[SCSI] fix error handling in scsi_io_completion

There was a logic fault in scsi_io_completion() where zero transfer
commands that complete successfully were sent to the block layer as
not up to date.  This patch removes the if (good_bytes > 0) gate
around the successful completion, since zero transfer commands do have
good_bytes == 0.

Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
This commit is contained in:
James Bottomley 2006-07-02 10:06:28 -05:00 committed by James Bottomley
parent f58f8313a6
commit d6b0c53723

View file

@ -920,22 +920,20 @@ void scsi_io_completion(struct scsi_cmnd *cmd, unsigned int good_bytes)
* Next deal with any sectors which we were able to correctly * Next deal with any sectors which we were able to correctly
* handle. * handle.
*/ */
if (good_bytes > 0) { SCSI_LOG_HLCOMPLETE(1, printk("%ld sectors total, "
SCSI_LOG_HLCOMPLETE(1, printk("%ld sectors total, " "%d bytes done.\n",
"%d bytes done.\n", req->nr_sectors, good_bytes));
req->nr_sectors, good_bytes)); SCSI_LOG_HLCOMPLETE(1, printk("use_sg is %d\n", cmd->use_sg));
SCSI_LOG_HLCOMPLETE(1, printk("use_sg is %d\n", cmd->use_sg));
if (clear_errors) if (clear_errors)
req->errors = 0; req->errors = 0;
/* A number of bytes were successfully read. If there /* A number of bytes were successfully read. If there
* is leftovers and there is some kind of error * are leftovers and there is some kind of error
* (result != 0), retry the rest. * (result != 0), retry the rest.
*/ */
if (scsi_end_request(cmd, 1, good_bytes, !!result) == NULL) if (scsi_end_request(cmd, 1, good_bytes, result == 0) == NULL)
return; return;
}
/* good_bytes = 0, or (inclusive) there were leftovers and /* good_bytes = 0, or (inclusive) there were leftovers and
* result = 0, so scsi_end_request couldn't retry. * result = 0, so scsi_end_request couldn't retry.