Date: Thu, 16 Jan 2003 08:03:37 +0200 (IST)
From: Eli Zaretskii <eliz AT is DOT elta DOT co DOT il>
X-Sender: eliz AT is
To: djgpp-workers AT delorie DOT com
Subject: RE: Re: your mail
In-Reply-To: <1042666298005458@lycos-europe.com>
Message-ID: <Pine.SUN.3.91.1030116080036.24436C-100000@is>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Reply-To: djgpp-workers AT delorie DOT com
Errors-To: nobody AT delorie DOT com
X-Mailing-List: djgpp-workers AT delorie DOT com
X-Unsubscribes-To: listserv AT delorie DOT com
Precedence: bulk


On Wed, 15 Jan 2003, Ben Decker wrote:

> The sigaltstack structure is defined as:
> struct sigaltstack {
>              char   *ss_sp;
>              int     ss_size;
>              int     ss_flags;
>      };
> (would this go in signal.h? ss_size may better be difined as size_t?)
> It is a bit confusing, because the structure is similar to stack_t (I notice DJGPP doesn't 
> have that either). The values seem normally defined as follows:
> ss_sp = malloc(SIGSTKSZ)
> ss_size = SIGSTKSZ;
> ss_flags = 0;
> SIGSTKSZ is "the number of bytes/chars that would be used to cover the usual case 
> when allocating an alternate stack area."
> 
> For Linux it was defined as:
> #define MINSIGSTKSZ     131027  /* min. stack size for sigaltstack() */
> #define SIGSTKSZ        262144  /* default stack size for sigaltstack() */

Please also look in the relevant Posix and OpenUnix standards, not only 
on GNU/Linux.  We would like to be Posix-compliant as much as possible.

> GNU platforms seem to store the sigaltstack() routine itself in a separate 'sigaltstack.c'. 
> Would this also be suitable for DJGPP?

Yes.  But please note that we generally prefer not to exceed the 8+3 
limitations of DOS file names.  So I'd suggest to consider something 
like `sigaltst.c' or maybe `sigastk.c'.

Thanks for working on this.