Skip to contentSkip to conversation

BUG: KASAN: global-out-of-bounds in strscpy+0x807/0x970

DISCUSSION Jakub Kicinski 4 participants 6 messages Oct 11, 2017 1 message not in the archive

lore
Share & tools
lore
#0 lore raw

I'm hitting this on sysctl -a with net-next (4.14-rc4).

I saw that seccomp_actions_logged_handler was introduced not-so-long-ago by Tyler, is there a fix for this?

[  753.503760] BUG: KASAN: global-out-of-bounds in strscpy+0x807/0x970
[  753.510892] Read of size 8 at addr ffffffffa96b97c8 by task sysctl/1787
[  753.518416] 
[  753.520198] CPU: 1 PID: 1787 Comm: sysctl Not tainted 4.14.0-rc4-debug-00822-g07ec06ef312a #316
[  753.538598] Call Trace:
[  753.541449]  dump_stack+0xb3/0x140
[  753.545372]  ? _atomic_dec_and_lock+0x2b0/0x2b0
[  753.550564]  ? show_regs_print_info+0x6d/0x6d
[  753.555562]  print_address_description+0x25f/0x440
[  753.561044]  ? strscpy+0x807/0x970
[  753.564967]  kasan_report+0x1b8/0x430
[  753.569182]  ? strscpy+0x807/0x970
[  753.573107]  strscpy+0x807/0x970
[  753.576838]  ? strncpy+0xf0/0xf0
[  753.580576]  ? sched_clock_cpu+0x18/0x290
[  753.585185]  seccomp_actions_logged_handler+0x384/0x8d0
[  753.591157]  ? seccomp_send_sigsys+0xa0/0xa0
[  753.596046]  ? lock_downgrade+0x740/0x740
[  753.600648]  ? get_lock_stats+0x18/0x160
[  753.605159]  ? sysctl_head_grab+0x1b/0xc0
[  753.609766]  ? lock_contended+0x1130/0x1130
[  753.614566]  ? lock_release+0x14e0/0x14e0
[  753.619171]  ? do_raw_spin_trylock+0x1c0/0x1c0
[  753.624261]  ? lock_downgrade+0x660/0x740
[  753.628858]  ? do_raw_spin_lock+0x1f0/0x1f0
[  753.633657]  proc_sys_call_handler+0x216/0x3c0
[  753.638752]  ? proc_sys_permission+0x250/0x250
[  753.643847]  ? do_mmap+0x622/0xf50
[  753.647772]  __vfs_read+0xea/0x9a0
[  753.651696]  ? vfs_copy_file_range+0xc60/0xc60
[  753.656787]  ? security_mmap_file+0x66/0x1d0
[  753.661686]  ? common_file_perm+0x1e0/0x6b0
[  753.666485]  ? vm_mmap_pgoff+0x1b6/0x250
[  753.670993]  ? fsnotify+0x1c80/0x1c80
[  753.675209]  ? apparmor_task_setrlimit+0x690/0x690
[  753.680687]  ? vma_is_stack_for_current+0xb0/0xb0
[  753.686073]  vfs_read+0x109/0x3e0
[  753.689900]  SyS_read+0xcc/0x1d0
[  753.693628]  ? kernel_write+0x1a0/0x1a0
[  753.698034]  ? trace_hardirqs_on_caller+0x3cb/0x680
[  753.703610]  ? trace_hardirqs_on_thunk+0x1a/0x1c
[  753.708898]  entry_SYSCALL_64_fastpath+0x1c/0xb1
[  753.714180] RIP: 0033:0x7f62c82296b0
[  753.718294] RSP: 002b:00007ffc2d6726c8 EFLAGS: 00000246 ORIG_RAX: 0000000000000000
[  753.726919] RAX: ffffffffffffffda RBX: 0000000000000046 RCX: 00007f62c82296b0
[  753.735028] RDX: 0000000000000400 RSI: 00007f62c8722000 RDI: 0000000000000006
[  753.743138] RBP: 0000000000000000 R08: 00000000ffffffff R09: 0000000000000000
[  753.751246] R10: 0000000000000022 R11: 0000000000000246 R12: 0000000000000000
[  753.759355] R13: 0000000001d63650 R14: 00007f62c87186a0 R15: 0000000000000028
[  753.767478] 
[  753.769256] The buggy address belongs to the variable:
[  753.775125]  kprobes_seq_ops+0x1268/0x15c0
[  753.779820] 
[  753.781595] Memory state around the buggy address:
[  753.787075]  ffffffffa96b9680: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
[  753.795301]  ffffffffa96b9700: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
[  753.803526] >ffffffffa96b9780: 07 fa fa fa fa fa fa fa 00 05 fa fa fa fa fa fa
[  753.811751]                                               ^
[  753.818105]  ffffffffa96b9800: 02 fa fa fa fa fa fa fa 00 00 04 fa fa fa fa fa
[  753.826330]  ffffffffa96b9880: 00 00 00 01 fa fa fa fa 00 00 00 00 01 fa fa fa
[  753.834554] ==================================================================
#1 lore raw
Quoted 1 line · Jakub Kicinski

I'm hitting this on sysctl -a with net-next (4.14-rc4).

Hey Jakub - thanks for the bug report!

Quoted 3 lines

I saw that seccomp_actions_logged_handler was introduced not-so-long-ago by Tyler, is there a fix for this?

No, this is the first I've heard of it. I'll have a look.

Tyler

Quoted 63 lines
[  753.503760] BUG: KASAN: global-out-of-bounds in strscpy+0x807/0x970
[  753.510892] Read of size 8 at addr ffffffffa96b97c8 by task sysctl/1787
[  753.518416] 
[  753.520198] CPU: 1 PID: 1787 Comm: sysctl Not tainted 4.14.0-rc4-debug-00822-g07ec06ef312a #316
[  753.538598] Call Trace:
[  753.541449]  dump_stack+0xb3/0x140
[  753.545372]  ? _atomic_dec_and_lock+0x2b0/0x2b0
[  753.550564]  ? show_regs_print_info+0x6d/0x6d
[  753.555562]  print_address_description+0x25f/0x440
[  753.561044]  ? strscpy+0x807/0x970
[  753.564967]  kasan_report+0x1b8/0x430
[  753.569182]  ? strscpy+0x807/0x970
[  753.573107]  strscpy+0x807/0x970
[  753.576838]  ? strncpy+0xf0/0xf0
[  753.580576]  ? sched_clock_cpu+0x18/0x290
[  753.585185]  seccomp_actions_logged_handler+0x384/0x8d0
[  753.591157]  ? seccomp_send_sigsys+0xa0/0xa0
[  753.596046]  ? lock_downgrade+0x740/0x740
[  753.600648]  ? get_lock_stats+0x18/0x160
[  753.605159]  ? sysctl_head_grab+0x1b/0xc0
[  753.609766]  ? lock_contended+0x1130/0x1130
[  753.614566]  ? lock_release+0x14e0/0x14e0
[  753.619171]  ? do_raw_spin_trylock+0x1c0/0x1c0
[  753.624261]  ? lock_downgrade+0x660/0x740
[  753.628858]  ? do_raw_spin_lock+0x1f0/0x1f0
[  753.633657]  proc_sys_call_handler+0x216/0x3c0
[  753.638752]  ? proc_sys_permission+0x250/0x250
[  753.643847]  ? do_mmap+0x622/0xf50
[  753.647772]  __vfs_read+0xea/0x9a0
[  753.651696]  ? vfs_copy_file_range+0xc60/0xc60
[  753.656787]  ? security_mmap_file+0x66/0x1d0
[  753.661686]  ? common_file_perm+0x1e0/0x6b0
[  753.666485]  ? vm_mmap_pgoff+0x1b6/0x250
[  753.670993]  ? fsnotify+0x1c80/0x1c80
[  753.675209]  ? apparmor_task_setrlimit+0x690/0x690
[  753.680687]  ? vma_is_stack_for_current+0xb0/0xb0
[  753.686073]  vfs_read+0x109/0x3e0
[  753.689900]  SyS_read+0xcc/0x1d0
[  753.693628]  ? kernel_write+0x1a0/0x1a0
[  753.698034]  ? trace_hardirqs_on_caller+0x3cb/0x680
[  753.703610]  ? trace_hardirqs_on_thunk+0x1a/0x1c
[  753.708898]  entry_SYSCALL_64_fastpath+0x1c/0xb1
[  753.714180] RIP: 0033:0x7f62c82296b0
[  753.718294] RSP: 002b:00007ffc2d6726c8 EFLAGS: 00000246 ORIG_RAX: 0000000000000000
[  753.726919] RAX: ffffffffffffffda RBX: 0000000000000046 RCX: 00007f62c82296b0
[  753.735028] RDX: 0000000000000400 RSI: 00007f62c8722000 RDI: 0000000000000006
[  753.743138] RBP: 0000000000000000 R08: 00000000ffffffff R09: 0000000000000000
[  753.751246] R10: 0000000000000022 R11: 0000000000000246 R12: 0000000000000000
[  753.759355] R13: 0000000001d63650 R14: 00007f62c87186a0 R15: 0000000000000028
[  753.767478] 
[  753.769256] The buggy address belongs to the variable:
[  753.775125]  kprobes_seq_ops+0x1268/0x15c0
[  753.779820] 
[  753.781595] Memory state around the buggy address:
[  753.787075]  ffffffffa96b9680: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
[  753.795301]  ffffffffa96b9700: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
[  753.803526] >ffffffffa96b9780: 07 fa fa fa fa fa fa fa 00 05 fa fa fa fa fa fa
[  753.811751]                                               ^
[  753.818105]  ffffffffa96b9800: 02 fa fa fa fa fa fa fa 00 00 04 fa fa fa fa fa
[  753.826330]  ffffffffa96b9880: 00 00 00 01 fa fa fa fa 00 00 00 00 01 fa fa fa
[  753.834554] ==================================================================
#2 lore raw
Quoted 10 lines · Tyler Hicks
Quoted 1 line · Jakub Kicinski

I'm hitting this on sysctl -a with net-next (4.14-rc4).

Hey Jakub - thanks for the bug report!

Quoted 3 lines

I saw that seccomp_actions_logged_handler was introduced not-so-long-ago by Tyler, is there a fix for this?

No, this is the first I've heard of it. I'll have a look.

Thanks! :)

#3 lore raw
Quoted 7 lines · Jakub Kicinski
Quoted 6 lines · Tyler Hicks
Quoted 1 line · Jakub Kicinski

I'm hitting this on sysctl -a with net-next (4.14-rc4).

Hey Jakub - thanks for the bug report!

Quoted 1 line
Quoted 6 lines
Quoted 4 lines
Quoted 2 lines

I saw that seccomp_actions_logged_handler was introduced not-so-long-ago by Tyler, is there a fix for this?

No, this is the first I've heard of it. I'll have a look.

Thanks! :)

Hint: start looking at this thread. https://lkml.org/lkml/2017/7/18/874

Summary: strscpy and KASAN are currently incompatible. strscpy does a 64 bit speculative fetch on a char pointer (for efficiency reasons). KASAN spots this and flags an error.

A message that is not in the archive (sent off-list or not yet received): CAEeGbKOtffFD9Fb3qRc-8DkkQsAhuVO3X3xuzFtxH2p1uT69RQ@mail.gmail.com

#5 lore raw
Quoted 5 lines · Simon Brewer

Hint start looking at this thread. https://lkml.org/lkml/2017/7/18/874

Summary: strscpy and KASAN are currently incompatible. strscpy does a 64 bit speculative fetch on a char pointer (for efficiency reasons). KASAN spots this and flags an error.

Thanks, Simon. I had already reviewed the loop in seccomp_names_from_actions_logged() and couldn't spot an issue so my next step was to take a look at strscpy() itself. Your reply was well timed. :)

@Kees, this is a false positive. I picked strscpy() because of its sane return codes for easy error handling but its word-at-a-time complexity is overkill for this sysctl. Are you alright with this KASAN false positive or would you like me to change over to strlcpy()?

Tyler

Quoted 19 lines

On 11 October 2017 at 12:46, Jakub Kicinski <[email protected] <mailto:[email protected]>> wrote:

    On Tue, 10 Oct 2017 21:44:01 -0400, Tyler Hicks wrote:
    > On 10/10/2017 09:28 PM, Jakub Kicinski wrote:
    > > I'm hitting this on sysctl -a with net-next (4.14-rc4).
    >
    > Hey Jakub - thanks for the bug report!
    >
    > >
    > > I saw that seccomp_actions_logged_handler was introduced
    > > not-so-long-ago by Tyler, is there a fix for this?
    >
    > No, this is the first I've heard of it. I'll have a look.

    Thanks! :)
#6 lore raw
Quoted 12 lines · Tyler Hicks
Quoted 5 lines · Simon Brewer

Hint start looking at this thread. https://lkml.org/lkml/2017/7/18/874

Summary: strscpy and KASAN are currently incompatible. strscpy does a 64 bit speculative fetch on a char pointer (for efficiency reasons). KASAN spots this and flags an error.

Thanks, Simon. I had already reviewed the loop in seccomp_names_from_actions_logged() and couldn't spot an issue so my next step was to take a look at strscpy() itself. Your reply was well timed. :)

Quoted 4 lines

@Kees, this is a false positive. I picked strscpy() because of its sane return codes for easy error handling but its word-at-a-time complexity is overkill for this sysctl. Are you alright with this KASAN false positive or would you like me to change over to strlcpy()?

I tend to prefer strscpy. I don't think we need a change here. Thanks!

-Kees

Signature · 2 lines
Kees Cook
Pixel Security