diff options
| author | James Smart <jsmart2021@gmail.com> | 2019-01-28 22:14:29 +0300 |
|---|---|---|
| committer | Martin K. Petersen <martin.petersen@oracle.com> | 2019-02-06 06:29:09 +0300 |
| commit | 45aa312e211f1c8c44a8e90c184a21bbb5b5572c (patch) | |
| tree | b205f600bf511b4ba1851bf7386085b9be0006ab /drivers/scsi/lpfc/lpfc_attr.c | |
| parent | c490850a094794e7515737a6939146966c826577 (diff) | |
| download | linux-45aa312e211f1c8c44a8e90c184a21bbb5b5572c.tar.xz | |
scsi: lpfc: Allow override of hardware queue selection policies
Default behavior is to use the information from the upper IO stacks to
select the hardware queue to use for IO submission. Which typically has
good cpu affinity.
However, the driver, when used on some variants of the upstream kernel, has
found queuing information to be suboptimal for FCP or IO completion locked
on particular cpus.
For command submission situations, the lpfc_fcp_io_sched module parameter
can be set to specify a hardware queue selection policy that overrides the
os stack information.
For IO completion situations, rather than queing cq processing based on the
cpu servicing the interrupting event, schedule the cq processing on the cpu
associated with the hardware queue's cq.
Signed-off-by: Dick Kennedy <dick.kennedy@broadcom.com>
Signed-off-by: James Smart <jsmart2021@gmail.com>
Reviewed-by: Hannes Reinecke <hare@suse.com>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Diffstat (limited to 'drivers/scsi/lpfc/lpfc_attr.c')
| -rw-r--r-- | drivers/scsi/lpfc/lpfc_attr.c | 11 |
1 files changed, 6 insertions, 5 deletions
diff --git a/drivers/scsi/lpfc/lpfc_attr.c b/drivers/scsi/lpfc/lpfc_attr.c index 47aa2af885a4..93a96491899c 100644 --- a/drivers/scsi/lpfc/lpfc_attr.c +++ b/drivers/scsi/lpfc/lpfc_attr.c @@ -5275,11 +5275,12 @@ LPFC_ATTR_R(xri_rebalancing, 1, 0, 1, "Enable/Disable XRI rebalancing"); /* * lpfc_io_sched: Determine scheduling algrithmn for issuing FCP cmds * range is [0,1]. Default value is 0. - * For [0], FCP commands are issued to Work Queues ina round robin fashion. + * For [0], FCP commands are issued to Work Queues based on upper layer + * hardware queue index. * For [1], FCP commands are issued to a Work Queue associated with the * current CPU. * - * LPFC_FCP_SCHED_ROUND_ROBIN == 0 + * LPFC_FCP_SCHED_BY_HDWQ == 0 * LPFC_FCP_SCHED_BY_CPU == 1 * * The driver dynamically sets this to 1 (BY_CPU) if it's able to set up cpu @@ -5287,11 +5288,11 @@ LPFC_ATTR_R(xri_rebalancing, 1, 0, 1, "Enable/Disable XRI rebalancing"); * CPU. Otherwise, the default 0 (Round Robin) scheduling of FCP/NVME I/Os * through WQs will be used. */ -LPFC_ATTR_RW(fcp_io_sched, LPFC_FCP_SCHED_ROUND_ROBIN, - LPFC_FCP_SCHED_ROUND_ROBIN, +LPFC_ATTR_RW(fcp_io_sched, LPFC_FCP_SCHED_BY_HDWQ, + LPFC_FCP_SCHED_BY_HDWQ, LPFC_FCP_SCHED_BY_CPU, "Determine scheduling algorithm for " - "issuing commands [0] - Round Robin, [1] - Current CPU"); + "issuing commands [0] - Hardware Queue, [1] - Current CPU"); /* * lpfc_ns_query: Determine algrithmn for NameServer queries after RSCN |