A lightweight WebAssembly runtime that is fast, secure, and standards-compliant
;;! target = "x86_64"
;;! test = "optimize"
;;! filter = "array_to_wasm_trampoline"
(module
(func (export "f") (param i32 i64) (result i32 i64)
(local.get 0)
(local.get 1)
)
)
;; function u268435456:0(i64 vmctx, i64, i64, i64) -> i8 system_v {
;; region0 = 17 ""
;; region1 = 123 ""
;; region2 = 231 ""
;; region3 = 243 ""
;; region4 = 209 ""
;; region5 = 68 ""
;; region6 = 30 ""
;; sig0 = (i64 vmctx, i64, i32, i64) -> i32, i64 tail
;; fn0 = colocated u0:0 sig0
;;
;; block0(v0: i64, v1: i64, v2: i64, v3: i64):
;; jump block1
;;
;; block1:
;; v4 = iconst.i64 2
;; v5 = icmp.i64 uge v3, v4 ; v4 = 2
;; trapz v5, user1
;; v6 = load.i32 notrap little region0 v2
;; v7 = load.i64 notrap little region0 v2+16
;; v9 = get_frame_pointer.i64
;; v8 = load.i64 notrap aligned readonly can_move region1 v0+8
;; store notrap aligned region2 v9, v8+72
;; v10 = get_stack_pointer.i64
;; store notrap aligned region3 v10, v8+64
;; v11 = get_exception_handler_address.i64 block1, 0
;; store notrap aligned region4 v11, v8+80
;; try_call fn0(v0, v1, v6, v7), sig0, block2(ret0, ret1), [ default: block3 ]
;;
;; block2(v12: i32, v13: i64):
;; store notrap little region0 v12, v2
;; store notrap little region0 v13, v2+16
;; v14 = iconst.i8 1
;; return v14 ; v14 = 1
;;
;; block3:
;; v15 = iconst.i64 1
;; store notrap aligned region5 v15, v8+136 ; v15 = 1
;; v16 = iconst.i64 0
;; store notrap aligned region6 v16, v8+152 ; v16 = 0
;; v17 = iconst.i8 0
;; return v17 ; v17 = 0
;; }