The libc-provided isnan/isinf/isfinite macro implementations are specifically
designed to function correctly, even in the presence of -ffast-math (or, more
specifically, -ffinite-math-only). As such, on most implementation, these
either always turn into external function calls (e.g. glibc) or are
specifically function calls when FINITE_MATH_ONLY is defined (e.g. Darwin).
Our implementation of complex arithmetic makes heavy use of isnan/isinf/isfinite
to deal with corner cases involving non-finite quantities. This was problematic
in two respects:
1. On systems where these are always function calls (e.g. Linux/glibc), there was a
performance penalty
2. When compiling with -ffast-math, there was a significant performance
penalty (in fact, on Darwin and systems with similar implementations, the code
may in fact be slower than not using -ffast-math, because the inline
definitions provided by libc become unavailable to prevent the checks from
being optimized out).
Eliding these inf/nan checks in -ffast-math mode is consistent with what
happens with libstdc++, and in my experience, what users expect. This is
critical to getting high-performance code when using complex<T>. This change
replaces uses of those functions on basic floating-point types with calls to
__builtin_isnan/isinf/isfinite, which Clang will always expand inline. When
using -ffast-math (or -ffinite-math-only), the optimizer will remove the checks
as expected.
Differential Revision: https://reviews.llvm.org/D18639
llvm-svn: 283051
The functions arg, conj, imag, norm, proj, and real have additional overloads
for arguments of integral or floating point types. However these overloads should
not allow conversions to the integral/floating point types, only exact matches.
This patch constrains these functions so they no longer allow conversions.
llvm-svn: 276067