diff options
| author | Tejun Heo <tj@kernel.org> | 2011-04-21 22:54:46 +0400 |
|---|---|---|
| committer | Jens Axboe <jaxboe@fusionio.com> | 2011-04-21 22:54:46 +0400 |
| commit | d4dc210f69bcb0b4bef5a83b1c323817be89bad1 (patch) | |
| tree | 58774a27fa6faff1260a416a86d2c121f0013431 /fs/block_dev.c | |
| parent | 1196f8b814f32cd04df334abf47648c2a9fd8324 (diff) | |
| download | linux-d4dc210f69bcb0b4bef5a83b1c323817be89bad1.tar.xz | |
block: don't block events on excl write for non-optical devices
Disk event code automatically blocks events on excl write. This is
primarily to avoid issuing polling commands while burning is in
progress. This behavior doesn't fit other types of devices with
removeable media where polling commands don't have adverse side
effects and door locking usually doesn't exist.
This patch introduces new genhd flag which controls the auto-blocking
behavior and uses it to enable auto-blocking only on optical devices.
Note for stable: 2.6.38 and later only
Cc: stable@kernel.org
Signed-off-by: Tejun Heo <tj@kernel.org>
Reported-by: Kay Sievers <kay.sievers@vrfy.org>
Signed-off-by: Jens Axboe <jaxboe@fusionio.com>
Diffstat (limited to 'fs/block_dev.c')
| -rw-r--r-- | fs/block_dev.c | 17 |
1 files changed, 10 insertions, 7 deletions
diff --git a/fs/block_dev.c b/fs/block_dev.c index 257b00e98428..d7c2e0fddc6f 100644 --- a/fs/block_dev.c +++ b/fs/block_dev.c @@ -1237,6 +1237,8 @@ int blkdev_get(struct block_device *bdev, fmode_t mode, void *holder) res = __blkdev_get(bdev, mode, 0); if (whole) { + struct gendisk *disk = whole->bd_disk; + /* finish claiming */ mutex_lock(&bdev->bd_mutex); spin_lock(&bdev_lock); @@ -1263,15 +1265,16 @@ int blkdev_get(struct block_device *bdev, fmode_t mode, void *holder) spin_unlock(&bdev_lock); /* - * Block event polling for write claims. Any write - * holder makes the write_holder state stick until all - * are released. This is good enough and tracking - * individual writeable reference is too fragile given - * the way @mode is used in blkdev_get/put(). + * Block event polling for write claims if requested. Any + * write holder makes the write_holder state stick until + * all are released. This is good enough and tracking + * individual writeable reference is too fragile given the + * way @mode is used in blkdev_get/put(). */ - if (!res && (mode & FMODE_WRITE) && !bdev->bd_write_holder) { + if ((disk->flags & GENHD_FL_BLOCK_EVENTS_ON_EXCL_WRITE) && + !res && (mode & FMODE_WRITE) && !bdev->bd_write_holder) { bdev->bd_write_holder = true; - disk_block_events(bdev->bd_disk); + disk_block_events(disk); } mutex_unlock(&bdev->bd_mutex); |