drm/amdgpu: call ras_debugfs_create_all in debugfs_init

and remove each ras IP's own debugfs creation

this is required to fix ras when the driver does not use the drm load
and unload callbacks due to ordering issues with the drm device node.

Signed-off-by: Tao Zhou <[email protected]>
Signed-off-by: Stanley.Yang <[email protected]>
Reviewed-by: Alex Deucher <[email protected]>
Signed-off-by: Alex Deucher <[email protected]>

Author: Tao Zhou

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

@@ -1294,6 +1294,7 @@ DEFINE_SIMPLE_ATTRIBUTE(fops_ib_preempt, NULL,
 DEFINE_SIMPLE_ATTRIBUTE(fops_sclk_set, NULL,
 			amdgpu_debugfs_sclk_set, "%llu\n");
 
+extern void amdgpu_ras_debugfs_create_all(struct amdgpu_device *adev);
 int amdgpu_debugfs_init(struct amdgpu_device *adev)
 {
 	int r, i;
@@ -1366,6 +1367,8 @@ int amdgpu_debugfs_init(struct amdgpu_device *adev)
 		}
 	}
 
+	amdgpu_ras_debugfs_create_all(adev);
+
 	return amdgpu_debugfs_add_files(adev, amdgpu_debugfs_list,
 					ARRAY_SIZE(amdgpu_debugfs_list));
 }

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

@@ -573,7 +573,6 @@ int amdgpu_gfx_ras_late_init(struct amdgpu_device *adev)
 	int r;
 	struct ras_fs_if fs_info = {
 		.sysfs_name = "gfx_err_count",
-		.debugfs_name = "gfx_err_inject",
 	};
 	struct ras_ih_if ih_info = {
 		.cb = amdgpu_gfx_process_ras_data_cb,

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

@@ -32,7 +32,6 @@ int amdgpu_mmhub_ras_late_init(struct amdgpu_device *adev)
 	};
 	struct ras_fs_if fs_info = {
 		.sysfs_name = "mmhub_err_count",
-		.debugfs_name = "mmhub_err_inject",
 	};
 
 	if (!adev->mmhub.ras_if) {

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

@@ -30,7 +30,6 @@ int amdgpu_nbio_ras_late_init(struct amdgpu_device *adev)
 	};
 	struct ras_fs_if fs_info = {
 		.sysfs_name = "pcie_bif_err_count",
-		.debugfs_name = "pcie_bif_err_inject",
 	};
 
 	if (!adev->nbio.ras_if) {

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

@@ -1174,7 +1174,6 @@ 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;
 }
@@ -1878,8 +1877,6 @@ 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;
@@ -1888,7 +1885,6 @@ 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:
@@ -1905,7 +1901,6 @@ 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);

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

@@ -93,7 +93,6 @@ int amdgpu_sdma_ras_late_init(struct amdgpu_device *adev,
 	struct ras_ih_if *ih_info = (struct ras_ih_if *)ras_ih_info;
 	struct ras_fs_if fs_info = {
 		.sysfs_name = "sdma_err_count",
-		.debugfs_name = "sdma_err_inject",
 	};
 
 	if (!ih_info)

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

@@ -28,7 +28,6 @@ int amdgpu_umc_ras_late_init(struct amdgpu_device *adev)
 	int r;
 	struct ras_fs_if fs_info = {
 		.sysfs_name = "umc_err_count",
-		.debugfs_name = "umc_err_inject",
 	};
 	struct ras_ih_if ih_info = {
 		.cb = amdgpu_umc_process_ras_data_cb,

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

@@ -598,7 +598,6 @@ int amdgpu_xgmi_ras_late_init(struct amdgpu_device *adev)
 	};
 	struct ras_fs_if fs_info = {
 		.sysfs_name = "xgmi_wafl_err_count",
-		.debugfs_name = "xgmi_wafl_err_inject",
 	};
 
 	if (!adev->gmc.xgmi.supported ||