On Sun, Aug 13, 2017 at 10:16:07PM +0100, Edd Barrett wrote: > On Sun, Aug 13, 2017 at 10:21:51AM +0300, Paul Irofti wrote: > > It's because of the stupid context.h header from mupdf > > > > #if defined(MEMENTO) || !defined(NDEBUG) > > #define FITZ_DEBUG_LOCKING > > #endif > > You could apply that to the mupdf port to make this one debuggable? What > do you think?
I submitted a bug report to mupdf https://bugs.ghostscript.com/show_bug.cgi?id=698353 and am waiting for them to confirm and fix it. In the meantime I am going to propose a local fix to sthen@. > You are not going to try and debug the crash? If the code is *that* bad, > then alright ;) But we should raise a bug upstream so we can apply any > patch they might provide. I am going to have another go at fixing the crash, but I do not have the time nor the nerves to debug that code. If I fail, which I am more then certain at this point, I will submit a bug. It is weird that the Windows version does not crash on the same document...
