blkcg: update blkcg_print_stat() to handle larger outputs
authorTejun Heo <tj@kernel.org>
Thu, 13 Jun 2019 22:30:38 +0000 (15:30 -0700)
committerJens Axboe <axboe@kernel.dk>
Sat, 15 Jun 2019 16:39:37 +0000 (10:39 -0600)
commitf539da82f2158916e154d206054e0efd5df7ab61
tree38d5600b740b77c771b1f7893fb29bf5b7ac67cc
parent5de0073fcd50cc1f150895a7bb04d3cf8067b1d7
blkcg: update blkcg_print_stat() to handle larger outputs

Depending on the number of devices, blkcg stats can go over the
default seqfile buf size.  seqfile normally retries with a larger
buffer but since the ->pd_stat() addition, blkcg_print_stat() doesn't
tell seqfile that overflow has happened and the output gets printed
truncated.  Fix it by calling seq_commit() w/ -1 on possible
overflows.

Signed-off-by: Tejun Heo <tj@kernel.org>
Fixes: 903d23f0a354 ("blk-cgroup: allow controllers to output their own stats")
Cc: stable@vger.kernel.org # v4.19+
Cc: Josef Bacik <jbacik@fb.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
block/blk-cgroup.c