linux-kernel - Re: [PATCH v2] locking/mutex: remove redundant argument from __mutex_lock

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: <ZPYrFeAzWjESRiJD@gmail.com>
Date:   Mon, 4 Sep 2023 21:08:05 +0200
From:   Ingo Molnar <mingo@...nel.org>
To:     Peter Zijlstra <peterz@...radead.org>
Cc:     Michał Mirosław <mirq-linux@...e.qmqm.pl>,
        Ingo Molnar <mingo@...hat.com>, Will Deacon <will@...nel.org>,
        Waiman Long <longman@...hat.com>,
        Boqun Feng <boqun.feng@...il.com>, linux-kernel@...r.kernel.org
Subject: Re: [PATCH v2] locking/mutex: remove redundant argument from
 __mutex_lock_common()


* Peter Zijlstra <peterz@...radead.org> wrote:

> On Sun, Sep 03, 2023 at 12:33:19AM +0200, Michał Mirosław wrote:
> > use_ww_ctx is equivalent to ww_ctx != NULL. The one case where
> > use_ww_ctx was true but ww_ctx == NULL leads to the same
> > __mutex_add_waiter() call via __ww_mutex_add_waiter().
> > 
> > Since now __ww_mutex_add_waiter() is called only with ww_ctx != NULL
> > (from both regular and PREEMPT_RT implementations), remove the
> > branch there.
> > 
> 
> There were compilers that failed to constant propagate the ww_ctx==NULL 
> thing properly and generated crap code, the use_ww_ctx thing fixed that.
> 
> I can't remember which compilers that were (my brain is saying <gcc-6 or 
> something, but I could be totally wrong) and if we still care about 
> people using them (probably not).

The changelog of the patch should probably include before/after generated 
code comparison & analysis - or at minimum a '/bin/size' comparison to 
quantify the changes to generated code.

Thanks,

	Ingo