Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]


Groups > linux.kernel > #1741506 > unrolled thread

Re: [PATCH 06/25] hrtimer: Store running timer in hrtimer_clock_base

Started byThomas Gleixner <tglx@linutronix.de>
First post2017-09-28 14:50 +0200
Last post2017-09-28 14:50 +0200
Articles 1 — 1 participant

Back to article view | Back to linux.kernel

This discussion starts older than the indexed window; earlier articles aren't shown. The article labeled Started by below is the oldest one visible, not the original post.


Contents

  Re: [PATCH 06/25] hrtimer: Store running timer in  hrtimer_clock_base Thomas Gleixner <tglx@linutronix.de> - 2017-09-28 14:50 +0200

#1741506 — Re: [PATCH 06/25] hrtimer: Store running timer in hrtimer_clock_base

FromThomas Gleixner <tglx@linutronix.de>
Date2017-09-28 14:50 +0200
SubjectRe: [PATCH 06/25] hrtimer: Store running timer in hrtimer_clock_base
Message-ID<uuGqn-C7-37@gated-at.bofh.it>
On Mon, 25 Sep 2017, Peter Zijlstra wrote:

> On Thu, Aug 31, 2017 at 12:23:36PM -0000, Anna-Maria Gleixner wrote:
> > Unfortunately this makes cpu_clock base larger than 32 bytes on 32bit
> > kernels. Instead of having huge gaps due to alignment, remove the alignment
> > and let the compiler pack cpu base for 32bit. The resulting cache access
> > patterns are fortunately not really different from the current
> > behaviour.
> 
> How did you determine this? Perf lacks much useful tooling right about
> there :/

Just by analyzing the resulting layout and looking at the number of cache
lines involved for the frequently used clocks (mono/real).

Thanks,

	tglx

[toc] | [standalone]


Back to top | Article view | linux.kernel


csiph-web