From: Jens Axboe <axboe@kernel.dk>
Date: Wed, 4 Oct 2017 11:22:24 -0600
Subject: [PATCH] blk-mq: document the need to have STARTED and COMPLETED share
Git-commit: fc13457f74dcf054b0d17efb7b94b46fdf17f412
Patch-mainline: v4.15-rc1
References: bsc#1104967,FATE#325924
a byte
For memory ordering guarantees on stores, we need to ensure that
these two bits share the same byte of storage in the unsigned
long. Add a comment as to why, and a BUILD_BUG_ON() to ensure that
we don't violate this requirement.
Suggested-by: Boqun Feng <boqun.feng@gmail.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Acked-by: Hannes Reinecke <hare@suse.com>
---
block/blk-mq.c | 6 ++++++
block/blk.h | 7 +++++++
2 files changed, 13 insertions(+)
diff --git a/block/blk-mq.c b/block/blk-mq.c
index a7b46b754a7c..076cbab9c3e0 100644
--- a/block/blk-mq.c
+++ b/block/blk-mq.c
@@ -2923,6 +2923,12 @@ EXPORT_SYMBOL_GPL(blk_mq_poll);
static int __init blk_mq_init(void)
{
+ /*
+ * See comment in block/blk.h rq_atomic_flags enum
+ */
+ BUILD_BUG_ON((REQ_ATOM_STARTED / BITS_PER_BYTE) !=
+ (REQ_ATOM_COMPLETE / BITS_PER_BYTE));
+
cpuhp_setup_state_multi(CPUHP_BLK_MQ_DEAD, "block/mq:dead", NULL,
blk_mq_hctx_notify_dead);
return 0;
diff --git a/block/blk.h b/block/blk.h
index fda5a4632aba..6ac43dfd68a7 100644
--- a/block/blk.h
+++ b/block/blk.h
@@ -122,8 +122,15 @@ void blk_account_io_done(struct request *req);
* Internal atomic flags for request handling
*/
enum rq_atomic_flags {
+ /*
+ * Keep these two bits first - not because we depend on the
+ * value of them, but we do depend on them being in the same
+ * byte of storage to ensure ordering on writes. Keeping them
+ * first will achieve that nicely.
+ */
REQ_ATOM_COMPLETE = 0,
REQ_ATOM_STARTED,
+
REQ_ATOM_POLL_SLEPT,
};
--
2.16.4