linux-kernel - Re: [PATCH] kernel: locking: add releases(lock) annotation

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]

Date:   Thu, 6 Feb 2020 14:59:15 +0000 (GMT)
From:   Jules Irenge <jbi.octave@...il.com>
To:     Peter Zijlstra <peterz@...radead.org>
cc:     Jules Irenge <jbi.octave@...il.com>, boqun.feng@...il.com,
        mingo@...hat.com, will@...nel.org, linux-kernel@...r.kernel.org
Subject: Re: [PATCH] kernel: locking: add releases(lock) annotation

On Tue, 17 Dec 2019, Peter Zijlstra wrote:

> On Mon, Dec 16, 2019 at 03:39:52PM +0000, Jules Irenge wrote:
> > Add releases(lock) annotation to remove issue detected by sparse tool.
> > warning: context imbalance in xxxxxxx() - unexpected unlock
> > 
> > Signed-off-by: Jules Irenge <jbi.octave@...il.com>
> 
> So, personally I detest these sparse things.
> 
> But I'm also confused, as that function already has the annotation, see
> spinlock_api_smp.h. In order for sparse to see these annotations at the
> usage size, they need to be on the declaration, not the definition.
> 

Yes, I completely agree with you, but acording to my short experience 
with Sparse, the warning at function definition will always be there 
despite the annotation being at the function declaration.

The annotation at declaration help fix the warning when the anotated 
function is called within another function, this function can be 
elsewhere. It fixes the warning at the function within which the 
annotated function has been called.

Please correct me if I am wrong.

Kind regards,
Jules