Dan Luu

Also see #355, #349.

This could theoretically be done now that `brew install llvm` installs llvm39.

This is related to the REPL memory leak (#283)

I fixed the TODO in 5d42ab16ea9bb78b17860688a481f7f1c31dec41. I haven't thought about it in order to determine if that fixes this bug "for free".

Note to self: run `bitfunnel termtable` under valgrind.

After #323, I manually checked `TermTable.cpp` to make sure it didn't have the same issue, and it doesn't. We don't know if this issue is still occurring, but if it...

I think this is user error on my part (using load instead of cache), but it sounds like I'm not the only person to do this. We could prevent this...

Ohh, I see what this is -- this is an artifact of the row configuration getting compressed into a size_t. This is kinda ok as long as we know for...

As you might expect, this is worse with the compiled version: 

This issue mostly goes away if I reduce the number of threads to match the number of physical cores instead of the number of hyperthreads. However, @MikeHopcroft observed different behavior...