Re: [PATCH v2 0/1] MR2126: ntdll: Print an error if we receive a signal not on the signal stack.

[Zebediah Figura (＠zfigura)]

On Wed Feb  8 19:01:54 2023 +0000, **** wrote:
> Paul Gofman replied on the mailing list:
> On 2/8/23 12:55, Rémi Bernon (@rbernon) wrote:
> >
> > USR1 / USR2 / INT for instance could be blocked in non-Wine threads,
> raising them again until they reach a Wine thread.
> I think USR1 may only be received by a non-Wine thread if it was 
> originated by some native library and not by wineserver which sends that 
> only to threads it knows? In that case, not sure how that can be handled 
> better in a Wine thread?
> Also, is there a good way to block signals in threads which are created 
> by native libraries, I think we don't know anything about those threads?

Yes, what Paul said. If a non-Wine thread gets USR1 or USR2 it didn't come from us, and handing it to a Wine thread is not going to do the right thing.

Note that signal masks are per-thread, but according to my understanding signal *actions* aren't.

QUIT is supposed to kill the entire process by default, not just a thread, so just killing that thread doesn't seem right. I suppose we *could* exit() in that case, but QUIT also dumps core by default, so it doesn't seem quite right to do a clean exit either. At least not right enough to be worth the extra complexity.

Rethrowing INT to a Wine thread is... nontrivial, but possible, I guess? I'd rather leave it for a follow-up patch if it's really worth doing.