linux-toradex.git/fs/ext4, branch v4.4.77 Linux kernel for Apalis and Colibri modules ext4: check return value of kstrtoull correctly in reserved_clusters_store 2017-07-15T09:57:50+00:00 Chao Yu yuchao0@huawei.com 2017-06-23T05:08:22+00:00 ad5a88c54cd299f3bb2bf53810dd5bb5805369c3 commit 1ea1516fbbab2b30bf98c534ecaacba579a35208 upstream. kstrtoull returns 0 on success, however, in reserved_clusters_store we will return -EINVAL if kstrtoull returns 0, it makes us fail to update reserved_clusters value through sysfs. Fixes: 76d33bca5581b1dd5c3157fa168db849a784ada4 Signed-off-by: Chao Yu <yuchao0@huawei.com> Signed-off-by: Miao Xie <miaoxie@huawei.com> Signed-off-by: Theodore Ts'o <tytso@mit.edu> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org> 
commit 1ea1516fbbab2b30bf98c534ecaacba579a35208 upstream.

kstrtoull returns 0 on success, however, in reserved_clusters_store we
will return -EINVAL if kstrtoull returns 0, it makes us fail to update
reserved_clusters value through sysfs.

Fixes: 76d33bca5581b1dd5c3157fa168db849a784ada4
Signed-off-by: Chao Yu <yuchao0@huawei.com>
Signed-off-by: Miao Xie <miaoxie@huawei.com>
Signed-off-by: Theodore Ts'o <tytso@mit.edu>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
fs: add i_blocksize() 2017-06-14T11:16:24+00:00 Fabian Frederick fabf@skynet.be 2017-02-27T22:28:32+00:00 044470266a5040585093e863f163f49024c3e459 commit 93407472a21b82f39c955ea7787e5bc7da100642 upstream. Replace all 1 << inode->i_blkbits and (1 << inode->i_blkbits) in fs branch. This patch also fixes multiple checkpatch warnings: WARNING: Prefer 'unsigned int' to bare use of 'unsigned' Thanks to Andrew Morton for suggesting more appropriate function instead of macro. [geliangtang@gmail.com: truncate: use i_blocksize()] Link: http://lkml.kernel.org/r/9c8b2cd83c8f5653805d43debde9fa8817e02fc4.1484895804.git.geliangtang@gmail.com Link: http://lkml.kernel.org/r/1481319905-10126-1-git-send-email-fabf@skynet.be Signed-off-by: Fabian Frederick <fabf@skynet.be> Signed-off-by: Geliang Tang <geliangtang@gmail.com> Cc: Alexander Viro <viro@zeniv.linux.org.uk> Cc: Ross Zwisler <ross.zwisler@linux.intel.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org> 
commit 93407472a21b82f39c955ea7787e5bc7da100642 upstream.

Replace all 1 << inode->i_blkbits and (1 << inode->i_blkbits) in fs
branch.

This patch also fixes multiple checkpatch warnings: WARNING: Prefer
'unsigned int' to bare use of 'unsigned'

Thanks to Andrew Morton for suggesting more appropriate function instead
of macro.

[geliangtang@gmail.com: truncate: use i_blocksize()]
  Link: http://lkml.kernel.org/r/9c8b2cd83c8f5653805d43debde9fa8817e02fc4.1484895804.git.geliangtang@gmail.com
Link: http://lkml.kernel.org/r/1481319905-10126-1-git-send-email-fabf@skynet.be
Signed-off-by: Fabian Frederick <fabf@skynet.be>
Signed-off-by: Geliang Tang <geliangtang@gmail.com>
Cc: Alexander Viro <viro@zeniv.linux.org.uk>
Cc: Ross Zwisler <ross.zwisler@linux.intel.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
ext4: fix fdatasync(2) after extent manipulation operations 2017-06-14T11:16:22+00:00 Jan Kara jack@suse.cz 2017-05-29T17:24:55+00:00 daa1357ff346a67b68d03e0450de2e87a71d2ddf commit 67a7d5f561f469ad2fa5154d2888258ab8e6df7c upstream. Currently, extent manipulation operations such as hole punch, range zeroing, or extent shifting do not record the fact that file data has changed and thus fdatasync(2) has a work to do. As a result if we crash e.g. after a punch hole and fdatasync, user can still possibly see the punched out data after journal replay. Test generic/392 fails due to these problems. Fix the problem by properly marking that file data has changed in these operations. Fixes: a4bb6b64e39abc0e41ca077725f2a72c868e7622 Signed-off-by: Jan Kara <jack@suse.cz> Signed-off-by: Theodore Ts'o <tytso@mit.edu> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org> 
commit 67a7d5f561f469ad2fa5154d2888258ab8e6df7c upstream.

Currently, extent manipulation operations such as hole punch, range
zeroing, or extent shifting do not record the fact that file data has
changed and thus fdatasync(2) has a work to do. As a result if we crash
e.g. after a punch hole and fdatasync, user can still possibly see the
punched out data after journal replay. Test generic/392 fails due to
these problems.

Fix the problem by properly marking that file data has changed in these
operations.

Fixes: a4bb6b64e39abc0e41ca077725f2a72c868e7622
Signed-off-by: Jan Kara <jack@suse.cz>
Signed-off-by: Theodore Ts'o <tytso@mit.edu>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
ext4: keep existing extra fields when inode expands 2017-06-14T11:16:22+00:00 Konstantin Khlebnikov khlebnikov@yandex-team.ru 2017-05-22T02:36:23+00:00 7b9694cb7bf2fcd2b443807423b4b09fbbc3c4ff commit 887a9730614727c4fff7cb756711b190593fc1df upstream. ext4_expand_extra_isize() should clear only space between old and new size. Fixes: 6dd4ee7cab7e # v2.6.23 Signed-off-by: Konstantin Khlebnikov <khlebnikov@yandex-team.ru> Signed-off-by: Theodore Ts'o <tytso@mit.edu> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org> 
commit 887a9730614727c4fff7cb756711b190593fc1df upstream.

ext4_expand_extra_isize() should clear only space between old and new
size.

Fixes: 6dd4ee7cab7e # v2.6.23
Signed-off-by: Konstantin Khlebnikov <khlebnikov@yandex-team.ru>
Signed-off-by: Theodore Ts'o <tytso@mit.edu>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
ext4: fix SEEK_HOLE 2017-06-14T11:16:22+00:00 Jan Kara jack@suse.cz 2017-05-22T02:33:23+00:00 08dc390b2745e0eb26375758534f52ce4eb983ae commit 7d95eddf313c88b24f99d4ca9c2411a4b82fef33 upstream. Currently, SEEK_HOLE implementation in ext4 may both return that there's a hole at some offset although that offset already has data and skip some holes during a search for the next hole. The first problem is demostrated by: xfs_io -c "falloc 0 256k" -c "pwrite 0 56k" -c "seek -h 0" file wrote 57344/57344 bytes at offset 0 56 KiB, 14 ops; 0.0000 sec (2.054 GiB/sec and 538461.5385 ops/sec) Whence Result HOLE 0 Where we can see that SEEK_HOLE wrongly returned offset 0 as containing a hole although we have written data there. The second problem can be demonstrated by: xfs_io -c "falloc 0 256k" -c "pwrite 0 56k" -c "pwrite 128k 8k" -c "seek -h 0" file wrote 57344/57344 bytes at offset 0 56 KiB, 14 ops; 0.0000 sec (1.978 GiB/sec and 518518.5185 ops/sec) wrote 8192/8192 bytes at offset 131072 8 KiB, 2 ops; 0.0000 sec (2 GiB/sec and 500000.0000 ops/sec) Whence Result HOLE 139264 Where we can see that hole at offsets 56k..128k has been ignored by the SEEK_HOLE call. The underlying problem is in the ext4_find_unwritten_pgoff() which is just buggy. In some cases it fails to update returned offset when it finds a hole (when no pages are found or when the first found page has higher index than expected), in some cases conditions for detecting hole are just missing (we fail to detect a situation where indices of returned pages are not contiguous). Fix ext4_find_unwritten_pgoff() to properly detect non-contiguous page indices and also handle all cases where we got less pages then expected in one place and handle it properly there. Fixes: c8c0df241cc2719b1262e627f999638411934f60 CC: Zheng Liu <wenqing.lz@taobao.com> Signed-off-by: Jan Kara <jack@suse.cz> Signed-off-by: Theodore Ts'o <tytso@mit.edu> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org> 
commit 7d95eddf313c88b24f99d4ca9c2411a4b82fef33 upstream.

Currently, SEEK_HOLE implementation in ext4 may both return that there's
a hole at some offset although that offset already has data and skip
some holes during a search for the next hole. The first problem is
demostrated by:

xfs_io -c "falloc 0 256k" -c "pwrite 0 56k" -c "seek -h 0" file
wrote 57344/57344 bytes at offset 0
56 KiB, 14 ops; 0.0000 sec (2.054 GiB/sec and 538461.5385 ops/sec)
Whence	Result
HOLE	0

Where we can see that SEEK_HOLE wrongly returned offset 0 as containing
a hole although we have written data there. The second problem can be
demonstrated by:

xfs_io -c "falloc 0 256k" -c "pwrite 0 56k" -c "pwrite 128k 8k"
       -c "seek -h 0" file

wrote 57344/57344 bytes at offset 0
56 KiB, 14 ops; 0.0000 sec (1.978 GiB/sec and 518518.5185 ops/sec)
wrote 8192/8192 bytes at offset 131072
8 KiB, 2 ops; 0.0000 sec (2 GiB/sec and 500000.0000 ops/sec)
Whence	Result
HOLE	139264

Where we can see that hole at offsets 56k..128k has been ignored by the
SEEK_HOLE call.

The underlying problem is in the ext4_find_unwritten_pgoff() which is
just buggy. In some cases it fails to update returned offset when it
finds a hole (when no pages are found or when the first found page has
higher index than expected), in some cases conditions for detecting hole
are just missing (we fail to detect a situation where indices of
returned pages are not contiguous).

Fix ext4_find_unwritten_pgoff() to properly detect non-contiguous page
indices and also handle all cases where we got less pages then expected
in one place and handle it properly there.

Fixes: c8c0df241cc2719b1262e627f999638411934f60
CC: Zheng Liu <wenqing.lz@taobao.com>
Signed-off-by: Jan Kara <jack@suse.cz>
Signed-off-by: Theodore Ts'o <tytso@mit.edu>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
fscrypt: avoid collisions when presenting long encrypted filenames 2017-05-25T12:30:11+00:00 Eric Biggers ebiggers@google.com 2017-04-24T17:00:09+00:00 ae3d7b8931eb3151d8ad4fb8be6fe885000d3838 commit 6b06cdee81d68a8a829ad8e8d0f31d6836744af9 upstream. When accessing an encrypted directory without the key, userspace must operate on filenames derived from the ciphertext names, which contain arbitrary bytes. Since we must support filenames as long as NAME_MAX, we can't always just base64-encode the ciphertext, since that may make it too long. Currently, this is solved by presenting long names in an abbreviated form containing any needed filesystem-specific hashes (e.g. to identify a directory block), then the last 16 bytes of ciphertext. This needs to be sufficient to identify the actual name on lookup. However, there is a bug. It seems to have been assumed that due to the use of a CBC (ciphertext block chaining)-based encryption mode, the last 16 bytes (i.e. the AES block size) of ciphertext would depend on the full plaintext, preventing collisions. However, we actually use CBC with ciphertext stealing (CTS), which handles the last two blocks specially, causing them to appear "flipped". Thus, it's actually the second-to-last block which depends on the full plaintext. This caused long filenames that differ only near the end of their plaintexts to, when observed without the key, point to the wrong inode and be undeletable. For example, with ext4: # echo pass | e4crypt add_key -p 16 edir/ # seq -f "edir/abcdefghijklmnopqrstuvwxyz012345%.0f" 100000 | xargs touch # find edir/ -type f | xargs stat -c %i | sort | uniq | wc -l 100000 # sync # echo 3 > /proc/sys/vm/drop_caches # keyctl new_session # find edir/ -type f | xargs stat -c %i | sort | uniq | wc -l 2004 # rm -rf edir/ rm: cannot remove 'edir/_A7nNFi3rhkEQlJ6P,hdzluhODKOeWx5V': Structure needs cleaning ... To fix this, when presenting long encrypted filenames, encode the second-to-last block of ciphertext rather than the last 16 bytes. Although it would be nice to solve this without depending on a specific encryption mode, that would mean doing a cryptographic hash like SHA-256 which would be much less efficient. This way is sufficient for now, and it's still compatible with encryption modes like HEH which are strong pseudorandom permutations. Also, changing the presented names is still allowed at any time because they are only provided to allow applications to do things like delete encrypted directories. They're not designed to be used to persistently identify files --- which would be hard to do anyway, given that they're encrypted after all. For ease of backports, this patch only makes the minimal fix to both ext4 and f2fs. It leaves ubifs as-is, since ubifs doesn't compare the ciphertext block yet. Follow-on patches will clean things up properly and make the filesystems use a shared helper function. Fixes: 5de0b4d0cd15 ("ext4 crypto: simplify and speed up filename encryption") Reported-by: Gwendal Grignou <gwendal@chromium.org> Signed-off-by: Eric Biggers <ebiggers@google.com> Signed-off-by: Theodore Ts'o <tytso@mit.edu> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org> 
commit 6b06cdee81d68a8a829ad8e8d0f31d6836744af9 upstream.

When accessing an encrypted directory without the key, userspace must
operate on filenames derived from the ciphertext names, which contain
arbitrary bytes.  Since we must support filenames as long as NAME_MAX,
we can't always just base64-encode the ciphertext, since that may make
it too long.  Currently, this is solved by presenting long names in an
abbreviated form containing any needed filesystem-specific hashes (e.g.
to identify a directory block), then the last 16 bytes of ciphertext.
This needs to be sufficient to identify the actual name on lookup.

However, there is a bug.  It seems to have been assumed that due to the
use of a CBC (ciphertext block chaining)-based encryption mode, the last
16 bytes (i.e. the AES block size) of ciphertext would depend on the
full plaintext, preventing collisions.  However, we actually use CBC
with ciphertext stealing (CTS), which handles the last two blocks
specially, causing them to appear "flipped".  Thus, it's actually the
second-to-last block which depends on the full plaintext.

This caused long filenames that differ only near the end of their
plaintexts to, when observed without the key, point to the wrong inode
and be undeletable.  For example, with ext4:

    # echo pass | e4crypt add_key -p 16 edir/
    # seq -f "edir/abcdefghijklmnopqrstuvwxyz012345%.0f" 100000 | xargs touch
    # find edir/ -type f | xargs stat -c %i | sort | uniq | wc -l
    100000
    # sync
    # echo 3 > /proc/sys/vm/drop_caches
    # keyctl new_session
    # find edir/ -type f | xargs stat -c %i | sort | uniq | wc -l
    2004
    # rm -rf edir/
    rm: cannot remove 'edir/_A7nNFi3rhkEQlJ6P,hdzluhODKOeWx5V': Structure needs cleaning
    ...

To fix this, when presenting long encrypted filenames, encode the
second-to-last block of ciphertext rather than the last 16 bytes.

Although it would be nice to solve this without depending on a specific
encryption mode, that would mean doing a cryptographic hash like SHA-256
which would be much less efficient.  This way is sufficient for now, and
it's still compatible with encryption modes like HEH which are strong
pseudorandom permutations.  Also, changing the presented names is still
allowed at any time because they are only provided to allow applications
to do things like delete encrypted directories.  They're not designed to
be used to persistently identify files --- which would be hard to do
anyway, given that they're encrypted after all.

For ease of backports, this patch only makes the minimal fix to both
ext4 and f2fs.  It leaves ubifs as-is, since ubifs doesn't compare the
ciphertext block yet.  Follow-on patches will clean things up properly
and make the filesystems use a shared helper function.

Fixes: 5de0b4d0cd15 ("ext4 crypto: simplify and speed up filename encryption")
Reported-by: Gwendal Grignou <gwendal@chromium.org>
Signed-off-by: Eric Biggers <ebiggers@google.com>
Signed-off-by: Theodore Ts'o <tytso@mit.edu>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
fscrypt: fix context consistency check when key(s) unavailable 2017-05-25T12:30:11+00:00 Eric Biggers ebiggers@google.com 2017-04-07T17:58:37+00:00 269d8211c400b42ff08cb1e047bd80e960c2705f commit 272f98f6846277378e1758a49a49d7bf39343c02 upstream. To mitigate some types of offline attacks, filesystem encryption is designed to enforce that all files in an encrypted directory tree use the same encryption policy (i.e. the same encryption context excluding the nonce). However, the fscrypt_has_permitted_context() function which enforces this relies on comparing struct fscrypt_info's, which are only available when we have the encryption keys. This can cause two incorrect behaviors: 1. If we have the parent directory's key but not the child's key, or vice versa, then fscrypt_has_permitted_context() returned false, causing applications to see EPERM or ENOKEY. This is incorrect if the encryption contexts are in fact consistent. Although we'd normally have either both keys or neither key in that case since the master_key_descriptors would be the same, this is not guaranteed because keys can be added or removed from keyrings at any time. 2. If we have neither the parent's key nor the child's key, then fscrypt_has_permitted_context() returned true, causing applications to see no error (or else an error for some other reason). This is incorrect if the encryption contexts are in fact inconsistent, since in that case we should deny access. To fix this, retrieve and compare the fscrypt_contexts if we are unable to set up both fscrypt_infos. While this slightly hurts performance when accessing an encrypted directory tree without the key, this isn't a case we really need to be optimizing for; access *with* the key is much more important. Furthermore, the performance hit is barely noticeable given that we are already retrieving the fscrypt_context and doing two keyring searches in fscrypt_get_encryption_info(). If we ever actually wanted to optimize this case we might start by caching the fscrypt_contexts. Signed-off-by: Eric Biggers <ebiggers@google.com> Signed-off-by: Theodore Ts'o <tytso@mit.edu> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org> 
commit 272f98f6846277378e1758a49a49d7bf39343c02 upstream.

To mitigate some types of offline attacks, filesystem encryption is
designed to enforce that all files in an encrypted directory tree use
the same encryption policy (i.e. the same encryption context excluding
the nonce).  However, the fscrypt_has_permitted_context() function which
enforces this relies on comparing struct fscrypt_info's, which are only
available when we have the encryption keys.  This can cause two
incorrect behaviors:

1. If we have the parent directory's key but not the child's key, or
   vice versa, then fscrypt_has_permitted_context() returned false,
   causing applications to see EPERM or ENOKEY.  This is incorrect if
   the encryption contexts are in fact consistent.  Although we'd
   normally have either both keys or neither key in that case since the
   master_key_descriptors would be the same, this is not guaranteed
   because keys can be added or removed from keyrings at any time.

2. If we have neither the parent's key nor the child's key, then
   fscrypt_has_permitted_context() returned true, causing applications
   to see no error (or else an error for some other reason).  This is
   incorrect if the encryption contexts are in fact inconsistent, since
   in that case we should deny access.

To fix this, retrieve and compare the fscrypt_contexts if we are unable
to set up both fscrypt_infos.

While this slightly hurts performance when accessing an encrypted
directory tree without the key, this isn't a case we really need to be
optimizing for; access *with* the key is much more important.
Furthermore, the performance hit is barely noticeable given that we are
already retrieving the fscrypt_context and doing two keyring searches in
fscrypt_get_encryption_info().  If we ever actually wanted to optimize
this case we might start by caching the fscrypt_contexts.

Signed-off-by: Eric Biggers <ebiggers@google.com>
Signed-off-by: Theodore Ts'o <tytso@mit.edu>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
ext4 crypto: fix some error handling 2017-05-25T12:30:11+00:00 Dan Carpenter dan.carpenter@oracle.com 2016-04-02T22:13:38+00:00 22823e95193a0598b5681481eee9940e4128ad35 commit 4762cc3fbbd89e5fd316d6e4d3244a8984444f8d upstream. We should be testing for -ENOMEM but the minus sign is missing. Fixes: c9af28fdd449 ('ext4 crypto: don't let data integrity writebacks fail with ENOMEM') Signed-off-by: Dan Carpenter <dan.carpenter@oracle.com> Signed-off-by: Theodore Ts'o <tytso@mit.edu> Signed-off-by: Eric Biggers <ebiggers@google.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org> 
commit 4762cc3fbbd89e5fd316d6e4d3244a8984444f8d upstream.

We should be testing for -ENOMEM but the minus sign is missing.

Fixes: c9af28fdd449 ('ext4 crypto: don't let data integrity writebacks fail with ENOMEM')
Signed-off-by: Dan Carpenter <dan.carpenter@oracle.com>
Signed-off-by: Theodore Ts'o <tytso@mit.edu>
Signed-off-by: Eric Biggers <ebiggers@google.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
ext4 crypto: don't let data integrity writebacks fail with ENOMEM 2017-05-25T12:30:11+00:00 Theodore Ts'o tytso@mit.edu 2016-03-26T20:14:34+00:00 0a76f023e6f2073936cc87ff839b2aaeccc4fb9a commit c9af28fdd44922a6c10c9f8315718408af98e315 upstream. We don't want the writeback triggered from the journal commit (in data=writeback mode) to cause the journal to abort due to generic_writepages() returning an ENOMEM error. In addition, if fsync() fails with ENOMEM, most applications will probably not do the right thing. So if we are doing a data integrity sync, and ext4_encrypt() returns ENOMEM, we will submit any queued I/O to date, and then retry the allocation using GFP_NOFAIL. Google-Bug-Id: 27641567 Signed-off-by: Theodore Ts'o <tytso@mit.edu> Signed-off-by: Eric Biggers <ebiggers@google.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org> 
commit c9af28fdd44922a6c10c9f8315718408af98e315 upstream.

We don't want the writeback triggered from the journal commit (in
data=writeback mode) to cause the journal to abort due to
generic_writepages() returning an ENOMEM error.  In addition, if
fsync() fails with ENOMEM, most applications will probably not do the
right thing.

So if we are doing a data integrity sync, and ext4_encrypt() returns
ENOMEM, we will submit any queued I/O to date, and then retry the
allocation using GFP_NOFAIL.

Google-Bug-Id: 27641567

Signed-off-by: Theodore Ts'o <tytso@mit.edu>
Signed-off-by: Eric Biggers <ebiggers@google.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
ext4: evict inline data when writing to memory map 2017-05-20T12:27:01+00:00 Eric Biggers ebiggers@google.com 2017-04-30T04:10:50+00:00 a3e6be0e9447c07e874fd8662026d7ae5c823757 commit 7b4cc9787fe35b3ee2dfb1c35e22eafc32e00c33 upstream. Currently the case of writing via mmap to a file with inline data is not handled. This is maybe a rare case since it requires a writable memory map of a very small file, but it is trivial to trigger with on inline_data filesystem, and it causes the 'BUG_ON(ext4_test_inode_state(inode, EXT4_STATE_MAY_INLINE_DATA));' in ext4_writepages() to be hit: mkfs.ext4 -O inline_data /dev/vdb mount /dev/vdb /mnt xfs_io -f /mnt/file \ -c 'pwrite 0 1' \ -c 'mmap -w 0 1m' \ -c 'mwrite 0 1' \ -c 'fsync' kernel BUG at fs/ext4/inode.c:2723! invalid opcode: 0000 [#1] SMP CPU: 1 PID: 2532 Comm: xfs_io Not tainted 4.11.0-rc1-xfstests-00301-g071d9acf3d1f #633 Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS 1.10.2-20170228_101828-anatol 04/01/2014 task: ffff88003d3a8040 task.stack: ffffc90000300000 RIP: 0010:ext4_writepages+0xc89/0xf8a RSP: 0018:ffffc90000303ca0 EFLAGS: 00010283 RAX: 0000028410000000 RBX: ffff8800383fa3b0 RCX: ffffffff812afcdc RDX: 00000a9d00000246 RSI: ffffffff81e660e0 RDI: 0000000000000246 RBP: ffffc90000303dc0 R08: 0000000000000002 R09: 869618e8f99b4fa5 R10: 00000000852287a2 R11: 00000000a03b49f4 R12: ffff88003808e698 R13: 0000000000000000 R14: 7fffffffffffffff R15: 7fffffffffffffff FS: 00007fd3e53094c0(0000) GS:ffff88003e400000(0000) knlGS:0000000000000000 CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 CR2: 00007fd3e4c51000 CR3: 000000003d554000 CR4: 00000000003406e0 Call Trace: ? _raw_spin_unlock+0x27/0x2a ? kvm_clock_read+0x1e/0x20 do_writepages+0x23/0x2c ? do_writepages+0x23/0x2c __filemap_fdatawrite_range+0x80/0x87 filemap_write_and_wait_range+0x67/0x8c ext4_sync_file+0x20e/0x472 vfs_fsync_range+0x8e/0x9f ? syscall_trace_enter+0x25b/0x2d0 vfs_fsync+0x1c/0x1e do_fsync+0x31/0x4a SyS_fsync+0x10/0x14 do_syscall_64+0x69/0x131 entry_SYSCALL64_slow_path+0x25/0x25 We could try to be smart and keep the inline data in this case, or at least support delayed allocation when allocating the block, but these solutions would be more complicated and don't seem worthwhile given how rare this case seems to be. So just fix the bug by calling ext4_convert_inline_data() when we're asked to make a page writable, so that any inline data gets evicted, with the block allocated immediately. Reported-by: Nick Alcock <nick.alcock@oracle.com> Reviewed-by: Andreas Dilger <adilger@dilger.ca> Signed-off-by: Eric Biggers <ebiggers@google.com> Signed-off-by: Theodore Ts'o <tytso@mit.edu> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org> 
commit 7b4cc9787fe35b3ee2dfb1c35e22eafc32e00c33 upstream.

Currently the case of writing via mmap to a file with inline data is not
handled.  This is maybe a rare case since it requires a writable memory
map of a very small file, but it is trivial to trigger with on
inline_data filesystem, and it causes the
'BUG_ON(ext4_test_inode_state(inode, EXT4_STATE_MAY_INLINE_DATA));' in
ext4_writepages() to be hit:

    mkfs.ext4 -O inline_data /dev/vdb
    mount /dev/vdb /mnt
    xfs_io -f /mnt/file \
	-c 'pwrite 0 1' \
	-c 'mmap -w 0 1m' \
	-c 'mwrite 0 1' \
	-c 'fsync'

	kernel BUG at fs/ext4/inode.c:2723!
	invalid opcode: 0000 [#1] SMP
	CPU: 1 PID: 2532 Comm: xfs_io Not tainted 4.11.0-rc1-xfstests-00301-g071d9acf3d1f #633
	Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS 1.10.2-20170228_101828-anatol 04/01/2014
	task: ffff88003d3a8040 task.stack: ffffc90000300000
	RIP: 0010:ext4_writepages+0xc89/0xf8a
	RSP: 0018:ffffc90000303ca0 EFLAGS: 00010283
	RAX: 0000028410000000 RBX: ffff8800383fa3b0 RCX: ffffffff812afcdc
	RDX: 00000a9d00000246 RSI: ffffffff81e660e0 RDI: 0000000000000246
	RBP: ffffc90000303dc0 R08: 0000000000000002 R09: 869618e8f99b4fa5
	R10: 00000000852287a2 R11: 00000000a03b49f4 R12: ffff88003808e698
	R13: 0000000000000000 R14: 7fffffffffffffff R15: 7fffffffffffffff
	FS:  00007fd3e53094c0(0000) GS:ffff88003e400000(0000) knlGS:0000000000000000
	CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
	CR2: 00007fd3e4c51000 CR3: 000000003d554000 CR4: 00000000003406e0
	Call Trace:
	 ? _raw_spin_unlock+0x27/0x2a
	 ? kvm_clock_read+0x1e/0x20
	 do_writepages+0x23/0x2c
	 ? do_writepages+0x23/0x2c
	 __filemap_fdatawrite_range+0x80/0x87
	 filemap_write_and_wait_range+0x67/0x8c
	 ext4_sync_file+0x20e/0x472
	 vfs_fsync_range+0x8e/0x9f
	 ? syscall_trace_enter+0x25b/0x2d0
	 vfs_fsync+0x1c/0x1e
	 do_fsync+0x31/0x4a
	 SyS_fsync+0x10/0x14
	 do_syscall_64+0x69/0x131
	 entry_SYSCALL64_slow_path+0x25/0x25

We could try to be smart and keep the inline data in this case, or at
least support delayed allocation when allocating the block, but these
solutions would be more complicated and don't seem worthwhile given how
rare this case seems to be.  So just fix the bug by calling
ext4_convert_inline_data() when we're asked to make a page writable, so
that any inline data gets evicted, with the block allocated immediately.

Reported-by: Nick Alcock <nick.alcock@oracle.com>
Reviewed-by: Andreas Dilger <adilger@dilger.ca>
Signed-off-by: Eric Biggers <ebiggers@google.com>
Signed-off-by: Theodore Ts'o <tytso@mit.edu>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>