Merge drm/drm-next into drm-misc-next-fixes

Alex needs v5.6 into drm-misc-next-fixes to merge a fix for a regression in
the scatterlist processing in PRIME.

Signed-off-by: Maxime Ripard <maxime@cerno.tech>

1 files changed, 56 insertions, 11 deletions

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c
index 930633a0ed64..3c32a94d2424 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c
@@ -281,6 +281,11 @@ static ssize_t amdgpu_ras_debugfs_ctrl_write(struct file *f, const char __user *
 	struct ras_debug_if data;
 	int ret = 0;
 
+	if (amdgpu_ras_intr_triggered()) {
+		DRM_WARN("RAS WARN: error injection currently inaccessible\n");
+		return size;
+	}
+
 	ret = amdgpu_ras_debugfs_ctrl_parse_data(f, buf, size, pos, &data);
 	if (ret)
 		return -EINVAL;
@@ -394,6 +399,10 @@ static ssize_t amdgpu_ras_sysfs_read(struct device *dev,
 		.head = obj->head,
 	};
 
+	if (amdgpu_ras_intr_triggered())
+		return snprintf(buf, PAGE_SIZE,
+				"Query currently inaccessible\n");
+
 	if (amdgpu_ras_error_query(obj->adev, &info))
 		return -EINVAL;
 
@@ -721,6 +730,9 @@ int amdgpu_ras_error_query(struct amdgpu_device *adev,
 		if (adev->nbio.funcs->query_ras_error_count)
 			adev->nbio.funcs->query_ras_error_count(adev, &err_data);
 		break;
+	case AMDGPU_RAS_BLOCK__XGMI_WAFL:
+		amdgpu_xgmi_query_ras_error_count(adev, &err_data);
+		break;
 	default:
 		break;
 	}
@@ -1110,6 +1122,32 @@ void amdgpu_ras_debugfs_create(struct amdgpu_device *adev,
 				       &amdgpu_ras_debugfs_ops);
 }
 
+void amdgpu_ras_debugfs_create_all(struct amdgpu_device *adev)
+{
+	struct amdgpu_ras *con = amdgpu_ras_get_context(adev);
+	struct ras_manager *obj;
+	struct ras_fs_if fs_info;
+
+	/*
+	 * it won't be called in resume path, no need to check
+	 * suspend and gpu reset status
+	 */
+	if (!con)
+		return;
+
+	amdgpu_ras_debugfs_create_ctrl_node(adev);
+
+	list_for_each_entry(obj, &con->head, node) {
+		if (amdgpu_ras_is_supported(adev, obj->head.block) &&
+			(obj->attr_inuse == 1)) {
+			sprintf(fs_info.debugfs_name, "%s_err_inject",
+					ras_block_str(obj->head.block));
+			fs_info.head = obj->head;
+			amdgpu_ras_debugfs_create(adev, &fs_info);
+		}
+	}
+}
+
 void amdgpu_ras_debugfs_remove(struct amdgpu_device *adev,
 		struct ras_common_if *head)
 {
@@ -1142,7 +1180,6 @@ static void amdgpu_ras_debugfs_remove_all(struct amdgpu_device *adev)
 static int amdgpu_ras_fs_init(struct amdgpu_device *adev)
 {
 	amdgpu_ras_sysfs_create_feature_node(adev);
-	amdgpu_ras_debugfs_create_ctrl_node(adev);
 
 	return 0;
 }
@@ -1734,18 +1771,30 @@ static void amdgpu_ras_check_supported(struct amdgpu_device *adev,
 	*hw_supported = 0;
 	*supported = 0;
 
-	if (amdgpu_sriov_vf(adev) ||
+	if (amdgpu_sriov_vf(adev) || !adev->is_atom_fw ||
 	    (adev->asic_type != CHIP_VEGA20 &&
 	     adev->asic_type != CHIP_ARCTURUS))
 		return;
 
-	if (adev->is_atom_fw &&
-			(amdgpu_atomfirmware_mem_ecc_supported(adev) ||
-			 amdgpu_atomfirmware_sram_ecc_supported(adev)))
-		*hw_supported = AMDGPU_RAS_BLOCK_MASK;
+	if (amdgpu_atomfirmware_mem_ecc_supported(adev)) {
+		DRM_INFO("HBM ECC is active.\n");
+		*hw_supported |= (1 << AMDGPU_RAS_BLOCK__UMC |
+				1 << AMDGPU_RAS_BLOCK__DF);
+	} else
+		DRM_INFO("HBM ECC is not presented.\n");
+
+	if (amdgpu_atomfirmware_sram_ecc_supported(adev)) {
+		DRM_INFO("SRAM ECC is active.\n");
+		*hw_supported |= ~(1 << AMDGPU_RAS_BLOCK__UMC |
+				1 << AMDGPU_RAS_BLOCK__DF);
+	} else
+		DRM_INFO("SRAM ECC is not presented.\n");
+
+	/* hw_supported needs to be aligned with RAS block mask. */
+	*hw_supported &= AMDGPU_RAS_BLOCK_MASK;
 
 	*supported = amdgpu_ras_enable == 0 ?
-				0 : *hw_supported & amdgpu_ras_mask;
+			0 : *hw_supported & amdgpu_ras_mask;
 }
 
 int amdgpu_ras_init(struct amdgpu_device *adev)
@@ -1846,8 +1895,6 @@ int amdgpu_ras_late_init(struct amdgpu_device *adev,
 			goto interrupt;
 	}
 
-	amdgpu_ras_debugfs_create(adev, fs_info);
-
 	r = amdgpu_ras_sysfs_create(adev, fs_info);
 	if (r)
 		goto sysfs;
@@ -1856,7 +1903,6 @@ int amdgpu_ras_late_init(struct amdgpu_device *adev,
 cleanup:
 	amdgpu_ras_sysfs_remove(adev, ras_block);
 sysfs:
-	amdgpu_ras_debugfs_remove(adev, ras_block);
 	if (ih_info->cb)
 		amdgpu_ras_interrupt_remove_handler(adev, ih_info);
 interrupt:
@@ -1873,7 +1919,6 @@ void amdgpu_ras_late_fini(struct amdgpu_device *adev,
 		return;
 
 	amdgpu_ras_sysfs_remove(adev, ras_block);
-	amdgpu_ras_debugfs_remove(adev, ras_block);
 	if (ih_info->cb)
                 amdgpu_ras_interrupt_remove_handler(adev, ih_info);
 	amdgpu_ras_feature_enable(adev, ras_block, 0);