drm/amdgpu: correct the calculation of RAS bad page

After the introduction of NPS RAS, one bad page record on eeprom may be
related to 1 or 16 bad pages, so the bad page record and bad page are
two different concepts, define a new variable to store bad page number.

Signed-off-by: Tao Zhou <[e-mail]>
Reviewed-by: Hawking Zhang <[e-mail]>
Signed-off-by: Alex Deucher <[e-mail]>

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

···
 	mutex_lock(&con->recovery_lock);
 	control = &con->eeprom_control;
 	data = con->eh_data;
-	bad_page_num = control->ras_num_recs;
-	/* one record on eeprom stands for all pages in one memory row
-	 * in this mode
-	 */
-	if (control->rec_type == AMDGPU_RAS_EEPROM_REC_MCA)
-		bad_page_num = control->ras_num_recs * adev->umc.retire_unit;
-
+	bad_page_num = control->ras_num_bad_pages;
 	save_count = data->count - bad_page_num;
 	mutex_unlock(&con->recovery_lock);
 
···
 			return ret;
 
 		amdgpu_dpm_send_hbm_bad_pages_num(
-			adev, control->ras_num_recs);
+			adev, control->ras_num_bad_pages);
 
 		if (con->update_channel_flag == true) {
 			amdgpu_dpm_send_hbm_bad_channel_flag(

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

···
 		res = __write_table_ras_info(control);
 
 	control->ras_num_recs = 0;
+	control->ras_num_bad_pages = 0;
 	control->ras_fri = 0;
 
-	amdgpu_dpm_send_hbm_bad_pages_num(adev, control->ras_num_recs);
+	amdgpu_dpm_send_hbm_bad_pages_num(adev, control->ras_num_bad_pages);
 
 	control->bad_channel_bitmap = 0;
 	amdgpu_dpm_send_hbm_bad_channel_flag(adev, control->bad_channel_bitmap);
···
 	if (con->eeprom_control.tbl_hdr.header == RAS_TABLE_HDR_BAD) {
 		if (amdgpu_bad_page_threshold == -1) {
 			dev_warn(adev->dev, "RAS records:%d exceed threshold:%d",
-				con->eeprom_control.ras_num_recs, con->bad_page_cnt_threshold);
+				con->eeprom_control.ras_num_bad_pages, con->bad_page_cnt_threshold);
 			dev_warn(adev->dev,
 				"But GPU can be operated due to bad_page_threshold = -1.\n");
 			return false;
···
 			       const u32 num)
 {
 	struct amdgpu_ras *con = amdgpu_ras_get_context(to_amdgpu_device(control));
+	struct amdgpu_device *adev = to_amdgpu_device(control);
 	u32 a, b, i;
 	u8 *buf, *pp;
 	int res;
···
 	control->ras_num_recs = 1 + (control->ras_max_record_count + b
 				     - control->ras_fri)
 		% control->ras_max_record_count;
+
+	if (control->rec_type == AMDGPU_RAS_EEPROM_REC_PA)
+		control->ras_num_bad_pages = control->ras_num_recs;
+	else
+		control->ras_num_bad_pages =
+			control->ras_num_recs * adev->umc.retire_unit;
 Out:
 	kfree(buf);
 	return res;
···
 	/* Modify the header if it exceeds.
 	 */
 	if (amdgpu_bad_page_threshold != 0 &&
-	    control->ras_num_recs >= ras->bad_page_cnt_threshold) {
+	    control->ras_num_bad_pages >= ras->bad_page_cnt_threshold) {
 		dev_warn(adev->dev,
 			"Saved bad pages %d reaches threshold value %d\n",
-			control->ras_num_recs, ras->bad_page_cnt_threshold);
+			control->ras_num_bad_pages, ras->bad_page_cnt_threshold);
 		control->tbl_hdr.header = RAS_TABLE_HDR_BAD;
 		if (control->tbl_hdr.version == RAS_TABLE_VER_V2_1) {
 			control->tbl_rai.rma_status = GPU_RETIRED__ECC_REACH_THRESHOLD;
···
 	 */
 	if (amdgpu_bad_page_threshold != 0 &&
 	    control->tbl_hdr.version == RAS_TABLE_VER_V2_1 &&
-	    control->ras_num_recs < ras->bad_page_cnt_threshold)
+	    control->ras_num_bad_pages < ras->bad_page_cnt_threshold)
 		control->tbl_rai.health_percent = ((ras->bad_page_cnt_threshold -
-						   control->ras_num_recs) * 100) /
+						   control->ras_num_bad_pages) * 100) /
 						   ras->bad_page_cnt_threshold;
 
 	/* Recalc the checksum.
···
 	if (!__get_eeprom_i2c_addr(adev, control))
 		return -EINVAL;
 
+	if (control->rec_type == AMDGPU_RAS_EEPROM_REC_PA)
+		control->ras_num_bad_pages = control->ras_num_recs;
+	else
+		control->ras_num_bad_pages =
+			control->ras_num_recs * adev->umc.retire_unit;
+
 	if (hdr->header == RAS_TABLE_HDR_VAL) {
 		DRM_DEBUG_DRIVER("Found existing EEPROM table with %d records",
-				 control->ras_num_recs);
+				 control->ras_num_bad_pages);
 
 		if (hdr->version == RAS_TABLE_VER_V2_1) {
 			res = __read_table_ras_info(control);
···
 
 		/* Warn if we are at 90% of the threshold or above
 		 */
-		if (10 * control->ras_num_recs >= 9 * ras->bad_page_cnt_threshold)
+		if (10 * control->ras_num_bad_pages >= 9 * ras->bad_page_cnt_threshold)
 			dev_warn(adev->dev, "RAS records:%u exceeds 90%% of threshold:%d",
-					control->ras_num_recs,
+					control->ras_num_bad_pages,
 					ras->bad_page_cnt_threshold);
 	} else if (hdr->header == RAS_TABLE_HDR_BAD &&
 		   amdgpu_bad_page_threshold != 0) {
···
 				  res);
 			return -EINVAL;
 		}
-		if (ras->bad_page_cnt_threshold > control->ras_num_recs) {
+		if (ras->bad_page_cnt_threshold > control->ras_num_bad_pages) {
 			/* This means that, the threshold was increased since
 			 * the last time the system was booted, and now,
 			 * ras->bad_page_cnt_threshold - control->num_recs > 0,
···
 			dev_info(adev->dev,
 				 "records:%d threshold:%d, resetting "
 				 "RAS table header signature",
-				 control->ras_num_recs,
+				 control->ras_num_bad_pages,
 				 ras->bad_page_cnt_threshold);
 			res = amdgpu_ras_eeprom_correct_header_tag(control,
 								   RAS_TABLE_HDR_VAL);
 		} else {
 			dev_err(adev->dev, "RAS records:%d exceed threshold:%d",
-				control->ras_num_recs, ras->bad_page_cnt_threshold);
+				control->ras_num_bad_pages, ras->bad_page_cnt_threshold);
 			if (amdgpu_bad_page_threshold == -1) {
 				dev_warn(adev->dev, "GPU will be initialized due to bad_page_threshold = -1.");
 				res = 0;
···
 				dev_err(adev->dev,
 					"RAS records:%d exceed threshold:%d, "
 					"GPU will not be initialized. Replace this GPU or increase the threshold",
-					control->ras_num_recs, ras->bad_page_cnt_threshold);
+					control->ras_num_bad_pages, ras->bad_page_cnt_threshold);
 			}
 		}
 	} else {

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

···
 	 */
 	u32 ras_num_recs;
 
+	/* the bad page number is ras_num_recs or
+	 * ras_num_recs * umc.retire_unit
+	 */
+	u32 ras_num_bad_pages;
+
 	/* First record index to read, 0-based.
 	 * Range is [0, num_recs-1]. This is
 	 * an absolute index, starting right after

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

···
 						err_data->err_addr_cnt, false);
 			amdgpu_ras_save_bad_pages(adev, &err_count);
 
-			amdgpu_dpm_send_hbm_bad_pages_num(adev, con->eeprom_control.ras_num_recs);
+			amdgpu_dpm_send_hbm_bad_pages_num(adev,
+					con->eeprom_control.ras_num_bad_pages);
 
 			if (con->update_channel_flag == true) {
 				amdgpu_dpm_send_hbm_bad_channel_flag(adev, con->eeprom_control.bad_channel_bitmap);