From 183cfb5720dfc393641b87710ce78561af3db6cd Mon Sep 17 00:00:00 2001 From: Wei Yongjun Date: Fri, 22 Mar 2013 08:59:19 -0600 Subject: loop: fix error return code in loop_add() Fix to return a negative error code from the error handling case, as returned elsewhere in this function. Signed-off-by: Wei Yongjun Signed-off-by: Jens Axboe --- drivers/block/loop.c | 1 + 1 file changed, 1 insertion(+) (limited to 'drivers/block/loop.c') diff --git a/drivers/block/loop.c b/drivers/block/loop.c index 747bb2af69d..ee13a82f3f5 100644 --- a/drivers/block/loop.c +++ b/drivers/block/loop.c @@ -1623,6 +1623,7 @@ static int loop_add(struct loop_device **l, int i) goto out_free_dev; i = err; + err = -ENOMEM; lo->lo_queue = blk_alloc_queue(GFP_KERNEL); if (!lo->lo_queue) goto out_free_dev; -- cgit v1.2.3-70-g09d2 From 8761a3dc1f07b163414e2215a2cadbb4cfe2a107 Mon Sep 17 00:00:00 2001 From: Phillip Susi Date: Fri, 22 Mar 2013 12:21:53 -0600 Subject: loop: cleanup partitions when detaching loop device Any partitions added by user space to the loop device were being left in place after detaching the loop device. This was because the detach path issued a BLKRRPART to clean up partitions if LO_FLAGS_PARTSCAN was set, meaning that the partitions were auto scanned on attach. Replace this BLKRRPART with code that unconditionally cleans up partitions on detach instead. Signed-off-by: Phillip Susi Modified by Jens to export delete_partition(). Signed-off-by: Jens Axboe --- block/partition-generic.c | 1 + drivers/block/loop.c | 21 +++++++++++++++++++-- 2 files changed, 20 insertions(+), 2 deletions(-) (limited to 'drivers/block/loop.c') diff --git a/block/partition-generic.c b/block/partition-generic.c index 789cdea0589..ae95ee6a58a 100644 --- a/block/partition-generic.c +++ b/block/partition-generic.c @@ -257,6 +257,7 @@ void delete_partition(struct gendisk *disk, int partno) hd_struct_put(part); } +EXPORT_SYMBOL(delete_partition); static ssize_t whole_disk_show(struct device *dev, struct device_attribute *attr, char *buf) diff --git a/drivers/block/loop.c b/drivers/block/loop.c index ee13a82f3f5..fe5f6403417 100644 --- a/drivers/block/loop.c +++ b/drivers/block/loop.c @@ -1044,12 +1044,29 @@ static int loop_clr_fd(struct loop_device *lo) lo->lo_state = Lo_unbound; /* This is safe: open() is still holding a reference. */ module_put(THIS_MODULE); - if (lo->lo_flags & LO_FLAGS_PARTSCAN && bdev) - ioctl_by_bdev(bdev, BLKRRPART, 0); lo->lo_flags = 0; if (!part_shift) lo->lo_disk->flags |= GENHD_FL_NO_PART_SCAN; mutex_unlock(&lo->lo_ctl_mutex); + + /* + * Remove all partitions, since BLKRRPART won't remove user + * added partitions when max_part=0 + */ + if (bdev) { + struct disk_part_iter piter; + struct hd_struct *part; + + mutex_lock_nested(&bdev->bd_mutex, 1); + invalidate_partition(bdev->bd_disk, 0); + disk_part_iter_init(&piter, bdev->bd_disk, + DISK_PITER_INCL_EMPTY); + while ((part = disk_part_iter_next(&piter))) + delete_partition(bdev->bd_disk, part->partno); + disk_part_iter_exit(&piter); + mutex_unlock(&bdev->bd_mutex); + } + /* * Need not hold lo_ctl_mutex to fput backing file. * Calling fput holding lo_ctl_mutex triggers a circular -- cgit v1.2.3-70-g09d2 From c1681bf8a7b1b98edee8b862a42c19c4e53205fd Mon Sep 17 00:00:00 2001 From: Anatol Pomozov Date: Mon, 1 Apr 2013 09:47:56 -0700 Subject: loop: prevent bdev freeing while device in use struct block_device lifecycle is defined by its inode (see fs/block_dev.c) - block_device allocated first time we access /dev/loopXX and deallocated on bdev_destroy_inode. When we create the device "losetup /dev/loopXX afile" we want that block_device stay alive until we destroy the loop device with "losetup -d". But because we do not hold /dev/loopXX inode its counter goes 0, and inode/bdev can be destroyed at any moment. Usually it happens at memory pressure or when user drops inode cache (like in the test below). When later in loop_clr_fd() we want to use bdev we have use-after-free error with following stack: BUG: unable to handle kernel NULL pointer dereference at 0000000000000280 bd_set_size+0x10/0xa0 loop_clr_fd+0x1f8/0x420 [loop] lo_ioctl+0x200/0x7e0 [loop] lo_compat_ioctl+0x47/0xe0 [loop] compat_blkdev_ioctl+0x341/0x1290 do_filp_open+0x42/0xa0 compat_sys_ioctl+0xc1/0xf20 do_sys_open+0x16e/0x1d0 sysenter_dispatch+0x7/0x1a To prevent use-after-free we need to grab the device in loop_set_fd() and put it later in loop_clr_fd(). The issue is reprodusible on current Linus head and v3.3. Here is the test: dd if=/dev/zero of=loop.file bs=1M count=1 while [ true ]; do losetup /dev/loop0 loop.file echo 2 > /proc/sys/vm/drop_caches losetup -d /dev/loop0 done [ Doing bdgrab/bput in loop_set_fd/loop_clr_fd is safe, because every time we call loop_set_fd() we check that loop_device->lo_state is Lo_unbound and set it to Lo_bound If somebody will try to set_fd again it will get EBUSY. And if we try to loop_clr_fd() on unbound loop device we'll get ENXIO. loop_set_fd/loop_clr_fd (and any other loop ioctl) is called under loop_device->lo_ctl_mutex. ] Signed-off-by: Anatol Pomozov Cc: Al Viro Signed-off-by: Linus Torvalds --- drivers/block/loop.c | 9 ++++++++- fs/block_dev.c | 1 + 2 files changed, 9 insertions(+), 1 deletion(-) (limited to 'drivers/block/loop.c') diff --git a/drivers/block/loop.c b/drivers/block/loop.c index fe5f6403417..2c127f9c3f3 100644 --- a/drivers/block/loop.c +++ b/drivers/block/loop.c @@ -922,6 +922,11 @@ static int loop_set_fd(struct loop_device *lo, fmode_t mode, lo->lo_flags |= LO_FLAGS_PARTSCAN; if (lo->lo_flags & LO_FLAGS_PARTSCAN) ioctl_by_bdev(bdev, BLKRRPART, 0); + + /* Grab the block_device to prevent its destruction after we + * put /dev/loopXX inode. Later in loop_clr_fd() we bdput(bdev). + */ + bdgrab(bdev); return 0; out_clr: @@ -1031,8 +1036,10 @@ static int loop_clr_fd(struct loop_device *lo) memset(lo->lo_encrypt_key, 0, LO_KEY_SIZE); memset(lo->lo_crypt_name, 0, LO_NAME_SIZE); memset(lo->lo_file_name, 0, LO_NAME_SIZE); - if (bdev) + if (bdev) { + bdput(bdev); invalidate_bdev(bdev); + } set_capacity(lo->lo_disk, 0); loop_sysfs_exit(lo); if (bdev) { diff --git a/fs/block_dev.c b/fs/block_dev.c index aea605c98ba..aae187a7f94 100644 --- a/fs/block_dev.c +++ b/fs/block_dev.c @@ -551,6 +551,7 @@ struct block_device *bdgrab(struct block_device *bdev) ihold(bdev->bd_inode); return bdev; } +EXPORT_SYMBOL(bdgrab); long nr_blockdev_pages(void) { -- cgit v1.2.3-70-g09d2 From c2fccc1c9f7c81700cbac2120a4ad5441dd37004 Mon Sep 17 00:00:00 2001 From: Jens Axboe Date: Mon, 8 Apr 2013 10:12:11 +0200 Subject: Revert "loop: cleanup partitions when detaching loop device" This reverts commit 8761a3dc1f07b163414e2215a2cadbb4cfe2a107. There are situations where the destruction path is called with the bdev->bd_mutex already held, which then deadlocks in loop_clr_fd(). The normal partition cleanup does a trylock() on the mutex, but it'd be nice to have a more bullet proof method in loop. So punt this more involved fix to the next merge window, and just back out this buggy fix for now. Signed-off-by: Jens Axboe --- block/partition-generic.c | 1 - drivers/block/loop.c | 21 ++------------------- 2 files changed, 2 insertions(+), 20 deletions(-) (limited to 'drivers/block/loop.c') diff --git a/block/partition-generic.c b/block/partition-generic.c index ae95ee6a58a..789cdea0589 100644 --- a/block/partition-generic.c +++ b/block/partition-generic.c @@ -257,7 +257,6 @@ void delete_partition(struct gendisk *disk, int partno) hd_struct_put(part); } -EXPORT_SYMBOL(delete_partition); static ssize_t whole_disk_show(struct device *dev, struct device_attribute *attr, char *buf) diff --git a/drivers/block/loop.c b/drivers/block/loop.c index fe5f6403417..ee13a82f3f5 100644 --- a/drivers/block/loop.c +++ b/drivers/block/loop.c @@ -1044,29 +1044,12 @@ static int loop_clr_fd(struct loop_device *lo) lo->lo_state = Lo_unbound; /* This is safe: open() is still holding a reference. */ module_put(THIS_MODULE); + if (lo->lo_flags & LO_FLAGS_PARTSCAN && bdev) + ioctl_by_bdev(bdev, BLKRRPART, 0); lo->lo_flags = 0; if (!part_shift) lo->lo_disk->flags |= GENHD_FL_NO_PART_SCAN; mutex_unlock(&lo->lo_ctl_mutex); - - /* - * Remove all partitions, since BLKRRPART won't remove user - * added partitions when max_part=0 - */ - if (bdev) { - struct disk_part_iter piter; - struct hd_struct *part; - - mutex_lock_nested(&bdev->bd_mutex, 1); - invalidate_partition(bdev->bd_disk, 0); - disk_part_iter_init(&piter, bdev->bd_disk, - DISK_PITER_INCL_EMPTY); - while ((part = disk_part_iter_next(&piter))) - delete_partition(bdev->bd_disk, part->partno); - disk_part_iter_exit(&piter); - mutex_unlock(&bdev->bd_mutex); - } - /* * Need not hold lo_ctl_mutex to fput backing file. * Calling fput holding lo_ctl_mutex triggers a circular -- cgit v1.2.3-70-g09d2