diff mbox series

[v4,22/36] scsi: Have scsi-ml retry read_capacity_16 errors

Message ID 20221016195946.7613-23-michael.christie@oracle.com
State New
Headers show
Series Allow scsi_execute users to control retries | expand

Commit Message

Mike Christie Oct. 16, 2022, 7:59 p.m. UTC 
This has read_capacity_16 have scsi-ml retry errors instead of driving
them itself.

Signed-off-by: Mike Christie <michael.christie@oracle.com>
Reviewed-by: Martin Wilck <mwilck@suse.com>
---
 drivers/scsi/sd.c | 81 +++++++++++++++++++++++------------------------
 1 file changed, 40 insertions(+), 41 deletions(-)

Comments

Bart Van Assche Oct. 19, 2022, 10:15 p.m. UTC | #1 
On 10/16/22 12:59, Mike Christie wrote:
> +	struct scsi_failure failures[] = {
> +		{
> +			.sense = UNIT_ATTENTION,
> +			.asc = 0x29,
> +			.ascq = 0,
> +			/* Device reset might occur several times */
> +			.allowed = READ_CAPACITY_RETRIES_ON_RESET,
> +			.result = SAM_STAT_CHECK_CONDITION,
> +		},
> +		{
> +			.result = SCMD_FAILURE_ANY,
> +			.allowed = 3,
> +		},
> +		{},
> +	};

Is the implication of the above that any failure reported in the .result 
field will result in a retry? I don't think that matches the current 
behavior of read_capacity_16().

Thanks,

Bart.
Mike Christie Oct. 20, 2022, 7:15 p.m. UTC | #2 
On 10/19/22 5:15 PM, Bart Van Assche wrote:
> On 10/16/22 12:59, Mike Christie wrote:
>> +    struct scsi_failure failures[] = {
>> +        {
>> +            .sense = UNIT_ATTENTION,
>> +            .asc = 0x29,
>> +            .ascq = 0,
>> +            /* Device reset might occur several times */
>> +            .allowed = READ_CAPACITY_RETRIES_ON_RESET,
>> +            .result = SAM_STAT_CHECK_CONDITION,
>> +        },
>> +        {
>> +            .result = SCMD_FAILURE_ANY,
>> +            .allowed = 3,
>> +        },
>> +        {},
>> +    };
> 
> Is the implication of the above that any failure reported in the .result field will result in a retry? I don't think that matches the current behavior of read_capacity_16().

I think I made 2 mistakes:

1. Retrying on media_not_present
2. Retrying on the specific ILLEGAL_REQUEST error.

I made a similar mistakes on some other patches so will fix
them as well.
diff mbox series

Patch

diff --git a/drivers/scsi/sd.c b/drivers/scsi/sd.c
index 37eafa968116..aedc37613c81 100644
--- a/drivers/scsi/sd.c
+++ b/drivers/scsi/sd.c
@@ -2279,59 +2279,58 @@  static void read_capacity_error(struct scsi_disk *sdkp, struct scsi_device *sdp,
 static int read_capacity_16(struct scsi_disk *sdkp, struct scsi_device *sdp,
 						unsigned char *buffer)
 {
-	unsigned char cmd[16];
+	static const u8 cmd[16] = { SERVICE_ACTION_IN_16, SAI_READ_CAPACITY_16,
+				    0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, RC16_LEN };
 	struct scsi_sense_hdr sshdr;
 	int sense_valid = 0;
 	int the_result;
-	int retries = 3, reset_retries = READ_CAPACITY_RETRIES_ON_RESET;
 	unsigned int alignment;
 	unsigned long long lba;
 	unsigned sector_size;
+	struct scsi_failure failures[] = {
+		{
+			.sense = UNIT_ATTENTION,
+			.asc = 0x29,
+			.ascq = 0,
+			/* Device reset might occur several times */
+			.allowed = READ_CAPACITY_RETRIES_ON_RESET,
+			.result = SAM_STAT_CHECK_CONDITION,
+		},
+		{
+			.result = SCMD_FAILURE_ANY,
+			.allowed = 3,
+		},
+		{},
+	};
 
 	if (sdp->no_read_capacity_16)
 		return -EINVAL;
 
-	do {
-		memset(cmd, 0, 16);
-		cmd[0] = SERVICE_ACTION_IN_16;
-		cmd[1] = SAI_READ_CAPACITY_16;
-		cmd[13] = RC16_LEN;
-		memset(buffer, 0, RC16_LEN);
-
-		the_result = scsi_exec_req(((struct scsi_exec_args) {
-						.sdev = sdp,
-						.cmd = cmd,
-						.data_dir = DMA_FROM_DEVICE,
-						.buf = buffer,
-						.buf_len = RC16_LEN,
-						.sshdr = &sshdr,
-						.timeout = SD_TIMEOUT,
-						.retries = sdkp->max_retries }));
-
-		if (media_not_present(sdkp, &sshdr))
-			return -ENODEV;
+	the_result = scsi_exec_req(((struct scsi_exec_args) {
+					.sdev = sdp,
+					.cmd = cmd,
+					.data_dir = DMA_FROM_DEVICE,
+					.buf = buffer,
+					.buf_len = RC16_LEN,
+					.sshdr = &sshdr,
+					.timeout = SD_TIMEOUT,
+					.retries = sdkp->max_retries,
+					.failures = failures }));
 
-		if (the_result > 0) {
-			sense_valid = scsi_sense_valid(&sshdr);
-			if (sense_valid &&
-			    sshdr.sense_key == ILLEGAL_REQUEST &&
-			    (sshdr.asc == 0x20 || sshdr.asc == 0x24) &&
-			    sshdr.ascq == 0x00)
-				/* Invalid Command Operation Code or
-				 * Invalid Field in CDB, just retry
-				 * silently with RC10 */
-				return -EINVAL;
-			if (sense_valid &&
-			    sshdr.sense_key == UNIT_ATTENTION &&
-			    sshdr.asc == 0x29 && sshdr.ascq == 0x00)
-				/* Device reset might occur several times,
-				 * give it one more chance */
-				if (--reset_retries > 0)
-					continue;
-		}
-		retries--;
+	if (media_not_present(sdkp, &sshdr))
+		return -ENODEV;
 
-	} while (the_result && retries);
+	if (the_result > 0) {
+		sense_valid = scsi_sense_valid(&sshdr);
+		if (sense_valid && sshdr.sense_key == ILLEGAL_REQUEST &&
+		    (sshdr.asc == 0x20 || sshdr.asc == 0x24) &&
+		     sshdr.ascq == 0x00)
+			/*
+			 * Invalid Command Operation Code or Invalid Field in
+			 * CDB, just retry silently with RC10
+			 */
+			return -EINVAL;
+	}
 
 	if (the_result) {
 		sd_print_result(sdkp, "Read Capacity(16) failed", the_result);