netdev - Re: [PATCH 1/2]: softirq: Add support for triggering softirq work on softirqs.

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: <20080919.132951.46369136.davem@davemloft.net>
Date:	Fri, 19 Sep 2008 13:29:51 -0700 (PDT)
From:	David Miller <davem@...emloft.net>
To:	stephen.hemminger@...tta.com
Cc:	jens.axboe@...cle.com, netdev@...r.kernel.org,
	nickpiggin@...oo.com.au, ilpo.jarvinen@...sinki.fi
Subject: Re: [PATCH 1/2]: softirq: Add support for triggering softirq work
 on softirqs.

From: Stephen Hemminger <stephen.hemminger@...tta.com>
Date: Fri, 19 Sep 2008 09:57:30 -0700 (PDT)

> I wonder if the added overhead of local_irq_disable/local_irq_enable is
> killing any performance gain. Certainly it adds more latency to a sensitive code path.
> 
> Is there a good lock-less solution?

The cross-cpu IPI is 1,000 times more expensive than the local IRQ
enable/disable (on x86 it's a bunch of APIC programming), so if you're
looking for cost analysis start there.
--
To unsubscribe from this list: send the line "unsubscribe netdev" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html