Bug 43114 - Can't set high engine clock for RadeonHD 6620G
Summary: Can't set high engine clock for RadeonHD 6620G
Status: RESOLVED PATCH_ALREADY_AVAILABLE
Alias: None
Product: Drivers
Classification: Unclassified
Component: Video(DRI - non Intel) (show other bugs)
Hardware: All Linux
: P1 normal
Assignee: drivers_video-dri
URL:
Keywords:
Depends on:
Blocks:
 
Reported: 2012-04-17 00:02 UTC by RussianNeuroMancer
Modified: 2013-07-04 09:48 UTC (History)
0 users

See Also:
Kernel Version:
Subsystem:
Regression: No
Bisected commit-id:


Attachments
dmesg (59.56 KB, application/octet-stream)
2012-04-17 00:03 UTC, RussianNeuroMancer
Details
cpuinfo (3.66 KB, application/octet-stream)
2012-04-17 00:03 UTC, RussianNeuroMancer
Details
lspci (26.57 KB, application/octet-stream)
2012-04-17 00:03 UTC, RussianNeuroMancer
Details

Description RussianNeuroMancer 2012-04-17 00:02:48 UTC
APU model: A8-3500M with Radeon HD 6620G.
Laptop model: Acer Aspire 7560G.
According to official specs of this APU
http://www.amd.com/us/products/notebook/apu/mainstream/Pages/mainstream.aspx#6
https://en.wikipedia.org/wiki/Comparison_of_AMD_graphics_processing_units#cite_ref-122
GPU core clock on high profile should be 444 MHz, but even when I set high profile manually

~# cat /sys/class/drm/card0/device/power_method
profile
~# cat /sys/class/drm/card0/device/power_profile
high

GPU core clock is not 444 MHz anyway, even below 200 MHz:
~$ cat /sys/kernel/debug/dri/0/radeon_pm_info
default engine clock: 200000 kHz
current engine clock: 7990 kHz
default memory clock: 667000 kHz
Maybe output of radeon_pm_info is incorrect, because change between different profiles and dynpm make no difference in radeon_pm_info information.

Reproduceable with Linux 3.2 and Linux 3.4rc3. dmesg of 3.4rc3 boot process attached. Is there any additional information I need to provide?
Comment 1 RussianNeuroMancer 2012-04-17 00:03:21 UTC
Created attachment 72934 [details]
dmesg
Comment 2 RussianNeuroMancer 2012-04-17 00:03:38 UTC
Created attachment 72935 [details]
cpuinfo
Comment 3 RussianNeuroMancer 2012-04-17 00:03:57 UTC
Created attachment 72936 [details]
lspci
Comment 4 RussianNeuroMancer 2013-07-04 09:48:10 UTC
Should be resolved now (in Linux 3.11) by AMD Power Management patches.

P.S. Hopefully I set right bugreport status.

Note You need to log in before you can comment on or make changes to this bug.