linux-kernel - Re: [PATCH] pinctrl: Include <linux/bug.h> to prevent compile errors

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

Message-ID: <4F5A8DA7.5070900@wwwdotorg.org>
Date:	Fri, 09 Mar 2012 16:09:27 -0700
From:	Stephen Warren <swarren@...dotorg.org>
To:	Paul Gortmaker <paul.gortmaker@...driver.com>
CC:	Linus Walleij <linus.walleij@...aro.org>,
	linux-kernel@...r.kernel.org
Subject: Re: [PATCH] pinctrl: Include <linux/bug.h> to prevent compile errors

On 03/09/2012 02:29 PM, Paul Gortmaker wrote:
> On Fri, Mar 9, 2012 at 3:43 PM, Stephen Warren <swarren@...dotorg.org> wrote:
>> On 03/09/2012 01:32 PM, Paul Gortmaker wrote:
>>> [[PATCH] pinctrl: Include <linux/bug.h> to prevent compile errors] On 09/03/2012 (Fri 13:18) Stephen Warren wrote:
>>>
>>>> Macros in <linux/pinctrl/machine.h> call ARRAY_SIZE(), the definition of
>>>> which eventually calls BUILD_BUG_ON_ZERO(), which is defined in
>>>> <linux/bug.h>. Include that so that every .c file using the pinctrl macros
>>>> doesn't have to do that itself.
>>>
>>> Which C files are failing?  The approach was to be adding it only
>>> for headers with static inlines which use it, and add it to C files
>>> that are actually *deploying* the macros.
>>
>> For me, the files that are failing haven't been committed upstream yet;
>> I expect them to be very soon after the 3.4 merge window completes.
>>
>> However, I expect you'll see this issue with arch/arm/mach-u300/core.c,
>> since it uses the same macros defined in pinctrl/machine.h that cause
>> the problem for me, unless one of the include files there picks up bug.h
>> already.
>>
>> As a general statement though, once pinctrl becomes more widely adopted,
>> I expect to see many files start to use macros from pinctrl/machine.h,
>> and hence implicitly use ARRAY_SIZE(), and hence even more implicitly
>> depend on linux/bug.h. I don't think it's at all reasonable to force the
>> author of every one of those files to track down 3 or 4 levels of header
>> file dependencies to find that they need linux/bug.h even though they
>> make no direct use of its features, when it's known that almost any user
>> of pinctrl/machine.h is going to hit this issue.
> 
> There was a similar discussion before, where there was concern over
> "every one of those files" but in the end the number is a fraction of
> a percent.   There is no harm in the author knowing they are making
> _use_ of BUG stuff -- in fact if you happened to be trying to write some
> fault tolerant code, you might really be glad to know, so you could take
> steps to avoid it....

The primary purpose for pinctrl/machine.h is files that define a certain
data structure that will almost invariably be initialized using the
macros in that header. Size those macros use ARRAY_SIZE, every one of
those .c files will require bug.h.

There are 1 or 2 files in drivers/pinctrl that include pinctrl/machine.h
solely for the data structure definitions or function prototypes, and
won't use those macros, but those are the exception rather than the rule
for this one header at least.
--
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/