aboutsummaryrefslogtreecommitdiff
path: root/block/blk-mq.c
diff options
context:
space:
mode:
authorGravatar Linus Torvalds <torvalds@linux-foundation.org> 2023-01-06 13:12:42 -0800
committerGravatar Linus Torvalds <torvalds@linux-foundation.org> 2023-01-06 13:12:42 -0800
commita689b938df39ab513026c53fb7011fd7cd594943 (patch)
tree190d683bba2d6e6f9e769d232cc2412dcbd0541e /block/blk-mq.c
parentMerge tag 'io_uring-2023-01-06' of git://git.kernel.dk/linux (diff)
parentblock: Remove "select SRCU" (diff)
downloadlinux-a689b938df39ab513026c53fb7011fd7cd594943.tar.gz
linux-a689b938df39ab513026c53fb7011fd7cd594943.tar.bz2
linux-a689b938df39ab513026c53fb7011fd7cd594943.zip
Merge tag 'block-2023-01-06' of git://git.kernel.dk/linux
Pull block fixes from Jens Axboe: "The big change here is obviously the revert of the pktcdvd driver removal. Outside of that, just minor tweaks. In detail: - Re-instate the pktcdvd driver, which necessitates adding back bio_copy_data_iter() and the fops->devnode() hook for now (me) - Fix for splitting of a bio marked as NOWAIT, causing either nowait reads or writes to error with EAGAIN even if parts of the IO completed (me) - Fix for ublk, punting management commands to io-wq as they can all easily block for extended periods of time (Ming) - Removal of SRCU dependency for the block layer (Paul)" * tag 'block-2023-01-06' of git://git.kernel.dk/linux: block: Remove "select SRCU" Revert "pktcdvd: remove driver." Revert "block: remove devnode callback from struct block_device_operations" Revert "block: bio_copy_data_iter" ublk: honor IO_URING_F_NONBLOCK for handling control command block: don't allow splitting of a REQ_NOWAIT bio block: handle bio_split_to_limits() NULL return
Diffstat (limited to 'block/blk-mq.c')
-rw-r--r--block/blk-mq.c5
1 files changed, 4 insertions, 1 deletions
diff --git a/block/blk-mq.c b/block/blk-mq.c
index c5cf0dbca1db..2c49b4151da1 100644
--- a/block/blk-mq.c
+++ b/block/blk-mq.c
@@ -2951,8 +2951,11 @@ void blk_mq_submit_bio(struct bio *bio)
blk_status_t ret;
bio = blk_queue_bounce(bio, q);
- if (bio_may_exceed_limits(bio, &q->limits))
+ if (bio_may_exceed_limits(bio, &q->limits)) {
bio = __bio_split_to_limits(bio, &q->limits, &nr_segs);
+ if (!bio)
+ return;
+ }
if (!bio_integrity_prep(bio))
return;