Merge tag 'drm-next-2023-05-05' of git://anongit.freedesktop.org/drm/drm

Pull more drm fixes from Dave Airlie:
 "This is the fixes for the last couple of weeks for i915 and last 3
  weeks for amdgpu, lots of them but pretty scattered around and all
  pretty small.

  amdgpu:
   - SR-IOV fixes
   - DCN 3.2 fixes
   - DC mclk handling fixes
   - eDP fixes
   - SubVP fixes
   - HDCP regression fix
   - DSC fixes
   - DC FP fixes
   - DCN 3.x fixes
   - Display flickering fix when switching between vram and gtt
   - Z8 power saving fix
   - Fix hang when skipping modeset
   - GPU reset fixes
   - Doorbell fix when resizing BARs
   - Fix spurious warnings in gmc
   - Locking fix for AMDGPU_SCHED IOCTL
   - SR-IOV fix
   - DCN 3.1.4 fix
   - DCN 3.2 fix
   - Fix job cleanup when CS is aborted

  i915:
   - skl pipe source size check
   - mtl transcoder mask fix
   - DSI power on sequence fix
   - GuC versioning corner case fix"

* tag 'drm-next-2023-05-05' of git://anongit.freedesktop.org/drm/drm: (48 commits)
  drm/amdgpu: drop redundant sched job cleanup when cs is aborted
  drm/amd/display: filter out invalid bits in pipe_fuses
  drm/amd/display: Change default Z8 watermark values
  drm/amdgpu: disable SDMA WPTR_POLL_ENABLE for SR-IOV
  drm/amdgpu: add a missing lock for AMDGPU_SCHED
  drm/amdgpu: fix an amdgpu_irq_put() issue in gmc_v9_0_hw_fini()
  drm/amdgpu: fix amdgpu_irq_put call trace in gmc_v10_0_hw_fini
  drm/amdgpu: fix amdgpu_irq_put call trace in gmc_v11_0_hw_fini
  drm/amdgpu: Enable doorbell selfring after resize FB BAR
  drm/amdgpu: Use the default reset when loading or reloading the driver
  drm/amdgpu: Fix mode2 reset for sienna cichlid
  drm/i915/dsi: Use unconditional msleep() instead of intel_dsi_msleep()
  drm/i915/mtl: Add the missing CPU transcoder mask in intel_device_info
  drm/i915/guc: Actually return an error if GuC version range check fails
  drm/amd/display: Lowering min Z8 residency time
  drm/amd/display: fix flickering caused by S/G mode
  drm/amd/display: Set min_width and min_height capability for DCN30
  drm/amd/display: Isolate remaining FPU code in DCN32
  drm/amd/display: Update bounding box values for DCN321
  drm/amd/display: Do not clear GPINT register when releasing DMUB from reset
  ...

Author: torvalds

drivers/gpu/drm/amd/amdgpu/amdgpu_cs.c

@@ -1276,7 +1276,7 @@ static int amdgpu_cs_submit(struct amdgpu_cs_parser *p,
 		r = drm_sched_job_add_dependency(&leader->base, fence);
 		if (r) {
 			dma_fence_put(fence);
-			goto error_cleanup;
+			return r;
 		}
 	}
 
@@ -1303,7 +1303,8 @@ static int amdgpu_cs_submit(struct amdgpu_cs_parser *p,
 	}
 	if (r) {
 		r = -EAGAIN;
-		goto error_unlock;
+		mutex_unlock(&p->adev->notifier_lock);
+		return r;
 	}
 
 	p->fence = dma_fence_get(&leader->base.s_fence->finished);
@@ -1350,14 +1351,6 @@ static int amdgpu_cs_submit(struct amdgpu_cs_parser *p,
 	mutex_unlock(&p->adev->notifier_lock);
 	mutex_unlock(&p->bo_list->bo_list_mutex);
 	return 0;
-
-error_unlock:
-	mutex_unlock(&p->adev->notifier_lock);
-
-error_cleanup:
-	for (i = 0; i < p->gang_size; ++i)
-		drm_sched_job_cleanup(&p->jobs[i]->base);
-	return r;
 }
 
 /* Cleanup the parser structure */

drivers/gpu/drm/amd/amdgpu/amdgpu_device.c

@@ -2539,8 +2539,6 @@ static int amdgpu_device_ip_init(struct amdgpu_device *adev)
 	amdgpu_fru_get_product_info(adev);
 
 init_failed:
-	if (amdgpu_sriov_vf(adev))
-		amdgpu_virt_release_full_gpu(adev, true);
 
 	return r;
 }
@@ -3580,6 +3578,7 @@ int amdgpu_device_init(struct amdgpu_device *adev,
 	int r, i;
 	bool px = false;
 	u32 max_MBps;
+	int tmp;
 
 	adev->shutdown = false;
 	adev->flags = flags;
@@ -3801,7 +3800,13 @@ int amdgpu_device_init(struct amdgpu_device *adev,
 				}
 			}
 		} else {
+			tmp = amdgpu_reset_method;
+			/* It should do a default reset when loading or reloading the driver,
+			 * regardless of the module parameter reset_method.
+			 */
+			amdgpu_reset_method = AMD_RESET_METHOD_NONE;
 			r = amdgpu_asic_reset(adev);
+			amdgpu_reset_method = tmp;
 			if (r) {
 				dev_err(adev->dev, "asic reset on init failed\n");
 				goto failed;
@@ -3859,18 +3864,6 @@ int amdgpu_device_init(struct amdgpu_device *adev,
 
 	r = amdgpu_device_ip_init(adev);
 	if (r) {
-		/* failed in exclusive mode due to timeout */
-		if (amdgpu_sriov_vf(adev) &&
-		    !amdgpu_sriov_runtime(adev) &&
-		    amdgpu_virt_mmio_blocked(adev) &&
-		    !amdgpu_virt_wait_reset(adev)) {
-			dev_err(adev->dev, "VF exclusive mode timeout\n");
-			/* Don't send request since VF is inactive. */
-			adev->virt.caps &= ~AMDGPU_SRIOV_CAPS_RUNTIME;
-			adev->virt.ops = NULL;
-			r = -EAGAIN;
-			goto release_ras_con;
-		}
 		dev_err(adev->dev, "amdgpu_device_ip_init failed\n");
 		amdgpu_vf_error_put(adev, AMDGIM_ERROR_VF_AMDGPU_INIT_FAIL, 0, 0);
 		goto release_ras_con;
@@ -3939,8 +3932,10 @@ int amdgpu_device_init(struct amdgpu_device *adev,
 				   msecs_to_jiffies(AMDGPU_RESUME_MS));
 	}
 
-	if (amdgpu_sriov_vf(adev))
+	if (amdgpu_sriov_vf(adev)) {
+		amdgpu_virt_release_full_gpu(adev, true);
 		flush_delayed_work(&adev->delayed_init_work);
+	}
 
 	r = sysfs_create_files(&adev->dev->kobj, amdgpu_dev_attributes);
 	if (r)
@@ -3980,6 +3975,20 @@ int amdgpu_device_init(struct amdgpu_device *adev,
 	return 0;
 
 release_ras_con:
+	if (amdgpu_sriov_vf(adev))
+		amdgpu_virt_release_full_gpu(adev, true);
+
+	/* failed in exclusive mode due to timeout */
+	if (amdgpu_sriov_vf(adev) &&
+		!amdgpu_sriov_runtime(adev) &&
+		amdgpu_virt_mmio_blocked(adev) &&
+		!amdgpu_virt_wait_reset(adev)) {
+		dev_err(adev->dev, "VF exclusive mode timeout\n");
+		/* Don't send request since VF is inactive. */
+		adev->virt.caps &= ~AMDGPU_SRIOV_CAPS_RUNTIME;
+		adev->virt.ops = NULL;
+		r = -EAGAIN;
+	}
 	amdgpu_release_ras_context(adev);
 
 failed:

drivers/gpu/drm/amd/amdgpu/amdgpu_sched.c

@@ -38,6 +38,7 @@ static int amdgpu_sched_process_priority_override(struct amdgpu_device *adev,
 {
 	struct fd f = fdget(fd);
 	struct amdgpu_fpriv *fpriv;
+	struct amdgpu_ctx_mgr *mgr;
 	struct amdgpu_ctx *ctx;
 	uint32_t id;
 	int r;
@@ -51,8 +52,11 @@ static int amdgpu_sched_process_priority_override(struct amdgpu_device *adev,
 		return r;
 	}
 
-	idr_for_each_entry(&fpriv->ctx_mgr.ctx_handles, ctx, id)
+	mgr = &fpriv->ctx_mgr;
+	mutex_lock(&mgr->lock);
+	idr_for_each_entry(&mgr->ctx_handles, ctx, id)
 		amdgpu_ctx_priority_override(ctx, priority);
+	mutex_unlock(&mgr->lock);
 
 	fdput(f);
 	return 0;

drivers/gpu/drm/amd/amdgpu/gmc_v10_0.c

@@ -1143,7 +1143,6 @@ static int gmc_v10_0_hw_fini(void *handle)
 		return 0;
 	}
 
-	amdgpu_irq_put(adev, &adev->gmc.ecc_irq, 0);
 	amdgpu_irq_put(adev, &adev->gmc.vm_fault, 0);
 
 	return 0;

drivers/gpu/drm/amd/amdgpu/gmc_v11_0.c

@@ -951,7 +951,6 @@ static int gmc_v11_0_hw_fini(void *handle)
 		return 0;
 	}
 
-	amdgpu_irq_put(adev, &adev->gmc.ecc_irq, 0);
 	amdgpu_irq_put(adev, &adev->gmc.vm_fault, 0);
 	gmc_v11_0_gart_disable(adev);
 

drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c

@@ -1999,7 +1999,6 @@ static int gmc_v9_0_hw_fini(void *handle)
 	if (adev->mmhub.funcs->update_power_gating)
 		adev->mmhub.funcs->update_power_gating(adev, false);
 
-	amdgpu_irq_put(adev, &adev->gmc.ecc_irq, 0);
 	amdgpu_irq_put(adev, &adev->gmc.vm_fault, 0);
 
 	return 0;

drivers/gpu/drm/amd/amdgpu/jpeg_v4_0.c

@@ -430,7 +430,7 @@ static int jpeg_v4_0_start_sriov(struct amdgpu_device *adev)
 		MMSCH_COMMAND__END;
 
 	header.version = MMSCH_VERSION;
-	header.total_size = sizeof(struct mmsch_v4_0_init_header) >> 2;
+	header.total_size = RREG32_SOC15(VCN, 0, regMMSCH_VF_CTX_SIZE);
 
 	header.jpegdec.init_status = 0;
 	header.jpegdec.table_offset = 0;

drivers/gpu/drm/amd/amdgpu/nv.c

@@ -531,13 +531,6 @@ static void nv_program_aspm(struct amdgpu_device *adev)
 
 }
 
-static void nv_enable_doorbell_aperture(struct amdgpu_device *adev,
-					bool enable)
-{
-	adev->nbio.funcs->enable_doorbell_aperture(adev, enable);
-	adev->nbio.funcs->enable_doorbell_selfring_aperture(adev, enable);
-}
-
 const struct amdgpu_ip_block_version nv_common_ip_block =
 {
 	.type = AMD_IP_BLOCK_TYPE_COMMON,
@@ -999,6 +992,11 @@ static int nv_common_late_init(void *handle)
 		}
 	}
 
+	/* Enable selfring doorbell aperture late because doorbell BAR
+	 * aperture will change if resize BAR successfully in gmc sw_init.
+	 */
+	adev->nbio.funcs->enable_doorbell_selfring_aperture(adev, true);
+
 	return 0;
 }
 
@@ -1038,7 +1036,7 @@ static int nv_common_hw_init(void *handle)
 	if (adev->nbio.funcs->remap_hdp_registers && !amdgpu_sriov_vf(adev))
 		adev->nbio.funcs->remap_hdp_registers(adev);
 	/* enable the doorbell aperture */
-	nv_enable_doorbell_aperture(adev, true);
+	adev->nbio.funcs->enable_doorbell_aperture(adev, true);
 
 	return 0;
 }
@@ -1047,8 +1045,13 @@ static int nv_common_hw_fini(void *handle)
 {
 	struct amdgpu_device *adev = (struct amdgpu_device *)handle;
 
-	/* disable the doorbell aperture */
-	nv_enable_doorbell_aperture(adev, false);
+	/* Disable the doorbell aperture and selfring doorbell aperture
+	 * separately in hw_fini because nv_enable_doorbell_aperture
+	 * has been removed and there is no need to delay disabling
+	 * selfring doorbell.
+	 */
+	adev->nbio.funcs->enable_doorbell_aperture(adev, false);
+	adev->nbio.funcs->enable_doorbell_selfring_aperture(adev, false);
 
 	return 0;
 }

drivers/gpu/drm/amd/amdgpu/sdma_v6_0.c

@@ -510,10 +510,7 @@ static int sdma_v6_0_gfx_resume(struct amdgpu_device *adev)
 		       lower_32_bits(ring->rptr_gpu_addr) & 0xFFFFFFFC);
 
 		rb_cntl = REG_SET_FIELD(rb_cntl, SDMA0_QUEUE0_RB_CNTL, RPTR_WRITEBACK_ENABLE, 1);
-		if (amdgpu_sriov_vf(adev))
-			rb_cntl = REG_SET_FIELD(rb_cntl, SDMA0_QUEUE0_RB_CNTL, WPTR_POLL_ENABLE, 1);
-		else
-			rb_cntl = REG_SET_FIELD(rb_cntl, SDMA0_QUEUE0_RB_CNTL, WPTR_POLL_ENABLE, 0);
+		rb_cntl = REG_SET_FIELD(rb_cntl, SDMA0_QUEUE0_RB_CNTL, WPTR_POLL_ENABLE, 0);
 		rb_cntl = REG_SET_FIELD(rb_cntl, SDMA0_QUEUE0_RB_CNTL, F32_WPTR_POLL_ENABLE, 1);
 
 		WREG32_SOC15_IP(GC, sdma_v6_0_get_reg_offset(adev, i, regSDMA0_QUEUE0_RB_BASE), ring->gpu_addr >> 8);

drivers/gpu/drm/amd/amdgpu/sienna_cichlid.c

@@ -40,7 +40,7 @@ static bool sienna_cichlid_is_mode2_default(struct amdgpu_reset_control *reset_c
 	    adev->pm.fw_version >= 0x3a5500 && !amdgpu_sriov_vf(adev))
 		return true;
 #endif
-	return false;
+	return amdgpu_reset_method == AMD_RESET_METHOD_MODE2;
 }
 
 static struct amdgpu_reset_handler *