Patch preparing the addition of more tests.
Some require 128 alignment, so update driver to allocate aligned memory.
Details
Diff Detail
Event Timeline
I tested the change on Linux and OSX.
Suggestions on if there are other architecture specific changes I missed are welcome.
Bitcode/simd_ops/simd_ops.cpp | ||
---|---|---|
19 | Will this work on Windows? |
Bitcode/simd_ops/simd_ops.cpp | ||
---|---|---|
19 | I looked it up and I think I need to use "_aligned_malloc()". |
Bitcode/simd_ops/simd_ops.cpp | ||
---|---|---|
19 | For portability there's nothing specific that will work across everything until C++17. _aligned_malloc is the right way for windows. Perhaps abstract it out into an aligned allocation function. Also, if you wouldn't mind commenting this routine as well as to why we're allocating aligned memory :) |
One inline comment :)
Bitcode/simd_ops/simd_ops.cpp | ||
---|---|---|
6 | Comment what's going on here and why we're allocating memory if you wouldn't mind? It might be nicer for this to be a function rather than a macro? Just call it alloc_aligned or something and have the functionality for all 3 platforms in there? |
Bitcode/simd_ops/simd_ops.cpp | ||
---|---|---|
6 | Moved the comment from below here and created a function instead of the macro. |
Comment what's going on here and why we're allocating memory if you wouldn't mind? It might be nicer for this to be a function rather than a macro? Just call it alloc_aligned or something and have the functionality for all 3 platforms in there?