httpd-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Greg Stein <gst...@lyra.org>
Subject Re: Eliminating NO_USE_SIGACTION
Date Sat, 08 Apr 2000 11:57:07 GMT
Rather than defining a *second* symbol (NO_USE_SIGACTION), the
HAVE_SIGACTION symbol should simply be disabled. Seems a bit ridiculous to
define two of them.

Cheers,
-g

On Wed, 5 Apr 2000, Jim Jagielski wrote:
> NO_USE_SIGACTION is for those platforms that have sigaction() but
> it shouldn't be used, because it's broken or unreliable. It's
> short for HAVE_SIGACTION_BUT_DONT_USE_IT :)
> 
> Manoj Kasichainula wrote:
> > 
> > Speaking of ap_signal, is there any reason to maintain
> > NO_USE_SIGACTION anymore? Autoconf can detect the presence of
> > sigaction(), so we should do so and use HAVE_SIGACTION instead. Also,
> > I'd like to eliminate even bothering to check for ap_signal in
> > mpmt_pthread and dexter. Since sigaction is a POSIX call, and pthreads
> > imply proper POSIX support, I think it's unnecessary for MPMs using
> > pthreads to bother supporting non-sigaction systems.
> 
> 
> -- 
> ===========================================================================
>    Jim Jagielski   [|]   jim@jaguNET.com   [|]   http://www.jaguNET.com/
>                 "Are you suggesting coconuts migrate??"

-- 
Greg Stein, http://www.lyra.org/



Mime
View raw message