Merge tag 'drm-next-5.5-2019-11-08' of git://people.freedesktop.org/~…

…agd5f/linux into drm-next

drm-next-5.5-2019-11-08:

amdgpu:
- Enable VCN dynamic powergating on RV/RV2
- Fixes for Navi14
- Misc Navi fixes
- Fix MSI-X tear down
- Misc Arturus fixes
- Fix xgmi powerstate handling
- Documenation fixes

scheduler:
- Fix static code checker warning
- Fix possible thread reactivation while thread is stopped
- Avoid cleanup if thread is parked

radeon:
- SI dpm fix ported from amdgpu

Signed-off-by: Dave Airlie <airlied@redhat.com>
From: Alex Deucher <alexdeucher@gmail.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20191108212713.5078-1-alexander.deucher@amd.com

Documentation/gpu/amdgpu.rst

@@ -82,12 +82,21 @@ AMDGPU XGMI Support
 AMDGPU RAS Support
 ==================
 
+The AMDGPU RAS interfaces are exposed via sysfs (for informational queries) and
+debugfs (for error injection).
+
 RAS debugfs/sysfs Control and Error Injection Interfaces
 --------------------------------------------------------
 
 .. kernel-doc:: drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c
    :doc: AMDGPU RAS debugfs control interface
 
+RAS Reboot Behavior for Unrecoverable Errors
+--------------------------------------------------------
+
+.. kernel-doc:: drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c
+   :doc: AMDGPU RAS Reboot Behavior for Unrecoverable Errors
+
 RAS Error Count sysfs Interface
 -------------------------------
 
@@ -109,6 +118,32 @@ RAS VRAM Bad Pages sysfs Interface
 .. kernel-doc:: drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c
    :internal:
 
+Sample Code
+-----------
+Sample code for testing error injection can be found here:
+https://cgit.freedesktop.org/mesa/drm/tree/tests/amdgpu/ras_tests.c
+
+This is part of the libdrm amdgpu unit tests which cover several areas of the GPU.
+There are four sets of tests:
+
+RAS Basic Test
+
+The test verifies the RAS feature enabled status and makes sure the necessary sysfs and debugfs files
+are present.
+
+RAS Query Test
+
+This test checks the RAS availability and enablement status for each supported IP block as well as
+the error counts.
+
+RAS Inject Test
+
+This test injects errors for each IP.
+
+RAS Disable Test
+
+This test tests disabling of RAS features for each IP block.
+
 
 GPU Power/Thermal Controls and Monitoring
 =========================================

drivers/gpu/drm/amd/amdgpu/amdgpu.h

@@ -977,6 +977,9 @@ struct amdgpu_device {
 
 	uint64_t			unique_id;
 	uint64_t	df_perfmon_config_assign_mask[AMDGPU_MAX_DF_PERFMONS];
+
+	/* device pstate */
+	int				pstate;
 };
 
 static inline struct amdgpu_device *amdgpu_ttm_adev(struct ttm_bo_device *bdev)

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

@@ -33,7 +33,7 @@ static int amdgpu_benchmark_do_move(struct amdgpu_device *adev, unsigned size,
 {
 	unsigned long start_jiffies;
 	unsigned long end_jiffies;
-	struct dma_fence *fence = NULL;
+	struct dma_fence *fence;
 	int i, r;
 
 	start_jiffies = jiffies;
@@ -44,16 +44,14 @@ static int amdgpu_benchmark_do_move(struct amdgpu_device *adev, unsigned size,
 		if (r)
 			goto exit_do_move;
 		r = dma_fence_wait(fence, false);
+		dma_fence_put(fence);
 		if (r)
 			goto exit_do_move;
-		dma_fence_put(fence);
 	}
 	end_jiffies = jiffies;
 	r = jiffies_to_msecs(end_jiffies - start_jiffies);
 
 exit_do_move:
-	if (fence)
-		dma_fence_put(fence);
 	return r;
 }
 

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

@@ -859,6 +859,9 @@ static int amdgpu_debugfs_test_ib(struct seq_file *m, void *data)
 	struct amdgpu_device *adev = dev->dev_private;
 	int r = 0, i;
 
+	/* Avoid accidently unparking the sched thread during GPU reset */
+	mutex_lock(&adev->lock_reset);
+
 	/* hold on the scheduler */
 	for (i = 0; i < AMDGPU_MAX_RINGS; i++) {
 		struct amdgpu_ring *ring = adev->rings[i];
@@ -884,6 +887,8 @@ static int amdgpu_debugfs_test_ib(struct seq_file *m, void *data)
 		kthread_unpark(ring->sched.thread);
 	}
 
+	mutex_unlock(&adev->lock_reset);
+
 	return 0;
 }
 
@@ -1036,6 +1041,9 @@ static int amdgpu_debugfs_ib_preempt(void *data, u64 val)
 	if (!fences)
 		return -ENOMEM;
 
+	/* Avoid accidently unparking the sched thread during GPU reset */
+	mutex_lock(&adev->lock_reset);
+
 	/* stop the scheduler */
 	kthread_park(ring->sched.thread);
 
@@ -1075,6 +1083,8 @@ static int amdgpu_debugfs_ib_preempt(void *data, u64 val)
 	/* restart the scheduler */
 	kthread_unpark(ring->sched.thread);
 
+	mutex_unlock(&adev->lock_reset);
+
 	ttm_bo_unlock_delayed_workqueue(&adev->mman.bdev, resched);
 
 	kfree(fences);

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

@@ -2057,6 +2057,7 @@ static int amdgpu_device_enable_mgpu_fan_boost(void)
  */
 static int amdgpu_device_ip_late_init(struct amdgpu_device *adev)
 {
+	struct amdgpu_gpu_instance *gpu_instance;
 	int i = 0, r;
 
 	for (i = 0; i < adev->num_ip_blocks; i++) {
@@ -2082,8 +2083,39 @@ static int amdgpu_device_ip_late_init(struct amdgpu_device *adev)
 	if (r)
 		DRM_ERROR("enable mgpu fan boost failed (%d).\n", r);
 
-	/* set to low pstate by default */
-	amdgpu_xgmi_set_pstate(adev, 0);
+
+	if (adev->gmc.xgmi.num_physical_nodes > 1) {
+		mutex_lock(&mgpu_info.mutex);
+
+		/*
+		 * Reset device p-state to low as this was booted with high.
+		 *
+		 * This should be performed only after all devices from the same
+		 * hive get initialized.
+		 *
+		 * However, it's unknown how many device in the hive in advance.
+		 * As this is counted one by one during devices initializations.
+		 *
+		 * So, we wait for all XGMI interlinked devices initialized.
+		 * This may bring some delays as those devices may come from
+		 * different hives. But that should be OK.
+		 */
+		if (mgpu_info.num_dgpu == adev->gmc.xgmi.num_physical_nodes) {
+			for (i = 0; i < mgpu_info.num_gpu; i++) {
+				gpu_instance = &(mgpu_info.gpu_ins[i]);
+				if (gpu_instance->adev->flags & AMD_IS_APU)
+					continue;
+
+				r = amdgpu_xgmi_set_pstate(gpu_instance->adev, 0);
+				if (r) {
+					DRM_ERROR("pstate setting failed (%d).\n", r);
+					break;
+				}
+			}
+		}
+
+		mutex_unlock(&mgpu_info.mutex);
+	}
 
 	return 0;
 }

drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.h

@@ -127,18 +127,48 @@ struct amdgpu_xgmi {
 };
 
 struct amdgpu_gmc {
+	/* FB's physical address in MMIO space (for CPU to
+	 * map FB). This is different compared to the agp/
+	 * gart/vram_start/end field as the later is from
+	 * GPU's view and aper_base is from CPU's view.
+	 */
 	resource_size_t		aper_size;
 	resource_size_t		aper_base;
 	/* for some chips with <= 32MB we need to lie
 	 * about vram size near mc fb location */
 	u64			mc_vram_size;
 	u64			visible_vram_size;
+	/* AGP aperture start and end in MC address space
+	 * Driver find a hole in the MC address space
+	 * to place AGP by setting MC_VM_AGP_BOT/TOP registers
+	 * Under VMID0, logical address == MC address. AGP
+	 * aperture maps to physical bus or IOVA addressed.
+	 * AGP aperture is used to simulate FB in ZFB case.
+	 * AGP aperture is also used for page table in system
+	 * memory (mainly for APU).
+	 *
+	 */
 	u64			agp_size;
 	u64			agp_start;
 	u64			agp_end;
+	/* GART aperture start and end in MC address space
+	 * Driver find a hole in the MC address space
+	 * to place GART by setting VM_CONTEXT0_PAGE_TABLE_START/END_ADDR
+	 * registers
+	 * Under VMID0, logical address inside GART aperture will
+	 * be translated through gpuvm gart page table to access
+	 * paged system memory
+	 */
 	u64			gart_size;
 	u64			gart_start;
 	u64			gart_end;
+	/* Frame buffer aperture of this GPU device. Different from
+	 * fb_start (see below), this only covers the local GPU device.
+	 * Driver get fb_start from MC_VM_FB_LOCATION_BASE (set by vbios)
+	 * and calculate vram_start of this local device by adding an
+	 * offset inside the XGMI hive.
+	 * Under VMID0, logical address == MC address
+	 */
 	u64			vram_start;
 	u64			vram_end;
 	/* FB region , it's same as local vram region in single GPU, in XGMI

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

@@ -311,7 +311,7 @@ void amdgpu_irq_fini(struct amdgpu_device *adev)
 		drm_irq_uninstall(adev->ddev);
 		adev->irq.installed = false;
 		if (adev->irq.msi_enabled)
-			pci_disable_msi(adev->pdev);
+			pci_free_irq_vectors(adev->pdev);
 		if (!amdgpu_device_has_dc_support(adev))
 			flush_work(&adev->hotplug_work);
 	}

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

@@ -220,22 +220,28 @@ static struct ras_manager *amdgpu_ras_find_obj(struct amdgpu_device *adev,
  * As their names indicate, inject operation will write the
  * value to the address.
  *
- * Second member: struct ras_debug_if::op.
+ * The second member: struct ras_debug_if::op.
  * It has three kinds of operations.
  *
  * - 0: disable RAS on the block. Take ::head as its data.
  * - 1: enable RAS on the block. Take ::head as its data.
  * - 2: inject errors on the block. Take ::inject as its data.
  *
  * How to use the interface?
- * programs:
- * copy the struct ras_debug_if in your codes and initialize it.
- * write the struct to the control node.
+ *
+ * Programs
+ *
+ * Copy the struct ras_debug_if in your codes and initialize it.
+ * Write the struct to the control node.
+ *
+ * Shells
  *
  * .. code-block:: bash
  *
  *	echo op block [error [sub_block address value]] > .../ras/ras_ctrl
  *
+ * Parameters:
+ *
  * op: disable, enable, inject
  *	disable: only block is needed
  *	enable: block and error are needed
@@ -265,8 +271,10 @@ static struct ras_manager *amdgpu_ras_find_obj(struct amdgpu_device *adev,
  * /sys/class/drm/card[0/1/2...]/device/ras/[gfx/sdma/...]_err_count
  *
  * .. note::
- *	Operation is only allowed on blocks which are supported.
+ *	Operations are only allowed on blocks which are supported.
  *	Please check ras mask at /sys/module/amdgpu/parameters/ras_mask
+ *	to see which blocks support RAS on a particular asic.
+ *
  */
 static ssize_t amdgpu_ras_debugfs_ctrl_write(struct file *f, const char __user *buf,
 		size_t size, loff_t *pos)
@@ -322,7 +330,7 @@ static ssize_t amdgpu_ras_debugfs_ctrl_write(struct file *f, const char __user *
  * DOC: AMDGPU RAS debugfs EEPROM table reset interface
  *
  * Some boards contain an EEPROM which is used to persistently store a list of
- * bad pages containing ECC errors detected in vram.  This interface provides
+ * bad pages which experiences ECC errors in vram.  This interface provides
  * a way to reset the EEPROM, e.g., after testing error injection.
  *
  * Usage:
@@ -362,7 +370,7 @@ static const struct file_operations amdgpu_ras_debugfs_eeprom_ops = {
 /**
  * DOC: AMDGPU RAS sysfs Error Count Interface
  *
- * It allows user to read the error count for each IP block on the gpu through
+ * It allows the user to read the error count for each IP block on the gpu through
  * /sys/class/drm/card[0/1/2...]/device/ras/[gfx/sdma/...]_err_count
  *
  * It outputs the multiple lines which report the uncorrected (ue) and corrected
@@ -1027,6 +1035,24 @@ static int amdgpu_ras_sysfs_remove_all(struct amdgpu_device *adev)
 }
 /* sysfs end */
 
+/**
+ * DOC: AMDGPU RAS Reboot Behavior for Unrecoverable Errors
+ *
+ * Normally when there is an uncorrectable error, the driver will reset
+ * the GPU to recover.  However, in the event of an unrecoverable error,
+ * the driver provides an interface to reboot the system automatically
+ * in that event.
+ *
+ * The following file in debugfs provides that interface:
+ * /sys/kernel/debug/dri/[0/1/2...]/ras/auto_reboot
+ *
+ * Usage:
+ *
+ * .. code-block:: bash
+ *
+ *	echo true > .../ras/auto_reboot
+ *
+ */
 /* debugfs begin */
 static void amdgpu_ras_debugfs_create_ctrl_node(struct amdgpu_device *adev)
 {

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

@@ -138,6 +138,7 @@ static void amdgpu_do_test_moves(struct amdgpu_device *adev)
 		}
 
 		dma_fence_put(fence);
+		fence = NULL;
 
 		r = amdgpu_bo_kmap(vram_obj, &vram_map);
 		if (r) {
@@ -183,6 +184,7 @@ static void amdgpu_do_test_moves(struct amdgpu_device *adev)
 		}
 
 		dma_fence_put(fence);
+		fence = NULL;
 
 		r = amdgpu_bo_kmap(gtt_obj[i], &gtt_map);
 		if (r) {

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

@@ -1906,9 +1906,6 @@ void amdgpu_ttm_late_init(struct amdgpu_device *adev)
 	void *stolen_vga_buf;
 	/* return the VGA stolen memory (if any) back to VRAM */
 	amdgpu_bo_free_kernel(&adev->stolen_vga_memory, NULL, &stolen_vga_buf);
-
-	/* return the IP Discovery TMR memory back to VRAM */
-	amdgpu_bo_free_kernel(&adev->discovery_memory, NULL, NULL);
 }
 
 /**
@@ -1921,7 +1918,10 @@ void amdgpu_ttm_fini(struct amdgpu_device *adev)
 
 	amdgpu_ttm_debugfs_fini(adev);
 	amdgpu_ttm_training_reserve_vram_fini(adev);
+	/* return the IP Discovery TMR memory back to VRAM */
+	amdgpu_bo_free_kernel(&adev->discovery_memory, NULL, NULL);
 	amdgpu_ttm_fw_reserve_vram_fini(adev);
+
 	if (adev->mman.aper_base_kaddr)
 		iounmap(adev->mman.aper_base_kaddr);
 	adev->mman.aper_base_kaddr = NULL;

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

@@ -1418,6 +1418,9 @@ static int amdgpu_vm_update_ptes(struct amdgpu_vm_update_params *params,
 		uint64_t incr, entry_end, pe_start;
 		struct amdgpu_bo *pt;
 
+		/* make sure that the page tables covering the address range are
+		 * actually allocated
+		 */
 		r = amdgpu_vm_alloc_pts(params->adev, params->vm, &cursor,
 					params->direct);
 		if (r)
@@ -1491,7 +1494,12 @@ static int amdgpu_vm_update_ptes(struct amdgpu_vm_update_params *params,
 		} while (frag_start < entry_end);
 
 		if (amdgpu_vm_pt_descendant(adev, &cursor)) {
-			/* Free all child entries */
+			/* Free all child entries.
+			 * Update the tables with the flags and addresses and free up subsequent
+			 * tables in the case of huge pages or freed up areas.
+			 * This is the maximum you can free, because all other page tables are not
+			 * completely covered by the range and so potentially still in use.
+			 */
 			while (cursor.pfn < frag_start) {
 				amdgpu_vm_free_pts(adev, params->vm, &cursor);
 				amdgpu_vm_pt_next(adev, &cursor);