6 ms·
If your code may be compiled, to use the Intel x87 numerical coprocessor, an important issue is the so called "excess precision": Different values on chip can c
by beyondCritics 5mo ago
If your code may be compiled, to use the Intel x87 numerical coprocessor, an important issue is the so called "excess precision": Different values on chip can collapse after being rounded and stored to their memory locations, invalidating previous comparisons. Spilling can happen unexpectedly. Note that Intel calls the x87 "legacy"
- pclmulqdq 5mo agoNobody's code will be compiled to use x87 any more.
- beyondCritics 5mo agoThere is plenty of demand for so called "secure code", where such coder arrogance will not be tolerated. Trust me on that, I know it.
- adgjlsfhk1 5mo agomodern compilers do just have options to disable using x87 registers entirely.
- pclmulqdq 5mo agoWhat is "coder arrogance"? The expectation that the 20-year-old API (SSE) will be used over the deprecated 40-year-old API that is currently losing compiler (and silicon) support? Can you point to a running x86 machine today that does not have SSE? It is also faster and more precise to use double-double for math over x87 extended precision. There is literally no reason to compile an x87 instruction today aside from a programmer not knowing better.