Testing the linux compatibility with the amdgpu driver and linux-doom3 fails

[Johannes Lundberg]

On Sat, Aug 18, 2018 at 8:38 PM Stefan Rumetshofer <sterum77 at gmail.com>
wrote:

>
>
> 2018-08-18 20:05 GMT+02:00 Johannes Lundberg <johalun0 at gmail.com>:
>
>>
>>
>> On Thu, Aug 16, 2018 at 8:15 PM Greg V <greg at unrelenting.technology>
>> wrote:
>>
>>>
>>>
>>> On Thu, Aug 16, 2018 at 9:42 PM, Stefan Rumetshofer
>>> <sterum77 at gmail.com> wrote:
>>> >
>>> >
>>> > Am Donnerstag, 16. August 2018 schrieb Greg V :
>>> >>
>>> >>
>>> >> On Thu, Aug 16, 2018 at 7:46 PM, Stefan Rumetshofer
>>> >> <sterum77 at gmail.com> wrote:
>>> >>> Firts of all i hope this is not the wrong mailing list.
>>> >>>
>>> >>> I was trying to run linux-doom3 from the ports collection with my
>>> >>> graphics
>>> >>> hardware but this fails with an error realted to OpenGL or so. Also
>>> >>> other
>>> >>> linux games are failing like Unreal or RTCW. It seems that there is
>>> >>> a general
>>> >>> Problem with the amd Graphics driver and the linux compatibility. I
>>> >>> also teted
>>> >>> the game wit the intel driver on my SandyBridge GPU with no error.
>>> >> Hi,
>>> >>
>>> >> was the sandybridge system running i915 also from drm-next or the
>>> >> old in-tree version?
>>> >
>>> > It was the same system with the Radeon card detached. The i915kms.ko
>>> > was loaded via the rc.conf and the xf86-video-intel port was
>>> > installed and loaded with the xorg.conf file. Tomorrow i can upload
>>> > the Xorg.0.log files if they are needed.
>>>
>>> If you just load it without the full /boot/modules path, it's the
>>> system one…
>>>
>>> >>
>>> >>> My graphics hardware is an AMD Radeon RX560. OS is a FreeBSD
>>> >>> 11.2-RELEASE with
>>> >>> drm-next-kmod and xf86-video-amdgpu. The kernel module is loaded
>>> >>> with
>>> >>> kld_list="amdgpu" in the rc.conf and the X11 driver is loaded with
>>> >>> 'Driver
>>> >>> "amdgpu"' in the Device Section of the xorg.conf. Xorg works fine
>>> >>> with this
>>> >>> configuration.
>>> >>>
>>> >>> Following linux related packages are installed:
>>> >>> linux-c7-dri-17.0.1                =   up-to-date with remote
>>> >>> linux-c7-elfutils-libelf-0.168     =   up-to-date with remote
>>> >>> linux-c7-expat-2.1.0_2             =   up-to-date with remote
>>> >>> linux-c7-fontconfig-2.10.95_3      =   up-to-date with remote
>>> >>> linux-c7-glx-utils-8.2.0_4         =   up-to-date with remote
>>> >>> linux-c7-libpciaccess-0.13.4_3     =   up-to-date with remote
>>> >>> linux-c7-xorg-libs-7.7_5           =   up-to-date with remote
>>> >>> linux-doom3-1.3.1.1304,1           =   up-to-date with remote
>>> >>> linux_base-c7-7.4.1708_6           =   up-to-date with remote
>>> >>>
>>> >>> When running linux-doom3 i get the following output:
>>> >>> % linux-doom3
>>> >>> [...]
>>> >>> -------------------------------
>>> >>> using ARB_vertex_buffer_object memory
>>> >>> using ARB2 renderSystem
>>> >>> signal caught: Segmentation fault
>>> >>> si_code 1
>>> >>
>>> >> A backtrace would be useful here.
>>> >> Was a core dump produced?
>>> >
>>> > No coredump because i disabled it with kern.coredump=0 but i can
>>> > change this and look if a dump will be created. Tomorrow...
>>> >>
>>> >>
>>> >> Anyway, I definitely have run Linux GL applications on amdgpu, but I
>>> >> was using an Ubuntu chroot instead of the linux-* ports.
>>> >
>>> > Which Ubuntu version should i try? How can i create such a chroot. Is
>>> > it possible to install an Ubuntu on a separate disk, tar it up and
>>> > unpack it to /compat/linux.
>>>
>>> 16.04 or older.
>>> Just download e.g.
>>>
>>> https://cloud-images.ubuntu.com/xenial/current/xenial-server-cloudimg-amd64-root.tar.xz
>>> untar somewhere (anywhere, e.g. under the home dir), mount linprocfs,
>>> linsysfs and devfs under there, chroot into that, and you can use apt
>>> and whatnot
>>>
>>
>> How do you manage this? apt just gives me
>> http: ../sysdeps/posix/getaddrinfo.c:1758: rfc3484_sort: Assertion
>> `a1->source_addr.sin6_family == PF_INET6' failed.
>> E: Method http has died unexpectedly!
>> E: Sub-process http received signal 6.
>>
>> with 12-current.
>> Do you have the linux and linux64 module loaded? Also in chroot make sure
>> that the /etc/resolv.conf has a correct entry.
>>
>
Yes and yes. The solution was to disable IPv6 on the native interface.


>>> _______________________________________________
>>> freebsd-x11 at freebsd.org mailing list
>>> https://lists.freebsd.org/mailman/listinfo/freebsd-x11
>>> To unsubscribe, send any mail to "freebsd-x11-unsubscribe at freebsd.org"
>>>
>>
>