David T Lewis

This along with VMMaker.oscog-mt.3184 looks like a straighforward solution to the problem.

It is a VM bug, sorry for overlooking this. Proposed fix in VMMaker inbox VMMaker.oscog-dtl.3185

Fix is in VMMaker (VMMaker.oscog-dtl.3185), still need to commit generated code and build VM.

@marceltaeumel the fix inVMMaker.oscog-dtl.3185 is trivial, but I need someone else (you or Eliot) to do the code generation this time. The reason is that my generated code has differences...

For my understanding - is the concern with imageSegmentVersion a problem for practical reasons (something does not work as expected)? It certainly makes sense that an image segment should have...

I am trying to understand the real-world scenario in which this matters. I have an image running Sista bytecodes and my image format is 68533. I save an image segment...

Regarding image side code and MultipleBytecodeSetsBitmask in VMMaker, on the image side you would use this (only if necessary, and hopefully it is not necessary): (ImageFormat fromInteger: 68021) requiresMultipleBytecodeSupport ==>...

The issue occurs with ByteString but not WideString, so "'aaa' asWideString tryPrimitive: 64 withArgs: {2. $b asciiValue}; yourself" does not show the problem, but "'aaa' copy tryPrimitive: 64 withArgs: {2....

EPOLL on Linux can be a bit problematic when fork() is used (OSProcess forkSqueak). I wonder if there may also be some issues related to EPOLL on Linux pthreads? Totally...

Hi David, I tried your patch on Linux 64x64 and it causes no problems. The net effect is that the "hint" for the mmap() call is an address about 150k...