Gmoccapy - spindle restarts at program stop

I assume this is the same issue:

2.10.0-pre0-3740-g5ec17ac2b8

The only way I've been able to reproduce it consistently is by starting and then stopping immediately, but when I recorded this video it happened without fail.
I think it's related to how the state is restored from a state tag after an abort(?).
If I comment out emcTaskStateRestore(); at line 2078 in emctaskmain.cc the spindle doesn't restart.

Here it is in gmoccapy:

If I abort almost immediately with the ESC key, the spindle stays on.
Does not happen with emcTaskStateRestore(); commented out.

Does this happen in QtDragon for you as well?

Yes. I have a real machine with QtDragon where this has been happening for some time now but I thought it was caused by something else until I noticed it happening in a simulation as well. I did some more experimenting and I discovered that if I have a G64 with a large P value, like 1 for example, the spindle doesn't restart. The smaller the P value is the further into a program it has to run before the spindle doesn't restart after an abort.With a G61 the spindle always restarts. I've only used 3D_Chips.ngc to test it.

I think this is related to what's happening:
github.com/LinuxCNC/linuxcnc/issues/579#issuecomment-640557582

This sure does sound related to #865 and #882, where readahead segments aren't dropped during task abort, and get flushed into the queue by commands following the abort: the ON_ABORT_COMMAND in this case, and the restored state tag in the case of #865.

However, in my quick test, I couldn't tickle the bug in the 2.8 branch by adding the ON_ABORT_COMMAND from @sheetcam's config. Anyway, if it is the same problem, hopefully #882 will fix it.

If it's related to that issue then we should expect this to also show up using the Axis GUI. Are you seeing it in Axis as well?

Yes. I tested it now on 2.10 RIP and it seems to behave like how I described.