From e525fd89d380c4a94c0d63913a1dd1a593ed25e7 Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Sat, 13 Nov 2010 11:55:17 +0100 Subject: block: make blkdev_get/put() handle exclusive access Over time, block layer has accumulated a set of APIs dealing with bdev open, close, claim and release. * blkdev_get/put() are the primary open and close functions. * bd_claim/release() deal with exclusive open. * open/close_bdev_exclusive() are combination of open and claim and the other way around, respectively. * bd_link/unlink_disk_holder() to create and remove holder/slave symlinks. * open_by_devnum() wraps bdget() + blkdev_get(). The interface is a bit confusing and the decoupling of open and claim makes it impossible to properly guarantee exclusive access as in-kernel open + claim sequence can disturb the existing exclusive open even before the block layer knows the current open if for another exclusive access. Reorganize the interface such that, * blkdev_get() is extended to include exclusive access management. @holder argument is added and, if is @FMODE_EXCL specified, it will gain exclusive access atomically w.r.t. other exclusive accesses. * blkdev_put() is similarly extended. It now takes @mode argument and if @FMODE_EXCL is set, it releases an exclusive access. Also, when the last exclusive claim is released, the holder/slave symlinks are removed automatically. * bd_claim/release() and close_bdev_exclusive() are no longer necessary and either made static or removed. * bd_link_disk_holder() remains the same but bd_unlink_disk_holder() is no longer necessary and removed. * open_bdev_exclusive() becomes a simple wrapper around lookup_bdev() and blkdev_get(). It also has an unexpected extra bdev_read_only() test which probably should be moved into blkdev_get(). * open_by_devnum() is modified to take @holder argument and pass it to blkdev_get(). Most of bdev open/close operations are unified into blkdev_get/put() and most exclusive accesses are tested atomically at the open time (as it should). This cleans up code and removes some, both valid and invalid, but unnecessary all the same, corner cases. open_bdev_exclusive() and open_by_devnum() can use further cleanup - rename to blkdev_get_by_path() and blkdev_get_by_devt() and drop special features. Well, let's leave them for another day. Most conversions are straight-forward. drbd conversion is a bit more involved as there was some reordering, but the logic should stay the same. Signed-off-by: Tejun Heo Acked-by: Neil Brown Acked-by: Ryusuke Konishi Acked-by: Mike Snitzer Acked-by: Philipp Reisner Cc: Peter Osterlund Cc: Martin Schwidefsky Cc: Heiko Carstens Cc: Jan Kara Cc: Andrew Morton Cc: Andreas Dilger Cc: "Theodore Ts'o" Cc: Mark Fasheh Cc: Joel Becker Cc: Alex Elder Cc: Christoph Hellwig Cc: dm-devel@redhat.com Cc: drbd-dev@lists.linbit.com Cc: Leo Chen Cc: Scott Branden Cc: Chris Mason Cc: Steven Whitehouse Cc: Dave Kleikamp Cc: Joern Engel Cc: reiserfs-devel@vger.kernel.org Cc: Alexander Viro --- kernel/power/swap.c | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-) (limited to 'kernel') diff --git a/kernel/power/swap.c b/kernel/power/swap.c index a0e4a86..513a77f1 100644 --- a/kernel/power/swap.c +++ b/kernel/power/swap.c @@ -223,7 +223,7 @@ static int swsusp_swap_check(void) return res; root_swap = res; - res = blkdev_get(hib_resume_bdev, FMODE_WRITE); + res = blkdev_get(hib_resume_bdev, FMODE_WRITE, NULL); if (res) return res; @@ -907,7 +907,8 @@ int swsusp_check(void) { int error; - hib_resume_bdev = open_by_devnum(swsusp_resume_device, FMODE_READ); + hib_resume_bdev = open_by_devnum(swsusp_resume_device, + FMODE_READ, NULL); if (!IS_ERR(hib_resume_bdev)) { set_blocksize(hib_resume_bdev, PAGE_SIZE); clear_page(swsusp_header); -- cgit v1.1 From d4d77629953eabd3c14f6fa5746f6b28babfc55f Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Sat, 13 Nov 2010 11:55:18 +0100 Subject: block: clean up blkdev_get() wrappers and their users After recent blkdev_get() modifications, open_by_devnum() and open_bdev_exclusive() are simple wrappers around blkdev_get(). Replace them with blkdev_get_by_dev() and blkdev_get_by_path(). blkdev_get_by_dev() is identical to open_by_devnum(). blkdev_get_by_path() is slightly different in that it doesn't automatically add %FMODE_EXCL to @mode. All users are converted. Most conversions are mechanical and don't introduce any behavior difference. There are several exceptions. * btrfs now sets FMODE_EXCL in btrfs_device->mode, so there's no reason to OR it explicitly on blkdev_put(). * gfs2, nilfs2 and the generic mount_bdev() now set FMODE_EXCL in sb->s_mode. * With the above changes, sb->s_mode now always should contain FMODE_EXCL. WARN_ON_ONCE() added to kill_block_super() to detect errors. The new blkdev_get_*() functions are with proper docbook comments. While at it, add function description to blkdev_get() too. Signed-off-by: Tejun Heo Cc: Philipp Reisner Cc: Neil Brown Cc: Mike Snitzer Cc: Joern Engel Cc: Chris Mason Cc: Jan Kara Cc: "Theodore Ts'o" Cc: KONISHI Ryusuke Cc: reiserfs-devel@vger.kernel.org Cc: xfs-masters@oss.sgi.com Cc: Alexander Viro --- kernel/power/swap.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'kernel') diff --git a/kernel/power/swap.c b/kernel/power/swap.c index 513a77f1..b019609 100644 --- a/kernel/power/swap.c +++ b/kernel/power/swap.c @@ -907,8 +907,8 @@ int swsusp_check(void) { int error; - hib_resume_bdev = open_by_devnum(swsusp_resume_device, - FMODE_READ, NULL); + hib_resume_bdev = blkdev_get_by_dev(swsusp_resume_device, + FMODE_READ, NULL); if (!IS_ERR(hib_resume_bdev)) { set_blocksize(hib_resume_bdev, PAGE_SIZE); clear_page(swsusp_header); -- cgit v1.1 From d07335e51df0c6dec202d315fc4f1f7e100eec4e Mon Sep 17 00:00:00 2001 From: Mike Snitzer Date: Tue, 16 Nov 2010 12:52:38 +0100 Subject: block: Rename "block_remap" tracepoint to "block_bio_remap" to clarify the event. Signed-off-by: Kiyoshi Ueda Signed-off-by: Jun'ichi Nomura Signed-off-by: Mike Snitzer Signed-off-by: Jens Axboe --- kernel/trace/blktrace.c | 12 ++++++------ 1 file changed, 6 insertions(+), 6 deletions(-) (limited to 'kernel') diff --git a/kernel/trace/blktrace.c b/kernel/trace/blktrace.c index 7b8ec02..2b8e2ee 100644 --- a/kernel/trace/blktrace.c +++ b/kernel/trace/blktrace.c @@ -887,7 +887,7 @@ static void blk_add_trace_split(void *ignore, } /** - * blk_add_trace_remap - Add a trace for a remap operation + * blk_add_trace_bio_remap - Add a trace for a bio-remap operation * @ignore: trace callback data parameter (not used) * @q: queue the io is for * @bio: the source bio @@ -899,9 +899,9 @@ static void blk_add_trace_split(void *ignore, * it spans a stripe (or similar). Add a trace for that action. * **/ -static void blk_add_trace_remap(void *ignore, - struct request_queue *q, struct bio *bio, - dev_t dev, sector_t from) +static void blk_add_trace_bio_remap(void *ignore, + struct request_queue *q, struct bio *bio, + dev_t dev, sector_t from) { struct blk_trace *bt = q->blk_trace; struct blk_io_trace_remap r; @@ -1016,7 +1016,7 @@ static void blk_register_tracepoints(void) WARN_ON(ret); ret = register_trace_block_split(blk_add_trace_split, NULL); WARN_ON(ret); - ret = register_trace_block_remap(blk_add_trace_remap, NULL); + ret = register_trace_block_bio_remap(blk_add_trace_bio_remap, NULL); WARN_ON(ret); ret = register_trace_block_rq_remap(blk_add_trace_rq_remap, NULL); WARN_ON(ret); @@ -1025,7 +1025,7 @@ static void blk_register_tracepoints(void) static void blk_unregister_tracepoints(void) { unregister_trace_block_rq_remap(blk_add_trace_rq_remap, NULL); - unregister_trace_block_remap(blk_add_trace_remap, NULL); + unregister_trace_block_bio_remap(blk_add_trace_bio_remap, NULL); unregister_trace_block_split(blk_add_trace_split, NULL); unregister_trace_block_unplug_io(blk_add_trace_unplug_io, NULL); unregister_trace_block_unplug_timer(blk_add_trace_unplug_timer, NULL); -- cgit v1.1 From 23036f1a340beec19cc451ba9719526c4ffb3a57 Mon Sep 17 00:00:00 2001 From: Mathieu Desnoyers Date: Fri, 7 Jan 2011 08:53:58 +0100 Subject: blktrace: add missing probe argument to block_bio_complete blktrace.c block bio complete callback needs to gain a new argument to reflect the newly added "error" tracepoint argument. This is needed to match the new block_bio_complete TRACE_EVENT as of commit de983a7bfcb7c020901ca6e2314cf55a4207ab5a. Signed-off-by: Mathieu Desnoyers CC: Jeff Moyer CC: Steven Rostedt CC: Frederic Weisbecker CC: Ingo Molnar CC: Thomas Gleixner CC: Li Zefan Signed-off-by: Jens Axboe --- kernel/trace/blktrace.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) (limited to 'kernel') diff --git a/kernel/trace/blktrace.c b/kernel/trace/blktrace.c index 2b8e2ee..fab31af 100644 --- a/kernel/trace/blktrace.c +++ b/kernel/trace/blktrace.c @@ -782,7 +782,8 @@ static void blk_add_trace_bio_bounce(void *ignore, } static void blk_add_trace_bio_complete(void *ignore, - struct request_queue *q, struct bio *bio) + struct request_queue *q, struct bio *bio, + int error) { blk_add_trace_bio(q, bio, BLK_TA_COMPLETE); } -- cgit v1.1 From 797a455d2c682476c3797dbfecf5bf84c1e3b9d3 Mon Sep 17 00:00:00 2001 From: Jens Axboe Date: Mon, 10 Jan 2011 09:06:44 +0100 Subject: block: ensure that completion error gets properly traced We normally just use the BIO_UPTODATE flag to signal 0/-EIO. If we have more information available, we should pass that along to the trace output. Signed-off-by: Jens Axboe --- kernel/trace/blktrace.c | 22 +++++++++++++--------- 1 file changed, 13 insertions(+), 9 deletions(-) (limited to 'kernel') diff --git a/kernel/trace/blktrace.c b/kernel/trace/blktrace.c index fab31af..153562d 100644 --- a/kernel/trace/blktrace.c +++ b/kernel/trace/blktrace.c @@ -758,54 +758,58 @@ static void blk_add_trace_rq_complete(void *ignore, * @q: queue the io is for * @bio: the source bio * @what: the action + * @error: error, if any * * Description: * Records an action against a bio. Will log the bio offset + size. * **/ static void blk_add_trace_bio(struct request_queue *q, struct bio *bio, - u32 what) + u32 what, int error) { struct blk_trace *bt = q->blk_trace; if (likely(!bt)) return; + if (!error && !bio_flagged(bio, BIO_UPTODATE)) + error = EIO; + __blk_add_trace(bt, bio->bi_sector, bio->bi_size, bio->bi_rw, what, - !bio_flagged(bio, BIO_UPTODATE), 0, NULL); + error, 0, NULL); } static void blk_add_trace_bio_bounce(void *ignore, struct request_queue *q, struct bio *bio) { - blk_add_trace_bio(q, bio, BLK_TA_BOUNCE); + blk_add_trace_bio(q, bio, BLK_TA_BOUNCE, 0); } static void blk_add_trace_bio_complete(void *ignore, struct request_queue *q, struct bio *bio, int error) { - blk_add_trace_bio(q, bio, BLK_TA_COMPLETE); + blk_add_trace_bio(q, bio, BLK_TA_COMPLETE, error); } static void blk_add_trace_bio_backmerge(void *ignore, struct request_queue *q, struct bio *bio) { - blk_add_trace_bio(q, bio, BLK_TA_BACKMERGE); + blk_add_trace_bio(q, bio, BLK_TA_BACKMERGE, 0); } static void blk_add_trace_bio_frontmerge(void *ignore, struct request_queue *q, struct bio *bio) { - blk_add_trace_bio(q, bio, BLK_TA_FRONTMERGE); + blk_add_trace_bio(q, bio, BLK_TA_FRONTMERGE, 0); } static void blk_add_trace_bio_queue(void *ignore, struct request_queue *q, struct bio *bio) { - blk_add_trace_bio(q, bio, BLK_TA_QUEUE); + blk_add_trace_bio(q, bio, BLK_TA_QUEUE, 0); } static void blk_add_trace_getrq(void *ignore, @@ -813,7 +817,7 @@ static void blk_add_trace_getrq(void *ignore, struct bio *bio, int rw) { if (bio) - blk_add_trace_bio(q, bio, BLK_TA_GETRQ); + blk_add_trace_bio(q, bio, BLK_TA_GETRQ, 0); else { struct blk_trace *bt = q->blk_trace; @@ -828,7 +832,7 @@ static void blk_add_trace_sleeprq(void *ignore, struct bio *bio, int rw) { if (bio) - blk_add_trace_bio(q, bio, BLK_TA_SLEEPRQ); + blk_add_trace_bio(q, bio, BLK_TA_SLEEPRQ, 0); else { struct blk_trace *bt = q->blk_trace; -- cgit v1.1