Re: [PATCH 1/4] signalfd: add ability to return siginfo in a raw format

[Oleg Nesterov <oleg@redhat.com>]

On 12/25, Andrey Wagin wrote:
>
> 2012/12/24 Oleg Nesterov <oleg@redhat.com>:
> > On 12/24, Andrey Vagin wrote:
> >>
> >> signalfd should be called with the flag SFD_RAW for that.
> >>
> >> signalfd_siginfo is not full for siginfo with a negative si_code.
> >> copy_siginfo_to_user() is copied a full siginfo to user-space, if
> >> si_code is negative.  signalfd_copyinfo() doesn't do that and can't be
> >> expanded, because it has not compatiable format with siginfo_t.
> >
> > Yes, but otoh perhaps we should change (fix) signalfd_siginfo/copyinfo,
> > its "default" case makes no sense if si_code < 0.
>
> Its "default" case makes sense if a signal is sent by sigqueue(pid,sig,ptr).

But it doesn't really work, that is what I was trying to say. And that
is why you want copy_siginfo_to_user ;)

> I'm afraid, we can change (fix) signalfd_copyinfo, because for
> negative si_code a whole siginfo should be copied to userspace.

Exactly, this is what I meant. We simply do not know what this info
contains if it was sent by sigqueueinfo().

> Currently if si_code is unknown, signalfd_copyinfo sets only ssi_ptr
> and that can't be changed due to backward compatibility. ssi_ptr is in
> the midle of signalfd_siginfo and a sizeof(signalfd_siginfo) is equal
> to sizeof(siginfo_t). We don't have space to copied siginfo into
> signalfd_siginfo.

Yes, I understand.

> If we want to have another format with SFD_RAW, I prefer to have
> siginfo_t instead of signalfd_siginfo. Because if si_code is negative,
> it should be siginfo_t in any case. A minor thing is that it can be
> sent back without modifications.

"without modifications" is not actually true, your patch changes the
meaning of ->si_code. Yes, I understand why do you do this, and I am
not going to argue. But it looks a bit sad that, say, sigtimedwait()
and read(SFD_RAW) will return the "same" siginfo_t except the subtle
difference in ->si_code.

What I am trying to say, is that SFD_RAW should be named
SFD_signalfd_siginfo_SUCKS_BUT_WE_CANT_CHANGE_IT_FOR_COMPATIBILITY ;) So
you need another format. And if we add another format we should think
twice. For example, if it is _RAW, why we can't simply do memcpy()
instead of copy_siginfo_to_user() ? Not that I really suggest this.

And if we change the meaning of ->si_code then, perhaps, we should
think what else we should change to avoid SFD_RAW_RAW in future.
Just for example, we can set MSB in ->si_signo if the signal was
private.

Oleg.