Hi,
When helping someone in the mumax forums , I also ran into a more serious issue (unrelated to #357 (comment)). Calling minimize() by hand in the GUI after relax() but before relax had finished relaxing (ie, interrupting relax) leads to a nil pointer error message:
I've attached a script that reliably reproduces this issue here
mumaxcrash.txt
In order to reproduce this issue, I called minimize() in the GUI manually, approximately ~10-15 seconds after relax() started running. I haven't found a script that will reproduce the error without having to manually type in the minimize. If you call minimize in the script after relax() has completely finished, I do not get the error. However I was able to consistently reproduce this behavior 10/10 attempts.
I don't know if this timing is sensitive to GPU hardware (I am using a 1070 ti, so a newer card will finish relaxing faster), but for reference this is around step ~10,000 or so. It doesn't seem to be particularly sensitive to timing, if i call minimize() around step 5,000 or step 20,000 it still seems to crash. I've definitely called minimize partway through a relax before without issues many many times, so I'm not sure what about this particular script is sensitive. Using the break button before calling minimize seems to prevent this issue.
Cheers,
Josh L.
Hi,
When helping someone in the mumax forums , I also ran into a more serious issue (unrelated to #357 (comment)). Calling minimize() by hand in the GUI after relax() but before relax had finished relaxing (ie, interrupting relax) leads to a nil pointer error message:
I've attached a script that reliably reproduces this issue here
mumaxcrash.txt
In order to reproduce this issue, I called minimize() in the GUI manually, approximately ~10-15 seconds after relax() started running. I haven't found a script that will reproduce the error without having to manually type in the minimize. If you call minimize in the script after relax() has completely finished, I do not get the error. However I was able to consistently reproduce this behavior 10/10 attempts.
I don't know if this timing is sensitive to GPU hardware (I am using a 1070 ti, so a newer card will finish relaxing faster), but for reference this is around step ~10,000 or so. It doesn't seem to be particularly sensitive to timing, if i call minimize() around step 5,000 or step 20,000 it still seems to crash. I've definitely called minimize partway through a relax before without issues many many times, so I'm not sure what about this particular script is sensitive. Using the break button before calling minimize seems to prevent this issue.
Cheers,
Josh L.