Radeon DPM Support Should Now Be In Good Shape

[ChrisXY]

Intel + Radeon works well with randr. What doesn't work with 3.11rc2 and rc3 is vgaswitcheroo. At least for me.

[droidhacker]

Not to belittle the wonderful dpm work, but has there been any progress made on the UVD for RS780/880 front?

[chithanh]

I think regarding RS780 UVD support, this post from Deathsimple is still valid.

[AJenbo]

This is amazing!
The open source AMD drivers are shaping up!
I can't wait until they rival the proprietary drivers.

Then why not make a donation towards that goal:

http://phoronix.com/forums/showthread.php?82762-Crowd-Funding-Mesa-Driver-Development&p=346365#post346365

To clarify, not only will I match your donation, but I will also make one my self. The least you could do is respond with why you do not want to donate.

[droidhacker]

Then why not make a donation towards that goal:

http://phoronix.com/forums/showthread.php?82762-Crowd-Funding-Mesa-Driver-Development&p=346365#post346365

To clarify, not only will I match your donation, but I will also make one my self. The least you could do is respond with why you do not want to donate.

I don't want to donate there, because I "donate" every single time I buy AMD hardware.

[Hamish Wilson]

Will the Radeon DPM support be made default by the time Linux 3.11 is given its final release or will that have to wait until a later Kernel version?

[duby229]

Will the Radeon DPM support be made default by the time Linux 3.11 is given its final release or will that have to wait until a later Kernel version?

This kernel version (3.11) it will be disabled by default. Being available to enable on a stable kernel release means that anybody that uses a distro which ships that kernel will be able to test out the code easily with only needing to enable a boot parameter. It's very likely it will be enabled by default in 3.12 though.

Still though only one kernel cycle is not a whole lot of testing.

[ChrisXY]

Aaand with the latest update, vgaswitcheroo works again. I think.

Thanks.

[mudig]

Uh, temperatures with DPM (3.11rc3) on my 5870 during idle are 10?C higher than with Catalyst on Windows and the the "low" profile of the previous power_method. Anybody know what's up?

[Death Knight]

I think DPM doesn't work well.
Just look GPU clock 10 times per seccond and divide by ten
After using computer for a while
Power state of my GPU HD6850 is like that:

P0: 703 Seconds
P1: 119 Seconds
P2: 9372 Seconds !!!

Means more than %90 of the time, my GPU is in High power state.

Indeed with latest Mesa and sitting on Idle state on my GTK3 desktop Power State switching P0 to P2 without apparent reason.

I watch with my GPU with readeontop and don't see any bottleneck on my GPU. There are no use but power state jumps to P2.

I think power switching algorithm need some tune up.

Edit: You can inspect your GPU power states with this Python script

#!/usr/bin/env python
import time
a=[0,0,0]
while 1:
   time.sleep(0.1)
   state=open('/sys/kernel/debug/dri/0/radeon_pm_info').read()
   x=int(state[state.find('power level')+12])
   if x >=0 and x <=2:
      a[x]+=1
   else:
      print state
   print "P0:",a[0]/10 ,"P1:", a[1]/10, "P2:", a[2]/10