>
> The new process image also inherits the following attributes from
> the calling process image:
>
> [...]
>
> (12) Process signal mask (see 3.3.5)
>
> (13) Pending signals (see 3.3.6)
>
> [...]
>
> So either part of your patch contravenes POSIX, I'm afraid.
I don't have POSIX at hand, but does it not also specify exec() from a
signal handler is OK? I am pretty sure it does. In this case, the new
image has the "current" signal blocked. Is this proper? It seems obtuse that
a process would have to do sigemptyset() and sigprocmask() before exec().
comments? should we perhaps clear the 'current' signal from the mask, or
is that just overinterpreting?
Tim
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu
Please read the FAQ at http://www.tux.org/lkml/
This archive was generated by hypermail 2b29 : Fri Jul 07 2000 - 21:00:17 EST