dix/dispatch.c:4049: AttachOffloadGPU: Assertion `new->current_master == pScreen' failed.

[Paul Menzel]

Dear Michel,


On 2017-08-15 03:19, Michel Dänzer wrote:
> On 15/08/17 02:48 AM, Paul Menzel wrote:
>> On 08/14/17 03:15, Michel Dänzer wrote:
>>> On 11/08/17 01:02 AM, Paul Menzel wrote:
>>>> On 08/10/17 15:14, Paul Menzel wrote:
>>>> 
>>>>> Xorg:
>>>>> /dev/shm/bee-root/xorg-server/xorg-server-1.19.3-0/source/dix/dispatch.c:4049:
>>>>> 
>>>>> AttachOffloadGPU: Assertion `new->current_master == pScreen' 
>>>>> failed.
>>>>> ```
>>>>> 
>>>>> The USB keyboard does not work, and the system has to be controlled
>>>>> remotely over SSH. Plugging in the DisplayPort cable into the 
>>>>> internal
>>>>> Intel graphics ports the Linux messages can be seen, but nothing 
>>>>> else
>>>>> works. Stopping GDM does not bring back the TTY screen. The 
>>>>> keyboard
>>>>> still does work. Plugging the USB cable out and back in doesn’t fix 
>>>>> it
>>>>> either.
>>>>> 
>>>>> Trying to disable the internal Intel graphics card, did not fix the
>>>>> problem either.
>>>> 
>>>> Blacklisting the module *i915* works around the problem.
>>>> 
>>>> Please tell me if you want me to create a bug report for this issue,
>>>> and, if yes, what part in the graphics stack this problem belongs 
>>>> to.
>>> 
>>> It's hard to be sure without seeing a backtrace of the assertion
>>> failure, but I'd say start at
>>> https://bugs.freedesktop.org/enter_bug.cgi?product=xorg&component=Server/Ext/RandR.
>>> 
>> 
>> Do you know why the stack trace is not present in the X server log?
> 
> Because your xserver build doesn't have
> https://cgit.freedesktop.org/xorg/xserver/commit/?id=27a6b9f7c84c914d0f5909ec1069d72f5035bc04
> (it's only in Git master yet).

With this patch the stack trace is indeed in the log.

```
[…]
Xorg: 
/dev/shm/bee-root/xorg-server/xorg-server-1.19.3-1/source/dix/dispatch.c:4049: 
AttachOffloadGPU: Assertion `new->current_master == pScreen' failed.
(EE)
(EE) Backtrace:
(EE) 0: /usr/libexec/Xorg (xorg_backtrace+0x41) [0x580461]
(EE) 1: /usr/libexec/Xorg (0x400000+0x183e09) [0x583e09]
(EE) 2: /lib/libpthread.so.0 (0x7f6f957d3000+0x11a50) [0x7f6f957e4a50]
(EE) 3: /lib/libc.so.6 (gsignal+0x104) [0x7f6f954682f4]
(EE) 4: /lib/libc.so.6 (abort+0x16a) [0x7f6f9546975a]
(EE) 5: /lib/libc.so.6 (0x7f6f95435000+0x2c027) [0x7f6f95461027]
(EE) 6: /lib/libc.so.6 (0x7f6f95435000+0x2c0d2) [0x7f6f954610d2]
(EE) 7: /usr/libexec/Xorg (AttachOffloadGPU+0x6e) [0x4357fe]
(EE) 8: /usr/libexec/Xorg (0x400000+0xa8b76) [0x4a8b76]
(EE) 9: /usr/libexec/Xorg (InitOutput+0x5b3) [0x4775b3]
(EE) 10: /usr/libexec/Xorg (0x400000+0x38cb6) [0x438cb6]
(EE) 11: /lib/libc.so.6 (__libc_start_main+0xf0) [0x7f6f95455520]
(EE) 12: /usr/libexec/Xorg (_start+0x2a) [0x42421a]
(EE)
(EE)
Fatal server error:
(EE) Caught signal 6 (Aborted). Server aborting
(EE)
(EE)
[…]
```

>> I was able to get a stack trace as root and with GDB. Issue #102222
>> (dix/dispatch.c:4049: AttachOffloadGPU: Assertion `new->current_master
>> == pScreen' failed.) tracks this problem now, and the stack traces are
>> attached there.
> 
> Thanks, though it looks like it might actually be a downstream issue.

I don’t have my login credentials handy right now, so I attach the X 
server
here, and go looking for the credentials.


Kind regards,

Paul
-------------- next part --------------
An embedded and charset-unspecified text was scrubbed...
Name: 20170815?Xorg.5.log
URL: <https://lists.freedesktop.org/archives/amd-gfx/attachments/20170815/da4bf3a3/attachment-0001.ksh>