Currently, ext4_set_inode_mapping_order() does not set max folio order
for files with the data journalling flag. For files that already have
large folios enabled, ext4_inode_journal_mode() ignores the data
journalling flag once max folio order is set.
This is not because data journalling cannot work with large folios, but
because credit estimates will go through the roof if there are too many
blocks per folio.
Since the real constraint is blocks-per-folio, to support data=journal
under LBS, we now set max folio order to be equal to min folio order for
files with the journalling flag. When LBS is disabled, the max folio order
remains unset as before.
Therefore, before ext4_change_inode_journal_flag() switches the journalling
mode, we call truncate_pagecache() to drop all page cache for that inode,
and filemap_write_and_wait() is called unconditionally.
After that, once the journalling mode has been switched, we can safely
reset the inode mapping order, and the mapping_large_folio_support() check
in ext4_inode_journal_mode() can be removed.
Suggested-by: Jan Kara <jack@suse.cz>
Suggested-by: Dan Carpenter <dan.carpenter@linaro.org>
Signed-off-by: Baokun Li <libaokun1@huawei.com>
Reviewed-by: Zhang Yi <yi.zhang@huawei.com>
Reviewed-by: Jan Kara <jack@suse.cz>
Reviewed-by: Ojaswin Mujoo <ojaswin@linux.ibm.com>
Message-ID: <20251121090654.631996-22-libaokun@huaweicloud.com>
Signed-off-by: Theodore Ts'o <tytso@mit.edu>
Use the EXT4_PG_TO_LBLK() macro to convert folio indexes to blocks to avoid
negative left shifts after supporting blocksize greater than PAGE_SIZE.
Signed-off-by: Zhihao Cheng <chengzhihao1@huawei.com>
Signed-off-by: Baokun Li <libaokun1@huawei.com>
Reviewed-by: Zhang Yi <yi.zhang@huawei.com>
Reviewed-by: Jan Kara <jack@suse.cz>
Reviewed-by: Ojaswin Mujoo <ojaswin@linux.ibm.com>
Message-ID: <20251121090654.631996-21-libaokun@huaweicloud.com>
Signed-off-by: Theodore Ts'o <tytso@mit.edu>
Use the EXT4_PG_TO_LBLK/EXT4_LBLK_TO_PG macros to complete the conversion
between folio indexes and blocks to avoid negative left/right shifts after
supporting blocksize greater than PAGE_SIZE.
Signed-off-by: Baokun Li <libaokun1@huawei.com>
Reviewed-by: Zhang Yi <yi.zhang@huawei.com>
Reviewed-by: Jan Kara <jack@suse.cz>
Reviewed-by: Ojaswin Mujoo <ojaswin@linux.ibm.com>
Message-ID: <20251121090654.631996-20-libaokun@huaweicloud.com>
Signed-off-by: Theodore Ts'o <tytso@mit.edu>
Use the EXT4_PG_TO_LBLK/EXT4_LBLK_TO_PG macros to complete the conversion
between folio indexes and blocks to avoid negative left/right shifts after
supporting blocksize greater than PAGE_SIZE.
Signed-off-by: Baokun Li <libaokun1@huawei.com>
Reviewed-by: Zhang Yi <yi.zhang@huawei.com>
Reviewed-by: Jan Kara <jack@suse.cz>
Reviewed-by: Ojaswin Mujoo <ojaswin@linux.ibm.com>
Message-ID: <20251121090654.631996-19-libaokun@huaweicloud.com>
Signed-off-by: Theodore Ts'o <tytso@mit.edu>
Use the EXT4_PG_TO_LBLK() macro to convert folio indexes to blocks to avoid
negative left shifts after supporting blocksize greater than PAGE_SIZE.
Signed-off-by: Baokun Li <libaokun1@huawei.com>
Reviewed-by: Zhang Yi <yi.zhang@huawei.com>
Reviewed-by: Jan Kara <jack@suse.cz>
Reviewed-by: Ojaswin Mujoo <ojaswin@linux.ibm.com>
Message-ID: <20251121090654.631996-18-libaokun@huaweicloud.com>
Signed-off-by: Theodore Ts'o <tytso@mit.edu>
Use the EXT4_PG_TO_LBLK() macro to convert folio indexes to blocks to avoid
negative left shifts after supporting blocksize greater than PAGE_SIZE.
Signed-off-by: Baokun Li <libaokun1@huawei.com>
Reviewed-by: Zhang Yi <yi.zhang@huawei.com>
Reviewed-by: Jan Kara <jack@suse.cz>
Reviewed-by: Ojaswin Mujoo <ojaswin@linux.ibm.com>
Message-ID: <20251121090654.631996-17-libaokun@huaweicloud.com>
Signed-off-by: Theodore Ts'o <tytso@mit.edu>
The ext4 multi-block allocator now fully supports folio objects. Update
all variable names, function names, and comments to replace legacy 'page'
terminology with 'folio', improving clarity and consistency.
No functional changes.
Signed-off-by: Zhihao Cheng <chengzhihao1@huawei.com>
Signed-off-by: Baokun Li <libaokun1@huawei.com>
Reviewed-by: Zhang Yi <yi.zhang@huawei.com>
Reviewed-by: Jan Kara <jack@suse.cz>
Reviewed-by: Ojaswin Mujoo <ojaswin@linux.ibm.com>
Message-ID: <20251121090654.631996-16-libaokun@huaweicloud.com>
Signed-off-by: Theodore Ts'o <tytso@mit.edu>
We use EXT4_BAD_INO for the buddy cache inode number. This inode is not
accessed via __ext4_new_inode() or __ext4_iget(), meaning
ext4_set_inode_mapping_order() is not called to set its folio order range.
However, future block size greater than page size support requires this
inode to support large folios, and the buddy cache code already handles
BS > PS. Therefore, ext4_set_inode_mapping_order() is now explicitly
called for this specific inode to set its folio order range.
Signed-off-by: Baokun Li <libaokun1@huawei.com>
Reviewed-by: Zhang Yi <yi.zhang@huawei.com>
Reviewed-by: Jan Kara <jack@suse.cz>
Reviewed-by: Ojaswin Mujoo <ojaswin@linux.ibm.com>
Message-ID: <20251121090654.631996-15-libaokun@huaweicloud.com>
Signed-off-by: Theodore Ts'o <tytso@mit.edu>
Currently, ext4_mb_init_cache() uses blocks_per_page to calculate the
folio index and offset. However, when blocksize is larger than PAGE_SIZE,
blocks_per_page becomes zero, leading to a potential division-by-zero bug.
Since we now have the folio, we know its exact size. This allows us to
convert {blocks, groups}_per_page to {blocks, groups}_per_folio, thus
supporting block sizes greater than page size.
Signed-off-by: Baokun Li <libaokun1@huawei.com>
Reviewed-by: Zhang Yi <yi.zhang@huawei.com>
Reviewed-by: Jan Kara <jack@suse.cz>
Reviewed-by: Ojaswin Mujoo <ojaswin@linux.ibm.com>
Message-ID: <20251121090654.631996-14-libaokun@huaweicloud.com>
Signed-off-by: Theodore Ts'o <tytso@mit.edu>
Currently, ext4_mb_get_buddy_page_lock() uses blocks_per_page to calculate
folio index and offset. However, when blocksize is larger than PAGE_SIZE,
blocks_per_page becomes zero, leading to a potential division-by-zero bug.
To support BS > PS, use bytes to compute folio index and offset within
folio to get rid of blocks_per_page.
Also, since ext4_mb_get_buddy_page_lock() already fully supports folio,
rename it to ext4_mb_get_buddy_folio_lock().
Signed-off-by: Baokun Li <libaokun1@huawei.com>
Reviewed-by: Zhang Yi <yi.zhang@huawei.com>
Reviewed-by: Jan Kara <jack@suse.cz>
Reviewed-by: Ojaswin Mujoo <ojaswin@linux.ibm.com>
Message-ID: <20251121090654.631996-13-libaokun@huaweicloud.com>
Signed-off-by: Theodore Ts'o <tytso@mit.edu>
Currently, ext4_mb_load_buddy_gfp() uses blocks_per_page to calculate the
folio index and offset. However, when blocksize is larger than PAGE_SIZE,
blocks_per_page becomes zero, leading to a potential division-by-zero bug.
To support BS > PS, use bytes to compute folio index and offset within
folio to get rid of blocks_per_page.
Also, if buddy and bitmap land in the same folio, we get that folio’s ref
instead of looking it up again before updating the buddy.
Signed-off-by: Baokun Li <libaokun1@huawei.com>
Reviewed-by: Zhang Yi <yi.zhang@huawei.com>
Reviewed-by: Jan Kara <jack@suse.cz>
Reviewed-by: Ojaswin Mujoo <ojaswin@linux.ibm.com>
Message-ID: <20251121090654.631996-12-libaokun@huaweicloud.com>
Signed-off-by: Theodore Ts'o <tytso@mit.edu>
As BS > PS support is coming, all block number to page index (and
vice-versa) conversions must now go via bytes. Added EXT4_LBLK_TO_PG()
and EXT4_PG_TO_LBLK() macros to simplify these conversions and handle
both BS <= PS and BS > PS scenarios cleanly.
Suggested-by: Jan Kara <jack@suse.cz>
Signed-off-by: Baokun Li <libaokun1@huawei.com>
Reviewed-by: Zhang Yi <yi.zhang@huawei.com>
Reviewed-by: Jan Kara <jack@suse.cz>
Reviewed-by: Ojaswin Mujoo <ojaswin@linux.ibm.com>
Message-ID: <20251121090654.631996-11-libaokun@huaweicloud.com>
Signed-off-by: Theodore Ts'o <tytso@mit.edu>
No functional changes.
Signed-off-by: Baokun Li <libaokun1@huawei.com>
Reviewed-by: Zhang Yi <yi.zhang@huawei.com>
Reviewed-by: Jan Kara <jack@suse.cz>
Reviewed-by: Ojaswin Mujoo <ojaswin@linux.ibm.com>
Message-ID: <20251121090654.631996-10-libaokun@huaweicloud.com>
Signed-off-by: Theodore Ts'o <tytso@mit.edu>
In ext4_readdir(), page_cache_sync_readahead() is used to readahead mapped
physical blocks. With LBS support, this can lead to a negative right shift.
To fix this, the page index is now calculated by first converting the
physical block number (pblk) to a file position (pos) before converting
it to a page index. Also, the correct number of pages to readahead is now
passed.
Signed-off-by: Baokun Li <libaokun1@huawei.com>
Reviewed-by: Zhang Yi <yi.zhang@huawei.com>
Reviewed-by: Jan Kara <jack@suse.cz>
Reviewed-by: Pankaj Raghav <p.raghav@samsung.com>
Reviewed-by: Ojaswin Mujoo <ojaswin@linux.ibm.com>
Message-ID: <20251121090654.631996-9-libaokun@huaweicloud.com>
Signed-off-by: Theodore Ts'o <tytso@mit.edu>
ext4_calculate_overhead() used a single page for its bitmap buffer, which
worked fine when PAGE_SIZE >= block size. However, with block size greater
than page size (BS > PS) support, the bitmap can exceed a single page.
To address this, we now use kvmalloc() to allocate memory of the filesystem
block size, to properly support BS > PS.
Suggested-by: Jan Kara <jack@suse.cz>
Signed-off-by: Baokun Li <libaokun1@huawei.com>
Reviewed-by: Zhang Yi <yi.zhang@huawei.com>
Reviewed-by: Jan Kara <jack@suse.cz>
Reviewed-by: Ojaswin Mujoo <ojaswin@linux.ibm.com>
Message-ID: <20251121090654.631996-8-libaokun@huaweicloud.com>
Signed-off-by: Theodore Ts'o <tytso@mit.edu>
This commit introduces the s_min_folio_order field to the ext4_sb_info
structure. This field will store the minimum folio order required by the
current filesystem, laying groundwork for future support of block sizes
greater than PAGE_SIZE.
Signed-off-by: Baokun Li <libaokun1@huawei.com>
Reviewed-by: Zhang Yi <yi.zhang@huawei.com>
Reviewed-by: Jan Kara <jack@suse.cz>
Reviewed-by: Pankaj Raghav <p.raghav@samsung.com>
Reviewed-by: Ojaswin Mujoo <ojaswin@linux.ibm.com>
Message-ID: <20251121090654.631996-7-libaokun@huaweicloud.com>
Signed-off-by: Theodore Ts'o <tytso@mit.edu>
The dioread_nolock related processes already support large folio, so
dioread_nolock is enabled by default regardless of whether the blocksize
is less than, equal to, or greater than PAGE_SIZE.
Signed-off-by: Baokun Li <libaokun1@huawei.com>
Reviewed-by: Zhang Yi <yi.zhang@huawei.com>
Reviewed-by: Jan Kara <jack@suse.cz>
Reviewed-by: Ojaswin Mujoo <ojaswin@linux.ibm.com>
Message-ID: <20251121090654.631996-6-libaokun@huaweicloud.com>
Signed-off-by: Theodore Ts'o <tytso@mit.edu>
When preparing for bs > ps support, clean up unnecessary PAGE_SIZE
references in ext4_punch_hole().
Previously, when a hole extended beyond i_size, we aligned the hole end
upwards to PAGE_SIZE to handle partial folio invalidation. Now that
truncate_inode_pages_range() already handles partial folio invalidation
correctly, this alignment is no longer required.
However, to save pointless tail block zeroing, we still keep rounding up
to the block size here.
In addition, as Honza pointed out, when the hole end equals i_size, it
should also be rounded up to the block size. This patch fixes that as well.
Suggested-by: Jan Kara <jack@suse.cz>
Signed-off-by: Baokun Li <libaokun1@huawei.com>
Reviewed-by: Zhang Yi <yi.zhang@huawei.com>
Reviewed-by: Jan Kara <jack@suse.cz>
Reviewed-by: Ojaswin Mujoo <ojaswin@linux.ibm.com>
Message-ID: <20251121090654.631996-5-libaokun@huaweicloud.com>
Signed-off-by: Theodore Ts'o <tytso@mit.edu>
Previously, ext4_rec_len_(to|from)_disk only performed complex rec_len
conversions when PAGE_SIZE >= 65536 to reduce complexity.
However, we are soon to support file system block sizes greater than
page size, which makes these conditional checks unnecessary. Thus, these
checks are now removed.
Signed-off-by: Baokun Li <libaokun1@huawei.com>
Reviewed-by: Zhang Yi <yi.zhang@huawei.com>
Reviewed-by: Jan Kara <jack@suse.cz>
Reviewed-by: Ojaswin Mujoo <ojaswin@linux.ibm.com>
Message-ID: <20251121090654.631996-4-libaokun@huaweicloud.com>
Signed-off-by: Theodore Ts'o <tytso@mit.edu>
For bs <= ps scenarios, calculating the offset within the block is
sufficient. For bs > ps, an initial page offset calculation can lead to
incorrect behavior. Thus this redundant calculation has been removed.
Signed-off-by: Baokun Li <libaokun1@huawei.com>
Reviewed-by: Zhang Yi <yi.zhang@huawei.com>
Reviewed-by: Jan Kara <jack@suse.cz>
Reviewed-by: Ojaswin Mujoo <ojaswin@linux.ibm.com>
Message-ID: <20251121090654.631996-3-libaokun@huaweicloud.com>
Signed-off-by: Theodore Ts'o <tytso@mit.edu>
For bs <= ps scenarios, calculating the offset within the block is
sufficient. For bs > ps, an initial page offset calculation can lead to
incorrect behavior. Thus this redundant calculation has been removed.
Signed-off-by: Zhihao Cheng <chengzhihao1@huawei.com>
Signed-off-by: Baokun Li <libaokun1@huawei.com>
Reviewed-by: Zhang Yi <yi.zhang@huawei.com>
Reviewed-by: Jan Kara <jack@suse.cz>
Reviewed-by: Ojaswin Mujoo <ojaswin@linux.ibm.com>
Message-ID: <20251121090654.631996-2-libaokun@huaweicloud.com>
Signed-off-by: Theodore Ts'o <tytso@mit.edu>
Kernel commit 0a6ce20c15 ("ext4: verify orphan file size is not too big")
limits the maximum supported orphan file size to 8 << 20.
However, in e2fsprogs, the orphan file size is set to 32–512 filesystem
blocks when creating a filesystem.
With 64k block size, formatting an ext4 fs >32G gives an orphan file bigger
than the kernel allows, so mount prints an error and fails:
EXT4-fs (vdb): orphan file too big: 8650752
EXT4-fs (vdb): mount failed
To prevent this issue and allow previously created 64KB filesystems to
mount, we updates the maximum allowed orphan file size in the kernel to
512 filesystem blocks.
Fixes: 0a6ce20c15 ("ext4: verify orphan file size is not too big")
Signed-off-by: Baokun Li <libaokun1@huawei.com>
Reviewed-by: Jan Kara <jack@suse.cz>
Message-ID: <20251120134233.2994147-1-libaokun@huaweicloud.com>
Signed-off-by: Theodore Ts'o <tytso@mit.edu>
Cc: stable@kernel.org
Based on ext4(5) and fs/ext4/ext4.h.
For INCOMPAT_ENCRYPT, it's possible to create a new filesystem with that
flag without creating any encrypted inodes. ext4(5) says it adds
"support" but doesn't say whether anything's actually present like
COMPAT_RESIZE_INODE does.
Signed-off-by: Daniel Tang <danielzgtg.opensource@gmail.com>
Message-ID: <4506189.9SDvczpPoe@daniel-desktop3>
Signed-off-by: Theodore Ts'o <tytso@mit.edu>
Move the comments just before we set EXT4_EXT_MAY_ZEROOUT in
ext4_split_convert_extents.
Signed-off-by: Yang Erkun <yangerkun@huawei.com>
Message-ID: <20251112084538.1658232-4-yangerkun@huawei.com>
Signed-off-by: Theodore Ts'o <tytso@mit.edu>
Retval from ext4_map_create_blocks means we really create some blocks,
cannot happened with m_flags without EXT4_MAP_UNWRITTEN and
EXT4_MAP_MAPPED.
Reviewed-by: Zhang Yi <yi.zhang@huawei.com>
Reviewed-by: Jan Kara <jack@suse.cz>
Reviewed-by: Baokun Li <libaokun1@huawei.com>
Signed-off-by: Yang Erkun <yangerkun@huawei.com>
Message-ID: <20251112084538.1658232-3-yangerkun@huawei.com>
Signed-off-by: Theodore Ts'o <tytso@mit.edu>
This flag has been generalized to split an unwritten extent when we do
dio or dioread_nolock writeback, or to avoid merge new extents which was
created by extents split. Update some related comments too.
Reviewed-by: Zhang Yi <yi.zhang@huawei.com>
Reviewed-by: Jan Kara <jack@suse.cz>
Reviewed-by: Baokun Li <libaokun1@huawei.com>
Signed-off-by: Yang Erkun <yangerkun@huawei.com>
Message-ID: <20251112084538.1658232-2-yangerkun@huawei.com>
Signed-off-by: Theodore Ts'o <tytso@mit.edu>
When the MB_CHECK_ASSERT macro is enabled, we found that the
current validation logic in __mb_check_buddy has a gap in
detecting certain invalid buddy states, particularly related
to order-0 (bitmap) bits.
The original logic consists of three steps:
1. Validates higher-order buddies: if a higher-order bit is
set, at most one of the two corresponding lower-order bits
may be free; if a higher-order bit is clear, both lower-order
bits must be allocated (and their bitmap bits must be 0).
2. For any set bit in order-0, ensures all corresponding
higher-order bits are not free.
3. Verifies that all preallocated blocks (pa) in the group
have pa_pstart within bounds and their bitmap bits marked as
allocated.
However, this approach fails to properly validate cases where
order-0 bits are incorrectly cleared (0), allowing some invalid
configurations to pass:
corrupt integral
order 3 1 1
order 2 1 1 1 1
order 1 1 1 1 1 1 1 1 1
order 0 0 0 1 1 1 1 1 1 1 1 1 1 1 1 1 1
Here we get two adjacent free blocks at order-0 with inconsistent
higher-order state, and the right one shows the correct scenario.
The root cause is insufficient validation of order-0 zero bits.
To fix this and improve completeness without significant performance
cost, we refine the logic:
1. Maintain the top-down higher-order validation, but we no longer
check the cases where the higher-order bit is 0, as this case will
be covered in step 2.
2. Enhance order-0 checking by examining pairs of bits:
- If either bit in a pair is set (1), all corresponding
higher-order bits must not be free.
- If both bits are clear (0), then exactly one of the
corresponding higher-order bits must be free
3. Keep the preallocation (pa) validation unchanged.
This change closes the validation gap, ensuring illegal buddy states
involving order-0 are correctly detected, while removing redundant
checks and maintaining efficiency.
Fixes: c9de560ded ("ext4: Add multi block allocator for ext4")
Suggested-by: Jan Kara <jack@suse.cz>
Signed-off-by: Yongjian Sun <sunyongjian1@huawei.com>
Reviewed-by: Baokun Li <libaokun1@huawei.com>
Reviewed-by: Jan Kara <jack@suse.cz>
Message-ID: <20251106060614.631382-3-sunyongjian@huaweicloud.com>
Signed-off-by: Theodore Ts'o <tytso@mit.edu>
When the MB_CHECK_ASSERT macro is enabled, an assertion failure can
occur in __mb_check_buddy when checking preallocated blocks (pa) in
a block group:
Assertion failure in mb_free_blocks() : "groupnr == e4b->bd_group"
This happens when a pa at the very end of a block group (e.g.,
pa_pstart=32765, pa_len=3 in a group of 32768 blocks) becomes
exhausted - its pa_pstart is advanced by pa_len to 32768, which
lies in the next block group. If this exhausted pa (with pa_len == 0)
is still in the bb_prealloc_list during the buddy check, the assertion
incorrectly flags it as belonging to the wrong group. A possible
sequence is as follows:
ext4_mb_new_blocks
ext4_mb_release_context
pa->pa_pstart += EXT4_C2B(sbi, ac->ac_b_ex.fe_len)
pa->pa_len -= ac->ac_b_ex.fe_len
__mb_check_buddy
for each pa in group
ext4_get_group_no_and_offset
MB_CHECK_ASSERT(groupnr == e4b->bd_group)
To fix this, we modify the check to skip block group validation for
exhausted preallocations (where pa_len == 0). Such entries are in a
transitional state and will be removed from the list soon, so they
should not trigger an assertion. This change prevents the false
positive while maintaining the integrity of the checks for active
allocations.
Fixes: c9de560ded ("ext4: Add multi block allocator for ext4")
Signed-off-by: Yongjian Sun <sunyongjian1@huawei.com>
Reviewed-by: Baokun Li <libaokun1@huawei.com>
Reviewed-by: Jan Kara <jack@suse.cz>
Message-ID: <20251106060614.631382-2-sunyongjian@huaweicloud.com>
Signed-off-by: Theodore Ts'o <tytso@mit.edu>
Cc: stable@kernel.org
i_state_flags used on 32-bit archs, need to clear this flag when
alloc inode.
Find this issue when umount ext4, sometimes track the inode as orphan
accidently, cause ext4 mesg dump.
Fixes: acf943e976 ("ext4: fix checks for orphan inodes")
Signed-off-by: Haibo Chen <haibo.chen@nxp.com>
Reviewed-by: Baokun Li <libaokun1@huawei.com>
Reviewed-by: Zhang Yi <yi.zhang@huawei.com>
Reviewed-by: Jan Kara <jack@suse.cz>
Message-ID: <20251104-ext4-v1-1-73691a0800f9@nxp.com>
Signed-off-by: Theodore Ts'o <tytso@mit.edu>
Cc: stable@kernel.org
Copying the file system while it is mounted as read-only results in
a mount failure:
[~]# mkfs.ext4 -F /dev/sdc
[~]# mount /dev/sdc -o ro /mnt/test
[~]# dd if=/dev/sdc of=/dev/sda bs=1M
[~]# mount /dev/sda /mnt/test1
[ 1094.849826] JBD2: journal checksum error
[ 1094.850927] EXT4-fs (sda): Could not load journal inode
mount: mount /dev/sda on /mnt/test1 failed: Bad message
The process described above is just an abstracted way I came up with to
reproduce the issue. In the actual scenario, the file system was mounted
read-only and then copied while it was still mounted. It was found that
the mount operation failed. The user intended to verify the data or use
it as a backup, and this action was performed during a version upgrade.
Above issue may happen as follows:
ext4_fill_super
set_journal_csum_feature_set(sb)
if (ext4_has_metadata_csum(sb))
incompat = JBD2_FEATURE_INCOMPAT_CSUM_V3;
if (test_opt(sb, JOURNAL_CHECKSUM)
jbd2_journal_set_features(sbi->s_journal, compat, 0, incompat);
lock_buffer(journal->j_sb_buffer);
sb->s_feature_incompat |= cpu_to_be32(incompat);
//The data in the journal sb was modified, but the checksum was not
updated, so the data remaining in memory has a mismatch between the
data and the checksum.
unlock_buffer(journal->j_sb_buffer);
In this case, the journal sb copied over is in a state where the checksum
and data are inconsistent, so mounting fails.
To solve the above issue, update the checksum in memory after modifying
the journal sb.
Fixes: 4fd5ea43bc ("jbd2: checksum journal superblock")
Signed-off-by: Ye Bin <yebin10@huawei.com>
Reviewed-by: Baokun Li <libaokun1@huawei.com>
Reviewed-by: Darrick J. Wong <djwong@kernel.org>
Reviewed-by: Jan Kara <jack@suse.cz>
Message-ID: <20251103010123.3753631-1-yebin@huaweicloud.com>
Signed-off-by: Theodore Ts'o <tytso@mit.edu>
Cc: stable@kernel.org
params.mount_opts may come as potentially non-NUL-term string. Userspace
is expected to pass a NUL-term string. Add an extra check to ensure this
holds true. Note that further code utilizes strscpy_pad() so this is just
for proper informing the user of incorrect data being provided.
Found by Linux Verification Center (linuxtesting.org).
Signed-off-by: Fedor Pchelkin <pchelkin@ispras.ru>
Reviewed-by: Baokun Li <libaokun1@huawei.com>
Reviewed-by: Jan Kara <jack@suse.cz>
Message-ID: <20251101160430.222297-2-pchelkin@ispras.ru>
Signed-off-by: Theodore Ts'o <tytso@mit.edu>
Cc: stable@kernel.org
strscpy_pad() can't be used to copy a non-NUL-term string into a NUL-term
string of possibly bigger size. Commit 0efc5990bc ("string.h: Introduce
memtostr() and memtostr_pad()") provides additional information in that
regard. So if this happens, the following warning is observed:
strnlen: detected buffer overflow: 65 byte read of buffer size 64
WARNING: CPU: 0 PID: 28655 at lib/string_helpers.c:1032 __fortify_report+0x96/0xc0 lib/string_helpers.c:1032
Modules linked in:
CPU: 0 UID: 0 PID: 28655 Comm: syz-executor.3 Not tainted 6.12.54-syzkaller-00144-g5f0270f1ba00 #0
Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS 1.16.3-debian-1.16.3-2 04/01/2014
RIP: 0010:__fortify_report+0x96/0xc0 lib/string_helpers.c:1032
Call Trace:
<TASK>
__fortify_panic+0x1f/0x30 lib/string_helpers.c:1039
strnlen include/linux/fortify-string.h:235 [inline]
sized_strscpy include/linux/fortify-string.h:309 [inline]
parse_apply_sb_mount_options fs/ext4/super.c:2504 [inline]
__ext4_fill_super fs/ext4/super.c:5261 [inline]
ext4_fill_super+0x3c35/0xad00 fs/ext4/super.c:5706
get_tree_bdev_flags+0x387/0x620 fs/super.c:1636
vfs_get_tree+0x93/0x380 fs/super.c:1814
do_new_mount fs/namespace.c:3553 [inline]
path_mount+0x6ae/0x1f70 fs/namespace.c:3880
do_mount fs/namespace.c:3893 [inline]
__do_sys_mount fs/namespace.c:4103 [inline]
__se_sys_mount fs/namespace.c:4080 [inline]
__x64_sys_mount+0x280/0x300 fs/namespace.c:4080
do_syscall_x64 arch/x86/entry/common.c:52 [inline]
do_syscall_64+0x64/0x140 arch/x86/entry/common.c:83
entry_SYSCALL_64_after_hwframe+0x76/0x7e
Since userspace is expected to provide s_mount_opts field to be at most 63
characters long with the ending byte being NUL-term, use a 64-byte buffer
which matches the size of s_mount_opts, so that strscpy_pad() does its job
properly. Return with error if the user still managed to provide a
non-NUL-term string here.
Found by Linux Verification Center (linuxtesting.org) with Syzkaller.
Fixes: 8ecb790ea8 ("ext4: avoid potential buffer over-read in parse_apply_sb_mount_options()")
Cc: stable@vger.kernel.org
Signed-off-by: Fedor Pchelkin <pchelkin@ispras.ru>
Reviewed-by: Baokun Li <libaokun1@huawei.com>
Reviewed-by: Jan Kara <jack@suse.cz>
Message-ID: <20251101160430.222297-1-pchelkin@ispras.ru>
Signed-off-by: Theodore Ts'o <tytso@mit.edu>
When jbd2_journal_abort() is called, the provided error code is stored
in the journal superblock. Some existing calls hard-code -EIO even when
the actual failure is not I/O related.
This patch updates those calls to pass more accurate error codes,
allowing the superblock to record the true cause of failure. This helps
improve diagnostics and debugging clarity when analyzing journal aborts.
Signed-off-by: Wengang Wang <wen.gang.wang@oracle.com>
Reviewed-by: Zhang Yi <yi.zhang@huawei.com>
Message-ID: <20251031210501.7337-1-wen.gang.wang@oracle.com>
Signed-off-by: Theodore Ts'o <tytso@mit.edu>
There's issue when file system corrupted:
------------[ cut here ]------------
kernel BUG at fs/jbd2/transaction.c:1289!
Oops: invalid opcode: 0000 [#1] SMP KASAN PTI
CPU: 5 UID: 0 PID: 2031 Comm: mkdir Not tainted 6.18.0-rc1-next
RIP: 0010:jbd2_journal_get_create_access+0x3b6/0x4d0
RSP: 0018:ffff888117aafa30 EFLAGS: 00010202
RAX: 0000000000000000 RBX: ffff88811a86b000 RCX: ffffffff89a63534
RDX: 1ffff110200ec602 RSI: 0000000000000004 RDI: ffff888100763010
RBP: ffff888100763000 R08: 0000000000000001 R09: ffff888100763028
R10: 0000000000000003 R11: 0000000000000000 R12: 0000000000000000
R13: ffff88812c432000 R14: ffff88812c608000 R15: ffff888120bfc000
CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
CR2: 00007f91d6970c99 CR3: 00000001159c4000 CR4: 00000000000006f0
Call Trace:
<TASK>
__ext4_journal_get_create_access+0x42/0x170
ext4_getblk+0x319/0x6f0
ext4_bread+0x11/0x100
ext4_append+0x1e6/0x4a0
ext4_init_new_dir+0x145/0x1d0
ext4_mkdir+0x326/0x920
vfs_mkdir+0x45c/0x740
do_mkdirat+0x234/0x2f0
__x64_sys_mkdir+0xd6/0x120
do_syscall_64+0x5f/0xfa0
entry_SYSCALL_64_after_hwframe+0x76/0x7e
The above issue occurs with us in errors=continue mode when accompanied by
storage failures. There have been many inconsistencies in the file system
data.
In the case of file system data inconsistency, for example, if the block
bitmap of a referenced block is not set, it can lead to the situation where
a block being committed is allocated and used again. As a result, the
following condition will not be satisfied then trigger BUG_ON. Of course,
it is entirely possible to construct a problematic image that can trigger
this BUG_ON through specific operations. In fact, I have constructed such
an image and easily reproduced this issue.
Therefore, J_ASSERT() holds true only under ideal conditions, but it may
not necessarily be satisfied in exceptional scenarios. Using J_ASSERT()
directly in abnormal situations would cause the system to crash, which is
clearly not what we want. So here we directly trigger a JBD abort instead
of immediately invoking BUG_ON.
Fixes: 470decc613 ("[PATCH] jbd2: initial copy of files from jbd")
Signed-off-by: Ye Bin <yebin10@huawei.com>
Reviewed-by: Jan Kara <jack@suse.cz>
Message-ID: <20251025072657.307851-1-yebin@huaweicloud.com>
Signed-off-by: Theodore Ts'o <tytso@mit.edu>
Cc: stable@kernel.org
jbd2 journal handling code doesn't want jbd2_might_wait_for_commit()
to be placed between start_this_handle() and stop_this_handle(). So it
marks the region with rwsem_acquire_read() and rwsem_release().
However, the annotation is too strong for that purpose. We don't have
to use more than try lock annotation for that.
rwsem_acquire_read() implies:
1. might be a waiter on contention of the lock.
2. enter to the critical section of the lock.
All we need in here is to act 2, not 1. So trylock version of
annotation is sufficient for that purpose. Now that dept partially
relies on lockdep annotaions, dept interpets rwsem_acquire_read() as a
potential wait and might report a deadlock by the wait.
Replace it with trylock version of annotation.
Signed-off-by: Byungchul Park <byungchul@sk.com>
Reviewed-by: Jan Kara <jack@suse.cz>
Cc: stable@kernel.org
Message-ID: <20251024073940.1063-1-byungchul@sk.com>
Signed-off-by: Theodore Ts'o <tytso@mit.edu>
If ext4_get_inode_loc() fails (e.g. if it returns -EFSCORRUPTED),
iloc.bh will remain set to NULL. Since ext4_xattr_inode_dec_ref_all()
lacks error checking, this will lead to a null pointer dereference
in ext4_raw_inode(), called right after ext4_get_inode_loc().
Found by Linux Verification Center (linuxtesting.org) with SVACE.
Fixes: c8e008b604 ("ext4: ignore xattrs past end")
Cc: stable@kernel.org
Signed-off-by: Karina Yankevich <k.yankevich@omp.ru>
Reviewed-by: Sergey Shtylyov <s.shtylyov@omp.ru>
Reviewed-by: Baokun Li <libaokun1@huawei.com>
Message-ID: <20251022093253.3546296-1-k.yankevich@omp.ru>
Signed-off-by: Theodore Ts'o <tytso@mit.edu>
The cached ei->i_inline_size can become stale between the initial size
check and when ext4_update_inline_data()/ext4_create_inline_data() use
it. Although ext4_get_max_inline_size() reads the correct value at the
time of the check, concurrent xattr operations can modify i_inline_size
before ext4_write_lock_xattr() is acquired.
This causes ext4_update_inline_data() and ext4_create_inline_data() to
work with stale capacity values, leading to a BUG_ON() crash in
ext4_write_inline_data():
kernel BUG at fs/ext4/inline.c:1331!
BUG_ON(pos + len > EXT4_I(inode)->i_inline_size);
The race window:
1. ext4_get_max_inline_size() reads i_inline_size = 60 (correct)
2. Size check passes for 50-byte write
3. [Another thread adds xattr, i_inline_size changes to 40]
4. ext4_write_lock_xattr() acquires lock
5. ext4_update_inline_data() uses stale i_inline_size = 60
6. Attempts to write 50 bytes but only 40 bytes actually available
7. BUG_ON() triggers
Fix this by recalculating i_inline_size via ext4_find_inline_data_nolock()
immediately after acquiring xattr_sem. This ensures ext4_update_inline_data()
and ext4_create_inline_data() work with current values that are protected
from concurrent modifications.
This is similar to commit a54c4613da ("ext4: fix race writing to an
inline_data file while its xattrs are changing") which fixed i_inline_off
staleness. This patch addresses the related i_inline_size staleness issue.
Reported-by: syzbot+f3185be57d7e8dda32b8@syzkaller.appspotmail.com
Link: https://syzkaller.appspot.com/bug?extid=f3185be57d7e8dda32b8
Cc: stable@kernel.org
Signed-off-by: Deepanshu Kartikey <kartikey406@gmail.com>
Message-ID: <20251020060936.474314-1-kartikey406@gmail.com>
Signed-off-by: Theodore Ts'o <tytso@mit.edu>
To facilitate tracking the length, type, and outcome of the move extent,
add a trace point at both the entry and exit of mext_move_extent().
Signed-off-by: Zhang Yi <yi.zhang@huawei.com>
Reviewed-by: Jan Kara <jack@suse.cz>
Message-ID: <20251013015128.499308-13-yi.zhang@huaweicloud.com>
Signed-off-by: Theodore Ts'o <tytso@mit.edu>
Pass the moving extent length into mext_folio_double_lock() so that it
can acquire a higher-order folio if the length exceeds PAGE_SIZE. This
can speed up extent moving when the extent is larger than one page.
Additionally, remove the unnecessary comments from
mext_folio_double_lock().
Signed-off-by: Zhang Yi <yi.zhang@huawei.com>
Reviewed-by: Jan Kara <jack@suse.cz>
Message-ID: <20251013015128.499308-12-yi.zhang@huaweicloud.com>
Signed-off-by: Theodore Ts'o <tytso@mit.edu>
Now that we have mext_move_extent(), we can switch to this new interface
and deprecate move_extent_per_page(). First, after acquiring the
i_rwsem, we can directly use ext4_map_blocks() to obtain a contiguous
extent from the original inode as the extent to be moved. It can and
it's safe to get mapping information from the extent status tree without
needing to access the ondisk extent tree, because ext4_move_extent()
will check the sequence cookie under the folio lock. Then, after
populating the mext_data structure, we call ext4_move_extent() to move
the extent. Finally, the length of the extent will be adjusted in
mext.orig_map.m_len and the actual length moved is returned through
m_len.
Signed-off-by: Zhang Yi <yi.zhang@huawei.com>
Reviewed-by: Jan Kara <jack@suse.cz>
Message-ID: <20251013015128.499308-11-yi.zhang@huaweicloud.com>
Signed-off-by: Theodore Ts'o <tytso@mit.edu>
When moving extents, the current move_extent_per_page() process can only
move extents of length PAGE_SIZE at a time, which is highly inefficient,
especially when the fragmentation of the file is not particularly
severe, this will result in a large number of unnecessary extent split
and merge operations. Moreover, since the ext4 file system now supports
large folios, using PAGE_SIZE as the processing unit is no longer
practical.
Therefore, introduce a new move extents method, mext_move_extent(). It
moves one extent of the origin inode at a time, but not exceeding the
size of a folio. The parameters for the move are passed through the new
mext_data data structure, which includes the origin inode, donor inode,
the mapping extent of the origin inode to be moved, and the starting
offset of the donor inode.
The move process is similar to move_extent_per_page() and can be
categorized into three types: MEXT_SKIP_EXTENT, MEXT_MOVE_EXTENT, and
MEXT_COPY_DATA. MEXT_SKIP_EXTENT indicates that the corresponding area
of the donor file is a hole, meaning no actual space is allocated, so
the move is skipped. MEXT_MOVE_EXTENT indicates that the corresponding
areas of both the origin and donor files are unwritten, so no data needs
to be copied; only the extents are swapped. MEXT_COPY_DATA indicates
that the corresponding areas of both the origin and donor files contain
data, so data must be copied. The data copying is performed in three
steps: first, the data from the original location is read into the page
cache; then, the extents are swapped, and the page cache is rebuilt to
reflect the index of the physical blocks; finally, the dirty page cache
is marked and written back to ensure that the data is written to disk
before the metadata is persisted.
One important point to note is that the folio lock and i_data_sem are
held only during the moving process. Therefore, before moving an extent,
it is necessary to check whether the sequence cookie of the area to be
moved has changed while holding the folio lock. If a change is detected,
it indicates that concurrent write-back operations may have occurred
during this period, and the type of the extent to be moved can no longer
be considered reliable. For example, it may have changed from unwritten
to written. In such cases, return -ESTALE, and the calling function
should reacquire the move extent of the original file and retry the
movement.
Signed-off-by: Zhang Yi <yi.zhang@huawei.com>
Reviewed-by: Jan Kara <jack@suse.cz>
Message-ID: <20251013015128.499308-10-yi.zhang@huaweicloud.com>
Signed-off-by: Theodore Ts'o <tytso@mit.edu>
mext_page_mkuptodate() no longer works on a single page, so rename it to
mext_folio_mkuptodate().
Signed-off-by: Zhang Yi <yi.zhang@huawei.com>
Reviewed-by: Jan Kara <jack@suse.cz>
Message-ID: <20251013015128.499308-9-yi.zhang@huaweicloud.com>
Signed-off-by: Theodore Ts'o <tytso@mit.edu>
When moving extents, mext_check_validity() performs some basic file
system and file checks. However, some essential checks need to be
performed after acquiring the i_rwsem are still scattered in
mext_check_arguments(). Move those checks into mext_check_validity() and
make it executes entirely under the i_rwsem to make the checks clearer.
Furthermore, rename mext_check_arguments() to mext_check_adjust_range(),
as it only performs checks and length adjustments on the move extent
range. Finally, also change the print message for the non-existent file
check to be consistent with other unsupported checks.
Signed-off-by: Zhang Yi <yi.zhang@huawei.com>
Reviewed-by: Jan Kara <jack@suse.cz>
Message-ID: <20251013015128.499308-8-yi.zhang@huaweicloud.com>
Signed-off-by: Theodore Ts'o <tytso@mit.edu>
Currently, the basic validation checks during the move extent operation
are scattered across __ext4_ioctl() and ext4_move_extents(), which makes
the code somewhat disorganized. Introduce a new helper,
mext_check_validity(), to handle these checks. This change involves only
code relocation without any logical modifications.
Signed-off-by: Zhang Yi <yi.zhang@huawei.com>
Reviewed-by: Jan Kara <jack@suse.cz>
Message-ID: <20251013015128.499308-7-yi.zhang@huaweicloud.com>
Signed-off-by: Theodore Ts'o <tytso@mit.edu>
Switch to using EXT4_B_TO_LBLK() to calculate the EOF position of the
origin and donor inodes, instead of using open-coded calculations.
Signed-off-by: Zhang Yi <yi.zhang@huawei.com>
Reviewed-by: Jan Kara <jack@suse.cz>
Message-ID: <20251013015128.499308-6-yi.zhang@huaweicloud.com>
Signed-off-by: Theodore Ts'o <tytso@mit.edu>
When creating or querying mapping blocks using the ext4_map_blocks() and
ext4_map_{query|create}_blocks() helpers, also pass out the extent
sequence number of the block mapping info through the ext4_map_blocks
structure. This sequence number can later serve as a valid cookie within
iomap infrastructure and the move extents procedure.
Signed-off-by: Zhang Yi <yi.zhang@huawei.com>
Reviewed-by: Jan Kara <jack@suse.cz>
Message-ID: <20251013015128.499308-5-yi.zhang@huaweicloud.com>
Signed-off-by: Theodore Ts'o <tytso@mit.edu>
When querying extents in the extent status tree, we should hold the
data_sem if we want to obtain the sequence number as a valid cookie
simultaneously. However, currently, ext4_map_blocks() calls
ext4_es_lookup_extent() without holding data_sem. Therefore, we should
acquire i_es_lock instead, which also ensures that the sequence cookie
and the extent remain consistent. Consequently, make
ext4_es_lookup_extent() to pass out the sequence number when necessary.
Signed-off-by: Zhang Yi <yi.zhang@huawei.com>
Reviewed-by: Jan Kara <jack@suse.cz>
Message-ID: <20251013015128.499308-4-yi.zhang@huaweicloud.com>
Signed-off-by: Theodore Ts'o <tytso@mit.edu>
Baokun Li
Zhihao Cheng
Daniel Tang
Haodong Tian
Yang Erkun
Yongjian Sun
Alexey Nepomnyashih
Haibo Chen
Ye Bin
Fedor Pchelkin
Wengang Wang
Byungchul Park
Tetsuo Handa
Karina Yankevich
Deepanshu Kartikey
Zhang Yi