Open sclimpro opened 6 years ago
I'm not aware of any problems with that test (or any others), beyond the benign output bug. This shouldn't affect whether the test runs or not, but if you are printing the output then it will.
My guess is that this is some kind of problem where you have some sort of sharing that
you haven't noticed, or a value is being captured by reference rather than as a copy? For
example, the j
and j2
values you need to be quite careful that each task gets a different
value, rather than some people having a reference that they then copy.
It's possible it is some kind of concurrency bug, but I'm not aware of any specific problems that will affect running Crush via the driver program.
Hi, I noted that for the mNP2S test in ClosePairs for Crush test suite; it runs occasionally but not on every iteration; despite the code being the same.
I can't seem to find the root cause of the issue. Has anyone found this issue too?