Messages in this thread Patch in this message | | | From | Liu Xiang <> | Subject | [PATCH] mmc: block: use REQ_HIPRI flag to complete request directly in own complete workqueue | Date | Thu, 21 Jan 2021 16:13:49 +0800 |
| |
After commit "40d09b53bfc557af7481b9d80f060a7ac9c7d314", request is completed in softirq. This may cause the system to suffer bad preemptoff time. The mmc driver has its own complete workqueue, but it can not work well now. The REQ_HIPRI flag can be used to complete request directly in its own complete workqueue and the preemptoff problem could be avoided.
Signed-off-by: Liu Xiang <liu.xiang@zlingsmart.com> --- drivers/mmc/core/block.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-)
diff --git a/drivers/mmc/core/block.c b/drivers/mmc/core/block.c index 42e27a298..c27239a89 100644 --- a/drivers/mmc/core/block.c +++ b/drivers/mmc/core/block.c @@ -1985,8 +1985,10 @@ static void mmc_blk_mq_post_req(struct mmc_queue *mq, struct request *req) */ if (mq->in_recovery) mmc_blk_mq_complete_rq(mq, req); - else if (likely(!blk_should_fake_timeout(req->q))) + else if (likely(!blk_should_fake_timeout(req->q))) { + req->cmd_flags |= REQ_HIPRI; blk_mq_complete_request(req); + } mmc_blk_mq_dec_in_flight(mq, req); } -- 2.17.1
| |