linux/fs/ext4/inode.c

5075 lines
147 KiB
C
Raw Normal View History

/*
* linux/fs/ext4/inode.c
*
* Copyright (C) 1992, 1993, 1994, 1995
* Remy Card (card@masi.ibp.fr)
* Laboratoire MASI - Institut Blaise Pascal
* Universite Pierre et Marie Curie (Paris VI)
*
* from
*
* linux/fs/minix/inode.c
*
* Copyright (C) 1991, 1992 Linus Torvalds
*
* 64-bit file support on 64-bit platforms by Jakub Jelinek
* (jj@sunsite.ms.mff.cuni.cz)
*
* Assorted race fixes, rewrite of ext4_get_block() by Al Viro, 2000
*/
#include <linux/fs.h>
#include <linux/time.h>
#include <linux/jbd2.h>
#include <linux/highuid.h>
#include <linux/pagemap.h>
#include <linux/quotaops.h>
#include <linux/string.h>
#include <linux/buffer_head.h>
#include <linux/writeback.h>
#include <linux/pagevec.h>
#include <linux/mpage.h>
#include <linux/namei.h>
#include <linux/uio.h>
#include <linux/bio.h>
#include <linux/workqueue.h>
#include <linux/kernel.h>
#include <linux/printk.h>
include cleanup: Update gfp.h and slab.h includes to prepare for breaking implicit slab.h inclusion from percpu.h percpu.h is included by sched.h and module.h and thus ends up being included when building most .c files. percpu.h includes slab.h which in turn includes gfp.h making everything defined by the two files universally available and complicating inclusion dependencies. percpu.h -> slab.h dependency is about to be removed. Prepare for this change by updating users of gfp and slab facilities include those headers directly instead of assuming availability. As this conversion needs to touch large number of source files, the following script is used as the basis of conversion. http://userweb.kernel.org/~tj/misc/slabh-sweep.py The script does the followings. * Scan files for gfp and slab usages and update includes such that only the necessary includes are there. ie. if only gfp is used, gfp.h, if slab is used, slab.h. * When the script inserts a new include, it looks at the include blocks and try to put the new include such that its order conforms to its surrounding. It's put in the include block which contains core kernel includes, in the same order that the rest are ordered - alphabetical, Christmas tree, rev-Xmas-tree or at the end if there doesn't seem to be any matching order. * If the script can't find a place to put a new include (mostly because the file doesn't have fitting include block), it prints out an error message indicating which .h file needs to be added to the file. The conversion was done in the following steps. 1. The initial automatic conversion of all .c files updated slightly over 4000 files, deleting around 700 includes and adding ~480 gfp.h and ~3000 slab.h inclusions. The script emitted errors for ~400 files. 2. Each error was manually checked. Some didn't need the inclusion, some needed manual addition while adding it to implementation .h or embedding .c file was more appropriate for others. This step added inclusions to around 150 files. 3. The script was run again and the output was compared to the edits from #2 to make sure no file was left behind. 4. Several build tests were done and a couple of problems were fixed. e.g. lib/decompress_*.c used malloc/free() wrappers around slab APIs requiring slab.h to be added manually. 5. The script was run on all .h files but without automatically editing them as sprinkling gfp.h and slab.h inclusions around .h files could easily lead to inclusion dependency hell. Most gfp.h inclusion directives were ignored as stuff from gfp.h was usually wildly available and often used in preprocessor macros. Each slab.h inclusion directive was examined and added manually as necessary. 6. percpu.h was updated not to include slab.h. 7. Build test were done on the following configurations and failures were fixed. CONFIG_GCOV_KERNEL was turned off for all tests (as my distributed build env didn't work with gcov compiles) and a few more options had to be turned off depending on archs to make things build (like ipr on powerpc/64 which failed due to missing writeq). * x86 and x86_64 UP and SMP allmodconfig and a custom test config. * powerpc and powerpc64 SMP allmodconfig * sparc and sparc64 SMP allmodconfig * ia64 SMP allmodconfig * s390 SMP allmodconfig * alpha SMP allmodconfig * um on x86_64 SMP allmodconfig 8. percpu.h modifications were reverted so that it could be applied as a separate patch and serve as bisection point. Given the fact that I had only a couple of failures from tests on step 6, I'm fairly confident about the coverage of this conversion patch. If there is a breakage, it's likely to be something in one of the arch headers which should be easily discoverable easily on most builds of the specific arch. Signed-off-by: Tejun Heo <tj@kernel.org> Guess-its-ok-by: Christoph Lameter <cl@linux-foundation.org> Cc: Ingo Molnar <mingo@redhat.com> Cc: Lee Schermerhorn <Lee.Schermerhorn@hp.com>
2010-03-24 08:04:11 +00:00
#include <linux/slab.h>
#include <linux/ratelimit.h>
#include <linux/aio.h>
#include <linux/bitops.h>
#include "ext4_jbd2.h"
#include "xattr.h"
#include "acl.h"
#include "truncate.h"
#include <trace/events/ext4.h>
#define MPAGE_DA_EXTENT_TAIL 0x01
static __u32 ext4_inode_csum(struct inode *inode, struct ext4_inode *raw,
struct ext4_inode_info *ei)
{
struct ext4_sb_info *sbi = EXT4_SB(inode->i_sb);
__u16 csum_lo;
__u16 csum_hi = 0;
__u32 csum;
csum_lo = le16_to_cpu(raw->i_checksum_lo);
raw->i_checksum_lo = 0;
if (EXT4_INODE_SIZE(inode->i_sb) > EXT4_GOOD_OLD_INODE_SIZE &&
EXT4_FITS_IN_INODE(raw, ei, i_checksum_hi)) {
csum_hi = le16_to_cpu(raw->i_checksum_hi);
raw->i_checksum_hi = 0;
}
csum = ext4_chksum(sbi, ei->i_csum_seed, (__u8 *)raw,
EXT4_INODE_SIZE(inode->i_sb));
raw->i_checksum_lo = cpu_to_le16(csum_lo);
if (EXT4_INODE_SIZE(inode->i_sb) > EXT4_GOOD_OLD_INODE_SIZE &&
EXT4_FITS_IN_INODE(raw, ei, i_checksum_hi))
raw->i_checksum_hi = cpu_to_le16(csum_hi);
return csum;
}
static int ext4_inode_csum_verify(struct inode *inode, struct ext4_inode *raw,
struct ext4_inode_info *ei)
{
__u32 provided, calculated;
if (EXT4_SB(inode->i_sb)->s_es->s_creator_os !=
cpu_to_le32(EXT4_OS_LINUX) ||
!EXT4_HAS_RO_COMPAT_FEATURE(inode->i_sb,
EXT4_FEATURE_RO_COMPAT_METADATA_CSUM))
return 1;
provided = le16_to_cpu(raw->i_checksum_lo);
calculated = ext4_inode_csum(inode, raw, ei);
if (EXT4_INODE_SIZE(inode->i_sb) > EXT4_GOOD_OLD_INODE_SIZE &&
EXT4_FITS_IN_INODE(raw, ei, i_checksum_hi))
provided |= ((__u32)le16_to_cpu(raw->i_checksum_hi)) << 16;
else
calculated &= 0xFFFF;
return provided == calculated;
}
static void ext4_inode_csum_set(struct inode *inode, struct ext4_inode *raw,
struct ext4_inode_info *ei)
{
__u32 csum;
if (EXT4_SB(inode->i_sb)->s_es->s_creator_os !=
cpu_to_le32(EXT4_OS_LINUX) ||
!EXT4_HAS_RO_COMPAT_FEATURE(inode->i_sb,
EXT4_FEATURE_RO_COMPAT_METADATA_CSUM))
return;
csum = ext4_inode_csum(inode, raw, ei);
raw->i_checksum_lo = cpu_to_le16(csum & 0xFFFF);
if (EXT4_INODE_SIZE(inode->i_sb) > EXT4_GOOD_OLD_INODE_SIZE &&
EXT4_FITS_IN_INODE(raw, ei, i_checksum_hi))
raw->i_checksum_hi = cpu_to_le16(csum >> 16);
}
static inline int ext4_begin_ordered_truncate(struct inode *inode,
loff_t new_size)
{
trace_ext4_begin_ordered_truncate(inode, new_size);
/*
* If jinode is zero, then we never opened the file for
* writing, so there's no need to call
* jbd2_journal_begin_ordered_truncate() since there's no
* outstanding writes we need to flush.
*/
if (!EXT4_I(inode)->jinode)
return 0;
return jbd2_journal_begin_ordered_truncate(EXT4_JOURNAL(inode),
EXT4_I(inode)->jinode,
new_size);
}
static void ext4_invalidatepage(struct page *page, unsigned int offset,
unsigned int length);
static int __ext4_journalled_writepage(struct page *page, unsigned int len);
static int ext4_bh_delay_or_unwritten(handle_t *handle, struct buffer_head *bh);
static int ext4_meta_trans_blocks(struct inode *inode, int lblocks,
int pextents);
/*
* Test whether an inode is a fast symlink.
*/
static int ext4_inode_is_fast_symlink(struct inode *inode)
{
int ea_blocks = EXT4_I(inode)->i_file_acl ?
EXT4_CLUSTER_SIZE(inode->i_sb) >> 9 : 0;
ext4: handle symlink properly with inline_data This commit tries to fix a bug that we can't read symlink properly with inline data feature when the length of symlink is greater than 60 bytes but less than extra space. The key issue is in ext4_inode_is_fast_symlink() that it doesn't check whether or not an inode has inline data. When the user creates a new symlink, an inode will be allocated with MAY_INLINE_DATA flag. Then symlink will be stored in ->i_block and extended attribute space. In the mean time, this inode is with inline data flag. After remounting it, ext4_inode_is_fast_symlink() function thinks that this inode is a fast symlink so that the data in ->i_block is copied to the user, and the data in extra space is trimmed. In fact this inode should be as a normal symlink. The following script can hit this bug. #!/bin/bash cd ${MNT} filename=ABCDEFGHIJKLMNOPQRSTUVWXYZabcdefghijklmnopqrstuvwxyz0123456789 rm -rf test mkdir test cd test echo "hello" >$filename ln -s $filename symlinkfile cd sudo umount /mnt/sda1 sudo mount -t ext4 /dev/sda1 /mnt/sda1 readlink /mnt/sda1/test/symlinkfile After applying this patch, it will break the assumption in e2fsck because the original implementation doesn't want to support symlink with inline data. Reported-by: "Darrick J. Wong" <darrick.wong@oracle.com> Reported-by: Ian Nartowicz <claws@nartowicz.co.uk> Cc: Ian Nartowicz <claws@nartowicz.co.uk> Cc: Tao Ma <tm@tao.ma> Cc: "Darrick J. Wong" <darrick.wong@oracle.com> Cc: Andreas Dilger <adilger.kernel@dilger.ca> Signed-off-by: Zheng Liu <wenqing.lz@taobao.com> Signed-off-by: Theodore Ts'o <tytso@mit.edu>
2014-06-02 14:48:22 +00:00
if (ext4_has_inline_data(inode))
return 0;
return (S_ISLNK(inode->i_mode) && inode->i_blocks - ea_blocks == 0);
}
/*
* Restart the transaction associated with *handle. This does a commit,
* so before we call here everything must be consistently dirtied against
* this transaction.
*/
int ext4_truncate_restart_trans(handle_t *handle, struct inode *inode,
int nblocks)
{
int ret;
/*
* Drop i_data_sem to avoid deadlock with ext4_map_blocks. At this
* moment, get_block can be called only for blocks inside i_size since
* page cache has been already dropped and writes are blocked by
* i_mutex. So we can safely drop the i_data_sem here.
*/
BUG_ON(EXT4_JOURNAL(inode) == NULL);
jbd_debug(2, "restarting handle %p\n", handle);
up_write(&EXT4_I(inode)->i_data_sem);
ret = ext4_journal_restart(handle, nblocks);
down_write(&EXT4_I(inode)->i_data_sem);
ext4_discard_preallocations(inode);
return ret;
}
/*
* Called at the last iput() if i_nlink is zero.
*/
void ext4_evict_inode(struct inode *inode)
{
handle_t *handle;
int err;
trace_ext4_evict_inode(inode);
ext4: call ext4_ioend_wait and ext4_flush_completed_IO in ext4_evict_inode Flush inode's i_completed_io_list before calling ext4_io_wait to prevent the following deadlock scenario: A page fault happens while some process is writing inode A. During page fault, shrink_icache_memory is called that in turn evicts another inode B. Inode B has some pending io_end work so it calls ext4_ioend_wait() that waits for inode B's i_ioend_count to become zero. However, inode B's ioend work was queued behind some of inode A's ioend work on the same cpu's ext4-dio-unwritten workqueue. As the ext4-dio-unwritten thread on that cpu is processing inode A's ioend work, it tries to grab inode A's i_mutex lock. Since the i_mutex lock of inode A is still hold before the page fault happened, we enter a deadlock. Also moves ext4_flush_completed_IO and ext4_ioend_wait from ext4_destroy_inode() to ext4_evict_inode(). During inode deleteion, ext4_evict_inode() is called before ext4_destroy_inode() and in ext4_evict_inode(), we may call ext4_truncate() without holding i_mutex lock. As a result, there is a race between flush_completed_IO that is called from ext4_ext_truncate() and ext4_end_io_work, which may cause corruption on an io_end structure. This change moves ext4_flush_completed_IO and ext4_ioend_wait from ext4_destroy_inode() to ext4_evict_inode() to resolve the race between ext4_truncate() and ext4_end_io_work during inode deletion. Signed-off-by: Jiaying Zhang <jiayingz@google.com> Signed-off-by: "Theodore Ts'o" <tytso@mit.edu> Cc: stable@kernel.org
2011-08-13 16:17:13 +00:00
if (inode->i_nlink) {
/*
* When journalling data dirty buffers are tracked only in the
* journal. So although mm thinks everything is clean and
* ready for reaping the inode might still have some pages to
* write in the running transaction or waiting to be
* checkpointed. Thus calling jbd2_journal_invalidatepage()
* (via truncate_inode_pages()) to discard these buffers can
* cause data loss. Also even if we did not discard these
* buffers, we would have no way to find them after the inode
* is reaped and thus user could see stale data if he tries to
* read them before the transaction is checkpointed. So be
* careful and force everything to disk here... We use
* ei->i_datasync_tid to store the newest transaction
* containing inode's data.
*
* Note that directories do not have this problem because they
* don't use page cache.
*/
if (ext4_should_journal_data(inode) &&
ext4: fix data=journal fast mount/umount hang In data=journal mode, if we unmount the file system before a transaction has a chance to complete, when the journal inode is being evicted, we can end up calling into jbd2_log_wait_commit() for the last transaction, after the journalling machinery has been shut down. Arguably we should adjust ext4_should_journal_data() to return FALSE for the journal inode, but the only place it matters is ext4_evict_inode(), and so to save a bit of CPU time, and to make the patch much more obviously correct by inspection(tm), we'll fix it by explicitly not trying to waiting for a journal commit when we are evicting the journal inode, since it's guaranteed to never succeed in this case. This can be easily replicated via: mount -t ext4 -o data=journal /dev/vdb /vdb ; umount /vdb ------------[ cut here ]------------ WARNING: at /usr/projects/linux/ext4/fs/jbd2/journal.c:542 __jbd2_log_start_commit+0xba/0xcd() Hardware name: Bochs JBD2: bad log_start_commit: 3005630206 3005630206 0 0 Modules linked in: Pid: 2909, comm: umount Not tainted 3.8.0-rc3 #1020 Call Trace: [<c015c0ef>] warn_slowpath_common+0x68/0x7d [<c02b7e7d>] ? __jbd2_log_start_commit+0xba/0xcd [<c015c177>] warn_slowpath_fmt+0x2b/0x2f [<c02b7e7d>] __jbd2_log_start_commit+0xba/0xcd [<c02b8075>] jbd2_log_start_commit+0x24/0x34 [<c0279ed5>] ext4_evict_inode+0x71/0x2e3 [<c021f0ec>] evict+0x94/0x135 [<c021f9aa>] iput+0x10a/0x110 [<c02b7836>] jbd2_journal_destroy+0x190/0x1ce [<c0175284>] ? bit_waitqueue+0x50/0x50 [<c028d23f>] ext4_put_super+0x52/0x294 [<c020efe3>] generic_shutdown_super+0x48/0xb4 [<c020f071>] kill_block_super+0x22/0x60 [<c020f3e0>] deactivate_locked_super+0x22/0x49 [<c020f5d6>] deactivate_super+0x30/0x33 [<c0222795>] mntput_no_expire+0x107/0x10c [<c02233a7>] sys_umount+0x2cf/0x2e0 [<c02233ca>] sys_oldumount+0x12/0x14 [<c08096b8>] syscall_call+0x7/0xb ---[ end trace 6a954cc790501c1f ]--- jbd2_log_wait_commit: error: j_commit_request=-1289337090, tid=0 Signed-off-by: "Theodore Ts'o" <tytso@mit.edu> Reviewed-by: Jan Kara <jack@suse.cz> Cc: stable@vger.kernel.org
2013-03-20 13:42:11 +00:00
(S_ISLNK(inode->i_mode) || S_ISREG(inode->i_mode)) &&
inode->i_ino != EXT4_JOURNAL_INO) {
journal_t *journal = EXT4_SB(inode->i_sb)->s_journal;
tid_t commit_tid = EXT4_I(inode)->i_datasync_tid;
jbd2_complete_transaction(journal, commit_tid);
filemap_write_and_wait(&inode->i_data);
}
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);
WARN_ON(atomic_read(&EXT4_I(inode)->i_ioend_count));
goto no_delete;
}
if (!is_bad_inode(inode))
dquot_initialize(inode);
if (ext4_should_order_data(inode))
ext4_begin_ordered_truncate(inode, 0);
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);
WARN_ON(atomic_read(&EXT4_I(inode)->i_ioend_count));
if (is_bad_inode(inode))
goto no_delete;
/*
* Protect us against freezing - iput() caller didn't have to have any
* protection against it
*/
sb_start_intwrite(inode->i_sb);
handle = ext4_journal_start(inode, EXT4_HT_TRUNCATE,
ext4_blocks_for_truncate(inode)+3);
if (IS_ERR(handle)) {
ext4_std_error(inode->i_sb, PTR_ERR(handle));
/*
* If we're going to skip the normal cleanup, we still need to
* make sure that the in-core orphan linked list is properly
* cleaned up.
*/
ext4_orphan_del(NULL, inode);
sb_end_intwrite(inode->i_sb);
goto no_delete;
}
if (IS_SYNC(inode))
ext4_handle_sync(handle);
inode->i_size = 0;
err = ext4_mark_inode_dirty(handle, inode);
if (err) {
ext4_warning(inode->i_sb,
"couldn't mark inode dirty (err %d)", err);
goto stop_handle;
}
if (inode->i_blocks)
ext4_truncate(inode);
/*
* ext4_ext_truncate() doesn't reserve any slop when it
* restarts journal transactions; therefore there may not be
* enough credits left in the handle to remove the inode from
* the orphan list and set the dtime field.
*/
if (!ext4_handle_has_enough_credits(handle, 3)) {
err = ext4_journal_extend(handle, 3);
if (err > 0)
err = ext4_journal_restart(handle, 3);
if (err != 0) {
ext4_warning(inode->i_sb,
"couldn't extend journal (err %d)", err);
stop_handle:
ext4_journal_stop(handle);
ext4_orphan_del(NULL, inode);
sb_end_intwrite(inode->i_sb);
goto no_delete;
}
}
/*
* Kill off the orphan record which ext4_truncate created.
* AKPM: I think this can be inside the above `if'.
* Note that ext4_orphan_del() has to be able to cope with the
* deletion of a non-existent orphan - this is because we don't
* know if ext4_truncate() actually created an orphan record.
* (Well, we could do this if we need to, but heck - it works)
*/
ext4_orphan_del(handle, inode);
EXT4_I(inode)->i_dtime = get_seconds();
/*
* One subtle ordering requirement: if anything has gone wrong
* (transaction abort, IO errors, whatever), then we can still
* do these next steps (the fs will already have been marked as
* having errors), but we can't free the inode if the mark_dirty
* fails.
*/
if (ext4_mark_inode_dirty(handle, inode))
/* If that failed, just do the required in-core inode clear. */
ext4_clear_inode(inode);
else
ext4_free_inode(handle, inode);
ext4_journal_stop(handle);
sb_end_intwrite(inode->i_sb);
return;
no_delete:
ext4_clear_inode(inode); /* We must guarantee clearing of inode... */
}
#ifdef CONFIG_QUOTA
qsize_t *ext4_get_reserved_space(struct inode *inode)
{
return &EXT4_I(inode)->i_reserved_quota;
}
#endif
/*
* Called with i_data_sem down, which is important since we can call
* ext4_discard_preallocations() from here.
*/
void ext4_da_update_reserve_space(struct inode *inode,
int used, int quota_claim)
{
struct ext4_sb_info *sbi = EXT4_SB(inode->i_sb);
struct ext4_inode_info *ei = EXT4_I(inode);
spin_lock(&ei->i_block_reservation_lock);
trace_ext4_da_update_reserve_space(inode, used, quota_claim);
if (unlikely(used > ei->i_reserved_data_blocks)) {
ext4_warning(inode->i_sb, "%s: ino %lu, used %d "
"with only %d reserved data blocks",
__func__, inode->i_ino, used,
ei->i_reserved_data_blocks);
WARN_ON(1);
used = ei->i_reserved_data_blocks;
}
/* Update per-inode reservations */
ei->i_reserved_data_blocks -= used;
percpu_counter_sub(&sbi->s_dirtyclusters_counter, used);
spin_unlock(&EXT4_I(inode)->i_block_reservation_lock);
/* Update quota subsystem for data blocks */
if (quota_claim)
dquot_claim_block(inode, EXT4_C2B(sbi, used));
else {
/*
* We did fallocate with an offset that is already delayed
* allocated. So on delayed allocated writeback we should
* not re-claim the quota for fallocated blocks.
*/
dquot_release_reservation_block(inode, EXT4_C2B(sbi, used));
}
/*
* If we have done all the pending block allocations and if
* there aren't any writers on the inode, we can discard the
* inode's preallocations.
*/
if ((ei->i_reserved_data_blocks == 0) &&
(atomic_read(&inode->i_writecount) == 0))
ext4_discard_preallocations(inode);
}
static int __check_block_validity(struct inode *inode, const char *func,
unsigned int line,
struct ext4_map_blocks *map)
{
if (!ext4_data_block_valid(EXT4_SB(inode->i_sb), map->m_pblk,
map->m_len)) {
ext4_error_inode(inode, func, line, map->m_pblk,
"lblock %lu mapped to illegal pblock "
"(length %d)", (unsigned long) map->m_lblk,
map->m_len);
return -EIO;
}
return 0;
}
#define check_block_validity(inode, map) \
__check_block_validity((inode), __func__, __LINE__, (map))
#ifdef ES_AGGRESSIVE_TEST
static void ext4_map_blocks_es_recheck(handle_t *handle,
struct inode *inode,
struct ext4_map_blocks *es_map,
struct ext4_map_blocks *map,
int flags)
{
int retval;
map->m_flags = 0;
/*
* There is a race window that the result is not the same.
* e.g. xfstests #223 when dioread_nolock enables. The reason
* is that we lookup a block mapping in extent status tree with
* out taking i_data_sem. So at the time the unwritten extent
* could be converted.
*/
if (!(flags & EXT4_GET_BLOCKS_NO_LOCK))
down_read(&EXT4_I(inode)->i_data_sem);
if (ext4_test_inode_flag(inode, EXT4_INODE_EXTENTS)) {
retval = ext4_ext_map_blocks(handle, inode, map, flags &
EXT4_GET_BLOCKS_KEEP_SIZE);
} else {
retval = ext4_ind_map_blocks(handle, inode, map, flags &
EXT4_GET_BLOCKS_KEEP_SIZE);
}
if (!(flags & EXT4_GET_BLOCKS_NO_LOCK))
up_read((&EXT4_I(inode)->i_data_sem));
/*
* Clear EXT4_MAP_FROM_CLUSTER and EXT4_MAP_BOUNDARY flag
* because it shouldn't be marked in es_map->m_flags.
*/
map->m_flags &= ~(EXT4_MAP_FROM_CLUSTER | EXT4_MAP_BOUNDARY);
/*
* We don't check m_len because extent will be collpased in status
* tree. So the m_len might not equal.
*/
if (es_map->m_lblk != map->m_lblk ||
es_map->m_flags != map->m_flags ||
es_map->m_pblk != map->m_pblk) {
printk("ES cache assertion failed for inode: %lu "
"es_cached ex [%d/%d/%llu/%x] != "
"found ex [%d/%d/%llu/%x] retval %d flags %x\n",
inode->i_ino, es_map->m_lblk, es_map->m_len,
es_map->m_pblk, es_map->m_flags, map->m_lblk,
map->m_len, map->m_pblk, map->m_flags,
retval, flags);
}
}
#endif /* ES_AGGRESSIVE_TEST */
/*
* The ext4_map_blocks() function tries to look up the requested blocks,
* and returns if the blocks are already mapped.
*
* Otherwise it takes the write lock of the i_data_sem and allocate blocks
* and store the allocated blocks in the result buffer head and mark it
* mapped.
*
* If file type is extents based, it will call ext4_ext_map_blocks(),
* Otherwise, call with ext4_ind_map_blocks() to handle indirect mapping
* based files
*
* On success, it returns the number of blocks being mapped or allocated.
* if create==0 and the blocks are pre-allocated and unwritten block,
* the result buffer head is unmapped. If the create ==1, it will make sure
* the buffer head is mapped.
*
* It returns 0 if plain look up failed (blocks have not been allocated), in
* that case, buffer head is unmapped
*
* It returns the error in case of allocation failure.
*/
int ext4_map_blocks(handle_t *handle, struct inode *inode,
struct ext4_map_blocks *map, int flags)
{
struct extent_status es;
int retval;
int ret = 0;
#ifdef ES_AGGRESSIVE_TEST
struct ext4_map_blocks orig_map;
memcpy(&orig_map, map, sizeof(*map));
#endif
map->m_flags = 0;
ext_debug("ext4_map_blocks(): inode %lu, flag %d, max_blocks %u,"
"logical block %lu\n", inode->i_ino, flags, map->m_len,
(unsigned long) map->m_lblk);
ext4: avoid possible overflow in ext4_map_blocks() The ext4_map_blocks() function returns the number of blocks which satisfying the caller's request. This number of blocks requested by the caller is specified by an unsigned integer, but the return value of ext4_map_blocks() is a signed integer (to accomodate error codes per the kernel's standard error signalling convention). Historically, overflows could never happen since mballoc() will refuse to allocate more than 2048 blocks at a time (which is something we should fix), and if the blocks were already allocated, the fact that there would be some number of intervening metadata blocks pretty much guaranteed that there could never be a contiguous region of data blocks that was greater than 2**31 blocks. However, this is now possible if there is a file system which is a bit bigger than 8TB, and is created using the new mke2fs hugeblock feature, which can create a perfectly contiguous file. In that case, if a userspace program attempted to call fallocate() on this already fully allocated file, it's possible that ext4_map_blocks() could return a number large enough that it would overflow a signed integer, resulting in a ext4 thinking that the ext4_map_blocks() call had failed with some strange error code. Since ext4_map_blocks() is always free to return a smaller number of blocks than what was requested by the caller, fix this by capping the number of blocks that ext4_map_blocks() will ever try to map to 2**31 - 1. In practice this should never get hit, except by someone deliberately trying to provke the above-described bug. Thanks to the PaX team for asking whethre this could possibly happen in some off-line discussions about using some static code checking technology they are developing to find bugs in kernel code. Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
2014-02-20 17:54:05 +00:00
/*
* ext4_map_blocks returns an int, and m_len is an unsigned int
*/
if (unlikely(map->m_len > INT_MAX))
map->m_len = INT_MAX;
/* We can handle the block number less than EXT_MAX_BLOCKS */
if (unlikely(map->m_lblk >= EXT_MAX_BLOCKS))
return -EIO;
/* Lookup extent status tree firstly */
if (ext4_es_lookup_extent(inode, map->m_lblk, &es)) {
ext4_es_lru_add(inode);
if (ext4_es_is_written(&es) || ext4_es_is_unwritten(&es)) {
map->m_pblk = ext4_es_pblock(&es) +
map->m_lblk - es.es_lblk;
map->m_flags |= ext4_es_is_written(&es) ?
EXT4_MAP_MAPPED : EXT4_MAP_UNWRITTEN;
retval = es.es_len - (map->m_lblk - es.es_lblk);
if (retval > map->m_len)
retval = map->m_len;
map->m_len = retval;
} else if (ext4_es_is_delayed(&es) || ext4_es_is_hole(&es)) {
retval = 0;
} else {
BUG_ON(1);
}
#ifdef ES_AGGRESSIVE_TEST
ext4_map_blocks_es_recheck(handle, inode, map,
&orig_map, flags);
#endif
goto found;
}
/*
* Try to see if we can get the block without requesting a new
* file system block.
*/
if (!(flags & EXT4_GET_BLOCKS_NO_LOCK))
down_read(&EXT4_I(inode)->i_data_sem);
if (ext4_test_inode_flag(inode, EXT4_INODE_EXTENTS)) {
retval = ext4_ext_map_blocks(handle, inode, map, flags &
EXT4_GET_BLOCKS_KEEP_SIZE);
} else {
retval = ext4_ind_map_blocks(handle, inode, map, flags &
EXT4_GET_BLOCKS_KEEP_SIZE);
}
if (retval > 0) {
unsigned int status;
if (unlikely(retval != map->m_len)) {
ext4_warning(inode->i_sb,
"ES len assertion failed for inode "
"%lu: retval %d != map->m_len %d",
inode->i_ino, retval, map->m_len);
WARN_ON(1);
}
status = map->m_flags & EXT4_MAP_UNWRITTEN ?
EXTENT_STATUS_UNWRITTEN : EXTENT_STATUS_WRITTEN;
if (!(flags & EXT4_GET_BLOCKS_DELALLOC_RESERVE) &&
ext4_find_delalloc_range(inode, map->m_lblk,
map->m_lblk + map->m_len - 1))
status |= EXTENT_STATUS_DELAYED;
ret = ext4_es_insert_extent(inode, map->m_lblk,
map->m_len, map->m_pblk, status);
if (ret < 0)
retval = ret;
}
if (!(flags & EXT4_GET_BLOCKS_NO_LOCK))
up_read((&EXT4_I(inode)->i_data_sem));
found:
if (retval > 0 && map->m_flags & EXT4_MAP_MAPPED) {
ret = check_block_validity(inode, map);
if (ret != 0)
return ret;
}
/* If it is only a block(s) look up */
if ((flags & EXT4_GET_BLOCKS_CREATE) == 0)
return retval;
/*
* Returns if the blocks have already allocated
*
* Note that if blocks have been preallocated
* ext4_ext_get_block() returns the create = 0
* with buffer head unmapped.
*/
if (retval > 0 && map->m_flags & EXT4_MAP_MAPPED)
/*
* If we need to convert extent to unwritten
* we continue and do the actual work in
* ext4_ext_map_blocks()
*/
if (!(flags & EXT4_GET_BLOCKS_CONVERT_UNWRITTEN))
return retval;
/*
* Here we clear m_flags because after allocating an new extent,
* it will be set again.
*/
map->m_flags &= ~EXT4_MAP_FLAGS;
/*
* New blocks allocate and/or writing to unwritten extent
* will possibly result in updating i_data, so we take
* the write lock of i_data_sem, and call get_blocks()
* with create == 1 flag.
*/
down_write(&EXT4_I(inode)->i_data_sem);
/*
* if the caller is from delayed allocation writeout path
* we have already reserved fs blocks for allocation
* let the underlying get_block() function know to
* avoid double accounting
*/
if (flags & EXT4_GET_BLOCKS_DELALLOC_RESERVE)
ext4_set_inode_state(inode, EXT4_STATE_DELALLOC_RESERVED);
/*
* We need to check for EXT4 here because migrate
* could have changed the inode type in between
*/
if (ext4_test_inode_flag(inode, EXT4_INODE_EXTENTS)) {
retval = ext4_ext_map_blocks(handle, inode, map, flags);
} else {
retval = ext4_ind_map_blocks(handle, inode, map, flags);
if (retval > 0 && map->m_flags & EXT4_MAP_NEW) {
/*
* We allocated new blocks which will result in
* i_data's format changing. Force the migrate
* to fail by clearing migrate flags
*/
ext4_clear_inode_state(inode, EXT4_STATE_EXT_MIGRATE);
}
/*
* Update reserved blocks/metadata blocks after successful
* block allocation which had been deferred till now. We don't
* support fallocate for non extent files. So we can update
* reserve space here.
*/
if ((retval > 0) &&
(flags & EXT4_GET_BLOCKS_DELALLOC_RESERVE))
ext4_da_update_reserve_space(inode, retval, 1);
}
if (flags & EXT4_GET_BLOCKS_DELALLOC_RESERVE)
ext4_clear_inode_state(inode, EXT4_STATE_DELALLOC_RESERVED);
if (retval > 0) {
unsigned int status;
if (unlikely(retval != map->m_len)) {
ext4_warning(inode->i_sb,
"ES len assertion failed for inode "
"%lu: retval %d != map->m_len %d",
inode->i_ino, retval, map->m_len);
WARN_ON(1);
}
/*
* If the extent has been zeroed out, we don't need to update
* extent status tree.
*/
if ((flags & EXT4_GET_BLOCKS_PRE_IO) &&
ext4_es_lookup_extent(inode, map->m_lblk, &es)) {
if (ext4_es_is_written(&es))
goto has_zeroout;
}
status = map->m_flags & EXT4_MAP_UNWRITTEN ?
EXTENT_STATUS_UNWRITTEN : EXTENT_STATUS_WRITTEN;
if (!(flags & EXT4_GET_BLOCKS_DELALLOC_RESERVE) &&
ext4_find_delalloc_range(inode, map->m_lblk,
map->m_lblk + map->m_len - 1))
status |= EXTENT_STATUS_DELAYED;
ret = ext4_es_insert_extent(inode, map->m_lblk, map->m_len,
map->m_pblk, status);
if (ret < 0)
retval = ret;
ext4: attempt to fix race in bigalloc code path Currently, there exists a race between delayed allocated writes and the writeback when bigalloc feature is in use. The race was because we wanted to determine what blocks in a cluster are under delayed allocation and we were using buffer_delayed(bh) check for it. But, the writeback codepath clears this bit without any synchronization which resulted in a race and an ext4 warning similar to: EXT4-fs (ram1): ext4_da_update_reserve_space: ino 13, used 1 with only 0 reserved data blocks The race existed in two places. (1) between ext4_find_delalloc_range() and ext4_map_blocks() when called from writeback code path. (2) between ext4_find_delalloc_range() and ext4_da_get_block_prep() (where buffer_delayed(bh) is set. To fix (1), this patch introduces a new buffer_head state bit - BH_Da_Mapped. This bit is set under the protection of EXT4_I(inode)->i_data_sem when we have actually mapped the delayed allocated blocks during the writeout time. We can now reliably check for this bit inside ext4_find_delalloc_range() to determine whether the reservation for the blocks have already been claimed or not. To fix (2), it was necessary to set buffer_delay(bh) under the protection of i_data_sem. So, I extracted the very beginning of ext4_map_blocks into a new function - ext4_da_map_blocks() - and performed the required setting of bh_delay bit and the quota reservation under the protection of i_data_sem. These two fixes makes the checking of buffer_delay(bh) and buffer_da_mapped(bh) consistent, thus removing the race. Tested: I was able to reproduce the problem by running 'dd' and 'fsync' in parallel. Also, xfstests sometimes used to reproduce this race. After the fix both my test and xfstests were successful and no race (warning message) was observed. Google-Bug-Id: 4997027 Signed-off-by: Aditya Kali <adityakali@google.com> Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
2011-09-09 23:20:51 +00:00
}
has_zeroout:
up_write((&EXT4_I(inode)->i_data_sem));
if (retval > 0 && map->m_flags & EXT4_MAP_MAPPED) {
ret = check_block_validity(inode, map);
if (ret != 0)
return ret;
}
return retval;
}
/* Maximum number of blocks we map for direct IO at once. */
#define DIO_MAX_BLOCKS 4096
static int _ext4_get_block(struct inode *inode, sector_t iblock,
struct buffer_head *bh, int flags)
{
handle_t *handle = ext4_journal_current_handle();
struct ext4_map_blocks map;
int ret = 0, started = 0;
int dio_credits;
if (ext4_has_inline_data(inode))
return -ERANGE;
map.m_lblk = iblock;
map.m_len = bh->b_size >> inode->i_blkbits;
if (flags && !(flags & EXT4_GET_BLOCKS_NO_LOCK) && !handle) {
/* Direct IO write... */
if (map.m_len > DIO_MAX_BLOCKS)
map.m_len = DIO_MAX_BLOCKS;
dio_credits = ext4_chunk_trans_blocks(inode, map.m_len);
handle = ext4_journal_start(inode, EXT4_HT_MAP_BLOCKS,
dio_credits);
if (IS_ERR(handle)) {
ret = PTR_ERR(handle);
return ret;
}
started = 1;
}
ret = ext4_map_blocks(handle, inode, &map, flags);
if (ret > 0) {
ext4_io_end_t *io_end = ext4_inode_aio(inode);
map_bh(bh, inode->i_sb, map.m_pblk);
bh->b_state = (bh->b_state & ~EXT4_MAP_FLAGS) | map.m_flags;
if (io_end && io_end->flag & EXT4_IO_END_UNWRITTEN)
set_buffer_defer_completion(bh);
bh->b_size = inode->i_sb->s_blocksize * map.m_len;
ret = 0;
}
if (started)
ext4_journal_stop(handle);
return ret;
}
int ext4_get_block(struct inode *inode, sector_t iblock,
struct buffer_head *bh, int create)
{
return _ext4_get_block(inode, iblock, bh,
create ? EXT4_GET_BLOCKS_CREATE : 0);
}
/*
* `handle' can be NULL if create is zero
*/
struct buffer_head *ext4_getblk(handle_t *handle, struct inode *inode,
ext4_lblk_t block, int create, int *errp)
{
struct ext4_map_blocks map;
struct buffer_head *bh;
int fatal = 0, err;
J_ASSERT(handle != NULL || create == 0);
map.m_lblk = block;
map.m_len = 1;
err = ext4_map_blocks(handle, inode, &map,
create ? EXT4_GET_BLOCKS_CREATE : 0);
ext4: fix possible non-initialized variable in htree_dirblock_to_tree() htree_dirblock_to_tree() declares a non-initialized 'err' variable, which is passed as a reference to another functions expecting them to set this variable with their error codes. It's passed to ext4_bread(), which then passes it to ext4_getblk(). If ext4_map_blocks() returns 0 due to a lookup failure, leaving the ext4_getblk() buffer_head uninitialized, it will make ext4_getblk() return to ext4_bread() without initialize the 'err' variable, and ext4_bread() will return to htree_dirblock_to_tree() with this variable still uninitialized. htree_dirblock_to_tree() will pass this variable with garbage back to ext4_htree_fill_tree(), which expects a number of directory entries added to the rb-tree. which, in case, might return a fake non-zero value due the garbage left in the 'err' variable, leading the kernel to an Oops in ext4_dx_readdir(), once this is expecting a filled rb-tree node, when in turn it will have a NULL-ed one, causing an invalid page request when trying to get a fname struct from this NULL-ed rb-tree node in this line: fname = rb_entry(info->curr_node, struct fname, rb_hash); The patch itself initializes the err variable in htree_dirblock_to_tree() to avoid usage mistakes by the called functions, and also fix ext4_getblk() to return a initialized 'err' variable when ext4_map_blocks() fails a lookup. Signed-off-by: Carlos Maiolino <cmaiolino@redhat.com> Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
2012-09-18 03:39:12 +00:00
/* ensure we send some value back into *errp */
*errp = 0;
if (create && err == 0)
err = -ENOSPC; /* should never happen */
if (err < 0)
*errp = err;
if (err <= 0)
return NULL;
bh = sb_getblk(inode->i_sb, map.m_pblk);
if (unlikely(!bh)) {
*errp = -ENOMEM;
return NULL;
}
if (map.m_flags & EXT4_MAP_NEW) {
J_ASSERT(create != 0);
J_ASSERT(handle != NULL);
/*
* Now that we do not always journal data, we should
* keep in mind whether this should always journal the
* new buffer as metadata. For now, regular file
* writes use ext4_get_block instead, so it's not a
* problem.
*/
lock_buffer(bh);
BUFFER_TRACE(bh, "call get_create_access");
fatal = ext4_journal_get_create_access(handle, bh);
if (!fatal && !buffer_uptodate(bh)) {
memset(bh->b_data, 0, inode->i_sb->s_blocksize);
set_buffer_uptodate(bh);
}
unlock_buffer(bh);
BUFFER_TRACE(bh, "call ext4_handle_dirty_metadata");
err = ext4_handle_dirty_metadata(handle, inode, bh);
if (!fatal)
fatal = err;
} else {
BUFFER_TRACE(bh, "not a new buffer");
}
if (fatal) {
*errp = fatal;
brelse(bh);
bh = NULL;
}
return bh;
}
struct buffer_head *ext4_bread(handle_t *handle, struct inode *inode,
ext4_lblk_t block, int create, int *err)
{
struct buffer_head *bh;
bh = ext4_getblk(handle, inode, block, create, err);
if (!bh)
return bh;
if (buffer_uptodate(bh))
return bh;
ll_rw_block(READ | REQ_META | REQ_PRIO, 1, &bh);
wait_on_buffer(bh);
if (buffer_uptodate(bh))
return bh;
put_bh(bh);
*err = -EIO;
return NULL;
}
int ext4_walk_page_buffers(handle_t *handle,
struct buffer_head *head,
unsigned from,
unsigned to,
int *partial,
int (*fn)(handle_t *handle,
struct buffer_head *bh))
{
struct buffer_head *bh;
unsigned block_start, block_end;
unsigned blocksize = head->b_size;
int err, ret = 0;
struct buffer_head *next;
for (bh = head, block_start = 0;
ret == 0 && (bh != head || !block_start);
block_start = block_end, bh = next) {
next = bh->b_this_page;
block_end = block_start + blocksize;
if (block_end <= from || block_start >= to) {
if (partial && !buffer_uptodate(bh))
*partial = 1;
continue;
}
err = (*fn)(handle, bh);
if (!ret)
ret = err;
}
return ret;
}
/*
* To preserve ordering, it is essential that the hole instantiation and
* the data write be encapsulated in a single transaction. We cannot
* close off a transaction and start a new one between the ext4_get_block()
* and the commit_write(). So doing the jbd2_journal_start at the start of
* prepare_write() is the right place.
*
* Also, this function can nest inside ext4_writepage(). In that case, we
* *know* that ext4_writepage() has generated enough buffer credits to do the
* whole page. So we won't block on the journal in that case, which is good,
* because the caller may be PF_MEMALLOC.
*
* By accident, ext4 can be reentered when a transaction is open via
* quota file writes. If we were to commit the transaction while thus
* reentered, there can be a deadlock - we would be holding a quota
* lock, and the commit would never complete if another thread had a
* transaction open and was blocking on the quota lock - a ranking
* violation.
*
* So what we do is to rely on the fact that jbd2_journal_stop/journal_start
* will _not_ run commit under these circumstances because handle->h_ref
* is elevated. We'll still have enough credits for the tiny quotafile
* write.
*/
int do_journal_get_write_access(handle_t *handle,
struct buffer_head *bh)
{
int dirty = buffer_dirty(bh);
int ret;
if (!buffer_mapped(bh) || buffer_freed(bh))
return 0;
/*
* __block_write_begin() could have dirtied some buffers. Clean
* the dirty bit as jbd2_journal_get_write_access() could complain
* otherwise about fs integrity issues. Setting of the dirty bit
* by __block_write_begin() isn't a real problem here as we clear
* the bit before releasing a page lock and thus writeback cannot
* ever write the buffer.
*/
if (dirty)
clear_buffer_dirty(bh);
BUFFER_TRACE(bh, "get write access");
ret = ext4_journal_get_write_access(handle, bh);
if (!ret && dirty)
ret = ext4_handle_dirty_metadata(handle, NULL, bh);
return ret;
}
static int ext4_get_block_write_nolock(struct inode *inode, sector_t iblock,
struct buffer_head *bh_result, int create);
static int ext4_write_begin(struct file *file, struct address_space *mapping,
loff_t pos, unsigned len, unsigned flags,
struct page **pagep, void **fsdata)
{
struct inode *inode = mapping->host;
int ret, needed_blocks;
handle_t *handle;
int retries = 0;
struct page *page;
pgoff_t index;
unsigned from, to;
trace_ext4_write_begin(inode, pos, len, flags);
/*
* Reserve one block more for addition to orphan list in case
* we allocate blocks but write fails for some reason
*/
needed_blocks = ext4_writepage_trans_blocks(inode) + 1;
index = pos >> PAGE_CACHE_SHIFT;
from = pos & (PAGE_CACHE_SIZE - 1);
to = from + len;
if (ext4_test_inode_state(inode, EXT4_STATE_MAY_INLINE_DATA)) {
ret = ext4_try_to_write_inline_data(mapping, inode, pos, len,
flags, pagep);
if (ret < 0)
return ret;
if (ret == 1)
return 0;
}
/*
* grab_cache_page_write_begin() can take a long time if the
* system is thrashing due to memory pressure, or if the page
* is being written back. So grab it first before we start
* the transaction handle. This also allows us to allocate
* the page (if needed) without using GFP_NOFS.
*/
retry_grab:
page = grab_cache_page_write_begin(mapping, index, flags);
if (!page)
return -ENOMEM;
unlock_page(page);
retry_journal:
handle = ext4_journal_start(inode, EXT4_HT_WRITE_PAGE, needed_blocks);
if (IS_ERR(handle)) {
page_cache_release(page);
return PTR_ERR(handle);
}
lock_page(page);
if (page->mapping != mapping) {
/* The page got truncated from under us */
unlock_page(page);
page_cache_release(page);
ext4_journal_stop(handle);
goto retry_grab;
}
/* In case writeback began while the page was unlocked */
wait_for_stable_page(page);
if (ext4_should_dioread_nolock(inode))
ret = __block_write_begin(page, pos, len, ext4_get_block_write);
else
ret = __block_write_begin(page, pos, len, ext4_get_block);
if (!ret && ext4_should_journal_data(inode)) {
ret = ext4_walk_page_buffers(handle, page_buffers(page),
from, to, NULL,
do_journal_get_write_access);
}
if (ret) {
unlock_page(page);
/*
* __block_write_begin may have instantiated a few blocks
* outside i_size. Trim these off again. Don't need
* i_size_read because we hold i_mutex.
*
* Add inode to orphan list in case we crash before
* truncate finishes
*/
if (pos + len > inode->i_size && ext4_can_truncate(inode))
ext4_orphan_add(handle, inode);
ext4_journal_stop(handle);
if (pos + len > inode->i_size) {
ext4_truncate_failed_write(inode);
/*
* If truncate failed early the inode might
* still be on the orphan list; we need to
* make sure the inode is removed from the
* orphan list in that case.
*/
if (inode->i_nlink)
ext4_orphan_del(NULL, inode);
}
if (ret == -ENOSPC &&
ext4_should_retry_alloc(inode->i_sb, &retries))
goto retry_journal;
page_cache_release(page);
return ret;
}
*pagep = page;
return ret;
}
/* For write_end() in data=journal mode */
static int write_end_fn(handle_t *handle, struct buffer_head *bh)
{
int ret;
if (!buffer_mapped(bh) || buffer_freed(bh))
return 0;
set_buffer_uptodate(bh);
ret = ext4_handle_dirty_metadata(handle, NULL, bh);
clear_buffer_meta(bh);
clear_buffer_prio(bh);
return ret;
}
/*
* We need to pick up the new inode size which generic_commit_write gave us
* `file' can be NULL - eg, when called from page_symlink().
*
* ext4 never places buffers on inode->i_mapping->private_list. metadata
* buffers are managed internally.
*/
static int ext4_write_end(struct file *file,
struct address_space *mapping,
loff_t pos, unsigned len, unsigned copied,
struct page *page, void *fsdata)
{
handle_t *handle = ext4_journal_current_handle();
struct inode *inode = mapping->host;
int ret = 0, ret2;
int i_size_changed = 0;
trace_ext4_write_end(inode, pos, len, copied);
if (ext4_test_inode_state(inode, EXT4_STATE_ORDERED_MODE)) {
ret = ext4_jbd2_file_inode(handle, inode);
if (ret) {
unlock_page(page);
page_cache_release(page);
goto errout;
}
}
if (ext4_has_inline_data(inode)) {
ret = ext4_write_inline_data_end(inode, pos, len,
copied, page);
if (ret < 0)
goto errout;
copied = ret;
} else
copied = block_write_end(file, mapping, pos,
len, copied, page, fsdata);
/*
* it's important to update i_size while still holding page lock:
* page writeout could otherwise come in and zero beyond i_size.
*/
i_size_changed = ext4_update_inode_size(inode, pos + copied);
unlock_page(page);
page_cache_release(page);
/*
* Don't mark the inode dirty under page lock. First, it unnecessarily
* makes the holding time of page lock longer. Second, it forces lock
* ordering of page lock and transaction start for journaling
* filesystems.
*/
if (i_size_changed)
ext4_mark_inode_dirty(handle, inode);
if (pos + len > inode->i_size && ext4_can_truncate(inode))
/* if we have allocated more blocks and copied
* less. We will have blocks allocated outside
* inode->i_size. So truncate them
*/
ext4_orphan_add(handle, inode);
errout:
ret2 = ext4_journal_stop(handle);
if (!ret)
ret = ret2;
if (pos + len > inode->i_size) {
ext4_truncate_failed_write(inode);
/*
* If truncate failed early the inode might still be
* on the orphan list; we need to make sure the inode
* is removed from the orphan list in that case.
*/
if (inode->i_nlink)
ext4_orphan_del(NULL, inode);
}
return ret ? ret : copied;
}
static int ext4_journalled_write_end(struct file *file,
struct address_space *mapping,
loff_t pos, unsigned len, unsigned copied,
struct page *page, void *fsdata)
{
handle_t *handle = ext4_journal_current_handle();
struct inode *inode = mapping->host;
int ret = 0, ret2;
int partial = 0;
unsigned from, to;
int size_changed = 0;
trace_ext4_journalled_write_end(inode, pos, len, copied);
from = pos & (PAGE_CACHE_SIZE - 1);
to = from + len;
BUG_ON(!ext4_handle_valid(handle));
if (ext4_has_inline_data(inode))
copied = ext4_write_inline_data_end(inode, pos, len,
copied, page);
else {
if (copied < len) {
if (!PageUptodate(page))
copied = 0;
page_zero_new_buffers(page, from+copied, to);
}
ret = ext4_walk_page_buffers(handle, page_buffers(page), from,
to, &partial, write_end_fn);
if (!partial)
SetPageUptodate(page);
}
size_changed = ext4_update_inode_size(inode, pos + copied);
ext4_set_inode_state(inode, EXT4_STATE_JDATA);
EXT4_I(inode)->i_datasync_tid = handle->h_transaction->t_tid;
unlock_page(page);
page_cache_release(page);
if (size_changed) {
ret2 = ext4_mark_inode_dirty(handle, inode);
if (!ret)
ret = ret2;
}
if (pos + len > inode->i_size && ext4_can_truncate(inode))
/* if we have allocated more blocks and copied
* less. We will have blocks allocated outside
* inode->i_size. So truncate them
*/
ext4_orphan_add(handle, inode);
ret2 = ext4_journal_stop(handle);
if (!ret)
ret = ret2;
if (pos + len > inode->i_size) {
ext4_truncate_failed_write(inode);
/*
* If truncate failed early the inode might still be
* on the orphan list; we need to make sure the inode
* is removed from the orphan list in that case.
*/
if (inode->i_nlink)
ext4_orphan_del(NULL, inode);
}
return ret ? ret : copied;
}
/*
* Reserve a single cluster located at lblock
*/
static int ext4_da_reserve_space(struct inode *inode, ext4_lblk_t lblock)
{
struct ext4_sb_info *sbi = EXT4_SB(inode->i_sb);
struct ext4_inode_info *ei = EXT4_I(inode);
unsigned int md_needed;
int ret;
/*
* We will charge metadata quota at writeout time; this saves
* us from metadata over-estimation, though we may go over by
* a small amount in the end. Here we just reserve for data.
*/
ret = dquot_reserve_block(inode, EXT4_C2B(sbi, 1));
if (ret)
return ret;
/*
* recalculate the amount of metadata blocks to reserve
* in order to allocate nrblocks
* worse case is one extent per block
*/
spin_lock(&ei->i_block_reservation_lock);
/*
* ext4_calc_metadata_amount() has side effects, which we have
* to be prepared undo if we fail to claim space.
*/
md_needed = 0;
trace_ext4_da_reserve_space(inode, 0);
if (ext4_claim_free_clusters(sbi, 1, 0)) {
spin_unlock(&ei->i_block_reservation_lock);
dquot_release_reservation_block(inode, EXT4_C2B(sbi, 1));
return -ENOSPC;
}
ei->i_reserved_data_blocks++;
spin_unlock(&ei->i_block_reservation_lock);
return 0; /* success */
}
static void ext4_da_release_space(struct inode *inode, int to_free)
{
struct ext4_sb_info *sbi = EXT4_SB(inode->i_sb);
struct ext4_inode_info *ei = EXT4_I(inode);
if (!to_free)
return; /* Nothing to release, exit */
spin_lock(&EXT4_I(inode)->i_block_reservation_lock);
trace_ext4_da_release_space(inode, to_free);
if (unlikely(to_free > ei->i_reserved_data_blocks)) {
/*
* if there aren't enough reserved blocks, then the
* counter is messed up somewhere. Since this
* function is called from invalidate page, it's
* harmless to return without any action.
*/
ext4_warning(inode->i_sb, "ext4_da_release_space: "
"ino %lu, to_free %d with only %d reserved "
"data blocks", inode->i_ino, to_free,
ei->i_reserved_data_blocks);
WARN_ON(1);
to_free = ei->i_reserved_data_blocks;
}
ei->i_reserved_data_blocks -= to_free;
/* update fs dirty data blocks counter */
percpu_counter_sub(&sbi->s_dirtyclusters_counter, to_free);
spin_unlock(&EXT4_I(inode)->i_block_reservation_lock);
dquot_release_reservation_block(inode, EXT4_C2B(sbi, to_free));
}
static void ext4_da_page_release_reservation(struct page *page,
unsigned int offset,
unsigned int length)
{
int to_release = 0;
struct buffer_head *head, *bh;
unsigned int curr_off = 0;
struct inode *inode = page->mapping->host;
struct ext4_sb_info *sbi = EXT4_SB(inode->i_sb);
unsigned int stop = offset + length;
int num_clusters;
ext4_fsblk_t lblk;
BUG_ON(stop > PAGE_CACHE_SIZE || stop < length);
head = page_buffers(page);
bh = head;
do {
unsigned int next_off = curr_off + bh->b_size;
if (next_off > stop)
break;
if ((offset <= curr_off) && (buffer_delay(bh))) {
to_release++;
clear_buffer_delay(bh);
}
curr_off = next_off;
} while ((bh = bh->b_this_page) != head);
if (to_release) {
lblk = page->index << (PAGE_CACHE_SHIFT - inode->i_blkbits);
ext4_es_remove_extent(inode, lblk, to_release);
}
/* If we have released all the blocks belonging to a cluster, then we
* need to release the reserved space for that cluster. */
num_clusters = EXT4_NUM_B2C(sbi, to_release);
while (num_clusters > 0) {
lblk = (page->index << (PAGE_CACHE_SHIFT - inode->i_blkbits)) +
((num_clusters - 1) << sbi->s_cluster_bits);
if (sbi->s_cluster_ratio == 1 ||
!ext4_find_delalloc_cluster(inode, lblk))
ext4_da_release_space(inode, 1);
num_clusters--;
}
}
/*
* Delayed allocation stuff
*/
struct mpage_da_data {
struct inode *inode;
struct writeback_control *wbc;
pgoff_t first_page; /* The first page to write */
pgoff_t next_page; /* Current page to examine */
pgoff_t last_page; /* Last page to examine */
/*
* Extent to map - this can be after first_page because that can be
* fully mapped. We somewhat abuse m_flags to store whether the extent
* is delalloc or unwritten.
*/
struct ext4_map_blocks map;
struct ext4_io_submit io_submit; /* IO submission data */
};
static void mpage_release_unused_pages(struct mpage_da_data *mpd,
bool invalidate)
{
int nr_pages, i;
pgoff_t index, end;
struct pagevec pvec;
struct inode *inode = mpd->inode;
struct address_space *mapping = inode->i_mapping;
/* This is necessary when next_page == 0. */
if (mpd->first_page >= mpd->next_page)
return;
index = mpd->first_page;
end = mpd->next_page - 1;
if (invalidate) {
ext4_lblk_t start, last;
start = index << (PAGE_CACHE_SHIFT - inode->i_blkbits);
last = end << (PAGE_CACHE_SHIFT - inode->i_blkbits);
ext4_es_remove_extent(inode, start, last - start + 1);
}
pagevec_init(&pvec, 0);
while (index <= end) {
nr_pages = pagevec_lookup(&pvec, mapping, index, PAGEVEC_SIZE);
if (nr_pages == 0)
break;
for (i = 0; i < nr_pages; i++) {
struct page *page = pvec.pages[i];
if (page->index > end)
break;
BUG_ON(!PageLocked(page));
BUG_ON(PageWriteback(page));
if (invalidate) {
block_invalidatepage(page, 0, PAGE_CACHE_SIZE);
ClearPageUptodate(page);
}
unlock_page(page);
}
index = pvec.pages[nr_pages - 1]->index + 1;
pagevec_release(&pvec);
}
}
static void ext4_print_free_blocks(struct inode *inode)
{
struct ext4_sb_info *sbi = EXT4_SB(inode->i_sb);
struct super_block *sb = inode->i_sb;
struct ext4_inode_info *ei = EXT4_I(inode);
ext4_msg(sb, KERN_CRIT, "Total free blocks count %lld",
EXT4_C2B(EXT4_SB(inode->i_sb),
ext4_count_free_clusters(sb)));
ext4_msg(sb, KERN_CRIT, "Free/Dirty block details");
ext4_msg(sb, KERN_CRIT, "free_blocks=%lld",
(long long) EXT4_C2B(EXT4_SB(sb),
percpu_counter_sum(&sbi->s_freeclusters_counter)));
ext4_msg(sb, KERN_CRIT, "dirty_blocks=%lld",
(long long) EXT4_C2B(EXT4_SB(sb),
percpu_counter_sum(&sbi->s_dirtyclusters_counter)));
ext4_msg(sb, KERN_CRIT, "Block reservation details");
ext4_msg(sb, KERN_CRIT, "i_reserved_data_blocks=%u",
ei->i_reserved_data_blocks);
return;
}
static int ext4_bh_delay_or_unwritten(handle_t *handle, struct buffer_head *bh)
{
return (buffer_delay(bh) || buffer_unwritten(bh)) && buffer_dirty(bh);
}
ext4: attempt to fix race in bigalloc code path Currently, there exists a race between delayed allocated writes and the writeback when bigalloc feature is in use. The race was because we wanted to determine what blocks in a cluster are under delayed allocation and we were using buffer_delayed(bh) check for it. But, the writeback codepath clears this bit without any synchronization which resulted in a race and an ext4 warning similar to: EXT4-fs (ram1): ext4_da_update_reserve_space: ino 13, used 1 with only 0 reserved data blocks The race existed in two places. (1) between ext4_find_delalloc_range() and ext4_map_blocks() when called from writeback code path. (2) between ext4_find_delalloc_range() and ext4_da_get_block_prep() (where buffer_delayed(bh) is set. To fix (1), this patch introduces a new buffer_head state bit - BH_Da_Mapped. This bit is set under the protection of EXT4_I(inode)->i_data_sem when we have actually mapped the delayed allocated blocks during the writeout time. We can now reliably check for this bit inside ext4_find_delalloc_range() to determine whether the reservation for the blocks have already been claimed or not. To fix (2), it was necessary to set buffer_delay(bh) under the protection of i_data_sem. So, I extracted the very beginning of ext4_map_blocks into a new function - ext4_da_map_blocks() - and performed the required setting of bh_delay bit and the quota reservation under the protection of i_data_sem. These two fixes makes the checking of buffer_delay(bh) and buffer_da_mapped(bh) consistent, thus removing the race. Tested: I was able to reproduce the problem by running 'dd' and 'fsync' in parallel. Also, xfstests sometimes used to reproduce this race. After the fix both my test and xfstests were successful and no race (warning message) was observed. Google-Bug-Id: 4997027 Signed-off-by: Aditya Kali <adityakali@google.com> Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
2011-09-09 23:20:51 +00:00
/*
* This function is grabs code from the very beginning of
* ext4_map_blocks, but assumes that the caller is from delayed write
* time. This function looks up the requested blocks and sets the
* buffer delay bit under the protection of i_data_sem.
*/
static int ext4_da_map_blocks(struct inode *inode, sector_t iblock,
struct ext4_map_blocks *map,
struct buffer_head *bh)
{
struct extent_status es;
ext4: attempt to fix race in bigalloc code path Currently, there exists a race between delayed allocated writes and the writeback when bigalloc feature is in use. The race was because we wanted to determine what blocks in a cluster are under delayed allocation and we were using buffer_delayed(bh) check for it. But, the writeback codepath clears this bit without any synchronization which resulted in a race and an ext4 warning similar to: EXT4-fs (ram1): ext4_da_update_reserve_space: ino 13, used 1 with only 0 reserved data blocks The race existed in two places. (1) between ext4_find_delalloc_range() and ext4_map_blocks() when called from writeback code path. (2) between ext4_find_delalloc_range() and ext4_da_get_block_prep() (where buffer_delayed(bh) is set. To fix (1), this patch introduces a new buffer_head state bit - BH_Da_Mapped. This bit is set under the protection of EXT4_I(inode)->i_data_sem when we have actually mapped the delayed allocated blocks during the writeout time. We can now reliably check for this bit inside ext4_find_delalloc_range() to determine whether the reservation for the blocks have already been claimed or not. To fix (2), it was necessary to set buffer_delay(bh) under the protection of i_data_sem. So, I extracted the very beginning of ext4_map_blocks into a new function - ext4_da_map_blocks() - and performed the required setting of bh_delay bit and the quota reservation under the protection of i_data_sem. These two fixes makes the checking of buffer_delay(bh) and buffer_da_mapped(bh) consistent, thus removing the race. Tested: I was able to reproduce the problem by running 'dd' and 'fsync' in parallel. Also, xfstests sometimes used to reproduce this race. After the fix both my test and xfstests were successful and no race (warning message) was observed. Google-Bug-Id: 4997027 Signed-off-by: Aditya Kali <adityakali@google.com> Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
2011-09-09 23:20:51 +00:00
int retval;
sector_t invalid_block = ~((sector_t) 0xffff);
#ifdef ES_AGGRESSIVE_TEST
struct ext4_map_blocks orig_map;
memcpy(&orig_map, map, sizeof(*map));
#endif
ext4: attempt to fix race in bigalloc code path Currently, there exists a race between delayed allocated writes and the writeback when bigalloc feature is in use. The race was because we wanted to determine what blocks in a cluster are under delayed allocation and we were using buffer_delayed(bh) check for it. But, the writeback codepath clears this bit without any synchronization which resulted in a race and an ext4 warning similar to: EXT4-fs (ram1): ext4_da_update_reserve_space: ino 13, used 1 with only 0 reserved data blocks The race existed in two places. (1) between ext4_find_delalloc_range() and ext4_map_blocks() when called from writeback code path. (2) between ext4_find_delalloc_range() and ext4_da_get_block_prep() (where buffer_delayed(bh) is set. To fix (1), this patch introduces a new buffer_head state bit - BH_Da_Mapped. This bit is set under the protection of EXT4_I(inode)->i_data_sem when we have actually mapped the delayed allocated blocks during the writeout time. We can now reliably check for this bit inside ext4_find_delalloc_range() to determine whether the reservation for the blocks have already been claimed or not. To fix (2), it was necessary to set buffer_delay(bh) under the protection of i_data_sem. So, I extracted the very beginning of ext4_map_blocks into a new function - ext4_da_map_blocks() - and performed the required setting of bh_delay bit and the quota reservation under the protection of i_data_sem. These two fixes makes the checking of buffer_delay(bh) and buffer_da_mapped(bh) consistent, thus removing the race. Tested: I was able to reproduce the problem by running 'dd' and 'fsync' in parallel. Also, xfstests sometimes used to reproduce this race. After the fix both my test and xfstests were successful and no race (warning message) was observed. Google-Bug-Id: 4997027 Signed-off-by: Aditya Kali <adityakali@google.com> Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
2011-09-09 23:20:51 +00:00
if (invalid_block < ext4_blocks_count(EXT4_SB(inode->i_sb)->s_es))
invalid_block = ~0;
map->m_flags = 0;
ext_debug("ext4_da_map_blocks(): inode %lu, max_blocks %u,"
"logical block %lu\n", inode->i_ino, map->m_len,
(unsigned long) map->m_lblk);
/* Lookup extent status tree firstly */
if (ext4_es_lookup_extent(inode, iblock, &es)) {
ext4_es_lru_add(inode);
if (ext4_es_is_hole(&es)) {
retval = 0;
down_read(&EXT4_I(inode)->i_data_sem);
goto add_delayed;
}
/*
* Delayed extent could be allocated by fallocate.
* So we need to check it.
*/
if (ext4_es_is_delayed(&es) && !ext4_es_is_unwritten(&es)) {
map_bh(bh, inode->i_sb, invalid_block);
set_buffer_new(bh);
set_buffer_delay(bh);
return 0;
}
map->m_pblk = ext4_es_pblock(&es) + iblock - es.es_lblk;
retval = es.es_len - (iblock - es.es_lblk);
if (retval > map->m_len)
retval = map->m_len;
map->m_len = retval;
if (ext4_es_is_written(&es))
map->m_flags |= EXT4_MAP_MAPPED;
else if (ext4_es_is_unwritten(&es))
map->m_flags |= EXT4_MAP_UNWRITTEN;
else
BUG_ON(1);
#ifdef ES_AGGRESSIVE_TEST
ext4_map_blocks_es_recheck(NULL, inode, map, &orig_map, 0);
#endif
return retval;
}
ext4: attempt to fix race in bigalloc code path Currently, there exists a race between delayed allocated writes and the writeback when bigalloc feature is in use. The race was because we wanted to determine what blocks in a cluster are under delayed allocation and we were using buffer_delayed(bh) check for it. But, the writeback codepath clears this bit without any synchronization which resulted in a race and an ext4 warning similar to: EXT4-fs (ram1): ext4_da_update_reserve_space: ino 13, used 1 with only 0 reserved data blocks The race existed in two places. (1) between ext4_find_delalloc_range() and ext4_map_blocks() when called from writeback code path. (2) between ext4_find_delalloc_range() and ext4_da_get_block_prep() (where buffer_delayed(bh) is set. To fix (1), this patch introduces a new buffer_head state bit - BH_Da_Mapped. This bit is set under the protection of EXT4_I(inode)->i_data_sem when we have actually mapped the delayed allocated blocks during the writeout time. We can now reliably check for this bit inside ext4_find_delalloc_range() to determine whether the reservation for the blocks have already been claimed or not. To fix (2), it was necessary to set buffer_delay(bh) under the protection of i_data_sem. So, I extracted the very beginning of ext4_map_blocks into a new function - ext4_da_map_blocks() - and performed the required setting of bh_delay bit and the quota reservation under the protection of i_data_sem. These two fixes makes the checking of buffer_delay(bh) and buffer_da_mapped(bh) consistent, thus removing the race. Tested: I was able to reproduce the problem by running 'dd' and 'fsync' in parallel. Also, xfstests sometimes used to reproduce this race. After the fix both my test and xfstests were successful and no race (warning message) was observed. Google-Bug-Id: 4997027 Signed-off-by: Aditya Kali <adityakali@google.com> Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
2011-09-09 23:20:51 +00:00
/*
* Try to see if we can get the block without requesting a new
* file system block.
*/
down_read(&EXT4_I(inode)->i_data_sem);
if (ext4_has_inline_data(inode)) {
/*
* We will soon create blocks for this page, and let
* us pretend as if the blocks aren't allocated yet.
* In case of clusters, we have to handle the work
* of mapping from cluster so that the reserved space
* is calculated properly.
*/
if ((EXT4_SB(inode->i_sb)->s_cluster_ratio > 1) &&
ext4_find_delalloc_cluster(inode, map->m_lblk))
map->m_flags |= EXT4_MAP_FROM_CLUSTER;
retval = 0;
} else if (ext4_test_inode_flag(inode, EXT4_INODE_EXTENTS))
retval = ext4_ext_map_blocks(NULL, inode, map,
EXT4_GET_BLOCKS_NO_PUT_HOLE);
ext4: attempt to fix race in bigalloc code path Currently, there exists a race between delayed allocated writes and the writeback when bigalloc feature is in use. The race was because we wanted to determine what blocks in a cluster are under delayed allocation and we were using buffer_delayed(bh) check for it. But, the writeback codepath clears this bit without any synchronization which resulted in a race and an ext4 warning similar to: EXT4-fs (ram1): ext4_da_update_reserve_space: ino 13, used 1 with only 0 reserved data blocks The race existed in two places. (1) between ext4_find_delalloc_range() and ext4_map_blocks() when called from writeback code path. (2) between ext4_find_delalloc_range() and ext4_da_get_block_prep() (where buffer_delayed(bh) is set. To fix (1), this patch introduces a new buffer_head state bit - BH_Da_Mapped. This bit is set under the protection of EXT4_I(inode)->i_data_sem when we have actually mapped the delayed allocated blocks during the writeout time. We can now reliably check for this bit inside ext4_find_delalloc_range() to determine whether the reservation for the blocks have already been claimed or not. To fix (2), it was necessary to set buffer_delay(bh) under the protection of i_data_sem. So, I extracted the very beginning of ext4_map_blocks into a new function - ext4_da_map_blocks() - and performed the required setting of bh_delay bit and the quota reservation under the protection of i_data_sem. These two fixes makes the checking of buffer_delay(bh) and buffer_da_mapped(bh) consistent, thus removing the race. Tested: I was able to reproduce the problem by running 'dd' and 'fsync' in parallel. Also, xfstests sometimes used to reproduce this race. After the fix both my test and xfstests were successful and no race (warning message) was observed. Google-Bug-Id: 4997027 Signed-off-by: Aditya Kali <adityakali@google.com> Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
2011-09-09 23:20:51 +00:00
else
retval = ext4_ind_map_blocks(NULL, inode, map,
EXT4_GET_BLOCKS_NO_PUT_HOLE);
ext4: attempt to fix race in bigalloc code path Currently, there exists a race between delayed allocated writes and the writeback when bigalloc feature is in use. The race was because we wanted to determine what blocks in a cluster are under delayed allocation and we were using buffer_delayed(bh) check for it. But, the writeback codepath clears this bit without any synchronization which resulted in a race and an ext4 warning similar to: EXT4-fs (ram1): ext4_da_update_reserve_space: ino 13, used 1 with only 0 reserved data blocks The race existed in two places. (1) between ext4_find_delalloc_range() and ext4_map_blocks() when called from writeback code path. (2) between ext4_find_delalloc_range() and ext4_da_get_block_prep() (where buffer_delayed(bh) is set. To fix (1), this patch introduces a new buffer_head state bit - BH_Da_Mapped. This bit is set under the protection of EXT4_I(inode)->i_data_sem when we have actually mapped the delayed allocated blocks during the writeout time. We can now reliably check for this bit inside ext4_find_delalloc_range() to determine whether the reservation for the blocks have already been claimed or not. To fix (2), it was necessary to set buffer_delay(bh) under the protection of i_data_sem. So, I extracted the very beginning of ext4_map_blocks into a new function - ext4_da_map_blocks() - and performed the required setting of bh_delay bit and the quota reservation under the protection of i_data_sem. These two fixes makes the checking of buffer_delay(bh) and buffer_da_mapped(bh) consistent, thus removing the race. Tested: I was able to reproduce the problem by running 'dd' and 'fsync' in parallel. Also, xfstests sometimes used to reproduce this race. After the fix both my test and xfstests were successful and no race (warning message) was observed. Google-Bug-Id: 4997027 Signed-off-by: Aditya Kali <adityakali@google.com> Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
2011-09-09 23:20:51 +00:00
add_delayed:
ext4: attempt to fix race in bigalloc code path Currently, there exists a race between delayed allocated writes and the writeback when bigalloc feature is in use. The race was because we wanted to determine what blocks in a cluster are under delayed allocation and we were using buffer_delayed(bh) check for it. But, the writeback codepath clears this bit without any synchronization which resulted in a race and an ext4 warning similar to: EXT4-fs (ram1): ext4_da_update_reserve_space: ino 13, used 1 with only 0 reserved data blocks The race existed in two places. (1) between ext4_find_delalloc_range() and ext4_map_blocks() when called from writeback code path. (2) between ext4_find_delalloc_range() and ext4_da_get_block_prep() (where buffer_delayed(bh) is set. To fix (1), this patch introduces a new buffer_head state bit - BH_Da_Mapped. This bit is set under the protection of EXT4_I(inode)->i_data_sem when we have actually mapped the delayed allocated blocks during the writeout time. We can now reliably check for this bit inside ext4_find_delalloc_range() to determine whether the reservation for the blocks have already been claimed or not. To fix (2), it was necessary to set buffer_delay(bh) under the protection of i_data_sem. So, I extracted the very beginning of ext4_map_blocks into a new function - ext4_da_map_blocks() - and performed the required setting of bh_delay bit and the quota reservation under the protection of i_data_sem. These two fixes makes the checking of buffer_delay(bh) and buffer_da_mapped(bh) consistent, thus removing the race. Tested: I was able to reproduce the problem by running 'dd' and 'fsync' in parallel. Also, xfstests sometimes used to reproduce this race. After the fix both my test and xfstests were successful and no race (warning message) was observed. Google-Bug-Id: 4997027 Signed-off-by: Aditya Kali <adityakali@google.com> Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
2011-09-09 23:20:51 +00:00
if (retval == 0) {
int ret;
ext4: attempt to fix race in bigalloc code path Currently, there exists a race between delayed allocated writes and the writeback when bigalloc feature is in use. The race was because we wanted to determine what blocks in a cluster are under delayed allocation and we were using buffer_delayed(bh) check for it. But, the writeback codepath clears this bit without any synchronization which resulted in a race and an ext4 warning similar to: EXT4-fs (ram1): ext4_da_update_reserve_space: ino 13, used 1 with only 0 reserved data blocks The race existed in two places. (1) between ext4_find_delalloc_range() and ext4_map_blocks() when called from writeback code path. (2) between ext4_find_delalloc_range() and ext4_da_get_block_prep() (where buffer_delayed(bh) is set. To fix (1), this patch introduces a new buffer_head state bit - BH_Da_Mapped. This bit is set under the protection of EXT4_I(inode)->i_data_sem when we have actually mapped the delayed allocated blocks during the writeout time. We can now reliably check for this bit inside ext4_find_delalloc_range() to determine whether the reservation for the blocks have already been claimed or not. To fix (2), it was necessary to set buffer_delay(bh) under the protection of i_data_sem. So, I extracted the very beginning of ext4_map_blocks into a new function - ext4_da_map_blocks() - and performed the required setting of bh_delay bit and the quota reservation under the protection of i_data_sem. These two fixes makes the checking of buffer_delay(bh) and buffer_da_mapped(bh) consistent, thus removing the race. Tested: I was able to reproduce the problem by running 'dd' and 'fsync' in parallel. Also, xfstests sometimes used to reproduce this race. After the fix both my test and xfstests were successful and no race (warning message) was observed. Google-Bug-Id: 4997027 Signed-off-by: Aditya Kali <adityakali@google.com> Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
2011-09-09 23:20:51 +00:00
/*
* XXX: __block_prepare_write() unmaps passed block,
* is it OK?
*/
/*
* If the block was allocated from previously allocated cluster,
* then we don't need to reserve it again. However we still need
* to reserve metadata for every block we're going to write.
*/
ext4: attempt to fix race in bigalloc code path Currently, there exists a race between delayed allocated writes and the writeback when bigalloc feature is in use. The race was because we wanted to determine what blocks in a cluster are under delayed allocation and we were using buffer_delayed(bh) check for it. But, the writeback codepath clears this bit without any synchronization which resulted in a race and an ext4 warning similar to: EXT4-fs (ram1): ext4_da_update_reserve_space: ino 13, used 1 with only 0 reserved data blocks The race existed in two places. (1) between ext4_find_delalloc_range() and ext4_map_blocks() when called from writeback code path. (2) between ext4_find_delalloc_range() and ext4_da_get_block_prep() (where buffer_delayed(bh) is set. To fix (1), this patch introduces a new buffer_head state bit - BH_Da_Mapped. This bit is set under the protection of EXT4_I(inode)->i_data_sem when we have actually mapped the delayed allocated blocks during the writeout time. We can now reliably check for this bit inside ext4_find_delalloc_range() to determine whether the reservation for the blocks have already been claimed or not. To fix (2), it was necessary to set buffer_delay(bh) under the protection of i_data_sem. So, I extracted the very beginning of ext4_map_blocks into a new function - ext4_da_map_blocks() - and performed the required setting of bh_delay bit and the quota reservation under the protection of i_data_sem. These two fixes makes the checking of buffer_delay(bh) and buffer_da_mapped(bh) consistent, thus removing the race. Tested: I was able to reproduce the problem by running 'dd' and 'fsync' in parallel. Also, xfstests sometimes used to reproduce this race. After the fix both my test and xfstests were successful and no race (warning message) was observed. Google-Bug-Id: 4997027 Signed-off-by: Aditya Kali <adityakali@google.com> Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
2011-09-09 23:20:51 +00:00
if (!(map->m_flags & EXT4_MAP_FROM_CLUSTER)) {
ret = ext4_da_reserve_space(inode, iblock);
if (ret) {
ext4: attempt to fix race in bigalloc code path Currently, there exists a race between delayed allocated writes and the writeback when bigalloc feature is in use. The race was because we wanted to determine what blocks in a cluster are under delayed allocation and we were using buffer_delayed(bh) check for it. But, the writeback codepath clears this bit without any synchronization which resulted in a race and an ext4 warning similar to: EXT4-fs (ram1): ext4_da_update_reserve_space: ino 13, used 1 with only 0 reserved data blocks The race existed in two places. (1) between ext4_find_delalloc_range() and ext4_map_blocks() when called from writeback code path. (2) between ext4_find_delalloc_range() and ext4_da_get_block_prep() (where buffer_delayed(bh) is set. To fix (1), this patch introduces a new buffer_head state bit - BH_Da_Mapped. This bit is set under the protection of EXT4_I(inode)->i_data_sem when we have actually mapped the delayed allocated blocks during the writeout time. We can now reliably check for this bit inside ext4_find_delalloc_range() to determine whether the reservation for the blocks have already been claimed or not. To fix (2), it was necessary to set buffer_delay(bh) under the protection of i_data_sem. So, I extracted the very beginning of ext4_map_blocks into a new function - ext4_da_map_blocks() - and performed the required setting of bh_delay bit and the quota reservation under the protection of i_data_sem. These two fixes makes the checking of buffer_delay(bh) and buffer_da_mapped(bh) consistent, thus removing the race. Tested: I was able to reproduce the problem by running 'dd' and 'fsync' in parallel. Also, xfstests sometimes used to reproduce this race. After the fix both my test and xfstests were successful and no race (warning message) was observed. Google-Bug-Id: 4997027 Signed-off-by: Aditya Kali <adityakali@google.com> Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
2011-09-09 23:20:51 +00:00
/* not enough space to reserve */
retval = ret;
ext4: attempt to fix race in bigalloc code path Currently, there exists a race between delayed allocated writes and the writeback when bigalloc feature is in use. The race was because we wanted to determine what blocks in a cluster are under delayed allocation and we were using buffer_delayed(bh) check for it. But, the writeback codepath clears this bit without any synchronization which resulted in a race and an ext4 warning similar to: EXT4-fs (ram1): ext4_da_update_reserve_space: ino 13, used 1 with only 0 reserved data blocks The race existed in two places. (1) between ext4_find_delalloc_range() and ext4_map_blocks() when called from writeback code path. (2) between ext4_find_delalloc_range() and ext4_da_get_block_prep() (where buffer_delayed(bh) is set. To fix (1), this patch introduces a new buffer_head state bit - BH_Da_Mapped. This bit is set under the protection of EXT4_I(inode)->i_data_sem when we have actually mapped the delayed allocated blocks during the writeout time. We can now reliably check for this bit inside ext4_find_delalloc_range() to determine whether the reservation for the blocks have already been claimed or not. To fix (2), it was necessary to set buffer_delay(bh) under the protection of i_data_sem. So, I extracted the very beginning of ext4_map_blocks into a new function - ext4_da_map_blocks() - and performed the required setting of bh_delay bit and the quota reservation under the protection of i_data_sem. These two fixes makes the checking of buffer_delay(bh) and buffer_da_mapped(bh) consistent, thus removing the race. Tested: I was able to reproduce the problem by running 'dd' and 'fsync' in parallel. Also, xfstests sometimes used to reproduce this race. After the fix both my test and xfstests were successful and no race (warning message) was observed. Google-Bug-Id: 4997027 Signed-off-by: Aditya Kali <adityakali@google.com> Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
2011-09-09 23:20:51 +00:00
goto out_unlock;
}
ext4: attempt to fix race in bigalloc code path Currently, there exists a race between delayed allocated writes and the writeback when bigalloc feature is in use. The race was because we wanted to determine what blocks in a cluster are under delayed allocation and we were using buffer_delayed(bh) check for it. But, the writeback codepath clears this bit without any synchronization which resulted in a race and an ext4 warning similar to: EXT4-fs (ram1): ext4_da_update_reserve_space: ino 13, used 1 with only 0 reserved data blocks The race existed in two places. (1) between ext4_find_delalloc_range() and ext4_map_blocks() when called from writeback code path. (2) between ext4_find_delalloc_range() and ext4_da_get_block_prep() (where buffer_delayed(bh) is set. To fix (1), this patch introduces a new buffer_head state bit - BH_Da_Mapped. This bit is set under the protection of EXT4_I(inode)->i_data_sem when we have actually mapped the delayed allocated blocks during the writeout time. We can now reliably check for this bit inside ext4_find_delalloc_range() to determine whether the reservation for the blocks have already been claimed or not. To fix (2), it was necessary to set buffer_delay(bh) under the protection of i_data_sem. So, I extracted the very beginning of ext4_map_blocks into a new function - ext4_da_map_blocks() - and performed the required setting of bh_delay bit and the quota reservation under the protection of i_data_sem. These two fixes makes the checking of buffer_delay(bh) and buffer_da_mapped(bh) consistent, thus removing the race. Tested: I was able to reproduce the problem by running 'dd' and 'fsync' in parallel. Also, xfstests sometimes used to reproduce this race. After the fix both my test and xfstests were successful and no race (warning message) was observed. Google-Bug-Id: 4997027 Signed-off-by: Aditya Kali <adityakali@google.com> Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
2011-09-09 23:20:51 +00:00
}
ret = ext4_es_insert_extent(inode, map->m_lblk, map->m_len,
~0, EXTENT_STATUS_DELAYED);
if (ret) {
retval = ret;
goto out_unlock;
}
ext4: attempt to fix race in bigalloc code path Currently, there exists a race between delayed allocated writes and the writeback when bigalloc feature is in use. The race was because we wanted to determine what blocks in a cluster are under delayed allocation and we were using buffer_delayed(bh) check for it. But, the writeback codepath clears this bit without any synchronization which resulted in a race and an ext4 warning similar to: EXT4-fs (ram1): ext4_da_update_reserve_space: ino 13, used 1 with only 0 reserved data blocks The race existed in two places. (1) between ext4_find_delalloc_range() and ext4_map_blocks() when called from writeback code path. (2) between ext4_find_delalloc_range() and ext4_da_get_block_prep() (where buffer_delayed(bh) is set. To fix (1), this patch introduces a new buffer_head state bit - BH_Da_Mapped. This bit is set under the protection of EXT4_I(inode)->i_data_sem when we have actually mapped the delayed allocated blocks during the writeout time. We can now reliably check for this bit inside ext4_find_delalloc_range() to determine whether the reservation for the blocks have already been claimed or not. To fix (2), it was necessary to set buffer_delay(bh) under the protection of i_data_sem. So, I extracted the very beginning of ext4_map_blocks into a new function - ext4_da_map_blocks() - and performed the required setting of bh_delay bit and the quota reservation under the protection of i_data_sem. These two fixes makes the checking of buffer_delay(bh) and buffer_da_mapped(bh) consistent, thus removing the race. Tested: I was able to reproduce the problem by running 'dd' and 'fsync' in parallel. Also, xfstests sometimes used to reproduce this race. After the fix both my test and xfstests were successful and no race (warning message) was observed. Google-Bug-Id: 4997027 Signed-off-by: Aditya Kali <adityakali@google.com> Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
2011-09-09 23:20:51 +00:00
/* Clear EXT4_MAP_FROM_CLUSTER flag since its purpose is served
* and it should not appear on the bh->b_state.
*/
map->m_flags &= ~EXT4_MAP_FROM_CLUSTER;
map_bh(bh, inode->i_sb, invalid_block);
set_buffer_new(bh);
set_buffer_delay(bh);
} else if (retval > 0) {
int ret;
unsigned int status;
if (unlikely(retval != map->m_len)) {
ext4_warning(inode->i_sb,
"ES len assertion failed for inode "
"%lu: retval %d != map->m_len %d",
inode->i_ino, retval, map->m_len);
WARN_ON(1);
}
status = map->m_flags & EXT4_MAP_UNWRITTEN ?
EXTENT_STATUS_UNWRITTEN : EXTENT_STATUS_WRITTEN;
ret = ext4_es_insert_extent(inode, map->m_lblk, map->m_len,
map->m_pblk, status);
if (ret != 0)
retval = ret;
ext4: attempt to fix race in bigalloc code path Currently, there exists a race between delayed allocated writes and the writeback when bigalloc feature is in use. The race was because we wanted to determine what blocks in a cluster are under delayed allocation and we were using buffer_delayed(bh) check for it. But, the writeback codepath clears this bit without any synchronization which resulted in a race and an ext4 warning similar to: EXT4-fs (ram1): ext4_da_update_reserve_space: ino 13, used 1 with only 0 reserved data blocks The race existed in two places. (1) between ext4_find_delalloc_range() and ext4_map_blocks() when called from writeback code path. (2) between ext4_find_delalloc_range() and ext4_da_get_block_prep() (where buffer_delayed(bh) is set. To fix (1), this patch introduces a new buffer_head state bit - BH_Da_Mapped. This bit is set under the protection of EXT4_I(inode)->i_data_sem when we have actually mapped the delayed allocated blocks during the writeout time. We can now reliably check for this bit inside ext4_find_delalloc_range() to determine whether the reservation for the blocks have already been claimed or not. To fix (2), it was necessary to set buffer_delay(bh) under the protection of i_data_sem. So, I extracted the very beginning of ext4_map_blocks into a new function - ext4_da_map_blocks() - and performed the required setting of bh_delay bit and the quota reservation under the protection of i_data_sem. These two fixes makes the checking of buffer_delay(bh) and buffer_da_mapped(bh) consistent, thus removing the race. Tested: I was able to reproduce the problem by running 'dd' and 'fsync' in parallel. Also, xfstests sometimes used to reproduce this race. After the fix both my test and xfstests were successful and no race (warning message) was observed. Google-Bug-Id: 4997027 Signed-off-by: Aditya Kali <adityakali@google.com> Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
2011-09-09 23:20:51 +00:00
}
out_unlock:
up_read((&EXT4_I(inode)->i_data_sem));
return retval;
}
/*
* This is a special get_blocks_t callback which is used by
* ext4_da_write_begin(). It will either return mapped block or
* reserve space for a single block.
*
* For delayed buffer_head we have BH_Mapped, BH_New, BH_Delay set.
* We also have b_blocknr = -1 and b_bdev initialized properly
*
* For unwritten buffer_head we have BH_Mapped, BH_New, BH_Unwritten set.
* We also have b_blocknr = physicalblock mapping unwritten extent and b_bdev
* initialized properly.
*/
int ext4_da_get_block_prep(struct inode *inode, sector_t iblock,
struct buffer_head *bh, int create)
{
struct ext4_map_blocks map;
int ret = 0;
BUG_ON(create == 0);
BUG_ON(bh->b_size != inode->i_sb->s_blocksize);
map.m_lblk = iblock;
map.m_len = 1;
/*
* first, we need to know whether the block is allocated already
* preallocated blocks are unmapped but should treated
* the same as allocated blocks.
*/
ext4: attempt to fix race in bigalloc code path Currently, there exists a race between delayed allocated writes and the writeback when bigalloc feature is in use. The race was because we wanted to determine what blocks in a cluster are under delayed allocation and we were using buffer_delayed(bh) check for it. But, the writeback codepath clears this bit without any synchronization which resulted in a race and an ext4 warning similar to: EXT4-fs (ram1): ext4_da_update_reserve_space: ino 13, used 1 with only 0 reserved data blocks The race existed in two places. (1) between ext4_find_delalloc_range() and ext4_map_blocks() when called from writeback code path. (2) between ext4_find_delalloc_range() and ext4_da_get_block_prep() (where buffer_delayed(bh) is set. To fix (1), this patch introduces a new buffer_head state bit - BH_Da_Mapped. This bit is set under the protection of EXT4_I(inode)->i_data_sem when we have actually mapped the delayed allocated blocks during the writeout time. We can now reliably check for this bit inside ext4_find_delalloc_range() to determine whether the reservation for the blocks have already been claimed or not. To fix (2), it was necessary to set buffer_delay(bh) under the protection of i_data_sem. So, I extracted the very beginning of ext4_map_blocks into a new function - ext4_da_map_blocks() - and performed the required setting of bh_delay bit and the quota reservation under the protection of i_data_sem. These two fixes makes the checking of buffer_delay(bh) and buffer_da_mapped(bh) consistent, thus removing the race. Tested: I was able to reproduce the problem by running 'dd' and 'fsync' in parallel. Also, xfstests sometimes used to reproduce this race. After the fix both my test and xfstests were successful and no race (warning message) was observed. Google-Bug-Id: 4997027 Signed-off-by: Aditya Kali <adityakali@google.com> Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
2011-09-09 23:20:51 +00:00
ret = ext4_da_map_blocks(inode, iblock, &map, bh);
if (ret <= 0)
return ret;
map_bh(bh, inode->i_sb, map.m_pblk);
bh->b_state = (bh->b_state & ~EXT4_MAP_FLAGS) | map.m_flags;
if (buffer_unwritten(bh)) {
/* A delayed write to unwritten bh should be marked
* new and mapped. Mapped ensures that we don't do
* get_block multiple times when we write to the same
* offset and new ensures that we do proper zero out
* for partial write.
*/
set_buffer_new(bh);
set_buffer_mapped(bh);
}
return 0;
}
static int bget_one(handle_t *handle, struct buffer_head *bh)
{
get_bh(bh);
return 0;
}
static int bput_one(handle_t *handle, struct buffer_head *bh)
{
put_bh(bh);
return 0;
}
static int __ext4_journalled_writepage(struct page *page,
unsigned int len)
{
struct address_space *mapping = page->mapping;
struct inode *inode = mapping->host;
struct buffer_head *page_bufs = NULL;
handle_t *handle = NULL;
int ret = 0, err = 0;
int inline_data = ext4_has_inline_data(inode);
struct buffer_head *inode_bh = NULL;
ClearPageChecked(page);
if (inline_data) {
BUG_ON(page->index != 0);
BUG_ON(len > ext4_get_max_inline_size(inode));
inode_bh = ext4_journalled_write_inline_data(inode, len, page);
if (inode_bh == NULL)
goto out;
} else {
page_bufs = page_buffers(page);
if (!page_bufs) {
BUG();
goto out;
}
ext4_walk_page_buffers(handle, page_bufs, 0, len,
NULL, bget_one);
}
/* As soon as we unlock the page, it can go away, but we have
* references to buffers so we are safe */
unlock_page(page);
handle = ext4_journal_start(inode, EXT4_HT_WRITE_PAGE,
ext4_writepage_trans_blocks(inode));
if (IS_ERR(handle)) {
ret = PTR_ERR(handle);
goto out;
}
BUG_ON(!ext4_handle_valid(handle));
if (inline_data) {
BUFFER_TRACE(inode_bh, "get write access");
ret = ext4_journal_get_write_access(handle, inode_bh);
err = ext4_handle_dirty_metadata(handle, inode, inode_bh);
} else {
ret = ext4_walk_page_buffers(handle, page_bufs, 0, len, NULL,
do_journal_get_write_access);
err = ext4_walk_page_buffers(handle, page_bufs, 0, len, NULL,
write_end_fn);
}
if (ret == 0)
ret = err;
EXT4_I(inode)->i_datasync_tid = handle->h_transaction->t_tid;
err = ext4_journal_stop(handle);
if (!ret)
ret = err;
if (!ext4_has_inline_data(inode))
ext4_walk_page_buffers(NULL, page_bufs, 0, len,
NULL, bput_one);
ext4_set_inode_state(inode, EXT4_STATE_JDATA);
out:
brelse(inode_bh);
return ret;
}
/*
* Note that we don't need to start a transaction unless we're journaling data
* because we should have holes filled from ext4_page_mkwrite(). We even don't
* need to file the inode to the transaction's list in ordered mode because if
* we are writing back data added by write(), the inode is already there and if
* we are writing back data modified via mmap(), no one guarantees in which
* transaction the data will hit the disk. In case we are journaling data, we
* cannot start transaction directly because transaction start ranks above page
* lock so we have to do some magic.
*
* This function can get called via...
* - ext4_writepages after taking page lock (have journal handle)
* - journal_submit_inode_data_buffers (no journal handle)
* - shrink_page_list via the kswapd/direct reclaim (no journal handle)
* - grab_page_cache when doing write_begin (have journal handle)
*
* We don't do any block allocation in this function. If we have page with
* multiple blocks we need to write those buffer_heads that are mapped. This
* is important for mmaped based write. So if we do with blocksize 1K
* truncate(f, 1024);
* a = mmap(f, 0, 4096);
* a[0] = 'a';
* truncate(f, 4096);
* we have in the page first buffer_head mapped via page_mkwrite call back
* but other buffer_heads would be unmapped but dirty (dirty done via the
* do_wp_page). So writepage should write the first block. If we modify
* the mmap area beyond 1024 we will again get a page_fault and the
* page_mkwrite callback will do the block allocation and mark the
* buffer_heads mapped.
*
* We redirty the page if we have any buffer_heads that is either delay or
* unwritten in the page.
*
* We can get recursively called as show below.
*
* ext4_writepage() -> kmalloc() -> __alloc_pages() -> page_launder() ->
* ext4_writepage()
*
* But since we don't do any block allocation we should not deadlock.
* Page also have the dirty flag cleared so we don't get recurive page_lock.
*/
static int ext4_writepage(struct page *page,
struct writeback_control *wbc)
{
int ret = 0;
loff_t size;
unsigned int len;
struct buffer_head *page_bufs = NULL;
struct inode *inode = page->mapping->host;
struct ext4_io_submit io_submit;
ext4: fix data integrity sync in ordered mode When we perform a data integrity sync we tag all the dirty pages with PAGECACHE_TAG_TOWRITE at start of ext4_da_writepages. Later we check for this tag in write_cache_pages_da and creates a struct mpage_da_data containing contiguously indexed pages tagged with this tag and sync these pages with a call to mpage_da_map_and_submit. This process is done in while loop until all the PAGECACHE_TAG_TOWRITE pages are synced. We also do journal start and stop in each iteration. journal_stop could initiate journal commit which would call ext4_writepage which in turn will call ext4_bio_write_page even for delayed OR unwritten buffers. When ext4_bio_write_page is called for such buffers, even though it does not sync them but it clears the PAGECACHE_TAG_TOWRITE of the corresponding page and hence these pages are also not synced by the currently running data integrity sync. We will end up with dirty pages although sync is completed. This could cause a potential data loss when the sync call is followed by a truncate_pagecache call, which is exactly the case in collapse_range. (It will cause generic/127 failure in xfstests) To avoid this issue, we can use set_page_writeback_keepwrite instead of set_page_writeback, which doesn't clear TOWRITE tag. Cc: stable@vger.kernel.org Signed-off-by: Namjae Jeon <namjae.jeon@samsung.com> Signed-off-by: Ashish Sangwan <a.sangwan@samsung.com> Signed-off-by: "Theodore Ts'o" <tytso@mit.edu> Reviewed-by: Jan Kara <jack@suse.cz>
2014-05-12 12:12:25 +00:00
bool keep_towrite = false;
trace_ext4_writepage(page);
size = i_size_read(inode);
if (page->index == size >> PAGE_CACHE_SHIFT)
len = size & ~PAGE_CACHE_MASK;
else
len = PAGE_CACHE_SIZE;
page_bufs = page_buffers(page);
/*
* We cannot do block allocation or other extent handling in this
* function. If there are buffers needing that, we have to redirty
* the page. But we may reach here when we do a journal commit via
* journal_submit_inode_data_buffers() and in that case we must write
* allocated buffers to achieve data=ordered mode guarantees.
*/
if (ext4_walk_page_buffers(NULL, page_bufs, 0, len, NULL,
ext4_bh_delay_or_unwritten)) {
redirty_page_for_writepage(wbc, page);
if (current->flags & PF_MEMALLOC) {
/*
* For memory cleaning there's no point in writing only
* some buffers. So just bail out. Warn if we came here
* from direct reclaim.
*/
WARN_ON_ONCE((current->flags & (PF_MEMALLOC|PF_KSWAPD))
== PF_MEMALLOC);
unlock_page(page);
return 0;
}
ext4: fix data integrity sync in ordered mode When we perform a data integrity sync we tag all the dirty pages with PAGECACHE_TAG_TOWRITE at start of ext4_da_writepages. Later we check for this tag in write_cache_pages_da and creates a struct mpage_da_data containing contiguously indexed pages tagged with this tag and sync these pages with a call to mpage_da_map_and_submit. This process is done in while loop until all the PAGECACHE_TAG_TOWRITE pages are synced. We also do journal start and stop in each iteration. journal_stop could initiate journal commit which would call ext4_writepage which in turn will call ext4_bio_write_page even for delayed OR unwritten buffers. When ext4_bio_write_page is called for such buffers, even though it does not sync them but it clears the PAGECACHE_TAG_TOWRITE of the corresponding page and hence these pages are also not synced by the currently running data integrity sync. We will end up with dirty pages although sync is completed. This could cause a potential data loss when the sync call is followed by a truncate_pagecache call, which is exactly the case in collapse_range. (It will cause generic/127 failure in xfstests) To avoid this issue, we can use set_page_writeback_keepwrite instead of set_page_writeback, which doesn't clear TOWRITE tag. Cc: stable@vger.kernel.org Signed-off-by: Namjae Jeon <namjae.jeon@samsung.com> Signed-off-by: Ashish Sangwan <a.sangwan@samsung.com> Signed-off-by: "Theodore Ts'o" <tytso@mit.edu> Reviewed-by: Jan Kara <jack@suse.cz>
2014-05-12 12:12:25 +00:00
keep_towrite = true;
}
if (PageChecked(page) && ext4_should_journal_data(inode))
/*
* It's mmapped pagecache. Add buffers and journal it. There
* doesn't seem much point in redirtying the page here.
*/
return __ext4_journalled_writepage(page, len);
ext4_io_submit_init(&io_submit, wbc);
io_submit.io_end = ext4_init_io_end(inode, GFP_NOFS);
if (!io_submit.io_end) {
redirty_page_for_writepage(wbc, page);
unlock_page(page);
return -ENOMEM;
}
ext4: fix data integrity sync in ordered mode When we perform a data integrity sync we tag all the dirty pages with PAGECACHE_TAG_TOWRITE at start of ext4_da_writepages. Later we check for this tag in write_cache_pages_da and creates a struct mpage_da_data containing contiguously indexed pages tagged with this tag and sync these pages with a call to mpage_da_map_and_submit. This process is done in while loop until all the PAGECACHE_TAG_TOWRITE pages are synced. We also do journal start and stop in each iteration. journal_stop could initiate journal commit which would call ext4_writepage which in turn will call ext4_bio_write_page even for delayed OR unwritten buffers. When ext4_bio_write_page is called for such buffers, even though it does not sync them but it clears the PAGECACHE_TAG_TOWRITE of the corresponding page and hence these pages are also not synced by the currently running data integrity sync. We will end up with dirty pages although sync is completed. This could cause a potential data loss when the sync call is followed by a truncate_pagecache call, which is exactly the case in collapse_range. (It will cause generic/127 failure in xfstests) To avoid this issue, we can use set_page_writeback_keepwrite instead of set_page_writeback, which doesn't clear TOWRITE tag. Cc: stable@vger.kernel.org Signed-off-by: Namjae Jeon <namjae.jeon@samsung.com> Signed-off-by: Ashish Sangwan <a.sangwan@samsung.com> Signed-off-by: "Theodore Ts'o" <tytso@mit.edu> Reviewed-by: Jan Kara <jack@suse.cz>
2014-05-12 12:12:25 +00:00
ret = ext4_bio_write_page(&io_submit, page, len, wbc, keep_towrite);
ext4_io_submit(&io_submit);
/* Drop io_end reference we got from init */
ext4_put_io_end_defer(io_submit.io_end);
return ret;
}
static int mpage_submit_page(struct mpage_da_data *mpd, struct page *page)
{
int len;
loff_t size = i_size_read(mpd->inode);
int err;
BUG_ON(page->index != mpd->first_page);
if (page->index == size >> PAGE_CACHE_SHIFT)
len = size & ~PAGE_CACHE_MASK;
else
len = PAGE_CACHE_SIZE;
clear_page_dirty_for_io(page);
ext4: fix data integrity sync in ordered mode When we perform a data integrity sync we tag all the dirty pages with PAGECACHE_TAG_TOWRITE at start of ext4_da_writepages. Later we check for this tag in write_cache_pages_da and creates a struct mpage_da_data containing contiguously indexed pages tagged with this tag and sync these pages with a call to mpage_da_map_and_submit. This process is done in while loop until all the PAGECACHE_TAG_TOWRITE pages are synced. We also do journal start and stop in each iteration. journal_stop could initiate journal commit which would call ext4_writepage which in turn will call ext4_bio_write_page even for delayed OR unwritten buffers. When ext4_bio_write_page is called for such buffers, even though it does not sync them but it clears the PAGECACHE_TAG_TOWRITE of the corresponding page and hence these pages are also not synced by the currently running data integrity sync. We will end up with dirty pages although sync is completed. This could cause a potential data loss when the sync call is followed by a truncate_pagecache call, which is exactly the case in collapse_range. (It will cause generic/127 failure in xfstests) To avoid this issue, we can use set_page_writeback_keepwrite instead of set_page_writeback, which doesn't clear TOWRITE tag. Cc: stable@vger.kernel.org Signed-off-by: Namjae Jeon <namjae.jeon@samsung.com> Signed-off-by: Ashish Sangwan <a.sangwan@samsung.com> Signed-off-by: "Theodore Ts'o" <tytso@mit.edu> Reviewed-by: Jan Kara <jack@suse.cz>
2014-05-12 12:12:25 +00:00
err = ext4_bio_write_page(&mpd->io_submit, page, len, mpd->wbc, false);
if (!err)
mpd->wbc->nr_to_write--;
mpd->first_page++;
return err;
}
#define BH_FLAGS ((1 << BH_Unwritten) | (1 << BH_Delay))
/*
* mballoc gives us at most this number of blocks...
* XXX: That seems to be only a limitation of ext4_mb_normalize_request().
* The rest of mballoc seems to handle chunks up to full group size.
*/
#define MAX_WRITEPAGES_EXTENT_LEN 2048
/*
* mpage_add_bh_to_extent - try to add bh to extent of blocks to map
*
* @mpd - extent of blocks
* @lblk - logical number of the block in the file
* @bh - buffer head we want to add to the extent
*
* The function is used to collect contig. blocks in the same state. If the
* buffer doesn't require mapping for writeback and we haven't started the
* extent of buffers to map yet, the function returns 'true' immediately - the
* caller can write the buffer right away. Otherwise the function returns true
* if the block has been added to the extent, false if the block couldn't be
* added.
*/
static bool mpage_add_bh_to_extent(struct mpage_da_data *mpd, ext4_lblk_t lblk,
struct buffer_head *bh)
{
struct ext4_map_blocks *map = &mpd->map;
/* Buffer that doesn't need mapping for writeback? */
if (!buffer_dirty(bh) || !buffer_mapped(bh) ||
(!buffer_delay(bh) && !buffer_unwritten(bh))) {
/* So far no extent to map => we write the buffer right away */
if (map->m_len == 0)
return true;
return false;
}
/* First block in the extent? */
if (map->m_len == 0) {
map->m_lblk = lblk;
map->m_len = 1;
map->m_flags = bh->b_state & BH_FLAGS;
return true;
}
/* Don't go larger than mballoc is willing to allocate */
if (map->m_len >= MAX_WRITEPAGES_EXTENT_LEN)
return false;
/* Can we merge the block to our big extent? */
if (lblk == map->m_lblk + map->m_len &&
(bh->b_state & BH_FLAGS) == map->m_flags) {
map->m_len++;
return true;
}
return false;
}
/*
* mpage_process_page_bufs - submit page buffers for IO or add them to extent
*
* @mpd - extent of blocks for mapping
* @head - the first buffer in the page
* @bh - buffer we should start processing from
* @lblk - logical number of the block in the file corresponding to @bh
*
* Walk through page buffers from @bh upto @head (exclusive) and either submit
* the page for IO if all buffers in this page were mapped and there's no
* accumulated extent of buffers to map or add buffers in the page to the
* extent of buffers to map. The function returns 1 if the caller can continue
* by processing the next page, 0 if it should stop adding buffers to the
* extent to map because we cannot extend it anymore. It can also return value
* < 0 in case of error during IO submission.
*/
static int mpage_process_page_bufs(struct mpage_da_data *mpd,
struct buffer_head *head,
struct buffer_head *bh,
ext4_lblk_t lblk)
{
struct inode *inode = mpd->inode;
int err;
ext4_lblk_t blocks = (i_size_read(inode) + (1 << inode->i_blkbits) - 1)
>> inode->i_blkbits;
do {
BUG_ON(buffer_locked(bh));
if (lblk >= blocks || !mpage_add_bh_to_extent(mpd, lblk, bh)) {
/* Found extent to map? */
if (mpd->map.m_len)
return 0;
/* Everything mapped so far and we hit EOF */
break;
}
} while (lblk++, (bh = bh->b_this_page) != head);
/* So far everything mapped? Submit the page for IO. */
if (mpd->map.m_len == 0) {
err = mpage_submit_page(mpd, head->b_page);
if (err < 0)
return err;
}
return lblk < blocks;
}
/*
* mpage_map_buffers - update buffers corresponding to changed extent and
* submit fully mapped pages for IO
*
* @mpd - description of extent to map, on return next extent to map
*
* Scan buffers corresponding to changed extent (we expect corresponding pages
* to be already locked) and update buffer state according to new extent state.
* We map delalloc buffers to their physical location, clear unwritten bits,
* and mark buffers as uninit when we perform writes to unwritten extents
* and do extent conversion after IO is finished. If the last page is not fully
* mapped, we update @map to the next extent in the last page that needs
* mapping. Otherwise we submit the page for IO.
*/
static int mpage_map_and_submit_buffers(struct mpage_da_data *mpd)
{
struct pagevec pvec;
int nr_pages, i;
struct inode *inode = mpd->inode;
struct buffer_head *head, *bh;
int bpp_bits = PAGE_CACHE_SHIFT - inode->i_blkbits;
pgoff_t start, end;
ext4_lblk_t lblk;
sector_t pblock;
int err;
start = mpd->map.m_lblk >> bpp_bits;
end = (mpd->map.m_lblk + mpd->map.m_len - 1) >> bpp_bits;
lblk = start << bpp_bits;
pblock = mpd->map.m_pblk;
pagevec_init(&pvec, 0);
while (start <= end) {
nr_pages = pagevec_lookup(&pvec, inode->i_mapping, start,
PAGEVEC_SIZE);
if (nr_pages == 0)
break;
for (i = 0; i < nr_pages; i++) {
struct page *page = pvec.pages[i];
if (page->index > end)
break;
/* Up to 'end' pages must be contiguous */
BUG_ON(page->index != start);
bh = head = page_buffers(page);
do {
if (lblk < mpd->map.m_lblk)
continue;
if (lblk >= mpd->map.m_lblk + mpd->map.m_len) {
/*
* Buffer after end of mapped extent.
* Find next buffer in the page to map.
*/
mpd->map.m_len = 0;
mpd->map.m_flags = 0;
/*
* FIXME: If dioread_nolock supports
* blocksize < pagesize, we need to make
* sure we add size mapped so far to
* io_end->size as the following call
* can submit the page for IO.
*/
err = mpage_process_page_bufs(mpd, head,
bh, lblk);
pagevec_release(&pvec);
if (err > 0)
err = 0;
return err;
}
if (buffer_delay(bh)) {
clear_buffer_delay(bh);
bh->b_blocknr = pblock++;
}
clear_buffer_unwritten(bh);
} while (lblk++, (bh = bh->b_this_page) != head);
/*
* FIXME: This is going to break if dioread_nolock
* supports blocksize < pagesize as we will try to
* convert potentially unmapped parts of inode.
*/
mpd->io_submit.io_end->size += PAGE_CACHE_SIZE;
/* Page fully mapped - let IO run! */
err = mpage_submit_page(mpd, page);
if (err < 0) {
pagevec_release(&pvec);
return err;
}
start++;
}
pagevec_release(&pvec);
}
/* Extent fully mapped and matches with page boundary. We are done. */
mpd->map.m_len = 0;
mpd->map.m_flags = 0;
return 0;
}
static int mpage_map_one_extent(handle_t *handle, struct mpage_da_data *mpd)
{
struct inode *inode = mpd->inode;
struct ext4_map_blocks *map = &mpd->map;
int get_blocks_flags;
int err, dioread_nolock;
trace_ext4_da_write_pages_extent(inode, map);
/*
* Call ext4_map_blocks() to allocate any delayed allocation blocks, or
* to convert an unwritten extent to be initialized (in the case
* where we have written into one or more preallocated blocks). It is
* possible that we're going to need more metadata blocks than
* previously reserved. However we must not fail because we're in
* writeback and there is nothing we can do about it so it might result
* in data loss. So use reserved blocks to allocate metadata if
* possible.
*
* We pass in the magic EXT4_GET_BLOCKS_DELALLOC_RESERVE if the blocks
* in question are delalloc blocks. This affects functions in many
* different parts of the allocation call path. This flag exists
* primarily because we don't want to change *many* call functions, so
* ext4_map_blocks() will set the EXT4_STATE_DELALLOC_RESERVED flag
* once the inode's allocation semaphore is taken.
*/
get_blocks_flags = EXT4_GET_BLOCKS_CREATE |
EXT4_GET_BLOCKS_METADATA_NOFAIL;
dioread_nolock = ext4_should_dioread_nolock(inode);
if (dioread_nolock)
get_blocks_flags |= EXT4_GET_BLOCKS_IO_CREATE_EXT;
if (map->m_flags & (1 << BH_Delay))
get_blocks_flags |= EXT4_GET_BLOCKS_DELALLOC_RESERVE;
err = ext4_map_blocks(handle, inode, map, get_blocks_flags);
if (err < 0)
return err;
if (dioread_nolock && (map->m_flags & EXT4_MAP_UNWRITTEN)) {
if (!mpd->io_submit.io_end->handle &&
ext4_handle_valid(handle)) {
mpd->io_submit.io_end->handle = handle->h_rsv_handle;
handle->h_rsv_handle = NULL;
}
ext4_set_io_unwritten_flag(inode, mpd->io_submit.io_end);
}
BUG_ON(map->m_len == 0);
if (map->m_flags & EXT4_MAP_NEW) {
struct block_device *bdev = inode->i_sb->s_bdev;
int i;
for (i = 0; i < map->m_len; i++)
unmap_underlying_metadata(bdev, map->m_pblk + i);
}
return 0;
}
/*
* mpage_map_and_submit_extent - map extent starting at mpd->lblk of length
* mpd->len and submit pages underlying it for IO
*
* @handle - handle for journal operations
* @mpd - extent to map
* @give_up_on_write - we set this to true iff there is a fatal error and there
* is no hope of writing the data. The caller should discard
* dirty pages to avoid infinite loops.
*
* The function maps extent starting at mpd->lblk of length mpd->len. If it is
* delayed, blocks are allocated, if it is unwritten, we may need to convert
* them to initialized or split the described range from larger unwritten
* extent. Note that we need not map all the described range since allocation
* can return less blocks or the range is covered by more unwritten extents. We
* cannot map more because we are limited by reserved transaction credits. On
* the other hand we always make sure that the last touched page is fully
* mapped so that it can be written out (and thus forward progress is
* guaranteed). After mapping we submit all mapped pages for IO.
*/
static int mpage_map_and_submit_extent(handle_t *handle,
struct mpage_da_data *mpd,
bool *give_up_on_write)
{
struct inode *inode = mpd->inode;
struct ext4_map_blocks *map = &mpd->map;
int err;
loff_t disksize;
int progress = 0;
mpd->io_submit.io_end->offset =
((loff_t)map->m_lblk) << inode->i_blkbits;
do {
err = mpage_map_one_extent(handle, mpd);
if (err < 0) {
struct super_block *sb = inode->i_sb;
if (EXT4_SB(sb)->s_mount_flags & EXT4_MF_FS_ABORTED)
goto invalidate_dirty_pages;
/*
* Let the uper layers retry transient errors.
* In the case of ENOSPC, if ext4_count_free_blocks()
* is non-zero, a commit should free up blocks.
*/
if ((err == -ENOMEM) ||
(err == -ENOSPC && ext4_count_free_clusters(sb))) {
if (progress)
goto update_disksize;
return err;
}
ext4_msg(sb, KERN_CRIT,
"Delayed block allocation failed for "
"inode %lu at logical offset %llu with"
" max blocks %u with error %d",
inode->i_ino,
(unsigned long long)map->m_lblk,
(unsigned)map->m_len, -err);
ext4_msg(sb, KERN_CRIT,
"This should not happen!! Data will "
"be lost\n");
if (err == -ENOSPC)
ext4_print_free_blocks(inode);
invalidate_dirty_pages:
*give_up_on_write = true;
return err;
}
progress = 1;
/*
* Update buffer state, submit mapped pages, and get us new
* extent to map
*/
err = mpage_map_and_submit_buffers(mpd);
if (err < 0)
goto update_disksize;
} while (map->m_len);
update_disksize:
/*
* Update on-disk size after IO is submitted. Races with
* truncate are avoided by checking i_size under i_data_sem.
*/
disksize = ((loff_t)mpd->first_page) << PAGE_CACHE_SHIFT;
if (disksize > EXT4_I(inode)->i_disksize) {
int err2;
loff_t i_size;
down_write(&EXT4_I(inode)->i_data_sem);
i_size = i_size_read(inode);
if (disksize > i_size)
disksize = i_size;
if (disksize > EXT4_I(inode)->i_disksize)
EXT4_I(inode)->i_disksize = disksize;
err2 = ext4_mark_inode_dirty(handle, inode);
up_write(&EXT4_I(inode)->i_data_sem);
if (err2)
ext4_error(inode->i_sb,
"Failed to mark inode %lu dirty",
inode->i_ino);
if (!err)
err = err2;
}
return err;
}
/*
* Calculate the total number of credits to reserve for one writepages
* iteration. This is called from ext4_writepages(). We map an extent of
* up to MAX_WRITEPAGES_EXTENT_LEN blocks and then we go on and finish mapping
* the last partial page. So in total we can map MAX_WRITEPAGES_EXTENT_LEN +
* bpp - 1 blocks in bpp different extents.
*/
static int ext4_da_writepages_trans_blocks(struct inode *inode)
{
int bpp = ext4_journal_blocks_per_page(inode);
return ext4_meta_trans_blocks(inode,
MAX_WRITEPAGES_EXTENT_LEN + bpp - 1, bpp);
}
/*
* mpage_prepare_extent_to_map - find & lock contiguous range of dirty pages
* and underlying extent to map
*
* @mpd - where to look for pages
*
* Walk dirty pages in the mapping. If they are fully mapped, submit them for
* IO immediately. When we find a page which isn't mapped we start accumulating
* extent of buffers underlying these pages that needs mapping (formed by
* either delayed or unwritten buffers). We also lock the pages containing
* these buffers. The extent found is returned in @mpd structure (starting at
* mpd->lblk with length mpd->len blocks).
*
* Note that this function can attach bios to one io_end structure which are
* neither logically nor physically contiguous. Although it may seem as an
* unnecessary complication, it is actually inevitable in blocksize < pagesize
* case as we need to track IO to all buffers underlying a page in one io_end.
*/
static int mpage_prepare_extent_to_map(struct mpage_da_data *mpd)
{
struct address_space *mapping = mpd->inode->i_mapping;
struct pagevec pvec;
unsigned int nr_pages;
long left = mpd->wbc->nr_to_write;
pgoff_t index = mpd->first_page;
pgoff_t end = mpd->last_page;
int tag;
int i, err = 0;
int blkbits = mpd->inode->i_blkbits;
ext4_lblk_t lblk;
struct buffer_head *head;
if (mpd->wbc->sync_mode == WB_SYNC_ALL || mpd->wbc->tagged_writepages)
tag = PAGECACHE_TAG_TOWRITE;
else
tag = PAGECACHE_TAG_DIRTY;
pagevec_init(&pvec, 0);
mpd->map.m_len = 0;
mpd->next_page = index;
while (index <= end) {
nr_pages = pagevec_lookup_tag(&pvec, mapping, &index, tag,
min(end - index, (pgoff_t)PAGEVEC_SIZE-1) + 1);
if (nr_pages == 0)
goto out;
for (i = 0; i < nr_pages; i++) {
struct page *page = pvec.pages[i];
/*
* At this point, the page may be truncated or
* invalidated (changing page->mapping to NULL), or
* even swizzled back from swapper_space to tmpfs file
* mapping. However, page->index will not change
* because we have a reference on the page.
*/
if (page->index > end)
goto out;
/*
* Accumulated enough dirty pages? This doesn't apply
* to WB_SYNC_ALL mode. For integrity sync we have to
* keep going because someone may be concurrently
* dirtying pages, and we might have synced a lot of
* newly appeared dirty pages, but have not synced all
* of the old dirty pages.
*/
if (mpd->wbc->sync_mode == WB_SYNC_NONE && left <= 0)
goto out;
/* If we can't merge this page, we are done. */
if (mpd->map.m_len > 0 && mpd->next_page != page->index)
goto out;
lock_page(page);
/*
* If the page is no longer dirty, or its mapping no
* longer corresponds to inode we are writing (which
* means it has been truncated or invalidated), or the
* page is already under writeback and we are not doing
* a data integrity writeback, skip the page
*/
if (!PageDirty(page) ||
(PageWriteback(page) &&
(mpd->wbc->sync_mode == WB_SYNC_NONE)) ||
unlikely(page->mapping != mapping)) {
unlock_page(page);
continue;
}
wait_on_page_writeback(page);
BUG_ON(PageWriteback(page));
if (mpd->map.m_len == 0)
mpd->first_page = page->index;
mpd->next_page = page->index + 1;
/* Add all dirty buffers to mpd */
lblk = ((ext4_lblk_t)page->index) <<
(PAGE_CACHE_SHIFT - blkbits);
head = page_buffers(page);
err = mpage_process_page_bufs(mpd, head, head, lblk);
if (err <= 0)
goto out;
err = 0;
left--;
}
pagevec_release(&pvec);
cond_resched();
}
return 0;
out:
pagevec_release(&pvec);
return err;
}
static int __writepage(struct page *page, struct writeback_control *wbc,
void *data)
{
struct address_space *mapping = data;
int ret = ext4_writepage(page, wbc);
mapping_set_error(mapping, ret);
return ret;
}
static int ext4_writepages(struct address_space *mapping,
struct writeback_control *wbc)
{
pgoff_t writeback_index = 0;
long nr_to_write = wbc->nr_to_write;
int range_whole = 0;
int cycled = 1;
handle_t *handle = NULL;
struct mpage_da_data mpd;
struct inode *inode = mapping->host;
int needed_blocks, rsv_blocks = 0, ret = 0;
struct ext4_sb_info *sbi = EXT4_SB(mapping->host->i_sb);
bool done;
struct blk_plug plug;
bool give_up_on_write = false;
trace_ext4_writepages(inode, wbc);
/*
* No pages to write? This is mainly a kludge to avoid starting
* a transaction for special inodes like journal inode on last iput()
* because that could violate lock ordering on umount
*/
if (!mapping->nrpages || !mapping_tagged(mapping, PAGECACHE_TAG_DIRTY))
goto out_writepages;
if (ext4_should_journal_data(inode)) {
struct blk_plug plug;
blk_start_plug(&plug);
ret = write_cache_pages(mapping, wbc, __writepage, mapping);
blk_finish_plug(&plug);
goto out_writepages;
}
/*
* If the filesystem has aborted, it is read-only, so return
* right away instead of dumping stack traces later on that
* will obscure the real source of the problem. We test
* EXT4_MF_FS_ABORTED instead of sb->s_flag's MS_RDONLY because
* the latter could be true if the filesystem is mounted
* read-only, and in that case, ext4_writepages should
* *never* be called, so if that ever happens, we would want
* the stack trace.
*/
if (unlikely(sbi->s_mount_flags & EXT4_MF_FS_ABORTED)) {
ret = -EROFS;
goto out_writepages;
}
if (ext4_should_dioread_nolock(inode)) {
/*
* We may need to convert up to one extent per block in
* the page and we may dirty the inode.
*/
rsv_blocks = 1 + (PAGE_CACHE_SIZE >> inode->i_blkbits);
}
/*
* If we have inline data and arrive here, it means that
* we will soon create the block for the 1st page, so
* we'd better clear the inline data here.
*/
if (ext4_has_inline_data(inode)) {
/* Just inode will be modified... */
handle = ext4_journal_start(inode, EXT4_HT_INODE, 1);
if (IS_ERR(handle)) {
ret = PTR_ERR(handle);
goto out_writepages;
}
BUG_ON(ext4_test_inode_state(inode,
EXT4_STATE_MAY_INLINE_DATA));
ext4_destroy_inline_data(handle, inode);
ext4_journal_stop(handle);
}
if (wbc->range_start == 0 && wbc->range_end == LLONG_MAX)
range_whole = 1;
if (wbc->range_cyclic) {
writeback_index = mapping->writeback_index;
if (writeback_index)
cycled = 0;
mpd.first_page = writeback_index;
mpd.last_page = -1;
} else {
mpd.first_page = wbc->range_start >> PAGE_CACHE_SHIFT;
mpd.last_page = wbc->range_end >> PAGE_CACHE_SHIFT;
}
mpd.inode = inode;
mpd.wbc = wbc;
ext4_io_submit_init(&mpd.io_submit, wbc);
retry:
if (wbc->sync_mode == WB_SYNC_ALL || wbc->tagged_writepages)
tag_pages_for_writeback(mapping, mpd.first_page, mpd.last_page);
done = false;
blk_start_plug(&plug);
while (!done && mpd.first_page <= mpd.last_page) {
/* For each extent of pages we use new io_end */
mpd.io_submit.io_end = ext4_init_io_end(inode, GFP_KERNEL);
if (!mpd.io_submit.io_end) {
ret = -ENOMEM;
break;
}
/*
* We have two constraints: We find one extent to map and we
* must always write out whole page (makes a difference when
* blocksize < pagesize) so that we don't block on IO when we
* try to write out the rest of the page. Journalled mode is
* not supported by delalloc.
*/
BUG_ON(ext4_should_journal_data(inode));
needed_blocks = ext4_da_writepages_trans_blocks(inode);
/* start a new transaction */
handle = ext4_journal_start_with_reserve(inode,
EXT4_HT_WRITE_PAGE, needed_blocks, rsv_blocks);
if (IS_ERR(handle)) {
ret = PTR_ERR(handle);
ext4_msg(inode->i_sb, KERN_CRIT, "%s: jbd2_start: "
"%ld pages, ino %lu; err %d", __func__,
wbc->nr_to_write, inode->i_ino, ret);
/* Release allocated io_end */
ext4_put_io_end(mpd.io_submit.io_end);
break;
}
trace_ext4_da_write_pages(inode, mpd.first_page, mpd.wbc);
ret = mpage_prepare_extent_to_map(&mpd);
if (!ret) {
if (mpd.map.m_len)
ret = mpage_map_and_submit_extent(handle, &mpd,
&give_up_on_write);
else {
/*
* We scanned the whole range (or exhausted
* nr_to_write), submitted what was mapped and
* didn't find anything needing mapping. We are
* done.
*/
done = true;
}
}
ext4_journal_stop(handle);
/* Submit prepared bio */
ext4_io_submit(&mpd.io_submit);
/* Unlock pages we didn't use */
mpage_release_unused_pages(&mpd, give_up_on_write);
/* Drop our io_end reference we got from init */
ext4_put_io_end(mpd.io_submit.io_end);
if (ret == -ENOSPC && sbi->s_journal) {
/*
* Commit the transaction which would
* free blocks released in the transaction
* and try again
*/
jbd2_journal_force_commit_nested(sbi->s_journal);
ret = 0;
continue;
}
/* Fatal error - ENOMEM, EIO... */
if (ret)
break;
}
blk_finish_plug(&plug);
if (!ret && !cycled && wbc->nr_to_write > 0) {
cycled = 1;
mpd.last_page = writeback_index - 1;
mpd.first_page = 0;
goto retry;
}
/* Update index */
if (wbc->range_cyclic || (range_whole && wbc->nr_to_write > 0))
/*
* Set the writeback_index so that range_cyclic
* mode will write it back later
*/
mapping->writeback_index = mpd.first_page;
out_writepages:
trace_ext4_writepages_result(inode, wbc, ret,
nr_to_write - wbc->nr_to_write);
return ret;
}
static int ext4_nonda_switch(struct super_block *sb)
{
s64 free_clusters, dirty_clusters;
struct ext4_sb_info *sbi = EXT4_SB(sb);
/*
* switch to non delalloc mode if we are running low
* on free block. The free block accounting via percpu
* counters can get slightly wrong with percpu_counter_batch getting
* accumulated on each CPU without updating global counters
* Delalloc need an accurate free block accounting. So switch
* to non delalloc when we are near to error range.
*/
free_clusters =
percpu_counter_read_positive(&sbi->s_freeclusters_counter);
dirty_clusters =
percpu_counter_read_positive(&sbi->s_dirtyclusters_counter);
ext4: fix potential deadlock in ext4_nonda_switch() In ext4_nonda_switch(), if the file system is getting full we used to call writeback_inodes_sb_if_idle(). The problem is that we can be holding i_mutex already, and this causes a potential deadlock when writeback_inodes_sb_if_idle() when it tries to take s_umount. (See lockdep output below). As it turns out we don't need need to hold s_umount; the fact that we are in the middle of the write(2) system call will keep the superblock pinned. Unfortunately writeback_inodes_sb() checks to make sure s_umount is taken, and the VFS uses a different mechanism for making sure the file system doesn't get unmounted out from under us. The simplest way of dealing with this is to just simply grab s_umount using a trylock, and skip kicking the writeback flusher thread in the very unlikely case that we can't take a read lock on s_umount without blocking. Also, we now check the cirteria for kicking the writeback thread before we decide to whether to fall back to non-delayed writeback, so if there are any outstanding delayed allocation writes, we try to get them resolved as soon as possible. [ INFO: possible circular locking dependency detected ] 3.6.0-rc1-00042-gce894ca #367 Not tainted ------------------------------------------------------- dd/8298 is trying to acquire lock: (&type->s_umount_key#18){++++..}, at: [<c02277d4>] writeback_inodes_sb_if_idle+0x28/0x46 but task is already holding lock: (&sb->s_type->i_mutex_key#8){+.+...}, at: [<c01ddcce>] generic_file_aio_write+0x5f/0xd3 which lock already depends on the new lock. 2 locks held by dd/8298: #0: (sb_writers#2){.+.+.+}, at: [<c01ddcc5>] generic_file_aio_write+0x56/0xd3 #1: (&sb->s_type->i_mutex_key#8){+.+...}, at: [<c01ddcce>] generic_file_aio_write+0x5f/0xd3 stack backtrace: Pid: 8298, comm: dd Not tainted 3.6.0-rc1-00042-gce894ca #367 Call Trace: [<c015b79c>] ? console_unlock+0x345/0x372 [<c06d62a1>] print_circular_bug+0x190/0x19d [<c019906c>] __lock_acquire+0x86d/0xb6c [<c01999db>] ? mark_held_locks+0x5c/0x7b [<c0199724>] lock_acquire+0x66/0xb9 [<c02277d4>] ? writeback_inodes_sb_if_idle+0x28/0x46 [<c06db935>] down_read+0x28/0x58 [<c02277d4>] ? writeback_inodes_sb_if_idle+0x28/0x46 [<c02277d4>] writeback_inodes_sb_if_idle+0x28/0x46 [<c026f3b2>] ext4_nonda_switch+0xe1/0xf4 [<c0271ece>] ext4_da_write_begin+0x27/0x193 [<c01dcdb0>] generic_file_buffered_write+0xc8/0x1bb [<c01ddc47>] __generic_file_aio_write+0x1dd/0x205 [<c01ddce7>] generic_file_aio_write+0x78/0xd3 [<c026d336>] ext4_file_write+0x480/0x4a6 [<c0198c1d>] ? __lock_acquire+0x41e/0xb6c [<c0180944>] ? sched_clock_cpu+0x11a/0x13e [<c01967e9>] ? trace_hardirqs_off+0xb/0xd [<c018099f>] ? local_clock+0x37/0x4e [<c0209f2c>] do_sync_write+0x67/0x9d [<c0209ec5>] ? wait_on_retry_sync_kiocb+0x44/0x44 [<c020a7b9>] vfs_write+0x7b/0xe6 [<c020a9a6>] sys_write+0x3b/0x64 [<c06dd4bd>] syscall_call+0x7/0xb Signed-off-by: "Theodore Ts'o" <tytso@mit.edu> Cc: stable@vger.kernel.org
2012-09-20 02:42:36 +00:00
/*
* Start pushing delalloc when 1/2 of free blocks are dirty.
*/
if (dirty_clusters && (free_clusters < 2 * dirty_clusters))
try_to_writeback_inodes_sb(sb, WB_REASON_FS_FREE_SPACE);
ext4: fix potential deadlock in ext4_nonda_switch() In ext4_nonda_switch(), if the file system is getting full we used to call writeback_inodes_sb_if_idle(). The problem is that we can be holding i_mutex already, and this causes a potential deadlock when writeback_inodes_sb_if_idle() when it tries to take s_umount. (See lockdep output below). As it turns out we don't need need to hold s_umount; the fact that we are in the middle of the write(2) system call will keep the superblock pinned. Unfortunately writeback_inodes_sb() checks to make sure s_umount is taken, and the VFS uses a different mechanism for making sure the file system doesn't get unmounted out from under us. The simplest way of dealing with this is to just simply grab s_umount using a trylock, and skip kicking the writeback flusher thread in the very unlikely case that we can't take a read lock on s_umount without blocking. Also, we now check the cirteria for kicking the writeback thread before we decide to whether to fall back to non-delayed writeback, so if there are any outstanding delayed allocation writes, we try to get them resolved as soon as possible. [ INFO: possible circular locking dependency detected ] 3.6.0-rc1-00042-gce894ca #367 Not tainted ------------------------------------------------------- dd/8298 is trying to acquire lock: (&type->s_umount_key#18){++++..}, at: [<c02277d4>] writeback_inodes_sb_if_idle+0x28/0x46 but task is already holding lock: (&sb->s_type->i_mutex_key#8){+.+...}, at: [<c01ddcce>] generic_file_aio_write+0x5f/0xd3 which lock already depends on the new lock. 2 locks held by dd/8298: #0: (sb_writers#2){.+.+.+}, at: [<c01ddcc5>] generic_file_aio_write+0x56/0xd3 #1: (&sb->s_type->i_mutex_key#8){+.+...}, at: [<c01ddcce>] generic_file_aio_write+0x5f/0xd3 stack backtrace: Pid: 8298, comm: dd Not tainted 3.6.0-rc1-00042-gce894ca #367 Call Trace: [<c015b79c>] ? console_unlock+0x345/0x372 [<c06d62a1>] print_circular_bug+0x190/0x19d [<c019906c>] __lock_acquire+0x86d/0xb6c [<c01999db>] ? mark_held_locks+0x5c/0x7b [<c0199724>] lock_acquire+0x66/0xb9 [<c02277d4>] ? writeback_inodes_sb_if_idle+0x28/0x46 [<c06db935>] down_read+0x28/0x58 [<c02277d4>] ? writeback_inodes_sb_if_idle+0x28/0x46 [<c02277d4>] writeback_inodes_sb_if_idle+0x28/0x46 [<c026f3b2>] ext4_nonda_switch+0xe1/0xf4 [<c0271ece>] ext4_da_write_begin+0x27/0x193 [<c01dcdb0>] generic_file_buffered_write+0xc8/0x1bb [<c01ddc47>] __generic_file_aio_write+0x1dd/0x205 [<c01ddce7>] generic_file_aio_write+0x78/0xd3 [<c026d336>] ext4_file_write+0x480/0x4a6 [<c0198c1d>] ? __lock_acquire+0x41e/0xb6c [<c0180944>] ? sched_clock_cpu+0x11a/0x13e [<c01967e9>] ? trace_hardirqs_off+0xb/0xd [<c018099f>] ? local_clock+0x37/0x4e [<c0209f2c>] do_sync_write+0x67/0x9d [<c0209ec5>] ? wait_on_retry_sync_kiocb+0x44/0x44 [<c020a7b9>] vfs_write+0x7b/0xe6 [<c020a9a6>] sys_write+0x3b/0x64 [<c06dd4bd>] syscall_call+0x7/0xb Signed-off-by: "Theodore Ts'o" <tytso@mit.edu> Cc: stable@vger.kernel.org
2012-09-20 02:42:36 +00:00
if (2 * free_clusters < 3 * dirty_clusters ||
free_clusters < (dirty_clusters + EXT4_FREECLUSTERS_WATERMARK)) {
/*
* free block count is less than 150% of dirty blocks
* or free blocks is less than watermark
*/
return 1;
}
return 0;
}
static int ext4_da_write_begin(struct file *file, struct address_space *mapping,
loff_t pos, unsigned len, unsigned flags,
struct page **pagep, void **fsdata)
{
int ret, retries = 0;
struct page *page;
pgoff_t index;
struct inode *inode = mapping->host;
handle_t *handle;
index = pos >> PAGE_CACHE_SHIFT;
if (ext4_nonda_switch(inode->i_sb)) {
*fsdata = (void *)FALL_BACK_TO_NONDELALLOC;
return ext4_write_begin(file, mapping, pos,
len, flags, pagep, fsdata);
}
*fsdata = (void *)0;
trace_ext4_da_write_begin(inode, pos, len, flags);
if (ext4_test_inode_state(inode, EXT4_STATE_MAY_INLINE_DATA)) {
ret = ext4_da_write_inline_data_begin(mapping, inode,
pos, len, flags,
pagep, fsdata);
if (ret < 0)
return ret;
if (ret == 1)
return 0;
}
/*
* grab_cache_page_write_begin() can take a long time if the
* system is thrashing due to memory pressure, or if the page
* is being written back. So grab it first before we start
* the transaction handle. This also allows us to allocate
* the page (if needed) without using GFP_NOFS.
*/
retry_grab:
page = grab_cache_page_write_begin(mapping, index, flags);
if (!page)
return -ENOMEM;
unlock_page(page);
/*
* With delayed allocation, we don't log the i_disksize update
* if there is delayed block allocation. But we still need
* to journalling the i_disksize update if writes to the end
* of file which has an already mapped buffer.
*/
retry_journal:
handle = ext4_journal_start(inode, EXT4_HT_WRITE_PAGE, 1);
if (IS_ERR(handle)) {
page_cache_release(page);
return PTR_ERR(handle);
}
lock_page(page);
if (page->mapping != mapping) {
/* The page got truncated from under us */
unlock_page(page);
page_cache_release(page);
ext4_journal_stop(handle);
goto retry_grab;
}
/* In case writeback began while the page was unlocked */
wait_for_stable_page(page);
ret = __block_write_begin(page, pos, len, ext4_da_get_block_prep);
if (ret < 0) {
unlock_page(page);
ext4_journal_stop(handle);
/*
* block_write_begin may have instantiated a few blocks
* outside i_size. Trim these off again. Don't need
* i_size_read because we hold i_mutex.
*/
if (pos + len > inode->i_size)
ext4_truncate_failed_write(inode);
if (ret == -ENOSPC &&
ext4_should_retry_alloc(inode->i_sb, &retries))
goto retry_journal;
page_cache_release(page);
return ret;
}
*pagep = page;
return ret;
}
ext4: fix delalloc i_disksize early update issue Ext4_da_write_end() used walk_page_buffers() with a callback function of ext4_bh_unmapped_or_delay() to check if it extended the file size without allocating any blocks (since in this case i_disksize needs to be updated). However, this is didn't work proprely because the buffer head has not been marked dirty yet --- this is done later in block_commit_write() --- which caused ext4_bh_unmapped_or_delay() to always return false. In addition, walk_page_buffers() checks all of the buffer heads covering the page, and the only buffer_head that should be checked is the one covering the end of the write. Otherwise, given a 1k blocksize filesystem and a 4k page size, the buffer head covering the first 1k stripe of the file could be unmapped (because it was a sparse file), and the second or third buffer_head covering that page could be mapped, and using walk_page_buffers() would fail in this case since it would stop at the first unmapped buffer_head and return true. The core problem is that walk_page_buffers() was intended to do work in a callback function, and a non-zero return value indicated a failure, which termined the walk of the buffer heads covering the page. It was not intended to be used with a boolean function, such as ext4_bh_unmapped_or_delay(). Add addtional fix from Aneesh to protect i_disksize update rave with truncate. Signed-off-by: Mingming Cao <cmm@us.ibm.com> Signed-off-by: Aneesh Kumar K.V <aneesh.kumar@linux.vnet.ibm.com> Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
2008-07-11 23:27:31 +00:00
/*
* Check if we should update i_disksize
* when write to the end of file but not require block allocation
*/
static int ext4_da_should_update_i_disksize(struct page *page,
unsigned long offset)
ext4: fix delalloc i_disksize early update issue Ext4_da_write_end() used walk_page_buffers() with a callback function of ext4_bh_unmapped_or_delay() to check if it extended the file size without allocating any blocks (since in this case i_disksize needs to be updated). However, this is didn't work proprely because the buffer head has not been marked dirty yet --- this is done later in block_commit_write() --- which caused ext4_bh_unmapped_or_delay() to always return false. In addition, walk_page_buffers() checks all of the buffer heads covering the page, and the only buffer_head that should be checked is the one covering the end of the write. Otherwise, given a 1k blocksize filesystem and a 4k page size, the buffer head covering the first 1k stripe of the file could be unmapped (because it was a sparse file), and the second or third buffer_head covering that page could be mapped, and using walk_page_buffers() would fail in this case since it would stop at the first unmapped buffer_head and return true. The core problem is that walk_page_buffers() was intended to do work in a callback function, and a non-zero return value indicated a failure, which termined the walk of the buffer heads covering the page. It was not intended to be used with a boolean function, such as ext4_bh_unmapped_or_delay(). Add addtional fix from Aneesh to protect i_disksize update rave with truncate. Signed-off-by: Mingming Cao <cmm@us.ibm.com> Signed-off-by: Aneesh Kumar K.V <aneesh.kumar@linux.vnet.ibm.com> Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
2008-07-11 23:27:31 +00:00
{
struct buffer_head *bh;
struct inode *inode = page->mapping->host;
unsigned int idx;
int i;
bh = page_buffers(page);
idx = offset >> inode->i_blkbits;
for (i = 0; i < idx; i++)
ext4: fix delalloc i_disksize early update issue Ext4_da_write_end() used walk_page_buffers() with a callback function of ext4_bh_unmapped_or_delay() to check if it extended the file size without allocating any blocks (since in this case i_disksize needs to be updated). However, this is didn't work proprely because the buffer head has not been marked dirty yet --- this is done later in block_commit_write() --- which caused ext4_bh_unmapped_or_delay() to always return false. In addition, walk_page_buffers() checks all of the buffer heads covering the page, and the only buffer_head that should be checked is the one covering the end of the write. Otherwise, given a 1k blocksize filesystem and a 4k page size, the buffer head covering the first 1k stripe of the file could be unmapped (because it was a sparse file), and the second or third buffer_head covering that page could be mapped, and using walk_page_buffers() would fail in this case since it would stop at the first unmapped buffer_head and return true. The core problem is that walk_page_buffers() was intended to do work in a callback function, and a non-zero return value indicated a failure, which termined the walk of the buffer heads covering the page. It was not intended to be used with a boolean function, such as ext4_bh_unmapped_or_delay(). Add addtional fix from Aneesh to protect i_disksize update rave with truncate. Signed-off-by: Mingming Cao <cmm@us.ibm.com> Signed-off-by: Aneesh Kumar K.V <aneesh.kumar@linux.vnet.ibm.com> Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
2008-07-11 23:27:31 +00:00
bh = bh->b_this_page;
if (!buffer_mapped(bh) || (buffer_delay(bh)) || buffer_unwritten(bh))
ext4: fix delalloc i_disksize early update issue Ext4_da_write_end() used walk_page_buffers() with a callback function of ext4_bh_unmapped_or_delay() to check if it extended the file size without allocating any blocks (since in this case i_disksize needs to be updated). However, this is didn't work proprely because the buffer head has not been marked dirty yet --- this is done later in block_commit_write() --- which caused ext4_bh_unmapped_or_delay() to always return false. In addition, walk_page_buffers() checks all of the buffer heads covering the page, and the only buffer_head that should be checked is the one covering the end of the write. Otherwise, given a 1k blocksize filesystem and a 4k page size, the buffer head covering the first 1k stripe of the file could be unmapped (because it was a sparse file), and the second or third buffer_head covering that page could be mapped, and using walk_page_buffers() would fail in this case since it would stop at the first unmapped buffer_head and return true. The core problem is that walk_page_buffers() was intended to do work in a callback function, and a non-zero return value indicated a failure, which termined the walk of the buffer heads covering the page. It was not intended to be used with a boolean function, such as ext4_bh_unmapped_or_delay(). Add addtional fix from Aneesh to protect i_disksize update rave with truncate. Signed-off-by: Mingming Cao <cmm@us.ibm.com> Signed-off-by: Aneesh Kumar K.V <aneesh.kumar@linux.vnet.ibm.com> Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
2008-07-11 23:27:31 +00:00
return 0;
return 1;
}
static int ext4_da_write_end(struct file *file,
struct address_space *mapping,
loff_t pos, unsigned len, unsigned copied,
struct page *page, void *fsdata)
{
struct inode *inode = mapping->host;
int ret = 0, ret2;
handle_t *handle = ext4_journal_current_handle();
loff_t new_i_size;
ext4: fix delalloc i_disksize early update issue Ext4_da_write_end() used walk_page_buffers() with a callback function of ext4_bh_unmapped_or_delay() to check if it extended the file size without allocating any blocks (since in this case i_disksize needs to be updated). However, this is didn't work proprely because the buffer head has not been marked dirty yet --- this is done later in block_commit_write() --- which caused ext4_bh_unmapped_or_delay() to always return false. In addition, walk_page_buffers() checks all of the buffer heads covering the page, and the only buffer_head that should be checked is the one covering the end of the write. Otherwise, given a 1k blocksize filesystem and a 4k page size, the buffer head covering the first 1k stripe of the file could be unmapped (because it was a sparse file), and the second or third buffer_head covering that page could be mapped, and using walk_page_buffers() would fail in this case since it would stop at the first unmapped buffer_head and return true. The core problem is that walk_page_buffers() was intended to do work in a callback function, and a non-zero return value indicated a failure, which termined the walk of the buffer heads covering the page. It was not intended to be used with a boolean function, such as ext4_bh_unmapped_or_delay(). Add addtional fix from Aneesh to protect i_disksize update rave with truncate. Signed-off-by: Mingming Cao <cmm@us.ibm.com> Signed-off-by: Aneesh Kumar K.V <aneesh.kumar@linux.vnet.ibm.com> Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
2008-07-11 23:27:31 +00:00
unsigned long start, end;
int write_mode = (int)(unsigned long)fsdata;
if (write_mode == FALL_BACK_TO_NONDELALLOC)
return ext4_write_end(file, mapping, pos,
len, copied, page, fsdata);
ext4: fix delalloc i_disksize early update issue Ext4_da_write_end() used walk_page_buffers() with a callback function of ext4_bh_unmapped_or_delay() to check if it extended the file size without allocating any blocks (since in this case i_disksize needs to be updated). However, this is didn't work proprely because the buffer head has not been marked dirty yet --- this is done later in block_commit_write() --- which caused ext4_bh_unmapped_or_delay() to always return false. In addition, walk_page_buffers() checks all of the buffer heads covering the page, and the only buffer_head that should be checked is the one covering the end of the write. Otherwise, given a 1k blocksize filesystem and a 4k page size, the buffer head covering the first 1k stripe of the file could be unmapped (because it was a sparse file), and the second or third buffer_head covering that page could be mapped, and using walk_page_buffers() would fail in this case since it would stop at the first unmapped buffer_head and return true. The core problem is that walk_page_buffers() was intended to do work in a callback function, and a non-zero return value indicated a failure, which termined the walk of the buffer heads covering the page. It was not intended to be used with a boolean function, such as ext4_bh_unmapped_or_delay(). Add addtional fix from Aneesh to protect i_disksize update rave with truncate. Signed-off-by: Mingming Cao <cmm@us.ibm.com> Signed-off-by: Aneesh Kumar K.V <aneesh.kumar@linux.vnet.ibm.com> Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
2008-07-11 23:27:31 +00:00
trace_ext4_da_write_end(inode, pos, len, copied);
ext4: fix delalloc i_disksize early update issue Ext4_da_write_end() used walk_page_buffers() with a callback function of ext4_bh_unmapped_or_delay() to check if it extended the file size without allocating any blocks (since in this case i_disksize needs to be updated). However, this is didn't work proprely because the buffer head has not been marked dirty yet --- this is done later in block_commit_write() --- which caused ext4_bh_unmapped_or_delay() to always return false. In addition, walk_page_buffers() checks all of the buffer heads covering the page, and the only buffer_head that should be checked is the one covering the end of the write. Otherwise, given a 1k blocksize filesystem and a 4k page size, the buffer head covering the first 1k stripe of the file could be unmapped (because it was a sparse file), and the second or third buffer_head covering that page could be mapped, and using walk_page_buffers() would fail in this case since it would stop at the first unmapped buffer_head and return true. The core problem is that walk_page_buffers() was intended to do work in a callback function, and a non-zero return value indicated a failure, which termined the walk of the buffer heads covering the page. It was not intended to be used with a boolean function, such as ext4_bh_unmapped_or_delay(). Add addtional fix from Aneesh to protect i_disksize update rave with truncate. Signed-off-by: Mingming Cao <cmm@us.ibm.com> Signed-off-by: Aneesh Kumar K.V <aneesh.kumar@linux.vnet.ibm.com> Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
2008-07-11 23:27:31 +00:00
start = pos & (PAGE_CACHE_SIZE - 1);
end = start + copied - 1;
/*
* generic_write_end() will run mark_inode_dirty() if i_size
* changes. So let's piggyback the i_disksize mark_inode_dirty
* into that.
*/
new_i_size = pos + copied;
ext4: avoid hangs in ext4_da_should_update_i_disksize() If the pte mapping in generic_perform_write() is unmapped between iov_iter_fault_in_readable() and iov_iter_copy_from_user_atomic(), the "copied" parameter to ->end_write can be zero. ext4 couldn't cope with it with delayed allocations enabled. This skips the i_disksize enlargement logic if copied is zero and no new data was appeneded to the inode. gdb> bt #0 0xffffffff811afe80 in ext4_da_should_update_i_disksize (file=0xffff88003f606a80, mapping=0xffff88001d3824e0, pos=0x1\ 08000, len=0x1000, copied=0x0, page=0xffffea0000d792e8, fsdata=0x0) at fs/ext4/inode.c:2467 #1 ext4_da_write_end (file=0xffff88003f606a80, mapping=0xffff88001d3824e0, pos=0x108000, len=0x1000, copied=0x0, page=0\ xffffea0000d792e8, fsdata=0x0) at fs/ext4/inode.c:2512 #2 0xffffffff810d97f1 in generic_perform_write (iocb=<value optimized out>, iov=<value optimized out>, nr_segs=<value o\ ptimized out>, pos=0x108000, ppos=0xffff88001e26be40, count=<value optimized out>, written=0x0) at mm/filemap.c:2440 #3 generic_file_buffered_write (iocb=<value optimized out>, iov=<value optimized out>, nr_segs=<value optimized out>, p\ os=0x108000, ppos=0xffff88001e26be40, count=<value optimized out>, written=0x0) at mm/filemap.c:2482 #4 0xffffffff810db5d1 in __generic_file_aio_write (iocb=0xffff88001e26bde8, iov=0xffff88001e26bec8, nr_segs=0x1, ppos=0\ xffff88001e26be40) at mm/filemap.c:2600 #5 0xffffffff810db853 in generic_file_aio_write (iocb=0xffff88001e26bde8, iov=0xffff88001e26bec8, nr_segs=<value optimi\ zed out>, pos=<value optimized out>) at mm/filemap.c:2632 #6 0xffffffff811a71aa in ext4_file_write (iocb=0xffff88001e26bde8, iov=0xffff88001e26bec8, nr_segs=0x1, pos=0x108000) a\ t fs/ext4/file.c:136 #7 0xffffffff811375aa in do_sync_write (filp=0xffff88003f606a80, buf=<value optimized out>, len=<value optimized out>, \ ppos=0xffff88001e26bf48) at fs/read_write.c:406 #8 0xffffffff81137e56 in vfs_write (file=0xffff88003f606a80, buf=0x1ec2960 <Address 0x1ec2960 out of bounds>, count=0x4\ 000, pos=0xffff88001e26bf48) at fs/read_write.c:435 #9 0xffffffff8113816c in sys_write (fd=<value optimized out>, buf=0x1ec2960 <Address 0x1ec2960 out of bounds>, count=0x\ 4000) at fs/read_write.c:487 #10 <signal handler called> #11 0x00007f120077a390 in __brk_reservation_fn_dmi_alloc__ () #12 0x0000000000000000 in ?? () gdb> print offset $22 = 0xffffffffffffffff gdb> print idx $23 = 0xffffffff gdb> print inode->i_blkbits $24 = 0xc gdb> up #1 ext4_da_write_end (file=0xffff88003f606a80, mapping=0xffff88001d3824e0, pos=0x108000, len=0x1000, copied=0x0, page=0\ xffffea0000d792e8, fsdata=0x0) at fs/ext4/inode.c:2512 2512 if (ext4_da_should_update_i_disksize(page, end)) { gdb> print start $25 = 0x0 gdb> print end $26 = 0xffffffffffffffff gdb> print pos $27 = 0x108000 gdb> print new_i_size $28 = 0x108000 gdb> print ((struct ext4_inode_info *)((char *)inode-((int)(&((struct ext4_inode_info *)0)->vfs_inode))))->i_disksize $29 = 0xd9000 gdb> down 2467 for (i = 0; i < idx; i++) gdb> print i $30 = 0xd44acbee This is 100% reproducible with some autonuma development code tuned in a very aggressive manner (not normal way even for knumad) which does "exotic" changes to the ptes. It wouldn't normally trigger but I don't see why it can't happen normally if the page is added to swap cache in between the two faults leading to "copied" being zero (which then hangs in ext4). So it should be fixed. Especially possible with lumpy reclaim (albeit disabled if compaction is enabled) as that would ignore the young bits in the ptes. Signed-off-by: Andrea Arcangeli <aarcange@redhat.com> Signed-off-by: "Theodore Ts'o" <tytso@mit.edu> Cc: stable@kernel.org
2011-12-14 02:41:15 +00:00
if (copied && new_i_size > EXT4_I(inode)->i_disksize) {
if (ext4_has_inline_data(inode) ||
ext4_da_should_update_i_disksize(page, end)) {
ext4: fix delalloc i_disksize early update issue Ext4_da_write_end() used walk_page_buffers() with a callback function of ext4_bh_unmapped_or_delay() to check if it extended the file size without allocating any blocks (since in this case i_disksize needs to be updated). However, this is didn't work proprely because the buffer head has not been marked dirty yet --- this is done later in block_commit_write() --- which caused ext4_bh_unmapped_or_delay() to always return false. In addition, walk_page_buffers() checks all of the buffer heads covering the page, and the only buffer_head that should be checked is the one covering the end of the write. Otherwise, given a 1k blocksize filesystem and a 4k page size, the buffer head covering the first 1k stripe of the file could be unmapped (because it was a sparse file), and the second or third buffer_head covering that page could be mapped, and using walk_page_buffers() would fail in this case since it would stop at the first unmapped buffer_head and return true. The core problem is that walk_page_buffers() was intended to do work in a callback function, and a non-zero return value indicated a failure, which termined the walk of the buffer heads covering the page. It was not intended to be used with a boolean function, such as ext4_bh_unmapped_or_delay(). Add addtional fix from Aneesh to protect i_disksize update rave with truncate. Signed-off-by: Mingming Cao <cmm@us.ibm.com> Signed-off-by: Aneesh Kumar K.V <aneesh.kumar@linux.vnet.ibm.com> Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
2008-07-11 23:27:31 +00:00
down_write(&EXT4_I(inode)->i_data_sem);
if (new_i_size > EXT4_I(inode)->i_disksize)
ext4: fix delalloc i_disksize early update issue Ext4_da_write_end() used walk_page_buffers() with a callback function of ext4_bh_unmapped_or_delay() to check if it extended the file size without allocating any blocks (since in this case i_disksize needs to be updated). However, this is didn't work proprely because the buffer head has not been marked dirty yet --- this is done later in block_commit_write() --- which caused ext4_bh_unmapped_or_delay() to always return false. In addition, walk_page_buffers() checks all of the buffer heads covering the page, and the only buffer_head that should be checked is the one covering the end of the write. Otherwise, given a 1k blocksize filesystem and a 4k page size, the buffer head covering the first 1k stripe of the file could be unmapped (because it was a sparse file), and the second or third buffer_head covering that page could be mapped, and using walk_page_buffers() would fail in this case since it would stop at the first unmapped buffer_head and return true. The core problem is that walk_page_buffers() was intended to do work in a callback function, and a non-zero return value indicated a failure, which termined the walk of the buffer heads covering the page. It was not intended to be used with a boolean function, such as ext4_bh_unmapped_or_delay(). Add addtional fix from Aneesh to protect i_disksize update rave with truncate. Signed-off-by: Mingming Cao <cmm@us.ibm.com> Signed-off-by: Aneesh Kumar K.V <aneesh.kumar@linux.vnet.ibm.com> Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
2008-07-11 23:27:31 +00:00
EXT4_I(inode)->i_disksize = new_i_size;
up_write(&EXT4_I(inode)->i_data_sem);
/* We need to mark inode dirty even if
* new_i_size is less that inode->i_size
* bu greater than i_disksize.(hint delalloc)
*/
ext4_mark_inode_dirty(handle, inode);
}
ext4: fix delalloc i_disksize early update issue Ext4_da_write_end() used walk_page_buffers() with a callback function of ext4_bh_unmapped_or_delay() to check if it extended the file size without allocating any blocks (since in this case i_disksize needs to be updated). However, this is didn't work proprely because the buffer head has not been marked dirty yet --- this is done later in block_commit_write() --- which caused ext4_bh_unmapped_or_delay() to always return false. In addition, walk_page_buffers() checks all of the buffer heads covering the page, and the only buffer_head that should be checked is the one covering the end of the write. Otherwise, given a 1k blocksize filesystem and a 4k page size, the buffer head covering the first 1k stripe of the file could be unmapped (because it was a sparse file), and the second or third buffer_head covering that page could be mapped, and using walk_page_buffers() would fail in this case since it would stop at the first unmapped buffer_head and return true. The core problem is that walk_page_buffers() was intended to do work in a callback function, and a non-zero return value indicated a failure, which termined the walk of the buffer heads covering the page. It was not intended to be used with a boolean function, such as ext4_bh_unmapped_or_delay(). Add addtional fix from Aneesh to protect i_disksize update rave with truncate. Signed-off-by: Mingming Cao <cmm@us.ibm.com> Signed-off-by: Aneesh Kumar K.V <aneesh.kumar@linux.vnet.ibm.com> Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
2008-07-11 23:27:31 +00:00
}
if (write_mode != CONVERT_INLINE_DATA &&
ext4_test_inode_state(inode, EXT4_STATE_MAY_INLINE_DATA) &&
ext4_has_inline_data(inode))
ret2 = ext4_da_write_inline_data_end(inode, pos, len, copied,
page);
else
ret2 = generic_write_end(file, mapping, pos, len, copied,
page, fsdata);
copied = ret2;
if (ret2 < 0)
ret = ret2;
ret2 = ext4_journal_stop(handle);
if (!ret)
ret = ret2;
return ret ? ret : copied;
}
static void ext4_da_invalidatepage(struct page *page, unsigned int offset,
unsigned int length)
{
/*
* Drop reserved blocks
*/
BUG_ON(!PageLocked(page));
if (!page_has_buffers(page))
goto out;
ext4_da_page_release_reservation(page, offset, length);
out:
ext4_invalidatepage(page, offset, length);
return;
}
/*
* Force all delayed allocation blocks to be allocated for a given inode.
*/
int ext4_alloc_da_blocks(struct inode *inode)
{
trace_ext4_alloc_da_blocks(inode);
if (!EXT4_I(inode)->i_reserved_data_blocks)
return 0;
/*
* We do something simple for now. The filemap_flush() will
* also start triggering a write of the data blocks, which is
* not strictly speaking necessary (and for users of
* laptop_mode, not even desirable). However, to do otherwise
* would require replicating code paths in:
*
* ext4_writepages() ->
* write_cache_pages() ---> (via passed in callback function)
* __mpage_da_writepage() -->
* mpage_add_bh_to_extent()
* mpage_da_map_blocks()
*
* The problem is that write_cache_pages(), located in
* mm/page-writeback.c, marks pages clean in preparation for
* doing I/O, which is not desirable if we're not planning on
* doing I/O at all.
*
* We could call write_cache_pages(), and then redirty all of
* the pages by calling redirty_page_for_writepage() but that
* would be ugly in the extreme. So instead we would need to
* replicate parts of the code in the above functions,
* simplifying them because we wouldn't actually intend to
* write out the pages, but rather only collect contiguous
* logical block extents, call the multi-block allocator, and
* then update the buffer heads with the block allocations.
*
* For now, though, we'll cheat by calling filemap_flush(),
* which will map the blocks, and start the I/O, but not
* actually wait for the I/O to complete.
*/
return filemap_flush(inode->i_mapping);
}
/*
* bmap() is special. It gets used by applications such as lilo and by
* the swapper to find the on-disk block of a specific piece of data.
*
* Naturally, this is dangerous if the block concerned is still in the
* journal. If somebody makes a swapfile on an ext4 data-journaling
* filesystem and enables swap, then they may get a nasty shock when the
* data getting swapped to that swapfile suddenly gets overwritten by
* the original zero's written out previously to the journal and
* awaiting writeback in the kernel's buffer cache.
*
* So, if we see any bmap calls here on a modified, data-journaled file,
* take extra steps to flush any blocks which might be in the cache.
*/
static sector_t ext4_bmap(struct address_space *mapping, sector_t block)
{
struct inode *inode = mapping->host;
journal_t *journal;
int err;
/*
* We can get here for an inline file via the FIBMAP ioctl
*/
if (ext4_has_inline_data(inode))
return 0;
if (mapping_tagged(mapping, PAGECACHE_TAG_DIRTY) &&
test_opt(inode->i_sb, DELALLOC)) {
/*
* With delalloc we want to sync the file
* so that we can make sure we allocate
* blocks for file
*/
filemap_write_and_wait(mapping);
}
if (EXT4_JOURNAL(inode) &&
ext4_test_inode_state(inode, EXT4_STATE_JDATA)) {
/*
* This is a REALLY heavyweight approach, but the use of
* bmap on dirty files is expected to be extremely rare:
* only if we run lilo or swapon on a freshly made file
* do we expect this to happen.
*
* (bmap requires CAP_SYS_RAWIO so this does not
* represent an unprivileged user DOS attack --- we'd be
* in trouble if mortal users could trigger this path at
* will.)
*
* NB. EXT4_STATE_JDATA is not set on files other than
* regular files. If somebody wants to bmap a directory
* or symlink and gets confused because the buffer
* hasn't yet been flushed to disk, they deserve
* everything they get.
*/
ext4_clear_inode_state(inode, EXT4_STATE_JDATA);
journal = EXT4_JOURNAL(inode);
jbd2_journal_lock_updates(journal);
err = jbd2_journal_flush(journal);
jbd2_journal_unlock_updates(journal);
if (err)
return 0;
}
return generic_block_bmap(mapping, block, ext4_get_block);
}
static int ext4_readpage(struct file *file, struct page *page)
{
int ret = -EAGAIN;
struct inode *inode = page->mapping->host;
trace_ext4_readpage(page);
if (ext4_has_inline_data(inode))
ret = ext4_readpage_inline(inode, page);
if (ret == -EAGAIN)
return mpage_readpage(page, ext4_get_block);
return ret;
}
static int
ext4_readpages(struct file *file, struct address_space *mapping,
struct list_head *pages, unsigned nr_pages)
{
struct inode *inode = mapping->host;
/* If the file has inline data, no need to do readpages. */
if (ext4_has_inline_data(inode))
return 0;
return mpage_readpages(mapping, pages, nr_pages, ext4_get_block);
}
static void ext4_invalidatepage(struct page *page, unsigned int offset,
unsigned int length)
{
trace_ext4_invalidatepage(page, offset, length);
/* No journalling happens on data buffers when this function is used */
WARN_ON(page_has_buffers(page) && buffer_jbd(page_buffers(page)));
block_invalidatepage(page, offset, length);
}
static int __ext4_journalled_invalidatepage(struct page *page,
unsigned int offset,
unsigned int length)
{
journal_t *journal = EXT4_JOURNAL(page->mapping->host);
trace_ext4_journalled_invalidatepage(page, offset, length);
/*
* If it's a full truncate we just forget about the pending dirtying
*/
if (offset == 0 && length == PAGE_CACHE_SIZE)
ClearPageChecked(page);
return jbd2_journal_invalidatepage(journal, page, offset, length);
}
/* Wrapper for aops... */
static void ext4_journalled_invalidatepage(struct page *page,
unsigned int offset,
unsigned int length)
{
WARN_ON(__ext4_journalled_invalidatepage(page, offset, length) < 0);
}
static int ext4_releasepage(struct page *page, gfp_t wait)
{
journal_t *journal = EXT4_JOURNAL(page->mapping->host);
trace_ext4_releasepage(page);
/* Page has dirty journalled data -> cannot release */
if (PageChecked(page))
return 0;
if (journal)
return jbd2_journal_try_to_free_buffers(journal, page, wait);
else
return try_to_free_buffers(page);
}
/*
* ext4_get_block used when preparing for a DIO write or buffer write.
* We allocate an uinitialized extent if blocks haven't been allocated.
* The extent will be converted to initialized after the IO is complete.
*/
int ext4_get_block_write(struct inode *inode, sector_t iblock,
struct buffer_head *bh_result, int create)
{
ext4_debug("ext4_get_block_write: inode %lu, create flag %d\n",
inode->i_ino, create);
return _ext4_get_block(inode, iblock, bh_result,
EXT4_GET_BLOCKS_IO_CREATE_EXT);
}
static int ext4_get_block_write_nolock(struct inode *inode, sector_t iblock,
struct buffer_head *bh_result, int create)
{
ext4_debug("ext4_get_block_write_nolock: inode %lu, create flag %d\n",
inode->i_ino, create);
return _ext4_get_block(inode, iblock, bh_result,
EXT4_GET_BLOCKS_NO_LOCK);
}
static void ext4_end_io_dio(struct kiocb *iocb, loff_t offset,
ssize_t size, void *private)
{
ext4_io_end_t *io_end = iocb->private;
/* if not async direct IO just return */
if (!io_end)
return;
ext_debug("ext4_end_io_dio(): io_end 0x%p "
"for inode %lu, iocb 0x%p, offset %llu, size %zd\n",
iocb->private, io_end->inode->i_ino, iocb, offset,
size);
iocb->private = NULL;
io_end->offset = offset;
io_end->size = size;
ext4_put_io_end(io_end);
}
/*
* For ext4 extent files, ext4 will do direct-io write to holes,
* preallocated extents, and those write extend the file, no need to
* fall back to buffered IO.
*
* For holes, we fallocate those blocks, mark them as unwritten
* If those blocks were preallocated, we mark sure they are split, but
* still keep the range to write as unwritten.
*
* The unwritten extents will be converted to written when DIO is completed.
* For async direct IO, since the IO may still pending when return, we
* set up an end_io call back function, which will do the conversion
* when async direct IO completed.
*
* If the O_DIRECT write will extend the file then add this inode to the
* orphan list. So recovery will truncate it back to the original size
* if the machine crashes during the write.
*
*/
static ssize_t ext4_ext_direct_IO(int rw, struct kiocb *iocb,
struct iov_iter *iter, loff_t offset)
{
struct file *file = iocb->ki_filp;
struct inode *inode = file->f_mapping->host;
ssize_t ret;
size_t count = iov_iter_count(iter);
int overwrite = 0;
get_block_t *get_block_func = NULL;
int dio_flags = 0;
loff_t final_size = offset + count;
ext4_io_end_t *io_end = NULL;
/* Use the old path for reads and writes beyond i_size. */
if (rw != WRITE || final_size > inode->i_size)
return ext4_ind_direct_IO(rw, iocb, iter, offset);
BUG_ON(iocb->private == NULL);
/*
* Make all waiters for direct IO properly wait also for extent
* conversion. This also disallows race between truncate() and
* overwrite DIO as i_dio_count needs to be incremented under i_mutex.
*/
if (rw == WRITE)
atomic_inc(&inode->i_dio_count);
/* If we do a overwrite dio, i_mutex locking can be released */
overwrite = *((int *)iocb->private);
if (overwrite) {
down_read(&EXT4_I(inode)->i_data_sem);
mutex_unlock(&inode->i_mutex);
}
/*
* We could direct write to holes and fallocate.
*
* Allocated blocks to fill the hole are marked as
* unwritten to prevent parallel buffered read to expose
* the stale data before DIO complete the data IO.
*
* As to previously fallocated extents, ext4 get_block will
* just simply mark the buffer mapped but still keep the
* extents unwritten.
*
* For non AIO case, we will convert those unwritten extents
* to written after return back from blockdev_direct_IO.
*
* For async DIO, the conversion needs to be deferred when the
* IO is completed. The ext4 end_io callback function will be
* called to take care of the conversion work. Here for async
* case, we allocate an io_end structure to hook to the iocb.
*/
iocb->private = NULL;
ext4_inode_aio_set(inode, NULL);
if (!is_sync_kiocb(iocb)) {
io_end = ext4_init_io_end(inode, GFP_NOFS);
if (!io_end) {
ret = -ENOMEM;
goto retake_lock;
}
/*
* Grab reference for DIO. Will be dropped in ext4_end_io_dio()
*/
iocb->private = ext4_get_io_end(io_end);
/*
* we save the io structure for current async direct
* IO, so that later ext4_map_blocks() could flag the
* io structure whether there is a unwritten extents
* needs to be converted when IO is completed.
*/
ext4_inode_aio_set(inode, io_end);
}
if (overwrite) {
get_block_func = ext4_get_block_write_nolock;
} else {
get_block_func = ext4_get_block_write;
dio_flags = DIO_LOCKING;
}
ret = __blockdev_direct_IO(rw, iocb, inode,
inode->i_sb->s_bdev, iter,
offset,
get_block_func,
ext4_end_io_dio,
NULL,
dio_flags);
/*
* Put our reference to io_end. This can free the io_end structure e.g.
* in sync IO case or in case of error. It can even perform extent
* conversion if all bios we submitted finished before we got here.
* Note that in that case iocb->private can be already set to NULL
* here.
*/
if (io_end) {
ext4_inode_aio_set(inode, NULL);
ext4_put_io_end(io_end);
/*
* When no IO was submitted ext4_end_io_dio() was not
* called so we have to put iocb's reference.
*/
if (ret <= 0 && ret != -EIOCBQUEUED && iocb->private) {
WARN_ON(iocb->private != io_end);
WARN_ON(io_end->flag & EXT4_IO_END_UNWRITTEN);
ext4_put_io_end(io_end);
iocb->private = NULL;
}
}
if (ret > 0 && !overwrite && ext4_test_inode_state(inode,
EXT4_STATE_DIO_UNWRITTEN)) {
int err;
/*
* for non AIO case, since the IO is already
* completed, we could do the conversion right here
*/
err = ext4_convert_unwritten_extents(NULL, inode,
offset, ret);
if (err < 0)
ret = err;
ext4_clear_inode_state(inode, EXT4_STATE_DIO_UNWRITTEN);
}
retake_lock:
if (rw == WRITE)
inode_dio_done(inode);
/* take i_mutex locking again if we do a ovewrite dio */
if (overwrite) {
up_read(&EXT4_I(inode)->i_data_sem);
mutex_lock(&inode->i_mutex);
}
return ret;
}
static ssize_t ext4_direct_IO(int rw, struct kiocb *iocb,
struct iov_iter *iter, loff_t offset)
{
struct file *file = iocb->ki_filp;
struct inode *inode = file->f_mapping->host;
size_t count = iov_iter_count(iter);
ssize_t ret;
/*
* If we are doing data journalling we don't support O_DIRECT
*/
if (ext4_should_journal_data(inode))
return 0;
/* Let buffer I/O handle the inline data case. */
if (ext4_has_inline_data(inode))
return 0;
trace_ext4_direct_IO_enter(inode, offset, count, rw);
if (ext4_test_inode_flag(inode, EXT4_INODE_EXTENTS))
ret = ext4_ext_direct_IO(rw, iocb, iter, offset);
else
ret = ext4_ind_direct_IO(rw, iocb, iter, offset);
trace_ext4_direct_IO_exit(inode, offset, count, rw, ret);
return ret;
}
/*
* Pages can be marked dirty completely asynchronously from ext4's journalling
* activity. By filemap_sync_pte(), try_to_unmap_one(), etc. We cannot do
* much here because ->set_page_dirty is called under VFS locks. The page is
* not necessarily locked.
*
* We cannot just dirty the page and leave attached buffers clean, because the
* buffers' dirty state is "definitive". We cannot just set the buffers dirty
* or jbddirty because all the journalling code will explode.
*
* So what we do is to mark the page "pending dirty" and next time writepage
* is called, propagate that into the buffers appropriately.
*/
static int ext4_journalled_set_page_dirty(struct page *page)
{
SetPageChecked(page);
return __set_page_dirty_nobuffers(page);
}
static const struct address_space_operations ext4_aops = {
vfs: pagecache usage optimization for pagesize!=blocksize When we read some part of a file through pagecache, if there is a pagecache of corresponding index but this page is not uptodate, read IO is issued and this page will be uptodate. I think this is good for pagesize == blocksize environment but there is room for improvement on pagesize != blocksize environment. Because in this case a page can have multiple buffers and even if a page is not uptodate, some buffers can be uptodate. So I suggest that when all buffers which correspond to a part of a file that we want to read are uptodate, use this pagecache and copy data from this pagecache to user buffer even if a page is not uptodate. This can reduce read IO and improve system throughput. I wrote a benchmark program and got result number with this program. This benchmark do: 1: mount and open a test file. 2: create a 512MB file. 3: close a file and umount. 4: mount and again open a test file. 5: pwrite randomly 300000 times on a test file. offset is aligned by IO size(1024bytes). 6: measure time of preading randomly 100000 times on a test file. The result was: 2.6.26 330 sec 2.6.26-patched 226 sec Arch:i386 Filesystem:ext3 Blocksize:1024 bytes Memory: 1GB On ext3/4, a file is written through buffer/block. So random read/write mixed workloads or random read after random write workloads are optimized with this patch under pagesize != blocksize environment. This test result showed this. The benchmark program is as follows: #include <stdio.h> #include <sys/types.h> #include <sys/stat.h> #include <fcntl.h> #include <unistd.h> #include <time.h> #include <stdlib.h> #include <string.h> #include <sys/mount.h> #define LEN 1024 #define LOOP 1024*512 /* 512MB */ main(void) { unsigned long i, offset, filesize; int fd; char buf[LEN]; time_t t1, t2; if (mount("/dev/sda1", "/root/test1/", "ext3", 0, 0) < 0) { perror("cannot mount\n"); exit(1); } memset(buf, 0, LEN); fd = open("/root/test1/testfile", O_CREAT|O_RDWR|O_TRUNC); if (fd < 0) { perror("cannot open file\n"); exit(1); } for (i = 0; i < LOOP; i++) write(fd, buf, LEN); close(fd); if (umount("/root/test1/") < 0) { perror("cannot umount\n"); exit(1); } if (mount("/dev/sda1", "/root/test1/", "ext3", 0, 0) < 0) { perror("cannot mount\n"); exit(1); } fd = open("/root/test1/testfile", O_RDWR); if (fd < 0) { perror("cannot open file\n"); exit(1); } filesize = LEN * LOOP; for (i = 0; i < 300000; i++){ offset = (random() % filesize) & (~(LEN - 1)); pwrite(fd, buf, LEN, offset); } printf("start test\n"); time(&t1); for (i = 0; i < 100000; i++){ offset = (random() % filesize) & (~(LEN - 1)); pread(fd, buf, LEN, offset); } time(&t2); printf("%ld sec\n", t2-t1); close(fd); if (umount("/root/test1/") < 0) { perror("cannot umount\n"); exit(1); } } Signed-off-by: Hisashi Hifumi <hifumi.hisashi@oss.ntt.co.jp> Cc: Nick Piggin <nickpiggin@yahoo.com.au> Cc: Christoph Hellwig <hch@infradead.org> Cc: Jan Kara <jack@ucw.cz> Cc: <linux-ext4@vger.kernel.org> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2008-07-28 22:46:36 +00:00
.readpage = ext4_readpage,
.readpages = ext4_readpages,
.writepage = ext4_writepage,
.writepages = ext4_writepages,
vfs: pagecache usage optimization for pagesize!=blocksize When we read some part of a file through pagecache, if there is a pagecache of corresponding index but this page is not uptodate, read IO is issued and this page will be uptodate. I think this is good for pagesize == blocksize environment but there is room for improvement on pagesize != blocksize environment. Because in this case a page can have multiple buffers and even if a page is not uptodate, some buffers can be uptodate. So I suggest that when all buffers which correspond to a part of a file that we want to read are uptodate, use this pagecache and copy data from this pagecache to user buffer even if a page is not uptodate. This can reduce read IO and improve system throughput. I wrote a benchmark program and got result number with this program. This benchmark do: 1: mount and open a test file. 2: create a 512MB file. 3: close a file and umount. 4: mount and again open a test file. 5: pwrite randomly 300000 times on a test file. offset is aligned by IO size(1024bytes). 6: measure time of preading randomly 100000 times on a test file. The result was: 2.6.26 330 sec 2.6.26-patched 226 sec Arch:i386 Filesystem:ext3 Blocksize:1024 bytes Memory: 1GB On ext3/4, a file is written through buffer/block. So random read/write mixed workloads or random read after random write workloads are optimized with this patch under pagesize != blocksize environment. This test result showed this. The benchmark program is as follows: #include <stdio.h> #include <sys/types.h> #include <sys/stat.h> #include <fcntl.h> #include <unistd.h> #include <time.h> #include <stdlib.h> #include <string.h> #include <sys/mount.h> #define LEN 1024 #define LOOP 1024*512 /* 512MB */ main(void) { unsigned long i, offset, filesize; int fd; char buf[LEN]; time_t t1, t2; if (mount("/dev/sda1", "/root/test1/", "ext3", 0, 0) < 0) { perror("cannot mount\n"); exit(1); } memset(buf, 0, LEN); fd = open("/root/test1/testfile", O_CREAT|O_RDWR|O_TRUNC); if (fd < 0) { perror("cannot open file\n"); exit(1); } for (i = 0; i < LOOP; i++) write(fd, buf, LEN); close(fd); if (umount("/root/test1/") < 0) { perror("cannot umount\n"); exit(1); } if (mount("/dev/sda1", "/root/test1/", "ext3", 0, 0) < 0) { perror("cannot mount\n"); exit(1); } fd = open("/root/test1/testfile", O_RDWR); if (fd < 0) { perror("cannot open file\n"); exit(1); } filesize = LEN * LOOP; for (i = 0; i < 300000; i++){ offset = (random() % filesize) & (~(LEN - 1)); pwrite(fd, buf, LEN, offset); } printf("start test\n"); time(&t1); for (i = 0; i < 100000; i++){ offset = (random() % filesize) & (~(LEN - 1)); pread(fd, buf, LEN, offset); } time(&t2); printf("%ld sec\n", t2-t1); close(fd); if (umount("/root/test1/") < 0) { perror("cannot umount\n"); exit(1); } } Signed-off-by: Hisashi Hifumi <hifumi.hisashi@oss.ntt.co.jp> Cc: Nick Piggin <nickpiggin@yahoo.com.au> Cc: Christoph Hellwig <hch@infradead.org> Cc: Jan Kara <jack@ucw.cz> Cc: <linux-ext4@vger.kernel.org> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2008-07-28 22:46:36 +00:00
.write_begin = ext4_write_begin,
.write_end = ext4_write_end,
vfs: pagecache usage optimization for pagesize!=blocksize When we read some part of a file through pagecache, if there is a pagecache of corresponding index but this page is not uptodate, read IO is issued and this page will be uptodate. I think this is good for pagesize == blocksize environment but there is room for improvement on pagesize != blocksize environment. Because in this case a page can have multiple buffers and even if a page is not uptodate, some buffers can be uptodate. So I suggest that when all buffers which correspond to a part of a file that we want to read are uptodate, use this pagecache and copy data from this pagecache to user buffer even if a page is not uptodate. This can reduce read IO and improve system throughput. I wrote a benchmark program and got result number with this program. This benchmark do: 1: mount and open a test file. 2: create a 512MB file. 3: close a file and umount. 4: mount and again open a test file. 5: pwrite randomly 300000 times on a test file. offset is aligned by IO size(1024bytes). 6: measure time of preading randomly 100000 times on a test file. The result was: 2.6.26 330 sec 2.6.26-patched 226 sec Arch:i386 Filesystem:ext3 Blocksize:1024 bytes Memory: 1GB On ext3/4, a file is written through buffer/block. So random read/write mixed workloads or random read after random write workloads are optimized with this patch under pagesize != blocksize environment. This test result showed this. The benchmark program is as follows: #include <stdio.h> #include <sys/types.h> #include <sys/stat.h> #include <fcntl.h> #include <unistd.h> #include <time.h> #include <stdlib.h> #include <string.h> #include <sys/mount.h> #define LEN 1024 #define LOOP 1024*512 /* 512MB */ main(void) { unsigned long i, offset, filesize; int fd; char buf[LEN]; time_t t1, t2; if (mount("/dev/sda1", "/root/test1/", "ext3", 0, 0) < 0) { perror("cannot mount\n"); exit(1); } memset(buf, 0, LEN); fd = open("/root/test1/testfile", O_CREAT|O_RDWR|O_TRUNC); if (fd < 0) { perror("cannot open file\n"); exit(1); } for (i = 0; i < LOOP; i++) write(fd, buf, LEN); close(fd); if (umount("/root/test1/") < 0) { perror("cannot umount\n"); exit(1); } if (mount("/dev/sda1", "/root/test1/", "ext3", 0, 0) < 0) { perror("cannot mount\n"); exit(1); } fd = open("/root/test1/testfile", O_RDWR); if (fd < 0) { perror("cannot open file\n"); exit(1); } filesize = LEN * LOOP; for (i = 0; i < 300000; i++){ offset = (random() % filesize) & (~(LEN - 1)); pwrite(fd, buf, LEN, offset); } printf("start test\n"); time(&t1); for (i = 0; i < 100000; i++){ offset = (random() % filesize) & (~(LEN - 1)); pread(fd, buf, LEN, offset); } time(&t2); printf("%ld sec\n", t2-t1); close(fd); if (umount("/root/test1/") < 0) { perror("cannot umount\n"); exit(1); } } Signed-off-by: Hisashi Hifumi <hifumi.hisashi@oss.ntt.co.jp> Cc: Nick Piggin <nickpiggin@yahoo.com.au> Cc: Christoph Hellwig <hch@infradead.org> Cc: Jan Kara <jack@ucw.cz> Cc: <linux-ext4@vger.kernel.org> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2008-07-28 22:46:36 +00:00
.bmap = ext4_bmap,
.invalidatepage = ext4_invalidatepage,
.releasepage = ext4_releasepage,
.direct_IO = ext4_direct_IO,
.migratepage = buffer_migrate_page,
.is_partially_uptodate = block_is_partially_uptodate,
.error_remove_page = generic_error_remove_page,
};
static const struct address_space_operations ext4_journalled_aops = {
vfs: pagecache usage optimization for pagesize!=blocksize When we read some part of a file through pagecache, if there is a pagecache of corresponding index but this page is not uptodate, read IO is issued and this page will be uptodate. I think this is good for pagesize == blocksize environment but there is room for improvement on pagesize != blocksize environment. Because in this case a page can have multiple buffers and even if a page is not uptodate, some buffers can be uptodate. So I suggest that when all buffers which correspond to a part of a file that we want to read are uptodate, use this pagecache and copy data from this pagecache to user buffer even if a page is not uptodate. This can reduce read IO and improve system throughput. I wrote a benchmark program and got result number with this program. This benchmark do: 1: mount and open a test file. 2: create a 512MB file. 3: close a file and umount. 4: mount and again open a test file. 5: pwrite randomly 300000 times on a test file. offset is aligned by IO size(1024bytes). 6: measure time of preading randomly 100000 times on a test file. The result was: 2.6.26 330 sec 2.6.26-patched 226 sec Arch:i386 Filesystem:ext3 Blocksize:1024 bytes Memory: 1GB On ext3/4, a file is written through buffer/block. So random read/write mixed workloads or random read after random write workloads are optimized with this patch under pagesize != blocksize environment. This test result showed this. The benchmark program is as follows: #include <stdio.h> #include <sys/types.h> #include <sys/stat.h> #include <fcntl.h> #include <unistd.h> #include <time.h> #include <stdlib.h> #include <string.h> #include <sys/mount.h> #define LEN 1024 #define LOOP 1024*512 /* 512MB */ main(void) { unsigned long i, offset, filesize; int fd; char buf[LEN]; time_t t1, t2; if (mount("/dev/sda1", "/root/test1/", "ext3", 0, 0) < 0) { perror("cannot mount\n"); exit(1); } memset(buf, 0, LEN); fd = open("/root/test1/testfile", O_CREAT|O_RDWR|O_TRUNC); if (fd < 0) { perror("cannot open file\n"); exit(1); } for (i = 0; i < LOOP; i++) write(fd, buf, LEN); close(fd); if (umount("/root/test1/") < 0) { perror("cannot umount\n"); exit(1); } if (mount("/dev/sda1", "/root/test1/", "ext3", 0, 0) < 0) { perror("cannot mount\n"); exit(1); } fd = open("/root/test1/testfile", O_RDWR); if (fd < 0) { perror("cannot open file\n"); exit(1); } filesize = LEN * LOOP; for (i = 0; i < 300000; i++){ offset = (random() % filesize) & (~(LEN - 1)); pwrite(fd, buf, LEN, offset); } printf("start test\n"); time(&t1); for (i = 0; i < 100000; i++){ offset = (random() % filesize) & (~(LEN - 1)); pread(fd, buf, LEN, offset); } time(&t2); printf("%ld sec\n", t2-t1); close(fd); if (umount("/root/test1/") < 0) { perror("cannot umount\n"); exit(1); } } Signed-off-by: Hisashi Hifumi <hifumi.hisashi@oss.ntt.co.jp> Cc: Nick Piggin <nickpiggin@yahoo.com.au> Cc: Christoph Hellwig <hch@infradead.org> Cc: Jan Kara <jack@ucw.cz> Cc: <linux-ext4@vger.kernel.org> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2008-07-28 22:46:36 +00:00
.readpage = ext4_readpage,
.readpages = ext4_readpages,
.writepage = ext4_writepage,
.writepages = ext4_writepages,
vfs: pagecache usage optimization for pagesize!=blocksize When we read some part of a file through pagecache, if there is a pagecache of corresponding index but this page is not uptodate, read IO is issued and this page will be uptodate. I think this is good for pagesize == blocksize environment but there is room for improvement on pagesize != blocksize environment. Because in this case a page can have multiple buffers and even if a page is not uptodate, some buffers can be uptodate. So I suggest that when all buffers which correspond to a part of a file that we want to read are uptodate, use this pagecache and copy data from this pagecache to user buffer even if a page is not uptodate. This can reduce read IO and improve system throughput. I wrote a benchmark program and got result number with this program. This benchmark do: 1: mount and open a test file. 2: create a 512MB file. 3: close a file and umount. 4: mount and again open a test file. 5: pwrite randomly 300000 times on a test file. offset is aligned by IO size(1024bytes). 6: measure time of preading randomly 100000 times on a test file. The result was: 2.6.26 330 sec 2.6.26-patched 226 sec Arch:i386 Filesystem:ext3 Blocksize:1024 bytes Memory: 1GB On ext3/4, a file is written through buffer/block. So random read/write mixed workloads or random read after random write workloads are optimized with this patch under pagesize != blocksize environment. This test result showed this. The benchmark program is as follows: #include <stdio.h> #include <sys/types.h> #include <sys/stat.h> #include <fcntl.h> #include <unistd.h> #include <time.h> #include <stdlib.h> #include <string.h> #include <sys/mount.h> #define LEN 1024 #define LOOP 1024*512 /* 512MB */ main(void) { unsigned long i, offset, filesize; int fd; char buf[LEN]; time_t t1, t2; if (mount("/dev/sda1", "/root/test1/", "ext3", 0, 0) < 0) { perror("cannot mount\n"); exit(1); } memset(buf, 0, LEN); fd = open("/root/test1/testfile", O_CREAT|O_RDWR|O_TRUNC); if (fd < 0) { perror("cannot open file\n"); exit(1); } for (i = 0; i < LOOP; i++) write(fd, buf, LEN); close(fd); if (umount("/root/test1/") < 0) { perror("cannot umount\n"); exit(1); } if (mount("/dev/sda1", "/root/test1/", "ext3", 0, 0) < 0) { perror("cannot mount\n"); exit(1); } fd = open("/root/test1/testfile", O_RDWR); if (fd < 0) { perror("cannot open file\n"); exit(1); } filesize = LEN * LOOP; for (i = 0; i < 300000; i++){ offset = (random() % filesize) & (~(LEN - 1)); pwrite(fd, buf, LEN, offset); } printf("start test\n"); time(&t1); for (i = 0; i < 100000; i++){ offset = (random() % filesize) & (~(LEN - 1)); pread(fd, buf, LEN, offset); } time(&t2); printf("%ld sec\n", t2-t1); close(fd); if (umount("/root/test1/") < 0) { perror("cannot umount\n"); exit(1); } } Signed-off-by: Hisashi Hifumi <hifumi.hisashi@oss.ntt.co.jp> Cc: Nick Piggin <nickpiggin@yahoo.com.au> Cc: Christoph Hellwig <hch@infradead.org> Cc: Jan Kara <jack@ucw.cz> Cc: <linux-ext4@vger.kernel.org> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2008-07-28 22:46:36 +00:00
.write_begin = ext4_write_begin,
.write_end = ext4_journalled_write_end,
.set_page_dirty = ext4_journalled_set_page_dirty,
.bmap = ext4_bmap,
.invalidatepage = ext4_journalled_invalidatepage,
vfs: pagecache usage optimization for pagesize!=blocksize When we read some part of a file through pagecache, if there is a pagecache of corresponding index but this page is not uptodate, read IO is issued and this page will be uptodate. I think this is good for pagesize == blocksize environment but there is room for improvement on pagesize != blocksize environment. Because in this case a page can have multiple buffers and even if a page is not uptodate, some buffers can be uptodate. So I suggest that when all buffers which correspond to a part of a file that we want to read are uptodate, use this pagecache and copy data from this pagecache to user buffer even if a page is not uptodate. This can reduce read IO and improve system throughput. I wrote a benchmark program and got result number with this program. This benchmark do: 1: mount and open a test file. 2: create a 512MB file. 3: close a file and umount. 4: mount and again open a test file. 5: pwrite randomly 300000 times on a test file. offset is aligned by IO size(1024bytes). 6: measure time of preading randomly 100000 times on a test file. The result was: 2.6.26 330 sec 2.6.26-patched 226 sec Arch:i386 Filesystem:ext3 Blocksize:1024 bytes Memory: 1GB On ext3/4, a file is written through buffer/block. So random read/write mixed workloads or random read after random write workloads are optimized with this patch under pagesize != blocksize environment. This test result showed this. The benchmark program is as follows: #include <stdio.h> #include <sys/types.h> #include <sys/stat.h> #include <fcntl.h> #include <unistd.h> #include <time.h> #include <stdlib.h> #include <string.h> #include <sys/mount.h> #define LEN 1024 #define LOOP 1024*512 /* 512MB */ main(void) { unsigned long i, offset, filesize; int fd; char buf[LEN]; time_t t1, t2; if (mount("/dev/sda1", "/root/test1/", "ext3", 0, 0) < 0) { perror("cannot mount\n"); exit(1); } memset(buf, 0, LEN); fd = open("/root/test1/testfile", O_CREAT|O_RDWR|O_TRUNC); if (fd < 0) { perror("cannot open file\n"); exit(1); } for (i = 0; i < LOOP; i++) write(fd, buf, LEN); close(fd); if (umount("/root/test1/") < 0) { perror("cannot umount\n"); exit(1); } if (mount("/dev/sda1", "/root/test1/", "ext3", 0, 0) < 0) { perror("cannot mount\n"); exit(1); } fd = open("/root/test1/testfile", O_RDWR); if (fd < 0) { perror("cannot open file\n"); exit(1); } filesize = LEN * LOOP; for (i = 0; i < 300000; i++){ offset = (random() % filesize) & (~(LEN - 1)); pwrite(fd, buf, LEN, offset); } printf("start test\n"); time(&t1); for (i = 0; i < 100000; i++){ offset = (random() % filesize) & (~(LEN - 1)); pread(fd, buf, LEN, offset); } time(&t2); printf("%ld sec\n", t2-t1); close(fd); if (umount("/root/test1/") < 0) { perror("cannot umount\n"); exit(1); } } Signed-off-by: Hisashi Hifumi <hifumi.hisashi@oss.ntt.co.jp> Cc: Nick Piggin <nickpiggin@yahoo.com.au> Cc: Christoph Hellwig <hch@infradead.org> Cc: Jan Kara <jack@ucw.cz> Cc: <linux-ext4@vger.kernel.org> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2008-07-28 22:46:36 +00:00
.releasepage = ext4_releasepage,
.direct_IO = ext4_direct_IO,
vfs: pagecache usage optimization for pagesize!=blocksize When we read some part of a file through pagecache, if there is a pagecache of corresponding index but this page is not uptodate, read IO is issued and this page will be uptodate. I think this is good for pagesize == blocksize environment but there is room for improvement on pagesize != blocksize environment. Because in this case a page can have multiple buffers and even if a page is not uptodate, some buffers can be uptodate. So I suggest that when all buffers which correspond to a part of a file that we want to read are uptodate, use this pagecache and copy data from this pagecache to user buffer even if a page is not uptodate. This can reduce read IO and improve system throughput. I wrote a benchmark program and got result number with this program. This benchmark do: 1: mount and open a test file. 2: create a 512MB file. 3: close a file and umount. 4: mount and again open a test file. 5: pwrite randomly 300000 times on a test file. offset is aligned by IO size(1024bytes). 6: measure time of preading randomly 100000 times on a test file. The result was: 2.6.26 330 sec 2.6.26-patched 226 sec Arch:i386 Filesystem:ext3 Blocksize:1024 bytes Memory: 1GB On ext3/4, a file is written through buffer/block. So random read/write mixed workloads or random read after random write workloads are optimized with this patch under pagesize != blocksize environment. This test result showed this. The benchmark program is as follows: #include <stdio.h> #include <sys/types.h> #include <sys/stat.h> #include <fcntl.h> #include <unistd.h> #include <time.h> #include <stdlib.h> #include <string.h> #include <sys/mount.h> #define LEN 1024 #define LOOP 1024*512 /* 512MB */ main(void) { unsigned long i, offset, filesize; int fd; char buf[LEN]; time_t t1, t2; if (mount("/dev/sda1", "/root/test1/", "ext3", 0, 0) < 0) { perror("cannot mount\n"); exit(1); } memset(buf, 0, LEN); fd = open("/root/test1/testfile", O_CREAT|O_RDWR|O_TRUNC); if (fd < 0) { perror("cannot open file\n"); exit(1); } for (i = 0; i < LOOP; i++) write(fd, buf, LEN); close(fd); if (umount("/root/test1/") < 0) { perror("cannot umount\n"); exit(1); } if (mount("/dev/sda1", "/root/test1/", "ext3", 0, 0) < 0) { perror("cannot mount\n"); exit(1); } fd = open("/root/test1/testfile", O_RDWR); if (fd < 0) { perror("cannot open file\n"); exit(1); } filesize = LEN * LOOP; for (i = 0; i < 300000; i++){ offset = (random() % filesize) & (~(LEN - 1)); pwrite(fd, buf, LEN, offset); } printf("start test\n"); time(&t1); for (i = 0; i < 100000; i++){ offset = (random() % filesize) & (~(LEN - 1)); pread(fd, buf, LEN, offset); } time(&t2); printf("%ld sec\n", t2-t1); close(fd); if (umount("/root/test1/") < 0) { perror("cannot umount\n"); exit(1); } } Signed-off-by: Hisashi Hifumi <hifumi.hisashi@oss.ntt.co.jp> Cc: Nick Piggin <nickpiggin@yahoo.com.au> Cc: Christoph Hellwig <hch@infradead.org> Cc: Jan Kara <jack@ucw.cz> Cc: <linux-ext4@vger.kernel.org> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2008-07-28 22:46:36 +00:00
.is_partially_uptodate = block_is_partially_uptodate,
.error_remove_page = generic_error_remove_page,
};
static const struct address_space_operations ext4_da_aops = {
vfs: pagecache usage optimization for pagesize!=blocksize When we read some part of a file through pagecache, if there is a pagecache of corresponding index but this page is not uptodate, read IO is issued and this page will be uptodate. I think this is good for pagesize == blocksize environment but there is room for improvement on pagesize != blocksize environment. Because in this case a page can have multiple buffers and even if a page is not uptodate, some buffers can be uptodate. So I suggest that when all buffers which correspond to a part of a file that we want to read are uptodate, use this pagecache and copy data from this pagecache to user buffer even if a page is not uptodate. This can reduce read IO and improve system throughput. I wrote a benchmark program and got result number with this program. This benchmark do: 1: mount and open a test file. 2: create a 512MB file. 3: close a file and umount. 4: mount and again open a test file. 5: pwrite randomly 300000 times on a test file. offset is aligned by IO size(1024bytes). 6: measure time of preading randomly 100000 times on a test file. The result was: 2.6.26 330 sec 2.6.26-patched 226 sec Arch:i386 Filesystem:ext3 Blocksize:1024 bytes Memory: 1GB On ext3/4, a file is written through buffer/block. So random read/write mixed workloads or random read after random write workloads are optimized with this patch under pagesize != blocksize environment. This test result showed this. The benchmark program is as follows: #include <stdio.h> #include <sys/types.h> #include <sys/stat.h> #include <fcntl.h> #include <unistd.h> #include <time.h> #include <stdlib.h> #include <string.h> #include <sys/mount.h> #define LEN 1024 #define LOOP 1024*512 /* 512MB */ main(void) { unsigned long i, offset, filesize; int fd; char buf[LEN]; time_t t1, t2; if (mount("/dev/sda1", "/root/test1/", "ext3", 0, 0) < 0) { perror("cannot mount\n"); exit(1); } memset(buf, 0, LEN); fd = open("/root/test1/testfile", O_CREAT|O_RDWR|O_TRUNC); if (fd < 0) { perror("cannot open file\n"); exit(1); } for (i = 0; i < LOOP; i++) write(fd, buf, LEN); close(fd); if (umount("/root/test1/") < 0) { perror("cannot umount\n"); exit(1); } if (mount("/dev/sda1", "/root/test1/", "ext3", 0, 0) < 0) { perror("cannot mount\n"); exit(1); } fd = open("/root/test1/testfile", O_RDWR); if (fd < 0) { perror("cannot open file\n"); exit(1); } filesize = LEN * LOOP; for (i = 0; i < 300000; i++){ offset = (random() % filesize) & (~(LEN - 1)); pwrite(fd, buf, LEN, offset); } printf("start test\n"); time(&t1); for (i = 0; i < 100000; i++){ offset = (random() % filesize) & (~(LEN - 1)); pread(fd, buf, LEN, offset); } time(&t2); printf("%ld sec\n", t2-t1); close(fd); if (umount("/root/test1/") < 0) { perror("cannot umount\n"); exit(1); } } Signed-off-by: Hisashi Hifumi <hifumi.hisashi@oss.ntt.co.jp> Cc: Nick Piggin <nickpiggin@yahoo.com.au> Cc: Christoph Hellwig <hch@infradead.org> Cc: Jan Kara <jack@ucw.cz> Cc: <linux-ext4@vger.kernel.org> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2008-07-28 22:46:36 +00:00
.readpage = ext4_readpage,
.readpages = ext4_readpages,
.writepage = ext4_writepage,
.writepages = ext4_writepages,
vfs: pagecache usage optimization for pagesize!=blocksize When we read some part of a file through pagecache, if there is a pagecache of corresponding index but this page is not uptodate, read IO is issued and this page will be uptodate. I think this is good for pagesize == blocksize environment but there is room for improvement on pagesize != blocksize environment. Because in this case a page can have multiple buffers and even if a page is not uptodate, some buffers can be uptodate. So I suggest that when all buffers which correspond to a part of a file that we want to read are uptodate, use this pagecache and copy data from this pagecache to user buffer even if a page is not uptodate. This can reduce read IO and improve system throughput. I wrote a benchmark program and got result number with this program. This benchmark do: 1: mount and open a test file. 2: create a 512MB file. 3: close a file and umount. 4: mount and again open a test file. 5: pwrite randomly 300000 times on a test file. offset is aligned by IO size(1024bytes). 6: measure time of preading randomly 100000 times on a test file. The result was: 2.6.26 330 sec 2.6.26-patched 226 sec Arch:i386 Filesystem:ext3 Blocksize:1024 bytes Memory: 1GB On ext3/4, a file is written through buffer/block. So random read/write mixed workloads or random read after random write workloads are optimized with this patch under pagesize != blocksize environment. This test result showed this. The benchmark program is as follows: #include <stdio.h> #include <sys/types.h> #include <sys/stat.h> #include <fcntl.h> #include <unistd.h> #include <time.h> #include <stdlib.h> #include <string.h> #include <sys/mount.h> #define LEN 1024 #define LOOP 1024*512 /* 512MB */ main(void) { unsigned long i, offset, filesize; int fd; char buf[LEN]; time_t t1, t2; if (mount("/dev/sda1", "/root/test1/", "ext3", 0, 0) < 0) { perror("cannot mount\n"); exit(1); } memset(buf, 0, LEN); fd = open("/root/test1/testfile", O_CREAT|O_RDWR|O_TRUNC); if (fd < 0) { perror("cannot open file\n"); exit(1); } for (i = 0; i < LOOP; i++) write(fd, buf, LEN); close(fd); if (umount("/root/test1/") < 0) { perror("cannot umount\n"); exit(1); } if (mount("/dev/sda1", "/root/test1/", "ext3", 0, 0) < 0) { perror("cannot mount\n"); exit(1); } fd = open("/root/test1/testfile", O_RDWR); if (fd < 0) { perror("cannot open file\n"); exit(1); } filesize = LEN * LOOP; for (i = 0; i < 300000; i++){ offset = (random() % filesize) & (~(LEN - 1)); pwrite(fd, buf, LEN, offset); } printf("start test\n"); time(&t1); for (i = 0; i < 100000; i++){ offset = (random() % filesize) & (~(LEN - 1)); pread(fd, buf, LEN, offset); } time(&t2); printf("%ld sec\n", t2-t1); close(fd); if (umount("/root/test1/") < 0) { perror("cannot umount\n"); exit(1); } } Signed-off-by: Hisashi Hifumi <hifumi.hisashi@oss.ntt.co.jp> Cc: Nick Piggin <nickpiggin@yahoo.com.au> Cc: Christoph Hellwig <hch@infradead.org> Cc: Jan Kara <jack@ucw.cz> Cc: <linux-ext4@vger.kernel.org> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2008-07-28 22:46:36 +00:00
.write_begin = ext4_da_write_begin,
.write_end = ext4_da_write_end,
.bmap = ext4_bmap,
.invalidatepage = ext4_da_invalidatepage,
.releasepage = ext4_releasepage,
.direct_IO = ext4_direct_IO,
.migratepage = buffer_migrate_page,
.is_partially_uptodate = block_is_partially_uptodate,
.error_remove_page = generic_error_remove_page,
};
void ext4_set_aops(struct inode *inode)
{
ext4: ignore EXT4_INODE_JOURNAL_DATA flag with delalloc Ext4 does not support data journalling with delayed allocation enabled. We even do not allow to mount the file system with delayed allocation and data journalling enabled, however it can be set via FS_IOC_SETFLAGS so we can hit the inode with EXT4_INODE_JOURNAL_DATA set even on file system mounted with delayed allocation (default) and that's where problem arises. The easies way to reproduce this problem is with the following set of commands: mkfs.ext4 /dev/sdd mount /dev/sdd /mnt/test1 dd if=/dev/zero of=/mnt/test1/file bs=1M count=4 chattr +j /mnt/test1/file dd if=/dev/zero of=/mnt/test1/file bs=1M count=4 conv=notrunc chattr -j /mnt/test1/file Additionally it can be reproduced quite reliably with xfstests 272 and 269. In fact the above reproducer is a part of test 272. To fix this we should ignore the EXT4_INODE_JOURNAL_DATA inode flag if the file system is mounted with delayed allocation. This can be easily done by fixing ext4_should_*_data() functions do ignore data journal flag when delalloc is set (suggested by Ted). We also have to set the appropriate address space operations for the inode (again, ignoring data journal flag if delalloc enabled). Additionally this commit introduces ext4_inode_journal_mode() function because ext4_should_*_data() has already had a lot of common code and this change is putting it all into one function so it is easier to read. Successfully tested with xfstests in following configurations: delalloc + data=ordered delalloc + data=writeback data=journal nodelalloc + data=ordered nodelalloc + data=writeback nodelalloc + data=journal Signed-off-by: Lukas Czerner <lczerner@redhat.com> Signed-off-by: "Theodore Ts'o" <tytso@mit.edu> Cc: stable@vger.kernel.org
2012-02-20 22:53:00 +00:00
switch (ext4_inode_journal_mode(inode)) {
case EXT4_INODE_ORDERED_DATA_MODE:
ext4_set_inode_state(inode, EXT4_STATE_ORDERED_MODE);
ext4: ignore EXT4_INODE_JOURNAL_DATA flag with delalloc Ext4 does not support data journalling with delayed allocation enabled. We even do not allow to mount the file system with delayed allocation and data journalling enabled, however it can be set via FS_IOC_SETFLAGS so we can hit the inode with EXT4_INODE_JOURNAL_DATA set even on file system mounted with delayed allocation (default) and that's where problem arises. The easies way to reproduce this problem is with the following set of commands: mkfs.ext4 /dev/sdd mount /dev/sdd /mnt/test1 dd if=/dev/zero of=/mnt/test1/file bs=1M count=4 chattr +j /mnt/test1/file dd if=/dev/zero of=/mnt/test1/file bs=1M count=4 conv=notrunc chattr -j /mnt/test1/file Additionally it can be reproduced quite reliably with xfstests 272 and 269. In fact the above reproducer is a part of test 272. To fix this we should ignore the EXT4_INODE_JOURNAL_DATA inode flag if the file system is mounted with delayed allocation. This can be easily done by fixing ext4_should_*_data() functions do ignore data journal flag when delalloc is set (suggested by Ted). We also have to set the appropriate address space operations for the inode (again, ignoring data journal flag if delalloc enabled). Additionally this commit introduces ext4_inode_journal_mode() function because ext4_should_*_data() has already had a lot of common code and this change is putting it all into one function so it is easier to read. Successfully tested with xfstests in following configurations: delalloc + data=ordered delalloc + data=writeback data=journal nodelalloc + data=ordered nodelalloc + data=writeback nodelalloc + data=journal Signed-off-by: Lukas Czerner <lczerner@redhat.com> Signed-off-by: "Theodore Ts'o" <tytso@mit.edu> Cc: stable@vger.kernel.org
2012-02-20 22:53:00 +00:00
break;
case EXT4_INODE_WRITEBACK_DATA_MODE:
ext4_clear_inode_state(inode, EXT4_STATE_ORDERED_MODE);
ext4: ignore EXT4_INODE_JOURNAL_DATA flag with delalloc Ext4 does not support data journalling with delayed allocation enabled. We even do not allow to mount the file system with delayed allocation and data journalling enabled, however it can be set via FS_IOC_SETFLAGS so we can hit the inode with EXT4_INODE_JOURNAL_DATA set even on file system mounted with delayed allocation (default) and that's where problem arises. The easies way to reproduce this problem is with the following set of commands: mkfs.ext4 /dev/sdd mount /dev/sdd /mnt/test1 dd if=/dev/zero of=/mnt/test1/file bs=1M count=4 chattr +j /mnt/test1/file dd if=/dev/zero of=/mnt/test1/file bs=1M count=4 conv=notrunc chattr -j /mnt/test1/file Additionally it can be reproduced quite reliably with xfstests 272 and 269. In fact the above reproducer is a part of test 272. To fix this we should ignore the EXT4_INODE_JOURNAL_DATA inode flag if the file system is mounted with delayed allocation. This can be easily done by fixing ext4_should_*_data() functions do ignore data journal flag when delalloc is set (suggested by Ted). We also have to set the appropriate address space operations for the inode (again, ignoring data journal flag if delalloc enabled). Additionally this commit introduces ext4_inode_journal_mode() function because ext4_should_*_data() has already had a lot of common code and this change is putting it all into one function so it is easier to read. Successfully tested with xfstests in following configurations: delalloc + data=ordered delalloc + data=writeback data=journal nodelalloc + data=ordered nodelalloc + data=writeback nodelalloc + data=journal Signed-off-by: Lukas Czerner <lczerner@redhat.com> Signed-off-by: "Theodore Ts'o" <tytso@mit.edu> Cc: stable@vger.kernel.org
2012-02-20 22:53:00 +00:00
break;
case EXT4_INODE_JOURNAL_DATA_MODE:
inode->i_mapping->a_ops = &ext4_journalled_aops;
return;
ext4: ignore EXT4_INODE_JOURNAL_DATA flag with delalloc Ext4 does not support data journalling with delayed allocation enabled. We even do not allow to mount the file system with delayed allocation and data journalling enabled, however it can be set via FS_IOC_SETFLAGS so we can hit the inode with EXT4_INODE_JOURNAL_DATA set even on file system mounted with delayed allocation (default) and that's where problem arises. The easies way to reproduce this problem is with the following set of commands: mkfs.ext4 /dev/sdd mount /dev/sdd /mnt/test1 dd if=/dev/zero of=/mnt/test1/file bs=1M count=4 chattr +j /mnt/test1/file dd if=/dev/zero of=/mnt/test1/file bs=1M count=4 conv=notrunc chattr -j /mnt/test1/file Additionally it can be reproduced quite reliably with xfstests 272 and 269. In fact the above reproducer is a part of test 272. To fix this we should ignore the EXT4_INODE_JOURNAL_DATA inode flag if the file system is mounted with delayed allocation. This can be easily done by fixing ext4_should_*_data() functions do ignore data journal flag when delalloc is set (suggested by Ted). We also have to set the appropriate address space operations for the inode (again, ignoring data journal flag if delalloc enabled). Additionally this commit introduces ext4_inode_journal_mode() function because ext4_should_*_data() has already had a lot of common code and this change is putting it all into one function so it is easier to read. Successfully tested with xfstests in following configurations: delalloc + data=ordered delalloc + data=writeback data=journal nodelalloc + data=ordered nodelalloc + data=writeback nodelalloc + data=journal Signed-off-by: Lukas Czerner <lczerner@redhat.com> Signed-off-by: "Theodore Ts'o" <tytso@mit.edu> Cc: stable@vger.kernel.org
2012-02-20 22:53:00 +00:00
default:
BUG();
}
if (test_opt(inode->i_sb, DELALLOC))
inode->i_mapping->a_ops = &ext4_da_aops;
else
inode->i_mapping->a_ops = &ext4_aops;
}
/*
* ext4_block_zero_page_range() zeros out a mapping of length 'length'
* starting from file offset 'from'. The range to be zero'd must
* be contained with in one block. If the specified range exceeds
* the end of the block it will be shortened to end of the block
* that cooresponds to 'from'
*/
static int ext4_block_zero_page_range(handle_t *handle,
struct address_space *mapping, loff_t from, loff_t length)
{
ext4_fsblk_t index = from >> PAGE_CACHE_SHIFT;
unsigned offset = from & (PAGE_CACHE_SIZE-1);
unsigned blocksize, max, pos;
ext4_lblk_t iblock;
struct inode *inode = mapping->host;
struct buffer_head *bh;
struct page *page;
int err = 0;
page = find_or_create_page(mapping, from >> PAGE_CACHE_SHIFT,
mapping_gfp_mask(mapping) & ~__GFP_FS);
if (!page)
return -ENOMEM;
blocksize = inode->i_sb->s_blocksize;
max = blocksize - (offset & (blocksize - 1));
/*
* correct length if it does not fall between
* 'from' and the end of the block
*/
if (length > max || length < 0)
length = max;
iblock = index << (PAGE_CACHE_SHIFT - inode->i_sb->s_blocksize_bits);
if (!page_has_buffers(page))
create_empty_buffers(page, blocksize, 0);
/* Find the buffer that contains "offset" */
bh = page_buffers(page);
pos = blocksize;
while (offset >= pos) {
bh = bh->b_this_page;
iblock++;
pos += blocksize;
}
if (buffer_freed(bh)) {
BUFFER_TRACE(bh, "freed: skip");
goto unlock;
}
if (!buffer_mapped(bh)) {
BUFFER_TRACE(bh, "unmapped");
ext4_get_block(inode, iblock, bh, 0);
/* unmapped? It's a hole - nothing to do */
if (!buffer_mapped(bh)) {
BUFFER_TRACE(bh, "still unmapped");
goto unlock;
}
}
/* Ok, it's mapped. Make sure it's up-to-date */
if (PageUptodate(page))
set_buffer_uptodate(bh);
if (!buffer_uptodate(bh)) {
err = -EIO;
ll_rw_block(READ, 1, &bh);
wait_on_buffer(bh);
/* Uhhuh. Read error. Complain and punt. */
if (!buffer_uptodate(bh))
goto unlock;
}
if (ext4_should_journal_data(inode)) {
BUFFER_TRACE(bh, "get write access");
err = ext4_journal_get_write_access(handle, bh);
if (err)
goto unlock;
}
zero_user(page, offset, length);
BUFFER_TRACE(bh, "zeroed end of block");
if (ext4_should_journal_data(inode)) {
err = ext4_handle_dirty_metadata(handle, inode, bh);
} else {
err = 0;
mark_buffer_dirty(bh);
if (ext4_test_inode_state(inode, EXT4_STATE_ORDERED_MODE))
err = ext4_jbd2_file_inode(handle, inode);
}
unlock:
unlock_page(page);
page_cache_release(page);
return err;
}
/*
* ext4_block_truncate_page() zeroes out a mapping from file offset `from'
* up to the end of the block which corresponds to `from'.
* This required during truncate. We need to physically zero the tail end
* of that block so it doesn't yield old data if the file is later grown.
*/
static int ext4_block_truncate_page(handle_t *handle,
struct address_space *mapping, loff_t from)
{
unsigned offset = from & (PAGE_CACHE_SIZE-1);
unsigned length;
unsigned blocksize;
struct inode *inode = mapping->host;
blocksize = inode->i_sb->s_blocksize;
length = blocksize - (offset & (blocksize - 1));
return ext4_block_zero_page_range(handle, mapping, from, length);
}
int ext4_zero_partial_blocks(handle_t *handle, struct inode *inode,
loff_t lstart, loff_t length)
{
struct super_block *sb = inode->i_sb;
struct address_space *mapping = inode->i_mapping;
unsigned partial_start, partial_end;
ext4_fsblk_t start, end;
loff_t byte_end = (lstart + length - 1);
int err = 0;
partial_start = lstart & (sb->s_blocksize - 1);
partial_end = byte_end & (sb->s_blocksize - 1);
start = lstart >> sb->s_blocksize_bits;
end = byte_end >> sb->s_blocksize_bits;
/* Handle partial zero within the single block */
if (start == end &&
(partial_start || (partial_end != sb->s_blocksize - 1))) {
err = ext4_block_zero_page_range(handle, mapping,
lstart, length);
return err;
}
/* Handle partial zero out on the start of the range */
if (partial_start) {
err = ext4_block_zero_page_range(handle, mapping,
lstart, sb->s_blocksize);
if (err)
return err;
}
/* Handle partial zero out on the end of the range */
if (partial_end != sb->s_blocksize - 1)
err = ext4_block_zero_page_range(handle, mapping,
byte_end - partial_end,
partial_end + 1);
return err;
}
int ext4_can_truncate(struct inode *inode)
{
if (S_ISREG(inode->i_mode))
return 1;
if (S_ISDIR(inode->i_mode))
return 1;
if (S_ISLNK(inode->i_mode))
return !ext4_inode_is_fast_symlink(inode);
return 0;
}
/*
* ext4_punch_hole: punches a hole in a file by releaseing the blocks
* associated with the given offset and length
*
* @inode: File inode
* @offset: The offset where the hole will begin
* @len: The length of the hole
*
* Returns: 0 on success or negative on failure
*/
int ext4_punch_hole(struct inode *inode, loff_t offset, loff_t length)
{
struct super_block *sb = inode->i_sb;
ext4_lblk_t first_block, stop_block;
struct address_space *mapping = inode->i_mapping;
loff_t first_block_offset, last_block_offset;
handle_t *handle;
unsigned int credits;
int ret = 0;
if (!S_ISREG(inode->i_mode))
return -EOPNOTSUPP;
trace_ext4_punch_hole(inode, offset, length, 0);
/*
* Write out all dirty pages to avoid race conditions
* Then release them.
*/
if (mapping->nrpages && mapping_tagged(mapping, PAGECACHE_TAG_DIRTY)) {
ret = filemap_write_and_wait_range(mapping, offset,
offset + length - 1);
if (ret)
return ret;
}
mutex_lock(&inode->i_mutex);
/* No need to punch hole beyond i_size */
if (offset >= inode->i_size)
goto out_mutex;
/*
* If the hole extends beyond i_size, set the hole
* to end after the page that contains i_size
*/
if (offset + length > inode->i_size) {
length = inode->i_size +
PAGE_CACHE_SIZE - (inode->i_size & (PAGE_CACHE_SIZE - 1)) -
offset;
}
if (offset & (sb->s_blocksize - 1) ||
(offset + length) & (sb->s_blocksize - 1)) {
/*
* Attach jinode to inode for jbd2 if we do any zeroing of
* partial block
*/
ret = ext4_inode_attach_jinode(inode);
if (ret < 0)
goto out_mutex;
}
first_block_offset = round_up(offset, sb->s_blocksize);
last_block_offset = round_down((offset + length), sb->s_blocksize) - 1;
/* Now release the pages and zero block aligned part of pages*/
if (last_block_offset > first_block_offset)
truncate_pagecache_range(inode, first_block_offset,
last_block_offset);
/* Wait all existing dio workers, newcomers will block on i_mutex */
ext4_inode_block_unlocked_dio(inode);
inode_dio_wait(inode);
if (ext4_test_inode_flag(inode, EXT4_INODE_EXTENTS))
credits = ext4_writepage_trans_blocks(inode);
else
credits = ext4_blocks_for_truncate(inode);
handle = ext4_journal_start(inode, EXT4_HT_TRUNCATE, credits);
if (IS_ERR(handle)) {
ret = PTR_ERR(handle);
ext4_std_error(sb, ret);
goto out_dio;
}
ret = ext4_zero_partial_blocks(handle, inode, offset,
length);
if (ret)
goto out_stop;
first_block = (offset + sb->s_blocksize - 1) >>
EXT4_BLOCK_SIZE_BITS(sb);
stop_block = (offset + length) >> EXT4_BLOCK_SIZE_BITS(sb);
/* If there are no blocks to remove, return now */
if (first_block >= stop_block)
goto out_stop;
down_write(&EXT4_I(inode)->i_data_sem);
ext4_discard_preallocations(inode);
ret = ext4_es_remove_extent(inode, first_block,
stop_block - first_block);
if (ret) {
up_write(&EXT4_I(inode)->i_data_sem);
goto out_stop;
}
if (ext4_test_inode_flag(inode, EXT4_INODE_EXTENTS))
ret = ext4_ext_remove_space(inode, first_block,
stop_block - 1);
else
ret = ext4_ind_remove_space(handle, inode, first_block,
stop_block);
up_write(&EXT4_I(inode)->i_data_sem);
if (IS_SYNC(inode))
ext4_handle_sync(handle);
/* Now release the pages again to reduce race window */
if (last_block_offset > first_block_offset)
truncate_pagecache_range(inode, first_block_offset,
last_block_offset);
inode->i_mtime = inode->i_ctime = ext4_current_time(inode);
ext4_mark_inode_dirty(handle, inode);
out_stop:
ext4_journal_stop(handle);
out_dio:
ext4_inode_resume_unlocked_dio(inode);
out_mutex:
mutex_unlock(&inode->i_mutex);
return ret;
}
int ext4_inode_attach_jinode(struct inode *inode)
{
struct ext4_inode_info *ei = EXT4_I(inode);
struct jbd2_inode *jinode;
if (ei->jinode || !EXT4_SB(inode->i_sb)->s_journal)
return 0;
jinode = jbd2_alloc_inode(GFP_KERNEL);
spin_lock(&inode->i_lock);
if (!ei->jinode) {
if (!jinode) {
spin_unlock(&inode->i_lock);
return -ENOMEM;
}
ei->jinode = jinode;
jbd2_journal_init_jbd_inode(ei->jinode, inode);
jinode = NULL;
}
spin_unlock(&inode->i_lock);
if (unlikely(jinode != NULL))
jbd2_free_inode(jinode);
return 0;
}
/*
* ext4_truncate()
*
* We block out ext4_get_block() block instantiations across the entire
* transaction, and VFS/VM ensures that ext4_truncate() cannot run
* simultaneously on behalf of the same inode.
*
* As we work through the truncate and commit bits of it to the journal there
* is one core, guiding principle: the file's tree must always be consistent on
* disk. We must be able to restart the truncate after a crash.
*
* The file's tree may be transiently inconsistent in memory (although it
* probably isn't), but whenever we close off and commit a journal transaction,
* the contents of (the filesystem + the journal) must be consistent and
* restartable. It's pretty simple, really: bottom up, right to left (although
* left-to-right works OK too).
*
* Note that at recovery time, journal replay occurs *before* the restart of
* truncate against the orphan inode list.
*
* The committed inode has the new, desired i_size (which is the same as
* i_disksize in this case). After a crash, ext4_orphan_cleanup() will see
* that this inode's truncate did not complete and it will again call
* ext4_truncate() to have another go. So there will be instantiated blocks
* to the right of the truncation point in a crashed ext4 filesystem. But
* that's fine - as long as they are linked from the inode, the post-crash
* ext4_truncate() run will find them and release them.
*/
void ext4_truncate(struct inode *inode)
{
struct ext4_inode_info *ei = EXT4_I(inode);
unsigned int credits;
handle_t *handle;
struct address_space *mapping = inode->i_mapping;
/*
* There is a possibility that we're either freeing the inode
* or it's a completely new inode. In those cases we might not
* have i_mutex locked because it's not necessary.
*/
if (!(inode->i_state & (I_NEW|I_FREEING)))
WARN_ON(!mutex_is_locked(&inode->i_mutex));
trace_ext4_truncate_enter(inode);
if (!ext4_can_truncate(inode))
return;
ext4_clear_inode_flag(inode, EXT4_INODE_EOFBLOCKS);
if (inode->i_size == 0 && !test_opt(inode->i_sb, NO_AUTO_DA_ALLOC))
ext4_set_inode_state(inode, EXT4_STATE_DA_ALLOC_CLOSE);
if (ext4_has_inline_data(inode)) {
int has_inline = 1;
ext4_inline_data_truncate(inode, &has_inline);
if (has_inline)
return;
}
/* If we zero-out tail of the page, we have to create jinode for jbd2 */
if (inode->i_size & (inode->i_sb->s_blocksize - 1)) {
if (ext4_inode_attach_jinode(inode) < 0)
return;
}
if (ext4_test_inode_flag(inode, EXT4_INODE_EXTENTS))
credits = ext4_writepage_trans_blocks(inode);
else
credits = ext4_blocks_for_truncate(inode);
handle = ext4_journal_start(inode, EXT4_HT_TRUNCATE, credits);
if (IS_ERR(handle)) {
ext4_std_error(inode->i_sb, PTR_ERR(handle));
return;
}
if (inode->i_size & (inode->i_sb->s_blocksize - 1))
ext4_block_truncate_page(handle, mapping, inode->i_size);
/*
* We add the inode to the orphan list, so that if this
* truncate spans multiple transactions, and we crash, we will
* resume the truncate when the filesystem recovers. It also
* marks the inode dirty, to catch the new size.
*
* Implication: the file must always be in a sane, consistent
* truncatable state while each transaction commits.
*/
if (ext4_orphan_add(handle, inode))
goto out_stop;
down_write(&EXT4_I(inode)->i_data_sem);
ext4_discard_preallocations(inode);
if (ext4_test_inode_flag(inode, EXT4_INODE_EXTENTS))
ext4_ext_truncate(handle, inode);
else
ext4_ind_truncate(handle, inode);
up_write(&ei->i_data_sem);
if (IS_SYNC(inode))
ext4_handle_sync(handle);
out_stop:
/*
* If this was a simple ftruncate() and the file will remain alive,
* then we need to clear up the orphan record which we created above.
* However, if this was a real unlink then we were called by
* ext4_delete_inode(), and we allow that function to clean up the
* orphan info for us.
*/
if (inode->i_nlink)
ext4_orphan_del(handle, inode);
inode->i_mtime = inode->i_ctime = ext4_current_time(inode);
ext4_mark_inode_dirty(handle, inode);
ext4_journal_stop(handle);
trace_ext4_truncate_exit(inode);
}
/*
* ext4_get_inode_loc returns with an extra refcount against the inode's
* underlying buffer_head on success. If 'in_mem' is true, we have all
* data in memory that is needed to recreate the on-disk version of this
* inode.
*/
static int __ext4_get_inode_loc(struct inode *inode,
struct ext4_iloc *iloc, int in_mem)
{
struct ext4_group_desc *gdp;
struct buffer_head *bh;
struct super_block *sb = inode->i_sb;
ext4_fsblk_t block;
int inodes_per_block, inode_offset;
iloc->bh = NULL;
if (!ext4_valid_inum(sb, inode->i_ino))
return -EIO;
iloc->block_group = (inode->i_ino - 1) / EXT4_INODES_PER_GROUP(sb);
gdp = ext4_get_group_desc(sb, iloc->block_group, NULL);
if (!gdp)
return -EIO;
/*
* Figure out the offset within the block group inode table
*/
inodes_per_block = EXT4_SB(sb)->s_inodes_per_block;
inode_offset = ((inode->i_ino - 1) %
EXT4_INODES_PER_GROUP(sb));
block = ext4_inode_table(sb, gdp) + (inode_offset / inodes_per_block);
iloc->offset = (inode_offset % inodes_per_block) * EXT4_INODE_SIZE(sb);
bh = sb_getblk(sb, block);
if (unlikely(!bh))
return -ENOMEM;
if (!buffer_uptodate(bh)) {
lock_buffer(bh);
/*
* If the buffer has the write error flag, we have failed
* to write out another inode in the same block. In this
* case, we don't have to read the block because we may
* read the old inode data successfully.
*/
if (buffer_write_io_error(bh) && !buffer_uptodate(bh))
set_buffer_uptodate(bh);
if (buffer_uptodate(bh)) {
/* someone brought it uptodate while we waited */
unlock_buffer(bh);
goto has_buffer;
}
/*
* If we have all information of the inode in memory and this
* is the only valid inode in the block, we need not read the
* block.
*/
if (in_mem) {
struct buffer_head *bitmap_bh;
int i, start;
start = inode_offset & ~(inodes_per_block - 1);
/* Is the inode bitmap in cache? */
bitmap_bh = sb_getblk(sb, ext4_inode_bitmap(sb, gdp));
if (unlikely(!bitmap_bh))
goto make_io;
/*
* If the inode bitmap isn't in cache then the
* optimisation may end up performing two reads instead
* of one, so skip it.
*/
if (!buffer_uptodate(bitmap_bh)) {
brelse(bitmap_bh);
goto make_io;
}
for (i = start; i < start + inodes_per_block; i++) {
if (i == inode_offset)
continue;
if (ext4_test_bit(i, bitmap_bh->b_data))
break;
}
brelse(bitmap_bh);
if (i == start + inodes_per_block) {
/* all other inodes are free, so skip I/O */
memset(bh->b_data, 0, bh->b_size);
set_buffer_uptodate(bh);
unlock_buffer(bh);
goto has_buffer;
}
}
make_io:
/*
* If we need to do any I/O, try to pre-readahead extra
* blocks from the inode table.
*/
if (EXT4_SB(sb)->s_inode_readahead_blks) {
ext4_fsblk_t b, end, table;
unsigned num;
__u32 ra_blks = EXT4_SB(sb)->s_inode_readahead_blks;
table = ext4_inode_table(sb, gdp);
/* s_inode_readahead_blks is always a power of 2 */
b = block & ~((ext4_fsblk_t) ra_blks - 1);
if (table > b)
b = table;
end = b + ra_blks;
num = EXT4_INODES_PER_GROUP(sb);
if (ext4_has_group_desc_csum(sb))
num -= ext4_itable_unused_count(sb, gdp);
table += num / inodes_per_block;
if (end > table)
end = table;
while (b <= end)
sb_breadahead(sb, b++);
}
/*
* There are other valid inodes in the buffer, this inode
* has in-inode xattrs, or we don't have this inode in memory.
* Read the block from disk.
*/
trace_ext4_load_inode(inode);
get_bh(bh);
bh->b_end_io = end_buffer_read_sync;
submit_bh(READ | REQ_META | REQ_PRIO, bh);
wait_on_buffer(bh);
if (!buffer_uptodate(bh)) {
EXT4_ERROR_INODE_BLOCK(inode, block,
"unable to read itable block");
brelse(bh);
return -EIO;
}
}
has_buffer:
iloc->bh = bh;
return 0;
}
int ext4_get_inode_loc(struct inode *inode, struct ext4_iloc *iloc)
{
/* We have all inode data except xattrs in memory here. */
return __ext4_get_inode_loc(inode, iloc,
!ext4_test_inode_state(inode, EXT4_STATE_XATTR));
}
void ext4_set_inode_flags(struct inode *inode)
{
unsigned int flags = EXT4_I(inode)->i_flags;
unsigned int new_fl = 0;
if (flags & EXT4_SYNC_FL)
new_fl |= S_SYNC;
if (flags & EXT4_APPEND_FL)
new_fl |= S_APPEND;
if (flags & EXT4_IMMUTABLE_FL)
new_fl |= S_IMMUTABLE;
if (flags & EXT4_NOATIME_FL)
new_fl |= S_NOATIME;
if (flags & EXT4_DIRSYNC_FL)
new_fl |= S_DIRSYNC;
inode_set_flags(inode, new_fl,
S_SYNC|S_APPEND|S_IMMUTABLE|S_NOATIME|S_DIRSYNC);
}
/* Propagate flags from i_flags to EXT4_I(inode)->i_flags */
void ext4_get_inode_flags(struct ext4_inode_info *ei)
{
unsigned int vfs_fl;
unsigned long old_fl, new_fl;
do {
vfs_fl = ei->vfs_inode.i_flags;
old_fl = ei->i_flags;
new_fl = old_fl & ~(EXT4_SYNC_FL|EXT4_APPEND_FL|
EXT4_IMMUTABLE_FL|EXT4_NOATIME_FL|
EXT4_DIRSYNC_FL);
if (vfs_fl & S_SYNC)
new_fl |= EXT4_SYNC_FL;
if (vfs_fl & S_APPEND)
new_fl |= EXT4_APPEND_FL;
if (vfs_fl & S_IMMUTABLE)
new_fl |= EXT4_IMMUTABLE_FL;
if (vfs_fl & S_NOATIME)
new_fl |= EXT4_NOATIME_FL;
if (vfs_fl & S_DIRSYNC)
new_fl |= EXT4_DIRSYNC_FL;
} while (cmpxchg(&ei->i_flags, old_fl, new_fl) != old_fl);
}
static blkcnt_t ext4_inode_blocks(struct ext4_inode *raw_inode,
struct ext4_inode_info *ei)
{
blkcnt_t i_blocks ;
struct inode *inode = &(ei->vfs_inode);
struct super_block *sb = inode->i_sb;
if (EXT4_HAS_RO_COMPAT_FEATURE(sb,
EXT4_FEATURE_RO_COMPAT_HUGE_FILE)) {
/* we are using combined 48 bit field */
i_blocks = ((u64)le16_to_cpu(raw_inode->i_blocks_high)) << 32 |
le32_to_cpu(raw_inode->i_blocks_lo);
if (ext4_test_inode_flag(inode, EXT4_INODE_HUGE_FILE)) {
/* i_blocks represent file system block size */
return i_blocks << (inode->i_blkbits - 9);
} else {
return i_blocks;
}
} else {
return le32_to_cpu(raw_inode->i_blocks_lo);
}
}
static inline void ext4_iget_extra_inode(struct inode *inode,
struct ext4_inode *raw_inode,
struct ext4_inode_info *ei)
{
__le32 *magic = (void *)raw_inode +
EXT4_GOOD_OLD_INODE_SIZE + ei->i_extra_isize;
if (*magic == cpu_to_le32(EXT4_XATTR_MAGIC)) {
ext4_set_inode_state(inode, EXT4_STATE_XATTR);
ext4_find_inline_data_nolock(inode);
} else
EXT4_I(inode)->i_inline_off = 0;
}
struct inode *ext4_iget(struct super_block *sb, unsigned long ino)
{
struct ext4_iloc iloc;
struct ext4_inode *raw_inode;
struct ext4_inode_info *ei;
struct inode *inode;
journal_t *journal = EXT4_SB(sb)->s_journal;
long ret;
int block;
uid_t i_uid;
gid_t i_gid;
inode = iget_locked(sb, ino);
if (!inode)
return ERR_PTR(-ENOMEM);
if (!(inode->i_state & I_NEW))
return inode;
ei = EXT4_I(inode);
iloc.bh = NULL;
ret = __ext4_get_inode_loc(inode, &iloc, 0);
if (ret < 0)
goto bad_inode;
raw_inode = ext4_raw_inode(&iloc);
if (EXT4_INODE_SIZE(inode->i_sb) > EXT4_GOOD_OLD_INODE_SIZE) {
ei->i_extra_isize = le16_to_cpu(raw_inode->i_extra_isize);
if (EXT4_GOOD_OLD_INODE_SIZE + ei->i_extra_isize >
EXT4_INODE_SIZE(inode->i_sb)) {
EXT4_ERROR_INODE(inode, "bad extra_isize (%u != %u)",
EXT4_GOOD_OLD_INODE_SIZE + ei->i_extra_isize,
EXT4_INODE_SIZE(inode->i_sb));
ret = -EIO;
goto bad_inode;
}
} else
ei->i_extra_isize = 0;
/* Precompute checksum seed for inode metadata */
if (EXT4_HAS_RO_COMPAT_FEATURE(sb,
EXT4_FEATURE_RO_COMPAT_METADATA_CSUM)) {
struct ext4_sb_info *sbi = EXT4_SB(inode->i_sb);
__u32 csum;
__le32 inum = cpu_to_le32(inode->i_ino);
__le32 gen = raw_inode->i_generation;
csum = ext4_chksum(sbi, sbi->s_csum_seed, (__u8 *)&inum,
sizeof(inum));
ei->i_csum_seed = ext4_chksum(sbi, csum, (__u8 *)&gen,
sizeof(gen));
}
if (!ext4_inode_csum_verify(inode, raw_inode, ei)) {
EXT4_ERROR_INODE(inode, "checksum invalid");
ret = -EIO;
goto bad_inode;
}
inode->i_mode = le16_to_cpu(raw_inode->i_mode);
i_uid = (uid_t)le16_to_cpu(raw_inode->i_uid_low);
i_gid = (gid_t)le16_to_cpu(raw_inode->i_gid_low);
if (!(test_opt(inode->i_sb, NO_UID32))) {
i_uid |= le16_to_cpu(raw_inode->i_uid_high) << 16;
i_gid |= le16_to_cpu(raw_inode->i_gid_high) << 16;
}
i_uid_write(inode, i_uid);
i_gid_write(inode, i_gid);
set_nlink(inode, le16_to_cpu(raw_inode->i_links_count));
ext4_clear_state_flags(ei); /* Only relevant on 32-bit archs */
ei->i_inline_off = 0;
ei->i_dir_start_lookup = 0;
ei->i_dtime = le32_to_cpu(raw_inode->i_dtime);
/* We now have enough fields to check if the inode was active or not.
* This is needed because nfsd might try to access dead inodes
* the test is that same one that e2fsck uses
* NeilBrown 1999oct15
*/
if (inode->i_nlink == 0) {
if ((inode->i_mode == 0 ||
!(EXT4_SB(inode->i_sb)->s_mount_state & EXT4_ORPHAN_FS)) &&
ino != EXT4_BOOT_LOADER_INO) {
/* this inode is deleted */
ret = -ESTALE;
goto bad_inode;
}
/* The only unlinked inodes we let through here have
* valid i_mode and are being read by the orphan
* recovery code: that's fine, we're about to complete
* the process of deleting those.
* OR it is the EXT4_BOOT_LOADER_INO which is
* not initialized on a new filesystem. */
}
ei->i_flags = le32_to_cpu(raw_inode->i_flags);
inode->i_blocks = ext4_inode_blocks(raw_inode, ei);
ei->i_file_acl = le32_to_cpu(raw_inode->i_file_acl_lo);
if (EXT4_HAS_INCOMPAT_FEATURE(sb, EXT4_FEATURE_INCOMPAT_64BIT))
ei->i_file_acl |=
((__u64)le16_to_cpu(raw_inode->i_file_acl_high)) << 32;
inode->i_size = ext4_isize(raw_inode);
ei->i_disksize = inode->i_size;
#ifdef CONFIG_QUOTA
ei->i_reserved_quota = 0;
#endif
inode->i_generation = le32_to_cpu(raw_inode->i_generation);
ei->i_block_group = iloc.block_group;
ei->i_last_alloc_group = ~0;
/*
* NOTE! The in-memory inode i_data array is in little-endian order
* even on big-endian machines: we do NOT byteswap the block numbers!
*/
for (block = 0; block < EXT4_N_BLOCKS; block++)
ei->i_data[block] = raw_inode->i_block[block];
INIT_LIST_HEAD(&ei->i_orphan);
/*
* Set transaction id's of transactions that have to be committed
* to finish f[data]sync. We set them to currently running transaction
* as we cannot be sure that the inode or some of its metadata isn't
* part of the transaction - the inode could have been reclaimed and
* now it is reread from disk.
*/
if (journal) {
transaction_t *transaction;
tid_t tid;
read_lock(&journal->j_state_lock);
if (journal->j_running_transaction)
transaction = journal->j_running_transaction;
else
transaction = journal->j_committing_transaction;
if (transaction)
tid = transaction->t_tid;
else
tid = journal->j_commit_sequence;
read_unlock(&journal->j_state_lock);
ei->i_sync_tid = tid;
ei->i_datasync_tid = tid;
}
if (EXT4_INODE_SIZE(inode->i_sb) > EXT4_GOOD_OLD_INODE_SIZE) {
if (ei->i_extra_isize == 0) {
/* The extra space is currently unused. Use it. */
ei->i_extra_isize = sizeof(struct ext4_inode) -
EXT4_GOOD_OLD_INODE_SIZE;
} else {
ext4_iget_extra_inode(inode, raw_inode, ei);
}
}
EXT4_INODE_GET_XTIME(i_ctime, inode, raw_inode);
EXT4_INODE_GET_XTIME(i_mtime, inode, raw_inode);
EXT4_INODE_GET_XTIME(i_atime, inode, raw_inode);
EXT4_EINODE_GET_XTIME(i_crtime, ei, raw_inode);
if (likely(!test_opt2(inode->i_sb, HURD_COMPAT))) {
inode->i_version = le32_to_cpu(raw_inode->i_disk_version);
if (EXT4_INODE_SIZE(inode->i_sb) > EXT4_GOOD_OLD_INODE_SIZE) {
if (EXT4_FITS_IN_INODE(raw_inode, ei, i_version_hi))
inode->i_version |=
(__u64)(le32_to_cpu(raw_inode->i_version_hi)) << 32;
}
}
ret = 0;
if (ei->i_file_acl &&
!ext4_data_block_valid(EXT4_SB(sb), ei->i_file_acl, 1)) {
EXT4_ERROR_INODE(inode, "bad extended attribute block %llu",
ei->i_file_acl);
ret = -EIO;
goto bad_inode;
} else if (!ext4_has_inline_data(inode)) {
if (ext4_test_inode_flag(inode, EXT4_INODE_EXTENTS)) {
if ((S_ISREG(inode->i_mode) || S_ISDIR(inode->i_mode) ||
(S_ISLNK(inode->i_mode) &&
!ext4_inode_is_fast_symlink(inode))))
/* Validate extent which is part of inode */
ret = ext4_ext_check_inode(inode);
} else if (S_ISREG(inode->i_mode) || S_ISDIR(inode->i_mode) ||
(S_ISLNK(inode->i_mode) &&
!ext4_inode_is_fast_symlink(inode))) {
/* Validate block references which are part of inode */
ret = ext4_ind_check_inode(inode);
}
}
if (ret)
goto bad_inode;
if (S_ISREG(inode->i_mode)) {
inode->i_op = &ext4_file_inode_operations;
inode->i_fop = &ext4_file_operations;
ext4_set_aops(inode);
} else if (S_ISDIR(inode->i_mode)) {
inode->i_op = &ext4_dir_inode_operations;
inode->i_fop = &ext4_dir_operations;
} else if (S_ISLNK(inode->i_mode)) {
if (ext4_inode_is_fast_symlink(inode)) {
inode->i_op = &ext4_fast_symlink_inode_operations;
nd_terminate_link(ei->i_data, inode->i_size,
sizeof(ei->i_data) - 1);
} else {
inode->i_op = &ext4_symlink_inode_operations;
ext4_set_aops(inode);
}
} else if (S_ISCHR(inode->i_mode) || S_ISBLK(inode->i_mode) ||
S_ISFIFO(inode->i_mode) || S_ISSOCK(inode->i_mode)) {
inode->i_op = &ext4_special_inode_operations;
if (raw_inode->i_block[0])
init_special_inode(inode, inode->i_mode,
old_decode_dev(le32_to_cpu(raw_inode->i_block[0])));
else
init_special_inode(inode, inode->i_mode,
new_decode_dev(le32_to_cpu(raw_inode->i_block[1])));
} else if (ino == EXT4_BOOT_LOADER_INO) {
make_bad_inode(inode);
} else {
ret = -EIO;
EXT4_ERROR_INODE(inode, "bogus i_mode (%o)", inode->i_mode);
goto bad_inode;
}
brelse(iloc.bh);
ext4_set_inode_flags(inode);
unlock_new_inode(inode);
return inode;
bad_inode:
brelse(iloc.bh);
iget_failed(inode);
return ERR_PTR(ret);
}
static int ext4_inode_blocks_set(handle_t *handle,
struct ext4_inode *raw_inode,
struct ext4_inode_info *ei)
{
struct inode *inode = &(ei->vfs_inode);
u64 i_blocks = inode->i_blocks;
struct super_block *sb = inode->i_sb;
if (i_blocks <= ~0U) {
/*
* i_blocks can be represented in a 32 bit variable
* as multiple of 512 bytes
*/
raw_inode->i_blocks_lo = cpu_to_le32(i_blocks);
raw_inode->i_blocks_high = 0;
ext4_clear_inode_flag(inode, EXT4_INODE_HUGE_FILE);
return 0;
}
if (!EXT4_HAS_RO_COMPAT_FEATURE(sb, EXT4_FEATURE_RO_COMPAT_HUGE_FILE))
return -EFBIG;
if (i_blocks <= 0xffffffffffffULL) {
/*
* i_blocks can be represented in a 48 bit variable
* as multiple of 512 bytes
*/
raw_inode->i_blocks_lo = cpu_to_le32(i_blocks);
raw_inode->i_blocks_high = cpu_to_le16(i_blocks >> 32);
ext4_clear_inode_flag(inode, EXT4_INODE_HUGE_FILE);
} else {
ext4_set_inode_flag(inode, EXT4_INODE_HUGE_FILE);
/* i_block is stored in file system block size */
i_blocks = i_blocks >> (inode->i_blkbits - 9);
raw_inode->i_blocks_lo = cpu_to_le32(i_blocks);
raw_inode->i_blocks_high = cpu_to_le16(i_blocks >> 32);
}
return 0;
}
/*
* Post the struct inode info into an on-disk inode location in the
* buffer-cache. This gobbles the caller's reference to the
* buffer_head in the inode location struct.
*
* The caller must have write access to iloc->bh.
*/
static int ext4_do_update_inode(handle_t *handle,
struct inode *inode,
struct ext4_iloc *iloc)
{
struct ext4_inode *raw_inode = ext4_raw_inode(iloc);
struct ext4_inode_info *ei = EXT4_I(inode);
struct buffer_head *bh = iloc->bh;
struct super_block *sb = inode->i_sb;
int err = 0, rc, block;
int need_datasync = 0, set_large_file = 0;
uid_t i_uid;
gid_t i_gid;
spin_lock(&ei->i_raw_lock);
/* For fields not tracked in the in-memory inode,
* initialise them to zero for new inodes. */
if (ext4_test_inode_state(inode, EXT4_STATE_NEW))
memset(raw_inode, 0, EXT4_SB(inode->i_sb)->s_inode_size);
ext4_get_inode_flags(ei);
raw_inode->i_mode = cpu_to_le16(inode->i_mode);
i_uid = i_uid_read(inode);
i_gid = i_gid_read(inode);
if (!(test_opt(inode->i_sb, NO_UID32))) {
raw_inode->i_uid_low = cpu_to_le16(low_16_bits(i_uid));
raw_inode->i_gid_low = cpu_to_le16(low_16_bits(i_gid));
/*
* Fix up interoperability with old kernels. Otherwise, old inodes get
* re-used with the upper 16 bits of the uid/gid intact
*/
if (!ei->i_dtime) {
raw_inode->i_uid_high =
cpu_to_le16(high_16_bits(i_uid));
raw_inode->i_gid_high =
cpu_to_le16(high_16_bits(i_gid));
} else {
raw_inode->i_uid_high = 0;
raw_inode->i_gid_high = 0;
}
} else {
raw_inode->i_uid_low = cpu_to_le16(fs_high2lowuid(i_uid));
raw_inode->i_gid_low = cpu_to_le16(fs_high2lowgid(i_gid));
raw_inode->i_uid_high = 0;
raw_inode->i_gid_high = 0;
}
raw_inode->i_links_count = cpu_to_le16(inode->i_nlink);
EXT4_INODE_SET_XTIME(i_ctime, inode, raw_inode);
EXT4_INODE_SET_XTIME(i_mtime, inode, raw_inode);
EXT4_INODE_SET_XTIME(i_atime, inode, raw_inode);
EXT4_EINODE_SET_XTIME(i_crtime, ei, raw_inode);
if (ext4_inode_blocks_set(handle, raw_inode, ei)) {
spin_unlock(&ei->i_raw_lock);
goto out_brelse;
}
raw_inode->i_dtime = cpu_to_le32(ei->i_dtime);
raw_inode->i_flags = cpu_to_le32(ei->i_flags & 0xFFFFFFFF);
if (likely(!test_opt2(inode->i_sb, HURD_COMPAT)))
raw_inode->i_file_acl_high =
cpu_to_le16(ei->i_file_acl >> 32);
raw_inode->i_file_acl_lo = cpu_to_le32(ei->i_file_acl);
if (ei->i_disksize != ext4_isize(raw_inode)) {
ext4_isize_set(raw_inode, ei->i_disksize);
need_datasync = 1;
}
if (ei->i_disksize > 0x7fffffffULL) {
if (!EXT4_HAS_RO_COMPAT_FEATURE(sb,
EXT4_FEATURE_RO_COMPAT_LARGE_FILE) ||
EXT4_SB(sb)->s_es->s_rev_level ==
cpu_to_le32(EXT4_GOOD_OLD_REV))
set_large_file = 1;
}
raw_inode->i_generation = cpu_to_le32(inode->i_generation);
if (S_ISCHR(inode->i_mode) || S_ISBLK(inode->i_mode)) {
if (old_valid_dev(inode->i_rdev)) {
raw_inode->i_block[0] =
cpu_to_le32(old_encode_dev(inode->i_rdev));
raw_inode->i_block[1] = 0;
} else {
raw_inode->i_block[0] = 0;
raw_inode->i_block[1] =
cpu_to_le32(new_encode_dev(inode->i_rdev));
raw_inode->i_block[2] = 0;
}
} else if (!ext4_has_inline_data(inode)) {
for (block = 0; block < EXT4_N_BLOCKS; block++)
raw_inode->i_block[block] = ei->i_data[block];
}
if (likely(!test_opt2(inode->i_sb, HURD_COMPAT))) {
raw_inode->i_disk_version = cpu_to_le32(inode->i_version);
if (ei->i_extra_isize) {
if (EXT4_FITS_IN_INODE(raw_inode, ei, i_version_hi))
raw_inode->i_version_hi =
cpu_to_le32(inode->i_version >> 32);
raw_inode->i_extra_isize =
cpu_to_le16(ei->i_extra_isize);
}
}
ext4_inode_csum_set(inode, raw_inode, ei);
spin_unlock(&ei->i_raw_lock);
BUFFER_TRACE(bh, "call ext4_handle_dirty_metadata");
rc = ext4_handle_dirty_metadata(handle, NULL, bh);
if (!err)
err = rc;
ext4_clear_inode_state(inode, EXT4_STATE_NEW);
if (set_large_file) {
BUFFER_TRACE(EXT4_SB(sb)->s_sbh, "get write access");
err = ext4_journal_get_write_access(handle, EXT4_SB(sb)->s_sbh);
if (err)
goto out_brelse;
ext4_update_dynamic_rev(sb);
EXT4_SET_RO_COMPAT_FEATURE(sb,
EXT4_FEATURE_RO_COMPAT_LARGE_FILE);
ext4_handle_sync(handle);
err = ext4_handle_dirty_super(handle, sb);
}
ext4_update_inode_fsync_trans(handle, inode, need_datasync);
out_brelse:
brelse(bh);
ext4_std_error(inode->i_sb, err);
return err;
}
/*
* ext4_write_inode()
*
* We are called from a few places:
*
* - Within generic_file_aio_write() -> generic_write_sync() for O_SYNC files.
* Here, there will be no transaction running. We wait for any running
* transaction to commit.
*
* - Within flush work (sys_sync(), kupdate and such).
* We wait on commit, if told to.
*
* - Within iput_final() -> write_inode_now()
* We wait on commit, if told to.
*
* In all cases it is actually safe for us to return without doing anything,
* because the inode has been copied into a raw inode buffer in
* ext4_mark_inode_dirty(). This is a correctness thing for WB_SYNC_ALL
* writeback.
*
* Note that we are absolutely dependent upon all inode dirtiers doing the
* right thing: they *must* call mark_inode_dirty() after dirtying info in
* which we are interested.
*
* It would be a bug for them to not do this. The code:
*
* mark_inode_dirty(inode)
* stuff();
* inode->i_size = expr;
*
* is in error because write_inode() could occur while `stuff()' is running,
* and the new i_size will be lost. Plus the inode will no longer be on the
* superblock's dirty inode list.
*/
int ext4_write_inode(struct inode *inode, struct writeback_control *wbc)
{
int err;
if (WARN_ON_ONCE(current->flags & PF_MEMALLOC))
return 0;
if (EXT4_SB(inode->i_sb)->s_journal) {
if (ext4_journal_current_handle()) {
jbd_debug(1, "called recursively, non-PF_MEMALLOC!\n");
dump_stack();
return -EIO;
}
/*
* No need to force transaction in WB_SYNC_NONE mode. Also
* ext4_sync_fs() will force the commit after everything is
* written.
*/
if (wbc->sync_mode != WB_SYNC_ALL || wbc->for_sync)
return 0;
err = ext4_force_commit(inode->i_sb);
} else {
struct ext4_iloc iloc;
err = __ext4_get_inode_loc(inode, &iloc, 0);
if (err)
return err;
/*
* sync(2) will flush the whole buffer cache. No need to do
* it here separately for each inode.
*/
if (wbc->sync_mode == WB_SYNC_ALL && !wbc->for_sync)
sync_dirty_buffer(iloc.bh);
if (buffer_req(iloc.bh) && !buffer_uptodate(iloc.bh)) {
EXT4_ERROR_INODE_BLOCK(inode, iloc.bh->b_blocknr,
"IO error syncing inode");
err = -EIO;
}
brelse(iloc.bh);
}
return err;
}
/*
* In data=journal mode ext4_journalled_invalidatepage() may fail to invalidate
* buffers that are attached to a page stradding i_size and are undergoing
* commit. In that case we have to wait for commit to finish and try again.
*/
static void ext4_wait_for_tail_page_commit(struct inode *inode)
{
struct page *page;
unsigned offset;
journal_t *journal = EXT4_SB(inode->i_sb)->s_journal;
tid_t commit_tid = 0;
int ret;
offset = inode->i_size & (PAGE_CACHE_SIZE - 1);
/*
* All buffers in the last page remain valid? Then there's nothing to
* do. We do the check mainly to optimize the common PAGE_CACHE_SIZE ==
* blocksize case
*/
if (offset > PAGE_CACHE_SIZE - (1 << inode->i_blkbits))
return;
while (1) {
page = find_lock_page(inode->i_mapping,
inode->i_size >> PAGE_CACHE_SHIFT);
if (!page)
return;
ret = __ext4_journalled_invalidatepage(page, offset,
PAGE_CACHE_SIZE - offset);
unlock_page(page);
page_cache_release(page);
if (ret != -EBUSY)
return;
commit_tid = 0;
read_lock(&journal->j_state_lock);
if (journal->j_committing_transaction)
commit_tid = journal->j_committing_transaction->t_tid;
read_unlock(&journal->j_state_lock);
if (commit_tid)
jbd2_log_wait_commit(journal, commit_tid);
}
}
/*
* ext4_setattr()
*
* Called from notify_change.
*
* We want to trap VFS attempts to truncate the file as soon as
* possible. In particular, we want to make sure that when the VFS
* shrinks i_size, we put the inode on the orphan list and modify
* i_disksize immediately, so that during the subsequent flushing of
* dirty pages and freeing of disk blocks, we can guarantee that any
* commit will leave the blocks being flushed in an unused state on
* disk. (On recovery, the inode will get truncated and the blocks will
* be freed, so we have a strong guarantee that no future commit will
* leave these blocks visible to the user.)
*
* Another thing we have to assure is that if we are in ordered mode
* and inode is still attached to the committing transaction, we must
* we start writeout of all the dirty pages which are being truncated.
* This way we are sure that all the data written in the previous
* transaction are already on disk (truncate waits for pages under
* writeback).
*
* Called with inode->i_mutex down.
*/
int ext4_setattr(struct dentry *dentry, struct iattr *attr)
{
struct inode *inode = dentry->d_inode;
int error, rc = 0;
int orphan = 0;
const unsigned int ia_valid = attr->ia_valid;
error = inode_change_ok(inode, attr);
if (error)
return error;
if (is_quota_modification(inode, attr))
dquot_initialize(inode);
if ((ia_valid & ATTR_UID && !uid_eq(attr->ia_uid, inode->i_uid)) ||
(ia_valid & ATTR_GID && !gid_eq(attr->ia_gid, inode->i_gid))) {
handle_t *handle;
/* (user+group)*(old+new) structure, inode write (sb,
* inode block, ? - but truncate inode update has it) */
handle = ext4_journal_start(inode, EXT4_HT_QUOTA,
(EXT4_MAXQUOTAS_INIT_BLOCKS(inode->i_sb) +
EXT4_MAXQUOTAS_DEL_BLOCKS(inode->i_sb)) + 3);
if (IS_ERR(handle)) {
error = PTR_ERR(handle);
goto err_out;
}
error = dquot_transfer(inode, attr);
if (error) {
ext4_journal_stop(handle);
return error;
}
/* Update corresponding info in inode so that everything is in
* one transaction */
if (attr->ia_valid & ATTR_UID)
inode->i_uid = attr->ia_uid;
if (attr->ia_valid & ATTR_GID)
inode->i_gid = attr->ia_gid;
error = ext4_mark_inode_dirty(handle, inode);
ext4_journal_stop(handle);
}
if (attr->ia_valid & ATTR_SIZE && attr->ia_size != inode->i_size) {
handle_t *handle;
if (!(ext4_test_inode_flag(inode, EXT4_INODE_EXTENTS))) {
struct ext4_sb_info *sbi = EXT4_SB(inode->i_sb);
if (attr->ia_size > sbi->s_bitmap_maxbytes)
return -EFBIG;
}
if (IS_I_VERSION(inode) && attr->ia_size != inode->i_size)
inode_inc_iversion(inode);
if (S_ISREG(inode->i_mode) &&
(attr->ia_size < inode->i_size)) {
if (ext4_should_order_data(inode)) {
error = ext4_begin_ordered_truncate(inode,
attr->ia_size);
if (error)
goto err_out;
}
handle = ext4_journal_start(inode, EXT4_HT_INODE, 3);
if (IS_ERR(handle)) {
error = PTR_ERR(handle);
goto err_out;
}
if (ext4_handle_valid(handle)) {
error = ext4_orphan_add(handle, inode);
orphan = 1;
}
down_write(&EXT4_I(inode)->i_data_sem);
EXT4_I(inode)->i_disksize = attr->ia_size;
rc = ext4_mark_inode_dirty(handle, inode);
if (!error)
error = rc;
/*
* We have to update i_size under i_data_sem together
* with i_disksize to avoid races with writeback code
* running ext4_wb_update_i_disksize().
*/
if (!error)
i_size_write(inode, attr->ia_size);
up_write(&EXT4_I(inode)->i_data_sem);
ext4_journal_stop(handle);
if (error) {
ext4_orphan_del(NULL, inode);
goto err_out;
}
} else
i_size_write(inode, attr->ia_size);
/*
* Blocks are going to be removed from the inode. Wait
* for dio in flight. Temporarily disable
* dioread_nolock to prevent livelock.
*/
if (orphan) {
if (!ext4_should_journal_data(inode)) {
ext4_inode_block_unlocked_dio(inode);
inode_dio_wait(inode);
ext4_inode_resume_unlocked_dio(inode);
} else
ext4_wait_for_tail_page_commit(inode);
}
/*
* Truncate pagecache after we've waited for commit
* in data=journal mode to make pages freeable.
*/
truncate_pagecache(inode, inode->i_size);
}
/*
* We want to call ext4_truncate() even if attr->ia_size ==
* inode->i_size for cases like truncation of fallocated space
*/
if (attr->ia_valid & ATTR_SIZE)
ext4_truncate(inode);
if (!rc) {
setattr_copy(inode, attr);
mark_inode_dirty(inode);
}
/*
* If the call to ext4_truncate failed to get a transaction handle at
* all, we need to clean up the in-core orphan list manually.
*/
if (orphan && inode->i_nlink)
ext4_orphan_del(NULL, inode);
if (!rc && (ia_valid & ATTR_MODE))
rc = posix_acl_chmod(inode, inode->i_mode);
err_out:
ext4_std_error(inode->i_sb, error);
if (!error)
error = rc;
return error;
}
int ext4_getattr(struct vfsmount *mnt, struct dentry *dentry,
struct kstat *stat)
{
struct inode *inode;
unsigned long long delalloc_blocks;
inode = dentry->d_inode;
generic_fillattr(inode, stat);
/*
* If there is inline data in the inode, the inode will normally not
* have data blocks allocated (it may have an external xattr block).
* Report at least one sector for such files, so tools like tar, rsync,
* others doen't incorrectly think the file is completely sparse.
*/
if (unlikely(ext4_has_inline_data(inode)))
stat->blocks += (stat->size + 511) >> 9;
/*
* We can't update i_blocks if the block allocation is delayed
* otherwise in the case of system crash before the real block
* allocation is done, we will have i_blocks inconsistent with
* on-disk file blocks.
* We always keep i_blocks updated together with real
* allocation. But to not confuse with user, stat
* will return the blocks that include the delayed allocation
* blocks for this file.
*/
delalloc_blocks = EXT4_C2B(EXT4_SB(inode->i_sb),
EXT4_I(inode)->i_reserved_data_blocks);
stat->blocks += delalloc_blocks << (inode->i_sb->s_blocksize_bits - 9);
return 0;
}
static int ext4_index_trans_blocks(struct inode *inode, int lblocks,
int pextents)
{
if (!(ext4_test_inode_flag(inode, EXT4_INODE_EXTENTS)))
return ext4_ind_trans_blocks(inode, lblocks);
return ext4_ext_index_trans_blocks(inode, pextents);
}
/*
* Account for index blocks, block groups bitmaps and block group
* descriptor blocks if modify datablocks and index blocks
* worse case, the indexs blocks spread over different block groups
*
* If datablocks are discontiguous, they are possible to spread over
* different block groups too. If they are contiguous, with flexbg,
* they could still across block group boundary.
*
* Also account for superblock, inode, quota and xattr blocks
*/
static int ext4_meta_trans_blocks(struct inode *inode, int lblocks,
int pextents)
{
ext4_group_t groups, ngroups = ext4_get_groups_count(inode->i_sb);
int gdpblocks;
int idxblocks;
int ret = 0;
/*
* How many index blocks need to touch to map @lblocks logical blocks
* to @pextents physical extents?
*/
idxblocks = ext4_index_trans_blocks(inode, lblocks, pextents);
ret = idxblocks;
/*
* Now let's see how many group bitmaps and group descriptors need
* to account
*/
groups = idxblocks + pextents;
gdpblocks = groups;
if (groups > ngroups)
groups = ngroups;
if (groups > EXT4_SB(inode->i_sb)->s_gdb_count)
gdpblocks = EXT4_SB(inode->i_sb)->s_gdb_count;
/* bitmaps and block group descriptor blocks */
ret += groups + gdpblocks;
/* Blocks for super block, inode, quota and xattr blocks */
ret += EXT4_META_TRANS_BLOCKS(inode->i_sb);
return ret;
}
/*
* Calculate the total number of credits to reserve to fit
* the modification of a single pages into a single transaction,
* which may include multiple chunks of block allocations.
*
* This could be called via ext4_write_begin()
*
* We need to consider the worse case, when
* one new block per extent.
*/
int ext4_writepage_trans_blocks(struct inode *inode)
{
int bpp = ext4_journal_blocks_per_page(inode);
int ret;
ret = ext4_meta_trans_blocks(inode, bpp, bpp);
/* Account for data blocks for journalled mode */
if (ext4_should_journal_data(inode))
ret += bpp;
return ret;
}
/*
* Calculate the journal credits for a chunk of data modification.
*
* This is called from DIO, fallocate or whoever calling
* ext4_map_blocks() to map/allocate a chunk of contiguous disk blocks.
*
* journal buffers for data blocks are not included here, as DIO
* and fallocate do no need to journal data buffers.
*/
int ext4_chunk_trans_blocks(struct inode *inode, int nrblocks)
{
return ext4_meta_trans_blocks(inode, nrblocks, 1);
}
/*
* The caller must have previously called ext4_reserve_inode_write().
* Give this, we know that the caller already has write access to iloc->bh.
*/
int ext4_mark_iloc_dirty(handle_t *handle,
struct inode *inode, struct ext4_iloc *iloc)
{
int err = 0;
if (IS_I_VERSION(inode))
inode_inc_iversion(inode);
/* the do_update_inode consumes one bh->b_count */
get_bh(iloc->bh);
/* ext4_do_update_inode() does jbd2_journal_dirty_metadata */
err = ext4_do_update_inode(handle, inode, iloc);
put_bh(iloc->bh);
return err;
}
/*
* On success, We end up with an outstanding reference count against
* iloc->bh. This _must_ be cleaned up later.
*/
int
ext4_reserve_inode_write(handle_t *handle, struct inode *inode,
struct ext4_iloc *iloc)
{
int err;
err = ext4_get_inode_loc(inode, iloc);
if (!err) {
BUFFER_TRACE(iloc->bh, "get_write_access");
err = ext4_journal_get_write_access(handle, iloc->bh);
if (err) {
brelse(iloc->bh);
iloc->bh = NULL;
}
}
ext4_std_error(inode->i_sb, err);
return err;
}
ext4: Expand extra_inodes space per the s_{want,min}_extra_isize fields We need to make sure that existing ext3 filesystems can also avail the new fields that have been added to the ext4 inode. We use s_want_extra_isize and s_min_extra_isize to decide by how much we should expand the inode. If EXT4_FEATURE_RO_COMPAT_EXTRA_ISIZE feature is set then we expand the inode by max(s_want_extra_isize, s_min_extra_isize , sizeof(ext4_inode) - EXT4_GOOD_OLD_INODE_SIZE) bytes. Actually it is still an open question about whether users should be able to set s_*_extra_isize smaller than the known fields or not. This patch also adds the functionality to expand inodes to include the newly added fields. We start by trying to expand by s_want_extra_isize bytes and if its fails we try to expand by s_min_extra_isize bytes. This is done by changing the i_extra_isize if enough space is available in the inode and no EAs are present. If EAs are present and there is enough space in the inode then the EAs in the inode are shifted to make space. If enough space is not available in the inode due to the EAs then 1 or more EAs are shifted to the external EA block. In the worst case when even the external EA block does not have enough space we inform the user that some EA would need to be deleted or s_min_extra_isize would have to be reduced. Signed-off-by: Andreas Dilger <adilger@clusterfs.com> Signed-off-by: Kalpak Shah <kalpak@clusterfs.com> Signed-off-by: Mingming Cao <cmm@us.ibm.com> Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
2007-07-18 13:19:57 +00:00
/*
* Expand an inode by new_extra_isize bytes.
* Returns 0 on success or negative error number on failure.
*/
static int ext4_expand_extra_isize(struct inode *inode,
unsigned int new_extra_isize,
struct ext4_iloc iloc,
handle_t *handle)
ext4: Expand extra_inodes space per the s_{want,min}_extra_isize fields We need to make sure that existing ext3 filesystems can also avail the new fields that have been added to the ext4 inode. We use s_want_extra_isize and s_min_extra_isize to decide by how much we should expand the inode. If EXT4_FEATURE_RO_COMPAT_EXTRA_ISIZE feature is set then we expand the inode by max(s_want_extra_isize, s_min_extra_isize , sizeof(ext4_inode) - EXT4_GOOD_OLD_INODE_SIZE) bytes. Actually it is still an open question about whether users should be able to set s_*_extra_isize smaller than the known fields or not. This patch also adds the functionality to expand inodes to include the newly added fields. We start by trying to expand by s_want_extra_isize bytes and if its fails we try to expand by s_min_extra_isize bytes. This is done by changing the i_extra_isize if enough space is available in the inode and no EAs are present. If EAs are present and there is enough space in the inode then the EAs in the inode are shifted to make space. If enough space is not available in the inode due to the EAs then 1 or more EAs are shifted to the external EA block. In the worst case when even the external EA block does not have enough space we inform the user that some EA would need to be deleted or s_min_extra_isize would have to be reduced. Signed-off-by: Andreas Dilger <adilger@clusterfs.com> Signed-off-by: Kalpak Shah <kalpak@clusterfs.com> Signed-off-by: Mingming Cao <cmm@us.ibm.com> Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
2007-07-18 13:19:57 +00:00
{
struct ext4_inode *raw_inode;
struct ext4_xattr_ibody_header *header;
if (EXT4_I(inode)->i_extra_isize >= new_extra_isize)
return 0;
raw_inode = ext4_raw_inode(&iloc);
header = IHDR(inode, raw_inode);
/* No extended attributes present */
if (!ext4_test_inode_state(inode, EXT4_STATE_XATTR) ||
header->h_magic != cpu_to_le32(EXT4_XATTR_MAGIC)) {
ext4: Expand extra_inodes space per the s_{want,min}_extra_isize fields We need to make sure that existing ext3 filesystems can also avail the new fields that have been added to the ext4 inode. We use s_want_extra_isize and s_min_extra_isize to decide by how much we should expand the inode. If EXT4_FEATURE_RO_COMPAT_EXTRA_ISIZE feature is set then we expand the inode by max(s_want_extra_isize, s_min_extra_isize , sizeof(ext4_inode) - EXT4_GOOD_OLD_INODE_SIZE) bytes. Actually it is still an open question about whether users should be able to set s_*_extra_isize smaller than the known fields or not. This patch also adds the functionality to expand inodes to include the newly added fields. We start by trying to expand by s_want_extra_isize bytes and if its fails we try to expand by s_min_extra_isize bytes. This is done by changing the i_extra_isize if enough space is available in the inode and no EAs are present. If EAs are present and there is enough space in the inode then the EAs in the inode are shifted to make space. If enough space is not available in the inode due to the EAs then 1 or more EAs are shifted to the external EA block. In the worst case when even the external EA block does not have enough space we inform the user that some EA would need to be deleted or s_min_extra_isize would have to be reduced. Signed-off-by: Andreas Dilger <adilger@clusterfs.com> Signed-off-by: Kalpak Shah <kalpak@clusterfs.com> Signed-off-by: Mingming Cao <cmm@us.ibm.com> Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
2007-07-18 13:19:57 +00:00
memset((void *)raw_inode + EXT4_GOOD_OLD_INODE_SIZE, 0,
new_extra_isize);
EXT4_I(inode)->i_extra_isize = new_extra_isize;
return 0;
}
/* try to expand with EAs present */
return ext4_expand_extra_isize_ea(inode, new_extra_isize,
raw_inode, handle);
}
/*
* What we do here is to mark the in-core inode as clean with respect to inode
* dirtiness (it may still be data-dirty).
* This means that the in-core inode may be reaped by prune_icache
* without having to perform any I/O. This is a very good thing,
* because *any* task may call prune_icache - even ones which
* have a transaction open against a different journal.
*
* Is this cheating? Not really. Sure, we haven't written the
* inode out, but prune_icache isn't a user-visible syncing function.
* Whenever the user wants stuff synced (sys_sync, sys_msync, sys_fsync)
* we start and wait on commits.
*/
int ext4_mark_inode_dirty(handle_t *handle, struct inode *inode)
{
struct ext4_iloc iloc;
ext4: Expand extra_inodes space per the s_{want,min}_extra_isize fields We need to make sure that existing ext3 filesystems can also avail the new fields that have been added to the ext4 inode. We use s_want_extra_isize and s_min_extra_isize to decide by how much we should expand the inode. If EXT4_FEATURE_RO_COMPAT_EXTRA_ISIZE feature is set then we expand the inode by max(s_want_extra_isize, s_min_extra_isize , sizeof(ext4_inode) - EXT4_GOOD_OLD_INODE_SIZE) bytes. Actually it is still an open question about whether users should be able to set s_*_extra_isize smaller than the known fields or not. This patch also adds the functionality to expand inodes to include the newly added fields. We start by trying to expand by s_want_extra_isize bytes and if its fails we try to expand by s_min_extra_isize bytes. This is done by changing the i_extra_isize if enough space is available in the inode and no EAs are present. If EAs are present and there is enough space in the inode then the EAs in the inode are shifted to make space. If enough space is not available in the inode due to the EAs then 1 or more EAs are shifted to the external EA block. In the worst case when even the external EA block does not have enough space we inform the user that some EA would need to be deleted or s_min_extra_isize would have to be reduced. Signed-off-by: Andreas Dilger <adilger@clusterfs.com> Signed-off-by: Kalpak Shah <kalpak@clusterfs.com> Signed-off-by: Mingming Cao <cmm@us.ibm.com> Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
2007-07-18 13:19:57 +00:00
struct ext4_sb_info *sbi = EXT4_SB(inode->i_sb);
static unsigned int mnt_count;
int err, ret;
might_sleep();
trace_ext4_mark_inode_dirty(inode, _RET_IP_);
err = ext4_reserve_inode_write(handle, inode, &iloc);
if (ext4_handle_valid(handle) &&
EXT4_I(inode)->i_extra_isize < sbi->s_want_extra_isize &&
!ext4_test_inode_state(inode, EXT4_STATE_NO_EXPAND)) {
ext4: Expand extra_inodes space per the s_{want,min}_extra_isize fields We need to make sure that existing ext3 filesystems can also avail the new fields that have been added to the ext4 inode. We use s_want_extra_isize and s_min_extra_isize to decide by how much we should expand the inode. If EXT4_FEATURE_RO_COMPAT_EXTRA_ISIZE feature is set then we expand the inode by max(s_want_extra_isize, s_min_extra_isize , sizeof(ext4_inode) - EXT4_GOOD_OLD_INODE_SIZE) bytes. Actually it is still an open question about whether users should be able to set s_*_extra_isize smaller than the known fields or not. This patch also adds the functionality to expand inodes to include the newly added fields. We start by trying to expand by s_want_extra_isize bytes and if its fails we try to expand by s_min_extra_isize bytes. This is done by changing the i_extra_isize if enough space is available in the inode and no EAs are present. If EAs are present and there is enough space in the inode then the EAs in the inode are shifted to make space. If enough space is not available in the inode due to the EAs then 1 or more EAs are shifted to the external EA block. In the worst case when even the external EA block does not have enough space we inform the user that some EA would need to be deleted or s_min_extra_isize would have to be reduced. Signed-off-by: Andreas Dilger <adilger@clusterfs.com> Signed-off-by: Kalpak Shah <kalpak@clusterfs.com> Signed-off-by: Mingming Cao <cmm@us.ibm.com> Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
2007-07-18 13:19:57 +00:00
/*
* We need extra buffer credits since we may write into EA block
* with this same handle. If journal_extend fails, then it will
* only result in a minor loss of functionality for that inode.
* If this is felt to be critical, then e2fsck should be run to
* force a large enough s_min_extra_isize.
*/
if ((jbd2_journal_extend(handle,
EXT4_DATA_TRANS_BLOCKS(inode->i_sb))) == 0) {
ret = ext4_expand_extra_isize(inode,
sbi->s_want_extra_isize,
iloc, handle);
if (ret) {
ext4_set_inode_state(inode,
EXT4_STATE_NO_EXPAND);
if (mnt_count !=
le16_to_cpu(sbi->s_es->s_mnt_count)) {
ext4_warning(inode->i_sb,
ext4: Expand extra_inodes space per the s_{want,min}_extra_isize fields We need to make sure that existing ext3 filesystems can also avail the new fields that have been added to the ext4 inode. We use s_want_extra_isize and s_min_extra_isize to decide by how much we should expand the inode. If EXT4_FEATURE_RO_COMPAT_EXTRA_ISIZE feature is set then we expand the inode by max(s_want_extra_isize, s_min_extra_isize , sizeof(ext4_inode) - EXT4_GOOD_OLD_INODE_SIZE) bytes. Actually it is still an open question about whether users should be able to set s_*_extra_isize smaller than the known fields or not. This patch also adds the functionality to expand inodes to include the newly added fields. We start by trying to expand by s_want_extra_isize bytes and if its fails we try to expand by s_min_extra_isize bytes. This is done by changing the i_extra_isize if enough space is available in the inode and no EAs are present. If EAs are present and there is enough space in the inode then the EAs in the inode are shifted to make space. If enough space is not available in the inode due to the EAs then 1 or more EAs are shifted to the external EA block. In the worst case when even the external EA block does not have enough space we inform the user that some EA would need to be deleted or s_min_extra_isize would have to be reduced. Signed-off-by: Andreas Dilger <adilger@clusterfs.com> Signed-off-by: Kalpak Shah <kalpak@clusterfs.com> Signed-off-by: Mingming Cao <cmm@us.ibm.com> Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
2007-07-18 13:19:57 +00:00
"Unable to expand inode %lu. Delete"
" some EAs or run e2fsck.",
inode->i_ino);
mnt_count =
le16_to_cpu(sbi->s_es->s_mnt_count);
ext4: Expand extra_inodes space per the s_{want,min}_extra_isize fields We need to make sure that existing ext3 filesystems can also avail the new fields that have been added to the ext4 inode. We use s_want_extra_isize and s_min_extra_isize to decide by how much we should expand the inode. If EXT4_FEATURE_RO_COMPAT_EXTRA_ISIZE feature is set then we expand the inode by max(s_want_extra_isize, s_min_extra_isize , sizeof(ext4_inode) - EXT4_GOOD_OLD_INODE_SIZE) bytes. Actually it is still an open question about whether users should be able to set s_*_extra_isize smaller than the known fields or not. This patch also adds the functionality to expand inodes to include the newly added fields. We start by trying to expand by s_want_extra_isize bytes and if its fails we try to expand by s_min_extra_isize bytes. This is done by changing the i_extra_isize if enough space is available in the inode and no EAs are present. If EAs are present and there is enough space in the inode then the EAs in the inode are shifted to make space. If enough space is not available in the inode due to the EAs then 1 or more EAs are shifted to the external EA block. In the worst case when even the external EA block does not have enough space we inform the user that some EA would need to be deleted or s_min_extra_isize would have to be reduced. Signed-off-by: Andreas Dilger <adilger@clusterfs.com> Signed-off-by: Kalpak Shah <kalpak@clusterfs.com> Signed-off-by: Mingming Cao <cmm@us.ibm.com> Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
2007-07-18 13:19:57 +00:00
}
}
}
}
if (!err)
err = ext4_mark_iloc_dirty(handle, inode, &iloc);
return err;
}
/*
* ext4_dirty_inode() is called from __mark_inode_dirty()
*
* We're really interested in the case where a file is being extended.
* i_size has been changed by generic_commit_write() and we thus need
* to include the updated inode in the current transaction.
*
* Also, dquot_alloc_block() will always dirty the inode when blocks
* are allocated to the file.
*
* If the inode is marked synchronous, we don't honour that here - doing
* so would cause a commit on atime updates, which we don't bother doing.
* We handle synchronous inodes at the highest possible level.
*/
void ext4_dirty_inode(struct inode *inode, int flags)
{
handle_t *handle;
handle = ext4_journal_start(inode, EXT4_HT_INODE, 2);
if (IS_ERR(handle))
goto out;
ext4_mark_inode_dirty(handle, inode);
ext4_journal_stop(handle);
out:
return;
}
#if 0
/*
* Bind an inode's backing buffer_head into this transaction, to prevent
* it from being flushed to disk early. Unlike
* ext4_reserve_inode_write, this leaves behind no bh reference and
* returns no iloc structure, so the caller needs to repeat the iloc
* lookup to mark the inode dirty later.
*/
static int ext4_pin_inode(handle_t *handle, struct inode *inode)
{
struct ext4_iloc iloc;
int err = 0;
if (handle) {
err = ext4_get_inode_loc(inode, &iloc);
if (!err) {
BUFFER_TRACE(iloc.bh, "get_write_access");
err = jbd2_journal_get_write_access(handle, iloc.bh);
if (!err)
err = ext4_handle_dirty_metadata(handle,
NULL,
iloc.bh);
brelse(iloc.bh);
}
}
ext4_std_error(inode->i_sb, err);
return err;
}
#endif
int ext4_change_inode_journal_flag(struct inode *inode, int val)
{
journal_t *journal;
handle_t *handle;
int err;
/*
* We have to be very careful here: changing a data block's
* journaling status dynamically is dangerous. If we write a
* data block to the journal, change the status and then delete
* that block, we risk forgetting to revoke the old log record
* from the journal and so a subsequent replay can corrupt data.
* So, first we make sure that the journal is empty and that
* nobody is changing anything.
*/
journal = EXT4_JOURNAL(inode);
if (!journal)
return 0;
if (is_journal_aborted(journal))
return -EROFS;
/* We have to allocate physical blocks for delalloc blocks
* before flushing journal. otherwise delalloc blocks can not
* be allocated any more. even more truncate on delalloc blocks
* could trigger BUG by flushing delalloc blocks in journal.
* There is no delalloc block in non-journal data mode.
*/
if (val && test_opt(inode->i_sb, DELALLOC)) {
err = ext4_alloc_da_blocks(inode);
if (err < 0)
return err;
}
/* Wait for all existing dio workers */
ext4_inode_block_unlocked_dio(inode);
inode_dio_wait(inode);
jbd2_journal_lock_updates(journal);
/*
* OK, there are no updates running now, and all cached data is
* synced to disk. We are now in a completely consistent state
* which doesn't have anything in the journal, and we know that
* no filesystem updates are running, so it is safe to modify
* the inode's in-core data-journaling state flag now.
*/
if (val)
ext4_set_inode_flag(inode, EXT4_INODE_JOURNAL_DATA);
else {
jbd2_journal_flush(journal);
ext4_clear_inode_flag(inode, EXT4_INODE_JOURNAL_DATA);
}
ext4_set_aops(inode);
jbd2_journal_unlock_updates(journal);
ext4_inode_resume_unlocked_dio(inode);
/* Finally we can mark the inode as dirty. */
handle = ext4_journal_start(inode, EXT4_HT_INODE, 1);
if (IS_ERR(handle))
return PTR_ERR(handle);
err = ext4_mark_inode_dirty(handle, inode);
ext4_handle_sync(handle);
ext4_journal_stop(handle);
ext4_std_error(inode->i_sb, err);
return err;
}
static int ext4_bh_unmapped(handle_t *handle, struct buffer_head *bh)
{
return !buffer_mapped(bh);
}
int ext4_page_mkwrite(struct vm_area_struct *vma, struct vm_fault *vmf)
{
struct page *page = vmf->page;
loff_t size;
unsigned long len;
int ret;
struct file *file = vma->vm_file;
struct inode *inode = file_inode(file);
struct address_space *mapping = inode->i_mapping;
handle_t *handle;
get_block_t *get_block;
int retries = 0;
sb_start_pagefault(inode->i_sb);
file_update_time(vma->vm_file);
/* Delalloc case is easy... */
if (test_opt(inode->i_sb, DELALLOC) &&
!ext4_should_journal_data(inode) &&
!ext4_nonda_switch(inode->i_sb)) {
do {
ret = __block_page_mkwrite(vma, vmf,
ext4_da_get_block_prep);
} while (ret == -ENOSPC &&
ext4_should_retry_alloc(inode->i_sb, &retries));
goto out_ret;
}
lock_page(page);
size = i_size_read(inode);
/* Page got truncated from under us? */
if (page->mapping != mapping || page_offset(page) > size) {
unlock_page(page);
ret = VM_FAULT_NOPAGE;
goto out;
}
if (page->index == size >> PAGE_CACHE_SHIFT)
len = size & ~PAGE_CACHE_MASK;
else
len = PAGE_CACHE_SIZE;
/*
* Return if we have all the buffers mapped. This avoids the need to do
* journal_start/journal_stop which can block and take a long time
*/
if (page_has_buffers(page)) {
if (!ext4_walk_page_buffers(NULL, page_buffers(page),
0, len, NULL,
ext4_bh_unmapped)) {
/* Wait so that we don't change page under IO */
mm: only enforce stable page writes if the backing device requires it Create a helper function to check if a backing device requires stable page writes and, if so, performs the necessary wait. Then, make it so that all points in the memory manager that handle making pages writable use the helper function. This should provide stable page write support to most filesystems, while eliminating unnecessary waiting for devices that don't require the feature. Before this patchset, all filesystems would block, regardless of whether or not it was necessary. ext3 would wait, but still generate occasional checksum errors. The network filesystems were left to do their own thing, so they'd wait too. After this patchset, all the disk filesystems except ext3 and btrfs will wait only if the hardware requires it. ext3 (if necessary) snapshots pages instead of blocking, and btrfs provides its own bdi so the mm will never wait. Network filesystems haven't been touched, so either they provide their own stable page guarantees or they don't block at all. The blocking behavior is back to what it was before 3.0 if you don't have a disk requiring stable page writes. Here's the result of using dbench to test latency on ext2: 3.8.0-rc3: Operation Count AvgLat MaxLat ---------------------------------------- WriteX 109347 0.028 59.817 ReadX 347180 0.004 3.391 Flush 15514 29.828 287.283 Throughput 57.429 MB/sec 4 clients 4 procs max_latency=287.290 ms 3.8.0-rc3 + patches: WriteX 105556 0.029 4.273 ReadX 335004 0.005 4.112 Flush 14982 30.540 298.634 Throughput 55.4496 MB/sec 4 clients 4 procs max_latency=298.650 ms As you can see, the maximum write latency drops considerably with this patch enabled. The other filesystems (ext3/ext4/xfs/btrfs) behave similarly, but see the cover letter for those results. Signed-off-by: Darrick J. Wong <darrick.wong@oracle.com> Acked-by: Steven Whitehouse <swhiteho@redhat.com> Reviewed-by: Jan Kara <jack@suse.cz> Cc: Adrian Hunter <adrian.hunter@intel.com> Cc: Andy Lutomirski <luto@amacapital.net> Cc: Artem Bityutskiy <dedekind1@gmail.com> Cc: Joel Becker <jlbec@evilplan.org> Cc: Mark Fasheh <mfasheh@suse.com> Cc: Jens Axboe <axboe@kernel.dk> Cc: Eric Van Hensbergen <ericvh@gmail.com> Cc: Ron Minnich <rminnich@sandia.gov> Cc: Latchesar Ionkov <lucho@ionkov.net> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2013-02-22 00:42:51 +00:00
wait_for_stable_page(page);
ret = VM_FAULT_LOCKED;
goto out;
}
}
unlock_page(page);
/* OK, we need to fill the hole... */
if (ext4_should_dioread_nolock(inode))
get_block = ext4_get_block_write;
else
get_block = ext4_get_block;
retry_alloc:
handle = ext4_journal_start(inode, EXT4_HT_WRITE_PAGE,
ext4_writepage_trans_blocks(inode));
if (IS_ERR(handle)) {
ret = VM_FAULT_SIGBUS;
goto out;
}
ret = __block_page_mkwrite(vma, vmf, get_block);
if (!ret && ext4_should_journal_data(inode)) {
if (ext4_walk_page_buffers(handle, page_buffers(page), 0,
PAGE_CACHE_SIZE, NULL, do_journal_get_write_access)) {
unlock_page(page);
ret = VM_FAULT_SIGBUS;
ext4_journal_stop(handle);
goto out;
}
ext4_set_inode_state(inode, EXT4_STATE_JDATA);
}
ext4_journal_stop(handle);
if (ret == -ENOSPC && ext4_should_retry_alloc(inode->i_sb, &retries))
goto retry_alloc;
out_ret:
ret = block_page_mkwrite_return(ret);
out:
sb_end_pagefault(inode->i_sb);
return ret;
}