Re: Q: user_enable_single_step() && update_debugctlmsr()

[Oleg Nesterov <oleg@redhat.com>]

On 08/01, Sebastian Andrzej Siewior wrote:
>
> On 08/01/2012 04:01 PM, Oleg Nesterov wrote:
>> On 08/01, Sebastian Andrzej Siewior wrote:
>>>
>>> On 08/01/2012 03:46 PM, Oleg Nesterov wrote:
>>>
>>>>>> But, worse, isn't it wrong? Suppose that debugger switches to
>>>>>> another TIF_SINGLESTEP&&    !TIF_BLOCKSTEP task, in this case
>>>>>> we "leak" DEBUGCTLMSR_BTF, no?
>>>>>
>>>>> __switch_to_xtra() should notice the difference in the TIF_BLOCKSTEP
>>>>> flag and disable it.
>>>>
>>>> And how it can notice the difference if there is no difference?
>>>>
>>>> (unless, of course debugger is TIF_BLOCKSTEP'ed).
>>>
>>> Yes. enable_step() sets DEBUGCTLMSR_BTF along with TIF_BLOCKSTEP.
>>> kprobes checks the same flag before touching DEBUGCTLMSR_BTF.
>>
>> It seems that you replied to the wrong email or I am confused ;)
>
> No I think I replied to the correct one :)
> enable_step() is the only place for ptrace/debugger which is touching
> DEBUGCTLMSR_BTF. It always sets DEBUGCTLMSR_BTF and TIF_BLOCKSTEP in
> sync so why should they both end up different? And once
> __switch_to_extra() notices that TIF_BLOCKSTEP from the previous task
> is different from the next task is different, then the CPU flag has
> to be changed.

OK, I was confuse by "kprobes" above.

And I think you missed my point. I'll try again.

We have the GDB process and the (stopped) tracee T. And we have
another task X which have TIF_SINGLESTEP but not TIF_BLOCKSTEP.
To simplify, suppose that X is already TASK_RUNNING but not on rq.

GDB does ptrace(PTRACE_SINGLEBLOCK, T). This sets X->TIF_BLOCKSTEP.
Now suppose that GDB is preempted right after it does
update_debugctlmsr(), and the scheduler choses X as the next task.

Both GDB and X do not have TIF_BLOCKSTEP, so __switch_to_extra()
does not update DEBUGCTLMSR_BTF.

X returns to the user-mode with TIF_SINGLESTEP and TIF_BLOCKSTEP,
the latter is wrong.

No?

Oleg.