scsi: iscsi: register sysfs for iscsi workqueue
RFC PATCH v2 1/1 scsi: iscsi: register sysfs for iscsi workqueue
Motivation:
This patch enable setting cpu affinity through "cpumask" for iscsi workqueues (iscsi_q_xx and iscsi_eh), so as to get performance isolation.
The max number of active worker was changed form 1 to 2, because "cpumask" of ordered workqueue isn't allowed to change.
Notes:
- Having 2 workers break the current ordering guarantees, please let me know if anyone depends on this.
- __WQ_LEGACY have to be left because of
23d11a5(workqueue: skip flush dependency checks for legacy workqueues)
--- drivers/scsi/libiscsi.c | 4 +++- drivers/scsi/scsi_transport_iscsi.c | 4 +++- 2 files changed, 6 insertions(+), 2 deletions(-)
drivers/scsi/libiscsi.c +3 -1
@@ -2627,7 +2627,9 @@ 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 =alloc_workqueue ("%s" , +WQ_SYSFS | __WQ_LEGACY |WQ_MEM_RECLAIM |WQ_UNBOUND , +2 , ihost->workq_name);if (!ihost->workq) goto free_host; }
drivers/scsi/scsi_transport_iscsi.c +3 -1
@@ -4602,7 +4602,9 @@ static __init int iscsi_transport_init(void) goto unregister_flashnode_bus; } - iscsi_eh_timer_workq =create_singlethread_workqueue ("iscsi_eh" ); + iscsi_eh_timer_workq =alloc_workqueue ("%s" , +WQ_SYSFS | __WQ_LEGACY |WQ_MEM_RECLAIM |WQ_UNBOUND , +2 ,"iscsi_eh" );if (!iscsi_eh_timer_workq) { err = -ENOMEM ; goto release_nls;
Signature · 1 line
2.9.5
Chris/Lee: Please review!
Quoted 50 lines
Motivation:
This patch enable setting cpu affinity through "cpumask" for iscsi workqueues (iscsi_q_xxandiscsi_eh), so as to get performance isolation.The max number of active worker was changed form 1 to 2, because "cpumask" of ordered workqueue isn't allowed to change.
Notes:
- Having 2 workers break the current ordering guarantees, please let me know if anyone depends on this.-
__WQ_LEGACYhave to be left because of
23d11a5(workqueue: skip flush dependency checks for legacy workqueues)Signed-off-by Bob Liu <[email protected]>--- drivers/scsi/libiscsi.c | 4 +++- drivers/scsi/scsi_transport_iscsi.c | 4 +++- 2 files changed, 6 insertions(+), 2 deletions(-)drivers/scsi/libiscsi.c +3
-1@@ -2627,7 +2627,9 @@ 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 =alloc_workqueue ("%s" , +WQ_SYSFS | __WQ_LEGACY |WQ_MEM_RECLAIM |WQ_UNBOUND , +2 , ihost->workq_name);if (!ihost->workq) goto free_host; }drivers/scsi/scsi_transport_iscsi.c +3
-1@@ -4602,7 +4602,9 @@ static __init int iscsi_transport_init(void) goto unregister_flashnode_bus; } - iscsi_eh_timer_workq =create_singlethread_workqueue ("iscsi_eh" ); + iscsi_eh_timer_workq =alloc_workqueue ("%s" , +WQ_SYSFS | __WQ_LEGACY |WQ_MEM_RECLAIM |WQ_UNBOUND , +2 ,"iscsi_eh" );if (!iscsi_eh_timer_workq) { err = -ENOMEM ; goto release_nls;
Signature · 1 line
Martin K. Petersen Oracle Linux Engineering
friendly ping.
Quoted 51 lines · Bob Liu
Motivation:
This patch enable setting cpu affinity through "cpumask" for iscsi workqueues (iscsi_q_xxandiscsi_eh), so as to get performance isolation.The max number of active worker was changed form 1 to 2, because "cpumask" of ordered workqueue isn't allowed to change.
Notes:
- Having 2 workers break the current ordering guarantees, please let me know if anyone depends on this.-
__WQ_LEGACYhave to be left because of
23d11a5(workqueue: skip flush dependency checks for legacy workqueues)Signed-off-by Bob Liu <[email protected]>--- drivers/scsi/libiscsi.c | 4 +++- drivers/scsi/scsi_transport_iscsi.c | 4 +++- 2 files changed, 6 insertions(+), 2 deletions(-)drivers/scsi/libiscsi.c +3
-1@@ -2627,7 +2627,9 @@ 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 =alloc_workqueue ("%s" , +WQ_SYSFS | __WQ_LEGACY |WQ_MEM_RECLAIM |WQ_UNBOUND , +2 , ihost->workq_name);if (!ihost->workq) goto free_host; }drivers/scsi/scsi_transport_iscsi.c +3
-1@@ -4602,7 +4602,9 @@ static __init int iscsi_transport_init(void) goto unregister_flashnode_bus; } - iscsi_eh_timer_workq =create_singlethread_workqueue ("iscsi_eh" ); + iscsi_eh_timer_workq =alloc_workqueue ("%s" , +WQ_SYSFS | __WQ_LEGACY |WQ_MEM_RECLAIM |WQ_UNBOUND , +2 ,"iscsi_eh" );if (!iscsi_eh_timer_workq) { err = -ENOMEM ; goto release_nls;
Quoted 46 lines · Bob Liu
friendly ping.
Quoted 43 lines · Bob Liu
Motivation:
This patch enable setting cpu affinity through "cpumask" for iscsi workqueues (iscsi_q_xxandiscsi_eh), so as to get performance isolation.The max number of active worker was changed form 1 to 2, because "cpumask" of ordered workqueue isn't allowed to change.
Notes:
- Having 2 workers break the current ordering guarantees, please let me know if anyone depends on this.-
__WQ_LEGACYhave to be left because of
23d11a5(workqueue: skip flush dependency checks for legacy workqueues)Signed-off-by Bob Liu <[email protected]>--- drivers/scsi/libiscsi.c | 4 +++- drivers/scsi/scsi_transport_iscsi.c | 4 +++- 2 files changed, 6 insertions(+), 2 deletions(-)drivers/scsi/libiscsi.c +3
-1@@ -2627,7 +2627,9 @@ 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 =alloc_workqueue ("%s" , +WQ_SYSFS | __WQ_LEGACY |WQ_MEM_RECLAIM |WQ_UNBOUND , +2 , ihost->workq_name);if (!ihost->workq) goto free_host; }drivers/scsi/scsi_transport_iscsi.c +0
-0@@ -4602,7 +4602,9 @@ static __init int iscsi_transport_init(void) goto unregister_flashnode_bus; }
Quoted 9 lines
drivers/scsi/scsi_transport_iscsi.c static __init int iscsi_transport_init(void) · line 4610
Quoted 8 lines
drivers/scsi/scsi_transport_iscsi.c +3
-1- iscsi_eh_timer_workq =create_singlethread_workqueue ("iscsi_eh" ); + iscsi_eh_timer_workq =alloc_workqueue ("%s" , +WQ_SYSFS | __WQ_LEGACY |WQ_MEM_RECLAIM |WQ_UNBOUND , +2 ,"iscsi_eh" );if (!iscsi_eh_timer_workq) { err = -ENOMEM ; goto release_nls;
Martin K. Petersen
AppliedQuoted 8 lines · Bob Liu
Motivation:
This patch enable setting cpu affinity through "cpumask" for iscsi workqueues (iscsi_q_xxandiscsi_eh), so as to get performance isolation.The max number of active worker was changed form 1 to 2, because "cpumask" of ordered workqueue isn't allowed to change.
[...]
Applied to 5.8/scsi-queue, thanks!
[1/1] scsi: iscsi: Register sysfs for iscsi workqueue
https://git.kernel.org/mkp/scsi/c/3ce419662dd4
Signature · 1 line
Martin K. Petersen Oracle Linux Engineering