Merge tag 'drm-fixes-2020-04-18' of git://anongit.freedesktop.org/drm/drm

torvalds · torvalds · commit fecca6899783 · 2020-04-18T11:25:58.000-07:00
Pull drm fixes from Dave Airlie:
 "Quiet enough for rc2, mostly amdgpu fixes, a couple of i915 fixes, and
  one nouveau module firmware fix:

  i915:
   - Fix guest page access by using the brand new VFIO dma r/w interface (Yan)
   - Fix for i915 perf read buffers (Ashutosh)

  amdgpu:
   - gfx10 fix
   - SMU7 overclocking fix
   - RAS fix
   - GPU reset fix
   - Fix a regression in a previous suspend/resume fix
   - Add a gfxoff quirk

  nouveau:
   - fix missing MODULE_FIRMWARE"

* tag 'drm-fixes-2020-04-18' of git://anongit.freedesktop.org/drm/drm:
  drm/nouveau/sec2/gv100-: add missing MODULE_FIRMWARE()
  drm/amdgpu/gfx9: add gfxoff quirk
  drm/amdgpu: fix the hw hang during perform system reboot and reset
  drm/i915/gvt: switch to user vfio_group_pin/upin_pages
  drm/i915/gvt: subsitute kvm_read/write_guest with vfio_dma_rw
  drm/i915/gvt: hold reference of VFIO group during opening of vgpu
  drm/i915/perf: Do not clear pollin for small user read buffers
  drm/amdgpu: fix wrong vram lost counter increment V2
  drm/amd/powerplay: unload mp1 for Arcturus RAS baco reset
  drm/amd/powerplay: force the trim of the mclk dpm_levels if OD is enabled
  Revert "drm/amdgpu: change SH MEM alignment mode for gfx10"
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
@@ -2008,8 +2008,24 @@ static void amdgpu_device_fill_reset_magic(struct amdgpu_device *adev)
  */
 static bool amdgpu_device_check_vram_lost(struct amdgpu_device *adev)
 {
-	return !!memcmp(adev->gart.ptr, adev->reset_magic,
-			AMDGPU_RESET_MAGIC_NUM);
+	if (memcmp(adev->gart.ptr, adev->reset_magic,
+			AMDGPU_RESET_MAGIC_NUM))
+		return true;
+
+	if (!adev->in_gpu_reset)
+		return false;
+
+	/*
+	 * For all ASICs with baco/mode1 reset, the VRAM is
+	 * always assumed to be lost.
+	 */
+	switch (amdgpu_asic_reset_method(adev)) {
+	case AMD_RESET_METHOD_BACO:
+	case AMD_RESET_METHOD_MODE1:
+		return true;
+	default:
+		return false;
+	}
 }
 
 /**
@@ -2340,6 +2356,8 @@ static int amdgpu_device_ip_suspend_phase1(struct amdgpu_device *adev)
 {
 	int i, r;
 
+	amdgpu_device_set_pg_state(adev, AMD_PG_STATE_UNGATE);
+	amdgpu_device_set_cg_state(adev, AMD_CG_STATE_UNGATE);
 
 	for (i = adev->num_ip_blocks - 1; i >= 0; i--) {
 		if (!adev->ip_blocks[i].status.valid)
diff --git a/drivers/gpu/drm/amd/amdgpu/cik.c b/drivers/gpu/drm/amd/amdgpu/cik.c
@@ -1358,8 +1358,6 @@ static int cik_asic_reset(struct amdgpu_device *adev)
 	int r;
 
 	if (cik_asic_reset_method(adev) == AMD_RESET_METHOD_BACO) {
-		if (!adev->in_suspend)
-			amdgpu_inc_vram_lost(adev);
 		r = amdgpu_dpm_baco_reset(adev);
 	} else {
 		r = cik_asic_pci_config_reset(adev);
diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c
@@ -279,7 +279,7 @@ static const struct soc15_reg_golden golden_settings_gc_10_1_2_nv12[] =
 
 #define DEFAULT_SH_MEM_CONFIG \
 	((SH_MEM_ADDRESS_MODE_64 << SH_MEM_CONFIG__ADDRESS_MODE__SHIFT) | \
-	 (SH_MEM_ALIGNMENT_MODE_DWORD << SH_MEM_CONFIG__ALIGNMENT_MODE__SHIFT) | \
+	 (SH_MEM_ALIGNMENT_MODE_UNALIGNED << SH_MEM_CONFIG__ALIGNMENT_MODE__SHIFT) | \
 	 (SH_MEM_RETRY_MODE_ALL << SH_MEM_CONFIG__RETRY_MODE__SHIFT) | \
 	 (3 << SH_MEM_CONFIG__INITIAL_INST_PREFETCH__SHIFT))
 
diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c
@@ -1234,6 +1234,8 @@ struct amdgpu_gfxoff_quirk {
 static const struct amdgpu_gfxoff_quirk amdgpu_gfxoff_quirk_list[] = {
 	/* https://bugzilla.kernel.org/show_bug.cgi?id=204689 */
 	{ 0x1002, 0x15dd, 0x1002, 0x15dd, 0xc8 },
+	/* https://bugzilla.kernel.org/show_bug.cgi?id=207171 */
+	{ 0x1002, 0x15dd, 0x103c, 0x83e7, 0xd3 },
 	{ 0, 0, 0, 0, 0 },
 };
 
diff --git a/drivers/gpu/drm/amd/amdgpu/nv.c b/drivers/gpu/drm/amd/amdgpu/nv.c
@@ -351,17 +351,13 @@ static int nv_asic_reset(struct amdgpu_device *adev)
 	struct smu_context *smu = &adev->smu;
 
 	if (nv_asic_reset_method(adev) == AMD_RESET_METHOD_BACO) {
-		if (!adev->in_suspend)
-			amdgpu_inc_vram_lost(adev);
 		ret = smu_baco_enter(smu);
 		if (ret)
 			return ret;
 		ret = smu_baco_exit(smu);
 		if (ret)
 			return ret;
 	} else {
-		if (!adev->in_suspend)
-			amdgpu_inc_vram_lost(adev);
 		ret = nv_asic_mode1_reset(adev);
 	}
 
diff --git a/drivers/gpu/drm/amd/amdgpu/soc15.c b/drivers/gpu/drm/amd/amdgpu/soc15.c
@@ -569,14 +569,10 @@ static int soc15_asic_reset(struct amdgpu_device *adev)
 
 	switch (soc15_asic_reset_method(adev)) {
 		case AMD_RESET_METHOD_BACO:
-			if (!adev->in_suspend)
-				amdgpu_inc_vram_lost(adev);
 			return soc15_asic_baco_reset(adev);
 		case AMD_RESET_METHOD_MODE2:
 			return amdgpu_dpm_mode2_reset(adev);
 		default:
-			if (!adev->in_suspend)
-				amdgpu_inc_vram_lost(adev);
 			return soc15_asic_mode1_reset(adev);
 	}
 }
diff --git a/drivers/gpu/drm/amd/amdgpu/vi.c b/drivers/gpu/drm/amd/amdgpu/vi.c
@@ -765,8 +765,6 @@ static int vi_asic_reset(struct amdgpu_device *adev)
 	int r;
 
 	if (vi_asic_reset_method(adev) == AMD_RESET_METHOD_BACO) {
-		if (!adev->in_suspend)
-			amdgpu_inc_vram_lost(adev);
 		r = amdgpu_dpm_baco_reset(adev);
 	} else {
 		r = vi_asic_pci_config_reset(adev);
diff --git a/drivers/gpu/drm/amd/powerplay/hwmgr/smu7_hwmgr.c b/drivers/gpu/drm/amd/powerplay/hwmgr/smu7_hwmgr.c
@@ -3804,9 +3804,12 @@ static int smu7_trim_single_dpm_states(struct pp_hwmgr *hwmgr,
 {
 	uint32_t i;
 
+	/* force the trim if mclk_switching is disabled to prevent flicker */
+	bool force_trim = (low_limit == high_limit);
 	for (i = 0; i < dpm_table->count; i++) {
 	/*skip the trim if od is enabled*/
-		if (!hwmgr->od_enabled && (dpm_table->dpm_levels[i].value < low_limit
+		if ((!hwmgr->od_enabled || force_trim)
+			&& (dpm_table->dpm_levels[i].value < low_limit
 			|| dpm_table->dpm_levels[i].value > high_limit))
 			dpm_table->dpm_levels[i].enabled = false;
 		else
diff --git a/drivers/gpu/drm/amd/powerplay/smu_v11_0.c b/drivers/gpu/drm/amd/powerplay/smu_v11_0.c
@@ -1718,6 +1718,12 @@ int smu_v11_0_baco_set_state(struct smu_context *smu, enum smu_baco_state state)
 		if (ret)
 			goto out;
 
+		if (ras && ras->supported) {
+			ret = smu_send_smc_msg(smu, SMU_MSG_PrepareMp1ForUnload, NULL);
+			if (ret)
+				goto out;
+		}
+
 		/* clear vbios scratch 6 and 7 for coming asic reinit */
 		WREG32(adev->bios_scratch_reg_offset + 6, 0);
 		WREG32(adev->bios_scratch_reg_offset + 7, 0);
diff --git a/drivers/gpu/drm/i915/gvt/kvmgt.c b/drivers/gpu/drm/i915/gvt/kvmgt.c
@@ -131,6 +131,7 @@ struct kvmgt_vdev {
 	struct work_struct release_work;
 	atomic_t released;
 	struct vfio_device *vfio_device;
+	struct vfio_group *vfio_group;
 };
 
 static inline struct kvmgt_vdev *kvmgt_vdev(struct intel_vgpu *vgpu)
@@ -151,6 +152,7 @@ static void gvt_unpin_guest_page(struct intel_vgpu *vgpu, unsigned long gfn,
 		unsigned long size)
 {
 	struct drm_i915_private *i915 = vgpu->gvt->gt->i915;
+	struct kvmgt_vdev *vdev = kvmgt_vdev(vgpu);
 	int total_pages;
 	int npage;
 	int ret;
@@ -160,7 +162,7 @@ static void gvt_unpin_guest_page(struct intel_vgpu *vgpu, unsigned long gfn,
 	for (npage = 0; npage < total_pages; npage++) {
 		unsigned long cur_gfn = gfn + npage;
 
-		ret = vfio_unpin_pages(mdev_dev(kvmgt_vdev(vgpu)->mdev), &cur_gfn, 1);
+		ret = vfio_group_unpin_pages(vdev->vfio_group, &cur_gfn, 1);
 		drm_WARN_ON(&i915->drm, ret != 1);
 	}
 }
@@ -169,6 +171,7 @@ static void gvt_unpin_guest_page(struct intel_vgpu *vgpu, unsigned long gfn,
 static int gvt_pin_guest_page(struct intel_vgpu *vgpu, unsigned long gfn,
 		unsigned long size, struct page **page)
 {
+	struct kvmgt_vdev *vdev = kvmgt_vdev(vgpu);
 	unsigned long base_pfn = 0;
 	int total_pages;
 	int npage;
@@ -183,8 +186,8 @@ static int gvt_pin_guest_page(struct intel_vgpu *vgpu, unsigned long gfn,
 		unsigned long cur_gfn = gfn + npage;
 		unsigned long pfn;
 
-		ret = vfio_pin_pages(mdev_dev(kvmgt_vdev(vgpu)->mdev), &cur_gfn, 1,
-				     IOMMU_READ | IOMMU_WRITE, &pfn);
+		ret = vfio_group_pin_pages(vdev->vfio_group, &cur_gfn, 1,
+					   IOMMU_READ | IOMMU_WRITE, &pfn);
 		if (ret != 1) {
 			gvt_vgpu_err("vfio_pin_pages failed for gfn 0x%lx, ret %d\n",
 				     cur_gfn, ret);
@@ -792,6 +795,7 @@ static int intel_vgpu_open(struct mdev_device *mdev)
 	struct kvmgt_vdev *vdev = kvmgt_vdev(vgpu);
 	unsigned long events;
 	int ret;
+	struct vfio_group *vfio_group;
 
 	vdev->iommu_notifier.notifier_call = intel_vgpu_iommu_notifier;
 	vdev->group_notifier.notifier_call = intel_vgpu_group_notifier;
@@ -814,6 +818,14 @@ static int intel_vgpu_open(struct mdev_device *mdev)
 		goto undo_iommu;
 	}
 
+	vfio_group = vfio_group_get_external_user_from_dev(mdev_dev(mdev));
+	if (IS_ERR_OR_NULL(vfio_group)) {
+		ret = !vfio_group ? -EFAULT : PTR_ERR(vfio_group);
+		gvt_vgpu_err("vfio_group_get_external_user_from_dev failed\n");
+		goto undo_register;
+	}
+	vdev->vfio_group = vfio_group;
+
 	/* Take a module reference as mdev core doesn't take
 	 * a reference for vendor driver.
 	 */
@@ -830,6 +842,10 @@ static int intel_vgpu_open(struct mdev_device *mdev)
 	return ret;
 
 undo_group:
+	vfio_group_put_external_user(vdev->vfio_group);
+	vdev->vfio_group = NULL;
+
+undo_register:
 	vfio_unregister_notifier(mdev_dev(mdev), VFIO_GROUP_NOTIFY,
 					&vdev->group_notifier);
 
@@ -884,6 +900,7 @@ static void __intel_vgpu_release(struct intel_vgpu *vgpu)
 	kvmgt_guest_exit(info);
 
 	intel_vgpu_release_msi_eventfd_ctx(vgpu);
+	vfio_group_put_external_user(vdev->vfio_group);
 
 	vdev->kvm = NULL;
 	vgpu->handle = 0;
@@ -2035,33 +2052,14 @@ static int kvmgt_rw_gpa(unsigned long handle, unsigned long gpa,
 			void *buf, unsigned long len, bool write)
 {
 	struct kvmgt_guest_info *info;
-	struct kvm *kvm;
-	int idx, ret;
-	bool kthread = current->mm == NULL;
 
 	if (!handle_valid(handle))
 		return -ESRCH;
 
 	info = (struct kvmgt_guest_info *)handle;
-	kvm = info->kvm;
-
-	if (kthread) {
-		if (!mmget_not_zero(kvm->mm))
-			return -EFAULT;
-		use_mm(kvm->mm);
-	}
-
-	idx = srcu_read_lock(&kvm->srcu);
-	ret = write ? kvm_write_guest(kvm, gpa, buf, len) :
-		      kvm_read_guest(kvm, gpa, buf, len);
-	srcu_read_unlock(&kvm->srcu, idx);
-
-	if (kthread) {
-		unuse_mm(kvm->mm);
-		mmput(kvm->mm);
-	}
 
-	return ret;
+	return vfio_dma_rw(kvmgt_vdev(info->vgpu)->vfio_group,
+			   gpa, buf, len, write);
 }
 
 static int kvmgt_read_gpa(unsigned long handle, unsigned long gpa,
diff --git a/drivers/gpu/drm/i915/i915_perf.c b/drivers/gpu/drm/i915/i915_perf.c
@@ -2940,49 +2940,6 @@ void i915_oa_init_reg_state(const struct intel_context *ce,
 		gen8_update_reg_state_unlocked(ce, stream);
 }
 
-/**
- * i915_perf_read_locked - &i915_perf_stream_ops->read with error normalisation
- * @stream: An i915 perf stream
- * @file: An i915 perf stream file
- * @buf: destination buffer given by userspace
- * @count: the number of bytes userspace wants to read
- * @ppos: (inout) file seek position (unused)
- *
- * Besides wrapping &i915_perf_stream_ops->read this provides a common place to
- * ensure that if we've successfully copied any data then reporting that takes
- * precedence over any internal error status, so the data isn't lost.
- *
- * For example ret will be -ENOSPC whenever there is more buffered data than
- * can be copied to userspace, but that's only interesting if we weren't able
- * to copy some data because it implies the userspace buffer is too small to
- * receive a single record (and we never split records).
- *
- * Another case with ret == -EFAULT is more of a grey area since it would seem
- * like bad form for userspace to ask us to overrun its buffer, but the user
- * knows best:
- *
- *   http://yarchive.net/comp/linux/partial_reads_writes.html
- *
- * Returns: The number of bytes copied or a negative error code on failure.
- */
-static ssize_t i915_perf_read_locked(struct i915_perf_stream *stream,
-				     struct file *file,
-				     char __user *buf,
-				     size_t count,
-				     loff_t *ppos)
-{
-	/* Note we keep the offset (aka bytes read) separate from any
-	 * error status so that the final check for whether we return
-	 * the bytes read with a higher precedence than any error (see
-	 * comment below) doesn't need to be handled/duplicated in
-	 * stream->ops->read() implementations.
-	 */
-	size_t offset = 0;
-	int ret = stream->ops->read(stream, buf, count, &offset);
-
-	return offset ?: (ret ?: -EAGAIN);
-}
-
 /**
  * i915_perf_read - handles read() FOP for i915 perf stream FDs
  * @file: An i915 perf stream file
@@ -3008,7 +2965,8 @@ static ssize_t i915_perf_read(struct file *file,
 {
 	struct i915_perf_stream *stream = file->private_data;
 	struct i915_perf *perf = stream->perf;
-	ssize_t ret;
+	size_t offset = 0;
+	int ret;
 
 	/* To ensure it's handled consistently we simply treat all reads of a
 	 * disabled stream as an error. In particular it might otherwise lead
@@ -3031,13 +2989,12 @@ static ssize_t i915_perf_read(struct file *file,
 				return ret;
 
 			mutex_lock(&perf->lock);
-			ret = i915_perf_read_locked(stream, file,
-						    buf, count, ppos);
+			ret = stream->ops->read(stream, buf, count, &offset);
 			mutex_unlock(&perf->lock);
-		} while (ret == -EAGAIN);
+		} while (!offset && !ret);
 	} else {
 		mutex_lock(&perf->lock);
-		ret = i915_perf_read_locked(stream, file, buf, count, ppos);
+		ret = stream->ops->read(stream, buf, count, &offset);
 		mutex_unlock(&perf->lock);
 	}
 
@@ -3048,15 +3005,15 @@ static ssize_t i915_perf_read(struct file *file,
 	 * and read() returning -EAGAIN. Clearing the oa.pollin state here
 	 * effectively ensures we back off until the next hrtimer callback
 	 * before reporting another EPOLLIN event.
+	 * The exception to this is if ops->read() returned -ENOSPC which means
+	 * that more OA data is available than could fit in the user provided
+	 * buffer. In this case we want the next poll() call to not block.
 	 */
-	if (ret >= 0 || ret == -EAGAIN) {
-		/* Maybe make ->pollin per-stream state if we support multiple
-		 * concurrent streams in the future.
-		 */
+	if (ret != -ENOSPC)
 		stream->pollin = false;
-	}
 
-	return ret;
+	/* Possible values for ret are 0, -EFAULT, -ENOSPC, -EIO, ... */
+	return offset ?: (ret ?: -EAGAIN);
 }
 
 static enum hrtimer_restart oa_poll_check_timer_cb(struct hrtimer *hrtimer)
diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/sec2/gp108.c b/drivers/gpu/drm/nouveau/nvkm/engine/sec2/gp108.c
@@ -25,6 +25,9 @@
 MODULE_FIRMWARE("nvidia/gp108/sec2/desc.bin");
 MODULE_FIRMWARE("nvidia/gp108/sec2/image.bin");
 MODULE_FIRMWARE("nvidia/gp108/sec2/sig.bin");
+MODULE_FIRMWARE("nvidia/gv100/sec2/desc.bin");
+MODULE_FIRMWARE("nvidia/gv100/sec2/image.bin");
+MODULE_FIRMWARE("nvidia/gv100/sec2/sig.bin");
 
 static const struct nvkm_sec2_fwif
 gp108_sec2_fwif[] = {
diff --git a/drivers/gpu/drm/nouveau/nvkm/engine/sec2/tu102.c b/drivers/gpu/drm/nouveau/nvkm/engine/sec2/tu102.c
@@ -56,6 +56,22 @@ tu102_sec2_nofw(struct nvkm_sec2 *sec2, int ver,
 	return 0;
 }
 
+MODULE_FIRMWARE("nvidia/tu102/sec2/desc.bin");
+MODULE_FIRMWARE("nvidia/tu102/sec2/image.bin");
+MODULE_FIRMWARE("nvidia/tu102/sec2/sig.bin");
+MODULE_FIRMWARE("nvidia/tu104/sec2/desc.bin");
+MODULE_FIRMWARE("nvidia/tu104/sec2/image.bin");
+MODULE_FIRMWARE("nvidia/tu104/sec2/sig.bin");
+MODULE_FIRMWARE("nvidia/tu106/sec2/desc.bin");
+MODULE_FIRMWARE("nvidia/tu106/sec2/image.bin");
+MODULE_FIRMWARE("nvidia/tu106/sec2/sig.bin");
+MODULE_FIRMWARE("nvidia/tu116/sec2/desc.bin");
+MODULE_FIRMWARE("nvidia/tu116/sec2/image.bin");
+MODULE_FIRMWARE("nvidia/tu116/sec2/sig.bin");
+MODULE_FIRMWARE("nvidia/tu117/sec2/desc.bin");
+MODULE_FIRMWARE("nvidia/tu117/sec2/image.bin");
+MODULE_FIRMWARE("nvidia/tu117/sec2/sig.bin");
+
 static const struct nvkm_sec2_fwif
 tu102_sec2_fwif[] = {
 	{  0, gp102_sec2_load, &tu102_sec2, &gp102_sec2_acr_1 },

Original file line number	Diff line number	Diff line change
`@@ -569,14 +569,10 @@ static int soc15_asic_reset(struct amdgpu_device *adev)`
`569`	`569`
`570`	`570`	`switch (soc15_asic_reset_method(adev)) {`
`571`	`571`	`case AMD_RESET_METHOD_BACO:`
`572`		`- if (!adev->in_suspend)`
`573`		`- amdgpu_inc_vram_lost(adev);`
`574`	`572`	`return soc15_asic_baco_reset(adev);`
`575`	`573`	`case AMD_RESET_METHOD_MODE2:`
`576`	`574`	`return amdgpu_dpm_mode2_reset(adev);`
`577`	`575`	`default:`
`578`		`- if (!adev->in_suspend)`
`579`		`- amdgpu_inc_vram_lost(adev);`
`580`	`576`	`return soc15_asic_mode1_reset(adev);`
`581`	`577`	`}`
`582`	`578`	`}`