linux-kernel - Re: linux-next: percpu tree build warning

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: <4B0F7224.9020509@kernel.org>
Date:	Fri, 27 Nov 2009 15:31:00 +0900
From:	Tejun Heo <tj@...nel.org>
To:	Ingo Molnar <mingo@...e.hu>
CC:	Rusty Russell <rusty@...tcorp.com.au>,
	Stephen Rothwell <sfr@...b.auug.org.au>,
	Fr??d??ric Weisbecker <fweisbec@...il.com>,
	Peter Zijlstra <a.p.zijlstra@...llo.nl>,
	Christoph Lameter <cl@...ux-foundation.org>,
	linux-next@...r.kernel.org, linux-kernel@...r.kernel.org,
	Linus Torvalds <torvalds@...ux-foundation.org>,
	Andrew Morton <akpm@...ux-foundation.org>
Subject: Re: linux-next: percpu tree build warning

Hello, Ingo.
> At least to me a typo like this would stick out like a sore thumb during 
> review.

Yeah, maybe, but it still shows why reusing the same name for global
and local variables behind compiler's back is a bad idea.

> I'd recognize &reg1 as a stack local variable immediately, and when i 
> see it being used in this_cpu_inc() i'd go 'huh' immediately.
> 
> OTOH, the two examples of confusion i gave you in my previous mail would 
> be far less obvious. The 'visual distance' to a percpu variable 
> definition is greater (it's at least file scope in 95% of the cases), so 
> i wouldnt be able to 'see' which the percpu variables are, from a code 
> context.

With proper __percpu annotations (which we desparately need for
dynamic percpu pointers anyway) the 'visual distance' should remain
fine in most cases, I think.

If we can manage the separate namespace thing without adding confusion
regarding different types of accessors and the actually non-existing
but yet visible differences between static and dynamic percpu
variables, I think it would be good.  But it costs us quite a bit and
__percpu sparse annotation has almost complete coverage over the issue
including the visible queue telling that something is percpu.  So,
given that, to me __percpu seems like a much better way to do it.

Thanks.

-- 
tejun
--
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/