Wunk

[Updated the readme with these!](https://github.com/Wunkolo/libsai/commit/cdee838a33d6424913ded74bf39aa40a7fb94f1b) Gonna dive in IDA Pro and get more info on some of these unknowns and behaviors, such as `peff`

Got a good map of the layer writing procedures and some dependent-tags. Will update the readme with some of this info 

`lmfl` is a 32-bit bitmask it looks like. It only actually uses the 2 least-significant bits, and only exists if the layer file is a mask. Bit 0 is set:...

Looking back at this old library, there are some decisions that I think I'd like to try and do to facilitate something like this without having to make redundant copies...

Sounds like a good pattern that is similar to how the structures themselves are laid out in memory. We can use the serialized layer headers or just encapsulate it into...

I've added a bit of an API related to iterating all of the layers and sublayers in a document. [Check out the new `Document.cpp` sample to see the new pattern.](https://github.com/Wunkolo/libsai/blob/master/samples/Document.cpp)...

I'll set up some sort of github-action or something sometime to house binaries of all the benchmark-tests executables! I have targets for each microarchitecture level in the cmake build though...

Not really... I don't have a MIPS or Risc-V machine to research code on but I am very open to PRs!

I suppose so. MIPS has a flag that can be enabled in user-space that causes all load/store instructions to use a reverse byte endian, so a pattern I would imagine...

Can't help but wonder how much performance gain the "real" [pdep](https://www.felixcloutier.com/x86/pdep) would do here. Though, there is a case to not use it on pre-Zen3 hardware.