Re: [rms@gnu.org: Re: [twurgler@goodyear.com: emacs-21.2.90 on HP 11.0]]

[Stef Van Vlierberghe <vvl@cfmu.eurocontrol.be>]

Tom Wurgler writes:
 > 
 > Recently Richard Stallman <rms@gnu.org> wrote:
 > 
 > > Date: Thu, 18 Jul 2002 08:55:08 -0600 (MDT)
 > > From: Richard Stallman <rms@gnu.org>
 > > CC: twurgler@goodyear.com, philippe.waroquiers@eurocontrol.int,
 > >         stef.van-vlierberghe@eurocontrol.int, emacs-devel@gnu.org,
 > >         bruce@perens.com
 > > Reply-to: rms@gnu.org
 > > 
 > >     If *both* patches are applied and POSIX_SIGNALS is defined,
 > >     all the problems above are disappearing.
 > >     In other words:
 > >        (while t)       is interrupted by C-g
 > >        (sleep-for 3)   really sleeps 3 seconds
 > >        abort running job in shell works (both with C-c C-c and with Signal ->
 > >     Break)
 > > 
 > > That is good.  Thanks.
 > > 
 > > 
 > 
 > I have that the all the problems I reported are fixed with these patches (the
 > BROKEN_SA_RESTART in sysdep.c and hpux11.h, and the process.c patch), but I
 > don't find that the (while t) is interruptable.  Did I miss a fix here?
 > 
 > thanks
 > 
 > tom

I think the (while t) problem was caused by the SA_RESTART.

If you use tusc (tusc.7.2.shar from HP devresource site) to look at
the system calls being made with emacs is hanging, you should see a 
repetition of select() calls being continuously interrupted by SIGALARMs. 

Use :

tusc -R -v -o /tmp/emacs.tusc emacs

That will show you the system call restarts (-R)
The options used when installing signal handlers (-v)
will put the output in /tmp/emacs.tusc (-o)

Good luck.
-- 
Stef Van Vlierberghe                   Eurocontrol - CFMU room 20115
stef.van-vlierberghe@eurocontrol.int   Raketstraat 96
Tel: +32 2 729 97 32                   B-1130 BRUSSELS
Fax: +32 2 729 90 22                   Belgium