Dylan Tuttle

@0xdaryl @hzongaro I apparently still have a bug in here somewhere because I'm getting a failure in the JCL during the build, but I thought I'd open up for a...

I fixed the bug! I squashed all of my review comment changes into one commit, just doing some more performance testing

@0xdaryl I believe I am ready for your review whenever you are!

@0xdaryl Does that mean to disable the optimization if `comp->om.canGenerateArraylets()`?

It definitely would be smart to write a unit test that verifies the results in all of the different cases. I've spent so much time using the benchmarks the last...

I included all of the tracelogs generated to cover all of the compilations: [log_CG_countPositives.txt](https://github.com/user-attachments/files/19095167/log_CG_countPositives.txt) [log_CG_countPositives_2.txt](https://github.com/user-attachments/files/19095169/log_CG_countPositives_2.txt) [log_CG_hasNegatives.txt](https://github.com/user-attachments/files/19095165/log_CG_hasNegatives.txt) [log_CG_hasNegatives_2.txt](https://github.com/user-attachments/files/19095166/log_CG_hasNegatives_2.txt)

I have implemented a simple iterative routine for handling the residual bits. Now that I have a unit test that checks all of the cases, I can confirm that it...

After doing some performance testing, the iterative approach is not going to cut it. I will try my xmm register + masking garbage bytes approach instead.

I have implemented a residue handling algorithm based on the original implementation, but with added logic to find the exact index. The code is getting pretty big and unwieldy, and...

Adding the exact index logic slows us down a little, but the performance is certainly still satisfactory, far outperforming OpenJ9. @hzongaro could I get a review on this new version?