linux-kernel - Re: [PATCH] perf: Fix broken build by rearranging some #includes

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [day] [month] [year] [list]

Message-ID: <4F15C2A6.3010708@gmail.com>
Date:	Tue, 17 Jan 2012 10:49:10 -0800
From:	David Daney <ddaney.cavm@...il.com>
To:	Namhyung Kim <namhyung@...il.com>
CC:	linux-kernel@...r.kernel.org, David Daney <david.daney@...ium.com>,
	Peter Zijlstra <a.p.zijlstra@...llo.nl>,
	Paul Mackerras <paulus@...ba.org>, Ingo Molnar <mingo@...e.hu>,
	Arnaldo Carvalho de Melo <acme@...stprotocols.net>,
	Joerg Roedel <joerg.roedel@....com>
Subject: Re: [PATCH] perf: Fix broken build by rearranging some #includes

On 01/16/2012 11:25 PM, Namhyung Kim wrote:
> Hi, again
>
> 2012-01-17 11:01 AM, David Daney wrote:
>> From: David Daney<david.daney@...ium.com>
>>
>> When building on my Debian/mips system, util/util.c fails to build
>> because commit 1aed2671738785e8f5aea663a6fda91aa7ef59b5 (perf kvm: Do
>> guest-only counting by default) indirectly includes stdio.h before the
>> feature selection in util.h is done. This prevents _GNU_SOURCE in
>> util.h from enabling the declaration of getline(), from now second
>> inclusion of stdio.h, and the build is broken.
>>
>> There is another breakage in util/evsel.c caused by include ordering,
>> but I didn't fully track down the commit that caused it.
>>
>> Since the #include situation is a little convoluted throughout perf, I
>> did the following:
>>
>> 1) Always include util/util.h before all other include files.
>>
>> 2) Remove all #define _GNU_SOURCE, as it is already done in util.h.
>>
>> 3) Remove all #undef _GNU_SOURCE, as they don't do what a naive person
>> might expect (or much of anything for that matter) and clutter up
>> the source.
>>
>> 4) #include ctype.h in util.h, so that ctype things have uniform
>> definitions throughout perf.
>>
>> 5) Remove #define of _FILE_OFFSET_BITS where util/util.h is already
>> providing it.
>>
>> This gives me a clean build on x86_64 (fc12) and mips (Debian).
>
> Instead of reordering, how about adding -D_GNU_SOURCE on Makefile and
> removing the line from util.h? If util.h would always be included at
> first, the end result should be same, right? And this way, we don't need
> to care about the ordering anymore.
>

Yeah, that is better.  I had tried it first, but then reversed course as 
I found so many random #define _GNU_SOURCE, that I thought there must 
have been a reason.

Upon further thought on the subject, I think you are right.  We need to 
present a consistent set of features.h defines to the entire perf program.

I withdraw this patch from consideration, and will send a new one soon, 
that does the -D_GNU_SOURCE.

David Daney
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/