Re: incorrect caching messages

[Ralf Wildenhues]

Hello Eric,

* Eric Blake wrote on Wed, Nov 15, 2006 at 06:02:00PM CET:
> 
> checking for intmax_t... (cached) yes
> (cached) (cached) checking whether to enable assertions... no
> 
> I tracked both of those spurious messages to this.  gl_ALLOCSA was calling 
> gl_FUNC_ALLOCA rather than AC_REQUIRE'ing it, so gl_FUNC_ALLOCA got expanded 
> twice in a row since the use of the allocsa module also pulls in the 
> invocation 
> of gl_FUNC_ALLOCA from the alloca-opt module.  Then gl_FUNC_ALLOCA was using 
> a 
> cache variable, but without any useful messages.

The alloca.m4 change fixes the bug I reported.  The allocsa.m4 change
contradicts
http://lists.gnu.org/archive/html/bug-gnulib/2006-10/msg00279.html

Cheers,
Ralf

> 2006-11-15  Eric Blake  <address@hidden>
> 
>       * m4/alloca.m4 (gl_FUNC_ALLOCA): Use AC_CACHE_CHECK to avoid a
>       random "(cached)" in configure output.
>       * m4/allocsa.m4 (gl_ALLOCSA): Don't invoke macro already picked
>       up by the module dependency.