Nathan Cole <nath@xxxxxxxxxxxxxxxxxxxxxx> writes:
> Apologies if I am missing something obvious, but I cannot understand why
> I am getting the following assembler output from gcc.
>
> Taking the following C function
>
> int myadd(int a, int b)
> {
> return a + b;
> }
>
> I compile with
>
> mips-linux-gnu-gcc -S test.c
>
> and the resulting assembly code is
>
> .file 1 "test.c"
> .section .mdebug.abi32
> .previous
> .nan legacy
> .module fp=32
> .module oddspreg
> .abicalls
> .option pic0
> .text
> .align 2
> .globl myadd
> .set nomips16
> .set nomicromips
> .ent myadd
> .type myadd, @function
> myadd:
> .frame $fp,8,$31 # vars= 0, regs= 1/0, args= 0,
> gp= 0
> .mask 0x40000000,-4
> .fmask 0x00000000,0
> .set noreorder
> .set nomacro
> addiu $sp,$sp,-8
> sw $fp,4($sp)
> move $fp,$sp
> sw $4,8($fp)
> sw $5,12($fp)
> lw $3,8($fp)
> lw $2,12($fp)
> addu $2,$3,$2
> move $sp,$fp
> lw $fp,4($sp)
> addiu $sp,$sp,8
> j $31
> nop
>
> .set macro
> .set reorder
> .end myadd
> .size myadd, .-myadd
> .ident "GCC: (Sourcery CodeBench Lite 2015.11-32) 5.2.0"
>
>
> What I do not understand is the use of the stack. The stack pointer is
> being moved by 8, so I would expect the code to use 0($sp) and 4($sp),
> but it uses 4, 8 and 12, which would appear to be in the callers stack
> frame?
Hi Nathan,
This is part of the o32 MIPS ABI. There is a minimum 16-byte argument
area on the stack for any call even if all arguments are passed in registers.
This stack space is used to spill arguments instead of creating new stack
slots in the caller's frame.
Take a look at the mipsabi.pdf found here for more details although it is
a tough read:
https://dmz-portal.mips.com/wiki/MIPS_ABI_Project#Where_are_the_current_and_historical_MIPS_ABI_related_documents
Hope that helps,
Matthew
