> > On 2021/06/02 21:29, Changheun Lee wrote: > > > Set max_bio_bytes same with queue max sectors. It will lead to fast bio > > > submit when bio size is over than queue max sectors. And it might be helpful > > > to align submit bio size in some case. > > > > > > Signed-off-by: Changheun Lee <nanich.lee@xxxxxxxxxxx> > > > --- > > > drivers/scsi/scsi_lib.c | 2 ++ > > > 1 file changed, 2 insertions(+) > > > > > > diff --git a/drivers/scsi/scsi_lib.c b/drivers/scsi/scsi_lib.c > > > index 532304d42f00..f6269268b0e0 100644 > > > --- a/drivers/scsi/scsi_lib.c > > > +++ b/drivers/scsi/scsi_lib.c > > > @@ -1837,6 +1837,8 @@ void __scsi_init_queue(struct Scsi_Host *shost, struct request_queue *q) > > > blk_queue_virt_boundary(q, shost->virt_boundary_mask); > > > dma_set_max_seg_size(dev, queue_max_segment_size(q)); > > > > > > + blk_queue_max_bio_bytes(q, queue_max_sectors(q)); > > > > Doing this unconditionally for all scsi block devices is probably not a good > > idea. Cannot this be moved to the LLD handling the devices that actually need it ? > > OK, I'll try to check more nice location in LLD. > > > > > > + > > > /* > > > * Set a reasonable default alignment: The larger of 32-byte (dword), > > > * which is a common minimum for HBAs, and the minimum DMA alignment, > > > > > > > > > -- > > Damien Le Moal > > Western Digital Research > > Thank you, > Changheun Lee I think below location might be good. feedback to me please. diff --git a/drivers/scsi/ufs/ufshcd.c b/drivers/scsi/ufs/ufshcd.c index 3eb54937f1d8..0f97b7d275ee 100644 --- a/drivers/scsi/ufs/ufshcd.c +++ b/drivers/scsi/ufs/ufshcd.c @@ -4831,6 +4831,9 @@ static int ufshcd_slave_alloc(struct scsi_device *sdev) ufshcd_get_lu_power_on_wp_status(hba, sdev); + blk_queue_max_bio_bytes(sdev->request_queue, + queue_max_sectors(sdev->request_queue)); + return 0; } Thank you, Changheun Lee
