Announcement

Collapse
No announcement yet.

Radeon DPM Support Should Now Be In Good Shape

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

  • #41
    Is this going to be enabled by default in kernel 3.12? I cant wait

    Comment


    • #42
      Originally posted by edoantonioco View Post
      Is this going to be enabled by default in kernel 3.12? I cant wait
      Kernel 3.11 already has full dpm support, you have to add "radeon.dpm=1" to the kernel command line.

      Comment


      • #43
        Originally posted by gradinaruvasile View Post
        Kernel 3.11 already has full dpm support, you have to add "radeon.dpm=1" to the kernel command line.
        Then that wouldn't be enabled by default, now would it?

        Comment


        • #44
          Originally posted by gradinaruvasile View Post
          Kernel 3.11 already has full dpm support, you have to add "radeon.dpm=1" to the kernel command line.
          Thanks for the info , but seems to be than I have more FPS in games like team fortress 2 and portal having this feature disabled, so, now I have it disabled
          (my GPU is Radeon HD 4200). Thats weird.

          Comment


          • #45
            Originally posted by edoantonioco View Post
            Thanks for the info , but seems to be than I have more FPS in games like team fortress 2 and portal having this feature disabled, so, now I have it disabled
            (my GPU is Radeon HD 4200). Thats weird.
            Did you download the appropiate firmware and put it in /lib/firmware/radeon ?
            Run dmesg | grep dpm in terminal, if it failed to intialize then you need the right ucode...

            I'm not sure which you should d/l... Seeing how you have a HD4000 series card you should download:
            RV710_smc.bin
            RV730_smc.bin
            RV740_smc.bin
            RV770_smc.bin

            and put it all in /lib/firmware/radeon
            Then reboot and run dmesg | grep dpm to verify if DPM intialized.
            Last edited by Bushido_root; 18 September 2013, 05:07 PM.

            Comment


            • #46
              @agd5f or bridgman

              Is the code that controls the fan speed of the card in the firmware or in mesa/kernel/somewhere open? I am just curious about something

              Thanks in advance.

              Comment


              • #47
                Originally posted by 89c51 View Post
                Is the code that controls the fan speed of the card in the firmware or in mesa/kernel/somewhere open?
                It depends on the card. The gpu has a built in fan controller. If the oem uses that, it's controlled by the fan controller hw. If the oem uses an external fan control chip, that controls the fan. The vbios loads a default fan profile when the asic is initialized. It raises the fan speed as the temperature increases and reduces it as the temperature decreases. There is no support for manual fan control at the moment.

                Comment


                • #48
                  Originally posted by agd5f View Post
                  It depends on the card. The gpu has a built in fan controller. If the oem uses that, it's controlled by the fan controller hw. If the oem uses an external fan control chip, that controls the fan. The vbios loads a default fan profile when the asic is initialized. It raises the fan speed as the temperature increases and reduces it as the temperature decreases. There is no support for manual fan control at the moment.
                  Thanks for the answer. I was just curious about the control algorithm of the thing.

                  Comment


                  • #49
                    I would love to use this feature as my card (ASUS EAH6870 DirectCU/2DI2S/1GD5) is really noisy but it freezes my PC when enabled and the monitor goes to sleep mode.
                    If I have radeon compiled as module and firmware in /lib/firmware/radeon freez occurs when the module loads and if I have it with firmware in kernel freez is right at load of kernel.
                    I have enabled intel integrated card to see if there is any output but there is nothing on the screen.
                    Do I need more than radeon module and firmware from http://people.freedesktop.org/~agd5f/radeon_ucode/ ?

                    Comment


                    • #50
                      Also this.

                      Sep 19 16:32:18 mainland kernel: radeon 0000:01:00.0: GPU lockup CP stall for more than 10000msec
                      Sep 19 16:32:18 mainland kernel: radeon 0000:01:00.0: GPU lockup (waiting for 0x000000000011333f)
                      Sep 19 16:32:18 mainland kernel: radeon 0000:01:00.0: failed to get a new IB (-35)
                      Sep 19 16:32:18 mainland kernel: [drm:radeon_cs_ib_chunk] *ERROR* Failed to get ib !
                      Sep 19 16:32:18 mainland kernel: radeon 0000:01:00.0: Saved 695 dwords of commands on ring 0.
                      Sep 19 16:32:18 mainland kernel: radeon 0000:01:00.0: GPU softreset: 0x00000009
                      Sep 19 16:32:18 mainland kernel: radeon 0000:01:00.0: GRBM_STATUS = 0xB2703828
                      Sep 19 16:32:18 mainland kernel: radeon 0000:01:00.0: GRBM_STATUS_SE0 = 0x1C000007
                      Sep 19 16:32:18 mainland kernel: radeon 0000:01:00.0: GRBM_STATUS_SE1 = 0x08000007
                      Sep 19 16:32:18 mainland kernel: radeon 0000:01:00.0: SRBM_STATUS = 0x200000C0
                      Sep 19 16:32:18 mainland kernel: radeon 0000:01:00.0: SRBM_STATUS2 = 0x00000000
                      Sep 19 16:32:18 mainland kernel: radeon 0000:01:00.0: R_008674_CP_STALLED_STAT1 = 0x00000000
                      Sep 19 16:32:18 mainland kernel: radeon 0000:01:00.0: R_008678_CP_STALLED_STAT2 = 0x40000000
                      Sep 19 16:32:18 mainland kernel: radeon 0000:01:00.0: R_00867C_CP_BUSY_STAT = 0x00008006
                      Sep 19 16:32:18 mainland kernel: radeon 0000:01:00.0: R_008680_CP_STAT = 0x80228647
                      Sep 19 16:32:18 mainland kernel: radeon 0000:01:00.0: R_00D034_DMA_STATUS_REG = 0x44C83D57
                      Sep 19 16:32:18 mainland kernel: radeon 0000:01:00.0: GRBM_SOFT_RESET=0x00007F6B
                      Sep 19 16:32:18 mainland kernel: radeon 0000:01:00.0: SRBM_SOFT_RESET=0x00000100
                      Sep 19 16:32:18 mainland kernel: radeon 0000:01:00.0: GRBM_STATUS = 0x00003828
                      Sep 19 16:32:18 mainland kernel: radeon 0000:01:00.0: GRBM_STATUS_SE0 = 0x00000007
                      Sep 19 16:32:18 mainland kernel: radeon 0000:01:00.0: GRBM_STATUS_SE1 = 0x00000007
                      Sep 19 16:32:18 mainland kernel: radeon 0000:01:00.0: SRBM_STATUS = 0x200000C0
                      Sep 19 16:32:18 mainland kernel: radeon 0000:01:00.0: SRBM_STATUS2 = 0x00000000
                      Sep 19 16:32:18 mainland kernel: radeon 0000:01:00.0: R_008674_CP_STALLED_STAT1 = 0x00000000
                      Sep 19 16:32:18 mainland kernel: radeon 0000:01:00.0: R_008678_CP_STALLED_STAT2 = 0x00000000
                      Sep 19 16:32:18 mainland kernel: radeon 0000:01:00.0: R_00867C_CP_BUSY_STAT = 0x00000000
                      Sep 19 16:32:18 mainland kernel: radeon 0000:01:00.0: R_008680_CP_STAT = 0x00000000
                      Sep 19 16:32:18 mainland kernel: radeon 0000:01:00.0: R_00D034_DMA_STATUS_REG = 0x44C83D57
                      Sep 19 16:32:18 mainland kernel: radeon 0000:01:00.0: GPU reset succeeded, trying to resume
                      Sep 19 16:32:18 mainland kernel: [drm] PCIE gen 2 link speeds already enabled
                      Sep 19 16:32:18 mainland kernel: [drm] PCIE GART of 512M enabled (table at 0x0000000000273000).
                      Sep 19 16:32:18 mainland kernel: radeon 0000:01:00.0: WB enabled
                      Sep 19 16:32:18 mainland kernel: radeon 0000:01:00.0: fence driver on ring 0 use gpu addr 0x0000000040000c00 and cpu addr 0xffff88030ac57c00
                      Sep 19 16:32:18 mainland kernel: radeon 0000:01:00.0: fence driver on ring 3 use gpu addr 0x0000000040000c0c and cpu addr 0xffff88030ac57c0c
                      Sep 19 16:32:18 mainland kernel: radeon 0000:01:00.0: fence driver on ring 5 use gpu addr 0x0000000000072118 and cpu addr 0xffffc9000b332118
                      Sep 19 16:32:18 mainland kernel: [drm] ring test on 0 succeeded in 3 usecs
                      Sep 19 16:32:18 mainland kernel: [drm] ring test on 3 succeeded in 1 usecs
                      Sep 19 16:32:19 mainland kernel: [drm] ring test on 5 succeeded in 2 usecs
                      Sep 19 16:32:19 mainland kernel: [drm] UVD initialized successfully.
                      Sep 19 16:32:19 mainland kernel: [drm] ib test on ring 0 succeeded in 0 usecs
                      Sep 19 16:32:19 mainland kernel: [drm] ib test on ring 3 succeeded in 1 usecs
                      Sep 19 16:32:19 mainland kernel: [drm] ib test on ring 5 succeeded
                      Sep 19 16:32:19 mainland kernel: switching from power state:
                      Sep 19 16:32:19 mainland kernel: ui class: none
                      Sep 19 16:32:19 mainland kernel: internal class: boot
                      Sep 19 16:32:19 mainland kernel: caps:
                      Sep 19 16:32:19 mainland kernel: uvd vclk: 0 dclk: 0
                      Sep 19 16:32:19 mainland kernel: power level 0 sclk: 10000 mclk: 15000 vddc: 950 vddci: 950
                      Sep 19 16:32:19 mainland kernel: power level 1 sclk: 10000 mclk: 15000 vddc: 950 vddci: 950
                      Sep 19 16:32:19 mainland kernel: power level 2 sclk: 10000 mclk: 15000 vddc: 950 vddci: 950
                      Sep 19 16:32:19 mainland kernel: status: c b
                      Sep 19 16:32:19 mainland kernel: switching to power state:
                      Sep 19 16:32:19 mainland kernel: ui class: performance
                      Sep 19 16:32:19 mainland kernel: internal class: none
                      Sep 19 16:32:19 mainland kernel: caps:
                      Sep 19 16:32:19 mainland kernel: uvd vclk: 0 dclk: 0
                      Sep 19 16:32:19 mainland kernel: power level 0 sclk: 10000 mclk: 15000 vddc: 950 vddci: 950
                      Sep 19 16:32:19 mainland kernel: power level 1 sclk: 60000 mclk: 100000 vddc: 1100 vddci: 1100
                      Sep 19 16:32:19 mainland kernel: power level 2 sclk: 77500 mclk: 100000 vddc: 1150 vddci: 1100
                      Sep 19 16:32:19 mainland kernel: status: r
                      Screen blinks from time to time. But thats rare. Not sure if its driver related or something else.

                      Comment

                      Working...
                      X