openkylin
/
linux
mirror of https://gitee.com/openkylin/linux.git
9
0
Fork 0
Code Issues Projects Releases Wiki Activity

IB/hfi1: Avoid hardlockup with flushlist_lock 

Heavy contention of the sde flushlist_lock can cause hard lockups at
extreme scale when the flushing logic is under stress.

Mitigate by replacing the item at a time copy to the local list with
an O(1) list_splice_init() and using the high priority work queue to
do the flushes.

Fixes: 7724105686 ("IB/hfi1: add driver files")
Cc: <stable@vger.kernel.org>
Reviewed-by: Dennis Dalessandro <dennis.dalessandro@intel.com>
Signed-off-by: Mike Marciniszyn <mike.marciniszyn@intel.com>
Signed-off-by: Dennis Dalessandro <dennis.dalessandro@intel.com>
Signed-off-by: Doug Ledford <dledford@redhat.com>
This commit is contained in:
Mike Marciniszyn 2019-06-14 12:32:26 -04:00 committed by Doug Ledford
parent cc78076af1
commit cf131a8196
1 changed files with 3 additions and 6 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

9
drivers/infiniband/hw/hfi1/sdma.c
View File

@ -410,10 +410,7 @@ static void sdma_flush(struct sdma_engine *sde)
sdma_flush_descq(sde); sdma_flush_descq(sde);
spin_lock_irqsave(&sde->flushlist_lock, flags); spin_lock_irqsave(&sde->flushlist_lock, flags);
/* copy flush list */ /* copy flush list */
list_for_each_entry_safe(txp, txp_next, &sde->flushlist, list) { list_splice_init(&sde->flushlist, &flushlist);
list_del_init(&txp->list);
list_add_tail(&txp->list, &flushlist);
}
spin_unlock_irqrestore(&sde->flushlist_lock, flags); spin_unlock_irqrestore(&sde->flushlist_lock, flags);
/* flush from flush list */ /* flush from flush list */
list_for_each_entry_safe(txp, txp_next, &flushlist, list) list_for_each_entry_safe(txp, txp_next, &flushlist, list)
@ -2413,7 +2410,7 @@ int sdma_send_txreq(struct sdma_engine *sde,
list_add_tail(&tx->list, &sde->flushlist); list_add_tail(&tx->list, &sde->flushlist);
spin_unlock(&sde->flushlist_lock); spin_unlock(&sde->flushlist_lock);
iowait_inc_wait_count(wait, tx->num_desc); iowait_inc_wait_count(wait, tx->num_desc);
schedule_work(&sde->flush_worker); queue_work_on(sde->cpu, system_highpri_wq, &sde->flush_worker);
ret = -ECOMM; ret = -ECOMM;
goto unlock; goto unlock;
nodesc: nodesc:
@ -2511,7 +2508,7 @@ int sdma_send_txlist(struct sdma_engine *sde, struct iowait_work *wait,
iowait_inc_wait_count(wait, tx->num_desc); iowait_inc_wait_count(wait, tx->num_desc);
} }
spin_unlock(&sde->flushlist_lock); spin_unlock(&sde->flushlist_lock);
schedule_work(&sde->flush_worker); queue_work_on(sde->cpu, system_highpri_wq, &sde->flush_worker);
ret = -ECOMM; ret = -ECOMM;
goto update_tail; goto update_tail;
nodesc: nodesc: