rtm
rtm copied to clipboard
nfrechette
Realtime Math
See here for details: https://developercommunity.visualstudio.com/content/problem/1076955/vs2019-vs2017-vs2015-crash-when-mm-set-epi64x-is-u.html
While functions are commented, browsing what they do is a bit awkward. It would be nice if we could generate markdown or html somehow. Perhaps switching to doxygen might work?
Hi! Thanks for this library, seems pretty cool! I just thought I'd drop in some feedback as I'm taking a stab at using this. I'm attempting to use your library...
quatd, vector4d, and the other types can use `__vectorcall` as well where available. We need to add the proper macro to function signatures and we need to add an equivalent...
This causes the matrix3x4 unit test to fail. See branch: /bug/vs2019-vectorcall-inline A work around has been made in the unit test by splitting the large function. Bug report here: https://developercommunity.visualstudio.com/content/problem/1055298/vs-2019-sometimes-incorrectly-reads-return-value-f.html
Depends on emscripten support. See #62. SIMD support is experimental but we should be able to port the library to use it when present.
To improve inlinability, the SSE version should be explicit and use the `shuffle_ps` intrinsic directly instead of relying on vector_mix. The NEON implementation also needs to be added. Even though...
Clang supports https://clang.llvm.org/docs/AttributeReference.html#vectorcall so, it could be used when compiling with clang instead in addition to just MSVC
Might need to go in a major release as it might be quite a big change. See #92 for prior discussion. The idea is to expose to the host runtime...
Armv8 now supports float64 SIMD with ARM64. Several types can benefit from this: `quatd, vector4d, mask4d, etc` This will lead to more efficient register usage and argument passing.