linux-kernel - Re: Solving M produces N consumers scalability problem

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: <20131104202806.GL29695@two.firstfloor.org>
Date:	Mon, 4 Nov 2013 21:28:06 +0100
From:	Andi Kleen <andi@...stfloor.org>
To:	Anatol Pomozov <anatol.pomozov@...il.com>
Cc:	Andi Kleen <andi@...stfloor.org>,
	LKML <linux-kernel@...r.kernel.org>
Subject: Re: Solving M produces N consumers scalability problem

> This solution pretty-much equivalent to per-CPU data structures. And

No it's not, it doesn't require one queue per CPU.

A CPU these days isn't really a CPU anymore, but often a CPU thread,
which is much more light weight. So having a queue per CPU is often
total overkill, and amplifies any per queue costs (like invalidation)

-Andi
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/