From: Coly Li <colyli@suse.de>
To: linux-bcache@vger.kernel.org
Cc: linux-block@vger.kernel.org, mlyle@lyle.org,
tang.junhui@zte.com.cn, Coly Li <colyli@suse.de>
Subject: [PATCH v1 08/10] bcache: fix misleading error message in bch_count_io_errors()
Date: Wed, 3 Jan 2018 22:03:23 +0800 [thread overview]
Message-ID: <20180103140325.63175-9-colyli@suse.de> (raw)
In-Reply-To: <20180103140325.63175-1-colyli@suse.de>
Bcache only does recoverable I/O for read operations by calling
cached_dev_read_error(). For write opertions there is no I/O recovery for
failed requests.
But in bch_count_io_errors() no matter read or write I/Os, before errors
counter reaches io error limit, pr_err() always prints "IO error on %,
recoverying". For write requests this information is misleading, because
there is no I/O recovery at all.
This patch adds a parameter 'is_read' to bch_count_io_errors(), and only
prints "recovering" by pr_err() when the bio direction is READ.
Signed-off-by: Coly Li <colyli@suse.de>
---
drivers/md/bcache/bcache.h | 2 +-
drivers/md/bcache/io.c | 13 +++++++++----
drivers/md/bcache/super.c | 4 +++-
drivers/md/bcache/writeback.c | 4 +++-
4 files changed, 16 insertions(+), 7 deletions(-)
diff --git a/drivers/md/bcache/bcache.h b/drivers/md/bcache/bcache.h
index a31dc3737dae..c53f312b2216 100644
--- a/drivers/md/bcache/bcache.h
+++ b/drivers/md/bcache/bcache.h
@@ -855,7 +855,7 @@ static inline void wake_up_allocators(struct cache_set *c)
/* Forward declarations */
-void bch_count_io_errors(struct cache *, blk_status_t, const char *);
+void bch_count_io_errors(struct cache *, blk_status_t, int, const char *);
void bch_bbio_count_io_errors(struct cache_set *, struct bio *,
blk_status_t, const char *);
void bch_bbio_endio(struct cache_set *, struct bio *, blk_status_t,
diff --git a/drivers/md/bcache/io.c b/drivers/md/bcache/io.c
index fac97ec2d0e2..a783c5a41ff1 100644
--- a/drivers/md/bcache/io.c
+++ b/drivers/md/bcache/io.c
@@ -51,7 +51,10 @@ void bch_submit_bbio(struct bio *bio, struct cache_set *c,
/* IO errors */
-void bch_count_io_errors(struct cache *ca, blk_status_t error, const char *m)
+void bch_count_io_errors(struct cache *ca,
+ blk_status_t error,
+ int is_read,
+ const char *m)
{
/*
* The halflife of an error is:
@@ -94,8 +97,9 @@ void bch_count_io_errors(struct cache *ca, blk_status_t error, const char *m)
errors >>= IO_ERROR_SHIFT;
if (errors < ca->set->error_limit)
- pr_err("%s: IO error on %s, recovering",
- bdevname(ca->bdev, buf), m);
+ pr_err("%s: IO error on %s%s",
+ bdevname(ca->bdev, buf), m,
+ is_read ? ", recovering." : ".");
else
bch_cache_set_error(ca->set,
"%s: too many IO errors %s",
@@ -108,6 +112,7 @@ void bch_bbio_count_io_errors(struct cache_set *c, struct bio *bio,
{
struct bbio *b = container_of(bio, struct bbio, bio);
struct cache *ca = PTR_CACHE(c, &b->key, 0);
+ int is_read = (bio_data_dir(bio) == READ ? 1 : 0);
unsigned threshold = op_is_write(bio_op(bio))
? c->congested_write_threshold_us
@@ -129,7 +134,7 @@ void bch_bbio_count_io_errors(struct cache_set *c, struct bio *bio,
atomic_inc(&c->congested);
}
- bch_count_io_errors(ca, error, m);
+ bch_count_io_errors(ca, error, is_read, m);
}
void bch_bbio_endio(struct cache_set *c, struct bio *bio,
diff --git a/drivers/md/bcache/super.c b/drivers/md/bcache/super.c
index 02d9d7110769..bbe911847eea 100644
--- a/drivers/md/bcache/super.c
+++ b/drivers/md/bcache/super.c
@@ -274,7 +274,9 @@ static void write_super_endio(struct bio *bio)
{
struct cache *ca = bio->bi_private;
- bch_count_io_errors(ca, bio->bi_status, "writing superblock");
+ /* is_read = 0 */
+ bch_count_io_errors(ca, bio->bi_status, 0,
+ "writing superblock");
closure_put(&ca->set->sb_write);
}
diff --git a/drivers/md/bcache/writeback.c b/drivers/md/bcache/writeback.c
index ab2ac3d72393..e58f9be5ae43 100644
--- a/drivers/md/bcache/writeback.c
+++ b/drivers/md/bcache/writeback.c
@@ -228,8 +228,10 @@ static void read_dirty_endio(struct bio *bio)
struct keybuf_key *w = bio->bi_private;
struct dirty_io *io = w->private;
+ /* is_read = 1 */
bch_count_io_errors(PTR_CACHE(io->dc->disk.c, &w->key, 0),
- bio->bi_status, "reading dirty data from cache");
+ bio->bi_status, 1,
+ "reading dirty data from cache");
dirty_endio(bio);
}
--
2.15.1
next prev parent reply other threads:[~2018-01-03 14:04 UTC|newest]
Thread overview: 34+ messages / expand[flat|nested] mbox.gz Atom feed top
2018-01-03 14:03 [PATCH v1 00/10] cache device failure handling improvement Coly Li
2018-01-03 14:03 ` [PATCH v1 01/10] bcache: exit bch_writeback_thread() with proper task state Coly Li
2018-01-03 17:08 ` Michael Lyle
2018-01-05 17:05 ` Coly Li
2018-01-05 17:09 ` Michael Lyle
2018-01-08 7:09 ` Hannes Reinecke
2018-01-08 13:50 ` Coly Li
2018-01-03 14:03 ` [PATCH v1 02/10] bcache: set task properly in allocator_wait() Coly Li
2018-01-03 17:09 ` Michael Lyle
2018-01-05 17:11 ` Coly Li
2018-01-08 7:10 ` Hannes Reinecke
2018-01-03 14:03 ` [PATCH v1 03/10] bcache: reduce cache_set devices iteration by devices_max_used Coly Li
2018-01-03 17:11 ` Michael Lyle
2018-01-08 7:12 ` Hannes Reinecke
2018-01-03 14:03 ` [PATCH v1 04/10] bcache: fix cached_dev->count usage for bch_cache_set_error() Coly Li
2018-01-08 7:16 ` Hannes Reinecke
2018-01-03 14:03 ` [PATCH v1 05/10] bcache: stop dc->writeback_rate_update if cache set is stopping Coly Li
2018-01-08 7:22 ` Hannes Reinecke
2018-01-08 16:01 ` Coly Li
2018-01-03 14:03 ` [PATCH v1 06/10] bcache: stop dc->writeback_rate_update, dc->writeback_thread earlier Coly Li
2018-01-08 7:25 ` Hannes Reinecke
2018-01-03 14:03 ` [PATCH v1 07/10] bcache: set error_limit correctly Coly Li
2018-01-08 7:26 ` Hannes Reinecke
2018-01-03 14:03 ` Coly Li [this message]
2018-01-03 17:14 ` [PATCH v1 08/10] bcache: fix misleading error message in bch_count_io_errors() Michael Lyle
2018-01-08 7:27 ` Hannes Reinecke
2018-01-03 14:03 ` [PATCH v1 09/10] bcache: add io_disable to struct cache_set Coly Li
2018-01-08 7:30 ` Hannes Reinecke
2018-01-03 14:03 ` [PATCH v1 10/10] bcache: stop all attached bcache devices for a retired cache set Coly Li
2018-01-08 7:31 ` Hannes Reinecke
2018-01-03 17:07 ` [PATCH v1 00/10] cache device failure handling improvement Michael Lyle
2018-01-04 2:20 ` Coly Li
2018-01-04 17:46 ` Michael Lyle
2018-01-05 4:04 ` Coly Li
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=20180103140325.63175-9-colyli@suse.de \
--to=colyli@suse.de \
--cc=linux-bcache@vger.kernel.org \
--cc=linux-block@vger.kernel.org \
--cc=mlyle@lyle.org \
--cc=tang.junhui@zte.com.cn \
/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