[Tinycc-devel] x86-64 floating point arguments use the wrong registers

[Phillip McNallen]

Hello,

 

Functions with floating point arguments(in x86-64) don’t use the proper calling convention.

Float values should be passed in XMM0 – XMM3

 

double sum(double a, double b)  {

    return (a+b);

}

Generates this:

mov         qword ptr [rbp+10h],rcx 

mov         qword ptr [rbp+18h],rdx 

movq        xmm0,mmword ptr [rbp+10h] 

addsd       xmm0,mmword ptr [rbp+18h] 

 

 

This same function does however leave the result in XMM0 as it should.