Re: [PATCH v3 1/2] scsi: Handle Unit Attention when issuing SCSI command

[James Bottomley <jejb@linux.vnet.ibm.com>]

On Tue, 2016-10-25 at 15:16 -0700, Bart Van Assche wrote:
> On 10/23/2016 08:20 PM, Gabriel Krisman Bertazi wrote:
> > +
> >  /**
> >   * scsi_execute - insert request and wait for the result
> >   * @sdev:	scsi device
> > @@ -187,7 +197,14 @@ int scsi_execute(struct scsi_device *sdev,
> > const unsigned char *cmd,
> >  	struct request *req;
> >  	int write = (data_direction == DMA_TO_DEVICE);
> >  	int ret = DRIVER_ERROR << 24;
> > +	unsigned char sense_buf[SCSI_SENSE_BUFFERSIZE];
> > 
> > +	if (!sense) {
> > +		sense = sense_buf;
> > +		memset(sense, 0, SCSI_SENSE_BUFFERSIZE);
> > +	}
> > +
> > + retry:
> >  	req = blk_get_request(sdev->request_queue, write,
> > __GFP_RECLAIM);
> >  	if (IS_ERR(req))
> >  		return ret;
> > @@ -210,6 +227,13 @@ int scsi_execute(struct scsi_device *sdev,
> > const unsigned char *cmd,
> >  	 */
> >  	blk_execute_rq(req->q, NULL, req, 1);
> > 
> > +	if (scsi_sense_unit_attention(sense) && req->retries > 0)
> > {
> > +		memset(sense, 0, SCSI_SENSE_BUFFERSIZE);
> > +		retries = req->retries - 1;
> > +		blk_put_request(req);
> > +		goto retry;
> > +	}
> > +
> >  	/*
> >  	 * Some devices (USB mass-storage in particular) may
> > transfer
> >  	 * garbage data together with a residue indicating that
> > the data
> 
>  From scsi_io_completion():
> 
> 	if (sense_valid && !sense_deferred) {
> 		switch (sshdr.sense_key) {
> 		case UNIT_ATTENTION:
> 			if (cmd->device->removable) {
> 				cmd->device->changed = 1;
> 				action = ACTION_FAIL;
> 			} else {
> 				action = ACTION_RETRY;
> 			}
> 			[ ... ]
> 
> Why do you think new retry code is needed in scsi_execute()?

Because scsi_execute uses REQ_BLOCK_PC which is completed before you
get to that code.

James