Message ID | 20190328171012.26425-5-hmadhani@marvell.com (mailing list archive) |
---|---|
State | Superseded |
Headers | show |
Series | qla2xxx: Misc updates and bug fixes for the driver. | expand |
On Thu, 2019-03-28 at 10:10 -0700, Himanshu Madhani wrote: > From: Giridhar Malavali <gmalavali@marvell.com> > > This patch increases max_sgl_segments value to max supported > which is 1024. Increase in max_sgl_segments will support larger > IO size from driver. > > Signed-off-by: Giridhar Malavali <gmalavali@marvell.com> > Signed-off-by: Himanshu Madhani <hmadhani@marvell.com> > --- > drivers/scsi/qla2xxx/qla_nvme.c | 2 + > 1 file changed, 1 insertion(), 1 deletion(-) > > diff --git a/drivers/scsi/qla2xxx/qla_nvme.c b/drivers/scsi/qla2xxx/qla_nvme.c > index 41c85da3ab32..cc2afc21a30d 100644 > --- a/drivers/scsi/qla2xxx/qla_nvme.c > +++ b/drivers/scsi/qla2xxx/qla_nvme.c > @@ -573,7 +573,7 @@ static struct nvme_fc_port_template qla_nvme_fc_transport = { > .fcp_io = qla_nvme_post_cmd, > .fcp_abort = qla_nvme_fcp_abort, > .max_hw_queues = 8, > - .max_sgl_segments = 128, > + .max_sgl_segments = 1024, > .max_dif_sgl_segments = 64, > .dma_boundary = 0xFFFFFFFF, > .local_priv_sz = 8, Where does the original value "128" come from? Where does the new value "1024" come from? Do all firmware versions support the new and larger value? Thanks, Bart.
On 3/28/19, 10:39 AM, "linux-scsi-owner@vger.kernel.org on behalf of Bart Van Assche" <linux-scsi-owner@vger.kernel.org on behalf of bvanassche@acm.org> wrote: On Thu, 2019-03-28 at 10:10 -0700, Himanshu Madhani wrote: > From: Giridhar Malavali <gmalavali@marvell.com> > > This patch increases max_sgl_segments value to max supported > which is 1024. Increase in max_sgl_segments will support larger > IO size from driver. > > Signed-off-by: Giridhar Malavali <gmalavali@marvell.com> > Signed-off-by: Himanshu Madhani <hmadhani@marvell.com> > --- > drivers/scsi/qla2xxx/qla_nvme.c | 2 + > 1 file changed, 1 insertion(), 1 deletion(-) > > diff --git a/drivers/scsi/qla2xxx/qla_nvme.c b/drivers/scsi/qla2xxx/qla_nvme.c > index 41c85da3ab32..cc2afc21a30d 100644 > --- a/drivers/scsi/qla2xxx/qla_nvme.c > +++ b/drivers/scsi/qla2xxx/qla_nvme.c > @@ -573,7 +573,7 @@ static struct nvme_fc_port_template qla_nvme_fc_transport = { > .fcp_io = qla_nvme_post_cmd, > .fcp_abort = qla_nvme_fcp_abort, > .max_hw_queues = 8, > - .max_sgl_segments = 128, > + .max_sgl_segments = 1024, > .max_dif_sgl_segments = 64, > .dma_boundary = 0xFFFFFFFF, > .local_priv_sz = 8, Where does the original value "128" come from? Where does the new value "1024" come from? Do all firmware versions support the new and larger value? The original 128 value was used during initial bring up. 1024 is what our hardware/firmware can support. Yes, all firmware supports the new larger value. -- Giri Thanks, Bart.
diff --git a/drivers/scsi/qla2xxx/qla_nvme.c b/drivers/scsi/qla2xxx/qla_nvme.c index 41c85da3ab32..cc2afc21a30d 100644 --- a/drivers/scsi/qla2xxx/qla_nvme.c +++ b/drivers/scsi/qla2xxx/qla_nvme.c @@ -573,7 +573,7 @@ static struct nvme_fc_port_template qla_nvme_fc_transport = { .fcp_io = qla_nvme_post_cmd, .fcp_abort = qla_nvme_fcp_abort, .max_hw_queues = 8, - .max_sgl_segments = 128, + .max_sgl_segments = 1024, .max_dif_sgl_segments = 64, .dma_boundary = 0xFFFFFFFF, .local_priv_sz = 8,