ToT CrOS-for-Linux Build Failing to Start |
|||
Issue description
Chrome Version: refs/heads/master@{#530876}
OS: CrOS on Linux
Compile CrOS on Linux, then try to run it. I get an endless loop of the following output:
[112056:112056:0122/113444.036475:ERROR:gl_context_glx.cc(187)] Failed to create GL context with glXCreateNewContext.
[112056:112056:0122/113444.036515:ERROR:gpu_info_collector.cc(57)] gl::init::CreateGLContext failed
[112056:112056:0122/113444.036522:ERROR:gpu_info_collector.cc(126)] Could not create context for info collection.
[112056:112056:0122/113444.036536:ERROR:gpu_init.cc(54)] gpu::CollectGraphicsInfo failed (fatal).
[112056:112056:0122/113444.040269:ERROR:viz_main_impl.cc(197)] Exiting GPU process due to errors during initialization
fsamuel@ - thanks for triaging!
,
Jan 22 2018
Are you running in xvfb/xdummy etc.?
,
Jan 22 2018
My Linux CrOS build works fine at ea7b84f594d16edfd2f91ca84b3e08518df87698 from this morning. 1. What happens if you try running in a new terminal? There are some env variables that change things. 2. Are you running locally, xvfb or over CRD? 3. Try running with --use-gl=egl and see if that helps. 4. If you run a Linux desktop build check about:gpu. If GLX isn't working it will fall back to software compositing with SwiftShader GL.
,
Jan 22 2018
Answers to questions: - running in X11 terminal, not xvfb, etc. - running locally - trying --use-gl=egl gives slightly different message, but same result: [175726:175757:0122/121458.574235:ERROR:service_manager_context.cc(252)] Attempting to run unsupported native service: /usr/local/google_ssd/projects/chromium/src/out/CrOSRelease/content_gpu.service [179059:179059:0122/121458.602735:ERROR:gl_surface_egl.cc(826)] EGL display query failed with error EGL_SUCCESS [179059:179059:0122/121458.602766:ERROR:gl_surface_egl.cc(840)] eglInitialize Default failed with error EGL_BAD_DISPLAY [179059:179059:0122/121458.602776:ERROR:gl_ozone_egl.cc(20)] GLSurfaceEGL::InitializeOneOff failed. [179059:179059:0122/121458.604445:ERROR:viz_main_impl.cc(197)] Exiting GPU process due to errors during initialization - about:gpu Interestingly, I get (event when running in a fresh profile): Graphics Feature Status Canvas: Software only, hardware acceleration unavailable CheckerImaging: Enabled Flash: Software only. Hardware acceleration disabled Flash Stage3D: Software only. Hardware acceleration disabled Flash Stage3D Baseline profile: Software only. Hardware acceleration disabled Compositing: Software only. Hardware acceleration disabled Multiple Raster Threads: Enabled Native GpuMemoryBuffers: Software only. Hardware acceleration disabled Rasterization: Software only. Hardware acceleration disabled Video Decode: Software only. Hardware acceleration disabled WebGL: Software only, hardware acceleration unavailable WebGL2: Software only, hardware acceleration unavailable Problems Detected Native GpuMemoryBuffers have been disabled, either via about:flags or command line. Disabled Features: native_gpu_memory_buffers Not sure why ...
,
Jan 22 2018
It seems like Chrome trying to use the Nvidia driver is broken on your machine. I don't have any ideas for why exactly that might be though. Can you run 'glxgears' at all?
,
Jan 22 2018
Is your driver maybe busted (e.g. update or something)? Can you attach the output of 'glxinfo' ?
,
Jan 22 2018
Ha ha .. ok ... probably the rodete upgrade that I tried to start the other day, and it early-outed and said I had to uninstall the NVidia drivers first and try again ... Except I haven't done that yet, but I wonder if the failed upgrade attempt did something. I'll try completing the update tonight and see what happens.
,
Jan 22 2018
$ glxinfo name of display: :0.0 X Error of failed request: GLXBadContext Major opcode of failed request: 154 (GLX) Minor opcode of failed request: 6 (X_GLXIsDirect) Serial number of failed request: 26 Current serial number in output stream: 26
,
Jan 22 2018
Yep, that's a busted driver. Feel free to reopen after this is fixed if the problem persists.
,
Feb 1 2018
I've seen other people with this problem recently, including erg@ and probably estade@. erg@ and I saw the issue go away after a reboot. I had recently (~2 days before) updated to rodete. Does this make sense to anyone?
,
Feb 1 2018
It sounds like something went wrong updating nvidia drivers and then on restart the nvidia driver update was successful.
,
Feb 1 2018
A reboot did fix it for me. I learned I should have searched for "all issues" instead of just open ones to find the solution. |
|||
►
Sign in to add a comment |
|||
Comment 1 by fsamuel@google.com
, Jan 22 2018