Both asserts should pass, but don't because dmd loads the nan onto the fp stack, stores it into memory _as a double_, then loads it into the xmm registers as a double and returns it. The comparison code treats the xmm regs as if they contain floats, giving the wrong result. void main() { cfloat a() { return cfloat.nan; } assert(a() != 0); }
https://github.com/D-Programming-Language/dmd/pull/760 This is fixed by pull 760
New pull https://github.com/D-Programming-Language/dmd/pull/769
Commit pushed to dmd-1.x at https://github.com/D-Programming-Language/dmd https://github.com/D-Programming-Language/dmd/commit/e45925dafe7a9c71fc6eaaa852ed526e01768981 yebblies' fix Issue 7581 - Compiler uses wrong instructions to move complex value from ST to xmm registers
Commit pushed to master at https://github.com/D-Programming-Language/dmd https://github.com/D-Programming-Language/dmd/commit/0cf3c08642bb0cae819464e6a1aaecfb09b2d03c yebblies' fix Issue 7581 - Compiler uses wrong instructions to move complex value from ST to xmm registers