Radeon Graphics Problems With CentOS 6.6

Home » CentOS » Radeon Graphics Problems With CentOS 6.6
CentOS 11 Comments

Hello,

I saw in the CentOS 6.6 release notes that the Xorg ABI has changed but that “This should not concern anyone who uses the default drivers shipped with CentOS.” I’m using the ATI drivers in Base but X still fails when I
boot a CentOS 6.6 kernel. I was hoping this is a known problem that would get fixed with a kernel update but I’m still having problems with the
2.6.32-504.1.3.el6.x86_64 kernel. Relevant info below. Any ideas?

Thanks, Darby

[root@c6 ~]# lspci | grep VGA
00:01.0 VGA compatible controller: Advanced Micro Devices, Inc. [AMD/ATI]
BeaverCreek [Radeon HD 6530D]
[root@c6 ~]# uname -a Linux c6 2.6.32-431.29.2.el6.x86_64 #1 SMP Tue Sep 9 21:36:05 UTC 2014
x86_64 x86_64 x86_64 GNU/Linux
[root@c6 ~]# yum list xorg-x11-drv-ati.x86_64
xorg-x11-drv-ati-firmware.noarch Loaded plugins: fastestmirror, refresh-packagekit, security Loading mirror speeds from cached hostfile
* base: repos.dfw.quadranet.com
* epel: mirror.compevo.com
* extras: repos.dfw.quadranet.com
* updates: repos.dfw.quadranet.com Installed Packages xorg-x11-drv-ati.x86_64 7.3.99-2.el6
@base xorg-x11-drv-ati-firmware.noarch 7.3.99-2.el6
@base
[root@c6 ~]#

Log messages containing “radeon” from booting into the CentOS 6.6 kernel:

c6 kernel: [drm] radeon kernel modesetting enabled. c6 kernel: fb: conflicting fb hw usage radeondrmfb vs EFI VGA – removing generic driver c6 kernel: radeon 0000:00:01.0: PCI INT A -> GSI 18 (level, low) -> IRQ 18
c6 kernel: radeon 0000:00:01.0: VRAM: 512M 0x0000000000000000 –
0x000000001FFFFFFF (512M used)
c6 kernel: radeon 0000:00:01.0: GTT: 1024M 0x0000000020000000 –
0x000000005FFFFFFF
c6 kernel: [drm] radeon: 512M of VRAM memory ready c6 kernel: [drm] radeon: 1024M of GTT memory ready. c6 kernel: radeon 0000:00:01.0: firmware: requesting radeon/SUMO_pfp.bin c6 kernel: radeon 0000:00:01.0: firmware: requesting radeon/SUMO_me.bin c6 kernel: radeon 0000:00:01.0: firmware: requesting radeon/SUMO_rlc.bin c6 kernel: [drm] radeon: dpm initialized c6 kernel: radeon 0000:00:01.0: firmware: requesting radeon/SUMO_uvd.bin c6 kernel: radeon 0000:00:01.0: WB enabled c6 kernel: radeon 0000:00:01.0: fence driver on ring 0 use gpu addr
0x0000000020000c00 and cpu addr 0xffff8801b6e81c00
c6 kernel: radeon 0000:00:01.0: fence driver on ring 3 use gpu addr
0x0000000020000c0c and cpu addr 0xffff8801b6e81c0c c6 kernel: radeon 0000:00:01.0: fence driver on ring 5 use gpu addr
0x0000000000072118 and cpu addr 0xffffc900229b2118
c6 kernel: radeon 0000:00:01.0: radeon: using MSI. c6 kernel: [drm] radeon: irq initialized. c6 kernel: [drm] Radeon Display Connectors c6 kernel: fbcon: radeondrmfb (fb0) is primary device c6 kernel: radeon 0000:00:01.0: fb0: radeondrmfb frame buffer device c6 kernel: radeon 0000:00:01.0: registered panic notifier c6 kernel: [drm] Initialized radeon 2.37.0 20080528 for 0000:00:01.0 on minor 0
c6 kernel: powernow-k8: Found 1 AMD A6-3600 APU with Radeon(tm) HD Graphics
(4 cpu cores) (version 2.20.00)

Log messages from booting into the CentOS 6.5 kernel:

c6 kernel: [drm] radeon kernel modesetting enabled. c6 kernel: fb: conflicting fb hw usage radeondrmfb vs EFI VGA – removing generic driver c6 kernel: radeon 0000:00:01.0: PCI INT A -> GSI 18 (level, low) -> IRQ 18
c6 kernel: radeon 0000:00:01.0: VRAM: 512M 0x0000000000000000 –
0x000000001FFFFFFF (512M used)
c6 kernel: radeon 0000:00:01.0: GTT: 512M 0x0000000020000000 –
0x000000003FFFFFFF
c6 kernel: [drm] radeon: 512M of VRAM memory ready c6 kernel: [drm] radeon: 512M of GTT memory ready. c6 kernel: platform radeon_cp.0: firmware: requesting radeon/SUMO_pfp.bin c6 kernel: platform radeon_cp.0: firmware: requesting radeon/SUMO_me.bin c6 kernel: platform radeon_cp.0: firmware: requesting radeon/SUMO_rlc.bin c6 kernel: radeon 0000:00:01.0: WB enabled c6 kernel: radeon 0000:00:01.0: fence driver on ring 0 use gpu addr
0x0000000020000c00 and cpu addr 0xffff8801b83aec00
c6 kernel: radeon 0000:00:01.0: fence driver on ring 3 use gpu addr
0x0000000020000c0c and cpu addr 0xffff8801b83aec0c c6 kernel: radeon 0000:00:01.0: radeon: using MSI. c6 kernel: [drm] radeon: irq initialized. c6 kernel: [drm] Radeon Display Connectors c6 kernel: [drm] radeon: power management initialized c6 kernel: fbcon: radeondrmfb (fb0) is primary device c6 kernel: radeon 0000:00:01.0: fb0: radeondrmfb frame buffer device c6 kernel: radeon 0000:00:01.0: registered panic notifier c6 kernel: [drm] Initialized radeon 2.30.0 20080528 for 0000:00:01.0 on minor 0
c6 kernel: powernow-k8: Found 1 AMD A6-3600 APU with Radeon(tm) HD Graphics
(4 cpu cores) (version 2.20.00)
c6 kernel: CPU0: AMD A6-3600 APU with Radeon(tm) HD Graphics stepping 00

11 thoughts on - Radeon Graphics Problems With CentOS 6.6

  • Thanks for the suggestion but unfortunately not. When I boot a 6.6
    kernel with the fglrx driver, my machine completely locks up. with the stock ATI driver X would lock but I could still ping/ssh to the machine. I can still boot using the fglrx driver with a 6.5 kernel
    (but performance seems worse than before…). Any other ideas?

  • I’m still struggling with this. I was hoping to boot a CentOS 6.6
    livecd to see if this is specific to my installation or a general problem with my hardware. I can’t find a 6.6 livecd on any of the several mirrors I’ve checked or on the CentOS vault. Does anyone know if one is available somewhere. Other ideas are welcome too.

  • I thought about trying CentOS 7. But in the end that doesn’t really help me since I’m trying to keep my current CentOS 6 installation working. I don’t really have the time/energy to reinstall everyting under CentOS 7 right now.

    I was able to get a 6.6 LiveCD built myself. The liveCD has the same problem – X doesn’t work. I can boot the “Basic Video” option from the LiveCD and I can at least get to the desktop but there are several odd video problems when I do that. But I think that means that there is a genuine problem with Xorg on my hardware. Does anyone know who I should submit this bug to – AMD or Xorg?

    Any other ideas for workarounds in the mean time?

  • Two things I think you should consider doing…
    1) in your next post (and any bugs you open), include the `uname -a` of the last known to work correctly CentOS/RHEL kernel, and the `uname -a` of the first known to not work correctly CentOS/RHEL kernel and the respective versions of radeon &| flrgx drivers.

    2) try out elrepo’s el6 kernel-lt & kernel-ml and see if they work correctly for you. I suggest this, because for me it did workaround an issue with intel chips (I guess I need to see what it will take to get the elrepo folks to make just a back-ported intel driver from 3.17 [first known to work again, 3.16 does not]) . http://lists.CentOS.org/pipermail/CentOS/2014-November/147579.html

    Even when this disclaimer is not here:
    I am not a contracting officer. I do not have authority to make or modify the terms of any contract.

  • This is correct .. we rebuild the RHEL 6.6 kernel source code as is, if this is a kernel problem, we don’t change it.

    But .. IF the kernel does NEED a change, then the CentOSPlus kernel would be the place where we have the ability to add patches and change something. We don’t change the base OS kernel except to remove branding.

    AKemi Yagi is our CentOS Plus maintainer, and bug/patch requests can be entered here:

    http://bugs.CentOS.org/view.php?id=4586

    But, we do not provide technical support .. we have a community forum and bugs system to allow the community to discuss and solve problems. We will try/roll in patches (into CentOS Plus kernel) to try and fix problems if a solution is proposed by the community. However, we don’t do the research or development.

    What I am trying to convey is that CentOS rebuilds upstream source code and provides resources for the community to solve their own problems. CentOS does not provide SLA level support. That is what RHEL is for.

    We will be glad to help the community find the problem, and feed back the fix upstream to Red Hat so they can fix it in RHEL and then we can roll it into our main OS kernel. And we will provide a temporary fix in the CentOS Plus kernel.

  • I sent this last night but it didn’t show up in the archives. Trying again without the attachment. Are messages with attachments blocked on this list?