| Message ID | 1602750425-240341-1-git-send-email-john.garry@huawei.com |
|---|---|
| State | Superseded |
| Headers | show |
| Series | scsi: hisi_sas: Stop using queue #0 always for v2 hw | expand |
John, > In commit 8d98416a55eb ("scsi: hisi_sas: Switch v3 hw to MQ"), the dispatch > function was changed to choose the delivery queue based on the request tag > HW queue index. Applied to 5.10/scsi-fixes, thanks! -- Martin K. Petersen Oracle Linux Engineering
diff --git a/drivers/scsi/hisi_sas/hisi_sas_main.c b/drivers/scsi/hisi_sas/hisi_sas_main.c index a994c7b8d26f..fd980a86aa21 100644 --- a/drivers/scsi/hisi_sas/hisi_sas_main.c +++ b/drivers/scsi/hisi_sas/hisi_sas_main.c @@ -444,7 +444,7 @@ static int hisi_sas_task_prep(struct sas_task *task, } } - if (scmd) { + if (scmd && hisi_hba->shost->nr_hw_queues) { unsigned int dq_index; u32 blk_tag;
In commit 8d98416a55eb ("scsi: hisi_sas: Switch v3 hw to MQ"), the dispatch function was changed to choose the delivery queue based on the request tag HW queue index. This heavily degrades performance for v2 hw, since the HW queues are not exposed there, and, as such, HW queue #0 is used for every command. Revert to previous behaviour for when nr_hw_queues is not set, that being to choose the HW queue based on target device index. Fixes: 8d98416a55eb ("scsi: hisi_sas: Switch v3 hw to MQ") Signed-off-by: John Garry <john.garry@huawei.com> --- Martin, James, this is a fix for the code which was went via Jens' tree, now merged. Maybe I need to wait until rc1 is released to send. Or maybe, if there is a second SCSI pull for v5.10 (and a rebase to Linus' tree), it can go then. It does not apply to Martin's 5.10/queue. Let me know, thanks!