Re: [RFC PATCH] ptrace: don't report syscall-exit if the tracee was killed by seccomp

[Oleg Nesterov]

On 04/04, Oleg Nesterov wrote:
>
> On 04/03, Kusaram Devineni wrote:
> >
> OK, this is another (although related) issue, lets discuss it separately.

Yes...

> > For that specific path, one approach that seems to work is making signalfd
> > exclude
> > SA_IMMUTABLE signals from the mask it passes to
>
> Perhaps... But this is nasty.

OK, lets do it. I'll send the patch in a minute. It was already tested by syzbot.
Sorry, forgot to CC you, see https://lore.kernel.org/all/adJvw9gEC9D1Gxtq@xxxxxxxxxx/

I don't see a better fix for now. Hopefully we can cleanup this later.
I think force_exit_sig / force_sig_seccomp should make fatal_signal_pending()
true, but we need a simple fix...

Oleg.