On 2021-10-19 14:31, Jaikiran Pai wrote:
The other option I experimented with was to make ModuleDescriptor#hashCode() generate the same hashcode across multiple JVM runs. Although I do have a "working" version of that change, I decided not to spend too much time on it because the java.lang.Object#hashCode() contract itself clearly states that this value isn't expected to be same across multiple JVM runs. So whatever I do here is going to be brittle.
I'm assuming the cause for ModuleDescriptor#hashCode being is due to the various enums not having an explicitly defined hashCode? I think this should be fixed.
Either way you're going to be brittle since the patch to emit a 0 is relying on the ModuleDescriptor#hashCode implementation disallowing 0 as a hash value (a 0 will force a recalculation). While it'll only happen at most once per module these relatively expensive calculations are something we want to avoid on startup if we can do so for free.
/Claes
