Re: [PATCH 1/7] Fix function argument and return value locations

[Simon Marchi <simark@simark.ca>]

On 2020-05-25 2:56 p.m., Hannes Domani via Gdb-patches wrote:
> diff --git a/gdb/amd64-windows-tdep.c b/gdb/amd64-windows-tdep.c
> index 487dfd45fc..db9845203f 100644
> --- a/gdb/amd64-windows-tdep.c
> +++ b/gdb/amd64-windows-tdep.c
> @@ -77,7 +77,8 @@ static int
>  amd64_windows_passed_by_xmm_register (struct type *type)
>  {
>    return ((type->code () == TYPE_CODE_FLT
> -	   || type->code () == TYPE_CODE_DECFLOAT)
> +	   || type->code () == TYPE_CODE_DECFLOAT
> +	   || type->code () == TYPE_CODE_COMPLEX)
>            && (TYPE_LENGTH (type) == 4 || TYPE_LENGTH (type) == 8));
>  }

I don't know much about ABIs, so I tried:

$ cat hello.c
#include <complex.h>

void other(int real, int imag);
void func (complex int n)
{
  other(creal(n), cimag(n));
}
$ x86_64-w64-mingw32-gcc hello.c -g3 -O0 -c
$ x86_64-w64-mingw32-objdump -d allo.o

allo.o:     file format elf64-x86-64


Disassembly of section .text:

0000000000000000 <func>:
   0:   55                      push   %rbp
   1:   48 89 e5                mov    %rsp,%rbp
   4:   48 83 ec 10             sub    $0x10,%rsp
   8:   48 89 7d f8             mov    %rdi,-0x8(%rbp)
   c:   8b 45 fc                mov    -0x4(%rbp),%eax
   f:   66 0f ef c0             pxor   %xmm0,%xmm0
  13:   f2 0f 2a c0             cvtsi2sd %eax,%xmm0
  17:   f2 0f 2c d0             cvttsd2si %xmm0,%edx
  1b:   8b 45 f8                mov    -0x8(%rbp),%eax
  1e:   66 0f ef c0             pxor   %xmm0,%xmm0
  22:   f2 0f 2a c0             cvtsi2sd %eax,%xmm0
  26:   f2 0f 2c c0             cvttsd2si %xmm0,%eax
  2a:   89 d6                   mov    %edx,%esi
  2c:   89 c7                   mov    %eax,%edi
  2e:   e8 00 00 00 00          callq  33 <func+0x33>
  33:   90                      nop
  34:   c9                      leaveq
  35:   c3                      retq


Doesn't this show that a `complex int` argument is passed through the rdi
register?  Am I missing something here?

Simon