Re: [PATCH v8 5/9] scsi: core: Retry unaligned zoned writes

On 8/16/23 02:29, Bart Van Assche wrote:

On 8/14/23 05:36, Damien Le Moal wrote:

quoted

On 8/12/23 06:35, Bart Van Assche wrote:

quoted

--- a/drivers/scsi/sd.c
+++ b/drivers/scsi/sd.c

@@ -1238,6 +1238,8 @@ static blk_status_t sd_setup_read_write_cmnd(struct scsi_cmnd *cmd)
  	cmd->transfersize = sdp->sector_size;
  	cmd->underflow = nr_blocks << 9;
  	cmd->allowed = sdkp->max_retries;
+	if (!rq->q->limits.use_zone_write_lock && blk_rq_is_seq_zoned_write(rq))

This condition could be written as a little inline helper
blk_req_need_zone_write_lock(), which could be used in mq-dealine patch 2.

The above test differs from the tests in the mq-deadline I/O scheduler.
The mq-deadline I/O scheduler uses write locking if
rq->q->limits.use_zone_write_lock && q->disk->seq_zones_wlock &&
blk_rq_is_seq_zoned_write(rq). The above test is different and occurs
two times in scsi_error.c and one time in sd.c. Do you still want me to
introduce a helper function for that expression?

May be not needed. But a comment explaining it may be good to add.
I still think that added the test for use_zone_write_lock in
blk_req_needs_zone_write_lock() is needed though, for consistency of all
functions related to zone locking.

Thanks,

Bart.

-- 
Damien Le Moal
Western Digital Research