Many hyperlinks are disabled.
Use anonymous login to enable hyperlinks.
|Comment:||Backup developer notes.|
|Downloads:||Tarball | ZIP archive | SQL archive|
|Timelines:||family | ancestors | descendants | both | trunk | x-date-201912|
|Files:||files | file ages | folders|
|User & Date:||squirreljme 2019-12-01 08:02:20|
|15:35||Setting of various class properties. check-in: 8fd9c6daec user: stephanie.gawroriski tags: trunk|
|08:02||Backup developer notes. check-in: 1e7d0af871 user: squirreljme tags: trunk, x-date-201912|
|21:38||Correct types to be the boxed types. check-in: ed840e0c6b user: stephanie.gawroriski tags: trunk|
> > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > >
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39
# 2019/11/30 ## 09:41 I had an idea for class loading, I can just do the `ClassInfo` accessing stuff in a helper class, then using that I can load more things from this. That way I can just not really have to worry too much about loading `ClassInfo`'s accessor each time when I just want to know some field information. I could also save some variable space for the vtable and pool pointers by just directly accessing the class as well. This would definitely be more efficient and easier to implement. Also another thing would be to actually implement that dynamic loading. ## 13:39 Okay so for the issue, register 14 is copied to 3. Then register 14 is decremented for the count. Register 15 also has our exception too. ## 13:43 Okay so on the invoke, `ATHROW` is keeping the exception which seems a bit wrong here as it should be cleared otherwise it might end up getting handled. Not sure which method is invoked however. ## 13:45 Since the invoke is right after count, I think it is calling garbage collect on it. ## 16:02 Okay so I wonder if there is a dup or similar before the throw, that causes it to be eligible for garbage collection when the stack is cleaned up. ## 16:05 Wonder what will happen if I disable enqueuing of exception handlers. Doing this just still causes too many uncounts.