ae5fe8a728
This PR removes all uses of modify-operands in the aarch64 backend, replacing them with reused-input operands instead. This has the nice effect of removing a bunch of move instructions and more clearly representing inputs and outputs. This PR also removes the explicit use of pinned vregs in the aarch64 backend, instead using fixed-register constraints on the operands when insts or pseudo-inst sequences require certain registers. This is the second PR in the regalloc-semantics cleanup series; after the remaining backend (s390x) and the ABI code are cleaned up as well, we'll be able to simplify the regalloc2 frontend.
844 lines
19 KiB
Plaintext
844 lines
19 KiB
Plaintext
test compile precise-output
|
|
target aarch64
|
|
|
|
function %atomic_rmw_add_i64(i64, i64) {
|
|
block0(v0: i64, v1: i64):
|
|
v2 = atomic_rmw.i64 add v0, v1
|
|
return
|
|
}
|
|
|
|
; stp fp, lr, [sp, #-16]!
|
|
; mov fp, sp
|
|
; str x28, [sp, #-16]!
|
|
; stp x26, x27, [sp, #-16]!
|
|
; stp x24, x25, [sp, #-16]!
|
|
; block0:
|
|
; mov x25, x0
|
|
; mov x26, x1
|
|
; atomic_rmw_loop_add_64 addr=x25 operand=x26 oldval=x27 scratch1=x24 scratch2=x28
|
|
; ldp x24, x25, [sp], #16
|
|
; ldp x26, x27, [sp], #16
|
|
; ldr x28, [sp], #16
|
|
; ldp fp, lr, [sp], #16
|
|
; ret
|
|
|
|
function %atomic_rmw_add_i32(i64, i32) {
|
|
block0(v0: i64, v1: i32):
|
|
v2 = atomic_rmw.i32 add v0, v1
|
|
return
|
|
}
|
|
|
|
; stp fp, lr, [sp, #-16]!
|
|
; mov fp, sp
|
|
; str x28, [sp, #-16]!
|
|
; stp x26, x27, [sp, #-16]!
|
|
; stp x24, x25, [sp, #-16]!
|
|
; block0:
|
|
; mov x25, x0
|
|
; mov x26, x1
|
|
; atomic_rmw_loop_add_32 addr=x25 operand=x26 oldval=x27 scratch1=x24 scratch2=x28
|
|
; ldp x24, x25, [sp], #16
|
|
; ldp x26, x27, [sp], #16
|
|
; ldr x28, [sp], #16
|
|
; ldp fp, lr, [sp], #16
|
|
; ret
|
|
|
|
function %atomic_rmw_add_i16(i64, i16) {
|
|
block0(v0: i64, v1: i16):
|
|
v2 = atomic_rmw.i16 add v0, v1
|
|
return
|
|
}
|
|
|
|
; stp fp, lr, [sp, #-16]!
|
|
; mov fp, sp
|
|
; str x28, [sp, #-16]!
|
|
; stp x26, x27, [sp, #-16]!
|
|
; stp x24, x25, [sp, #-16]!
|
|
; block0:
|
|
; mov x25, x0
|
|
; mov x26, x1
|
|
; atomic_rmw_loop_add_16 addr=x25 operand=x26 oldval=x27 scratch1=x24 scratch2=x28
|
|
; ldp x24, x25, [sp], #16
|
|
; ldp x26, x27, [sp], #16
|
|
; ldr x28, [sp], #16
|
|
; ldp fp, lr, [sp], #16
|
|
; ret
|
|
|
|
function %atomic_rmw_add_i8(i64, i8) {
|
|
block0(v0: i64, v1: i8):
|
|
v2 = atomic_rmw.i8 add v0, v1
|
|
return
|
|
}
|
|
|
|
; stp fp, lr, [sp, #-16]!
|
|
; mov fp, sp
|
|
; str x28, [sp, #-16]!
|
|
; stp x26, x27, [sp, #-16]!
|
|
; stp x24, x25, [sp, #-16]!
|
|
; block0:
|
|
; mov x25, x0
|
|
; mov x26, x1
|
|
; atomic_rmw_loop_add_8 addr=x25 operand=x26 oldval=x27 scratch1=x24 scratch2=x28
|
|
; ldp x24, x25, [sp], #16
|
|
; ldp x26, x27, [sp], #16
|
|
; ldr x28, [sp], #16
|
|
; ldp fp, lr, [sp], #16
|
|
; ret
|
|
|
|
function %atomic_rmw_sub_i64(i64, i64) {
|
|
block0(v0: i64, v1: i64):
|
|
v2 = atomic_rmw.i64 sub v0, v1
|
|
return
|
|
}
|
|
|
|
; stp fp, lr, [sp, #-16]!
|
|
; mov fp, sp
|
|
; str x28, [sp, #-16]!
|
|
; stp x26, x27, [sp, #-16]!
|
|
; stp x24, x25, [sp, #-16]!
|
|
; block0:
|
|
; mov x25, x0
|
|
; mov x26, x1
|
|
; atomic_rmw_loop_sub_64 addr=x25 operand=x26 oldval=x27 scratch1=x24 scratch2=x28
|
|
; ldp x24, x25, [sp], #16
|
|
; ldp x26, x27, [sp], #16
|
|
; ldr x28, [sp], #16
|
|
; ldp fp, lr, [sp], #16
|
|
; ret
|
|
|
|
function %atomic_rmw_sub_i32(i64, i32) {
|
|
block0(v0: i64, v1: i32):
|
|
v2 = atomic_rmw.i32 sub v0, v1
|
|
return
|
|
}
|
|
|
|
; stp fp, lr, [sp, #-16]!
|
|
; mov fp, sp
|
|
; str x28, [sp, #-16]!
|
|
; stp x26, x27, [sp, #-16]!
|
|
; stp x24, x25, [sp, #-16]!
|
|
; block0:
|
|
; mov x25, x0
|
|
; mov x26, x1
|
|
; atomic_rmw_loop_sub_32 addr=x25 operand=x26 oldval=x27 scratch1=x24 scratch2=x28
|
|
; ldp x24, x25, [sp], #16
|
|
; ldp x26, x27, [sp], #16
|
|
; ldr x28, [sp], #16
|
|
; ldp fp, lr, [sp], #16
|
|
; ret
|
|
|
|
function %atomic_rmw_sub_i16(i64, i16) {
|
|
block0(v0: i64, v1: i16):
|
|
v2 = atomic_rmw.i16 sub v0, v1
|
|
return
|
|
}
|
|
|
|
; stp fp, lr, [sp, #-16]!
|
|
; mov fp, sp
|
|
; str x28, [sp, #-16]!
|
|
; stp x26, x27, [sp, #-16]!
|
|
; stp x24, x25, [sp, #-16]!
|
|
; block0:
|
|
; mov x25, x0
|
|
; mov x26, x1
|
|
; atomic_rmw_loop_sub_16 addr=x25 operand=x26 oldval=x27 scratch1=x24 scratch2=x28
|
|
; ldp x24, x25, [sp], #16
|
|
; ldp x26, x27, [sp], #16
|
|
; ldr x28, [sp], #16
|
|
; ldp fp, lr, [sp], #16
|
|
; ret
|
|
|
|
function %atomic_rmw_sub_i8(i64, i8) {
|
|
block0(v0: i64, v1: i8):
|
|
v2 = atomic_rmw.i8 sub v0, v1
|
|
return
|
|
}
|
|
|
|
; stp fp, lr, [sp, #-16]!
|
|
; mov fp, sp
|
|
; str x28, [sp, #-16]!
|
|
; stp x26, x27, [sp, #-16]!
|
|
; stp x24, x25, [sp, #-16]!
|
|
; block0:
|
|
; mov x25, x0
|
|
; mov x26, x1
|
|
; atomic_rmw_loop_sub_8 addr=x25 operand=x26 oldval=x27 scratch1=x24 scratch2=x28
|
|
; ldp x24, x25, [sp], #16
|
|
; ldp x26, x27, [sp], #16
|
|
; ldr x28, [sp], #16
|
|
; ldp fp, lr, [sp], #16
|
|
; ret
|
|
|
|
function %atomic_rmw_and_i64(i64, i64) {
|
|
block0(v0: i64, v1: i64):
|
|
v2 = atomic_rmw.i64 and v0, v1
|
|
return
|
|
}
|
|
|
|
; stp fp, lr, [sp, #-16]!
|
|
; mov fp, sp
|
|
; str x28, [sp, #-16]!
|
|
; stp x26, x27, [sp, #-16]!
|
|
; stp x24, x25, [sp, #-16]!
|
|
; block0:
|
|
; mov x25, x0
|
|
; mov x26, x1
|
|
; atomic_rmw_loop_and_64 addr=x25 operand=x26 oldval=x27 scratch1=x24 scratch2=x28
|
|
; ldp x24, x25, [sp], #16
|
|
; ldp x26, x27, [sp], #16
|
|
; ldr x28, [sp], #16
|
|
; ldp fp, lr, [sp], #16
|
|
; ret
|
|
|
|
function %atomic_rmw_and_i32(i64, i32) {
|
|
block0(v0: i64, v1: i32):
|
|
v2 = atomic_rmw.i32 and v0, v1
|
|
return
|
|
}
|
|
|
|
; stp fp, lr, [sp, #-16]!
|
|
; mov fp, sp
|
|
; str x28, [sp, #-16]!
|
|
; stp x26, x27, [sp, #-16]!
|
|
; stp x24, x25, [sp, #-16]!
|
|
; block0:
|
|
; mov x25, x0
|
|
; mov x26, x1
|
|
; atomic_rmw_loop_and_32 addr=x25 operand=x26 oldval=x27 scratch1=x24 scratch2=x28
|
|
; ldp x24, x25, [sp], #16
|
|
; ldp x26, x27, [sp], #16
|
|
; ldr x28, [sp], #16
|
|
; ldp fp, lr, [sp], #16
|
|
; ret
|
|
|
|
function %atomic_rmw_and_i16(i64, i16) {
|
|
block0(v0: i64, v1: i16):
|
|
v2 = atomic_rmw.i16 and v0, v1
|
|
return
|
|
}
|
|
|
|
; stp fp, lr, [sp, #-16]!
|
|
; mov fp, sp
|
|
; str x28, [sp, #-16]!
|
|
; stp x26, x27, [sp, #-16]!
|
|
; stp x24, x25, [sp, #-16]!
|
|
; block0:
|
|
; mov x25, x0
|
|
; mov x26, x1
|
|
; atomic_rmw_loop_and_16 addr=x25 operand=x26 oldval=x27 scratch1=x24 scratch2=x28
|
|
; ldp x24, x25, [sp], #16
|
|
; ldp x26, x27, [sp], #16
|
|
; ldr x28, [sp], #16
|
|
; ldp fp, lr, [sp], #16
|
|
; ret
|
|
|
|
function %atomic_rmw_and_i8(i64, i8) {
|
|
block0(v0: i64, v1: i8):
|
|
v2 = atomic_rmw.i8 and v0, v1
|
|
return
|
|
}
|
|
|
|
; stp fp, lr, [sp, #-16]!
|
|
; mov fp, sp
|
|
; str x28, [sp, #-16]!
|
|
; stp x26, x27, [sp, #-16]!
|
|
; stp x24, x25, [sp, #-16]!
|
|
; block0:
|
|
; mov x25, x0
|
|
; mov x26, x1
|
|
; atomic_rmw_loop_and_8 addr=x25 operand=x26 oldval=x27 scratch1=x24 scratch2=x28
|
|
; ldp x24, x25, [sp], #16
|
|
; ldp x26, x27, [sp], #16
|
|
; ldr x28, [sp], #16
|
|
; ldp fp, lr, [sp], #16
|
|
; ret
|
|
|
|
function %atomic_rmw_nand_i64(i64, i64) {
|
|
block0(v0: i64, v1: i64):
|
|
v2 = atomic_rmw.i64 nand v0, v1
|
|
return
|
|
}
|
|
|
|
; stp fp, lr, [sp, #-16]!
|
|
; mov fp, sp
|
|
; str x28, [sp, #-16]!
|
|
; stp x26, x27, [sp, #-16]!
|
|
; stp x24, x25, [sp, #-16]!
|
|
; block0:
|
|
; mov x25, x0
|
|
; mov x26, x1
|
|
; atomic_rmw_loop_nand_64 addr=x25 operand=x26 oldval=x27 scratch1=x24 scratch2=x28
|
|
; ldp x24, x25, [sp], #16
|
|
; ldp x26, x27, [sp], #16
|
|
; ldr x28, [sp], #16
|
|
; ldp fp, lr, [sp], #16
|
|
; ret
|
|
|
|
function %atomic_rmw_nand_i32(i64, i32) {
|
|
block0(v0: i64, v1: i32):
|
|
v2 = atomic_rmw.i32 nand v0, v1
|
|
return
|
|
}
|
|
|
|
; stp fp, lr, [sp, #-16]!
|
|
; mov fp, sp
|
|
; str x28, [sp, #-16]!
|
|
; stp x26, x27, [sp, #-16]!
|
|
; stp x24, x25, [sp, #-16]!
|
|
; block0:
|
|
; mov x25, x0
|
|
; mov x26, x1
|
|
; atomic_rmw_loop_nand_32 addr=x25 operand=x26 oldval=x27 scratch1=x24 scratch2=x28
|
|
; ldp x24, x25, [sp], #16
|
|
; ldp x26, x27, [sp], #16
|
|
; ldr x28, [sp], #16
|
|
; ldp fp, lr, [sp], #16
|
|
; ret
|
|
|
|
function %atomic_rmw_nand_i16(i64, i16) {
|
|
block0(v0: i64, v1: i16):
|
|
v2 = atomic_rmw.i16 nand v0, v1
|
|
return
|
|
}
|
|
|
|
; stp fp, lr, [sp, #-16]!
|
|
; mov fp, sp
|
|
; str x28, [sp, #-16]!
|
|
; stp x26, x27, [sp, #-16]!
|
|
; stp x24, x25, [sp, #-16]!
|
|
; block0:
|
|
; mov x25, x0
|
|
; mov x26, x1
|
|
; atomic_rmw_loop_nand_16 addr=x25 operand=x26 oldval=x27 scratch1=x24 scratch2=x28
|
|
; ldp x24, x25, [sp], #16
|
|
; ldp x26, x27, [sp], #16
|
|
; ldr x28, [sp], #16
|
|
; ldp fp, lr, [sp], #16
|
|
; ret
|
|
|
|
function %atomic_rmw_nand_i8(i64, i8) {
|
|
block0(v0: i64, v1: i8):
|
|
v2 = atomic_rmw.i8 nand v0, v1
|
|
return
|
|
}
|
|
|
|
; stp fp, lr, [sp, #-16]!
|
|
; mov fp, sp
|
|
; str x28, [sp, #-16]!
|
|
; stp x26, x27, [sp, #-16]!
|
|
; stp x24, x25, [sp, #-16]!
|
|
; block0:
|
|
; mov x25, x0
|
|
; mov x26, x1
|
|
; atomic_rmw_loop_nand_8 addr=x25 operand=x26 oldval=x27 scratch1=x24 scratch2=x28
|
|
; ldp x24, x25, [sp], #16
|
|
; ldp x26, x27, [sp], #16
|
|
; ldr x28, [sp], #16
|
|
; ldp fp, lr, [sp], #16
|
|
; ret
|
|
|
|
function %atomic_rmw_or_i64(i64, i64) {
|
|
block0(v0: i64, v1: i64):
|
|
v2 = atomic_rmw.i64 or v0, v1
|
|
return
|
|
}
|
|
|
|
; stp fp, lr, [sp, #-16]!
|
|
; mov fp, sp
|
|
; str x28, [sp, #-16]!
|
|
; stp x26, x27, [sp, #-16]!
|
|
; stp x24, x25, [sp, #-16]!
|
|
; block0:
|
|
; mov x25, x0
|
|
; mov x26, x1
|
|
; atomic_rmw_loop_orr_64 addr=x25 operand=x26 oldval=x27 scratch1=x24 scratch2=x28
|
|
; ldp x24, x25, [sp], #16
|
|
; ldp x26, x27, [sp], #16
|
|
; ldr x28, [sp], #16
|
|
; ldp fp, lr, [sp], #16
|
|
; ret
|
|
|
|
function %atomic_rmw_or_i32(i64, i32) {
|
|
block0(v0: i64, v1: i32):
|
|
v2 = atomic_rmw.i32 or v0, v1
|
|
return
|
|
}
|
|
|
|
; stp fp, lr, [sp, #-16]!
|
|
; mov fp, sp
|
|
; str x28, [sp, #-16]!
|
|
; stp x26, x27, [sp, #-16]!
|
|
; stp x24, x25, [sp, #-16]!
|
|
; block0:
|
|
; mov x25, x0
|
|
; mov x26, x1
|
|
; atomic_rmw_loop_orr_32 addr=x25 operand=x26 oldval=x27 scratch1=x24 scratch2=x28
|
|
; ldp x24, x25, [sp], #16
|
|
; ldp x26, x27, [sp], #16
|
|
; ldr x28, [sp], #16
|
|
; ldp fp, lr, [sp], #16
|
|
; ret
|
|
|
|
function %atomic_rmw_or_i16(i64, i16) {
|
|
block0(v0: i64, v1: i16):
|
|
v2 = atomic_rmw.i16 or v0, v1
|
|
return
|
|
}
|
|
|
|
; stp fp, lr, [sp, #-16]!
|
|
; mov fp, sp
|
|
; str x28, [sp, #-16]!
|
|
; stp x26, x27, [sp, #-16]!
|
|
; stp x24, x25, [sp, #-16]!
|
|
; block0:
|
|
; mov x25, x0
|
|
; mov x26, x1
|
|
; atomic_rmw_loop_orr_16 addr=x25 operand=x26 oldval=x27 scratch1=x24 scratch2=x28
|
|
; ldp x24, x25, [sp], #16
|
|
; ldp x26, x27, [sp], #16
|
|
; ldr x28, [sp], #16
|
|
; ldp fp, lr, [sp], #16
|
|
; ret
|
|
|
|
function %atomic_rmw_or_i8(i64, i8) {
|
|
block0(v0: i64, v1: i8):
|
|
v2 = atomic_rmw.i8 or v0, v1
|
|
return
|
|
}
|
|
|
|
; stp fp, lr, [sp, #-16]!
|
|
; mov fp, sp
|
|
; str x28, [sp, #-16]!
|
|
; stp x26, x27, [sp, #-16]!
|
|
; stp x24, x25, [sp, #-16]!
|
|
; block0:
|
|
; mov x25, x0
|
|
; mov x26, x1
|
|
; atomic_rmw_loop_orr_8 addr=x25 operand=x26 oldval=x27 scratch1=x24 scratch2=x28
|
|
; ldp x24, x25, [sp], #16
|
|
; ldp x26, x27, [sp], #16
|
|
; ldr x28, [sp], #16
|
|
; ldp fp, lr, [sp], #16
|
|
; ret
|
|
|
|
function %atomic_rmw_xor_i64(i64, i64) {
|
|
block0(v0: i64, v1: i64):
|
|
v2 = atomic_rmw.i64 xor v0, v1
|
|
return
|
|
}
|
|
|
|
; stp fp, lr, [sp, #-16]!
|
|
; mov fp, sp
|
|
; str x28, [sp, #-16]!
|
|
; stp x26, x27, [sp, #-16]!
|
|
; stp x24, x25, [sp, #-16]!
|
|
; block0:
|
|
; mov x25, x0
|
|
; mov x26, x1
|
|
; atomic_rmw_loop_eor_64 addr=x25 operand=x26 oldval=x27 scratch1=x24 scratch2=x28
|
|
; ldp x24, x25, [sp], #16
|
|
; ldp x26, x27, [sp], #16
|
|
; ldr x28, [sp], #16
|
|
; ldp fp, lr, [sp], #16
|
|
; ret
|
|
|
|
function %atomic_rmw_xor_i32(i64, i32) {
|
|
block0(v0: i64, v1: i32):
|
|
v2 = atomic_rmw.i32 xor v0, v1
|
|
return
|
|
}
|
|
|
|
; stp fp, lr, [sp, #-16]!
|
|
; mov fp, sp
|
|
; str x28, [sp, #-16]!
|
|
; stp x26, x27, [sp, #-16]!
|
|
; stp x24, x25, [sp, #-16]!
|
|
; block0:
|
|
; mov x25, x0
|
|
; mov x26, x1
|
|
; atomic_rmw_loop_eor_32 addr=x25 operand=x26 oldval=x27 scratch1=x24 scratch2=x28
|
|
; ldp x24, x25, [sp], #16
|
|
; ldp x26, x27, [sp], #16
|
|
; ldr x28, [sp], #16
|
|
; ldp fp, lr, [sp], #16
|
|
; ret
|
|
|
|
function %atomic_rmw_xor_i16(i64, i16) {
|
|
block0(v0: i64, v1: i16):
|
|
v2 = atomic_rmw.i16 xor v0, v1
|
|
return
|
|
}
|
|
|
|
; stp fp, lr, [sp, #-16]!
|
|
; mov fp, sp
|
|
; str x28, [sp, #-16]!
|
|
; stp x26, x27, [sp, #-16]!
|
|
; stp x24, x25, [sp, #-16]!
|
|
; block0:
|
|
; mov x25, x0
|
|
; mov x26, x1
|
|
; atomic_rmw_loop_eor_16 addr=x25 operand=x26 oldval=x27 scratch1=x24 scratch2=x28
|
|
; ldp x24, x25, [sp], #16
|
|
; ldp x26, x27, [sp], #16
|
|
; ldr x28, [sp], #16
|
|
; ldp fp, lr, [sp], #16
|
|
; ret
|
|
|
|
function %atomic_rmw_xor_i8(i64, i8) {
|
|
block0(v0: i64, v1: i8):
|
|
v2 = atomic_rmw.i8 xor v0, v1
|
|
return
|
|
}
|
|
|
|
; stp fp, lr, [sp, #-16]!
|
|
; mov fp, sp
|
|
; str x28, [sp, #-16]!
|
|
; stp x26, x27, [sp, #-16]!
|
|
; stp x24, x25, [sp, #-16]!
|
|
; block0:
|
|
; mov x25, x0
|
|
; mov x26, x1
|
|
; atomic_rmw_loop_eor_8 addr=x25 operand=x26 oldval=x27 scratch1=x24 scratch2=x28
|
|
; ldp x24, x25, [sp], #16
|
|
; ldp x26, x27, [sp], #16
|
|
; ldr x28, [sp], #16
|
|
; ldp fp, lr, [sp], #16
|
|
; ret
|
|
|
|
function %atomic_rmw_smax_i64(i64, i64) {
|
|
block0(v0: i64, v1: i64):
|
|
v2 = atomic_rmw.i64 smax v0, v1
|
|
return
|
|
}
|
|
|
|
; stp fp, lr, [sp, #-16]!
|
|
; mov fp, sp
|
|
; str x28, [sp, #-16]!
|
|
; stp x26, x27, [sp, #-16]!
|
|
; stp x24, x25, [sp, #-16]!
|
|
; block0:
|
|
; mov x25, x0
|
|
; mov x26, x1
|
|
; atomic_rmw_loop_smax_64 addr=x25 operand=x26 oldval=x27 scratch1=x24 scratch2=x28
|
|
; ldp x24, x25, [sp], #16
|
|
; ldp x26, x27, [sp], #16
|
|
; ldr x28, [sp], #16
|
|
; ldp fp, lr, [sp], #16
|
|
; ret
|
|
|
|
function %atomic_rmw_smax_i32(i64, i32) {
|
|
block0(v0: i64, v1: i32):
|
|
v2 = atomic_rmw.i32 smax v0, v1
|
|
return
|
|
}
|
|
|
|
; stp fp, lr, [sp, #-16]!
|
|
; mov fp, sp
|
|
; str x28, [sp, #-16]!
|
|
; stp x26, x27, [sp, #-16]!
|
|
; stp x24, x25, [sp, #-16]!
|
|
; block0:
|
|
; mov x25, x0
|
|
; mov x26, x1
|
|
; atomic_rmw_loop_smax_32 addr=x25 operand=x26 oldval=x27 scratch1=x24 scratch2=x28
|
|
; ldp x24, x25, [sp], #16
|
|
; ldp x26, x27, [sp], #16
|
|
; ldr x28, [sp], #16
|
|
; ldp fp, lr, [sp], #16
|
|
; ret
|
|
|
|
function %atomic_rmw_smax_i16(i64, i16) {
|
|
block0(v0: i64, v1: i16):
|
|
v2 = atomic_rmw.i16 smax v0, v1
|
|
return
|
|
}
|
|
|
|
; stp fp, lr, [sp, #-16]!
|
|
; mov fp, sp
|
|
; str x28, [sp, #-16]!
|
|
; stp x26, x27, [sp, #-16]!
|
|
; stp x24, x25, [sp, #-16]!
|
|
; block0:
|
|
; mov x25, x0
|
|
; mov x26, x1
|
|
; atomic_rmw_loop_smax_16 addr=x25 operand=x26 oldval=x27 scratch1=x24 scratch2=x28
|
|
; ldp x24, x25, [sp], #16
|
|
; ldp x26, x27, [sp], #16
|
|
; ldr x28, [sp], #16
|
|
; ldp fp, lr, [sp], #16
|
|
; ret
|
|
|
|
function %atomic_rmw_smax_i8(i64, i8) {
|
|
block0(v0: i64, v1: i8):
|
|
v2 = atomic_rmw.i8 smax v0, v1
|
|
return
|
|
}
|
|
|
|
; stp fp, lr, [sp, #-16]!
|
|
; mov fp, sp
|
|
; str x28, [sp, #-16]!
|
|
; stp x26, x27, [sp, #-16]!
|
|
; stp x24, x25, [sp, #-16]!
|
|
; block0:
|
|
; mov x25, x0
|
|
; mov x26, x1
|
|
; atomic_rmw_loop_smax_8 addr=x25 operand=x26 oldval=x27 scratch1=x24 scratch2=x28
|
|
; ldp x24, x25, [sp], #16
|
|
; ldp x26, x27, [sp], #16
|
|
; ldr x28, [sp], #16
|
|
; ldp fp, lr, [sp], #16
|
|
; ret
|
|
|
|
function %atomic_rmw_umax_i64(i64, i64) {
|
|
block0(v0: i64, v1: i64):
|
|
v2 = atomic_rmw.i64 umax v0, v1
|
|
return
|
|
}
|
|
|
|
; stp fp, lr, [sp, #-16]!
|
|
; mov fp, sp
|
|
; str x28, [sp, #-16]!
|
|
; stp x26, x27, [sp, #-16]!
|
|
; stp x24, x25, [sp, #-16]!
|
|
; block0:
|
|
; mov x25, x0
|
|
; mov x26, x1
|
|
; atomic_rmw_loop_umax_64 addr=x25 operand=x26 oldval=x27 scratch1=x24 scratch2=x28
|
|
; ldp x24, x25, [sp], #16
|
|
; ldp x26, x27, [sp], #16
|
|
; ldr x28, [sp], #16
|
|
; ldp fp, lr, [sp], #16
|
|
; ret
|
|
|
|
function %atomic_rmw_umax_i32(i64, i32) {
|
|
block0(v0: i64, v1: i32):
|
|
v2 = atomic_rmw.i32 umax v0, v1
|
|
return
|
|
}
|
|
|
|
; stp fp, lr, [sp, #-16]!
|
|
; mov fp, sp
|
|
; str x28, [sp, #-16]!
|
|
; stp x26, x27, [sp, #-16]!
|
|
; stp x24, x25, [sp, #-16]!
|
|
; block0:
|
|
; mov x25, x0
|
|
; mov x26, x1
|
|
; atomic_rmw_loop_umax_32 addr=x25 operand=x26 oldval=x27 scratch1=x24 scratch2=x28
|
|
; ldp x24, x25, [sp], #16
|
|
; ldp x26, x27, [sp], #16
|
|
; ldr x28, [sp], #16
|
|
; ldp fp, lr, [sp], #16
|
|
; ret
|
|
|
|
function %atomic_rmw_umax_i16(i64, i16) {
|
|
block0(v0: i64, v1: i16):
|
|
v2 = atomic_rmw.i16 umax v0, v1
|
|
return
|
|
}
|
|
|
|
; stp fp, lr, [sp, #-16]!
|
|
; mov fp, sp
|
|
; str x28, [sp, #-16]!
|
|
; stp x26, x27, [sp, #-16]!
|
|
; stp x24, x25, [sp, #-16]!
|
|
; block0:
|
|
; mov x25, x0
|
|
; mov x26, x1
|
|
; atomic_rmw_loop_umax_16 addr=x25 operand=x26 oldval=x27 scratch1=x24 scratch2=x28
|
|
; ldp x24, x25, [sp], #16
|
|
; ldp x26, x27, [sp], #16
|
|
; ldr x28, [sp], #16
|
|
; ldp fp, lr, [sp], #16
|
|
; ret
|
|
|
|
function %atomic_rmw_umax_i8(i64, i8) {
|
|
block0(v0: i64, v1: i8):
|
|
v2 = atomic_rmw.i8 umax v0, v1
|
|
return
|
|
}
|
|
|
|
; stp fp, lr, [sp, #-16]!
|
|
; mov fp, sp
|
|
; str x28, [sp, #-16]!
|
|
; stp x26, x27, [sp, #-16]!
|
|
; stp x24, x25, [sp, #-16]!
|
|
; block0:
|
|
; mov x25, x0
|
|
; mov x26, x1
|
|
; atomic_rmw_loop_umax_8 addr=x25 operand=x26 oldval=x27 scratch1=x24 scratch2=x28
|
|
; ldp x24, x25, [sp], #16
|
|
; ldp x26, x27, [sp], #16
|
|
; ldr x28, [sp], #16
|
|
; ldp fp, lr, [sp], #16
|
|
; ret
|
|
|
|
function %atomic_rmw_smin_i64(i64, i64) {
|
|
block0(v0: i64, v1: i64):
|
|
v2 = atomic_rmw.i64 smin v0, v1
|
|
return
|
|
}
|
|
|
|
; stp fp, lr, [sp, #-16]!
|
|
; mov fp, sp
|
|
; str x28, [sp, #-16]!
|
|
; stp x26, x27, [sp, #-16]!
|
|
; stp x24, x25, [sp, #-16]!
|
|
; block0:
|
|
; mov x25, x0
|
|
; mov x26, x1
|
|
; atomic_rmw_loop_smin_64 addr=x25 operand=x26 oldval=x27 scratch1=x24 scratch2=x28
|
|
; ldp x24, x25, [sp], #16
|
|
; ldp x26, x27, [sp], #16
|
|
; ldr x28, [sp], #16
|
|
; ldp fp, lr, [sp], #16
|
|
; ret
|
|
|
|
function %atomic_rmw_smin_i32(i64, i32) {
|
|
block0(v0: i64, v1: i32):
|
|
v2 = atomic_rmw.i32 smin v0, v1
|
|
return
|
|
}
|
|
|
|
; stp fp, lr, [sp, #-16]!
|
|
; mov fp, sp
|
|
; str x28, [sp, #-16]!
|
|
; stp x26, x27, [sp, #-16]!
|
|
; stp x24, x25, [sp, #-16]!
|
|
; block0:
|
|
; mov x25, x0
|
|
; mov x26, x1
|
|
; atomic_rmw_loop_smin_32 addr=x25 operand=x26 oldval=x27 scratch1=x24 scratch2=x28
|
|
; ldp x24, x25, [sp], #16
|
|
; ldp x26, x27, [sp], #16
|
|
; ldr x28, [sp], #16
|
|
; ldp fp, lr, [sp], #16
|
|
; ret
|
|
|
|
function %atomic_rmw_smin_i16(i64, i16) {
|
|
block0(v0: i64, v1: i16):
|
|
v2 = atomic_rmw.i16 smin v0, v1
|
|
return
|
|
}
|
|
|
|
; stp fp, lr, [sp, #-16]!
|
|
; mov fp, sp
|
|
; str x28, [sp, #-16]!
|
|
; stp x26, x27, [sp, #-16]!
|
|
; stp x24, x25, [sp, #-16]!
|
|
; block0:
|
|
; mov x25, x0
|
|
; mov x26, x1
|
|
; atomic_rmw_loop_smin_16 addr=x25 operand=x26 oldval=x27 scratch1=x24 scratch2=x28
|
|
; ldp x24, x25, [sp], #16
|
|
; ldp x26, x27, [sp], #16
|
|
; ldr x28, [sp], #16
|
|
; ldp fp, lr, [sp], #16
|
|
; ret
|
|
|
|
function %atomic_rmw_smin_i8(i64, i8) {
|
|
block0(v0: i64, v1: i8):
|
|
v2 = atomic_rmw.i8 smin v0, v1
|
|
return
|
|
}
|
|
|
|
; stp fp, lr, [sp, #-16]!
|
|
; mov fp, sp
|
|
; str x28, [sp, #-16]!
|
|
; stp x26, x27, [sp, #-16]!
|
|
; stp x24, x25, [sp, #-16]!
|
|
; block0:
|
|
; mov x25, x0
|
|
; mov x26, x1
|
|
; atomic_rmw_loop_smin_8 addr=x25 operand=x26 oldval=x27 scratch1=x24 scratch2=x28
|
|
; ldp x24, x25, [sp], #16
|
|
; ldp x26, x27, [sp], #16
|
|
; ldr x28, [sp], #16
|
|
; ldp fp, lr, [sp], #16
|
|
; ret
|
|
|
|
function %atomic_rmw_umin_i64(i64, i64) {
|
|
block0(v0: i64, v1: i64):
|
|
v2 = atomic_rmw.i64 umin v0, v1
|
|
return
|
|
}
|
|
|
|
; stp fp, lr, [sp, #-16]!
|
|
; mov fp, sp
|
|
; str x28, [sp, #-16]!
|
|
; stp x26, x27, [sp, #-16]!
|
|
; stp x24, x25, [sp, #-16]!
|
|
; block0:
|
|
; mov x25, x0
|
|
; mov x26, x1
|
|
; atomic_rmw_loop_umin_64 addr=x25 operand=x26 oldval=x27 scratch1=x24 scratch2=x28
|
|
; ldp x24, x25, [sp], #16
|
|
; ldp x26, x27, [sp], #16
|
|
; ldr x28, [sp], #16
|
|
; ldp fp, lr, [sp], #16
|
|
; ret
|
|
|
|
function %atomic_rmw_umin_i32(i64, i32) {
|
|
block0(v0: i64, v1: i32):
|
|
v2 = atomic_rmw.i32 umin v0, v1
|
|
return
|
|
}
|
|
|
|
; stp fp, lr, [sp, #-16]!
|
|
; mov fp, sp
|
|
; str x28, [sp, #-16]!
|
|
; stp x26, x27, [sp, #-16]!
|
|
; stp x24, x25, [sp, #-16]!
|
|
; block0:
|
|
; mov x25, x0
|
|
; mov x26, x1
|
|
; atomic_rmw_loop_umin_32 addr=x25 operand=x26 oldval=x27 scratch1=x24 scratch2=x28
|
|
; ldp x24, x25, [sp], #16
|
|
; ldp x26, x27, [sp], #16
|
|
; ldr x28, [sp], #16
|
|
; ldp fp, lr, [sp], #16
|
|
; ret
|
|
|
|
function %atomic_rmw_umin_i16(i64, i16) {
|
|
block0(v0: i64, v1: i16):
|
|
v2 = atomic_rmw.i16 umin v0, v1
|
|
return
|
|
}
|
|
|
|
; stp fp, lr, [sp, #-16]!
|
|
; mov fp, sp
|
|
; str x28, [sp, #-16]!
|
|
; stp x26, x27, [sp, #-16]!
|
|
; stp x24, x25, [sp, #-16]!
|
|
; block0:
|
|
; mov x25, x0
|
|
; mov x26, x1
|
|
; atomic_rmw_loop_umin_16 addr=x25 operand=x26 oldval=x27 scratch1=x24 scratch2=x28
|
|
; ldp x24, x25, [sp], #16
|
|
; ldp x26, x27, [sp], #16
|
|
; ldr x28, [sp], #16
|
|
; ldp fp, lr, [sp], #16
|
|
; ret
|
|
|
|
function %atomic_rmw_umin_i8(i64, i8) {
|
|
block0(v0: i64, v1: i8):
|
|
v2 = atomic_rmw.i8 umin v0, v1
|
|
return
|
|
}
|
|
|
|
; stp fp, lr, [sp, #-16]!
|
|
; mov fp, sp
|
|
; str x28, [sp, #-16]!
|
|
; stp x26, x27, [sp, #-16]!
|
|
; stp x24, x25, [sp, #-16]!
|
|
; block0:
|
|
; mov x25, x0
|
|
; mov x26, x1
|
|
; atomic_rmw_loop_umin_8 addr=x25 operand=x26 oldval=x27 scratch1=x24 scratch2=x28
|
|
; ldp x24, x25, [sp], #16
|
|
; ldp x26, x27, [sp], #16
|
|
; ldr x28, [sp], #16
|
|
; ldp fp, lr, [sp], #16
|
|
; ret
|
|
|