Announcement

Collapse
No announcement yet.

Linux 2.6.34-rc1 Kernel Is Out w/ New Features

Collapse
X
 
  • Filter
  • Time
  • Show
Clear All
new posts

  • luksow
    replied
    I have:
    Code:
    state: PM_STATE_MINIMUM
    What does it mean? The new kernel dropped about 20C (from 80C to 60C) which is already very good result but I still have a hairdryer instead of fan (it's sooooooo annoying). Maybe with ACTIVE instead of MINIMUM it'll be better? Is there any way to change it?

    Leave a comment:


  • bridgman
    replied
    Manufacturers do this already - isn't the issue that you want them to do it with *your* Linux distro rather than just with Windows ?

    Leave a comment:


  • squirrl
    replied
    The phoronix baseline test

    PHORONIX BASELINE TEST

    Have a series of test to check basic functionality.
    Suspend / Resume, Hibernate Same test really.<br>
    Acceleration<br>
    Sound<br>
    etc.<br>

    Manufacturers could begin testing before they release new hardware.

    Leave a comment:


  • icek
    replied
    hm, with this I have
    Code:
    state: PM_STATE_ACTIVE
    , but engine clock is always on default value...

    Leave a comment:


  • Melcar
    replied
    From the grub2 menu, I just added:

    Code:
    radeon.dynpm=1
    ... to the kernel parameter line. It works on both my HD4850 and my laptop's integrated 200M.

    Leave a comment:


  • icek
    replied
    Melcar: how did you enable PM ? I tried everything but I'm still on default clocks...

    I have integrated 3200 HD, may this be the problem? Is PM supposed to work with integrated cards?

    Leave a comment:


  • Melcar
    replied
    On my laptop (200M) it works a bit better, in that it's not unbearably slow. Thing is that when the core downclocks, response becomes "jumpy"; the mouse cursor seems to get "stuck" as I move it across the screen, and window animations sometimes leave a noticeable trace behind them. This again, is due to KMS making things slow and the slower frequencies only worsening the problem. Should be better once I move to a newer xserver.

    Leave a comment:


  • Loris
    replied
    Originally posted by Melcar View Post
    As far as PM, it does seem to work. The core is downclocking fine as far as I can tell, which is making things slower than usual; KMS makes 2D slow as is, so the downclocking is making things worse .
    Mobility 3470 here. The first DRM power management patches did just that, slowing 2D to unacceptable levels. Since 2.6.33 + patches, it's become very good, with no noticeable slowness at all. Now, it seems to change core and memory frequencies very fast.

    And my notebook stopped hanging with a black screen due to overheating, as well.

    Now, I'm just hoping the current power management, or the future advancements, will be enough to cope with the summer's heat.

    Thank you developers.

    Leave a comment:


  • Melcar
    replied
    As far as PM, it does seem to work. The core is downclocking fine as far as I can tell, which is making things slower than usual; KMS makes 2D slow as is, so the downclocking is making things worse .

    Leave a comment:


  • Melcar
    replied
    With my HD4850, I can see the 4 powerstates being listed. Each one has (or should) 3 modes. This I learned before from messing around with Radeon Bios Editor. In the kernel log, state 0 and 3 each have three modes, and modes 1 and 2 show only two. Looking at the listed frequencies, it seems that the actual "3D" frequencies are missing. I see the low "2D" clocks in there, but not the high power ones. Is this a current limitation, or a problem with my particular setup?
    Also, this rc causes a weird kernel panic on my laptop (Presario V2000) and I'm unable to go past the grub2 menu.

    Leave a comment:

Working...
X