linux-kernel - Re: Proper kernel irq time accounting -v3

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]

Date:	Thu, 30 Sep 2010 19:36:52 +0200
From:	Andi Kleen <andi@...stfloor.org>
To:	Venkatesh Pallipadi <venki@...gle.com>
Cc:	Peter Zijlstra <peterz@...radead.org>, Ingo Molnar <mingo@...e.hu>,
	"H. Peter Anvin" <hpa@...or.com>,
	Thomas Gleixner <tglx@...utronix.de>,
	Balbir Singh <balbir@...ux.vnet.ibm.com>,
	linux-kernel@...r.kernel.org, Paul Turner <pjt@...gle.com>,
	Eric Dumazet <eric.dumazet@...il.com>
Subject: Re: Proper kernel irq time accounting -v3

Venkatesh Pallipadi <venki@...gle.com> writes:
>
> Yes. Overhead will be RDTSC.
> notsc is a boot option that will disable this. I didn't want to have a

notsc is a very heavy hammer.

> specific run time option to disable this as we would already have
> other RDTSC users like native_sched_clock. Do you think we need a
> specific option to disable this feature? I don't like adding a runtime
> option because it complicates the code a bit (to handle things like
> someone turning it off and turning it on later, etc).

How about some benchmarks on a known slow TSC system, like a P4?

If you can't measure it even with high interrupt loads it is probably
not relevant. If it is it would be better to have a on/off switch.

-Andi
-- 
ak@...ux.intel.com -- Speaking for myself only.
--
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/