Hello,
I've tested the hrt7 patch on a Debian 2.6.22-rc4-686
(snapshot.8973), applying also the radeon patch from
http://lkml.org/lkml/2007/5/16/53,
and I tried it on my laptop, a ThinkPad X31 :
- the HPET seems to be recognized, I've in the boot log :
...
Force enabled HPET at base address 0xfed00000
...
hpet clockevent registered
hpet0: at MMIO 0xfed00000, IRQs 2, 8, 0
hpet0: 3 64-bit timers, 14318180 Hz
...
speedstep-centrino with X86_SPEEDSTEP_CENTRINO_ACPI config is deprecated.
Use X86_ACPI_CPUFREQ (acpi-cpufreq) instead.
Marking TSC unstable due to: cpufreq changes.
Time: hpet clocksource has been installed.
...
- the s2ram works nice :-)
- but as Carlos, the powertop seems to be out of order, reporting no
C-states and telling CPU is always at 100% (which is wrong) ; moreover,
the power consumption is higher than with the 2.6.21-4, and the number
of wake-ups is at least double ! :
PowerTOP version 1.5 (C) 2007 Intel Corporation
< CPU was 100% busy; no C-states were entered >
Wakeups-from-idle per second : 284.0
Power usage (ACPI estimate) : 10.2 W (0.9 hours left)
Top causes for wakeups:
9.7% ( 17.0) fast-user-switc : schedule_timeout (process_timeout)
9.1% ( 16.0) nautilus : schedule_timeout (process_timeout)
9.1% ( 16.0) gnome-vfs-daemo : schedule_timeout (process_timeout)
8.5% ( 15.0) update-notifier : schedule_timeout (process_timeout)
8.0% ( 14.0) mixer_applet2 : schedule_timeout (process_timeout)
8.0% ( 14.0) gnome-panel : schedule_timeout (process_timeout)
- as a side effect, I couldn't test effect of the radeon patch...
with regards,
Fred.
_______________________________________________
Power mailing list
[email protected]
http://www.bughost.org/mailman/listinfo/power
