linux/fs/block_dev.c

2108 lines
52 KiB
C
Raw Normal View History

// SPDX-License-Identifier: GPL-2.0-only
/*
* linux/fs/block_dev.c
*
* Copyright (C) 1991, 1992 Linus Torvalds
* Copyright (C) 2001 Andrea Arcangeli <andrea@suse.de> SuSE
*/
#include <linux/init.h>
#include <linux/mm.h>
#include <linux/fcntl.h>
#include <linux/slab.h>
#include <linux/kmod.h>
#include <linux/major.h>
#include <linux/device_cgroup.h>
#include <linux/highmem.h>
#include <linux/blkdev.h>
#include <linux/backing-dev.h>
#include <linux/module.h>
#include <linux/blkpg.h>
#include <linux/magic.h>
#include <linux/buffer_head.h>
#include <linux/swap.h>
#include <linux/pagevec.h>
#include <linux/writeback.h>
#include <linux/mpage.h>
#include <linux/mount.h>
#include <linux/pseudo_fs.h>
#include <linux/uio.h>
#include <linux/namei.h>
#include <linux/log2.h>
#include <linux/cleancache.h>
#include <linux/task_io_accounting_ops.h>
#include <linux/falloc.h>
#include <linux/uaccess.h>
#include <linux/suspend.h>
#include "internal.h"
struct bdev_inode {
struct block_device bdev;
struct inode vfs_inode;
};
static const struct address_space_operations def_blk_aops;
static inline struct bdev_inode *BDEV_I(struct inode *inode)
{
return container_of(inode, struct bdev_inode, vfs_inode);
}
struct block_device *I_BDEV(struct inode *inode)
{
return &BDEV_I(inode)->bdev;
}
EXPORT_SYMBOL(I_BDEV);
static void bdev_write_inode(struct block_device *bdev)
{
struct inode *inode = bdev->bd_inode;
int ret;
spin_lock(&inode->i_lock);
while (inode->i_state & I_DIRTY) {
spin_unlock(&inode->i_lock);
ret = write_inode_now(inode, true);
if (ret) {
char name[BDEVNAME_SIZE];
pr_warn_ratelimited("VFS: Dirty inode writeback failed "
"for block device %s (err=%d).\n",
bdevname(bdev, name), ret);
}
spin_lock(&inode->i_lock);
}
spin_unlock(&inode->i_lock);
}
/* Kill _all_ buffers and pagecache , dirty or not.. */
static void kill_bdev(struct block_device *bdev)
{
struct address_space *mapping = bdev->bd_inode->i_mapping;
dax: support dirty DAX entries in radix tree Add support for tracking dirty DAX entries in the struct address_space radix tree. This tree is already used for dirty page writeback, and it already supports the use of exceptional (non struct page*) entries. In order to properly track dirty DAX pages we will insert new exceptional entries into the radix tree that represent dirty DAX PTE or PMD pages. These exceptional entries will also contain the writeback addresses for the PTE or PMD faults that we can use at fsync/msync time. There are currently two types of exceptional entries (shmem and shadow) that can be placed into the radix tree, and this adds a third. We rely on the fact that only one type of exceptional entry can be found in a given radix tree based on its usage. This happens for free with DAX vs shmem but we explicitly prevent shadow entries from being added to radix trees for DAX mappings. The only shadow entries that would be generated for DAX radix trees would be to track zero page mappings that were created for holes. These pages would receive minimal benefit from having shadow entries, and the choice to have only one type of exceptional entry in a given radix tree makes the logic simpler both in clear_exceptional_entry() and in the rest of DAX. Signed-off-by: Ross Zwisler <ross.zwisler@linux.intel.com> Cc: "H. Peter Anvin" <hpa@zytor.com> Cc: "J. Bruce Fields" <bfields@fieldses.org> Cc: "Theodore Ts'o" <tytso@mit.edu> Cc: Alexander Viro <viro@zeniv.linux.org.uk> Cc: Andreas Dilger <adilger.kernel@dilger.ca> Cc: Dave Chinner <david@fromorbit.com> Cc: Ingo Molnar <mingo@redhat.com> Cc: Jan Kara <jack@suse.com> Cc: Jeff Layton <jlayton@poochiereds.net> Cc: Matthew Wilcox <willy@linux.intel.com> Cc: Thomas Gleixner <tglx@linutronix.de> Cc: Dan Williams <dan.j.williams@intel.com> Cc: Matthew Wilcox <matthew.r.wilcox@intel.com> Cc: Dave Hansen <dave.hansen@linux.intel.com> Cc: Hugh Dickins <hughd@google.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2016-01-22 23:10:40 +00:00
if (mapping->nrpages == 0 && mapping->nrexceptional == 0)
return;
invalidate_bh_lrus();
truncate_inode_pages(mapping, 0);
}
/* Invalidate clean unused buffers and pagecache. */
void invalidate_bdev(struct block_device *bdev)
{
struct address_space *mapping = bdev->bd_inode->i_mapping;
if (mapping->nrpages) {
invalidate_bh_lrus();
lru_add_drain_all(); /* make sure all lru add caches are flushed */
invalidate_mapping_pages(mapping, 0, -1);
}
/* 99% of the time, we don't need to flush the cleancache on the bdev.
* But, for the strange corners, lets be cautious
*/
cleancache_invalidate_inode(mapping);
}
EXPORT_SYMBOL(invalidate_bdev);
/*
* Drop all buffers & page cache for given bdev range. This function bails
* with error if bdev has other exclusive owner (such as filesystem).
*/
int truncate_bdev_range(struct block_device *bdev, fmode_t mode,
loff_t lstart, loff_t lend)
{
struct block_device *claimed_bdev = NULL;
int err;
/*
* If we don't hold exclusive handle for the device, upgrade to it
* while we discard the buffer cache to avoid discarding buffers
* under live filesystem.
*/
if (!(mode & FMODE_EXCL)) {
claimed_bdev = bdev->bd_contains;
err = bd_prepare_to_claim(bdev, claimed_bdev,
truncate_bdev_range);
if (err)
return err;
}
truncate_inode_pages_range(bdev->bd_inode->i_mapping, lstart, lend);
if (claimed_bdev)
bd_abort_claiming(bdev, claimed_bdev, truncate_bdev_range);
return 0;
}
EXPORT_SYMBOL(truncate_bdev_range);
blockdev: Fix livelocks on loop device bd_set_size() updates also block device's block size. This is somewhat unexpected from its name and at this point, only blkdev_open() uses this functionality. Furthermore, this can result in changing block size under a filesystem mounted on a loop device which leads to livelocks inside __getblk_gfp() like: Sending NMI from CPU 0 to CPUs 1: NMI backtrace for cpu 1 CPU: 1 PID: 10863 Comm: syz-executor0 Not tainted 4.18.0-rc5+ #151 Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 01/01/2011 RIP: 0010:__sanitizer_cov_trace_pc+0x3f/0x50 kernel/kcov.c:106 ... Call Trace: init_page_buffers+0x3e2/0x530 fs/buffer.c:904 grow_dev_page fs/buffer.c:947 [inline] grow_buffers fs/buffer.c:1009 [inline] __getblk_slow fs/buffer.c:1036 [inline] __getblk_gfp+0x906/0xb10 fs/buffer.c:1313 __bread_gfp+0x2d/0x310 fs/buffer.c:1347 sb_bread include/linux/buffer_head.h:307 [inline] fat12_ent_bread+0x14e/0x3d0 fs/fat/fatent.c:75 fat_ent_read_block fs/fat/fatent.c:441 [inline] fat_alloc_clusters+0x8ce/0x16e0 fs/fat/fatent.c:489 fat_add_cluster+0x7a/0x150 fs/fat/inode.c:101 __fat_get_block fs/fat/inode.c:148 [inline] ... Trivial reproducer for the problem looks like: truncate -s 1G /tmp/image losetup /dev/loop0 /tmp/image mkfs.ext4 -b 1024 /dev/loop0 mount -t ext4 /dev/loop0 /mnt losetup -c /dev/loop0 l /mnt Fix the problem by moving initialization of a block device block size into a separate function and call it when needed. Thanks to Tetsuo Handa <penguin-kernel@I-love.SAKURA.ne.jp> for help with debugging the problem. Reported-by: syzbot+9933e4476f365f5d5a1b@syzkaller.appspotmail.com Signed-off-by: Jan Kara <jack@suse.cz> Signed-off-by: Jens Axboe <axboe@kernel.dk>
2019-01-14 08:48:10 +00:00
static void set_init_blocksize(struct block_device *bdev)
{
bdev->bd_inode->i_blkbits = blksize_bits(bdev_logical_block_size(bdev));
blockdev: Fix livelocks on loop device bd_set_size() updates also block device's block size. This is somewhat unexpected from its name and at this point, only blkdev_open() uses this functionality. Furthermore, this can result in changing block size under a filesystem mounted on a loop device which leads to livelocks inside __getblk_gfp() like: Sending NMI from CPU 0 to CPUs 1: NMI backtrace for cpu 1 CPU: 1 PID: 10863 Comm: syz-executor0 Not tainted 4.18.0-rc5+ #151 Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 01/01/2011 RIP: 0010:__sanitizer_cov_trace_pc+0x3f/0x50 kernel/kcov.c:106 ... Call Trace: init_page_buffers+0x3e2/0x530 fs/buffer.c:904 grow_dev_page fs/buffer.c:947 [inline] grow_buffers fs/buffer.c:1009 [inline] __getblk_slow fs/buffer.c:1036 [inline] __getblk_gfp+0x906/0xb10 fs/buffer.c:1313 __bread_gfp+0x2d/0x310 fs/buffer.c:1347 sb_bread include/linux/buffer_head.h:307 [inline] fat12_ent_bread+0x14e/0x3d0 fs/fat/fatent.c:75 fat_ent_read_block fs/fat/fatent.c:441 [inline] fat_alloc_clusters+0x8ce/0x16e0 fs/fat/fatent.c:489 fat_add_cluster+0x7a/0x150 fs/fat/inode.c:101 __fat_get_block fs/fat/inode.c:148 [inline] ... Trivial reproducer for the problem looks like: truncate -s 1G /tmp/image losetup /dev/loop0 /tmp/image mkfs.ext4 -b 1024 /dev/loop0 mount -t ext4 /dev/loop0 /mnt losetup -c /dev/loop0 l /mnt Fix the problem by moving initialization of a block device block size into a separate function and call it when needed. Thanks to Tetsuo Handa <penguin-kernel@I-love.SAKURA.ne.jp> for help with debugging the problem. Reported-by: syzbot+9933e4476f365f5d5a1b@syzkaller.appspotmail.com Signed-off-by: Jan Kara <jack@suse.cz> Signed-off-by: Jens Axboe <axboe@kernel.dk>
2019-01-14 08:48:10 +00:00
}
int set_blocksize(struct block_device *bdev, int size)
{
/* Size must be a power of two, and between 512 and PAGE_SIZE */
if (size > PAGE_SIZE || size < 512 || !is_power_of_2(size))
return -EINVAL;
/* Size cannot be smaller than the size supported by the device */
if (size < bdev_logical_block_size(bdev))
return -EINVAL;
/* Don't change the size if it is same as current */
if (bdev->bd_inode->i_blkbits != blksize_bits(size)) {
sync_blockdev(bdev);
bdev->bd_inode->i_blkbits = blksize_bits(size);
kill_bdev(bdev);
}
return 0;
}
EXPORT_SYMBOL(set_blocksize);
int sb_set_blocksize(struct super_block *sb, int size)
{
if (set_blocksize(sb->s_bdev, size))
return 0;
/* If we get here, we know size is power of two
* and it's value is between 512 and PAGE_SIZE */
sb->s_blocksize = size;
sb->s_blocksize_bits = blksize_bits(size);
return sb->s_blocksize;
}
EXPORT_SYMBOL(sb_set_blocksize);
int sb_min_blocksize(struct super_block *sb, int size)
{
int minsize = bdev_logical_block_size(sb->s_bdev);
if (size < minsize)
size = minsize;
return sb_set_blocksize(sb, size);
}
EXPORT_SYMBOL(sb_min_blocksize);
static int
blkdev_get_block(struct inode *inode, sector_t iblock,
struct buffer_head *bh, int create)
{
bh->b_bdev = I_BDEV(inode);
bh->b_blocknr = iblock;
set_buffer_mapped(bh);
return 0;
}
static struct inode *bdev_file_inode(struct file *file)
{
return file->f_mapping->host;
}
static unsigned int dio_bio_write_op(struct kiocb *iocb)
{
unsigned int op = REQ_OP_WRITE | REQ_SYNC | REQ_IDLE;
/* avoid the need for a I/O completion work item */
if (iocb->ki_flags & IOCB_DSYNC)
op |= REQ_FUA;
return op;
}
#define DIO_INLINE_BIO_VECS 4
static void blkdev_bio_end_io_simple(struct bio *bio)
{
struct task_struct *waiter = bio->bi_private;
WRITE_ONCE(bio->bi_private, NULL);
blk_wake_io_task(waiter);
}
static ssize_t
__blkdev_direct_IO_simple(struct kiocb *iocb, struct iov_iter *iter,
int nr_pages)
{
struct file *file = iocb->ki_filp;
struct block_device *bdev = I_BDEV(bdev_file_inode(file));
struct bio_vec inline_vecs[DIO_INLINE_BIO_VECS], *vecs;
loff_t pos = iocb->ki_pos;
bool should_dirty = false;
struct bio bio;
ssize_t ret;
blk_qc_t qc;
if ((pos | iov_iter_alignment(iter)) &
(bdev_logical_block_size(bdev) - 1))
return -EINVAL;
if (nr_pages <= DIO_INLINE_BIO_VECS)
vecs = inline_vecs;
else {
treewide: kmalloc() -> kmalloc_array() The kmalloc() function has a 2-factor argument form, kmalloc_array(). This patch replaces cases of: kmalloc(a * b, gfp) with: kmalloc_array(a * b, gfp) as well as handling cases of: kmalloc(a * b * c, gfp) with: kmalloc(array3_size(a, b, c), gfp) as it's slightly less ugly than: kmalloc_array(array_size(a, b), c, gfp) This does, however, attempt to ignore constant size factors like: kmalloc(4 * 1024, gfp) though any constants defined via macros get caught up in the conversion. Any factors with a sizeof() of "unsigned char", "char", and "u8" were dropped, since they're redundant. The tools/ directory was manually excluded, since it has its own implementation of kmalloc(). The Coccinelle script used for this was: // Fix redundant parens around sizeof(). @@ type TYPE; expression THING, E; @@ ( kmalloc( - (sizeof(TYPE)) * E + sizeof(TYPE) * E , ...) | kmalloc( - (sizeof(THING)) * E + sizeof(THING) * E , ...) ) // Drop single-byte sizes and redundant parens. @@ expression COUNT; typedef u8; typedef __u8; @@ ( kmalloc( - sizeof(u8) * (COUNT) + COUNT , ...) | kmalloc( - sizeof(__u8) * (COUNT) + COUNT , ...) | kmalloc( - sizeof(char) * (COUNT) + COUNT , ...) | kmalloc( - sizeof(unsigned char) * (COUNT) + COUNT , ...) | kmalloc( - sizeof(u8) * COUNT + COUNT , ...) | kmalloc( - sizeof(__u8) * COUNT + COUNT , ...) | kmalloc( - sizeof(char) * COUNT + COUNT , ...) | kmalloc( - sizeof(unsigned char) * COUNT + COUNT , ...) ) // 2-factor product with sizeof(type/expression) and identifier or constant. @@ type TYPE; expression THING; identifier COUNT_ID; constant COUNT_CONST; @@ ( - kmalloc + kmalloc_array ( - sizeof(TYPE) * (COUNT_ID) + COUNT_ID, sizeof(TYPE) , ...) | - kmalloc + kmalloc_array ( - sizeof(TYPE) * COUNT_ID + COUNT_ID, sizeof(TYPE) , ...) | - kmalloc + kmalloc_array ( - sizeof(TYPE) * (COUNT_CONST) + COUNT_CONST, sizeof(TYPE) , ...) | - kmalloc + kmalloc_array ( - sizeof(TYPE) * COUNT_CONST + COUNT_CONST, sizeof(TYPE) , ...) | - kmalloc + kmalloc_array ( - sizeof(THING) * (COUNT_ID) + COUNT_ID, sizeof(THING) , ...) | - kmalloc + kmalloc_array ( - sizeof(THING) * COUNT_ID + COUNT_ID, sizeof(THING) , ...) | - kmalloc + kmalloc_array ( - sizeof(THING) * (COUNT_CONST) + COUNT_CONST, sizeof(THING) , ...) | - kmalloc + kmalloc_array ( - sizeof(THING) * COUNT_CONST + COUNT_CONST, sizeof(THING) , ...) ) // 2-factor product, only identifiers. @@ identifier SIZE, COUNT; @@ - kmalloc + kmalloc_array ( - SIZE * COUNT + COUNT, SIZE , ...) // 3-factor product with 1 sizeof(type) or sizeof(expression), with // redundant parens removed. @@ expression THING; identifier STRIDE, COUNT; type TYPE; @@ ( kmalloc( - sizeof(TYPE) * (COUNT) * (STRIDE) + array3_size(COUNT, STRIDE, sizeof(TYPE)) , ...) | kmalloc( - sizeof(TYPE) * (COUNT) * STRIDE + array3_size(COUNT, STRIDE, sizeof(TYPE)) , ...) | kmalloc( - sizeof(TYPE) * COUNT * (STRIDE) + array3_size(COUNT, STRIDE, sizeof(TYPE)) , ...) | kmalloc( - sizeof(TYPE) * COUNT * STRIDE + array3_size(COUNT, STRIDE, sizeof(TYPE)) , ...) | kmalloc( - sizeof(THING) * (COUNT) * (STRIDE) + array3_size(COUNT, STRIDE, sizeof(THING)) , ...) | kmalloc( - sizeof(THING) * (COUNT) * STRIDE + array3_size(COUNT, STRIDE, sizeof(THING)) , ...) | kmalloc( - sizeof(THING) * COUNT * (STRIDE) + array3_size(COUNT, STRIDE, sizeof(THING)) , ...) | kmalloc( - sizeof(THING) * COUNT * STRIDE + array3_size(COUNT, STRIDE, sizeof(THING)) , ...) ) // 3-factor product with 2 sizeof(variable), with redundant parens removed. @@ expression THING1, THING2; identifier COUNT; type TYPE1, TYPE2; @@ ( kmalloc( - sizeof(TYPE1) * sizeof(TYPE2) * COUNT + array3_size(COUNT, sizeof(TYPE1), sizeof(TYPE2)) , ...) | kmalloc( - sizeof(TYPE1) * sizeof(THING2) * (COUNT) + array3_size(COUNT, sizeof(TYPE1), sizeof(TYPE2)) , ...) | kmalloc( - sizeof(THING1) * sizeof(THING2) * COUNT + array3_size(COUNT, sizeof(THING1), sizeof(THING2)) , ...) | kmalloc( - sizeof(THING1) * sizeof(THING2) * (COUNT) + array3_size(COUNT, sizeof(THING1), sizeof(THING2)) , ...) | kmalloc( - sizeof(TYPE1) * sizeof(THING2) * COUNT + array3_size(COUNT, sizeof(TYPE1), sizeof(THING2)) , ...) | kmalloc( - sizeof(TYPE1) * sizeof(THING2) * (COUNT) + array3_size(COUNT, sizeof(TYPE1), sizeof(THING2)) , ...) ) // 3-factor product, only identifiers, with redundant parens removed. @@ identifier STRIDE, SIZE, COUNT; @@ ( kmalloc( - (COUNT) * STRIDE * SIZE + array3_size(COUNT, STRIDE, SIZE) , ...) | kmalloc( - COUNT * (STRIDE) * SIZE + array3_size(COUNT, STRIDE, SIZE) , ...) | kmalloc( - COUNT * STRIDE * (SIZE) + array3_size(COUNT, STRIDE, SIZE) , ...) | kmalloc( - (COUNT) * (STRIDE) * SIZE + array3_size(COUNT, STRIDE, SIZE) , ...) | kmalloc( - COUNT * (STRIDE) * (SIZE) + array3_size(COUNT, STRIDE, SIZE) , ...) | kmalloc( - (COUNT) * STRIDE * (SIZE) + array3_size(COUNT, STRIDE, SIZE) , ...) | kmalloc( - (COUNT) * (STRIDE) * (SIZE) + array3_size(COUNT, STRIDE, SIZE) , ...) | kmalloc( - COUNT * STRIDE * SIZE + array3_size(COUNT, STRIDE, SIZE) , ...) ) // Any remaining multi-factor products, first at least 3-factor products, // when they're not all constants... @@ expression E1, E2, E3; constant C1, C2, C3; @@ ( kmalloc(C1 * C2 * C3, ...) | kmalloc( - (E1) * E2 * E3 + array3_size(E1, E2, E3) , ...) | kmalloc( - (E1) * (E2) * E3 + array3_size(E1, E2, E3) , ...) | kmalloc( - (E1) * (E2) * (E3) + array3_size(E1, E2, E3) , ...) | kmalloc( - E1 * E2 * E3 + array3_size(E1, E2, E3) , ...) ) // And then all remaining 2 factors products when they're not all constants, // keeping sizeof() as the second factor argument. @@ expression THING, E1, E2; type TYPE; constant C1, C2, C3; @@ ( kmalloc(sizeof(THING) * C2, ...) | kmalloc(sizeof(TYPE) * C2, ...) | kmalloc(C1 * C2 * C3, ...) | kmalloc(C1 * C2, ...) | - kmalloc + kmalloc_array ( - sizeof(TYPE) * (E2) + E2, sizeof(TYPE) , ...) | - kmalloc + kmalloc_array ( - sizeof(TYPE) * E2 + E2, sizeof(TYPE) , ...) | - kmalloc + kmalloc_array ( - sizeof(THING) * (E2) + E2, sizeof(THING) , ...) | - kmalloc + kmalloc_array ( - sizeof(THING) * E2 + E2, sizeof(THING) , ...) | - kmalloc + kmalloc_array ( - (E1) * E2 + E1, E2 , ...) | - kmalloc + kmalloc_array ( - (E1) * (E2) + E1, E2 , ...) | - kmalloc + kmalloc_array ( - E1 * E2 + E1, E2 , ...) ) Signed-off-by: Kees Cook <keescook@chromium.org>
2018-06-12 20:55:00 +00:00
vecs = kmalloc_array(nr_pages, sizeof(struct bio_vec),
GFP_KERNEL);
if (!vecs)
return -ENOMEM;
}
bio_init(&bio, vecs, nr_pages);
bio_set_dev(&bio, bdev);
bio.bi_iter.bi_sector = pos >> 9;
bio.bi_write_hint = iocb->ki_hint;
bio.bi_private = current;
bio.bi_end_io = blkdev_bio_end_io_simple;
bio.bi_ioprio = iocb->ki_ioprio;
ret = bio_iov_iter_get_pages(&bio, iter);
if (unlikely(ret))
goto out;
ret = bio.bi_iter.bi_size;
if (iov_iter_rw(iter) == READ) {
bio.bi_opf = REQ_OP_READ;
if (iter_is_iovec(iter))
should_dirty = true;
} else {
bio.bi_opf = dio_bio_write_op(iocb);
task_io_account_write(ret);
}
if (iocb->ki_flags & IOCB_HIPRI)
bio_set_polled(&bio, iocb);
qc = submit_bio(&bio);
for (;;) {
set_current_state(TASK_UNINTERRUPTIBLE);
if (!READ_ONCE(bio.bi_private))
break;
if (!(iocb->ki_flags & IOCB_HIPRI) ||
!blk_poll(bdev_get_queue(bdev), qc, true))
blk_io_schedule();
}
__set_current_state(TASK_RUNNING);
bio_release_pages(&bio, should_dirty);
if (unlikely(bio.bi_status))
Merge branch 'for-4.13/block' of git://git.kernel.dk/linux-block Pull core block/IO updates from Jens Axboe: "This is the main pull request for the block layer for 4.13. Not a huge round in terms of features, but there's a lot of churn related to some core cleanups. Note this depends on the UUID tree pull request, that Christoph already sent out. This pull request contains: - A series from Christoph, unifying the error/stats codes in the block layer. We now use blk_status_t everywhere, instead of using different schemes for different places. - Also from Christoph, some cleanups around request allocation and IO scheduler interactions in blk-mq. - And yet another series from Christoph, cleaning up how we handle and do bounce buffering in the block layer. - A blk-mq debugfs series from Bart, further improving on the support we have for exporting internal information to aid debugging IO hangs or stalls. - Also from Bart, a series that cleans up the request initialization differences across types of devices. - A series from Goldwyn Rodrigues, allowing the block layer to return failure if we will block and the user asked for non-blocking. - Patch from Hannes for supporting setting loop devices block size to that of the underlying device. - Two series of patches from Javier, fixing various issues with lightnvm, particular around pblk. - A series from me, adding support for write hints. This comes with NVMe support as well, so applications can help guide data placement on flash to improve performance, latencies, and write amplification. - A series from Ming, improving and hardening blk-mq support for stopping/starting and quiescing hardware queues. - Two pull requests for NVMe updates. Nothing major on the feature side, but lots of cleanups and bug fixes. From the usual crew. - A series from Neil Brown, greatly improving the bio rescue set support. Most notably, this kills the bio rescue work queues, if we don't really need them. - Lots of other little bug fixes that are all over the place" * 'for-4.13/block' of git://git.kernel.dk/linux-block: (217 commits) lightnvm: pblk: set line bitmap check under debug lightnvm: pblk: verify that cache read is still valid lightnvm: pblk: add initialization check lightnvm: pblk: remove target using async. I/Os lightnvm: pblk: use vmalloc for GC data buffer lightnvm: pblk: use right metadata buffer for recovery lightnvm: pblk: schedule if data is not ready lightnvm: pblk: remove unused return variable lightnvm: pblk: fix double-free on pblk init lightnvm: pblk: fix bad le64 assignations nvme: Makefile: remove dead build rule blk-mq: map all HWQ also in hyperthreaded system nvmet-rdma: register ib_client to not deadlock in device removal nvme_fc: fix error recovery on link down. nvmet_fc: fix crashes on bad opcodes nvme_fc: Fix crash when nvme controller connection fails. nvme_fc: replace ioabort msleep loop with completion nvme_fc: fix double calls to nvme_cleanup_cmd() nvme-fabrics: verify that a controller returns the correct NQN nvme: simplify nvme_dev_attrs_are_visible ...
2017-07-03 17:34:51 +00:00
ret = blk_status_to_errno(bio.bi_status);
block: provide bio_uninit() free freeing integrity/task associations Wen reports significant memory leaks with DIF and O_DIRECT: "With nvme devive + T10 enabled, On a system it has 256GB and started logging /proc/meminfo & /proc/slabinfo for every minute and in an hour it increased by 15968128 kB or ~15+GB.. Approximately 256 MB / minute leaking. /proc/meminfo | grep SUnreclaim... SUnreclaim: 6752128 kB SUnreclaim: 6874880 kB SUnreclaim: 7238080 kB .... SUnreclaim: 22307264 kB SUnreclaim: 22485888 kB SUnreclaim: 22720256 kB When testcases with T10 enabled call into __blkdev_direct_IO_simple, code doesn't free memory allocated by bio_integrity_alloc. The patch fixes the issue. HTX has been run with +60 hours without failure." Since __blkdev_direct_IO_simple() allocates the bio on the stack, it doesn't go through the regular bio free. This means that any ancillary data allocated with the bio through the stack is not freed. Hence, we can leak the integrity data associated with the bio, if the device is using DIF/DIX. Fix this by providing a bio_uninit() and export it, so that we can use it to free this data. Note that this is a minimal fix for this issue. Any current user of bio's that are allocated outside of bio_alloc_bioset() suffers from this issue, most notably some drivers. We will fix those in a more comprehensive patch for 4.13. This also means that the commit marked as being fixed by this isn't the real culprit, it's just the most obvious one out there. Fixes: 542ff7bf18c6 ("block: new direct I/O implementation") Reported-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Reviewed-by: Christoph Hellwig <hch@lst.de> Signed-off-by: Jens Axboe <axboe@kernel.dk>
2017-06-28 21:30:13 +00:00
out:
if (vecs != inline_vecs)
kfree(vecs);
block: provide bio_uninit() free freeing integrity/task associations Wen reports significant memory leaks with DIF and O_DIRECT: "With nvme devive + T10 enabled, On a system it has 256GB and started logging /proc/meminfo & /proc/slabinfo for every minute and in an hour it increased by 15968128 kB or ~15+GB.. Approximately 256 MB / minute leaking. /proc/meminfo | grep SUnreclaim... SUnreclaim: 6752128 kB SUnreclaim: 6874880 kB SUnreclaim: 7238080 kB .... SUnreclaim: 22307264 kB SUnreclaim: 22485888 kB SUnreclaim: 22720256 kB When testcases with T10 enabled call into __blkdev_direct_IO_simple, code doesn't free memory allocated by bio_integrity_alloc. The patch fixes the issue. HTX has been run with +60 hours without failure." Since __blkdev_direct_IO_simple() allocates the bio on the stack, it doesn't go through the regular bio free. This means that any ancillary data allocated with the bio through the stack is not freed. Hence, we can leak the integrity data associated with the bio, if the device is using DIF/DIX. Fix this by providing a bio_uninit() and export it, so that we can use it to free this data. Note that this is a minimal fix for this issue. Any current user of bio's that are allocated outside of bio_alloc_bioset() suffers from this issue, most notably some drivers. We will fix those in a more comprehensive patch for 4.13. This also means that the commit marked as being fixed by this isn't the real culprit, it's just the most obvious one out there. Fixes: 542ff7bf18c6 ("block: new direct I/O implementation") Reported-by: Wen Xiong <wenxiong@linux.vnet.ibm.com> Reviewed-by: Christoph Hellwig <hch@lst.de> Signed-off-by: Jens Axboe <axboe@kernel.dk>
2017-06-28 21:30:13 +00:00
bio_uninit(&bio);
return ret;
}
struct blkdev_dio {
union {
struct kiocb *iocb;
struct task_struct *waiter;
};
size_t size;
atomic_t ref;
bool multi_bio : 1;
bool should_dirty : 1;
bool is_sync : 1;
struct bio bio;
};
static struct bio_set blkdev_dio_pool;
static int blkdev_iopoll(struct kiocb *kiocb, bool wait)
{
struct block_device *bdev = I_BDEV(kiocb->ki_filp->f_mapping->host);
struct request_queue *q = bdev_get_queue(bdev);
return blk_poll(q, READ_ONCE(kiocb->ki_cookie), wait);
}
static void blkdev_bio_end_io(struct bio *bio)
{
struct blkdev_dio *dio = bio->bi_private;
bool should_dirty = dio->should_dirty;
if (bio->bi_status && !dio->bio.bi_status)
dio->bio.bi_status = bio->bi_status;
if (!dio->multi_bio || atomic_dec_and_test(&dio->ref)) {
if (!dio->is_sync) {
struct kiocb *iocb = dio->iocb;
ssize_t ret;
if (likely(!dio->bio.bi_status)) {
ret = dio->size;
iocb->ki_pos += ret;
} else {
ret = blk_status_to_errno(dio->bio.bi_status);
}
dio->iocb->ki_complete(iocb, ret, 0);
if (dio->multi_bio)
bio_put(&dio->bio);
} else {
struct task_struct *waiter = dio->waiter;
WRITE_ONCE(dio->waiter, NULL);
blk_wake_io_task(waiter);
}
}
if (should_dirty) {
bio_check_pages_dirty(bio);
} else {
bio_release_pages(bio, false);
bio_put(bio);
}
}
static ssize_t
__blkdev_direct_IO(struct kiocb *iocb, struct iov_iter *iter, int nr_pages)
{
struct file *file = iocb->ki_filp;
struct inode *inode = bdev_file_inode(file);
struct block_device *bdev = I_BDEV(inode);
struct blk_plug plug;
struct blkdev_dio *dio;
struct bio *bio;
bool is_poll = (iocb->ki_flags & IOCB_HIPRI) != 0;
bool is_read = (iov_iter_rw(iter) == READ), is_sync;
loff_t pos = iocb->ki_pos;
blk_qc_t qc = BLK_QC_T_NONE;
int ret = 0;
if ((pos | iov_iter_alignment(iter)) &
(bdev_logical_block_size(bdev) - 1))
return -EINVAL;
bio = bio_alloc_bioset(GFP_KERNEL, nr_pages, &blkdev_dio_pool);
dio = container_of(bio, struct blkdev_dio, bio);
dio->is_sync = is_sync = is_sync_kiocb(iocb);
if (dio->is_sync) {
dio->waiter = current;
bio_get(bio);
} else {
dio->iocb = iocb;
}
dio->size = 0;
dio->multi_bio = false;
dio->should_dirty = is_read && iter_is_iovec(iter);
/*
* Don't plug for HIPRI/polled IO, as those should go straight
* to issue
*/
if (!is_poll)
blk_start_plug(&plug);
for (;;) {
bio_set_dev(bio, bdev);
bio->bi_iter.bi_sector = pos >> 9;
bio->bi_write_hint = iocb->ki_hint;
bio->bi_private = dio;
bio->bi_end_io = blkdev_bio_end_io;
bio->bi_ioprio = iocb->ki_ioprio;
block: fix O_DIRECT error handling for bio fragments 0eb6ddfb865c tried to fix this up, but introduced a use-after-free of dio. Additionally, we still had an issue with error handling, as reported by Darrick: "I noticed a regression in xfs/747 (an unreleased xfstest for the xfs_scrub media scanning feature) on 5.3-rc3. I'll condense that down to a simpler reproducer: error-test: 0 209 linear 8:48 0 error-test: 209 1 error error-test: 210 6446894 linear 8:48 210 Basically we have a ~3G /dev/sdd and we set up device mapper to fail IO for sector 209 and to pass the io to the scsi device everywhere else. On 5.3-rc3, performing a directio pread of this range with a < 1M buffer (in other words, a request for fewer than MAX_BIO_PAGES bytes) yields EIO like you'd expect: pread64(3, 0x7f880e1c7000, 1048576, 0) = -1 EIO (Input/output error) pread: Input/output error +++ exited with 0 +++ But doing it with a larger buffer succeeds(!): pread64(3, "XFSB\0\0\20\0\0\0\0\0\0\fL\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0"..., 1146880, 0) = 1146880 read 1146880/1146880 bytes at offset 0 1 MiB, 1 ops; 0.0009 sec (1.124 GiB/sec and 1052.6316 ops/sec) +++ exited with 0 +++ (Note that the part of the buffer corresponding to the dm-error area is uninitialized) On 5.3-rc2, both commands would fail with EIO like you'd expect. The only change between rc2 and rc3 is commit 0eb6ddfb865c ("block: Fix __blkdev_direct_IO() for bio fragments"). AFAICT we end up in __blkdev_direct_IO with a 1120K buffer, which gets split into two bios: one for the first BIO_MAX_PAGES worth of data (1MB) and a second one for the 96k after that." Fix this by noting that it's always safe to dereference dio if we get BLK_QC_T_EAGAIN returned, as end_io hasn't been run for that case. So we can safely increment the dio size before calling submit_bio(), and then decrement it on failure (not that it really matters, as the bio and dio are going away). For error handling, return to the original method of just using 'ret' for tracking the error, and the size tracking in dio->size. Fixes: 0eb6ddfb865c ("block: Fix __blkdev_direct_IO() for bio fragments") Fixes: 6a43074e2f46 ("block: properly handle IOCB_NOWAIT for async O_DIRECT IO") Reported-by: Darrick J. Wong <darrick.wong@oracle.com> Signed-off-by: Jens Axboe <axboe@kernel.dk>
2019-08-06 19:34:31 +00:00
ret = bio_iov_iter_get_pages(bio, iter);
if (unlikely(ret)) {
bio->bi_status = BLK_STS_IOERR;
bio_endio(bio);
break;
}
if (is_read) {
bio->bi_opf = REQ_OP_READ;
if (dio->should_dirty)
bio_set_pages_dirty(bio);
} else {
bio->bi_opf = dio_bio_write_op(iocb);
task_io_account_write(bio->bi_iter.bi_size);
}
dio->size += bio->bi_iter.bi_size;
pos += bio->bi_iter.bi_size;
nr_pages = iov_iter_npages(iter, BIO_MAX_PAGES);
if (!nr_pages) {
bool polled = false;
if (iocb->ki_flags & IOCB_HIPRI) {
bio_set_polled(bio, iocb);
polled = true;
}
qc = submit_bio(bio);
if (polled)
WRITE_ONCE(iocb->ki_cookie, qc);
break;
}
if (!dio->multi_bio) {
/*
* AIO needs an extra reference to ensure the dio
* structure which is embedded into the first bio
* stays around.
*/
if (!is_sync)
bio_get(bio);
dio->multi_bio = true;
atomic_set(&dio->ref, 2);
} else {
atomic_inc(&dio->ref);
}
submit_bio(bio);
bio = bio_alloc(GFP_KERNEL, nr_pages);
}
if (!is_poll)
blk_finish_plug(&plug);
if (!is_sync)
return -EIOCBQUEUED;
for (;;) {
set_current_state(TASK_UNINTERRUPTIBLE);
if (!READ_ONCE(dio->waiter))
break;
if (!(iocb->ki_flags & IOCB_HIPRI) ||
!blk_poll(bdev_get_queue(bdev), qc, true))
blk_io_schedule();
}
__set_current_state(TASK_RUNNING);
if (!ret)
ret = blk_status_to_errno(dio->bio.bi_status);
block: fix O_DIRECT error handling for bio fragments 0eb6ddfb865c tried to fix this up, but introduced a use-after-free of dio. Additionally, we still had an issue with error handling, as reported by Darrick: "I noticed a regression in xfs/747 (an unreleased xfstest for the xfs_scrub media scanning feature) on 5.3-rc3. I'll condense that down to a simpler reproducer: error-test: 0 209 linear 8:48 0 error-test: 209 1 error error-test: 210 6446894 linear 8:48 210 Basically we have a ~3G /dev/sdd and we set up device mapper to fail IO for sector 209 and to pass the io to the scsi device everywhere else. On 5.3-rc3, performing a directio pread of this range with a < 1M buffer (in other words, a request for fewer than MAX_BIO_PAGES bytes) yields EIO like you'd expect: pread64(3, 0x7f880e1c7000, 1048576, 0) = -1 EIO (Input/output error) pread: Input/output error +++ exited with 0 +++ But doing it with a larger buffer succeeds(!): pread64(3, "XFSB\0\0\20\0\0\0\0\0\0\fL\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0"..., 1146880, 0) = 1146880 read 1146880/1146880 bytes at offset 0 1 MiB, 1 ops; 0.0009 sec (1.124 GiB/sec and 1052.6316 ops/sec) +++ exited with 0 +++ (Note that the part of the buffer corresponding to the dm-error area is uninitialized) On 5.3-rc2, both commands would fail with EIO like you'd expect. The only change between rc2 and rc3 is commit 0eb6ddfb865c ("block: Fix __blkdev_direct_IO() for bio fragments"). AFAICT we end up in __blkdev_direct_IO with a 1120K buffer, which gets split into two bios: one for the first BIO_MAX_PAGES worth of data (1MB) and a second one for the 96k after that." Fix this by noting that it's always safe to dereference dio if we get BLK_QC_T_EAGAIN returned, as end_io hasn't been run for that case. So we can safely increment the dio size before calling submit_bio(), and then decrement it on failure (not that it really matters, as the bio and dio are going away). For error handling, return to the original method of just using 'ret' for tracking the error, and the size tracking in dio->size. Fixes: 0eb6ddfb865c ("block: Fix __blkdev_direct_IO() for bio fragments") Fixes: 6a43074e2f46 ("block: properly handle IOCB_NOWAIT for async O_DIRECT IO") Reported-by: Darrick J. Wong <darrick.wong@oracle.com> Signed-off-by: Jens Axboe <axboe@kernel.dk>
2019-08-06 19:34:31 +00:00
if (likely(!ret))
ret = dio->size;
bio_put(&dio->bio);
return ret;
}
static ssize_t
blkdev_direct_IO(struct kiocb *iocb, struct iov_iter *iter)
{
int nr_pages;
nr_pages = iov_iter_npages(iter, BIO_MAX_PAGES + 1);
if (!nr_pages)
return 0;
if (is_sync_kiocb(iocb) && nr_pages <= BIO_MAX_PAGES)
return __blkdev_direct_IO_simple(iocb, iter, nr_pages);
return __blkdev_direct_IO(iocb, iter, min(nr_pages, BIO_MAX_PAGES));
}
static __init int blkdev_init(void)
{
return bioset_init(&blkdev_dio_pool, 4, offsetof(struct blkdev_dio, bio), BIOSET_NEED_BVECS);
}
module_init(blkdev_init);
int __sync_blockdev(struct block_device *bdev, int wait)
{
if (!bdev)
return 0;
if (!wait)
return filemap_flush(bdev->bd_inode->i_mapping);
return filemap_write_and_wait(bdev->bd_inode->i_mapping);
}
/*
* Write out and wait upon all the dirty data associated with a block
* device via its mapping. Does not take the superblock lock.
*/
int sync_blockdev(struct block_device *bdev)
{
return __sync_blockdev(bdev, 1);
}
EXPORT_SYMBOL(sync_blockdev);
/*
* Write out and wait upon all dirty data associated with this
* device. Filesystem data as well as the underlying block
* device. Takes the superblock lock.
*/
int fsync_bdev(struct block_device *bdev)
{
struct super_block *sb = get_super(bdev);
if (sb) {
int res = sync_filesystem(sb);
drop_super(sb);
return res;
}
return sync_blockdev(bdev);
}
EXPORT_SYMBOL(fsync_bdev);
/**
* freeze_bdev -- lock a filesystem and force it into a consistent state
* @bdev: blockdevice to lock
*
* If a superblock is found on this device, we take the s_umount semaphore
* on it to make sure nobody unmounts until the snapshot creation is done.
* The reference counter (bd_fsfreeze_count) guarantees that only the last
* unfreeze process can unfreeze the frozen filesystem actually when multiple
* freeze requests arrive simultaneously. It counts up in freeze_bdev() and
* count down in thaw_bdev(). When it becomes 0, thaw_bdev() will unfreeze
* actually.
*/
int freeze_bdev(struct block_device *bdev)
{
struct super_block *sb;
int error = 0;
mutex_lock(&bdev->bd_fsfreeze_mutex);
if (++bdev->bd_fsfreeze_count > 1)
goto done;
sb = get_active_super(bdev);
if (!sb)
goto sync;
if (sb->s_op->freeze_super)
error = sb->s_op->freeze_super(sb);
else
error = freeze_super(sb);
deactivate_super(sb);
if (error) {
bdev->bd_fsfreeze_count--;
goto done;
}
bdev->bd_fsfreeze_sb = sb;
sync:
sync_blockdev(bdev);
done:
mutex_unlock(&bdev->bd_fsfreeze_mutex);
return error;
}
EXPORT_SYMBOL(freeze_bdev);
/**
* thaw_bdev -- unlock filesystem
* @bdev: blockdevice to unlock
*
* Unlocks the filesystem and marks it writeable again after freeze_bdev().
*/
int thaw_bdev(struct block_device *bdev)
{
struct super_block *sb;
int error = -EINVAL;
mutex_lock(&bdev->bd_fsfreeze_mutex);
if (!bdev->bd_fsfreeze_count)
goto out;
error = 0;
if (--bdev->bd_fsfreeze_count > 0)
goto out;
sb = bdev->bd_fsfreeze_sb;
if (!sb)
goto out;
if (sb->s_op->thaw_super)
error = sb->s_op->thaw_super(sb);
else
error = thaw_super(sb);
if (error)
bdev->bd_fsfreeze_count++;
out:
mutex_unlock(&bdev->bd_fsfreeze_mutex);
return error;
}
EXPORT_SYMBOL(thaw_bdev);
static int blkdev_writepage(struct page *page, struct writeback_control *wbc)
{
return block_write_full_page(page, blkdev_get_block, wbc);
}
static int blkdev_readpage(struct file * file, struct page * page)
{
return block_read_full_page(page, blkdev_get_block);
}
static void blkdev_readahead(struct readahead_control *rac)
{
mpage_readahead(rac, blkdev_get_block);
}
static int blkdev_write_begin(struct file *file, struct address_space *mapping,
loff_t pos, unsigned len, unsigned flags,
struct page **pagep, void **fsdata)
{
return block_write_begin(mapping, pos, len, flags, pagep,
blkdev_get_block);
}
static int blkdev_write_end(struct file *file, struct address_space *mapping,
loff_t pos, unsigned len, unsigned copied,
struct page *page, void *fsdata)
{
int ret;
ret = block_write_end(file, mapping, pos, len, copied, page, fsdata);
unlock_page(page);
mm, fs: get rid of PAGE_CACHE_* and page_cache_{get,release} macros PAGE_CACHE_{SIZE,SHIFT,MASK,ALIGN} macros were introduced *long* time ago with promise that one day it will be possible to implement page cache with bigger chunks than PAGE_SIZE. This promise never materialized. And unlikely will. We have many places where PAGE_CACHE_SIZE assumed to be equal to PAGE_SIZE. And it's constant source of confusion on whether PAGE_CACHE_* or PAGE_* constant should be used in a particular case, especially on the border between fs and mm. Global switching to PAGE_CACHE_SIZE != PAGE_SIZE would cause to much breakage to be doable. Let's stop pretending that pages in page cache are special. They are not. The changes are pretty straight-forward: - <foo> << (PAGE_CACHE_SHIFT - PAGE_SHIFT) -> <foo>; - <foo> >> (PAGE_CACHE_SHIFT - PAGE_SHIFT) -> <foo>; - PAGE_CACHE_{SIZE,SHIFT,MASK,ALIGN} -> PAGE_{SIZE,SHIFT,MASK,ALIGN}; - page_cache_get() -> get_page(); - page_cache_release() -> put_page(); This patch contains automated changes generated with coccinelle using script below. For some reason, coccinelle doesn't patch header files. I've called spatch for them manually. The only adjustment after coccinelle is revert of changes to PAGE_CAHCE_ALIGN definition: we are going to drop it later. There are few places in the code where coccinelle didn't reach. I'll fix them manually in a separate patch. Comments and documentation also will be addressed with the separate patch. virtual patch @@ expression E; @@ - E << (PAGE_CACHE_SHIFT - PAGE_SHIFT) + E @@ expression E; @@ - E >> (PAGE_CACHE_SHIFT - PAGE_SHIFT) + E @@ @@ - PAGE_CACHE_SHIFT + PAGE_SHIFT @@ @@ - PAGE_CACHE_SIZE + PAGE_SIZE @@ @@ - PAGE_CACHE_MASK + PAGE_MASK @@ expression E; @@ - PAGE_CACHE_ALIGN(E) + PAGE_ALIGN(E) @@ expression E; @@ - page_cache_get(E) + get_page(E) @@ expression E; @@ - page_cache_release(E) + put_page(E) Signed-off-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com> Acked-by: Michal Hocko <mhocko@suse.com> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2016-04-01 12:29:47 +00:00
put_page(page);
return ret;
}
/*
* private llseek:
* for a block special file file_inode(file)->i_size is zero
* so we compute the size by hand (just as in block_read/write above)
*/
static loff_t block_llseek(struct file *file, loff_t offset, int whence)
{
struct inode *bd_inode = bdev_file_inode(file);
loff_t retval;
inode_lock(bd_inode);
retval = fixed_size_llseek(file, offset, whence, i_size_read(bd_inode));
inode_unlock(bd_inode);
return retval;
}
int blkdev_fsync(struct file *filp, loff_t start, loff_t end, int datasync)
{
struct inode *bd_inode = bdev_file_inode(filp);
fs/block_dev.c: fix performance regression in O_DIRECT|O_SYNC writes to block devices We are seeing a large regression in database performance on recent kernels. The database opens a block device with O_DIRECT|O_SYNC and a number of threads write to different regions of the file at the same time. A simple test case is below. I haven't defined DEVICE since getting it wrong will destroy your data :) On an 3 disk LVM with a 64k chunk size we see about 17MB/sec and only a few threads in IO wait: procs -----io---- -system-- -----cpu------ r b bi bo in cs us sy id wa st 0 3 0 16170 656 2259 0 0 86 14 0 0 2 0 16704 695 2408 0 0 92 8 0 0 2 0 17308 744 2653 0 0 86 14 0 0 2 0 17933 759 2777 0 0 89 10 0 Most threads are blocking in vfs_fsync_range, which has: mutex_lock(&mapping->host->i_mutex); err = fop->fsync(file, dentry, datasync); if (!ret) ret = err; mutex_unlock(&mapping->host->i_mutex); commit 148f948ba877f4d3cdef036b1ff6d9f68986706a (vfs: Introduce new helpers for syncing after writing to O_SYNC file or IS_SYNC inode) offers some explanation of what is going on: Use these new helpers for syncing from generic VFS functions. This makes O_SYNC writes to block devices acquire i_mutex for syncing. If we really care about this, we can make block_fsync() drop the i_mutex and reacquire it before it returns. Thanks Jan for such a good commit message! As well as dropping i_mutex, Christoph suggests we should remove the call to sync_blockdev(): > sync_blockdev is an overcomplicated alias for filemap_write_and_wait on > the block device inode, which is exactly what we did just before calling > into ->fsync The patch below incorporates both suggestions. With it the testcase improves from 17MB/s to 68M/sec: procs -----io---- -system-- -----cpu------ r b bi bo in cs us sy id wa st 0 7 0 65536 1000 3878 0 0 70 30 0 0 34 0 69632 1016 3921 0 1 46 53 0 0 57 0 69632 1000 3921 0 0 55 45 0 0 53 0 69640 754 4111 0 0 81 19 0 Testcase: #define _GNU_SOURCE #include <stdio.h> #include <pthread.h> #include <unistd.h> #include <stdlib.h> #include <string.h> #include <sys/types.h> #include <sys/stat.h> #include <fcntl.h> #define NR_THREADS 64 #define BUFSIZE (64 * 1024) #define DEVICE "/dev/mapper/XXXXXX" #define ALIGN(VAL, SIZE) (((VAL)+(SIZE)-1) & ~((SIZE)-1)) static int fd; static void *doit(void *arg) { unsigned long offset = (long)arg; char *b, *buf; b = malloc(BUFSIZE + 1024); buf = (char *)ALIGN((unsigned long)b, 1024); memset(buf, 0, BUFSIZE); while (1) pwrite(fd, buf, BUFSIZE, offset); } int main(int argc, char *argv[]) { int flags = O_RDWR|O_DIRECT; int i; unsigned long offset = 0; if (argc > 1 && !strcmp(argv[1], "O_SYNC")) flags |= O_SYNC; fd = open(DEVICE, flags); if (fd == -1) { perror("open"); exit(1); } for (i = 0; i < NR_THREADS-1; i++) { pthread_t tid; pthread_create(&tid, NULL, doit, (void *)offset); offset += BUFSIZE; } doit((void *)offset); return 0; } Signed-off-by: Anton Blanchard <anton@samba.org> Acked-by: Jan Kara <jack@suse.cz> Cc: Christoph Hellwig <hch@lst.de> Cc: Alexander Viro <viro@zeniv.linux.org.uk> Cc: Jens Axboe <jens.axboe@oracle.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2010-04-23 17:18:06 +00:00
struct block_device *bdev = I_BDEV(bd_inode);
int error;
error = file_write_and_wait_range(filp, start, end);
if (error)
return error;
fs/block_dev.c: fix performance regression in O_DIRECT|O_SYNC writes to block devices We are seeing a large regression in database performance on recent kernels. The database opens a block device with O_DIRECT|O_SYNC and a number of threads write to different regions of the file at the same time. A simple test case is below. I haven't defined DEVICE since getting it wrong will destroy your data :) On an 3 disk LVM with a 64k chunk size we see about 17MB/sec and only a few threads in IO wait: procs -----io---- -system-- -----cpu------ r b bi bo in cs us sy id wa st 0 3 0 16170 656 2259 0 0 86 14 0 0 2 0 16704 695 2408 0 0 92 8 0 0 2 0 17308 744 2653 0 0 86 14 0 0 2 0 17933 759 2777 0 0 89 10 0 Most threads are blocking in vfs_fsync_range, which has: mutex_lock(&mapping->host->i_mutex); err = fop->fsync(file, dentry, datasync); if (!ret) ret = err; mutex_unlock(&mapping->host->i_mutex); commit 148f948ba877f4d3cdef036b1ff6d9f68986706a (vfs: Introduce new helpers for syncing after writing to O_SYNC file or IS_SYNC inode) offers some explanation of what is going on: Use these new helpers for syncing from generic VFS functions. This makes O_SYNC writes to block devices acquire i_mutex for syncing. If we really care about this, we can make block_fsync() drop the i_mutex and reacquire it before it returns. Thanks Jan for such a good commit message! As well as dropping i_mutex, Christoph suggests we should remove the call to sync_blockdev(): > sync_blockdev is an overcomplicated alias for filemap_write_and_wait on > the block device inode, which is exactly what we did just before calling > into ->fsync The patch below incorporates both suggestions. With it the testcase improves from 17MB/s to 68M/sec: procs -----io---- -system-- -----cpu------ r b bi bo in cs us sy id wa st 0 7 0 65536 1000 3878 0 0 70 30 0 0 34 0 69632 1016 3921 0 1 46 53 0 0 57 0 69632 1000 3921 0 0 55 45 0 0 53 0 69640 754 4111 0 0 81 19 0 Testcase: #define _GNU_SOURCE #include <stdio.h> #include <pthread.h> #include <unistd.h> #include <stdlib.h> #include <string.h> #include <sys/types.h> #include <sys/stat.h> #include <fcntl.h> #define NR_THREADS 64 #define BUFSIZE (64 * 1024) #define DEVICE "/dev/mapper/XXXXXX" #define ALIGN(VAL, SIZE) (((VAL)+(SIZE)-1) & ~((SIZE)-1)) static int fd; static void *doit(void *arg) { unsigned long offset = (long)arg; char *b, *buf; b = malloc(BUFSIZE + 1024); buf = (char *)ALIGN((unsigned long)b, 1024); memset(buf, 0, BUFSIZE); while (1) pwrite(fd, buf, BUFSIZE, offset); } int main(int argc, char *argv[]) { int flags = O_RDWR|O_DIRECT; int i; unsigned long offset = 0; if (argc > 1 && !strcmp(argv[1], "O_SYNC")) flags |= O_SYNC; fd = open(DEVICE, flags); if (fd == -1) { perror("open"); exit(1); } for (i = 0; i < NR_THREADS-1; i++) { pthread_t tid; pthread_create(&tid, NULL, doit, (void *)offset); offset += BUFSIZE; } doit((void *)offset); return 0; } Signed-off-by: Anton Blanchard <anton@samba.org> Acked-by: Jan Kara <jack@suse.cz> Cc: Christoph Hellwig <hch@lst.de> Cc: Alexander Viro <viro@zeniv.linux.org.uk> Cc: Jens Axboe <jens.axboe@oracle.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2010-04-23 17:18:06 +00:00
/*
* There is no need to serialise calls to blkdev_issue_flush with
* i_mutex and doing so causes performance issues with concurrent
* O_SYNC writers to a block device.
*/
error = blkdev_issue_flush(bdev, GFP_KERNEL);
if (error == -EOPNOTSUPP)
error = 0;
fs/block_dev.c: fix performance regression in O_DIRECT|O_SYNC writes to block devices We are seeing a large regression in database performance on recent kernels. The database opens a block device with O_DIRECT|O_SYNC and a number of threads write to different regions of the file at the same time. A simple test case is below. I haven't defined DEVICE since getting it wrong will destroy your data :) On an 3 disk LVM with a 64k chunk size we see about 17MB/sec and only a few threads in IO wait: procs -----io---- -system-- -----cpu------ r b bi bo in cs us sy id wa st 0 3 0 16170 656 2259 0 0 86 14 0 0 2 0 16704 695 2408 0 0 92 8 0 0 2 0 17308 744 2653 0 0 86 14 0 0 2 0 17933 759 2777 0 0 89 10 0 Most threads are blocking in vfs_fsync_range, which has: mutex_lock(&mapping->host->i_mutex); err = fop->fsync(file, dentry, datasync); if (!ret) ret = err; mutex_unlock(&mapping->host->i_mutex); commit 148f948ba877f4d3cdef036b1ff6d9f68986706a (vfs: Introduce new helpers for syncing after writing to O_SYNC file or IS_SYNC inode) offers some explanation of what is going on: Use these new helpers for syncing from generic VFS functions. This makes O_SYNC writes to block devices acquire i_mutex for syncing. If we really care about this, we can make block_fsync() drop the i_mutex and reacquire it before it returns. Thanks Jan for such a good commit message! As well as dropping i_mutex, Christoph suggests we should remove the call to sync_blockdev(): > sync_blockdev is an overcomplicated alias for filemap_write_and_wait on > the block device inode, which is exactly what we did just before calling > into ->fsync The patch below incorporates both suggestions. With it the testcase improves from 17MB/s to 68M/sec: procs -----io---- -system-- -----cpu------ r b bi bo in cs us sy id wa st 0 7 0 65536 1000 3878 0 0 70 30 0 0 34 0 69632 1016 3921 0 1 46 53 0 0 57 0 69632 1000 3921 0 0 55 45 0 0 53 0 69640 754 4111 0 0 81 19 0 Testcase: #define _GNU_SOURCE #include <stdio.h> #include <pthread.h> #include <unistd.h> #include <stdlib.h> #include <string.h> #include <sys/types.h> #include <sys/stat.h> #include <fcntl.h> #define NR_THREADS 64 #define BUFSIZE (64 * 1024) #define DEVICE "/dev/mapper/XXXXXX" #define ALIGN(VAL, SIZE) (((VAL)+(SIZE)-1) & ~((SIZE)-1)) static int fd; static void *doit(void *arg) { unsigned long offset = (long)arg; char *b, *buf; b = malloc(BUFSIZE + 1024); buf = (char *)ALIGN((unsigned long)b, 1024); memset(buf, 0, BUFSIZE); while (1) pwrite(fd, buf, BUFSIZE, offset); } int main(int argc, char *argv[]) { int flags = O_RDWR|O_DIRECT; int i; unsigned long offset = 0; if (argc > 1 && !strcmp(argv[1], "O_SYNC")) flags |= O_SYNC; fd = open(DEVICE, flags); if (fd == -1) { perror("open"); exit(1); } for (i = 0; i < NR_THREADS-1; i++) { pthread_t tid; pthread_create(&tid, NULL, doit, (void *)offset); offset += BUFSIZE; } doit((void *)offset); return 0; } Signed-off-by: Anton Blanchard <anton@samba.org> Acked-by: Jan Kara <jack@suse.cz> Cc: Christoph Hellwig <hch@lst.de> Cc: Alexander Viro <viro@zeniv.linux.org.uk> Cc: Jens Axboe <jens.axboe@oracle.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2010-04-23 17:18:06 +00:00
return error;
}
EXPORT_SYMBOL(blkdev_fsync);
/**
* bdev_read_page() - Start reading a page from a block device
* @bdev: The device to read the page from
* @sector: The offset on the device to read the page to (need not be aligned)
* @page: The page to read
*
* On entry, the page should be locked. It will be unlocked when the page
* has been read. If the block driver implements rw_page synchronously,
* that will be true on exit from this function, but it need not be.
*
* Errors returned by this function are usually "soft", eg out of memory, or
* queue full; callers should try a different route to read this page rather
* than propagate an error back up the stack.
*
* Return: negative errno if an error occurs, 0 if submission was successful.
*/
int bdev_read_page(struct block_device *bdev, sector_t sector,
struct page *page)
{
const struct block_device_operations *ops = bdev->bd_disk->fops;
int result = -EOPNOTSUPP;
if (!ops->rw_page || bdev_get_integrity(bdev))
return result;
result = blk_queue_enter(bdev->bd_disk->queue, 0);
if (result)
return result;
result = ops->rw_page(bdev, sector + get_start_sect(bdev), page,
REQ_OP_READ);
blk_queue_exit(bdev->bd_disk->queue);
return result;
}
/**
* bdev_write_page() - Start writing a page to a block device
* @bdev: The device to write the page to
* @sector: The offset on the device to write the page to (need not be aligned)
* @page: The page to write
* @wbc: The writeback_control for the write
*
* On entry, the page should be locked and not currently under writeback.
* On exit, if the write started successfully, the page will be unlocked and
* under writeback. If the write failed already (eg the driver failed to
* queue the page to the device), the page will still be locked. If the
* caller is a ->writepage implementation, it will need to unlock the page.
*
* Errors returned by this function are usually "soft", eg out of memory, or
* queue full; callers should try a different route to write this page rather
* than propagate an error back up the stack.
*
* Return: negative errno if an error occurs, 0 if submission was successful.
*/
int bdev_write_page(struct block_device *bdev, sector_t sector,
struct page *page, struct writeback_control *wbc)
{
int result;
const struct block_device_operations *ops = bdev->bd_disk->fops;
if (!ops->rw_page || bdev_get_integrity(bdev))
return -EOPNOTSUPP;
result = blk_queue_enter(bdev->bd_disk->queue, 0);
if (result)
return result;
set_page_writeback(page);
result = ops->rw_page(bdev, sector + get_start_sect(bdev), page,
REQ_OP_WRITE);
if (result) {
end_page_writeback(page);
} else {
clean_page_buffers(page);
unlock_page(page);
}
blk_queue_exit(bdev->bd_disk->queue);
return result;
}
/*
* pseudo-fs
*/
static __cacheline_aligned_in_smp DEFINE_SPINLOCK(bdev_lock);
static struct kmem_cache * bdev_cachep __read_mostly;
static struct inode *bdev_alloc_inode(struct super_block *sb)
{
struct bdev_inode *ei = kmem_cache_alloc(bdev_cachep, GFP_KERNEL);
if (!ei)
return NULL;
return &ei->vfs_inode;
}
static void bdev_free_inode(struct inode *inode)
{
kmem_cache_free(bdev_cachep, BDEV_I(inode));
2011-01-07 06:49:49 +00:00
}
static void init_once(void *foo)
{
struct bdev_inode *ei = (struct bdev_inode *) foo;
struct block_device *bdev = &ei->bdev;
memset(bdev, 0, sizeof(*bdev));
mutex_init(&bdev->bd_mutex);
#ifdef CONFIG_SYSFS
INIT_LIST_HEAD(&bdev->bd_holder_disks);
#endif
bdev->bd_bdi = &noop_backing_dev_info;
inode_init_once(&ei->vfs_inode);
/* Initialize mutex for freeze. */
mutex_init(&bdev->bd_fsfreeze_mutex);
}
static void bdev_evict_inode(struct inode *inode)
{
struct block_device *bdev = &BDEV_I(inode)->bdev;
mm + fs: store shadow entries in page cache Reclaim will be leaving shadow entries in the page cache radix tree upon evicting the real page. As those pages are found from the LRU, an iput() can lead to the inode being freed concurrently. At this point, reclaim must no longer install shadow pages because the inode freeing code needs to ensure the page tree is really empty. Add an address_space flag, AS_EXITING, that the inode freeing code sets under the tree lock before doing the final truncate. Reclaim will check for this flag before installing shadow pages. Signed-off-by: Johannes Weiner <hannes@cmpxchg.org> Reviewed-by: Rik van Riel <riel@redhat.com> Reviewed-by: Minchan Kim <minchan@kernel.org> Cc: Andrea Arcangeli <aarcange@redhat.com> Cc: Bob Liu <bob.liu@oracle.com> Cc: Christoph Hellwig <hch@infradead.org> Cc: Dave Chinner <david@fromorbit.com> Cc: Greg Thelen <gthelen@google.com> Cc: Hugh Dickins <hughd@google.com> Cc: Jan Kara <jack@suse.cz> Cc: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com> Cc: Luigi Semenzato <semenzato@google.com> Cc: Mel Gorman <mgorman@suse.de> Cc: Metin Doslu <metin@citusdata.com> Cc: Michel Lespinasse <walken@google.com> Cc: Ozgun Erdogan <ozgun@citusdata.com> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Roman Gushchin <klamm@yandex-team.ru> Cc: Ryan Mallon <rmallon@gmail.com> Cc: Tejun Heo <tj@kernel.org> Cc: Vlastimil Babka <vbabka@suse.cz> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2014-04-03 21:47:49 +00:00
truncate_inode_pages_final(&inode->i_data);
invalidate_inode_buffers(inode); /* is it needed here? */
clear_inode(inode);
/* Detach inode from wb early as bdi_put() may free bdi->wb */
inode_detach_wb(inode);
if (bdev->bd_bdi != &noop_backing_dev_info) {
bdi_put(bdev->bd_bdi);
bdev->bd_bdi = &noop_backing_dev_info;
}
}
static const struct super_operations bdev_sops = {
.statfs = simple_statfs,
.alloc_inode = bdev_alloc_inode,
.free_inode = bdev_free_inode,
.drop_inode = generic_delete_inode,
.evict_inode = bdev_evict_inode,
};
static int bd_init_fs_context(struct fs_context *fc)
{
struct pseudo_fs_context *ctx = init_pseudo(fc, BDEVFS_MAGIC);
if (!ctx)
return -ENOMEM;
fc->s_iflags |= SB_I_CGROUPWB;
ctx->ops = &bdev_sops;
return 0;
}
static struct file_system_type bd_type = {
.name = "bdev",
.init_fs_context = bd_init_fs_context,
.kill_sb = kill_anon_super,
};
struct super_block *blockdev_superblock __read_mostly;
EXPORT_SYMBOL_GPL(blockdev_superblock);
void __init bdev_cache_init(void)
{
int err;
block_dev: Suppress bdev_cache_init() kmemleak warninig Kmemleak reports the following warning in bdev_cache_init() [ 0.003738] kmemleak: Object 0xffff880153035200 (size 256): [ 0.003823] kmemleak: comm "swapper/0", pid 0, jiffies 4294667299 [ 0.003909] kmemleak: min_count = 1 [ 0.003988] kmemleak: count = 0 [ 0.004066] kmemleak: flags = 0x1 [ 0.004144] kmemleak: checksum = 0 [ 0.004224] kmemleak: backtrace: [ 0.004303] [<ffffffff814755ac>] kmemleak_alloc+0x21/0x3e [ 0.004446] [<ffffffff811100ba>] kmem_cache_alloc+0xca/0x1dc [ 0.004592] [<ffffffff811371b1>] alloc_vfsmnt+0x1f/0x198 [ 0.004736] [<ffffffff811375c5>] vfs_kern_mount+0x36/0xd2 [ 0.004879] [<ffffffff8113929a>] kern_mount_data+0x18/0x32 [ 0.005025] [<ffffffff81ab9075>] bdev_cache_init+0x51/0x81 [ 0.005169] [<ffffffff81ab8abf>] vfs_caches_init+0x101/0x10d [ 0.005313] [<ffffffff81a9bae3>] start_kernel+0x344/0x383 [ 0.005456] [<ffffffff81a9b2a7>] x86_64_start_reservations+0xae/0xb2 [ 0.005602] [<ffffffff81a9b3ad>] x86_64_start_kernel+0x102/0x111 [ 0.005747] [<ffffffffffffffff>] 0xffffffffffffffff [ 0.008653] kmemleak: Trying to color unknown object at 0xffff880153035220 as Grey [ 0.008754] Pid: 0, comm: swapper/0 Not tainted 3.3.0-rc0-dbg-04200-g8180888-dirty #888 [ 0.008856] Call Trace: [ 0.008934] [<ffffffff81118704>] ? find_and_get_object+0x44/0x118 [ 0.009023] [<ffffffff81118fe6>] paint_ptr+0x57/0x8f [ 0.009109] [<ffffffff81475935>] kmemleak_not_leak+0x23/0x42 [ 0.009195] [<ffffffff81ab9096>] bdev_cache_init+0x72/0x81 [ 0.009282] [<ffffffff81ab8abf>] vfs_caches_init+0x101/0x10d [ 0.009368] [<ffffffff81a9bae3>] start_kernel+0x344/0x383 [ 0.009466] [<ffffffff81a9b2a7>] x86_64_start_reservations+0xae/0xb2 [ 0.009555] [<ffffffff81a9b140>] ? early_idt_handlers+0x140/0x140 [ 0.009643] [<ffffffff81a9b3ad>] x86_64_start_kernel+0x102/0x111 due to attempt to mark pointer to `struct vfsmount' as a gray object, which is embedded into `struct mount' returned from alloc_vfsmnt(). Make `bd_mnt' static, avoiding need to tell kmemleak to mark it gray, as suggested by Al Viro. Signed-off-by: Sergey Senozhatsky <sergey.senozhatsky@gmail.com> Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
2012-01-09 23:43:59 +00:00
static struct vfsmount *bd_mnt;
bdev_cachep = kmem_cache_create("bdev_cache", sizeof(struct bdev_inode),
0, (SLAB_HWCACHE_ALIGN|SLAB_RECLAIM_ACCOUNT|
2016-01-14 23:18:21 +00:00
SLAB_MEM_SPREAD|SLAB_ACCOUNT|SLAB_PANIC),
init_once);
err = register_filesystem(&bd_type);
if (err)
panic("Cannot register bdev pseudo-fs");
bd_mnt = kern_mount(&bd_type);
if (IS_ERR(bd_mnt))
panic("Cannot create bdev pseudo-fs");
block_dev: Suppress bdev_cache_init() kmemleak warninig Kmemleak reports the following warning in bdev_cache_init() [ 0.003738] kmemleak: Object 0xffff880153035200 (size 256): [ 0.003823] kmemleak: comm "swapper/0", pid 0, jiffies 4294667299 [ 0.003909] kmemleak: min_count = 1 [ 0.003988] kmemleak: count = 0 [ 0.004066] kmemleak: flags = 0x1 [ 0.004144] kmemleak: checksum = 0 [ 0.004224] kmemleak: backtrace: [ 0.004303] [<ffffffff814755ac>] kmemleak_alloc+0x21/0x3e [ 0.004446] [<ffffffff811100ba>] kmem_cache_alloc+0xca/0x1dc [ 0.004592] [<ffffffff811371b1>] alloc_vfsmnt+0x1f/0x198 [ 0.004736] [<ffffffff811375c5>] vfs_kern_mount+0x36/0xd2 [ 0.004879] [<ffffffff8113929a>] kern_mount_data+0x18/0x32 [ 0.005025] [<ffffffff81ab9075>] bdev_cache_init+0x51/0x81 [ 0.005169] [<ffffffff81ab8abf>] vfs_caches_init+0x101/0x10d [ 0.005313] [<ffffffff81a9bae3>] start_kernel+0x344/0x383 [ 0.005456] [<ffffffff81a9b2a7>] x86_64_start_reservations+0xae/0xb2 [ 0.005602] [<ffffffff81a9b3ad>] x86_64_start_kernel+0x102/0x111 [ 0.005747] [<ffffffffffffffff>] 0xffffffffffffffff [ 0.008653] kmemleak: Trying to color unknown object at 0xffff880153035220 as Grey [ 0.008754] Pid: 0, comm: swapper/0 Not tainted 3.3.0-rc0-dbg-04200-g8180888-dirty #888 [ 0.008856] Call Trace: [ 0.008934] [<ffffffff81118704>] ? find_and_get_object+0x44/0x118 [ 0.009023] [<ffffffff81118fe6>] paint_ptr+0x57/0x8f [ 0.009109] [<ffffffff81475935>] kmemleak_not_leak+0x23/0x42 [ 0.009195] [<ffffffff81ab9096>] bdev_cache_init+0x72/0x81 [ 0.009282] [<ffffffff81ab8abf>] vfs_caches_init+0x101/0x10d [ 0.009368] [<ffffffff81a9bae3>] start_kernel+0x344/0x383 [ 0.009466] [<ffffffff81a9b2a7>] x86_64_start_reservations+0xae/0xb2 [ 0.009555] [<ffffffff81a9b140>] ? early_idt_handlers+0x140/0x140 [ 0.009643] [<ffffffff81a9b3ad>] x86_64_start_kernel+0x102/0x111 due to attempt to mark pointer to `struct vfsmount' as a gray object, which is embedded into `struct mount' returned from alloc_vfsmnt(). Make `bd_mnt' static, avoiding need to tell kmemleak to mark it gray, as suggested by Al Viro. Signed-off-by: Sergey Senozhatsky <sergey.senozhatsky@gmail.com> Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
2012-01-09 23:43:59 +00:00
blockdev_superblock = bd_mnt->mnt_sb; /* For writeback */
}
static struct block_device *bdget(dev_t dev)
{
struct block_device *bdev;
struct inode *inode;
inode = iget_locked(blockdev_superblock, dev);
if (!inode)
return NULL;
bdev = &BDEV_I(inode)->bdev;
if (inode->i_state & I_NEW) {
spin_lock_init(&bdev->bd_size_lock);
bdev->bd_contains = NULL;
bdev->bd_super = NULL;
bdev->bd_inode = inode;
bdev->bd_part_count = 0;
bdev->bd_dev = dev;
inode->i_mode = S_IFBLK;
inode->i_rdev = dev;
inode->i_bdev = bdev;
inode->i_data.a_ops = &def_blk_aops;
mapping_set_gfp_mask(&inode->i_data, GFP_USER);
unlock_new_inode(inode);
}
return bdev;
}
/**
* bdgrab -- Grab a reference to an already referenced block device
* @bdev: Block device to grab a reference to.
*
* Returns the block_device with an additional reference when successful,
* or NULL if the inode is already beeing freed.
*/
struct block_device *bdgrab(struct block_device *bdev)
{
if (!igrab(bdev->bd_inode))
return NULL;
return bdev;
}
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 <anatol.pomozov@gmail.com> Cc: Al Viro <viro@zeniv.linux.org.uk> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2013-04-01 16:47:56 +00:00
EXPORT_SYMBOL(bdgrab);
struct block_device *bdget_part(struct hd_struct *part)
{
return bdget(part_devt(part));
}
long nr_blockdev_pages(void)
{
struct inode *inode;
long ret = 0;
spin_lock(&blockdev_superblock->s_inode_list_lock);
list_for_each_entry(inode, &blockdev_superblock->s_inodes, i_sb_list)
ret += inode->i_mapping->nrpages;
spin_unlock(&blockdev_superblock->s_inode_list_lock);
return ret;
}
void bdput(struct block_device *bdev)
{
iput(bdev->bd_inode);
}
EXPORT_SYMBOL(bdput);
static struct block_device *bd_acquire(struct inode *inode)
{
struct block_device *bdev;
spin_lock(&bdev_lock);
bdev = inode->i_bdev;
if (bdev && !inode_unhashed(bdev->bd_inode)) {
bdgrab(bdev);
spin_unlock(&bdev_lock);
return bdev;
}
spin_unlock(&bdev_lock);
/*
* i_bdev references block device inode that was already shut down
* (corresponding device got removed). Remove the reference and look
* up block device inode again just in case new device got
* reestablished under the same device number.
*/
if (bdev)
bd_forget(inode);
bdev = bdget(inode->i_rdev);
if (bdev) {
spin_lock(&bdev_lock);
if (!inode->i_bdev) {
/*
* We take an additional reference to bd_inode,
* and it's released in clear_inode() of inode.
* So, we can access it via ->i_mapping always
* without igrab().
*/
bdgrab(bdev);
inode->i_bdev = bdev;
inode->i_mapping = bdev->bd_inode->i_mapping;
}
spin_unlock(&bdev_lock);
}
return bdev;
}
/* Call when you free inode */
void bd_forget(struct inode *inode)
{
struct block_device *bdev = NULL;
spin_lock(&bdev_lock);
if (!sb_is_blkdev_sb(inode->i_sb))
bdev = inode->i_bdev;
inode->i_bdev = NULL;
inode->i_mapping = &inode->i_data;
spin_unlock(&bdev_lock);
if (bdev)
bdput(bdev);
}
/**
* bd_may_claim - test whether a block device can be claimed
* @bdev: block device of interest
* @whole: whole block device containing @bdev, may equal @bdev
* @holder: holder trying to claim @bdev
*
* Test whether @bdev can be claimed by @holder.
*
* CONTEXT:
* spin_lock(&bdev_lock).
*
* RETURNS:
* %true if @bdev can be claimed, %false otherwise.
*/
static bool bd_may_claim(struct block_device *bdev, struct block_device *whole,
void *holder)
{
if (bdev->bd_holder == holder)
return true; /* already a holder */
else if (bdev->bd_holder != NULL)
return false; /* held by someone else */
block_dev: don't test bdev->bd_contains when it is not stable bdev->bd_contains is not stable before calling __blkdev_get(). When __blkdev_get() is called on a parition with ->bd_openers == 0 it sets bdev->bd_contains = bdev; which is not correct for a partition. After a call to __blkdev_get() succeeds, ->bd_openers will be > 0 and then ->bd_contains is stable. When FMODE_EXCL is used, blkdev_get() calls bd_start_claiming() -> bd_prepare_to_claim() -> bd_may_claim() This call happens before __blkdev_get() is called, so ->bd_contains is not stable. So bd_may_claim() cannot safely use ->bd_contains. It currently tries to use it, and this can lead to a BUG_ON(). This happens when a whole device is already open with a bd_holder (in use by dm in my particular example) and two threads race to open a partition of that device for the first time, one opening with O_EXCL and one without. The thread that doesn't use O_EXCL gets through blkdev_get() to __blkdev_get(), gains the ->bd_mutex, and sets bdev->bd_contains = bdev; Immediately thereafter the other thread, using FMODE_EXCL, calls bd_start_claiming() from blkdev_get(). This should fail because the whole device has a holder, but because bdev->bd_contains == bdev bd_may_claim() incorrectly reports success. This thread continues and blocks on bd_mutex. The first thread then sets bdev->bd_contains correctly and drops the mutex. The thread using FMODE_EXCL then continues and when it calls bd_may_claim() again in: BUG_ON(!bd_may_claim(bdev, whole, holder)); The BUG_ON fires. Fix this by removing the dependency on ->bd_contains in bd_may_claim(). As bd_may_claim() has direct access to the whole device, it can simply test if the target bdev is the whole device. Fixes: 6b4517a7913a ("block: implement bd_claiming and claiming block") Cc: stable@vger.kernel.org (v2.6.35+) Signed-off-by: NeilBrown <neilb@suse.com> Signed-off-by: Jens Axboe <axboe@fb.com>
2016-12-12 15:21:51 +00:00
else if (whole == bdev)
return true; /* is a whole device which isn't held */
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 <tj@kernel.org> Acked-by: Neil Brown <neilb@suse.de> Acked-by: Ryusuke Konishi <konishi.ryusuke@lab.ntt.co.jp> Acked-by: Mike Snitzer <snitzer@redhat.com> Acked-by: Philipp Reisner <philipp.reisner@linbit.com> Cc: Peter Osterlund <petero2@telia.com> Cc: Martin Schwidefsky <schwidefsky@de.ibm.com> Cc: Heiko Carstens <heiko.carstens@de.ibm.com> Cc: Jan Kara <jack@suse.cz> Cc: Andrew Morton <akpm@linux-foundation.org> Cc: Andreas Dilger <adilger.kernel@dilger.ca> Cc: "Theodore Ts'o" <tytso@mit.edu> Cc: Mark Fasheh <mfasheh@suse.com> Cc: Joel Becker <joel.becker@oracle.com> Cc: Alex Elder <aelder@sgi.com> Cc: Christoph Hellwig <hch@infradead.org> Cc: dm-devel@redhat.com Cc: drbd-dev@lists.linbit.com Cc: Leo Chen <leochen@broadcom.com> Cc: Scott Branden <sbranden@broadcom.com> Cc: Chris Mason <chris.mason@oracle.com> Cc: Steven Whitehouse <swhiteho@redhat.com> Cc: Dave Kleikamp <shaggy@linux.vnet.ibm.com> Cc: Joern Engel <joern@logfs.org> Cc: reiserfs-devel@vger.kernel.org Cc: Alexander Viro <viro@zeniv.linux.org.uk>
2010-11-13 10:55:17 +00:00
else if (whole->bd_holder == bd_may_claim)
return true; /* is a partition of a device that is being partitioned */
else if (whole->bd_holder != NULL)
return false; /* is a partition of a held device */
else
return true; /* is a partition of an un-held device */
}
block: implement bd_claiming and claiming block Currently, device claiming for exclusive open is done after low level open - disk->fops->open() - has completed successfully. This means that exclusive open attempts while a device is already exclusively open will fail only after disk->fops->open() is called. cdrom driver issues commands during open() which means that O_EXCL open attempt can unintentionally inject commands to in-progress command stream for burning thus disturbing burning process. In most cases, this doesn't cause problems because the first command to be issued is TUR which most devices can process in the middle of burning. However, depending on how a device replies to TUR during burning, cdrom driver may end up issuing further commands. This can't be resolved trivially by moving bd_claim() before doing actual open() because that means an open attempt which will end up failing could interfere other legit O_EXCL open attempts. ie. unconfirmed open attempts can fail others. This patch resolves the problem by introducing claiming block which is started by bd_start_claiming() and terminated either by bd_claim() or bd_abort_claiming(). bd_claim() from inside a claiming block is guaranteed to succeed and once a claiming block is started, other bd_start_claiming() or bd_claim() attempts block till the current claiming block is terminated. bd_claim() can still be used standalone although now it always synchronizes against claiming blocks, so the existing users will keep working without any change. blkdev_open() and open_bdev_exclusive() are converted to use claiming blocks so that exclusive open attempts from these functions don't interfere with the existing exclusive open. This problem was discovered while investigating bko#15403. https://bugzilla.kernel.org/show_bug.cgi?id=15403 The burning problem itself can be resolved by updating userspace probing tools to always open w/ O_EXCL. Signed-off-by: Tejun Heo <tj@kernel.org> Reported-by: Matthias-Christian Ott <ott@mirix.org> Cc: Kay Sievers <kay.sievers@vrfy.org> Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
2010-04-07 09:53:59 +00:00
/**
* bd_prepare_to_claim - claim a block device
block: implement bd_claiming and claiming block Currently, device claiming for exclusive open is done after low level open - disk->fops->open() - has completed successfully. This means that exclusive open attempts while a device is already exclusively open will fail only after disk->fops->open() is called. cdrom driver issues commands during open() which means that O_EXCL open attempt can unintentionally inject commands to in-progress command stream for burning thus disturbing burning process. In most cases, this doesn't cause problems because the first command to be issued is TUR which most devices can process in the middle of burning. However, depending on how a device replies to TUR during burning, cdrom driver may end up issuing further commands. This can't be resolved trivially by moving bd_claim() before doing actual open() because that means an open attempt which will end up failing could interfere other legit O_EXCL open attempts. ie. unconfirmed open attempts can fail others. This patch resolves the problem by introducing claiming block which is started by bd_start_claiming() and terminated either by bd_claim() or bd_abort_claiming(). bd_claim() from inside a claiming block is guaranteed to succeed and once a claiming block is started, other bd_start_claiming() or bd_claim() attempts block till the current claiming block is terminated. bd_claim() can still be used standalone although now it always synchronizes against claiming blocks, so the existing users will keep working without any change. blkdev_open() and open_bdev_exclusive() are converted to use claiming blocks so that exclusive open attempts from these functions don't interfere with the existing exclusive open. This problem was discovered while investigating bko#15403. https://bugzilla.kernel.org/show_bug.cgi?id=15403 The burning problem itself can be resolved by updating userspace probing tools to always open w/ O_EXCL. Signed-off-by: Tejun Heo <tj@kernel.org> Reported-by: Matthias-Christian Ott <ott@mirix.org> Cc: Kay Sievers <kay.sievers@vrfy.org> Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
2010-04-07 09:53:59 +00:00
* @bdev: block device of interest
* @whole: the whole device containing @bdev, may equal @bdev
* @holder: holder trying to claim @bdev
*
* Claim @bdev. This function fails if @bdev is already claimed by another
* holder and waits if another claiming is in progress. return, the caller
* has ownership of bd_claiming and bd_holder[s].
block: implement bd_claiming and claiming block Currently, device claiming for exclusive open is done after low level open - disk->fops->open() - has completed successfully. This means that exclusive open attempts while a device is already exclusively open will fail only after disk->fops->open() is called. cdrom driver issues commands during open() which means that O_EXCL open attempt can unintentionally inject commands to in-progress command stream for burning thus disturbing burning process. In most cases, this doesn't cause problems because the first command to be issued is TUR which most devices can process in the middle of burning. However, depending on how a device replies to TUR during burning, cdrom driver may end up issuing further commands. This can't be resolved trivially by moving bd_claim() before doing actual open() because that means an open attempt which will end up failing could interfere other legit O_EXCL open attempts. ie. unconfirmed open attempts can fail others. This patch resolves the problem by introducing claiming block which is started by bd_start_claiming() and terminated either by bd_claim() or bd_abort_claiming(). bd_claim() from inside a claiming block is guaranteed to succeed and once a claiming block is started, other bd_start_claiming() or bd_claim() attempts block till the current claiming block is terminated. bd_claim() can still be used standalone although now it always synchronizes against claiming blocks, so the existing users will keep working without any change. blkdev_open() and open_bdev_exclusive() are converted to use claiming blocks so that exclusive open attempts from these functions don't interfere with the existing exclusive open. This problem was discovered while investigating bko#15403. https://bugzilla.kernel.org/show_bug.cgi?id=15403 The burning problem itself can be resolved by updating userspace probing tools to always open w/ O_EXCL. Signed-off-by: Tejun Heo <tj@kernel.org> Reported-by: Matthias-Christian Ott <ott@mirix.org> Cc: Kay Sievers <kay.sievers@vrfy.org> Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
2010-04-07 09:53:59 +00:00
*
* RETURNS:
* 0 if @bdev can be claimed, -EBUSY otherwise.
*/
int bd_prepare_to_claim(struct block_device *bdev, struct block_device *whole,
void *holder)
block: implement bd_claiming and claiming block Currently, device claiming for exclusive open is done after low level open - disk->fops->open() - has completed successfully. This means that exclusive open attempts while a device is already exclusively open will fail only after disk->fops->open() is called. cdrom driver issues commands during open() which means that O_EXCL open attempt can unintentionally inject commands to in-progress command stream for burning thus disturbing burning process. In most cases, this doesn't cause problems because the first command to be issued is TUR which most devices can process in the middle of burning. However, depending on how a device replies to TUR during burning, cdrom driver may end up issuing further commands. This can't be resolved trivially by moving bd_claim() before doing actual open() because that means an open attempt which will end up failing could interfere other legit O_EXCL open attempts. ie. unconfirmed open attempts can fail others. This patch resolves the problem by introducing claiming block which is started by bd_start_claiming() and terminated either by bd_claim() or bd_abort_claiming(). bd_claim() from inside a claiming block is guaranteed to succeed and once a claiming block is started, other bd_start_claiming() or bd_claim() attempts block till the current claiming block is terminated. bd_claim() can still be used standalone although now it always synchronizes against claiming blocks, so the existing users will keep working without any change. blkdev_open() and open_bdev_exclusive() are converted to use claiming blocks so that exclusive open attempts from these functions don't interfere with the existing exclusive open. This problem was discovered while investigating bko#15403. https://bugzilla.kernel.org/show_bug.cgi?id=15403 The burning problem itself can be resolved by updating userspace probing tools to always open w/ O_EXCL. Signed-off-by: Tejun Heo <tj@kernel.org> Reported-by: Matthias-Christian Ott <ott@mirix.org> Cc: Kay Sievers <kay.sievers@vrfy.org> Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
2010-04-07 09:53:59 +00:00
{
retry:
spin_lock(&bdev_lock);
block: implement bd_claiming and claiming block Currently, device claiming for exclusive open is done after low level open - disk->fops->open() - has completed successfully. This means that exclusive open attempts while a device is already exclusively open will fail only after disk->fops->open() is called. cdrom driver issues commands during open() which means that O_EXCL open attempt can unintentionally inject commands to in-progress command stream for burning thus disturbing burning process. In most cases, this doesn't cause problems because the first command to be issued is TUR which most devices can process in the middle of burning. However, depending on how a device replies to TUR during burning, cdrom driver may end up issuing further commands. This can't be resolved trivially by moving bd_claim() before doing actual open() because that means an open attempt which will end up failing could interfere other legit O_EXCL open attempts. ie. unconfirmed open attempts can fail others. This patch resolves the problem by introducing claiming block which is started by bd_start_claiming() and terminated either by bd_claim() or bd_abort_claiming(). bd_claim() from inside a claiming block is guaranteed to succeed and once a claiming block is started, other bd_start_claiming() or bd_claim() attempts block till the current claiming block is terminated. bd_claim() can still be used standalone although now it always synchronizes against claiming blocks, so the existing users will keep working without any change. blkdev_open() and open_bdev_exclusive() are converted to use claiming blocks so that exclusive open attempts from these functions don't interfere with the existing exclusive open. This problem was discovered while investigating bko#15403. https://bugzilla.kernel.org/show_bug.cgi?id=15403 The burning problem itself can be resolved by updating userspace probing tools to always open w/ O_EXCL. Signed-off-by: Tejun Heo <tj@kernel.org> Reported-by: Matthias-Christian Ott <ott@mirix.org> Cc: Kay Sievers <kay.sievers@vrfy.org> Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
2010-04-07 09:53:59 +00:00
/* if someone else claimed, fail */
if (!bd_may_claim(bdev, whole, holder)) {
spin_unlock(&bdev_lock);
block: implement bd_claiming and claiming block Currently, device claiming for exclusive open is done after low level open - disk->fops->open() - has completed successfully. This means that exclusive open attempts while a device is already exclusively open will fail only after disk->fops->open() is called. cdrom driver issues commands during open() which means that O_EXCL open attempt can unintentionally inject commands to in-progress command stream for burning thus disturbing burning process. In most cases, this doesn't cause problems because the first command to be issued is TUR which most devices can process in the middle of burning. However, depending on how a device replies to TUR during burning, cdrom driver may end up issuing further commands. This can't be resolved trivially by moving bd_claim() before doing actual open() because that means an open attempt which will end up failing could interfere other legit O_EXCL open attempts. ie. unconfirmed open attempts can fail others. This patch resolves the problem by introducing claiming block which is started by bd_start_claiming() and terminated either by bd_claim() or bd_abort_claiming(). bd_claim() from inside a claiming block is guaranteed to succeed and once a claiming block is started, other bd_start_claiming() or bd_claim() attempts block till the current claiming block is terminated. bd_claim() can still be used standalone although now it always synchronizes against claiming blocks, so the existing users will keep working without any change. blkdev_open() and open_bdev_exclusive() are converted to use claiming blocks so that exclusive open attempts from these functions don't interfere with the existing exclusive open. This problem was discovered while investigating bko#15403. https://bugzilla.kernel.org/show_bug.cgi?id=15403 The burning problem itself can be resolved by updating userspace probing tools to always open w/ O_EXCL. Signed-off-by: Tejun Heo <tj@kernel.org> Reported-by: Matthias-Christian Ott <ott@mirix.org> Cc: Kay Sievers <kay.sievers@vrfy.org> Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
2010-04-07 09:53:59 +00:00
return -EBUSY;
}
block: implement bd_claiming and claiming block Currently, device claiming for exclusive open is done after low level open - disk->fops->open() - has completed successfully. This means that exclusive open attempts while a device is already exclusively open will fail only after disk->fops->open() is called. cdrom driver issues commands during open() which means that O_EXCL open attempt can unintentionally inject commands to in-progress command stream for burning thus disturbing burning process. In most cases, this doesn't cause problems because the first command to be issued is TUR which most devices can process in the middle of burning. However, depending on how a device replies to TUR during burning, cdrom driver may end up issuing further commands. This can't be resolved trivially by moving bd_claim() before doing actual open() because that means an open attempt which will end up failing could interfere other legit O_EXCL open attempts. ie. unconfirmed open attempts can fail others. This patch resolves the problem by introducing claiming block which is started by bd_start_claiming() and terminated either by bd_claim() or bd_abort_claiming(). bd_claim() from inside a claiming block is guaranteed to succeed and once a claiming block is started, other bd_start_claiming() or bd_claim() attempts block till the current claiming block is terminated. bd_claim() can still be used standalone although now it always synchronizes against claiming blocks, so the existing users will keep working without any change. blkdev_open() and open_bdev_exclusive() are converted to use claiming blocks so that exclusive open attempts from these functions don't interfere with the existing exclusive open. This problem was discovered while investigating bko#15403. https://bugzilla.kernel.org/show_bug.cgi?id=15403 The burning problem itself can be resolved by updating userspace probing tools to always open w/ O_EXCL. Signed-off-by: Tejun Heo <tj@kernel.org> Reported-by: Matthias-Christian Ott <ott@mirix.org> Cc: Kay Sievers <kay.sievers@vrfy.org> Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
2010-04-07 09:53:59 +00:00
/* if claiming is already in progress, wait for it to finish */
if (whole->bd_claiming) {
block: implement bd_claiming and claiming block Currently, device claiming for exclusive open is done after low level open - disk->fops->open() - has completed successfully. This means that exclusive open attempts while a device is already exclusively open will fail only after disk->fops->open() is called. cdrom driver issues commands during open() which means that O_EXCL open attempt can unintentionally inject commands to in-progress command stream for burning thus disturbing burning process. In most cases, this doesn't cause problems because the first command to be issued is TUR which most devices can process in the middle of burning. However, depending on how a device replies to TUR during burning, cdrom driver may end up issuing further commands. This can't be resolved trivially by moving bd_claim() before doing actual open() because that means an open attempt which will end up failing could interfere other legit O_EXCL open attempts. ie. unconfirmed open attempts can fail others. This patch resolves the problem by introducing claiming block which is started by bd_start_claiming() and terminated either by bd_claim() or bd_abort_claiming(). bd_claim() from inside a claiming block is guaranteed to succeed and once a claiming block is started, other bd_start_claiming() or bd_claim() attempts block till the current claiming block is terminated. bd_claim() can still be used standalone although now it always synchronizes against claiming blocks, so the existing users will keep working without any change. blkdev_open() and open_bdev_exclusive() are converted to use claiming blocks so that exclusive open attempts from these functions don't interfere with the existing exclusive open. This problem was discovered while investigating bko#15403. https://bugzilla.kernel.org/show_bug.cgi?id=15403 The burning problem itself can be resolved by updating userspace probing tools to always open w/ O_EXCL. Signed-off-by: Tejun Heo <tj@kernel.org> Reported-by: Matthias-Christian Ott <ott@mirix.org> Cc: Kay Sievers <kay.sievers@vrfy.org> Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
2010-04-07 09:53:59 +00:00
wait_queue_head_t *wq = bit_waitqueue(&whole->bd_claiming, 0);
DEFINE_WAIT(wait);
prepare_to_wait(wq, &wait, TASK_UNINTERRUPTIBLE);
spin_unlock(&bdev_lock);
schedule();
finish_wait(wq, &wait);
goto retry;
}
/* yay, all mine */
whole->bd_claiming = holder;
spin_unlock(&bdev_lock);
block: implement bd_claiming and claiming block Currently, device claiming for exclusive open is done after low level open - disk->fops->open() - has completed successfully. This means that exclusive open attempts while a device is already exclusively open will fail only after disk->fops->open() is called. cdrom driver issues commands during open() which means that O_EXCL open attempt can unintentionally inject commands to in-progress command stream for burning thus disturbing burning process. In most cases, this doesn't cause problems because the first command to be issued is TUR which most devices can process in the middle of burning. However, depending on how a device replies to TUR during burning, cdrom driver may end up issuing further commands. This can't be resolved trivially by moving bd_claim() before doing actual open() because that means an open attempt which will end up failing could interfere other legit O_EXCL open attempts. ie. unconfirmed open attempts can fail others. This patch resolves the problem by introducing claiming block which is started by bd_start_claiming() and terminated either by bd_claim() or bd_abort_claiming(). bd_claim() from inside a claiming block is guaranteed to succeed and once a claiming block is started, other bd_start_claiming() or bd_claim() attempts block till the current claiming block is terminated. bd_claim() can still be used standalone although now it always synchronizes against claiming blocks, so the existing users will keep working without any change. blkdev_open() and open_bdev_exclusive() are converted to use claiming blocks so that exclusive open attempts from these functions don't interfere with the existing exclusive open. This problem was discovered while investigating bko#15403. https://bugzilla.kernel.org/show_bug.cgi?id=15403 The burning problem itself can be resolved by updating userspace probing tools to always open w/ O_EXCL. Signed-off-by: Tejun Heo <tj@kernel.org> Reported-by: Matthias-Christian Ott <ott@mirix.org> Cc: Kay Sievers <kay.sievers@vrfy.org> Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
2010-04-07 09:53:59 +00:00
return 0;
}
EXPORT_SYMBOL_GPL(bd_prepare_to_claim); /* only for the loop driver */
block: implement bd_claiming and claiming block Currently, device claiming for exclusive open is done after low level open - disk->fops->open() - has completed successfully. This means that exclusive open attempts while a device is already exclusively open will fail only after disk->fops->open() is called. cdrom driver issues commands during open() which means that O_EXCL open attempt can unintentionally inject commands to in-progress command stream for burning thus disturbing burning process. In most cases, this doesn't cause problems because the first command to be issued is TUR which most devices can process in the middle of burning. However, depending on how a device replies to TUR during burning, cdrom driver may end up issuing further commands. This can't be resolved trivially by moving bd_claim() before doing actual open() because that means an open attempt which will end up failing could interfere other legit O_EXCL open attempts. ie. unconfirmed open attempts can fail others. This patch resolves the problem by introducing claiming block which is started by bd_start_claiming() and terminated either by bd_claim() or bd_abort_claiming(). bd_claim() from inside a claiming block is guaranteed to succeed and once a claiming block is started, other bd_start_claiming() or bd_claim() attempts block till the current claiming block is terminated. bd_claim() can still be used standalone although now it always synchronizes against claiming blocks, so the existing users will keep working without any change. blkdev_open() and open_bdev_exclusive() are converted to use claiming blocks so that exclusive open attempts from these functions don't interfere with the existing exclusive open. This problem was discovered while investigating bko#15403. https://bugzilla.kernel.org/show_bug.cgi?id=15403 The burning problem itself can be resolved by updating userspace probing tools to always open w/ O_EXCL. Signed-off-by: Tejun Heo <tj@kernel.org> Reported-by: Matthias-Christian Ott <ott@mirix.org> Cc: Kay Sievers <kay.sievers@vrfy.org> Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
2010-04-07 09:53:59 +00:00
static struct gendisk *bdev_get_gendisk(struct block_device *bdev, int *partno)
{
struct gendisk *disk = get_gendisk(bdev->bd_dev, partno);
if (!disk)
return NULL;
/*
* Now that we hold gendisk reference we make sure bdev we looked up is
* not stale. If it is, it means device got removed and created before
* we looked up gendisk and we fail open in such case. Associating
* unhashed bdev with newly created gendisk could lead to two bdevs
* (and thus two independent caches) being associated with one device
* which is bad.
*/
if (inode_unhashed(bdev->bd_inode)) {
put_disk_and_module(disk);
return NULL;
}
return disk;
}
static void bd_clear_claiming(struct block_device *whole, void *holder)
{
lockdep_assert_held(&bdev_lock);
/* tell others that we're done */
BUG_ON(whole->bd_claiming != holder);
whole->bd_claiming = NULL;
wake_up_bit(&whole->bd_claiming, 0);
}
/**
* bd_finish_claiming - finish claiming of a block device
* @bdev: block device of interest
* @whole: whole block device
* @holder: holder that has claimed @bdev
*
* Finish exclusive open of a block device. Mark the device as exlusively
* open by the holder and wake up all waiters for exclusive open to finish.
*/
static void bd_finish_claiming(struct block_device *bdev,
struct block_device *whole, void *holder)
{
spin_lock(&bdev_lock);
BUG_ON(!bd_may_claim(bdev, whole, holder));
/*
* Note that for a whole device bd_holders will be incremented twice,
* and bd_holder will be set to bd_may_claim before being set to holder
*/
whole->bd_holders++;
whole->bd_holder = bd_may_claim;
bdev->bd_holders++;
bdev->bd_holder = holder;
bd_clear_claiming(whole, holder);
spin_unlock(&bdev_lock);
}
/**
* bd_abort_claiming - abort claiming of a block device
* @bdev: block device of interest
* @whole: whole block device
* @holder: holder that has claimed @bdev
*
* Abort claiming of a block device when the exclusive open failed. This can be
* also used when exclusive open is not actually desired and we just needed
* to block other exclusive openers for a while.
*/
void bd_abort_claiming(struct block_device *bdev, struct block_device *whole,
void *holder)
{
spin_lock(&bdev_lock);
bd_clear_claiming(whole, holder);
spin_unlock(&bdev_lock);
}
EXPORT_SYMBOL(bd_abort_claiming);
block: implement bd_claiming and claiming block Currently, device claiming for exclusive open is done after low level open - disk->fops->open() - has completed successfully. This means that exclusive open attempts while a device is already exclusively open will fail only after disk->fops->open() is called. cdrom driver issues commands during open() which means that O_EXCL open attempt can unintentionally inject commands to in-progress command stream for burning thus disturbing burning process. In most cases, this doesn't cause problems because the first command to be issued is TUR which most devices can process in the middle of burning. However, depending on how a device replies to TUR during burning, cdrom driver may end up issuing further commands. This can't be resolved trivially by moving bd_claim() before doing actual open() because that means an open attempt which will end up failing could interfere other legit O_EXCL open attempts. ie. unconfirmed open attempts can fail others. This patch resolves the problem by introducing claiming block which is started by bd_start_claiming() and terminated either by bd_claim() or bd_abort_claiming(). bd_claim() from inside a claiming block is guaranteed to succeed and once a claiming block is started, other bd_start_claiming() or bd_claim() attempts block till the current claiming block is terminated. bd_claim() can still be used standalone although now it always synchronizes against claiming blocks, so the existing users will keep working without any change. blkdev_open() and open_bdev_exclusive() are converted to use claiming blocks so that exclusive open attempts from these functions don't interfere with the existing exclusive open. This problem was discovered while investigating bko#15403. https://bugzilla.kernel.org/show_bug.cgi?id=15403 The burning problem itself can be resolved by updating userspace probing tools to always open w/ O_EXCL. Signed-off-by: Tejun Heo <tj@kernel.org> Reported-by: Matthias-Christian Ott <ott@mirix.org> Cc: Kay Sievers <kay.sievers@vrfy.org> Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
2010-04-07 09:53:59 +00:00
#ifdef CONFIG_SYSFS
struct bd_holder_disk {
struct list_head list;
struct gendisk *disk;
int refcnt;
};
static struct bd_holder_disk *bd_find_holder_disk(struct block_device *bdev,
struct gendisk *disk)
{
struct bd_holder_disk *holder;
list_for_each_entry(holder, &bdev->bd_holder_disks, list)
if (holder->disk == disk)
return holder;
return NULL;
}
[PATCH] blockdev.c: check driver layer errors Check driver layer errors. Fix from: "Jun'ichi Nomura" <j-nomura@ce.jp.nec.com> In blockdevc-check-errors.patch, add_bd_holder() is modified to return error values when some of its operation failed. Among them, it returns -EEXIST when a given bd_holder object already exists in the list. However, in this case, the function completed its work successfully and need no action by its caller other than freeing unused bd_holder object. So I think it's better to return success after freeing by itself. Otherwise, bd_claim-ing with same claim pointer will fail. Typically, lvresize will fails with following message: device-mapper: reload ioctl failed: Invalid argument and you'll see messages like below in kernel log: device-mapper: table: 254:13: linear: dm-linear: Device lookup failed device-mapper: ioctl: error adding target to table Similarly, it should not add bd_holder to the list if either one of symlinking fails. I don't have a test case for this to happen but it should cause dereference of freed pointer. If a matching bd_holder is found in bd_holder_list, add_bd_holder() completes its job by just incrementing the reference count. In this case, it should be considered as success but it used to return 'fail' to let the caller free temporary bd_holder. Fixed it to return success and free given object by itself. Also, if either one of symlinking fails, the bd_holder should not be added to the list so that it can be discarded later. Otherwise, the caller will free bd_holder which is in the list. Signed-off-by: Jun'ichi Nomura <j-nomura@ce.jp.nec.com> Cc: "Randy.Dunlap" <rdunlap@xenotime.net> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-09-29 08:58:56 +00:00
static int add_symlink(struct kobject *from, struct kobject *to)
{
[PATCH] blockdev.c: check driver layer errors Check driver layer errors. Fix from: "Jun'ichi Nomura" <j-nomura@ce.jp.nec.com> In blockdevc-check-errors.patch, add_bd_holder() is modified to return error values when some of its operation failed. Among them, it returns -EEXIST when a given bd_holder object already exists in the list. However, in this case, the function completed its work successfully and need no action by its caller other than freeing unused bd_holder object. So I think it's better to return success after freeing by itself. Otherwise, bd_claim-ing with same claim pointer will fail. Typically, lvresize will fails with following message: device-mapper: reload ioctl failed: Invalid argument and you'll see messages like below in kernel log: device-mapper: table: 254:13: linear: dm-linear: Device lookup failed device-mapper: ioctl: error adding target to table Similarly, it should not add bd_holder to the list if either one of symlinking fails. I don't have a test case for this to happen but it should cause dereference of freed pointer. If a matching bd_holder is found in bd_holder_list, add_bd_holder() completes its job by just incrementing the reference count. In this case, it should be considered as success but it used to return 'fail' to let the caller free temporary bd_holder. Fixed it to return success and free given object by itself. Also, if either one of symlinking fails, the bd_holder should not be added to the list so that it can be discarded later. Otherwise, the caller will free bd_holder which is in the list. Signed-off-by: Jun'ichi Nomura <j-nomura@ce.jp.nec.com> Cc: "Randy.Dunlap" <rdunlap@xenotime.net> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-09-29 08:58:56 +00:00
return sysfs_create_link(from, to, kobject_name(to));
}
static void del_symlink(struct kobject *from, struct kobject *to)
{
sysfs_remove_link(from, kobject_name(to));
}
/**
* bd_link_disk_holder - create symlinks between holding disk and slave bdev
* @bdev: the claimed slave bdev
* @disk: the holding disk
*
* DON'T USE THIS UNLESS YOU'RE ALREADY USING IT.
*
* This functions creates the following sysfs symlinks.
*
* - from "slaves" directory of the holder @disk to the claimed @bdev
* - from "holders" directory of the @bdev to the holder @disk
*
* For example, if /dev/dm-0 maps to /dev/sda and disk for dm-0 is
* passed to bd_link_disk_holder(), then:
*
* /sys/block/dm-0/slaves/sda --> /sys/block/sda
* /sys/block/sda/holders/dm-0 --> /sys/block/dm-0
*
* The caller must have claimed @bdev before calling this function and
* ensure that both @bdev and @disk are valid during the creation and
* lifetime of these symlinks.
*
* CONTEXT:
* Might sleep.
*
* RETURNS:
* 0 on success, -errno on failure.
*/
int bd_link_disk_holder(struct block_device *bdev, struct gendisk *disk)
{
struct bd_holder_disk *holder;
int ret = 0;
mutex_lock(&bdev->bd_mutex);
WARN_ON_ONCE(!bdev->bd_holder);
/* FIXME: remove the following once add_disk() handles errors */
if (WARN_ON(!disk->slave_dir || !bdev->bd_part->holder_dir))
goto out_unlock;
holder = bd_find_holder_disk(bdev, disk);
if (holder) {
holder->refcnt++;
goto out_unlock;
}
holder = kzalloc(sizeof(*holder), GFP_KERNEL);
if (!holder) {
ret = -ENOMEM;
goto out_unlock;
}
INIT_LIST_HEAD(&holder->list);
holder->disk = disk;
holder->refcnt = 1;
ret = add_symlink(disk->slave_dir, bdev_kobj(bdev));
if (ret)
goto out_free;
ret = add_symlink(bdev->bd_part->holder_dir, &disk_to_dev(disk)->kobj);
if (ret)
goto out_del;
/*
* bdev could be deleted beneath us which would implicitly destroy
* the holder directory. Hold on to it.
*/
kobject_get(bdev->bd_part->holder_dir);
list_add(&holder->list, &bdev->bd_holder_disks);
goto out_unlock;
out_del:
del_symlink(disk->slave_dir, bdev_kobj(bdev));
out_free:
kfree(holder);
out_unlock:
mutex_unlock(&bdev->bd_mutex);
return ret;
}
EXPORT_SYMBOL_GPL(bd_link_disk_holder);
/**
* bd_unlink_disk_holder - destroy symlinks created by bd_link_disk_holder()
* @bdev: the calimed slave bdev
* @disk: the holding disk
*
* DON'T USE THIS UNLESS YOU'RE ALREADY USING IT.
*
* CONTEXT:
* Might sleep.
*/
void bd_unlink_disk_holder(struct block_device *bdev, struct gendisk *disk)
{
struct bd_holder_disk *holder;
mutex_lock(&bdev->bd_mutex);
holder = bd_find_holder_disk(bdev, disk);
if (!WARN_ON_ONCE(holder == NULL) && !--holder->refcnt) {
del_symlink(disk->slave_dir, bdev_kobj(bdev));
del_symlink(bdev->bd_part->holder_dir,
&disk_to_dev(disk)->kobj);
kobject_put(bdev->bd_part->holder_dir);
list_del_init(&holder->list);
kfree(holder);
}
mutex_unlock(&bdev->bd_mutex);
}
EXPORT_SYMBOL_GPL(bd_unlink_disk_holder);
#endif
/**
* check_disk_size_change - checks for disk size change and adjusts bdev size.
* @disk: struct gendisk to check
* @bdev: struct bdev to adjust.
* @verbose: if %true log a message about a size change if there is any
*
* This routine checks to see if the bdev size does not match the disk size
* and adjusts it if it differs. When shrinking the bdev size, its all caches
* are freed.
*/
static void check_disk_size_change(struct gendisk *disk,
struct block_device *bdev, bool verbose)
{
loff_t disk_size, bdev_size;
spin_lock(&bdev->bd_size_lock);
disk_size = (loff_t)get_capacity(disk) << 9;
bdev_size = i_size_read(bdev->bd_inode);
if (disk_size != bdev_size) {
if (verbose) {
printk(KERN_INFO
"%s: detected capacity change from %lld to %lld\n",
disk->disk_name, bdev_size, disk_size);
}
i_size_write(bdev->bd_inode, disk_size);
}
spin_unlock(&bdev->bd_size_lock);
}
/**
* revalidate_disk_size - checks for disk size change and adjusts bdev size.
* @disk: struct gendisk to check
* @verbose: if %true log a message about a size change if there is any
*
* This routine checks to see if the bdev size does not match the disk size
* and adjusts it if it differs. When shrinking the bdev size, its all caches
* are freed.
*/
void revalidate_disk_size(struct gendisk *disk, bool verbose)
{
struct block_device *bdev;
/*
* Hidden disks don't have associated bdev so there's no point in
* revalidating them.
*/
if (disk->flags & GENHD_FL_HIDDEN)
return;
bdev = bdget_disk(disk, 0);
if (bdev) {
check_disk_size_change(disk, bdev, verbose);
bdput(bdev);
}
}
void bd_set_nr_sectors(struct block_device *bdev, sector_t sectors)
{
spin_lock(&bdev->bd_size_lock);
i_size_write(bdev->bd_inode, (loff_t)sectors << SECTOR_SHIFT);
spin_unlock(&bdev->bd_size_lock);
}
EXPORT_SYMBOL(bd_set_nr_sectors);
static void __blkdev_put(struct block_device *bdev, fmode_t mode, int for_part);
[PATCH] lockdep: simplify some aspects of bd_mutex nesting When we open (actually blkdev_get) a partition we need to also open (get) the whole device that holds the partition. The involves some limited recursion. This patch tries to simplify some aspects of this. As well as opening the whole device, we need to increment ->bd_part_count when a partition is opened (this is used by rescan_partitions to avoid a rescan if any partition is active, as that would be confusing). The main change this patch makes is to move the inc/dec of bd_part_count into blkdev_{get,put} for the whole rather than doing it in blkdev_{get,put} for the partition. More specifically, we introduce __blkdev_get and __blkdev_put which do exactly what blkdev_{get,put} did, only with an extra "for_part" argument (blkget_{get,put} then call the __ version with a '0' for the extra argument). If for_part is 1, then the blkdev is being get(put) because a partition is being opened(closed) for the first(last) time, and so bd_part_count should be updated (on success). The particular advantage of pushing this function down is that the bd_mutex lock (which is needed to update bd_part_count) is already held at the lower level. Note that this slightly changes the semantics of bd_part_count. Instead of updating it whenever a partition is opened or released, it is now only updated on the first open or last release. This is an adequate semantic as it is only ever tested for "== 0". Having introduced these functions we remove the current bd_part_count updates from do_open (which is really the body of blkdev_get) and call __blkdev_get(... 1). Similarly in blkget_put we remove the old bd_part_count updates and call __blkget_put(..., 1). This call is moved to the end of __blkdev_put to avoid nested locks of bd_mutex. Finally the mutex_lock on whole->bd_mutex in do_open can be removed. It was only really needed to protect bd_part_count, and that is now managed (and protected) within the recursive call. The observation that bd_part_count is central to the locking issues, and the modifications to create __blkdev_put are from Peter Zijlstra. Cc: Ingo Molnar <mingo@elte.hu> Acked-by: Peter Zijlstra <a.p.zijlstra@chello.nl> Signed-off-by: Neil Brown <neilb@suse.de> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-12-08 10:36:16 +00:00
int bdev_disk_changed(struct block_device *bdev, bool invalidate)
{
struct gendisk *disk = bdev->bd_disk;
int ret;
lockdep_assert_held(&bdev->bd_mutex);
clear_bit(GD_NEED_PART_SCAN, &bdev->bd_disk->state);
rescan:
ret = blk_drop_partitions(bdev);
if (ret)
return ret;
/*
* Historically we only set the capacity to zero for devices that
* support partitions (independ of actually having partitions created).
* Doing that is rather inconsistent, but changing it broke legacy
* udisks polling for legacy ide-cdrom devices. Use the crude check
* below to get the sane behavior for most device while not breaking
* userspace for this particular setup.
*/
if (invalidate) {
if (disk_part_scan_enabled(disk) ||
!(disk->flags & GENHD_FL_REMOVABLE))
set_capacity(disk, 0);
} else {
if (disk->fops->revalidate_disk)
disk->fops->revalidate_disk(disk);
}
check_disk_size_change(disk, bdev, !invalidate);
if (get_capacity(disk)) {
ret = blk_add_partitions(disk, bdev);
if (ret == -EAGAIN)
goto rescan;
} else if (invalidate) {
/*
* Tell userspace that the media / partition table may have
* changed.
*/
kobject_uevent(&disk_to_dev(disk)->kobj, KOBJ_CHANGE);
}
return ret;
}
/*
* Only exported for for loop and dasd for historic reasons. Don't use in new
* code!
*/
EXPORT_SYMBOL_GPL(bdev_disk_changed);
/*
* bd_mutex locking:
*
* mutex_lock(part->bd_mutex)
* mutex_lock_nested(whole->bd_mutex, 1)
*/
static int __blkdev_get(struct block_device *bdev, struct gendisk *disk,
int partno, fmode_t mode)
{
int ret;
if (!bdev->bd_openers) {
bdev->bd_disk = disk;
bdev->bd_contains = bdev;
bdev->bd_partno = partno;
if (!partno) {
ret = -ENXIO;
bdev->bd_part = disk_get_part(disk, partno);
if (!bdev->bd_part)
goto out_clear;
ret = 0;
if (disk->fops->open)
ret = disk->fops->open(bdev, mode);
blockdev: Fix livelocks on loop device bd_set_size() updates also block device's block size. This is somewhat unexpected from its name and at this point, only blkdev_open() uses this functionality. Furthermore, this can result in changing block size under a filesystem mounted on a loop device which leads to livelocks inside __getblk_gfp() like: Sending NMI from CPU 0 to CPUs 1: NMI backtrace for cpu 1 CPU: 1 PID: 10863 Comm: syz-executor0 Not tainted 4.18.0-rc5+ #151 Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 01/01/2011 RIP: 0010:__sanitizer_cov_trace_pc+0x3f/0x50 kernel/kcov.c:106 ... Call Trace: init_page_buffers+0x3e2/0x530 fs/buffer.c:904 grow_dev_page fs/buffer.c:947 [inline] grow_buffers fs/buffer.c:1009 [inline] __getblk_slow fs/buffer.c:1036 [inline] __getblk_gfp+0x906/0xb10 fs/buffer.c:1313 __bread_gfp+0x2d/0x310 fs/buffer.c:1347 sb_bread include/linux/buffer_head.h:307 [inline] fat12_ent_bread+0x14e/0x3d0 fs/fat/fatent.c:75 fat_ent_read_block fs/fat/fatent.c:441 [inline] fat_alloc_clusters+0x8ce/0x16e0 fs/fat/fatent.c:489 fat_add_cluster+0x7a/0x150 fs/fat/inode.c:101 __fat_get_block fs/fat/inode.c:148 [inline] ... Trivial reproducer for the problem looks like: truncate -s 1G /tmp/image losetup /dev/loop0 /tmp/image mkfs.ext4 -b 1024 /dev/loop0 mount -t ext4 /dev/loop0 /mnt losetup -c /dev/loop0 l /mnt Fix the problem by moving initialization of a block device block size into a separate function and call it when needed. Thanks to Tetsuo Handa <penguin-kernel@I-love.SAKURA.ne.jp> for help with debugging the problem. Reported-by: syzbot+9933e4476f365f5d5a1b@syzkaller.appspotmail.com Signed-off-by: Jan Kara <jack@suse.cz> Signed-off-by: Jens Axboe <axboe@kernel.dk>
2019-01-14 08:48:10 +00:00
if (!ret) {
bd_set_nr_sectors(bdev, get_capacity(disk));
blockdev: Fix livelocks on loop device bd_set_size() updates also block device's block size. This is somewhat unexpected from its name and at this point, only blkdev_open() uses this functionality. Furthermore, this can result in changing block size under a filesystem mounted on a loop device which leads to livelocks inside __getblk_gfp() like: Sending NMI from CPU 0 to CPUs 1: NMI backtrace for cpu 1 CPU: 1 PID: 10863 Comm: syz-executor0 Not tainted 4.18.0-rc5+ #151 Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 01/01/2011 RIP: 0010:__sanitizer_cov_trace_pc+0x3f/0x50 kernel/kcov.c:106 ... Call Trace: init_page_buffers+0x3e2/0x530 fs/buffer.c:904 grow_dev_page fs/buffer.c:947 [inline] grow_buffers fs/buffer.c:1009 [inline] __getblk_slow fs/buffer.c:1036 [inline] __getblk_gfp+0x906/0xb10 fs/buffer.c:1313 __bread_gfp+0x2d/0x310 fs/buffer.c:1347 sb_bread include/linux/buffer_head.h:307 [inline] fat12_ent_bread+0x14e/0x3d0 fs/fat/fatent.c:75 fat_ent_read_block fs/fat/fatent.c:441 [inline] fat_alloc_clusters+0x8ce/0x16e0 fs/fat/fatent.c:489 fat_add_cluster+0x7a/0x150 fs/fat/inode.c:101 __fat_get_block fs/fat/inode.c:148 [inline] ... Trivial reproducer for the problem looks like: truncate -s 1G /tmp/image losetup /dev/loop0 /tmp/image mkfs.ext4 -b 1024 /dev/loop0 mount -t ext4 /dev/loop0 /mnt losetup -c /dev/loop0 l /mnt Fix the problem by moving initialization of a block device block size into a separate function and call it when needed. Thanks to Tetsuo Handa <penguin-kernel@I-love.SAKURA.ne.jp> for help with debugging the problem. Reported-by: syzbot+9933e4476f365f5d5a1b@syzkaller.appspotmail.com Signed-off-by: Jan Kara <jack@suse.cz> Signed-off-by: Jens Axboe <axboe@kernel.dk>
2019-01-14 08:48:10 +00:00
set_init_blocksize(bdev);
}
/*
* If the device is invalidated, rescan partition
* if open succeeded or failed with -ENOMEDIUM.
* The latter is necessary to prevent ghost
* partitions on a removed medium.
*/
if (test_bit(GD_NEED_PART_SCAN, &disk->state) &&
(!ret || ret == -ENOMEDIUM))
bdev_disk_changed(bdev, ret == -ENOMEDIUM);
if (ret)
goto out_clear;
} else {
struct block_device *whole = bdget_disk(disk, 0);
mutex_lock_nested(&whole->bd_mutex, 1);
ret = __blkdev_get(whole, disk, 0, mode);
if (ret) {
mutex_unlock(&whole->bd_mutex);
bdput(whole);
goto out_clear;
}
whole->bd_part_count++;
mutex_unlock(&whole->bd_mutex);
bdev->bd_contains = whole;
bdev->bd_part = disk_get_part(disk, partno);
if (!(disk->flags & GENHD_FL_UP) ||
!bdev->bd_part || !bdev->bd_part->nr_sects) {
__blkdev_put(whole, mode, 1);
bdput(whole);
ret = -ENXIO;
goto out_clear;
}
bd_set_nr_sectors(bdev, bdev->bd_part->nr_sects);
blockdev: Fix livelocks on loop device bd_set_size() updates also block device's block size. This is somewhat unexpected from its name and at this point, only blkdev_open() uses this functionality. Furthermore, this can result in changing block size under a filesystem mounted on a loop device which leads to livelocks inside __getblk_gfp() like: Sending NMI from CPU 0 to CPUs 1: NMI backtrace for cpu 1 CPU: 1 PID: 10863 Comm: syz-executor0 Not tainted 4.18.0-rc5+ #151 Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 01/01/2011 RIP: 0010:__sanitizer_cov_trace_pc+0x3f/0x50 kernel/kcov.c:106 ... Call Trace: init_page_buffers+0x3e2/0x530 fs/buffer.c:904 grow_dev_page fs/buffer.c:947 [inline] grow_buffers fs/buffer.c:1009 [inline] __getblk_slow fs/buffer.c:1036 [inline] __getblk_gfp+0x906/0xb10 fs/buffer.c:1313 __bread_gfp+0x2d/0x310 fs/buffer.c:1347 sb_bread include/linux/buffer_head.h:307 [inline] fat12_ent_bread+0x14e/0x3d0 fs/fat/fatent.c:75 fat_ent_read_block fs/fat/fatent.c:441 [inline] fat_alloc_clusters+0x8ce/0x16e0 fs/fat/fatent.c:489 fat_add_cluster+0x7a/0x150 fs/fat/inode.c:101 __fat_get_block fs/fat/inode.c:148 [inline] ... Trivial reproducer for the problem looks like: truncate -s 1G /tmp/image losetup /dev/loop0 /tmp/image mkfs.ext4 -b 1024 /dev/loop0 mount -t ext4 /dev/loop0 /mnt losetup -c /dev/loop0 l /mnt Fix the problem by moving initialization of a block device block size into a separate function and call it when needed. Thanks to Tetsuo Handa <penguin-kernel@I-love.SAKURA.ne.jp> for help with debugging the problem. Reported-by: syzbot+9933e4476f365f5d5a1b@syzkaller.appspotmail.com Signed-off-by: Jan Kara <jack@suse.cz> Signed-off-by: Jens Axboe <axboe@kernel.dk>
2019-01-14 08:48:10 +00:00
set_init_blocksize(bdev);
}
if (bdev->bd_bdi == &noop_backing_dev_info)
bdev->bd_bdi = bdi_get(disk->queue->backing_dev_info);
} else {
if (bdev->bd_contains == bdev) {
ret = 0;
if (bdev->bd_disk->fops->open)
ret = bdev->bd_disk->fops->open(bdev, mode);
/* the same as first opener case, read comment there */
if (test_bit(GD_NEED_PART_SCAN, &disk->state) &&
(!ret || ret == -ENOMEDIUM))
bdev_disk_changed(bdev, ret == -ENOMEDIUM);
if (ret)
return ret;
}
}
bdev->bd_openers++;
return 0;
out_clear:
disk_put_part(bdev->bd_part);
bdev->bd_disk = NULL;
bdev->bd_part = NULL;
bdev->bd_contains = NULL;
return ret;
}
/**
* blkdev_get - open a block device
* @bdev: block_device to open
* @mode: FMODE_* mask
* @holder: exclusive holder identifier
*
* Open @bdev with @mode. If @mode includes %FMODE_EXCL, @bdev is
* open with exclusive access. Specifying %FMODE_EXCL with %NULL
* @holder is invalid. Exclusive opens may nest for the same @holder.
*
* On success, the reference count of @bdev is unchanged. On failure,
* @bdev is put.
*
* CONTEXT:
* Might sleep.
*
* RETURNS:
* 0 on success, -errno on failure.
*/
static int blkdev_get(struct block_device *bdev, fmode_t mode, void *holder)
{
struct block_device *claiming;
bool unblock_events = true;
struct gendisk *disk;
int partno;
int ret;
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 <tj@kernel.org> Acked-by: Neil Brown <neilb@suse.de> Acked-by: Ryusuke Konishi <konishi.ryusuke@lab.ntt.co.jp> Acked-by: Mike Snitzer <snitzer@redhat.com> Acked-by: Philipp Reisner <philipp.reisner@linbit.com> Cc: Peter Osterlund <petero2@telia.com> Cc: Martin Schwidefsky <schwidefsky@de.ibm.com> Cc: Heiko Carstens <heiko.carstens@de.ibm.com> Cc: Jan Kara <jack@suse.cz> Cc: Andrew Morton <akpm@linux-foundation.org> Cc: Andreas Dilger <adilger.kernel@dilger.ca> Cc: "Theodore Ts'o" <tytso@mit.edu> Cc: Mark Fasheh <mfasheh@suse.com> Cc: Joel Becker <joel.becker@oracle.com> Cc: Alex Elder <aelder@sgi.com> Cc: Christoph Hellwig <hch@infradead.org> Cc: dm-devel@redhat.com Cc: drbd-dev@lists.linbit.com Cc: Leo Chen <leochen@broadcom.com> Cc: Scott Branden <sbranden@broadcom.com> Cc: Chris Mason <chris.mason@oracle.com> Cc: Steven Whitehouse <swhiteho@redhat.com> Cc: Dave Kleikamp <shaggy@linux.vnet.ibm.com> Cc: Joern Engel <joern@logfs.org> Cc: reiserfs-devel@vger.kernel.org Cc: Alexander Viro <viro@zeniv.linux.org.uk>
2010-11-13 10:55:17 +00:00
ret = devcgroup_check_permission(DEVCG_DEV_BLOCK,
imajor(bdev->bd_inode), iminor(bdev->bd_inode),
((mode & FMODE_READ) ? DEVCG_ACC_READ : 0) |
((mode & FMODE_WRITE) ? DEVCG_ACC_WRITE : 0));
if (ret)
goto bdput;
/*
* If we lost a race with 'disk' being deleted, try again. See md.c.
*/
retry:
ret = -ENXIO;
disk = bdev_get_gendisk(bdev, &partno);
if (!disk)
goto bdput;
if (mode & FMODE_EXCL) {
WARN_ON_ONCE(!holder);
ret = -ENOMEM;
claiming = bdget_disk(disk, 0);
if (!claiming)
goto put_disk;
ret = bd_prepare_to_claim(bdev, claiming, holder);
if (ret)
goto put_claiming;
}
disk_block_events(disk);
mutex_lock(&bdev->bd_mutex);
ret =__blkdev_get(bdev, disk, partno, mode);
if (!(mode & FMODE_EXCL)) {
; /* nothing to do here */
} else if (ret) {
bd_abort_claiming(bdev, claiming, holder);
} else {
bd_finish_claiming(bdev, claiming, holder);
/*
* 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 ((mode & FMODE_WRITE) && !bdev->bd_write_holder &&
(disk->flags & GENHD_FL_BLOCK_EVENTS_ON_EXCL_WRITE)) {
bdev->bd_write_holder = true;
unblock_events = false;
}
}
mutex_unlock(&bdev->bd_mutex);
if (unblock_events)
disk_unblock_events(disk);
put_claiming:
if (mode & FMODE_EXCL)
bdput(claiming);
put_disk:
if (ret)
put_disk_and_module(disk);
if (ret == -ERESTARTSYS)
goto retry;
bdput:
if (ret)
bdput(bdev);
return ret;
[PATCH] lockdep: simplify some aspects of bd_mutex nesting When we open (actually blkdev_get) a partition we need to also open (get) the whole device that holds the partition. The involves some limited recursion. This patch tries to simplify some aspects of this. As well as opening the whole device, we need to increment ->bd_part_count when a partition is opened (this is used by rescan_partitions to avoid a rescan if any partition is active, as that would be confusing). The main change this patch makes is to move the inc/dec of bd_part_count into blkdev_{get,put} for the whole rather than doing it in blkdev_{get,put} for the partition. More specifically, we introduce __blkdev_get and __blkdev_put which do exactly what blkdev_{get,put} did, only with an extra "for_part" argument (blkget_{get,put} then call the __ version with a '0' for the extra argument). If for_part is 1, then the blkdev is being get(put) because a partition is being opened(closed) for the first(last) time, and so bd_part_count should be updated (on success). The particular advantage of pushing this function down is that the bd_mutex lock (which is needed to update bd_part_count) is already held at the lower level. Note that this slightly changes the semantics of bd_part_count. Instead of updating it whenever a partition is opened or released, it is now only updated on the first open or last release. This is an adequate semantic as it is only ever tested for "== 0". Having introduced these functions we remove the current bd_part_count updates from do_open (which is really the body of blkdev_get) and call __blkdev_get(... 1). Similarly in blkget_put we remove the old bd_part_count updates and call __blkget_put(..., 1). This call is moved to the end of __blkdev_put to avoid nested locks of bd_mutex. Finally the mutex_lock on whole->bd_mutex in do_open can be removed. It was only really needed to protect bd_part_count, and that is now managed (and protected) within the recursive call. The observation that bd_part_count is central to the locking issues, and the modifications to create __blkdev_put are from Peter Zijlstra. Cc: Ingo Molnar <mingo@elte.hu> Acked-by: Peter Zijlstra <a.p.zijlstra@chello.nl> Signed-off-by: Neil Brown <neilb@suse.de> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-12-08 10:36:16 +00:00
}
/**
* blkdev_get_by_path - open a block device by name
* @path: path to the block device to open
* @mode: FMODE_* mask
* @holder: exclusive holder identifier
*
* Open the blockdevice described by the device file at @path. @mode
* and @holder are identical to blkdev_get().
*
* On success, the returned block_device has reference count of one.
*
* CONTEXT:
* Might sleep.
*
* RETURNS:
* Pointer to block_device on success, ERR_PTR(-errno) on failure.
*/
struct block_device *blkdev_get_by_path(const char *path, fmode_t mode,
void *holder)
{
struct block_device *bdev;
int err;
bdev = lookup_bdev(path);
if (IS_ERR(bdev))
return bdev;
err = blkdev_get(bdev, mode, holder);
if (err)
return ERR_PTR(err);
if ((mode & FMODE_WRITE) && bdev_read_only(bdev)) {
blkdev_put(bdev, mode);
return ERR_PTR(-EACCES);
}
return bdev;
}
EXPORT_SYMBOL(blkdev_get_by_path);
/**
* blkdev_get_by_dev - open a block device by device number
* @dev: device number of block device to open
* @mode: FMODE_* mask
* @holder: exclusive holder identifier
*
* Open the blockdevice described by device number @dev. @mode and
* @holder are identical to blkdev_get().
*
* Use it ONLY if you really do not have anything better - i.e. when
* you are behind a truly sucky interface and all you are given is a
* device number. _Never_ to be used for internal purposes. If you
* ever need it - reconsider your API.
*
* On success, the returned block_device has reference count of one.
*
* CONTEXT:
* Might sleep.
*
* RETURNS:
* Pointer to block_device on success, ERR_PTR(-errno) on failure.
*/
struct block_device *blkdev_get_by_dev(dev_t dev, fmode_t mode, void *holder)
{
struct block_device *bdev;
int err;
bdev = bdget(dev);
if (!bdev)
return ERR_PTR(-ENOMEM);
err = blkdev_get(bdev, mode, holder);
if (err)
return ERR_PTR(err);
return bdev;
}
EXPORT_SYMBOL(blkdev_get_by_dev);
static int blkdev_open(struct inode * inode, struct file * filp)
{
struct block_device *bdev;
/*
* Preserve backwards compatibility and allow large file access
* even if userspace doesn't ask for it explicitly. Some mkfs
* binary needs it. We might want to drop this workaround
* during an unstable branch.
*/
filp->f_flags |= O_LARGEFILE;
filp->f_mode |= FMODE_NOWAIT | FMODE_BUF_RASYNC;
if (filp->f_flags & O_NDELAY)
filp->f_mode |= FMODE_NDELAY;
if (filp->f_flags & O_EXCL)
filp->f_mode |= FMODE_EXCL;
if ((filp->f_flags & O_ACCMODE) == 3)
filp->f_mode |= FMODE_WRITE_IOCTL;
bdev = bd_acquire(inode);
if (bdev == NULL)
return -ENOMEM;
filp->f_mapping = bdev->bd_inode->i_mapping;
fs: new infrastructure for writeback error handling and reporting Most filesystems currently use mapping_set_error and filemap_check_errors for setting and reporting/clearing writeback errors at the mapping level. filemap_check_errors is indirectly called from most of the filemap_fdatawait_* functions and from filemap_write_and_wait*. These functions are called from all sorts of contexts to wait on writeback to finish -- e.g. mostly in fsync, but also in truncate calls, getattr, etc. The non-fsync callers are problematic. We should be reporting writeback errors during fsync, but many places spread over the tree clear out errors before they can be properly reported, or report errors at nonsensical times. If I get -EIO on a stat() call, there is no reason for me to assume that it is because some previous writeback failed. The fact that it also clears out the error such that a subsequent fsync returns 0 is a bug, and a nasty one since that's potentially silent data corruption. This patch adds a small bit of new infrastructure for setting and reporting errors during address_space writeback. While the above was my original impetus for adding this, I think it's also the case that current fsync semantics are just problematic for userland. Most applications that call fsync do so to ensure that the data they wrote has hit the backing store. In the case where there are multiple writers to the file at the same time, this is really hard to determine. The first one to call fsync will see any stored error, and the rest get back 0. The processes with open fds may not be associated with one another in any way. They could even be in different containers, so ensuring coordination between all fsync callers is not really an option. One way to remedy this would be to track what file descriptor was used to dirty the file, but that's rather cumbersome and would likely be slow. However, there is a simpler way to improve the semantics here without incurring too much overhead. This set adds an errseq_t to struct address_space, and a corresponding one is added to struct file. Writeback errors are recorded in the mapping's errseq_t, and the one in struct file is used as the "since" value. This changes the semantics of the Linux fsync implementation such that applications can now use it to determine whether there were any writeback errors since fsync(fd) was last called (or since the file was opened in the case of fsync having never been called). Note that those writeback errors may have occurred when writing data that was dirtied via an entirely different fd, but that's the case now with the current mapping_set_error/filemap_check_error infrastructure. This will at least prevent you from getting a false report of success. The new behavior is still consistent with the POSIX spec, and is more reliable for application developers. This patch just adds some basic infrastructure for doing this, and ensures that the f_wb_err "cursor" is properly set when a file is opened. Later patches will change the existing code to use this new infrastructure for reporting errors at fsync time. Signed-off-by: Jeff Layton <jlayton@redhat.com> Reviewed-by: Jan Kara <jack@suse.cz>
2017-07-06 11:02:25 +00:00
filp->f_wb_err = filemap_sample_wb_err(filp->f_mapping);
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 <tj@kernel.org> Acked-by: Neil Brown <neilb@suse.de> Acked-by: Ryusuke Konishi <konishi.ryusuke@lab.ntt.co.jp> Acked-by: Mike Snitzer <snitzer@redhat.com> Acked-by: Philipp Reisner <philipp.reisner@linbit.com> Cc: Peter Osterlund <petero2@telia.com> Cc: Martin Schwidefsky <schwidefsky@de.ibm.com> Cc: Heiko Carstens <heiko.carstens@de.ibm.com> Cc: Jan Kara <jack@suse.cz> Cc: Andrew Morton <akpm@linux-foundation.org> Cc: Andreas Dilger <adilger.kernel@dilger.ca> Cc: "Theodore Ts'o" <tytso@mit.edu> Cc: Mark Fasheh <mfasheh@suse.com> Cc: Joel Becker <joel.becker@oracle.com> Cc: Alex Elder <aelder@sgi.com> Cc: Christoph Hellwig <hch@infradead.org> Cc: dm-devel@redhat.com Cc: drbd-dev@lists.linbit.com Cc: Leo Chen <leochen@broadcom.com> Cc: Scott Branden <sbranden@broadcom.com> Cc: Chris Mason <chris.mason@oracle.com> Cc: Steven Whitehouse <swhiteho@redhat.com> Cc: Dave Kleikamp <shaggy@linux.vnet.ibm.com> Cc: Joern Engel <joern@logfs.org> Cc: reiserfs-devel@vger.kernel.org Cc: Alexander Viro <viro@zeniv.linux.org.uk>
2010-11-13 10:55:17 +00:00
return blkdev_get(bdev, filp->f_mode, filp);
}
static void __blkdev_put(struct block_device *bdev, fmode_t mode, int for_part)
{
struct gendisk *disk = bdev->bd_disk;
[PATCH] lockdep: simplify some aspects of bd_mutex nesting When we open (actually blkdev_get) a partition we need to also open (get) the whole device that holds the partition. The involves some limited recursion. This patch tries to simplify some aspects of this. As well as opening the whole device, we need to increment ->bd_part_count when a partition is opened (this is used by rescan_partitions to avoid a rescan if any partition is active, as that would be confusing). The main change this patch makes is to move the inc/dec of bd_part_count into blkdev_{get,put} for the whole rather than doing it in blkdev_{get,put} for the partition. More specifically, we introduce __blkdev_get and __blkdev_put which do exactly what blkdev_{get,put} did, only with an extra "for_part" argument (blkget_{get,put} then call the __ version with a '0' for the extra argument). If for_part is 1, then the blkdev is being get(put) because a partition is being opened(closed) for the first(last) time, and so bd_part_count should be updated (on success). The particular advantage of pushing this function down is that the bd_mutex lock (which is needed to update bd_part_count) is already held at the lower level. Note that this slightly changes the semantics of bd_part_count. Instead of updating it whenever a partition is opened or released, it is now only updated on the first open or last release. This is an adequate semantic as it is only ever tested for "== 0". Having introduced these functions we remove the current bd_part_count updates from do_open (which is really the body of blkdev_get) and call __blkdev_get(... 1). Similarly in blkget_put we remove the old bd_part_count updates and call __blkget_put(..., 1). This call is moved to the end of __blkdev_put to avoid nested locks of bd_mutex. Finally the mutex_lock on whole->bd_mutex in do_open can be removed. It was only really needed to protect bd_part_count, and that is now managed (and protected) within the recursive call. The observation that bd_part_count is central to the locking issues, and the modifications to create __blkdev_put are from Peter Zijlstra. Cc: Ingo Molnar <mingo@elte.hu> Acked-by: Peter Zijlstra <a.p.zijlstra@chello.nl> Signed-off-by: Neil Brown <neilb@suse.de> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-12-08 10:36:16 +00:00
struct block_device *victim = NULL;
bdev: Reduce time holding bd_mutex in sync in blkdev_close() While trying to "dd" to the block device for a USB stick, I encountered a hung task warning (blocked for > 120 seconds). I managed to come up with an easy way to reproduce this on my system (where /dev/sdb is the block device for my USB stick) with: while true; do dd if=/dev/zero of=/dev/sdb bs=4M; done With my reproduction here are the relevant bits from the hung task detector: INFO: task udevd:294 blocked for more than 122 seconds. ... udevd D 0 294 1 0x00400008 Call trace: ... mutex_lock_nested+0x40/0x50 __blkdev_get+0x7c/0x3d4 blkdev_get+0x118/0x138 blkdev_open+0x94/0xa8 do_dentry_open+0x268/0x3a0 vfs_open+0x34/0x40 path_openat+0x39c/0xdf4 do_filp_open+0x90/0x10c do_sys_open+0x150/0x3c8 ... ... Showing all locks held in the system: ... 1 lock held by dd/2798: #0: ffffff814ac1a3b8 (&bdev->bd_mutex){+.+.}, at: __blkdev_put+0x50/0x204 ... dd D 0 2798 2764 0x00400208 Call trace: ... schedule+0x8c/0xbc io_schedule+0x1c/0x40 wait_on_page_bit_common+0x238/0x338 __lock_page+0x5c/0x68 write_cache_pages+0x194/0x500 generic_writepages+0x64/0xa4 blkdev_writepages+0x24/0x30 do_writepages+0x48/0xa8 __filemap_fdatawrite_range+0xac/0xd8 filemap_write_and_wait+0x30/0x84 __blkdev_put+0x88/0x204 blkdev_put+0xc4/0xe4 blkdev_close+0x28/0x38 __fput+0xe0/0x238 ____fput+0x1c/0x28 task_work_run+0xb0/0xe4 do_notify_resume+0xfc0/0x14bc work_pending+0x8/0x14 The problem appears related to the fact that my USB disk is terribly slow and that I have a lot of RAM in my system to cache things. Specifically my writes seem to be happening at ~15 MB/s and I've got ~4 GB of RAM in my system that can be used for buffering. To write 4 GB of buffer to disk thus takes ~4000 MB / ~15 MB/s = ~267 seconds. The 267 second number is a problem because in __blkdev_put() we call sync_blockdev() while holding the bd_mutex. Any other callers who want the bd_mutex will be blocked for the whole time. The problem is made worse because I believe blkdev_put() specifically tells other tasks (namely udev) to go try to access the device at right around the same time we're going to hold the mutex for a long time. Putting some traces around this (after disabling the hung task detector), I could confirm: dd: 437.608600: __blkdev_put() right before sync_blockdev() for sdb udevd: 437.623901: blkdev_open() right before blkdev_get() for sdb dd: 661.468451: __blkdev_put() right after sync_blockdev() for sdb udevd: 663.820426: blkdev_open() right after blkdev_get() for sdb A simple fix for this is to realize that sync_blockdev() works fine if you're not holding the mutex. Also, it's not the end of the world if you sync a little early (though it can have performance impacts). Thus we can make a guess that we're going to need to do the sync and then do it without holding the mutex. We still do one last sync with the mutex but it should be much, much faster. With this, my hung task warnings for my test case are gone. Signed-off-by: Douglas Anderson <dianders@chromium.org> Reviewed-by: Guenter Roeck <groeck@chromium.org> Reviewed-by: Christoph Hellwig <hch@lst.de> Signed-off-by: Jens Axboe <axboe@kernel.dk>
2020-03-24 21:48:27 +00:00
/*
* Sync early if it looks like we're the last one. If someone else
* opens the block device between now and the decrement of bd_openers
* then we did a sync that we didn't need to, but that's not the end
* of the world and we want to avoid long (could be several minute)
* syncs while holding the mutex.
*/
if (bdev->bd_openers == 1)
sync_blockdev(bdev);
mutex_lock_nested(&bdev->bd_mutex, for_part);
[PATCH] lockdep: simplify some aspects of bd_mutex nesting When we open (actually blkdev_get) a partition we need to also open (get) the whole device that holds the partition. The involves some limited recursion. This patch tries to simplify some aspects of this. As well as opening the whole device, we need to increment ->bd_part_count when a partition is opened (this is used by rescan_partitions to avoid a rescan if any partition is active, as that would be confusing). The main change this patch makes is to move the inc/dec of bd_part_count into blkdev_{get,put} for the whole rather than doing it in blkdev_{get,put} for the partition. More specifically, we introduce __blkdev_get and __blkdev_put which do exactly what blkdev_{get,put} did, only with an extra "for_part" argument (blkget_{get,put} then call the __ version with a '0' for the extra argument). If for_part is 1, then the blkdev is being get(put) because a partition is being opened(closed) for the first(last) time, and so bd_part_count should be updated (on success). The particular advantage of pushing this function down is that the bd_mutex lock (which is needed to update bd_part_count) is already held at the lower level. Note that this slightly changes the semantics of bd_part_count. Instead of updating it whenever a partition is opened or released, it is now only updated on the first open or last release. This is an adequate semantic as it is only ever tested for "== 0". Having introduced these functions we remove the current bd_part_count updates from do_open (which is really the body of blkdev_get) and call __blkdev_get(... 1). Similarly in blkget_put we remove the old bd_part_count updates and call __blkget_put(..., 1). This call is moved to the end of __blkdev_put to avoid nested locks of bd_mutex. Finally the mutex_lock on whole->bd_mutex in do_open can be removed. It was only really needed to protect bd_part_count, and that is now managed (and protected) within the recursive call. The observation that bd_part_count is central to the locking issues, and the modifications to create __blkdev_put are from Peter Zijlstra. Cc: Ingo Molnar <mingo@elte.hu> Acked-by: Peter Zijlstra <a.p.zijlstra@chello.nl> Signed-off-by: Neil Brown <neilb@suse.de> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-12-08 10:36:16 +00:00
if (for_part)
bdev->bd_part_count--;
if (!--bdev->bd_openers) {
WARN_ON_ONCE(bdev->bd_holders);
sync_blockdev(bdev);
kill_bdev(bdev);
block: detach bdev inode from its wb in __blkdev_put() Since 52ebea749aae ("writeback: make backing_dev_info host cgroup-specific bdi_writebacks") inode, at some point in its lifetime, gets attached to a wb (struct bdi_writeback). Detaching happens on evict, in inode_detach_wb() called from __destroy_inode(), and involves updating wb. However, detaching an internal bdev inode from its wb in __destroy_inode() is too late. Its bdi and by extension root wb are embedded into struct request_queue, which has different lifetime rules and can be freed long before the final bdput() is called (can be from __fput() of a corresponding /dev inode, through dput() - evict() - bd_forget(). bdevs hold onto the underlying disk/queue pair only while opened; as soon as bdev is closed all bets are off. In fact, disk/queue can be gone before __blkdev_put() even returns: 1499 static void __blkdev_put(struct block_device *bdev, fmode_t mode, int for_part) 1500 { ... 1518 if (bdev->bd_contains == bdev) { 1519 if (disk->fops->release) 1520 disk->fops->release(disk, mode); [ Driver puts its references to disk/queue ] 1521 } 1522 if (!bdev->bd_openers) { 1523 struct module *owner = disk->fops->owner; 1524 1525 disk_put_part(bdev->bd_part); 1526 bdev->bd_part = NULL; 1527 bdev->bd_disk = NULL; 1528 if (bdev != bdev->bd_contains) 1529 victim = bdev->bd_contains; 1530 bdev->bd_contains = NULL; 1531 1532 put_disk(disk); [ We put ours, the queue is gone The last bdput() would result in a write to invalid memory ] 1533 module_put(owner); ... 1539 } Since bdev inodes are special anyway, detach them in __blkdev_put() after clearing inode's dirty bits, turning the problematic inode_detach_wb() in __destroy_inode() into a noop. add_disk() grabs its disk->queue since 523e1d399ce0 ("block: make gendisk hold a reference to its queue"), so the old ->release comment is removed in favor of the new inode_detach_wb() comment. Cc: stable@vger.kernel.org # 4.2+, needs backporting Signed-off-by: Ilya Dryomov <idryomov@gmail.com> Acked-by: Tejun Heo <tj@kernel.org> Tested-by: Raghavendra K T <raghavendra.kt@linux.vnet.ibm.com> Signed-off-by: Jens Axboe <axboe@fb.com>
2015-11-20 21:22:34 +00:00
bdev_write_inode(bdev);
if (!bdev_is_partition(bdev) && disk->fops->release)
disk->fops->release(disk, mode);
disk_put_part(bdev->bd_part);
bdev->bd_part = NULL;
bdev->bd_disk = NULL;
if (bdev_is_partition(bdev))
[PATCH] lockdep: simplify some aspects of bd_mutex nesting When we open (actually blkdev_get) a partition we need to also open (get) the whole device that holds the partition. The involves some limited recursion. This patch tries to simplify some aspects of this. As well as opening the whole device, we need to increment ->bd_part_count when a partition is opened (this is used by rescan_partitions to avoid a rescan if any partition is active, as that would be confusing). The main change this patch makes is to move the inc/dec of bd_part_count into blkdev_{get,put} for the whole rather than doing it in blkdev_{get,put} for the partition. More specifically, we introduce __blkdev_get and __blkdev_put which do exactly what blkdev_{get,put} did, only with an extra "for_part" argument (blkget_{get,put} then call the __ version with a '0' for the extra argument). If for_part is 1, then the blkdev is being get(put) because a partition is being opened(closed) for the first(last) time, and so bd_part_count should be updated (on success). The particular advantage of pushing this function down is that the bd_mutex lock (which is needed to update bd_part_count) is already held at the lower level. Note that this slightly changes the semantics of bd_part_count. Instead of updating it whenever a partition is opened or released, it is now only updated on the first open or last release. This is an adequate semantic as it is only ever tested for "== 0". Having introduced these functions we remove the current bd_part_count updates from do_open (which is really the body of blkdev_get) and call __blkdev_get(... 1). Similarly in blkget_put we remove the old bd_part_count updates and call __blkget_put(..., 1). This call is moved to the end of __blkdev_put to avoid nested locks of bd_mutex. Finally the mutex_lock on whole->bd_mutex in do_open can be removed. It was only really needed to protect bd_part_count, and that is now managed (and protected) within the recursive call. The observation that bd_part_count is central to the locking issues, and the modifications to create __blkdev_put are from Peter Zijlstra. Cc: Ingo Molnar <mingo@elte.hu> Acked-by: Peter Zijlstra <a.p.zijlstra@chello.nl> Signed-off-by: Neil Brown <neilb@suse.de> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-12-08 10:36:16 +00:00
victim = bdev->bd_contains;
bdev->bd_contains = NULL;
} else {
if (!bdev_is_partition(bdev) && disk->fops->release)
disk->fops->release(disk, mode);
}
mutex_unlock(&bdev->bd_mutex);
if (victim) {
__blkdev_put(victim, mode, 1);
bdput(victim);
}
}
void blkdev_put(struct block_device *bdev, fmode_t mode)
[PATCH] lockdep: simplify some aspects of bd_mutex nesting When we open (actually blkdev_get) a partition we need to also open (get) the whole device that holds the partition. The involves some limited recursion. This patch tries to simplify some aspects of this. As well as opening the whole device, we need to increment ->bd_part_count when a partition is opened (this is used by rescan_partitions to avoid a rescan if any partition is active, as that would be confusing). The main change this patch makes is to move the inc/dec of bd_part_count into blkdev_{get,put} for the whole rather than doing it in blkdev_{get,put} for the partition. More specifically, we introduce __blkdev_get and __blkdev_put which do exactly what blkdev_{get,put} did, only with an extra "for_part" argument (blkget_{get,put} then call the __ version with a '0' for the extra argument). If for_part is 1, then the blkdev is being get(put) because a partition is being opened(closed) for the first(last) time, and so bd_part_count should be updated (on success). The particular advantage of pushing this function down is that the bd_mutex lock (which is needed to update bd_part_count) is already held at the lower level. Note that this slightly changes the semantics of bd_part_count. Instead of updating it whenever a partition is opened or released, it is now only updated on the first open or last release. This is an adequate semantic as it is only ever tested for "== 0". Having introduced these functions we remove the current bd_part_count updates from do_open (which is really the body of blkdev_get) and call __blkdev_get(... 1). Similarly in blkget_put we remove the old bd_part_count updates and call __blkget_put(..., 1). This call is moved to the end of __blkdev_put to avoid nested locks of bd_mutex. Finally the mutex_lock on whole->bd_mutex in do_open can be removed. It was only really needed to protect bd_part_count, and that is now managed (and protected) within the recursive call. The observation that bd_part_count is central to the locking issues, and the modifications to create __blkdev_put are from Peter Zijlstra. Cc: Ingo Molnar <mingo@elte.hu> Acked-by: Peter Zijlstra <a.p.zijlstra@chello.nl> Signed-off-by: Neil Brown <neilb@suse.de> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-12-08 10:36:16 +00:00
{
struct gendisk *disk = bdev->bd_disk;
2011-07-01 14:17:47 +00:00
mutex_lock(&bdev->bd_mutex);
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 <tj@kernel.org> Acked-by: Neil Brown <neilb@suse.de> Acked-by: Ryusuke Konishi <konishi.ryusuke@lab.ntt.co.jp> Acked-by: Mike Snitzer <snitzer@redhat.com> Acked-by: Philipp Reisner <philipp.reisner@linbit.com> Cc: Peter Osterlund <petero2@telia.com> Cc: Martin Schwidefsky <schwidefsky@de.ibm.com> Cc: Heiko Carstens <heiko.carstens@de.ibm.com> Cc: Jan Kara <jack@suse.cz> Cc: Andrew Morton <akpm@linux-foundation.org> Cc: Andreas Dilger <adilger.kernel@dilger.ca> Cc: "Theodore Ts'o" <tytso@mit.edu> Cc: Mark Fasheh <mfasheh@suse.com> Cc: Joel Becker <joel.becker@oracle.com> Cc: Alex Elder <aelder@sgi.com> Cc: Christoph Hellwig <hch@infradead.org> Cc: dm-devel@redhat.com Cc: drbd-dev@lists.linbit.com Cc: Leo Chen <leochen@broadcom.com> Cc: Scott Branden <sbranden@broadcom.com> Cc: Chris Mason <chris.mason@oracle.com> Cc: Steven Whitehouse <swhiteho@redhat.com> Cc: Dave Kleikamp <shaggy@linux.vnet.ibm.com> Cc: Joern Engel <joern@logfs.org> Cc: reiserfs-devel@vger.kernel.org Cc: Alexander Viro <viro@zeniv.linux.org.uk>
2010-11-13 10:55:17 +00:00
if (mode & FMODE_EXCL) {
bool bdev_free;
/*
* Release a claim on the device. The holder fields
* are protected with bdev_lock. bd_mutex is to
* synchronize disk_holder unlinking.
*/
spin_lock(&bdev_lock);
WARN_ON_ONCE(--bdev->bd_holders < 0);
WARN_ON_ONCE(--bdev->bd_contains->bd_holders < 0);
/* bd_contains might point to self, check in a separate step */
if ((bdev_free = !bdev->bd_holders))
bdev->bd_holder = NULL;
if (!bdev->bd_contains->bd_holders)
bdev->bd_contains->bd_holder = NULL;
spin_unlock(&bdev_lock);
implement in-kernel gendisk events handling Currently, media presence polling for removeable block devices is done from userland. There are several issues with this. * Polling is done by periodically opening the device. For SCSI devices, the command sequence generated by such action involves a few different commands including TEST_UNIT_READY. This behavior, while perfectly legal, is different from Windows which only issues single command, GET_EVENT_STATUS_NOTIFICATION. Unfortunately, some ATAPI devices lock up after being periodically queried such command sequences. * There is no reliable and unintrusive way for a userland program to tell whether the target device is safe for media presence polling. For example, polling for media presence during an on-going burning session can make it fail. The polling program can avoid this by opening the device with O_EXCL but then it risks making a valid exclusive user of the device fail w/ -EBUSY. * Userland polling is unnecessarily heavy and in-kernel implementation is lighter and better coordinated (workqueue, timer slack). This patch implements framework for in-kernel disk event handling, which includes media presence polling. * bdops->check_events() is added, which supercedes ->media_changed(). It should check whether there's any pending event and return if so. Currently, two events are defined - DISK_EVENT_MEDIA_CHANGE and DISK_EVENT_EJECT_REQUEST. ->check_events() is guaranteed not to be called parallelly. * gendisk->events and ->async_events are added. These should be initialized by block driver before passing the device to add_disk(). The former contains the mask of all supported events and the latter the mask of all events which the device can report without polling. /sys/block/*/events[_async] export these to userland. * Kernel parameter block.events_dfl_poll_msecs controls the system polling interval (default is 0 which means disable) and /sys/block/*/events_poll_msecs control polling intervals for individual devices (default is -1 meaning use system setting). Note that if a device can report all supported events asynchronously and its polling interval isn't explicitly set, the device won't be polled regardless of the system polling interval. * If a device is opened exclusively with write access, event checking is automatically disabled until all write exclusive accesses are released. * There are event 'clearing' events. For example, both of currently defined events are cleared after the device has been successfully opened. This information is passed to ->check_events() callback using @clearing argument as a hint. * Event checking is always performed from system_nrt_wq and timer slack is set to 25% for polling. * Nothing changes for drivers which implement ->media_changed() but not ->check_events(). Going forward, all drivers will be converted to ->check_events() and ->media_change() will be dropped. Signed-off-by: Tejun Heo <tj@kernel.org> Cc: Kay Sievers <kay.sievers@vrfy.org> Cc: Jan Kara <jack@suse.cz> Signed-off-by: Jens Axboe <jaxboe@fusionio.com>
2010-12-08 19:57:37 +00:00
/*
* If this was the last claim, remove holder link and
* unblock evpoll if it was a write holder.
*/
2011-07-01 14:17:47 +00:00
if (bdev_free && bdev->bd_write_holder) {
disk_unblock_events(disk);
2011-07-01 14:17:47 +00:00
bdev->bd_write_holder = false;
implement in-kernel gendisk events handling Currently, media presence polling for removeable block devices is done from userland. There are several issues with this. * Polling is done by periodically opening the device. For SCSI devices, the command sequence generated by such action involves a few different commands including TEST_UNIT_READY. This behavior, while perfectly legal, is different from Windows which only issues single command, GET_EVENT_STATUS_NOTIFICATION. Unfortunately, some ATAPI devices lock up after being periodically queried such command sequences. * There is no reliable and unintrusive way for a userland program to tell whether the target device is safe for media presence polling. For example, polling for media presence during an on-going burning session can make it fail. The polling program can avoid this by opening the device with O_EXCL but then it risks making a valid exclusive user of the device fail w/ -EBUSY. * Userland polling is unnecessarily heavy and in-kernel implementation is lighter and better coordinated (workqueue, timer slack). This patch implements framework for in-kernel disk event handling, which includes media presence polling. * bdops->check_events() is added, which supercedes ->media_changed(). It should check whether there's any pending event and return if so. Currently, two events are defined - DISK_EVENT_MEDIA_CHANGE and DISK_EVENT_EJECT_REQUEST. ->check_events() is guaranteed not to be called parallelly. * gendisk->events and ->async_events are added. These should be initialized by block driver before passing the device to add_disk(). The former contains the mask of all supported events and the latter the mask of all events which the device can report without polling. /sys/block/*/events[_async] export these to userland. * Kernel parameter block.events_dfl_poll_msecs controls the system polling interval (default is 0 which means disable) and /sys/block/*/events_poll_msecs control polling intervals for individual devices (default is -1 meaning use system setting). Note that if a device can report all supported events asynchronously and its polling interval isn't explicitly set, the device won't be polled regardless of the system polling interval. * If a device is opened exclusively with write access, event checking is automatically disabled until all write exclusive accesses are released. * There are event 'clearing' events. For example, both of currently defined events are cleared after the device has been successfully opened. This information is passed to ->check_events() callback using @clearing argument as a hint. * Event checking is always performed from system_nrt_wq and timer slack is set to 25% for polling. * Nothing changes for drivers which implement ->media_changed() but not ->check_events(). Going forward, all drivers will be converted to ->check_events() and ->media_change() will be dropped. Signed-off-by: Tejun Heo <tj@kernel.org> Cc: Kay Sievers <kay.sievers@vrfy.org> Cc: Jan Kara <jack@suse.cz> Signed-off-by: Jens Axboe <jaxboe@fusionio.com>
2010-12-08 19:57:37 +00:00
}
}
implement in-kernel gendisk events handling Currently, media presence polling for removeable block devices is done from userland. There are several issues with this. * Polling is done by periodically opening the device. For SCSI devices, the command sequence generated by such action involves a few different commands including TEST_UNIT_READY. This behavior, while perfectly legal, is different from Windows which only issues single command, GET_EVENT_STATUS_NOTIFICATION. Unfortunately, some ATAPI devices lock up after being periodically queried such command sequences. * There is no reliable and unintrusive way for a userland program to tell whether the target device is safe for media presence polling. For example, polling for media presence during an on-going burning session can make it fail. The polling program can avoid this by opening the device with O_EXCL but then it risks making a valid exclusive user of the device fail w/ -EBUSY. * Userland polling is unnecessarily heavy and in-kernel implementation is lighter and better coordinated (workqueue, timer slack). This patch implements framework for in-kernel disk event handling, which includes media presence polling. * bdops->check_events() is added, which supercedes ->media_changed(). It should check whether there's any pending event and return if so. Currently, two events are defined - DISK_EVENT_MEDIA_CHANGE and DISK_EVENT_EJECT_REQUEST. ->check_events() is guaranteed not to be called parallelly. * gendisk->events and ->async_events are added. These should be initialized by block driver before passing the device to add_disk(). The former contains the mask of all supported events and the latter the mask of all events which the device can report without polling. /sys/block/*/events[_async] export these to userland. * Kernel parameter block.events_dfl_poll_msecs controls the system polling interval (default is 0 which means disable) and /sys/block/*/events_poll_msecs control polling intervals for individual devices (default is -1 meaning use system setting). Note that if a device can report all supported events asynchronously and its polling interval isn't explicitly set, the device won't be polled regardless of the system polling interval. * If a device is opened exclusively with write access, event checking is automatically disabled until all write exclusive accesses are released. * There are event 'clearing' events. For example, both of currently defined events are cleared after the device has been successfully opened. This information is passed to ->check_events() callback using @clearing argument as a hint. * Event checking is always performed from system_nrt_wq and timer slack is set to 25% for polling. * Nothing changes for drivers which implement ->media_changed() but not ->check_events(). Going forward, all drivers will be converted to ->check_events() and ->media_change() will be dropped. Signed-off-by: Tejun Heo <tj@kernel.org> Cc: Kay Sievers <kay.sievers@vrfy.org> Cc: Jan Kara <jack@suse.cz> Signed-off-by: Jens Axboe <jaxboe@fusionio.com>
2010-12-08 19:57:37 +00:00
2011-07-01 14:17:47 +00:00
/*
* Trigger event checking and tell drivers to flush MEDIA_CHANGE
* event. This is to ensure detection of media removal commanded
* from userland - e.g. eject(1).
*/
disk_flush_events(disk, DISK_EVENT_MEDIA_CHANGE);
2011-07-01 14:17:47 +00:00
mutex_unlock(&bdev->bd_mutex);
__blkdev_put(bdev, mode, 0);
bdput(bdev);
put_disk_and_module(disk);
[PATCH] lockdep: simplify some aspects of bd_mutex nesting When we open (actually blkdev_get) a partition we need to also open (get) the whole device that holds the partition. The involves some limited recursion. This patch tries to simplify some aspects of this. As well as opening the whole device, we need to increment ->bd_part_count when a partition is opened (this is used by rescan_partitions to avoid a rescan if any partition is active, as that would be confusing). The main change this patch makes is to move the inc/dec of bd_part_count into blkdev_{get,put} for the whole rather than doing it in blkdev_{get,put} for the partition. More specifically, we introduce __blkdev_get and __blkdev_put which do exactly what blkdev_{get,put} did, only with an extra "for_part" argument (blkget_{get,put} then call the __ version with a '0' for the extra argument). If for_part is 1, then the blkdev is being get(put) because a partition is being opened(closed) for the first(last) time, and so bd_part_count should be updated (on success). The particular advantage of pushing this function down is that the bd_mutex lock (which is needed to update bd_part_count) is already held at the lower level. Note that this slightly changes the semantics of bd_part_count. Instead of updating it whenever a partition is opened or released, it is now only updated on the first open or last release. This is an adequate semantic as it is only ever tested for "== 0". Having introduced these functions we remove the current bd_part_count updates from do_open (which is really the body of blkdev_get) and call __blkdev_get(... 1). Similarly in blkget_put we remove the old bd_part_count updates and call __blkget_put(..., 1). This call is moved to the end of __blkdev_put to avoid nested locks of bd_mutex. Finally the mutex_lock on whole->bd_mutex in do_open can be removed. It was only really needed to protect bd_part_count, and that is now managed (and protected) within the recursive call. The observation that bd_part_count is central to the locking issues, and the modifications to create __blkdev_put are from Peter Zijlstra. Cc: Ingo Molnar <mingo@elte.hu> Acked-by: Peter Zijlstra <a.p.zijlstra@chello.nl> Signed-off-by: Neil Brown <neilb@suse.de> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-12-08 10:36:16 +00:00
}
EXPORT_SYMBOL(blkdev_put);
static int blkdev_close(struct inode * inode, struct file * filp)
{
struct block_device *bdev = I_BDEV(bdev_file_inode(filp));
blkdev_put(bdev, filp->f_mode);
return 0;
}
static long block_ioctl(struct file *file, unsigned cmd, unsigned long arg)
{
struct block_device *bdev = I_BDEV(bdev_file_inode(file));
fmode_t mode = file->f_mode;
/*
* O_NDELAY can be altered using fcntl(.., F_SETFL, ..), so we have
* to updated it before every ioctl.
*/
if (file->f_flags & O_NDELAY)
mode |= FMODE_NDELAY;
else
mode &= ~FMODE_NDELAY;
return blkdev_ioctl(bdev, mode, cmd, arg);
}
/*
* Write data to the block device. Only intended for the block device itself
* and the raw driver which basically is a fake block device.
*
* Does not take i_mutex for the write and thus is not for general purpose
* use.
*/
ssize_t blkdev_write_iter(struct kiocb *iocb, struct iov_iter *from)
{
struct file *file = iocb->ki_filp;
struct inode *bd_inode = bdev_file_inode(file);
loff_t size = i_size_read(bd_inode);
struct blk_plug plug;
ssize_t ret;
if (bdev_read_only(I_BDEV(bd_inode)))
return -EPERM;
if (IS_SWAPFILE(bd_inode) && !is_hibernate_resume_dev(bd_inode->i_rdev))
return -ETXTBSY;
if (!iov_iter_count(from))
return 0;
if (iocb->ki_pos >= size)
return -ENOSPC;
if ((iocb->ki_flags & (IOCB_NOWAIT | IOCB_DIRECT)) == IOCB_NOWAIT)
return -EOPNOTSUPP;
iov_iter_truncate(from, size - iocb->ki_pos);
blk_start_plug(&plug);
ret = __generic_file_write_iter(iocb, from);
if (ret > 0)
ret = generic_write_sync(iocb, ret);
blk_finish_plug(&plug);
return ret;
}
EXPORT_SYMBOL_GPL(blkdev_write_iter);
ssize_t blkdev_read_iter(struct kiocb *iocb, struct iov_iter *to)
{
struct file *file = iocb->ki_filp;
struct inode *bd_inode = bdev_file_inode(file);
loff_t size = i_size_read(bd_inode);
loff_t pos = iocb->ki_pos;
if (pos >= size)
return 0;
size -= pos;
iov_iter_truncate(to, size);
return generic_file_read_iter(iocb, to);
}
EXPORT_SYMBOL_GPL(blkdev_read_iter);
/*
* Try to release a page associated with block device when the system
* is under memory pressure.
*/
static int blkdev_releasepage(struct page *page, gfp_t wait)
{
struct super_block *super = BDEV_I(page->mapping->host)->bdev.bd_super;
if (super && super->s_op->bdev_try_to_free_page)
return super->s_op->bdev_try_to_free_page(super, page, wait);
return try_to_free_buffers(page);
}
static int blkdev_writepages(struct address_space *mapping,
struct writeback_control *wbc)
{
return generic_writepages(mapping, wbc);
}
static const struct address_space_operations def_blk_aops = {
.readpage = blkdev_readpage,
.readahead = blkdev_readahead,
.writepage = blkdev_writepage,
.write_begin = blkdev_write_begin,
.write_end = blkdev_write_end,
.writepages = blkdev_writepages,
.releasepage = blkdev_releasepage,
.direct_IO = blkdev_direct_IO,
.migratepage = buffer_migrate_page_norefs,
mm: vmscan: take page buffers dirty and locked state into account Page reclaim keeps track of dirty and under writeback pages and uses it to determine if wait_iff_congested() should stall or if kswapd should begin writing back pages. This fails to account for buffer pages that can be under writeback but not PageWriteback which is the case for filesystems like ext3 ordered mode. Furthermore, PageDirty buffer pages can have all the buffers clean and writepage does no IO so it should not be accounted as congested. This patch adds an address_space operation that filesystems may optionally use to check if a page is really dirty or really under writeback. An implementation is provided for for buffer_heads is added and used for block operations and ext3 in ordered mode. By default the page flags are obeyed. Credit goes to Jan Kara for identifying that the page flags alone are not sufficient for ext3 and sanity checking a number of ideas on how the problem could be addressed. Signed-off-by: Mel Gorman <mgorman@suse.de> Cc: Johannes Weiner <hannes@cmpxchg.org> Cc: Michal Hocko <mhocko@suse.cz> Cc: Rik van Riel <riel@redhat.com> Cc: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com> Cc: Jiri Slaby <jslaby@suse.cz> Cc: Valdis Kletnieks <Valdis.Kletnieks@vt.edu> Cc: Zlatko Calusic <zcalusic@bitsync.net> Cc: dormando <dormando@rydia.net> Cc: Trond Myklebust <trond.myklebust@fys.uio.no> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2013-07-03 22:02:05 +00:00
.is_dirty_writeback = buffer_check_dirty_writeback,
};
#define BLKDEV_FALLOC_FL_SUPPORTED \
(FALLOC_FL_KEEP_SIZE | FALLOC_FL_PUNCH_HOLE | \
FALLOC_FL_ZERO_RANGE | FALLOC_FL_NO_HIDE_STALE)
static long blkdev_fallocate(struct file *file, int mode, loff_t start,
loff_t len)
{
struct block_device *bdev = I_BDEV(bdev_file_inode(file));
loff_t end = start + len - 1;
loff_t isize;
int error;
/* Fail if we don't recognize the flags. */
if (mode & ~BLKDEV_FALLOC_FL_SUPPORTED)
return -EOPNOTSUPP;
/* Don't go off the end of the device. */
isize = i_size_read(bdev->bd_inode);
if (start >= isize)
return -EINVAL;
if (end >= isize) {
if (mode & FALLOC_FL_KEEP_SIZE) {
len = isize - start;
end = start + len - 1;
} else
return -EINVAL;
}
/*
* Don't allow IO that isn't aligned to logical block size.
*/
if ((start | len) & (bdev_logical_block_size(bdev) - 1))
return -EINVAL;
/* Invalidate the page cache, including dirty pages. */
error = truncate_bdev_range(bdev, file->f_mode, start, end);
if (error)
return error;
switch (mode) {
case FALLOC_FL_ZERO_RANGE:
case FALLOC_FL_ZERO_RANGE | FALLOC_FL_KEEP_SIZE:
error = blkdev_issue_zeroout(bdev, start >> 9, len >> 9,
GFP_KERNEL, BLKDEV_ZERO_NOUNMAP);
break;
case FALLOC_FL_PUNCH_HOLE | FALLOC_FL_KEEP_SIZE:
error = blkdev_issue_zeroout(bdev, start >> 9, len >> 9,
GFP_KERNEL, BLKDEV_ZERO_NOFALLBACK);
break;
case FALLOC_FL_PUNCH_HOLE | FALLOC_FL_KEEP_SIZE | FALLOC_FL_NO_HIDE_STALE:
error = blkdev_issue_discard(bdev, start >> 9, len >> 9,
GFP_KERNEL, 0);
break;
default:
return -EOPNOTSUPP;
}
if (error)
return error;
/*
* Invalidate again; if someone wandered in and dirtied a page,
* the caller will be given -EBUSY. The third argument is
* inclusive, so the rounding here is safe.
*/
return invalidate_inode_pages2_range(bdev->bd_inode->i_mapping,
start >> PAGE_SHIFT,
end >> PAGE_SHIFT);
}
const struct file_operations def_blk_fops = {
.open = blkdev_open,
.release = blkdev_close,
.llseek = block_llseek,
.read_iter = blkdev_read_iter,
.write_iter = blkdev_write_iter,
.iopoll = blkdev_iopoll,
.mmap = generic_file_mmap,
.fsync = blkdev_fsync,
.unlocked_ioctl = block_ioctl,
#ifdef CONFIG_COMPAT
.compat_ioctl = compat_blkdev_ioctl,
#endif
.splice_read = generic_file_splice_read,
.splice_write = iter_file_splice_write,
.fallocate = blkdev_fallocate,
};
/**
* lookup_bdev - lookup a struct block_device by name
* @pathname: special file representing the block device
*
* Get a reference to the blockdevice at @pathname in the current
* namespace if possible and return it. Return ERR_PTR(error)
* otherwise.
*/
struct block_device *lookup_bdev(const char *pathname)
{
struct block_device *bdev;
struct inode *inode;
struct path path;
int error;
if (!pathname || !*pathname)
return ERR_PTR(-EINVAL);
error = kern_path(pathname, LOOKUP_FOLLOW, &path);
if (error)
return ERR_PTR(error);
inode = d_backing_inode(path.dentry);
error = -ENOTBLK;
if (!S_ISBLK(inode->i_mode))
goto fail;
error = -EACCES;
if (!may_open_dev(&path))
goto fail;
error = -ENOMEM;
bdev = bd_acquire(inode);
if (!bdev)
goto fail;
out:
path_put(&path);
return bdev;
fail:
bdev = ERR_PTR(error);
goto out;
}
EXPORT_SYMBOL(lookup_bdev);
Fix over-zealous flush_disk when changing device size. There are two cases when we call flush_disk. In one, the device has disappeared (check_disk_change) so any data will hold becomes irrelevant. In the oter, the device has changed size (check_disk_size_change) so data we hold may be irrelevant. In both cases it makes sense to discard any 'clean' buffers, so they will be read back from the device if needed. In the former case it makes sense to discard 'dirty' buffers as there will never be anywhere safe to write the data. In the second case it *does*not* make sense to discard dirty buffers as that will lead to file system corruption when you simply enlarge the containing devices. flush_disk calls __invalidate_devices. __invalidate_device calls both invalidate_inodes and invalidate_bdev. invalidate_inodes *does* discard I_DIRTY inodes and this does lead to fs corruption. invalidate_bev *does*not* discard dirty pages, but I don't really care about that at present. So this patch adds a flag to __invalidate_device (calling it __invalidate_device2) to indicate whether dirty buffers should be killed, and this is passed to invalidate_inodes which can choose to skip dirty inodes. flusk_disk then passes true from check_disk_change and false from check_disk_size_change. dm avoids tripping over this problem by calling i_size_write directly rathher than using check_disk_size_change. md does use check_disk_size_change and so is affected. This regression was introduced by commit 608aeef17a which causes check_disk_size_change to call flush_disk, so it is suitable for any kernel since 2.6.27. Cc: stable@kernel.org Acked-by: Jeff Moyer <jmoyer@redhat.com> Cc: Andrew Patterson <andrew.patterson@hp.com> Cc: Jens Axboe <axboe@kernel.dk> Signed-off-by: NeilBrown <neilb@suse.de>
2011-02-24 06:25:47 +00:00
int __invalidate_device(struct block_device *bdev, bool kill_dirty)
{
struct super_block *sb = get_super(bdev);
int res = 0;
if (sb) {
/*
* no need to lock the super, get_super holds the
* read mutex so the filesystem cannot go away
* under us (->put_super runs with the write lock
* hold).
*/
shrink_dcache_sb(sb);
Fix over-zealous flush_disk when changing device size. There are two cases when we call flush_disk. In one, the device has disappeared (check_disk_change) so any data will hold becomes irrelevant. In the oter, the device has changed size (check_disk_size_change) so data we hold may be irrelevant. In both cases it makes sense to discard any 'clean' buffers, so they will be read back from the device if needed. In the former case it makes sense to discard 'dirty' buffers as there will never be anywhere safe to write the data. In the second case it *does*not* make sense to discard dirty buffers as that will lead to file system corruption when you simply enlarge the containing devices. flush_disk calls __invalidate_devices. __invalidate_device calls both invalidate_inodes and invalidate_bdev. invalidate_inodes *does* discard I_DIRTY inodes and this does lead to fs corruption. invalidate_bev *does*not* discard dirty pages, but I don't really care about that at present. So this patch adds a flag to __invalidate_device (calling it __invalidate_device2) to indicate whether dirty buffers should be killed, and this is passed to invalidate_inodes which can choose to skip dirty inodes. flusk_disk then passes true from check_disk_change and false from check_disk_size_change. dm avoids tripping over this problem by calling i_size_write directly rathher than using check_disk_size_change. md does use check_disk_size_change and so is affected. This regression was introduced by commit 608aeef17a which causes check_disk_size_change to call flush_disk, so it is suitable for any kernel since 2.6.27. Cc: stable@kernel.org Acked-by: Jeff Moyer <jmoyer@redhat.com> Cc: Andrew Patterson <andrew.patterson@hp.com> Cc: Jens Axboe <axboe@kernel.dk> Signed-off-by: NeilBrown <neilb@suse.de>
2011-02-24 06:25:47 +00:00
res = invalidate_inodes(sb, kill_dirty);
drop_super(sb);
}
invalidate_bdev(bdev);
return res;
}
EXPORT_SYMBOL(__invalidate_device);
void iterate_bdevs(void (*func)(struct block_device *, void *), void *arg)
{
struct inode *inode, *old_inode = NULL;
spin_lock(&blockdev_superblock->s_inode_list_lock);
list_for_each_entry(inode, &blockdev_superblock->s_inodes, i_sb_list) {
struct address_space *mapping = inode->i_mapping;
block: protect iterate_bdevs() against concurrent close If a block device is closed while iterate_bdevs() is handling it, the following NULL pointer dereference occurs because bdev->b_disk is NULL in bdev_get_queue(), which is called from blk_get_backing_dev_info() (in turn called by the mapping_cap_writeback_dirty() call in __filemap_fdatawrite_range()): BUG: unable to handle kernel NULL pointer dereference at 0000000000000508 IP: [<ffffffff81314790>] blk_get_backing_dev_info+0x10/0x20 PGD 9e62067 PUD 9ee8067 PMD 0 Oops: 0000 [#1] PREEMPT SMP DEBUG_PAGEALLOC Modules linked in: CPU: 1 PID: 2422 Comm: sync Not tainted 4.5.0-rc7+ #400 Hardware name: QEMU Standard PC (i440FX + PIIX, 1996) task: ffff880009f4d700 ti: ffff880009f5c000 task.ti: ffff880009f5c000 RIP: 0010:[<ffffffff81314790>] [<ffffffff81314790>] blk_get_backing_dev_info+0x10/0x20 RSP: 0018:ffff880009f5fe68 EFLAGS: 00010246 RAX: 0000000000000000 RBX: ffff88000ec17a38 RCX: ffffffff81a4e940 RDX: 7fffffffffffffff RSI: 0000000000000000 RDI: ffff88000ec176c0 RBP: ffff880009f5fe68 R08: 0000000000000000 R09: 0000000000000000 R10: 0000000000000001 R11: 0000000000000000 R12: ffff88000ec17860 R13: ffffffff811b25c0 R14: ffff88000ec178e0 R15: ffff88000ec17a38 FS: 00007faee505d700(0000) GS:ffff88000fb00000(0000) knlGS:0000000000000000 CS: 0010 DS: 0000 ES: 0000 CR0: 000000008005003b CR2: 0000000000000508 CR3: 0000000009e8a000 CR4: 00000000000006e0 Stack: ffff880009f5feb8 ffffffff8112e7f5 0000000000000000 7fffffffffffffff 0000000000000000 0000000000000000 7fffffffffffffff 0000000000000001 ffff88000ec178e0 ffff88000ec17860 ffff880009f5fec8 ffffffff8112e81f Call Trace: [<ffffffff8112e7f5>] __filemap_fdatawrite_range+0x85/0x90 [<ffffffff8112e81f>] filemap_fdatawrite+0x1f/0x30 [<ffffffff811b25d6>] fdatawrite_one_bdev+0x16/0x20 [<ffffffff811bc402>] iterate_bdevs+0xf2/0x130 [<ffffffff811b2763>] sys_sync+0x63/0x90 [<ffffffff815d4272>] entry_SYSCALL_64_fastpath+0x12/0x76 Code: 0f 1f 44 00 00 48 8b 87 f0 00 00 00 55 48 89 e5 <48> 8b 80 08 05 00 00 5d RIP [<ffffffff81314790>] blk_get_backing_dev_info+0x10/0x20 RSP <ffff880009f5fe68> CR2: 0000000000000508 ---[ end trace 2487336ceb3de62d ]--- The crash is easily reproducible by running the following command, if an msleep(100) is inserted before the call to func() in iterate_devs(): while :; do head -c1 /dev/nullb0; done > /dev/null & while :; do sync; done Fix it by holding the bd_mutex across the func() call and only calling func() if the bdev is opened. Cc: stable@vger.kernel.org Fixes: 5c0d6b60a0ba ("vfs: Create function for iterating over block devices") Reported-and-tested-by: Wei Fang <fangwei1@huawei.com> Signed-off-by: Rabin Vincent <rabinv@axis.com> Signed-off-by: Jan Kara <jack@suse.cz> Reviewed-by: Christoph Hellwig <hch@lst.de> Signed-off-by: Jens Axboe <axboe@fb.com>
2016-12-01 08:18:28 +00:00
struct block_device *bdev;
spin_lock(&inode->i_lock);
if (inode->i_state & (I_FREEING|I_WILL_FREE|I_NEW) ||
mapping->nrpages == 0) {
spin_unlock(&inode->i_lock);
continue;
}
__iget(inode);
spin_unlock(&inode->i_lock);
spin_unlock(&blockdev_superblock->s_inode_list_lock);
/*
* We hold a reference to 'inode' so it couldn't have been
* removed from s_inodes list while we dropped the
* s_inode_list_lock We cannot iput the inode now as we can
* be holding the last reference and we cannot iput it under
* s_inode_list_lock. So we keep the reference and iput it
* later.
*/
iput(old_inode);
old_inode = inode;
block: protect iterate_bdevs() against concurrent close If a block device is closed while iterate_bdevs() is handling it, the following NULL pointer dereference occurs because bdev->b_disk is NULL in bdev_get_queue(), which is called from blk_get_backing_dev_info() (in turn called by the mapping_cap_writeback_dirty() call in __filemap_fdatawrite_range()): BUG: unable to handle kernel NULL pointer dereference at 0000000000000508 IP: [<ffffffff81314790>] blk_get_backing_dev_info+0x10/0x20 PGD 9e62067 PUD 9ee8067 PMD 0 Oops: 0000 [#1] PREEMPT SMP DEBUG_PAGEALLOC Modules linked in: CPU: 1 PID: 2422 Comm: sync Not tainted 4.5.0-rc7+ #400 Hardware name: QEMU Standard PC (i440FX + PIIX, 1996) task: ffff880009f4d700 ti: ffff880009f5c000 task.ti: ffff880009f5c000 RIP: 0010:[<ffffffff81314790>] [<ffffffff81314790>] blk_get_backing_dev_info+0x10/0x20 RSP: 0018:ffff880009f5fe68 EFLAGS: 00010246 RAX: 0000000000000000 RBX: ffff88000ec17a38 RCX: ffffffff81a4e940 RDX: 7fffffffffffffff RSI: 0000000000000000 RDI: ffff88000ec176c0 RBP: ffff880009f5fe68 R08: 0000000000000000 R09: 0000000000000000 R10: 0000000000000001 R11: 0000000000000000 R12: ffff88000ec17860 R13: ffffffff811b25c0 R14: ffff88000ec178e0 R15: ffff88000ec17a38 FS: 00007faee505d700(0000) GS:ffff88000fb00000(0000) knlGS:0000000000000000 CS: 0010 DS: 0000 ES: 0000 CR0: 000000008005003b CR2: 0000000000000508 CR3: 0000000009e8a000 CR4: 00000000000006e0 Stack: ffff880009f5feb8 ffffffff8112e7f5 0000000000000000 7fffffffffffffff 0000000000000000 0000000000000000 7fffffffffffffff 0000000000000001 ffff88000ec178e0 ffff88000ec17860 ffff880009f5fec8 ffffffff8112e81f Call Trace: [<ffffffff8112e7f5>] __filemap_fdatawrite_range+0x85/0x90 [<ffffffff8112e81f>] filemap_fdatawrite+0x1f/0x30 [<ffffffff811b25d6>] fdatawrite_one_bdev+0x16/0x20 [<ffffffff811bc402>] iterate_bdevs+0xf2/0x130 [<ffffffff811b2763>] sys_sync+0x63/0x90 [<ffffffff815d4272>] entry_SYSCALL_64_fastpath+0x12/0x76 Code: 0f 1f 44 00 00 48 8b 87 f0 00 00 00 55 48 89 e5 <48> 8b 80 08 05 00 00 5d RIP [<ffffffff81314790>] blk_get_backing_dev_info+0x10/0x20 RSP <ffff880009f5fe68> CR2: 0000000000000508 ---[ end trace 2487336ceb3de62d ]--- The crash is easily reproducible by running the following command, if an msleep(100) is inserted before the call to func() in iterate_devs(): while :; do head -c1 /dev/nullb0; done > /dev/null & while :; do sync; done Fix it by holding the bd_mutex across the func() call and only calling func() if the bdev is opened. Cc: stable@vger.kernel.org Fixes: 5c0d6b60a0ba ("vfs: Create function for iterating over block devices") Reported-and-tested-by: Wei Fang <fangwei1@huawei.com> Signed-off-by: Rabin Vincent <rabinv@axis.com> Signed-off-by: Jan Kara <jack@suse.cz> Reviewed-by: Christoph Hellwig <hch@lst.de> Signed-off-by: Jens Axboe <axboe@fb.com>
2016-12-01 08:18:28 +00:00
bdev = I_BDEV(inode);
block: protect iterate_bdevs() against concurrent close If a block device is closed while iterate_bdevs() is handling it, the following NULL pointer dereference occurs because bdev->b_disk is NULL in bdev_get_queue(), which is called from blk_get_backing_dev_info() (in turn called by the mapping_cap_writeback_dirty() call in __filemap_fdatawrite_range()): BUG: unable to handle kernel NULL pointer dereference at 0000000000000508 IP: [<ffffffff81314790>] blk_get_backing_dev_info+0x10/0x20 PGD 9e62067 PUD 9ee8067 PMD 0 Oops: 0000 [#1] PREEMPT SMP DEBUG_PAGEALLOC Modules linked in: CPU: 1 PID: 2422 Comm: sync Not tainted 4.5.0-rc7+ #400 Hardware name: QEMU Standard PC (i440FX + PIIX, 1996) task: ffff880009f4d700 ti: ffff880009f5c000 task.ti: ffff880009f5c000 RIP: 0010:[<ffffffff81314790>] [<ffffffff81314790>] blk_get_backing_dev_info+0x10/0x20 RSP: 0018:ffff880009f5fe68 EFLAGS: 00010246 RAX: 0000000000000000 RBX: ffff88000ec17a38 RCX: ffffffff81a4e940 RDX: 7fffffffffffffff RSI: 0000000000000000 RDI: ffff88000ec176c0 RBP: ffff880009f5fe68 R08: 0000000000000000 R09: 0000000000000000 R10: 0000000000000001 R11: 0000000000000000 R12: ffff88000ec17860 R13: ffffffff811b25c0 R14: ffff88000ec178e0 R15: ffff88000ec17a38 FS: 00007faee505d700(0000) GS:ffff88000fb00000(0000) knlGS:0000000000000000 CS: 0010 DS: 0000 ES: 0000 CR0: 000000008005003b CR2: 0000000000000508 CR3: 0000000009e8a000 CR4: 00000000000006e0 Stack: ffff880009f5feb8 ffffffff8112e7f5 0000000000000000 7fffffffffffffff 0000000000000000 0000000000000000 7fffffffffffffff 0000000000000001 ffff88000ec178e0 ffff88000ec17860 ffff880009f5fec8 ffffffff8112e81f Call Trace: [<ffffffff8112e7f5>] __filemap_fdatawrite_range+0x85/0x90 [<ffffffff8112e81f>] filemap_fdatawrite+0x1f/0x30 [<ffffffff811b25d6>] fdatawrite_one_bdev+0x16/0x20 [<ffffffff811bc402>] iterate_bdevs+0xf2/0x130 [<ffffffff811b2763>] sys_sync+0x63/0x90 [<ffffffff815d4272>] entry_SYSCALL_64_fastpath+0x12/0x76 Code: 0f 1f 44 00 00 48 8b 87 f0 00 00 00 55 48 89 e5 <48> 8b 80 08 05 00 00 5d RIP [<ffffffff81314790>] blk_get_backing_dev_info+0x10/0x20 RSP <ffff880009f5fe68> CR2: 0000000000000508 ---[ end trace 2487336ceb3de62d ]--- The crash is easily reproducible by running the following command, if an msleep(100) is inserted before the call to func() in iterate_devs(): while :; do head -c1 /dev/nullb0; done > /dev/null & while :; do sync; done Fix it by holding the bd_mutex across the func() call and only calling func() if the bdev is opened. Cc: stable@vger.kernel.org Fixes: 5c0d6b60a0ba ("vfs: Create function for iterating over block devices") Reported-and-tested-by: Wei Fang <fangwei1@huawei.com> Signed-off-by: Rabin Vincent <rabinv@axis.com> Signed-off-by: Jan Kara <jack@suse.cz> Reviewed-by: Christoph Hellwig <hch@lst.de> Signed-off-by: Jens Axboe <axboe@fb.com>
2016-12-01 08:18:28 +00:00
mutex_lock(&bdev->bd_mutex);
if (bdev->bd_openers)
func(bdev, arg);
mutex_unlock(&bdev->bd_mutex);
spin_lock(&blockdev_superblock->s_inode_list_lock);
}
spin_unlock(&blockdev_superblock->s_inode_list_lock);
iput(old_inode);
}