cpufreq Archive on lore.kernel.org
 help / color / mirror / Atom feed
From: Eric Piel <Eric.Piel@tremplin-utc.net>
To: bugme-daemon@bugzilla.kernel.org
Cc: linux@brodo.de, cpufreq@www.linux.org.uk
Subject: Re: [Bug 4379] Default sampling rates for ondemand governor are too high on a amd64
Date: Wed, 16 Nov 2005 08:50:42 +0100	[thread overview]
Message-ID: <437AE4D2.2030503@tremplin-utc.net> (raw)
In-Reply-To: <200511152159.jAFLxHL4003020@fire-2.osdl.org>

[-- Attachment #1: Type: text/plain, Size: 883 bytes --]

15.11.2005 22:59, bugme-daemon@bugzilla.kernel.org wrote/a écrit:
> http://bugzilla.kernel.org/show_bug.cgi?id=4379
> 
> 
> 
> 
> 
> ------- Additional Comments From linux@brodo.de  2005-11-15 13:53 -------
> Is this problem still present in 2.6.14?
> 
I think so.

Well, I've got a patch that should fix it. It make the polling frequency 
10 times faster than it used to be. This is valid _only_ for kernel >= 
2.6.15-rc1 because there is now a check to avoid polling too fast. I've 
tested it here, on my pentium III it works.

Venki, do you see any problem with this patch?

Eric

-- 
From: Eric Piel <eric.piel@tremplin-utc.net>

Make the polling frequency 10 times faster than it used to be, now that 
there is a check to avoid too high freqencies. Solve problems with 
hardwares which have a long transition latency.

Signed-off-by: Eric Piel <eric.piel@tremplin-utc.net>
-- 

[-- Attachment #2: cpufreq-ondemand-more-frequent-polling-2.6.15-rc1.patch --]
[-- Type: text/x-patch, Size: 1074 bytes --]

--- linux-2.6.14/drivers/cpufreq/cpufreq_ondemand.c.bak	2005-11-16 08:35:14.000000000 +0100
+++ linux-2.6.14/drivers/cpufreq/cpufreq_ondemand.c	2005-11-16 08:36:57.000000000 +0100
@@ -39,7 +39,7 @@
 
 /* 
  * The polling frequency of this governor depends on the capability of 
- * the processor. Default polling frequency is 1000 times the transition
+ * the processor. Default polling frequency is 100 times the transition
  * latency of the processor. The governor will work on any processor with 
  * transition latency <= 10mS, using appropriate sampling 
  * rate.
@@ -53,7 +53,7 @@ static unsigned int 				def_sampling_rat
 #define MIN_STAT_SAMPLING_RATE			(MIN_SAMPLING_RATE_RATIO * jiffies_to_usecs(10))
 #define MIN_SAMPLING_RATE			(def_sampling_rate / MIN_SAMPLING_RATE_RATIO)
 #define MAX_SAMPLING_RATE			(500 * def_sampling_rate)
-#define DEF_SAMPLING_RATE_LATENCY_MULTIPLIER	(1000)
+#define DEF_SAMPLING_RATE_LATENCY_MULTIPLIER	(100)
 #define DEF_SAMPLING_DOWN_FACTOR		(1)
 #define MAX_SAMPLING_DOWN_FACTOR		(10)
 #define TRANSITION_LATENCY_LIMIT		(10 * 1000)

[-- Attachment #3: Type: text/plain, Size: 147 bytes --]

_______________________________________________
Cpufreq mailing list
Cpufreq@lists.linux.org.uk
http://lists.linux.org.uk/mailman/listinfo/cpufreq

  reply	other threads:[~2005-11-16  7:50 UTC|newest]

Thread overview: 10+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2005-11-15 21:59 [Bug 4379] Default sampling rates for ondemand governor are too high on a amd64 bugme-daemon
2005-11-16  7:50 ` Eric Piel [this message]
  -- strict thread matches above, loose matches on Subject: below --
2006-02-05 10:46 bugme-daemon
2007-02-11  9:10 bugme-daemon
     [not found] <bug-4379-3570@http.bugzilla.kernel.org/>
2007-11-18  7:02 ` bugme-daemon
2007-11-19 13:17 ` bugme-daemon
2008-05-14 13:39 ` bugme-daemon
2008-05-14 13:44 ` bugme-daemon
2008-05-14 13:54 ` bugme-daemon
2008-05-15 18:12 ` bugme-daemon

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=437AE4D2.2030503@tremplin-utc.net \
    --to=eric.piel@tremplin-utc.net \
    --cc=bugme-daemon@bugzilla.kernel.org \
    --cc=cpufreq@www.linux.org.uk \
    --cc=linux@brodo.de \
    /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