[wasm-simd] Move i64x2.widen_i32x4_{s,u} to mvp and remove ifdefs

These instructions were accepted into the proposal:
https://github.com/WebAssembly/simd/pull/290

Bug: v8:10972
Change-Id: Ia2cce2df575786babe770b043b1e90bf953c5f9b
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/2643658Reviewed-by: Deepti Gandluri <gdeepti@chromium.org>
Commit-Queue: Zhi An Ng <zhin@chromium.org>
Cr-Commit-Position: refs/heads/master@{#72243}

src/compiler/backend/instruction-selector.cc

@@ -2749,28 +2749,6 @@ void InstructionSelector::VisitI64x2Eq(Node* node) { UNIMPLEMENTED(); }
         // && !V8_TARGET_ARCH_IA32 && !V8_TARGET_ARCH_ARM
         // && !V8_TARGET_ARCH_MIPS64 && !V8_TARGET_ARCH_MIPS
 
-#if !V8_TARGET_ARCH_ARM64 && !V8_TARGET_ARCH_X64 && !V8_TARGET_ARCH_IA32 && \
-    !V8_TARGET_ARCH_ARM && !V8_TARGET_ARCH_MIPS64 && !V8_TARGET_ARCH_MIPS
-// TODO(v8:10972) Prototype i64x2 widen i32x4.
-void InstructionSelector::VisitI64x2SConvertI32x4Low(Node* node) {
-  UNIMPLEMENTED();
-}
-
-void InstructionSelector::VisitI64x2SConvertI32x4High(Node* node) {
-  UNIMPLEMENTED();
-}
-
-void InstructionSelector::VisitI64x2UConvertI32x4Low(Node* node) {
-  UNIMPLEMENTED();
-}
-
-void InstructionSelector::VisitI64x2UConvertI32x4High(Node* node) {
-  UNIMPLEMENTED();
-}
-#endif  // !V8_TARGET_ARCH_ARM64 || !V8_TARGET_ARCH_X64 && !V8_TARGET_ARCH_IA32
-        // && !V8_TARGET_ARCH_ARM && !V8_TARGET_ARCH_MIPS64 &&
-        // !V8_TARGET_ARCH_MIPS
-
 #if !V8_TARGET_ARCH_ARM64 && !V8_TARGET_ARCH_X64 && !V8_TARGET_ARCH_IA32
 // TODO(v8:11168): Prototyping prefetch.
 void InstructionSelector::VisitPrefetchTemporal(Node* node) { UNIMPLEMENTED(); }

src/wasm/wasm-opcodes.h

@@ -443,6 +443,10 @@ bool V8_EXPORT_PRIVATE IsJSCompatibleSignature(const FunctionSig* sig,
   V(I64x2ExtMulHighI32x4S, 0xfdd3, s_ss)     \
   V(I64x2ExtMulLowI32x4U, 0xfdd6, s_ss)      \
   V(I64x2ExtMulHighI32x4U, 0xfdd7, s_ss)     \
+  V(I64x2SConvertI32x4Low, 0xfdc7, s_s)      \
+  V(I64x2SConvertI32x4High, 0xfdc8, s_s)     \
+  V(I64x2UConvertI32x4Low, 0xfdc9, s_s)      \
+  V(I64x2UConvertI32x4High, 0xfdca, s_s)     \
   V(F32x4Abs, 0xfde0, s_s)                   \
   V(F32x4Neg, 0xfde1, s_s)                   \
   V(F32x4Sqrt, 0xfde3, s_s)                  \
@@ -503,10 +507,6 @@ bool V8_EXPORT_PRIVATE IsJSCompatibleSignature(const FunctionSig* sig,
   V(I16x8ExtAddPairwiseI8x16U, 0xfdc3, s_s) \
   V(I64x2Eq, 0xfdc0, s_ss)                  \
   V(F32x4Qfma, 0xfdb4, s_sss)               \
-  V(I64x2SConvertI32x4Low, 0xfdc7, s_s)     \
-  V(I64x2SConvertI32x4High, 0xfdc8, s_s)    \
-  V(I64x2UConvertI32x4Low, 0xfdc9, s_s)     \
-  V(I64x2UConvertI32x4High, 0xfdca, s_s)    \
   V(F32x4Qfms, 0xfdd4, s_sss)               \
   V(F64x2Qfma, 0xfdfe, s_sss)               \
   V(F64x2Qfms, 0xfdff, s_sss)               \

test/cctest/wasm/test-run-wasm-simd.cc

@@ -1939,12 +1939,8 @@ WASM_SIMD_TEST(I32x4ConvertI16x8) {
   }
 }
 
-// TODO(v8:10972) Prototyping i64x2 convert from i32x4.
 // Tests both signed and unsigned conversion from I32x4 (unpacking).
-#if V8_TARGET_ARCH_ARM64 || V8_TARGET_ARCH_X64 || V8_TARGET_ARCH_IA32 || \
-    V8_TARGET_ARCH_ARM || V8_TARGET_ARCH_MIPS64 || V8_TARGET_ARCH_MIPS
 WASM_SIMD_TEST_NO_LOWERING(I64x2ConvertI32x4) {
-  FLAG_SCOPE(wasm_simd_post_mvp);
   WasmRunner<int32_t, int32_t> r(execution_tier, lower_simd);
   // Create four output vectors to hold signed and unsigned results.
   int64_t* g0 = r.builder().AddGlobal<int64_t>(kWasmS128);
@@ -1978,8 +1974,6 @@ WASM_SIMD_TEST_NO_LOWERING(I64x2ConvertI32x4) {
     }
   }
 }
-#endif  // V8_TARGET_ARCH_ARM64 || V8_TARGET_ARCH_X64 || V8_TARGET_ARCH_IA32 ||
-        // V8_TARGET_ARCH_ARM || V8_TARGET_ARCH_MIPS64 || V8_TARGET_ARCH_MIPS
 
 void RunI32x4UnOpTest(TestExecutionTier execution_tier, LowerSimd lower_simd,
                       WasmOpcode opcode, Int32UnOp expected_op) {