Re: [PATCH v3] arm64: ptrace: don't warn on missing siginfo in a group-stop

[Oleg Nesterov <oleg@redhat.com>]

On 11/09, Zhao Gongyi wrote:
>
> On 11/9/26 2:59 AM, Oleg Nesterov wrote:
> >
> > ----------------------------------------------------------------------------
> > But can't resist... The old comment was obviously wrong, the new one doesn't
> > look clear to me. It correctly explains why info can be NULL, but not why we
> > should not update orig_x0 in this case. May be it should just say that exit-stop
> > always comes with info != NULL?
> >
>
> You're right: the comment explains why info can be NULL, but not why we
> return. The early return is intentional: it keeps orig_x0 unchanged
> during a non-seized group-stop.

Yep, this is what I meant. And this looks confusing to me. Lets suppose that
the tracee was stopped because of JOBCTL_TRAP_STOP. Why should the behaviour
of update_syscall_orig_x0_after_ptrace() depend on whether that tracee is
PT_SEIZED or not?

IOW. To me your V2 looks more understandable to me, I don't understand the
Catalin's concerns.

But then again, I can't really argue, I know nothing about arch/arm64. And I
do remember that I tried to confuse you when I reviewed your V2 ;) Thanks again
for correcting me.

> I'll wait for Catalin and Will's thoughts on the wording.

Me too.

Oleg.