* [syzbot] upstream test error: KFENCE: use-after-free in kvm_fastop_exception
@ 2021-09-04 18:58 syzbot
2021-09-17 10:00 ` Dmitry Vyukov
0 siblings, 1 reply; 17+ messages in thread
From: syzbot @ 2021-09-04 18:58 UTC (permalink / raw)
To: linux-fsdevel, linux-kernel, syzkaller-bugs, viro
Hello,
syzbot found the following issue on:
HEAD commit: 835d31d319d9 Merge tag 'media/v5.15-1' of git://git.kernel..
git tree: upstream
console output: https://syzkaller.appspot.com/x/log.txt?x=1189fe49300000
kernel config: https://syzkaller.appspot.com/x/.config?x=d1a7a34dc082816f
dashboard link: https://syzkaller.appspot.com/bug?extid=d08efd12a2905a344291
compiler: gcc (Debian 10.2.1-6) 10.2.1 20210110, GNU ld (GNU Binutils for Debian) 2.35.1
IMPORTANT: if you fix the issue, please add the following tag to the commit:
Reported-by: syzbot+d08efd12a2905a344291@syzkaller.appspotmail.com
==================================================================
BUG: KFENCE: use-after-free read in kvm_fastop_exception+0xf6d/0x105b
Use-after-free read at 0xffff88823bc0c020 (in kfence-#5):
kvm_fastop_exception+0xf6d/0x105b
d_lookup+0xd8/0x170 fs/dcache.c:2370
lookup_dcache+0x1e/0x130 fs/namei.c:1520
__lookup_hash+0x29/0x180 fs/namei.c:1543
kern_path_locked+0x17e/0x320 fs/namei.c:2567
handle_remove+0xa2/0x5fe drivers/base/devtmpfs.c:312
handle drivers/base/devtmpfs.c:382 [inline]
devtmpfs_work_loop drivers/base/devtmpfs.c:395 [inline]
devtmpfsd+0x1b9/0x2a3 drivers/base/devtmpfs.c:437
kthread+0x3e5/0x4d0 kernel/kthread.c:319
ret_from_fork+0x1f/0x30 arch/x86/entry/entry_64.S:295
kfence-#5 [0xffff88823bc0c000-0xffff88823bc0cfff, size=4096, cache=names_cache] allocated by task 22:
getname_kernel+0x4e/0x370 fs/namei.c:226
kern_path_locked+0x71/0x320 fs/namei.c:2558
handle_remove+0xa2/0x5fe drivers/base/devtmpfs.c:312
handle drivers/base/devtmpfs.c:382 [inline]
devtmpfs_work_loop drivers/base/devtmpfs.c:395 [inline]
devtmpfsd+0x1b9/0x2a3 drivers/base/devtmpfs.c:437
kthread+0x3e5/0x4d0 kernel/kthread.c:319
ret_from_fork+0x1f/0x30 arch/x86/entry/entry_64.S:295
freed by task 22:
putname.part.0+0xe1/0x120 fs/namei.c:270
putname include/linux/err.h:41 [inline]
filename_parentat fs/namei.c:2547 [inline]
kern_path_locked+0xc2/0x320 fs/namei.c:2558
handle_remove+0xa2/0x5fe drivers/base/devtmpfs.c:312
handle drivers/base/devtmpfs.c:382 [inline]
devtmpfs_work_loop drivers/base/devtmpfs.c:395 [inline]
devtmpfsd+0x1b9/0x2a3 drivers/base/devtmpfs.c:437
kthread+0x3e5/0x4d0 kernel/kthread.c:319
ret_from_fork+0x1f/0x30 arch/x86/entry/entry_64.S:295
CPU: 1 PID: 22 Comm: kdevtmpfs Not tainted 5.14.0-syzkaller #0
Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 01/01/2011
RIP: 0010:kvm_fastop_exception+0xf6d/0x105b
Code: d3 ed e9 14 1b 6d f8 49 8d 0e 48 83 e1 f8 4c 8b 21 41 8d 0e 83 e1 07 c1 e1 03 49 d3 ec e9 6a 28 6d f8 49 8d 4d 00 48 83 e1 f8 <4c> 8b 21 41 8d 4d 00 83 e1 07 c1 e1 03 49 d3 ec e9 5a 32 6d f8 bd
RSP: 0018:ffffc90000fe7ae8 EFLAGS: 00010282
RAX: 0000000035736376 RBX: ffff88803b141cc0 RCX: ffff88823bc0c020
RDX: ffffed100762839f RSI: 0000000000000004 RDI: 0000000000000007
RBP: 0000000000000004 R08: 0000000000000000 R09: ffff88803b141cf0
R10: ffffed100762839e R11: 0000000000000000 R12: ffff88823bc0c020
R13: ffff88823bc0c020 R14: ffff88803b141cf0 R15: dffffc0000000000
FS: 0000000000000000(0000) GS:ffff8880b9d00000(0000) knlGS:0000000000000000
CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
CR2: ffff88823bc0c020 CR3: 0000000029892000 CR4: 00000000001506e0
DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400
Call Trace:
d_lookup+0xd8/0x170 fs/dcache.c:2370
lookup_dcache+0x1e/0x130 fs/namei.c:1520
__lookup_hash+0x29/0x180 fs/namei.c:1543
kern_path_locked+0x17e/0x320 fs/namei.c:2567
handle_remove+0xa2/0x5fe drivers/base/devtmpfs.c:312
handle drivers/base/devtmpfs.c:382 [inline]
devtmpfs_work_loop drivers/base/devtmpfs.c:395 [inline]
devtmpfsd+0x1b9/0x2a3 drivers/base/devtmpfs.c:437
kthread+0x3e5/0x4d0 kernel/kthread.c:319
ret_from_fork+0x1f/0x30 arch/x86/entry/entry_64.S:295
==================================================================
----------------
Code disassembly (best guess):
0: d3 ed shr %cl,%ebp
2: e9 14 1b 6d f8 jmpq 0xf86d1b1b
7: 49 8d 0e lea (%r14),%rcx
a: 48 83 e1 f8 and $0xfffffffffffffff8,%rcx
e: 4c 8b 21 mov (%rcx),%r12
11: 41 8d 0e lea (%r14),%ecx
14: 83 e1 07 and $0x7,%ecx
17: c1 e1 03 shl $0x3,%ecx
1a: 49 d3 ec shr %cl,%r12
1d: e9 6a 28 6d f8 jmpq 0xf86d288c
22: 49 8d 4d 00 lea 0x0(%r13),%rcx
26: 48 83 e1 f8 and $0xfffffffffffffff8,%rcx
* 2a: 4c 8b 21 mov (%rcx),%r12 <-- trapping instruction
2d: 41 8d 4d 00 lea 0x0(%r13),%ecx
31: 83 e1 07 and $0x7,%ecx
34: c1 e1 03 shl $0x3,%ecx
37: 49 d3 ec shr %cl,%r12
3a: e9 5a 32 6d f8 jmpq 0xf86d3299
3f: bd .byte 0xbd
---
This report 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 issue. See:
https://goo.gl/tpsmEJ#status for how to communicate with syzbot.
^ permalink raw reply [flat|nested] 17+ messages in thread
* Re: [syzbot] upstream test error: KFENCE: use-after-free in kvm_fastop_exception
2021-09-04 18:58 [syzbot] upstream test error: KFENCE: use-after-free in kvm_fastop_exception syzbot
@ 2021-09-17 10:00 ` Dmitry Vyukov
0 siblings, 0 replies; 17+ messages in thread
From: Dmitry Vyukov @ 2021-09-17 10:00 UTC (permalink / raw)
To: syzbot
Cc: linux-fsdevel, linux-kernel, syzkaller-bugs, viro,
the arch/x86 maintainers, Linux ARM, Marco Elver, kasan-dev
On Sat, 4 Sept 2021 at 20:58, syzbot
<syzbot+d08efd12a2905a344291@syzkaller.appspotmail.com> wrote:
>
> Hello,
>
> syzbot found the following issue on:
>
> HEAD commit: 835d31d319d9 Merge tag 'media/v5.15-1' of git://git.kernel..
> git tree: upstream
> console output: https://syzkaller.appspot.com/x/log.txt?x=1189fe49300000
> kernel config: https://syzkaller.appspot.com/x/.config?x=d1a7a34dc082816f
> dashboard link: https://syzkaller.appspot.com/bug?extid=d08efd12a2905a344291
> compiler: gcc (Debian 10.2.1-6) 10.2.1 20210110, GNU ld (GNU Binutils for Debian) 2.35.1
>
> IMPORTANT: if you fix the issue, please add the following tag to the commit:
> Reported-by: syzbot+d08efd12a2905a344291@syzkaller.appspotmail.com
>
> ==================================================================
> BUG: KFENCE: use-after-free read in kvm_fastop_exception+0xf6d/0x105b
>
> Use-after-free read at 0xffff88823bc0c020 (in kfence-#5):
> kvm_fastop_exception+0xf6d/0x105b
There is probably some bug in d_lookup, but there is also something
wrong with the unwinder. It prints an unrelated kvm_fastop_exception
frame instead of __d_lookup and interestingly a very similar thing
happens on arm64 with HWASAN and a similar bug in d_lookup. The
corresponding report is:
https://syzkaller.appspot.com/bug?extid=488ddf8087564d6de6e2
BUG: KASAN: invalid-access in __entry_tramp_text_end+0xddc/0xd000
CPU: 0 PID: 22 Comm: kdevtmpfs Not tainted
5.14.0-syzkaller-11152-g78e709522d2c #0
Hardware name: linux,dummy-virt (DT)
Call trace:
dump_backtrace+0x0/0x1ac arch/arm64/kernel/stacktrace.c:76
show_stack+0x18/0x24 arch/arm64/kernel/stacktrace.c:215
__dump_stack lib/dump_stack.c:88 [inline]
dump_stack_lvl+0x68/0x84 lib/dump_stack.c:106
print_address_description+0x7c/0x2b4 mm/kasan/report.c:256
__kasan_report mm/kasan/report.c:442 [inline]
kasan_report+0x134/0x380 mm/kasan/report.c:459
__do_kernel_fault+0x128/0x1bc arch/arm64/mm/fault.c:317
do_bad_area arch/arm64/mm/fault.c:466 [inline]
do_tag_check_fault+0x74/0x90 arch/arm64/mm/fault.c:737
do_mem_abort+0x44/0xb4 arch/arm64/mm/fault.c:813
el1_abort+0x40/0x60 arch/arm64/kernel/entry-common.c:357
el1h_64_sync_handler+0xb0/0xd0 arch/arm64/kernel/entry-common.c:408
el1h_64_sync+0x78/0x7c arch/arm64/kernel/entry.S:567
__entry_tramp_text_end+0xddc/0xd000
d_lookup+0x44/0x70 fs/dcache.c:2370
lookup_dcache+0x24/0x84 fs/namei.c:1520
__lookup_hash+0x24/0xd0 fs/namei.c:1543
kern_path_locked+0x90/0x10c fs/namei.c:2567
handle_remove+0x38/0x284 drivers/base/devtmpfs.c:312
handle drivers/base/devtmpfs.c:382 [inline]
devtmpfs_work_loop drivers/base/devtmpfs.c:395 [inline]
devtmpfsd+0x8c/0xd0 drivers/base/devtmpfs.c:437
kthread+0x150/0x15c kernel/kthread.c:319
ret_from_fork+0x10/0x20 arch/arm64/kernel/entry.S:756
Here kernel unwinder prints __entry_tramp_text_end instead of __d_lookup.
I've looked in more detail into the arm64 case:
d_lookup contains a static call to __d_lookup as expected:
ffff8000102e0780 <d_lookup>:
...
ffff8000102e07c0: 97ffffa4 bl ffff8000102e0650 <__d_lookup>
...
ffff8000102e07e8: d65f03c0 ret
and these symbols don't overlap or something:
$ aarch64-linux-gnu-nm -nS vmlinux | egrep -C 1 " (t|T)
(__entry_tramp_text|__d_lookup)"
ffff8000102e01f0 0000000000000458 T d_alloc_parallel
ffff8000102e0650 0000000000000128 T __d_lookup
ffff8000102e0780 000000000000006c T d_lookup
--
ffff8000117a1f88 T __hibernate_exit_text_end
ffff8000117a2000 T __entry_tramp_text_start
ffff8000117a2000 00000000000007c8 T tramp_vectors
--
ffff8000117a27f0 0000000000000024 T tramp_exit_compat
ffff8000117a3000 T __entry_tramp_text_end
ffff8000117b0000 D _etext
So it looks like in both cases the top fault frame is just wrong. But
I would assume it's extracted by arch-dependent code, so it's
suspicious that it affects both x86 and arm64...
Any ideas what's happening?
> d_lookup+0xd8/0x170 fs/dcache.c:2370
> lookup_dcache+0x1e/0x130 fs/namei.c:1520
> __lookup_hash+0x29/0x180 fs/namei.c:1543
> kern_path_locked+0x17e/0x320 fs/namei.c:2567
> handle_remove+0xa2/0x5fe drivers/base/devtmpfs.c:312
> handle drivers/base/devtmpfs.c:382 [inline]
> devtmpfs_work_loop drivers/base/devtmpfs.c:395 [inline]
> devtmpfsd+0x1b9/0x2a3 drivers/base/devtmpfs.c:437
> kthread+0x3e5/0x4d0 kernel/kthread.c:319
> ret_from_fork+0x1f/0x30 arch/x86/entry/entry_64.S:295
>
> kfence-#5 [0xffff88823bc0c000-0xffff88823bc0cfff, size=4096, cache=names_cache] allocated by task 22:
> getname_kernel+0x4e/0x370 fs/namei.c:226
> kern_path_locked+0x71/0x320 fs/namei.c:2558
> handle_remove+0xa2/0x5fe drivers/base/devtmpfs.c:312
> handle drivers/base/devtmpfs.c:382 [inline]
> devtmpfs_work_loop drivers/base/devtmpfs.c:395 [inline]
> devtmpfsd+0x1b9/0x2a3 drivers/base/devtmpfs.c:437
> kthread+0x3e5/0x4d0 kernel/kthread.c:319
> ret_from_fork+0x1f/0x30 arch/x86/entry/entry_64.S:295
>
> freed by task 22:
> putname.part.0+0xe1/0x120 fs/namei.c:270
> putname include/linux/err.h:41 [inline]
> filename_parentat fs/namei.c:2547 [inline]
> kern_path_locked+0xc2/0x320 fs/namei.c:2558
> handle_remove+0xa2/0x5fe drivers/base/devtmpfs.c:312
> handle drivers/base/devtmpfs.c:382 [inline]
> devtmpfs_work_loop drivers/base/devtmpfs.c:395 [inline]
> devtmpfsd+0x1b9/0x2a3 drivers/base/devtmpfs.c:437
> kthread+0x3e5/0x4d0 kernel/kthread.c:319
> ret_from_fork+0x1f/0x30 arch/x86/entry/entry_64.S:295
>
> CPU: 1 PID: 22 Comm: kdevtmpfs Not tainted 5.14.0-syzkaller #0
> Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 01/01/2011
> RIP: 0010:kvm_fastop_exception+0xf6d/0x105b
> Code: d3 ed e9 14 1b 6d f8 49 8d 0e 48 83 e1 f8 4c 8b 21 41 8d 0e 83 e1 07 c1 e1 03 49 d3 ec e9 6a 28 6d f8 49 8d 4d 00 48 83 e1 f8 <4c> 8b 21 41 8d 4d 00 83 e1 07 c1 e1 03 49 d3 ec e9 5a 32 6d f8 bd
> RSP: 0018:ffffc90000fe7ae8 EFLAGS: 00010282
> RAX: 0000000035736376 RBX: ffff88803b141cc0 RCX: ffff88823bc0c020
> RDX: ffffed100762839f RSI: 0000000000000004 RDI: 0000000000000007
> RBP: 0000000000000004 R08: 0000000000000000 R09: ffff88803b141cf0
> R10: ffffed100762839e R11: 0000000000000000 R12: ffff88823bc0c020
> R13: ffff88823bc0c020 R14: ffff88803b141cf0 R15: dffffc0000000000
> FS: 0000000000000000(0000) GS:ffff8880b9d00000(0000) knlGS:0000000000000000
> CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
> CR2: ffff88823bc0c020 CR3: 0000000029892000 CR4: 00000000001506e0
> DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
> DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400
> Call Trace:
> d_lookup+0xd8/0x170 fs/dcache.c:2370
> lookup_dcache+0x1e/0x130 fs/namei.c:1520
> __lookup_hash+0x29/0x180 fs/namei.c:1543
> kern_path_locked+0x17e/0x320 fs/namei.c:2567
> handle_remove+0xa2/0x5fe drivers/base/devtmpfs.c:312
> handle drivers/base/devtmpfs.c:382 [inline]
> devtmpfs_work_loop drivers/base/devtmpfs.c:395 [inline]
> devtmpfsd+0x1b9/0x2a3 drivers/base/devtmpfs.c:437
> kthread+0x3e5/0x4d0 kernel/kthread.c:319
> ret_from_fork+0x1f/0x30 arch/x86/entry/entry_64.S:295
> ==================================================================
> ----------------
> Code disassembly (best guess):
> 0: d3 ed shr %cl,%ebp
> 2: e9 14 1b 6d f8 jmpq 0xf86d1b1b
> 7: 49 8d 0e lea (%r14),%rcx
> a: 48 83 e1 f8 and $0xfffffffffffffff8,%rcx
> e: 4c 8b 21 mov (%rcx),%r12
> 11: 41 8d 0e lea (%r14),%ecx
> 14: 83 e1 07 and $0x7,%ecx
> 17: c1 e1 03 shl $0x3,%ecx
> 1a: 49 d3 ec shr %cl,%r12
> 1d: e9 6a 28 6d f8 jmpq 0xf86d288c
> 22: 49 8d 4d 00 lea 0x0(%r13),%rcx
> 26: 48 83 e1 f8 and $0xfffffffffffffff8,%rcx
> * 2a: 4c 8b 21 mov (%rcx),%r12 <-- trapping instruction
> 2d: 41 8d 4d 00 lea 0x0(%r13),%ecx
> 31: 83 e1 07 and $0x7,%ecx
> 34: c1 e1 03 shl $0x3,%ecx
> 37: 49 d3 ec shr %cl,%r12
> 3a: e9 5a 32 6d f8 jmpq 0xf86d3299
> 3f: bd .byte 0xbd
>
>
> ---
> This report 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 issue. See:
> https://goo.gl/tpsmEJ#status for how to communicate with syzbot.
^ permalink raw reply [flat|nested] 17+ messages in thread
* Re: [syzbot] upstream test error: KFENCE: use-after-free in kvm_fastop_exception
@ 2021-09-17 10:00 ` Dmitry Vyukov
0 siblings, 0 replies; 17+ messages in thread
From: Dmitry Vyukov @ 2021-09-17 10:00 UTC (permalink / raw)
To: syzbot
Cc: linux-fsdevel, linux-kernel, syzkaller-bugs, viro,
the arch/x86 maintainers, Linux ARM, Marco Elver, kasan-dev
On Sat, 4 Sept 2021 at 20:58, syzbot
<syzbot+d08efd12a2905a344291@syzkaller.appspotmail.com> wrote:
>
> Hello,
>
> syzbot found the following issue on:
>
> HEAD commit: 835d31d319d9 Merge tag 'media/v5.15-1' of git://git.kernel..
> git tree: upstream
> console output: https://syzkaller.appspot.com/x/log.txt?x=1189fe49300000
> kernel config: https://syzkaller.appspot.com/x/.config?x=d1a7a34dc082816f
> dashboard link: https://syzkaller.appspot.com/bug?extid=d08efd12a2905a344291
> compiler: gcc (Debian 10.2.1-6) 10.2.1 20210110, GNU ld (GNU Binutils for Debian) 2.35.1
>
> IMPORTANT: if you fix the issue, please add the following tag to the commit:
> Reported-by: syzbot+d08efd12a2905a344291@syzkaller.appspotmail.com
>
> ==================================================================
> BUG: KFENCE: use-after-free read in kvm_fastop_exception+0xf6d/0x105b
>
> Use-after-free read at 0xffff88823bc0c020 (in kfence-#5):
> kvm_fastop_exception+0xf6d/0x105b
There is probably some bug in d_lookup, but there is also something
wrong with the unwinder. It prints an unrelated kvm_fastop_exception
frame instead of __d_lookup and interestingly a very similar thing
happens on arm64 with HWASAN and a similar bug in d_lookup. The
corresponding report is:
https://syzkaller.appspot.com/bug?extid=488ddf8087564d6de6e2
BUG: KASAN: invalid-access in __entry_tramp_text_end+0xddc/0xd000
CPU: 0 PID: 22 Comm: kdevtmpfs Not tainted
5.14.0-syzkaller-11152-g78e709522d2c #0
Hardware name: linux,dummy-virt (DT)
Call trace:
dump_backtrace+0x0/0x1ac arch/arm64/kernel/stacktrace.c:76
show_stack+0x18/0x24 arch/arm64/kernel/stacktrace.c:215
__dump_stack lib/dump_stack.c:88 [inline]
dump_stack_lvl+0x68/0x84 lib/dump_stack.c:106
print_address_description+0x7c/0x2b4 mm/kasan/report.c:256
__kasan_report mm/kasan/report.c:442 [inline]
kasan_report+0x134/0x380 mm/kasan/report.c:459
__do_kernel_fault+0x128/0x1bc arch/arm64/mm/fault.c:317
do_bad_area arch/arm64/mm/fault.c:466 [inline]
do_tag_check_fault+0x74/0x90 arch/arm64/mm/fault.c:737
do_mem_abort+0x44/0xb4 arch/arm64/mm/fault.c:813
el1_abort+0x40/0x60 arch/arm64/kernel/entry-common.c:357
el1h_64_sync_handler+0xb0/0xd0 arch/arm64/kernel/entry-common.c:408
el1h_64_sync+0x78/0x7c arch/arm64/kernel/entry.S:567
__entry_tramp_text_end+0xddc/0xd000
d_lookup+0x44/0x70 fs/dcache.c:2370
lookup_dcache+0x24/0x84 fs/namei.c:1520
__lookup_hash+0x24/0xd0 fs/namei.c:1543
kern_path_locked+0x90/0x10c fs/namei.c:2567
handle_remove+0x38/0x284 drivers/base/devtmpfs.c:312
handle drivers/base/devtmpfs.c:382 [inline]
devtmpfs_work_loop drivers/base/devtmpfs.c:395 [inline]
devtmpfsd+0x8c/0xd0 drivers/base/devtmpfs.c:437
kthread+0x150/0x15c kernel/kthread.c:319
ret_from_fork+0x10/0x20 arch/arm64/kernel/entry.S:756
Here kernel unwinder prints __entry_tramp_text_end instead of __d_lookup.
I've looked in more detail into the arm64 case:
d_lookup contains a static call to __d_lookup as expected:
ffff8000102e0780 <d_lookup>:
...
ffff8000102e07c0: 97ffffa4 bl ffff8000102e0650 <__d_lookup>
...
ffff8000102e07e8: d65f03c0 ret
and these symbols don't overlap or something:
$ aarch64-linux-gnu-nm -nS vmlinux | egrep -C 1 " (t|T)
(__entry_tramp_text|__d_lookup)"
ffff8000102e01f0 0000000000000458 T d_alloc_parallel
ffff8000102e0650 0000000000000128 T __d_lookup
ffff8000102e0780 000000000000006c T d_lookup
--
ffff8000117a1f88 T __hibernate_exit_text_end
ffff8000117a2000 T __entry_tramp_text_start
ffff8000117a2000 00000000000007c8 T tramp_vectors
--
ffff8000117a27f0 0000000000000024 T tramp_exit_compat
ffff8000117a3000 T __entry_tramp_text_end
ffff8000117b0000 D _etext
So it looks like in both cases the top fault frame is just wrong. But
I would assume it's extracted by arch-dependent code, so it's
suspicious that it affects both x86 and arm64...
Any ideas what's happening?
> d_lookup+0xd8/0x170 fs/dcache.c:2370
> lookup_dcache+0x1e/0x130 fs/namei.c:1520
> __lookup_hash+0x29/0x180 fs/namei.c:1543
> kern_path_locked+0x17e/0x320 fs/namei.c:2567
> handle_remove+0xa2/0x5fe drivers/base/devtmpfs.c:312
> handle drivers/base/devtmpfs.c:382 [inline]
> devtmpfs_work_loop drivers/base/devtmpfs.c:395 [inline]
> devtmpfsd+0x1b9/0x2a3 drivers/base/devtmpfs.c:437
> kthread+0x3e5/0x4d0 kernel/kthread.c:319
> ret_from_fork+0x1f/0x30 arch/x86/entry/entry_64.S:295
>
> kfence-#5 [0xffff88823bc0c000-0xffff88823bc0cfff, size=4096, cache=names_cache] allocated by task 22:
> getname_kernel+0x4e/0x370 fs/namei.c:226
> kern_path_locked+0x71/0x320 fs/namei.c:2558
> handle_remove+0xa2/0x5fe drivers/base/devtmpfs.c:312
> handle drivers/base/devtmpfs.c:382 [inline]
> devtmpfs_work_loop drivers/base/devtmpfs.c:395 [inline]
> devtmpfsd+0x1b9/0x2a3 drivers/base/devtmpfs.c:437
> kthread+0x3e5/0x4d0 kernel/kthread.c:319
> ret_from_fork+0x1f/0x30 arch/x86/entry/entry_64.S:295
>
> freed by task 22:
> putname.part.0+0xe1/0x120 fs/namei.c:270
> putname include/linux/err.h:41 [inline]
> filename_parentat fs/namei.c:2547 [inline]
> kern_path_locked+0xc2/0x320 fs/namei.c:2558
> handle_remove+0xa2/0x5fe drivers/base/devtmpfs.c:312
> handle drivers/base/devtmpfs.c:382 [inline]
> devtmpfs_work_loop drivers/base/devtmpfs.c:395 [inline]
> devtmpfsd+0x1b9/0x2a3 drivers/base/devtmpfs.c:437
> kthread+0x3e5/0x4d0 kernel/kthread.c:319
> ret_from_fork+0x1f/0x30 arch/x86/entry/entry_64.S:295
>
> CPU: 1 PID: 22 Comm: kdevtmpfs Not tainted 5.14.0-syzkaller #0
> Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 01/01/2011
> RIP: 0010:kvm_fastop_exception+0xf6d/0x105b
> Code: d3 ed e9 14 1b 6d f8 49 8d 0e 48 83 e1 f8 4c 8b 21 41 8d 0e 83 e1 07 c1 e1 03 49 d3 ec e9 6a 28 6d f8 49 8d 4d 00 48 83 e1 f8 <4c> 8b 21 41 8d 4d 00 83 e1 07 c1 e1 03 49 d3 ec e9 5a 32 6d f8 bd
> RSP: 0018:ffffc90000fe7ae8 EFLAGS: 00010282
> RAX: 0000000035736376 RBX: ffff88803b141cc0 RCX: ffff88823bc0c020
> RDX: ffffed100762839f RSI: 0000000000000004 RDI: 0000000000000007
> RBP: 0000000000000004 R08: 0000000000000000 R09: ffff88803b141cf0
> R10: ffffed100762839e R11: 0000000000000000 R12: ffff88823bc0c020
> R13: ffff88823bc0c020 R14: ffff88803b141cf0 R15: dffffc0000000000
> FS: 0000000000000000(0000) GS:ffff8880b9d00000(0000) knlGS:0000000000000000
> CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
> CR2: ffff88823bc0c020 CR3: 0000000029892000 CR4: 00000000001506e0
> DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
> DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400
> Call Trace:
> d_lookup+0xd8/0x170 fs/dcache.c:2370
> lookup_dcache+0x1e/0x130 fs/namei.c:1520
> __lookup_hash+0x29/0x180 fs/namei.c:1543
> kern_path_locked+0x17e/0x320 fs/namei.c:2567
> handle_remove+0xa2/0x5fe drivers/base/devtmpfs.c:312
> handle drivers/base/devtmpfs.c:382 [inline]
> devtmpfs_work_loop drivers/base/devtmpfs.c:395 [inline]
> devtmpfsd+0x1b9/0x2a3 drivers/base/devtmpfs.c:437
> kthread+0x3e5/0x4d0 kernel/kthread.c:319
> ret_from_fork+0x1f/0x30 arch/x86/entry/entry_64.S:295
> ==================================================================
> ----------------
> Code disassembly (best guess):
> 0: d3 ed shr %cl,%ebp
> 2: e9 14 1b 6d f8 jmpq 0xf86d1b1b
> 7: 49 8d 0e lea (%r14),%rcx
> a: 48 83 e1 f8 and $0xfffffffffffffff8,%rcx
> e: 4c 8b 21 mov (%rcx),%r12
> 11: 41 8d 0e lea (%r14),%ecx
> 14: 83 e1 07 and $0x7,%ecx
> 17: c1 e1 03 shl $0x3,%ecx
> 1a: 49 d3 ec shr %cl,%r12
> 1d: e9 6a 28 6d f8 jmpq 0xf86d288c
> 22: 49 8d 4d 00 lea 0x0(%r13),%rcx
> 26: 48 83 e1 f8 and $0xfffffffffffffff8,%rcx
> * 2a: 4c 8b 21 mov (%rcx),%r12 <-- trapping instruction
> 2d: 41 8d 4d 00 lea 0x0(%r13),%ecx
> 31: 83 e1 07 and $0x7,%ecx
> 34: c1 e1 03 shl $0x3,%ecx
> 37: 49 d3 ec shr %cl,%r12
> 3a: e9 5a 32 6d f8 jmpq 0xf86d3299
> 3f: bd .byte 0xbd
>
>
> ---
> This report 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 issue. See:
> https://goo.gl/tpsmEJ#status for how to communicate with syzbot.
_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel
^ permalink raw reply [flat|nested] 17+ messages in thread
* Re: [syzbot] upstream test error: KFENCE: use-after-free in kvm_fastop_exception
2021-09-17 10:00 ` Dmitry Vyukov
@ 2021-09-17 11:04 ` Marco Elver
-1 siblings, 0 replies; 17+ messages in thread
From: Marco Elver @ 2021-09-17 11:04 UTC (permalink / raw)
To: Dmitry Vyukov
Cc: syzbot, linux-fsdevel, linux-kernel, syzkaller-bugs, viro,
the arch/x86 maintainers, Linux ARM, kasan-dev
On Fri, 17 Sept 2021 at 12:01, Dmitry Vyukov <dvyukov@google.com> wrote:
>
> On Sat, 4 Sept 2021 at 20:58, syzbot
> <syzbot+d08efd12a2905a344291@syzkaller.appspotmail.com> wrote:
> >
> > Hello,
> >
> > syzbot found the following issue on:
> >
> > HEAD commit: 835d31d319d9 Merge tag 'media/v5.15-1' of git://git.kernel..
> > git tree: upstream
> > console output: https://syzkaller.appspot.com/x/log.txt?x=1189fe49300000
> > kernel config: https://syzkaller.appspot.com/x/.config?x=d1a7a34dc082816f
> > dashboard link: https://syzkaller.appspot.com/bug?extid=d08efd12a2905a344291
> > compiler: gcc (Debian 10.2.1-6) 10.2.1 20210110, GNU ld (GNU Binutils for Debian) 2.35.1
> >
> > IMPORTANT: if you fix the issue, please add the following tag to the commit:
> > Reported-by: syzbot+d08efd12a2905a344291@syzkaller.appspotmail.com
> >
> > ==================================================================
> > BUG: KFENCE: use-after-free read in kvm_fastop_exception+0xf6d/0x105b
> >
> > Use-after-free read at 0xffff88823bc0c020 (in kfence-#5):
> > kvm_fastop_exception+0xf6d/0x105b
>
> There is probably some bug in d_lookup, but there is also something
> wrong with the unwinder. It prints an unrelated kvm_fastop_exception
> frame instead of __d_lookup and interestingly a very similar thing
> happens on arm64 with HWASAN and a similar bug in d_lookup. The
> corresponding report is:
> https://syzkaller.appspot.com/bug?extid=488ddf8087564d6de6e2
>
> BUG: KASAN: invalid-access in __entry_tramp_text_end+0xddc/0xd000
> CPU: 0 PID: 22 Comm: kdevtmpfs Not tainted
> 5.14.0-syzkaller-11152-g78e709522d2c #0
> Hardware name: linux,dummy-virt (DT)
> Call trace:
> dump_backtrace+0x0/0x1ac arch/arm64/kernel/stacktrace.c:76
> show_stack+0x18/0x24 arch/arm64/kernel/stacktrace.c:215
> __dump_stack lib/dump_stack.c:88 [inline]
> dump_stack_lvl+0x68/0x84 lib/dump_stack.c:106
> print_address_description+0x7c/0x2b4 mm/kasan/report.c:256
> __kasan_report mm/kasan/report.c:442 [inline]
> kasan_report+0x134/0x380 mm/kasan/report.c:459
> __do_kernel_fault+0x128/0x1bc arch/arm64/mm/fault.c:317
> do_bad_area arch/arm64/mm/fault.c:466 [inline]
> do_tag_check_fault+0x74/0x90 arch/arm64/mm/fault.c:737
> do_mem_abort+0x44/0xb4 arch/arm64/mm/fault.c:813
> el1_abort+0x40/0x60 arch/arm64/kernel/entry-common.c:357
> el1h_64_sync_handler+0xb0/0xd0 arch/arm64/kernel/entry-common.c:408
> el1h_64_sync+0x78/0x7c arch/arm64/kernel/entry.S:567
> __entry_tramp_text_end+0xddc/0xd000
> d_lookup+0x44/0x70 fs/dcache.c:2370
> lookup_dcache+0x24/0x84 fs/namei.c:1520
> __lookup_hash+0x24/0xd0 fs/namei.c:1543
> kern_path_locked+0x90/0x10c fs/namei.c:2567
> handle_remove+0x38/0x284 drivers/base/devtmpfs.c:312
> handle drivers/base/devtmpfs.c:382 [inline]
> devtmpfs_work_loop drivers/base/devtmpfs.c:395 [inline]
> devtmpfsd+0x8c/0xd0 drivers/base/devtmpfs.c:437
> kthread+0x150/0x15c kernel/kthread.c:319
> ret_from_fork+0x10/0x20 arch/arm64/kernel/entry.S:756
>
> Here kernel unwinder prints __entry_tramp_text_end instead of __d_lookup.
>
> I've looked in more detail into the arm64 case:
> d_lookup contains a static call to __d_lookup as expected:
>
> ffff8000102e0780 <d_lookup>:
> ...
> ffff8000102e07c0: 97ffffa4 bl ffff8000102e0650 <__d_lookup>
> ...
> ffff8000102e07e8: d65f03c0 ret
>
> and these symbols don't overlap or something:
>
> $ aarch64-linux-gnu-nm -nS vmlinux | egrep -C 1 " (t|T)
> (__entry_tramp_text|__d_lookup)"
> ffff8000102e01f0 0000000000000458 T d_alloc_parallel
> ffff8000102e0650 0000000000000128 T __d_lookup
> ffff8000102e0780 000000000000006c T d_lookup
> --
> ffff8000117a1f88 T __hibernate_exit_text_end
> ffff8000117a2000 T __entry_tramp_text_start
> ffff8000117a2000 00000000000007c8 T tramp_vectors
> --
> ffff8000117a27f0 0000000000000024 T tramp_exit_compat
> ffff8000117a3000 T __entry_tramp_text_end
> ffff8000117b0000 D _etext
>
> So it looks like in both cases the top fault frame is just wrong. But
> I would assume it's extracted by arch-dependent code, so it's
> suspicious that it affects both x86 and arm64...
>
> Any ideas what's happening?
My suspicion for the x86 case is that kvm_fastop_exception is related
to instruction emulation and the fault occurs in an emulated
instruction?
But I can't explain the arm64 case.
> > d_lookup+0xd8/0x170 fs/dcache.c:2370
> > lookup_dcache+0x1e/0x130 fs/namei.c:1520
> > __lookup_hash+0x29/0x180 fs/namei.c:1543
> > kern_path_locked+0x17e/0x320 fs/namei.c:2567
> > handle_remove+0xa2/0x5fe drivers/base/devtmpfs.c:312
> > handle drivers/base/devtmpfs.c:382 [inline]
> > devtmpfs_work_loop drivers/base/devtmpfs.c:395 [inline]
> > devtmpfsd+0x1b9/0x2a3 drivers/base/devtmpfs.c:437
> > kthread+0x3e5/0x4d0 kernel/kthread.c:319
> > ret_from_fork+0x1f/0x30 arch/x86/entry/entry_64.S:295
> >
> > kfence-#5 [0xffff88823bc0c000-0xffff88823bc0cfff, size=4096, cache=names_cache] allocated by task 22:
> > getname_kernel+0x4e/0x370 fs/namei.c:226
> > kern_path_locked+0x71/0x320 fs/namei.c:2558
> > handle_remove+0xa2/0x5fe drivers/base/devtmpfs.c:312
> > handle drivers/base/devtmpfs.c:382 [inline]
> > devtmpfs_work_loop drivers/base/devtmpfs.c:395 [inline]
> > devtmpfsd+0x1b9/0x2a3 drivers/base/devtmpfs.c:437
> > kthread+0x3e5/0x4d0 kernel/kthread.c:319
> > ret_from_fork+0x1f/0x30 arch/x86/entry/entry_64.S:295
> >
> > freed by task 22:
> > putname.part.0+0xe1/0x120 fs/namei.c:270
> > putname include/linux/err.h:41 [inline]
> > filename_parentat fs/namei.c:2547 [inline]
> > kern_path_locked+0xc2/0x320 fs/namei.c:2558
> > handle_remove+0xa2/0x5fe drivers/base/devtmpfs.c:312
> > handle drivers/base/devtmpfs.c:382 [inline]
> > devtmpfs_work_loop drivers/base/devtmpfs.c:395 [inline]
> > devtmpfsd+0x1b9/0x2a3 drivers/base/devtmpfs.c:437
> > kthread+0x3e5/0x4d0 kernel/kthread.c:319
> > ret_from_fork+0x1f/0x30 arch/x86/entry/entry_64.S:295
> >
> > CPU: 1 PID: 22 Comm: kdevtmpfs Not tainted 5.14.0-syzkaller #0
> > Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 01/01/2011
> > RIP: 0010:kvm_fastop_exception+0xf6d/0x105b
> > Code: d3 ed e9 14 1b 6d f8 49 8d 0e 48 83 e1 f8 4c 8b 21 41 8d 0e 83 e1 07 c1 e1 03 49 d3 ec e9 6a 28 6d f8 49 8d 4d 00 48 83 e1 f8 <4c> 8b 21 41 8d 4d 00 83 e1 07 c1 e1 03 49 d3 ec e9 5a 32 6d f8 bd
> > RSP: 0018:ffffc90000fe7ae8 EFLAGS: 00010282
> > RAX: 0000000035736376 RBX: ffff88803b141cc0 RCX: ffff88823bc0c020
> > RDX: ffffed100762839f RSI: 0000000000000004 RDI: 0000000000000007
> > RBP: 0000000000000004 R08: 0000000000000000 R09: ffff88803b141cf0
> > R10: ffffed100762839e R11: 0000000000000000 R12: ffff88823bc0c020
> > R13: ffff88823bc0c020 R14: ffff88803b141cf0 R15: dffffc0000000000
> > FS: 0000000000000000(0000) GS:ffff8880b9d00000(0000) knlGS:0000000000000000
> > CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
> > CR2: ffff88823bc0c020 CR3: 0000000029892000 CR4: 00000000001506e0
> > DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
> > DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400
> > Call Trace:
> > d_lookup+0xd8/0x170 fs/dcache.c:2370
> > lookup_dcache+0x1e/0x130 fs/namei.c:1520
> > __lookup_hash+0x29/0x180 fs/namei.c:1543
> > kern_path_locked+0x17e/0x320 fs/namei.c:2567
> > handle_remove+0xa2/0x5fe drivers/base/devtmpfs.c:312
> > handle drivers/base/devtmpfs.c:382 [inline]
> > devtmpfs_work_loop drivers/base/devtmpfs.c:395 [inline]
> > devtmpfsd+0x1b9/0x2a3 drivers/base/devtmpfs.c:437
> > kthread+0x3e5/0x4d0 kernel/kthread.c:319
> > ret_from_fork+0x1f/0x30 arch/x86/entry/entry_64.S:295
> > ==================================================================
> > ----------------
> > Code disassembly (best guess):
> > 0: d3 ed shr %cl,%ebp
> > 2: e9 14 1b 6d f8 jmpq 0xf86d1b1b
> > 7: 49 8d 0e lea (%r14),%rcx
> > a: 48 83 e1 f8 and $0xfffffffffffffff8,%rcx
> > e: 4c 8b 21 mov (%rcx),%r12
> > 11: 41 8d 0e lea (%r14),%ecx
> > 14: 83 e1 07 and $0x7,%ecx
> > 17: c1 e1 03 shl $0x3,%ecx
> > 1a: 49 d3 ec shr %cl,%r12
> > 1d: e9 6a 28 6d f8 jmpq 0xf86d288c
> > 22: 49 8d 4d 00 lea 0x0(%r13),%rcx
> > 26: 48 83 e1 f8 and $0xfffffffffffffff8,%rcx
> > * 2a: 4c 8b 21 mov (%rcx),%r12 <-- trapping instruction
> > 2d: 41 8d 4d 00 lea 0x0(%r13),%ecx
> > 31: 83 e1 07 and $0x7,%ecx
> > 34: c1 e1 03 shl $0x3,%ecx
> > 37: 49 d3 ec shr %cl,%r12
> > 3a: e9 5a 32 6d f8 jmpq 0xf86d3299
> > 3f: bd .byte 0xbd
> >
> >
> > ---
> > This report 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 issue. See:
> > https://goo.gl/tpsmEJ#status for how to communicate with syzbot.
^ permalink raw reply [flat|nested] 17+ messages in thread
* Re: [syzbot] upstream test error: KFENCE: use-after-free in kvm_fastop_exception
@ 2021-09-17 11:04 ` Marco Elver
0 siblings, 0 replies; 17+ messages in thread
From: Marco Elver @ 2021-09-17 11:04 UTC (permalink / raw)
To: Dmitry Vyukov
Cc: syzbot, linux-fsdevel, linux-kernel, syzkaller-bugs, viro,
the arch/x86 maintainers, Linux ARM, kasan-dev
On Fri, 17 Sept 2021 at 12:01, Dmitry Vyukov <dvyukov@google.com> wrote:
>
> On Sat, 4 Sept 2021 at 20:58, syzbot
> <syzbot+d08efd12a2905a344291@syzkaller.appspotmail.com> wrote:
> >
> > Hello,
> >
> > syzbot found the following issue on:
> >
> > HEAD commit: 835d31d319d9 Merge tag 'media/v5.15-1' of git://git.kernel..
> > git tree: upstream
> > console output: https://syzkaller.appspot.com/x/log.txt?x=1189fe49300000
> > kernel config: https://syzkaller.appspot.com/x/.config?x=d1a7a34dc082816f
> > dashboard link: https://syzkaller.appspot.com/bug?extid=d08efd12a2905a344291
> > compiler: gcc (Debian 10.2.1-6) 10.2.1 20210110, GNU ld (GNU Binutils for Debian) 2.35.1
> >
> > IMPORTANT: if you fix the issue, please add the following tag to the commit:
> > Reported-by: syzbot+d08efd12a2905a344291@syzkaller.appspotmail.com
> >
> > ==================================================================
> > BUG: KFENCE: use-after-free read in kvm_fastop_exception+0xf6d/0x105b
> >
> > Use-after-free read at 0xffff88823bc0c020 (in kfence-#5):
> > kvm_fastop_exception+0xf6d/0x105b
>
> There is probably some bug in d_lookup, but there is also something
> wrong with the unwinder. It prints an unrelated kvm_fastop_exception
> frame instead of __d_lookup and interestingly a very similar thing
> happens on arm64 with HWASAN and a similar bug in d_lookup. The
> corresponding report is:
> https://syzkaller.appspot.com/bug?extid=488ddf8087564d6de6e2
>
> BUG: KASAN: invalid-access in __entry_tramp_text_end+0xddc/0xd000
> CPU: 0 PID: 22 Comm: kdevtmpfs Not tainted
> 5.14.0-syzkaller-11152-g78e709522d2c #0
> Hardware name: linux,dummy-virt (DT)
> Call trace:
> dump_backtrace+0x0/0x1ac arch/arm64/kernel/stacktrace.c:76
> show_stack+0x18/0x24 arch/arm64/kernel/stacktrace.c:215
> __dump_stack lib/dump_stack.c:88 [inline]
> dump_stack_lvl+0x68/0x84 lib/dump_stack.c:106
> print_address_description+0x7c/0x2b4 mm/kasan/report.c:256
> __kasan_report mm/kasan/report.c:442 [inline]
> kasan_report+0x134/0x380 mm/kasan/report.c:459
> __do_kernel_fault+0x128/0x1bc arch/arm64/mm/fault.c:317
> do_bad_area arch/arm64/mm/fault.c:466 [inline]
> do_tag_check_fault+0x74/0x90 arch/arm64/mm/fault.c:737
> do_mem_abort+0x44/0xb4 arch/arm64/mm/fault.c:813
> el1_abort+0x40/0x60 arch/arm64/kernel/entry-common.c:357
> el1h_64_sync_handler+0xb0/0xd0 arch/arm64/kernel/entry-common.c:408
> el1h_64_sync+0x78/0x7c arch/arm64/kernel/entry.S:567
> __entry_tramp_text_end+0xddc/0xd000
> d_lookup+0x44/0x70 fs/dcache.c:2370
> lookup_dcache+0x24/0x84 fs/namei.c:1520
> __lookup_hash+0x24/0xd0 fs/namei.c:1543
> kern_path_locked+0x90/0x10c fs/namei.c:2567
> handle_remove+0x38/0x284 drivers/base/devtmpfs.c:312
> handle drivers/base/devtmpfs.c:382 [inline]
> devtmpfs_work_loop drivers/base/devtmpfs.c:395 [inline]
> devtmpfsd+0x8c/0xd0 drivers/base/devtmpfs.c:437
> kthread+0x150/0x15c kernel/kthread.c:319
> ret_from_fork+0x10/0x20 arch/arm64/kernel/entry.S:756
>
> Here kernel unwinder prints __entry_tramp_text_end instead of __d_lookup.
>
> I've looked in more detail into the arm64 case:
> d_lookup contains a static call to __d_lookup as expected:
>
> ffff8000102e0780 <d_lookup>:
> ...
> ffff8000102e07c0: 97ffffa4 bl ffff8000102e0650 <__d_lookup>
> ...
> ffff8000102e07e8: d65f03c0 ret
>
> and these symbols don't overlap or something:
>
> $ aarch64-linux-gnu-nm -nS vmlinux | egrep -C 1 " (t|T)
> (__entry_tramp_text|__d_lookup)"
> ffff8000102e01f0 0000000000000458 T d_alloc_parallel
> ffff8000102e0650 0000000000000128 T __d_lookup
> ffff8000102e0780 000000000000006c T d_lookup
> --
> ffff8000117a1f88 T __hibernate_exit_text_end
> ffff8000117a2000 T __entry_tramp_text_start
> ffff8000117a2000 00000000000007c8 T tramp_vectors
> --
> ffff8000117a27f0 0000000000000024 T tramp_exit_compat
> ffff8000117a3000 T __entry_tramp_text_end
> ffff8000117b0000 D _etext
>
> So it looks like in both cases the top fault frame is just wrong. But
> I would assume it's extracted by arch-dependent code, so it's
> suspicious that it affects both x86 and arm64...
>
> Any ideas what's happening?
My suspicion for the x86 case is that kvm_fastop_exception is related
to instruction emulation and the fault occurs in an emulated
instruction?
But I can't explain the arm64 case.
> > d_lookup+0xd8/0x170 fs/dcache.c:2370
> > lookup_dcache+0x1e/0x130 fs/namei.c:1520
> > __lookup_hash+0x29/0x180 fs/namei.c:1543
> > kern_path_locked+0x17e/0x320 fs/namei.c:2567
> > handle_remove+0xa2/0x5fe drivers/base/devtmpfs.c:312
> > handle drivers/base/devtmpfs.c:382 [inline]
> > devtmpfs_work_loop drivers/base/devtmpfs.c:395 [inline]
> > devtmpfsd+0x1b9/0x2a3 drivers/base/devtmpfs.c:437
> > kthread+0x3e5/0x4d0 kernel/kthread.c:319
> > ret_from_fork+0x1f/0x30 arch/x86/entry/entry_64.S:295
> >
> > kfence-#5 [0xffff88823bc0c000-0xffff88823bc0cfff, size=4096, cache=names_cache] allocated by task 22:
> > getname_kernel+0x4e/0x370 fs/namei.c:226
> > kern_path_locked+0x71/0x320 fs/namei.c:2558
> > handle_remove+0xa2/0x5fe drivers/base/devtmpfs.c:312
> > handle drivers/base/devtmpfs.c:382 [inline]
> > devtmpfs_work_loop drivers/base/devtmpfs.c:395 [inline]
> > devtmpfsd+0x1b9/0x2a3 drivers/base/devtmpfs.c:437
> > kthread+0x3e5/0x4d0 kernel/kthread.c:319
> > ret_from_fork+0x1f/0x30 arch/x86/entry/entry_64.S:295
> >
> > freed by task 22:
> > putname.part.0+0xe1/0x120 fs/namei.c:270
> > putname include/linux/err.h:41 [inline]
> > filename_parentat fs/namei.c:2547 [inline]
> > kern_path_locked+0xc2/0x320 fs/namei.c:2558
> > handle_remove+0xa2/0x5fe drivers/base/devtmpfs.c:312
> > handle drivers/base/devtmpfs.c:382 [inline]
> > devtmpfs_work_loop drivers/base/devtmpfs.c:395 [inline]
> > devtmpfsd+0x1b9/0x2a3 drivers/base/devtmpfs.c:437
> > kthread+0x3e5/0x4d0 kernel/kthread.c:319
> > ret_from_fork+0x1f/0x30 arch/x86/entry/entry_64.S:295
> >
> > CPU: 1 PID: 22 Comm: kdevtmpfs Not tainted 5.14.0-syzkaller #0
> > Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 01/01/2011
> > RIP: 0010:kvm_fastop_exception+0xf6d/0x105b
> > Code: d3 ed e9 14 1b 6d f8 49 8d 0e 48 83 e1 f8 4c 8b 21 41 8d 0e 83 e1 07 c1 e1 03 49 d3 ec e9 6a 28 6d f8 49 8d 4d 00 48 83 e1 f8 <4c> 8b 21 41 8d 4d 00 83 e1 07 c1 e1 03 49 d3 ec e9 5a 32 6d f8 bd
> > RSP: 0018:ffffc90000fe7ae8 EFLAGS: 00010282
> > RAX: 0000000035736376 RBX: ffff88803b141cc0 RCX: ffff88823bc0c020
> > RDX: ffffed100762839f RSI: 0000000000000004 RDI: 0000000000000007
> > RBP: 0000000000000004 R08: 0000000000000000 R09: ffff88803b141cf0
> > R10: ffffed100762839e R11: 0000000000000000 R12: ffff88823bc0c020
> > R13: ffff88823bc0c020 R14: ffff88803b141cf0 R15: dffffc0000000000
> > FS: 0000000000000000(0000) GS:ffff8880b9d00000(0000) knlGS:0000000000000000
> > CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
> > CR2: ffff88823bc0c020 CR3: 0000000029892000 CR4: 00000000001506e0
> > DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
> > DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400
> > Call Trace:
> > d_lookup+0xd8/0x170 fs/dcache.c:2370
> > lookup_dcache+0x1e/0x130 fs/namei.c:1520
> > __lookup_hash+0x29/0x180 fs/namei.c:1543
> > kern_path_locked+0x17e/0x320 fs/namei.c:2567
> > handle_remove+0xa2/0x5fe drivers/base/devtmpfs.c:312
> > handle drivers/base/devtmpfs.c:382 [inline]
> > devtmpfs_work_loop drivers/base/devtmpfs.c:395 [inline]
> > devtmpfsd+0x1b9/0x2a3 drivers/base/devtmpfs.c:437
> > kthread+0x3e5/0x4d0 kernel/kthread.c:319
> > ret_from_fork+0x1f/0x30 arch/x86/entry/entry_64.S:295
> > ==================================================================
> > ----------------
> > Code disassembly (best guess):
> > 0: d3 ed shr %cl,%ebp
> > 2: e9 14 1b 6d f8 jmpq 0xf86d1b1b
> > 7: 49 8d 0e lea (%r14),%rcx
> > a: 48 83 e1 f8 and $0xfffffffffffffff8,%rcx
> > e: 4c 8b 21 mov (%rcx),%r12
> > 11: 41 8d 0e lea (%r14),%ecx
> > 14: 83 e1 07 and $0x7,%ecx
> > 17: c1 e1 03 shl $0x3,%ecx
> > 1a: 49 d3 ec shr %cl,%r12
> > 1d: e9 6a 28 6d f8 jmpq 0xf86d288c
> > 22: 49 8d 4d 00 lea 0x0(%r13),%rcx
> > 26: 48 83 e1 f8 and $0xfffffffffffffff8,%rcx
> > * 2a: 4c 8b 21 mov (%rcx),%r12 <-- trapping instruction
> > 2d: 41 8d 4d 00 lea 0x0(%r13),%ecx
> > 31: 83 e1 07 and $0x7,%ecx
> > 34: c1 e1 03 shl $0x3,%ecx
> > 37: 49 d3 ec shr %cl,%r12
> > 3a: e9 5a 32 6d f8 jmpq 0xf86d3299
> > 3f: bd .byte 0xbd
> >
> >
> > ---
> > This report 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 issue. See:
> > https://goo.gl/tpsmEJ#status for how to communicate with syzbot.
_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel
^ permalink raw reply [flat|nested] 17+ messages in thread
* Re: [syzbot] upstream test error: KFENCE: use-after-free in kvm_fastop_exception
2021-09-17 11:04 ` Marco Elver
@ 2021-09-17 12:43 ` Dmitry Vyukov
-1 siblings, 0 replies; 17+ messages in thread
From: Dmitry Vyukov @ 2021-09-17 12:43 UTC (permalink / raw)
To: Marco Elver
Cc: syzbot, linux-fsdevel, linux-kernel, syzkaller-bugs, viro,
the arch/x86 maintainers, Linux ARM, kasan-dev
On Fri, 17 Sept 2021 at 13:04, Marco Elver <elver@google.com> wrote:
> > On Sat, 4 Sept 2021 at 20:58, syzbot
> > <syzbot+d08efd12a2905a344291@syzkaller.appspotmail.com> wrote:
> > >
> > > Hello,
> > >
> > > syzbot found the following issue on:
> > >
> > > HEAD commit: 835d31d319d9 Merge tag 'media/v5.15-1' of git://git.kernel..
> > > git tree: upstream
> > > console output: https://syzkaller.appspot.com/x/log.txt?x=1189fe49300000
> > > kernel config: https://syzkaller.appspot.com/x/.config?x=d1a7a34dc082816f
> > > dashboard link: https://syzkaller.appspot.com/bug?extid=d08efd12a2905a344291
> > > compiler: gcc (Debian 10.2.1-6) 10.2.1 20210110, GNU ld (GNU Binutils for Debian) 2.35.1
> > >
> > > IMPORTANT: if you fix the issue, please add the following tag to the commit:
> > > Reported-by: syzbot+d08efd12a2905a344291@syzkaller.appspotmail.com
> > >
> > > ==================================================================
> > > BUG: KFENCE: use-after-free read in kvm_fastop_exception+0xf6d/0x105b
> > >
> > > Use-after-free read at 0xffff88823bc0c020 (in kfence-#5):
> > > kvm_fastop_exception+0xf6d/0x105b
> >
> > There is probably some bug in d_lookup, but there is also something
> > wrong with the unwinder. It prints an unrelated kvm_fastop_exception
> > frame instead of __d_lookup and interestingly a very similar thing
> > happens on arm64 with HWASAN and a similar bug in d_lookup. The
> > corresponding report is:
> > https://syzkaller.appspot.com/bug?extid=488ddf8087564d6de6e2
> >
> > BUG: KASAN: invalid-access in __entry_tramp_text_end+0xddc/0xd000
> > CPU: 0 PID: 22 Comm: kdevtmpfs Not tainted
> > 5.14.0-syzkaller-11152-g78e709522d2c #0
> > Hardware name: linux,dummy-virt (DT)
> > Call trace:
> > dump_backtrace+0x0/0x1ac arch/arm64/kernel/stacktrace.c:76
> > show_stack+0x18/0x24 arch/arm64/kernel/stacktrace.c:215
> > __dump_stack lib/dump_stack.c:88 [inline]
> > dump_stack_lvl+0x68/0x84 lib/dump_stack.c:106
> > print_address_description+0x7c/0x2b4 mm/kasan/report.c:256
> > __kasan_report mm/kasan/report.c:442 [inline]
> > kasan_report+0x134/0x380 mm/kasan/report.c:459
> > __do_kernel_fault+0x128/0x1bc arch/arm64/mm/fault.c:317
> > do_bad_area arch/arm64/mm/fault.c:466 [inline]
> > do_tag_check_fault+0x74/0x90 arch/arm64/mm/fault.c:737
> > do_mem_abort+0x44/0xb4 arch/arm64/mm/fault.c:813
> > el1_abort+0x40/0x60 arch/arm64/kernel/entry-common.c:357
> > el1h_64_sync_handler+0xb0/0xd0 arch/arm64/kernel/entry-common.c:408
> > el1h_64_sync+0x78/0x7c arch/arm64/kernel/entry.S:567
> > __entry_tramp_text_end+0xddc/0xd000
> > d_lookup+0x44/0x70 fs/dcache.c:2370
> > lookup_dcache+0x24/0x84 fs/namei.c:1520
> > __lookup_hash+0x24/0xd0 fs/namei.c:1543
> > kern_path_locked+0x90/0x10c fs/namei.c:2567
> > handle_remove+0x38/0x284 drivers/base/devtmpfs.c:312
> > handle drivers/base/devtmpfs.c:382 [inline]
> > devtmpfs_work_loop drivers/base/devtmpfs.c:395 [inline]
> > devtmpfsd+0x8c/0xd0 drivers/base/devtmpfs.c:437
> > kthread+0x150/0x15c kernel/kthread.c:319
> > ret_from_fork+0x10/0x20 arch/arm64/kernel/entry.S:756
> >
> > Here kernel unwinder prints __entry_tramp_text_end instead of __d_lookup.
> >
> > I've looked in more detail into the arm64 case:
> > d_lookup contains a static call to __d_lookup as expected:
> >
> > ffff8000102e0780 <d_lookup>:
> > ...
> > ffff8000102e07c0: 97ffffa4 bl ffff8000102e0650 <__d_lookup>
> > ...
> > ffff8000102e07e8: d65f03c0 ret
> >
> > and these symbols don't overlap or something:
> >
> > $ aarch64-linux-gnu-nm -nS vmlinux | egrep -C 1 " (t|T)
> > (__entry_tramp_text|__d_lookup)"
> > ffff8000102e01f0 0000000000000458 T d_alloc_parallel
> > ffff8000102e0650 0000000000000128 T __d_lookup
> > ffff8000102e0780 000000000000006c T d_lookup
> > --
> > ffff8000117a1f88 T __hibernate_exit_text_end
> > ffff8000117a2000 T __entry_tramp_text_start
> > ffff8000117a2000 00000000000007c8 T tramp_vectors
> > --
> > ffff8000117a27f0 0000000000000024 T tramp_exit_compat
> > ffff8000117a3000 T __entry_tramp_text_end
> > ffff8000117b0000 D _etext
> >
> > So it looks like in both cases the top fault frame is just wrong. But
> > I would assume it's extracted by arch-dependent code, so it's
> > suspicious that it affects both x86 and arm64...
> >
> > Any ideas what's happening?
>
> My suspicion for the x86 case is that kvm_fastop_exception is related
> to instruction emulation and the fault occurs in an emulated
> instruction?
Why would the kernel emulate a plain MOV?
2a: 4c 8b 21 mov (%rcx),%r12
And it would also mean a broken unwind because the emulated
instruction is in __d_lookup, so it should be in the stack trace.
> But I can't explain the arm64 case.
>
> > > d_lookup+0xd8/0x170 fs/dcache.c:2370
> > > lookup_dcache+0x1e/0x130 fs/namei.c:1520
> > > __lookup_hash+0x29/0x180 fs/namei.c:1543
> > > kern_path_locked+0x17e/0x320 fs/namei.c:2567
> > > handle_remove+0xa2/0x5fe drivers/base/devtmpfs.c:312
> > > handle drivers/base/devtmpfs.c:382 [inline]
> > > devtmpfs_work_loop drivers/base/devtmpfs.c:395 [inline]
> > > devtmpfsd+0x1b9/0x2a3 drivers/base/devtmpfs.c:437
> > > kthread+0x3e5/0x4d0 kernel/kthread.c:319
> > > ret_from_fork+0x1f/0x30 arch/x86/entry/entry_64.S:295
> > >
> > > kfence-#5 [0xffff88823bc0c000-0xffff88823bc0cfff, size=4096, cache=names_cache] allocated by task 22:
> > > getname_kernel+0x4e/0x370 fs/namei.c:226
> > > kern_path_locked+0x71/0x320 fs/namei.c:2558
> > > handle_remove+0xa2/0x5fe drivers/base/devtmpfs.c:312
> > > handle drivers/base/devtmpfs.c:382 [inline]
> > > devtmpfs_work_loop drivers/base/devtmpfs.c:395 [inline]
> > > devtmpfsd+0x1b9/0x2a3 drivers/base/devtmpfs.c:437
> > > kthread+0x3e5/0x4d0 kernel/kthread.c:319
> > > ret_from_fork+0x1f/0x30 arch/x86/entry/entry_64.S:295
> > >
> > > freed by task 22:
> > > putname.part.0+0xe1/0x120 fs/namei.c:270
> > > putname include/linux/err.h:41 [inline]
> > > filename_parentat fs/namei.c:2547 [inline]
> > > kern_path_locked+0xc2/0x320 fs/namei.c:2558
> > > handle_remove+0xa2/0x5fe drivers/base/devtmpfs.c:312
> > > handle drivers/base/devtmpfs.c:382 [inline]
> > > devtmpfs_work_loop drivers/base/devtmpfs.c:395 [inline]
> > > devtmpfsd+0x1b9/0x2a3 drivers/base/devtmpfs.c:437
> > > kthread+0x3e5/0x4d0 kernel/kthread.c:319
> > > ret_from_fork+0x1f/0x30 arch/x86/entry/entry_64.S:295
> > >
> > > CPU: 1 PID: 22 Comm: kdevtmpfs Not tainted 5.14.0-syzkaller #0
> > > Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 01/01/2011
> > > RIP: 0010:kvm_fastop_exception+0xf6d/0x105b
> > > Code: d3 ed e9 14 1b 6d f8 49 8d 0e 48 83 e1 f8 4c 8b 21 41 8d 0e 83 e1 07 c1 e1 03 49 d3 ec e9 6a 28 6d f8 49 8d 4d 00 48 83 e1 f8 <4c> 8b 21 41 8d 4d 00 83 e1 07 c1 e1 03 49 d3 ec e9 5a 32 6d f8 bd
> > > RSP: 0018:ffffc90000fe7ae8 EFLAGS: 00010282
> > > RAX: 0000000035736376 RBX: ffff88803b141cc0 RCX: ffff88823bc0c020
> > > RDX: ffffed100762839f RSI: 0000000000000004 RDI: 0000000000000007
> > > RBP: 0000000000000004 R08: 0000000000000000 R09: ffff88803b141cf0
> > > R10: ffffed100762839e R11: 0000000000000000 R12: ffff88823bc0c020
> > > R13: ffff88823bc0c020 R14: ffff88803b141cf0 R15: dffffc0000000000
> > > FS: 0000000000000000(0000) GS:ffff8880b9d00000(0000) knlGS:0000000000000000
> > > CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
> > > CR2: ffff88823bc0c020 CR3: 0000000029892000 CR4: 00000000001506e0
> > > DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
> > > DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400
> > > Call Trace:
> > > d_lookup+0xd8/0x170 fs/dcache.c:2370
> > > lookup_dcache+0x1e/0x130 fs/namei.c:1520
> > > __lookup_hash+0x29/0x180 fs/namei.c:1543
> > > kern_path_locked+0x17e/0x320 fs/namei.c:2567
> > > handle_remove+0xa2/0x5fe drivers/base/devtmpfs.c:312
> > > handle drivers/base/devtmpfs.c:382 [inline]
> > > devtmpfs_work_loop drivers/base/devtmpfs.c:395 [inline]
> > > devtmpfsd+0x1b9/0x2a3 drivers/base/devtmpfs.c:437
> > > kthread+0x3e5/0x4d0 kernel/kthread.c:319
> > > ret_from_fork+0x1f/0x30 arch/x86/entry/entry_64.S:295
> > > ==================================================================
> > > ----------------
> > > Code disassembly (best guess):
> > > 0: d3 ed shr %cl,%ebp
> > > 2: e9 14 1b 6d f8 jmpq 0xf86d1b1b
> > > 7: 49 8d 0e lea (%r14),%rcx
> > > a: 48 83 e1 f8 and $0xfffffffffffffff8,%rcx
> > > e: 4c 8b 21 mov (%rcx),%r12
> > > 11: 41 8d 0e lea (%r14),%ecx
> > > 14: 83 e1 07 and $0x7,%ecx
> > > 17: c1 e1 03 shl $0x3,%ecx
> > > 1a: 49 d3 ec shr %cl,%r12
> > > 1d: e9 6a 28 6d f8 jmpq 0xf86d288c
> > > 22: 49 8d 4d 00 lea 0x0(%r13),%rcx
> > > 26: 48 83 e1 f8 and $0xfffffffffffffff8,%rcx
> > > * 2a: 4c 8b 21 mov (%rcx),%r12 <-- trapping instruction
> > > 2d: 41 8d 4d 00 lea 0x0(%r13),%ecx
> > > 31: 83 e1 07 and $0x7,%ecx
> > > 34: c1 e1 03 shl $0x3,%ecx
> > > 37: 49 d3 ec shr %cl,%r12
> > > 3a: e9 5a 32 6d f8 jmpq 0xf86d3299
> > > 3f: bd .byte 0xbd
> > >
> > >
> > > ---
> > > This report 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 issue. See:
> > > https://goo.gl/tpsmEJ#status for how to communicate with syzbot.
^ permalink raw reply [flat|nested] 17+ messages in thread
* Re: [syzbot] upstream test error: KFENCE: use-after-free in kvm_fastop_exception
@ 2021-09-17 12:43 ` Dmitry Vyukov
0 siblings, 0 replies; 17+ messages in thread
From: Dmitry Vyukov @ 2021-09-17 12:43 UTC (permalink / raw)
To: Marco Elver
Cc: syzbot, linux-fsdevel, linux-kernel, syzkaller-bugs, viro,
the arch/x86 maintainers, Linux ARM, kasan-dev
On Fri, 17 Sept 2021 at 13:04, Marco Elver <elver@google.com> wrote:
> > On Sat, 4 Sept 2021 at 20:58, syzbot
> > <syzbot+d08efd12a2905a344291@syzkaller.appspotmail.com> wrote:
> > >
> > > Hello,
> > >
> > > syzbot found the following issue on:
> > >
> > > HEAD commit: 835d31d319d9 Merge tag 'media/v5.15-1' of git://git.kernel..
> > > git tree: upstream
> > > console output: https://syzkaller.appspot.com/x/log.txt?x=1189fe49300000
> > > kernel config: https://syzkaller.appspot.com/x/.config?x=d1a7a34dc082816f
> > > dashboard link: https://syzkaller.appspot.com/bug?extid=d08efd12a2905a344291
> > > compiler: gcc (Debian 10.2.1-6) 10.2.1 20210110, GNU ld (GNU Binutils for Debian) 2.35.1
> > >
> > > IMPORTANT: if you fix the issue, please add the following tag to the commit:
> > > Reported-by: syzbot+d08efd12a2905a344291@syzkaller.appspotmail.com
> > >
> > > ==================================================================
> > > BUG: KFENCE: use-after-free read in kvm_fastop_exception+0xf6d/0x105b
> > >
> > > Use-after-free read at 0xffff88823bc0c020 (in kfence-#5):
> > > kvm_fastop_exception+0xf6d/0x105b
> >
> > There is probably some bug in d_lookup, but there is also something
> > wrong with the unwinder. It prints an unrelated kvm_fastop_exception
> > frame instead of __d_lookup and interestingly a very similar thing
> > happens on arm64 with HWASAN and a similar bug in d_lookup. The
> > corresponding report is:
> > https://syzkaller.appspot.com/bug?extid=488ddf8087564d6de6e2
> >
> > BUG: KASAN: invalid-access in __entry_tramp_text_end+0xddc/0xd000
> > CPU: 0 PID: 22 Comm: kdevtmpfs Not tainted
> > 5.14.0-syzkaller-11152-g78e709522d2c #0
> > Hardware name: linux,dummy-virt (DT)
> > Call trace:
> > dump_backtrace+0x0/0x1ac arch/arm64/kernel/stacktrace.c:76
> > show_stack+0x18/0x24 arch/arm64/kernel/stacktrace.c:215
> > __dump_stack lib/dump_stack.c:88 [inline]
> > dump_stack_lvl+0x68/0x84 lib/dump_stack.c:106
> > print_address_description+0x7c/0x2b4 mm/kasan/report.c:256
> > __kasan_report mm/kasan/report.c:442 [inline]
> > kasan_report+0x134/0x380 mm/kasan/report.c:459
> > __do_kernel_fault+0x128/0x1bc arch/arm64/mm/fault.c:317
> > do_bad_area arch/arm64/mm/fault.c:466 [inline]
> > do_tag_check_fault+0x74/0x90 arch/arm64/mm/fault.c:737
> > do_mem_abort+0x44/0xb4 arch/arm64/mm/fault.c:813
> > el1_abort+0x40/0x60 arch/arm64/kernel/entry-common.c:357
> > el1h_64_sync_handler+0xb0/0xd0 arch/arm64/kernel/entry-common.c:408
> > el1h_64_sync+0x78/0x7c arch/arm64/kernel/entry.S:567
> > __entry_tramp_text_end+0xddc/0xd000
> > d_lookup+0x44/0x70 fs/dcache.c:2370
> > lookup_dcache+0x24/0x84 fs/namei.c:1520
> > __lookup_hash+0x24/0xd0 fs/namei.c:1543
> > kern_path_locked+0x90/0x10c fs/namei.c:2567
> > handle_remove+0x38/0x284 drivers/base/devtmpfs.c:312
> > handle drivers/base/devtmpfs.c:382 [inline]
> > devtmpfs_work_loop drivers/base/devtmpfs.c:395 [inline]
> > devtmpfsd+0x8c/0xd0 drivers/base/devtmpfs.c:437
> > kthread+0x150/0x15c kernel/kthread.c:319
> > ret_from_fork+0x10/0x20 arch/arm64/kernel/entry.S:756
> >
> > Here kernel unwinder prints __entry_tramp_text_end instead of __d_lookup.
> >
> > I've looked in more detail into the arm64 case:
> > d_lookup contains a static call to __d_lookup as expected:
> >
> > ffff8000102e0780 <d_lookup>:
> > ...
> > ffff8000102e07c0: 97ffffa4 bl ffff8000102e0650 <__d_lookup>
> > ...
> > ffff8000102e07e8: d65f03c0 ret
> >
> > and these symbols don't overlap or something:
> >
> > $ aarch64-linux-gnu-nm -nS vmlinux | egrep -C 1 " (t|T)
> > (__entry_tramp_text|__d_lookup)"
> > ffff8000102e01f0 0000000000000458 T d_alloc_parallel
> > ffff8000102e0650 0000000000000128 T __d_lookup
> > ffff8000102e0780 000000000000006c T d_lookup
> > --
> > ffff8000117a1f88 T __hibernate_exit_text_end
> > ffff8000117a2000 T __entry_tramp_text_start
> > ffff8000117a2000 00000000000007c8 T tramp_vectors
> > --
> > ffff8000117a27f0 0000000000000024 T tramp_exit_compat
> > ffff8000117a3000 T __entry_tramp_text_end
> > ffff8000117b0000 D _etext
> >
> > So it looks like in both cases the top fault frame is just wrong. But
> > I would assume it's extracted by arch-dependent code, so it's
> > suspicious that it affects both x86 and arm64...
> >
> > Any ideas what's happening?
>
> My suspicion for the x86 case is that kvm_fastop_exception is related
> to instruction emulation and the fault occurs in an emulated
> instruction?
Why would the kernel emulate a plain MOV?
2a: 4c 8b 21 mov (%rcx),%r12
And it would also mean a broken unwind because the emulated
instruction is in __d_lookup, so it should be in the stack trace.
> But I can't explain the arm64 case.
>
> > > d_lookup+0xd8/0x170 fs/dcache.c:2370
> > > lookup_dcache+0x1e/0x130 fs/namei.c:1520
> > > __lookup_hash+0x29/0x180 fs/namei.c:1543
> > > kern_path_locked+0x17e/0x320 fs/namei.c:2567
> > > handle_remove+0xa2/0x5fe drivers/base/devtmpfs.c:312
> > > handle drivers/base/devtmpfs.c:382 [inline]
> > > devtmpfs_work_loop drivers/base/devtmpfs.c:395 [inline]
> > > devtmpfsd+0x1b9/0x2a3 drivers/base/devtmpfs.c:437
> > > kthread+0x3e5/0x4d0 kernel/kthread.c:319
> > > ret_from_fork+0x1f/0x30 arch/x86/entry/entry_64.S:295
> > >
> > > kfence-#5 [0xffff88823bc0c000-0xffff88823bc0cfff, size=4096, cache=names_cache] allocated by task 22:
> > > getname_kernel+0x4e/0x370 fs/namei.c:226
> > > kern_path_locked+0x71/0x320 fs/namei.c:2558
> > > handle_remove+0xa2/0x5fe drivers/base/devtmpfs.c:312
> > > handle drivers/base/devtmpfs.c:382 [inline]
> > > devtmpfs_work_loop drivers/base/devtmpfs.c:395 [inline]
> > > devtmpfsd+0x1b9/0x2a3 drivers/base/devtmpfs.c:437
> > > kthread+0x3e5/0x4d0 kernel/kthread.c:319
> > > ret_from_fork+0x1f/0x30 arch/x86/entry/entry_64.S:295
> > >
> > > freed by task 22:
> > > putname.part.0+0xe1/0x120 fs/namei.c:270
> > > putname include/linux/err.h:41 [inline]
> > > filename_parentat fs/namei.c:2547 [inline]
> > > kern_path_locked+0xc2/0x320 fs/namei.c:2558
> > > handle_remove+0xa2/0x5fe drivers/base/devtmpfs.c:312
> > > handle drivers/base/devtmpfs.c:382 [inline]
> > > devtmpfs_work_loop drivers/base/devtmpfs.c:395 [inline]
> > > devtmpfsd+0x1b9/0x2a3 drivers/base/devtmpfs.c:437
> > > kthread+0x3e5/0x4d0 kernel/kthread.c:319
> > > ret_from_fork+0x1f/0x30 arch/x86/entry/entry_64.S:295
> > >
> > > CPU: 1 PID: 22 Comm: kdevtmpfs Not tainted 5.14.0-syzkaller #0
> > > Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 01/01/2011
> > > RIP: 0010:kvm_fastop_exception+0xf6d/0x105b
> > > Code: d3 ed e9 14 1b 6d f8 49 8d 0e 48 83 e1 f8 4c 8b 21 41 8d 0e 83 e1 07 c1 e1 03 49 d3 ec e9 6a 28 6d f8 49 8d 4d 00 48 83 e1 f8 <4c> 8b 21 41 8d 4d 00 83 e1 07 c1 e1 03 49 d3 ec e9 5a 32 6d f8 bd
> > > RSP: 0018:ffffc90000fe7ae8 EFLAGS: 00010282
> > > RAX: 0000000035736376 RBX: ffff88803b141cc0 RCX: ffff88823bc0c020
> > > RDX: ffffed100762839f RSI: 0000000000000004 RDI: 0000000000000007
> > > RBP: 0000000000000004 R08: 0000000000000000 R09: ffff88803b141cf0
> > > R10: ffffed100762839e R11: 0000000000000000 R12: ffff88823bc0c020
> > > R13: ffff88823bc0c020 R14: ffff88803b141cf0 R15: dffffc0000000000
> > > FS: 0000000000000000(0000) GS:ffff8880b9d00000(0000) knlGS:0000000000000000
> > > CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
> > > CR2: ffff88823bc0c020 CR3: 0000000029892000 CR4: 00000000001506e0
> > > DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
> > > DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400
> > > Call Trace:
> > > d_lookup+0xd8/0x170 fs/dcache.c:2370
> > > lookup_dcache+0x1e/0x130 fs/namei.c:1520
> > > __lookup_hash+0x29/0x180 fs/namei.c:1543
> > > kern_path_locked+0x17e/0x320 fs/namei.c:2567
> > > handle_remove+0xa2/0x5fe drivers/base/devtmpfs.c:312
> > > handle drivers/base/devtmpfs.c:382 [inline]
> > > devtmpfs_work_loop drivers/base/devtmpfs.c:395 [inline]
> > > devtmpfsd+0x1b9/0x2a3 drivers/base/devtmpfs.c:437
> > > kthread+0x3e5/0x4d0 kernel/kthread.c:319
> > > ret_from_fork+0x1f/0x30 arch/x86/entry/entry_64.S:295
> > > ==================================================================
> > > ----------------
> > > Code disassembly (best guess):
> > > 0: d3 ed shr %cl,%ebp
> > > 2: e9 14 1b 6d f8 jmpq 0xf86d1b1b
> > > 7: 49 8d 0e lea (%r14),%rcx
> > > a: 48 83 e1 f8 and $0xfffffffffffffff8,%rcx
> > > e: 4c 8b 21 mov (%rcx),%r12
> > > 11: 41 8d 0e lea (%r14),%ecx
> > > 14: 83 e1 07 and $0x7,%ecx
> > > 17: c1 e1 03 shl $0x3,%ecx
> > > 1a: 49 d3 ec shr %cl,%r12
> > > 1d: e9 6a 28 6d f8 jmpq 0xf86d288c
> > > 22: 49 8d 4d 00 lea 0x0(%r13),%rcx
> > > 26: 48 83 e1 f8 and $0xfffffffffffffff8,%rcx
> > > * 2a: 4c 8b 21 mov (%rcx),%r12 <-- trapping instruction
> > > 2d: 41 8d 4d 00 lea 0x0(%r13),%ecx
> > > 31: 83 e1 07 and $0x7,%ecx
> > > 34: c1 e1 03 shl $0x3,%ecx
> > > 37: 49 d3 ec shr %cl,%r12
> > > 3a: e9 5a 32 6d f8 jmpq 0xf86d3299
> > > 3f: bd .byte 0xbd
> > >
> > >
> > > ---
> > > This report 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 issue. See:
> > > https://goo.gl/tpsmEJ#status for how to communicate with syzbot.
_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel
^ permalink raw reply [flat|nested] 17+ messages in thread
* Re: [syzbot] upstream test error: KFENCE: use-after-free in kvm_fastop_exception
2021-09-17 12:43 ` Dmitry Vyukov
@ 2021-09-21 23:34 ` Sean Christopherson
-1 siblings, 0 replies; 17+ messages in thread
From: Sean Christopherson @ 2021-09-21 23:34 UTC (permalink / raw)
To: Dmitry Vyukov
Cc: Marco Elver, syzbot, linux-fsdevel, linux-kernel, syzkaller-bugs,
viro, the arch/x86 maintainers, Linux ARM, kasan-dev
On Fri, Sep 17, 2021, Dmitry Vyukov wrote:
> On Fri, 17 Sept 2021 at 13:04, Marco Elver <elver@google.com> wrote:
> > > So it looks like in both cases the top fault frame is just wrong. But
> > > I would assume it's extracted by arch-dependent code, so it's
> > > suspicious that it affects both x86 and arm64...
> > >
> > > Any ideas what's happening?
> >
> > My suspicion for the x86 case is that kvm_fastop_exception is related
> > to instruction emulation and the fault occurs in an emulated
> > instruction?
>
> Why would the kernel emulate a plain MOV?
> 2a: 4c 8b 21 mov (%rcx),%r12
>
> And it would also mean a broken unwind because the emulated
> instruction is in __d_lookup, so it should be in the stack trace.
kvm_fastop_exception is a red herring. It's indeed related to emulation, and
while MOV emulation is common in KVM, that emulation is for KVM guests not for
the host kernel where this splat occurs (ignoring the fact that the "host" is
itself a guest).
kvm_fastop_exception is out-of-line fixup, and certainly shouldn't be reachable
via d_lookup. It's also two instruction, XOR+RET, neither of which are in the
code stream.
IIRC, the unwinder gets confused when given an IP that's in out-of-line code,
e.g. exception fixup like this. If you really want to find out what code blew
up, you might be able to objdump -D the kernel and search for unique, matching
disassembly, e.g. find "jmpq 0xf86d288c" and go from there.
^ permalink raw reply [flat|nested] 17+ messages in thread
* Re: [syzbot] upstream test error: KFENCE: use-after-free in kvm_fastop_exception
@ 2021-09-21 23:34 ` Sean Christopherson
0 siblings, 0 replies; 17+ messages in thread
From: Sean Christopherson @ 2021-09-21 23:34 UTC (permalink / raw)
To: Dmitry Vyukov
Cc: Marco Elver, syzbot, linux-fsdevel, linux-kernel, syzkaller-bugs,
viro, the arch/x86 maintainers, Linux ARM, kasan-dev
On Fri, Sep 17, 2021, Dmitry Vyukov wrote:
> On Fri, 17 Sept 2021 at 13:04, Marco Elver <elver@google.com> wrote:
> > > So it looks like in both cases the top fault frame is just wrong. But
> > > I would assume it's extracted by arch-dependent code, so it's
> > > suspicious that it affects both x86 and arm64...
> > >
> > > Any ideas what's happening?
> >
> > My suspicion for the x86 case is that kvm_fastop_exception is related
> > to instruction emulation and the fault occurs in an emulated
> > instruction?
>
> Why would the kernel emulate a plain MOV?
> 2a: 4c 8b 21 mov (%rcx),%r12
>
> And it would also mean a broken unwind because the emulated
> instruction is in __d_lookup, so it should be in the stack trace.
kvm_fastop_exception is a red herring. It's indeed related to emulation, and
while MOV emulation is common in KVM, that emulation is for KVM guests not for
the host kernel where this splat occurs (ignoring the fact that the "host" is
itself a guest).
kvm_fastop_exception is out-of-line fixup, and certainly shouldn't be reachable
via d_lookup. It's also two instruction, XOR+RET, neither of which are in the
code stream.
IIRC, the unwinder gets confused when given an IP that's in out-of-line code,
e.g. exception fixup like this. If you really want to find out what code blew
up, you might be able to objdump -D the kernel and search for unique, matching
disassembly, e.g. find "jmpq 0xf86d288c" and go from there.
_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel
^ permalink raw reply [flat|nested] 17+ messages in thread
* Re: [syzbot] upstream test error: KFENCE: use-after-free in kvm_fastop_exception
2021-09-21 23:34 ` Sean Christopherson
@ 2021-09-27 14:16 ` Dmitry Vyukov
-1 siblings, 0 replies; 17+ messages in thread
From: Dmitry Vyukov @ 2021-09-27 14:16 UTC (permalink / raw)
To: Sean Christopherson
Cc: Marco Elver, syzbot, linux-fsdevel, linux-kernel, syzkaller-bugs,
viro, the arch/x86 maintainers, Linux ARM, kasan-dev
On Wed, 22 Sept 2021 at 01:34, 'Sean Christopherson' via
syzkaller-bugs <syzkaller-bugs@googlegroups.com> wrote:
>
> On Fri, Sep 17, 2021, Dmitry Vyukov wrote:
> > On Fri, 17 Sept 2021 at 13:04, Marco Elver <elver@google.com> wrote:
> > > > So it looks like in both cases the top fault frame is just wrong. But
> > > > I would assume it's extracted by arch-dependent code, so it's
> > > > suspicious that it affects both x86 and arm64...
> > > >
> > > > Any ideas what's happening?
> > >
> > > My suspicion for the x86 case is that kvm_fastop_exception is related
> > > to instruction emulation and the fault occurs in an emulated
> > > instruction?
> >
> > Why would the kernel emulate a plain MOV?
> > 2a: 4c 8b 21 mov (%rcx),%r12
> >
> > And it would also mean a broken unwind because the emulated
> > instruction is in __d_lookup, so it should be in the stack trace.
>
> kvm_fastop_exception is a red herring. It's indeed related to emulation, and
> while MOV emulation is common in KVM, that emulation is for KVM guests not for
> the host kernel where this splat occurs (ignoring the fact that the "host" is
> itself a guest).
>
> kvm_fastop_exception is out-of-line fixup, and certainly shouldn't be reachable
> via d_lookup. It's also two instruction, XOR+RET, neither of which are in the
> code stream.
>
> IIRC, the unwinder gets confused when given an IP that's in out-of-line code,
> e.g. exception fixup like this. If you really want to find out what code blew
> up, you might be able to objdump -D the kernel and search for unique, matching
> disassembly, e.g. find "jmpq 0xf86d288c" and go from there.
Hi Sean,
Thanks for the info.
I don't want to find out what code blew (it's __d_lookup).
I am interested in getting the unwinder fixed to output truthful and
useful frames.
Is there more info on this "the unwinder gets confused"? Bug filed
somewhere or an email thread? Is it on anybody's radar?
^ permalink raw reply [flat|nested] 17+ messages in thread
* Re: [syzbot] upstream test error: KFENCE: use-after-free in kvm_fastop_exception
@ 2021-09-27 14:16 ` Dmitry Vyukov
0 siblings, 0 replies; 17+ messages in thread
From: Dmitry Vyukov @ 2021-09-27 14:16 UTC (permalink / raw)
To: Sean Christopherson
Cc: Marco Elver, syzbot, linux-fsdevel, linux-kernel, syzkaller-bugs,
viro, the arch/x86 maintainers, Linux ARM, kasan-dev
On Wed, 22 Sept 2021 at 01:34, 'Sean Christopherson' via
syzkaller-bugs <syzkaller-bugs@googlegroups.com> wrote:
>
> On Fri, Sep 17, 2021, Dmitry Vyukov wrote:
> > On Fri, 17 Sept 2021 at 13:04, Marco Elver <elver@google.com> wrote:
> > > > So it looks like in both cases the top fault frame is just wrong. But
> > > > I would assume it's extracted by arch-dependent code, so it's
> > > > suspicious that it affects both x86 and arm64...
> > > >
> > > > Any ideas what's happening?
> > >
> > > My suspicion for the x86 case is that kvm_fastop_exception is related
> > > to instruction emulation and the fault occurs in an emulated
> > > instruction?
> >
> > Why would the kernel emulate a plain MOV?
> > 2a: 4c 8b 21 mov (%rcx),%r12
> >
> > And it would also mean a broken unwind because the emulated
> > instruction is in __d_lookup, so it should be in the stack trace.
>
> kvm_fastop_exception is a red herring. It's indeed related to emulation, and
> while MOV emulation is common in KVM, that emulation is for KVM guests not for
> the host kernel where this splat occurs (ignoring the fact that the "host" is
> itself a guest).
>
> kvm_fastop_exception is out-of-line fixup, and certainly shouldn't be reachable
> via d_lookup. It's also two instruction, XOR+RET, neither of which are in the
> code stream.
>
> IIRC, the unwinder gets confused when given an IP that's in out-of-line code,
> e.g. exception fixup like this. If you really want to find out what code blew
> up, you might be able to objdump -D the kernel and search for unique, matching
> disassembly, e.g. find "jmpq 0xf86d288c" and go from there.
Hi Sean,
Thanks for the info.
I don't want to find out what code blew (it's __d_lookup).
I am interested in getting the unwinder fixed to output truthful and
useful frames.
Is there more info on this "the unwinder gets confused"? Bug filed
somewhere or an email thread? Is it on anybody's radar?
_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel
^ permalink raw reply [flat|nested] 17+ messages in thread
* Re: [syzbot] upstream test error: KFENCE: use-after-free in kvm_fastop_exception
2021-09-27 14:16 ` Dmitry Vyukov
@ 2021-09-27 16:07 ` Sean Christopherson
-1 siblings, 0 replies; 17+ messages in thread
From: Sean Christopherson @ 2021-09-27 16:07 UTC (permalink / raw)
To: Dmitry Vyukov
Cc: Marco Elver, syzbot, linux-fsdevel, linux-kernel, syzkaller-bugs,
viro, the arch/x86 maintainers, Linux ARM, kasan-dev,
Josh Poimboeuf, Peter Zijlstra
+Josh and PeterZ
On Mon, Sep 27, 2021, Dmitry Vyukov wrote:
> On Wed, 22 Sept 2021 at 01:34, 'Sean Christopherson' via
> syzkaller-bugs <syzkaller-bugs@googlegroups.com> wrote:
> >
> > On Fri, Sep 17, 2021, Dmitry Vyukov wrote:
> > > On Fri, 17 Sept 2021 at 13:04, Marco Elver <elver@google.com> wrote:
> > > > > So it looks like in both cases the top fault frame is just wrong. But
> > > > > I would assume it's extracted by arch-dependent code, so it's
> > > > > suspicious that it affects both x86 and arm64...
> > > > >
> > > > > Any ideas what's happening?
> > > >
> > > > My suspicion for the x86 case is that kvm_fastop_exception is related
> > > > to instruction emulation and the fault occurs in an emulated
> > > > instruction?
> > >
> > > Why would the kernel emulate a plain MOV?
> > > 2a: 4c 8b 21 mov (%rcx),%r12
> > >
> > > And it would also mean a broken unwind because the emulated
> > > instruction is in __d_lookup, so it should be in the stack trace.
> >
> > kvm_fastop_exception is a red herring. It's indeed related to emulation, and
> > while MOV emulation is common in KVM, that emulation is for KVM guests not for
> > the host kernel where this splat occurs (ignoring the fact that the "host" is
> > itself a guest).
> >
> > kvm_fastop_exception is out-of-line fixup, and certainly shouldn't be reachable
> > via d_lookup. It's also two instruction, XOR+RET, neither of which are in the
> > code stream.
> >
> > IIRC, the unwinder gets confused when given an IP that's in out-of-line code,
> > e.g. exception fixup like this. If you really want to find out what code blew
> > up, you might be able to objdump -D the kernel and search for unique, matching
> > disassembly, e.g. find "jmpq 0xf86d288c" and go from there.
>
> Hi Sean,
>
> Thanks for the info.
>
> I don't want to find out what code blew (it's __d_lookup).
> I am interested in getting the unwinder fixed to output truthful and
> useful frames.
I was asking about the exact location to confirm that the explosion is indeed
from exception fixup, which is the "unwinder scenario get confused" I was thinking
of. Based on the disassembly from syzbot, that does indeed appear to be the case
here, i.e. this
2a: 4c 8b 21 mov (%rcx),%r12
is from exception fixup from somewhere in __d_lookup (can't tell exactly what
it's from, maybe KASAN?).
> Is there more info on this "the unwinder gets confused"? Bug filed
> somewhere or an email thread? Is it on anybody's radar?
I don't know if there's a bug report or if this is on anyone's radar. The issue
I've encountered in the past, and what I'm pretty sure is being hit here, is that
the ORC unwinder doesn't play nice with out-of-line fixup code, presumably because
there are no tables for the fixup. I believe kvm_fastop_exception() gets blamed
because it's the first label that's found when searching back through the tables.
^ permalink raw reply [flat|nested] 17+ messages in thread
* Re: [syzbot] upstream test error: KFENCE: use-after-free in kvm_fastop_exception
@ 2021-09-27 16:07 ` Sean Christopherson
0 siblings, 0 replies; 17+ messages in thread
From: Sean Christopherson @ 2021-09-27 16:07 UTC (permalink / raw)
To: Dmitry Vyukov
Cc: Marco Elver, syzbot, linux-fsdevel, linux-kernel, syzkaller-bugs,
viro, the arch/x86 maintainers, Linux ARM, kasan-dev,
Josh Poimboeuf, Peter Zijlstra
+Josh and PeterZ
On Mon, Sep 27, 2021, Dmitry Vyukov wrote:
> On Wed, 22 Sept 2021 at 01:34, 'Sean Christopherson' via
> syzkaller-bugs <syzkaller-bugs@googlegroups.com> wrote:
> >
> > On Fri, Sep 17, 2021, Dmitry Vyukov wrote:
> > > On Fri, 17 Sept 2021 at 13:04, Marco Elver <elver@google.com> wrote:
> > > > > So it looks like in both cases the top fault frame is just wrong. But
> > > > > I would assume it's extracted by arch-dependent code, so it's
> > > > > suspicious that it affects both x86 and arm64...
> > > > >
> > > > > Any ideas what's happening?
> > > >
> > > > My suspicion for the x86 case is that kvm_fastop_exception is related
> > > > to instruction emulation and the fault occurs in an emulated
> > > > instruction?
> > >
> > > Why would the kernel emulate a plain MOV?
> > > 2a: 4c 8b 21 mov (%rcx),%r12
> > >
> > > And it would also mean a broken unwind because the emulated
> > > instruction is in __d_lookup, so it should be in the stack trace.
> >
> > kvm_fastop_exception is a red herring. It's indeed related to emulation, and
> > while MOV emulation is common in KVM, that emulation is for KVM guests not for
> > the host kernel where this splat occurs (ignoring the fact that the "host" is
> > itself a guest).
> >
> > kvm_fastop_exception is out-of-line fixup, and certainly shouldn't be reachable
> > via d_lookup. It's also two instruction, XOR+RET, neither of which are in the
> > code stream.
> >
> > IIRC, the unwinder gets confused when given an IP that's in out-of-line code,
> > e.g. exception fixup like this. If you really want to find out what code blew
> > up, you might be able to objdump -D the kernel and search for unique, matching
> > disassembly, e.g. find "jmpq 0xf86d288c" and go from there.
>
> Hi Sean,
>
> Thanks for the info.
>
> I don't want to find out what code blew (it's __d_lookup).
> I am interested in getting the unwinder fixed to output truthful and
> useful frames.
I was asking about the exact location to confirm that the explosion is indeed
from exception fixup, which is the "unwinder scenario get confused" I was thinking
of. Based on the disassembly from syzbot, that does indeed appear to be the case
here, i.e. this
2a: 4c 8b 21 mov (%rcx),%r12
is from exception fixup from somewhere in __d_lookup (can't tell exactly what
it's from, maybe KASAN?).
> Is there more info on this "the unwinder gets confused"? Bug filed
> somewhere or an email thread? Is it on anybody's radar?
I don't know if there's a bug report or if this is on anyone's radar. The issue
I've encountered in the past, and what I'm pretty sure is being hit here, is that
the ORC unwinder doesn't play nice with out-of-line fixup code, presumably because
there are no tables for the fixup. I believe kvm_fastop_exception() gets blamed
because it's the first label that's found when searching back through the tables.
_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel
^ permalink raw reply [flat|nested] 17+ messages in thread
* Re: [syzbot] upstream test error: KFENCE: use-after-free in kvm_fastop_exception
2021-09-27 16:07 ` Sean Christopherson
@ 2021-09-27 23:45 ` Josh Poimboeuf
-1 siblings, 0 replies; 17+ messages in thread
From: Josh Poimboeuf @ 2021-09-27 23:45 UTC (permalink / raw)
To: Sean Christopherson
Cc: Dmitry Vyukov, Marco Elver, syzbot, linux-fsdevel, linux-kernel,
syzkaller-bugs, viro, the arch/x86 maintainers, Linux ARM,
kasan-dev, Peter Zijlstra
On Mon, Sep 27, 2021 at 04:07:51PM +0000, Sean Christopherson wrote:
> I was asking about the exact location to confirm that the explosion is indeed
> from exception fixup, which is the "unwinder scenario get confused" I was thinking
> of. Based on the disassembly from syzbot, that does indeed appear to be the case
> here, i.e. this
>
> 2a: 4c 8b 21 mov (%rcx),%r12
>
> is from exception fixup from somewhere in __d_lookup (can't tell exactly what
> it's from, maybe KASAN?).
>
> > Is there more info on this "the unwinder gets confused"? Bug filed
> > somewhere or an email thread? Is it on anybody's radar?
>
> I don't know if there's a bug report or if this is on anyone's radar. The issue
> I've encountered in the past, and what I'm pretty sure is being hit here, is that
> the ORC unwinder doesn't play nice with out-of-line fixup code, presumably because
> there are no tables for the fixup. I believe kvm_fastop_exception() gets blamed
> because it's the first label that's found when searching back through the tables.
The ORC unwinder actually knows about .fixup, and unwinding through the
.fixup code worked here, as evidenced by the entire stacktrace getting
printed. Otherwise there would have been a bunch of question marks in
the stack trace.
The problem reported here -- falsely printing kvm_fastop_exception -- is
actually in the arch-independent printing of symbol names, done by
__sprint_symbol(). Most .fixup code fragments are anonymous, in the
sense that they don't have symbols associated with them. For x86, here
are the only defined symbols in .fixup:
ffffffff81e02408 T kvm_fastop_exception
ffffffff81e02728 t .E_read_words
ffffffff81e0272b t .E_leading_bytes
ffffffff81e0272d t .E_trailing_bytes
ffffffff81e02734 t .E_write_words
ffffffff81e02740 t .E_copy
There's a lot of anonymous .fixup code which happens to be placed in the
gap between "kvm_fastop_exception" and ".E_read_words". The kernel
symbol printing code will go backwards from the given address and will
print the first symbol it finds. So any anonymous code in that gap will
falsely be reported as kvm_fastop_exception().
I'm thinking the ideal way to fix this would be getting rid of the
.fixup section altogether, and instead place a function's corresponding
fixup code in a cold part of the original function, with the help of
asm_goto and cold label attributes.
That way, the original faulting function would be printed instead of an
obscure reference to an anonymous .fixup code fragment. It would have
other benefits as well. For example, not breaking livepatch...
I'll try to play around with it.
--
Josh
^ permalink raw reply [flat|nested] 17+ messages in thread
* Re: [syzbot] upstream test error: KFENCE: use-after-free in kvm_fastop_exception
@ 2021-09-27 23:45 ` Josh Poimboeuf
0 siblings, 0 replies; 17+ messages in thread
From: Josh Poimboeuf @ 2021-09-27 23:45 UTC (permalink / raw)
To: Sean Christopherson
Cc: Dmitry Vyukov, Marco Elver, syzbot, linux-fsdevel, linux-kernel,
syzkaller-bugs, viro, the arch/x86 maintainers, Linux ARM,
kasan-dev, Peter Zijlstra
On Mon, Sep 27, 2021 at 04:07:51PM +0000, Sean Christopherson wrote:
> I was asking about the exact location to confirm that the explosion is indeed
> from exception fixup, which is the "unwinder scenario get confused" I was thinking
> of. Based on the disassembly from syzbot, that does indeed appear to be the case
> here, i.e. this
>
> 2a: 4c 8b 21 mov (%rcx),%r12
>
> is from exception fixup from somewhere in __d_lookup (can't tell exactly what
> it's from, maybe KASAN?).
>
> > Is there more info on this "the unwinder gets confused"? Bug filed
> > somewhere or an email thread? Is it on anybody's radar?
>
> I don't know if there's a bug report or if this is on anyone's radar. The issue
> I've encountered in the past, and what I'm pretty sure is being hit here, is that
> the ORC unwinder doesn't play nice with out-of-line fixup code, presumably because
> there are no tables for the fixup. I believe kvm_fastop_exception() gets blamed
> because it's the first label that's found when searching back through the tables.
The ORC unwinder actually knows about .fixup, and unwinding through the
.fixup code worked here, as evidenced by the entire stacktrace getting
printed. Otherwise there would have been a bunch of question marks in
the stack trace.
The problem reported here -- falsely printing kvm_fastop_exception -- is
actually in the arch-independent printing of symbol names, done by
__sprint_symbol(). Most .fixup code fragments are anonymous, in the
sense that they don't have symbols associated with them. For x86, here
are the only defined symbols in .fixup:
ffffffff81e02408 T kvm_fastop_exception
ffffffff81e02728 t .E_read_words
ffffffff81e0272b t .E_leading_bytes
ffffffff81e0272d t .E_trailing_bytes
ffffffff81e02734 t .E_write_words
ffffffff81e02740 t .E_copy
There's a lot of anonymous .fixup code which happens to be placed in the
gap between "kvm_fastop_exception" and ".E_read_words". The kernel
symbol printing code will go backwards from the given address and will
print the first symbol it finds. So any anonymous code in that gap will
falsely be reported as kvm_fastop_exception().
I'm thinking the ideal way to fix this would be getting rid of the
.fixup section altogether, and instead place a function's corresponding
fixup code in a cold part of the original function, with the help of
asm_goto and cold label attributes.
That way, the original faulting function would be printed instead of an
obscure reference to an anonymous .fixup code fragment. It would have
other benefits as well. For example, not breaking livepatch...
I'll try to play around with it.
--
Josh
_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel
^ permalink raw reply [flat|nested] 17+ messages in thread
* Re: [syzbot] upstream test error: KFENCE: use-after-free in kvm_fastop_exception
2021-09-27 23:45 ` Josh Poimboeuf
@ 2021-09-28 10:16 ` Dmitry Vyukov
-1 siblings, 0 replies; 17+ messages in thread
From: Dmitry Vyukov @ 2021-09-28 10:16 UTC (permalink / raw)
To: Josh Poimboeuf
Cc: Sean Christopherson, Marco Elver, syzbot, linux-fsdevel,
linux-kernel, syzkaller-bugs, viro, the arch/x86 maintainers,
Linux ARM, kasan-dev, Peter Zijlstra
On Tue, 28 Sept 2021 at 01:45, Josh Poimboeuf <jpoimboe@redhat.com> wrote:
>
> On Mon, Sep 27, 2021 at 04:07:51PM +0000, Sean Christopherson wrote:
> > I was asking about the exact location to confirm that the explosion is indeed
> > from exception fixup, which is the "unwinder scenario get confused" I was thinking
> > of. Based on the disassembly from syzbot, that does indeed appear to be the case
> > here, i.e. this
> >
> > 2a: 4c 8b 21 mov (%rcx),%r12
> >
> > is from exception fixup from somewhere in __d_lookup (can't tell exactly what
> > it's from, maybe KASAN?).
> >
> > > Is there more info on this "the unwinder gets confused"? Bug filed
> > > somewhere or an email thread? Is it on anybody's radar?
> >
> > I don't know if there's a bug report or if this is on anyone's radar. The issue
> > I've encountered in the past, and what I'm pretty sure is being hit here, is that
> > the ORC unwinder doesn't play nice with out-of-line fixup code, presumably because
> > there are no tables for the fixup. I believe kvm_fastop_exception() gets blamed
> > because it's the first label that's found when searching back through the tables.
>
> The ORC unwinder actually knows about .fixup, and unwinding through the
> .fixup code worked here, as evidenced by the entire stacktrace getting
> printed. Otherwise there would have been a bunch of question marks in
> the stack trace.
>
> The problem reported here -- falsely printing kvm_fastop_exception -- is
> actually in the arch-independent printing of symbol names, done by
> __sprint_symbol(). Most .fixup code fragments are anonymous, in the
> sense that they don't have symbols associated with them. For x86, here
> are the only defined symbols in .fixup:
>
> ffffffff81e02408 T kvm_fastop_exception
> ffffffff81e02728 t .E_read_words
> ffffffff81e0272b t .E_leading_bytes
> ffffffff81e0272d t .E_trailing_bytes
> ffffffff81e02734 t .E_write_words
> ffffffff81e02740 t .E_copy
>
> There's a lot of anonymous .fixup code which happens to be placed in the
> gap between "kvm_fastop_exception" and ".E_read_words". The kernel
> symbol printing code will go backwards from the given address and will
> print the first symbol it finds. So any anonymous code in that gap will
> falsely be reported as kvm_fastop_exception().
>
> I'm thinking the ideal way to fix this would be getting rid of the
> .fixup section altogether, and instead place a function's corresponding
> fixup code in a cold part of the original function, with the help of
> asm_goto and cold label attributes.
>
> That way, the original faulting function would be printed instead of an
> obscure reference to an anonymous .fixup code fragment. It would have
> other benefits as well. For example, not breaking livepatch...
>
> I'll try to play around with it.
Thanks for debugging this, Josh.
I think your solution can also help arm64 as it has the same issue.
^ permalink raw reply [flat|nested] 17+ messages in thread
* Re: [syzbot] upstream test error: KFENCE: use-after-free in kvm_fastop_exception
@ 2021-09-28 10:16 ` Dmitry Vyukov
0 siblings, 0 replies; 17+ messages in thread
From: Dmitry Vyukov @ 2021-09-28 10:16 UTC (permalink / raw)
To: Josh Poimboeuf
Cc: Sean Christopherson, Marco Elver, syzbot, linux-fsdevel,
linux-kernel, syzkaller-bugs, viro, the arch/x86 maintainers,
Linux ARM, kasan-dev, Peter Zijlstra
On Tue, 28 Sept 2021 at 01:45, Josh Poimboeuf <jpoimboe@redhat.com> wrote:
>
> On Mon, Sep 27, 2021 at 04:07:51PM +0000, Sean Christopherson wrote:
> > I was asking about the exact location to confirm that the explosion is indeed
> > from exception fixup, which is the "unwinder scenario get confused" I was thinking
> > of. Based on the disassembly from syzbot, that does indeed appear to be the case
> > here, i.e. this
> >
> > 2a: 4c 8b 21 mov (%rcx),%r12
> >
> > is from exception fixup from somewhere in __d_lookup (can't tell exactly what
> > it's from, maybe KASAN?).
> >
> > > Is there more info on this "the unwinder gets confused"? Bug filed
> > > somewhere or an email thread? Is it on anybody's radar?
> >
> > I don't know if there's a bug report or if this is on anyone's radar. The issue
> > I've encountered in the past, and what I'm pretty sure is being hit here, is that
> > the ORC unwinder doesn't play nice with out-of-line fixup code, presumably because
> > there are no tables for the fixup. I believe kvm_fastop_exception() gets blamed
> > because it's the first label that's found when searching back through the tables.
>
> The ORC unwinder actually knows about .fixup, and unwinding through the
> .fixup code worked here, as evidenced by the entire stacktrace getting
> printed. Otherwise there would have been a bunch of question marks in
> the stack trace.
>
> The problem reported here -- falsely printing kvm_fastop_exception -- is
> actually in the arch-independent printing of symbol names, done by
> __sprint_symbol(). Most .fixup code fragments are anonymous, in the
> sense that they don't have symbols associated with them. For x86, here
> are the only defined symbols in .fixup:
>
> ffffffff81e02408 T kvm_fastop_exception
> ffffffff81e02728 t .E_read_words
> ffffffff81e0272b t .E_leading_bytes
> ffffffff81e0272d t .E_trailing_bytes
> ffffffff81e02734 t .E_write_words
> ffffffff81e02740 t .E_copy
>
> There's a lot of anonymous .fixup code which happens to be placed in the
> gap between "kvm_fastop_exception" and ".E_read_words". The kernel
> symbol printing code will go backwards from the given address and will
> print the first symbol it finds. So any anonymous code in that gap will
> falsely be reported as kvm_fastop_exception().
>
> I'm thinking the ideal way to fix this would be getting rid of the
> .fixup section altogether, and instead place a function's corresponding
> fixup code in a cold part of the original function, with the help of
> asm_goto and cold label attributes.
>
> That way, the original faulting function would be printed instead of an
> obscure reference to an anonymous .fixup code fragment. It would have
> other benefits as well. For example, not breaking livepatch...
>
> I'll try to play around with it.
Thanks for debugging this, Josh.
I think your solution can also help arm64 as it has the same issue.
_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel
^ permalink raw reply [flat|nested] 17+ messages in thread
end of thread, other threads:[~2021-09-28 10:18 UTC | newest]
Thread overview: 17+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2021-09-04 18:58 [syzbot] upstream test error: KFENCE: use-after-free in kvm_fastop_exception syzbot
2021-09-17 10:00 ` Dmitry Vyukov
2021-09-17 10:00 ` Dmitry Vyukov
2021-09-17 11:04 ` Marco Elver
2021-09-17 11:04 ` Marco Elver
2021-09-17 12:43 ` Dmitry Vyukov
2021-09-17 12:43 ` Dmitry Vyukov
2021-09-21 23:34 ` Sean Christopherson
2021-09-21 23:34 ` Sean Christopherson
2021-09-27 14:16 ` Dmitry Vyukov
2021-09-27 14:16 ` Dmitry Vyukov
2021-09-27 16:07 ` Sean Christopherson
2021-09-27 16:07 ` Sean Christopherson
2021-09-27 23:45 ` Josh Poimboeuf
2021-09-27 23:45 ` Josh Poimboeuf
2021-09-28 10:16 ` Dmitry Vyukov
2021-09-28 10:16 ` Dmitry Vyukov
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.