Bug 197940
| Summary: | games/openra (also lang/mono ): Crashes on load with apparent Mono runtime error | ||||||||
|---|---|---|---|---|---|---|---|---|---|
| Product: | Ports & Packages | Reporter: | David Mackay <davidjx8p> | ||||||
| Component: | Individual Port(s) | Assignee: | freebsd-mono (Nobody) <mono> | ||||||
| Status: | Closed Unable to Reproduce | ||||||||
| Severity: | Affects Some People | CC: | dbn, jbeich, mono | ||||||
| Priority: | --- | Keywords: | i915, needs-qa | ||||||
| Version: | Latest | ||||||||
| Hardware: | Any | ||||||||
| OS: | Any | ||||||||
| Attachments: |
|
||||||||
Maintainers CC'd Created attachment 153347 [details]
GDB attempt with mono running OpenRA
I have just tried running OpenRA after switching my laptop into nVIDIA GPU mode and using nvidia-driver, and it worked. It seems that the i915 OpenGL is problematic. OpenRA requires OpenGL 2.0 or greater. It seems i915 is enough to pass the check but not run the game. Handing over to x11@ to reproduce and analyze the issue in relation to the graphics stack. Comment on attachment 153347 [details]
GDB attempt with mono running OpenRA
With many symbols missing the backtrace isn't useful. Maybe rebuild everything with symbols (DEBUG_FLAGS=-g for base, WITH_DEBUG=1 for ports) or just graphics/libdrm, graphics/dri, graphics/libGL, x11-servers/xorg-server, x11-drivers/xf86-video-intel.
I can't reproduce on i915/Skylake, so the issue may have been fixed in drm-next-4.6/xserver-next branches.
Feedback timeout |
Created attachment 153346 [details] OpenRA running with Mono verbosity on debug games/openra will crash on attempting to run after getting partway through the splashscreen. This happens after the contents for the 'Red Alert' mod have been downloaded and it tries to proceed. Stacktrace: ================================================================= Got a SIGSEGV while executing native code. This usually indicates a fatal error in the mono runtime or one of the native libraries used by your application. ================================================================= Abort trap I have tried several things to deal with this: downgrading from Mono 3.10 to 3.6, recompiling Mono with gcc instead of clang, and so forth. None of this has made any difference. I have attached logs of running OpenRA with Mono's debugging verbosity enabled, as well as of an attempt to glean some information on mono-sgen's segfault from GDB.