Its amazing =) I can't imagine that a new driver would have a NULL glGenFramebuffers. Hmm.
Any chance you can get a dbgutil build (these are usually quite a large download) and get the console output from:
SAL_LOG=1 ./soffice --writer
or whatever with GL enabled; that should help us go a bit further; but ... this guy is quite odd. More ideally - it'd be nice to have the crash in gdb [ which looks like a simple NULL ptr de-reference ] and ...
I fear bug#97700 makes any testing on 5.1.0* hard to predict anyway; we can corrupt memory without valgrind seeing =)
Its amazing =) I can't imagine that a new driver would have a NULL glGenFramebuffers. Hmm.
Any chance you can get a dbgutil build (these are usually quite a large download) and get the console output from:
SAL_LOG=1 ./soffice --writer
or whatever with GL enabled; that should help us go a bit further; but ... this guy is quite odd. More ideally - it'd be nice to have the crash in gdb [ which looks like a simple NULL ptr de-reference ] and ...
I fear bug#97700 makes any testing on 5.1.0* hard to predict anyway; we can corrupt memory without valgrind seeing =)
Could you re-test with 5.1.1.2 ? =)