https://bugs.freedesktop.org/show_bug.cgi?id=33301
Summary: dynpm doesn't downclock Product: DRI Version: XOrg 6.7.0 Platform: Other OS/Version: Linux (All) Status: NEW Severity: normal Priority: medium Component: DRM/Radeon AssignedTo: dri-devel@lists.freedesktop.org ReportedBy: cristiklein@gmail.com
Hello,
I have an ATI Mobility Radeon HD 2400. I am using Linux 2.6.37. Currently, dynpm does not down-clock the GPU for me. dmesg is filled with the following messages:
[245199.552075] [drm:r600_pm_get_dynpm_state], Requested: e: 30000 m: 40000 p: 16 (about 10 messages / second)
I attach a script with all the necessary details.
Cristi.
https://bugs.freedesktop.org/show_bug.cgi?id=33301
Jerome Glisse glisse@freedesktop.org changed:
What |Removed |Added ---------------------------------------------------------------------------- Summary|dynpm doesn't downclock |[RADEON:KMS:R600:DYNPM] | |dynpm doesn't downclock
https://bugs.freedesktop.org/show_bug.cgi?id=33301
--- Comment #1 from Cristian KLEIN cristiklein+fontconfig@gmail.com 2011-03-08 11:58:14 PST --- Created an attachment (id=44245) --> (https://bugs.freedesktop.org/attachment.cgi?id=44245) Part of dmesg with power states
The problem is as follows. On r600, dynpm selects power state 1 and switches between clock modes to upclock / downclock. One can see that power state 1 is a performance power state, thus it does not include a low-power clock mode. In my case, it would be preferable for dynpm to use power state 2. Ideally, users should be able to select between low-power dynpm and performance dynpm.
https://bugs.freedesktop.org/show_bug.cgi?id=33301
Martin Peres martin.peres@free.fr changed:
What |Removed |Added ---------------------------------------------------------------------------- Status|NEW |RESOLVED Resolution|--- |MOVED
--- Comment #2 from Martin Peres martin.peres@free.fr --- -- GitLab Migration Automatic Message --
This bug has been migrated to freedesktop.org's GitLab instance and has been closed from further activity.
You can subscribe and participate further through the new bug through this link to our GitLab instance: https://gitlab.freedesktop.org/drm/amd/issues/174.
dri-devel@lists.freedesktop.org
-
bugzilla-daemon＠freedesktop.org