Consider two bfq_queues, say Q1 and Q2, with Q2 empty. If a request of Q1 gets completed shortly before a new request arrives for Q2, then BFQ flags Q1 as a candidate waker for Q2. Yet, the arrival of this new request may have a different cause, in the following case. If also Q2 has requests in flight while waiting for the arrival of a new request, then the completion of its own requests may be the actual cause of the awakening of the process that sends I/O to Q2. So Q1 may be flagged wrongly as a candidate waker. This commit avoids this deceptive flagging, by disabling candidate-waker flagging for Q2, if Q2 has in-flight I/O. Signed-off-by: Paolo Valente <paolo.valente@xxxxxxxxxx> --- block/bfq-iosched.c | 21 +++++++++++++-------- 1 file changed, 13 insertions(+), 8 deletions(-) diff --git a/block/bfq-iosched.c b/block/bfq-iosched.c index 7bf073ef9443..a273b2bcea2a 100644 --- a/block/bfq-iosched.c +++ b/block/bfq-iosched.c @@ -1985,14 +1985,18 @@ static void bfq_update_io_intensity(struct bfq_queue *bfqq, u64 now_ns) * Turning back to the detection of a waker queue, a queue Q is deemed * as a waker queue for bfqq if, for three consecutive times, bfqq * happens to become non empty right after a request of Q has been - * completed. In particular, on the first time, Q is tentatively set - * as a candidate waker queue, while on the third consecutive time - * that Q is detected, the field waker_bfqq is set to Q, to confirm - * that Q is a waker queue for bfqq. These detection steps are - * performed only if bfqq has a long think time, so as to make it more - * likely that bfqq's I/O is actually being blocked by a - * synchronization. This last filter, plus the above three-times - * requirement, make false positives less likely. + * completed. In this respect, even if bfqq is empty, we do not check + * for a waker if it still has some in-flight I/O. In fact, in this + * case bfqq is actually still being served by the drive, and may + * receive new I/O on the completion of some of the in-flight + * requests. In particular, on the first time, Q is tentatively set as + * a candidate waker queue, while on the third consecutive time that Q + * is detected, the field waker_bfqq is set to Q, to confirm that Q is + * a waker queue for bfqq. These detection steps are performed only if + * bfqq has a long think time, so as to make it more likely that + * bfqq's I/O is actually being blocked by a synchronization. This + * last filter, plus the above three-times requirement, make false + * positives less likely. * * NOTE * @@ -2018,6 +2022,7 @@ static void bfq_check_waker(struct bfq_data *bfqd, struct bfq_queue *bfqq, if (!bfqd->last_completed_rq_bfqq || bfqd->last_completed_rq_bfqq == bfqq || bfq_bfqq_has_short_ttime(bfqq) || + bfqq->dispatched > 0 || now_ns - bfqd->last_completion >= 4 * NSEC_PER_MSEC || bfqd->last_completed_rq_bfqq == bfqq->waker_bfqq) return; -- 2.20.1