From: Peter Zijlstra <peterz@infradead.org>
To: Arjan van de Ven <arjan@infradead.org>
Cc: linux-kernel@vger.kernel.org, torvalds@linux-foundation.org,
dwmw2@infradead.org, drepper@redhat.com, mingo@elte.hu,
tglx@tglx.de
Subject: Re: [PATCH 13/13] hrtimer: make select() and poll() use the hrtimer range feature
Date: Tue, 02 Sep 2008 10:22:20 +0200 [thread overview]
Message-ID: <1220343740.8609.21.camel@twins> (raw)
In-Reply-To: <20080901161459.0990a16b@infradead.org>
On Mon, 2008-09-01 at 16:14 -0700, Arjan van de Ven wrote:
> From: Arjan van de Ven <arjan@linux.intel.com>
> Subject: [PATCH] hrtimer: make select() and poll() use the hrtimer range feature
>
> This patch makes the select() and poll() hrtimers use the new range
> feature and settings from the task struct.
>
> In addition, this includes the estimate_accuracy() function that Linus
> posted to lkml (but with a few steps added based on experiments).
>
> Signed-off-by: Arjan van de Ven <arjan@linux.intel.com>
> ---
> fs/select.c | 68 +++++++++++++++++++++++++++++++++++++++++++++++++++++++++-
> 1 files changed, 66 insertions(+), 2 deletions(-)
>
> diff --git a/fs/select.c b/fs/select.c
> index f6dceb5..21bf77d 100644
> --- a/fs/select.c
> +++ b/fs/select.c
> @@ -28,6 +28,62 @@
>
> #include <asm/uaccess.h>
>
> +
> +/* Estimate expected accuracy in ns from a timeval */
> +
> +static unsigned long __estimate_accuracy(struct timespec *tv)
> +{
> + /*
> + * Tens of ms if we're looking at seconds, even
> + * more for 10s+ sleeping
> + */
> + if (tv->tv_sec) {
> + /* 100 milliseconds for long sleeps */
> + if (tv->tv_sec > 10)
> + return 100 * NSEC_PER_MSEC;
> +
> + /*
> + * Tens of ms for second-granularity sleeps. This,
> + * btw, is the historical Linux 100Hz timer range.
> + */
> + return 10 * NSEC_PER_MSEC;
> + }
> +
> + /* 5 msec if we're looking at 100+ milliseconds */
> + if (tv->tv_nsec > 100 * NSEC_PER_MSEC)
> + return 5 * NSEC_PER_MSEC;
> +
> + /* A msec if we're looking at 10+ milliseconds */
> + if (tv->tv_nsec > 10 * NSEC_PER_MSEC)
> + return NSEC_PER_MSEC;
> +
> + /* half a msec if we're looking at milliseconds */
> + if (tv->tv_nsec > NSEC_PER_MSEC)
> + return NSEC_PER_MSEC/2;
> +
> + /* Single usecs if we're looking at microseconds */
> + if (tv->tv_nsec > NSEC_PER_USEC)
> + return NSEC_PER_USEC;
> +
> + /* Aim for tenths of nanosecs otherwise */
> + return 10;
> +}
Why not use a simple logarithmic decay to drive this estimate?
next prev parent reply other threads:[~2008-09-02 8:22 UTC|newest]
Thread overview: 46+ messages / expand[flat|nested] mbox.gz Atom feed top
2008-09-01 23:03 [PATCH 0/13] Turn hrtimers into a range capable timer Arjan van de Ven
2008-09-01 23:05 ` [PATCH 1/13] hrtimer: add abstraction functions for accessing the "expires" member Arjan van de Ven
2008-09-01 23:05 ` [PATCH 2/13] hrtimer: convert kvm to the new hrtimer apis Arjan van de Ven
2008-09-01 23:06 ` [PATCH 3/13] hrtimer: convert timerfd " Arjan van de Ven
2008-09-01 23:07 ` [PATCH 4/13] hrtimer: convert net::sched_cbq " Arjan van de Ven
2008-09-01 23:08 ` [PATCH 5/13] hrtimer: convert kernel/* " Arjan van de Ven
2008-09-01 23:09 ` [PATCH 6/13] hrtimer: convert powerpc/oprofile " Arjan van de Ven
2008-09-01 23:09 ` [PATCH 7/13] hrtimer: convert kvm-ia64 " Arjan van de Ven
2008-09-01 23:10 ` [PATCH 8/13] hrtimer: convert s390 " Arjan van de Ven
2008-09-01 23:11 ` [PATCH 9/13] hrtimer: convert sound/ " Arjan van de Ven
2008-09-01 23:12 ` [PATCH 10/13] hrtimer: rename the "expires" struct member to avoid accidental usage Arjan van de Ven
2008-09-01 23:12 ` Arjan van de Ven
2008-09-01 23:13 ` [PATCH 11/13] hrtimer: turn hrtimers into range timers Arjan van de Ven
2008-09-02 8:22 ` Peter Zijlstra
2008-09-02 11:08 ` Peter Zijlstra
2008-09-02 11:15 ` Peter Zijlstra
2008-09-02 13:06 ` Arjan van de Ven
2008-09-02 13:05 ` Arjan van de Ven
2008-09-02 13:47 ` Peter Zijlstra
2008-09-02 16:02 ` Arjan van de Ven
2008-09-01 23:14 ` [PATCH 12/13] hrtimer: create a "timer_slack" field in the task struct Arjan van de Ven
2008-09-02 10:04 ` Pavel Machek
2008-09-02 13:03 ` Arjan van de Ven
2008-09-08 13:27 ` Pavel Machek
2008-09-08 13:40 ` Arjan van de Ven
2008-09-08 14:15 ` Pavel Machek
2008-09-08 14:22 ` Arjan van de Ven
2008-09-13 16:24 ` Pavel Machek
2008-09-14 15:21 ` Ulrich Drepper
2008-09-14 15:27 ` Arjan van de Ven
2008-09-14 15:57 ` Pavel Machek
2008-09-14 16:04 ` Ulrich Drepper
2008-09-14 16:14 ` Arjan van de Ven
2008-09-17 7:42 ` Pavel Machek
2008-09-30 5:16 ` KOSAKI Motohiro
2008-09-30 8:28 ` Arjan van de Ven
2008-09-30 8:54 ` KOSAKI Motohiro
2008-09-01 23:14 ` [PATCH 13/13] hrtimer: make select() and poll() use the hrtimer range feature Arjan van de Ven
2008-09-02 8:22 ` Peter Zijlstra [this message]
2008-09-02 16:03 ` Arjan van de Ven
2008-09-06 14:56 ` [PATCH 0/13] Turn hrtimers into a range capable timer Ingo Molnar
2008-09-06 16:30 ` Arjan van de Ven
2008-09-06 16:33 ` Ingo Molnar
2008-09-12 3:39 ` Rusty Russell
2008-09-12 5:42 ` Arjan van de Ven
2008-09-12 20:24 ` Thomas Gleixner
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=1220343740.8609.21.camel@twins \
--to=peterz@infradead.org \
--cc=arjan@infradead.org \
--cc=drepper@redhat.com \
--cc=dwmw2@infradead.org \
--cc=linux-kernel@vger.kernel.org \
--cc=mingo@elte.hu \
--cc=tglx@tglx.de \
--cc=torvalds@linux-foundation.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).