lkml.org 
[lkml]   [2018]   [Apr]   [17]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
    Patch in this message
    /
    From
    Subject[PATCH 4.15 31/53] blk-mq: order getting budget and driver tag
    Date
    4.15-stable review patch.  If anyone has any objections, please let me know.

    ------------------

    From: Ming Lei <ming.lei@redhat.com>

    commit 0bca799b92807ee9be0890690f5dde7d8c6a8e25 upstream.

    This patch orders getting budget and driver tag by making sure to acquire
    driver tag after budget is got, this way can help to avoid the following
    race:

    1) before dispatch request from scheduler queue, get one budget first, then
    dequeue a request, call it request A.

    2) in another IO path for dispatching request B which is from hctx->dispatch,
    driver tag is got, then try to get budget in blk_mq_dispatch_rq_list(),
    unfortunately the budget is held by request A.

    3) meantime blk_mq_dispatch_rq_list() is called for dispatching request
    A, and try to get driver tag first, unfortunately no driver tag is
    available because the driver tag is held by request B

    4) both two IO pathes can't move on, and IO stall is caused.

    This issue can be observed when running dbench on USB storage.

    This patch fixes this issue by always getting budget before getting
    driver tag.

    Cc: stable@vger.kernel.org
    Fixes: de1482974080ec9e ("blk-mq: introduce .get_budget and .put_budget in blk_mq_ops")
    Cc: Christoph Hellwig <hch@lst.de>
    Cc: Bart Van Assche <bart.vanassche@wdc.com>
    Cc: Omar Sandoval <osandov@fb.com>
    Signed-off-by: Ming Lei <ming.lei@redhat.com>
    Signed-off-by: Jens Axboe <axboe@kernel.dk>
    Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

    ---
    block/blk-mq.c | 21 ++++++++++-----------
    1 file changed, 10 insertions(+), 11 deletions(-)

    --- a/block/blk-mq.c
    +++ b/block/blk-mq.c
    @@ -1096,7 +1096,12 @@ bool blk_mq_dispatch_rq_list(struct requ
    blk_status_t ret;

    rq = list_first_entry(list, struct request, queuelist);
    - if (!blk_mq_get_driver_tag(rq, &hctx, false)) {
    +
    + hctx = blk_mq_map_queue(rq->q, rq->mq_ctx->cpu);
    + if (!got_budget && !blk_mq_get_dispatch_budget(hctx))
    + break;
    +
    + if (!blk_mq_get_driver_tag(rq, NULL, false)) {
    /*
    * The initial allocation attempt failed, so we need to
    * rerun the hardware queue when a tag is freed. The
    @@ -1105,8 +1110,7 @@ bool blk_mq_dispatch_rq_list(struct requ
    * we'll re-run it below.
    */
    if (!blk_mq_mark_tag_wait(&hctx, rq)) {
    - if (got_budget)
    - blk_mq_put_dispatch_budget(hctx);
    + blk_mq_put_dispatch_budget(hctx);
    /*
    * For non-shared tags, the RESTART check
    * will suffice.
    @@ -1117,11 +1121,6 @@ bool blk_mq_dispatch_rq_list(struct requ
    }
    }

    - if (!got_budget && !blk_mq_get_dispatch_budget(hctx)) {
    - blk_mq_put_driver_tag(rq);
    - break;
    - }
    -
    list_del_init(&rq->queuelist);

    bd.rq = rq;
    @@ -1619,11 +1618,11 @@ static void __blk_mq_try_issue_directly(
    if (q->elevator)
    goto insert;

    - if (!blk_mq_get_driver_tag(rq, NULL, false))
    + if (!blk_mq_get_dispatch_budget(hctx))
    goto insert;

    - if (!blk_mq_get_dispatch_budget(hctx)) {
    - blk_mq_put_driver_tag(rq);
    + if (!blk_mq_get_driver_tag(rq, NULL, false)) {
    + blk_mq_put_dispatch_budget(hctx);
    goto insert;
    }


    \
     
     \ /
      Last update: 2018-04-17 18:50    [W:6.557 / U:0.088 seconds]
    ©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site