linux-kernel - Re: [RFC PATCH-tip 2/6] locking/rwsem: Enable optional count-based spinning on reader

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: <576056DB.5050701@hpe.com>
Date:	Tue, 14 Jun 2016 15:11:23 -0400
From:	Waiman Long <waiman.long@....com>
To:	Davidlohr Bueso <dave@...olabs.net>
CC:	Peter Zijlstra <peterz@...radead.org>,
	Ingo Molnar <mingo@...hat.com>, <linux-kernel@...r.kernel.org>,
	<x86@...nel.org>, <linux-alpha@...r.kernel.org>,
	<linux-ia64@...r.kernel.org>, <linux-s390@...r.kernel.org>,
	<linux-arch@...r.kernel.org>, <xfs@....sgi.com>,
	Jason Low <jason.low2@...com>,
	Dave Chinner <david@...morbit.com>,
	Scott J Norton <scott.norton@....com>,
	Douglas Hatch <doug.hatch@....com>
Subject: Re: [RFC PATCH-tip 2/6] locking/rwsem: Enable optional count-based
 spinning on reader

On 06/14/2016 02:27 PM, Davidlohr Bueso wrote:
> On Tue, 14 Jun 2016, Waiman Long wrote:
>
>> This patch provides a way for the kernel code to designate specific
>> rwsems to be more aggressive in term of optimistic spinning that the
>> writers will continue to spin for some additional count-based time to
>> see if it can get the lock before sleeping. This aggressive spinning
>> mode should only be used on rwsems where the readers are unlikely to
>> go to sleep.
>
> Yikes, exposing this sort of thing makes me _very_ uneasy, not to mention
> the ad-hoc nature and its easiness to mess up. I'm not really for 
> this, even
> if it shows extraordinary performance boosts on benchmarks.
>
> Thanks,
> Davidlohr

I understand your concern. I will see if there is a way to autotune 
instead of using explicit enablement.

Cheers,
Longman