Re: [PATCH] ext4: fix uninit-value in 'ext4_evict_inode'

[Eric Biggers]

On Thu, Nov 17, 2022 at 03:36:03PM +0800, Ye Bin wrote:
> From: Ye Bin <yebin10@xxxxxxxxxx>
> 
> Syzbot found the following issue:
> =====================================================
> BUG: KMSAN: uninit-value in ext4_evict_inode+0xdd/0x26b0 fs/ext4/inode.c:180
>  ext4_evict_inode+0xdd/0x26b0 fs/ext4/inode.c:180
>  evict+0x365/0x9a0 fs/inode.c:664
>  iput_final fs/inode.c:1747 [inline]
>  iput+0x985/0xdd0 fs/inode.c:1773
>  __ext4_new_inode+0xe54/0x7ec0 fs/ext4/ialloc.c:1361
>  ext4_mknod+0x376/0x840 fs/ext4/namei.c:2844
>  vfs_mknod+0x79d/0x830 fs/namei.c:3914
>  do_mknodat+0x47d/0xaa0
>  __do_sys_mknodat fs/namei.c:3992 [inline]
>  __se_sys_mknodat fs/namei.c:3989 [inline]
>  __ia32_sys_mknodat+0xeb/0x150 fs/namei.c:3989
>  do_syscall_32_irqs_on arch/x86/entry/common.c:112 [inline]
>  __do_fast_syscall_32+0xa2/0x100 arch/x86/entry/common.c:178
>  do_fast_syscall_32+0x33/0x70 arch/x86/entry/common.c:203
>  do_SYSENTER_32+0x1b/0x20 arch/x86/entry/common.c:246
>  entry_SYSENTER_compat_after_hwframe+0x70/0x82
> 
> Uninit was created at:
>  __alloc_pages+0x9f1/0xe80 mm/page_alloc.c:5578
>  alloc_pages+0xaae/0xd80 mm/mempolicy.c:2285
>  alloc_slab_page mm/slub.c:1794 [inline]
>  allocate_slab+0x1b5/0x1010 mm/slub.c:1939
>  new_slab mm/slub.c:1992 [inline]
>  ___slab_alloc+0x10c3/0x2d60 mm/slub.c:3180
>  __slab_alloc mm/slub.c:3279 [inline]
>  slab_alloc_node mm/slub.c:3364 [inline]
>  slab_alloc mm/slub.c:3406 [inline]
>  __kmem_cache_alloc_lru mm/slub.c:3413 [inline]
>  kmem_cache_alloc_lru+0x6f3/0xb30 mm/slub.c:3429
>  alloc_inode_sb include/linux/fs.h:3117 [inline]
>  ext4_alloc_inode+0x5f/0x860 fs/ext4/super.c:1321
>  alloc_inode+0x83/0x440 fs/inode.c:259
>  new_inode_pseudo fs/inode.c:1018 [inline]
>  new_inode+0x3b/0x430 fs/inode.c:1046
>  __ext4_new_inode+0x2a7/0x7ec0 fs/ext4/ialloc.c:959
>  ext4_mkdir+0x4d5/0x1560 fs/ext4/namei.c:2992
>  vfs_mkdir+0x62a/0x870 fs/namei.c:4035
>  do_mkdirat+0x466/0x7b0 fs/namei.c:4060
>  __do_sys_mkdirat fs/namei.c:4075 [inline]
>  __se_sys_mkdirat fs/namei.c:4073 [inline]
>  __ia32_sys_mkdirat+0xc4/0x120 fs/namei.c:4073
>  do_syscall_32_irqs_on arch/x86/entry/common.c:112 [inline]
>  __do_fast_syscall_32+0xa2/0x100 arch/x86/entry/common.c:178
>  do_fast_syscall_32+0x33/0x70 arch/x86/entry/common.c:203
>  do_SYSENTER_32+0x1b/0x20 arch/x86/entry/common.c:246
>  entry_SYSENTER_compat_after_hwframe+0x70/0x82
> 
> CPU: 1 PID: 4625 Comm: syz-executor.2 Not tainted 6.1.0-rc4-syzkaller-62821-gcb231e2f67ec #0
> Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 10/26/2022
> =====================================================
> 
> Now, 'ext4_alloc_inode()' didn't init 'ei->i_flags'. If new inode failed
> before set 'ei->i_flags' in '__ext4_new_inode()', then do 'iput()'. As after
> 6bc0d63dad7f commit will access 'ei->i_flags' in 'ext4_evict_inode()' which
> will lead to access uninit-value.
> To solve above issue just init 'ei->i_flags' in 'ext4_alloc_inode()'.
> 
> Reported-by: syzbot+57b25da729eb0b88177d@xxxxxxxxxxxxxxxxxxxxxxxxx
> Fixes:6bc0d63dad7f("ext4: remove EA inode entry from mbcache on inode eviction")
> Signed-off-by: Ye Bin <yebin10@xxxxxxxxxx>
> ---
>  fs/ext4/super.c | 1 +
>  1 file changed, 1 insertion(+)
> 
> diff --git a/fs/ext4/super.c b/fs/ext4/super.c
> index ae433e1337ed..cd2590489392 100644
> --- a/fs/ext4/super.c
> +++ b/fs/ext4/super.c
> @@ -1323,6 +1323,7 @@ static struct inode *ext4_alloc_inode(struct super_block *sb)
>  		return NULL;
>  
>  	inode_set_iversion(&ei->vfs_inode, 1);
> +	ei->i_flags = 0;
>  	spin_lock_init(&ei->i_raw_lock);
>  	INIT_LIST_HEAD(&ei->i_prealloc_list);
>  	atomic_set(&ei->i_prealloc_active, 0);

Reviewed-by: Eric Biggers <ebiggers@xxxxxxxxxx>

I'm seeing this when running xfstests on a kernel with KMSAN enabled, so it
would be nice to get this fixed.

- Eric