Re: [PATCH 00/15] Hurd signals: please review

[Samuel Thibault <samuel dot thibault at gnu dot org>]

Hello,

Jeremie Koenig, le Wed 29 Jun 2011 18:30:12 +0200, a écrit :
> I believe my signal-related series of patches for Hurd are mature enough
> to be reviewed and considered for inclusion.

This looks quite good indeed.  Roland, I'd like to apply the patch
series to the Debian package to get more thorough testing.  Do you agree
on the interface below, so we will not have ABI changes?

> * #define SA_SIGINFO 0x40
> 
>   This follows the current approach of using BSD values for
>   <bits/sigaction.h> constants.
> 
> * void _hurd_sigstate_set_global_rcv (struct hurd_sigstate *ss);
> 
>   Used by libpthread to mark a thread as a global receiver, meaning the
>   thread should receive global signal it doesn't block, and use the
>   global action vector instead of a thread-specific one.
> 
> * void _hurd_sigstate_lock (struct hurd_sigstate *ss);
> * sigset_t _hurd_sigstate_pending (const struct hurd_sigstate *ss);
> * void _hurd_sigstate_unlock (struct hurd_sigstate *ss);
> 
>   Used to access a thread's set of pending signals, including global
>   signals if appropriate.
> 
> * void _hurd_sigstate_delete (thread_t thread);
> 
>   Delete the sigstate structure associated to THREAD. We need this to avoid
>   attempting the delivery of global signals to terminated threads.

Samuel