| Message ID | 20200611100717.27506-2-bob.liu@oracle.com (mailing list archive) |
|---|---|
| State | Deferred |
| Headers | show |
| Series | [1/2] workqueue: don't always set __WQ_ORDERED implicitly | expand |
On 6/11/20 5:07 AM, Bob Liu wrote: > This patch enable setting cpu affinity through "cpumask" for below > scsi/iscsi workqueues, so as to get better isolation. > - scsi_wq_* > - scsi_tmf_* > - iscsi_q_xx > - iscsi_eh > > Signed-off-by: Bob Liu <bob.liu@oracle.com> > --- > drivers/scsi/hosts.c | 4 ++-- > drivers/scsi/libiscsi.c | 2 +- > drivers/scsi/scsi_transport_iscsi.c | 2 +- > 3 files changed, 4 insertions(+), 4 deletions(-) > > diff --git a/drivers/scsi/hosts.c b/drivers/scsi/hosts.c > index 1d669e4..4b9f80d 100644 > --- a/drivers/scsi/hosts.c > +++ b/drivers/scsi/hosts.c > @@ -272,7 +272,7 @@ int scsi_add_host_with_dma(struct Scsi_Host *shost, struct device *dev, > if (shost->transportt->create_work_queue) { > snprintf(shost->work_q_name, sizeof(shost->work_q_name), > "scsi_wq_%d", shost->host_no); > - shost->work_q = create_singlethread_workqueue( > + shost->work_q = create_singlethread_workqueue_noorder( > shost->work_q_name); > if (!shost->work_q) { > error = -EINVAL; This patch seems ok for the iscsi, fc, tmf, and non transport class scan uses. We are either heavy handed with flushes or did not need ordering. I don't know about the zfcp use though, so I cc'd the developers listed as maintainers. It looks like for zfcp we can do: zfcp_scsi_rport_register->fc_remote_port_add->fc_remote_port_create->scsi_queue_work to scan the scsi target on the rport. and then zfcp_scsi_rport_register can call zfcp_unit_queue_scsi_scan-> scsi_queue_work which will scan for a specific lun. It looks ok if those are not ordered, but I would get their review to make sure.
On Mon, Jun 22, 2020 at 10:40:09AM -0500, Mike Christie wrote: > On 6/11/20 5:07 AM, Bob Liu wrote: > > This patch enable setting cpu affinity through "cpumask" for below > > scsi/iscsi workqueues, so as to get better isolation. > > - scsi_wq_* > > - scsi_tmf_* > > - iscsi_q_xx > > - iscsi_eh > > > > Signed-off-by: Bob Liu <bob.liu@oracle.com> > > --- > > drivers/scsi/hosts.c | 4 ++-- > > drivers/scsi/libiscsi.c | 2 +- > > drivers/scsi/scsi_transport_iscsi.c | 2 +- > > 3 files changed, 4 insertions(+), 4 deletions(-) > > > > diff --git a/drivers/scsi/hosts.c b/drivers/scsi/hosts.c > > index 1d669e4..4b9f80d 100644 > > --- a/drivers/scsi/hosts.c > > +++ b/drivers/scsi/hosts.c > > @@ -272,7 +272,7 @@ int scsi_add_host_with_dma(struct Scsi_Host *shost, struct device *dev, > > if (shost->transportt->create_work_queue) { > > snprintf(shost->work_q_name, sizeof(shost->work_q_name), > > "scsi_wq_%d", shost->host_no); > > - shost->work_q = create_singlethread_workqueue( > > + shost->work_q = create_singlethread_workqueue_noorder( > > shost->work_q_name); > > if (!shost->work_q) { > > error = -EINVAL; > > This patch seems ok for the iscsi, fc, tmf, and non transport class scan > uses. We are either heavy handed with flushes or did not need ordering. > > I don't know about the zfcp use though, so I cc'd the developers listed as > maintainers. It looks like for zfcp we can do: Thx for the notice. > > zfcp_scsi_rport_register->fc_remote_port_add->fc_remote_port_create->scsi_queue_work > to scan the scsi target on the rport. > > and then zfcp_scsi_rport_register can call zfcp_unit_queue_scsi_scan-> > scsi_queue_work which will scan for a specific lun. > > It looks ok if those are not ordered, but I would get their review to make > sure. I am not aware of any temporal requirements of those LUN-scans, so I think making them not explicitly ordered shouldn't hurt us. The target scan itself is protected again by `shost->scan_mutex`.. so all fine I think.
diff --git a/drivers/scsi/hosts.c b/drivers/scsi/hosts.c index 1d669e4..4b9f80d 100644 --- a/drivers/scsi/hosts.c +++ b/drivers/scsi/hosts.c @@ -272,7 +272,7 @@ int scsi_add_host_with_dma(struct Scsi_Host *shost, struct device *dev, if (shost->transportt->create_work_queue) { snprintf(shost->work_q_name, sizeof(shost->work_q_name), "scsi_wq_%d", shost->host_no); - shost->work_q = create_singlethread_workqueue( + shost->work_q = create_singlethread_workqueue_noorder( shost->work_q_name); if (!shost->work_q) { error = -EINVAL; @@ -487,7 +487,7 @@ struct Scsi_Host *scsi_host_alloc(struct scsi_host_template *sht, int privsize) } shost->tmf_work_q = alloc_workqueue("scsi_tmf_%d", - WQ_UNBOUND | WQ_MEM_RECLAIM, + WQ_UNBOUND | WQ_MEM_RECLAIM | WQ_SYSFS | __WQ_ORDERED_DISABLE, 1, shost->host_no); if (!shost->tmf_work_q) { shost_printk(KERN_WARNING, shost, diff --git a/drivers/scsi/libiscsi.c b/drivers/scsi/libiscsi.c index 70b99c0..6808cf3 100644 --- a/drivers/scsi/libiscsi.c +++ b/drivers/scsi/libiscsi.c @@ -2627,7 +2627,7 @@ struct Scsi_Host *iscsi_host_alloc(struct scsi_host_template *sht, if (xmit_can_sleep) { snprintf(ihost->workq_name, sizeof(ihost->workq_name), "iscsi_q_%d", shost->host_no); - ihost->workq = create_singlethread_workqueue(ihost->workq_name); + ihost->workq = create_singlethread_workqueue_noorder(ihost->workq_name); if (!ihost->workq) goto free_host; } diff --git a/drivers/scsi/scsi_transport_iscsi.c b/drivers/scsi/scsi_transport_iscsi.c index dfc726f..d07a0e4 100644 --- a/drivers/scsi/scsi_transport_iscsi.c +++ b/drivers/scsi/scsi_transport_iscsi.c @@ -4602,7 +4602,7 @@ static __init int iscsi_transport_init(void) goto unregister_flashnode_bus; } - iscsi_eh_timer_workq = create_singlethread_workqueue("iscsi_eh"); + iscsi_eh_timer_workq = create_singlethread_workqueue_noorder("iscsi_eh"); if (!iscsi_eh_timer_workq) { err = -ENOMEM; goto release_nls;
This patch enable setting cpu affinity through "cpumask" for below scsi/iscsi workqueues, so as to get better isolation. - scsi_wq_* - scsi_tmf_* - iscsi_q_xx - iscsi_eh Signed-off-by: Bob Liu <bob.liu@oracle.com> --- drivers/scsi/hosts.c | 4 ++-- drivers/scsi/libiscsi.c | 2 +- drivers/scsi/scsi_transport_iscsi.c | 2 +- 3 files changed, 4 insertions(+), 4 deletions(-)