linux-kernel - Re: [PATCH 4.20 079/145] ext4: check for shutdown and r/o file system in ext4_write

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives
Hash Suite: Windows password security audit tool. GUI, reports in PDF.
[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <20190109134418.GJ166797@sasha-vm>
Date:   Wed, 9 Jan 2019 08:44:18 -0500
From:   Sasha Levin <sashal@...nel.org>
To:     Greg Kroah-Hartman <gregkh@...uxfoundation.org>
Cc:     linux-kernel@...r.kernel.org, stable@...r.kernel.org,
        Theodore Tso <tytso@....edu>, stable@...nel.org
Subject: Re: [PATCH 4.20 079/145] ext4: check for shutdown and r/o file
 system in ext4_write_inode()

On Mon, Jan 07, 2019 at 04:56:06PM -0500, Sasha Levin wrote:
>On Mon, Jan 07, 2019 at 01:31:56PM +0100, Greg Kroah-Hartman wrote:
>>4.20-stable review patch.  If anyone has any objections, please let me know.
>>
>>------------------
>>
>>From: Theodore Ts'o <tytso@....edu>
>>
>>commit 18f2c4fcebf2582f96cbd5f2238f4f354a0e4847 upstream.
>>
>>If the file system has been shut down or is read-only, then
>>ext4_write_inode() needs to bail out early.
>>
>>Also use jbd2_complete_transaction() instead of ext4_force_commit() so
>>we only force a commit if it is needed.
>>
>>Signed-off-by: Theodore Ts'o <tytso@....edu>
>>Cc: stable@...nel.org
>>Signed-off-by: Greg Kroah-Hartman <gregkh@...uxfoundation.org>
>>
>>---
>>fs/ext4/inode.c |    9 +++++++--
>>1 file changed, 7 insertions(+), 2 deletions(-)
>>
>>--- a/fs/ext4/inode.c
>>+++ b/fs/ext4/inode.c
>>@@ -5400,9 +5400,13 @@ int ext4_write_inode(struct inode *inode
>>{
>>	int err;
>>
>>-	if (WARN_ON_ONCE(current->flags & PF_MEMALLOC))
>>+	if (WARN_ON_ONCE(current->flags & PF_MEMALLOC) ||
>>+	    sb_rdonly(inode->i_sb))
>>		return 0;
>>
>>+	if (unlikely(ext4_forced_shutdown(EXT4_SB(inode->i_sb))))
>>+		return -EIO;
>>+
>>	if (EXT4_SB(inode->i_sb)->s_journal) {
>>		if (ext4_journal_current_handle()) {
>>			jbd_debug(1, "called recursively, non-PF_MEMALLOC!\n");
>>@@ -5418,7 +5422,8 @@ int ext4_write_inode(struct inode *inode
>>		if (wbc->sync_mode != WB_SYNC_ALL || wbc->for_sync)
>>			return 0;
>>
>>-		err = ext4_force_commit(inode->i_sb);
>>+		err = jbd2_complete_transaction(EXT4_SB(inode->i_sb)->s_journal,
>>+						EXT4_I(inode)->i_sync_tid);
>>	} else {
>>		struct ext4_iloc iloc;
>
>Hi Ted,
>
>I'm not sure if this patch is the culprit or no, but testing with
>xfstests on 4.20.1-rc1 started failing generic/390 with:
>
>[ 3405.656893] run fstests generic/390 at 2019-01-07 20:26:55
>[ 3406.524380] EXT4-fs (nvme0n1p2): mounted filesystem without journal. Opts: acl,user_xattr
>[ 3406.731410] WARNING: CPU: 0 PID: 93128 at fs/ext4/ext4_jbd2.c:53 ext4_journal_check_start+0x121/0x190
>[ 3406.731418] CPU: 0 PID: 93128 Comm: fsstress Tainted: G    B   W         4.20.1-rc1+ #1
>[ 3406.731420] Hardware name: Microsoft Corporation Virtual Machine/Virtual Machine, BIOS 090007  05/18/2018
>[ 3406.731425] RIP: 0010:ext4_journal_check_start+0x121/0x190
>[ 3406.731430] Code: 4d 85 e4 74 1e 4c 89 e2 48 b8 00 00 00 00 00 fc ff df 48 c1 ea 03 80 3c 02 00 75 4f 41 f6 04 24 02 75 0b 31 c0 5b 41 5c 5d c3 <0f> 0b eb ab 48 c7 c1 60 b4 f9 a4 ba 3d 00 00 00 48 c7 c6 a0 b8 f9
>[ 3406.731432] RSP: 0018:ffff888f30317648 EFLAGS: 00010246
>[ 3406.731437] RAX: 0000000000000000 RBX: ffff888ee6ce12c8 RCX: 000000000000000c
>[ 3406.731440] RDX: 1ffff111dcd9c2a0 RSI: 0000000000000b33 RDI: ffff888ee6ce1500
>[ 3406.731442] RBP: ffff888f30317658 R08: 0000000000000000 R09: ffffed11db5f01a7
>[ 3406.731444] R10: 0000000000000001 R11: ffffed11db5f01a6 R12: ffff888ee6ce2ee8
>[ 3406.731447] R13: dffffc0000000000 R14: 0000000000000000 R15: ffff888f303178e0
>[ 3406.731450] FS:  00007fc015d5cb80(0000) GS:ffff888f6b800000(0000) knlGS:0000000000000000
>[ 3406.731452] CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
>[ 3406.731455] CR2: 0000557f34651000 CR3: 0000000f2c722000 CR4: 00000000003406f0
>[ 3406.731459] Call Trace:
>[ 3406.731465]  __ext4_journal_start_sb+0x95/0x380
>[ 3406.731479]  ext4_writepages+0x10e1/0x3170
>[ 3406.731567]  do_writepages+0xc7/0x120
>[ 3406.731577]  __filemap_fdatawrite_range+0x313/0x470
>[ 3406.731594]  file_write_and_wait_range+0x7c/0xd0
>[ 3406.731598]  __generic_file_fsync+0x6c/0x180
>[ 3406.731601]  ext4_sync_file+0x60b/0xf90
>[ 3406.731626]  vfs_fsync_range+0xf5/0x210
>[ 3406.731630]  do_fsync+0x3d/0x70
>[ 3406.731634]  __x64_sys_fdatasync+0x36/0x50
>[ 3406.731638]  do_syscall_64+0x153/0x460
>[ 3406.731666]  entry_SYSCALL_64_after_hwframe+0x44/0xa9
>[ 3406.731669] RIP: 0033:0x7fc0152412c4
>[ 3406.731673] Code: 89 01 48 83 c8 ff c3 66 2e 0f 1f 84 00 00 00 00 00 0f 1f 44 00 00 48 8d 05 41 96 2d 00 8b 00 85 c0 75 13 b8 4b 00 00 00 0f 05 <48> 3d 00 f0 ff ff 77 3c f3 c3 66 90 53 89 fb 48 83 ec 10 e8 04 94
>[ 3406.731675] RSP: 002b:00007ffc445f72f8 EFLAGS: 00000246 ORIG_RAX: 000000000000004b
>[ 3406.731679] RAX: ffffffffffffffda RBX: 0000000000000003 RCX: 00007fc0152412c4
>[ 3406.731681] RDX: 00007ffc445f7250 RSI: 00007ffc445f7250 RDI: 0000000000000003
>[ 3406.731683] RBP: 0000000000000034 R08: 00007fc015515c40 R09: 0000000000000000
>[ 3406.731685] R10: 0000000000000000 R11: 0000000000000246 R12: 00000000000001f4
>[ 3406.731687] R13: 0000000051eb851f R14: 00007ffc445f73f6 R15: 0000557f33ad1930
>[ 3406.731690] ---[ end trace 698eba2a0467d312 ]---
>
>I'm also slightly confused here because this test is for ext4 without
>journal (see even "EXT4-fs (nvme0n1p2): mounted filesystem without
>journal") but for some reason it seems that this still goes to the
>journal?

Following up on this, I can reproduce this failure somewhat reliably
with 4.20.1-rc1 but can't with 4.20. If there's anything I can do to
help debug this please let me know.

--
Thanks,
Sasha