I periodically wonder whether Itanium or even the Pentium 4 would have been more successful had Intel released icc under an open source license. I’m assuming that they were trying to keep AMD from using it but I’m still not sure it was worth it from the number of times I knew scientists to avoid it because it was hard to share.
Yeah, I haven't used ICC for 7 years now, but at the time it was much better than clang/gcc at keeping SSE/AVX intrinsic types in registers through function calls (i.e. clang/gcc used to spill out onto the stack and re-load), and things like this in the article.
Were you testing on the same platform? The Microsoft ABI has callee-save XMM registers, whereas the Linux/macOS ABI does not. Regardless, it would be nice if more compilers could do interprocedural register allocation in cases where all callers are known.
I've heard similar opinions that people could just recompile their soft and receive significant speed boost