Re: [PATCH] scsi: Let scsi_execute_cmd() mark args->sshdr as invalid

On 5/20/23 8:19 PM, Martin K. Petersen wrote:

As I alluded to in the tracing thread, I'd like to see SK/ASC/ASCQ being generally available in the scsi_cmnd results instead of all this sense

Do you mean you just want a scsi_sense_hdr struct in the scsi_cmnd?

If so, I can do this when I resend the scsi retries patches since I have to touch every scsi_execute_cmd user and test them (at least what I can because I couldn't test things like scsi_dh_hp/rdac).

For the scsi_execute_cmd issue would we just do:

scsi_mq_init_request() { cmd->sshdr = { 0 }; .... }

scsi_execute_cmd()

blk_execute_rq()

if (sshdr) memcpy(sshdr, &scmd->sshdr, sizeof(*sshdr);

?

buffer and sense header micromanagement in every caller. That's a pretty heavy lift, though.