Re: [PATCH 06/14] x86/ptrace: run seccomp after ptrace

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

 



On Thu, Jun 9, 2016 at 3:52 PM, Andy Lutomirski <luto@xxxxxxxxxxxxxx> wrote:
> On Thu, Jun 9, 2016 at 2:01 PM, Kees Cook <keescook@xxxxxxxxxxxx> wrote:
>> This moves seccomp after ptrace on x86 to that seccomp can catch changes
>> made by ptrace. Emulation should skip the rest of processing too.
>>
>> We can get rid of test_thread_flag because there's no longer any
>> opportunity for seccomp to mess with ptrace state before invoking
>> ptrace.
>>
>> Suggested-by: Andy Lutomirski <luto@xxxxxxxxxx>
>> Signed-off-by: Kees Cook <keescook@xxxxxxxxxxxx>
>> Cc: x86@xxxxxxxxxx
>> Cc: Andy Lutomirski <luto@xxxxxxxxxx>
>> ---
>>  arch/x86/entry/common.c | 22 ++++++++++++----------
>>  1 file changed, 12 insertions(+), 10 deletions(-)
>>
>> diff --git a/arch/x86/entry/common.c b/arch/x86/entry/common.c
>> index df56ca394877..81c0e12d831c 100644
>> --- a/arch/x86/entry/common.c
>> +++ b/arch/x86/entry/common.c
>> @@ -73,6 +73,7 @@ static long syscall_trace_enter(struct pt_regs *regs)
>>
>>         struct thread_info *ti = pt_regs_to_thread_info(regs);
>>         unsigned long ret = 0;
>> +       bool emulated = false;
>>         u32 work;
>>
>>         if (IS_ENABLED(CONFIG_DEBUG_ENTRY))
>> @@ -80,11 +81,19 @@ static long syscall_trace_enter(struct pt_regs *regs)
>>
>>         work = ACCESS_ONCE(ti->flags) & _TIF_WORK_SYSCALL_ENTRY;
>>
>> +       if (unlikely(work & _TIF_SYSCALL_EMU))
>> +               emulated = true;
>> +
>> +       if ((emulated || (work & _TIF_SYSCALL_TRACE)) &&
>> +           tracehook_report_syscall_entry(regs))
>> +               return -1L;
>> +
>> +       if (emulated)
>> +               return -1L;
>> +
>
> I think that this code will result in ptrace-induced skips calling the
> audit exit hook but not the audit entry hook.  I don't know whether
> this is a problem.  It's also worth making sure that ptracing a
> seccomp-skipped syscall calls the exit hook with the right regs.
>
> I suspect it's fine, but I want to think about it a little bit more.

I don't think this is true, since all architectures already needed to
handle an immediate return from seccomp, so audit shouldn't be touched
on the exit path either.

-Kees

-- 
Kees Cook
Chrome OS & Brillo Security




[Index of Archives]     [Linux MIPS Home]     [LKML Archive]     [Linux ARM Kernel]     [Linux ARM]     [Linux]     [Git]     [Yosemite News]     [Linux SCSI]     [Linux Hams]

  Powered by Linux