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: <20241031091306.GU9767@noisy.programming.kicks-ass.net>
Date: Thu, 31 Oct 2024 10:13:06 +0100
From: Peter Zijlstra <peterz@...radead.org>
To: Thomas Gleixner <tglx@...utronix.de>
Cc: Sebastian Andrzej Siewior <bigeasy@...utronix.de>,
	linux-kernel@...r.kernel.org,
	André Almeida <andrealmeid@...lia.com>,
	Darren Hart <dvhart@...radead.org>,
	Davidlohr Bueso <dave@...olabs.net>, Ingo Molnar <mingo@...hat.com>,
	Juri Lelli <juri.lelli@...hat.com>,
	Valentin Schneider <vschneid@...hat.com>,
	Waiman Long <longman@...hat.com>
Subject: Re: [RFC PATCH 2/3] futex: Add basic infrastructure for local task
 local hash.

On Thu, Oct 31, 2024 at 12:14:16AM +0100, Thomas Gleixner wrote:

> If that's handled, then it falls flat when a futex is used as a wait
> queue. There might be a thread sitting on it waiting for an event which
> never happens, which means the old hash never gets removed. Then the
> process gets another pile of threads and can't expand the hash because
> there are already two instances.

Damn... I mean it means a userspace thread is basically stuck forever,
but that's not our problem. Yes this is annoying.

The whole requeue on lookup and hb->waiters thing looks simple enough,
but this is a bit annoying. I can probably make it work though,
however..

> I really want to start simple and let the process' futex usage come to a
> grinding halt when the resizing and rehashing takes place.

Fair enough. Lets do the simple thing first.

> I really want to know who thought that futexes are a good idea to begin
> with. Once we figured that out I need to find those who added all the
> other complexity on top of that bad idea :)

If memory serves me, it was some tall German dude that did most of that.

Powered by blists - more mailing lists

Powered by Openwall GNU/*/Linux Powered by OpenVZ