Nathan Moinvaziri
Nathan Moinvaziri
> nmake is still getting new features... I can't locate a nmake change log anywhere..
Here is the testing I did on various hash functions a long time ago: https://gist.github.com/nmoinvaz/26daed85a714e7c55cb29e61a3ee82fe
I think we should do some performance testing against `update_hash_c` and `update_hash_sse42` to see if it still makes sense to have the architecture specific functions.
### update_hash_c ``` OS: Windows 10 10.0.22631 AMD64 CPU: Intel64 Family 6 Model 151 Stepping 2, GenuineIntel Tool: ../zlib-ng/build/Release/minigzip-c.exe Size: 98,816 B Levels: 0-9 Runs: 70 Trim worst: 40 Level...
Ok I was going to rebase after #1644 anyways.
Actually, if we are just going to remove the sse42 & acle methods then there is no need for this PR.
Here is another run again using silestia, but also including chromium's hash. ### update_hash_c ``` OS: Windows 10 10.0.22631 AMD64 CPU: Intel64 Family 6 Model 151 Stepping 2, GenuineIntel Tool:...
I'm not really convinced that it makes sense to continue this PR after the recent changes. There is only one hashing method we now support - knuth. `update_hash` and `update_hash_roll`...
I have rebased the PR, so you can easily test against #1686 if you want it for that purpose.
I'm not familar enough with ARM, why rename ACLE to ARMv8?