Hi all, we installed VMD-1.9.3-intel-2016b-Python-2.7.12.eb, but it crashes with a segfault when started. The core dump shows:
Program terminated with signal 11, Segmentation fault. #0 0x00002ab05072246e in try_update_scene_state () from /hpc/eb/RedHatEnterpriseServer7/Mesa/12.0.2-intel-2016b/lib/libGL.so.1 ... (gdb) bt #0 0x00002ab05072246e in try_update_scene_state () from /hpc/eb/RedHatEnterpriseServer7/Mesa/12.0.2-intel-2016b/lib/libGL.so.1 #1 0x00002ab05071fa9d in begin_binning () from /hpc/eb/RedHatEnterpriseServer7/Mesa/12.0.2-intel-2016b/lib/libGL.so.1 #2 0x00002ab05071e67f in lp_setup_flush () from /hpc/eb/RedHatEnterpriseServer7/Mesa/12.0.2-intel-2016b/lib/libGL.so.1 #3 0x00002ab050470fb9 in st_context_flush () from /hpc/eb/RedHatEnterpriseServer7/Mesa/12.0.2-intel-2016b/lib/libGL.so.1 #4 0x00002ab050177bfe in XMesaSwapBuffers () from /hpc/eb/RedHatEnterpriseServer7/Mesa/12.0.2-intel-2016b/lib/libGL.so.1 #5 0x00000000004988f4 in OpenGLDisplayDevice::update(int) () #6 0x00000000004997c8 in OpenGLDisplayDevice::init(int, char**, VMDApp*, int*, int*) () #7 0x00000000007b850a in VMDApp::VMDinit(int, char**, char const*, int*, int*) () #8 0x00000000007e5361 in main () the problem seems to be in LLVM compiled with the intel compiler when used as backend for Mesa. It can be circumvented when using GALLIUM_DRIVER=swr vmd but that is slower, so not preferrable. Did anyone get VMD running with LLVM as driver for gallium (which is the default)? Cheers, John -- We are ISO 27001 certified and meet the high requirements for information security. | John Donners | Senior consultant | Compute | SURFsara | Science Park 140 | 1098 XG Amsterdam | Nederland | T (31)6 19039023 | [email protected] | www.surfsara.nl | Aanwezig op | ma | di | wo | do | vr
