1002:68b8 Radeon won't work with KMS enabled

[Charlie McMackin]

After upgrading to Xenial today and ensuring the xorg radeon driver was enabled, when booting both monitor lose signal input and the keyboard becomes unresponsive.

My hardware:
02:00.0 VGA compatible controller [0300]: Advanced Micro Devices, Inc. [AMD/ATI] Juniper XT [Radeon HD 5770] [1002:68b8] (prog-if 00 [VGA controller])
 Subsystem: XFX Pine Group Inc. Juniper XT [Radeon HD 5770] [1682:2990]

WORKAROUND: Use one of the following kernel parameters:
nomodeset
radeon.modeset=0

The machine will boot but the radeon driver will not load. Attempting to load radeon driver with:
sudo modprobe radeon modeset=1

after booting with modesetting off results in the same fail state explained above.

WORKAROUND:
1. Computer in OFF state, press power.
2. At boot select screen, press `e` and edit kernel line by appending:
modprobe.blacklist=radeon
3. Press `F10` to boot.
4. At login screen, click gear icon in upper right and select `Suspend`.
5. After computer is in suspended state, press any key to awaken.
6. The display does not "reawaken" and still appears off.
7. From another computer, ssh into the problem computer.
8. Execute:
sudo service lightdm stop
sudo modprobe radeon
sudo service lightdm start

ProblemType: Bug
DistroRelease: Ubuntu 16.04
Package: xorg 1:7.7+13ubuntu3
ProcVersionSignature: Ubuntu 4.4.0-18.34-generic 4.4.6
Uname: Linux 4.4.0-18-generic x86_64
.tmp.unity.support.test.1:

ApportVersion: 2.20.1-0ubuntu1
Architecture: amd64
CompizPlugins: [core,bailer,detection,composite,opengl,compiztoolbox,decor,grid,mousepoll,widget,regex,vpswitch,resize,imgpng,place,gnomecompat,wall,move,animation,fade,snap,expo,scale,unitymtgrabhandles,ezoom,session,workarounds]
CompositorRunning: None
Date: Sat Apr 9 20:02:22 2016
DistUpgraded: 2016-04-09 16:49:55,340 ERROR got error from PostInstallScript ./xorg_fix_proprietary.py (g-exec-error-quark: Failed to execute child process "./xorg_fix_proprietary.py" (No such file or directory) (8))
DistroCodename: xenial
DistroVariant: ubuntu
DkmsStatus:
 vboxhost, 5.0.12, 4.2.0-35-generic, x86_64: installed
 vboxhost, 5.0.12, 4.4.0-18-generic, x86_64: installed
ExtraDebuggingInterest: Yes
GpuHangFrequency: Continuously
GpuHangReproducibility: Yes, I can easily reproduce it
GpuHangStarted: Immediately after installing this version of Ubuntu
GraphicsCard:
 Advanced Micro Devices, Inc. [AMD/ATI] Juniper XT [Radeon HD 5770] [1002:68b8] (prog-if 00 [VGA controller])
   Subsystem: XFX Pine Group Inc. Juniper XT [Radeon HD 5770] [1682:2990]
InstallationDate: Installed on 2013-12-29 (832 days ago)
InstallationMedia: Ubuntu 13.10 "Saucy Salamander" - Release amd64 (20131016.1)
MachineType: System manufacturer System Product Name
ProcKernelCmdLine: BOOT_IMAGE=/boot/vmlinuz-4.4.0-18-generic root=UUID=e7310b2e-03b9-4e94-ba8a-d384330592cf ro quiet splash radeon.modeset=0
Renderer: Software
SourcePackage: xorg
Symptom: display
Title: Xorg freeze
UpgradeStatus: Upgraded to xenial on 2016-04-09 (0 days ago)
dmi.bios.date: 04/29/2010
dmi.bios.vendor: American Megatrends Inc.
dmi.bios.version: 0519
dmi.board.asset.tag: To Be Filled By O.E.M.
dmi.board.name: P5N7A-VM
dmi.board.vendor: ASUSTeK Computer INC.
dmi.board.version: Rev 1.xx
dmi.chassis.asset.tag: Asset-1234567890
dmi.chassis.type: 3
dmi.chassis.vendor: Chassis Manufacture
dmi.chassis.version: Chassis Version
dmi.modalias: dmi:bvnAmericanMegatrendsInc.:bvr0519:bd04/29/2010:svnSystemmanufacturer:pnSystemProductName:pvrSystemVersion:rvnASUSTeKComputerINC.:rnP5N7A-VM:rvrRev1.xx:cvnChassisManufacture:ct3:cvrChassisVersion:
dmi.product.name: System Product Name
dmi.product.version: System Version
dmi.sys.vendor: System manufacturer
version.compiz: compiz 1:0.9.12.2+16.04.20160331-0ubuntu1
version.ia32-libs: ia32-libs N/A
version.libdrm2: libdrm2 2.4.67-1
version.libgl1-mesa-dri: libgl1-mesa-dri 11.2.0-1ubuntu1
version.libgl1-mesa-dri-experimental: libgl1-mesa-dri-experimental N/A
version.libgl1-mesa-glx: libgl1-mesa-glx 11.2.0-1ubuntu1
version.xserver-xorg-core: xserver-xorg-core 2:1.18.3-1ubuntu2
version.xserver-xorg-input-evdev: xserver-xorg-input-evdev 1:2.10.1-1ubuntu2
version.xserver-xorg-video-ati: xserver-xorg-video-ati 1:7.7.0-1
version.xserver-xorg-video-intel: xserver-xorg-video-intel 2:2.99.917+git20160325-1ubuntu1
version.xserver-xorg-video-nouveau: xserver-xorg-video-nouveau 1:1.0.12-1build2
xserver.bootTime: Sat Apr 9 19:49:46 2016
xserver.configfile: default
xserver.logfile: /var/log/Xorg.0.log
xserver.outputs:

xserver.version: 2:1.18.3-1ubuntu2

[Jonathan (z3-jonathan)]

I read you Phoronix comment while updating my own copy from 14.04. Personal update went fine while using almost same card (HD 5750.) but was already using radeon driver.

From what I gather you were using catalyst before update. Any idea if radeon driver worked with older version?

X with disabled KMS isn't supported. Your X logs essentially are from trying this.

Something to try;
Add "modprobe.blacklist=radeon" to GRUB command line.
From terminal;
sudo modprobe radeon ; sleep 1 ; dmesg > dmesg.log
If no errors startx.

[Charlie McMackin (charliemac+bugs)]

Yes, Catalyst has been the only driver that has been working for me so far. I don't believe the radeon driver was working before either, I remember having a similar problem when I upgraded to 15.10.

Thanks for the debugging tip. However it didn't work. Immediately after running the modprobe command the screens go blank, lose signal input, and then after about 5 seconds the machine reboots. After logging in again, I found that dmesg didn't have a chance to pipe its output to the file either.

Definitely seems to be a bug closer to the kernel / drm

[broadcast23 (kenn06401)]

I have an A8 3820 AMD cpu (HD 6550 graphics) I have the same problem. First I want to say that I have been booting a usb hdd since 16.04 alpha. when I booted kernal 4.4.0-10 no problem, but I then booted to 4.4.0-12 that's when the problem started. Then I waited for the release of 16.04, there was no change. I just booted the live dvd using nomodset option and the screen showed an error "no ums support on radeon module." I don't have any other driver installed except what's in the kernel.

[Launchpad Janitor (janitor)]

Status changed to 'Confirmed' because the bug affects multiple users.

[Luis (luisepbueno)]

I had to set radeon.modeset=0 to be able to boot. That prevents the module from loading at startup because it needs modeset enabled.

After that, I tried to load the module with:
# modprobe radeon modeset=1
Then I got the black screen.

I logged in to the system using SSH. The system became extremely slow after loading the module.

The attached file is the output of dmesg after loading the radeon module. It was loaded at line 1061.

[Charlie McMackin (charliemac+bugs)]

Is there anything more I can do to help debug this? In my latest attempt to get any useful information, I removed my usb keyboard in favor of a ps2 one. On boot and subsequent fail, the caps lock and scroll lock flash 10 times. I thought maybe getting a serial console going would help but this computer doesn't have a db9/rs232 port and it appears to be failing before the USB subsystem is initialized so USB serial console appears to be a no-go.

I'm pretty new / green on debugging at this low a level, but I'm willing to give it a go if anyone has ideas to try...

[penalvch (penalvch)]

Charlie McMackin, to clarify, which Ubuntu release did you update from where this problem did not occur?

[Luis (luisepbueno)]

I upgrade from 15.10 to 16.04.

[penalvch (penalvch)]

Luis, in order to track your issue and hardware, it is helpful if you filed a new report with Ubuntu 15.10 by ensuring you have the package xdiagnose installed, and that you click the Yes button for attaching additional debugging information running the following from a terminal:
ubuntu-bug xorg

Also, please feel free to subscribe me to it.

For more on why this is helpful, please see https://wiki.ubuntu.com/ReportingBugs.

[Charlie McMackin (charliemac+bugs)]

Chistopher Penalver, it is hard to say as this is the first release where I can no longer use Catalyst -- I have been using the the binary drivers before this release. I remember my upgrading from 15.04 to 15.10 having a similar looking issue until I reinstalled the binary drivers. I never tried modesetting kernel parameters at that time so I can't say 100% if the error was the same.

[penalvch (penalvch)]

Charlie McMackin, could you please provide the missing information from https://wiki.ubuntu.com/DebuggingKernelBoot ?

[Charlie McMackin (charliemac+bugs)]

I'm still unable to boot following the linked directions by removing the vt_handoff, quiet, and splash kernel params and also setting gfxmode to "text". It follows nearly the same behavior black screen, constant hdd activity light on -- the monitor however still detects an input signal, it's just black.

I can boot by using the linked method but also supplying radeon.modeset=0 flag. I'm attaching the dmesg log from that boot up. I'm assuming dmesg output is the "missing information" requested. If there is something else missing please let me know.

[penalvch (penalvch)]

Charlie McMackin, to advise, the article provides information on how to obtain details from a session when the issue occurs.

Posting logs from a working session unfortunately doesn't provide what is needed.

[Charlie McMackin (charliemac+bugs)]

"debug break" => black screen, not responsive, no initramfs shell
"debug break=modules" => black screen, not responsive, no initramfs shell
"debug break=top" => initramfs shell but I haven't figured out how to mount a read-write filesystem to persist the initramfs.debug or dmesg output, yet.

[penalvch (penalvch)]

Charlie McMackin, when in initramfs can you mount it via:
mount -o remount,rw /root

[Charlie McMackin (charliemac+bugs)]

I receive:

mount: can't find /root in /proc/mounts

`more /proc/mounts` shows:

rootfs / rootfs rw 0 0
sysfs /sys sysfs rw,nosuid,nodev,noexec,relatime 0 0
proc /proc proc rw,nosuid,nodev,noexec,relatime 0 0
udev /dev devtmpfs rw,nosuid,relatime,size=4067424k,nr_inodes-1016856,mode=755 0 0
devpts /dev/pts devpts rw,nosuid,noexec,relatime,gid=5,mode=620,ptmxmode=000 0 0
tmpfs /run tmpfs rw,nosuid,noexec,relatime,size=817336k,mode=755 0 0

this is at initramfs shell on break=top

[Alberto (albertop)]

I have the same problem after upgrading to 16.04 from 14.04.
I am using a Kaveri APU, that is supposed to work with both "radeon" and "amdgpu" drivers.
The "radeon" driver doesn't work: after the very initial steps of the bootstrap, the screen goes black, although a may access the system via SSH.
Setting "nomodeset" prevents the "radeon" driver to start, reverting back to basic driver.
However, the new "amdgpu" driver seems working fine.
I managed to fix the issue in the following way:
1. blacklist "radeon", by creating a new file /etc/modprobe.d/blacklist-radeon.conf having a single line "blacklist radeon";
2. force "amdgpu", by appending to /etc/modules the line "amdgpu";
3. setting the screen resolution in grub, by uncomment the line GRUB_GFXMODE, e.g. "GRUB_GFXMODE=1280x1024";
4. create a new file /etc/X11/xorg.conf, having a block
     Section "Device"
       Identifier "VideoCard"
       DriveR "amdgpu"
     EndSection
5. update initramfs, via "update-initramfs -u -k all" with sudo;
6. update grub configuration, via "update-grub" with sudo.
Maybe other user having a GCN-enabled AMD video card may fix the issue in the same way.

[seeker5528 (seeker5528)]

Don't know where mode setting broke, had my box set aside for a while, so didn't get all my normal updating in.

I'm on Yakkety.

The most recent Ubuntu kernel that I have on my machine that did KMS was 4.3.0-5.16

None of the 4.4 kernels images work without adding nomodeset to the grub options.

Here is a listing showing the versions of the Ubuntu 4.4 kernel images I have installed.

  Installed: 4.4.0-37.56
  Installed: 4.4.0-21.37
  Installed: 4.4.0-22.40
  Installed: 4.4.0-9136.55
  Installed: 4.4.0-25.44
  Installed: 4.4.0-33.52
  Installed: 4.4.0-18.34

Currently have these Debian kernel images installed and mode setting works with both of them.

seeker@ubuntu:~$ apt policy linux-image-4*amd64 | grep Installed -B 1

WARNING: apt does not have a stable CLI interface. Use with caution in scripts.

linux-image-4.4.0-1-amd64:
  Installed: 4.4.2-1
--
linux-image-4.5.0-2-amd64:
  Installed: 4.5.5-1

Before installing 4.4.2-1 I tried 4.4.6-1 and it also did mode setting.

I have and AMD FM1 chip on an MSI A55M-P33 Motherboard. Video hardware is listed this way by lspci.

VGA compatible controller [0300]: Advanced Micro Devices, Inc. [AMD/ATI] Sumo [Radeon HD 6410D] [1002:9644] (prog-if 00 [VGA controller])
Subsystem: Micro-Star International Co., Ltd. [MSI] Sumo [Radeon HD 6410D] [1462:7786]

[Launchpad Janitor (janitor)]

Status changed to 'Confirmed' because the bug affects multiple users.

[seeker5528 (seeker5528)]

Symptoms are as reported by others. Whether it's during boot or attempting to enable after boot up. Screen goes black, reports no video, then goes to sleep.

When it happens on boot, I can hit 'Ctrl'+'Alt'+'Del' or hit the power button and the machine will reboot or shut down so things are still running, just the video output that's dead.

[penalvch (penalvch)]

seeker5528, it will help immensely if you filed a new report with Ubuntu by ensuring you have the package xdiagnose installed, and that you click the Yes button for attaching additional debugging information running the following from a terminal:
ubuntu-bug xorg

Also, please feel free to subscribe me to it.

For more on why this is helpful, please see https://wiki.ubuntu.com/ReportingBugs.

[seeker5528 (seeker5528)]

Christopher, I filed the report and subscribled you.

For others following here.

Did also find out in my case it happens when the monitor is attached with a VGA cable.

It does not happen if I use a DVI cable to connect the monitor.

Still have to test have to test with monitor connected on HDMI port.

[Charlie McMackin (charliemac+bugs)]

Thanks for the information seeker5528. Unfortunately for me, I'm not seeing similar results. My XFX 5570 only has DVI, HDMI and HDCP outputs. I just tested both DVI and HDMI alone and still the same issue for each.

[Charlie McMackin (charliemac+bugs)]

I meant XFX 5770

[penalvch (penalvch)]

Charlie McMackin, in order to allow additional upstream developers to examine the issue, at your earliest convenience, could you please test the latest upstream kernel available from http://kernel.ubuntu.com/~kernel-ppa/mainline/?C=N;O=D ? Please keep in mind the following:
1) The one to test is at the very top line at the top of the page (not the daily folder).
2) The release names are irrelevant.
3) The folder time stamps aren't indicative of when the kernel actually was released upstream.
4) Install instructions are available at https://wiki.ubuntu.com/Kernel/MainlineBuilds .

If testing on your main install would be inconvenient, one may:
1) Install Ubuntu to a different partition and then test this there.
2) Backup, or clone the primary install.

If the latest kernel did not allow you to test to the issue (ex. you couldn't boot into the OS) please make a comment in your report about this, and continue to test the next most recent kernel version until you can test to the issue. Once you've tested the upstream kernel, please comment on which kernel version specifically you tested. If this issue is fixed in the mainline kernel, please add the following tags by clicking on the yellow circle with a black pencil icon, next to the word Tags, located at the bottom of the report description:
kernel-fixed-upstream
kernel-fixed-upstream-X.Y-rcZ

Where X, and Y are the first two numbers of the kernel version, and Z is the release candidate number if it exists.

If the mainline kernel does not fix the issue, please add the following tags:
kernel-bug-exists-upstream
kernel-bug-exists-upstream-X.Y-rcZ

Please note, an error to install the kernel does not fit the criteria of kernel-bug-exists-upstream.

Also, you don't need to apport-collect further unless specifically requested to do so.

It is most helpful that after testing of the latest upstream kernel is complete, you mark this report Status Confirmed.

Lastly, to keep this issue relevant to upstream, please continue to test the latest mainline kernel as it becomes available.

Thank you for your help.

[Charlie McMackin (charliemac+bugs)]

Good news: I have a had a small eureka moment that results my computer fully booted with graphics hardware fully accelerated. Bad news: it is still a workaround -- I still can't push power button and boot up normally.

Here is the workaround sequence -- it requires another computer that can ssh in:

1. Computer in OFF state, press power.
2. At boot select screen, press `e` and edit kernel line by appending `modprobe.blacklist=radeon`
3. Press `F10` to boot
4. At login screen, click gear icon in upper right and select `Suspend`
5. After computer is in suspended state, press any key to awaken.
6. The display does not "reawaken" and still appears off.
7. From another computer, ssh into the first.
8. Issue `sudo service lightdm stop`
9. Issue `sudo modprobe radeon`. This should succeed! \o/
10. Issue `sudo service lightdm start`

Now the computer should be using correct drm drivers and have full resolution display. Furthermore, Suspend and Wake work successfully.

I discovered this using ubuntu mainline kernel 4.8.0-rc5 but have verified it works with mainline 4.8.6 and also with 4.4.0-45. Unfortunately, I am also using the oibaf/graphics-drivers PPA with more recent mesa and libdrm drivers. I plan to test again after removing the PPA but it may take a while. If anyone with this problem is still watching, I'd be curious to hear if my workaround steps work for you.

[Launchpad Janitor (janitor)]

[Expired for xorg (Ubuntu) because there has been no activity for 60 days.]