Re: bogus configure output for strstr

[Eric Blake]

On 08/30/2010 05:20 PM, Bruno Haible wrote:
> > static void quit (int sig) { exit (sig + 128); }
> > ...
> >       signal (SIGALRM, quit);

I tested this on FreeBSD, and it indeed works.  And given that sleep.m4 
already installs a handler, it's no less portable than what we are 
already doing in other tests.

> If you write it like this, it will be a no-op (because the shell does
> not see the signal that killed the process, it sees the exit code of the
> process). You need to write
>
>    static void quit (int sig) { exit (1); }

Huh?  It won't be a signal, but it WILL be a valid exit code, and it may 
make post-mortem analysis easier (that is, most shells set $? to 
signal+128 when they detect exit via a signal; we've lost the detection 
of exit from a signal, but having a $? that matches in the config.log 
output may still be handy).  Exit codes do not need to be between 1 and 
127 in order to be non-zero.

> Affected files: strstr.m4, strcasestr.m4, memmem.m4.

Yep, I came to the same conclusion.  Pushing my patch soon.

--
Eric Blake   address@hidden    +1-801-349-2682
Libvirt virtualization library http://libvirt.org