emacs-bug-tracker
[Top][All Lists]
Advanced

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

[Emacs-bug-tracker] bug#6170: closed (24.0.50; Compiling on solaris2.10


From: GNU bug Tracking System
Subject: [Emacs-bug-tracker] bug#6170: closed (24.0.50; Compiling on solaris2.10 with gcc doesn't define alloca)
Date: Wed, 02 Jun 2010 09:25:02 +0000

Your message dated Wed, 02 Jun 2010 05:24:36 -0400
with message-id <address@hidden>
and subject line Re: bug#6170: 24.0.50; Compiling on solaris2.10 with gcc 
doesn't define alloca
has caused the GNU bug report #6170,
regarding 24.0.50; Compiling on solaris2.10 with gcc doesn't define alloca
to be marked as done.

(If you believe you have received this mail in error, please contact
address@hidden)


-- 
6170: http://debbugs.gnu.org/cgi/bugreport.cgi?bug=6170
GNU Bug Tracking System
Contact address@hidden with problems
--- Begin Message --- Subject: 24.0.50; Compiling on solaris2.10 with gcc doesn't define alloca Date: Tue, 11 May 2010 11:57:17 +0100
On this system, <stdlib.h> is provided by Sun and therefore
doesn't define alloca, unlike on a typical GNU/linux system where
<stdlib.h> contains the following:

| #if defined __USE_GNU || defined __USE_BSD || defined __USE_MISC
| # include <alloca.h>
| #endif /* Use GNU, BSD, or misc.  */

When compiling emacs with gcc, alloca is therefore undefined.
The culprit is this snippet in configure.in:

| #ifndef __GNUC__
| # ifdef HAVE_ALLOCA_H
| #  include <alloca.h>
| # else /* AIX files deal with #pragma.  */
| #  ifndef alloca /* predefined by HP cc +Olibcalls */
| char *alloca ();
| #  endif
| # endif /* HAVE_ALLOCA_H */
| #endif /* __GNUC__ */

On this system, configure correctly defines both HAVE_ALLOCA_H
and HAVE_ALLOCA, but since the inclusion is guarded by __GNUC__
not being defined, we never include <alloca.h>.

I think the correct fix is to unconditionally define alloca if
__GNUC__ is detected, see patch below.  Note that this is closer
to how autoconf checks for the presence of the alloca function.

Possible changelog entry:

2010-05-11  Lawrence Mitchell  <address@hidden>

        * configure.in: Unconditionally define alloca if __GNUC__ is
        detected.

diff --git a/configure.in b/configure.in
index 8a7d9be..98b6abb 100644
--- a/configure.in
+++ b/configure.in
@@ -3360,6 +3360,9 @@ extern char *getenv ();
 char *alloca ();
 #  endif
 # endif /* HAVE_ALLOCA_H */
+#else
+# undef alloca
+# define alloca __builtin_alloca
 #endif /* __GNUC__ */
 #ifndef HAVE_SIZE_T
 typedef unsigned size_t;



--- End Message ---
--- Begin Message --- Subject: Re: bug#6170: 24.0.50; Compiling on solaris2.10 with gcc doesn't define alloca Date: Wed, 02 Jun 2010 05:24:36 -0400 User-agent: Gnus/5.11 (Gnus v5.11) Emacs/22.1 (gnu/linux)
Lawrence Mitchell <address@hidden> writes:

> Dan Nicolaescu wrote:
>> Lawrence Mitchell <address@hidden> writes:
>
>>> On this system, <stdlib.h> is provided by Sun and therefore
>>> doesn't define alloca, unlike on a typical GNU/linux system where
>>> <stdlib.h> contains the following:
>
>>> | #if defined __USE_GNU || defined __USE_BSD || defined __USE_MISC
>>> | # include <alloca.h>
>>> | #endif /* Use GNU, BSD, or misc.  */
>
>>> When compiling emacs with gcc, alloca is therefore undefined.
>>> The culprit is this snippet in configure.in:
>
>>> | #ifndef __GNUC__
>>> | # ifdef HAVE_ALLOCA_H
>>> | #  include <alloca.h>
>>> | # else /* AIX files deal with #pragma.  */
>>> | #  ifndef alloca /* predefined by HP cc +Olibcalls */
>>> | char *alloca ();
>>> | #  endif
>>> | # endif /* HAVE_ALLOCA_H */
>>> | #endif /* __GNUC__ */
>
>
>> "info autoconf" says that this is the proper way to do it:
>
>>           #ifdef HAVE_ALLOCA_H
>>           # include <alloca.h>
>>           #elif defined __GNUC__
>>           # define alloca __builtin_alloca
>>           #elif defined _AIX
>>           # define alloca __alloca
>>           #elif defined _MSC_VER
>>           # include <malloc.h>
>>           # define alloca _alloca
>>           #else
>>           # include <stddef.h>
>>           # ifdef  __cplusplus
>>           extern "C"
>>           # endif
>>           void *alloca (size_t);
>>           #endif
>
>> Not sure we need the last #else part, or the _MSC_VER part...
>
> Is there any movement on getting this fix, or something like it,
> installed?  It doesn't seem like a controversial change.

Should be fixed now.


--- End Message ---

reply via email to

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