nvme: fix race condition in split bio completion path

Fixes race condition observed under following circumstances:

1) I/O split on 128KB boundary with Intel NVMe controller.
   Current Intel controllers produce better latency when
   I/Os do not span a 128KB boundary - even if the I/O size
   itself is less than 128KB.
2) Per-CPU I/O queues are enabled.
3) Child I/Os are submitted on different submission queues.
4) Interrupts for child I/O completions occur almost
   simultaneously.
5) ithread for child I/O A increments bio_inbed, then
   immediately is preempted (rendezvous IPI, higher priority
   interrupt).
6) ithread for child I/O B increments bio_inbed, then completes
   parent bio since all children are now completed.
7) parent bio is freed, and immediately reallocated for a VFS
   or gpart bio (including setting bio_children to 1 and
   clearing bio_driver1).
8) ithread for child I/O A resumes processing.  bio_children
   for what it thinks is the parent bio is set to 1, so it
   thinks it needs to complete the parent bio.

Result is either calling a NULL callback function, or double freeing
the bio to its uma zone.

PR:		203746
Reported by:	Drew Gallatin <gallatin@netflix.com>,
		Marc Goroff <mgoroff@quorum.net>
Tested by:	Drew Gallatin <gallatin@netflix.com>
MFC after:	3 days
Sponsored by:	Intel
This commit is contained in:
jimharris 2015-10-30 16:06:34 +00:00
parent e52d6f1e4a
commit f472a0a088

View File

@ -239,7 +239,7 @@ static void
nvme_bio_child_inbed(struct bio *parent, int bio_error)
{
struct nvme_completion parent_cpl;
int inbed;
int children, inbed;
if (bio_error != 0) {
parent->bio_flags |= BIO_ERROR;
@ -248,10 +248,13 @@ nvme_bio_child_inbed(struct bio *parent, int bio_error)
/*
* atomic_fetchadd will return value before adding 1, so we still
* must add 1 to get the updated inbed number.
* must add 1 to get the updated inbed number. Save bio_children
* before incrementing to guard against race conditions when
* two children bios complete on different queues.
*/
children = atomic_load_acq_int(&parent->bio_children);
inbed = atomic_fetchadd_int(&parent->bio_inbed, 1) + 1;
if (inbed == parent->bio_children) {
if (inbed == children) {
bzero(&parent_cpl, sizeof(parent_cpl));
if (parent->bio_flags & BIO_ERROR)
parent_cpl.status.sc = NVME_SC_DATA_TRANSFER_ERROR;