Re: memory leak in path_openat

Re: memory leak in path_openat

[Eric Biggers]

On Thu, Jan 02, 2020 at 04:35:09AM -0800, syzbot wrote:
> Hello,
> 
> syzbot found the following crash on:
> 
> HEAD commit:    bf8d1cd4 Merge tag 'scsi-fixes' of git://git.kernel.org/pu..
> git tree:       upstream
> console output: https://syzkaller.appspot.com/x/log.txt?x=16386971e00000
> kernel config:  https://syzkaller.appspot.com/x/.config?x=328af7338803d39a
> dashboard link: https://syzkaller.appspot.com/bug?extid=9c064b9ab4dbb724c806
> compiler:       gcc (GCC) 9.0.0 20181231 (experimental)
> syz repro:      https://syzkaller.appspot.com/x/repro.syz?x=17f4ce15e00000
> C reproducer:   https://syzkaller.appspot.com/x/repro.c?x=153c8971e00000
> 
> IMPORTANT: if you fix the bug, please add the following tag to the commit:
> Reported-by: syzbot+9c064b9ab4dbb724c806@syzkaller.appspotmail.com
> 
> BUG: memory leak
> unreferenced object 0xffff88811f95b400 (size 256):
>   comm "syz-executor609", pid 6975, jiffies 4294945087 (age 7.980s)
>   hex dump (first 32 bytes):
>     00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00  ................
>     a0 2b ab 1b 82 88 ff ff c0 3c 80 2b 81 88 ff ff  .+.......<.+....
>   backtrace:
>     [<00000000aa112990>] kmemleak_alloc_recursive
> include/linux/kmemleak.h:43 [inline]
>     [<00000000aa112990>] slab_post_alloc_hook mm/slab.h:586 [inline]
>     [<00000000aa112990>] slab_alloc mm/slab.c:3320 [inline]
>     [<00000000aa112990>] kmem_cache_alloc+0x13f/0x2c0 mm/slab.c:3484
>     [<00000000a62a216f>] kmem_cache_zalloc include/linux/slab.h:660 [inline]
>     [<00000000a62a216f>] __alloc_file+0x28/0x130 fs/file_table.c:101
>     [<00000000db4f5560>] alloc_empty_file+0x50/0xd0 fs/file_table.c:151
>     [<00000000178121b2>] path_openat+0x52/0x1dd0 fs/namei.c:3526
>     [<00000000b9f51901>] do_filp_open+0xaa/0x130 fs/namei.c:3567
>     [<000000008b6c278b>] do_sys_open+0x253/0x330 fs/open.c:1097
>     [<00000000de529158>] __do_sys_openat fs/open.c:1124 [inline]
>     [<00000000de529158>] __se_sys_openat fs/open.c:1118 [inline]
>     [<00000000de529158>] __x64_sys_openat+0x24/0x30 fs/open.c:1118
>     [<000000002f0aeb7b>] do_syscall_64+0x73/0x220
> arch/x86/entry/common.c:294
>     [<00000000720f3b5c>] entry_SYSCALL_64_after_hwframe+0x44/0xa9
> 
> 
> 
> ---
> This bug is generated by a bot. It may contain errors.
> See https://goo.gl/tpsmEJ for more information about syzbot.
> syzbot engineers can be reached at syzkaller@googlegroups.com.
> 
> syzbot will keep track of this bug report. See:
> https://goo.gl/tpsmEJ#status for how to communicate with syzbot.
> syzbot can test patches for this bug, for details see:
> https://goo.gl/tpsmEJ#testing-patches

Looks like this was an io_uring bug, but it's not reproducible anymore and the
relevant code was heavily changed recently.  So, invalidating this bug report:

#syz invalid