[Top][All Lists]

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: Bison 1.30f

From: Hans Aberg
Subject: Re: Bison 1.30f
Date: Wed, 12 Dec 2001 23:31:47 +0100

At 20:33 +0100 2001/12/12, Akim Demaille wrote:
>Hans> That Bison works the same way as in earlier versions,
>Hans> specifically, I had a brief look at the alloca test section in
>Hans> bison.simple and got the impression that if YYSTACK_USE_ALLOCA
>Hans> was undefined, the generated parser would attempt to use alloca,
>Hans> but if not possible, it would use malloc instead.
>Like it always did.  What's new is that you can decide for it.
>YYSTACK_USE_ALLOCA didn't exist as a user tunable symbol before.

When I do not define YYSTACK_USE_ALLOCA, I do not get alloca, but malloc; I
think this happens because my platform has not been included. So I was
under the illusion it had been changed, which it probably not has been.

The only change I would wish for is that the bison.simple alloca test
section starts
#  include <alloca.h>          /* add this */
#  define YYSTACK_ALLOC alloca
#  define YYSIZE_T size_t
because then one does not have to add #include <alloca.h> separatly. Is
there any platform with alloca in a header with a different name?

  Hans Aberg

reply via email to

[Prev in Thread] Current Thread [Next in Thread]