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

Groups > linux.kernel > #1739827 > unrolled thread

Re: [PATCH 08/25] hrtimer: Reduce conditional code (expires_next, next_timer)

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 08/25] hrtimer: Reduce conditional code (expires_next,  next_timer) Peter Zijlstra <peterz@infradead.org> - 2017-09-26 14:20 +0200
    Re: [PATCH 08/25] hrtimer: Reduce conditional code (expires_next,  next_timer) Thomas Gleixner <tglx@linutronix.de> - 2017-09-28 10:10 +0200

#1739827 — Re: [PATCH 08/25] hrtimer: Reduce conditional code (expires_next, next_timer)

On Thu, Aug 31, 2017 at 12:23:37PM -0000, Anna-Maria Gleixner wrote:
> The hrtimer_cpu_base struct member expires_next and next_timer are
> conditional members (CONFIG_HIGH_RES_TIMERS). This makes the hrtimer code
> more complex and harder to understand than it actually is.
> 
> Reduce the conditionals related to those two struct members.

So this grows the data structure for !HIGH_RES for easier code? We no
longer care about archs that don't support HIGH_RES?

[toc] | [next] | [standalone]

#1741257

On Tue, 26 Sep 2017, Peter Zijlstra wrote:

> On Thu, Aug 31, 2017 at 12:23:37PM -0000, Anna-Maria Gleixner wrote:
> > The hrtimer_cpu_base struct member expires_next and next_timer are
> > conditional members (CONFIG_HIGH_RES_TIMERS). This makes the hrtimer code
> > more complex and harder to understand than it actually is.
> > 
> > Reduce the conditionals related to those two struct members.
> 
> So this grows the data structure for !HIGH_RES for easier code? We no
> longer care about archs that don't support HIGH_RES?

We care, but there is a tradeoff between the required conditional code and
the marginal storage overhead. Easier to follow code which does a few
superflous things for the !highres case is certainly preferred.

Thanks,

	tglx

[toc] | [prev] | [standalone]

Back to top | Article view | linux.kernel

csiph-web