Adam Stylinski

Is CMake just suddenly failing the test for whether or not that compiler builtin exists and is now not defining this macro? That has to be it, somehow. Wow, yeah,...

Anyhow, assuming that cmake issue gets fixed: ``` adam@g5box ~/zlib-ng-unaligned/build $ make test Running tests... Test project /home/adam/zlib-ng-unaligned/build Start 1: example 1/70 Test #1: example .......................... Passed 0.01 sec Start...

```adam@g5box ~/zlib-ng-unaligned/build $ grep -R "HAVE_ATTRIBUTE_ALIGNED" . ./CMakeFiles/CMakeConfigureLog.yaml: - "Performing Test HAVE_ATTRIBUTE_ALIGNED" ./CMakeFiles/CMakeConfigureLog.yaml: variable: "HAVE_ATTRIBUTE_ALIGNED" ./CMakeFiles/CMakeConfigureLog.yaml: /usr/bin/cc -DHAVE_ATTRIBUTE_ALIGNED -std=c11 -o CMakeFiles/cmTC_e2feb.dir/src.c.o -c /home/adam/zlib-ng-unaligned/build/CMakeFiles/CMakeScratch/TryCompile-lRaiIX/src.c ./CMakeCache.txt://Test HAVE_ATTRIBUTE_ALIGNED ./CMakeCache.txt:HAVE_ATTRIBUTE_ALIGNED:INTERNAL= ```

Something is up with this failing this cmake test, I'm not really sure why it's suddenly failing this test.

CHUNKMEMSET_SAFE does byte at a time copies up until an alignment is meant or if the remaining len in the buffer is less than 3x whatever size a "chunk" is...

If I had to guess it'd probably be TLS overhead. Initializing the tables before an lwp fork would certainly fix that, I would think. TLS requires this weird indirection through...

Oh yeah, 64 bit atomics would also be a much more straightforward and obvious approach, hah. At least it prevents the torn store situation. To date I don't think I've...

Issues probably isn't the correct place to put this question. I have run benchmarks against their adler32 checksum implementation, and there at least, we are several times faster. This is...

We'd need C bindings to directly plug this into our existing benchmarks. That having been said, it's definitely possible to leverage python's zlib interfaces with zlib-ng for a direct comparison....

For your edification, I managed to build this project (though, it seems rely on Makefiles that aren't present, so I had to switch it to point to my distro's package...