| Age | Commit message (Collapse) | Author | Files | Lines |
|---|
|
In monitoring a transcode pipeline that is latency sensitive (it waits
between submitting frames, and each frame requires work on rcs/vcs/vecs
engines), it is found that it took longer than a single jiffy for it to
sustain its workload. Allowing an extra jiffy headroom for the userspace
prevents us from prematurely parking and having to exit powersaving
immediately.
Link: https://gitlab.freedesktop.org/drm/intel/-/issues/6284
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Signed-off-by: Karolina Drobnik <karolina.drobnik@intel.com>
Reviewed-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
Reviewed-by: Andi Shyti <andi.shyti@linux.intel.com>
Signed-off-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/e37911ec087a9ce50630d6faf61fa2c0d5f96d44.1657289332.git.karolina.drobnik@intel.com
|
|
We employ a "waitboost" heuristic to detect when userspace is stalled
waiting for results from earlier execution. Under latency sensitive work
mixed between the gpu/cpu, the GPU is typically under-utilised and so
RPS sees that low utilisation as a reason to downclock the frequency,
causing longer stalls and lower throughput. The user left waiting for
the results is not impressed.
On applying commit 047a1b877ed4 ("dma-buf & drm/amdgpu: remove dma_resv
workaround") it was observed that deinterlacing h264 on Haswell
performance dropped by 2-5x. The reason being that the natural workload
was not intense enough to trigger RPS (using HW evaluation intervals) to
upclock, and so it was depending on waitboosting for the throughput.
Commit 047a1b877ed4 ("dma-buf & drm/amdgpu: remove dma_resv workaround")
changes the composition of dma-resv from keeping a single write fence +
multiple read fences, to a single array of multiple write and read
fences (a maximum of one pair of write/read fences per context). The
iteration order was also changed implicitly from all-read fences then
the single write fence, to a mix of write fences followed by read
fences. It is that ordering change that belied the fragility of
waitboosting.
Currently, a waitboost is inspected at the point of waiting on an
outstanding fence. If the GPU is backlogged such that we haven't yet
stated the request we need to wait on, we force the GPU to upclock until
the completion of that request. By changing the order in which we waited
upon requests, we ended up waiting on those requests in sequence and as
such we saw that each request was already started and so not a suitable
candidate for waitboosting.
Instead of asking whether to boost each fence in turn, we can look at
whether boosting is required for the dma-resv ensemble prior to waiting
on any fence, making the heuristic more robust to the order in which
fences are stored in the dma-resv.
Reported-by: Thomas Voegtle <tv@lio96.de>
Closes: https://gitlab.freedesktop.org/drm/intel/-/issues/6284
Fixes: 047a1b877ed4 ("dma-buf & drm/amdgpu: remove dma_resv workaround")
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Signed-off-by: Karolina Drobnik <karolina.drobnik@intel.com>
Tested-by: Thomas Voegtle <tv@lio96.de>
Reviewed-by: Andi Shyti <andi.shyti@linux.intel.com>
Acked-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
Signed-off-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/07e05518d9f6620d20cc1101ec1849203fe973f9.1657289332.git.karolina.drobnik@intel.com
|
|
Avoid trying to invalidate the TLB in the middle of performing an
engine reset, as this may result in the reset timing out. Currently,
the TLB invalidate is only serialised by its own mutex, forgoing the
uncore lock, but we can take the uncore->lock as well to serialise
the mmio access, thereby serialising with the GDRST.
Tested on a NUC5i7RYB, BIOS RYBDWi35.86A.0380.2019.0517.1530 with
i915 selftest/hangcheck.
Cc: stable@vger.kernel.org # v4.4 and upper
Fixes: 7938d61591d3 ("drm/i915: Flush TLBs before releasing backing store")
Reported-by: Mauro Carvalho Chehab <mchehab@kernel.org>
Tested-by: Mauro Carvalho Chehab <mchehab@kernel.org>
Reviewed-by: Mauro Carvalho Chehab <mchehab@kernel.org>
Signed-off-by: Chris Wilson <chris.p.wilson@intel.com>
Cc: Tvrtko Ursulin <tvrtko.ursulin@linux.intel.com>
Reviewed-by: Andi Shyti <andi.shyti@linux.intel.com>
Acked-by: Thomas Hellström <thomas.hellstrom@linux.intel.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@kernel.org>
Signed-off-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/1e59a7c45dd919a530256b9ac721ac6ea86c0677.1657639152.git.mchehab@kernel.org
|
|
Don't allow two engines to be reset in parallel, as they would both
try to select a reset bit (and send requests to common registers)
and wait on that register, at the same time. Serialize control of
the reset requests/acks using the uncore->lock, which will also ensure
that no other GT state changes at the same time as the actual reset.
Cc: stable@vger.kernel.org # v4.4 and upper
Reported-by: Mika Kuoppala <mika.kuoppala@linux.intel.com>
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Acked-by: Mika Kuoppala <mika.kuoppala@linux.intel.com>
Reviewed-by: Andi Shyti <andi.shyti@intel.com>
Reviewed-by: Andrzej Hajda <andrzej.hajda@intel.com>
Acked-by: Thomas Hellström <thomas.hellstrom@linux.intel.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@kernel.org>
Signed-off-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/e0a2d894e77aed7c2e36b0d1abdc7dbac3011729.1657639152.git.mchehab@kernel.org
|
|
When I look into implements of create_hist_fields(), I think there can be
following two simplifications:
1. If something wrong happened in parse_var_defs(), free_var_defs() would
have been called in it, so no need goto free again after calling it;
2. After calling create_key_fields(), regardless of the value of 'ret', it
then always runs into 'out: ', so the judge of 'ret' is redundant.
Link: https://lkml.kernel.org/r/20220630013152.164871-1-zhengyejian1@huawei.com
Signed-off-by: Zheng Yejian <zhengyejian1@huawei.com>
Reviewed-by: Tom Rix <trix@redhat.com>
Signed-off-by: Steven Rostedt (Google) <rostedt@goodmis.org>
|
|
Delete the redundant word 'have'.
Link: https://lkml.kernel.org/r/20220606023007.23377-1-wangxiang@cdjrlc.com
Signed-off-by: Xiang wangx <wangxiang@cdjrlc.com>
Signed-off-by: Steven Rostedt (Google) <rostedt@goodmis.org>
|
|
Qualcomm ARM64 defconfig updates for v5.20
This enables the drivers for the Light Pulse Generator and Bandwidth
Monitor hardware, found in a wide variety of Qualcomm PMICs and SoCs.
It enables interconnect providers for MSM8996, QCS404 and SC7180 and it
enables the USB PHYs for the QCS404 platform.
* tag 'qcom-arm64-defconfig-for-5.20' of git://git.kernel.org/pub/scm/linux/kernel/git/qcom/linux:
arm64: defconfig: enable Qualcomm Bandwidth Monitor
arm64: defconfig: Enable qcom interconnect drivers
arm64: defconfig: Enable qcom ss & hs usb phy
arm64: defconfig: enable Qualcomm LPG leds driver
Link: https://lore.kernel.org/r/20220712031812.1275182-1-bjorn.andersson@linaro.org
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
|
|
This pull request contains Broadcom ARM64-based SoCs defconfig changes
for 5.10, please pull the following:
- Nicolas enables CONFIG_DRM_V3D in order for the Raspberry Pi 4 systems
to load the proper GPU module driver
- William enables the newly introduced CONFIG_ARCH_BCMBCA 64-bit ARM
counterpart to support the newly added BCA SocS
* tag 'arm-soc/for-5.20/defconfig-arm64' of https://github.com/Broadcom/stblinux:
arm64: defconfig: Enable DRM_V3D
arm64: defconfig: enable bcmbca soc support
Link: https://lore.kernel.org/r/20220711164451.3542127-3-f.fainelli@gmail.com
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
|
|
This pull request contains ARM SoCs defconfig changes for 5.20, please
pull the following:
- Peter enables CONFIG_DRM_V3D in order for the Raspberry Pi 4 systems
to load the proper GPU module driver
- William updates the multi_v7_defconfig to replace the enabling of
CONFIG_ARCH_BCM_63XX with CONFIG_ARCH_BCMBCA which is how this platform
is now supported
* tag 'arm-soc/for-5.20/defconfig' of https://github.com/Broadcom/stblinux:
ARM: multi_v7_defconfig: Update configs for BCM63138
ARM: configs: Enable DRM_V3D
Link: https://lore.kernel.org/r/20220711164451.3542127-2-f.fainelli@gmail.com
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
|
|
Add a separate YAML binding to act as a superset of the PAS and non-PAS
compatible for SC7180 MSS PIL. This also serves as a way to increase
readability.
Signed-off-by: Sibi Sankar <quic_sibis@quicinc.com>
Reviewed-by: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
Signed-off-by: Bjorn Andersson <bjorn.andersson@linaro.org>
Link: https://lore.kernel.org/r/1657020721-24939-3-git-send-email-quic_sibis@quicinc.com
|
|
i.MX defconfig update for 5.20:
- Enable i.MX93 clock and pinctrl driver in arm64 defconfig.
- Enable gpio-fan driver in arm64 defconfig, as it's needed by
phyBOARD-Polis-i.MX8M Mini board.
* tag 'imx-defconfig-5.20' of git://git.kernel.org/pub/scm/linux/kernel/git/shawnguo/linux:
arm64: defconfig: Enable gpio-fan support
arm64: defconfig: enable i.MX93 clk & pinctrl
Link: https://lore.kernel.org/r/20220709082951.15123-6-shawnguo@kernel.org
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
|
|
Add a separate YAML binding to act as a superset of the PAS and non-PAS
compatible for SC7280 MSS PIL. This also serves as a way to increase
readability.
Signed-off-by: Sibi Sankar <quic_sibis@quicinc.com>
Reviewed-by: Rob Herring <robh@kernel.org>
Signed-off-by: Bjorn Andersson <bjorn.andersson@linaro.org>
Link: https://lore.kernel.org/r/1657020721-24939-2-git-send-email-quic_sibis@quicinc.com
|
|
arm64: tegra: Default configuration updates for v5.20-rc1
Updates the arm64 default configuration to enable the OPE driver as a
module.
* tag 'tegra-for-5.20-arm64-defconfig' of git://git.kernel.org/pub/scm/linux/kernel/git/tegra/linux:
arm64: defconfig: Build Tegra OPE module
Link: https://lore.kernel.org/r/20220708185608.676474-8-thierry.reding@gmail.com
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
|
|
- enable Allwinner analogue audio codecs
* tag 'sunxi-config64-for-5.20-1' of git://git.kernel.org/pub/scm/linux/kernel/git/sunxi/linux:
arm64: defconfig: Enable Allwinner built in CODECs
Link: https://lore.kernel.org/r/YsheC0hgyJB9kPsS@kista.localdomain
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
|
|
Qualcomm driver updates for v5.20
This introduces a new driver that requests interconnect bandwidth based
on throughput measurements of the bwmon hardware blocks found associated
with, among other things, the CPU subsystem on many Qualcomm platforms.
It introduces support for the SCM wrapper driver to vote for
interconnect bandwidth for operations that needs bandwidth to the crypto
engine. This ensures both performance and guards against issues caused
by lacking votes for this path.
The socinfo driver gains knowledge about the SC7180P SoC.
It contains a range of fixes for spelling mistakes, refcount leaks in
various drivers and removes some redundant code from the apr remove
path.
The SCM DT bindings are updated to declare support for QCS404, SM6125
and SDX65.
The command db driver has a strncpy() converted to strscpy_pad() and
then back again with proper documentation to why this was the right API.
* tag 'qcom-drivers-for-5.20' of git://git.kernel.org/pub/scm/linux/kernel/git/qcom/linux:
soc: qcom: icc-bwmon: Add bandwidth monitoring driver
dt-bindings: interconnect: qcom,msm8998-cpu-bwmon: add BWMON device
soc/qcom: Make QCOM_RPMPD select PM_GENERIC_DOMAINS/_OF
soc: qcom: aoss: Fix refcount leak in qmp_cooling_devices_register
soc: qcom: llcc: Fix syntax errors in comments
soc: qcom: ocmem: Fix refcount leak in of_get_ocmem
dt-bindings: soc: qcom,wcnss: remove unneeded ref for names
firmware: qcom_scm: Add bw voting support to the SCM interface
dt-bindings: firmware: qcom-scm: Add interconnects property
soc: qcom: cmd-db: replace strscpy_pad() with strncpy()
dt-bindings: firmware: scm: Add compatible for SDX65
dt-bindings: arm: qcom: switch maintainer to Bjorn
soc: qcom: rpmhpd: fix typos in comment
soc: qcom: correct kerneldoc
soc: qcom: cmd-db: replace strncpy() with strscpy_pad()
firmware: qcom_scm-legacy: correct kerneldoc
dt-bindings: firmware: document Qualcomm QCS404 and SM6125 SCM
soc: qcom: apr: Drop redundant check in .remove()
firmware: qcom_scm: drop unexpected word "the"
soc: qcom: socinfo: Add an ID for sc7180P
Link: https://lore.kernel.org/r/20220712021830.1271398-1-bjorn.andersson@linaro.org
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
|
|
This pull request contains Broadcom SoC drivers updatse for 5.20, please
pull the following:
- Julia fixes a typo in the Broadcom STB legacy power management code
- Liang fixes a device_node reference count leak in the Broadcom STB BIU
driver code error path(s)
- Nicolas and Stefan provide updates to the BCM2835 power management
driver allowing its use on BCM2711 (Raspberry Pi 4) and to enable the
use of the V3D GPU driver on such platforms. This is a merge of an
immutable branch from Lee Jones' MFD tree
- William removes the use of CONFIG_ARCH_BCM_63XX which is removed and
replaces the dependencies with CONFIG_ARCH_BCMBCA which is how all of
the DSL/PON SoCs from Broadcom are now supported in the upstream
kernel.
* tag 'arm-soc/for-5.20/drivers' of https://github.com/Broadcom/stblinux:
tty: serial: bcm63xx: bcmbca: Replace ARCH_BCM_63XX with ARCH_BCMBCA
spi: bcm63xx-hsspi: bcmbca: Replace ARCH_BCM_63XX with ARCH_BCMBCA
clk: bcm: bcmbca: Replace ARCH_BCM_63XX with ARCH_BCMBCA
hwrng: bcm2835: bcmbca: Replace ARCH_BCM_63XX with ARCH_BCMBCA
phy: brcm-sata: bcmbca: Replace ARCH_BCM_63XX with ARCH_BCMBCA
i2c: brcmstb: bcmbca: Replace ARCH_BCM_63XX with ARCH_BCMBCA
ata: ahci_brcm: bcmbca: Replace ARCH_BCM_63XX with ARCH_BCMBCA
soc: bcm: bcm2835-power: Bypass power_on/off() calls
soc: bcm: bcm2835-power: Add support for BCM2711's RPiVid ASB
soc: bcm: bcm2835-power: Resolve ASB register macros
soc: bcm: bcm2835-power: Refactor ASB control
mfd: bcm2835-pm: Add support for BCM2711
mfd: bcm2835-pm: Use 'reg-names' to get resources
soc: bcm: brcmstb: biuctrl: Add missing of_node_put()
soc: bcm: brcmstb: pm: pm-arm: fix typo in comment
Link: https://lore.kernel.org/r/20220711164451.3542127-6-f.fainelli@gmail.com
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
|
|
i.MX drivers change for 5.20:
- Update imx8m-blk-ctrl driver to print the name of the power domain,
so that the error message becomes more useful.
- Update the error messages in gpcv2 driver by printing the reason why
a regulator fails to be enabled or disabled.
- A series from Michael Walle to convert the guts driver from a platform
driver to an core_initcall for getting rid of all those global static
variables.
- A couple of follow-up fixes on guts driver series.
* tag 'imx-drivers-5.20' of git://git.kernel.org/pub/scm/linux/kernel/git/shawnguo/linux:
soc: fsl: guts: check return value after calling of_iomap() in fsl_guts_get_soc_uid()
soc: fsl: guts: fix return value check in fsl_guts_init()
soc: imx: imx8m-blk-ctrl: Make error prints useful
soc: fsl: guts: add serial_number support
soc: fsl: guts: drop platform driver
soc: fsl: guts: use of_root instead of own reference
soc: fsl: guts: allocate soc_dev_attr on the heap
soc: fsl: guts: embed fsl_guts_get_svr() in probe()
soc: fsl: guts: remove module_exit() and fsl_guts_remove()
soc: fsl: guts: machine variable might be unset
soc: imx: gpcv2: print errno for regulator errors
Link: https://lore.kernel.org/r/20220709082951.15123-1-shawnguo@kernel.org
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
|
|
TI Driver updates for v5.20
* pruss: Enable support for AM62, bindings cleanups
* wkup_m3_ipc: cosmetic cleanup
* tag 'ti-driver-soc-for-v5.20' of git://git.kernel.org/pub/scm/linux/kernel/git/ti/linux:
soc: ti: pruss: Enable support for PRUSS-M subsystem on K3 AM62x SoCs
dt-bindings: soc: ti: pruss: Update bindings for K3 AM62x SoCs
dt-bindings: soc: ti: pruss: Re-arrange "compatible" in alphabetic order
soc: ti: wkup_m3_ipc: Remove unneeded semicolon
Link: https://lore.kernel.org/r/20220708232655.reqjdbib52ejdja4@unadvised
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
|
|
memory: tegra: Changes for v5.20-rc1
Add memory client definitions for the Multi-Gigabit Ethernet (MGBE)
controllers found on Tegra234.
* tag 'tegra-for-5.20-memory' of git://git.kernel.org/pub/scm/linux/kernel/git/tegra/linux:
memory: tegra: Add MGBE memory clients for Tegra234
dt-bindings: arm: tegra: Add NVIDIA Tegra234 CBB 2.0 binding
dt-bindings: arm: tegra: Add NVIDIA Tegra194 AXI2APB binding
dt-bindings: arm: tegra: Add NVIDIA Tegra194 CBB 1.0 binding
dt-bindings: memory: Add Tegra234 MGBE memory clients
dt-bindings: Add Tegra234 MGBE clocks and resets
dt-bindings: power: Add Tegra234 MGBE power domains
dt-bindings: Add headers for Tegra234 GPCDMA
Link: https://lore.kernel.org/r/20220708185608.676474-5-thierry.reding@gmail.com
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
|
|
firmware: tegra: Changes for v5.20-rc1
Contains a fix that avoids unaligned accesses to IPC memory, which in
turn helps in suppressing speculative reads from out-of-bounds memory.
Another patch performs some minor cleanup around debugfs API usage.
* tag 'tegra-for-5.20-firmware' of git://git.kernel.org/pub/scm/linux/kernel/git/tegra/linux:
firmware: tegra: Fix error check return value of debugfs_create_file()
firmware: tegra: bpmp: Do only aligned access to IPC memory area
Link: https://lore.kernel.org/r/20220708185608.676474-3-thierry.reding@gmail.com
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
|
|
To expose unified memory for ctx save/resotre area feature
availablity to libhsakmt.
Proposed userspace:
https://patchwork.freedesktop.org/series/106218/
Signed-off-by: Eric Huang <jinhuieric.huang@amd.com>
Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
|
|
[Why]
When playing NV12 1080p MPO video, it is pipe splitting so
we see two pipes in fullscreen and four pipes in windowed
mode. Pipe split is happening because we are setting
MaximumMPCCombine = 1
[How]
Algorithm for MaximumMPCCombine has extra conditions we do
not need. Use DCN31 algorithm instead
Signed-off-by: Samson Tam <Samson.Tam@amd.com>
Reviewed-by: Rodrigo Siqueira <Rodrigo.Siqueira@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
|
|
[Why]
Black screen encountered when disabling Freesync through OSD on some
displays.
[How]
Set the should_disable flag when new top pipe has no plane state to
ensure that pipes get cleaned up.
Reviewed-by: Chris Park <Chris.Park@amd.com>
Acked-by: Solomon Chiu <solomon.chiu@amd.com>
Signed-off-by: Ilya Bakoulin <Ilya.Bakoulin@amd.com>
Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
|
|
[Why]
Some specific sink is not able to support PSRSU when DSC is turned on.
For this case, fall-back to use PSR1.
Reviewed-by: Anthony Koo <Anthony.Koo@amd.com>
Acked-by: Solomon Chiu <solomon.chiu@amd.com>
Signed-off-by: Robin Chen <po-tchen@amd.com>
Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
|
|
[Why]
When unplug one sst monitor from a mst hub and plug in the same
port with another sst monitor, we don't read the corresponding
edid. That's because we detect there is already an edid stored in
aconnector->edid which is a stale one.
[How]
Clean up aconnector->edid when unplug mst connector.
Reviewed-by: Hersen Wu <hersen.wu@amd.com>
Acked-by: Solomon Chiu <solomon.chiu@amd.com>
Signed-off-by: Wayne Lin <Wayne.Lin@amd.com>
Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
|
|
[Why]
In single display configuration, windowed MPO does not work
with ODM combine.
[How]
For ODM + MPO window on one half of ODM, only 3 pipes should
be allocated and scaling parameters adjusted to handle this case.
Otherwise, we use 4 pipes.
Move copy_surface_update_to_plane() before dc_add_plane_to_context()
so that it gets the updated rect information when setting up
the pipes.
Add dc_check_boundary_crossing_for_windowed_mpo_with_odm() to force
a full update when we cross a boundary requiring us to reconfigure
the number of pipes between 3 and 4 pipes.
Set config.enable_windowed_mpo_odm to true when we have the
debug.enable_single_display_2to1_odm_policy set to true.
Don't fail validating ODM with windowed MPO if
config.enable_windowed_mpo_odm is true.
Reviewed-by: Aric Cyr <Aric.Cyr@amd.com>
Acked-by: Solomon Chiu <solomon.chiu@amd.com>
Signed-off-by: Samson Tam <Samson.Tam@amd.com>
Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
|
|
[WHY]
Assert statements causing several bugs on Linux DM
[HOW]
Removing assert statement for Linux DM
(ASSERT(result == VBIOSSMC_Result_OK)). Also adding
logging statements for setting dcfclk.
Bug: https://bugzilla.kernel.org/show_bug.cgi?id=216092
Fixes: c1b972a18d05 ("drm/amd/display: Insert pulling smu busy status before sending another request")
Reviewed-by: Gabe Teeger <Gabe.Teeger@amd.com>
Acked-by: Solomon Chiu <solomon.chiu@amd.com>
Signed-off-by: Saaem Rizvi <SyedSaaem.Rizvi@amd.com>
Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
|
|
[WHY]
Needed a helper function for ALPM DPCD initialization
[HOW]
Refactoring to put ALPM initialization in a helper function
Reviewed-by: Anthony Koo <Anthony.Koo@amd.com>
Acked-by: Solomon Chiu <solomon.chiu@amd.com>
Signed-off-by: Muhammad Ansari <muansari@amd.com>
Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
|
|
[WHY]
Certain DP 2.0 modes may fail validation if DP 2.0 is not considered for
ODM combine.
Reviewed-by: Wenjing Liu <Wenjing.Liu@amd.com>
Acked-by: Solomon Chiu <solomon.chiu@amd.com>
Signed-off-by: Wesley Chalmers <Wesley.Chalmers@amd.com>
Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
|
|
[Description]
Exit SubVP if MPO is in use since SubVP + MPO together is not supported.
- Don't add SubVP at validation time if we see MPO is in use
Issues fixed in the SubVP / MPO transition:
1. Enable phantom pipes in post unlock function to prevent underflow
when an active pipe is being transitioned to be a phantom pipe (VTG
updates take place right away). Also must wait for VUPDATE of the main
pipe to complete first
2. Don't wait for MPCC idle when transitioning a phantom pipe to an
actual pipe. MPCC_STATUS is never asserted due to OTG being off for
phantom pipes
3. When transitioning an active pipe to phantom, program DET right away
(same as disabling the pipe) or the DET update will only take when
the phantom pipe is enabled which can cause DET allocation errors.
4. For K1/K2 programming of phantom pipes, use same settings as the
main pipe. Also don't program K1 / K2 = 0xF ever since the field is only
1 / 2 bits wide.
Reviewed-by: Jun Lei <Jun.Lei@amd.com>
Acked-by: Solomon Chiu <solomon.chiu@amd.com>
Signed-off-by: Alvin Lee <Alvin.Lee2@amd.com>
Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
|
|
Use try_cmpxchg instead of cmpxchg (*ptr, old, new) == old in
update_io_ticks. x86 CMPXCHG instruction returns success in ZF flag,
so this change saves a compare after cmpxchg (and related
move instruction in front of cmpxchg).
No functional change intended.
Signed-off-by: Uros Bizjak <ubizjak@gmail.com>
Cc: Jens Axboe <axboe@kernel.dk>
Link: https://lore.kernel.org/r/20220712152741.7324-1-ubizjak@gmail.com
Signed-off-by: Jens Axboe <axboe@kernel.dk>
|
|
Use atomic_try_cmpxchg instead of atomic_cmpxchg (*ptr, old, new) == old in
atomic_inc_below. x86 CMPXCHG instruction returns success in ZF flag,
so this change saves a compare after cmpxchg (and related move instruction
in front of cmpxchg).
Also, atomic_try_cmpxchg implicitly assigns old *ptr value to "old" when
cmpxchg fails, enabling further code simplifications.
No functional change intended.
Signed-off-by: Uros Bizjak <ubizjak@gmail.com>
Cc: Jens Axboe <axboe@kernel.dk>
Link: https://lore.kernel.org/r/20220712150547.5786-1-ubizjak@gmail.com
Signed-off-by: Jens Axboe <axboe@kernel.dk>
|
|
It is better and enough to use KSYM_NAME_LEN for kprobes
in samples, no need to define and use the other values.
Link: https://lkml.kernel.org/r/1654651402-21552-1-git-send-email-yangtiezhu@loongson.cn
Signed-off-by: Tiezhu Yang <yangtiezhu@loongson.cn>
Acked-by: Masami Hiramatsu (Google) <mhiramat@kernel.org>
Signed-off-by: Steven Rostedt (Google) <rostedt@goodmis.org>
|
|
This symbol is not used outside of fprobe_example.c, so marks it static.
Fixes the following warning:
sparse warnings: (new ones prefixed by >>)
>> samples/fprobe/fprobe_example.c:23:15: sparse: sparse: symbol 'sample_probe'
was not declared. Should it be static?
Link: https://lkml.kernel.org/r/20220606075659.674556-1-sunliming@kylinos.cn
Reported-by: kernel test robot <lkp@intel.com>
Signed-off-by: sunliming <sunliming@kylinos.cn>
Acked-by: Masami Hiramatsu (Google) <mhiramat@kernel.org>
Signed-off-by: Steven Rostedt (Google) <rostedt@goodmis.org>
|
|
Remove extra semicolon.
Link: https://lkml.kernel.org/r/20220629030013.10362-1-kunyu@nfschina.com
Cc: Tejun Heo <tj@kernel.org>
Cc: Jens Axboe <axboe@kernel.dk>
Signed-off-by: Li kunyu <kunyu@nfschina.com>
Signed-off-by: Steven Rostedt (Google) <rostedt@goodmis.org>
|
|
It was brought up that on ARMv7, that because the FUNCTION_TRACER does not
use nops to keep function tracing disabled because of the use of a link
register, it does have some performance impact.
The start of functions when -pg is used to compile the kernel is:
push {lr}
bl 8010e7c0 <__gnu_mcount_nc>
When function tracing is tuned off, it becomes:
push {lr}
add sp, sp, #4
Which just puts the stack back to its normal location. But these two
instructions at the start of every function does incur some overhead.
Be more honest in the Kconfig FUNCTION_TRACER description and specify that
the overhead being in the noise was x86 specific, but other architectures
may vary.
Link: https://lore.kernel.org/all/20220705105416.GE5208@pengutronix.de/
Link: https://lkml.kernel.org/r/20220706161231.085a83da@gandalf.local.home
Reported-by: Sascha Hauer <sha@pengutronix.de>
Acked-by: Sascha Hauer <s.hauer@pengutronix.de>
Signed-off-by: Steven Rostedt (Google) <rostedt@goodmis.org>
|
|
If you drop into kdb and type "ftdump" you'll get a sleeping while
atomic warning from memory allocation in trace_find_next_entry().
This appears to have been caused by commit ff895103a84a ("tracing:
Save off entry when peeking at next entry"), which added the
allocation in that path. The problematic commit was already fixed by
commit 8e99cf91b99b ("tracing: Do not allocate buffer in
trace_find_next_entry() in atomic") but that fix missed the kdb case.
The fix here is easy: just move the assignment of the static buffer to
the place where it should have been to begin with:
trace_init_global_iter(). That function is called in two places, once
is right before the assignment of the static buffer added by the
previous fix and once is in kdb.
Note that it appears that there's a second static buffer that we need
to assign that was added in commit efbbdaa22bb7 ("tracing: Show real
address for trace event arguments"), so we'll move that too.
Link: https://lkml.kernel.org/r/20220708170919.1.I75844e5038d9425add2ad853a608cb44bb39df40@changeid
Fixes: ff895103a84a ("tracing: Save off entry when peeking at next entry")
Fixes: efbbdaa22bb7 ("tracing: Show real address for trace event arguments")
Signed-off-by: Douglas Anderson <dianders@chromium.org>
Signed-off-by: Steven Rostedt (Google) <rostedt@goodmis.org>
|
|
This reverts commit 46bbe5c671e06f070428b9be142cc4ee5cedebac.
As commit 46bbe5c671e0 ("tracing: fix double free") said, the
"double free" problem reported by clang static analyzer is:
> In parse_var_defs() if there is a problem allocating
> var_defs.expr, the earlier var_defs.name is freed.
> This free is duplicated by free_var_defs() which frees
> the rest of the list.
However, if there is a problem allocating N-th var_defs.expr:
+ in parse_var_defs(), the freed 'earlier var_defs.name' is
actually the N-th var_defs.name;
+ then in free_var_defs(), the names from 0th to (N-1)-th are freed;
IF ALLOCATING PROBLEM HAPPENED HERE!!! -+
\
|
0th 1th (N-1)-th N-th V
+-------------+-------------+-----+-------------+-----------
var_defs: | name | expr | name | expr | ... | name | expr | name | ///
+-------------+-------------+-----+-------------+-----------
These two frees don't act on same name, so there was no "double free"
problem before. Conversely, after that commit, we get a "memory leak"
problem because the above "N-th var_defs.name" is not freed.
If enable CONFIG_DEBUG_KMEMLEAK and inject a fault at where the N-th
var_defs.expr allocated, then execute on shell like:
$ echo 'hist:key=call_site:val=$v1,$v2:v1=bytes_req,v2=bytes_alloc' > \
/sys/kernel/debug/tracing/events/kmem/kmalloc/trigger
Then kmemleak reports:
unreferenced object 0xffff8fb100ef3518 (size 8):
comm "bash", pid 196, jiffies 4295681690 (age 28.538s)
hex dump (first 8 bytes):
76 31 00 00 b1 8f ff ff v1......
backtrace:
[<0000000038fe4895>] kstrdup+0x2d/0x60
[<00000000c99c049a>] event_hist_trigger_parse+0x206f/0x20e0
[<00000000ae70d2cc>] trigger_process_regex+0xc0/0x110
[<0000000066737a4c>] event_trigger_write+0x75/0xd0
[<000000007341e40c>] vfs_write+0xbb/0x2a0
[<0000000087fde4c2>] ksys_write+0x59/0xd0
[<00000000581e9cdf>] do_syscall_64+0x3a/0x80
[<00000000cf3b065c>] entry_SYSCALL_64_after_hwframe+0x46/0xb0
Link: https://lkml.kernel.org/r/20220711014731.69520-1-zhengyejian1@huawei.com
Cc: stable@vger.kernel.org
Fixes: 46bbe5c671e0 ("tracing: fix double free")
Reported-by: Hulk Robot <hulkci@huawei.com>
Suggested-by: Steven Rostedt <rostedt@goodmis.org>
Reviewed-by: Tom Zanussi <tom.zanussi@linux.intel.com>
Signed-off-by: Zheng Yejian <zhengyejian1@huawei.com>
Signed-off-by: Steven Rostedt (Google) <rostedt@goodmis.org>
|
|
Set corresponding ready flag for mes ring when enable or disable
mes ring.
Signed-off-by: Jack Xiao <Jack.Xiao@amd.com>
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
|
|
That has been done in BO release notify.
Signed-off-by: xinhui pan <xinhui.pan@amd.com>
Acked-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
|
|
In init/Kconfig, the part dedicated to modules is quite large.
Move it into a dedicated Kconfig in kernel/module/
MODULES_TREE_LOOKUP was outside of the 'if MODULES', but as it is
only used when MODULES are set, move it in with everything else to
avoid confusion.
MODULE_SIG_FORMAT is left in init/Kconfig because this configuration
item is not used in kernel/modules/ but in kernel/ and can be
selected independently from CONFIG_MODULES. It is for instance
selected from security/integrity/ima/Kconfig.
Signed-off-by: Christophe Leroy <christophe.leroy@csgroup.eu>
Signed-off-by: Luis Chamberlain <mcgrof@kernel.org>
|
|
Windows 10/11 guests with Hyper-V role (WSL2) enabled are observed to
hang upon boot or shortly after when a non-default TSC frequency was
set for L1. The issue is observed on a host where TSC scaling is
supported. The problem appears to be that Windows doesn't use TSC
scaling for its guests, even when the feature is advertised, and KVM
filters SECONDARY_EXEC_TSC_SCALING out when creating L2 controls from
L1's VMCS. This leads to L2 running with the default frequency (matching
host's) while L1 is running with an altered one.
Keep SECONDARY_EXEC_TSC_SCALING in secondary exec controls for L2 when
it was set for L1. TSC_MULTIPLIER is already correctly computed and
written by prepare_vmcs02().
Signed-off-by: Vitaly Kuznetsov <vkuznets@redhat.com>
Fixes: d041b5ea93352b ("KVM: nVMX: Enable nested TSC scaling")
Cc: stable@vger.kernel.org
Reviewed-by: Maxim Levitsky <mlevitsk@redhat.com>
Link: https://lore.kernel.org/r/20220712135009.952805-1-vkuznets@redhat.com
Signed-off-by: Sean Christopherson <seanjc@google.com>
|
|
There is an unexpected word 'is' in a comments that need to be dropped
file: ./drivers/thermal/intel/x86_pkg_temp_thermal.c
line: 108
* tj-max is is interesting because threshold is set relative to this
changed to:
* tj-max is interesting because threshold is set relative to this
Signed-off-by: Jiang Jian <jiangjian@cdjrlc.com>
[ rjw: Subject and changelog edits ]
Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
|
|
Merge series from Peter Ujfalusi <peter.ujfalusi@linux.intel.com>:
The FSR (Firmware State Register) holds the ROM state information, it does not
contain error information.
The FSR itself is a bit more complicated as well as the state depends on the
module currently in use.
The error code from ROM or the status code from the firmware is located at the
next register.
Fix the handling of the FSR in order to provide usable and human readable (in
most cases) report on the status and error.
|
|
Merge series from Daniel Baluta <daniel.baluta@oss.nxp.com>
Daniel Baluta <daniel.baluta@nxp.com>:
From: Daniel Baluta <daniel.baluta@nxp.com>
We need a way to send extra parameters to DSP firmware. In order to do
this, we introduce ext_data array at the end of ipc_stream_params.
With this new addition we can send compress parameters.
This requires SOF ABI bump.
|
|
Merge series from Peter Ujfalusi <peter.ujfalusi@linux.intel.com>:
By mistake a developer managed to create a 'corrupted' IPC4 firmware image which
loaded fine to the DSP and after boot it sent an IPC reply before we would have
received the FW_READY message.
It turned out that the image was an IPC3 firmware and the IPC reply was the IPC3
FW_READY notification message which got understood as an IPC4 reply message due
to the difference between the two IPC mechanism.
This caused a NULL pointer dereference since the reply memory will be allocated
after the FW_READY message.
To make sure this will not bite again, skip any spurious reply messages before
the FW_READY.
|
|
Merge series from Peter Ujfalusi <peter.ujfalusi@linux.intel.com>:
There is no need to decouple a decoupled stream twice.
Keep the decoupling in hda_link_stream_assign() only as it is going to be
executed in all cases.
Drop the outdated comment from hda_link_dma_hw_params() as well since the code
has changed around it.
|
|
pcie_aspm_support_enabled() is used only by the acpi/pci_root.c driver,
which cannot be built as a module, so it does not need to be exported.
Unexport it.
Signed-off-by: Bjorn Helgaas <bhelgaas@google.com>
Reviewed-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
|
|
Replace the shouty macros here with typechecked helper functions.
Signed-off-by: Darrick J. Wong <djwong@kernel.org>
Reviewed-by: Dave Chinner <dchinner@redhat.com>
|
|
Replace this shouty macro with a real C function that has a more
descriptive name.
Signed-off-by: Darrick J. Wong <djwong@kernel.org>
Reviewed-by: Dave Chinner <dchinner@redhat.com>
|
Chris Wilson
Chris Wilson
Zheng Yejian
Xiang wangx
Arnd Bergmann
Sibi Sankar
Eric Huang
Samson Tam
Ilya Bakoulin
Robin Chen
Wayne Lin
Saaem Rizvi
Muhammad Ansari
Wesley Chalmers
Alvin Lee
Uros Bizjak
Tiezhu Yang
sunliming
Li kunyu
Steven Rostedt (Google)
Douglas Anderson
Jack Xiao
xinhui pan
Christophe Leroy
Vitaly Kuznetsov
Jiang Jian
Mark Brown
Bjorn Helgaas
Darrick J. Wong