I wonder if we should have Spike as a submodule instead of a fixed hash in the workflow files. That way Dependabot will update Spike on a regular basis. The expected results for the vector tests come from Spike, so if there are fixes to Spike that impact it's vector unit, we currently won't see those in the compiled tests.
I wonder if we should have Spike as a submodule instead of a fixed hash in the workflow files. That way Dependabot will update Spike on a regular basis. The expected results for the vector tests come from Spike, so if there are fixes to Spike that impact it's vector unit, we currently won't see those in the compiled tests.