When there isn't enough vring descriptor for adding to vq, blk-mq will be put as stopped state until some of pending descriptors are completed & freed. Unfortunately, the vq's interrupt may come just before blk-mq's BLK_MQ_S_STOPPED flag is set, so the blk-mq will still be kept as stopped even though lots of descriptors are completed and freed in the interrupt handler. The worst case is that all pending descriptors are freed in the interrupt handler, and the queue is kept as stopped forever. This patch fixes the problem by starting/stopping blk-mq with holding vq_lock. Cc: Jens Axboe <axboe@kernel.dk> Cc: Rusty Russell <rusty@rustcorp.com.au> Signed-off-by: Ming Lei <tom.leiming@gmail.com> Cc: stable@kernel.org Signed-off-by: Jens Axboe <axboe@fb.com> Conflicts: drivers/block/virtio_blk.c |
||
---|---|---|
.. | ||
aoe | ||
drbd | ||
mtip32xx | ||
paride | ||
rsxx | ||
xen-blkback | ||
zram | ||
DAC960.c | ||
DAC960.h | ||
Kconfig | ||
Makefile | ||
amiflop.c | ||
ataflop.c | ||
brd.c | ||
cciss.c | ||
cciss.h | ||
cciss_cmd.h | ||
cciss_scsi.c | ||
cciss_scsi.h | ||
cpqarray.c | ||
cpqarray.h | ||
cryptoloop.c | ||
floppy.c | ||
hd.c | ||
ida_cmd.h | ||
ida_ioctl.h | ||
loop.c | ||
loop.h | ||
mg_disk.c | ||
nbd.c | ||
null_blk.c | ||
nvme-core.c | ||
nvme-scsi.c | ||
osdblk.c | ||
pktcdvd.c | ||
ps3disk.c | ||
ps3vram.c | ||
rbd.c | ||
rbd_types.h | ||
skd_main.c | ||
skd_s1120.h | ||
smart1,2.h | ||
sunvdc.c | ||
swim.c | ||
swim3.c | ||
swim_asm.S | ||
sx8.c | ||
umem.c | ||
umem.h | ||
virtio_blk.c | ||
xen-blkfront.c | ||
xsysace.c | ||
z2ram.c |