You are not logged in.
@lordheavy: I'm not sure why, but I don't get compute shaders in glxinfo even with your latest builds.
Do I need to install something on top of opencl-mesa for it to show or a newer kernel than in testing? or maybe is there a new setting for the autogen step?Thanks!
Probably it depends on your video card?
Here:
$ glxinfo | grep compute
GL_ARB_compute_shader, GL_ARB_conditional_render_inverted,
GL_ARB_compute_shader, GL_ARB_conditional_render_inverted,
Offline
It's a 280x/7970HD it should support it I believe. Am I wrong?
glxinfo: Device: AMD TAHITI (DRM 2.43.0, LLVM 3.9.0) (0x6798).
lspci: 01:00.0 VGA compatible controller: Advanced Micro Devices, Inc. [AMD/ATI] Tahiti XT [Radeon HD 7970/8970 OEM / R9 280X] (prog-if 00 [VGA controller])
Last edited by gee (2016-04-24 09:13:28)
Offline
It's a 280x/7970HD it should support it I believe. Am I wrong?
glxinfo: Device: AMD TAHITI (DRM 2.43.0, LLVM 3.9.0) (0x6798).
lspci: 01:00.0 VGA compatible controller: Advanced Micro Devices, Inc. [AMD/ATI] Tahiti XT [Radeon HD 7970/8970 OEM / R9 280X] (prog-if 00 [VGA controller])
For this card you need kernel from Alexander Deucher (the future kernel 4.7), because current kernels are missing bits to expose compute shaders
See https://cgit.freedesktop.org/~agd5f/lin … xt-4.7-wip (this patch https://cgit.freedesktop.org/~agd5f/lin … d53afab21)
Offline
Ooooh I see, thank you so much!
I was going crazy not understanding this!
I am building the wip kernel and will report.
Last edited by gee (2016-04-24 09:57:21)
Offline
That worked thank you!
Now I need to find something that can use this new extension
I'll try patching 4.5.2 and see if it works as well. I applied the patch you linked, and the previous one that changed the drm version.
edit: Well that seems to work as well!
Last edited by gee (2016-04-24 10:34:36)
Offline
a new build fail with intel-vulkan : https://bugs.freedesktop.org/show_bug.cgi?id=95182
I commented vulkan support out for Aur mesa-git & lib32-mesa-git .
EDIT : solved by upstream
Last edited by Lone_Wolf (2016-04-29 12:22:39)
Disliking systemd intensely, but not satisfied with alternatives so focusing on taming systemd.
clean chroot building not flexible enough ?
Try clean chroot manager by graysky
Offline
Anyone else getting
LLVM ERROR: branch size exceeds simm16
with the last one or two updates?
Downgraded from 269966 to 269746 and it works again.
Whatever exactly this does... http://reviews.llvm.org/D20227
Offline
Anyone else getting
LLVM ERROR: branch size exceeds simm16
with the last one or two updates?Downgraded from 269966 to 269746 and it works again.
Whatever exactly this does... http://reviews.llvm.org/D20227
Yes, i get this with Tomb Raider.
Offline
For mesa-git related posts,please use https://bbs.archlinux.org/viewtopic.php?id=212819 from now on.
Disliking systemd intensely, but not satisfied with alternatives so focusing on taming systemd.
clean chroot building not flexible enough ?
Try clean chroot manager by graysky
Offline
Can anyone actually use VCE encoding? I'm trying a simple pipeline like
gst-launch-1.0 videotestsrc ! omxh264enc ! "video/x-h264,profile=high" ! h264parse ! matroskamux ! filesink location=output.avi
and get this puzzling message:
OMX-could not load /usr/lib/bellagio/libomxvideosched.so ==> OMX.st.video.scheduler ==> OMX.st.video.scheduler: ==> 2 1,456192 1,304128: /usr/lib/bellagio/libomxvideosched.so ==> OMX.st.video.scheduler ==> OMX.st.video.scheduler: ==> 2 1,456192 1,304128: cannot open shared object file: No such file or directory
It's confusing until you try to run it with strace:
open("/usr/lib/bellagio/libomxvideosched.so ==> OMX.st.video.scheduler ==> OMX.st.video.scheduler: ==> 2 1,456192 1,304128", O_RDONLY|O_CLOEXEC) = -1 ENOENT (No such file or directory)
Yes, it tries to open the whole thing as a filename.
In ~/.omxregister I have
/usr/lib/bellagio/libomxvideosched.so ==> OMX.st.video.scheduler ==> OMX.st.video.scheduler: ==> 2 1,456192 1,304128
/usr/lib/bellagio/libomx_mesa.so
BELLAGIO_SEARCH_PATHOMX-Failed to write %zu bytes to fd %d
OMX-Component % ==> OMX.mesa.video_decoder ==> OMX.mesa.video_decoder.mpeg2:OMX.mesa.video_decoder.avc:
==> OMX.mesa.video_encoder ==> OMX.mesa.video_encoder.avc:
/usr/lib/bellagio/libomxaudio_effects.so
BELLAGIO_SEARCH_PATHOMX-Failed to write %zu bytes to fd ==> OMX.st.volume.component ==> OMX.st.volume.component: ==> 2 1,65536 1,32768
==> OMX.st.audio.mixer ==> OMX.st.audio.mixer: ==> 1 50,60000
/usr/lib/bellagio/libomxclocksrc.so
BELLAGIO_SEARCH_PATHOMX-Failed to write %zu bytes to fd %d ==> OMX.st.clocksrc ==> OMX.st.clocksrc:
so I think that omxregister-bellagio screws up. But why? And how? When I remove the thing starting with ==> from the first line so only /usr/lib/bellagio/libomxvideosched.so remains and then try to run the gstreamer pipeline again I get
OMX-Component not found with current ST static component loader.
OMX-Component not found with current ST static component loader.
But... What does that mean? Does VCE encoding work for *anyone*? If yes, can someone with it working post ~/.omxregister to compare how a working one looks like?
Edit:
https://sourceforge.net/p/omxil/omxil/c … ter.c#l187
Last edited by haagch (2016-05-22 20:57:39)
Offline
This has now been superseded. Relevant thread here: https://bbs.archlinux.org/viewtopic.php?id=212819
Closing
Offline