GNU bug report logs - #15739
Doesn't support clang as well as it could.

Previous Next

To add a comment to this bug, you must first unarchive it, by sending
a message to control AT debbugs.gnu.org, with unarchive 15739 in the body.
You can then email your comments to 15739 AT debbugs.gnu.org in the normal way.

Toggle the display of automated, internal messages from the tracker.

Message #5 received at submit <at> debbugs.gnu.org (full text, mbox):

From: Matt Sicker <boards <at> gmail.com>
To: bug-guile <at> gnu.org
Subject: Doesn't support clang as well as it could.
Date: Sun, 27 Oct 2013 14:53:20 -0500

[Message part 1 (text/plain, inline)]

There are plenty of macros that only check for GNU GCC and don't bother
checking support from other popular compilers. For example, in
libguile/__scm.h, the macros SCM_NORETURN, SCM_UNUSED, SCM_EXPECT, and many
more, all only check for GCC support. Most (if not all) of these language
extensions are supported by Clang, and usually using the same exact syntax!

I'll submit a patch that adds Clang support to these macros. See <
http://clang.llvm.org/docs/LanguageExtensions.html> for the documentation
about this.

-- 
Matt Sicker <boards <at> gmail.com>

[Message part 2 (text/html, inline)]

Message #12 received at 15739-done <at> debbugs.gnu.org (full text, mbox):

From: Mark H Weaver <mhw <at> netris.org>
To: Matt Sicker <boards <at> gmail.com>
Cc: 15739-done <at> debbugs.gnu.org
Subject: Re: bug#15739: Doesn't support clang as well as it could.
Date: Fri, 01 Nov 2013 18:00:23 -0400

Hi Matt,

Matt Sicker <boards <at> gmail.com> writes:
> I noticed that clang reports to be GCC version 4.2, so most of what I
> thought might be wrong wasn't. Therefore, the patch was far easier and
> smaller than I expected. This isn't too high a priority, but it does
> add support for clang where the code would normally check for GCC
> 4.3+. Clang even supports C++ static_assert supposedly, so there's a
> check for that as well.
>
> -- 
> Matt Sicker <boards <at> gmail.com> 
>
>
> From 2b02509f53a247784b4bb0be96061e46b53b0c4e Mon Sep 17 00:00:00 2001
> From: Matt Sicker <boards <at> gmail.com>
> Date: Sun, 27 Oct 2013 16:54:06 -0500
> Subject: [PATCH] Updated a couple compiler checks to support clang.
>
> * lib/poll.c [__clang__]: Same functionality as GCC 4.3+ for ignoring
>   the (nfd < 0) tests.
> * lib/verify.h [__has_feature(c_static_assert)]: Enables
>   _GL_HAVE__STATIC_ASSERT when clang feature is available for C.
> * lib/verify.h [__has_feature(cxx_static_assert)]: Enables
> * _GL_HAVE_STATIC_ASSERT when clang feature is available for C++.

Both of these files are part of gnulib.  As a matter of policy, we do
not change those files in our repository.  Instead we update from
gnulib's repository from time to time.

Can you please send this problem report and patch to
<bug-gnulib <at> gnu.org> ?

I'm closing this bug, since there's nothing for us to do in Guile.  If
you find any clang-related problems in Guile itself, please let us know.

     Thanks,
       Mark

Message #20 received at 15739 <at> debbugs.gnu.org (full text, mbox):

From: ludo <at> gnu.org (Ludovic Courtès)
To: Matt Sicker <boards <at> gmail.com>
Cc: 15739 <at> debbugs.gnu.org
Subject: Re: bug#15739: Doesn't support clang as well as it could.
Date: Thu, 28 Nov 2013 20:35:37 +0100

Hi,

Matt Sicker <boards <at> gmail.com> skribis:

> There are plenty of macros that only check for GNU GCC and don't bother
> checking support from other popular compilers. For example, in
> libguile/__scm.h, the macros SCM_NORETURN, SCM_UNUSED, SCM_EXPECT, and many
> more, all only check for GCC support. Most (if not all) of these language
> extensions are supported by Clang, and usually using the same exact syntax!

Doesn’t Clang still define __GNUC__?  If it does, then it has to behave
like GCC, and thus nothing Clang-specific needs to be added.

Ludo’.

Message #23 received at 15739 <at> debbugs.gnu.org (full text, mbox):

From: Matt Sicker <boards <at> gmail.com>
To: Ludovic Courtès <ludo <at> gnu.org>
Cc: 15739 <at> debbugs.gnu.org
Subject: Re: bug#15739: Doesn't support clang as well as it could.
Date: Sat, 30 Nov 2013 19:30:57 -0600

[Message part 1 (text/plain, inline)]

On 28 November 2013 13:35, Ludovic Courtès <ludo <at> gnu.org> wrote:
>
> Doesn’t Clang still define __GNUC__?  If it does, then it has to behave
> like GCC, and thus nothing Clang-specific needs to be added.
>

Yeah, turns out it's a different issue.

-- 
Matt Sicker <boards <at> gmail.com>

[Message part 2 (text/html, inline)]

Message #26 received at 15739-done <at> debbugs.gnu.org (full text, mbox):

From: ludo <at> gnu.org (Ludovic Courtès)
To: Matt Sicker <boards <at> gmail.com>
Cc: 15739-done <at> debbugs.gnu.org
Subject: Re: bug#15739: Doesn't support clang as well as it could.
Date: Sun, 01 Dec 2013 23:09:20 +0100

Matt Sicker <boards <at> gmail.com> skribis:

> On 28 November 2013 13:35, Ludovic Courtès <ludo <at> gnu.org> wrote:
>>
>> Doesn’t Clang still define __GNUC__?  If it does, then it has to behave
>> like GCC, and thus nothing Clang-specific needs to be added.
>>
>
> Yeah, turns out it's a different issue.

Thanks for letting us know.  I’m closing the bug now.

Ludo’.

Previous Next

GNU bug tracking system
Copyright (C) 1999 Darren O. Benham, 1997,2003 nCipher Corporation Ltd, 1994-97 Ian Jackson.