Merge tag 'asoc-fix-v4.2-rc5' of git://git.kernel.org/pub/scm/linux/kernel/git/broonie/sound into for-linus

ASoC: Fixes for v4.2

There are a couple of small driver specific fixes here but the
overwhelming bulk of these changes are fixes to the topology ABI that
has been newly introduced in v4.2.  Once this makes it into a release we
will have to firm this up but for now getting enhancements in before
they've made it into a release is the most expedient thing.

54 files changed, 724 insertions, 319 deletions

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu.h b/drivers/gpu/drm/amd/amdgpu/amdgpu.h
index 01657830b470..31b00f91cfcd 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu.h
@@ -1614,6 +1614,9 @@ struct amdgpu_uvd {
 #define AMDGPU_MAX_VCE_HANDLES	16
 #define AMDGPU_VCE_FIRMWARE_OFFSET 256
 
+#define AMDGPU_VCE_HARVEST_VCE0 (1 << 0)
+#define AMDGPU_VCE_HARVEST_VCE1 (1 << 1)
+
 struct amdgpu_vce {
 	struct amdgpu_bo	*vcpu_bo;
 	uint64_t		gpu_addr;
@@ -1626,6 +1629,7 @@ struct amdgpu_vce {
 	const struct firmware	*fw;	/* VCE firmware */
 	struct amdgpu_ring	ring[AMDGPU_MAX_VCE_RINGS];
 	struct amdgpu_irq_src	irq;
+	unsigned		harvest_config;
 };
 
 /*
@@ -1862,6 +1866,12 @@ typedef void (*amdgpu_wreg_t)(struct amdgpu_device*, uint32_t, uint32_t);
 typedef uint32_t (*amdgpu_block_rreg_t)(struct amdgpu_device*, uint32_t, uint32_t);
 typedef void (*amdgpu_block_wreg_t)(struct amdgpu_device*, uint32_t, uint32_t, uint32_t);
 
+struct amdgpu_ip_block_status {
+	bool valid;
+	bool sw;
+	bool hw;
+};
+
 struct amdgpu_device {
 	struct device			*dev;
 	struct drm_device		*ddev;
@@ -2004,7 +2014,7 @@ struct amdgpu_device {
 
 	const struct amdgpu_ip_block_version *ip_blocks;
 	int				num_ip_blocks;
-	bool				*ip_block_enabled;
+	struct amdgpu_ip_block_status	*ip_block_status;
 	struct mutex	mn_lock;
 	DECLARE_HASHTABLE(mn_hash, 7);
 
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
index d79009b65867..99f158e1baff 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
@@ -1191,8 +1191,9 @@ static int amdgpu_early_init(struct amdgpu_device *adev)
 		return -EINVAL;
 	}
 
-	adev->ip_block_enabled = kcalloc(adev->num_ip_blocks, sizeof(bool), GFP_KERNEL);
-	if (adev->ip_block_enabled == NULL)
+	adev->ip_block_status = kcalloc(adev->num_ip_blocks,
+					sizeof(struct amdgpu_ip_block_status), GFP_KERNEL);
+	if (adev->ip_block_status == NULL)
 		return -ENOMEM;
 
 	if (adev->ip_blocks == NULL) {
@@ -1203,18 +1204,18 @@ static int amdgpu_early_init(struct amdgpu_device *adev)
 	for (i = 0; i < adev->num_ip_blocks; i++) {
 		if ((amdgpu_ip_block_mask & (1 << i)) == 0) {
 			DRM_ERROR("disabled ip block: %d\n", i);
-			adev->ip_block_enabled[i] = false;
+			adev->ip_block_status[i].valid = false;
 		} else {
 			if (adev->ip_blocks[i].funcs->early_init) {
 				r = adev->ip_blocks[i].funcs->early_init((void *)adev);
 				if (r == -ENOENT)
-					adev->ip_block_enabled[i] = false;
+					adev->ip_block_status[i].valid = false;
 				else if (r)
 					return r;
 				else
-					adev->ip_block_enabled[i] = true;
+					adev->ip_block_status[i].valid = true;
 			} else {
-				adev->ip_block_enabled[i] = true;
+				adev->ip_block_status[i].valid = true;
 			}
 		}
 	}
@@ -1227,11 +1228,12 @@ static int amdgpu_init(struct amdgpu_device *adev)
 	int i, r;
 
 	for (i = 0; i < adev->num_ip_blocks; i++) {
-		if (!adev->ip_block_enabled[i])
+		if (!adev->ip_block_status[i].valid)
 			continue;
 		r = adev->ip_blocks[i].funcs->sw_init((void *)adev);
 		if (r)
 			return r;
+		adev->ip_block_status[i].sw = true;
 		/* need to do gmc hw init early so we can allocate gpu mem */
 		if (adev->ip_blocks[i].type == AMD_IP_BLOCK_TYPE_GMC) {
 			r = amdgpu_vram_scratch_init(adev);
@@ -1243,11 +1245,12 @@ static int amdgpu_init(struct amdgpu_device *adev)
 			r = amdgpu_wb_init(adev);
 			if (r)
 				return r;
+			adev->ip_block_status[i].hw = true;
 		}
 	}
 
 	for (i = 0; i < adev->num_ip_blocks; i++) {
-		if (!adev->ip_block_enabled[i])
+		if (!adev->ip_block_status[i].sw)
 			continue;
 		/* gmc hw init is done early */
 		if (adev->ip_blocks[i].type == AMD_IP_BLOCK_TYPE_GMC)
@@ -1255,6 +1258,7 @@ static int amdgpu_init(struct amdgpu_device *adev)
 		r = adev->ip_blocks[i].funcs->hw_init((void *)adev);
 		if (r)
 			return r;
+		adev->ip_block_status[i].hw = true;
 	}
 
 	return 0;
@@ -1265,7 +1269,7 @@ static int amdgpu_late_init(struct amdgpu_device *adev)
 	int i = 0, r;
 
 	for (i = 0; i < adev->num_ip_blocks; i++) {
-		if (!adev->ip_block_enabled[i])
+		if (!adev->ip_block_status[i].valid)
 			continue;
 		/* enable clockgating to save power */
 		r = adev->ip_blocks[i].funcs->set_clockgating_state((void *)adev,
@@ -1287,7 +1291,7 @@ static int amdgpu_fini(struct amdgpu_device *adev)
 	int i, r;
 
 	for (i = adev->num_ip_blocks - 1; i >= 0; i--) {
-		if (!adev->ip_block_enabled[i])
+		if (!adev->ip_block_status[i].hw)
 			continue;
 		if (adev->ip_blocks[i].type == AMD_IP_BLOCK_TYPE_GMC) {
 			amdgpu_wb_fini(adev);
@@ -1300,14 +1304,16 @@ static int amdgpu_fini(struct amdgpu_device *adev)
 			return r;
 		r = adev->ip_blocks[i].funcs->hw_fini((void *)adev);
 		/* XXX handle errors */
+		adev->ip_block_status[i].hw = false;
 	}
 
 	for (i = adev->num_ip_blocks - 1; i >= 0; i--) {
-		if (!adev->ip_block_enabled[i])
+		if (!adev->ip_block_status[i].sw)
 			continue;
 		r = adev->ip_blocks[i].funcs->sw_fini((void *)adev);
 		/* XXX handle errors */
-		adev->ip_block_enabled[i] = false;
+		adev->ip_block_status[i].sw = false;
+		adev->ip_block_status[i].valid = false;
 	}
 
 	return 0;
@@ -1318,7 +1324,7 @@ static int amdgpu_suspend(struct amdgpu_device *adev)
 	int i, r;
 
 	for (i = adev->num_ip_blocks - 1; i >= 0; i--) {
-		if (!adev->ip_block_enabled[i])
+		if (!adev->ip_block_status[i].valid)
 			continue;
 		/* ungate blocks so that suspend can properly shut them down */
 		r = adev->ip_blocks[i].funcs->set_clockgating_state((void *)adev,
@@ -1336,7 +1342,7 @@ static int amdgpu_resume(struct amdgpu_device *adev)
 	int i, r;
 
 	for (i = 0; i < adev->num_ip_blocks; i++) {
-		if (!adev->ip_block_enabled[i])
+		if (!adev->ip_block_status[i].valid)
 			continue;
 		r = adev->ip_blocks[i].funcs->resume(adev);
 		if (r)
@@ -1582,8 +1588,8 @@ void amdgpu_device_fini(struct amdgpu_device *adev)
 	amdgpu_fence_driver_fini(adev);
 	amdgpu_fbdev_fini(adev);
 	r = amdgpu_fini(adev);
-	kfree(adev->ip_block_enabled);
-	adev->ip_block_enabled = NULL;
+	kfree(adev->ip_block_status);
+	adev->ip_block_status = NULL;
 	adev->accel_working = false;
 	/* free i2c buses */
 	amdgpu_i2c_fini(adev);
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c
index ae43b58c9733..4afc507820c0 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c
@@ -449,7 +449,7 @@ out:
  * vital here, so they are not reported back to userspace.
  */
 static void amdgpu_gem_va_update_vm(struct amdgpu_device *adev,
-				    struct amdgpu_bo_va *bo_va)
+				    struct amdgpu_bo_va *bo_va, uint32_t operation)
 {
 	struct ttm_validate_buffer tv, *entry;
 	struct amdgpu_bo_list_entry *vm_bos;
@@ -485,7 +485,9 @@ static void amdgpu_gem_va_update_vm(struct amdgpu_device *adev,
 	if (r)
 		goto error_unlock;
 
-	r = amdgpu_vm_bo_update(adev, bo_va, &bo_va->bo->tbo.mem);
+
+	if (operation == AMDGPU_VA_OP_MAP)
+		r = amdgpu_vm_bo_update(adev, bo_va, &bo_va->bo->tbo.mem);
 
 error_unlock:
 	mutex_unlock(&bo_va->vm->mutex);
@@ -580,7 +582,7 @@ int amdgpu_gem_va_ioctl(struct drm_device *dev, void *data,
 	}
 
 	if (!r && !(args->flags & AMDGPU_VM_DELAY_UPDATE))
-		amdgpu_gem_va_update_vm(adev, bo_va);
+		amdgpu_gem_va_update_vm(adev, bo_va, args->operation);
 
 	drm_gem_object_unreference_unlocked(gobj);
 	return r;
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ib.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ib.c
index 52dff75aac6f..bc0fac618a3f 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ib.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ib.c
@@ -180,16 +180,16 @@ int amdgpu_ib_schedule(struct amdgpu_device *adev, unsigned num_ibs,
 	if (vm) {
 		/* do context switch */
 		amdgpu_vm_flush(ring, vm, ib->sync.last_vm_update);
-	}
 
-	if (vm && 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);
+		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);
 
-	if (ring->funcs->emit_hdp_flush)
-		amdgpu_ring_emit_hdp_flush(ring);
+		if (ring->funcs->emit_hdp_flush)
+			amdgpu_ring_emit_hdp_flush(ring);
+	}
 
 	old_ctx = ring->current_ctx;
 	for (i = 0; i < num_ibs; ++i) {
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c
index 5533434c7a8f..9736892bcdf9 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c
@@ -235,7 +235,7 @@ static int amdgpu_info_ioctl(struct drm_device *dev, void *data, struct drm_file
 
 		for (i = 0; i < adev->num_ip_blocks; i++) {
 			if (adev->ip_blocks[i].type == type &&
-			    adev->ip_block_enabled[i]) {
+			    adev->ip_block_status[i].valid) {
 				ip.hw_ip_version_major = adev->ip_blocks[i].major;
 				ip.hw_ip_version_minor = adev->ip_blocks[i].minor;
 				ip.capabilities_flags = 0;
@@ -274,7 +274,7 @@ static int amdgpu_info_ioctl(struct drm_device *dev, void *data, struct drm_file
 
 		for (i = 0; i < adev->num_ip_blocks; i++)
 			if (adev->ip_blocks[i].type == type &&
-			    adev->ip_block_enabled[i] &&
+			    adev->ip_block_status[i].valid &&
 			    count < AMDGPU_HW_IP_INSTANCE_MAX_COUNT)
 				count++;
 
@@ -416,7 +416,7 @@ static int amdgpu_info_ioctl(struct drm_device *dev, void *data, struct drm_file
 		return n ? -EFAULT : 0;
 	}
 	case AMDGPU_INFO_DEV_INFO: {
-		struct drm_amdgpu_info_device dev_info;
+		struct drm_amdgpu_info_device dev_info = {};
 		struct amdgpu_cu_info cu_info;
 
 		dev_info.device_id = dev->pdev->device;
@@ -459,6 +459,7 @@ static int amdgpu_info_ioctl(struct drm_device *dev, void *data, struct drm_file
 		memcpy(&dev_info.cu_bitmap[0], &cu_info.bitmap[0], sizeof(cu_info.bitmap));
 		dev_info.vram_type = adev->mc.vram_type;
 		dev_info.vram_bit_width = adev->mc.vram_width;
+		dev_info.vce_harvest_config = adev->vce.harvest_config;
 
 		return copy_to_user(out, &dev_info,
 				    min((size_t)size, sizeof(dev_info))) ? -EFAULT : 0;
diff --git a/drivers/gpu/drm/amd/amdgpu/cz_dpm.c b/drivers/gpu/drm/amd/amdgpu/cz_dpm.c
index 1a2d419cbf16..ace870afc7d4 100644
--- a/drivers/gpu/drm/amd/amdgpu/cz_dpm.c
+++ b/drivers/gpu/drm/amd/amdgpu/cz_dpm.c
@@ -494,29 +494,67 @@ static void cz_dpm_fini(struct amdgpu_device *adev)
 	amdgpu_free_extended_power_table(adev);
 }
 
+#define ixSMUSVI_NB_CURRENTVID 0xD8230044
+#define CURRENT_NB_VID_MASK 0xff000000
+#define CURRENT_NB_VID__SHIFT 24
+#define ixSMUSVI_GFX_CURRENTVID  0xD8230048
+#define CURRENT_GFX_VID_MASK 0xff000000
+#define CURRENT_GFX_VID__SHIFT 24
+
 static void
 cz_dpm_debugfs_print_current_performance_level(struct amdgpu_device *adev,
 					       struct seq_file *m)
 {
+	struct cz_power_info *pi = cz_get_pi(adev);
 	struct amdgpu_clock_voltage_dependency_table *table =
 		&adev->pm.dpm.dyn_state.vddc_dependency_on_sclk;
-	u32 current_index =
-		(RREG32_SMC(ixTARGET_AND_CURRENT_PROFILE_INDEX) &
-		TARGET_AND_CURRENT_PROFILE_INDEX__CURR_SCLK_INDEX_MASK) >>
-		TARGET_AND_CURRENT_PROFILE_INDEX__CURR_SCLK_INDEX__SHIFT;
-	u32 sclk, tmp;
-	u16 vddc;
-
-	if (current_index >= NUM_SCLK_LEVELS) {
-		seq_printf(m, "invalid dpm profile %d\n", current_index);
+	struct amdgpu_uvd_clock_voltage_dependency_table *uvd_table =
+		&adev->pm.dpm.dyn_state.uvd_clock_voltage_dependency_table;
+	struct amdgpu_vce_clock_voltage_dependency_table *vce_table =
+		&adev->pm.dpm.dyn_state.vce_clock_voltage_dependency_table;
+	u32 sclk_index = REG_GET_FIELD(RREG32_SMC(ixTARGET_AND_CURRENT_PROFILE_INDEX),
+				       TARGET_AND_CURRENT_PROFILE_INDEX, CURR_SCLK_INDEX);
+	u32 uvd_index = REG_GET_FIELD(RREG32_SMC(ixTARGET_AND_CURRENT_PROFILE_INDEX_2),
+				      TARGET_AND_CURRENT_PROFILE_INDEX_2, CURR_UVD_INDEX);
+	u32 vce_index = REG_GET_FIELD(RREG32_SMC(ixTARGET_AND_CURRENT_PROFILE_INDEX_2),
+				      TARGET_AND_CURRENT_PROFILE_INDEX_2, CURR_VCE_INDEX);
+	u32 sclk, vclk, dclk, ecclk, tmp;
+	u16 vddnb, vddgfx;
+
+	if (sclk_index >= NUM_SCLK_LEVELS) {
+		seq_printf(m, "invalid sclk dpm profile %d\n", sclk_index);
 	} else {
-		sclk = table->entries[current_index].clk;
-		tmp = (RREG32_SMC(ixSMU_VOLTAGE_STATUS) &
-			SMU_VOLTAGE_STATUS__SMU_VOLTAGE_CURRENT_LEVEL_MASK) >>
-			SMU_VOLTAGE_STATUS__SMU_VOLTAGE_CURRENT_LEVEL__SHIFT;
-		vddc = cz_convert_8bit_index_to_voltage(adev, (u16)tmp);
-		seq_printf(m, "power level %d    sclk: %u vddc: %u\n",
-			   current_index, sclk, vddc);
+		sclk = table->entries[sclk_index].clk;
+		seq_printf(m, "%u sclk: %u\n", sclk_index, sclk);
+	}
+
+	tmp = (RREG32_SMC(ixSMUSVI_NB_CURRENTVID) &
+	       CURRENT_NB_VID_MASK) >> CURRENT_NB_VID__SHIFT;
+	vddnb = cz_convert_8bit_index_to_voltage(adev, (u16)tmp);
+	tmp = (RREG32_SMC(ixSMUSVI_GFX_CURRENTVID) &
+	       CURRENT_GFX_VID_MASK) >> CURRENT_GFX_VID__SHIFT;
+	vddgfx = cz_convert_8bit_index_to_voltage(adev, (u16)tmp);
+	seq_printf(m, "vddnb: %u vddgfx: %u\n", vddnb, vddgfx);
+
+	seq_printf(m, "uvd    %sabled\n", pi->uvd_power_gated ? "dis" : "en");
+	if (!pi->uvd_power_gated) {
+		if (uvd_index >= CZ_MAX_HARDWARE_POWERLEVELS) {
+			seq_printf(m, "invalid uvd dpm level %d\n", uvd_index);
+		} else {
+			vclk = uvd_table->entries[uvd_index].vclk;
+			dclk = uvd_table->entries[uvd_index].dclk;
+			seq_printf(m, "%u uvd vclk: %u dclk: %u\n", uvd_index, vclk, dclk);
+		}
+	}
+
+	seq_printf(m, "vce    %sabled\n", pi->vce_power_gated ? "dis" : "en");
+	if (!pi->vce_power_gated) {
+		if (vce_index >= CZ_MAX_HARDWARE_POWERLEVELS) {
+			seq_printf(m, "invalid vce dpm level %d\n", vce_index);
+		} else {
+			ecclk = vce_table->entries[vce_index].ecclk;
+			seq_printf(m, "%u vce ecclk: %u\n", vce_index, ecclk);
+		}
 	}
 }
 
diff --git a/drivers/gpu/drm/amd/amdgpu/dce_v10_0.c b/drivers/gpu/drm/amd/amdgpu/dce_v10_0.c
index 6e77964f1b64..e70a26f587a0 100644
--- a/drivers/gpu/drm/amd/amdgpu/dce_v10_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/dce_v10_0.c
@@ -2632,6 +2632,7 @@ static void dce_v10_0_crtc_dpms(struct drm_crtc *crtc, int mode)
 	struct drm_device *dev = crtc->dev;
 	struct amdgpu_device *adev = dev->dev_private;
 	struct amdgpu_crtc *amdgpu_crtc = to_amdgpu_crtc(crtc);
+	unsigned type;
 
 	switch (mode) {
 	case DRM_MODE_DPMS_ON:
@@ -2640,6 +2641,9 @@ static void dce_v10_0_crtc_dpms(struct drm_crtc *crtc, int mode)
 		dce_v10_0_vga_enable(crtc, true);
 		amdgpu_atombios_crtc_blank(crtc, ATOM_DISABLE);
 		dce_v10_0_vga_enable(crtc, false);
+		/* Make sure VBLANK interrupt is still enabled */
+		type = amdgpu_crtc_idx_to_irq_type(adev, amdgpu_crtc->crtc_id);
+		amdgpu_irq_update(adev, &adev->crtc_irq, type);
 		drm_vblank_post_modeset(dev, amdgpu_crtc->crtc_id);
 		dce_v10_0_crtc_load_lut(crtc);
 		break;
diff --git a/drivers/gpu/drm/amd/amdgpu/dce_v11_0.c b/drivers/gpu/drm/amd/amdgpu/dce_v11_0.c
index 7f7abb0e0be5..dcb402ee048a 100644
--- a/drivers/gpu/drm/amd/amdgpu/dce_v11_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/dce_v11_0.c
@@ -2631,6 +2631,7 @@ static void dce_v11_0_crtc_dpms(struct drm_crtc *crtc, int mode)
 	struct drm_device *dev = crtc->dev;
 	struct amdgpu_device *adev = dev->dev_private;
 	struct amdgpu_crtc *amdgpu_crtc = to_amdgpu_crtc(crtc);
+	unsigned type;
 
 	switch (mode) {
 	case DRM_MODE_DPMS_ON:
@@ -2639,6 +2640,9 @@ static void dce_v11_0_crtc_dpms(struct drm_crtc *crtc, int mode)
 		dce_v11_0_vga_enable(crtc, true);
 		amdgpu_atombios_crtc_blank(crtc, ATOM_DISABLE);
 		dce_v11_0_vga_enable(crtc, false);
+		/* Make sure VBLANK interrupt is still enabled */
+		type = amdgpu_crtc_idx_to_irq_type(adev, amdgpu_crtc->crtc_id);
+		amdgpu_irq_update(adev, &adev->crtc_irq, type);
 		drm_vblank_post_modeset(dev, amdgpu_crtc->crtc_id);
 		dce_v11_0_crtc_load_lut(crtc);
 		break;
diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v7_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v7_0.c
index 2c188fb9fd22..2db6ab0a543d 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfx_v7_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfx_v7_0.c
@@ -2561,7 +2561,7 @@ static bool gfx_v7_0_ring_emit_semaphore(struct amdgpu_ring *ring,
  * sheduling on the ring.  This function schedules the IB
  * on the gfx ring for execution by the GPU.
  */
-static void gfx_v7_0_ring_emit_ib(struct amdgpu_ring *ring,
+static void gfx_v7_0_ring_emit_ib_gfx(struct amdgpu_ring *ring,
 				  struct amdgpu_ib *ib)
 {
 	bool need_ctx_switch = ring->current_ctx != ib->ctx;
@@ -2569,15 +2569,10 @@ static void gfx_v7_0_ring_emit_ib(struct amdgpu_ring *ring,
 	u32 next_rptr = ring->wptr + 5;
 
 	/* drop the CE preamble IB for the same context */
-	if ((ring->type == AMDGPU_RING_TYPE_GFX) &&
-	    (ib->flags & AMDGPU_IB_FLAG_PREAMBLE) &&
-	    !need_ctx_switch)
+	if ((ib->flags & AMDGPU_IB_FLAG_PREAMBLE) && !need_ctx_switch)
 		return;
 
-	if (ring->type == AMDGPU_RING_TYPE_COMPUTE)
-		control |= INDIRECT_BUFFER_VALID;
-
-	if (need_ctx_switch && ring->type == AMDGPU_RING_TYPE_GFX)
+	if (need_ctx_switch)
 		next_rptr += 2;
 
 	next_rptr += 4;
@@ -2588,7 +2583,7 @@ static void gfx_v7_0_ring_emit_ib(struct amdgpu_ring *ring,
 	amdgpu_ring_write(ring, next_rptr);
 
 	/* insert SWITCH_BUFFER packet before first IB in the ring frame */
-	if (need_ctx_switch && ring->type == AMDGPU_RING_TYPE_GFX) {
+	if (need_ctx_switch) {
 		amdgpu_ring_write(ring, PACKET3(PACKET3_SWITCH_BUFFER, 0));
 		amdgpu_ring_write(ring, 0);
 	}
@@ -2611,6 +2606,35 @@ static void gfx_v7_0_ring_emit_ib(struct amdgpu_ring *ring,
 	amdgpu_ring_write(ring, control);
 }
 
+static void gfx_v7_0_ring_emit_ib_compute(struct amdgpu_ring *ring,
+				  struct amdgpu_ib *ib)
+{
+	u32 header, control = 0;
+	u32 next_rptr = ring->wptr + 5;
+
+	control |= INDIRECT_BUFFER_VALID;
+	next_rptr += 4;
+	amdgpu_ring_write(ring, PACKET3(PACKET3_WRITE_DATA, 3));
+	amdgpu_ring_write(ring, WRITE_DATA_DST_SEL(5) | WR_CONFIRM);
+	amdgpu_ring_write(ring, ring->next_rptr_gpu_addr & 0xfffffffc);
+	amdgpu_ring_write(ring, upper_32_bits(ring->next_rptr_gpu_addr) & 0xffffffff);
+	amdgpu_ring_write(ring, next_rptr);
+
+	header = PACKET3(PACKET3_INDIRECT_BUFFER, 2);
+
+	control |= ib->length_dw |
+			   (ib->vm ? (ib->vm->ids[ring->idx].id << 24) : 0);
+
+	amdgpu_ring_write(ring, header);
+	amdgpu_ring_write(ring,
+#ifdef __BIG_ENDIAN
+					  (2 << 0) |
+#endif
+					  (ib->gpu_addr & 0xFFFFFFFC));
+	amdgpu_ring_write(ring, upper_32_bits(ib->gpu_addr) & 0xFFFF);
+	amdgpu_ring_write(ring, control);
+}
+
 /**
  * gfx_v7_0_ring_test_ib - basic ring IB test
  *
@@ -5555,7 +5579,7 @@ static const struct amdgpu_ring_funcs gfx_v7_0_ring_funcs_gfx = {
 	.get_wptr = gfx_v7_0_ring_get_wptr_gfx,
 	.set_wptr = gfx_v7_0_ring_set_wptr_gfx,
 	.parse_cs = NULL,
-	.emit_ib = gfx_v7_0_ring_emit_ib,
+	.emit_ib = gfx_v7_0_ring_emit_ib_gfx,
 	.emit_fence = gfx_v7_0_ring_emit_fence_gfx,
 	.emit_semaphore = gfx_v7_0_ring_emit_semaphore,
 	.emit_vm_flush = gfx_v7_0_ring_emit_vm_flush,
@@ -5571,7 +5595,7 @@ static const struct amdgpu_ring_funcs gfx_v7_0_ring_funcs_compute = {
 	.get_wptr = gfx_v7_0_ring_get_wptr_compute,
 	.set_wptr = gfx_v7_0_ring_set_wptr_compute,
 	.parse_cs = NULL,
-	.emit_ib = gfx_v7_0_ring_emit_ib,
+	.emit_ib = gfx_v7_0_ring_emit_ib_compute,
 	.emit_fence = gfx_v7_0_ring_emit_fence_compute,
 	.emit_semaphore = gfx_v7_0_ring_emit_semaphore,
 	.emit_vm_flush = gfx_v7_0_ring_emit_vm_flush,
diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c
index 1c7c992dea37..9e1d4ddbf475 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfx_v8_0.c
@@ -3753,7 +3753,7 @@ 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_ib(struct amdgpu_ring *ring,
+static void gfx_v8_0_ring_emit_ib_gfx(struct amdgpu_ring *ring,
 				  struct amdgpu_ib *ib)
 {
 	bool need_ctx_switch = ring->current_ctx != ib->ctx;
@@ -3761,15 +3761,10 @@ static void gfx_v8_0_ring_emit_ib(struct amdgpu_ring *ring,
 	u32 next_rptr = ring->wptr + 5;
 
 	/* drop the CE preamble IB for the same context */
-	if ((ring->type == AMDGPU_RING_TYPE_GFX) &&
-	    (ib->flags & AMDGPU_IB_FLAG_PREAMBLE) &&
-	    !need_ctx_switch)
+	if ((ib->flags & AMDGPU_IB_FLAG_PREAMBLE) && !need_ctx_switch)
 		return;
 
-	if (ring->type == AMDGPU_RING_TYPE_COMPUTE)
-		control |= INDIRECT_BUFFER_VALID;
-
-	if (need_ctx_switch && ring->type == AMDGPU_RING_TYPE_GFX)
+	if (need_ctx_switch)
 		next_rptr += 2;
 
 	next_rptr += 4;
@@ -3780,7 +3775,7 @@ static void gfx_v8_0_ring_emit_ib(struct amdgpu_ring *ring,
 	amdgpu_ring_write(ring, next_rptr);
 
 	/* insert SWITCH_BUFFER packet before first IB in the ring frame */
-	if (need_ctx_switch && ring->type == AMDGPU_RING_TYPE_GFX) {
+	if (need_ctx_switch) {
 		amdgpu_ring_write(ring, PACKET3(PACKET3_SWITCH_BUFFER, 0));
 		amdgpu_ring_write(ring, 0);
 	}
@@ -3803,6 +3798,36 @@ static void gfx_v8_0_ring_emit_ib(struct amdgpu_ring *ring,
 	amdgpu_ring_write(ring, control);
 }
 
+static void gfx_v8_0_ring_emit_ib_compute(struct amdgpu_ring *ring,
+				  struct amdgpu_ib *ib)
+{
+	u32 header, control = 0;
+	u32 next_rptr = ring->wptr + 5;
+
+	control |= INDIRECT_BUFFER_VALID;
+
+	next_rptr += 4;
+	amdgpu_ring_write(ring, PACKET3(PACKET3_WRITE_DATA, 3));
+	amdgpu_ring_write(ring, WRITE_DATA_DST_SEL(5) | WR_CONFIRM);
+	amdgpu_ring_write(ring, ring->next_rptr_gpu_addr & 0xfffffffc);
+	amdgpu_ring_write(ring, upper_32_bits(ring->next_rptr_gpu_addr) & 0xffffffff);
+	amdgpu_ring_write(ring, next_rptr);
+
+	header = PACKET3(PACKET3_INDIRECT_BUFFER, 2);
+
+	control |= ib->length_dw |
+			   (ib->vm ? (ib->vm->ids[ring->idx].id << 24) : 0);
+
+	amdgpu_ring_write(ring, header);
+	amdgpu_ring_write(ring,
+#ifdef __BIG_ENDIAN
+					  (2 << 0) |
+#endif
+					  (ib->gpu_addr & 0xFFFFFFFC));
+	amdgpu_ring_write(ring, upper_32_bits(ib->gpu_addr) & 0xFFFF);
+	amdgpu_ring_write(ring, control);
+}
+
 static void gfx_v8_0_ring_emit_fence_gfx(struct amdgpu_ring *ring, u64 addr,
 					 u64 seq, unsigned flags)
 {
@@ -4224,7 +4249,7 @@ static const struct amdgpu_ring_funcs gfx_v8_0_ring_funcs_gfx = {
 	.get_wptr = gfx_v8_0_ring_get_wptr_gfx,
 	.set_wptr = gfx_v8_0_ring_set_wptr_gfx,
 	.parse_cs = NULL,
-	.emit_ib = gfx_v8_0_ring_emit_ib,
+	.emit_ib = gfx_v8_0_ring_emit_ib_gfx,
 	.emit_fence = gfx_v8_0_ring_emit_fence_gfx,
 	.emit_semaphore = gfx_v8_0_ring_emit_semaphore,
 	.emit_vm_flush = gfx_v8_0_ring_emit_vm_flush,
@@ -4240,7 +4265,7 @@ static const struct amdgpu_ring_funcs gfx_v8_0_ring_funcs_compute = {
 	.get_wptr = gfx_v8_0_ring_get_wptr_compute,
 	.set_wptr = gfx_v8_0_ring_set_wptr_compute,
 	.parse_cs = NULL,
-	.emit_ib = gfx_v8_0_ring_emit_ib,
+	.emit_ib = gfx_v8_0_ring_emit_ib_compute,
 	.emit_fence = gfx_v8_0_ring_emit_fence_compute,
 	.emit_semaphore = gfx_v8_0_ring_emit_semaphore,
 	.emit_vm_flush = gfx_v8_0_ring_emit_vm_flush,
diff --git a/drivers/gpu/drm/amd/amdgpu/vce_v3_0.c b/drivers/gpu/drm/amd/amdgpu/vce_v3_0.c
index d62c4002e39c..d1064ca3670e 100644
--- a/drivers/gpu/drm/amd/amdgpu/vce_v3_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/vce_v3_0.c
@@ -35,6 +35,8 @@
 #include "oss/oss_2_0_d.h"
 #include "oss/oss_2_0_sh_mask.h"
 #include "gca/gfx_8_0_d.h"
+#include "smu/smu_7_1_2_d.h"
+#include "smu/smu_7_1_2_sh_mask.h"
 
 #define GRBM_GFX_INDEX__VCE_INSTANCE__SHIFT	0x04
 #define GRBM_GFX_INDEX__VCE_INSTANCE_MASK	0x10
@@ -112,6 +114,10 @@ static int vce_v3_0_start(struct amdgpu_device *adev)
 
 	mutex_lock(&adev->grbm_idx_mutex);
 	for (idx = 0; idx < 2; ++idx) {
+
+		if (adev->vce.harvest_config & (1 << idx))
+			continue;
+
 		if(idx == 0)
 			WREG32_P(mmGRBM_GFX_INDEX, 0,
 				~GRBM_GFX_INDEX__VCE_INSTANCE_MASK);
@@ -190,10 +196,52 @@ static int vce_v3_0_start(struct amdgpu_device *adev)
 	return 0;
 }
 
+#define ixVCE_HARVEST_FUSE_MACRO__ADDRESS     0xC0014074
+#define VCE_HARVEST_FUSE_MACRO__SHIFT       27
+#define VCE_HARVEST_FUSE_MACRO__MASK        0x18000000
+
+static unsigned vce_v3_0_get_harvest_config(struct amdgpu_device *adev)
+{
+	u32 tmp;
+	unsigned ret;
+
+	if (adev->flags & AMDGPU_IS_APU)
+		tmp = (RREG32_SMC(ixVCE_HARVEST_FUSE_MACRO__ADDRESS) &
+		       VCE_HARVEST_FUSE_MACRO__MASK) >>
+			VCE_HARVEST_FUSE_MACRO__SHIFT;
+	else
+		tmp = (RREG32_SMC(ixCC_HARVEST_FUSES) &
+		       CC_HARVEST_FUSES__VCE_DISABLE_MASK) >>
+			CC_HARVEST_FUSES__VCE_DISABLE__SHIFT;
+
+	switch (tmp) {
+	case 1:
+		ret = AMDGPU_VCE_HARVEST_VCE0;
+		break;
+	case 2:
+		ret = AMDGPU_VCE_HARVEST_VCE1;
+		break;
+	case 3:
+		ret = AMDGPU_VCE_HARVEST_VCE0 | AMDGPU_VCE_HARVEST_VCE1;
+		break;
+	default:
+		ret = 0;
+	}
+
+	return ret;
+}
+
 static int vce_v3_0_early_init(void *handle)
 {
 	struct amdgpu_device *adev = (struct amdgpu_device *)handle;
 
+	adev->vce.harvest_config = vce_v3_0_get_harvest_config(adev);
+
+	if ((adev->vce.harvest_config &
+	     (AMDGPU_VCE_HARVEST_VCE0 | AMDGPU_VCE_HARVEST_VCE1)) ==
+	    (AMDGPU_VCE_HARVEST_VCE0 | AMDGPU_VCE_HARVEST_VCE1))
+		return -ENOENT;
+
 	vce_v3_0_set_ring_funcs(adev);
 	vce_v3_0_set_irq_funcs(adev);
 
diff --git a/drivers/gpu/drm/atmel-hlcdc/atmel_hlcdc_crtc.c b/drivers/gpu/drm/atmel-hlcdc/atmel_hlcdc_crtc.c
index f69b92535505..5ae5c6923128 100644
--- a/drivers/gpu/drm/atmel-hlcdc/atmel_hlcdc_crtc.c
+++ b/drivers/gpu/drm/atmel-hlcdc/atmel_hlcdc_crtc.c
@@ -355,6 +355,7 @@ int atmel_hlcdc_crtc_create(struct drm_device *dev)
 		planes->overlays[i]->base.possible_crtcs = 1 << crtc->id;
 
 	drm_crtc_helper_add(&crtc->base, &lcdc_crtc_helper_funcs);
+	drm_crtc_vblank_reset(&crtc->base);
 
 	dc->crtc = &crtc->base;
 
diff --git a/drivers/gpu/drm/atmel-hlcdc/atmel_hlcdc_dc.c b/drivers/gpu/drm/atmel-hlcdc/atmel_hlcdc_dc.c
index 60b0c13d7ff5..6fad1f9648f3 100644
--- a/drivers/gpu/drm/atmel-hlcdc/atmel_hlcdc_dc.c
+++ b/drivers/gpu/drm/atmel-hlcdc/atmel_hlcdc_dc.c
@@ -313,20 +313,20 @@ static int atmel_hlcdc_dc_load(struct drm_device *dev)
 
 	pm_runtime_enable(dev->dev);
 
-	ret = atmel_hlcdc_dc_modeset_init(dev);
+	ret = drm_vblank_init(dev, 1);
 	if (ret < 0) {
-		dev_err(dev->dev, "failed to initialize mode setting\n");
+		dev_err(dev->dev, "failed to initialize vblank\n");
 		goto err_periph_clk_disable;
 	}
 
-	drm_mode_config_reset(dev);
-
-	ret = drm_vblank_init(dev, 1);
+	ret = atmel_hlcdc_dc_modeset_init(dev);
 	if (ret < 0) {
-		dev_err(dev->dev, "failed to initialize vblank\n");
+		dev_err(dev->dev, "failed to initialize mode setting\n");
 		goto err_periph_clk_disable;
 	}
 
+	drm_mode_config_reset(dev);
+
 	pm_runtime_get_sync(dev->dev);
 	ret = drm_irq_install(dev, dc->hlcdc->irq);
 	pm_runtime_put_sync(dev->dev);
diff --git a/drivers/gpu/drm/drm_atomic_helper.c b/drivers/gpu/drm/drm_atomic_helper.c
index 5b59d5ad7d1c..aac212297b49 100644
--- a/drivers/gpu/drm/drm_atomic_helper.c
+++ b/drivers/gpu/drm/drm_atomic_helper.c
@@ -230,10 +230,12 @@ update_connector_routing(struct drm_atomic_state *state, int conn_idx)
 	}
 
 	connector_state->best_encoder = new_encoder;
-	idx = drm_crtc_index(connector_state->crtc);
+	if (connector_state->crtc) {
+		idx = drm_crtc_index(connector_state->crtc);
 
-	crtc_state = state->crtc_states[idx];
-	crtc_state->mode_changed = true;
+		crtc_state = state->crtc_states[idx];
+		crtc_state->mode_changed = true;
+	}
 
 	DRM_DEBUG_ATOMIC("[CONNECTOR:%d:%s] using [ENCODER:%d:%s] on [CRTC:%d]\n",
 			 connector->base.id,
diff --git a/drivers/gpu/drm/drm_crtc.c b/drivers/gpu/drm/drm_crtc.c
index 357bd04a173b..fed748311b92 100644
--- a/drivers/gpu/drm/drm_crtc.c
+++ b/drivers/gpu/drm/drm_crtc.c
@@ -5398,12 +5398,9 @@ void drm_mode_config_reset(struct drm_device *dev)
 		if (encoder->funcs->reset)
 			encoder->funcs->reset(encoder);
 
-	list_for_each_entry(connector, &dev->mode_config.connector_list, head) {
-		connector->status = connector_status_unknown;
-
+	list_for_each_entry(connector, &dev->mode_config.connector_list, head)
 		if (connector->funcs->reset)
 			connector->funcs->reset(connector);
-	}
 }
 EXPORT_SYMBOL(drm_mode_config_reset);
 
diff --git a/drivers/gpu/drm/i915/i915_drv.h b/drivers/gpu/drm/i915/i915_drv.h
index 5f27290201e0..fd1de451c8c6 100644
--- a/drivers/gpu/drm/i915/i915_drv.h
+++ b/drivers/gpu/drm/i915/i915_drv.h
@@ -3303,15 +3303,14 @@ int intel_freq_opcode(struct drm_i915_private *dev_priv, int val);
 #define I915_READ64(reg)	dev_priv->uncore.funcs.mmio_readq(dev_priv, (reg), true)
 
 #define I915_READ64_2x32(lower_reg, upper_reg) ({			\
-		u32 upper = I915_READ(upper_reg);			\
-		u32 lower = I915_READ(lower_reg);			\
-		u32 tmp = I915_READ(upper_reg);				\
-		if (upper != tmp) {					\
-			upper = tmp;					\
-			lower = I915_READ(lower_reg);			\
-			WARN_ON(I915_READ(upper_reg) != upper);		\
-		}							\
-		(u64)upper << 32 | lower; })
+	u32 upper, lower, tmp;						\
+	tmp = I915_READ(upper_reg);					\
+	do {								\
+		upper = tmp;						\
+		lower = I915_READ(lower_reg);				\
+		tmp = I915_READ(upper_reg);				\
+	} while (upper != tmp);						\
+	(u64)upper << 32 | lower; })
 
 #define POSTING_READ(reg)	(void)I915_READ_NOTRACE(reg)
 #define POSTING_READ16(reg)	(void)I915_READ16_NOTRACE(reg)
diff --git a/drivers/gpu/drm/i915/i915_gem_gtt.c b/drivers/gpu/drm/i915/i915_gem_gtt.c
index 56b52a4767d4..31e8269e6e3d 100644
--- a/drivers/gpu/drm/i915/i915_gem_gtt.c
+++ b/drivers/gpu/drm/i915/i915_gem_gtt.c
@@ -1923,6 +1923,17 @@ static int ggtt_bind_vma(struct i915_vma *vma,
 		vma->vm->insert_entries(vma->vm, pages,
 					vma->node.start,
 					cache_level, pte_flags);
+
+		/* Note the inconsistency here is due to absence of the
+		 * aliasing ppgtt on gen4 and earlier. Though we always
+		 * request PIN_USER for execbuffer (translated to LOCAL_BIND),
+		 * without the appgtt, we cannot honour that request and so
+		 * must substitute it with a global binding. Since we do this
+		 * behind the upper layers back, we need to explicitly set
+		 * the bound flag ourselves.
+		 */
+		vma->bound |= GLOBAL_BIND;
+
 	}
 
 	if (dev_priv->mm.aliasing_ppgtt && flags & LOCAL_BIND) {
diff --git a/drivers/gpu/drm/i915/i915_gem_tiling.c b/drivers/gpu/drm/i915/i915_gem_tiling.c
index 633bd1fcab69..d19c9db5e18c 100644
--- a/drivers/gpu/drm/i915/i915_gem_tiling.c
+++ b/drivers/gpu/drm/i915/i915_gem_tiling.c
@@ -464,7 +464,10 @@ i915_gem_get_tiling(struct drm_device *dev, void *data,
 	}
 
 	/* Hide bit 17 from the user -- see comment in i915_gem_set_tiling */
-	args->phys_swizzle_mode = args->swizzle_mode;
+	if (dev_priv->quirks & QUIRK_PIN_SWIZZLED_PAGES)
+		args->phys_swizzle_mode = I915_BIT_6_SWIZZLE_UNKNOWN;
+	else
+		args->phys_swizzle_mode = args->swizzle_mode;
 	if (args->swizzle_mode == I915_BIT_6_SWIZZLE_9_17)
 		args->swizzle_mode = I915_BIT_6_SWIZZLE_9;
 	if (args->swizzle_mode == I915_BIT_6_SWIZZLE_9_10_17)
diff --git a/drivers/gpu/drm/i915/intel_uncore.c b/drivers/gpu/drm/i915/intel_uncore.c
index a6d8a3ee7750..260389acfb77 100644
--- a/drivers/gpu/drm/i915/intel_uncore.c
+++ b/drivers/gpu/drm/i915/intel_uncore.c
@@ -1274,10 +1274,12 @@ int i915_reg_read_ioctl(struct drm_device *dev,
 	struct drm_i915_private *dev_priv = dev->dev_private;
 	struct drm_i915_reg_read *reg = data;
 	struct register_whitelist const *entry = whitelist;
+	unsigned size;
+	u64 offset;
 	int i, ret = 0;
 
 	for (i = 0; i < ARRAY_SIZE(whitelist); i++, entry++) {
-		if (entry->offset == reg->offset &&
+		if (entry->offset == (reg->offset & -entry->size) &&
 		    (1 << INTEL_INFO(dev)->gen & entry->gen_bitmask))
 			break;
 	}
@@ -1285,23 +1287,33 @@ int i915_reg_read_ioctl(struct drm_device *dev,
 	if (i == ARRAY_SIZE(whitelist))
 		return -EINVAL;
 
+	/* We use the low bits to encode extra flags as the register should
+	 * be naturally aligned (and those that are not so aligned merely
+	 * limit the available flags for that register).
+	 */
+	offset = entry->offset;
+	size = entry->size;
+	size |= reg->offset ^ offset;
+
 	intel_runtime_pm_get(dev_priv);
 
-	switch (entry->size) {
+	switch (size) {
+	case 8 | 1:
+		reg->val = I915_READ64_2x32(offset, offset+4);
+		break;
 	case 8:
-		reg->val = I915_READ64(reg->offset);
+		reg->val = I915_READ64(offset);
 		break;
 	case 4:
-		reg->val = I915_READ(reg->offset);
+		reg->val = I915_READ(offset);
 		break;
 	case 2:
-		reg->val = I915_READ16(reg->offset);
+		reg->val = I915_READ16(offset);
 		break;
 	case 1:
-		reg->val = I915_READ8(reg->offset);
+		reg->val = I915_READ8(offset);
 		break;
 	default:
-		MISSING_CASE(entry->size);
 		ret = -EINVAL;
 		goto out;
 	}
diff --git a/drivers/gpu/drm/msm/mdp/mdp4/mdp4_plane.c b/drivers/gpu/drm/msm/mdp/mdp4/mdp4_plane.c
index 0d1dbb737933..247a424445f7 100644
--- a/drivers/gpu/drm/msm/mdp/mdp4/mdp4_plane.c
+++ b/drivers/gpu/drm/msm/mdp/mdp4/mdp4_plane.c
@@ -220,13 +220,15 @@ static int mdp4_plane_mode_set(struct drm_plane *plane,
 	uint32_t op_mode = 0;
 	uint32_t phasex_step = MDP4_VG_PHASE_STEP_DEFAULT;
 	uint32_t phasey_step = MDP4_VG_PHASE_STEP_DEFAULT;
-	enum mdp4_frame_format frame_type = mdp4_get_frame_format(fb);
+	enum mdp4_frame_format frame_type;
 
 	if (!(crtc && fb)) {
 		DBG("%s: disabled!", mdp4_plane->name);
 		return 0;
 	}
 
+	frame_type = mdp4_get_frame_format(fb);
+
 	/* src values are in Q16 fixed point, convert to integer: */
 	src_x = src_x >> 16;
 	src_y = src_y >> 16;
diff --git a/drivers/gpu/drm/msm/mdp/mdp5/mdp5_kms.c b/drivers/gpu/drm/msm/mdp/mdp5/mdp5_kms.c
index 206f758f7d64..e253db5de5aa 100644
--- a/drivers/gpu/drm/msm/mdp/mdp5/mdp5_kms.c
+++ b/drivers/gpu/drm/msm/mdp/mdp5/mdp5_kms.c
@@ -76,7 +76,20 @@ static void mdp5_prepare_commit(struct msm_kms *kms, struct drm_atomic_state *st
 
 static void mdp5_complete_commit(struct msm_kms *kms, struct drm_atomic_state *state)
 {
+	int i;
 	struct mdp5_kms *mdp5_kms = to_mdp5_kms(to_mdp_kms(kms));
+	int nplanes = mdp5_kms->dev->mode_config.num_total_plane;
+
+	for (i = 0; i < nplanes; i++) {
+		struct drm_plane *plane = state->planes[i];
+		struct drm_plane_state *plane_state = state->plane_states[i];
+
+		if (!plane)
+			continue;
+
+		mdp5_plane_complete_commit(plane, plane_state);
+	}
+
 	mdp5_disable(mdp5_kms);
 }
 
diff --git a/drivers/gpu/drm/msm/mdp/mdp5/mdp5_kms.h b/drivers/gpu/drm/msm/mdp/mdp5/mdp5_kms.h
index e0eb24587c84..e79ac09b7216 100644
--- a/drivers/gpu/drm/msm/mdp/mdp5/mdp5_kms.h
+++ b/drivers/gpu/drm/msm/mdp/mdp5/mdp5_kms.h
@@ -227,6 +227,8 @@ void mdp5_plane_install_properties(struct drm_plane *plane,
 		struct drm_mode_object *obj);
 uint32_t mdp5_plane_get_flush(struct drm_plane *plane);
 void mdp5_plane_complete_flip(struct drm_plane *plane);
+void mdp5_plane_complete_commit(struct drm_plane *plane,
+	struct drm_plane_state *state);
 enum mdp5_pipe mdp5_plane_pipe(struct drm_plane *plane);
 struct drm_plane *mdp5_plane_init(struct drm_device *dev,
 		enum mdp5_pipe pipe, bool private_plane, uint32_t reg_offset);
diff --git a/drivers/gpu/drm/msm/mdp/mdp5/mdp5_plane.c b/drivers/gpu/drm/msm/mdp/mdp5/mdp5_plane.c
index 57b8f56ae9d0..22275568ab8b 100644
--- a/drivers/gpu/drm/msm/mdp/mdp5/mdp5_plane.c
+++ b/drivers/gpu/drm/msm/mdp/mdp5/mdp5_plane.c
@@ -31,8 +31,6 @@ struct mdp5_plane {
 
 	uint32_t nformats;
 	uint32_t formats[32];
-
-	bool enabled;
 };
 #define to_mdp5_plane(x) container_of(x, struct mdp5_plane, base)
 
@@ -56,22 +54,6 @@ static bool plane_enabled(struct drm_plane_state *state)
 	return state->fb && state->crtc;
 }
 
-static int mdp5_plane_disable(struct drm_plane *plane)
-{
-	struct mdp5_plane *mdp5_plane = to_mdp5_plane(plane);
-	struct mdp5_kms *mdp5_kms = get_kms(plane);
-	enum mdp5_pipe pipe = mdp5_plane->pipe;
-
-	DBG("%s: disable", mdp5_plane->name);
-
-	if (mdp5_kms) {
-		/* Release the memory we requested earlier from the SMP: */
-		mdp5_smp_release(mdp5_kms->smp, pipe);
-	}
-
-	return 0;
-}
-
 static void mdp5_plane_destroy(struct drm_plane *plane)
 {
 	struct mdp5_plane *mdp5_plane = to_mdp5_plane(plane);
@@ -224,7 +206,6 @@ static void mdp5_plane_atomic_update(struct drm_plane *plane,
 
 	if (!plane_enabled(state)) {
 		to_mdp5_plane_state(state)->pending = true;
-		mdp5_plane_disable(plane);
 	} else if (to_mdp5_plane_state(state)->mode_changed) {
 		int ret;
 		to_mdp5_plane_state(state)->pending = true;
@@ -602,6 +583,20 @@ uint32_t mdp5_plane_get_flush(struct drm_plane *plane)
 	return mdp5_plane->flush_mask;
 }
 
+/* called after vsync in thread context */
+void mdp5_plane_complete_commit(struct drm_plane *plane,
+	struct drm_plane_state *state)
+{
+	struct mdp5_kms *mdp5_kms = get_kms(plane);
+	struct mdp5_plane *mdp5_plane = to_mdp5_plane(plane);
+	enum mdp5_pipe pipe = mdp5_plane->pipe;
+
+	if (!plane_enabled(plane->state)) {
+		DBG("%s: free SMP", mdp5_plane->name);
+		mdp5_smp_release(mdp5_kms->smp, pipe);
+	}
+}
+
 /* initialize plane */
 struct drm_plane *mdp5_plane_init(struct drm_device *dev,
 		enum mdp5_pipe pipe, bool private_plane, uint32_t reg_offset)
diff --git a/drivers/gpu/drm/msm/mdp/mdp5/mdp5_smp.c b/drivers/gpu/drm/msm/mdp/mdp5/mdp5_smp.c
index 16702aecf0df..64a27d86f2f5 100644
--- a/drivers/gpu/drm/msm/mdp/mdp5/mdp5_smp.c
+++ b/drivers/gpu/drm/msm/mdp/mdp5/mdp5_smp.c
@@ -34,22 +34,44 @@
  * and CANNOT be re-allocated (eg: MMB0 and MMB1 both tied to RGB0).
  *
  * For each block that can be dynamically allocated, it can be either
- * free, or pending/in-use by a client. The updates happen in three steps:
+ *     free:
+ *     The block is free.
+ *
+ *     pending:
+ *     The block is allocated to some client and not free.
+ *
+ *     configured:
+ *     The block is allocated to some client, and assigned to that
+ *     client in MDP5_MDP_SMP_ALLOC registers.
+ *
+ *     inuse:
+ *     The block is being actively used by a client.
+ *
+ * The updates happen in the following steps:
  *
  *  1) mdp5_smp_request():
  *     When plane scanout is setup, calculate required number of
- *     blocks needed per client, and request.  Blocks not inuse or
- *     pending by any other client are added to client's pending
- *     set.
+ *     blocks needed per client, and request. Blocks neither inuse nor
+ *     configured nor pending by any other client are added to client's
+ *     pending set.
+ *     For shrinking, blocks in pending but not in configured can be freed
+ *     directly, but those already in configured will be freed later by
+ *     mdp5_smp_commit.
  *
  *  2) mdp5_smp_configure():
  *     As hw is programmed, before FLUSH, MDP5_MDP_SMP_ALLOC registers
  *     are configured for the union(pending, inuse)
+ *     Current pending is copied to configured.
+ *     It is assumed that mdp5_smp_request and mdp5_smp_configure not run
+ *     concurrently for the same pipe.
  *
  *  3) mdp5_smp_commit():
- *     After next vblank, copy pending -> inuse.  Optionally update
+ *     After next vblank, copy configured -> inuse.  Optionally update
  *     MDP5_SMP_ALLOC registers if there are newly unused blocks
  *
+ *  4) mdp5_smp_release():
+ *     Must be called after the pipe is disabled and no longer uses any SMB
+ *
  * On the next vblank after changes have been committed to hw, the
  * client's pending blocks become it's in-use blocks (and no-longer
  * in-use blocks become available to other clients).
@@ -77,6 +99,9 @@ struct mdp5_smp {
 	struct mdp5_client_smp_state client_state[MAX_CLIENTS];
 };
 
+static void update_smp_state(struct mdp5_smp *smp,
+		u32 cid, mdp5_smp_state_t *assigned);
+
 static inline
 struct mdp5_kms *get_kms(struct mdp5_smp *smp)
 {
@@ -149,7 +174,12 @@ static int smp_request_block(struct mdp5_smp *smp,
 		for (i = cur_nblks; i > nblks; i--) {
 			int blk = find_first_bit(ps->pending, cnt);
 			clear_bit(blk, ps->pending);
-			/* don't clear in global smp_state until _commit() */
+
+			/* clear in global smp_state if not in configured
+			 * otherwise until _commit()
+			 */
+			if (!test_bit(blk, ps->configured))
+				clear_bit(blk, smp->state);
 		}
 	}
 
@@ -223,10 +253,33 @@ int mdp5_smp_request(struct mdp5_smp *smp, enum mdp5_pipe pipe, u32 fmt, u32 wid
 /* Release SMP blocks for all clients of the pipe */
 void mdp5_smp_release(struct mdp5_smp *smp, enum mdp5_pipe pipe)
 {
-	int i, nblks;
+	int i;
+	unsigned long flags;
+	int cnt = smp->blk_cnt;
+
+	for (i = 0; i < pipe2nclients(pipe); i++) {
+		mdp5_smp_state_t assigned;
+		u32 cid = pipe2client(pipe, i);
+		struct mdp5_client_smp_state *ps = &smp->client_state[cid];
+
+		spin_lock_irqsave(&smp->state_lock, flags);
+
+		/* clear hw assignment */
+		bitmap_or(assigned, ps->inuse, ps->configured, cnt);
+		update_smp_state(smp, CID_UNUSED, &assigned);
+
+		/* free to global pool */
+		bitmap_andnot(smp->state, smp->state, ps->pending, cnt);
+		bitmap_andnot(smp->state, smp->state, assigned, cnt);
+
+		/* clear client's infor */
+		bitmap_zero(ps->pending, cnt);
+		bitmap_zero(ps->configured, cnt);
+		bitmap_zero(ps->inuse, cnt);
+
+		spin_unlock_irqrestore(&smp->state_lock, flags);
+	}
 
-	for (i = 0, nblks = 0; i < pipe2nclients(pipe); i++)
-		smp_request_block(smp, pipe2client(pipe, i), 0);
 	set_fifo_thresholds(smp, pipe, 0);
 }
 
@@ -274,12 +327,20 @@ void mdp5_smp_configure(struct mdp5_smp *smp, enum mdp5_pipe pipe)
 		u32 cid = pipe2client(pipe, i);
 		struct mdp5_client_smp_state *ps = &smp->client_state[cid];
 
-		bitmap_or(assigned, ps->inuse, ps->pending, cnt);
+		/*
+		 * if vblank has not happened since last smp_configure
+		 * skip the configure for now
+		 */
+		if (!bitmap_equal(ps->inuse, ps->configured, cnt))
+			continue;
+
+		bitmap_copy(ps->configured, ps->pending, cnt);
+		bitmap_or(assigned, ps->inuse, ps->configured, cnt);
 		update_smp_state(smp, cid, &assigned);
 	}
 }
 
-/* step #3: after vblank, copy pending -> inuse: */
+/* step #3: after vblank, copy configured -> inuse: */
 void mdp5_smp_commit(struct mdp5_smp *smp, enum mdp5_pipe pipe)
 {
 	int cnt = smp->blk_cnt;
@@ -295,7 +356,7 @@ void mdp5_smp_commit(struct mdp5_smp *smp, enum mdp5_pipe pipe)
 		 * using, which can be released and made available to other
 		 * clients:
 		 */
-		if (bitmap_andnot(released, ps->inuse, ps->pending, cnt)) {
+		if (bitmap_andnot(released, ps->inuse, ps->configured, cnt)) {
 			unsigned long flags;
 
 			spin_lock_irqsave(&smp->state_lock, flags);
@@ -306,7 +367,7 @@ void mdp5_smp_commit(struct mdp5_smp *smp, enum mdp5_pipe pipe)
 			update_smp_state(smp, CID_UNUSED, &released);
 		}
 
-		bitmap_copy(ps->inuse, ps->pending, cnt);
+		bitmap_copy(ps->inuse, ps->configured, cnt);
 	}
 }
 
diff --git a/drivers/gpu/drm/msm/mdp/mdp5/mdp5_smp.h b/drivers/gpu/drm/msm/mdp/mdp5/mdp5_smp.h
index e47179f63585..5b6c2363f592 100644
--- a/drivers/gpu/drm/msm/mdp/mdp5/mdp5_smp.h
+++ b/drivers/gpu/drm/msm/mdp/mdp5/mdp5_smp.h
@@ -23,6 +23,7 @@
 
 struct mdp5_client_smp_state {
 	mdp5_smp_state_t inuse;
+	mdp5_smp_state_t configured;
 	mdp5_smp_state_t pending;
 };
 
diff --git a/drivers/gpu/drm/msm/msm_atomic.c b/drivers/gpu/drm/msm/msm_atomic.c
index 1b22d8bfe142..1ceb4f22dd89 100644
--- a/drivers/gpu/drm/msm/msm_atomic.c
+++ b/drivers/gpu/drm/msm/msm_atomic.c
@@ -283,12 +283,8 @@ int msm_atomic_commit(struct drm_device *dev,
 
 	timeout = ktime_add_ms(ktime_get(), 1000);
 
-	ret = msm_wait_fence_interruptable(dev, c->fence, &timeout);
-	if (ret) {
-		WARN_ON(ret);  // TODO unswap state back?  or??
-		commit_destroy(c);
-		return ret;
-	}
+	/* uninterruptible wait */
+	msm_wait_fence(dev, c->fence, &timeout, false);
 
 	complete_commit(c);
 
diff --git a/drivers/gpu/drm/msm/msm_drv.c b/drivers/gpu/drm/msm/msm_drv.c
index b7ef56ed8d1c..d3467b115e04 100644
--- a/drivers/gpu/drm/msm/msm_drv.c
+++ b/drivers/gpu/drm/msm/msm_drv.c
@@ -637,8 +637,8 @@ static void msm_debugfs_cleanup(struct drm_minor *minor)
  * Fences:
  */
 
-int msm_wait_fence_interruptable(struct drm_device *dev, uint32_t fence,
-		ktime_t *timeout)
+int msm_wait_fence(struct drm_device *dev, uint32_t fence,
+		ktime_t *timeout , bool interruptible)
 {
 	struct msm_drm_private *priv = dev->dev_private;
 	int ret;
@@ -667,7 +667,12 @@ int msm_wait_fence_interruptable(struct drm_device *dev, uint32_t fence,
 			remaining_jiffies = timespec_to_jiffies(&ts);
 		}
 
-		ret = wait_event_interruptible_timeout(priv->fence_event,
+		if (interruptible)
+			ret = wait_event_interruptible_timeout(priv->fence_event,
+				fence_completed(dev, fence),
+				remaining_jiffies);
+		else
+			ret = wait_event_timeout(priv->fence_event,
 				fence_completed(dev, fence),
 				remaining_jiffies);
 
@@ -853,7 +858,7 @@ static int msm_ioctl_wait_fence(struct drm_device *dev, void *data,
 		return -EINVAL;
 	}
 
-	return msm_wait_fence_interruptable(dev, args->fence, &timeout);
+	return msm_wait_fence(dev, args->fence, &timeout, true);
 }
 
 static const struct drm_ioctl_desc msm_ioctls[] = {
diff --git a/drivers/gpu/drm/msm/msm_drv.h b/drivers/gpu/drm/msm/msm_drv.h
index e7c5ea125d45..4ff0ec9c994b 100644
--- a/drivers/gpu/drm/msm/msm_drv.h
+++ b/drivers/gpu/drm/msm/msm_drv.h
@@ -164,8 +164,8 @@ int msm_atomic_commit(struct drm_device *dev,
 
 int msm_register_mmu(struct drm_device *dev, struct msm_mmu *mmu);
 
-int msm_wait_fence_interruptable(struct drm_device *dev, uint32_t fence,
-		ktime_t *timeout);
+int msm_wait_fence(struct drm_device *dev, uint32_t fence,
+		ktime_t *timeout, bool interruptible);
 int msm_queue_fence_cb(struct drm_device *dev,
 		struct msm_fence_cb *cb, uint32_t fence);
 void msm_update_fence(struct drm_device *dev, uint32_t fence);
diff --git a/drivers/gpu/drm/msm/msm_gem.c b/drivers/gpu/drm/msm/msm_gem.c
index f211b80e3a1e..c76cc853b08a 100644
--- a/drivers/gpu/drm/msm/msm_gem.c
+++ b/drivers/gpu/drm/msm/msm_gem.c
@@ -460,7 +460,7 @@ int msm_gem_cpu_prep(struct drm_gem_object *obj, uint32_t op, ktime_t *timeout)
 		if (op & MSM_PREP_NOSYNC)
 			timeout = NULL;
 
-		ret = msm_wait_fence_interruptable(dev, fence, timeout);
+		ret = msm_wait_fence(dev, fence, timeout, true);
 	}
 
 	/* TODO cache maintenance */
diff --git a/drivers/gpu/drm/msm/msm_gem_prime.c b/drivers/gpu/drm/msm/msm_gem_prime.c
index dd7a7ab603e2..831461bc98a5 100644
--- a/drivers/gpu/drm/msm/msm_gem_prime.c
+++ b/drivers/gpu/drm/msm/msm_gem_prime.c
@@ -23,8 +23,12 @@
 struct sg_table *msm_gem_prime_get_sg_table(struct drm_gem_object *obj)
 {
 	struct msm_gem_object *msm_obj = to_msm_bo(obj);
-	BUG_ON(!msm_obj->sgt);  /* should have already pinned! */
-	return msm_obj->sgt;
+	int npages = obj->size >> PAGE_SHIFT;
+
+	if (WARN_ON(!msm_obj->pages))  /* should have already pinned! */
+		return NULL;
+
+	return drm_prime_pages_to_sg(msm_obj->pages, npages);
 }
 
 void *msm_gem_prime_vmap(struct drm_gem_object *obj)
diff --git a/drivers/gpu/drm/nouveau/nouveau_drm.c b/drivers/gpu/drm/nouveau/nouveau_drm.c
index 649024d4daf1..477cbb12809b 100644
--- a/drivers/gpu/drm/nouveau/nouveau_drm.c
+++ b/drivers/gpu/drm/nouveau/nouveau_drm.c
@@ -128,6 +128,7 @@ nouveau_cli_destroy(struct nouveau_cli *cli)
 	nvkm_vm_ref(NULL, &nvxx_client(&cli->base)->vm, NULL);
 	nvif_client_fini(&cli->base);
 	usif_client_fini(cli);
+	kfree(cli);
 }
 
 static void
@@ -865,8 +866,10 @@ nouveau_drm_preclose(struct drm_device *dev, struct drm_file *fpriv)
 
 	pm_runtime_get_sync(dev->dev);
 
+	mutex_lock(&cli->mutex);
 	if (cli->abi16)
 		nouveau_abi16_fini(cli->abi16);
+	mutex_unlock(&cli->mutex);
 
 	mutex_lock(&drm->client.mutex);
 	list_del(&cli->head);
diff --git a/drivers/gpu/drm/nouveau/nouveau_platform.c b/drivers/gpu/drm/nouveau/nouveau_platform.c
index 775277f1edb0..dcfbbfaf1739 100644
--- a/drivers/gpu/drm/nouveau/nouveau_platform.c
+++ b/drivers/gpu/drm/nouveau/nouveau_platform.c
@@ -92,6 +92,8 @@ static int nouveau_platform_power_down(struct nouveau_platform_gpu *gpu)
 	return 0;
 }
 
+#if IS_ENABLED(CONFIG_IOMMU_API)
+
 static void nouveau_platform_probe_iommu(struct device *dev,
 					 struct nouveau_platform_gpu *gpu)
 {
@@ -158,6 +160,20 @@ static void nouveau_platform_remove_iommu(struct device *dev,
 	}
 }
 
+#else
+
+static void nouveau_platform_probe_iommu(struct device *dev,
+					 struct nouveau_platform_gpu *gpu)
+{
+}
+
+static void nouveau_platform_remove_iommu(struct device *dev,
+					  struct nouveau_platform_gpu *gpu)
+{
+}
+
+#endif
+
 static int nouveau_platform_probe(struct platform_device *pdev)
 {
 	struct nouveau_platform_gpu *gpu;
diff --git a/drivers/gpu/drm/nouveau/nouveau_ttm.c b/drivers/gpu/drm/nouveau/nouveau_ttm.c
index 18f449715788..7464aef34674 100644
--- a/drivers/gpu/drm/nouveau/nouveau_ttm.c
+++ b/drivers/gpu/drm/nouveau/nouveau_ttm.c
@@ -175,15 +175,24 @@ nouveau_gart_manager_new(struct ttm_mem_type_manager *man,
 	node->page_shift = 12;
 
 	switch (drm->device.info.family) {
+	case NV_DEVICE_INFO_V0_TNT:
+	case NV_DEVICE_INFO_V0_CELSIUS:
+	case NV_DEVICE_INFO_V0_KELVIN:
+	case NV_DEVICE_INFO_V0_RANKINE:
+	case NV_DEVICE_INFO_V0_CURIE:
+		break;
 	case NV_DEVICE_INFO_V0_TESLA:
 		if (drm->device.info.chipset != 0x50)
 			node->memtype = (nvbo->tile_flags & 0x7f00) >> 8;
 		break;
 	case NV_DEVICE_INFO_V0_FERMI:
 	case NV_DEVICE_INFO_V0_KEPLER:
+	case NV_DEVICE_INFO_V0_MAXWELL:
 		node->memtype = (nvbo->tile_flags & 0xff00) >> 8;
 		break;
 	default:
+		NV_WARN(drm, "%s: unhandled family type %x\n", __func__,
+			drm->device.info.family);
 		break;
 	}
 
diff --git a/drivers/gpu/drm/nouveau/nv04_fbcon.c b/drivers/gpu/drm/nouveau/nv04_fbcon.c
index 4ef602c5469d..495c57644ced 100644
--- a/drivers/gpu/drm/nouveau/nv04_fbcon.c
+++ b/drivers/gpu/drm/nouveau/nv04_fbcon.c
@@ -203,7 +203,7 @@ nv04_fbcon_accel_init(struct fb_info *info)
 	if (ret)
 		return ret;
 
-	if (RING_SPACE(chan, 49)) {
+	if (RING_SPACE(chan, 49 + (device->info.chipset >= 0x11 ? 4 : 0))) {
 		nouveau_fbcon_gpu_lockup(info);
 		return 0;
 	}
diff --git a/drivers/gpu/drm/nouveau/nv50_display.c b/drivers/gpu/drm/nouveau/nv50_display.c
index 7da7958556a3..981342d142ff 100644
--- a/drivers/gpu/drm/nouveau/nv50_display.c
+++ b/drivers/gpu/drm/nouveau/nv50_display.c
@@ -979,7 +979,7 @@ nv50_crtc_cursor_show_hide(struct nouveau_crtc *nv_crtc, bool show, bool update)
 {
 	struct nv50_mast *mast = nv50_mast(nv_crtc->base.dev);
 
-	if (show && nv_crtc->cursor.nvbo)
+	if (show && nv_crtc->cursor.nvbo && nv_crtc->base.enabled)
 		nv50_crtc_cursor_show(nv_crtc);
 	else
 		nv50_crtc_cursor_hide(nv_crtc);
diff --git a/drivers/gpu/drm/nouveau/nv50_fbcon.c b/drivers/gpu/drm/nouveau/nv50_fbcon.c
index 394c89abcc97..901130b06072 100644
--- a/drivers/gpu/drm/nouveau/nv50_fbcon.c
+++ b/drivers/gpu/drm/nouveau/nv50_fbcon.c
@@ -188,7 +188,7 @@ nv50_fbcon_accel_init(struct fb_info *info)
 	if (ret)
 		return ret;
 
-	ret = RING_SPACE(chan, 59);
+	ret = RING_SPACE(chan, 58);
 	if (ret) {
 		nouveau_fbcon_gpu_lockup(info);
 		return ret;
@@ -252,6 +252,7 @@ nv50_fbcon_accel_init(struct fb_info *info)
 	OUT_RING(chan, info->var.yres_virtual);
 	OUT_RING(chan, upper_32_bits(fb->vma.offset));
 	OUT_RING(chan, lower_32_bits(fb->vma.offset));
+	FIRE_RING(chan);
 
 	return 0;
 }
diff --git a/drivers/gpu/drm/nouveau/nvc0_fbcon.c b/drivers/gpu/drm/nouveau/nvc0_fbcon.c
index 61246677e8dc..fcd2e5f27bb9 100644
--- a/drivers/gpu/drm/nouveau/nvc0_fbcon.c
+++ b/drivers/gpu/drm/nouveau/nvc0_fbcon.c
@@ -188,7 +188,7 @@ nvc0_fbcon_accel_init(struct fb_info *info)
 		return -EINVAL;
 	}
 
-	ret = RING_SPACE(chan, 60);
+	ret = RING_SPACE(chan, 58);
 	if (ret) {
 		WARN_ON(1);
 		nouveau_fbcon_gpu_lockup(info);
diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/disp/gf110.c b/drivers/gpu/drm/nouveau/nvkm/engine/disp/gf110.c
index 9ef6728c528d..7f2f05f78cc8 100644
--- a/drivers/gpu/drm/nouveau/nvkm/engine/disp/gf110.c
+++ b/drivers/gpu/drm/nouveau/nvkm/engine/disp/gf110.c
@@ -809,7 +809,7 @@ exec_lookup(struct nv50_disp_priv *priv, int head, int or, u32 ctrl,
 		case 0x00000900: type = DCB_OUTPUT_DP; mask = 2; break;
 		default:
 			nv_error(priv, "unknown SOR mc 0x%08x\n", ctrl);
-			return 0x0000;
+			return NULL;
 		}
 	}
 
diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/fifo/gk104.c b/drivers/gpu/drm/nouveau/nvkm/engine/fifo/gk104.c
index e10f9644140f..52c22b026005 100644
--- a/drivers/gpu/drm/nouveau/nvkm/engine/fifo/gk104.c
+++ b/drivers/gpu/drm/nouveau/nvkm/engine/fifo/gk104.c
@@ -166,14 +166,30 @@ gk104_fifo_context_attach(struct nvkm_object *parent,
 }
 
 static int
+gk104_fifo_chan_kick(struct gk104_fifo_chan *chan)
+{
+	struct nvkm_object *obj = (void *)chan;
+	struct gk104_fifo_priv *priv = (void *)obj->engine;
+
+	nv_wr32(priv, 0x002634, chan->base.chid);
+	if (!nv_wait(priv, 0x002634, 0x100000, 0x000000)) {
+		nv_error(priv, "channel %d [%s] kick timeout\n",
+			 chan->base.chid, nvkm_client_name(chan));
+		return -EBUSY;
+	}
+
+	return 0;
+}
+
+static int
 gk104_fifo_context_detach(struct nvkm_object *parent, bool suspend,
 			  struct nvkm_object *object)
 {
 	struct nvkm_bar *bar = nvkm_bar(parent);
-	struct gk104_fifo_priv *priv = (void *)parent->engine;
 	struct gk104_fifo_base *base = (void *)parent->parent;
 	struct gk104_fifo_chan *chan = (void *)parent;
 	u32 addr;
+	int ret;
 
 	switch (nv_engidx(object->engine)) {
 	case NVDEV_ENGINE_SW    : return 0;
@@ -188,13 +204,9 @@ gk104_fifo_context_detach(struct nvkm_object *parent, bool suspend,
 		return -EINVAL;
 	}
 
-	nv_wr32(priv, 0x002634, chan->base.chid);
-	if (!nv_wait(priv, 0x002634, 0xffffffff, chan->base.chid)) {
-		nv_error(priv, "channel %d [%s] kick timeout\n",
-			 chan->base.chid, nvkm_client_name(chan));
-		if (suspend)
-			return -EBUSY;
-	}
+	ret = gk104_fifo_chan_kick(chan);
+	if (ret && suspend)
+		return ret;
 
 	if (addr) {
 		nv_wo32(base, addr + 0x00, 0x00000000);
@@ -319,6 +331,7 @@ gk104_fifo_chan_fini(struct nvkm_object *object, bool suspend)
 		gk104_fifo_runlist_update(priv, chan->engine);
 	}
 
+	gk104_fifo_chan_kick(chan);
 	nv_wr32(priv, 0x800000 + (chid * 8), 0x00000000);
 	return nvkm_fifo_channel_fini(&chan->base, suspend);
 }
diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/gr/gf100.c b/drivers/gpu/drm/nouveau/nvkm/engine/gr/gf100.c
index 5606c25e5d02..ca11ddb6ed46 100644
--- a/drivers/gpu/drm/nouveau/nvkm/engine/gr/gf100.c
+++ b/drivers/gpu/drm/nouveau/nvkm/engine/gr/gf100.c
@@ -663,6 +663,37 @@ gf100_gr_zbc_init(struct gf100_gr_priv *priv)
 		gf100_gr_zbc_clear_depth(priv, index);
 }
 
+/**
+ * Wait until GR goes idle. GR is considered idle if it is disabled by the
+ * MC (0x200) register, or GR is not busy and a context switch is not in
+ * progress.
+ */
+int
+gf100_gr_wait_idle(struct gf100_gr_priv *priv)
+{
+	unsigned long end_jiffies = jiffies + msecs_to_jiffies(2000);
+	bool gr_enabled, ctxsw_active, gr_busy;
+
+	do {
+		/*
+		 * required to make sure FIFO_ENGINE_STATUS (0x2640) is
+		 * up-to-date
+		 */
+		nv_rd32(priv, 0x400700);
+
+		gr_enabled = nv_rd32(priv, 0x200) & 0x1000;
+		ctxsw_active = nv_rd32(priv, 0x2640) & 0x8000;
+		gr_busy = nv_rd32(priv, 0x40060c) & 0x1;
+
+		if (!gr_enabled || (!gr_busy && !ctxsw_active))
+			return 0;
+	} while (time_before(jiffies, end_jiffies));
+
+	nv_error(priv, "wait for idle timeout (en: %d, ctxsw: %d, busy: %d)\n",
+		 gr_enabled, ctxsw_active, gr_busy);
+	return -EAGAIN;
+}
+
 void
 gf100_gr_mmio(struct gf100_gr_priv *priv, const struct gf100_gr_pack *p)
 {
@@ -699,7 +730,13 @@ gf100_gr_icmd(struct gf100_gr_priv *priv, const struct gf100_gr_pack *p)
 
 		while (addr < next) {
 			nv_wr32(priv, 0x400200, addr);
-			nv_wait(priv, 0x400700, 0x00000002, 0x00000000);
+			/**
+			 * Wait for GR to go idle after submitting a
+			 * GO_IDLE bundle
+			 */
+			if ((addr & 0xffff) == 0xe100)
+				gf100_gr_wait_idle(priv);
+			nv_wait(priv, 0x400700, 0x00000004, 0x00000000);
 			addr += init->pitch;
 		}
 	}
diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/gr/gf100.h b/drivers/gpu/drm/nouveau/nvkm/engine/gr/gf100.h
index 8af1a89eda84..c9533fdac4fc 100644
--- a/drivers/gpu/drm/nouveau/nvkm/engine/gr/gf100.h
+++ b/drivers/gpu/drm/nouveau/nvkm/engine/gr/gf100.h
@@ -181,6 +181,7 @@ struct gf100_gr_oclass {
 	int ppc_nr;
 };
 
+int  gf100_gr_wait_idle(struct gf100_gr_priv *);
 void gf100_gr_mmio(struct gf100_gr_priv *, const struct gf100_gr_pack *);
 void gf100_gr_icmd(struct gf100_gr_priv *, const struct gf100_gr_pack *);
 void gf100_gr_mthd(struct gf100_gr_priv *, const struct gf100_gr_pack *);
diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/pm/base.c b/drivers/gpu/drm/nouveau/nvkm/engine/pm/base.c
index 2006c445938d..4cf36a3aa814 100644
--- a/drivers/gpu/drm/nouveau/nvkm/engine/pm/base.c
+++ b/drivers/gpu/drm/nouveau/nvkm/engine/pm/base.c
@@ -332,9 +332,12 @@ static void
 nvkm_perfctx_dtor(struct nvkm_object *object)
 {
 	struct nvkm_pm *ppm = (void *)object->engine;
+	struct nvkm_perfctx *ctx = (void *)object;
+
 	mutex_lock(&nv_subdev(ppm)->mutex);
-	nvkm_engctx_destroy(&ppm->context->base);
-	ppm->context = NULL;
+	nvkm_engctx_destroy(&ctx->base);
+	if (ppm->context == ctx)
+		ppm->context = NULL;
 	mutex_unlock(&nv_subdev(ppm)->mutex);
 }
 
@@ -355,12 +358,11 @@ nvkm_perfctx_ctor(struct nvkm_object *parent, struct nvkm_object *engine,
 	mutex_lock(&nv_subdev(ppm)->mutex);
 	if (ppm->context == NULL)
 		ppm->context = ctx;
-	mutex_unlock(&nv_subdev(ppm)->mutex);
-
 	if (ctx != ppm->context)
-		return -EBUSY;
+		ret = -EBUSY;
+	mutex_unlock(&nv_subdev(ppm)->mutex);
 
-	return 0;
+	return ret;
 }
 
 struct nvkm_oclass
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/bios/init.c b/drivers/gpu/drm/nouveau/nvkm/subdev/bios/init.c
index f67cdae1e90a..f4611e3f0971 100644
--- a/drivers/gpu/drm/nouveau/nvkm/subdev/bios/init.c
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/bios/init.c
@@ -1285,6 +1285,44 @@ init_zm_reg_sequence(struct nvbios_init *init)
 }
 
 /**
+ * INIT_PLL_INDIRECT - opcode 0x59
+ *
+ */
+static void
+init_pll_indirect(struct nvbios_init *init)
+{
+	struct nvkm_bios *bios = init->bios;
+	u32  reg = nv_ro32(bios, init->offset + 1);
+	u16 addr = nv_ro16(bios, init->offset + 5);
+	u32 freq = (u32)nv_ro16(bios, addr) * 1000;
+
+	trace("PLL_INDIRECT\tR[0x%06x] =PLL= VBIOS[%04x] = %dkHz\n",
+	      reg, addr, freq);
+	init->offset += 7;
+
+	init_prog_pll(init, reg, freq);
+}
+
+/**
+ * INIT_ZM_REG_INDIRECT - opcode 0x5a
+ *
+ */
+static void
+init_zm_reg_indirect(struct nvbios_init *init)
+{
+	struct nvkm_bios *bios = init->bios;
+	u32  reg = nv_ro32(bios, init->offset + 1);
+	u16 addr = nv_ro16(bios, init->offset + 5);
+	u32 data = nv_ro32(bios, addr);
+
+	trace("ZM_REG_INDIRECT\tR[0x%06x] = VBIOS[0x%04x] = 0x%08x\n",
+	      reg, addr, data);
+	init->offset += 7;
+
+	init_wr32(init, addr, data);
+}
+
+/**
  * INIT_SUB_DIRECT - opcode 0x5b
  *
  */
@@ -2145,6 +2183,8 @@ static struct nvbios_init_opcode {
 	[0x56] = { init_condition_time },
 	[0x57] = { init_ltime },
 	[0x58] = { init_zm_reg_sequence },
+	[0x59] = { init_pll_indirect },
+	[0x5a] = { init_zm_reg_indirect },
 	[0x5b] = { init_sub_direct },
 	[0x5c] = { init_jump },
 	[0x5e] = { init_i2c_if },
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/clk/gt215.c b/drivers/gpu/drm/nouveau/nvkm/subdev/clk/gt215.c
index 822d32a28d6e..065e9f5c8db9 100644
--- a/drivers/gpu/drm/nouveau/nvkm/subdev/clk/gt215.c
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/clk/gt215.c
@@ -180,7 +180,8 @@ gt215_clk_info(struct nvkm_clk *clock, int clk, u32 khz,
 	       struct gt215_clk_info *info)
 {
 	struct gt215_clk_priv *priv = (void *)clock;
-	u32 oclk, sclk, sdiv, diff;
+	u32 oclk, sclk, sdiv;
+	s32 diff;
 
 	info->clk = 0;
 
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/ibus/gk20a.c b/drivers/gpu/drm/nouveau/nvkm/subdev/ibus/gk20a.c
index c0fdb89e74ac..24dcdfb58a8d 100644
--- a/drivers/gpu/drm/nouveau/nvkm/subdev/ibus/gk20a.c
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/ibus/gk20a.c
@@ -38,6 +38,14 @@ gk20a_ibus_init_priv_ring(struct gk20a_ibus_priv *priv)
 	nv_wr32(priv, 0x12004c, 0x4);
 	nv_wr32(priv, 0x122204, 0x2);
 	nv_rd32(priv, 0x122204);
+
+	/*
+	 * Bug: increase clock timeout to avoid operation failure at high
+	 * gpcclk rate.
+	 */
+	nv_wr32(priv, 0x122354, 0x800);
+	nv_wr32(priv, 0x128328, 0x800);
+	nv_wr32(priv, 0x124320, 0x800);
 }
 
 static void
diff --git a/drivers/gpu/drm/nouveau/nvkm/subdev/instmem/nv04.c b/drivers/gpu/drm/nouveau/nvkm/subdev/instmem/nv04.c
index 80614f1b2074..282143f49d72 100644
--- a/drivers/gpu/drm/nouveau/nvkm/subdev/instmem/nv04.c
+++ b/drivers/gpu/drm/nouveau/nvkm/subdev/instmem/nv04.c
@@ -50,7 +50,12 @@ nv04_instobj_dtor(struct nvkm_object *object)
 {
 	struct nv04_instmem_priv *priv = (void *)nvkm_instmem(object);
 	struct nv04_instobj_priv *node = (void *)object;
+	struct nvkm_subdev *subdev = (void *)priv;
+
+	mutex_lock(&subdev->mutex);
 	nvkm_mm_free(&priv->heap, &node->mem);
+	mutex_unlock(&subdev->mutex);
+
 	nvkm_instobj_destroy(&node->base);
 }
 
@@ -62,6 +67,7 @@ nv04_instobj_ctor(struct nvkm_object *parent, struct nvkm_object *engine,
 	struct nv04_instmem_priv *priv = (void *)nvkm_instmem(parent);
 	struct nv04_instobj_priv *node;
 	struct nvkm_instobj_args *args = data;
+	struct nvkm_subdev *subdev = (void *)priv;
 	int ret;
 
 	if (!args->align)
@@ -72,8 +78,10 @@ nv04_instobj_ctor(struct nvkm_object *parent, struct nvkm_object *engine,
 	if (ret)
 		return ret;
 
+	mutex_lock(&subdev->mutex);
 	ret = nvkm_mm_head(&priv->heap, 0, 1, args->size, args->size,
 			   args->align, &node->mem);
+	mutex_unlock(&subdev->mutex);
 	if (ret)
 		return ret;
 
diff --git a/drivers/gpu/drm/radeon/atombios_encoders.c b/drivers/gpu/drm/radeon/atombios_encoders.c
index dd39f434b4a7..c3872598b85a 100644
--- a/drivers/gpu/drm/radeon/atombios_encoders.c
+++ b/drivers/gpu/drm/radeon/atombios_encoders.c
@@ -2299,8 +2299,7 @@ radeon_atom_encoder_mode_set(struct drm_encoder *encoder,
 	encoder_mode = atombios_get_encoder_mode(encoder);
 	if (connector && (radeon_audio != 0) &&
 	    ((encoder_mode == ATOM_ENCODER_MODE_HDMI) ||
-	     (ENCODER_MODE_IS_DP(encoder_mode) &&
-	      drm_detect_monitor_audio(radeon_connector_edid(connector)))))
+	     ENCODER_MODE_IS_DP(encoder_mode)))
 		radeon_audio_mode_set(encoder, adjusted_mode);
 }
 
diff --git a/drivers/gpu/drm/radeon/dce6_afmt.c b/drivers/gpu/drm/radeon/dce6_afmt.c
index 68fd9fc677e3..44480c1b9738 100644
--- a/drivers/gpu/drm/radeon/dce6_afmt.c
+++ b/drivers/gpu/drm/radeon/dce6_afmt.c
@@ -93,30 +93,26 @@ void dce6_afmt_select_pin(struct drm_encoder *encoder)
 	struct radeon_device *rdev = encoder->dev->dev_private;
 	struct radeon_encoder *radeon_encoder = to_radeon_encoder(encoder);
 	struct radeon_encoder_atom_dig *dig = radeon_encoder->enc_priv;
-	u32 offset;
 
-	if (!dig || !dig->afmt || !dig->afmt->pin)
+	if (!dig || !dig->afmt || !dig->pin)
 		return;
 
-	offset = dig->afmt->offset;
-
-	WREG32(AFMT_AUDIO_SRC_CONTROL + offset,
-	       AFMT_AUDIO_SRC_SELECT(dig->afmt->pin->id));
+	WREG32(AFMT_AUDIO_SRC_CONTROL +  dig->afmt->offset,
+	       AFMT_AUDIO_SRC_SELECT(dig->pin->id));
 }
 
 void dce6_afmt_write_latency_fields(struct drm_encoder *encoder,
-		struct drm_connector *connector, struct drm_display_mode *mode)
+				    struct drm_connector *connector,
+				    struct drm_display_mode *mode)
 {
 	struct radeon_device *rdev = encoder->dev->dev_private;
 	struct radeon_encoder *radeon_encoder = to_radeon_encoder(encoder);
 	struct radeon_encoder_atom_dig *dig = radeon_encoder->enc_priv;
-	u32 tmp = 0, offset;
+	u32 tmp = 0;
 
-	if (!dig || !dig->afmt || !dig->afmt->pin)
+	if (!dig || !dig->afmt || !dig->pin)
 		return;
 
-	offset = dig->afmt->pin->offset;
-
 	if (mode->flags & DRM_MODE_FLAG_INTERLACE) {
 		if (connector->latency_present[1])
 			tmp = VIDEO_LIPSYNC(connector->video_latency[1]) |
@@ -130,24 +126,24 @@ void dce6_afmt_write_latency_fields(struct drm_encoder *encoder,
 		else
 			tmp = VIDEO_LIPSYNC(0) | AUDIO_LIPSYNC(0);
 	}
-	WREG32_ENDPOINT(offset, AZ_F0_CODEC_PIN_CONTROL_RESPONSE_LIPSYNC, tmp);
+	WREG32_ENDPOINT(dig->pin->offset,
+			AZ_F0_CODEC_PIN_CONTROL_RESPONSE_LIPSYNC, tmp);
 }
 
 void dce6_afmt_hdmi_write_speaker_allocation(struct drm_encoder *encoder,
-	u8 *sadb, int sad_count)
+					     u8 *sadb, int sad_count)
 {
 	struct radeon_device *rdev = encoder->dev->dev_private;
 	struct radeon_encoder *radeon_encoder = to_radeon_encoder(encoder);
 	struct radeon_encoder_atom_dig *dig = radeon_encoder->enc_priv;
-	u32 offset, tmp;
+	u32 tmp;
 
-	if (!dig || !dig->afmt || !dig->afmt->pin)
+	if (!dig || !dig->afmt || !dig->pin)
 		return;
 
-	offset = dig->afmt->pin->offset;
-
 	/* program the speaker allocation */
-	tmp = RREG32_ENDPOINT(offset, AZ_F0_CODEC_PIN_CONTROL_CHANNEL_SPEAKER);
+	tmp = RREG32_ENDPOINT(dig->pin->offset,
+			      AZ_F0_CODEC_PIN_CONTROL_CHANNEL_SPEAKER);
 	tmp &= ~(DP_CONNECTION | SPEAKER_ALLOCATION_MASK);
 	/* set HDMI mode */
 	tmp |= HDMI_CONNECTION;
@@ -155,24 +151,24 @@ void dce6_afmt_hdmi_write_speaker_allocation(struct drm_encoder *encoder,
 		tmp |= SPEAKER_ALLOCATION(sadb[0]);
 	else
 		tmp |= SPEAKER_ALLOCATION(5); /* stereo */
-	WREG32_ENDPOINT(offset, AZ_F0_CODEC_PIN_CONTROL_CHANNEL_SPEAKER, tmp);
+	WREG32_ENDPOINT(dig->pin->offset,
+			AZ_F0_CODEC_PIN_CONTROL_CHANNEL_SPEAKER, tmp);
 }
 
 void dce6_afmt_dp_write_speaker_allocation(struct drm_encoder *encoder,
-	u8 *sadb, int sad_count)
+					   u8 *sadb, int sad_count)
 {
 	struct radeon_device *rdev = encoder->dev->dev_private;
 	struct radeon_encoder *radeon_encoder = to_radeon_encoder(encoder);
 	struct radeon_encoder_atom_dig *dig = radeon_encoder->enc_priv;
-	u32 offset, tmp;
+	u32 tmp;
 
-	if (!dig || !dig->afmt || !dig->afmt->pin)
+	if (!dig || !dig->afmt || !dig->pin)
 		return;
 
-	offset = dig->afmt->pin->offset;
-
 	/* program the speaker allocation */
-	tmp = RREG32_ENDPOINT(offset, AZ_F0_CODEC_PIN_CONTROL_CHANNEL_SPEAKER);
+	tmp = RREG32_ENDPOINT(dig->pin->offset,
+			      AZ_F0_CODEC_PIN_CONTROL_CHANNEL_SPEAKER);
 	tmp &= ~(HDMI_CONNECTION | SPEAKER_ALLOCATION_MASK);
 	/* set DP mode */
 	tmp |= DP_CONNECTION;
@@ -180,13 +176,13 @@ void dce6_afmt_dp_write_speaker_allocation(struct drm_encoder *encoder,
 		tmp |= SPEAKER_ALLOCATION(sadb[0]);
 	else
 		tmp |= SPEAKER_ALLOCATION(5); /* stereo */
-	WREG32_ENDPOINT(offset, AZ_F0_CODEC_PIN_CONTROL_CHANNEL_SPEAKER, tmp);
+	WREG32_ENDPOINT(dig->pin->offset,
+			AZ_F0_CODEC_PIN_CONTROL_CHANNEL_SPEAKER, tmp);
 }
 
 void dce6_afmt_write_sad_regs(struct drm_encoder *encoder,
-	struct cea_sad *sads, int sad_count)
+			      struct cea_sad *sads, int sad_count)
 {
-	u32 offset;
 	int i;
 	struct radeon_encoder *radeon_encoder = to_radeon_encoder(encoder);
 	struct radeon_encoder_atom_dig *dig = radeon_encoder->enc_priv;
@@ -206,11 +202,9 @@ void dce6_afmt_write_sad_regs(struct drm_encoder *encoder,
 		{ AZ_F0_CODEC_PIN_CONTROL_AUDIO_DESCRIPTOR13, HDMI_AUDIO_CODING_TYPE_WMA_PRO },
 	};
 
-	if (!dig || !dig->afmt || !dig->afmt->pin)
+	if (!dig || !dig->afmt || !dig->pin)
 		return;
 
-	offset = dig->afmt->pin->offset;
-
 	for (i = 0; i < ARRAY_SIZE(eld_reg_to_type); i++) {
 		u32 value = 0;
 		u8 stereo_freqs = 0;
@@ -237,7 +231,7 @@ void dce6_afmt_write_sad_regs(struct drm_encoder *encoder,
 
 		value |= SUPPORTED_FREQUENCIES_STEREO(stereo_freqs);
 
-		WREG32_ENDPOINT(offset, eld_reg_to_type[i][0], value);
+		WREG32_ENDPOINT(dig->pin->offset, eld_reg_to_type[i][0], value);
 	}
 }
 
@@ -253,7 +247,7 @@ void dce6_audio_enable(struct radeon_device *rdev,
 }
 
 void dce6_hdmi_audio_set_dto(struct radeon_device *rdev,
-	struct radeon_crtc *crtc, unsigned int clock)
+			     struct radeon_crtc *crtc, unsigned int clock)
 {
 	/* Two dtos; generally use dto0 for HDMI */
 	u32 value = 0;
@@ -272,7 +266,7 @@ void dce6_hdmi_audio_set_dto(struct radeon_device *rdev,
 }
 
 void dce6_dp_audio_set_dto(struct radeon_device *rdev,
-	struct radeon_crtc *crtc, unsigned int clock)
+			   struct radeon_crtc *crtc, unsigned int clock)
 {
 	/* Two dtos; generally use dto1 for DP */
 	u32 value = 0;
diff --git a/drivers/gpu/drm/radeon/radeon_audio.c b/drivers/gpu/drm/radeon/radeon_audio.c
index fa719c53449b..fbc8d88d6e5d 100644
--- a/drivers/gpu/drm/radeon/radeon_audio.c
+++ b/drivers/gpu/drm/radeon/radeon_audio.c
@@ -245,6 +245,28 @@ static struct radeon_audio_funcs dce6_dp_funcs = {
 static void radeon_audio_enable(struct radeon_device *rdev,
 				struct r600_audio_pin *pin, u8 enable_mask)
 {
+	struct drm_encoder *encoder;
+	struct radeon_encoder *radeon_encoder;
+	struct radeon_encoder_atom_dig *dig;
+	int pin_count = 0;
+
+	if (!pin)
+		return;
+
+	if (rdev->mode_info.mode_config_initialized) {
+		list_for_each_entry(encoder, &rdev->ddev->mode_config.encoder_list, head) {
+			if (radeon_encoder_is_digital(encoder)) {
+				radeon_encoder = to_radeon_encoder(encoder);
+				dig = radeon_encoder->enc_priv;
+				if (dig->pin == pin)
+					pin_count++;
+			}
+		}
+
+		if ((pin_count > 1) && (enable_mask == 0))
+			return;
+	}
+
 	if (rdev->audio.funcs->enable)
 		rdev->audio.funcs->enable(rdev, pin, enable_mask);
 }
@@ -336,24 +358,13 @@ void radeon_audio_endpoint_wreg(struct radeon_device *rdev, u32 offset,
 
 static void radeon_audio_write_sad_regs(struct drm_encoder *encoder)
 {
-	struct radeon_encoder *radeon_encoder;
-	struct drm_connector *connector;
-	struct radeon_connector *radeon_connector = NULL;
+	struct drm_connector *connector = radeon_get_connector_for_encoder(encoder);
+	struct radeon_encoder *radeon_encoder = to_radeon_encoder(encoder);
 	struct cea_sad *sads;
 	int sad_count;
 
-	list_for_each_entry(connector,
-		&encoder->dev->mode_config.connector_list, head) {
-		if (connector->encoder == encoder) {
-			radeon_connector = to_radeon_connector(connector);
-			break;
-		}
-	}
-
-	if (!radeon_connector) {
-		DRM_ERROR("Couldn't find encoder's connector\n");
+	if (!connector)
 		return;
-	}
 
 	sad_count = drm_edid_to_sad(radeon_connector_edid(connector), &sads);
 	if (sad_count <= 0) {
@@ -362,8 +373,6 @@ static void radeon_audio_write_sad_regs(struct drm_encoder *encoder)
 	}
 	BUG_ON(!sads);
 
-	radeon_encoder = to_radeon_encoder(encoder);
-
 	if (radeon_encoder->audio && radeon_encoder->audio->write_sad_regs)
 		radeon_encoder->audio->write_sad_regs(encoder, sads, sad_count);
 
@@ -372,27 +381,16 @@ static void radeon_audio_write_sad_regs(struct drm_encoder *encoder)
 
 static void radeon_audio_write_speaker_allocation(struct drm_encoder *encoder)
 {
+	struct drm_connector *connector = radeon_get_connector_for_encoder(encoder);
 	struct radeon_encoder *radeon_encoder = to_radeon_encoder(encoder);
-	struct drm_connector *connector;
-	struct radeon_connector *radeon_connector = NULL;
 	u8 *sadb = NULL;
 	int sad_count;
 
-	list_for_each_entry(connector,
-			    &encoder->dev->mode_config.connector_list, head) {
-		if (connector->encoder == encoder) {
-			radeon_connector = to_radeon_connector(connector);
-			break;
-		}
-	}
-
-	if (!radeon_connector) {
-		DRM_ERROR("Couldn't find encoder's connector\n");
+	if (!connector)
 		return;
-	}
 
-	sad_count = drm_edid_to_speaker_allocation(
-		radeon_connector_edid(connector), &sadb);
+	sad_count = drm_edid_to_speaker_allocation(radeon_connector_edid(connector),
+						   &sadb);
 	if (sad_count < 0) {
 		DRM_DEBUG("Couldn't read Speaker Allocation Data Block: %d\n",
 			  sad_count);
@@ -406,26 +404,13 @@ static void radeon_audio_write_speaker_allocation(struct drm_encoder *encoder)
 }
 
 static void radeon_audio_write_latency_fields(struct drm_encoder *encoder,
-	struct drm_display_mode *mode)
+					      struct drm_display_mode *mode)
 {
-	struct radeon_encoder *radeon_encoder;
-	struct drm_connector *connector;
-	struct radeon_connector *radeon_connector = 0;
-
-	list_for_each_entry(connector,
-		&encoder->dev->mode_config.connector_list, head) {
-		if (connector->encoder == encoder) {
-			radeon_connector = to_radeon_connector(connector);
-			break;
-		}
-	}
+	struct drm_connector *connector = radeon_get_connector_for_encoder(encoder);
+	struct radeon_encoder *radeon_encoder = to_radeon_encoder(encoder);
 
-	if (!radeon_connector) {
-		DRM_ERROR("Couldn't find encoder's connector\n");
+	if (!connector)
 		return;
-	}
-
-	radeon_encoder = to_radeon_encoder(encoder);
 
 	if (radeon_encoder->audio && radeon_encoder->audio->write_latency_fields)
 		radeon_encoder->audio->write_latency_fields(encoder, connector, mode);
@@ -451,29 +436,23 @@ static void radeon_audio_select_pin(struct drm_encoder *encoder)
 }
 
 void radeon_audio_detect(struct drm_connector *connector,
+			 struct drm_encoder *encoder,
 			 enum drm_connector_status status)
 {
-	struct radeon_device *rdev;
-	struct radeon_encoder *radeon_encoder;
+	struct drm_device *dev = connector->dev;
+	struct radeon_device *rdev = dev->dev_private;
+	struct radeon_encoder *radeon_encoder = to_radeon_encoder(encoder);
 	struct radeon_encoder_atom_dig *dig;
 
-	if (!connector || !connector->encoder)
+	if (!radeon_audio_chipset_supported(rdev))
 		return;
 
-	rdev = connector->encoder->dev->dev_private;
-
-	if (!radeon_audio_chipset_supported(rdev))
+	if (!radeon_encoder_is_digital(encoder))
 		return;
 
-	radeon_encoder = to_radeon_encoder(connector->encoder);
 	dig = radeon_encoder->enc_priv;
 
 	if (status == connector_status_connected) {
-		if (!drm_detect_monitor_audio(radeon_connector_edid(connector))) {
-			radeon_encoder->audio = NULL;
-			return;
-		}
-
 		if (connector->connector_type == DRM_MODE_CONNECTOR_DisplayPort) {
 			struct radeon_connector *radeon_connector = to_radeon_connector(connector);
 
@@ -486,11 +465,17 @@ void radeon_audio_detect(struct drm_connector *connector,
 			radeon_encoder->audio = rdev->audio.hdmi_funcs;
 		}
 
-		dig->afmt->pin = radeon_audio_get_pin(connector->encoder);
-		radeon_audio_enable(rdev, dig->afmt->pin, 0xf);
+		if (drm_detect_monitor_audio(radeon_connector_edid(connector))) {
+			if (!dig->pin)
+				dig->pin = radeon_audio_get_pin(encoder);
+			radeon_audio_enable(rdev, dig->pin, 0xf);
+		} else {
+			radeon_audio_enable(rdev, dig->pin, 0);
+			dig->pin = NULL;
+		}
 	} else {
-		radeon_audio_enable(rdev, dig->afmt->pin, 0);
-		dig->afmt->pin = NULL;
+		radeon_audio_enable(rdev, dig->pin, 0);
+		dig->pin = NULL;
 	}
 }
 
@@ -518,29 +503,18 @@ static void radeon_audio_set_dto(struct drm_encoder *encoder, unsigned int clock
 }
 
 static int radeon_audio_set_avi_packet(struct drm_encoder *encoder,
-	struct drm_display_mode *mode)
+				       struct drm_display_mode *mode)
 {
 	struct radeon_device *rdev = encoder->dev->dev_private;
 	struct radeon_encoder *radeon_encoder = to_radeon_encoder(encoder);
 	struct radeon_encoder_atom_dig *dig = radeon_encoder->enc_priv;
-	struct drm_connector *connector;
-	struct radeon_connector *radeon_connector = NULL;
+	struct drm_connector *connector = radeon_get_connector_for_encoder(encoder);
 	u8 buffer[HDMI_INFOFRAME_HEADER_SIZE + HDMI_AVI_INFOFRAME_SIZE];
 	struct hdmi_avi_infoframe frame;
 	int err;
 
-	list_for_each_entry(connector,
-		&encoder->dev->mode_config.connector_list, head) {
-		if (connector->encoder == encoder) {
-			radeon_connector = to_radeon_connector(connector);
-			break;
-		}
-	}
-
-	if (!radeon_connector) {
-		DRM_ERROR("Couldn't find encoder's connector\n");
-		return -ENOENT;
-	}
+	if (!connector)
+		return -EINVAL;
 
 	err = drm_hdmi_avi_infoframe_from_display_mode(&frame, mode);
 	if (err < 0) {
@@ -563,8 +537,8 @@ static int radeon_audio_set_avi_packet(struct drm_encoder *encoder,
 		return err;
 	}
 
-	if (dig && dig->afmt &&
-		radeon_encoder->audio && radeon_encoder->audio->set_avi_packet)
+	if (dig && dig->afmt && radeon_encoder->audio &&
+	    radeon_encoder->audio->set_avi_packet)
 		radeon_encoder->audio->set_avi_packet(rdev, dig->afmt->offset,
 			buffer, sizeof(buffer));
 
@@ -722,30 +696,41 @@ static void radeon_audio_hdmi_mode_set(struct drm_encoder *encoder,
 {
 	struct radeon_encoder *radeon_encoder = to_radeon_encoder(encoder);
 	struct radeon_encoder_atom_dig *dig = radeon_encoder->enc_priv;
+	struct drm_connector *connector = radeon_get_connector_for_encoder(encoder);
 
 	if (!dig || !dig->afmt)
 		return;
 
-	radeon_audio_set_mute(encoder, true);
+	if (!connector)
+		return;
 
-	radeon_audio_write_speaker_allocation(encoder);
-	radeon_audio_write_sad_regs(encoder);
-	radeon_audio_write_latency_fields(encoder, mode);
-	radeon_audio_set_dto(encoder, mode->clock);
-	radeon_audio_set_vbi_packet(encoder);
-	radeon_hdmi_set_color_depth(encoder);
-	radeon_audio_update_acr(encoder, mode->clock);
-	radeon_audio_set_audio_packet(encoder);
-	radeon_audio_select_pin(encoder);
+	if (drm_detect_monitor_audio(radeon_connector_edid(connector))) {
+		radeon_audio_set_mute(encoder, true);
 
-	if (radeon_audio_set_avi_packet(encoder, mode) < 0)
-		return;
+		radeon_audio_write_speaker_allocation(encoder);
+		radeon_audio_write_sad_regs(encoder);
+		radeon_audio_write_latency_fields(encoder, mode);
+		radeon_audio_set_dto(encoder, mode->clock);
+		radeon_audio_set_vbi_packet(encoder);
+		radeon_hdmi_set_color_depth(encoder);
+		radeon_audio_update_acr(encoder, mode->clock);
+		radeon_audio_set_audio_packet(encoder);
+		radeon_audio_select_pin(encoder);
+
+		if (radeon_audio_set_avi_packet(encoder, mode) < 0)
+			return;
 
-	radeon_audio_set_mute(encoder, false);
+		radeon_audio_set_mute(encoder, false);
+	} else {
+		radeon_hdmi_set_color_depth(encoder);
+
+		if (radeon_audio_set_avi_packet(encoder, mode) < 0)
+			return;
+	}
 }
 
 static void radeon_audio_dp_mode_set(struct drm_encoder *encoder,
-	struct drm_display_mode *mode)
+				     struct drm_display_mode *mode)
 {
 	struct drm_device *dev = encoder->dev;
 	struct radeon_device *rdev = dev->dev_private;
@@ -759,22 +744,27 @@ static void radeon_audio_dp_mode_set(struct drm_encoder *encoder,
 	if (!dig || !dig->afmt)
 		return;
 
-	radeon_audio_write_speaker_allocation(encoder);
-	radeon_audio_write_sad_regs(encoder);
-	radeon_audio_write_latency_fields(encoder, mode);
-	if (rdev->clock.dp_extclk || ASIC_IS_DCE5(rdev))
-		radeon_audio_set_dto(encoder, rdev->clock.default_dispclk * 10);
-	else
-		radeon_audio_set_dto(encoder, dig_connector->dp_clock);
-	radeon_audio_set_audio_packet(encoder);
-	radeon_audio_select_pin(encoder);
-
-	if (radeon_audio_set_avi_packet(encoder, mode) < 0)
+	if (!connector)
 		return;
+
+	if (drm_detect_monitor_audio(radeon_connector_edid(connector))) {
+		radeon_audio_write_speaker_allocation(encoder);
+		radeon_audio_write_sad_regs(encoder);
+		radeon_audio_write_latency_fields(encoder, mode);
+		if (rdev->clock.dp_extclk || ASIC_IS_DCE5(rdev))
+			radeon_audio_set_dto(encoder, rdev->clock.default_dispclk * 10);
+		else
+			radeon_audio_set_dto(encoder, dig_connector->dp_clock);
+		radeon_audio_set_audio_packet(encoder);
+		radeon_audio_select_pin(encoder);
+
+		if (radeon_audio_set_avi_packet(encoder, mode) < 0)
+			return;
+	}
 }
 
 void radeon_audio_mode_set(struct drm_encoder *encoder,
-	struct drm_display_mode *mode)
+			   struct drm_display_mode *mode)
 {
 	struct radeon_encoder *radeon_encoder = to_radeon_encoder(encoder);
 
diff --git a/drivers/gpu/drm/radeon/radeon_audio.h b/drivers/gpu/drm/radeon/radeon_audio.h
index 8438304f7139..059cc3012062 100644
--- a/drivers/gpu/drm/radeon/radeon_audio.h
+++ b/drivers/gpu/drm/radeon/radeon_audio.h
@@ -68,7 +68,8 @@ struct radeon_audio_funcs
 
 int radeon_audio_init(struct radeon_device *rdev);
 void radeon_audio_detect(struct drm_connector *connector,
-	enum drm_connector_status status);
+			 struct drm_encoder *encoder,
+			 enum drm_connector_status status);
 u32 radeon_audio_endpoint_rreg(struct radeon_device *rdev,
 	u32 offset, u32 reg);
 void radeon_audio_endpoint_wreg(struct radeon_device *rdev,
diff --git a/drivers/gpu/drm/radeon/radeon_combios.c b/drivers/gpu/drm/radeon/radeon_combios.c
index 3e5f6b71f3ad..c097d3a82bda 100644
--- a/drivers/gpu/drm/radeon/radeon_combios.c
+++ b/drivers/gpu/drm/radeon/radeon_combios.c
@@ -1255,10 +1255,15 @@ struct radeon_encoder_lvds *radeon_combios_get_lvds_info(struct radeon_encoder
 
 			if ((RBIOS16(tmp) == lvds->native_mode.hdisplay) &&
 			    (RBIOS16(tmp + 2) == lvds->native_mode.vdisplay)) {
+				u32 hss = (RBIOS16(tmp + 21) - RBIOS16(tmp + 19) - 1) * 8;
+
+				if (hss > lvds->native_mode.hdisplay)
+					hss = (10 - 1) * 8;
+
 				lvds->native_mode.htotal = lvds->native_mode.hdisplay +
 					(RBIOS16(tmp + 17) - RBIOS16(tmp + 19)) * 8;
 				lvds->native_mode.hsync_start = lvds->native_mode.hdisplay +
-					(RBIOS16(tmp + 21) - RBIOS16(tmp + 19) - 1) * 8;
+					hss;
 				lvds->native_mode.hsync_end = lvds->native_mode.hsync_start +
 					(RBIOS8(tmp + 23) * 8);
 
diff --git a/drivers/gpu/drm/radeon/radeon_connectors.c b/drivers/gpu/drm/radeon/radeon_connectors.c
index cebb65e07e1d..94b21ae70ef7 100644
--- a/drivers/gpu/drm/radeon/radeon_connectors.c
+++ b/drivers/gpu/drm/radeon/radeon_connectors.c
@@ -1379,8 +1379,16 @@ out:
 	/* updated in get modes as well since we need to know if it's analog or digital */
 	radeon_connector_update_scratch_regs(connector, ret);
 
-	if (radeon_audio != 0)
-		radeon_audio_detect(connector, ret);
+	if ((radeon_audio != 0) && radeon_connector->use_digital) {
+		const struct drm_connector_helper_funcs *connector_funcs =
+			connector->helper_private;
+
+		encoder = connector_funcs->best_encoder(connector);
+		if (encoder && (encoder->encoder_type == DRM_MODE_ENCODER_TMDS)) {
+			radeon_connector_get_edid(connector);
+			radeon_audio_detect(connector, encoder, ret);
+		}
+	}
 
 exit:
 	pm_runtime_mark_last_busy(connector->dev->dev);
@@ -1717,8 +1725,10 @@ radeon_dp_detect(struct drm_connector *connector, bool force)
 
 	radeon_connector_update_scratch_regs(connector, ret);
 
-	if (radeon_audio != 0)
-		radeon_audio_detect(connector, ret);
+	if ((radeon_audio != 0) && encoder) {
+		radeon_connector_get_edid(connector);
+		radeon_audio_detect(connector, encoder, ret);
+	}
 
 out:
 	pm_runtime_mark_last_busy(connector->dev->dev);
diff --git a/drivers/gpu/drm/radeon/radeon_mode.h b/drivers/gpu/drm/radeon/radeon_mode.h
index 07909d817381..aecc3e3dec0c 100644
--- a/drivers/gpu/drm/radeon/radeon_mode.h
+++ b/drivers/gpu/drm/radeon/radeon_mode.h
@@ -237,7 +237,6 @@ struct radeon_afmt {
 	int offset;
 	bool last_buffer_filled_status;
 	int id;
-	struct r600_audio_pin *pin;
 };
 
 struct radeon_mode_info {
@@ -439,6 +438,7 @@ struct radeon_encoder_atom_dig {
 	uint8_t backlight_level;
 	int panel_mode;
 	struct radeon_afmt *afmt;
+	struct r600_audio_pin *pin;
 	int active_mst_links;
 };
 
diff --git a/drivers/gpu/drm/ttm/ttm_bo_util.c b/drivers/gpu/drm/ttm/ttm_bo_util.c
index 882cccdad272..ac6fe40b99f7 100644
--- a/drivers/gpu/drm/ttm/ttm_bo_util.c
+++ b/drivers/gpu/drm/ttm/ttm_bo_util.c
@@ -490,7 +490,8 @@ pgprot_t ttm_io_prot(uint32_t caching_flags, pgprot_t tmp)
 	else if (boot_cpu_data.x86 > 3)
 		tmp = pgprot_noncached(tmp);
 #endif
-#if defined(__ia64__) || defined(__arm__) || defined(__powerpc__)
+#if defined(__ia64__) || defined(__arm__) || defined(__aarch64__) || \
+    defined(__powerpc__)
 	if (caching_flags & TTM_PL_FLAG_WC)
 		tmp = pgprot_writecombine(tmp);
 	else