linux-kernel - Re: [PATCH] irqdomain: move revmap_trees_mutex to struct irq

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: <2a8b1556-f39c-7a05-0b86-4d8f2d3bc1ed@arm.com>
Date:   Wed, 4 Oct 2017 15:59:48 +0100
From:   Marc Zyngier <marc.zyngier@....com>
To:     Masahiro Yamada <yamada.masahiro@...ionext.com>,
        Thomas Gleixner <tglx@...utronix.de>,
        Jason Cooper <jason@...edaemon.net>
Cc:     linux-kernel@...r.kernel.org
Subject: Re: [PATCH] irqdomain: move revmap_trees_mutex to struct irq_domain

On 04/10/17 15:46, Masahiro Yamada wrote:
> The revmap_trees_mutex protects domain->revmap_tree.  It is allowed
> to modify revmap_tree of two different domains concurrently.  Change
> the global mutex to per-domain mutex.
> 
> Signed-off-by: Masahiro Yamada <yamada.masahiro@...ionext.com>

Whilst this doesn't seem completely outlandish from a data-structure
PoV, I really wonder if that's an actual bottleneck. Do you have a
use-case where the rate of interrupt being mapped/unmapped is so high
that this becomes an actual contention point? If so, it would be good to
mention it in the commit message.

Thanks,

	M.
-- 
Jazz is not dead. It just smells funny...