Re: [PATCH] x86/ptrace: Remove questionable TS_COMPAT usage in ptrace

[Andy Lutomirski]

On Sun, Jun 19, 2016 at 2:19 PM, Oleg Nesterov <oleg@xxxxxxxxxx> wrote:
> Let me first thank Pedro who has already replied!
>
> And I have to admit I will need to re-read his explanations after
> sleep to (try to) convince myself I fully understans the problems ;)
> Too late for me.
>
> Right now I have nothing to add, but
>
> On 06/18, Andy Lutomirski wrote:
>>
>> @@ -922,16 +922,7 @@ static int putreg32(struct task_struct *child, unsigned regno, u32 value)
>>       R32(esp, sp);
>>
>>       case offsetof(struct user32, regs.orig_eax):
>> -             /*
>> -              * A 32-bit debugger setting orig_eax means to restore
>> -              * the state of the task restarting a 32-bit syscall.
>> -              * Make sure we interpret the -ERESTART* codes correctly
>> -              * in case the task is not actually still sitting at the
>> -              * exit from a 32-bit syscall with TS_COMPAT still set.
>> -              */
>>               regs->orig_ax = value;
>> -             if (syscall_get_nr(child, regs) >= 0)
>> -                     task_thread_info(child)->status |= TS_COMPAT;
>
> I agree it would be nice to remove this code, but then it is not clear
> how/when we should sign-extend regs->ax..
>
> And this leads to another question, why do we actually need to set/clear
> TS_COMPAT in set_personality_ia32() ??

Something's clearly buggy there, considering that
set_personality_64bit() does *not* clear it.

--Andy