drm/amdgpu: Modify umc block to fit for the unified ras block data and ops

1.Modify umc block to fit for the unified ras block data and ops.
2.Change amdgpu_umc_ras_funcs to amdgpu_umc_ras, and the corresponding variable name remove _funcs suffix.
3.Remove the const flag of umc ras variable so that umc ras block can be able to be inserted into amdgpu device ras block link list.
4.Invoke amdgpu_ras_register_ras_block function to register umc ras block into amdgpu device ras block link list.
5.Remove the redundant code about umc in amdgpu_ras.c after using the unified ras block.
6.Fill unified ras block .name .block .ras_late_init and .ras_fini for all of umc versions. If .ras_late_init and .ras_fini had been defined by the selected umc version, the defined functions will take effect; if not defined, default fill them with amdgpu_umc_ras_late_init and amdgpu_umc_ras_fini.
Signed-off-by: yipechai <YiPeng.Chai@amd.com>
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Reviewed-by: John Clements <john.clements@amd.com>
Reviewed-by: Tao Zhou <tao.zhou1@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

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

@@ -440,9 +440,8 @@ int amdgpu_gmc_ras_late_init(struct amdgpu_device *adev)
 {
 	int r;
 
-	if (adev->umc.ras_funcs &&
-	    adev->umc.ras_funcs->ras_late_init) {
-		r = adev->umc.ras_funcs->ras_late_init(adev);
+	if (adev->umc.ras && adev->umc.ras->ras_block.ras_late_init) {
+		r = adev->umc.ras->ras_block.ras_late_init(adev, NULL);
 		if (r)
 			return r;
 	}
@@ -496,9 +495,8 @@ int amdgpu_gmc_ras_late_init(struct amdgpu_device *adev)
 
 void amdgpu_gmc_ras_fini(struct amdgpu_device *adev)
 {
-	if (adev->umc.ras_funcs &&
-	    adev->umc.ras_funcs->ras_fini)
-		adev->umc.ras_funcs->ras_fini(adev);
+	if (adev->umc.ras && adev->umc.ras->ras_block.ras_fini)
+		adev->umc.ras->ras_block.ras_fini(adev);
 
 	if (adev->mmhub.ras && adev->mmhub.ras->ras_block.ras_fini)
 		adev->mmhub.ras->ras_block.ras_fini(adev);

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

@@ -939,24 +939,24 @@ static void amdgpu_ras_get_ecc_info(struct amdgpu_device *adev, struct ras_err_d
 	 */
 	ret = amdgpu_dpm_get_ecc_info(adev, (void *)&(ras->umc_ecc));
 	if (ret == -EOPNOTSUPP) {
-		if (adev->umc.ras_funcs &&
-			adev->umc.ras_funcs->query_ras_error_count)
-			adev->umc.ras_funcs->query_ras_error_count(adev, err_data);
+		if (adev->umc.ras && adev->umc.ras->ras_block.hw_ops &&
+			adev->umc.ras->ras_block.hw_ops->query_ras_error_count)
+			adev->umc.ras->ras_block.hw_ops->query_ras_error_count(adev, err_data);
 
 		/* umc query_ras_error_address is also responsible for clearing
 		 * error status
 		 */
-		if (adev->umc.ras_funcs &&
-		    adev->umc.ras_funcs->query_ras_error_address)
-			adev->umc.ras_funcs->query_ras_error_address(adev, err_data);
+		if (adev->umc.ras && adev->umc.ras->ras_block.hw_ops &&
+		    adev->umc.ras->ras_block.hw_ops->query_ras_error_address)
+			adev->umc.ras->ras_block.hw_ops->query_ras_error_address(adev, err_data);
 	} else if (!ret) {
-		if (adev->umc.ras_funcs &&
-			adev->umc.ras_funcs->ecc_info_query_ras_error_count)
-			adev->umc.ras_funcs->ecc_info_query_ras_error_count(adev, err_data);
+		if (adev->umc.ras &&
+			adev->umc.ras->ecc_info_query_ras_error_count)
+			adev->umc.ras->ecc_info_query_ras_error_count(adev, err_data);
 
-		if (adev->umc.ras_funcs &&
-			adev->umc.ras_funcs->ecc_info_query_ras_error_address)
-			adev->umc.ras_funcs->ecc_info_query_ras_error_address(adev, err_data);
+		if (adev->umc.ras &&
+			adev->umc.ras->ecc_info_query_ras_error_address)
+			adev->umc.ras->ecc_info_query_ras_error_address(adev, err_data);
 	}
 }
 
@@ -2412,12 +2412,12 @@ int amdgpu_ras_init(struct amdgpu_device *adev)
 	}
 	else if (adev->df.funcs &&
 	    adev->df.funcs->query_ras_poison_mode &&
-	    adev->umc.ras_funcs &&
-	    adev->umc.ras_funcs->query_ras_poison_mode) {
+	    adev->umc.ras &&
+	    adev->umc.ras->query_ras_poison_mode) {
 		df_poison =
 			adev->df.funcs->query_ras_poison_mode(adev);
 		umc_poison =
-			adev->umc.ras_funcs->query_ras_poison_mode(adev);
+			adev->umc.ras->query_ras_poison_mode(adev);
 		/* Only poison is set in both DF and UMC, we can support it */
 		if (df_poison && umc_poison)
 			con->poison_supported = true;

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

@@ -35,12 +35,12 @@ static int amdgpu_umc_do_page_retirement(struct amdgpu_device *adev,
 	kgd2kfd_set_sram_ecc_flag(adev->kfd.dev);
 	ret = amdgpu_dpm_get_ecc_info(adev, (void *)&(con->umc_ecc));
 	if (ret == -EOPNOTSUPP) {
-		if (adev->umc.ras_funcs &&
-		    adev->umc.ras_funcs->query_ras_error_count)
-		    adev->umc.ras_funcs->query_ras_error_count(adev, ras_error_status);
+		if (adev->umc.ras && adev->umc.ras->ras_block.hw_ops &&
+		    adev->umc.ras->ras_block.hw_ops->query_ras_error_count)
+		    adev->umc.ras->ras_block.hw_ops->query_ras_error_count(adev, ras_error_status);
 
-		if (adev->umc.ras_funcs &&
-		    adev->umc.ras_funcs->query_ras_error_address &&
+		if (adev->umc.ras && adev->umc.ras->ras_block.hw_ops &&
+		    adev->umc.ras->ras_block.hw_ops->query_ras_error_address &&
 		    adev->umc.max_ras_err_cnt_per_query) {
 			err_data->err_addr =
 				kcalloc(adev->umc.max_ras_err_cnt_per_query,
@@ -56,15 +56,15 @@ static int amdgpu_umc_do_page_retirement(struct amdgpu_device *adev,
 			/* umc query_ras_error_address is also responsible for clearing
 			 * error status
 			 */
-			adev->umc.ras_funcs->query_ras_error_address(adev, ras_error_status);
+			adev->umc.ras->ras_block.hw_ops->query_ras_error_address(adev, ras_error_status);
 		}
 	} else if (!ret) {
-		if (adev->umc.ras_funcs &&
-		    adev->umc.ras_funcs->ecc_info_query_ras_error_count)
-		    adev->umc.ras_funcs->ecc_info_query_ras_error_count(adev, ras_error_status);
+		if (adev->umc.ras &&
+		    adev->umc.ras->ecc_info_query_ras_error_count)
+		    adev->umc.ras->ecc_info_query_ras_error_count(adev, ras_error_status);
 
-		if (adev->umc.ras_funcs &&
-		    adev->umc.ras_funcs->ecc_info_query_ras_error_address &&
+		if (adev->umc.ras &&
+		    adev->umc.ras->ecc_info_query_ras_error_address &&
 		    adev->umc.max_ras_err_cnt_per_query) {
 			err_data->err_addr =
 				kcalloc(adev->umc.max_ras_err_cnt_per_query,
@@ -80,7 +80,7 @@ static int amdgpu_umc_do_page_retirement(struct amdgpu_device *adev,
 			/* umc query_ras_error_address is also responsible for clearing
 			 * error status
 			 */
-			adev->umc.ras_funcs->ecc_info_query_ras_error_address(adev, ras_error_status);
+			adev->umc.ras->ecc_info_query_ras_error_address(adev, ras_error_status);
 		}
 	}
 
@@ -136,7 +136,7 @@ static int amdgpu_umc_process_ras_data_cb(struct amdgpu_device *adev,
 	return amdgpu_umc_do_page_retirement(adev, ras_error_status, entry, true);
 }
 
-int amdgpu_umc_ras_late_init(struct amdgpu_device *adev)
+int amdgpu_umc_ras_late_init(struct amdgpu_device *adev, void *ras_info)
 {
 	int r;
 	struct ras_fs_if fs_info = {
@@ -172,9 +172,9 @@ int amdgpu_umc_ras_late_init(struct amdgpu_device *adev)
 	}
 
 	/* ras init of specific umc version */
-	if (adev->umc.ras_funcs &&
-	    adev->umc.ras_funcs->err_cnt_init)
-		adev->umc.ras_funcs->err_cnt_init(adev);
+	if (adev->umc.ras &&
+	    adev->umc.ras->err_cnt_init)
+		adev->umc.ras->err_cnt_init(adev);
 
 	return 0;
 

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

@@ -20,6 +20,7 @@
  */
 #ifndef __AMDGPU_UMC_H__
 #define __AMDGPU_UMC_H__
+#include "amdgpu_ras.h"
 
 /*
  * (addr / 256) * 4096, the higher 26 bits in ErrorAddr
@@ -40,14 +41,9 @@
 #define LOOP_UMC_CH_INST(ch_inst) for ((ch_inst) = 0; (ch_inst) < adev->umc.channel_inst_num; (ch_inst)++)
 #define LOOP_UMC_INST_AND_CH(umc_inst, ch_inst) LOOP_UMC_INST((umc_inst)) LOOP_UMC_CH_INST((ch_inst))
 
-struct amdgpu_umc_ras_funcs {
+struct amdgpu_umc_ras {
+	struct amdgpu_ras_block_object ras_block;
 	void (*err_cnt_init)(struct amdgpu_device *adev);
-	int (*ras_late_init)(struct amdgpu_device *adev);
-	void (*ras_fini)(struct amdgpu_device *adev);
-	void (*query_ras_error_count)(struct amdgpu_device *adev,
-				      void *ras_error_status);
-	void (*query_ras_error_address)(struct amdgpu_device *adev,
-					void *ras_error_status);
 	bool (*query_ras_poison_mode)(struct amdgpu_device *adev);
 	void (*ecc_info_query_ras_error_count)(struct amdgpu_device *adev,
 				      void *ras_error_status);
@@ -73,10 +69,10 @@ struct amdgpu_umc {
 	struct ras_common_if *ras_if;
 
 	const struct amdgpu_umc_funcs *funcs;
-	const struct amdgpu_umc_ras_funcs *ras_funcs;
+	struct amdgpu_umc_ras *ras;
 };
 
-int amdgpu_umc_ras_late_init(struct amdgpu_device *adev);
+int amdgpu_umc_ras_late_init(struct amdgpu_device *adev, void *ras_info);
 void amdgpu_umc_ras_fini(struct amdgpu_device *adev);
 int amdgpu_umc_poison_handler(struct amdgpu_device *adev,
 		void *ras_error_status,

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

@@ -664,11 +664,25 @@ static void gmc_v10_0_set_umc_funcs(struct amdgpu_device *adev)
 		adev->umc.umc_inst_num = UMC_V8_7_UMC_INSTANCE_NUM;
 		adev->umc.channel_offs = UMC_V8_7_PER_CHANNEL_OFFSET_SIENNA;
 		adev->umc.channel_idx_tbl = &umc_v8_7_channel_idx_tbl[0][0];
-		adev->umc.ras_funcs = &umc_v8_7_ras_funcs;
+		adev->umc.ras = &umc_v8_7_ras;
 		break;
 	default:
 		break;
 	}
+	if (adev->umc.ras) {
+		amdgpu_ras_register_ras_block(adev, &adev->umc.ras->ras_block);
+
+		strcpy(adev->umc.ras->ras_block.name,"umc");
+		adev->umc.ras->ras_block.block = AMDGPU_RAS_BLOCK__UMC;
+
+		/* If don't define special ras_late_init function, use default ras_late_init */
+		if (!adev->umc.ras->ras_block.ras_late_init)
+				adev->umc.ras->ras_block.ras_late_init = amdgpu_umc_ras_late_init;
+
+		/* If don't define special ras_fini function, use default ras_fini */
+		if (!adev->umc.ras->ras_block.ras_fini)
+				adev->umc.ras->ras_block.ras_fini = amdgpu_umc_ras_fini;
+	}
 }
 
 

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

@@ -1202,7 +1202,7 @@ static void gmc_v9_0_set_umc_funcs(struct amdgpu_device *adev)
 		adev->umc.umc_inst_num = UMC_V6_1_UMC_INSTANCE_NUM;
 		adev->umc.channel_offs = UMC_V6_1_PER_CHANNEL_OFFSET_VG20;
 		adev->umc.channel_idx_tbl = &umc_v6_1_channel_idx_tbl[0][0];
-		adev->umc.ras_funcs = &umc_v6_1_ras_funcs;
+		adev->umc.ras = &umc_v6_1_ras;
 		break;
 	case IP_VERSION(6, 1, 2):
 		adev->umc.max_ras_err_cnt_per_query = UMC_V6_1_TOTAL_CHANNEL_NUM;
@@ -1210,7 +1210,7 @@ static void gmc_v9_0_set_umc_funcs(struct amdgpu_device *adev)
 		adev->umc.umc_inst_num = UMC_V6_1_UMC_INSTANCE_NUM;
 		adev->umc.channel_offs = UMC_V6_1_PER_CHANNEL_OFFSET_ARCT;
 		adev->umc.channel_idx_tbl = &umc_v6_1_channel_idx_tbl[0][0];
-		adev->umc.ras_funcs = &umc_v6_1_ras_funcs;
+		adev->umc.ras = &umc_v6_1_ras;
 		break;
 	case IP_VERSION(6, 7, 0):
 		adev->umc.max_ras_err_cnt_per_query = UMC_V6_7_TOTAL_CHANNEL_NUM;
@@ -1218,7 +1218,7 @@ static void gmc_v9_0_set_umc_funcs(struct amdgpu_device *adev)
 		adev->umc.umc_inst_num = UMC_V6_7_UMC_INSTANCE_NUM;
 		adev->umc.channel_offs = UMC_V6_7_PER_CHANNEL_OFFSET;
 		if (!adev->gmc.xgmi.connected_to_cpu)
-			adev->umc.ras_funcs = &umc_v6_7_ras_funcs;
+			adev->umc.ras = &umc_v6_7_ras;
 		if (1 & adev->smuio.funcs->get_die_id(adev))
 			adev->umc.channel_idx_tbl = &umc_v6_7_channel_idx_tbl_first[0][0];
 		else
@@ -1227,6 +1227,21 @@ static void gmc_v9_0_set_umc_funcs(struct amdgpu_device *adev)
 	default:
 		break;
 	}
+
+	if (adev->umc.ras) {
+		amdgpu_ras_register_ras_block(adev, &adev->umc.ras->ras_block);
+
+		strcpy(adev->umc.ras->ras_block.name,"umc");
+		adev->umc.ras->ras_block.block = AMDGPU_RAS_BLOCK__UMC;
+
+		/* If don't define special ras_late_init function, use default ras_late_init */
+		if (!adev->umc.ras->ras_block.ras_late_init)
+				adev->umc.ras->ras_block.ras_late_init = amdgpu_umc_ras_late_init;
+
+		/* If don't define special ras_fini function, use default ras_fini */
+		if (!adev->umc.ras->ras_block.ras_fini)
+				adev->umc.ras->ras_block.ras_fini = amdgpu_umc_ras_fini;
+	}
 }
 
 static void gmc_v9_0_set_mmhub_funcs(struct amdgpu_device *adev)

drivers/gpu/drm/amd/amdgpu/umc_v6_1.c

@@ -465,10 +465,14 @@ static void umc_v6_1_err_cnt_init(struct amdgpu_device *adev)
 		umc_v6_1_enable_umc_index_mode(adev);
 }
 
-const struct amdgpu_umc_ras_funcs umc_v6_1_ras_funcs = {
-	.err_cnt_init = umc_v6_1_err_cnt_init,
-	.ras_late_init = amdgpu_umc_ras_late_init,
-	.ras_fini = amdgpu_umc_ras_fini,
+const struct amdgpu_ras_block_hw_ops umc_v6_1_ras_hw_ops = {
 	.query_ras_error_count = umc_v6_1_query_ras_error_count,
 	.query_ras_error_address = umc_v6_1_query_ras_error_address,
 };
+
+struct amdgpu_umc_ras umc_v6_1_ras = {
+	.ras_block = {
+		.hw_ops = &umc_v6_1_ras_hw_ops,
+	},
+	.err_cnt_init = umc_v6_1_err_cnt_init,
+};
\ No newline at end of file

drivers/gpu/drm/amd/amdgpu/umc_v6_1.h

@@ -45,7 +45,7 @@
 /* umc ce count initial value */
 #define UMC_V6_1_CE_CNT_INIT	(UMC_V6_1_CE_CNT_MAX - UMC_V6_1_CE_INT_THRESHOLD)
 
-extern const struct amdgpu_umc_ras_funcs umc_v6_1_ras_funcs;
+extern struct amdgpu_umc_ras umc_v6_1_ras;
 extern const uint32_t
 	umc_v6_1_channel_idx_tbl[UMC_V6_1_UMC_INSTANCE_NUM][UMC_V6_1_CHANNEL_INSTANCE_NUM];
 

drivers/gpu/drm/amd/amdgpu/umc_v6_7.c

@@ -480,11 +480,15 @@ static bool umc_v6_7_query_ras_poison_mode(struct amdgpu_device *adev)
 	return true;
 }
 
-const struct amdgpu_umc_ras_funcs umc_v6_7_ras_funcs = {
-	.ras_late_init = amdgpu_umc_ras_late_init,
-	.ras_fini = amdgpu_umc_ras_fini,
+const struct amdgpu_ras_block_hw_ops umc_v6_7_ras_hw_ops = {
 	.query_ras_error_count = umc_v6_7_query_ras_error_count,
 	.query_ras_error_address = umc_v6_7_query_ras_error_address,
+};
+
+struct amdgpu_umc_ras umc_v6_7_ras = {
+	.ras_block = {
+		.hw_ops = &umc_v6_7_ras_hw_ops,
+	},
 	.query_ras_poison_mode = umc_v6_7_query_ras_poison_mode,
 	.ecc_info_query_ras_error_count = umc_v6_7_ecc_info_query_ras_error_count,
 	.ecc_info_query_ras_error_address = umc_v6_7_ecc_info_query_ras_error_address,

drivers/gpu/drm/amd/amdgpu/umc_v6_7.h

@@ -43,7 +43,7 @@
 #define UMC_V6_7_TOTAL_CHANNEL_NUM	(UMC_V6_7_CHANNEL_INSTANCE_NUM * UMC_V6_7_UMC_INSTANCE_NUM)
 /* UMC regiser per channel offset */
 #define UMC_V6_7_PER_CHANNEL_OFFSET		0x400
-extern const struct amdgpu_umc_ras_funcs umc_v6_7_ras_funcs;
+extern struct amdgpu_umc_ras umc_v6_7_ras;
 extern const uint32_t
 	umc_v6_7_channel_idx_tbl_second[UMC_V6_7_UMC_INSTANCE_NUM][UMC_V6_7_CHANNEL_INSTANCE_NUM];
 extern const uint32_t

drivers/gpu/drm/amd/amdgpu/umc_v8_7.c

@@ -324,10 +324,14 @@ static void umc_v8_7_err_cnt_init(struct amdgpu_device *adev)
 	}
 }
 
-const struct amdgpu_umc_ras_funcs umc_v8_7_ras_funcs = {
-	.err_cnt_init = umc_v8_7_err_cnt_init,
-	.ras_late_init = amdgpu_umc_ras_late_init,
-	.ras_fini = amdgpu_umc_ras_fini,
+const struct amdgpu_ras_block_hw_ops umc_v8_7_ras_hw_ops = {
 	.query_ras_error_count = umc_v8_7_query_ras_error_count,
 	.query_ras_error_address = umc_v8_7_query_ras_error_address,
 };
+
+struct amdgpu_umc_ras umc_v8_7_ras = {
+	.ras_block = {
+		.hw_ops = &umc_v8_7_ras_hw_ops,
+	},
+	.err_cnt_init = umc_v8_7_err_cnt_init,
+};

drivers/gpu/drm/amd/amdgpu/umc_v8_7.h

@@ -44,7 +44,7 @@
 /* umc ce count initial value */
 #define UMC_V8_7_CE_CNT_INIT	(UMC_V8_7_CE_CNT_MAX - UMC_V8_7_CE_INT_THRESHOLD)
 
-extern const struct amdgpu_umc_ras_funcs umc_v8_7_ras_funcs;
+extern struct amdgpu_umc_ras umc_v8_7_ras;
 extern const uint32_t
 	umc_v8_7_channel_idx_tbl[UMC_V8_7_UMC_INSTANCE_NUM][UMC_V8_7_CHANNEL_INSTANCE_NUM];