From: Alexey Kardashevskiy <aik@ozlabs.ru>
To: Peter Maydell <peter.maydell@linaro.org>
Cc: Blue Swirl <blauwirbel@gmail.com>,
David Gibson <david@gibson.dropbear.id.au>,
qemu-devel@nongnu.org, Stefan Hajnoczi <stefanha@redhat.com>,
Gerd Hoffmann <kraxel@redhat.com>
Subject: Re: [Qemu-devel] [PATCH] pixman: remove -Wredundand-decls
Date: Mon, 15 Apr 2013 17:45:02 +1000 [thread overview]
Message-ID: <516BAFFE.9020306@ozlabs.ru> (raw)
In-Reply-To: <CAFEAcA-x18vsMT9E2gm8yVr633e+zNRjd3rFrzG69Tpp==r4OA@mail.gmail.com>
On 04/15/2013 05:30 PM, Peter Maydell wrote:
> On 15 April 2013 08:26, Alexey Kardashevskiy <aik@ozlabs.ru> wrote:
>> On 04/15/2013 05:18 PM, Peter Maydell wrote:
>>>
>>> On 15 April 2013 03:31, Alexey Kardashevskiy <aik@ozlabs.ru> wrote:
>>>>
>>>> The assert.h header file from Fedora18 does not have #ifdef-#endif
>>>> brackets around __assertXXXX function so it cannot compile with
>>>> the -Wredundant-decls switch on.
>>>>
>>>> Some Linux distributions (such as Debian Wheezy) still do have those
>>>> brackets arounb __assertXXXX functions (#ifndef _ASSERT_H_DECLS) but
>>>> the version of assert.h on http://sourceware.org/git/?p=glibc.git does
>>>> not
>>>> so we should not be using -Wredundant-decls.
>>>>
>>>> The patch removes it.
>>>
>>>
>>> This commit message seems to be missing any mention of
>>> which versions of pixman this change breaks and why
>>> it's OK now to break compiling against them...
>>
>>
>>
>> The change does not _break_ anything.
>
> The change breaks the fix that the code was put in to deal with,
> ie that pixman-0.16's header files generate warnings, which the
> default development qemu will turn into errors with -Werror, which
> means we won't compile.
>
>> The second removed chunk in the patch is the problem as it:
>> 1) enables -Wredundant-decls even if it was not enabled before;
>> 2) makes -Wredundant-decls an error, not just a warning.
>
> This is because not all versions of gcc have the push/pop warning
> pragma.
>
>> Default assert.h shipped with Fedora Core 18 (pretty recent and popular
>> distribution, I would say) cannot compile with -Wredundant-decls as an error
>
> This appears to be a bug in Fedora.
Rather glibc as it is still in glibc's master branch:
http://sourceware.org/git/?p=glibc.git;a=blob;f=assert/assert.h;h=1bb9b2d50f92da3b17cc0b937ecd6141a87dc196;hb=HEAD
Or am I looking at wrong git?
> I guess the correct workaround for that bug is going to be either:
>
> * wrap assert.h the same way we wrap pixman.h
> * drop -Wredundant-decls :-(
The second option seems less ugly.
--
Alexey
prev parent reply other threads:[~2013-04-15 7:59 UTC|newest]
Thread overview: 31+ messages / expand[flat|nested] mbox.gz Atom feed top
2013-04-15 2:31 [Qemu-devel] [PATCH] pixman: remove -Wredundand-decls Alexey Kardashevskiy
2013-04-15 7:08 ` Markus Armbruster
2013-04-15 7:12 ` Alexey Kardashevskiy
2013-04-15 8:23 ` Markus Armbruster
2013-04-15 9:50 ` Alexey Kardashevskiy
2013-04-15 10:01 ` Peter Maydell
2013-04-15 10:40 ` Peter Maydell
2013-04-15 10:52 ` Alexey Kardashevskiy
2013-04-15 11:00 ` Peter Maydell
2013-04-15 11:08 ` Alexey Kardashevskiy
2013-04-15 11:14 ` Peter Maydell
2013-04-15 11:20 ` Alexey Kardashevskiy
2013-04-15 12:57 ` Markus Armbruster
2013-04-15 13:03 ` Alexey Kardashevskiy
2013-04-15 15:55 ` Markus Armbruster
2013-04-15 22:48 ` Alexey Kardashevskiy
2013-04-16 3:16 ` Alexey Kardashevskiy
2013-04-16 7:57 ` Markus Armbruster
2013-04-16 9:22 ` Paolo Bonzini
2013-04-16 10:29 ` Markus Armbruster
2013-04-16 10:54 ` Alexey Kardashevskiy
2013-04-16 11:12 ` Paolo Bonzini
2013-04-16 11:52 ` Markus Armbruster
2013-04-16 9:00 ` Markus Armbruster
2013-04-16 10:54 ` Alexey Kardashevskiy
2013-04-16 11:51 ` Kevin Wolf
2013-04-16 12:12 ` Markus Armbruster
2013-04-15 7:18 ` Peter Maydell
2013-04-15 7:26 ` Alexey Kardashevskiy
2013-04-15 7:30 ` Peter Maydell
2013-04-15 7:45 ` Alexey Kardashevskiy [this message]
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=516BAFFE.9020306@ozlabs.ru \
--to=aik@ozlabs.ru \
--cc=blauwirbel@gmail.com \
--cc=david@gibson.dropbear.id.au \
--cc=kraxel@redhat.com \
--cc=peter.maydell@linaro.org \
--cc=qemu-devel@nongnu.org \
--cc=stefanha@redhat.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).