News for hackers. Stuff that matters.
Re:Intel C/C++ compilers complete adoption of LLVM by gpderetta on Thursday August 12, 2021 @06:25AM (#28154440)
My understanding (but I'm not an ICC user) is the majority of the performance improvement are from the vectorized math libraries (that technically you can use from GCC as well).
Re:Intel C/C++ compilers complete adoption of LLVM by physicsguy on Thursday August 12, 2021 @06:32AM (#28154483)
It uses a default lower accuracy floating point model than GCC, so initially people think it's faster til they realise their results are slightly different and have to mess around with the `fp-model` flag
Re:Intel C/C++ compilers complete adoption of LLVM by gpderetta on Thursday August 12, 2021 @07:25AM (#28154879)
Right, of course, IIRC it does the equivalent of -ffast-math by default.