From: Jens Axboe <axboe@kernel.dk>
To: linux-block@vger.kernel.org
Cc: Filipp.Mikoian@acronis.com, Jens Axboe <axboe@kernel.dk>
Subject: [PATCH 1/2] blk-mq: allow REQ_NOWAIT to return an error inline
Date: Wed, 17 Jul 2019 09:04:44 -0600 [thread overview]
Message-ID: <20190717150445.1131-2-axboe@kernel.dk> (raw)
In-Reply-To: <20190717150445.1131-1-axboe@kernel.dk>
By default, if a caller sets REQ_NOWAIT and we need to block, we'll
return -EAGAIN through the bio->bi_end_io() callback. For some use
cases, this makes it hard to use.
Allow a caller to ask for inline return of errors related to
blocking by also setting REQ_NOWAIT_INLINE.
Signed-off-by: Jens Axboe <axboe@kernel.dk>
---
block/blk-mq.c | 9 +++++++--
include/linux/blk_types.h | 5 ++++-
2 files changed, 11 insertions(+), 3 deletions(-)
diff --git a/block/blk-mq.c b/block/blk-mq.c
index b038ec680e84..ac827e7e3bbe 100644
--- a/block/blk-mq.c
+++ b/block/blk-mq.c
@@ -1960,9 +1960,14 @@ static blk_qc_t blk_mq_make_request(struct request_queue *q, struct bio *bio)
rq = blk_mq_get_request(q, bio, &data);
if (unlikely(!rq)) {
rq_qos_cleanup(q, bio);
- if (bio->bi_opf & REQ_NOWAIT)
+
+ if (bio->bi_opf & REQ_NOWAIT_INLINE) {
+ cookie = BLK_QC_T_EAGAIN;
+ } else if (bio->bi_opf & REQ_NOWAIT) {
bio_wouldblock_error(bio);
- return BLK_QC_T_NONE;
+ cookie = BLK_QC_T_NONE;
+ }
+ return cookie;
}
trace_block_getrq(q, bio, bio->bi_opf);
diff --git a/include/linux/blk_types.h b/include/linux/blk_types.h
index feff3fe4467e..1b1fa1557e68 100644
--- a/include/linux/blk_types.h
+++ b/include/linux/blk_types.h
@@ -311,6 +311,7 @@ enum req_flag_bits {
__REQ_RAHEAD, /* read ahead, can fail anytime */
__REQ_BACKGROUND, /* background IO */
__REQ_NOWAIT, /* Don't wait if request will block */
+ __REQ_NOWAIT_INLINE, /* Return would-block error inline */
/*
* When a shared kthread needs to issue a bio for a cgroup, doing
* so synchronously can lead to priority inversions as the kthread
@@ -345,6 +346,7 @@ enum req_flag_bits {
#define REQ_RAHEAD (1ULL << __REQ_RAHEAD)
#define REQ_BACKGROUND (1ULL << __REQ_BACKGROUND)
#define REQ_NOWAIT (1ULL << __REQ_NOWAIT)
+#define REQ_NOWAIT_INLINE (1ULL << __REQ_NOWAIT_INLINE)
#define REQ_CGROUP_PUNT (1ULL << __REQ_CGROUP_PUNT)
#define REQ_NOUNMAP (1ULL << __REQ_NOUNMAP)
@@ -418,12 +420,13 @@ static inline int op_stat_group(unsigned int op)
typedef unsigned int blk_qc_t;
#define BLK_QC_T_NONE -1U
+#define BLK_QC_T_EAGAIN -2U
#define BLK_QC_T_SHIFT 16
#define BLK_QC_T_INTERNAL (1U << 31)
static inline bool blk_qc_t_valid(blk_qc_t cookie)
{
- return cookie != BLK_QC_T_NONE;
+ return cookie != BLK_QC_T_NONE && cookie != BLK_QC_T_EAGAIN;
}
static inline unsigned int blk_qc_t_to_queue_num(blk_qc_t cookie)
--
2.17.1
next parent reply other threads:[~2019-07-17 15:04 UTC|newest]
Thread overview: 2+ messages / expand[flat|nested] mbox.gz Atom feed top
[not found] <20190717150445.1131-1-axboe@kernel.dk>
2019-07-17 15:04 ` Jens Axboe [this message]
2019-07-17 15:04 ` [PATCH 2/2] block: properly handle IOCB_NOWAIT for async O_DIRECT IO Jens Axboe
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20190717150445.1131-2-axboe@kernel.dk \
--to=axboe@kernel.dk \
--cc="[RFC PATCHSET 0/2]"@vger.kernel.org \
--cc=Filipp.Mikoian@acronis.com \
--cc=Fix@vger.kernel.org \
--cc=IOCB_NOWAIT@vger.kernel.org \
--cc=O_DIRECT@vger.kernel.org \
--cc=blocking@vger.kernel.org \
--cc=linux-block@vger.kernel.org \
--cc=with@vger.kernel.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).