/export.h>
 #include <linux/idr.h>
@@ -119,6 +122,26 @@ static long media_device_enum_entities(struct media_device *mdev,
 	u_ent.group_id = 0;		/* Unused */
 	u_ent.pads = ent->num_pads;
 	u_ent.links = ent->num_links - ent->num_backlinks;
+
+	/*
+	 * Workaround for a bug at media-ctl <= v1.10 that makes it to
+	 * do the wrong thing if the entity function doesn't belong to
+	 * either MEDIA_ENT_F_OLD_BASE or MEDIA_ENT_F_OLD_SUBDEV_BASE
+	 * Ranges.
+	 *
+	 * Non-subdevices are expected to be at the MEDIA_ENT_F_OLD_BASE,
+	 * or, otherwise, will be silently ignored by media-ctl when
+	 * printing the graphviz diagram. So, map them into the devnode
+	 * old range.
+	 */
+	if (ent->function < MEDIA_ENT_F_OLD_BASE ||
+	    ent->function > MEDIA_ENT_T_DEVNODE_UNKNOWN) {
+		if (is_media_entity_v4l2_subdev(ent))
+			u_ent.type = MEDIA_ENT_F_V4L2_SUBDEV_UNKNOWN;
+		else if (ent->function != MEDIA_ENT_F_IO_V4L)
+			u_ent.type = MEDIA_ENT_T_DEVNODE_UNKNOWN;
+	}
+
 	memcpy(&u_ent.raw, &ent->info, sizeof(ent->info));
 	if (copy_to_user(uent, &u_ent, sizeof(u_ent)))
 		return -EFAULT;
diff --git a/drivers/mfd/db8500-prcmu.c b/drivers/mfd/db8500-prcmu.c
index e6e4bacb09ee..12099b09a9a7 100644
--- a/drivers/mfd/db8500-prcmu.c
+++ b/drivers/mfd/db8500-prcmu.c
@@ -2048,6 +2048,7 @@ int db8500_prcmu_config_hotmon(u8 low, u8 high)
 
 	return 0;
 }
+EXPORT_SYMBOL_GPL(db8500_prcmu_config_hotmon);
 
 static int config_hot_period(u16 val)
 {
@@ -2074,11 +2075,13 @@ int db8500_prcmu_start_temp_sense(u16 cycles32k)
 
 	return config_hot_period(cycles32k);
 }
+EXPORT_SYMBOL_GPL(db8500_prcmu_start_temp_sense);
 
 int db8500_prcmu_stop_temp_sense(void)
 {
 	return config_hot_period(0xFFFF);
 }
+EXPORT_SYMBOL_GPL(db8500_prcmu_stop_temp_sense);
 
 static int prcmu_a9wdog(u8 cmd, u8 d0, u8 d1, u8 d2, u8 d3)
 {
diff --git a/drivers/misc/cxl/pci.c b/drivers/misc/cxl/pci.c
index 4c1903f781fc..0c6c17a1c59e 100644
--- a/drivers/misc/cxl/pci.c
+++ b/drivers/misc/cxl/pci.c
@@ -415,7 +415,7 @@ static int cxl_setup_psl_timebase(struct cxl *adapter, struct pci_dev *dev)
 		delta = mftb() - psl_tb;
 		if (delta < 0)
 			delta = -delta;
-	} while (cputime_to_usecs(delta) > 16);
+	} while (tb_to_ns(delta) > 16000);
 
 	return 0;
 }
diff --git a/drivers/misc/mei/main.c b/drivers/misc/mei/main.c
index 677d0362f334..80f9afcb1382 100644
--- a/drivers/misc/mei/main.c
+++ b/drivers/misc/mei/main.c
@@ -458,7 +458,11 @@ static int mei_ioctl_client_notify_request(struct file *file, u32 request)
 {
 	struct mei_cl *cl = file->private_data;
 
-	return mei_cl_notify_request(cl, file, request);
+	if (request != MEI_HBM_NOTIFICATION_START &&
+	    request != MEI_HBM_NOTIFICATION_STOP)
+		return -EINVAL;
+
+	return mei_cl_notify_request(cl, file, (u8)request);
 }
 
 /**
diff --git a/drivers/mmc/card/block.c b/drivers/mmc/card/block.c
index 5914263090fc..fe207e542032 100644
--- a/drivers/mmc/card/block.c
+++ b/drivers/mmc/card/block.c
@@ -47,13 +47,10 @@
 #include "queue.h"
 
 MODULE_ALIAS("mmc:block");
-
-#ifdef KERNEL
 #ifdef MODULE_PARAM_PREFIX
 #undef MODULE_PARAM_PREFIX
 #endif
 #define MODULE_PARAM_PREFIX "mmcblk."
-#endif
 
 #define INAND_CMD38_ARG_EXT_CSD  113
 #define INAND_CMD38_ARG_ERASE    0x00
@@ -655,8 +652,10 @@ static int mmc_blk_ioctl_multi_cmd(struct block_device *bdev,
 	}
 
 	md = mmc_blk_get(bdev->bd_disk);
-	if (!md)
+	if (!md) {
+		err = -EINVAL;
 		goto cmd_err;
+	}
 
 	card = md->queue.card;
 	if (IS_ERR(card)) {
diff --git a/drivers/mmc/host/mmc_spi.c b/drivers/mmc/host/mmc_spi.c
index 1c1b45ef3faf..3446097a43c0 100644
--- a/drivers/mmc/host/mmc_spi.c
+++ b/drivers/mmc/host/mmc_spi.c
@@ -925,6 +925,10 @@ mmc_spi_data_do(struct mmc_spi_host *host, struct mmc_command *cmd,
 
 			dma_addr = dma_map_page(dma_dev, sg_page(sg), 0,
 						PAGE_SIZE, dir);
+			if (dma_mapping_error(dma_dev, dma_addr)) {
+				data->error = -EFAULT;
+				break;
+			}
 			if (direction == DMA_TO_DEVICE)
 				t->tx_dma = dma_addr + sg->offset;
 			else
@@ -1393,10 +1397,12 @@ static int mmc_spi_probe(struct spi_device *spi)
 		host->dma_dev = dev;
 		host->ones_dma = dma_map_single(dev, ones,
 				MMC_SPI_BLOCKSIZE, DMA_TO_DEVICE);
+		if (dma_mapping_error(dev, host->ones_dma))
+			goto fail_ones_dma;
 		host->data_dma = dma_map_single(dev, host->data,
 				sizeof(*host->data), DMA_BIDIRECTIONAL);
-
-		/* REVISIT in theory those map operations can fail... */
+		if (dma_mapping_error(dev, host->data_dma))
+			goto fail_data_dma;
 
 		dma_sync_single_for_cpu(host->dma_dev,
 				host->data_dma, sizeof(*host->data),
@@ -1462,6 +1468,11 @@ fail_glue_init:
 	if (host->dma_dev)
 		dma_unmap_single(host->dma_dev, host->data_dma,
 				sizeof(*host->data), DMA_BIDIRECTIONAL);
+fail_data_dma:
+	if (host->dma_dev)
+		dma_unmap_single(host->dma_dev, host->ones_dma,
+				MMC_SPI_BLOCKSIZE, DMA_TO_DEVICE);
+fail_ones_dma:
 	kfree(host->data);
 
 fail_nobuf1:
diff --git a/drivers/mmc/host/omap_hsmmc.c b/drivers/mmc/host/omap_hsmmc.c
index b6639ea0bf18..f6e4d9718035 100644
--- a/drivers/mmc/host/omap_hsmmc.c
+++ b/drivers/mmc/host/omap_hsmmc.c
@@ -2232,6 +2232,7 @@ err_irq:
 		dma_release_channel(host->tx_chan);
 	if (host->rx_chan)
 		dma_release_channel(host->rx_chan);
+	pm_runtime_dont_use_autosuspend(host->dev);
 	pm_runtime_put_sync(host->dev);
 	pm_runtime_disable(host->dev);
 	if (host->dbclk)
@@ -2253,6 +2254,7 @@ static int omap_hsmmc_remove(struct platform_device *pdev)
 	dma_release_channel(host->tx_chan);
 	dma_release_channel(host->rx_chan);
 
+	pm_runtime_dont_use_autosuspend(host->dev);
 	pm_runtime_put_sync(host->dev);
 	pm_runtime_disable(host->dev);
 	device_init_wakeup(&pdev->dev, false);
diff --git a/drivers/mmc/host/pxamci.c b/drivers/mmc/host/pxamci.c
index ce08896b9d69..da824772bbb4 100644
--- a/drivers/mmc/host/pxamci.c
+++ b/drivers/mmc/host/pxamci.c
@@ -86,7 +86,7 @@ struct pxamci_host {
 static inline void pxamci_init_ocr(struct pxamci_host *host)
 {
 #ifdef CONFIG_REGULATOR
-	host->vcc = regulator_get_optional(mmc_dev(host->mmc), "vmmc");
+	host->vcc = devm_regulator_get_optional(mmc_dev(host->mmc), "vmmc");
 
 	if (IS_ERR(host->vcc))
 		host->vcc = NULL;
@@ -654,12 +654,8 @@ static int pxamci_probe(struct platform_device *pdev)
 
 	r = platform_get_resource(pdev, IORESOURCE_MEM, 0);
 	irq = platform_get_irq(pdev, 0);
-	if (!r || irq < 0)
-		return -ENXIO;
-
-	r = request_mem_region(r->start, SZ_4K, DRIVER_NAME);
-	if (!r)
-		return -EBUSY;
+	if (irq < 0)
+		return irq;
 
 	mmc = mmc_alloc_host(sizeof(struct pxamci_host), &pdev->dev);
 	if (!mmc) {
@@ -695,7 +691,7 @@ static int pxamci_probe(struct platform_device *pdev)
 	host->pdata = pdev->dev.platform_data;
 	host->clkrt = CLKRT_OFF;
 
-	host->clk = clk_get(&pdev->dev, NULL);
+	host->clk = devm_clk_get(&pdev->dev, NULL);
 	if (IS_ERR(host->clk)) {
 		ret = PTR_ERR(host->clk);
 		host->clk = NULL;
@@ -727,9 +723,9 @@ static int pxamci_probe(struct platform_device *pdev)
 	host->irq = irq;
 	host->imask = MMC_I_MASK_ALL;
 
-	host->base = ioremap(r->start, SZ_4K);
-	if (!host->base) {
-		ret = -ENOMEM;
+	host->base = devm_ioremap_resource(&pdev->dev, r);
+	if (IS_ERR(host->base)) {
+		ret = PTR_ERR(host->base);
 		goto out;
 	}
 
@@ -742,7 +738,8 @@ static int pxamci_probe(struct platform_device *pdev)
 	writel(64, host->base + MMC_RESTO);
 	writel(host->imask, host->base + MMC_I_MASK);
 
-	ret = request_irq(host->irq, pxamci_irq, 0, DRIVER_NAME, host);
+	ret = devm_request_irq(&pdev->dev, host->irq, pxamci_irq, 0,
+			       DRIVER_NAME, host);
 	if (ret)
 		goto out;
 
@@ -804,7 +801,7 @@ static int pxamci_probe(struct platform_device *pdev)
 		dev_err(&pdev->dev, "Failed requesting gpio_ro %d\n", gpio_ro);
 		goto out;
 	} else {
-		mmc->caps |= host->pdata->gpio_card_ro_invert ?
+		mmc->caps2 |= host->pdata->gpio_card_ro_invert ?
 			0 : MMC_CAP2_RO_ACTIVE_HIGH;
 	}
 
@@ -833,14 +830,9 @@ out:
 			dma_release_channel(host->dma_chan_rx);
 		if (host->dma_chan_tx)
 			dma_release_channel(host->dma_chan_tx);
-		if (host->base)
-			iounmap(host->base);
-		if (host->clk)
-			clk_put(host->clk);
 	}
 	if (mmc)
 		mmc_free_host(mmc);
-	release_resource(r);
 	return ret;
 }
 
@@ -859,9 +851,6 @@ static int pxamci_remove(struct platform_device *pdev)
 			gpio_ro = host->pdata->gpio_card_ro;
 			gpio_power = host->pdata->gpio_power;
 		}
-		if (host->vcc)
-			regulator_put(host->vcc);
-
 		if (host->pdata && host->pdata->exit)
 			host->pdata->exit(&pdev->dev, mmc);
 
@@ -870,16 +859,10 @@ static int pxamci_remove(struct platform_device *pdev)
 		       END_CMD_RES|PRG_DONE|DATA_TRAN_DONE,
 		       host->base + MMC_I_MASK);
 
-		free_irq(host->irq, host);
 		dmaengine_terminate_all(host->dma_chan_rx);
 		dmaengine_terminate_all(host->dma_chan_tx);
 		dma_release_channel(host->dma_chan_rx);
 		dma_release_channel(host->dma_chan_tx);
-		iounmap(host->base);
-
-		clk_put(host->clk);
-
-		release_resource(host->res);
 
 		mmc_free_host(mmc);
 	}
diff --git a/drivers/mmc/host/sdhci-acpi.c b/drivers/mmc/host/sdhci-acpi.c
index f6047fc94062..a5cda926d38e 100644
--- a/drivers/mmc/host/sdhci-acpi.c
+++ b/drivers/mmc/host/sdhci-acpi.c
@@ -146,6 +146,33 @@ static const struct sdhci_acpi_chip sdhci_acpi_chip_int = {
 	.ops = &sdhci_acpi_ops_int,
 };
 
+static int bxt_get_cd(struct mmc_host *mmc)
+{
+	int gpio_cd = mmc_gpio_get_cd(mmc);
+	struct sdhci_host *host = mmc_priv(mmc);
+	unsigned long flags;
+	int ret = 0;
+
+	if (!gpio_cd)
+		return 0;
+
+	pm_runtime_get_sync(mmc->parent);
+
+	spin_lock_irqsave(&host->lock, flags);
+
+	if (host->flags & SDHCI_DEVICE_DEAD)
+		goto out;
+
+	ret = !!(sdhci_readl(host, SDHCI_PRESENT_STATE) & SDHCI_CARD_PRESENT);
+out:
+	spin_unlock_irqrestore(&host->lock, flags);
+
+	pm_runtime_mark_last_busy(mmc->parent);
+	pm_runtime_put_autosuspend(mmc->parent);
+
+	return ret;
+}
+
 static int sdhci_acpi_emmc_probe_slot(struct platform_device *pdev,
 				      const char *hid, const char *uid)
 {
@@ -196,6 +223,9 @@ static int sdhci_acpi_sd_probe_slot(struct platform_device *pdev,
 
 	/* Platform specific code during sd probe slot goes here */
 
+	if (hid && !strcmp(hid, "80865ACA"))
+		host->mmc_host_ops.get_cd = bxt_get_cd;
+
 	return 0;
 }
 
diff --git a/drivers/mmc/host/sdhci-of-at91.c b/drivers/mmc/host/sdhci-of-at91.c
index 7e7d8f0c9438..9cb86fb25976 100644
--- a/drivers/mmc/host/sdhci-of-at91.c
+++ b/drivers/mmc/host/sdhci-of-at91.c
@@ -217,6 +217,7 @@ static int sdhci_at91_probe(struct platform_device *pdev)
 pm_runtime_disable:
 	pm_runtime_disable(&pdev->dev);
 	pm_runtime_set_suspended(&pdev->dev);
+	pm_runtime_put_noidle(&pdev->dev);
 clocks_disable_unprepare:
 	clk_disable_unprepare(priv->gck);
 	clk_disable_unprepare(priv->mainck);
diff --git a/drivers/mmc/host/sdhci-pci-core.c b/drivers/mmc/host/sdhci-pci-core.c
index cc851b065d0a..df3b8eced8c4 100644
--- a/drivers/mmc/host/sdhci-pci-core.c
+++ b/drivers/mmc/host/sdhci-pci-core.c
@@ -330,6 +330,33 @@ static void spt_read_drive_strength(struct sdhci_host *host)
 	sdhci_pci_spt_drive_strength = 0x10 | ((val >> 12) & 0xf);
 }
 
+static int bxt_get_cd(struct mmc_host *mmc)
+{
+	int gpio_cd = mmc_gpio_get_cd(mmc);
+	struct sdhci_host *host = mmc_priv(mmc);
+	unsigned long flags;
+	int ret = 0;
+
+	if (!gpio_cd)
+		return 0;
+
+	pm_runtime_get_sync(mmc->parent);
+
+	spin_lock_irqsave(&host->lock, flags);
+
+	if (host->flags & SDHCI_DEVICE_DEAD)
+		goto out;
+
+	ret = !!(sdhci_readl(host, SDHCI_PRESENT_STATE) & SDHCI_CARD_PRESENT);
+out:
+	spin_unlock_irqrestore(&host->lock, flags);
+
+	pm_runtime_mark_last_busy(mmc->parent);
+	pm_runtime_put_autosuspend(mmc->parent);
+
+	return ret;
+}
+
 static int byt_emmc_probe_slot(struct sdhci_pci_slot *slot)
 {
 	slot->host->mmc->caps |= MMC_CAP_8_BIT_DATA | MMC_CAP_NONREMOVABLE |
@@ -362,6 +389,10 @@ static int byt_sd_probe_slot(struct sdhci_pci_slot *slot)
 	slot->cd_con_id = NULL;
 	slot->cd_idx = 0;
 	slot->cd_override_level = true;
+	if (slot->chip->pdev->device == PCI_DEVICE_ID_INTEL_BXT_SD ||
+	    slot->chip->pdev->device == PCI_DEVICE_ID_INTEL_APL_SD)
+		slot->host->mmc_host_ops.get_cd = bxt_get_cd;
+
 	return 0;
 }
 
diff --git a/drivers/mmc/host/sdhci.c b/drivers/mmc/host/sdhci.c
index d622435d1bcc..add9fdfd1d8f 100644
--- a/drivers/mmc/host/sdhci.c
+++ b/drivers/mmc/host/sdhci.c
@@ -1360,7 +1360,7 @@ static void sdhci_request(struct mmc_host *mmc, struct mmc_request *mrq)
 	sdhci_runtime_pm_get(host);
 
 	/* Firstly check card presence */
-	present = sdhci_do_get_cd(host);
+	present = mmc->ops->get_cd(mmc);
 
 	spin_lock_irqsave(&host->lock, flags);
 
@@ -2849,6 +2849,8 @@ struct sdhci_host *sdhci_alloc_host(struct device *dev,
 
 	host = mmc_priv(mmc);
 	host->mmc = mmc;
+	host->mmc_host_ops = sdhci_ops;
+	mmc->ops = &host->mmc_host_ops;
 
 	return host;
 }
@@ -3037,7 +3039,6 @@ int sdhci_add_host(struct sdhci_host *host)
 	/*
 	 * Set host parameters.
 	 */
-	mmc->ops = &sdhci_ops;
 	max_clk = host->max_clk;
 
 	if (host->ops->get_min_clock)
diff --git a/drivers/mmc/host/sdhci.h b/drivers/mmc/host/sdhci.h
index 7654ae5d2b4e..0115e9907bf8 100644
--- a/drivers/mmc/host/sdhci.h
+++ b/drivers/mmc/host/sdhci.h
@@ -430,6 +430,7 @@ struct sdhci_host {
 
 	/* Internal data */
 	struct mmc_host *mmc;	/* MMC structure */
+	struct mmc_host_ops mmc_host_ops;	/* MMC host ops */
 	u64 dma_mask;		/* custom DMA mask */
 
 #if defined(CONFIG_LEDS_CLASS) || defined(CONFIG_LEDS_CLASS_MODULE)
diff --git a/drivers/mmc/host/sh_mmcif.c b/drivers/mmc/host/sh_mmcif.c
index 1ca8a1359cbc..6234eab38ff3 100644
--- a/drivers/mmc/host/sh_mmcif.c
+++ b/drivers/mmc/host/sh_mmcif.c
@@ -445,7 +445,7 @@ static void sh_mmcif_request_dma(struct sh_mmcif_host *host)
 							pdata->slave_id_rx);
 	} else {
 		host->chan_tx = dma_request_slave_channel(dev, "tx");
-		host->chan_tx = dma_request_slave_channel(dev, "rx");
+		host->chan_rx = dma_request_slave_channel(dev, "rx");
 	}
 	dev_dbg(dev, "%s: got channel TX %p RX %p\n", __func__, host->chan_tx,
 		host->chan_rx);
diff --git a/drivers/mtd/ubi/upd.c b/drivers/mtd/ubi/upd.c
index 2a1b6e037e1a..0134ba32a057 100644
--- a/drivers/mtd/ubi/upd.c
+++ b/drivers/mtd/ubi/upd.c
@@ -193,7 +193,7 @@ int ubi_start_leb_change(struct ubi_device *ubi, struct ubi_volume *vol,
 	vol->changing_leb = 1;
 	vol->ch_lnum = req->lnum;
 
-	vol->upd_buf = vmalloc(req->bytes);
+	vol->upd_buf = vmalloc(ALIGN((int)req->bytes, ubi->min_io_size));
 	if (!vol->upd_buf)
 		return -ENOMEM;
 
diff --git a/drivers/net/bonding/bond_main.c b/drivers/net/bonding/bond_main.c
index 56b560558884..b7f1a9919033 100644
--- a/drivers/net/bonding/bond_main.c
+++ b/drivers/net/bonding/bond_main.c
@@ -214,6 +214,8 @@ static void bond_uninit(struct net_device *bond_dev);
 static struct rtnl_link_stats64 *bond_get_stats(struct net_device *bond_dev,
 						struct rtnl_link_stats64 *stats);
 static void bond_slave_arr_handler(struct work_struct *work);
+static bool bond_time_in_interval(struct bonding *bond, unsigned long last_act,
+				  int mod);
 
 /*---------------------------- General routines -----------------------------*/
 
@@ -2127,6 +2129,7 @@ static void bond_miimon_commit(struct bonding *bond)
 			continue;
 
 		case BOND_LINK_UP:
+			bond_update_speed_duplex(slave);
 			bond_set_slave_link_state(slave, BOND_LINK_UP,
 						  BOND_SLAVE_NOTIFY_NOW);
 			slave->last_link_up = jiffies;
@@ -2459,7 +2462,7 @@ int bond_arp_rcv(const struct sk_buff *skb, struct bonding *bond,
 		 struct slave *slave)
 {
 	struct arphdr *arp = (struct arphdr *)skb->data;
-	struct slave *curr_active_slave;
+	struct slave *curr_active_slave, *curr_arp_slave;
 	unsigned char *arp_ptr;
 	__be32 sip, tip;
 	int alen, is_arp = skb->protocol == __cpu_to_be16(ETH_P_ARP);
@@ -2506,26 +2509,41 @@ int bond_arp_rcv(const struct sk_buff *skb, struct bonding *bond,
 		     &sip, &tip);
 
 	curr_active_slave = rcu_dereference(bond->curr_active_slave);
+	curr_arp_slave = rcu_dereference(bond->current_arp_slave);
 
-	/* Backup slaves won't see the ARP reply, but do come through
-	 * here for each ARP probe (so we swap the sip/tip to validate
-	 * the probe).  In a "redundant switch, common router" type of
-	 * configuration, the ARP probe will (hopefully) travel from
-	 * the active, through one switch, the router, then the other
-	 * switch before reaching the backup.
+	/* We 'trust' the received ARP enough to validate it if:
+	 *
+	 * (a) the slave receiving the ARP is active (which includes the
+	 * current ARP slave, if any), or
+	 *
+	 * (b) the receiving slave isn't active, but there is a currently
+	 * active slave and it received valid arp reply(s) after it became
+	 * the currently active slave, or
+	 *
+	 * (c) there is an ARP slave that sent an ARP during the prior ARP
+	 * interval, and we receive an ARP reply on any slave.  We accept
+	 * these because switch FDB update delays may deliver the ARP
+	 * reply to a slave other than the sender of the ARP request.
 	 *
-	 * We 'trust' the arp requests if there is an active slave and
-	 * it received valid arp reply(s) after it became active. This
-	 * is done to avoid endless looping when we can't reach the
+	 * Note: for (b), backup slaves are receiving the broadcast ARP
+	 * request, not a reply.  This request passes from the sending
+	 * slave through the L2 switch(es) to the receiving slave.  Since
+	 * this is checking the request, sip/tip are swapped for
+	 * validation.
+	 *
+	 * This is done to avoid endless looping when we can't reach the
 	 * arp_ip_target and fool ourselves with our own arp requests.
 	 */
-
 	if (bond_is_active_slave(slave))
 		bond_validate_arp(bond, slave, sip, tip);
 	else if (curr_active_slave &&
 		 time_after(slave_last_rx(bond, curr_active_slave),
 			    curr_active_slave->last_link_up))
 		bond_validate_arp(bond, slave, tip, sip);
+	else if (curr_arp_slave && (arp->ar_op == htons(ARPOP_REPLY)) &&
+		 bond_time_in_interval(bond,
+				       dev_trans_start(curr_arp_slave->dev), 1))
+		bond_validate_arp(bond, slave, sip, tip);
 
 out_unlock:
 	if (arp != (struct arphdr *)skb->data)
diff --git a/drivers/net/can/spi/mcp251x.c b/drivers/net/can/spi/mcp251x.c
index 575790e8a75a..74a7dfecee27 100644
--- a/drivers/net/can/spi/mcp251x.c
+++ b/drivers/net/can/spi/mcp251x.c
@@ -843,7 +843,7 @@ static irqreturn_t mcp251x_can_ist(int irq, void *dev_id)
 		if (clear_intf)
 			mcp251x_write_bits(spi, CANINTF, clear_intf, 0x00);
 
-		if (eflag)
+		if (eflag & (EFLG_RX0OVR | EFLG_RX1OVR))
 			mcp251x_write_bits(spi, EFLG, eflag, 0x00);
 
 		/* Update can state */
diff --git a/drivers/net/can/usb/ems_usb.c b/drivers/net/can/usb/ems_usb.c
index fc5b75675cd8..eb7192fab593 100644
--- a/drivers/net/can/usb/ems_usb.c
+++ b/drivers/net/can/usb/ems_usb.c
@@ -117,6 +117,9 @@ MODULE_LICENSE("GPL v2");
  */
 #define EMS_USB_ARM7_CLOCK 8000000
 
+#define CPC_TX_QUEUE_TRIGGER_LOW	25
+#define CPC_TX_QUEUE_TRIGGER_HIGH	35
+
 /*
  * CAN-Message representation in a CPC_MSG. Message object type is
  * CPC_MSG_TYPE_CAN_FRAME or CPC_MSG_TYPE_RTR_FRAME or
@@ -278,6 +281,11 @@ static void ems_usb_read_interrupt_callback(struct urb *urb)
 	switch (urb->status) {
 	case 0:
 		dev->free_slots = dev->intr_in_buffer[1];
+		if(dev->free_slots > CPC_TX_QUEUE_TRIGGER_HIGH){
+			if (netif_queue_stopped(netdev)){
+				netif_wake_queue(netdev);
+			}
+		}
 		break;
 
 	case -ECONNRESET: /* unlink */
@@ -526,8 +534,6 @@ static void ems_usb_write_bulk_callback(struct urb *urb)
 	/* Release context */
 	context->echo_index = MAX_TX_URBS;
 
-	if (netif_queue_stopped(netdev))
-		netif_wake_queue(netdev);
 }
 
 /*
@@ -587,7 +593,7 @@ static int ems_usb_start(struct ems_usb *dev)
 	int err, i;
 
 	dev->intr_in_buffer[0] = 0;
-	dev->free_slots = 15; /* initial size */
+	dev->free_slots = 50; /* initial size */
 
 	for (i = 0; i < MAX_RX_URBS; i++) {
 		struct urb *urb = NULL;
@@ -835,7 +841,7 @@ static netdev_tx_t ems_usb_start_xmit(struct sk_buff *skb, struct net_device *ne
 
 		/* Slow down tx path */
 		if (atomic_read(&dev->active_tx_urbs) >= MAX_TX_URBS ||
-		    dev->free_slots < 5) {
+		    dev->free_slots < CPC_TX_QUEUE_TRIGGER_LOW) {
 			netif_stop_queue(netdev);
 		}
 	}
diff --git a/drivers/net/can/usb/gs_usb.c b/drivers/net/can/usb/gs_usb.c
index 5eee62badf45..cbc99d5649af 100644
--- a/drivers/net/can/usb/gs_usb.c
+++ b/drivers/net/can/usb/gs_usb.c
@@ -826,9 +826,8 @@ static struct gs_can *gs_make_candev(unsigned int channel, struct usb_interface
 static void gs_destroy_candev(struct gs_can *dev)
 {
 	unregister_candev(dev->netdev);
-	free_candev(dev->netdev);
 	usb_kill_anchored_urbs(&dev->tx_submitted);
-	kfree(dev);
+	free_candev(dev->netdev);
 }
 
 static int gs_usb_probe(struct usb_interface *intf, const struct usb_device_id *id)
@@ -913,12 +912,15 @@ static int gs_usb_probe(struct usb_interface *intf, const struct usb_device_id *
 	for (i = 0; i < icount; i++) {
 		dev->canch[i] = gs_make_candev(i, intf);
 		if (IS_ERR_OR_NULL(dev->canch[i])) {
+			/* save error code to return later */
+			rc = PTR_ERR(dev->canch[i]);
+
 			/* on failure destroy previously created candevs */
 			icount = i;
-			for (i = 0; i < icount; i++) {
+			for (i = 0; i < icount; i++)
 				gs_destroy_candev(dev->canch[i]);
-				dev->canch[i] = NULL;
-			}
+
+			usb_kill_anchored_urbs(&dev->rx_submitted);
 			kfree(dev);
 			return rc;
 		}
@@ -939,16 +941,12 @@ static void gs_usb_disconnect(struct usb_interface *intf)
 		return;
 	}
 
-	for (i = 0; i < GS_MAX_INTF; i++) {
-		struct gs_can *can = dev->canch[i];
-
-		if (!can)
-			continue;
-
-		gs_destroy_candev(can);
-	}
+	for (i = 0; i < GS_MAX_INTF; i++)
+		if (dev->canch[i])
+			gs_destroy_candev(dev->canch[i]);
 
 	usb_kill_anchored_urbs(&dev->rx_submitted);
+	kfree(dev);
 }
 
 static const struct usb_device_id gs_usb_table[] = {
diff --git a/drivers/net/dsa/mv88e6352.c b/drivers/net/dsa/mv88e6352.c
index cc6c54553418..a47f52f44b0d 100644
--- a/drivers/net/dsa/mv88e6352.c
+++ b/drivers/net/dsa/mv88e6352.c
@@ -25,6 +25,7 @@
 static const struct mv88e6xxx_switch_id mv88e6352_table[] = {
 	{ PORT_SWITCH_ID_6172, "Marvell 88E6172" },
 	{ PORT_SWITCH_ID_6176, "Marvell 88E6176" },
+	{ PORT_SWITCH_ID_6240, "Marvell 88E6240" },
 	{ PORT_SWITCH_ID_6320, "Marvell 88E6320" },
 	{ PORT_SWITCH_ID_6320_A1, "Marvell 88E6320 (A1)" },
 	{ PORT_SWITCH_ID_6320_A2, "Marvell 88e6320 (A2)" },
diff --git a/drivers/net/dsa/mv88e6xxx.c b/drivers/net/dsa/mv88e6xxx.c
index cf34681af4f6..512c8c0be1b4 100644
--- a/drivers/net/dsa/mv88e6xxx.c
+++ b/drivers/net/dsa/mv88e6xxx.c
@@ -1555,7 +1555,7 @@ static int _mv88e6xxx_port_vlan_del(struct dsa_switch *ds, int port, u16 vid)
 
 	if (vlan.vid != vid || !vlan.valid ||
 	    vlan.data[port] == GLOBAL_VTU_DATA_MEMBER_TAG_NON_MEMBER)
-		return -ENOENT;
+		return -EOPNOTSUPP;
 
 	vlan.data[port] = GLOBAL_VTU_DATA_MEMBER_TAG_NON_MEMBER;
 
@@ -1582,6 +1582,7 @@ int mv88e6xxx_port_vlan_del(struct dsa_switch *ds, int port,
 			    const struct switchdev_obj_port_vlan *vlan)
 {
 	struct mv88e6xxx_priv_state *ps = ds_to_priv(ds);
+	const u16 defpvid = 4000 + ds->index * DSA_MAX_PORTS + port;
 	u16 pvid, vid;
 	int err = 0;
 
@@ -1597,7 +1598,8 @@ int mv88e6xxx_port_vlan_del(struct dsa_switch *ds, int port,
 			goto unlock;
 
 		if (vid == pvid) {
-			err = _mv88e6xxx_port_pvid_set(ds, port, 0);
+			/* restore reserved VLAN ID */
+			err = _mv88e6xxx_port_pvid_set(ds, port, defpvid);
 			if (err)
 				goto unlock;
 		}
@@ -1889,26 +1891,20 @@ unlock:
 
 int mv88e6xxx_port_bridge_join(struct dsa_switch *ds, int port, u32 members)
 {
-	struct mv88e6xxx_priv_state *ps = ds_to_priv(ds);
-	const u16 pvid = 4000 + ds->index * DSA_MAX_PORTS + port;
-	int err;
-
-	/* The port joined a bridge, so leave its reserved VLAN */
-	mutex_lock(&ps->smi_mutex);
-	err = _mv88e6xxx_port_vlan_del(ds, port, pvid);
-	if (!err)
-		err = _mv88e6xxx_port_pvid_set(ds, port, 0);
-	mutex_unlock(&ps->smi_mutex);
-	return err;
+	return 0;
 }
 
 int mv88e6xxx_port_bridge_leave(struct dsa_switch *ds, int port, u32 members)
 {
+	return 0;
+}
+
+static int mv88e6xxx_setup_port_default_vlan(struct dsa_switch *ds, int port)
+{
 	struct mv88e6xxx_priv_state *ps = ds_to_priv(ds);
 	const u16 pvid = 4000 + ds->index * DSA_MAX_PORTS + port;
 	int err;
 
-	/* The port left the bridge, so join its reserved VLAN */
 	mutex_lock(&ps->smi_mutex);
 	err = _mv88e6xxx_port_vlan_add(ds, port, pvid, true);
 	if (!err)
@@ -2192,8 +2188,7 @@ int mv88e6xxx_setup_ports(struct dsa_switch *ds)
 		if (dsa_is_cpu_port(ds, i) || dsa_is_dsa_port(ds, i))
 			continue;
 
-		/* setup the unbridged state */
-		ret = mv88e6xxx_port_bridge_leave(ds, i, 0);
+		ret = mv88e6xxx_setup_port_default_vlan(ds, i);
 		if (ret < 0)
 			return ret;
 	}
diff --git a/drivers/net/ethernet/3com/3c59x.c b/drivers/net/ethernet/3com/3c59x.c
index 79e1a0282163..17b2126075e0 100644
--- a/drivers/net/ethernet/3com/3c59x.c
+++ b/drivers/net/ethernet/3com/3c59x.c
@@ -2461,7 +2461,7 @@ boomerang_interrupt(int irq, void *dev_id)
 					int i;
 					pci_unmap_single(VORTEX_PCI(vp),
 							le32_to_cpu(vp->tx_ring[entry].frag[0].addr),
-							le32_to_cpu(vp->tx_ring[entry].frag[0].length),
+							le32_to_cpu(vp->tx_ring[entry].frag[0].length)&0xFFF,
 							PCI_DMA_TODEVICE);
 
 					for (i=1; i<=skb_shinfo(skb)->nr_frags; i++)
diff --git a/drivers/net/ethernet/8390/pcnet_cs.c b/drivers/net/ethernet/8390/pcnet_cs.c
index 2777289a26c0..2f79d29f17f2 100644
--- a/drivers/net/ethernet/8390/pcnet_cs.c
+++ b/drivers/net/ethernet/8390/pcnet_cs.c
@@ -1501,6 +1501,7 @@ static const struct pcmcia_device_id pcnet_ids[] = {
 	PCMCIA_DEVICE_MANF_CARD(0x026f, 0x030a),
 	PCMCIA_DEVICE_MANF_CARD(0x0274, 0x1103),
 	PCMCIA_DEVICE_MANF_CARD(0x0274, 0x1121),
+	PCMCIA_DEVICE_MANF_CARD(0xc001, 0x0009),
 	PCMCIA_DEVICE_PROD_ID12("2408LAN", "Ethernet", 0x352fff7f, 0x00b2e941),
 	PCMCIA_DEVICE_PROD_ID1234("Socket", "CF 10/100 Ethernet Card", "Revision B", "05/11/06", 0xb38bcc2e, 0x4de88352, 0xeaca6c8d, 0x7e57c22e),
 	PCMCIA_DEVICE_PROD_ID123("Cardwell", "PCMCIA", "ETHERNET", 0x9533672e, 0x281f1c5d, 0x3ff7175b),
diff --git a/drivers/net/ethernet/agere/et131x.c b/drivers/net/ethernet/agere/et131x.c
index 3f3bcbea15bd..0907ab6ff309 100644
--- a/drivers/net/ethernet/agere/et131x.c
+++ b/drivers/net/ethernet/agere/et131x.c
@@ -2380,7 +2380,7 @@ static int et131x_tx_dma_memory_alloc(struct et131x_adapter *adapter)
 						    sizeof(u32),
 						    &tx_ring->tx_status_pa,
 						    GFP_KERNEL);
-	if (!tx_ring->tx_status_pa) {
+	if (!tx_ring->tx_status) {
 		dev_err(&adapter->pdev->dev,
 			"Cannot alloc memory for Tx status block\n");
 		return -ENOMEM;
diff --git a/drivers/net/ethernet/altera/altera_tse_main.c b/drivers/net/ethernet/altera/altera_tse_main.c
index 17472851674f..f749e4d389eb 100644
--- a/drivers/net/ethernet/altera/altera_tse_main.c
+++ b/drivers/net/ethernet/altera/altera_tse_main.c
@@ -193,7 +193,6 @@ static void altera_tse_mdio_destroy(struct net_device *dev)
 			    priv->mdio->id);
 
 	mdiobus_unregister(priv->mdio);
-	kfree(priv->mdio->irq);
 	mdiobus_free(priv->mdio);
 	priv->mdio = NULL;
 }
diff --git a/drivers/net/ethernet/amd/am79c961a.c b/drivers/net/ethernet/amd/am79c961a.c
index 87e727b921dc..fcdf5dda448f 100644
--- a/drivers/net/ethernet/amd/am79c961a.c
+++ b/drivers/net/ethernet/amd/am79c961a.c
@@ -50,8 +50,8 @@ static const char version[] =
 static void write_rreg(u_long base, u_int reg, u_int val)
 {
 	asm volatile(
-	"str%?h	%1, [%2]	@ NET_RAP\n\t"
-	"str%?h	%0, [%2, #-4]	@ NET_RDP"
+	"strh	%1, [%2]	@ NET_RAP\n\t"
+	"strh	%0, [%2, #-4]	@ NET_RDP"
 	:
 	: "r" (val), "r" (reg), "r" (ISAIO_BASE + 0x0464));
 }
@@ -60,8 +60,8 @@ static inline unsigned short read_rreg(u_long base_addr, u_int reg)
 {
 	unsigned short v;
 	asm volatile(
-	"str%?h	%1, [%2]	@ NET_RAP\n\t"
-	"ldr%?h	%0, [%2, #-4]	@ NET_RDP"
+	"strh	%1, [%2]	@ NET_RAP\n\t"
+	"ldrh	%0, [%2, #-4]	@ NET_RDP"
 	: "=r" (v)
 	: "r" (reg), "r" (ISAIO_BASE + 0x0464));
 	return v;
@@ -70,8 +70,8 @@ static inline unsigned short read_rreg(u_long base_addr, u_int reg)
 static inline void write_ireg(u_long base, u_int reg, u_int val)
 {
 	asm volatile(
-	"str%?h	%1, [%2]	@ NET_RAP\n\t"
-	"str%?h	%0, [%2, #8]	@ NET_IDP"
+	"strh	%1, [%2]	@ NET_RAP\n\t"
+	"strh	%0, [%2, #8]	@ NET_IDP"
 	:
 	: "r" (val), "r" (reg), "r" (ISAIO_BASE + 0x0464));
 }
@@ -80,8 +80,8 @@ static inline unsigned short read_ireg(u_long base_addr, u_int reg)
 {
 	u_short v;
 	asm volatile(
-	"str%?h	%1, [%2]	@ NAT_RAP\n\t"
-	"ldr%?h	%0, [%2, #8]	@ NET_IDP\n\t"
+	"strh	%1, [%2]	@ NAT_RAP\n\t"
+	"ldrh	%0, [%2, #8]	@ NET_IDP\n\t"
 	: "=r" (v)
 	: "r" (reg), "r" (ISAIO_BASE + 0x0464));
 	return v;
@@ -96,7 +96,7 @@ am_writebuffer(struct net_device *dev, u_int offset, unsigned char *buf, unsigne
 	offset = ISAMEM_BASE + (offset << 1);
 	length = (length + 1) & ~1;
 	if ((int)buf & 2) {
-		asm volatile("str%?h	%2, [%0], #4"
+		asm volatile("strh	%2, [%0], #4"
 		 : "=&r" (offset) : "0" (offset), "r" (buf[0] | (buf[1] << 8)));
 		buf += 2;
 		length -= 2;
@@ -104,20 +104,20 @@ am_writebuffer(struct net_device *dev, u_int offset, unsigned char *buf, unsigne
 	while (length > 8) {
 		register unsigned int tmp asm("r2"), tmp2 asm("r3");
 		asm volatile(
-			"ldm%?ia	%0!, {%1, %2}"
+			"ldmia	%0!, {%1, %2}"
 			: "+r" (buf), "=&r" (tmp), "=&r" (tmp2));
 		length -= 8;
 		asm volatile(
-			"str%?h	%1, [%0], #4\n\t"
-			"mov%?	%1, %1, lsr #16\n\t"
-			"str%?h	%1, [%0], #4\n\t"
-			"str%?h	%2, [%0], #4\n\t"
-			"mov%?	%2, %2, lsr #16\n\t"
-			"str%?h	%2, [%0], #4"
+			"strh	%1, [%0], #4\n\t"
+			"mov	%1, %1, lsr #16\n\t"
+			"strh	%1, [%0], #4\n\t"
+			"strh	%2, [%0], #4\n\t"
+			"mov	%2, %2, lsr #16\n\t"
+			"strh	%2, [%0], #4"
 		: "+r" (offset), "=&r" (tmp), "=&r" (tmp2));
 	}
 	while (length > 0) {
-		asm volatile("str%?h	%2, [%0], #4"
+		asm volatile("strh	%2, [%0], #4"
 		 : "=&r" (offset) : "0" (offset), "r" (buf[0] | (buf[1] << 8)));
 		buf += 2;
 		length -= 2;
@@ -132,23 +132,23 @@ am_readbuffer(struct net_device *dev, u_int offset, unsigned char *buf, unsigned
 	if ((int)buf & 2) {
 		unsigned int tmp;
 		asm volatile(
-			"ldr%?h	%2, [%0], #4\n\t"
-			"str%?b	%2, [%1], #1\n\t"
-			"mov%?	%2, %2, lsr #8\n\t"
-			"str%?b	%2, [%1], #1"
+			"ldrh	%2, [%0], #4\n\t"
+			"strb	%2, [%1], #1\n\t"
+			"mov	%2, %2, lsr #8\n\t"
+			"strb	%2, [%1], #1"
 		: "=&r" (offset), "=&r" (buf), "=r" (tmp): "0" (offset), "1" (buf));
 		length -= 2;
 	}
 	while (length > 8) {
 		register unsigned int tmp asm("r2"), tmp2 asm("r3"), tmp3;
 		asm volatile(
-			"ldr%?h	%2, [%0], #4\n\t"
-			"ldr%?h	%4, [%0], #4\n\t"
-			"ldr%?h	%3, [%0], #4\n\t"
-			"orr%?	%2, %2, %4, lsl #16\n\t"
-			"ldr%?h	%4, [%0], #4\n\t"
-			"orr%?	%3, %3, %4, lsl #16\n\t"
-			"stm%?ia	%1!, {%2, %3}"
+			"ldrh	%2, [%0], #4\n\t"
+			"ldrh	%4, [%0], #4\n\t"
+			"ldrh	%3, [%0], #4\n\t"
+			"orr	%2, %2, %4, lsl #16\n\t"
+			"ldrh	%4, [%0], #4\n\t"
+			"orr	%3, %3, %4, lsl #16\n\t"
+			"stmia	%1!, {%2, %3}"
 		: "=&r" (offset), "=&r" (buf), "=r" (tmp), "=r" (tmp2), "=r" (tmp3)
 		: "0" (offset), "1" (buf));
 		length -= 8;
@@ -156,10 +156,10 @@ am_readbuffer(struct net_device *dev, u_int offset, unsigned char *buf, unsigned
 	while (length > 0) {
 		unsigned int tmp;
 		asm volatile(
-			"ldr%?h	%2, [%0], #4\n\t"
-			"str%?b	%2, [%1], #1\n\t"
-			"mov%?	%2, %2, lsr #8\n\t"
-			"str%?b	%2, [%1], #1"
+			"ldrh	%2, [%0], #4\n\t"
+			"strb	%2, [%1], #1\n\t"
+			"mov	%2, %2, lsr #8\n\t"
+			"strb	%2, [%1], #1"
 		: "=&r" (offset), "=&r" (buf), "=r" (tmp) : "0" (offset), "1" (buf));
 		length -= 2;
 	}
diff --git a/drivers/net/ethernet/amd/lance.c b/drivers/net/ethernet/amd/lance.c
index 256f590f6bb1..3a7ebfdda57d 100644
--- a/drivers/net/ethernet/amd/lance.c
+++ b/drivers/net/ethernet/amd/lance.c
@@ -547,8 +547,8 @@ static int __init lance_probe1(struct net_device *dev, int ioaddr, int irq, int
 	/* Make certain the data structures used by the LANCE are aligned and DMAble. */
 
 	lp = kzalloc(sizeof(*lp), GFP_DMA | GFP_KERNEL);
-	if(lp==NULL)
-		return -ENODEV;
+	if (!lp)
+		return -ENOMEM;
 	if (lance_debug > 6) printk(" (#0x%05lx)", (unsigned long)lp);
 	dev->ml_priv = lp;
 	lp->name = chipname;
diff --git a/drivers/net/ethernet/arc/emac_main.c b/drivers/net/ethernet/arc/emac_main.c
index abe1eabc0171..6446af1403f7 100644
--- a/drivers/net/ethernet/arc/emac_main.c
+++ b/drivers/net/ethernet/arc/emac_main.c
@@ -163,7 +163,7 @@ static void arc_emac_tx_clean(struct net_device *ndev)
 		struct sk_buff *skb = tx_buff->skb;
 		unsigned int info = le32_to_cpu(txbd->info);
 
-		if ((info & FOR_EMAC) || !txbd->data)
+		if ((info & FOR_EMAC) || !txbd->data || !skb)
 			break;
 
 		if (unlikely(info & (DROP | DEFR | LTCL | UFLO))) {
@@ -191,6 +191,7 @@ static void arc_emac_tx_clean(struct net_device *ndev)
 
 		txbd->data = 0;
 		txbd->info = 0;
+		tx_buff->skb = NULL;
 
 		*txbd_dirty = (*txbd_dirty + 1) % TX_BD_NUM;
 	}
@@ -446,6 +447,9 @@ static int arc_emac_open(struct net_device *ndev)
 		*last_rx_bd = (*last_rx_bd + 1) % RX_BD_NUM;
 	}
 
+	priv->txbd_curr = 0;
+	priv->txbd_dirty = 0;
+
 	/* Clean Tx BD's */
 	memset(priv->txbd, 0, TX_RING_SZ);
 
@@ -514,6 +518,64 @@ static void arc_emac_set_rx_mode(struct net_device *ndev)
 }
 
 /**
+ * arc_free_tx_queue - free skb from tx queue
+ * @ndev:	Pointer to the network device.
+ *
+ * This function must be called while EMAC disable
+ */
+static void arc_free_tx_queue(struct net_device *ndev)
+{
+	struct arc_emac_priv *priv = netdev_priv(ndev);
+	unsigned int i;
+
+	for (i = 0; i < TX_BD_NUM; i++) {
+		struct arc_emac_bd *txbd = &priv->txbd[i];
+		struct buffer_state *tx_buff = &priv->tx_buff[i];
+
+		if (tx_buff->skb) {
+			dma_unmap_single(&ndev->dev, dma_unmap_addr(tx_buff, addr),
+					 dma_unmap_len(tx_buff, len), DMA_TO_DEVICE);
+
+			/* return the sk_buff to system */
+			dev_kfree_skb_irq(tx_buff->skb);
+		}
+
+		txbd->info = 0;
+		txbd->data = 0;
+		tx_buff->skb = NULL;
+	}
+}
+
+/**
+ * arc_free_rx_queue - free skb from rx queue
+ * @ndev:	Pointer to the network device.
+ *
+ * This function must be called while EMAC disable
+ */
+static void arc_free_rx_queue(struct net_device *ndev)
+{
+	struct arc_emac_priv *priv = netdev_priv(ndev);
+	unsigned int i;
+
+	for (i = 0; i < RX_BD_NUM; i++) {
+		struct arc_emac_bd *rxbd = &priv->rxbd[i];
+		struct buffer_state *rx_buff = &priv->rx_buff[i];
+
+		if (rx_buff->skb) {
+			dma_unmap_single(&ndev->dev, dma_unmap_addr(rx_buff, addr),
+					dma_unmap_len(rx_buff, len), DMA_FROM_DEVICE);
+
+			/* return the sk_buff to system */
+			dev_kfree_skb_irq(rx_buff->skb);
+		}
+
+		rxbd->info = 0;
+		rxbd->data = 0;
+		rx_buff->skb = NULL;
+	}
+}
+
+/**
  * arc_emac_stop - Close the network device.
  * @ndev:	Pointer to the network device.
  *
@@ -534,6 +596,10 @@ static int arc_emac_stop(struct net_device *ndev)
 	/* Disable EMAC */
 	arc_reg_clr(priv, R_CTRL, EN_MASK);
 
+	/* Return the sk_buff to system */
+	arc_free_tx_queue(ndev);
+	arc_free_rx_queue(ndev);
+
 	return 0;
 }
 
@@ -610,7 +676,6 @@ static int arc_emac_tx(struct sk_buff *skb, struct net_device *ndev)
 	dma_unmap_addr_set(&priv->tx_buff[*txbd_curr], addr, addr);
 	dma_unmap_len_set(&priv->tx_buff[*txbd_curr], len, len);
 
-	priv->tx_buff[*txbd_curr].skb = skb;
 	priv->txbd[*txbd_curr].data = cpu_to_le32(addr);
 
 	/* Make sure pointer to data buffer is set */
@@ -620,6 +685,11 @@ static int arc_emac_tx(struct sk_buff *skb, struct net_device *ndev)
 
 	*info = cpu_to_le32(FOR_EMAC | FIRST_OR_LAST_MASK | len);
 
+	/* Make sure info word is set */
+	wmb();
+
+	priv->tx_buff[*txbd_curr].skb = skb;
+
 	/* Increment index to point to the next BD */
 	*txbd_curr = (*txbd_curr + 1) % TX_BD_NUM;
 
diff --git a/drivers/net/ethernet/aurora/nb8800.c b/drivers/net/ethernet/aurora/nb8800.c
index f71ab2647a3b..08a23e6b60e9 100644
--- a/drivers/net/ethernet/aurora/nb8800.c
+++ b/drivers/net/ethernet/aurora/nb8800.c
@@ -1460,7 +1460,19 @@ static int nb8800_probe(struct platform_device *pdev)
 		goto err_disable_clk;
 	}
 
-	priv->phy_node = of_parse_phandle(pdev->dev.of_node, "phy-handle", 0);
+	if (of_phy_is_fixed_link(pdev->dev.of_node)) {
+		ret = of_phy_register_fixed_link(pdev->dev.of_node);
+		if (ret < 0) {
+			dev_err(&pdev->dev, "bad fixed-link spec\n");
+			goto err_free_bus;
+		}
+		priv->phy_node = of_node_get(pdev->dev.of_node);
+	}
+
+	if (!priv->phy_node)
+		priv->phy_node = of_parse_phandle(pdev->dev.of_node,
+						  "phy-handle", 0);
+
 	if (!priv->phy_node) {
 		dev_err(&pdev->dev, "no PHY specified\n");
 		ret = -ENODEV;
diff --git a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_hsi.h b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_hsi.h
index 27aa0802d87d..91874d24fd56 100644
--- a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_hsi.h
+++ b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_hsi.h
@@ -4896,9 +4896,9 @@ struct c2s_pri_trans_table_entry {
  * cfc delete event data
  */
 struct cfc_del_event_data {
-	u32 cid;
-	u32 reserved0;
-	u32 reserved1;
+	__le32 cid;
+	__le32 reserved0;
+	__le32 reserved1;
 };
 
 
@@ -5114,15 +5114,9 @@ struct vf_pf_channel_zone_trigger {
  * zone that triggers the in-bound interrupt
  */
 struct trigger_vf_zone {
-#if defined(__BIG_ENDIAN)
-	u16 reserved1;
-	u8 reserved0;
-	struct vf_pf_channel_zone_trigger vf_pf_channel;
-#elif defined(__LITTLE_ENDIAN)
 	struct vf_pf_channel_zone_trigger vf_pf_channel;
 	u8 reserved0;
 	u16 reserved1;
-#endif
 	u32 reserved2;
 };
 
@@ -5207,9 +5201,9 @@ struct e2_integ_data {
  * set mac event data
  */
 struct eth_event_data {
-	u32 echo;
-	u32 reserved0;
-	u32 reserved1;
+	__le32 echo;
+	__le32 reserved0;
+	__le32 reserved1;
 };
 
 
@@ -5219,9 +5213,9 @@ struct eth_event_data {
 struct vf_pf_event_data {
 	u8 vf_id;
 	u8 reserved0;
-	u16 reserved1;
-	u32 msg_addr_lo;
-	u32 msg_addr_hi;
+	__le16 reserved1;
+	__le32 msg_addr_lo;
+	__le32 msg_addr_hi;
 };
 
 /*
@@ -5230,9 +5224,9 @@ struct vf_pf_event_data {
 struct vf_flr_event_data {
 	u8 vf_id;
 	u8 reserved0;
-	u16 reserved1;
-	u32 reserved2;
-	u32 reserved3;
+	__le16 reserved1;
+	__le32 reserved2;
+	__le32 reserved3;
 };
 
 /*
@@ -5241,9 +5235,9 @@ struct vf_flr_event_data {
 struct malicious_vf_event_data {
 	u8 vf_id;
 	u8 err_id;
-	u16 reserved1;
-	u32 reserved2;
-	u32 reserved3;
+	__le16 reserved1;
+	__le32 reserved2;
+	__le32 reserved3;
 };
 
 /*
diff --git a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_link.c b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_link.c
index d946bba43726..1fb80100e5e7 100644
--- a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_link.c
+++ b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_link.c
@@ -6185,26 +6185,80 @@ static int bnx2x_format_ver(u32 num, u8 *str, u16 *len)
 		shift -= 4;
 		digit = ((num & mask) >> shift);
 		if (digit == 0 && remove_leading_zeros) {
-			mask = mask >> 4;
-			continue;
-		} else if (digit < 0xa)
-			*str_ptr = digit + '0';
-		else
-			*str_ptr = digit - 0xa + 'a';
-		remove_leading_zeros = 0;
-		str_ptr++;
-		(*len)--;
+			*str_ptr = '0';
+		} else {
+			if (digit < 0xa)
+				*str_ptr = digit + '0';
+			else
+				*str_ptr = digit - 0xa + 'a';
+
+			remove_leading_zeros = 0;
+			str_ptr++;
+			(*len)--;
+		}
 		mask = mask >> 4;
 		if (shift == 4*4) {
+			if (remove_leading_zeros) {
+				str_ptr++;
+				(*len)--;
+			}
 			*str_ptr = '.';
 			str_ptr++;
 			(*len)--;
 			remove_leading_zeros = 1;
 		}
 	}
+	if (remove_leading_zeros)
+		(*len)--;
 	return 0;
 }
 
+static int bnx2x_3_seq_format_ver(u32 num, u8 *str, u16 *len)
+{
+	u8 *str_ptr = str;
+	u32 mask = 0x00f00000;
+	u8 shift = 8*3;
+	u8 digit;
+	u8 remove_leading_zeros = 1;
+
+	if (*len < 10) {
+		/* Need more than 10chars for this format */
+		*str_ptr = '\0';
+		(*len)--;
+		return -EINVAL;
+	}
+
+	while (shift > 0) {
+		shift -= 4;
+		digit = ((num & mask) >> shift);
+		if (digit == 0 && remove_leading_zeros) {
+			*str_ptr = '0';
+		} else {
+			if (digit < 0xa)
+				*str_ptr = digit + '0';
+			else
+				*str_ptr = digit - 0xa + 'a';
+
+			remove_leading_zeros = 0;
+			str_ptr++;
+			(*len)--;
+		}
+		mask = mask >> 4;
+		if ((shift == 4*4) || (shift == 4*2)) {
+			if (remove_leading_zeros) {
+				str_ptr++;
+				(*len)--;
+			}
+			*str_ptr = '.';
+			str_ptr++;
+			(*len)--;
+			remove_leading_zeros = 1;
+		}
+	}
+	if (remove_leading_zeros)
+		(*len)--;
+	return 0;
+}
 
 static int bnx2x_null_format_ver(u32 spirom_ver, u8 *str, u16 *len)
 {
@@ -9677,8 +9731,9 @@ static void bnx2x_save_848xx_spirom_version(struct bnx2x_phy *phy,
 
 	if (bnx2x_is_8483x_8485x(phy)) {
 		bnx2x_cl45_read(bp, phy, MDIO_CTL_DEVAD, 0x400f, &fw_ver1);
-		bnx2x_save_spirom_version(bp, port, fw_ver1 & 0xfff,
-				phy->ver_addr);
+		if (phy->type != PORT_HW_CFG_XGXS_EXT_PHY_TYPE_BCM84858)
+			fw_ver1 &= 0xfff;
+		bnx2x_save_spirom_version(bp, port, fw_ver1, phy->ver_addr);
 	} else {
 		/* For 32-bit registers in 848xx, access via MDIO2ARM i/f. */
 		/* (1) set reg 0xc200_0014(SPI_BRIDGE_CTRL_2) to 0x03000000 */
@@ -9732,16 +9787,32 @@ static void bnx2x_save_848xx_spirom_version(struct bnx2x_phy *phy,
 static void bnx2x_848xx_set_led(struct bnx2x *bp,
 				struct bnx2x_phy *phy)
 {
-	u16 val, offset, i;
+	u16 val, led3_blink_rate, offset, i;
 	static struct bnx2x_reg_set reg_set[] = {
 		{MDIO_PMA_DEVAD, MDIO_PMA_REG_8481_LED1_MASK, 0x0080},
 		{MDIO_PMA_DEVAD, MDIO_PMA_REG_8481_LED2_MASK, 0x0018},
 		{MDIO_PMA_DEVAD, MDIO_PMA_REG_8481_LED3_MASK, 0x0006},
-		{MDIO_PMA_DEVAD, MDIO_PMA_REG_8481_LED3_BLINK, 0x0000},
 		{MDIO_PMA_DEVAD, MDIO_PMA_REG_84823_CTL_SLOW_CLK_CNT_HIGH,
 			MDIO_PMA_REG_84823_BLINK_RATE_VAL_15P9HZ},
 		{MDIO_AN_DEVAD, 0xFFFB, 0xFFFD}
 	};
+
+	if (phy->type == PORT_HW_CFG_XGXS_EXT_PHY_TYPE_BCM84858) {
+		/* Set LED5 source */
+		bnx2x_cl45_write(bp, phy,
+				 MDIO_PMA_DEVAD,
+				 MDIO_PMA_REG_8481_LED5_MASK,
+				 0x90);
+		led3_blink_rate = 0x000f;
+	} else {
+		led3_blink_rate = 0x0000;
+	}
+	/* Set LED3 BLINK */
+	bnx2x_cl45_write(bp, phy,
+			 MDIO_PMA_DEVAD,
+			 MDIO_PMA_REG_8481_LED3_BLINK,
+			 led3_blink_rate);
+
 	/* PHYC_CTL_LED_CTL */
 	bnx2x_cl45_read(bp, phy,
 			MDIO_PMA_DEVAD,
@@ -9749,6 +9820,9 @@ static void bnx2x_848xx_set_led(struct bnx2x *bp,
 	val &= 0xFE00;
 	val |= 0x0092;
 
+	if (phy->type == PORT_HW_CFG_XGXS_EXT_PHY_TYPE_BCM84858)
+		val |= 2 << 12; /* LED5 ON based on source */
+
 	bnx2x_cl45_write(bp, phy,
 			 MDIO_PMA_DEVAD,
 			 MDIO_PMA_REG_8481_LINK_SIGNAL, val);
@@ -9762,10 +9836,17 @@ static void bnx2x_848xx_set_led(struct bnx2x *bp,
 	else
 		offset = MDIO_PMA_REG_84823_CTL_LED_CTL_1;
 
-	/* stretch_en for LED3*/
+	if (phy->type == PORT_HW_CFG_XGXS_EXT_PHY_TYPE_BCM84858)
+		val = MDIO_PMA_REG_84858_ALLOW_GPHY_ACT |
+		      MDIO_PMA_REG_84823_LED3_STRETCH_EN;
+	else
+		val = MDIO_PMA_REG_84823_LED3_STRETCH_EN;
+
+	/* stretch_en for LEDs */
 	bnx2x_cl45_read_or_write(bp, phy,
-				 MDIO_PMA_DEVAD, offset,
-				 MDIO_PMA_REG_84823_LED3_STRETCH_EN);
+				 MDIO_PMA_DEVAD,
+				 offset,
+				 val);
 }
 
 static void bnx2x_848xx_specific_func(struct bnx2x_phy *phy,
@@ -9775,7 +9856,7 @@ static void bnx2x_848xx_specific_func(struct bnx2x_phy *phy,
 	struct bnx2x *bp = params->bp;
 	switch (action) {
 	case PHY_INIT:
-		if (!bnx2x_is_8483x_8485x(phy)) {
+		if (bnx2x_is_8483x_8485x(phy)) {
 			/* Save spirom version */
 			bnx2x_save_848xx_spirom_version(phy, bp, params->port);
 		}
@@ -10036,15 +10117,20 @@ static int bnx2x_84858_cmd_hdlr(struct bnx2x_phy *phy,
 
 static int bnx2x_84833_cmd_hdlr(struct bnx2x_phy *phy,
 				struct link_params *params, u16 fw_cmd,
-				u16 cmd_args[], int argc)
+				u16 cmd_args[], int argc, int process)
 {
 	int idx;
 	u16 val;
 	struct bnx2x *bp = params->bp;
-	/* Write CMD_OPEN_OVERRIDE to STATUS reg */
-	bnx2x_cl45_write(bp, phy, MDIO_CTL_DEVAD,
-			MDIO_848xx_CMD_HDLR_STATUS,
-			PHY84833_STATUS_CMD_OPEN_OVERRIDE);
+	int rc = 0;
+
+	if (process == PHY84833_MB_PROCESS2) {
+		/* Write CMD_OPEN_OVERRIDE to STATUS reg */
+		bnx2x_cl45_write(bp, phy, MDIO_CTL_DEVAD,
+				 MDIO_848xx_CMD_HDLR_STATUS,
+				 PHY84833_STATUS_CMD_OPEN_OVERRIDE);
+	}
+
 	for (idx = 0; idx < PHY848xx_CMDHDLR_WAIT; idx++) {
 		bnx2x_cl45_read(bp, phy, MDIO_CTL_DEVAD,
 				MDIO_848xx_CMD_HDLR_STATUS, &val);
@@ -10054,15 +10140,27 @@ static int bnx2x_84833_cmd_hdlr(struct bnx2x_phy *phy,
 	}
 	if (idx >= PHY848xx_CMDHDLR_WAIT) {
 		DP(NETIF_MSG_LINK, "FW cmd: FW not ready.\n");
+		/* if the status is CMD_COMPLETE_PASS or CMD_COMPLETE_ERROR
+		 * clear the status to CMD_CLEAR_COMPLETE
+		 */
+		if (val == PHY84833_STATUS_CMD_COMPLETE_PASS ||
+		    val == PHY84833_STATUS_CMD_COMPLETE_ERROR) {
+			bnx2x_cl45_write(bp, phy, MDIO_CTL_DEVAD,
+					 MDIO_848xx_CMD_HDLR_STATUS,
+					 PHY84833_STATUS_CMD_CLEAR_COMPLETE);
+		}
 		return -EINVAL;
 	}
-
-	/* Prepare argument(s) and issue command */
-	for (idx = 0; idx < argc; idx++) {
-		bnx2x_cl45_write(bp, phy, MDIO_CTL_DEVAD,
-				MDIO_848xx_CMD_HDLR_DATA1 + idx,
-				cmd_args[idx]);
+	if (process == PHY84833_MB_PROCESS1 ||
+	    process == PHY84833_MB_PROCESS2) {
+		/* Prepare argument(s) */
+		for (idx = 0; idx < argc; idx++) {
+			bnx2x_cl45_write(bp, phy, MDIO_CTL_DEVAD,
+					 MDIO_848xx_CMD_HDLR_DATA1 + idx,
+					 cmd_args[idx]);
+		}
 	}
+
 	bnx2x_cl45_write(bp, phy, MDIO_CTL_DEVAD,
 			MDIO_848xx_CMD_HDLR_COMMAND, fw_cmd);
 	for (idx = 0; idx < PHY848xx_CMDHDLR_WAIT; idx++) {
@@ -10076,24 +10174,30 @@ static int bnx2x_84833_cmd_hdlr(struct bnx2x_phy *phy,
 	if ((idx >= PHY848xx_CMDHDLR_WAIT) ||
 	    (val == PHY84833_STATUS_CMD_COMPLETE_ERROR)) {
 		DP(NETIF_MSG_LINK, "FW cmd failed.\n");
-		return -EINVAL;
+		rc = -EINVAL;
 	}
-	/* Gather returning data */
-	for (idx = 0; idx < argc; idx++) {
-		bnx2x_cl45_read(bp, phy, MDIO_CTL_DEVAD,
-				MDIO_848xx_CMD_HDLR_DATA1 + idx,
-				&cmd_args[idx]);
+	if (process == PHY84833_MB_PROCESS3 && rc == 0) {
+		/* Gather returning data */
+		for (idx = 0; idx < argc; idx++) {
+			bnx2x_cl45_read(bp, phy, MDIO_CTL_DEVAD,
+					MDIO_848xx_CMD_HDLR_DATA1 + idx,
+					&cmd_args[idx]);
+		}
 	}
-	bnx2x_cl45_write(bp, phy, MDIO_CTL_DEVAD,
-			MDIO_848xx_CMD_HDLR_STATUS,
-			PHY84833_STATUS_CMD_CLEAR_COMPLETE);
-	return 0;
+	if (val == PHY84833_STATUS_CMD_COMPLETE_ERROR ||
+	    val == PHY84833_STATUS_CMD_COMPLETE_PASS) {
+		bnx2x_cl45_write(bp, phy, MDIO_CTL_DEVAD,
+				 MDIO_848xx_CMD_HDLR_STATUS,
+				 PHY84833_STATUS_CMD_CLEAR_COMPLETE);
+	}
+	return rc;
 }
 
 static int bnx2x_848xx_cmd_hdlr(struct bnx2x_phy *phy,
 				struct link_params *params,
 				u16 fw_cmd,
-				u16 cmd_args[], int argc)
+					   u16 cmd_args[], int argc,
+					   int process)
 {
 	struct bnx2x *bp = params->bp;
 
@@ -10106,7 +10210,7 @@ static int bnx2x_848xx_cmd_hdlr(struct bnx2x_phy *phy,
 					    argc);
 	} else {
 		return bnx2x_84833_cmd_hdlr(phy, params, fw_cmd, cmd_args,
-					    argc);
+					    argc, process);
 	}
 }
 
@@ -10133,7 +10237,7 @@ static int bnx2x_848xx_pair_swap_cfg(struct bnx2x_phy *phy,
 
 	status = bnx2x_848xx_cmd_hdlr(phy, params,
 				      PHY848xx_CMD_SET_PAIR_SWAP, data,
-				      PHY848xx_CMDHDLR_MAX_ARGS);
+				      2, PHY84833_MB_PROCESS2);
 	if (status == 0)
 		DP(NETIF_MSG_LINK, "Pairswap OK, val=0x%x\n", data[1]);
 
@@ -10222,8 +10326,8 @@ static int bnx2x_8483x_disable_eee(struct bnx2x_phy *phy,
 	DP(NETIF_MSG_LINK, "Don't Advertise 10GBase-T EEE\n");
 
 	/* Prevent Phy from working in EEE and advertising it */
-	rc = bnx2x_848xx_cmd_hdlr(phy, params,
-				  PHY848xx_CMD_SET_EEE_MODE, &cmd_args, 1);
+	rc = bnx2x_848xx_cmd_hdlr(phy, params, PHY848xx_CMD_SET_EEE_MODE,
+				  &cmd_args, 1, PHY84833_MB_PROCESS1);
 	if (rc) {
 		DP(NETIF_MSG_LINK, "EEE disable failed.\n");
 		return rc;
@@ -10240,8 +10344,8 @@ static int bnx2x_8483x_enable_eee(struct bnx2x_phy *phy,
 	struct bnx2x *bp = params->bp;
 	u16 cmd_args = 1;
 
-	rc = bnx2x_848xx_cmd_hdlr(phy, params,
-				  PHY848xx_CMD_SET_EEE_MODE, &cmd_args, 1);
+	rc = bnx2x_848xx_cmd_hdlr(phy, params, PHY848xx_CMD_SET_EEE_MODE,
+				  &cmd_args, 1, PHY84833_MB_PROCESS1);
 	if (rc) {
 		DP(NETIF_MSG_LINK, "EEE enable failed.\n");
 		return rc;
@@ -10362,7 +10466,7 @@ static int bnx2x_848x3_config_init(struct bnx2x_phy *phy,
 		cmd_args[3] = PHY84833_CONSTANT_LATENCY;
 		rc = bnx2x_848xx_cmd_hdlr(phy, params,
 					  PHY848xx_CMD_SET_EEE_MODE, cmd_args,
-					  PHY848xx_CMDHDLR_MAX_ARGS);
+					  4, PHY84833_MB_PROCESS1);
 		if (rc)
 			DP(NETIF_MSG_LINK, "Cfg AutogrEEEn failed.\n");
 	}
@@ -10416,6 +10520,32 @@ static int bnx2x_848x3_config_init(struct bnx2x_phy *phy,
 		vars->eee_status &= ~SHMEM_EEE_SUPPORTED_MASK;
 	}
 
+	if (phy->type == PORT_HW_CFG_XGXS_EXT_PHY_TYPE_BCM84833) {
+		/* Additional settings for jumbo packets in 1000BASE-T mode */
+		/* Allow rx extended length */
+		bnx2x_cl45_read(bp, phy, MDIO_AN_DEVAD,
+				MDIO_AN_REG_8481_AUX_CTRL, &val);
+		val |= 0x4000;
+		bnx2x_cl45_write(bp, phy, MDIO_AN_DEVAD,
+				 MDIO_AN_REG_8481_AUX_CTRL, val);
+		/* TX FIFO Elasticity LSB */
+		bnx2x_cl45_read(bp, phy, MDIO_AN_DEVAD,
+				MDIO_AN_REG_8481_1G_100T_EXT_CTRL, &val);
+		val |= 0x1;
+		bnx2x_cl45_write(bp, phy, MDIO_AN_DEVAD,
+				 MDIO_AN_REG_8481_1G_100T_EXT_CTRL, val);
+		/* TX FIFO Elasticity MSB */
+		/* Enable expansion register 0x46 (Pattern Generator status) */
+		bnx2x_cl45_write(bp, phy, MDIO_AN_DEVAD,
+				 MDIO_AN_REG_8481_EXPANSION_REG_ACCESS, 0xf46);
+
+		bnx2x_cl45_read(bp, phy, MDIO_AN_DEVAD,
+				MDIO_AN_REG_8481_EXPANSION_REG_RD_RW, &val);
+		val |= 0x4000;
+		bnx2x_cl45_write(bp, phy, MDIO_AN_DEVAD,
+				 MDIO_AN_REG_8481_EXPANSION_REG_RD_RW, val);
+	}
+
 	if (bnx2x_is_8483x_8485x(phy)) {
 		/* Bring PHY out of super isolate mode as the final step. */
 		bnx2x_cl45_read_and_write(bp, phy,
@@ -10555,6 +10685,17 @@ static u8 bnx2x_848xx_read_status(struct bnx2x_phy *phy,
 	return link_up;
 }
 
+static int bnx2x_8485x_format_ver(u32 raw_ver, u8 *str, u16 *len)
+{
+	int status = 0;
+	u32 num;
+
+	num = ((raw_ver & 0xF80) >> 7) << 16 | ((raw_ver & 0x7F) << 8) |
+	      ((raw_ver & 0xF000) >> 12);
+	status = bnx2x_3_seq_format_ver(num, str, len);
+	return status;
+}
+
 static int bnx2x_848xx_format_ver(u32 raw_ver, u8 *str, u16 *len)
 {
 	int status = 0;
@@ -10651,10 +10792,25 @@ static void bnx2x_848xx_set_link_led(struct bnx2x_phy *phy,
 					0x0);
 
 		} else {
+			/* LED 1 OFF */
 			bnx2x_cl45_write(bp, phy,
 					 MDIO_PMA_DEVAD,
 					 MDIO_PMA_REG_8481_LED1_MASK,
 					 0x0);
+
+			if (phy->type ==
+				PORT_HW_CFG_XGXS_EXT_PHY_TYPE_BCM84858) {
+				/* LED 2 OFF */
+				bnx2x_cl45_write(bp, phy,
+						 MDIO_PMA_DEVAD,
+						 MDIO_PMA_REG_8481_LED2_MASK,
+						 0x0);
+				/* LED 3 OFF */
+				bnx2x_cl45_write(bp, phy,
+						 MDIO_PMA_DEVAD,
+						 MDIO_PMA_REG_8481_LED3_MASK,
+						 0x0);
+			}
 		}
 		break;
 	case LED_MODE_FRONT_PANEL_OFF:
@@ -10713,6 +10869,19 @@ static void bnx2x_848xx_set_link_led(struct bnx2x_phy *phy,
 						 MDIO_PMA_REG_8481_SIGNAL_MASK,
 						 0x0);
 			}
+			if (phy->type ==
+				PORT_HW_CFG_XGXS_EXT_PHY_TYPE_BCM84858) {
+				/* LED 2 OFF */
+				bnx2x_cl45_write(bp, phy,
+						 MDIO_PMA_DEVAD,
+						 MDIO_PMA_REG_8481_LED2_MASK,
+						 0x0);
+				/* LED 3 OFF */
+				bnx2x_cl45_write(bp, phy,
+						 MDIO_PMA_DEVAD,
+						 MDIO_PMA_REG_8481_LED3_MASK,
+						 0x0);
+			}
 		}
 		break;
 	case LED_MODE_ON:
@@ -10776,6 +10945,25 @@ static void bnx2x_848xx_set_link_led(struct bnx2x_phy *phy,
 						params->port*4,
 						NIG_MASK_MI_INT);
 				}
+			}
+			if (phy->type ==
+			    PORT_HW_CFG_XGXS_EXT_PHY_TYPE_BCM84858) {
+				/* Tell LED3 to constant on */
+				bnx2x_cl45_read(bp, phy,
+						MDIO_PMA_DEVAD,
+						MDIO_PMA_REG_8481_LINK_SIGNAL,
+						&val);
+				val &= ~(7<<6);
+				val |= (2<<6);  /* A83B[8:6]= 2 */
+				bnx2x_cl45_write(bp, phy,
+						 MDIO_PMA_DEVAD,
+						 MDIO_PMA_REG_8481_LINK_SIGNAL,
+						 val);
+				bnx2x_cl45_write(bp, phy,
+						 MDIO_PMA_DEVAD,
+						 MDIO_PMA_REG_8481_LED3_MASK,
+						 0x20);
+			} else {
 				bnx2x_cl45_write(bp, phy,
 						 MDIO_PMA_DEVAD,
 						 MDIO_PMA_REG_8481_SIGNAL_MASK,
@@ -10854,6 +11042,17 @@ static void bnx2x_848xx_set_link_led(struct bnx2x_phy *phy,
 					 MDIO_PMA_REG_8481_LINK_SIGNAL,
 					 val);
 			if (phy->type ==
+			    PORT_HW_CFG_XGXS_EXT_PHY_TYPE_BCM84858) {
+				bnx2x_cl45_write(bp, phy,
+						 MDIO_PMA_DEVAD,
+						 MDIO_PMA_REG_8481_LED2_MASK,
+						 0x18);
+				bnx2x_cl45_write(bp, phy,
+						 MDIO_PMA_DEVAD,
+						 MDIO_PMA_REG_8481_LED3_MASK,
+						 0x06);
+			}
+			if (phy->type ==
 			    PORT_HW_CFG_XGXS_EXT_PHY_TYPE_BCM84834) {
 				/* Restore LED4 source to external link,
 				 * and re-enable interrupts.
@@ -11982,7 +12181,7 @@ static const struct bnx2x_phy phy_84858 = {
 	.read_status	= (read_status_t)bnx2x_848xx_read_status,
 	.link_reset	= (link_reset_t)bnx2x_848x3_link_reset,
 	.config_loopback = (config_loopback_t)NULL,
-	.format_fw_ver	= (format_fw_ver_t)bnx2x_848xx_format_ver,
+	.format_fw_ver	= (format_fw_ver_t)bnx2x_8485x_format_ver,
 	.hw_reset	= (hw_reset_t)bnx2x_84833_hw_reset_phy,
 	.set_link_led	= (set_link_led_t)bnx2x_848xx_set_link_led,
 	.phy_specific_func = (phy_specific_func_t)bnx2x_848xx_specific_func
@@ -13807,8 +14006,10 @@ void bnx2x_period_func(struct link_params *params, struct link_vars *vars)
 	if (CHIP_IS_E3(bp)) {
 		struct bnx2x_phy *phy = &params->phy[INT_PHY];
 		bnx2x_set_aer_mmd(params, phy);
-		if ((phy->supported & SUPPORTED_20000baseKR2_Full) &&
-		    (phy->speed_cap_mask & PORT_HW_CFG_SPEED_CAPABILITY_D0_20G))
+		if (((phy->req_line_speed == SPEED_AUTO_NEG) &&
+		     (phy->speed_cap_mask &
+		      PORT_HW_CFG_SPEED_CAPABILITY_D0_20G)) ||
+		    (phy->req_line_speed == SPEED_20000))
 			bnx2x_check_kr2_wa(params, vars, phy);
 		bnx2x_check_over_curr(params, vars);
 		if (vars->rx_tx_asic_rst)
diff --git a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_main.c b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_main.c
index 6c4e3a69976f..2bf9c871144f 100644
--- a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_main.c
+++ b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_main.c
@@ -5280,14 +5280,14 @@ static void bnx2x_handle_classification_eqe(struct bnx2x *bp,
 {
 	unsigned long ramrod_flags = 0;
 	int rc = 0;
-	u32 cid = elem->message.data.eth_event.echo & BNX2X_SWCID_MASK;
+	u32 echo = le32_to_cpu(elem->message.data.eth_event.echo);
+	u32 cid = echo & BNX2X_SWCID_MASK;
 	struct bnx2x_vlan_mac_obj *vlan_mac_obj;
 
 	/* Always push next commands out, don't wait here */
 	__set_bit(RAMROD_CONT, &ramrod_flags);
 
-	switch (le32_to_cpu((__force __le32)elem->message.data.eth_event.echo)
-			    >> BNX2X_SWCID_SHIFT) {
+	switch (echo >> BNX2X_SWCID_SHIFT) {
 	case BNX2X_FILTER_MAC_PENDING:
 		DP(BNX2X_MSG_SP, "Got SETUP_MAC completions\n");
 		if (CNIC_LOADED(bp) && (cid == BNX2X_ISCSI_ETH_CID(bp)))
@@ -5308,8 +5308,7 @@ static void bnx2x_handle_classification_eqe(struct bnx2x *bp,
 		bnx2x_handle_mcast_eqe(bp);
 		return;
 	default:
-		BNX2X_ERR("Unsupported classification command: %d\n",
-			  elem->message.data.eth_event.echo);
+		BNX2X_ERR("Unsupported classification command: 0x%x\n", echo);
 		return;
 	}
 
@@ -5478,9 +5477,6 @@ static void bnx2x_eq_int(struct bnx2x *bp)
 			goto next_spqe;
 		}
 
-		/* elem CID originates from FW; actually LE */
-		cid = SW_CID((__force __le32)
-			     elem->message.data.cfc_del_event.cid);
 		opcode = elem->message.opcode;
 
 		/* handle eq element */
@@ -5503,6 +5499,10 @@ static void bnx2x_eq_int(struct bnx2x *bp)
 			 * we may want to verify here that the bp state is
 			 * HALTING
 			 */
+
+			/* elem CID originates from FW; actually LE */
+			cid = SW_CID(elem->message.data.cfc_del_event.cid);
+
 			DP(BNX2X_MSG_SP,
 			   "got delete ramrod for MULTI[%d]\n", cid);
 
@@ -5596,10 +5596,8 @@ static void bnx2x_eq_int(struct bnx2x *bp)
 		      BNX2X_STATE_OPENING_WAIT4_PORT):
 		case (EVENT_RING_OPCODE_RSS_UPDATE_RULES |
 		      BNX2X_STATE_CLOSING_WAIT4_HALT):
-			cid = elem->message.data.eth_event.echo &
-				BNX2X_SWCID_MASK;
 			DP(BNX2X_MSG_SP, "got RSS_UPDATE ramrod. CID %d\n",
-			   cid);
+			   SW_CID(elem->message.data.eth_event.echo));
 			rss_raw->clear_pending(rss_raw);
 			break;
 
@@ -5684,7 +5682,7 @@ static void bnx2x_sp_task(struct work_struct *work)
 		if (status & BNX2X_DEF_SB_IDX) {
 			struct bnx2x_fastpath *fp = bnx2x_fcoe_fp(bp);
 
-		if (FCOE_INIT(bp) &&
+			if (FCOE_INIT(bp) &&
 			    (bnx2x_has_rx_work(fp) || bnx2x_has_tx_work(fp))) {
 				/* Prevent local bottom-halves from running as
 				 * we are going to change the local NAPI list.
diff --git a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_reg.h b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_reg.h
index 4dead49bd5cb..a43dea259b12 100644
--- a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_reg.h
+++ b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_reg.h
@@ -7296,6 +7296,8 @@ Theotherbitsarereservedandshouldbezero*/
 #define MDIO_PMA_REG_84823_CTL_LED_CTL_1			0xa8e3
 #define MDIO_PMA_REG_84833_CTL_LED_CTL_1			0xa8ec
 #define MDIO_PMA_REG_84823_LED3_STRETCH_EN			0x0080
+/* BCM84858 only */
+#define MDIO_PMA_REG_84858_ALLOW_GPHY_ACT			0x8000
 
 /* BCM84833 only */
 #define MDIO_84833_TOP_CFG_FW_REV			0x400f
@@ -7337,6 +7339,10 @@ Theotherbitsarereservedandshouldbezero*/
 #define PHY84833_STATUS_CMD_NOT_OPEN_FOR_CMDS		0x0040
 #define PHY84833_STATUS_CMD_CLEAR_COMPLETE		0x0080
 #define PHY84833_STATUS_CMD_OPEN_OVERRIDE		0xa5a5
+/* Mailbox Process */
+#define PHY84833_MB_PROCESS1				1
+#define PHY84833_MB_PROCESS2				2
+#define PHY84833_MB_PROCESS3				3
 
 /* Mailbox status set used by 84858 only */
 #define PHY84858_STATUS_CMD_RECEIVED			0x0001
diff --git a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_sriov.c b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_sriov.c
index 9d027348cd09..632daff117d3 100644
--- a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_sriov.c
+++ b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_sriov.c
@@ -1672,11 +1672,12 @@ void bnx2x_vf_handle_classification_eqe(struct bnx2x *bp,
 {
 	unsigned long ramrod_flags = 0;
 	int rc = 0;
+	u32 echo = le32_to_cpu(elem->message.data.eth_event.echo);
 
 	/* Always push next commands out, don't wait here */
 	set_bit(RAMROD_CONT, &ramrod_flags);
 
-	switch (elem->message.data.eth_event.echo >> BNX2X_SWCID_SHIFT) {
+	switch (echo >> BNX2X_SWCID_SHIFT) {
 	case BNX2X_FILTER_MAC_PENDING:
 		rc = vfq->mac_obj.complete(bp, &vfq->mac_obj, elem,
 					   &ramrod_flags);
@@ -1686,8 +1687,7 @@ void bnx2x_vf_handle_classification_eqe(struct bnx2x *bp,
 					    &ramrod_flags);
 		break;
 	default:
-		BNX2X_ERR("Unsupported classification command: %d\n",
-			  elem->message.data.eth_event.echo);
+		BNX2X_ERR("Unsupported classification command: 0x%x\n", echo);
 		return;
 	}
 	if (rc < 0)
@@ -1747,16 +1747,14 @@ int bnx2x_iov_eq_sp_event(struct bnx2x *bp, union event_ring_elem *elem)
 
 	switch (opcode) {
 	case EVENT_RING_OPCODE_CFC_DEL:
-		cid = SW_CID((__force __le32)
-			     elem->message.data.cfc_del_event.cid);
+		cid = SW_CID(elem->message.data.cfc_del_event.cid);
 		DP(BNX2X_MSG_IOV, "checking cfc-del comp cid=%d\n", cid);
 		break;
 	case EVENT_RING_OPCODE_CLASSIFICATION_RULES:
 	case EVENT_RING_OPCODE_MULTICAST_RULES:
 	case EVENT_RING_OPCODE_FILTERS_RULES:
 	case EVENT_RING_OPCODE_RSS_UPDATE_RULES:
-		cid = (elem->message.data.eth_event.echo &
-		       BNX2X_SWCID_MASK);
+		cid = SW_CID(elem->message.data.eth_event.echo);
 		DP(BNX2X_MSG_IOV, "checking filtering comp cid=%d\n", cid);
 		break;
 	case EVENT_RING_OPCODE_VF_FLR:
diff --git a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_vfpf.c b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_vfpf.c
index 1374e5394a79..bfae300cf25f 100644
--- a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_vfpf.c
+++ b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_vfpf.c
@@ -2187,8 +2187,10 @@ void bnx2x_vf_mbx_schedule(struct bnx2x *bp,
 
 	/* Update VFDB with current message and schedule its handling */
 	mutex_lock(&BP_VFDB(bp)->event_mutex);
-	BP_VF_MBX(bp, vf_idx)->vf_addr_hi = vfpf_event->msg_addr_hi;
-	BP_VF_MBX(bp, vf_idx)->vf_addr_lo = vfpf_event->msg_addr_lo;
+	BP_VF_MBX(bp, vf_idx)->vf_addr_hi =
+		le32_to_cpu(vfpf_event->msg_addr_hi);
+	BP_VF_MBX(bp, vf_idx)->vf_addr_lo =
+		le32_to_cpu(vfpf_event->msg_addr_lo);
 	BP_VFDB(bp)->event_occur |= (1ULL << vf_idx);
 	mutex_unlock(&BP_VFDB(bp)->event_mutex);
 
diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt.c b/drivers/net/ethernet/broadcom/bnxt/bnxt.c
index 5dc89e527e7d..82f191382989 100644
--- a/drivers/net/ethernet/broadcom/bnxt/bnxt.c
+++ b/drivers/net/ethernet/broadcom/bnxt/bnxt.c
@@ -69,7 +69,7 @@ MODULE_VERSION(DRV_MODULE_VERSION);
 #define BNXT_RX_DMA_OFFSET NET_SKB_PAD
 #define BNXT_RX_COPY_THRESH 256
 
-#define BNXT_TX_PUSH_THRESH 92
+#define BNXT_TX_PUSH_THRESH 164
 
 enum board_idx {
 	BCM57301,
@@ -223,11 +223,12 @@ static netdev_tx_t bnxt_start_xmit(struct sk_buff *skb, struct net_device *dev)
 	}
 
 	if (free_size == bp->tx_ring_size && length <= bp->tx_push_thresh) {
-		struct tx_push_bd *push = txr->tx_push;
-		struct tx_bd *tx_push = &push->txbd1;
-		struct tx_bd_ext *tx_push1 = &push->txbd2;
-		void *pdata = tx_push1 + 1;
-		int j;
+		struct tx_push_buffer *tx_push_buf = txr->tx_push;
+		struct tx_push_bd *tx_push = &tx_push_buf->push_bd;
+		struct tx_bd_ext *tx_push1 = &tx_push->txbd2;
+		void *pdata = tx_push_buf->data;
+		u64 *end;
+		int j, push_len;
 
 		/* Set COAL_NOW to be ready quickly for the next push */
 		tx_push->tx_bd_len_flags_type =
@@ -247,6 +248,10 @@ static netdev_tx_t bnxt_start_xmit(struct sk_buff *skb, struct net_device *dev)
 		tx_push1->tx_bd_cfa_meta = cpu_to_le32(vlan_tag_flags);
 		tx_push1->tx_bd_cfa_action = cpu_to_le32(cfa_action);
 
+		end = pdata + length;
+		end = PTR_ALIGN(end, 8) - 1;
+		*end = 0;
+
 		skb_copy_from_linear_data(skb, pdata, len);
 		pdata += len;
 		for (j = 0; j < last_frag; j++) {
@@ -261,22 +266,29 @@ static netdev_tx_t bnxt_start_xmit(struct sk_buff *skb, struct net_device *dev)
 			pdata += skb_frag_size(frag);
 		}
 
-		memcpy(txbd, tx_push, sizeof(*txbd));
+		txbd->tx_bd_len_flags_type = tx_push->tx_bd_len_flags_type;
+		txbd->tx_bd_haddr = txr->data_mapping;
 		prod = NEXT_TX(prod);
 		txbd = &txr->tx_desc_ring[TX_RING(prod)][TX_IDX(prod)];
 		memcpy(txbd, tx_push1, sizeof(*txbd));
 		prod = NEXT_TX(prod);
-		push->doorbell =
+		tx_push->doorbell =
 			cpu_to_le32(DB_KEY_TX_PUSH | DB_LONG_TX_PUSH | prod);
 		txr->tx_prod = prod;
 
 		netdev_tx_sent_queue(txq, skb->len);
 
-		__iowrite64_copy(txr->tx_doorbell, push,
-				 (length + sizeof(*push) + 8) / 8);
+		push_len = (length + sizeof(*tx_push) + 7) / 8;
+		if (push_len > 16) {
+			__iowrite64_copy(txr->tx_doorbell, tx_push_buf, 16);
+			__iowrite64_copy(txr->tx_doorbell + 4, tx_push_buf + 1,
+					 push_len - 16);
+		} else {
+			__iowrite64_copy(txr->tx_doorbell, tx_push_buf,
+					 push_len);
+		}
 
 		tx_buf->is_push = 1;
-
 		goto tx_done;
 	}
 
@@ -1753,7 +1765,7 @@ static int bnxt_alloc_tx_rings(struct bnxt *bp)
 		push_size  = L1_CACHE_ALIGN(sizeof(struct tx_push_bd) +
 					bp->tx_push_thresh);
 
-		if (push_size > 128) {
+		if (push_size > 256) {
 			push_size = 0;
 			bp->tx_push_thresh = 0;
 		}
@@ -1772,7 +1784,6 @@ static int bnxt_alloc_tx_rings(struct bnxt *bp)
 			return rc;
 
 		if (bp->tx_push_size) {
-			struct tx_bd *txbd;
 			dma_addr_t mapping;
 
 			/* One pre-allocated DMA buffer to backup
@@ -1786,13 +1797,11 @@ static int bnxt_alloc_tx_rings(struct bnxt *bp)
 			if (!txr->tx_push)
 				return -ENOMEM;
 
-			txbd = &txr->tx_push->txbd1;
-
 			mapping = txr->tx_push_mapping +
 				sizeof(struct tx_push_bd);
-			txbd->tx_bd_haddr = cpu_to_le64(mapping);
+			txr->data_mapping = cpu_to_le64(mapping);
 
-			memset(txbd + 1, 0, sizeof(struct tx_bd_ext));
+			memset(txr->tx_push, 0, sizeof(struct tx_push_bd));
 		}
 		ring->queue_id = bp->q_info[j].queue_id;
 		if (i % bp->tx_nr_rings_per_tc == (bp->tx_nr_rings_per_tc - 1))
@@ -4546,20 +4555,18 @@ static int bnxt_update_phy_setting(struct bnxt *bp)
 	if (!(link_info->autoneg & BNXT_AUTONEG_FLOW_CTRL) &&
 	    link_info->force_pause_setting != link_info->req_flow_ctrl)
 		update_pause = true;
-	if (link_info->req_duplex != link_info->duplex_setting)
-		update_link = true;
 	if (!(link_info->autoneg & BNXT_AUTONEG_SPEED)) {
 		if (BNXT_AUTO_MODE(link_info->auto_mode))
 			update_link = true;
 		if (link_info->req_link_speed != link_info->force_link_speed)
 			update_link = true;
+		if (link_info->req_duplex != link_info->duplex_setting)
+			update_link = true;
 	} else {
 		if (link_info->auto_mode == BNXT_LINK_AUTO_NONE)
 			update_link = true;
 		if (link_info->advertising != link_info->auto_link_speeds)
 			update_link = true;
-		if (link_info->req_link_speed != link_info->auto_link_speed)
-			update_link = true;
 	}
 
 	if (update_link)
@@ -4636,7 +4643,7 @@ static int __bnxt_open_nic(struct bnxt *bp, bool irq_re_init, bool link_re_init)
 	if (link_re_init) {
 		rc = bnxt_update_phy_setting(bp);
 		if (rc)
-			goto open_err;
+			netdev_warn(bp->dev, "failed to update phy settings\n");
 	}
 
 	if (irq_re_init) {
@@ -4654,6 +4661,7 @@ static int __bnxt_open_nic(struct bnxt *bp, bool irq_re_init, bool link_re_init)
 	/* Enable TX queues */
 	bnxt_tx_enable(bp);
 	mod_timer(&bp->timer, jiffies + bp->current_interval);
+	bnxt_update_link(bp, true);
 
 	return 0;
 
@@ -5670,22 +5678,16 @@ static int bnxt_probe_phy(struct bnxt *bp)
 	}
 
 	/*initialize the ethool setting copy with NVM settings */
-	if (BNXT_AUTO_MODE(link_info->auto_mode))
-		link_info->autoneg |= BNXT_AUTONEG_SPEED;
-
-	if (link_info->auto_pause_setting & BNXT_LINK_PAUSE_BOTH) {
-		if (link_info->auto_pause_setting == BNXT_LINK_PAUSE_BOTH)
-			link_info->autoneg |= BNXT_AUTONEG_FLOW_CTRL;
+	if (BNXT_AUTO_MODE(link_info->auto_mode)) {
+		link_info->autoneg = BNXT_AUTONEG_SPEED |
+				     BNXT_AUTONEG_FLOW_CTRL;
+		link_info->advertising = link_info->auto_link_speeds;
 		link_info->req_flow_ctrl = link_info->auto_pause_setting;
-	} else if (link_info->force_pause_setting & BNXT_LINK_PAUSE_BOTH) {
+	} else {
+		link_info->req_link_speed = link_info->force_link_speed;
+		link_info->req_duplex = link_info->duplex_setting;
 		link_info->req_flow_ctrl = link_info->force_pause_setting;
 	}
-	link_info->req_duplex = link_info->duplex_setting;
-	if (link_info->autoneg & BNXT_AUTONEG_SPEED)
-		link_info->req_link_speed = link_info->auto_link_speed;
-	else
-		link_info->req_link_speed = link_info->force_link_speed;
-	link_info->advertising = link_info->auto_link_speeds;
 	snprintf(phy_ver, PHY_VER_STR_LEN, " ph %d.%d.%d",
 		 link_info->phy_ver[0],
 		 link_info->phy_ver[1],
diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt.h b/drivers/net/ethernet/broadcom/bnxt/bnxt.h
index 8af3ca8efcef..2be51b332652 100644
--- a/drivers/net/ethernet/broadcom/bnxt/bnxt.h
+++ b/drivers/net/ethernet/broadcom/bnxt/bnxt.h
@@ -411,8 +411,8 @@ struct rx_tpa_end_cmp_ext {
 
 #define BNXT_NUM_TESTS(bp)	0
 
-#define BNXT_DEFAULT_RX_RING_SIZE	1023
-#define BNXT_DEFAULT_TX_RING_SIZE	512
+#define BNXT_DEFAULT_RX_RING_SIZE	511
+#define BNXT_DEFAULT_TX_RING_SIZE	511
 
 #define MAX_TPA		64
 
@@ -523,10 +523,16 @@ struct bnxt_ring_struct {
 
 struct tx_push_bd {
 	__le32			doorbell;
-	struct tx_bd		txbd1;
+	__le32			tx_bd_len_flags_type;
+	u32			tx_bd_opaque;
 	struct tx_bd_ext	txbd2;
 };
 
+struct tx_push_buffer {
+	struct tx_push_bd	push_bd;
+	u32			data[25];
+};
+
 struct bnxt_tx_ring_info {
 	struct bnxt_napi	*bnapi;
 	u16			tx_prod;
@@ -538,8 +544,9 @@ struct bnxt_tx_ring_info {
 
 	dma_addr_t		tx_desc_mapping[MAX_TX_PAGES];
 
-	struct tx_push_bd	*tx_push;
+	struct tx_push_buffer	*tx_push;
 	dma_addr_t		tx_push_mapping;
+	__le64			data_mapping;
 
 #define BNXT_DEV_STATE_CLOSING	0x1
 	u32			dev_state;
diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.c b/drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.c
index 922b898e7a32..3238817dfd5f 100644
--- a/drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.c
+++ b/drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.c
@@ -486,15 +486,8 @@ static u32 bnxt_fw_to_ethtool_support_spds(struct bnxt_link_info *link_info)
 		speed_mask |= SUPPORTED_2500baseX_Full;
 	if (fw_speeds & BNXT_LINK_SPEED_MSK_10GB)
 		speed_mask |= SUPPORTED_10000baseT_Full;
-	/* TODO: support 25GB, 50GB with different cable type */
-	if (fw_speeds & BNXT_LINK_SPEED_MSK_20GB)
-		speed_mask |= SUPPORTED_20000baseMLD2_Full |
-			SUPPORTED_20000baseKR2_Full;
 	if (fw_speeds & BNXT_LINK_SPEED_MSK_40GB)
-		speed_mask |= SUPPORTED_40000baseKR4_Full |
-			SUPPORTED_40000baseCR4_Full |
-			SUPPORTED_40000baseSR4_Full |
-			SUPPORTED_40000baseLR4_Full;
+		speed_mask |= SUPPORTED_40000baseCR4_Full;
 
 	return speed_mask;
 }
@@ -514,15 +507,8 @@ static u32 bnxt_fw_to_ethtool_advertised_spds(struct bnxt_link_info *link_info)
 		speed_mask |= ADVERTISED_2500baseX_Full;
 	if (fw_speeds & BNXT_LINK_SPEED_MSK_10GB)
 		speed_mask |= ADVERTISED_10000baseT_Full;
-	/* TODO: how to advertise 20, 25, 40, 50GB with different cable type ?*/
-	if (fw_speeds & BNXT_LINK_SPEED_MSK_20GB)
-		speed_mask |= ADVERTISED_20000baseMLD2_Full |
-			      ADVERTISED_20000baseKR2_Full;
 	if (fw_speeds & BNXT_LINK_SPEED_MSK_40GB)
-		speed_mask |= ADVERTISED_40000baseKR4_Full |
-			      ADVERTISED_40000baseCR4_Full |
-			      ADVERTISED_40000baseSR4_Full |
-			      ADVERTISED_40000baseLR4_Full;
+		speed_mask |= ADVERTISED_40000baseCR4_Full;
 	return speed_mask;
 }
 
@@ -557,11 +543,12 @@ static int bnxt_get_settings(struct net_device *dev, struct ethtool_cmd *cmd)
 	u16 ethtool_speed;
 
 	cmd->supported = bnxt_fw_to_ethtool_support_spds(link_info);
+	cmd->supported |= SUPPORTED_Pause | SUPPORTED_Asym_Pause;
 
 	if (link_info->auto_link_speeds)
 		cmd->supported |= SUPPORTED_Autoneg;
 
-	if (BNXT_AUTO_MODE(link_info->auto_mode)) {
+	if (link_info->autoneg) {
 		cmd->advertising =
 			bnxt_fw_to_ethtool_advertised_spds(link_info);
 		cmd->advertising |= ADVERTISED_Autoneg;
@@ -570,28 +557,16 @@ static int bnxt_get_settings(struct net_device *dev, struct ethtool_cmd *cmd)
 		cmd->autoneg = AUTONEG_DISABLE;
 		cmd->advertising = 0;
 	}
-	if (link_info->auto_pause_setting & BNXT_LINK_PAUSE_BOTH) {
+	if (link_info->autoneg & BNXT_AUTONEG_FLOW_CTRL) {
 		if ((link_info->auto_pause_setting & BNXT_LINK_PAUSE_BOTH) ==
 		    BNXT_LINK_PAUSE_BOTH) {
 			cmd->advertising |= ADVERTISED_Pause;
-			cmd->supported |= SUPPORTED_Pause;
 		} else {
 			cmd->advertising |= ADVERTISED_Asym_Pause;
-			cmd->supported |= SUPPORTED_Asym_Pause;
 			if (link_info->auto_pause_setting &
 			    BNXT_LINK_PAUSE_RX)
 				cmd->advertising |= ADVERTISED_Pause;
 		}
-	} else if (link_info->force_pause_setting & BNXT_LINK_PAUSE_BOTH) {
-		if ((link_info->force_pause_setting & BNXT_LINK_PAUSE_BOTH) ==
-		    BNXT_LINK_PAUSE_BOTH) {
-			cmd->supported |= SUPPORTED_Pause;
-		} else {
-			cmd->supported |= SUPPORTED_Asym_Pause;
-			if (link_info->force_pause_setting &
-			    BNXT_LINK_PAUSE_RX)
-				cmd->supported |= SUPPORTED_Pause;
-		}
 	}
 
 	cmd->port = PORT_NONE;
@@ -670,6 +645,9 @@ static u16 bnxt_get_fw_auto_link_speeds(u32 advertising)
 	if (advertising & ADVERTISED_10000baseT_Full)
 		fw_speed_mask |= BNXT_LINK_SPEED_MSK_10GB;
 
+	if (advertising & ADVERTISED_40000baseCR4_Full)
+		fw_speed_mask |= BNXT_LINK_SPEED_MSK_40GB;
+
 	return fw_speed_mask;
 }
 
@@ -729,7 +707,7 @@ static int bnxt_set_settings(struct net_device *dev, struct ethtool_cmd *cmd)
 		speed = ethtool_cmd_speed(cmd);
 		link_info->req_link_speed = bnxt_get_fw_speed(dev, speed);
 		link_info->req_duplex = BNXT_LINK_DUPLEX_FULL;
-		link_info->autoneg &= ~BNXT_AUTONEG_SPEED;
+		link_info->autoneg = 0;
 		link_info->advertising = 0;
 	}
 
@@ -748,8 +726,7 @@ static void bnxt_get_pauseparam(struct net_device *dev,
 
 	if (BNXT_VF(bp))
 		return;
-	epause->autoneg = !!(link_info->auto_pause_setting &
-			     BNXT_LINK_PAUSE_BOTH);
+	epause->autoneg = !!(link_info->autoneg & BNXT_AUTONEG_FLOW_CTRL);
 	epause->rx_pause = ((link_info->pause & BNXT_LINK_PAUSE_RX) != 0);
 	epause->tx_pause = ((link_info->pause & BNXT_LINK_PAUSE_TX) != 0);
 }
@@ -765,6 +742,9 @@ static int bnxt_set_pauseparam(struct net_device *dev,
 		return rc;
 
 	if (epause->autoneg) {
+		if (!(link_info->autoneg & BNXT_AUTONEG_SPEED))
+			return -EINVAL;
+
 		link_info->autoneg |= BNXT_AUTONEG_FLOW_CTRL;
 		link_info->req_flow_ctrl |= BNXT_LINK_PAUSE_BOTH;
 	} else {
diff --git a/drivers/net/ethernet/broadcom/genet/bcmgenet.c b/drivers/net/ethernet/broadcom/genet/bcmgenet.c
index b15a60d787c7..d7e01a74e927 100644
--- a/drivers/net/ethernet/broadcom/genet/bcmgenet.c
+++ b/drivers/net/ethernet/broadcom/genet/bcmgenet.c
@@ -2445,8 +2445,7 @@ static void bcmgenet_irq_task(struct work_struct *work)
 	}
 
 	/* Link UP/DOWN event */
-	if ((priv->hw_params->flags & GENET_HAS_MDIO_INTR) &&
-	    (priv->irq0_stat & UMAC_IRQ_LINK_EVENT)) {
+	if (priv->irq0_stat & UMAC_IRQ_LINK_EVENT) {
 		phy_mac_interrupt(priv->phydev,
 				  !!(priv->irq0_stat & UMAC_IRQ_LINK_UP));
 		priv->irq0_stat &= ~UMAC_IRQ_LINK_EVENT;
diff --git a/drivers/net/ethernet/broadcom/tg3.c b/drivers/net/ethernet/broadcom/tg3.c
index 49eea8981332..3010080cfeee 100644
--- a/drivers/net/ethernet/broadcom/tg3.c
+++ b/drivers/net/ethernet/broadcom/tg3.c
@@ -7831,6 +7831,14 @@ static int tigon3_dma_hwbug_workaround(struct tg3_napi *tnapi,
 	return ret;
 }
 
+static bool tg3_tso_bug_gso_check(struct tg3_napi *tnapi, struct sk_buff *skb)
+{
+	/* Check if we will never have enough descriptors,
+	 * as gso_segs can be more than current ring size
+	 */
+	return skb_shinfo(skb)->gso_segs < tnapi->tx_pending / 3;
+}
+
 static netdev_tx_t tg3_start_xmit(struct sk_buff *, struct net_device *);
 
 /* Use GSO to workaround all TSO packets that meet HW bug conditions
@@ -7934,14 +7942,19 @@ static netdev_tx_t tg3_start_xmit(struct sk_buff *skb, struct net_device *dev)
 		 * vlan encapsulated.
 		 */
 		if (skb->protocol == htons(ETH_P_8021Q) ||
-		    skb->protocol == htons(ETH_P_8021AD))
-			return tg3_tso_bug(tp, tnapi, txq, skb);
+		    skb->protocol == htons(ETH_P_8021AD)) {
+			if (tg3_tso_bug_gso_check(tnapi, skb))
+				return tg3_tso_bug(tp, tnapi, txq, skb);
+			goto drop;
+		}
 
 		if (!skb_is_gso_v6(skb)) {
 			if (unlikely((ETH_HLEN + hdr_len) > 80) &&
-			    tg3_flag(tp, TSO_BUG))
-				return tg3_tso_bug(tp, tnapi, txq, skb);
-
+			    tg3_flag(tp, TSO_BUG)) {
+				if (tg3_tso_bug_gso_check(tnapi, skb))
+					return tg3_tso_bug(tp, tnapi, txq, skb);
+				goto drop;
+			}
 			ip_csum = iph->check;
 			ip_tot_len = iph->tot_len;
 			iph->check = 0;
@@ -8073,7 +8086,7 @@ static netdev_tx_t tg3_start_xmit(struct sk_buff *skb, struct net_device *dev)
 	if (would_hit_hwbug) {
 		tg3_tx_skb_unmap(tnapi, tnapi->tx_prod, i);
 
-		if (mss) {
+		if (mss && tg3_tso_bug_gso_check(tnapi, skb)) {
 			/* If it's a TSO packet, do GSO instead of
 			 * allocating and copying to a large linear SKB
 			 */
diff --git a/drivers/net/ethernet/brocade/bna/bna_tx_rx.c b/drivers/net/ethernet/brocade/bna/bna_tx_rx.c
index 04b0d16b210e..95bc470ae441 100644
--- a/drivers/net/ethernet/brocade/bna/bna_tx_rx.c
+++ b/drivers/net/ethernet/brocade/bna/bna_tx_rx.c
@@ -987,7 +987,7 @@ bna_rxf_ucast_cfg_apply(struct bna_rxf *rxf)
 	if (!list_empty(&rxf->ucast_pending_add_q)) {
 		mac = list_first_entry(&rxf->ucast_pending_add_q,
 				       struct bna_mac, qe);
-		list_add_tail(&mac->qe, &rxf->ucast_active_q);
+		list_move_tail(&mac->qe, &rxf->ucast_active_q);
 		bna_bfi_ucast_req(rxf, mac, BFI_ENET_H2I_MAC_UCAST_ADD_REQ);
 		return 1;
 	}
diff --git a/drivers/net/ethernet/cavium/liquidio/lio_main.c b/drivers/net/ethernet/cavium/liquidio/lio_main.c
index 872765527081..34d269cd5579 100644
--- a/drivers/net/ethernet/cavium/liquidio/lio_main.c
+++ b/drivers/net/ethernet/cavium/liquidio/lio_main.c
@@ -1683,7 +1683,7 @@ static int octeon_setup_droq(struct octeon_device *oct, int q_no, int num_descs,
 	dev_dbg(&oct->pci_dev->dev, "Creating Droq: %d\n", q_no);
 	/* droq creation and local register settings. */
 	ret_val = octeon_create_droq(oct, q_no, num_descs, desc_size, app_ctx);
-	if (ret_val == -1)
+	if (ret_val < 0)
 		return ret_val;
 
 	if (ret_val == 1) {
@@ -2524,7 +2524,7 @@ static void handle_timestamp(struct octeon_device *oct,
 
 	octeon_swap_8B_data(&resp->timestamp, 1);
 
-	if (unlikely((skb_shinfo(skb)->tx_flags | SKBTX_IN_PROGRESS) != 0)) {
+	if (unlikely((skb_shinfo(skb)->tx_flags & SKBTX_IN_PROGRESS) != 0)) {
 		struct skb_shared_hwtstamps ts;
 		u64 ns = resp->timestamp;
 
diff --git a/drivers/net/ethernet/cavium/liquidio/octeon_droq.c b/drivers/net/ethernet/cavium/liquidio/octeon_droq.c
index 4dba86eaa045..174072b3740b 100644
--- a/drivers/net/ethernet/cavium/liquidio/octeon_droq.c
+++ b/drivers/net/ethernet/cavium/liquidio/octeon_droq.c
@@ -983,5 +983,5 @@ int octeon_create_droq(struct octeon_device *oct,
 
 create_droq_fail:
 	octeon_delete_droq(oct, q_no);
-	return -1;
+	return -ENOMEM;
 }
diff --git a/drivers/net/ethernet/cavium/thunder/nic.h b/drivers/net/ethernet/cavium/thunder/nic.h
index 688828865c48..34e9acea8747 100644
--- a/drivers/net/ethernet/cavium/thunder/nic.h
+++ b/drivers/net/ethernet/cavium/thunder/nic.h
@@ -116,6 +116,15 @@
 #define NIC_PF_INTR_ID_MBOX0		8
 #define NIC_PF_INTR_ID_MBOX1		9
 
+/* Minimum FIFO level before all packets for the CQ are dropped
+ *
+ * This value ensures that once a packet has been "accepted"
+ * for reception it will not get dropped due to non-availability
+ * of CQ descriptor. An errata in HW mandates this value to be
+ * atleast 0x100.
+ */
+#define NICPF_CQM_MIN_DROP_LEVEL       0x100
+
 /* Global timer for CQ timer thresh interrupts
  * Calculated for SCLK of 700Mhz
  * value written should be a 1/16th of what is expected
diff --git a/drivers/net/ethernet/cavium/thunder/nic_main.c b/drivers/net/ethernet/cavium/thunder/nic_main.c
index 4dded90076c8..95f17f8cadac 100644
--- a/drivers/net/ethernet/cavium/thunder/nic_main.c
+++ b/drivers/net/ethernet/cavium/thunder/nic_main.c
@@ -304,6 +304,7 @@ static void nic_set_lmac_vf_mapping(struct nicpf *nic)
 static void nic_init_hw(struct nicpf *nic)
 {
 	int i;
+	u64 cqm_cfg;
 
 	/* Enable NIC HW block */
 	nic_reg_write(nic, NIC_PF_CFG, 0x3);
@@ -340,6 +341,11 @@ static void nic_init_hw(struct nicpf *nic)
 	/* Enable VLAN ethertype matching and stripping */
 	nic_reg_write(nic, NIC_PF_RX_ETYPE_0_7,
 		      (2 << 19) | (ETYPE_ALG_VLAN_STRIP << 16) | ETH_P_8021Q);
+
+	/* Check if HW expected value is higher (could be in future chips) */
+	cqm_cfg = nic_reg_read(nic, NIC_PF_CQM_CFG);
+	if (cqm_cfg < NICPF_CQM_MIN_DROP_LEVEL)
+		nic_reg_write(nic, NIC_PF_CQM_CFG, NICPF_CQM_MIN_DROP_LEVEL);
 }
 
 /* Channel parse index configuration */
diff --git a/drivers/net/ethernet/cavium/thunder/nic_reg.h b/drivers/net/ethernet/cavium/thunder/nic_reg.h
index dd536be20193..afb10e326b4f 100644
--- a/drivers/net/ethernet/cavium/thunder/nic_reg.h
+++ b/drivers/net/ethernet/cavium/thunder/nic_reg.h
@@ -21,7 +21,7 @@
 #define   NIC_PF_TCP_TIMER			(0x0060)
 #define   NIC_PF_BP_CFG				(0x0080)
 #define   NIC_PF_RRM_CFG			(0x0088)
-#define   NIC_PF_CQM_CF				(0x00A0)
+#define   NIC_PF_CQM_CFG			(0x00A0)
 #define   NIC_PF_CNM_CF				(0x00A8)
 #define   NIC_PF_CNM_STATUS			(0x00B0)
 #define   NIC_PF_CQ_AVG_CFG			(0x00C0)
diff --git a/drivers/net/ethernet/cavium/thunder/nicvf_main.c b/drivers/net/ethernet/cavium/thunder/nicvf_main.c
index c24cb2a86a42..a009bc30dc4d 100644
--- a/drivers/net/ethernet/cavium/thunder/nicvf_main.c
+++ b/drivers/net/ethernet/cavium/thunder/nicvf_main.c
@@ -574,8 +574,7 @@ static inline void nicvf_set_rxhash(struct net_device *netdev,
 
 static void nicvf_rcv_pkt_handler(struct net_device *netdev,
 				  struct napi_struct *napi,
-				  struct cmp_queue *cq,
-				  struct cqe_rx_t *cqe_rx, int cqe_type)
+				  struct cqe_rx_t *cqe_rx)
 {
 	struct sk_buff *skb;
 	struct nicvf *nic = netdev_priv(netdev);
@@ -591,7 +590,7 @@ static void nicvf_rcv_pkt_handler(struct net_device *netdev,
 	}
 
 	/* Check for errors */
-	err = nicvf_check_cqe_rx_errs(nic, cq, cqe_rx);
+	err = nicvf_check_cqe_rx_errs(nic, cqe_rx);
 	if (err && !cqe_rx->rb_cnt)
 		return;
 
@@ -682,8 +681,7 @@ loop:
 			   cq_idx, cq_desc->cqe_type);
 		switch (cq_desc->cqe_type) {
 		case CQE_TYPE_RX:
-			nicvf_rcv_pkt_handler(netdev, napi, cq,
-					      cq_desc, CQE_TYPE_RX);
+			nicvf_rcv_pkt_handler(netdev, napi, cq_desc);
 			work_done++;
 		break;
 		case CQE_TYPE_SEND:
@@ -1125,7 +1123,6 @@ int nicvf_stop(struct net_device *netdev)
 
 	/* Clear multiqset info */
 	nic->pnicvf = nic;
-	nic->sqs_count = 0;
 
 	return 0;
 }
@@ -1354,6 +1351,9 @@ void nicvf_update_stats(struct nicvf *nic)
 	drv_stats->tx_frames_ok = stats->tx_ucast_frames_ok +
 				  stats->tx_bcast_frames_ok +
 				  stats->tx_mcast_frames_ok;
+	drv_stats->rx_frames_ok = stats->rx_ucast_frames +
+				  stats->rx_bcast_frames +
+				  stats->rx_mcast_frames;
 	drv_stats->rx_drops = stats->rx_drop_red +
 			      stats->rx_drop_overrun;
 	drv_stats->tx_drops = stats->tx_drops;
@@ -1538,6 +1538,9 @@ static int nicvf_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
 
 	nicvf_send_vf_struct(nic);
 
+	if (!pass1_silicon(nic->pdev))
+		nic->hw_tso = true;
+
 	/* Check if this VF is in QS only mode */
 	if (nic->sqs_mode)
 		return 0;
@@ -1557,9 +1560,6 @@ static int nicvf_probe(struct pci_dev *pdev, const struct pci_device_id *ent)
 
 	netdev->vlan_features = NETIF_F_SG | NETIF_F_IP_CSUM | NETIF_F_TSO;
 
-	if (!pass1_silicon(nic->pdev))
-		nic->hw_tso = true;
-
 	netdev->netdev_ops = &nicvf_netdev_ops;
 	netdev->watchdog_timeo = NICVF_TX_TIMEOUT;
 
diff --git a/drivers/net/ethernet/cavium/thunder/nicvf_queues.c b/drivers/net/ethernet/cavium/thunder/nicvf_queues.c
index d0d1b5490061..767347b1f631 100644
--- a/drivers/net/ethernet/cavium/thunder/nicvf_queues.c
+++ b/drivers/net/ethernet/cavium/thunder/nicvf_queues.c
@@ -1329,16 +1329,12 @@ void nicvf_update_sq_stats(struct nicvf *nic, int sq_idx)
 }
 
 /* Check for errors in the receive cmp.queue entry */
-int nicvf_check_cqe_rx_errs(struct nicvf *nic,
-			    struct cmp_queue *cq, struct cqe_rx_t *cqe_rx)
+int nicvf_check_cqe_rx_errs(struct nicvf *nic, struct cqe_rx_t *cqe_rx)
 {
 	struct nicvf_hw_stats *stats = &nic->hw_stats;
-	struct nicvf_drv_stats *drv_stats = &nic->drv_stats;
 
-	if (!cqe_rx->err_level && !cqe_rx->err_opcode) {
-		drv_stats->rx_frames_ok++;
+	if (!cqe_rx->err_level && !cqe_rx->err_opcode)
 		return 0;
-	}
 
 	if (netif_msg_rx_err(nic))
 		netdev_err(nic->netdev,
diff --git a/drivers/net/ethernet/cavium/thunder/nicvf_queues.h b/drivers/net/ethernet/cavium/thunder/nicvf_queues.h
index c5030a7f213a..6673e1133523 100644
--- a/drivers/net/ethernet/cavium/thunder/nicvf_queues.h
+++ b/drivers/net/ethernet/cavium/thunder/nicvf_queues.h
@@ -338,8 +338,7 @@ u64  nicvf_queue_reg_read(struct nicvf *nic,
 /* Stats */
 void nicvf_update_rq_stats(struct nicvf *nic, int rq_idx);
 void nicvf_update_sq_stats(struct nicvf *nic, int sq_idx);
-int nicvf_check_cqe_rx_errs(struct nicvf *nic,
-			    struct cmp_queue *cq, struct cqe_rx_t *cqe_rx);
+int nicvf_check_cqe_rx_errs(struct nicvf *nic, struct cqe_rx_t *cqe_rx);
 int nicvf_check_cqe_tx_errs(struct nicvf *nic,
 			    struct cmp_queue *cq, struct cqe_send_t *cqe_tx);
 #endif /* NICVF_QUEUES_H */
diff --git a/drivers/net/ethernet/chelsio/cxgb3/t3_hw.c b/drivers/net/ethernet/chelsio/cxgb3/t3_hw.c
index ee04caa6c4d8..a89721fad633 100644
--- a/drivers/net/ethernet/chelsio/cxgb3/t3_hw.c
+++ b/drivers/net/ethernet/chelsio/cxgb3/t3_hw.c
@@ -681,6 +681,24 @@ int t3_seeprom_wp(struct adapter *adapter, int enable)
 	return t3_seeprom_write(adapter, EEPROM_STAT_ADDR, enable ? 0xc : 0);
 }
 
+static int vpdstrtouint(char *s, int len, unsigned int base, unsigned int *val)
+{
+	char tok[len + 1];
+
+	memcpy(tok, s, len);
+	tok[len] = 0;
+	return kstrtouint(strim(tok), base, val);
+}
+
+static int vpdstrtou16(char *s, int len, unsigned int base, u16 *val)
+{
+	char tok[len + 1];
+
+	memcpy(tok, s, len);
+	tok[len] = 0;
+	return kstrtou16(strim(tok), base, val);
+}
+
 /**
  *	get_vpd_params - read VPD parameters from VPD EEPROM
  *	@adapter: adapter to read
@@ -709,19 +727,19 @@ static int get_vpd_params(struct adapter *adapter, struct vpd_params *p)
 			return ret;
 	}
 
-	ret = kstrtouint(vpd.cclk_data, 10, &p->cclk);
+	ret = vpdstrtouint(vpd.cclk_data, vpd.cclk_len, 10, &p->cclk);
 	if (ret)
 		return ret;
-	ret = kstrtouint(vpd.mclk_data, 10, &p->mclk);
+	ret = vpdstrtouint(vpd.mclk_data, vpd.mclk_len, 10, &p->mclk);
 	if (ret)
 		return ret;
-	ret = kstrtouint(vpd.uclk_data, 10, &p->uclk);
+	ret = vpdstrtouint(vpd.uclk_data, vpd.uclk_len, 10, &p->uclk);
 	if (ret)
 		return ret;
-	ret = kstrtouint(vpd.mdc_data, 10, &p->mdc);
+	ret = vpdstrtouint(vpd.mdc_data, vpd.mdc_len, 10, &p->mdc);
 	if (ret)
 		return ret;
-	ret = kstrtouint(vpd.mt_data, 10, &p->mem_timing);
+	ret = vpdstrtouint(vpd.mt_data, vpd.mt_len, 10, &p->mem_timing);
 	if (ret)
 		return ret;
 	memcpy(p->sn, vpd.sn_data, SERNUM_LEN);
@@ -733,10 +751,12 @@ static int get_vpd_params(struct adapter *adapter, struct vpd_params *p)
 	} else {
 		p->port_type[0] = hex_to_bin(vpd.port0_data[0]);
 		p->port_type[1] = hex_to_bin(vpd.port1_data[0]);
-		ret = kstrtou16(vpd.xaui0cfg_data, 16, &p->xauicfg[0]);
+		ret = vpdstrtou16(vpd.xaui0cfg_data, vpd.xaui0cfg_len, 16,
+				  &p->xauicfg[0]);
 		if (ret)
 			return ret;
-		ret = kstrtou16(vpd.xaui1cfg_data, 16, &p->xauicfg[1]);
+		ret = vpdstrtou16(vpd.xaui1cfg_data, vpd.xaui1cfg_len, 16,
+				  &p->xauicfg[1]);
 		if (ret)
 			return ret;
 	}
diff --git a/drivers/net/ethernet/chelsio/cxgb4/t4_pci_id_tbl.h b/drivers/net/ethernet/chelsio/cxgb4/t4_pci_id_tbl.h
index a8dda635456d..06bc2d2e7a73 100644
--- a/drivers/net/ethernet/chelsio/cxgb4/t4_pci_id_tbl.h
+++ b/drivers/net/ethernet/chelsio/cxgb4/t4_pci_id_tbl.h
@@ -165,6 +165,7 @@ CH_PCI_DEVICE_ID_TABLE_DEFINE_BEGIN
 	CH_PCI_ID_TABLE_FENTRY(0x5098),	/* Custom 2x40G QSFP */
 	CH_PCI_ID_TABLE_FENTRY(0x5099),	/* Custom 2x40G QSFP */
 	CH_PCI_ID_TABLE_FENTRY(0x509a),	/* Custom T520-CR */
+	CH_PCI_ID_TABLE_FENTRY(0x509b),	/* Custom T540-CR LOM */
 
 	/* T6 adapters:
 	 */
diff --git a/drivers/net/ethernet/cisco/enic/enic.h b/drivers/net/ethernet/cisco/enic/enic.h
index 1671fa3332c2..7ba6d530b0c0 100644
--- a/drivers/net/ethernet/cisco/enic/enic.h
+++ b/drivers/net/ethernet/cisco/enic/enic.h
@@ -33,7 +33,7 @@
 
 #define DRV_NAME		"enic"
 #define DRV_DESCRIPTION		"Cisco VIC Ethernet NIC Driver"
-#define DRV_VERSION		"2.3.0.12"
+#define DRV_VERSION		"2.3.0.20"
 #define DRV_COPYRIGHT		"Copyright 2008-2013 Cisco Systems, Inc"
 
 #define ENIC_BARS_MAX		6
diff --git a/drivers/net/ethernet/cisco/enic/vnic_dev.c b/drivers/net/ethernet/cisco/enic/vnic_dev.c
index 1ffd1050860b..1fdf5fe12a95 100644
--- a/drivers/net/ethernet/cisco/enic/vnic_dev.c
+++ b/drivers/net/ethernet/cisco/enic/vnic_dev.c
@@ -298,7 +298,8 @@ static int _vnic_dev_cmd2(struct vnic_dev *vdev, enum vnic_devcmd_cmd cmd,
 			  int wait)
 {
 	struct devcmd2_controller *dc2c = vdev->devcmd2;
-	struct devcmd2_result *result = dc2c->result + dc2c->next_result;
+	struct devcmd2_result *result;
+	u8 color;
 	unsigned int i;
 	int delay, err;
 	u32 fetch_index, new_posted;
@@ -336,13 +337,17 @@ static int _vnic_dev_cmd2(struct vnic_dev *vdev, enum vnic_devcmd_cmd cmd,
 	if (dc2c->cmd_ring[posted].flags & DEVCMD2_FNORESULT)
 		return 0;
 
+	result = dc2c->result + dc2c->next_result;
+	color = dc2c->color;
+
+	dc2c->next_result++;
+	if (dc2c->next_result == dc2c->result_size) {
+		dc2c->next_result = 0;
+		dc2c->color = dc2c->color ? 0 : 1;
+	}
+
 	for (delay = 0; delay < wait; delay++) {
-		if (result->color == dc2c->color) {
-			dc2c->next_result++;
-			if (dc2c->next_result == dc2c->result_size) {
-				dc2c->next_result = 0;
-				dc2c->color = dc2c->color ? 0 : 1;
-			}
+		if (result->color == color) {
 			if (result->error) {
 				err = result->error;
 				if (err != ERR_ECMDUNKNOWN ||
diff --git a/drivers/net/ethernet/davicom/dm9000.c b/drivers/net/ethernet/davicom/dm9000.c
index cf94b72dbacd..48d91941408d 100644
--- a/drivers/net/ethernet/davicom/dm9000.c
+++ b/drivers/net/ethernet/davicom/dm9000.c
@@ -128,7 +128,6 @@ struct board_info {
 	struct resource *data_res;
 	struct resource	*addr_req;   /* resources requested */
 	struct resource *data_req;
-	struct resource *irq_res;
 
 	int		 irq_wake;
 
@@ -1300,22 +1299,16 @@ static int
 dm9000_open(struct net_device *dev)
 {
 	struct board_info *db = netdev_priv(dev);
-	unsigned long irqflags = db->irq_res->flags & IRQF_TRIGGER_MASK;
 
 	if (netif_msg_ifup(db))
 		dev_dbg(db->dev, "enabling %s\n", dev->name);
 
-	/* If there is no IRQ type specified, default to something that
-	 * may work, and tell the user that this is a problem */
-
-	if (irqflags == IRQF_TRIGGER_NONE)
-		irqflags = irq_get_trigger_type(dev->irq);
-
-	if (irqflags == IRQF_TRIGGER_NONE)
+	/* If there is no IRQ type specified, tell the user that this is a
+	 * problem
+	 */
+	if (irq_get_trigger_type(dev->irq) == IRQF_TRIGGER_NONE)
 		dev_warn(db->dev, "WARNING: no IRQ resource flags set.\n");
 
-	irqflags |= IRQF_SHARED;
-
 	/* GPIO0 on pre-activate PHY, Reg 1F is not set by reset */
 	iow(db, DM9000_GPR, 0);	/* REG_1F bit0 activate phyxcer */
 	mdelay(1); /* delay needs by DM9000B */
@@ -1323,7 +1316,8 @@ dm9000_open(struct net_device *dev)
 	/* Initialize DM9000 board */
 	dm9000_init_dm9000(dev);
 
-	if (request_irq(dev->irq, dm9000_interrupt, irqflags, dev->name, dev))
+	if (request_irq(dev->irq, dm9000_interrupt, IRQF_SHARED,
+			dev->name, dev))
 		return -EAGAIN;
 	/* Now that we have an interrupt handler hooked up we can unmask
 	 * our interrupts
@@ -1500,15 +1494,22 @@ dm9000_probe(struct platform_device *pdev)
 
 	db->addr_res = platform_get_resource(pdev, IORESOURCE_MEM, 0);
 	db->data_res = platform_get_resource(pdev, IORESOURCE_MEM, 1);
-	db->irq_res  = platform_get_resource(pdev, IORESOURCE_IRQ, 0);
 
-	if (db->addr_res == NULL || db->data_res == NULL ||
-	    db->irq_res == NULL) {
-		dev_err(db->dev, "insufficient resources\n");
+	if (!db->addr_res || !db->data_res) {
+		dev_err(db->dev, "insufficient resources addr=%p data=%p\n",
+			db->addr_res, db->data_res);
 		ret = -ENOENT;
 		goto out;
 	}
 
+	ndev->irq = platform_get_irq(pdev, 0);
+	if (ndev->irq < 0) {
+		dev_err(db->dev, "interrupt resource unavailable: %d\n",
+			ndev->irq);
+		ret = ndev->irq;
+		goto out;
+	}
+
 	db->irq_wake = platform_get_irq(pdev, 1);
 	if (db->irq_wake >= 0) {
 		dev_dbg(db->dev, "wakeup irq %d\n", db->irq_wake);
@@ -1570,7 +1571,6 @@ dm9000_probe(struct platform_device *pdev)
 
 	/* fill in parameters for net-dev structure */
 	ndev->base_addr = (unsigned long)db->io_addr;
-	ndev->irq	= db->irq_res->start;
 
 	/* ensure at least we have a default set of IO routines */
 	dm9000_set_io(db, iosize);
diff --git a/drivers/net/ethernet/emulex/benet/be.h b/drivers/net/ethernet/emulex/benet/be.h
index cf837831304b..f9751294ece7 100644
--- a/drivers/net/ethernet/emulex/benet/be.h
+++ b/drivers/net/ethernet/emulex/benet/be.h
@@ -531,6 +531,7 @@ struct be_adapter {
 
 	struct delayed_work be_err_detection_work;
 	u8 err_flags;
+	bool pcicfg_mapped;	/* pcicfg obtained via pci_iomap() */
 	u32 flags;
 	u32 cmd_privileges;
 	/* Ethtool knobs and info */
diff --git a/drivers/net/ethernet/emulex/benet/be_cmds.h b/drivers/net/ethernet/emulex/benet/be_cmds.h
index 241819b36ca7..6d9a8d78e8ad 100644
--- a/drivers/net/ethernet/emulex/benet/be_cmds.h
+++ b/drivers/net/ethernet/emulex/benet/be_cmds.h
@@ -622,10 +622,13 @@ enum be_if_flags {
 					 BE_IF_FLAGS_VLAN_PROMISCUOUS |\
 					 BE_IF_FLAGS_MCAST_PROMISCUOUS)
 
-#define BE_IF_EN_FLAGS	(BE_IF_FLAGS_BROADCAST | BE_IF_FLAGS_PASS_L3L4_ERRORS |\
-			BE_IF_FLAGS_MULTICAST | BE_IF_FLAGS_UNTAGGED)
+#define BE_IF_FILT_FLAGS_BASIC (BE_IF_FLAGS_BROADCAST | \
+				BE_IF_FLAGS_PASS_L3L4_ERRORS | \
+				BE_IF_FLAGS_UNTAGGED)
 
-#define BE_IF_ALL_FILT_FLAGS	(BE_IF_EN_FLAGS | BE_IF_FLAGS_ALL_PROMISCUOUS)
+#define BE_IF_ALL_FILT_FLAGS	(BE_IF_FILT_FLAGS_BASIC | \
+				 BE_IF_FLAGS_MULTICAST | \
+				 BE_IF_FLAGS_ALL_PROMISCUOUS)
 
 /* An RX interface is an object with one or more MAC addresses and
  * filtering capabilities. */
diff --git a/drivers/net/ethernet/emulex/benet/be_main.c b/drivers/net/ethernet/emulex/benet/be_main.c
index f99de3657ce3..d1cf1274fc2f 100644
--- a/drivers/net/ethernet/emulex/benet/be_main.c
+++ b/drivers/net/ethernet/emulex/benet/be_main.c
@@ -125,6 +125,11 @@ static const char * const ue_status_hi_desc[] = {
 	"Unknown"
 };
 
+#define BE_VF_IF_EN_FLAGS	(BE_IF_FLAGS_UNTAGGED | \
+				 BE_IF_FLAGS_BROADCAST | \
+				 BE_IF_FLAGS_MULTICAST | \
+				 BE_IF_FLAGS_PASS_L3L4_ERRORS)
+
 static void be_queue_free(struct be_adapter *adapter, struct be_queue_info *q)
 {
 	struct be_dma_mem *mem = &q->dma_mem;
@@ -3537,7 +3542,7 @@ static int be_enable_if_filters(struct be_adapter *adapter)
 {
 	int status;
 
-	status = be_cmd_rx_filter(adapter, BE_IF_EN_FLAGS, ON);
+	status = be_cmd_rx_filter(adapter, BE_IF_FILT_FLAGS_BASIC, ON);
 	if (status)
 		return status;
 
@@ -3857,8 +3862,7 @@ static int be_vfs_if_create(struct be_adapter *adapter)
 	int status;
 
 	/* If a FW profile exists, then cap_flags are updated */
-	cap_flags = BE_IF_FLAGS_UNTAGGED | BE_IF_FLAGS_BROADCAST |
-		    BE_IF_FLAGS_MULTICAST | BE_IF_FLAGS_PASS_L3L4_ERRORS;
+	cap_flags = BE_VF_IF_EN_FLAGS;
 
 	for_all_vfs(adapter, vf_cfg, vf) {
 		if (!BE3_chip(adapter)) {
@@ -3874,10 +3878,8 @@ static int be_vfs_if_create(struct be_adapter *adapter)
 			}
 		}
 
-		en_flags = cap_flags & (BE_IF_FLAGS_UNTAGGED |
-					BE_IF_FLAGS_BROADCAST |
-					BE_IF_FLAGS_MULTICAST |
-					BE_IF_FLAGS_PASS_L3L4_ERRORS);
+		/* PF should enable IF flags during proxy if_create call */
+		en_flags = cap_flags & BE_VF_IF_EN_FLAGS;
 		status = be_cmd_if_create(adapter, cap_flags, en_flags,
 					  &vf_cfg->if_handle, vf + 1);
 		if (status)
@@ -4968,6 +4970,8 @@ static void be_unmap_pci_bars(struct be_adapter *adapter)
 		pci_iounmap(adapter->pdev, adapter->csr);
 	if (adapter->db)
 		pci_iounmap(adapter->pdev, adapter->db);
+	if (adapter->pcicfg && adapter->pcicfg_mapped)
+		pci_iounmap(adapter->pdev, adapter->pcicfg);
 }
 
 static int db_bar(struct be_adapter *adapter)
@@ -5019,8 +5023,10 @@ static int be_map_pci_bars(struct be_adapter *adapter)
 			if (!addr)
 				goto pci_map_err;
 			adapter->pcicfg = addr;
+			adapter->pcicfg_mapped = true;
 		} else {
 			adapter->pcicfg = adapter->db + SRIOV_VF_PCICFG_OFFSET;
+			adapter->pcicfg_mapped = false;
 		}
 	}
 
diff --git a/drivers/net/ethernet/ethoc.c b/drivers/net/ethernet/ethoc.c
index 62fa136554ac..41b010645100 100644
--- a/drivers/net/ethernet/ethoc.c
+++ b/drivers/net/ethernet/ethoc.c
@@ -1265,7 +1265,6 @@ static int ethoc_remove(struct platform_device *pdev)
 
 		if (priv->mdio) {
 			mdiobus_unregister(priv->mdio);
-			kfree(priv->mdio->irq);
 			mdiobus_free(priv->mdio);
 		}
 		if (priv->clk)
diff --git a/drivers/net/ethernet/freescale/fman/fman.c b/drivers/net/ethernet/freescale/fman/fman.c
index 623aa1c8ebc6..79a210aaf0bb 100644
--- a/drivers/net/ethernet/freescale/fman/fman.c
+++ b/drivers/net/ethernet/freescale/fman/fman.c
@@ -2791,6 +2791,8 @@ static struct fman *read_dts_node(struct platform_device *of_dev)
 		goto fman_free;
 	}
 
+	fman->dev = &of_dev->dev;
+
 	return fman;
 
 fman_node_put:
@@ -2845,8 +2847,6 @@ static int fman_probe(struct platform_device *of_dev)
 
 	dev_set_drvdata(dev, fman);
 
-	fman->dev = dev;
-
 	dev_dbg(dev, "FMan%d probed\n", fman->dts_params.id);
 
 	return 0;
diff --git a/drivers/net/ethernet/freescale/gianfar.c b/drivers/net/ethernet/freescale/gianfar.c
index 2aa7b401cc3b..b9ecf197ad11 100644
--- a/drivers/net/ethernet/freescale/gianfar.c
+++ b/drivers/net/ethernet/freescale/gianfar.c
@@ -1111,8 +1111,10 @@ static void __gfar_detect_errata_85xx(struct gfar_private *priv)
 
 	if ((SVR_SOC_VER(svr) == SVR_8548) && (SVR_REV(svr) == 0x20))
 		priv->errata |= GFAR_ERRATA_12;
+	/* P2020/P1010 Rev 1; MPC8548 Rev 2 */
 	if (((SVR_SOC_VER(svr) == SVR_P2020) && (SVR_REV(svr) < 0x20)) ||
-	    ((SVR_SOC_VER(svr) == SVR_P2010) && (SVR_REV(svr) < 0x20)))
+	    ((SVR_SOC_VER(svr) == SVR_P2010) && (SVR_REV(svr) < 0x20)) ||
+	    ((SVR_SOC_VER(svr) == SVR_8548) && (SVR_REV(svr) < 0x31)))
 		priv->errata |= GFAR_ERRATA_76; /* aka eTSEC 20 */
 }
 #endif
diff --git a/drivers/net/ethernet/fujitsu/fmvj18x_cs.c b/drivers/net/ethernet/fujitsu/fmvj18x_cs.c
index a7139f588ad2..678f5018d0be 100644
--- a/drivers/net/ethernet/fujitsu/fmvj18x_cs.c
+++ b/drivers/net/ethernet/fujitsu/fmvj18x_cs.c
@@ -469,8 +469,8 @@ static int fmvj18x_config(struct pcmcia_device *link)
 		    goto failed;
 	    }
 	    /* Read MACID from CIS */
-	    for (i = 5; i < 11; i++)
-		    dev->dev_addr[i] = buf[i];
+	    for (i = 0; i < 6; i++)
+		    dev->dev_addr[i] = buf[i + 5];
 	    kfree(buf);
 	} else {
 	    if (pcmcia_get_mac_from_cis(link, dev))
diff --git a/drivers/net/ethernet/hisilicon/Kconfig b/drivers/net/ethernet/hisilicon/Kconfig
index 74beb1867230..4ccc032633c4 100644
--- a/drivers/net/ethernet/hisilicon/Kconfig
+++ b/drivers/net/ethernet/hisilicon/Kconfig
@@ -25,6 +25,7 @@ config HIX5HD2_GMAC
 
 config HIP04_ETH
 	tristate "HISILICON P04 Ethernet support"
+	depends on HAS_IOMEM	# For MFD_SYSCON
 	select MARVELL_PHY
 	select MFD_SYSCON
 	select HNS_MDIO
diff --git a/drivers/net/ethernet/hisilicon/hns/hns_ae_adapt.c b/drivers/net/ethernet/hisilicon/hns/hns_ae_adapt.c
index a0070d0e740d..d4f92ed322d6 100644
--- a/drivers/net/ethernet/hisilicon/hns/hns_ae_adapt.c
+++ b/drivers/net/ethernet/hisilicon/hns/hns_ae_adapt.c
@@ -675,8 +675,12 @@ static int hns_ae_config_loopback(struct hnae_handle *handle,
 {
 	int ret;
 	struct hnae_vf_cb *vf_cb = hns_ae_get_vf_cb(handle);
+	struct hns_mac_cb *mac_cb = hns_get_mac_cb(handle);
 
 	switch (loop) {
+	case MAC_INTERNALLOOP_PHY:
+		ret = 0;
+		break;
 	case MAC_INTERNALLOOP_SERDES:
 		ret = hns_mac_config_sds_loopback(vf_cb->mac_cb, en);
 		break;
@@ -686,6 +690,10 @@ static int hns_ae_config_loopback(struct hnae_handle *handle,
 	default:
 		ret = -EINVAL;
 	}
+
+	if (!ret)
+		hns_dsaf_set_inner_lb(mac_cb->dsaf_dev, mac_cb->mac_id, en);
+
 	return ret;
 }
 
diff --git a/drivers/net/ethernet/hisilicon/hns/hns_dsaf_main.c b/drivers/net/ethernet/hisilicon/hns/hns_dsaf_main.c
index 9439f04962e1..38fc5be3870c 100644
--- a/drivers/net/ethernet/hisilicon/hns/hns_dsaf_main.c
+++ b/drivers/net/ethernet/hisilicon/hns/hns_dsaf_main.c
@@ -230,6 +230,30 @@ static void hns_dsaf_mix_def_qid_cfg(struct dsaf_device *dsaf_dev)
 	}
 }
 
+static void hns_dsaf_inner_qid_cfg(struct dsaf_device *dsaf_dev)
+{
+	u16 max_q_per_vf, max_vfn;
+	u32 q_id, q_num_per_port;
+	u32 mac_id;
+
+	if (AE_IS_VER1(dsaf_dev->dsaf_ver))
+		return;
+
+	hns_rcb_get_queue_mode(dsaf_dev->dsaf_mode,
+			       HNS_DSAF_COMM_SERVICE_NW_IDX,
+			       &max_vfn, &max_q_per_vf);
+	q_num_per_port = max_vfn * max_q_per_vf;
+
+	for (mac_id = 0, q_id = 0; mac_id < DSAF_SERVICE_NW_NUM; mac_id++) {
+		dsaf_set_dev_field(dsaf_dev,
+				   DSAFV2_SERDES_LBK_0_REG + 4 * mac_id,
+				   DSAFV2_SERDES_LBK_QID_M,
+				   DSAFV2_SERDES_LBK_QID_S,
+				   q_id);
+		q_id += q_num_per_port;
+	}
+}
+
 /**
  * hns_dsaf_sw_port_type_cfg - cfg sw type
  * @dsaf_id: dsa fabric id
@@ -691,6 +715,16 @@ void hns_dsaf_set_promisc_mode(struct dsaf_device *dsaf_dev, u32 en)
 	dsaf_set_dev_bit(dsaf_dev, DSAF_CFG_0_REG, DSAF_CFG_MIX_MODE_S, !!en);
 }
 
+void hns_dsaf_set_inner_lb(struct dsaf_device *dsaf_dev, u32 mac_id, u32 en)
+{
+	if (AE_IS_VER1(dsaf_dev->dsaf_ver) ||
+	    dsaf_dev->mac_cb[mac_id].mac_type == HNAE_PORT_DEBUG)
+		return;
+
+	dsaf_set_dev_bit(dsaf_dev, DSAFV2_SERDES_LBK_0_REG + 4 * mac_id,
+			 DSAFV2_SERDES_LBK_EN_B, !!en);
+}
+
 /**
  * hns_dsaf_tbl_stat_en - tbl
  * @dsaf_id: dsa fabric id
@@ -1022,6 +1056,9 @@ static void hns_dsaf_comm_init(struct dsaf_device *dsaf_dev)
 	/* set promisc def queue id */
 	hns_dsaf_mix_def_qid_cfg(dsaf_dev);
 
+	/* set inner loopback queue id */
+	hns_dsaf_inner_qid_cfg(dsaf_dev);
+
 	/* in non switch mode, set all port to access mode */
 	hns_dsaf_sw_port_type_cfg(dsaf_dev, DSAF_SW_PORT_TYPE_NON_VLAN);
 
diff --git a/drivers/net/ethernet/hisilicon/hns/hns_dsaf_main.h b/drivers/net/ethernet/hisilicon/hns/hns_dsaf_main.h
index 40205b910f80..5fea226efaf3 100644
--- a/drivers/net/ethernet/hisilicon/hns/hns_dsaf_main.h
+++ b/drivers/net/ethernet/hisilicon/hns/hns_dsaf_main.h
@@ -417,5 +417,6 @@ void hns_dsaf_get_strings(int stringset, u8 *data, int port);
 void hns_dsaf_get_regs(struct dsaf_device *ddev, u32 port, void *data);
 int hns_dsaf_get_regs_count(void);
 void hns_dsaf_set_promisc_mode(struct dsaf_device *dsaf_dev, u32 en);
+void hns_dsaf_set_inner_lb(struct dsaf_device *dsaf_dev, u32 mac_id, u32 en);
 
 #endif /* __HNS_DSAF_MAIN_H__ */
diff --git a/drivers/net/ethernet/hisilicon/hns/hns_dsaf_reg.h b/drivers/net/ethernet/hisilicon/hns/hns_dsaf_reg.h
index f0c4f9b09d5b..60d695daa471 100644
--- a/drivers/net/ethernet/hisilicon/hns/hns_dsaf_reg.h
+++ b/drivers/net/ethernet/hisilicon/hns/hns_dsaf_reg.h
@@ -134,6 +134,7 @@
 #define DSAF_XGE_INT_STS_0_REG		0x1C0
 #define DSAF_PPE_INT_STS_0_REG		0x1E0
 #define DSAF_ROCEE_INT_STS_0_REG	0x200
+#define DSAFV2_SERDES_LBK_0_REG         0x220
 #define DSAF_PPE_QID_CFG_0_REG		0x300
 #define DSAF_SW_PORT_TYPE_0_REG		0x320
 #define DSAF_STP_PORT_TYPE_0_REG	0x340
@@ -857,6 +858,10 @@
 #define PPEV2_CFG_RSS_TBL_4N3_S	24
 #define PPEV2_CFG_RSS_TBL_4N3_M	(((1UL << 5) - 1) << PPEV2_CFG_RSS_TBL_4N3_S)
 
+#define DSAFV2_SERDES_LBK_EN_B  8
+#define DSAFV2_SERDES_LBK_QID_S 0
+#define DSAFV2_SERDES_LBK_QID_M	(((1UL << 8) - 1) << DSAFV2_SERDES_LBK_QID_S)
+
 #define PPE_CNT_CLR_CE_B	0
 #define PPE_CNT_CLR_SNAP_EN_B	1
 
diff --git a/drivers/net/ethernet/hisilicon/hns/hns_ethtool.c b/drivers/net/ethernet/hisilicon/hns/hns_ethtool.c
index 3df22840fcd1..3c4a3bc31a89 100644
--- a/drivers/net/ethernet/hisilicon/hns/hns_ethtool.c
+++ b/drivers/net/ethernet/hisilicon/hns/hns_ethtool.c
@@ -295,8 +295,10 @@ static int __lb_setup(struct net_device *ndev,
 
 	switch (loop) {
 	case MAC_INTERNALLOOP_PHY:
-		if ((phy_dev) && (!phy_dev->is_c45))
+		if ((phy_dev) && (!phy_dev->is_c45)) {
 			ret = hns_nic_config_phy_loopback(phy_dev, 0x1);
+			ret |= h->dev->ops->set_loopback(h, loop, 0x1);
+		}
 		break;
 	case MAC_INTERNALLOOP_MAC:
 		if ((h->dev->ops->set_loopback) &&
@@ -376,6 +378,7 @@ static void __lb_other_process(struct hns_nic_ring_data *ring_data,
 			       struct sk_buff *skb)
 {
 	struct net_device *ndev;
+	struct hns_nic_priv *priv;
 	struct hnae_ring *ring;
 	struct netdev_queue *dev_queue;
 	struct sk_buff *new_skb;
@@ -385,8 +388,17 @@ static void __lb_other_process(struct hns_nic_ring_data *ring_data,
 	char buff[33]; /* 32B data and the last character '\0' */
 
 	if (!ring_data) { /* Just for doing create frame*/
+		ndev = skb->dev;
+		priv = netdev_priv(ndev);
+
 		frame_size = skb->len;
 		memset(skb->data, 0xFF, frame_size);
+		if ((!AE_IS_VER1(priv->enet_ver)) &&
+		    (priv->ae_handle->port_type == HNAE_PORT_SERVICE)) {
+			memcpy(skb->data, ndev->dev_addr, 6);
+			skb->data[5] += 0x1f;
+		}
+
 		frame_size &= ~1ul;
 		memset(&skb->data[frame_size / 2], 0xAA, frame_size / 2 - 1);
 		memset(&skb->data[frame_size / 2 + 10], 0xBE,
@@ -486,6 +498,7 @@ static int __lb_run_test(struct net_device *ndev,
 
 	/* place data into test skb */
 	(void)skb_put(skb, size);
+	skb->dev = ndev;
 	__lb_other_process(NULL, skb);
 	skb->queue_mapping = NIC_LB_TEST_RING_ID;
 
diff --git a/drivers/net/ethernet/ibm/ibmveth.c b/drivers/net/ethernet/ibm/ibmveth.c
index 335417b4756b..ebe60719e489 100644
--- a/drivers/net/ethernet/ibm/ibmveth.c
+++ b/drivers/net/ethernet/ibm/ibmveth.c
@@ -1166,7 +1166,10 @@ map_failed:
 	if (!firmware_has_feature(FW_FEATURE_CMO))
 		netdev_err(netdev, "tx: unable to map xmit buffer\n");
 	adapter->tx_map_failed++;
-	skb_linearize(skb);
+	if (skb_linearize(skb)) {
+		netdev->stats.tx_dropped++;
+		goto out;
+	}
 	force_bounce = 1;
 	goto retry_bounce;
 }
diff --git a/drivers/net/ethernet/ibm/ibmvnic.c b/drivers/net/ethernet/ibm/ibmvnic.c
index 7d6570843723..6e9e16eee5d0 100644
--- a/drivers/net/ethernet/ibm/ibmvnic.c
+++ b/drivers/net/ethernet/ibm/ibmvnic.c
@@ -1348,44 +1348,44 @@ static void init_sub_crqs(struct ibmvnic_adapter *adapter, int retry)
 	crq.request_capability.cmd = REQUEST_CAPABILITY;
 
 	crq.request_capability.capability = cpu_to_be16(REQ_TX_QUEUES);
-	crq.request_capability.number = cpu_to_be32(adapter->req_tx_queues);
+	crq.request_capability.number = cpu_to_be64(adapter->req_tx_queues);
 	ibmvnic_send_crq(adapter, &crq);
 
 	crq.request_capability.capability = cpu_to_be16(REQ_RX_QUEUES);
-	crq.request_capability.number = cpu_to_be32(adapter->req_rx_queues);
+	crq.request_capability.number = cpu_to_be64(adapter->req_rx_queues);
 	ibmvnic_send_crq(adapter, &crq);
 
 	crq.request_capability.capability = cpu_to_be16(REQ_RX_ADD_QUEUES);
-	crq.request_capability.number = cpu_to_be32(adapter->req_rx_add_queues);
+	crq.request_capability.number = cpu_to_be64(adapter->req_rx_add_queues);
 	ibmvnic_send_crq(adapter, &crq);
 
 	crq.request_capability.capability =
 	    cpu_to_be16(REQ_TX_ENTRIES_PER_SUBCRQ);
 	crq.request_capability.number =
-	    cpu_to_be32(adapter->req_tx_entries_per_subcrq);
+	    cpu_to_be64(adapter->req_tx_entries_per_subcrq);
 	ibmvnic_send_crq(adapter, &crq);
 
 	crq.request_capability.capability =
 	    cpu_to_be16(REQ_RX_ADD_ENTRIES_PER_SUBCRQ);
 	crq.request_capability.number =
-	    cpu_to_be32(adapter->req_rx_add_entries_per_subcrq);
+	    cpu_to_be64(adapter->req_rx_add_entries_per_subcrq);
 	ibmvnic_send_crq(adapter, &crq);
 
 	crq.request_capability.capability = cpu_to_be16(REQ_MTU);
-	crq.request_capability.number = cpu_to_be32(adapter->req_mtu);
+	crq.request_capability.number = cpu_to_be64(adapter->req_mtu);
 	ibmvnic_send_crq(adapter, &crq);
 
 	if (adapter->netdev->flags & IFF_PROMISC) {
 		if (adapter->promisc_supported) {
 			crq.request_capability.capability =
 			    cpu_to_be16(PROMISC_REQUESTED);
-			crq.request_capability.number = cpu_to_be32(1);
+			crq.request_capability.number = cpu_to_be64(1);
 			ibmvnic_send_crq(adapter, &crq);
 		}
 	} else {
 		crq.request_capability.capability =
 		    cpu_to_be16(PROMISC_REQUESTED);
-		crq.request_capability.number = cpu_to_be32(0);
+		crq.request_capability.number = cpu_to_be64(0);
 		ibmvnic_send_crq(adapter, &crq);
 	}
 
@@ -2312,93 +2312,93 @@ static void handle_query_cap_rsp(union ibmvnic_crq *crq,
 	switch (be16_to_cpu(crq->query_capability.capability)) {
 	case MIN_TX_QUEUES:
 		adapter->min_tx_queues =
-		    be32_to_cpu(crq->query_capability.number);
+		    be64_to_cpu(crq->query_capability.number);
 		netdev_dbg(netdev, "min_tx_queues = %lld\n",
 			   adapter->min_tx_queues);
 		break;
 	case MIN_RX_QUEUES:
 		adapter->min_rx_queues =
-		    be32_to_cpu(crq->query_capability.number);
+		    be64_to_cpu(crq->query_capability.number);
 		netdev_dbg(netdev, "min_rx_queues = %lld\n",
 			   adapter->min_rx_queues);
 		break;
 	case MIN_RX_ADD_QUEUES:
 		adapter->min_rx_add_queues =
-		    be32_to_cpu(crq->query_capability.number);
+		    be64_to_cpu(crq->query_capability.number);
 		netdev_dbg(netdev, "min_rx_add_queues = %lld\n",
 			   adapter->min_rx_add_queues);
 		break;
 	case MAX_TX_QUEUES:
 		adapter->max_tx_queues =
-		    be32_to_cpu(crq->query_capability.number);
+		    be64_to_cpu(crq->query_capability.number);
 		netdev_dbg(netdev, "max_tx_queues = %lld\n",
 			   adapter->max_tx_queues);
 		break;
 	case MAX_RX_QUEUES:
 		adapter->max_rx_queues =
-		    be32_to_cpu(crq->query_capability.number);
+		    be64_to_cpu(crq->query_capability.number);
 		netdev_dbg(netdev, "max_rx_queues = %lld\n",
 			   adapter->max_rx_queues);
 		break;
 	case MAX_RX_ADD_QUEUES:
 		adapter->max_rx_add_queues =
-		    be32_to_cpu(crq->query_capability.number);
+		    be64_to_cpu(crq->query_capability.number);
 		netdev_dbg(netdev, "max_rx_add_queues = %lld\n",
 			   adapter->max_rx_add_queues);
 		break;
 	case MIN_TX_ENTRIES_PER_SUBCRQ:
 		adapter->min_tx_entries_per_subcrq =
-		    be32_to_cpu(crq->query_capability.number);
+		    be64_to_cpu(crq->query_capability.number);
 		netdev_dbg(netdev, "min_tx_entries_per_subcrq = %lld\n",
 			   adapter->min_tx_entries_per_subcrq);
 		break;
 	case MIN_RX_ADD_ENTRIES_PER_SUBCRQ:
 		adapter->min_rx_add_entries_per_subcrq =
-		    be32_to_cpu(crq->query_capability.number);
+		    be64_to_cpu(crq->query_capability.number);
 		netdev_dbg(netdev, "min_rx_add_entrs_per_subcrq = %lld\n",
 			   adapter->min_rx_add_entries_per_subcrq);
 		break;
 	case MAX_TX_ENTRIES_PER_SUBCRQ:
 		adapter->max_tx_entries_per_subcrq =
-		    be32_to_cpu(crq->query_capability.number);
+		    be64_to_cpu(crq->query_capability.number);
 		netdev_dbg(netdev, "max_tx_entries_per_subcrq = %lld\n",
 			   adapter->max_tx_entries_per_subcrq);
 		break;
 	case MAX_RX_ADD_ENTRIES_PER_SUBCRQ:
 		adapter->max_rx_add_entries_per_subcrq =
-		    be32_to_cpu(crq->query_capability.number);
+		    be64_to_cpu(crq->query_capability.number);
 		netdev_dbg(netdev, "max_rx_add_entrs_per_subcrq = %lld\n",
 			   adapter->max_rx_add_entries_per_subcrq);
 		break;
 	case TCP_IP_OFFLOAD:
 		adapter->tcp_ip_offload =
-		    be32_to_cpu(crq->query_capability.number);
+		    be64_to_cpu(crq->query_capability.number);
 		netdev_dbg(netdev, "tcp_ip_offload = %lld\n",
 			   adapter->tcp_ip_offload);
 		break;
 	case PROMISC_SUPPORTED:
 		adapter->promisc_supported =
-		    be32_to_cpu(crq->query_capability.number);
+		    be64_to_cpu(crq->query_capability.number);
 		netdev_dbg(netdev, "promisc_supported = %lld\n",
 			   adapter->promisc_supported);
 		break;
 	case MIN_MTU:
-		adapter->min_mtu = be32_to_cpu(crq->query_capability.number);
+		adapter->min_mtu = be64_to_cpu(crq->query_capability.number);
 		netdev_dbg(netdev, "min_mtu = %lld\n", adapter->min_mtu);
 		break;
 	case MAX_MTU:
-		adapter->max_mtu = be32_to_cpu(crq->query_capability.number);
+		adapter->max_mtu = be64_to_cpu(crq->query_capability.number);
 		netdev_dbg(netdev, "max_mtu = %lld\n", adapter->max_mtu);
 		break;
 	case MAX_MULTICAST_FILTERS:
 		adapter->max_multicast_filters =
-		    be32_to_cpu(crq->query_capability.number);
+		    be64_to_cpu(crq->query_capability.number);
 		netdev_dbg(netdev, "max_multicast_filters = %lld\n",
 			   adapter->max_multicast_filters);
 		break;
 	case VLAN_HEADER_INSERTION:
 		adapter->vlan_header_insertion =
-		    be32_to_cpu(crq->query_capability.number);
+		    be64_to_cpu(crq->query_capability.number);
 		if (adapter->vlan_header_insertion)
 			netdev->features |= NETIF_F_HW_VLAN_STAG_TX;
 		netdev_dbg(netdev, "vlan_header_insertion = %lld\n",
@@ -2406,43 +2406,43 @@ static void handle_query_cap_rsp(union ibmvnic_crq *crq,
 		break;
 	case MAX_TX_SG_ENTRIES:
 		adapter->max_tx_sg_entries =
-		    be32_to_cpu(crq->query_capability.number);
+		    be64_to_cpu(crq->query_capability.number);
 		netdev_dbg(netdev, "max_tx_sg_entries = %lld\n",
 			   adapter->max_tx_sg_entries);
 		break;
 	case RX_SG_SUPPORTED:
 		adapter->rx_sg_supported =
-		    be32_to_cpu(crq->query_capability.number);
+		    be64_to_cpu(crq->query_capability.number);
 		netdev_dbg(netdev, "rx_sg_supported = %lld\n",
 			   adapter->rx_sg_supported);
 		break;
 	case OPT_TX_COMP_SUB_QUEUES:
 		adapter->opt_tx_comp_sub_queues =
-		    be32_to_cpu(crq->query_capability.number);
+		    be64_to_cpu(crq->query_capability.number);
 		netdev_dbg(netdev, "opt_tx_comp_sub_queues = %lld\n",
 			   adapter->opt_tx_comp_sub_queues);
 		break;
 	case OPT_RX_COMP_QUEUES:
 		adapter->opt_rx_comp_queues =
-		    be32_to_cpu(crq->query_capability.number);
+		    be64_to_cpu(crq->query_capability.number);
 		netdev_dbg(netdev, "opt_rx_comp_queues = %lld\n",
 			   adapter->opt_rx_comp_queues);
 		break;
 	case OPT_RX_BUFADD_Q_PER_RX_COMP_Q:
 		adapter->opt_rx_bufadd_q_per_rx_comp_q =
-		    be32_to_cpu(crq->query_capability.number);
+		    be64_to_cpu(crq->query_capability.number);
 		netdev_dbg(netdev, "opt_rx_bufadd_q_per_rx_comp_q = %lld\n",
 			   adapter->opt_rx_bufadd_q_per_rx_comp_q);
 		break;
 	case OPT_TX_ENTRIES_PER_SUBCRQ:
 		adapter->opt_tx_entries_per_subcrq =
-		    be32_to_cpu(crq->query_capability.number);
+		    be64_to_cpu(crq->query_capability.number);
 		netdev_dbg(netdev, "opt_tx_entries_per_subcrq = %lld\n",
 			   adapter->opt_tx_entries_per_subcrq);
 		break;
 	case OPT_RXBA_ENTRIES_PER_SUBCRQ:
 		adapter->opt_rxba_entries_per_subcrq =
-		    be32_to_cpu(crq->query_capability.number);
+		    be64_to_cpu(crq->query_capability.number);
 		netdev_dbg(netdev, "opt_rxba_entries_per_subcrq = %lld\n",
 			   adapter->opt_rxba_entries_per_subcrq);
 		break;
diff --git a/drivers/net/ethernet/ibm/ibmvnic.h b/drivers/net/ethernet/ibm/ibmvnic.h
index 1242925ad34c..1a9993cc79b5 100644
--- a/drivers/net/ethernet/ibm/ibmvnic.h
+++ b/drivers/net/ethernet/ibm/ibmvnic.h
@@ -319,10 +319,8 @@ struct ibmvnic_capability {
 	u8 first;
 	u8 cmd;
 	__be16 capability; /* one of ibmvnic_capabilities */
+	__be64 number;
 	struct ibmvnic_rc rc;
-	__be32 number; /*FIX: should be __be64, but I'm getting the least
-			* significant word first
-			*/
 } __packed __aligned(8);
 
 struct ibmvnic_login {
diff --git a/drivers/net/ethernet/jme.c b/drivers/net/ethernet/jme.c
index b1de7afd4116..3ddf657bc10b 100644
--- a/drivers/net/ethernet/jme.c
+++ b/drivers/net/ethernet/jme.c
@@ -270,11 +270,17 @@ jme_reset_mac_processor(struct jme_adapter *jme)
 }
 
 static inline void
-jme_clear_pm(struct jme_adapter *jme)
+jme_clear_pm_enable_wol(struct jme_adapter *jme)
 {
 	jwrite32(jme, JME_PMCS, PMCS_STMASK | jme->reg_pmcs);
 }
 
+static inline void
+jme_clear_pm_disable_wol(struct jme_adapter *jme)
+{
+	jwrite32(jme, JME_PMCS, PMCS_STMASK);
+}
+
 static int
 jme_reload_eeprom(struct jme_adapter *jme)
 {
@@ -1853,7 +1859,7 @@ jme_open(struct net_device *netdev)
 	struct jme_adapter *jme = netdev_priv(netdev);
 	int rc;
 
-	jme_clear_pm(jme);
+	jme_clear_pm_disable_wol(jme);
 	JME_NAPI_ENABLE(jme);
 
 	tasklet_init(&jme->linkch_task, jme_link_change_tasklet,
@@ -1925,11 +1931,11 @@ jme_wait_link(struct jme_adapter *jme)
 static void
 jme_powersave_phy(struct jme_adapter *jme)
 {
-	if (jme->reg_pmcs) {
+	if (jme->reg_pmcs && device_may_wakeup(&jme->pdev->dev)) {
 		jme_set_100m_half(jme);
 		if (jme->reg_pmcs & (PMCS_LFEN | PMCS_LREN))
 			jme_wait_link(jme);
-		jme_clear_pm(jme);
+		jme_clear_pm_enable_wol(jme);
 	} else {
 		jme_phy_off(jme);
 	}
@@ -2646,9 +2652,6 @@ jme_set_wol(struct net_device *netdev,
 	if (wol->wolopts & WAKE_MAGIC)
 		jme->reg_pmcs |= PMCS_MFEN;
 
-	jwrite32(jme, JME_PMCS, jme->reg_pmcs);
-	device_set_wakeup_enable(&jme->pdev->dev, !!(jme->reg_pmcs));
-
 	return 0;
 }
 
@@ -3172,8 +3175,8 @@ jme_init_one(struct pci_dev *pdev,
 	jme->mii_if.mdio_read = jme_mdio_read;
 	jme->mii_if.mdio_write = jme_mdio_write;
 
-	jme_clear_pm(jme);
-	device_set_wakeup_enable(&pdev->dev, true);
+	jme_clear_pm_disable_wol(jme);
+	device_init_wakeup(&pdev->dev, true);
 
 	jme_set_phyfifo_5level(jme);
 	jme->pcirev = pdev->revision;
@@ -3304,7 +3307,7 @@ jme_resume(struct device *dev)
 	if (!netif_running(netdev))
 		return 0;
 
-	jme_clear_pm(jme);
+	jme_clear_pm_disable_wol(jme);
 	jme_phy_on(jme);
 	if (test_bit(JME_FLAG_SSET, &jme->flags))
 		jme_set_settings(netdev, &jme->old_ecmd);
@@ -3312,13 +3315,14 @@ jme_resume(struct device *dev)
 		jme_reset_phy_processor(jme);
 	jme_phy_calibration(jme);
 	jme_phy_setEA(jme);
-	jme_start_irq(jme);
 	netif_device_attach(netdev);
 
 	atomic_inc(&jme->link_changing);
 
 	jme_reset_link(jme);
 
+	jme_start_irq(jme);
+
 	return 0;
 }
 
diff --git a/drivers/net/ethernet/marvell/mvneta.c b/drivers/net/ethernet/marvell/mvneta.c
index 662c2ee268c7..b0ae69f84493 100644
--- a/drivers/net/ethernet/marvell/mvneta.c
+++ b/drivers/net/ethernet/marvell/mvneta.c
@@ -370,6 +370,11 @@ struct mvneta_port {
 	struct net_device *dev;
 	struct notifier_block cpu_notifier;
 	int rxq_def;
+	/* Protect the access to the percpu interrupt registers,
+	 * ensuring that the configuration remains coherent.
+	 */
+	spinlock_t lock;
+	bool is_stopped;
 
 	/* Core clock */
 	struct clk *clk;
@@ -1038,6 +1043,43 @@ static void mvneta_set_autoneg(struct mvneta_port *pp, int enable)
 	}
 }
 
+static void mvneta_percpu_unmask_interrupt(void *arg)
+{
+	struct mvneta_port *pp = arg;
+
+	/* All the queue are unmasked, but actually only the ones
+	 * mapped to this CPU will be unmasked
+	 */
+	mvreg_write(pp, MVNETA_INTR_NEW_MASK,
+		    MVNETA_RX_INTR_MASK_ALL |
+		    MVNETA_TX_INTR_MASK_ALL |
+		    MVNETA_MISCINTR_INTR_MASK);
+}
+
+static void mvneta_percpu_mask_interrupt(void *arg)
+{
+	struct mvneta_port *pp = arg;
+
+	/* All the queue are masked, but actually only the ones
+	 * mapped to this CPU will be masked
+	 */
+	mvreg_write(pp, MVNETA_INTR_NEW_MASK, 0);
+	mvreg_write(pp, MVNETA_INTR_OLD_MASK, 0);
+	mvreg_write(pp, MVNETA_INTR_MISC_MASK, 0);
+}
+
+static void mvneta_percpu_clear_intr_cause(void *arg)
+{
+	struct mvneta_port *pp = arg;
+
+	/* All the queue are cleared, but actually only the ones
+	 * mapped to this CPU will be cleared
+	 */
+	mvreg_write(pp, MVNETA_INTR_NEW_CAUSE, 0);
+	mvreg_write(pp, MVNETA_INTR_MISC_CAUSE, 0);
+	mvreg_write(pp, MVNETA_INTR_OLD_CAUSE, 0);
+}
+
 /* This method sets defaults to the NETA port:
  *	Clears interrupt Cause and Mask registers.
  *	Clears all MAC tables.
@@ -1055,14 +1097,10 @@ static void mvneta_defaults_set(struct mvneta_port *pp)
 	int max_cpu = num_present_cpus();
 
 	/* Clear all Cause registers */
-	mvreg_write(pp, MVNETA_INTR_NEW_CAUSE, 0);
-	mvreg_write(pp, MVNETA_INTR_OLD_CAUSE, 0);
-	mvreg_write(pp, MVNETA_INTR_MISC_CAUSE, 0);
+	on_each_cpu(mvneta_percpu_clear_intr_cause, pp, true);
 
 	/* Mask all interrupts */
-	mvreg_write(pp, MVNETA_INTR_NEW_MASK, 0);
-	mvreg_write(pp, MVNETA_INTR_OLD_MASK, 0);
-	mvreg_write(pp, MVNETA_INTR_MISC_MASK, 0);
+	on_each_cpu(mvneta_percpu_mask_interrupt, pp, true);
 	mvreg_write(pp, MVNETA_INTR_ENABLE, 0);
 
 	/* Enable MBUS Retry bit16 */
@@ -2528,34 +2566,9 @@ static int mvneta_setup_txqs(struct mvneta_port *pp)
 	return 0;
 }
 
-static void mvneta_percpu_unmask_interrupt(void *arg)
-{
-	struct mvneta_port *pp = arg;
-
-	/* All the queue are unmasked, but actually only the ones
-	 * maped to this CPU will be unmasked
-	 */
-	mvreg_write(pp, MVNETA_INTR_NEW_MASK,
-		    MVNETA_RX_INTR_MASK_ALL |
-		    MVNETA_TX_INTR_MASK_ALL |
-		    MVNETA_MISCINTR_INTR_MASK);
-}
-
-static void mvneta_percpu_mask_interrupt(void *arg)
-{
-	struct mvneta_port *pp = arg;
-
-	/* All the queue are masked, but actually only the ones
-	 * maped to this CPU will be masked
-	 */
-	mvreg_write(pp, MVNETA_INTR_NEW_MASK, 0);
-	mvreg_write(pp, MVNETA_INTR_OLD_MASK, 0);
-	mvreg_write(pp, MVNETA_INTR_MISC_MASK, 0);
-}
-
 static void mvneta_start_dev(struct mvneta_port *pp)
 {
-	unsigned int cpu;
+	int cpu;
 
 	mvneta_max_rx_size_set(pp, pp->pkt_size);
 	mvneta_txq_max_tx_size_set(pp, pp->pkt_size);
@@ -2564,16 +2577,15 @@ static void mvneta_start_dev(struct mvneta_port *pp)
 	mvneta_port_enable(pp);
 
 	/* Enable polling on the port */
-	for_each_present_cpu(cpu) {
+	for_each_online_cpu(cpu) {
 		struct mvneta_pcpu_port *port = per_cpu_ptr(pp->ports, cpu);
 
 		napi_enable(&port->napi);
 	}
 
 	/* Unmask interrupts. It has to be done from each CPU */
-	for_each_online_cpu(cpu)
-		smp_call_function_single(cpu, mvneta_percpu_unmask_interrupt,
-					 pp, true);
+	on_each_cpu(mvneta_percpu_unmask_interrupt, pp, true);
+
 	mvreg_write(pp, MVNETA_INTR_MISC_MASK,
 		    MVNETA_CAUSE_PHY_STATUS_CHANGE |
 		    MVNETA_CAUSE_LINK_CHANGE |
@@ -2589,7 +2601,7 @@ static void mvneta_stop_dev(struct mvneta_port *pp)
 
 	phy_stop(pp->phy_dev);
 
-	for_each_present_cpu(cpu) {
+	for_each_online_cpu(cpu) {
 		struct mvneta_pcpu_port *port = per_cpu_ptr(pp->ports, cpu);
 
 		napi_disable(&port->napi);
@@ -2604,13 +2616,10 @@ static void mvneta_stop_dev(struct mvneta_port *pp)
 	mvneta_port_disable(pp);
 
 	/* Clear all ethernet port interrupts */
-	mvreg_write(pp, MVNETA_INTR_MISC_CAUSE, 0);
-	mvreg_write(pp, MVNETA_INTR_OLD_CAUSE, 0);
+	on_each_cpu(mvneta_percpu_clear_intr_cause, pp, true);
 
 	/* Mask all ethernet port interrupts */
-	mvreg_write(pp, MVNETA_INTR_NEW_MASK, 0);
-	mvreg_write(pp, MVNETA_INTR_OLD_MASK, 0);
-	mvreg_write(pp, MVNETA_INTR_MISC_MASK, 0);
+	on_each_cpu(mvneta_percpu_mask_interrupt, pp, true);
 
 	mvneta_tx_reset(pp);
 	mvneta_rx_reset(pp);
@@ -2847,11 +2856,20 @@ static void mvneta_percpu_disable(void *arg)
 	disable_percpu_irq(pp->dev->irq);
 }
 
+/* Electing a CPU must be done in an atomic way: it should be done
+ * after or before the removal/insertion of a CPU and this function is
+ * not reentrant.
+ */
 static void mvneta_percpu_elect(struct mvneta_port *pp)
 {
-	int online_cpu_idx, max_cpu, cpu, i = 0;
+	int elected_cpu = 0, max_cpu, cpu, i = 0;
+
+	/* Use the cpu associated to the rxq when it is online, in all
+	 * the other cases, use the cpu 0 which can't be offline.
+	 */
+	if (cpu_online(pp->rxq_def))
+		elected_cpu = pp->rxq_def;
 
-	online_cpu_idx = pp->rxq_def % num_online_cpus();
 	max_cpu = num_present_cpus();
 
 	for_each_online_cpu(cpu) {
@@ -2862,7 +2880,7 @@ static void mvneta_percpu_elect(struct mvneta_port *pp)
 			if ((rxq % max_cpu) == cpu)
 				rxq_map |= MVNETA_CPU_RXQ_ACCESS(rxq);
 
-		if (i == online_cpu_idx)
+		if (cpu == elected_cpu)
 			/* Map the default receive queue queue to the
 			 * elected CPU
 			 */
@@ -2873,7 +2891,7 @@ static void mvneta_percpu_elect(struct mvneta_port *pp)
 		 * the CPU bound to the default RX queue
 		 */
 		if (txq_number == 1)
-			txq_map = (i == online_cpu_idx) ?
+			txq_map = (cpu == elected_cpu) ?
 				MVNETA_CPU_TXQ_ACCESS(1) : 0;
 		else
 			txq_map = mvreg_read(pp, MVNETA_CPU_MAP(cpu)) &
@@ -2902,6 +2920,14 @@ static int mvneta_percpu_notifier(struct notifier_block *nfb,
 	switch (action) {
 	case CPU_ONLINE:
 	case CPU_ONLINE_FROZEN:
+		spin_lock(&pp->lock);
+		/* Configuring the driver for a new CPU while the
+		 * driver is stopping is racy, so just avoid it.
+		 */
+		if (pp->is_stopped) {
+			spin_unlock(&pp->lock);
+			break;
+		}
 		netif_tx_stop_all_queues(pp->dev);
 
 		/* We have to synchronise on tha napi of each CPU
@@ -2917,9 +2943,7 @@ static int mvneta_percpu_notifier(struct notifier_block *nfb,
 		}
 
 		/* Mask all ethernet port interrupts */
-		mvreg_write(pp, MVNETA_INTR_NEW_MASK, 0);
-		mvreg_write(pp, MVNETA_INTR_OLD_MASK, 0);
-		mvreg_write(pp, MVNETA_INTR_MISC_MASK, 0);
+		on_each_cpu(mvneta_percpu_mask_interrupt, pp, true);
 		napi_enable(&port->napi);
 
 
@@ -2934,27 +2958,25 @@ static int mvneta_percpu_notifier(struct notifier_block *nfb,
 		 */
 		mvneta_percpu_elect(pp);
 
-		/* Unmask all ethernet port interrupts, as this
-		 * notifier is called for each CPU then the CPU to
-		 * Queue mapping is applied
-		 */
-		mvreg_write(pp, MVNETA_INTR_NEW_MASK,
-			MVNETA_RX_INTR_MASK(rxq_number) |
-			MVNETA_TX_INTR_MASK(txq_number) |
-			MVNETA_MISCINTR_INTR_MASK);
+		/* Unmask all ethernet port interrupts */
+		on_each_cpu(mvneta_percpu_unmask_interrupt, pp, true);
 		mvreg_write(pp, MVNETA_INTR_MISC_MASK,
 			MVNETA_CAUSE_PHY_STATUS_CHANGE |
 			MVNETA_CAUSE_LINK_CHANGE |
 			MVNETA_CAUSE_PSC_SYNC_CHANGE);
 		netif_tx_start_all_queues(pp->dev);
+		spin_unlock(&pp->lock);
 		break;
 	case CPU_DOWN_PREPARE:
 	case CPU_DOWN_PREPARE_FROZEN:
 		netif_tx_stop_all_queues(pp->dev);
+		/* Thanks to this lock we are sure that any pending
+		 * cpu election is done
+		 */
+		spin_lock(&pp->lock);
 		/* Mask all ethernet port interrupts */
-		mvreg_write(pp, MVNETA_INTR_NEW_MASK, 0);
-		mvreg_write(pp, MVNETA_INTR_OLD_MASK, 0);
-		mvreg_write(pp, MVNETA_INTR_MISC_MASK, 0);
+		on_each_cpu(mvneta_percpu_mask_interrupt, pp, true);
+		spin_unlock(&pp->lock);
 
 		napi_synchronize(&port->napi);
 		napi_disable(&port->napi);
@@ -2968,12 +2990,11 @@ static int mvneta_percpu_notifier(struct notifier_block *nfb,
 	case CPU_DEAD:
 	case CPU_DEAD_FROZEN:
 		/* Check if a new CPU must be elected now this on is down */
+		spin_lock(&pp->lock);
 		mvneta_percpu_elect(pp);
+		spin_unlock(&pp->lock);
 		/* Unmask all ethernet port interrupts */
-		mvreg_write(pp, MVNETA_INTR_NEW_MASK,
-			MVNETA_RX_INTR_MASK(rxq_number) |
-			MVNETA_TX_INTR_MASK(txq_number) |
-			MVNETA_MISCINTR_INTR_MASK);
+		on_each_cpu(mvneta_percpu_unmask_interrupt, pp, true);
 		mvreg_write(pp, MVNETA_INTR_MISC_MASK,
 			MVNETA_CAUSE_PHY_STATUS_CHANGE |
 			MVNETA_CAUSE_LINK_CHANGE |
@@ -2988,7 +3009,7 @@ static int mvneta_percpu_notifier(struct notifier_block *nfb,
 static int mvneta_open(struct net_device *dev)
 {
 	struct mvneta_port *pp = netdev_priv(dev);
-	int ret, cpu;
+	int ret;
 
 	pp->pkt_size = MVNETA_RX_PKT_SIZE(pp->dev->mtu);
 	pp->frag_size = SKB_DATA_ALIGN(MVNETA_RX_BUF_SIZE(pp->pkt_size)) +
@@ -3010,22 +3031,12 @@ static int mvneta_open(struct net_device *dev)
 		goto err_cleanup_txqs;
 	}
 
-	/* Even though the documentation says that request_percpu_irq
-	 * doesn't enable the interrupts automatically, it actually
-	 * does so on the local CPU.
-	 *
-	 * Make sure it's disabled.
-	 */
-	mvneta_percpu_disable(pp);
-
 	/* Enable per-CPU interrupt on all the CPU to handle our RX
 	 * queue interrupts
 	 */
-	for_each_online_cpu(cpu)
-		smp_call_function_single(cpu, mvneta_percpu_enable,
-					 pp, true);
-
+	on_each_cpu(mvneta_percpu_enable, pp, true);
 
+	pp->is_stopped = false;
 	/* Register a CPU notifier to handle the case where our CPU
 	 * might be taken offline.
 	 */
@@ -3057,13 +3068,20 @@ err_cleanup_rxqs:
 static int mvneta_stop(struct net_device *dev)
 {
 	struct mvneta_port *pp = netdev_priv(dev);
-	int cpu;
 
+	/* Inform that we are stopping so we don't want to setup the
+	 * driver for new CPUs in the notifiers
+	 */
+	spin_lock(&pp->lock);
+	pp->is_stopped = true;
 	mvneta_stop_dev(pp);
 	mvneta_mdio_remove(pp);
 	unregister_cpu_notifier(&pp->cpu_notifier);
-	for_each_present_cpu(cpu)
-		smp_call_function_single(cpu, mvneta_percpu_disable, pp, true);
+	/* Now that the notifier are unregistered, we can release le
+	 * lock
+	 */
+	spin_unlock(&pp->lock);
+	on_each_cpu(mvneta_percpu_disable, pp, true);
 	free_percpu_irq(dev->irq, pp->ports);
 	mvneta_cleanup_rxqs(pp);
 	mvneta_cleanup_txqs(pp);
@@ -3312,9 +3330,7 @@ static int  mvneta_config_rss(struct mvneta_port *pp)
 
 	netif_tx_stop_all_queues(pp->dev);
 
-	for_each_online_cpu(cpu)
-		smp_call_function_single(cpu, mvneta_percpu_mask_interrupt,
-					 pp, true);
+	on_each_cpu(mvneta_percpu_mask_interrupt, pp, true);
 
 	/* We have to synchronise on the napi of each CPU */
 	for_each_online_cpu(cpu) {
@@ -3335,7 +3351,9 @@ static int  mvneta_config_rss(struct mvneta_port *pp)
 	mvreg_write(pp, MVNETA_PORT_CONFIG, val);
 
 	/* Update the elected CPU matching the new rxq_def */
+	spin_lock(&pp->lock);
 	mvneta_percpu_elect(pp);
+	spin_unlock(&pp->lock);
 
 	/* We have to synchronise on the napi of each CPU */
 	for_each_online_cpu(cpu) {
diff --git a/drivers/net/ethernet/marvell/mvpp2.c b/drivers/net/ethernet/marvell/mvpp2.c
index a4beccf1fd46..c797971aefab 100644
--- a/drivers/net/ethernet/marvell/mvpp2.c
+++ b/drivers/net/ethernet/marvell/mvpp2.c
@@ -3061,7 +3061,7 @@ static int mvpp2_prs_mac_da_accept(struct mvpp2 *priv, int port,
 
 		pe = kzalloc(sizeof(*pe), GFP_KERNEL);
 		if (!pe)
-			return -1;
+			return -ENOMEM;
 		mvpp2_prs_tcam_lu_set(pe, MVPP2_PRS_LU_MAC);
 		pe->index = tid;
 
@@ -3077,7 +3077,7 @@ static int mvpp2_prs_mac_da_accept(struct mvpp2 *priv, int port,
 	if (pmap == 0) {
 		if (add) {
 			kfree(pe);
-			return -1;
+			return -EINVAL;
 		}
 		mvpp2_prs_hw_inv(priv, pe->index);
 		priv->prs_shadow[pe->index].valid = false;
diff --git a/drivers/net/ethernet/mellanox/mlx4/catas.c b/drivers/net/ethernet/mellanox/mlx4/catas.c
index 715de8affcc9..c7e939945259 100644
--- a/drivers/net/ethernet/mellanox/mlx4/catas.c
+++ b/drivers/net/ethernet/mellanox/mlx4/catas.c
@@ -182,10 +182,17 @@ void mlx4_enter_error_state(struct mlx4_dev_persistent *persist)
 		err = mlx4_reset_slave(dev);
 	else
 		err = mlx4_reset_master(dev);
-	BUG_ON(err != 0);
 
+	if (!err) {
+		mlx4_err(dev, "device was reset successfully\n");
+	} else {
+		/* EEH could have disabled the PCI channel during reset. That's
+		 * recoverable and the PCI error flow will handle it.
+		 */
+		if (!pci_channel_offline(dev->persist->pdev))
+			BUG_ON(1);
+	}
 	dev->persist->state |= MLX4_DEVICE_STATE_INTERNAL_ERROR;
-	mlx4_err(dev, "device was reset successfully\n");
 	mutex_unlock(&persist->device_state_mutex);
 
 	/* At that step HW was already reset, now notify clients */
diff --git a/drivers/net/ethernet/mellanox/mlx4/cmd.c b/drivers/net/ethernet/mellanox/mlx4/cmd.c
index d48d5793407d..e94ca1c3fc7c 100644
--- a/drivers/net/ethernet/mellanox/mlx4/cmd.c
+++ b/drivers/net/ethernet/mellanox/mlx4/cmd.c
@@ -2429,7 +2429,7 @@ err_thread:
 	flush_workqueue(priv->mfunc.master.comm_wq);
 	destroy_workqueue(priv->mfunc.master.comm_wq);
 err_slaves:
-	while (--i) {
+	while (i--) {
 		for (port = 1; port <= MLX4_MAX_PORTS; port++)
 			kfree(priv->mfunc.master.slave_state[i].vlan_filter[port]);
 	}
diff --git a/drivers/net/ethernet/mellanox/mlx4/cq.c b/drivers/net/ethernet/mellanox/mlx4/cq.c
index 3348e646db70..a849da92f857 100644
--- a/drivers/net/ethernet/mellanox/mlx4/cq.c
+++ b/drivers/net/ethernet/mellanox/mlx4/cq.c
@@ -318,7 +318,9 @@ int mlx4_cq_alloc(struct mlx4_dev *dev, int nent,
 	if (timestamp_en)
 		cq_context->flags  |= cpu_to_be32(1 << 19);
 
-	cq_context->logsize_usrpage = cpu_to_be32((ilog2(nent) << 24) | uar->index);
+	cq_context->logsize_usrpage =
+		cpu_to_be32((ilog2(nent) << 24) |
+			    mlx4_to_hw_uar_index(dev, uar->index));
 	cq_context->comp_eqn	    = priv->eq_table.eq[MLX4_CQ_TO_EQ_VECTOR(vector)].eqn;
 	cq_context->log_page_size   = mtt->page_shift - MLX4_ICM_PAGE_SHIFT;
 
diff --git a/drivers/net/ethernet/mellanox/mlx4/en_clock.c b/drivers/net/ethernet/mellanox/mlx4/en_clock.c
index 038f9ce391e6..1494997c4f7e 100644
--- a/drivers/net/ethernet/mellanox/mlx4/en_clock.c
+++ b/drivers/net/ethernet/mellanox/mlx4/en_clock.c
@@ -236,6 +236,24 @@ static const struct ptp_clock_info mlx4_en_ptp_clock_info = {
 	.enable		= mlx4_en_phc_enable,
 };
 
+#define MLX4_EN_WRAP_AROUND_SEC	10ULL
+
+/* This function calculates the max shift that enables the user range
+ * of MLX4_EN_WRAP_AROUND_SEC values in the cycles register.
+ */
+static u32 freq_to_shift(u16 freq)
+{
+	u32 freq_khz = freq * 1000;
+	u64 max_val_cycles = freq_khz * 1000 * MLX4_EN_WRAP_AROUND_SEC;
+	u64 max_val_cycles_rounded = is_power_of_2(max_val_cycles + 1) ?
+		max_val_cycles : roundup_pow_of_two(max_val_cycles) - 1;
+	/* calculate max possible multiplier in order to fit in 64bit */
+	u64 max_mul = div_u64(0xffffffffffffffffULL, max_val_cycles_rounded);
+
+	/* This comes from the reverse of clocksource_khz2mult */
+	return ilog2(div_u64(max_mul * freq_khz, 1000000));
+}
+
 void mlx4_en_init_timestamp(struct mlx4_en_dev *mdev)
 {
 	struct mlx4_dev *dev = mdev->dev;
@@ -254,12 +272,7 @@ void mlx4_en_init_timestamp(struct mlx4_en_dev *mdev)
 	memset(&mdev->cycles, 0, sizeof(mdev->cycles));
 	mdev->cycles.read = mlx4_en_read_clock;
 	mdev->cycles.mask = CLOCKSOURCE_MASK(48);
-	/* Using shift to make calculation more accurate. Since current HW
-	 * clock frequency is 427 MHz, and cycles are given using a 48 bits
-	 * register, the biggest shift when calculating using u64, is 14
-	 * (max_cycles * multiplier < 2^64)
-	 */
-	mdev->cycles.shift = 14;
+	mdev->cycles.shift = freq_to_shift(dev->caps.hca_core_clock);
 	mdev->cycles.mult =
 		clocksource_khz2mult(1000 * dev->caps.hca_core_clock, mdev->cycles.shift);
 	mdev->nominal_c_mult = mdev->cycles.mult;
diff --git a/drivers/net/ethernet/mellanox/mlx4/en_netdev.c b/drivers/net/ethernet/mellanox/mlx4/en_netdev.c
index 0c7e3f69a73b..21e2c0960271 100644
--- a/drivers/net/ethernet/mellanox/mlx4/en_netdev.c
+++ b/drivers/net/ethernet/mellanox/mlx4/en_netdev.c
@@ -2245,7 +2245,7 @@ static int mlx4_en_set_vf_mac(struct net_device *dev, int queue, u8 *mac)
 	struct mlx4_en_dev *mdev = en_priv->mdev;
 	u64 mac_u64 = mlx4_mac_to_u64(mac);
 
-	if (!is_valid_ether_addr(mac))
+	if (is_multicast_ether_addr(mac))
 		return -EINVAL;
 
 	return mlx4_set_vf_mac(mdev->dev, en_priv->port, queue, mac_u64);
@@ -2344,8 +2344,6 @@ out:
 	/* set offloads */
 	priv->dev->hw_enc_features |= NETIF_F_IP_CSUM | NETIF_F_RXCSUM |
 				      NETIF_F_TSO | NETIF_F_GSO_UDP_TUNNEL;
-	priv->dev->hw_features |= NETIF_F_GSO_UDP_TUNNEL;
-	priv->dev->features    |= NETIF_F_GSO_UDP_TUNNEL;
 }
 
 static void mlx4_en_del_vxlan_offloads(struct work_struct *work)
@@ -2356,8 +2354,6 @@ static void mlx4_en_del_vxlan_offloads(struct work_struct *work)
 	/* unset offloads */
 	priv->dev->hw_enc_features &= ~(NETIF_F_IP_CSUM | NETIF_F_RXCSUM |
 				      NETIF_F_TSO | NETIF_F_GSO_UDP_TUNNEL);
-	priv->dev->hw_features &= ~NETIF_F_GSO_UDP_TUNNEL;
-	priv->dev->features    &= ~NETIF_F_GSO_UDP_TUNNEL;
 
 	ret = mlx4_SET_PORT_VXLAN(priv->mdev->dev, priv->port,
 				  VXLAN_STEER_BY_OUTER_MAC, 0);
@@ -2980,6 +2976,11 @@ int mlx4_en_init_netdev(struct mlx4_en_dev *mdev, int port,
 		priv->rss_hash_fn = ETH_RSS_HASH_TOP;
 	}
 
+	if (mdev->dev->caps.tunnel_offload_mode == MLX4_TUNNEL_OFFLOAD_MODE_VXLAN) {
+		dev->hw_features |= NETIF_F_GSO_UDP_TUNNEL;
+		dev->features    |= NETIF_F_GSO_UDP_TUNNEL;
+	}
+
 	mdev->pndev[port] = dev;
 	mdev->upper[port] = NULL;
 
diff --git a/drivers/net/ethernet/mellanox/mlx4/en_port.c b/drivers/net/ethernet/mellanox/mlx4/en_port.c
index ee99e67187f5..3904b5fc0b7c 100644
--- a/drivers/net/ethernet/mellanox/mlx4/en_port.c
+++ b/drivers/net/ethernet/mellanox/mlx4/en_port.c
@@ -238,11 +238,11 @@ int mlx4_en_DUMP_ETH_STATS(struct mlx4_en_dev *mdev, u8 port, u8 reset)
 	stats->collisions = 0;
 	stats->rx_dropped = be32_to_cpu(mlx4_en_stats->RDROP);
 	stats->rx_length_errors = be32_to_cpu(mlx4_en_stats->RdropLength);
-	stats->rx_over_errors = be32_to_cpu(mlx4_en_stats->RdropOvflw);
+	stats->rx_over_errors = 0;
 	stats->rx_crc_errors = be32_to_cpu(mlx4_en_stats->RCRC);
 	stats->rx_frame_errors = 0;
 	stats->rx_fifo_errors = be32_to_cpu(mlx4_en_stats->RdropOvflw);
-	stats->rx_missed_errors = be32_to_cpu(mlx4_en_stats->RdropOvflw);
+	stats->rx_missed_errors = 0;
 	stats->tx_aborted_errors = 0;
 	stats->tx_carrier_errors = 0;
 	stats->tx_fifo_errors = 0;
diff --git a/drivers/net/ethernet/mellanox/mlx4/en_resources.c b/drivers/net/ethernet/mellanox/mlx4/en_resources.c
index 12aab5a659d3..02e925d6f734 100644
--- a/drivers/net/ethernet/mellanox/mlx4/en_resources.c
+++ b/drivers/net/ethernet/mellanox/mlx4/en_resources.c
@@ -58,7 +58,8 @@ void mlx4_en_fill_qp_context(struct mlx4_en_priv *priv, int size, int stride,
 	} else {
 		context->sq_size_stride = ilog2(TXBB_SIZE) - 4;
 	}
-	context->usr_page = cpu_to_be32(mdev->priv_uar.index);
+	context->usr_page = cpu_to_be32(mlx4_to_hw_uar_index(mdev->dev,
+					mdev->priv_uar.index));
 	context->local_qpn = cpu_to_be32(qpn);
 	context->pri_path.ackto = 1 & 0x07;
 	context->pri_path.sched_queue = 0x83 | (priv->port - 1) << 6;
diff --git a/drivers/net/ethernet/mellanox/mlx4/en_tx.c b/drivers/net/ethernet/mellanox/mlx4/en_tx.c
index 4421bf5463f6..e0946ab22010 100644
--- a/drivers/net/ethernet/mellanox/mlx4/en_tx.c
+++ b/drivers/net/ethernet/mellanox/mlx4/en_tx.c
@@ -213,7 +213,9 @@ int mlx4_en_activate_tx_ring(struct mlx4_en_priv *priv,
 	mlx4_en_fill_qp_context(priv, ring->size, ring->stride, 1, 0, ring->qpn,
 				ring->cqn, user_prio, &ring->context);
 	if (ring->bf_alloced)
-		ring->context.usr_page = cpu_to_be32(ring->bf.uar->index);
+		ring->context.usr_page =
+			cpu_to_be32(mlx4_to_hw_uar_index(mdev->dev,
+							 ring->bf.uar->index));
 
 	err = mlx4_qp_to_ready(mdev->dev, &ring->wqres.mtt, &ring->context,
 			       &ring->qp, &ring->qp_state);
diff --git a/drivers/net/ethernet/mellanox/mlx4/eq.c b/drivers/net/ethernet/mellanox/mlx4/eq.c
index 4696053165f8..f613977455e0 100644
--- a/drivers/net/ethernet/mellanox/mlx4/eq.c
+++ b/drivers/net/ethernet/mellanox/mlx4/eq.c
@@ -940,9 +940,10 @@ static void __iomem *mlx4_get_eq_uar(struct mlx4_dev *dev, struct mlx4_eq *eq)
 
 	if (!priv->eq_table.uar_map[index]) {
 		priv->eq_table.uar_map[index] =
-			ioremap(pci_resource_start(dev->persist->pdev, 2) +
-				((eq->eqn / 4) << PAGE_SHIFT),
-				PAGE_SIZE);
+			ioremap(
+				pci_resource_start(dev->persist->pdev, 2) +
+				((eq->eqn / 4) << (dev->uar_page_shift)),
+				(1 << (dev->uar_page_shift)));
 		if (!priv->eq_table.uar_map[index]) {
 			mlx4_err(dev, "Couldn't map EQ doorbell for EQN 0x%06x\n",
 				 eq->eqn);
diff --git a/drivers/net/ethernet/mellanox/mlx4/main.c b/drivers/net/ethernet/mellanox/mlx4/main.c
index f1b6d219e445..f8674ae62752 100644
--- a/drivers/net/ethernet/mellanox/mlx4/main.c
+++ b/drivers/net/ethernet/mellanox/mlx4/main.c
@@ -168,6 +168,20 @@ struct mlx4_port_config {
 
 static atomic_t pf_loading = ATOMIC_INIT(0);
 
+static inline void mlx4_set_num_reserved_uars(struct mlx4_dev *dev,
+					      struct mlx4_dev_cap *dev_cap)
+{
+	/* The reserved_uars is calculated by system page size unit.
+	 * Therefore, adjustment is added when the uar page size is less
+	 * than the system page size
+	 */
+	dev->caps.reserved_uars	=
+		max_t(int,
+		      mlx4_get_num_reserved_uar(dev),
+		      dev_cap->reserved_uars /
+			(1 << (PAGE_SHIFT - dev->uar_page_shift)));
+}
+
 int mlx4_check_port_params(struct mlx4_dev *dev,
 			   enum mlx4_port_type *port_type)
 {
@@ -386,8 +400,6 @@ static int mlx4_dev_cap(struct mlx4_dev *dev, struct mlx4_dev_cap *dev_cap)
 	dev->caps.reserved_mtts      = dev_cap->reserved_mtts;
 	dev->caps.reserved_mrws	     = dev_cap->reserved_mrws;
 
-	/* The first 128 UARs are used for EQ doorbells */
-	dev->caps.reserved_uars	     = max_t(int, 128, dev_cap->reserved_uars);
 	dev->caps.reserved_pds	     = dev_cap->reserved_pds;
 	dev->caps.reserved_xrcds     = (dev->caps.flags & MLX4_DEV_CAP_FLAG_XRC) ?
 					dev_cap->reserved_xrcds : 0;
@@ -405,6 +417,15 @@ static int mlx4_dev_cap(struct mlx4_dev *dev, struct mlx4_dev_cap *dev_cap)
 	dev->caps.max_gso_sz	     = dev_cap->max_gso_sz;
 	dev->caps.max_rss_tbl_sz     = dev_cap->max_rss_tbl_sz;
 
+	/* Save uar page shift */
+	if (!mlx4_is_slave(dev)) {
+		/* Virtual PCI function needs to determine UAR page size from
+		 * firmware. Only master PCI function can set the uar page size
+		 */
+		dev->uar_page_shift = DEFAULT_UAR_PAGE_SHIFT;
+		mlx4_set_num_reserved_uars(dev, dev_cap);
+	}
+
 	if (dev->caps.flags2 & MLX4_DEV_CAP_FLAG2_PHV_EN) {
 		struct mlx4_init_hca_param hca_param;
 
@@ -815,16 +836,25 @@ static int mlx4_slave_cap(struct mlx4_dev *dev)
 		return -ENODEV;
 	}
 
-	/* slave gets uar page size from QUERY_HCA fw command */
-	dev->caps.uar_page_size = 1 << (hca_param.uar_page_sz + 12);
+	/* Set uar_page_shift for VF */
+	dev->uar_page_shift = hca_param.uar_page_sz + 12;
 
-	/* TODO: relax this assumption */
-	if (dev->caps.uar_page_size != PAGE_SIZE) {
-		mlx4_err(dev, "UAR size:%d != kernel PAGE_SIZE of %ld\n",
-			 dev->caps.uar_page_size, PAGE_SIZE);
-		return -ENODEV;
+	/* Make sure the master uar page size is valid */
+	if (dev->uar_page_shift > PAGE_SHIFT) {
+		mlx4_err(dev,
+			 "Invalid configuration: uar page size is larger than system page size\n");
+		return  -ENODEV;
 	}
 
+	/* Set reserved_uars based on the uar_page_shift */
+	mlx4_set_num_reserved_uars(dev, &dev_cap);
+
+	/* Although uar page size in FW differs from system page size,
+	 * upper software layers (mlx4_ib, mlx4_en and part of mlx4_core)
+	 * still works with assumption that uar page size == system page size
+	 */
+	dev->caps.uar_page_size = PAGE_SIZE;
+
 	memset(&func_cap, 0, sizeof(func_cap));
 	err = mlx4_QUERY_FUNC_CAP(dev, 0, &func_cap);
 	if (err) {
@@ -1226,6 +1256,7 @@ err_set_port:
 static int mlx4_mf_bond(struct mlx4_dev *dev)
 {
 	int err = 0;
+	int nvfs;
 	struct mlx4_slaves_pport slaves_port1;
 	struct mlx4_slaves_pport slaves_port2;
 	DECLARE_BITMAP(slaves_port_1_2, MLX4_MFUNC_MAX);
@@ -1242,11 +1273,18 @@ static int mlx4_mf_bond(struct mlx4_dev *dev)
 		return -EINVAL;
 	}
 
+	/* number of virtual functions is number of total functions minus one
+	 * physical function for each port.
+	 */
+	nvfs = bitmap_weight(slaves_port1.slaves, dev->persist->num_vfs + 1) +
+		bitmap_weight(slaves_port2.slaves, dev->persist->num_vfs + 1) - 2;
+
 	/* limit on maximum allowed VFs */
-	if ((bitmap_weight(slaves_port1.slaves, dev->persist->num_vfs + 1) +
-	    bitmap_weight(slaves_port2.slaves, dev->persist->num_vfs + 1)) >
-	    MAX_MF_BOND_ALLOWED_SLAVES)
+	if (nvfs > MAX_MF_BOND_ALLOWED_SLAVES) {
+		mlx4_warn(dev, "HA mode is not supported for %d VFs (max %d are allowed)\n",
+			  nvfs, MAX_MF_BOND_ALLOWED_SLAVES);
 		return -EINVAL;
+	}
 
 	if (dev->caps.steering_mode != MLX4_STEERING_MODE_DEVICE_MANAGED) {
 		mlx4_warn(dev, "HA mode unsupported for NON DMFS steering\n");
@@ -2179,8 +2217,12 @@ static int mlx4_init_hca(struct mlx4_dev *dev)
 
 		dev->caps.max_fmr_maps = (1 << (32 - ilog2(dev->caps.num_mpts))) - 1;
 
-		init_hca.log_uar_sz = ilog2(dev->caps.num_uars);
-		init_hca.uar_page_sz = PAGE_SHIFT - 12;
+		/* Always set UAR page size 4KB, set log_uar_sz accordingly */
+		init_hca.log_uar_sz = ilog2(dev->caps.num_uars) +
+				      PAGE_SHIFT -
+				      DEFAULT_UAR_PAGE_SHIFT;
+		init_hca.uar_page_sz = DEFAULT_UAR_PAGE_SHIFT - 12;
+
 		init_hca.mw_enabled = 0;
 		if (dev->caps.flags & MLX4_DEV_CAP_FLAG_MEM_WINDOW ||
 		    dev->caps.bmme_flags & MLX4_BMME_FLAG_TYPE_2_WIN)
diff --git a/drivers/net/ethernet/mellanox/mlx4/pd.c b/drivers/net/ethernet/mellanox/mlx4/pd.c
index 609c59dc854e..b3cc3ab63799 100644
--- a/drivers/net/ethernet/mellanox/mlx4/pd.c
+++ b/drivers/net/ethernet/mellanox/mlx4/pd.c
@@ -269,9 +269,15 @@ EXPORT_SYMBOL_GPL(mlx4_bf_free);
 
 int mlx4_init_uar_table(struct mlx4_dev *dev)
 {
-	if (dev->caps.num_uars <= 128) {
-		mlx4_err(dev, "Only %d UAR pages (need more than 128)\n",
-			 dev->caps.num_uars);
+	int num_reserved_uar = mlx4_get_num_reserved_uar(dev);
+
+	mlx4_dbg(dev, "uar_page_shift = %d", dev->uar_page_shift);
+	mlx4_dbg(dev, "Effective reserved_uars=%d", dev->caps.reserved_uars);
+
+	if (dev->caps.num_uars <= num_reserved_uar) {
+		mlx4_err(
+			dev, "Only %d UAR pages (need more than %d)\n",
+			dev->caps.num_uars, num_reserved_uar);
 		mlx4_err(dev, "Increase firmware log2_uar_bar_megabytes?\n");
 		return -ENODEV;
 	}
diff --git a/drivers/net/ethernet/mellanox/mlx4/port.c b/drivers/net/ethernet/mellanox/mlx4/port.c
index 787b7bb54d52..211c65087997 100644
--- a/drivers/net/ethernet/mellanox/mlx4/port.c
+++ b/drivers/net/ethernet/mellanox/mlx4/port.c
@@ -193,10 +193,10 @@ int __mlx4_register_mac(struct mlx4_dev *dev, u8 port, u64 mac)
 	if (need_mf_bond) {
 		if (port == 1) {
 			mutex_lock(&table->mutex);
-			mutex_lock(&dup_table->mutex);
+			mutex_lock_nested(&dup_table->mutex, SINGLE_DEPTH_NESTING);
 		} else {
 			mutex_lock(&dup_table->mutex);
-			mutex_lock(&table->mutex);
+			mutex_lock_nested(&table->mutex, SINGLE_DEPTH_NESTING);
 		}
 	} else {
 		mutex_lock(&table->mutex);
@@ -389,10 +389,10 @@ void __mlx4_unregister_mac(struct mlx4_dev *dev, u8 port, u64 mac)
 	if (dup) {
 		if (port == 1) {
 			mutex_lock(&table->mutex);
-			mutex_lock(&dup_table->mutex);
+			mutex_lock_nested(&dup_table->mutex, SINGLE_DEPTH_NESTING);
 		} else {
 			mutex_lock(&dup_table->mutex);
-			mutex_lock(&table->mutex);
+			mutex_lock_nested(&table->mutex, SINGLE_DEPTH_NESTING);
 		}
 	} else {
 		mutex_lock(&table->mutex);
@@ -479,10 +479,10 @@ int __mlx4_replace_mac(struct mlx4_dev *dev, u8 port, int qpn, u64 new_mac)
 	if (dup) {
 		if (port == 1) {
 			mutex_lock(&table->mutex);
-			mutex_lock(&dup_table->mutex);
+			mutex_lock_nested(&dup_table->mutex, SINGLE_DEPTH_NESTING);
 		} else {
 			mutex_lock(&dup_table->mutex);
-			mutex_lock(&table->mutex);
+			mutex_lock_nested(&table->mutex, SINGLE_DEPTH_NESTING);
 		}
 	} else {
 		mutex_lock(&table->mutex);
@@ -588,10 +588,10 @@ int __mlx4_register_vlan(struct mlx4_dev *dev, u8 port, u16 vlan,
 	if (need_mf_bond) {
 		if (port == 1) {
 			mutex_lock(&table->mutex);
-			mutex_lock(&dup_table->mutex);
+			mutex_lock_nested(&dup_table->mutex, SINGLE_DEPTH_NESTING);
 		} else {
 			mutex_lock(&dup_table->mutex);
-			mutex_lock(&table->mutex);
+			mutex_lock_nested(&table->mutex, SINGLE_DEPTH_NESTING);
 		}
 	} else {
 		mutex_lock(&table->mutex);
@@ -764,10 +764,10 @@ void __mlx4_unregister_vlan(struct mlx4_dev *dev, u8 port, u16 vlan)
 	if (dup) {
 		if (port == 1) {
 			mutex_lock(&table->mutex);
-			mutex_lock(&dup_table->mutex);
+			mutex_lock_nested(&dup_table->mutex, SINGLE_DEPTH_NESTING);
 		} else {
 			mutex_lock(&dup_table->mutex);
-			mutex_lock(&table->mutex);
+			mutex_lock_nested(&table->mutex, SINGLE_DEPTH_NESTING);
 		}
 	} else {
 		mutex_lock(&table->mutex);
diff --git a/drivers/net/ethernet/mellanox/mlx4/resource_tracker.c b/drivers/net/ethernet/mellanox/mlx4/resource_tracker.c
index b46dbe29ef6c..25ce1b030a00 100644
--- a/drivers/net/ethernet/mellanox/mlx4/resource_tracker.c
+++ b/drivers/net/ethernet/mellanox/mlx4/resource_tracker.c
@@ -915,11 +915,13 @@ static int handle_existing_counter(struct mlx4_dev *dev, u8 slave, int port,
 
 	spin_lock_irq(mlx4_tlock(dev));
 	r = find_res(dev, counter_index, RES_COUNTER);
-	if (!r || r->owner != slave)
+	if (!r || r->owner != slave) {
 		ret = -EINVAL;
-	counter = container_of(r, struct res_counter, com);
-	if (!counter->port)
-		counter->port = port;
+	} else {
+		counter = container_of(r, struct res_counter, com);
+		if (!counter->port)
+			counter->port = port;
+	}
 
 	spin_unlock_irq(mlx4_tlock(dev));
 	return ret;
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en.h b/drivers/net/ethernet/mellanox/mlx5/core/en.h
index aac071a7e830..5b1753233c5d 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en.h
@@ -223,6 +223,7 @@ struct mlx5e_pport_stats {
 
 static const char rq_stats_strings[][ETH_GSTRING_LEN] = {
 	"packets",
+	"bytes",
 	"csum_none",
 	"csum_sw",
 	"lro_packets",
@@ -232,16 +233,18 @@ static const char rq_stats_strings[][ETH_GSTRING_LEN] = {
 
 struct mlx5e_rq_stats {
 	u64 packets;
+	u64 bytes;
 	u64 csum_none;
 	u64 csum_sw;
 	u64 lro_packets;
 	u64 lro_bytes;
 	u64 wqe_err;
-#define NUM_RQ_STATS 6
+#define NUM_RQ_STATS 7
 };
 
 static const char sq_stats_strings[][ETH_GSTRING_LEN] = {
 	"packets",
+	"bytes",
 	"tso_packets",
 	"tso_bytes",
 	"csum_offload_none",
@@ -253,6 +256,7 @@ static const char sq_stats_strings[][ETH_GSTRING_LEN] = {
 
 struct mlx5e_sq_stats {
 	u64 packets;
+	u64 bytes;
 	u64 tso_packets;
 	u64 tso_bytes;
 	u64 csum_offload_none;
@@ -260,7 +264,7 @@ struct mlx5e_sq_stats {
 	u64 wake;
 	u64 dropped;
 	u64 nop;
-#define NUM_SQ_STATS 8
+#define NUM_SQ_STATS 9
 };
 
 struct mlx5e_stats {
@@ -304,14 +308,9 @@ enum {
 	MLX5E_RQ_STATE_POST_WQES_ENABLE,
 };
 
-enum cq_flags {
-	MLX5E_CQ_HAS_CQES = 1,
-};
-
 struct mlx5e_cq {
 	/* data path - accessed per cqe */
 	struct mlx5_cqwq           wq;
-	unsigned long              flags;
 
 	/* data path - accessed per napi poll */
 	struct napi_struct        *napi;
@@ -452,6 +451,8 @@ enum mlx5e_traffic_types {
 	MLX5E_NUM_TT,
 };
 
+#define IS_HASHING_TT(tt) (tt != MLX5E_TT_ANY)
+
 enum mlx5e_rqt_ix {
 	MLX5E_INDIRECTION_RQT,
 	MLX5E_SINGLE_RQ_RQT,
@@ -618,9 +619,12 @@ void mlx5e_enable_vlan_filter(struct mlx5e_priv *priv);
 void mlx5e_disable_vlan_filter(struct mlx5e_priv *priv);
 
 int mlx5e_redirect_rqt(struct mlx5e_priv *priv, enum mlx5e_rqt_ix rqt_ix);
+void mlx5e_build_tir_ctx_hash(void *tirc, struct mlx5e_priv *priv);
 
 int mlx5e_open_locked(struct net_device *netdev);
 int mlx5e_close_locked(struct net_device *netdev);
+void mlx5e_build_default_indir_rqt(u32 *indirection_rqt, int len,
+				   int num_channels);
 
 static inline void mlx5e_tx_notify_hw(struct mlx5e_sq *sq,
 				      struct mlx5e_tx_wqe *wqe, int bf_sz)
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_clock.c b/drivers/net/ethernet/mellanox/mlx5/core/en_clock.c
index be6543570b2b..2018eebe1531 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_clock.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_clock.c
@@ -62,10 +62,11 @@ static void mlx5e_timestamp_overflow(struct work_struct *work)
 	struct delayed_work *dwork = to_delayed_work(work);
 	struct mlx5e_tstamp *tstamp = container_of(dwork, struct mlx5e_tstamp,
 						   overflow_work);
+	unsigned long flags;
 
-	write_lock(&tstamp->lock);
+	write_lock_irqsave(&tstamp->lock, flags);
 	timecounter_read(&tstamp->clock);
-	write_unlock(&tstamp->lock);
+	write_unlock_irqrestore(&tstamp->lock, flags);
 	schedule_delayed_work(&tstamp->overflow_work, tstamp->overflow_period);
 }
 
@@ -136,10 +137,11 @@ static int mlx5e_ptp_settime(struct ptp_clock_info *ptp,
 	struct mlx5e_tstamp *tstamp = container_of(ptp, struct mlx5e_tstamp,
 						   ptp_info);
 	u64 ns = timespec64_to_ns(ts);
+	unsigned long flags;
 
-	write_lock(&tstamp->lock);
+	write_lock_irqsave(&tstamp->lock, flags);
 	timecounter_init(&tstamp->clock, &tstamp->cycles, ns);
-	write_unlock(&tstamp->lock);
+	write_unlock_irqrestore(&tstamp->lock, flags);
 
 	return 0;
 }
@@ -150,10 +152,11 @@ static int mlx5e_ptp_gettime(struct ptp_clock_info *ptp,
 	struct mlx5e_tstamp *tstamp = container_of(ptp, struct mlx5e_tstamp,
 						   ptp_info);
 	u64 ns;
+	unsigned long flags;
 
-	write_lock(&tstamp->lock);
+	write_lock_irqsave(&tstamp->lock, flags);
 	ns = timecounter_read(&tstamp->clock);
-	write_unlock(&tstamp->lock);
+	write_unlock_irqrestore(&tstamp->lock, flags);
 
 	*ts = ns_to_timespec64(ns);
 
@@ -164,10 +167,11 @@ static int mlx5e_ptp_adjtime(struct ptp_clock_info *ptp, s64 delta)
 {
 	struct mlx5e_tstamp *tstamp = container_of(ptp, struct mlx5e_tstamp,
 						   ptp_info);
+	unsigned long flags;
 
-	write_lock(&tstamp->lock);
+	write_lock_irqsave(&tstamp->lock, flags);
 	timecounter_adjtime(&tstamp->clock, delta);
-	write_unlock(&tstamp->lock);
+	write_unlock_irqrestore(&tstamp->lock, flags);
 
 	return 0;
 }
@@ -176,6 +180,7 @@ static int mlx5e_ptp_adjfreq(struct ptp_clock_info *ptp, s32 delta)
 {
 	u64 adj;
 	u32 diff;
+	unsigned long flags;
 	int neg_adj = 0;
 	struct mlx5e_tstamp *tstamp = container_of(ptp, struct mlx5e_tstamp,
 						  ptp_info);
@@ -189,11 +194,11 @@ static int mlx5e_ptp_adjfreq(struct ptp_clock_info *ptp, s32 delta)
 	adj *= delta;
 	diff = div_u64(adj, 1000000000ULL);
 
-	write_lock(&tstamp->lock);
+	write_lock_irqsave(&tstamp->lock, flags);
 	timecounter_read(&tstamp->clock);
 	tstamp->cycles.mult = neg_adj ? tstamp->nominal_c_mult - diff :
 					tstamp->nominal_c_mult + diff;
-	write_unlock(&tstamp->lock);
+	write_unlock_irqrestore(&tstamp->lock, flags);
 
 	return 0;
 }
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c b/drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c
index 65624ac65b4c..5abeb00fceb8 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c
@@ -385,6 +385,8 @@ static int mlx5e_set_channels(struct net_device *dev,
 		mlx5e_close_locked(dev);
 
 	priv->params.num_channels = count;
+	mlx5e_build_default_indir_rqt(priv->params.indirection_rqt,
+				      MLX5E_INDIR_RQT_SIZE, count);
 
 	if (was_opened)
 		err = mlx5e_open_locked(dev);
@@ -703,18 +705,36 @@ static int mlx5e_get_rxfh(struct net_device *netdev, u32 *indir, u8 *key,
 	return 0;
 }
 
+static void mlx5e_modify_tirs_hash(struct mlx5e_priv *priv, void *in, int inlen)
+{
+	struct mlx5_core_dev *mdev = priv->mdev;
+	void *tirc = MLX5_ADDR_OF(modify_tir_in, in, ctx);
+	int i;
+
+	MLX5_SET(modify_tir_in, in, bitmask.hash, 1);
+	mlx5e_build_tir_ctx_hash(tirc, priv);
+
+	for (i = 0; i < MLX5E_NUM_TT; i++)
+		if (IS_HASHING_TT(i))
+			mlx5_core_modify_tir(mdev, priv->tirn[i], in, inlen);
+}
+
 static int mlx5e_set_rxfh(struct net_device *dev, const u32 *indir,
 			  const u8 *key, const u8 hfunc)
 {
 	struct mlx5e_priv *priv = netdev_priv(dev);
-	bool close_open;
-	int err = 0;
+	int inlen = MLX5_ST_SZ_BYTES(modify_tir_in);
+	void *in;
 
 	if ((hfunc != ETH_RSS_HASH_NO_CHANGE) &&
 	    (hfunc != ETH_RSS_HASH_XOR) &&
 	    (hfunc != ETH_RSS_HASH_TOP))
 		return -EINVAL;
 
+	in = mlx5_vzalloc(inlen);
+	if (!in)
+		return -ENOMEM;
+
 	mutex_lock(&priv->state_lock);
 
 	if (indir) {
@@ -723,11 +743,6 @@ static int mlx5e_set_rxfh(struct net_device *dev, const u32 *indir,
 		mlx5e_redirect_rqt(priv, MLX5E_INDIRECTION_RQT);
 	}
 
-	close_open = (key || (hfunc != ETH_RSS_HASH_NO_CHANGE)) &&
-		     test_bit(MLX5E_STATE_OPENED, &priv->state);
-	if (close_open)
-		mlx5e_close_locked(dev);
-
 	if (key)
 		memcpy(priv->params.toeplitz_hash_key, key,
 		       sizeof(priv->params.toeplitz_hash_key));
@@ -735,12 +750,13 @@ static int mlx5e_set_rxfh(struct net_device *dev, const u32 *indir,
 	if (hfunc != ETH_RSS_HASH_NO_CHANGE)
 		priv->params.rss_hfunc = hfunc;
 
-	if (close_open)
-		err = mlx5e_open_locked(priv->netdev);
+	mlx5e_modify_tirs_hash(priv, in, inlen);
 
 	mutex_unlock(&priv->state_lock);
 
-	return err;
+	kvfree(in);
+
+	return 0;
 }
 
 static int mlx5e_get_rxnfc(struct net_device *netdev,
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c
index 6a3e430f1062..402994bf7e16 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c
@@ -141,6 +141,10 @@ void mlx5e_update_stats(struct mlx5e_priv *priv)
 		return;
 
 	/* Collect firts the SW counters and then HW for consistency */
+	s->rx_packets		= 0;
+	s->rx_bytes		= 0;
+	s->tx_packets		= 0;
+	s->tx_bytes		= 0;
 	s->tso_packets		= 0;
 	s->tso_bytes		= 0;
 	s->tx_queue_stopped	= 0;
@@ -155,6 +159,8 @@ void mlx5e_update_stats(struct mlx5e_priv *priv)
 	for (i = 0; i < priv->params.num_channels; i++) {
 		rq_stats = &priv->channel[i]->rq.stats;
 
+		s->rx_packets	+= rq_stats->packets;
+		s->rx_bytes	+= rq_stats->bytes;
 		s->lro_packets	+= rq_stats->lro_packets;
 		s->lro_bytes	+= rq_stats->lro_bytes;
 		s->rx_csum_none	+= rq_stats->csum_none;
@@ -164,6 +170,8 @@ void mlx5e_update_stats(struct mlx5e_priv *priv)
 		for (j = 0; j < priv->params.num_tc; j++) {
 			sq_stats = &priv->channel[i]->sq[j].stats;
 
+			s->tx_packets		+= sq_stats->packets;
+			s->tx_bytes		+= sq_stats->bytes;
 			s->tso_packets		+= sq_stats->tso_packets;
 			s->tso_bytes		+= sq_stats->tso_bytes;
 			s->tx_queue_stopped	+= sq_stats->stopped;
@@ -225,23 +233,6 @@ void mlx5e_update_stats(struct mlx5e_priv *priv)
 	s->tx_broadcast_bytes   =
 		MLX5_GET_CTR(out, transmitted_eth_broadcast.octets);
 
-	s->rx_packets =
-		s->rx_unicast_packets +
-		s->rx_multicast_packets +
-		s->rx_broadcast_packets;
-	s->rx_bytes =
-		s->rx_unicast_bytes +
-		s->rx_multicast_bytes +
-		s->rx_broadcast_bytes;
-	s->tx_packets =
-		s->tx_unicast_packets +
-		s->tx_multicast_packets +
-		s->tx_broadcast_packets;
-	s->tx_bytes =
-		s->tx_unicast_bytes +
-		s->tx_multicast_bytes +
-		s->tx_broadcast_bytes;
-
 	/* Update calculated offload counters */
 	s->tx_csum_offload = s->tx_packets - tx_offload_none;
 	s->rx_csum_good    = s->rx_packets - s->rx_csum_none -
@@ -1199,7 +1190,6 @@ static void mlx5e_fill_indir_rqt_rqns(struct mlx5e_priv *priv, void *rqtc)
 			ix = mlx5e_bits_invert(i, MLX5E_LOG_INDIR_RQT_SIZE);
 
 		ix = priv->params.indirection_rqt[ix];
-		ix = ix % priv->params.num_channels;
 		MLX5_SET(rqtc, rqtc, rq_num[i],
 			 test_bit(MLX5E_STATE_OPENED, &priv->state) ?
 			 priv->channel[ix]->rq.rqn :
@@ -1317,7 +1307,22 @@ static void mlx5e_build_tir_ctx_lro(void *tirc, struct mlx5e_priv *priv)
 			      lro_timer_supported_periods[2]));
 }
 
-static int mlx5e_modify_tir_lro(struct mlx5e_priv *priv, int tt)
+void mlx5e_build_tir_ctx_hash(void *tirc, struct mlx5e_priv *priv)
+{
+	MLX5_SET(tirc, tirc, rx_hash_fn,
+		 mlx5e_rx_hash_fn(priv->params.rss_hfunc));
+	if (priv->params.rss_hfunc == ETH_RSS_HASH_TOP) {
+		void *rss_key = MLX5_ADDR_OF(tirc, tirc,
+					     rx_hash_toeplitz_key);
+		size_t len = MLX5_FLD_SZ_BYTES(tirc,
+					       rx_hash_toeplitz_key);
+
+		MLX5_SET(tirc, tirc, rx_hash_symmetric, 1);
+		memcpy(rss_key, priv->params.toeplitz_hash_key, len);
+	}
+}
+
+static int mlx5e_modify_tirs_lro(struct mlx5e_priv *priv)
 {
 	struct mlx5_core_dev *mdev = priv->mdev;
 
@@ -1325,6 +1330,7 @@ static int mlx5e_modify_tir_lro(struct mlx5e_priv *priv, int tt)
 	void *tirc;
 	int inlen;
 	int err;
+	int tt;
 
 	inlen = MLX5_ST_SZ_BYTES(modify_tir_in);
 	in = mlx5_vzalloc(inlen);
@@ -1336,7 +1342,11 @@ static int mlx5e_modify_tir_lro(struct mlx5e_priv *priv, int tt)
 
 	mlx5e_build_tir_ctx_lro(tirc, priv);
 
-	err = mlx5_core_modify_tir(mdev, priv->tirn[tt], in, inlen);
+	for (tt = 0; tt < MLX5E_NUM_TT; tt++) {
+		err = mlx5_core_modify_tir(mdev, priv->tirn[tt], in, inlen);
+		if (err)
+			break;
+	}
 
 	kvfree(in);
 
@@ -1672,17 +1682,7 @@ static void mlx5e_build_tir_ctx(struct mlx5e_priv *priv, u32 *tirc, int tt)
 	default:
 		MLX5_SET(tirc, tirc, indirect_table,
 			 priv->rqtn[MLX5E_INDIRECTION_RQT]);
-		MLX5_SET(tirc, tirc, rx_hash_fn,
-			 mlx5e_rx_hash_fn(priv->params.rss_hfunc));
-		if (priv->params.rss_hfunc == ETH_RSS_HASH_TOP) {
-			void *rss_key = MLX5_ADDR_OF(tirc, tirc,
-						     rx_hash_toeplitz_key);
-			size_t len = MLX5_FLD_SZ_BYTES(tirc,
-						       rx_hash_toeplitz_key);
-
-			MLX5_SET(tirc, tirc, rx_hash_symmetric, 1);
-			memcpy(rss_key, priv->params.toeplitz_hash_key, len);
-		}
+		mlx5e_build_tir_ctx_hash(tirc, priv);
 		break;
 	}
 
@@ -1885,8 +1885,10 @@ static int mlx5e_set_features(struct net_device *netdev,
 			mlx5e_close_locked(priv->netdev);
 
 		priv->params.lro_en = !!(features & NETIF_F_LRO);
-		mlx5e_modify_tir_lro(priv, MLX5E_TT_IPV4_TCP);
-		mlx5e_modify_tir_lro(priv, MLX5E_TT_IPV6_TCP);
+		err = mlx5e_modify_tirs_lro(priv);
+		if (err)
+			mlx5_core_warn(priv->mdev, "lro modify failed, %d\n",
+				       err);
 
 		if (was_opened)
 			err = mlx5e_open_locked(priv->netdev);
@@ -2024,18 +2026,37 @@ static int mlx5e_get_vf_stats(struct net_device *dev,
 					    vf_stats);
 }
 
-static struct net_device_ops mlx5e_netdev_ops = {
+static const struct net_device_ops mlx5e_netdev_ops_basic = {
+	.ndo_open                = mlx5e_open,
+	.ndo_stop                = mlx5e_close,
+	.ndo_start_xmit          = mlx5e_xmit,
+	.ndo_get_stats64         = mlx5e_get_stats,
+	.ndo_set_rx_mode         = mlx5e_set_rx_mode,
+	.ndo_set_mac_address     = mlx5e_set_mac,
+	.ndo_vlan_rx_add_vid     = mlx5e_vlan_rx_add_vid,
+	.ndo_vlan_rx_kill_vid    = mlx5e_vlan_rx_kill_vid,
+	.ndo_set_features        = mlx5e_set_features,
+	.ndo_change_mtu          = mlx5e_change_mtu,
+	.ndo_do_ioctl            = mlx5e_ioctl,
+};
+
+static const struct net_device_ops mlx5e_netdev_ops_sriov = {
 	.ndo_open                = mlx5e_open,
 	.ndo_stop                = mlx5e_close,
 	.ndo_start_xmit          = mlx5e_xmit,
 	.ndo_get_stats64         = mlx5e_get_stats,
 	.ndo_set_rx_mode         = mlx5e_set_rx_mode,
 	.ndo_set_mac_address     = mlx5e_set_mac,
-	.ndo_vlan_rx_add_vid	 = mlx5e_vlan_rx_add_vid,
-	.ndo_vlan_rx_kill_vid	 = mlx5e_vlan_rx_kill_vid,
+	.ndo_vlan_rx_add_vid     = mlx5e_vlan_rx_add_vid,
+	.ndo_vlan_rx_kill_vid    = mlx5e_vlan_rx_kill_vid,
 	.ndo_set_features        = mlx5e_set_features,
-	.ndo_change_mtu		 = mlx5e_change_mtu,
-	.ndo_do_ioctl		 = mlx5e_ioctl,
+	.ndo_change_mtu          = mlx5e_change_mtu,
+	.ndo_do_ioctl            = mlx5e_ioctl,
+	.ndo_set_vf_mac          = mlx5e_set_vf_mac,
+	.ndo_set_vf_vlan         = mlx5e_set_vf_vlan,
+	.ndo_get_vf_config       = mlx5e_get_vf_config,
+	.ndo_set_vf_link_state   = mlx5e_set_vf_link_state,
+	.ndo_get_vf_stats        = mlx5e_get_vf_stats,
 };
 
 static int mlx5e_check_required_hca_cap(struct mlx5_core_dev *mdev)
@@ -2070,12 +2091,20 @@ u16 mlx5e_get_max_inline_cap(struct mlx5_core_dev *mdev)
 	       2 /*sizeof(mlx5e_tx_wqe.inline_hdr_start)*/;
 }
 
+void mlx5e_build_default_indir_rqt(u32 *indirection_rqt, int len,
+				   int num_channels)
+{
+	int i;
+
+	for (i = 0; i < len; i++)
+		indirection_rqt[i] = i % num_channels;
+}
+
 static void mlx5e_build_netdev_priv(struct mlx5_core_dev *mdev,
 				    struct net_device *netdev,
 				    int num_channels)
 {
 	struct mlx5e_priv *priv = netdev_priv(netdev);
-	int i;
 
 	priv->params.log_sq_size           =
 		MLX5E_PARAMS_DEFAULT_LOG_SQ_SIZE;
@@ -2099,8 +2128,8 @@ static void mlx5e_build_netdev_priv(struct mlx5_core_dev *mdev,
 	netdev_rss_key_fill(priv->params.toeplitz_hash_key,
 			    sizeof(priv->params.toeplitz_hash_key));
 
-	for (i = 0; i < MLX5E_INDIR_RQT_SIZE; i++)
-		priv->params.indirection_rqt[i] = i % num_channels;
+	mlx5e_build_default_indir_rqt(priv->params.indirection_rqt,
+				      MLX5E_INDIR_RQT_SIZE, num_channels);
 
 	priv->params.lro_wqe_sz            =
 		MLX5E_PARAMS_DEFAULT_LRO_WQE_SZ;
@@ -2137,18 +2166,11 @@ static void mlx5e_build_netdev(struct net_device *netdev)
 
 	SET_NETDEV_DEV(netdev, &mdev->pdev->dev);
 
-	if (priv->params.num_tc > 1)
-		mlx5e_netdev_ops.ndo_select_queue = mlx5e_select_queue;
-
-	if (MLX5_CAP_GEN(mdev, vport_group_manager)) {
-		mlx5e_netdev_ops.ndo_set_vf_mac = mlx5e_set_vf_mac;
-		mlx5e_netdev_ops.ndo_set_vf_vlan = mlx5e_set_vf_vlan;
-		mlx5e_netdev_ops.ndo_get_vf_config = mlx5e_get_vf_config;
-		mlx5e_netdev_ops.ndo_set_vf_link_state = mlx5e_set_vf_link_state;
-		mlx5e_netdev_ops.ndo_get_vf_stats = mlx5e_get_vf_stats;
-	}
+	if (MLX5_CAP_GEN(mdev, vport_group_manager))
+		netdev->netdev_ops = &mlx5e_netdev_ops_sriov;
+	else
+		netdev->netdev_ops = &mlx5e_netdev_ops_basic;
 
-	netdev->netdev_ops        = &mlx5e_netdev_ops;
 	netdev->watchdog_timeo    = 15 * HZ;
 
 	netdev->ethtool_ops	  = &mlx5e_ethtool_ops;
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_rx.c b/drivers/net/ethernet/mellanox/mlx5/core/en_rx.c
index dd959d929aad..59658b9d05d1 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_rx.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_rx.c
@@ -230,10 +230,6 @@ int mlx5e_poll_rx_cq(struct mlx5e_cq *cq, int budget)
 	struct mlx5e_rq *rq = container_of(cq, struct mlx5e_rq, cq);
 	int work_done;
 
-	/* avoid accessing cq (dma coherent memory) if not needed */
-	if (!test_and_clear_bit(MLX5E_CQ_HAS_CQES, &cq->flags))
-		return 0;
-
 	for (work_done = 0; work_done < budget; work_done++) {
 		struct mlx5e_rx_wqe *wqe;
 		struct mlx5_cqe64 *cqe;
@@ -267,6 +263,7 @@ int mlx5e_poll_rx_cq(struct mlx5e_cq *cq, int budget)
 
 		mlx5e_build_rx_skb(cqe, rq, skb);
 		rq->stats.packets++;
+		rq->stats.bytes += be32_to_cpu(cqe->byte_cnt);
 		napi_gro_receive(cq->napi, skb);
 
 wq_ll_pop:
@@ -279,8 +276,5 @@ wq_ll_pop:
 	/* ensure cq space is freed before enabling more cqes */
 	wmb();
 
-	if (work_done == budget)
-		set_bit(MLX5E_CQ_HAS_CQES, &cq->flags);
-
 	return work_done;
 }
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_tx.c b/drivers/net/ethernet/mellanox/mlx5/core/en_tx.c
index 2c3fba0fff54..bb4eeeb007de 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_tx.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_tx.c
@@ -179,6 +179,7 @@ static netdev_tx_t mlx5e_sq_xmit(struct mlx5e_sq *sq, struct sk_buff *skb)
 	unsigned int skb_len = skb->len;
 	u8  opcode = MLX5_OPCODE_SEND;
 	dma_addr_t dma_addr = 0;
+	unsigned int num_bytes;
 	bool bf = false;
 	u16 headlen;
 	u16 ds_cnt;
@@ -204,8 +205,7 @@ static netdev_tx_t mlx5e_sq_xmit(struct mlx5e_sq *sq, struct sk_buff *skb)
 		opcode       = MLX5_OPCODE_LSO;
 		ihs          = skb_transport_offset(skb) + tcp_hdrlen(skb);
 		payload_len  = skb->len - ihs;
-		wi->num_bytes = skb->len +
-				(skb_shinfo(skb)->gso_segs - 1) * ihs;
+		num_bytes = skb->len + (skb_shinfo(skb)->gso_segs - 1) * ihs;
 		sq->stats.tso_packets++;
 		sq->stats.tso_bytes += payload_len;
 	} else {
@@ -213,9 +213,11 @@ static netdev_tx_t mlx5e_sq_xmit(struct mlx5e_sq *sq, struct sk_buff *skb)
 		     !skb->xmit_more &&
 		     !skb_shinfo(skb)->nr_frags;
 		ihs = mlx5e_get_inline_hdr_size(sq, skb, bf);
-		wi->num_bytes = max_t(unsigned int, skb->len, ETH_ZLEN);
+		num_bytes = max_t(unsigned int, skb->len, ETH_ZLEN);
 	}
 
+	wi->num_bytes = num_bytes;
+
 	if (skb_vlan_tag_present(skb)) {
 		mlx5e_insert_vlan(eseg->inline_hdr_start, skb, ihs, &skb_data,
 				  &skb_len);
@@ -307,6 +309,7 @@ static netdev_tx_t mlx5e_sq_xmit(struct mlx5e_sq *sq, struct sk_buff *skb)
 	sq->bf_budget = bf ? sq->bf_budget - 1 : 0;
 
 	sq->stats.packets++;
+	sq->stats.bytes += num_bytes;
 	return NETDEV_TX_OK;
 
 dma_unmap_wqe_err:
@@ -335,10 +338,6 @@ bool mlx5e_poll_tx_cq(struct mlx5e_cq *cq)
 	u16 sqcc;
 	int i;
 
-	/* avoid accessing cq (dma coherent memory) if not needed */
-	if (!test_and_clear_bit(MLX5E_CQ_HAS_CQES, &cq->flags))
-		return false;
-
 	sq = container_of(cq, struct mlx5e_sq, cq);
 
 	npkts = 0;
@@ -422,10 +421,6 @@ bool mlx5e_poll_tx_cq(struct mlx5e_cq *cq)
 				netif_tx_wake_queue(sq->txq);
 				sq->stats.wake++;
 	}
-	if (i == MLX5E_TX_CQ_POLL_BUDGET) {
-		set_bit(MLX5E_CQ_HAS_CQES, &cq->flags);
-		return true;
-	}
 
-	return false;
+	return (i == MLX5E_TX_CQ_POLL_BUDGET);
 }
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_txrx.c b/drivers/net/ethernet/mellanox/mlx5/core/en_txrx.c
index 4ac8d716dbdd..66d51a77609e 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_txrx.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_txrx.c
@@ -88,7 +88,6 @@ void mlx5e_completion_event(struct mlx5_core_cq *mcq)
 {
 	struct mlx5e_cq *cq = container_of(mcq, struct mlx5e_cq, mcq);
 
-	set_bit(MLX5E_CQ_HAS_CQES, &cq->flags);
 	set_bit(MLX5E_CHANNEL_NAPI_SCHED, &cq->channel->flags);
 	barrier();
 	napi_schedule(cq->napi);
diff --git a/drivers/net/ethernet/mellanox/mlxsw/pci.c b/drivers/net/ethernet/mellanox/mlxsw/pci.c
index c071077aafbd..7992c553c1f5 100644
--- a/drivers/net/ethernet/mellanox/mlxsw/pci.c
+++ b/drivers/net/ethernet/mellanox/mlxsw/pci.c
@@ -215,7 +215,7 @@ mlxsw_pci_queue_elem_info_producer_get(struct mlxsw_pci_queue *q)
 {
 	int index = q->producer_counter & (q->count - 1);
 
-	if ((q->producer_counter - q->consumer_counter) == q->count)
+	if ((u16) (q->producer_counter - q->consumer_counter) == q->count)
 		return NULL;
 	return mlxsw_pci_queue_elem_info_get(q, index);
 }
diff --git a/drivers/net/ethernet/mellanox/mlxsw/port.h b/drivers/net/ethernet/mellanox/mlxsw/port.h
index 726f5435b32f..ae65b9940aed 100644
--- a/drivers/net/ethernet/mellanox/mlxsw/port.h
+++ b/drivers/net/ethernet/mellanox/mlxsw/port.h
@@ -49,7 +49,7 @@
 #define MLXSW_PORT_MID			0xd000
 
 #define MLXSW_PORT_MAX_PHY_PORTS	0x40
-#define MLXSW_PORT_MAX_PORTS		MLXSW_PORT_MAX_PHY_PORTS
+#define MLXSW_PORT_MAX_PORTS		(MLXSW_PORT_MAX_PHY_PORTS + 1)
 
 #define MLXSW_PORT_DEVID_BITS_OFFSET	10
 #define MLXSW_PORT_PHY_BITS_OFFSET	4
diff --git a/drivers/net/ethernet/mellanox/mlxsw/reg.h b/drivers/net/ethernet/mellanox/mlxsw/reg.h
index bb77e2207804..ffe4c0305733 100644
--- a/drivers/net/ethernet/mellanox/mlxsw/reg.h
+++ b/drivers/net/ethernet/mellanox/mlxsw/reg.h
@@ -873,6 +873,62 @@ static inline void mlxsw_reg_spvm_pack(char *payload, u8 local_port,
 	}
 }
 
+/* SPAFT - Switch Port Acceptable Frame Types
+ * ------------------------------------------
+ * The Switch Port Acceptable Frame Types register configures the frame
+ * admittance of the port.
+ */
+#define MLXSW_REG_SPAFT_ID 0x2010
+#define MLXSW_REG_SPAFT_LEN 0x08
+
+static const struct mlxsw_reg_info mlxsw_reg_spaft = {
+	.id = MLXSW_REG_SPAFT_ID,
+	.len = MLXSW_REG_SPAFT_LEN,
+};
+
+/* reg_spaft_local_port
+ * Local port number.
+ * Access: Index
+ *
+ * Note: CPU port is not supported (all tag types are allowed).
+ */
+MLXSW_ITEM32(reg, spaft, local_port, 0x00, 16, 8);
+
+/* reg_spaft_sub_port
+ * Virtual port within the physical port.
+ * Should be set to 0 when virtual ports are not enabled on the port.
+ * Access: RW
+ */
+MLXSW_ITEM32(reg, spaft, sub_port, 0x00, 8, 8);
+
+/* reg_spaft_allow_untagged
+ * When set, untagged frames on the ingress are allowed (default).
+ * Access: RW
+ */
+MLXSW_ITEM32(reg, spaft, allow_untagged, 0x04, 31, 1);
+
+/* reg_spaft_allow_prio_tagged
+ * When set, priority tagged frames on the ingress are allowed (default).
+ * Access: RW
+ */
+MLXSW_ITEM32(reg, spaft, allow_prio_tagged, 0x04, 30, 1);
+
+/* reg_spaft_allow_tagged
+ * When set, tagged frames on the ingress are allowed (default).
+ * Access: RW
+ */
+MLXSW_ITEM32(reg, spaft, allow_tagged, 0x04, 29, 1);
+
+static inline void mlxsw_reg_spaft_pack(char *payload, u8 local_port,
+					bool allow_untagged)
+{
+	MLXSW_REG_ZERO(spaft, payload);
+	mlxsw_reg_spaft_local_port_set(payload, local_port);
+	mlxsw_reg_spaft_allow_untagged_set(payload, allow_untagged);
+	mlxsw_reg_spaft_allow_prio_tagged_set(payload, true);
+	mlxsw_reg_spaft_allow_tagged_set(payload, true);
+}
+
 /* SFGC - Switch Flooding Group Configuration
  * ------------------------------------------
  * The following register controls the association of flooding tables and MIDs
@@ -3203,6 +3259,8 @@ static inline const char *mlxsw_reg_id_str(u16 reg_id)
 		return "SPVID";
 	case MLXSW_REG_SPVM_ID:
 		return "SPVM";
+	case MLXSW_REG_SPAFT_ID:
+		return "SPAFT";
 	case MLXSW_REG_SFGC_ID:
 		return "SFGC";
 	case MLXSW_REG_SFTR_ID:
diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum.c b/drivers/net/ethernet/mellanox/mlxsw/spectrum.c
index 217856bdd400..a94daa8c346c 100644
--- a/drivers/net/ethernet/mellanox/mlxsw/spectrum.c
+++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum.c
@@ -2123,6 +2123,8 @@ static int mlxsw_sp_port_bridge_leave(struct mlxsw_sp_port *mlxsw_sp_port,
 	if (flush_fdb && mlxsw_sp_port_fdb_flush(mlxsw_sp_port))
 		netdev_err(mlxsw_sp_port->dev, "Failed to flush FDB\n");
 
+	mlxsw_sp_port_pvid_set(mlxsw_sp_port, 1);
+
 	mlxsw_sp_port->learning = 0;
 	mlxsw_sp_port->learning_sync = 0;
 	mlxsw_sp_port->uc_flood = 0;
@@ -2356,9 +2358,7 @@ static int mlxsw_sp_port_lag_leave(struct mlxsw_sp_port *mlxsw_sp_port,
 	if (mlxsw_sp_port->bridged) {
 		mlxsw_sp_port_active_vlans_del(mlxsw_sp_port);
 		mlxsw_sp_port_bridge_leave(mlxsw_sp_port, false);
-
-		if (lag->ref_count == 1)
-			mlxsw_sp_master_bridge_dec(mlxsw_sp, NULL);
+		mlxsw_sp_master_bridge_dec(mlxsw_sp, NULL);
 	}
 
 	if (lag->ref_count == 1) {
@@ -2746,6 +2746,13 @@ static int mlxsw_sp_vport_bridge_leave(struct mlxsw_sp_port *mlxsw_sp_vport,
 		goto err_vport_flood_set;
 	}
 
+	err = mlxsw_sp_port_stp_state_set(mlxsw_sp_vport, vid,
+					  MLXSW_REG_SPMS_STATE_FORWARDING);
+	if (err) {
+		netdev_err(dev, "Failed to set STP state\n");
+		goto err_port_stp_state_set;
+	}
+
 	if (flush_fdb && mlxsw_sp_vport_fdb_flush(mlxsw_sp_vport))
 		netdev_err(dev, "Failed to flush FDB\n");
 
@@ -2763,6 +2770,7 @@ static int mlxsw_sp_vport_bridge_leave(struct mlxsw_sp_port *mlxsw_sp_vport,
 
 	return 0;
 
+err_port_stp_state_set:
 err_vport_flood_set:
 err_port_vid_learning_set:
 err_port_vid_to_fid_validate:
diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum.h b/drivers/net/ethernet/mellanox/mlxsw/spectrum.h
index 7f42eb1c320e..3b89ed2f3c76 100644
--- a/drivers/net/ethernet/mellanox/mlxsw/spectrum.h
+++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum.h
@@ -254,5 +254,6 @@ int mlxsw_sp_port_kill_vid(struct net_device *dev,
 int mlxsw_sp_vport_flood_set(struct mlxsw_sp_port *mlxsw_sp_vport, u16 vfid,
 			     bool set, bool only_uc);
 void mlxsw_sp_port_active_vlans_del(struct mlxsw_sp_port *mlxsw_sp_port);
+int mlxsw_sp_port_pvid_set(struct mlxsw_sp_port *mlxsw_sp_port, u16 vid);
 
 #endif
diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum_switchdev.c b/drivers/net/ethernet/mellanox/mlxsw/spectrum_switchdev.c
index e492ca2cdecd..7b56098acc58 100644
--- a/drivers/net/ethernet/mellanox/mlxsw/spectrum_switchdev.c
+++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum_switchdev.c
@@ -370,7 +370,8 @@ static int mlxsw_sp_port_attr_set(struct net_device *dev,
 	return err;
 }
 
-static int mlxsw_sp_port_pvid_set(struct mlxsw_sp_port *mlxsw_sp_port, u16 vid)
+static int __mlxsw_sp_port_pvid_set(struct mlxsw_sp_port *mlxsw_sp_port,
+				    u16 vid)
 {
 	struct mlxsw_sp *mlxsw_sp = mlxsw_sp_port->mlxsw_sp;
 	char spvid_pl[MLXSW_REG_SPVID_LEN];
@@ -379,6 +380,53 @@ static int mlxsw_sp_port_pvid_set(struct mlxsw_sp_port *mlxsw_sp_port, u16 vid)
 	return mlxsw_reg_write(mlxsw_sp->core, MLXSW_REG(spvid), spvid_pl);
 }
 
+static int mlxsw_sp_port_allow_untagged_set(struct mlxsw_sp_port *mlxsw_sp_port,
+					    bool allow)
+{
+	struct mlxsw_sp *mlxsw_sp = mlxsw_sp_port->mlxsw_sp;
+	char spaft_pl[MLXSW_REG_SPAFT_LEN];
+
+	mlxsw_reg_spaft_pack(spaft_pl, mlxsw_sp_port->local_port, allow);
+	return mlxsw_reg_write(mlxsw_sp->core, MLXSW_REG(spaft), spaft_pl);
+}
+
+int mlxsw_sp_port_pvid_set(struct mlxsw_sp_port *mlxsw_sp_port, u16 vid)
+{
+	struct net_device *dev = mlxsw_sp_port->dev;
+	int err;
+
+	if (!vid) {
+		err = mlxsw_sp_port_allow_untagged_set(mlxsw_sp_port, false);
+		if (err) {
+			netdev_err(dev, "Failed to disallow untagged traffic\n");
+			return err;
+		}
+	} else {
+		err = __mlxsw_sp_port_pvid_set(mlxsw_sp_port, vid);
+		if (err) {
+			netdev_err(dev, "Failed to set PVID\n");
+			return err;
+		}
+
+		/* Only allow if not already allowed. */
+		if (!mlxsw_sp_port->pvid) {
+			err = mlxsw_sp_port_allow_untagged_set(mlxsw_sp_port,
+							       true);
+			if (err) {
+				netdev_err(dev, "Failed to allow untagged traffic\n");
+				goto err_port_allow_untagged_set;
+			}
+		}
+	}
+
+	mlxsw_sp_port->pvid = vid;
+	return 0;
+
+err_port_allow_untagged_set:
+	__mlxsw_sp_port_pvid_set(mlxsw_sp_port, mlxsw_sp_port->pvid);
+	return err;
+}
+
 static int mlxsw_sp_fid_create(struct mlxsw_sp *mlxsw_sp, u16 fid)
 {
 	char sfmr_pl[MLXSW_REG_SFMR_LEN];
@@ -540,7 +588,12 @@ static int __mlxsw_sp_port_vlans_add(struct mlxsw_sp_port *mlxsw_sp_port,
 			netdev_err(dev, "Unable to add PVID %d\n", vid_begin);
 			goto err_port_pvid_set;
 		}
-		mlxsw_sp_port->pvid = vid_begin;
+	} else if (!flag_pvid && old_pvid >= vid_begin && old_pvid <= vid_end) {
+		err = mlxsw_sp_port_pvid_set(mlxsw_sp_port, 0);
+		if (err) {
+			netdev_err(dev, "Unable to del PVID\n");
+			goto err_port_pvid_set;
+		}
 	}
 
 	/* Changing activity bits only if HW operation succeded */
@@ -892,20 +945,18 @@ static int __mlxsw_sp_port_vlans_del(struct mlxsw_sp_port *mlxsw_sp_port,
 		return err;
 	}
 
+	if (init)
+		goto out;
+
 	pvid = mlxsw_sp_port->pvid;
-	if (pvid >= vid_begin && pvid <= vid_end && pvid != 1) {
-		/* Default VLAN is always 1 */
-		err = mlxsw_sp_port_pvid_set(mlxsw_sp_port, 1);
+	if (pvid >= vid_begin && pvid <= vid_end) {
+		err = mlxsw_sp_port_pvid_set(mlxsw_sp_port, 0);
 		if (err) {
 			netdev_err(dev, "Unable to del PVID %d\n", pvid);
 			return err;
 		}
-		mlxsw_sp_port->pvid = 1;
 	}
 
-	if (init)
-		goto out;
-
 	err = __mlxsw_sp_port_flood_set(mlxsw_sp_port, vid_begin, vid_end,
 					false, false);
 	if (err) {
diff --git a/drivers/net/ethernet/moxa/moxart_ether.c b/drivers/net/ethernet/moxa/moxart_ether.c
index 00cfd95ca59d..3e67f451f2ab 100644
--- a/drivers/net/ethernet/moxa/moxart_ether.c
+++ b/drivers/net/ethernet/moxa/moxart_ether.c
@@ -474,9 +474,9 @@ static int moxart_mac_probe(struct platform_device *pdev)
 	res = platform_get_resource(pdev, IORESOURCE_MEM, 0);
 	ndev->base_addr = res->start;
 	priv->base = devm_ioremap_resource(p_dev, res);
-	ret = IS_ERR(priv->base);
-	if (ret) {
+	if (IS_ERR(priv->base)) {
 		dev_err(p_dev, "devm_ioremap_resource failed\n");
+		ret = PTR_ERR(priv->base);
 		goto init_fail;
 	}
 
diff --git a/drivers/net/ethernet/qualcomm/qca_spi.c b/drivers/net/ethernet/qualcomm/qca_spi.c
index 689a4a5c8dcf..1ef03939d25f 100644
--- a/drivers/net/ethernet/qualcomm/qca_spi.c
+++ b/drivers/net/ethernet/qualcomm/qca_spi.c
@@ -811,7 +811,7 @@ qcaspi_netdev_setup(struct net_device *dev)
 	dev->netdev_ops = &qcaspi_netdev_ops;
 	qcaspi_set_ethtool_ops(dev);
 	dev->watchdog_timeo = QCASPI_TX_TIMEOUT;
-	dev->flags = IFF_MULTICAST;
+	dev->priv_flags &= ~IFF_TX_SKB_SHARING;
 	dev->tx_queue_len = 100;
 
 	qca = netdev_priv(dev);
diff --git a/drivers/net/ethernet/realtek/r8169.c b/drivers/net/ethernet/realtek/r8169.c
index 17d5571d0432..dd2cf3738b73 100644
--- a/drivers/net/ethernet/realtek/r8169.c
+++ b/drivers/net/ethernet/realtek/r8169.c
@@ -4933,8 +4933,6 @@ static void rtl_init_rxcfg(struct rtl8169_private *tp)
 		RTL_W32(RxConfig, RX128_INT_EN | RX_MULTI_EN | RX_DMA_BURST);
 		break;
 	case RTL_GIGA_MAC_VER_40:
-		RTL_W32(RxConfig, RX128_INT_EN | RX_MULTI_EN | RX_DMA_BURST | RX_EARLY_OFF);
-		break;
 	case RTL_GIGA_MAC_VER_41:
 	case RTL_GIGA_MAC_VER_42:
 	case RTL_GIGA_MAC_VER_43:
@@ -4943,8 +4941,6 @@ static void rtl_init_rxcfg(struct rtl8169_private *tp)
 	case RTL_GIGA_MAC_VER_46:
 	case RTL_GIGA_MAC_VER_47:
 	case RTL_GIGA_MAC_VER_48:
-		RTL_W32(RxConfig, RX128_INT_EN | RX_DMA_BURST | RX_EARLY_OFF);
-		break;
 	case RTL_GIGA_MAC_VER_49:
 	case RTL_GIGA_MAC_VER_50:
 	case RTL_GIGA_MAC_VER_51:
@@ -6137,28 +6133,28 @@ static void rtl_hw_start_8168h_1(struct rtl8169_private *tp)
 		sw_cnt_1ms_ini = 16000000/rg_saw_cnt;
 		sw_cnt_1ms_ini &= 0x0fff;
 		data = r8168_mac_ocp_read(tp, 0xd412);
-		data &= 0x0fff;
+		data &= ~0x0fff;
 		data |= sw_cnt_1ms_ini;
 		r8168_mac_ocp_write(tp, 0xd412, data);
 	}
 
 	data = r8168_mac_ocp_read(tp, 0xe056);
-	data &= 0xf0;
-	data |= 0x07;
+	data &= ~0xf0;
+	data |= 0x70;
 	r8168_mac_ocp_write(tp, 0xe056, data);
 
 	data = r8168_mac_ocp_read(tp, 0xe052);
-	data &= 0x8008;
-	data |= 0x6000;
+	data &= ~0x6000;
+	data |= 0x8008;
 	r8168_mac_ocp_write(tp, 0xe052, data);
 
 	data = r8168_mac_ocp_read(tp, 0xe0d6);
-	data &= 0x01ff;
+	data &= ~0x01ff;
 	data |= 0x017f;
 	r8168_mac_ocp_write(tp, 0xe0d6, data);
 
 	data = r8168_mac_ocp_read(tp, 0xd420);
-	data &= 0x0fff;
+	data &= ~0x0fff;
 	data |= 0x047f;
 	r8168_mac_ocp_write(tp, 0xd420, data);
 
@@ -7730,10 +7726,13 @@ rtl8169_get_stats64(struct net_device *dev, struct rtnl_link_stats64 *stats)
 {
 	struct rtl8169_private *tp = netdev_priv(dev);
 	void __iomem *ioaddr = tp->mmio_addr;
+	struct pci_dev *pdev = tp->pci_dev;
 	struct rtl8169_counters *counters = tp->counters;
 	unsigned int start;
 
-	if (netif_running(dev))
+	pm_runtime_get_noresume(&pdev->dev);
+
+	if (netif_running(dev) && pm_runtime_active(&pdev->dev))
 		rtl8169_rx_missed(dev, ioaddr);
 
 	do {
@@ -7761,7 +7760,8 @@ rtl8169_get_stats64(struct net_device *dev, struct rtnl_link_stats64 *stats)
 	 * Fetch additonal counter values missing in stats collected by driver
 	 * from tally counters.
 	 */
-	rtl8169_update_counters(dev);
+	if (pm_runtime_active(&pdev->dev))
+		rtl8169_update_counters(dev);
 
 	/*
 	 * Subtract values fetched during initalization.
@@ -7774,6 +7774,8 @@ rtl8169_get_stats64(struct net_device *dev, struct rtnl_link_stats64 *stats)
 	stats->tx_aborted_errors = le16_to_cpu(counters->tx_aborted) -
 		le16_to_cpu(tp->tc_offset.tx_aborted);
 
+	pm_runtime_put_noidle(&pdev->dev);
+
 	return stats;
 }
 
@@ -7853,6 +7855,10 @@ static int rtl8169_runtime_suspend(struct device *device)
 
 	rtl8169_net_suspend(dev);
 
+	/* Update counters before going runtime suspend */
+	rtl8169_rx_missed(dev, tp->mmio_addr);
+	rtl8169_update_counters(dev);
+
 	return 0;
 }
 
diff --git a/drivers/net/ethernet/renesas/ravb_main.c b/drivers/net/ethernet/renesas/ravb_main.c
index ac43ed914fcf..86449c357168 100644
--- a/drivers/net/ethernet/renesas/ravb_main.c
+++ b/drivers/net/ethernet/renesas/ravb_main.c
@@ -1139,7 +1139,8 @@ static int ravb_set_ringparam(struct net_device *ndev,
 	if (netif_running(ndev)) {
 		netif_device_detach(ndev);
 		/* Stop PTP Clock driver */
-		ravb_ptp_stop(ndev);
+		if (priv->chip_id == RCAR_GEN2)
+			ravb_ptp_stop(ndev);
 		/* Wait for DMA stopping */
 		error = ravb_stop_dma(ndev);
 		if (error) {
@@ -1170,7 +1171,8 @@ static int ravb_set_ringparam(struct net_device *ndev,
 		ravb_emac_init(ndev);
 
 		/* Initialise PTP Clock driver */
-		ravb_ptp_init(ndev, priv->pdev);
+		if (priv->chip_id == RCAR_GEN2)
+			ravb_ptp_init(ndev, priv->pdev);
 
 		netif_device_attach(ndev);
 	}
@@ -1298,7 +1300,8 @@ static void ravb_tx_timeout_work(struct work_struct *work)
 	netif_tx_stop_all_queues(ndev);
 
 	/* Stop PTP Clock driver */
-	ravb_ptp_stop(ndev);
+	if (priv->chip_id == RCAR_GEN2)
+		ravb_ptp_stop(ndev);
 
 	/* Wait for DMA stopping */
 	ravb_stop_dma(ndev);
@@ -1311,7 +1314,8 @@ static void ravb_tx_timeout_work(struct work_struct *work)
 	ravb_emac_init(ndev);
 
 	/* Initialise PTP Clock driver */
-	ravb_ptp_init(ndev, priv->pdev);
+	if (priv->chip_id == RCAR_GEN2)
+		ravb_ptp_init(ndev, priv->pdev);
 
 	netif_tx_start_all_queues(ndev);
 }
@@ -1718,7 +1722,6 @@ static int ravb_set_gti(struct net_device *ndev)
 static int ravb_probe(struct platform_device *pdev)
 {
 	struct device_node *np = pdev->dev.of_node;
-	const struct of_device_id *match;
 	struct ravb_private *priv;
 	enum ravb_chip_id chip_id;
 	struct net_device *ndev;
@@ -1750,8 +1753,7 @@ static int ravb_probe(struct platform_device *pdev)
 	ndev->base_addr = res->start;
 	ndev->dma = -1;
 
-	match = of_match_device(of_match_ptr(ravb_match_table), &pdev->dev);
-	chip_id = (enum ravb_chip_id)match->data;
+	chip_id = (enum ravb_chip_id)of_device_get_match_data(&pdev->dev);
 
 	if (chip_id == RCAR_GEN3)
 		irq = platform_get_irq_byname(pdev, "ch22");
@@ -1814,10 +1816,6 @@ static int ravb_probe(struct platform_device *pdev)
 			   CCC_OPC_CONFIG | CCC_GAC | CCC_CSEL_HPB, CCC);
 	}
 
-	/* Set CSEL value */
-	ravb_write(ndev, (ravb_read(ndev, CCC) & ~CCC_CSEL) | CCC_CSEL_HPB,
-		   CCC);
-
 	/* Set GTI value */
 	error = ravb_set_gti(ndev);
 	if (error)
diff --git a/drivers/net/ethernet/renesas/sh_eth.c b/drivers/net/ethernet/renesas/sh_eth.c
index dfa9e59c9442..738449992876 100644
--- a/drivers/net/ethernet/renesas/sh_eth.c
+++ b/drivers/net/ethernet/renesas/sh_eth.c
@@ -3061,15 +3061,11 @@ static int sh_eth_drv_probe(struct platform_device *pdev)
 	mdp->ether_link_active_low = pd->ether_link_active_low;
 
 	/* set cpu data */
-	if (id) {
+	if (id)
 		mdp->cd = (struct sh_eth_cpu_data *)id->driver_data;
-	} else	{
-		const struct of_device_id *match;
+	else
+		mdp->cd = (struct sh_eth_cpu_data *)of_device_get_match_data(&pdev->dev);
 
-		match = of_match_device(of_match_ptr(sh_eth_match_table),
-					&pdev->dev);
-		mdp->cd = (struct sh_eth_cpu_data *)match->data;
-	}
 	mdp->reg_offset = sh_eth_get_register_offset(mdp->cd->register_type);
 	if (!mdp->reg_offset) {
 		dev_err(&pdev->dev, "Unknown register type (%d)\n",
diff --git a/drivers/net/ethernet/smsc/smc91x.c b/drivers/net/ethernet/smsc/smc91x.c
index 0e2fc1a844ab..db7db8ac4ca3 100644
--- a/drivers/net/ethernet/smsc/smc91x.c
+++ b/drivers/net/ethernet/smsc/smc91x.c
@@ -2342,8 +2342,8 @@ static int smc_drv_probe(struct platform_device *pdev)
 	}
 
 	ndev->irq = platform_get_irq(pdev, 0);
-	if (ndev->irq <= 0) {
-		ret = -ENODEV;
+	if (ndev->irq < 0) {
+		ret = ndev->irq;
 		goto out_release_io;
 	}
 	/*
diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac_mdio.c b/drivers/net/ethernet/stmicro/stmmac/stmmac_mdio.c
index 0faf16336035..efb54f356a67 100644
--- a/drivers/net/ethernet/stmicro/stmmac/stmmac_mdio.c
+++ b/drivers/net/ethernet/stmicro/stmmac/stmmac_mdio.c
@@ -199,21 +199,12 @@ int stmmac_mdio_register(struct net_device *ndev)
 	struct stmmac_priv *priv = netdev_priv(ndev);
 	struct stmmac_mdio_bus_data *mdio_bus_data = priv->plat->mdio_bus_data;
 	int addr, found;
-	struct device_node *mdio_node = NULL;
-	struct device_node *child_node = NULL;
+	struct device_node *mdio_node = priv->plat->mdio_node;
 
 	if (!mdio_bus_data)
 		return 0;
 
 	if (IS_ENABLED(CONFIG_OF)) {
-		for_each_child_of_node(priv->device->of_node, child_node) {
-			if (of_device_is_compatible(child_node,
-						    "snps,dwmac-mdio")) {
-				mdio_node = child_node;
-				break;
-			}
-		}
-
 		if (mdio_node) {
 			netdev_dbg(ndev, "FOUND MDIO subnode\n");
 		} else {
diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac_platform.c b/drivers/net/ethernet/stmicro/stmmac/stmmac_platform.c
index 6a52fa18cbf2..4514ba73d961 100644
--- a/drivers/net/ethernet/stmicro/stmmac/stmmac_platform.c
+++ b/drivers/net/ethernet/stmicro/stmmac/stmmac_platform.c
@@ -110,6 +110,7 @@ stmmac_probe_config_dt(struct platform_device *pdev, const char **mac)
 	struct device_node *np = pdev->dev.of_node;
 	struct plat_stmmacenet_data *plat;
 	struct stmmac_dma_cfg *dma_cfg;
+	struct device_node *child_node = NULL;
 
 	plat = devm_kzalloc(&pdev->dev, sizeof(*plat), GFP_KERNEL);
 	if (!plat)
@@ -140,13 +141,19 @@ stmmac_probe_config_dt(struct platform_device *pdev, const char **mac)
 		plat->phy_node = of_node_get(np);
 	}
 
+	for_each_child_of_node(np, child_node)
+		if (of_device_is_compatible(child_node,	"snps,dwmac-mdio")) {
+			plat->mdio_node = child_node;
+			break;
+		}
+
 	/* "snps,phy-addr" is not a standard property. Mark it as deprecated
 	 * and warn of its use. Remove this when phy node support is added.
 	 */
 	if (of_property_read_u32(np, "snps,phy-addr", &plat->phy_addr) == 0)
 		dev_warn(&pdev->dev, "snps,phy-addr property is deprecated\n");
 
-	if ((plat->phy_node && !of_phy_is_fixed_link(np)) || plat->phy_bus_name)
+	if ((plat->phy_node && !of_phy_is_fixed_link(np)) || !plat->mdio_node)
 		plat->mdio_bus_data = NULL;
 	else
 		plat->mdio_bus_data =
diff --git a/drivers/net/ethernet/synopsys/dwc_eth_qos.c b/drivers/net/ethernet/synopsys/dwc_eth_qos.c
index 70814b7386b3..af11ed1e0bcc 100644
--- a/drivers/net/ethernet/synopsys/dwc_eth_qos.c
+++ b/drivers/net/ethernet/synopsys/dwc_eth_qos.c
@@ -426,7 +426,7 @@
 #define DWC_MMC_RXOCTETCOUNT_GB          0x0784
 #define DWC_MMC_RXPACKETCOUNT_GB         0x0780
 
-static int debug = 3;
+static int debug = -1;
 module_param(debug, int, 0);
 MODULE_PARM_DESC(debug, "DWC_eth_qos debug level (0=none,...,16=all)");
 
@@ -650,6 +650,11 @@ struct net_local {
 	u32 mmc_tx_counters_mask;
 
 	struct dwceqos_flowcontrol flowcontrol;
+
+	/* Tracks the intermediate state of phy started but hardware
+	 * init not finished yet.
+	 */
+	bool phy_defer;
 };
 
 static void dwceqos_read_mmc_counters(struct net_local *lp, u32 rx_mask,
@@ -901,6 +906,9 @@ static void dwceqos_adjust_link(struct net_device *ndev)
 	struct phy_device *phydev = lp->phy_dev;
 	int status_change = 0;
 
+	if (lp->phy_defer)
+		return;
+
 	if (phydev->link) {
 		if ((lp->speed != phydev->speed) ||
 		    (lp->duplex != phydev->duplex)) {
@@ -1113,7 +1121,7 @@ static int dwceqos_descriptor_init(struct net_local *lp)
 	/* Allocate DMA descriptors */
 	size = DWCEQOS_RX_DCNT * sizeof(struct dwceqos_dma_desc);
 	lp->rx_descs = dma_alloc_coherent(lp->ndev->dev.parent, size,
-			&lp->rx_descs_addr, 0);
+			&lp->rx_descs_addr, GFP_KERNEL);
 	if (!lp->rx_descs)
 		goto err_out;
 	lp->rx_descs_tail_addr = lp->rx_descs_addr +
@@ -1121,7 +1129,7 @@ static int dwceqos_descriptor_init(struct net_local *lp)
 
 	size = DWCEQOS_TX_DCNT * sizeof(struct dwceqos_dma_desc);
 	lp->tx_descs = dma_alloc_coherent(lp->ndev->dev.parent, size,
-			&lp->tx_descs_addr, 0);
+			&lp->tx_descs_addr, GFP_KERNEL);
 	if (!lp->tx_descs)
 		goto err_out;
 	lp->tx_descs_tail_addr = lp->tx_descs_addr +
@@ -1635,6 +1643,12 @@ static void dwceqos_init_hw(struct net_local *lp)
 	regval = dwceqos_read(lp, REG_DWCEQOS_MAC_CFG);
 	dwceqos_write(lp, REG_DWCEQOS_MAC_CFG,
 		      regval | DWCEQOS_MAC_CFG_TE | DWCEQOS_MAC_CFG_RE);
+
+	lp->phy_defer = false;
+	mutex_lock(&lp->phy_dev->lock);
+	phy_read_status(lp->phy_dev);
+	dwceqos_adjust_link(lp->ndev);
+	mutex_unlock(&lp->phy_dev->lock);
 }
 
 static void dwceqos_tx_reclaim(unsigned long data)
@@ -1880,9 +1894,13 @@ static int dwceqos_open(struct net_device *ndev)
 	}
 	netdev_reset_queue(ndev);
 
-	napi_enable(&lp->napi);
+	/* The dwceqos reset state machine requires all phy clocks to complete,
+	 * hence the unusual init order with phy_start first.
+	 */
+	lp->phy_defer = true;
 	phy_start(lp->phy_dev);
 	dwceqos_init_hw(lp);
+	napi_enable(&lp->napi);
 
 	netif_start_queue(ndev);
 	tasklet_enable(&lp->tx_bdreclaim_tasklet);
@@ -1915,18 +1933,19 @@ static int dwceqos_stop(struct net_device *ndev)
 {
 	struct net_local *lp = netdev_priv(ndev);
 
-	phy_stop(lp->phy_dev);
-
 	tasklet_disable(&lp->tx_bdreclaim_tasklet);
-	netif_stop_queue(ndev);
 	napi_disable(&lp->napi);
 
-	dwceqos_drain_dma(lp);
+	/* Stop all tx before we drain the tx dma. */
+	netif_tx_lock_bh(lp->ndev);
+	netif_stop_queue(ndev);
+	netif_tx_unlock_bh(lp->ndev);
 
-	netif_tx_lock(lp->ndev);
+	dwceqos_drain_dma(lp);
 	dwceqos_reset_hw(lp);
+	phy_stop(lp->phy_dev);
+
 	dwceqos_descriptor_free(lp);
-	netif_tx_unlock(lp->ndev);
 
 	return 0;
 }
@@ -2178,12 +2197,10 @@ static int dwceqos_start_xmit(struct sk_buff *skb, struct net_device *ndev)
 		((trans.initial_descriptor + trans.nr_descriptors) %
 		 DWCEQOS_TX_DCNT));
 
-	dwceqos_tx_finalize(skb, lp, &trans);
-
-	netdev_sent_queue(ndev, skb->len);
-
 	spin_lock_bh(&lp->tx_lock);
 	lp->tx_free -= trans.nr_descriptors;
+	dwceqos_tx_finalize(skb, lp, &trans);
+	netdev_sent_queue(ndev, skb->len);
 	spin_unlock_bh(&lp->tx_lock);
 
 	ndev->trans_start = jiffies;
diff --git a/drivers/net/ethernet/ti/cpsw-phy-sel.c b/drivers/net/ethernet/ti/cpsw-phy-sel.c
index e9cc61e1ec74..c3e85acfdc70 100644
--- a/drivers/net/ethernet/ti/cpsw-phy-sel.c
+++ b/drivers/net/ethernet/ti/cpsw-phy-sel.c
@@ -63,8 +63,12 @@ static void cpsw_gmii_sel_am3352(struct cpsw_phy_sel_priv *priv,
 		mode = AM33XX_GMII_SEL_MODE_RGMII;
 		break;
 
-	case PHY_INTERFACE_MODE_MII:
 	default:
+		dev_warn(priv->dev,
+			 "Unsupported PHY mode: \"%s\". Defaulting to MII.\n",
+			phy_modes(phy_mode));
+		/* fallthrough */
+	case PHY_INTERFACE_MODE_MII:
 		mode = AM33XX_GMII_SEL_MODE_MII;
 		break;
 	};
@@ -106,8 +110,12 @@ static void cpsw_gmii_sel_dra7xx(struct cpsw_phy_sel_priv *priv,
 		mode = AM33XX_GMII_SEL_MODE_RGMII;
 		break;
 
-	case PHY_INTERFACE_MODE_MII:
 	default:
+		dev_warn(priv->dev,
+			 "Unsupported PHY mode: \"%s\". Defaulting to MII.\n",
+			phy_modes(phy_mode));
+		/* fallthrough */
+	case PHY_INTERFACE_MODE_MII:
 		mode = AM33XX_GMII_SEL_MODE_MII;
 		break;
 	};
diff --git a/drivers/net/ethernet/ti/netcp_core.c b/drivers/net/ethernet/ti/netcp_core.c
index c61d66d38634..029841f98c32 100644
--- a/drivers/net/ethernet/ti/netcp_core.c
+++ b/drivers/net/ethernet/ti/netcp_core.c
@@ -117,21 +117,17 @@ static void get_pkt_info(dma_addr_t *buff, u32 *buff_len, dma_addr_t *ndesc,
 	*ndesc = le32_to_cpu(desc->next_desc);
 }
 
-static void get_pad_info(u32 *pad0, u32 *pad1, u32 *pad2, struct knav_dma_desc *desc)
+static u32 get_sw_data(int index, struct knav_dma_desc *desc)
 {
-	*pad0 = le32_to_cpu(desc->pad[0]);
-	*pad1 = le32_to_cpu(desc->pad[1]);
-	*pad2 = le32_to_cpu(desc->pad[2]);
+	/* No Endian conversion needed as this data is untouched by hw */
+	return desc->sw_data[index];
 }
 
-static void get_pad_ptr(void **padptr, struct knav_dma_desc *desc)
-{
-	u64 pad64;
-
-	pad64 = le32_to_cpu(desc->pad[0]) +
-		((u64)le32_to_cpu(desc->pad[1]) << 32);
-	*padptr = (void *)(uintptr_t)pad64;
-}
+/* use these macros to get sw data */
+#define GET_SW_DATA0(desc) get_sw_data(0, desc)
+#define GET_SW_DATA1(desc) get_sw_data(1, desc)
+#define GET_SW_DATA2(desc) get_sw_data(2, desc)
+#define GET_SW_DATA3(desc) get_sw_data(3, desc)
 
 static void get_org_pkt_info(dma_addr_t *buff, u32 *buff_len,
 			     struct knav_dma_desc *desc)
@@ -163,13 +159,18 @@ static void set_desc_info(u32 desc_info, u32 pkt_info,
 	desc->packet_info = cpu_to_le32(pkt_info);
 }
 
-static void set_pad_info(u32 pad0, u32 pad1, u32 pad2, struct knav_dma_desc *desc)
+static void set_sw_data(int index, u32 data, struct knav_dma_desc *desc)
 {
-	desc->pad[0] = cpu_to_le32(pad0);
-	desc->pad[1] = cpu_to_le32(pad1);
-	desc->pad[2] = cpu_to_le32(pad1);
+	/* No Endian conversion needed as this data is untouched by hw */
+	desc->sw_data[index] = data;
 }
 
+/* use these macros to set sw data */
+#define SET_SW_DATA0(data, desc) set_sw_data(0, data, desc)
+#define SET_SW_DATA1(data, desc) set_sw_data(1, data, desc)
+#define SET_SW_DATA2(data, desc) set_sw_data(2, data, desc)
+#define SET_SW_DATA3(data, desc) set_sw_data(3, data, desc)
+
 static void set_org_pkt_info(dma_addr_t buff, u32 buff_len,
 			     struct knav_dma_desc *desc)
 {
@@ -581,7 +582,6 @@ static void netcp_free_rx_desc_chain(struct netcp_intf *netcp,
 	dma_addr_t dma_desc, dma_buf;
 	unsigned int buf_len, dma_sz = sizeof(*ndesc);
 	void *buf_ptr;
-	u32 pad[2];
 	u32 tmp;
 
 	get_words(&dma_desc, 1, &desc->next_desc);
@@ -593,14 +593,20 @@ static void netcp_free_rx_desc_chain(struct netcp_intf *netcp,
 			break;
 		}
 		get_pkt_info(&dma_buf, &tmp, &dma_desc, ndesc);
-		get_pad_ptr(&buf_ptr, ndesc);
+		/* warning!!!! We are retrieving the virtual ptr in the sw_data
+		 * field as a 32bit value. Will not work on 64bit machines
+		 */
+		buf_ptr = (void *)GET_SW_DATA0(ndesc);
+		buf_len = (int)GET_SW_DATA1(desc);
 		dma_unmap_page(netcp->dev, dma_buf, PAGE_SIZE, DMA_FROM_DEVICE);
 		__free_page(buf_ptr);
 		knav_pool_desc_put(netcp->rx_pool, desc);
 	}
-
-	get_pad_info(&pad[0], &pad[1], &buf_len, desc);
-	buf_ptr = (void *)(uintptr_t)(pad[0] + ((u64)pad[1] << 32));
+	/* warning!!!! We are retrieving the virtual ptr in the sw_data
+	 * field as a 32bit value. Will not work on 64bit machines
+	 */
+	buf_ptr = (void *)GET_SW_DATA0(desc);
+	buf_len = (int)GET_SW_DATA1(desc);
 
 	if (buf_ptr)
 		netcp_frag_free(buf_len <= PAGE_SIZE, buf_ptr);
@@ -639,7 +645,6 @@ static int netcp_process_one_rx_packet(struct netcp_intf *netcp)
 	dma_addr_t dma_desc, dma_buff;
 	struct netcp_packet p_info;
 	struct sk_buff *skb;
-	u32 pad[2];
 	void *org_buf_ptr;
 
 	dma_desc = knav_queue_pop(netcp->rx_queue, &dma_sz);
@@ -653,8 +658,11 @@ static int netcp_process_one_rx_packet(struct netcp_intf *netcp)
 	}
 
 	get_pkt_info(&dma_buff, &buf_len, &dma_desc, desc);
-	get_pad_info(&pad[0], &pad[1], &org_buf_len, desc);
-	org_buf_ptr = (void *)(uintptr_t)(pad[0] + ((u64)pad[1] << 32));
+	/* warning!!!! We are retrieving the virtual ptr in the sw_data
+	 * field as a 32bit value. Will not work on 64bit machines
+	 */
+	org_buf_ptr = (void *)GET_SW_DATA0(desc);
+	org_buf_len = (int)GET_SW_DATA1(desc);
 
 	if (unlikely(!org_buf_ptr)) {
 		dev_err(netcp->ndev_dev, "NULL bufptr in desc\n");
@@ -679,7 +687,6 @@ static int netcp_process_one_rx_packet(struct netcp_intf *netcp)
 	/* Fill in the page fragment list */
 	while (dma_desc) {
 		struct page *page;
-		void *ptr;
 
 		ndesc = knav_pool_desc_unmap(netcp->rx_pool, dma_desc, dma_sz);
 		if (unlikely(!ndesc)) {
@@ -688,8 +695,10 @@ static int netcp_process_one_rx_packet(struct netcp_intf *netcp)
 		}
 
 		get_pkt_info(&dma_buff, &buf_len, &dma_desc, ndesc);
-		get_pad_ptr(&ptr, ndesc);
-		page = ptr;
+		/* warning!!!! We are retrieving the virtual ptr in the sw_data
+		 * field as a 32bit value. Will not work on 64bit machines
+		 */
+		page = (struct page *)GET_SW_DATA0(desc);
 
 		if (likely(dma_buff && buf_len && page)) {
 			dma_unmap_page(netcp->dev, dma_buff, PAGE_SIZE,
@@ -777,7 +786,10 @@ static void netcp_free_rx_buf(struct netcp_intf *netcp, int fdq)
 		}
 
 		get_org_pkt_info(&dma, &buf_len, desc);
-		get_pad_ptr(&buf_ptr, desc);
+		/* warning!!!! We are retrieving the virtual ptr in the sw_data
+		 * field as a 32bit value. Will not work on 64bit machines
+		 */
+		buf_ptr = (void *)GET_SW_DATA0(desc);
 
 		if (unlikely(!dma)) {
 			dev_err(netcp->ndev_dev, "NULL orig_buff in desc\n");
@@ -829,7 +841,7 @@ static int netcp_allocate_rx_buf(struct netcp_intf *netcp, int fdq)
 	struct page *page;
 	dma_addr_t dma;
 	void *bufptr;
-	u32 pad[3];
+	u32 sw_data[2];
 
 	/* Allocate descriptor */
 	hwdesc = knav_pool_desc_get(netcp->rx_pool);
@@ -846,7 +858,7 @@ static int netcp_allocate_rx_buf(struct netcp_intf *netcp, int fdq)
 				SKB_DATA_ALIGN(sizeof(struct skb_shared_info));
 
 		bufptr = netdev_alloc_frag(primary_buf_len);
-		pad[2] = primary_buf_len;
+		sw_data[1] = primary_buf_len;
 
 		if (unlikely(!bufptr)) {
 			dev_warn_ratelimited(netcp->ndev_dev,
@@ -858,9 +870,10 @@ static int netcp_allocate_rx_buf(struct netcp_intf *netcp, int fdq)
 		if (unlikely(dma_mapping_error(netcp->dev, dma)))
 			goto fail;
 
-		pad[0] = lower_32_bits((uintptr_t)bufptr);
-		pad[1] = upper_32_bits((uintptr_t)bufptr);
-
+		/* warning!!!! We are saving the virtual ptr in the sw_data
+		 * field as a 32bit value. Will not work on 64bit machines
+		 */
+		sw_data[0] = (u32)bufptr;
 	} else {
 		/* Allocate a secondary receive queue entry */
 		page = alloc_page(GFP_ATOMIC | GFP_DMA | __GFP_COLD);
@@ -870,9 +883,11 @@ static int netcp_allocate_rx_buf(struct netcp_intf *netcp, int fdq)
 		}
 		buf_len = PAGE_SIZE;
 		dma = dma_map_page(netcp->dev, page, 0, buf_len, DMA_TO_DEVICE);
-		pad[0] = lower_32_bits(dma);
-		pad[1] = upper_32_bits(dma);
-		pad[2] = 0;
+		/* warning!!!! We are saving the virtual ptr in the sw_data
+		 * field as a 32bit value. Will not work on 64bit machines
+		 */
+		sw_data[0] = (u32)page;
+		sw_data[1] = 0;
 	}
 
 	desc_info =  KNAV_DMA_DESC_PS_INFO_IN_DESC;
@@ -882,7 +897,8 @@ static int netcp_allocate_rx_buf(struct netcp_intf *netcp, int fdq)
 	pkt_info |= (netcp->rx_queue_id & KNAV_DMA_DESC_RETQ_MASK) <<
 		    KNAV_DMA_DESC_RETQ_SHIFT;
 	set_org_pkt_info(dma, buf_len, hwdesc);
-	set_pad_info(pad[0], pad[1], pad[2], hwdesc);
+	SET_SW_DATA0(sw_data[0], hwdesc);
+	SET_SW_DATA1(sw_data[1], hwdesc);
 	set_desc_info(desc_info, pkt_info, hwdesc);
 
 	/* Push to FDQs */
@@ -971,7 +987,6 @@ static int netcp_process_tx_compl_packets(struct netcp_intf *netcp,
 					  unsigned int budget)
 {
 	struct knav_dma_desc *desc;
-	void *ptr;
 	struct sk_buff *skb;
 	unsigned int dma_sz;
 	dma_addr_t dma;
@@ -988,8 +1003,10 @@ static int netcp_process_tx_compl_packets(struct netcp_intf *netcp,
 			continue;
 		}
 
-		get_pad_ptr(&ptr, desc);
-		skb = ptr;
+		/* warning!!!! We are retrieving the virtual ptr in the sw_data
+		 * field as a 32bit value. Will not work on 64bit machines
+		 */
+		skb = (struct sk_buff *)GET_SW_DATA0(desc);
 		netcp_free_tx_desc_chain(netcp, desc, dma_sz);
 		if (!skb) {
 			dev_err(netcp->ndev_dev, "No skb in Tx desc\n");
@@ -1194,10 +1211,10 @@ static int netcp_tx_submit_skb(struct netcp_intf *netcp,
 	}
 
 	set_words(&tmp, 1, &desc->packet_info);
-	tmp = lower_32_bits((uintptr_t)&skb);
-	set_words(&tmp, 1, &desc->pad[0]);
-	tmp = upper_32_bits((uintptr_t)&skb);
-	set_words(&tmp, 1, &desc->pad[1]);
+	/* warning!!!! We are saving the virtual ptr in the sw_data
+	 * field as a 32bit value. Will not work on 64bit machines
+	 */
+	SET_SW_DATA0((u32)skb, desc);
 
 	if (tx_pipe->flags & SWITCH_TO_PORT_IN_TAGINFO) {
 		tmp = tx_pipe->switch_to_port;
diff --git a/drivers/net/geneve.c b/drivers/net/geneve.c
index 0b14ac3b8d11..0bf7edd99573 100644
--- a/drivers/net/geneve.c
+++ b/drivers/net/geneve.c
@@ -1039,6 +1039,34 @@ static netdev_tx_t geneve_xmit(struct sk_buff *skb, struct net_device *dev)
 	return geneve_xmit_skb(skb, dev, info);
 }
 
+static int __geneve_change_mtu(struct net_device *dev, int new_mtu, bool strict)
+{
+	/* The max_mtu calculation does not take account of GENEVE
+	 * options, to avoid excluding potentially valid
+	 * configurations.
+	 */
+	int max_mtu = IP_MAX_MTU - GENEVE_BASE_HLEN - sizeof(struct iphdr)
+		- dev->hard_header_len;
+
+	if (new_mtu < 68)
+		return -EINVAL;
+
+	if (new_mtu > max_mtu) {
+		if (strict)
+			return -EINVAL;
+
+		new_mtu = max_mtu;
+	}
+
+	dev->mtu = new_mtu;
+	return 0;
+}
+
+static int geneve_change_mtu(struct net_device *dev, int new_mtu)
+{
+	return __geneve_change_mtu(dev, new_mtu, true);
+}
+
 static int geneve_fill_metadata_dst(struct net_device *dev, struct sk_buff *skb)
 {
 	struct ip_tunnel_info *info = skb_tunnel_info(skb);
@@ -1083,7 +1111,7 @@ static const struct net_device_ops geneve_netdev_ops = {
 	.ndo_stop		= geneve_stop,
 	.ndo_start_xmit		= geneve_xmit,
 	.ndo_get_stats64	= ip_tunnel_get_stats64,
-	.ndo_change_mtu		= eth_change_mtu,
+	.ndo_change_mtu		= geneve_change_mtu,
 	.ndo_validate_addr	= eth_validate_addr,
 	.ndo_set_mac_address	= eth_mac_addr,
 	.ndo_fill_metadata_dst	= geneve_fill_metadata_dst,
@@ -1150,6 +1178,7 @@ static void geneve_setup(struct net_device *dev)
 	dev->hw_features |= NETIF_F_GSO_SOFTWARE;
 
 	netif_keep_dst(dev);
+	dev->priv_flags &= ~IFF_TX_SKB_SHARING;
 	dev->priv_flags |= IFF_LIVE_ADDR_CHANGE | IFF_NO_QUEUE;
 	eth_hw_addr_random(dev);
 }
@@ -1441,12 +1470,23 @@ struct net_device *geneve_dev_create_fb(struct net *net, const char *name,
 		return dev;
 
 	err = geneve_configure(net, dev, &geneve_remote_unspec,
-			       0, 0, 0, htons(dst_port), true, 0);
-	if (err) {
-		free_netdev(dev);
-		return ERR_PTR(err);
-	}
+			       0, 0, 0, htons(dst_port), true,
+			       GENEVE_F_UDP_ZERO_CSUM6_RX);
+	if (err)
+		goto err;
+
+	/* openvswitch users expect packet sizes to be unrestricted,
+	 * so set the largest MTU we can.
+	 */
+	err = __geneve_change_mtu(dev, IP_MAX_MTU, false);
+	if (err)
+		goto err;
+
 	return dev;
+
+ err:
+	free_netdev(dev);
+	return ERR_PTR(err);
 }
 EXPORT_SYMBOL_GPL(geneve_dev_create_fb);
 
diff --git a/drivers/net/hyperv/netvsc_drv.c b/drivers/net/hyperv/netvsc_drv.c
index 1d3a66563bac..98e34fee45c7 100644
--- a/drivers/net/hyperv/netvsc_drv.c
+++ b/drivers/net/hyperv/netvsc_drv.c
@@ -1089,6 +1089,9 @@ static int netvsc_probe(struct hv_device *dev,
 	net->ethtool_ops = &ethtool_ops;
 	SET_NETDEV_DEV(net, &dev->device);
 
+	/* We always need headroom for rndis header */
+	net->needed_headroom = RNDIS_AND_PPI_SIZE;
+
 	/* Notify the netvsc driver of the new device */
 	memset(&device_info, 0, sizeof(device_info));
 	device_info.ring_size = ring_size;
diff --git a/drivers/net/phy/bcm7xxx.c b/drivers/net/phy/bcm7xxx.c
index bf241a3ec5e5..db507e3bcab9 100644
--- a/drivers/net/phy/bcm7xxx.c
+++ b/drivers/net/phy/bcm7xxx.c
@@ -250,10 +250,6 @@ static int bcm7xxx_config_init(struct phy_device *phydev)
 	phy_write(phydev, MII_BCM7XXX_AUX_MODE, MII_BCM7XX_64CLK_MDIO);
 	phy_read(phydev, MII_BCM7XXX_AUX_MODE);
 
-	/* Workaround only required for 100Mbits/sec capable PHYs */
-	if (phydev->supported & PHY_GBIT_FEATURES)
-		return 0;
-
 	/* set shadow mode 2 */
 	ret = phy_set_clr_bits(phydev, MII_BCM7XXX_TEST,
 			MII_BCM7XXX_SHD_MODE_2, MII_BCM7XXX_SHD_MODE_2);
@@ -270,7 +266,7 @@ static int bcm7xxx_config_init(struct phy_device *phydev)
 	phy_write(phydev, MII_BCM7XXX_100TX_FALSE_CAR, 0x7555);
 
 	/* reset shadow mode 2 */
-	ret = phy_set_clr_bits(phydev, MII_BCM7XXX_TEST, MII_BCM7XXX_SHD_MODE_2, 0);
+	ret = phy_set_clr_bits(phydev, MII_BCM7XXX_TEST, 0, MII_BCM7XXX_SHD_MODE_2);
 	if (ret < 0)
 		return ret;
 
@@ -307,11 +303,6 @@ static int bcm7xxx_suspend(struct phy_device *phydev)
 	return 0;
 }
 
-static int bcm7xxx_dummy_config_init(struct phy_device *phydev)
-{
-	return 0;
-}
-
 #define BCM7XXX_28NM_GPHY(_oui, _name)					\
 {									\
 	.phy_id		= (_oui),					\
@@ -337,7 +328,7 @@ static struct phy_driver bcm7xxx_driver[] = {
 	.phy_id         = PHY_ID_BCM7425,
 	.phy_id_mask    = 0xfffffff0,
 	.name           = "Broadcom BCM7425",
-	.features       = PHY_GBIT_FEATURES |
+	.features       = PHY_BASIC_FEATURES |
 			  SUPPORTED_Pause | SUPPORTED_Asym_Pause,
 	.flags          = PHY_IS_INTERNAL,
 	.config_init    = bcm7xxx_config_init,
@@ -349,7 +340,7 @@ static struct phy_driver bcm7xxx_driver[] = {
 	.phy_id         = PHY_ID_BCM7429,
 	.phy_id_mask    = 0xfffffff0,
 	.name           = "Broadcom BCM7429",
-	.features       = PHY_GBIT_FEATURES |
+	.features       = PHY_BASIC_FEATURES |
 			  SUPPORTED_Pause | SUPPORTED_Asym_Pause,
 	.flags          = PHY_IS_INTERNAL,
 	.config_init    = bcm7xxx_config_init,
@@ -361,7 +352,7 @@ static struct phy_driver bcm7xxx_driver[] = {
 	.phy_id         = PHY_ID_BCM7435,
 	.phy_id_mask    = 0xfffffff0,
 	.name           = "Broadcom BCM7435",
-	.features       = PHY_GBIT_FEATURES |
+	.features       = PHY_BASIC_FEATURES |
 			  SUPPORTED_Pause | SUPPORTED_Asym_Pause,
 	.flags          = PHY_IS_INTERNAL,
 	.config_init    = bcm7xxx_config_init,
@@ -369,30 +360,6 @@ static struct phy_driver bcm7xxx_driver[] = {
 	.read_status    = genphy_read_status,
 	.suspend        = bcm7xxx_suspend,
 	.resume         = bcm7xxx_config_init,
-}, {
-	.phy_id		= PHY_BCM_OUI_4,
-	.phy_id_mask	= 0xffff0000,
-	.name		= "Broadcom BCM7XXX 40nm",
-	.features	= PHY_GBIT_FEATURES |
-			  SUPPORTED_Pause | SUPPORTED_Asym_Pause,
-	.flags		= PHY_IS_INTERNAL,
-	.config_init	= bcm7xxx_config_init,
-	.config_aneg	= genphy_config_aneg,
-	.read_status	= genphy_read_status,
-	.suspend	= bcm7xxx_suspend,
-	.resume		= bcm7xxx_config_init,
-}, {
-	.phy_id		= PHY_BCM_OUI_5,
-	.phy_id_mask	= 0xffffff00,
-	.name		= "Broadcom BCM7XXX 65nm",
-	.features	= PHY_BASIC_FEATURES |
-			  SUPPORTED_Pause | SUPPORTED_Asym_Pause,
-	.flags		= PHY_IS_INTERNAL,
-	.config_init	= bcm7xxx_dummy_config_init,
-	.config_aneg	= genphy_config_aneg,
-	.read_status	= genphy_read_status,
-	.suspend	= bcm7xxx_suspend,
-	.resume		= bcm7xxx_config_init,
 } };
 
 static struct mdio_device_id __maybe_unused bcm7xxx_tbl[] = {
@@ -404,8 +371,6 @@ static struct mdio_device_id __maybe_unused bcm7xxx_tbl[] = {
 	{ PHY_ID_BCM7439, 0xfffffff0, },
 	{ PHY_ID_BCM7435, 0xfffffff0, },
 	{ PHY_ID_BCM7445, 0xfffffff0, },
-	{ PHY_BCM_OUI_4, 0xffff0000 },
-	{ PHY_BCM_OUI_5, 0xffffff00 },
 	{ }
 };
 
diff --git a/drivers/net/phy/marvell.c b/drivers/net/phy/marvell.c
index e3eb96443c97..ab1d0fcaf1d9 100644
--- a/drivers/net/phy/marvell.c
+++ b/drivers/net/phy/marvell.c
@@ -446,6 +446,12 @@ static int m88e1510_config_aneg(struct phy_device *phydev)
 	if (err < 0)
 		return err;
 
+	return 0;
+}
+
+static int marvell_config_init(struct phy_device *phydev)
+{
+	/* Set registers from marvell,reg-init DT property */
 	return marvell_of_reg_init(phydev);
 }
 
@@ -495,7 +501,7 @@ static int m88e1116r_config_init(struct phy_device *phydev)
 
 	mdelay(500);
 
-	return 0;
+	return marvell_config_init(phydev);
 }
 
 static int m88e3016_config_init(struct phy_device *phydev)
@@ -514,7 +520,7 @@ static int m88e3016_config_init(struct phy_device *phydev)
 	if (reg < 0)
 		return reg;
 
-	return 0;
+	return marvell_config_init(phydev);
 }
 
 static int m88e1111_config_init(struct phy_device *phydev)
@@ -1078,6 +1084,7 @@ static struct phy_driver marvell_drivers[] = {
 		.features = PHY_GBIT_FEATURES,
 		.probe = marvell_probe,
 		.flags = PHY_HAS_INTERRUPT,
+		.config_init = &marvell_config_init,
 		.config_aneg = &marvell_config_aneg,
 		.read_status = &genphy_read_status,
 		.ack_interrupt = &marvell_ack_interrupt,
@@ -1149,6 +1156,7 @@ static struct phy_driver marvell_drivers[] = {
 		.features = PHY_GBIT_FEATURES,
 		.flags = PHY_HAS_INTERRUPT,
 		.probe = marvell_probe,
+		.config_init = &marvell_config_init,
 		.config_aneg = &m88e1121_config_aneg,
 		.read_status = &marvell_read_status,
 		.ack_interrupt = &marvell_ack_interrupt,
@@ -1167,6 +1175,7 @@ static struct phy_driver marvell_drivers[] = {
 		.features = PHY_GBIT_FEATURES,
 		.flags = PHY_HAS_INTERRUPT,
 		.probe = marvell_probe,
+		.config_init = &marvell_config_init,
 		.config_aneg = &m88e1318_config_aneg,
 		.read_status = &marvell_read_status,
 		.ack_interrupt = &marvell_ack_interrupt,
@@ -1259,6 +1268,7 @@ static struct phy_driver marvell_drivers[] = {
 		.features = PHY_GBIT_FEATURES,
 		.flags = PHY_HAS_INTERRUPT,
 		.probe = marvell_probe,
+		.config_init = &marvell_config_init,
 		.config_aneg = &m88e1510_config_aneg,
 		.read_status = &marvell_read_status,
 		.ack_interrupt = &marvell_ack_interrupt,
@@ -1277,6 +1287,7 @@ static struct phy_driver marvell_drivers[] = {
 		.features = PHY_GBIT_FEATURES,
 		.flags = PHY_HAS_INTERRUPT,
 		.probe = marvell_probe,
+		.config_init = &marvell_config_init,
 		.config_aneg = &m88e1510_config_aneg,
 		.read_status = &marvell_read_status,
 		.ack_interrupt = &marvell_ack_interrupt,
diff --git a/drivers/net/phy/micrel.c b/drivers/net/phy/micrel.c
index 03833dbfca67..dc85f7095e51 100644
--- a/drivers/net/phy/micrel.c
+++ b/drivers/net/phy/micrel.c
@@ -297,6 +297,17 @@ static int kszphy_config_init(struct phy_device *phydev)
 	if (priv->led_mode >= 0)
 		kszphy_setup_led(phydev, type->led_mode_reg, priv->led_mode);
 
+	if (phy_interrupt_is_valid(phydev)) {
+		int ctl = phy_read(phydev, MII_BMCR);
+
+		if (ctl < 0)
+			return ctl;
+
+		ret = phy_write(phydev, MII_BMCR, ctl & ~BMCR_ANENABLE);
+		if (ret < 0)
+			return ret;
+	}
+
 	return 0;
 }
 
@@ -635,6 +646,21 @@ static void kszphy_get_stats(struct phy_device *phydev,
 		data[i] = kszphy_get_stat(phydev, i);
 }
 
+static int kszphy_resume(struct phy_device *phydev)
+{
+	int value;
+
+	mutex_lock(&phydev->lock);
+
+	value = phy_read(phydev, MII_BMCR);
+	phy_write(phydev, MII_BMCR, value & ~BMCR_PDOWN);
+
+	kszphy_config_intr(phydev);
+	mutex_unlock(&phydev->lock);
+
+	return 0;
+}
+
 static int kszphy_probe(struct phy_device *phydev)
 {
 	const struct kszphy_type *type = phydev->drv->driver_data;
@@ -844,7 +870,7 @@ static struct phy_driver ksphy_driver[] = {
 	.get_strings	= kszphy_get_strings,
 	.get_stats	= kszphy_get_stats,
 	.suspend	= genphy_suspend,
-	.resume		= genphy_resume,
+	.resume		= kszphy_resume,
 }, {
 	.phy_id		= PHY_ID_KSZ8061,
 	.name		= "Micrel KSZ8061",
diff --git a/drivers/net/phy/phy_device.c b/drivers/net/phy/phy_device.c
index bad3f005faee..e551f3a89cfd 100644
--- a/drivers/net/phy/phy_device.c
+++ b/drivers/net/phy/phy_device.c
@@ -1410,7 +1410,7 @@ int genphy_config_init(struct phy_device *phydev)
 
 	features = (SUPPORTED_TP | SUPPORTED_MII
 			| SUPPORTED_AUI | SUPPORTED_FIBRE |
-			SUPPORTED_BNC);
+			SUPPORTED_BNC | SUPPORTED_Pause | SUPPORTED_Asym_Pause);
 
 	/* Do we support autonegotiation? */
 	val = phy_read(phydev, MII_BMSR);
diff --git a/drivers/net/ppp/ppp_generic.c b/drivers/net/ppp/ppp_generic.c
index fc8ad001bc94..d61da9ece3ba 100644
--- a/drivers/net/ppp/ppp_generic.c
+++ b/drivers/net/ppp/ppp_generic.c
@@ -443,9 +443,14 @@ static ssize_t ppp_read(struct file *file, char __user *buf,
 			 * network traffic (demand mode).
 			 */
 			struct ppp *ppp = PF_TO_PPP(pf);
+
+			ppp_recv_lock(ppp);
 			if (ppp->n_channels == 0 &&
-			    (ppp->flags & SC_LOOP_TRAFFIC) == 0)
+			    (ppp->flags & SC_LOOP_TRAFFIC) == 0) {
+				ppp_recv_unlock(ppp);
 				break;
+			}
+			ppp_recv_unlock(ppp);
 		}
 		ret = -EAGAIN;
 		if (file->f_flags & O_NONBLOCK)
@@ -532,9 +537,12 @@ static unsigned int ppp_poll(struct file *file, poll_table *wait)
 	else if (pf->kind == INTERFACE) {
 		/* see comment in ppp_read */
 		struct ppp *ppp = PF_TO_PPP(pf);
+
+		ppp_recv_lock(ppp);
 		if (ppp->n_channels == 0 &&
 		    (ppp->flags & SC_LOOP_TRAFFIC) == 0)
 			mask |= POLLIN | POLLRDNORM;
+		ppp_recv_unlock(ppp);
 	}
 
 	return mask;
@@ -2808,6 +2816,7 @@ static struct ppp *ppp_create_interface(struct net *net, int unit,
 
 out2:
 	mutex_unlock(&pn->all_ppp_mutex);
+	rtnl_unlock();
 	free_netdev(dev);
 out1:
 	*retp = ret;
diff --git a/drivers/net/ppp/pppoe.c b/drivers/net/ppp/pppoe.c
index f3c63022eb3c..4ddae8118c85 100644
--- a/drivers/net/ppp/pppoe.c
+++ b/drivers/net/ppp/pppoe.c
@@ -395,6 +395,8 @@ static int pppoe_rcv_core(struct sock *sk, struct sk_buff *skb)
 
 		if (!__pppoe_xmit(sk_pppox(relay_po), skb))
 			goto abort_put;
+
+		sock_put(sk_pppox(relay_po));
 	} else {
 		if (sock_queue_rcv_skb(sk, skb))
 			goto abort_kfree;
diff --git a/drivers/net/usb/Kconfig b/drivers/net/usb/Kconfig
index 7f83504dfa69..cdde59089f72 100644
--- a/drivers/net/usb/Kconfig
+++ b/drivers/net/usb/Kconfig
@@ -395,6 +395,10 @@ config USB_NET_RNDIS_HOST
 	  The protocol specification is incomplete, and is controlled by
 	  (and for) Microsoft; it isn't an "Open" ecosystem or market.
 
+config USB_NET_CDC_SUBSET_ENABLE
+	tristate
+	depends on USB_NET_CDC_SUBSET
+
 config USB_NET_CDC_SUBSET
 	tristate "Simple USB Network Links (CDC Ethernet subset)"
 	depends on USB_USBNET
@@ -413,6 +417,7 @@ config USB_NET_CDC_SUBSET
 config USB_ALI_M5632
 	bool "ALi M5632 based 'USB 2.0 Data Link' cables"
 	depends on USB_NET_CDC_SUBSET
+	select USB_NET_CDC_SUBSET_ENABLE
 	help
 	  Choose this option if you're using a host-to-host cable
 	  based on this design, which supports USB 2.0 high speed.
@@ -420,6 +425,7 @@ config USB_ALI_M5632
 config USB_AN2720
 	bool "AnchorChips 2720 based cables (Xircom PGUNET, ...)"
 	depends on USB_NET_CDC_SUBSET
+	select USB_NET_CDC_SUBSET_ENABLE
 	help
 	  Choose this option if you're using a host-to-host cable
 	  based on this design.  Note that AnchorChips is now a
@@ -428,6 +434,7 @@ config USB_AN2720
 config USB_BELKIN
 	bool "eTEK based host-to-host cables (Advance, Belkin, ...)"
 	depends on USB_NET_CDC_SUBSET
+	select USB_NET_CDC_SUBSET_ENABLE
 	default y
 	help
 	  Choose this option if you're using a host-to-host cable
@@ -437,6 +444,7 @@ config USB_BELKIN
 config USB_ARMLINUX
 	bool "Embedded ARM Linux links (iPaq, ...)"
 	depends on USB_NET_CDC_SUBSET
+	select USB_NET_CDC_SUBSET_ENABLE
 	default y
 	help
 	  Choose this option to support the "usb-eth" networking driver
@@ -454,6 +462,7 @@ config USB_ARMLINUX
 config USB_EPSON2888
 	bool "Epson 2888 based firmware (DEVELOPMENT)"
 	depends on USB_NET_CDC_SUBSET
+	select USB_NET_CDC_SUBSET_ENABLE
 	help
 	  Choose this option to support the usb networking links used
 	  by some sample firmware from Epson.
@@ -461,6 +470,7 @@ config USB_EPSON2888
 config USB_KC2190
 	bool "KT Technology KC2190 based cables (InstaNet)"
 	depends on USB_NET_CDC_SUBSET
+	select USB_NET_CDC_SUBSET_ENABLE
 	help
 	  Choose this option if you're using a host-to-host cable
 	  with one of these chips.
diff --git a/drivers/net/usb/Makefile b/drivers/net/usb/Makefile
index b5f04068dbe4..37fb46aee341 100644
--- a/drivers/net/usb/Makefile
+++ b/drivers/net/usb/Makefile
@@ -23,7 +23,7 @@ obj-$(CONFIG_USB_NET_GL620A)	+= gl620a.o
 obj-$(CONFIG_USB_NET_NET1080)	+= net1080.o
 obj-$(CONFIG_USB_NET_PLUSB)	+= plusb.o
 obj-$(CONFIG_USB_NET_RNDIS_HOST)	+= rndis_host.o
-obj-$(CONFIG_USB_NET_CDC_SUBSET)	+= cdc_subset.o
+obj-$(CONFIG_USB_NET_CDC_SUBSET_ENABLE)	+= cdc_subset.o
 obj-$(CONFIG_USB_NET_ZAURUS)	+= zaurus.o
 obj-$(CONFIG_USB_NET_MCS7830)	+= mcs7830.o
 obj-$(CONFIG_USB_USBNET)	+= usbnet.o
diff --git a/drivers/net/usb/ax88172a.c b/drivers/net/usb/ax88172a.c
index 224e7d82de6d..cf77f2dffa69 100644
--- a/drivers/net/usb/ax88172a.c
+++ b/drivers/net/usb/ax88172a.c
@@ -134,7 +134,6 @@ static void ax88172a_remove_mdio(struct usbnet *dev)
 
 	netdev_info(dev->net, "deregistering mdio bus %s\n", priv->mdio->id);
 	mdiobus_unregister(priv->mdio);
-	kfree(priv->mdio->irq);
 	mdiobus_free(priv->mdio);
 }
 
diff --git a/drivers/net/usb/cdc_ncm.c b/drivers/net/usb/cdc_ncm.c
index dc0212c3cc28..86ba30ba35e8 100644
--- a/drivers/net/usb/cdc_ncm.c
+++ b/drivers/net/usb/cdc_ncm.c
@@ -837,7 +837,11 @@ int cdc_ncm_bind_common(struct usbnet *dev, struct usb_interface *intf, u8 data_
 
 	iface_no = ctx->data->cur_altsetting->desc.bInterfaceNumber;
 
-	/* reset data interface */
+	/* Reset data interface. Some devices will not reset properly
+	 * unless they are configured first.  Toggle the altsetting to
+	 * force a reset
+	 */
+	usb_set_interface(dev->udev, iface_no, data_altsetting);
 	temp = usb_set_interface(dev->udev, iface_no, 0);
 	if (temp) {
 		dev_dbg(&intf->dev, "set interface failed\n");
@@ -984,8 +988,6 @@ EXPORT_SYMBOL_GPL(cdc_ncm_select_altsetting);
 
 static int cdc_ncm_bind(struct usbnet *dev, struct usb_interface *intf)
 {
-	int ret;
-
 	/* MBIM backwards compatible function? */
 	if (cdc_ncm_select_altsetting(intf) != CDC_NCM_COMM_ALTSETTING_NCM)
 		return -ENODEV;
@@ -994,16 +996,7 @@ static int cdc_ncm_bind(struct usbnet *dev, struct usb_interface *intf)
 	 * Additionally, generic NCM devices are assumed to accept arbitrarily
 	 * placed NDP.
 	 */
-	ret = cdc_ncm_bind_common(dev, intf, CDC_NCM_DATA_ALTSETTING_NCM, 0);
-
-	/*
-	 * We should get an event when network connection is "connected" or
-	 * "disconnected". Set network connection in "disconnected" state
-	 * (carrier is OFF) during attach, so the IP network stack does not
-	 * start IPv6 negotiation and more.
-	 */
-	usbnet_link_change(dev, 0, 0);
-	return ret;
+	return cdc_ncm_bind_common(dev, intf, CDC_NCM_DATA_ALTSETTING_NCM, 0);
 }
 
 static void cdc_ncm_align_tail(struct sk_buff *skb, size_t modulus, size_t remainder, size_t max)
@@ -1586,7 +1579,8 @@ static void cdc_ncm_status(struct usbnet *dev, struct urb *urb)
 
 static const struct driver_info cdc_ncm_info = {
 	.description = "CDC NCM",
-	.flags = FLAG_POINTTOPOINT | FLAG_NO_SETINT | FLAG_MULTI_PACKET,
+	.flags = FLAG_POINTTOPOINT | FLAG_NO_SETINT | FLAG_MULTI_PACKET
+			| FLAG_LINK_INTR,
 	.bind = cdc_ncm_bind,
 	.unbind = cdc_ncm_unbind,
 	.manage_power = usbnet_manage_power,
@@ -1599,7 +1593,7 @@ static const struct driver_info cdc_ncm_info = {
 static const struct driver_info wwan_info = {
 	.description = "Mobile Broadband Network Device",
 	.flags = FLAG_POINTTOPOINT | FLAG_NO_SETINT | FLAG_MULTI_PACKET
-			| FLAG_WWAN,
+			| FLAG_LINK_INTR | FLAG_WWAN,
 	.bind = cdc_ncm_bind,
 	.unbind = cdc_ncm_unbind,
 	.manage_power = usbnet_manage_power,
@@ -1612,7 +1606,7 @@ static const struct driver_info wwan_info = {
 static const struct driver_info wwan_noarp_info = {
 	.description = "Mobile Broadband Network Device (NO ARP)",
 	.flags = FLAG_POINTTOPOINT | FLAG_NO_SETINT | FLAG_MULTI_PACKET
-			| FLAG_WWAN | FLAG_NOARP,
+			| FLAG_LINK_INTR | FLAG_WWAN | FLAG_NOARP,
 	.bind = cdc_ncm_bind,
 	.unbind = cdc_ncm_unbind,
 	.manage_power = usbnet_manage_power,
diff --git a/drivers/net/usb/qmi_wwan.c b/drivers/net/usb/qmi_wwan.c
index 23e9880791fc..a3a4ccf7cf52 100644
--- a/drivers/net/usb/qmi_wwan.c
+++ b/drivers/net/usb/qmi_wwan.c
@@ -637,6 +637,7 @@ static const struct usb_device_id products[] = {
 
 	/* 3. Combined interface devices matching on interface number */
 	{QMI_FIXED_INTF(0x0408, 0xea42, 4)},	/* Yota / Megafon M100-1 */
+	{QMI_FIXED_INTF(0x05c6, 0x6001, 3)},	/* 4G LTE usb-modem U901 */
 	{QMI_FIXED_INTF(0x05c6, 0x7000, 0)},
 	{QMI_FIXED_INTF(0x05c6, 0x7001, 1)},
 	{QMI_FIXED_INTF(0x05c6, 0x7002, 1)},
@@ -860,8 +861,10 @@ static const struct usb_device_id products[] = {
 	{QMI_FIXED_INTF(0x1199, 0x9056, 8)},	/* Sierra Wireless Modem */
 	{QMI_FIXED_INTF(0x1199, 0x9057, 8)},
 	{QMI_FIXED_INTF(0x1199, 0x9061, 8)},	/* Sierra Wireless Modem */
-	{QMI_FIXED_INTF(0x1199, 0x9071, 8)},	/* Sierra Wireless MC74xx/EM74xx */
-	{QMI_FIXED_INTF(0x1199, 0x9071, 10)},	/* Sierra Wireless MC74xx/EM74xx */
+	{QMI_FIXED_INTF(0x1199, 0x9071, 8)},	/* Sierra Wireless MC74xx */
+	{QMI_FIXED_INTF(0x1199, 0x9071, 10)},	/* Sierra Wireless MC74xx */
+	{QMI_FIXED_INTF(0x1199, 0x9079, 8)},	/* Sierra Wireless EM74xx */
+	{QMI_FIXED_INTF(0x1199, 0x9079, 10)},	/* Sierra Wireless EM74xx */
 	{QMI_FIXED_INTF(0x1bbb, 0x011e, 4)},	/* Telekom Speedstick LTE II (Alcatel One Touch L100V LTE) */
 	{QMI_FIXED_INTF(0x1bbb, 0x0203, 2)},	/* Alcatel L800MA */
 	{QMI_FIXED_INTF(0x2357, 0x0201, 4)},	/* TP-LINK HSUPA Modem MA180 */
@@ -884,6 +887,7 @@ static const struct usb_device_id products[] = {
 	{QMI_FIXED_INTF(0x413c, 0x81a8, 8)},	/* Dell Wireless 5808 Gobi(TM) 4G LTE Mobile Broadband Card */
 	{QMI_FIXED_INTF(0x413c, 0x81a9, 8)},	/* Dell Wireless 5808e Gobi(TM) 4G LTE Mobile Broadband Card */
 	{QMI_FIXED_INTF(0x413c, 0x81b1, 8)},	/* Dell Wireless 5809e Gobi(TM) 4G LTE Mobile Broadband Card */
+	{QMI_FIXED_INTF(0x413c, 0x81b3, 8)},	/* Dell Wireless 5809e Gobi(TM) 4G LTE Mobile Broadband Card (rev3) */
 	{QMI_FIXED_INTF(0x03f0, 0x4e1d, 8)},	/* HP lt4111 LTE/EV-DO/HSPA+ Gobi 4G Module */
 	{QMI_FIXED_INTF(0x22de, 0x9061, 3)},	/* WeTelecom WPD-600N */
 	{QMI_FIXED_INTF(0x1e0e, 0x9001, 5)},	/* SIMCom 7230E */
diff --git a/drivers/net/usb/usbnet.c b/drivers/net/usb/usbnet.c
index 0b0ba7ef14e4..10798128c03f 100644
--- a/drivers/net/usb/usbnet.c
+++ b/drivers/net/usb/usbnet.c
@@ -1769,6 +1769,13 @@ out3:
 	if (info->unbind)
 		info->unbind (dev, udev);
 out1:
+	/* subdrivers must undo all they did in bind() if they
+	 * fail it, but we may fail later and a deferred kevent
+	 * may trigger an error resubmitting itself and, worse,
+	 * schedule a timer. So we kill it all just in case.
+	 */
+	cancel_work_sync(&dev->kevent);
+	del_timer_sync(&dev->delay);
 	free_netdev(net);
 out:
 	return status;
diff --git a/drivers/net/vmxnet3/vmxnet3_defs.h b/drivers/net/vmxnet3/vmxnet3_defs.h
index 221a53025fd0..72ba8ae7f09a 100644
--- a/drivers/net/vmxnet3/vmxnet3_defs.h
+++ b/drivers/net/vmxnet3/vmxnet3_defs.h
@@ -377,7 +377,7 @@ union Vmxnet3_GenericDesc {
 #define VMXNET3_TX_RING_MAX_SIZE   4096
 #define VMXNET3_TC_RING_MAX_SIZE   4096
 #define VMXNET3_RX_RING_MAX_SIZE   4096
-#define VMXNET3_RX_RING2_MAX_SIZE  2048
+#define VMXNET3_RX_RING2_MAX_SIZE  4096
 #define VMXNET3_RC_RING_MAX_SIZE   8192
 
 /* a list of reasons for queue stop */
diff --git a/drivers/net/vmxnet3/vmxnet3_drv.c b/drivers/net/vmxnet3/vmxnet3_drv.c
index 0cbf520cea77..fc895d0e85d9 100644
--- a/drivers/net/vmxnet3/vmxnet3_drv.c
+++ b/drivers/net/vmxnet3/vmxnet3_drv.c
@@ -814,7 +814,7 @@ vmxnet3_tq_init_all(struct vmxnet3_adapter *adapter)
 
 
 /*
- *    parse and copy relevant protocol headers:
+ *    parse relevant protocol headers:
  *      For a tso pkt, relevant headers are L2/3/4 including options
  *      For a pkt requesting csum offloading, they are L2/3 and may include L4
  *      if it's a TCP/UDP pkt
@@ -827,15 +827,14 @@ vmxnet3_tq_init_all(struct vmxnet3_adapter *adapter)
  * Other effects:
  *    1. related *ctx fields are updated.
  *    2. ctx->copy_size is # of bytes copied
- *    3. the portion copied is guaranteed to be in the linear part
+ *    3. the portion to be copied is guaranteed to be in the linear part
  *
  */
 static int
-vmxnet3_parse_and_copy_hdr(struct sk_buff *skb, struct vmxnet3_tx_queue *tq,
-			   struct vmxnet3_tx_ctx *ctx,
-			   struct vmxnet3_adapter *adapter)
+vmxnet3_parse_hdr(struct sk_buff *skb, struct vmxnet3_tx_queue *tq,
+		  struct vmxnet3_tx_ctx *ctx,
+		  struct vmxnet3_adapter *adapter)
 {
-	struct Vmxnet3_TxDataDesc *tdd;
 	u8 protocol = 0;
 
 	if (ctx->mss) {	/* TSO */
@@ -892,16 +891,34 @@ vmxnet3_parse_and_copy_hdr(struct sk_buff *skb, struct vmxnet3_tx_queue *tq,
 		return 0;
 	}
 
+	return 1;
+err:
+	return -1;
+}
+
+/*
+ *    copy relevant protocol headers to the transmit ring:
+ *      For a tso pkt, relevant headers are L2/3/4 including options
+ *      For a pkt requesting csum offloading, they are L2/3 and may include L4
+ *      if it's a TCP/UDP pkt
+ *
+ *
+ *    Note that this requires that vmxnet3_parse_hdr be called first to set the
+ *      appropriate bits in ctx first
+ */
+static void
+vmxnet3_copy_hdr(struct sk_buff *skb, struct vmxnet3_tx_queue *tq,
+		 struct vmxnet3_tx_ctx *ctx,
+		 struct vmxnet3_adapter *adapter)
+{
+	struct Vmxnet3_TxDataDesc *tdd;
+
 	tdd = tq->data_ring.base + tq->tx_ring.next2fill;
 
 	memcpy(tdd->data, skb->data, ctx->copy_size);
 	netdev_dbg(adapter->netdev,
 		"copy %u bytes to dataRing[%u]\n",
 		ctx->copy_size, tq->tx_ring.next2fill);
-	return 1;
-
-err:
-	return -1;
 }
 
 
@@ -998,22 +1015,7 @@ vmxnet3_tq_xmit(struct sk_buff *skb, struct vmxnet3_tx_queue *tq,
 		}
 	}
 
-	spin_lock_irqsave(&tq->tx_lock, flags);
-
-	if (count > vmxnet3_cmd_ring_desc_avail(&tq->tx_ring)) {
-		tq->stats.tx_ring_full++;
-		netdev_dbg(adapter->netdev,
-			"tx queue stopped on %s, next2comp %u"
-			" next2fill %u\n", adapter->netdev->name,
-			tq->tx_ring.next2comp, tq->tx_ring.next2fill);
-
-		vmxnet3_tq_stop(tq, adapter);
-		spin_unlock_irqrestore(&tq->tx_lock, flags);
-		return NETDEV_TX_BUSY;
-	}
-
-
-	ret = vmxnet3_parse_and_copy_hdr(skb, tq, &ctx, adapter);
+	ret = vmxnet3_parse_hdr(skb, tq, &ctx, adapter);
 	if (ret >= 0) {
 		BUG_ON(ret <= 0 && ctx.copy_size != 0);
 		/* hdrs parsed, check against other limits */
@@ -1033,9 +1035,26 @@ vmxnet3_tq_xmit(struct sk_buff *skb, struct vmxnet3_tx_queue *tq,
 		}
 	} else {
 		tq->stats.drop_hdr_inspect_err++;
-		goto unlock_drop_pkt;
+		goto drop_pkt;
 	}
 
+	spin_lock_irqsave(&tq->tx_lock, flags);
+
+	if (count > vmxnet3_cmd_ring_desc_avail(&tq->tx_ring)) {
+		tq->stats.tx_ring_full++;
+		netdev_dbg(adapter->netdev,
+			"tx queue stopped on %s, next2comp %u"
+			" next2fill %u\n", adapter->netdev->name,
+			tq->tx_ring.next2comp, tq->tx_ring.next2fill);
+
+		vmxnet3_tq_stop(tq, adapter);
+		spin_unlock_irqrestore(&tq->tx_lock, flags);
+		return NETDEV_TX_BUSY;
+	}
+
+
+	vmxnet3_copy_hdr(skb, tq, &ctx, adapter);
+
 	/* fill tx descs related to addr & len */
 	if (vmxnet3_map_pkt(skb, &ctx, tq, adapter->pdev, adapter))
 		goto unlock_drop_pkt;
diff --git a/drivers/net/vmxnet3/vmxnet3_int.h b/drivers/net/vmxnet3/vmxnet3_int.h
index bdb8a6c0f8aa..729c344e6774 100644
--- a/drivers/net/vmxnet3/vmxnet3_int.h
+++ b/drivers/net/vmxnet3/vmxnet3_int.h
@@ -69,10 +69,10 @@
 /*
  * Version numbers
  */
-#define VMXNET3_DRIVER_VERSION_STRING   "1.4.5.0-k"
+#define VMXNET3_DRIVER_VERSION_STRING   "1.4.6.0-k"
 
 /* a 32-bit int, each byte encode a verion number in VMXNET3_DRIVER_VERSION */
-#define VMXNET3_DRIVER_VERSION_NUM      0x01040500
+#define VMXNET3_DRIVER_VERSION_NUM      0x01040600
 
 #if defined(CONFIG_PCI_MSI)
 	/* RSS only makes sense if MSI-X is supported. */
diff --git a/drivers/net/vrf.c b/drivers/net/vrf.c
index 66addb7a7911..bdcf617a9d52 100644
--- a/drivers/net/vrf.c
+++ b/drivers/net/vrf.c
@@ -104,20 +104,23 @@ static struct dst_ops vrf_dst_ops = {
 #if IS_ENABLED(CONFIG_IPV6)
 static bool check_ipv6_frame(const struct sk_buff *skb)
 {
-	const struct ipv6hdr *ipv6h = (struct ipv6hdr *)skb->data;
-	size_t hlen = sizeof(*ipv6h);
+	const struct ipv6hdr *ipv6h;
+	struct ipv6hdr _ipv6h;
 	bool rc = true;
 
-	if (skb->len < hlen)
+	ipv6h = skb_header_pointer(skb, 0, sizeof(_ipv6h), &_ipv6h);
+	if (!ipv6h)
 		goto out;
 
 	if (ipv6h->nexthdr == NEXTHDR_ICMP) {
 		const struct icmp6hdr *icmph;
+		struct icmp6hdr _icmph;
 
-		if (skb->len < hlen + sizeof(*icmph))
+		icmph = skb_header_pointer(skb, sizeof(_ipv6h),
+					   sizeof(_icmph), &_icmph);
+		if (!icmph)
 			goto out;
 
-		icmph = (struct icmp6hdr *)(skb->data + sizeof(*ipv6h));
 		switch (icmph->icmp6_type) {
 		case NDISC_ROUTER_SOLICITATION:
 		case NDISC_ROUTER_ADVERTISEMENT:
diff --git a/drivers/net/vxlan.c b/drivers/net/vxlan.c
index 65439188c582..1c32bd104797 100644
--- a/drivers/net/vxlan.c
+++ b/drivers/net/vxlan.c
@@ -931,8 +931,10 @@ static int vxlan_fdb_dump(struct sk_buff *skb, struct netlink_callback *cb,
 						     cb->nlh->nlmsg_seq,
 						     RTM_NEWNEIGH,
 						     NLM_F_MULTI, rd);
-				if (err < 0)
+				if (err < 0) {
+					cb->args[1] = err;
 					goto out;
+				}
 skip:
 				++idx;
 			}
@@ -1306,8 +1308,10 @@ static int vxlan_udp_encap_recv(struct sock *sk, struct sk_buff *skb)
 		gbp = (struct vxlanhdr_gbp *)vxh;
 		md->gbp = ntohs(gbp->policy_id);
 
-		if (tun_dst)
+		if (tun_dst) {
 			tun_dst->u.tun_info.key.tun_flags |= TUNNEL_VXLAN_OPT;
+			tun_dst->u.tun_info.options_len = sizeof(*md);
+		}
 
 		if (gbp->dont_learn)
 			md->gbp |= VXLAN_GBP_DONT_LEARN;
@@ -2171,9 +2175,11 @@ static netdev_tx_t vxlan_xmit(struct sk_buff *skb, struct net_device *dev)
 #endif
 	}
 
-	if (vxlan->flags & VXLAN_F_COLLECT_METADATA &&
-	    info && info->mode & IP_TUNNEL_INFO_TX) {
-		vxlan_xmit_one(skb, dev, NULL, false);
+	if (vxlan->flags & VXLAN_F_COLLECT_METADATA) {
+		if (info && info->mode & IP_TUNNEL_INFO_TX)
+			vxlan_xmit_one(skb, dev, NULL, false);
+		else
+			kfree_skb(skb);
 		return NETDEV_TX_OK;
 	}
 
@@ -2367,29 +2373,43 @@ static void vxlan_set_multicast_list(struct net_device *dev)
 {
 }
 
-static int vxlan_change_mtu(struct net_device *dev, int new_mtu)
+static int __vxlan_change_mtu(struct net_device *dev,
+			      struct net_device *lowerdev,
+			      struct vxlan_rdst *dst, int new_mtu, bool strict)
 {
-	struct vxlan_dev *vxlan = netdev_priv(dev);
-	struct vxlan_rdst *dst = &vxlan->default_dst;
-	struct net_device *lowerdev;
-	int max_mtu;
+	int max_mtu = IP_MAX_MTU;
 
-	lowerdev = __dev_get_by_index(vxlan->net, dst->remote_ifindex);
-	if (lowerdev == NULL)
-		return eth_change_mtu(dev, new_mtu);
+	if (lowerdev)
+		max_mtu = lowerdev->mtu;
 
 	if (dst->remote_ip.sa.sa_family == AF_INET6)
-		max_mtu = lowerdev->mtu - VXLAN6_HEADROOM;
+		max_mtu -= VXLAN6_HEADROOM;
 	else
-		max_mtu = lowerdev->mtu - VXLAN_HEADROOM;
+		max_mtu -= VXLAN_HEADROOM;
 
-	if (new_mtu < 68 || new_mtu > max_mtu)
+	if (new_mtu < 68)
 		return -EINVAL;
 
+	if (new_mtu > max_mtu) {
+		if (strict)
+			return -EINVAL;
+
+		new_mtu = max_mtu;
+	}
+
 	dev->mtu = new_mtu;
 	return 0;
 }
 
+static int vxlan_change_mtu(struct net_device *dev, int new_mtu)
+{
+	struct vxlan_dev *vxlan = netdev_priv(dev);
+	struct vxlan_rdst *dst = &vxlan->default_dst;
+	struct net_device *lowerdev = __dev_get_by_index(vxlan->net,
+							 dst->remote_ifindex);
+	return __vxlan_change_mtu(dev, lowerdev, dst, new_mtu, true);
+}
+
 static int egress_ipv4_tun_info(struct net_device *dev, struct sk_buff *skb,
 				struct ip_tunnel_info *info,
 				__be16 sport, __be16 dport)
@@ -2523,6 +2543,7 @@ static void vxlan_setup(struct net_device *dev)
 	dev->hw_features |= NETIF_F_GSO_SOFTWARE;
 	dev->hw_features |= NETIF_F_HW_VLAN_CTAG_TX | NETIF_F_HW_VLAN_STAG_TX;
 	netif_keep_dst(dev);
+	dev->priv_flags &= ~IFF_TX_SKB_SHARING;
 	dev->priv_flags |= IFF_LIVE_ADDR_CHANGE | IFF_NO_QUEUE;
 
 	INIT_LIST_HEAD(&vxlan->next);
@@ -2765,6 +2786,7 @@ static int vxlan_dev_configure(struct net *src_net, struct net_device *dev,
 	int err;
 	bool use_ipv6 = false;
 	__be16 default_port = vxlan->cfg.dst_port;
+	struct net_device *lowerdev = NULL;
 
 	vxlan->net = src_net;
 
@@ -2785,9 +2807,7 @@ static int vxlan_dev_configure(struct net *src_net, struct net_device *dev,
 	}
 
 	if (conf->remote_ifindex) {
-		struct net_device *lowerdev
-			 = __dev_get_by_index(src_net, conf->remote_ifindex);
-
+		lowerdev = __dev_get_by_index(src_net, conf->remote_ifindex);
 		dst->remote_ifindex = conf->remote_ifindex;
 
 		if (!lowerdev) {
@@ -2811,6 +2831,12 @@ static int vxlan_dev_configure(struct net *src_net, struct net_device *dev,
 		needed_headroom = lowerdev->hard_header_len;
 	}
 
+	if (conf->mtu) {
+		err = __vxlan_change_mtu(dev, lowerdev, dst, conf->mtu, false);
+		if (err)
+			return err;
+	}
+
 	if (use_ipv6 || conf->flags & VXLAN_F_COLLECT_METADATA)
 		needed_headroom += VXLAN6_HEADROOM;
 	else
diff --git a/drivers/net/wan/dscc4.c b/drivers/net/wan/dscc4.c
index 7a72407208b1..629225980463 100644
--- a/drivers/net/wan/dscc4.c
+++ b/drivers/net/wan/dscc4.c
@@ -1626,7 +1626,7 @@ try:
 		if (state & Xpr) {
 			void __iomem *scc_addr;
 			unsigned long ring;
-			int i;
+			unsigned int i;
 
 			/*
 			 * - the busy condition happens (sometimes);
diff --git a/drivers/net/wireless/intel/iwlwifi/Kconfig b/drivers/net/wireless/intel/iwlwifi/Kconfig
index 866067789330..7438fbeef744 100644
--- a/drivers/net/wireless/intel/iwlwifi/Kconfig
+++ b/drivers/net/wireless/intel/iwlwifi/Kconfig
@@ -53,7 +53,6 @@ config IWLWIFI_LEDS
 
 config IWLDVM
 	tristate "Intel Wireless WiFi DVM Firmware support"
-	depends on m
 	help
 	  This is the driver that supports the DVM firmware. The list
 	  of the devices that use this firmware is available here:
diff --git a/drivers/net/wireless/intel/iwlwifi/iwl-8000.c b/drivers/net/wireless/intel/iwlwifi/iwl-8000.c
index c84a0299d43e..bce9b3420a13 100644
--- a/drivers/net/wireless/intel/iwlwifi/iwl-8000.c
+++ b/drivers/net/wireless/intel/iwlwifi/iwl-8000.c
@@ -7,6 +7,7 @@
  *
  * Copyright(c) 2014 Intel Corporation. All rights reserved.
  * Copyright(c) 2014 - 2015 Intel Mobile Communications GmbH
+ * Copyright(c) 2016 Intel Deutschland GmbH
  *
  * This program is free software; you can redistribute it and/or modify
  * it under the terms of version 2 of the GNU General Public License as
@@ -70,12 +71,15 @@
 
 /* Highest firmware API version supported */
 #define IWL8000_UCODE_API_MAX	20
+#define IWL8265_UCODE_API_MAX	20
 
 /* Oldest version we won't warn about */
 #define IWL8000_UCODE_API_OK	13
+#define IWL8265_UCODE_API_OK	20
 
 /* Lowest firmware API version supported */
 #define IWL8000_UCODE_API_MIN	13
+#define IWL8265_UCODE_API_MIN	20
 
 /* NVM versions */
 #define IWL8000_NVM_VERSION		0x0a1d
@@ -93,6 +97,10 @@
 #define IWL8000_MODULE_FIRMWARE(api) \
 	IWL8000_FW_PRE "-" __stringify(api) ".ucode"
 
+#define IWL8265_FW_PRE "iwlwifi-8265-"
+#define IWL8265_MODULE_FIRMWARE(api) \
+	IWL8265_FW_PRE __stringify(api) ".ucode"
+
 #define NVM_HW_SECTION_NUM_FAMILY_8000		10
 #define DEFAULT_NVM_FILE_FAMILY_8000B		"nvmData-8000B"
 #define DEFAULT_NVM_FILE_FAMILY_8000C		"nvmData-8000C"
@@ -144,10 +152,7 @@ static const struct iwl_tt_params iwl8000_tt_params = {
 	.support_tx_backoff = true,
 };
 
-#define IWL_DEVICE_8000							\
-	.ucode_api_max = IWL8000_UCODE_API_MAX,				\
-	.ucode_api_ok = IWL8000_UCODE_API_OK,				\
-	.ucode_api_min = IWL8000_UCODE_API_MIN,				\
+#define IWL_DEVICE_8000_COMMON						\
 	.device_family = IWL_DEVICE_FAMILY_8000,			\
 	.max_inst_size = IWL60_RTC_INST_SIZE,				\
 	.max_data_size = IWL60_RTC_DATA_SIZE,				\
@@ -167,10 +172,28 @@ static const struct iwl_tt_params iwl8000_tt_params = {
 	.thermal_params = &iwl8000_tt_params,				\
 	.apmg_not_supported = true
 
+#define IWL_DEVICE_8000							\
+	IWL_DEVICE_8000_COMMON,						\
+	.ucode_api_max = IWL8000_UCODE_API_MAX,				\
+	.ucode_api_ok = IWL8000_UCODE_API_OK,				\
+	.ucode_api_min = IWL8000_UCODE_API_MIN				\
+
+#define IWL_DEVICE_8260							\
+	IWL_DEVICE_8000_COMMON,						\
+	.ucode_api_max = IWL8000_UCODE_API_MAX,				\
+	.ucode_api_ok = IWL8000_UCODE_API_OK,				\
+	.ucode_api_min = IWL8000_UCODE_API_MIN				\
+
+#define IWL_DEVICE_8265							\
+	IWL_DEVICE_8000_COMMON,						\
+	.ucode_api_max = IWL8265_UCODE_API_MAX,				\
+	.ucode_api_ok = IWL8265_UCODE_API_OK,				\
+	.ucode_api_min = IWL8265_UCODE_API_MIN				\
+
 const struct iwl_cfg iwl8260_2n_cfg = {
 	.name = "Intel(R) Dual Band Wireless N 8260",
 	.fw_name_pre = IWL8000_FW_PRE,
-	IWL_DEVICE_8000,
+	IWL_DEVICE_8260,
 	.ht_params = &iwl8000_ht_params,
 	.nvm_ver = IWL8000_NVM_VERSION,
 	.nvm_calib_ver = IWL8000_TX_POWER_VERSION,
@@ -179,7 +202,7 @@ const struct iwl_cfg iwl8260_2n_cfg = {
 const struct iwl_cfg iwl8260_2ac_cfg = {
 	.name = "Intel(R) Dual Band Wireless AC 8260",
 	.fw_name_pre = IWL8000_FW_PRE,
-	IWL_DEVICE_8000,
+	IWL_DEVICE_8260,
 	.ht_params = &iwl8000_ht_params,
 	.nvm_ver = IWL8000_NVM_VERSION,
 	.nvm_calib_ver = IWL8000_TX_POWER_VERSION,
@@ -188,8 +211,8 @@ const struct iwl_cfg iwl8260_2ac_cfg = {
 
 const struct iwl_cfg iwl8265_2ac_cfg = {
 	.name = "Intel(R) Dual Band Wireless AC 8265",
-	.fw_name_pre = IWL8000_FW_PRE,
-	IWL_DEVICE_8000,
+	.fw_name_pre = IWL8265_FW_PRE,
+	IWL_DEVICE_8265,
 	.ht_params = &iwl8000_ht_params,
 	.nvm_ver = IWL8000_NVM_VERSION,
 	.nvm_calib_ver = IWL8000_TX_POWER_VERSION,
@@ -209,7 +232,7 @@ const struct iwl_cfg iwl4165_2ac_cfg = {
 const struct iwl_cfg iwl8260_2ac_sdio_cfg = {
 	.name = "Intel(R) Dual Band Wireless-AC 8260",
 	.fw_name_pre = IWL8000_FW_PRE,
-	IWL_DEVICE_8000,
+	IWL_DEVICE_8260,
 	.ht_params = &iwl8000_ht_params,
 	.nvm_ver = IWL8000_NVM_VERSION,
 	.nvm_calib_ver = IWL8000_TX_POWER_VERSION,
@@ -236,3 +259,4 @@ const struct iwl_cfg iwl4165_2ac_sdio_cfg = {
 };
 
 MODULE_FIRMWARE(IWL8000_MODULE_FIRMWARE(IWL8000_UCODE_API_OK));
+MODULE_FIRMWARE(IWL8265_MODULE_FIRMWARE(IWL8265_UCODE_API_OK));
diff --git a/drivers/net/wireless/intel/iwlwifi/iwl-drv.c b/drivers/net/wireless/intel/iwlwifi/iwl-drv.c
index 7acb49075683..ab4c2a0470b2 100644
--- a/drivers/net/wireless/intel/iwlwifi/iwl-drv.c
+++ b/drivers/net/wireless/intel/iwlwifi/iwl-drv.c
@@ -243,8 +243,10 @@ static int iwl_request_firmware(struct iwl_drv *drv, bool first)
 	if (drv->trans->cfg->device_family == IWL_DEVICE_FAMILY_8000) {
 		char rev_step = 'A' + CSR_HW_REV_STEP(drv->trans->hw_rev);
 
-		snprintf(drv->firmware_name, sizeof(drv->firmware_name),
-			 "%s%c-%s.ucode", name_pre, rev_step, tag);
+		if (rev_step != 'A')
+			snprintf(drv->firmware_name,
+				 sizeof(drv->firmware_name), "%s%c-%s.ucode",
+				 name_pre, rev_step, tag);
 	}
 
 	IWL_DEBUG_INFO(drv, "attempting to load firmware %s'%s'\n",
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/fw.c b/drivers/net/wireless/intel/iwlwifi/mvm/fw.c
index 4ed5180c547b..0ccc697fef76 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/fw.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/fw.c
@@ -107,7 +107,7 @@ static int iwl_send_tx_ant_cfg(struct iwl_mvm *mvm, u8 valid_tx_ant)
 				    sizeof(tx_ant_cmd), &tx_ant_cmd);
 }
 
-static void iwl_free_fw_paging(struct iwl_mvm *mvm)
+void iwl_free_fw_paging(struct iwl_mvm *mvm)
 {
 	int i;
 
@@ -127,6 +127,8 @@ static void iwl_free_fw_paging(struct iwl_mvm *mvm)
 			     get_order(mvm->fw_paging_db[i].fw_paging_size));
 	}
 	kfree(mvm->trans->paging_download_buf);
+	mvm->trans->paging_download_buf = NULL;
+
 	memset(mvm->fw_paging_db, 0, sizeof(mvm->fw_paging_db));
 }
 
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h b/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h
index 5f3ac8cccf49..ff7c6df9f941 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/mvm.h
@@ -1225,6 +1225,9 @@ void iwl_mvm_rx_umac_scan_complete_notif(struct iwl_mvm *mvm,
 void iwl_mvm_rx_umac_scan_iter_complete_notif(struct iwl_mvm *mvm,
 					      struct iwl_rx_cmd_buffer *rxb);
 
+/* Paging */
+void iwl_free_fw_paging(struct iwl_mvm *mvm);
+
 /* MVM debugfs */
 #ifdef CONFIG_IWLWIFI_DEBUGFS
 int iwl_mvm_dbgfs_register(struct iwl_mvm *mvm, struct dentry *dbgfs_dir);
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/ops.c b/drivers/net/wireless/intel/iwlwifi/mvm/ops.c
index 89ea70deeb84..e80be9a59520 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/ops.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/ops.c
@@ -684,6 +684,8 @@ static void iwl_op_mode_mvm_stop(struct iwl_op_mode *op_mode)
 	for (i = 0; i < NVM_MAX_NUM_SECTIONS; i++)
 		kfree(mvm->nvm_sections[i].data);
 
+	iwl_free_fw_paging(mvm);
+
 	iwl_mvm_tof_clean(mvm);
 
 	ieee80211_free_hw(mvm->hw);
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/scan.c b/drivers/net/wireless/intel/iwlwifi/mvm/scan.c
index 9a15642f80dd..ea1e177c2ea1 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/scan.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/scan.c
@@ -1298,6 +1298,10 @@ int iwl_mvm_sched_scan_start(struct iwl_mvm *mvm,
 		return -EBUSY;
 	}
 
+	/* we don't support "match all" in the firmware */
+	if (!req->n_match_sets)
+		return -EOPNOTSUPP;
+
 	ret = iwl_mvm_check_running_scans(mvm, type);
 	if (ret)
 		return ret;
diff --git a/drivers/net/wireless/intel/iwlwifi/mvm/tx.c b/drivers/net/wireless/intel/iwlwifi/mvm/tx.c
index 0914ec2fd574..a040edc55057 100644
--- a/drivers/net/wireless/intel/iwlwifi/mvm/tx.c
+++ b/drivers/net/wireless/intel/iwlwifi/mvm/tx.c
@@ -423,6 +423,15 @@ int iwl_mvm_tx_skb_non_sta(struct iwl_mvm *mvm, struct sk_buff *skb)
 		return -1;
 	}
 
+	/*
+	 * Increase the pending frames counter, so that later when a reply comes
+	 * in and the counter is decreased - we don't start getting negative
+	 * values.
+	 * Note that we don't need to make sure it isn't agg'd, since we're
+	 * TXing non-sta
+	 */
+	atomic_inc(&mvm->pending_frames[sta_id]);
+
 	return 0;
 }
 
diff --git a/drivers/net/wireless/intel/iwlwifi/pcie/internal.h b/drivers/net/wireless/intel/iwlwifi/pcie/internal.h
index cc3888e2700d..73c95594eabe 100644
--- a/drivers/net/wireless/intel/iwlwifi/pcie/internal.h
+++ b/drivers/net/wireless/intel/iwlwifi/pcie/internal.h
@@ -490,6 +490,15 @@ static inline void iwl_enable_interrupts(struct iwl_trans *trans)
 	iwl_write32(trans, CSR_INT_MASK, trans_pcie->inta_mask);
 }
 
+static inline void iwl_enable_fw_load_int(struct iwl_trans *trans)
+{
+	struct iwl_trans_pcie *trans_pcie = IWL_TRANS_GET_PCIE_TRANS(trans);
+
+	IWL_DEBUG_ISR(trans, "Enabling FW load interrupt\n");
+	trans_pcie->inta_mask = CSR_INT_BIT_FH_TX;
+	iwl_write32(trans, CSR_INT_MASK, trans_pcie->inta_mask);
+}
+
 static inline void iwl_enable_rfkill_int(struct iwl_trans *trans)
 {
 	struct iwl_trans_pcie *trans_pcie = IWL_TRANS_GET_PCIE_TRANS(trans);
diff --git a/drivers/net/wireless/intel/iwlwifi/pcie/rx.c b/drivers/net/wireless/intel/iwlwifi/pcie/rx.c
index ccafbd8cf4b3..152cf9ad9566 100644
--- a/drivers/net/wireless/intel/iwlwifi/pcie/rx.c
+++ b/drivers/net/wireless/intel/iwlwifi/pcie/rx.c
@@ -1438,9 +1438,11 @@ irqreturn_t iwl_pcie_irq_handler(int irq, void *dev_id)
 			 inta & ~trans_pcie->inta_mask);
 	}
 
-	/* Re-enable all interrupts */
-	/* only Re-enable if disabled by irq */
-	if (test_bit(STATUS_INT_ENABLED, &trans->status))
+	/* we are loading the firmware, enable FH_TX interrupt only */
+	if (handled & CSR_INT_BIT_FH_TX)
+		iwl_enable_fw_load_int(trans);
+	/* only Re-enable all interrupt if disabled by irq */
+	else if (test_bit(STATUS_INT_ENABLED, &trans->status))
 		iwl_enable_interrupts(trans);
 	/* Re-enable RF_KILL if it occurred */
 	else if (handled & CSR_INT_BIT_RF_KILL)
diff --git a/drivers/net/wireless/intel/iwlwifi/pcie/trans.c b/drivers/net/wireless/intel/iwlwifi/pcie/trans.c
index d60a467a983c..5a854c609477 100644
--- a/drivers/net/wireless/intel/iwlwifi/pcie/trans.c
+++ b/drivers/net/wireless/intel/iwlwifi/pcie/trans.c
@@ -1021,82 +1021,6 @@ static int iwl_pcie_load_given_ucode_8000(struct iwl_trans *trans,
 					       &first_ucode_section);
 }
 
-static int iwl_trans_pcie_start_fw(struct iwl_trans *trans,
-				   const struct fw_img *fw, bool run_in_rfkill)
-{
-	struct iwl_trans_pcie *trans_pcie = IWL_TRANS_GET_PCIE_TRANS(trans);
-	bool hw_rfkill;
-	int ret;
-
-	mutex_lock(&trans_pcie->mutex);
-
-	/* Someone called stop_device, don't try to start_fw */
-	if (trans_pcie->is_down) {
-		IWL_WARN(trans,
-			 "Can't start_fw since the HW hasn't been started\n");
-		ret = EIO;
-		goto out;
-	}
-
-	/* This may fail if AMT took ownership of the device */
-	if (iwl_pcie_prepare_card_hw(trans)) {
-		IWL_WARN(trans, "Exit HW not ready\n");
-		ret = -EIO;
-		goto out;
-	}
-
-	iwl_enable_rfkill_int(trans);
-
-	/* If platform's RF_KILL switch is NOT set to KILL */
-	hw_rfkill = iwl_is_rfkill_set(trans);
-	if (hw_rfkill)
-		set_bit(STATUS_RFKILL, &trans->status);
-	else
-		clear_bit(STATUS_RFKILL, &trans->status);
-	iwl_trans_pcie_rf_kill(trans, hw_rfkill);
-	if (hw_rfkill && !run_in_rfkill) {
-		ret = -ERFKILL;
-		goto out;
-	}
-
-	iwl_write32(trans, CSR_INT, 0xFFFFFFFF);
-
-	ret = iwl_pcie_nic_init(trans);
-	if (ret) {
-		IWL_ERR(trans, "Unable to init nic\n");
-		goto out;
-	}
-
-	/* make sure rfkill handshake bits are cleared */
-	iwl_write32(trans, CSR_UCODE_DRV_GP1_CLR, CSR_UCODE_SW_BIT_RFKILL);
-	iwl_write32(trans, CSR_UCODE_DRV_GP1_CLR,
-		    CSR_UCODE_DRV_GP1_BIT_CMD_BLOCKED);
-
-	/* clear (again), then enable host interrupts */
-	iwl_write32(trans, CSR_INT, 0xFFFFFFFF);
-	iwl_enable_interrupts(trans);
-
-	/* really make sure rfkill handshake bits are cleared */
-	iwl_write32(trans, CSR_UCODE_DRV_GP1_CLR, CSR_UCODE_SW_BIT_RFKILL);
-	iwl_write32(trans, CSR_UCODE_DRV_GP1_CLR, CSR_UCODE_SW_BIT_RFKILL);
-
-	/* Load the given image to the HW */
-	if (trans->cfg->device_family == IWL_DEVICE_FAMILY_8000)
-		ret = iwl_pcie_load_given_ucode_8000(trans, fw);
-	else
-		ret = iwl_pcie_load_given_ucode(trans, fw);
-
-out:
-	mutex_unlock(&trans_pcie->mutex);
-	return ret;
-}
-
-static void iwl_trans_pcie_fw_alive(struct iwl_trans *trans, u32 scd_addr)
-{
-	iwl_pcie_reset_ict(trans);
-	iwl_pcie_tx_start(trans, scd_addr);
-}
-
 static void _iwl_trans_pcie_stop_device(struct iwl_trans *trans, bool low_power)
 {
 	struct iwl_trans_pcie *trans_pcie = IWL_TRANS_GET_PCIE_TRANS(trans);
@@ -1127,7 +1051,8 @@ static void _iwl_trans_pcie_stop_device(struct iwl_trans *trans, bool low_power)
 	 * already dead.
 	 */
 	if (test_and_clear_bit(STATUS_DEVICE_ENABLED, &trans->status)) {
-		IWL_DEBUG_INFO(trans, "DEVICE_ENABLED bit was set and is now cleared\n");
+		IWL_DEBUG_INFO(trans,
+			       "DEVICE_ENABLED bit was set and is now cleared\n");
 		iwl_pcie_tx_stop(trans);
 		iwl_pcie_rx_stop(trans);
 
@@ -1161,7 +1086,6 @@ static void _iwl_trans_pcie_stop_device(struct iwl_trans *trans, bool low_power)
 	iwl_disable_interrupts(trans);
 	spin_unlock(&trans_pcie->irq_lock);
 
-
 	/* clear all status bits */
 	clear_bit(STATUS_SYNC_HCMD_ACTIVE, &trans->status);
 	clear_bit(STATUS_INT_ENABLED, &trans->status);
@@ -1194,10 +1118,116 @@ static void _iwl_trans_pcie_stop_device(struct iwl_trans *trans, bool low_power)
 	if (hw_rfkill != was_hw_rfkill)
 		iwl_trans_pcie_rf_kill(trans, hw_rfkill);
 
-	/* re-take ownership to prevent other users from stealing the deivce */
+	/* re-take ownership to prevent other users from stealing the device */
 	iwl_pcie_prepare_card_hw(trans);
 }
 
+static int iwl_trans_pcie_start_fw(struct iwl_trans *trans,
+				   const struct fw_img *fw, bool run_in_rfkill)
+{
+	struct iwl_trans_pcie *trans_pcie = IWL_TRANS_GET_PCIE_TRANS(trans);
+	bool hw_rfkill;
+	int ret;
+
+	/* This may fail if AMT took ownership of the device */
+	if (iwl_pcie_prepare_card_hw(trans)) {
+		IWL_WARN(trans, "Exit HW not ready\n");
+		ret = -EIO;
+		goto out;
+	}
+
+	iwl_enable_rfkill_int(trans);
+
+	iwl_write32(trans, CSR_INT, 0xFFFFFFFF);
+
+	/*
+	 * We enabled the RF-Kill interrupt and the handler may very
+	 * well be running. Disable the interrupts to make sure no other
+	 * interrupt can be fired.
+	 */
+	iwl_disable_interrupts(trans);
+
+	/* Make sure it finished running */
+	synchronize_irq(trans_pcie->pci_dev->irq);
+
+	mutex_lock(&trans_pcie->mutex);
+
+	/* If platform's RF_KILL switch is NOT set to KILL */
+	hw_rfkill = iwl_is_rfkill_set(trans);
+	if (hw_rfkill)
+		set_bit(STATUS_RFKILL, &trans->status);
+	else
+		clear_bit(STATUS_RFKILL, &trans->status);
+	iwl_trans_pcie_rf_kill(trans, hw_rfkill);
+	if (hw_rfkill && !run_in_rfkill) {
+		ret = -ERFKILL;
+		goto out;
+	}
+
+	/* Someone called stop_device, don't try to start_fw */
+	if (trans_pcie->is_down) {
+		IWL_WARN(trans,
+			 "Can't start_fw since the HW hasn't been started\n");
+		ret = -EIO;
+		goto out;
+	}
+
+	/* make sure rfkill handshake bits are cleared */
+	iwl_write32(trans, CSR_UCODE_DRV_GP1_CLR, CSR_UCODE_SW_BIT_RFKILL);
+	iwl_write32(trans, CSR_UCODE_DRV_GP1_CLR,
+		    CSR_UCODE_DRV_GP1_BIT_CMD_BLOCKED);
+
+	/* clear (again), then enable host interrupts */
+	iwl_write32(trans, CSR_INT, 0xFFFFFFFF);
+
+	ret = iwl_pcie_nic_init(trans);
+	if (ret) {
+		IWL_ERR(trans, "Unable to init nic\n");
+		goto out;
+	}
+
+	/*
+	 * Now, we load the firmware and don't want to be interrupted, even
+	 * by the RF-Kill interrupt (hence mask all the interrupt besides the
+	 * FH_TX interrupt which is needed to load the firmware). If the
+	 * RF-Kill switch is toggled, we will find out after having loaded
+	 * the firmware and return the proper value to the caller.
+	 */
+	iwl_enable_fw_load_int(trans);
+
+	/* really make sure rfkill handshake bits are cleared */
+	iwl_write32(trans, CSR_UCODE_DRV_GP1_CLR, CSR_UCODE_SW_BIT_RFKILL);
+	iwl_write32(trans, CSR_UCODE_DRV_GP1_CLR, CSR_UCODE_SW_BIT_RFKILL);
+
+	/* Load the given image to the HW */
+	if (trans->cfg->device_family == IWL_DEVICE_FAMILY_8000)
+		ret = iwl_pcie_load_given_ucode_8000(trans, fw);
+	else
+		ret = iwl_pcie_load_given_ucode(trans, fw);
+	iwl_enable_interrupts(trans);
+
+	/* re-check RF-Kill state since we may have missed the interrupt */
+	hw_rfkill = iwl_is_rfkill_set(trans);
+	if (hw_rfkill)
+		set_bit(STATUS_RFKILL, &trans->status);
+	else
+		clear_bit(STATUS_RFKILL, &trans->status);
+
+	iwl_trans_pcie_rf_kill(trans, hw_rfkill);
+	if (hw_rfkill && !run_in_rfkill)
+		ret = -ERFKILL;
+
+out:
+	mutex_unlock(&trans_pcie->mutex);
+	return ret;
+}
+
+static void iwl_trans_pcie_fw_alive(struct iwl_trans *trans, u32 scd_addr)
+{
+	iwl_pcie_reset_ict(trans);
+	iwl_pcie_tx_start(trans, scd_addr);
+}
+
 static void iwl_trans_pcie_stop_device(struct iwl_trans *trans, bool low_power)
 {
 	struct iwl_trans_pcie *trans_pcie = IWL_TRANS_GET_PCIE_TRANS(trans);
diff --git a/drivers/net/wireless/realtek/rtlwifi/rc.c b/drivers/net/wireless/realtek/rtlwifi/rc.c
index 74c14ce28238..28f7010e7108 100644
--- a/drivers/net/wireless/realtek/rtlwifi/rc.c
+++ b/drivers/net/wireless/realtek/rtlwifi/rc.c
@@ -138,6 +138,11 @@ static void _rtl_rc_rate_set_series(struct rtl_priv *rtlpriv,
 		    ((wireless_mode == WIRELESS_MODE_N_5G) ||
 		     (wireless_mode == WIRELESS_MODE_N_24G)))
 			rate->flags |= IEEE80211_TX_RC_MCS;
+		if (sta && sta->vht_cap.vht_supported &&
+		    (wireless_mode == WIRELESS_MODE_AC_5G ||
+		     wireless_mode == WIRELESS_MODE_AC_24G ||
+		     wireless_mode == WIRELESS_MODE_AC_ONLY))
+			rate->flags |= IEEE80211_TX_RC_VHT_MCS;
 	}
 }
 
diff --git a/drivers/net/wireless/ti/wlcore/io.c b/drivers/net/wireless/ti/wlcore/io.c
index 9ac118e727e9..564ca750c5ee 100644
--- a/drivers/net/wireless/ti/wlcore/io.c
+++ b/drivers/net/wireless/ti/wlcore/io.c
@@ -175,14 +175,14 @@ int wlcore_set_partition(struct wl1271 *wl,
 	if (ret < 0)
 		goto out;
 
+	/* We don't need the size of the last partition, as it is
+	 * automatically calculated based on the total memory size and
+	 * the sizes of the previous partitions.
+	 */
 	ret = wlcore_raw_write32(wl, HW_PART3_START_ADDR, p->mem3.start);
 	if (ret < 0)
 		goto out;
 
-	ret = wlcore_raw_write32(wl, HW_PART3_SIZE_ADDR, p->mem3.size);
-	if (ret < 0)
-		goto out;
-
 out:
 	return ret;
 }
diff --git a/drivers/net/wireless/ti/wlcore/io.h b/drivers/net/wireless/ti/wlcore/io.h
index 6c257b54f415..10cf3747694d 100644
--- a/drivers/net/wireless/ti/wlcore/io.h
+++ b/drivers/net/wireless/ti/wlcore/io.h
@@ -36,8 +36,8 @@
 #define HW_PART1_START_ADDR             (HW_PARTITION_REGISTERS_ADDR + 12)
 #define HW_PART2_SIZE_ADDR              (HW_PARTITION_REGISTERS_ADDR + 16)
 #define HW_PART2_START_ADDR             (HW_PARTITION_REGISTERS_ADDR + 20)
-#define HW_PART3_SIZE_ADDR              (HW_PARTITION_REGISTERS_ADDR + 24)
-#define HW_PART3_START_ADDR             (HW_PARTITION_REGISTERS_ADDR + 28)
+#define HW_PART3_START_ADDR             (HW_PARTITION_REGISTERS_ADDR + 24)
+
 #define HW_ACCESS_REGISTER_SIZE         4
 
 #define HW_ACCESS_PRAM_MAX_RANGE	0x3c000
diff --git a/drivers/nvdimm/bus.c b/drivers/nvdimm/bus.c
index 7e2c43f701bc..5d28e9405f32 100644
--- a/drivers/nvdimm/bus.c
+++ b/drivers/nvdimm/bus.c
@@ -382,18 +382,18 @@ static const struct nd_cmd_desc __nd_cmd_bus_descs[] = {
 	[ND_CMD_ARS_CAP] = {
 		.in_num = 2,
 		.in_sizes = { 8, 8, },
-		.out_num = 2,
-		.out_sizes = { 4, 4, },
+		.out_num = 4,
+		.out_sizes = { 4, 4, 4, 4, },
 	},
 	[ND_CMD_ARS_START] = {
-		.in_num = 4,
-		.in_sizes = { 8, 8, 2, 6, },
-		.out_num = 1,
-		.out_sizes = { 4, },
+		.in_num = 5,
+		.in_sizes = { 8, 8, 2, 1, 5, },
+		.out_num = 2,
+		.out_sizes = { 4, 4, },
 	},
 	[ND_CMD_ARS_STATUS] = {
-		.out_num = 2,
-		.out_sizes = { 4, UINT_MAX, },
+		.out_num = 3,
+		.out_sizes = { 4, 4, UINT_MAX, },
 	},
 };
 
@@ -442,8 +442,8 @@ u32 nd_cmd_out_size(struct nvdimm *nvdimm, int cmd,
 		return in_field[1];
 	else if (nvdimm && cmd == ND_CMD_VENDOR && idx == 2)
 		return out_field[1];
-	else if (!nvdimm && cmd == ND_CMD_ARS_STATUS && idx == 1)
-		return ND_CMD_ARS_STATUS_MAX;
+	else if (!nvdimm && cmd == ND_CMD_ARS_STATUS && idx == 2)
+		return out_field[1] - 8;
 
 	return UINT_MAX;
 }
diff --git a/drivers/nvdimm/pmem.c b/drivers/nvdimm/pmem.c
index 7edf31671dab..8d0b54670184 100644
--- a/drivers/nvdimm/pmem.c
+++ b/drivers/nvdimm/pmem.c
@@ -41,7 +41,7 @@ struct pmem_device {
 	phys_addr_t		phys_addr;
 	/* when non-zero this device is hosting a 'pfn' instance */
 	phys_addr_t		data_offset;
-	unsigned long		pfn_flags;
+	u64			pfn_flags;
 	void __pmem		*virt_addr;
 	size_t			size;
 	struct badblocks	bb;
diff --git a/drivers/nvme/host/Kconfig b/drivers/nvme/host/Kconfig
index 5d6237391dcd..b586d84f2518 100644
--- a/drivers/nvme/host/Kconfig
+++ b/drivers/nvme/host/Kconfig
@@ -17,5 +17,6 @@ config BLK_DEV_NVME_SCSI
 	  and block devices nodes, as well a a translation for a small
 	  number of selected SCSI commands to NVMe commands to the NVMe
 	  driver.  If you don't know what this means you probably want
-	  to say N here, and if you know what it means you probably
-	  want to say N as well.
+	  to say N here, unless you run a distro that abuses the SCSI
+	  emulation to provide stable device names for mount by id, like
+	  some OpenSuSE and SLES versions.
diff --git a/drivers/nvme/host/core.c b/drivers/nvme/host/core.c
index c5bf001af559..03c46412fff4 100644
--- a/drivers/nvme/host/core.c
+++ b/drivers/nvme/host/core.c
@@ -55,8 +55,9 @@ static void nvme_free_ns(struct kref *kref)
 	ns->disk->private_data = NULL;
 	spin_unlock(&dev_list_lock);
 
-	nvme_put_ctrl(ns->ctrl);
 	put_disk(ns->disk);
+	ida_simple_remove(&ns->ctrl->ns_ida, ns->instance);
+	nvme_put_ctrl(ns->ctrl);
 	kfree(ns);
 }
 
@@ -183,7 +184,7 @@ int __nvme_submit_user_cmd(struct request_queue *q, struct nvme_command *cmd,
 			goto out_unmap;
 		}
 
-		if (meta_buffer) {
+		if (meta_buffer && meta_len) {
 			struct bio_integrity_payload *bip;
 
 			meta = kmalloc(meta_len, GFP_KERNEL);
@@ -373,6 +374,8 @@ static int nvme_submit_io(struct nvme_ns *ns, struct nvme_user_io __user *uio)
 
 	if (copy_from_user(&io, uio, sizeof(io)))
 		return -EFAULT;
+	if (io.flags)
+		return -EINVAL;
 
 	switch (io.opcode) {
 	case nvme_cmd_write:
@@ -424,6 +427,8 @@ static int nvme_user_cmd(struct nvme_ctrl *ctrl, struct nvme_ns *ns,
 		return -EACCES;
 	if (copy_from_user(&cmd, ucmd, sizeof(cmd)))
 		return -EFAULT;
+	if (cmd.flags)
+		return -EINVAL;
 
 	memset(&c, 0, sizeof(c));
 	c.common.opcode = cmd.opcode;
@@ -556,6 +561,10 @@ static int nvme_revalidate_disk(struct gendisk *disk)
 	u16 old_ms;
 	unsigned short bs;
 
+	if (test_bit(NVME_NS_DEAD, &ns->flags)) {
+		set_capacity(disk, 0);
+		return -ENODEV;
+	}
 	if (nvme_identify_ns(ns->ctrl, ns->ns_id, &id)) {
 		dev_warn(ns->ctrl->dev, "%s: Identify failure nvme%dn%d\n",
 				__func__, ns->ctrl->instance, ns->ns_id);
@@ -831,6 +840,23 @@ int nvme_shutdown_ctrl(struct nvme_ctrl *ctrl)
 	return ret;
 }
 
+static void nvme_set_queue_limits(struct nvme_ctrl *ctrl,
+		struct request_queue *q)
+{
+	if (ctrl->max_hw_sectors) {
+		u32 max_segments =
+			(ctrl->max_hw_sectors / (ctrl->page_size >> 9)) + 1;
+
+		blk_queue_max_hw_sectors(q, ctrl->max_hw_sectors);
+		blk_queue_max_segments(q, min_t(u32, max_segments, USHRT_MAX));
+	}
+	if (ctrl->stripe_size)
+		blk_queue_chunk_sectors(q, ctrl->stripe_size >> 9);
+	if (ctrl->vwc & NVME_CTRL_VWC_PRESENT)
+		blk_queue_flush(q, REQ_FLUSH | REQ_FUA);
+	blk_queue_virt_boundary(q, ctrl->page_size - 1);
+}
+
 /*
  * Initialize the cached copies of the Identify data and various controller
  * register in our nvme_ctrl structure.  This should be called as soon as
@@ -888,6 +914,8 @@ int nvme_init_identify(struct nvme_ctrl *ctrl)
 		}
 	}
 
+	nvme_set_queue_limits(ctrl, ctrl->admin_q);
+
 	kfree(id);
 	return 0;
 }
@@ -1118,10 +1146,13 @@ static void nvme_alloc_ns(struct nvme_ctrl *ctrl, unsigned nsid)
 	if (!ns)
 		return;
 
+	ns->instance = ida_simple_get(&ctrl->ns_ida, 1, 0, GFP_KERNEL);
+	if (ns->instance < 0)
+		goto out_free_ns;
+
 	ns->queue = blk_mq_init_queue(ctrl->tagset);
 	if (IS_ERR(ns->queue))
-		goto out_free_ns;
-	queue_flag_set_unlocked(QUEUE_FLAG_NOMERGES, ns->queue);
+		goto out_release_instance;
 	queue_flag_set_unlocked(QUEUE_FLAG_NONROT, ns->queue);
 	ns->queue->queuedata = ns;
 	ns->ctrl = ctrl;
@@ -1135,17 +1166,9 @@ static void nvme_alloc_ns(struct nvme_ctrl *ctrl, unsigned nsid)
 	ns->disk = disk;
 	ns->lba_shift = 9; /* set to a default value for 512 until disk is validated */
 
+
 	blk_queue_logical_block_size(ns->queue, 1 << ns->lba_shift);
-	if (ctrl->max_hw_sectors) {
-		blk_queue_max_hw_sectors(ns->queue, ctrl->max_hw_sectors);
-		blk_queue_max_segments(ns->queue,
-			(ctrl->max_hw_sectors / (ctrl->page_size >> 9)) + 1);
-	}
-	if (ctrl->stripe_size)
-		blk_queue_chunk_sectors(ns->queue, ctrl->stripe_size >> 9);
-	if (ctrl->vwc & NVME_CTRL_VWC_PRESENT)
-		blk_queue_flush(ns->queue, REQ_FLUSH | REQ_FUA);
-	blk_queue_virt_boundary(ns->queue, ctrl->page_size - 1);
+	nvme_set_queue_limits(ctrl, ns->queue);
 
 	disk->major = nvme_major;
 	disk->first_minor = 0;
@@ -1154,7 +1177,7 @@ static void nvme_alloc_ns(struct nvme_ctrl *ctrl, unsigned nsid)
 	disk->queue = ns->queue;
 	disk->driverfs_dev = ctrl->device;
 	disk->flags = GENHD_FL_EXT_DEVT;
-	sprintf(disk->disk_name, "nvme%dn%d", ctrl->instance, nsid);
+	sprintf(disk->disk_name, "nvme%dn%d", ctrl->instance, ns->instance);
 
 	if (nvme_revalidate_disk(ns->disk))
 		goto out_free_disk;
@@ -1174,40 +1197,29 @@ static void nvme_alloc_ns(struct nvme_ctrl *ctrl, unsigned nsid)
 	kfree(disk);
  out_free_queue:
 	blk_cleanup_queue(ns->queue);
+ out_release_instance:
+	ida_simple_remove(&ctrl->ns_ida, ns->instance);
  out_free_ns:
 	kfree(ns);
 }
 
 static void nvme_ns_remove(struct nvme_ns *ns)
 {
-	bool kill = nvme_io_incapable(ns->ctrl) &&
-			!blk_queue_dying(ns->queue);
-
-	lockdep_assert_held(&ns->ctrl->namespaces_mutex);
-
-	if (kill) {
-		blk_set_queue_dying(ns->queue);
+	if (test_and_set_bit(NVME_NS_REMOVING, &ns->flags))
+		return;
 
-		/*
-		 * The controller was shutdown first if we got here through
-		 * device removal. The shutdown may requeue outstanding
-		 * requests. These need to be aborted immediately so
-		 * del_gendisk doesn't block indefinitely for their completion.
-		 */
-		blk_mq_abort_requeue_list(ns->queue);
-	}
 	if (ns->disk->flags & GENHD_FL_UP) {
 		if (blk_get_integrity(ns->disk))
 			blk_integrity_unregister(ns->disk);
 		sysfs_remove_group(&disk_to_dev(ns->disk)->kobj,
 					&nvme_ns_attr_group);
 		del_gendisk(ns->disk);
-	}
-	if (kill || !blk_queue_dying(ns->queue)) {
 		blk_mq_abort_requeue_list(ns->queue);
 		blk_cleanup_queue(ns->queue);
 	}
+	mutex_lock(&ns->ctrl->namespaces_mutex);
 	list_del_init(&ns->list);
+	mutex_unlock(&ns->ctrl->namespaces_mutex);
 	nvme_put_ns(ns);
 }
 
@@ -1301,10 +1313,8 @@ void nvme_remove_namespaces(struct nvme_ctrl *ctrl)
 {
 	struct nvme_ns *ns, *next;
 
-	mutex_lock(&ctrl->namespaces_mutex);
 	list_for_each_entry_safe(ns, next, &ctrl->namespaces, list)
 		nvme_ns_remove(ns);
-	mutex_unlock(&ctrl->namespaces_mutex);
 }
 
 static DEFINE_IDA(nvme_instance_ida);
@@ -1351,6 +1361,7 @@ static void nvme_free_ctrl(struct kref *kref)
 
 	put_device(ctrl->device);
 	nvme_release_instance(ctrl);
+	ida_destroy(&ctrl->ns_ida);
 
 	ctrl->ops->free_ctrl(ctrl);
 }
@@ -1391,6 +1402,7 @@ int nvme_init_ctrl(struct nvme_ctrl *ctrl, struct device *dev,
 	}
 	get_device(ctrl->device);
 	dev_set_drvdata(ctrl->device, ctrl);
+	ida_init(&ctrl->ns_ida);
 
 	spin_lock(&dev_list_lock);
 	list_add_tail(&ctrl->node, &nvme_ctrl_list);
@@ -1403,6 +1415,38 @@ out:
 	return ret;
 }
 
+/**
+ * nvme_kill_queues(): Ends all namespace queues
+ * @ctrl: the dead controller that needs to end
+ *
+ * Call this function when the driver determines it is unable to get the
+ * controller in a state capable of servicing IO.
+ */
+void nvme_kill_queues(struct nvme_ctrl *ctrl)
+{
+	struct nvme_ns *ns;
+
+	mutex_lock(&ctrl->namespaces_mutex);
+	list_for_each_entry(ns, &ctrl->namespaces, list) {
+		if (!kref_get_unless_zero(&ns->kref))
+			continue;
+
+		/*
+		 * Revalidating a dead namespace sets capacity to 0. This will
+		 * end buffered writers dirtying pages that can't be synced.
+		 */
+		if (!test_and_set_bit(NVME_NS_DEAD, &ns->flags))
+			revalidate_disk(ns->disk);
+
+		blk_set_queue_dying(ns->queue);
+		blk_mq_abort_requeue_list(ns->queue);
+		blk_mq_start_stopped_hw_queues(ns->queue, true);
+
+		nvme_put_ns(ns);
+	}
+	mutex_unlock(&ctrl->namespaces_mutex);
+}
+
 void nvme_stop_queues(struct nvme_ctrl *ctrl)
 {
 	struct nvme_ns *ns;
diff --git a/drivers/nvme/host/lightnvm.c b/drivers/nvme/host/lightnvm.c
index 5cd3725e2fa4..6bb15e4926dc 100644
--- a/drivers/nvme/host/lightnvm.c
+++ b/drivers/nvme/host/lightnvm.c
@@ -146,9 +146,10 @@ struct nvme_nvm_command {
 	};
 };
 
+#define NVME_NVM_LP_MLC_PAIRS 886
 struct nvme_nvm_lp_mlc {
 	__u16			num_pairs;
-	__u8			pairs[886];
+	__u8			pairs[NVME_NVM_LP_MLC_PAIRS];
 };
 
 struct nvme_nvm_lp_tbl {
@@ -282,9 +283,14 @@ static int init_grps(struct nvm_id *nvm_id, struct nvme_nvm_id *nvme_nvm_id)
 			memcpy(dst->lptbl.id, src->lptbl.id, 8);
 			dst->lptbl.mlc.num_pairs =
 					le16_to_cpu(src->lptbl.mlc.num_pairs);
-			/* 4 bits per pair */
+
+			if (dst->lptbl.mlc.num_pairs > NVME_NVM_LP_MLC_PAIRS) {
+				pr_err("nvm: number of MLC pairs not supported\n");
+				return -EINVAL;
+			}
+
 			memcpy(dst->lptbl.mlc.pairs, src->lptbl.mlc.pairs,
-						dst->lptbl.mlc.num_pairs >> 1);
+						dst->lptbl.mlc.num_pairs);
 		}
 	}
 
diff --git a/drivers/nvme/host/nvme.h b/drivers/nvme/host/nvme.h
index 4fb5bb737868..fb15ba5f5d19 100644
--- a/drivers/nvme/host/nvme.h
+++ b/drivers/nvme/host/nvme.h
@@ -72,6 +72,7 @@ struct nvme_ctrl {
 	struct mutex namespaces_mutex;
 	struct device *device;	/* char device */
 	struct list_head node;
+	struct ida ns_ida;
 
 	char name[12];
 	char serial[20];
@@ -102,6 +103,7 @@ struct nvme_ns {
 	struct request_queue *queue;
 	struct gendisk *disk;
 	struct kref kref;
+	int instance;
 
 	u8 eui[8];
 	u8 uuid[16];
@@ -112,6 +114,11 @@ struct nvme_ns {
 	bool ext;
 	u8 pi_type;
 	int type;
+	unsigned long flags;
+
+#define NVME_NS_REMOVING 0
+#define NVME_NS_DEAD     1
+
 	u64 mode_select_num_blocks;
 	u32 mode_select_block_len;
 };
@@ -139,9 +146,9 @@ static inline bool nvme_io_incapable(struct nvme_ctrl *ctrl)
 	u32 val = 0;
 
 	if (ctrl->ops->io_incapable(ctrl))
-		return false;
+		return true;
 	if (ctrl->ops->reg_read32(ctrl, NVME_REG_CSTS, &val))
-		return false;
+		return true;
 	return val & NVME_CSTS_CFS;
 }
 
@@ -240,6 +247,7 @@ void nvme_remove_namespaces(struct nvme_ctrl *ctrl);
 
 void nvme_stop_queues(struct nvme_ctrl *ctrl);
 void nvme_start_queues(struct nvme_ctrl *ctrl);
+void nvme_kill_queues(struct nvme_ctrl *ctrl);
 
 struct request *nvme_alloc_request(struct request_queue *q,
 		struct nvme_command *cmd, unsigned int flags);
diff --git a/drivers/nvme/host/pci.c b/drivers/nvme/host/pci.c
index 72ef8322d32a..680f5780750c 100644
--- a/drivers/nvme/host/pci.c
+++ b/drivers/nvme/host/pci.c
@@ -86,7 +86,6 @@ struct nvme_queue;
 
 static int nvme_reset(struct nvme_dev *dev);
 static void nvme_process_cq(struct nvme_queue *nvmeq);
-static void nvme_remove_dead_ctrl(struct nvme_dev *dev);
 static void nvme_dev_disable(struct nvme_dev *dev, bool shutdown);
 
 /*
@@ -120,6 +119,7 @@ struct nvme_dev {
 	unsigned long flags;
 
 #define NVME_CTRL_RESETTING    0
+#define NVME_CTRL_REMOVING     1
 
 	struct nvme_ctrl ctrl;
 	struct completion ioq_wait;
@@ -286,6 +286,17 @@ static int nvme_init_request(void *data, struct request *req,
 	return 0;
 }
 
+static void nvme_queue_scan(struct nvme_dev *dev)
+{
+	/*
+	 * Do not queue new scan work when a controller is reset during
+	 * removal.
+	 */
+	if (test_bit(NVME_CTRL_REMOVING, &dev->flags))
+		return;
+	queue_work(nvme_workq, &dev->scan_work);
+}
+
 static void nvme_complete_async_event(struct nvme_dev *dev,
 		struct nvme_completion *cqe)
 {
@@ -300,7 +311,7 @@ static void nvme_complete_async_event(struct nvme_dev *dev,
 	switch (result & 0xff07) {
 	case NVME_AER_NOTICE_NS_CHANGED:
 		dev_info(dev->dev, "rescanning\n");
-		queue_work(nvme_workq, &dev->scan_work);
+		nvme_queue_scan(dev);
 	default:
 		dev_warn(dev->dev, "async event result %08x\n", result);
 	}
@@ -678,6 +689,14 @@ static int nvme_queue_rq(struct blk_mq_hw_ctx *hctx,
 	blk_mq_start_request(req);
 
 	spin_lock_irq(&nvmeq->q_lock);
+	if (unlikely(nvmeq->cq_vector < 0)) {
+		if (ns && !test_bit(NVME_NS_DEAD, &ns->flags))
+			ret = BLK_MQ_RQ_QUEUE_BUSY;
+		else
+			ret = BLK_MQ_RQ_QUEUE_ERROR;
+		spin_unlock_irq(&nvmeq->q_lock);
+		goto out;
+	}
 	__nvme_submit_cmd(nvmeq, &cmnd);
 	nvme_process_cq(nvmeq);
 	spin_unlock_irq(&nvmeq->q_lock);
@@ -999,7 +1018,7 @@ static void nvme_cancel_queue_ios(struct request *req, void *data, bool reserved
 	if (!blk_mq_request_started(req))
 		return;
 
-	dev_warn(nvmeq->q_dmadev,
+	dev_dbg_ratelimited(nvmeq->q_dmadev,
 		 "Cancelling I/O %d QID %d\n", req->tag, nvmeq->qid);
 
 	status = NVME_SC_ABORT_REQ;
@@ -1245,6 +1264,12 @@ static struct blk_mq_ops nvme_mq_ops = {
 static void nvme_dev_remove_admin(struct nvme_dev *dev)
 {
 	if (dev->ctrl.admin_q && !blk_queue_dying(dev->ctrl.admin_q)) {
+		/*
+		 * If the controller was reset during removal, it's possible
+		 * user requests may be waiting on a stopped queue. Start the
+		 * queue to flush these to completion.
+		 */
+		blk_mq_start_stopped_hw_queues(dev->ctrl.admin_q, true);
 		blk_cleanup_queue(dev->ctrl.admin_q);
 		blk_mq_free_tag_set(&dev->admin_tagset);
 	}
@@ -1685,14 +1710,14 @@ static int nvme_dev_add(struct nvme_dev *dev)
 			return 0;
 		dev->ctrl.tagset = &dev->tagset;
 	}
-	queue_work(nvme_workq, &dev->scan_work);
+	nvme_queue_scan(dev);
 	return 0;
 }
 
-static int nvme_dev_map(struct nvme_dev *dev)
+static int nvme_pci_enable(struct nvme_dev *dev)
 {
 	u64 cap;
-	int bars, result = -ENOMEM;
+	int result = -ENOMEM;
 	struct pci_dev *pdev = to_pci_dev(dev->dev);
 
 	if (pci_enable_device_mem(pdev))
@@ -1700,24 +1725,14 @@ static int nvme_dev_map(struct nvme_dev *dev)
 
 	dev->entry[0].vector = pdev->irq;
 	pci_set_master(pdev);
-	bars = pci_select_bars(pdev, IORESOURCE_MEM);
-	if (!bars)
-		goto disable_pci;
-
-	if (pci_request_selected_regions(pdev, bars, "nvme"))
-		goto disable_pci;
 
 	if (dma_set_mask_and_coherent(dev->dev, DMA_BIT_MASK(64)) &&
 	    dma_set_mask_and_coherent(dev->dev, DMA_BIT_MASK(32)))
 		goto disable;
 
-	dev->bar = ioremap(pci_resource_start(pdev, 0), 8192);
-	if (!dev->bar)
-		goto disable;
-
 	if (readl(dev->bar + NVME_REG_CSTS) == -1) {
 		result = -ENODEV;
-		goto unmap;
+		goto disable;
 	}
 
 	/*
@@ -1727,7 +1742,7 @@ static int nvme_dev_map(struct nvme_dev *dev)
 	if (!pdev->irq) {
 		result = pci_enable_msix(pdev, dev->entry, 1);
 		if (result < 0)
-			goto unmap;
+			goto disable;
 	}
 
 	cap = lo_hi_readq(dev->bar + NVME_REG_CAP);
@@ -1754,18 +1769,20 @@ static int nvme_dev_map(struct nvme_dev *dev)
 	pci_save_state(pdev);
 	return 0;
 
- unmap:
-	iounmap(dev->bar);
-	dev->bar = NULL;
  disable:
-	pci_release_regions(pdev);
- disable_pci:
 	pci_disable_device(pdev);
 	return result;
 }
 
 static void nvme_dev_unmap(struct nvme_dev *dev)
 {
+	if (dev->bar)
+		iounmap(dev->bar);
+	pci_release_regions(to_pci_dev(dev->dev));
+}
+
+static void nvme_pci_disable(struct nvme_dev *dev)
+{
 	struct pci_dev *pdev = to_pci_dev(dev->dev);
 
 	if (pdev->msi_enabled)
@@ -1773,12 +1790,6 @@ static void nvme_dev_unmap(struct nvme_dev *dev)
 	else if (pdev->msix_enabled)
 		pci_disable_msix(pdev);
 
-	if (dev->bar) {
-		iounmap(dev->bar);
-		dev->bar = NULL;
-		pci_release_regions(pdev);
-	}
-
 	if (pci_is_enabled(pdev)) {
 		pci_disable_pcie_error_reporting(pdev);
 		pci_disable_device(pdev);
@@ -1837,7 +1848,7 @@ static void nvme_dev_disable(struct nvme_dev *dev, bool shutdown)
 	nvme_dev_list_remove(dev);
 
 	mutex_lock(&dev->shutdown_lock);
-	if (dev->bar) {
+	if (pci_is_enabled(to_pci_dev(dev->dev))) {
 		nvme_stop_queues(&dev->ctrl);
 		csts = readl(dev->bar + NVME_REG_CSTS);
 	}
@@ -1850,7 +1861,7 @@ static void nvme_dev_disable(struct nvme_dev *dev, bool shutdown)
 		nvme_disable_io_queues(dev);
 		nvme_disable_admin_queue(dev, shutdown);
 	}
-	nvme_dev_unmap(dev);
+	nvme_pci_disable(dev);
 
 	for (i = dev->queue_count - 1; i >= 0; i--)
 		nvme_clear_queue(dev->queues[i]);
@@ -1894,10 +1905,20 @@ static void nvme_pci_free_ctrl(struct nvme_ctrl *ctrl)
 	kfree(dev);
 }
 
+static void nvme_remove_dead_ctrl(struct nvme_dev *dev, int status)
+{
+	dev_warn(dev->dev, "Removing after probe failure status: %d\n", status);
+
+	kref_get(&dev->ctrl.kref);
+	nvme_dev_disable(dev, false);
+	if (!schedule_work(&dev->remove_work))
+		nvme_put_ctrl(&dev->ctrl);
+}
+
 static void nvme_reset_work(struct work_struct *work)
 {
 	struct nvme_dev *dev = container_of(work, struct nvme_dev, reset_work);
-	int result;
+	int result = -ENODEV;
 
 	if (WARN_ON(test_bit(NVME_CTRL_RESETTING, &dev->flags)))
 		goto out;
@@ -1906,37 +1927,37 @@ static void nvme_reset_work(struct work_struct *work)
 	 * If we're called to reset a live controller first shut it down before
 	 * moving on.
 	 */
-	if (dev->bar)
+	if (dev->ctrl.ctrl_config & NVME_CC_ENABLE)
 		nvme_dev_disable(dev, false);
 
 	set_bit(NVME_CTRL_RESETTING, &dev->flags);
 
-	result = nvme_dev_map(dev);
+	result = nvme_pci_enable(dev);
 	if (result)
 		goto out;
 
 	result = nvme_configure_admin_queue(dev);
 	if (result)
-		goto unmap;
+		goto out;
 
 	nvme_init_queue(dev->queues[0], 0);
 	result = nvme_alloc_admin_tags(dev);
 	if (result)
-		goto disable;
+		goto out;
 
 	result = nvme_init_identify(&dev->ctrl);
 	if (result)
-		goto free_tags;
+		goto out;
 
 	result = nvme_setup_io_queues(dev);
 	if (result)
-		goto free_tags;
+		goto out;
 
 	dev->ctrl.event_limit = NVME_NR_AEN_COMMANDS;
 
 	result = nvme_dev_list_add(dev);
 	if (result)
-		goto remove;
+		goto out;
 
 	/*
 	 * Keep the controller around but remove all namespaces if we don't have
@@ -1953,19 +1974,8 @@ static void nvme_reset_work(struct work_struct *work)
 	clear_bit(NVME_CTRL_RESETTING, &dev->flags);
 	return;
 
- remove:
-	nvme_dev_list_remove(dev);
- free_tags:
-	nvme_dev_remove_admin(dev);
-	blk_put_queue(dev->ctrl.admin_q);
-	dev->ctrl.admin_q = NULL;
-	dev->queues[0]->tags = NULL;
- disable:
-	nvme_disable_admin_queue(dev, false);
- unmap:
-	nvme_dev_unmap(dev);
  out:
-	nvme_remove_dead_ctrl(dev);
+	nvme_remove_dead_ctrl(dev, result);
 }
 
 static void nvme_remove_dead_ctrl_work(struct work_struct *work)
@@ -1973,19 +1983,12 @@ static void nvme_remove_dead_ctrl_work(struct work_struct *work)
 	struct nvme_dev *dev = container_of(work, struct nvme_dev, remove_work);
 	struct pci_dev *pdev = to_pci_dev(dev->dev);
 
+	nvme_kill_queues(&dev->ctrl);
 	if (pci_get_drvdata(pdev))
 		pci_stop_and_remove_bus_device_locked(pdev);
 	nvme_put_ctrl(&dev->ctrl);
 }
 
-static void nvme_remove_dead_ctrl(struct nvme_dev *dev)
-{
-	dev_warn(dev->dev, "Removing after probe failure\n");
-	kref_get(&dev->ctrl.kref);
-	if (!schedule_work(&dev->remove_work))
-		nvme_put_ctrl(&dev->ctrl);
-}
-
 static int nvme_reset(struct nvme_dev *dev)
 {
 	if (!dev->ctrl.admin_q || blk_queue_dying(dev->ctrl.admin_q))
@@ -2037,6 +2040,27 @@ static const struct nvme_ctrl_ops nvme_pci_ctrl_ops = {
 	.free_ctrl		= nvme_pci_free_ctrl,
 };
 
+static int nvme_dev_map(struct nvme_dev *dev)
+{
+	int bars;
+	struct pci_dev *pdev = to_pci_dev(dev->dev);
+
+	bars = pci_select_bars(pdev, IORESOURCE_MEM);
+	if (!bars)
+		return -ENODEV;
+	if (pci_request_selected_regions(pdev, bars, "nvme"))
+		return -ENODEV;
+
+	dev->bar = ioremap(pci_resource_start(pdev, 0), 8192);
+	if (!dev->bar)
+		goto release;
+
+       return 0;
+  release:
+       pci_release_regions(pdev);
+       return -ENODEV;
+}
+
 static int nvme_probe(struct pci_dev *pdev, const struct pci_device_id *id)
 {
 	int node, result = -ENOMEM;
@@ -2061,6 +2085,10 @@ static int nvme_probe(struct pci_dev *pdev, const struct pci_device_id *id)
 	dev->dev = get_device(&pdev->dev);
 	pci_set_drvdata(pdev, dev);
 
+	result = nvme_dev_map(dev);
+	if (result)
+		goto free;
+
 	INIT_LIST_HEAD(&dev->node);
 	INIT_WORK(&dev->scan_work, nvme_dev_scan);
 	INIT_WORK(&dev->reset_work, nvme_reset_work);
@@ -2084,6 +2112,7 @@ static int nvme_probe(struct pci_dev *pdev, const struct pci_device_id *id)
 	nvme_release_prp_pools(dev);
  put_pci:
 	put_device(dev->dev);
+	nvme_dev_unmap(dev);
  free:
 	kfree(dev->queues);
 	kfree(dev->entry);
@@ -2107,24 +2136,27 @@ static void nvme_shutdown(struct pci_dev *pdev)
 	nvme_dev_disable(dev, true);
 }
 
+/*
+ * The driver's remove may be called on a device in a partially initialized
+ * state. This function must not have any dependencies on the device state in
+ * order to proceed.
+ */
 static void nvme_remove(struct pci_dev *pdev)
 {
 	struct nvme_dev *dev = pci_get_drvdata(pdev);
 
-	spin_lock(&dev_list_lock);
-	list_del_init(&dev->node);
-	spin_unlock(&dev_list_lock);
-
+	set_bit(NVME_CTRL_REMOVING, &dev->flags);
 	pci_set_drvdata(pdev, NULL);
-	flush_work(&dev->reset_work);
 	flush_work(&dev->scan_work);
 	nvme_remove_namespaces(&dev->ctrl);
 	nvme_uninit_ctrl(&dev->ctrl);
 	nvme_dev_disable(dev, true);
+	flush_work(&dev->reset_work);
 	nvme_dev_remove_admin(dev);
 	nvme_free_queues(dev, 0);
 	nvme_release_cmb(dev);
 	nvme_release_prp_pools(dev);
+	nvme_dev_unmap(dev);
 	nvme_put_ctrl(&dev->ctrl);
 }
 
diff --git a/drivers/nvmem/core.c b/drivers/nvmem/core.c
index 6fd4e5a5ef4a..9d11d9837312 100644
--- a/drivers/nvmem/core.c
+++ b/drivers/nvmem/core.c
@@ -70,6 +70,9 @@ static ssize_t bin_attr_nvmem_read(struct file *filp, struct kobject *kobj,
 	if (pos >= nvmem->size)
 		return 0;
 
+	if (count < nvmem->word_size)
+		return -EINVAL;
+
 	if (pos + count > nvmem->size)
 		count = nvmem->size - pos;
 
@@ -95,6 +98,9 @@ static ssize_t bin_attr_nvmem_write(struct file *filp, struct kobject *kobj,
 	if (pos >= nvmem->size)
 		return 0;
 
+	if (count < nvmem->word_size)
+		return -EINVAL;
+
 	if (pos + count > nvmem->size)
 		count = nvmem->size - pos;
 
diff --git a/drivers/nvmem/qfprom.c b/drivers/nvmem/qfprom.c
index afb67e7eeee4..3829e5fbf8c3 100644
--- a/drivers/nvmem/qfprom.c
+++ b/drivers/nvmem/qfprom.c
@@ -21,6 +21,7 @@ static struct regmap_config qfprom_regmap_config = {
 	.reg_bits = 32,
 	.val_bits = 8,
 	.reg_stride = 1,
+	.val_format_endian = REGMAP_ENDIAN_LITTLE,
 };
 
 static struct nvmem_config econfig = {
diff --git a/drivers/of/irq.c b/drivers/of/irq.c
index 7ee21ae305ae..e7bfc175b8e1 100644
--- a/drivers/of/irq.c
+++ b/drivers/of/irq.c
@@ -635,6 +635,13 @@ static u32 __of_msi_map_rid(struct device *dev, struct device_node **np,
 		msi_base = be32_to_cpup(msi_map + 2);
 		rid_len = be32_to_cpup(msi_map + 3);
 
+		if (rid_base & ~map_mask) {
+			dev_err(parent_dev,
+				"Invalid msi-map translation - msi-map-mask (0x%x) ignores rid-base (0x%x)\n",
+				map_mask, rid_base);
+			return rid_out;
+		}
+
 		msi_controller_node = of_find_node_by_phandle(phandle);
 
 		matched = (masked_rid >= rid_base &&
@@ -654,7 +661,7 @@ static u32 __of_msi_map_rid(struct device *dev, struct device_node **np,
 	if (!matched)
 		return rid_out;
 
-	rid_out = masked_rid + msi_base;
+	rid_out = masked_rid - rid_base + msi_base;
 	dev_dbg(dev,
 		"msi-map at: %s, using mask %08x, rid-base: %08x, msi-base: %08x, length: %08x, rid: %08x -> %08x\n",
 		dev_name(parent_dev), map_mask, rid_base, msi_base,
diff --git a/drivers/of/of_mdio.c b/drivers/of/of_mdio.c
index 5648317d355f..365dc7e83ab4 100644
--- a/drivers/of/of_mdio.c
+++ b/drivers/of/of_mdio.c
@@ -154,6 +154,7 @@ static const struct of_device_id whitelist_phys[] = {
 	{ .compatible = "marvell,88E1111", },
 	{ .compatible = "marvell,88e1116", },
 	{ .compatible = "marvell,88e1118", },
+	{ .compatible = "marvell,88e1145", },
 	{ .compatible = "marvell,88e1149r", },
 	{ .compatible = "marvell,88e1310", },
 	{ .compatible = "marvell,88E1510", },
@@ -304,6 +305,7 @@ EXPORT_SYMBOL(of_phy_find_device);
  * @dev: pointer to net_device claiming the phy
  * @phy_np: Pointer to device tree node for the PHY
  * @hndlr: Link state callback for the network device
+ * @flags: flags to pass to the PHY
  * @iface: PHY data interface type
  *
  * If successful, returns a pointer to the phy_device with the embedded
diff --git a/drivers/pci/host/Kconfig b/drivers/pci/host/Kconfig
index 75a605426538..d1cdd9c992ac 100644
--- a/drivers/pci/host/Kconfig
+++ b/drivers/pci/host/Kconfig
@@ -14,6 +14,7 @@ config PCI_DRA7XX
 config PCI_MVEBU
 	bool "Marvell EBU PCIe controller"
 	depends on ARCH_MVEBU || ARCH_DOVE
+	depends on ARM
 	depends on OF
 
 config PCIE_DW
diff --git a/drivers/pci/host/pci-keystone-dw.c b/drivers/pci/host/pci-keystone-dw.c
index ed34c9520a02..6153853ca9c3 100644
--- a/drivers/pci/host/pci-keystone-dw.c
+++ b/drivers/pci/host/pci-keystone-dw.c
@@ -58,11 +58,6 @@
 
 #define to_keystone_pcie(x)	container_of(x, struct keystone_pcie, pp)
 
-static inline struct pcie_port *sys_to_pcie(struct pci_sys_data *sys)
-{
-	return sys->private_data;
-}
-
 static inline void update_reg_offset_bit_pos(u32 offset, u32 *reg_offset,
 					     u32 *bit_pos)
 {
@@ -108,7 +103,7 @@ static void ks_dw_pcie_msi_irq_ack(struct irq_data *d)
 	struct pcie_port *pp;
 
 	msi = irq_data_get_msi_desc(d);
-	pp = sys_to_pcie(msi_desc_to_pci_sysdata(msi));
+	pp = (struct pcie_port *) msi_desc_to_pci_sysdata(msi);
 	ks_pcie = to_keystone_pcie(pp);
 	offset = d->irq - irq_linear_revmap(pp->irq_domain, 0);
 	update_reg_offset_bit_pos(offset, &reg_offset, &bit_pos);
@@ -146,7 +141,7 @@ static void ks_dw_pcie_msi_irq_mask(struct irq_data *d)
 	u32 offset;
 
 	msi = irq_data_get_msi_desc(d);
-	pp = sys_to_pcie(msi_desc_to_pci_sysdata(msi));
+	pp = (struct pcie_port *) msi_desc_to_pci_sysdata(msi);
 	ks_pcie = to_keystone_pcie(pp);
 	offset = d->irq - irq_linear_revmap(pp->irq_domain, 0);
 
@@ -167,7 +162,7 @@ static void ks_dw_pcie_msi_irq_unmask(struct irq_data *d)
 	u32 offset;
 
 	msi = irq_data_get_msi_desc(d);
-	pp = sys_to_pcie(msi_desc_to_pci_sysdata(msi));
+	pp = (struct pcie_port *) msi_desc_to_pci_sysdata(msi);
 	ks_pcie = to_keystone_pcie(pp);
 	offset = d->irq - irq_linear_revmap(pp->irq_domain, 0);
 
diff --git a/drivers/pci/host/pci-layerscape.c b/drivers/pci/host/pci-layerscape.c
index 3923bed93c7e..f39961bcf7aa 100644
--- a/drivers/pci/host/pci-layerscape.c
+++ b/drivers/pci/host/pci-layerscape.c
@@ -77,6 +77,16 @@ static void ls_pcie_fix_class(struct ls_pcie *pcie)
 	iowrite16(PCI_CLASS_BRIDGE_PCI, pcie->dbi + PCI_CLASS_DEVICE);
 }
 
+/* Drop MSG TLP except for Vendor MSG */
+static void ls_pcie_drop_msg_tlp(struct ls_pcie *pcie)
+{
+	u32 val;
+
+	val = ioread32(pcie->dbi + PCIE_STRFMR1);
+	val &= 0xDFFFFFFF;
+	iowrite32(val, pcie->dbi + PCIE_STRFMR1);
+}
+
 static int ls1021_pcie_link_up(struct pcie_port *pp)
 {
 	u32 state;
@@ -97,7 +107,7 @@ static int ls1021_pcie_link_up(struct pcie_port *pp)
 static void ls1021_pcie_host_init(struct pcie_port *pp)
 {
 	struct ls_pcie *pcie = to_ls_pcie(pp);
-	u32 val, index[2];
+	u32 index[2];
 
 	pcie->scfg = syscon_regmap_lookup_by_phandle(pp->dev->of_node,
 						     "fsl,pcie-scfg");
@@ -116,13 +126,7 @@ static void ls1021_pcie_host_init(struct pcie_port *pp)
 
 	dw_pcie_setup_rc(pp);
 
-	/*
-	 * LS1021A Workaround for internal TKT228622
-	 * to fix the INTx hang issue
-	 */
-	val = ioread32(pcie->dbi + PCIE_STRFMR1);
-	val &= 0xffff;
-	iowrite32(val, pcie->dbi + PCIE_STRFMR1);
+	ls_pcie_drop_msg_tlp(pcie);
 }
 
 static int ls_pcie_link_up(struct pcie_port *pp)
@@ -147,6 +151,7 @@ static void ls_pcie_host_init(struct pcie_port *pp)
 	iowrite32(1, pcie->dbi + PCIE_DBI_RO_WR_EN);
 	ls_pcie_fix_class(pcie);
 	ls_pcie_clear_multifunction(pcie);
+	ls_pcie_drop_msg_tlp(pcie);
 	iowrite32(0, pcie->dbi + PCIE_DBI_RO_WR_EN);
 }
 
diff --git a/drivers/pci/host/pcie-iproc.c b/drivers/pci/host/pcie-iproc.c
index 5816bceddb65..a576aeeb22da 100644
--- a/drivers/pci/host/pcie-iproc.c
+++ b/drivers/pci/host/pcie-iproc.c
@@ -64,7 +64,6 @@
 #define OARR_SIZE_CFG                BIT(OARR_SIZE_CFG_SHIFT)
 
 #define MAX_NUM_OB_WINDOWS           2
-#define MAX_NUM_PAXC_PF              4
 
 #define IPROC_PCIE_REG_INVALID 0xffff
 
@@ -170,20 +169,6 @@ static inline void iproc_pcie_ob_write(struct iproc_pcie *pcie,
 	writel(val, pcie->base + offset + (window * 8));
 }
 
-static inline bool iproc_pcie_device_is_valid(struct iproc_pcie *pcie,
-					      unsigned int slot,
-					      unsigned int fn)
-{
-	if (slot > 0)
-		return false;
-
-	/* PAXC can only support limited number of functions */
-	if (pcie->type == IPROC_PCIE_PAXC && fn >= MAX_NUM_PAXC_PF)
-		return false;
-
-	return true;
-}
-
 /**
  * Note access to the configuration registers are protected at the higher layer
  * by 'pci_lock' in drivers/pci/access.c
@@ -199,11 +184,11 @@ static void __iomem *iproc_pcie_map_cfg_bus(struct pci_bus *bus,
 	u32 val;
 	u16 offset;
 
-	if (!iproc_pcie_device_is_valid(pcie, slot, fn))
-		return NULL;
-
 	/* root complex access */
 	if (busno == 0) {
+		if (slot > 0 || fn > 0)
+			return NULL;
+
 		iproc_pcie_write_reg(pcie, IPROC_PCIE_CFG_IND_ADDR,
 				     where & CFG_IND_ADDR_MASK);
 		offset = iproc_pcie_reg_offset(pcie, IPROC_PCIE_CFG_IND_DATA);
@@ -213,6 +198,14 @@ static void __iomem *iproc_pcie_map_cfg_bus(struct pci_bus *bus,
 			return (pcie->base + offset);
 	}
 
+	/*
+	 * PAXC is connected to an internally emulated EP within the SoC.  It
+	 * allows only one device.
+	 */
+	if (pcie->type == IPROC_PCIE_PAXC)
+		if (slot > 0)
+			return NULL;
+
 	/* EP device access */
 	val = (busno << CFG_ADDR_BUS_NUM_SHIFT) |
 		(slot << CFG_ADDR_DEV_NUM_SHIFT) |
diff --git a/drivers/pci/pci.c b/drivers/pci/pci.c
index 602eb4223510..f89db3af0607 100644
--- a/drivers/pci/pci.c
+++ b/drivers/pci/pci.c
@@ -4772,8 +4772,10 @@ int pci_get_new_domain_nr(void)
 void pci_bus_assign_domain_nr(struct pci_bus *bus, struct device *parent)
 {
 	static int use_dt_domains = -1;
-	int domain = of_get_pci_domain_nr(parent->of_node);
+	int domain = -1;
 
+	if (parent)
+		domain = of_get_pci_domain_nr(parent->of_node);
 	/*
 	 * Check DT domain and use_dt_domains values.
 	 *
diff --git a/drivers/pci/pcie/aer/aerdrv.c b/drivers/pci/pcie/aer/aerdrv.c
index 0bf82a20a0fb..48d21e0edd56 100644
--- a/drivers/pci/pcie/aer/aerdrv.c
+++ b/drivers/pci/pcie/aer/aerdrv.c
@@ -262,7 +262,6 @@ static struct aer_rpc *aer_alloc_rpc(struct pcie_device *dev)
 	rpc->rpd = dev;
 	INIT_WORK(&rpc->dpc_handler, aer_isr);
 	mutex_init(&rpc->rpc_mutex);
-	init_waitqueue_head(&rpc->wait_release);
 
 	/* Use PCIe bus function to store rpc into PCIe device */
 	set_service_data(dev, rpc);
@@ -285,8 +284,7 @@ static void aer_remove(struct pcie_device *dev)
 		if (rpc->isr)
 			free_irq(dev->irq, dev);
 
-		wait_event(rpc->wait_release, rpc->prod_idx == rpc->cons_idx);
-
+		flush_work(&rpc->dpc_handler);
 		aer_disable_rootport(rpc);
 		kfree(rpc);
 		set_service_data(dev, NULL);
diff --git a/drivers/pci/pcie/aer/aerdrv.h b/drivers/pci/pcie/aer/aerdrv.h
index 84420b7c9456..945c939a86c5 100644
--- a/drivers/pci/pcie/aer/aerdrv.h
+++ b/drivers/pci/pcie/aer/aerdrv.h
@@ -72,7 +72,6 @@ struct aer_rpc {
 					 * recovery on the same
 					 * root port hierarchy
 					 */
-	wait_queue_head_t wait_release;
 };
 
 struct aer_broadcast_data {
diff --git a/drivers/pci/pcie/aer/aerdrv_core.c b/drivers/pci/pcie/aer/aerdrv_core.c
index 712392504ed9..521e39c1b66d 100644
--- a/drivers/pci/pcie/aer/aerdrv_core.c
+++ b/drivers/pci/pcie/aer/aerdrv_core.c
@@ -811,8 +811,6 @@ void aer_isr(struct work_struct *work)
 	while (get_e_source(rpc, &e_src))
 		aer_isr_one_error(p_device, &e_src);
 	mutex_unlock(&rpc->rpc_mutex);
-
-	wake_up(&rpc->wait_release);
 }
 
 /**
diff --git a/drivers/pci/xen-pcifront.c b/drivers/pci/xen-pcifront.c
index c777b97207d5..5f70fee59a94 100644
--- a/drivers/pci/xen-pcifront.c
+++ b/drivers/pci/xen-pcifront.c
@@ -53,7 +53,7 @@ struct pcifront_device {
 };
 
 struct pcifront_sd {
-	int domain;
+	struct pci_sysdata sd;
 	struct pcifront_device *pdev;
 };
 
@@ -67,7 +67,9 @@ static inline void pcifront_init_sd(struct pcifront_sd *sd,
 				    unsigned int domain, unsigned int bus,
 				    struct pcifront_device *pdev)
 {
-	sd->domain = domain;
+	/* Because we do not expose that information via XenBus. */
+	sd->sd.node = first_online_node;
+	sd->sd.domain = domain;
 	sd->pdev = pdev;
 }
 
@@ -468,8 +470,8 @@ static int pcifront_scan_root(struct pcifront_device *pdev,
 	dev_info(&pdev->xdev->dev, "Creating PCI Frontend Bus %04x:%02x\n",
 		 domain, bus);
 
-	bus_entry = kmalloc(sizeof(*bus_entry), GFP_KERNEL);
-	sd = kmalloc(sizeof(*sd), GFP_KERNEL);
+	bus_entry = kzalloc(sizeof(*bus_entry), GFP_KERNEL);
+	sd = kzalloc(sizeof(*sd), GFP_KERNEL);
 	if (!bus_entry || !sd) {
 		err = -ENOMEM;
 		goto err_out;
diff --git a/drivers/phy/Kconfig b/drivers/phy/Kconfig
index e7e117d5dbbe..0124d17bd9fe 100644
--- a/drivers/phy/Kconfig
+++ b/drivers/phy/Kconfig
@@ -224,6 +224,7 @@ config PHY_MT65XX_USB3
 
 config PHY_HI6220_USB
 	tristate "hi6220 USB PHY support"
+	depends on (ARCH_HISI && ARM64) || COMPILE_TEST
 	select GENERIC_PHY
 	select MFD_SYSCON
 	help
diff --git a/drivers/phy/phy-core.c b/drivers/phy/phy-core.c
index 8c7f27db6ad3..e7e574dc667a 100644
--- a/drivers/phy/phy-core.c
+++ b/drivers/phy/phy-core.c
@@ -275,20 +275,21 @@ EXPORT_SYMBOL_GPL(phy_exit);
 
 int phy_power_on(struct phy *phy)
 {
-	int ret;
+	int ret = 0;
 
 	if (!phy)
-		return 0;
+		goto out;
 
 	if (phy->pwr) {
 		ret = regulator_enable(phy->pwr);
 		if (ret)
-			return ret;
+			goto out;
 	}
 
 	ret = phy_pm_runtime_get_sync(phy);
 	if (ret < 0 && ret != -ENOTSUPP)
-		return ret;
+		goto err_pm_sync;
+
 	ret = 0; /* Override possible ret == -ENOTSUPP */
 
 	mutex_lock(&phy->mutex);
@@ -296,19 +297,20 @@ int phy_power_on(struct phy *phy)
 		ret = phy->ops->power_on(phy);
 		if (ret < 0) {
 			dev_err(&phy->dev, "phy poweron failed --> %d\n", ret);
-			goto out;
+			goto err_pwr_on;
 		}
 	}
 	++phy->power_count;
 	mutex_unlock(&phy->mutex);
 	return 0;
 
-out:
+err_pwr_on:
 	mutex_unlock(&phy->mutex);
 	phy_pm_runtime_put_sync(phy);
+err_pm_sync:
 	if (phy->pwr)
 		regulator_disable(phy->pwr);
-
+out:
 	return ret;
 }
 EXPORT_SYMBOL_GPL(phy_power_on);
diff --git a/drivers/phy/phy-twl4030-usb.c b/drivers/phy/phy-twl4030-usb.c
index 4a3fc6e59f8e..840f3eae428b 100644
--- a/drivers/phy/phy-twl4030-usb.c
+++ b/drivers/phy/phy-twl4030-usb.c
@@ -715,6 +715,7 @@ static int twl4030_usb_probe(struct platform_device *pdev)
 	pm_runtime_use_autosuspend(&pdev->dev);
 	pm_runtime_set_autosuspend_delay(&pdev->dev, 2000);
 	pm_runtime_enable(&pdev->dev);
+	pm_runtime_get_sync(&pdev->dev);
 
 	/* Our job is to use irqs and status from the power module
 	 * to keep the transceiver disabled when nothing's connected.
@@ -750,6 +751,7 @@ static int twl4030_usb_remove(struct platform_device *pdev)
 	struct twl4030_usb *twl = platform_get_drvdata(pdev);
 	int val;
 
+	usb_remove_phy(&twl->phy);
 	pm_runtime_get_sync(twl->dev);
 	cancel_delayed_work(&twl->id_workaround_work);
 	device_remove_file(twl->dev, &dev_attr_vbus);
@@ -757,6 +759,13 @@ static int twl4030_usb_remove(struct platform_device *pdev)
 	/* set transceiver mode to power on defaults */
 	twl4030_usb_set_mode(twl, -1);
 
+	/* idle ulpi before powering off */
+	if (cable_present(twl->linkstat))
+		pm_runtime_put_noidle(twl->dev);
+	pm_runtime_mark_last_busy(twl->dev);
+	pm_runtime_put_sync_suspend(twl->dev);
+	pm_runtime_disable(twl->dev);
+
 	/* autogate 60MHz ULPI clock,
 	 * clear dpll clock request for i2c access,
 	 * disable 32KHz
@@ -771,11 +780,6 @@ static int twl4030_usb_remove(struct platform_device *pdev)
 	/* disable complete OTG block */
 	twl4030_usb_clear_bits(twl, POWER_CTRL, POWER_CTRL_OTG_ENAB);
 
-	if (cable_present(twl->linkstat))
-		pm_runtime_put_noidle(twl->dev);
-	pm_runtime_mark_last_busy(twl->dev);
-	pm_runtime_put(twl->dev);
-
 	return 0;
 }
 
diff --git a/drivers/pinctrl/mediatek/pinctrl-mtk-common.c b/drivers/pinctrl/mediatek/pinctrl-mtk-common.c
index 16d48a4ed225..e96e86d2e745 100644
--- a/drivers/pinctrl/mediatek/pinctrl-mtk-common.c
+++ b/drivers/pinctrl/mediatek/pinctrl-mtk-common.c
@@ -347,6 +347,7 @@ static int mtk_pconf_parse_conf(struct pinctrl_dev *pctldev,
 		ret = mtk_pconf_set_pull_select(pctl, pin, true, false, arg);
 		break;
 	case PIN_CONFIG_INPUT_ENABLE:
+		mtk_pmx_gpio_set_direction(pctldev, NULL, pin, true);
 		ret = mtk_pconf_set_ies_smt(pctl, pin, arg, param);
 		break;
 	case PIN_CONFIG_OUTPUT:
@@ -354,6 +355,7 @@ static int mtk_pconf_parse_conf(struct pinctrl_dev *pctldev,
 		ret = mtk_pmx_gpio_set_direction(pctldev, NULL, pin, false);
 		break;
 	case PIN_CONFIG_INPUT_SCHMITT_ENABLE:
+		mtk_pmx_gpio_set_direction(pctldev, NULL, pin, true);
 		ret = mtk_pconf_set_ies_smt(pctl, pin, arg, param);
 		break;
 	case PIN_CONFIG_DRIVE_STRENGTH:
diff --git a/drivers/pinctrl/mvebu/pinctrl-mvebu.c b/drivers/pinctrl/mvebu/pinctrl-mvebu.c
index e4d473811bb3..3ef798fac81b 100644
--- a/drivers/pinctrl/mvebu/pinctrl-mvebu.c
+++ b/drivers/pinctrl/mvebu/pinctrl-mvebu.c
@@ -666,16 +666,19 @@ int mvebu_pinctrl_probe(struct platform_device *pdev)
 		struct mvebu_mpp_ctrl_setting *set = &mode->settings[0];
 		struct mvebu_pinctrl_group *grp;
 		unsigned num_settings;
+		unsigned supp_settings;
 
-		for (num_settings = 0; ; set++) {
+		for (num_settings = 0, supp_settings = 0; ; set++) {
 			if (!set->name)
 				break;
 
+			num_settings++;
+
 			/* skip unsupported settings for this variant */
 			if (pctl->variant && !(pctl->variant & set->variant))
 				continue;
 
-			num_settings++;
+			supp_settings++;
 
 			/* find gpio/gpo/gpi settings */
 			if (strcmp(set->name, "gpio") == 0)
@@ -688,7 +691,7 @@ int mvebu_pinctrl_probe(struct platform_device *pdev)
 		}
 
 		/* skip modes with no settings for this variant */
-		if (!num_settings)
+		if (!supp_settings)
 			continue;
 
 		grp = mvebu_pinctrl_find_group_by_pid(pctl, mode->pid);
diff --git a/drivers/pinctrl/nomadik/pinctrl-abx500.c b/drivers/pinctrl/nomadik/pinctrl-abx500.c
index 085e60106ec2..1f7469c9857d 100644
--- a/drivers/pinctrl/nomadik/pinctrl-abx500.c
+++ b/drivers/pinctrl/nomadik/pinctrl-abx500.c
@@ -191,6 +191,7 @@ static void abx500_gpio_set(struct gpio_chip *chip, unsigned offset, int val)
 		dev_err(pct->dev, "%s write failed (%d)\n", __func__, ret);
 }
 
+#ifdef CONFIG_DEBUG_FS
 static int abx500_get_pull_updown(struct abx500_pinctrl *pct, int offset,
 				  enum abx500_gpio_pull_updown *pull_updown)
 {
@@ -226,6 +227,7 @@ out:
 
 	return ret;
 }
+#endif
 
 static int abx500_set_pull_updown(struct abx500_pinctrl *pct,
 				  int offset, enum abx500_gpio_pull_updown val)
@@ -468,6 +470,7 @@ out:
 	return ret;
 }
 
+#ifdef CONFIG_DEBUG_FS
 static int abx500_get_mode(struct pinctrl_dev *pctldev, struct gpio_chip *chip,
 			  unsigned gpio)
 {
@@ -553,8 +556,6 @@ out:
 	return ret;
 }
 
-#ifdef CONFIG_DEBUG_FS
-
 #include <linux/seq_file.h>
 
 static void abx500_gpio_dbg_show_one(struct seq_file *s,
diff --git a/drivers/pinctrl/pxa/pinctrl-pxa2xx.c b/drivers/pinctrl/pxa/pinctrl-pxa2xx.c
index d90e205cf809..216f227c6009 100644
--- a/drivers/pinctrl/pxa/pinctrl-pxa2xx.c
+++ b/drivers/pinctrl/pxa/pinctrl-pxa2xx.c
@@ -426,6 +426,7 @@ int pxa2xx_pinctrl_init(struct platform_device *pdev,
 
 	return 0;
 }
+EXPORT_SYMBOL(pxa2xx_pinctrl_init);
 
 int pxa2xx_pinctrl_exit(struct platform_device *pdev)
 {
diff --git a/drivers/pinctrl/samsung/pinctrl-samsung.c b/drivers/pinctrl/samsung/pinctrl-samsung.c
index f67b1e958589..5cc97f85db02 100644
--- a/drivers/pinctrl/samsung/pinctrl-samsung.c
+++ b/drivers/pinctrl/samsung/pinctrl-samsung.c
@@ -514,25 +514,35 @@ static const struct pinconf_ops samsung_pinconf_ops = {
 	.pin_config_group_set	= samsung_pinconf_group_set,
 };
 
-/* gpiolib gpio_set callback function */
-static void samsung_gpio_set(struct gpio_chip *gc, unsigned offset, int value)
+/*
+ * The samsung_gpio_set_vlaue() should be called with "bank->slock" held
+ * to avoid race condition.
+ */
+static void samsung_gpio_set_value(struct gpio_chip *gc,
+					  unsigned offset, int value)
 {
 	struct samsung_pin_bank *bank = gpiochip_get_data(gc);
 	const struct samsung_pin_bank_type *type = bank->type;
-	unsigned long flags;
 	void __iomem *reg;
 	u32 data;
 
 	reg = bank->drvdata->virt_base + bank->pctl_offset;
 
-	spin_lock_irqsave(&bank->slock, flags);
-
 	data = readl(reg + type->reg_offset[PINCFG_TYPE_DAT]);
 	data &= ~(1 << offset);
 	if (value)
 		data |= 1 << offset;
 	writel(data, reg + type->reg_offset[PINCFG_TYPE_DAT]);
+}
+
+/* gpiolib gpio_set callback function */
+static void samsung_gpio_set(struct gpio_chip *gc, unsigned offset, int value)
+{
+	struct samsung_pin_bank *bank = gpiochip_get_data(gc);
+	unsigned long flags;
 
+	spin_lock_irqsave(&bank->slock, flags);
+	samsung_gpio_set_value(gc, offset, value);
 	spin_unlock_irqrestore(&bank->slock, flags);
 }
 
@@ -553,6 +563,8 @@ static int samsung_gpio_get(struct gpio_chip *gc, unsigned offset)
 }
 
 /*
+ * The samsung_gpio_set_direction() should be called with "bank->slock" held
+ * to avoid race condition.
  * The calls to gpio_direction_output() and gpio_direction_input()
  * leads to this function call.
  */
@@ -564,7 +576,6 @@ static int samsung_gpio_set_direction(struct gpio_chip *gc,
 	struct samsung_pinctrl_drv_data *drvdata;
 	void __iomem *reg;
 	u32 data, mask, shift;
-	unsigned long flags;
 
 	bank = gpiochip_get_data(gc);
 	type = bank->type;
@@ -581,31 +592,42 @@ static int samsung_gpio_set_direction(struct gpio_chip *gc,
 		reg += 4;
 	}
 
-	spin_lock_irqsave(&bank->slock, flags);
-
 	data = readl(reg);
 	data &= ~(mask << shift);
 	if (!input)
 		data |= FUNC_OUTPUT << shift;
 	writel(data, reg);
 
-	spin_unlock_irqrestore(&bank->slock, flags);
-
 	return 0;
 }
 
 /* gpiolib gpio_direction_input callback function. */
 static int samsung_gpio_direction_input(struct gpio_chip *gc, unsigned offset)
 {
-	return samsung_gpio_set_direction(gc, offset, true);
+	struct samsung_pin_bank *bank = gpiochip_get_data(gc);
+	unsigned long flags;
+	int ret;
+
+	spin_lock_irqsave(&bank->slock, flags);
+	ret = samsung_gpio_set_direction(gc, offset, true);
+	spin_unlock_irqrestore(&bank->slock, flags);
+	return ret;
 }
 
 /* gpiolib gpio_direction_output callback function. */
 static int samsung_gpio_direction_output(struct gpio_chip *gc, unsigned offset,
 							int value)
 {
-	samsung_gpio_set(gc, offset, value);
-	return samsung_gpio_set_direction(gc, offset, false);
+	struct samsung_pin_bank *bank = gpiochip_get_data(gc);
+	unsigned long flags;
+	int ret;
+
+	spin_lock_irqsave(&bank->slock, flags);
+	samsung_gpio_set_value(gc, offset, value);
+	ret = samsung_gpio_set_direction(gc, offset, false);
+	spin_unlock_irqrestore(&bank->slock, flags);
+
+	return ret;
 }
 
 /*
diff --git a/drivers/pinctrl/sunxi/pinctrl-sun8i-h3.c b/drivers/pinctrl/sunxi/pinctrl-sun8i-h3.c
index 77d4cf047cee..11760bbe9d51 100644
--- a/drivers/pinctrl/sunxi/pinctrl-sun8i-h3.c
+++ b/drivers/pinctrl/sunxi/pinctrl-sun8i-h3.c
@@ -492,6 +492,7 @@ static const struct sunxi_pinctrl_desc sun8i_h3_pinctrl_data = {
 	.pins = sun8i_h3_pins,
 	.npins = ARRAY_SIZE(sun8i_h3_pins),
 	.irq_banks = 2,
+	.irq_read_needs_mux = true
 };
 
 static int sun8i_h3_pinctrl_probe(struct platform_device *pdev)
diff --git a/drivers/platform/x86/intel-hid.c b/drivers/platform/x86/intel-hid.c
index 20f0ad9bb9f3..e20f23e04c24 100644
--- a/drivers/platform/x86/intel-hid.c
+++ b/drivers/platform/x86/intel-hid.c
@@ -41,8 +41,7 @@ static const struct key_entry intel_hid_keymap[] = {
 	{ KE_KEY, 4, { KEY_HOME } },
 	{ KE_KEY, 5, { KEY_END } },
 	{ KE_KEY, 6, { KEY_PAGEUP } },
-	{ KE_KEY, 4, { KEY_PAGEDOWN } },
-	{ KE_KEY, 4, { KEY_HOME } },
+	{ KE_KEY, 7, { KEY_PAGEDOWN } },
 	{ KE_KEY, 8, { KEY_RFKILL } },
 	{ KE_KEY, 9, { KEY_POWER } },
 	{ KE_KEY, 11, { KEY_SLEEP } },
diff --git a/drivers/platform/x86/intel_scu_ipcutil.c b/drivers/platform/x86/intel_scu_ipcutil.c
index 02bc5a6343c3..aa454241489c 100644
--- a/drivers/platform/x86/intel_scu_ipcutil.c
+++ b/drivers/platform/x86/intel_scu_ipcutil.c
@@ -49,7 +49,7 @@ struct scu_ipc_data {
 
 static int scu_reg_access(u32 cmd, struct scu_ipc_data  *data)
 {
-	int count = data->count;
+	unsigned int count = data->count;
 
 	if (count == 0 || count == 3 || count > 4)
 		return -EINVAL;
diff --git a/drivers/power/bq27xxx_battery_i2c.c b/drivers/power/bq27xxx_battery_i2c.c
index 9429e66be096..8eafc6f0df88 100644
--- a/drivers/power/bq27xxx_battery_i2c.c
+++ b/drivers/power/bq27xxx_battery_i2c.c
@@ -21,6 +21,9 @@
 
 #include <linux/power/bq27xxx_battery.h>
 
+static DEFINE_IDR(battery_id);
+static DEFINE_MUTEX(battery_mutex);
+
 static irqreturn_t bq27xxx_battery_irq_handler_thread(int irq, void *data)
 {
 	struct bq27xxx_device_info *di = data;
@@ -70,19 +73,33 @@ static int bq27xxx_battery_i2c_probe(struct i2c_client *client,
 {
 	struct bq27xxx_device_info *di;
 	int ret;
+	char *name;
+	int num;
+
+	/* Get new ID for the new battery device */
+	mutex_lock(&battery_mutex);
+	num = idr_alloc(&battery_id, client, 0, 0, GFP_KERNEL);
+	mutex_unlock(&battery_mutex);
+	if (num < 0)
+		return num;
+
+	name = devm_kasprintf(&client->dev, GFP_KERNEL, "%s-%d", id->name, num);
+	if (!name)
+		goto err_mem;
 
 	di = devm_kzalloc(&client->dev, sizeof(*di), GFP_KERNEL);
 	if (!di)
-		return -ENOMEM;
+		goto err_mem;
 
+	di->id = num;
 	di->dev = &client->dev;
 	di->chip = id->driver_data;
-	di->name = id->name;
+	di->name = name;
 	di->bus.read = bq27xxx_battery_i2c_read;
 
 	ret = bq27xxx_battery_setup(di);
 	if (ret)
-		return ret;
+		goto err_failed;
 
 	/* Schedule a polling after about 1 min */
 	schedule_delayed_work(&di->work, 60 * HZ);
@@ -103,6 +120,16 @@ static int bq27xxx_battery_i2c_probe(struct i2c_client *client,
 	}
 
 	return 0;
+
+err_mem:
+	ret = -ENOMEM;
+
+err_failed:
+	mutex_lock(&battery_mutex);
+	idr_remove(&battery_id, num);
+	mutex_unlock(&battery_mutex);
+
+	return ret;
 }
 
 static int bq27xxx_battery_i2c_remove(struct i2c_client *client)
@@ -111,6 +138,10 @@ static int bq27xxx_battery_i2c_remove(struct i2c_client *client)
 
 	bq27xxx_battery_teardown(di);
 
+	mutex_lock(&battery_mutex);
+	idr_remove(&battery_id, di->id);
+	mutex_unlock(&battery_mutex);
+
 	return 0;
 }
 
diff --git a/drivers/s390/block/dasd.c b/drivers/s390/block/dasd.c
index 41605dac8309..c78db05e75b1 100644
--- a/drivers/s390/block/dasd.c
+++ b/drivers/s390/block/dasd.c
@@ -3035,6 +3035,7 @@ static void dasd_setup_queue(struct dasd_block *block)
 		max = block->base->discipline->max_blocks << block->s2b_shift;
 	}
 	queue_flag_set_unlocked(QUEUE_FLAG_NONROT, block->request_queue);
+	block->request_queue->limits.max_dev_sectors = max;
 	blk_queue_logical_block_size(block->request_queue,
 				     block->bp_block);
 	blk_queue_max_hw_sectors(block->request_queue, max);
diff --git a/drivers/s390/block/dasd_alias.c b/drivers/s390/block/dasd_alias.c
index 184b1dbeb554..286782c60da4 100644
--- a/drivers/s390/block/dasd_alias.c
+++ b/drivers/s390/block/dasd_alias.c
@@ -264,8 +264,10 @@ void dasd_alias_disconnect_device_from_lcu(struct dasd_device *device)
 		spin_unlock_irqrestore(&lcu->lock, flags);
 		cancel_work_sync(&lcu->suc_data.worker);
 		spin_lock_irqsave(&lcu->lock, flags);
-		if (device == lcu->suc_data.device)
+		if (device == lcu->suc_data.device) {
+			dasd_put_device(device);
 			lcu->suc_data.device = NULL;
+		}
 	}
 	was_pending = 0;
 	if (device == lcu->ruac_data.device) {
@@ -273,8 +275,10 @@ void dasd_alias_disconnect_device_from_lcu(struct dasd_device *device)
 		was_pending = 1;
 		cancel_delayed_work_sync(&lcu->ruac_data.dwork);
 		spin_lock_irqsave(&lcu->lock, flags);
-		if (device == lcu->ruac_data.device)
+		if (device == lcu->ruac_data.device) {
+			dasd_put_device(device);
 			lcu->ruac_data.device = NULL;
+		}
 	}
 	private->lcu = NULL;
 	spin_unlock_irqrestore(&lcu->lock, flags);
@@ -549,8 +553,10 @@ static void lcu_update_work(struct work_struct *work)
 	if ((rc && (rc != -EOPNOTSUPP)) || (lcu->flags & NEED_UAC_UPDATE)) {
 		DBF_DEV_EVENT(DBF_WARNING, device, "could not update"
 			    " alias data in lcu (rc = %d), retry later", rc);
-		schedule_delayed_work(&lcu->ruac_data.dwork, 30*HZ);
+		if (!schedule_delayed_work(&lcu->ruac_data.dwork, 30*HZ))
+			dasd_put_device(device);
 	} else {
+		dasd_put_device(device);
 		lcu->ruac_data.device = NULL;
 		lcu->flags &= ~UPDATE_PENDING;
 	}
@@ -593,8 +599,10 @@ static int _schedule_lcu_update(struct alias_lcu *lcu,
 	 */
 	if (!usedev)
 		return -EINVAL;
+	dasd_get_device(usedev);
 	lcu->ruac_data.device = usedev;
-	schedule_delayed_work(&lcu->ruac_data.dwork, 0);
+	if (!schedule_delayed_work(&lcu->ruac_data.dwork, 0))
+		dasd_put_device(usedev);
 	return 0;
 }
 
@@ -723,7 +731,7 @@ static int reset_summary_unit_check(struct alias_lcu *lcu,
 	ASCEBC((char *) &cqr->magic, 4);
 	ccw = cqr->cpaddr;
 	ccw->cmd_code = DASD_ECKD_CCW_RSCK;
-	ccw->flags = 0 ;
+	ccw->flags = CCW_FLAG_SLI;
 	ccw->count = 16;
 	ccw->cda = (__u32)(addr_t) cqr->data;
 	((char *)cqr->data)[0] = reason;
@@ -930,6 +938,7 @@ static void summary_unit_check_handling_work(struct work_struct *work)
 	/* 3. read new alias configuration */
 	_schedule_lcu_update(lcu, device);
 	lcu->suc_data.device = NULL;
+	dasd_put_device(device);
 	spin_unlock_irqrestore(&lcu->lock, flags);
 }
 
@@ -989,6 +998,8 @@ void dasd_alias_handle_summary_unit_check(struct dasd_device *device,
 	}
 	lcu->suc_data.reason = reason;
 	lcu->suc_data.device = device;
+	dasd_get_device(device);
 	spin_unlock(&lcu->lock);
-	schedule_work(&lcu->suc_data.worker);
+	if (!schedule_work(&lcu->suc_data.worker))
+		dasd_put_device(device);
 };
diff --git a/drivers/s390/block/dasd_diag.c b/drivers/s390/block/dasd_diag.c
index cb61f300f8b5..277b5c8c825c 100644
--- a/drivers/s390/block/dasd_diag.c
+++ b/drivers/s390/block/dasd_diag.c
@@ -67,7 +67,7 @@ static const u8 DASD_DIAG_CMS1[] = { 0xc3, 0xd4, 0xe2, 0xf1 };/* EBCDIC CMS1 */
  * and function code cmd.
  * In case of an exception return 3. Otherwise return result of bitwise OR of
  * resulting condition code and DIAG return code. */
-static inline int dia250(void *iob, int cmd)
+static inline int __dia250(void *iob, int cmd)
 {
 	register unsigned long reg2 asm ("2") = (unsigned long) iob;
 	typedef union {
@@ -77,7 +77,6 @@ static inline int dia250(void *iob, int cmd)
 	int rc;
 
 	rc = 3;
-	diag_stat_inc(DIAG_STAT_X250);
 	asm volatile(
 		"	diag	2,%2,0x250\n"
 		"0:	ipm	%0\n"
@@ -91,6 +90,12 @@ static inline int dia250(void *iob, int cmd)
 	return rc;
 }
 
+static inline int dia250(void *iob, int cmd)
+{
+	diag_stat_inc(DIAG_STAT_X250);
+	return __dia250(iob, cmd);
+}
+
 /* Initialize block I/O to DIAG device using the specified blocksize and
  * block offset. On success, return zero and set end_block to contain the
  * number of blocks on the device minus the specified offset. Return non-zero
diff --git a/drivers/scsi/device_handler/scsi_dh_rdac.c b/drivers/scsi/device_handler/scsi_dh_rdac.c
index 361358134315..93880ed6291c 100644
--- a/drivers/scsi/device_handler/scsi_dh_rdac.c
+++ b/drivers/scsi/device_handler/scsi_dh_rdac.c
@@ -562,7 +562,7 @@ static int mode_select_handle_sense(struct scsi_device *sdev,
 			/*
 			 * Command Lock contention
 			 */
-			err = SCSI_DH_RETRY;
+			err = SCSI_DH_IMM_RETRY;
 		break;
 	default:
 		break;
@@ -612,6 +612,8 @@ retry:
 		err = mode_select_handle_sense(sdev, h->sense);
 		if (err == SCSI_DH_RETRY && retry_cnt--)
 			goto retry;
+		if (err == SCSI_DH_IMM_RETRY)
+			goto retry;
 	}
 	if (err == SCSI_DH_OK) {
 		h->state = RDAC_STATE_ACTIVE;
diff --git a/drivers/scsi/hisi_sas/Kconfig b/drivers/scsi/hisi_sas/Kconfig
index b67661836c9f..d1dd1616f983 100644
--- a/drivers/scsi/hisi_sas/Kconfig
+++ b/drivers/scsi/hisi_sas/Kconfig
@@ -1,6 +1,6 @@
 config SCSI_HISI_SAS
 	tristate "HiSilicon SAS"
-	depends on HAS_DMA
+	depends on HAS_DMA && HAS_IOMEM
 	depends on ARM64 || COMPILE_TEST
 	select SCSI_SAS_LIBSAS
 	select BLK_DEV_INTEGRITY
diff --git a/drivers/scsi/hisi_sas/hisi_sas_v1_hw.c b/drivers/scsi/hisi_sas/hisi_sas_v1_hw.c
index 057fdeb720ac..eea24d7531cf 100644
--- a/drivers/scsi/hisi_sas/hisi_sas_v1_hw.c
+++ b/drivers/scsi/hisi_sas/hisi_sas_v1_hw.c
@@ -1289,13 +1289,10 @@ static int slot_complete_v1_hw(struct hisi_hba *hisi_hba,
 		goto out;
 	}
 
-	if (cmplt_hdr_data & CMPLT_HDR_ERR_RCRD_XFRD_MSK) {
-		if (!(cmplt_hdr_data & CMPLT_HDR_CMD_CMPLT_MSK) ||
-		    !(cmplt_hdr_data & CMPLT_HDR_RSPNS_XFRD_MSK))
-			ts->stat = SAS_DATA_OVERRUN;
-		else
-			slot_err_v1_hw(hisi_hba, task, slot);
+	if (cmplt_hdr_data & CMPLT_HDR_ERR_RCRD_XFRD_MSK &&
+		!(cmplt_hdr_data & CMPLT_HDR_RSPNS_XFRD_MSK)) {
 
+		slot_err_v1_hw(hisi_hba, task, slot);
 		goto out;
 	}
 
diff --git a/drivers/scsi/ipr.c b/drivers/scsi/ipr.c
index 3b3e0998fa6e..d6a691e27d33 100644
--- a/drivers/scsi/ipr.c
+++ b/drivers/scsi/ipr.c
@@ -4002,6 +4002,7 @@ static ssize_t ipr_store_update_fw(struct device *dev,
 	struct ipr_sglist *sglist;
 	char fname[100];
 	char *src;
+	char *endline;
 	int result, dnld_size;
 
 	if (!capable(CAP_SYS_ADMIN))
@@ -4009,6 +4010,10 @@ static ssize_t ipr_store_update_fw(struct device *dev,
 
 	snprintf(fname, sizeof(fname), "%s", buf);
 
+	endline = strchr(fname, '\n');
+	if (endline)
+		*endline = '\0';
+
 	if (request_firmware(&fw_entry, fname, &ioa_cfg->pdev->dev)) {
 		dev_err(&ioa_cfg->pdev->dev, "Firmware file %s not found\n", fname);
 		return -EIO;
diff --git a/drivers/scsi/qla2xxx/qla_init.c b/drivers/scsi/qla2xxx/qla_init.c
index 52a87657c7dd..692a7570b5e1 100644
--- a/drivers/scsi/qla2xxx/qla_init.c
+++ b/drivers/scsi/qla2xxx/qla_init.c
@@ -2204,7 +2204,7 @@ qla2x00_init_rings(scsi_qla_host_t *vha)
 	/* Clear outstanding commands array. */
 	for (que = 0; que < ha->max_req_queues; que++) {
 		req = ha->req_q_map[que];
-		if (!req)
+		if (!req || !test_bit(que, ha->req_qid_map))
 			continue;
 		req->out_ptr = (void *)(req->ring + req->length);
 		*req->out_ptr = 0;
@@ -2221,7 +2221,7 @@ qla2x00_init_rings(scsi_qla_host_t *vha)
 
 	for (que = 0; que < ha->max_rsp_queues; que++) {
 		rsp = ha->rsp_q_map[que];
-		if (!rsp)
+		if (!rsp || !test_bit(que, ha->rsp_qid_map))
 			continue;
 		rsp->in_ptr = (void *)(rsp->ring + rsp->length);
 		*rsp->in_ptr = 0;
@@ -4981,7 +4981,7 @@ qla25xx_init_queues(struct qla_hw_data *ha)
 
 	for (i = 1; i < ha->max_rsp_queues; i++) {
 		rsp = ha->rsp_q_map[i];
-		if (rsp) {
+		if (rsp && test_bit(i, ha->rsp_qid_map)) {
 			rsp->options &= ~BIT_0;
 			ret = qla25xx_init_rsp_que(base_vha, rsp);
 			if (ret != QLA_SUCCESS)
@@ -4996,8 +4996,8 @@ qla25xx_init_queues(struct qla_hw_data *ha)
 	}
 	for (i = 1; i < ha->max_req_queues; i++) {
 		req = ha->req_q_map[i];
-		if (req) {
-		/* Clear outstanding commands array. */
+		if (req && test_bit(i, ha->req_qid_map)) {
+			/* Clear outstanding commands array. */
 			req->options &= ~BIT_0;
 			ret = qla25xx_init_req_que(base_vha, req);
 			if (ret != QLA_SUCCESS)
diff --git a/drivers/scsi/qla2xxx/qla_isr.c b/drivers/scsi/qla2xxx/qla_isr.c
index d4d65eb0e9b4..4af95479a9db 100644
--- a/drivers/scsi/qla2xxx/qla_isr.c
+++ b/drivers/scsi/qla2xxx/qla_isr.c
@@ -3063,9 +3063,9 @@ qla24xx_enable_msix(struct qla_hw_data *ha, struct rsp_que *rsp)
 		    "MSI-X: Failed to enable support "
 		    "-- %d/%d\n Retry with %d vectors.\n",
 		    ha->msix_count, ret, ret);
+		ha->msix_count = ret;
+		ha->max_rsp_queues = ha->msix_count - 1;
 	}
-	ha->msix_count = ret;
-	ha->max_rsp_queues = ha->msix_count - 1;
 	ha->msix_entries = kzalloc(sizeof(struct qla_msix_entry) *
 				ha->msix_count, GFP_KERNEL);
 	if (!ha->msix_entries) {
diff --git a/drivers/scsi/qla2xxx/qla_mid.c b/drivers/scsi/qla2xxx/qla_mid.c
index c5dd594f6c31..cf7ba52bae66 100644
--- a/drivers/scsi/qla2xxx/qla_mid.c
+++ b/drivers/scsi/qla2xxx/qla_mid.c
@@ -600,7 +600,7 @@ qla25xx_delete_queues(struct scsi_qla_host *vha)
 	/* Delete request queues */
 	for (cnt = 1; cnt < ha->max_req_queues; cnt++) {
 		req = ha->req_q_map[cnt];
-		if (req) {
+		if (req && test_bit(cnt, ha->req_qid_map)) {
 			ret = qla25xx_delete_req_que(vha, req);
 			if (ret != QLA_SUCCESS) {
 				ql_log(ql_log_warn, vha, 0x00ea,
@@ -614,7 +614,7 @@ qla25xx_delete_queues(struct scsi_qla_host *vha)
 	/* Delete response queues */
 	for (cnt = 1; cnt < ha->max_rsp_queues; cnt++) {
 		rsp = ha->rsp_q_map[cnt];
-		if (rsp) {
+		if (rsp && test_bit(cnt, ha->rsp_qid_map)) {
 			ret = qla25xx_delete_rsp_que(vha, rsp);
 			if (ret != QLA_SUCCESS) {
 				ql_log(ql_log_warn, vha, 0x00eb,
diff --git a/drivers/scsi/qla2xxx/qla_os.c b/drivers/scsi/qla2xxx/qla_os.c
index f1788db43195..f6c7ce35b542 100644
--- a/drivers/scsi/qla2xxx/qla_os.c
+++ b/drivers/scsi/qla2xxx/qla_os.c
@@ -409,6 +409,9 @@ static void qla2x00_free_queues(struct qla_hw_data *ha)
 	int cnt;
 
 	for (cnt = 0; cnt < ha->max_req_queues; cnt++) {
+		if (!test_bit(cnt, ha->req_qid_map))
+			continue;
+
 		req = ha->req_q_map[cnt];
 		qla2x00_free_req_que(ha, req);
 	}
@@ -416,6 +419,9 @@ static void qla2x00_free_queues(struct qla_hw_data *ha)
 	ha->req_q_map = NULL;
 
 	for (cnt = 0; cnt < ha->max_rsp_queues; cnt++) {
+		if (!test_bit(cnt, ha->rsp_qid_map))
+			continue;
+
 		rsp = ha->rsp_q_map[cnt];
 		qla2x00_free_rsp_que(ha, rsp);
 	}
diff --git a/drivers/scsi/qla2xxx/qla_target.c b/drivers/scsi/qla2xxx/qla_target.c
index 8075a4cdb45c..ee967becd257 100644
--- a/drivers/scsi/qla2xxx/qla_target.c
+++ b/drivers/scsi/qla2xxx/qla_target.c
@@ -105,7 +105,7 @@ static void qlt_response_pkt(struct scsi_qla_host *ha, response_t *pkt);
 static int qlt_issue_task_mgmt(struct qla_tgt_sess *sess, uint32_t lun,
 	int fn, void *iocb, int flags);
 static void qlt_send_term_exchange(struct scsi_qla_host *ha, struct qla_tgt_cmd
-	*cmd, struct atio_from_isp *atio, int ha_locked);
+	*cmd, struct atio_from_isp *atio, int ha_locked, int ul_abort);
 static void qlt_reject_free_srr_imm(struct scsi_qla_host *ha,
 	struct qla_tgt_srr_imm *imm, int ha_lock);
 static void qlt_abort_cmd_on_host_reset(struct scsi_qla_host *vha,
@@ -1756,7 +1756,7 @@ void qlt_xmit_tm_rsp(struct qla_tgt_mgmt_cmd *mcmd)
 		qlt_send_notify_ack(vha, &mcmd->orig_iocb.imm_ntfy,
 		    0, 0, 0, 0, 0, 0);
 	else {
-		if (mcmd->se_cmd.se_tmr_req->function == TMR_ABORT_TASK)
+		if (mcmd->orig_iocb.atio.u.raw.entry_type == ABTS_RECV_24XX)
 			qlt_24xx_send_abts_resp(vha, &mcmd->orig_iocb.abts,
 			    mcmd->fc_tm_rsp, false);
 		else
@@ -2665,7 +2665,7 @@ int qlt_xmit_response(struct qla_tgt_cmd *cmd, int xmit_type,
 			/* no need to terminate. FW already freed exchange. */
 			qlt_abort_cmd_on_host_reset(cmd->vha, cmd);
 		else
-			qlt_send_term_exchange(vha, cmd, &cmd->atio, 1);
+			qlt_send_term_exchange(vha, cmd, &cmd->atio, 1, 0);
 		spin_unlock_irqrestore(&ha->hardware_lock, flags);
 		return 0;
 	}
@@ -3173,7 +3173,8 @@ static int __qlt_send_term_exchange(struct scsi_qla_host *vha,
 }
 
 static void qlt_send_term_exchange(struct scsi_qla_host *vha,
-	struct qla_tgt_cmd *cmd, struct atio_from_isp *atio, int ha_locked)
+	struct qla_tgt_cmd *cmd, struct atio_from_isp *atio, int ha_locked,
+	int ul_abort)
 {
 	unsigned long flags = 0;
 	int rc;
@@ -3193,8 +3194,7 @@ static void qlt_send_term_exchange(struct scsi_qla_host *vha,
 		qlt_alloc_qfull_cmd(vha, atio, 0, 0);
 
 done:
-	if (cmd && (!cmd->aborted ||
-	    !cmd->cmd_sent_to_fw)) {
+	if (cmd && !ul_abort && !cmd->aborted) {
 		if (cmd->sg_mapped)
 			qlt_unmap_sg(vha, cmd);
 		vha->hw->tgt.tgt_ops->free_cmd(cmd);
@@ -3253,21 +3253,38 @@ static void qlt_chk_exch_leak_thresh_hold(struct scsi_qla_host *vha)
 
 }
 
-void qlt_abort_cmd(struct qla_tgt_cmd *cmd)
+int qlt_abort_cmd(struct qla_tgt_cmd *cmd)
 {
 	struct qla_tgt *tgt = cmd->tgt;
 	struct scsi_qla_host *vha = tgt->vha;
 	struct se_cmd *se_cmd = &cmd->se_cmd;
+	unsigned long flags;
 
 	ql_dbg(ql_dbg_tgt_mgt, vha, 0xf014,
 	    "qla_target(%d): terminating exchange for aborted cmd=%p "
 	    "(se_cmd=%p, tag=%llu)", vha->vp_idx, cmd, &cmd->se_cmd,
 	    se_cmd->tag);
 
+	spin_lock_irqsave(&cmd->cmd_lock, flags);
+	if (cmd->aborted) {
+		spin_unlock_irqrestore(&cmd->cmd_lock, flags);
+		/*
+		 * It's normal to see 2 calls in this path:
+		 *  1) XFER Rdy completion + CMD_T_ABORT
+		 *  2) TCM TMR - drain_state_list
+		 */
+	        ql_dbg(ql_dbg_tgt_mgt, vha, 0xffff,
+			"multiple abort. %p transport_state %x, t_state %x,"
+			" se_cmd_flags %x \n", cmd, cmd->se_cmd.transport_state,
+			cmd->se_cmd.t_state,cmd->se_cmd.se_cmd_flags);
+		return EIO;
+	}
 	cmd->aborted = 1;
 	cmd->cmd_flags |= BIT_6;
+	spin_unlock_irqrestore(&cmd->cmd_lock, flags);
 
-	qlt_send_term_exchange(vha, cmd, &cmd->atio, 0);
+	qlt_send_term_exchange(vha, cmd, &cmd->atio, 0, 1);
+	return 0;
 }
 EXPORT_SYMBOL(qlt_abort_cmd);
 
@@ -3282,6 +3299,9 @@ void qlt_free_cmd(struct qla_tgt_cmd *cmd)
 
 	BUG_ON(cmd->cmd_in_wq);
 
+	if (cmd->sg_mapped)
+		qlt_unmap_sg(cmd->vha, cmd);
+
 	if (!cmd->q_full)
 		qlt_decr_num_pend_cmds(cmd->vha);
 
@@ -3399,7 +3419,7 @@ static int qlt_term_ctio_exchange(struct scsi_qla_host *vha, void *ctio,
 		term = 1;
 
 	if (term)
-		qlt_send_term_exchange(vha, cmd, &cmd->atio, 1);
+		qlt_send_term_exchange(vha, cmd, &cmd->atio, 1, 0);
 
 	return term;
 }
@@ -3580,12 +3600,13 @@ static void qlt_do_ctio_completion(struct scsi_qla_host *vha, uint32_t handle,
 		case CTIO_PORT_LOGGED_OUT:
 		case CTIO_PORT_UNAVAILABLE:
 		{
-			int logged_out = (status & 0xFFFF);
+			int logged_out =
+				(status & 0xFFFF) == CTIO_PORT_LOGGED_OUT;
+
 			ql_dbg(ql_dbg_tgt_mgt, vha, 0xf059,
 			    "qla_target(%d): CTIO with %s status %x "
 			    "received (state %x, se_cmd %p)\n", vha->vp_idx,
-			    (logged_out == CTIO_PORT_LOGGED_OUT) ?
-			    "PORT LOGGED OUT" : "PORT UNAVAILABLE",
+			    logged_out ? "PORT LOGGED OUT" : "PORT UNAVAILABLE",
 			    status, cmd->state, se_cmd);
 
 			if (logged_out && cmd->sess) {
@@ -3754,6 +3775,7 @@ static void __qlt_do_work(struct qla_tgt_cmd *cmd)
 		goto out_term;
 	}
 
+	spin_lock_init(&cmd->cmd_lock);
 	cdb = &atio->u.isp24.fcp_cmnd.cdb[0];
 	cmd->se_cmd.tag = atio->u.isp24.exchange_addr;
 	cmd->unpacked_lun = scsilun_to_int(
@@ -3796,7 +3818,7 @@ out_term:
 	 */
 	cmd->cmd_flags |= BIT_2;
 	spin_lock_irqsave(&ha->hardware_lock, flags);
-	qlt_send_term_exchange(vha, NULL, &cmd->atio, 1);
+	qlt_send_term_exchange(vha, NULL, &cmd->atio, 1, 0);
 
 	qlt_decr_num_pend_cmds(vha);
 	percpu_ida_free(&sess->se_sess->sess_tag_pool, cmd->se_cmd.map_tag);
@@ -3918,7 +3940,7 @@ static void qlt_create_sess_from_atio(struct work_struct *work)
 
 out_term:
 	spin_lock_irqsave(&ha->hardware_lock, flags);
-	qlt_send_term_exchange(vha, NULL, &op->atio, 1);
+	qlt_send_term_exchange(vha, NULL, &op->atio, 1, 0);
 	spin_unlock_irqrestore(&ha->hardware_lock, flags);
 	kfree(op);
 
@@ -3982,7 +4004,8 @@ static int qlt_handle_cmd_for_atio(struct scsi_qla_host *vha,
 
 	cmd->cmd_in_wq = 1;
 	cmd->cmd_flags |= BIT_0;
-	cmd->se_cmd.cpuid = -1;
+	cmd->se_cmd.cpuid = ha->msix_count ?
+		ha->tgt.rspq_vector_cpuid : WORK_CPU_UNBOUND;
 
 	spin_lock(&vha->cmd_list_lock);
 	list_add_tail(&cmd->cmd_list, &vha->qla_cmd_list);
@@ -3990,7 +4013,6 @@ static int qlt_handle_cmd_for_atio(struct scsi_qla_host *vha,
 
 	INIT_WORK(&cmd->work, qlt_do_work);
 	if (ha->msix_count) {
-		cmd->se_cmd.cpuid = ha->tgt.rspq_vector_cpuid;
 		if (cmd->atio.u.isp24.fcp_cmnd.rddata)
 			queue_work_on(smp_processor_id(), qla_tgt_wq,
 			    &cmd->work);
@@ -4771,7 +4793,7 @@ out_reject:
 		dump_stack();
 	} else {
 		cmd->cmd_flags |= BIT_9;
-		qlt_send_term_exchange(vha, cmd, &cmd->atio, 1);
+		qlt_send_term_exchange(vha, cmd, &cmd->atio, 1, 0);
 	}
 	spin_unlock_irqrestore(&ha->hardware_lock, flags);
 }
@@ -4950,7 +4972,7 @@ static void qlt_prepare_srr_imm(struct scsi_qla_host *vha,
 				    sctio, sctio->srr_id);
 				list_del(&sctio->srr_list_entry);
 				qlt_send_term_exchange(vha, sctio->cmd,
-				    &sctio->cmd->atio, 1);
+				    &sctio->cmd->atio, 1, 0);
 				kfree(sctio);
 			}
 		}
@@ -5123,7 +5145,7 @@ static int __qlt_send_busy(struct scsi_qla_host *vha,
 	    atio->u.isp24.fcp_hdr.s_id);
 	spin_unlock_irqrestore(&ha->tgt.sess_lock, flags);
 	if (!sess) {
-		qlt_send_term_exchange(vha, NULL, atio, 1);
+		qlt_send_term_exchange(vha, NULL, atio, 1, 0);
 		return 0;
 	}
 	/* Sending marker isn't necessary, since we called from ISR */
@@ -5406,7 +5428,7 @@ static void qlt_24xx_atio_pkt(struct scsi_qla_host *vha,
 #if 1 /* With TERM EXCHANGE some FC cards refuse to boot */
 				qlt_send_busy(vha, atio, SAM_STAT_BUSY);
 #else
-				qlt_send_term_exchange(vha, NULL, atio, 1);
+				qlt_send_term_exchange(vha, NULL, atio, 1, 0);
 #endif
 
 				if (!ha_locked)
@@ -5523,7 +5545,7 @@ static void qlt_response_pkt(struct scsi_qla_host *vha, response_t *pkt)
 #if 1 /* With TERM EXCHANGE some FC cards refuse to boot */
 				qlt_send_busy(vha, atio, 0);
 #else
-				qlt_send_term_exchange(vha, NULL, atio, 1);
+				qlt_send_term_exchange(vha, NULL, atio, 1, 0);
 #endif
 			} else {
 				if (tgt->tgt_stop) {
@@ -5532,7 +5554,7 @@ static void qlt_response_pkt(struct scsi_qla_host *vha, response_t *pkt)
 					    "command to target, sending TERM "
 					    "EXCHANGE for rsp\n");
 					qlt_send_term_exchange(vha, NULL,
-					    atio, 1);
+					    atio, 1, 0);
 				} else {
 					ql_dbg(ql_dbg_tgt, vha, 0xe060,
 					    "qla_target(%d): Unable to send "
@@ -5960,7 +5982,7 @@ static void qlt_tmr_work(struct qla_tgt *tgt,
 	return;
 
 out_term:
-	qlt_send_term_exchange(vha, NULL, &prm->tm_iocb2, 0);
+	qlt_send_term_exchange(vha, NULL, &prm->tm_iocb2, 1, 0);
 	if (sess)
 		ha->tgt.tgt_ops->put_sess(sess);
 	spin_unlock_irqrestore(&ha->tgt.sess_lock, flags);
diff --git a/drivers/scsi/qla2xxx/qla_target.h b/drivers/scsi/qla2xxx/qla_target.h
index 71b2865ba3c8..22a6a767fe07 100644
--- a/drivers/scsi/qla2xxx/qla_target.h
+++ b/drivers/scsi/qla2xxx/qla_target.h
@@ -943,6 +943,36 @@ struct qla_tgt_sess {
 	qlt_plogi_ack_t *plogi_link[QLT_PLOGI_LINK_MAX];
 };
 
+typedef enum {
+	/*
+	 * BIT_0 - Atio Arrival / schedule to work
+	 * BIT_1 - qlt_do_work
+	 * BIT_2 - qlt_do work failed
+	 * BIT_3 - xfer rdy/tcm_qla2xxx_write_pending
+	 * BIT_4 - read respond/tcm_qla2xx_queue_data_in
+	 * BIT_5 - status respond / tcm_qla2xx_queue_status
+	 * BIT_6 - tcm request to abort/Term exchange.
+	 *	pre_xmit_response->qlt_send_term_exchange
+	 * BIT_7 - SRR received (qlt_handle_srr->qlt_xmit_response)
+	 * BIT_8 - SRR received (qlt_handle_srr->qlt_rdy_to_xfer)
+	 * BIT_9 - SRR received (qla_handle_srr->qlt_send_term_exchange)
+	 * BIT_10 - Data in - hanlde_data->tcm_qla2xxx_handle_data
+
+	 * BIT_12 - good completion - qlt_ctio_do_completion -->free_cmd
+	 * BIT_13 - Bad completion -
+	 *	qlt_ctio_do_completion --> qlt_term_ctio_exchange
+	 * BIT_14 - Back end data received/sent.
+	 * BIT_15 - SRR prepare ctio
+	 * BIT_16 - complete free
+	 * BIT_17 - flush - qlt_abort_cmd_on_host_reset
+	 * BIT_18 - completion w/abort status
+	 * BIT_19 - completion w/unknown status
+	 * BIT_20 - tcm_qla2xxx_free_cmd
+	 */
+	CMD_FLAG_DATA_WORK = BIT_11,
+	CMD_FLAG_DATA_WORK_FREE = BIT_21,
+} cmd_flags_t;
+
 struct qla_tgt_cmd {
 	struct se_cmd se_cmd;
 	struct qla_tgt_sess *sess;
@@ -952,6 +982,7 @@ struct qla_tgt_cmd {
 	/* Sense buffer that will be mapped into outgoing status */
 	unsigned char sense_buffer[TRANSPORT_SENSE_BUFFER];
 
+	spinlock_t cmd_lock;
 	/* to save extra sess dereferences */
 	unsigned int conf_compl_supported:1;
 	unsigned int sg_mapped:1;
@@ -986,30 +1017,8 @@ struct qla_tgt_cmd {
 
 	uint64_t jiffies_at_alloc;
 	uint64_t jiffies_at_free;
-	/* BIT_0 - Atio Arrival / schedule to work
-	 * BIT_1 - qlt_do_work
-	 * BIT_2 - qlt_do work failed
-	 * BIT_3 - xfer rdy/tcm_qla2xxx_write_pending
-	 * BIT_4 - read respond/tcm_qla2xx_queue_data_in
-	 * BIT_5 - status respond / tcm_qla2xx_queue_status
-	 * BIT_6 - tcm request to abort/Term exchange.
-	 *	pre_xmit_response->qlt_send_term_exchange
-	 * BIT_7 - SRR received (qlt_handle_srr->qlt_xmit_response)
-	 * BIT_8 - SRR received (qlt_handle_srr->qlt_rdy_to_xfer)
-	 * BIT_9 - SRR received (qla_handle_srr->qlt_send_term_exchange)
-	 * BIT_10 - Data in - hanlde_data->tcm_qla2xxx_handle_data
-	 * BIT_11 - Data actually going to TCM : tcm_qla2xx_handle_data_work
-	 * BIT_12 - good completion - qlt_ctio_do_completion -->free_cmd
-	 * BIT_13 - Bad completion -
-	 *	qlt_ctio_do_completion --> qlt_term_ctio_exchange
-	 * BIT_14 - Back end data received/sent.
-	 * BIT_15 - SRR prepare ctio
-	 * BIT_16 - complete free
-	 * BIT_17 - flush - qlt_abort_cmd_on_host_reset
-	 * BIT_18 - completion w/abort status
-	 * BIT_19 - completion w/unknown status
-	 */
-	uint32_t cmd_flags;
+
+	cmd_flags_t cmd_flags;
 };
 
 struct qla_tgt_sess_work_param {
@@ -1148,7 +1157,7 @@ static inline void sid_to_portid(const uint8_t *s_id, port_id_t *p)
 extern void qlt_response_pkt_all_vps(struct scsi_qla_host *, response_t *);
 extern int qlt_rdy_to_xfer(struct qla_tgt_cmd *);
 extern int qlt_xmit_response(struct qla_tgt_cmd *, int, uint8_t);
-extern void qlt_abort_cmd(struct qla_tgt_cmd *);
+extern int qlt_abort_cmd(struct qla_tgt_cmd *);
 extern void qlt_xmit_tm_rsp(struct qla_tgt_mgmt_cmd *);
 extern void qlt_free_mcmd(struct qla_tgt_mgmt_cmd *);
 extern void qlt_free_cmd(struct qla_tgt_cmd *cmd);
diff --git a/drivers/scsi/qla2xxx/qla_tmpl.c b/drivers/scsi/qla2xxx/qla_tmpl.c
index ddbe2e7ac14d..c3e622524604 100644
--- a/drivers/scsi/qla2xxx/qla_tmpl.c
+++ b/drivers/scsi/qla2xxx/qla_tmpl.c
@@ -395,6 +395,10 @@ qla27xx_fwdt_entry_t263(struct scsi_qla_host *vha,
 	if (ent->t263.queue_type == T263_QUEUE_TYPE_REQ) {
 		for (i = 0; i < vha->hw->max_req_queues; i++) {
 			struct req_que *req = vha->hw->req_q_map[i];
+
+			if (!test_bit(i, vha->hw->req_qid_map))
+				continue;
+
 			if (req || !buf) {
 				length = req ?
 				    req->length : REQUEST_ENTRY_CNT_24XX;
@@ -408,6 +412,10 @@ qla27xx_fwdt_entry_t263(struct scsi_qla_host *vha,
 	} else if (ent->t263.queue_type == T263_QUEUE_TYPE_RSP) {
 		for (i = 0; i < vha->hw->max_rsp_queues; i++) {
 			struct rsp_que *rsp = vha->hw->rsp_q_map[i];
+
+			if (!test_bit(i, vha->hw->rsp_qid_map))
+				continue;
+
 			if (rsp || !buf) {
 				length = rsp ?
 				    rsp->length : RESPONSE_ENTRY_CNT_MQ;
@@ -634,6 +642,10 @@ qla27xx_fwdt_entry_t274(struct scsi_qla_host *vha,
 	if (ent->t274.queue_type == T274_QUEUE_TYPE_REQ_SHAD) {
 		for (i = 0; i < vha->hw->max_req_queues; i++) {
 			struct req_que *req = vha->hw->req_q_map[i];
+
+			if (!test_bit(i, vha->hw->req_qid_map))
+				continue;
+
 			if (req || !buf) {
 				qla27xx_insert16(i, buf, len);
 				qla27xx_insert16(1, buf, len);
@@ -645,6 +657,10 @@ qla27xx_fwdt_entry_t274(struct scsi_qla_host *vha,
 	} else if (ent->t274.queue_type == T274_QUEUE_TYPE_RSP_SHAD) {
 		for (i = 0; i < vha->hw->max_rsp_queues; i++) {
 			struct rsp_que *rsp = vha->hw->rsp_q_map[i];
+
+			if (!test_bit(i, vha->hw->rsp_qid_map))
+				continue;
+
 			if (rsp || !buf) {
 				qla27xx_insert16(i, buf, len);
 				qla27xx_insert16(1, buf, len);
diff --git a/drivers/scsi/qla2xxx/tcm_qla2xxx.c b/drivers/scsi/qla2xxx/tcm_qla2xxx.c
index faf0a126627f..1808a01cfb7e 100644
--- a/drivers/scsi/qla2xxx/tcm_qla2xxx.c
+++ b/drivers/scsi/qla2xxx/tcm_qla2xxx.c
@@ -298,6 +298,10 @@ static void tcm_qla2xxx_free_cmd(struct qla_tgt_cmd *cmd)
 {
 	cmd->vha->tgt_counters.core_qla_free_cmd++;
 	cmd->cmd_in_wq = 1;
+
+	BUG_ON(cmd->cmd_flags & BIT_20);
+	cmd->cmd_flags |= BIT_20;
+
 	INIT_WORK(&cmd->work, tcm_qla2xxx_complete_free);
 	queue_work_on(smp_processor_id(), tcm_qla2xxx_free_wq, &cmd->work);
 }
@@ -374,6 +378,20 @@ static int tcm_qla2xxx_write_pending(struct se_cmd *se_cmd)
 {
 	struct qla_tgt_cmd *cmd = container_of(se_cmd,
 				struct qla_tgt_cmd, se_cmd);
+
+	if (cmd->aborted) {
+		/* Cmd can loop during Q-full.  tcm_qla2xxx_aborted_task
+		 * can get ahead of this cmd. tcm_qla2xxx_aborted_task
+		 * already kick start the free.
+		 */
+		pr_debug("write_pending aborted cmd[%p] refcount %d "
+			"transport_state %x, t_state %x, se_cmd_flags %x\n",
+			cmd,cmd->se_cmd.cmd_kref.refcount.counter,
+			cmd->se_cmd.transport_state,
+			cmd->se_cmd.t_state,
+			cmd->se_cmd.se_cmd_flags);
+		return 0;
+	}
 	cmd->cmd_flags |= BIT_3;
 	cmd->bufflen = se_cmd->data_length;
 	cmd->dma_data_direction = target_reverse_dma_direction(se_cmd);
@@ -405,7 +423,7 @@ static int tcm_qla2xxx_write_pending_status(struct se_cmd *se_cmd)
 	    se_cmd->t_state == TRANSPORT_COMPLETE_QF_WP) {
 		spin_unlock_irqrestore(&se_cmd->t_state_lock, flags);
 		wait_for_completion_timeout(&se_cmd->t_transport_stop_comp,
-					    3 * HZ);
+						50);
 		return 0;
 	}
 	spin_unlock_irqrestore(&se_cmd->t_state_lock, flags);
@@ -444,6 +462,9 @@ static int tcm_qla2xxx_handle_cmd(scsi_qla_host_t *vha, struct qla_tgt_cmd *cmd,
 	if (bidi)
 		flags |= TARGET_SCF_BIDI_OP;
 
+	if (se_cmd->cpuid != WORK_CPU_UNBOUND)
+		flags |= TARGET_SCF_USE_CPUID;
+
 	sess = cmd->sess;
 	if (!sess) {
 		pr_err("Unable to locate struct qla_tgt_sess from qla_tgt_cmd\n");
@@ -465,13 +486,25 @@ static int tcm_qla2xxx_handle_cmd(scsi_qla_host_t *vha, struct qla_tgt_cmd *cmd,
 static void tcm_qla2xxx_handle_data_work(struct work_struct *work)
 {
 	struct qla_tgt_cmd *cmd = container_of(work, struct qla_tgt_cmd, work);
+	unsigned long flags;
 
 	/*
 	 * Ensure that the complete FCP WRITE payload has been received.
 	 * Otherwise return an exception via CHECK_CONDITION status.
 	 */
 	cmd->cmd_in_wq = 0;
-	cmd->cmd_flags |= BIT_11;
+
+	spin_lock_irqsave(&cmd->cmd_lock, flags);
+	cmd->cmd_flags |= CMD_FLAG_DATA_WORK;
+	if (cmd->aborted) {
+		cmd->cmd_flags |= CMD_FLAG_DATA_WORK_FREE;
+		spin_unlock_irqrestore(&cmd->cmd_lock, flags);
+
+		tcm_qla2xxx_free_cmd(cmd);
+		return;
+	}
+	spin_unlock_irqrestore(&cmd->cmd_lock, flags);
+
 	cmd->vha->tgt_counters.qla_core_ret_ctio++;
 	if (!cmd->write_data_transferred) {
 		/*
@@ -546,6 +579,20 @@ static int tcm_qla2xxx_queue_data_in(struct se_cmd *se_cmd)
 	struct qla_tgt_cmd *cmd = container_of(se_cmd,
 				struct qla_tgt_cmd, se_cmd);
 
+	if (cmd->aborted) {
+		/* Cmd can loop during Q-full.  tcm_qla2xxx_aborted_task
+		 * can get ahead of this cmd. tcm_qla2xxx_aborted_task
+		 * already kick start the free.
+		 */
+		pr_debug("queue_data_in aborted cmd[%p] refcount %d "
+			"transport_state %x, t_state %x, se_cmd_flags %x\n",
+			cmd,cmd->se_cmd.cmd_kref.refcount.counter,
+			cmd->se_cmd.transport_state,
+			cmd->se_cmd.t_state,
+			cmd->se_cmd.se_cmd_flags);
+		return 0;
+	}
+
 	cmd->cmd_flags |= BIT_4;
 	cmd->bufflen = se_cmd->data_length;
 	cmd->dma_data_direction = target_reverse_dma_direction(se_cmd);
@@ -637,11 +684,34 @@ static void tcm_qla2xxx_queue_tm_rsp(struct se_cmd *se_cmd)
 	qlt_xmit_tm_rsp(mcmd);
 }
 
+
+#define DATA_WORK_NOT_FREE(_flags) \
+	(( _flags & (CMD_FLAG_DATA_WORK|CMD_FLAG_DATA_WORK_FREE)) == \
+	 CMD_FLAG_DATA_WORK)
 static void tcm_qla2xxx_aborted_task(struct se_cmd *se_cmd)
 {
 	struct qla_tgt_cmd *cmd = container_of(se_cmd,
 				struct qla_tgt_cmd, se_cmd);
-	qlt_abort_cmd(cmd);
+	unsigned long flags;
+
+	if (qlt_abort_cmd(cmd))
+		return;
+
+	spin_lock_irqsave(&cmd->cmd_lock, flags);
+	if ((cmd->state == QLA_TGT_STATE_NEW)||
+		((cmd->state == QLA_TGT_STATE_DATA_IN) &&
+		 DATA_WORK_NOT_FREE(cmd->cmd_flags)) ) {
+
+		cmd->cmd_flags |= CMD_FLAG_DATA_WORK_FREE;
+		spin_unlock_irqrestore(&cmd->cmd_lock, flags);
+		/* Cmd have not reached firmware.
+		 * Use this trigger to free it. */
+		tcm_qla2xxx_free_cmd(cmd);
+		return;
+	}
+	spin_unlock_irqrestore(&cmd->cmd_lock, flags);
+	return;
+
 }
 
 static void tcm_qla2xxx_clear_sess_lookup(struct tcm_qla2xxx_lport *,
diff --git a/drivers/scsi/scsi_devinfo.c b/drivers/scsi/scsi_devinfo.c
index 47b9d13f97b8..bbfbfd9e5aa3 100644
--- a/drivers/scsi/scsi_devinfo.c
+++ b/drivers/scsi/scsi_devinfo.c
@@ -205,6 +205,8 @@ static struct {
 	{"Intel", "Multi-Flex", NULL, BLIST_NO_RSOC},
 	{"iRiver", "iFP Mass Driver", NULL, BLIST_NOT_LOCKABLE | BLIST_INQUIRY_36},
 	{"LASOUND", "CDX7405", "3.10", BLIST_MAX5LUN | BLIST_SINGLELUN},
+	{"Marvell", "Console", NULL, BLIST_SKIP_VPD_PAGES},
+	{"Marvell", "91xx Config", "1.01", BLIST_SKIP_VPD_PAGES},
 	{"MATSHITA", "PD-1", NULL, BLIST_FORCELUN | BLIST_SINGLELUN},
 	{"MATSHITA", "DMC-LC5", NULL, BLIST_NOT_LOCKABLE | BLIST_INQUIRY_36},
 	{"MATSHITA", "DMC-LC40", NULL, BLIST_NOT_LOCKABLE | BLIST_INQUIRY_36},
diff --git a/drivers/scsi/scsi_lib.c b/drivers/scsi/scsi_lib.c
index fa6b2c4eb7a2..8c6e31874171 100644
--- a/drivers/scsi/scsi_lib.c
+++ b/drivers/scsi/scsi_lib.c
@@ -1344,6 +1344,7 @@ scsi_prep_return(struct request_queue *q, struct request *req, int ret)
 
 	switch (ret) {
 	case BLKPREP_KILL:
+	case BLKPREP_INVALID:
 		req->errors = DID_NO_CONNECT << 16;
 		/* release the command and kill it */
 		if (req->special) {
diff --git a/drivers/scsi/scsi_sysfs.c b/drivers/scsi/scsi_sysfs.c
index 4f18a851e2c7..00bc7218a7f8 100644
--- a/drivers/scsi/scsi_sysfs.c
+++ b/drivers/scsi/scsi_sysfs.c
@@ -1272,16 +1272,18 @@ static void __scsi_remove_target(struct scsi_target *starget)
 void scsi_remove_target(struct device *dev)
 {
 	struct Scsi_Host *shost = dev_to_shost(dev->parent);
-	struct scsi_target *starget;
+	struct scsi_target *starget, *last_target = NULL;
 	unsigned long flags;
 
 restart:
 	spin_lock_irqsave(shost->host_lock, flags);
 	list_for_each_entry(starget, &shost->__targets, siblings) {
-		if (starget->state == STARGET_DEL)
+		if (starget->state == STARGET_DEL ||
+		    starget == last_target)
 			continue;
 		if (starget->dev.parent == dev || &starget->dev == dev) {
 			kref_get(&starget->reap_ref);
+			last_target = starget;
 			spin_unlock_irqrestore(shost->host_lock, flags);
 			__scsi_remove_target(starget);
 			scsi_target_reap(starget);
diff --git a/drivers/scsi/sd.c b/drivers/scsi/sd.c
index bb669d32ccd0..d749da765df1 100644
--- a/drivers/scsi/sd.c
+++ b/drivers/scsi/sd.c
@@ -761,7 +761,7 @@ static int sd_setup_discard_cmnd(struct scsi_cmnd *cmd)
 		break;
 
 	default:
-		ret = BLKPREP_KILL;
+		ret = BLKPREP_INVALID;
 		goto out;
 	}
 
@@ -839,7 +839,7 @@ static int sd_setup_write_same_cmnd(struct scsi_cmnd *cmd)
 	int ret;
 
 	if (sdkp->device->no_write_same)
-		return BLKPREP_KILL;
+		return BLKPREP_INVALID;
 
 	BUG_ON(bio_offset(bio) || bio_iovec(bio).bv_len != sdp->sector_size);
 
diff --git a/drivers/scsi/storvsc_drv.c b/drivers/scsi/storvsc_drv.c
index 55627d097873..292c04eec9ad 100644
--- a/drivers/scsi/storvsc_drv.c
+++ b/drivers/scsi/storvsc_drv.c
@@ -42,6 +42,7 @@
 #include <scsi/scsi_devinfo.h>
 #include <scsi/scsi_dbg.h>
 #include <scsi/scsi_transport_fc.h>
+#include <scsi/scsi_transport.h>
 
 /*
  * All wire protocol details (storage protocol between the guest and the host)
@@ -477,19 +478,18 @@ struct hv_host_device {
 struct storvsc_scan_work {
 	struct work_struct work;
 	struct Scsi_Host *host;
-	uint lun;
+	u8 lun;
+	u8 tgt_id;
 };
 
 static void storvsc_device_scan(struct work_struct *work)
 {
 	struct storvsc_scan_work *wrk;
-	uint lun;
 	struct scsi_device *sdev;
 
 	wrk = container_of(work, struct storvsc_scan_work, work);
-	lun = wrk->lun;
 
-	sdev = scsi_device_lookup(wrk->host, 0, 0, lun);
+	sdev = scsi_device_lookup(wrk->host, 0, wrk->tgt_id, wrk->lun);
 	if (!sdev)
 		goto done;
 	scsi_rescan_device(&sdev->sdev_gendev);
@@ -540,7 +540,7 @@ static void storvsc_remove_lun(struct work_struct *work)
 	if (!scsi_host_get(wrk->host))
 		goto done;
 
-	sdev = scsi_device_lookup(wrk->host, 0, 0, wrk->lun);
+	sdev = scsi_device_lookup(wrk->host, 0, wrk->tgt_id, wrk->lun);
 
 	if (sdev) {
 		scsi_remove_device(sdev);
@@ -940,6 +940,7 @@ static void storvsc_handle_error(struct vmscsi_request *vm_srb,
 
 	wrk->host = host;
 	wrk->lun = vm_srb->lun;
+	wrk->tgt_id = vm_srb->target_id;
 	INIT_WORK(&wrk->work, process_err_fn);
 	schedule_work(&wrk->work);
 }
@@ -1770,6 +1771,11 @@ static int __init storvsc_drv_init(void)
 	fc_transport_template = fc_attach_transport(&fc_transport_functions);
 	if (!fc_transport_template)
 		return -ENODEV;
+
+	/*
+	 * Install Hyper-V specific timeout handler.
+	 */
+	fc_transport_template->eh_timed_out = storvsc_eh_timed_out;
 #endif
 
 	ret = vmbus_driver_register(&storvsc_drv);
diff --git a/drivers/sh/pm_runtime.c b/drivers/sh/pm_runtime.c
index 91a003011acf..a9bac3bf20de 100644
--- a/drivers/sh/pm_runtime.c
+++ b/drivers/sh/pm_runtime.c
@@ -34,7 +34,7 @@ static struct pm_clk_notifier_block platform_bus_notifier = {
 
 static int __init sh_pm_runtime_init(void)
 {
-	if (IS_ENABLED(CONFIG_ARCH_SHMOBILE)) {
+	if (IS_ENABLED(CONFIG_OF) && IS_ENABLED(CONFIG_ARCH_SHMOBILE)) {
 		if (!of_find_compatible_node(NULL, NULL,
 					     "renesas,cpg-mstp-clocks"))
 			return 0;
diff --git a/drivers/spi/spi-atmel.c b/drivers/spi/spi-atmel.c
index aebad36391c9..8feac599e9ab 100644
--- a/drivers/spi/spi-atmel.c
+++ b/drivers/spi/spi-atmel.c
@@ -1571,6 +1571,7 @@ static int atmel_spi_probe(struct platform_device *pdev)
 
 	as->use_cs_gpios = true;
 	if (atmel_spi_is_v2(as) &&
+	    pdev->dev.of_node &&
 	    !of_get_property(pdev->dev.of_node, "cs-gpios", NULL)) {
 		as->use_cs_gpios = false;
 		master->num_chipselect = 4;
diff --git a/drivers/spi/spi-bcm2835aux.c b/drivers/spi/spi-bcm2835aux.c
index 7de6f8472a81..ecc73c0a97cf 100644
--- a/drivers/spi/spi-bcm2835aux.c
+++ b/drivers/spi/spi-bcm2835aux.c
@@ -73,8 +73,8 @@
 
 /* Bitfields in CNTL1 */
 #define BCM2835_AUX_SPI_CNTL1_CSHIGH	0x00000700
-#define BCM2835_AUX_SPI_CNTL1_IDLE	0x00000080
-#define BCM2835_AUX_SPI_CNTL1_TXEMPTY	0x00000040
+#define BCM2835_AUX_SPI_CNTL1_TXEMPTY	0x00000080
+#define BCM2835_AUX_SPI_CNTL1_IDLE	0x00000040
 #define BCM2835_AUX_SPI_CNTL1_MSBF_IN	0x00000002
 #define BCM2835_AUX_SPI_CNTL1_KEEP_IN	0x00000001
 
diff --git a/drivers/spi/spi-fsl-espi.c b/drivers/spi/spi-fsl-espi.c
index 7fd6a4c009d2..7cb0c1921495 100644
--- a/drivers/spi/spi-fsl-espi.c
+++ b/drivers/spi/spi-fsl-espi.c
@@ -84,7 +84,7 @@ struct fsl_espi_transfer {
 /* SPCOM register values */
 #define SPCOM_CS(x)		((x) << 30)
 #define SPCOM_TRANLEN(x)	((x) << 0)
-#define	SPCOM_TRANLEN_MAX	0xFFFF	/* Max transaction length */
+#define	SPCOM_TRANLEN_MAX	0x10000	/* Max transaction length */
 
 #define AUTOSUSPEND_TIMEOUT 2000
 
@@ -233,7 +233,7 @@ static int fsl_espi_bufs(struct spi_device *spi, struct spi_transfer *t)
 	reinit_completion(&mpc8xxx_spi->done);
 
 	/* Set SPCOM[CS] and SPCOM[TRANLEN] field */
-	if ((t->len - 1) > SPCOM_TRANLEN_MAX) {
+	if (t->len > SPCOM_TRANLEN_MAX) {
 		dev_err(mpc8xxx_spi->dev, "Transaction length (%d)"
 				" beyond the SPCOM[TRANLEN] field\n", t->len);
 		return -EINVAL;
diff --git a/drivers/spi/spi-imx.c b/drivers/spi/spi-imx.c
index d98c33cb64f9..c688efa95e29 100644
--- a/drivers/spi/spi-imx.c
+++ b/drivers/spi/spi-imx.c
@@ -204,8 +204,8 @@ static bool spi_imx_can_dma(struct spi_master *master, struct spi_device *spi,
 {
 	struct spi_imx_data *spi_imx = spi_master_get_devdata(master);
 
-	if (spi_imx->dma_is_inited &&
-	    transfer->len > spi_imx->wml * sizeof(u32))
+	if (spi_imx->dma_is_inited && transfer->len >= spi_imx->wml &&
+	    (transfer->len % spi_imx->wml) == 0)
 		return true;
 	return false;
 }
@@ -919,8 +919,6 @@ static int spi_imx_dma_transfer(struct spi_imx_data *spi_imx,
 	struct dma_async_tx_descriptor *desc_tx = NULL, *desc_rx = NULL;
 	int ret;
 	unsigned long timeout;
-	u32 dma;
-	int left;
 	struct spi_master *master = spi_imx->bitbang.master;
 	struct sg_table *tx = &transfer->tx_sg, *rx = &transfer->rx_sg;
 
@@ -929,7 +927,7 @@ static int spi_imx_dma_transfer(struct spi_imx_data *spi_imx,
 					tx->sgl, tx->nents, DMA_MEM_TO_DEV,
 					DMA_PREP_INTERRUPT | DMA_CTRL_ACK);
 		if (!desc_tx)
-			goto no_dma;
+			goto tx_nodma;
 
 		desc_tx->callback = spi_imx_dma_tx_callback;
 		desc_tx->callback_param = (void *)spi_imx;
@@ -941,7 +939,7 @@ static int spi_imx_dma_transfer(struct spi_imx_data *spi_imx,
 					rx->sgl, rx->nents, DMA_DEV_TO_MEM,
 					DMA_PREP_INTERRUPT | DMA_CTRL_ACK);
 		if (!desc_rx)
-			goto no_dma;
+			goto rx_nodma;
 
 		desc_rx->callback = spi_imx_dma_rx_callback;
 		desc_rx->callback_param = (void *)spi_imx;
@@ -954,13 +952,6 @@ static int spi_imx_dma_transfer(struct spi_imx_data *spi_imx,
 	/* Trigger the cspi module. */
 	spi_imx->dma_finished = 0;
 
-	dma = readl(spi_imx->base + MX51_ECSPI_DMA);
-	dma = dma & (~MX51_ECSPI_DMA_RXT_WML_MASK);
-	/* Change RX_DMA_LENGTH trigger dma fetch tail data */
-	left = transfer->len % spi_imx->wml;
-	if (left)
-		writel(dma | (left << MX51_ECSPI_DMA_RXT_WML_OFFSET),
-				spi_imx->base + MX51_ECSPI_DMA);
 	/*
 	 * Set these order to avoid potential RX overflow. The overflow may
 	 * happen if we enable SPI HW before starting RX DMA due to rescheduling
@@ -992,10 +983,6 @@ static int spi_imx_dma_transfer(struct spi_imx_data *spi_imx,
 			spi_imx->devtype_data->reset(spi_imx);
 			dmaengine_terminate_all(master->dma_rx);
 		}
-		dma &= ~MX51_ECSPI_DMA_RXT_WML_MASK;
-		writel(dma |
-		       spi_imx->wml << MX51_ECSPI_DMA_RXT_WML_OFFSET,
-		       spi_imx->base + MX51_ECSPI_DMA);
 	}
 
 	spi_imx->dma_finished = 1;
@@ -1008,7 +995,9 @@ static int spi_imx_dma_transfer(struct spi_imx_data *spi_imx,
 
 	return ret;
 
-no_dma:
+rx_nodma:
+	dmaengine_terminate_all(master->dma_tx);
+tx_nodma:
 	pr_warn_once("%s %s: DMA not available, falling back to PIO\n",
 		     dev_driver_string(&master->dev),
 		     dev_name(&master->dev));
diff --git a/drivers/spi/spi-loopback-test.c b/drivers/spi/spi-loopback-test.c
index 894616f687b0..cf4bb36bee25 100644
--- a/drivers/spi/spi-loopback-test.c
+++ b/drivers/spi/spi-loopback-test.c
@@ -761,6 +761,7 @@ static int spi_test_run_iter(struct spi_device *spi,
 		test.iterate_transfer_mask = 1;
 
 	/* count number of transfers with tx/rx_buf != NULL */
+	rx_count = tx_count = 0;
 	for (i = 0; i < test.transfer_count; i++) {
 		if (test.transfers[i].tx_buf)
 			tx_count++;
diff --git a/drivers/spi/spi-omap2-mcspi.c b/drivers/spi/spi-omap2-mcspi.c
index 7273820275e9..0caa3c8bef46 100644
--- a/drivers/spi/spi-omap2-mcspi.c
+++ b/drivers/spi/spi-omap2-mcspi.c
@@ -1490,6 +1490,8 @@ static int omap2_mcspi_probe(struct platform_device *pdev)
 	return status;
 
 disable_pm:
+	pm_runtime_dont_use_autosuspend(&pdev->dev);
+	pm_runtime_put_sync(&pdev->dev);
 	pm_runtime_disable(&pdev->dev);
 free_master:
 	spi_master_put(master);
@@ -1501,6 +1503,7 @@ static int omap2_mcspi_remove(struct platform_device *pdev)
 	struct spi_master *master = platform_get_drvdata(pdev);
 	struct omap2_mcspi *mcspi = spi_master_get_devdata(master);
 
+	pm_runtime_dont_use_autosuspend(mcspi->dev);
 	pm_runtime_put_sync(mcspi->dev);
 	pm_runtime_disable(&pdev->dev);
 
diff --git a/drivers/spi/spi-rockchip.c b/drivers/spi/spi-rockchip.c
index 79a8bc4f6cec..7cb1b2d710c1 100644
--- a/drivers/spi/spi-rockchip.c
+++ b/drivers/spi/spi-rockchip.c
@@ -749,6 +749,7 @@ static int rockchip_spi_probe(struct platform_device *pdev)
 	return 0;
 
 err_register_master:
+	pm_runtime_disable(&pdev->dev);
 	if (rs->dma_tx.ch)
 		dma_release_channel(rs->dma_tx.ch);
 	if (rs->dma_rx.ch)
@@ -778,6 +779,8 @@ static int rockchip_spi_remove(struct platform_device *pdev)
 	if (rs->dma_rx.ch)
 		dma_release_channel(rs->dma_rx.ch);
 
+	spi_master_put(master);
+
 	return 0;
 }
 
diff --git a/drivers/ssb/Kconfig b/drivers/ssb/Kconfig
index 0c675861623f..d8e4219c2324 100644
--- a/drivers/ssb/Kconfig
+++ b/drivers/ssb/Kconfig
@@ -83,6 +83,7 @@ config SSB_SDIOHOST
 config SSB_HOST_SOC
 	bool "Support for SSB bus on SoC"
 	depends on SSB && BCM47XX_NVRAM
+	select SSB_SPROM
 	help
 	  Host interface for a SSB directly mapped into memory. This is
 	  for some Broadcom SoCs from the BCM47xx and BCM53xx lines.
diff --git a/drivers/target/target_core_configfs.c b/drivers/target/target_core_configfs.c
index 3327c49674d3..713c63d9681b 100644
--- a/drivers/target/target_core_configfs.c
+++ b/drivers/target/target_core_configfs.c
@@ -898,7 +898,7 @@ static ssize_t unmap_zeroes_data_store(struct config_item *item,
 	da->unmap_zeroes_data = flag;
 	pr_debug("dev[%p]: SE Device Thin Provisioning LBPRZ bit: %d\n",
 		 da->da_dev, flag);
-	return 0;
+	return count;
 }
 
 /*
diff --git a/drivers/target/target_core_device.c b/drivers/target/target_core_device.c
index cacd97a8cbd0..da457e25717a 100644
--- a/drivers/target/target_core_device.c
+++ b/drivers/target/target_core_device.c
@@ -828,6 +828,50 @@ struct se_device *target_alloc_device(struct se_hba *hba, const char *name)
 	return dev;
 }
 
+/*
+ * Check if the underlying struct block_device request_queue supports
+ * the QUEUE_FLAG_DISCARD bit for UNMAP/WRITE_SAME in SCSI + TRIM
+ * in ATA and we need to set TPE=1
+ */
+bool target_configure_unmap_from_queue(struct se_dev_attrib *attrib,
+				       struct request_queue *q, int block_size)
+{
+	if (!blk_queue_discard(q))
+		return false;
+
+	attrib->max_unmap_lba_count = (q->limits.max_discard_sectors << 9) /
+								block_size;
+	/*
+	 * Currently hardcoded to 1 in Linux/SCSI code..
+	 */
+	attrib->max_unmap_block_desc_count = 1;
+	attrib->unmap_granularity = q->limits.discard_granularity / block_size;
+	attrib->unmap_granularity_alignment = q->limits.discard_alignment /
+								block_size;
+	attrib->unmap_zeroes_data = q->limits.discard_zeroes_data;
+	return true;
+}
+EXPORT_SYMBOL(target_configure_unmap_from_queue);
+
+/*
+ * Convert from blocksize advertised to the initiator to the 512 byte
+ * units unconditionally used by the Linux block layer.
+ */
+sector_t target_to_linux_sector(struct se_device *dev, sector_t lb)
+{
+	switch (dev->dev_attrib.block_size) {
+	case 4096:
+		return lb << 3;
+	case 2048:
+		return lb << 2;
+	case 1024:
+		return lb << 1;
+	default:
+		return lb;
+	}
+}
+EXPORT_SYMBOL(target_to_linux_sector);
+
 int target_configure_device(struct se_device *dev)
 {
 	struct se_hba *hba = dev->se_hba;
diff --git a/drivers/target/target_core_file.c b/drivers/target/target_core_file.c
index e3195700211a..75f0f08b2a34 100644
--- a/drivers/target/target_core_file.c
+++ b/drivers/target/target_core_file.c
@@ -160,25 +160,11 @@ static int fd_configure_device(struct se_device *dev)
 			" block_device blocks: %llu logical_block_size: %d\n",
 			dev_size, div_u64(dev_size, fd_dev->fd_block_size),
 			fd_dev->fd_block_size);
-		/*
-		 * Check if the underlying struct block_device request_queue supports
-		 * the QUEUE_FLAG_DISCARD bit for UNMAP/WRITE_SAME in SCSI + TRIM
-		 * in ATA and we need to set TPE=1
-		 */
-		if (blk_queue_discard(q)) {
-			dev->dev_attrib.max_unmap_lba_count =
-				q->limits.max_discard_sectors;
-			/*
-			 * Currently hardcoded to 1 in Linux/SCSI code..
-			 */
-			dev->dev_attrib.max_unmap_block_desc_count = 1;
-			dev->dev_attrib.unmap_granularity =
-				q->limits.discard_granularity >> 9;
-			dev->dev_attrib.unmap_granularity_alignment =
-				q->limits.discard_alignment;
+
+		if (target_configure_unmap_from_queue(&dev->dev_attrib, q,
+						      fd_dev->fd_block_size))
 			pr_debug("IFILE: BLOCK Discard support available,"
-					" disabled by default\n");
-		}
+				 " disabled by default\n");
 		/*
 		 * Enable write same emulation for IBLOCK and use 0xFFFF as
 		 * the smaller WRITE_SAME(10) only has a two-byte block count.
@@ -490,9 +476,12 @@ fd_execute_unmap(struct se_cmd *cmd, sector_t lba, sector_t nolb)
 	if (S_ISBLK(inode->i_mode)) {
 		/* The backend is block device, use discard */
 		struct block_device *bdev = inode->i_bdev;
+		struct se_device *dev = cmd->se_dev;
 
-		ret = blkdev_issue_discard(bdev, lba,
-				nolb, GFP_KERNEL, 0);
+		ret = blkdev_issue_discard(bdev,
+					   target_to_linux_sector(dev, lba),
+					   target_to_linux_sector(dev,  nolb),
+					   GFP_KERNEL, 0);
 		if (ret < 0) {
 			pr_warn("FILEIO: blkdev_issue_discard() failed: %d\n",
 				ret);
diff --git a/drivers/target/target_core_iblock.c b/drivers/target/target_core_iblock.c
index 5a2899f9f50b..abe4eb997a84 100644
--- a/drivers/target/target_core_iblock.c
+++ b/drivers/target/target_core_iblock.c
@@ -121,29 +121,11 @@ static int iblock_configure_device(struct se_device *dev)
 	dev->dev_attrib.hw_max_sectors = queue_max_hw_sectors(q);
 	dev->dev_attrib.hw_queue_depth = q->nr_requests;
 
-	/*
-	 * Check if the underlying struct block_device request_queue supports
-	 * the QUEUE_FLAG_DISCARD bit for UNMAP/WRITE_SAME in SCSI + TRIM
-	 * in ATA and we need to set TPE=1
-	 */
-	if (blk_queue_discard(q)) {
-		dev->dev_attrib.max_unmap_lba_count =
-				q->limits.max_discard_sectors;
-
-		/*
-		 * Currently hardcoded to 1 in Linux/SCSI code..
-		 */
-		dev->dev_attrib.max_unmap_block_desc_count = 1;
-		dev->dev_attrib.unmap_granularity =
-				q->limits.discard_granularity >> 9;
-		dev->dev_attrib.unmap_granularity_alignment =
-				q->limits.discard_alignment;
-		dev->dev_attrib.unmap_zeroes_data =
-				q->limits.discard_zeroes_data;
-
+	if (target_configure_unmap_from_queue(&dev->dev_attrib, q,
+					      dev->dev_attrib.hw_block_size))
 		pr_debug("IBLOCK: BLOCK Discard support available,"
-				" disabled by default\n");
-	}
+			 " disabled by default\n");
+
 	/*
 	 * Enable write same emulation for IBLOCK and use 0xFFFF as
 	 * the smaller WRITE_SAME(10) only has a two-byte block count.
@@ -415,9 +397,13 @@ static sense_reason_t
 iblock_execute_unmap(struct se_cmd *cmd, sector_t lba, sector_t nolb)
 {
 	struct block_device *bdev = IBLOCK_DEV(cmd->se_dev)->ibd_bd;
+	struct se_device *dev = cmd->se_dev;
 	int ret;
 
-	ret = blkdev_issue_discard(bdev, lba, nolb, GFP_KERNEL, 0);
+	ret = blkdev_issue_discard(bdev,
+				   target_to_linux_sector(dev, lba),
+				   target_to_linux_sector(dev,  nolb),
+				   GFP_KERNEL, 0);
 	if (ret < 0) {
 		pr_err("blkdev_issue_discard() failed: %d\n", ret);
 		return TCM_LOGICAL_UNIT_COMMUNICATION_FAILURE;
@@ -433,8 +419,10 @@ iblock_execute_write_same(struct se_cmd *cmd)
 	struct scatterlist *sg;
 	struct bio *bio;
 	struct bio_list list;
-	sector_t block_lba = cmd->t_task_lba;
-	sector_t sectors = sbc_get_write_same_sectors(cmd);
+	struct se_device *dev = cmd->se_dev;
+	sector_t block_lba = target_to_linux_sector(dev, cmd->t_task_lba);
+	sector_t sectors = target_to_linux_sector(dev,
+					sbc_get_write_same_sectors(cmd));
 
 	if (cmd->prot_op) {
 		pr_err("WRITE_SAME: Protection information with IBLOCK"
@@ -648,12 +636,12 @@ iblock_execute_rw(struct se_cmd *cmd, struct scatterlist *sgl, u32 sgl_nents,
 		  enum dma_data_direction data_direction)
 {
 	struct se_device *dev = cmd->se_dev;
+	sector_t block_lba = target_to_linux_sector(dev, cmd->t_task_lba);
 	struct iblock_req *ibr;
 	struct bio *bio, *bio_start;
 	struct bio_list list;
 	struct scatterlist *sg;
 	u32 sg_num = sgl_nents;
-	sector_t block_lba;
 	unsigned bio_cnt;
 	int rw = 0;
 	int i;
@@ -679,24 +667,6 @@ iblock_execute_rw(struct se_cmd *cmd, struct scatterlist *sgl, u32 sgl_nents,
 		rw = READ;
 	}
 
-	/*
-	 * Convert the blocksize advertised to the initiator to the 512 byte
-	 * units unconditionally used by the Linux block layer.
-	 */
-	if (dev->dev_attrib.block_size == 4096)
-		block_lba = (cmd->t_task_lba << 3);
-	else if (dev->dev_attrib.block_size == 2048)
-		block_lba = (cmd->t_task_lba << 2);
-	else if (dev->dev_attrib.block_size == 1024)
-		block_lba = (cmd->t_task_lba << 1);
-	else if (dev->dev_attrib.block_size == 512)
-		block_lba = cmd->t_task_lba;
-	else {
-		pr_err("Unsupported SCSI -> BLOCK LBA conversion:"
-				" %u\n", dev->dev_attrib.block_size);
-		return TCM_LOGICAL_UNIT_COMMUNICATION_FAILURE;
-	}
-
 	ibr = kzalloc(sizeof(struct iblock_req), GFP_KERNEL);
 	if (!ibr)
 		goto fail;
diff --git a/drivers/target/target_core_internal.h b/drivers/target/target_core_internal.h
index dae0750c2032..db4412fe6b8a 100644
--- a/drivers/target/target_core_internal.h
+++ b/drivers/target/target_core_internal.h
@@ -141,7 +141,6 @@ void	transport_dump_vpd_proto_id(struct t10_vpd *, unsigned char *, int);
 int	transport_dump_vpd_assoc(struct t10_vpd *, unsigned char *, int);
 int	transport_dump_vpd_ident_type(struct t10_vpd *, unsigned char *, int);
 int	transport_dump_vpd_ident(struct t10_vpd *, unsigned char *, int);
-bool	target_stop_cmd(struct se_cmd *cmd, unsigned long *flags);
 void	transport_clear_lun_ref(struct se_lun *);
 void	transport_send_task_abort(struct se_cmd *);
 sense_reason_t	target_cmd_size_check(struct se_cmd *cmd, unsigned int size);
diff --git a/drivers/target/target_core_tmr.c b/drivers/target/target_core_tmr.c
index fcdcb117c60d..82a663ba9800 100644
--- a/drivers/target/target_core_tmr.c
+++ b/drivers/target/target_core_tmr.c
@@ -68,23 +68,25 @@ void core_tmr_release_req(struct se_tmr_req *tmr)
 
 	if (dev) {
 		spin_lock_irqsave(&dev->se_tmr_lock, flags);
-		list_del(&tmr->tmr_list);
+		list_del_init(&tmr->tmr_list);
 		spin_unlock_irqrestore(&dev->se_tmr_lock, flags);
 	}
 
 	kfree(tmr);
 }
 
-static void core_tmr_handle_tas_abort(
-	struct se_node_acl *tmr_nacl,
-	struct se_cmd *cmd,
-	int tas)
+static void core_tmr_handle_tas_abort(struct se_cmd *cmd, int tas)
 {
-	bool remove = true;
+	unsigned long flags;
+	bool remove = true, send_tas;
 	/*
 	 * TASK ABORTED status (TAS) bit support
 	 */
-	if ((tmr_nacl && (tmr_nacl != cmd->se_sess->se_node_acl)) && tas) {
+	spin_lock_irqsave(&cmd->t_state_lock, flags);
+	send_tas = (cmd->transport_state & CMD_T_TAS);
+	spin_unlock_irqrestore(&cmd->t_state_lock, flags);
+
+	if (send_tas) {
 		remove = false;
 		transport_send_task_abort(cmd);
 	}
@@ -107,6 +109,46 @@ static int target_check_cdb_and_preempt(struct list_head *list,
 	return 1;
 }
 
+static bool __target_check_io_state(struct se_cmd *se_cmd,
+				    struct se_session *tmr_sess, int tas)
+{
+	struct se_session *sess = se_cmd->se_sess;
+
+	assert_spin_locked(&sess->sess_cmd_lock);
+	WARN_ON_ONCE(!irqs_disabled());
+	/*
+	 * If command already reached CMD_T_COMPLETE state within
+	 * target_complete_cmd() or CMD_T_FABRIC_STOP due to shutdown,
+	 * this se_cmd has been passed to fabric driver and will
+	 * not be aborted.
+	 *
+	 * Otherwise, obtain a local se_cmd->cmd_kref now for TMR
+	 * ABORT_TASK + LUN_RESET for CMD_T_ABORTED processing as
+	 * long as se_cmd->cmd_kref is still active unless zero.
+	 */
+	spin_lock(&se_cmd->t_state_lock);
+	if (se_cmd->transport_state & (CMD_T_COMPLETE | CMD_T_FABRIC_STOP)) {
+		pr_debug("Attempted to abort io tag: %llu already complete or"
+			" fabric stop, skipping\n", se_cmd->tag);
+		spin_unlock(&se_cmd->t_state_lock);
+		return false;
+	}
+	if (sess->sess_tearing_down || se_cmd->cmd_wait_set) {
+		pr_debug("Attempted to abort io tag: %llu already shutdown,"
+			" skipping\n", se_cmd->tag);
+		spin_unlock(&se_cmd->t_state_lock);
+		return false;
+	}
+	se_cmd->transport_state |= CMD_T_ABORTED;
+
+	if ((tmr_sess != se_cmd->se_sess) && tas)
+		se_cmd->transport_state |= CMD_T_TAS;
+
+	spin_unlock(&se_cmd->t_state_lock);
+
+	return kref_get_unless_zero(&se_cmd->cmd_kref);
+}
+
 void core_tmr_abort_task(
 	struct se_device *dev,
 	struct se_tmr_req *tmr,
@@ -130,34 +172,22 @@ void core_tmr_abort_task(
 		if (tmr->ref_task_tag != ref_tag)
 			continue;
 
-		if (!kref_get_unless_zero(&se_cmd->cmd_kref))
-			continue;
-
 		printk("ABORT_TASK: Found referenced %s task_tag: %llu\n",
 			se_cmd->se_tfo->get_fabric_name(), ref_tag);
 
-		spin_lock(&se_cmd->t_state_lock);
-		if (se_cmd->transport_state & CMD_T_COMPLETE) {
-			printk("ABORT_TASK: ref_tag: %llu already complete,"
-			       " skipping\n", ref_tag);
-			spin_unlock(&se_cmd->t_state_lock);
+		if (!__target_check_io_state(se_cmd, se_sess, 0)) {
 			spin_unlock_irqrestore(&se_sess->sess_cmd_lock, flags);
-
 			target_put_sess_cmd(se_cmd);
-
 			goto out;
 		}
-		se_cmd->transport_state |= CMD_T_ABORTED;
-		spin_unlock(&se_cmd->t_state_lock);
-
 		list_del_init(&se_cmd->se_cmd_list);
 		spin_unlock_irqrestore(&se_sess->sess_cmd_lock, flags);
 
 		cancel_work_sync(&se_cmd->work);
 		transport_wait_for_tasks(se_cmd);
 
-		target_put_sess_cmd(se_cmd);
 		transport_cmd_finish_abort(se_cmd, true);
+		target_put_sess_cmd(se_cmd);
 
 		printk("ABORT_TASK: Sending TMR_FUNCTION_COMPLETE for"
 				" ref_tag: %llu\n", ref_tag);
@@ -178,9 +208,11 @@ static void core_tmr_drain_tmr_list(
 	struct list_head *preempt_and_abort_list)
 {
 	LIST_HEAD(drain_tmr_list);
+	struct se_session *sess;
 	struct se_tmr_req *tmr_p, *tmr_pp;
 	struct se_cmd *cmd;
 	unsigned long flags;
+	bool rc;
 	/*
 	 * Release all pending and outgoing TMRs aside from the received
 	 * LUN_RESET tmr..
@@ -206,17 +238,39 @@ static void core_tmr_drain_tmr_list(
 		if (target_check_cdb_and_preempt(preempt_and_abort_list, cmd))
 			continue;
 
+		sess = cmd->se_sess;
+		if (WARN_ON_ONCE(!sess))
+			continue;
+
+		spin_lock(&sess->sess_cmd_lock);
 		spin_lock(&cmd->t_state_lock);
-		if (!(cmd->transport_state & CMD_T_ACTIVE)) {
+		if (!(cmd->transport_state & CMD_T_ACTIVE) ||
+		     (cmd->transport_state & CMD_T_FABRIC_STOP)) {
 			spin_unlock(&cmd->t_state_lock);
+			spin_unlock(&sess->sess_cmd_lock);
 			continue;
 		}
 		if (cmd->t_state == TRANSPORT_ISTATE_PROCESSING) {
 			spin_unlock(&cmd->t_state_lock);
+			spin_unlock(&sess->sess_cmd_lock);
 			continue;
 		}
+		if (sess->sess_tearing_down || cmd->cmd_wait_set) {
+			spin_unlock(&cmd->t_state_lock);
+			spin_unlock(&sess->sess_cmd_lock);
+			continue;
+		}
+		cmd->transport_state |= CMD_T_ABORTED;
 		spin_unlock(&cmd->t_state_lock);
 
+		rc = kref_get_unless_zero(&cmd->cmd_kref);
+		if (!rc) {
+			printk("LUN_RESET TMR: non-zero kref_get_unless_zero\n");
+			spin_unlock(&sess->sess_cmd_lock);
+			continue;
+		}
+		spin_unlock(&sess->sess_cmd_lock);
+
 		list_move_tail(&tmr_p->tmr_list, &drain_tmr_list);
 	}
 	spin_unlock_irqrestore(&dev->se_tmr_lock, flags);
@@ -230,20 +284,26 @@ static void core_tmr_drain_tmr_list(
 			(preempt_and_abort_list) ? "Preempt" : "", tmr_p,
 			tmr_p->function, tmr_p->response, cmd->t_state);
 
+		cancel_work_sync(&cmd->work);
+		transport_wait_for_tasks(cmd);
+
 		transport_cmd_finish_abort(cmd, 1);
+		target_put_sess_cmd(cmd);
 	}
 }
 
 static void core_tmr_drain_state_list(
 	struct se_device *dev,
 	struct se_cmd *prout_cmd,
-	struct se_node_acl *tmr_nacl,
+	struct se_session *tmr_sess,
 	int tas,
 	struct list_head *preempt_and_abort_list)
 {
 	LIST_HEAD(drain_task_list);
+	struct se_session *sess;
 	struct se_cmd *cmd, *next;
 	unsigned long flags;
+	int rc;
 
 	/*
 	 * Complete outstanding commands with TASK_ABORTED SAM status.
@@ -282,6 +342,16 @@ static void core_tmr_drain_state_list(
 		if (prout_cmd == cmd)
 			continue;
 
+		sess = cmd->se_sess;
+		if (WARN_ON_ONCE(!sess))
+			continue;
+
+		spin_lock(&sess->sess_cmd_lock);
+		rc = __target_check_io_state(cmd, tmr_sess, tas);
+		spin_unlock(&sess->sess_cmd_lock);
+		if (!rc)
+			continue;
+
 		list_move_tail(&cmd->state_list, &drain_task_list);
 		cmd->state_active = false;
 	}
@@ -289,7 +359,7 @@ static void core_tmr_drain_state_list(
 
 	while (!list_empty(&drain_task_list)) {
 		cmd = list_entry(drain_task_list.next, struct se_cmd, state_list);
-		list_del(&cmd->state_list);
+		list_del_init(&cmd->state_list);
 
 		pr_debug("LUN_RESET: %s cmd: %p"
 			" ITT/CmdSN: 0x%08llx/0x%08x, i_state: %d, t_state: %d"
@@ -313,16 +383,11 @@ static void core_tmr_drain_state_list(
 		 * loop above, but we do it down here given that
 		 * cancel_work_sync may block.
 		 */
-		if (cmd->t_state == TRANSPORT_COMPLETE)
-			cancel_work_sync(&cmd->work);
-
-		spin_lock_irqsave(&cmd->t_state_lock, flags);
-		target_stop_cmd(cmd, &flags);
-
-		cmd->transport_state |= CMD_T_ABORTED;
-		spin_unlock_irqrestore(&cmd->t_state_lock, flags);
+		cancel_work_sync(&cmd->work);
+		transport_wait_for_tasks(cmd);
 
-		core_tmr_handle_tas_abort(tmr_nacl, cmd, tas);
+		core_tmr_handle_tas_abort(cmd, tas);
+		target_put_sess_cmd(cmd);
 	}
 }
 
@@ -334,6 +399,7 @@ int core_tmr_lun_reset(
 {
 	struct se_node_acl *tmr_nacl = NULL;
 	struct se_portal_group *tmr_tpg = NULL;
+	struct se_session *tmr_sess = NULL;
 	int tas;
         /*
 	 * TASK_ABORTED status bit, this is configurable via ConfigFS
@@ -352,8 +418,9 @@ int core_tmr_lun_reset(
 	 * or struct se_device passthrough..
 	 */
 	if (tmr && tmr->task_cmd && tmr->task_cmd->se_sess) {
-		tmr_nacl = tmr->task_cmd->se_sess->se_node_acl;
-		tmr_tpg = tmr->task_cmd->se_sess->se_tpg;
+		tmr_sess = tmr->task_cmd->se_sess;
+		tmr_nacl = tmr_sess->se_node_acl;
+		tmr_tpg = tmr_sess->se_tpg;
 		if (tmr_nacl && tmr_tpg) {
 			pr_debug("LUN_RESET: TMR caller fabric: %s"
 				" initiator port %s\n",
@@ -366,7 +433,7 @@ int core_tmr_lun_reset(
 		dev->transport->name, tas);
 
 	core_tmr_drain_tmr_list(dev, tmr, preempt_and_abort_list);
-	core_tmr_drain_state_list(dev, prout_cmd, tmr_nacl, tas,
+	core_tmr_drain_state_list(dev, prout_cmd, tmr_sess, tas,
 				preempt_and_abort_list);
 
 	/*
diff --git a/drivers/target/target_core_transport.c b/drivers/target/target_core_transport.c
index 9f3608e10f25..867bc6d0a68a 100644
--- a/drivers/target/target_core_transport.c
+++ b/drivers/target/target_core_transport.c
@@ -534,9 +534,6 @@ void transport_deregister_session(struct se_session *se_sess)
 }
 EXPORT_SYMBOL(transport_deregister_session);
 
-/*
- * Called with cmd->t_state_lock held.
- */
 static void target_remove_from_state_list(struct se_cmd *cmd)
 {
 	struct se_device *dev = cmd->se_dev;
@@ -561,10 +558,6 @@ static int transport_cmd_check_stop(struct se_cmd *cmd, bool remove_from_lists,
 {
 	unsigned long flags;
 
-	spin_lock_irqsave(&cmd->t_state_lock, flags);
-	if (write_pending)
-		cmd->t_state = TRANSPORT_WRITE_PENDING;
-
 	if (remove_from_lists) {
 		target_remove_from_state_list(cmd);
 
@@ -574,6 +567,10 @@ static int transport_cmd_check_stop(struct se_cmd *cmd, bool remove_from_lists,
 		cmd->se_lun = NULL;
 	}
 
+	spin_lock_irqsave(&cmd->t_state_lock, flags);
+	if (write_pending)
+		cmd->t_state = TRANSPORT_WRITE_PENDING;
+
 	/*
 	 * Determine if frontend context caller is requesting the stopping of
 	 * this command for frontend exceptions.
@@ -627,6 +624,8 @@ static void transport_lun_remove_cmd(struct se_cmd *cmd)
 
 void transport_cmd_finish_abort(struct se_cmd *cmd, int remove)
 {
+	bool ack_kref = (cmd->se_cmd_flags & SCF_ACK_KREF);
+
 	if (cmd->se_cmd_flags & SCF_SE_LUN_CMD)
 		transport_lun_remove_cmd(cmd);
 	/*
@@ -638,7 +637,7 @@ void transport_cmd_finish_abort(struct se_cmd *cmd, int remove)
 
 	if (transport_cmd_check_stop_to_fabric(cmd))
 		return;
-	if (remove)
+	if (remove && ack_kref)
 		transport_put_cmd(cmd);
 }
 
@@ -694,19 +693,10 @@ void target_complete_cmd(struct se_cmd *cmd, u8 scsi_status)
 	}
 
 	/*
-	 * See if we are waiting to complete for an exception condition.
-	 */
-	if (cmd->transport_state & CMD_T_REQUEST_STOP) {
-		spin_unlock_irqrestore(&cmd->t_state_lock, flags);
-		complete(&cmd->task_stop_comp);
-		return;
-	}
-
-	/*
 	 * Check for case where an explicit ABORT_TASK has been received
 	 * and transport_wait_for_tasks() will be waiting for completion..
 	 */
-	if (cmd->transport_state & CMD_T_ABORTED &&
+	if (cmd->transport_state & CMD_T_ABORTED ||
 	    cmd->transport_state & CMD_T_STOP) {
 		spin_unlock_irqrestore(&cmd->t_state_lock, flags);
 		complete_all(&cmd->t_transport_stop_comp);
@@ -721,10 +711,10 @@ void target_complete_cmd(struct se_cmd *cmd, u8 scsi_status)
 	cmd->transport_state |= (CMD_T_COMPLETE | CMD_T_ACTIVE);
 	spin_unlock_irqrestore(&cmd->t_state_lock, flags);
 
-	if (cmd->cpuid == -1)
-		queue_work(target_completion_wq, &cmd->work);
-	else
+	if (cmd->se_cmd_flags & SCF_USE_CPUID)
 		queue_work_on(cmd->cpuid, target_completion_wq, &cmd->work);
+	else
+		queue_work(target_completion_wq, &cmd->work);
 }
 EXPORT_SYMBOL(target_complete_cmd);
 
@@ -1203,7 +1193,6 @@ void transport_init_se_cmd(
 	INIT_LIST_HEAD(&cmd->state_list);
 	init_completion(&cmd->t_transport_stop_comp);
 	init_completion(&cmd->cmd_wait_comp);
-	init_completion(&cmd->task_stop_comp);
 	spin_lock_init(&cmd->t_state_lock);
 	kref_init(&cmd->cmd_kref);
 	cmd->transport_state = CMD_T_DEV_ACTIVE;
@@ -1437,6 +1426,12 @@ int target_submit_cmd_map_sgls(struct se_cmd *se_cmd, struct se_session *se_sess
 	 */
 	transport_init_se_cmd(se_cmd, se_tpg->se_tpg_tfo, se_sess,
 				data_length, data_dir, task_attr, sense);
+
+	if (flags & TARGET_SCF_USE_CPUID)
+		se_cmd->se_cmd_flags |= SCF_USE_CPUID;
+	else
+		se_cmd->cpuid = WORK_CPU_UNBOUND;
+
 	if (flags & TARGET_SCF_UNKNOWN_SIZE)
 		se_cmd->unknown_data_length = 1;
 	/*
@@ -1635,33 +1630,6 @@ int target_submit_tmr(struct se_cmd *se_cmd, struct se_session *se_sess,
 EXPORT_SYMBOL(target_submit_tmr);
 
 /*
- * If the cmd is active, request it to be stopped and sleep until it
- * has completed.
- */
-bool target_stop_cmd(struct se_cmd *cmd, unsigned long *flags)
-	__releases(&cmd->t_state_lock)
-	__acquires(&cmd->t_state_lock)
-{
-	bool was_active = false;
-
-	if (cmd->transport_state & CMD_T_BUSY) {
-		cmd->transport_state |= CMD_T_REQUEST_STOP;
-		spin_unlock_irqrestore(&cmd->t_state_lock, *flags);
-
-		pr_debug("cmd %p waiting to complete\n", cmd);
-		wait_for_completion(&cmd->task_stop_comp);
-		pr_debug("cmd %p stopped successfully\n", cmd);
-
-		spin_lock_irqsave(&cmd->t_state_lock, *flags);
-		cmd->transport_state &= ~CMD_T_REQUEST_STOP;
-		cmd->transport_state &= ~CMD_T_BUSY;
-		was_active = true;
-	}
-
-	return was_active;
-}
-
-/*
  * Handle SAM-esque emulation for generic transport request failures.
  */
 void transport_generic_request_failure(struct se_cmd *cmd,
@@ -1859,19 +1827,21 @@ static bool target_handle_task_attr(struct se_cmd *cmd)
 	return true;
 }
 
+static int __transport_check_aborted_status(struct se_cmd *, int);
+
 void target_execute_cmd(struct se_cmd *cmd)
 {
 	/*
-	 * If the received CDB has aleady been aborted stop processing it here.
-	 */
-	if (transport_check_aborted_status(cmd, 1))
-		return;
-
-	/*
 	 * Determine if frontend context caller is requesting the stopping of
 	 * this command for frontend exceptions.
+	 *
+	 * If the received CDB has aleady been aborted stop processing it here.
 	 */
 	spin_lock_irq(&cmd->t_state_lock);
+	if (__transport_check_aborted_status(cmd, 1)) {
+		spin_unlock_irq(&cmd->t_state_lock);
+		return;
+	}
 	if (cmd->transport_state & CMD_T_STOP) {
 		pr_debug("%s:%d CMD_T_STOP for ITT: 0x%08llx\n",
 			__func__, __LINE__, cmd->tag);
@@ -2222,20 +2192,14 @@ static inline void transport_free_pages(struct se_cmd *cmd)
 }
 
 /**
- * transport_release_cmd - free a command
- * @cmd:       command to free
+ * transport_put_cmd - release a reference to a command
+ * @cmd:       command to release
  *
- * This routine unconditionally frees a command, and reference counting
- * or list removal must be done in the caller.
+ * This routine releases our reference to the command and frees it if possible.
  */
-static int transport_release_cmd(struct se_cmd *cmd)
+static int transport_put_cmd(struct se_cmd *cmd)
 {
 	BUG_ON(!cmd->se_tfo);
-
-	if (cmd->se_cmd_flags & SCF_SCSI_TMR_CDB)
-		core_tmr_release_req(cmd->se_tmr_req);
-	if (cmd->t_task_cdb != cmd->__t_task_cdb)
-		kfree(cmd->t_task_cdb);
 	/*
 	 * If this cmd has been setup with target_get_sess_cmd(), drop
 	 * the kref and call ->release_cmd() in kref callback.
@@ -2243,18 +2207,6 @@ static int transport_release_cmd(struct se_cmd *cmd)
 	return target_put_sess_cmd(cmd);
 }
 
-/**
- * transport_put_cmd - release a reference to a command
- * @cmd:       command to release
- *
- * This routine releases our reference to the command and frees it if possible.
- */
-static int transport_put_cmd(struct se_cmd *cmd)
-{
-	transport_free_pages(cmd);
-	return transport_release_cmd(cmd);
-}
-
 void *transport_kmap_data_sg(struct se_cmd *cmd)
 {
 	struct scatterlist *sg = cmd->t_data_sg;
@@ -2450,34 +2402,58 @@ static void transport_write_pending_qf(struct se_cmd *cmd)
 	}
 }
 
-int transport_generic_free_cmd(struct se_cmd *cmd, int wait_for_tasks)
+static bool
+__transport_wait_for_tasks(struct se_cmd *, bool, bool *, bool *,
+			   unsigned long *flags);
+
+static void target_wait_free_cmd(struct se_cmd *cmd, bool *aborted, bool *tas)
 {
 	unsigned long flags;
+
+	spin_lock_irqsave(&cmd->t_state_lock, flags);
+	__transport_wait_for_tasks(cmd, true, aborted, tas, &flags);
+	spin_unlock_irqrestore(&cmd->t_state_lock, flags);
+}
+
+int transport_generic_free_cmd(struct se_cmd *cmd, int wait_for_tasks)
+{
 	int ret = 0;
+	bool aborted = false, tas = false;
 
 	if (!(cmd->se_cmd_flags & SCF_SE_LUN_CMD)) {
 		if (wait_for_tasks && (cmd->se_cmd_flags & SCF_SCSI_TMR_CDB))
-			 transport_wait_for_tasks(cmd);
+			target_wait_free_cmd(cmd, &aborted, &tas);
 
-		ret = transport_release_cmd(cmd);
+		if (!aborted || tas)
+			ret = transport_put_cmd(cmd);
 	} else {
 		if (wait_for_tasks)
-			transport_wait_for_tasks(cmd);
+			target_wait_free_cmd(cmd, &aborted, &tas);
 		/*
 		 * Handle WRITE failure case where transport_generic_new_cmd()
 		 * has already added se_cmd to state_list, but fabric has
 		 * failed command before I/O submission.
 		 */
-		if (cmd->state_active) {
-			spin_lock_irqsave(&cmd->t_state_lock, flags);
+		if (cmd->state_active)
 			target_remove_from_state_list(cmd);
-			spin_unlock_irqrestore(&cmd->t_state_lock, flags);
-		}
 
 		if (cmd->se_lun)
 			transport_lun_remove_cmd(cmd);
 
-		ret = transport_put_cmd(cmd);
+		if (!aborted || tas)
+			ret = transport_put_cmd(cmd);
+	}
+	/*
+	 * If the task has been internally aborted due to TMR ABORT_TASK
+	 * or LUN_RESET, target_core_tmr.c is responsible for performing
+	 * the remaining calls to target_put_sess_cmd(), and not the
+	 * callers of this function.
+	 */
+	if (aborted) {
+		pr_debug("Detected CMD_T_ABORTED for ITT: %llu\n", cmd->tag);
+		wait_for_completion(&cmd->cmd_wait_comp);
+		cmd->se_tfo->release_cmd(cmd);
+		ret = 1;
 	}
 	return ret;
 }
@@ -2517,26 +2493,46 @@ out:
 }
 EXPORT_SYMBOL(target_get_sess_cmd);
 
+static void target_free_cmd_mem(struct se_cmd *cmd)
+{
+	transport_free_pages(cmd);
+
+	if (cmd->se_cmd_flags & SCF_SCSI_TMR_CDB)
+		core_tmr_release_req(cmd->se_tmr_req);
+	if (cmd->t_task_cdb != cmd->__t_task_cdb)
+		kfree(cmd->t_task_cdb);
+}
+
 static void target_release_cmd_kref(struct kref *kref)
 {
 	struct se_cmd *se_cmd = container_of(kref, struct se_cmd, cmd_kref);
 	struct se_session *se_sess = se_cmd->se_sess;
 	unsigned long flags;
+	bool fabric_stop;
 
 	spin_lock_irqsave(&se_sess->sess_cmd_lock, flags);
 	if (list_empty(&se_cmd->se_cmd_list)) {
 		spin_unlock_irqrestore(&se_sess->sess_cmd_lock, flags);
+		target_free_cmd_mem(se_cmd);
 		se_cmd->se_tfo->release_cmd(se_cmd);
 		return;
 	}
-	if (se_sess->sess_tearing_down && se_cmd->cmd_wait_set) {
+
+	spin_lock(&se_cmd->t_state_lock);
+	fabric_stop = (se_cmd->transport_state & CMD_T_FABRIC_STOP);
+	spin_unlock(&se_cmd->t_state_lock);
+
+	if (se_cmd->cmd_wait_set || fabric_stop) {
+		list_del_init(&se_cmd->se_cmd_list);
 		spin_unlock_irqrestore(&se_sess->sess_cmd_lock, flags);
+		target_free_cmd_mem(se_cmd);
 		complete(&se_cmd->cmd_wait_comp);
 		return;
 	}
-	list_del(&se_cmd->se_cmd_list);
+	list_del_init(&se_cmd->se_cmd_list);
 	spin_unlock_irqrestore(&se_sess->sess_cmd_lock, flags);
 
+	target_free_cmd_mem(se_cmd);
 	se_cmd->se_tfo->release_cmd(se_cmd);
 }
 
@@ -2548,6 +2544,7 @@ int target_put_sess_cmd(struct se_cmd *se_cmd)
 	struct se_session *se_sess = se_cmd->se_sess;
 
 	if (!se_sess) {
+		target_free_cmd_mem(se_cmd);
 		se_cmd->se_tfo->release_cmd(se_cmd);
 		return 1;
 	}
@@ -2564,6 +2561,7 @@ void target_sess_cmd_list_set_waiting(struct se_session *se_sess)
 {
 	struct se_cmd *se_cmd;
 	unsigned long flags;
+	int rc;
 
 	spin_lock_irqsave(&se_sess->sess_cmd_lock, flags);
 	if (se_sess->sess_tearing_down) {
@@ -2573,8 +2571,15 @@ void target_sess_cmd_list_set_waiting(struct se_session *se_sess)
 	se_sess->sess_tearing_down = 1;
 	list_splice_init(&se_sess->sess_cmd_list, &se_sess->sess_wait_list);
 
-	list_for_each_entry(se_cmd, &se_sess->sess_wait_list, se_cmd_list)
-		se_cmd->cmd_wait_set = 1;
+	list_for_each_entry(se_cmd, &se_sess->sess_wait_list, se_cmd_list) {
+		rc = kref_get_unless_zero(&se_cmd->cmd_kref);
+		if (rc) {
+			se_cmd->cmd_wait_set = 1;
+			spin_lock(&se_cmd->t_state_lock);
+			se_cmd->transport_state |= CMD_T_FABRIC_STOP;
+			spin_unlock(&se_cmd->t_state_lock);
+		}
+	}
 
 	spin_unlock_irqrestore(&se_sess->sess_cmd_lock, flags);
 }
@@ -2587,15 +2592,25 @@ void target_wait_for_sess_cmds(struct se_session *se_sess)
 {
 	struct se_cmd *se_cmd, *tmp_cmd;
 	unsigned long flags;
+	bool tas;
 
 	list_for_each_entry_safe(se_cmd, tmp_cmd,
 				&se_sess->sess_wait_list, se_cmd_list) {
-		list_del(&se_cmd->se_cmd_list);
+		list_del_init(&se_cmd->se_cmd_list);
 
 		pr_debug("Waiting for se_cmd: %p t_state: %d, fabric state:"
 			" %d\n", se_cmd, se_cmd->t_state,
 			se_cmd->se_tfo->get_cmd_state(se_cmd));
 
+		spin_lock_irqsave(&se_cmd->t_state_lock, flags);
+		tas = (se_cmd->transport_state & CMD_T_TAS);
+		spin_unlock_irqrestore(&se_cmd->t_state_lock, flags);
+
+		if (!target_put_sess_cmd(se_cmd)) {
+			if (tas)
+				target_put_sess_cmd(se_cmd);
+		}
+
 		wait_for_completion(&se_cmd->cmd_wait_comp);
 		pr_debug("After cmd_wait_comp: se_cmd: %p t_state: %d"
 			" fabric state: %d\n", se_cmd, se_cmd->t_state,
@@ -2617,53 +2632,75 @@ void transport_clear_lun_ref(struct se_lun *lun)
 	wait_for_completion(&lun->lun_ref_comp);
 }
 
-/**
- * transport_wait_for_tasks - wait for completion to occur
- * @cmd:	command to wait
- *
- * Called from frontend fabric context to wait for storage engine
- * to pause and/or release frontend generated struct se_cmd.
- */
-bool transport_wait_for_tasks(struct se_cmd *cmd)
+static bool
+__transport_wait_for_tasks(struct se_cmd *cmd, bool fabric_stop,
+			   bool *aborted, bool *tas, unsigned long *flags)
+	__releases(&cmd->t_state_lock)
+	__acquires(&cmd->t_state_lock)
 {
-	unsigned long flags;
 
-	spin_lock_irqsave(&cmd->t_state_lock, flags);
+	assert_spin_locked(&cmd->t_state_lock);
+	WARN_ON_ONCE(!irqs_disabled());
+
+	if (fabric_stop)
+		cmd->transport_state |= CMD_T_FABRIC_STOP;
+
+	if (cmd->transport_state & CMD_T_ABORTED)
+		*aborted = true;
+
+	if (cmd->transport_state & CMD_T_TAS)
+		*tas = true;
+
 	if (!(cmd->se_cmd_flags & SCF_SE_LUN_CMD) &&
-	    !(cmd->se_cmd_flags & SCF_SCSI_TMR_CDB)) {
-		spin_unlock_irqrestore(&cmd->t_state_lock, flags);
+	    !(cmd->se_cmd_flags & SCF_SCSI_TMR_CDB))
 		return false;
-	}
 
 	if (!(cmd->se_cmd_flags & SCF_SUPPORTED_SAM_OPCODE) &&
-	    !(cmd->se_cmd_flags & SCF_SCSI_TMR_CDB)) {
-		spin_unlock_irqrestore(&cmd->t_state_lock, flags);
+	    !(cmd->se_cmd_flags & SCF_SCSI_TMR_CDB))
 		return false;
-	}
 
-	if (!(cmd->transport_state & CMD_T_ACTIVE)) {
-		spin_unlock_irqrestore(&cmd->t_state_lock, flags);
+	if (!(cmd->transport_state & CMD_T_ACTIVE))
+		return false;
+
+	if (fabric_stop && *aborted)
 		return false;
-	}
 
 	cmd->transport_state |= CMD_T_STOP;
 
-	pr_debug("wait_for_tasks: Stopping %p ITT: 0x%08llx i_state: %d, t_state: %d, CMD_T_STOP\n",
-		cmd, cmd->tag, cmd->se_tfo->get_cmd_state(cmd), cmd->t_state);
+	pr_debug("wait_for_tasks: Stopping %p ITT: 0x%08llx i_state: %d,"
+		 " t_state: %d, CMD_T_STOP\n", cmd, cmd->tag,
+		 cmd->se_tfo->get_cmd_state(cmd), cmd->t_state);
 
-	spin_unlock_irqrestore(&cmd->t_state_lock, flags);
+	spin_unlock_irqrestore(&cmd->t_state_lock, *flags);
 
 	wait_for_completion(&cmd->t_transport_stop_comp);
 
-	spin_lock_irqsave(&cmd->t_state_lock, flags);
+	spin_lock_irqsave(&cmd->t_state_lock, *flags);
 	cmd->transport_state &= ~(CMD_T_ACTIVE | CMD_T_STOP);
 
-	pr_debug("wait_for_tasks: Stopped wait_for_completion(&cmd->t_transport_stop_comp) for ITT: 0x%08llx\n",
-		cmd->tag);
+	pr_debug("wait_for_tasks: Stopped wait_for_completion(&cmd->"
+		 "t_transport_stop_comp) for ITT: 0x%08llx\n", cmd->tag);
 
+	return true;
+}
+
+/**
+ * transport_wait_for_tasks - wait for completion to occur
+ * @cmd:	command to wait
+ *
+ * Called from frontend fabric context to wait for storage engine
+ * to pause and/or release frontend generated struct se_cmd.
+ */
+bool transport_wait_for_tasks(struct se_cmd *cmd)
+{
+	unsigned long flags;
+	bool ret, aborted = false, tas = false;
+
+	spin_lock_irqsave(&cmd->t_state_lock, flags);
+	ret = __transport_wait_for_tasks(cmd, false, &aborted, &tas, &flags);
 	spin_unlock_irqrestore(&cmd->t_state_lock, flags);
 
-	return true;
+	return ret;
 }
 EXPORT_SYMBOL(transport_wait_for_tasks);
 
@@ -2845,28 +2882,49 @@ transport_send_check_condition_and_sense(struct se_cmd *cmd,
 }
 EXPORT_SYMBOL(transport_send_check_condition_and_sense);
 
-int transport_check_aborted_status(struct se_cmd *cmd, int send_status)
+static int __transport_check_aborted_status(struct se_cmd *cmd, int send_status)
+	__releases(&cmd->t_state_lock)
+	__acquires(&cmd->t_state_lock)
 {
+	assert_spin_locked(&cmd->t_state_lock);
+	WARN_ON_ONCE(!irqs_disabled());
+
 	if (!(cmd->transport_state & CMD_T_ABORTED))
 		return 0;
-
 	/*
 	 * If cmd has been aborted but either no status is to be sent or it has
 	 * already been sent, just return
 	 */
-	if (!send_status || !(cmd->se_cmd_flags & SCF_SEND_DELAYED_TAS))
+	if (!send_status || !(cmd->se_cmd_flags & SCF_SEND_DELAYED_TAS)) {
+		if (send_status)
+			cmd->se_cmd_flags |= SCF_SEND_DELAYED_TAS;
 		return 1;
+	}
 
-	pr_debug("Sending delayed SAM_STAT_TASK_ABORTED status for CDB: 0x%02x ITT: 0x%08llx\n",
-		 cmd->t_task_cdb[0], cmd->tag);
+	pr_debug("Sending delayed SAM_STAT_TASK_ABORTED status for CDB:"
+		" 0x%02x ITT: 0x%08llx\n", cmd->t_task_cdb[0], cmd->tag);
 
 	cmd->se_cmd_flags &= ~SCF_SEND_DELAYED_TAS;
 	cmd->scsi_status = SAM_STAT_TASK_ABORTED;
 	trace_target_cmd_complete(cmd);
+
+	spin_unlock_irq(&cmd->t_state_lock);
 	cmd->se_tfo->queue_status(cmd);
+	spin_lock_irq(&cmd->t_state_lock);
 
 	return 1;
 }
+
+int transport_check_aborted_status(struct se_cmd *cmd, int send_status)
+{
+	int ret;
+
+	spin_lock_irq(&cmd->t_state_lock);
+	ret = __transport_check_aborted_status(cmd, send_status);
+	spin_unlock_irq(&cmd->t_state_lock);
+
+	return ret;
+}
 EXPORT_SYMBOL(transport_check_aborted_status);
 
 void transport_send_task_abort(struct se_cmd *cmd)
@@ -2888,11 +2946,17 @@ void transport_send_task_abort(struct se_cmd *cmd)
 	 */
 	if (cmd->data_direction == DMA_TO_DEVICE) {
 		if (cmd->se_tfo->write_pending_status(cmd) != 0) {
-			cmd->transport_state |= CMD_T_ABORTED;
+			spin_lock_irqsave(&cmd->t_state_lock, flags);
+			if (cmd->se_cmd_flags & SCF_SEND_DELAYED_TAS) {
+				spin_unlock_irqrestore(&cmd->t_state_lock, flags);
+				goto send_abort;
+			}
 			cmd->se_cmd_flags |= SCF_SEND_DELAYED_TAS;
+			spin_unlock_irqrestore(&cmd->t_state_lock, flags);
 			return;
 		}
 	}
+send_abort:
 	cmd->scsi_status = SAM_STAT_TASK_ABORTED;
 
 	transport_lun_remove_cmd(cmd);
@@ -2909,8 +2973,17 @@ static void target_tmr_work(struct work_struct *work)
 	struct se_cmd *cmd = container_of(work, struct se_cmd, work);
 	struct se_device *dev = cmd->se_dev;
 	struct se_tmr_req *tmr = cmd->se_tmr_req;
+	unsigned long flags;
 	int ret;
 
+	spin_lock_irqsave(&cmd->t_state_lock, flags);
+	if (cmd->transport_state & CMD_T_ABORTED) {
+		tmr->response = TMR_FUNCTION_REJECTED;
+		spin_unlock_irqrestore(&cmd->t_state_lock, flags);
+		goto check_stop;
+	}
+	spin_unlock_irqrestore(&cmd->t_state_lock, flags);
+
 	switch (tmr->function) {
 	case TMR_ABORT_TASK:
 		core_tmr_abort_task(dev, tmr, cmd->se_sess);
@@ -2943,9 +3016,17 @@ static void target_tmr_work(struct work_struct *work)
 		break;
 	}
 
+	spin_lock_irqsave(&cmd->t_state_lock, flags);
+	if (cmd->transport_state & CMD_T_ABORTED) {
+		spin_unlock_irqrestore(&cmd->t_state_lock, flags);
+		goto check_stop;
+	}
 	cmd->t_state = TRANSPORT_ISTATE_PROCESSING;
+	spin_unlock_irqrestore(&cmd->t_state_lock, flags);
+
 	cmd->se_tfo->queue_tm_rsp(cmd);
 
+check_stop:
 	transport_cmd_check_stop_to_fabric(cmd);
 }
 
diff --git a/drivers/target/target_core_user.c b/drivers/target/target_core_user.c
index dd600e5ead71..94f5154ac788 100644
--- a/drivers/target/target_core_user.c
+++ b/drivers/target/target_core_user.c
@@ -903,7 +903,7 @@ static int tcmu_configure_device(struct se_device *dev)
 	info->version = __stringify(TCMU_MAILBOX_VERSION);
 
 	info->mem[0].name = "tcm-user command & data buffer";
-	info->mem[0].addr = (phys_addr_t) udev->mb_addr;
+	info->mem[0].addr = (phys_addr_t)(uintptr_t)udev->mb_addr;
 	info->mem[0].size = TCMU_RING_SIZE;
 	info->mem[0].memtype = UIO_MEM_VIRTUAL;
 
diff --git a/drivers/thermal/Kconfig b/drivers/thermal/Kconfig
index 8cc4ac64a91c..7c92c09be213 100644
--- a/drivers/thermal/Kconfig
+++ b/drivers/thermal/Kconfig
@@ -195,7 +195,7 @@ config IMX_THERMAL
 	  passive trip is crossed.
 
 config SPEAR_THERMAL
-	bool "SPEAr thermal sensor driver"
+	tristate "SPEAr thermal sensor driver"
 	depends on PLAT_SPEAR || COMPILE_TEST
 	depends on OF
 	help
@@ -237,8 +237,8 @@ config DOVE_THERMAL
 	  framework.
 
 config DB8500_THERMAL
-	bool "DB8500 thermal management"
-	depends on ARCH_U8500
+	tristate "DB8500 thermal management"
+	depends on MFD_DB8500_PRCMU
 	default y
 	help
 	  Adds DB8500 thermal management implementation according to the thermal
diff --git a/drivers/thermal/cpu_cooling.c b/drivers/thermal/cpu_cooling.c
index e3fbc5a5d88f..6ceac4f2d4b2 100644
--- a/drivers/thermal/cpu_cooling.c
+++ b/drivers/thermal/cpu_cooling.c
@@ -377,26 +377,28 @@ static u32 cpu_power_to_freq(struct cpufreq_cooling_device *cpufreq_device,
  * get_load() - get load for a cpu since last updated
  * @cpufreq_device:	&struct cpufreq_cooling_device for this cpu
  * @cpu:	cpu number
+ * @cpu_idx:	index of the cpu in cpufreq_device->allowed_cpus
  *
  * Return: The average load of cpu @cpu in percentage since this
  * function was last called.
  */
-static u32 get_load(struct cpufreq_cooling_device *cpufreq_device, int cpu)
+static u32 get_load(struct cpufreq_cooling_device *cpufreq_device, int cpu,
+		    int cpu_idx)
 {
 	u32 load;
 	u64 now, now_idle, delta_time, delta_idle;
 
 	now_idle = get_cpu_idle_time(cpu, &now, 0);
-	delta_idle = now_idle - cpufreq_device->time_in_idle[cpu];
-	delta_time = now - cpufreq_device->time_in_idle_timestamp[cpu];
+	delta_idle = now_idle - cpufreq_device->time_in_idle[cpu_idx];
+	delta_time = now - cpufreq_device->time_in_idle_timestamp[cpu_idx];
 
 	if (delta_time <= delta_idle)
 		load = 0;
 	else
 		load = div64_u64(100 * (delta_time - delta_idle), delta_time);
 
-	cpufreq_device->time_in_idle[cpu] = now_idle;
-	cpufreq_device->time_in_idle_timestamp[cpu] = now;
+	cpufreq_device->time_in_idle[cpu_idx] = now_idle;
+	cpufreq_device->time_in_idle_timestamp[cpu_idx] = now;
 
 	return load;
 }
@@ -598,7 +600,7 @@ static int cpufreq_get_requested_power(struct thermal_cooling_device *cdev,
 		u32 load;
 
 		if (cpu_online(cpu))
-			load = get_load(cpufreq_device, cpu);
+			load = get_load(cpufreq_device, cpu, i);
 		else
 			load = 0;
 
diff --git a/drivers/thermal/of-thermal.c b/drivers/thermal/of-thermal.c
index be4eedcb839a..9043f8f91852 100644
--- a/drivers/thermal/of-thermal.c
+++ b/drivers/thermal/of-thermal.c
@@ -475,14 +475,10 @@ thermal_zone_of_sensor_register(struct device *dev, int sensor_id, void *data,
 
 	sensor_np = of_node_get(dev->of_node);
 
-	for_each_child_of_node(np, child) {
+	for_each_available_child_of_node(np, child) {
 		struct of_phandle_args sensor_specs;
 		int ret, id;
 
-		/* Check whether child is enabled or not */
-		if (!of_device_is_available(child))
-			continue;
-
 		/* For now, thermal framework supports only 1 sensor per zone */
 		ret = of_parse_phandle_with_args(child, "thermal-sensors",
 						 "#thermal-sensor-cells",
@@ -881,16 +877,12 @@ int __init of_parse_thermal_zones(void)
 		return 0; /* Run successfully on systems without thermal DT */
 	}
 
-	for_each_child_of_node(np, child) {
+	for_each_available_child_of_node(np, child) {
 		struct thermal_zone_device *zone;
 		struct thermal_zone_params *tzp;
 		int i, mask = 0;
 		u32 prop;
 
-		/* Check whether child is enabled or not */
-		if (!of_device_is_available(child))
-			continue;
-
 		tz = thermal_of_build_thermal_zone(child);
 		if (IS_ERR(tz)) {
 			pr_err("failed to build thermal zone %s: %ld\n",
@@ -968,13 +960,9 @@ void of_thermal_destroy_zones(void)
 		return;
 	}
 
-	for_each_child_of_node(np, child) {
+	for_each_available_child_of_node(np, child) {
 		struct thermal_zone_device *zone;
 
-		/* Check whether child is enabled or not */
-		if (!of_device_is_available(child))
-			continue;
-
 		zone = thermal_zone_get_zone_by_name(child->name);
 		if (IS_ERR(zone))
 			continue;
diff --git a/drivers/thermal/rcar_thermal.c b/drivers/thermal/rcar_thermal.c
index 44b9c485157d..0e735acea33a 100644
--- a/drivers/thermal/rcar_thermal.c
+++ b/drivers/thermal/rcar_thermal.c
@@ -23,6 +23,7 @@
 #include <linux/interrupt.h>
 #include <linux/io.h>
 #include <linux/module.h>
+#include <linux/of_device.h>
 #include <linux/platform_device.h>
 #include <linux/pm_runtime.h>
 #include <linux/reboot.h>
@@ -75,8 +76,10 @@ struct rcar_thermal_priv {
 #define rcar_has_irq_support(priv)	((priv)->common->base)
 #define rcar_id_to_shift(priv)		((priv)->id * 8)
 
+#define USE_OF_THERMAL	1
 static const struct of_device_id rcar_thermal_dt_ids[] = {
 	{ .compatible = "renesas,rcar-thermal", },
+	{ .compatible = "renesas,rcar-gen2-thermal", .data = (void *)USE_OF_THERMAL },
 	{},
 };
 MODULE_DEVICE_TABLE(of, rcar_thermal_dt_ids);
@@ -200,9 +203,9 @@ err_out_unlock:
 	return ret;
 }
 
-static int rcar_thermal_get_temp(struct thermal_zone_device *zone, int *temp)
+static int rcar_thermal_get_current_temp(struct rcar_thermal_priv *priv,
+					 int *temp)
 {
-	struct rcar_thermal_priv *priv = rcar_zone_to_priv(zone);
 	int tmp;
 	int ret;
 
@@ -226,6 +229,20 @@ static int rcar_thermal_get_temp(struct thermal_zone_device *zone, int *temp)
 	return 0;
 }
 
+static int rcar_thermal_of_get_temp(void *data, int *temp)
+{
+	struct rcar_thermal_priv *priv = data;
+
+	return rcar_thermal_get_current_temp(priv, temp);
+}
+
+static int rcar_thermal_get_temp(struct thermal_zone_device *zone, int *temp)
+{
+	struct rcar_thermal_priv *priv = rcar_zone_to_priv(zone);
+
+	return rcar_thermal_get_current_temp(priv, temp);
+}
+
 static int rcar_thermal_get_trip_type(struct thermal_zone_device *zone,
 				      int trip, enum thermal_trip_type *type)
 {
@@ -282,6 +299,10 @@ static int rcar_thermal_notify(struct thermal_zone_device *zone,
 	return 0;
 }
 
+static const struct thermal_zone_of_device_ops rcar_thermal_zone_of_ops = {
+	.get_temp	= rcar_thermal_of_get_temp,
+};
+
 static struct thermal_zone_device_ops rcar_thermal_zone_ops = {
 	.get_temp	= rcar_thermal_get_temp,
 	.get_trip_type	= rcar_thermal_get_trip_type,
@@ -318,14 +339,20 @@ static void rcar_thermal_work(struct work_struct *work)
 
 	priv = container_of(work, struct rcar_thermal_priv, work.work);
 
-	rcar_thermal_get_temp(priv->zone, &cctemp);
+	ret = rcar_thermal_get_current_temp(priv, &cctemp);
+	if (ret < 0)
+		return;
+
 	ret = rcar_thermal_update_temp(priv);
 	if (ret < 0)
 		return;
 
 	rcar_thermal_irq_enable(priv);
 
-	rcar_thermal_get_temp(priv->zone, &nctemp);
+	ret = rcar_thermal_get_current_temp(priv, &nctemp);
+	if (ret < 0)
+		return;
+
 	if (nctemp != cctemp)
 		thermal_zone_device_update(priv->zone);
 }
@@ -403,6 +430,8 @@ static int rcar_thermal_probe(struct platform_device *pdev)
 	struct rcar_thermal_priv *priv;
 	struct device *dev = &pdev->dev;
 	struct resource *res, *irq;
+	const struct of_device_id *of_id = of_match_device(rcar_thermal_dt_ids, dev);
+	unsigned long of_data = (unsigned long)of_id->data;
 	int mres = 0;
 	int i;
 	int ret = -ENODEV;
@@ -463,7 +492,13 @@ static int rcar_thermal_probe(struct platform_device *pdev)
 		if (ret < 0)
 			goto error_unregister;
 
-		priv->zone = thermal_zone_device_register("rcar_thermal",
+		if (of_data == USE_OF_THERMAL)
+			priv->zone = thermal_zone_of_sensor_register(
+						dev, i, priv,
+						&rcar_thermal_zone_of_ops);
+		else
+			priv->zone = thermal_zone_device_register(
+						"rcar_thermal",
 						1, 0, priv,
 						&rcar_thermal_zone_ops, NULL, 0,
 						idle);
diff --git a/drivers/thermal/spear_thermal.c b/drivers/thermal/spear_thermal.c
index 534dd9136662..81b35aace9de 100644
--- a/drivers/thermal/spear_thermal.c
+++ b/drivers/thermal/spear_thermal.c
@@ -54,8 +54,7 @@ static struct thermal_zone_device_ops ops = {
 	.get_temp = thermal_get_temp,
 };
 
-#ifdef CONFIG_PM
-static int spear_thermal_suspend(struct device *dev)
+static int __maybe_unused spear_thermal_suspend(struct device *dev)
 {
 	struct platform_device *pdev = to_platform_device(dev);
 	struct thermal_zone_device *spear_thermal = platform_get_drvdata(pdev);
@@ -72,7 +71,7 @@ static int spear_thermal_suspend(struct device *dev)
 	return 0;
 }
 
-static int spear_thermal_resume(struct device *dev)
+static int __maybe_unused spear_thermal_resume(struct device *dev)
 {
 	struct platform_device *pdev = to_platform_device(dev);
 	struct thermal_zone_device *spear_thermal = platform_get_drvdata(pdev);
@@ -94,7 +93,6 @@ static int spear_thermal_resume(struct device *dev)
 
 	return 0;
 }
-#endif
 
 static SIMPLE_DEV_PM_OPS(spear_thermal_pm_ops, spear_thermal_suspend,
 		spear_thermal_resume);
diff --git a/drivers/tty/pty.c b/drivers/tty/pty.c
index b3110040164a..2348fa613707 100644
--- a/drivers/tty/pty.c
+++ b/drivers/tty/pty.c
@@ -681,7 +681,14 @@ static void pty_unix98_remove(struct tty_driver *driver, struct tty_struct *tty)
 /* this is called once with whichever end is closed last */
 static void pty_unix98_shutdown(struct tty_struct *tty)
 {
-	devpts_kill_index(tty->driver_data, tty->index);
+	struct inode *ptmx_inode;
+
+	if (tty->driver->subtype == PTY_TYPE_MASTER)
+		ptmx_inode = tty->driver_data;
+	else
+		ptmx_inode = tty->link->driver_data;
+	devpts_kill_index(ptmx_inode, tty->index);
+	devpts_del_ref(ptmx_inode);
 }
 
 static const struct tty_operations ptm_unix98_ops = {
@@ -773,6 +780,18 @@ static int ptmx_open(struct inode *inode, struct file *filp)
 	set_bit(TTY_PTY_LOCK, &tty->flags); /* LOCK THE SLAVE */
 	tty->driver_data = inode;
 
+	/*
+	 * In the case where all references to ptmx inode are dropped and we
+	 * still have /dev/tty opened pointing to the master/slave pair (ptmx
+	 * is closed/released before /dev/tty), we must make sure that the inode
+	 * is still valid when we call the final pty_unix98_shutdown, thus we
+	 * hold an additional reference to the ptmx inode. For the same /dev/tty
+	 * last close case, we also need to make sure the super_block isn't
+	 * destroyed (devpts instance unmounted), before /dev/tty is closed and
+	 * on its release devpts_kill_index is called.
+	 */
+	devpts_add_ref(inode);
+
 	tty_add_file(tty, filp);
 
 	slave_inode = devpts_pty_new(inode,
diff --git a/drivers/tty/serial/8250/8250_pci.c b/drivers/tty/serial/8250/8250_pci.c
index e71ec78fc11e..7cd6f9a90542 100644
--- a/drivers/tty/serial/8250/8250_pci.c
+++ b/drivers/tty/serial/8250/8250_pci.c
@@ -1941,6 +1941,7 @@ pci_wch_ch38x_setup(struct serial_private *priv,
 #define PCIE_VENDOR_ID_WCH		0x1c00
 #define PCIE_DEVICE_ID_WCH_CH382_2S1P	0x3250
 #define PCIE_DEVICE_ID_WCH_CH384_4S	0x3470
+#define PCIE_DEVICE_ID_WCH_CH382_2S	0x3253
 
 #define PCI_VENDOR_ID_PERICOM			0x12D8
 #define PCI_DEVICE_ID_PERICOM_PI7C9X7951	0x7951
@@ -2637,6 +2638,14 @@ static struct pci_serial_quirk pci_serial_quirks[] __refdata = {
 		.subdevice	= PCI_ANY_ID,
 		.setup		= pci_wch_ch353_setup,
 	},
+	/* WCH CH382 2S card (16850 clone) */
+	{
+		.vendor         = PCIE_VENDOR_ID_WCH,
+		.device         = PCIE_DEVICE_ID_WCH_CH382_2S,
+		.subvendor      = PCI_ANY_ID,
+		.subdevice      = PCI_ANY_ID,
+		.setup          = pci_wch_ch38x_setup,
+	},
 	/* WCH CH382 2S1P card (16850 clone) */
 	{
 		.vendor         = PCIE_VENDOR_ID_WCH,
@@ -2955,6 +2964,7 @@ enum pci_board_num_t {
 	pbn_fintek_4,
 	pbn_fintek_8,
 	pbn_fintek_12,
+	pbn_wch382_2,
 	pbn_wch384_4,
 	pbn_pericom_PI7C9X7951,
 	pbn_pericom_PI7C9X7952,
@@ -3775,6 +3785,13 @@ static struct pciserial_board pci_boards[] = {
 		.base_baud	= 115200,
 		.first_offset	= 0x40,
 	},
+	[pbn_wch382_2] = {
+		.flags		= FL_BASE0,
+		.num_ports	= 2,
+		.base_baud	= 115200,
+		.uart_offset	= 8,
+		.first_offset	= 0xC0,
+	},
 	[pbn_wch384_4] = {
 		.flags		= FL_BASE0,
 		.num_ports	= 4,
@@ -5574,6 +5591,10 @@ static struct pci_device_id serial_pci_tbl[] = {
 		PCI_ANY_ID, PCI_ANY_ID,
 		0, 0, pbn_b0_bt_2_115200 },
 
+	{	PCIE_VENDOR_ID_WCH, PCIE_DEVICE_ID_WCH_CH382_2S,
+		PCI_ANY_ID, PCI_ANY_ID,
+		0, 0, pbn_wch382_2 },
+
 	{	PCIE_VENDOR_ID_WCH, PCIE_DEVICE_ID_WCH_CH384_4S,
 		PCI_ANY_ID, PCI_ANY_ID,
 		0, 0, pbn_wch384_4 },
diff --git a/drivers/tty/serial/omap-serial.c b/drivers/tty/serial/omap-serial.c
index b645f9228ed7..fa49eb1e2fa2 100644
--- a/drivers/tty/serial/omap-serial.c
+++ b/drivers/tty/serial/omap-serial.c
@@ -1165,7 +1165,7 @@ serial_omap_type(struct uart_port *port)
 
 #define BOTH_EMPTY (UART_LSR_TEMT | UART_LSR_THRE)
 
-static void wait_for_xmitr(struct uart_omap_port *up)
+static void __maybe_unused wait_for_xmitr(struct uart_omap_port *up)
 {
 	unsigned int status, tmout = 10000;
 
@@ -1343,7 +1343,7 @@ static inline void serial_omap_add_console_port(struct uart_omap_port *up)
 
 /* Enable or disable the rs485 support */
 static int
-serial_omap_config_rs485(struct uart_port *port, struct serial_rs485 *rs485conf)
+serial_omap_config_rs485(struct uart_port *port, struct serial_rs485 *rs485)
 {
 	struct uart_omap_port *up = to_uart_omap_port(port);
 	unsigned int mode;
@@ -1356,8 +1356,12 @@ serial_omap_config_rs485(struct uart_port *port, struct serial_rs485 *rs485conf)
 	up->ier = 0;
 	serial_out(up, UART_IER, 0);
 
+	/* Clamp the delays to [0, 100ms] */
+	rs485->delay_rts_before_send = min(rs485->delay_rts_before_send, 100U);
+	rs485->delay_rts_after_send  = min(rs485->delay_rts_after_send, 100U);
+
 	/* store new config */
-	port->rs485 = *rs485conf;
+	port->rs485 = *rs485;
 
 	/*
 	 * Just as a precaution, only allow rs485
diff --git a/drivers/tty/tty_io.c b/drivers/tty/tty_io.c
index 5cec01c75691..a7eacef1bd22 100644
--- a/drivers/tty/tty_io.c
+++ b/drivers/tty/tty_io.c
@@ -2066,13 +2066,12 @@ retry_open:
 		if (tty) {
 			mutex_unlock(&tty_mutex);
 			retval = tty_lock_interruptible(tty);
+			tty_kref_put(tty);  /* drop kref from tty_driver_lookup_tty() */
 			if (retval) {
 				if (retval == -EINTR)
 					retval = -ERESTARTSYS;
 				goto err_unref;
 			}
-			/* safe to drop the kref from tty_driver_lookup_tty() */
-			tty_kref_put(tty);
 			retval = tty_reopen(tty);
 			if (retval < 0) {
 				tty_unlock(tty);
diff --git a/drivers/tty/tty_mutex.c b/drivers/tty/tty_mutex.c
index d2f3c4cd697f..dfa9ec03fa8e 100644
--- a/drivers/tty/tty_mutex.c
+++ b/drivers/tty/tty_mutex.c
@@ -21,10 +21,15 @@ EXPORT_SYMBOL(tty_lock);
 
 int tty_lock_interruptible(struct tty_struct *tty)
 {
+	int ret;
+
 	if (WARN(tty->magic != TTY_MAGIC, "L Bad %p\n", tty))
 		return -EIO;
 	tty_kref_get(tty);
-	return mutex_lock_interruptible(&tty->legacy_mutex);
+	ret = mutex_lock_interruptible(&tty->legacy_mutex);
+	if (ret)
+		tty_kref_put(tty);
+	return ret;
 }
 
 void __lockfunc tty_unlock(struct tty_struct *tty)
diff --git a/drivers/usb/chipidea/ci_hdrc_pci.c b/drivers/usb/chipidea/ci_hdrc_pci.c
index b59195edf636..b635ab67490d 100644
--- a/drivers/usb/chipidea/ci_hdrc_pci.c
+++ b/drivers/usb/chipidea/ci_hdrc_pci.c
@@ -85,8 +85,8 @@ static int ci_hdrc_pci_probe(struct pci_dev *pdev,
 
 	/* register a nop PHY */
 	ci->phy = usb_phy_generic_register();
-	if (!ci->phy)
-		return -ENOMEM;
+	if (IS_ERR(ci->phy))
+		return PTR_ERR(ci->phy);
 
 	memset(res, 0, sizeof(res));
 	res[0].start	= pci_resource_start(pdev, 0);
diff --git a/drivers/usb/chipidea/debug.c b/drivers/usb/chipidea/debug.c
index a4f7db2e18dd..df47110bad2d 100644
--- a/drivers/usb/chipidea/debug.c
+++ b/drivers/usb/chipidea/debug.c
@@ -100,6 +100,9 @@ static ssize_t ci_port_test_write(struct file *file, const char __user *ubuf,
 	if (sscanf(buf, "%u", &mode) != 1)
 		return -EINVAL;
 
+	if (mode > 255)
+		return -EBADRQC;
+
 	pm_runtime_get_sync(ci->dev);
 	spin_lock_irqsave(&ci->lock, flags);
 	ret = hw_port_test_set(ci, mode);
diff --git a/drivers/usb/chipidea/otg.c b/drivers/usb/chipidea/otg.c
index 45f86da1d6d3..03b6743461d1 100644
--- a/drivers/usb/chipidea/otg.c
+++ b/drivers/usb/chipidea/otg.c
@@ -158,7 +158,7 @@ static void ci_otg_work(struct work_struct *work)
 int ci_hdrc_otg_init(struct ci_hdrc *ci)
 {
 	INIT_WORK(&ci->work, ci_otg_work);
-	ci->wq = create_singlethread_workqueue("ci_otg");
+	ci->wq = create_freezable_workqueue("ci_otg");
 	if (!ci->wq) {
 		dev_err(ci->dev, "can't create workqueue\n");
 		return -ENODEV;
diff --git a/drivers/usb/core/hub.c b/drivers/usb/core/hub.c
index 350dcd9af5d8..51b436918f78 100644
--- a/drivers/usb/core/hub.c
+++ b/drivers/usb/core/hub.c
@@ -5401,6 +5401,7 @@ static int usb_reset_and_verify_device(struct usb_device *udev)
 	}
 
 	bos = udev->bos;
+	udev->bos = NULL;
 
 	for (i = 0; i < SET_CONFIG_TRIES; ++i) {
 
@@ -5493,11 +5494,8 @@ done:
 	usb_set_usb2_hardware_lpm(udev, 1);
 	usb_unlocked_enable_lpm(udev);
 	usb_enable_ltm(udev);
-	/* release the new BOS descriptor allocated  by hub_port_init() */
-	if (udev->bos != bos) {
-		usb_release_bos_descriptor(udev);
-		udev->bos = bos;
-	}
+	usb_release_bos_descriptor(udev);
+	udev->bos = bos;
 	return 0;
 
 re_enumerate:
diff --git a/drivers/usb/dwc2/Kconfig b/drivers/usb/dwc2/Kconfig
index fd95ba6ec317..f0decc0d69b5 100644
--- a/drivers/usb/dwc2/Kconfig
+++ b/drivers/usb/dwc2/Kconfig
@@ -1,5 +1,6 @@
 config USB_DWC2
 	tristate "DesignWare USB2 DRD Core Support"
+	depends on HAS_DMA
 	depends on USB || USB_GADGET
 	help
 	  Say Y here if your system has a Dual Role Hi-Speed USB
diff --git a/drivers/usb/dwc2/core.c b/drivers/usb/dwc2/core.c
index e991d55914db..46c4ba75dc2a 100644
--- a/drivers/usb/dwc2/core.c
+++ b/drivers/usb/dwc2/core.c
@@ -619,6 +619,12 @@ void dwc2_force_dr_mode(struct dwc2_hsotg *hsotg)
 			 __func__, hsotg->dr_mode);
 		break;
 	}
+
+	/*
+	 * NOTE: This is required for some rockchip soc based
+	 * platforms.
+	 */
+	msleep(50);
 }
 
 /*
diff --git a/drivers/usb/dwc2/hcd_ddma.c b/drivers/usb/dwc2/hcd_ddma.c
index 36606fc33c0d..a41274aa52ad 100644
--- a/drivers/usb/dwc2/hcd_ddma.c
+++ b/drivers/usb/dwc2/hcd_ddma.c
@@ -1174,14 +1174,11 @@ static int dwc2_process_non_isoc_desc(struct dwc2_hsotg *hsotg,
 	failed = dwc2_update_non_isoc_urb_state_ddma(hsotg, chan, qtd, dma_desc,
 						     halt_status, n_bytes,
 						     xfer_done);
-	if (*xfer_done && urb->status != -EINPROGRESS)
-		failed = 1;
-
-	if (failed) {
+	if (failed || (*xfer_done && urb->status != -EINPROGRESS)) {
 		dwc2_host_complete(hsotg, qtd, urb->status);
 		dwc2_hcd_qtd_unlink_and_free(hsotg, qtd, qh);
-		dev_vdbg(hsotg->dev, "failed=%1x xfer_done=%1x status=%08x\n",
-			 failed, *xfer_done, urb->status);
+		dev_vdbg(hsotg->dev, "failed=%1x xfer_done=%1x\n",
+			 failed, *xfer_done);
 		return failed;
 	}
 
@@ -1236,21 +1233,23 @@ static void dwc2_complete_non_isoc_xfer_ddma(struct dwc2_hsotg *hsotg,
 
 	list_for_each_safe(qtd_item, qtd_tmp, &qh->qtd_list) {
 		int i;
+		int qtd_desc_count;
 
 		qtd = list_entry(qtd_item, struct dwc2_qtd, qtd_list_entry);
 		xfer_done = 0;
+		qtd_desc_count = qtd->n_desc;
 
-		for (i = 0; i < qtd->n_desc; i++) {
+		for (i = 0; i < qtd_desc_count; i++) {
 			if (dwc2_process_non_isoc_desc(hsotg, chan, chnum, qtd,
 						       desc_num, halt_status,
-						       &xfer_done)) {
-				qtd = NULL;
-				break;
-			}
+						       &xfer_done))
+				goto stop_scan;
+
 			desc_num++;
 		}
 	}
 
+stop_scan:
 	if (qh->ep_type != USB_ENDPOINT_XFER_CONTROL) {
 		/*
 		 * Resetting the data toggle for bulk and interrupt endpoints
@@ -1258,7 +1257,7 @@ static void dwc2_complete_non_isoc_xfer_ddma(struct dwc2_hsotg *hsotg,
 		 */
 		if (halt_status == DWC2_HC_XFER_STALL)
 			qh->data_toggle = DWC2_HC_PID_DATA0;
-		else if (qtd)
+		else
 			dwc2_hcd_save_data_toggle(hsotg, chan, chnum, qtd);
 	}
 
diff --git a/drivers/usb/dwc2/hcd_intr.c b/drivers/usb/dwc2/hcd_intr.c
index f8253803a050..cadba8b13c48 100644
--- a/drivers/usb/dwc2/hcd_intr.c
+++ b/drivers/usb/dwc2/hcd_intr.c
@@ -525,11 +525,19 @@ void dwc2_hcd_save_data_toggle(struct dwc2_hsotg *hsotg,
 	u32 pid = (hctsiz & TSIZ_SC_MC_PID_MASK) >> TSIZ_SC_MC_PID_SHIFT;
 
 	if (chan->ep_type != USB_ENDPOINT_XFER_CONTROL) {
+		if (WARN(!chan || !chan->qh,
+			 "chan->qh must be specified for non-control eps\n"))
+			return;
+
 		if (pid == TSIZ_SC_MC_PID_DATA0)
 			chan->qh->data_toggle = DWC2_HC_PID_DATA0;
 		else
 			chan->qh->data_toggle = DWC2_HC_PID_DATA1;
 	} else {
+		if (WARN(!qtd,
+			 "qtd must be specified for control eps\n"))
+			return;
+
 		if (pid == TSIZ_SC_MC_PID_DATA0)
 			qtd->data_toggle = DWC2_HC_PID_DATA0;
 		else
diff --git a/drivers/usb/dwc3/core.h b/drivers/usb/dwc3/core.h
index 29130682e547..e4f8b90d9627 100644
--- a/drivers/usb/dwc3/core.h
+++ b/drivers/usb/dwc3/core.h
@@ -856,7 +856,6 @@ struct dwc3 {
 	unsigned		pullups_connected:1;
 	unsigned		resize_fifos:1;
 	unsigned		setup_packet_pending:1;
-	unsigned		start_config_issued:1;
 	unsigned		three_stage_setup:1;
 	unsigned		usb3_lpm_capable:1;
 
diff --git a/drivers/usb/dwc3/ep0.c b/drivers/usb/dwc3/ep0.c
index 3a9354abcb68..8d6b75c2f53b 100644
--- a/drivers/usb/dwc3/ep0.c
+++ b/drivers/usb/dwc3/ep0.c
@@ -555,7 +555,6 @@ static int dwc3_ep0_set_config(struct dwc3 *dwc, struct usb_ctrlrequest *ctrl)
 	int ret;
 	u32 reg;
 
-	dwc->start_config_issued = false;
 	cfg = le16_to_cpu(ctrl->wValue);
 
 	switch (state) {
@@ -737,10 +736,6 @@ static int dwc3_ep0_std_request(struct dwc3 *dwc, struct usb_ctrlrequest *ctrl)
 		dwc3_trace(trace_dwc3_ep0, "USB_REQ_SET_ISOCH_DELAY");
 		ret = dwc3_ep0_set_isoch_delay(dwc, ctrl);
 		break;
-	case USB_REQ_SET_INTERFACE:
-		dwc3_trace(trace_dwc3_ep0, "USB_REQ_SET_INTERFACE");
-		dwc->start_config_issued = false;
-		/* Fall through */
 	default:
 		dwc3_trace(trace_dwc3_ep0, "Forwarding to gadget driver");
 		ret = dwc3_ep0_delegate_req(dwc, ctrl);
diff --git a/drivers/usb/dwc3/gadget.c b/drivers/usb/dwc3/gadget.c
index 7d1dd82a95ac..2363bad45af8 100644
--- a/drivers/usb/dwc3/gadget.c
+++ b/drivers/usb/dwc3/gadget.c
@@ -385,24 +385,66 @@ static void dwc3_free_trb_pool(struct dwc3_ep *dep)
 	dep->trb_pool_dma = 0;
 }
 
+static int dwc3_gadget_set_xfer_resource(struct dwc3 *dwc, struct dwc3_ep *dep);
+
+/**
+ * dwc3_gadget_start_config - Configure EP resources
+ * @dwc: pointer to our controller context structure
+ * @dep: endpoint that is being enabled
+ *
+ * The assignment of transfer resources cannot perfectly follow the
+ * data book due to the fact that the controller driver does not have
+ * all knowledge of the configuration in advance. It is given this
+ * information piecemeal by the composite gadget framework after every
+ * SET_CONFIGURATION and SET_INTERFACE. Trying to follow the databook
+ * programming model in this scenario can cause errors. For two
+ * reasons:
+ *
+ * 1) The databook says to do DEPSTARTCFG for every SET_CONFIGURATION
+ * and SET_INTERFACE (8.1.5). This is incorrect in the scenario of
+ * multiple interfaces.
+ *
+ * 2) The databook does not mention doing more DEPXFERCFG for new
+ * endpoint on alt setting (8.1.6).
+ *
+ * The following simplified method is used instead:
+ *
+ * All hardware endpoints can be assigned a transfer resource and this
+ * setting will stay persistent until either a core reset or
+ * hibernation. So whenever we do a DEPSTARTCFG(0) we can go ahead and
+ * do DEPXFERCFG for every hardware endpoint as well. We are
+ * guaranteed that there are as many transfer resources as endpoints.
+ *
+ * This function is called for each endpoint when it is being enabled
+ * but is triggered only when called for EP0-out, which always happens
+ * first, and which should only happen in one of the above conditions.
+ */
 static int dwc3_gadget_start_config(struct dwc3 *dwc, struct dwc3_ep *dep)
 {
 	struct dwc3_gadget_ep_cmd_params params;
 	u32			cmd;
+	int			i;
+	int			ret;
+
+	if (dep->number)
+		return 0;
 
 	memset(&params, 0x00, sizeof(params));
+	cmd = DWC3_DEPCMD_DEPSTARTCFG;
 
-	if (dep->number != 1) {
-		cmd = DWC3_DEPCMD_DEPSTARTCFG;
-		/* XferRscIdx == 0 for ep0 and 2 for the remaining */
-		if (dep->number > 1) {
-			if (dwc->start_config_issued)
-				return 0;
-			dwc->start_config_issued = true;
-			cmd |= DWC3_DEPCMD_PARAM(2);
-		}
+	ret = dwc3_send_gadget_ep_cmd(dwc, 0, cmd, &params);
+	if (ret)
+		return ret;
 
-		return dwc3_send_gadget_ep_cmd(dwc, 0, cmd, &params);
+	for (i = 0; i < DWC3_ENDPOINTS_NUM; i++) {
+		struct dwc3_ep *dep = dwc->eps[i];
+
+		if (!dep)
+			continue;
+
+		ret = dwc3_gadget_set_xfer_resource(dwc, dep);
+		if (ret)
+			return ret;
 	}
 
 	return 0;
@@ -516,10 +558,6 @@ static int __dwc3_gadget_ep_enable(struct dwc3_ep *dep,
 		struct dwc3_trb	*trb_st_hw;
 		struct dwc3_trb	*trb_link;
 
-		ret = dwc3_gadget_set_xfer_resource(dwc, dep);
-		if (ret)
-			return ret;
-
 		dep->endpoint.desc = desc;
 		dep->comp_desc = comp_desc;
 		dep->type = usb_endpoint_type(desc);
@@ -1636,8 +1674,6 @@ static int dwc3_gadget_start(struct usb_gadget *g,
 	}
 	dwc3_writel(dwc->regs, DWC3_DCFG, reg);
 
-	dwc->start_config_issued = false;
-
 	/* Start with SuperSpeed Default */
 	dwc3_gadget_ep0_desc.wMaxPacketSize = cpu_to_le16(512);
 
@@ -2237,7 +2273,6 @@ static void dwc3_gadget_disconnect_interrupt(struct dwc3 *dwc)
 	dwc3_writel(dwc->regs, DWC3_DCTL, reg);
 
 	dwc3_disconnect_gadget(dwc);
-	dwc->start_config_issued = false;
 
 	dwc->gadget.speed = USB_SPEED_UNKNOWN;
 	dwc->setup_packet_pending = false;
@@ -2288,7 +2323,6 @@ static void dwc3_gadget_reset_interrupt(struct dwc3 *dwc)
 
 	dwc3_stop_active_transfers(dwc);
 	dwc3_clear_stall_all_ep(dwc);
-	dwc->start_config_issued = false;
 
 	/* Reset device address to zero */
 	reg = dwc3_readl(dwc->regs, DWC3_DCFG);
diff --git a/drivers/usb/gadget/legacy/inode.c b/drivers/usb/gadget/legacy/inode.c
index 7e179f81d05c..87fb0fd6aaab 100644
--- a/drivers/usb/gadget/legacy/inode.c
+++ b/drivers/usb/gadget/legacy/inode.c
@@ -130,7 +130,8 @@ struct dev_data {
 					setup_can_stall : 1,
 					setup_out_ready : 1,
 					setup_out_error : 1,
-					setup_abort : 1;
+					setup_abort : 1,
+					gadget_registered : 1;
 	unsigned			setup_wLength;
 
 	/* the rest is basically write-once */
@@ -1179,7 +1180,8 @@ dev_release (struct inode *inode, struct file *fd)
 
 	/* closing ep0 === shutdown all */
 
-	usb_gadget_unregister_driver (&gadgetfs_driver);
+	if (dev->gadget_registered)
+		usb_gadget_unregister_driver (&gadgetfs_driver);
 
 	/* at this point "good" hardware has disconnected the
 	 * device from USB; the host won't see it any more.
@@ -1847,6 +1849,7 @@ dev_config (struct file *fd, const char __user *buf, size_t len, loff_t *ptr)
 		 * kick in after the ep0 descriptor is closed.
 		 */
 		value = len;
+		dev->gadget_registered = true;
 	}
 	return value;
 
diff --git a/drivers/usb/gadget/udc/fsl_qe_udc.c b/drivers/usb/gadget/udc/fsl_qe_udc.c
index 53c0692f1b09..93d28cb00b76 100644
--- a/drivers/usb/gadget/udc/fsl_qe_udc.c
+++ b/drivers/usb/gadget/udc/fsl_qe_udc.c
@@ -2340,7 +2340,7 @@ static struct qe_udc *qe_udc_config(struct platform_device *ofdev)
 {
 	struct qe_udc *udc;
 	struct device_node *np = ofdev->dev.of_node;
-	unsigned int tmp_addr = 0;
+	unsigned long tmp_addr = 0;
 	struct usb_device_para __iomem *usbpram;
 	unsigned int i;
 	u64 size;
diff --git a/drivers/usb/gadget/udc/net2280.h b/drivers/usb/gadget/udc/net2280.h
index 4dff60d34f73..0d32052bf16f 100644
--- a/drivers/usb/gadget/udc/net2280.h
+++ b/drivers/usb/gadget/udc/net2280.h
@@ -369,9 +369,20 @@ static inline void set_max_speed(struct net2280_ep *ep, u32 max)
 	static const u32 ep_enhanced[9] = { 0x10, 0x60, 0x30, 0x80,
 					  0x50, 0x20, 0x70, 0x40, 0x90 };
 
-	if (ep->dev->enhanced_mode)
+	if (ep->dev->enhanced_mode) {
 		reg = ep_enhanced[ep->num];
-	else{
+		switch (ep->dev->gadget.speed) {
+		case USB_SPEED_SUPER:
+			reg += 2;
+			break;
+		case USB_SPEED_FULL:
+			reg += 1;
+			break;
+		case USB_SPEED_HIGH:
+		default:
+			break;
+		}
+	} else {
 		reg = (ep->num + 1) * 0x10;
 		if (ep->dev->gadget.speed != USB_SPEED_HIGH)
 			reg += 1;
diff --git a/drivers/usb/gadget/udc/udc-core.c b/drivers/usb/gadget/udc/udc-core.c
index fd73a3ea07c2..b86a6f03592e 100644
--- a/drivers/usb/gadget/udc/udc-core.c
+++ b/drivers/usb/gadget/udc/udc-core.c
@@ -413,9 +413,10 @@ int usb_add_gadget_udc_release(struct device *parent, struct usb_gadget *gadget,
 		if (!driver->udc_name || strcmp(driver->udc_name,
 						dev_name(&udc->dev)) == 0) {
 			ret = udc_bind_to_driver(udc, driver);
+			if (ret != -EPROBE_DEFER)
+				list_del(&driver->pending);
 			if (ret)
 				goto err4;
-			list_del(&driver->pending);
 			break;
 		}
 	}
diff --git a/drivers/usb/musb/musb_host.c b/drivers/usb/musb/musb_host.c
index 795a45b1b25b..58487a473521 100644
--- a/drivers/usb/musb/musb_host.c
+++ b/drivers/usb/musb/musb_host.c
@@ -662,7 +662,7 @@ static int musb_tx_dma_set_mode_mentor(struct dma_controller *dma,
 		csr &= ~(MUSB_TXCSR_AUTOSET | MUSB_TXCSR_DMAMODE);
 		csr |= MUSB_TXCSR_DMAENAB; /* against programmer's guide */
 	}
-	channel->desired_mode = mode;
+	channel->desired_mode = *mode;
 	musb_writew(epio, MUSB_TXCSR, csr);
 
 	return 0;
@@ -2003,10 +2003,8 @@ void musb_host_rx(struct musb *musb, u8 epnum)
 				qh->offset,
 				urb->transfer_buffer_length);
 
-			done = musb_rx_dma_in_inventra_cppi41(c, hw_ep, qh,
-							      urb, xfer_len,
-							      iso_err);
-			if (done)
+			if (musb_rx_dma_in_inventra_cppi41(c, hw_ep, qh, urb,
+							   xfer_len, iso_err))
 				goto finish;
 			else
 				dev_err(musb->controller, "error: rx_dma failed\n");
diff --git a/drivers/usb/phy/phy-msm-usb.c b/drivers/usb/phy/phy-msm-usb.c
index 970a30e155cb..72b387d592c2 100644
--- a/drivers/usb/phy/phy-msm-usb.c
+++ b/drivers/usb/phy/phy-msm-usb.c
@@ -757,14 +757,8 @@ static int msm_otg_set_host(struct usb_otg *otg, struct usb_bus *host)
 	otg->host = host;
 	dev_dbg(otg->usb_phy->dev, "host driver registered w/ tranceiver\n");
 
-	/*
-	 * Kick the state machine work, if peripheral is not supported
-	 * or peripheral is already registered with us.
-	 */
-	if (motg->pdata->mode == USB_DR_MODE_HOST || otg->gadget) {
-		pm_runtime_get_sync(otg->usb_phy->dev);
-		schedule_work(&motg->sm_work);
-	}
+	pm_runtime_get_sync(otg->usb_phy->dev);
+	schedule_work(&motg->sm_work);
 
 	return 0;
 }
@@ -827,14 +821,8 @@ static int msm_otg_set_peripheral(struct usb_otg *otg,
 	dev_dbg(otg->usb_phy->dev,
 		"peripheral driver registered w/ tranceiver\n");
 
-	/*
-	 * Kick the state machine work, if host is not supported
-	 * or host is already registered with us.
-	 */
-	if (motg->pdata->mode == USB_DR_MODE_PERIPHERAL || otg->host) {
-		pm_runtime_get_sync(otg->usb_phy->dev);
-		schedule_work(&motg->sm_work);
-	}
+	pm_runtime_get_sync(otg->usb_phy->dev);
+	schedule_work(&motg->sm_work);
 
 	return 0;
 }
diff --git a/drivers/usb/serial/Kconfig b/drivers/usb/serial/Kconfig
index f612dda9c977..56ecb8b5115d 100644
--- a/drivers/usb/serial/Kconfig
+++ b/drivers/usb/serial/Kconfig
@@ -475,22 +475,6 @@ config USB_SERIAL_MOS7840
 	  To compile this driver as a module, choose M here: the
 	  module will be called mos7840.  If unsure, choose N.
 
-config USB_SERIAL_MXUPORT11
-	tristate "USB Moxa UPORT 11x0 Serial Driver"
-	---help---
-	  Say Y here if you want to use a MOXA UPort 11x0 Serial hub.
-
-	  This driver supports:
-
-	  - UPort 1110  : 1 port RS-232 USB to Serial Hub.
-	  - UPort 1130  : 1 port RS-422/485 USB to Serial Hub.
-	  - UPort 1130I : 1 port RS-422/485 USB to Serial Hub with Isolation.
-	  - UPort 1150  : 1 port RS-232/422/485 USB to Serial Hub.
-	  - UPort 1150I : 1 port RS-232/422/485 USB to Serial Hub with Isolation.
-
-	  To compile this driver as a module, choose M here: the
-	  module will be called mxu11x0.
-
 config USB_SERIAL_MXUPORT
 	tristate "USB Moxa UPORT Serial Driver"
 	---help---
diff --git a/drivers/usb/serial/Makefile b/drivers/usb/serial/Makefile
index f3fa5e53702d..349d9df0895f 100644
--- a/drivers/usb/serial/Makefile
+++ b/drivers/usb/serial/Makefile
@@ -38,7 +38,6 @@ obj-$(CONFIG_USB_SERIAL_METRO)			+= metro-usb.o
 obj-$(CONFIG_USB_SERIAL_MOS7720)		+= mos7720.o
 obj-$(CONFIG_USB_SERIAL_MOS7840)		+= mos7840.o
 obj-$(CONFIG_USB_SERIAL_MXUPORT)		+= mxuport.o
-obj-$(CONFIG_USB_SERIAL_MXUPORT11)		+= mxu11x0.o
 obj-$(CONFIG_USB_SERIAL_NAVMAN)			+= navman.o
 obj-$(CONFIG_USB_SERIAL_OMNINET)		+= omninet.o
 obj-$(CONFIG_USB_SERIAL_OPTICON)		+= opticon.o
diff --git a/drivers/usb/serial/cp210x.c b/drivers/usb/serial/cp210x.c
index 987813b8a7f9..73a366de5102 100644
--- a/drivers/usb/serial/cp210x.c
+++ b/drivers/usb/serial/cp210x.c
@@ -163,6 +163,9 @@ static const struct usb_device_id id_table[] = {
 	{ USB_DEVICE(0x1843, 0x0200) }, /* Vaisala USB Instrument Cable */
 	{ USB_DEVICE(0x18EF, 0xE00F) }, /* ELV USB-I2C-Interface */
 	{ USB_DEVICE(0x18EF, 0xE025) }, /* ELV Marble Sound Board 1 */
+	{ USB_DEVICE(0x1901, 0x0190) }, /* GE B850 CP2105 Recorder interface */
+	{ USB_DEVICE(0x1901, 0x0193) }, /* GE B650 CP2104 PMC interface */
+	{ USB_DEVICE(0x19CF, 0x3000) }, /* Parrot NMEA GPS Flight Recorder */
 	{ USB_DEVICE(0x1ADB, 0x0001) }, /* Schweitzer Engineering C662 Cable */
 	{ USB_DEVICE(0x1B1C, 0x1C00) }, /* Corsair USB Dongle */
 	{ USB_DEVICE(0x1BA4, 0x0002) },	/* Silicon Labs 358x factory default */
diff --git a/drivers/usb/serial/mxu11x0.c b/drivers/usb/serial/mxu11x0.c
deleted file mode 100644
index 619607323bfd..000000000000
--- a/drivers/usb/serial/mxu11x0.c
+++ /dev/null
@@ -1,1006 +0,0 @@
-/*
- * USB Moxa UPORT 11x0 Serial Driver
- *
- * Copyright (C) 2007 MOXA Technologies Co., Ltd.
- * Copyright (C) 2015 Mathieu Othacehe <m.othacehe@gmail.com>
- *
- * This program is free software; you can redistribute it and/or modify
- * it under the terms of the GNU General Public License as published by
- * the Free Software Foundation; either version 2 of the License, or
- * (at your option) any later version.
- *
- *
- * Supports the following Moxa USB to serial converters:
- *  UPort 1110,  1 port RS-232 USB to Serial Hub.
- *  UPort 1130,  1 port RS-422/485 USB to Serial Hub.
- *  UPort 1130I, 1 port RS-422/485 USB to Serial Hub with isolation
- *    protection.
- *  UPort 1150,  1 port RS-232/422/485 USB to Serial Hub.
- *  UPort 1150I, 1 port RS-232/422/485 USB to Serial Hub with isolation
- *  protection.
- */
-
-#include <linux/kernel.h>
-#include <linux/module.h>
-#include <linux/firmware.h>
-#include <linux/jiffies.h>
-#include <linux/serial.h>
-#include <linux/serial_reg.h>
-#include <linux/slab.h>
-#include <linux/spinlock.h>
-#include <linux/mutex.h>
-#include <linux/tty.h>
-#include <linux/tty_driver.h>
-#include <linux/tty_flip.h>
-#include <linux/uaccess.h>
-#include <linux/usb.h>
-#include <linux/usb/serial.h>
-
-/* Vendor and product ids */
-#define MXU1_VENDOR_ID				0x110a
-#define MXU1_1110_PRODUCT_ID			0x1110
-#define MXU1_1130_PRODUCT_ID			0x1130
-#define MXU1_1150_PRODUCT_ID			0x1150
-#define MXU1_1151_PRODUCT_ID			0x1151
-#define MXU1_1131_PRODUCT_ID			0x1131
-
-/* Commands */
-#define MXU1_GET_VERSION			0x01
-#define MXU1_GET_PORT_STATUS			0x02
-#define MXU1_GET_PORT_DEV_INFO			0x03
-#define MXU1_GET_CONFIG				0x04
-#define MXU1_SET_CONFIG				0x05
-#define MXU1_OPEN_PORT				0x06
-#define MXU1_CLOSE_PORT				0x07
-#define MXU1_START_PORT				0x08
-#define MXU1_STOP_PORT				0x09
-#define MXU1_TEST_PORT				0x0A
-#define MXU1_PURGE_PORT				0x0B
-#define MXU1_RESET_EXT_DEVICE			0x0C
-#define MXU1_GET_OUTQUEUE			0x0D
-#define MXU1_WRITE_DATA				0x80
-#define MXU1_READ_DATA				0x81
-#define MXU1_REQ_TYPE_CLASS			0x82
-
-/* Module identifiers */
-#define MXU1_I2C_PORT				0x01
-#define MXU1_IEEE1284_PORT			0x02
-#define MXU1_UART1_PORT				0x03
-#define MXU1_UART2_PORT				0x04
-#define MXU1_RAM_PORT				0x05
-
-/* Modem status */
-#define MXU1_MSR_DELTA_CTS			0x01
-#define MXU1_MSR_DELTA_DSR			0x02
-#define MXU1_MSR_DELTA_RI			0x04
-#define MXU1_MSR_DELTA_CD			0x08
-#define MXU1_MSR_CTS				0x10
-#define MXU1_MSR_DSR				0x20
-#define MXU1_MSR_RI				0x40
-#define MXU1_MSR_CD				0x80
-#define MXU1_MSR_DELTA_MASK			0x0F
-#define MXU1_MSR_MASK				0xF0
-
-/* Line status */
-#define MXU1_LSR_OVERRUN_ERROR			0x01
-#define MXU1_LSR_PARITY_ERROR			0x02
-#define MXU1_LSR_FRAMING_ERROR			0x04
-#define MXU1_LSR_BREAK				0x08
-#define MXU1_LSR_ERROR				0x0F
-#define MXU1_LSR_RX_FULL			0x10
-#define MXU1_LSR_TX_EMPTY			0x20
-
-/* Modem control */
-#define MXU1_MCR_LOOP				0x04
-#define MXU1_MCR_DTR				0x10
-#define MXU1_MCR_RTS				0x20
-
-/* Mask settings */
-#define MXU1_UART_ENABLE_RTS_IN			0x0001
-#define MXU1_UART_DISABLE_RTS			0x0002
-#define MXU1_UART_ENABLE_PARITY_CHECKING	0x0008
-#define MXU1_UART_ENABLE_DSR_OUT		0x0010
-#define MXU1_UART_ENABLE_CTS_OUT		0x0020
-#define MXU1_UART_ENABLE_X_OUT			0x0040
-#define MXU1_UART_ENABLE_XA_OUT			0x0080
-#define MXU1_UART_ENABLE_X_IN			0x0100
-#define MXU1_UART_ENABLE_DTR_IN			0x0800
-#define MXU1_UART_DISABLE_DTR			0x1000
-#define MXU1_UART_ENABLE_MS_INTS		0x2000
-#define MXU1_UART_ENABLE_AUTO_START_DMA		0x4000
-#define MXU1_UART_SEND_BREAK_SIGNAL		0x8000
-
-/* Parity */
-#define MXU1_UART_NO_PARITY			0x00
-#define MXU1_UART_ODD_PARITY			0x01
-#define MXU1_UART_EVEN_PARITY			0x02
-#define MXU1_UART_MARK_PARITY			0x03
-#define MXU1_UART_SPACE_PARITY			0x04
-
-/* Stop bits */
-#define MXU1_UART_1_STOP_BITS			0x00
-#define MXU1_UART_1_5_STOP_BITS			0x01
-#define MXU1_UART_2_STOP_BITS			0x02
-
-/* Bits per character */
-#define MXU1_UART_5_DATA_BITS			0x00
-#define MXU1_UART_6_DATA_BITS			0x01
-#define MXU1_UART_7_DATA_BITS			0x02
-#define MXU1_UART_8_DATA_BITS			0x03
-
-/* Operation modes */
-#define MXU1_UART_232				0x00
-#define MXU1_UART_485_RECEIVER_DISABLED		0x01
-#define MXU1_UART_485_RECEIVER_ENABLED		0x02
-
-/* Pipe transfer mode and timeout */
-#define MXU1_PIPE_MODE_CONTINUOUS		0x01
-#define MXU1_PIPE_MODE_MASK			0x03
-#define MXU1_PIPE_TIMEOUT_MASK			0x7C
-#define MXU1_PIPE_TIMEOUT_ENABLE		0x80
-
-/* Config struct */
-struct mxu1_uart_config {
-	__be16	wBaudRate;
-	__be16	wFlags;
-	u8	bDataBits;
-	u8	bParity;
-	u8	bStopBits;
-	char	cXon;
-	char	cXoff;
-	u8	bUartMode;
-} __packed;
-
-/* Purge modes */
-#define MXU1_PURGE_OUTPUT			0x00
-#define MXU1_PURGE_INPUT			0x80
-
-/* Read/Write data */
-#define MXU1_RW_DATA_ADDR_SFR			0x10
-#define MXU1_RW_DATA_ADDR_IDATA			0x20
-#define MXU1_RW_DATA_ADDR_XDATA			0x30
-#define MXU1_RW_DATA_ADDR_CODE			0x40
-#define MXU1_RW_DATA_ADDR_GPIO			0x50
-#define MXU1_RW_DATA_ADDR_I2C			0x60
-#define MXU1_RW_DATA_ADDR_FLASH			0x70
-#define MXU1_RW_DATA_ADDR_DSP			0x80
-
-#define MXU1_RW_DATA_UNSPECIFIED		0x00
-#define MXU1_RW_DATA_BYTE			0x01
-#define MXU1_RW_DATA_WORD			0x02
-#define MXU1_RW_DATA_DOUBLE_WORD		0x04
-
-struct mxu1_write_data_bytes {
-	u8	bAddrType;
-	u8	bDataType;
-	u8	bDataCounter;
-	__be16	wBaseAddrHi;
-	__be16	wBaseAddrLo;
-	u8	bData[0];
-} __packed;
-
-/* Interrupt codes */
-#define MXU1_CODE_HARDWARE_ERROR		0xFF
-#define MXU1_CODE_DATA_ERROR			0x03
-#define MXU1_CODE_MODEM_STATUS			0x04
-
-static inline int mxu1_get_func_from_code(unsigned char code)
-{
-	return code & 0x0f;
-}
-
-/* Download firmware max packet size */
-#define MXU1_DOWNLOAD_MAX_PACKET_SIZE		64
-
-/* Firmware image header */
-struct mxu1_firmware_header {
-	__le16 wLength;
-	u8 bCheckSum;
-} __packed;
-
-#define MXU1_UART_BASE_ADDR	    0xFFA0
-#define MXU1_UART_OFFSET_MCR	    0x0004
-
-#define MXU1_BAUD_BASE              923077
-
-#define MXU1_TRANSFER_TIMEOUT	    2
-#define MXU1_DOWNLOAD_TIMEOUT       1000
-#define MXU1_DEFAULT_CLOSING_WAIT   4000 /* in .01 secs */
-
-struct mxu1_port {
-	u8 msr;
-	u8 mcr;
-	u8 uart_mode;
-	spinlock_t spinlock; /* Protects msr */
-	struct mutex mutex; /* Protects mcr */
-	bool send_break;
-};
-
-struct mxu1_device {
-	u16 mxd_model;
-};
-
-static const struct usb_device_id mxu1_idtable[] = {
-	{ USB_DEVICE(MXU1_VENDOR_ID, MXU1_1110_PRODUCT_ID) },
-	{ USB_DEVICE(MXU1_VENDOR_ID, MXU1_1130_PRODUCT_ID) },
-	{ USB_DEVICE(MXU1_VENDOR_ID, MXU1_1150_PRODUCT_ID) },
-	{ USB_DEVICE(MXU1_VENDOR_ID, MXU1_1151_PRODUCT_ID) },
-	{ USB_DEVICE(MXU1_VENDOR_ID, MXU1_1131_PRODUCT_ID) },
-	{ }
-};
-
-MODULE_DEVICE_TABLE(usb, mxu1_idtable);
-
-/* Write the given buffer out to the control pipe.  */
-static int mxu1_send_ctrl_data_urb(struct usb_serial *serial,
-				   u8 request,
-				   u16 value, u16 index,
-				   void *data, size_t size)
-{
-	int status;
-
-	status = usb_control_msg(serial->dev,
-				 usb_sndctrlpipe(serial->dev, 0),
-				 request,
-				 (USB_DIR_OUT | USB_TYPE_VENDOR |
-				  USB_RECIP_DEVICE), value, index,
-				 data, size,
-				 USB_CTRL_SET_TIMEOUT);
-	if (status < 0) {
-		dev_err(&serial->interface->dev,
-			"%s - usb_control_msg failed: %d\n",
-			__func__, status);
-		return status;
-	}
-
-	if (status != size) {
-		dev_err(&serial->interface->dev,
-			"%s - short write (%d / %zd)\n",
-			__func__, status, size);
-		return -EIO;
-	}
-
-	return 0;
-}
-
-/* Send a vendor request without any data */
-static int mxu1_send_ctrl_urb(struct usb_serial *serial,
-			      u8 request, u16 value, u16 index)
-{
-	return mxu1_send_ctrl_data_urb(serial, request, value, index,
-				       NULL, 0);
-}
-
-static int mxu1_download_firmware(struct usb_serial *serial,
-				  const struct firmware *fw_p)
-{
-	int status = 0;
-	int buffer_size;
-	int pos;
-	int len;
-	int done;
-	u8 cs = 0;
-	u8 *buffer;
-	struct usb_device *dev = serial->dev;
-	struct mxu1_firmware_header *header;
-	unsigned int pipe;
-
-	pipe = usb_sndbulkpipe(dev, serial->port[0]->bulk_out_endpointAddress);
-
-	buffer_size = fw_p->size + sizeof(*header);
-	buffer = kmalloc(buffer_size, GFP_KERNEL);
-	if (!buffer)
-		return -ENOMEM;
-
-	memcpy(buffer, fw_p->data, fw_p->size);
-	memset(buffer + fw_p->size, 0xff, buffer_size - fw_p->size);
-
-	for (pos = sizeof(*header); pos < buffer_size; pos++)
-		cs = (u8)(cs + buffer[pos]);
-
-	header = (struct mxu1_firmware_header *)buffer;
-	header->wLength = cpu_to_le16(buffer_size - sizeof(*header));
-	header->bCheckSum = cs;
-
-	dev_dbg(&dev->dev, "%s - downloading firmware\n", __func__);
-
-	for (pos = 0; pos < buffer_size; pos += done) {
-		len = min(buffer_size - pos, MXU1_DOWNLOAD_MAX_PACKET_SIZE);
-
-		status = usb_bulk_msg(dev, pipe, buffer + pos, len, &done,
-				MXU1_DOWNLOAD_TIMEOUT);
-		if (status)
-			break;
-	}
-
-	kfree(buffer);
-
-	if (status) {
-		dev_err(&dev->dev, "failed to download firmware: %d\n", status);
-		return status;
-	}
-
-	msleep_interruptible(100);
-	usb_reset_device(dev);
-
-	dev_dbg(&dev->dev, "%s - download successful\n", __func__);
-
-	return 0;
-}
-
-static int mxu1_port_probe(struct usb_serial_port *port)
-{
-	struct mxu1_port *mxport;
-	struct mxu1_device *mxdev;
-
-	if (!port->interrupt_in_urb) {
-		dev_err(&port->dev, "no interrupt urb\n");
-		return -ENODEV;
-	}
-
-	mxport = kzalloc(sizeof(struct mxu1_port), GFP_KERNEL);
-	if (!mxport)
-		return -ENOMEM;
-
-	spin_lock_init(&mxport->spinlock);
-	mutex_init(&mxport->mutex);
-
-	mxdev = usb_get_serial_data(port->serial);
-
-	switch (mxdev->mxd_model) {
-	case MXU1_1110_PRODUCT_ID:
-	case MXU1_1150_PRODUCT_ID:
-	case MXU1_1151_PRODUCT_ID:
-		mxport->uart_mode = MXU1_UART_232;
-		break;
-	case MXU1_1130_PRODUCT_ID:
-	case MXU1_1131_PRODUCT_ID:
-		mxport->uart_mode = MXU1_UART_485_RECEIVER_DISABLED;
-		break;
-	}
-
-	usb_set_serial_port_data(port, mxport);
-
-	port->port.closing_wait =
-			msecs_to_jiffies(MXU1_DEFAULT_CLOSING_WAIT * 10);
-	port->port.drain_delay = 1;
-
-	return 0;
-}
-
-static int mxu1_port_remove(struct usb_serial_port *port)
-{
-	struct mxu1_port *mxport;
-
-	mxport = usb_get_serial_port_data(port);
-	kfree(mxport);
-
-	return 0;
-}
-
-static int mxu1_startup(struct usb_serial *serial)
-{
-	struct mxu1_device *mxdev;
-	struct usb_device *dev = serial->dev;
-	struct usb_host_interface *cur_altsetting;
-	char fw_name[32];
-	const struct firmware *fw_p = NULL;
-	int err;
-
-	dev_dbg(&serial->interface->dev, "%s - product 0x%04X, num configurations %d, configuration value %d\n",
-		__func__, le16_to_cpu(dev->descriptor.idProduct),
-		dev->descriptor.bNumConfigurations,
-		dev->actconfig->desc.bConfigurationValue);
-
-	/* create device structure */
-	mxdev = kzalloc(sizeof(struct mxu1_device), GFP_KERNEL);
-	if (!mxdev)
-		return -ENOMEM;
-
-	usb_set_serial_data(serial, mxdev);
-
-	mxdev->mxd_model = le16_to_cpu(dev->descriptor.idProduct);
-
-	cur_altsetting = serial->interface->cur_altsetting;
-
-	/* if we have only 1 configuration, download firmware */
-	if (cur_altsetting->desc.bNumEndpoints == 1) {
-
-		snprintf(fw_name,
-			 sizeof(fw_name),
-			 "moxa/moxa-%04x.fw",
-			 mxdev->mxd_model);
-
-		err = request_firmware(&fw_p, fw_name, &serial->interface->dev);
-		if (err) {
-			dev_err(&serial->interface->dev, "failed to request firmware: %d\n",
-				err);
-			goto err_free_mxdev;
-		}
-
-		err = mxu1_download_firmware(serial, fw_p);
-		if (err)
-			goto err_release_firmware;
-
-		/* device is being reset */
-		err = -ENODEV;
-		goto err_release_firmware;
-	}
-
-	return 0;
-
-err_release_firmware:
-	release_firmware(fw_p);
-err_free_mxdev:
-	kfree(mxdev);
-
-	return err;
-}
-
-static void mxu1_release(struct usb_serial *serial)
-{
-	struct mxu1_device *mxdev;
-
-	mxdev = usb_get_serial_data(serial);
-	kfree(mxdev);
-}
-
-static int mxu1_write_byte(struct usb_serial_port *port, u32 addr,
-			   u8 mask, u8 byte)
-{
-	int status;
-	size_t size;
-	struct mxu1_write_data_bytes *data;
-
-	dev_dbg(&port->dev, "%s - addr 0x%08X, mask 0x%02X, byte 0x%02X\n",
-		__func__, addr, mask, byte);
-
-	size = sizeof(struct mxu1_write_data_bytes) + 2;
-	data = kzalloc(size, GFP_KERNEL);
-	if (!data)
-		return -ENOMEM;
-
-	data->bAddrType = MXU1_RW_DATA_ADDR_XDATA;
-	data->bDataType = MXU1_RW_DATA_BYTE;
-	data->bDataCounter = 1;
-	data->wBaseAddrHi = cpu_to_be16(addr >> 16);
-	data->wBaseAddrLo = cpu_to_be16(addr);
-	data->bData[0] = mask;
-	data->bData[1] = byte;
-
-	status = mxu1_send_ctrl_data_urb(port->serial, MXU1_WRITE_DATA, 0,
-					 MXU1_RAM_PORT, data, size);
-	if (status < 0)
-		dev_err(&port->dev, "%s - failed: %d\n", __func__, status);
-
-	kfree(data);
-
-	return status;
-}
-
-static int mxu1_set_mcr(struct usb_serial_port *port, unsigned int mcr)
-{
-	int status;
-
-	status = mxu1_write_byte(port,
-				 MXU1_UART_BASE_ADDR + MXU1_UART_OFFSET_MCR,
-				 MXU1_MCR_RTS | MXU1_MCR_DTR | MXU1_MCR_LOOP,
-				 mcr);
-	return status;
-}
-
-static void mxu1_set_termios(struct tty_struct *tty,
-			     struct usb_serial_port *port,
-			     struct ktermios *old_termios)
-{
-	struct mxu1_port *mxport = usb_get_serial_port_data(port);
-	struct mxu1_uart_config *config;
-	tcflag_t cflag, iflag;
-	speed_t baud;
-	int status;
-	unsigned int mcr;
-
-	cflag = tty->termios.c_cflag;
-	iflag = tty->termios.c_iflag;
-
-	if (old_termios &&
-	    !tty_termios_hw_change(&tty->termios, old_termios) &&
-	    tty->termios.c_iflag == old_termios->c_iflag) {
-		dev_dbg(&port->dev, "%s - nothing to change\n", __func__);
-		return;
-	}
-
-	dev_dbg(&port->dev,
-		"%s - cflag 0x%08x, iflag 0x%08x\n", __func__, cflag, iflag);
-
-	if (old_termios) {
-		dev_dbg(&port->dev, "%s - old cflag 0x%08x, old iflag 0x%08x\n",
-			__func__,
-			old_termios->c_cflag,
-			old_termios->c_iflag);
-	}
-
-	config = kzalloc(sizeof(*config), GFP_KERNEL);
-	if (!config)
-		return;
-
-	/* these flags must be set */
-	config->wFlags |= MXU1_UART_ENABLE_MS_INTS;
-	config->wFlags |= MXU1_UART_ENABLE_AUTO_START_DMA;
-	if (mxport->send_break)
-		config->wFlags |= MXU1_UART_SEND_BREAK_SIGNAL;
-	config->bUartMode = mxport->uart_mode;
-
-	switch (C_CSIZE(tty)) {
-	case CS5:
-		config->bDataBits = MXU1_UART_5_DATA_BITS;
-		break;
-	case CS6:
-		config->bDataBits = MXU1_UART_6_DATA_BITS;
-		break;
-	case CS7:
-		config->bDataBits = MXU1_UART_7_DATA_BITS;
-		break;
-	default:
-	case CS8:
-		config->bDataBits = MXU1_UART_8_DATA_BITS;
-		break;
-	}
-
-	if (C_PARENB(tty)) {
-		config->wFlags |= MXU1_UART_ENABLE_PARITY_CHECKING;
-		if (C_CMSPAR(tty)) {
-			if (C_PARODD(tty))
-				config->bParity = MXU1_UART_MARK_PARITY;
-			else
-				config->bParity = MXU1_UART_SPACE_PARITY;
-		} else {
-			if (C_PARODD(tty))
-				config->bParity = MXU1_UART_ODD_PARITY;
-			else
-				config->bParity = MXU1_UART_EVEN_PARITY;
-		}
-	} else {
-		config->bParity = MXU1_UART_NO_PARITY;
-	}
-
-	if (C_CSTOPB(tty))
-		config->bStopBits = MXU1_UART_2_STOP_BITS;
-	else
-		config->bStopBits = MXU1_UART_1_STOP_BITS;
-
-	if (C_CRTSCTS(tty)) {
-		/* RTS flow control must be off to drop RTS for baud rate B0 */
-		if (C_BAUD(tty) != B0)
-			config->wFlags |= MXU1_UART_ENABLE_RTS_IN;
-		config->wFlags |= MXU1_UART_ENABLE_CTS_OUT;
-	}
-
-	if (I_IXOFF(tty) || I_IXON(tty)) {
-		config->cXon  = START_CHAR(tty);
-		config->cXoff = STOP_CHAR(tty);
-
-		if (I_IXOFF(tty))
-			config->wFlags |= MXU1_UART_ENABLE_X_IN;
-
-		if (I_IXON(tty))
-			config->wFlags |= MXU1_UART_ENABLE_X_OUT;
-	}
-
-	baud = tty_get_baud_rate(tty);
-	if (!baud)
-		baud = 9600;
-	config->wBaudRate = MXU1_BAUD_BASE / baud;
-
-	dev_dbg(&port->dev, "%s - BaudRate=%d, wBaudRate=%d, wFlags=0x%04X, bDataBits=%d, bParity=%d, bStopBits=%d, cXon=%d, cXoff=%d, bUartMode=%d\n",
-		__func__, baud, config->wBaudRate, config->wFlags,
-		config->bDataBits, config->bParity, config->bStopBits,
-		config->cXon, config->cXoff, config->bUartMode);
-
-	cpu_to_be16s(&config->wBaudRate);
-	cpu_to_be16s(&config->wFlags);
-
-	status = mxu1_send_ctrl_data_urb(port->serial, MXU1_SET_CONFIG, 0,
-					 MXU1_UART1_PORT, config,
-					 sizeof(*config));
-	if (status)
-		dev_err(&port->dev, "cannot set config: %d\n", status);
-
-	mutex_lock(&mxport->mutex);
-	mcr = mxport->mcr;
-
-	if (C_BAUD(tty) == B0)
-		mcr &= ~(MXU1_MCR_DTR | MXU1_MCR_RTS);
-	else if (old_termios && (old_termios->c_cflag & CBAUD) == B0)
-		mcr |= MXU1_MCR_DTR | MXU1_MCR_RTS;
-
-	status = mxu1_set_mcr(port, mcr);
-	if (status)
-		dev_err(&port->dev, "cannot set modem control: %d\n", status);
-	else
-		mxport->mcr = mcr;
-
-	mutex_unlock(&mxport->mutex);
-
-	kfree(config);
-}
-
-static int mxu1_get_serial_info(struct usb_serial_port *port,
-				struct serial_struct __user *ret_arg)
-{
-	struct serial_struct ret_serial;
-	unsigned cwait;
-
-	if (!ret_arg)
-		return -EFAULT;
-
-	cwait = port->port.closing_wait;
-	if (cwait != ASYNC_CLOSING_WAIT_NONE)
-		cwait = jiffies_to_msecs(cwait) / 10;
-
-	memset(&ret_serial, 0, sizeof(ret_serial));
-
-	ret_serial.type = PORT_16550A;
-	ret_serial.line = port->minor;
-	ret_serial.port = 0;
-	ret_serial.xmit_fifo_size = port->bulk_out_size;
-	ret_serial.baud_base = MXU1_BAUD_BASE;
-	ret_serial.close_delay = 5*HZ;
-	ret_serial.closing_wait = cwait;
-
-	if (copy_to_user(ret_arg, &ret_serial, sizeof(*ret_arg)))
-		return -EFAULT;
-
-	return 0;
-}
-
-
-static int mxu1_set_serial_info(struct usb_serial_port *port,
-				struct serial_struct __user *new_arg)
-{
-	struct serial_struct new_serial;
-	unsigned cwait;
-
-	if (copy_from_user(&new_serial, new_arg, sizeof(new_serial)))
-		return -EFAULT;
-
-	cwait = new_serial.closing_wait;
-	if (cwait != ASYNC_CLOSING_WAIT_NONE)
-		cwait = msecs_to_jiffies(10 * new_serial.closing_wait);
-
-	port->port.closing_wait = cwait;
-
-	return 0;
-}
-
-static int mxu1_ioctl(struct tty_struct *tty,
-		      unsigned int cmd, unsigned long arg)
-{
-	struct usb_serial_port *port = tty->driver_data;
-
-	switch (cmd) {
-	case TIOCGSERIAL:
-		return mxu1_get_serial_info(port,
-					    (struct serial_struct __user *)arg);
-	case TIOCSSERIAL:
-		return mxu1_set_serial_info(port,
-					    (struct serial_struct __user *)arg);
-	}
-
-	return -ENOIOCTLCMD;
-}
-
-static int mxu1_tiocmget(struct tty_struct *tty)
-{
-	struct usb_serial_port *port = tty->driver_data;
-	struct mxu1_port *mxport = usb_get_serial_port_data(port);
-	unsigned int result;
-	unsigned int msr;
-	unsigned int mcr;
-	unsigned long flags;
-
-	mutex_lock(&mxport->mutex);
-	spin_lock_irqsave(&mxport->spinlock, flags);
-
-	msr = mxport->msr;
-	mcr = mxport->mcr;
-
-	spin_unlock_irqrestore(&mxport->spinlock, flags);
-	mutex_unlock(&mxport->mutex);
-
-	result = ((mcr & MXU1_MCR_DTR)	? TIOCM_DTR	: 0) |
-		 ((mcr & MXU1_MCR_RTS)	? TIOCM_RTS	: 0) |
-		 ((mcr & MXU1_MCR_LOOP) ? TIOCM_LOOP	: 0) |
-		 ((msr & MXU1_MSR_CTS)	? TIOCM_CTS	: 0) |
-		 ((msr & MXU1_MSR_CD)	? TIOCM_CAR	: 0) |
-		 ((msr & MXU1_MSR_RI)	? TIOCM_RI	: 0) |
-		 ((msr & MXU1_MSR_DSR)	? TIOCM_DSR	: 0);
-
-	dev_dbg(&port->dev, "%s - 0x%04X\n", __func__, result);
-
-	return result;
-}
-
-static int mxu1_tiocmset(struct tty_struct *tty,
-			 unsigned int set, unsigned int clear)
-{
-	struct usb_serial_port *port = tty->driver_data;
-	struct mxu1_port *mxport = usb_get_serial_port_data(port);
-	int err;
-	unsigned int mcr;
-
-	mutex_lock(&mxport->mutex);
-	mcr = mxport->mcr;
-
-	if (set & TIOCM_RTS)
-		mcr |= MXU1_MCR_RTS;
-	if (set & TIOCM_DTR)
-		mcr |= MXU1_MCR_DTR;
-	if (set & TIOCM_LOOP)
-		mcr |= MXU1_MCR_LOOP;
-
-	if (clear & TIOCM_RTS)
-		mcr &= ~MXU1_MCR_RTS;
-	if (clear & TIOCM_DTR)
-		mcr &= ~MXU1_MCR_DTR;
-	if (clear & TIOCM_LOOP)
-		mcr &= ~MXU1_MCR_LOOP;
-
-	err = mxu1_set_mcr(port, mcr);
-	if (!err)
-		mxport->mcr = mcr;
-
-	mutex_unlock(&mxport->mutex);
-
-	return err;
-}
-
-static void mxu1_break(struct tty_struct *tty, int break_state)
-{
-	struct usb_serial_port *port = tty->driver_data;
-	struct mxu1_port *mxport = usb_get_serial_port_data(port);
-
-	if (break_state == -1)
-		mxport->send_break = true;
-	else
-		mxport->send_break = false;
-
-	mxu1_set_termios(tty, port, NULL);
-}
-
-static int mxu1_open(struct tty_struct *tty, struct usb_serial_port *port)
-{
-	struct mxu1_port *mxport = usb_get_serial_port_data(port);
-	struct usb_serial *serial = port->serial;
-	int status;
-	u16 open_settings;
-
-	open_settings = (MXU1_PIPE_MODE_CONTINUOUS |
-			 MXU1_PIPE_TIMEOUT_ENABLE |
-			 (MXU1_TRANSFER_TIMEOUT << 2));
-
-	mxport->msr = 0;
-
-	status = usb_submit_urb(port->interrupt_in_urb, GFP_KERNEL);
-	if (status) {
-		dev_err(&port->dev, "failed to submit interrupt urb: %d\n",
-			status);
-		return status;
-	}
-
-	if (tty)
-		mxu1_set_termios(tty, port, NULL);
-
-	status = mxu1_send_ctrl_urb(serial, MXU1_OPEN_PORT,
-				    open_settings, MXU1_UART1_PORT);
-	if (status) {
-		dev_err(&port->dev, "cannot send open command: %d\n", status);
-		goto unlink_int_urb;
-	}
-
-	status = mxu1_send_ctrl_urb(serial, MXU1_START_PORT,
-				    0, MXU1_UART1_PORT);
-	if (status) {
-		dev_err(&port->dev, "cannot send start command: %d\n", status);
-		goto unlink_int_urb;
-	}
-
-	status = mxu1_send_ctrl_urb(serial, MXU1_PURGE_PORT,
-				    MXU1_PURGE_INPUT, MXU1_UART1_PORT);
-	if (status) {
-		dev_err(&port->dev, "cannot clear input buffers: %d\n",
-			status);
-
-		goto unlink_int_urb;
-	}
-
-	status = mxu1_send_ctrl_urb(serial, MXU1_PURGE_PORT,
-				    MXU1_PURGE_OUTPUT, MXU1_UART1_PORT);
-	if (status) {
-		dev_err(&port->dev, "cannot clear output buffers: %d\n",
-			status);
-
-		goto unlink_int_urb;
-	}
-
-	/*
-	 * reset the data toggle on the bulk endpoints to work around bug in
-	 * host controllers where things get out of sync some times
-	 */
-	usb_clear_halt(serial->dev, port->write_urb->pipe);
-	usb_clear_halt(serial->dev, port->read_urb->pipe);
-
-	if (tty)
-		mxu1_set_termios(tty, port, NULL);
-
-	status = mxu1_send_ctrl_urb(serial, MXU1_OPEN_PORT,
-				    open_settings, MXU1_UART1_PORT);
-	if (status) {
-		dev_err(&port->dev, "cannot send open command: %d\n", status);
-		goto unlink_int_urb;
-	}
-
-	status = mxu1_send_ctrl_urb(serial, MXU1_START_PORT,
-				    0, MXU1_UART1_PORT);
-	if (status) {
-		dev_err(&port->dev, "cannot send start command: %d\n", status);
-		goto unlink_int_urb;
-	}
-
-	status = usb_serial_generic_open(tty, port);
-	if (status)
-		goto unlink_int_urb;
-
-	return 0;
-
-unlink_int_urb:
-	usb_kill_urb(port->interrupt_in_urb);
-
-	return status;
-}
-
-static void mxu1_close(struct usb_serial_port *port)
-{
-	int status;
-
-	usb_serial_generic_close(port);
-	usb_kill_urb(port->interrupt_in_urb);
-
-	status = mxu1_send_ctrl_urb(port->serial, MXU1_CLOSE_PORT,
-				    0, MXU1_UART1_PORT);
-	if (status) {
-		dev_err(&port->dev, "failed to send close port command: %d\n",
-			status);
-	}
-}
-
-static void mxu1_handle_new_msr(struct usb_serial_port *port, u8 msr)
-{
-	struct mxu1_port *mxport = usb_get_serial_port_data(port);
-	struct async_icount *icount;
-	unsigned long flags;
-
-	dev_dbg(&port->dev, "%s - msr 0x%02X\n", __func__, msr);
-
-	spin_lock_irqsave(&mxport->spinlock, flags);
-	mxport->msr = msr & MXU1_MSR_MASK;
-	spin_unlock_irqrestore(&mxport->spinlock, flags);
-
-	if (msr & MXU1_MSR_DELTA_MASK) {
-		icount = &port->icount;
-		if (msr & MXU1_MSR_DELTA_CTS)
-			icount->cts++;
-		if (msr & MXU1_MSR_DELTA_DSR)
-			icount->dsr++;
-		if (msr & MXU1_MSR_DELTA_CD)
-			icount->dcd++;
-		if (msr & MXU1_MSR_DELTA_RI)
-			icount->rng++;
-
-		wake_up_interruptible(&port->port.delta_msr_wait);
-	}
-}
-
-static void mxu1_interrupt_callback(struct urb *urb)
-{
-	struct usb_serial_port *port = urb->context;
-	unsigned char *data = urb->transfer_buffer;
-	int length = urb->actual_length;
-	int function;
-	int status;
-	u8 msr;
-
-	switch (urb->status) {
-	case 0:
-		break;
-	case -ECONNRESET:
-	case -ENOENT:
-	case -ESHUTDOWN:
-		dev_dbg(&port->dev, "%s - urb shutting down: %d\n",
-			__func__, urb->status);
-		return;
-	default:
-		dev_dbg(&port->dev, "%s - nonzero urb status: %d\n",
-			__func__, urb->status);
-		goto exit;
-	}
-
-	if (length != 2) {
-		dev_dbg(&port->dev, "%s - bad packet size: %d\n",
-			__func__, length);
-		goto exit;
-	}
-
-	if (data[0] == MXU1_CODE_HARDWARE_ERROR) {
-		dev_err(&port->dev, "hardware error: %d\n", data[1]);
-		goto exit;
-	}
-
-	function = mxu1_get_func_from_code(data[0]);
-
-	dev_dbg(&port->dev, "%s - function %d, data 0x%02X\n",
-		 __func__, function, data[1]);
-
-	switch (function) {
-	case MXU1_CODE_DATA_ERROR:
-		dev_dbg(&port->dev, "%s - DATA ERROR, data 0x%02X\n",
-			 __func__, data[1]);
-		break;
-
-	case MXU1_CODE_MODEM_STATUS:
-		msr = data[1];
-		mxu1_handle_new_msr(port, msr);
-		break;
-
-	default:
-		dev_err(&port->dev, "unknown interrupt code: 0x%02X\n",
-			data[1]);
-		break;
-	}
-
-exit:
-	status = usb_submit_urb(urb, GFP_ATOMIC);
-	if (status) {
-		dev_err(&port->dev, "resubmit interrupt urb failed: %d\n",
-			status);
-	}
-}
-
-static struct usb_serial_driver mxu11x0_device = {
-	.driver = {
-		.owner		= THIS_MODULE,
-		.name		= "mxu11x0",
-	},
-	.description		= "MOXA UPort 11x0",
-	.id_table		= mxu1_idtable,
-	.num_ports		= 1,
-	.port_probe             = mxu1_port_probe,
-	.port_remove            = mxu1_port_remove,
-	.attach			= mxu1_startup,
-	.release                = mxu1_release,
-	.open			= mxu1_open,
-	.close			= mxu1_close,
-	.ioctl			= mxu1_ioctl,
-	.set_termios		= mxu1_set_termios,
-	.tiocmget		= mxu1_tiocmget,
-	.tiocmset		= mxu1_tiocmset,
-	.tiocmiwait		= usb_serial_generic_tiocmiwait,
-	.get_icount		= usb_serial_generic_get_icount,
-	.break_ctl		= mxu1_break,
-	.read_int_callback	= mxu1_interrupt_callback,
-};
-
-static struct usb_serial_driver *const serial_drivers[] = {
-	&mxu11x0_device, NULL
-};
-
-module_usb_serial_driver(serial_drivers, mxu1_idtable);
-
-MODULE_AUTHOR("Mathieu Othacehe <m.othacehe@gmail.com>");
-MODULE_DESCRIPTION("MOXA UPort 11x0 USB to Serial Hub Driver");
-MODULE_LICENSE("GPL");
-MODULE_FIRMWARE("moxa/moxa-1110.fw");
-MODULE_FIRMWARE("moxa/moxa-1130.fw");
-MODULE_FIRMWARE("moxa/moxa-1131.fw");
-MODULE_FIRMWARE("moxa/moxa-1150.fw");
-MODULE_FIRMWARE("moxa/moxa-1151.fw");
diff --git a/drivers/usb/serial/option.c b/drivers/usb/serial/option.c
index db86e512e0fc..348e19834b83 100644
--- a/drivers/usb/serial/option.c
+++ b/drivers/usb/serial/option.c
@@ -270,6 +270,7 @@ static void option_instat_callback(struct urb *urb);
 #define TELIT_PRODUCT_UE910_V2			0x1012
 #define TELIT_PRODUCT_LE922_USBCFG0		0x1042
 #define TELIT_PRODUCT_LE922_USBCFG3		0x1043
+#define TELIT_PRODUCT_LE922_USBCFG5		0x1045
 #define TELIT_PRODUCT_LE920			0x1200
 #define TELIT_PRODUCT_LE910			0x1201
 
@@ -315,6 +316,7 @@ static void option_instat_callback(struct urb *urb);
 #define TOSHIBA_PRODUCT_G450			0x0d45
 
 #define ALINK_VENDOR_ID				0x1e0e
+#define SIMCOM_PRODUCT_SIM7100E			0x9001 /* Yes, ALINK_VENDOR_ID */
 #define ALINK_PRODUCT_PH300			0x9100
 #define ALINK_PRODUCT_3GU			0x9200
 
@@ -607,6 +609,10 @@ static const struct option_blacklist_info zte_1255_blacklist = {
 	.reserved = BIT(3) | BIT(4),
 };
 
+static const struct option_blacklist_info simcom_sim7100e_blacklist = {
+	.reserved = BIT(5) | BIT(6),
+};
+
 static const struct option_blacklist_info telit_le910_blacklist = {
 	.sendsetup = BIT(0),
 	.reserved = BIT(1) | BIT(2),
@@ -1122,9 +1128,13 @@ static const struct usb_device_id option_ids[] = {
 	{ USB_DEVICE(KYOCERA_VENDOR_ID, KYOCERA_PRODUCT_KPC650) },
 	{ USB_DEVICE(KYOCERA_VENDOR_ID, KYOCERA_PRODUCT_KPC680) },
 	{ USB_DEVICE(QUALCOMM_VENDOR_ID, 0x6000)}, /* ZTE AC8700 */
+	{ USB_DEVICE_AND_INTERFACE_INFO(QUALCOMM_VENDOR_ID, 0x6001, 0xff, 0xff, 0xff), /* 4G LTE usb-modem U901 */
+	  .driver_info = (kernel_ulong_t)&net_intf3_blacklist },
 	{ USB_DEVICE(QUALCOMM_VENDOR_ID, 0x6613)}, /* Onda H600/ZTE MF330 */
 	{ USB_DEVICE(QUALCOMM_VENDOR_ID, 0x0023)}, /* ONYX 3G device */
 	{ USB_DEVICE(QUALCOMM_VENDOR_ID, 0x9000)}, /* SIMCom SIM5218 */
+	{ USB_DEVICE(QUALCOMM_VENDOR_ID, 0x9003), /* Quectel UC20 */
+	  .driver_info = (kernel_ulong_t)&net_intf4_blacklist },
 	{ USB_DEVICE(CMOTECH_VENDOR_ID, CMOTECH_PRODUCT_6001) },
 	{ USB_DEVICE(CMOTECH_VENDOR_ID, CMOTECH_PRODUCT_CMU_300) },
 	{ USB_DEVICE(CMOTECH_VENDOR_ID, CMOTECH_PRODUCT_6003),
@@ -1176,6 +1186,8 @@ static const struct usb_device_id option_ids[] = {
 		.driver_info = (kernel_ulong_t)&telit_le922_blacklist_usbcfg0 },
 	{ USB_DEVICE(TELIT_VENDOR_ID, TELIT_PRODUCT_LE922_USBCFG3),
 		.driver_info = (kernel_ulong_t)&telit_le922_blacklist_usbcfg3 },
+	{ USB_DEVICE_INTERFACE_CLASS(TELIT_VENDOR_ID, TELIT_PRODUCT_LE922_USBCFG5, 0xff),
+		.driver_info = (kernel_ulong_t)&telit_le922_blacklist_usbcfg0 },
 	{ USB_DEVICE(TELIT_VENDOR_ID, TELIT_PRODUCT_LE910),
 		.driver_info = (kernel_ulong_t)&telit_le910_blacklist },
 	{ USB_DEVICE(TELIT_VENDOR_ID, TELIT_PRODUCT_LE920),
@@ -1645,6 +1657,8 @@ static const struct usb_device_id option_ids[] = {
 	{ USB_DEVICE(ALINK_VENDOR_ID, 0x9000) },
 	{ USB_DEVICE(ALINK_VENDOR_ID, ALINK_PRODUCT_PH300) },
 	{ USB_DEVICE_AND_INTERFACE_INFO(ALINK_VENDOR_ID, ALINK_PRODUCT_3GU, 0xff, 0xff, 0xff) },
+	{ USB_DEVICE(ALINK_VENDOR_ID, SIMCOM_PRODUCT_SIM7100E),
+	  .driver_info = (kernel_ulong_t)&simcom_sim7100e_blacklist },
 	{ USB_DEVICE(ALCATEL_VENDOR_ID, ALCATEL_PRODUCT_X060S_X200),
 	  .driver_info = (kernel_ulong_t)&alcatel_x200_blacklist
 	},
diff --git a/drivers/usb/serial/qcserial.c b/drivers/usb/serial/qcserial.c
index 9919d2a9faf2..1bc6089b9008 100644
--- a/drivers/usb/serial/qcserial.c
+++ b/drivers/usb/serial/qcserial.c
@@ -157,14 +157,17 @@ static const struct usb_device_id id_table[] = {
 	{DEVICE_SWI(0x1199, 0x9056)},	/* Sierra Wireless Modem */
 	{DEVICE_SWI(0x1199, 0x9060)},	/* Sierra Wireless Modem */
 	{DEVICE_SWI(0x1199, 0x9061)},	/* Sierra Wireless Modem */
-	{DEVICE_SWI(0x1199, 0x9070)},	/* Sierra Wireless MC74xx/EM74xx */
-	{DEVICE_SWI(0x1199, 0x9071)},	/* Sierra Wireless MC74xx/EM74xx */
+	{DEVICE_SWI(0x1199, 0x9070)},	/* Sierra Wireless MC74xx */
+	{DEVICE_SWI(0x1199, 0x9071)},	/* Sierra Wireless MC74xx */
+	{DEVICE_SWI(0x1199, 0x9078)},	/* Sierra Wireless EM74xx */
+	{DEVICE_SWI(0x1199, 0x9079)},	/* Sierra Wireless EM74xx */
 	{DEVICE_SWI(0x413c, 0x81a2)},	/* Dell Wireless 5806 Gobi(TM) 4G LTE Mobile Broadband Card */
 	{DEVICE_SWI(0x413c, 0x81a3)},	/* Dell Wireless 5570 HSPA+ (42Mbps) Mobile Broadband Card */
 	{DEVICE_SWI(0x413c, 0x81a4)},	/* Dell Wireless 5570e HSPA+ (42Mbps) Mobile Broadband Card */
 	{DEVICE_SWI(0x413c, 0x81a8)},	/* Dell Wireless 5808 Gobi(TM) 4G LTE Mobile Broadband Card */
 	{DEVICE_SWI(0x413c, 0x81a9)},	/* Dell Wireless 5808e Gobi(TM) 4G LTE Mobile Broadband Card */
 	{DEVICE_SWI(0x413c, 0x81b1)},	/* Dell Wireless 5809e Gobi(TM) 4G LTE Mobile Broadband Card */
+	{DEVICE_SWI(0x413c, 0x81b3)},	/* Dell Wireless 5809e Gobi(TM) 4G LTE Mobile Broadband Card (rev3) */
 
 	/* Huawei devices */
 	{DEVICE_HWI(0x03f0, 0x581d)},	/* HP lt4112 LTE/HSPA+ Gobi 4G Modem (Huawei me906e) */
diff --git a/drivers/vfio/pci/vfio_pci.c b/drivers/vfio/pci/vfio_pci.c
index 2760a7ba3f30..8c80a48e3233 100644
--- a/drivers/vfio/pci/vfio_pci.c
+++ b/drivers/vfio/pci/vfio_pci.c
@@ -446,7 +446,8 @@ static long vfio_pci_ioctl(void *device_data,
 		info.num_regions = VFIO_PCI_NUM_REGIONS;
 		info.num_irqs = VFIO_PCI_NUM_IRQS;
 
-		return copy_to_user((void __user *)arg, &info, minsz);
+		return copy_to_user((void __user *)arg, &info, minsz) ?
+			-EFAULT : 0;
 
 	} else if (cmd == VFIO_DEVICE_GET_REGION_INFO) {
 		struct pci_dev *pdev = vdev->pdev;
@@ -520,7 +521,8 @@ static long vfio_pci_ioctl(void *device_data,
 			return -EINVAL;
 		}
 
-		return copy_to_user((void __user *)arg, &info, minsz);
+		return copy_to_user((void __user *)arg, &info, minsz) ?
+			-EFAULT : 0;
 
 	} else if (cmd == VFIO_DEVICE_GET_IRQ_INFO) {
 		struct vfio_irq_info info;
@@ -555,7 +557,8 @@ static long vfio_pci_ioctl(void *device_data,
 		else
 			info.flags |= VFIO_IRQ_INFO_NORESIZE;
 
-		return copy_to_user((void __user *)arg, &info, minsz);
+		return copy_to_user((void __user *)arg, &info, minsz) ?
+			-EFAULT : 0;
 
 	} else if (cmd == VFIO_DEVICE_SET_IRQS) {
 		struct vfio_irq_set hdr;
diff --git a/drivers/vfio/platform/vfio_platform_common.c b/drivers/vfio/platform/vfio_platform_common.c
index 418cdd9ba3f4..e65b142d3422 100644
--- a/drivers/vfio/platform/vfio_platform_common.c
+++ b/drivers/vfio/platform/vfio_platform_common.c
@@ -219,7 +219,8 @@ static long vfio_platform_ioctl(void *device_data,
 		info.num_regions = vdev->num_regions;
 		info.num_irqs = vdev->num_irqs;
 
-		return copy_to_user((void __user *)arg, &info, minsz);
+		return copy_to_user((void __user *)arg, &info, minsz) ?
+			-EFAULT : 0;
 
 	} else if (cmd == VFIO_DEVICE_GET_REGION_INFO) {
 		struct vfio_region_info info;
@@ -240,7 +241,8 @@ static long vfio_platform_ioctl(void *device_data,
 		info.size = vdev->regions[info.index].size;
 		info.flags = vdev->regions[info.index].flags;
 
-		return copy_to_user((void __user *)arg, &info, minsz);
+		return copy_to_user((void __user *)arg, &info, minsz) ?
+			-EFAULT : 0;
 
 	} else if (cmd == VFIO_DEVICE_GET_IRQ_INFO) {
 		struct vfio_irq_info info;
@@ -259,7 +261,8 @@ static long vfio_platform_ioctl(void *device_data,
 		info.flags = vdev->irqs[info.index].flags;
 		info.count = vdev->irqs[info.index].count;
 
-		return copy_to_user((void __user *)arg, &info, minsz);
+		return copy_to_user((void __user *)arg, &info, minsz) ?
+			-EFAULT : 0;
 
 	} else if (cmd == VFIO_DEVICE_SET_IRQS) {
 		struct vfio_irq_set hdr;
diff --git a/drivers/vfio/vfio_iommu_type1.c b/drivers/vfio/vfio_iommu_type1.c
index 6f1ea3dddbad..75b24e93cedb 100644
--- a/drivers/vfio/vfio_iommu_type1.c
+++ b/drivers/vfio/vfio_iommu_type1.c
@@ -999,7 +999,8 @@ static long vfio_iommu_type1_ioctl(void *iommu_data,
 
 		info.iova_pgsizes = vfio_pgsize_bitmap(iommu);
 
-		return copy_to_user((void __user *)arg, &info, minsz);
+		return copy_to_user((void __user *)arg, &info, minsz) ?
+			-EFAULT : 0;
 
 	} else if (cmd == VFIO_IOMMU_MAP_DMA) {
 		struct vfio_iommu_type1_dma_map map;
@@ -1032,7 +1033,8 @@ static long vfio_iommu_type1_ioctl(void *iommu_data,
 		if (ret)
 			return ret;
 
-		return copy_to_user((void __user *)arg, &unmap, minsz);
+		return copy_to_user((void __user *)arg, &unmap, minsz) ?
+			-EFAULT : 0;
 	}
 
 	return -ENOTTY;
diff --git a/drivers/vhost/vhost.c b/drivers/vhost/vhost.c
index ad2146a9ab2d..236553e81027 100644
--- a/drivers/vhost/vhost.c
+++ b/drivers/vhost/vhost.c
@@ -1156,6 +1156,8 @@ int vhost_init_used(struct vhost_virtqueue *vq)
 {
 	__virtio16 last_used_idx;
 	int r;
+	bool is_le = vq->is_le;
+
 	if (!vq->private_data) {
 		vq->is_le = virtio_legacy_is_little_endian();
 		return 0;
@@ -1165,15 +1167,20 @@ int vhost_init_used(struct vhost_virtqueue *vq)
 
 	r = vhost_update_used_flags(vq);
 	if (r)
-		return r;
+		goto err;
 	vq->signalled_used_valid = false;
-	if (!access_ok(VERIFY_READ, &vq->used->idx, sizeof vq->used->idx))
-		return -EFAULT;
+	if (!access_ok(VERIFY_READ, &vq->used->idx, sizeof vq->used->idx)) {
+		r = -EFAULT;
+		goto err;
+	}
 	r = __get_user(last_used_idx, &vq->used->idx);
 	if (r)
-		return r;
+		goto err;
 	vq->last_used_idx = vhost16_to_cpu(vq, last_used_idx);
 	return 0;
+err:
+	vq->is_le = is_le;
+	return r;
 }
 EXPORT_SYMBOL_GPL(vhost_init_used);
 
diff --git a/drivers/video/console/fbcon.c b/drivers/video/console/fbcon.c
index 92f394927f24..6e92917ba77a 100644
--- a/drivers/video/console/fbcon.c
+++ b/drivers/video/console/fbcon.c
@@ -709,6 +709,7 @@ static int con2fb_acquire_newinfo(struct vc_data *vc, struct fb_info *info,
 	}
 
 	if (!err) {
+		ops->cur_blink_jiffies = HZ / 5;
 		info->fbcon_par = ops;
 
 		if (vc)
@@ -956,6 +957,7 @@ static const char *fbcon_startup(void)
 	ops->currcon = -1;
 	ops->graphics = 1;
 	ops->cur_rotate = -1;
+	ops->cur_blink_jiffies = HZ / 5;
 	info->fbcon_par = ops;
 	p->con_rotate = initial_rotation;
 	set_blitting_type(vc, info);
diff --git a/drivers/video/fbdev/da8xx-fb.c b/drivers/video/fbdev/da8xx-fb.c
index 0081725c6b5b..6b2a06d09f2b 100644
--- a/drivers/video/fbdev/da8xx-fb.c
+++ b/drivers/video/fbdev/da8xx-fb.c
@@ -152,7 +152,7 @@ static void lcdc_write(unsigned int val, unsigned int addr)
 
 struct da8xx_fb_par {
 	struct device		*dev;
-	resource_size_t p_palette_base;
+	dma_addr_t		p_palette_base;
 	unsigned char *v_palette_base;
 	dma_addr_t		vram_phys;
 	unsigned long		vram_size;
@@ -1428,7 +1428,7 @@ static int fb_probe(struct platform_device *device)
 
 	par->vram_virt = dma_alloc_coherent(NULL,
 					    par->vram_size,
-					    (resource_size_t *) &par->vram_phys,
+					    &par->vram_phys,
 					    GFP_KERNEL | GFP_DMA);
 	if (!par->vram_virt) {
 		dev_err(&device->dev,
@@ -1448,7 +1448,7 @@ static int fb_probe(struct platform_device *device)
 
 	/* allocate palette buffer */
 	par->v_palette_base = dma_zalloc_coherent(NULL, PALETTE_SIZE,
-						  (resource_size_t *)&par->p_palette_base,
+						  &par->p_palette_base,
 						  GFP_KERNEL | GFP_DMA);
 	if (!par->v_palette_base) {
 		dev_err(&device->dev,
diff --git a/drivers/video/fbdev/exynos/s6e8ax0.c b/drivers/video/fbdev/exynos/s6e8ax0.c
index 95873f26e39c..de2f3e793786 100644
--- a/drivers/video/fbdev/exynos/s6e8ax0.c
+++ b/drivers/video/fbdev/exynos/s6e8ax0.c
@@ -829,8 +829,7 @@ static int s6e8ax0_probe(struct mipi_dsim_lcd_device *dsim_dev)
 	return 0;
 }
 
-#ifdef CONFIG_PM
-static int s6e8ax0_suspend(struct mipi_dsim_lcd_device *dsim_dev)
+static int __maybe_unused s6e8ax0_suspend(struct mipi_dsim_lcd_device *dsim_dev)
 {
 	struct s6e8ax0 *lcd = dev_get_drvdata(&dsim_dev->dev);
 
@@ -843,7 +842,7 @@ static int s6e8ax0_suspend(struct mipi_dsim_lcd_device *dsim_dev)
 	return 0;
 }
 
-static int s6e8ax0_resume(struct mipi_dsim_lcd_device *dsim_dev)
+static int __maybe_unused s6e8ax0_resume(struct mipi_dsim_lcd_device *dsim_dev)
 {
 	struct s6e8ax0 *lcd = dev_get_drvdata(&dsim_dev->dev);
 
@@ -855,10 +854,6 @@ static int s6e8ax0_resume(struct mipi_dsim_lcd_device *dsim_dev)
 
 	return 0;
 }
-#else
-#define s6e8ax0_suspend		NULL
-#define s6e8ax0_resume		NULL
-#endif
 
 static struct mipi_dsim_lcd_driver s6e8ax0_dsim_ddi_driver = {
 	.name = "s6e8ax0",
@@ -867,8 +862,8 @@ static struct mipi_dsim_lcd_driver s6e8ax0_dsim_ddi_driver = {
 	.power_on = s6e8ax0_power_on,
 	.set_sequence = s6e8ax0_set_sequence,
 	.probe = s6e8ax0_probe,
-	.suspend = s6e8ax0_suspend,
-	.resume = s6e8ax0_resume,
+	.suspend = IS_ENABLED(CONFIG_PM) ? s6e8ax0_suspend : NULL,
+	.resume = IS_ENABLED(CONFIG_PM) ? s6e8ax0_resume : NULL,
 };
 
 static int s6e8ax0_init(void)
diff --git a/drivers/video/fbdev/imxfb.c b/drivers/video/fbdev/imxfb.c
index cee88603efc9..bb2f1e866020 100644
--- a/drivers/video/fbdev/imxfb.c
+++ b/drivers/video/fbdev/imxfb.c
@@ -902,6 +902,21 @@ static int imxfb_probe(struct platform_device *pdev)
 		goto failed_getclock;
 	}
 
+	/*
+	 * The LCDC controller does not have an enable bit. The
+	 * controller starts directly when the clocks are enabled.
+	 * If the clocks are enabled when the controller is not yet
+	 * programmed with proper register values (enabled at the
+	 * bootloader, for example) then it just goes into some undefined
+	 * state.
+	 * To avoid this issue, let's enable and disable LCDC IPG clock
+	 * so that we force some kind of 'reset' to the LCDC block.
+	 */
+	ret = clk_prepare_enable(fbi->clk_ipg);
+	if (ret)
+		goto failed_getclock;
+	clk_disable_unprepare(fbi->clk_ipg);
+
 	fbi->clk_ahb = devm_clk_get(&pdev->dev, "ahb");
 	if (IS_ERR(fbi->clk_ahb)) {
 		ret = PTR_ERR(fbi->clk_ahb);
diff --git a/drivers/video/fbdev/mmp/hw/mmp_ctrl.c b/drivers/video/fbdev/mmp/hw/mmp_ctrl.c
index de54a4748065..b6f83d5df9fd 100644
--- a/drivers/video/fbdev/mmp/hw/mmp_ctrl.c
+++ b/drivers/video/fbdev/mmp/hw/mmp_ctrl.c
@@ -503,8 +503,7 @@ static int mmphw_probe(struct platform_device *pdev)
 	ctrl->reg_base = devm_ioremap_nocache(ctrl->dev,
 			res->start, resource_size(res));
 	if (ctrl->reg_base == NULL) {
-		dev_err(ctrl->dev, "%s: res %x - %x map failed\n", __func__,
-			res->start, res->end);
+		dev_err(ctrl->dev, "%s: res %pR map failed\n", __func__, res);
 		ret = -ENOMEM;
 		goto failed;
 	}
diff --git a/drivers/video/fbdev/ocfb.c b/drivers/video/fbdev/ocfb.c
index c9293aea8ec3..a970edc2a6f8 100644
--- a/drivers/video/fbdev/ocfb.c
+++ b/drivers/video/fbdev/ocfb.c
@@ -123,11 +123,11 @@ static int ocfb_setupfb(struct ocfb_dev *fbdev)
 
 	/* Horizontal timings */
 	ocfb_writereg(fbdev, OCFB_HTIM, (var->hsync_len - 1) << 24 |
-		      (var->right_margin - 1) << 16 | (var->xres - 1));
+		      (var->left_margin - 1) << 16 | (var->xres - 1));
 
 	/* Vertical timings */
 	ocfb_writereg(fbdev, OCFB_VTIM, (var->vsync_len - 1) << 24 |
-		      (var->lower_margin - 1) << 16 | (var->yres - 1));
+		      (var->upper_margin - 1) << 16 | (var->yres - 1));
 
 	/* Total length of frame */
 	hlen = var->left_margin + var->right_margin + var->hsync_len +
diff --git a/drivers/video/fbdev/omap2/omapfb/dss/dispc.h b/drivers/video/fbdev/omap2/omapfb/dss/dispc.h
index 483744223dd1..e014d0419c58 100644
--- a/drivers/video/fbdev/omap2/omapfb/dss/dispc.h
+++ b/drivers/video/fbdev/omap2/omapfb/dss/dispc.h
@@ -915,4 +915,5 @@ static inline u16 DISPC_MFLAG_THRESHOLD_OFFSET(enum omap_plane plane)
 		return 0;
 	}
 }
+
 #endif
diff --git a/drivers/video/fbdev/omap2/omapfb/dss/dss.h b/drivers/video/fbdev/omap2/omapfb/dss/dss.h
index b9066afee301..0184a8461df1 100644
--- a/drivers/video/fbdev/omap2/omapfb/dss/dss.h
+++ b/drivers/video/fbdev/omap2/omapfb/dss/dss.h
@@ -412,6 +412,44 @@ void dispc_wb_set_channel_in(enum dss_writeback_channel channel);
 int dispc_wb_setup(const struct omap_dss_writeback_info *wi,
 		bool mem_to_mem, const struct omap_video_timings *timings);
 
+u32 dispc_read_irqstatus(void);
+void dispc_clear_irqstatus(u32 mask);
+u32 dispc_read_irqenable(void);
+void dispc_write_irqenable(u32 mask);
+
+int dispc_request_irq(irq_handler_t handler, void *dev_id);
+void dispc_free_irq(void *dev_id);
+
+int dispc_runtime_get(void);
+void dispc_runtime_put(void);
+
+void dispc_mgr_enable(enum omap_channel channel, bool enable);
+bool dispc_mgr_is_enabled(enum omap_channel channel);
+u32 dispc_mgr_get_vsync_irq(enum omap_channel channel);
+u32 dispc_mgr_get_framedone_irq(enum omap_channel channel);
+u32 dispc_mgr_get_sync_lost_irq(enum omap_channel channel);
+bool dispc_mgr_go_busy(enum omap_channel channel);
+void dispc_mgr_go(enum omap_channel channel);
+void dispc_mgr_set_lcd_config(enum omap_channel channel,
+		const struct dss_lcd_mgr_config *config);
+void dispc_mgr_set_timings(enum omap_channel channel,
+		const struct omap_video_timings *timings);
+void dispc_mgr_setup(enum omap_channel channel,
+		const struct omap_overlay_manager_info *info);
+
+int dispc_ovl_check(enum omap_plane plane, enum omap_channel channel,
+		const struct omap_overlay_info *oi,
+		const struct omap_video_timings *timings,
+		int *x_predecim, int *y_predecim);
+
+int dispc_ovl_enable(enum omap_plane plane, bool enable);
+bool dispc_ovl_enabled(enum omap_plane plane);
+void dispc_ovl_set_channel_out(enum omap_plane plane,
+		enum omap_channel channel);
+int dispc_ovl_setup(enum omap_plane plane, const struct omap_overlay_info *oi,
+		bool replication, const struct omap_video_timings *mgr_timings,
+		bool mem_to_mem);
+
 /* VENC */
 int venc_init_platform_driver(void) __init;
 void venc_uninit_platform_driver(void);
@@ -465,4 +503,44 @@ int dss_pll_write_config_type_b(struct dss_pll *pll,
 		const struct dss_pll_clock_info *cinfo);
 int dss_pll_wait_reset_done(struct dss_pll *pll);
 
+/* compat */
+
+struct dss_mgr_ops {
+	int (*connect)(struct omap_overlay_manager *mgr,
+		struct omap_dss_device *dst);
+	void (*disconnect)(struct omap_overlay_manager *mgr,
+		struct omap_dss_device *dst);
+
+	void (*start_update)(struct omap_overlay_manager *mgr);
+	int (*enable)(struct omap_overlay_manager *mgr);
+	void (*disable)(struct omap_overlay_manager *mgr);
+	void (*set_timings)(struct omap_overlay_manager *mgr,
+			const struct omap_video_timings *timings);
+	void (*set_lcd_config)(struct omap_overlay_manager *mgr,
+			const struct dss_lcd_mgr_config *config);
+	int (*register_framedone_handler)(struct omap_overlay_manager *mgr,
+			void (*handler)(void *), void *data);
+	void (*unregister_framedone_handler)(struct omap_overlay_manager *mgr,
+			void (*handler)(void *), void *data);
+};
+
+int dss_install_mgr_ops(const struct dss_mgr_ops *mgr_ops);
+void dss_uninstall_mgr_ops(void);
+
+int dss_mgr_connect(struct omap_overlay_manager *mgr,
+		struct omap_dss_device *dst);
+void dss_mgr_disconnect(struct omap_overlay_manager *mgr,
+		struct omap_dss_device *dst);
+void dss_mgr_set_timings(struct omap_overlay_manager *mgr,
+		const struct omap_video_timings *timings);
+void dss_mgr_set_lcd_config(struct omap_overlay_manager *mgr,
+		const struct dss_lcd_mgr_config *config);
+int dss_mgr_enable(struct omap_overlay_manager *mgr);
+void dss_mgr_disable(struct omap_overlay_manager *mgr);
+void dss_mgr_start_update(struct omap_overlay_manager *mgr);
+int dss_mgr_register_framedone_handler(struct omap_overlay_manager *mgr,
+		void (*handler)(void *), void *data);
+void dss_mgr_unregister_framedone_handler(struct omap_overlay_manager *mgr,
+		void (*handler)(void *), void *data);
+
 #endif
diff --git a/drivers/virtio/virtio_pci_modern.c b/drivers/virtio/virtio_pci_modern.c
index c0c11fad4611..7760fc1a2218 100644
--- a/drivers/virtio/virtio_pci_modern.c
+++ b/drivers/virtio/virtio_pci_modern.c
@@ -679,7 +679,7 @@ int virtio_pci_modern_probe(struct virtio_pci_device *vp_dev)
 
 	pci_read_config_dword(pci_dev,
 			      notify + offsetof(struct virtio_pci_notify_cap,
-						cap.length),
+						cap.offset),
 			      &notify_offset);
 
 	/* We don't know how many VQs we'll map, ahead of the time.
diff --git a/drivers/watchdog/Kconfig b/drivers/watchdog/Kconfig
index 0f6d8515ba4f..80825a7e8e48 100644
--- a/drivers/watchdog/Kconfig
+++ b/drivers/watchdog/Kconfig
@@ -1569,6 +1569,17 @@ config WATCHDOG_RIO
 	  machines.  The watchdog timeout period is normally one minute but
 	  can be changed with a boot-time parameter.
 
+config WATCHDOG_SUN4V
+	tristate "Sun4v Watchdog support"
+	select WATCHDOG_CORE
+	depends on SPARC64
+	help
+	  Say Y here to support the hypervisor watchdog capability embedded
+	  in the SPARC sun4v architecture.
+
+	  To compile this driver as a module, choose M here. The module will
+	  be called sun4v_wdt.
+
 # XTENSA Architecture
 
 # Xen Architecture
diff --git a/drivers/watchdog/Makefile b/drivers/watchdog/Makefile
index f566753256ab..f6a6a387c6c7 100644
--- a/drivers/watchdog/Makefile
+++ b/drivers/watchdog/Makefile
@@ -179,6 +179,7 @@ obj-$(CONFIG_SH_WDT) += shwdt.o
 
 obj-$(CONFIG_WATCHDOG_RIO)		+= riowd.o
 obj-$(CONFIG_WATCHDOG_CP1XXX)		+= cpwd.o
+obj-$(CONFIG_WATCHDOG_SUN4V)		+= sun4v_wdt.o
 
 # XTENSA Architecture
 
diff --git a/drivers/watchdog/sun4v_wdt.c b/drivers/watchdog/sun4v_wdt.c
new file mode 100644
index 000000000000..1467fe50a76f
--- /dev/null
+++ b/drivers/watchdog/sun4v_wdt.c
@@ -0,0 +1,191 @@
+/*
+ *	sun4v watchdog timer
+ *	(c) Copyright 2016 Oracle Corporation
+ *
+ *	Implement a simple watchdog driver using the built-in sun4v hypervisor
+ *	watchdog support. If time expires, the hypervisor stops or bounces
+ *	the guest domain.
+ *
+ *	This program is free software; you can redistribute it and/or
+ *	modify it under the terms of the GNU General Public License
+ *	as published by the Free Software Foundation; either version
+ *	2 of the License, or (at your option) any later version.
+ */
+
+#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt
+
+#include <linux/errno.h>
+#include <linux/init.h>
+#include <linux/kernel.h>
+#include <linux/module.h>
+#include <linux/moduleparam.h>
+#include <linux/watchdog.h>
+#include <asm/hypervisor.h>
+#include <asm/mdesc.h>
+
+#define WDT_TIMEOUT			60
+#define WDT_MAX_TIMEOUT			31536000
+#define WDT_MIN_TIMEOUT			1
+#define WDT_DEFAULT_RESOLUTION_MS	1000	/* 1 second */
+
+static unsigned int timeout;
+module_param(timeout, uint, 0);
+MODULE_PARM_DESC(timeout, "Watchdog timeout in seconds (default="
+	__MODULE_STRING(WDT_TIMEOUT) ")");
+
+static bool nowayout = WATCHDOG_NOWAYOUT;
+module_param(nowayout, bool, S_IRUGO);
+MODULE_PARM_DESC(nowayout, "Watchdog cannot be stopped once started (default="
+	__MODULE_STRING(WATCHDOG_NOWAYOUT) ")");
+
+static int sun4v_wdt_stop(struct watchdog_device *wdd)
+{
+	sun4v_mach_set_watchdog(0, NULL);
+
+	return 0;
+}
+
+static int sun4v_wdt_ping(struct watchdog_device *wdd)
+{
+	int hverr;
+
+	/*
+	 * HV watchdog timer will round up the timeout
+	 * passed in to the nearest multiple of the
+	 * watchdog resolution in milliseconds.
+	 */
+	hverr = sun4v_mach_set_watchdog(wdd->timeout * 1000, NULL);
+	if (hverr == HV_EINVAL)
+		return -EINVAL;
+
+	return 0;
+}
+
+static int sun4v_wdt_set_timeout(struct watchdog_device *wdd,
+				 unsigned int timeout)
+{
+	wdd->timeout = timeout;
+
+	return 0;
+}
+
+static const struct watchdog_info sun4v_wdt_ident = {
+	.options =	WDIOF_SETTIMEOUT |
+			WDIOF_MAGICCLOSE |
+			WDIOF_KEEPALIVEPING,
+	.identity =	"sun4v hypervisor watchdog",
+	.firmware_version = 0,
+};
+
+static struct watchdog_ops sun4v_wdt_ops = {
+	.owner =	THIS_MODULE,
+	.start =	sun4v_wdt_ping,
+	.stop =		sun4v_wdt_stop,
+	.ping =		sun4v_wdt_ping,
+	.set_timeout =	sun4v_wdt_set_timeout,
+};
+
+static struct watchdog_device wdd = {
+	.info = &sun4v_wdt_ident,
+	.ops = &sun4v_wdt_ops,
+	.min_timeout = WDT_MIN_TIMEOUT,
+	.max_timeout = WDT_MAX_TIMEOUT,
+	.timeout = WDT_TIMEOUT,
+};
+
+static int __init sun4v_wdt_init(void)
+{
+	struct mdesc_handle *handle;
+	u64 node;
+	const u64 *value;
+	int err = 0;
+	unsigned long major = 1, minor = 1;
+
+	/*
+	 * There are 2 properties that can be set from the control
+	 * domain for the watchdog.
+	 * watchdog-resolution
+	 * watchdog-max-timeout
+	 *
+	 * We can expect a handle to be returned otherwise something
+	 * serious is wrong. Correct to return -ENODEV here.
+	 */
+
+	handle = mdesc_grab();
+	if (!handle)
+		return -ENODEV;
+
+	node = mdesc_node_by_name(handle, MDESC_NODE_NULL, "platform");
+	err = -ENODEV;
+	if (node == MDESC_NODE_NULL)
+		goto out_release;
+
+	/*
+	 * This is a safe way to validate if we are on the right
+	 * platform.
+	 */
+	if (sun4v_hvapi_register(HV_GRP_CORE, major, &minor))
+		goto out_hv_unreg;
+
+	/* Allow value of watchdog-resolution up to 1s (default) */
+	value = mdesc_get_property(handle, node, "watchdog-resolution", NULL);
+	err = -EINVAL;
+	if (value) {
+		if (*value == 0 ||
+		    *value > WDT_DEFAULT_RESOLUTION_MS)
+			goto out_hv_unreg;
+	}
+
+	value = mdesc_get_property(handle, node, "watchdog-max-timeout", NULL);
+	if (value) {
+		/*
+		 * If the property value (in ms) is smaller than
+		 * min_timeout, return -EINVAL.
+		 */
+		if (*value < wdd.min_timeout * 1000)
+			goto out_hv_unreg;
+
+		/*
+		 * If the property value is smaller than
+		 * default max_timeout  then set watchdog max_timeout to
+		 * the value of the property in seconds.
+		 */
+		if (*value < wdd.max_timeout * 1000)
+			wdd.max_timeout = *value  / 1000;
+	}
+
+	watchdog_init_timeout(&wdd, timeout, NULL);
+
+	watchdog_set_nowayout(&wdd, nowayout);
+
+	err = watchdog_register_device(&wdd);
+	if (err)
+		goto out_hv_unreg;
+
+	pr_info("initialized (timeout=%ds, nowayout=%d)\n",
+		 wdd.timeout, nowayout);
+
+	mdesc_release(handle);
+
+	return 0;
+
+out_hv_unreg:
+	sun4v_hvapi_unregister(HV_GRP_CORE);
+
+out_release:
+	mdesc_release(handle);
+	return err;
+}
+
+static void __exit sun4v_wdt_exit(void)
+{
+	sun4v_hvapi_unregister(HV_GRP_CORE);
+	watchdog_unregister_device(&wdd);
+}
+
+module_init(sun4v_wdt_init);
+module_exit(sun4v_wdt_exit);
+
+MODULE_AUTHOR("Wim Coekaerts <wim.coekaerts@oracle.com>");
+MODULE_DESCRIPTION("sun4v watchdog driver");
+MODULE_LICENSE("GPL");
diff --git a/drivers/xen/xen-pciback/pciback_ops.c b/drivers/xen/xen-pciback/pciback_ops.c
index 73dafdc494aa..fb0221434f81 100644
--- a/drivers/xen/xen-pciback/pciback_ops.c
+++ b/drivers/xen/xen-pciback/pciback_ops.c
@@ -227,8 +227,9 @@ int xen_pcibk_enable_msix(struct xen_pcibk_device *pdev,
 	/*
 	 * PCI_COMMAND_MEMORY must be enabled, otherwise we may not be able
 	 * to access the BARs where the MSI-X entries reside.
+	 * But VF devices are unique in which the PF needs to be checked.
 	 */
-	pci_read_config_word(dev, PCI_COMMAND, &cmd);
+	pci_read_config_word(pci_physfn(dev), PCI_COMMAND, &cmd);
 	if (dev->msi_enabled || !(cmd & PCI_COMMAND_MEMORY))
 		return -ENXIO;
 
@@ -332,6 +333,9 @@ void xen_pcibk_do_op(struct work_struct *data)
 	struct xen_pcibk_dev_data *dev_data = NULL;
 	struct xen_pci_op *op = &pdev->op;
 	int test_intx = 0;
+#ifdef CONFIG_PCI_MSI
+	unsigned int nr = 0;
+#endif
 
 	*op = pdev->sh_info->op;
 	barrier();
@@ -360,6 +364,7 @@ void xen_pcibk_do_op(struct work_struct *data)
 			op->err = xen_pcibk_disable_msi(pdev, dev, op);
 			break;
 		case XEN_PCI_OP_enable_msix:
+			nr = op->value;
 			op->err = xen_pcibk_enable_msix(pdev, dev, op);
 			break;
 		case XEN_PCI_OP_disable_msix:
@@ -382,7 +387,7 @@ void xen_pcibk_do_op(struct work_struct *data)
 	if (op->cmd == XEN_PCI_OP_enable_msix && op->err == 0) {
 		unsigned int i;
 
-		for (i = 0; i < op->value; i++)
+		for (i = 0; i < nr; i++)
 			pdev->sh_info->op.msix_entries[i].vector =
 				op->msix_entries[i].vector;
 	}
diff --git a/drivers/xen/xen-scsiback.c b/drivers/xen/xen-scsiback.c
index ad4eb1024d1f..c46ee189466f 100644
--- a/drivers/xen/xen-scsiback.c
+++ b/drivers/xen/xen-scsiback.c
@@ -849,15 +849,31 @@ static int scsiback_map(struct vscsibk_info *info)
 }
 
 /*
+  Check for a translation entry being present
+*/
+static struct v2p_entry *scsiback_chk_translation_entry(
+	struct vscsibk_info *info, struct ids_tuple *v)
+{
+	struct list_head *head = &(info->v2p_entry_lists);
+	struct v2p_entry *entry;
+
+	list_for_each_entry(entry, head, l)
+		if ((entry->v.chn == v->chn) &&
+		    (entry->v.tgt == v->tgt) &&
+		    (entry->v.lun == v->lun))
+			return entry;
+
+	return NULL;
+}
+
+/*
   Add a new translation entry
 */
 static int scsiback_add_translation_entry(struct vscsibk_info *info,
 					  char *phy, struct ids_tuple *v)
 {
 	int err = 0;
-	struct v2p_entry *entry;
 	struct v2p_entry *new;
-	struct list_head *head = &(info->v2p_entry_lists);
 	unsigned long flags;
 	char *lunp;
 	unsigned long long unpacked_lun;
@@ -917,15 +933,10 @@ static int scsiback_add_translation_entry(struct vscsibk_info *info,
 	spin_lock_irqsave(&info->v2p_lock, flags);
 
 	/* Check double assignment to identical virtual ID */
-	list_for_each_entry(entry, head, l) {
-		if ((entry->v.chn == v->chn) &&
-		    (entry->v.tgt == v->tgt) &&
-		    (entry->v.lun == v->lun)) {
-			pr_warn("Virtual ID is already used. Assignment was not performed.\n");
-			err = -EEXIST;
-			goto out;
-		}
-
+	if (scsiback_chk_translation_entry(info, v)) {
+		pr_warn("Virtual ID is already used. Assignment was not performed.\n");
+		err = -EEXIST;
+		goto out;
 	}
 
 	/* Create a new translation entry and add to the list */
@@ -933,18 +944,18 @@ static int scsiback_add_translation_entry(struct vscsibk_info *info,
 	new->v = *v;
 	new->tpg = tpg;
 	new->lun = unpacked_lun;
-	list_add_tail(&new->l, head);
+	list_add_tail(&new->l, &info->v2p_entry_lists);
 
 out:
 	spin_unlock_irqrestore(&info->v2p_lock, flags);
 
 out_free:
-	mutex_lock(&tpg->tv_tpg_mutex);
-	tpg->tv_tpg_fe_count--;
-	mutex_unlock(&tpg->tv_tpg_mutex);
-
-	if (err)
+	if (err) {
+		mutex_lock(&tpg->tv_tpg_mutex);
+		tpg->tv_tpg_fe_count--;
+		mutex_unlock(&tpg->tv_tpg_mutex);
 		kfree(new);
+	}
 
 	return err;
 }
@@ -956,39 +967,40 @@ static void __scsiback_del_translation_entry(struct v2p_entry *entry)
 }
 
 /*
-  Delete the translation entry specfied
+  Delete the translation entry specified
 */
 static int scsiback_del_translation_entry(struct vscsibk_info *info,
 					  struct ids_tuple *v)
 {
 	struct v2p_entry *entry;
-	struct list_head *head = &(info->v2p_entry_lists);
 	unsigned long flags;
+	int ret = 0;
 
 	spin_lock_irqsave(&info->v2p_lock, flags);
 	/* Find out the translation entry specified */
-	list_for_each_entry(entry, head, l) {
-		if ((entry->v.chn == v->chn) &&
-		    (entry->v.tgt == v->tgt) &&
-		    (entry->v.lun == v->lun)) {
-			goto found;
-		}
-	}
-
-	spin_unlock_irqrestore(&info->v2p_lock, flags);
-	return 1;
-
-found:
-	/* Delete the translation entry specfied */
-	__scsiback_del_translation_entry(entry);
+	entry = scsiback_chk_translation_entry(info, v);
+	if (entry)
+		__scsiback_del_translation_entry(entry);
+	else
+		ret = -ENOENT;
 
 	spin_unlock_irqrestore(&info->v2p_lock, flags);
-	return 0;
+	return ret;
 }
 
 static void scsiback_do_add_lun(struct vscsibk_info *info, const char *state,
 				char *phy, struct ids_tuple *vir, int try)
 {
+	struct v2p_entry *entry;
+	unsigned long flags;
+
+	if (try) {
+		spin_lock_irqsave(&info->v2p_lock, flags);
+		entry = scsiback_chk_translation_entry(info, vir);
+		spin_unlock_irqrestore(&info->v2p_lock, flags);
+		if (entry)
+			return;
+	}
 	if (!scsiback_add_translation_entry(info, phy, vir)) {
 		if (xenbus_printf(XBT_NIL, info->dev->nodename, state,
 				  "%d", XenbusStateInitialised)) {
diff --git a/drivers/xen/xenbus/xenbus_dev_frontend.c b/drivers/xen/xenbus/xenbus_dev_frontend.c
index 9433e46518c8..912b64edb42b 100644
--- a/drivers/xen/xenbus/xenbus_dev_frontend.c
+++ b/drivers/xen/xenbus/xenbus_dev_frontend.c
@@ -188,6 +188,8 @@ static int queue_reply(struct list_head *queue, const void *data, size_t len)
 
 	if (len == 0)
 		return 0;
+	if (len > XENSTORE_PAYLOAD_MAX)
+		return -EINVAL;
 
 	rb = kmalloc(sizeof(*rb) + len, GFP_KERNEL);
 	if (rb == NULL)
diff --git a/fs/affs/file.c b/fs/affs/file.c
index 0548c53f41d5..22fc7c802d69 100644
--- a/fs/affs/file.c
+++ b/fs/affs/file.c
@@ -511,8 +511,6 @@ affs_do_readpage_ofs(struct page *page, unsigned to)
 	pr_debug("%s(%lu, %ld, 0, %d)\n", __func__, inode->i_ino,
 		 page->index, to);
 	BUG_ON(to > PAGE_CACHE_SIZE);
-	kmap(page);
-	data = page_address(page);
 	bsize = AFFS_SB(sb)->s_data_blksize;
 	tmp = page->index << PAGE_CACHE_SHIFT;
 	bidx = tmp / bsize;
@@ -524,14 +522,15 @@ affs_do_readpage_ofs(struct page *page, unsigned to)
 			return PTR_ERR(bh);
 		tmp = min(bsize - boff, to - pos);
 		BUG_ON(pos + tmp > to || tmp > bsize);
+		data = kmap_atomic(page);
 		memcpy(data + pos, AFFS_DATA(bh) + boff, tmp);
+		kunmap_atomic(data);
 		affs_brelse(bh);
 		bidx++;
 		pos += tmp;
 		boff = 0;
 	}
 	flush_dcache_page(page);
-	kunmap(page);
 	return 0;
 }
 
diff --git a/fs/binfmt_elf.c b/fs/binfmt_elf.c
index 051ea4809c14..7d914c67a9d0 100644
--- a/fs/binfmt_elf.c
+++ b/fs/binfmt_elf.c
@@ -653,7 +653,7 @@ static unsigned long randomize_stack_top(unsigned long stack_top)
 
 	if ((current->flags & PF_RANDOMIZE) &&
 		!(current->personality & ADDR_NO_RANDOMIZE)) {
-		random_variable = (unsigned long) get_random_int();
+		random_variable = get_random_long();
 		random_variable &= STACK_RND_MASK;
 		random_variable <<= PAGE_SHIFT;
 	}
diff --git a/fs/block_dev.c b/fs/block_dev.c
index 39b3a174a425..826b164a4b5b 100644
--- a/fs/block_dev.c
+++ b/fs/block_dev.c
@@ -1201,7 +1201,11 @@ static int __blkdev_get(struct block_device *bdev, fmode_t mode, int for_part)
 		bdev->bd_disk = disk;
 		bdev->bd_queue = disk->queue;
 		bdev->bd_contains = bdev;
-		bdev->bd_inode->i_flags = disk->fops->direct_access ? S_DAX : 0;
+		if (IS_ENABLED(CONFIG_BLK_DEV_DAX) && disk->fops->direct_access)
+			bdev->bd_inode->i_flags = S_DAX;
+		else
+			bdev->bd_inode->i_flags = 0;
+
 		if (!partno) {
 			ret = -ENXIO;
 			bdev->bd_part = disk_get_part(disk, partno);
@@ -1693,13 +1697,24 @@ static int blkdev_releasepage(struct page *page, gfp_t wait)
 	return try_to_free_buffers(page);
 }
 
+static int blkdev_writepages(struct address_space *mapping,
+			     struct writeback_control *wbc)
+{
+	if (dax_mapping(mapping)) {
+		struct block_device *bdev = I_BDEV(mapping->host);
+
+		return dax_writeback_mapping_range(mapping, bdev, wbc);
+	}
+	return generic_writepages(mapping, wbc);
+}
+
 static const struct address_space_operations def_blk_aops = {
 	.readpage	= blkdev_readpage,
 	.readpages	= blkdev_readpages,
 	.writepage	= blkdev_writepage,
 	.write_begin	= blkdev_write_begin,
 	.write_end	= blkdev_write_end,
-	.writepages	= generic_writepages,
+	.writepages	= blkdev_writepages,
 	.releasepage	= blkdev_releasepage,
 	.direct_IO	= blkdev_direct_IO,
 	.is_dirty_writeback = buffer_check_dirty_writeback,
diff --git a/fs/btrfs/backref.c b/fs/btrfs/backref.c
index b90cd3776f8e..f6dac40f87ff 100644
--- a/fs/btrfs/backref.c
+++ b/fs/btrfs/backref.c
@@ -1406,7 +1406,8 @@ char *btrfs_ref_to_path(struct btrfs_root *fs_root, struct btrfs_path *path,
 			read_extent_buffer(eb, dest + bytes_left,
 					   name_off, name_len);
 		if (eb != eb_in) {
-			btrfs_tree_read_unlock_blocking(eb);
+			if (!path->skip_locking)
+				btrfs_tree_read_unlock_blocking(eb);
 			free_extent_buffer(eb);
 		}
 		ret = btrfs_find_item(fs_root, path, parent, 0,
@@ -1426,9 +1427,10 @@ char *btrfs_ref_to_path(struct btrfs_root *fs_root, struct btrfs_path *path,
 		eb = path->nodes[0];
 		/* make sure we can use eb after releasing the path */
 		if (eb != eb_in) {
-			atomic_inc(&eb->refs);
-			btrfs_tree_read_lock(eb);
-			btrfs_set_lock_blocking_rw(eb, BTRFS_READ_LOCK);
+			if (!path->skip_locking)
+				btrfs_set_lock_blocking_rw(eb, BTRFS_READ_LOCK);
+			path->nodes[0] = NULL;
+			path->locks[0] = 0;
 		}
 		btrfs_release_path(path);
 		iref = btrfs_item_ptr(eb, slot, struct btrfs_inode_ref);
diff --git a/fs/btrfs/compression.c b/fs/btrfs/compression.c
index c473c42d7d6c..3346cd8f9910 100644
--- a/fs/btrfs/compression.c
+++ b/fs/btrfs/compression.c
@@ -637,11 +637,7 @@ int btrfs_submit_compressed_read(struct inode *inode, struct bio *bio,
 	faili = nr_pages - 1;
 	cb->nr_pages = nr_pages;
 
-	/* In the parent-locked case, we only locked the range we are
-	 * interested in.  In all other cases, we can opportunistically
-	 * cache decompressed data that goes beyond the requested range. */
-	if (!(bio_flags & EXTENT_BIO_PARENT_LOCKED))
-		add_ra_bio_pages(inode, em_start + em_len, cb);
+	add_ra_bio_pages(inode, em_start + em_len, cb);
 
 	/* include any pages we added in add_ra-bio_pages */
 	uncompressed_len = bio->bi_vcnt * PAGE_CACHE_SIZE;
diff --git a/fs/btrfs/delayed-inode.c b/fs/btrfs/delayed-inode.c
index 0be47e4b8136..b57daa895cea 100644
--- a/fs/btrfs/delayed-inode.c
+++ b/fs/btrfs/delayed-inode.c
@@ -1689,7 +1689,7 @@ int btrfs_should_delete_dir_index(struct list_head *del_list,
  *
  */
 int btrfs_readdir_delayed_dir_index(struct dir_context *ctx,
-				    struct list_head *ins_list)
+				    struct list_head *ins_list, bool *emitted)
 {
 	struct btrfs_dir_item *di;
 	struct btrfs_delayed_item *curr, *next;
@@ -1733,6 +1733,7 @@ int btrfs_readdir_delayed_dir_index(struct dir_context *ctx,
 
 		if (over)
 			return 1;
+		*emitted = true;
 	}
 	return 0;
 }
diff --git a/fs/btrfs/delayed-inode.h b/fs/btrfs/delayed-inode.h
index f70119f25421..0167853c84ae 100644
--- a/fs/btrfs/delayed-inode.h
+++ b/fs/btrfs/delayed-inode.h
@@ -144,7 +144,7 @@ void btrfs_put_delayed_items(struct list_head *ins_list,
 int btrfs_should_delete_dir_index(struct list_head *del_list,
 				  u64 index);
 int btrfs_readdir_delayed_dir_index(struct dir_context *ctx,
-				    struct list_head *ins_list);
+				    struct list_head *ins_list, bool *emitted);
 
 /* for init */
 int __init btrfs_delayed_inode_init(void);
diff --git a/fs/btrfs/extent_io.c b/fs/btrfs/extent_io.c
index 2e7c97a3f344..392592dc7010 100644
--- a/fs/btrfs/extent_io.c
+++ b/fs/btrfs/extent_io.c
@@ -2897,12 +2897,11 @@ static int __do_readpage(struct extent_io_tree *tree,
 	struct block_device *bdev;
 	int ret;
 	int nr = 0;
-	int parent_locked = *bio_flags & EXTENT_BIO_PARENT_LOCKED;
 	size_t pg_offset = 0;
 	size_t iosize;
 	size_t disk_io_size;
 	size_t blocksize = inode->i_sb->s_blocksize;
-	unsigned long this_bio_flag = *bio_flags & EXTENT_BIO_PARENT_LOCKED;
+	unsigned long this_bio_flag = 0;
 
 	set_page_extent_mapped(page);
 
@@ -2942,18 +2941,16 @@ static int __do_readpage(struct extent_io_tree *tree,
 			kunmap_atomic(userpage);
 			set_extent_uptodate(tree, cur, cur + iosize - 1,
 					    &cached, GFP_NOFS);
-			if (!parent_locked)
-				unlock_extent_cached(tree, cur,
-						     cur + iosize - 1,
-						     &cached, GFP_NOFS);
+			unlock_extent_cached(tree, cur,
+					     cur + iosize - 1,
+					     &cached, GFP_NOFS);
 			break;
 		}
 		em = __get_extent_map(inode, page, pg_offset, cur,
 				      end - cur + 1, get_extent, em_cached);
 		if (IS_ERR_OR_NULL(em)) {
 			SetPageError(page);
-			if (!parent_locked)
-				unlock_extent(tree, cur, end);
+			unlock_extent(tree, cur, end);
 			break;
 		}
 		extent_offset = cur - em->start;
@@ -3038,12 +3035,9 @@ static int __do_readpage(struct extent_io_tree *tree,
 
 			set_extent_uptodate(tree, cur, cur + iosize - 1,
 					    &cached, GFP_NOFS);
-			if (parent_locked)
-				free_extent_state(cached);
-			else
-				unlock_extent_cached(tree, cur,
-						     cur + iosize - 1,
-						     &cached, GFP_NOFS);
+			unlock_extent_cached(tree, cur,
+					     cur + iosize - 1,
+					     &cached, GFP_NOFS);
 			cur = cur + iosize;
 			pg_offset += iosize;
 			continue;
@@ -3052,8 +3046,7 @@ static int __do_readpage(struct extent_io_tree *tree,
 		if (test_range_bit(tree, cur, cur_end,
 				   EXTENT_UPTODATE, 1, NULL)) {
 			check_page_uptodate(tree, page);
-			if (!parent_locked)
-				unlock_extent(tree, cur, cur + iosize - 1);
+			unlock_extent(tree, cur, cur + iosize - 1);
 			cur = cur + iosize;
 			pg_offset += iosize;
 			continue;
@@ -3063,8 +3056,7 @@ static int __do_readpage(struct extent_io_tree *tree,
 		 */
 		if (block_start == EXTENT_MAP_INLINE) {
 			SetPageError(page);
-			if (!parent_locked)
-				unlock_extent(tree, cur, cur + iosize - 1);
+			unlock_extent(tree, cur, cur + iosize - 1);
 			cur = cur + iosize;
 			pg_offset += iosize;
 			continue;
@@ -3083,8 +3075,7 @@ static int __do_readpage(struct extent_io_tree *tree,
 			*bio_flags = this_bio_flag;
 		} else {
 			SetPageError(page);
-			if (!parent_locked)
-				unlock_extent(tree, cur, cur + iosize - 1);
+			unlock_extent(tree, cur, cur + iosize - 1);
 		}
 		cur = cur + iosize;
 		pg_offset += iosize;
@@ -3213,20 +3204,6 @@ int extent_read_full_page(struct extent_io_tree *tree, struct page *page,
 	return ret;
 }
 
-int extent_read_full_page_nolock(struct extent_io_tree *tree, struct page *page,
-				 get_extent_t *get_extent, int mirror_num)
-{
-	struct bio *bio = NULL;
-	unsigned long bio_flags = EXTENT_BIO_PARENT_LOCKED;
-	int ret;
-
-	ret = __do_readpage(tree, page, get_extent, NULL, &bio, mirror_num,
-			    &bio_flags, READ, NULL);
-	if (bio)
-		ret = submit_one_bio(READ, bio, mirror_num, bio_flags);
-	return ret;
-}
-
 static noinline void update_nr_written(struct page *page,
 				      struct writeback_control *wbc,
 				      unsigned long nr_written)
diff --git a/fs/btrfs/extent_io.h b/fs/btrfs/extent_io.h
index 0377413bd4b9..880d5292e972 100644
--- a/fs/btrfs/extent_io.h
+++ b/fs/btrfs/extent_io.h
@@ -29,7 +29,6 @@
  */
 #define EXTENT_BIO_COMPRESSED 1
 #define EXTENT_BIO_TREE_LOG 2
-#define EXTENT_BIO_PARENT_LOCKED 4
 #define EXTENT_BIO_FLAG_SHIFT 16
 
 /* these are bit numbers for test/set bit */
@@ -210,8 +209,6 @@ static inline int lock_extent(struct extent_io_tree *tree, u64 start, u64 end)
 int try_lock_extent(struct extent_io_tree *tree, u64 start, u64 end);
 int extent_read_full_page(struct extent_io_tree *tree, struct page *page,
 			  get_extent_t *get_extent, int mirror_num);
-int extent_read_full_page_nolock(struct extent_io_tree *tree, struct page *page,
-				 get_extent_t *get_extent, int mirror_num);
 int __init extent_io_init(void);
 void extent_io_exit(void);
 
diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c
index 5f06eb1f4384..d96f5cf38a2d 100644
--- a/fs/btrfs/inode.c
+++ b/fs/btrfs/inode.c
@@ -5717,6 +5717,7 @@ static int btrfs_real_readdir(struct file *file, struct dir_context *ctx)
 	char *name_ptr;
 	int name_len;
 	int is_curr = 0;	/* ctx->pos points to the current index? */
+	bool emitted;
 
 	/* FIXME, use a real flag for deciding about the key type */
 	if (root->fs_info->tree_root == root)
@@ -5745,6 +5746,7 @@ static int btrfs_real_readdir(struct file *file, struct dir_context *ctx)
 	if (ret < 0)
 		goto err;
 
+	emitted = false;
 	while (1) {
 		leaf = path->nodes[0];
 		slot = path->slots[0];
@@ -5824,6 +5826,7 @@ skip:
 
 			if (over)
 				goto nopos;
+			emitted = true;
 			di_len = btrfs_dir_name_len(leaf, di) +
 				 btrfs_dir_data_len(leaf, di) + sizeof(*di);
 			di_cur += di_len;
@@ -5836,11 +5839,20 @@ next:
 	if (key_type == BTRFS_DIR_INDEX_KEY) {
 		if (is_curr)
 			ctx->pos++;
-		ret = btrfs_readdir_delayed_dir_index(ctx, &ins_list);
+		ret = btrfs_readdir_delayed_dir_index(ctx, &ins_list, &emitted);
 		if (ret)
 			goto nopos;
 	}
 
+	/*
+	 * If we haven't emitted any dir entry, we must not touch ctx->pos as
+	 * it was was set to the termination value in previous call. We assume
+	 * that "." and ".." were emitted if we reach this point and set the
+	 * termination value as well for an empty directory.
+	 */
+	if (ctx->pos > 2 && !emitted)
+		goto nopos;
+
 	/* Reached end of directory/root. Bump pos past the last item. */
 	ctx->pos++;
 
@@ -7974,6 +7986,7 @@ static void btrfs_endio_direct_read(struct bio *bio)
 
 	kfree(dip);
 
+	dio_bio->bi_error = bio->bi_error;
 	dio_end_io(dio_bio, bio->bi_error);
 
 	if (io_bio->end_io)
@@ -8028,6 +8041,7 @@ static void btrfs_endio_direct_write(struct bio *bio)
 
 	kfree(dip);
 
+	dio_bio->bi_error = bio->bi_error;
 	dio_end_io(dio_bio, bio->bi_error);
 	bio_put(bio);
 }
diff --git a/fs/btrfs/ioctl.c b/fs/btrfs/ioctl.c
index 952172ca7e45..48aee9846329 100644
--- a/fs/btrfs/ioctl.c
+++ b/fs/btrfs/ioctl.c
@@ -2794,24 +2794,29 @@ out:
 static struct page *extent_same_get_page(struct inode *inode, pgoff_t index)
 {
 	struct page *page;
-	struct extent_io_tree *tree = &BTRFS_I(inode)->io_tree;
 
 	page = grab_cache_page(inode->i_mapping, index);
 	if (!page)
-		return NULL;
+		return ERR_PTR(-ENOMEM);
 
 	if (!PageUptodate(page)) {
-		if (extent_read_full_page_nolock(tree, page, btrfs_get_extent,
-						 0))
-			return NULL;
+		int ret;
+
+		ret = btrfs_readpage(NULL, page);
+		if (ret)
+			return ERR_PTR(ret);
 		lock_page(page);
 		if (!PageUptodate(page)) {
 			unlock_page(page);
 			page_cache_release(page);
-			return NULL;
+			return ERR_PTR(-EIO);
+		}
+		if (page->mapping != inode->i_mapping) {
+			unlock_page(page);
+			page_cache_release(page);
+			return ERR_PTR(-EAGAIN);
 		}
 	}
-	unlock_page(page);
 
 	return page;
 }
@@ -2823,17 +2828,31 @@ static int gather_extent_pages(struct inode *inode, struct page **pages,
 	pgoff_t index = off >> PAGE_CACHE_SHIFT;
 
 	for (i = 0; i < num_pages; i++) {
+again:
 		pages[i] = extent_same_get_page(inode, index + i);
-		if (!pages[i])
-			return -ENOMEM;
+		if (IS_ERR(pages[i])) {
+			int err = PTR_ERR(pages[i]);
+
+			if (err == -EAGAIN)
+				goto again;
+			pages[i] = NULL;
+			return err;
+		}
 	}
 	return 0;
 }
 
-static inline void lock_extent_range(struct inode *inode, u64 off, u64 len)
+static int lock_extent_range(struct inode *inode, u64 off, u64 len,
+			     bool retry_range_locking)
 {
-	/* do any pending delalloc/csum calc on src, one way or
-	   another, and lock file content */
+	/*
+	 * Do any pending delalloc/csum calculations on inode, one way or
+	 * another, and lock file content.
+	 * The locking order is:
+	 *
+	 *   1) pages
+	 *   2) range in the inode's io tree
+	 */
 	while (1) {
 		struct btrfs_ordered_extent *ordered;
 		lock_extent(&BTRFS_I(inode)->io_tree, off, off + len - 1);
@@ -2851,8 +2870,11 @@ static inline void lock_extent_range(struct inode *inode, u64 off, u64 len)
 		unlock_extent(&BTRFS_I(inode)->io_tree, off, off + len - 1);
 		if (ordered)
 			btrfs_put_ordered_extent(ordered);
+		if (!retry_range_locking)
+			return -EAGAIN;
 		btrfs_wait_ordered_range(inode, off, len);
 	}
+	return 0;
 }
 
 static void btrfs_double_inode_unlock(struct inode *inode1, struct inode *inode2)
@@ -2877,15 +2899,24 @@ static void btrfs_double_extent_unlock(struct inode *inode1, u64 loff1,
 	unlock_extent(&BTRFS_I(inode2)->io_tree, loff2, loff2 + len - 1);
 }
 
-static void btrfs_double_extent_lock(struct inode *inode1, u64 loff1,
-				     struct inode *inode2, u64 loff2, u64 len)
+static int btrfs_double_extent_lock(struct inode *inode1, u64 loff1,
+				    struct inode *inode2, u64 loff2, u64 len,
+				    bool retry_range_locking)
 {
+	int ret;
+
 	if (inode1 < inode2) {
 		swap(inode1, inode2);
 		swap(loff1, loff2);
 	}
-	lock_extent_range(inode1, loff1, len);
-	lock_extent_range(inode2, loff2, len);
+	ret = lock_extent_range(inode1, loff1, len, retry_range_locking);
+	if (ret)
+		return ret;
+	ret = lock_extent_range(inode2, loff2, len, retry_range_locking);
+	if (ret)
+		unlock_extent(&BTRFS_I(inode1)->io_tree, loff1,
+			      loff1 + len - 1);
+	return ret;
 }
 
 struct cmp_pages {
@@ -2901,11 +2932,15 @@ static void btrfs_cmp_data_free(struct cmp_pages *cmp)
 
 	for (i = 0; i < cmp->num_pages; i++) {
 		pg = cmp->src_pages[i];
-		if (pg)
+		if (pg) {
+			unlock_page(pg);
 			page_cache_release(pg);
+		}
 		pg = cmp->dst_pages[i];
-		if (pg)
+		if (pg) {
+			unlock_page(pg);
 			page_cache_release(pg);
+		}
 	}
 	kfree(cmp->src_pages);
 	kfree(cmp->dst_pages);
@@ -2966,6 +3001,8 @@ static int btrfs_cmp_data(struct inode *src, u64 loff, struct inode *dst,
 
 		src_page = cmp->src_pages[i];
 		dst_page = cmp->dst_pages[i];
+		ASSERT(PageLocked(src_page));
+		ASSERT(PageLocked(dst_page));
 
 		addr = kmap_atomic(src_page);
 		dst_addr = kmap_atomic(dst_page);
@@ -3078,14 +3115,46 @@ static int btrfs_extent_same(struct inode *src, u64 loff, u64 olen,
 		goto out_unlock;
 	}
 
+again:
 	ret = btrfs_cmp_data_prepare(src, loff, dst, dst_loff, olen, &cmp);
 	if (ret)
 		goto out_unlock;
 
 	if (same_inode)
-		lock_extent_range(src, same_lock_start, same_lock_len);
+		ret = lock_extent_range(src, same_lock_start, same_lock_len,
+					false);
 	else
-		btrfs_double_extent_lock(src, loff, dst, dst_loff, len);
+		ret = btrfs_double_extent_lock(src, loff, dst, dst_loff, len,
+					       false);
+	/*
+	 * If one of the inodes has dirty pages in the respective range or
+	 * ordered extents, we need to flush dellaloc and wait for all ordered
+	 * extents in the range. We must unlock the pages and the ranges in the
+	 * io trees to avoid deadlocks when flushing delalloc (requires locking
+	 * pages) and when waiting for ordered extents to complete (they require
+	 * range locking).
+	 */
+	if (ret == -EAGAIN) {
+		/*
+		 * Ranges in the io trees already unlocked. Now unlock all
+		 * pages before waiting for all IO to complete.
+		 */
+		btrfs_cmp_data_free(&cmp);
+		if (same_inode) {
+			btrfs_wait_ordered_range(src, same_lock_start,
+						 same_lock_len);
+		} else {
+			btrfs_wait_ordered_range(src, loff, len);
+			btrfs_wait_ordered_range(dst, dst_loff, len);
+		}
+		goto again;
+	}
+	ASSERT(ret == 0);
+	if (WARN_ON(ret)) {
+		/* ranges in the io trees already unlocked */
+		btrfs_cmp_data_free(&cmp);
+		return ret;
+	}
 
 	/* pass original length for comparison so we stay within i_size */
 	ret = btrfs_cmp_data(src, loff, dst, dst_loff, olen, &cmp);
@@ -3795,9 +3864,15 @@ static noinline int btrfs_clone_files(struct file *file, struct file *file_src,
 		u64 lock_start = min_t(u64, off, destoff);
 		u64 lock_len = max_t(u64, off, destoff) + len - lock_start;
 
-		lock_extent_range(src, lock_start, lock_len);
+		ret = lock_extent_range(src, lock_start, lock_len, true);
 	} else {
-		btrfs_double_extent_lock(src, off, inode, destoff, len);
+		ret = btrfs_double_extent_lock(src, off, inode, destoff, len,
+					       true);
+	}
+	ASSERT(ret == 0);
+	if (WARN_ON(ret)) {
+		/* ranges in the io trees already unlocked */
+		goto out_unlock;
 	}
 
 	ret = btrfs_clone(src, inode, off, olen, len, destoff, 0);
diff --git a/fs/btrfs/root-tree.c b/fs/btrfs/root-tree.c
index 7cf8509deda7..2c849b08a91b 100644
--- a/fs/btrfs/root-tree.c
+++ b/fs/btrfs/root-tree.c
@@ -310,8 +310,16 @@ int btrfs_find_orphan_roots(struct btrfs_root *tree_root)
 		set_bit(BTRFS_ROOT_ORPHAN_ITEM_INSERTED, &root->state);
 
 		err = btrfs_insert_fs_root(root->fs_info, root);
+		/*
+		 * The root might have been inserted already, as before we look
+		 * for orphan roots, log replay might have happened, which
+		 * triggers a transaction commit and qgroup accounting, which
+		 * in turn reads and inserts fs roots while doing backref
+		 * walking.
+		 */
+		if (err == -EEXIST)
+			err = 0;
 		if (err) {
-			BUG_ON(err == -EEXIST);
 			btrfs_free_fs_root(root);
 			break;
 		}
diff --git a/fs/ceph/addr.c b/fs/ceph/addr.c
index c22213789090..19adeb0ef82a 100644
--- a/fs/ceph/addr.c
+++ b/fs/ceph/addr.c
@@ -1756,6 +1756,10 @@ int ceph_pool_perm_check(struct ceph_inode_info *ci, int need)
 	u32 pool;
 	int ret, flags;
 
+	/* does not support pool namespace yet */
+	if (ci->i_pool_ns_len)
+		return -EIO;
+
 	if (ceph_test_mount_opt(ceph_inode_to_client(&ci->vfs_inode),
 				NOPOOLPERM))
 		return 0;
diff --git a/fs/ceph/caps.c b/fs/ceph/caps.c
index cdbf8cf3d52c..6fe0ad26a7df 100644
--- a/fs/ceph/caps.c
+++ b/fs/ceph/caps.c
@@ -2753,7 +2753,8 @@ static void handle_cap_grant(struct ceph_mds_client *mdsc,
 			     void *inline_data, int inline_len,
 			     struct ceph_buffer *xattr_buf,
 			     struct ceph_mds_session *session,
-			     struct ceph_cap *cap, int issued)
+			     struct ceph_cap *cap, int issued,
+			     u32 pool_ns_len)
 	__releases(ci->i_ceph_lock)
 	__releases(mdsc->snap_rwsem)
 {
@@ -2873,6 +2874,8 @@ static void handle_cap_grant(struct ceph_mds_client *mdsc,
 	if (newcaps & (CEPH_CAP_ANY_FILE_RD | CEPH_CAP_ANY_FILE_WR)) {
 		/* file layout may have changed */
 		ci->i_layout = grant->layout;
+		ci->i_pool_ns_len = pool_ns_len;
+
 		/* size/truncate_seq? */
 		queue_trunc = ceph_fill_file_size(inode, issued,
 					le32_to_cpu(grant->truncate_seq),
@@ -3411,6 +3414,7 @@ void ceph_handle_caps(struct ceph_mds_session *session,
 	u32  inline_len = 0;
 	void *snaptrace;
 	size_t snaptrace_len;
+	u32 pool_ns_len = 0;
 	void *p, *end;
 
 	dout("handle_caps from mds%d\n", mds);
@@ -3463,6 +3467,21 @@ void ceph_handle_caps(struct ceph_mds_session *session,
 		p += inline_len;
 	}
 
+	if (le16_to_cpu(msg->hdr.version) >= 8) {
+		u64 flush_tid;
+		u32 caller_uid, caller_gid;
+		u32 osd_epoch_barrier;
+		/* version >= 5 */
+		ceph_decode_32_safe(&p, end, osd_epoch_barrier, bad);
+		/* version >= 6 */
+		ceph_decode_64_safe(&p, end, flush_tid, bad);
+		/* version >= 7 */
+		ceph_decode_32_safe(&p, end, caller_uid, bad);
+		ceph_decode_32_safe(&p, end, caller_gid, bad);
+		/* version >= 8 */
+		ceph_decode_32_safe(&p, end, pool_ns_len, bad);
+	}
+
 	/* lookup ino */
 	inode = ceph_find_inode(sb, vino);
 	ci = ceph_inode(inode);
@@ -3518,7 +3537,8 @@ void ceph_handle_caps(struct ceph_mds_session *session,
 				  &cap, &issued);
 		handle_cap_grant(mdsc, inode, h,
 				 inline_version, inline_data, inline_len,
-				 msg->middle, session, cap, issued);
+				 msg->middle, session, cap, issued,
+				 pool_ns_len);
 		if (realm)
 			ceph_put_snap_realm(mdsc, realm);
 		goto done_unlocked;
@@ -3542,7 +3562,8 @@ void ceph_handle_caps(struct ceph_mds_session *session,
 		issued |= __ceph_caps_dirty(ci);
 		handle_cap_grant(mdsc, inode, h,
 				 inline_version, inline_data, inline_len,
-				 msg->middle, session, cap, issued);
+				 msg->middle, session, cap, issued,
+				 pool_ns_len);
 		goto done_unlocked;
 
 	case CEPH_CAP_OP_FLUSH_ACK:
diff --git a/fs/ceph/inode.c b/fs/ceph/inode.c
index fb4ba2e4e2a5..5849b88bbed3 100644
--- a/fs/ceph/inode.c
+++ b/fs/ceph/inode.c
@@ -396,6 +396,7 @@ struct inode *ceph_alloc_inode(struct super_block *sb)
 	ci->i_symlink = NULL;
 
 	memset(&ci->i_dir_layout, 0, sizeof(ci->i_dir_layout));
+	ci->i_pool_ns_len = 0;
 
 	ci->i_fragtree = RB_ROOT;
 	mutex_init(&ci->i_fragtree_mutex);
@@ -756,6 +757,7 @@ static int fill_inode(struct inode *inode, struct page *locked_page,
 		if (ci->i_layout.fl_pg_pool != info->layout.fl_pg_pool)
 			ci->i_ceph_flags &= ~CEPH_I_POOL_PERM;
 		ci->i_layout = info->layout;
+		ci->i_pool_ns_len = iinfo->pool_ns_len;
 
 		queue_trunc = ceph_fill_file_size(inode, issued,
 					le32_to_cpu(info->truncate_seq),
diff --git a/fs/ceph/mds_client.c b/fs/ceph/mds_client.c
index e7b130a637f9..911d64d865f1 100644
--- a/fs/ceph/mds_client.c
+++ b/fs/ceph/mds_client.c
@@ -100,6 +100,14 @@ static int parse_reply_info_in(void **p, void *end,
 	} else
 		info->inline_version = CEPH_INLINE_NONE;
 
+	if (features & CEPH_FEATURE_FS_FILE_LAYOUT_V2) {
+		ceph_decode_32_safe(p, end, info->pool_ns_len, bad);
+		ceph_decode_need(p, end, info->pool_ns_len, bad);
+		*p += info->pool_ns_len;
+	} else {
+		info->pool_ns_len = 0;
+	}
+
 	return 0;
 bad:
 	return err;
@@ -2298,6 +2306,14 @@ int ceph_mdsc_do_request(struct ceph_mds_client *mdsc,
 		ceph_get_cap_refs(ceph_inode(req->r_old_dentry_dir),
 				  CEPH_CAP_PIN);
 
+	/* deny access to directories with pool_ns layouts */
+	if (req->r_inode && S_ISDIR(req->r_inode->i_mode) &&
+	    ceph_inode(req->r_inode)->i_pool_ns_len)
+		return -EIO;
+	if (req->r_locked_dir &&
+	    ceph_inode(req->r_locked_dir)->i_pool_ns_len)
+		return -EIO;
+
 	/* issue */
 	mutex_lock(&mdsc->mutex);
 	__register_request(mdsc, req, dir);
diff --git a/fs/ceph/mds_client.h b/fs/ceph/mds_client.h
index ccf11ef0ca87..37712ccffcc6 100644
--- a/fs/ceph/mds_client.h
+++ b/fs/ceph/mds_client.h
@@ -44,6 +44,7 @@ struct ceph_mds_reply_info_in {
 	u64 inline_version;
 	u32 inline_len;
 	char *inline_data;
+	u32 pool_ns_len;
 };
 
 /*
diff --git a/fs/ceph/super.h b/fs/ceph/super.h
index 75b7d125ce66..9c458eb52245 100644
--- a/fs/ceph/super.h
+++ b/fs/ceph/super.h
@@ -287,6 +287,7 @@ struct ceph_inode_info {
 
 	struct ceph_dir_layout i_dir_layout;
 	struct ceph_file_layout i_layout;
+	size_t i_pool_ns_len;
 	char *i_symlink;
 
 	/* for dirs */
diff --git a/fs/cifs/cifs_dfs_ref.c b/fs/cifs/cifs_dfs_ref.c
index 7dc886c9a78f..e956cba94338 100644
--- a/fs/cifs/cifs_dfs_ref.c
+++ b/fs/cifs/cifs_dfs_ref.c
@@ -175,7 +175,7 @@ char *cifs_compose_mount_options(const char *sb_mountdata,
 	 * string to the length of the original string to allow for worst case.
 	 */
 	md_len = strlen(sb_mountdata) + INET6_ADDRSTRLEN;
-	mountdata = kzalloc(md_len + 1, GFP_KERNEL);
+	mountdata = kzalloc(md_len + sizeof("ip=") + 1, GFP_KERNEL);
 	if (mountdata == NULL) {
 		rc = -ENOMEM;
 		goto compose_mount_options_err;
diff --git a/fs/cifs/cifsencrypt.c b/fs/cifs/cifsencrypt.c
index afa09fce8151..e682b36a210f 100644
--- a/fs/cifs/cifsencrypt.c
+++ b/fs/cifs/cifsencrypt.c
@@ -714,7 +714,7 @@ setup_ntlmv2_rsp(struct cifs_ses *ses, const struct nls_table *nls_cp)
 
 	ses->auth_key.response = kmalloc(baselen + tilen, GFP_KERNEL);
 	if (!ses->auth_key.response) {
-		rc = ENOMEM;
+		rc = -ENOMEM;
 		ses->auth_key.len = 0;
 		goto setup_ntlmv2_rsp_ret;
 	}
diff --git a/fs/cifs/cifsfs.c b/fs/cifs/cifsfs.c
index c48ca13673e3..2eea40353e60 100644
--- a/fs/cifs/cifsfs.c
+++ b/fs/cifs/cifsfs.c
@@ -1013,7 +1013,6 @@ const struct file_operations cifs_file_strict_ops = {
 	.llseek = cifs_llseek,
 	.unlocked_ioctl	= cifs_ioctl,
 	.clone_file_range = cifs_clone_file_range,
-	.clone_file_range = cifs_clone_file_range,
 	.setlease = cifs_setlease,
 	.fallocate = cifs_fallocate,
 };
diff --git a/fs/cifs/cifsfs.h b/fs/cifs/cifsfs.h
index 68c4547528c4..83aac8ba50b0 100644
--- a/fs/cifs/cifsfs.h
+++ b/fs/cifs/cifsfs.h
@@ -31,19 +31,15 @@
  * so that it will fit. We use hash_64 to convert the value to 31 bits, and
  * then add 1, to ensure that we don't end up with a 0 as the value.
  */
-#if BITS_PER_LONG == 64
 static inline ino_t
 cifs_uniqueid_to_ino_t(u64 fileid)
 {
+	if ((sizeof(ino_t)) < (sizeof(u64)))
+		return (ino_t)hash_64(fileid, (sizeof(ino_t) * 8) - 1) + 1;
+
 	return (ino_t)fileid;
+
 }
-#else
-static inline ino_t
-cifs_uniqueid_to_ino_t(u64 fileid)
-{
-	return (ino_t)hash_64(fileid, (sizeof(ino_t) * 8) - 1) + 1;
-}
-#endif
 
 extern struct file_system_type cifs_fs_type;
 extern const struct address_space_operations cifs_addr_ops;
diff --git a/fs/cifs/cifssmb.c b/fs/cifs/cifssmb.c
index 90b4f9f7de66..76fcb50295a3 100644
--- a/fs/cifs/cifssmb.c
+++ b/fs/cifs/cifssmb.c
@@ -1396,11 +1396,10 @@ openRetry:
  * current bigbuf.
  */
 static int
-cifs_readv_discard(struct TCP_Server_Info *server, struct mid_q_entry *mid)
+discard_remaining_data(struct TCP_Server_Info *server)
 {
 	unsigned int rfclen = get_rfc1002_length(server->smallbuf);
 	int remaining = rfclen + 4 - server->total_read;
-	struct cifs_readdata *rdata = mid->callback_data;
 
 	while (remaining > 0) {
 		int length;
@@ -1414,10 +1413,20 @@ cifs_readv_discard(struct TCP_Server_Info *server, struct mid_q_entry *mid)
 		remaining -= length;
 	}
 
-	dequeue_mid(mid, rdata->result);
 	return 0;
 }
 
+static int
+cifs_readv_discard(struct TCP_Server_Info *server, struct mid_q_entry *mid)
+{
+	int length;
+	struct cifs_readdata *rdata = mid->callback_data;
+
+	length = discard_remaining_data(server);
+	dequeue_mid(mid, rdata->result);
+	return length;
+}
+
 int
 cifs_readv_receive(struct TCP_Server_Info *server, struct mid_q_entry *mid)
 {
@@ -1446,6 +1455,12 @@ cifs_readv_receive(struct TCP_Server_Info *server, struct mid_q_entry *mid)
 		return length;
 	server->total_read += length;
 
+	if (server->ops->is_status_pending &&
+	    server->ops->is_status_pending(buf, server, 0)) {
+		discard_remaining_data(server);
+		return -1;
+	}
+
 	/* Was the SMB read successful? */
 	rdata->result = server->ops->map_error(buf, false);
 	if (rdata->result != 0) {
diff --git a/fs/cifs/connect.c b/fs/cifs/connect.c
index 4fbd92d2e113..a763cd3d9e7c 100644
--- a/fs/cifs/connect.c
+++ b/fs/cifs/connect.c
@@ -2999,8 +2999,7 @@ ip_rfc1001_connect(struct TCP_Server_Info *server)
 	if (ses_init_buf) {
 		ses_init_buf->trailer.session_req.called_len = 32;
 
-		if (server->server_RFC1001_name &&
-		    server->server_RFC1001_name[0] != 0)
+		if (server->server_RFC1001_name[0] != 0)
 			rfc1002mangle(ses_init_buf->trailer.
 				      session_req.called_name,
 				      server->server_RFC1001_name,
diff --git a/fs/cifs/smb2pdu.c b/fs/cifs/smb2pdu.c
index 10f8d5cf5681..42e1f440eb1e 100644
--- a/fs/cifs/smb2pdu.c
+++ b/fs/cifs/smb2pdu.c
@@ -1106,21 +1106,25 @@ parse_lease_state(struct TCP_Server_Info *server, struct smb2_create_rsp *rsp,
 {
 	char *data_offset;
 	struct create_context *cc;
-	unsigned int next = 0;
+	unsigned int next;
+	unsigned int remaining;
 	char *name;
 
 	data_offset = (char *)rsp + 4 + le32_to_cpu(rsp->CreateContextsOffset);
+	remaining = le32_to_cpu(rsp->CreateContextsLength);
 	cc = (struct create_context *)data_offset;
-	do {
-		cc = (struct create_context *)((char *)cc + next);
+	while (remaining >= sizeof(struct create_context)) {
 		name = le16_to_cpu(cc->NameOffset) + (char *)cc;
-		if (le16_to_cpu(cc->NameLength) != 4 ||
-		    strncmp(name, "RqLs", 4)) {
-			next = le32_to_cpu(cc->Next);
-			continue;
-		}
-		return server->ops->parse_lease_buf(cc, epoch);
-	} while (next != 0);
+		if (le16_to_cpu(cc->NameLength) == 4 &&
+		    strncmp(name, "RqLs", 4) == 0)
+			return server->ops->parse_lease_buf(cc, epoch);
+
+		next = le32_to_cpu(cc->Next);
+		if (!next)
+			break;
+		remaining -= next;
+		cc = (struct create_context *)((char *)cc + next);
+	}
 
 	return 0;
 }
diff --git a/fs/dax.c b/fs/dax.c
index fc2e3141138b..bbb2ad783770 100644
--- a/fs/dax.c
+++ b/fs/dax.c
@@ -79,15 +79,14 @@ struct page *read_dax_sector(struct block_device *bdev, sector_t n)
 }
 
 /*
- * dax_clear_blocks() is called from within transaction context from XFS,
+ * dax_clear_sectors() is called from within transaction context from XFS,
  * and hence this means the stack from this point must follow GFP_NOFS
  * semantics for all operations.
  */
-int dax_clear_blocks(struct inode *inode, sector_t block, long _size)
+int dax_clear_sectors(struct block_device *bdev, sector_t _sector, long _size)
 {
-	struct block_device *bdev = inode->i_sb->s_bdev;
 	struct blk_dax_ctl dax = {
-		.sector = block << (inode->i_blkbits - 9),
+		.sector = _sector,
 		.size = _size,
 	};
 
@@ -109,7 +108,7 @@ int dax_clear_blocks(struct inode *inode, sector_t block, long _size)
 	wmb_pmem();
 	return 0;
 }
-EXPORT_SYMBOL_GPL(dax_clear_blocks);
+EXPORT_SYMBOL_GPL(dax_clear_sectors);
 
 /* the clear_pmem() calls are ordered by a wmb_pmem() in the caller */
 static void dax_new_buf(void __pmem *addr, unsigned size, unsigned first,
@@ -485,11 +484,10 @@ static int dax_writeback_one(struct block_device *bdev,
  * end]. This is required by data integrity operations to ensure file data is
  * on persistent storage prior to completion of the operation.
  */
-int dax_writeback_mapping_range(struct address_space *mapping, loff_t start,
-		loff_t end)
+int dax_writeback_mapping_range(struct address_space *mapping,
+		struct block_device *bdev, struct writeback_control *wbc)
 {
 	struct inode *inode = mapping->host;
-	struct block_device *bdev = inode->i_sb->s_bdev;
 	pgoff_t start_index, end_index, pmd_index;
 	pgoff_t indices[PAGEVEC_SIZE];
 	struct pagevec pvec;
@@ -500,8 +498,11 @@ int dax_writeback_mapping_range(struct address_space *mapping, loff_t start,
 	if (WARN_ON_ONCE(inode->i_blkbits != PAGE_SHIFT))
 		return -EIO;
 
-	start_index = start >> PAGE_CACHE_SHIFT;
-	end_index = end >> PAGE_CACHE_SHIFT;
+	if (!mapping->nrexceptional || wbc->sync_mode != WB_SYNC_ALL)
+		return 0;
+
+	start_index = wbc->range_start >> PAGE_CACHE_SHIFT;
+	end_index = wbc->range_end >> PAGE_CACHE_SHIFT;
 	pmd_index = DAX_PMD_INDEX(start_index);
 
 	rcu_read_lock();
@@ -1055,6 +1056,7 @@ EXPORT_SYMBOL_GPL(dax_pmd_fault);
 int dax_pfn_mkwrite(struct vm_area_struct *vma, struct vm_fault *vmf)
 {
 	struct file *file = vma->vm_file;
+	int error;
 
 	/*
 	 * We pass NO_SECTOR to dax_radix_entry() because we expect that a
@@ -1064,7 +1066,13 @@ int dax_pfn_mkwrite(struct vm_area_struct *vma, struct vm_fault *vmf)
 	 * saves us from having to make a call to get_block() here to look
 	 * up the sector.
 	 */
-	dax_radix_entry(file->f_mapping, vmf->pgoff, NO_SECTOR, false, true);
+	error = dax_radix_entry(file->f_mapping, vmf->pgoff, NO_SECTOR, false,
+			true);
+
+	if (error == -ENOMEM)
+		return VM_FAULT_OOM;
+	if (error)
+		return VM_FAULT_SIGBUS;
 	return VM_FAULT_NOPAGE;
 }
 EXPORT_SYMBOL_GPL(dax_pfn_mkwrite);
diff --git a/fs/dcache.c b/fs/dcache.c
index 92d5140de851..2398f9f94337 100644
--- a/fs/dcache.c
+++ b/fs/dcache.c
@@ -269,9 +269,6 @@ static inline int dname_external(const struct dentry *dentry)
 	return dentry->d_name.name != dentry->d_iname;
 }
 
-/*
- * Make sure other CPUs see the inode attached before the type is set.
- */
 static inline void __d_set_inode_and_type(struct dentry *dentry,
 					  struct inode *inode,
 					  unsigned type_flags)
@@ -279,28 +276,18 @@ static inline void __d_set_inode_and_type(struct dentry *dentry,
 	unsigned flags;
 
 	dentry->d_inode = inode;
-	smp_wmb();
 	flags = READ_ONCE(dentry->d_flags);
 	flags &= ~(DCACHE_ENTRY_TYPE | DCACHE_FALLTHRU);
 	flags |= type_flags;
 	WRITE_ONCE(dentry->d_flags, flags);
 }
 
-/*
- * Ideally, we want to make sure that other CPUs see the flags cleared before
- * the inode is detached, but this is really a violation of RCU principles
- * since the ordering suggests we should always set inode before flags.
- *
- * We should instead replace or discard the entire dentry - but that sucks
- * performancewise on mass deletion/rename.
- */
 static inline void __d_clear_type_and_inode(struct dentry *dentry)
 {
 	unsigned flags = READ_ONCE(dentry->d_flags);
 
 	flags &= ~(DCACHE_ENTRY_TYPE | DCACHE_FALLTHRU);
 	WRITE_ONCE(dentry->d_flags, flags);
-	smp_wmb();
 	dentry->d_inode = NULL;
 }
 
@@ -370,9 +357,11 @@ static void dentry_unlink_inode(struct dentry * dentry)
 	__releases(dentry->d_inode->i_lock)
 {
 	struct inode *inode = dentry->d_inode;
+
+	raw_write_seqcount_begin(&dentry->d_seq);
 	__d_clear_type_and_inode(dentry);
 	hlist_del_init(&dentry->d_u.d_alias);
-	dentry_rcuwalk_invalidate(dentry);
+	raw_write_seqcount_end(&dentry->d_seq);
 	spin_unlock(&dentry->d_lock);
 	spin_unlock(&inode->i_lock);
 	if (!inode->i_nlink)
@@ -1758,8 +1747,9 @@ static void __d_instantiate(struct dentry *dentry, struct inode *inode)
 	spin_lock(&dentry->d_lock);
 	if (inode)
 		hlist_add_head(&dentry->d_u.d_alias, &inode->i_dentry);
+	raw_write_seqcount_begin(&dentry->d_seq);
 	__d_set_inode_and_type(dentry, inode, add_flags);
-	dentry_rcuwalk_invalidate(dentry);
+	raw_write_seqcount_end(&dentry->d_seq);
 	spin_unlock(&dentry->d_lock);
 	fsnotify_d_instantiate(dentry, inode);
 }
diff --git a/fs/devpts/inode.c b/fs/devpts/inode.c
index 1f107fd51328..655f21f99160 100644
--- a/fs/devpts/inode.c
+++ b/fs/devpts/inode.c
@@ -575,6 +575,26 @@ void devpts_kill_index(struct inode *ptmx_inode, int idx)
 	mutex_unlock(&allocated_ptys_lock);
 }
 
+/*
+ * pty code needs to hold extra references in case of last /dev/tty close
+ */
+
+void devpts_add_ref(struct inode *ptmx_inode)
+{
+	struct super_block *sb = pts_sb_from_inode(ptmx_inode);
+
+	atomic_inc(&sb->s_active);
+	ihold(ptmx_inode);
+}
+
+void devpts_del_ref(struct inode *ptmx_inode)
+{
+	struct super_block *sb = pts_sb_from_inode(ptmx_inode);
+
+	iput(ptmx_inode);
+	deactivate_super(sb);
+}
+
 /**
  * devpts_pty_new -- create a new inode in /dev/pts/
  * @ptmx_inode: inode of the master
diff --git a/fs/direct-io.c b/fs/direct-io.c
index 1b2f7ffc8b84..d6a9012d42ad 100644
--- a/fs/direct-io.c
+++ b/fs/direct-io.c
@@ -472,8 +472,8 @@ static int dio_bio_complete(struct dio *dio, struct bio *bio)
 		dio->io_error = -EIO;
 
 	if (dio->is_async && dio->rw == READ && dio->should_dirty) {
-		bio_check_pages_dirty(bio);	/* transfers ownership */
 		err = bio->bi_error;
+		bio_check_pages_dirty(bio);	/* transfers ownership */
 	} else {
 		bio_for_each_segment_all(bvec, bio, i) {
 			struct page *page = bvec->bv_page;
diff --git a/fs/efivarfs/file.c b/fs/efivarfs/file.c
index c424e4813ec8..d48e0d261d78 100644
--- a/fs/efivarfs/file.c
+++ b/fs/efivarfs/file.c
@@ -10,6 +10,7 @@
 #include <linux/efi.h>
 #include <linux/fs.h>
 #include <linux/slab.h>
+#include <linux/mount.h>
 
 #include "internal.h"
 
@@ -103,9 +104,78 @@ out_free:
 	return size;
 }
 
+static int
+efivarfs_ioc_getxflags(struct file *file, void __user *arg)
+{
+	struct inode *inode = file->f_mapping->host;
+	unsigned int i_flags;
+	unsigned int flags = 0;
+
+	i_flags = inode->i_flags;
+	if (i_flags & S_IMMUTABLE)
+		flags |= FS_IMMUTABLE_FL;
+
+	if (copy_to_user(arg, &flags, sizeof(flags)))
+		return -EFAULT;
+	return 0;
+}
+
+static int
+efivarfs_ioc_setxflags(struct file *file, void __user *arg)
+{
+	struct inode *inode = file->f_mapping->host;
+	unsigned int flags;
+	unsigned int i_flags = 0;
+	int error;
+
+	if (!inode_owner_or_capable(inode))
+		return -EACCES;
+
+	if (copy_from_user(&flags, arg, sizeof(flags)))
+		return -EFAULT;
+
+	if (flags & ~FS_IMMUTABLE_FL)
+		return -EOPNOTSUPP;
+
+	if (!capable(CAP_LINUX_IMMUTABLE))
+		return -EPERM;
+
+	if (flags & FS_IMMUTABLE_FL)
+		i_flags |= S_IMMUTABLE;
+
+
+	error = mnt_want_write_file(file);
+	if (error)
+		return error;
+
+	inode_lock(inode);
+	inode_set_flags(inode, i_flags, S_IMMUTABLE);
+	inode_unlock(inode);
+
+	mnt_drop_write_file(file);
+
+	return 0;
+}
+
+long
+efivarfs_file_ioctl(struct file *file, unsigned int cmd, unsigned long p)
+{
+	void __user *arg = (void __user *)p;
+
+	switch (cmd) {
+	case FS_IOC_GETFLAGS:
+		return efivarfs_ioc_getxflags(file, arg);
+	case FS_IOC_SETFLAGS:
+		return efivarfs_ioc_setxflags(file, arg);
+	}
+
+	return -ENOTTY;
+}
+
 const struct file_operations efivarfs_file_operations = {
 	.open	= simple_open,
 	.read	= efivarfs_file_read,
 	.write	= efivarfs_file_write,
 	.llseek	= no_llseek,
+	.unlocked_ioctl = efivarfs_file_ioctl,
 };
diff --git a/fs/efivarfs/inode.c b/fs/efivarfs/inode.c
index 3381b9da9ee6..e2ab6d0497f2 100644
--- a/fs/efivarfs/inode.c
+++ b/fs/efivarfs/inode.c
@@ -15,7 +15,8 @@
 #include "internal.h"
 
 struct inode *efivarfs_get_inode(struct super_block *sb,
-				const struct inode *dir, int mode, dev_t dev)
+				const struct inode *dir, int mode,
+				dev_t dev, bool is_removable)
 {
 	struct inode *inode = new_inode(sb);
 
@@ -23,6 +24,7 @@ struct inode *efivarfs_get_inode(struct super_block *sb,
 		inode->i_ino = get_next_ino();
 		inode->i_mode = mode;
 		inode->i_atime = inode->i_mtime = inode->i_ctime = CURRENT_TIME;
+		inode->i_flags = is_removable ? 0 : S_IMMUTABLE;
 		switch (mode & S_IFMT) {
 		case S_IFREG:
 			inode->i_fop = &efivarfs_file_operations;
@@ -102,22 +104,17 @@ static void efivarfs_hex_to_guid(const char *str, efi_guid_t *guid)
 static int efivarfs_create(struct inode *dir, struct dentry *dentry,
 			  umode_t mode, bool excl)
 {
-	struct inode *inode;
+	struct inode *inode = NULL;
 	struct efivar_entry *var;
 	int namelen, i = 0, err = 0;
+	bool is_removable = false;
 
 	if (!efivarfs_valid_name(dentry->d_name.name, dentry->d_name.len))
 		return -EINVAL;
 
-	inode = efivarfs_get_inode(dir->i_sb, dir, mode, 0);
-	if (!inode)
-		return -ENOMEM;
-
 	var = kzalloc(sizeof(struct efivar_entry), GFP_KERNEL);
-	if (!var) {
-		err = -ENOMEM;
-		goto out;
-	}
+	if (!var)
+		return -ENOMEM;
 
 	/* length of the variable name itself: remove GUID and separator */
 	namelen = dentry->d_name.len - EFI_VARIABLE_GUID_LEN - 1;
@@ -125,6 +122,16 @@ static int efivarfs_create(struct inode *dir, struct dentry *dentry,
 	efivarfs_hex_to_guid(dentry->d_name.name + namelen + 1,
 			&var->var.VendorGuid);
 
+	if (efivar_variable_is_removable(var->var.VendorGuid,
+					 dentry->d_name.name, namelen))
+		is_removable = true;
+
+	inode = efivarfs_get_inode(dir->i_sb, dir, mode, 0, is_removable);
+	if (!inode) {
+		err = -ENOMEM;
+		goto out;
+	}
+
 	for (i = 0; i < namelen; i++)
 		var->var.VariableName[i] = dentry->d_name.name[i];
 
@@ -138,7 +145,8 @@ static int efivarfs_create(struct inode *dir, struct dentry *dentry,
 out:
 	if (err) {
 		kfree(var);
-		iput(inode);
+		if (inode)
+			iput(inode);
 	}
 	return err;
 }
diff --git a/fs/efivarfs/internal.h b/fs/efivarfs/internal.h
index b5ff16addb7c..b4505188e799 100644
--- a/fs/efivarfs/internal.h
+++ b/fs/efivarfs/internal.h
@@ -15,7 +15,8 @@ extern const struct file_operations efivarfs_file_operations;
 extern const struct inode_operations efivarfs_dir_inode_operations;
 extern bool efivarfs_valid_name(const char *str, int len);
 extern struct inode *efivarfs_get_inode(struct super_block *sb,
-			const struct inode *dir, int mode, dev_t dev);
+			const struct inode *dir, int mode, dev_t dev,
+			bool is_removable);
 
 extern struct list_head efivarfs_list;
 
diff --git a/fs/efivarfs/super.c b/fs/efivarfs/super.c
index b8a564f29107..dd029d13ea61 100644
--- a/fs/efivarfs/super.c
+++ b/fs/efivarfs/super.c
@@ -118,8 +118,9 @@ static int efivarfs_callback(efi_char16_t *name16, efi_guid_t vendor,
 	struct dentry *dentry, *root = sb->s_root;
 	unsigned long size = 0;
 	char *name;
-	int len, i;
+	int len;
 	int err = -ENOMEM;
+	bool is_removable = false;
 
 	entry = kzalloc(sizeof(*entry), GFP_KERNEL);
 	if (!entry)
@@ -128,15 +129,17 @@ static int efivarfs_callback(efi_char16_t *name16, efi_guid_t vendor,
 	memcpy(entry->var.VariableName, name16, name_size);
 	memcpy(&(entry->var.VendorGuid), &vendor, sizeof(efi_guid_t));
 
-	len = ucs2_strlen(entry->var.VariableName);
+	len = ucs2_utf8size(entry->var.VariableName);
 
 	/* name, plus '-', plus GUID, plus NUL*/
 	name = kmalloc(len + 1 + EFI_VARIABLE_GUID_LEN + 1, GFP_KERNEL);
 	if (!name)
 		goto fail;
 
-	for (i = 0; i < len; i++)
-		name[i] = entry->var.VariableName[i] & 0xFF;
+	ucs2_as_utf8(name, entry->var.VariableName, len);
+
+	if (efivar_variable_is_removable(entry->var.VendorGuid, name, len))
+		is_removable = true;
 
 	name[len] = '-';
 
@@ -144,7 +147,8 @@ static int efivarfs_callback(efi_char16_t *name16, efi_guid_t vendor,
 
 	name[len + EFI_VARIABLE_GUID_LEN+1] = '\0';
 
-	inode = efivarfs_get_inode(sb, d_inode(root), S_IFREG | 0644, 0);
+	inode = efivarfs_get_inode(sb, d_inode(root), S_IFREG | 0644, 0,
+				   is_removable);
 	if (!inode)
 		goto fail_name;
 
@@ -200,7 +204,7 @@ static int efivarfs_fill_super(struct super_block *sb, void *data, int silent)
 	sb->s_d_op		= &efivarfs_d_ops;
 	sb->s_time_gran         = 1;
 
-	inode = efivarfs_get_inode(sb, NULL, S_IFDIR | 0755, 0);
+	inode = efivarfs_get_inode(sb, NULL, S_IFDIR | 0755, 0, true);
 	if (!inode)
 		return -ENOMEM;
 	inode->i_op = &efivarfs_dir_inode_operations;
diff --git a/fs/ext2/file.c b/fs/ext2/file.c
index 2c88d683cd91..c1400b109805 100644
--- a/fs/ext2/file.c
+++ b/fs/ext2/file.c
@@ -80,23 +80,6 @@ static int ext2_dax_pmd_fault(struct vm_area_struct *vma, unsigned long addr,
 	return ret;
 }
 
-static int ext2_dax_mkwrite(struct vm_area_struct *vma, struct vm_fault *vmf)
-{
-	struct inode *inode = file_inode(vma->vm_file);
-	struct ext2_inode_info *ei = EXT2_I(inode);
-	int ret;
-
-	sb_start_pagefault(inode->i_sb);
-	file_update_time(vma->vm_file);
-	down_read(&ei->dax_sem);
-
-	ret = __dax_mkwrite(vma, vmf, ext2_get_block, NULL);
-
-	up_read(&ei->dax_sem);
-	sb_end_pagefault(inode->i_sb);
-	return ret;
-}
-
 static int ext2_dax_pfn_mkwrite(struct vm_area_struct *vma,
 		struct vm_fault *vmf)
 {
@@ -124,7 +107,7 @@ static int ext2_dax_pfn_mkwrite(struct vm_area_struct *vma,
 static const struct vm_operations_struct ext2_dax_vm_ops = {
 	.fault		= ext2_dax_fault,
 	.pmd_fault	= ext2_dax_pmd_fault,
-	.page_mkwrite	= ext2_dax_mkwrite,
+	.page_mkwrite	= ext2_dax_fault,
 	.pfn_mkwrite	= ext2_dax_pfn_mkwrite,
 };
 
diff --git a/fs/ext2/inode.c b/fs/ext2/inode.c
index 338eefda70c6..6bd58e6ff038 100644
--- a/fs/ext2/inode.c
+++ b/fs/ext2/inode.c
@@ -737,8 +737,10 @@ static int ext2_get_blocks(struct inode *inode,
 		 * so that it's not found by another thread before it's
 		 * initialised
 		 */
-		err = dax_clear_blocks(inode, le32_to_cpu(chain[depth-1].key),
-						1 << inode->i_blkbits);
+		err = dax_clear_sectors(inode->i_sb->s_bdev,
+				le32_to_cpu(chain[depth-1].key) <<
+				(inode->i_blkbits - 9),
+				1 << inode->i_blkbits);
 		if (err) {
 			mutex_unlock(&ei->truncate_mutex);
 			goto cleanup;
@@ -874,6 +876,14 @@ ext2_direct_IO(struct kiocb *iocb, struct iov_iter *iter, loff_t offset)
 static int
 ext2_writepages(struct address_space *mapping, struct writeback_control *wbc)
 {
+#ifdef CONFIG_FS_DAX
+	if (dax_mapping(mapping)) {
+		return dax_writeback_mapping_range(mapping,
+						   mapping->host->i_sb->s_bdev,
+						   wbc);
+	}
+#endif
+
 	return mpage_writepages(mapping, wbc, ext2_get_block);
 }
 
@@ -1296,7 +1306,7 @@ void ext2_set_inode_flags(struct inode *inode)
 		inode->i_flags |= S_NOATIME;
 	if (flags & EXT2_DIRSYNC_FL)
 		inode->i_flags |= S_DIRSYNC;
-	if (test_opt(inode->i_sb, DAX))
+	if (test_opt(inode->i_sb, DAX) && S_ISREG(inode->i_mode))
 		inode->i_flags |= S_DAX;
 }
 
diff --git a/fs/ext4/balloc.c b/fs/ext4/balloc.c
index ec0668a60678..fe1f50fe764f 100644
--- a/fs/ext4/balloc.c
+++ b/fs/ext4/balloc.c
@@ -191,7 +191,6 @@ static int ext4_init_block_bitmap(struct super_block *sb,
 	/* If checksum is bad mark all blocks used to prevent allocation
 	 * essentially implementing a per-group read-only flag. */
 	if (!ext4_group_desc_csum_verify(sb, block_group, gdp)) {
-		ext4_error(sb, "Checksum bad for group %u", block_group);
 		grp = ext4_get_group_info(sb, block_group);
 		if (!EXT4_MB_GRP_BBITMAP_CORRUPT(grp))
 			percpu_counter_sub(&sbi->s_freeclusters_counter,
@@ -442,14 +441,16 @@ ext4_read_block_bitmap_nowait(struct super_block *sb, ext4_group_t block_group)
 	}
 	ext4_lock_group(sb, block_group);
 	if (desc->bg_flags & cpu_to_le16(EXT4_BG_BLOCK_UNINIT)) {
-
 		err = ext4_init_block_bitmap(sb, bh, block_group, desc);
 		set_bitmap_uptodate(bh);
 		set_buffer_uptodate(bh);
 		ext4_unlock_group(sb, block_group);
 		unlock_buffer(bh);
-		if (err)
+		if (err) {
+			ext4_error(sb, "Failed to init block bitmap for group "
+				   "%u: %d", block_group, err);
 			goto out;
+		}
 		goto verify;
 	}
 	ext4_unlock_group(sb, block_group);
diff --git a/fs/ext4/crypto.c b/fs/ext4/crypto.c
index c8021208a7eb..38f7562489bb 100644
--- a/fs/ext4/crypto.c
+++ b/fs/ext4/crypto.c
@@ -467,3 +467,59 @@ uint32_t ext4_validate_encryption_key_size(uint32_t mode, uint32_t size)
 		return size;
 	return 0;
 }
+
+/*
+ * Validate dentries for encrypted directories to make sure we aren't
+ * potentially caching stale data after a key has been added or
+ * removed.
+ */
+static int ext4_d_revalidate(struct dentry *dentry, unsigned int flags)
+{
+	struct inode *dir = d_inode(dentry->d_parent);
+	struct ext4_crypt_info *ci = EXT4_I(dir)->i_crypt_info;
+	int dir_has_key, cached_with_key;
+
+	if (!ext4_encrypted_inode(dir))
+		return 0;
+
+	if (ci && ci->ci_keyring_key &&
+	    (ci->ci_keyring_key->flags & ((1 << KEY_FLAG_INVALIDATED) |
+					  (1 << KEY_FLAG_REVOKED) |
+					  (1 << KEY_FLAG_DEAD))))
+		ci = NULL;
+
+	/* this should eventually be an flag in d_flags */
+	cached_with_key = dentry->d_fsdata != NULL;
+	dir_has_key = (ci != NULL);
+
+	/*
+	 * If the dentry was cached without the key, and it is a
+	 * negative dentry, it might be a valid name.  We can't check
+	 * if the key has since been made available due to locking
+	 * reasons, so we fail the validation so ext4_lookup() can do
+	 * this check.
+	 *
+	 * We also fail the validation if the dentry was created with
+	 * the key present, but we no longer have the key, or vice versa.
+	 */
+	if ((!cached_with_key && d_is_negative(dentry)) ||
+	    (!cached_with_key && dir_has_key) ||
+	    (cached_with_key && !dir_has_key)) {
+#if 0				/* Revalidation debug */
+		char buf[80];
+		char *cp = simple_dname(dentry, buf, sizeof(buf));
+
+		if (IS_ERR(cp))
+			cp = (char *) "???";
+		pr_err("revalidate: %s %p %d %d %d\n", cp, dentry->d_fsdata,
+		       cached_with_key, d_is_negative(dentry),
+		       dir_has_key);
+#endif
+		return 0;
+	}
+	return 1;
+}
+
+const struct dentry_operations ext4_encrypted_d_ops = {
+	.d_revalidate = ext4_d_revalidate,
+};
diff --git a/fs/ext4/dir.c b/fs/ext4/dir.c
index 1d1bca74f844..33f5e2a50cf8 100644
--- a/fs/ext4/dir.c
+++ b/fs/ext4/dir.c
@@ -111,6 +111,12 @@ static int ext4_readdir(struct file *file, struct dir_context *ctx)
 	int dir_has_error = 0;
 	struct ext4_str fname_crypto_str = {.name = NULL, .len = 0};
 
+	if (ext4_encrypted_inode(inode)) {
+		err = ext4_get_encryption_info(inode);
+		if (err && err != -ENOKEY)
+			return err;
+	}
+
 	if (is_dx_dir(inode)) {
 		err = ext4_dx_readdir(file, ctx);
 		if (err != ERR_BAD_DX_DIR) {
@@ -157,8 +163,11 @@ static int ext4_readdir(struct file *file, struct dir_context *ctx)
 					index, 1);
 			file->f_ra.prev_pos = (loff_t)index << PAGE_CACHE_SHIFT;
 			bh = ext4_bread(NULL, inode, map.m_lblk, 0);
-			if (IS_ERR(bh))
-				return PTR_ERR(bh);
+			if (IS_ERR(bh)) {
+				err = PTR_ERR(bh);
+				bh = NULL;
+				goto errout;
+			}
 		}
 
 		if (!bh) {
diff --git a/fs/ext4/ext4.h b/fs/ext4/ext4.h
index 0662b285dc8a..157b458a69d4 100644
--- a/fs/ext4/ext4.h
+++ b/fs/ext4/ext4.h
@@ -2302,6 +2302,7 @@ struct page *ext4_encrypt(struct inode *inode,
 int ext4_decrypt(struct page *page);
 int ext4_encrypted_zeroout(struct inode *inode, ext4_lblk_t lblk,
 			   ext4_fsblk_t pblk, ext4_lblk_t len);
+extern const struct dentry_operations ext4_encrypted_d_ops;
 
 #ifdef CONFIG_EXT4_FS_ENCRYPTION
 int ext4_init_crypto(void);
diff --git a/fs/ext4/extents.c b/fs/ext4/extents.c
index 0ffabaf90aa5..3753ceb0b0dd 100644
--- a/fs/ext4/extents.c
+++ b/fs/ext4/extents.c
@@ -3928,7 +3928,7 @@ static int
 convert_initialized_extent(handle_t *handle, struct inode *inode,
 			   struct ext4_map_blocks *map,
 			   struct ext4_ext_path **ppath, int flags,
-			   unsigned int allocated, ext4_fsblk_t newblock)
+			   unsigned int allocated)
 {
 	struct ext4_ext_path *path = *ppath;
 	struct ext4_extent *ex;
@@ -4347,7 +4347,7 @@ int ext4_ext_map_blocks(handle_t *handle, struct inode *inode,
 			    (flags & EXT4_GET_BLOCKS_CONVERT_UNWRITTEN)) {
 				allocated = convert_initialized_extent(
 						handle, inode, map, &path,
-						flags, allocated, newblock);
+						flags, allocated);
 				goto out2;
 			} else if (!ext4_ext_is_unwritten(ex))
 				goto out;
diff --git a/fs/ext4/file.c b/fs/ext4/file.c
index 1126436dada1..4cd318f31cbe 100644
--- a/fs/ext4/file.c
+++ b/fs/ext4/file.c
@@ -262,23 +262,8 @@ static int ext4_dax_pmd_fault(struct vm_area_struct *vma, unsigned long addr,
 	return result;
 }
 
-static int ext4_dax_mkwrite(struct vm_area_struct *vma, struct vm_fault *vmf)
-{
-	int err;
-	struct inode *inode = file_inode(vma->vm_file);
-
-	sb_start_pagefault(inode->i_sb);
-	file_update_time(vma->vm_file);
-	down_read(&EXT4_I(inode)->i_mmap_sem);
-	err = __dax_mkwrite(vma, vmf, ext4_dax_mmap_get_block, NULL);
-	up_read(&EXT4_I(inode)->i_mmap_sem);
-	sb_end_pagefault(inode->i_sb);
-
-	return err;
-}
-
 /*
- * Handle write fault for VM_MIXEDMAP mappings. Similarly to ext4_dax_mkwrite()
+ * Handle write fault for VM_MIXEDMAP mappings. Similarly to ext4_dax_fault()
  * handler we check for races agaist truncate. Note that since we cycle through
  * i_mmap_sem, we are sure that also any hole punching that began before we
  * were called is finished by now and so if it included part of the file we
@@ -311,7 +296,7 @@ static int ext4_dax_pfn_mkwrite(struct vm_area_struct *vma,
 static const struct vm_operations_struct ext4_dax_vm_ops = {
 	.fault		= ext4_dax_fault,
 	.pmd_fault	= ext4_dax_pmd_fault,
-	.page_mkwrite	= ext4_dax_mkwrite,
+	.page_mkwrite	= ext4_dax_fault,
 	.pfn_mkwrite	= ext4_dax_pfn_mkwrite,
 };
 #else
@@ -350,6 +335,7 @@ static int ext4_file_open(struct inode * inode, struct file * filp)
 	struct super_block *sb = inode->i_sb;
 	struct ext4_sb_info *sbi = EXT4_SB(inode->i_sb);
 	struct vfsmount *mnt = filp->f_path.mnt;
+	struct inode *dir = filp->f_path.dentry->d_parent->d_inode;
 	struct path path;
 	char buf[64], *cp;
 	int ret;
@@ -393,6 +379,14 @@ static int ext4_file_open(struct inode * inode, struct file * filp)
 		if (ext4_encryption_info(inode) == NULL)
 			return -ENOKEY;
 	}
+	if (ext4_encrypted_inode(dir) &&
+	    !ext4_is_child_context_consistent_with_parent(dir, inode)) {
+		ext4_warning(inode->i_sb,
+			     "Inconsistent encryption contexts: %lu/%lu\n",
+			     (unsigned long) dir->i_ino,
+			     (unsigned long) inode->i_ino);
+		return -EPERM;
+	}
 	/*
 	 * Set up the jbd2_inode if we are opening the inode for
 	 * writing and the journal is present
diff --git a/fs/ext4/ialloc.c b/fs/ext4/ialloc.c
index 3fcfd50a2e8a..acc0ad56bf2f 100644
--- a/fs/ext4/ialloc.c
+++ b/fs/ext4/ialloc.c
@@ -76,7 +76,6 @@ static int ext4_init_inode_bitmap(struct super_block *sb,
 	/* If checksum is bad mark all blocks and inodes use to prevent
 	 * allocation, essentially implementing a per-group read-only flag. */
 	if (!ext4_group_desc_csum_verify(sb, block_group, gdp)) {
-		ext4_error(sb, "Checksum bad for group %u", block_group);
 		grp = ext4_get_group_info(sb, block_group);
 		if (!EXT4_MB_GRP_BBITMAP_CORRUPT(grp))
 			percpu_counter_sub(&sbi->s_freeclusters_counter,
@@ -191,8 +190,11 @@ ext4_read_inode_bitmap(struct super_block *sb, ext4_group_t block_group)
 		set_buffer_verified(bh);
 		ext4_unlock_group(sb, block_group);
 		unlock_buffer(bh);
-		if (err)
+		if (err) {
+			ext4_error(sb, "Failed to init inode bitmap for group "
+				   "%u: %d", block_group, err);
 			goto out;
+		}
 		return bh;
 	}
 	ext4_unlock_group(sb, block_group);
diff --git a/fs/ext4/inode.c b/fs/ext4/inode.c
index 83bc8bfb3bea..aee960b1af34 100644
--- a/fs/ext4/inode.c
+++ b/fs/ext4/inode.c
@@ -686,6 +686,34 @@ out_sem:
 	return retval;
 }
 
+/*
+ * Update EXT4_MAP_FLAGS in bh->b_state. For buffer heads attached to pages
+ * we have to be careful as someone else may be manipulating b_state as well.
+ */
+static void ext4_update_bh_state(struct buffer_head *bh, unsigned long flags)
+{
+	unsigned long old_state;
+	unsigned long new_state;
+
+	flags &= EXT4_MAP_FLAGS;
+
+	/* Dummy buffer_head? Set non-atomically. */
+	if (!bh->b_page) {
+		bh->b_state = (bh->b_state & ~EXT4_MAP_FLAGS) | flags;
+		return;
+	}
+	/*
+	 * Someone else may be modifying b_state. Be careful! This is ugly but
+	 * once we get rid of using bh as a container for mapping information
+	 * to pass to / from get_block functions, this can go away.
+	 */
+	do {
+		old_state = READ_ONCE(bh->b_state);
+		new_state = (old_state & ~EXT4_MAP_FLAGS) | flags;
+	} while (unlikely(
+		 cmpxchg(&bh->b_state, old_state, new_state) != old_state));
+}
+
 /* Maximum number of blocks we map for direct IO at once. */
 #define DIO_MAX_BLOCKS 4096
 
@@ -722,7 +750,7 @@ static int _ext4_get_block(struct inode *inode, sector_t iblock,
 		ext4_io_end_t *io_end = ext4_inode_aio(inode);
 
 		map_bh(bh, inode->i_sb, map.m_pblk);
-		bh->b_state = (bh->b_state & ~EXT4_MAP_FLAGS) | map.m_flags;
+		ext4_update_bh_state(bh, map.m_flags);
 		if (io_end && io_end->flag & EXT4_IO_END_UNWRITTEN)
 			set_buffer_defer_completion(bh);
 		bh->b_size = inode->i_sb->s_blocksize * map.m_len;
@@ -1685,7 +1713,7 @@ int ext4_da_get_block_prep(struct inode *inode, sector_t iblock,
 		return ret;
 
 	map_bh(bh, inode->i_sb, map.m_pblk);
-	bh->b_state = (bh->b_state & ~EXT4_MAP_FLAGS) | map.m_flags;
+	ext4_update_bh_state(bh, map.m_flags);
 
 	if (buffer_unwritten(bh)) {
 		/* A delayed write to unwritten bh should be marked
@@ -2450,6 +2478,10 @@ static int ext4_writepages(struct address_space *mapping,
 
 	trace_ext4_writepages(inode, wbc);
 
+	if (dax_mapping(mapping))
+		return dax_writeback_mapping_range(mapping, inode->i_sb->s_bdev,
+						   wbc);
+
 	/*
 	 * No pages to write? This is mainly a kludge to avoid starting
 	 * a transaction for special inodes like journal inode on last iput()
@@ -3253,29 +3285,29 @@ static ssize_t ext4_ext_direct_IO(struct kiocb *iocb, struct iov_iter *iter,
 	 * case, we allocate an io_end structure to hook to the iocb.
 	 */
 	iocb->private = NULL;
-	ext4_inode_aio_set(inode, NULL);
-	if (!is_sync_kiocb(iocb)) {
-		io_end = ext4_init_io_end(inode, GFP_NOFS);
-		if (!io_end) {
-			ret = -ENOMEM;
-			goto retake_lock;
-		}
-		/*
-		 * Grab reference for DIO. Will be dropped in ext4_end_io_dio()
-		 */
-		iocb->private = ext4_get_io_end(io_end);
-		/*
-		 * we save the io structure for current async direct
-		 * IO, so that later ext4_map_blocks() could flag the
-		 * io structure whether there is a unwritten extents
-		 * needs to be converted when IO is completed.
-		 */
-		ext4_inode_aio_set(inode, io_end);
-	}
-
 	if (overwrite) {
 		get_block_func = ext4_get_block_overwrite;
 	} else {
+		ext4_inode_aio_set(inode, NULL);
+		if (!is_sync_kiocb(iocb)) {
+			io_end = ext4_init_io_end(inode, GFP_NOFS);
+			if (!io_end) {
+				ret = -ENOMEM;
+				goto retake_lock;
+			}
+			/*
+			 * Grab reference for DIO. Will be dropped in
+			 * ext4_end_io_dio()
+			 */
+			iocb->private = ext4_get_io_end(io_end);
+			/*
+			 * we save the io structure for current async direct
+			 * IO, so that later ext4_map_blocks() could flag the
+			 * io structure whether there is a unwritten extents
+			 * needs to be converted when IO is completed.
+			 */
+			ext4_inode_aio_set(inode, io_end);
+		}
 		get_block_func = ext4_get_block_write;
 		dio_flags = DIO_LOCKING;
 	}
@@ -4127,7 +4159,7 @@ void ext4_set_inode_flags(struct inode *inode)
 		new_fl |= S_NOATIME;
 	if (flags & EXT4_DIRSYNC_FL)
 		new_fl |= S_DIRSYNC;
-	if (test_opt(inode->i_sb, DAX))
+	if (test_opt(inode->i_sb, DAX) && S_ISREG(inode->i_mode))
 		new_fl |= S_DAX;
 	inode_set_flags(inode, new_fl,
 			S_SYNC|S_APPEND|S_IMMUTABLE|S_NOATIME|S_DIRSYNC|S_DAX);
diff --git a/fs/ext4/ioctl.c b/fs/ext4/ioctl.c
index 0f6c36922c24..eae5917c534e 100644
--- a/fs/ext4/ioctl.c
+++ b/fs/ext4/ioctl.c
@@ -208,7 +208,7 @@ static int ext4_ioctl_setflags(struct inode *inode,
 {
 	struct ext4_inode_info *ei = EXT4_I(inode);
 	handle_t *handle = NULL;
-	int err = EPERM, migrate = 0;
+	int err = -EPERM, migrate = 0;
 	struct ext4_iloc iloc;
 	unsigned int oldflags, mask, i;
 	unsigned int jflag;
@@ -583,6 +583,11 @@ group_extend_out:
 				 "Online defrag not supported with bigalloc");
 			err = -EOPNOTSUPP;
 			goto mext_out;
+		} else if (IS_DAX(inode)) {
+			ext4_msg(sb, KERN_ERR,
+				 "Online defrag not supported with DAX");
+			err = -EOPNOTSUPP;
+			goto mext_out;
 		}
 
 		err = mnt_want_write_file(filp);
diff --git a/fs/ext4/mballoc.c b/fs/ext4/mballoc.c
index 61eaf74dca37..4424b7bf8ac6 100644
--- a/fs/ext4/mballoc.c
+++ b/fs/ext4/mballoc.c
@@ -2285,7 +2285,7 @@ static int ext4_mb_seq_groups_show(struct seq_file *seq, void *v)
 	if (group == 0)
 		seq_puts(seq, "#group: free  frags first ["
 			      " 2^0   2^1   2^2   2^3   2^4   2^5   2^6  "
-			      " 2^7   2^8   2^9   2^10  2^11  2^12  2^13  ]");
+			      " 2^7   2^8   2^9   2^10  2^11  2^12  2^13  ]\n");
 
 	i = (sb->s_blocksize_bits + 2) * sizeof(sg.info.bb_counters[0]) +
 		sizeof(struct ext4_group_info);
diff --git a/fs/ext4/move_extent.c b/fs/ext4/move_extent.c
index fb6f11709ae6..4098acc701c3 100644
--- a/fs/ext4/move_extent.c
+++ b/fs/ext4/move_extent.c
@@ -265,11 +265,12 @@ move_extent_per_page(struct file *o_filp, struct inode *donor_inode,
 	ext4_lblk_t orig_blk_offset, donor_blk_offset;
 	unsigned long blocksize = orig_inode->i_sb->s_blocksize;
 	unsigned int tmp_data_size, data_size, replaced_size;
-	int err2, jblocks, retries = 0;
+	int i, err2, jblocks, retries = 0;
 	int replaced_count = 0;
 	int from = data_offset_in_page << orig_inode->i_blkbits;
 	int blocks_per_page = PAGE_CACHE_SIZE >> orig_inode->i_blkbits;
 	struct super_block *sb = orig_inode->i_sb;
+	struct buffer_head *bh = NULL;
 
 	/*
 	 * It needs twice the amount of ordinary journal buffers because
@@ -380,8 +381,17 @@ data_copy:
 	}
 	/* Perform all necessary steps similar write_begin()/write_end()
 	 * but keeping in mind that i_size will not change */
-	*err = __block_write_begin(pagep[0], from, replaced_size,
-				   ext4_get_block);
+	if (!page_has_buffers(pagep[0]))
+		create_empty_buffers(pagep[0], 1 << orig_inode->i_blkbits, 0);
+	bh = page_buffers(pagep[0]);
+	for (i = 0; i < data_offset_in_page; i++)
+		bh = bh->b_this_page;
+	for (i = 0; i < block_len_in_page; i++) {
+		*err = ext4_get_block(orig_inode, orig_blk_offset + i, bh, 0);
+		if (*err < 0)
+			break;
+		bh = bh->b_this_page;
+	}
 	if (!*err)
 		*err = block_commit_write(pagep[0], from, from + replaced_size);
 
diff --git a/fs/ext4/namei.c b/fs/ext4/namei.c
index 06574dd77614..48e4b8907826 100644
--- a/fs/ext4/namei.c
+++ b/fs/ext4/namei.c
@@ -1558,6 +1558,24 @@ static struct dentry *ext4_lookup(struct inode *dir, struct dentry *dentry, unsi
 	struct ext4_dir_entry_2 *de;
 	struct buffer_head *bh;
 
+       if (ext4_encrypted_inode(dir)) {
+               int res = ext4_get_encryption_info(dir);
+
+		/*
+		 * This should be a properly defined flag for
+		 * dentry->d_flags when we uplift this to the VFS.
+		 * d_fsdata is set to (void *) 1 if if the dentry is
+		 * created while the directory was encrypted and we
+		 * don't have access to the key.
+		 */
+	       dentry->d_fsdata = NULL;
+	       if (ext4_encryption_info(dir))
+		       dentry->d_fsdata = (void *) 1;
+	       d_set_d_op(dentry, &ext4_encrypted_d_ops);
+	       if (res && res != -ENOKEY)
+		       return ERR_PTR(res);
+       }
+
 	if (dentry->d_name.len > EXT4_NAME_LEN)
 		return ERR_PTR(-ENAMETOOLONG);
 
@@ -1585,11 +1603,15 @@ static struct dentry *ext4_lookup(struct inode *dir, struct dentry *dentry, unsi
 			return ERR_PTR(-EFSCORRUPTED);
 		}
 		if (!IS_ERR(inode) && ext4_encrypted_inode(dir) &&
-		    (S_ISREG(inode->i_mode) || S_ISDIR(inode->i_mode) ||
-		     S_ISLNK(inode->i_mode)) &&
+		    (S_ISDIR(inode->i_mode) || S_ISLNK(inode->i_mode)) &&
 		    !ext4_is_child_context_consistent_with_parent(dir,
 								  inode)) {
+			int nokey = ext4_encrypted_inode(inode) &&
+				!ext4_encryption_info(inode);
+
 			iput(inode);
+			if (nokey)
+				return ERR_PTR(-ENOKEY);
 			ext4_warning(inode->i_sb,
 				     "Inconsistent encryption contexts: %lu/%lu\n",
 				     (unsigned long) dir->i_ino,
diff --git a/fs/ext4/resize.c b/fs/ext4/resize.c
index ad62d7acc315..34038e3598d5 100644
--- a/fs/ext4/resize.c
+++ b/fs/ext4/resize.c
@@ -198,7 +198,7 @@ static struct ext4_new_flex_group_data *alloc_flex_gd(unsigned long flexbg_size)
 	if (flex_gd == NULL)
 		goto out3;
 
-	if (flexbg_size >= UINT_MAX / sizeof(struct ext4_new_flex_group_data))
+	if (flexbg_size >= UINT_MAX / sizeof(struct ext4_new_group_data))
 		goto out2;
 	flex_gd->count = flexbg_size;
 
diff --git a/fs/fs-writeback.c b/fs/fs-writeback.c
index 6915c950e6e8..5c46ed9f3e14 100644
--- a/fs/fs-writeback.c
+++ b/fs/fs-writeback.c
@@ -223,6 +223,9 @@ static void wb_wait_for_completion(struct backing_dev_info *bdi,
 #define WB_FRN_HIST_MAX_SLOTS	(WB_FRN_HIST_THR_SLOTS / 2 + 1)
 					/* one round can affect upto 5 slots */
 
+static atomic_t isw_nr_in_flight = ATOMIC_INIT(0);
+static struct workqueue_struct *isw_wq;
+
 void __inode_attach_wb(struct inode *inode, struct page *page)
 {
 	struct backing_dev_info *bdi = inode_to_bdi(inode);
@@ -424,6 +427,8 @@ skip_switch:
 
 	iput(inode);
 	kfree(isw);
+
+	atomic_dec(&isw_nr_in_flight);
 }
 
 static void inode_switch_wbs_rcu_fn(struct rcu_head *rcu_head)
@@ -433,7 +438,7 @@ static void inode_switch_wbs_rcu_fn(struct rcu_head *rcu_head)
 
 	/* needs to grab bh-unsafe locks, bounce to work item */
 	INIT_WORK(&isw->work, inode_switch_wbs_work_fn);
-	schedule_work(&isw->work);
+	queue_work(isw_wq, &isw->work);
 }
 
 /**
@@ -469,7 +474,8 @@ static void inode_switch_wbs(struct inode *inode, int new_wb_id)
 
 	/* while holding I_WB_SWITCH, no one else can update the association */
 	spin_lock(&inode->i_lock);
-	if (inode->i_state & (I_WB_SWITCH | I_FREEING) ||
+	if (!(inode->i_sb->s_flags & MS_ACTIVE) ||
+	    inode->i_state & (I_WB_SWITCH | I_FREEING) ||
 	    inode_to_wb(inode) == isw->new_wb) {
 		spin_unlock(&inode->i_lock);
 		goto out_free;
@@ -480,6 +486,8 @@ static void inode_switch_wbs(struct inode *inode, int new_wb_id)
 	ihold(inode);
 	isw->inode = inode;
 
+	atomic_inc(&isw_nr_in_flight);
+
 	/*
 	 * In addition to synchronizing among switchers, I_WB_SWITCH tells
 	 * the RCU protected stat update paths to grab the mapping's
@@ -840,6 +848,33 @@ restart:
 		wb_put(last_wb);
 }
 
+/**
+ * cgroup_writeback_umount - flush inode wb switches for umount
+ *
+ * This function is called when a super_block is about to be destroyed and
+ * flushes in-flight inode wb switches.  An inode wb switch goes through
+ * RCU and then workqueue, so the two need to be flushed in order to ensure
+ * that all previously scheduled switches are finished.  As wb switches are
+ * rare occurrences and synchronize_rcu() can take a while, perform
+ * flushing iff wb switches are in flight.
+ */
+void cgroup_writeback_umount(void)
+{
+	if (atomic_read(&isw_nr_in_flight)) {
+		synchronize_rcu();
+		flush_workqueue(isw_wq);
+	}
+}
+
+static int __init cgroup_writeback_init(void)
+{
+	isw_wq = alloc_workqueue("inode_switch_wbs", 0, 0);
+	if (!isw_wq)
+		return -ENOMEM;
+	return 0;
+}
+fs_initcall(cgroup_writeback_init);
+
 #else	/* CONFIG_CGROUP_WRITEBACK */
 
 static struct bdi_writeback *
diff --git a/fs/hpfs/namei.c b/fs/hpfs/namei.c
index 506765afa1a3..bb8d67e2740a 100644
--- a/fs/hpfs/namei.c
+++ b/fs/hpfs/namei.c
@@ -376,12 +376,11 @@ static int hpfs_unlink(struct inode *dir, struct dentry *dentry)
 	struct inode *inode = d_inode(dentry);
 	dnode_secno dno;
 	int r;
-	int rep = 0;
 	int err;
 
 	hpfs_lock(dir->i_sb);
 	hpfs_adjust_length(name, &len);
-again:
+
 	err = -ENOENT;
 	de = map_dirent(dir, hpfs_i(dir)->i_dno, name, len, &dno, &qbh);
 	if (!de)
@@ -401,33 +400,9 @@ again:
 		hpfs_error(dir->i_sb, "there was error when removing dirent");
 		err = -EFSERROR;
 		break;
-	case 2:		/* no space for deleting, try to truncate file */
-
+	case 2:		/* no space for deleting */
 		err = -ENOSPC;
-		if (rep++)
-			break;
-
-		dentry_unhash(dentry);
-		if (!d_unhashed(dentry)) {
-			hpfs_unlock(dir->i_sb);
-			return -ENOSPC;
-		}
-		if (generic_permission(inode, MAY_WRITE) ||
-		    !S_ISREG(inode->i_mode) ||
-		    get_write_access(inode)) {
-			d_rehash(dentry);
-		} else {
-			struct iattr newattrs;
-			/*pr_info("truncating file before delete.\n");*/
-			newattrs.ia_size = 0;
-			newattrs.ia_valid = ATTR_SIZE | ATTR_CTIME;
-			err = notify_change(dentry, &newattrs, NULL);
-			put_write_access(inode);
-			if (!err)
-				goto again;
-		}
-		hpfs_unlock(dir->i_sb);
-		return -ENOSPC;
+		break;
 	default:
 		drop_nlink(inode);
 		err = 0;
diff --git a/fs/inode.c b/fs/inode.c
index 9f62db3bcc3e..69b8b526c194 100644
--- a/fs/inode.c
+++ b/fs/inode.c
@@ -154,6 +154,12 @@ int inode_init_always(struct super_block *sb, struct inode *inode)
 	inode->i_rdev = 0;
 	inode->dirtied_when = 0;
 
+#ifdef CONFIG_CGROUP_WRITEBACK
+	inode->i_wb_frn_winner = 0;
+	inode->i_wb_frn_avg_time = 0;
+	inode->i_wb_frn_history = 0;
+#endif
+
 	if (security_inode_alloc(inode))
 		goto out;
 	spin_lock_init(&inode->i_lock);
diff --git a/fs/jffs2/README.Locking b/fs/jffs2/README.Locking
index 3ea36554107f..8918ac905a3b 100644
--- a/fs/jffs2/README.Locking
+++ b/fs/jffs2/README.Locking
@@ -2,10 +2,6 @@
 	JFFS2 LOCKING DOCUMENTATION
 	---------------------------
 
-At least theoretically, JFFS2 does not require the Big Kernel Lock
-(BKL), which was always helpfully obtained for it by Linux 2.4 VFS
-code. It has its own locking, as described below.
-
 This document attempts to describe the existing locking rules for
 JFFS2. It is not expected to remain perfectly up to date, but ought to
 be fairly close.
@@ -69,6 +65,7 @@ Ordering constraints:
 	   any f->sem held.
 	2. Never attempt to lock two file mutexes in one thread.
 	   No ordering rules have been made for doing so.
+	3. Never lock a page cache page with f->sem held.
 
 
 	erase_completion_lock spinlock
diff --git a/fs/jffs2/build.c b/fs/jffs2/build.c
index 0ae91ad6df2d..b288c8ae1236 100644
--- a/fs/jffs2/build.c
+++ b/fs/jffs2/build.c
@@ -50,7 +50,8 @@ next_inode(int *i, struct jffs2_inode_cache *ic, struct jffs2_sb_info *c)
 
 
 static void jffs2_build_inode_pass1(struct jffs2_sb_info *c,
-				    struct jffs2_inode_cache *ic)
+				    struct jffs2_inode_cache *ic,
+				    int *dir_hardlinks)
 {
 	struct jffs2_full_dirent *fd;
 
@@ -69,19 +70,21 @@ static void jffs2_build_inode_pass1(struct jffs2_sb_info *c,
 			dbg_fsbuild("child \"%s\" (ino #%u) of dir ino #%u doesn't exist!\n",
 				  fd->name, fd->ino, ic->ino);
 			jffs2_mark_node_obsolete(c, fd->raw);
+			/* Clear the ic/raw union so it doesn't cause problems later. */
+			fd->ic = NULL;
 			continue;
 		}
 
+		/* From this point, fd->raw is no longer used so we can set fd->ic */
+		fd->ic = child_ic;
+		child_ic->pino_nlink++;
+		/* If we appear (at this stage) to have hard-linked directories,
+		 * set a flag to trigger a scan later */
 		if (fd->type == DT_DIR) {
-			if (child_ic->pino_nlink) {
-				JFFS2_ERROR("child dir \"%s\" (ino #%u) of dir ino #%u appears to be a hard link\n",
-					    fd->name, fd->ino, ic->ino);
-				/* TODO: What do we do about it? */
-			} else {
-				child_ic->pino_nlink = ic->ino;
-			}
-		} else
-			child_ic->pino_nlink++;
+			child_ic->flags |= INO_FLAGS_IS_DIR;
+			if (child_ic->pino_nlink > 1)
+				*dir_hardlinks = 1;
+		}
 
 		dbg_fsbuild("increased nlink for child \"%s\" (ino #%u)\n", fd->name, fd->ino);
 		/* Can't free scan_dents so far. We might need them in pass 2 */
@@ -95,8 +98,7 @@ static void jffs2_build_inode_pass1(struct jffs2_sb_info *c,
 */
 static int jffs2_build_filesystem(struct jffs2_sb_info *c)
 {
-	int ret;
-	int i;
+	int ret, i, dir_hardlinks = 0;
 	struct jffs2_inode_cache *ic;
 	struct jffs2_full_dirent *fd;
 	struct jffs2_full_dirent *dead_fds = NULL;
@@ -120,7 +122,7 @@ static int jffs2_build_filesystem(struct jffs2_sb_info *c)
 	/* Now scan the directory tree, increasing nlink according to every dirent found. */
 	for_each_inode(i, c, ic) {
 		if (ic->scan_dents) {
-			jffs2_build_inode_pass1(c, ic);
+			jffs2_build_inode_pass1(c, ic, &dir_hardlinks);
 			cond_resched();
 		}
 	}
@@ -156,6 +158,20 @@ static int jffs2_build_filesystem(struct jffs2_sb_info *c)
 	}
 
 	dbg_fsbuild("pass 2a complete\n");
+
+	if (dir_hardlinks) {
+		/* If we detected directory hardlinks earlier, *hopefully*
+		 * they are gone now because some of the links were from
+		 * dead directories which still had some old dirents lying
+		 * around and not yet garbage-collected, but which have
+		 * been discarded above. So clear the pino_nlink field
+		 * in each directory, so that the final scan below can
+		 * print appropriate warnings. */
+		for_each_inode(i, c, ic) {
+			if (ic->flags & INO_FLAGS_IS_DIR)
+				ic->pino_nlink = 0;
+		}
+	}
 	dbg_fsbuild("freeing temporary data structures\n");
 
 	/* Finally, we can scan again and free the dirent structs */
@@ -163,6 +179,33 @@ static int jffs2_build_filesystem(struct jffs2_sb_info *c)
 		while(ic->scan_dents) {
 			fd = ic->scan_dents;
 			ic->scan_dents = fd->next;
+			/* We do use the pino_nlink field to count nlink of
+			 * directories during fs build, so set it to the
+			 * parent ino# now. Now that there's hopefully only
+			 * one. */
+			if (fd->type == DT_DIR) {
+				if (!fd->ic) {
+					/* We'll have complained about it and marked the coresponding
+					   raw node obsolete already. Just skip it. */
+					continue;
+				}
+
+				/* We *have* to have set this in jffs2_build_inode_pass1() */
+				BUG_ON(!(fd->ic->flags & INO_FLAGS_IS_DIR));
+
+				/* We clear ic->pino_nlink ∀ directories' ic *only* if dir_hardlinks
+				 * is set. Otherwise, we know this should never trigger anyway, so
+				 * we don't do the check. And ic->pino_nlink still contains the nlink
+				 * value (which is 1). */
+				if (dir_hardlinks && fd->ic->pino_nlink) {
+					JFFS2_ERROR("child dir \"%s\" (ino #%u) of dir ino #%u is also hard linked from dir ino #%u\n",
+						    fd->name, fd->ino, ic->ino, fd->ic->pino_nlink);
+					/* Should we unlink it from its previous parent? */
+				}
+
+				/* For directories, ic->pino_nlink holds that parent inode # */
+				fd->ic->pino_nlink = ic->ino;
+			}
 			jffs2_free_full_dirent(fd);
 		}
 		ic->scan_dents = NULL;
@@ -241,11 +284,7 @@ static void jffs2_build_remove_unlinked_inode(struct jffs2_sb_info *c,
 
 			/* Reduce nlink of the child. If it's now zero, stick it on the
 			   dead_fds list to be cleaned up later. Else just free the fd */
-
-			if (fd->type == DT_DIR)
-				child_ic->pino_nlink = 0;
-			else
-				child_ic->pino_nlink--;
+			child_ic->pino_nlink--;
 
 			if (!child_ic->pino_nlink) {
 				dbg_fsbuild("inode #%u (\"%s\") now has no links; adding to dead_fds list.\n",
diff --git a/fs/jffs2/dir.c b/fs/jffs2/dir.c
index d211b8e18566..30c4c9ebb693 100644
--- a/fs/jffs2/dir.c
+++ b/fs/jffs2/dir.c
@@ -843,9 +843,14 @@ static int jffs2_rename (struct inode *old_dir_i, struct dentry *old_dentry,
 
 		pr_notice("%s(): Link succeeded, unlink failed (err %d). You now have a hard link\n",
 			  __func__, ret);
-		/* Might as well let the VFS know */
-		d_instantiate(new_dentry, d_inode(old_dentry));
-		ihold(d_inode(old_dentry));
+		/*
+		 * We can't keep the target in dcache after that.
+		 * For one thing, we can't afford dentry aliases for directories.
+		 * For another, if there was a victim, we _can't_ set new inode
+		 * for that sucker and we have to trigger mount eviction - the
+		 * caller won't do it on its own since we are returning an error.
+		 */
+		d_invalidate(new_dentry);
 		new_dir_i->i_mtime = new_dir_i->i_ctime = ITIME(now);
 		return ret;
 	}
diff --git a/fs/jffs2/file.c b/fs/jffs2/file.c
index c5ac5944bc1b..cad86bac3453 100644
--- a/fs/jffs2/file.c
+++ b/fs/jffs2/file.c
@@ -137,39 +137,33 @@ static int jffs2_write_begin(struct file *filp, struct address_space *mapping,
 	struct page *pg;
 	struct inode *inode = mapping->host;
 	struct jffs2_inode_info *f = JFFS2_INODE_INFO(inode);
-	struct jffs2_sb_info *c = JFFS2_SB_INFO(inode->i_sb);
-	struct jffs2_raw_inode ri;
-	uint32_t alloc_len = 0;
 	pgoff_t index = pos >> PAGE_CACHE_SHIFT;
 	uint32_t pageofs = index << PAGE_CACHE_SHIFT;
 	int ret = 0;
 
-	jffs2_dbg(1, "%s()\n", __func__);
-
-	if (pageofs > inode->i_size) {
-		ret = jffs2_reserve_space(c, sizeof(ri), &alloc_len,
-					  ALLOC_NORMAL, JFFS2_SUMMARY_INODE_SIZE);
-		if (ret)
-			return ret;
-	}
-
-	mutex_lock(&f->sem);
 	pg = grab_cache_page_write_begin(mapping, index, flags);
-	if (!pg) {
-		if (alloc_len)
-			jffs2_complete_reservation(c);
-		mutex_unlock(&f->sem);
+	if (!pg)
 		return -ENOMEM;
-	}
 	*pagep = pg;
 
-	if (alloc_len) {
+	jffs2_dbg(1, "%s()\n", __func__);
+
+	if (pageofs > inode->i_size) {
 		/* Make new hole frag from old EOF to new page */
+		struct jffs2_sb_info *c = JFFS2_SB_INFO(inode->i_sb);
+		struct jffs2_raw_inode ri;
 		struct jffs2_full_dnode *fn;
+		uint32_t alloc_len;
 
 		jffs2_dbg(1, "Writing new hole frag 0x%x-0x%x between current EOF and new page\n",
 			  (unsigned int)inode->i_size, pageofs);
 
+		ret = jffs2_reserve_space(c, sizeof(ri), &alloc_len,
+					  ALLOC_NORMAL, JFFS2_SUMMARY_INODE_SIZE);
+		if (ret)
+			goto out_page;
+
+		mutex_lock(&f->sem);
 		memset(&ri, 0, sizeof(ri));
 
 		ri.magic = cpu_to_je16(JFFS2_MAGIC_BITMASK);
@@ -196,6 +190,7 @@ static int jffs2_write_begin(struct file *filp, struct address_space *mapping,
 		if (IS_ERR(fn)) {
 			ret = PTR_ERR(fn);
 			jffs2_complete_reservation(c);
+			mutex_unlock(&f->sem);
 			goto out_page;
 		}
 		ret = jffs2_add_full_dnode_to_inode(c, f, fn);
@@ -210,10 +205,12 @@ static int jffs2_write_begin(struct file *filp, struct address_space *mapping,
 			jffs2_mark_node_obsolete(c, fn->raw);
 			jffs2_free_full_dnode(fn);
 			jffs2_complete_reservation(c);
+			mutex_unlock(&f->sem);
 			goto out_page;
 		}
 		jffs2_complete_reservation(c);
 		inode->i_size = pageofs;
+		mutex_unlock(&f->sem);
 	}
 
 	/*
@@ -222,18 +219,18 @@ static int jffs2_write_begin(struct file *filp, struct address_space *mapping,
 	 * case of a short-copy.
 	 */
 	if (!PageUptodate(pg)) {
+		mutex_lock(&f->sem);
 		ret = jffs2_do_readpage_nolock(inode, pg);
+		mutex_unlock(&f->sem);
 		if (ret)
 			goto out_page;
 	}
-	mutex_unlock(&f->sem);
 	jffs2_dbg(1, "end write_begin(). pg->flags %lx\n", pg->flags);
 	return ret;
 
 out_page:
 	unlock_page(pg);
 	page_cache_release(pg);
-	mutex_unlock(&f->sem);
 	return ret;
 }
 
diff --git a/fs/jffs2/gc.c b/fs/jffs2/gc.c
index 5a2dec2b064c..95d5880a63ee 100644
--- a/fs/jffs2/gc.c
+++ b/fs/jffs2/gc.c
@@ -1296,14 +1296,17 @@ static int jffs2_garbage_collect_dnode(struct jffs2_sb_info *c, struct jffs2_era
 		BUG_ON(start > orig_start);
 	}
 
-	/* First, use readpage() to read the appropriate page into the page cache */
-	/* Q: What happens if we actually try to GC the _same_ page for which commit_write()
-	 *    triggered garbage collection in the first place?
-	 * A: I _think_ it's OK. read_cache_page shouldn't deadlock, we'll write out the
-	 *    page OK. We'll actually write it out again in commit_write, which is a little
-	 *    suboptimal, but at least we're correct.
-	 */
+	/* The rules state that we must obtain the page lock *before* f->sem, so
+	 * drop f->sem temporarily. Since we also hold c->alloc_sem, nothing's
+	 * actually going to *change* so we're safe; we only allow reading.
+	 *
+	 * It is important to note that jffs2_write_begin() will ensure that its
+	 * page is marked Uptodate before allocating space. That means that if we
+	 * end up here trying to GC the *same* page that jffs2_write_begin() is
+	 * trying to write out, read_cache_page() will not deadlock. */
+	mutex_unlock(&f->sem);
 	pg_ptr = jffs2_gc_fetch_page(c, f, start, &pg);
+	mutex_lock(&f->sem);
 
 	if (IS_ERR(pg_ptr)) {
 		pr_warn("read_cache_page() returned error: %ld\n",
diff --git a/fs/jffs2/nodelist.h b/fs/jffs2/nodelist.h
index fa35ff79ab35..0637271f3770 100644
--- a/fs/jffs2/nodelist.h
+++ b/fs/jffs2/nodelist.h
@@ -194,6 +194,7 @@ struct jffs2_inode_cache {
 #define INO_STATE_CLEARING	6	/* In clear_inode() */
 
 #define INO_FLAGS_XATTR_CHECKED	0x01	/* has no duplicate xattr_ref */
+#define INO_FLAGS_IS_DIR	0x02	/* is a directory */
 
 #define RAWNODE_CLASS_INODE_CACHE	0
 #define RAWNODE_CLASS_XATTR_DATUM	1
@@ -249,7 +250,10 @@ struct jffs2_readinode_info
 
 struct jffs2_full_dirent
 {
-	struct jffs2_raw_node_ref *raw;
+	union {
+		struct jffs2_raw_node_ref *raw;
+		struct jffs2_inode_cache *ic; /* Just during part of build */
+	};
 	struct jffs2_full_dirent *next;
 	uint32_t version;
 	uint32_t ino; /* == zero for unlink */
diff --git a/fs/namei.c b/fs/namei.c
index f624d132e01e..9c590e0f66e9 100644
--- a/fs/namei.c
+++ b/fs/namei.c
@@ -1712,6 +1712,11 @@ static inline int should_follow_link(struct nameidata *nd, struct path *link,
 		return 0;
 	if (!follow)
 		return 0;
+	/* make sure that d_is_symlink above matches inode */
+	if (nd->flags & LOOKUP_RCU) {
+		if (read_seqcount_retry(&link->dentry->d_seq, seq))
+			return -ECHILD;
+	}
 	return pick_link(nd, link, inode, seq);
 }
 
@@ -1743,11 +1748,11 @@ static int walk_component(struct nameidata *nd, int flags)
 		if (err < 0)
 			return err;
 
-		inode = d_backing_inode(path.dentry);
 		seq = 0;	/* we are already out of RCU mode */
 		err = -ENOENT;
 		if (d_is_negative(path.dentry))
 			goto out_path_put;
+		inode = d_backing_inode(path.dentry);
 	}
 
 	if (flags & WALK_PUT)
@@ -3192,12 +3197,12 @@ retry_lookup:
 		return error;
 
 	BUG_ON(nd->flags & LOOKUP_RCU);
-	inode = d_backing_inode(path.dentry);
 	seq = 0;	/* out of RCU mode, so the value doesn't matter */
 	if (unlikely(d_is_negative(path.dentry))) {
 		path_to_nameidata(&path, nd);
 		return -ENOENT;
 	}
+	inode = d_backing_inode(path.dentry);
 finish_lookup:
 	if (nd->depth)
 		put_link(nd);
@@ -3206,11 +3211,6 @@ finish_lookup:
 	if (unlikely(error))
 		return error;
 
-	if (unlikely(d_is_symlink(path.dentry)) && !(open_flag & O_PATH)) {
-		path_to_nameidata(&path, nd);
-		return -ELOOP;
-	}
-
 	if ((nd->flags & LOOKUP_RCU) || nd->path.mnt != path.mnt) {
 		path_to_nameidata(&path, nd);
 	} else {
@@ -3229,6 +3229,10 @@ finish_open:
 		return error;
 	}
 	audit_inode(nd->name, nd->path.dentry, 0);
+	if (unlikely(d_is_symlink(nd->path.dentry)) && !(open_flag & O_PATH)) {
+		error = -ELOOP;
+		goto out;
+	}
 	error = -EISDIR;
 	if ((open_flag & O_CREAT) && d_is_dir(nd->path.dentry))
 		goto out;
@@ -3273,6 +3277,10 @@ opened:
 			goto exit_fput;
 	}
 out:
+	if (unlikely(error > 0)) {
+		WARN_ON(1);
+		error = -EINVAL;
+	}
 	if (got_write)
 		mnt_drop_write(nd->path.mnt);
 	path_put(&save_parent);
diff --git a/fs/ncpfs/dir.c b/fs/ncpfs/dir.c
index 26c2de2de13f..b7f8eaeea5d8 100644
--- a/fs/ncpfs/dir.c
+++ b/fs/ncpfs/dir.c
@@ -633,7 +633,7 @@ ncp_fill_cache(struct file *file, struct dir_context *ctx,
 				d_rehash(newdent);
 		} else {
 			spin_lock(&dentry->d_lock);
-			NCP_FINFO(inode)->flags &= ~NCPI_DIR_CACHE;
+			NCP_FINFO(dir)->flags &= ~NCPI_DIR_CACHE;
 			spin_unlock(&dentry->d_lock);
 		}
 	} else {
diff --git a/fs/nfs/blocklayout/extent_tree.c b/fs/nfs/blocklayout/extent_tree.c
index c59a59c37f3d..35ab51c04814 100644
--- a/fs/nfs/blocklayout/extent_tree.c
+++ b/fs/nfs/blocklayout/extent_tree.c
@@ -476,6 +476,7 @@ static void ext_tree_free_commitdata(struct nfs4_layoutcommit_args *arg,
 
 		for (i = 0; i < nr_pages; i++)
 			put_page(arg->layoutupdate_pages[i]);
+		vfree(arg->start_p);
 		kfree(arg->layoutupdate_pages);
 	} else {
 		put_page(arg->layoutupdate_page);
@@ -559,10 +560,15 @@ retry:
 
 	if (unlikely(arg->layoutupdate_pages != &arg->layoutupdate_page)) {
 		void *p = start_p, *end = p + arg->layoutupdate_len;
+		struct page *page = NULL;
 		int i = 0;
 
-		for ( ; p < end; p += PAGE_SIZE)
-			arg->layoutupdate_pages[i++] = vmalloc_to_page(p);
+		arg->start_p = start_p;
+		for ( ; p < end; p += PAGE_SIZE) {
+			page = vmalloc_to_page(p);
+			arg->layoutupdate_pages[i++] = page;
+			get_page(page);
+		}
 	}
 
 	dprintk("%s found %zu ranges\n", __func__, count);
diff --git a/fs/nfs/nfs42proc.c b/fs/nfs/nfs42proc.c
index bd25dc7077f7..dff83460e5a6 100644
--- a/fs/nfs/nfs42proc.c
+++ b/fs/nfs/nfs42proc.c
@@ -16,29 +16,8 @@
 
 #define NFSDBG_FACILITY NFSDBG_PROC
 
-static int nfs42_set_rw_stateid(nfs4_stateid *dst, struct file *file,
-				fmode_t fmode)
-{
-	struct nfs_open_context *open;
-	struct nfs_lock_context *lock;
-	int ret;
-
-	open = get_nfs_open_context(nfs_file_open_context(file));
-	lock = nfs_get_lock_context(open);
-	if (IS_ERR(lock)) {
-		put_nfs_open_context(open);
-		return PTR_ERR(lock);
-	}
-
-	ret = nfs4_set_rw_stateid(dst, open, lock, fmode);
-
-	nfs_put_lock_context(lock);
-	put_nfs_open_context(open);
-	return ret;
-}
-
 static int _nfs42_proc_fallocate(struct rpc_message *msg, struct file *filep,
-				 loff_t offset, loff_t len)
+		struct nfs_lock_context *lock, loff_t offset, loff_t len)
 {
 	struct inode *inode = file_inode(filep);
 	struct nfs_server *server = NFS_SERVER(inode);
@@ -56,7 +35,8 @@ static int _nfs42_proc_fallocate(struct rpc_message *msg, struct file *filep,
 	msg->rpc_argp = &args;
 	msg->rpc_resp = &res;
 
-	status = nfs42_set_rw_stateid(&args.falloc_stateid, filep, FMODE_WRITE);
+	status = nfs4_set_rw_stateid(&args.falloc_stateid, lock->open_context,
+			lock, FMODE_WRITE);
 	if (status)
 		return status;
 
@@ -78,15 +58,26 @@ static int nfs42_proc_fallocate(struct rpc_message *msg, struct file *filep,
 {
 	struct nfs_server *server = NFS_SERVER(file_inode(filep));
 	struct nfs4_exception exception = { };
+	struct nfs_lock_context *lock;
 	int err;
 
+	lock = nfs_get_lock_context(nfs_file_open_context(filep));
+	if (IS_ERR(lock))
+		return PTR_ERR(lock);
+
+	exception.inode = file_inode(filep);
+	exception.state = lock->open_context->state;
+
 	do {
-		err = _nfs42_proc_fallocate(msg, filep, offset, len);
-		if (err == -ENOTSUPP)
-			return -EOPNOTSUPP;
+		err = _nfs42_proc_fallocate(msg, filep, lock, offset, len);
+		if (err == -ENOTSUPP) {
+			err = -EOPNOTSUPP;
+			break;
+		}
 		err = nfs4_handle_exception(server, err, &exception);
 	} while (exception.retry);
 
+	nfs_put_lock_context(lock);
 	return err;
 }
 
@@ -135,7 +126,8 @@ int nfs42_proc_deallocate(struct file *filep, loff_t offset, loff_t len)
 	return err;
 }
 
-static loff_t _nfs42_proc_llseek(struct file *filep, loff_t offset, int whence)
+static loff_t _nfs42_proc_llseek(struct file *filep,
+		struct nfs_lock_context *lock, loff_t offset, int whence)
 {
 	struct inode *inode = file_inode(filep);
 	struct nfs42_seek_args args = {
@@ -156,7 +148,8 @@ static loff_t _nfs42_proc_llseek(struct file *filep, loff_t offset, int whence)
 	if (!nfs_server_capable(inode, NFS_CAP_SEEK))
 		return -ENOTSUPP;
 
-	status = nfs42_set_rw_stateid(&args.sa_stateid, filep, FMODE_READ);
+	status = nfs4_set_rw_stateid(&args.sa_stateid, lock->open_context,
+			lock, FMODE_READ);
 	if (status)
 		return status;
 
@@ -175,17 +168,28 @@ loff_t nfs42_proc_llseek(struct file *filep, loff_t offset, int whence)
 {
 	struct nfs_server *server = NFS_SERVER(file_inode(filep));
 	struct nfs4_exception exception = { };
+	struct nfs_lock_context *lock;
 	loff_t err;
 
+	lock = nfs_get_lock_context(nfs_file_open_context(filep));
+	if (IS_ERR(lock))
+		return PTR_ERR(lock);
+
+	exception.inode = file_inode(filep);
+	exception.state = lock->open_context->state;
+
 	do {
-		err = _nfs42_proc_llseek(filep, offset, whence);
+		err = _nfs42_proc_llseek(filep, lock, offset, whence);
 		if (err >= 0)
 			break;
-		if (err == -ENOTSUPP)
-			return -EOPNOTSUPP;
+		if (err == -ENOTSUPP) {
+			err = -EOPNOTSUPP;
+			break;
+		}
 		err = nfs4_handle_exception(server, err, &exception);
 	} while (exception.retry);
 
+	nfs_put_lock_context(lock);
 	return err;
 }
 
@@ -298,8 +302,9 @@ int nfs42_proc_layoutstats_generic(struct nfs_server *server,
 }
 
 static int _nfs42_proc_clone(struct rpc_message *msg, struct file *src_f,
-			     struct file *dst_f, loff_t src_offset,
-			     loff_t dst_offset, loff_t count)
+		struct file *dst_f, struct nfs_lock_context *src_lock,
+		struct nfs_lock_context *dst_lock, loff_t src_offset,
+		loff_t dst_offset, loff_t count)
 {
 	struct inode *src_inode = file_inode(src_f);
 	struct inode *dst_inode = file_inode(dst_f);
@@ -320,11 +325,13 @@ static int _nfs42_proc_clone(struct rpc_message *msg, struct file *src_f,
 	msg->rpc_argp = &args;
 	msg->rpc_resp = &res;
 
-	status = nfs42_set_rw_stateid(&args.src_stateid, src_f, FMODE_READ);
+	status = nfs4_set_rw_stateid(&args.src_stateid, src_lock->open_context,
+			src_lock, FMODE_READ);
 	if (status)
 		return status;
 
-	status = nfs42_set_rw_stateid(&args.dst_stateid, dst_f, FMODE_WRITE);
+	status = nfs4_set_rw_stateid(&args.dst_stateid, dst_lock->open_context,
+			dst_lock, FMODE_WRITE);
 	if (status)
 		return status;
 
@@ -349,22 +356,48 @@ int nfs42_proc_clone(struct file *src_f, struct file *dst_f,
 	};
 	struct inode *inode = file_inode(src_f);
 	struct nfs_server *server = NFS_SERVER(file_inode(src_f));
-	struct nfs4_exception exception = { };
-	int err;
+	struct nfs_lock_context *src_lock;
+	struct nfs_lock_context *dst_lock;
+	struct nfs4_exception src_exception = { };
+	struct nfs4_exception dst_exception = { };
+	int err, err2;
 
 	if (!nfs_server_capable(inode, NFS_CAP_CLONE))
 		return -EOPNOTSUPP;
 
+	src_lock = nfs_get_lock_context(nfs_file_open_context(src_f));
+	if (IS_ERR(src_lock))
+		return PTR_ERR(src_lock);
+
+	src_exception.inode = file_inode(src_f);
+	src_exception.state = src_lock->open_context->state;
+
+	dst_lock = nfs_get_lock_context(nfs_file_open_context(dst_f));
+	if (IS_ERR(dst_lock)) {
+		err = PTR_ERR(dst_lock);
+		goto out_put_src_lock;
+	}
+
+	dst_exception.inode = file_inode(dst_f);
+	dst_exception.state = dst_lock->open_context->state;
+
 	do {
-		err = _nfs42_proc_clone(&msg, src_f, dst_f, src_offset,
-					dst_offset, count);
+		err = _nfs42_proc_clone(&msg, src_f, dst_f, src_lock, dst_lock,
+					src_offset, dst_offset, count);
 		if (err == -ENOTSUPP || err == -EOPNOTSUPP) {
 			NFS_SERVER(inode)->caps &= ~NFS_CAP_CLONE;
-			return -EOPNOTSUPP;
+			err = -EOPNOTSUPP;
+			break;
 		}
-		err = nfs4_handle_exception(server, err, &exception);
-	} while (exception.retry);
 
-	return err;
+		err2 = nfs4_handle_exception(server, err, &src_exception);
+		err = nfs4_handle_exception(server, err, &dst_exception);
+		if (!err)
+			err = err2;
+	} while (src_exception.retry || dst_exception.retry);
 
+	nfs_put_lock_context(dst_lock);
+out_put_src_lock:
+	nfs_put_lock_context(src_lock);
+	return err;
 }
diff --git a/fs/nfs/nfs4proc.c b/fs/nfs/nfs4proc.c
index 4bfc33ad0563..14881594dd07 100644
--- a/fs/nfs/nfs4proc.c
+++ b/fs/nfs/nfs4proc.c
@@ -2466,9 +2466,9 @@ static int _nfs4_open_and_get_state(struct nfs4_opendata *opendata,
 		dentry = d_add_unique(dentry, igrab(state->inode));
 		if (dentry == NULL) {
 			dentry = opendata->dentry;
-		} else if (dentry != ctx->dentry) {
+		} else {
 			dput(ctx->dentry);
-			ctx->dentry = dget(dentry);
+			ctx->dentry = dentry;
 		}
 		nfs_set_verifier(dentry,
 				nfs_save_change_attribute(d_inode(opendata->dir)));
diff --git a/fs/nfs/pnfs.c b/fs/nfs/pnfs.c
index 482b6e94bb37..2fa483e6dbe2 100644
--- a/fs/nfs/pnfs.c
+++ b/fs/nfs/pnfs.c
@@ -252,6 +252,27 @@ pnfs_put_layout_hdr(struct pnfs_layout_hdr *lo)
 	}
 }
 
+/*
+ * Mark a pnfs_layout_hdr and all associated layout segments as invalid
+ *
+ * In order to continue using the pnfs_layout_hdr, a full recovery
+ * is required.
+ * Note that caller must hold inode->i_lock.
+ */
+static int
+pnfs_mark_layout_stateid_invalid(struct pnfs_layout_hdr *lo,
+		struct list_head *lseg_list)
+{
+	struct pnfs_layout_range range = {
+		.iomode = IOMODE_ANY,
+		.offset = 0,
+		.length = NFS4_MAX_UINT64,
+	};
+
+	set_bit(NFS_LAYOUT_INVALID_STID, &lo->plh_flags);
+	return pnfs_mark_matching_lsegs_invalid(lo, lseg_list, &range);
+}
+
 static int
 pnfs_iomode_to_fail_bit(u32 iomode)
 {
@@ -554,9 +575,8 @@ pnfs_destroy_layout(struct nfs_inode *nfsi)
 	spin_lock(&nfsi->vfs_inode.i_lock);
 	lo = nfsi->layout;
 	if (lo) {
-		lo->plh_block_lgets++; /* permanently block new LAYOUTGETs */
-		pnfs_mark_matching_lsegs_invalid(lo, &tmp_list, NULL);
 		pnfs_get_layout_hdr(lo);
+		pnfs_mark_layout_stateid_invalid(lo, &tmp_list);
 		pnfs_layout_clear_fail_bit(lo, NFS_LAYOUT_RO_FAILED);
 		pnfs_layout_clear_fail_bit(lo, NFS_LAYOUT_RW_FAILED);
 		spin_unlock(&nfsi->vfs_inode.i_lock);
@@ -617,11 +637,6 @@ pnfs_layout_free_bulk_destroy_list(struct list_head *layout_list,
 {
 	struct pnfs_layout_hdr *lo;
 	struct inode *inode;
-	struct pnfs_layout_range range = {
-		.iomode = IOMODE_ANY,
-		.offset = 0,
-		.length = NFS4_MAX_UINT64,
-	};
 	LIST_HEAD(lseg_list);
 	int ret = 0;
 
@@ -636,11 +651,11 @@ pnfs_layout_free_bulk_destroy_list(struct list_head *layout_list,
 
 		spin_lock(&inode->i_lock);
 		list_del_init(&lo->plh_bulk_destroy);
-		lo->plh_block_lgets++; /* permanently block new LAYOUTGETs */
-		if (is_bulk_recall)
-			set_bit(NFS_LAYOUT_BULK_RECALL, &lo->plh_flags);
-		if (pnfs_mark_matching_lsegs_invalid(lo, &lseg_list, &range))
+		if (pnfs_mark_layout_stateid_invalid(lo, &lseg_list)) {
+			if (is_bulk_recall)
+				set_bit(NFS_LAYOUT_BULK_RECALL, &lo->plh_flags);
 			ret = -EAGAIN;
+		}
 		spin_unlock(&inode->i_lock);
 		pnfs_free_lseg_list(&lseg_list);
 		/* Free all lsegs that are attached to commit buckets */
@@ -1738,8 +1753,19 @@ pnfs_set_plh_return_iomode(struct pnfs_layout_hdr *lo, enum pnfs_iomode iomode)
 	if (lo->plh_return_iomode != 0)
 		iomode = IOMODE_ANY;
 	lo->plh_return_iomode = iomode;
+	set_bit(NFS_LAYOUT_RETURN_REQUESTED, &lo->plh_flags);
 }
 
+/**
+ * pnfs_mark_matching_lsegs_return - Free or return matching layout segments
+ * @lo: pointer to layout header
+ * @tmp_list: list header to be used with pnfs_free_lseg_list()
+ * @return_range: describe layout segment ranges to be returned
+ *
+ * This function is mainly intended for use by layoutrecall. It attempts
+ * to free the layout segment immediately, or else to mark it for return
+ * as soon as its reference count drops to zero.
+ */
 int
 pnfs_mark_matching_lsegs_return(struct pnfs_layout_hdr *lo,
 				struct list_head *tmp_list,
@@ -1762,12 +1788,11 @@ pnfs_mark_matching_lsegs_return(struct pnfs_layout_hdr *lo,
 				lseg, lseg->pls_range.iomode,
 				lseg->pls_range.offset,
 				lseg->pls_range.length);
+			if (mark_lseg_invalid(lseg, tmp_list))
+				continue;
+			remaining++;
 			set_bit(NFS_LSEG_LAYOUTRETURN, &lseg->pls_flags);
 			pnfs_set_plh_return_iomode(lo, return_range->iomode);
-			if (!mark_lseg_invalid(lseg, tmp_list))
-				remaining++;
-			set_bit(NFS_LAYOUT_RETURN_REQUESTED,
-					&lo->plh_flags);
 		}
 	return remaining;
 }
diff --git a/fs/notify/mark.c b/fs/notify/mark.c
index cfcbf114676e..7115c5d7d373 100644
--- a/fs/notify/mark.c
+++ b/fs/notify/mark.c
@@ -91,7 +91,14 @@
 #include <linux/fsnotify_backend.h>
 #include "fsnotify.h"
 
+#define FSNOTIFY_REAPER_DELAY	(1)	/* 1 jiffy */
+
 struct srcu_struct fsnotify_mark_srcu;
+static DEFINE_SPINLOCK(destroy_lock);
+static LIST_HEAD(destroy_list);
+
+static void fsnotify_mark_destroy(struct work_struct *work);
+static DECLARE_DELAYED_WORK(reaper_work, fsnotify_mark_destroy);
 
 void fsnotify_get_mark(struct fsnotify_mark *mark)
 {
@@ -165,19 +172,10 @@ void fsnotify_detach_mark(struct fsnotify_mark *mark)
 	atomic_dec(&group->num_marks);
 }
 
-static void
-fsnotify_mark_free_rcu(struct rcu_head *rcu)
-{
-	struct fsnotify_mark	*mark;
-
-	mark = container_of(rcu, struct fsnotify_mark, g_rcu);
-	fsnotify_put_mark(mark);
-}
-
 /*
- * Free fsnotify mark. The freeing is actually happening from a call_srcu
- * callback. Caller must have a reference to the mark or be protected by
- * fsnotify_mark_srcu.
+ * Free fsnotify mark. The freeing is actually happening from a kthread which
+ * first waits for srcu period end. Caller must have a reference to the mark
+ * or be protected by fsnotify_mark_srcu.
  */
 void fsnotify_free_mark(struct fsnotify_mark *mark)
 {
@@ -192,7 +190,11 @@ void fsnotify_free_mark(struct fsnotify_mark *mark)
 	mark->flags &= ~FSNOTIFY_MARK_FLAG_ALIVE;
 	spin_unlock(&mark->lock);
 
-	call_srcu(&fsnotify_mark_srcu, &mark->g_rcu, fsnotify_mark_free_rcu);
+	spin_lock(&destroy_lock);
+	list_add(&mark->g_list, &destroy_list);
+	spin_unlock(&destroy_lock);
+	queue_delayed_work(system_unbound_wq, &reaper_work,
+				FSNOTIFY_REAPER_DELAY);
 
 	/*
 	 * Some groups like to know that marks are being freed.  This is a
@@ -388,7 +390,12 @@ err:
 
 	spin_unlock(&mark->lock);
 
-	call_srcu(&fsnotify_mark_srcu, &mark->g_rcu, fsnotify_mark_free_rcu);
+	spin_lock(&destroy_lock);
+	list_add(&mark->g_list, &destroy_list);
+	spin_unlock(&destroy_lock);
+	queue_delayed_work(system_unbound_wq, &reaper_work,
+				FSNOTIFY_REAPER_DELAY);
+
 	return ret;
 }
 
@@ -491,3 +498,21 @@ void fsnotify_init_mark(struct fsnotify_mark *mark,
 	atomic_set(&mark->refcnt, 1);
 	mark->free_mark = free_mark;
 }
+
+static void fsnotify_mark_destroy(struct work_struct *work)
+{
+	struct fsnotify_mark *mark, *next;
+	struct list_head private_destroy_list;
+
+	spin_lock(&destroy_lock);
+	/* exchange the list head */
+	list_replace_init(&destroy_list, &private_destroy_list);
+	spin_unlock(&destroy_lock);
+
+	synchronize_srcu(&fsnotify_mark_srcu);
+
+	list_for_each_entry_safe(mark, next, &private_destroy_list, g_list) {
+		list_del_init(&mark->g_list);
+		fsnotify_put_mark(mark);
+	}
+}
diff --git a/fs/ocfs2/aops.c b/fs/ocfs2/aops.c
index 794fd1587f34..cda0361e95a4 100644
--- a/fs/ocfs2/aops.c
+++ b/fs/ocfs2/aops.c
@@ -956,6 +956,7 @@ clean_orphan:
 		tmp_ret = ocfs2_del_inode_from_orphan(osb, inode, di_bh,
 				update_isize, end);
 		if (tmp_ret < 0) {
+			ocfs2_inode_unlock(inode, 1);
 			ret = tmp_ret;
 			mlog_errno(ret);
 			brelse(di_bh);
diff --git a/fs/ocfs2/mmap.c b/fs/ocfs2/mmap.c
index 9581d190f6e1..77ebc2bc1cca 100644
--- a/fs/ocfs2/mmap.c
+++ b/fs/ocfs2/mmap.c
@@ -147,6 +147,10 @@ static int ocfs2_page_mkwrite(struct vm_area_struct *vma, struct vm_fault *vmf)
 	ret = ocfs2_inode_lock(inode, &di_bh, 1);
 	if (ret < 0) {
 		mlog_errno(ret);
+		if (ret == -ENOMEM)
+			ret = VM_FAULT_OOM;
+		else
+			ret = VM_FAULT_SIGBUS;
 		goto out;
 	}
 
diff --git a/fs/overlayfs/dir.c b/fs/overlayfs/dir.c
index ed95272d57a6..52f6de5d40a9 100644
--- a/fs/overlayfs/dir.c
+++ b/fs/overlayfs/dir.c
@@ -618,7 +618,8 @@ static int ovl_remove_upper(struct dentry *dentry, bool is_dir)
 	 * sole user of this dentry.  Too tricky...  Just unhash for
 	 * now.
 	 */
-	d_drop(dentry);
+	if (!err)
+		d_drop(dentry);
 	inode_unlock(dir);
 
 	return err;
@@ -903,6 +904,13 @@ static int ovl_rename2(struct inode *olddir, struct dentry *old,
 	if (!overwrite && new_is_dir && !old_opaque && new_opaque)
 		ovl_remove_opaque(newdentry);
 
+	/*
+	 * Old dentry now lives in different location. Dentries in
+	 * lowerstack are stale. We cannot drop them here because
+	 * access to them is lockless. This could be only pure upper
+	 * or opaque directory - numlower is zero. Or upper non-dir
+	 * entry - its pureness is tracked by flag opaque.
+	 */
 	if (old_opaque != new_opaque) {
 		ovl_dentry_set_opaque(old, new_opaque);
 		if (!overwrite)
diff --git a/fs/overlayfs/inode.c b/fs/overlayfs/inode.c
index 49e204560655..a4ff5d0d7db9 100644
--- a/fs/overlayfs/inode.c
+++ b/fs/overlayfs/inode.c
@@ -65,6 +65,8 @@ int ovl_setattr(struct dentry *dentry, struct iattr *attr)
 
 		inode_lock(upperdentry->d_inode);
 		err = notify_change(upperdentry, attr, NULL);
+		if (!err)
+			ovl_copyattr(upperdentry->d_inode, dentry->d_inode);
 		inode_unlock(upperdentry->d_inode);
 	}
 	ovl_drop_write(dentry);
diff --git a/fs/overlayfs/super.c b/fs/overlayfs/super.c
index 8d826bd56b26..619ad4b016d2 100644
--- a/fs/overlayfs/super.c
+++ b/fs/overlayfs/super.c
@@ -76,12 +76,14 @@ enum ovl_path_type ovl_path_type(struct dentry *dentry)
 	if (oe->__upperdentry) {
 		type = __OVL_PATH_UPPER;
 
-		if (oe->numlower) {
-			if (S_ISDIR(dentry->d_inode->i_mode))
-				type |= __OVL_PATH_MERGE;
-		} else if (!oe->opaque) {
+		/*
+		 * Non-dir dentry can hold lower dentry from previous
+		 * location. Its purity depends only on opaque flag.
+		 */
+		if (oe->numlower && S_ISDIR(dentry->d_inode->i_mode))
+			type |= __OVL_PATH_MERGE;
+		else if (!oe->opaque)
 			type |= __OVL_PATH_PURE;
-		}
 	} else {
 		if (oe->numlower > 1)
 			type |= __OVL_PATH_MERGE;
@@ -341,6 +343,7 @@ static const struct dentry_operations ovl_dentry_operations = {
 
 static const struct dentry_operations ovl_reval_dentry_operations = {
 	.d_release = ovl_dentry_release,
+	.d_select_inode = ovl_d_select_inode,
 	.d_revalidate = ovl_dentry_revalidate,
 	.d_weak_revalidate = ovl_dentry_weak_revalidate,
 };
diff --git a/fs/pnode.c b/fs/pnode.c
index 6367e1e435c6..c524fdddc7fb 100644
--- a/fs/pnode.c
+++ b/fs/pnode.c
@@ -202,6 +202,11 @@ static struct mount *last_dest, *last_source, *dest_master;
 static struct mountpoint *mp;
 static struct hlist_head *list;
 
+static inline bool peers(struct mount *m1, struct mount *m2)
+{
+	return m1->mnt_group_id == m2->mnt_group_id && m1->mnt_group_id;
+}
+
 static int propagate_one(struct mount *m)
 {
 	struct mount *child;
@@ -212,7 +217,7 @@ static int propagate_one(struct mount *m)
 	/* skip if mountpoint isn't covered by it */
 	if (!is_subdir(mp->m_dentry, m->mnt.mnt_root))
 		return 0;
-	if (m->mnt_group_id == last_dest->mnt_group_id) {
+	if (peers(m, last_dest)) {
 		type = CL_MAKE_SHARED;
 	} else {
 		struct mount *n, *p;
@@ -223,7 +228,7 @@ static int propagate_one(struct mount *m)
 					last_source = last_source->mnt_master;
 					last_dest = last_source->mnt_parent;
 				}
-				if (n->mnt_group_id != last_dest->mnt_group_id) {
+				if (!peers(n, last_dest)) {
 					last_source = last_source->mnt_master;
 					last_dest = last_source->mnt_parent;
 				}
diff --git a/fs/read_write.c b/fs/read_write.c
index 324ec271cc4e..dadf24e5c95b 100644
--- a/fs/read_write.c
+++ b/fs/read_write.c
@@ -17,6 +17,7 @@
 #include <linux/splice.h>
 #include <linux/compat.h>
 #include <linux/mount.h>
+#include <linux/fs.h>
 #include "internal.h"
 
 #include <asm/uaccess.h>
@@ -183,7 +184,7 @@ loff_t no_seek_end_llseek(struct file *file, loff_t offset, int whence)
 	switch (whence) {
 	case SEEK_SET: case SEEK_CUR:
 		return generic_file_llseek_size(file, offset, whence,
-						~0ULL, 0);
+						OFFSET_MAX, 0);
 	default:
 		return -EINVAL;
 	}
@@ -1532,10 +1533,12 @@ int vfs_clone_file_range(struct file *file_in, loff_t pos_in,
 
 	if (!(file_in->f_mode & FMODE_READ) ||
 	    !(file_out->f_mode & FMODE_WRITE) ||
-	    (file_out->f_flags & O_APPEND) ||
-	    !file_in->f_op->clone_file_range)
+	    (file_out->f_flags & O_APPEND))
 		return -EBADF;
 
+	if (!file_in->f_op->clone_file_range)
+		return -EOPNOTSUPP;
+
 	ret = clone_verify_area(file_in, pos_in, len, false);
 	if (ret)
 		return ret;
diff --git a/fs/super.c b/fs/super.c
index 1182af8fd5ff..74914b1bae70 100644
--- a/fs/super.c
+++ b/fs/super.c
@@ -415,6 +415,7 @@ void generic_shutdown_super(struct super_block *sb)
 		sb->s_flags &= ~MS_ACTIVE;
 
 		fsnotify_unmount_inodes(sb);
+		cgroup_writeback_umount();
 
 		evict_inodes(sb);
 
diff --git a/fs/userfaultfd.c b/fs/userfaultfd.c
index 50311703135b..66cdb44616d5 100644
--- a/fs/userfaultfd.c
+++ b/fs/userfaultfd.c
@@ -287,6 +287,12 @@ int handle_userfault(struct vm_area_struct *vma, unsigned long address,
 		goto out;
 
 	/*
+	 * We don't do userfault handling for the final child pid update.
+	 */
+	if (current->flags & PF_EXITING)
+		goto out;
+
+	/*
 	 * Check that we can return VM_FAULT_RETRY.
 	 *
 	 * NOTE: it should become possible to return VM_FAULT_RETRY
diff --git a/fs/xattr.c b/fs/xattr.c
index 07d0e47f6a7f..4861322e28e8 100644
--- a/fs/xattr.c
+++ b/fs/xattr.c
@@ -940,7 +940,7 @@ ssize_t simple_xattr_list(struct inode *inode, struct simple_xattrs *xattrs,
 	bool trusted = capable(CAP_SYS_ADMIN);
 	struct simple_xattr *xattr;
 	ssize_t remaining_size = size;
-	int err;
+	int err = 0;
 
 #ifdef CONFIG_FS_POSIX_ACL
 	if (inode->i_acl) {
@@ -965,11 +965,11 @@ ssize_t simple_xattr_list(struct inode *inode, struct simple_xattrs *xattrs,
 
 		err = xattr_list_one(&buffer, &remaining_size, xattr->name);
 		if (err)
-			return err;
+			break;
 	}
 	spin_unlock(&xattrs->lock);
 
-	return size - remaining_size;
+	return err ? err : size - remaining_size;
 }
 
 /*
diff --git a/fs/xfs/xfs_aops.c b/fs/xfs/xfs_aops.c
index 379c089fb051..a9ebabfe7587 100644
--- a/fs/xfs/xfs_aops.c
+++ b/fs/xfs/xfs_aops.c
@@ -55,7 +55,7 @@ xfs_count_page_state(
 	} while ((bh = bh->b_this_page) != head);
 }
 
-STATIC struct block_device *
+struct block_device *
 xfs_find_bdev_for_inode(
 	struct inode		*inode)
 {
@@ -1208,6 +1208,10 @@ xfs_vm_writepages(
 	struct writeback_control *wbc)
 {
 	xfs_iflags_clear(XFS_I(mapping->host), XFS_ITRUNCATED);
+	if (dax_mapping(mapping))
+		return dax_writeback_mapping_range(mapping,
+				xfs_find_bdev_for_inode(mapping->host), wbc);
+
 	return generic_writepages(mapping, wbc);
 }
 
diff --git a/fs/xfs/xfs_aops.h b/fs/xfs/xfs_aops.h
index f6ffc9ae5ceb..a4343c63fb38 100644
--- a/fs/xfs/xfs_aops.h
+++ b/fs/xfs/xfs_aops.h
@@ -62,5 +62,6 @@ int	xfs_get_blocks_dax_fault(struct inode *inode, sector_t offset,
 			         struct buffer_head *map_bh, int create);
 
 extern void xfs_count_page_state(struct page *, int *, int *);
+extern struct block_device *xfs_find_bdev_for_inode(struct inode *);
 
 #endif /* __XFS_AOPS_H__ */
diff --git a/fs/xfs/xfs_bmap_util.c b/fs/xfs/xfs_bmap_util.c
index 45ec9e40150c..6c876012b2e5 100644
--- a/fs/xfs/xfs_bmap_util.c
+++ b/fs/xfs/xfs_bmap_util.c
@@ -75,7 +75,8 @@ xfs_zero_extent(
 	ssize_t		size = XFS_FSB_TO_B(mp, count_fsb);
 
 	if (IS_DAX(VFS_I(ip)))
-		return dax_clear_blocks(VFS_I(ip), block, size);
+		return dax_clear_sectors(xfs_find_bdev_for_inode(VFS_I(ip)),
+				sector, size);
 
 	/*
 	 * let the block layer decide on the fastest method of
diff --git a/fs/xfs/xfs_log_recover.c b/fs/xfs/xfs_log_recover.c
index da37beb76f6e..be5568839442 100644
--- a/fs/xfs/xfs_log_recover.c
+++ b/fs/xfs/xfs_log_recover.c
@@ -1109,27 +1109,10 @@ xlog_verify_head(
 	bool			tmp_wrapped;
 
 	/*
-	 * Search backwards through the log looking for the log record header
-	 * block. This wraps all the way back around to the head so something is
-	 * seriously wrong if we can't find it.
-	 */
-	found = xlog_rseek_logrec_hdr(log, *head_blk, *head_blk, 1, bp, rhead_blk,
-				      rhead, wrapped);
-	if (found < 0)
-		return found;
-	if (!found) {
-		xfs_warn(log->l_mp, "%s: couldn't find sync record", __func__);
-		return -EIO;
-	}
-
-	*tail_blk = BLOCK_LSN(be64_to_cpu((*rhead)->h_tail_lsn));
-
-	/*
-	 * Now that we have a tail block, check the head of the log for torn
-	 * writes. Search again until we hit the tail or the maximum number of
-	 * log record I/Os that could have been in flight at one time. Use a
-	 * temporary buffer so we don't trash the rhead/bp pointer from the
-	 * call above.
+	 * Check the head of the log for torn writes. Search backwards from the
+	 * head until we hit the tail or the maximum number of log record I/Os
+	 * that could have been in flight at one time. Use a temporary buffer so
+	 * we don't trash the rhead/bp pointers from the caller.
 	 */
 	tmp_bp = xlog_get_bp(log, 1);
 	if (!tmp_bp)
@@ -1216,6 +1199,115 @@ xlog_verify_head(
 }
 
 /*
+ * Check whether the head of the log points to an unmount record. In other
+ * words, determine whether the log is clean. If so, update the in-core state
+ * appropriately.
+ */
+static int
+xlog_check_unmount_rec(
+	struct xlog		*log,
+	xfs_daddr_t		*head_blk,
+	xfs_daddr_t		*tail_blk,
+	struct xlog_rec_header	*rhead,
+	xfs_daddr_t		rhead_blk,
+	struct xfs_buf		*bp,
+	bool			*clean)
+{
+	struct xlog_op_header	*op_head;
+	xfs_daddr_t		umount_data_blk;
+	xfs_daddr_t		after_umount_blk;
+	int			hblks;
+	int			error;
+	char			*offset;
+
+	*clean = false;
+
+	/*
+	 * Look for unmount record. If we find it, then we know there was a
+	 * clean unmount. Since 'i' could be the last block in the physical
+	 * log, we convert to a log block before comparing to the head_blk.
+	 *
+	 * Save the current tail lsn to use to pass to xlog_clear_stale_blocks()
+	 * below. We won't want to clear the unmount record if there is one, so
+	 * we pass the lsn of the unmount record rather than the block after it.
+	 */
+	if (xfs_sb_version_haslogv2(&log->l_mp->m_sb)) {
+		int	h_size = be32_to_cpu(rhead->h_size);
+		int	h_version = be32_to_cpu(rhead->h_version);
+
+		if ((h_version & XLOG_VERSION_2) &&
+		    (h_size > XLOG_HEADER_CYCLE_SIZE)) {
+			hblks = h_size / XLOG_HEADER_CYCLE_SIZE;
+			if (h_size % XLOG_HEADER_CYCLE_SIZE)
+				hblks++;
+		} else {
+			hblks = 1;
+		}
+	} else {
+		hblks = 1;
+	}
+	after_umount_blk = rhead_blk + hblks + BTOBB(be32_to_cpu(rhead->h_len));
+	after_umount_blk = do_mod(after_umount_blk, log->l_logBBsize);
+	if (*head_blk == after_umount_blk &&
+	    be32_to_cpu(rhead->h_num_logops) == 1) {
+		umount_data_blk = rhead_blk + hblks;
+		umount_data_blk = do_mod(umount_data_blk, log->l_logBBsize);
+		error = xlog_bread(log, umount_data_blk, 1, bp, &offset);
+		if (error)
+			return error;
+
+		op_head = (struct xlog_op_header *)offset;
+		if (op_head->oh_flags & XLOG_UNMOUNT_TRANS) {
+			/*
+			 * Set tail and last sync so that newly written log
+			 * records will point recovery to after the current
+			 * unmount record.
+			 */
+			xlog_assign_atomic_lsn(&log->l_tail_lsn,
+					log->l_curr_cycle, after_umount_blk);
+			xlog_assign_atomic_lsn(&log->l_last_sync_lsn,
+					log->l_curr_cycle, after_umount_blk);
+			*tail_blk = after_umount_blk;
+
+			*clean = true;
+		}
+	}
+
+	return 0;
+}
+
+static void
+xlog_set_state(
+	struct xlog		*log,
+	xfs_daddr_t		head_blk,
+	struct xlog_rec_header	*rhead,
+	xfs_daddr_t		rhead_blk,
+	bool			bump_cycle)
+{
+	/*
+	 * Reset log values according to the state of the log when we
+	 * crashed.  In the case where head_blk == 0, we bump curr_cycle
+	 * one because the next write starts a new cycle rather than
+	 * continuing the cycle of the last good log record.  At this
+	 * point we have guaranteed that all partial log records have been
+	 * accounted for.  Therefore, we know that the last good log record
+	 * written was complete and ended exactly on the end boundary
+	 * of the physical log.
+	 */
+	log->l_prev_block = rhead_blk;
+	log->l_curr_block = (int)head_blk;
+	log->l_curr_cycle = be32_to_cpu(rhead->h_cycle);
+	if (bump_cycle)
+		log->l_curr_cycle++;
+	atomic64_set(&log->l_tail_lsn, be64_to_cpu(rhead->h_tail_lsn));
+	atomic64_set(&log->l_last_sync_lsn, be64_to_cpu(rhead->h_lsn));
+	xlog_assign_grant_head(&log->l_reserve_head.grant, log->l_curr_cycle,
+					BBTOB(log->l_curr_block));
+	xlog_assign_grant_head(&log->l_write_head.grant, log->l_curr_cycle,
+					BBTOB(log->l_curr_block));
+}
+
+/*
  * Find the sync block number or the tail of the log.
  *
  * This will be the block number of the last record to have its
@@ -1238,22 +1330,20 @@ xlog_find_tail(
 	xfs_daddr_t		*tail_blk)
 {
 	xlog_rec_header_t	*rhead;
-	xlog_op_header_t	*op_head;
 	char			*offset = NULL;
 	xfs_buf_t		*bp;
 	int			error;
-	xfs_daddr_t		umount_data_blk;
-	xfs_daddr_t		after_umount_blk;
 	xfs_daddr_t		rhead_blk;
 	xfs_lsn_t		tail_lsn;
-	int			hblks;
 	bool			wrapped = false;
+	bool			clean = false;
 
 	/*
 	 * Find previous log record
 	 */
 	if ((error = xlog_find_head(log, head_blk)))
 		return error;
+	ASSERT(*head_blk < INT_MAX);
 
 	bp = xlog_get_bp(log, 1);
 	if (!bp)
@@ -1271,100 +1361,75 @@ xlog_find_tail(
 	}
 
 	/*
-	 * Trim the head block back to skip over torn records. We can have
-	 * multiple log I/Os in flight at any time, so we assume CRC failures
-	 * back through the previous several records are torn writes and skip
-	 * them.
+	 * Search backwards through the log looking for the log record header
+	 * block. This wraps all the way back around to the head so something is
+	 * seriously wrong if we can't find it.
 	 */
-	ASSERT(*head_blk < INT_MAX);
-	error = xlog_verify_head(log, head_blk, tail_blk, bp, &rhead_blk,
-				 &rhead, &wrapped);
-	if (error)
-		goto done;
+	error = xlog_rseek_logrec_hdr(log, *head_blk, *head_blk, 1, bp,
+				      &rhead_blk, &rhead, &wrapped);
+	if (error < 0)
+		return error;
+	if (!error) {
+		xfs_warn(log->l_mp, "%s: couldn't find sync record", __func__);
+		return -EIO;
+	}
+	*tail_blk = BLOCK_LSN(be64_to_cpu(rhead->h_tail_lsn));
 
 	/*
-	 * Reset log values according to the state of the log when we
-	 * crashed.  In the case where head_blk == 0, we bump curr_cycle
-	 * one because the next write starts a new cycle rather than
-	 * continuing the cycle of the last good log record.  At this
-	 * point we have guaranteed that all partial log records have been
-	 * accounted for.  Therefore, we know that the last good log record
-	 * written was complete and ended exactly on the end boundary
-	 * of the physical log.
+	 * Set the log state based on the current head record.
 	 */
-	log->l_prev_block = rhead_blk;
-	log->l_curr_block = (int)*head_blk;
-	log->l_curr_cycle = be32_to_cpu(rhead->h_cycle);
-	if (wrapped)
-		log->l_curr_cycle++;
-	atomic64_set(&log->l_tail_lsn, be64_to_cpu(rhead->h_tail_lsn));
-	atomic64_set(&log->l_last_sync_lsn, be64_to_cpu(rhead->h_lsn));
-	xlog_assign_grant_head(&log->l_reserve_head.grant, log->l_curr_cycle,
-					BBTOB(log->l_curr_block));
-	xlog_assign_grant_head(&log->l_write_head.grant, log->l_curr_cycle,
-					BBTOB(log->l_curr_block));
+	xlog_set_state(log, *head_blk, rhead, rhead_blk, wrapped);
+	tail_lsn = atomic64_read(&log->l_tail_lsn);
 
 	/*
-	 * Look for unmount record.  If we find it, then we know there
-	 * was a clean unmount.  Since 'i' could be the last block in
-	 * the physical log, we convert to a log block before comparing
-	 * to the head_blk.
+	 * Look for an unmount record at the head of the log. This sets the log
+	 * state to determine whether recovery is necessary.
+	 */
+	error = xlog_check_unmount_rec(log, head_blk, tail_blk, rhead,
+				       rhead_blk, bp, &clean);
+	if (error)
+		goto done;
+
+	/*
+	 * Verify the log head if the log is not clean (e.g., we have anything
+	 * but an unmount record at the head). This uses CRC verification to
+	 * detect and trim torn writes. If discovered, CRC failures are
+	 * considered torn writes and the log head is trimmed accordingly.
 	 *
-	 * Save the current tail lsn to use to pass to
-	 * xlog_clear_stale_blocks() below.  We won't want to clear the
-	 * unmount record if there is one, so we pass the lsn of the
-	 * unmount record rather than the block after it.
+	 * Note that we can only run CRC verification when the log is dirty
+	 * because there's no guarantee that the log data behind an unmount
+	 * record is compatible with the current architecture.
 	 */
-	if (xfs_sb_version_haslogv2(&log->l_mp->m_sb)) {
-		int	h_size = be32_to_cpu(rhead->h_size);
-		int	h_version = be32_to_cpu(rhead->h_version);
+	if (!clean) {
+		xfs_daddr_t	orig_head = *head_blk;
 
-		if ((h_version & XLOG_VERSION_2) &&
-		    (h_size > XLOG_HEADER_CYCLE_SIZE)) {
-			hblks = h_size / XLOG_HEADER_CYCLE_SIZE;
-			if (h_size % XLOG_HEADER_CYCLE_SIZE)
-				hblks++;
-		} else {
-			hblks = 1;
-		}
-	} else {
-		hblks = 1;
-	}
-	after_umount_blk = rhead_blk + hblks + BTOBB(be32_to_cpu(rhead->h_len));
-	after_umount_blk = do_mod(after_umount_blk, log->l_logBBsize);
-	tail_lsn = atomic64_read(&log->l_tail_lsn);
-	if (*head_blk == after_umount_blk &&
-	    be32_to_cpu(rhead->h_num_logops) == 1) {
-		umount_data_blk = rhead_blk + hblks;
-		umount_data_blk = do_mod(umount_data_blk, log->l_logBBsize);
-		error = xlog_bread(log, umount_data_blk, 1, bp, &offset);
+		error = xlog_verify_head(log, head_blk, tail_blk, bp,
+					 &rhead_blk, &rhead, &wrapped);
 		if (error)
 			goto done;
 
-		op_head = (xlog_op_header_t *)offset;
-		if (op_head->oh_flags & XLOG_UNMOUNT_TRANS) {
-			/*
-			 * Set tail and last sync so that newly written
-			 * log records will point recovery to after the
-			 * current unmount record.
-			 */
-			xlog_assign_atomic_lsn(&log->l_tail_lsn,
-					log->l_curr_cycle, after_umount_blk);
-			xlog_assign_atomic_lsn(&log->l_last_sync_lsn,
-					log->l_curr_cycle, after_umount_blk);
-			*tail_blk = after_umount_blk;
-
-			/*
-			 * Note that the unmount was clean. If the unmount
-			 * was not clean, we need to know this to rebuild the
-			 * superblock counters from the perag headers if we
-			 * have a filesystem using non-persistent counters.
-			 */
-			log->l_mp->m_flags |= XFS_MOUNT_WAS_CLEAN;
+		/* update in-core state again if the head changed */
+		if (*head_blk != orig_head) {
+			xlog_set_state(log, *head_blk, rhead, rhead_blk,
+				       wrapped);
+			tail_lsn = atomic64_read(&log->l_tail_lsn);
+			error = xlog_check_unmount_rec(log, head_blk, tail_blk,
+						       rhead, rhead_blk, bp,
+						       &clean);
+			if (error)
+				goto done;
 		}
 	}
 
 	/*
+	 * Note that the unmount was clean. If the unmount was not clean, we
+	 * need to know this to rebuild the superblock counters from the perag
+	 * headers if we have a filesystem using non-persistent counters.
+	 */
+	if (clean)
+		log->l_mp->m_flags |= XFS_MOUNT_WAS_CLEAN;
+
+	/*
 	 * Make sure that there are no blocks in front of the head
 	 * with the same cycle number as the head.  This can happen
 	 * because we allow multiple outstanding log writes concurrently,
@@ -4491,7 +4556,7 @@ xlog_recover_process(
 	 * know precisely what failed.
 	 */
 	if (pass == XLOG_RECOVER_CRCPASS) {
-		if (rhead->h_crc && crc != le32_to_cpu(rhead->h_crc))
+		if (rhead->h_crc && crc != rhead->h_crc)
 			return -EFSBADCRC;
 		return 0;
 	}
@@ -4502,7 +4567,7 @@ xlog_recover_process(
 	 * zero CRC check prevents warnings from being emitted when upgrading
 	 * the kernel from one that does not add CRCs by default.
 	 */
-	if (crc != le32_to_cpu(rhead->h_crc)) {
+	if (crc != rhead->h_crc) {
 		if (rhead->h_crc || xfs_sb_version_hascrc(&log->l_mp->m_sb)) {
 			xfs_alert(log->l_mp,
 		"log record CRC mismatch: found 0x%x, expected 0x%x.",
diff --git a/include/asm-generic/cputime_nsecs.h b/include/asm-generic/cputime_nsecs.h
index 0419485891f2..0f1c6f315cdc 100644
--- a/include/asm-generic/cputime_nsecs.h
+++ b/include/asm-generic/cputime_nsecs.h
@@ -75,7 +75,7 @@ typedef u64 __nocast cputime64_t;
  */
 static inline cputime_t timespec_to_cputime(const struct timespec *val)
 {
-	u64 ret = val->tv_sec * NSEC_PER_SEC + val->tv_nsec;
+	u64 ret = (u64)val->tv_sec * NSEC_PER_SEC + val->tv_nsec;
 	return (__force cputime_t) ret;
 }
 static inline void cputime_to_timespec(const cputime_t ct, struct timespec *val)
@@ -91,7 +91,8 @@ static inline void cputime_to_timespec(const cputime_t ct, struct timespec *val)
  */
 static inline cputime_t timeval_to_cputime(const struct timeval *val)
 {
-	u64 ret = val->tv_sec * NSEC_PER_SEC + val->tv_usec * NSEC_PER_USEC;
+	u64 ret = (u64)val->tv_sec * NSEC_PER_SEC +
+			val->tv_usec * NSEC_PER_USEC;
 	return (__force cputime_t) ret;
 }
 static inline void cputime_to_timeval(const cputime_t ct, struct timeval *val)
diff --git a/include/asm-generic/pgtable.h b/include/asm-generic/pgtable.h
index 0b3c0d39ef75..c370b261c720 100644
--- a/include/asm-generic/pgtable.h
+++ b/include/asm-generic/pgtable.h
@@ -239,6 +239,14 @@ extern void pmdp_invalidate(struct vm_area_struct *vma, unsigned long address,
 			    pmd_t *pmdp);
 #endif
 
+#ifndef __HAVE_ARCH_PMDP_HUGE_SPLIT_PREPARE
+static inline void pmdp_huge_split_prepare(struct vm_area_struct *vma,
+					   unsigned long address, pmd_t *pmdp)
+{
+
+}
+#endif
+
 #ifndef __HAVE_ARCH_PTE_SAME
 static inline int pte_same(pte_t pte_a, pte_t pte_b)
 {
diff --git a/include/drm/drm_crtc.h b/include/drm/drm_crtc.h
index b10eba23a744..e0170bf80bb0 100644
--- a/include/drm/drm_crtc.h
+++ b/include/drm/drm_crtc.h
@@ -1181,6 +1181,7 @@ struct drm_connector {
 	struct drm_mode_object base;
 
 	char *name;
+	int connector_id;
 	int connector_type;
 	int connector_type_id;
 	bool interlace_allowed;
@@ -1596,6 +1597,8 @@ struct drm_bridge_funcs {
 	 *
 	 * The bridge can assume that the display pipe (i.e. clocks and timing
 	 * signals) feeding it is still running when this callback is called.
+	 *
+	 * The disable callback is optional.
 	 */
 	void (*disable)(struct drm_bridge *bridge);
 
@@ -1612,6 +1615,8 @@ struct drm_bridge_funcs {
 	 * The bridge must assume that the display pipe (i.e. clocks and timing
 	 * singals) feeding it is no longer running when this callback is
 	 * called.
+	 *
+	 * The post_disable callback is optional.
 	 */
 	void (*post_disable)(struct drm_bridge *bridge);
 
@@ -1640,6 +1645,8 @@ struct drm_bridge_funcs {
 	 * will not yet be running when this callback is called. The bridge must
 	 * not enable the display link feeding the next bridge in the chain (if
 	 * there is one) when this callback is called.
+	 *
+	 * The pre_enable callback is optional.
 	 */
 	void (*pre_enable)(struct drm_bridge *bridge);
 
@@ -1657,6 +1664,8 @@ struct drm_bridge_funcs {
 	 * signals) feeding it is running when this callback is called. This
 	 * callback must enable the display link feeding the next bridge in the
 	 * chain if there is one.
+	 *
+	 * The enable callback is optional.
 	 */
 	void (*enable)(struct drm_bridge *bridge);
 };
@@ -2073,6 +2082,7 @@ struct drm_mode_config {
 	struct list_head fb_list;
 
 	int num_connector;
+	struct ida connector_ida;
 	struct list_head connector_list;
 	int num_encoder;
 	struct list_head encoder_list;
@@ -2244,7 +2254,11 @@ int drm_connector_register(struct drm_connector *connector);
 void drm_connector_unregister(struct drm_connector *connector);
 
 extern void drm_connector_cleanup(struct drm_connector *connector);
-extern unsigned int drm_connector_index(struct drm_connector *connector);
+static inline unsigned drm_connector_index(struct drm_connector *connector)
+{
+	return connector->connector_id;
+}
+
 /* helper to unplug all connectors from sysfs for device */
 extern void drm_connector_unplug_all(struct drm_device *dev);
 
diff --git a/include/drm/drm_mipi_dsi.h b/include/drm/drm_mipi_dsi.h
index 1b3b1f8c8cdf..7a9840f8b38e 100644
--- a/include/drm/drm_mipi_dsi.h
+++ b/include/drm/drm_mipi_dsi.h
@@ -96,14 +96,17 @@ struct mipi_dsi_host_ops {
  * struct mipi_dsi_host - DSI host device
  * @dev: driver model device node for this DSI host
  * @ops: DSI host operations
+ * @list: list management
  */
 struct mipi_dsi_host {
 	struct device *dev;
 	const struct mipi_dsi_host_ops *ops;
+	struct list_head list;
 };
 
 int mipi_dsi_host_register(struct mipi_dsi_host *host);
 void mipi_dsi_host_unregister(struct mipi_dsi_host *host);
+struct mipi_dsi_host *of_find_mipi_dsi_host_by_node(struct device_node *node);
 
 /* DSI mode flags */
 
@@ -139,10 +142,28 @@ enum mipi_dsi_pixel_format {
 	MIPI_DSI_FMT_RGB565,
 };
 
+#define DSI_DEV_NAME_SIZE		20
+
+/**
+ * struct mipi_dsi_device_info - template for creating a mipi_dsi_device
+ * @type: DSI peripheral chip type
+ * @channel: DSI virtual channel assigned to peripheral
+ * @node: pointer to OF device node or NULL
+ *
+ * This is populated and passed to mipi_dsi_device_new to create a new
+ * DSI device
+ */
+struct mipi_dsi_device_info {
+	char type[DSI_DEV_NAME_SIZE];
+	u32 channel;
+	struct device_node *node;
+};
+
 /**
  * struct mipi_dsi_device - DSI peripheral device
  * @host: DSI host for this peripheral
  * @dev: driver model device node for this peripheral
+ * @name: DSI peripheral chip type
  * @channel: virtual channel assigned to the peripheral
  * @format: pixel format for video mode
  * @lanes: number of active data lanes
@@ -152,6 +173,7 @@ struct mipi_dsi_device {
 	struct mipi_dsi_host *host;
 	struct device dev;
 
+	char name[DSI_DEV_NAME_SIZE];
 	unsigned int channel;
 	unsigned int lanes;
 	enum mipi_dsi_pixel_format format;
@@ -188,6 +210,10 @@ static inline int mipi_dsi_pixel_format_to_bpp(enum mipi_dsi_pixel_format fmt)
 	return -EINVAL;
 }
 
+struct mipi_dsi_device *
+mipi_dsi_device_register_full(struct mipi_dsi_host *host,
+			      const struct mipi_dsi_device_info *info);
+void mipi_dsi_device_unregister(struct mipi_dsi_device *dsi);
 struct mipi_dsi_device *of_find_mipi_dsi_device_by_node(struct device_node *np);
 int mipi_dsi_attach(struct mipi_dsi_device *dsi);
 int mipi_dsi_detach(struct mipi_dsi_device *dsi);
diff --git a/include/drm/drm_of.h b/include/drm/drm_of.h
index 8544665ee4f4..3fd87b386ed7 100644
--- a/include/drm/drm_of.h
+++ b/include/drm/drm_of.h
@@ -1,9 +1,12 @@
 #ifndef __DRM_OF_H__
 #define __DRM_OF_H__
 
+#include <linux/of_graph.h>
+
 struct component_master_ops;
 struct device;
 struct drm_device;
+struct drm_encoder;
 struct device_node;
 
 #ifdef CONFIG_OF
@@ -12,6 +15,9 @@ extern uint32_t drm_of_find_possible_crtcs(struct drm_device *dev,
 extern int drm_of_component_probe(struct device *dev,
 				  int (*compare_of)(struct device *, void *),
 				  const struct component_master_ops *m_ops);
+extern int drm_of_encoder_active_endpoint(struct device_node *node,
+					  struct drm_encoder *encoder,
+					  struct of_endpoint *endpoint);
 #else
 static inline uint32_t drm_of_find_possible_crtcs(struct drm_device *dev,
 						  struct device_node *port)
@@ -26,6 +32,33 @@ drm_of_component_probe(struct device *dev,
 {
 	return -EINVAL;
 }
+
+static inline int drm_of_encoder_active_endpoint(struct device_node *node,
+						 struct drm_encoder *encoder,
+						 struct of_endpoint *endpoint)
+{
+	return -EINVAL;
+}
 #endif
 
+static inline int drm_of_encoder_active_endpoint_id(struct device_node *node,
+						    struct drm_encoder *encoder)
+{
+	struct of_endpoint endpoint;
+	int ret = drm_of_encoder_active_endpoint(node, encoder,
+						 &endpoint);
+
+	return ret ?: endpoint.id;
+}
+
+static inline int drm_of_encoder_active_port_id(struct device_node *node,
+						struct drm_encoder *encoder)
+{
+	struct of_endpoint endpoint;
+	int ret = drm_of_encoder_active_endpoint(node, encoder,
+						 &endpoint);
+
+	return ret ?: endpoint.port;
+}
+
 #endif /* __DRM_OF_H__ */
diff --git a/include/drm/exynos_drm.h b/include/drm/exynos_drm.h
deleted file mode 100644
index cb65fa14acfc..000000000000
--- a/include/drm/exynos_drm.h
+++ /dev/null
@@ -1,101 +0,0 @@
-/* exynos_drm.h
- *
- * Copyright (c) 2011 Samsung Electronics Co., Ltd.
- * Authors:
- *	Inki Dae <inki.dae@samsung.com>
- *	Joonyoung Shim <jy0922.shim@samsung.com>
- *	Seung-Woo Kim <sw0312.kim@samsung.com>
- *
- * This program is free software; you can redistribute  it and/or modify it
- * under  the terms of  the GNU General  Public License as published by the
- * Free Software Foundation;  either version 2 of the  License, or (at your
- * option) any later version.
- */
-#ifndef _EXYNOS_DRM_H_
-#define _EXYNOS_DRM_H_
-
-#include <uapi/drm/exynos_drm.h>
-#include <video/videomode.h>
-
-/**
- * A structure for lcd panel information.
- *
- * @timing: default video mode for initializing
- * @width_mm: physical size of lcd width.
- * @height_mm: physical size of lcd height.
- */
-struct exynos_drm_panel_info {
-	struct videomode vm;
-	u32 width_mm;
-	u32 height_mm;
-};
-
-/**
- * Platform Specific Structure for DRM based FIMD.
- *
- * @panel: default panel info for initializing
- * @default_win: default window layer number to be used for UI.
- * @bpp: default bit per pixel.
- */
-struct exynos_drm_fimd_pdata {
-	struct exynos_drm_panel_info panel;
-	u32				vidcon0;
-	u32				vidcon1;
-	unsigned int			default_win;
-	unsigned int			bpp;
-};
-
-/**
- * Platform Specific Structure for DRM based HDMI.
- *
- * @hdmi_dev: device point to specific hdmi driver.
- * @mixer_dev: device point to specific mixer driver.
- *
- * this structure is used for common hdmi driver and each device object
- * would be used to access specific device driver(hdmi or mixer driver)
- */
-struct exynos_drm_common_hdmi_pd {
-	struct device *hdmi_dev;
-	struct device *mixer_dev;
-};
-
-/**
- * Platform Specific Structure for DRM based HDMI core.
- *
- * @is_v13: set if hdmi version 13 is.
- * @cfg_hpd: function pointer to configure hdmi hotplug detection pin
- * @get_hpd: function pointer to get value of hdmi hotplug detection pin
- */
-struct exynos_drm_hdmi_pdata {
-	bool is_v13;
-	void (*cfg_hpd)(bool external);
-	int (*get_hpd)(void);
-};
-
-/**
- * Platform Specific Structure for DRM based IPP.
- *
- * @inv_pclk: if set 1. invert pixel clock
- * @inv_vsync: if set 1. invert vsync signal for wb
- * @inv_href: if set 1. invert href signal
- * @inv_hsync: if set 1. invert hsync signal for wb
- */
-struct exynos_drm_ipp_pol {
-	unsigned int inv_pclk;
-	unsigned int inv_vsync;
-	unsigned int inv_href;
-	unsigned int inv_hsync;
-};
-
-/**
- * Platform Specific Structure for DRM based FIMC.
- *
- * @pol: current hardware block polarity settings.
- * @clk_rate: current hardware clock rate.
- */
-struct exynos_drm_fimc_pdata {
-	struct exynos_drm_ipp_pol pol;
-	int clk_rate;
-};
-
-#endif	/* _EXYNOS_DRM_H_ */
diff --git a/include/drm/i915_pciids.h b/include/drm/i915_pciids.h
index f97020904717..9094599a1150 100644
--- a/include/drm/i915_pciids.h
+++ b/include/drm/i915_pciids.h
@@ -277,7 +277,9 @@
 	INTEL_VGA_DEVICE(0x191D, info)  /* WKS GT2 */
 
 #define INTEL_SKL_GT3_IDS(info) \
+	INTEL_VGA_DEVICE(0x1923, info), /* ULT GT3 */ \
 	INTEL_VGA_DEVICE(0x1926, info), /* ULT GT3 */ \
+	INTEL_VGA_DEVICE(0x1927, info), /* ULT GT3 */ \
 	INTEL_VGA_DEVICE(0x192B, info), /* Halo GT3 */ \
 	INTEL_VGA_DEVICE(0x192A, info)  /* SRV GT3 */
 
@@ -296,7 +298,9 @@
 #define INTEL_BXT_IDS(info) \
 	INTEL_VGA_DEVICE(0x0A84, info), \
 	INTEL_VGA_DEVICE(0x1A84, info), \
-	INTEL_VGA_DEVICE(0x5A84, info)
+	INTEL_VGA_DEVICE(0x1A85, info), \
+	INTEL_VGA_DEVICE(0x5A84, info), /* APL HD Graphics 505 */ \
+	INTEL_VGA_DEVICE(0x5A85, info)  /* APL HD Graphics 500 */
 
 #define INTEL_KBL_GT1_IDS(info)	\
 	INTEL_VGA_DEVICE(0x5913, info), /* ULT GT1.5 */ \
diff --git a/include/dt-bindings/clock/tegra210-car.h b/include/dt-bindings/clock/tegra210-car.h
index 6f45aea49e4f..0a05b0d36ae7 100644
--- a/include/dt-bindings/clock/tegra210-car.h
+++ b/include/dt-bindings/clock/tegra210-car.h
@@ -126,7 +126,7 @@
 /* 104 */
 /* 105 */
 #define TEGRA210_CLK_D_AUDIO 106
-/* 107 ( affects abp -> ape) */
+#define TEGRA210_CLK_APB2APE 107
 /* 108 */
 /* 109 */
 /* 110 */
diff --git a/include/linux/ata.h b/include/linux/ata.h
index d2992bfa1706..c1a2f345cbe6 100644
--- a/include/linux/ata.h
+++ b/include/linux/ata.h
@@ -487,8 +487,8 @@ enum ata_tf_protocols {
 };
 
 enum ata_ioctls {
-	ATA_IOC_GET_IO32	= 0x309,
-	ATA_IOC_SET_IO32	= 0x324,
+	ATA_IOC_GET_IO32	= 0x309, /* HDIO_GET_32BIT */
+	ATA_IOC_SET_IO32	= 0x324, /* HDIO_SET_32BIT */
 };
 
 /* core structures */
diff --git a/include/linux/bio.h b/include/linux/bio.h
index 5349e6816cbb..cb6888824108 100644
--- a/include/linux/bio.h
+++ b/include/linux/bio.h
@@ -310,6 +310,43 @@ static inline void bio_clear_flag(struct bio *bio, unsigned int bit)
 	bio->bi_flags &= ~(1U << bit);
 }
 
+static inline void bio_get_first_bvec(struct bio *bio, struct bio_vec *bv)
+{
+	*bv = bio_iovec(bio);
+}
+
+static inline void bio_get_last_bvec(struct bio *bio, struct bio_vec *bv)
+{
+	struct bvec_iter iter = bio->bi_iter;
+	int idx;
+
+	if (!bio_flagged(bio, BIO_CLONED)) {
+		*bv = bio->bi_io_vec[bio->bi_vcnt - 1];
+		return;
+	}
+
+	if (unlikely(!bio_multiple_segments(bio))) {
+		*bv = bio_iovec(bio);
+		return;
+	}
+
+	bio_advance_iter(bio, &iter, iter.bi_size);
+
+	if (!iter.bi_bvec_done)
+		idx = iter.bi_idx - 1;
+	else	/* in the middle of bvec */
+		idx = iter.bi_idx;
+
+	*bv = bio->bi_io_vec[idx];
+
+	/*
+	 * iter.bi_bvec_done records actual length of the last bvec
+	 * if this bio ends in the middle of one io vector
+	 */
+	if (iter.bi_bvec_done)
+		bv->bv_len = iter.bi_bvec_done;
+}
+
 enum bip_flags {
 	BIP_BLOCK_INTEGRITY	= 1 << 0, /* block layer owns integrity data */
 	BIP_MAPPED_INTEGRITY	= 1 << 1, /* ref tag has been remapped */
diff --git a/include/linux/blkdev.h b/include/linux/blkdev.h
index 29189aeace19..413c84fbc4ed 100644
--- a/include/linux/blkdev.h
+++ b/include/linux/blkdev.h
@@ -682,9 +682,12 @@ static inline bool blk_write_same_mergeable(struct bio *a, struct bio *b)
 /*
  * q->prep_rq_fn return values
  */
-#define BLKPREP_OK		0	/* serve it */
-#define BLKPREP_KILL		1	/* fatal error, kill */
-#define BLKPREP_DEFER		2	/* leave on queue */
+enum {
+	BLKPREP_OK,		/* serve it */
+	BLKPREP_KILL,		/* fatal error, kill, return -EIO */
+	BLKPREP_DEFER,		/* leave on queue */
+	BLKPREP_INVALID,	/* invalid command, kill, return -EREMOTEIO */
+};
 
 extern unsigned long blk_max_low_pfn, blk_max_pfn;
 
@@ -892,7 +895,7 @@ static inline unsigned int blk_rq_get_max_sectors(struct request *rq)
 {
 	struct request_queue *q = rq->q;
 
-	if (unlikely(rq->cmd_type == REQ_TYPE_BLOCK_PC))
+	if (unlikely(rq->cmd_type != REQ_TYPE_FS))
 		return q->limits.max_hw_sectors;
 
 	if (!q->limits.chunk_sectors || (rq->cmd_flags & REQ_DISCARD))
@@ -1369,6 +1372,13 @@ static inline void put_dev_sector(Sector p)
 	page_cache_release(p.v);
 }
 
+static inline bool __bvec_gap_to_prev(struct request_queue *q,
+				struct bio_vec *bprv, unsigned int offset)
+{
+	return offset ||
+		((bprv->bv_offset + bprv->bv_len) & queue_virt_boundary(q));
+}
+
 /*
  * Check if adding a bio_vec after bprv with offset would create a gap in
  * the SG list. Most drivers don't care about this, but some do.
@@ -1378,18 +1388,22 @@ static inline bool bvec_gap_to_prev(struct request_queue *q,
 {
 	if (!queue_virt_boundary(q))
 		return false;
-	return offset ||
-		((bprv->bv_offset + bprv->bv_len) & queue_virt_boundary(q));
+	return __bvec_gap_to_prev(q, bprv, offset);
 }
 
 static inline bool bio_will_gap(struct request_queue *q, struct bio *prev,
 			 struct bio *next)
 {
-	if (!bio_has_data(prev))
-		return false;
+	if (bio_has_data(prev) && queue_virt_boundary(q)) {
+		struct bio_vec pb, nb;
+
+		bio_get_last_bvec(prev, &pb);
+		bio_get_first_bvec(next, &nb);
 
-	return bvec_gap_to_prev(q, &prev->bi_io_vec[prev->bi_vcnt - 1],
-				next->bi_io_vec[0].bv_offset);
+		return __bvec_gap_to_prev(q, &pb, nb.bv_offset);
+	}
+
+	return false;
 }
 
 static inline bool req_gap_back_merge(struct request *req, struct bio *bio)
diff --git a/include/linux/ceph/ceph_features.h b/include/linux/ceph/ceph_features.h
index c1ef6f14e7be..15151f3c4120 100644
--- a/include/linux/ceph/ceph_features.h
+++ b/include/linux/ceph/ceph_features.h
@@ -75,6 +75,7 @@
 #define CEPH_FEATURE_CRUSH_TUNABLES5	(1ULL<<58) /* chooseleaf stable mode */
 // duplicated since it was introduced at the same time as CEPH_FEATURE_CRUSH_TUNABLES5
 #define CEPH_FEATURE_NEW_OSDOPREPLY_ENCODING   (1ULL<<58) /* New, v7 encoding */
+#define CEPH_FEATURE_FS_FILE_LAYOUT_V2       (1ULL<<58) /* file_layout_t */
 
 /*
  * The introduction of CEPH_FEATURE_OSD_SNAPMAPPER caused the feature
diff --git a/include/linux/cgroup-defs.h b/include/linux/cgroup-defs.h
index 7f540f7f588d..789471dba6fb 100644
--- a/include/linux/cgroup-defs.h
+++ b/include/linux/cgroup-defs.h
@@ -127,6 +127,12 @@ struct cgroup_subsys_state {
 	 */
 	u64 serial_nr;
 
+	/*
+	 * Incremented by online self and children.  Used to guarantee that
+	 * parents are not offlined before their children.
+	 */
+	atomic_t online_cnt;
+
 	/* percpu_ref killing and RCU release */
 	struct rcu_head rcu_head;
 	struct work_struct destroy_work;
diff --git a/include/linux/compiler.h b/include/linux/compiler.h
index 00b042c49ccd..48f5aab117ae 100644
--- a/include/linux/compiler.h
+++ b/include/linux/compiler.h
@@ -144,7 +144,7 @@ void ftrace_likely_update(struct ftrace_branch_data *f, int val, int expect);
  */
 #define if(cond, ...) __trace_if( (cond , ## __VA_ARGS__) )
 #define __trace_if(cond) \
-	if (__builtin_constant_p((cond)) ? !!(cond) :			\
+	if (__builtin_constant_p(!!(cond)) ? !!(cond) :			\
 	({								\
 		int ______r;						\
 		static struct ftrace_branch_data			\
diff --git a/include/linux/cpuset.h b/include/linux/cpuset.h
index 85a868ccb493..fea160ee5803 100644
--- a/include/linux/cpuset.h
+++ b/include/linux/cpuset.h
@@ -137,6 +137,8 @@ static inline void set_mems_allowed(nodemask_t nodemask)
 	task_unlock(current);
 }
 
+extern void cpuset_post_attach_flush(void);
+
 #else /* !CONFIG_CPUSETS */
 
 static inline bool cpusets_enabled(void) { return false; }
@@ -243,6 +245,10 @@ static inline bool read_mems_allowed_retry(unsigned int seq)
 	return false;
 }
 
+static inline void cpuset_post_attach_flush(void)
+{
+}
+
 #endif /* !CONFIG_CPUSETS */
 
 #endif /* _LINUX_CPUSET_H */
diff --git a/include/linux/dax.h b/include/linux/dax.h
index 818e45078929..636dd59ab505 100644
--- a/include/linux/dax.h
+++ b/include/linux/dax.h
@@ -7,7 +7,7 @@
 
 ssize_t dax_do_io(struct kiocb *, struct inode *, struct iov_iter *, loff_t,
 		  get_block_t, dio_iodone_t, int flags);
-int dax_clear_blocks(struct inode *, sector_t block, long size);
+int dax_clear_sectors(struct block_device *bdev, sector_t _sector, long _size);
 int dax_zero_page_range(struct inode *, loff_t from, unsigned len, get_block_t);
 int dax_truncate_page(struct inode *, loff_t from, get_block_t);
 int dax_fault(struct vm_area_struct *, struct vm_fault *, get_block_t,
@@ -52,6 +52,8 @@ static inline bool dax_mapping(struct address_space *mapping)
 {
 	return mapping->host && IS_DAX(mapping->host);
 }
-int dax_writeback_mapping_range(struct address_space *mapping, loff_t start,
-		loff_t end);
+
+struct writeback_control;
+int dax_writeback_mapping_range(struct address_space *mapping,
+		struct block_device *bdev, struct writeback_control *wbc);
 #endif
diff --git a/include/linux/dcache.h b/include/linux/dcache.h
index 7781ce110503..c4b5f4b3f8f8 100644
--- a/include/linux/dcache.h
+++ b/include/linux/dcache.h
@@ -409,9 +409,7 @@ static inline bool d_mountpoint(const struct dentry *dentry)
  */
 static inline unsigned __d_entry_type(const struct dentry *dentry)
 {
-	unsigned type = READ_ONCE(dentry->d_flags);
-	smp_rmb();
-	return type & DCACHE_ENTRY_TYPE;
+	return dentry->d_flags & DCACHE_ENTRY_TYPE;
 }
 
 static inline bool d_is_miss(const struct dentry *dentry)
diff --git a/include/linux/devpts_fs.h b/include/linux/devpts_fs.h
index 251a2090a554..e0ee0b3000b2 100644
--- a/include/linux/devpts_fs.h
+++ b/include/linux/devpts_fs.h
@@ -19,6 +19,8 @@
 
 int devpts_new_index(struct inode *ptmx_inode);
 void devpts_kill_index(struct inode *ptmx_inode, int idx);
+void devpts_add_ref(struct inode *ptmx_inode);
+void devpts_del_ref(struct inode *ptmx_inode);
 /* mknod in devpts */
 struct inode *devpts_pty_new(struct inode *ptmx_inode, dev_t device, int index,
 		void *priv);
@@ -32,6 +34,8 @@ void devpts_pty_kill(struct inode *inode);
 /* Dummy stubs in the no-pty case */
 static inline int devpts_new_index(struct inode *ptmx_inode) { return -EINVAL; }
 static inline void devpts_kill_index(struct inode *ptmx_inode, int idx) { }
+static inline void devpts_add_ref(struct inode *ptmx_inode) { }
+static inline void devpts_del_ref(struct inode *ptmx_inode) { }
 static inline struct inode *devpts_pty_new(struct inode *ptmx_inode,
 		dev_t device, int index, void *priv)
 {
diff --git a/include/linux/dma-mapping.h b/include/linux/dma-mapping.h
index 75857cda38e9..728ef074602a 100644
--- a/include/linux/dma-mapping.h
+++ b/include/linux/dma-mapping.h
@@ -386,7 +386,7 @@ static inline void dma_free_attrs(struct device *dev, size_t size,
 	if (dma_release_from_coherent(dev, get_order(size), cpu_addr))
 		return;
 
-	if (!ops->free)
+	if (!ops->free || !cpu_addr)
 		return;
 
 	debug_dma_free_coherent(dev, size, cpu_addr, dma_handle);
diff --git a/include/linux/efi.h b/include/linux/efi.h
index 569b5a866bb1..47be3ad7d3e5 100644
--- a/include/linux/efi.h
+++ b/include/linux/efi.h
@@ -1199,7 +1199,10 @@ int efivar_entry_iter(int (*func)(struct efivar_entry *, void *),
 struct efivar_entry *efivar_entry_find(efi_char16_t *name, efi_guid_t guid,
 				       struct list_head *head, bool remove);
 
-bool efivar_validate(efi_char16_t *var_name, u8 *data, unsigned long len);
+bool efivar_validate(efi_guid_t vendor, efi_char16_t *var_name, u8 *data,
+		     unsigned long data_size);
+bool efivar_variable_is_removable(efi_guid_t vendor, const char *name,
+				  size_t len);
 
 extern struct work_struct efivar_work;
 void efivar_run_worker(void);
diff --git a/include/linux/fsnotify_backend.h b/include/linux/fsnotify_backend.h
index 6b7e89f45aa4..533c4408529a 100644
--- a/include/linux/fsnotify_backend.h
+++ b/include/linux/fsnotify_backend.h
@@ -220,10 +220,7 @@ struct fsnotify_mark {
 	/* List of marks by group->i_fsnotify_marks. Also reused for queueing
 	 * mark into destroy_list when it's waiting for the end of SRCU period
 	 * before it can be freed. [group->mark_mutex] */
-	union {
-		struct list_head g_list;
-		struct rcu_head g_rcu;
-	};
+	struct list_head g_list;
 	/* Protects inode / mnt pointers, flags, masks */
 	spinlock_t lock;
 	/* List of marks for inode / vfsmount [obj_lock] */
diff --git a/include/linux/ftrace.h b/include/linux/ftrace.h
index 81de7123959d..c2b340e23f62 100644
--- a/include/linux/ftrace.h
+++ b/include/linux/ftrace.h
@@ -603,6 +603,7 @@ extern int ftrace_arch_read_dyn_info(char *buf, int size);
 
 extern int skip_trace(unsigned long ip);
 extern void ftrace_module_init(struct module *mod);
+extern void ftrace_module_enable(struct module *mod);
 extern void ftrace_release_mod(struct module *mod);
 
 extern void ftrace_disable_daemon(void);
@@ -612,8 +613,9 @@ static inline int skip_trace(unsigned long ip) { return 0; }
 static inline int ftrace_force_update(void) { return 0; }
 static inline void ftrace_disable_daemon(void) { }
 static inline void ftrace_enable_daemon(void) { }
-static inline void ftrace_release_mod(struct module *mod) {}
-static inline void ftrace_module_init(struct module *mod) {}
+static inline void ftrace_module_init(struct module *mod) { }
+static inline void ftrace_module_enable(struct module *mod) { }
+static inline void ftrace_release_mod(struct module *mod) { }
 static inline __init int register_ftrace_command(struct ftrace_func_command *cmd)
 {
 	return -EINVAL;
diff --git a/include/linux/intel-iommu.h b/include/linux/intel-iommu.h
index 821273ca4873..2d9b650047a5 100644
--- a/include/linux/intel-iommu.h
+++ b/include/linux/intel-iommu.h
@@ -235,6 +235,9 @@ static inline void dmar_writeq(void __iomem *addr, u64 val)
 /* low 64 bit */
 #define dma_frcd_page_addr(d) (d & (((u64)-1) << PAGE_SHIFT))
 
+/* PRS_REG */
+#define DMA_PRS_PPR	((u32)1)
+
 #define IOMMU_WAIT_OP(iommu, offset, op, cond, sts)			\
 do {									\
 	cycles_t start_time = get_cycles();				\
diff --git a/include/linux/kasan.h b/include/linux/kasan.h
index 4b9f85c963d0..0fdc798e3ff7 100644
--- a/include/linux/kasan.h
+++ b/include/linux/kasan.h
@@ -1,6 +1,7 @@
 #ifndef _LINUX_KASAN_H
 #define _LINUX_KASAN_H
 
+#include <linux/sched.h>
 #include <linux/types.h>
 
 struct kmem_cache;
@@ -13,7 +14,6 @@ struct vm_struct;
 
 #include <asm/kasan.h>
 #include <asm/pgtable.h>
-#include <linux/sched.h>
 
 extern unsigned char kasan_zero_page[PAGE_SIZE];
 extern pte_t kasan_zero_pte[PTRS_PER_PTE];
@@ -43,6 +43,8 @@ static inline void kasan_disable_current(void)
 
 void kasan_unpoison_shadow(const void *address, size_t size);
 
+void kasan_unpoison_task_stack(struct task_struct *task);
+
 void kasan_alloc_pages(struct page *page, unsigned int order);
 void kasan_free_pages(struct page *page, unsigned int order);
 
@@ -66,6 +68,8 @@ void kasan_free_shadow(const struct vm_struct *vm);
 
 static inline void kasan_unpoison_shadow(const void *address, size_t size) {}
 
+static inline void kasan_unpoison_task_stack(struct task_struct *task) {}
+
 static inline void kasan_enable_current(void) {}
 static inline void kasan_disable_current(void) {}
 
diff --git a/include/linux/libata.h b/include/linux/libata.h
index 851821bfd553..2c4ebef79d0c 100644
--- a/include/linux/libata.h
+++ b/include/linux/libata.h
@@ -526,6 +526,7 @@ enum ata_lpm_policy {
 enum ata_lpm_hints {
 	ATA_LPM_EMPTY		= (1 << 0), /* port empty/probing */
 	ATA_LPM_HIPM		= (1 << 1), /* may use HIPM */
+	ATA_LPM_WAKE_ONLY	= (1 << 2), /* only wake up link */
 };
 
 /* forward declarations */
@@ -719,7 +720,7 @@ struct ata_device {
 	union {
 		u16		id[ATA_ID_WORDS]; /* IDENTIFY xxx DEVICE data */
 		u32		gscr[SATA_PMP_GSCR_DWORDS]; /* PMP GSCR block */
-	};
+	} ____cacheline_aligned;
 
 	/* DEVSLP Timing Variables from Identify Device Data Log */
 	u8			devslp_timing[ATA_LOG_DEVSLP_SIZE];
diff --git a/include/linux/libnvdimm.h b/include/linux/libnvdimm.h
index bed40dff0e86..141ffdd59960 100644
--- a/include/linux/libnvdimm.h
+++ b/include/linux/libnvdimm.h
@@ -26,9 +26,8 @@ enum {
 
 	/* need to set a limit somewhere, but yes, this is likely overkill */
 	ND_IOCTL_MAX_BUFLEN = SZ_4M,
-	ND_CMD_MAX_ELEM = 4,
+	ND_CMD_MAX_ELEM = 5,
 	ND_CMD_MAX_ENVELOPE = 16,
-	ND_CMD_ARS_STATUS_MAX = SZ_4K,
 	ND_MAX_MAPPINGS = 32,
 
 	/* region flag indicating to direct-map persistent memory by default */
diff --git a/include/linux/lightnvm.h b/include/linux/lightnvm.h
index d6750111e48e..2190419bdf0a 100644
--- a/include/linux/lightnvm.h
+++ b/include/linux/lightnvm.h
@@ -135,6 +135,10 @@ enum {
 	/* Memory types */
 	NVM_ID_FMTYPE_SLC	= 0,
 	NVM_ID_FMTYPE_MLC	= 1,
+
+	/* Device capabilities */
+	NVM_ID_DCAP_BBLKMGMT	= 0x1,
+	NVM_UD_DCAP_ECC		= 0x2,
 };
 
 struct nvm_id_lp_mlc {
diff --git a/include/linux/list.h b/include/linux/list.h
index 30cf4200ab40..5356f4d661a7 100644
--- a/include/linux/list.h
+++ b/include/linux/list.h
@@ -113,17 +113,6 @@ extern void __list_del_entry(struct list_head *entry);
 extern void list_del(struct list_head *entry);
 #endif
 
-#ifdef CONFIG_DEBUG_LIST
-/*
- * See devm_memremap_pages() which wants DEBUG_LIST=y to assert if one
- * of the pages it allocates is ever passed to list_add()
- */
-extern void list_force_poison(struct list_head *entry);
-#else
-/* fallback to the less strict LIST_POISON* definitions */
-#define list_force_poison list_del
-#endif
-
 /**
  * list_replace - replace old entry by new one
  * @old : the element to be replaced
diff --git a/include/linux/lockdep.h b/include/linux/lockdep.h
index c57e424d914b..4dca42fd32f5 100644
--- a/include/linux/lockdep.h
+++ b/include/linux/lockdep.h
@@ -66,7 +66,7 @@ struct lock_class {
 	/*
 	 * class-hash:
 	 */
-	struct list_head		hash_entry;
+	struct hlist_node		hash_entry;
 
 	/*
 	 * global list of all lock-classes:
@@ -199,7 +199,7 @@ struct lock_chain {
 	u8				irq_context;
 	u8				depth;
 	u16				base;
-	struct list_head		entry;
+	struct hlist_node		entry;
 	u64				chain_key;
 };
 
diff --git a/include/linux/mlx4/device.h b/include/linux/mlx4/device.h
index 430a929f048b..a0e8cc8dcc67 100644
--- a/include/linux/mlx4/device.h
+++ b/include/linux/mlx4/device.h
@@ -44,6 +44,8 @@
 
 #include <linux/timecounter.h>
 
+#define DEFAULT_UAR_PAGE_SHIFT  12
+
 #define MAX_MSIX_P_PORT		17
 #define MAX_MSIX		64
 #define MIN_MSIX_P_PORT		5
@@ -856,6 +858,7 @@ struct mlx4_dev {
 	u64			regid_promisc_array[MLX4_MAX_PORTS + 1];
 	u64			regid_allmulti_array[MLX4_MAX_PORTS + 1];
 	struct mlx4_vf_dev     *dev_vfs;
+	u8  uar_page_shift;
 };
 
 struct mlx4_clock_params {
@@ -1528,4 +1531,14 @@ int mlx4_ACCESS_PTYS_REG(struct mlx4_dev *dev,
 int mlx4_get_internal_clock_params(struct mlx4_dev *dev,
 				   struct mlx4_clock_params *params);
 
+static inline int mlx4_to_hw_uar_index(struct mlx4_dev *dev, int index)
+{
+	return (index << (PAGE_SHIFT - dev->uar_page_shift));
+}
+
+static inline int mlx4_get_num_reserved_uar(struct mlx4_dev *dev)
+{
+	/* The first 128 UARs are used for EQ doorbells */
+	return (128 >> (PAGE_SHIFT - dev->uar_page_shift));
+}
 #endif /* MLX4_DEVICE_H */
diff --git a/include/linux/mlx5/mlx5_ifc.h b/include/linux/mlx5/mlx5_ifc.h
index 231ab6bcea76..58eef02edc7e 100644
--- a/include/linux/mlx5/mlx5_ifc.h
+++ b/include/linux/mlx5/mlx5_ifc.h
@@ -207,15 +207,15 @@ struct mlx5_ifc_flow_table_fields_supported_bits {
 	u8         outer_dmac[0x1];
 	u8         outer_smac[0x1];
 	u8         outer_ether_type[0x1];
-	u8         reserved_0[0x1];
+	u8         reserved_at_3[0x1];
 	u8         outer_first_prio[0x1];
 	u8         outer_first_cfi[0x1];
 	u8         outer_first_vid[0x1];
-	u8         reserved_1[0x1];
+	u8         reserved_at_7[0x1];
 	u8         outer_second_prio[0x1];
 	u8         outer_second_cfi[0x1];
 	u8         outer_second_vid[0x1];
-	u8         reserved_2[0x1];
+	u8         reserved_at_b[0x1];
 	u8         outer_sip[0x1];
 	u8         outer_dip[0x1];
 	u8         outer_frag[0x1];
@@ -230,21 +230,21 @@ struct mlx5_ifc_flow_table_fields_supported_bits {
 	u8         outer_gre_protocol[0x1];
 	u8         outer_gre_key[0x1];
 	u8         outer_vxlan_vni[0x1];
-	u8         reserved_3[0x5];
+	u8         reserved_at_1a[0x5];
 	u8         source_eswitch_port[0x1];
 
 	u8         inner_dmac[0x1];
 	u8         inner_smac[0x1];
 	u8         inner_ether_type[0x1];
-	u8         reserved_4[0x1];
+	u8         reserved_at_23[0x1];
 	u8         inner_first_prio[0x1];
 	u8         inner_first_cfi[0x1];
 	u8         inner_first_vid[0x1];
-	u8         reserved_5[0x1];
+	u8         reserved_at_27[0x1];
 	u8         inner_second_prio[0x1];
 	u8         inner_second_cfi[0x1];
 	u8         inner_second_vid[0x1];
-	u8         reserved_6[0x1];
+	u8         reserved_at_2b[0x1];
 	u8         inner_sip[0x1];
 	u8         inner_dip[0x1];
 	u8         inner_frag[0x1];
@@ -256,37 +256,37 @@ struct mlx5_ifc_flow_table_fields_supported_bits {
 	u8         inner_tcp_sport[0x1];
 	u8         inner_tcp_dport[0x1];
 	u8         inner_tcp_flags[0x1];
-	u8         reserved_7[0x9];
+	u8         reserved_at_37[0x9];
 
-	u8         reserved_8[0x40];
+	u8         reserved_at_40[0x40];
 };
 
 struct mlx5_ifc_flow_table_prop_layout_bits {
 	u8         ft_support[0x1];
-	u8         reserved_0[0x2];
+	u8         reserved_at_1[0x2];
 	u8	   flow_modify_en[0x1];
 	u8         modify_root[0x1];
 	u8         identified_miss_table_mode[0x1];
 	u8         flow_table_modify[0x1];
-	u8         reserved_1[0x19];
+	u8         reserved_at_7[0x19];
 
-	u8         reserved_2[0x2];
+	u8         reserved_at_20[0x2];
 	u8         log_max_ft_size[0x6];
-	u8         reserved_3[0x10];
+	u8         reserved_at_28[0x10];
 	u8         max_ft_level[0x8];
 
-	u8         reserved_4[0x20];
+	u8         reserved_at_40[0x20];
 
-	u8         reserved_5[0x18];
+	u8         reserved_at_60[0x18];
 	u8         log_max_ft_num[0x8];
 
-	u8         reserved_6[0x18];
+	u8         reserved_at_80[0x18];
 	u8         log_max_destination[0x8];
 
-	u8         reserved_7[0x18];
+	u8         reserved_at_a0[0x18];
 	u8         log_max_flow[0x8];
 
-	u8         reserved_8[0x40];
+	u8         reserved_at_c0[0x40];
 
 	struct mlx5_ifc_flow_table_fields_supported_bits ft_field_support;
 
@@ -298,13 +298,13 @@ struct mlx5_ifc_odp_per_transport_service_cap_bits {
 	u8         receive[0x1];
 	u8         write[0x1];
 	u8         read[0x1];
-	u8         reserved_0[0x1];
+	u8         reserved_at_4[0x1];
 	u8         srq_receive[0x1];
-	u8         reserved_1[0x1a];
+	u8         reserved_at_6[0x1a];
 };
 
 struct mlx5_ifc_ipv4_layout_bits {
-	u8         reserved_0[0x60];
+	u8         reserved_at_0[0x60];
 
 	u8         ipv4[0x20];
 };
@@ -316,7 +316,7 @@ struct mlx5_ifc_ipv6_layout_bits {
 union mlx5_ifc_ipv6_layout_ipv4_layout_auto_bits {
 	struct mlx5_ifc_ipv6_layout_bits ipv6_layout;
 	struct mlx5_ifc_ipv4_layout_bits ipv4_layout;
-	u8         reserved_0[0x80];
+	u8         reserved_at_0[0x80];
 };
 
 struct mlx5_ifc_fte_match_set_lyr_2_4_bits {
@@ -336,15 +336,15 @@ struct mlx5_ifc_fte_match_set_lyr_2_4_bits {
 	u8         ip_dscp[0x6];
 	u8         ip_ecn[0x2];
 	u8         vlan_tag[0x1];
-	u8         reserved_0[0x1];
+	u8         reserved_at_91[0x1];
 	u8         frag[0x1];
-	u8         reserved_1[0x4];
+	u8         reserved_at_93[0x4];
 	u8         tcp_flags[0x9];
 
 	u8         tcp_sport[0x10];
 	u8         tcp_dport[0x10];
 
-	u8         reserved_2[0x20];
+	u8         reserved_at_c0[0x20];
 
 	u8         udp_sport[0x10];
 	u8         udp_dport[0x10];
@@ -355,9 +355,9 @@ struct mlx5_ifc_fte_match_set_lyr_2_4_bits {
 };
 
 struct mlx5_ifc_fte_match_set_misc_bits {
-	u8         reserved_0[0x20];
+	u8         reserved_at_0[0x20];
 
-	u8         reserved_1[0x10];
+	u8         reserved_at_20[0x10];
 	u8         source_port[0x10];
 
 	u8         outer_second_prio[0x3];
@@ -369,31 +369,31 @@ struct mlx5_ifc_fte_match_set_misc_bits {
 
 	u8         outer_second_vlan_tag[0x1];
 	u8         inner_second_vlan_tag[0x1];
-	u8         reserved_2[0xe];
+	u8         reserved_at_62[0xe];
 	u8         gre_protocol[0x10];
 
 	u8         gre_key_h[0x18];
 	u8         gre_key_l[0x8];
 
 	u8         vxlan_vni[0x18];
-	u8         reserved_3[0x8];
+	u8         reserved_at_b8[0x8];
 
-	u8         reserved_4[0x20];
+	u8         reserved_at_c0[0x20];
 
-	u8         reserved_5[0xc];
+	u8         reserved_at_e0[0xc];
 	u8         outer_ipv6_flow_label[0x14];
 
-	u8         reserved_6[0xc];
+	u8         reserved_at_100[0xc];
 	u8         inner_ipv6_flow_label[0x14];
 
-	u8         reserved_7[0xe0];
+	u8         reserved_at_120[0xe0];
 };
 
 struct mlx5_ifc_cmd_pas_bits {
 	u8         pa_h[0x20];
 
 	u8         pa_l[0x14];
-	u8         reserved_0[0xc];
+	u8         reserved_at_34[0xc];
 };
 
 struct mlx5_ifc_uint64_bits {
@@ -418,31 +418,31 @@ enum {
 struct mlx5_ifc_ads_bits {
 	u8         fl[0x1];
 	u8         free_ar[0x1];
-	u8         reserved_0[0xe];
+	u8         reserved_at_2[0xe];
 	u8         pkey_index[0x10];
 
-	u8         reserved_1[0x8];
+	u8         reserved_at_20[0x8];
 	u8         grh[0x1];
 	u8         mlid[0x7];
 	u8         rlid[0x10];
 
 	u8         ack_timeout[0x5];
-	u8         reserved_2[0x3];
+	u8         reserved_at_45[0x3];
 	u8         src_addr_index[0x8];
-	u8         reserved_3[0x4];
+	u8         reserved_at_50[0x4];
 	u8         stat_rate[0x4];
 	u8         hop_limit[0x8];
 
-	u8         reserved_4[0x4];
+	u8         reserved_at_60[0x4];
 	u8         tclass[0x8];
 	u8         flow_label[0x14];
 
 	u8         rgid_rip[16][0x8];
 
-	u8         reserved_5[0x4];
+	u8         reserved_at_100[0x4];
 	u8         f_dscp[0x1];
 	u8         f_ecn[0x1];
-	u8         reserved_6[0x1];
+	u8         reserved_at_106[0x1];
 	u8         f_eth_prio[0x1];
 	u8         ecn[0x2];
 	u8         dscp[0x6];
@@ -458,25 +458,25 @@ struct mlx5_ifc_ads_bits {
 };
 
 struct mlx5_ifc_flow_table_nic_cap_bits {
-	u8         reserved_0[0x200];
+	u8         reserved_at_0[0x200];
 
 	struct mlx5_ifc_flow_table_prop_layout_bits flow_table_properties_nic_receive;
 
-	u8         reserved_1[0x200];
+	u8         reserved_at_400[0x200];
 
 	struct mlx5_ifc_flow_table_prop_layout_bits flow_table_properties_nic_receive_sniffer;
 
 	struct mlx5_ifc_flow_table_prop_layout_bits flow_table_properties_nic_transmit;
 
-	u8         reserved_2[0x200];
+	u8         reserved_at_a00[0x200];
 
 	struct mlx5_ifc_flow_table_prop_layout_bits flow_table_properties_nic_transmit_sniffer;
 
-	u8         reserved_3[0x7200];
+	u8         reserved_at_e00[0x7200];
 };
 
 struct mlx5_ifc_flow_table_eswitch_cap_bits {
-	u8     reserved_0[0x200];
+	u8     reserved_at_0[0x200];
 
 	struct mlx5_ifc_flow_table_prop_layout_bits flow_table_properties_nic_esw_fdb;
 
@@ -484,7 +484,7 @@ struct mlx5_ifc_flow_table_eswitch_cap_bits {
 
 	struct mlx5_ifc_flow_table_prop_layout_bits flow_table_properties_esw_acl_egress;
 
-	u8      reserved_1[0x7800];
+	u8      reserved_at_800[0x7800];
 };
 
 struct mlx5_ifc_e_switch_cap_bits {
@@ -493,9 +493,9 @@ struct mlx5_ifc_e_switch_cap_bits {
 	u8         vport_svlan_insert[0x1];
 	u8         vport_cvlan_insert_if_not_exist[0x1];
 	u8         vport_cvlan_insert_overwrite[0x1];
-	u8         reserved_0[0x1b];
+	u8         reserved_at_5[0x1b];
 
-	u8         reserved_1[0x7e0];
+	u8         reserved_at_20[0x7e0];
 };
 
 struct mlx5_ifc_per_protocol_networking_offload_caps_bits {
@@ -504,51 +504,51 @@ struct mlx5_ifc_per_protocol_networking_offload_caps_bits {
 	u8         lro_cap[0x1];
 	u8         lro_psh_flag[0x1];
 	u8         lro_time_stamp[0x1];
-	u8         reserved_0[0x3];
+	u8         reserved_at_5[0x3];
 	u8         self_lb_en_modifiable[0x1];
-	u8         reserved_1[0x2];
+	u8         reserved_at_9[0x2];
 	u8         max_lso_cap[0x5];
-	u8         reserved_2[0x4];
+	u8         reserved_at_10[0x4];
 	u8         rss_ind_tbl_cap[0x4];
-	u8         reserved_3[0x3];
+	u8         reserved_at_18[0x3];
 	u8         tunnel_lso_const_out_ip_id[0x1];
-	u8         reserved_4[0x2];
+	u8         reserved_at_1c[0x2];
 	u8         tunnel_statless_gre[0x1];
 	u8         tunnel_stateless_vxlan[0x1];
 
-	u8         reserved_5[0x20];
+	u8         reserved_at_20[0x20];
 
-	u8         reserved_6[0x10];
+	u8         reserved_at_40[0x10];
 	u8         lro_min_mss_size[0x10];
 
-	u8         reserved_7[0x120];
+	u8         reserved_at_60[0x120];
 
 	u8         lro_timer_supported_periods[4][0x20];
 
-	u8         reserved_8[0x600];
+	u8         reserved_at_200[0x600];
 };
 
 struct mlx5_ifc_roce_cap_bits {
 	u8         roce_apm[0x1];
-	u8         reserved_0[0x1f];
+	u8         reserved_at_1[0x1f];
 
-	u8         reserved_1[0x60];
+	u8         reserved_at_20[0x60];
 
-	u8         reserved_2[0xc];
+	u8         reserved_at_80[0xc];
 	u8         l3_type[0x4];
-	u8         reserved_3[0x8];
+	u8         reserved_at_90[0x8];
 	u8         roce_version[0x8];
 
-	u8         reserved_4[0x10];
+	u8         reserved_at_a0[0x10];
 	u8         r_roce_dest_udp_port[0x10];
 
 	u8         r_roce_max_src_udp_port[0x10];
 	u8         r_roce_min_src_udp_port[0x10];
 
-	u8         reserved_5[0x10];
+	u8         reserved_at_e0[0x10];
 	u8         roce_address_table_size[0x10];
 
-	u8         reserved_6[0x700];
+	u8         reserved_at_100[0x700];
 };
 
 enum {
@@ -576,35 +576,35 @@ enum {
 };
 
 struct mlx5_ifc_atomic_caps_bits {
-	u8         reserved_0[0x40];
+	u8         reserved_at_0[0x40];
 
 	u8         atomic_req_8B_endianess_mode[0x2];
-	u8         reserved_1[0x4];
+	u8         reserved_at_42[0x4];
 	u8         supported_atomic_req_8B_endianess_mode_1[0x1];
 
-	u8         reserved_2[0x19];
+	u8         reserved_at_47[0x19];
 
-	u8         reserved_3[0x20];
+	u8         reserved_at_60[0x20];
 
-	u8         reserved_4[0x10];
+	u8         reserved_at_80[0x10];
 	u8         atomic_operations[0x10];
 
-	u8         reserved_5[0x10];
+	u8         reserved_at_a0[0x10];
 	u8         atomic_size_qp[0x10];
 
-	u8         reserved_6[0x10];
+	u8         reserved_at_c0[0x10];
 	u8         atomic_size_dc[0x10];
 
-	u8         reserved_7[0x720];
+	u8         reserved_at_e0[0x720];
 };
 
 struct mlx5_ifc_odp_cap_bits {
-	u8         reserved_0[0x40];
+	u8         reserved_at_0[0x40];
 
 	u8         sig[0x1];
-	u8         reserved_1[0x1f];
+	u8         reserved_at_41[0x1f];
 
-	u8         reserved_2[0x20];
+	u8         reserved_at_60[0x20];
 
 	struct mlx5_ifc_odp_per_transport_service_cap_bits rc_odp_caps;
 
@@ -612,7 +612,7 @@ struct mlx5_ifc_odp_cap_bits {
 
 	struct mlx5_ifc_odp_per_transport_service_cap_bits ud_odp_caps;
 
-	u8         reserved_3[0x720];
+	u8         reserved_at_e0[0x720];
 };
 
 enum {
@@ -660,55 +660,55 @@ enum {
 };
 
 struct mlx5_ifc_cmd_hca_cap_bits {
-	u8         reserved_0[0x80];
+	u8         reserved_at_0[0x80];
 
 	u8         log_max_srq_sz[0x8];
 	u8         log_max_qp_sz[0x8];
-	u8         reserved_1[0xb];
+	u8         reserved_at_90[0xb];
 	u8         log_max_qp[0x5];
 
-	u8         reserved_2[0xb];
+	u8         reserved_at_a0[0xb];
 	u8         log_max_srq[0x5];
-	u8         reserved_3[0x10];
+	u8         reserved_at_b0[0x10];
 
-	u8         reserved_4[0x8];
+	u8         reserved_at_c0[0x8];
 	u8         log_max_cq_sz[0x8];
-	u8         reserved_5[0xb];
+	u8         reserved_at_d0[0xb];
 	u8         log_max_cq[0x5];
 
 	u8         log_max_eq_sz[0x8];
-	u8         reserved_6[0x2];
+	u8         reserved_at_e8[0x2];
 	u8         log_max_mkey[0x6];
-	u8         reserved_7[0xc];
+	u8         reserved_at_f0[0xc];
 	u8         log_max_eq[0x4];
 
 	u8         max_indirection[0x8];
-	u8         reserved_8[0x1];
+	u8         reserved_at_108[0x1];
 	u8         log_max_mrw_sz[0x7];
-	u8         reserved_9[0x2];
+	u8         reserved_at_110[0x2];
 	u8         log_max_bsf_list_size[0x6];
-	u8         reserved_10[0x2];
+	u8         reserved_at_118[0x2];
 	u8         log_max_klm_list_size[0x6];
 
-	u8         reserved_11[0xa];
+	u8         reserved_at_120[0xa];
 	u8         log_max_ra_req_dc[0x6];
-	u8         reserved_12[0xa];
+	u8         reserved_at_130[0xa];
 	u8         log_max_ra_res_dc[0x6];
 
-	u8         reserved_13[0xa];
+	u8         reserved_at_140[0xa];
 	u8         log_max_ra_req_qp[0x6];
-	u8         reserved_14[0xa];
+	u8         reserved_at_150[0xa];
 	u8         log_max_ra_res_qp[0x6];
 
 	u8         pad_cap[0x1];
 	u8         cc_query_allowed[0x1];
 	u8         cc_modify_allowed[0x1];
-	u8         reserved_15[0xd];
+	u8         reserved_at_163[0xd];
 	u8         gid_table_size[0x10];
 
 	u8         out_of_seq_cnt[0x1];
 	u8         vport_counters[0x1];
-	u8         reserved_16[0x4];
+	u8         reserved_at_182[0x4];
 	u8         max_qp_cnt[0xa];
 	u8         pkey_table_size[0x10];
 
@@ -716,158 +716,158 @@ struct mlx5_ifc_cmd_hca_cap_bits {
 	u8         vhca_group_manager[0x1];
 	u8         ib_virt[0x1];
 	u8         eth_virt[0x1];
-	u8         reserved_17[0x1];
+	u8         reserved_at_1a4[0x1];
 	u8         ets[0x1];
 	u8         nic_flow_table[0x1];
 	u8         eswitch_flow_table[0x1];
 	u8	   early_vf_enable;
-	u8         reserved_18[0x2];
+	u8         reserved_at_1a8[0x2];
 	u8         local_ca_ack_delay[0x5];
-	u8         reserved_19[0x6];
+	u8         reserved_at_1af[0x6];
 	u8         port_type[0x2];
 	u8         num_ports[0x8];
 
-	u8         reserved_20[0x3];
+	u8         reserved_at_1bf[0x3];
 	u8         log_max_msg[0x5];
-	u8         reserved_21[0x18];
+	u8         reserved_at_1c7[0x18];
 
 	u8         stat_rate_support[0x10];
-	u8         reserved_22[0xc];
+	u8         reserved_at_1ef[0xc];
 	u8         cqe_version[0x4];
 
 	u8         compact_address_vector[0x1];
-	u8         reserved_23[0xe];
+	u8         reserved_at_200[0xe];
 	u8         drain_sigerr[0x1];
 	u8         cmdif_checksum[0x2];
 	u8         sigerr_cqe[0x1];
-	u8         reserved_24[0x1];
+	u8         reserved_at_212[0x1];
 	u8         wq_signature[0x1];
 	u8         sctr_data_cqe[0x1];
-	u8         reserved_25[0x1];
+	u8         reserved_at_215[0x1];
 	u8         sho[0x1];
 	u8         tph[0x1];
 	u8         rf[0x1];
 	u8         dct[0x1];
-	u8         reserved_26[0x1];
+	u8         reserved_at_21a[0x1];
 	u8         eth_net_offloads[0x1];
 	u8         roce[0x1];
 	u8         atomic[0x1];
-	u8         reserved_27[0x1];
+	u8         reserved_at_21e[0x1];
 
 	u8         cq_oi[0x1];
 	u8         cq_resize[0x1];
 	u8         cq_moderation[0x1];
-	u8         reserved_28[0x3];
+	u8         reserved_at_222[0x3];
 	u8         cq_eq_remap[0x1];
 	u8         pg[0x1];
 	u8         block_lb_mc[0x1];
-	u8         reserved_29[0x1];
+	u8         reserved_at_228[0x1];
 	u8         scqe_break_moderation[0x1];
-	u8         reserved_30[0x1];
+	u8         reserved_at_22a[0x1];
 	u8         cd[0x1];
-	u8         reserved_31[0x1];
+	u8         reserved_at_22c[0x1];
 	u8         apm[0x1];
-	u8         reserved_32[0x7];
+	u8         reserved_at_22e[0x7];
 	u8         qkv[0x1];
 	u8         pkv[0x1];
-	u8         reserved_33[0x4];
+	u8         reserved_at_237[0x4];
 	u8         xrc[0x1];
 	u8         ud[0x1];
 	u8         uc[0x1];
 	u8         rc[0x1];
 
-	u8         reserved_34[0xa];
+	u8         reserved_at_23f[0xa];
 	u8         uar_sz[0x6];
-	u8         reserved_35[0x8];
+	u8         reserved_at_24f[0x8];
 	u8         log_pg_sz[0x8];
 
 	u8         bf[0x1];
-	u8         reserved_36[0x1];
+	u8         reserved_at_260[0x1];
 	u8         pad_tx_eth_packet[0x1];
-	u8         reserved_37[0x8];
+	u8         reserved_at_262[0x8];
 	u8         log_bf_reg_size[0x5];
-	u8         reserved_38[0x10];
+	u8         reserved_at_26f[0x10];
 
-	u8         reserved_39[0x10];
+	u8         reserved_at_27f[0x10];
 	u8         max_wqe_sz_sq[0x10];
 
-	u8         reserved_40[0x10];
+	u8         reserved_at_29f[0x10];
 	u8         max_wqe_sz_rq[0x10];
 
-	u8         reserved_41[0x10];
+	u8         reserved_at_2bf[0x10];
 	u8         max_wqe_sz_sq_dc[0x10];
 
-	u8         reserved_42[0x7];
+	u8         reserved_at_2df[0x7];
 	u8         max_qp_mcg[0x19];
 
-	u8         reserved_43[0x18];
+	u8         reserved_at_2ff[0x18];
 	u8         log_max_mcg[0x8];
 
-	u8         reserved_44[0x3];
+	u8         reserved_at_31f[0x3];
 	u8         log_max_transport_domain[0x5];
-	u8         reserved_45[0x3];
+	u8         reserved_at_327[0x3];
 	u8         log_max_pd[0x5];
-	u8         reserved_46[0xb];
+	u8         reserved_at_32f[0xb];
 	u8         log_max_xrcd[0x5];
 
-	u8         reserved_47[0x20];
+	u8         reserved_at_33f[0x20];
 
-	u8         reserved_48[0x3];
+	u8         reserved_at_35f[0x3];
 	u8         log_max_rq[0x5];
-	u8         reserved_49[0x3];
+	u8         reserved_at_367[0x3];
 	u8         log_max_sq[0x5];
-	u8         reserved_50[0x3];
+	u8         reserved_at_36f[0x3];
 	u8         log_max_tir[0x5];
-	u8         reserved_51[0x3];
+	u8         reserved_at_377[0x3];
 	u8         log_max_tis[0x5];
 
 	u8         basic_cyclic_rcv_wqe[0x1];
-	u8         reserved_52[0x2];
+	u8         reserved_at_380[0x2];
 	u8         log_max_rmp[0x5];
-	u8         reserved_53[0x3];
+	u8         reserved_at_387[0x3];
 	u8         log_max_rqt[0x5];
-	u8         reserved_54[0x3];
+	u8         reserved_at_38f[0x3];
 	u8         log_max_rqt_size[0x5];
-	u8         reserved_55[0x3];
+	u8         reserved_at_397[0x3];
 	u8         log_max_tis_per_sq[0x5];
 
-	u8         reserved_56[0x3];
+	u8         reserved_at_39f[0x3];
 	u8         log_max_stride_sz_rq[0x5];
-	u8         reserved_57[0x3];
+	u8         reserved_at_3a7[0x3];
 	u8         log_min_stride_sz_rq[0x5];
-	u8         reserved_58[0x3];
+	u8         reserved_at_3af[0x3];
 	u8         log_max_stride_sz_sq[0x5];
-	u8         reserved_59[0x3];
+	u8         reserved_at_3b7[0x3];
 	u8         log_min_stride_sz_sq[0x5];
 
-	u8         reserved_60[0x1b];
+	u8         reserved_at_3bf[0x1b];
 	u8         log_max_wq_sz[0x5];
 
 	u8         nic_vport_change_event[0x1];
-	u8         reserved_61[0xa];
+	u8         reserved_at_3e0[0xa];
 	u8         log_max_vlan_list[0x5];
-	u8         reserved_62[0x3];
+	u8         reserved_at_3ef[0x3];
 	u8         log_max_current_mc_list[0x5];
-	u8         reserved_63[0x3];
+	u8         reserved_at_3f7[0x3];
 	u8         log_max_current_uc_list[0x5];
 
-	u8         reserved_64[0x80];
+	u8         reserved_at_3ff[0x80];
 
-	u8         reserved_65[0x3];
+	u8         reserved_at_47f[0x3];
 	u8         log_max_l2_table[0x5];
-	u8         reserved_66[0x8];
+	u8         reserved_at_487[0x8];
 	u8         log_uar_page_sz[0x10];
 
-	u8         reserved_67[0x20];
+	u8         reserved_at_49f[0x20];
 	u8         device_frequency_mhz[0x20];
 	u8         device_frequency_khz[0x20];
-	u8         reserved_68[0x5f];
+	u8         reserved_at_4ff[0x5f];
 	u8         cqe_zip[0x1];
 
 	u8         cqe_zip_timeout[0x10];
 	u8         cqe_zip_max_num[0x10];
 
-	u8         reserved_69[0x220];
+	u8         reserved_at_57f[0x220];
 };
 
 enum mlx5_flow_destination_type {
@@ -880,7 +880,7 @@ struct mlx5_ifc_dest_format_struct_bits {
 	u8         destination_type[0x8];
 	u8         destination_id[0x18];
 
-	u8         reserved_0[0x20];
+	u8         reserved_at_20[0x20];
 };
 
 struct mlx5_ifc_fte_match_param_bits {
@@ -890,7 +890,7 @@ struct mlx5_ifc_fte_match_param_bits {
 
 	struct mlx5_ifc_fte_match_set_lyr_2_4_bits inner_headers;
 
-	u8         reserved_0[0xa00];
+	u8         reserved_at_600[0xa00];
 };
 
 enum {
@@ -922,18 +922,18 @@ struct mlx5_ifc_wq_bits {
 	u8         wq_signature[0x1];
 	u8         end_padding_mode[0x2];
 	u8         cd_slave[0x1];
-	u8         reserved_0[0x18];
+	u8         reserved_at_8[0x18];
 
 	u8         hds_skip_first_sge[0x1];
 	u8         log2_hds_buf_size[0x3];
-	u8         reserved_1[0x7];
+	u8         reserved_at_24[0x7];
 	u8         page_offset[0x5];
 	u8         lwm[0x10];
 
-	u8         reserved_2[0x8];
+	u8         reserved_at_40[0x8];
 	u8         pd[0x18];
 
-	u8         reserved_3[0x8];
+	u8         reserved_at_60[0x8];
 	u8         uar_page[0x18];
 
 	u8         dbr_addr[0x40];
@@ -942,60 +942,60 @@ struct mlx5_ifc_wq_bits {
 
 	u8         sw_counter[0x20];
 
-	u8         reserved_4[0xc];
+	u8         reserved_at_100[0xc];
 	u8         log_wq_stride[0x4];
-	u8         reserved_5[0x3];
+	u8         reserved_at_110[0x3];
 	u8         log_wq_pg_sz[0x5];
-	u8         reserved_6[0x3];
+	u8         reserved_at_118[0x3];
 	u8         log_wq_sz[0x5];
 
-	u8         reserved_7[0x4e0];
+	u8         reserved_at_120[0x4e0];
 
 	struct mlx5_ifc_cmd_pas_bits pas[0];
 };
 
 struct mlx5_ifc_rq_num_bits {
-	u8         reserved_0[0x8];
+	u8         reserved_at_0[0x8];
 	u8         rq_num[0x18];
 };
 
 struct mlx5_ifc_mac_address_layout_bits {
-	u8         reserved_0[0x10];
+	u8         reserved_at_0[0x10];
 	u8         mac_addr_47_32[0x10];
 
 	u8         mac_addr_31_0[0x20];
 };
 
 struct mlx5_ifc_vlan_layout_bits {
-	u8         reserved_0[0x14];
+	u8         reserved_at_0[0x14];
 	u8         vlan[0x0c];
 
-	u8         reserved_1[0x20];
+	u8         reserved_at_20[0x20];
 };
 
 struct mlx5_ifc_cong_control_r_roce_ecn_np_bits {
-	u8         reserved_0[0xa0];
+	u8         reserved_at_0[0xa0];
 
 	u8         min_time_between_cnps[0x20];
 
-	u8         reserved_1[0x12];
+	u8         reserved_at_c0[0x12];
 	u8         cnp_dscp[0x6];
-	u8         reserved_2[0x5];
+	u8         reserved_at_d8[0x5];
 	u8         cnp_802p_prio[0x3];
 
-	u8         reserved_3[0x720];
+	u8         reserved_at_e0[0x720];
 };
 
 struct mlx5_ifc_cong_control_r_roce_ecn_rp_bits {
-	u8         reserved_0[0x60];
+	u8         reserved_at_0[0x60];
 
-	u8         reserved_1[0x4];
+	u8         reserved_at_60[0x4];
 	u8         clamp_tgt_rate[0x1];
-	u8         reserved_2[0x3];
+	u8         reserved_at_65[0x3];
 	u8         clamp_tgt_rate_after_time_inc[0x1];
-	u8         reserved_3[0x17];
+	u8         reserved_at_69[0x17];
 
-	u8         reserved_4[0x20];
+	u8         reserved_at_80[0x20];
 
 	u8         rpg_time_reset[0x20];
 
@@ -1015,7 +1015,7 @@ struct mlx5_ifc_cong_control_r_roce_ecn_rp_bits {
 
 	u8         rpg_min_rate[0x20];
 
-	u8         reserved_5[0xe0];
+	u8         reserved_at_1c0[0xe0];
 
 	u8         rate_to_set_on_first_cnp[0x20];
 
@@ -1025,15 +1025,15 @@ struct mlx5_ifc_cong_control_r_roce_ecn_rp_bits {
 
 	u8         rate_reduce_monitor_period[0x20];
 
-	u8         reserved_6[0x20];
+	u8         reserved_at_320[0x20];
 
 	u8         initial_alpha_value[0x20];
 
-	u8         reserved_7[0x4a0];
+	u8         reserved_at_360[0x4a0];
 };
 
 struct mlx5_ifc_cong_control_802_1qau_rp_bits {
-	u8         reserved_0[0x80];
+	u8         reserved_at_0[0x80];
 
 	u8         rppp_max_rps[0x20];
 
@@ -1055,7 +1055,7 @@ struct mlx5_ifc_cong_control_802_1qau_rp_bits {
 
 	u8         rpg_min_rate[0x20];
 
-	u8         reserved_1[0x640];
+	u8         reserved_at_1c0[0x640];
 };
 
 enum {
@@ -1205,7 +1205,7 @@ struct mlx5_ifc_phys_layer_cntrs_bits {
 
 	u8         successful_recovery_events[0x20];
 
-	u8         reserved_0[0x180];
+	u8         reserved_at_640[0x180];
 };
 
 struct mlx5_ifc_eth_per_traffic_grp_data_layout_bits {
@@ -1213,7 +1213,7 @@ struct mlx5_ifc_eth_per_traffic_grp_data_layout_bits {
 
 	u8         transmit_queue_low[0x20];
 
-	u8         reserved_0[0x780];
+	u8         reserved_at_40[0x780];
 };
 
 struct mlx5_ifc_eth_per_prio_grp_data_layout_bits {
@@ -1221,7 +1221,7 @@ struct mlx5_ifc_eth_per_prio_grp_data_layout_bits {
 
 	u8         rx_octets_low[0x20];
 
-	u8         reserved_0[0xc0];
+	u8         reserved_at_40[0xc0];
 
 	u8         rx_frames_high[0x20];
 
@@ -1231,7 +1231,7 @@ struct mlx5_ifc_eth_per_prio_grp_data_layout_bits {
 
 	u8         tx_octets_low[0x20];
 
-	u8         reserved_1[0xc0];
+	u8         reserved_at_180[0xc0];
 
 	u8         tx_frames_high[0x20];
 
@@ -1257,7 +1257,7 @@ struct mlx5_ifc_eth_per_prio_grp_data_layout_bits {
 
 	u8         rx_pause_transition_low[0x20];
 
-	u8         reserved_2[0x400];
+	u8         reserved_at_3c0[0x400];
 };
 
 struct mlx5_ifc_eth_extended_cntrs_grp_data_layout_bits {
@@ -1265,7 +1265,7 @@ struct mlx5_ifc_eth_extended_cntrs_grp_data_layout_bits {
 
 	u8         port_transmit_wait_low[0x20];
 
-	u8         reserved_0[0x780];
+	u8         reserved_at_40[0x780];
 };
 
 struct mlx5_ifc_eth_3635_cntrs_grp_data_layout_bits {
@@ -1333,7 +1333,7 @@ struct mlx5_ifc_eth_3635_cntrs_grp_data_layout_bits {
 
 	u8         dot3out_pause_frames_low[0x20];
 
-	u8         reserved_0[0x3c0];
+	u8         reserved_at_400[0x3c0];
 };
 
 struct mlx5_ifc_eth_2819_cntrs_grp_data_layout_bits {
@@ -1421,7 +1421,7 @@ struct mlx5_ifc_eth_2819_cntrs_grp_data_layout_bits {
 
 	u8         ether_stats_pkts8192to10239octets_low[0x20];
 
-	u8         reserved_0[0x280];
+	u8         reserved_at_540[0x280];
 };
 
 struct mlx5_ifc_eth_2863_cntrs_grp_data_layout_bits {
@@ -1477,7 +1477,7 @@ struct mlx5_ifc_eth_2863_cntrs_grp_data_layout_bits {
 
 	u8         if_out_broadcast_pkts_low[0x20];
 
-	u8         reserved_0[0x480];
+	u8         reserved_at_340[0x480];
 };
 
 struct mlx5_ifc_eth_802_3_cntrs_grp_data_layout_bits {
@@ -1557,54 +1557,54 @@ struct mlx5_ifc_eth_802_3_cntrs_grp_data_layout_bits {
 
 	u8         a_pause_mac_ctrl_frames_transmitted_low[0x20];
 
-	u8         reserved_0[0x300];
+	u8         reserved_at_4c0[0x300];
 };
 
 struct mlx5_ifc_cmd_inter_comp_event_bits {
 	u8         command_completion_vector[0x20];
 
-	u8         reserved_0[0xc0];
+	u8         reserved_at_20[0xc0];
 };
 
 struct mlx5_ifc_stall_vl_event_bits {
-	u8         reserved_0[0x18];
+	u8         reserved_at_0[0x18];
 	u8         port_num[0x1];
-	u8         reserved_1[0x3];
+	u8         reserved_at_19[0x3];
 	u8         vl[0x4];
 
-	u8         reserved_2[0xa0];
+	u8         reserved_at_20[0xa0];
 };
 
 struct mlx5_ifc_db_bf_congestion_event_bits {
 	u8         event_subtype[0x8];
-	u8         reserved_0[0x8];
+	u8         reserved_at_8[0x8];
 	u8         congestion_level[0x8];
-	u8         reserved_1[0x8];
+	u8         reserved_at_18[0x8];
 
-	u8         reserved_2[0xa0];
+	u8         reserved_at_20[0xa0];
 };
 
 struct mlx5_ifc_gpio_event_bits {
-	u8         reserved_0[0x60];
+	u8         reserved_at_0[0x60];
 
 	u8         gpio_event_hi[0x20];
 
 	u8         gpio_event_lo[0x20];
 
-	u8         reserved_1[0x40];
+	u8         reserved_at_a0[0x40];
 };
 
 struct mlx5_ifc_port_state_change_event_bits {
-	u8         reserved_0[0x40];
+	u8         reserved_at_0[0x40];
 
 	u8         port_num[0x4];
-	u8         reserved_1[0x1c];
+	u8         reserved_at_44[0x1c];
 
-	u8         reserved_2[0x80];
+	u8         reserved_at_60[0x80];
 };
 
 struct mlx5_ifc_dropped_packet_logged_bits {
-	u8         reserved_0[0xe0];
+	u8         reserved_at_0[0xe0];
 };
 
 enum {
@@ -1613,15 +1613,15 @@ enum {
 };
 
 struct mlx5_ifc_cq_error_bits {
-	u8         reserved_0[0x8];
+	u8         reserved_at_0[0x8];
 	u8         cqn[0x18];
 
-	u8         reserved_1[0x20];
+	u8         reserved_at_20[0x20];
 
-	u8         reserved_2[0x18];
+	u8         reserved_at_40[0x18];
 	u8         syndrome[0x8];
 
-	u8         reserved_3[0x80];
+	u8         reserved_at_60[0x80];
 };
 
 struct mlx5_ifc_rdma_page_fault_event_bits {
@@ -1629,14 +1629,14 @@ struct mlx5_ifc_rdma_page_fault_event_bits {
 
 	u8         r_key[0x20];
 
-	u8         reserved_0[0x10];
+	u8         reserved_at_40[0x10];
 	u8         packet_len[0x10];
 
 	u8         rdma_op_len[0x20];
 
 	u8         rdma_va[0x40];
 
-	u8         reserved_1[0x5];
+	u8         reserved_at_c0[0x5];
 	u8         rdma[0x1];
 	u8         write[0x1];
 	u8         requestor[0x1];
@@ -1646,15 +1646,15 @@ struct mlx5_ifc_rdma_page_fault_event_bits {
 struct mlx5_ifc_wqe_associated_page_fault_event_bits {
 	u8         bytes_committed[0x20];
 
-	u8         reserved_0[0x10];
+	u8         reserved_at_20[0x10];
 	u8         wqe_index[0x10];
 
-	u8         reserved_1[0x10];
+	u8         reserved_at_40[0x10];
 	u8         len[0x10];
 
-	u8         reserved_2[0x60];
+	u8         reserved_at_60[0x60];
 
-	u8         reserved_3[0x5];
+	u8         reserved_at_c0[0x5];
 	u8         rdma[0x1];
 	u8         write_read[0x1];
 	u8         requestor[0x1];
@@ -1662,26 +1662,26 @@ struct mlx5_ifc_wqe_associated_page_fault_event_bits {
 };
 
 struct mlx5_ifc_qp_events_bits {
-	u8         reserved_0[0xa0];
+	u8         reserved_at_0[0xa0];
 
 	u8         type[0x8];
-	u8         reserved_1[0x18];
+	u8         reserved_at_a8[0x18];
 
-	u8         reserved_2[0x8];
+	u8         reserved_at_c0[0x8];
 	u8         qpn_rqn_sqn[0x18];
 };
 
 struct mlx5_ifc_dct_events_bits {
-	u8         reserved_0[0xc0];
+	u8         reserved_at_0[0xc0];
 
-	u8         reserved_1[0x8];
+	u8         reserved_at_c0[0x8];
 	u8         dct_number[0x18];
 };
 
 struct mlx5_ifc_comp_event_bits {
-	u8         reserved_0[0xc0];
+	u8         reserved_at_0[0xc0];
 
-	u8         reserved_1[0x8];
+	u8         reserved_at_c0[0x8];
 	u8         cq_number[0x18];
 };
 
@@ -1754,41 +1754,41 @@ enum {
 
 struct mlx5_ifc_qpc_bits {
 	u8         state[0x4];
-	u8         reserved_0[0x4];
+	u8         reserved_at_4[0x4];
 	u8         st[0x8];
-	u8         reserved_1[0x3];
+	u8         reserved_at_10[0x3];
 	u8         pm_state[0x2];
-	u8         reserved_2[0x7];
+	u8         reserved_at_15[0x7];
 	u8         end_padding_mode[0x2];
-	u8         reserved_3[0x2];
+	u8         reserved_at_1e[0x2];
 
 	u8         wq_signature[0x1];
 	u8         block_lb_mc[0x1];
 	u8         atomic_like_write_en[0x1];
 	u8         latency_sensitive[0x1];
-	u8         reserved_4[0x1];
+	u8         reserved_at_24[0x1];
 	u8         drain_sigerr[0x1];
-	u8         reserved_5[0x2];
+	u8         reserved_at_26[0x2];
 	u8         pd[0x18];
 
 	u8         mtu[0x3];
 	u8         log_msg_max[0x5];
-	u8         reserved_6[0x1];
+	u8         reserved_at_48[0x1];
 	u8         log_rq_size[0x4];
 	u8         log_rq_stride[0x3];
 	u8         no_sq[0x1];
 	u8         log_sq_size[0x4];
-	u8         reserved_7[0x6];
+	u8         reserved_at_55[0x6];
 	u8         rlky[0x1];
-	u8         reserved_8[0x4];
+	u8         reserved_at_5c[0x4];
 
 	u8         counter_set_id[0x8];
 	u8         uar_page[0x18];
 
-	u8         reserved_9[0x8];
+	u8         reserved_at_80[0x8];
 	u8         user_index[0x18];
 
-	u8         reserved_10[0x3];
+	u8         reserved_at_a0[0x3];
 	u8         log_page_size[0x5];
 	u8         remote_qpn[0x18];
 
@@ -1797,66 +1797,66 @@ struct mlx5_ifc_qpc_bits {
 	struct mlx5_ifc_ads_bits secondary_address_path;
 
 	u8         log_ack_req_freq[0x4];
-	u8         reserved_11[0x4];
+	u8         reserved_at_384[0x4];
 	u8         log_sra_max[0x3];
-	u8         reserved_12[0x2];
+	u8         reserved_at_38b[0x2];
 	u8         retry_count[0x3];
 	u8         rnr_retry[0x3];
-	u8         reserved_13[0x1];
+	u8         reserved_at_393[0x1];
 	u8         fre[0x1];
 	u8         cur_rnr_retry[0x3];
 	u8         cur_retry_count[0x3];
-	u8         reserved_14[0x5];
+	u8         reserved_at_39b[0x5];
 
-	u8         reserved_15[0x20];
+	u8         reserved_at_3a0[0x20];
 
-	u8         reserved_16[0x8];
+	u8         reserved_at_3c0[0x8];
 	u8         next_send_psn[0x18];
 
-	u8         reserved_17[0x8];
+	u8         reserved_at_3e0[0x8];
 	u8         cqn_snd[0x18];
 
-	u8         reserved_18[0x40];
+	u8         reserved_at_400[0x40];
 
-	u8         reserved_19[0x8];
+	u8         reserved_at_440[0x8];
 	u8         last_acked_psn[0x18];
 
-	u8         reserved_20[0x8];
+	u8         reserved_at_460[0x8];
 	u8         ssn[0x18];
 
-	u8         reserved_21[0x8];
+	u8         reserved_at_480[0x8];
 	u8         log_rra_max[0x3];
-	u8         reserved_22[0x1];
+	u8         reserved_at_48b[0x1];
 	u8         atomic_mode[0x4];
 	u8         rre[0x1];
 	u8         rwe[0x1];
 	u8         rae[0x1];
-	u8         reserved_23[0x1];
+	u8         reserved_at_493[0x1];
 	u8         page_offset[0x6];
-	u8         reserved_24[0x3];
+	u8         reserved_at_49a[0x3];
 	u8         cd_slave_receive[0x1];
 	u8         cd_slave_send[0x1];
 	u8         cd_master[0x1];
 
-	u8         reserved_25[0x3];
+	u8         reserved_at_4a0[0x3];
 	u8         min_rnr_nak[0x5];
 	u8         next_rcv_psn[0x18];
 
-	u8         reserved_26[0x8];
+	u8         reserved_at_4c0[0x8];
 	u8         xrcd[0x18];
 
-	u8         reserved_27[0x8];
+	u8         reserved_at_4e0[0x8];
 	u8         cqn_rcv[0x18];
 
 	u8         dbr_addr[0x40];
 
 	u8         q_key[0x20];
 
-	u8         reserved_28[0x5];
+	u8         reserved_at_560[0x5];
 	u8         rq_type[0x3];
 	u8         srqn_rmpn[0x18];
 
-	u8         reserved_29[0x8];
+	u8         reserved_at_580[0x8];
 	u8         rmsn[0x18];
 
 	u8         hw_sq_wqebb_counter[0x10];
@@ -1866,33 +1866,33 @@ struct mlx5_ifc_qpc_bits {
 
 	u8         sw_rq_counter[0x20];
 
-	u8         reserved_30[0x20];
+	u8         reserved_at_600[0x20];
 
-	u8         reserved_31[0xf];
+	u8         reserved_at_620[0xf];
 	u8         cgs[0x1];
 	u8         cs_req[0x8];
 	u8         cs_res[0x8];
 
 	u8         dc_access_key[0x40];
 
-	u8         reserved_32[0xc0];
+	u8         reserved_at_680[0xc0];
 };
 
 struct mlx5_ifc_roce_addr_layout_bits {
 	u8         source_l3_address[16][0x8];
 
-	u8         reserved_0[0x3];
+	u8         reserved_at_80[0x3];
 	u8         vlan_valid[0x1];
 	u8         vlan_id[0xc];
 	u8         source_mac_47_32[0x10];
 
 	u8         source_mac_31_0[0x20];
 
-	u8         reserved_1[0x14];
+	u8         reserved_at_c0[0x14];
 	u8         roce_l3_type[0x4];
 	u8         roce_version[0x8];
 
-	u8         reserved_2[0x20];
+	u8         reserved_at_e0[0x20];
 };
 
 union mlx5_ifc_hca_cap_union_bits {
@@ -1904,7 +1904,7 @@ union mlx5_ifc_hca_cap_union_bits {
 	struct mlx5_ifc_flow_table_nic_cap_bits flow_table_nic_cap;
 	struct mlx5_ifc_flow_table_eswitch_cap_bits flow_table_eswitch_cap;
 	struct mlx5_ifc_e_switch_cap_bits e_switch_cap;
-	u8         reserved_0[0x8000];
+	u8         reserved_at_0[0x8000];
 };
 
 enum {
@@ -1914,24 +1914,24 @@ enum {
 };
 
 struct mlx5_ifc_flow_context_bits {
-	u8         reserved_0[0x20];
+	u8         reserved_at_0[0x20];
 
 	u8         group_id[0x20];
 
-	u8         reserved_1[0x8];
+	u8         reserved_at_40[0x8];
 	u8         flow_tag[0x18];
 
-	u8         reserved_2[0x10];
+	u8         reserved_at_60[0x10];
 	u8         action[0x10];
 
-	u8         reserved_3[0x8];
+	u8         reserved_at_80[0x8];
 	u8         destination_list_size[0x18];
 
-	u8         reserved_4[0x160];
+	u8         reserved_at_a0[0x160];
 
 	struct mlx5_ifc_fte_match_param_bits match_value;
 
-	u8         reserved_5[0x600];
+	u8         reserved_at_1200[0x600];
 
 	struct mlx5_ifc_dest_format_struct_bits destination[0];
 };
@@ -1944,43 +1944,43 @@ enum {
 struct mlx5_ifc_xrc_srqc_bits {
 	u8         state[0x4];
 	u8         log_xrc_srq_size[0x4];
-	u8         reserved_0[0x18];
+	u8         reserved_at_8[0x18];
 
 	u8         wq_signature[0x1];
 	u8         cont_srq[0x1];
-	u8         reserved_1[0x1];
+	u8         reserved_at_22[0x1];
 	u8         rlky[0x1];
 	u8         basic_cyclic_rcv_wqe[0x1];
 	u8         log_rq_stride[0x3];
 	u8         xrcd[0x18];
 
 	u8         page_offset[0x6];
-	u8         reserved_2[0x2];
+	u8         reserved_at_46[0x2];
 	u8         cqn[0x18];
 
-	u8         reserved_3[0x20];
+	u8         reserved_at_60[0x20];
 
 	u8         user_index_equal_xrc_srqn[0x1];
-	u8         reserved_4[0x1];
+	u8         reserved_at_81[0x1];
 	u8         log_page_size[0x6];
 	u8         user_index[0x18];
 
-	u8         reserved_5[0x20];
+	u8         reserved_at_a0[0x20];
 
-	u8         reserved_6[0x8];
+	u8         reserved_at_c0[0x8];
 	u8         pd[0x18];
 
 	u8         lwm[0x10];
 	u8         wqe_cnt[0x10];
 
-	u8         reserved_7[0x40];
+	u8         reserved_at_100[0x40];
 
 	u8         db_record_addr_h[0x20];
 
 	u8         db_record_addr_l[0x1e];
-	u8         reserved_8[0x2];
+	u8         reserved_at_17e[0x2];
 
-	u8         reserved_9[0x80];
+	u8         reserved_at_180[0x80];
 };
 
 struct mlx5_ifc_traffic_counter_bits {
@@ -1990,16 +1990,16 @@ struct mlx5_ifc_traffic_counter_bits {
 };
 
 struct mlx5_ifc_tisc_bits {
-	u8         reserved_0[0xc];
+	u8         reserved_at_0[0xc];
 	u8         prio[0x4];
-	u8         reserved_1[0x10];
+	u8         reserved_at_10[0x10];
 
-	u8         reserved_2[0x100];
+	u8         reserved_at_20[0x100];
 
-	u8         reserved_3[0x8];
+	u8         reserved_at_120[0x8];
 	u8         transport_domain[0x18];
 
-	u8         reserved_4[0x3c0];
+	u8         reserved_at_140[0x3c0];
 };
 
 enum {
@@ -2024,31 +2024,31 @@ enum {
 };
 
 struct mlx5_ifc_tirc_bits {
-	u8         reserved_0[0x20];
+	u8         reserved_at_0[0x20];
 
 	u8         disp_type[0x4];
-	u8         reserved_1[0x1c];
+	u8         reserved_at_24[0x1c];
 
-	u8         reserved_2[0x40];
+	u8         reserved_at_40[0x40];
 
-	u8         reserved_3[0x4];
+	u8         reserved_at_80[0x4];
 	u8         lro_timeout_period_usecs[0x10];
 	u8         lro_enable_mask[0x4];
 	u8         lro_max_ip_payload_size[0x8];
 
-	u8         reserved_4[0x40];
+	u8         reserved_at_a0[0x40];
 
-	u8         reserved_5[0x8];
+	u8         reserved_at_e0[0x8];
 	u8         inline_rqn[0x18];
 
 	u8         rx_hash_symmetric[0x1];
-	u8         reserved_6[0x1];
+	u8         reserved_at_101[0x1];
 	u8         tunneled_offload_en[0x1];
-	u8         reserved_7[0x5];
+	u8         reserved_at_103[0x5];
 	u8         indirect_table[0x18];
 
 	u8         rx_hash_fn[0x4];
-	u8         reserved_8[0x2];
+	u8         reserved_at_124[0x2];
 	u8         self_lb_block[0x2];
 	u8         transport_domain[0x18];
 
@@ -2058,7 +2058,7 @@ struct mlx5_ifc_tirc_bits {
 
 	struct mlx5_ifc_rx_hash_field_select_bits rx_hash_field_selector_inner;
 
-	u8         reserved_9[0x4c0];
+	u8         reserved_at_2c0[0x4c0];
 };
 
 enum {
@@ -2069,39 +2069,39 @@ enum {
 struct mlx5_ifc_srqc_bits {
 	u8         state[0x4];
 	u8         log_srq_size[0x4];
-	u8         reserved_0[0x18];
+	u8         reserved_at_8[0x18];
 
 	u8         wq_signature[0x1];
 	u8         cont_srq[0x1];
-	u8         reserved_1[0x1];
+	u8         reserved_at_22[0x1];
 	u8         rlky[0x1];
-	u8         reserved_2[0x1];
+	u8         reserved_at_24[0x1];
 	u8         log_rq_stride[0x3];
 	u8         xrcd[0x18];
 
 	u8         page_offset[0x6];
-	u8         reserved_3[0x2];
+	u8         reserved_at_46[0x2];
 	u8         cqn[0x18];
 
-	u8         reserved_4[0x20];
+	u8         reserved_at_60[0x20];
 
-	u8         reserved_5[0x2];
+	u8         reserved_at_80[0x2];
 	u8         log_page_size[0x6];
-	u8         reserved_6[0x18];
+	u8         reserved_at_88[0x18];
 
-	u8         reserved_7[0x20];
+	u8         reserved_at_a0[0x20];
 
-	u8         reserved_8[0x8];
+	u8         reserved_at_c0[0x8];
 	u8         pd[0x18];
 
 	u8         lwm[0x10];
 	u8         wqe_cnt[0x10];
 
-	u8         reserved_9[0x40];
+	u8         reserved_at_100[0x40];
 
 	u8         dbr_addr[0x40];
 
-	u8         reserved_10[0x80];
+	u8         reserved_at_180[0x80];
 };
 
 enum {
@@ -2115,39 +2115,39 @@ struct mlx5_ifc_sqc_bits {
 	u8         cd_master[0x1];
 	u8         fre[0x1];
 	u8         flush_in_error_en[0x1];
-	u8         reserved_0[0x4];
+	u8         reserved_at_4[0x4];
 	u8         state[0x4];
-	u8         reserved_1[0x14];
+	u8         reserved_at_c[0x14];
 
-	u8         reserved_2[0x8];
+	u8         reserved_at_20[0x8];
 	u8         user_index[0x18];
 
-	u8         reserved_3[0x8];
+	u8         reserved_at_40[0x8];
 	u8         cqn[0x18];
 
-	u8         reserved_4[0xa0];
+	u8         reserved_at_60[0xa0];
 
 	u8         tis_lst_sz[0x10];
-	u8         reserved_5[0x10];
+	u8         reserved_at_110[0x10];
 
-	u8         reserved_6[0x40];
+	u8         reserved_at_120[0x40];
 
-	u8         reserved_7[0x8];
+	u8         reserved_at_160[0x8];
 	u8         tis_num_0[0x18];
 
 	struct mlx5_ifc_wq_bits wq;
 };
 
 struct mlx5_ifc_rqtc_bits {
-	u8         reserved_0[0xa0];
+	u8         reserved_at_0[0xa0];
 
-	u8         reserved_1[0x10];
+	u8         reserved_at_a0[0x10];
 	u8         rqt_max_size[0x10];
 
-	u8         reserved_2[0x10];
+	u8         reserved_at_c0[0x10];
 	u8         rqt_actual_size[0x10];
 
-	u8         reserved_3[0x6a0];
+	u8         reserved_at_e0[0x6a0];
 
 	struct mlx5_ifc_rq_num_bits rq_num[0];
 };
@@ -2165,27 +2165,27 @@ enum {
 
 struct mlx5_ifc_rqc_bits {
 	u8         rlky[0x1];
-	u8         reserved_0[0x2];
+	u8         reserved_at_1[0x2];
 	u8         vsd[0x1];
 	u8         mem_rq_type[0x4];
 	u8         state[0x4];
-	u8         reserved_1[0x1];
+	u8         reserved_at_c[0x1];
 	u8         flush_in_error_en[0x1];
-	u8         reserved_2[0x12];
+	u8         reserved_at_e[0x12];
 
-	u8         reserved_3[0x8];
+	u8         reserved_at_20[0x8];
 	u8         user_index[0x18];
 
-	u8         reserved_4[0x8];
+	u8         reserved_at_40[0x8];
 	u8         cqn[0x18];
 
 	u8         counter_set_id[0x8];
-	u8         reserved_5[0x18];
+	u8         reserved_at_68[0x18];
 
-	u8         reserved_6[0x8];
+	u8         reserved_at_80[0x8];
 	u8         rmpn[0x18];
 
-	u8         reserved_7[0xe0];
+	u8         reserved_at_a0[0xe0];
 
 	struct mlx5_ifc_wq_bits wq;
 };
@@ -2196,31 +2196,31 @@ enum {
 };
 
 struct mlx5_ifc_rmpc_bits {
-	u8         reserved_0[0x8];
+	u8         reserved_at_0[0x8];
 	u8         state[0x4];
-	u8         reserved_1[0x14];
+	u8         reserved_at_c[0x14];
 
 	u8         basic_cyclic_rcv_wqe[0x1];
-	u8         reserved_2[0x1f];
+	u8         reserved_at_21[0x1f];
 
-	u8         reserved_3[0x140];
+	u8         reserved_at_40[0x140];
 
 	struct mlx5_ifc_wq_bits wq;
 };
 
 struct mlx5_ifc_nic_vport_context_bits {
-	u8         reserved_0[0x1f];
+	u8         reserved_at_0[0x1f];
 	u8         roce_en[0x1];
 
 	u8         arm_change_event[0x1];
-	u8         reserved_1[0x1a];
+	u8         reserved_at_21[0x1a];
 	u8         event_on_mtu[0x1];
 	u8         event_on_promisc_change[0x1];
 	u8         event_on_vlan_change[0x1];
 	u8         event_on_mc_address_change[0x1];
 	u8         event_on_uc_address_change[0x1];
 
-	u8         reserved_2[0xf0];
+	u8         reserved_at_40[0xf0];
 
 	u8         mtu[0x10];
 
@@ -2228,21 +2228,21 @@ struct mlx5_ifc_nic_vport_context_bits {
 	u8         port_guid[0x40];
 	u8         node_guid[0x40];
 
-	u8         reserved_3[0x140];
+	u8         reserved_at_200[0x140];
 	u8         qkey_violation_counter[0x10];
-	u8         reserved_4[0x430];
+	u8         reserved_at_350[0x430];
 
 	u8         promisc_uc[0x1];
 	u8         promisc_mc[0x1];
 	u8         promisc_all[0x1];
-	u8         reserved_5[0x2];
+	u8         reserved_at_783[0x2];
 	u8         allowed_list_type[0x3];
-	u8         reserved_6[0xc];
+	u8         reserved_at_788[0xc];
 	u8         allowed_list_size[0xc];
 
 	struct mlx5_ifc_mac_address_layout_bits permanent_address;
 
-	u8         reserved_7[0x20];
+	u8         reserved_at_7e0[0x20];
 
 	u8         current_uc_mac_address[0][0x40];
 };
@@ -2254,9 +2254,9 @@ enum {
 };
 
 struct mlx5_ifc_mkc_bits {
-	u8         reserved_0[0x1];
+	u8         reserved_at_0[0x1];
 	u8         free[0x1];
-	u8         reserved_1[0xd];
+	u8         reserved_at_2[0xd];
 	u8         small_fence_on_rdma_read_response[0x1];
 	u8         umr_en[0x1];
 	u8         a[0x1];
@@ -2265,19 +2265,19 @@ struct mlx5_ifc_mkc_bits {
 	u8         lw[0x1];
 	u8         lr[0x1];
 	u8         access_mode[0x2];
-	u8         reserved_2[0x8];
+	u8         reserved_at_18[0x8];
 
 	u8         qpn[0x18];
 	u8         mkey_7_0[0x8];
 
-	u8         reserved_3[0x20];
+	u8         reserved_at_40[0x20];
 
 	u8         length64[0x1];
 	u8         bsf_en[0x1];
 	u8         sync_umr[0x1];
-	u8         reserved_4[0x2];
+	u8         reserved_at_63[0x2];
 	u8         expected_sigerr_count[0x1];
-	u8         reserved_5[0x1];
+	u8         reserved_at_66[0x1];
 	u8         en_rinval[0x1];
 	u8         pd[0x18];
 
@@ -2287,18 +2287,18 @@ struct mlx5_ifc_mkc_bits {
 
 	u8         bsf_octword_size[0x20];
 
-	u8         reserved_6[0x80];
+	u8         reserved_at_120[0x80];
 
 	u8         translations_octword_size[0x20];
 
-	u8         reserved_7[0x1b];
+	u8         reserved_at_1c0[0x1b];
 	u8         log_page_size[0x5];
 
-	u8         reserved_8[0x20];
+	u8         reserved_at_1e0[0x20];
 };
 
 struct mlx5_ifc_pkey_bits {
-	u8         reserved_0[0x10];
+	u8         reserved_at_0[0x10];
 	u8         pkey[0x10];
 };
 
@@ -2309,19 +2309,19 @@ struct mlx5_ifc_array128_auto_bits {
 struct mlx5_ifc_hca_vport_context_bits {
 	u8         field_select[0x20];
 
-	u8         reserved_0[0xe0];
+	u8         reserved_at_20[0xe0];
 
 	u8         sm_virt_aware[0x1];
 	u8         has_smi[0x1];
 	u8         has_raw[0x1];
 	u8         grh_required[0x1];
-	u8         reserved_1[0xc];
+	u8         reserved_at_104[0xc];
 	u8         port_physical_state[0x4];
 	u8         vport_state_policy[0x4];
 	u8         port_state[0x4];
 	u8         vport_state[0x4];
 
-	u8         reserved_2[0x20];
+	u8         reserved_at_120[0x20];
 
 	u8         system_image_guid[0x40];
 
@@ -2337,33 +2337,33 @@ struct mlx5_ifc_hca_vport_context_bits {
 
 	u8         cap_mask2_field_select[0x20];
 
-	u8         reserved_3[0x80];
+	u8         reserved_at_280[0x80];
 
 	u8         lid[0x10];
-	u8         reserved_4[0x4];
+	u8         reserved_at_310[0x4];
 	u8         init_type_reply[0x4];
 	u8         lmc[0x3];
 	u8         subnet_timeout[0x5];
 
 	u8         sm_lid[0x10];
 	u8         sm_sl[0x4];
-	u8         reserved_5[0xc];
+	u8         reserved_at_334[0xc];
 
 	u8         qkey_violation_counter[0x10];
 	u8         pkey_violation_counter[0x10];
 
-	u8         reserved_6[0xca0];
+	u8         reserved_at_360[0xca0];
 };
 
 struct mlx5_ifc_esw_vport_context_bits {
-	u8         reserved_0[0x3];
+	u8         reserved_at_0[0x3];
 	u8         vport_svlan_strip[0x1];
 	u8         vport_cvlan_strip[0x1];
 	u8         vport_svlan_insert[0x1];
 	u8         vport_cvlan_insert[0x2];
-	u8         reserved_1[0x18];
+	u8         reserved_at_8[0x18];
 
-	u8         reserved_2[0x20];
+	u8         reserved_at_20[0x20];
 
 	u8         svlan_cfi[0x1];
 	u8         svlan_pcp[0x3];
@@ -2372,7 +2372,7 @@ struct mlx5_ifc_esw_vport_context_bits {
 	u8         cvlan_pcp[0x3];
 	u8         cvlan_id[0xc];
 
-	u8         reserved_3[0x7a0];
+	u8         reserved_at_60[0x7a0];
 };
 
 enum {
@@ -2387,41 +2387,41 @@ enum {
 
 struct mlx5_ifc_eqc_bits {
 	u8         status[0x4];
-	u8         reserved_0[0x9];
+	u8         reserved_at_4[0x9];
 	u8         ec[0x1];
 	u8         oi[0x1];
-	u8         reserved_1[0x5];
+	u8         reserved_at_f[0x5];
 	u8         st[0x4];
-	u8         reserved_2[0x8];
+	u8         reserved_at_18[0x8];
 
-	u8         reserved_3[0x20];
+	u8         reserved_at_20[0x20];
 
-	u8         reserved_4[0x14];
+	u8         reserved_at_40[0x14];
 	u8         page_offset[0x6];
-	u8         reserved_5[0x6];
+	u8         reserved_at_5a[0x6];
 
-	u8         reserved_6[0x3];
+	u8         reserved_at_60[0x3];
 	u8         log_eq_size[0x5];
 	u8         uar_page[0x18];
 
-	u8         reserved_7[0x20];
+	u8         reserved_at_80[0x20];
 
-	u8         reserved_8[0x18];
+	u8         reserved_at_a0[0x18];
 	u8         intr[0x8];
 
-	u8         reserved_9[0x3];
+	u8         reserved_at_c0[0x3];
 	u8         log_page_size[0x5];
-	u8         reserved_10[0x18];
+	u8         reserved_at_c8[0x18];
 
-	u8         reserved_11[0x60];
+	u8         reserved_at_e0[0x60];
 
-	u8         reserved_12[0x8];
+	u8         reserved_at_140[0x8];
 	u8         consumer_counter[0x18];
 
-	u8         reserved_13[0x8];
+	u8         reserved_at_160[0x8];
 	u8         producer_counter[0x18];
 
-	u8         reserved_14[0x80];
+	u8         reserved_at_180[0x80];
 };
 
 enum {
@@ -2445,14 +2445,14 @@ enum {
 };
 
 struct mlx5_ifc_dctc_bits {
-	u8         reserved_0[0x4];
+	u8         reserved_at_0[0x4];
 	u8         state[0x4];
-	u8         reserved_1[0x18];
+	u8         reserved_at_8[0x18];
 
-	u8         reserved_2[0x8];
+	u8         reserved_at_20[0x8];
 	u8         user_index[0x18];
 
-	u8         reserved_3[0x8];
+	u8         reserved_at_40[0x8];
 	u8         cqn[0x18];
 
 	u8         counter_set_id[0x8];
@@ -2464,45 +2464,45 @@ struct mlx5_ifc_dctc_bits {
 	u8         latency_sensitive[0x1];
 	u8         rlky[0x1];
 	u8         free_ar[0x1];
-	u8         reserved_4[0xd];
+	u8         reserved_at_73[0xd];
 
-	u8         reserved_5[0x8];
+	u8         reserved_at_80[0x8];
 	u8         cs_res[0x8];
-	u8         reserved_6[0x3];
+	u8         reserved_at_90[0x3];
 	u8         min_rnr_nak[0x5];
-	u8         reserved_7[0x8];
+	u8         reserved_at_98[0x8];
 
-	u8         reserved_8[0x8];
+	u8         reserved_at_a0[0x8];
 	u8         srqn[0x18];
 
-	u8         reserved_9[0x8];
+	u8         reserved_at_c0[0x8];
 	u8         pd[0x18];
 
 	u8         tclass[0x8];
-	u8         reserved_10[0x4];
+	u8         reserved_at_e8[0x4];
 	u8         flow_label[0x14];
 
 	u8         dc_access_key[0x40];
 
-	u8         reserved_11[0x5];
+	u8         reserved_at_140[0x5];
 	u8         mtu[0x3];
 	u8         port[0x8];
 	u8         pkey_index[0x10];
 
-	u8         reserved_12[0x8];
+	u8         reserved_at_160[0x8];
 	u8         my_addr_index[0x8];
-	u8         reserved_13[0x8];
+	u8         reserved_at_170[0x8];
 	u8         hop_limit[0x8];
 
 	u8         dc_access_key_violation_count[0x20];
 
-	u8         reserved_14[0x14];
+	u8         reserved_at_1a0[0x14];
 	u8         dei_cfi[0x1];
 	u8         eth_prio[0x3];
 	u8         ecn[0x2];
 	u8         dscp[0x6];
 
-	u8         reserved_15[0x40];
+	u8         reserved_at_1c0[0x40];
 };
 
 enum {
@@ -2524,54 +2524,54 @@ enum {
 
 struct mlx5_ifc_cqc_bits {
 	u8         status[0x4];
-	u8         reserved_0[0x4];
+	u8         reserved_at_4[0x4];
 	u8         cqe_sz[0x3];
 	u8         cc[0x1];
-	u8         reserved_1[0x1];
+	u8         reserved_at_c[0x1];
 	u8         scqe_break_moderation_en[0x1];
 	u8         oi[0x1];
-	u8         reserved_2[0x2];
+	u8         reserved_at_f[0x2];
 	u8         cqe_zip_en[0x1];
 	u8         mini_cqe_res_format[0x2];
 	u8         st[0x4];
-	u8         reserved_3[0x8];
+	u8         reserved_at_18[0x8];
 
-	u8         reserved_4[0x20];
+	u8         reserved_at_20[0x20];
 
-	u8         reserved_5[0x14];
+	u8         reserved_at_40[0x14];
 	u8         page_offset[0x6];
-	u8         reserved_6[0x6];
+	u8         reserved_at_5a[0x6];
 
-	u8         reserved_7[0x3];
+	u8         reserved_at_60[0x3];
 	u8         log_cq_size[0x5];
 	u8         uar_page[0x18];
 
-	u8         reserved_8[0x4];
+	u8         reserved_at_80[0x4];
 	u8         cq_period[0xc];
 	u8         cq_max_count[0x10];
 
-	u8         reserved_9[0x18];
+	u8         reserved_at_a0[0x18];
 	u8         c_eqn[0x8];
 
-	u8         reserved_10[0x3];
+	u8         reserved_at_c0[0x3];
 	u8         log_page_size[0x5];
-	u8         reserved_11[0x18];
+	u8         reserved_at_c8[0x18];
 
-	u8         reserved_12[0x20];
+	u8         reserved_at_e0[0x20];
 
-	u8         reserved_13[0x8];
+	u8         reserved_at_100[0x8];
 	u8         last_notified_index[0x18];
 
-	u8         reserved_14[0x8];
+	u8         reserved_at_120[0x8];
 	u8         last_solicit_index[0x18];
 
-	u8         reserved_15[0x8];
+	u8         reserved_at_140[0x8];
 	u8         consumer_counter[0x18];
 
-	u8         reserved_16[0x8];
+	u8         reserved_at_160[0x8];
 	u8         producer_counter[0x18];
 
-	u8         reserved_17[0x40];
+	u8         reserved_at_180[0x40];
 
 	u8         dbr_addr[0x40];
 };
@@ -2580,16 +2580,16 @@ union mlx5_ifc_cong_control_roce_ecn_auto_bits {
 	struct mlx5_ifc_cong_control_802_1qau_rp_bits cong_control_802_1qau_rp;
 	struct mlx5_ifc_cong_control_r_roce_ecn_rp_bits cong_control_r_roce_ecn_rp;
 	struct mlx5_ifc_cong_control_r_roce_ecn_np_bits cong_control_r_roce_ecn_np;
-	u8         reserved_0[0x800];
+	u8         reserved_at_0[0x800];
 };
 
 struct mlx5_ifc_query_adapter_param_block_bits {
-	u8         reserved_0[0xc0];
+	u8         reserved_at_0[0xc0];
 
-	u8         reserved_1[0x8];
+	u8         reserved_at_c0[0x8];
 	u8         ieee_vendor_id[0x18];
 
-	u8         reserved_2[0x10];
+	u8         reserved_at_e0[0x10];
 	u8         vsd_vendor_id[0x10];
 
 	u8         vsd[208][0x8];
@@ -2600,14 +2600,14 @@ struct mlx5_ifc_query_adapter_param_block_bits {
 union mlx5_ifc_modify_field_select_resize_field_select_auto_bits {
 	struct mlx5_ifc_modify_field_select_bits modify_field_select;
 	struct mlx5_ifc_resize_field_select_bits resize_field_select;
-	u8         reserved_0[0x20];
+	u8         reserved_at_0[0x20];
 };
 
 union mlx5_ifc_field_select_802_1_r_roce_auto_bits {
 	struct mlx5_ifc_field_select_802_1qau_rp_bits field_select_802_1qau_rp;
 	struct mlx5_ifc_field_select_r_roce_rp_bits field_select_r_roce_rp;
 	struct mlx5_ifc_field_select_r_roce_np_bits field_select_r_roce_np;
-	u8         reserved_0[0x20];
+	u8         reserved_at_0[0x20];
 };
 
 union mlx5_ifc_eth_cntrs_grp_data_layout_auto_bits {
@@ -2619,7 +2619,7 @@ union mlx5_ifc_eth_cntrs_grp_data_layout_auto_bits {
 	struct mlx5_ifc_eth_per_prio_grp_data_layout_bits eth_per_prio_grp_data_layout;
 	struct mlx5_ifc_eth_per_traffic_grp_data_layout_bits eth_per_traffic_grp_data_layout;
 	struct mlx5_ifc_phys_layer_cntrs_bits phys_layer_cntrs;
-	u8         reserved_0[0x7c0];
+	u8         reserved_at_0[0x7c0];
 };
 
 union mlx5_ifc_event_auto_bits {
@@ -2635,23 +2635,23 @@ union mlx5_ifc_event_auto_bits {
 	struct mlx5_ifc_db_bf_congestion_event_bits db_bf_congestion_event;
 	struct mlx5_ifc_stall_vl_event_bits stall_vl_event;
 	struct mlx5_ifc_cmd_inter_comp_event_bits cmd_inter_comp_event;
-	u8         reserved_0[0xe0];
+	u8         reserved_at_0[0xe0];
 };
 
 struct mlx5_ifc_health_buffer_bits {
-	u8         reserved_0[0x100];
+	u8         reserved_at_0[0x100];
 
 	u8         assert_existptr[0x20];
 
 	u8         assert_callra[0x20];
 
-	u8         reserved_1[0x40];
+	u8         reserved_at_140[0x40];
 
 	u8         fw_version[0x20];
 
 	u8         hw_id[0x20];
 
-	u8         reserved_2[0x20];
+	u8         reserved_at_1c0[0x20];
 
 	u8         irisc_index[0x8];
 	u8         synd[0x8];
@@ -2660,20 +2660,20 @@ struct mlx5_ifc_health_buffer_bits {
 
 struct mlx5_ifc_register_loopback_control_bits {
 	u8         no_lb[0x1];
-	u8         reserved_0[0x7];
+	u8         reserved_at_1[0x7];
 	u8         port[0x8];
-	u8         reserved_1[0x10];
+	u8         reserved_at_10[0x10];
 
-	u8         reserved_2[0x60];
+	u8         reserved_at_20[0x60];
 };
 
 struct mlx5_ifc_teardown_hca_out_bits {
 	u8         status[0x8];
-	u8         reserved_0[0x18];
+	u8         reserved_at_8[0x18];
 
 	u8         syndrome[0x20];
 
-	u8         reserved_1[0x40];
+	u8         reserved_at_40[0x40];
 };
 
 enum {
@@ -2683,108 +2683,108 @@ enum {
 
 struct mlx5_ifc_teardown_hca_in_bits {
 	u8         opcode[0x10];
-	u8         reserved_0[0x10];
+	u8         reserved_at_10[0x10];
 
-	u8         reserved_1[0x10];
+	u8         reserved_at_20[0x10];
 	u8         op_mod[0x10];
 
-	u8         reserved_2[0x10];
+	u8         reserved_at_40[0x10];
 	u8         profile[0x10];
 
-	u8         reserved_3[0x20];
+	u8         reserved_at_60[0x20];
 };
 
 struct mlx5_ifc_sqerr2rts_qp_out_bits {
 	u8         status[0x8];
-	u8         reserved_0[0x18];
+	u8         reserved_at_8[0x18];
 
 	u8         syndrome[0x20];
 
-	u8         reserved_1[0x40];
+	u8         reserved_at_40[0x40];
 };
 
 struct mlx5_ifc_sqerr2rts_qp_in_bits {
 	u8         opcode[0x10];
-	u8         reserved_0[0x10];
+	u8         reserved_at_10[0x10];
 
-	u8         reserved_1[0x10];
+	u8         reserved_at_20[0x10];
 	u8         op_mod[0x10];
 
-	u8         reserved_2[0x8];
+	u8         reserved_at_40[0x8];
 	u8         qpn[0x18];
 
-	u8         reserved_3[0x20];
+	u8         reserved_at_60[0x20];
 
 	u8         opt_param_mask[0x20];
 
-	u8         reserved_4[0x20];
+	u8         reserved_at_a0[0x20];
 
 	struct mlx5_ifc_qpc_bits qpc;
 
-	u8         reserved_5[0x80];
+	u8         reserved_at_800[0x80];
 };
 
 struct mlx5_ifc_sqd2rts_qp_out_bits {
 	u8         status[0x8];
-	u8         reserved_0[0x18];
+	u8         reserved_at_8[0x18];
 
 	u8         syndrome[0x20];
 
-	u8         reserved_1[0x40];
+	u8         reserved_at_40[0x40];
 };
 
 struct mlx5_ifc_sqd2rts_qp_in_bits {
 	u8         opcode[0x10];
-	u8         reserved_0[0x10];
+	u8         reserved_at_10[0x10];
 
-	u8         reserved_1[0x10];
+	u8         reserved_at_20[0x10];
 	u8         op_mod[0x10];
 
-	u8         reserved_2[0x8];
+	u8         reserved_at_40[0x8];
 	u8         qpn[0x18];
 
-	u8         reserved_3[0x20];
+	u8         reserved_at_60[0x20];
 
 	u8         opt_param_mask[0x20];
 
-	u8         reserved_4[0x20];
+	u8         reserved_at_a0[0x20];
 
 	struct mlx5_ifc_qpc_bits qpc;
 
-	u8         reserved_5[0x80];
+	u8         reserved_at_800[0x80];
 };
 
 struct mlx5_ifc_set_roce_address_out_bits {
 	u8         status[0x8];
-	u8         reserved_0[0x18];
+	u8         reserved_at_8[0x18];
 
 	u8         syndrome[0x20];
 
-	u8         reserved_1[0x40];
+	u8         reserved_at_40[0x40];
 };
 
 struct mlx5_ifc_set_roce_address_in_bits {
 	u8         opcode[0x10];
-	u8         reserved_0[0x10];
+	u8         reserved_at_10[0x10];
 
-	u8         reserved_1[0x10];
+	u8         reserved_at_20[0x10];
 	u8         op_mod[0x10];
 
 	u8         roce_address_index[0x10];
-	u8         reserved_2[0x10];
+	u8         reserved_at_50[0x10];
 
-	u8         reserved_3[0x20];
+	u8         reserved_at_60[0x20];
 
 	struct mlx5_ifc_roce_addr_layout_bits roce_address;
 };
 
 struct mlx5_ifc_set_mad_demux_out_bits {
 	u8         status[0x8];
-	u8         reserved_0[0x18];
+	u8         reserved_at_8[0x18];
 
 	u8         syndrome[0x20];
 
-	u8         reserved_1[0x40];
+	u8         reserved_at_40[0x40];
 };
 
 enum {
@@ -2794,89 +2794,89 @@ enum {
 
 struct mlx5_ifc_set_mad_demux_in_bits {
 	u8         opcode[0x10];
-	u8         reserved_0[0x10];
+	u8         reserved_at_10[0x10];
 
-	u8         reserved_1[0x10];
+	u8         reserved_at_20[0x10];
 	u8         op_mod[0x10];
 
-	u8         reserved_2[0x20];
+	u8         reserved_at_40[0x20];
 
-	u8         reserved_3[0x6];
+	u8         reserved_at_60[0x6];
 	u8         demux_mode[0x2];
-	u8         reserved_4[0x18];
+	u8         reserved_at_68[0x18];
 };
 
 struct mlx5_ifc_set_l2_table_entry_out_bits {
 	u8         status[0x8];
-	u8         reserved_0[0x18];
+	u8         reserved_at_8[0x18];
 
 	u8         syndrome[0x20];
 
-	u8         reserved_1[0x40];
+	u8         reserved_at_40[0x40];
 };
 
 struct mlx5_ifc_set_l2_table_entry_in_bits {
 	u8         opcode[0x10];
-	u8         reserved_0[0x10];
+	u8         reserved_at_10[0x10];
 
-	u8         reserved_1[0x10];
+	u8         reserved_at_20[0x10];
 	u8         op_mod[0x10];
 
-	u8         reserved_2[0x60];
+	u8         reserved_at_40[0x60];
 
-	u8         reserved_3[0x8];
+	u8         reserved_at_a0[0x8];
 	u8         table_index[0x18];
 
-	u8         reserved_4[0x20];
+	u8         reserved_at_c0[0x20];
 
-	u8         reserved_5[0x13];
+	u8         reserved_at_e0[0x13];
 	u8         vlan_valid[0x1];
 	u8         vlan[0xc];
 
 	struct mlx5_ifc_mac_address_layout_bits mac_address;
 
-	u8         reserved_6[0xc0];
+	u8         reserved_at_140[0xc0];
 };
 
 struct mlx5_ifc_set_issi_out_bits {
 	u8         status[0x8];
-	u8         reserved_0[0x18];
+	u8         reserved_at_8[0x18];
 
 	u8         syndrome[0x20];
 
-	u8         reserved_1[0x40];
+	u8         reserved_at_40[0x40];
 };
 
 struct mlx5_ifc_set_issi_in_bits {
 	u8         opcode[0x10];
-	u8         reserved_0[0x10];
+	u8         reserved_at_10[0x10];
 
-	u8         reserved_1[0x10];
+	u8         reserved_at_20[0x10];
 	u8         op_mod[0x10];
 
-	u8         reserved_2[0x10];
+	u8         reserved_at_40[0x10];
 	u8         current_issi[0x10];
 
-	u8         reserved_3[0x20];
+	u8         reserved_at_60[0x20];
 };
 
 struct mlx5_ifc_set_hca_cap_out_bits {
 	u8         status[0x8];
-	u8         reserved_0[0x18];
+	u8         reserved_at_8[0x18];
 
 	u8         syndrome[0x20];
 
-	u8         reserved_1[0x40];
+	u8         reserved_at_40[0x40];
 };
 
 struct mlx5_ifc_set_hca_cap_in_bits {
 	u8         opcode[0x10];
-	u8         reserved_0[0x10];
+	u8         reserved_at_10[0x10];
 
-	u8         reserved_1[0x10];
+	u8         reserved_at_20[0x10];
 	u8         op_mod[0x10];
 
-	u8         reserved_2[0x40];
+	u8         reserved_at_40[0x40];
 
 	union mlx5_ifc_hca_cap_union_bits capability;
 };
@@ -2890,156 +2890,156 @@ enum {
 
 struct mlx5_ifc_set_fte_out_bits {
 	u8         status[0x8];
-	u8         reserved_0[0x18];
+	u8         reserved_at_8[0x18];
 
 	u8         syndrome[0x20];
 
-	u8         reserved_1[0x40];
+	u8         reserved_at_40[0x40];
 };
 
 struct mlx5_ifc_set_fte_in_bits {
 	u8         opcode[0x10];
-	u8         reserved_0[0x10];
+	u8         reserved_at_10[0x10];
 
-	u8         reserved_1[0x10];
+	u8         reserved_at_20[0x10];
 	u8         op_mod[0x10];
 
-	u8         reserved_2[0x40];
+	u8         reserved_at_40[0x40];
 
 	u8         table_type[0x8];
-	u8         reserved_3[0x18];
+	u8         reserved_at_88[0x18];
 
-	u8         reserved_4[0x8];
+	u8         reserved_at_a0[0x8];
 	u8         table_id[0x18];
 
-	u8         reserved_5[0x18];
+	u8         reserved_at_c0[0x18];
 	u8         modify_enable_mask[0x8];
 
-	u8         reserved_6[0x20];
+	u8         reserved_at_e0[0x20];
 
 	u8         flow_index[0x20];
 
-	u8         reserved_7[0xe0];
+	u8         reserved_at_120[0xe0];
 
 	struct mlx5_ifc_flow_context_bits flow_context;
 };
 
 struct mlx5_ifc_rts2rts_qp_out_bits {
 	u8         status[0x8];
-	u8         reserved_0[0x18];
+	u8         reserved_at_8[0x18];
 
 	u8         syndrome[0x20];
 
-	u8         reserved_1[0x40];
+	u8         reserved_at_40[0x40];
 };
 
 struct mlx5_ifc_rts2rts_qp_in_bits {
 	u8         opcode[0x10];
-	u8         reserved_0[0x10];
+	u8         reserved_at_10[0x10];
 
-	u8         reserved_1[0x10];
+	u8         reserved_at_20[0x10];
 	u8         op_mod[0x10];
 
-	u8         reserved_2[0x8];
+	u8         reserved_at_40[0x8];
 	u8         qpn[0x18];
 
-	u8         reserved_3[0x20];
+	u8         reserved_at_60[0x20];
 
 	u8         opt_param_mask[0x20];
 
-	u8         reserved_4[0x20];
+	u8         reserved_at_a0[0x20];
 
 	struct mlx5_ifc_qpc_bits qpc;
 
-	u8         reserved_5[0x80];
+	u8         reserved_at_800[0x80];
 };
 
 struct mlx5_ifc_rtr2rts_qp_out_bits {
 	u8         status[0x8];
-	u8         reserved_0[0x18];
+	u8         reserved_at_8[0x18];
 
 	u8         syndrome[0x20];
 
-	u8         reserved_1[0x40];
+	u8         reserved_at_40[0x40];
 };
 
 struct mlx5_ifc_rtr2rts_qp_in_bits {
 	u8         opcode[0x10];
-	u8         reserved_0[0x10];
+	u8         reserved_at_10[0x10];
 
-	u8         reserved_1[0x10];
+	u8         reserved_at_20[0x10];
 	u8         op_mod[0x10];
 
-	u8         reserved_2[0x8];
+	u8         reserved_at_40[0x8];
 	u8         qpn[0x18];
 
-	u8         reserved_3[0x20];
+	u8         reserved_at_60[0x20];
 
 	u8         opt_param_mask[0x20];
 
-	u8         reserved_4[0x20];
+	u8         reserved_at_a0[0x20];
 
 	struct mlx5_ifc_qpc_bits qpc;
 
-	u8         reserved_5[0x80];
+	u8         reserved_at_800[0x80];
 };
 
 struct mlx5_ifc_rst2init_qp_out_bits {
 	u8         status[0x8];
-	u8         reserved_0[0x18];
+	u8         reserved_at_8[0x18];
 
 	u8         syndrome[0x20];
 
-	u8         reserved_1[0x40];
+	u8         reserved_at_40[0x40];
 };
 
 struct mlx5_ifc_rst2init_qp_in_bits {
 	u8         opcode[0x10];
-	u8         reserved_0[0x10];
+	u8         reserved_at_10[0x10];
 
-	u8         reserved_1[0x10];
+	u8         reserved_at_20[0x10];
 	u8         op_mod[0x10];
 
-	u8         reserved_2[0x8];
+	u8         reserved_at_40[0x8];
 	u8         qpn[0x18];
 
-	u8         reserved_3[0x20];
+	u8         reserved_at_60[0x20];
 
 	u8         opt_param_mask[0x20];
 
-	u8         reserved_4[0x20];
+	u8         reserved_at_a0[0x20];
 
 	struct mlx5_ifc_qpc_bits qpc;
 
-	u8         reserved_5[0x80];
+	u8         reserved_at_800[0x80];
 };
 
 struct mlx5_ifc_query_xrc_srq_out_bits {
 	u8         status[0x8];
-	u8         reserved_0[0x18];
+	u8         reserved_at_8[0x18];
 
 	u8         syndrome[0x20];
 
-	u8         reserved_1[0x40];
+	u8         reserved_at_40[0x40];
 
 	struct mlx5_ifc_xrc_srqc_bits xrc_srq_context_entry;
 
-	u8         reserved_2[0x600];
+	u8         reserved_at_280[0x600];
 
 	u8         pas[0][0x40];
 };
 
 struct mlx5_ifc_query_xrc_srq_in_bits {
 	u8         opcode[0x10];
-	u8         reserved_0[0x10];
+	u8         reserved_at_10[0x10];
 
-	u8         reserved_1[0x10];
+	u8         reserved_at_20[0x10];
 	u8         op_mod[0x10];
 
-	u8         reserved_2[0x8];
+	u8         reserved_at_40[0x8];
 	u8         xrc_srqn[0x18];
 
-	u8         reserved_3[0x20];
+	u8         reserved_at_60[0x20];
 };
 
 enum {
@@ -3049,13 +3049,13 @@ enum {
 
 struct mlx5_ifc_query_vport_state_out_bits {
 	u8         status[0x8];
-	u8         reserved_0[0x18];
+	u8         reserved_at_8[0x18];
 
 	u8         syndrome[0x20];
 
-	u8         reserved_1[0x20];
+	u8         reserved_at_40[0x20];
 
-	u8         reserved_2[0x18];
+	u8         reserved_at_60[0x18];
 	u8         admin_state[0x4];
 	u8         state[0x4];
 };
@@ -3067,25 +3067,25 @@ enum {
 
 struct mlx5_ifc_query_vport_state_in_bits {
 	u8         opcode[0x10];
-	u8         reserved_0[0x10];
+	u8         reserved_at_10[0x10];
 
-	u8         reserved_1[0x10];
+	u8         reserved_at_20[0x10];
 	u8         op_mod[0x10];
 
 	u8         other_vport[0x1];
-	u8         reserved_2[0xf];
+	u8         reserved_at_41[0xf];
 	u8         vport_number[0x10];
 
-	u8         reserved_3[0x20];
+	u8         reserved_at_60[0x20];
 };
 
 struct mlx5_ifc_query_vport_counter_out_bits {
 	u8         status[0x8];
-	u8         reserved_0[0x18];
+	u8         reserved_at_8[0x18];
 
 	u8         syndrome[0x20];
 
-	u8         reserved_1[0x40];
+	u8         reserved_at_40[0x40];
 
 	struct mlx5_ifc_traffic_counter_bits received_errors;
 
@@ -3111,7 +3111,7 @@ struct mlx5_ifc_query_vport_counter_out_bits {
 
 	struct mlx5_ifc_traffic_counter_bits transmitted_eth_multicast;
 
-	u8         reserved_2[0xa00];
+	u8         reserved_at_680[0xa00];
 };
 
 enum {
@@ -3120,328 +3120,328 @@ enum {
 
 struct mlx5_ifc_query_vport_counter_in_bits {
 	u8         opcode[0x10];
-	u8         reserved_0[0x10];
+	u8         reserved_at_10[0x10];
 
-	u8         reserved_1[0x10];
+	u8         reserved_at_20[0x10];
 	u8         op_mod[0x10];
 
 	u8         other_vport[0x1];
-	u8         reserved_2[0xf];
+	u8         reserved_at_41[0xf];
 	u8         vport_number[0x10];
 
-	u8         reserved_3[0x60];
+	u8         reserved_at_60[0x60];
 
 	u8         clear[0x1];
-	u8         reserved_4[0x1f];
+	u8         reserved_at_c1[0x1f];
 
-	u8         reserved_5[0x20];
+	u8         reserved_at_e0[0x20];
 };
 
 struct mlx5_ifc_query_tis_out_bits {
 	u8         status[0x8];
-	u8         reserved_0[0x18];
+	u8         reserved_at_8[0x18];
 
 	u8         syndrome[0x20];
 
-	u8         reserved_1[0x40];
+	u8         reserved_at_40[0x40];
 
 	struct mlx5_ifc_tisc_bits tis_context;
 };
 
 struct mlx5_ifc_query_tis_in_bits {
 	u8         opcode[0x10];
-	u8         reserved_0[0x10];
+	u8         reserved_at_10[0x10];
 
-	u8         reserved_1[0x10];
+	u8         reserved_at_20[0x10];
 	u8         op_mod[0x10];
 
-	u8         reserved_2[0x8];
+	u8         reserved_at_40[0x8];
 	u8         tisn[0x18];
 
-	u8         reserved_3[0x20];
+	u8         reserved_at_60[0x20];
 };
 
 struct mlx5_ifc_query_tir_out_bits {
 	u8         status[0x8];
-	u8         reserved_0[0x18];
+	u8         reserved_at_8[0x18];
 
 	u8         syndrome[0x20];
 
-	u8         reserved_1[0xc0];
+	u8         reserved_at_40[0xc0];
 
 	struct mlx5_ifc_tirc_bits tir_context;
 };
 
 struct mlx5_ifc_query_tir_in_bits {
 	u8         opcode[0x10];
-	u8         reserved_0[0x10];
+	u8         reserved_at_10[0x10];
 
-	u8         reserved_1[0x10];
+	u8         reserved_at_20[0x10];
 	u8         op_mod[0x10];
 
-	u8         reserved_2[0x8];
+	u8         reserved_at_40[0x8];
 	u8         tirn[0x18];
 
-	u8         reserved_3[0x20];
+	u8         reserved_at_60[0x20];
 };
 
 struct mlx5_ifc_query_srq_out_bits {
 	u8         status[0x8];
-	u8         reserved_0[0x18];
+	u8         reserved_at_8[0x18];
 
 	u8         syndrome[0x20];
 
-	u8         reserved_1[0x40];
+	u8         reserved_at_40[0x40];
 
 	struct mlx5_ifc_srqc_bits srq_context_entry;
 
-	u8         reserved_2[0x600];
+	u8         reserved_at_280[0x600];
 
 	u8         pas[0][0x40];
 };
 
 struct mlx5_ifc_query_srq_in_bits {
 	u8         opcode[0x10];
-	u8         reserved_0[0x10];
+	u8         reserved_at_10[0x10];
 
-	u8         reserved_1[0x10];
+	u8         reserved_at_20[0x10];
 	u8         op_mod[0x10];
 
-	u8         reserved_2[0x8];
+	u8         reserved_at_40[0x8];
 	u8         srqn[0x18];
 
-	u8         reserved_3[0x20];
+	u8         reserved_at_60[0x20];
 };
 
 struct mlx5_ifc_query_sq_out_bits {
 	u8         status[0x8];
-	u8         reserved_0[0x18];
+	u8         reserved_at_8[0x18];
 
 	u8         syndrome[0x20];
 
-	u8         reserved_1[0xc0];
+	u8         reserved_at_40[0xc0];
 
 	struct mlx5_ifc_sqc_bits sq_context;
 };
 
 struct mlx5_ifc_query_sq_in_bits {
 	u8         opcode[0x10];
-	u8         reserved_0[0x10];
+	u8         reserved_at_10[0x10];
 
-	u8         reserved_1[0x10];
+	u8         reserved_at_20[0x10];
 	u8         op_mod[0x10];
 
-	u8         reserved_2[0x8];
+	u8         reserved_at_40[0x8];
 	u8         sqn[0x18];
 
-	u8         reserved_3[0x20];
+	u8         reserved_at_60[0x20];
 };
 
 struct mlx5_ifc_query_special_contexts_out_bits {
 	u8         status[0x8];
-	u8         reserved_0[0x18];
+	u8         reserved_at_8[0x18];
 
 	u8         syndrome[0x20];
 
-	u8         reserved_1[0x20];
+	u8         reserved_at_40[0x20];
 
 	u8         resd_lkey[0x20];
 };
 
 struct mlx5_ifc_query_special_contexts_in_bits {
 	u8         opcode[0x10];
-	u8         reserved_0[0x10];
+	u8         reserved_at_10[0x10];
 
-	u8         reserved_1[0x10];
+	u8         reserved_at_20[0x10];
 	u8         op_mod[0x10];
 
-	u8         reserved_2[0x40];
+	u8         reserved_at_40[0x40];
 };
 
 struct mlx5_ifc_query_rqt_out_bits {
 	u8         status[0x8];
-	u8         reserved_0[0x18];
+	u8         reserved_at_8[0x18];
 
 	u8         syndrome[0x20];
 
-	u8         reserved_1[0xc0];
+	u8         reserved_at_40[0xc0];
 
 	struct mlx5_ifc_rqtc_bits rqt_context;
 };
 
 struct mlx5_ifc_query_rqt_in_bits {
 	u8         opcode[0x10];
-	u8         reserved_0[0x10];
+	u8         reserved_at_10[0x10];
 
-	u8         reserved_1[0x10];
+	u8         reserved_at_20[0x10];
 	u8         op_mod[0x10];
 
-	u8         reserved_2[0x8];
+	u8         reserved_at_40[0x8];
 	u8         rqtn[0x18];
 
-	u8         reserved_3[0x20];
+	u8         reserved_at_60[0x20];
 };
 
 struct mlx5_ifc_query_rq_out_bits {
 	u8         status[0x8];
-	u8         reserved_0[0x18];
+	u8         reserved_at_8[0x18];
 
 	u8         syndrome[0x20];
 
-	u8         reserved_1[0xc0];
+	u8         reserved_at_40[0xc0];
 
 	struct mlx5_ifc_rqc_bits rq_context;
 };
 
 struct mlx5_ifc_query_rq_in_bits {
 	u8         opcode[0x10];
-	u8         reserved_0[0x10];
+	u8         reserved_at_10[0x10];
 
-	u8         reserved_1[0x10];
+	u8         reserved_at_20[0x10];
 	u8         op_mod[0x10];
 
-	u8         reserved_2[0x8];
+	u8         reserved_at_40[0x8];
 	u8         rqn[0x18];
 
-	u8         reserved_3[0x20];
+	u8         reserved_at_60[0x20];
 };
 
 struct mlx5_ifc_query_roce_address_out_bits {
 	u8         status[0x8];
-	u8         reserved_0[0x18];
+	u8         reserved_at_8[0x18];
 
 	u8         syndrome[0x20];
 
-	u8         reserved_1[0x40];
+	u8         reserved_at_40[0x40];
 
 	struct mlx5_ifc_roce_addr_layout_bits roce_address;
 };
 
 struct mlx5_ifc_query_roce_address_in_bits {
 	u8         opcode[0x10];
-	u8         reserved_0[0x10];
+	u8         reserved_at_10[0x10];
 
-	u8         reserved_1[0x10];
+	u8         reserved_at_20[0x10];
 	u8         op_mod[0x10];
 
 	u8         roce_address_index[0x10];
-	u8         reserved_2[0x10];
+	u8         reserved_at_50[0x10];
 
-	u8         reserved_3[0x20];
+	u8         reserved_at_60[0x20];
 };
 
 struct mlx5_ifc_query_rmp_out_bits {
 	u8         status[0x8];
-	u8         reserved_0[0x18];
+	u8         reserved_at_8[0x18];
 
 	u8         syndrome[0x20];
 
-	u8         reserved_1[0xc0];
+	u8         reserved_at_40[0xc0];
 
 	struct mlx5_ifc_rmpc_bits rmp_context;
 };
 
 struct mlx5_ifc_query_rmp_in_bits {
 	u8         opcode[0x10];
-	u8         reserved_0[0x10];
+	u8         reserved_at_10[0x10];
 
-	u8         reserved_1[0x10];
+	u8         reserved_at_20[0x10];
 	u8         op_mod[0x10];
 
-	u8         reserved_2[0x8];
+	u8         reserved_at_40[0x8];
 	u8         rmpn[0x18];
 
-	u8         reserved_3[0x20];
+	u8         reserved_at_60[0x20];
 };
 
 struct mlx5_ifc_query_qp_out_bits {
 	u8         status[0x8];
-	u8         reserved_0[0x18];
+	u8         reserved_at_8[0x18];
 
 	u8         syndrome[0x20];
 
-	u8         reserved_1[0x40];
+	u8         reserved_at_40[0x40];
 
 	u8         opt_param_mask[0x20];
 
-	u8         reserved_2[0x20];
+	u8         reserved_at_a0[0x20];
 
 	struct mlx5_ifc_qpc_bits qpc;
 
-	u8         reserved_3[0x80];
+	u8         reserved_at_800[0x80];
 
 	u8         pas[0][0x40];
 };
 
 struct mlx5_ifc_query_qp_in_bits {
 	u8         opcode[0x10];
-	u8         reserved_0[0x10];
+	u8         reserved_at_10[0x10];
 
-	u8         reserved_1[0x10];
+	u8         reserved_at_20[0x10];
 	u8         op_mod[0x10];
 
-	u8         reserved_2[0x8];
+	u8         reserved_at_40[0x8];
 	u8         qpn[0x18];
 
-	u8         reserved_3[0x20];
+	u8         reserved_at_60[0x20];
 };
 
 struct mlx5_ifc_query_q_counter_out_bits {
 	u8         status[0x8];
-	u8         reserved_0[0x18];
+	u8         reserved_at_8[0x18];
 
 	u8         syndrome[0x20];
 
-	u8         reserved_1[0x40];
+	u8         reserved_at_40[0x40];
 
 	u8         rx_write_requests[0x20];
 
-	u8         reserved_2[0x20];
+	u8         reserved_at_a0[0x20];
 
 	u8         rx_read_requests[0x20];
 
-	u8         reserved_3[0x20];
+	u8         reserved_at_e0[0x20];
 
 	u8         rx_atomic_requests[0x20];
 
-	u8         reserved_4[0x20];
+	u8         reserved_at_120[0x20];
 
 	u8         rx_dct_connect[0x20];
 
-	u8         reserved_5[0x20];
+	u8         reserved_at_160[0x20];
 
 	u8         out_of_buffer[0x20];
 
-	u8         reserved_6[0x20];
+	u8         reserved_at_1a0[0x20];
 
 	u8         out_of_sequence[0x20];
 
-	u8         reserved_7[0x620];
+	u8         reserved_at_1e0[0x620];
 };
 
 struct mlx5_ifc_query_q_counter_in_bits {
 	u8         opcode[0x10];
-	u8         reserved_0[0x10];
+	u8         reserved_at_10[0x10];
 
-	u8         reserved_1[0x10];
+	u8         reserved_at_20[0x10];
 	u8         op_mod[0x10];
 
-	u8         reserved_2[0x80];
+	u8         reserved_at_40[0x80];
 
 	u8         clear[0x1];
-	u8         reserved_3[0x1f];
+	u8         reserved_at_c1[0x1f];
 
-	u8         reserved_4[0x18];
+	u8         reserved_at_e0[0x18];
 	u8         counter_set_id[0x8];
 };
 
 struct mlx5_ifc_query_pages_out_bits {
 	u8         status[0x8];
-	u8         reserved_0[0x18];
+	u8         reserved_at_8[0x18];
 
 	u8         syndrome[0x20];
 
-	u8         reserved_1[0x10];
+	u8         reserved_at_40[0x10];
 	u8         function_id[0x10];
 
 	u8         num_pages[0x20];
@@ -3455,55 +3455,55 @@ enum {
 
 struct mlx5_ifc_query_pages_in_bits {
 	u8         opcode[0x10];
-	u8         reserved_0[0x10];
+	u8         reserved_at_10[0x10];
 
-	u8         reserved_1[0x10];
+	u8         reserved_at_20[0x10];
 	u8         op_mod[0x10];
 
-	u8         reserved_2[0x10];
+	u8         reserved_at_40[0x10];
 	u8         function_id[0x10];
 
-	u8         reserved_3[0x20];
+	u8         reserved_at_60[0x20];
 };
 
 struct mlx5_ifc_query_nic_vport_context_out_bits {
 	u8         status[0x8];
-	u8         reserved_0[0x18];
+	u8         reserved_at_8[0x18];
 
 	u8         syndrome[0x20];
 
-	u8         reserved_1[0x40];
+	u8         reserved_at_40[0x40];
 
 	struct mlx5_ifc_nic_vport_context_bits nic_vport_context;
 };
 
 struct mlx5_ifc_query_nic_vport_context_in_bits {
 	u8         opcode[0x10];
-	u8         reserved_0[0x10];
+	u8         reserved_at_10[0x10];
 
-	u8         reserved_1[0x10];
+	u8         reserved_at_20[0x10];
 	u8         op_mod[0x10];
 
 	u8         other_vport[0x1];
-	u8         reserved_2[0xf];
+	u8         reserved_at_41[0xf];
 	u8         vport_number[0x10];
 
-	u8         reserved_3[0x5];
+	u8         reserved_at_60[0x5];
 	u8         allowed_list_type[0x3];
-	u8         reserved_4[0x18];
+	u8         reserved_at_68[0x18];
 };
 
 struct mlx5_ifc_query_mkey_out_bits {
 	u8         status[0x8];
-	u8         reserved_0[0x18];
+	u8         reserved_at_8[0x18];
 
 	u8         syndrome[0x20];
 
-	u8         reserved_1[0x40];
+	u8         reserved_at_40[0x40];
 
 	struct mlx5_ifc_mkc_bits memory_key_mkey_entry;
 
-	u8         reserved_2[0x600];
+	u8         reserved_at_280[0x600];
 
 	u8         bsf0_klm0_pas_mtt0_1[16][0x8];
 
@@ -3512,265 +3512,265 @@ struct mlx5_ifc_query_mkey_out_bits {
 
 struct mlx5_ifc_query_mkey_in_bits {
 	u8         opcode[0x10];
-	u8         reserved_0[0x10];
+	u8         reserved_at_10[0x10];
 
-	u8         reserved_1[0x10];
+	u8         reserved_at_20[0x10];
 	u8         op_mod[0x10];
 
-	u8         reserved_2[0x8];
+	u8         reserved_at_40[0x8];
 	u8         mkey_index[0x18];
 
 	u8         pg_access[0x1];
-	u8         reserved_3[0x1f];
+	u8         reserved_at_61[0x1f];
 };
 
 struct mlx5_ifc_query_mad_demux_out_bits {
 	u8         status[0x8];
-	u8         reserved_0[0x18];
+	u8         reserved_at_8[0x18];
 
 	u8         syndrome[0x20];
 
-	u8         reserved_1[0x40];
+	u8         reserved_at_40[0x40];
 
 	u8         mad_dumux_parameters_block[0x20];
 };
 
 struct mlx5_ifc_query_mad_demux_in_bits {
 	u8         opcode[0x10];
-	u8         reserved_0[0x10];
+	u8         reserved_at_10[0x10];
 
-	u8         reserved_1[0x10];
+	u8         reserved_at_20[0x10];
 	u8         op_mod[0x10];
 
-	u8         reserved_2[0x40];
+	u8         reserved_at_40[0x40];
 };
 
 struct mlx5_ifc_query_l2_table_entry_out_bits {
 	u8         status[0x8];
-	u8         reserved_0[0x18];
+	u8         reserved_at_8[0x18];
 
 	u8         syndrome[0x20];
 
-	u8         reserved_1[0xa0];
+	u8         reserved_at_40[0xa0];
 
-	u8         reserved_2[0x13];
+	u8         reserved_at_e0[0x13];
 	u8         vlan_valid[0x1];
 	u8         vlan[0xc];
 
 	struct mlx5_ifc_mac_address_layout_bits mac_address;
 
-	u8         reserved_3[0xc0];
+	u8         reserved_at_140[0xc0];
 };
 
 struct mlx5_ifc_query_l2_table_entry_in_bits {
 	u8         opcode[0x10];
-	u8         reserved_0[0x10];
+	u8         reserved_at_10[0x10];
 
-	u8         reserved_1[0x10];
+	u8         reserved_at_20[0x10];
 	u8         op_mod[0x10];
 
-	u8         reserved_2[0x60];
+	u8         reserved_at_40[0x60];
 
-	u8         reserved_3[0x8];
+	u8         reserved_at_a0[0x8];
 	u8         table_index[0x18];
 
-	u8         reserved_4[0x140];
+	u8         reserved_at_c0[0x140];
 };
 
 struct mlx5_ifc_query_issi_out_bits {
 	u8         status[0x8];
-	u8         reserved_0[0x18];
+	u8         reserved_at_8[0x18];
 
 	u8         syndrome[0x20];
 
-	u8         reserved_1[0x10];
+	u8         reserved_at_40[0x10];
 	u8         current_issi[0x10];
 
-	u8         reserved_2[0xa0];
+	u8         reserved_at_60[0xa0];
 
-	u8         supported_issi_reserved[76][0x8];
+	u8         reserved_at_100[76][0x8];
 	u8         supported_issi_dw0[0x20];
 };
 
 struct mlx5_ifc_query_issi_in_bits {
 	u8         opcode[0x10];
-	u8         reserved_0[0x10];
+	u8         reserved_at_10[0x10];
 
-	u8         reserved_1[0x10];
+	u8         reserved_at_20[0x10];
 	u8         op_mod[0x10];
 
-	u8         reserved_2[0x40];
+	u8         reserved_at_40[0x40];
 };
 
 struct mlx5_ifc_query_hca_vport_pkey_out_bits {
 	u8         status[0x8];
-	u8         reserved_0[0x18];
+	u8         reserved_at_8[0x18];
 
 	u8         syndrome[0x20];
 
-	u8         reserved_1[0x40];
+	u8         reserved_at_40[0x40];
 
 	struct mlx5_ifc_pkey_bits pkey[0];
 };
 
 struct mlx5_ifc_query_hca_vport_pkey_in_bits {
 	u8         opcode[0x10];
-	u8         reserved_0[0x10];
+	u8         reserved_at_10[0x10];
 
-	u8         reserved_1[0x10];
+	u8         reserved_at_20[0x10];
 	u8         op_mod[0x10];
 
 	u8         other_vport[0x1];
-	u8         reserved_2[0xb];
+	u8         reserved_at_41[0xb];
 	u8         port_num[0x4];
 	u8         vport_number[0x10];
 
-	u8         reserved_3[0x10];
+	u8         reserved_at_60[0x10];
 	u8         pkey_index[0x10];
 };
 
 struct mlx5_ifc_query_hca_vport_gid_out_bits {
 	u8         status[0x8];
-	u8         reserved_0[0x18];
+	u8         reserved_at_8[0x18];
 
 	u8         syndrome[0x20];
 
-	u8         reserved_1[0x20];
+	u8         reserved_at_40[0x20];
 
 	u8         gids_num[0x10];
-	u8         reserved_2[0x10];
+	u8         reserved_at_70[0x10];
 
 	struct mlx5_ifc_array128_auto_bits gid[0];
 };
 
 struct mlx5_ifc_query_hca_vport_gid_in_bits {
 	u8         opcode[0x10];
-	u8         reserved_0[0x10];
+	u8         reserved_at_10[0x10];
 
-	u8         reserved_1[0x10];
+	u8         reserved_at_20[0x10];
 	u8         op_mod[0x10];
 
 	u8         other_vport[0x1];
-	u8         reserved_2[0xb];
+	u8         reserved_at_41[0xb];
 	u8         port_num[0x4];
 	u8         vport_number[0x10];
 
-	u8         reserved_3[0x10];
+	u8         reserved_at_60[0x10];
 	u8         gid_index[0x10];
 };
 
 struct mlx5_ifc_query_hca_vport_context_out_bits {
 	u8         status[0x8];
-	u8         reserved_0[0x18];
+	u8         reserved_at_8[0x18];
 
 	u8         syndrome[0x20];
 
-	u8         reserved_1[0x40];
+	u8         reserved_at_40[0x40];
 
 	struct mlx5_ifc_hca_vport_context_bits hca_vport_context;
 };
 
 struct mlx5_ifc_query_hca_vport_context_in_bits {
 	u8         opcode[0x10];
-	u8         reserved_0[0x10];
+	u8         reserved_at_10[0x10];
 
-	u8         reserved_1[0x10];
+	u8         reserved_at_20[0x10];
 	u8         op_mod[0x10];
 
 	u8         other_vport[0x1];
-	u8         reserved_2[0xb];
+	u8         reserved_at_41[0xb];
 	u8         port_num[0x4];
 	u8         vport_number[0x10];
 
-	u8         reserved_3[0x20];
+	u8         reserved_at_60[0x20];
 };
 
 struct mlx5_ifc_query_hca_cap_out_bits {
 	u8         status[0x8];
-	u8         reserved_0[0x18];
+	u8         reserved_at_8[0x18];
 
 	u8         syndrome[0x20];
 
-	u8         reserved_1[0x40];
+	u8         reserved_at_40[0x40];
 
 	union mlx5_ifc_hca_cap_union_bits capability;
 };
 
 struct mlx5_ifc_query_hca_cap_in_bits {
 	u8         opcode[0x10];
-	u8         reserved_0[0x10];
+	u8         reserved_at_10[0x10];
 
-	u8         reserved_1[0x10];
+	u8         reserved_at_20[0x10];
 	u8         op_mod[0x10];
 
-	u8         reserved_2[0x40];
+	u8         reserved_at_40[0x40];
 };
 
 struct mlx5_ifc_query_flow_table_out_bits {
 	u8         status[0x8];
-	u8         reserved_0[0x18];
+	u8         reserved_at_8[0x18];
 
 	u8         syndrome[0x20];
 
-	u8         reserved_1[0x80];
+	u8         reserved_at_40[0x80];
 
-	u8         reserved_2[0x8];
+	u8         reserved_at_c0[0x8];
 	u8         level[0x8];
-	u8         reserved_3[0x8];
+	u8         reserved_at_d0[0x8];
 	u8         log_size[0x8];
 
-	u8         reserved_4[0x120];
+	u8         reserved_at_e0[0x120];
 };
 
 struct mlx5_ifc_query_flow_table_in_bits {
 	u8         opcode[0x10];
-	u8         reserved_0[0x10];
+	u8         reserved_at_10[0x10];
 
-	u8         reserved_1[0x10];
+	u8         reserved_at_20[0x10];
 	u8         op_mod[0x10];
 
-	u8         reserved_2[0x40];
+	u8         reserved_at_40[0x40];
 
 	u8         table_type[0x8];
-	u8         reserved_3[0x18];
+	u8         reserved_at_88[0x18];
 
-	u8         reserved_4[0x8];
+	u8         reserved_at_a0[0x8];
 	u8         table_id[0x18];
 
-	u8         reserved_5[0x140];
+	u8         reserved_at_c0[0x140];
 };
 
 struct mlx5_ifc_query_fte_out_bits {
 	u8         status[0x8];
-	u8         reserved_0[0x18];
+	u8         reserved_at_8[0x18];
 
 	u8         syndrome[0x20];
 
-	u8         reserved_1[0x1c0];
+	u8         reserved_at_40[0x1c0];
 
 	struct mlx5_ifc_flow_context_bits flow_context;
 };
 
 struct mlx5_ifc_query_fte_in_bits {
 	u8         opcode[0x10];
-	u8         reserved_0[0x10];
+	u8         reserved_at_10[0x10];
 
-	u8         reserved_1[0x10];
+	u8         reserved_at_20[0x10];
 	u8         op_mod[0x10];
 
-	u8         reserved_2[0x40];
+	u8         reserved_at_40[0x40];
 
 	u8         table_type[0x8];
-	u8         reserved_3[0x18];
+	u8         reserved_at_88[0x18];
 
-	u8         reserved_4[0x8];
+	u8         reserved_at_a0[0x8];
 	u8         table_id[0x18];
 
-	u8         reserved_5[0x40];
+	u8         reserved_at_c0[0x40];
 
 	u8         flow_index[0x20];
 
-	u8         reserved_6[0xe0];
+	u8         reserved_at_120[0xe0];
 };
 
 enum {
@@ -3781,84 +3781,84 @@ enum {
 
 struct mlx5_ifc_query_flow_group_out_bits {
 	u8         status[0x8];
-	u8         reserved_0[0x18];
+	u8         reserved_at_8[0x18];
 
 	u8         syndrome[0x20];
 
-	u8         reserved_1[0xa0];
+	u8         reserved_at_40[0xa0];
 
 	u8         start_flow_index[0x20];
 
-	u8         reserved_2[0x20];
+	u8         reserved_at_100[0x20];
 
 	u8         end_flow_index[0x20];
 
-	u8         reserved_3[0xa0];
+	u8         reserved_at_140[0xa0];
 
-	u8         reserved_4[0x18];
+	u8         reserved_at_1e0[0x18];
 	u8         match_criteria_enable[0x8];
 
 	struct mlx5_ifc_fte_match_param_bits match_criteria;
 
-	u8         reserved_5[0xe00];
+	u8         reserved_at_1200[0xe00];
 };
 
 struct mlx5_ifc_query_flow_group_in_bits {
 	u8         opcode[0x10];
-	u8         reserved_0[0x10];
+	u8         reserved_at_10[0x10];
 
-	u8         reserved_1[0x10];
+	u8         reserved_at_20[0x10];
 	u8         op_mod[0x10];
 
-	u8         reserved_2[0x40];
+	u8         reserved_at_40[0x40];
 
 	u8         table_type[0x8];
-	u8         reserved_3[0x18];
+	u8         reserved_at_88[0x18];
 
-	u8         reserved_4[0x8];
+	u8         reserved_at_a0[0x8];
 	u8         table_id[0x18];
 
 	u8         group_id[0x20];
 
-	u8         reserved_5[0x120];
+	u8         reserved_at_e0[0x120];
 };
 
 struct mlx5_ifc_query_esw_vport_context_out_bits {
 	u8         status[0x8];
-	u8         reserved_0[0x18];
+	u8         reserved_at_8[0x18];
 
 	u8         syndrome[0x20];
 
-	u8         reserved_1[0x40];
+	u8         reserved_at_40[0x40];
 
 	struct mlx5_ifc_esw_vport_context_bits esw_vport_context;
 };
 
 struct mlx5_ifc_query_esw_vport_context_in_bits {
 	u8         opcode[0x10];
-	u8         reserved_0[0x10];
+	u8         reserved_at_10[0x10];
 
-	u8         reserved_1[0x10];
+	u8         reserved_at_20[0x10];
 	u8         op_mod[0x10];
 
 	u8         other_vport[0x1];
-	u8         reserved_2[0xf];
+	u8         reserved_at_41[0xf];
 	u8         vport_number[0x10];
 
-	u8         reserved_3[0x20];
+	u8         reserved_at_60[0x20];
 };
 
 struct mlx5_ifc_modify_esw_vport_context_out_bits {
 	u8         status[0x8];
-	u8         reserved_0[0x18];
+	u8         reserved_at_8[0x18];
 
 	u8         syndrome[0x20];
 
-	u8         reserved_1[0x40];
+	u8         reserved_at_40[0x40];
 };
 
 struct mlx5_ifc_esw_vport_context_fields_select_bits {
-	u8         reserved[0x1c];
+	u8         reserved_at_0[0x1c];
 	u8         vport_cvlan_insert[0x1];
 	u8         vport_svlan_insert[0x1];
 	u8         vport_cvlan_strip[0x1];
@@ -3867,13 +3867,13 @@ struct mlx5_ifc_esw_vport_context_fields_select_bits {
 
 struct mlx5_ifc_modify_esw_vport_context_in_bits {
 	u8         opcode[0x10];
-	u8         reserved_0[0x10];
+	u8         reserved_at_10[0x10];
 
-	u8         reserved_1[0x10];
+	u8         reserved_at_20[0x10];
 	u8         op_mod[0x10];
 
 	u8         other_vport[0x1];
-	u8         reserved_2[0xf];
+	u8         reserved_at_41[0xf];
 	u8         vport_number[0x10];
 
 	struct mlx5_ifc_esw_vport_context_fields_select_bits field_select;
@@ -3883,124 +3883,124 @@ struct mlx5_ifc_modify_esw_vport_context_in_bits {
 
 struct mlx5_ifc_query_eq_out_bits {
 	u8         status[0x8];
-	u8         reserved_0[0x18];
+	u8         reserved_at_8[0x18];
 
 	u8         syndrome[0x20];
 
-	u8         reserved_1[0x40];
+	u8         reserved_at_40[0x40];
 
 	struct mlx5_ifc_eqc_bits eq_context_entry;
 
-	u8         reserved_2[0x40];
+	u8         reserved_at_280[0x40];
 
 	u8         event_bitmask[0x40];
 
-	u8         reserved_3[0x580];
+	u8         reserved_at_300[0x580];
 
 	u8         pas[0][0x40];
 };
 
 struct mlx5_ifc_query_eq_in_bits {
 	u8         opcode[0x10];
-	u8         reserved_0[0x10];
+	u8         reserved_at_10[0x10];
 
-	u8         reserved_1[0x10];
+	u8         reserved_at_20[0x10];
 	u8         op_mod[0x10];
 
-	u8         reserved_2[0x18];
+	u8         reserved_at_40[0x18];
 	u8         eq_number[0x8];
 
-	u8         reserved_3[0x20];
+	u8         reserved_at_60[0x20];
 };
 
 struct mlx5_ifc_query_dct_out_bits {
 	u8         status[0x8];
-	u8         reserved_0[0x18];
+	u8         reserved_at_8[0x18];
 
 	u8         syndrome[0x20];
 
-	u8         reserved_1[0x40];
+	u8         reserved_at_40[0x40];
 
 	struct mlx5_ifc_dctc_bits dct_context_entry;
 
-	u8         reserved_2[0x180];
+	u8         reserved_at_280[0x180];
 };
 
 struct mlx5_ifc_query_dct_in_bits {
 	u8         opcode[0x10];
-	u8         reserved_0[0x10];
+	u8         reserved_at_10[0x10];
 
-	u8         reserved_1[0x10];
+	u8         reserved_at_20[0x10];
 	u8         op_mod[0x10];
 
-	u8         reserved_2[0x8];
+	u8         reserved_at_40[0x8];
 	u8         dctn[0x18];
 
-	u8         reserved_3[0x20];
+	u8         reserved_at_60[0x20];
 };
 
 struct mlx5_ifc_query_cq_out_bits {
 	u8         status[0x8];
-	u8         reserved_0[0x18];
+	u8         reserved_at_8[0x18];
 
 	u8         syndrome[0x20];
 
-	u8         reserved_1[0x40];
+	u8         reserved_at_40[0x40];
 
 	struct mlx5_ifc_cqc_bits cq_context;
 
-	u8         reserved_2[0x600];
+	u8         reserved_at_280[0x600];
 
 	u8         pas[0][0x40];
 };
 
 struct mlx5_ifc_query_cq_in_bits {
 	u8         opcode[0x10];
-	u8         reserved_0[0x10];
+	u8         reserved_at_10[0x10];
 
-	u8         reserved_1[0x10];
+	u8         reserved_at_20[0x10];
 	u8         op_mod[0x10];
 
-	u8         reserved_2[0x8];
+	u8         reserved_at_40[0x8];
 	u8         cqn[0x18];
 
-	u8         reserved_3[0x20];
+	u8         reserved_at_60[0x20];
 };
 
 struct mlx5_ifc_query_cong_status_out_bits {
 	u8         status[0x8];
-	u8         reserved_0[0x18];
+	u8         reserved_at_8[0x18];
 
 	u8         syndrome[0x20];
 
-	u8         reserved_1[0x20];
+	u8         reserved_at_40[0x20];
 
 	u8         enable[0x1];
 	u8         tag_enable[0x1];
-	u8         reserved_2[0x1e];
+	u8         reserved_at_62[0x1e];
 };
 
 struct mlx5_ifc_query_cong_status_in_bits {
 	u8         opcode[0x10];
-	u8         reserved_0[0x10];
+	u8         reserved_at_10[0x10];
 
-	u8         reserved_1[0x10];
+	u8         reserved_at_20[0x10];
 	u8         op_mod[0x10];
 
-	u8         reserved_2[0x18];
+	u8         reserved_at_40[0x18];
 	u8         priority[0x4];
 	u8         cong_protocol[0x4];
 
-	u8         reserved_3[0x20];
+	u8         reserved_at_60[0x20];
 };
 
 struct mlx5_ifc_query_cong_statistics_out_bits {
 	u8         status[0x8];
-	u8         reserved_0[0x18];
+	u8         reserved_at_8[0x18];
 
 	u8         syndrome[0x20];
 
-	u8         reserved_1[0x40];
+	u8         reserved_at_40[0x40];
 
 	u8         cur_flows[0x20];
 
@@ -4014,7 +4014,7 @@ struct mlx5_ifc_query_cong_statistics_out_bits {
 
 	u8         cnp_handled_low[0x20];
 
-	u8         reserved_2[0x100];
+	u8         reserved_at_140[0x100];
 
 	u8         time_stamp_high[0x20];
 
@@ -4030,453 +4030,455 @@ struct mlx5_ifc_query_cong_statistics_out_bits {
 
 	u8         cnps_sent_low[0x20];
 
-	u8         reserved_3[0x560];
+	u8         reserved_at_320[0x560];
 };
 
 struct mlx5_ifc_query_cong_statistics_in_bits {
 	u8         opcode[0x10];
-	u8         reserved_0[0x10];
+	u8         reserved_at_10[0x10];
 
-	u8         reserved_1[0x10];
+	u8         reserved_at_20[0x10];
 	u8         op_mod[0x10];
 
 	u8         clear[0x1];
-	u8         reserved_2[0x1f];
+	u8         reserved_at_41[0x1f];
 
-	u8         reserved_3[0x20];
+	u8         reserved_at_60[0x20];
 };
 
 struct mlx5_ifc_query_cong_params_out_bits {
 	u8         status[0x8];
-	u8         reserved_0[0x18];
+	u8         reserved_at_8[0x18];
 
 	u8         syndrome[0x20];
 
-	u8         reserved_1[0x40];
+	u8         reserved_at_40[0x40];
 
 	union mlx5_ifc_cong_control_roce_ecn_auto_bits congestion_parameters;
 };
 
 struct mlx5_ifc_query_cong_params_in_bits {
 	u8         opcode[0x10];
-	u8         reserved_0[0x10];
+	u8         reserved_at_10[0x10];
 
-	u8         reserved_1[0x10];
+	u8         reserved_at_20[0x10];
 	u8         op_mod[0x10];
 
-	u8         reserved_2[0x1c];
+	u8         reserved_at_40[0x1c];
 	u8         cong_protocol[0x4];
 
-	u8         reserved_3[0x20];
+	u8         reserved_at_60[0x20];
 };
 
 struct mlx5_ifc_query_adapter_out_bits {
 	u8         status[0x8];
-	u8         reserved_0[0x18];
+	u8         reserved_at_8[0x18];
 
 	u8         syndrome[0x20];
 
-	u8         reserved_1[0x40];
+	u8         reserved_at_40[0x40];
 
 	struct mlx5_ifc_query_adapter_param_block_bits query_adapter_struct;
 };
 
 struct mlx5_ifc_query_adapter_in_bits {
 	u8         opcode[0x10];
-	u8         reserved_0[0x10];
+	u8         reserved_at_10[0x10];
 
-	u8         reserved_1[0x10];
+	u8         reserved_at_20[0x10];
 	u8         op_mod[0x10];
 
-	u8         reserved_2[0x40];
+	u8         reserved_at_40[0x40];
 };
 
 struct mlx5_ifc_qp_2rst_out_bits {
 	u8         status[0x8];
-	u8         reserved_0[0x18];
+	u8         reserved_at_8[0x18];
 
 	u8         syndrome[0x20];
 
-	u8         reserved_1[0x40];
+	u8         reserved_at_40[0x40];
 };
 
 struct mlx5_ifc_qp_2rst_in_bits {
 	u8         opcode[0x10];
-	u8         reserved_0[0x10];
+	u8         reserved_at_10[0x10];
 
-	u8         reserved_1[0x10];
+	u8         reserved_at_20[0x10];
 	u8         op_mod[0x10];
 
-	u8         reserved_2[0x8];
+	u8         reserved_at_40[0x8];
 	u8         qpn[0x18];
 
-	u8         reserved_3[0x20];
+	u8         reserved_at_60[0x20];
 };
 
 struct mlx5_ifc_qp_2err_out_bits {
 	u8         status[0x8];
-	u8         reserved_0[0x18];
+	u8         reserved_at_8[0x18];
 
 	u8         syndrome[0x20];
 
-	u8         reserved_1[0x40];
+	u8         reserved_at_40[0x40];
 };
 
 struct mlx5_ifc_qp_2err_in_bits {
 	u8         opcode[0x10];
-	u8         reserved_0[0x10];
+	u8         reserved_at_10[0x10];
 
-	u8         reserved_1[0x10];
+	u8         reserved_at_20[0x10];
 	u8         op_mod[0x10];
 
-	u8         reserved_2[0x8];
+	u8         reserved_at_40[0x8];
 	u8         qpn[0x18];
 
-	u8         reserved_3[0x20];
+	u8         reserved_at_60[0x20];
 };
 
 struct mlx5_ifc_page_fault_resume_out_bits {
 	u8         status[0x8];
-	u8         reserved_0[0x18];
+	u8         reserved_at_8[0x18];
 
 	u8         syndrome[0x20];
 
-	u8         reserved_1[0x40];
+	u8         reserved_at_40[0x40];
 };
 
 struct mlx5_ifc_page_fault_resume_in_bits {
 	u8         opcode[0x10];
-	u8         reserved_0[0x10];
+	u8         reserved_at_10[0x10];
 
-	u8         reserved_1[0x10];
+	u8         reserved_at_20[0x10];
 	u8         op_mod[0x10];
 
 	u8         error[0x1];
-	u8         reserved_2[0x4];
+	u8         reserved_at_41[0x4];
 	u8         rdma[0x1];
 	u8         read_write[0x1];
 	u8         req_res[0x1];
 	u8         qpn[0x18];
 
-	u8         reserved_3[0x20];
+	u8         reserved_at_60[0x20];
 };
 
 struct mlx5_ifc_nop_out_bits {
 	u8         status[0x8];
-	u8         reserved_0[0x18];
+	u8         reserved_at_8[0x18];
 
 	u8         syndrome[0x20];
 
-	u8         reserved_1[0x40];
+	u8         reserved_at_40[0x40];
 };
 
 struct mlx5_ifc_nop_in_bits {
 	u8         opcode[0x10];
-	u8         reserved_0[0x10];
+	u8         reserved_at_10[0x10];
 
-	u8         reserved_1[0x10];
+	u8         reserved_at_20[0x10];
 	u8         op_mod[0x10];
 
-	u8         reserved_2[0x40];
+	u8         reserved_at_40[0x40];
 };
 
 struct mlx5_ifc_modify_vport_state_out_bits {
 	u8         status[0x8];
-	u8         reserved_0[0x18];
+	u8         reserved_at_8[0x18];
 
 	u8         syndrome[0x20];
 
-	u8         reserved_1[0x40];
+	u8         reserved_at_40[0x40];
 };
 
 struct mlx5_ifc_modify_vport_state_in_bits {
 	u8         opcode[0x10];
-	u8         reserved_0[0x10];
+	u8         reserved_at_10[0x10];
 
-	u8         reserved_1[0x10];
+	u8         reserved_at_20[0x10];
 	u8         op_mod[0x10];
 
 	u8         other_vport[0x1];
-	u8         reserved_2[0xf];
+	u8         reserved_at_41[0xf];
 	u8         vport_number[0x10];
 
-	u8         reserved_3[0x18];
+	u8         reserved_at_60[0x18];
 	u8         admin_state[0x4];
-	u8         reserved_4[0x4];
+	u8         reserved_at_7c[0x4];
 };
 
 struct mlx5_ifc_modify_tis_out_bits {
 	u8         status[0x8];
-	u8         reserved_0[0x18];
+	u8         reserved_at_8[0x18];
 
 	u8         syndrome[0x20];
 
-	u8         reserved_1[0x40];
+	u8         reserved_at_40[0x40];
 };
 
 struct mlx5_ifc_modify_tis_bitmask_bits {
-	u8         reserved_0[0x20];
+	u8         reserved_at_0[0x20];
 
-	u8         reserved_1[0x1f];
+	u8         reserved_at_20[0x1f];
 	u8         prio[0x1];
 };
 
 struct mlx5_ifc_modify_tis_in_bits {
 	u8         opcode[0x10];
-	u8         reserved_0[0x10];
+	u8         reserved_at_10[0x10];
 
-	u8         reserved_1[0x10];
+	u8         reserved_at_20[0x10];
 	u8         op_mod[0x10];
 
-	u8         reserved_2[0x8];
+	u8         reserved_at_40[0x8];
 	u8         tisn[0x18];
 
-	u8         reserved_3[0x20];
+	u8         reserved_at_60[0x20];
 
 	struct mlx5_ifc_modify_tis_bitmask_bits bitmask;
 
-	u8         reserved_4[0x40];
+	u8         reserved_at_c0[0x40];
 
 	struct mlx5_ifc_tisc_bits ctx;
 };
 
 struct mlx5_ifc_modify_tir_bitmask_bits {
-	u8	   reserved_0[0x20];
+	u8	   reserved_at_0[0x20];
 
-	u8         reserved_1[0x1b];
+	u8         reserved_at_20[0x1b];
 	u8         self_lb_en[0x1];
-	u8         reserved_2[0x3];
+	u8         reserved_at_3c[0x1];
+	u8         hash[0x1];
+	u8         reserved_at_3e[0x1];
 	u8         lro[0x1];
 };
 
 struct mlx5_ifc_modify_tir_out_bits {
 	u8         status[0x8];
-	u8         reserved_0[0x18];
+	u8         reserved_at_8[0x18];
 
 	u8         syndrome[0x20];
 
-	u8         reserved_1[0x40];
+	u8         reserved_at_40[0x40];
 };
 
 struct mlx5_ifc_modify_tir_in_bits {
 	u8         opcode[0x10];
-	u8         reserved_0[0x10];
+	u8         reserved_at_10[0x10];
 
-	u8         reserved_1[0x10];
+	u8         reserved_at_20[0x10];
 	u8         op_mod[0x10];
 
-	u8         reserved_2[0x8];
+	u8         reserved_at_40[0x8];
 	u8         tirn[0x18];
 
-	u8         reserved_3[0x20];
+	u8         reserved_at_60[0x20];
 
 	struct mlx5_ifc_modify_tir_bitmask_bits bitmask;
 
-	u8         reserved_4[0x40];
+	u8         reserved_at_c0[0x40];
 
 	struct mlx5_ifc_tirc_bits ctx;
 };
 
 struct mlx5_ifc_modify_sq_out_bits {
 	u8         status[0x8];
-	u8         reserved_0[0x18];
+	u8         reserved_at_8[0x18];
 
 	u8         syndrome[0x20];
 
-	u8         reserved_1[0x40];
+	u8         reserved_at_40[0x40];
 };
 
 struct mlx5_ifc_modify_sq_in_bits {
 	u8         opcode[0x10];
-	u8         reserved_0[0x10];
+	u8         reserved_at_10[0x10];
 
-	u8         reserved_1[0x10];
+	u8         reserved_at_20[0x10];
 	u8         op_mod[0x10];
 
 	u8         sq_state[0x4];
-	u8         reserved_2[0x4];
+	u8         reserved_at_44[0x4];
 	u8         sqn[0x18];
 
-	u8         reserved_3[0x20];
+	u8         reserved_at_60[0x20];
 
 	u8         modify_bitmask[0x40];
 
-	u8         reserved_4[0x40];
+	u8         reserved_at_c0[0x40];
 
 	struct mlx5_ifc_sqc_bits ctx;
 };
 
 struct mlx5_ifc_modify_rqt_out_bits {
 	u8         status[0x8];
-	u8         reserved_0[0x18];
+	u8         reserved_at_8[0x18];
 
 	u8         syndrome[0x20];
 
-	u8         reserved_1[0x40];
+	u8         reserved_at_40[0x40];
 };
 
 struct mlx5_ifc_rqt_bitmask_bits {
-	u8	   reserved[0x20];
+	u8	   reserved_at_0[0x20];
 
-	u8         reserved1[0x1f];
+	u8         reserved_at_20[0x1f];
 	u8         rqn_list[0x1];
 };
 
 struct mlx5_ifc_modify_rqt_in_bits {
 	u8         opcode[0x10];
-	u8         reserved_0[0x10];
+	u8         reserved_at_10[0x10];
 
-	u8         reserved_1[0x10];
+	u8         reserved_at_20[0x10];
 	u8         op_mod[0x10];
 
-	u8         reserved_2[0x8];
+	u8         reserved_at_40[0x8];
 	u8         rqtn[0x18];
 
-	u8         reserved_3[0x20];
+	u8         reserved_at_60[0x20];
 
 	struct mlx5_ifc_rqt_bitmask_bits bitmask;
 
-	u8         reserved_4[0x40];
+	u8         reserved_at_c0[0x40];
 
 	struct mlx5_ifc_rqtc_bits ctx;
 };
 
 struct mlx5_ifc_modify_rq_out_bits {
 	u8         status[0x8];
-	u8         reserved_0[0x18];
+	u8         reserved_at_8[0x18];
 
 	u8         syndrome[0x20];
 
-	u8         reserved_1[0x40];
+	u8         reserved_at_40[0x40];
 };
 
 struct mlx5_ifc_modify_rq_in_bits {
 	u8         opcode[0x10];
-	u8         reserved_0[0x10];
+	u8         reserved_at_10[0x10];
 
-	u8         reserved_1[0x10];
+	u8         reserved_at_20[0x10];
 	u8         op_mod[0x10];
 
 	u8         rq_state[0x4];
-	u8         reserved_2[0x4];
+	u8         reserved_at_44[0x4];
 	u8         rqn[0x18];
 
-	u8         reserved_3[0x20];
+	u8         reserved_at_60[0x20];
 
 	u8         modify_bitmask[0x40];
 
-	u8         reserved_4[0x40];
+	u8         reserved_at_c0[0x40];
 
 	struct mlx5_ifc_rqc_bits ctx;
 };
 
 struct mlx5_ifc_modify_rmp_out_bits {
 	u8         status[0x8];
-	u8         reserved_0[0x18];
+	u8         reserved_at_8[0x18];
 
 	u8         syndrome[0x20];
 
-	u8         reserved_1[0x40];
+	u8         reserved_at_40[0x40];
 };
 
 struct mlx5_ifc_rmp_bitmask_bits {
-	u8	   reserved[0x20];
+	u8	   reserved_at_0[0x20];
 
-	u8         reserved1[0x1f];
+	u8         reserved_at_20[0x1f];
 	u8         lwm[0x1];
 };
 
 struct mlx5_ifc_modify_rmp_in_bits {
 	u8         opcode[0x10];
-	u8         reserved_0[0x10];
+	u8         reserved_at_10[0x10];
 
-	u8         reserved_1[0x10];
+	u8         reserved_at_20[0x10];
 	u8         op_mod[0x10];
 
 	u8         rmp_state[0x4];
-	u8         reserved_2[0x4];
+	u8         reserved_at_44[0x4];
 	u8         rmpn[0x18];
 
-	u8         reserved_3[0x20];
+	u8         reserved_at_60[0x20];
 
 	struct mlx5_ifc_rmp_bitmask_bits bitmask;
 
-	u8         reserved_4[0x40];
+	u8         reserved_at_c0[0x40];
 
 	struct mlx5_ifc_rmpc_bits ctx;
 };
 
 struct mlx5_ifc_modify_nic_vport_context_out_bits {
 	u8         status[0x8];
-	u8         reserved_0[0x18];
+	u8         reserved_at_8[0x18];
 
 	u8         syndrome[0x20];
 
-	u8         reserved_1[0x40];
+	u8         reserved_at_40[0x40];
 };
 
 struct mlx5_ifc_modify_nic_vport_field_select_bits {
-	u8         reserved_0[0x19];
+	u8         reserved_at_0[0x19];
 	u8         mtu[0x1];
 	u8         change_event[0x1];
 	u8         promisc[0x1];
 	u8         permanent_address[0x1];
 	u8         addresses_list[0x1];
 	u8         roce_en[0x1];
-	u8         reserved_1[0x1];
+	u8         reserved_at_1f[0x1];
 };
 
 struct mlx5_ifc_modify_nic_vport_context_in_bits {
 	u8         opcode[0x10];
-	u8         reserved_0[0x10];
+	u8         reserved_at_10[0x10];
 
-	u8         reserved_1[0x10];
+	u8         reserved_at_20[0x10];
 	u8         op_mod[0x10];
 
 	u8         other_vport[0x1];
-	u8         reserved_2[0xf];
+	u8         reserved_at_41[0xf];
 	u8         vport_number[0x10];
 
 	struct mlx5_ifc_modify_nic_vport_field_select_bits field_select;
 
-	u8         reserved_3[0x780];
+	u8         reserved_at_80[0x780];
 
 	struct mlx5_ifc_nic_vport_context_bits nic_vport_context;
 };
 
 struct mlx5_ifc_modify_hca_vport_context_out_bits {
 	u8         status[0x8];
-	u8         reserved_0[0x18];
+	u8         reserved_at_8[0x18];
 
 	u8         syndrome[0x20];
 
-	u8         reserved_1[0x40];
+	u8         reserved_at_40[0x40];
 };
 
 struct mlx5_ifc_modify_hca_vport_context_in_bits {
 	u8         opcode[0x10];
-	u8         reserved_0[0x10];
+	u8         reserved_at_10[0x10];
 
-	u8         reserved_1[0x10];
+	u8         reserved_at_20[0x10];
 	u8         op_mod[0x10];
 
 	u8         other_vport[0x1];
-	u8         reserved_2[0xb];
+	u8         reserved_at_41[0xb];
 	u8         port_num[0x4];
 	u8         vport_number[0x10];
 
-	u8         reserved_3[0x20];
+	u8         reserved_at_60[0x20];
 
 	struct mlx5_ifc_hca_vport_context_bits hca_vport_context;
 };
 
 struct mlx5_ifc_modify_cq_out_bits {
 	u8         status[0x8];
-	u8         reserved_0[0x18];
+	u8         reserved_at_8[0x18];
 
 	u8         syndrome[0x20];
 
-	u8         reserved_1[0x40];
+	u8         reserved_at_40[0x40];
 };
 
 enum {
@@ -4486,83 +4488,83 @@ enum {
 
 struct mlx5_ifc_modify_cq_in_bits {
 	u8         opcode[0x10];
-	u8         reserved_0[0x10];
+	u8         reserved_at_10[0x10];
 
-	u8         reserved_1[0x10];
+	u8         reserved_at_20[0x10];
 	u8         op_mod[0x10];
 
-	u8         reserved_2[0x8];
+	u8         reserved_at_40[0x8];
 	u8         cqn[0x18];
 
 	union mlx5_ifc_modify_field_select_resize_field_select_auto_bits modify_field_select_resize_field_select;
 
 	struct mlx5_ifc_cqc_bits cq_context;
 
-	u8         reserved_3[0x600];
+	u8         reserved_at_280[0x600];
 
 	u8         pas[0][0x40];
 };
 
 struct mlx5_ifc_modify_cong_status_out_bits {
 	u8         status[0x8];
-	u8         reserved_0[0x18];
+	u8         reserved_at_8[0x18];
 
 	u8         syndrome[0x20];
 
-	u8         reserved_1[0x40];
+	u8         reserved_at_40[0x40];
 };
 
 struct mlx5_ifc_modify_cong_status_in_bits {
 	u8         opcode[0x10];
-	u8         reserved_0[0x10];
+	u8         reserved_at_10[0x10];
 
-	u8         reserved_1[0x10];
+	u8         reserved_at_20[0x10];
 	u8         op_mod[0x10];
 
-	u8         reserved_2[0x18];
+	u8         reserved_at_40[0x18];
 	u8         priority[0x4];
 	u8         cong_protocol[0x4];
 
 	u8         enable[0x1];
 	u8         tag_enable[0x1];
-	u8         reserved_3[0x1e];
+	u8         reserved_at_62[0x1e];
 };
 
 struct mlx5_ifc_modify_cong_params_out_bits {
 	u8         status[0x8];
-	u8         reserved_0[0x18];
+	u8         reserved_at_8[0x18];
 
 	u8         syndrome[0x20];
 
-	u8         reserved_1[0x40];
+	u8         reserved_at_40[0x40];
 };
 
 struct mlx5_ifc_modify_cong_params_in_bits {
 	u8         opcode[0x10];
-	u8         reserved_0[0x10];
+	u8         reserved_at_10[0x10];
 
-	u8         reserved_1[0x10];
+	u8         reserved_at_20[0x10];
 	u8         op_mod[0x10];
 
-	u8         reserved_2[0x1c];
+	u8         reserved_at_40[0x1c];
 	u8         cong_protocol[0x4];
 
 	union mlx5_ifc_field_select_802_1_r_roce_auto_bits field_select;
 
-	u8         reserved_3[0x80];
+	u8         reserved_at_80[0x80];
 
 	union mlx5_ifc_cong_control_roce_ecn_auto_bits congestion_parameters;
 };
 
 struct mlx5_ifc_manage_pages_out_bits {
 	u8         status[0x8];
-	u8         reserved_0[0x18];
+	u8         reserved_at_8[0x18];
 
 	u8         syndrome[0x20];
 
 	u8         output_num_entries[0x20];
 
-	u8         reserved_1[0x20];
+	u8         reserved_at_60[0x20];
 
 	u8         pas[0][0x40];
 };
@@ -4575,12 +4577,12 @@ enum {
 
 struct mlx5_ifc_manage_pages_in_bits {
 	u8         opcode[0x10];
-	u8         reserved_0[0x10];
+	u8         reserved_at_10[0x10];
 
-	u8         reserved_1[0x10];
+	u8         reserved_at_20[0x10];
 	u8         op_mod[0x10];
 
-	u8         reserved_2[0x10];
+	u8         reserved_at_40[0x10];
 	u8         function_id[0x10];
 
 	u8         input_num_entries[0x20];
@@ -4590,117 +4592,117 @@ struct mlx5_ifc_manage_pages_in_bits {
 
 struct mlx5_ifc_mad_ifc_out_bits {
 	u8         status[0x8];
-	u8         reserved_0[0x18];
+	u8         reserved_at_8[0x18];
 
 	u8         syndrome[0x20];
 
-	u8         reserved_1[0x40];
+	u8         reserved_at_40[0x40];
 
 	u8         response_mad_packet[256][0x8];
 };
 
 struct mlx5_ifc_mad_ifc_in_bits {
 	u8         opcode[0x10];
-	u8         reserved_0[0x10];
+	u8         reserved_at_10[0x10];
 
-	u8         reserved_1[0x10];
+	u8         reserved_at_20[0x10];
 	u8         op_mod[0x10];
 
 	u8         remote_lid[0x10];
-	u8         reserved_2[0x8];
+	u8         reserved_at_50[0x8];
 	u8         port[0x8];
 
-	u8         reserved_3[0x20];
+	u8         reserved_at_60[0x20];
 
 	u8         mad[256][0x8];
 };
 
 struct mlx5_ifc_init_hca_out_bits {
 	u8         status[0x8];
-	u8         reserved_0[0x18];
+	u8         reserved_at_8[0x18];
 
 	u8         syndrome[0x20];
 
-	u8         reserved_1[0x40];
+	u8         reserved_at_40[0x40];
 };
 
 struct mlx5_ifc_init_hca_in_bits {
 	u8         opcode[0x10];
-	u8         reserved_0[0x10];
+	u8         reserved_at_10[0x10];
 
-	u8         reserved_1[0x10];
+	u8         reserved_at_20[0x10];
 	u8         op_mod[0x10];
 
-	u8         reserved_2[0x40];
+	u8         reserved_at_40[0x40];
 };
 
 struct mlx5_ifc_init2rtr_qp_out_bits {
 	u8         status[0x8];
-	u8         reserved_0[0x18];
+	u8         reserved_at_8[0x18];
 
 	u8         syndrome[0x20];
 
-	u8         reserved_1[0x40];
+	u8         reserved_at_40[0x40];
 };
 
 struct mlx5_ifc_init2rtr_qp_in_bits {
 	u8         opcode[0x10];
-	u8         reserved_0[0x10];
+	u8         reserved_at_10[0x10];
 
-	u8         reserved_1[0x10];
+	u8         reserved_at_20[0x10];
 	u8         op_mod[0x10];
 
-	u8         reserved_2[0x8];
+	u8         reserved_at_40[0x8];
 	u8         qpn[0x18];
 
-	u8         reserved_3[0x20];
+	u8         reserved_at_60[0x20];
 
 	u8         opt_param_mask[0x20];
 
-	u8         reserved_4[0x20];
+	u8         reserved_at_a0[0x20];
 
 	struct mlx5_ifc_qpc_bits qpc;
 
-	u8         reserved_5[0x80];
+	u8         reserved_at_800[0x80];
 };
 
 struct mlx5_ifc_init2init_qp_out_bits {
 	u8         status[0x8];
-	u8         reserved_0[0x18];
+	u8         reserved_at_8[0x18];
 
 	u8         syndrome[0x20];
 
-	u8         reserved_1[0x40];
+	u8         reserved_at_40[0x40];
 };
 
 struct mlx5_ifc_init2init_qp_in_bits {
 	u8         opcode[0x10];
-	u8         reserved_0[0x10];
+	u8         reserved_at_10[0x10];
 
-	u8         reserved_1[0x10];
+	u8         reserved_at_20[0x10];
 	u8         op_mod[0x10];
 
-	u8         reserved_2[0x8];
+	u8         reserved_at_40[0x8];
 	u8         qpn[0x18];
 
-	u8         reserved_3[0x20];
+	u8         reserved_at_60[0x20];
 
 	u8         opt_param_mask[0x20];
 
-	u8         reserved_4[0x20];
+	u8         reserved_at_a0[0x20];
 
 	struct mlx5_ifc_qpc_bits qpc;
 
-	u8         reserved_5[0x80];
+	u8         reserved_at_800[0x80];
 };
 
 struct mlx5_ifc_get_dropped_packet_log_out_bits {
 	u8         status[0x8];
-	u8         reserved_0[0x18];
+	u8         reserved_at_8[0x18];
 
 	u8         syndrome[0x20];
 
-	u8         reserved_1[0x40];
+	u8         reserved_at_40[0x40];
 
 	u8         packet_headers_log[128][0x8];
 
@@ -4709,1029 +4711,1029 @@ struct mlx5_ifc_get_dropped_packet_log_out_bits {
 
 struct mlx5_ifc_get_dropped_packet_log_in_bits {
 	u8         opcode[0x10];
-	u8         reserved_0[0x10];
+	u8         reserved_at_10[0x10];
 
-	u8         reserved_1[0x10];
+	u8         reserved_at_20[0x10];
 	u8         op_mod[0x10];
 
-	u8         reserved_2[0x40];
+	u8         reserved_at_40[0x40];
 };
 
 struct mlx5_ifc_gen_eqe_in_bits {
 	u8         opcode[0x10];
-	u8         reserved_0[0x10];
+	u8         reserved_at_10[0x10];
 
-	u8         reserved_1[0x10];
+	u8         reserved_at_20[0x10];
 	u8         op_mod[0x10];
 
-	u8         reserved_2[0x18];
+	u8         reserved_at_40[0x18];
 	u8         eq_number[0x8];
 
-	u8         reserved_3[0x20];
+	u8         reserved_at_60[0x20];
 
 	u8         eqe[64][0x8];
 };
 
 struct mlx5_ifc_gen_eq_out_bits {
 	u8         status[0x8];
-	u8         reserved_0[0x18];
+	u8         reserved_at_8[0x18];
 
 	u8         syndrome[0x20];
 
-	u8         reserved_1[0x40];
+	u8         reserved_at_40[0x40];
 };
 
 struct mlx5_ifc_enable_hca_out_bits {
 	u8         status[0x8];
-	u8         reserved_0[0x18];
+	u8         reserved_at_8[0x18];
 
 	u8         syndrome[0x20];
 
-	u8         reserved_1[0x20];
+	u8         reserved_at_40[0x20];
 };
 
 struct mlx5_ifc_enable_hca_in_bits {
 	u8         opcode[0x10];
-	u8         reserved_0[0x10];
+	u8         reserved_at_10[0x10];
 
-	u8         reserved_1[0x10];
+	u8         reserved_at_20[0x10];
 	u8         op_mod[0x10];
 
-	u8         reserved_2[0x10];
+	u8         reserved_at_40[0x10];
 	u8         function_id[0x10];
 
-	u8         reserved_3[0x20];
+	u8         reserved_at_60[0x20];
 };
 
 struct mlx5_ifc_drain_dct_out_bits {
 	u8         status[0x8];
-	u8         reserved_0[0x18];
+	u8         reserved_at_8[0x18];
 
 	u8         syndrome[0x20];
 
-	u8         reserved_1[0x40];
+	u8         reserved_at_40[0x40];
 };
 
 struct mlx5_ifc_drain_dct_in_bits {
 	u8         opcode[0x10];
-	u8         reserved_0[0x10];
+	u8         reserved_at_10[0x10];
 
-	u8         reserved_1[0x10];
+	u8         reserved_at_20[0x10];
 	u8         op_mod[0x10];
 
-	u8         reserved_2[0x8];
+	u8         reserved_at_40[0x8];
 	u8         dctn[0x18];
 
-	u8         reserved_3[0x20];
+	u8         reserved_at_60[0x20];
 };
 
 struct mlx5_ifc_disable_hca_out_bits {
 	u8         status[0x8];
-	u8         reserved_0[0x18];
+	u8         reserved_at_8[0x18];
 
 	u8         syndrome[0x20];
 
-	u8         reserved_1[0x20];
+	u8         reserved_at_40[0x20];
 };
 
 struct mlx5_ifc_disable_hca_in_bits {
 	u8         opcode[0x10];
-	u8         reserved_0[0x10];
+	u8         reserved_at_10[0x10];
 
-	u8         reserved_1[0x10];
+	u8         reserved_at_20[0x10];
 	u8         op_mod[0x10];
 
-	u8         reserved_2[0x10];
+	u8         reserved_at_40[0x10];
 	u8         function_id[0x10];
 
-	u8         reserved_3[0x20];
+	u8         reserved_at_60[0x20];
 };
 
 struct mlx5_ifc_detach_from_mcg_out_bits {
 	u8         status[0x8];
-	u8         reserved_0[0x18];
+	u8         reserved_at_8[0x18];
 
 	u8         syndrome[0x20];
 
-	u8         reserved_1[0x40];
+	u8         reserved_at_40[0x40];
 };
 
 struct mlx5_ifc_detach_from_mcg_in_bits {
 	u8         opcode[0x10];
-	u8         reserved_0[0x10];
+	u8         reserved_at_10[0x10];
 
-	u8         reserved_1[0x10];
+	u8         reserved_at_20[0x10];
 	u8         op_mod[0x10];
 
-	u8         reserved_2[0x8];
+	u8         reserved_at_40[0x8];
 	u8         qpn[0x18];
 
-	u8         reserved_3[0x20];
+	u8         reserved_at_60[0x20];
 
 	u8         multicast_gid[16][0x8];
 };
 
 struct mlx5_ifc_destroy_xrc_srq_out_bits {
 	u8         status[0x8];
-	u8         reserved_0[0x18];
+	u8         reserved_at_8[0x18];
 
 	u8         syndrome[0x20];
 
-	u8         reserved_1[0x40];
+	u8         reserved_at_40[0x40];
 };
 
 struct mlx5_ifc_destroy_xrc_srq_in_bits {
 	u8         opcode[0x10];
-	u8         reserved_0[0x10];
+	u8         reserved_at_10[0x10];
 
-	u8         reserved_1[0x10];
+	u8         reserved_at_20[0x10];
 	u8         op_mod[0x10];
 
-	u8         reserved_2[0x8];
+	u8         reserved_at_40[0x8];
 	u8         xrc_srqn[0x18];
 
-	u8         reserved_3[0x20];
+	u8         reserved_at_60[0x20];
 };
 
 struct mlx5_ifc_destroy_tis_out_bits {
 	u8         status[0x8];
-	u8         reserved_0[0x18];
+	u8         reserved_at_8[0x18];
 
 	u8         syndrome[0x20];
 
-	u8         reserved_1[0x40];
+	u8         reserved_at_40[0x40];
 };
 
 struct mlx5_ifc_destroy_tis_in_bits {
 	u8         opcode[0x10];
-	u8         reserved_0[0x10];
+	u8         reserved_at_10[0x10];
 
-	u8         reserved_1[0x10];
+	u8         reserved_at_20[0x10];
 	u8         op_mod[0x10];
 
-	u8         reserved_2[0x8];
+	u8         reserved_at_40[0x8];
 	u8         tisn[0x18];
 
-	u8         reserved_3[0x20];
+	u8         reserved_at_60[0x20];
 };
 
 struct mlx5_ifc_destroy_tir_out_bits {
 	u8         status[0x8];
-	u8         reserved_0[0x18];
+	u8         reserved_at_8[0x18];
 
 	u8         syndrome[0x20];
 
-	u8         reserved_1[0x40];
+	u8         reserved_at_40[0x40];
 };
 
 struct mlx5_ifc_destroy_tir_in_bits {
 	u8         opcode[0x10];
-	u8         reserved_0[0x10];
+	u8         reserved_at_10[0x10];
 
-	u8         reserved_1[0x10];
+	u8         reserved_at_20[0x10];
 	u8         op_mod[0x10];
 
-	u8         reserved_2[0x8];
+	u8         reserved_at_40[0x8];
 	u8         tirn[0x18];
 
-	u8         reserved_3[0x20];
+	u8         reserved_at_60[0x20];
 };
 
 struct mlx5_ifc_destroy_srq_out_bits {
 	u8         status[0x8];
-	u8         reserved_0[0x18];
+	u8         reserved_at_8[0x18];
 
 	u8         syndrome[0x20];
 
-	u8         reserved_1[0x40];
+	u8         reserved_at_40[0x40];
 };
 
 struct mlx5_ifc_destroy_srq_in_bits {
 	u8         opcode[0x10];
-	u8         reserved_0[0x10];
+	u8         reserved_at_10[0x10];
 
-	u8         reserved_1[0x10];
+	u8         reserved_at_20[0x10];
 	u8         op_mod[0x10];
 
-	u8         reserved_2[0x8];
+	u8         reserved_at_40[0x8];
 	u8         srqn[0x18];
 
-	u8         reserved_3[0x20];
+	u8         reserved_at_60[0x20];
 };
 
 struct mlx5_ifc_destroy_sq_out_bits {
 	u8         status[0x8];
-	u8         reserved_0[0x18];
+	u8         reserved_at_8[0x18];
 
 	u8         syndrome[0x20];
 
-	u8         reserved_1[0x40];
+	u8         reserved_at_40[0x40];
 };
 
 struct mlx5_ifc_destroy_sq_in_bits {
 	u8         opcode[0x10];
-	u8         reserved_0[0x10];
+	u8         reserved_at_10[0x10];
 
-	u8         reserved_1[0x10];
+	u8         reserved_at_20[0x10];
 	u8         op_mod[0x10];
 
-	u8         reserved_2[0x8];
+	u8         reserved_at_40[0x8];
 	u8         sqn[0x18];
 
-	u8         reserved_3[0x20];
+	u8         reserved_at_60[0x20];
 };
 
 struct mlx5_ifc_destroy_rqt_out_bits {
 	u8         status[0x8];
-	u8         reserved_0[0x18];
+	u8         reserved_at_8[0x18];
 
 	u8         syndrome[0x20];
 
-	u8         reserved_1[0x40];
+	u8         reserved_at_40[0x40];
 };
 
 struct mlx5_ifc_destroy_rqt_in_bits {
 	u8         opcode[0x10];
-	u8         reserved_0[0x10];
+	u8         reserved_at_10[0x10];
 
-	u8         reserved_1[0x10];
+	u8         reserved_at_20[0x10];
 	u8         op_mod[0x10];
 
-	u8         reserved_2[0x8];
+	u8         reserved_at_40[0x8];
 	u8         rqtn[0x18];
 
-	u8         reserved_3[0x20];
+	u8         reserved_at_60[0x20];
 };
 
 struct mlx5_ifc_destroy_rq_out_bits {
 	u8         status[0x8];
-	u8         reserved_0[0x18];
+	u8         reserved_at_8[0x18];
 
 	u8         syndrome[0x20];
 
-	u8         reserved_1[0x40];
+	u8         reserved_at_40[0x40];
 };
 
 struct mlx5_ifc_destroy_rq_in_bits {
 	u8         opcode[0x10];
-	u8         reserved_0[0x10];
+	u8         reserved_at_10[0x10];
 
-	u8         reserved_1[0x10];
+	u8         reserved_at_20[0x10];
 	u8         op_mod[0x10];
 
-	u8         reserved_2[0x8];
+	u8         reserved_at_40[0x8];
 	u8         rqn[0x18];
 
-	u8         reserved_3[0x20];
+	u8         reserved_at_60[0x20];
 };
 
 struct mlx5_ifc_destroy_rmp_out_bits {
 	u8         status[0x8];
-	u8         reserved_0[0x18];
+	u8         reserved_at_8[0x18];
 
 	u8         syndrome[0x20];
 
-	u8         reserved_1[0x40];
+	u8         reserved_at_40[0x40];
 };
 
 struct mlx5_ifc_destroy_rmp_in_bits {
 	u8         opcode[0x10];
-	u8         reserved_0[0x10];
+	u8         reserved_at_10[0x10];
 
-	u8         reserved_1[0x10];
+	u8         reserved_at_20[0x10];
 	u8         op_mod[0x10];
 
-	u8         reserved_2[0x8];
+	u8         reserved_at_40[0x8];
 	u8         rmpn[0x18];
 
-	u8         reserved_3[0x20];
+	u8         reserved_at_60[0x20];
 };
 
 struct mlx5_ifc_destroy_qp_out_bits {
 	u8         status[0x8];
-	u8         reserved_0[0x18];
+	u8         reserved_at_8[0x18];
 
 	u8         syndrome[0x20];
 
-	u8         reserved_1[0x40];
+	u8         reserved_at_40[0x40];
 };
 
 struct mlx5_ifc_destroy_qp_in_bits {
 	u8         opcode[0x10];
-	u8         reserved_0[0x10];
+	u8         reserved_at_10[0x10];
 
-	u8         reserved_1[0x10];
+	u8         reserved_at_20[0x10];
 	u8         op_mod[0x10];
 
-	u8         reserved_2[0x8];
+	u8         reserved_at_40[0x8];
 	u8         qpn[0x18];
 
-	u8         reserved_3[0x20];
+	u8         reserved_at_60[0x20];
 };
 
 struct mlx5_ifc_destroy_psv_out_bits {
 	u8         status[0x8];
-	u8         reserved_0[0x18];
+	u8         reserved_at_8[0x18];
 
 	u8         syndrome[0x20];
 
-	u8         reserved_1[0x40];
+	u8         reserved_at_40[0x40];
 };
 
 struct mlx5_ifc_destroy_psv_in_bits {
 	u8         opcode[0x10];
-	u8         reserved_0[0x10];
+	u8         reserved_at_10[0x10];
 
-	u8         reserved_1[0x10];
+	u8         reserved_at_20[0x10];
 	u8         op_mod[0x10];
 
-	u8         reserved_2[0x8];
+	u8         reserved_at_40[0x8];
 	u8         psvn[0x18];
 
-	u8         reserved_3[0x20];
+	u8         reserved_at_60[0x20];
 };
 
 struct mlx5_ifc_destroy_mkey_out_bits {
 	u8         status[0x8];
-	u8         reserved_0[0x18];
+	u8         reserved_at_8[0x18];
 
 	u8         syndrome[0x20];
 
-	u8         reserved_1[0x40];
+	u8         reserved_at_40[0x40];
 };
 
 struct mlx5_ifc_destroy_mkey_in_bits {
 	u8         opcode[0x10];
-	u8         reserved_0[0x10];
+	u8         reserved_at_10[0x10];
 
-	u8         reserved_1[0x10];
+	u8         reserved_at_20[0x10];
 	u8         op_mod[0x10];
 
-	u8         reserved_2[0x8];
+	u8         reserved_at_40[0x8];
 	u8         mkey_index[0x18];
 
-	u8         reserved_3[0x20];
+	u8         reserved_at_60[0x20];
 };
 
 struct mlx5_ifc_destroy_flow_table_out_bits {
 	u8         status[0x8];
-	u8         reserved_0[0x18];
+	u8         reserved_at_8[0x18];
 
 	u8         syndrome[0x20];
 
-	u8         reserved_1[0x40];
+	u8         reserved_at_40[0x40];
 };
 
 struct mlx5_ifc_destroy_flow_table_in_bits {
 	u8         opcode[0x10];
-	u8         reserved_0[0x10];
+	u8         reserved_at_10[0x10];
 
-	u8         reserved_1[0x10];
+	u8         reserved_at_20[0x10];
 	u8         op_mod[0x10];
 
-	u8         reserved_2[0x40];
+	u8         reserved_at_40[0x40];
 
 	u8         table_type[0x8];
-	u8         reserved_3[0x18];
+	u8         reserved_at_88[0x18];
 
-	u8         reserved_4[0x8];
+	u8         reserved_at_a0[0x8];
 	u8         table_id[0x18];
 
-	u8         reserved_5[0x140];
+	u8         reserved_at_c0[0x140];
 };
 
 struct mlx5_ifc_destroy_flow_group_out_bits {
 	u8         status[0x8];
-	u8         reserved_0[0x18];
+	u8         reserved_at_8[0x18];
 
 	u8         syndrome[0x20];
 
-	u8         reserved_1[0x40];
+	u8         reserved_at_40[0x40];
 };
 
 struct mlx5_ifc_destroy_flow_group_in_bits {
 	u8         opcode[0x10];
-	u8         reserved_0[0x10];
+	u8         reserved_at_10[0x10];
 
-	u8         reserved_1[0x10];
+	u8         reserved_at_20[0x10];
 	u8         op_mod[0x10];
 
-	u8         reserved_2[0x40];
+	u8         reserved_at_40[0x40];
 
 	u8         table_type[0x8];
-	u8         reserved_3[0x18];
+	u8         reserved_at_88[0x18];
 
-	u8         reserved_4[0x8];
+	u8         reserved_at_a0[0x8];
 	u8         table_id[0x18];
 
 	u8         group_id[0x20];
 
-	u8         reserved_5[0x120];
+	u8         reserved_at_e0[0x120];
 };
 
 struct mlx5_ifc_destroy_eq_out_bits {
 	u8         status[0x8];
-	u8         reserved_0[0x18];
+	u8         reserved_at_8[0x18];
 
 	u8         syndrome[0x20];
 
-	u8         reserved_1[0x40];
+	u8         reserved_at_40[0x40];
 };
 
 struct mlx5_ifc_destroy_eq_in_bits {
 	u8         opcode[0x10];
-	u8         reserved_0[0x10];
+	u8         reserved_at_10[0x10];
 
-	u8         reserved_1[0x10];
+	u8         reserved_at_20[0x10];
 	u8         op_mod[0x10];
 
-	u8         reserved_2[0x18];
+	u8         reserved_at_40[0x18];
 	u8         eq_number[0x8];
 
-	u8         reserved_3[0x20];
+	u8         reserved_at_60[0x20];
 };
 
 struct mlx5_ifc_destroy_dct_out_bits {
 	u8         status[0x8];
-	u8         reserved_0[0x18];
+	u8         reserved_at_8[0x18];
 
 	u8         syndrome[0x20];
 
-	u8         reserved_1[0x40];
+	u8         reserved_at_40[0x40];
 };
 
 struct mlx5_ifc_destroy_dct_in_bits {
 	u8         opcode[0x10];
-	u8         reserved_0[0x10];
+	u8         reserved_at_10[0x10];
 
-	u8         reserved_1[0x10];
+	u8         reserved_at_20[0x10];
 	u8         op_mod[0x10];
 
-	u8         reserved_2[0x8];
+	u8         reserved_at_40[0x8];
 	u8         dctn[0x18];
 
-	u8         reserved_3[0x20];
+	u8         reserved_at_60[0x20];
 };
 
 struct mlx5_ifc_destroy_cq_out_bits {
 	u8         status[0x8];
-	u8         reserved_0[0x18];
+	u8         reserved_at_8[0x18];
 
 	u8         syndrome[0x20];
 
-	u8         reserved_1[0x40];
+	u8         reserved_at_40[0x40];
 };
 
 struct mlx5_ifc_destroy_cq_in_bits {
 	u8         opcode[0x10];
-	u8         reserved_0[0x10];
+	u8         reserved_at_10[0x10];
 
-	u8         reserved_1[0x10];
+	u8         reserved_at_20[0x10];
 	u8         op_mod[0x10];
 
-	u8         reserved_2[0x8];
+	u8         reserved_at_40[0x8];
 	u8         cqn[0x18];
 
-	u8         reserved_3[0x20];
+	u8         reserved_at_60[0x20];
 };
 
 struct mlx5_ifc_delete_vxlan_udp_dport_out_bits {
 	u8         status[0x8];
-	u8         reserved_0[0x18];
+	u8         reserved_at_8[0x18];
 
 	u8         syndrome[0x20];
 
-	u8         reserved_1[0x40];
+	u8         reserved_at_40[0x40];
 };
 
 struct mlx5_ifc_delete_vxlan_udp_dport_in_bits {
 	u8         opcode[0x10];
-	u8         reserved_0[0x10];
+	u8         reserved_at_10[0x10];
 
-	u8         reserved_1[0x10];
+	u8         reserved_at_20[0x10];
 	u8         op_mod[0x10];
 
-	u8         reserved_2[0x20];
+	u8         reserved_at_40[0x20];
 
-	u8         reserved_3[0x10];
+	u8         reserved_at_60[0x10];
 	u8         vxlan_udp_port[0x10];
 };
 
 struct mlx5_ifc_delete_l2_table_entry_out_bits {
 	u8         status[0x8];
-	u8         reserved_0[0x18];
+	u8         reserved_at_8[0x18];
 
 	u8         syndrome[0x20];
 
-	u8         reserved_1[0x40];
+	u8         reserved_at_40[0x40];
 };
 
 struct mlx5_ifc_delete_l2_table_entry_in_bits {
 	u8         opcode[0x10];
-	u8         reserved_0[0x10];
+	u8         reserved_at_10[0x10];
 
-	u8         reserved_1[0x10];
+	u8         reserved_at_20[0x10];
 	u8         op_mod[0x10];
 
-	u8         reserved_2[0x60];
+	u8         reserved_at_40[0x60];
 
-	u8         reserved_3[0x8];
+	u8         reserved_at_a0[0x8];
 	u8         table_index[0x18];
 
-	u8         reserved_4[0x140];
+	u8         reserved_at_c0[0x140];
 };
 
 struct mlx5_ifc_delete_fte_out_bits {
 	u8         status[0x8];
-	u8         reserved_0[0x18];
+	u8         reserved_at_8[0x18];
 
 	u8         syndrome[0x20];
 
-	u8         reserved_1[0x40];
+	u8         reserved_at_40[0x40];
 };
 
 struct mlx5_ifc_delete_fte_in_bits {
 	u8         opcode[0x10];
-	u8         reserved_0[0x10];
+	u8         reserved_at_10[0x10];
 
-	u8         reserved_1[0x10];
+	u8         reserved_at_20[0x10];
 	u8         op_mod[0x10];
 
-	u8         reserved_2[0x40];
+	u8         reserved_at_40[0x40];
 
 	u8         table_type[0x8];
-	u8         reserved_3[0x18];
+	u8         reserved_at_88[0x18];
 
-	u8         reserved_4[0x8];
+	u8         reserved_at_a0[0x8];
 	u8         table_id[0x18];
 
-	u8         reserved_5[0x40];
+	u8         reserved_at_c0[0x40];
 
 	u8         flow_index[0x20];
 
-	u8         reserved_6[0xe0];
+	u8         reserved_at_120[0xe0];
 };
 
 struct mlx5_ifc_dealloc_xrcd_out_bits {
 	u8         status[0x8];
-	u8         reserved_0[0x18];
+	u8         reserved_at_8[0x18];
 
 	u8         syndrome[0x20];
 
-	u8         reserved_1[0x40];
+	u8         reserved_at_40[0x40];
 };
 
 struct mlx5_ifc_dealloc_xrcd_in_bits {
 	u8         opcode[0x10];
-	u8         reserved_0[0x10];
+	u8         reserved_at_10[0x10];
 
-	u8         reserved_1[0x10];
+	u8         reserved_at_20[0x10];
 	u8         op_mod[0x10];
 
-	u8         reserved_2[0x8];
+	u8         reserved_at_40[0x8];
 	u8         xrcd[0x18];
 
-	u8         reserved_3[0x20];
+	u8         reserved_at_60[0x20];
 };
 
 struct mlx5_ifc_dealloc_uar_out_bits {
 	u8         status[0x8];
-	u8         reserved_0[0x18];
+	u8         reserved_at_8[0x18];
 
 	u8         syndrome[0x20];
 
-	u8         reserved_1[0x40];
+	u8         reserved_at_40[0x40];
 };
 
 struct mlx5_ifc_dealloc_uar_in_bits {
 	u8         opcode[0x10];
-	u8         reserved_0[0x10];
+	u8         reserved_at_10[0x10];
 
-	u8         reserved_1[0x10];
+	u8         reserved_at_20[0x10];
 	u8         op_mod[0x10];
 
-	u8         reserved_2[0x8];
+	u8         reserved_at_40[0x8];
 	u8         uar[0x18];
 
-	u8         reserved_3[0x20];
+	u8         reserved_at_60[0x20];
 };
 
 struct mlx5_ifc_dealloc_transport_domain_out_bits {
 	u8         status[0x8];
-	u8         reserved_0[0x18];
+	u8         reserved_at_8[0x18];
 
 	u8         syndrome[0x20];
 
-	u8         reserved_1[0x40];
+	u8         reserved_at_40[0x40];
 };
 
 struct mlx5_ifc_dealloc_transport_domain_in_bits {
 	u8         opcode[0x10];
-	u8         reserved_0[0x10];
+	u8         reserved_at_10[0x10];
 
-	u8         reserved_1[0x10];
+	u8         reserved_at_20[0x10];
 	u8         op_mod[0x10];
 
-	u8         reserved_2[0x8];
+	u8         reserved_at_40[0x8];
 	u8         transport_domain[0x18];
 
-	u8         reserved_3[0x20];
+	u8         reserved_at_60[0x20];
 };
 
 struct mlx5_ifc_dealloc_q_counter_out_bits {
 	u8         status[0x8];
-	u8         reserved_0[0x18];
+	u8         reserved_at_8[0x18];
 
 	u8         syndrome[0x20];
 
-	u8         reserved_1[0x40];
+	u8         reserved_at_40[0x40];
 };
 
 struct mlx5_ifc_dealloc_q_counter_in_bits {
 	u8         opcode[0x10];
-	u8         reserved_0[0x10];
+	u8         reserved_at_10[0x10];
 
-	u8         reserved_1[0x10];
+	u8         reserved_at_20[0x10];
 	u8         op_mod[0x10];
 
-	u8         reserved_2[0x18];
+	u8         reserved_at_40[0x18];
 	u8         counter_set_id[0x8];
 
-	u8         reserved_3[0x20];
+	u8         reserved_at_60[0x20];
 };
 
 struct mlx5_ifc_dealloc_pd_out_bits {
 	u8         status[0x8];
-	u8         reserved_0[0x18];
+	u8         reserved_at_8[0x18];
 
 	u8         syndrome[0x20];
 
-	u8         reserved_1[0x40];
+	u8         reserved_at_40[0x40];
 };
 
 struct mlx5_ifc_dealloc_pd_in_bits {
 	u8         opcode[0x10];
-	u8         reserved_0[0x10];
+	u8         reserved_at_10[0x10];
 
-	u8         reserved_1[0x10];
+	u8         reserved_at_20[0x10];
 	u8         op_mod[0x10];
 
-	u8         reserved_2[0x8];
+	u8         reserved_at_40[0x8];
 	u8         pd[0x18];
 
-	u8         reserved_3[0x20];
+	u8         reserved_at_60[0x20];
 };
 
 struct mlx5_ifc_create_xrc_srq_out_bits {
 	u8         status[0x8];
-	u8         reserved_0[0x18];
+	u8         reserved_at_8[0x18];
 
 	u8         syndrome[0x20];
 
-	u8         reserved_1[0x8];
+	u8         reserved_at_40[0x8];
 	u8         xrc_srqn[0x18];
 
-	u8         reserved_2[0x20];
+	u8         reserved_at_60[0x20];
 };
 
 struct mlx5_ifc_create_xrc_srq_in_bits {
 	u8         opcode[0x10];
-	u8         reserved_0[0x10];
+	u8         reserved_at_10[0x10];
 
-	u8         reserved_1[0x10];
+	u8         reserved_at_20[0x10];
 	u8         op_mod[0x10];
 
-	u8         reserved_2[0x40];
+	u8         reserved_at_40[0x40];
 
 	struct mlx5_ifc_xrc_srqc_bits xrc_srq_context_entry;
 
-	u8         reserved_3[0x600];
+	u8         reserved_at_280[0x600];
 
 	u8         pas[0][0x40];
 };
 
 struct mlx5_ifc_create_tis_out_bits {
 	u8         status[0x8];
-	u8         reserved_0[0x18];
+	u8         reserved_at_8[0x18];
 
 	u8         syndrome[0x20];
 
-	u8         reserved_1[0x8];
+	u8         reserved_at_40[0x8];
 	u8         tisn[0x18];
 
-	u8         reserved_2[0x20];
+	u8         reserved_at_60[0x20];
 };
 
 struct mlx5_ifc_create_tis_in_bits {
 	u8         opcode[0x10];
-	u8         reserved_0[0x10];
+	u8         reserved_at_10[0x10];
 
-	u8         reserved_1[0x10];
+	u8         reserved_at_20[0x10];
 	u8         op_mod[0x10];
 
-	u8         reserved_2[0xc0];
+	u8         reserved_at_40[0xc0];
 
 	struct mlx5_ifc_tisc_bits ctx;
 };
 
 struct mlx5_ifc_create_tir_out_bits {
 	u8         status[0x8];
-	u8         reserved_0[0x18];
+	u8         reserved_at_8[0x18];
 
 	u8         syndrome[0x20];
 
-	u8         reserved_1[0x8];
+	u8         reserved_at_40[0x8];
 	u8         tirn[0x18];
 
-	u8         reserved_2[0x20];
+	u8         reserved_at_60[0x20];
 };
 
 struct mlx5_ifc_create_tir_in_bits {
 	u8         opcode[0x10];
-	u8         reserved_0[0x10];
+	u8         reserved_at_10[0x10];
 
-	u8         reserved_1[0x10];
+	u8         reserved_at_20[0x10];
 	u8         op_mod[0x10];
 
-	u8         reserved_2[0xc0];
+	u8         reserved_at_40[0xc0];
 
 	struct mlx5_ifc_tirc_bits ctx;
 };
 
 struct mlx5_ifc_create_srq_out_bits {
 	u8         status[0x8];
-	u8         reserved_0[0x18];
+	u8         reserved_at_8[0x18];
 
 	u8         syndrome[0x20];
 
-	u8         reserved_1[0x8];
+	u8         reserved_at_40[0x8];
 	u8         srqn[0x18];
 
-	u8         reserved_2[0x20];
+	u8         reserved_at_60[0x20];
 };
 
 struct mlx5_ifc_create_srq_in_bits {
 	u8         opcode[0x10];
-	u8         reserved_0[0x10];
+	u8         reserved_at_10[0x10];
 
-	u8         reserved_1[0x10];
+	u8         reserved_at_20[0x10];
 	u8         op_mod[0x10];
 
-	u8         reserved_2[0x40];
+	u8         reserved_at_40[0x40];
 
 	struct mlx5_ifc_srqc_bits srq_context_entry;
 
-	u8         reserved_3[0x600];
+	u8         reserved_at_280[0x600];
 
 	u8         pas[0][0x40];
 };
 
 struct mlx5_ifc_create_sq_out_bits {
 	u8         status[0x8];
-	u8         reserved_0[0x18];
+	u8         reserved_at_8[0x18];
 
 	u8         syndrome[0x20];
 
-	u8         reserved_1[0x8];
+	u8         reserved_at_40[0x8];
 	u8         sqn[0x18];
 
-	u8         reserved_2[0x20];
+	u8         reserved_at_60[0x20];
 };
 
 struct mlx5_ifc_create_sq_in_bits {
 	u8         opcode[0x10];
-	u8         reserved_0[0x10];
+	u8         reserved_at_10[0x10];
 
-	u8         reserved_1[0x10];
+	u8         reserved_at_20[0x10];
 	u8         op_mod[0x10];
 
-	u8         reserved_2[0xc0];
+	u8         reserved_at_40[0xc0];
 
 	struct mlx5_ifc_sqc_bits ctx;
 };
 
 struct mlx5_ifc_create_rqt_out_bits {
 	u8         status[0x8];
-	u8         reserved_0[0x18];
+	u8         reserved_at_8[0x18];
 
 	u8         syndrome[0x20];
 
-	u8         reserved_1[0x8];
+	u8         reserved_at_40[0x8];
 	u8         rqtn[0x18];
 
-	u8         reserved_2[0x20];
+	u8         reserved_at_60[0x20];
 };
 
 struct mlx5_ifc_create_rqt_in_bits {
 	u8         opcode[0x10];
-	u8         reserved_0[0x10];
+	u8         reserved_at_10[0x10];
 
-	u8         reserved_1[0x10];
+	u8         reserved_at_20[0x10];
 	u8         op_mod[0x10];
 
-	u8         reserved_2[0xc0];
+	u8         reserved_at_40[0xc0];
 
 	struct mlx5_ifc_rqtc_bits rqt_context;
 };
 
 struct mlx5_ifc_create_rq_out_bits {
 	u8         status[0x8];
-	u8         reserved_0[0x18];
+	u8         reserved_at_8[0x18];
 
 	u8         syndrome[0x20];
 
-	u8         reserved_1[0x8];
+	u8         reserved_at_40[0x8];
 	u8         rqn[0x18];
 
-	u8         reserved_2[0x20];
+	u8         reserved_at_60[0x20];
 };
 
 struct mlx5_ifc_create_rq_in_bits {
 	u8         opcode[0x10];
-	u8         reserved_0[0x10];
+	u8         reserved_at_10[0x10];
 
-	u8         reserved_1[0x10];
+	u8         reserved_at_20[0x10];
 	u8         op_mod[0x10];
 
-	u8         reserved_2[0xc0];
+	u8         reserved_at_40[0xc0];
 
 	struct mlx5_ifc_rqc_bits ctx;
 };
 
 struct mlx5_ifc_create_rmp_out_bits {
 	u8         status[0x8];
-	u8         reserved_0[0x18];
+	u8         reserved_at_8[0x18];
 
 	u8         syndrome[0x20];
 
-	u8         reserved_1[0x8];
+	u8         reserved_at_40[0x8];
 	u8         rmpn[0x18];
 
-	u8         reserved_2[0x20];
+	u8         reserved_at_60[0x20];
 };
 
 struct mlx5_ifc_create_rmp_in_bits {
 	u8         opcode[0x10];
-	u8         reserved_0[0x10];
+	u8         reserved_at_10[0x10];
 
-	u8         reserved_1[0x10];
+	u8         reserved_at_20[0x10];
 	u8         op_mod[0x10];
 
-	u8         reserved_2[0xc0];
+	u8         reserved_at_40[0xc0];
 
 	struct mlx5_ifc_rmpc_bits ctx;
 };
 
 struct mlx5_ifc_create_qp_out_bits {
 	u8         status[0x8];
-	u8         reserved_0[0x18];
+	u8         reserved_at_8[0x18];
 
 	u8         syndrome[0x20];
 
-	u8         reserved_1[0x8];
+	u8         reserved_at_40[0x8];
 	u8         qpn[0x18];
 
-	u8         reserved_2[0x20];
+	u8         reserved_at_60[0x20];
 };
 
 struct mlx5_ifc_create_qp_in_bits {
 	u8         opcode[0x10];
-	u8         reserved_0[0x10];
+	u8         reserved_at_10[0x10];
 
-	u8         reserved_1[0x10];
+	u8         reserved_at_20[0x10];
 	u8         op_mod[0x10];
 
-	u8         reserved_2[0x40];
+	u8         reserved_at_40[0x40];
 
 	u8         opt_param_mask[0x20];
 
-	u8         reserved_3[0x20];
+	u8         reserved_at_a0[0x20];
 
 	struct mlx5_ifc_qpc_bits qpc;
 
-	u8         reserved_4[0x80];
+	u8         reserved_at_800[0x80];
 
 	u8         pas[0][0x40];
 };
 
 struct mlx5_ifc_create_psv_out_bits {
 	u8         status[0x8];
-	u8         reserved_0[0x18];
+	u8         reserved_at_8[0x18];
 
 	u8         syndrome[0x20];
 
-	u8         reserved_1[0x40];
+	u8         reserved_at_40[0x40];
 
-	u8         reserved_2[0x8];
+	u8         reserved_at_80[0x8];
 	u8         psv0_index[0x18];
 
-	u8         reserved_3[0x8];
+	u8         reserved_at_a0[0x8];
 	u8         psv1_index[0x18];
 
-	u8         reserved_4[0x8];
+	u8         reserved_at_c0[0x8];
 	u8         psv2_index[0x18];
 
-	u8         reserved_5[0x8];
+	u8         reserved_at_e0[0x8];
 	u8         psv3_index[0x18];
 };
 
 struct mlx5_ifc_create_psv_in_bits {
 	u8         opcode[0x10];
-	u8         reserved_0[0x10];
+	u8         reserved_at_10[0x10];
 
-	u8         reserved_1[0x10];
+	u8         reserved_at_20[0x10];
 	u8         op_mod[0x10];
 
 	u8         num_psv[0x4];
-	u8         reserved_2[0x4];
+	u8         reserved_at_44[0x4];
 	u8         pd[0x18];
 
-	u8         reserved_3[0x20];
+	u8         reserved_at_60[0x20];
 };
 
 struct mlx5_ifc_create_mkey_out_bits {
 	u8         status[0x8];
-	u8         reserved_0[0x18];
+	u8         reserved_at_8[0x18];
 
 	u8         syndrome[0x20];
 
-	u8         reserved_1[0x8];
+	u8         reserved_at_40[0x8];
 	u8         mkey_index[0x18];
 
-	u8         reserved_2[0x20];
+	u8         reserved_at_60[0x20];
 };
 
 struct mlx5_ifc_create_mkey_in_bits {
 	u8         opcode[0x10];
-	u8         reserved_0[0x10];
+	u8         reserved_at_10[0x10];
 
-	u8         reserved_1[0x10];
+	u8         reserved_at_20[0x10];
 	u8         op_mod[0x10];
 
-	u8         reserved_2[0x20];
+	u8         reserved_at_40[0x20];
 
 	u8         pg_access[0x1];
-	u8         reserved_3[0x1f];
+	u8         reserved_at_61[0x1f];
 
 	struct mlx5_ifc_mkc_bits memory_key_mkey_entry;
 
-	u8         reserved_4[0x80];
+	u8         reserved_at_280[0x80];
 
 	u8         translations_octword_actual_size[0x20];
 
-	u8         reserved_5[0x560];
+	u8         reserved_at_320[0x560];
 
 	u8         klm_pas_mtt[0][0x20];
 };
 
 struct mlx5_ifc_create_flow_table_out_bits {
 	u8         status[0x8];
-	u8         reserved_0[0x18];
+	u8         reserved_at_8[0x18];
 
 	u8         syndrome[0x20];
 
-	u8         reserved_1[0x8];
+	u8         reserved_at_40[0x8];
 	u8         table_id[0x18];
 
-	u8         reserved_2[0x20];
+	u8         reserved_at_60[0x20];
 };
 
 struct mlx5_ifc_create_flow_table_in_bits {
 	u8         opcode[0x10];
-	u8         reserved_0[0x10];
+	u8         reserved_at_10[0x10];
 
-	u8         reserved_1[0x10];
+	u8         reserved_at_20[0x10];
 	u8         op_mod[0x10];
 
-	u8         reserved_2[0x40];
+	u8         reserved_at_40[0x40];
 
 	u8         table_type[0x8];
-	u8         reserved_3[0x18];
+	u8         reserved_at_88[0x18];
 
-	u8         reserved_4[0x20];
+	u8         reserved_at_a0[0x20];
 
-	u8         reserved_5[0x4];
+	u8         reserved_at_c0[0x4];
 	u8         table_miss_mode[0x4];
 	u8         level[0x8];
-	u8         reserved_6[0x8];
+	u8         reserved_at_d0[0x8];
 	u8         log_size[0x8];
 
-	u8         reserved_7[0x8];
+	u8         reserved_at_e0[0x8];
 	u8         table_miss_id[0x18];
 
-	u8         reserved_8[0x100];
+	u8         reserved_at_100[0x100];
 };
 
 struct mlx5_ifc_create_flow_group_out_bits {
 	u8         status[0x8];
-	u8         reserved_0[0x18];
+	u8         reserved_at_8[0x18];
 
 	u8         syndrome[0x20];
 
-	u8         reserved_1[0x8];
+	u8         reserved_at_40[0x8];
 	u8         group_id[0x18];
 
-	u8         reserved_2[0x20];
+	u8         reserved_at_60[0x20];
 };
 
 enum {
@@ -5742,134 +5744,134 @@ enum {
 
 struct mlx5_ifc_create_flow_group_in_bits {
 	u8         opcode[0x10];
-	u8         reserved_0[0x10];
+	u8         reserved_at_10[0x10];
 
-	u8         reserved_1[0x10];
+	u8         reserved_at_20[0x10];
 	u8         op_mod[0x10];
 
-	u8         reserved_2[0x40];
+	u8         reserved_at_40[0x40];
 
 	u8         table_type[0x8];
-	u8         reserved_3[0x18];
+	u8         reserved_at_88[0x18];
 
-	u8         reserved_4[0x8];
+	u8         reserved_at_a0[0x8];
 	u8         table_id[0x18];
 
-	u8         reserved_5[0x20];
+	u8         reserved_at_c0[0x20];
 
 	u8         start_flow_index[0x20];
 
-	u8         reserved_6[0x20];
+	u8         reserved_at_100[0x20];
 
 	u8         end_flow_index[0x20];
 
-	u8         reserved_7[0xa0];
+	u8         reserved_at_140[0xa0];
 
-	u8         reserved_8[0x18];
+	u8         reserved_at_1e0[0x18];
 	u8         match_criteria_enable[0x8];
 
 	struct mlx5_ifc_fte_match_param_bits match_criteria;
 
-	u8         reserved_9[0xe00];
+	u8         reserved_at_1200[0xe00];
 };
 
 struct mlx5_ifc_create_eq_out_bits {
 	u8         status[0x8];
-	u8         reserved_0[0x18];
+	u8         reserved_at_8[0x18];
 
 	u8         syndrome[0x20];
 
-	u8         reserved_1[0x18];
+	u8         reserved_at_40[0x18];
 	u8         eq_number[0x8];
 
-	u8         reserved_2[0x20];
+	u8         reserved_at_60[0x20];
 };
 
 struct mlx5_ifc_create_eq_in_bits {
 	u8         opcode[0x10];
-	u8         reserved_0[0x10];
+	u8         reserved_at_10[0x10];
 
-	u8         reserved_1[0x10];
+	u8         reserved_at_20[0x10];
 	u8         op_mod[0x10];
 
-	u8         reserved_2[0x40];
+	u8         reserved_at_40[0x40];
 
 	struct mlx5_ifc_eqc_bits eq_context_entry;
 
-	u8         reserved_3[0x40];
+	u8         reserved_at_280[0x40];
 
 	u8         event_bitmask[0x40];
 
-	u8         reserved_4[0x580];
+	u8         reserved_at_300[0x580];
 
 	u8         pas[0][0x40];
 };
 
 struct mlx5_ifc_create_dct_out_bits {
 	u8         status[0x8];
-	u8         reserved_0[0x18];
+	u8         reserved_at_8[0x18];
 
 	u8         syndrome[0x20];
 
-	u8         reserved_1[0x8];
+	u8         reserved_at_40[0x8];
 	u8         dctn[0x18];
 
-	u8         reserved_2[0x20];
+	u8         reserved_at_60[0x20];
 };
 
 struct mlx5_ifc_create_dct_in_bits {
 	u8         opcode[0x10];
-	u8         reserved_0[0x10];
+	u8         reserved_at_10[0x10];
 
-	u8         reserved_1[0x10];
+	u8         reserved_at_20[0x10];
 	u8         op_mod[0x10];
 
-	u8         reserved_2[0x40];
+	u8         reserved_at_40[0x40];
 
 	struct mlx5_ifc_dctc_bits dct_context_entry;
 
-	u8         reserved_3[0x180];
+	u8         reserved_at_280[0x180];
 };
 
 struct mlx5_ifc_create_cq_out_bits {
 	u8         status[0x8];
-	u8         reserved_0[0x18];
+	u8         reserved_at_8[0x18];
 
 	u8         syndrome[0x20];
 
-	u8         reserved_1[0x8];
+	u8         reserved_at_40[0x8];
 	u8         cqn[0x18];
 
-	u8         reserved_2[0x20];
+	u8         reserved_at_60[0x20];
 };
 
 struct mlx5_ifc_create_cq_in_bits {
 	u8         opcode[0x10];
-	u8         reserved_0[0x10];
+	u8         reserved_at_10[0x10];
 
-	u8         reserved_1[0x10];
+	u8         reserved_at_20[0x10];
 	u8         op_mod[0x10];
 
-	u8         reserved_2[0x40];
+	u8         reserved_at_40[0x40];
 
 	struct mlx5_ifc_cqc_bits cq_context;
 
-	u8         reserved_3[0x600];
+	u8         reserved_at_280[0x600];
 
 	u8         pas[0][0x40];
 };
 
 struct mlx5_ifc_config_int_moderation_out_bits {
 	u8         status[0x8];
-	u8         reserved_0[0x18];
+	u8         reserved_at_8[0x18];
 
 	u8         syndrome[0x20];
 
-	u8         reserved_1[0x4];
+	u8         reserved_at_40[0x4];
 	u8         min_delay[0xc];
 	u8         int_vector[0x10];
 
-	u8         reserved_2[0x20];
+	u8         reserved_at_60[0x20];
 };
 
 enum {
@@ -5879,49 +5881,49 @@ enum {
 
 struct mlx5_ifc_config_int_moderation_in_bits {
 	u8         opcode[0x10];
-	u8         reserved_0[0x10];
+	u8         reserved_at_10[0x10];
 
-	u8         reserved_1[0x10];
+	u8         reserved_at_20[0x10];
 	u8         op_mod[0x10];
 
-	u8         reserved_2[0x4];
+	u8         reserved_at_40[0x4];
 	u8         min_delay[0xc];
 	u8         int_vector[0x10];
 
-	u8         reserved_3[0x20];
+	u8         reserved_at_60[0x20];
 };
 
 struct mlx5_ifc_attach_to_mcg_out_bits {
 	u8         status[0x8];
-	u8         reserved_0[0x18];
+	u8         reserved_at_8[0x18];
 
 	u8         syndrome[0x20];
 
-	u8         reserved_1[0x40];
+	u8         reserved_at_40[0x40];
 };
 
 struct mlx5_ifc_attach_to_mcg_in_bits {
 	u8         opcode[0x10];
-	u8         reserved_0[0x10];
+	u8         reserved_at_10[0x10];
 
-	u8         reserved_1[0x10];
+	u8         reserved_at_20[0x10];
 	u8         op_mod[0x10];
 
-	u8         reserved_2[0x8];
+	u8         reserved_at_40[0x8];
 	u8         qpn[0x18];
 
-	u8         reserved_3[0x20];
+	u8         reserved_at_60[0x20];
 
 	u8         multicast_gid[16][0x8];
 };
 
 struct mlx5_ifc_arm_xrc_srq_out_bits {
 	u8         status[0x8];
-	u8         reserved_0[0x18];
+	u8         reserved_at_8[0x18];
 
 	u8         syndrome[0x20];
 
-	u8         reserved_1[0x40];
+	u8         reserved_at_40[0x40];
 };
 
 enum {
@@ -5930,25 +5932,25 @@ enum {
 
 struct mlx5_ifc_arm_xrc_srq_in_bits {
 	u8         opcode[0x10];
-	u8         reserved_0[0x10];
+	u8         reserved_at_10[0x10];
 
-	u8         reserved_1[0x10];
+	u8         reserved_at_20[0x10];
 	u8         op_mod[0x10];
 
-	u8         reserved_2[0x8];
+	u8         reserved_at_40[0x8];
 	u8         xrc_srqn[0x18];
 
-	u8         reserved_3[0x10];
+	u8         reserved_at_60[0x10];
 	u8         lwm[0x10];
 };
 
 struct mlx5_ifc_arm_rq_out_bits {
 	u8         status[0x8];
-	u8         reserved_0[0x18];
+	u8         reserved_at_8[0x18];
 
 	u8         syndrome[0x20];
 
-	u8         reserved_1[0x40];
+	u8         reserved_at_40[0x40];
 };
 
 enum {
@@ -5957,179 +5959,179 @@ enum {
 
 struct mlx5_ifc_arm_rq_in_bits {
 	u8         opcode[0x10];
-	u8         reserved_0[0x10];
+	u8         reserved_at_10[0x10];
 
-	u8         reserved_1[0x10];
+	u8         reserved_at_20[0x10];
 	u8         op_mod[0x10];
 
-	u8         reserved_2[0x8];
+	u8         reserved_at_40[0x8];
 	u8         srq_number[0x18];
 
-	u8         reserved_3[0x10];
+	u8         reserved_at_60[0x10];
 	u8         lwm[0x10];
 };
 
 struct mlx5_ifc_arm_dct_out_bits {
 	u8         status[0x8];
-	u8         reserved_0[0x18];
+	u8         reserved_at_8[0x18];
 
 	u8         syndrome[0x20];
 
-	u8         reserved_1[0x40];
+	u8         reserved_at_40[0x40];
 };
 
 struct mlx5_ifc_arm_dct_in_bits {
 	u8         opcode[0x10];
-	u8         reserved_0[0x10];
+	u8         reserved_at_10[0x10];
 
-	u8         reserved_1[0x10];
+	u8         reserved_at_20[0x10];
 	u8         op_mod[0x10];
 
-	u8         reserved_2[0x8];
+	u8         reserved_at_40[0x8];
 	u8         dct_number[0x18];
 
-	u8         reserved_3[0x20];
+	u8         reserved_at_60[0x20];
 };
 
 struct mlx5_ifc_alloc_xrcd_out_bits {
 	u8         status[0x8];
-	u8         reserved_0[0x18];
+	u8         reserved_at_8[0x18];
 
 	u8         syndrome[0x20];
 
-	u8         reserved_1[0x8];
+	u8         reserved_at_40[0x8];
 	u8         xrcd[0x18];
 
-	u8         reserved_2[0x20];
+	u8         reserved_at_60[0x20];
 };
 
 struct mlx5_ifc_alloc_xrcd_in_bits {
 	u8         opcode[0x10];
-	u8         reserved_0[0x10];
+	u8         reserved_at_10[0x10];
 
-	u8         reserved_1[0x10];
+	u8         reserved_at_20[0x10];
 	u8         op_mod[0x10];
 
-	u8         reserved_2[0x40];
+	u8         reserved_at_40[0x40];
 };
 
 struct mlx5_ifc_alloc_uar_out_bits {
 	u8         status[0x8];
-	u8         reserved_0[0x18];
+	u8         reserved_at_8[0x18];
 
 	u8         syndrome[0x20];
 
-	u8         reserved_1[0x8];
+	u8         reserved_at_40[0x8];
 	u8         uar[0x18];
 
-	u8         reserved_2[0x20];
+	u8         reserved_at_60[0x20];
 };
 
 struct mlx5_ifc_alloc_uar_in_bits {
 	u8         opcode[0x10];
-	u8         reserved_0[0x10];
+	u8         reserved_at_10[0x10];
 
-	u8         reserved_1[0x10];
+	u8         reserved_at_20[0x10];
 	u8         op_mod[0x10];
 
-	u8         reserved_2[0x40];
+	u8         reserved_at_40[0x40];
 };
 
 struct mlx5_ifc_alloc_transport_domain_out_bits {
 	u8         status[0x8];
-	u8         reserved_0[0x18];
+	u8         reserved_at_8[0x18];
 
 	u8         syndrome[0x20];
 
-	u8         reserved_1[0x8];
+	u8         reserved_at_40[0x8];
 	u8         transport_domain[0x18];
 
-	u8         reserved_2[0x20];
+	u8         reserved_at_60[0x20];
 };
 
 struct mlx5_ifc_alloc_transport_domain_in_bits {
 	u8         opcode[0x10];
-	u8         reserved_0[0x10];
+	u8         reserved_at_10[0x10];
 
-	u8         reserved_1[0x10];
+	u8         reserved_at_20[0x10];
 	u8         op_mod[0x10];
 
-	u8         reserved_2[0x40];
+	u8         reserved_at_40[0x40];
 };
 
 struct mlx5_ifc_alloc_q_counter_out_bits {
 	u8         status[0x8];
-	u8         reserved_0[0x18];
+	u8         reserved_at_8[0x18];
 
 	u8         syndrome[0x20];
 
-	u8         reserved_1[0x18];
+	u8         reserved_at_40[0x18];
 	u8         counter_set_id[0x8];
 
-	u8         reserved_2[0x20];
+	u8         reserved_at_60[0x20];
 };
 
 struct mlx5_ifc_alloc_q_counter_in_bits {
 	u8         opcode[0x10];
-	u8         reserved_0[0x10];
+	u8         reserved_at_10[0x10];
 
-	u8         reserved_1[0x10];
+	u8         reserved_at_20[0x10];
 	u8         op_mod[0x10];
 
-	u8         reserved_2[0x40];
+	u8         reserved_at_40[0x40];
 };
 
 struct mlx5_ifc_alloc_pd_out_bits {
 	u8         status[0x8];
-	u8         reserved_0[0x18];
+	u8         reserved_at_8[0x18];
 
 	u8         syndrome[0x20];
 
-	u8         reserved_1[0x8];
+	u8         reserved_at_40[0x8];
 	u8         pd[0x18];
 
-	u8         reserved_2[0x20];
+	u8         reserved_at_60[0x20];
 };
 
 struct mlx5_ifc_alloc_pd_in_bits {
 	u8         opcode[0x10];
-	u8         reserved_0[0x10];
+	u8         reserved_at_10[0x10];
 
-	u8         reserved_1[0x10];
+	u8         reserved_at_20[0x10];
 	u8         op_mod[0x10];
 
-	u8         reserved_2[0x40];
+	u8         reserved_at_40[0x40];
 };
 
 struct mlx5_ifc_add_vxlan_udp_dport_out_bits {
 	u8         status[0x8];
-	u8         reserved_0[0x18];
+	u8         reserved_at_8[0x18];
 
 	u8         syndrome[0x20];
 
-	u8         reserved_1[0x40];
+	u8         reserved_at_40[0x40];
 };
 
 struct mlx5_ifc_add_vxlan_udp_dport_in_bits {
 	u8         opcode[0x10];
-	u8         reserved_0[0x10];
+	u8         reserved_at_10[0x10];
 
-	u8         reserved_1[0x10];
+	u8         reserved_at_20[0x10];
 	u8         op_mod[0x10];
 
-	u8         reserved_2[0x20];
+	u8         reserved_at_40[0x20];
 
-	u8         reserved_3[0x10];
+	u8         reserved_at_60[0x10];
 	u8         vxlan_udp_port[0x10];
 };
 
 struct mlx5_ifc_access_register_out_bits {
 	u8         status[0x8];
-	u8         reserved_0[0x18];
+	u8         reserved_at_8[0x18];
 
 	u8         syndrome[0x20];
 
-	u8         reserved_1[0x40];
+	u8         reserved_at_40[0x40];
 
 	u8         register_data[0][0x20];
 };
@@ -6141,12 +6143,12 @@ enum {
 
 struct mlx5_ifc_access_register_in_bits {
 	u8         opcode[0x10];
-	u8         reserved_0[0x10];
+	u8         reserved_at_10[0x10];
 
-	u8         reserved_1[0x10];
+	u8         reserved_at_20[0x10];
 	u8         op_mod[0x10];
 
-	u8         reserved_2[0x10];
+	u8         reserved_at_40[0x10];
 	u8         register_id[0x10];
 
 	u8         argument[0x20];
@@ -6159,24 +6161,24 @@ struct mlx5_ifc_sltp_reg_bits {
 	u8         version[0x4];
 	u8         local_port[0x8];
 	u8         pnat[0x2];
-	u8         reserved_0[0x2];
+	u8         reserved_at_12[0x2];
 	u8         lane[0x4];
-	u8         reserved_1[0x8];
+	u8         reserved_at_18[0x8];
 
-	u8         reserved_2[0x20];
+	u8         reserved_at_20[0x20];
 
-	u8         reserved_3[0x7];
+	u8         reserved_at_40[0x7];
 	u8         polarity[0x1];
 	u8         ob_tap0[0x8];
 	u8         ob_tap1[0x8];
 	u8         ob_tap2[0x8];
 
-	u8         reserved_4[0xc];
+	u8         reserved_at_60[0xc];
 	u8         ob_preemp_mode[0x4];
 	u8         ob_reg[0x8];
 	u8         ob_bias[0x8];
 
-	u8         reserved_5[0x20];
+	u8         reserved_at_80[0x20];
 };
 
 struct mlx5_ifc_slrg_reg_bits {
@@ -6184,36 +6186,36 @@ struct mlx5_ifc_slrg_reg_bits {
 	u8         version[0x4];
 	u8         local_port[0x8];
 	u8         pnat[0x2];
-	u8         reserved_0[0x2];
+	u8         reserved_at_12[0x2];
 	u8         lane[0x4];
-	u8         reserved_1[0x8];
+	u8         reserved_at_18[0x8];
 
 	u8         time_to_link_up[0x10];
-	u8         reserved_2[0xc];
+	u8         reserved_at_30[0xc];
 	u8         grade_lane_speed[0x4];
 
 	u8         grade_version[0x8];
 	u8         grade[0x18];
 
-	u8         reserved_3[0x4];
+	u8         reserved_at_60[0x4];
 	u8         height_grade_type[0x4];
 	u8         height_grade[0x18];
 
 	u8         height_dz[0x10];
 	u8         height_dv[0x10];
 
-	u8         reserved_4[0x10];
+	u8         reserved_at_a0[0x10];
 	u8         height_sigma[0x10];
 
-	u8         reserved_5[0x20];
+	u8         reserved_at_c0[0x20];
 
-	u8         reserved_6[0x4];
+	u8         reserved_at_e0[0x4];
 	u8         phase_grade_type[0x4];
 	u8         phase_grade[0x18];
 
-	u8         reserved_7[0x8];
+	u8         reserved_at_100[0x8];
 	u8         phase_eo_pos[0x8];
-	u8         reserved_8[0x8];
+	u8         reserved_at_110[0x8];
 	u8         phase_eo_neg[0x8];
 
 	u8         ffe_set_tested[0x10];
@@ -6221,70 +6223,70 @@ struct mlx5_ifc_slrg_reg_bits {
 };
 
 struct mlx5_ifc_pvlc_reg_bits {
-	u8         reserved_0[0x8];
+	u8         reserved_at_0[0x8];
 	u8         local_port[0x8];
-	u8         reserved_1[0x10];
+	u8         reserved_at_10[0x10];
 
-	u8         reserved_2[0x1c];
+	u8         reserved_at_20[0x1c];
 	u8         vl_hw_cap[0x4];
 
-	u8         reserved_3[0x1c];
+	u8         reserved_at_40[0x1c];
 	u8         vl_admin[0x4];
 
-	u8         reserved_4[0x1c];
+	u8         reserved_at_60[0x1c];
 	u8         vl_operational[0x4];
 };
 
 struct mlx5_ifc_pude_reg_bits {
 	u8         swid[0x8];
 	u8         local_port[0x8];
-	u8         reserved_0[0x4];
+	u8         reserved_at_10[0x4];
 	u8         admin_status[0x4];
-	u8         reserved_1[0x4];
+	u8         reserved_at_18[0x4];
 	u8         oper_status[0x4];
 
-	u8         reserved_2[0x60];
+	u8         reserved_at_20[0x60];
 };
 
 struct mlx5_ifc_ptys_reg_bits {
-	u8         reserved_0[0x8];
+	u8         reserved_at_0[0x8];
 	u8         local_port[0x8];
-	u8         reserved_1[0xd];
+	u8         reserved_at_10[0xd];
 	u8         proto_mask[0x3];
 
-	u8         reserved_2[0x40];
+	u8         reserved_at_20[0x40];
 
 	u8         eth_proto_capability[0x20];
 
 	u8         ib_link_width_capability[0x10];
 	u8         ib_proto_capability[0x10];
 
-	u8         reserved_3[0x20];
+	u8         reserved_at_a0[0x20];
 
 	u8         eth_proto_admin[0x20];
 
 	u8         ib_link_width_admin[0x10];
 	u8         ib_proto_admin[0x10];
 
-	u8         reserved_4[0x20];
+	u8         reserved_at_100[0x20];
 
 	u8         eth_proto_oper[0x20];
 
 	u8         ib_link_width_oper[0x10];
 	u8         ib_proto_oper[0x10];
 
-	u8         reserved_5[0x20];
+	u8         reserved_at_160[0x20];
 
 	u8         eth_proto_lp_advertise[0x20];
 
-	u8         reserved_6[0x60];
+	u8         reserved_at_1a0[0x60];
 };
 
 struct mlx5_ifc_ptas_reg_bits {
-	u8         reserved_0[0x20];
+	u8         reserved_at_0[0x20];
 
 	u8         algorithm_options[0x10];
-	u8         reserved_1[0x4];
+	u8         reserved_at_30[0x4];
 	u8         repetitions_mode[0x4];
 	u8         num_of_repetitions[0x8];
 
@@ -6310,13 +6312,13 @@ struct mlx5_ifc_ptas_reg_bits {
 	u8         ndeo_error_threshold[0x10];
 
 	u8         mixer_offset_step_size[0x10];
-	u8         reserved_2[0x8];
+	u8         reserved_at_110[0x8];
 	u8         mix90_phase_for_voltage_bath[0x8];
 
 	u8         mixer_offset_start[0x10];
 	u8         mixer_offset_end[0x10];
 
-	u8         reserved_3[0x15];
+	u8         reserved_at_140[0x15];
 	u8         ber_test_time[0xb];
 };
 
@@ -6324,154 +6326,154 @@ struct mlx5_ifc_pspa_reg_bits {
 	u8         swid[0x8];
 	u8         local_port[0x8];
 	u8         sub_port[0x8];
-	u8         reserved_0[0x8];
+	u8         reserved_at_18[0x8];
 
-	u8         reserved_1[0x20];
+	u8         reserved_at_20[0x20];
 };
 
 struct mlx5_ifc_pqdr_reg_bits {
-	u8         reserved_0[0x8];
+	u8         reserved_at_0[0x8];
 	u8         local_port[0x8];
-	u8         reserved_1[0x5];
+	u8         reserved_at_10[0x5];
 	u8         prio[0x3];
-	u8         reserved_2[0x6];
+	u8         reserved_at_18[0x6];
 	u8         mode[0x2];
 
-	u8         reserved_3[0x20];
+	u8         reserved_at_20[0x20];
 
-	u8         reserved_4[0x10];
+	u8         reserved_at_40[0x10];
 	u8         min_threshold[0x10];
 
-	u8         reserved_5[0x10];
+	u8         reserved_at_60[0x10];
 	u8         max_threshold[0x10];
 
-	u8         reserved_6[0x10];
+	u8         reserved_at_80[0x10];
 	u8         mark_probability_denominator[0x10];
 
-	u8         reserved_7[0x60];
+	u8         reserved_at_a0[0x60];
 };
 
 struct mlx5_ifc_ppsc_reg_bits {
-	u8         reserved_0[0x8];
+	u8         reserved_at_0[0x8];
 	u8         local_port[0x8];
-	u8         reserved_1[0x10];
+	u8         reserved_at_10[0x10];
 
-	u8         reserved_2[0x60];
+	u8         reserved_at_20[0x60];
 
-	u8         reserved_3[0x1c];
+	u8         reserved_at_80[0x1c];
 	u8         wrps_admin[0x4];
 
-	u8         reserved_4[0x1c];
+	u8         reserved_at_a0[0x1c];
 	u8         wrps_status[0x4];
 
-	u8         reserved_5[0x8];
+	u8         reserved_at_c0[0x8];
 	u8         up_threshold[0x8];
-	u8         reserved_6[0x8];
+	u8         reserved_at_d0[0x8];
 	u8         down_threshold[0x8];
 
-	u8         reserved_7[0x20];
+	u8         reserved_at_e0[0x20];
 
-	u8         reserved_8[0x1c];
+	u8         reserved_at_100[0x1c];
 	u8         srps_admin[0x4];
 
-	u8         reserved_9[0x1c];
+	u8         reserved_at_120[0x1c];
 	u8         srps_status[0x4];
 
-	u8         reserved_10[0x40];
+	u8         reserved_at_140[0x40];
 };
 
 struct mlx5_ifc_pplr_reg_bits {
-	u8         reserved_0[0x8];
+	u8         reserved_at_0[0x8];
 	u8         local_port[0x8];
-	u8         reserved_1[0x10];
+	u8         reserved_at_10[0x10];
 
-	u8         reserved_2[0x8];
+	u8         reserved_at_20[0x8];
 	u8         lb_cap[0x8];
-	u8         reserved_3[0x8];
+	u8         reserved_at_30[0x8];
 	u8         lb_en[0x8];
 };
 
 struct mlx5_ifc_pplm_reg_bits {
-	u8         reserved_0[0x8];
+	u8         reserved_at_0[0x8];
 	u8         local_port[0x8];
-	u8         reserved_1[0x10];
+	u8         reserved_at_10[0x10];
 
-	u8         reserved_2[0x20];
+	u8         reserved_at_20[0x20];
 
 	u8         port_profile_mode[0x8];
 	u8         static_port_profile[0x8];
 	u8         active_port_profile[0x8];
-	u8         reserved_3[0x8];
+	u8         reserved_at_58[0x8];
 
 	u8         retransmission_active[0x8];
 	u8         fec_mode_active[0x18];
 
-	u8         reserved_4[0x20];
+	u8         reserved_at_80[0x20];
 };
 
 struct mlx5_ifc_ppcnt_reg_bits {
 	u8         swid[0x8];
 	u8         local_port[0x8];
 	u8         pnat[0x2];
-	u8         reserved_0[0x8];
+	u8         reserved_at_12[0x8];
 	u8         grp[0x6];
 
 	u8         clr[0x1];
-	u8         reserved_1[0x1c];
+	u8         reserved_at_21[0x1c];
 	u8         prio_tc[0x3];
 
 	union mlx5_ifc_eth_cntrs_grp_data_layout_auto_bits counter_set;
 };
 
 struct mlx5_ifc_ppad_reg_bits {
-	u8         reserved_0[0x3];
+	u8         reserved_at_0[0x3];
 	u8         single_mac[0x1];
-	u8         reserved_1[0x4];
+	u8         reserved_at_4[0x4];
 	u8         local_port[0x8];
 	u8         mac_47_32[0x10];
 
 	u8         mac_31_0[0x20];
 
-	u8         reserved_2[0x40];
+	u8         reserved_at_40[0x40];
 };
 
 struct mlx5_ifc_pmtu_reg_bits {
-	u8         reserved_0[0x8];
+	u8         reserved_at_0[0x8];
 	u8         local_port[0x8];
-	u8         reserved_1[0x10];
+	u8         reserved_at_10[0x10];
 
 	u8         max_mtu[0x10];
-	u8         reserved_2[0x10];
+	u8         reserved_at_30[0x10];
 
 	u8         admin_mtu[0x10];
-	u8         reserved_3[0x10];
+	u8         reserved_at_50[0x10];
 
 	u8         oper_mtu[0x10];
-	u8         reserved_4[0x10];
+	u8         reserved_at_70[0x10];
 };
 
 struct mlx5_ifc_pmpr_reg_bits {
-	u8         reserved_0[0x8];
+	u8         reserved_at_0[0x8];
 	u8         module[0x8];
-	u8         reserved_1[0x10];
+	u8         reserved_at_10[0x10];
 
-	u8         reserved_2[0x18];
+	u8         reserved_at_20[0x18];
 	u8         attenuation_5g[0x8];
 
-	u8         reserved_3[0x18];
+	u8         reserved_at_40[0x18];
 	u8         attenuation_7g[0x8];
 
-	u8         reserved_4[0x18];
+	u8         reserved_at_60[0x18];
 	u8         attenuation_12g[0x8];
 };
 
 struct mlx5_ifc_pmpe_reg_bits {
-	u8         reserved_0[0x8];
+	u8         reserved_at_0[0x8];
 	u8         module[0x8];
-	u8         reserved_1[0xc];
+	u8         reserved_at_10[0xc];
 	u8         module_status[0x4];
 
-	u8         reserved_2[0x60];
+	u8         reserved_at_20[0x60];
 };
 
 struct mlx5_ifc_pmpc_reg_bits {
@@ -6479,20 +6481,20 @@ struct mlx5_ifc_pmpc_reg_bits {
 };
 
 struct mlx5_ifc_pmlpn_reg_bits {
-	u8         reserved_0[0x4];
+	u8         reserved_at_0[0x4];
 	u8         mlpn_status[0x4];
 	u8         local_port[0x8];
-	u8         reserved_1[0x10];
+	u8         reserved_at_10[0x10];
 
 	u8         e[0x1];
-	u8         reserved_2[0x1f];
+	u8         reserved_at_21[0x1f];
 };
 
 struct mlx5_ifc_pmlp_reg_bits {
 	u8         rxtx[0x1];
-	u8         reserved_0[0x7];
+	u8         reserved_at_1[0x7];
 	u8         local_port[0x8];
-	u8         reserved_1[0x8];
+	u8         reserved_at_10[0x8];
 	u8         width[0x8];
 
 	u8         lane0_module_mapping[0x20];
@@ -6503,36 +6505,36 @@ struct mlx5_ifc_pmlp_reg_bits {
 
 	u8         lane3_module_mapping[0x20];
 
-	u8         reserved_2[0x160];
+	u8         reserved_at_a0[0x160];
 };
 
 struct mlx5_ifc_pmaos_reg_bits {
-	u8         reserved_0[0x8];
+	u8         reserved_at_0[0x8];
 	u8         module[0x8];
-	u8         reserved_1[0x4];
+	u8         reserved_at_10[0x4];
 	u8         admin_status[0x4];
-	u8         reserved_2[0x4];
+	u8         reserved_at_18[0x4];
 	u8         oper_status[0x4];
 
 	u8         ase[0x1];
 	u8         ee[0x1];
-	u8         reserved_3[0x1c];
+	u8         reserved_at_22[0x1c];
 	u8         e[0x2];
 
-	u8         reserved_4[0x40];
+	u8         reserved_at_40[0x40];
 };
 
 struct mlx5_ifc_plpc_reg_bits {
-	u8         reserved_0[0x4];
+	u8         reserved_at_0[0x4];
 	u8         profile_id[0xc];
-	u8         reserved_1[0x4];
+	u8         reserved_at_10[0x4];
 	u8         proto_mask[0x4];
-	u8         reserved_2[0x8];
+	u8         reserved_at_18[0x8];
 
-	u8         reserved_3[0x10];
+	u8         reserved_at_20[0x10];
 	u8         lane_speed[0x10];
 
-	u8         reserved_4[0x17];
+	u8         reserved_at_40[0x17];
 	u8         lpbf[0x1];
 	u8         fec_mode_policy[0x8];
 
@@ -6545,44 +6547,44 @@ struct mlx5_ifc_plpc_reg_bits {
 	u8         retransmission_request_admin[0x8];
 	u8         fec_mode_request_admin[0x18];
 
-	u8         reserved_5[0x80];
+	u8         reserved_at_c0[0x80];
 };
 
 struct mlx5_ifc_plib_reg_bits {
-	u8         reserved_0[0x8];
+	u8         reserved_at_0[0x8];
 	u8         local_port[0x8];
-	u8         reserved_1[0x8];
+	u8         reserved_at_10[0x8];
 	u8         ib_port[0x8];
 
-	u8         reserved_2[0x60];
+	u8         reserved_at_20[0x60];
 };
 
 struct mlx5_ifc_plbf_reg_bits {
-	u8         reserved_0[0x8];
+	u8         reserved_at_0[0x8];
 	u8         local_port[0x8];
-	u8         reserved_1[0xd];
+	u8         reserved_at_10[0xd];
 	u8         lbf_mode[0x3];
 
-	u8         reserved_2[0x20];
+	u8         reserved_at_20[0x20];
 };
 
 struct mlx5_ifc_pipg_reg_bits {
-	u8         reserved_0[0x8];
+	u8         reserved_at_0[0x8];
 	u8         local_port[0x8];
-	u8         reserved_1[0x10];
+	u8         reserved_at_10[0x10];
 
 	u8         dic[0x1];
-	u8         reserved_2[0x19];
+	u8         reserved_at_21[0x19];
 	u8         ipg[0x4];
-	u8         reserved_3[0x2];
+	u8         reserved_at_3e[0x2];
 };
 
 struct mlx5_ifc_pifr_reg_bits {
-	u8         reserved_0[0x8];
+	u8         reserved_at_0[0x8];
 	u8         local_port[0x8];
-	u8         reserved_1[0x10];
+	u8         reserved_at_10[0x10];
 
-	u8         reserved_2[0xe0];
+	u8         reserved_at_20[0xe0];
 
 	u8         port_filter[8][0x20];
 
@@ -6590,36 +6592,36 @@ struct mlx5_ifc_pifr_reg_bits {
 };
 
 struct mlx5_ifc_pfcc_reg_bits {
-	u8         reserved_0[0x8];
+	u8         reserved_at_0[0x8];
 	u8         local_port[0x8];
-	u8         reserved_1[0x10];
+	u8         reserved_at_10[0x10];
 
 	u8         ppan[0x4];
-	u8         reserved_2[0x4];
+	u8         reserved_at_24[0x4];
 	u8         prio_mask_tx[0x8];
-	u8         reserved_3[0x8];
+	u8         reserved_at_30[0x8];
 	u8         prio_mask_rx[0x8];
 
 	u8         pptx[0x1];
 	u8         aptx[0x1];
-	u8         reserved_4[0x6];
+	u8         reserved_at_42[0x6];
 	u8         pfctx[0x8];
-	u8         reserved_5[0x10];
+	u8         reserved_at_50[0x10];
 
 	u8         pprx[0x1];
 	u8         aprx[0x1];
-	u8         reserved_6[0x6];
+	u8         reserved_at_62[0x6];
 	u8         pfcrx[0x8];
-	u8         reserved_7[0x10];
+	u8         reserved_at_70[0x10];
 
-	u8         reserved_8[0x80];
+	u8         reserved_at_80[0x80];
 };
 
 struct mlx5_ifc_pelc_reg_bits {
 	u8         op[0x4];
-	u8         reserved_0[0x4];
+	u8         reserved_at_4[0x4];
 	u8         local_port[0x8];
-	u8         reserved_1[0x10];
+	u8         reserved_at_10[0x10];
 
 	u8         op_admin[0x8];
 	u8         op_capability[0x8];
@@ -6634,28 +6636,28 @@ struct mlx5_ifc_pelc_reg_bits {
 
 	u8         active[0x40];
 
-	u8         reserved_2[0x80];
+	u8         reserved_at_140[0x80];
 };
 
 struct mlx5_ifc_peir_reg_bits {
-	u8         reserved_0[0x8];
+	u8         reserved_at_0[0x8];
 	u8         local_port[0x8];
-	u8         reserved_1[0x10];
+	u8         reserved_at_10[0x10];
 
-	u8         reserved_2[0xc];
+	u8         reserved_at_20[0xc];
 	u8         error_count[0x4];
-	u8         reserved_3[0x10];
+	u8         reserved_at_30[0x10];
 
-	u8         reserved_4[0xc];
+	u8         reserved_at_40[0xc];
 	u8         lane[0x4];
-	u8         reserved_5[0x8];
+	u8         reserved_at_50[0x8];
 	u8         error_type[0x8];
 };
 
 struct mlx5_ifc_pcap_reg_bits {
-	u8         reserved_0[0x8];
+	u8         reserved_at_0[0x8];
 	u8         local_port[0x8];
-	u8         reserved_1[0x10];
+	u8         reserved_at_10[0x10];
 
 	u8         port_capability_mask[4][0x20];
 };
@@ -6663,46 +6665,46 @@ struct mlx5_ifc_pcap_reg_bits {
 struct mlx5_ifc_paos_reg_bits {
 	u8         swid[0x8];
 	u8         local_port[0x8];
-	u8         reserved_0[0x4];
+	u8         reserved_at_10[0x4];
 	u8         admin_status[0x4];
-	u8         reserved_1[0x4];
+	u8         reserved_at_18[0x4];
 	u8         oper_status[0x4];
 
 	u8         ase[0x1];
 	u8         ee[0x1];
-	u8         reserved_2[0x1c];
+	u8         reserved_at_22[0x1c];
 	u8         e[0x2];
 
-	u8         reserved_3[0x40];
+	u8         reserved_at_40[0x40];
 };
 
 struct mlx5_ifc_pamp_reg_bits {
-	u8         reserved_0[0x8];
+	u8         reserved_at_0[0x8];
 	u8         opamp_group[0x8];
-	u8         reserved_1[0xc];
+	u8         reserved_at_10[0xc];
 	u8         opamp_group_type[0x4];
 
 	u8         start_index[0x10];
-	u8         reserved_2[0x4];
+	u8         reserved_at_30[0x4];
 	u8         num_of_indices[0xc];
 
 	u8         index_data[18][0x10];
 };
 
 struct mlx5_ifc_lane_2_module_mapping_bits {
-	u8         reserved_0[0x6];
+	u8         reserved_at_0[0x6];
 	u8         rx_lane[0x2];
-	u8         reserved_1[0x6];
+	u8         reserved_at_8[0x6];
 	u8         tx_lane[0x2];
-	u8         reserved_2[0x8];
+	u8         reserved_at_10[0x8];
 	u8         module[0x8];
 };
 
 struct mlx5_ifc_bufferx_reg_bits {
-	u8         reserved_0[0x6];
+	u8         reserved_at_0[0x6];
 	u8         lossy[0x1];
 	u8         epsb[0x1];
-	u8         reserved_1[0xc];
+	u8         reserved_at_8[0xc];
 	u8         size[0xc];
 
 	u8         xoff_threshold[0x10];
@@ -6714,21 +6716,21 @@ struct mlx5_ifc_set_node_in_bits {
 };
 
 struct mlx5_ifc_register_power_settings_bits {
-	u8         reserved_0[0x18];
+	u8         reserved_at_0[0x18];
 	u8         power_settings_level[0x8];
 
-	u8         reserved_1[0x60];
+	u8         reserved_at_20[0x60];
 };
 
 struct mlx5_ifc_register_host_endianness_bits {
 	u8         he[0x1];
-	u8         reserved_0[0x1f];
+	u8         reserved_at_1[0x1f];
 
-	u8         reserved_1[0x60];
+	u8         reserved_at_20[0x60];
 };
 
 struct mlx5_ifc_umr_pointer_desc_argument_bits {
-	u8         reserved_0[0x20];
+	u8         reserved_at_0[0x20];
 
 	u8         mkey[0x20];
 
@@ -6741,7 +6743,7 @@ struct mlx5_ifc_ud_adrs_vector_bits {
 	u8         dc_key[0x40];
 
 	u8         ext[0x1];
-	u8         reserved_0[0x7];
+	u8         reserved_at_41[0x7];
 	u8         destination_qp_dct[0x18];
 
 	u8         static_rate[0x4];
@@ -6750,7 +6752,7 @@ struct mlx5_ifc_ud_adrs_vector_bits {
 	u8         mlid[0x7];
 	u8         rlid_udp_sport[0x10];
 
-	u8         reserved_1[0x20];
+	u8         reserved_at_80[0x20];
 
 	u8         rmac_47_16[0x20];
 
@@ -6758,9 +6760,9 @@ struct mlx5_ifc_ud_adrs_vector_bits {
 	u8         tclass[0x8];
 	u8         hop_limit[0x8];
 
-	u8         reserved_2[0x1];
+	u8         reserved_at_e0[0x1];
 	u8         grh[0x1];
-	u8         reserved_3[0x2];
+	u8         reserved_at_e2[0x2];
 	u8         src_addr_index[0x8];
 	u8         flow_label[0x14];
 
@@ -6768,27 +6770,27 @@ struct mlx5_ifc_ud_adrs_vector_bits {
 };
 
 struct mlx5_ifc_pages_req_event_bits {
-	u8         reserved_0[0x10];
+	u8         reserved_at_0[0x10];
 	u8         function_id[0x10];
 
 	u8         num_pages[0x20];
 
-	u8         reserved_1[0xa0];
+	u8         reserved_at_40[0xa0];
 };
 
 struct mlx5_ifc_eqe_bits {
-	u8         reserved_0[0x8];
+	u8         reserved_at_0[0x8];
 	u8         event_type[0x8];
-	u8         reserved_1[0x8];
+	u8         reserved_at_10[0x8];
 	u8         event_sub_type[0x8];
 
-	u8         reserved_2[0xe0];
+	u8         reserved_at_20[0xe0];
 
 	union mlx5_ifc_event_auto_bits event_data;
 
-	u8         reserved_3[0x10];
+	u8         reserved_at_1e0[0x10];
 	u8         signature[0x8];
-	u8         reserved_4[0x7];
+	u8         reserved_at_1f8[0x7];
 	u8         owner[0x1];
 };
 
@@ -6798,14 +6800,14 @@ enum {
 
 struct mlx5_ifc_cmd_queue_entry_bits {
 	u8         type[0x8];
-	u8         reserved_0[0x18];
+	u8         reserved_at_8[0x18];
 
 	u8         input_length[0x20];
 
 	u8         input_mailbox_pointer_63_32[0x20];
 
 	u8         input_mailbox_pointer_31_9[0x17];
-	u8         reserved_1[0x9];
+	u8         reserved_at_77[0x9];
 
 	u8         command_input_inline_data[16][0x8];
 
@@ -6814,20 +6816,20 @@ struct mlx5_ifc_cmd_queue_entry_bits {
 	u8         output_mailbox_pointer_63_32[0x20];
 
 	u8         output_mailbox_pointer_31_9[0x17];
-	u8         reserved_2[0x9];
+	u8         reserved_at_1b7[0x9];
 
 	u8         output_length[0x20];
 
 	u8         token[0x8];
 	u8         signature[0x8];
-	u8         reserved_3[0x8];
+	u8         reserved_at_1f0[0x8];
 	u8         status[0x7];
 	u8         ownership[0x1];
 };
 
 struct mlx5_ifc_cmd_out_bits {
 	u8         status[0x8];
-	u8         reserved_0[0x18];
+	u8         reserved_at_8[0x18];
 
 	u8         syndrome[0x20];
 
@@ -6836,9 +6838,9 @@ struct mlx5_ifc_cmd_out_bits {
 
 struct mlx5_ifc_cmd_in_bits {
 	u8         opcode[0x10];
-	u8         reserved_0[0x10];
+	u8         reserved_at_10[0x10];
 
-	u8         reserved_1[0x10];
+	u8         reserved_at_20[0x10];
 	u8         op_mod[0x10];
 
 	u8         command[0][0x20];
@@ -6847,16 +6849,16 @@ struct mlx5_ifc_cmd_in_bits {
 struct mlx5_ifc_cmd_if_box_bits {
 	u8         mailbox_data[512][0x8];
 
-	u8         reserved_0[0x180];
+	u8         reserved_at_1000[0x180];
 
 	u8         next_pointer_63_32[0x20];
 
 	u8         next_pointer_31_10[0x16];
-	u8         reserved_1[0xa];
+	u8         reserved_at_11b6[0xa];
 
 	u8         block_number[0x20];
 
-	u8         reserved_2[0x8];
+	u8         reserved_at_11e0[0x8];
 	u8         token[0x8];
 	u8         ctrl_signature[0x8];
 	u8         signature[0x8];
@@ -6866,7 +6868,7 @@ struct mlx5_ifc_mtt_bits {
 	u8         ptag_63_32[0x20];
 
 	u8         ptag_31_8[0x18];
-	u8         reserved_0[0x6];
+	u8         reserved_at_38[0x6];
 	u8         wr_en[0x1];
 	u8         rd_en[0x1];
 };
@@ -6904,38 +6906,38 @@ struct mlx5_ifc_initial_seg_bits {
 	u8         cmd_interface_rev[0x10];
 	u8         fw_rev_subminor[0x10];
 
-	u8         reserved_0[0x40];
+	u8         reserved_at_40[0x40];
 
 	u8         cmdq_phy_addr_63_32[0x20];
 
 	u8         cmdq_phy_addr_31_12[0x14];
-	u8         reserved_1[0x2];
+	u8         reserved_at_b4[0x2];
 	u8         nic_interface[0x2];
 	u8         log_cmdq_size[0x4];
 	u8         log_cmdq_stride[0x4];
 
 	u8         command_doorbell_vector[0x20];
 
-	u8         reserved_2[0xf00];
+	u8         reserved_at_e0[0xf00];
 
 	u8         initializing[0x1];
-	u8         reserved_3[0x4];
+	u8         reserved_at_fe1[0x4];
 	u8         nic_interface_supported[0x3];
-	u8         reserved_4[0x18];
+	u8         reserved_at_fe8[0x18];
 
 	struct mlx5_ifc_health_buffer_bits health_buffer;
 
 	u8         no_dram_nic_offset[0x20];
 
-	u8         reserved_5[0x6e40];
+	u8         reserved_at_1220[0x6e40];
 
-	u8         reserved_6[0x1f];
+	u8         reserved_at_8060[0x1f];
 	u8         clear_int[0x1];
 
 	u8         health_syndrome[0x8];
 	u8         health_counter[0x18];
 
-	u8         reserved_7[0x17fc0];
+	u8         reserved_at_80a0[0x17fc0];
 };
 
 union mlx5_ifc_ports_control_registers_document_bits {
@@ -6980,44 +6982,44 @@ union mlx5_ifc_ports_control_registers_document_bits {
 	struct mlx5_ifc_pvlc_reg_bits pvlc_reg;
 	struct mlx5_ifc_slrg_reg_bits slrg_reg;
 	struct mlx5_ifc_sltp_reg_bits sltp_reg;
-	u8         reserved_0[0x60e0];
+	u8         reserved_at_0[0x60e0];
 };
 
 union mlx5_ifc_debug_enhancements_document_bits {
 	struct mlx5_ifc_health_buffer_bits health_buffer;
-	u8         reserved_0[0x200];
+	u8         reserved_at_0[0x200];
 };
 
 union mlx5_ifc_uplink_pci_interface_document_bits {
 	struct mlx5_ifc_initial_seg_bits initial_seg;
-	u8         reserved_0[0x20060];
+	u8         reserved_at_0[0x20060];
 };
 
 struct mlx5_ifc_set_flow_table_root_out_bits {
 	u8         status[0x8];
-	u8         reserved_0[0x18];
+	u8         reserved_at_8[0x18];
 
 	u8         syndrome[0x20];
 
-	u8         reserved_1[0x40];
+	u8         reserved_at_40[0x40];
 };
 
 struct mlx5_ifc_set_flow_table_root_in_bits {
 	u8         opcode[0x10];
-	u8         reserved_0[0x10];
+	u8         reserved_at_10[0x10];
 
-	u8         reserved_1[0x10];
+	u8         reserved_at_20[0x10];
 	u8         op_mod[0x10];
 
-	u8         reserved_2[0x40];
+	u8         reserved_at_40[0x40];
 
 	u8         table_type[0x8];
-	u8         reserved_3[0x18];
+	u8         reserved_at_88[0x18];
 
-	u8         reserved_4[0x8];
+	u8         reserved_at_a0[0x8];
 	u8         table_id[0x18];
 
-	u8         reserved_5[0x140];
+	u8         reserved_at_c0[0x140];
 };
 
 enum {
@@ -7026,39 +7028,39 @@ enum {
 
 struct mlx5_ifc_modify_flow_table_out_bits {
 	u8         status[0x8];
-	u8         reserved_0[0x18];
+	u8         reserved_at_8[0x18];
 
 	u8         syndrome[0x20];
 
-	u8         reserved_1[0x40];
+	u8         reserved_at_40[0x40];
 };
 
 struct mlx5_ifc_modify_flow_table_in_bits {
 	u8         opcode[0x10];
-	u8         reserved_0[0x10];
+	u8         reserved_at_10[0x10];
 
-	u8         reserved_1[0x10];
+	u8         reserved_at_20[0x10];
 	u8         op_mod[0x10];
 
-	u8         reserved_2[0x20];
+	u8         reserved_at_40[0x20];
 
-	u8         reserved_3[0x10];
+	u8         reserved_at_60[0x10];
 	u8         modify_field_select[0x10];
 
 	u8         table_type[0x8];
-	u8         reserved_4[0x18];
+	u8         reserved_at_88[0x18];
 
-	u8         reserved_5[0x8];
+	u8         reserved_at_a0[0x8];
 	u8         table_id[0x18];
 
-	u8         reserved_6[0x4];
+	u8         reserved_at_c0[0x4];
 	u8         table_miss_mode[0x4];
-	u8         reserved_7[0x18];
+	u8         reserved_at_c8[0x18];
 
-	u8         reserved_8[0x8];
+	u8         reserved_at_e0[0x8];
 	u8         table_miss_id[0x18];
 
-	u8         reserved_9[0x100];
+	u8         reserved_at_100[0x100];
 };
 
 #endif /* MLX5_IFC_H */
diff --git a/include/linux/module.h b/include/linux/module.h
index 4560d8f1545d..2bb0c3085706 100644
--- a/include/linux/module.h
+++ b/include/linux/module.h
@@ -324,6 +324,12 @@ struct module_layout {
 #define __module_layout_align
 #endif
 
+struct mod_kallsyms {
+	Elf_Sym *symtab;
+	unsigned int num_symtab;
+	char *strtab;
+};
+
 struct module {
 	enum module_state state;
 
@@ -405,15 +411,10 @@ struct module {
 #endif
 
 #ifdef CONFIG_KALLSYMS
-	/*
-	 * We keep the symbol and string tables for kallsyms.
-	 * The core_* fields below are temporary, loader-only (they
-	 * could really be discarded after module init).
-	 */
-	Elf_Sym *symtab, *core_symtab;
-	unsigned int num_symtab, core_num_syms;
-	char *strtab, *core_strtab;
-
+	/* Protected by RCU and/or module_mutex: use rcu_dereference() */
+	struct mod_kallsyms *kallsyms;
+	struct mod_kallsyms core_kallsyms;
+	
 	/* Section attributes */
 	struct module_sect_attrs *sect_attrs;
 
diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h
index 289c2314d766..5440b7b705eb 100644
--- a/include/linux/netdevice.h
+++ b/include/linux/netdevice.h
@@ -3718,7 +3718,7 @@ void *netdev_lower_get_next_private_rcu(struct net_device *dev,
 void *netdev_lower_get_next(struct net_device *dev,
 				struct list_head **iter);
 #define netdev_for_each_lower_dev(dev, ldev, iter) \
-	for (iter = &(dev)->adj_list.lower, \
+	for (iter = (dev)->adj_list.lower.next, \
 	     ldev = netdev_lower_get_next(dev, &(iter)); \
 	     ldev; \
 	     ldev = netdev_lower_get_next(dev, &(iter)))
diff --git a/include/linux/nfs_fs.h b/include/linux/nfs_fs.h
index 48e0320cd643..67300f8e5f2f 100644
--- a/include/linux/nfs_fs.h
+++ b/include/linux/nfs_fs.h
@@ -550,9 +550,7 @@ extern int  nfs_readpage_async(struct nfs_open_context *, struct inode *,
 
 static inline loff_t nfs_size_to_loff_t(__u64 size)
 {
-	if (size > (__u64) OFFSET_MAX - 1)
-		return OFFSET_MAX - 1;
-	return (loff_t) size;
+	return min_t(u64, size, OFFSET_MAX);
 }
 
 static inline ino_t
diff --git a/include/linux/nfs_xdr.h b/include/linux/nfs_xdr.h
index 791098a08a87..d320906cf13e 100644
--- a/include/linux/nfs_xdr.h
+++ b/include/linux/nfs_xdr.h
@@ -275,6 +275,7 @@ struct nfs4_layoutcommit_args {
 	size_t layoutupdate_len;
 	struct page *layoutupdate_page;
 	struct page **layoutupdate_pages;
+	__be32 *start_p;
 };
 
 struct nfs4_layoutcommit_res {
diff --git a/include/linux/pci.h b/include/linux/pci.h
index 27df4a6585da..27716254dcc5 100644
--- a/include/linux/pci.h
+++ b/include/linux/pci.h
@@ -988,23 +988,6 @@ static inline int pci_is_managed(struct pci_dev *pdev)
 	return pdev->is_managed;
 }
 
-static inline void pci_set_managed_irq(struct pci_dev *pdev, unsigned int irq)
-{
-	pdev->irq = irq;
-	pdev->irq_managed = 1;
-}
-
-static inline void pci_reset_managed_irq(struct pci_dev *pdev)
-{
-	pdev->irq = 0;
-	pdev->irq_managed = 0;
-}
-
-static inline bool pci_has_managed_irq(struct pci_dev *pdev)
-{
-	return pdev->irq_managed && pdev->irq > 0;
-}
-
 void pci_disable_device(struct pci_dev *dev);
 
 extern unsigned int pcibios_max_latency;
diff --git a/include/linux/perf_event.h b/include/linux/perf_event.h
index b35a61a481fa..f5c5a3fa2c81 100644
--- a/include/linux/perf_event.h
+++ b/include/linux/perf_event.h
@@ -397,6 +397,7 @@ struct pmu {
  * enum perf_event_active_state - the states of a event
  */
 enum perf_event_active_state {
+	PERF_EVENT_STATE_DEAD		= -4,
 	PERF_EVENT_STATE_EXIT		= -3,
 	PERF_EVENT_STATE_ERROR		= -2,
 	PERF_EVENT_STATE_OFF		= -1,
@@ -905,7 +906,7 @@ perf_sw_event_sched(u32 event_id, u64 nr, u64 addr)
 	}
 }
 
-extern struct static_key_deferred perf_sched_events;
+extern struct static_key_false perf_sched_events;
 
 static __always_inline bool
 perf_sw_migrate_enabled(void)
@@ -924,7 +925,7 @@ static inline void perf_event_task_migrate(struct task_struct *task)
 static inline void perf_event_task_sched_in(struct task_struct *prev,
 					    struct task_struct *task)
 {
-	if (static_key_false(&perf_sched_events.key))
+	if (static_branch_unlikely(&perf_sched_events))
 		__perf_event_task_sched_in(prev, task);
 
 	if (perf_sw_migrate_enabled() && task->sched_migrated) {
@@ -941,7 +942,7 @@ static inline void perf_event_task_sched_out(struct task_struct *prev,
 {
 	perf_sw_event_sched(PERF_COUNT_SW_CONTEXT_SWITCHES, 1, 0);
 
-	if (static_key_false(&perf_sched_events.key))
+	if (static_branch_unlikely(&perf_sched_events))
 		__perf_event_task_sched_out(prev, next);
 }
 
diff --git a/include/linux/pfn.h b/include/linux/pfn.h
index 2d8e49711b63..1132953235c0 100644
--- a/include/linux/pfn.h
+++ b/include/linux/pfn.h
@@ -10,7 +10,7 @@
  * backing is indicated by flags in the high bits of the value.
  */
 typedef struct {
-	unsigned long val;
+	u64 val;
 } pfn_t;
 #endif
 
diff --git a/include/linux/pfn_t.h b/include/linux/pfn_t.h
index 37448ab5fb5c..94994810c7c0 100644
--- a/include/linux/pfn_t.h
+++ b/include/linux/pfn_t.h
@@ -9,14 +9,13 @@
  * PFN_DEV - pfn is not covered by system memmap by default
  * PFN_MAP - pfn has a dynamic page mapping established by a device driver
  */
-#define PFN_FLAGS_MASK (((unsigned long) ~PAGE_MASK) \
-		<< (BITS_PER_LONG - PAGE_SHIFT))
-#define PFN_SG_CHAIN (1UL << (BITS_PER_LONG - 1))
-#define PFN_SG_LAST (1UL << (BITS_PER_LONG - 2))
-#define PFN_DEV (1UL << (BITS_PER_LONG - 3))
-#define PFN_MAP (1UL << (BITS_PER_LONG - 4))
-
-static inline pfn_t __pfn_to_pfn_t(unsigned long pfn, unsigned long flags)
+#define PFN_FLAGS_MASK (((u64) ~PAGE_MASK) << (BITS_PER_LONG_LONG - PAGE_SHIFT))
+#define PFN_SG_CHAIN (1ULL << (BITS_PER_LONG_LONG - 1))
+#define PFN_SG_LAST (1ULL << (BITS_PER_LONG_LONG - 2))
+#define PFN_DEV (1ULL << (BITS_PER_LONG_LONG - 3))
+#define PFN_MAP (1ULL << (BITS_PER_LONG_LONG - 4))
+
+static inline pfn_t __pfn_to_pfn_t(unsigned long pfn, u64 flags)
 {
 	pfn_t pfn_t = { .val = pfn | (flags & PFN_FLAGS_MASK), };
 
@@ -29,7 +28,7 @@ static inline pfn_t pfn_to_pfn_t(unsigned long pfn)
 	return __pfn_to_pfn_t(pfn, 0);
 }
 
-extern pfn_t phys_to_pfn_t(phys_addr_t addr, unsigned long flags);
+extern pfn_t phys_to_pfn_t(phys_addr_t addr, u64 flags);
 
 static inline bool pfn_t_has_page(pfn_t pfn)
 {
@@ -87,7 +86,7 @@ static inline pmd_t pfn_t_pmd(pfn_t pfn, pgprot_t pgprot)
 #ifdef __HAVE_ARCH_PTE_DEVMAP
 static inline bool pfn_t_devmap(pfn_t pfn)
 {
-	const unsigned long flags = PFN_DEV|PFN_MAP;
+	const u64 flags = PFN_DEV|PFN_MAP;
 
 	return (pfn.val & flags) == flags;
 }
diff --git a/include/linux/power/bq27xxx_battery.h b/include/linux/power/bq27xxx_battery.h
index 998d8f1c3c91..b50c0492629d 100644
--- a/include/linux/power/bq27xxx_battery.h
+++ b/include/linux/power/bq27xxx_battery.h
@@ -49,6 +49,7 @@ struct bq27xxx_reg_cache {
 
 struct bq27xxx_device_info {
 	struct device *dev;
+	int id;
 	enum bq27xxx_chip chip;
 	const char *name;
 	struct bq27xxx_access_methods bus;
diff --git a/include/linux/random.h b/include/linux/random.h
index a75840c1aa71..9c29122037f9 100644
--- a/include/linux/random.h
+++ b/include/linux/random.h
@@ -34,6 +34,7 @@ extern const struct file_operations random_fops, urandom_fops;
 #endif
 
 unsigned int get_random_int(void);
+unsigned long get_random_long(void);
 unsigned long randomize_range(unsigned long start, unsigned long end, unsigned long len);
 
 u32 prandom_u32(void);
diff --git a/include/linux/skbuff.h b/include/linux/skbuff.h
index 11f935c1a090..d3fcd4591ce4 100644
--- a/include/linux/skbuff.h
+++ b/include/linux/skbuff.h
@@ -299,6 +299,7 @@ struct sk_buff;
 #else
 #define MAX_SKB_FRAGS (65536/PAGE_SIZE + 1)
 #endif
+extern int sysctl_max_skb_frags;
 
 typedef struct skb_frag_struct skb_frag_t;
 
@@ -1984,6 +1985,30 @@ static inline void skb_reserve(struct sk_buff *skb, int len)
 	skb->tail += len;
 }
 
+/**
+ *	skb_tailroom_reserve - adjust reserved_tailroom
+ *	@skb: buffer to alter
+ *	@mtu: maximum amount of headlen permitted
+ *	@needed_tailroom: minimum amount of reserved_tailroom
+ *
+ *	Set reserved_tailroom so that headlen can be as large as possible but
+ *	not larger than mtu and tailroom cannot be smaller than
+ *	needed_tailroom.
+ *	The required headroom should already have been reserved before using
+ *	this function.
+ */
+static inline void skb_tailroom_reserve(struct sk_buff *skb, unsigned int mtu,
+					unsigned int needed_tailroom)
+{
+	SKB_LINEAR_ASSERT(skb);
+	if (mtu < skb_tailroom(skb) - needed_tailroom)
+		/* use at most mtu */
+		skb->reserved_tailroom = skb_tailroom(skb) - mtu;
+	else
+		/* use up to all available space */
+		skb->reserved_tailroom = needed_tailroom;
+}
+
 #define ENCAP_TYPE_ETHER	0
 #define ENCAP_TYPE_IPPROTO	1
 
diff --git a/include/linux/soc/ti/knav_dma.h b/include/linux/soc/ti/knav_dma.h
index 343c13ac4f71..35cb9264e0d5 100644
--- a/include/linux/soc/ti/knav_dma.h
+++ b/include/linux/soc/ti/knav_dma.h
@@ -44,6 +44,7 @@
 
 #define KNAV_DMA_NUM_EPIB_WORDS			4
 #define KNAV_DMA_NUM_PS_WORDS			16
+#define KNAV_DMA_NUM_SW_DATA_WORDS		4
 #define KNAV_DMA_FDQ_PER_CHAN			4
 
 /* Tx channel scheduling priority */
@@ -142,6 +143,7 @@ struct knav_dma_cfg {
  * @orig_buff:			buff pointer since 'buff' can be overwritten
  * @epib:			Extended packet info block
  * @psdata:			Protocol specific
+ * @sw_data:			Software private data not touched by h/w
  */
 struct knav_dma_desc {
 	__le32	desc_info;
@@ -154,7 +156,7 @@ struct knav_dma_desc {
 	__le32	orig_buff;
 	__le32	epib[KNAV_DMA_NUM_EPIB_WORDS];
 	__le32	psdata[KNAV_DMA_NUM_PS_WORDS];
-	__le32	pad[4];
+	u32	sw_data[KNAV_DMA_NUM_SW_DATA_WORDS];
 } ____cacheline_aligned;
 
 #if IS_ENABLED(CONFIG_KEYSTONE_NAVIGATOR_DMA)
diff --git a/include/linux/stmmac.h b/include/linux/stmmac.h
index eead8ab93c0a..881a79d52467 100644
--- a/include/linux/stmmac.h
+++ b/include/linux/stmmac.h
@@ -100,6 +100,7 @@ struct plat_stmmacenet_data {
 	int interface;
 	struct stmmac_mdio_bus_data *mdio_bus_data;
 	struct device_node *phy_node;
+	struct device_node *mdio_node;
 	struct stmmac_dma_cfg *dma_cfg;
 	int clk_csr;
 	int has_gmac;
diff --git a/include/linux/trace_events.h b/include/linux/trace_events.h
index 429fdfc3baf5..925730bc9fc1 100644
--- a/include/linux/trace_events.h
+++ b/include/linux/trace_events.h
@@ -568,6 +568,8 @@ enum {
 	FILTER_DYN_STRING,
 	FILTER_PTR_STRING,
 	FILTER_TRACE_FN,
+	FILTER_COMM,
+	FILTER_CPU,
 };
 
 extern int trace_event_raw_init(struct trace_event_call *call);
diff --git a/include/linux/tracepoint.h b/include/linux/tracepoint.h
index acd522a91539..be586c632a0c 100644
--- a/include/linux/tracepoint.h
+++ b/include/linux/tracepoint.h
@@ -14,8 +14,10 @@
  * See the file COPYING for more details.
  */
 
+#include <linux/smp.h>
 #include <linux/errno.h>
 #include <linux/types.h>
+#include <linux/cpumask.h>
 #include <linux/rcupdate.h>
 #include <linux/tracepoint-defs.h>
 
@@ -338,15 +340,19 @@ extern void syscall_unregfunc(void);
  * "void *__data, proto" as the callback prototype.
  */
 #define DECLARE_TRACE_NOARGS(name)					\
-		__DECLARE_TRACE(name, void, , 1, void *__data, __data)
+	__DECLARE_TRACE(name, void, ,					\
+			cpu_online(raw_smp_processor_id()),		\
+			void *__data, __data)
 
 #define DECLARE_TRACE(name, proto, args)				\
-		__DECLARE_TRACE(name, PARAMS(proto), PARAMS(args), 1,	\
-				PARAMS(void *__data, proto),		\
-				PARAMS(__data, args))
+	__DECLARE_TRACE(name, PARAMS(proto), PARAMS(args),		\
+			cpu_online(raw_smp_processor_id()),		\
+			PARAMS(void *__data, proto),			\
+			PARAMS(__data, args))
 
 #define DECLARE_TRACE_CONDITION(name, proto, args, cond)		\
-	__DECLARE_TRACE(name, PARAMS(proto), PARAMS(args), PARAMS(cond), \
+	__DECLARE_TRACE(name, PARAMS(proto), PARAMS(args),		\
+			cpu_online(raw_smp_processor_id()) && (PARAMS(cond)), \
 			PARAMS(void *__data, proto),			\
 			PARAMS(__data, args))
 
diff --git a/include/linux/ucs2_string.h b/include/linux/ucs2_string.h
index cbb20afdbc01..bb679b48f408 100644
--- a/include/linux/ucs2_string.h
+++ b/include/linux/ucs2_string.h
@@ -11,4 +11,8 @@ unsigned long ucs2_strlen(const ucs2_char_t *s);
 unsigned long ucs2_strsize(const ucs2_char_t *data, unsigned long maxlength);
 int ucs2_strncmp(const ucs2_char_t *a, const ucs2_char_t *b, size_t len);
 
+unsigned long ucs2_utf8size(const ucs2_char_t *src);
+unsigned long ucs2_as_utf8(u8 *dest, const ucs2_char_t *src,
+			   unsigned long maxlength);
+
 #endif /* _LINUX_UCS2_STRING_H_ */
diff --git a/include/linux/workqueue.h b/include/linux/workqueue.h
index 0e32bc71245e..ca73c503b92a 100644
--- a/include/linux/workqueue.h
+++ b/include/linux/workqueue.h
@@ -311,6 +311,7 @@ enum {
 
 	__WQ_DRAINING		= 1 << 16, /* internal: workqueue is draining */
 	__WQ_ORDERED		= 1 << 17, /* internal: workqueue is ordered */
+	__WQ_LEGACY		= 1 << 18, /* internal: create*_workqueue() */
 
 	WQ_MAX_ACTIVE		= 512,	  /* I like 512, better ideas? */
 	WQ_MAX_UNBOUND_PER_CPU	= 4,	  /* 4 * #cpus for unbound wq */
@@ -411,12 +412,12 @@ __alloc_workqueue_key(const char *fmt, unsigned int flags, int max_active,
 	alloc_workqueue(fmt, WQ_UNBOUND | __WQ_ORDERED | (flags), 1, ##args)
 
 #define create_workqueue(name)						\
-	alloc_workqueue("%s", WQ_MEM_RECLAIM, 1, (name))
+	alloc_workqueue("%s", __WQ_LEGACY | WQ_MEM_RECLAIM, 1, (name))
 #define create_freezable_workqueue(name)				\
-	alloc_workqueue("%s", WQ_FREEZABLE | WQ_UNBOUND | WQ_MEM_RECLAIM, \
-			1, (name))
+	alloc_workqueue("%s", __WQ_LEGACY | WQ_FREEZABLE | WQ_UNBOUND |	\
+			WQ_MEM_RECLAIM, 1, (name))
 #define create_singlethread_workqueue(name)				\
-	alloc_ordered_workqueue("%s", WQ_MEM_RECLAIM, name)
+	alloc_ordered_workqueue("%s", __WQ_LEGACY | WQ_MEM_RECLAIM, name)
 
 extern void destroy_workqueue(struct workqueue_struct *wq);
 
diff --git a/include/linux/writeback.h b/include/linux/writeback.h
index b333c945e571..d0b5ca5d4e08 100644
--- a/include/linux/writeback.h
+++ b/include/linux/writeback.h
@@ -198,6 +198,7 @@ void wbc_attach_and_unlock_inode(struct writeback_control *wbc,
 void wbc_detach_inode(struct writeback_control *wbc);
 void wbc_account_io(struct writeback_control *wbc, struct page *page,
 		    size_t bytes);
+void cgroup_writeback_umount(void);
 
 /**
  * inode_attach_wb - associate an inode with its wb
@@ -301,6 +302,10 @@ static inline void wbc_account_io(struct writeback_control *wbc,
 {
 }
 
+static inline void cgroup_writeback_umount(void)
+{
+}
+
 #endif	/* CONFIG_CGROUP_WRITEBACK */
 
 /*
diff --git a/include/net/af_unix.h b/include/net/af_unix.h
index 2a91a0561a47..9b4c418bebd8 100644
--- a/include/net/af_unix.h
+++ b/include/net/af_unix.h
@@ -6,8 +6,8 @@
 #include <linux/mutex.h>
 #include <net/sock.h>
 
-void unix_inflight(struct file *fp);
-void unix_notinflight(struct file *fp);
+void unix_inflight(struct user_struct *user, struct file *fp);
+void unix_notinflight(struct user_struct *user, struct file *fp);
 void unix_gc(void);
 void wait_for_unix_gc(void);
 struct sock *unix_get_socket(struct file *filp);
diff --git a/include/net/inet_connection_sock.h b/include/net/inet_connection_sock.h
index 481fe1c9044c..49dcad4fe99e 100644
--- a/include/net/inet_connection_sock.h
+++ b/include/net/inet_connection_sock.h
@@ -270,8 +270,9 @@ struct dst_entry *inet_csk_route_child_sock(const struct sock *sk,
 					    struct sock *newsk,
 					    const struct request_sock *req);
 
-void inet_csk_reqsk_queue_add(struct sock *sk, struct request_sock *req,
-			      struct sock *child);
+struct sock *inet_csk_reqsk_queue_add(struct sock *sk,
+				      struct request_sock *req,
+				      struct sock *child);
 void inet_csk_reqsk_queue_hash_add(struct sock *sk, struct request_sock *req,
 				   unsigned long timeout);
 struct sock *inet_csk_complete_hashdance(struct sock *sk, struct sock *child,
diff --git a/include/net/ip_fib.h b/include/net/ip_fib.h
index 7029527725dd..4079fc18ffe4 100644
--- a/include/net/ip_fib.h
+++ b/include/net/ip_fib.h
@@ -61,6 +61,7 @@ struct fib_nh_exception {
 	struct rtable __rcu		*fnhe_rth_input;
 	struct rtable __rcu		*fnhe_rth_output;
 	unsigned long			fnhe_stamp;
+	struct rcu_head			rcu;
 };
 
 struct fnhe_hash_bucket {
diff --git a/include/net/ip_tunnels.h b/include/net/ip_tunnels.h
index 6db96ea0144f..dda9abf6b89c 100644
--- a/include/net/ip_tunnels.h
+++ b/include/net/ip_tunnels.h
@@ -230,6 +230,7 @@ void ip_tunnel_xmit(struct sk_buff *skb, struct net_device *dev,
 int ip_tunnel_ioctl(struct net_device *dev, struct ip_tunnel_parm *p, int cmd);
 int ip_tunnel_encap(struct sk_buff *skb, struct ip_tunnel *t,
 		    u8 *protocol, struct flowi4 *fl4);
+int __ip_tunnel_change_mtu(struct net_device *dev, int new_mtu, bool strict);
 int ip_tunnel_change_mtu(struct net_device *dev, int new_mtu);
 
 struct rtnl_link_stats64 *ip_tunnel_get_stats64(struct net_device *dev,
diff --git a/include/net/iw_handler.h b/include/net/iw_handler.h
index 8f81bbbc38fc..e0f4109e64c6 100644
--- a/include/net/iw_handler.h
+++ b/include/net/iw_handler.h
@@ -439,6 +439,12 @@ int dev_get_wireless_info(char *buffer, char **start, off_t offset, int length);
 /* Send a single event to user space */
 void wireless_send_event(struct net_device *dev, unsigned int cmd,
 			 union iwreq_data *wrqu, const char *extra);
+#ifdef CONFIG_WEXT_CORE
+/* flush all previous wext events - if work is done from netdev notifiers */
+void wireless_nlevent_flush(void);
+#else
+static inline void wireless_nlevent_flush(void) {}
+#endif
 
 /* We may need a function to send a stream of events to user space.
  * More on that later... */
diff --git a/include/net/scm.h b/include/net/scm.h
index 262532d111f5..59fa93c01d2a 100644
--- a/include/net/scm.h
+++ b/include/net/scm.h
@@ -21,6 +21,7 @@ struct scm_creds {
 struct scm_fp_list {
 	short			count;
 	short			max;
+	struct user_struct	*user;
 	struct file		*fp[SCM_MAX_FD];
 };
 
diff --git a/include/net/tcp.h b/include/net/tcp.h
index f6f8f032c73e..ae6468f5c9f3 100644
--- a/include/net/tcp.h
+++ b/include/net/tcp.h
@@ -447,7 +447,7 @@ const u8 *tcp_parse_md5sig_option(const struct tcphdr *th);
 
 void tcp_v4_send_check(struct sock *sk, struct sk_buff *skb);
 void tcp_v4_mtu_reduced(struct sock *sk);
-void tcp_req_err(struct sock *sk, u32 seq);
+void tcp_req_err(struct sock *sk, u32 seq, bool abort);
 int tcp_v4_conn_request(struct sock *sk, struct sk_buff *skb);
 struct sock *tcp_create_openreq_child(const struct sock *sk,
 				      struct request_sock *req,
diff --git a/include/sound/hdaudio.h b/include/sound/hdaudio.h
index e2b712c90d3f..c21c38ce7450 100644
--- a/include/sound/hdaudio.h
+++ b/include/sound/hdaudio.h
@@ -343,7 +343,7 @@ void snd_hdac_bus_enter_link_reset(struct hdac_bus *bus);
 void snd_hdac_bus_exit_link_reset(struct hdac_bus *bus);
 
 void snd_hdac_bus_update_rirb(struct hdac_bus *bus);
-void snd_hdac_bus_handle_stream_irq(struct hdac_bus *bus, unsigned int status,
+int snd_hdac_bus_handle_stream_irq(struct hdac_bus *bus, unsigned int status,
 				    void (*ack)(struct hdac_bus *,
 						struct hdac_stream *));
 
diff --git a/include/target/target_core_backend.h b/include/target/target_core_backend.h
index 56cf8e485ef2..28ee5c2e6bcd 100644
--- a/include/target/target_core_backend.h
+++ b/include/target/target_core_backend.h
@@ -94,5 +94,8 @@ sense_reason_t passthrough_parse_cdb(struct se_cmd *cmd,
 	sense_reason_t (*exec_cmd)(struct se_cmd *cmd));
 
 bool target_sense_desc_format(struct se_device *dev);
+sector_t target_to_linux_sector(struct se_device *dev, sector_t lb);
+bool target_configure_unmap_from_queue(struct se_dev_attrib *attrib,
+				       struct request_queue *q, int block_size);
 
 #endif /* TARGET_CORE_BACKEND_H */
diff --git a/include/target/target_core_base.h b/include/target/target_core_base.h
index 5d82816cc4e3..e8c8c08bf575 100644
--- a/include/target/target_core_base.h
+++ b/include/target/target_core_base.h
@@ -140,6 +140,8 @@ enum se_cmd_flags_table {
 	SCF_COMPARE_AND_WRITE		= 0x00080000,
 	SCF_COMPARE_AND_WRITE_POST	= 0x00100000,
 	SCF_PASSTHROUGH_PROT_SG_TO_MEM_NOALLOC = 0x00200000,
+	SCF_ACK_KREF			= 0x00400000,
+	SCF_USE_CPUID			= 0x00800000,
 };
 
 /* struct se_dev_entry->lun_flags and struct se_lun->lun_access */
@@ -187,6 +189,7 @@ enum target_sc_flags_table {
 	TARGET_SCF_BIDI_OP		= 0x01,
 	TARGET_SCF_ACK_KREF		= 0x02,
 	TARGET_SCF_UNKNOWN_SIZE		= 0x04,
+	TARGET_SCF_USE_CPUID	= 0x08,
 };
 
 /* fabric independent task management function values */
@@ -490,8 +493,9 @@ struct se_cmd {
 #define CMD_T_SENT		(1 << 4)
 #define CMD_T_STOP		(1 << 5)
 #define CMD_T_DEV_ACTIVE	(1 << 7)
-#define CMD_T_REQUEST_STOP	(1 << 8)
 #define CMD_T_BUSY		(1 << 9)
+#define CMD_T_TAS		(1 << 10)
+#define CMD_T_FABRIC_STOP	(1 << 11)
 	spinlock_t		t_state_lock;
 	struct kref		cmd_kref;
 	struct completion	t_transport_stop_comp;
@@ -511,9 +515,6 @@ struct se_cmd {
 
 	struct list_head	state_list;
 
-	/* old task stop completion, consider merging with some of the above */
-	struct completion	task_stop_comp;
-
 	/* backend private data */
 	void			*priv;
 
diff --git a/include/trace/events/asoc.h b/include/trace/events/asoc.h
index 317a1ed2f4ac..9130dd5a184a 100644
--- a/include/trace/events/asoc.h
+++ b/include/trace/events/asoc.h
@@ -231,13 +231,13 @@ TRACE_EVENT(snd_soc_jack_report,
 	TP_ARGS(jack, mask, val),
 
 	TP_STRUCT__entry(
-		__string(	name,		jack->jack->name	)
+		__string(	name,		jack->jack->id		)
 		__field(	int,		mask			)
 		__field(	int,		val			)
 	),
 
 	TP_fast_assign(
-		__assign_str(name, jack->jack->name);
+		__assign_str(name, jack->jack->id);
 		__entry->mask = mask;
 		__entry->val = val;
 	),
@@ -253,12 +253,12 @@ TRACE_EVENT(snd_soc_jack_notify,
 	TP_ARGS(jack, val),
 
 	TP_STRUCT__entry(
-		__string(	name,		jack->jack->name	)
+		__string(	name,		jack->jack->id		)
 		__field(	int,		val			)
 	),
 
 	TP_fast_assign(
-		__assign_str(name, jack->jack->name);
+		__assign_str(name, jack->jack->id);
 		__entry->val = val;
 	),
 
diff --git a/include/uapi/drm/exynos_drm.h b/include/uapi/drm/exynos_drm.h
index 312c67d744ae..3947c2eb8d69 100644
--- a/include/uapi/drm/exynos_drm.h
+++ b/include/uapi/drm/exynos_drm.h
@@ -28,8 +28,21 @@
  */
 struct drm_exynos_gem_create {
 	__u64 size;
-	unsigned int flags;
-	unsigned int handle;
+	__u32 flags;
+	__u32 handle;
+};
+
+/**
+ * A structure for getting a fake-offset that can be used with mmap.
+ *
+ * @handle: handle of gem object.
+ * @reserved: just padding to be 64-bit aligned.
+ * @offset: a fake-offset of gem object.
+ */
+struct drm_exynos_gem_map {
+	__u32 handle;
+	__u32 reserved;
+	__u64 offset;
 };
 
 /**
@@ -42,8 +55,8 @@ struct drm_exynos_gem_create {
  *	be set by driver.
  */
 struct drm_exynos_gem_info {
-	unsigned int handle;
-	unsigned int flags;
+	__u32 handle;
+	__u32 flags;
 	__u64 size;
 };
 
@@ -56,8 +69,8 @@ struct drm_exynos_gem_info {
  * @edid: the edid data pointer from user side.
  */
 struct drm_exynos_vidi_connection {
-	unsigned int connection;
-	unsigned int extensions;
+	__u32 connection;
+	__u32 extensions;
 	__u64 edid;
 };
 
@@ -206,9 +219,9 @@ struct drm_exynos_ipp_prop_list {
  * @pos: property of image position(src-cropped,dst-scaler).
  */
 struct drm_exynos_ipp_config {
-	enum drm_exynos_ops_id ops_id;
-	enum drm_exynos_flip	flip;
-	enum drm_exynos_degree	degree;
+	__u32 ops_id;
+	__u32 flip;
+	__u32 degree;
 	__u32	fmt;
 	struct drm_exynos_sz	sz;
 	struct drm_exynos_pos	pos;
@@ -233,7 +246,7 @@ enum drm_exynos_ipp_cmd {
  */
 struct drm_exynos_ipp_property {
 	struct drm_exynos_ipp_config config[EXYNOS_DRM_OPS_MAX];
-	enum drm_exynos_ipp_cmd	cmd;
+	__u32	cmd;
 	__u32	ipp_id;
 	__u32	prop_id;
 	__u32	refresh_rate;
@@ -255,8 +268,8 @@ enum drm_exynos_ipp_buf_type {
  * @user_data: user data.
  */
 struct drm_exynos_ipp_queue_buf {
-	enum drm_exynos_ops_id	ops_id;
-	enum drm_exynos_ipp_buf_type	buf_type;
+	__u32	ops_id;
+	__u32	buf_type;
 	__u32	prop_id;
 	__u32	buf_id;
 	__u32	handle[EXYNOS_DRM_PLANAR_MAX];
@@ -280,10 +293,11 @@ enum drm_exynos_ipp_ctrl {
  */
 struct drm_exynos_ipp_cmd_ctrl {
 	__u32	prop_id;
-	enum drm_exynos_ipp_ctrl	ctrl;
+	__u32	ctrl;
 };
 
 #define DRM_EXYNOS_GEM_CREATE		0x00
+#define DRM_EXYNOS_GEM_MAP		0x01
 /* Reserved 0x03 ~ 0x05 for exynos specific gem ioctl */
 #define DRM_EXYNOS_GEM_GET		0x04
 #define DRM_EXYNOS_VIDI_CONNECTION	0x07
@@ -301,7 +315,8 @@ struct drm_exynos_ipp_cmd_ctrl {
 
 #define DRM_IOCTL_EXYNOS_GEM_CREATE		DRM_IOWR(DRM_COMMAND_BASE + \
 		DRM_EXYNOS_GEM_CREATE, struct drm_exynos_gem_create)
-
+#define DRM_IOCTL_EXYNOS_GEM_MAP		DRM_IOWR(DRM_COMMAND_BASE + \
+		DRM_EXYNOS_GEM_MAP, struct drm_exynos_gem_map)
 #define DRM_IOCTL_EXYNOS_GEM_GET	DRM_IOWR(DRM_COMMAND_BASE + \
 		DRM_EXYNOS_GEM_GET,	struct drm_exynos_gem_info)
 
diff --git a/include/uapi/drm/i915_drm.h b/include/uapi/drm/i915_drm.h
index 6a19371391fa..a5524cc95ff8 100644
--- a/include/uapi/drm/i915_drm.h
+++ b/include/uapi/drm/i915_drm.h
@@ -772,10 +772,12 @@ struct drm_i915_gem_execbuffer2 {
 #define I915_EXEC_HANDLE_LUT		(1<<12)
 
 /** Used for switching BSD rings on the platforms with two BSD rings */
-#define I915_EXEC_BSD_MASK		(3<<13)
-#define I915_EXEC_BSD_DEFAULT		(0<<13) /* default ping-pong mode */
-#define I915_EXEC_BSD_RING1		(1<<13)
-#define I915_EXEC_BSD_RING2		(2<<13)
+#define I915_EXEC_BSD_SHIFT	 (13)
+#define I915_EXEC_BSD_MASK	 (3 << I915_EXEC_BSD_SHIFT)
+/* default ping-pong mode */
+#define I915_EXEC_BSD_DEFAULT	 (0 << I915_EXEC_BSD_SHIFT)
+#define I915_EXEC_BSD_RING1	 (1 << I915_EXEC_BSD_SHIFT)
+#define I915_EXEC_BSD_RING2	 (2 << I915_EXEC_BSD_SHIFT)
 
 /** Tell the kernel that the batchbuffer is processed by
  *  the resource streamer.
diff --git a/include/uapi/drm/msm_drm.h b/include/uapi/drm/msm_drm.h
index 0e3c4c48b326..254d3e92d18e 100644
--- a/include/uapi/drm/msm_drm.h
+++ b/include/uapi/drm/msm_drm.h
@@ -51,6 +51,7 @@ struct drm_msm_timespec {
 #define MSM_PARAM_GMEM_SIZE  0x02
 #define MSM_PARAM_CHIP_ID    0x03
 #define MSM_PARAM_MAX_FREQ   0x04
+#define MSM_PARAM_TIMESTAMP  0x05
 
 struct drm_msm_param {
 	__u32 pipe;           /* in, MSM_PIPE_x */
diff --git a/include/uapi/linux/bpf.h b/include/uapi/linux/bpf.h
index aa6f8571de13..5df4881dea7b 100644
--- a/include/uapi/linux/bpf.h
+++ b/include/uapi/linux/bpf.h
@@ -292,6 +292,9 @@ enum bpf_func_id {
 /* BPF_FUNC_skb_set_tunnel_key and BPF_FUNC_skb_get_tunnel_key flags. */
 #define BPF_F_TUNINFO_IPV6		(1ULL << 0)
 
+/* BPF_FUNC_skb_set_tunnel_key flags. */
+#define BPF_F_ZERO_CSUM_TX		(1ULL << 1)
+
 /* user accessible mirror of in-kernel sk_buff.
  * new fields can only be added to the end of this structure
  */
diff --git a/include/uapi/linux/media.h b/include/uapi/linux/media.h
index 6aac2f035a5d..a2f1af049370 100644
--- a/include/uapi/linux/media.h
+++ b/include/uapi/linux/media.h
@@ -66,17 +66,17 @@ struct media_device_info {
 /*
  * DVB entities
  */
-#define MEDIA_ENT_F_DTV_DEMOD		(MEDIA_ENT_F_BASE + 1)
-#define MEDIA_ENT_F_TS_DEMUX		(MEDIA_ENT_F_BASE + 2)
-#define MEDIA_ENT_F_DTV_CA		(MEDIA_ENT_F_BASE + 3)
-#define MEDIA_ENT_F_DTV_NET_DECAP	(MEDIA_ENT_F_BASE + 4)
+#define MEDIA_ENT_F_DTV_DEMOD		(MEDIA_ENT_F_BASE + 0x00001)
+#define MEDIA_ENT_F_TS_DEMUX		(MEDIA_ENT_F_BASE + 0x00002)
+#define MEDIA_ENT_F_DTV_CA		(MEDIA_ENT_F_BASE + 0x00003)
+#define MEDIA_ENT_F_DTV_NET_DECAP	(MEDIA_ENT_F_BASE + 0x00004)
 
 /*
  * I/O entities
  */
-#define MEDIA_ENT_F_IO_DTV		(MEDIA_ENT_F_BASE + 1001)
-#define MEDIA_ENT_F_IO_VBI		(MEDIA_ENT_F_BASE + 1002)
-#define MEDIA_ENT_F_IO_SWRADIO		(MEDIA_ENT_F_BASE + 1003)
+#define MEDIA_ENT_F_IO_DTV		(MEDIA_ENT_F_BASE + 0x01001)
+#define MEDIA_ENT_F_IO_VBI		(MEDIA_ENT_F_BASE + 0x01002)
+#define MEDIA_ENT_F_IO_SWRADIO		(MEDIA_ENT_F_BASE + 0x01003)
 
 /*
  * Analog TV IF-PLL decoders
@@ -91,9 +91,17 @@ struct media_device_info {
  * Connectors
  */
 /* It is a responsibility of the entity drivers to add connectors and links */
-#define MEDIA_ENT_F_CONN_RF		(MEDIA_ENT_F_BASE + 10001)
-#define MEDIA_ENT_F_CONN_SVIDEO		(MEDIA_ENT_F_BASE + 10002)
-#define MEDIA_ENT_F_CONN_COMPOSITE	(MEDIA_ENT_F_BASE + 10003)
+#ifdef __KERNEL__
+	/*
+	 * For now, it should not be used in userspace, as some
+	 * definitions may change
+	 */
+
+#define MEDIA_ENT_F_CONN_RF		(MEDIA_ENT_F_BASE + 0x30001)
+#define MEDIA_ENT_F_CONN_SVIDEO		(MEDIA_ENT_F_BASE + 0x30002)
+#define MEDIA_ENT_F_CONN_COMPOSITE	(MEDIA_ENT_F_BASE + 0x30003)
+
+#endif
 
 /*
  * Don't touch on those. The ranges MEDIA_ENT_F_OLD_BASE and
@@ -125,7 +133,7 @@ struct media_device_info {
 
 #define MEDIA_ENT_F_V4L2_SUBDEV_UNKNOWN	MEDIA_ENT_F_OLD_SUBDEV_BASE
 
-#ifndef __KERNEL__
+#if !defined(__KERNEL__) || defined(__NEED_MEDIA_LEGACY_API)
 
 /*
  * Legacy symbols used to avoid userspace compilation breakages
@@ -138,6 +146,10 @@ struct media_device_info {
 #define MEDIA_ENT_TYPE_MASK		0x00ff0000
 #define MEDIA_ENT_SUBTYPE_MASK		0x0000ffff
 
+/* End of the old subdev reserved numberspace */
+#define MEDIA_ENT_T_DEVNODE_UNKNOWN	(MEDIA_ENT_T_DEVNODE | \
+					 MEDIA_ENT_SUBTYPE_MASK)
+
 #define MEDIA_ENT_T_DEVNODE		MEDIA_ENT_F_OLD_BASE
 #define MEDIA_ENT_T_DEVNODE_V4L		MEDIA_ENT_F_IO_V4L
 #define MEDIA_ENT_T_DEVNODE_FB		(MEDIA_ENT_T_DEVNODE + 2)
@@ -302,14 +314,14 @@ struct media_v2_entity {
 	__u32 id;
 	char name[64];		/* FIXME: move to a property? (RFC says so) */
 	__u32 function;		/* Main function of the entity */
-	__u16 reserved[12];
-};
+	__u32 reserved[6];
+} __attribute__ ((packed));
 
 /* Should match the specific fields at media_intf_devnode */
 struct media_v2_intf_devnode {
 	__u32 major;
 	__u32 minor;
-};
+} __attribute__ ((packed));
 
 struct media_v2_interface {
 	__u32 id;
@@ -321,22 +333,22 @@ struct media_v2_interface {
 		struct media_v2_intf_devnode devnode;
 		__u32 raw[16];
 	};
-};
+} __attribute__ ((packed));
 
 struct media_v2_pad {
 	__u32 id;
 	__u32 entity_id;
 	__u32 flags;
-	__u16 reserved[9];
-};
+	__u32 reserved[5];
+} __attribute__ ((packed));
 
 struct media_v2_link {
 	__u32 id;
 	__u32 source_id;
 	__u32 sink_id;
 	__u32 flags;
-	__u32 reserved[5];
-};
+	__u32 reserved[6];
+} __attribute__ ((packed));
 
 struct media_v2_topology {
 	__u64 topology_version;
@@ -356,7 +368,7 @@ struct media_v2_topology {
 	__u32 num_links;
 	__u32 reserved4;
 	__u64 ptr_links;
-};
+} __attribute__ ((packed));
 
 static inline void __user *media_get_uptr(__u64 arg)
 {
diff --git a/include/uapi/linux/ndctl.h b/include/uapi/linux/ndctl.h
index 5b4a4be06e2b..cc68b92124d4 100644
--- a/include/uapi/linux/ndctl.h
+++ b/include/uapi/linux/ndctl.h
@@ -66,14 +66,18 @@ struct nd_cmd_ars_cap {
 	__u64 length;
 	__u32 status;
 	__u32 max_ars_out;
+	__u32 clear_err_unit;
+	__u32 reserved;
 } __packed;
 
 struct nd_cmd_ars_start {
 	__u64 address;
 	__u64 length;
 	__u16 type;
-	__u8 reserved[6];
+	__u8 flags;
+	__u8 reserved[5];
 	__u32 status;
+	__u32 scrub_time;
 } __packed;
 
 struct nd_cmd_ars_status {
@@ -81,11 +85,14 @@ struct nd_cmd_ars_status {
 	__u32 out_length;
 	__u64 address;
 	__u64 length;
+	__u64 restart_address;
+	__u64 restart_length;
 	__u16 type;
+	__u16 flags;
 	__u32 num_records;
 	struct nd_ars_record {
 		__u32 handle;
-		__u32 flags;
+		__u32 reserved;
 		__u64 err_address;
 		__u64 length;
 	} __packed records[0];
diff --git a/include/video/omap-panel-data.h b/include/video/omap-panel-data.h
index 69279c013ac4..56830d1dc762 100644
--- a/include/video/omap-panel-data.h
+++ b/include/video/omap-panel-data.h
@@ -45,21 +45,6 @@ struct encoder_tfp410_platform_data {
 	int data_lines;
 };
 
-/**
- * encoder_tpd12s015 platform data
- * @name: name for this display entity
- * @ct_cp_hpd_gpio: CT_CP_HPD gpio number
- * @ls_oe_gpio: LS_OE gpio number
- * @hpd_gpio: HPD gpio number
- */
-struct encoder_tpd12s015_platform_data {
-	const char *name;
-	const char *source;
-
-	int ct_cp_hpd_gpio;
-	int ls_oe_gpio;
-	int hpd_gpio;
-};
 
 /**
  * connector_dvi platform data
diff --git a/include/video/omapdss.h b/include/video/omapdss.h
index 295b41e20d8e..8e14ad7327c9 100644
--- a/include/video/omapdss.h
+++ b/include/video/omapdss.h
@@ -51,6 +51,7 @@
 #define DISPC_IRQ_FRAMEDONEWB		(1 << 23)
 #define DISPC_IRQ_FRAMEDONETV		(1 << 24)
 #define DISPC_IRQ_WBBUFFEROVERFLOW	(1 << 25)
+#define DISPC_IRQ_WBUNCOMPLETEERROR	(1 << 26)
 #define DISPC_IRQ_SYNC_LOST3		(1 << 27)
 #define DISPC_IRQ_VSYNC3		(1 << 28)
 #define DISPC_IRQ_ACBIAS_COUNT_STAT3	(1 << 29)
@@ -331,8 +332,6 @@ struct omap_dss_board_info {
 
 /* Init with the board info */
 extern int omap_display_init(struct omap_dss_board_info *board_data);
-/* HDMI mux init*/
-extern int omap_hdmi_init(enum omap_hdmi_flags flags);
 
 struct omap_video_timings {
 	/* Unit: pixels */
@@ -366,6 +365,8 @@ struct omap_video_timings {
 	enum omap_dss_signal_level de_level;
 	/* Pixel clock edges to drive HSYNC and VSYNC signals */
 	enum omap_dss_signal_edge sync_pclk_edge;
+
+	bool double_pixel;
 };
 
 /* Hardcoded timings for tv modes. Venc only uses these to
@@ -769,6 +770,7 @@ struct omap_dss_device {
 
 	/* DISPC channel for this output */
 	enum omap_channel dispc_channel;
+	bool dispc_channel_connected;
 
 	/* output instance */
 	enum omap_dss_output_id id;
@@ -782,13 +784,6 @@ struct omap_dss_device {
 	struct omap_dss_device *dst;
 };
 
-struct omap_dss_hdmi_data
-{
-	int ct_cp_hpd_gpio;
-	int ls_oe_gpio;
-	int hpd_gpio;
-};
-
 struct omap_dss_driver {
 	int (*probe)(struct omap_dss_device *);
 	void (*remove)(struct omap_dss_device *);
@@ -897,85 +892,9 @@ typedef void (*omap_dispc_isr_t) (void *arg, u32 mask);
 int omap_dispc_register_isr(omap_dispc_isr_t isr, void *arg, u32 mask);
 int omap_dispc_unregister_isr(omap_dispc_isr_t isr, void *arg, u32 mask);
 
-u32 dispc_read_irqstatus(void);
-void dispc_clear_irqstatus(u32 mask);
-u32 dispc_read_irqenable(void);
-void dispc_write_irqenable(u32 mask);
-
-int dispc_request_irq(irq_handler_t handler, void *dev_id);
-void dispc_free_irq(void *dev_id);
-
-int dispc_runtime_get(void);
-void dispc_runtime_put(void);
-
-void dispc_mgr_enable(enum omap_channel channel, bool enable);
-bool dispc_mgr_is_enabled(enum omap_channel channel);
-u32 dispc_mgr_get_vsync_irq(enum omap_channel channel);
-u32 dispc_mgr_get_framedone_irq(enum omap_channel channel);
-u32 dispc_mgr_get_sync_lost_irq(enum omap_channel channel);
-bool dispc_mgr_go_busy(enum omap_channel channel);
-void dispc_mgr_go(enum omap_channel channel);
-void dispc_mgr_set_lcd_config(enum omap_channel channel,
-		const struct dss_lcd_mgr_config *config);
-void dispc_mgr_set_timings(enum omap_channel channel,
-		const struct omap_video_timings *timings);
-void dispc_mgr_setup(enum omap_channel channel,
-		const struct omap_overlay_manager_info *info);
-
-int dispc_ovl_check(enum omap_plane plane, enum omap_channel channel,
-		const struct omap_overlay_info *oi,
-		const struct omap_video_timings *timings,
-		int *x_predecim, int *y_predecim);
-
-int dispc_ovl_enable(enum omap_plane plane, bool enable);
-bool dispc_ovl_enabled(enum omap_plane plane);
-void dispc_ovl_set_channel_out(enum omap_plane plane,
-		enum omap_channel channel);
-int dispc_ovl_setup(enum omap_plane plane, const struct omap_overlay_info *oi,
-		bool replication, const struct omap_video_timings *mgr_timings,
-		bool mem_to_mem);
-
 int omapdss_compat_init(void);
 void omapdss_compat_uninit(void);
 
-struct dss_mgr_ops {
-	int (*connect)(struct omap_overlay_manager *mgr,
-		struct omap_dss_device *dst);
-	void (*disconnect)(struct omap_overlay_manager *mgr,
-		struct omap_dss_device *dst);
-
-	void (*start_update)(struct omap_overlay_manager *mgr);
-	int (*enable)(struct omap_overlay_manager *mgr);
-	void (*disable)(struct omap_overlay_manager *mgr);
-	void (*set_timings)(struct omap_overlay_manager *mgr,
-			const struct omap_video_timings *timings);
-	void (*set_lcd_config)(struct omap_overlay_manager *mgr,
-			const struct dss_lcd_mgr_config *config);
-	int (*register_framedone_handler)(struct omap_overlay_manager *mgr,
-			void (*handler)(void *), void *data);
-	void (*unregister_framedone_handler)(struct omap_overlay_manager *mgr,
-			void (*handler)(void *), void *data);
-};
-
-int dss_install_mgr_ops(const struct dss_mgr_ops *mgr_ops);
-void dss_uninstall_mgr_ops(void);
-
-int dss_mgr_connect(struct omap_overlay_manager *mgr,
-		struct omap_dss_device *dst);
-void dss_mgr_disconnect(struct omap_overlay_manager *mgr,
-		struct omap_dss_device *dst);
-void dss_mgr_set_timings(struct omap_overlay_manager *mgr,
-		const struct omap_video_timings *timings);
-void dss_mgr_set_lcd_config(struct omap_overlay_manager *mgr,
-		const struct dss_lcd_mgr_config *config);
-int dss_mgr_enable(struct omap_overlay_manager *mgr);
-void dss_mgr_disable(struct omap_overlay_manager *mgr);
-void dss_mgr_start_update(struct omap_overlay_manager *mgr);
-int dss_mgr_register_framedone_handler(struct omap_overlay_manager *mgr,
-		void (*handler)(void *), void *data);
-void dss_mgr_unregister_framedone_handler(struct omap_overlay_manager *mgr,
-		void (*handler)(void *), void *data);
-
 static inline bool omapdss_device_is_connected(struct omap_dss_device *dssdev)
 {
 	return dssdev->src;
diff --git a/ipc/shm.c b/ipc/shm.c
index ed3027d0f277..331fc1b0b3c7 100644
--- a/ipc/shm.c
+++ b/ipc/shm.c
@@ -156,11 +156,12 @@ static inline struct shmid_kernel *shm_lock(struct ipc_namespace *ns, int id)
 	struct kern_ipc_perm *ipcp = ipc_lock(&shm_ids(ns), id);
 
 	/*
-	 * We raced in the idr lookup or with shm_destroy().  Either way, the
-	 * ID is busted.
+	 * Callers of shm_lock() must validate the status of the returned ipc
+	 * object pointer (as returned by ipc_lock()), and error out as
+	 * appropriate.
 	 */
-	WARN_ON(IS_ERR(ipcp));
-
+	if (IS_ERR(ipcp))
+		return (void *)ipcp;
 	return container_of(ipcp, struct shmid_kernel, shm_perm);
 }
 
@@ -186,18 +187,33 @@ static inline void shm_rmid(struct ipc_namespace *ns, struct shmid_kernel *s)
 }
 
 
-/* This is called by fork, once for every shm attach. */
-static void shm_open(struct vm_area_struct *vma)
+static int __shm_open(struct vm_area_struct *vma)
 {
 	struct file *file = vma->vm_file;
 	struct shm_file_data *sfd = shm_file_data(file);
 	struct shmid_kernel *shp;
 
 	shp = shm_lock(sfd->ns, sfd->id);
+
+	if (IS_ERR(shp))
+		return PTR_ERR(shp);
+
 	shp->shm_atim = get_seconds();
 	shp->shm_lprid = task_tgid_vnr(current);
 	shp->shm_nattch++;
 	shm_unlock(shp);
+	return 0;
+}
+
+/* This is called by fork, once for every shm attach. */
+static void shm_open(struct vm_area_struct *vma)
+{
+	int err = __shm_open(vma);
+	/*
+	 * We raced in the idr lookup or with shm_destroy().
+	 * Either way, the ID is busted.
+	 */
+	WARN_ON_ONCE(err);
 }
 
 /*
@@ -260,6 +276,14 @@ static void shm_close(struct vm_area_struct *vma)
 	down_write(&shm_ids(ns).rwsem);
 	/* remove from the list of attaches of the shm segment */
 	shp = shm_lock(ns, sfd->id);
+
+	/*
+	 * We raced in the idr lookup or with shm_destroy().
+	 * Either way, the ID is busted.
+	 */
+	if (WARN_ON_ONCE(IS_ERR(shp)))
+		goto done; /* no-op */
+
 	shp->shm_lprid = task_tgid_vnr(current);
 	shp->shm_dtim = get_seconds();
 	shp->shm_nattch--;
@@ -267,6 +291,7 @@ static void shm_close(struct vm_area_struct *vma)
 		shm_destroy(ns, shp);
 	else
 		shm_unlock(shp);
+done:
 	up_write(&shm_ids(ns).rwsem);
 }
 
@@ -388,17 +413,25 @@ static int shm_mmap(struct file *file, struct vm_area_struct *vma)
 	struct shm_file_data *sfd = shm_file_data(file);
 	int ret;
 
+	/*
+	 * In case of remap_file_pages() emulation, the file can represent
+	 * removed IPC ID: propogate shm_lock() error to caller.
+	 */
+	ret =__shm_open(vma);
+	if (ret)
+		return ret;
+
 	ret = sfd->file->f_op->mmap(sfd->file, vma);
-	if (ret != 0)
+	if (ret) {
+		shm_close(vma);
 		return ret;
+	}
 	sfd->vm_ops = vma->vm_ops;
 #ifdef CONFIG_MMU
 	WARN_ON(!sfd->vm_ops->fault);
 #endif
 	vma->vm_ops = &shm_vm_ops;
-	shm_open(vma);
-
-	return ret;
+	return 0;
 }
 
 static int shm_release(struct inode *ino, struct file *file)
diff --git a/kernel/bpf/verifier.c b/kernel/bpf/verifier.c
index d1d3e8f57de9..2e7f7ab739e4 100644
--- a/kernel/bpf/verifier.c
+++ b/kernel/bpf/verifier.c
@@ -2082,7 +2082,7 @@ static void adjust_branches(struct bpf_prog *prog, int pos, int delta)
 		/* adjust offset of jmps if necessary */
 		if (i < pos && i + insn->off + 1 > pos)
 			insn->off += delta;
-		else if (i > pos && i + insn->off + 1 < pos)
+		else if (i > pos + delta && i + insn->off + 1 <= pos + delta)
 			insn->off -= delta;
 	}
 }
diff --git a/kernel/cgroup.c b/kernel/cgroup.c
index c03a640ef6da..d27904c193da 100644
--- a/kernel/cgroup.c
+++ b/kernel/cgroup.c
@@ -58,6 +58,7 @@
 #include <linux/kthread.h>
 #include <linux/delay.h>
 #include <linux/atomic.h>
+#include <linux/cpuset.h>
 #include <net/sock.h>
 
 /*
@@ -2739,6 +2740,7 @@ out_unlock_rcu:
 out_unlock_threadgroup:
 	percpu_up_write(&cgroup_threadgroup_rwsem);
 	cgroup_kn_unlock(of->kn);
+	cpuset_post_attach_flush();
 	return ret ?: nbytes;
 }
 
@@ -4655,14 +4657,15 @@ static void css_free_work_fn(struct work_struct *work)
 
 	if (ss) {
 		/* css free path */
+		struct cgroup_subsys_state *parent = css->parent;
 		int id = css->id;
 
-		if (css->parent)
-			css_put(css->parent);
-
 		ss->css_free(css);
 		cgroup_idr_remove(&ss->css_idr, id);
 		cgroup_put(cgrp);
+
+		if (parent)
+			css_put(parent);
 	} else {
 		/* cgroup free path */
 		atomic_dec(&cgrp->root->nr_cgrps);
@@ -4758,6 +4761,7 @@ static void init_and_link_css(struct cgroup_subsys_state *css,
 	INIT_LIST_HEAD(&css->sibling);
 	INIT_LIST_HEAD(&css->children);
 	css->serial_nr = css_serial_nr_next++;
+	atomic_set(&css->online_cnt, 0);
 
 	if (cgroup_parent(cgrp)) {
 		css->parent = cgroup_css(cgroup_parent(cgrp), ss);
@@ -4780,6 +4784,10 @@ static int online_css(struct cgroup_subsys_state *css)
 	if (!ret) {
 		css->flags |= CSS_ONLINE;
 		rcu_assign_pointer(css->cgroup->subsys[ss->id], css);
+
+		atomic_inc(&css->online_cnt);
+		if (css->parent)
+			atomic_inc(&css->parent->online_cnt);
 	}
 	return ret;
 }
@@ -5017,10 +5025,15 @@ static void css_killed_work_fn(struct work_struct *work)
 		container_of(work, struct cgroup_subsys_state, destroy_work);
 
 	mutex_lock(&cgroup_mutex);
-	offline_css(css);
-	mutex_unlock(&cgroup_mutex);
 
-	css_put(css);
+	do {
+		offline_css(css);
+		css_put(css);
+		/* @css can't go away while we're holding cgroup_mutex */
+		css = css->parent;
+	} while (css && atomic_dec_and_test(&css->online_cnt));
+
+	mutex_unlock(&cgroup_mutex);
 }
 
 /* css kill confirmation processing requires process context, bounce */
@@ -5029,8 +5042,10 @@ static void css_killed_ref_fn(struct percpu_ref *ref)
 	struct cgroup_subsys_state *css =
 		container_of(ref, struct cgroup_subsys_state, refcnt);
 
-	INIT_WORK(&css->destroy_work, css_killed_work_fn);
-	queue_work(cgroup_destroy_wq, &css->destroy_work);
+	if (atomic_dec_and_test(&css->online_cnt)) {
+		INIT_WORK(&css->destroy_work, css_killed_work_fn);
+		queue_work(cgroup_destroy_wq, &css->destroy_work);
+	}
 }
 
 /**
diff --git a/kernel/cpuset.c b/kernel/cpuset.c
index 3e945fcd8179..41989ab4db57 100644
--- a/kernel/cpuset.c
+++ b/kernel/cpuset.c
@@ -287,6 +287,8 @@ static struct cpuset top_cpuset = {
 static DEFINE_MUTEX(cpuset_mutex);
 static DEFINE_SPINLOCK(callback_lock);
 
+static struct workqueue_struct *cpuset_migrate_mm_wq;
+
 /*
  * CPU / memory hotplug is handled asynchronously.
  */
@@ -972,31 +974,51 @@ static int update_cpumask(struct cpuset *cs, struct cpuset *trialcs,
 }
 
 /*
- * cpuset_migrate_mm
- *
- *    Migrate memory region from one set of nodes to another.
- *
- *    Temporarilly set tasks mems_allowed to target nodes of migration,
- *    so that the migration code can allocate pages on these nodes.
- *
- *    While the mm_struct we are migrating is typically from some
- *    other task, the task_struct mems_allowed that we are hacking
- *    is for our current task, which must allocate new pages for that
- *    migrating memory region.
+ * Migrate memory region from one set of nodes to another.  This is
+ * performed asynchronously as it can be called from process migration path
+ * holding locks involved in process management.  All mm migrations are
+ * performed in the queued order and can be waited for by flushing
+ * cpuset_migrate_mm_wq.
  */
 
+struct cpuset_migrate_mm_work {
+	struct work_struct	work;
+	struct mm_struct	*mm;
+	nodemask_t		from;
+	nodemask_t		to;
+};
+
+static void cpuset_migrate_mm_workfn(struct work_struct *work)
+{
+	struct cpuset_migrate_mm_work *mwork =
+		container_of(work, struct cpuset_migrate_mm_work, work);
+
+	/* on a wq worker, no need to worry about %current's mems_allowed */
+	do_migrate_pages(mwork->mm, &mwork->from, &mwork->to, MPOL_MF_MOVE_ALL);
+	mmput(mwork->mm);
+	kfree(mwork);
+}
+
 static void cpuset_migrate_mm(struct mm_struct *mm, const nodemask_t *from,
 							const nodemask_t *to)
 {
-	struct task_struct *tsk = current;
-
-	tsk->mems_allowed = *to;
+	struct cpuset_migrate_mm_work *mwork;
 
-	do_migrate_pages(mm, from, to, MPOL_MF_MOVE_ALL);
+	mwork = kzalloc(sizeof(*mwork), GFP_KERNEL);
+	if (mwork) {
+		mwork->mm = mm;
+		mwork->from = *from;
+		mwork->to = *to;
+		INIT_WORK(&mwork->work, cpuset_migrate_mm_workfn);
+		queue_work(cpuset_migrate_mm_wq, &mwork->work);
+	} else {
+		mmput(mm);
+	}
+}
 
-	rcu_read_lock();
-	guarantee_online_mems(task_cs(tsk), &tsk->mems_allowed);
-	rcu_read_unlock();
+void cpuset_post_attach_flush(void)
+{
+	flush_workqueue(cpuset_migrate_mm_wq);
 }
 
 /*
@@ -1097,7 +1119,8 @@ static void update_tasks_nodemask(struct cpuset *cs)
 		mpol_rebind_mm(mm, &cs->mems_allowed);
 		if (migrate)
 			cpuset_migrate_mm(mm, &cs->old_mems_allowed, &newmems);
-		mmput(mm);
+		else
+			mmput(mm);
 	}
 	css_task_iter_end(&it);
 
@@ -1545,11 +1568,11 @@ static void cpuset_attach(struct cgroup_taskset *tset)
 			 * @old_mems_allowed is the right nodesets that we
 			 * migrate mm from.
 			 */
-			if (is_memory_migrate(cs)) {
+			if (is_memory_migrate(cs))
 				cpuset_migrate_mm(mm, &oldcs->old_mems_allowed,
 						  &cpuset_attach_nodemask_to);
-			}
-			mmput(mm);
+			else
+				mmput(mm);
 		}
 	}
 
@@ -1714,6 +1737,7 @@ out_unlock:
 	mutex_unlock(&cpuset_mutex);
 	kernfs_unbreak_active_protection(of->kn);
 	css_put(&cs->css);
+	flush_workqueue(cpuset_migrate_mm_wq);
 	return retval ?: nbytes;
 }
 
@@ -2359,6 +2383,9 @@ void __init cpuset_init_smp(void)
 	top_cpuset.effective_mems = node_states[N_MEMORY];
 
 	register_hotmemory_notifier(&cpuset_track_online_nodes_nb);
+
+	cpuset_migrate_mm_wq = alloc_ordered_workqueue("cpuset_migrate_mm", 0);
+	BUG_ON(!cpuset_migrate_mm_wq);
 }
 
 /**
diff --git a/kernel/events/core.c b/kernel/events/core.c
index 5946460b2425..614614821f00 100644
--- a/kernel/events/core.c
+++ b/kernel/events/core.c
@@ -64,8 +64,17 @@ static void remote_function(void *data)
 	struct task_struct *p = tfc->p;
 
 	if (p) {
-		tfc->ret = -EAGAIN;
-		if (task_cpu(p) != smp_processor_id() || !task_curr(p))
+		/* -EAGAIN */
+		if (task_cpu(p) != smp_processor_id())
+			return;
+
+		/*
+		 * Now that we're on right CPU with IRQs disabled, we can test
+		 * if we hit the right task without races.
+		 */
+
+		tfc->ret = -ESRCH; /* No such (running) process */
+		if (p != current)
 			return;
 	}
 
@@ -92,13 +101,17 @@ task_function_call(struct task_struct *p, remote_function_f func, void *info)
 		.p	= p,
 		.func	= func,
 		.info	= info,
-		.ret	= -ESRCH, /* No such (running) process */
+		.ret	= -EAGAIN,
 	};
+	int ret;
 
-	if (task_curr(p))
-		smp_call_function_single(task_cpu(p), remote_function, &data, 1);
+	do {
+		ret = smp_call_function_single(task_cpu(p), remote_function, &data, 1);
+		if (!ret)
+			ret = data.ret;
+	} while (ret == -EAGAIN);
 
-	return data.ret;
+	return ret;
 }
 
 /**
@@ -169,19 +182,6 @@ static bool is_kernel_event(struct perf_event *event)
  *    rely on ctx->is_active and therefore cannot use event_function_call().
  *    See perf_install_in_context().
  *
- * This is because we need a ctx->lock serialized variable (ctx->is_active)
- * to reliably determine if a particular task/context is scheduled in. The
- * task_curr() use in task_function_call() is racy in that a remote context
- * switch is not a single atomic operation.
- *
- * As is, the situation is 'safe' because we set rq->curr before we do the
- * actual context switch. This means that task_curr() will fail early, but
- * we'll continue spinning on ctx->is_active until we've passed
- * perf_event_task_sched_out().
- *
- * Without this ctx->lock serialized variable we could have race where we find
- * the task (and hence the context) would not be active while in fact they are.
- *
  * If ctx->nr_events, then ctx->is_active and cpuctx->task_ctx are set.
  */
 
@@ -212,7 +212,7 @@ static int event_function(void *info)
 	 */
 	if (ctx->task) {
 		if (ctx->task != current) {
-			ret = -EAGAIN;
+			ret = -ESRCH;
 			goto unlock;
 		}
 
@@ -276,10 +276,10 @@ static void event_function_call(struct perf_event *event, event_f func, void *da
 		return;
 	}
 
-again:
 	if (task == TASK_TOMBSTONE)
 		return;
 
+again:
 	if (!task_function_call(task, event_function, &efs))
 		return;
 
@@ -289,13 +289,15 @@ again:
 	 * a concurrent perf_event_context_sched_out().
 	 */
 	task = ctx->task;
-	if (task != TASK_TOMBSTONE) {
-		if (ctx->is_active) {
-			raw_spin_unlock_irq(&ctx->lock);
-			goto again;
-		}
-		func(event, NULL, ctx, data);
+	if (task == TASK_TOMBSTONE) {
+		raw_spin_unlock_irq(&ctx->lock);
+		return;
 	}
+	if (ctx->is_active) {
+		raw_spin_unlock_irq(&ctx->lock);
+		goto again;
+	}
+	func(event, NULL, ctx, data);
 	raw_spin_unlock_irq(&ctx->lock);
 }
 
@@ -314,6 +316,7 @@ again:
 enum event_type_t {
 	EVENT_FLEXIBLE = 0x1,
 	EVENT_PINNED = 0x2,
+	EVENT_TIME = 0x4,
 	EVENT_ALL = EVENT_FLEXIBLE | EVENT_PINNED,
 };
 
@@ -321,7 +324,13 @@ enum event_type_t {
  * perf_sched_events : >0 events exist
  * perf_cgroup_events: >0 per-cpu cgroup events exist on this cpu
  */
-struct static_key_deferred perf_sched_events __read_mostly;
+
+static void perf_sched_delayed(struct work_struct *work);
+DEFINE_STATIC_KEY_FALSE(perf_sched_events);
+static DECLARE_DELAYED_WORK(perf_sched_work, perf_sched_delayed);
+static DEFINE_MUTEX(perf_sched_mutex);
+static atomic_t perf_sched_count;
+
 static DEFINE_PER_CPU(atomic_t, perf_cgroup_events);
 static DEFINE_PER_CPU(int, perf_sched_cb_usages);
 
@@ -1288,16 +1297,18 @@ static u64 perf_event_time(struct perf_event *event)
 
 /*
  * Update the total_time_enabled and total_time_running fields for a event.
- * The caller of this function needs to hold the ctx->lock.
  */
 static void update_event_times(struct perf_event *event)
 {
 	struct perf_event_context *ctx = event->ctx;
 	u64 run_end;
 
+	lockdep_assert_held(&ctx->lock);
+
 	if (event->state < PERF_EVENT_STATE_INACTIVE ||
 	    event->group_leader->state < PERF_EVENT_STATE_INACTIVE)
 		return;
+
 	/*
 	 * in cgroup mode, time_enabled represents
 	 * the time the event was enabled AND active
@@ -1645,7 +1656,7 @@ out:
 
 static bool is_orphaned_event(struct perf_event *event)
 {
-	return event->state == PERF_EVENT_STATE_EXIT;
+	return event->state == PERF_EVENT_STATE_DEAD;
 }
 
 static inline int pmu_filter_match(struct perf_event *event)
@@ -1690,14 +1701,14 @@ event_sched_out(struct perf_event *event,
 
 	perf_pmu_disable(event->pmu);
 
+	event->tstamp_stopped = tstamp;
+	event->pmu->del(event, 0);
+	event->oncpu = -1;
 	event->state = PERF_EVENT_STATE_INACTIVE;
 	if (event->pending_disable) {
 		event->pending_disable = 0;
 		event->state = PERF_EVENT_STATE_OFF;
 	}
-	event->tstamp_stopped = tstamp;
-	event->pmu->del(event, 0);
-	event->oncpu = -1;
 
 	if (!is_software_event(event))
 		cpuctx->active_oncpu--;
@@ -1732,7 +1743,6 @@ group_sched_out(struct perf_event *group_event,
 }
 
 #define DETACH_GROUP	0x01UL
-#define DETACH_STATE	0x02UL
 
 /*
  * Cross CPU call to remove a performance event
@@ -1752,8 +1762,6 @@ __perf_remove_from_context(struct perf_event *event,
 	if (flags & DETACH_GROUP)
 		perf_group_detach(event);
 	list_del_event(event, ctx);
-	if (flags & DETACH_STATE)
-		event->state = PERF_EVENT_STATE_EXIT;
 
 	if (!ctx->nr_events && ctx->is_active) {
 		ctx->is_active = 0;
@@ -2063,14 +2071,27 @@ static void add_event_to_ctx(struct perf_event *event,
 	event->tstamp_stopped = tstamp;
 }
 
-static void task_ctx_sched_out(struct perf_cpu_context *cpuctx,
-			       struct perf_event_context *ctx);
+static void ctx_sched_out(struct perf_event_context *ctx,
+			  struct perf_cpu_context *cpuctx,
+			  enum event_type_t event_type);
 static void
 ctx_sched_in(struct perf_event_context *ctx,
 	     struct perf_cpu_context *cpuctx,
 	     enum event_type_t event_type,
 	     struct task_struct *task);
 
+static void task_ctx_sched_out(struct perf_cpu_context *cpuctx,
+			       struct perf_event_context *ctx)
+{
+	if (!cpuctx->task_ctx)
+		return;
+
+	if (WARN_ON_ONCE(ctx != cpuctx->task_ctx))
+		return;
+
+	ctx_sched_out(ctx, cpuctx, EVENT_ALL);
+}
+
 static void perf_event_sched_in(struct perf_cpu_context *cpuctx,
 				struct perf_event_context *ctx,
 				struct task_struct *task)
@@ -2097,49 +2118,68 @@ static void ctx_resched(struct perf_cpu_context *cpuctx,
 /*
  * Cross CPU call to install and enable a performance event
  *
- * Must be called with ctx->mutex held
+ * Very similar to remote_function() + event_function() but cannot assume that
+ * things like ctx->is_active and cpuctx->task_ctx are set.
  */
 static int  __perf_install_in_context(void *info)
 {
-	struct perf_event_context *ctx = info;
+	struct perf_event *event = info;
+	struct perf_event_context *ctx = event->ctx;
 	struct perf_cpu_context *cpuctx = __get_cpu_context(ctx);
 	struct perf_event_context *task_ctx = cpuctx->task_ctx;
+	bool activate = true;
+	int ret = 0;
 
 	raw_spin_lock(&cpuctx->ctx.lock);
 	if (ctx->task) {
 		raw_spin_lock(&ctx->lock);
-		/*
-		 * If we hit the 'wrong' task, we've since scheduled and
-		 * everything should be sorted, nothing to do!
-		 */
 		task_ctx = ctx;
-		if (ctx->task != current)
+
+		/* If we're on the wrong CPU, try again */
+		if (task_cpu(ctx->task) != smp_processor_id()) {
+			ret = -ESRCH;
 			goto unlock;
+		}
 
 		/*
-		 * If task_ctx is set, it had better be to us.
+		 * If we're on the right CPU, see if the task we target is
+		 * current, if not we don't have to activate the ctx, a future
+		 * context switch will do that for us.
 		 */
-		WARN_ON_ONCE(cpuctx->task_ctx != ctx && cpuctx->task_ctx);
+		if (ctx->task != current)
+			activate = false;
+		else
+			WARN_ON_ONCE(cpuctx->task_ctx && cpuctx->task_ctx != ctx);
+
 	} else if (task_ctx) {
 		raw_spin_lock(&task_ctx->lock);
 	}
 
-	ctx_resched(cpuctx, task_ctx);
+	if (activate) {
+		ctx_sched_out(ctx, cpuctx, EVENT_TIME);
+		add_event_to_ctx(event, ctx);
+		ctx_resched(cpuctx, task_ctx);
+	} else {
+		add_event_to_ctx(event, ctx);
+	}
+
 unlock:
 	perf_ctx_unlock(cpuctx, task_ctx);
 
-	return 0;
+	return ret;
 }
 
 /*
- * Attach a performance event to a context
+ * Attach a performance event to a context.
+ *
+ * Very similar to event_function_call, see comment there.
  */
 static void
 perf_install_in_context(struct perf_event_context *ctx,
 			struct perf_event *event,
 			int cpu)
 {
-	struct task_struct *task = NULL;
+	struct task_struct *task = READ_ONCE(ctx->task);
 
 	lockdep_assert_held(&ctx->mutex);
 
@@ -2147,40 +2187,46 @@ perf_install_in_context(struct perf_event_context *ctx,
 	if (event->cpu != -1)
 		event->cpu = cpu;
 
+	if (!task) {
+		cpu_function_call(cpu, __perf_install_in_context, event);
+		return;
+	}
+
+	/*
+	 * Should not happen, we validate the ctx is still alive before calling.
+	 */
+	if (WARN_ON_ONCE(task == TASK_TOMBSTONE))
+		return;
+
 	/*
 	 * Installing events is tricky because we cannot rely on ctx->is_active
 	 * to be set in case this is the nr_events 0 -> 1 transition.
-	 *
-	 * So what we do is we add the event to the list here, which will allow
-	 * a future context switch to DTRT and then send a racy IPI. If the IPI
-	 * fails to hit the right task, this means a context switch must have
-	 * happened and that will have taken care of business.
 	 */
-	raw_spin_lock_irq(&ctx->lock);
-	task = ctx->task;
+again:
 	/*
-	 * Worse, we cannot even rely on the ctx actually existing anymore. If
-	 * between find_get_context() and perf_install_in_context() the task
-	 * went through perf_event_exit_task() its dead and we should not be
-	 * adding new events.
+	 * Cannot use task_function_call() because we need to run on the task's
+	 * CPU regardless of whether its current or not.
 	 */
-	if (task == TASK_TOMBSTONE) {
+	if (!cpu_function_call(task_cpu(task), __perf_install_in_context, event))
+		return;
+
+	raw_spin_lock_irq(&ctx->lock);
+	task = ctx->task;
+	if (WARN_ON_ONCE(task == TASK_TOMBSTONE)) {
+		/*
+		 * Cannot happen because we already checked above (which also
+		 * cannot happen), and we hold ctx->mutex, which serializes us
+		 * against perf_event_exit_task_context().
+		 */
 		raw_spin_unlock_irq(&ctx->lock);
 		return;
 	}
-	update_context_time(ctx);
+	raw_spin_unlock_irq(&ctx->lock);
 	/*
-	 * Update cgrp time only if current cgrp matches event->cgrp.
-	 * Must be done before calling add_event_to_ctx().
+	 * Since !ctx->is_active doesn't mean anything, we must IPI
+	 * unconditionally.
 	 */
-	update_cgrp_time_from_event(event);
-	add_event_to_ctx(event, ctx);
-	raw_spin_unlock_irq(&ctx->lock);
-
-	if (task)
-		task_function_call(task, __perf_install_in_context, ctx);
-	else
-		cpu_function_call(cpu, __perf_install_in_context, ctx);
+	goto again;
 }
 
 /*
@@ -2219,17 +2265,18 @@ static void __perf_event_enable(struct perf_event *event,
 	    event->state <= PERF_EVENT_STATE_ERROR)
 		return;
 
-	update_context_time(ctx);
+	if (ctx->is_active)
+		ctx_sched_out(ctx, cpuctx, EVENT_TIME);
+
 	__perf_event_mark_enabled(event);
 
 	if (!ctx->is_active)
 		return;
 
 	if (!event_filter_match(event)) {
-		if (is_cgroup_event(event)) {
-			perf_cgroup_set_timestamp(current, ctx); // XXX ?
+		if (is_cgroup_event(event))
 			perf_cgroup_defer_enabled(event);
-		}
+		ctx_sched_in(ctx, cpuctx, EVENT_TIME, current);
 		return;
 	}
 
@@ -2237,8 +2284,10 @@ static void __perf_event_enable(struct perf_event *event,
 	 * If the event is in a group and isn't the group leader,
 	 * then don't put it on unless the group is on.
 	 */
-	if (leader != event && leader->state != PERF_EVENT_STATE_ACTIVE)
+	if (leader != event && leader->state != PERF_EVENT_STATE_ACTIVE) {
+		ctx_sched_in(ctx, cpuctx, EVENT_TIME, current);
 		return;
+	}
 
 	task_ctx = cpuctx->task_ctx;
 	if (ctx->task)
@@ -2344,24 +2393,33 @@ static void ctx_sched_out(struct perf_event_context *ctx,
 	}
 
 	ctx->is_active &= ~event_type;
+	if (!(ctx->is_active & EVENT_ALL))
+		ctx->is_active = 0;
+
 	if (ctx->task) {
 		WARN_ON_ONCE(cpuctx->task_ctx != ctx);
 		if (!ctx->is_active)
 			cpuctx->task_ctx = NULL;
 	}
 
-	update_context_time(ctx);
-	update_cgrp_time_from_cpuctx(cpuctx);
-	if (!ctx->nr_active)
+	is_active ^= ctx->is_active; /* changed bits */
+
+	if (is_active & EVENT_TIME) {
+		/* update (and stop) ctx time */
+		update_context_time(ctx);
+		update_cgrp_time_from_cpuctx(cpuctx);
+	}
+
+	if (!ctx->nr_active || !(is_active & EVENT_ALL))
 		return;
 
 	perf_pmu_disable(ctx->pmu);
-	if ((is_active & EVENT_PINNED) && (event_type & EVENT_PINNED)) {
+	if (is_active & EVENT_PINNED) {
 		list_for_each_entry(event, &ctx->pinned_groups, group_entry)
 			group_sched_out(event, cpuctx, ctx);
 	}
 
-	if ((is_active & EVENT_FLEXIBLE) && (event_type & EVENT_FLEXIBLE)) {
+	if (is_active & EVENT_FLEXIBLE) {
 		list_for_each_entry(event, &ctx->flexible_groups, group_entry)
 			group_sched_out(event, cpuctx, ctx);
 	}
@@ -2641,18 +2699,6 @@ void __perf_event_task_sched_out(struct task_struct *task,
 		perf_cgroup_sched_out(task, next);
 }
 
-static void task_ctx_sched_out(struct perf_cpu_context *cpuctx,
-			       struct perf_event_context *ctx)
-{
-	if (!cpuctx->task_ctx)
-		return;
-
-	if (WARN_ON_ONCE(ctx != cpuctx->task_ctx))
-		return;
-
-	ctx_sched_out(ctx, cpuctx, EVENT_ALL);
-}
-
 /*
  * Called with IRQs disabled
  */
@@ -2735,7 +2781,7 @@ ctx_sched_in(struct perf_event_context *ctx,
 	if (likely(!ctx->nr_events))
 		return;
 
-	ctx->is_active |= event_type;
+	ctx->is_active |= (event_type | EVENT_TIME);
 	if (ctx->task) {
 		if (!is_active)
 			cpuctx->task_ctx = ctx;
@@ -2743,18 +2789,24 @@ ctx_sched_in(struct perf_event_context *ctx,
 			WARN_ON_ONCE(cpuctx->task_ctx != ctx);
 	}
 
-	now = perf_clock();
-	ctx->timestamp = now;
-	perf_cgroup_set_timestamp(task, ctx);
+	is_active ^= ctx->is_active; /* changed bits */
+
+	if (is_active & EVENT_TIME) {
+		/* start ctx time */
+		now = perf_clock();
+		ctx->timestamp = now;
+		perf_cgroup_set_timestamp(task, ctx);
+	}
+
 	/*
 	 * First go through the list and put on any pinned groups
 	 * in order to give them the best chance of going on.
 	 */
-	if (!(is_active & EVENT_PINNED) && (event_type & EVENT_PINNED))
+	if (is_active & EVENT_PINNED)
 		ctx_pinned_sched_in(ctx, cpuctx);
 
 	/* Then walk through the lower prio flexible groups */
-	if (!(is_active & EVENT_FLEXIBLE) && (event_type & EVENT_FLEXIBLE))
+	if (is_active & EVENT_FLEXIBLE)
 		ctx_flexible_sched_in(ctx, cpuctx);
 }
 
@@ -3120,6 +3172,7 @@ static void perf_event_enable_on_exec(int ctxn)
 
 	cpuctx = __get_cpu_context(ctx);
 	perf_ctx_lock(cpuctx, ctx);
+	ctx_sched_out(ctx, cpuctx, EVENT_TIME);
 	list_for_each_entry(event, &ctx->event_list, event_entry)
 		enabled |= event_enable_on_exec(event, ctx);
 
@@ -3537,12 +3590,22 @@ static void unaccount_event(struct perf_event *event)
 	if (has_branch_stack(event))
 		dec = true;
 
-	if (dec)
-		static_key_slow_dec_deferred(&perf_sched_events);
+	if (dec) {
+		if (!atomic_add_unless(&perf_sched_count, -1, 1))
+			schedule_delayed_work(&perf_sched_work, HZ);
+	}
 
 	unaccount_event_cpu(event, event->cpu);
 }
 
+static void perf_sched_delayed(struct work_struct *work)
+{
+	mutex_lock(&perf_sched_mutex);
+	if (atomic_dec_and_test(&perf_sched_count))
+		static_branch_disable(&perf_sched_events);
+	mutex_unlock(&perf_sched_mutex);
+}
+
 /*
  * The following implement mutual exclusion of events on "exclusive" pmus
  * (PERF_PMU_CAP_EXCLUSIVE). Such pmus can only have one event scheduled
@@ -3752,30 +3815,42 @@ static void put_event(struct perf_event *event)
  */
 int perf_event_release_kernel(struct perf_event *event)
 {
-	struct perf_event_context *ctx;
+	struct perf_event_context *ctx = event->ctx;
 	struct perf_event *child, *tmp;
 
+	/*
+	 * If we got here through err_file: fput(event_file); we will not have
+	 * attached to a context yet.
+	 */
+	if (!ctx) {
+		WARN_ON_ONCE(event->attach_state &
+				(PERF_ATTACH_CONTEXT|PERF_ATTACH_GROUP));
+		goto no_ctx;
+	}
+
 	if (!is_kernel_event(event))
 		perf_remove_from_owner(event);
 
 	ctx = perf_event_ctx_lock(event);
 	WARN_ON_ONCE(ctx->parent_ctx);
-	perf_remove_from_context(event, DETACH_GROUP | DETACH_STATE);
-	perf_event_ctx_unlock(event, ctx);
+	perf_remove_from_context(event, DETACH_GROUP);
 
+	raw_spin_lock_irq(&ctx->lock);
 	/*
-	 * At this point we must have event->state == PERF_EVENT_STATE_EXIT,
-	 * either from the above perf_remove_from_context() or through
-	 * perf_event_exit_event().
+	 * Mark this even as STATE_DEAD, there is no external reference to it
+	 * anymore.
 	 *
-	 * Therefore, anybody acquiring event->child_mutex after the below
-	 * loop _must_ also see this, most importantly inherit_event() which
-	 * will avoid placing more children on the list.
+	 * Anybody acquiring event->child_mutex after the below loop _must_
+	 * also see this, most importantly inherit_event() which will avoid
+	 * placing more children on the list.
 	 *
 	 * Thus this guarantees that we will in fact observe and kill _ALL_
 	 * child events.
 	 */
-	WARN_ON_ONCE(event->state != PERF_EVENT_STATE_EXIT);
+	event->state = PERF_EVENT_STATE_DEAD;
+	raw_spin_unlock_irq(&ctx->lock);
+
+	perf_event_ctx_unlock(event, ctx);
 
 again:
 	mutex_lock(&event->child_mutex);
@@ -3830,8 +3905,8 @@ again:
 	}
 	mutex_unlock(&event->child_mutex);
 
-	/* Must be the last reference */
-	put_event(event);
+no_ctx:
+	put_event(event); /* Must be the 'last' reference */
 	return 0;
 }
 EXPORT_SYMBOL_GPL(perf_event_release_kernel);
@@ -3988,7 +4063,7 @@ static bool is_event_hup(struct perf_event *event)
 {
 	bool no_children;
 
-	if (event->state != PERF_EVENT_STATE_EXIT)
+	if (event->state > PERF_EVENT_STATE_EXIT)
 		return false;
 
 	mutex_lock(&event->child_mutex);
@@ -7769,8 +7844,28 @@ static void account_event(struct perf_event *event)
 	if (is_cgroup_event(event))
 		inc = true;
 
-	if (inc)
-		static_key_slow_inc(&perf_sched_events.key);
+	if (inc) {
+		if (atomic_inc_not_zero(&perf_sched_count))
+			goto enabled;
+
+		mutex_lock(&perf_sched_mutex);
+		if (!atomic_read(&perf_sched_count)) {
+			static_branch_enable(&perf_sched_events);
+			/*
+			 * Guarantee that all CPUs observe they key change and
+			 * call the perf scheduling hooks before proceeding to
+			 * install events that need them.
+			 */
+			synchronize_sched();
+		}
+		/*
+		 * Now that we have waited for the sync_sched(), allow further
+		 * increments to by-pass the mutex.
+		 */
+		atomic_inc(&perf_sched_count);
+		mutex_unlock(&perf_sched_mutex);
+	}
+enabled:
 
 	account_event_cpu(event, event->cpu);
 }
@@ -8389,10 +8484,19 @@ SYSCALL_DEFINE5(perf_event_open,
 	if (move_group) {
 		gctx = group_leader->ctx;
 		mutex_lock_double(&gctx->mutex, &ctx->mutex);
+		if (gctx->task == TASK_TOMBSTONE) {
+			err = -ESRCH;
+			goto err_locked;
+		}
 	} else {
 		mutex_lock(&ctx->mutex);
 	}
 
+	if (ctx->task == TASK_TOMBSTONE) {
+		err = -ESRCH;
+		goto err_locked;
+	}
+
 	if (!perf_event_validate_size(event)) {
 		err = -E2BIG;
 		goto err_locked;
@@ -8509,7 +8613,12 @@ err_context:
 	perf_unpin_context(ctx);
 	put_ctx(ctx);
 err_alloc:
-	free_event(event);
+	/*
+	 * If event_file is set, the fput() above will have called ->release()
+	 * and that will take care of freeing the event.
+	 */
+	if (!event_file)
+		free_event(event);
 err_cpus:
 	put_online_cpus();
 err_task:
@@ -8563,12 +8672,14 @@ perf_event_create_kernel_counter(struct perf_event_attr *attr, int cpu,
 
 	WARN_ON_ONCE(ctx->parent_ctx);
 	mutex_lock(&ctx->mutex);
+	if (ctx->task == TASK_TOMBSTONE) {
+		err = -ESRCH;
+		goto err_unlock;
+	}
+
 	if (!exclusive_event_installable(event, ctx)) {
-		mutex_unlock(&ctx->mutex);
-		perf_unpin_context(ctx);
-		put_ctx(ctx);
 		err = -EBUSY;
-		goto err_free;
+		goto err_unlock;
 	}
 
 	perf_install_in_context(ctx, event, cpu);
@@ -8577,6 +8688,10 @@ perf_event_create_kernel_counter(struct perf_event_attr *attr, int cpu,
 
 	return event;
 
+err_unlock:
+	mutex_unlock(&ctx->mutex);
+	perf_unpin_context(ctx);
+	put_ctx(ctx);
 err_free:
 	free_event(event);
 err:
@@ -8695,7 +8810,7 @@ perf_event_exit_event(struct perf_event *child_event,
 	if (parent_event)
 		perf_group_detach(child_event);
 	list_del_event(child_event, child_ctx);
-	child_event->state = PERF_EVENT_STATE_EXIT; /* see perf_event_release_kernel() */
+	child_event->state = PERF_EVENT_STATE_EXIT; /* is_event_hup() */
 	raw_spin_unlock_irq(&child_ctx->lock);
 
 	/*
@@ -9206,7 +9321,7 @@ static void perf_event_init_cpu(int cpu)
 	struct swevent_htable *swhash = &per_cpu(swevent_htable, cpu);
 
 	mutex_lock(&swhash->hlist_mutex);
-	if (swhash->hlist_refcount > 0) {
+	if (swhash->hlist_refcount > 0 && !swevent_hlist_deref(swhash)) {
 		struct swevent_hlist *hlist;
 
 		hlist = kzalloc_node(sizeof(*hlist), GFP_KERNEL, cpu_to_node(cpu));
@@ -9282,11 +9397,9 @@ perf_cpu_notify(struct notifier_block *self, unsigned long action, void *hcpu)
 	switch (action & ~CPU_TASKS_FROZEN) {
 
 	case CPU_UP_PREPARE:
-	case CPU_DOWN_FAILED:
 		perf_event_init_cpu(cpu);
 		break;
 
-	case CPU_UP_CANCELED:
 	case CPU_DOWN_PREPARE:
 		perf_event_exit_cpu(cpu);
 		break;
@@ -9315,9 +9428,6 @@ void __init perf_event_init(void)
 	ret = init_hw_breakpoint();
 	WARN(ret, "hw_breakpoint initialization failed with: %d", ret);
 
-	/* do not patch jump label more than once per second */
-	jump_label_rate_limit(&perf_sched_events, HZ);
-
 	/*
 	 * Build time assertion that we keep the data_head at the intended
 	 * location.  IOW, validation we got the __reserved[] size right.
diff --git a/kernel/locking/lockdep.c b/kernel/locking/lockdep.c
index 60ace56618f6..716547fdb873 100644
--- a/kernel/locking/lockdep.c
+++ b/kernel/locking/lockdep.c
@@ -292,7 +292,7 @@ LIST_HEAD(all_lock_classes);
 #define __classhashfn(key)	hash_long((unsigned long)key, CLASSHASH_BITS)
 #define classhashentry(key)	(classhash_table + __classhashfn((key)))
 
-static struct list_head classhash_table[CLASSHASH_SIZE];
+static struct hlist_head classhash_table[CLASSHASH_SIZE];
 
 /*
  * We put the lock dependency chains into a hash-table as well, to cache
@@ -303,7 +303,7 @@ static struct list_head classhash_table[CLASSHASH_SIZE];
 #define __chainhashfn(chain)	hash_long(chain, CHAINHASH_BITS)
 #define chainhashentry(chain)	(chainhash_table + __chainhashfn((chain)))
 
-static struct list_head chainhash_table[CHAINHASH_SIZE];
+static struct hlist_head chainhash_table[CHAINHASH_SIZE];
 
 /*
  * The hash key of the lock dependency chains is a hash itself too:
@@ -666,7 +666,7 @@ static inline struct lock_class *
 look_up_lock_class(struct lockdep_map *lock, unsigned int subclass)
 {
 	struct lockdep_subclass_key *key;
-	struct list_head *hash_head;
+	struct hlist_head *hash_head;
 	struct lock_class *class;
 
 #ifdef CONFIG_DEBUG_LOCKDEP
@@ -719,7 +719,7 @@ look_up_lock_class(struct lockdep_map *lock, unsigned int subclass)
 	if (DEBUG_LOCKS_WARN_ON(!irqs_disabled()))
 		return NULL;
 
-	list_for_each_entry_rcu(class, hash_head, hash_entry) {
+	hlist_for_each_entry_rcu(class, hash_head, hash_entry) {
 		if (class->key == key) {
 			/*
 			 * Huh! same key, different name? Did someone trample
@@ -742,7 +742,7 @@ static inline struct lock_class *
 register_lock_class(struct lockdep_map *lock, unsigned int subclass, int force)
 {
 	struct lockdep_subclass_key *key;
-	struct list_head *hash_head;
+	struct hlist_head *hash_head;
 	struct lock_class *class;
 
 	DEBUG_LOCKS_WARN_ON(!irqs_disabled());
@@ -774,7 +774,7 @@ register_lock_class(struct lockdep_map *lock, unsigned int subclass, int force)
 	 * We have to do the hash-walk again, to avoid races
 	 * with another CPU:
 	 */
-	list_for_each_entry_rcu(class, hash_head, hash_entry) {
+	hlist_for_each_entry_rcu(class, hash_head, hash_entry) {
 		if (class->key == key)
 			goto out_unlock_set;
 	}
@@ -805,7 +805,7 @@ register_lock_class(struct lockdep_map *lock, unsigned int subclass, int force)
 	 * We use RCU's safe list-add method to make
 	 * parallel walking of the hash-list safe:
 	 */
-	list_add_tail_rcu(&class->hash_entry, hash_head);
+	hlist_add_head_rcu(&class->hash_entry, hash_head);
 	/*
 	 * Add it to the global list of classes:
 	 */
@@ -1822,7 +1822,7 @@ check_deadlock(struct task_struct *curr, struct held_lock *next,
  */
 static int
 check_prev_add(struct task_struct *curr, struct held_lock *prev,
-	       struct held_lock *next, int distance, int trylock_loop)
+	       struct held_lock *next, int distance, int *stack_saved)
 {
 	struct lock_list *entry;
 	int ret;
@@ -1883,8 +1883,11 @@ check_prev_add(struct task_struct *curr, struct held_lock *prev,
 		}
 	}
 
-	if (!trylock_loop && !save_trace(&trace))
-		return 0;
+	if (!*stack_saved) {
+		if (!save_trace(&trace))
+			return 0;
+		*stack_saved = 1;
+	}
 
 	/*
 	 * Ok, all validations passed, add the new lock
@@ -1907,6 +1910,8 @@ check_prev_add(struct task_struct *curr, struct held_lock *prev,
 	 * Debugging printouts:
 	 */
 	if (verbose(hlock_class(prev)) || verbose(hlock_class(next))) {
+		/* We drop graph lock, so another thread can overwrite trace. */
+		*stack_saved = 0;
 		graph_unlock();
 		printk("\n new dependency: ");
 		print_lock_name(hlock_class(prev));
@@ -1929,7 +1934,7 @@ static int
 check_prevs_add(struct task_struct *curr, struct held_lock *next)
 {
 	int depth = curr->lockdep_depth;
-	int trylock_loop = 0;
+	int stack_saved = 0;
 	struct held_lock *hlock;
 
 	/*
@@ -1956,7 +1961,7 @@ check_prevs_add(struct task_struct *curr, struct held_lock *next)
 		 */
 		if (hlock->read != 2 && hlock->check) {
 			if (!check_prev_add(curr, hlock, next,
-						distance, trylock_loop))
+						distance, &stack_saved))
 				return 0;
 			/*
 			 * Stop after the first non-trylock entry,
@@ -1979,7 +1984,6 @@ check_prevs_add(struct task_struct *curr, struct held_lock *next)
 		if (curr->held_locks[depth].irq_context !=
 				curr->held_locks[depth-1].irq_context)
 			break;
-		trylock_loop = 1;
 	}
 	return 1;
 out_bug:
@@ -2017,7 +2021,7 @@ static inline int lookup_chain_cache(struct task_struct *curr,
 				     u64 chain_key)
 {
 	struct lock_class *class = hlock_class(hlock);
-	struct list_head *hash_head = chainhashentry(chain_key);
+	struct hlist_head *hash_head = chainhashentry(chain_key);
 	struct lock_chain *chain;
 	struct held_lock *hlock_curr;
 	int i, j;
@@ -2033,7 +2037,7 @@ static inline int lookup_chain_cache(struct task_struct *curr,
 	 * We can walk it lock-free, because entries only get added
 	 * to the hash:
 	 */
-	list_for_each_entry_rcu(chain, hash_head, entry) {
+	hlist_for_each_entry_rcu(chain, hash_head, entry) {
 		if (chain->chain_key == chain_key) {
 cache_hit:
 			debug_atomic_inc(chain_lookup_hits);
@@ -2057,7 +2061,7 @@ cache_hit:
 	/*
 	 * We have to walk the chain again locked - to avoid duplicates:
 	 */
-	list_for_each_entry(chain, hash_head, entry) {
+	hlist_for_each_entry(chain, hash_head, entry) {
 		if (chain->chain_key == chain_key) {
 			graph_unlock();
 			goto cache_hit;
@@ -2091,7 +2095,7 @@ cache_hit:
 		}
 		chain_hlocks[chain->base + j] = class - lock_classes;
 	}
-	list_add_tail_rcu(&chain->entry, hash_head);
+	hlist_add_head_rcu(&chain->entry, hash_head);
 	debug_atomic_inc(chain_lookup_misses);
 	inc_chains();
 
@@ -3875,7 +3879,7 @@ void lockdep_reset(void)
 	nr_process_chains = 0;
 	debug_locks = 1;
 	for (i = 0; i < CHAINHASH_SIZE; i++)
-		INIT_LIST_HEAD(chainhash_table + i);
+		INIT_HLIST_HEAD(chainhash_table + i);
 	raw_local_irq_restore(flags);
 }
 
@@ -3894,7 +3898,7 @@ static void zap_class(struct lock_class *class)
 	/*
 	 * Unhash the class and remove it from the all_lock_classes list:
 	 */
-	list_del_rcu(&class->hash_entry);
+	hlist_del_rcu(&class->hash_entry);
 	list_del_rcu(&class->lock_entry);
 
 	RCU_INIT_POINTER(class->key, NULL);
@@ -3917,7 +3921,7 @@ static inline int within(const void *addr, void *start, unsigned long size)
 void lockdep_free_key_range(void *start, unsigned long size)
 {
 	struct lock_class *class;
-	struct list_head *head;
+	struct hlist_head *head;
 	unsigned long flags;
 	int i;
 	int locked;
@@ -3930,9 +3934,7 @@ void lockdep_free_key_range(void *start, unsigned long size)
 	 */
 	for (i = 0; i < CLASSHASH_SIZE; i++) {
 		head = classhash_table + i;
-		if (list_empty(head))
-			continue;
-		list_for_each_entry_rcu(class, head, hash_entry) {
+		hlist_for_each_entry_rcu(class, head, hash_entry) {
 			if (within(class->key, start, size))
 				zap_class(class);
 			else if (within(class->name, start, size))
@@ -3962,7 +3964,7 @@ void lockdep_free_key_range(void *start, unsigned long size)
 void lockdep_reset_lock(struct lockdep_map *lock)
 {
 	struct lock_class *class;
-	struct list_head *head;
+	struct hlist_head *head;
 	unsigned long flags;
 	int i, j;
 	int locked;
@@ -3987,9 +3989,7 @@ void lockdep_reset_lock(struct lockdep_map *lock)
 	locked = graph_lock();
 	for (i = 0; i < CLASSHASH_SIZE; i++) {
 		head = classhash_table + i;
-		if (list_empty(head))
-			continue;
-		list_for_each_entry_rcu(class, head, hash_entry) {
+		hlist_for_each_entry_rcu(class, head, hash_entry) {
 			int match = 0;
 
 			for (j = 0; j < NR_LOCKDEP_CACHING_CLASSES; j++)
@@ -4027,10 +4027,10 @@ void lockdep_init(void)
 		return;
 
 	for (i = 0; i < CLASSHASH_SIZE; i++)
-		INIT_LIST_HEAD(classhash_table + i);
+		INIT_HLIST_HEAD(classhash_table + i);
 
 	for (i = 0; i < CHAINHASH_SIZE; i++)
-		INIT_LIST_HEAD(chainhash_table + i);
+		INIT_HLIST_HEAD(chainhash_table + i);
 
 	lockdep_initialized = 1;
 }
diff --git a/kernel/memremap.c b/kernel/memremap.c
index 70ee3775de24..6cf54615a9c4 100644
--- a/kernel/memremap.c
+++ b/kernel/memremap.c
@@ -29,10 +29,10 @@ __weak void __iomem *ioremap_cache(resource_size_t offset, unsigned long size)
 
 static void *try_ram_remap(resource_size_t offset, size_t size)
 {
-	struct page *page = pfn_to_page(offset >> PAGE_SHIFT);
+	unsigned long pfn = PHYS_PFN(offset);
 
 	/* In the simple case just return the existing linear address */
-	if (!PageHighMem(page))
+	if (pfn_valid(pfn) && !PageHighMem(pfn_to_page(pfn)))
 		return __va(offset);
 	return NULL; /* fallback to ioremap_cache */
 }
@@ -114,7 +114,7 @@ EXPORT_SYMBOL(memunmap);
 
 static void devm_memremap_release(struct device *dev, void *res)
 {
-	memunmap(res);
+	memunmap(*(void **)res);
 }
 
 static int devm_memremap_match(struct device *dev, void *res, void *match_data)
@@ -136,8 +136,10 @@ void *devm_memremap(struct device *dev, resource_size_t offset,
 	if (addr) {
 		*ptr = addr;
 		devres_add(dev, ptr);
-	} else
+	} else {
 		devres_free(ptr);
+		return ERR_PTR(-ENXIO);
+	}
 
 	return addr;
 }
@@ -150,7 +152,7 @@ void devm_memunmap(struct device *dev, void *addr)
 }
 EXPORT_SYMBOL(devm_memunmap);
 
-pfn_t phys_to_pfn_t(phys_addr_t addr, unsigned long flags)
+pfn_t phys_to_pfn_t(phys_addr_t addr, u64 flags)
 {
 	return __pfn_to_pfn_t(addr >> PAGE_SHIFT, flags);
 }
@@ -268,13 +270,16 @@ struct dev_pagemap *find_dev_pagemap(resource_size_t phys)
 void *devm_memremap_pages(struct device *dev, struct resource *res,
 		struct percpu_ref *ref, struct vmem_altmap *altmap)
 {
-	int is_ram = region_intersects(res->start, resource_size(res),
-			"System RAM");
 	resource_size_t key, align_start, align_size, align_end;
 	struct dev_pagemap *pgmap;
 	struct page_map *page_map;
+	int error, nid, is_ram;
 	unsigned long pfn;
-	int error, nid;
+
+	align_start = res->start & ~(SECTION_SIZE - 1);
+	align_size = ALIGN(res->start + resource_size(res), SECTION_SIZE)
+		- align_start;
+	is_ram = region_intersects(align_start, align_size, "System RAM");
 
 	if (is_ram == REGION_MIXED) {
 		WARN_ONCE(1, "%s attempted on mixed region %pr\n",
@@ -312,8 +317,6 @@ void *devm_memremap_pages(struct device *dev, struct resource *res,
 
 	mutex_lock(&pgmap_lock);
 	error = 0;
-	align_start = res->start & ~(SECTION_SIZE - 1);
-	align_size = ALIGN(resource_size(res), SECTION_SIZE);
 	align_end = align_start + align_size - 1;
 	for (key = align_start; key <= align_end; key += SECTION_SIZE) {
 		struct dev_pagemap *dup;
@@ -349,8 +352,13 @@ void *devm_memremap_pages(struct device *dev, struct resource *res,
 	for_each_device_pfn(pfn, page_map) {
 		struct page *page = pfn_to_page(pfn);
 
-		/* ZONE_DEVICE pages must never appear on a slab lru */
-		list_force_poison(&page->lru);
+		/*
+		 * ZONE_DEVICE pages union ->lru with a ->pgmap back
+		 * pointer.  It is a bug if a ZONE_DEVICE page is ever
+		 * freed or placed on a driver-private list.  Seed the
+		 * storage with LIST_POISON* values.
+		 */
+		list_del(&page->lru);
 		page->pgmap = pgmap;
 	}
 	devres_add(dev, page_map);
diff --git a/kernel/module.c b/kernel/module.c
index 8358f4697c0c..794ebe8e878d 100644
--- a/kernel/module.c
+++ b/kernel/module.c
@@ -303,6 +303,9 @@ struct load_info {
 	struct _ddebug *debug;
 	unsigned int num_debug;
 	bool sig_ok;
+#ifdef CONFIG_KALLSYMS
+	unsigned long mod_kallsyms_init_off;
+#endif
 	struct {
 		unsigned int sym, str, mod, vers, info, pcpu;
 	} index;
@@ -981,6 +984,8 @@ SYSCALL_DEFINE2(delete_module, const char __user *, name_user,
 		mod->exit();
 	blocking_notifier_call_chain(&module_notify_list,
 				     MODULE_STATE_GOING, mod);
+	ftrace_release_mod(mod);
+
 	async_synchronize_full();
 
 	/* Store the name of the last unloaded module for diagnostic purposes */
@@ -2480,10 +2485,21 @@ static void layout_symtab(struct module *mod, struct load_info *info)
 	strsect->sh_flags |= SHF_ALLOC;
 	strsect->sh_entsize = get_offset(mod, &mod->init_layout.size, strsect,
 					 info->index.str) | INIT_OFFSET_MASK;
-	mod->init_layout.size = debug_align(mod->init_layout.size);
 	pr_debug("\t%s\n", info->secstrings + strsect->sh_name);
+
+	/* We'll tack temporary mod_kallsyms on the end. */
+	mod->init_layout.size = ALIGN(mod->init_layout.size,
+				      __alignof__(struct mod_kallsyms));
+	info->mod_kallsyms_init_off = mod->init_layout.size;
+	mod->init_layout.size += sizeof(struct mod_kallsyms);
+	mod->init_layout.size = debug_align(mod->init_layout.size);
 }
 
+/*
+ * We use the full symtab and strtab which layout_symtab arranged to
+ * be appended to the init section.  Later we switch to the cut-down
+ * core-only ones.
+ */
 static void add_kallsyms(struct module *mod, const struct load_info *info)
 {
 	unsigned int i, ndst;
@@ -2492,29 +2508,34 @@ static void add_kallsyms(struct module *mod, const struct load_info *info)
 	char *s;
 	Elf_Shdr *symsec = &info->sechdrs[info->index.sym];
 
-	mod->symtab = (void *)symsec->sh_addr;
-	mod->num_symtab = symsec->sh_size / sizeof(Elf_Sym);
+	/* Set up to point into init section. */
+	mod->kallsyms = mod->init_layout.base + info->mod_kallsyms_init_off;
+
+	mod->kallsyms->symtab = (void *)symsec->sh_addr;
+	mod->kallsyms->num_symtab = symsec->sh_size / sizeof(Elf_Sym);
 	/* Make sure we get permanent strtab: don't use info->strtab. */
-	mod->strtab = (void *)info->sechdrs[info->index.str].sh_addr;
+	mod->kallsyms->strtab = (void *)info->sechdrs[info->index.str].sh_addr;
 
 	/* Set types up while we still have access to sections. */
-	for (i = 0; i < mod->num_symtab; i++)
-		mod->symtab[i].st_info = elf_type(&mod->symtab[i], info);
-
-	mod->core_symtab = dst = mod->core_layout.base + info->symoffs;
-	mod->core_strtab = s = mod->core_layout.base + info->stroffs;
-	src = mod->symtab;
-	for (ndst = i = 0; i < mod->num_symtab; i++) {
+	for (i = 0; i < mod->kallsyms->num_symtab; i++)
+		mod->kallsyms->symtab[i].st_info
+			= elf_type(&mod->kallsyms->symtab[i], info);
+
+	/* Now populate the cut down core kallsyms for after init. */
+	mod->core_kallsyms.symtab = dst = mod->core_layout.base + info->symoffs;
+	mod->core_kallsyms.strtab = s = mod->core_layout.base + info->stroffs;
+	src = mod->kallsyms->symtab;
+	for (ndst = i = 0; i < mod->kallsyms->num_symtab; i++) {
 		if (i == 0 ||
 		    is_core_symbol(src+i, info->sechdrs, info->hdr->e_shnum,
 				   info->index.pcpu)) {
 			dst[ndst] = src[i];
-			dst[ndst++].st_name = s - mod->core_strtab;
-			s += strlcpy(s, &mod->strtab[src[i].st_name],
+			dst[ndst++].st_name = s - mod->core_kallsyms.strtab;
+			s += strlcpy(s, &mod->kallsyms->strtab[src[i].st_name],
 				     KSYM_NAME_LEN) + 1;
 		}
 	}
-	mod->core_num_syms = ndst;
+	mod->core_kallsyms.num_symtab = ndst;
 }
 #else
 static inline void layout_symtab(struct module *mod, struct load_info *info)
@@ -3263,9 +3284,8 @@ static noinline int do_init_module(struct module *mod)
 	module_put(mod);
 	trim_init_extable(mod);
 #ifdef CONFIG_KALLSYMS
-	mod->num_symtab = mod->core_num_syms;
-	mod->symtab = mod->core_symtab;
-	mod->strtab = mod->core_strtab;
+	/* Switch to core kallsyms now init is done: kallsyms may be walking! */
+	rcu_assign_pointer(mod->kallsyms, &mod->core_kallsyms);
 #endif
 	mod_tree_remove_init(mod);
 	disable_ro_nx(&mod->init_layout);
@@ -3295,6 +3315,7 @@ fail:
 	module_put(mod);
 	blocking_notifier_call_chain(&module_notify_list,
 				     MODULE_STATE_GOING, mod);
+	ftrace_release_mod(mod);
 	free_module(mod);
 	wake_up_all(&module_wq);
 	return ret;
@@ -3371,6 +3392,7 @@ static int complete_formation(struct module *mod, struct load_info *info)
 	mod->state = MODULE_STATE_COMING;
 	mutex_unlock(&module_mutex);
 
+	ftrace_module_enable(mod);
 	blocking_notifier_call_chain(&module_notify_list,
 				     MODULE_STATE_COMING, mod);
 	return 0;
@@ -3496,7 +3518,7 @@ static int load_module(struct load_info *info, const char __user *uargs,
 
 	/* Module is ready to execute: parsing args may do that. */
 	after_dashes = parse_args(mod->name, mod->args, mod->kp, mod->num_kp,
-				  -32768, 32767, NULL,
+				  -32768, 32767, mod,
 				  unknown_module_param_cb);
 	if (IS_ERR(after_dashes)) {
 		err = PTR_ERR(after_dashes);
@@ -3627,6 +3649,11 @@ static inline int is_arm_mapping_symbol(const char *str)
 	       && (str[2] == '\0' || str[2] == '.');
 }
 
+static const char *symname(struct mod_kallsyms *kallsyms, unsigned int symnum)
+{
+	return kallsyms->strtab + kallsyms->symtab[symnum].st_name;
+}
+
 static const char *get_ksymbol(struct module *mod,
 			       unsigned long addr,
 			       unsigned long *size,
@@ -3634,6 +3661,7 @@ static const char *get_ksymbol(struct module *mod,
 {
 	unsigned int i, best = 0;
 	unsigned long nextval;
+	struct mod_kallsyms *kallsyms = rcu_dereference_sched(mod->kallsyms);
 
 	/* At worse, next value is at end of module */
 	if (within_module_init(addr, mod))
@@ -3643,32 +3671,32 @@ static const char *get_ksymbol(struct module *mod,
 
 	/* Scan for closest preceding symbol, and next symbol. (ELF
 	   starts real symbols at 1). */
-	for (i = 1; i < mod->num_symtab; i++) {
-		if (mod->symtab[i].st_shndx == SHN_UNDEF)
+	for (i = 1; i < kallsyms->num_symtab; i++) {
+		if (kallsyms->symtab[i].st_shndx == SHN_UNDEF)
 			continue;
 
 		/* We ignore unnamed symbols: they're uninformative
 		 * and inserted at a whim. */
-		if (mod->symtab[i].st_value <= addr
-		    && mod->symtab[i].st_value > mod->symtab[best].st_value
-		    && *(mod->strtab + mod->symtab[i].st_name) != '\0'
-		    && !is_arm_mapping_symbol(mod->strtab + mod->symtab[i].st_name))
+		if (*symname(kallsyms, i) == '\0'
+		    || is_arm_mapping_symbol(symname(kallsyms, i)))
+			continue;
+
+		if (kallsyms->symtab[i].st_value <= addr
+		    && kallsyms->symtab[i].st_value > kallsyms->symtab[best].st_value)
 			best = i;
-		if (mod->symtab[i].st_value > addr
-		    && mod->symtab[i].st_value < nextval
-		    && *(mod->strtab + mod->symtab[i].st_name) != '\0'
-		    && !is_arm_mapping_symbol(mod->strtab + mod->symtab[i].st_name))
-			nextval = mod->symtab[i].st_value;
+		if (kallsyms->symtab[i].st_value > addr
+		    && kallsyms->symtab[i].st_value < nextval)
+			nextval = kallsyms->symtab[i].st_value;
 	}
 
 	if (!best)
 		return NULL;
 
 	if (size)
-		*size = nextval - mod->symtab[best].st_value;
+		*size = nextval - kallsyms->symtab[best].st_value;
 	if (offset)
-		*offset = addr - mod->symtab[best].st_value;
-	return mod->strtab + mod->symtab[best].st_name;
+		*offset = addr - kallsyms->symtab[best].st_value;
+	return symname(kallsyms, best);
 }
 
 /* For kallsyms to ask for address resolution.  NULL means not found.  Careful
@@ -3758,19 +3786,21 @@ int module_get_kallsym(unsigned int symnum, unsigned long *value, char *type,
 
 	preempt_disable();
 	list_for_each_entry_rcu(mod, &modules, list) {
+		struct mod_kallsyms *kallsyms;
+
 		if (mod->state == MODULE_STATE_UNFORMED)
 			continue;
-		if (symnum < mod->num_symtab) {
-			*value = mod->symtab[symnum].st_value;
-			*type = mod->symtab[symnum].st_info;
-			strlcpy(name, mod->strtab + mod->symtab[symnum].st_name,
-				KSYM_NAME_LEN);
+		kallsyms = rcu_dereference_sched(mod->kallsyms);
+		if (symnum < kallsyms->num_symtab) {
+			*value = kallsyms->symtab[symnum].st_value;
+			*type = kallsyms->symtab[symnum].st_info;
+			strlcpy(name, symname(kallsyms, symnum), KSYM_NAME_LEN);
 			strlcpy(module_name, mod->name, MODULE_NAME_LEN);
 			*exported = is_exported(name, *value, mod);
 			preempt_enable();
 			return 0;
 		}
-		symnum -= mod->num_symtab;
+		symnum -= kallsyms->num_symtab;
 	}
 	preempt_enable();
 	return -ERANGE;
@@ -3779,11 +3809,12 @@ int module_get_kallsym(unsigned int symnum, unsigned long *value, char *type,
 static unsigned long mod_find_symname(struct module *mod, const char *name)
 {
 	unsigned int i;
+	struct mod_kallsyms *kallsyms = rcu_dereference_sched(mod->kallsyms);
 
-	for (i = 0; i < mod->num_symtab; i++)
-		if (strcmp(name, mod->strtab+mod->symtab[i].st_name) == 0 &&
-		    mod->symtab[i].st_info != 'U')
-			return mod->symtab[i].st_value;
+	for (i = 0; i < kallsyms->num_symtab; i++)
+		if (strcmp(name, symname(kallsyms, i)) == 0 &&
+		    kallsyms->symtab[i].st_info != 'U')
+			return kallsyms->symtab[i].st_value;
 	return 0;
 }
 
@@ -3822,11 +3853,14 @@ int module_kallsyms_on_each_symbol(int (*fn)(void *, const char *,
 	module_assert_mutex();
 
 	list_for_each_entry(mod, &modules, list) {
+		/* We hold module_mutex: no need for rcu_dereference_sched */
+		struct mod_kallsyms *kallsyms = mod->kallsyms;
+
 		if (mod->state == MODULE_STATE_UNFORMED)
 			continue;
-		for (i = 0; i < mod->num_symtab; i++) {
-			ret = fn(data, mod->strtab + mod->symtab[i].st_name,
-				 mod, mod->symtab[i].st_value);
+		for (i = 0; i < kallsyms->num_symtab; i++) {
+			ret = fn(data, symname(kallsyms, i),
+				 mod, kallsyms->symtab[i].st_value);
 			if (ret != 0)
 				return ret;
 		}
diff --git a/kernel/resource.c b/kernel/resource.c
index 09c0597840b0..3669d1bfc425 100644
--- a/kernel/resource.c
+++ b/kernel/resource.c
@@ -1083,9 +1083,10 @@ struct resource * __request_region(struct resource *parent,
 		if (!conflict)
 			break;
 		if (conflict != parent) {
-			parent = conflict;
-			if (!(conflict->flags & IORESOURCE_BUSY))
+			if (!(conflict->flags & IORESOURCE_BUSY)) {
+				parent = conflict;
 				continue;
+			}
 		}
 		if (conflict->flags & flags & IORESOURCE_MUXED) {
 			add_wait_queue(&muxed_resource_wait, &wait);
diff --git a/kernel/sched/core.c b/kernel/sched/core.c
index 9503d590e5ef..41f6b2215aa8 100644
--- a/kernel/sched/core.c
+++ b/kernel/sched/core.c
@@ -26,6 +26,7 @@
  *              Thomas Gleixner, Mike Kravetz
  */
 
+#include <linux/kasan.h>
 #include <linux/mm.h>
 #include <linux/module.h>
 #include <linux/nmi.h>
@@ -5096,6 +5097,8 @@ void init_idle(struct task_struct *idle, int cpu)
 	idle->state = TASK_RUNNING;
 	idle->se.exec_start = sched_clock();
 
+	kasan_unpoison_task_stack(idle);
+
 #ifdef CONFIG_SMP
 	/*
 	 * Its possible that init_idle() gets called multiple times on a task,
diff --git a/kernel/sched/deadline.c b/kernel/sched/deadline.c
index cd64c979d0e1..57b939c81bce 100644
--- a/kernel/sched/deadline.c
+++ b/kernel/sched/deadline.c
@@ -420,7 +420,7 @@ static void replenish_dl_entity(struct sched_dl_entity *dl_se,
 	 * entity.
 	 */
 	if (dl_time_before(dl_se->deadline, rq_clock(rq))) {
-		printk_deferred_once("sched: DL replenish lagged to much\n");
+		printk_deferred_once("sched: DL replenish lagged too much\n");
 		dl_se->deadline = rq_clock(rq) + pi_se->dl_deadline;
 		dl_se->runtime = pi_se->dl_runtime;
 	}
diff --git a/kernel/trace/ftrace.c b/kernel/trace/ftrace.c
index eca592f977b2..57a6eea84694 100644
--- a/kernel/trace/ftrace.c
+++ b/kernel/trace/ftrace.c
@@ -4961,7 +4961,7 @@ void ftrace_release_mod(struct module *mod)
 	mutex_unlock(&ftrace_lock);
 }
 
-static void ftrace_module_enable(struct module *mod)
+void ftrace_module_enable(struct module *mod)
 {
 	struct dyn_ftrace *rec;
 	struct ftrace_page *pg;
@@ -5038,38 +5038,8 @@ void ftrace_module_init(struct module *mod)
 	ftrace_process_locs(mod, mod->ftrace_callsites,
 			    mod->ftrace_callsites + mod->num_ftrace_callsites);
 }
-
-static int ftrace_module_notify(struct notifier_block *self,
-				unsigned long val, void *data)
-{
-	struct module *mod = data;
-
-	switch (val) {
-	case MODULE_STATE_COMING:
-		ftrace_module_enable(mod);
-		break;
-	case MODULE_STATE_GOING:
-		ftrace_release_mod(mod);
-		break;
-	default:
-		break;
-	}
-
-	return 0;
-}
-#else
-static int ftrace_module_notify(struct notifier_block *self,
-				unsigned long val, void *data)
-{
-	return 0;
-}
 #endif /* CONFIG_MODULES */
 
-struct notifier_block ftrace_module_nb = {
-	.notifier_call = ftrace_module_notify,
-	.priority = INT_MIN,	/* Run after anything that can remove kprobes */
-};
-
 void __init ftrace_init(void)
 {
 	extern unsigned long __start_mcount_loc[];
@@ -5098,10 +5068,6 @@ void __init ftrace_init(void)
 				  __start_mcount_loc,
 				  __stop_mcount_loc);
 
-	ret = register_module_notifier(&ftrace_module_nb);
-	if (ret)
-		pr_warning("Failed to register trace ftrace module exit notifier\n");
-
 	set_ftrace_early_filters();
 
 	return;
diff --git a/kernel/trace/trace_events.c b/kernel/trace/trace_events.c
index f333e57c4614..05ddc0820771 100644
--- a/kernel/trace/trace_events.c
+++ b/kernel/trace/trace_events.c
@@ -97,16 +97,16 @@ trace_find_event_field(struct trace_event_call *call, char *name)
 	struct ftrace_event_field *field;
 	struct list_head *head;
 
-	field = __find_event_field(&ftrace_generic_fields, name);
+	head = trace_get_fields(call);
+	field = __find_event_field(head, name);
 	if (field)
 		return field;
 
-	field = __find_event_field(&ftrace_common_fields, name);
+	field = __find_event_field(&ftrace_generic_fields, name);
 	if (field)
 		return field;
 
-	head = trace_get_fields(call);
-	return __find_event_field(head, name);
+	return __find_event_field(&ftrace_common_fields, name);
 }
 
 static int __trace_define_field(struct list_head *head, const char *type,
@@ -171,8 +171,10 @@ static int trace_define_generic_fields(void)
 {
 	int ret;
 
-	__generic_field(int, cpu, FILTER_OTHER);
-	__generic_field(char *, comm, FILTER_PTR_STRING);
+	__generic_field(int, CPU, FILTER_CPU);
+	__generic_field(int, cpu, FILTER_CPU);
+	__generic_field(char *, COMM, FILTER_COMM);
+	__generic_field(char *, comm, FILTER_COMM);
 
 	return ret;
 }
@@ -869,7 +871,8 @@ t_next(struct seq_file *m, void *v, loff_t *pos)
 		 * The ftrace subsystem is for showing formats only.
 		 * They can not be enabled or disabled via the event files.
 		 */
-		if (call->class && call->class->reg)
+		if (call->class && call->class->reg &&
+		    !(call->flags & TRACE_EVENT_FL_IGNORE_ENABLE))
 			return file;
 	}
 
diff --git a/kernel/trace/trace_events_filter.c b/kernel/trace/trace_events_filter.c
index f93a219b18da..6816302542b2 100644
--- a/kernel/trace/trace_events_filter.c
+++ b/kernel/trace/trace_events_filter.c
@@ -1043,13 +1043,14 @@ static int init_pred(struct filter_parse_state *ps,
 		return -EINVAL;
 	}
 
-	if (is_string_field(field)) {
+	if (field->filter_type == FILTER_COMM) {
+		filter_build_regex(pred);
+		fn = filter_pred_comm;
+		pred->regex.field_len = TASK_COMM_LEN;
+	} else if (is_string_field(field)) {
 		filter_build_regex(pred);
 
-		if (!strcmp(field->name, "comm")) {
-			fn = filter_pred_comm;
-			pred->regex.field_len = TASK_COMM_LEN;
-		} else if (field->filter_type == FILTER_STATIC_STRING) {
+		if (field->filter_type == FILTER_STATIC_STRING) {
 			fn = filter_pred_string;
 			pred->regex.field_len = field->size;
 		} else if (field->filter_type == FILTER_DYN_STRING)
@@ -1072,7 +1073,7 @@ static int init_pred(struct filter_parse_state *ps,
 		}
 		pred->val = val;
 
-		if (!strcmp(field->name, "cpu"))
+		if (field->filter_type == FILTER_CPU)
 			fn = filter_pred_cpu;
 		else
 			fn = select_comparison_fn(pred->op, field->size,
diff --git a/kernel/trace/trace_stack.c b/kernel/trace/trace_stack.c
index 202df6cffcca..2a1abbaca10e 100644
--- a/kernel/trace/trace_stack.c
+++ b/kernel/trace/trace_stack.c
@@ -156,7 +156,11 @@ check_stack(unsigned long ip, unsigned long *stack)
 		for (; p < top && i < stack_trace_max.nr_entries; p++) {
 			if (stack_dump_trace[i] == ULONG_MAX)
 				break;
-			if (*p == stack_dump_trace[i]) {
+			/*
+			 * The READ_ONCE_NOCHECK is used to let KASAN know that
+			 * this is not a stack-out-of-bounds error.
+			 */
+			if ((READ_ONCE_NOCHECK(*p)) == stack_dump_trace[i]) {
 				stack_dump_trace[x] = stack_dump_trace[i++];
 				this_size = stack_trace_index[x++] =
 					(top - p) * sizeof(unsigned long);
diff --git a/kernel/workqueue.c b/kernel/workqueue.c
index 61a0264e28f9..7ff5dc7d2ac5 100644
--- a/kernel/workqueue.c
+++ b/kernel/workqueue.c
@@ -301,7 +301,23 @@ static DEFINE_SPINLOCK(wq_mayday_lock);	/* protects wq->maydays list */
 static LIST_HEAD(workqueues);		/* PR: list of all workqueues */
 static bool workqueue_freezing;		/* PL: have wqs started freezing? */
 
-static cpumask_var_t wq_unbound_cpumask; /* PL: low level cpumask for all unbound wqs */
+/* PL: allowable cpus for unbound wqs and work items */
+static cpumask_var_t wq_unbound_cpumask;
+
+/* CPU where unbound work was last round robin scheduled from this CPU */
+static DEFINE_PER_CPU(int, wq_rr_cpu_last);
+
+/*
+ * Local execution of unbound work items is no longer guaranteed.  The
+ * following always forces round-robin CPU selection on unbound work items
+ * to uncover usages which depend on it.
+ */
+#ifdef CONFIG_DEBUG_WQ_FORCE_RR_CPU
+static bool wq_debug_force_rr_cpu = true;
+#else
+static bool wq_debug_force_rr_cpu = false;
+#endif
+module_param_named(debug_force_rr_cpu, wq_debug_force_rr_cpu, bool, 0644);
 
 /* the per-cpu worker pools */
 static DEFINE_PER_CPU_SHARED_ALIGNED(struct worker_pool [NR_STD_WORKER_POOLS],
@@ -570,6 +586,16 @@ static struct pool_workqueue *unbound_pwq_by_node(struct workqueue_struct *wq,
 						  int node)
 {
 	assert_rcu_or_wq_mutex_or_pool_mutex(wq);
+
+	/*
+	 * XXX: @node can be NUMA_NO_NODE if CPU goes offline while a
+	 * delayed item is pending.  The plan is to keep CPU -> NODE
+	 * mapping valid and stable across CPU on/offlines.  Once that
+	 * happens, this workaround can be removed.
+	 */
+	if (unlikely(node == NUMA_NO_NODE))
+		return wq->dfl_pwq;
+
 	return rcu_dereference_raw(wq->numa_pwq_tbl[node]);
 }
 
@@ -1298,6 +1324,39 @@ static bool is_chained_work(struct workqueue_struct *wq)
 	return worker && worker->current_pwq->wq == wq;
 }
 
+/*
+ * When queueing an unbound work item to a wq, prefer local CPU if allowed
+ * by wq_unbound_cpumask.  Otherwise, round robin among the allowed ones to
+ * avoid perturbing sensitive tasks.
+ */
+static int wq_select_unbound_cpu(int cpu)
+{
+	static bool printed_dbg_warning;
+	int new_cpu;
+
+	if (likely(!wq_debug_force_rr_cpu)) {
+		if (cpumask_test_cpu(cpu, wq_unbound_cpumask))
+			return cpu;
+	} else if (!printed_dbg_warning) {
+		pr_warn("workqueue: round-robin CPU selection forced, expect performance impact\n");
+		printed_dbg_warning = true;
+	}
+
+	if (cpumask_empty(wq_unbound_cpumask))
+		return cpu;
+
+	new_cpu = __this_cpu_read(wq_rr_cpu_last);
+	new_cpu = cpumask_next_and(new_cpu, wq_unbound_cpumask, cpu_online_mask);
+	if (unlikely(new_cpu >= nr_cpu_ids)) {
+		new_cpu = cpumask_first_and(wq_unbound_cpumask, cpu_online_mask);
+		if (unlikely(new_cpu >= nr_cpu_ids))
+			return cpu;
+	}
+	__this_cpu_write(wq_rr_cpu_last, new_cpu);
+
+	return new_cpu;
+}
+
 static void __queue_work(int cpu, struct workqueue_struct *wq,
 			 struct work_struct *work)
 {
@@ -1323,7 +1382,7 @@ static void __queue_work(int cpu, struct workqueue_struct *wq,
 		return;
 retry:
 	if (req_cpu == WORK_CPU_UNBOUND)
-		cpu = raw_smp_processor_id();
+		cpu = wq_select_unbound_cpu(raw_smp_processor_id());
 
 	/* pwq which will be used unless @work is executing elsewhere */
 	if (!(wq->flags & WQ_UNBOUND))
@@ -1464,13 +1523,13 @@ static void __queue_delayed_work(int cpu, struct workqueue_struct *wq,
 	timer_stats_timer_set_start_info(&dwork->timer);
 
 	dwork->wq = wq;
-	/* timer isn't guaranteed to run in this cpu, record earlier */
-	if (cpu == WORK_CPU_UNBOUND)
-		cpu = raw_smp_processor_id();
 	dwork->cpu = cpu;
 	timer->expires = jiffies + delay;
 
-	add_timer_on(timer, cpu);
+	if (unlikely(cpu != WORK_CPU_UNBOUND))
+		add_timer_on(timer, cpu);
+	else
+		add_timer(timer);
 }
 
 /**
@@ -2355,7 +2414,8 @@ static void check_flush_dependency(struct workqueue_struct *target_wq,
 	WARN_ONCE(current->flags & PF_MEMALLOC,
 		  "workqueue: PF_MEMALLOC task %d(%s) is flushing !WQ_MEM_RECLAIM %s:%pf",
 		  current->pid, current->comm, target_wq->name, target_func);
-	WARN_ONCE(worker && (worker->current_pwq->wq->flags & WQ_MEM_RECLAIM),
+	WARN_ONCE(worker && ((worker->current_pwq->wq->flags &
+			      (WQ_MEM_RECLAIM | __WQ_LEGACY)) == WQ_MEM_RECLAIM),
 		  "workqueue: WQ_MEM_RECLAIM %s:%pf is flushing !WQ_MEM_RECLAIM %s:%pf",
 		  worker->current_pwq->wq->name, worker->current_func,
 		  target_wq->name, target_func);
diff --git a/lib/Kconfig.debug b/lib/Kconfig.debug
index ecb9e75614bf..8bfd1aca7a3d 100644
--- a/lib/Kconfig.debug
+++ b/lib/Kconfig.debug
@@ -1400,6 +1400,21 @@ config RCU_EQS_DEBUG
 
 endmenu # "RCU Debugging"
 
+config DEBUG_WQ_FORCE_RR_CPU
+	bool "Force round-robin CPU selection for unbound work items"
+	depends on DEBUG_KERNEL
+	default n
+	help
+	  Workqueue used to implicitly guarantee that work items queued
+	  without explicit CPU specified are put on the local CPU.  This
+	  guarantee is no longer true and while local CPU is still
+	  preferred work items may be put on foreign CPUs.  Kernel
+	  parameter "workqueue.debug_force_rr_cpu" is added to force
+	  round-robin CPU selection to flush out usages which depend on the
+	  now broken guarantee.  This config option enables the debug
+	  feature by default.  When enabled, memory and cache locality will
+	  be impacted.
+
 config DEBUG_BLOCK_EXT_DEVT
         bool "Force extended block device numbers and spread them"
 	depends on DEBUG_KERNEL
diff --git a/lib/Kconfig.ubsan b/lib/Kconfig.ubsan
index 49518fb48cab..e07c1ba9ba13 100644
--- a/lib/Kconfig.ubsan
+++ b/lib/Kconfig.ubsan
@@ -18,6 +18,8 @@ config UBSAN_SANITIZE_ALL
 	  This option activates instrumentation for the entire kernel.
 	  If you don't enable this option, you have to explicitly specify
 	  UBSAN_SANITIZE := y for the files/directories you want to check for UB.
+	  Enabling this option will get kernel image size increased
+	  significantly.
 
 config UBSAN_ALIGNMENT
 	bool "Enable checking of pointers alignment"
@@ -25,5 +27,5 @@ config UBSAN_ALIGNMENT
 	default y if !HAVE_EFFICIENT_UNALIGNED_ACCESS
 	help
 	  This option enables detection of unaligned memory accesses.
-	  Enabling this option on architectures that support unalligned
+	  Enabling this option on architectures that support unaligned
 	  accesses may produce a lot of false positives.
diff --git a/lib/klist.c b/lib/klist.c
index d74cf7a29afd..0507fa5d84c5 100644
--- a/lib/klist.c
+++ b/lib/klist.c
@@ -282,9 +282,9 @@ void klist_iter_init_node(struct klist *k, struct klist_iter *i,
 			  struct klist_node *n)
 {
 	i->i_klist = k;
-	i->i_cur = n;
-	if (n)
-		kref_get(&n->n_ref);
+	i->i_cur = NULL;
+	if (n && kref_get_unless_zero(&n->n_ref))
+		i->i_cur = n;
 }
 EXPORT_SYMBOL_GPL(klist_iter_init_node);
 
diff --git a/lib/list_debug.c b/lib/list_debug.c
index 3345a089ef7b..3859bf63561c 100644
--- a/lib/list_debug.c
+++ b/lib/list_debug.c
@@ -12,13 +12,6 @@
 #include <linux/kernel.h>
 #include <linux/rculist.h>
 
-static struct list_head force_poison;
-void list_force_poison(struct list_head *entry)
-{
-	entry->next = &force_poison;
-	entry->prev = &force_poison;
-}
-
 /*
  * Insert a new entry between two known consecutive entries.
  *
@@ -30,8 +23,6 @@ void __list_add(struct list_head *new,
 			      struct list_head *prev,
 			      struct list_head *next)
 {
-	WARN(new->next == &force_poison || new->prev == &force_poison,
-		"list_add attempted on force-poisoned entry\n");
 	WARN(next->prev != prev,
 		"list_add corruption. next->prev should be "
 		"prev (%p), but was %p. (next=%p).\n",
diff --git a/lib/scatterlist.c b/lib/scatterlist.c
index bafa9933fa76..004fc70fc56a 100644
--- a/lib/scatterlist.c
+++ b/lib/scatterlist.c
@@ -598,9 +598,9 @@ EXPORT_SYMBOL(sg_miter_next);
  *
  * Description:
  *   Stops mapping iterator @miter.  @miter should have been started
- *   started using sg_miter_start().  A stopped iteration can be
- *   resumed by calling sg_miter_next() on it.  This is useful when
- *   resources (kmap) need to be released during iteration.
+ *   using sg_miter_start().  A stopped iteration can be resumed by
+ *   calling sg_miter_next() on it.  This is useful when resources (kmap)
+ *   need to be released during iteration.
  *
  * Context:
  *   Preemption disabled if the SG_MITER_ATOMIC is set.  Don't care
diff --git a/lib/ucs2_string.c b/lib/ucs2_string.c
index 6f500ef2301d..f0b323abb4c6 100644
--- a/lib/ucs2_string.c
+++ b/lib/ucs2_string.c
@@ -49,3 +49,65 @@ ucs2_strncmp(const ucs2_char_t *a, const ucs2_char_t *b, size_t len)
         }
 }
 EXPORT_SYMBOL(ucs2_strncmp);
+
+unsigned long
+ucs2_utf8size(const ucs2_char_t *src)
+{
+	unsigned long i;
+	unsigned long j = 0;
+
+	for (i = 0; i < ucs2_strlen(src); i++) {
+		u16 c = src[i];
+
+		if (c >= 0x800)
+			j += 3;
+		else if (c >= 0x80)
+			j += 2;
+		else
+			j += 1;
+	}
+
+	return j;
+}
+EXPORT_SYMBOL(ucs2_utf8size);
+
+/*
+ * copy at most maxlength bytes of whole utf8 characters to dest from the
+ * ucs2 string src.
+ *
+ * The return value is the number of characters copied, not including the
+ * final NUL character.
+ */
+unsigned long
+ucs2_as_utf8(u8 *dest, const ucs2_char_t *src, unsigned long maxlength)
+{
+	unsigned int i;
+	unsigned long j = 0;
+	unsigned long limit = ucs2_strnlen(src, maxlength);
+
+	for (i = 0; maxlength && i < limit; i++) {
+		u16 c = src[i];
+
+		if (c >= 0x800) {
+			if (maxlength < 3)
+				break;
+			maxlength -= 3;
+			dest[j++] = 0xe0 | (c & 0xf000) >> 12;
+			dest[j++] = 0x80 | (c & 0x0fc0) >> 6;
+			dest[j++] = 0x80 | (c & 0x003f);
+		} else if (c >= 0x80) {
+			if (maxlength < 2)
+				break;
+			maxlength -= 2;
+			dest[j++] = 0xc0 | (c & 0x7c0) >> 6;
+			dest[j++] = 0x80 | (c & 0x03f);
+		} else {
+			maxlength -= 1;
+			dest[j++] = c & 0x7f;
+		}
+	}
+	if (maxlength)
+		dest[j] = '\0';
+	return j;
+}
+EXPORT_SYMBOL(ucs2_as_utf8);
diff --git a/lib/vsprintf.c b/lib/vsprintf.c
index 48ff9c36644d..f44e178e6ede 100644
--- a/lib/vsprintf.c
+++ b/lib/vsprintf.c
@@ -1590,22 +1590,23 @@ char *pointer(const char *fmt, char *buf, char *end, void *ptr,
 			return buf;
 		}
 	case 'K':
-		/*
-		 * %pK cannot be used in IRQ context because its test
-		 * for CAP_SYSLOG would be meaningless.
-		 */
-		if (kptr_restrict && (in_irq() || in_serving_softirq() ||
-				      in_nmi())) {
-			if (spec.field_width == -1)
-				spec.field_width = default_width;
-			return string(buf, end, "pK-error", spec);
-		}
-
 		switch (kptr_restrict) {
 		case 0:
 			/* Always print %pK values */
 			break;
 		case 1: {
+			const struct cred *cred;
+
+			/*
+			 * kptr_restrict==1 cannot be used in IRQ context
+			 * because its test for CAP_SYSLOG would be meaningless.
+			 */
+			if (in_irq() || in_serving_softirq() || in_nmi()) {
+				if (spec.field_width == -1)
+					spec.field_width = default_width;
+				return string(buf, end, "pK-error", spec);
+			}
+
 			/*
 			 * Only print the real pointer value if the current
 			 * process has CAP_SYSLOG and is running with the
@@ -1615,8 +1616,7 @@ char *pointer(const char *fmt, char *buf, char *end, void *ptr,
 			 * leak pointer values if a binary opens a file using
 			 * %pK and then elevates privileges before reading it.
 			 */
-			const struct cred *cred = current_cred();
-
+			cred = current_cred();
 			if (!has_capability_noaudit(current, CAP_SYSLOG) ||
 			    !uid_eq(cred->euid, cred->uid) ||
 			    !gid_eq(cred->egid, cred->gid))
diff --git a/mm/backing-dev.c b/mm/backing-dev.c
index 926c76d56388..c554d173a65f 100644
--- a/mm/backing-dev.c
+++ b/mm/backing-dev.c
@@ -328,7 +328,7 @@ static int wb_init(struct bdi_writeback *wb, struct backing_dev_info *bdi,
 	return 0;
 
 out_destroy_stat:
-	while (--i)
+	while (i--)
 		percpu_counter_destroy(&wb->stat[i]);
 	fprop_local_destroy_percpu(&wb->completions);
 out_put_cong:
diff --git a/mm/filemap.c b/mm/filemap.c
index bc943867d68c..da7a35d83de7 100644
--- a/mm/filemap.c
+++ b/mm/filemap.c
@@ -195,6 +195,30 @@ void __delete_from_page_cache(struct page *page, void *shadow,
 	else
 		cleancache_invalidate_page(mapping, page);
 
+	VM_BUG_ON_PAGE(page_mapped(page), page);
+	if (!IS_ENABLED(CONFIG_DEBUG_VM) && unlikely(page_mapped(page))) {
+		int mapcount;
+
+		pr_alert("BUG: Bad page cache in process %s  pfn:%05lx\n",
+			 current->comm, page_to_pfn(page));
+		dump_page(page, "still mapped when deleted");
+		dump_stack();
+		add_taint(TAINT_BAD_PAGE, LOCKDEP_NOW_UNRELIABLE);
+
+		mapcount = page_mapcount(page);
+		if (mapping_exiting(mapping) &&
+		    page_count(page) >= mapcount + 2) {
+			/*
+			 * All vmas have already been torn down, so it's
+			 * a good bet that actually the page is unmapped,
+			 * and we'd prefer not to leak it: if we're wrong,
+			 * some other bad page check should catch it later.
+			 */
+			page_mapcount_reset(page);
+			atomic_sub(mapcount, &page->_count);
+		}
+	}
+
 	page_cache_tree_delete(mapping, page, shadow);
 
 	page->mapping = NULL;
@@ -205,7 +229,6 @@ void __delete_from_page_cache(struct page *page, void *shadow,
 		__dec_zone_page_state(page, NR_FILE_PAGES);
 	if (PageSwapBacked(page))
 		__dec_zone_page_state(page, NR_SHMEM);
-	VM_BUG_ON_PAGE(page_mapped(page), page);
 
 	/*
 	 * At this point page must be either written or cleaned by truncate.
@@ -446,7 +469,8 @@ int filemap_write_and_wait(struct address_space *mapping)
 {
 	int err = 0;
 
-	if (mapping->nrpages) {
+	if ((!dax_mapping(mapping) && mapping->nrpages) ||
+	    (dax_mapping(mapping) && mapping->nrexceptional)) {
 		err = filemap_fdatawrite(mapping);
 		/*
 		 * Even if the above returned error, the pages may be
@@ -482,13 +506,8 @@ int filemap_write_and_wait_range(struct address_space *mapping,
 {
 	int err = 0;
 
-	if (dax_mapping(mapping) && mapping->nrexceptional) {
-		err = dax_writeback_mapping_range(mapping, lstart, lend);
-		if (err)
-			return err;
-	}
-
-	if (mapping->nrpages) {
+	if ((!dax_mapping(mapping) && mapping->nrpages) ||
+	    (dax_mapping(mapping) && mapping->nrexceptional)) {
 		err = __filemap_fdatawrite_range(mapping, lstart, lend,
 						 WB_SYNC_ALL);
 		/* See comment of filemap_write_and_wait() */
@@ -1890,6 +1909,7 @@ EXPORT_SYMBOL(generic_file_read_iter);
  * page_cache_read - adds requested page to the page cache if not already there
  * @file:	file to read
  * @offset:	page index
+ * @gfp_mask:	memory allocation flags
  *
  * This adds the requested page to the page cache if it isn't already there,
  * and schedules an I/O to read in its contents from disk.
diff --git a/mm/huge_memory.c b/mm/huge_memory.c
index 08fc0ba2207e..e10a4fee88d2 100644
--- a/mm/huge_memory.c
+++ b/mm/huge_memory.c
@@ -1700,7 +1700,8 @@ bool move_huge_pmd(struct vm_area_struct *vma, struct vm_area_struct *new_vma,
 		pmd = pmdp_huge_get_and_clear(mm, old_addr, old_pmd);
 		VM_BUG_ON(!pmd_none(*new_pmd));
 
-		if (pmd_move_must_withdraw(new_ptl, old_ptl)) {
+		if (pmd_move_must_withdraw(new_ptl, old_ptl) &&
+				vma_is_anonymous(vma)) {
 			pgtable_t pgtable;
 			pgtable = pgtable_trans_huge_withdraw(mm, old_pmd);
 			pgtable_trans_huge_deposit(mm, new_pmd, pgtable);
@@ -2835,6 +2836,7 @@ static void __split_huge_pmd_locked(struct vm_area_struct *vma, pmd_t *pmd,
 	pgtable_t pgtable;
 	pmd_t _pmd;
 	bool young, write, dirty;
+	unsigned long addr;
 	int i;
 
 	VM_BUG_ON(haddr & ~HPAGE_PMD_MASK);
@@ -2860,10 +2862,11 @@ static void __split_huge_pmd_locked(struct vm_area_struct *vma, pmd_t *pmd,
 	young = pmd_young(*pmd);
 	dirty = pmd_dirty(*pmd);
 
+	pmdp_huge_split_prepare(vma, haddr, pmd);
 	pgtable = pgtable_trans_huge_withdraw(mm, pmd);
 	pmd_populate(mm, &_pmd, pgtable);
 
-	for (i = 0; i < HPAGE_PMD_NR; i++, haddr += PAGE_SIZE) {
+	for (i = 0, addr = haddr; i < HPAGE_PMD_NR; i++, addr += PAGE_SIZE) {
 		pte_t entry, *pte;
 		/*
 		 * Note that NUMA hinting access restrictions are not
@@ -2884,9 +2887,9 @@ static void __split_huge_pmd_locked(struct vm_area_struct *vma, pmd_t *pmd,
 		}
 		if (dirty)
 			SetPageDirty(page + i);
-		pte = pte_offset_map(&_pmd, haddr);
+		pte = pte_offset_map(&_pmd, addr);
 		BUG_ON(!pte_none(*pte));
-		set_pte_at(mm, haddr, pte, entry);
+		set_pte_at(mm, addr, pte, entry);
 		atomic_inc(&page[i]._mapcount);
 		pte_unmap(pte);
 	}
@@ -2936,7 +2939,7 @@ static void __split_huge_pmd_locked(struct vm_area_struct *vma, pmd_t *pmd,
 	pmd_populate(mm, pmd, pgtable);
 
 	if (freeze) {
-		for (i = 0; i < HPAGE_PMD_NR; i++, haddr += PAGE_SIZE) {
+		for (i = 0; i < HPAGE_PMD_NR; i++) {
 			page_remove_rmap(page + i, false);
 			put_page(page + i);
 		}
diff --git a/mm/hugetlb.c b/mm/hugetlb.c
index 06ae13e869d0..aefba5a9cc47 100644
--- a/mm/hugetlb.c
+++ b/mm/hugetlb.c
@@ -2630,8 +2630,10 @@ static int __init hugetlb_init(void)
 			hugetlb_add_hstate(HUGETLB_PAGE_ORDER);
 	}
 	default_hstate_idx = hstate_index(size_to_hstate(default_hstate_size));
-	if (default_hstate_max_huge_pages)
-		default_hstate.max_huge_pages = default_hstate_max_huge_pages;
+	if (default_hstate_max_huge_pages) {
+		if (!default_hstate.max_huge_pages)
+			default_hstate.max_huge_pages = default_hstate_max_huge_pages;
+	}
 
 	hugetlb_init_hstates();
 	gather_bootmem_prealloc();
@@ -2749,7 +2751,7 @@ static int hugetlb_sysctl_handler_common(bool obey_mempolicy,
 	int ret;
 
 	if (!hugepages_supported())
-		return -ENOTSUPP;
+		return -EOPNOTSUPP;
 
 	table->data = &tmp;
 	table->maxlen = sizeof(unsigned long);
@@ -2790,7 +2792,7 @@ int hugetlb_overcommit_handler(struct ctl_table *table, int write,
 	int ret;
 
 	if (!hugepages_supported())
-		return -ENOTSUPP;
+		return -EOPNOTSUPP;
 
 	tmp = h->nr_overcommit_huge_pages;
 
@@ -3500,7 +3502,7 @@ static int hugetlb_no_page(struct mm_struct *mm, struct vm_area_struct *vma,
 	 * COW. Warn that such a situation has occurred as it may not be obvious
 	 */
 	if (is_vma_resv_set(vma, HPAGE_RESV_UNMAPPED)) {
-		pr_warning("PID %d killed due to inadequate hugepage pool\n",
+		pr_warn_ratelimited("PID %d killed due to inadequate hugepage pool\n",
 			   current->pid);
 		return ret;
 	}
diff --git a/mm/kasan/kasan.c b/mm/kasan/kasan.c
index bc0a8d8b8f42..1ad20ade8c91 100644
--- a/mm/kasan/kasan.c
+++ b/mm/kasan/kasan.c
@@ -20,6 +20,7 @@
 #include <linux/init.h>
 #include <linux/kernel.h>
 #include <linux/kmemleak.h>
+#include <linux/linkage.h>
 #include <linux/memblock.h>
 #include <linux/memory.h>
 #include <linux/mm.h>
@@ -60,6 +61,25 @@ void kasan_unpoison_shadow(const void *address, size_t size)
 	}
 }
 
+static void __kasan_unpoison_stack(struct task_struct *task, void *sp)
+{
+	void *base = task_stack_page(task);
+	size_t size = sp - base;
+
+	kasan_unpoison_shadow(base, size);
+}
+
+/* Unpoison the entire stack for a task. */
+void kasan_unpoison_task_stack(struct task_struct *task)
+{
+	__kasan_unpoison_stack(task, task_stack_page(task) + THREAD_SIZE);
+}
+
+/* Unpoison the stack for the current task beyond a watermark sp value. */
+asmlinkage void kasan_unpoison_remaining_stack(void *sp)
+{
+	__kasan_unpoison_stack(current, sp);
+}
 
 /*
  * All functions below always inlined so compiler could
diff --git a/mm/memory.c b/mm/memory.c
index 635451abc8f7..8132787ae4d5 100644
--- a/mm/memory.c
+++ b/mm/memory.c
@@ -3404,8 +3404,18 @@ static int __handle_mm_fault(struct mm_struct *mm, struct vm_area_struct *vma,
 	if (unlikely(pmd_none(*pmd)) &&
 	    unlikely(__pte_alloc(mm, vma, pmd, address)))
 		return VM_FAULT_OOM;
-	/* if an huge pmd materialized from under us just retry later */
-	if (unlikely(pmd_trans_huge(*pmd) || pmd_devmap(*pmd)))
+	/*
+	 * If a huge pmd materialized under us just retry later.  Use
+	 * pmd_trans_unstable() instead of pmd_trans_huge() to ensure the pmd
+	 * didn't become pmd_trans_huge under us and then back to pmd_none, as
+	 * a result of MADV_DONTNEED running immediately after a huge pmd fault
+	 * in a different thread of this mm, in turn leading to a misleading
+	 * pmd_trans_huge() retval.  All we have to ensure is that it is a
+	 * regular pmd that we can walk with pte_offset_map() and we can do that
+	 * through an atomic read in C, which is what pmd_trans_unstable()
+	 * provides.
+	 */
+	if (unlikely(pmd_trans_unstable(pmd) || pmd_devmap(*pmd)))
 		return 0;
 	/*
 	 * A regular pmd is established and it can't morph into a huge pmd
diff --git a/mm/mempolicy.c b/mm/mempolicy.c
index 4c4187c0e1de..9a3f6b90e628 100644
--- a/mm/mempolicy.c
+++ b/mm/mempolicy.c
@@ -532,7 +532,7 @@ retry:
 		nid = page_to_nid(page);
 		if (node_isset(nid, *qp->nmask) == !!(flags & MPOL_MF_INVERT))
 			continue;
-		if (PageTail(page) && PageAnon(page)) {
+		if (PageTransCompound(page) && PageAnon(page)) {
 			get_page(page);
 			pte_unmap_unlock(pte, ptl);
 			lock_page(page);
diff --git a/mm/migrate.c b/mm/migrate.c
index b1034f9c77e7..3ad0fea5c438 100644
--- a/mm/migrate.c
+++ b/mm/migrate.c
@@ -1582,7 +1582,7 @@ static struct page *alloc_misplaced_dst_page(struct page *page,
 					 (GFP_HIGHUSER_MOVABLE |
 					  __GFP_THISNODE | __GFP_NOMEMALLOC |
 					  __GFP_NORETRY | __GFP_NOWARN) &
-					 ~(__GFP_IO | __GFP_FS), 0);
+					 ~__GFP_RECLAIM, 0);
 
 	return newpage;
 }
diff --git a/mm/mmap.c b/mm/mmap.c
index 2f2415a7a688..76d1ec29149b 100644
--- a/mm/mmap.c
+++ b/mm/mmap.c
@@ -2664,12 +2664,29 @@ SYSCALL_DEFINE5(remap_file_pages, unsigned long, start, unsigned long, size,
 	if (!vma || !(vma->vm_flags & VM_SHARED))
 		goto out;
 
-	if (start < vma->vm_start || start + size > vma->vm_end)
+	if (start < vma->vm_start)
 		goto out;
 
-	if (pgoff == linear_page_index(vma, start)) {
-		ret = 0;
-		goto out;
+	if (start + size > vma->vm_end) {
+		struct vm_area_struct *next;
+
+		for (next = vma->vm_next; next; next = next->vm_next) {
+			/* hole between vmas ? */
+			if (next->vm_start != next->vm_prev->vm_end)
+				goto out;
+
+			if (next->vm_file != vma->vm_file)
+				goto out;
+
+			if (next->vm_flags != vma->vm_flags)
+				goto out;
+
+			if (start + size <= next->vm_end)
+				break;
+		}
+
+		if (!next)
+			goto out;
 	}
 
 	prot |= vma->vm_flags & VM_READ ? PROT_READ : 0;
@@ -2679,9 +2696,16 @@ SYSCALL_DEFINE5(remap_file_pages, unsigned long, start, unsigned long, size,
 	flags &= MAP_NONBLOCK;
 	flags |= MAP_SHARED | MAP_FIXED | MAP_POPULATE;
 	if (vma->vm_flags & VM_LOCKED) {
+		struct vm_area_struct *tmp;
 		flags |= MAP_LOCKED;
+
 		/* drop PG_Mlocked flag for over-mapped range */
-		munlock_vma_pages_range(vma, start, start + size);
+		for (tmp = vma; tmp->vm_start >= start + size;
+				tmp = tmp->vm_next) {
+			munlock_vma_pages_range(tmp,
+					max(tmp->vm_start, start),
+					min(tmp->vm_end, start + size));
+		}
 	}
 
 	file = get_file(vma->vm_file);
diff --git a/mm/mprotect.c b/mm/mprotect.c
index 8eb7bb40dc40..f7cb3d4d9c2e 100644
--- a/mm/mprotect.c
+++ b/mm/mprotect.c
@@ -160,9 +160,11 @@ static inline unsigned long change_pmd_range(struct vm_area_struct *vma,
 		}
 
 		if (pmd_trans_huge(*pmd) || pmd_devmap(*pmd)) {
-			if (next - addr != HPAGE_PMD_SIZE)
+			if (next - addr != HPAGE_PMD_SIZE) {
 				split_huge_pmd(vma, pmd, addr);
-			else {
+				if (pmd_none(*pmd))
+					continue;
+			} else {
 				int nr_ptes = change_huge_pmd(vma, pmd, addr,
 						newprot, prot_numa);
 
diff --git a/mm/mremap.c b/mm/mremap.c
index d77946a997f7..8eeba02fc991 100644
--- a/mm/mremap.c
+++ b/mm/mremap.c
@@ -210,6 +210,8 @@ unsigned long move_page_tables(struct vm_area_struct *vma,
 				}
 			}
 			split_huge_pmd(vma, old_pmd, old_addr);
+			if (pmd_none(*old_pmd))
+				continue;
 			VM_BUG_ON(pmd_trans_huge(*old_pmd));
 		}
 		if (pmd_none(*new_pmd) && __pte_alloc(new_vma->vm_mm, new_vma,
diff --git a/mm/pgtable-generic.c b/mm/pgtable-generic.c
index 9d4767698a1c..06a005b979a7 100644
--- a/mm/pgtable-generic.c
+++ b/mm/pgtable-generic.c
@@ -90,9 +90,9 @@ pte_t ptep_clear_flush(struct vm_area_struct *vma, unsigned long address,
  * ARCHes with special requirements for evicting THP backing TLB entries can
  * implement this. Otherwise also, it can help optimize normal TLB flush in
  * THP regime. stock flush_tlb_range() typically has optimization to nuke the
- * entire TLB TLB if flush span is greater than a threshhold, which will
+ * entire TLB if flush span is greater than a threshold, which will
  * likely be true for a single huge page. Thus a single thp flush will
- * invalidate the entire TLB which is not desitable.
+ * invalidate the entire TLB which is not desirable.
  * e.g. see arch/arc: flush_pmd_tlb_range
  */
 #define flush_pmd_tlb_range(vma, addr, end)	flush_tlb_range(vma, addr, end)
@@ -195,7 +195,9 @@ pmd_t pmdp_collapse_flush(struct vm_area_struct *vma, unsigned long address,
 	VM_BUG_ON(address & ~HPAGE_PMD_MASK);
 	VM_BUG_ON(pmd_trans_huge(*pmdp));
 	pmd = pmdp_huge_get_and_clear(vma->vm_mm, address, pmdp);
-	flush_pmd_tlb_range(vma, address, address + HPAGE_PMD_SIZE);
+
+	/* collapse entails shooting down ptes not pmd */
+	flush_tlb_range(vma, address, address + HPAGE_PMD_SIZE);
 	return pmd;
 }
 #endif
diff --git a/mm/slab.c b/mm/slab.c
index 6ecc697a8bc4..621fbcb35a36 100644
--- a/mm/slab.c
+++ b/mm/slab.c
@@ -2275,7 +2275,7 @@ __kmem_cache_create (struct kmem_cache *cachep, unsigned long flags)
 
 	err = setup_cpu_cache(cachep, gfp);
 	if (err) {
-		__kmem_cache_shutdown(cachep);
+		__kmem_cache_release(cachep);
 		return err;
 	}
 
@@ -2414,12 +2414,13 @@ int __kmem_cache_shrink(struct kmem_cache *cachep, bool deactivate)
 
 int __kmem_cache_shutdown(struct kmem_cache *cachep)
 {
+	return __kmem_cache_shrink(cachep, false);
+}
+
+void __kmem_cache_release(struct kmem_cache *cachep)
+{
 	int i;
 	struct kmem_cache_node *n;
-	int rc = __kmem_cache_shrink(cachep, false);
-
-	if (rc)
-		return rc;
 
 	free_percpu(cachep->cpu_cache);
 
@@ -2430,7 +2431,6 @@ int __kmem_cache_shutdown(struct kmem_cache *cachep)
 		kfree(n);
 		cachep->node[i] = NULL;
 	}
-	return 0;
 }
 
 /*
diff --git a/mm/slab.h b/mm/slab.h
index 834ad240c0bb..2eedacea439d 100644
--- a/mm/slab.h
+++ b/mm/slab.h
@@ -140,6 +140,7 @@ static inline unsigned long kmem_cache_flags(unsigned long object_size,
 #define CACHE_CREATE_MASK (SLAB_CORE_FLAGS | SLAB_DEBUG_FLAGS | SLAB_CACHE_FLAGS)
 
 int __kmem_cache_shutdown(struct kmem_cache *);
+void __kmem_cache_release(struct kmem_cache *);
 int __kmem_cache_shrink(struct kmem_cache *, bool);
 void slab_kmem_cache_release(struct kmem_cache *);
 
diff --git a/mm/slab_common.c b/mm/slab_common.c
index b50aef01ccf7..065b7bdabdc3 100644
--- a/mm/slab_common.c
+++ b/mm/slab_common.c
@@ -693,6 +693,7 @@ static inline int shutdown_memcg_caches(struct kmem_cache *s,
 
 void slab_kmem_cache_release(struct kmem_cache *s)
 {
+	__kmem_cache_release(s);
 	destroy_memcg_params(s);
 	kfree_const(s->name);
 	kmem_cache_free(kmem_cache, s);
diff --git a/mm/slob.c b/mm/slob.c
index 17e8f8cc7c53..5ec158054ffe 100644
--- a/mm/slob.c
+++ b/mm/slob.c
@@ -630,6 +630,10 @@ int __kmem_cache_shutdown(struct kmem_cache *c)
 	return 0;
 }
 
+void __kmem_cache_release(struct kmem_cache *c)
+{
+}
+
 int __kmem_cache_shrink(struct kmem_cache *d, bool deactivate)
 {
 	return 0;
diff --git a/mm/slub.c b/mm/slub.c
index 2e1355ac056b..d8fbd4a6ed59 100644
--- a/mm/slub.c
+++ b/mm/slub.c
@@ -1592,18 +1592,12 @@ static inline void add_partial(struct kmem_cache_node *n,
 	__add_partial(n, page, tail);
 }
 
-static inline void
-__remove_partial(struct kmem_cache_node *n, struct page *page)
-{
-	list_del(&page->lru);
-	n->nr_partial--;
-}
-
 static inline void remove_partial(struct kmem_cache_node *n,
 					struct page *page)
 {
 	lockdep_assert_held(&n->list_lock);
-	__remove_partial(n, page);
+	list_del(&page->lru);
+	n->nr_partial--;
 }
 
 /*
@@ -3184,6 +3178,12 @@ static void free_kmem_cache_nodes(struct kmem_cache *s)
 	}
 }
 
+void __kmem_cache_release(struct kmem_cache *s)
+{
+	free_percpu(s->cpu_slab);
+	free_kmem_cache_nodes(s);
+}
+
 static int init_kmem_cache_nodes(struct kmem_cache *s)
 {
 	int node;
@@ -3443,28 +3443,31 @@ static void list_slab_objects(struct kmem_cache *s, struct page *page,
 
 /*
  * Attempt to free all partial slabs on a node.
- * This is called from kmem_cache_close(). We must be the last thread
- * using the cache and therefore we do not need to lock anymore.
+ * This is called from __kmem_cache_shutdown(). We must take list_lock
+ * because sysfs file might still access partial list after the shutdowning.
  */
 static void free_partial(struct kmem_cache *s, struct kmem_cache_node *n)
 {
 	struct page *page, *h;
 
+	BUG_ON(irqs_disabled());
+	spin_lock_irq(&n->list_lock);
 	list_for_each_entry_safe(page, h, &n->partial, lru) {
 		if (!page->inuse) {
-			__remove_partial(n, page);
+			remove_partial(n, page);
 			discard_slab(s, page);
 		} else {
 			list_slab_objects(s, page,
-			"Objects remaining in %s on kmem_cache_close()");
+			"Objects remaining in %s on __kmem_cache_shutdown()");
 		}
 	}
+	spin_unlock_irq(&n->list_lock);
 }
 
 /*
  * Release all resources used by a slab cache.
  */
-static inline int kmem_cache_close(struct kmem_cache *s)
+int __kmem_cache_shutdown(struct kmem_cache *s)
 {
 	int node;
 	struct kmem_cache_node *n;
@@ -3476,16 +3479,9 @@ static inline int kmem_cache_close(struct kmem_cache *s)
 		if (n->nr_partial || slabs_node(s, node))
 			return 1;
 	}
-	free_percpu(s->cpu_slab);
-	free_kmem_cache_nodes(s);
 	return 0;
 }
 
-int __kmem_cache_shutdown(struct kmem_cache *s)
-{
-	return kmem_cache_close(s);
-}
-
 /********************************************************************
  *		Kmalloc subsystem
  *******************************************************************/
@@ -3980,7 +3976,7 @@ int __kmem_cache_create(struct kmem_cache *s, unsigned long flags)
 	memcg_propagate_slab_attrs(s);
 	err = sysfs_slab_add(s);
 	if (err)
-		kmem_cache_close(s);
+		__kmem_cache_release(s);
 
 	return err;
 }
diff --git a/net/appletalk/ddp.c b/net/appletalk/ddp.c
index d5871ac493eb..f066781be3c8 100644
--- a/net/appletalk/ddp.c
+++ b/net/appletalk/ddp.c
@@ -1625,7 +1625,7 @@ static int atalk_sendmsg(struct socket *sock, struct msghdr *msg, size_t len)
 
 		rt = atrtr_find(&at_hint);
 	}
-	err = ENETUNREACH;
+	err = -ENETUNREACH;
 	if (!rt)
 		goto out;
 
diff --git a/net/batman-adv/gateway_client.c b/net/batman-adv/gateway_client.c
index e6c8382c79ba..ccf70bed0d0c 100644
--- a/net/batman-adv/gateway_client.c
+++ b/net/batman-adv/gateway_client.c
@@ -527,11 +527,12 @@ void batadv_gw_node_update(struct batadv_priv *bat_priv,
 		 * gets dereferenced.
 		 */
 		spin_lock_bh(&bat_priv->gw.list_lock);
-		hlist_del_init_rcu(&gw_node->list);
+		if (!hlist_unhashed(&gw_node->list)) {
+			hlist_del_init_rcu(&gw_node->list);
+			batadv_gw_node_free_ref(gw_node);
+		}
 		spin_unlock_bh(&bat_priv->gw.list_lock);
 
-		batadv_gw_node_free_ref(gw_node);
-
 		curr_gw = batadv_gw_get_selected_gw_node(bat_priv);
 		if (gw_node == curr_gw)
 			batadv_gw_reselect(bat_priv);
diff --git a/net/batman-adv/hard-interface.c b/net/batman-adv/hard-interface.c
index 01acccc4d218..57f7107169f5 100644
--- a/net/batman-adv/hard-interface.c
+++ b/net/batman-adv/hard-interface.c
@@ -76,6 +76,28 @@ out:
 }
 
 /**
+ * batadv_mutual_parents - check if two devices are each others parent
+ * @dev1: 1st net_device
+ * @dev2: 2nd net_device
+ *
+ * veth devices come in pairs and each is the parent of the other!
+ *
+ * Return: true if the devices are each others parent, otherwise false
+ */
+static bool batadv_mutual_parents(const struct net_device *dev1,
+				  const struct net_device *dev2)
+{
+	int dev1_parent_iflink = dev_get_iflink(dev1);
+	int dev2_parent_iflink = dev_get_iflink(dev2);
+
+	if (!dev1_parent_iflink || !dev2_parent_iflink)
+		return false;
+
+	return (dev1_parent_iflink == dev2->ifindex) &&
+	       (dev2_parent_iflink == dev1->ifindex);
+}
+
+/**
  * batadv_is_on_batman_iface - check if a device is a batman iface descendant
  * @net_dev: the device to check
  *
@@ -108,6 +130,9 @@ static bool batadv_is_on_batman_iface(const struct net_device *net_dev)
 	if (WARN(!parent_dev, "Cannot find parent device"))
 		return false;
 
+	if (batadv_mutual_parents(net_dev, parent_dev))
+		return false;
+
 	ret = batadv_is_on_batman_iface(parent_dev);
 
 	return ret;
diff --git a/net/batman-adv/translation-table.c b/net/batman-adv/translation-table.c
index cdfc85fa2743..0e80fd1461ab 100644
--- a/net/batman-adv/translation-table.c
+++ b/net/batman-adv/translation-table.c
@@ -303,9 +303,11 @@ static void batadv_tt_global_size_mod(struct batadv_orig_node *orig_node,
 
 	if (atomic_add_return(v, &vlan->tt.num_entries) == 0) {
 		spin_lock_bh(&orig_node->vlan_list_lock);
-		hlist_del_init_rcu(&vlan->list);
+		if (!hlist_unhashed(&vlan->list)) {
+			hlist_del_init_rcu(&vlan->list);
+			batadv_orig_node_vlan_free_ref(vlan);
+		}
 		spin_unlock_bh(&orig_node->vlan_list_lock);
-		batadv_orig_node_vlan_free_ref(vlan);
 	}
 
 	batadv_orig_node_vlan_free_ref(vlan);
diff --git a/net/bluetooth/hci_core.c b/net/bluetooth/hci_core.c
index 47bcef754796..883c821a9e78 100644
--- a/net/bluetooth/hci_core.c
+++ b/net/bluetooth/hci_core.c
@@ -4112,8 +4112,10 @@ void hci_req_cmd_complete(struct hci_dev *hdev, u16 opcode, u8 status,
 			break;
 		}
 
-		*req_complete = bt_cb(skb)->hci.req_complete;
-		*req_complete_skb = bt_cb(skb)->hci.req_complete_skb;
+		if (bt_cb(skb)->hci.req_flags & HCI_REQ_SKB)
+			*req_complete_skb = bt_cb(skb)->hci.req_complete_skb;
+		else
+			*req_complete = bt_cb(skb)->hci.req_complete;
 		kfree_skb(skb);
 	}
 	spin_unlock_irqrestore(&hdev->cmd_q.lock, flags);
diff --git a/net/bridge/br_fdb.c b/net/bridge/br_fdb.c
index 82e3e9705017..dcea4f4c62b3 100644
--- a/net/bridge/br_fdb.c
+++ b/net/bridge/br_fdb.c
@@ -723,6 +723,8 @@ int br_fdb_dump(struct sk_buff *skb,
 		struct net_bridge_fdb_entry *f;
 
 		hlist_for_each_entry_rcu(f, &br->hash[i], hlist) {
+			int err;
+
 			if (idx < cb->args[0])
 				goto skip;
 
@@ -741,12 +743,15 @@ int br_fdb_dump(struct sk_buff *skb,
 			if (!filter_dev && f->dst)
 				goto skip;
 
-			if (fdb_fill_info(skb, br, f,
-					  NETLINK_CB(cb->skb).portid,
-					  cb->nlh->nlmsg_seq,
-					  RTM_NEWNEIGH,
-					  NLM_F_MULTI) < 0)
+			err = fdb_fill_info(skb, br, f,
+					    NETLINK_CB(cb->skb).portid,
+					    cb->nlh->nlmsg_seq,
+					    RTM_NEWNEIGH,
+					    NLM_F_MULTI);
+			if (err < 0) {
+				cb->args[1] = err;
 				break;
+			}
 skip:
 			++idx;
 		}
diff --git a/net/bridge/br_mdb.c b/net/bridge/br_mdb.c
index 30e105f57f0d..74c278e00225 100644
--- a/net/bridge/br_mdb.c
+++ b/net/bridge/br_mdb.c
@@ -425,8 +425,8 @@ static int br_mdb_add_group(struct net_bridge *br, struct net_bridge_port *port,
 	mp = br_mdb_ip_get(mdb, group);
 	if (!mp) {
 		mp = br_multicast_new_group(br, port, group);
-		err = PTR_ERR(mp);
-		if (IS_ERR(mp))
+		err = PTR_ERR_OR_ZERO(mp);
+		if (err)
 			return err;
 	}
 
diff --git a/net/caif/cfrfml.c b/net/caif/cfrfml.c
index 61d7617d9249..b82440e1fcb4 100644
--- a/net/caif/cfrfml.c
+++ b/net/caif/cfrfml.c
@@ -159,7 +159,7 @@ static int cfrfml_receive(struct cflayer *layr, struct cfpkt *pkt)
 		tmppkt = NULL;
 
 		/* Verify that length is correct */
-		err = EPROTO;
+		err = -EPROTO;
 		if (rfml->pdu_size != cfpkt_getlen(pkt) - RFM_HEAD_SIZE + 1)
 			goto out;
 	}
diff --git a/net/ceph/messenger.c b/net/ceph/messenger.c
index 9cfedf565f5b..9382619a405b 100644
--- a/net/ceph/messenger.c
+++ b/net/ceph/messenger.c
@@ -1197,6 +1197,13 @@ static bool ceph_msg_data_advance(struct ceph_msg_data_cursor *cursor,
 	return new_piece;
 }
 
+static size_t sizeof_footer(struct ceph_connection *con)
+{
+	return (con->peer_features & CEPH_FEATURE_MSG_AUTH) ?
+	    sizeof(struct ceph_msg_footer) :
+	    sizeof(struct ceph_msg_footer_old);
+}
+
 static void prepare_message_data(struct ceph_msg *msg, u32 data_len)
 {
 	BUG_ON(!msg);
@@ -2335,9 +2342,9 @@ static int read_partial_message(struct ceph_connection *con)
 			ceph_pr_addr(&con->peer_addr.in_addr),
 			seq, con->in_seq + 1);
 		con->in_base_pos = -front_len - middle_len - data_len -
-			sizeof(m->footer);
+			sizeof_footer(con);
 		con->in_tag = CEPH_MSGR_TAG_READY;
-		return 0;
+		return 1;
 	} else if ((s64)seq - (s64)con->in_seq > 1) {
 		pr_err("read_partial_message bad seq %lld expected %lld\n",
 		       seq, con->in_seq + 1);
@@ -2360,10 +2367,10 @@ static int read_partial_message(struct ceph_connection *con)
 			/* skip this message */
 			dout("alloc_msg said skip message\n");
 			con->in_base_pos = -front_len - middle_len - data_len -
-				sizeof(m->footer);
+				sizeof_footer(con);
 			con->in_tag = CEPH_MSGR_TAG_READY;
 			con->in_seq++;
-			return 0;
+			return 1;
 		}
 
 		BUG_ON(!con->in_msg);
diff --git a/net/ceph/osd_client.c b/net/ceph/osd_client.c
index 3534e12683d3..5bc053778fed 100644
--- a/net/ceph/osd_client.c
+++ b/net/ceph/osd_client.c
@@ -2853,8 +2853,8 @@ static struct ceph_msg *get_reply(struct ceph_connection *con,
 	mutex_lock(&osdc->request_mutex);
 	req = __lookup_request(osdc, tid);
 	if (!req) {
-		pr_warn("%s osd%d tid %llu unknown, skipping\n",
-			__func__, osd->o_osd, tid);
+		dout("%s osd%d tid %llu unknown, skipping\n", __func__,
+		     osd->o_osd, tid);
 		m = NULL;
 		*skip = 1;
 		goto out;
diff --git a/net/core/dev.c b/net/core/dev.c
index 8cba3d852f25..0ef061b2badc 100644
--- a/net/core/dev.c
+++ b/net/core/dev.c
@@ -5379,12 +5379,12 @@ void *netdev_lower_get_next(struct net_device *dev, struct list_head **iter)
 {
 	struct netdev_adjacent *lower;
 
-	lower = list_entry((*iter)->next, struct netdev_adjacent, list);
+	lower = list_entry(*iter, struct netdev_adjacent, list);
 
 	if (&lower->list == &dev->adj_list.lower)
 		return NULL;
 
-	*iter = &lower->list;
+	*iter = lower->list.next;
 
 	return lower->dev;
 }
@@ -7422,8 +7422,10 @@ struct net_device *alloc_netdev_mqs(int sizeof_priv, const char *name,
 	dev->priv_flags = IFF_XMIT_DST_RELEASE | IFF_XMIT_DST_RELEASE_PERM;
 	setup(dev);
 
-	if (!dev->tx_queue_len)
+	if (!dev->tx_queue_len) {
 		dev->priv_flags |= IFF_NO_QUEUE;
+		dev->tx_queue_len = 1;
+	}
 
 	dev->num_tx_queues = txqs;
 	dev->real_num_tx_queues = txqs;
diff --git a/net/core/filter.c b/net/core/filter.c
index 94d26201080d..bba502f7cd57 100644
--- a/net/core/filter.c
+++ b/net/core/filter.c
@@ -1752,7 +1752,7 @@ static u64 bpf_skb_set_tunnel_key(u64 r1, u64 r2, u64 size, u64 flags, u64 r5)
 	u8 compat[sizeof(struct bpf_tunnel_key)];
 	struct ip_tunnel_info *info;
 
-	if (unlikely(flags & ~(BPF_F_TUNINFO_IPV6)))
+	if (unlikely(flags & ~(BPF_F_TUNINFO_IPV6 | BPF_F_ZERO_CSUM_TX)))
 		return -EINVAL;
 	if (unlikely(size != sizeof(struct bpf_tunnel_key))) {
 		switch (size) {
@@ -1776,7 +1776,7 @@ static u64 bpf_skb_set_tunnel_key(u64 r1, u64 r2, u64 size, u64 flags, u64 r5)
 	info = &md->u.tun_info;
 	info->mode = IP_TUNNEL_INFO_TX;
 
-	info->key.tun_flags = TUNNEL_KEY;
+	info->key.tun_flags = TUNNEL_KEY | TUNNEL_CSUM;
 	info->key.tun_id = cpu_to_be64(from->tunnel_id);
 	info->key.tos = from->tunnel_tos;
 	info->key.ttl = from->tunnel_ttl;
@@ -1787,6 +1787,8 @@ static u64 bpf_skb_set_tunnel_key(u64 r1, u64 r2, u64 size, u64 flags, u64 r5)
 		       sizeof(from->remote_ipv6));
 	} else {
 		info->key.u.ipv4.dst = cpu_to_be32(from->remote_ipv4);
+		if (flags & BPF_F_ZERO_CSUM_TX)
+			info->key.tun_flags &= ~TUNNEL_CSUM;
 	}
 
 	return 0;
diff --git a/net/core/flow_dissector.c b/net/core/flow_dissector.c
index d79699c9d1b9..12e700332010 100644
--- a/net/core/flow_dissector.c
+++ b/net/core/flow_dissector.c
@@ -208,7 +208,6 @@ ip:
 	case htons(ETH_P_IPV6): {
 		const struct ipv6hdr *iph;
 		struct ipv6hdr _iph;
-		__be32 flow_label;
 
 ipv6:
 		iph = __skb_header_pointer(skb, nhoff, sizeof(_iph), data, hlen, &_iph);
@@ -230,8 +229,12 @@ ipv6:
 			key_control->addr_type = FLOW_DISSECTOR_KEY_IPV6_ADDRS;
 		}
 
-		flow_label = ip6_flowlabel(iph);
-		if (flow_label) {
+		if ((dissector_uses_key(flow_dissector,
+					FLOW_DISSECTOR_KEY_FLOW_LABEL) ||
+		     (flags & FLOW_DISSECTOR_F_STOP_AT_FLOW_LABEL)) &&
+		    ip6_flowlabel(iph)) {
+			__be32 flow_label = ip6_flowlabel(iph);
+
 			if (dissector_uses_key(flow_dissector,
 					       FLOW_DISSECTOR_KEY_FLOW_LABEL)) {
 				key_tags = skb_flow_dissector_target(flow_dissector,
@@ -396,6 +399,13 @@ ip_proto_again:
 				goto out_bad;
 			proto = eth->h_proto;
 			nhoff += sizeof(*eth);
+
+			/* Cap headers that we access via pointers at the
+			 * end of the Ethernet header as our maximum alignment
+			 * at that point is only 2 bytes.
+			 */
+			if (NET_IP_ALIGN)
+				hlen = nhoff;
 		}
 
 		key_control->flags |= FLOW_DIS_ENCAPSULATION;
diff --git a/net/core/rtnetlink.c b/net/core/rtnetlink.c
index d735e854f916..8261d95dd846 100644
--- a/net/core/rtnetlink.c
+++ b/net/core/rtnetlink.c
@@ -2911,6 +2911,7 @@ int ndo_dflt_fdb_dump(struct sk_buff *skb,
 	nlmsg_populate_fdb(skb, cb, dev, &idx, &dev->mc);
 out:
 	netif_addr_unlock_bh(dev);
+	cb->args[1] = err;
 	return idx;
 }
 EXPORT_SYMBOL(ndo_dflt_fdb_dump);
@@ -2944,6 +2945,7 @@ static int rtnl_fdb_dump(struct sk_buff *skb, struct netlink_callback *cb)
 		ops = br_dev->netdev_ops;
 	}
 
+	cb->args[1] = 0;
 	for_each_netdev(net, dev) {
 		if (brport_idx && (dev->ifindex != brport_idx))
 			continue;
@@ -2971,12 +2973,16 @@ static int rtnl_fdb_dump(struct sk_buff *skb, struct netlink_callback *cb)
 				idx = cops->ndo_fdb_dump(skb, cb, br_dev, dev,
 							 idx);
 		}
+		if (cb->args[1] == -EMSGSIZE)
+			break;
 
 		if (dev->netdev_ops->ndo_fdb_dump)
 			idx = dev->netdev_ops->ndo_fdb_dump(skb, cb, dev, NULL,
 							    idx);
 		else
 			idx = ndo_dflt_fdb_dump(skb, cb, dev, NULL, idx);
+		if (cb->args[1] == -EMSGSIZE)
+			break;
 
 		cops = NULL;
 	}
diff --git a/net/core/scm.c b/net/core/scm.c
index 14596fb37172..2696aefdc148 100644
--- a/net/core/scm.c
+++ b/net/core/scm.c
@@ -87,6 +87,7 @@ static int scm_fp_copy(struct cmsghdr *cmsg, struct scm_fp_list **fplp)
 		*fplp = fpl;
 		fpl->count = 0;
 		fpl->max = SCM_MAX_FD;
+		fpl->user = NULL;
 	}
 	fpp = &fpl->fp[fpl->count];
 
@@ -107,6 +108,10 @@ static int scm_fp_copy(struct cmsghdr *cmsg, struct scm_fp_list **fplp)
 		*fpp++ = file;
 		fpl->count++;
 	}
+
+	if (!fpl->user)
+		fpl->user = get_uid(current_user());
+
 	return num;
 }
 
@@ -119,6 +124,7 @@ void __scm_destroy(struct scm_cookie *scm)
 		scm->fp = NULL;
 		for (i=fpl->count-1; i>=0; i--)
 			fput(fpl->fp[i]);
+		free_uid(fpl->user);
 		kfree(fpl);
 	}
 }
@@ -336,6 +342,7 @@ struct scm_fp_list *scm_fp_dup(struct scm_fp_list *fpl)
 		for (i = 0; i < fpl->count; i++)
 			get_file(fpl->fp[i]);
 		new_fpl->max = new_fpl->count;
+		new_fpl->user = get_uid(fpl->user);
 	}
 	return new_fpl;
 }
diff --git a/net/core/skbuff.c b/net/core/skbuff.c
index b2df375ec9c2..8616d1147c93 100644
--- a/net/core/skbuff.c
+++ b/net/core/skbuff.c
@@ -79,6 +79,8 @@
 
 struct kmem_cache *skbuff_head_cache __read_mostly;
 static struct kmem_cache *skbuff_fclone_cache __read_mostly;
+int sysctl_max_skb_frags __read_mostly = MAX_SKB_FRAGS;
+EXPORT_SYMBOL(sysctl_max_skb_frags);
 
 /**
  *	skb_panic - private function for out-of-line support
@@ -2946,6 +2948,24 @@ int skb_append_pagefrags(struct sk_buff *skb, struct page *page,
 EXPORT_SYMBOL_GPL(skb_append_pagefrags);
 
 /**
+ *	skb_push_rcsum - push skb and update receive checksum
+ *	@skb: buffer to update
+ *	@len: length of data pulled
+ *
+ *	This function performs an skb_push on the packet and updates
+ *	the CHECKSUM_COMPLETE checksum.  It should be used on
+ *	receive path processing instead of skb_push unless you know
+ *	that the checksum difference is zero (e.g., a valid IP header)
+ *	or you are setting ip_summed to CHECKSUM_NONE.
+ */
+static unsigned char *skb_push_rcsum(struct sk_buff *skb, unsigned len)
+{
+	skb_push(skb, len);
+	skb_postpush_rcsum(skb, skb->data, len);
+	return skb->data;
+}
+
+/**
  *	skb_pull_rcsum - pull skb and update receive checksum
  *	@skb: buffer to update
  *	@len: length of data pulled
@@ -4082,9 +4102,9 @@ struct sk_buff *skb_checksum_trimmed(struct sk_buff *skb,
 	if (!pskb_may_pull(skb_chk, offset))
 		goto err;
 
-	__skb_pull(skb_chk, offset);
+	skb_pull_rcsum(skb_chk, offset);
 	ret = skb_chkf(skb_chk);
-	__skb_push(skb_chk, offset);
+	skb_push_rcsum(skb_chk, offset);
 
 	if (ret)
 		goto err;
diff --git a/net/core/sysctl_net_core.c b/net/core/sysctl_net_core.c
index 95b6139d710c..a6beb7b6ae55 100644
--- a/net/core/sysctl_net_core.c
+++ b/net/core/sysctl_net_core.c
@@ -26,6 +26,7 @@ static int zero = 0;
 static int one = 1;
 static int min_sndbuf = SOCK_MIN_SNDBUF;
 static int min_rcvbuf = SOCK_MIN_RCVBUF;
+static int max_skb_frags = MAX_SKB_FRAGS;
 
 static int net_msg_warn;	/* Unused, but still a sysctl */
 
@@ -392,6 +393,15 @@ static struct ctl_table net_core_table[] = {
 		.mode		= 0644,
 		.proc_handler	= proc_dointvec
 	},
+	{
+		.procname	= "max_skb_frags",
+		.data		= &sysctl_max_skb_frags,
+		.maxlen		= sizeof(int),
+		.mode		= 0644,
+		.proc_handler	= proc_dointvec_minmax,
+		.extra1		= &one,
+		.extra2		= &max_skb_frags,
+	},
 	{ }
 };
 
diff --git a/net/dccp/ipv4.c b/net/dccp/ipv4.c
index 5684e14932bd..902d606324a0 100644
--- a/net/dccp/ipv4.c
+++ b/net/dccp/ipv4.c
@@ -824,26 +824,26 @@ lookup:
 
 	if (sk->sk_state == DCCP_NEW_SYN_RECV) {
 		struct request_sock *req = inet_reqsk(sk);
-		struct sock *nsk = NULL;
+		struct sock *nsk;
 
 		sk = req->rsk_listener;
-		if (likely(sk->sk_state == DCCP_LISTEN)) {
-			nsk = dccp_check_req(sk, skb, req);
-		} else {
+		if (unlikely(sk->sk_state != DCCP_LISTEN)) {
 			inet_csk_reqsk_queue_drop_and_put(sk, req);
 			goto lookup;
 		}
+		sock_hold(sk);
+		nsk = dccp_check_req(sk, skb, req);
 		if (!nsk) {
 			reqsk_put(req);
-			goto discard_it;
+			goto discard_and_relse;
 		}
 		if (nsk == sk) {
-			sock_hold(sk);
 			reqsk_put(req);
 		} else if (dccp_child_process(sk, nsk, skb)) {
 			dccp_v4_ctl_send_reset(sk, skb);
-			goto discard_it;
+			goto discard_and_relse;
 		} else {
+			sock_put(sk);
 			return 0;
 		}
 	}
diff --git a/net/dccp/ipv6.c b/net/dccp/ipv6.c
index 9c6d0508e63a..b8608b71a66d 100644
--- a/net/dccp/ipv6.c
+++ b/net/dccp/ipv6.c
@@ -691,26 +691,26 @@ lookup:
 
 	if (sk->sk_state == DCCP_NEW_SYN_RECV) {
 		struct request_sock *req = inet_reqsk(sk);
-		struct sock *nsk = NULL;
+		struct sock *nsk;
 
 		sk = req->rsk_listener;
-		if (likely(sk->sk_state == DCCP_LISTEN)) {
-			nsk = dccp_check_req(sk, skb, req);
-		} else {
+		if (unlikely(sk->sk_state != DCCP_LISTEN)) {
 			inet_csk_reqsk_queue_drop_and_put(sk, req);
 			goto lookup;
 		}
+		sock_hold(sk);
+		nsk = dccp_check_req(sk, skb, req);
 		if (!nsk) {
 			reqsk_put(req);
-			goto discard_it;
+			goto discard_and_relse;
 		}
 		if (nsk == sk) {
-			sock_hold(sk);
 			reqsk_put(req);
 		} else if (dccp_child_process(sk, nsk, skb)) {
 			dccp_v6_ctl_send_reset(sk, skb);
-			goto discard_it;
+			goto discard_and_relse;
 		} else {
+			sock_put(sk);
 			return 0;
 		}
 	}
diff --git a/net/dsa/slave.c b/net/dsa/slave.c
index 40b9ca72aae3..ab24521beb4d 100644
--- a/net/dsa/slave.c
+++ b/net/dsa/slave.c
@@ -1194,7 +1194,6 @@ int dsa_slave_create(struct dsa_switch *ds, struct device *parent,
 	if (ret) {
 		netdev_err(master, "error %d registering interface %s\n",
 			   ret, slave_dev->name);
-		phy_disconnect(p->phy);
 		ds->ports[port] = NULL;
 		free_netdev(slave_dev);
 		return ret;
@@ -1205,6 +1204,7 @@ int dsa_slave_create(struct dsa_switch *ds, struct device *parent,
 	ret = dsa_slave_phy_setup(p, slave_dev);
 	if (ret) {
 		netdev_err(master, "error %d setting up slave phy\n", ret);
+		unregister_netdev(slave_dev);
 		free_netdev(slave_dev);
 		return ret;
 	}
diff --git a/net/ipv4/devinet.c b/net/ipv4/devinet.c
index cebd9d31e65a..f6303b17546b 100644
--- a/net/ipv4/devinet.c
+++ b/net/ipv4/devinet.c
@@ -1847,7 +1847,7 @@ static int inet_netconf_get_devconf(struct sk_buff *in_skb,
 	if (err < 0)
 		goto errout;
 
-	err = EINVAL;
+	err = -EINVAL;
 	if (!tb[NETCONFA_IFINDEX])
 		goto errout;
 
diff --git a/net/ipv4/igmp.c b/net/ipv4/igmp.c
index 05e4cba14162..b3086cf27027 100644
--- a/net/ipv4/igmp.c
+++ b/net/ipv4/igmp.c
@@ -356,9 +356,8 @@ static struct sk_buff *igmpv3_newpack(struct net_device *dev, unsigned int mtu)
 	skb_dst_set(skb, &rt->dst);
 	skb->dev = dev;
 
-	skb->reserved_tailroom = skb_end_offset(skb) -
-				 min(mtu, skb_end_offset(skb));
 	skb_reserve(skb, hlen);
+	skb_tailroom_reserve(skb, mtu, tlen);
 
 	skb_reset_network_header(skb);
 	pip = ip_hdr(skb);
diff --git a/net/ipv4/inet_connection_sock.c b/net/ipv4/inet_connection_sock.c
index 46b9c887bede..64148914803a 100644
--- a/net/ipv4/inet_connection_sock.c
+++ b/net/ipv4/inet_connection_sock.c
@@ -789,14 +789,16 @@ static void inet_child_forget(struct sock *sk, struct request_sock *req,
 	reqsk_put(req);
 }
 
-void inet_csk_reqsk_queue_add(struct sock *sk, struct request_sock *req,
-			      struct sock *child)
+struct sock *inet_csk_reqsk_queue_add(struct sock *sk,
+				      struct request_sock *req,
+				      struct sock *child)
 {
 	struct request_sock_queue *queue = &inet_csk(sk)->icsk_accept_queue;
 
 	spin_lock(&queue->rskq_lock);
 	if (unlikely(sk->sk_state != TCP_LISTEN)) {
 		inet_child_forget(sk, req, child);
+		child = NULL;
 	} else {
 		req->sk = child;
 		req->dl_next = NULL;
@@ -808,6 +810,7 @@ void inet_csk_reqsk_queue_add(struct sock *sk, struct request_sock *req,
 		sk_acceptq_added(sk);
 	}
 	spin_unlock(&queue->rskq_lock);
+	return child;
 }
 EXPORT_SYMBOL(inet_csk_reqsk_queue_add);
 
@@ -817,11 +820,8 @@ struct sock *inet_csk_complete_hashdance(struct sock *sk, struct sock *child,
 	if (own_req) {
 		inet_csk_reqsk_queue_drop(sk, req);
 		reqsk_queue_removed(&inet_csk(sk)->icsk_accept_queue, req);
-		inet_csk_reqsk_queue_add(sk, req, child);
-		/* Warning: caller must not call reqsk_put(req);
-		 * child stole last reference on it.
-		 */
-		return child;
+		if (inet_csk_reqsk_queue_add(sk, req, child))
+			return child;
 	}
 	/* Too bad, another child took ownership of the request, undo. */
 	bh_unlock_sock(child);
diff --git a/net/ipv4/ip_gre.c b/net/ipv4/ip_gre.c
index 7c51c4e1661f..41ba68de46d8 100644
--- a/net/ipv4/ip_gre.c
+++ b/net/ipv4/ip_gre.c
@@ -1054,8 +1054,9 @@ static const struct net_device_ops gre_tap_netdev_ops = {
 static void ipgre_tap_setup(struct net_device *dev)
 {
 	ether_setup(dev);
-	dev->netdev_ops		= &gre_tap_netdev_ops;
-	dev->priv_flags 	|= IFF_LIVE_ADDR_CHANGE;
+	dev->netdev_ops	= &gre_tap_netdev_ops;
+	dev->priv_flags &= ~IFF_TX_SKB_SHARING;
+	dev->priv_flags	|= IFF_LIVE_ADDR_CHANGE;
 	ip_tunnel_setup(dev, gre_tap_net_id);
 }
 
@@ -1240,6 +1241,14 @@ struct net_device *gretap_fb_dev_create(struct net *net, const char *name,
 	err = ipgre_newlink(net, dev, tb, NULL);
 	if (err < 0)
 		goto out;
+
+	/* openvswitch users expect packet sizes to be unrestricted,
+	 * so set the largest MTU we can.
+	 */
+	err = __ip_tunnel_change_mtu(dev, IP_MAX_MTU, false);
+	if (err)
+		goto out;
+
 	return dev;
 out:
 	free_netdev(dev);
diff --git a/net/ipv4/ip_output.c b/net/ipv4/ip_output.c
index 64878efa045c..565bf64b2b7d 100644
--- a/net/ipv4/ip_output.c
+++ b/net/ipv4/ip_output.c
@@ -1236,13 +1236,16 @@ ssize_t	ip_append_page(struct sock *sk, struct flowi4 *fl4, struct page *page,
 	if (!skb)
 		return -EINVAL;
 
-	cork->length += size;
 	if ((size + skb->len > mtu) &&
 	    (sk->sk_protocol == IPPROTO_UDP) &&
 	    (rt->dst.dev->features & NETIF_F_UFO)) {
+		if (skb->ip_summed != CHECKSUM_PARTIAL)
+			return -EOPNOTSUPP;
+
 		skb_shinfo(skb)->gso_size = mtu - fragheaderlen;
 		skb_shinfo(skb)->gso_type = SKB_GSO_UDP;
 	}
+	cork->length += size;
 
 	while (size > 0) {
 		if (skb_is_gso(skb)) {
diff --git a/net/ipv4/ip_sockglue.c b/net/ipv4/ip_sockglue.c
index 5f73a7c03e27..a50124260f5a 100644
--- a/net/ipv4/ip_sockglue.c
+++ b/net/ipv4/ip_sockglue.c
@@ -249,6 +249,8 @@ int ip_cmsg_send(struct net *net, struct msghdr *msg, struct ipcm_cookie *ipc,
 		switch (cmsg->cmsg_type) {
 		case IP_RETOPTS:
 			err = cmsg->cmsg_len - CMSG_ALIGN(sizeof(struct cmsghdr));
+
+			/* Our caller is responsible for freeing ipc->opt */
 			err = ip_options_get(net, &ipc->opt, CMSG_DATA(cmsg),
 					     err < 40 ? err : 40);
 			if (err)
diff --git a/net/ipv4/ip_tunnel.c b/net/ipv4/ip_tunnel.c
index c7bd72e9b544..336e6892a93c 100644
--- a/net/ipv4/ip_tunnel.c
+++ b/net/ipv4/ip_tunnel.c
@@ -661,6 +661,8 @@ void ip_tunnel_xmit(struct sk_buff *skb, struct net_device *dev,
 	inner_iph = (const struct iphdr *)skb_inner_network_header(skb);
 	connected = (tunnel->parms.iph.daddr != 0);
 
+	memset(&(IPCB(skb)->opt), 0, sizeof(IPCB(skb)->opt));
+
 	dst = tnl_params->daddr;
 	if (dst == 0) {
 		/* NBMA tunnel */
@@ -758,7 +760,6 @@ void ip_tunnel_xmit(struct sk_buff *skb, struct net_device *dev,
 				tunnel->err_time + IPTUNNEL_ERR_TIMEO)) {
 			tunnel->err_count--;
 
-			memset(IPCB(skb), 0, sizeof(*IPCB(skb)));
 			dst_link_failure(skb);
 		} else
 			tunnel->err_count = 0;
@@ -943,17 +944,31 @@ done:
 }
 EXPORT_SYMBOL_GPL(ip_tunnel_ioctl);
 
-int ip_tunnel_change_mtu(struct net_device *dev, int new_mtu)
+int __ip_tunnel_change_mtu(struct net_device *dev, int new_mtu, bool strict)
 {
 	struct ip_tunnel *tunnel = netdev_priv(dev);
 	int t_hlen = tunnel->hlen + sizeof(struct iphdr);
+	int max_mtu = 0xFFF8 - dev->hard_header_len - t_hlen;
 
-	if (new_mtu < 68 ||
-	    new_mtu > 0xFFF8 - dev->hard_header_len - t_hlen)
+	if (new_mtu < 68)
 		return -EINVAL;
+
+	if (new_mtu > max_mtu) {
+		if (strict)
+			return -EINVAL;
+
+		new_mtu = max_mtu;
+	}
+
 	dev->mtu = new_mtu;
 	return 0;
 }
+EXPORT_SYMBOL_GPL(__ip_tunnel_change_mtu);
+
+int ip_tunnel_change_mtu(struct net_device *dev, int new_mtu)
+{
+	return __ip_tunnel_change_mtu(dev, new_mtu, true);
+}
 EXPORT_SYMBOL_GPL(ip_tunnel_change_mtu);
 
 static void ip_tunnel_dev_free(struct net_device *dev)
diff --git a/net/ipv4/ping.c b/net/ipv4/ping.c
index c117b21b937d..d3a27165f9cc 100644
--- a/net/ipv4/ping.c
+++ b/net/ipv4/ping.c
@@ -746,8 +746,10 @@ static int ping_v4_sendmsg(struct sock *sk, struct msghdr *msg, size_t len)
 
 	if (msg->msg_controllen) {
 		err = ip_cmsg_send(sock_net(sk), msg, &ipc, false);
-		if (err)
+		if (unlikely(err)) {
+			kfree(ipc.opt);
 			return err;
+		}
 		if (ipc.opt)
 			free = 1;
 	}
diff --git a/net/ipv4/raw.c b/net/ipv4/raw.c
index bc35f1842512..7113bae4e6a0 100644
--- a/net/ipv4/raw.c
+++ b/net/ipv4/raw.c
@@ -547,8 +547,10 @@ static int raw_sendmsg(struct sock *sk, struct msghdr *msg, size_t len)
 
 	if (msg->msg_controllen) {
 		err = ip_cmsg_send(net, msg, &ipc, false);
-		if (err)
+		if (unlikely(err)) {
+			kfree(ipc.opt);
 			goto out;
+		}
 		if (ipc.opt)
 			free = 1;
 	}
diff --git a/net/ipv4/route.c b/net/ipv4/route.c
index 85f184e429c6..02c62299d717 100644
--- a/net/ipv4/route.c
+++ b/net/ipv4/route.c
@@ -129,6 +129,7 @@ static int ip_rt_mtu_expires __read_mostly	= 10 * 60 * HZ;
 static int ip_rt_min_pmtu __read_mostly		= 512 + 20 + 20;
 static int ip_rt_min_advmss __read_mostly	= 256;
 
+static int ip_rt_gc_timeout __read_mostly	= RT_GC_TIMEOUT;
 /*
  *	Interface to generic destination cache.
  */
@@ -755,7 +756,7 @@ static void __ip_do_redirect(struct rtable *rt, struct sk_buff *skb, struct flow
 				struct fib_nh *nh = &FIB_RES_NH(res);
 
 				update_or_create_fnhe(nh, fl4->daddr, new_gw,
-						      0, 0);
+						0, jiffies + ip_rt_gc_timeout);
 			}
 			if (kill_route)
 				rt->dst.obsolete = DST_OBSOLETE_KILL;
@@ -1556,6 +1557,36 @@ static void ip_handle_martian_source(struct net_device *dev,
 #endif
 }
 
+static void ip_del_fnhe(struct fib_nh *nh, __be32 daddr)
+{
+	struct fnhe_hash_bucket *hash;
+	struct fib_nh_exception *fnhe, __rcu **fnhe_p;
+	u32 hval = fnhe_hashfun(daddr);
+
+	spin_lock_bh(&fnhe_lock);
+
+	hash = rcu_dereference_protected(nh->nh_exceptions,
+					 lockdep_is_held(&fnhe_lock));
+	hash += hval;
+
+	fnhe_p = &hash->chain;
+	fnhe = rcu_dereference_protected(*fnhe_p, lockdep_is_held(&fnhe_lock));
+	while (fnhe) {
+		if (fnhe->fnhe_daddr == daddr) {
+			rcu_assign_pointer(*fnhe_p, rcu_dereference_protected(
+				fnhe->fnhe_next, lockdep_is_held(&fnhe_lock)));
+			fnhe_flush_routes(fnhe);
+			kfree_rcu(fnhe, rcu);
+			break;
+		}
+		fnhe_p = &fnhe->fnhe_next;
+		fnhe = rcu_dereference_protected(fnhe->fnhe_next,
+						 lockdep_is_held(&fnhe_lock));
+	}
+
+	spin_unlock_bh(&fnhe_lock);
+}
+
 /* called in rcu_read_lock() section */
 static int __mkroute_input(struct sk_buff *skb,
 			   const struct fib_result *res,
@@ -1609,11 +1640,20 @@ static int __mkroute_input(struct sk_buff *skb,
 
 	fnhe = find_exception(&FIB_RES_NH(*res), daddr);
 	if (do_cache) {
-		if (fnhe)
+		if (fnhe) {
 			rth = rcu_dereference(fnhe->fnhe_rth_input);
-		else
-			rth = rcu_dereference(FIB_RES_NH(*res).nh_rth_input);
+			if (rth && rth->dst.expires &&
+			    time_after(jiffies, rth->dst.expires)) {
+				ip_del_fnhe(&FIB_RES_NH(*res), daddr);
+				fnhe = NULL;
+			} else {
+				goto rt_cache;
+			}
+		}
+
+		rth = rcu_dereference(FIB_RES_NH(*res).nh_rth_input);
 
+rt_cache:
 		if (rt_cache_valid(rth)) {
 			skb_dst_set_noref(skb, &rth->dst);
 			goto out;
@@ -2014,19 +2054,29 @@ static struct rtable *__mkroute_output(const struct fib_result *res,
 		struct fib_nh *nh = &FIB_RES_NH(*res);
 
 		fnhe = find_exception(nh, fl4->daddr);
-		if (fnhe)
+		if (fnhe) {
 			prth = &fnhe->fnhe_rth_output;
-		else {
-			if (unlikely(fl4->flowi4_flags &
-				     FLOWI_FLAG_KNOWN_NH &&
-				     !(nh->nh_gw &&
-				       nh->nh_scope == RT_SCOPE_LINK))) {
-				do_cache = false;
-				goto add;
+			rth = rcu_dereference(*prth);
+			if (rth && rth->dst.expires &&
+			    time_after(jiffies, rth->dst.expires)) {
+				ip_del_fnhe(nh, fl4->daddr);
+				fnhe = NULL;
+			} else {
+				goto rt_cache;
 			}
-			prth = raw_cpu_ptr(nh->nh_pcpu_rth_output);
 		}
+
+		if (unlikely(fl4->flowi4_flags &
+			     FLOWI_FLAG_KNOWN_NH &&
+			     !(nh->nh_gw &&
+			       nh->nh_scope == RT_SCOPE_LINK))) {
+			do_cache = false;
+			goto add;
+		}
+		prth = raw_cpu_ptr(nh->nh_pcpu_rth_output);
 		rth = rcu_dereference(*prth);
+
+rt_cache:
 		if (rt_cache_valid(rth)) {
 			dst_hold(&rth->dst);
 			return rth;
@@ -2569,7 +2619,6 @@ void ip_rt_multicast_event(struct in_device *in_dev)
 }
 
 #ifdef CONFIG_SYSCTL
-static int ip_rt_gc_timeout __read_mostly	= RT_GC_TIMEOUT;
 static int ip_rt_gc_interval __read_mostly  = 60 * HZ;
 static int ip_rt_gc_min_interval __read_mostly	= HZ / 2;
 static int ip_rt_gc_elasticity __read_mostly	= 8;
diff --git a/net/ipv4/tcp.c b/net/ipv4/tcp.c
index 19746b3fcbbe..483ffdf5aa4d 100644
--- a/net/ipv4/tcp.c
+++ b/net/ipv4/tcp.c
@@ -940,7 +940,7 @@ new_segment:
 
 		i = skb_shinfo(skb)->nr_frags;
 		can_coalesce = skb_can_coalesce(skb, i, page, offset);
-		if (!can_coalesce && i >= MAX_SKB_FRAGS) {
+		if (!can_coalesce && i >= sysctl_max_skb_frags) {
 			tcp_mark_push(tp, skb);
 			goto new_segment;
 		}
@@ -1213,7 +1213,7 @@ new_segment:
 
 			if (!skb_can_coalesce(skb, i, pfrag->page,
 					      pfrag->offset)) {
-				if (i == MAX_SKB_FRAGS || !sg) {
+				if (i == sysctl_max_skb_frags || !sg) {
 					tcp_mark_push(tp, skb);
 					goto new_segment;
 				}
@@ -2950,7 +2950,7 @@ static void __tcp_alloc_md5sig_pool(void)
 			struct crypto_hash *hash;
 
 			hash = crypto_alloc_hash("md5", 0, CRYPTO_ALG_ASYNC);
-			if (IS_ERR_OR_NULL(hash))
+			if (IS_ERR(hash))
 				return;
 			per_cpu(tcp_md5sig_pool, cpu).md5_desc.tfm = hash;
 		}
diff --git a/net/ipv4/tcp_input.c b/net/ipv4/tcp_input.c
index 1c2a73406261..3b2c8e90a475 100644
--- a/net/ipv4/tcp_input.c
+++ b/net/ipv4/tcp_input.c
@@ -2896,7 +2896,10 @@ static void tcp_update_rtt_min(struct sock *sk, u32 rtt_us)
 {
 	const u32 now = tcp_time_stamp, wlen = sysctl_tcp_min_rtt_wlen * HZ;
 	struct rtt_meas *m = tcp_sk(sk)->rtt_min;
-	struct rtt_meas rttm = { .rtt = (rtt_us ? : 1), .ts = now };
+	struct rtt_meas rttm = {
+		.rtt = likely(rtt_us) ? rtt_us : jiffies_to_usecs(1),
+		.ts = now,
+	};
 	u32 elapsed;
 
 	/* Check if the new measurement updates the 1st, 2nd, or 3rd choices */
diff --git a/net/ipv4/tcp_ipv4.c b/net/ipv4/tcp_ipv4.c
index a4d523709ab3..487ac67059e2 100644
--- a/net/ipv4/tcp_ipv4.c
+++ b/net/ipv4/tcp_ipv4.c
@@ -311,7 +311,7 @@ static void do_redirect(struct sk_buff *skb, struct sock *sk)
 
 
 /* handle ICMP messages on TCP_NEW_SYN_RECV request sockets */
-void tcp_req_err(struct sock *sk, u32 seq)
+void tcp_req_err(struct sock *sk, u32 seq, bool abort)
 {
 	struct request_sock *req = inet_reqsk(sk);
 	struct net *net = sock_net(sk);
@@ -323,7 +323,7 @@ void tcp_req_err(struct sock *sk, u32 seq)
 
 	if (seq != tcp_rsk(req)->snt_isn) {
 		NET_INC_STATS_BH(net, LINUX_MIB_OUTOFWINDOWICMPS);
-	} else {
+	} else if (abort) {
 		/*
 		 * Still in SYN_RECV, just remove it silently.
 		 * There is no good way to pass the error to the newly
@@ -383,7 +383,12 @@ void tcp_v4_err(struct sk_buff *icmp_skb, u32 info)
 	}
 	seq = ntohl(th->seq);
 	if (sk->sk_state == TCP_NEW_SYN_RECV)
-		return tcp_req_err(sk, seq);
+		return tcp_req_err(sk, seq,
+				  type == ICMP_PARAMETERPROB ||
+				  type == ICMP_TIME_EXCEEDED ||
+				  (type == ICMP_DEST_UNREACH &&
+				   (code == ICMP_NET_UNREACH ||
+				    code == ICMP_HOST_UNREACH)));
 
 	bh_lock_sock(sk);
 	/* If too many ICMPs get dropped on busy
@@ -1592,28 +1597,30 @@ process:
 
 	if (sk->sk_state == TCP_NEW_SYN_RECV) {
 		struct request_sock *req = inet_reqsk(sk);
-		struct sock *nsk = NULL;
+		struct sock *nsk;
 
 		sk = req->rsk_listener;
-		if (tcp_v4_inbound_md5_hash(sk, skb))
-			goto discard_and_relse;
-		if (likely(sk->sk_state == TCP_LISTEN)) {
-			nsk = tcp_check_req(sk, skb, req, false);
-		} else {
+		if (unlikely(tcp_v4_inbound_md5_hash(sk, skb))) {
+			reqsk_put(req);
+			goto discard_it;
+		}
+		if (unlikely(sk->sk_state != TCP_LISTEN)) {
 			inet_csk_reqsk_queue_drop_and_put(sk, req);
 			goto lookup;
 		}
+		sock_hold(sk);
+		nsk = tcp_check_req(sk, skb, req, false);
 		if (!nsk) {
 			reqsk_put(req);
-			goto discard_it;
+			goto discard_and_relse;
 		}
 		if (nsk == sk) {
-			sock_hold(sk);
 			reqsk_put(req);
 		} else if (tcp_child_process(sk, nsk, skb)) {
 			tcp_v4_send_reset(nsk, skb);
-			goto discard_it;
+			goto discard_and_relse;
 		} else {
+			sock_put(sk);
 			return 0;
 		}
 	}
diff --git a/net/ipv4/tcp_metrics.c b/net/ipv4/tcp_metrics.c
index c8cbc2b4b792..a726d7853ce5 100644
--- a/net/ipv4/tcp_metrics.c
+++ b/net/ipv4/tcp_metrics.c
@@ -550,7 +550,7 @@ reset:
 	 */
 	if (crtt > tp->srtt_us) {
 		/* Set RTO like tcp_rtt_estimator(), but from cached RTT. */
-		crtt /= 8 * USEC_PER_MSEC;
+		crtt /= 8 * USEC_PER_SEC / HZ;
 		inet_csk(sk)->icsk_rto = crtt + max(2 * crtt, tcp_rto_min(sk));
 	} else if (tp->srtt_us == 0) {
 		/* RFC6298: 5.7 We've failed to get a valid RTT sample from
diff --git a/net/ipv4/tcp_minisocks.c b/net/ipv4/tcp_minisocks.c
index 75632a925824..9b02af2139d3 100644
--- a/net/ipv4/tcp_minisocks.c
+++ b/net/ipv4/tcp_minisocks.c
@@ -455,7 +455,7 @@ struct sock *tcp_create_openreq_child(const struct sock *sk,
 
 		newtp->rcv_wup = newtp->copied_seq =
 		newtp->rcv_nxt = treq->rcv_isn + 1;
-		newtp->segs_in = 0;
+		newtp->segs_in = 1;
 
 		newtp->snd_sml = newtp->snd_una =
 		newtp->snd_nxt = newtp->snd_up = treq->snt_isn + 1;
@@ -815,6 +815,7 @@ int tcp_child_process(struct sock *parent, struct sock *child,
 	int ret = 0;
 	int state = child->sk_state;
 
+	tcp_sk(child)->segs_in += max_t(u16, 1, skb_shinfo(skb)->gso_segs);
 	if (!sock_owned_by_user(child)) {
 		ret = tcp_rcv_state_process(child, skb);
 		/* Wakeup parent, send SIGIO */
diff --git a/net/ipv4/udp.c b/net/ipv4/udp.c
index be0b21852b13..95d2f198017e 100644
--- a/net/ipv4/udp.c
+++ b/net/ipv4/udp.c
@@ -1048,8 +1048,10 @@ int udp_sendmsg(struct sock *sk, struct msghdr *msg, size_t len)
 	if (msg->msg_controllen) {
 		err = ip_cmsg_send(sock_net(sk), msg, &ipc,
 				   sk->sk_family == AF_INET6);
-		if (err)
+		if (unlikely(err)) {
+			kfree(ipc.opt);
 			return err;
+		}
 		if (ipc.opt)
 			free = 1;
 		connected = 0;
diff --git a/net/ipv4/udp_tunnel.c b/net/ipv4/udp_tunnel.c
index 0ec08814f37d..96599d1a1318 100644
--- a/net/ipv4/udp_tunnel.c
+++ b/net/ipv4/udp_tunnel.c
@@ -89,6 +89,8 @@ void udp_tunnel_xmit_skb(struct rtable *rt, struct sock *sk, struct sk_buff *skb
 	uh->source = src_port;
 	uh->len = htons(skb->len);
 
+	memset(&(IPCB(skb)->opt), 0, sizeof(IPCB(skb)->opt));
+
 	udp_set_csum(nocheck, skb, src, dst, skb->len);
 
 	iptunnel_xmit(sk, rt, skb, src, dst, IPPROTO_UDP, tos, ttl, df, xnet);
diff --git a/net/ipv6/addrconf.c b/net/ipv6/addrconf.c
index 38eeddedfc21..bdd7eac4307a 100644
--- a/net/ipv6/addrconf.c
+++ b/net/ipv6/addrconf.c
@@ -583,7 +583,7 @@ static int inet6_netconf_get_devconf(struct sk_buff *in_skb,
 	if (err < 0)
 		goto errout;
 
-	err = EINVAL;
+	err = -EINVAL;
 	if (!tb[NETCONFA_IFINDEX])
 		goto errout;
 
@@ -3538,6 +3538,7 @@ static void addrconf_dad_begin(struct inet6_ifaddr *ifp)
 {
 	struct inet6_dev *idev = ifp->idev;
 	struct net_device *dev = idev->dev;
+	bool notify = false;
 
 	addrconf_join_solict(dev, &ifp->addr);
 
@@ -3583,7 +3584,7 @@ static void addrconf_dad_begin(struct inet6_ifaddr *ifp)
 			/* Because optimistic nodes can use this address,
 			 * notify listeners. If DAD fails, RTM_DELADDR is sent.
 			 */
-			ipv6_ifa_notify(RTM_NEWADDR, ifp);
+			notify = true;
 		}
 	}
 
@@ -3591,6 +3592,8 @@ static void addrconf_dad_begin(struct inet6_ifaddr *ifp)
 out:
 	spin_unlock(&ifp->lock);
 	read_unlock_bh(&idev->lock);
+	if (notify)
+		ipv6_ifa_notify(RTM_NEWADDR, ifp);
 }
 
 static void addrconf_dad_start(struct inet6_ifaddr *ifp)
diff --git a/net/ipv6/exthdrs_core.c b/net/ipv6/exthdrs_core.c
index 5c5d23e59da5..9508a20fbf61 100644
--- a/net/ipv6/exthdrs_core.c
+++ b/net/ipv6/exthdrs_core.c
@@ -257,7 +257,11 @@ int ipv6_find_hdr(const struct sk_buff *skb, unsigned int *offset,
 						*fragoff = _frag_off;
 					return hp->nexthdr;
 				}
-				return -ENOENT;
+				if (!found)
+					return -ENOENT;
+				if (fragoff)
+					*fragoff = _frag_off;
+				break;
 			}
 			hdrlen = 8;
 		} else if (nexthdr == NEXTHDR_AUTH) {
diff --git a/net/ipv6/ip6_flowlabel.c b/net/ipv6/ip6_flowlabel.c
index 1f9ebe3cbb4a..dc2db4f7b182 100644
--- a/net/ipv6/ip6_flowlabel.c
+++ b/net/ipv6/ip6_flowlabel.c
@@ -540,12 +540,13 @@ int ipv6_flowlabel_opt(struct sock *sk, char __user *optval, int optlen)
 		}
 		spin_lock_bh(&ip6_sk_fl_lock);
 		for (sflp = &np->ipv6_fl_list;
-		     (sfl = rcu_dereference(*sflp)) != NULL;
+		     (sfl = rcu_dereference_protected(*sflp,
+						      lockdep_is_held(&ip6_sk_fl_lock))) != NULL;
 		     sflp = &sfl->next) {
 			if (sfl->fl->label == freq.flr_label) {
 				if (freq.flr_label == (np->flow_label&IPV6_FLOWLABEL_MASK))
 					np->flow_label &= ~IPV6_FLOWLABEL_MASK;
-				*sflp = rcu_dereference(sfl->next);
+				*sflp = sfl->next;
 				spin_unlock_bh(&ip6_sk_fl_lock);
 				fl_release(sfl->fl);
 				kfree_rcu(sfl, rcu);
diff --git a/net/ipv6/ip6_gre.c b/net/ipv6/ip6_gre.c
index f37f18b6b40c..c0d4dc1c5ea4 100644
--- a/net/ipv6/ip6_gre.c
+++ b/net/ipv6/ip6_gre.c
@@ -777,6 +777,8 @@ static inline int ip6gre_xmit_ipv4(struct sk_buff *skb, struct net_device *dev)
 	__u32 mtu;
 	int err;
 
+	memset(&(IPCB(skb)->opt), 0, sizeof(IPCB(skb)->opt));
+
 	if (!(t->parms.flags & IP6_TNL_F_IGN_ENCAP_LIMIT))
 		encap_limit = t->parms.encap_limit;
 
@@ -1512,6 +1514,7 @@ static void ip6gre_tap_setup(struct net_device *dev)
 	dev->destructor = ip6gre_dev_free;
 
 	dev->features |= NETIF_F_NETNS_LOCAL;
+	dev->priv_flags &= ~IFF_TX_SKB_SHARING;
 }
 
 static int ip6gre_newlink(struct net *src_net, struct net_device *dev,
diff --git a/net/ipv6/ip6_tunnel.c b/net/ipv6/ip6_tunnel.c
index 137fca42aaa6..6c5dfec7a377 100644
--- a/net/ipv6/ip6_tunnel.c
+++ b/net/ipv6/ip6_tunnel.c
@@ -1180,6 +1180,8 @@ ip4ip6_tnl_xmit(struct sk_buff *skb, struct net_device *dev)
 	u8 tproto;
 	int err;
 
+	memset(&(IPCB(skb)->opt), 0, sizeof(IPCB(skb)->opt));
+
 	tproto = ACCESS_ONCE(t->parms.proto);
 	if (tproto != IPPROTO_IPIP && tproto != 0)
 		return -1;
diff --git a/net/ipv6/mcast.c b/net/ipv6/mcast.c
index 5ee56d0a8699..d64ee7e83664 100644
--- a/net/ipv6/mcast.c
+++ b/net/ipv6/mcast.c
@@ -1574,9 +1574,8 @@ static struct sk_buff *mld_newpack(struct inet6_dev *idev, unsigned int mtu)
 		return NULL;
 
 	skb->priority = TC_PRIO_CONTROL;
-	skb->reserved_tailroom = skb_end_offset(skb) -
-				 min(mtu, skb_end_offset(skb));
 	skb_reserve(skb, hlen);
+	skb_tailroom_reserve(skb, mtu, tlen);
 
 	if (__ipv6_get_lladdr(idev, &addr_buf, IFA_F_TENTATIVE)) {
 		/* <draft-ietf-magma-mld-source-05.txt>:
diff --git a/net/ipv6/netfilter/nf_nat_masquerade_ipv6.c b/net/ipv6/netfilter/nf_nat_masquerade_ipv6.c
index 31ba7ca19757..051b6a6bfff6 100644
--- a/net/ipv6/netfilter/nf_nat_masquerade_ipv6.c
+++ b/net/ipv6/netfilter/nf_nat_masquerade_ipv6.c
@@ -21,6 +21,10 @@
 #include <net/ipv6.h>
 #include <net/netfilter/ipv6/nf_nat_masquerade.h>
 
+#define MAX_WORK_COUNT	16
+
+static atomic_t v6_worker_count;
+
 unsigned int
 nf_nat_masquerade_ipv6(struct sk_buff *skb, const struct nf_nat_range *range,
 		       const struct net_device *out)
@@ -78,14 +82,78 @@ static struct notifier_block masq_dev_notifier = {
 	.notifier_call	= masq_device_event,
 };
 
+struct masq_dev_work {
+	struct work_struct work;
+	struct net *net;
+	int ifindex;
+};
+
+static void iterate_cleanup_work(struct work_struct *work)
+{
+	struct masq_dev_work *w;
+	long index;
+
+	w = container_of(work, struct masq_dev_work, work);
+
+	index = w->ifindex;
+	nf_ct_iterate_cleanup(w->net, device_cmp, (void *)index, 0, 0);
+
+	put_net(w->net);
+	kfree(w);
+	atomic_dec(&v6_worker_count);
+	module_put(THIS_MODULE);
+}
+
+/* ipv6 inet notifier is an atomic notifier, i.e. we cannot
+ * schedule.
+ *
+ * Unfortunately, nf_ct_iterate_cleanup can run for a long
+ * time if there are lots of conntracks and the system
+ * handles high softirq load, so it frequently calls cond_resched
+ * while iterating the conntrack table.
+ *
+ * So we defer nf_ct_iterate_cleanup walk to the system workqueue.
+ *
+ * As we can have 'a lot' of inet_events (depending on amount
+ * of ipv6 addresses being deleted), we also need to add an upper
+ * limit to the number of queued work items.
+ */
 static int masq_inet_event(struct notifier_block *this,
 			   unsigned long event, void *ptr)
 {
 	struct inet6_ifaddr *ifa = ptr;
-	struct netdev_notifier_info info;
+	const struct net_device *dev;
+	struct masq_dev_work *w;
+	struct net *net;
+
+	if (event != NETDEV_DOWN ||
+	    atomic_read(&v6_worker_count) >= MAX_WORK_COUNT)
+		return NOTIFY_DONE;
+
+	dev = ifa->idev->dev;
+	net = maybe_get_net(dev_net(dev));
+	if (!net)
+		return NOTIFY_DONE;
 
-	netdev_notifier_info_init(&info, ifa->idev->dev);
-	return masq_device_event(this, event, &info);
+	if (!try_module_get(THIS_MODULE))
+		goto err_module;
+
+	w = kmalloc(sizeof(*w), GFP_ATOMIC);
+	if (w) {
+		atomic_inc(&v6_worker_count);
+
+		INIT_WORK(&w->work, iterate_cleanup_work);
+		w->ifindex = dev->ifindex;
+		w->net = net;
+		schedule_work(&w->work);
+
+		return NOTIFY_DONE;
+	}
+
+	module_put(THIS_MODULE);
+ err_module:
+	put_net(net);
+	return NOTIFY_DONE;
 }
 
 static struct notifier_block masq_inet_notifier = {
diff --git a/net/ipv6/tcp_ipv6.c b/net/ipv6/tcp_ipv6.c
index 006396e31cb0..5c8c84273028 100644
--- a/net/ipv6/tcp_ipv6.c
+++ b/net/ipv6/tcp_ipv6.c
@@ -327,6 +327,7 @@ static void tcp_v6_err(struct sk_buff *skb, struct inet6_skb_parm *opt,
 	struct tcp_sock *tp;
 	__u32 seq, snd_una;
 	struct sock *sk;
+	bool fatal;
 	int err;
 
 	sk = __inet6_lookup_established(net, &tcp_hashinfo,
@@ -345,8 +346,9 @@ static void tcp_v6_err(struct sk_buff *skb, struct inet6_skb_parm *opt,
 		return;
 	}
 	seq = ntohl(th->seq);
+	fatal = icmpv6_err_convert(type, code, &err);
 	if (sk->sk_state == TCP_NEW_SYN_RECV)
-		return tcp_req_err(sk, seq);
+		return tcp_req_err(sk, seq, fatal);
 
 	bh_lock_sock(sk);
 	if (sock_owned_by_user(sk) && type != ICMPV6_PKT_TOOBIG)
@@ -400,7 +402,6 @@ static void tcp_v6_err(struct sk_buff *skb, struct inet6_skb_parm *opt,
 		goto out;
 	}
 
-	icmpv6_err_convert(type, code, &err);
 
 	/* Might be for an request_sock */
 	switch (sk->sk_state) {
@@ -1386,7 +1387,7 @@ process:
 
 	if (sk->sk_state == TCP_NEW_SYN_RECV) {
 		struct request_sock *req = inet_reqsk(sk);
-		struct sock *nsk = NULL;
+		struct sock *nsk;
 
 		sk = req->rsk_listener;
 		tcp_v6_fill_cb(skb, hdr, th);
@@ -1394,24 +1395,24 @@ process:
 			reqsk_put(req);
 			goto discard_it;
 		}
-		if (likely(sk->sk_state == TCP_LISTEN)) {
-			nsk = tcp_check_req(sk, skb, req, false);
-		} else {
+		if (unlikely(sk->sk_state != TCP_LISTEN)) {
 			inet_csk_reqsk_queue_drop_and_put(sk, req);
 			goto lookup;
 		}
+		sock_hold(sk);
+		nsk = tcp_check_req(sk, skb, req, false);
 		if (!nsk) {
 			reqsk_put(req);
-			goto discard_it;
+			goto discard_and_relse;
 		}
 		if (nsk == sk) {
-			sock_hold(sk);
 			reqsk_put(req);
 			tcp_v6_restore_cb(skb);
 		} else if (tcp_child_process(sk, nsk, skb)) {
 			tcp_v6_send_reset(nsk, skb);
-			goto discard_it;
+			goto discard_and_relse;
 		} else {
+			sock_put(sk);
 			return 0;
 		}
 	}
diff --git a/net/ipv6/udp.c b/net/ipv6/udp.c
index 22e28a44e3c8..422dd014aa2c 100644
--- a/net/ipv6/udp.c
+++ b/net/ipv6/udp.c
@@ -962,11 +962,9 @@ int __udp6_lib_rcv(struct sk_buff *skb, struct udp_table *udptable,
 		ret = udpv6_queue_rcv_skb(sk, skb);
 		sock_put(sk);
 
-		/* a return value > 0 means to resubmit the input, but
-		 * it wants the return to be -protocol, or 0
-		 */
+		/* a return value > 0 means to resubmit the input */
 		if (ret > 0)
-			return -ret;
+			return ret;
 
 		return 0;
 	}
diff --git a/net/l2tp/l2tp_netlink.c b/net/l2tp/l2tp_netlink.c
index f93c5be612a7..2caaa84ce92d 100644
--- a/net/l2tp/l2tp_netlink.c
+++ b/net/l2tp/l2tp_netlink.c
@@ -124,8 +124,13 @@ static int l2tp_tunnel_notify(struct genl_family *family,
 	ret = l2tp_nl_tunnel_send(msg, info->snd_portid, info->snd_seq,
 				  NLM_F_ACK, tunnel, cmd);
 
-	if (ret >= 0)
-		return genlmsg_multicast_allns(family, msg, 0,	0, GFP_ATOMIC);
+	if (ret >= 0) {
+		ret = genlmsg_multicast_allns(family, msg, 0, 0, GFP_ATOMIC);
+		/* We don't care if no one is listening */
+		if (ret == -ESRCH)
+			ret = 0;
+		return ret;
+	}
 
 	nlmsg_free(msg);
 
@@ -147,8 +152,13 @@ static int l2tp_session_notify(struct genl_family *family,
 	ret = l2tp_nl_session_send(msg, info->snd_portid, info->snd_seq,
 				   NLM_F_ACK, session, cmd);
 
-	if (ret >= 0)
-		return genlmsg_multicast_allns(family, msg, 0,	0, GFP_ATOMIC);
+	if (ret >= 0) {
+		ret = genlmsg_multicast_allns(family, msg, 0, 0, GFP_ATOMIC);
+		/* We don't care if no one is listening */
+		if (ret == -ESRCH)
+			ret = 0;
+		return ret;
+	}
 
 	nlmsg_free(msg);
 
diff --git a/net/mac80211/agg-rx.c b/net/mac80211/agg-rx.c
index 10ad4ac1fa0b..367784be5df2 100644
--- a/net/mac80211/agg-rx.c
+++ b/net/mac80211/agg-rx.c
@@ -291,7 +291,7 @@ void __ieee80211_start_rx_ba_session(struct sta_info *sta,
 	}
 
 	/* prepare A-MPDU MLME for Rx aggregation */
-	tid_agg_rx = kmalloc(sizeof(struct tid_ampdu_rx), GFP_KERNEL);
+	tid_agg_rx = kzalloc(sizeof(*tid_agg_rx), GFP_KERNEL);
 	if (!tid_agg_rx)
 		goto end;
 
diff --git a/net/mac80211/ieee80211_i.h b/net/mac80211/ieee80211_i.h
index b84f6aa32c08..f006f4a44c0e 100644
--- a/net/mac80211/ieee80211_i.h
+++ b/net/mac80211/ieee80211_i.h
@@ -92,7 +92,7 @@ struct ieee80211_fragment_entry {
 	u16 extra_len;
 	u16 last_frag;
 	u8 rx_queue;
-	bool ccmp; /* Whether fragments were encrypted with CCMP */
+	bool check_sequential_pn; /* needed for CCMP/GCMP */
 	u8 last_pn[6]; /* PN of the last fragment if CCMP was used */
 };
 
diff --git a/net/mac80211/rc80211_minstrel.c b/net/mac80211/rc80211_minstrel.c
index 3ece7d1034c8..b54f398cda5d 100644
--- a/net/mac80211/rc80211_minstrel.c
+++ b/net/mac80211/rc80211_minstrel.c
@@ -711,7 +711,7 @@ static u32 minstrel_get_expected_throughput(void *priv_sta)
 	 * computing cur_tp
 	 */
 	tmp_mrs = &mi->r[idx].stats;
-	tmp_cur_tp = minstrel_get_tp_avg(&mi->r[idx], tmp_mrs->prob_ewma);
+	tmp_cur_tp = minstrel_get_tp_avg(&mi->r[idx], tmp_mrs->prob_ewma) * 10;
 	tmp_cur_tp = tmp_cur_tp * 1200 * 8 / 1024;
 
 	return tmp_cur_tp;
diff --git a/net/mac80211/rc80211_minstrel_ht.c b/net/mac80211/rc80211_minstrel_ht.c
index 3928dbd24e25..370d677b547b 100644
--- a/net/mac80211/rc80211_minstrel_ht.c
+++ b/net/mac80211/rc80211_minstrel_ht.c
@@ -414,15 +414,16 @@ minstrel_ht_set_best_prob_rate(struct minstrel_ht_sta *mi, u16 index)
 	    (max_tp_group != MINSTREL_CCK_GROUP))
 		return;
 
+	max_gpr_group = mg->max_group_prob_rate / MCS_GROUP_RATES;
+	max_gpr_idx = mg->max_group_prob_rate % MCS_GROUP_RATES;
+	max_gpr_prob = mi->groups[max_gpr_group].rates[max_gpr_idx].prob_ewma;
+
 	if (mrs->prob_ewma > MINSTREL_FRAC(75, 100)) {
 		cur_tp_avg = minstrel_ht_get_tp_avg(mi, cur_group, cur_idx,
 						    mrs->prob_ewma);
 		if (cur_tp_avg > tmp_tp_avg)
 			mi->max_prob_rate = index;
 
-		max_gpr_group = mg->max_group_prob_rate / MCS_GROUP_RATES;
-		max_gpr_idx = mg->max_group_prob_rate %	MCS_GROUP_RATES;
-		max_gpr_prob = mi->groups[max_gpr_group].rates[max_gpr_idx].prob_ewma;
 		max_gpr_tp_avg = minstrel_ht_get_tp_avg(mi, max_gpr_group,
 							max_gpr_idx,
 							max_gpr_prob);
@@ -431,7 +432,7 @@ minstrel_ht_set_best_prob_rate(struct minstrel_ht_sta *mi, u16 index)
 	} else {
 		if (mrs->prob_ewma > tmp_prob)
 			mi->max_prob_rate = index;
-		if (mrs->prob_ewma > mg->rates[mg->max_group_prob_rate].prob_ewma)
+		if (mrs->prob_ewma > max_gpr_prob)
 			mg->max_group_prob_rate = index;
 	}
 }
@@ -691,7 +692,7 @@ minstrel_aggr_check(struct ieee80211_sta *pubsta, struct sk_buff *skb)
 	if (likely(sta->ampdu_mlme.tid_tx[tid]))
 		return;
 
-	ieee80211_start_tx_ba_session(pubsta, tid, 5000);
+	ieee80211_start_tx_ba_session(pubsta, tid, 0);
 }
 
 static void
@@ -871,7 +872,7 @@ minstrel_ht_set_rate(struct minstrel_priv *mp, struct minstrel_ht_sta *mi,
 	 *  - if station is in dynamic SMPS (and streams > 1)
 	 *  - for fallback rates, to increase chances of getting through
 	 */
-	if (offset > 0 &&
+	if (offset > 0 ||
 	    (mi->sta->smps_mode == IEEE80211_SMPS_DYNAMIC &&
 	     group->streams > 1)) {
 		ratetbl->rate[offset].count = ratetbl->rate[offset].count_rts;
@@ -1334,7 +1335,8 @@ static u32 minstrel_ht_get_expected_throughput(void *priv_sta)
 	prob = mi->groups[i].rates[j].prob_ewma;
 
 	/* convert tp_avg from pkt per second in kbps */
-	tp_avg = minstrel_ht_get_tp_avg(mi, i, j, prob) * AVG_PKT_SIZE * 8 / 1024;
+	tp_avg = minstrel_ht_get_tp_avg(mi, i, j, prob) * 10;
+	tp_avg = tp_avg * AVG_PKT_SIZE * 8 / 1024;
 
 	return tp_avg;
 }
diff --git a/net/mac80211/rx.c b/net/mac80211/rx.c
index bc081850ac0e..60d093f40f1d 100644
--- a/net/mac80211/rx.c
+++ b/net/mac80211/rx.c
@@ -1753,7 +1753,7 @@ ieee80211_reassemble_add(struct ieee80211_sub_if_data *sdata,
 	entry->seq = seq;
 	entry->rx_queue = rx_queue;
 	entry->last_frag = frag;
-	entry->ccmp = 0;
+	entry->check_sequential_pn = false;
 	entry->extra_len = 0;
 
 	return entry;
@@ -1849,15 +1849,27 @@ ieee80211_rx_h_defragment(struct ieee80211_rx_data *rx)
 						 rx->seqno_idx, &(rx->skb));
 		if (rx->key &&
 		    (rx->key->conf.cipher == WLAN_CIPHER_SUITE_CCMP ||
-		     rx->key->conf.cipher == WLAN_CIPHER_SUITE_CCMP_256) &&
+		     rx->key->conf.cipher == WLAN_CIPHER_SUITE_CCMP_256 ||
+		     rx->key->conf.cipher == WLAN_CIPHER_SUITE_GCMP ||
+		     rx->key->conf.cipher == WLAN_CIPHER_SUITE_GCMP_256) &&
 		    ieee80211_has_protected(fc)) {
 			int queue = rx->security_idx;
-			/* Store CCMP PN so that we can verify that the next
-			 * fragment has a sequential PN value. */
-			entry->ccmp = 1;
+
+			/* Store CCMP/GCMP PN so that we can verify that the
+			 * next fragment has a sequential PN value.
+			 */
+			entry->check_sequential_pn = true;
 			memcpy(entry->last_pn,
 			       rx->key->u.ccmp.rx_pn[queue],
 			       IEEE80211_CCMP_PN_LEN);
+			BUILD_BUG_ON(offsetof(struct ieee80211_key,
+					      u.ccmp.rx_pn) !=
+				     offsetof(struct ieee80211_key,
+					      u.gcmp.rx_pn));
+			BUILD_BUG_ON(sizeof(rx->key->u.ccmp.rx_pn[queue]) !=
+				     sizeof(rx->key->u.gcmp.rx_pn[queue]));
+			BUILD_BUG_ON(IEEE80211_CCMP_PN_LEN !=
+				     IEEE80211_GCMP_PN_LEN);
 		}
 		return RX_QUEUED;
 	}
@@ -1872,15 +1884,21 @@ ieee80211_rx_h_defragment(struct ieee80211_rx_data *rx)
 		return RX_DROP_MONITOR;
 	}
 
-	/* Verify that MPDUs within one MSDU have sequential PN values.
-	 * (IEEE 802.11i, 8.3.3.4.5) */
-	if (entry->ccmp) {
+	/* "The receiver shall discard MSDUs and MMPDUs whose constituent
+	 *  MPDU PN values are not incrementing in steps of 1."
+	 * see IEEE P802.11-REVmc/D5.0, 12.5.3.4.4, item d (for CCMP)
+	 * and IEEE P802.11-REVmc/D5.0, 12.5.5.4.4, item d (for GCMP)
+	 */
+	if (entry->check_sequential_pn) {
 		int i;
 		u8 pn[IEEE80211_CCMP_PN_LEN], *rpn;
 		int queue;
+
 		if (!rx->key ||
 		    (rx->key->conf.cipher != WLAN_CIPHER_SUITE_CCMP &&
-		     rx->key->conf.cipher != WLAN_CIPHER_SUITE_CCMP_256))
+		     rx->key->conf.cipher != WLAN_CIPHER_SUITE_CCMP_256 &&
+		     rx->key->conf.cipher != WLAN_CIPHER_SUITE_GCMP &&
+		     rx->key->conf.cipher != WLAN_CIPHER_SUITE_GCMP_256))
 			return RX_DROP_UNUSABLE;
 		memcpy(pn, entry->last_pn, IEEE80211_CCMP_PN_LEN);
 		for (i = IEEE80211_CCMP_PN_LEN - 1; i >= 0; i--) {
@@ -3366,6 +3384,7 @@ static bool ieee80211_accept_frame(struct ieee80211_rx_data *rx)
 				return false;
 			/* ignore action frames to TDLS-peers */
 			if (ieee80211_is_action(hdr->frame_control) &&
+			    !is_broadcast_ether_addr(bssid) &&
 			    !ether_addr_equal(bssid, hdr->addr1))
 				return false;
 		}
diff --git a/net/netfilter/Kconfig b/net/netfilter/Kconfig
index 8c067e6663a1..95e757c377f9 100644
--- a/net/netfilter/Kconfig
+++ b/net/netfilter/Kconfig
@@ -891,7 +891,7 @@ config NETFILTER_XT_TARGET_TEE
 	depends on IPV6 || IPV6=n
 	depends on !NF_CONNTRACK || NF_CONNTRACK
 	select NF_DUP_IPV4
-	select NF_DUP_IPV6 if IP6_NF_IPTABLES != n
+	select NF_DUP_IPV6 if IPV6
 	---help---
 	This option adds a "TEE" target with which a packet can be cloned and
 	this clone be rerouted to another nexthop.
diff --git a/net/netfilter/nf_conntrack_core.c b/net/netfilter/nf_conntrack_core.c
index 58882de06bd7..f60b4fdeeb8c 100644
--- a/net/netfilter/nf_conntrack_core.c
+++ b/net/netfilter/nf_conntrack_core.c
@@ -1412,6 +1412,7 @@ get_next_corpse(struct net *net, int (*iter)(struct nf_conn *i, void *data),
 		}
 		spin_unlock(lockp);
 		local_bh_enable();
+		cond_resched();
 	}
 
 	for_each_possible_cpu(cpu) {
@@ -1424,6 +1425,7 @@ get_next_corpse(struct net *net, int (*iter)(struct nf_conn *i, void *data),
 				set_bit(IPS_DYING_BIT, &ct->status);
 		}
 		spin_unlock_bh(&pcpu->lock);
+		cond_resched();
 	}
 	return NULL;
 found:
@@ -1440,6 +1442,8 @@ void nf_ct_iterate_cleanup(struct net *net,
 	struct nf_conn *ct;
 	unsigned int bucket = 0;
 
+	might_sleep();
+
 	while ((ct = get_next_corpse(net, iter, data, &bucket)) != NULL) {
 		/* Time to push up daises... */
 		if (del_timer(&ct->timeout))
@@ -1448,6 +1452,7 @@ void nf_ct_iterate_cleanup(struct net *net,
 		/* ... else the timer will get him soon. */
 
 		nf_ct_put(ct);
+		cond_resched();
 	}
 }
 EXPORT_SYMBOL_GPL(nf_ct_iterate_cleanup);
diff --git a/net/netfilter/nfnetlink.c b/net/netfilter/nfnetlink.c
index a7ba23353dab..857ae89633af 100644
--- a/net/netfilter/nfnetlink.c
+++ b/net/netfilter/nfnetlink.c
@@ -311,14 +311,14 @@ replay:
 #endif
 		{
 			nfnl_unlock(subsys_id);
-			netlink_ack(skb, nlh, -EOPNOTSUPP);
+			netlink_ack(oskb, nlh, -EOPNOTSUPP);
 			return kfree_skb(skb);
 		}
 	}
 
 	if (!ss->commit || !ss->abort) {
 		nfnl_unlock(subsys_id);
-		netlink_ack(skb, nlh, -EOPNOTSUPP);
+		netlink_ack(oskb, nlh, -EOPNOTSUPP);
 		return kfree_skb(skb);
 	}
 
@@ -328,10 +328,12 @@ replay:
 		nlh = nlmsg_hdr(skb);
 		err = 0;
 
-		if (nlmsg_len(nlh) < sizeof(struct nfgenmsg) ||
-		    skb->len < nlh->nlmsg_len) {
-			err = -EINVAL;
-			goto ack;
+		if (nlh->nlmsg_len < NLMSG_HDRLEN ||
+		    skb->len < nlh->nlmsg_len ||
+		    nlmsg_len(nlh) < sizeof(struct nfgenmsg)) {
+			nfnl_err_reset(&err_list);
+			status |= NFNL_BATCH_FAILURE;
+			goto done;
 		}
 
 		/* Only requests are handled by the kernel */
@@ -406,7 +408,7 @@ ack:
 				 * pointing to the batch header.
 				 */
 				nfnl_err_reset(&err_list);
-				netlink_ack(skb, nlmsg_hdr(oskb), -ENOMEM);
+				netlink_ack(oskb, nlmsg_hdr(oskb), -ENOMEM);
 				status |= NFNL_BATCH_FAILURE;
 				goto done;
 			}
diff --git a/net/netfilter/nfnetlink_cttimeout.c b/net/netfilter/nfnetlink_cttimeout.c
index 94837d236ab0..2671b9deb103 100644
--- a/net/netfilter/nfnetlink_cttimeout.c
+++ b/net/netfilter/nfnetlink_cttimeout.c
@@ -312,7 +312,7 @@ static void ctnl_untimeout(struct net *net, struct ctnl_timeout *timeout)
 			hlist_nulls_for_each_entry(h, nn, &net->ct.hash[i], hnnode)
 				untimeout(h, timeout);
 		}
-		nf_conntrack_lock(&nf_conntrack_locks[i % CONNTRACK_LOCKS]);
+		spin_unlock(&nf_conntrack_locks[i % CONNTRACK_LOCKS]);
 	}
 	local_bh_enable();
 }
diff --git a/net/netfilter/nft_counter.c b/net/netfilter/nft_counter.c
index c7808fc19719..c9743f78f219 100644
--- a/net/netfilter/nft_counter.c
+++ b/net/netfilter/nft_counter.c
@@ -100,7 +100,7 @@ static int nft_counter_init(const struct nft_ctx *ctx,
 
 	cpu_stats = netdev_alloc_pcpu_stats(struct nft_counter_percpu);
 	if (cpu_stats == NULL)
-		return ENOMEM;
+		return -ENOMEM;
 
 	preempt_disable();
 	this_cpu = this_cpu_ptr(cpu_stats);
@@ -138,7 +138,7 @@ static int nft_counter_clone(struct nft_expr *dst, const struct nft_expr *src)
 	cpu_stats = __netdev_alloc_pcpu_stats(struct nft_counter_percpu,
 					      GFP_ATOMIC);
 	if (cpu_stats == NULL)
-		return ENOMEM;
+		return -ENOMEM;
 
 	preempt_disable();
 	this_cpu = this_cpu_ptr(cpu_stats);
diff --git a/net/netfilter/xt_TEE.c b/net/netfilter/xt_TEE.c
index 3eff7b67cdf2..6e57a3966dc5 100644
--- a/net/netfilter/xt_TEE.c
+++ b/net/netfilter/xt_TEE.c
@@ -38,7 +38,7 @@ tee_tg4(struct sk_buff *skb, const struct xt_action_param *par)
 	return XT_CONTINUE;
 }
 
-#if IS_ENABLED(CONFIG_NF_DUP_IPV6)
+#if IS_ENABLED(CONFIG_IPV6)
 static unsigned int
 tee_tg6(struct sk_buff *skb, const struct xt_action_param *par)
 {
@@ -131,7 +131,7 @@ static struct xt_target tee_tg_reg[] __read_mostly = {
 		.destroy    = tee_tg_destroy,
 		.me         = THIS_MODULE,
 	},
-#if IS_ENABLED(CONFIG_NF_DUP_IPV6)
+#if IS_ENABLED(CONFIG_IPV6)
 	{
 		.name       = "TEE",
 		.revision   = 1,
diff --git a/net/openvswitch/vport-vxlan.c b/net/openvswitch/vport-vxlan.c
index 1605691d9414..5eb7694348b5 100644
--- a/net/openvswitch/vport-vxlan.c
+++ b/net/openvswitch/vport-vxlan.c
@@ -90,7 +90,9 @@ static struct vport *vxlan_tnl_create(const struct vport_parms *parms)
 	int err;
 	struct vxlan_config conf = {
 		.no_share = true,
-		.flags = VXLAN_F_COLLECT_METADATA,
+		.flags = VXLAN_F_COLLECT_METADATA | VXLAN_F_UDP_ZERO_CSUM6_RX,
+		/* Don't restrict the packets that can be sent by MTU */
+		.mtu = IP_MAX_MTU,
 	};
 
 	if (!options) {
diff --git a/net/sched/act_ipt.c b/net/sched/act_ipt.c
index d05869646515..6b70399ab781 100644
--- a/net/sched/act_ipt.c
+++ b/net/sched/act_ipt.c
@@ -62,6 +62,7 @@ static void ipt_destroy_target(struct xt_entry_target *t)
 	struct xt_tgdtor_param par = {
 		.target   = t->u.kernel.target,
 		.targinfo = t->data,
+		.family   = NFPROTO_IPV4,
 	};
 	if (par.target->destroy != NULL)
 		par.target->destroy(&par);
@@ -195,6 +196,7 @@ static int tcf_ipt(struct sk_buff *skb, const struct tc_action *a,
 	par.hooknum  = ipt->tcfi_hook;
 	par.target   = ipt->tcfi_t->u.kernel.target;
 	par.targinfo = ipt->tcfi_t->data;
+	par.family   = NFPROTO_IPV4;
 	ret = par.target->target(skb, &par);
 
 	switch (ret) {
diff --git a/net/sched/sch_api.c b/net/sched/sch_api.c
index b5c2cf2aa6d4..af1acf009866 100644
--- a/net/sched/sch_api.c
+++ b/net/sched/sch_api.c
@@ -1852,6 +1852,7 @@ reset:
 	}
 
 	tp = old_tp;
+	protocol = tc_skb_protocol(skb);
 	goto reclassify;
 #endif
 }
diff --git a/net/sctp/ipv6.c b/net/sctp/ipv6.c
index ec529121f38a..ce46f1c7f133 100644
--- a/net/sctp/ipv6.c
+++ b/net/sctp/ipv6.c
@@ -526,6 +526,8 @@ static int sctp_v6_cmp_addr(const union sctp_addr *addr1,
 		}
 		return 0;
 	}
+	if (addr1->v6.sin6_port != addr2->v6.sin6_port)
+		return 0;
 	if (!ipv6_addr_equal(&addr1->v6.sin6_addr, &addr2->v6.sin6_addr))
 		return 0;
 	/* If this is a linklocal address, compare the scope_id. */
diff --git a/net/sctp/proc.c b/net/sctp/proc.c
index ded7d931a6a5..963dffcc2618 100644
--- a/net/sctp/proc.c
+++ b/net/sctp/proc.c
@@ -482,7 +482,7 @@ static void sctp_remaddr_seq_stop(struct seq_file *seq, void *v)
 static int sctp_remaddr_seq_show(struct seq_file *seq, void *v)
 {
 	struct sctp_association *assoc;
-	struct sctp_transport *tsp;
+	struct sctp_transport *transport, *tsp;
 
 	if (v == SEQ_START_TOKEN) {
 		seq_printf(seq, "ADDR ASSOC_ID HB_ACT RTO MAX_PATH_RTX "
@@ -490,10 +490,10 @@ static int sctp_remaddr_seq_show(struct seq_file *seq, void *v)
 		return 0;
 	}
 
-	tsp = (struct sctp_transport *)v;
-	if (!sctp_transport_hold(tsp))
+	transport = (struct sctp_transport *)v;
+	if (!sctp_transport_hold(transport))
 		return 0;
-	assoc = tsp->asoc;
+	assoc = transport->asoc;
 
 	list_for_each_entry_rcu(tsp, &assoc->peer.transport_addr_list,
 				transports) {
@@ -546,7 +546,7 @@ static int sctp_remaddr_seq_show(struct seq_file *seq, void *v)
 		seq_printf(seq, "\n");
 	}
 
-	sctp_transport_put(tsp);
+	sctp_transport_put(transport);
 
 	return 0;
 }
diff --git a/net/sctp/protocol.c b/net/sctp/protocol.c
index ab0d538a74ed..1099e99a53c4 100644
--- a/net/sctp/protocol.c
+++ b/net/sctp/protocol.c
@@ -60,6 +60,8 @@
 #include <net/inet_common.h>
 #include <net/inet_ecn.h>
 
+#define MAX_SCTP_PORT_HASH_ENTRIES (64 * 1024)
+
 /* Global data structures. */
 struct sctp_globals sctp_globals __read_mostly;
 
@@ -1355,6 +1357,8 @@ static __init int sctp_init(void)
 	unsigned long limit;
 	int max_share;
 	int order;
+	int num_entries;
+	int max_entry_order;
 
 	sock_skb_cb_check_size(sizeof(struct sctp_ulpevent));
 
@@ -1407,14 +1411,24 @@ static __init int sctp_init(void)
 
 	/* Size and allocate the association hash table.
 	 * The methodology is similar to that of the tcp hash tables.
+	 * Though not identical.  Start by getting a goal size
 	 */
 	if (totalram_pages >= (128 * 1024))
 		goal = totalram_pages >> (22 - PAGE_SHIFT);
 	else
 		goal = totalram_pages >> (24 - PAGE_SHIFT);
 
-	for (order = 0; (1UL << order) < goal; order++)
-		;
+	/* Then compute the page order for said goal */
+	order = get_order(goal);
+
+	/* Now compute the required page order for the maximum sized table we
+	 * want to create
+	 */
+	max_entry_order = get_order(MAX_SCTP_PORT_HASH_ENTRIES *
+				    sizeof(struct sctp_bind_hashbucket));
+
+	/* Limit the page order by that maximum hash table size */
+	order = min(order, max_entry_order);
 
 	/* Allocate and initialize the endpoint hash table.  */
 	sctp_ep_hashsize = 64;
@@ -1430,20 +1444,35 @@ static __init int sctp_init(void)
 		INIT_HLIST_HEAD(&sctp_ep_hashtable[i].chain);
 	}
 
-	/* Allocate and initialize the SCTP port hash table.  */
+	/* Allocate and initialize the SCTP port hash table.
+	 * Note that order is initalized to start at the max sized
+	 * table we want to support.  If we can't get that many pages
+	 * reduce the order and try again
+	 */
 	do {
-		sctp_port_hashsize = (1UL << order) * PAGE_SIZE /
-					sizeof(struct sctp_bind_hashbucket);
-		if ((sctp_port_hashsize > (64 * 1024)) && order > 0)
-			continue;
 		sctp_port_hashtable = (struct sctp_bind_hashbucket *)
 			__get_free_pages(GFP_KERNEL | __GFP_NOWARN, order);
 	} while (!sctp_port_hashtable && --order > 0);
+
 	if (!sctp_port_hashtable) {
 		pr_err("Failed bind hash alloc\n");
 		status = -ENOMEM;
 		goto err_bhash_alloc;
 	}
+
+	/* Now compute the number of entries that will fit in the
+	 * port hash space we allocated
+	 */
+	num_entries = (1UL << order) * PAGE_SIZE /
+		      sizeof(struct sctp_bind_hashbucket);
+
+	/* And finish by rounding it down to the nearest power of two
+	 * this wastes some memory of course, but its needed because
+	 * the hash function operates based on the assumption that
+	 * that the number of entries is a power of two
+	 */
+	sctp_port_hashsize = rounddown_pow_of_two(num_entries);
+
 	for (i = 0; i < sctp_port_hashsize; i++) {
 		spin_lock_init(&sctp_port_hashtable[i].lock);
 		INIT_HLIST_HEAD(&sctp_port_hashtable[i].chain);
@@ -1452,7 +1481,8 @@ static __init int sctp_init(void)
 	if (sctp_transport_hashtable_init())
 		goto err_thash_alloc;
 
-	pr_info("Hash tables configured (bind %d)\n", sctp_port_hashsize);
+	pr_info("Hash tables configured (bind %d/%d)\n", sctp_port_hashsize,
+		num_entries);
 
 	sctp_sysctl_register();
 
diff --git a/net/sctp/socket.c b/net/sctp/socket.c
index 5ca2ebfe0be8..e878da0949db 100644
--- a/net/sctp/socket.c
+++ b/net/sctp/socket.c
@@ -5538,6 +5538,7 @@ static int sctp_getsockopt_hmac_ident(struct sock *sk, int len,
 	struct sctp_hmac_algo_param *hmacs;
 	__u16 data_len = 0;
 	u32 num_idents;
+	int i;
 
 	if (!ep->auth_enable)
 		return -EACCES;
@@ -5555,8 +5556,12 @@ static int sctp_getsockopt_hmac_ident(struct sock *sk, int len,
 		return -EFAULT;
 	if (put_user(num_idents, &p->shmac_num_idents))
 		return -EFAULT;
-	if (copy_to_user(p->shmac_idents, hmacs->hmac_ids, data_len))
-		return -EFAULT;
+	for (i = 0; i < num_idents; i++) {
+		__u16 hmacid = ntohs(hmacs->hmac_ids[i]);
+
+		if (copy_to_user(&p->shmac_idents[i], &hmacid, sizeof(__u16)))
+			return -EFAULT;
+	}
 	return 0;
 }
 
diff --git a/net/sunrpc/auth_gss/auth_gss.c b/net/sunrpc/auth_gss/auth_gss.c
index 799e65b944b9..cabf586f47d7 100644
--- a/net/sunrpc/auth_gss/auth_gss.c
+++ b/net/sunrpc/auth_gss/auth_gss.c
@@ -740,7 +740,7 @@ gss_pipe_downcall(struct file *filp, const char __user *src, size_t mlen)
 		default:
 			printk(KERN_CRIT "%s: bad return from "
 				"gss_fill_context: %zd\n", __func__, err);
-			BUG();
+			gss_msg->msg.errno = -EIO;
 		}
 		goto err_release_msg;
 	}
diff --git a/net/sunrpc/cache.c b/net/sunrpc/cache.c
index 2b32fd602669..273bc3a35425 100644
--- a/net/sunrpc/cache.c
+++ b/net/sunrpc/cache.c
@@ -1225,7 +1225,7 @@ int qword_get(char **bpp, char *dest, int bufsize)
 	if (bp[0] == '\\' && bp[1] == 'x') {
 		/* HEX STRING */
 		bp += 2;
-		while (len < bufsize) {
+		while (len < bufsize - 1) {
 			int h, l;
 
 			h = hex_to_bin(bp[0]);
diff --git a/net/sunrpc/xprtrdma/backchannel.c b/net/sunrpc/xprtrdma/backchannel.c
index cc1251d07297..2dcd7640eeb5 100644
--- a/net/sunrpc/xprtrdma/backchannel.c
+++ b/net/sunrpc/xprtrdma/backchannel.c
@@ -341,6 +341,8 @@ void rpcrdma_bc_receive_call(struct rpcrdma_xprt *r_xprt,
 	rqst->rq_reply_bytes_recvd = 0;
 	rqst->rq_bytes_sent = 0;
 	rqst->rq_xid = headerp->rm_xid;
+
+	rqst->rq_private_buf.len = size;
 	set_bit(RPC_BC_PA_IN_USE, &rqst->rq_bc_pa_state);
 
 	buf = &rqst->rq_rcv_buf;
diff --git a/net/switchdev/switchdev.c b/net/switchdev/switchdev.c
index 47f7da58a7f0..8b5833c1ff2e 100644
--- a/net/switchdev/switchdev.c
+++ b/net/switchdev/switchdev.c
@@ -1093,8 +1093,11 @@ int switchdev_port_fdb_dump(struct sk_buff *skb, struct netlink_callback *cb,
 		.cb = cb,
 		.idx = idx,
 	};
+	int err;
 
-	switchdev_port_obj_dump(dev, &dump.fdb.obj, switchdev_port_fdb_dump_cb);
+	err = switchdev_port_obj_dump(dev, &dump.fdb.obj,
+				      switchdev_port_fdb_dump_cb);
+	cb->args[1] = err;
 	return dump.idx;
 }
 EXPORT_SYMBOL_GPL(switchdev_port_fdb_dump);
diff --git a/net/tipc/link.c b/net/tipc/link.c
index 0c2944fb9ae0..347cdc99ed09 100644
--- a/net/tipc/link.c
+++ b/net/tipc/link.c
@@ -1973,8 +1973,10 @@ int tipc_nl_add_bc_link(struct net *net, struct tipc_nl_msg *msg)
 
 	hdr = genlmsg_put(msg->skb, msg->portid, msg->seq, &tipc_genl_family,
 			  NLM_F_MULTI, TIPC_NL_LINK_GET);
-	if (!hdr)
+	if (!hdr) {
+		tipc_bcast_unlock(net);
 		return -EMSGSIZE;
+	}
 
 	attrs = nla_nest_start(msg->skb, TIPC_NLA_LINK);
 	if (!attrs)
diff --git a/net/tipc/node.c b/net/tipc/node.c
index fa97d9649a28..9d7a16fc5ca4 100644
--- a/net/tipc/node.c
+++ b/net/tipc/node.c
@@ -346,12 +346,6 @@ struct tipc_node *tipc_node_create(struct net *net, u32 addr, u16 capabilities)
 	skb_queue_head_init(&n->bc_entry.inputq2);
 	for (i = 0; i < MAX_BEARERS; i++)
 		spin_lock_init(&n->links[i].lock);
-	hlist_add_head_rcu(&n->hash, &tn->node_htable[tipc_hashfn(addr)]);
-	list_for_each_entry_rcu(temp_node, &tn->node_list, list) {
-		if (n->addr < temp_node->addr)
-			break;
-	}
-	list_add_tail_rcu(&n->list, &temp_node->list);
 	n->state = SELF_DOWN_PEER_LEAVING;
 	n->signature = INVALID_NODE_SIG;
 	n->active_links[0] = INVALID_BEARER_ID;
@@ -372,6 +366,12 @@ struct tipc_node *tipc_node_create(struct net *net, u32 addr, u16 capabilities)
 	tipc_node_get(n);
 	setup_timer(&n->timer, tipc_node_timeout, (unsigned long)n);
 	n->keepalive_intv = U32_MAX;
+	hlist_add_head_rcu(&n->hash, &tn->node_htable[tipc_hashfn(addr)]);
+	list_for_each_entry_rcu(temp_node, &tn->node_list, list) {
+		if (n->addr < temp_node->addr)
+			break;
+	}
+	list_add_tail_rcu(&n->list, &temp_node->list);
 exit:
 	spin_unlock_bh(&tn->node_list_lock);
 	return n;
diff --git a/net/tipc/socket.c b/net/tipc/socket.c
index 69c29050f14a..4d420bb27396 100644
--- a/net/tipc/socket.c
+++ b/net/tipc/socket.c
@@ -673,7 +673,7 @@ static int tipc_sendmcast(struct  socket *sock, struct tipc_name_seq *seq,
 	struct tipc_sock *tsk = tipc_sk(sk);
 	struct net *net = sock_net(sk);
 	struct tipc_msg *mhdr = &tsk->phdr;
-	struct sk_buff_head *pktchain = &sk->sk_write_queue;
+	struct sk_buff_head pktchain;
 	struct iov_iter save = msg->msg_iter;
 	uint mtu;
 	int rc;
@@ -687,14 +687,16 @@ static int tipc_sendmcast(struct  socket *sock, struct tipc_name_seq *seq,
 	msg_set_nameupper(mhdr, seq->upper);
 	msg_set_hdr_sz(mhdr, MCAST_H_SIZE);
 
+	skb_queue_head_init(&pktchain);
+
 new_mtu:
 	mtu = tipc_bcast_get_mtu(net);
-	rc = tipc_msg_build(mhdr, msg, 0, dsz, mtu, pktchain);
+	rc = tipc_msg_build(mhdr, msg, 0, dsz, mtu, &pktchain);
 	if (unlikely(rc < 0))
 		return rc;
 
 	do {
-		rc = tipc_bcast_xmit(net, pktchain);
+		rc = tipc_bcast_xmit(net, &pktchain);
 		if (likely(!rc))
 			return dsz;
 
@@ -704,7 +706,7 @@ new_mtu:
 			if (!rc)
 				continue;
 		}
-		__skb_queue_purge(pktchain);
+		__skb_queue_purge(&pktchain);
 		if (rc == -EMSGSIZE) {
 			msg->msg_iter = save;
 			goto new_mtu;
@@ -863,7 +865,7 @@ static int __tipc_sendmsg(struct socket *sock, struct msghdr *m, size_t dsz)
 	struct net *net = sock_net(sk);
 	struct tipc_msg *mhdr = &tsk->phdr;
 	u32 dnode, dport;
-	struct sk_buff_head *pktchain = &sk->sk_write_queue;
+	struct sk_buff_head pktchain;
 	struct sk_buff *skb;
 	struct tipc_name_seq *seq;
 	struct iov_iter save;
@@ -924,17 +926,18 @@ static int __tipc_sendmsg(struct socket *sock, struct msghdr *m, size_t dsz)
 		msg_set_hdr_sz(mhdr, BASIC_H_SIZE);
 	}
 
+	skb_queue_head_init(&pktchain);
 	save = m->msg_iter;
 new_mtu:
 	mtu = tipc_node_get_mtu(net, dnode, tsk->portid);
-	rc = tipc_msg_build(mhdr, m, 0, dsz, mtu, pktchain);
+	rc = tipc_msg_build(mhdr, m, 0, dsz, mtu, &pktchain);
 	if (rc < 0)
 		return rc;
 
 	do {
-		skb = skb_peek(pktchain);
+		skb = skb_peek(&pktchain);
 		TIPC_SKB_CB(skb)->wakeup_pending = tsk->link_cong;
-		rc = tipc_node_xmit(net, pktchain, dnode, tsk->portid);
+		rc = tipc_node_xmit(net, &pktchain, dnode, tsk->portid);
 		if (likely(!rc)) {
 			if (sock->state != SS_READY)
 				sock->state = SS_CONNECTING;
@@ -946,7 +949,7 @@ new_mtu:
 			if (!rc)
 				continue;
 		}
-		__skb_queue_purge(pktchain);
+		__skb_queue_purge(&pktchain);
 		if (rc == -EMSGSIZE) {
 			m->msg_iter = save;
 			goto new_mtu;
@@ -1016,7 +1019,7 @@ static int __tipc_send_stream(struct socket *sock, struct msghdr *m, size_t dsz)
 	struct net *net = sock_net(sk);
 	struct tipc_sock *tsk = tipc_sk(sk);
 	struct tipc_msg *mhdr = &tsk->phdr;
-	struct sk_buff_head *pktchain = &sk->sk_write_queue;
+	struct sk_buff_head pktchain;
 	DECLARE_SOCKADDR(struct sockaddr_tipc *, dest, m->msg_name);
 	u32 portid = tsk->portid;
 	int rc = -EINVAL;
@@ -1044,17 +1047,19 @@ static int __tipc_send_stream(struct socket *sock, struct msghdr *m, size_t dsz)
 
 	timeo = sock_sndtimeo(sk, m->msg_flags & MSG_DONTWAIT);
 	dnode = tsk_peer_node(tsk);
+	skb_queue_head_init(&pktchain);
 
 next:
 	save = m->msg_iter;
 	mtu = tsk->max_pkt;
 	send = min_t(uint, dsz - sent, TIPC_MAX_USER_MSG_SIZE);
-	rc = tipc_msg_build(mhdr, m, sent, send, mtu, pktchain);
+	rc = tipc_msg_build(mhdr, m, sent, send, mtu, &pktchain);
 	if (unlikely(rc < 0))
 		return rc;
+
 	do {
 		if (likely(!tsk_conn_cong(tsk))) {
-			rc = tipc_node_xmit(net, pktchain, dnode, portid);
+			rc = tipc_node_xmit(net, &pktchain, dnode, portid);
 			if (likely(!rc)) {
 				tsk->sent_unacked++;
 				sent += send;
@@ -1063,7 +1068,7 @@ next:
 				goto next;
 			}
 			if (rc == -EMSGSIZE) {
-				__skb_queue_purge(pktchain);
+				__skb_queue_purge(&pktchain);
 				tsk->max_pkt = tipc_node_get_mtu(net, dnode,
 								 portid);
 				m->msg_iter = save;
@@ -1077,7 +1082,7 @@ next:
 		rc = tipc_wait_for_sndpkt(sock, &timeo);
 	} while (!rc);
 
-	__skb_queue_purge(pktchain);
+	__skb_queue_purge(&pktchain);
 	return sent ? sent : rc;
 }
 
diff --git a/net/tipc/subscr.c b/net/tipc/subscr.c
index 69ee2eeef968..f9ff73a8d815 100644
--- a/net/tipc/subscr.c
+++ b/net/tipc/subscr.c
@@ -296,7 +296,8 @@ static void tipc_subscrb_rcv_cb(struct net *net, int conid,
 	if (tipc_subscrp_create(net, (struct tipc_subscr *)buf, subscrb, &sub))
 		return tipc_conn_terminate(tn->topsrv, subscrb->conid);
 
-	tipc_nametbl_subscribe(sub);
+	if (sub)
+		tipc_nametbl_subscribe(sub);
 }
 
 /* Handle one request to establish a new subscriber */
diff --git a/net/unix/af_unix.c b/net/unix/af_unix.c
index 49d5093eb055..f75f847e688d 100644
--- a/net/unix/af_unix.c
+++ b/net/unix/af_unix.c
@@ -1496,7 +1496,7 @@ static void unix_detach_fds(struct scm_cookie *scm, struct sk_buff *skb)
 	UNIXCB(skb).fp = NULL;
 
 	for (i = scm->fp->count-1; i >= 0; i--)
-		unix_notinflight(scm->fp->fp[i]);
+		unix_notinflight(scm->fp->user, scm->fp->fp[i]);
 }
 
 static void unix_destruct_scm(struct sk_buff *skb)
@@ -1561,7 +1561,7 @@ static int unix_attach_fds(struct scm_cookie *scm, struct sk_buff *skb)
 		return -ENOMEM;
 
 	for (i = scm->fp->count - 1; i >= 0; i--)
-		unix_inflight(scm->fp->fp[i]);
+		unix_inflight(scm->fp->user, scm->fp->fp[i]);
 	return max_level;
 }
 
@@ -1781,7 +1781,12 @@ restart_locked:
 			goto out_unlock;
 	}
 
-	if (unlikely(unix_peer(other) != sk && unix_recvq_full(other))) {
+	/* other == sk && unix_peer(other) != sk if
+	 * - unix_peer(sk) == NULL, destination address bound to sk
+	 * - unix_peer(sk) == sk by time of get but disconnected before lock
+	 */
+	if (other != sk &&
+	    unlikely(unix_peer(other) != sk && unix_recvq_full(other))) {
 		if (timeo) {
 			timeo = unix_wait_for_peer(other, timeo);
 
@@ -2277,13 +2282,15 @@ static int unix_stream_read_generic(struct unix_stream_read_state *state)
 	size_t size = state->size;
 	unsigned int last_len;
 
-	err = -EINVAL;
-	if (sk->sk_state != TCP_ESTABLISHED)
+	if (unlikely(sk->sk_state != TCP_ESTABLISHED)) {
+		err = -EINVAL;
 		goto out;
+	}
 
-	err = -EOPNOTSUPP;
-	if (flags & MSG_OOB)
+	if (unlikely(flags & MSG_OOB)) {
+		err = -EOPNOTSUPP;
 		goto out;
+	}
 
 	target = sock_rcvlowat(sk, flags & MSG_WAITALL, size);
 	timeo = sock_rcvtimeo(sk, noblock);
@@ -2305,6 +2312,7 @@ static int unix_stream_read_generic(struct unix_stream_read_state *state)
 		bool drop_skb;
 		struct sk_buff *skb, *last;
 
+redo:
 		unix_state_lock(sk);
 		if (sock_flag(sk, SOCK_DEAD)) {
 			err = -ECONNRESET;
@@ -2329,9 +2337,11 @@ again:
 				goto unlock;
 
 			unix_state_unlock(sk);
-			err = -EAGAIN;
-			if (!timeo)
+			if (!timeo) {
+				err = -EAGAIN;
 				break;
+			}
+
 			mutex_unlock(&u->readlock);
 
 			timeo = unix_stream_data_wait(sk, timeo, last,
@@ -2344,7 +2354,7 @@ again:
 			}
 
 			mutex_lock(&u->readlock);
-			continue;
+			goto redo;
 unlock:
 			unix_state_unlock(sk);
 			break;
diff --git a/net/unix/diag.c b/net/unix/diag.c
index c512f64d5287..4d9679701a6d 100644
--- a/net/unix/diag.c
+++ b/net/unix/diag.c
@@ -220,7 +220,7 @@ done:
 	return skb->len;
 }
 
-static struct sock *unix_lookup_by_ino(int ino)
+static struct sock *unix_lookup_by_ino(unsigned int ino)
 {
 	int i;
 	struct sock *sk;
diff --git a/net/unix/garbage.c b/net/unix/garbage.c
index 8fcdc2283af5..6a0d48525fcf 100644
--- a/net/unix/garbage.c
+++ b/net/unix/garbage.c
@@ -116,7 +116,7 @@ struct sock *unix_get_socket(struct file *filp)
  * descriptor if it is for an AF_UNIX socket.
  */
 
-void unix_inflight(struct file *fp)
+void unix_inflight(struct user_struct *user, struct file *fp)
 {
 	struct sock *s = unix_get_socket(fp);
 
@@ -133,11 +133,11 @@ void unix_inflight(struct file *fp)
 		}
 		unix_tot_inflight++;
 	}
-	fp->f_cred->user->unix_inflight++;
+	user->unix_inflight++;
 	spin_unlock(&unix_gc_lock);
 }
 
-void unix_notinflight(struct file *fp)
+void unix_notinflight(struct user_struct *user, struct file *fp)
 {
 	struct sock *s = unix_get_socket(fp);
 
@@ -152,7 +152,7 @@ void unix_notinflight(struct file *fp)
 			list_del_init(&u->link);
 		unix_tot_inflight--;
 	}
-	fp->f_cred->user->unix_inflight--;
+	user->unix_inflight--;
 	spin_unlock(&unix_gc_lock);
 }
 
diff --git a/net/vmw_vsock/af_vsock.c b/net/vmw_vsock/af_vsock.c
index 7fd1220fbfa0..bbe65dcb9738 100644
--- a/net/vmw_vsock/af_vsock.c
+++ b/net/vmw_vsock/af_vsock.c
@@ -1557,8 +1557,6 @@ static int vsock_stream_sendmsg(struct socket *sock, struct msghdr *msg,
 	if (err < 0)
 		goto out;
 
-	prepare_to_wait(sk_sleep(sk), &wait, TASK_INTERRUPTIBLE);
-
 	while (total_written < len) {
 		ssize_t written;
 
@@ -1578,7 +1576,9 @@ static int vsock_stream_sendmsg(struct socket *sock, struct msghdr *msg,
 				goto out_wait;
 
 			release_sock(sk);
+			prepare_to_wait(sk_sleep(sk), &wait, TASK_INTERRUPTIBLE);
 			timeout = schedule_timeout(timeout);
+			finish_wait(sk_sleep(sk), &wait);
 			lock_sock(sk);
 			if (signal_pending(current)) {
 				err = sock_intr_errno(timeout);
@@ -1588,8 +1588,6 @@ static int vsock_stream_sendmsg(struct socket *sock, struct msghdr *msg,
 				goto out_wait;
 			}
 
-			prepare_to_wait(sk_sleep(sk), &wait,
-					TASK_INTERRUPTIBLE);
 		}
 
 		/* These checks occur both as part of and after the loop
@@ -1635,7 +1633,6 @@ static int vsock_stream_sendmsg(struct socket *sock, struct msghdr *msg,
 out_wait:
 	if (total_written > 0)
 		err = total_written;
-	finish_wait(sk_sleep(sk), &wait);
 out:
 	release_sock(sk);
 	return err;
@@ -1716,7 +1713,6 @@ vsock_stream_recvmsg(struct socket *sock, struct msghdr *msg, size_t len,
 	if (err < 0)
 		goto out;
 
-	prepare_to_wait(sk_sleep(sk), &wait, TASK_INTERRUPTIBLE);
 
 	while (1) {
 		s64 ready = vsock_stream_has_data(vsk);
@@ -1727,7 +1723,7 @@ vsock_stream_recvmsg(struct socket *sock, struct msghdr *msg, size_t len,
 			 */
 
 			err = -ENOMEM;
-			goto out_wait;
+			goto out;
 		} else if (ready > 0) {
 			ssize_t read;
 
@@ -1750,7 +1746,7 @@ vsock_stream_recvmsg(struct socket *sock, struct msghdr *msg, size_t len,
 					vsk, target, read,
 					!(flags & MSG_PEEK), &recv_data);
 			if (err < 0)
-				goto out_wait;
+				goto out;
 
 			if (read >= target || flags & MSG_PEEK)
 				break;
@@ -1773,7 +1769,9 @@ vsock_stream_recvmsg(struct socket *sock, struct msghdr *msg, size_t len,
 				break;
 
 			release_sock(sk);
+			prepare_to_wait(sk_sleep(sk), &wait, TASK_INTERRUPTIBLE);
 			timeout = schedule_timeout(timeout);
+			finish_wait(sk_sleep(sk), &wait);
 			lock_sock(sk);
 
 			if (signal_pending(current)) {
@@ -1783,9 +1781,6 @@ vsock_stream_recvmsg(struct socket *sock, struct msghdr *msg, size_t len,
 				err = -EAGAIN;
 				break;
 			}
-
-			prepare_to_wait(sk_sleep(sk), &wait,
-					TASK_INTERRUPTIBLE);
 		}
 	}
 
@@ -1816,8 +1811,6 @@ vsock_stream_recvmsg(struct socket *sock, struct msghdr *msg, size_t len,
 		err = copied;
 	}
 
-out_wait:
-	finish_wait(sk_sleep(sk), &wait);
 out:
 	release_sock(sk);
 	return err;
diff --git a/net/wireless/core.c b/net/wireless/core.c
index b0915515640e..8f0bac7e03c4 100644
--- a/net/wireless/core.c
+++ b/net/wireless/core.c
@@ -1147,6 +1147,8 @@ static int cfg80211_netdev_notifier_call(struct notifier_block *nb,
 		return NOTIFY_DONE;
 	}
 
+	wireless_nlevent_flush();
+
 	return NOTIFY_OK;
 }
 
diff --git a/net/wireless/nl80211.c b/net/wireless/nl80211.c
index d4786f2802aa..711cb7ad6ae0 100644
--- a/net/wireless/nl80211.c
+++ b/net/wireless/nl80211.c
@@ -7547,7 +7547,7 @@ static int nl80211_join_ibss(struct sk_buff *skb, struct genl_info *info)
 
 		if ((ibss.chandef.width != NL80211_CHAN_WIDTH_20_NOHT) &&
 		    no_ht) {
-			kfree(connkeys);
+			kzfree(connkeys);
 			return -EINVAL;
 		}
 	}
diff --git a/net/wireless/sme.c b/net/wireless/sme.c
index 8020b5b094d4..d49ed7666d4c 100644
--- a/net/wireless/sme.c
+++ b/net/wireless/sme.c
@@ -917,6 +917,12 @@ void __cfg80211_disconnected(struct net_device *dev, const u8 *ie,
 
 	nl80211_send_disconnected(rdev, dev, reason, ie, ie_len, from_ap);
 
+	/* stop critical protocol if supported */
+	if (rdev->ops->crit_proto_stop && rdev->crit_proto_nlportid) {
+		rdev->crit_proto_nlportid = 0;
+		rdev_crit_proto_stop(rdev, wdev);
+	}
+
 	/*
 	 * Delete all the keys ... pairwise keys can't really
 	 * exist any more anyway, but default keys might.
diff --git a/net/wireless/wext-core.c b/net/wireless/wext-core.c
index c8717c1d082e..b50ee5d622e1 100644
--- a/net/wireless/wext-core.c
+++ b/net/wireless/wext-core.c
@@ -342,6 +342,40 @@ static const int compat_event_type_size[] = {
 
 /* IW event code */
 
+void wireless_nlevent_flush(void)
+{
+	struct sk_buff *skb;
+	struct net *net;
+
+	ASSERT_RTNL();
+
+	for_each_net(net) {
+		while ((skb = skb_dequeue(&net->wext_nlevents)))
+			rtnl_notify(skb, net, 0, RTNLGRP_LINK, NULL,
+				    GFP_KERNEL);
+	}
+}
+EXPORT_SYMBOL_GPL(wireless_nlevent_flush);
+
+static int wext_netdev_notifier_call(struct notifier_block *nb,
+				     unsigned long state, void *ptr)
+{
+	/*
+	 * When a netdev changes state in any way, flush all pending messages
+	 * to avoid them going out in a strange order, e.g. RTM_NEWLINK after
+	 * RTM_DELLINK, or with IFF_UP after without IFF_UP during dev_close()
+	 * or similar - all of which could otherwise happen due to delays from
+	 * schedule_work().
+	 */
+	wireless_nlevent_flush();
+
+	return NOTIFY_OK;
+}
+
+static struct notifier_block wext_netdev_notifier = {
+	.notifier_call = wext_netdev_notifier_call,
+};
+
 static int __net_init wext_pernet_init(struct net *net)
 {
 	skb_queue_head_init(&net->wext_nlevents);
@@ -360,7 +394,12 @@ static struct pernet_operations wext_pernet_ops = {
 
 static int __init wireless_nlevent_init(void)
 {
-	return register_pernet_subsys(&wext_pernet_ops);
+	int err = register_pernet_subsys(&wext_pernet_ops);
+
+	if (err)
+		return err;
+
+	return register_netdevice_notifier(&wext_netdev_notifier);
 }
 
 subsys_initcall(wireless_nlevent_init);
@@ -368,17 +407,8 @@ subsys_initcall(wireless_nlevent_init);
 /* Process events generated by the wireless layer or the driver. */
 static void wireless_nlevent_process(struct work_struct *work)
 {
-	struct sk_buff *skb;
-	struct net *net;
-
 	rtnl_lock();
-
-	for_each_net(net) {
-		while ((skb = skb_dequeue(&net->wext_nlevents)))
-			rtnl_notify(skb, net, 0, RTNLGRP_LINK, NULL,
-				    GFP_KERNEL);
-	}
-
+	wireless_nlevent_flush();
 	rtnl_unlock();
 }
 
diff --git a/scripts/prune-kernel b/scripts/prune-kernel
new file mode 100755
index 000000000000..ab5034e1d081
--- /dev/null
+++ b/scripts/prune-kernel
@@ -0,0 +1,20 @@
+#!/bin/bash
+
+# because I use CONFIG_LOCALVERSION_AUTO, not the same version again and
+# again, /boot and /lib/modules/ eventually fill up.
+# Dumb script to purge that stuff:
+
+for f in "$@"
+do
+        if rpm -qf "/lib/modules/$f" >/dev/null; then
+                echo "keeping $f (installed from rpm)"
+        elif [ $(uname -r) = "$f" ]; then
+                echo "keeping $f (running kernel) "
+        else
+                echo "removing $f"
+                rm -f "/boot/initramfs-$f.img" "/boot/System.map-$f"
+                rm -f "/boot/vmlinuz-$f"   "/boot/config-$f"
+                rm -rf "/lib/modules/$f"
+                new-kernel-pkg --remove $f
+        fi
+done
diff --git a/security/integrity/evm/evm_main.c b/security/integrity/evm/evm_main.c
index f7160253f17f..e6ea9d4b1de9 100644
--- a/security/integrity/evm/evm_main.c
+++ b/security/integrity/evm/evm_main.c
@@ -23,6 +23,7 @@
 #include <linux/integrity.h>
 #include <linux/evm.h>
 #include <crypto/hash.h>
+#include <crypto/algapi.h>
 #include "evm.h"
 
 int evm_initialized;
@@ -148,7 +149,7 @@ static enum integrity_status evm_verify_hmac(struct dentry *dentry,
 				   xattr_value_len, calc.digest);
 		if (rc)
 			break;
-		rc = memcmp(xattr_data->digest, calc.digest,
+		rc = crypto_memneq(xattr_data->digest, calc.digest,
 			    sizeof(calc.digest));
 		if (rc)
 			rc = -EINVAL;
diff --git a/security/selinux/hooks.c b/security/selinux/hooks.c
index f8110cfd80ff..f1ab71504e1d 100644
--- a/security/selinux/hooks.c
+++ b/security/selinux/hooks.c
@@ -3249,7 +3249,7 @@ static int selinux_inode_listsecurity(struct inode *inode, char *buffer, size_t
 
 static void selinux_inode_getsecid(struct inode *inode, u32 *secid)
 {
-	struct inode_security_struct *isec = inode_security(inode);
+	struct inode_security_struct *isec = inode_security_novalidate(inode);
 	*secid = isec->sid;
 }
 
diff --git a/security/selinux/nlmsgtab.c b/security/selinux/nlmsgtab.c
index 2bbb41822d8e..8495b9368190 100644
--- a/security/selinux/nlmsgtab.c
+++ b/security/selinux/nlmsgtab.c
@@ -83,6 +83,7 @@ static struct nlmsg_perm nlmsg_tcpdiag_perms[] =
 	{ TCPDIAG_GETSOCK,	NETLINK_TCPDIAG_SOCKET__NLMSG_READ },
 	{ DCCPDIAG_GETSOCK,	NETLINK_TCPDIAG_SOCKET__NLMSG_READ },
 	{ SOCK_DIAG_BY_FAMILY,	NETLINK_TCPDIAG_SOCKET__NLMSG_READ },
+	{ SOCK_DESTROY,		NETLINK_TCPDIAG_SOCKET__NLMSG_WRITE },
 };
 
 static struct nlmsg_perm nlmsg_xfrm_perms[] =
diff --git a/sound/core/control_compat.c b/sound/core/control_compat.c
index b9c0910fb8c4..0608f216f359 100644
--- a/sound/core/control_compat.c
+++ b/sound/core/control_compat.c
@@ -170,6 +170,19 @@ struct snd_ctl_elem_value32 {
         unsigned char reserved[128];
 };
 
+#ifdef CONFIG_X86_X32
+/* x32 has a different alignment for 64bit values from ia32 */
+struct snd_ctl_elem_value_x32 {
+	struct snd_ctl_elem_id id;
+	unsigned int indirect;	/* bit-field causes misalignment */
+	union {
+		s32 integer[128];
+		unsigned char data[512];
+		s64 integer64[64];
+	} value;
+	unsigned char reserved[128];
+};
+#endif /* CONFIG_X86_X32 */
 
 /* get the value type and count of the control */
 static int get_ctl_type(struct snd_card *card, struct snd_ctl_elem_id *id,
@@ -219,9 +232,11 @@ static int get_elem_size(int type, int count)
 
 static int copy_ctl_value_from_user(struct snd_card *card,
 				    struct snd_ctl_elem_value *data,
-				    struct snd_ctl_elem_value32 __user *data32,
+				    void __user *userdata,
+				    void __user *valuep,
 				    int *typep, int *countp)
 {
+	struct snd_ctl_elem_value32 __user *data32 = userdata;
 	int i, type, size;
 	int uninitialized_var(count);
 	unsigned int indirect;
@@ -239,8 +254,9 @@ static int copy_ctl_value_from_user(struct snd_card *card,
 	if (type == SNDRV_CTL_ELEM_TYPE_BOOLEAN ||
 	    type == SNDRV_CTL_ELEM_TYPE_INTEGER) {
 		for (i = 0; i < count; i++) {
+			s32 __user *intp = valuep;
 			int val;
-			if (get_user(val, &data32->value.integer[i]))
+			if (get_user(val, &intp[i]))
 				return -EFAULT;
 			data->value.integer.value[i] = val;
 		}
@@ -250,8 +266,7 @@ static int copy_ctl_value_from_user(struct snd_card *card,
 			dev_err(card->dev, "snd_ioctl32_ctl_elem_value: unknown type %d\n", type);
 			return -EINVAL;
 		}
-		if (copy_from_user(data->value.bytes.data,
-				   data32->value.data, size))
+		if (copy_from_user(data->value.bytes.data, valuep, size))
 			return -EFAULT;
 	}
 
@@ -261,7 +276,8 @@ static int copy_ctl_value_from_user(struct snd_card *card,
 }
 
 /* restore the value to 32bit */
-static int copy_ctl_value_to_user(struct snd_ctl_elem_value32 __user *data32,
+static int copy_ctl_value_to_user(void __user *userdata,
+				  void __user *valuep,
 				  struct snd_ctl_elem_value *data,
 				  int type, int count)
 {
@@ -270,22 +286,22 @@ static int copy_ctl_value_to_user(struct snd_ctl_elem_value32 __user *data32,
 	if (type == SNDRV_CTL_ELEM_TYPE_BOOLEAN ||
 	    type == SNDRV_CTL_ELEM_TYPE_INTEGER) {
 		for (i = 0; i < count; i++) {
+			s32 __user *intp = valuep;
 			int val;
 			val = data->value.integer.value[i];
-			if (put_user(val, &data32->value.integer[i]))
+			if (put_user(val, &intp[i]))
 				return -EFAULT;
 		}
 	} else {
 		size = get_elem_size(type, count);
-		if (copy_to_user(data32->value.data,
-				 data->value.bytes.data, size))
+		if (copy_to_user(valuep, data->value.bytes.data, size))
 			return -EFAULT;
 	}
 	return 0;
 }
 
-static int snd_ctl_elem_read_user_compat(struct snd_card *card, 
-					 struct snd_ctl_elem_value32 __user *data32)
+static int ctl_elem_read_user(struct snd_card *card,
+			      void __user *userdata, void __user *valuep)
 {
 	struct snd_ctl_elem_value *data;
 	int err, type, count;
@@ -294,7 +310,9 @@ static int snd_ctl_elem_read_user_compat(struct snd_card *card,
 	if (data == NULL)
 		return -ENOMEM;
 
-	if ((err = copy_ctl_value_from_user(card, data, data32, &type, &count)) < 0)
+	err = copy_ctl_value_from_user(card, data, userdata, valuep,
+				       &type, &count);
+	if (err < 0)
 		goto error;
 
 	snd_power_lock(card);
@@ -303,14 +321,15 @@ static int snd_ctl_elem_read_user_compat(struct snd_card *card,
 		err = snd_ctl_elem_read(card, data);
 	snd_power_unlock(card);
 	if (err >= 0)
-		err = copy_ctl_value_to_user(data32, data, type, count);
+		err = copy_ctl_value_to_user(userdata, valuep, data,
+					     type, count);
  error:
 	kfree(data);
 	return err;
 }
 
-static int snd_ctl_elem_write_user_compat(struct snd_ctl_file *file,
-					  struct snd_ctl_elem_value32 __user *data32)
+static int ctl_elem_write_user(struct snd_ctl_file *file,
+			       void __user *userdata, void __user *valuep)
 {
 	struct snd_ctl_elem_value *data;
 	struct snd_card *card = file->card;
@@ -320,7 +339,9 @@ static int snd_ctl_elem_write_user_compat(struct snd_ctl_file *file,
 	if (data == NULL)
 		return -ENOMEM;
 
-	if ((err = copy_ctl_value_from_user(card, data, data32, &type, &count)) < 0)
+	err = copy_ctl_value_from_user(card, data, userdata, valuep,
+				       &type, &count);
+	if (err < 0)
 		goto error;
 
 	snd_power_lock(card);
@@ -329,12 +350,39 @@ static int snd_ctl_elem_write_user_compat(struct snd_ctl_file *file,
 		err = snd_ctl_elem_write(card, file, data);
 	snd_power_unlock(card);
 	if (err >= 0)
-		err = copy_ctl_value_to_user(data32, data, type, count);
+		err = copy_ctl_value_to_user(userdata, valuep, data,
+					     type, count);
  error:
 	kfree(data);
 	return err;
 }
 
+static int snd_ctl_elem_read_user_compat(struct snd_card *card,
+					 struct snd_ctl_elem_value32 __user *data32)
+{
+	return ctl_elem_read_user(card, data32, &data32->value);
+}
+
+static int snd_ctl_elem_write_user_compat(struct snd_ctl_file *file,
+					  struct snd_ctl_elem_value32 __user *data32)
+{
+	return ctl_elem_write_user(file, data32, &data32->value);
+}
+
+#ifdef CONFIG_X86_X32
+static int snd_ctl_elem_read_user_x32(struct snd_card *card,
+				      struct snd_ctl_elem_value_x32 __user *data32)
+{
+	return ctl_elem_read_user(card, data32, &data32->value);
+}
+
+static int snd_ctl_elem_write_user_x32(struct snd_ctl_file *file,
+				       struct snd_ctl_elem_value_x32 __user *data32)
+{
+	return ctl_elem_write_user(file, data32, &data32->value);
+}
+#endif /* CONFIG_X86_X32 */
+
 /* add or replace a user control */
 static int snd_ctl_elem_add_compat(struct snd_ctl_file *file,
 				   struct snd_ctl_elem_info32 __user *data32,
@@ -393,6 +441,10 @@ enum {
 	SNDRV_CTL_IOCTL_ELEM_WRITE32 = _IOWR('U', 0x13, struct snd_ctl_elem_value32),
 	SNDRV_CTL_IOCTL_ELEM_ADD32 = _IOWR('U', 0x17, struct snd_ctl_elem_info32),
 	SNDRV_CTL_IOCTL_ELEM_REPLACE32 = _IOWR('U', 0x18, struct snd_ctl_elem_info32),
+#ifdef CONFIG_X86_X32
+	SNDRV_CTL_IOCTL_ELEM_READ_X32 = _IOWR('U', 0x12, struct snd_ctl_elem_value_x32),
+	SNDRV_CTL_IOCTL_ELEM_WRITE_X32 = _IOWR('U', 0x13, struct snd_ctl_elem_value_x32),
+#endif /* CONFIG_X86_X32 */
 };
 
 static inline long snd_ctl_ioctl_compat(struct file *file, unsigned int cmd, unsigned long arg)
@@ -431,6 +483,12 @@ static inline long snd_ctl_ioctl_compat(struct file *file, unsigned int cmd, uns
 		return snd_ctl_elem_add_compat(ctl, argp, 0);
 	case SNDRV_CTL_IOCTL_ELEM_REPLACE32:
 		return snd_ctl_elem_add_compat(ctl, argp, 1);
+#ifdef CONFIG_X86_X32
+	case SNDRV_CTL_IOCTL_ELEM_READ_X32:
+		return snd_ctl_elem_read_user_x32(ctl->card, argp);
+	case SNDRV_CTL_IOCTL_ELEM_WRITE_X32:
+		return snd_ctl_elem_write_user_x32(ctl, argp);
+#endif /* CONFIG_X86_X32 */
 	}
 
 	down_read(&snd_ioctl_rwsem);
diff --git a/sound/core/pcm_compat.c b/sound/core/pcm_compat.c
index 9630e9f72b7b..1f64ab0c2a95 100644
--- a/sound/core/pcm_compat.c
+++ b/sound/core/pcm_compat.c
@@ -183,6 +183,14 @@ static int snd_pcm_ioctl_channel_info_compat(struct snd_pcm_substream *substream
 	return err;
 }
 
+#ifdef CONFIG_X86_X32
+/* X32 ABI has the same struct as x86-64 for snd_pcm_channel_info */
+static int snd_pcm_channel_info_user(struct snd_pcm_substream *substream,
+				     struct snd_pcm_channel_info __user *src);
+#define snd_pcm_ioctl_channel_info_x32(s, p)	\
+	snd_pcm_channel_info_user(s, p)
+#endif /* CONFIG_X86_X32 */
+
 struct snd_pcm_status32 {
 	s32 state;
 	struct compat_timespec trigger_tstamp;
@@ -243,6 +251,71 @@ static int snd_pcm_status_user_compat(struct snd_pcm_substream *substream,
 	return err;
 }
 
+#ifdef CONFIG_X86_X32
+/* X32 ABI has 64bit timespec and 64bit alignment */
+struct snd_pcm_status_x32 {
+	s32 state;
+	u32 rsvd; /* alignment */
+	struct timespec trigger_tstamp;
+	struct timespec tstamp;
+	u32 appl_ptr;
+	u32 hw_ptr;
+	s32 delay;
+	u32 avail;
+	u32 avail_max;
+	u32 overrange;
+	s32 suspended_state;
+	u32 audio_tstamp_data;
+	struct timespec audio_tstamp;
+	struct timespec driver_tstamp;
+	u32 audio_tstamp_accuracy;
+	unsigned char reserved[52-2*sizeof(struct timespec)];
+} __packed;
+
+#define put_timespec(src, dst) copy_to_user(dst, src, sizeof(*dst))
+
+static int snd_pcm_status_user_x32(struct snd_pcm_substream *substream,
+				   struct snd_pcm_status_x32 __user *src,
+				   bool ext)
+{
+	struct snd_pcm_status status;
+	int err;
+
+	memset(&status, 0, sizeof(status));
+	/*
+	 * with extension, parameters are read/write,
+	 * get audio_tstamp_data from user,
+	 * ignore rest of status structure
+	 */
+	if (ext && get_user(status.audio_tstamp_data,
+				(u32 __user *)(&src->audio_tstamp_data)))
+		return -EFAULT;
+	err = snd_pcm_status(substream, &status);
+	if (err < 0)
+		return err;
+
+	if (clear_user(src, sizeof(*src)))
+		return -EFAULT;
+	if (put_user(status.state, &src->state) ||
+	    put_timespec(&status.trigger_tstamp, &src->trigger_tstamp) ||
+	    put_timespec(&status.tstamp, &src->tstamp) ||
+	    put_user(status.appl_ptr, &src->appl_ptr) ||
+	    put_user(status.hw_ptr, &src->hw_ptr) ||
+	    put_user(status.delay, &src->delay) ||
+	    put_user(status.avail, &src->avail) ||
+	    put_user(status.avail_max, &src->avail_max) ||
+	    put_user(status.overrange, &src->overrange) ||
+	    put_user(status.suspended_state, &src->suspended_state) ||
+	    put_user(status.audio_tstamp_data, &src->audio_tstamp_data) ||
+	    put_timespec(&status.audio_tstamp, &src->audio_tstamp) ||
+	    put_timespec(&status.driver_tstamp, &src->driver_tstamp) ||
+	    put_user(status.audio_tstamp_accuracy, &src->audio_tstamp_accuracy))
+		return -EFAULT;
+
+	return err;
+}
+#endif /* CONFIG_X86_X32 */
+
 /* both for HW_PARAMS and HW_REFINE */
 static int snd_pcm_ioctl_hw_params_compat(struct snd_pcm_substream *substream,
 					  int refine, 
@@ -469,6 +542,93 @@ static int snd_pcm_ioctl_sync_ptr_compat(struct snd_pcm_substream *substream,
 	return 0;
 }
 
+#ifdef CONFIG_X86_X32
+/* X32 ABI has 64bit timespec and 64bit alignment */
+struct snd_pcm_mmap_status_x32 {
+	s32 state;
+	s32 pad1;
+	u32 hw_ptr;
+	u32 pad2; /* alignment */
+	struct timespec tstamp;
+	s32 suspended_state;
+	struct timespec audio_tstamp;
+} __packed;
+
+struct snd_pcm_mmap_control_x32 {
+	u32 appl_ptr;
+	u32 avail_min;
+};
+
+struct snd_pcm_sync_ptr_x32 {
+	u32 flags;
+	u32 rsvd; /* alignment */
+	union {
+		struct snd_pcm_mmap_status_x32 status;
+		unsigned char reserved[64];
+	} s;
+	union {
+		struct snd_pcm_mmap_control_x32 control;
+		unsigned char reserved[64];
+	} c;
+} __packed;
+
+static int snd_pcm_ioctl_sync_ptr_x32(struct snd_pcm_substream *substream,
+				      struct snd_pcm_sync_ptr_x32 __user *src)
+{
+	struct snd_pcm_runtime *runtime = substream->runtime;
+	volatile struct snd_pcm_mmap_status *status;
+	volatile struct snd_pcm_mmap_control *control;
+	u32 sflags;
+	struct snd_pcm_mmap_control scontrol;
+	struct snd_pcm_mmap_status sstatus;
+	snd_pcm_uframes_t boundary;
+	int err;
+
+	if (snd_BUG_ON(!runtime))
+		return -EINVAL;
+
+	if (get_user(sflags, &src->flags) ||
+	    get_user(scontrol.appl_ptr, &src->c.control.appl_ptr) ||
+	    get_user(scontrol.avail_min, &src->c.control.avail_min))
+		return -EFAULT;
+	if (sflags & SNDRV_PCM_SYNC_PTR_HWSYNC) {
+		err = snd_pcm_hwsync(substream);
+		if (err < 0)
+			return err;
+	}
+	status = runtime->status;
+	control = runtime->control;
+	boundary = recalculate_boundary(runtime);
+	if (!boundary)
+		boundary = 0x7fffffff;
+	snd_pcm_stream_lock_irq(substream);
+	/* FIXME: we should consider the boundary for the sync from app */
+	if (!(sflags & SNDRV_PCM_SYNC_PTR_APPL))
+		control->appl_ptr = scontrol.appl_ptr;
+	else
+		scontrol.appl_ptr = control->appl_ptr % boundary;
+	if (!(sflags & SNDRV_PCM_SYNC_PTR_AVAIL_MIN))
+		control->avail_min = scontrol.avail_min;
+	else
+		scontrol.avail_min = control->avail_min;
+	sstatus.state = status->state;
+	sstatus.hw_ptr = status->hw_ptr % boundary;
+	sstatus.tstamp = status->tstamp;
+	sstatus.suspended_state = status->suspended_state;
+	sstatus.audio_tstamp = status->audio_tstamp;
+	snd_pcm_stream_unlock_irq(substream);
+	if (put_user(sstatus.state, &src->s.status.state) ||
+	    put_user(sstatus.hw_ptr, &src->s.status.hw_ptr) ||
+	    put_timespec(&sstatus.tstamp, &src->s.status.tstamp) ||
+	    put_user(sstatus.suspended_state, &src->s.status.suspended_state) ||
+	    put_timespec(&sstatus.audio_tstamp, &src->s.status.audio_tstamp) ||
+	    put_user(scontrol.appl_ptr, &src->c.control.appl_ptr) ||
+	    put_user(scontrol.avail_min, &src->c.control.avail_min))
+		return -EFAULT;
+
+	return 0;
+}
+#endif /* CONFIG_X86_X32 */
 
 /*
  */
@@ -487,7 +647,12 @@ enum {
 	SNDRV_PCM_IOCTL_WRITEN_FRAMES32 = _IOW('A', 0x52, struct snd_xfern32),
 	SNDRV_PCM_IOCTL_READN_FRAMES32 = _IOR('A', 0x53, struct snd_xfern32),
 	SNDRV_PCM_IOCTL_SYNC_PTR32 = _IOWR('A', 0x23, struct snd_pcm_sync_ptr32),
-
+#ifdef CONFIG_X86_X32
+	SNDRV_PCM_IOCTL_CHANNEL_INFO_X32 = _IOR('A', 0x32, struct snd_pcm_channel_info),
+	SNDRV_PCM_IOCTL_STATUS_X32 = _IOR('A', 0x20, struct snd_pcm_status_x32),
+	SNDRV_PCM_IOCTL_STATUS_EXT_X32 = _IOWR('A', 0x24, struct snd_pcm_status_x32),
+	SNDRV_PCM_IOCTL_SYNC_PTR_X32 = _IOWR('A', 0x23, struct snd_pcm_sync_ptr_x32),
+#endif /* CONFIG_X86_X32 */
 };
 
 static long snd_pcm_ioctl_compat(struct file *file, unsigned int cmd, unsigned long arg)
@@ -559,6 +724,16 @@ static long snd_pcm_ioctl_compat(struct file *file, unsigned int cmd, unsigned l
 		return snd_pcm_ioctl_rewind_compat(substream, argp);
 	case SNDRV_PCM_IOCTL_FORWARD32:
 		return snd_pcm_ioctl_forward_compat(substream, argp);
+#ifdef CONFIG_X86_X32
+	case SNDRV_PCM_IOCTL_STATUS_X32:
+		return snd_pcm_status_user_x32(substream, argp, false);
+	case SNDRV_PCM_IOCTL_STATUS_EXT_X32:
+		return snd_pcm_status_user_x32(substream, argp, true);
+	case SNDRV_PCM_IOCTL_SYNC_PTR_X32:
+		return snd_pcm_ioctl_sync_ptr_x32(substream, argp);
+	case SNDRV_PCM_IOCTL_CHANNEL_INFO_X32:
+		return snd_pcm_ioctl_channel_info_x32(substream, argp);
+#endif /* CONFIG_X86_X32 */
 	}
 
 	return -ENOIOCTLCMD;
diff --git a/sound/core/pcm_native.c b/sound/core/pcm_native.c
index fadd3eb8e8bb..9106d8e2300e 100644
--- a/sound/core/pcm_native.c
+++ b/sound/core/pcm_native.c
@@ -74,6 +74,18 @@ static int snd_pcm_open(struct file *file, struct snd_pcm *pcm, int stream);
 static DEFINE_RWLOCK(snd_pcm_link_rwlock);
 static DECLARE_RWSEM(snd_pcm_link_rwsem);
 
+/* Writer in rwsem may block readers even during its waiting in queue,
+ * and this may lead to a deadlock when the code path takes read sem
+ * twice (e.g. one in snd_pcm_action_nonatomic() and another in
+ * snd_pcm_stream_lock()).  As a (suboptimal) workaround, let writer to
+ * spin until it gets the lock.
+ */
+static inline void down_write_nonblock(struct rw_semaphore *lock)
+{
+	while (!down_write_trylock(lock))
+		cond_resched();
+}
+
 /**
  * snd_pcm_stream_lock - Lock the PCM stream
  * @substream: PCM substream
@@ -1813,7 +1825,7 @@ static int snd_pcm_link(struct snd_pcm_substream *substream, int fd)
 		res = -ENOMEM;
 		goto _nolock;
 	}
-	down_write(&snd_pcm_link_rwsem);
+	down_write_nonblock(&snd_pcm_link_rwsem);
 	write_lock_irq(&snd_pcm_link_rwlock);
 	if (substream->runtime->status->state == SNDRV_PCM_STATE_OPEN ||
 	    substream->runtime->status->state != substream1->runtime->status->state ||
@@ -1860,7 +1872,7 @@ static int snd_pcm_unlink(struct snd_pcm_substream *substream)
 	struct snd_pcm_substream *s;
 	int res = 0;
 
-	down_write(&snd_pcm_link_rwsem);
+	down_write_nonblock(&snd_pcm_link_rwsem);
 	write_lock_irq(&snd_pcm_link_rwlock);
 	if (!snd_pcm_stream_linked(substream)) {
 		res = -EALREADY;
diff --git a/sound/core/rawmidi_compat.c b/sound/core/rawmidi_compat.c
index 5268c1f58c25..f69764d7cdd7 100644
--- a/sound/core/rawmidi_compat.c
+++ b/sound/core/rawmidi_compat.c
@@ -85,8 +85,7 @@ static int snd_rawmidi_ioctl_status_compat(struct snd_rawmidi_file *rfile,
 	if (err < 0)
 		return err;
 
-	if (put_user(status.tstamp.tv_sec, &src->tstamp.tv_sec) ||
-	    put_user(status.tstamp.tv_nsec, &src->tstamp.tv_nsec) ||
+	if (compat_put_timespec(&status.tstamp, &src->tstamp) ||
 	    put_user(status.avail, &src->avail) ||
 	    put_user(status.xruns, &src->xruns))
 		return -EFAULT;
@@ -94,9 +93,58 @@ static int snd_rawmidi_ioctl_status_compat(struct snd_rawmidi_file *rfile,
 	return 0;
 }
 
+#ifdef CONFIG_X86_X32
+/* X32 ABI has 64bit timespec and 64bit alignment */
+struct snd_rawmidi_status_x32 {
+	s32 stream;
+	u32 rsvd; /* alignment */
+	struct timespec tstamp;
+	u32 avail;
+	u32 xruns;
+	unsigned char reserved[16];
+} __attribute__((packed));
+
+#define put_timespec(src, dst) copy_to_user(dst, src, sizeof(*dst))
+
+static int snd_rawmidi_ioctl_status_x32(struct snd_rawmidi_file *rfile,
+					struct snd_rawmidi_status_x32 __user *src)
+{
+	int err;
+	struct snd_rawmidi_status status;
+
+	if (rfile->output == NULL)
+		return -EINVAL;
+	if (get_user(status.stream, &src->stream))
+		return -EFAULT;
+
+	switch (status.stream) {
+	case SNDRV_RAWMIDI_STREAM_OUTPUT:
+		err = snd_rawmidi_output_status(rfile->output, &status);
+		break;
+	case SNDRV_RAWMIDI_STREAM_INPUT:
+		err = snd_rawmidi_input_status(rfile->input, &status);
+		break;
+	default:
+		return -EINVAL;
+	}
+	if (err < 0)
+		return err;
+
+	if (put_timespec(&status.tstamp, &src->tstamp) ||
+	    put_user(status.avail, &src->avail) ||
+	    put_user(status.xruns, &src->xruns))
+		return -EFAULT;
+
+	return 0;
+}
+#endif /* CONFIG_X86_X32 */
+
 enum {
 	SNDRV_RAWMIDI_IOCTL_PARAMS32 = _IOWR('W', 0x10, struct snd_rawmidi_params32),
 	SNDRV_RAWMIDI_IOCTL_STATUS32 = _IOWR('W', 0x20, struct snd_rawmidi_status32),
+#ifdef CONFIG_X86_X32
+	SNDRV_RAWMIDI_IOCTL_STATUS_X32 = _IOWR('W', 0x20, struct snd_rawmidi_status_x32),
+#endif /* CONFIG_X86_X32 */
 };
 
 static long snd_rawmidi_ioctl_compat(struct file *file, unsigned int cmd, unsigned long arg)
@@ -115,6 +163,10 @@ static long snd_rawmidi_ioctl_compat(struct file *file, unsigned int cmd, unsign
 		return snd_rawmidi_ioctl_params_compat(rfile, argp);
 	case SNDRV_RAWMIDI_IOCTL_STATUS32:
 		return snd_rawmidi_ioctl_status_compat(rfile, argp);
+#ifdef CONFIG_X86_X32
+	case SNDRV_RAWMIDI_IOCTL_STATUS_X32:
+		return snd_rawmidi_ioctl_status_x32(rfile, argp);
+#endif /* CONFIG_X86_X32 */
 	}
 	return -ENOIOCTLCMD;
 }
diff --git a/sound/core/seq/oss/seq_oss.c b/sound/core/seq/oss/seq_oss.c
index 8db156b207f1..8cdf489df80e 100644
--- a/sound/core/seq/oss/seq_oss.c
+++ b/sound/core/seq/oss/seq_oss.c
@@ -149,8 +149,6 @@ odev_release(struct inode *inode, struct file *file)
 	if ((dp = file->private_data) == NULL)
 		return 0;
 
-	snd_seq_oss_drain_write(dp);
-
 	mutex_lock(&register_mutex);
 	snd_seq_oss_release(dp);
 	mutex_unlock(&register_mutex);
diff --git a/sound/core/seq/oss/seq_oss_device.h b/sound/core/seq/oss/seq_oss_device.h
index b43924325249..d7b4d016b547 100644
--- a/sound/core/seq/oss/seq_oss_device.h
+++ b/sound/core/seq/oss/seq_oss_device.h
@@ -127,7 +127,6 @@ int snd_seq_oss_write(struct seq_oss_devinfo *dp, const char __user *buf, int co
 unsigned int snd_seq_oss_poll(struct seq_oss_devinfo *dp, struct file *file, poll_table * wait);
 
 void snd_seq_oss_reset(struct seq_oss_devinfo *dp);
-void snd_seq_oss_drain_write(struct seq_oss_devinfo *dp);
 
 /* */
 void snd_seq_oss_process_queue(struct seq_oss_devinfo *dp, abstime_t time);
diff --git a/sound/core/seq/oss/seq_oss_init.c b/sound/core/seq/oss/seq_oss_init.c
index 6779e82b46dd..92c96a95a903 100644
--- a/sound/core/seq/oss/seq_oss_init.c
+++ b/sound/core/seq/oss/seq_oss_init.c
@@ -436,22 +436,6 @@ snd_seq_oss_release(struct seq_oss_devinfo *dp)
 
 
 /*
- * Wait until the queue is empty (if we don't have nonblock)
- */
-void
-snd_seq_oss_drain_write(struct seq_oss_devinfo *dp)
-{
-	if (! dp->timer->running)
-		return;
-	if (is_write_mode(dp->file_mode) && !is_nonblock_mode(dp->file_mode) &&
-	    dp->writeq) {
-		while (snd_seq_oss_writeq_sync(dp->writeq))
-			;
-	}
-}
-
-
-/*
  * reset sequencer devices
  */
 void
diff --git a/sound/core/seq/seq_memory.c b/sound/core/seq/seq_memory.c
index 801076687bb1..c850345c43b5 100644
--- a/sound/core/seq/seq_memory.c
+++ b/sound/core/seq/seq_memory.c
@@ -383,15 +383,20 @@ int snd_seq_pool_init(struct snd_seq_pool *pool)
 
 	if (snd_BUG_ON(!pool))
 		return -EINVAL;
-	if (pool->ptr)			/* should be atomic? */
-		return 0;
 
-	pool->ptr = vmalloc(sizeof(struct snd_seq_event_cell) * pool->size);
-	if (!pool->ptr)
+	cellptr = vmalloc(sizeof(struct snd_seq_event_cell) * pool->size);
+	if (!cellptr)
 		return -ENOMEM;
 
 	/* add new cells to the free cell list */
 	spin_lock_irqsave(&pool->lock, flags);
+	if (pool->ptr) {
+		spin_unlock_irqrestore(&pool->lock, flags);
+		vfree(cellptr);
+		return 0;
+	}
+
+	pool->ptr = cellptr;
 	pool->free = NULL;
 
 	for (cell = 0; cell < pool->size; cell++) {
diff --git a/sound/core/seq/seq_ports.c b/sound/core/seq/seq_ports.c
index 921fb2bd8fad..fe686ee41c6d 100644
--- a/sound/core/seq/seq_ports.c
+++ b/sound/core/seq/seq_ports.c
@@ -535,19 +535,22 @@ static void delete_and_unsubscribe_port(struct snd_seq_client *client,
 					bool is_src, bool ack)
 {
 	struct snd_seq_port_subs_info *grp;
+	struct list_head *list;
+	bool empty;
 
 	grp = is_src ? &port->c_src : &port->c_dest;
+	list = is_src ? &subs->src_list : &subs->dest_list;
 	down_write(&grp->list_mutex);
 	write_lock_irq(&grp->list_lock);
-	if (is_src)
-		list_del(&subs->src_list);
-	else
-		list_del(&subs->dest_list);
+	empty = list_empty(list);
+	if (!empty)
+		list_del_init(list);
 	grp->exclusive = 0;
 	write_unlock_irq(&grp->list_lock);
 	up_write(&grp->list_mutex);
 
-	unsubscribe_port(client, port, grp, &subs->info, ack);
+	if (!empty)
+		unsubscribe_port(client, port, grp, &subs->info, ack);
 }
 
 /* connect two ports */
diff --git a/sound/core/timer.c b/sound/core/timer.c
index 9b513a05765a..dca817fc7894 100644
--- a/sound/core/timer.c
+++ b/sound/core/timer.c
@@ -422,7 +422,7 @@ static void snd_timer_notify1(struct snd_timer_instance *ti, int event)
 	spin_lock_irqsave(&timer->lock, flags);
 	list_for_each_entry(ts, &ti->slave_active_head, active_list)
 		if (ts->ccallback)
-			ts->ccallback(ti, event + 100, &tstamp, resolution);
+			ts->ccallback(ts, event + 100, &tstamp, resolution);
 	spin_unlock_irqrestore(&timer->lock, flags);
 }
 
@@ -518,9 +518,13 @@ static int _snd_timer_stop(struct snd_timer_instance *timeri, int event)
 			spin_unlock_irqrestore(&slave_active_lock, flags);
 			return -EBUSY;
 		}
+		if (timeri->timer)
+			spin_lock(&timeri->timer->lock);
 		timeri->flags &= ~SNDRV_TIMER_IFLG_RUNNING;
 		list_del_init(&timeri->ack_list);
 		list_del_init(&timeri->active_list);
+		if (timeri->timer)
+			spin_unlock(&timeri->timer->lock);
 		spin_unlock_irqrestore(&slave_active_lock, flags);
 		goto __end;
 	}
@@ -1929,6 +1933,7 @@ static ssize_t snd_timer_user_read(struct file *file, char __user *buffer,
 {
 	struct snd_timer_user *tu;
 	long result = 0, unit;
+	int qhead;
 	int err = 0;
 
 	tu = file->private_data;
@@ -1940,7 +1945,7 @@ static ssize_t snd_timer_user_read(struct file *file, char __user *buffer,
 
 			if ((file->f_flags & O_NONBLOCK) != 0 || result > 0) {
 				err = -EAGAIN;
-				break;
+				goto _error;
 			}
 
 			set_current_state(TASK_INTERRUPTIBLE);
@@ -1955,42 +1960,37 @@ static ssize_t snd_timer_user_read(struct file *file, char __user *buffer,
 
 			if (tu->disconnected) {
 				err = -ENODEV;
-				break;
+				goto _error;
 			}
 			if (signal_pending(current)) {
 				err = -ERESTARTSYS;
-				break;
+				goto _error;
 			}
 		}
 
+		qhead = tu->qhead++;
+		tu->qhead %= tu->queue_size;
 		spin_unlock_irq(&tu->qlock);
-		if (err < 0)
-			goto _error;
 
 		if (tu->tread) {
-			if (copy_to_user(buffer, &tu->tqueue[tu->qhead++],
-					 sizeof(struct snd_timer_tread))) {
+			if (copy_to_user(buffer, &tu->tqueue[qhead],
+					 sizeof(struct snd_timer_tread)))
 				err = -EFAULT;
-				goto _error;
-			}
 		} else {
-			if (copy_to_user(buffer, &tu->queue[tu->qhead++],
-					 sizeof(struct snd_timer_read))) {
+			if (copy_to_user(buffer, &tu->queue[qhead],
+					 sizeof(struct snd_timer_read)))
 				err = -EFAULT;
-				goto _error;
-			}
 		}
 
-		tu->qhead %= tu->queue_size;
-
-		result += unit;
-		buffer += unit;
-
 		spin_lock_irq(&tu->qlock);
 		tu->qused--;
+		if (err < 0)
+			goto _error;
+		result += unit;
+		buffer += unit;
 	}
-	spin_unlock_irq(&tu->qlock);
  _error:
+	spin_unlock_irq(&tu->qlock);
 	return result > 0 ? result : err;
 }
 
diff --git a/sound/core/timer_compat.c b/sound/core/timer_compat.c
index e05802ae6e1b..2e908225d754 100644
--- a/sound/core/timer_compat.c
+++ b/sound/core/timer_compat.c
@@ -70,13 +70,14 @@ static int snd_timer_user_status_compat(struct file *file,
 					struct snd_timer_status32 __user *_status)
 {
 	struct snd_timer_user *tu;
-	struct snd_timer_status status;
+	struct snd_timer_status32 status;
 	
 	tu = file->private_data;
 	if (snd_BUG_ON(!tu->timeri))
 		return -ENXIO;
 	memset(&status, 0, sizeof(status));
-	status.tstamp = tu->tstamp;
+	status.tstamp.tv_sec = tu->tstamp.tv_sec;
+	status.tstamp.tv_nsec = tu->tstamp.tv_nsec;
 	status.resolution = snd_timer_resolution(tu->timeri);
 	status.lost = tu->timeri->lost;
 	status.overrun = tu->overrun;
@@ -88,12 +89,21 @@ static int snd_timer_user_status_compat(struct file *file,
 	return 0;
 }
 
+#ifdef CONFIG_X86_X32
+/* X32 ABI has the same struct as x86-64 */
+#define snd_timer_user_status_x32(file, s) \
+	snd_timer_user_status(file, s)
+#endif /* CONFIG_X86_X32 */
+
 /*
  */
 
 enum {
 	SNDRV_TIMER_IOCTL_INFO32 = _IOR('T', 0x11, struct snd_timer_info32),
 	SNDRV_TIMER_IOCTL_STATUS32 = _IOW('T', 0x14, struct snd_timer_status32),
+#ifdef CONFIG_X86_X32
+	SNDRV_TIMER_IOCTL_STATUS_X32 = _IOW('T', 0x14, struct snd_timer_status),
+#endif /* CONFIG_X86_X32 */
 };
 
 static long snd_timer_user_ioctl_compat(struct file *file, unsigned int cmd, unsigned long arg)
@@ -122,6 +132,10 @@ static long snd_timer_user_ioctl_compat(struct file *file, unsigned int cmd, uns
 		return snd_timer_user_info_compat(file, argp);
 	case SNDRV_TIMER_IOCTL_STATUS32:
 		return snd_timer_user_status_compat(file, argp);
+#ifdef CONFIG_X86_X32
+	case SNDRV_TIMER_IOCTL_STATUS_X32:
+		return snd_timer_user_status_x32(file, argp);
+#endif /* CONFIG_X86_X32 */
 	}
 	return -ENOIOCTLCMD;
 }
diff --git a/sound/drivers/dummy.c b/sound/drivers/dummy.c
index bde33308f0d6..c0f8f613f1f1 100644
--- a/sound/drivers/dummy.c
+++ b/sound/drivers/dummy.c
@@ -87,7 +87,7 @@ MODULE_PARM_DESC(pcm_substreams, "PCM substreams # (1-128) for dummy driver.");
 module_param(fake_buffer, bool, 0444);
 MODULE_PARM_DESC(fake_buffer, "Fake buffer allocations.");
 #ifdef CONFIG_HIGH_RES_TIMERS
-module_param(hrtimer, bool, 0444);
+module_param(hrtimer, bool, 0644);
 MODULE_PARM_DESC(hrtimer, "Use hrtimer as the timer source.");
 #endif
 
@@ -109,6 +109,9 @@ struct dummy_timer_ops {
 	snd_pcm_uframes_t (*pointer)(struct snd_pcm_substream *);
 };
 
+#define get_dummy_ops(substream) \
+	(*(const struct dummy_timer_ops **)(substream)->runtime->private_data)
+
 struct dummy_model {
 	const char *name;
 	int (*playback_constraints)(struct snd_pcm_runtime *runtime);
@@ -137,7 +140,6 @@ struct snd_dummy {
 	int iobox;
 	struct snd_kcontrol *cd_volume_ctl;
 	struct snd_kcontrol *cd_switch_ctl;
-	const struct dummy_timer_ops *timer_ops;
 };
 
 /*
@@ -231,6 +233,8 @@ static struct dummy_model *dummy_models[] = {
  */
 
 struct dummy_systimer_pcm {
+	/* ops must be the first item */
+	const struct dummy_timer_ops *timer_ops;
 	spinlock_t lock;
 	struct timer_list timer;
 	unsigned long base_time;
@@ -366,6 +370,8 @@ static const struct dummy_timer_ops dummy_systimer_ops = {
  */
 
 struct dummy_hrtimer_pcm {
+	/* ops must be the first item */
+	const struct dummy_timer_ops *timer_ops;
 	ktime_t base_time;
 	ktime_t period_time;
 	atomic_t running;
@@ -492,31 +498,25 @@ static const struct dummy_timer_ops dummy_hrtimer_ops = {
 
 static int dummy_pcm_trigger(struct snd_pcm_substream *substream, int cmd)
 {
-	struct snd_dummy *dummy = snd_pcm_substream_chip(substream);
-
 	switch (cmd) {
 	case SNDRV_PCM_TRIGGER_START:
 	case SNDRV_PCM_TRIGGER_RESUME:
-		return dummy->timer_ops->start(substream);
+		return get_dummy_ops(substream)->start(substream);
 	case SNDRV_PCM_TRIGGER_STOP:
 	case SNDRV_PCM_TRIGGER_SUSPEND:
-		return dummy->timer_ops->stop(substream);
+		return get_dummy_ops(substream)->stop(substream);
 	}
 	return -EINVAL;
 }
 
 static int dummy_pcm_prepare(struct snd_pcm_substream *substream)
 {
-	struct snd_dummy *dummy = snd_pcm_substream_chip(substream);
-
-	return dummy->timer_ops->prepare(substream);
+	return get_dummy_ops(substream)->prepare(substream);
 }
 
 static snd_pcm_uframes_t dummy_pcm_pointer(struct snd_pcm_substream *substream)
 {
-	struct snd_dummy *dummy = snd_pcm_substream_chip(substream);
-
-	return dummy->timer_ops->pointer(substream);
+	return get_dummy_ops(substream)->pointer(substream);
 }
 
 static struct snd_pcm_hardware dummy_pcm_hardware = {
@@ -562,17 +562,19 @@ static int dummy_pcm_open(struct snd_pcm_substream *substream)
 	struct snd_dummy *dummy = snd_pcm_substream_chip(substream);
 	struct dummy_model *model = dummy->model;
 	struct snd_pcm_runtime *runtime = substream->runtime;
+	const struct dummy_timer_ops *ops;
 	int err;
 
-	dummy->timer_ops = &dummy_systimer_ops;
+	ops = &dummy_systimer_ops;
 #ifdef CONFIG_HIGH_RES_TIMERS
 	if (hrtimer)
-		dummy->timer_ops = &dummy_hrtimer_ops;
+		ops = &dummy_hrtimer_ops;
 #endif
 
-	err = dummy->timer_ops->create(substream);
+	err = ops->create(substream);
 	if (err < 0)
 		return err;
+	get_dummy_ops(substream) = ops;
 
 	runtime->hw = dummy->pcm_hw;
 	if (substream->pcm->device & 1) {
@@ -594,7 +596,7 @@ static int dummy_pcm_open(struct snd_pcm_substream *substream)
 			err = model->capture_constraints(substream->runtime);
 	}
 	if (err < 0) {
-		dummy->timer_ops->free(substream);
+		get_dummy_ops(substream)->free(substream);
 		return err;
 	}
 	return 0;
@@ -602,8 +604,7 @@ static int dummy_pcm_open(struct snd_pcm_substream *substream)
 
 static int dummy_pcm_close(struct snd_pcm_substream *substream)
 {
-	struct snd_dummy *dummy = snd_pcm_substream_chip(substream);
-	dummy->timer_ops->free(substream);
+	get_dummy_ops(substream)->free(substream);
 	return 0;
 }
 
diff --git a/sound/firewire/digi00x/amdtp-dot.c b/sound/firewire/digi00x/amdtp-dot.c
index b02a5e8cad44..0ac92aba5bc1 100644
--- a/sound/firewire/digi00x/amdtp-dot.c
+++ b/sound/firewire/digi00x/amdtp-dot.c
@@ -63,7 +63,7 @@ struct amdtp_dot {
 #define BYTE_PER_SAMPLE (4)
 #define MAGIC_DOT_BYTE (2)
 #define MAGIC_BYTE_OFF(x) (((x) * BYTE_PER_SAMPLE) + MAGIC_DOT_BYTE)
-static const u8 dot_scrt(const u8 idx, const unsigned int off)
+static u8 dot_scrt(const u8 idx, const unsigned int off)
 {
 	/*
 	 * the length of the added pattern only depends on the lower nibble
diff --git a/sound/firewire/tascam/tascam-transaction.c b/sound/firewire/tascam/tascam-transaction.c
index 904ce0329fa1..040a96d1ba8e 100644
--- a/sound/firewire/tascam/tascam-transaction.c
+++ b/sound/firewire/tascam/tascam-transaction.c
@@ -230,6 +230,7 @@ int snd_tscm_transaction_register(struct snd_tscm *tscm)
 	return err;
 error:
 	fw_core_remove_address_handler(&tscm->async_handler);
+	tscm->async_handler.callback_data = NULL;
 	return err;
 }
 
@@ -276,6 +277,9 @@ void snd_tscm_transaction_unregister(struct snd_tscm *tscm)
 	__be32 reg;
 	unsigned int i;
 
+	if (tscm->async_handler.callback_data == NULL)
+		return;
+
 	/* Turn off FireWire LED. */
 	reg = cpu_to_be32(0x0000008e);
 	snd_fw_transaction(tscm->unit, TCODE_WRITE_QUADLET_REQUEST,
@@ -297,6 +301,8 @@ void snd_tscm_transaction_unregister(struct snd_tscm *tscm)
 			   &reg, sizeof(reg), 0);
 
 	fw_core_remove_address_handler(&tscm->async_handler);
+	tscm->async_handler.callback_data = NULL;
+
 	for (i = 0; i < TSCM_MIDI_OUT_PORT_MAX; i++)
 		snd_fw_async_midi_port_destroy(&tscm->out_ports[i]);
 }
diff --git a/sound/firewire/tascam/tascam.c b/sound/firewire/tascam/tascam.c
index ee0bc1839508..e281c338e562 100644
--- a/sound/firewire/tascam/tascam.c
+++ b/sound/firewire/tascam/tascam.c
@@ -21,7 +21,6 @@ static struct snd_tscm_spec model_specs[] = {
 		.pcm_playback_analog_channels = 8,
 		.midi_capture_ports = 4,
 		.midi_playback_ports = 4,
-		.is_controller = true,
 	},
 	{
 		.name = "FW-1082",
@@ -31,9 +30,16 @@ static struct snd_tscm_spec model_specs[] = {
 		.pcm_playback_analog_channels = 2,
 		.midi_capture_ports = 2,
 		.midi_playback_ports = 2,
-		.is_controller = true,
 	},
-	/* FW-1804 may be supported. */
+	{
+		.name = "FW-1804",
+		.has_adat = true,
+		.has_spdif = true,
+		.pcm_capture_analog_channels = 8,
+		.pcm_playback_analog_channels = 2,
+		.midi_capture_ports = 2,
+		.midi_playback_ports = 4,
+	},
 };
 
 static int identify_model(struct snd_tscm *tscm)
diff --git a/sound/firewire/tascam/tascam.h b/sound/firewire/tascam/tascam.h
index 2d028d2bd3bd..30ab77e924f7 100644
--- a/sound/firewire/tascam/tascam.h
+++ b/sound/firewire/tascam/tascam.h
@@ -39,7 +39,6 @@ struct snd_tscm_spec {
 	unsigned int pcm_playback_analog_channels;
 	unsigned int midi_capture_ports;
 	unsigned int midi_playback_ports;
-	bool is_controller;
 };
 
 #define TSCM_MIDI_IN_PORT_MAX	4
@@ -72,9 +71,6 @@ struct snd_tscm {
 	struct snd_fw_async_midi_port out_ports[TSCM_MIDI_OUT_PORT_MAX];
 	u8 running_status[TSCM_MIDI_OUT_PORT_MAX];
 	bool on_sysex[TSCM_MIDI_OUT_PORT_MAX];
-
-	/* For control messages. */
-	struct snd_firewire_tascam_status *status;
 };
 
 #define TSCM_ADDR_BASE			0xffff00000000ull
diff --git a/sound/hda/hdac_controller.c b/sound/hda/hdac_controller.c
index b5a17cb510a0..8c486235c905 100644
--- a/sound/hda/hdac_controller.c
+++ b/sound/hda/hdac_controller.c
@@ -426,18 +426,22 @@ EXPORT_SYMBOL_GPL(snd_hdac_bus_stop_chip);
  * @bus: HD-audio core bus
  * @status: INTSTS register value
  * @ask: callback to be called for woken streams
+ *
+ * Returns the bits of handled streams, or zero if no stream is handled.
  */
-void snd_hdac_bus_handle_stream_irq(struct hdac_bus *bus, unsigned int status,
+int snd_hdac_bus_handle_stream_irq(struct hdac_bus *bus, unsigned int status,
 				    void (*ack)(struct hdac_bus *,
 						struct hdac_stream *))
 {
 	struct hdac_stream *azx_dev;
 	u8 sd_status;
+	int handled = 0;
 
 	list_for_each_entry(azx_dev, &bus->stream_list, list) {
 		if (status & azx_dev->sd_int_sta_mask) {
 			sd_status = snd_hdac_stream_readb(azx_dev, SD_STS);
 			snd_hdac_stream_writeb(azx_dev, SD_STS, SD_INT_MASK);
+			handled |= 1 << azx_dev->index;
 			if (!azx_dev->substream || !azx_dev->running ||
 			    !(sd_status & SD_INT_COMPLETE))
 				continue;
@@ -445,6 +449,7 @@ void snd_hdac_bus_handle_stream_irq(struct hdac_bus *bus, unsigned int status,
 				ack(bus, azx_dev);
 		}
 	}
+	return handled;
 }
 EXPORT_SYMBOL_GPL(snd_hdac_bus_handle_stream_irq);
 
diff --git a/sound/pci/hda/hda_controller.c b/sound/pci/hda/hda_controller.c
index 37cf9cee9835..27de8015717d 100644
--- a/sound/pci/hda/hda_controller.c
+++ b/sound/pci/hda/hda_controller.c
@@ -930,6 +930,8 @@ irqreturn_t azx_interrupt(int irq, void *dev_id)
 	struct azx *chip = dev_id;
 	struct hdac_bus *bus = azx_bus(chip);
 	u32 status;
+	bool active, handled = false;
+	int repeat = 0; /* count for avoiding endless loop */
 
 #ifdef CONFIG_PM
 	if (azx_has_pm_runtime(chip))
@@ -939,33 +941,36 @@ irqreturn_t azx_interrupt(int irq, void *dev_id)
 
 	spin_lock(&bus->reg_lock);
 
-	if (chip->disabled) {
-		spin_unlock(&bus->reg_lock);
-		return IRQ_NONE;
-	}
-
-	status = azx_readl(chip, INTSTS);
-	if (status == 0 || status == 0xffffffff) {
-		spin_unlock(&bus->reg_lock);
-		return IRQ_NONE;
-	}
+	if (chip->disabled)
+		goto unlock;
 
-	snd_hdac_bus_handle_stream_irq(bus, status, stream_update);
+	do {
+		status = azx_readl(chip, INTSTS);
+		if (status == 0 || status == 0xffffffff)
+			break;
 
-	/* clear rirb int */
-	status = azx_readb(chip, RIRBSTS);
-	if (status & RIRB_INT_MASK) {
-		if (status & RIRB_INT_RESPONSE) {
-			if (chip->driver_caps & AZX_DCAPS_CTX_WORKAROUND)
-				udelay(80);
-			snd_hdac_bus_update_rirb(bus);
+		handled = true;
+		active = false;
+		if (snd_hdac_bus_handle_stream_irq(bus, status, stream_update))
+			active = true;
+
+		/* clear rirb int */
+		status = azx_readb(chip, RIRBSTS);
+		if (status & RIRB_INT_MASK) {
+			active = true;
+			if (status & RIRB_INT_RESPONSE) {
+				if (chip->driver_caps & AZX_DCAPS_CTX_WORKAROUND)
+					udelay(80);
+				snd_hdac_bus_update_rirb(bus);
+			}
+			azx_writeb(chip, RIRBSTS, RIRB_INT_MASK);
 		}
-		azx_writeb(chip, RIRBSTS, RIRB_INT_MASK);
-	}
+	} while (active && ++repeat < 10);
 
+ unlock:
 	spin_unlock(&bus->reg_lock);
 
-	return IRQ_HANDLED;
+	return IRQ_RETVAL(handled);
 }
 EXPORT_SYMBOL_GPL(azx_interrupt);
 
diff --git a/sound/pci/hda/hda_generic.c b/sound/pci/hda/hda_generic.c
index 30c8efe0f80a..7ca5b89f088a 100644
--- a/sound/pci/hda/hda_generic.c
+++ b/sound/pci/hda/hda_generic.c
@@ -4028,9 +4028,9 @@ static void pin_power_callback(struct hda_codec *codec,
 			       struct hda_jack_callback *jack,
 			       bool on)
 {
-	if (jack && jack->tbl->nid)
+	if (jack && jack->nid)
 		sync_power_state_change(codec,
-					set_pin_power_jack(codec, jack->tbl->nid, on));
+					set_pin_power_jack(codec, jack->nid, on));
 }
 
 /* callback only doing power up -- called at first */
diff --git a/sound/pci/hda/hda_intel.c b/sound/pci/hda/hda_intel.c
index 4045dca3d699..e5240cb3749f 100644
--- a/sound/pci/hda/hda_intel.c
+++ b/sound/pci/hda/hda_intel.c
@@ -363,7 +363,10 @@ enum {
 					((pci)->device == 0x0d0c) || \
 					((pci)->device == 0x160c))
 
-#define IS_BROXTON(pci)	((pci)->device == 0x5a98)
+#define IS_SKL(pci) ((pci)->vendor == 0x8086 && (pci)->device == 0xa170)
+#define IS_SKL_LP(pci) ((pci)->vendor == 0x8086 && (pci)->device == 0x9d70)
+#define IS_BXT(pci) ((pci)->vendor == 0x8086 && (pci)->device == 0x5a98)
+#define IS_SKL_PLUS(pci) (IS_SKL(pci) || IS_SKL_LP(pci) || IS_BXT(pci))
 
 static char *driver_short_names[] = {
 	[AZX_DRIVER_ICH] = "HDA Intel",
@@ -540,13 +543,13 @@ static void hda_intel_init_chip(struct azx *chip, bool full_reset)
 
 	if (chip->driver_caps & AZX_DCAPS_I915_POWERWELL)
 		snd_hdac_set_codec_wakeup(bus, true);
-	if (IS_BROXTON(pci)) {
+	if (IS_SKL_PLUS(pci)) {
 		pci_read_config_dword(pci, INTEL_HDA_CGCTL, &val);
 		val = val & ~INTEL_HDA_CGCTL_MISCBDCGE;
 		pci_write_config_dword(pci, INTEL_HDA_CGCTL, val);
 	}
 	azx_init_chip(chip, full_reset);
-	if (IS_BROXTON(pci)) {
+	if (IS_SKL_PLUS(pci)) {
 		pci_read_config_dword(pci, INTEL_HDA_CGCTL, &val);
 		val = val | INTEL_HDA_CGCTL_MISCBDCGE;
 		pci_write_config_dword(pci, INTEL_HDA_CGCTL, val);
@@ -555,7 +558,7 @@ static void hda_intel_init_chip(struct azx *chip, bool full_reset)
 		snd_hdac_set_codec_wakeup(bus, false);
 
 	/* reduce dma latency to avoid noise */
-	if (IS_BROXTON(pci))
+	if (IS_BXT(pci))
 		bxt_reduce_dma_latency(chip);
 }
 
@@ -977,11 +980,6 @@ static int azx_resume(struct device *dev)
 /* put codec down to D3 at hibernation for Intel SKL+;
  * otherwise BIOS may still access the codec and screw up the driver
  */
-#define IS_SKL(pci) ((pci)->vendor == 0x8086 && (pci)->device == 0xa170)
-#define IS_SKL_LP(pci) ((pci)->vendor == 0x8086 && (pci)->device == 0x9d70)
-#define IS_BXT(pci) ((pci)->vendor == 0x8086 && (pci)->device == 0x5a98)
-#define IS_SKL_PLUS(pci) (IS_SKL(pci) || IS_SKL_LP(pci) || IS_BXT(pci))
-
 static int azx_freeze_noirq(struct device *dev)
 {
 	struct pci_dev *pci = to_pci_dev(dev);
@@ -2168,10 +2166,10 @@ static void azx_remove(struct pci_dev *pci)
 	struct hda_intel *hda;
 
 	if (card) {
-		/* flush the pending probing work */
+		/* cancel the pending probing work */
 		chip = card->private_data;
 		hda = container_of(chip, struct hda_intel, chip);
-		flush_work(&hda->probe_work);
+		cancel_work_sync(&hda->probe_work);
 
 		snd_card_free(card);
 	}
diff --git a/sound/pci/hda/hda_jack.c b/sound/pci/hda/hda_jack.c
index c945e257d368..a33234e04d4f 100644
--- a/sound/pci/hda/hda_jack.c
+++ b/sound/pci/hda/hda_jack.c
@@ -259,7 +259,7 @@ snd_hda_jack_detect_enable_callback(struct hda_codec *codec, hda_nid_t nid,
 		if (!callback)
 			return ERR_PTR(-ENOMEM);
 		callback->func = func;
-		callback->tbl = jack;
+		callback->nid = jack->nid;
 		callback->next = jack->callback;
 		jack->callback = callback;
 	}
diff --git a/sound/pci/hda/hda_jack.h b/sound/pci/hda/hda_jack.h
index 858708a044f5..e9814c0168ea 100644
--- a/sound/pci/hda/hda_jack.h
+++ b/sound/pci/hda/hda_jack.h
@@ -21,7 +21,7 @@ struct hda_jack_callback;
 typedef void (*hda_jack_callback_fn) (struct hda_codec *, struct hda_jack_callback *);
 
 struct hda_jack_callback {
-	struct hda_jack_tbl *tbl;
+	hda_nid_t nid;
 	hda_jack_callback_fn func;
 	unsigned int private_data;	/* arbitrary data */
 	struct hda_jack_callback *next;
diff --git a/sound/pci/hda/patch_ca0132.c b/sound/pci/hda/patch_ca0132.c
index 4ef2259f88ca..9ceb2bc36e68 100644
--- a/sound/pci/hda/patch_ca0132.c
+++ b/sound/pci/hda/patch_ca0132.c
@@ -4427,13 +4427,16 @@ static void ca0132_process_dsp_response(struct hda_codec *codec,
 static void hp_callback(struct hda_codec *codec, struct hda_jack_callback *cb)
 {
 	struct ca0132_spec *spec = codec->spec;
+	struct hda_jack_tbl *tbl;
 
 	/* Delay enabling the HP amp, to let the mic-detection
 	 * state machine run.
 	 */
 	cancel_delayed_work_sync(&spec->unsol_hp_work);
 	schedule_delayed_work(&spec->unsol_hp_work, msecs_to_jiffies(500));
-	cb->tbl->block_report = 1;
+	tbl = snd_hda_jack_tbl_get(codec, cb->nid);
+	if (tbl)
+		tbl->block_report = 1;
 }
 
 static void amic_callback(struct hda_codec *codec, struct hda_jack_callback *cb)
diff --git a/sound/pci/hda/patch_hdmi.c b/sound/pci/hda/patch_hdmi.c
index 1f52b55d77c9..bcbc4ee10130 100644
--- a/sound/pci/hda/patch_hdmi.c
+++ b/sound/pci/hda/patch_hdmi.c
@@ -448,7 +448,8 @@ static int hdmi_eld_ctl_get(struct snd_kcontrol *kcontrol,
 	eld = &per_pin->sink_eld;
 
 	mutex_lock(&per_pin->lock);
-	if (eld->eld_size > ARRAY_SIZE(ucontrol->value.bytes.data)) {
+	if (eld->eld_size > ARRAY_SIZE(ucontrol->value.bytes.data) ||
+	    eld->eld_size > ELD_MAX_SIZE) {
 		mutex_unlock(&per_pin->lock);
 		snd_BUG();
 		return -EINVAL;
@@ -1193,7 +1194,7 @@ static void check_presence_and_report(struct hda_codec *codec, hda_nid_t nid)
 static void jack_callback(struct hda_codec *codec,
 			  struct hda_jack_callback *jack)
 {
-	check_presence_and_report(codec, jack->tbl->nid);
+	check_presence_and_report(codec, jack->nid);
 }
 
 static void hdmi_intrinsic_event(struct hda_codec *codec, unsigned int res)
@@ -2476,13 +2477,6 @@ static int patch_generic_hdmi(struct hda_codec *codec)
 			is_broxton(codec))
 		codec->core.link_power_control = 1;
 
-	if (codec_has_acomp(codec)) {
-		codec->depop_delay = 0;
-		spec->i915_audio_ops.audio_ptr = codec;
-		spec->i915_audio_ops.pin_eld_notify = intel_pin_eld_notify;
-		snd_hdac_i915_register_notifier(&spec->i915_audio_ops);
-	}
-
 	if (hdmi_parse_codec(codec) < 0) {
 		if (spec->i915_bound)
 			snd_hdac_i915_exit(&codec->bus->core);
@@ -2504,6 +2498,18 @@ static int patch_generic_hdmi(struct hda_codec *codec)
 
 	init_channel_allocations();
 
+	if (codec_has_acomp(codec)) {
+		codec->depop_delay = 0;
+		spec->i915_audio_ops.audio_ptr = codec;
+		/* intel_audio_codec_enable() or intel_audio_codec_disable()
+		 * will call pin_eld_notify with using audio_ptr pointer
+		 * We need make sure audio_ptr is really setup
+		 */
+		wmb();
+		spec->i915_audio_ops.pin_eld_notify = intel_pin_eld_notify;
+		snd_hdac_i915_register_notifier(&spec->i915_audio_ops);
+	}
+
 	return 0;
 }
 
diff --git a/sound/pci/hda/patch_realtek.c b/sound/pci/hda/patch_realtek.c
index 21992fb7035d..93d2156b6241 100644
--- a/sound/pci/hda/patch_realtek.c
+++ b/sound/pci/hda/patch_realtek.c
@@ -282,7 +282,7 @@ static void alc_update_knob_master(struct hda_codec *codec,
 	uctl = kzalloc(sizeof(*uctl), GFP_KERNEL);
 	if (!uctl)
 		return;
-	val = snd_hda_codec_read(codec, jack->tbl->nid, 0,
+	val = snd_hda_codec_read(codec, jack->nid, 0,
 				 AC_VERB_GET_VOLUME_KNOB_CONTROL, 0);
 	val &= HDA_AMP_VOLMASK;
 	uctl->value.integer.value[0] = val;
@@ -1787,7 +1787,6 @@ enum {
 	ALC882_FIXUP_NO_PRIMARY_HP,
 	ALC887_FIXUP_ASUS_BASS,
 	ALC887_FIXUP_BASS_CHMAP,
-	ALC882_FIXUP_DISABLE_AAMIX,
 };
 
 static void alc889_fixup_coef(struct hda_codec *codec,
@@ -1949,8 +1948,6 @@ static void alc882_fixup_no_primary_hp(struct hda_codec *codec,
 
 static void alc_fixup_bass_chmap(struct hda_codec *codec,
 				 const struct hda_fixup *fix, int action);
-static void alc_fixup_disable_aamix(struct hda_codec *codec,
-				    const struct hda_fixup *fix, int action);
 
 static const struct hda_fixup alc882_fixups[] = {
 	[ALC882_FIXUP_ABIT_AW9D_MAX] = {
@@ -2188,10 +2185,6 @@ static const struct hda_fixup alc882_fixups[] = {
 		.type = HDA_FIXUP_FUNC,
 		.v.func = alc_fixup_bass_chmap,
 	},
-	[ALC882_FIXUP_DISABLE_AAMIX] = {
-		.type = HDA_FIXUP_FUNC,
-		.v.func = alc_fixup_disable_aamix,
-	},
 };
 
 static const struct snd_pci_quirk alc882_fixup_tbl[] = {
@@ -2230,6 +2223,7 @@ static const struct snd_pci_quirk alc882_fixup_tbl[] = {
 	SND_PCI_QUIRK(0x104d, 0x9047, "Sony Vaio TT", ALC889_FIXUP_VAIO_TT),
 	SND_PCI_QUIRK(0x104d, 0x905a, "Sony Vaio Z", ALC882_FIXUP_NO_PRIMARY_HP),
 	SND_PCI_QUIRK(0x104d, 0x9043, "Sony Vaio VGC-LN51JGB", ALC882_FIXUP_NO_PRIMARY_HP),
+	SND_PCI_QUIRK(0x104d, 0x9044, "Sony VAIO AiO", ALC882_FIXUP_NO_PRIMARY_HP),
 
 	/* All Apple entries are in codec SSIDs */
 	SND_PCI_QUIRK(0x106b, 0x00a0, "MacBookPro 3,1", ALC889_FIXUP_MBP_VREF),
@@ -2259,7 +2253,6 @@ static const struct snd_pci_quirk alc882_fixup_tbl[] = {
 	SND_PCI_QUIRK(0x1462, 0x7350, "MSI-7350", ALC889_FIXUP_CD),
 	SND_PCI_QUIRK_VENDOR(0x1462, "MSI", ALC882_FIXUP_GPIO3),
 	SND_PCI_QUIRK(0x1458, 0xa002, "Gigabyte EP45-DS3/Z87X-UD3H", ALC889_FIXUP_FRONT_HP_NO_PRESENCE),
-	SND_PCI_QUIRK(0x1458, 0xa182, "Gigabyte Z170X-UD3", ALC882_FIXUP_DISABLE_AAMIX),
 	SND_PCI_QUIRK(0x147b, 0x107a, "Abit AW9D-MAX", ALC882_FIXUP_ABIT_AW9D_MAX),
 	SND_PCI_QUIRK_VENDOR(0x1558, "Clevo laptop", ALC882_FIXUP_EAPD),
 	SND_PCI_QUIRK(0x161f, 0x2054, "Medion laptop", ALC883_FIXUP_EAPD),
@@ -3808,6 +3801,10 @@ static void alc_headset_mode_mic_in(struct hda_codec *codec, hda_nid_t hp_pin,
 
 static void alc_headset_mode_default(struct hda_codec *codec)
 {
+	static struct coef_fw coef0225[] = {
+		UPDATE_COEF(0x45, 0x3f<<10, 0x34<<10),
+		{}
+	};
 	static struct coef_fw coef0255[] = {
 		WRITE_COEF(0x45, 0xc089),
 		WRITE_COEF(0x45, 0xc489),
@@ -3849,6 +3846,9 @@ static void alc_headset_mode_default(struct hda_codec *codec)
 	};
 
 	switch (codec->core.vendor_id) {
+	case 0x10ec0225:
+		alc_process_coef_fw(codec, coef0225);
+		break;
 	case 0x10ec0255:
 	case 0x10ec0256:
 		alc_process_coef_fw(codec, coef0255);
@@ -4756,6 +4756,9 @@ enum {
 	ALC256_FIXUP_DELL_XPS_13_HEADPHONE_NOISE,
 	ALC293_FIXUP_LENOVO_SPK_NOISE,
 	ALC233_FIXUP_LENOVO_LINE2_MIC_HOTKEY,
+	ALC255_FIXUP_DELL_SPK_NOISE,
+	ALC225_FIXUP_DELL1_MIC_NO_PRESENCE,
+	ALC280_FIXUP_HP_HEADSET_MIC,
 };
 
 static const struct hda_fixup alc269_fixups[] = {
@@ -5375,6 +5378,29 @@ static const struct hda_fixup alc269_fixups[] = {
 		.type = HDA_FIXUP_FUNC,
 		.v.func = alc233_fixup_lenovo_line2_mic_hotkey,
 	},
+	[ALC255_FIXUP_DELL_SPK_NOISE] = {
+		.type = HDA_FIXUP_FUNC,
+		.v.func = alc_fixup_disable_aamix,
+		.chained = true,
+		.chain_id = ALC255_FIXUP_DELL1_MIC_NO_PRESENCE
+	},
+	[ALC225_FIXUP_DELL1_MIC_NO_PRESENCE] = {
+		.type = HDA_FIXUP_VERBS,
+		.v.verbs = (const struct hda_verb[]) {
+			/* Disable pass-through path for FRONT 14h */
+			{ 0x20, AC_VERB_SET_COEF_INDEX, 0x36 },
+			{ 0x20, AC_VERB_SET_PROC_COEF, 0x57d7 },
+			{}
+		},
+		.chained = true,
+		.chain_id = ALC269_FIXUP_DELL1_MIC_NO_PRESENCE
+	},
+	[ALC280_FIXUP_HP_HEADSET_MIC] = {
+		.type = HDA_FIXUP_FUNC,
+		.v.func = alc_fixup_disable_aamix,
+		.chained = true,
+		.chain_id = ALC269_FIXUP_HEADSET_MIC,
+	},
 };
 
 static const struct snd_pci_quirk alc269_fixup_tbl[] = {
@@ -5386,6 +5412,7 @@ static const struct snd_pci_quirk alc269_fixup_tbl[] = {
 	SND_PCI_QUIRK(0x1025, 0x080d, "Acer Aspire V5-122P", ALC269_FIXUP_ASPIRE_HEADSET_MIC),
 	SND_PCI_QUIRK(0x1025, 0x0740, "Acer AO725", ALC271_FIXUP_HP_GATE_MIC_JACK),
 	SND_PCI_QUIRK(0x1025, 0x0742, "Acer AO756", ALC271_FIXUP_HP_GATE_MIC_JACK),
+	SND_PCI_QUIRK(0x1025, 0x0762, "Acer Aspire E1-472", ALC271_FIXUP_HP_GATE_MIC_JACK_E1_572),
 	SND_PCI_QUIRK(0x1025, 0x0775, "Acer Aspire E1-572", ALC271_FIXUP_HP_GATE_MIC_JACK_E1_572),
 	SND_PCI_QUIRK(0x1025, 0x079b, "Acer Aspire V5-573G", ALC282_FIXUP_ASPIRE_V5_PINS),
 	SND_PCI_QUIRK(0x1025, 0x106d, "Acer Cloudbook 14", ALC283_FIXUP_CHROME_BOOK),
@@ -5417,6 +5444,7 @@ static const struct snd_pci_quirk alc269_fixup_tbl[] = {
 	SND_PCI_QUIRK(0x1028, 0x06df, "Dell", ALC293_FIXUP_DISABLE_AAMIX_MULTIJACK),
 	SND_PCI_QUIRK(0x1028, 0x06e0, "Dell", ALC293_FIXUP_DISABLE_AAMIX_MULTIJACK),
 	SND_PCI_QUIRK(0x1028, 0x0704, "Dell XPS 13", ALC256_FIXUP_DELL_XPS_13_HEADPHONE_NOISE),
+	SND_PCI_QUIRK(0x1028, 0x0725, "Dell Inspiron 3162", ALC255_FIXUP_DELL_SPK_NOISE),
 	SND_PCI_QUIRK(0x1028, 0x164a, "Dell", ALC293_FIXUP_DELL1_MIC_NO_PRESENCE),
 	SND_PCI_QUIRK(0x1028, 0x164b, "Dell", ALC293_FIXUP_DELL1_MIC_NO_PRESENCE),
 	SND_PCI_QUIRK(0x103c, 0x1586, "HP", ALC269_FIXUP_HP_MUTE_LED_MIC2),
@@ -5477,6 +5505,7 @@ static const struct snd_pci_quirk alc269_fixup_tbl[] = {
 	SND_PCI_QUIRK(0x103c, 0x2335, "HP", ALC269_FIXUP_HP_MUTE_LED_MIC1),
 	SND_PCI_QUIRK(0x103c, 0x2336, "HP", ALC269_FIXUP_HP_MUTE_LED_MIC1),
 	SND_PCI_QUIRK(0x103c, 0x2337, "HP", ALC269_FIXUP_HP_MUTE_LED_MIC1),
+	SND_PCI_QUIRK(0x103c, 0x221c, "HP EliteBook 755 G2", ALC280_FIXUP_HP_HEADSET_MIC),
 	SND_PCI_QUIRK(0x1043, 0x103f, "ASUS TX300", ALC282_FIXUP_ASUS_TX300),
 	SND_PCI_QUIRK(0x1043, 0x106d, "Asus K53BE", ALC269_FIXUP_LIMIT_INT_MIC_BOOST),
 	SND_PCI_QUIRK(0x1043, 0x115d, "Asus 1015E", ALC269_FIXUP_LIMIT_INT_MIC_BOOST),
@@ -5645,10 +5674,10 @@ static const struct hda_model_fixup alc269_fixup_models[] = {
 	{0x21, 0x03211020}
 
 static const struct snd_hda_pin_quirk alc269_pin_fixup_tbl[] = {
-	SND_HDA_PIN_QUIRK(0x10ec0225, 0x1028, "Dell", ALC269_FIXUP_DELL1_MIC_NO_PRESENCE,
+	SND_HDA_PIN_QUIRK(0x10ec0225, 0x1028, "Dell", ALC225_FIXUP_DELL1_MIC_NO_PRESENCE,
 		ALC225_STANDARD_PINS,
 		{0x14, 0x901701a0}),
-	SND_HDA_PIN_QUIRK(0x10ec0225, 0x1028, "Dell", ALC269_FIXUP_DELL1_MIC_NO_PRESENCE,
+	SND_HDA_PIN_QUIRK(0x10ec0225, 0x1028, "Dell", ALC225_FIXUP_DELL1_MIC_NO_PRESENCE,
 		ALC225_STANDARD_PINS,
 		{0x14, 0x901701b0}),
 	SND_HDA_PIN_QUIRK(0x10ec0255, 0x1028, "Dell", ALC255_FIXUP_DELL2_MIC_NO_PRESENCE,
diff --git a/sound/pci/hda/patch_sigmatel.c b/sound/pci/hda/patch_sigmatel.c
index 2c7c5eb8b1e9..37b70f8e878f 100644
--- a/sound/pci/hda/patch_sigmatel.c
+++ b/sound/pci/hda/patch_sigmatel.c
@@ -493,9 +493,9 @@ static void jack_update_power(struct hda_codec *codec,
 	if (!spec->num_pwrs)
 		return;
 
-	if (jack && jack->tbl->nid) {
-		stac_toggle_power_map(codec, jack->tbl->nid,
-				      snd_hda_jack_detect(codec, jack->tbl->nid),
+	if (jack && jack->nid) {
+		stac_toggle_power_map(codec, jack->nid,
+				      snd_hda_jack_detect(codec, jack->nid),
 				      true);
 		return;
 	}
diff --git a/sound/pci/rme9652/hdsp.c b/sound/pci/rme9652/hdsp.c
index 2875b4f6d8c9..7c8941b8b2de 100644
--- a/sound/pci/rme9652/hdsp.c
+++ b/sound/pci/rme9652/hdsp.c
@@ -2879,7 +2879,7 @@ static int snd_hdsp_get_dds_offset(struct snd_kcontrol *kcontrol, struct snd_ctl
 {
 	struct hdsp *hdsp = snd_kcontrol_chip(kcontrol);
 
-	ucontrol->value.enumerated.item[0] = hdsp_dds_offset(hdsp);
+	ucontrol->value.integer.value[0] = hdsp_dds_offset(hdsp);
 	return 0;
 }
 
@@ -2891,7 +2891,7 @@ static int snd_hdsp_put_dds_offset(struct snd_kcontrol *kcontrol, struct snd_ctl
 
 	if (!snd_hdsp_use_is_exclusive(hdsp))
 		return -EBUSY;
-	val = ucontrol->value.enumerated.item[0];
+	val = ucontrol->value.integer.value[0];
 	spin_lock_irq(&hdsp->lock);
 	if (val != hdsp_dds_offset(hdsp))
 		change = (hdsp_set_dds_offset(hdsp, val) == 0) ? 1 : 0;
diff --git a/sound/pci/rme9652/hdspm.c b/sound/pci/rme9652/hdspm.c
index 8bc8016c173d..a4a999a0317e 100644
--- a/sound/pci/rme9652/hdspm.c
+++ b/sound/pci/rme9652/hdspm.c
@@ -1601,6 +1601,9 @@ static void hdspm_set_dds_value(struct hdspm *hdspm, int rate)
 {
 	u64 n;
 
+	if (snd_BUG_ON(rate <= 0))
+		return;
+
 	if (rate >= 112000)
 		rate /= 4;
 	else if (rate >= 56000)
@@ -2215,6 +2218,8 @@ static int hdspm_get_system_sample_rate(struct hdspm *hdspm)
 		} else {
 			/* slave mode, return external sample rate */
 			rate = hdspm_external_sample_rate(hdspm);
+			if (!rate)
+				rate = hdspm->system_sample_rate;
 		}
 	}
 
@@ -2260,8 +2265,11 @@ static int snd_hdspm_put_system_sample_rate(struct snd_kcontrol *kcontrol,
 					    ucontrol)
 {
 	struct hdspm *hdspm = snd_kcontrol_chip(kcontrol);
+	int rate = ucontrol->value.integer.value[0];
 
-	hdspm_set_dds_value(hdspm, ucontrol->value.enumerated.item[0]);
+	if (rate < 27000 || rate > 207000)
+		return -EINVAL;
+	hdspm_set_dds_value(hdspm, ucontrol->value.integer.value[0]);
 	return 0;
 }
 
@@ -4449,7 +4457,7 @@ static int snd_hdspm_get_tco_word_term(struct snd_kcontrol *kcontrol,
 {
 	struct hdspm *hdspm = snd_kcontrol_chip(kcontrol);
 
-	ucontrol->value.enumerated.item[0] = hdspm->tco->term;
+	ucontrol->value.integer.value[0] = hdspm->tco->term;
 
 	return 0;
 }
@@ -4460,8 +4468,8 @@ static int snd_hdspm_put_tco_word_term(struct snd_kcontrol *kcontrol,
 {
 	struct hdspm *hdspm = snd_kcontrol_chip(kcontrol);
 
-	if (hdspm->tco->term != ucontrol->value.enumerated.item[0]) {
-		hdspm->tco->term = ucontrol->value.enumerated.item[0];
+	if (hdspm->tco->term != ucontrol->value.integer.value[0]) {
+		hdspm->tco->term = ucontrol->value.integer.value[0];
 
 		hdspm_tco_write(hdspm);
 
diff --git a/sound/soc/amd/acp-pcm-dma.c b/sound/soc/amd/acp-pcm-dma.c
index 3191e0a7d273..d1fb035f44db 100644
--- a/sound/soc/amd/acp-pcm-dma.c
+++ b/sound/soc/amd/acp-pcm-dma.c
@@ -635,6 +635,7 @@ static int acp_dma_open(struct snd_pcm_substream *substream)
 					    SNDRV_PCM_HW_PARAM_PERIODS);
 	if (ret < 0) {
 		dev_err(prtd->platform->dev, "set integer constraint failed\n");
+		kfree(adata);
 		return ret;
 	}
 
diff --git a/sound/soc/codecs/ab8500-codec.c b/sound/soc/codecs/ab8500-codec.c
index affb192238a4..faae6936bae4 100644
--- a/sound/soc/codecs/ab8500-codec.c
+++ b/sound/soc/codecs/ab8500-codec.c
@@ -1130,7 +1130,7 @@ static int sid_status_control_get(struct snd_kcontrol *kcontrol,
 	struct ab8500_codec_drvdata *drvdata = dev_get_drvdata(codec->dev);
 
 	mutex_lock(&drvdata->ctrl_lock);
-	ucontrol->value.integer.value[0] = drvdata->sid_status;
+	ucontrol->value.enumerated.item[0] = drvdata->sid_status;
 	mutex_unlock(&drvdata->ctrl_lock);
 
 	return 0;
@@ -1147,7 +1147,7 @@ static int sid_status_control_put(struct snd_kcontrol *kcontrol,
 
 	dev_dbg(codec->dev, "%s: Enter\n", __func__);
 
-	if (ucontrol->value.integer.value[0] != SID_APPLY_FIR) {
+	if (ucontrol->value.enumerated.item[0] != SID_APPLY_FIR) {
 		dev_err(codec->dev,
 			"%s: ERROR: This control supports '%s' only!\n",
 			__func__, enum_sid_state[SID_APPLY_FIR]);
@@ -1199,7 +1199,7 @@ static int anc_status_control_get(struct snd_kcontrol *kcontrol,
 	struct ab8500_codec_drvdata *drvdata = dev_get_drvdata(codec->dev);
 
 	mutex_lock(&drvdata->ctrl_lock);
-	ucontrol->value.integer.value[0] = drvdata->anc_status;
+	ucontrol->value.enumerated.item[0] = drvdata->anc_status;
 	mutex_unlock(&drvdata->ctrl_lock);
 
 	return 0;
@@ -1220,7 +1220,7 @@ static int anc_status_control_put(struct snd_kcontrol *kcontrol,
 
 	mutex_lock(&drvdata->ctrl_lock);
 
-	req = ucontrol->value.integer.value[0];
+	req = ucontrol->value.enumerated.item[0];
 	if (req >= ARRAY_SIZE(enum_anc_state)) {
 		status = -EINVAL;
 		goto cleanup;
diff --git a/sound/soc/codecs/adau17x1.h b/sound/soc/codecs/adau17x1.h
index e13583e6ff56..5ae87a084d97 100644
--- a/sound/soc/codecs/adau17x1.h
+++ b/sound/soc/codecs/adau17x1.h
@@ -103,9 +103,9 @@ bool adau17x1_has_dsp(struct adau *adau);
 #define ADAU17X1_CLOCK_CONTROL_CORECLK_SRC_PLL	BIT(3)
 #define ADAU17X1_CLOCK_CONTROL_SYSCLK_EN	BIT(0)
 
-#define ADAU17X1_SERIAL_PORT1_BCLK32		(0x0 << 5)
-#define ADAU17X1_SERIAL_PORT1_BCLK48		(0x1 << 5)
-#define ADAU17X1_SERIAL_PORT1_BCLK64		(0x2 << 5)
+#define ADAU17X1_SERIAL_PORT1_BCLK64		(0x0 << 5)
+#define ADAU17X1_SERIAL_PORT1_BCLK32		(0x1 << 5)
+#define ADAU17X1_SERIAL_PORT1_BCLK48		(0x2 << 5)
 #define ADAU17X1_SERIAL_PORT1_BCLK128		(0x3 << 5)
 #define ADAU17X1_SERIAL_PORT1_BCLK256		(0x4 << 5)
 #define ADAU17X1_SERIAL_PORT1_BCLK_MASK		(0x7 << 5)
diff --git a/sound/soc/codecs/arizona.c b/sound/soc/codecs/arizona.c
index 33143fe1de0b..91785318b283 100644
--- a/sound/soc/codecs/arizona.c
+++ b/sound/soc/codecs/arizona.c
@@ -1929,6 +1929,25 @@ static struct {
 	{ 1000000, 13500000, 0,  1 },
 };
 
+static const unsigned int pseudo_fref_max[ARIZONA_FLL_MAX_FRATIO] = {
+	13500000,
+	 6144000,
+	 6144000,
+	 3072000,
+	 3072000,
+	 2822400,
+	 2822400,
+	 1536000,
+	 1536000,
+	 1536000,
+	 1536000,
+	 1536000,
+	 1536000,
+	 1536000,
+	 1536000,
+	  768000,
+};
+
 static struct {
 	unsigned int min;
 	unsigned int max;
@@ -2042,16 +2061,32 @@ static int arizona_calc_fratio(struct arizona_fll *fll,
 	/* Adjust FRATIO/refdiv to avoid integer mode if possible */
 	refdiv = cfg->refdiv;
 
+	arizona_fll_dbg(fll, "pseudo: initial ratio=%u fref=%u refdiv=%u\n",
+			init_ratio, Fref, refdiv);
+
 	while (div <= ARIZONA_FLL_MAX_REFDIV) {
 		for (ratio = init_ratio; ratio <= ARIZONA_FLL_MAX_FRATIO;
 		     ratio++) {
 			if ((ARIZONA_FLL_VCO_CORNER / 2) /
-			    (fll->vco_mult * ratio) < Fref)
+			    (fll->vco_mult * ratio) < Fref) {
+				arizona_fll_dbg(fll, "pseudo: hit VCO corner\n");
 				break;
+			}
+
+			if (Fref > pseudo_fref_max[ratio - 1]) {
+				arizona_fll_dbg(fll,
+					"pseudo: exceeded max fref(%u) for ratio=%u\n",
+					pseudo_fref_max[ratio - 1],
+					ratio);
+				break;
+			}
 
 			if (target % (ratio * Fref)) {
 				cfg->refdiv = refdiv;
 				cfg->fratio = ratio - 1;
+				arizona_fll_dbg(fll,
+					"pseudo: found fref=%u refdiv=%d(%d) ratio=%d\n",
+					Fref, refdiv, div, ratio);
 				return ratio;
 			}
 		}
@@ -2060,6 +2095,9 @@ static int arizona_calc_fratio(struct arizona_fll *fll,
 			if (target % (ratio * Fref)) {
 				cfg->refdiv = refdiv;
 				cfg->fratio = ratio - 1;
+				arizona_fll_dbg(fll,
+					"pseudo: found fref=%u refdiv=%d(%d) ratio=%d\n",
+					Fref, refdiv, div, ratio);
 				return ratio;
 			}
 		}
@@ -2068,6 +2106,9 @@ static int arizona_calc_fratio(struct arizona_fll *fll,
 		Fref /= 2;
 		refdiv++;
 		init_ratio = arizona_find_fratio(Fref, NULL);
+		arizona_fll_dbg(fll,
+				"pseudo: change fref=%u refdiv=%d(%d) ratio=%u\n",
+				Fref, refdiv, div, init_ratio);
 	}
 
 	arizona_fll_warn(fll, "Falling back to integer mode operation\n");
diff --git a/sound/soc/codecs/cs42l51.c b/sound/soc/codecs/cs42l51.c
index b3951524339f..35488f14e237 100644
--- a/sound/soc/codecs/cs42l51.c
+++ b/sound/soc/codecs/cs42l51.c
@@ -60,15 +60,15 @@ static int cs42l51_get_chan_mix(struct snd_kcontrol *kcontrol,
 	switch (value) {
 	default:
 	case 0:
-		ucontrol->value.integer.value[0] = 0;
+		ucontrol->value.enumerated.item[0] = 0;
 		break;
 	/* same value : (L+R)/2 and (R+L)/2 */
 	case 1:
 	case 2:
-		ucontrol->value.integer.value[0] = 1;
+		ucontrol->value.enumerated.item[0] = 1;
 		break;
 	case 3:
-		ucontrol->value.integer.value[0] = 2;
+		ucontrol->value.enumerated.item[0] = 2;
 		break;
 	}
 
@@ -85,7 +85,7 @@ static int cs42l51_set_chan_mix(struct snd_kcontrol *kcontrol,
 	struct snd_soc_codec *codec = snd_soc_kcontrol_codec(kcontrol);
 	unsigned char val;
 
-	switch (ucontrol->value.integer.value[0]) {
+	switch (ucontrol->value.enumerated.item[0]) {
 	default:
 	case 0:
 		val = CHAN_MIX_NORMAL;
diff --git a/sound/soc/codecs/da732x.c b/sound/soc/codecs/da732x.c
index 1d5a89c5164b..461506a4ca6a 100644
--- a/sound/soc/codecs/da732x.c
+++ b/sound/soc/codecs/da732x.c
@@ -334,7 +334,7 @@ static int da732x_hpf_set(struct snd_kcontrol *kcontrol,
 	struct snd_soc_codec *codec = snd_soc_kcontrol_codec(kcontrol);
 	struct soc_enum *enum_ctrl = (struct soc_enum *)kcontrol->private_value;
 	unsigned int reg = enum_ctrl->reg;
-	unsigned int sel = ucontrol->value.integer.value[0];
+	unsigned int sel = ucontrol->value.enumerated.item[0];
 	unsigned int bits;
 
 	switch (sel) {
@@ -368,13 +368,13 @@ static int da732x_hpf_get(struct snd_kcontrol *kcontrol,
 
 	switch (val) {
 	case DA732X_HPF_VOICE_EN:
-		ucontrol->value.integer.value[0] = DA732X_HPF_VOICE;
+		ucontrol->value.enumerated.item[0] = DA732X_HPF_VOICE;
 		break;
 	case DA732X_HPF_MUSIC_EN:
-		ucontrol->value.integer.value[0] = DA732X_HPF_MUSIC;
+		ucontrol->value.enumerated.item[0] = DA732X_HPF_MUSIC;
 		break;
 	default:
-		ucontrol->value.integer.value[0] = DA732X_HPF_DISABLED;
+		ucontrol->value.enumerated.item[0] = DA732X_HPF_DISABLED;
 		break;
 	}
 
diff --git a/sound/soc/codecs/max98088.c b/sound/soc/codecs/max98088.c
index 20dcc496d39c..fc22804cabc5 100644
--- a/sound/soc/codecs/max98088.c
+++ b/sound/soc/codecs/max98088.c
@@ -1496,7 +1496,7 @@ static int max98088_put_eq_enum(struct snd_kcontrol *kcontrol,
        struct max98088_pdata *pdata = max98088->pdata;
        int channel = max98088_get_channel(codec, kcontrol->id.name);
        struct max98088_cdata *cdata;
-       int sel = ucontrol->value.integer.value[0];
+	int sel = ucontrol->value.enumerated.item[0];
 
        if (channel < 0)
 	       return channel;
diff --git a/sound/soc/codecs/max98095.c b/sound/soc/codecs/max98095.c
index 1fedac50355e..3577003f39cf 100644
--- a/sound/soc/codecs/max98095.c
+++ b/sound/soc/codecs/max98095.c
@@ -1499,7 +1499,7 @@ static int max98095_put_eq_enum(struct snd_kcontrol *kcontrol,
 	struct max98095_pdata *pdata = max98095->pdata;
 	int channel = max98095_get_eq_channel(kcontrol->id.name);
 	struct max98095_cdata *cdata;
-	unsigned int sel = ucontrol->value.integer.value[0];
+	unsigned int sel = ucontrol->value.enumerated.item[0];
 	struct max98095_eq_cfg *coef_set;
 	int fs, best, best_val, i;
 	int regmask, regsave;
@@ -1653,7 +1653,7 @@ static int max98095_put_bq_enum(struct snd_kcontrol *kcontrol,
 	struct max98095_pdata *pdata = max98095->pdata;
 	int channel = max98095_get_bq_channel(codec, kcontrol->id.name);
 	struct max98095_cdata *cdata;
-	unsigned int sel = ucontrol->value.integer.value[0];
+	unsigned int sel = ucontrol->value.enumerated.item[0];
 	struct max98095_biquad_cfg *coef_set;
 	int fs, best, best_val, i;
 	int regmask, regsave;
diff --git a/sound/soc/codecs/rt286.c b/sound/soc/codecs/rt286.c
index bc08f0c5a5f6..1bd31644a782 100644
--- a/sound/soc/codecs/rt286.c
+++ b/sound/soc/codecs/rt286.c
@@ -266,6 +266,8 @@ static int rt286_jack_detect(struct rt286_priv *rt286, bool *hp, bool *mic)
 		} else {
 			*mic = false;
 			regmap_write(rt286->regmap, RT286_SET_MIC1, 0x20);
+			regmap_update_bits(rt286->regmap,
+				RT286_CBJ_CTRL1, 0x0400, 0x0000);
 		}
 	} else {
 		regmap_read(rt286->regmap, RT286_GET_HP_SENSE, &buf);
@@ -470,24 +472,6 @@ static int rt286_set_dmic1_event(struct snd_soc_dapm_widget *w,
 	return 0;
 }
 
-static int rt286_vref_event(struct snd_soc_dapm_widget *w,
-			     struct snd_kcontrol *kcontrol, int event)
-{
-	struct snd_soc_codec *codec = snd_soc_dapm_to_codec(w->dapm);
-
-	switch (event) {
-	case SND_SOC_DAPM_PRE_PMU:
-		snd_soc_update_bits(codec,
-			RT286_CBJ_CTRL1, 0x0400, 0x0000);
-		mdelay(50);
-		break;
-	default:
-		return 0;
-	}
-
-	return 0;
-}
-
 static int rt286_ldo2_event(struct snd_soc_dapm_widget *w,
 			     struct snd_kcontrol *kcontrol, int event)
 {
@@ -536,7 +520,7 @@ static const struct snd_soc_dapm_widget rt286_dapm_widgets[] = {
 	SND_SOC_DAPM_SUPPLY_S("HV", 1, RT286_POWER_CTRL1,
 		12, 1, NULL, 0),
 	SND_SOC_DAPM_SUPPLY("VREF", RT286_POWER_CTRL1,
-		0, 1, rt286_vref_event, SND_SOC_DAPM_PRE_PMU),
+		0, 1, NULL, 0),
 	SND_SOC_DAPM_SUPPLY_S("LDO1", 1, RT286_POWER_CTRL2,
 		2, 0, NULL, 0),
 	SND_SOC_DAPM_SUPPLY_S("LDO2", 2, RT286_POWER_CTRL1,
@@ -911,8 +895,6 @@ static int rt286_set_bias_level(struct snd_soc_codec *codec,
 	case SND_SOC_BIAS_ON:
 		mdelay(10);
 		snd_soc_update_bits(codec,
-			RT286_CBJ_CTRL1, 0x0400, 0x0400);
-		snd_soc_update_bits(codec,
 			RT286_DC_GAIN, 0x200, 0x0);
 
 		break;
@@ -920,8 +902,6 @@ static int rt286_set_bias_level(struct snd_soc_codec *codec,
 	case SND_SOC_BIAS_STANDBY:
 		snd_soc_write(codec,
 			RT286_SET_AUDIO_POWER, AC_PWRST_D3);
-		snd_soc_update_bits(codec,
-			RT286_CBJ_CTRL1, 0x0400, 0x0000);
 		break;
 
 	default:
diff --git a/sound/soc/codecs/rt5645.c b/sound/soc/codecs/rt5645.c
index c61d38b585fb..93e8c9017633 100644
--- a/sound/soc/codecs/rt5645.c
+++ b/sound/soc/codecs/rt5645.c
@@ -776,7 +776,7 @@ static const struct snd_kcontrol_new rt5645_snd_controls[] = {
 
 	/* IN1/IN2 Control */
 	SOC_SINGLE_TLV("IN1 Boost", RT5645_IN1_CTRL1,
-		RT5645_BST_SFT1, 8, 0, bst_tlv),
+		RT5645_BST_SFT1, 12, 0, bst_tlv),
 	SOC_SINGLE_TLV("IN2 Boost", RT5645_IN2_CTRL,
 		RT5645_BST_SFT2, 8, 0, bst_tlv),
 
diff --git a/sound/soc/codecs/rt5659.c b/sound/soc/codecs/rt5659.c
index 820d8fa62b5e..fb8ea05c0de1 100644
--- a/sound/soc/codecs/rt5659.c
+++ b/sound/soc/codecs/rt5659.c
@@ -3985,7 +3985,6 @@ static int rt5659_i2c_probe(struct i2c_client *i2c,
 	if (rt5659 == NULL)
 		return -ENOMEM;
 
-	rt5659->i2c = i2c;
 	i2c_set_clientdata(i2c, rt5659);
 
 	if (pdata)
@@ -4157,24 +4156,17 @@ static int rt5659_i2c_probe(struct i2c_client *i2c,
 
 	INIT_DELAYED_WORK(&rt5659->jack_detect_work, rt5659_jack_detect_work);
 
-	if (rt5659->i2c->irq) {
-		ret = request_threaded_irq(rt5659->i2c->irq, NULL, rt5659_irq,
-			IRQF_TRIGGER_RISING | IRQF_TRIGGER_FALLING
+	if (i2c->irq) {
+		ret = devm_request_threaded_irq(&i2c->dev, i2c->irq, NULL,
+			rt5659_irq, IRQF_TRIGGER_RISING | IRQF_TRIGGER_FALLING
 			| IRQF_ONESHOT, "rt5659", rt5659);
 		if (ret)
 			dev_err(&i2c->dev, "Failed to reguest IRQ: %d\n", ret);
 
 	}
 
-	ret = snd_soc_register_codec(&i2c->dev, &soc_codec_dev_rt5659,
+	return snd_soc_register_codec(&i2c->dev, &soc_codec_dev_rt5659,
 			rt5659_dai, ARRAY_SIZE(rt5659_dai));
-
-	if (ret) {
-		if (rt5659->i2c->irq)
-			free_irq(rt5659->i2c->irq, rt5659);
-	}
-
-	return 0;
 }
 
 static int rt5659_i2c_remove(struct i2c_client *i2c)
@@ -4191,24 +4183,29 @@ void rt5659_i2c_shutdown(struct i2c_client *client)
 	regmap_write(rt5659->regmap, RT5659_RESET, 0);
 }
 
+#ifdef CONFIG_OF
 static const struct of_device_id rt5659_of_match[] = {
 	{ .compatible = "realtek,rt5658", },
 	{ .compatible = "realtek,rt5659", },
-	{},
+	{ },
 };
+MODULE_DEVICE_TABLE(of, rt5659_of_match);
+#endif
 
+#ifdef CONFIG_ACPI
 static struct acpi_device_id rt5659_acpi_match[] = {
-		{ "10EC5658", 0},
-		{ "10EC5659", 0},
-		{ },
+	{ "10EC5658", 0, },
+	{ "10EC5659", 0, },
+	{ },
 };
 MODULE_DEVICE_TABLE(acpi, rt5659_acpi_match);
+#endif
 
 struct i2c_driver rt5659_i2c_driver = {
 	.driver = {
 		.name = "rt5659",
 		.owner = THIS_MODULE,
-		.of_match_table = rt5659_of_match,
+		.of_match_table = of_match_ptr(rt5659_of_match),
 		.acpi_match_table = ACPI_PTR(rt5659_acpi_match),
 	},
 	.probe = rt5659_i2c_probe,
diff --git a/sound/soc/codecs/rt5659.h b/sound/soc/codecs/rt5659.h
index 8f07ee903eaa..d31c9e5bcec8 100644
--- a/sound/soc/codecs/rt5659.h
+++ b/sound/soc/codecs/rt5659.h
@@ -1792,7 +1792,6 @@ struct rt5659_priv {
 	struct snd_soc_codec *codec;
 	struct rt5659_platform_data pdata;
 	struct regmap *regmap;
-	struct i2c_client *i2c;
 	struct gpio_desc *gpiod_ldo1_en;
 	struct gpio_desc *gpiod_reset;
 	struct snd_soc_jack *hs_jack;
diff --git a/sound/soc/codecs/sigmadsp-i2c.c b/sound/soc/codecs/sigmadsp-i2c.c
index 21ca3a5e9f66..d374c18d4db7 100644
--- a/sound/soc/codecs/sigmadsp-i2c.c
+++ b/sound/soc/codecs/sigmadsp-i2c.c
@@ -31,7 +31,10 @@ static int sigmadsp_write_i2c(void *control_data,
 
 	kfree(buf);
 
-	return ret;
+	if (ret < 0)
+		return ret;
+
+	return 0;
 }
 
 static int sigmadsp_read_i2c(void *control_data,
diff --git a/sound/soc/codecs/tlv320dac33.c b/sound/soc/codecs/tlv320dac33.c
index 781398fb2841..f7a6ce7e5fb1 100644
--- a/sound/soc/codecs/tlv320dac33.c
+++ b/sound/soc/codecs/tlv320dac33.c
@@ -446,7 +446,7 @@ static int dac33_get_fifo_mode(struct snd_kcontrol *kcontrol,
 	struct snd_soc_codec *codec = snd_soc_kcontrol_codec(kcontrol);
 	struct tlv320dac33_priv *dac33 = snd_soc_codec_get_drvdata(codec);
 
-	ucontrol->value.integer.value[0] = dac33->fifo_mode;
+	ucontrol->value.enumerated.item[0] = dac33->fifo_mode;
 
 	return 0;
 }
@@ -458,17 +458,16 @@ static int dac33_set_fifo_mode(struct snd_kcontrol *kcontrol,
 	struct tlv320dac33_priv *dac33 = snd_soc_codec_get_drvdata(codec);
 	int ret = 0;
 
-	if (dac33->fifo_mode == ucontrol->value.integer.value[0])
+	if (dac33->fifo_mode == ucontrol->value.enumerated.item[0])
 		return 0;
 	/* Do not allow changes while stream is running*/
 	if (snd_soc_codec_is_active(codec))
 		return -EPERM;
 
-	if (ucontrol->value.integer.value[0] < 0 ||
-	    ucontrol->value.integer.value[0] >= DAC33_FIFO_LAST_MODE)
+	if (ucontrol->value.enumerated.item[0] >= DAC33_FIFO_LAST_MODE)
 		ret = -EINVAL;
 	else
-		dac33->fifo_mode = ucontrol->value.integer.value[0];
+		dac33->fifo_mode = ucontrol->value.enumerated.item[0];
 
 	return ret;
 }
diff --git a/sound/soc/codecs/wl1273.c b/sound/soc/codecs/wl1273.c
index 7693c1129bab..1b79778098d2 100644
--- a/sound/soc/codecs/wl1273.c
+++ b/sound/soc/codecs/wl1273.c
@@ -175,7 +175,7 @@ static int snd_wl1273_get_audio_route(struct snd_kcontrol *kcontrol,
 	struct snd_soc_codec *codec = snd_soc_kcontrol_codec(kcontrol);
 	struct wl1273_priv *wl1273 = snd_soc_codec_get_drvdata(codec);
 
-	ucontrol->value.integer.value[0] = wl1273->mode;
+	ucontrol->value.enumerated.item[0] = wl1273->mode;
 
 	return 0;
 }
@@ -193,18 +193,17 @@ static int snd_wl1273_set_audio_route(struct snd_kcontrol *kcontrol,
 	struct snd_soc_codec *codec = snd_soc_kcontrol_codec(kcontrol);
 	struct wl1273_priv *wl1273 = snd_soc_codec_get_drvdata(codec);
 
-	if (wl1273->mode == ucontrol->value.integer.value[0])
+	if (wl1273->mode == ucontrol->value.enumerated.item[0])
 		return 0;
 
 	/* Do not allow changes while stream is running */
 	if (snd_soc_codec_is_active(codec))
 		return -EPERM;
 
-	if (ucontrol->value.integer.value[0] < 0 ||
-	    ucontrol->value.integer.value[0] >=  ARRAY_SIZE(wl1273_audio_route))
+	if (ucontrol->value.enumerated.item[0] >=  ARRAY_SIZE(wl1273_audio_route))
 		return -EINVAL;
 
-	wl1273->mode = ucontrol->value.integer.value[0];
+	wl1273->mode = ucontrol->value.enumerated.item[0];
 
 	return 1;
 }
@@ -219,7 +218,7 @@ static int snd_wl1273_fm_audio_get(struct snd_kcontrol *kcontrol,
 
 	dev_dbg(codec->dev, "%s: enter.\n", __func__);
 
-	ucontrol->value.integer.value[0] = wl1273->core->audio_mode;
+	ucontrol->value.enumerated.item[0] = wl1273->core->audio_mode;
 
 	return 0;
 }
@@ -233,7 +232,7 @@ static int snd_wl1273_fm_audio_put(struct snd_kcontrol *kcontrol,
 
 	dev_dbg(codec->dev, "%s: enter.\n", __func__);
 
-	val = ucontrol->value.integer.value[0];
+	val = ucontrol->value.enumerated.item[0];
 	if (wl1273->core->audio_mode == val)
 		return 0;
 
diff --git a/sound/soc/codecs/wm5110.c b/sound/soc/codecs/wm5110.c
index 6088d30962a9..97c0f1e23886 100644
--- a/sound/soc/codecs/wm5110.c
+++ b/sound/soc/codecs/wm5110.c
@@ -2382,6 +2382,7 @@ error:
 
 static int wm5110_remove(struct platform_device *pdev)
 {
+	snd_soc_unregister_platform(&pdev->dev);
 	snd_soc_unregister_codec(&pdev->dev);
 	pm_runtime_disable(&pdev->dev);
 
diff --git a/sound/soc/codecs/wm8753.c b/sound/soc/codecs/wm8753.c
index 61299ca372ff..6f1024f48b19 100644
--- a/sound/soc/codecs/wm8753.c
+++ b/sound/soc/codecs/wm8753.c
@@ -233,7 +233,7 @@ static int wm8753_get_dai(struct snd_kcontrol *kcontrol,
 	struct snd_soc_codec *codec = snd_soc_kcontrol_codec(kcontrol);
 	struct wm8753_priv *wm8753 = snd_soc_codec_get_drvdata(codec);
 
-	ucontrol->value.integer.value[0] = wm8753->dai_func;
+	ucontrol->value.enumerated.item[0] = wm8753->dai_func;
 	return 0;
 }
 
@@ -244,7 +244,7 @@ static int wm8753_set_dai(struct snd_kcontrol *kcontrol,
 	struct wm8753_priv *wm8753 = snd_soc_codec_get_drvdata(codec);
 	u16 ioctl;
 
-	if (wm8753->dai_func == ucontrol->value.integer.value[0])
+	if (wm8753->dai_func == ucontrol->value.enumerated.item[0])
 		return 0;
 
 	if (snd_soc_codec_is_active(codec))
@@ -252,7 +252,7 @@ static int wm8753_set_dai(struct snd_kcontrol *kcontrol,
 
 	ioctl = snd_soc_read(codec, WM8753_IOCTL);
 
-	wm8753->dai_func = ucontrol->value.integer.value[0];
+	wm8753->dai_func = ucontrol->value.enumerated.item[0];
 
 	if (((ioctl >> 2) & 0x3) == wm8753->dai_func)
 		return 1;
diff --git a/sound/soc/codecs/wm8904.c b/sound/soc/codecs/wm8904.c
index 8172e499e6ed..edd7a7709194 100644
--- a/sound/soc/codecs/wm8904.c
+++ b/sound/soc/codecs/wm8904.c
@@ -396,7 +396,7 @@ static int wm8904_put_drc_enum(struct snd_kcontrol *kcontrol,
 	struct snd_soc_codec *codec = snd_soc_kcontrol_codec(kcontrol);
 	struct wm8904_priv *wm8904 = snd_soc_codec_get_drvdata(codec);
 	struct wm8904_pdata *pdata = wm8904->pdata;
-	int value = ucontrol->value.integer.value[0];
+	int value = ucontrol->value.enumerated.item[0];
 
 	if (value >= pdata->num_drc_cfgs)
 		return -EINVAL;
@@ -467,7 +467,7 @@ static int wm8904_put_retune_mobile_enum(struct snd_kcontrol *kcontrol,
 	struct snd_soc_codec *codec = snd_soc_kcontrol_codec(kcontrol);
 	struct wm8904_priv *wm8904 = snd_soc_codec_get_drvdata(codec);
 	struct wm8904_pdata *pdata = wm8904->pdata;
-	int value = ucontrol->value.integer.value[0];
+	int value = ucontrol->value.enumerated.item[0];
 
 	if (value >= pdata->num_retune_mobile_cfgs)
 		return -EINVAL;
diff --git a/sound/soc/codecs/wm8958-dsp2.c b/sound/soc/codecs/wm8958-dsp2.c
index c799cca5abeb..6b864c0fc2b6 100644
--- a/sound/soc/codecs/wm8958-dsp2.c
+++ b/sound/soc/codecs/wm8958-dsp2.c
@@ -459,7 +459,7 @@ static int wm8958_put_mbc_enum(struct snd_kcontrol *kcontrol,
 	struct snd_soc_codec *codec = snd_soc_kcontrol_codec(kcontrol);
 	struct wm8994_priv *wm8994 = snd_soc_codec_get_drvdata(codec);
 	struct wm8994 *control = wm8994->wm8994;
-	int value = ucontrol->value.integer.value[0];
+	int value = ucontrol->value.enumerated.item[0];
 	int reg;
 
 	/* Don't allow on the fly reconfiguration */
@@ -549,7 +549,7 @@ static int wm8958_put_vss_enum(struct snd_kcontrol *kcontrol,
 	struct snd_soc_codec *codec = snd_soc_kcontrol_codec(kcontrol);
 	struct wm8994_priv *wm8994 = snd_soc_codec_get_drvdata(codec);
 	struct wm8994 *control = wm8994->wm8994;
-	int value = ucontrol->value.integer.value[0];
+	int value = ucontrol->value.enumerated.item[0];
 	int reg;
 
 	/* Don't allow on the fly reconfiguration */
@@ -582,7 +582,7 @@ static int wm8958_put_vss_hpf_enum(struct snd_kcontrol *kcontrol,
 	struct snd_soc_codec *codec = snd_soc_kcontrol_codec(kcontrol);
 	struct wm8994_priv *wm8994 = snd_soc_codec_get_drvdata(codec);
 	struct wm8994 *control = wm8994->wm8994;
-	int value = ucontrol->value.integer.value[0];
+	int value = ucontrol->value.enumerated.item[0];
 	int reg;
 
 	/* Don't allow on the fly reconfiguration */
@@ -749,7 +749,7 @@ static int wm8958_put_enh_eq_enum(struct snd_kcontrol *kcontrol,
 	struct snd_soc_codec *codec = snd_soc_kcontrol_codec(kcontrol);
 	struct wm8994_priv *wm8994 = snd_soc_codec_get_drvdata(codec);
 	struct wm8994 *control = wm8994->wm8994;
-	int value = ucontrol->value.integer.value[0];
+	int value = ucontrol->value.enumerated.item[0];
 	int reg;
 
 	/* Don't allow on the fly reconfiguration */
diff --git a/sound/soc/codecs/wm8960.c b/sound/soc/codecs/wm8960.c
index ff237726775a..d7f444f87460 100644
--- a/sound/soc/codecs/wm8960.c
+++ b/sound/soc/codecs/wm8960.c
@@ -240,13 +240,13 @@ SOC_DOUBLE_R("Capture Volume ZC Switch", WM8960_LINVOL, WM8960_RINVOL,
 SOC_DOUBLE_R("Capture Switch", WM8960_LINVOL, WM8960_RINVOL,
 	7, 1, 1),
 
-SOC_SINGLE_TLV("Right Input Boost Mixer RINPUT3 Volume",
+SOC_SINGLE_TLV("Left Input Boost Mixer LINPUT3 Volume",
 	       WM8960_INBMIX1, 4, 7, 0, lineinboost_tlv),
-SOC_SINGLE_TLV("Right Input Boost Mixer RINPUT2 Volume",
+SOC_SINGLE_TLV("Left Input Boost Mixer LINPUT2 Volume",
 	       WM8960_INBMIX1, 1, 7, 0, lineinboost_tlv),
-SOC_SINGLE_TLV("Left Input Boost Mixer LINPUT3 Volume",
+SOC_SINGLE_TLV("Right Input Boost Mixer RINPUT3 Volume",
 	       WM8960_INBMIX2, 4, 7, 0, lineinboost_tlv),
-SOC_SINGLE_TLV("Left Input Boost Mixer LINPUT2 Volume",
+SOC_SINGLE_TLV("Right Input Boost Mixer RINPUT2 Volume",
 	       WM8960_INBMIX2, 1, 7, 0, lineinboost_tlv),
 SOC_SINGLE_TLV("Right Input Boost Mixer RINPUT1 Volume",
 		WM8960_RINPATH, 4, 3, 0, micboost_tlv),
@@ -643,29 +643,31 @@ static int wm8960_configure_clocking(struct snd_soc_codec *codec)
 		return -EINVAL;
 	}
 
-	/* check if the sysclk frequency is available. */
-	for (i = 0; i < ARRAY_SIZE(sysclk_divs); ++i) {
-		if (sysclk_divs[i] == -1)
-			continue;
-		sysclk = freq_out / sysclk_divs[i];
-		for (j = 0; j < ARRAY_SIZE(dac_divs); ++j) {
-			if (sysclk == dac_divs[j] * lrclk) {
+	if (wm8960->clk_id != WM8960_SYSCLK_PLL) {
+		/* check if the sysclk frequency is available. */
+		for (i = 0; i < ARRAY_SIZE(sysclk_divs); ++i) {
+			if (sysclk_divs[i] == -1)
+				continue;
+			sysclk = freq_out / sysclk_divs[i];
+			for (j = 0; j < ARRAY_SIZE(dac_divs); ++j) {
+				if (sysclk != dac_divs[j] * lrclk)
+					continue;
 				for (k = 0; k < ARRAY_SIZE(bclk_divs); ++k)
 					if (sysclk == bclk * bclk_divs[k] / 10)
 						break;
 				if (k != ARRAY_SIZE(bclk_divs))
 					break;
 			}
+			if (j != ARRAY_SIZE(dac_divs))
+				break;
 		}
-		if (j != ARRAY_SIZE(dac_divs))
-			break;
-	}
 
-	if (i != ARRAY_SIZE(sysclk_divs)) {
-		goto configure_clock;
-	} else if (wm8960->clk_id != WM8960_SYSCLK_AUTO) {
-		dev_err(codec->dev, "failed to configure clock\n");
-		return -EINVAL;
+		if (i != ARRAY_SIZE(sysclk_divs)) {
+			goto configure_clock;
+		} else if (wm8960->clk_id != WM8960_SYSCLK_AUTO) {
+			dev_err(codec->dev, "failed to configure clock\n");
+			return -EINVAL;
+		}
 	}
 	/* get a available pll out frequency and set pll */
 	for (i = 0; i < ARRAY_SIZE(sysclk_divs); ++i) {
diff --git a/sound/soc/codecs/wm8983.c b/sound/soc/codecs/wm8983.c
index 7350ff654bbf..0c002a5712cb 100644
--- a/sound/soc/codecs/wm8983.c
+++ b/sound/soc/codecs/wm8983.c
@@ -497,9 +497,9 @@ static int eqmode_get(struct snd_kcontrol *kcontrol,
 
 	reg = snd_soc_read(codec, WM8983_EQ1_LOW_SHELF);
 	if (reg & WM8983_EQ3DMODE)
-		ucontrol->value.integer.value[0] = 1;
+		ucontrol->value.enumerated.item[0] = 1;
 	else
-		ucontrol->value.integer.value[0] = 0;
+		ucontrol->value.enumerated.item[0] = 0;
 
 	return 0;
 }
@@ -511,18 +511,18 @@ static int eqmode_put(struct snd_kcontrol *kcontrol,
 	unsigned int regpwr2, regpwr3;
 	unsigned int reg_eq;
 
-	if (ucontrol->value.integer.value[0] != 0
-	    && ucontrol->value.integer.value[0] != 1)
+	if (ucontrol->value.enumerated.item[0] != 0
+	    && ucontrol->value.enumerated.item[0] != 1)
 		return -EINVAL;
 
 	reg_eq = snd_soc_read(codec, WM8983_EQ1_LOW_SHELF);
 	switch ((reg_eq & WM8983_EQ3DMODE) >> WM8983_EQ3DMODE_SHIFT) {
 	case 0:
-		if (!ucontrol->value.integer.value[0])
+		if (!ucontrol->value.enumerated.item[0])
 			return 0;
 		break;
 	case 1:
-		if (ucontrol->value.integer.value[0])
+		if (ucontrol->value.enumerated.item[0])
 			return 0;
 		break;
 	}
@@ -537,7 +537,7 @@ static int eqmode_put(struct snd_kcontrol *kcontrol,
 	/* set the desired eqmode */
 	snd_soc_update_bits(codec, WM8983_EQ1_LOW_SHELF,
 			    WM8983_EQ3DMODE_MASK,
-			    ucontrol->value.integer.value[0]
+			    ucontrol->value.enumerated.item[0]
 			    << WM8983_EQ3DMODE_SHIFT);
 	/* restore DAC/ADC configuration */
 	snd_soc_write(codec, WM8983_POWER_MANAGEMENT_2, regpwr2);
diff --git a/sound/soc/codecs/wm8985.c b/sound/soc/codecs/wm8985.c
index 9918152a03c7..6ac76fe116b0 100644
--- a/sound/soc/codecs/wm8985.c
+++ b/sound/soc/codecs/wm8985.c
@@ -531,9 +531,9 @@ static int eqmode_get(struct snd_kcontrol *kcontrol,
 
 	reg = snd_soc_read(codec, WM8985_EQ1_LOW_SHELF);
 	if (reg & WM8985_EQ3DMODE)
-		ucontrol->value.integer.value[0] = 1;
+		ucontrol->value.enumerated.item[0] = 1;
 	else
-		ucontrol->value.integer.value[0] = 0;
+		ucontrol->value.enumerated.item[0] = 0;
 
 	return 0;
 }
@@ -545,18 +545,18 @@ static int eqmode_put(struct snd_kcontrol *kcontrol,
 	unsigned int regpwr2, regpwr3;
 	unsigned int reg_eq;
 
-	if (ucontrol->value.integer.value[0] != 0
-			&& ucontrol->value.integer.value[0] != 1)
+	if (ucontrol->value.enumerated.item[0] != 0
+			&& ucontrol->value.enumerated.item[0] != 1)
 		return -EINVAL;
 
 	reg_eq = snd_soc_read(codec, WM8985_EQ1_LOW_SHELF);
 	switch ((reg_eq & WM8985_EQ3DMODE) >> WM8985_EQ3DMODE_SHIFT) {
 	case 0:
-		if (!ucontrol->value.integer.value[0])
+		if (!ucontrol->value.enumerated.item[0])
 			return 0;
 		break;
 	case 1:
-		if (ucontrol->value.integer.value[0])
+		if (ucontrol->value.enumerated.item[0])
 			return 0;
 		break;
 	}
@@ -573,7 +573,7 @@ static int eqmode_put(struct snd_kcontrol *kcontrol,
 	/* set the desired eqmode */
 	snd_soc_update_bits(codec, WM8985_EQ1_LOW_SHELF,
 			    WM8985_EQ3DMODE_MASK,
-			    ucontrol->value.integer.value[0]
+			    ucontrol->value.enumerated.item[0]
 			    << WM8985_EQ3DMODE_SHIFT);
 	/* restore DAC/ADC configuration */
 	snd_soc_write(codec, WM8985_POWER_MANAGEMENT_2, regpwr2);
diff --git a/sound/soc/codecs/wm8994.c b/sound/soc/codecs/wm8994.c
index 2ccbb322df77..a18aecb49935 100644
--- a/sound/soc/codecs/wm8994.c
+++ b/sound/soc/codecs/wm8994.c
@@ -362,7 +362,7 @@ static int wm8994_put_drc_enum(struct snd_kcontrol *kcontrol,
 	struct wm8994 *control = wm8994->wm8994;
 	struct wm8994_pdata *pdata = &control->pdata;
 	int drc = wm8994_get_drc(kcontrol->id.name);
-	int value = ucontrol->value.integer.value[0];
+	int value = ucontrol->value.enumerated.item[0];
 
 	if (drc < 0)
 		return drc;
@@ -469,7 +469,7 @@ static int wm8994_put_retune_mobile_enum(struct snd_kcontrol *kcontrol,
 	struct wm8994 *control = wm8994->wm8994;
 	struct wm8994_pdata *pdata = &control->pdata;
 	int block = wm8994_get_retune_mobile_block(kcontrol->id.name);
-	int value = ucontrol->value.integer.value[0];
+	int value = ucontrol->value.enumerated.item[0];
 
 	if (block < 0)
 		return block;
diff --git a/sound/soc/codecs/wm8996.c b/sound/soc/codecs/wm8996.c
index 8d7d6c01a2f7..f99b34f7647b 100644
--- a/sound/soc/codecs/wm8996.c
+++ b/sound/soc/codecs/wm8996.c
@@ -416,7 +416,7 @@ static int wm8996_put_retune_mobile_enum(struct snd_kcontrol *kcontrol,
 	struct wm8996_priv *wm8996 = snd_soc_codec_get_drvdata(codec);
 	struct wm8996_pdata *pdata = &wm8996->pdata;
 	int block = wm8996_get_retune_mobile_block(kcontrol->id.name);
-	int value = ucontrol->value.integer.value[0];
+	int value = ucontrol->value.enumerated.item[0];
 
 	if (block < 0)
 		return block;
diff --git a/sound/soc/codecs/wm9081.c b/sound/soc/codecs/wm9081.c
index ccb3b15139ad..363b3b667616 100644
--- a/sound/soc/codecs/wm9081.c
+++ b/sound/soc/codecs/wm9081.c
@@ -344,9 +344,9 @@ static int speaker_mode_get(struct snd_kcontrol *kcontrol,
 
 	reg = snd_soc_read(codec, WM9081_ANALOGUE_SPEAKER_2);
 	if (reg & WM9081_SPK_MODE)
-		ucontrol->value.integer.value[0] = 1;
+		ucontrol->value.enumerated.item[0] = 1;
 	else
-		ucontrol->value.integer.value[0] = 0;
+		ucontrol->value.enumerated.item[0] = 0;
 
 	return 0;
 }
@@ -365,7 +365,7 @@ static int speaker_mode_put(struct snd_kcontrol *kcontrol,
 	unsigned int reg2 = snd_soc_read(codec, WM9081_ANALOGUE_SPEAKER_2);
 
 	/* Are we changing anything? */
-	if (ucontrol->value.integer.value[0] ==
+	if (ucontrol->value.enumerated.item[0] ==
 	    ((reg2 & WM9081_SPK_MODE) != 0))
 		return 0;
 
@@ -373,7 +373,7 @@ static int speaker_mode_put(struct snd_kcontrol *kcontrol,
 	if (reg_pwr & WM9081_SPK_ENA)
 		return -EINVAL;
 
-	if (ucontrol->value.integer.value[0]) {
+	if (ucontrol->value.enumerated.item[0]) {
 		/* Class AB */
 		reg2 &= ~(WM9081_SPK_INV_MUTE | WM9081_OUT_SPK_CTRL);
 		reg2 |= WM9081_SPK_MODE;
diff --git a/sound/soc/codecs/wm9713.c b/sound/soc/codecs/wm9713.c
index 79e143625ac3..9849643ef809 100644
--- a/sound/soc/codecs/wm9713.c
+++ b/sound/soc/codecs/wm9713.c
@@ -1212,7 +1212,7 @@ static int wm9713_soc_probe(struct snd_soc_codec *codec)
 	if (IS_ERR(wm9713->ac97))
 		return PTR_ERR(wm9713->ac97);
 
-	regmap = devm_regmap_init_ac97(wm9713->ac97, &wm9713_regmap_config);
+	regmap = regmap_init_ac97(wm9713->ac97, &wm9713_regmap_config);
 	if (IS_ERR(regmap)) {
 		snd_soc_free_ac97_codec(wm9713->ac97);
 		return PTR_ERR(regmap);
diff --git a/sound/soc/codecs/wm_adsp.c b/sound/soc/codecs/wm_adsp.c
index 33806d487b8a..b9195b9c2b05 100644
--- a/sound/soc/codecs/wm_adsp.c
+++ b/sound/soc/codecs/wm_adsp.c
@@ -586,7 +586,7 @@ static int wm_adsp_fw_get(struct snd_kcontrol *kcontrol,
 	struct soc_enum *e = (struct soc_enum *)kcontrol->private_value;
 	struct wm_adsp *dsp = snd_soc_codec_get_drvdata(codec);
 
-	ucontrol->value.integer.value[0] = dsp[e->shift_l].fw;
+	ucontrol->value.enumerated.item[0] = dsp[e->shift_l].fw;
 
 	return 0;
 }
@@ -599,10 +599,10 @@ static int wm_adsp_fw_put(struct snd_kcontrol *kcontrol,
 	struct wm_adsp *dsp = snd_soc_codec_get_drvdata(codec);
 	int ret = 0;
 
-	if (ucontrol->value.integer.value[0] == dsp[e->shift_l].fw)
+	if (ucontrol->value.enumerated.item[0] == dsp[e->shift_l].fw)
 		return 0;
 
-	if (ucontrol->value.integer.value[0] >= WM_ADSP_NUM_FW)
+	if (ucontrol->value.enumerated.item[0] >= WM_ADSP_NUM_FW)
 		return -EINVAL;
 
 	mutex_lock(&dsp[e->shift_l].pwr_lock);
@@ -610,7 +610,7 @@ static int wm_adsp_fw_put(struct snd_kcontrol *kcontrol,
 	if (dsp[e->shift_l].running || dsp[e->shift_l].compr)
 		ret = -EBUSY;
 	else
-		dsp[e->shift_l].fw = ucontrol->value.integer.value[0];
+		dsp[e->shift_l].fw = ucontrol->value.enumerated.item[0];
 
 	mutex_unlock(&dsp[e->shift_l].pwr_lock);
 
diff --git a/sound/soc/dwc/designware_i2s.c b/sound/soc/dwc/designware_i2s.c
index ce664c239be3..bff258d7bcea 100644
--- a/sound/soc/dwc/designware_i2s.c
+++ b/sound/soc/dwc/designware_i2s.c
@@ -645,6 +645,8 @@ static int dw_i2s_probe(struct platform_device *pdev)
 
 	dev->dev = &pdev->dev;
 
+	dev->i2s_reg_comp1 = I2S_COMP_PARAM_1;
+	dev->i2s_reg_comp2 = I2S_COMP_PARAM_2;
 	if (pdata) {
 		dev->capability = pdata->cap;
 		clk_id = NULL;
@@ -652,9 +654,6 @@ static int dw_i2s_probe(struct platform_device *pdev)
 		if (dev->quirks & DW_I2S_QUIRK_COMP_REG_OFFSET) {
 			dev->i2s_reg_comp1 = pdata->i2s_reg_comp1;
 			dev->i2s_reg_comp2 = pdata->i2s_reg_comp2;
-		} else {
-			dev->i2s_reg_comp1 = I2S_COMP_PARAM_1;
-			dev->i2s_reg_comp2 = I2S_COMP_PARAM_2;
 		}
 		ret = dw_configure_dai_by_pd(dev, dw_i2s_dai, res, pdata);
 	} else {
diff --git a/sound/soc/fsl/imx-spdif.c b/sound/soc/fsl/imx-spdif.c
index a407e833c612..fb896b2c9ba3 100644
--- a/sound/soc/fsl/imx-spdif.c
+++ b/sound/soc/fsl/imx-spdif.c
@@ -72,8 +72,6 @@ static int imx_spdif_audio_probe(struct platform_device *pdev)
 		goto end;
 	}
 
-	platform_set_drvdata(pdev, data);
-
 end:
 	of_node_put(spdif_np);
 
diff --git a/sound/soc/generic/simple-card.c b/sound/soc/generic/simple-card.c
index 1ded8811598e..2389ab47e25f 100644
--- a/sound/soc/generic/simple-card.c
+++ b/sound/soc/generic/simple-card.c
@@ -99,7 +99,7 @@ static int asoc_simple_card_hw_params(struct snd_pcm_substream *substream,
 		if (ret && ret != -ENOTSUPP)
 			goto err;
 	}
-
+	return 0;
 err:
 	return ret;
 }
diff --git a/sound/soc/intel/Kconfig b/sound/soc/intel/Kconfig
index 803f95e40679..7d7c872c280d 100644
--- a/sound/soc/intel/Kconfig
+++ b/sound/soc/intel/Kconfig
@@ -30,11 +30,15 @@ config SND_SST_IPC_ACPI
 config SND_SOC_INTEL_SST
 	tristate
 	select SND_SOC_INTEL_SST_ACPI if ACPI
+	select SND_SOC_INTEL_SST_MATCH if ACPI
 	depends on (X86 || COMPILE_TEST)
 
 config SND_SOC_INTEL_SST_ACPI
 	tristate
 
+config SND_SOC_INTEL_SST_MATCH
+	tristate
+
 config SND_SOC_INTEL_HASWELL
 	tristate
 
@@ -57,7 +61,7 @@ config SND_SOC_INTEL_HASWELL_MACH
 config SND_SOC_INTEL_BYT_RT5640_MACH
 	tristate "ASoC Audio driver for Intel Baytrail with RT5640 codec"
 	depends on X86_INTEL_LPSS && I2C
-	depends on DW_DMAC_CORE=y && (SND_SOC_INTEL_BYTCR_RT5640_MACH = n)
+	depends on DW_DMAC_CORE=y && (SND_SST_IPC_ACPI = n)
 	select SND_SOC_INTEL_SST
 	select SND_SOC_INTEL_BAYTRAIL
 	select SND_SOC_RT5640
@@ -69,7 +73,7 @@ config SND_SOC_INTEL_BYT_RT5640_MACH
 config SND_SOC_INTEL_BYT_MAX98090_MACH
 	tristate "ASoC Audio driver for Intel Baytrail with MAX98090 codec"
 	depends on X86_INTEL_LPSS && I2C
-	depends on DW_DMAC_CORE=y
+	depends on DW_DMAC_CORE=y && (SND_SST_IPC_ACPI = n)
 	select SND_SOC_INTEL_SST
 	select SND_SOC_INTEL_BAYTRAIL
 	select SND_SOC_MAX98090
@@ -97,6 +101,7 @@ config SND_SOC_INTEL_BYTCR_RT5640_MACH
 	select SND_SOC_RT5640
 	select SND_SST_MFLD_PLATFORM
 	select SND_SST_IPC_ACPI
+	select SND_SOC_INTEL_SST_MATCH if ACPI
 	help
           This adds support for ASoC machine driver for Intel(R) Baytrail and Baytrail-CR
           platforms with RT5640 audio codec.
@@ -109,6 +114,7 @@ config SND_SOC_INTEL_BYTCR_RT5651_MACH
 	select SND_SOC_RT5651
 	select SND_SST_MFLD_PLATFORM
 	select SND_SST_IPC_ACPI
+	select SND_SOC_INTEL_SST_MATCH if ACPI
 	help
           This adds support for ASoC machine driver for Intel(R) Baytrail and Baytrail-CR
           platforms with RT5651 audio codec.
@@ -121,6 +127,7 @@ config SND_SOC_INTEL_CHT_BSW_RT5672_MACH
         select SND_SOC_RT5670
         select SND_SST_MFLD_PLATFORM
         select SND_SST_IPC_ACPI
+	select SND_SOC_INTEL_SST_MATCH if ACPI
         help
           This adds support for ASoC machine driver for Intel(R) Cherrytrail & Braswell
           platforms with RT5672 audio codec.
@@ -133,6 +140,7 @@ config SND_SOC_INTEL_CHT_BSW_RT5645_MACH
 	select SND_SOC_RT5645
 	select SND_SST_MFLD_PLATFORM
 	select SND_SST_IPC_ACPI
+	select SND_SOC_INTEL_SST_MATCH if ACPI
 	help
 	  This adds support for ASoC machine driver for Intel(R) Cherrytrail & Braswell
 	  platforms with RT5645/5650 audio codec.
@@ -145,6 +153,7 @@ config SND_SOC_INTEL_CHT_BSW_MAX98090_TI_MACH
 	select SND_SOC_TS3A227E
 	select SND_SST_MFLD_PLATFORM
 	select SND_SST_IPC_ACPI
+	select SND_SOC_INTEL_SST_MATCH if ACPI
 	help
       This adds support for ASoC machine driver for Intel(R) Cherrytrail & Braswell
       platforms with MAX98090 audio codec it also can support TI jack chip as aux device.
diff --git a/sound/soc/intel/atom/sst-mfld-platform-pcm.c b/sound/soc/intel/atom/sst-mfld-platform-pcm.c
index 55c33dc76ce4..52ed434cbca6 100644
--- a/sound/soc/intel/atom/sst-mfld-platform-pcm.c
+++ b/sound/soc/intel/atom/sst-mfld-platform-pcm.c
@@ -528,6 +528,7 @@ static struct snd_soc_dai_driver sst_platform_dai[] = {
 	.ops = &sst_compr_dai_ops,
 	.playback = {
 		.stream_name = "Compress Playback",
+		.channels_min = 1,
 	},
 },
 /* BE CPU  Dais */
diff --git a/sound/soc/intel/boards/cht_bsw_rt5645.c b/sound/soc/intel/boards/cht_bsw_rt5645.c
index 2d3afddb0a2e..a7b96a9a4e0e 100644
--- a/sound/soc/intel/boards/cht_bsw_rt5645.c
+++ b/sound/soc/intel/boards/cht_bsw_rt5645.c
@@ -367,8 +367,12 @@ static int snd_cht_mc_probe(struct platform_device *pdev)
 	}
 	card->dev = &pdev->dev;
 	sprintf(codec_name, "i2c-%s:00", drv->acpi_card->codec_id);
+
 	/* set correct codec name */
-	strcpy((char *)card->dai_link[2].codec_name, codec_name);
+	for (i = 0; i < ARRAY_SIZE(cht_dailink); i++)
+		if (!strcmp(card->dai_link[i].codec_name, "i2c-10EC5645:00"))
+			card->dai_link[i].codec_name = kstrdup(codec_name, GFP_KERNEL);
+
 	snd_soc_card_set_drvdata(card, drv);
 	ret_val = devm_snd_soc_register_card(&pdev->dev, card);
 	if (ret_val) {
diff --git a/sound/soc/intel/boards/mfld_machine.c b/sound/soc/intel/boards/mfld_machine.c
index 49c09a0add79..34f46c72a0e2 100644
--- a/sound/soc/intel/boards/mfld_machine.c
+++ b/sound/soc/intel/boards/mfld_machine.c
@@ -94,7 +94,7 @@ static const struct soc_enum lo_enum =
 static int headset_get_switch(struct snd_kcontrol *kcontrol,
 	struct snd_ctl_elem_value *ucontrol)
 {
-	ucontrol->value.integer.value[0] = hs_switch;
+	ucontrol->value.enumerated.item[0] = hs_switch;
 	return 0;
 }
 
@@ -104,12 +104,12 @@ static int headset_set_switch(struct snd_kcontrol *kcontrol,
 	struct snd_soc_card *card = snd_kcontrol_chip(kcontrol);
 	struct snd_soc_dapm_context *dapm = &card->dapm;
 
-	if (ucontrol->value.integer.value[0] == hs_switch)
+	if (ucontrol->value.enumerated.item[0] == hs_switch)
 		return 0;
 
 	snd_soc_dapm_mutex_lock(dapm);
 
-	if (ucontrol->value.integer.value[0]) {
+	if (ucontrol->value.enumerated.item[0]) {
 		pr_debug("hs_set HS path\n");
 		snd_soc_dapm_enable_pin_unlocked(dapm, "Headphones");
 		snd_soc_dapm_disable_pin_unlocked(dapm, "EPOUT");
@@ -123,7 +123,7 @@ static int headset_set_switch(struct snd_kcontrol *kcontrol,
 
 	snd_soc_dapm_mutex_unlock(dapm);
 
-	hs_switch = ucontrol->value.integer.value[0];
+	hs_switch = ucontrol->value.enumerated.item[0];
 
 	return 0;
 }
@@ -148,7 +148,7 @@ static void lo_enable_out_pins(struct snd_soc_dapm_context *dapm)
 static int lo_get_switch(struct snd_kcontrol *kcontrol,
 	struct snd_ctl_elem_value *ucontrol)
 {
-	ucontrol->value.integer.value[0] = lo_dac;
+	ucontrol->value.enumerated.item[0] = lo_dac;
 	return 0;
 }
 
@@ -158,7 +158,7 @@ static int lo_set_switch(struct snd_kcontrol *kcontrol,
 	struct snd_soc_card *card = snd_kcontrol_chip(kcontrol);
 	struct snd_soc_dapm_context *dapm = &card->dapm;
 
-	if (ucontrol->value.integer.value[0] == lo_dac)
+	if (ucontrol->value.enumerated.item[0] == lo_dac)
 		return 0;
 
 	snd_soc_dapm_mutex_lock(dapm);
@@ -168,7 +168,7 @@ static int lo_set_switch(struct snd_kcontrol *kcontrol,
 	 */
 	lo_enable_out_pins(dapm);
 
-	switch (ucontrol->value.integer.value[0]) {
+	switch (ucontrol->value.enumerated.item[0]) {
 	case 0:
 		pr_debug("set vibra path\n");
 		snd_soc_dapm_disable_pin_unlocked(dapm, "VIB1OUT");
@@ -202,7 +202,7 @@ static int lo_set_switch(struct snd_kcontrol *kcontrol,
 
 	snd_soc_dapm_mutex_unlock(dapm);
 
-	lo_dac = ucontrol->value.integer.value[0];
+	lo_dac = ucontrol->value.enumerated.item[0];
 	return 0;
 }
 
diff --git a/sound/soc/intel/boards/skl_rt286.c b/sound/soc/intel/boards/skl_rt286.c
index 7396ddb427d8..2cbcbe412661 100644
--- a/sound/soc/intel/boards/skl_rt286.c
+++ b/sound/soc/intel/boards/skl_rt286.c
@@ -212,7 +212,10 @@ static int skylake_dmic_fixup(struct snd_soc_pcm_runtime *rtd,
 {
 	struct snd_interval *channels = hw_param_interval(params,
 						SNDRV_PCM_HW_PARAM_CHANNELS);
-	channels->min = channels->max = 4;
+	if (params_channels(params) == 2)
+		channels->min = channels->max = 2;
+	else
+		channels->min = channels->max = 4;
 
 	return 0;
 }
diff --git a/sound/soc/intel/common/Makefile b/sound/soc/intel/common/Makefile
index 668fdeee195e..fbbb25c2ceed 100644
--- a/sound/soc/intel/common/Makefile
+++ b/sound/soc/intel/common/Makefile
@@ -1,13 +1,10 @@
 snd-soc-sst-dsp-objs := sst-dsp.o
-ifneq ($(CONFIG_SND_SST_IPC_ACPI),)
-snd-soc-sst-acpi-objs := sst-match-acpi.o
-else
-snd-soc-sst-acpi-objs := sst-acpi.o sst-match-acpi.o
-endif
-
+snd-soc-sst-acpi-objs := sst-acpi.o
+snd-soc-sst-match-objs := sst-match-acpi.o
 snd-soc-sst-ipc-objs := sst-ipc.o
 
 snd-soc-sst-dsp-$(CONFIG_DW_DMAC_CORE) += sst-firmware.o
 
 obj-$(CONFIG_SND_SOC_INTEL_SST) += snd-soc-sst-dsp.o snd-soc-sst-ipc.o
 obj-$(CONFIG_SND_SOC_INTEL_SST_ACPI) += snd-soc-sst-acpi.o
+obj-$(CONFIG_SND_SOC_INTEL_SST_MATCH) += snd-soc-sst-match.o
diff --git a/sound/soc/intel/common/sst-acpi.c b/sound/soc/intel/common/sst-acpi.c
index 7a85c576dad3..2c5eda14d510 100644
--- a/sound/soc/intel/common/sst-acpi.c
+++ b/sound/soc/intel/common/sst-acpi.c
@@ -215,6 +215,7 @@ static struct sst_acpi_desc sst_acpi_broadwell_desc = {
 	.dma_size = SST_LPT_DSP_DMA_SIZE,
 };
 
+#if !IS_ENABLED(CONFIG_SND_SST_IPC_ACPI)
 static struct sst_acpi_mach baytrail_machines[] = {
 	{ "10EC5640", "byt-rt5640", "intel/fw_sst_0f28.bin-48kHz_i2s_master", NULL, NULL, NULL },
 	{ "193C9890", "byt-max98090", "intel/fw_sst_0f28.bin-48kHz_i2s_master", NULL, NULL, NULL },
@@ -231,11 +232,14 @@ static struct sst_acpi_desc sst_acpi_baytrail_desc = {
 	.sst_id = SST_DEV_ID_BYT,
 	.resindex_dma_base = -1,
 };
+#endif
 
 static const struct acpi_device_id sst_acpi_match[] = {
 	{ "INT33C8", (unsigned long)&sst_acpi_haswell_desc },
 	{ "INT3438", (unsigned long)&sst_acpi_broadwell_desc },
+#if !IS_ENABLED(CONFIG_SND_SST_IPC_ACPI)
 	{ "80860F28", (unsigned long)&sst_acpi_baytrail_desc },
+#endif
 	{ }
 };
 MODULE_DEVICE_TABLE(acpi, sst_acpi_match);
diff --git a/sound/soc/intel/common/sst-match-acpi.c b/sound/soc/intel/common/sst-match-acpi.c
index dd077e116d25..3b4539d21492 100644
--- a/sound/soc/intel/common/sst-match-acpi.c
+++ b/sound/soc/intel/common/sst-match-acpi.c
@@ -41,3 +41,6 @@ struct sst_acpi_mach *sst_acpi_find_machine(struct sst_acpi_mach *machines)
 	return NULL;
 }
 EXPORT_SYMBOL_GPL(sst_acpi_find_machine);
+
+MODULE_LICENSE("GPL v2");
+MODULE_DESCRIPTION("Intel Common ACPI Match module");
diff --git a/sound/soc/intel/skylake/skl-messages.c b/sound/soc/intel/skylake/skl-messages.c
index de6dac496a0d..4629372d7c8e 100644
--- a/sound/soc/intel/skylake/skl-messages.c
+++ b/sound/soc/intel/skylake/skl-messages.c
@@ -688,14 +688,14 @@ int skl_unbind_modules(struct skl_sst *ctx,
 	/* get src queue index */
 	src_index = skl_get_queue_index(src_mcfg->m_out_pin, dst_id, out_max);
 	if (src_index < 0)
-		return -EINVAL;
+		return 0;
 
 	msg.src_queue = src_index;
 
 	/* get dst queue index */
 	dst_index  = skl_get_queue_index(dst_mcfg->m_in_pin, src_id, in_max);
 	if (dst_index < 0)
-		return -EINVAL;
+		return 0;
 
 	msg.dst_queue = dst_index;
 
@@ -747,7 +747,7 @@ int skl_bind_modules(struct skl_sst *ctx,
 
 	skl_dump_bind_info(ctx, src_mcfg, dst_mcfg);
 
-	if (src_mcfg->m_state < SKL_MODULE_INIT_DONE &&
+	if (src_mcfg->m_state < SKL_MODULE_INIT_DONE ||
 		dst_mcfg->m_state < SKL_MODULE_INIT_DONE)
 		return 0;
 
diff --git a/sound/soc/intel/skylake/skl-pcm.c b/sound/soc/intel/skylake/skl-pcm.c
index f3553258091a..b6e6b61d10ec 100644
--- a/sound/soc/intel/skylake/skl-pcm.c
+++ b/sound/soc/intel/skylake/skl-pcm.c
@@ -863,6 +863,7 @@ static int skl_get_delay_from_lpib(struct hdac_ext_bus *ebus,
 		else
 			delay += hstream->bufsize;
 	}
+	delay = (hstream->bufsize == delay) ? 0 : delay;
 
 	if (delay >= hstream->period_bytes) {
 		dev_info(bus->dev,
diff --git a/sound/soc/intel/skylake/skl-topology.c b/sound/soc/intel/skylake/skl-topology.c
index 4624556f486d..5a4837dcfce3 100644
--- a/sound/soc/intel/skylake/skl-topology.c
+++ b/sound/soc/intel/skylake/skl-topology.c
@@ -54,12 +54,9 @@ static int is_skl_dsp_widget_type(struct snd_soc_dapm_widget *w)
 
 /*
  * Each pipelines needs memory to be allocated. Check if we have free memory
- * from available pool. Then only add this to pool
- * This is freed when pipe is deleted
- * Note: DSP does actual memory management we only keep track for complete
- * pool
+ * from available pool.
  */
-static bool skl_tplg_alloc_pipe_mem(struct skl *skl,
+static bool skl_is_pipe_mem_avail(struct skl *skl,
 				struct skl_module_cfg *mconfig)
 {
 	struct skl_sst *ctx = skl->skl_sst;
@@ -74,10 +71,20 @@ static bool skl_tplg_alloc_pipe_mem(struct skl *skl,
 				"exceeds ppl memory available %d mem %d\n",
 				skl->resource.max_mem, skl->resource.mem);
 		return false;
+	} else {
+		return true;
 	}
+}
 
+/*
+ * Add the mem to the mem pool. This is freed when pipe is deleted.
+ * Note: DSP does actual memory management we only keep track for complete
+ * pool
+ */
+static void skl_tplg_alloc_pipe_mem(struct skl *skl,
+				struct skl_module_cfg *mconfig)
+{
 	skl->resource.mem += mconfig->pipe->memory_pages;
-	return true;
 }
 
 /*
@@ -85,10 +92,10 @@ static bool skl_tplg_alloc_pipe_mem(struct skl *skl,
  * quantified in MCPS (Million Clocks Per Second) required for module/pipe
  *
  * Each pipelines needs mcps to be allocated. Check if we have mcps for this
- * pipe. This adds the mcps to driver counter
- * This is removed on pipeline delete
+ * pipe.
  */
-static bool skl_tplg_alloc_pipe_mcps(struct skl *skl,
+
+static bool skl_is_pipe_mcps_avail(struct skl *skl,
 				struct skl_module_cfg *mconfig)
 {
 	struct skl_sst *ctx = skl->skl_sst;
@@ -98,13 +105,18 @@ static bool skl_tplg_alloc_pipe_mcps(struct skl *skl,
 			"%s: module_id %d instance %d\n", __func__,
 			mconfig->id.module_id, mconfig->id.instance_id);
 		dev_err(ctx->dev,
-			"exceeds ppl memory available %d > mem %d\n",
+			"exceeds ppl mcps available %d > mem %d\n",
 			skl->resource.max_mcps, skl->resource.mcps);
 		return false;
+	} else {
+		return true;
 	}
+}
 
+static void skl_tplg_alloc_pipe_mcps(struct skl *skl,
+				struct skl_module_cfg *mconfig)
+{
 	skl->resource.mcps += mconfig->mcps;
-	return true;
 }
 
 /*
@@ -411,7 +423,7 @@ skl_tplg_init_pipe_modules(struct skl *skl, struct skl_pipe *pipe)
 		mconfig = w->priv;
 
 		/* check resource available */
-		if (!skl_tplg_alloc_pipe_mcps(skl, mconfig))
+		if (!skl_is_pipe_mcps_avail(skl, mconfig))
 			return -ENOMEM;
 
 		if (mconfig->is_loadable && ctx->dsp->fw_ops.load_mod) {
@@ -435,6 +447,7 @@ skl_tplg_init_pipe_modules(struct skl *skl, struct skl_pipe *pipe)
 		ret = skl_tplg_set_module_params(w, ctx);
 		if (ret < 0)
 			return ret;
+		skl_tplg_alloc_pipe_mcps(skl, mconfig);
 	}
 
 	return 0;
@@ -477,10 +490,10 @@ static int skl_tplg_mixer_dapm_pre_pmu_event(struct snd_soc_dapm_widget *w,
 	struct skl_sst *ctx = skl->skl_sst;
 
 	/* check resource available */
-	if (!skl_tplg_alloc_pipe_mcps(skl, mconfig))
+	if (!skl_is_pipe_mcps_avail(skl, mconfig))
 		return -EBUSY;
 
-	if (!skl_tplg_alloc_pipe_mem(skl, mconfig))
+	if (!skl_is_pipe_mem_avail(skl, mconfig))
 		return -ENOMEM;
 
 	/*
@@ -526,11 +539,15 @@ static int skl_tplg_mixer_dapm_pre_pmu_event(struct snd_soc_dapm_widget *w,
 		src_module = dst_module;
 	}
 
+	skl_tplg_alloc_pipe_mem(skl, mconfig);
+	skl_tplg_alloc_pipe_mcps(skl, mconfig);
+
 	return 0;
 }
 
 static int skl_tplg_bind_sinks(struct snd_soc_dapm_widget *w,
 				struct skl *skl,
+				struct snd_soc_dapm_widget *src_w,
 				struct skl_module_cfg *src_mconfig)
 {
 	struct snd_soc_dapm_path *p;
@@ -547,6 +564,10 @@ static int skl_tplg_bind_sinks(struct snd_soc_dapm_widget *w,
 		dev_dbg(ctx->dev, "%s: sink widget=%s\n", __func__, p->sink->name);
 
 		next_sink = p->sink;
+
+		if (!is_skl_dsp_widget_type(p->sink))
+			return skl_tplg_bind_sinks(p->sink, skl, src_w, src_mconfig);
+
 		/*
 		 * here we will check widgets in sink pipelines, so that
 		 * can be any widgets type and we are only interested if
@@ -576,7 +597,7 @@ static int skl_tplg_bind_sinks(struct snd_soc_dapm_widget *w,
 	}
 
 	if (!sink)
-		return skl_tplg_bind_sinks(next_sink, skl, src_mconfig);
+		return skl_tplg_bind_sinks(next_sink, skl, src_w, src_mconfig);
 
 	return 0;
 }
@@ -605,7 +626,7 @@ static int skl_tplg_pga_dapm_pre_pmu_event(struct snd_soc_dapm_widget *w,
 	 * if sink is not started, start sink pipe first, then start
 	 * this pipe
 	 */
-	ret = skl_tplg_bind_sinks(w, skl, src_mconfig);
+	ret = skl_tplg_bind_sinks(w, skl, w, src_mconfig);
 	if (ret)
 		return ret;
 
@@ -773,10 +794,7 @@ static int skl_tplg_mixer_dapm_post_pmd_event(struct snd_soc_dapm_widget *w,
 			continue;
 		}
 
-		ret = skl_unbind_modules(ctx, src_module, dst_module);
-		if (ret < 0)
-			return ret;
-
+		skl_unbind_modules(ctx, src_module, dst_module);
 		src_module = dst_module;
 	}
 
@@ -814,9 +832,6 @@ static int skl_tplg_pga_dapm_post_pmd_event(struct snd_soc_dapm_widget *w,
 			 * This is a connecter and if path is found that means
 			 * unbind between source and sink has not happened yet
 			 */
-			ret = skl_stop_pipe(ctx, sink_mconfig->pipe);
-			if (ret < 0)
-				return ret;
 			ret = skl_unbind_modules(ctx, src_mconfig,
 							sink_mconfig);
 		}
@@ -842,6 +857,12 @@ static int skl_tplg_vmixer_event(struct snd_soc_dapm_widget *w,
 	case SND_SOC_DAPM_PRE_PMU:
 		return skl_tplg_mixer_dapm_pre_pmu_event(w, skl);
 
+	case SND_SOC_DAPM_POST_PMU:
+		return skl_tplg_mixer_dapm_post_pmu_event(w, skl);
+
+	case SND_SOC_DAPM_PRE_PMD:
+		return skl_tplg_mixer_dapm_pre_pmd_event(w, skl);
+
 	case SND_SOC_DAPM_POST_PMD:
 		return skl_tplg_mixer_dapm_post_pmd_event(w, skl);
 	}
@@ -916,6 +937,13 @@ static int skl_tplg_tlv_control_get(struct snd_kcontrol *kcontrol,
 		skl_get_module_params(skl->skl_sst, (u32 *)bc->params,
 				      bc->max, bc->param_id, mconfig);
 
+	/* decrement size for TLV header */
+	size -= 2 * sizeof(u32);
+
+	/* check size as we don't want to send kernel data */
+	if (size > bc->max)
+		size = bc->max;
+
 	if (bc->params) {
 		if (copy_to_user(data, &bc->param_id, sizeof(u32)))
 			return -EFAULT;
@@ -950,7 +978,7 @@ static int skl_tplg_tlv_control_set(struct snd_kcontrol *kcontrol,
 				return -EFAULT;
 		} else {
 			if (copy_from_user(ac->params,
-					   data + 2 * sizeof(u32), size))
+					   data + 2, size))
 				return -EFAULT;
 		}
 
@@ -1510,6 +1538,7 @@ int skl_tplg_init(struct snd_soc_platform *platform, struct hdac_ext_bus *ebus)
 					&skl_tplg_ops, fw, 0);
 	if (ret < 0) {
 		dev_err(bus->dev, "tplg component load failed%d\n", ret);
+		release_firmware(fw);
 		return -EINVAL;
 	}
 
diff --git a/sound/soc/intel/skylake/skl.c b/sound/soc/intel/skylake/skl.c
index 443a15de94b5..092705e73db4 100644
--- a/sound/soc/intel/skylake/skl.c
+++ b/sound/soc/intel/skylake/skl.c
@@ -614,8 +614,6 @@ static int skl_probe(struct pci_dev *pci,
 		goto out_unregister;
 
 	/*configure PM */
-	pm_runtime_set_autosuspend_delay(bus->dev, SKL_SUSPEND_DELAY);
-	pm_runtime_use_autosuspend(bus->dev);
 	pm_runtime_put_noidle(bus->dev);
 	pm_runtime_allow(bus->dev);
 
diff --git a/sound/soc/mediatek/Kconfig b/sound/soc/mediatek/Kconfig
index 15c04e2eae34..976967675387 100644
--- a/sound/soc/mediatek/Kconfig
+++ b/sound/soc/mediatek/Kconfig
@@ -9,7 +9,7 @@ config SND_SOC_MEDIATEK
 
 config SND_SOC_MT8173_MAX98090
 	tristate "ASoC Audio driver for MT8173 with MAX98090 codec"
-	depends on SND_SOC_MEDIATEK
+	depends on SND_SOC_MEDIATEK && I2C
 	select SND_SOC_MAX98090
 	help
 	  This adds ASoC driver for Mediatek MT8173 boards
@@ -19,7 +19,7 @@ config SND_SOC_MT8173_MAX98090
 
 config SND_SOC_MT8173_RT5650_RT5676
 	tristate "ASoC Audio driver for MT8173 with RT5650 RT5676 codecs"
-	depends on SND_SOC_MEDIATEK
+	depends on SND_SOC_MEDIATEK && I2C
 	select SND_SOC_RT5645
 	select SND_SOC_RT5677
 	help
diff --git a/sound/soc/mxs/mxs-saif.c b/sound/soc/mxs/mxs-saif.c
index c866ade28ad0..a6c7b8d87cd2 100644
--- a/sound/soc/mxs/mxs-saif.c
+++ b/sound/soc/mxs/mxs-saif.c
@@ -381,9 +381,19 @@ static int mxs_saif_startup(struct snd_pcm_substream *substream,
 	__raw_writel(BM_SAIF_CTRL_CLKGATE,
 		saif->base + SAIF_CTRL + MXS_CLR_ADDR);
 
+	clk_prepare(saif->clk);
+
 	return 0;
 }
 
+static void mxs_saif_shutdown(struct snd_pcm_substream *substream,
+			      struct snd_soc_dai *cpu_dai)
+{
+	struct mxs_saif *saif = snd_soc_dai_get_drvdata(cpu_dai);
+
+	clk_unprepare(saif->clk);
+}
+
 /*
  * Should only be called when port is inactive.
  * although can be called multiple times by upper layers.
@@ -424,8 +434,6 @@ static int mxs_saif_hw_params(struct snd_pcm_substream *substream,
 		return ret;
 	}
 
-	/* prepare clk in hw_param, enable in trigger */
-	clk_prepare(saif->clk);
 	if (saif != master_saif) {
 		/*
 		* Set an initial clock rate for the saif internal logic to work
@@ -611,6 +619,7 @@ static int mxs_saif_trigger(struct snd_pcm_substream *substream, int cmd,
 
 static const struct snd_soc_dai_ops mxs_saif_dai_ops = {
 	.startup = mxs_saif_startup,
+	.shutdown = mxs_saif_shutdown,
 	.trigger = mxs_saif_trigger,
 	.prepare = mxs_saif_prepare,
 	.hw_params = mxs_saif_hw_params,
diff --git a/sound/soc/omap/n810.c b/sound/soc/omap/n810.c
index 190f868e78b2..fdecb7043174 100644
--- a/sound/soc/omap/n810.c
+++ b/sound/soc/omap/n810.c
@@ -133,7 +133,7 @@ static struct snd_soc_ops n810_ops = {
 static int n810_get_spk(struct snd_kcontrol *kcontrol,
 			struct snd_ctl_elem_value *ucontrol)
 {
-	ucontrol->value.integer.value[0] = n810_spk_func;
+	ucontrol->value.enumerated.item[0] = n810_spk_func;
 
 	return 0;
 }
@@ -143,10 +143,10 @@ static int n810_set_spk(struct snd_kcontrol *kcontrol,
 {
 	struct snd_soc_card *card =  snd_kcontrol_chip(kcontrol);
 
-	if (n810_spk_func == ucontrol->value.integer.value[0])
+	if (n810_spk_func == ucontrol->value.enumerated.item[0])
 		return 0;
 
-	n810_spk_func = ucontrol->value.integer.value[0];
+	n810_spk_func = ucontrol->value.enumerated.item[0];
 	n810_ext_control(&card->dapm);
 
 	return 1;
@@ -155,7 +155,7 @@ static int n810_set_spk(struct snd_kcontrol *kcontrol,
 static int n810_get_jack(struct snd_kcontrol *kcontrol,
 			 struct snd_ctl_elem_value *ucontrol)
 {
-	ucontrol->value.integer.value[0] = n810_jack_func;
+	ucontrol->value.enumerated.item[0] = n810_jack_func;
 
 	return 0;
 }
@@ -165,10 +165,10 @@ static int n810_set_jack(struct snd_kcontrol *kcontrol,
 {
 	struct snd_soc_card *card =  snd_kcontrol_chip(kcontrol);
 
-	if (n810_jack_func == ucontrol->value.integer.value[0])
+	if (n810_jack_func == ucontrol->value.enumerated.item[0])
 		return 0;
 
-	n810_jack_func = ucontrol->value.integer.value[0];
+	n810_jack_func = ucontrol->value.enumerated.item[0];
 	n810_ext_control(&card->dapm);
 
 	return 1;
@@ -177,7 +177,7 @@ static int n810_set_jack(struct snd_kcontrol *kcontrol,
 static int n810_get_input(struct snd_kcontrol *kcontrol,
 			  struct snd_ctl_elem_value *ucontrol)
 {
-	ucontrol->value.integer.value[0] = n810_dmic_func;
+	ucontrol->value.enumerated.item[0] = n810_dmic_func;
 
 	return 0;
 }
@@ -187,10 +187,10 @@ static int n810_set_input(struct snd_kcontrol *kcontrol,
 {
 	struct snd_soc_card *card =  snd_kcontrol_chip(kcontrol);
 
-	if (n810_dmic_func == ucontrol->value.integer.value[0])
+	if (n810_dmic_func == ucontrol->value.enumerated.item[0])
 		return 0;
 
-	n810_dmic_func = ucontrol->value.integer.value[0];
+	n810_dmic_func = ucontrol->value.enumerated.item[0];
 	n810_ext_control(&card->dapm);
 
 	return 1;
diff --git a/sound/soc/omap/rx51.c b/sound/soc/omap/rx51.c
index 5e21f08579d8..54949242bc70 100644
--- a/sound/soc/omap/rx51.c
+++ b/sound/soc/omap/rx51.c
@@ -132,7 +132,7 @@ static struct snd_soc_ops rx51_ops = {
 static int rx51_get_spk(struct snd_kcontrol *kcontrol,
 			struct snd_ctl_elem_value *ucontrol)
 {
-	ucontrol->value.integer.value[0] = rx51_spk_func;
+	ucontrol->value.enumerated.item[0] = rx51_spk_func;
 
 	return 0;
 }
@@ -142,10 +142,10 @@ static int rx51_set_spk(struct snd_kcontrol *kcontrol,
 {
 	struct snd_soc_card *card = snd_kcontrol_chip(kcontrol);
 
-	if (rx51_spk_func == ucontrol->value.integer.value[0])
+	if (rx51_spk_func == ucontrol->value.enumerated.item[0])
 		return 0;
 
-	rx51_spk_func = ucontrol->value.integer.value[0];
+	rx51_spk_func = ucontrol->value.enumerated.item[0];
 	rx51_ext_control(&card->dapm);
 
 	return 1;
@@ -180,7 +180,7 @@ static int rx51_hp_event(struct snd_soc_dapm_widget *w,
 static int rx51_get_input(struct snd_kcontrol *kcontrol,
 			  struct snd_ctl_elem_value *ucontrol)
 {
-	ucontrol->value.integer.value[0] = rx51_dmic_func;
+	ucontrol->value.enumerated.item[0] = rx51_dmic_func;
 
 	return 0;
 }
@@ -190,10 +190,10 @@ static int rx51_set_input(struct snd_kcontrol *kcontrol,
 {
 	struct snd_soc_card *card = snd_kcontrol_chip(kcontrol);
 
-	if (rx51_dmic_func == ucontrol->value.integer.value[0])
+	if (rx51_dmic_func == ucontrol->value.enumerated.item[0])
 		return 0;
 
-	rx51_dmic_func = ucontrol->value.integer.value[0];
+	rx51_dmic_func = ucontrol->value.enumerated.item[0];
 	rx51_ext_control(&card->dapm);
 
 	return 1;
@@ -202,7 +202,7 @@ static int rx51_set_input(struct snd_kcontrol *kcontrol,
 static int rx51_get_jack(struct snd_kcontrol *kcontrol,
 			 struct snd_ctl_elem_value *ucontrol)
 {
-	ucontrol->value.integer.value[0] = rx51_jack_func;
+	ucontrol->value.enumerated.item[0] = rx51_jack_func;
 
 	return 0;
 }
@@ -212,10 +212,10 @@ static int rx51_set_jack(struct snd_kcontrol *kcontrol,
 {
 	struct snd_soc_card *card = snd_kcontrol_chip(kcontrol);
 
-	if (rx51_jack_func == ucontrol->value.integer.value[0])
+	if (rx51_jack_func == ucontrol->value.enumerated.item[0])
 		return 0;
 
-	rx51_jack_func = ucontrol->value.integer.value[0];
+	rx51_jack_func = ucontrol->value.enumerated.item[0];
 	rx51_ext_control(&card->dapm);
 
 	return 1;
diff --git a/sound/soc/pxa/corgi.c b/sound/soc/pxa/corgi.c
index c97dc13d3608..dcbb7aa9830c 100644
--- a/sound/soc/pxa/corgi.c
+++ b/sound/soc/pxa/corgi.c
@@ -163,7 +163,7 @@ static struct snd_soc_ops corgi_ops = {
 static int corgi_get_jack(struct snd_kcontrol *kcontrol,
 	struct snd_ctl_elem_value *ucontrol)
 {
-	ucontrol->value.integer.value[0] = corgi_jack_func;
+	ucontrol->value.enumerated.item[0] = corgi_jack_func;
 	return 0;
 }
 
@@ -172,10 +172,10 @@ static int corgi_set_jack(struct snd_kcontrol *kcontrol,
 {
 	struct snd_soc_card *card = snd_kcontrol_chip(kcontrol);
 
-	if (corgi_jack_func == ucontrol->value.integer.value[0])
+	if (corgi_jack_func == ucontrol->value.enumerated.item[0])
 		return 0;
 
-	corgi_jack_func = ucontrol->value.integer.value[0];
+	corgi_jack_func = ucontrol->value.enumerated.item[0];
 	corgi_ext_control(&card->dapm);
 	return 1;
 }
@@ -183,7 +183,7 @@ static int corgi_set_jack(struct snd_kcontrol *kcontrol,
 static int corgi_get_spk(struct snd_kcontrol *kcontrol,
 	struct snd_ctl_elem_value *ucontrol)
 {
-	ucontrol->value.integer.value[0] = corgi_spk_func;
+	ucontrol->value.enumerated.item[0] = corgi_spk_func;
 	return 0;
 }
 
@@ -192,10 +192,10 @@ static int corgi_set_spk(struct snd_kcontrol *kcontrol,
 {
 	struct snd_soc_card *card =  snd_kcontrol_chip(kcontrol);
 
-	if (corgi_spk_func == ucontrol->value.integer.value[0])
+	if (corgi_spk_func == ucontrol->value.enumerated.item[0])
 		return 0;
 
-	corgi_spk_func = ucontrol->value.integer.value[0];
+	corgi_spk_func = ucontrol->value.enumerated.item[0];
 	corgi_ext_control(&card->dapm);
 	return 1;
 }
diff --git a/sound/soc/pxa/magician.c b/sound/soc/pxa/magician.c
index 241d0be42d7a..62b8377a9d2b 100644
--- a/sound/soc/pxa/magician.c
+++ b/sound/soc/pxa/magician.c
@@ -308,17 +308,17 @@ static int magician_set_spk(struct snd_kcontrol *kcontrol,
 static int magician_get_input(struct snd_kcontrol *kcontrol,
 			      struct snd_ctl_elem_value *ucontrol)
 {
-	ucontrol->value.integer.value[0] = magician_in_sel;
+	ucontrol->value.enumerated.item[0] = magician_in_sel;
 	return 0;
 }
 
 static int magician_set_input(struct snd_kcontrol *kcontrol,
 			      struct snd_ctl_elem_value *ucontrol)
 {
-	if (magician_in_sel == ucontrol->value.integer.value[0])
+	if (magician_in_sel == ucontrol->value.enumerated.item[0])
 		return 0;
 
-	magician_in_sel = ucontrol->value.integer.value[0];
+	magician_in_sel = ucontrol->value.enumerated.item[0];
 
 	switch (magician_in_sel) {
 	case MAGICIAN_MIC:
diff --git a/sound/soc/pxa/poodle.c b/sound/soc/pxa/poodle.c
index 84d0e2e50808..4b3b714f5ee7 100644
--- a/sound/soc/pxa/poodle.c
+++ b/sound/soc/pxa/poodle.c
@@ -138,7 +138,7 @@ static struct snd_soc_ops poodle_ops = {
 static int poodle_get_jack(struct snd_kcontrol *kcontrol,
 	struct snd_ctl_elem_value *ucontrol)
 {
-	ucontrol->value.integer.value[0] = poodle_jack_func;
+	ucontrol->value.enumerated.item[0] = poodle_jack_func;
 	return 0;
 }
 
@@ -147,10 +147,10 @@ static int poodle_set_jack(struct snd_kcontrol *kcontrol,
 {
 	struct snd_soc_card *card =  snd_kcontrol_chip(kcontrol);
 
-	if (poodle_jack_func == ucontrol->value.integer.value[0])
+	if (poodle_jack_func == ucontrol->value.enumerated.item[0])
 		return 0;
 
-	poodle_jack_func = ucontrol->value.integer.value[0];
+	poodle_jack_func = ucontrol->value.enumerated.item[0];
 	poodle_ext_control(&card->dapm);
 	return 1;
 }
@@ -158,7 +158,7 @@ static int poodle_set_jack(struct snd_kcontrol *kcontrol,
 static int poodle_get_spk(struct snd_kcontrol *kcontrol,
 	struct snd_ctl_elem_value *ucontrol)
 {
-	ucontrol->value.integer.value[0] = poodle_spk_func;
+	ucontrol->value.enumerated.item[0] = poodle_spk_func;
 	return 0;
 }
 
@@ -167,10 +167,10 @@ static int poodle_set_spk(struct snd_kcontrol *kcontrol,
 {
 	struct snd_soc_card *card =  snd_kcontrol_chip(kcontrol);
 
-	if (poodle_spk_func == ucontrol->value.integer.value[0])
+	if (poodle_spk_func == ucontrol->value.enumerated.item[0])
 		return 0;
 
-	poodle_spk_func = ucontrol->value.integer.value[0];
+	poodle_spk_func = ucontrol->value.enumerated.item[0];
 	poodle_ext_control(&card->dapm);
 	return 1;
 }
diff --git a/sound/soc/pxa/spitz.c b/sound/soc/pxa/spitz.c
index b00222620fd0..0e02634c8b7f 100644
--- a/sound/soc/pxa/spitz.c
+++ b/sound/soc/pxa/spitz.c
@@ -164,7 +164,7 @@ static struct snd_soc_ops spitz_ops = {
 static int spitz_get_jack(struct snd_kcontrol *kcontrol,
 	struct snd_ctl_elem_value *ucontrol)
 {
-	ucontrol->value.integer.value[0] = spitz_jack_func;
+	ucontrol->value.enumerated.item[0] = spitz_jack_func;
 	return 0;
 }
 
@@ -173,10 +173,10 @@ static int spitz_set_jack(struct snd_kcontrol *kcontrol,
 {
 	struct snd_soc_card *card = snd_kcontrol_chip(kcontrol);
 
-	if (spitz_jack_func == ucontrol->value.integer.value[0])
+	if (spitz_jack_func == ucontrol->value.enumerated.item[0])
 		return 0;
 
-	spitz_jack_func = ucontrol->value.integer.value[0];
+	spitz_jack_func = ucontrol->value.enumerated.item[0];
 	spitz_ext_control(&card->dapm);
 	return 1;
 }
@@ -184,7 +184,7 @@ static int spitz_set_jack(struct snd_kcontrol *kcontrol,
 static int spitz_get_spk(struct snd_kcontrol *kcontrol,
 	struct snd_ctl_elem_value *ucontrol)
 {
-	ucontrol->value.integer.value[0] = spitz_spk_func;
+	ucontrol->value.enumerated.item[0] = spitz_spk_func;
 	return 0;
 }
 
@@ -193,10 +193,10 @@ static int spitz_set_spk(struct snd_kcontrol *kcontrol,
 {
 	struct snd_soc_card *card = snd_kcontrol_chip(kcontrol);
 
-	if (spitz_spk_func == ucontrol->value.integer.value[0])
+	if (spitz_spk_func == ucontrol->value.enumerated.item[0])
 		return 0;
 
-	spitz_spk_func = ucontrol->value.integer.value[0];
+	spitz_spk_func = ucontrol->value.enumerated.item[0];
 	spitz_ext_control(&card->dapm);
 	return 1;
 }
diff --git a/sound/soc/pxa/tosa.c b/sound/soc/pxa/tosa.c
index 49518dd642aa..c508f024ecfb 100644
--- a/sound/soc/pxa/tosa.c
+++ b/sound/soc/pxa/tosa.c
@@ -95,7 +95,7 @@ static struct snd_soc_ops tosa_ops = {
 static int tosa_get_jack(struct snd_kcontrol *kcontrol,
 	struct snd_ctl_elem_value *ucontrol)
 {
-	ucontrol->value.integer.value[0] = tosa_jack_func;
+	ucontrol->value.enumerated.item[0] = tosa_jack_func;
 	return 0;
 }
 
@@ -104,10 +104,10 @@ static int tosa_set_jack(struct snd_kcontrol *kcontrol,
 {
 	struct snd_soc_card *card = snd_kcontrol_chip(kcontrol);
 
-	if (tosa_jack_func == ucontrol->value.integer.value[0])
+	if (tosa_jack_func == ucontrol->value.enumerated.item[0])
 		return 0;
 
-	tosa_jack_func = ucontrol->value.integer.value[0];
+	tosa_jack_func = ucontrol->value.enumerated.item[0];
 	tosa_ext_control(&card->dapm);
 	return 1;
 }
@@ -115,7 +115,7 @@ static int tosa_set_jack(struct snd_kcontrol *kcontrol,
 static int tosa_get_spk(struct snd_kcontrol *kcontrol,
 	struct snd_ctl_elem_value *ucontrol)
 {
-	ucontrol->value.integer.value[0] = tosa_spk_func;
+	ucontrol->value.enumerated.item[0] = tosa_spk_func;
 	return 0;
 }
 
@@ -124,10 +124,10 @@ static int tosa_set_spk(struct snd_kcontrol *kcontrol,
 {
 	struct snd_soc_card *card = snd_kcontrol_chip(kcontrol);
 
-	if (tosa_spk_func == ucontrol->value.integer.value[0])
+	if (tosa_spk_func == ucontrol->value.enumerated.item[0])
 		return 0;
 
-	tosa_spk_func = ucontrol->value.integer.value[0];
+	tosa_spk_func = ucontrol->value.enumerated.item[0];
 	tosa_ext_control(&card->dapm);
 	return 1;
 }
diff --git a/sound/soc/qcom/lpass-cpu.c b/sound/soc/qcom/lpass-cpu.c
index 00b6c9d039cf..e5101e0d2d37 100644
--- a/sound/soc/qcom/lpass-cpu.c
+++ b/sound/soc/qcom/lpass-cpu.c
@@ -355,7 +355,6 @@ static struct regmap_config lpass_cpu_regmap_config = {
 	.readable_reg = lpass_cpu_regmap_readable,
 	.volatile_reg = lpass_cpu_regmap_volatile,
 	.cache_type = REGCACHE_FLAT,
-	.val_format_endian = REGMAP_ENDIAN_LITTLE,
 };
 
 int asoc_qcom_lpass_cpu_platform_probe(struct platform_device *pdev)
diff --git a/sound/soc/qcom/lpass-platform.c b/sound/soc/qcom/lpass-platform.c
index 79688aa1941a..4aeb8e1a7160 100644
--- a/sound/soc/qcom/lpass-platform.c
+++ b/sound/soc/qcom/lpass-platform.c
@@ -440,18 +440,18 @@ static irqreturn_t lpass_platform_lpaif_irq(int irq, void *data)
 }
 
 static int lpass_platform_alloc_buffer(struct snd_pcm_substream *substream,
-		struct snd_soc_pcm_runtime *soc_runtime)
+		struct snd_soc_pcm_runtime *rt)
 {
 	struct snd_dma_buffer *buf = &substream->dma_buffer;
 	size_t size = lpass_platform_pcm_hardware.buffer_bytes_max;
 
 	buf->dev.type = SNDRV_DMA_TYPE_DEV;
-	buf->dev.dev = soc_runtime->dev;
+	buf->dev.dev = rt->platform->dev;
 	buf->private_data = NULL;
-	buf->area = dma_alloc_coherent(soc_runtime->dev, size, &buf->addr,
+	buf->area = dma_alloc_coherent(rt->platform->dev, size, &buf->addr,
 			GFP_KERNEL);
 	if (!buf->area) {
-		dev_err(soc_runtime->dev, "%s: Could not allocate DMA buffer\n",
+		dev_err(rt->platform->dev, "%s: Could not allocate DMA buffer\n",
 				__func__);
 		return -ENOMEM;
 	}
@@ -461,12 +461,12 @@ static int lpass_platform_alloc_buffer(struct snd_pcm_substream *substream,
 }
 
 static void lpass_platform_free_buffer(struct snd_pcm_substream *substream,
-		struct snd_soc_pcm_runtime *soc_runtime)
+		struct snd_soc_pcm_runtime *rt)
 {
 	struct snd_dma_buffer *buf = &substream->dma_buffer;
 
 	if (buf->area) {
-		dma_free_coherent(soc_runtime->dev, buf->bytes, buf->area,
+		dma_free_coherent(rt->dev, buf->bytes, buf->area,
 				buf->addr);
 	}
 	buf->area = NULL;
@@ -499,9 +499,6 @@ static int lpass_platform_pcm_new(struct snd_soc_pcm_runtime *soc_runtime)
 
 	snd_soc_pcm_set_drvdata(soc_runtime, data);
 
-	soc_runtime->dev->coherent_dma_mask = DMA_BIT_MASK(32);
-	soc_runtime->dev->dma_mask = &soc_runtime->dev->coherent_dma_mask;
-
 	ret = lpass_platform_alloc_buffer(substream, soc_runtime);
 	if (ret)
 		return ret;
diff --git a/sound/soc/samsung/i2s.c b/sound/soc/samsung/i2s.c
index 84d9e77c0fbe..70a2559b63f9 100644
--- a/sound/soc/samsung/i2s.c
+++ b/sound/soc/samsung/i2s.c
@@ -481,10 +481,11 @@ static int i2s_set_sysclk(struct snd_soc_dai *dai,
 	unsigned int cdcon_mask = 1 << i2s_regs->cdclkcon_off;
 	unsigned int rsrc_mask = 1 << i2s_regs->rclksrc_off;
 	u32 mod, mask, val = 0;
+	unsigned long flags;
 
-	spin_lock(i2s->lock);
+	spin_lock_irqsave(i2s->lock, flags);
 	mod = readl(i2s->addr + I2SMOD);
-	spin_unlock(i2s->lock);
+	spin_unlock_irqrestore(i2s->lock, flags);
 
 	switch (clk_id) {
 	case SAMSUNG_I2S_OPCLK:
@@ -575,11 +576,11 @@ static int i2s_set_sysclk(struct snd_soc_dai *dai,
 		return -EINVAL;
 	}
 
-	spin_lock(i2s->lock);
+	spin_lock_irqsave(i2s->lock, flags);
 	mod = readl(i2s->addr + I2SMOD);
 	mod = (mod & ~mask) | val;
 	writel(mod, i2s->addr + I2SMOD);
-	spin_unlock(i2s->lock);
+	spin_unlock_irqrestore(i2s->lock, flags);
 
 	return 0;
 }
@@ -590,6 +591,7 @@ static int i2s_set_fmt(struct snd_soc_dai *dai,
 	struct i2s_dai *i2s = to_info(dai);
 	int lrp_shift, sdf_shift, sdf_mask, lrp_rlow, mod_slave;
 	u32 mod, tmp = 0;
+	unsigned long flags;
 
 	lrp_shift = i2s->variant_regs->lrp_off;
 	sdf_shift = i2s->variant_regs->sdf_off;
@@ -649,7 +651,7 @@ static int i2s_set_fmt(struct snd_soc_dai *dai,
 		return -EINVAL;
 	}
 
-	spin_lock(i2s->lock);
+	spin_lock_irqsave(i2s->lock, flags);
 	mod = readl(i2s->addr + I2SMOD);
 	/*
 	 * Don't change the I2S mode if any controller is active on this
@@ -657,7 +659,7 @@ static int i2s_set_fmt(struct snd_soc_dai *dai,
 	 */
 	if (any_active(i2s) &&
 		((mod & (sdf_mask | lrp_rlow | mod_slave)) != tmp)) {
-		spin_unlock(i2s->lock);
+		spin_unlock_irqrestore(i2s->lock, flags);
 		dev_err(&i2s->pdev->dev,
 				"%s:%d Other DAI busy\n", __func__, __LINE__);
 		return -EAGAIN;
@@ -666,7 +668,7 @@ static int i2s_set_fmt(struct snd_soc_dai *dai,
 	mod &= ~(sdf_mask | lrp_rlow | mod_slave);
 	mod |= tmp;
 	writel(mod, i2s->addr + I2SMOD);
-	spin_unlock(i2s->lock);
+	spin_unlock_irqrestore(i2s->lock, flags);
 
 	return 0;
 }
@@ -676,6 +678,7 @@ static int i2s_hw_params(struct snd_pcm_substream *substream,
 {
 	struct i2s_dai *i2s = to_info(dai);
 	u32 mod, mask = 0, val = 0;
+	unsigned long flags;
 
 	if (!is_secondary(i2s))
 		mask |= (MOD_DC2_EN | MOD_DC1_EN);
@@ -744,11 +747,11 @@ static int i2s_hw_params(struct snd_pcm_substream *substream,
 		return -EINVAL;
 	}
 
-	spin_lock(i2s->lock);
+	spin_lock_irqsave(i2s->lock, flags);
 	mod = readl(i2s->addr + I2SMOD);
 	mod = (mod & ~mask) | val;
 	writel(mod, i2s->addr + I2SMOD);
-	spin_unlock(i2s->lock);
+	spin_unlock_irqrestore(i2s->lock, flags);
 
 	samsung_asoc_init_dma_data(dai, &i2s->dma_playback, &i2s->dma_capture);
 
diff --git a/sound/soc/soc-dapm.c b/sound/soc/soc-dapm.c
index 5a2812fa8946..581175a51ecf 100644
--- a/sound/soc/soc-dapm.c
+++ b/sound/soc/soc-dapm.c
@@ -310,7 +310,7 @@ struct dapm_kcontrol_data {
 };
 
 static int dapm_kcontrol_data_alloc(struct snd_soc_dapm_widget *widget,
-	struct snd_kcontrol *kcontrol)
+	struct snd_kcontrol *kcontrol, const char *ctrl_name)
 {
 	struct dapm_kcontrol_data *data;
 	struct soc_mixer_control *mc;
@@ -333,7 +333,7 @@ static int dapm_kcontrol_data_alloc(struct snd_soc_dapm_widget *widget,
 		if (mc->autodisable) {
 			struct snd_soc_dapm_widget template;
 
-			name = kasprintf(GFP_KERNEL, "%s %s", kcontrol->id.name,
+			name = kasprintf(GFP_KERNEL, "%s %s", ctrl_name,
 					 "Autodisable");
 			if (!name) {
 				ret = -ENOMEM;
@@ -371,7 +371,7 @@ static int dapm_kcontrol_data_alloc(struct snd_soc_dapm_widget *widget,
 		if (e->autodisable) {
 			struct snd_soc_dapm_widget template;
 
-			name = kasprintf(GFP_KERNEL, "%s %s", kcontrol->id.name,
+			name = kasprintf(GFP_KERNEL, "%s %s", ctrl_name,
 					 "Autodisable");
 			if (!name) {
 				ret = -ENOMEM;
@@ -871,7 +871,7 @@ static int dapm_create_or_share_kcontrol(struct snd_soc_dapm_widget *w,
 
 		kcontrol->private_free = dapm_kcontrol_free;
 
-		ret = dapm_kcontrol_data_alloc(w, kcontrol);
+		ret = dapm_kcontrol_data_alloc(w, kcontrol, name);
 		if (ret) {
 			snd_ctl_free_one(kcontrol);
 			goto exit_free;
@@ -3573,7 +3573,7 @@ static int snd_soc_dapm_dai_link_get(struct snd_kcontrol *kcontrol,
 {
 	struct snd_soc_dapm_widget *w = snd_kcontrol_chip(kcontrol);
 
-	ucontrol->value.integer.value[0] = w->params_select;
+	ucontrol->value.enumerated.item[0] = w->params_select;
 
 	return 0;
 }
@@ -3587,13 +3587,13 @@ static int snd_soc_dapm_dai_link_put(struct snd_kcontrol *kcontrol,
 	if (w->power)
 		return -EBUSY;
 
-	if (ucontrol->value.integer.value[0] == w->params_select)
+	if (ucontrol->value.enumerated.item[0] == w->params_select)
 		return 0;
 
-	if (ucontrol->value.integer.value[0] >= w->num_params)
+	if (ucontrol->value.enumerated.item[0] >= w->num_params)
 		return -EINVAL;
 
-	w->params_select = ucontrol->value.integer.value[0];
+	w->params_select = ucontrol->value.enumerated.item[0];
 
 	return 0;
 }
diff --git a/sound/soc/soc-pcm.c b/sound/soc/soc-pcm.c
index e898b427be7e..1af4f23697a7 100644
--- a/sound/soc/soc-pcm.c
+++ b/sound/soc/soc-pcm.c
@@ -1810,7 +1810,8 @@ int dpcm_be_dai_hw_free(struct snd_soc_pcm_runtime *fe, int stream)
 		    (be->dpcm[stream].state != SND_SOC_DPCM_STATE_PREPARE) &&
 		    (be->dpcm[stream].state != SND_SOC_DPCM_STATE_HW_FREE) &&
 		    (be->dpcm[stream].state != SND_SOC_DPCM_STATE_PAUSED) &&
-		    (be->dpcm[stream].state != SND_SOC_DPCM_STATE_STOP))
+		    (be->dpcm[stream].state != SND_SOC_DPCM_STATE_STOP) &&
+		    (be->dpcm[stream].state != SND_SOC_DPCM_STATE_SUSPEND))
 			continue;
 
 		dev_dbg(be->dev, "ASoC: hw_free BE %s\n",
diff --git a/sound/usb/midi.c b/sound/usb/midi.c
index cc39f63299ef..007cf5831121 100644
--- a/sound/usb/midi.c
+++ b/sound/usb/midi.c
@@ -2455,7 +2455,6 @@ int snd_usbmidi_create(struct snd_card *card,
 	else
 		err = snd_usbmidi_create_endpoints(umidi, endpoints);
 	if (err < 0) {
-		snd_usbmidi_free(umidi);
 		return err;
 	}
 
diff --git a/sound/usb/quirks.c b/sound/usb/quirks.c
index 4f6ce1cac8e2..c458d60d5030 100644
--- a/sound/usb/quirks.c
+++ b/sound/usb/quirks.c
@@ -1124,6 +1124,7 @@ bool snd_usb_get_sample_rate_quirk(struct snd_usb_audio *chip)
 	case USB_ID(0x045E, 0x076F): /* MS Lifecam HD-6000 */
 	case USB_ID(0x045E, 0x0772): /* MS Lifecam Studio */
 	case USB_ID(0x045E, 0x0779): /* MS Lifecam HD-3000 */
+	case USB_ID(0x047F, 0xAA05): /* Plantronics DA45 */
 	case USB_ID(0x04D8, 0xFEEA): /* Benchmark DAC1 Pre */
 	case USB_ID(0x074D, 0x3553): /* Outlaw RR2150 (Micronas UAC3553B) */
 	case USB_ID(0x21B4, 0x0081): /* AudioQuest DragonFly */
diff --git a/tools/perf/util/intel-pt.c b/tools/perf/util/intel-pt.c
index 81a2eb77ba7f..05d815851be1 100644
--- a/tools/perf/util/intel-pt.c
+++ b/tools/perf/util/intel-pt.c
@@ -2068,6 +2068,15 @@ int intel_pt_process_auxtrace_info(union perf_event *event,
 		err = -ENOMEM;
 		goto err_free_queues;
 	}
+
+	/*
+	 * Since this thread will not be kept in any rbtree not in a
+	 * list, initialize its list node so that at thread__put() the
+	 * current thread lifetime assuption is kept and we don't segfault
+	 * at list_del_init().
+	 */
+	INIT_LIST_HEAD(&pt->unknown_thread->node);
+
 	err = thread__set_comm(pt->unknown_thread, "unknown", 0);
 	if (err)
 		goto err_delete_thread;
diff --git a/tools/perf/util/parse-events.c b/tools/perf/util/parse-events.c
index 4f7b0efdde2f..813d9b272c81 100644
--- a/tools/perf/util/parse-events.c
+++ b/tools/perf/util/parse-events.c
@@ -399,6 +399,9 @@ static void tracepoint_error(struct parse_events_error *e, int err,
 {
 	char help[BUFSIZ];
 
+	if (!e)
+		return;
+
 	/*
 	 * We get error directly from syscall errno ( > 0),
 	 * or from encoded pointer's error ( < 0).
diff --git a/tools/perf/util/probe-finder.c b/tools/perf/util/probe-finder.c
index 2be10fb27172..4ce5c5e18f48 100644
--- a/tools/perf/util/probe-finder.c
+++ b/tools/perf/util/probe-finder.c
@@ -686,8 +686,9 @@ static int call_probe_finder(Dwarf_Die *sc_die, struct probe_finder *pf)
 		pf->fb_ops = NULL;
 #if _ELFUTILS_PREREQ(0, 142)
 	} else if (nops == 1 && pf->fb_ops[0].atom == DW_OP_call_frame_cfa &&
-		   pf->cfi != NULL) {
-		if (dwarf_cfi_addrframe(pf->cfi, pf->addr, &frame) != 0 ||
+		   (pf->cfi_eh != NULL || pf->cfi_dbg != NULL)) {
+		if ((dwarf_cfi_addrframe(pf->cfi_eh, pf->addr, &frame) != 0 &&
+		     (dwarf_cfi_addrframe(pf->cfi_dbg, pf->addr, &frame) != 0)) ||
 		    dwarf_frame_cfa(frame, &pf->fb_ops, &nops) != 0) {
 			pr_warning("Failed to get call frame on 0x%jx\n",
 				   (uintmax_t)pf->addr);
@@ -1015,8 +1016,7 @@ static int pubname_search_cb(Dwarf *dbg, Dwarf_Global *gl, void *data)
 	return DWARF_CB_OK;
 }
 
-/* Find probe points from debuginfo */
-static int debuginfo__find_probes(struct debuginfo *dbg,
+static int debuginfo__find_probe_location(struct debuginfo *dbg,
 				  struct probe_finder *pf)
 {
 	struct perf_probe_point *pp = &pf->pev->point;
@@ -1025,27 +1025,6 @@ static int debuginfo__find_probes(struct debuginfo *dbg,
 	Dwarf_Die *diep;
 	int ret = 0;
 
-#if _ELFUTILS_PREREQ(0, 142)
-	Elf *elf;
-	GElf_Ehdr ehdr;
-	GElf_Shdr shdr;
-
-	/* Get the call frame information from this dwarf */
-	elf = dwarf_getelf(dbg->dbg);
-	if (elf == NULL)
-		return -EINVAL;
-
-	if (gelf_getehdr(elf, &ehdr) == NULL)
-		return -EINVAL;
-
-	if (elf_section_by_name(elf, &ehdr, &shdr, ".eh_frame", NULL) &&
-	    shdr.sh_type == SHT_PROGBITS) {
-		pf->cfi = dwarf_getcfi_elf(elf);
-	} else {
-		pf->cfi = dwarf_getcfi(dbg->dbg);
-	}
-#endif
-
 	off = 0;
 	pf->lcache = intlist__new(NULL);
 	if (!pf->lcache)
@@ -1108,6 +1087,39 @@ found:
 	return ret;
 }
 
+/* Find probe points from debuginfo */
+static int debuginfo__find_probes(struct debuginfo *dbg,
+				  struct probe_finder *pf)
+{
+	int ret = 0;
+
+#if _ELFUTILS_PREREQ(0, 142)
+	Elf *elf;
+	GElf_Ehdr ehdr;
+	GElf_Shdr shdr;
+
+	if (pf->cfi_eh || pf->cfi_dbg)
+		return debuginfo__find_probe_location(dbg, pf);
+
+	/* Get the call frame information from this dwarf */
+	elf = dwarf_getelf(dbg->dbg);
+	if (elf == NULL)
+		return -EINVAL;
+
+	if (gelf_getehdr(elf, &ehdr) == NULL)
+		return -EINVAL;
+
+	if (elf_section_by_name(elf, &ehdr, &shdr, ".eh_frame", NULL) &&
+	    shdr.sh_type == SHT_PROGBITS)
+		pf->cfi_eh = dwarf_getcfi_elf(elf);
+
+	pf->cfi_dbg = dwarf_getcfi(dbg->dbg);
+#endif
+
+	ret = debuginfo__find_probe_location(dbg, pf);
+	return ret;
+}
+
 struct local_vars_finder {
 	struct probe_finder *pf;
 	struct perf_probe_arg *args;
diff --git a/tools/perf/util/probe-finder.h b/tools/perf/util/probe-finder.h
index bed82716e1b4..0aec7704e395 100644
--- a/tools/perf/util/probe-finder.h
+++ b/tools/perf/util/probe-finder.h
@@ -76,7 +76,10 @@ struct probe_finder {
 
 	/* For variable searching */
 #if _ELFUTILS_PREREQ(0, 142)
-	Dwarf_CFI		*cfi;		/* Call Frame Information */
+	/* Call Frame Information from .eh_frame */
+	Dwarf_CFI		*cfi_eh;
+	/* Call Frame Information from .debug_frame */
+	Dwarf_CFI		*cfi_dbg;
 #endif
 	Dwarf_Op		*fb_ops;	/* Frame base attribute */
 	struct perf_probe_arg	*pvar;		/* Current target variable */
diff --git a/tools/perf/util/stat.c b/tools/perf/util/stat.c
index 2b58edccd56f..afb0c45eba34 100644
--- a/tools/perf/util/stat.c
+++ b/tools/perf/util/stat.c
@@ -311,6 +311,16 @@ int perf_stat_process_counter(struct perf_stat_config *config,
 
 	aggr->val = aggr->ena = aggr->run = 0;
 
+	/*
+	 * We calculate counter's data every interval,
+	 * and the display code shows ps->res_stats
+	 * avg value. We need to zero the stats for
+	 * interval mode, otherwise overall avg running
+	 * averages will be shown for each interval.
+	 */
+	if (config->interval)
+		init_stats(ps->res_stats);
+
 	if (counter->per_pkg)
 		zero_per_pkg(counter);
 
diff --git a/tools/testing/nvdimm/test/nfit.c b/tools/testing/nvdimm/test/nfit.c
index 90bd2ea41032..b3281dcd4a5d 100644
--- a/tools/testing/nvdimm/test/nfit.c
+++ b/tools/testing/nvdimm/test/nfit.c
@@ -217,13 +217,16 @@ static int nfit_test_cmd_set_config_data(struct nd_cmd_set_config_hdr *nd_cmd,
 	return rc;
 }
 
+#define NFIT_TEST_ARS_RECORDS 4
+
 static int nfit_test_cmd_ars_cap(struct nd_cmd_ars_cap *nd_cmd,
 		unsigned int buf_len)
 {
 	if (buf_len < sizeof(*nd_cmd))
 		return -EINVAL;
 
-	nd_cmd->max_ars_out = 256;
+	nd_cmd->max_ars_out = sizeof(struct nd_cmd_ars_status)
+		+ NFIT_TEST_ARS_RECORDS * sizeof(struct nd_ars_record);
 	nd_cmd->status = (ND_ARS_PERSISTENT | ND_ARS_VOLATILE) << 16;
 
 	return 0;
@@ -246,7 +249,8 @@ static int nfit_test_cmd_ars_status(struct nd_cmd_ars_status *nd_cmd,
 	if (buf_len < sizeof(*nd_cmd))
 		return -EINVAL;
 
-	nd_cmd->out_length = 256;
+	nd_cmd->out_length = sizeof(struct nd_cmd_ars_status);
+	/* TODO: emit error records */
 	nd_cmd->num_records = 0;
 	nd_cmd->address = 0;
 	nd_cmd->length = -1ULL;
diff --git a/tools/testing/selftests/efivarfs/efivarfs.sh b/tools/testing/selftests/efivarfs/efivarfs.sh
index 77edcdcc016b..057278448515 100755
--- a/tools/testing/selftests/efivarfs/efivarfs.sh
+++ b/tools/testing/selftests/efivarfs/efivarfs.sh
@@ -88,7 +88,11 @@ test_delete()
 		exit 1
 	fi
 
-	rm $file
+	rm $file 2>/dev/null
+	if [ $? -ne 0 ]; then
+		chattr -i $file
+		rm $file
+	fi
 
 	if [ -e $file ]; then
 		echo "$file couldn't be deleted" >&2
@@ -111,6 +115,7 @@ test_zero_size_delete()
 		exit 1
 	fi
 
+	chattr -i $file
 	printf "$attrs" > $file
 
 	if [ -e $file ]; then
@@ -141,7 +146,11 @@ test_valid_filenames()
 			echo "$file could not be created" >&2
 			ret=1
 		else
-			rm $file
+			rm $file 2>/dev/null
+			if [ $? -ne 0 ]; then
+				chattr -i $file
+				rm $file
+			fi
 		fi
 	done
 
@@ -174,7 +183,11 @@ test_invalid_filenames()
 
 		if [ -e $file ]; then
 			echo "Creating $file should have failed" >&2
-			rm $file
+			rm $file 2>/dev/null
+			if [ $? -ne 0 ]; then
+				chattr -i $file
+				rm $file
+			fi
 			ret=1
 		fi
 	done
diff --git a/tools/testing/selftests/efivarfs/open-unlink.c b/tools/testing/selftests/efivarfs/open-unlink.c
index 8c0764407b3c..4af74f733036 100644
--- a/tools/testing/selftests/efivarfs/open-unlink.c
+++ b/tools/testing/selftests/efivarfs/open-unlink.c
@@ -1,10 +1,68 @@
+#include <errno.h>
 #include <stdio.h>
 #include <stdint.h>
 #include <stdlib.h>
 #include <unistd.h>
+#include <sys/ioctl.h>
 #include <sys/types.h>
 #include <sys/stat.h>
 #include <fcntl.h>
+#include <linux/fs.h>
+
+static int set_immutable(const char *path, int immutable)
+{
+	unsigned int flags;
+	int fd;
+	int rc;
+	int error;
+
+	fd = open(path, O_RDONLY);
+	if (fd < 0)
+		return fd;
+
+	rc = ioctl(fd, FS_IOC_GETFLAGS, &flags);
+	if (rc < 0) {
+		error = errno;
+		close(fd);
+		errno = error;
+		return rc;
+	}
+
+	if (immutable)
+		flags |= FS_IMMUTABLE_FL;
+	else
+		flags &= ~FS_IMMUTABLE_FL;
+
+	rc = ioctl(fd, FS_IOC_SETFLAGS, &flags);
+	error = errno;
+	close(fd);
+	errno = error;
+	return rc;
+}
+
+static int get_immutable(const char *path)
+{
+	unsigned int flags;
+	int fd;
+	int rc;
+	int error;
+
+	fd = open(path, O_RDONLY);
+	if (fd < 0)
+		return fd;
+
+	rc = ioctl(fd, FS_IOC_GETFLAGS, &flags);
+	if (rc < 0) {
+		error = errno;
+		close(fd);
+		errno = error;
+		return rc;
+	}
+	close(fd);
+	if (flags & FS_IMMUTABLE_FL)
+		return 1;
+	return 0;
+}
 
 int main(int argc, char **argv)
 {
@@ -27,7 +85,7 @@ int main(int argc, char **argv)
 	buf[4] = 0;
 
 	/* create a test variable */
-	fd = open(path, O_WRONLY | O_CREAT);
+	fd = open(path, O_WRONLY | O_CREAT, 0600);
 	if (fd < 0) {
 		perror("open(O_WRONLY)");
 		return EXIT_FAILURE;
@@ -41,6 +99,18 @@ int main(int argc, char **argv)
 
 	close(fd);
 
+	rc = get_immutable(path);
+	if (rc < 0) {
+		perror("ioctl(FS_IOC_GETFLAGS)");
+		return EXIT_FAILURE;
+	} else if (rc) {
+		rc = set_immutable(path, 0);
+		if (rc < 0) {
+			perror("ioctl(FS_IOC_SETFLAGS)");
+			return EXIT_FAILURE;
+		}
+	}
+
 	fd = open(path, O_RDONLY);
 	if (fd < 0) {
 		perror("open");
diff --git a/tools/testing/selftests/ftrace/test.d/instances/instance.tc b/tools/testing/selftests/ftrace/test.d/instances/instance.tc
index 773e276ff90b..1e1abe0ad354 100644
--- a/tools/testing/selftests/ftrace/test.d/instances/instance.tc
+++ b/tools/testing/selftests/ftrace/test.d/instances/instance.tc
@@ -39,28 +39,23 @@ instance_slam() {
 }
 
 instance_slam &
-x=`jobs -l`
-p1=`echo $x | cut -d' ' -f2`
+p1=$!
 echo $p1
 
 instance_slam &
-x=`jobs -l | tail -1`
-p2=`echo $x | cut -d' ' -f2`
+p2=$!
 echo $p2
 
 instance_slam &
-x=`jobs -l | tail -1`
-p3=`echo $x | cut -d' ' -f2`
+p3=$!
 echo $p3
 
 instance_slam &
-x=`jobs -l | tail -1`
-p4=`echo $x | cut -d' ' -f2`
+p4=$!
 echo $p4
 
 instance_slam &
-x=`jobs -l | tail -1`
-p5=`echo $x | cut -d' ' -f2`
+p5=$!
 echo $p5
 
 ls -lR >/dev/null
diff --git a/virt/kvm/arm/arch_timer.c b/virt/kvm/arm/arch_timer.c
index 69bca185c471..ea6064696fe4 100644
--- a/virt/kvm/arm/arch_timer.c
+++ b/virt/kvm/arm/arch_timer.c
@@ -143,7 +143,7 @@ static void kvm_timer_update_irq(struct kvm_vcpu *vcpu, bool new_level)
  * Check if there was a change in the timer state (should we raise or lower
  * the line level to the GIC).
  */
-static void kvm_timer_update_state(struct kvm_vcpu *vcpu)
+static int kvm_timer_update_state(struct kvm_vcpu *vcpu)
 {
 	struct arch_timer_cpu *timer = &vcpu->arch.timer_cpu;
 
@@ -154,10 +154,12 @@ static void kvm_timer_update_state(struct kvm_vcpu *vcpu)
 	 * until we call this function from kvm_timer_flush_hwstate.
 	 */
 	if (!vgic_initialized(vcpu->kvm))
-	    return;
+		return -ENODEV;
 
 	if (kvm_timer_should_fire(vcpu) != timer->irq.level)
 		kvm_timer_update_irq(vcpu, !timer->irq.level);
+
+	return 0;
 }
 
 /*
@@ -218,7 +220,8 @@ void kvm_timer_flush_hwstate(struct kvm_vcpu *vcpu)
 	bool phys_active;
 	int ret;
 
-	kvm_timer_update_state(vcpu);
+	if (kvm_timer_update_state(vcpu))
+		return;
 
 	/*
 	* If we enter the guest with the virtual input level to the VGIC
diff --git a/virt/kvm/arm/vgic.c b/virt/kvm/arm/vgic.c
index 043032c6a5a4..00429b392c61 100644
--- a/virt/kvm/arm/vgic.c
+++ b/virt/kvm/arm/vgic.c
@@ -1875,8 +1875,8 @@ void kvm_vgic_vcpu_destroy(struct kvm_vcpu *vcpu)
 static int vgic_vcpu_init_maps(struct kvm_vcpu *vcpu, int nr_irqs)
 {
 	struct vgic_cpu *vgic_cpu = &vcpu->arch.vgic_cpu;
-
-	int sz = (nr_irqs - VGIC_NR_PRIVATE_IRQS) / 8;
+	int nr_longs = BITS_TO_LONGS(nr_irqs - VGIC_NR_PRIVATE_IRQS);
+	int sz = nr_longs * sizeof(unsigned long);
 	vgic_cpu->pending_shared = kzalloc(sz, GFP_KERNEL);
 	vgic_cpu->active_shared = kzalloc(sz, GFP_KERNEL);
 	vgic_cpu->pend_act_shared = kzalloc(sz, GFP_KERNEL);
diff --git a/virt/kvm/async_pf.c b/virt/kvm/async_pf.c
index 353159922456..db2dd3335c6a 100644
--- a/virt/kvm/async_pf.c
+++ b/virt/kvm/async_pf.c
@@ -172,7 +172,7 @@ int kvm_setup_async_pf(struct kvm_vcpu *vcpu, gva_t gva, unsigned long hva,
 	 * do alloc nowait since if we are going to sleep anyway we
 	 * may as well sleep faulting in page
 	 */
-	work = kmem_cache_zalloc(async_pf_cache, GFP_NOWAIT);
+	work = kmem_cache_zalloc(async_pf_cache, GFP_NOWAIT | __GFP_NOWARN);
 	if (!work)
 		return 0;
 
diff --git a/virt/kvm/kvm_main.c b/virt/kvm/kvm_main.c
index a11cfd20a6a0..9102ae172d2a 100644
--- a/virt/kvm/kvm_main.c
+++ b/virt/kvm/kvm_main.c
@@ -1952,6 +1952,9 @@ static void grow_halt_poll_ns(struct kvm_vcpu *vcpu)
 	else
 		val *= halt_poll_ns_grow;
 
+	if (val > halt_poll_ns)
+		val = halt_poll_ns;
+
 	vcpu->halt_poll_ns = val;
 	trace_kvm_halt_poll_ns_grow(vcpu->vcpu_id, val, old);
 }