8, 16, 32 bits integers are always passed, on the stack, as full width 32 bits values1.
No extension, signed or zeroed, is needed.
The callee will just use the lower part of
the full width values.
//C prototype of the callee
void __attribute__((cdecl)) foo(char a, short b, int c, long d);
foo(-1, 2, -3, 4);
;Call to foo in assembly
push DWORD 4 ;d, long is 32 bits, nothing special here
push DWORD 0fffffffdh ;c, int is 32 bits, nothing special here
push DWORD 0badb0002h ;b, short is 16 bits, higher WORD can be any value
push DWORD 0badbadffh ;a, char is 8 bits, higher three bytes can be any value
call foo
add esp, 10h ;Clean up the stack
64 bits values are passed on the stack using two pushes, respecting the littel endian convention2, pushing first the higher 32 bits then the lower ones.
//C prototype of the callee
void __attribute__((cdecl)) foo(char a, short b, int c, long d);
foo(0x0123456789abcdefLL);
;Call to foo in assembly
push DWORD 89abcdefh ;Higher DWORD of 0123456789abcdef
push DWORD 01234567h ;Lower DWORD of 0123456789abcdef
call foo
add esp, 08h
8 bits integers are returned in AL
, eventually clobbering the whole eax
.
16 bits integers are returned in AX
, eventually clobbering the whole eax
.
32 bits integers are returned in EAX
.
64 bits integers are returned in EDX:EAX
, where EAX
holds the lower 32 bits and EDX
the upper ones.
//C
char foo() { return -1; }
;Assembly
mov al, 0ffh
ret
//C
unsigned short foo() { return 2; }
;Assembly
mov ax, 2
ret
//C
int foo() { return -3; }
;Assembly
mov eax, 0fffffffdh
ret
//C
int foo() { return 4; }
;Assembly
xor edx, edx ;EDX = 0
mov eax, 4 ;EAX = 4
ret
1 This keep the stack aligned on 4 bytes, the natural word size. Also an x86 CPU can only push 2 or 4 bytes when not in long mode.
2 Lower DWORD at lower address