|
2010-04-24
, 09:00
|
|
Posts: 415 |
Thanked: 161 times |
Joined on Apr 2010
@ San Francisco, CA
|
#62
|
Titan
In your ULV kernel, it seems that after a call /sys/devices/system/cpu/cpu0/cpufreq/scaling_min_freq is reset to 125000 from 124999 and after that the minimum frequency I can see in /sys/devices/system/cpu/cpu0/cpufreq/stats/time_in_state is 250MHz instead of 125MHz
The only way to fix the issue is to do
echo 124999 > /sys/devices/system/cpu/cpu0/cpufreq/scaling_min_freq
after every call
|
2010-04-24
, 15:36
|
Posts: 946 |
Thanked: 1,650 times |
Joined on Oct 2009
@ Germany
|
#63
|
Titan:
Will you be able to make a fix for this so the minimum frequency won't have to be changed after every phone call? Right now I am using ULV kernel with the min at 250 because that kernel uses the lowest voltage at that frequency. I would perfer to use the XLV kernel and set the minimum frequency at 125 but having to change the frquency back to 125 after every call is too much of a hassle. Thank you again for your awesome kernels!
The Following User Says Thank You to titan For This Useful Post: | ||
|
2010-04-25
, 10:59
|
Posts: 24 |
Thanked: 6 times |
Joined on Feb 2010
|
#64
|
|
2010-04-25
, 11:47
|
Posts: 946 |
Thanked: 1,650 times |
Joined on Oct 2009
@ Germany
|
#65
|
I have the 25 kernel, using ideal settings, but my n900 idles at 500mhz? Is that correct....or how do I make it idle at the default 250mhz?
The Following User Says Thank You to titan For This Useful Post: | ||
|
2010-04-25
, 15:37
|
Posts: 44 |
Thanked: 5 times |
Joined on Apr 2010
|
#66
|
I assume you use the kernel-power-flasher version 24 from extras-testing
and you the ULV or XLV settings?
The version maemo25 from extras-devel should have fixed the problems with
the minimum frequency using avoid_frequencies (see the wiki page).
I'm going to upload a new kernel-power-settings package that also configures this
new kernel option soon.
|
2010-04-25
, 18:21
|
Posts: 203 |
Thanked: 152 times |
Joined on May 2009
@ Austria
|
#67
|
echo > /sys/devices/system/cpu/cpu0/cpufreq/ondemand/avoid_frequencies
echo 1 > /sys/devices/system/cpu/cpu0/cpufreq/ondemand/ignore_nice_load
|
2010-04-26
, 04:57
|
Posts: 24 |
Thanked: 6 times |
Joined on Feb 2010
|
#68
|
It idles at 0MHz. The lowest frequency is just used for load workload.
500MHz as minimum appears to be better for battery life than 250.
|
2010-04-26
, 05:13
|
Posts: 24 |
Thanked: 6 times |
Joined on Feb 2010
|
#69
|
|
2010-04-26
, 06:14
|
Posts: 946 |
Thanked: 1,650 times |
Joined on Oct 2009
@ Germany
|
#70
|
your kernel is too buggy. I did:
rm -f /etc/default/kernel-power
ln -s /usr/share/kernel-power-settings/ideal /etc/default/kernel-power
to permanently change the settings and it still goes back to 600mhz after reboot. How do i fix this?
This may be a bug....rebooting while charging will reset to default 250-600mhz (even if I did the ln command.)
Anyway, BFS was written for desktop computers with only one or few cores and it is said to work better than the default cpu scheduler in these systems, but it works worst than the default in computers with lots of cores. I won't try to explain in detail as it is better reading this: http://ck.kolivas.org/patches/bfs/bfs-faq.txt
Before the BFS, Con Kolivas wrote de Staircase CPU Scheduler, witch I used to compile a custom kernel for Diablo, if you have a N800 or N810 you can download it here: https://garage.maemo.org/frs/?group_id=1200
Unfortunately, I'm an electric engineering student, not a performance products specialist, so I'm not familiar with benchmarking performance. But I can try describing my experience with the Staircase CPU Scheduler in my N810.
I was trying to make the N810 a better multitasking device, but it had a 128MB RAM + 128MB swap limitation in it's default configuration. So I first tried increasing the swap space by moving (all) the system files to the internal flash card and formatting the /dev/mtdblock4 device as swap. That didn't work well because the system would take too much time to move data from ram to swap a from swap to ram, making the device unresponsive, and the watchdog would reboot it. So I compiled the patched kernel and not only the multitasking experience was faster, but it was also stable, the reboots were gone because the Staircase CPU scheduler wouldn't let the system become unresponsive anymore. That's the goal of the schedulers Con Kolivas wrote: make the system more responsive increasing the end-user experience.
So, that's a lot of talking but I won't be able to provide you test results to prove what I'm saying, but I hope I convinced you to try it yourself.
Sorry for my English, I hope you could understand it.