Fix f32x4 min and max in wasm-interpreter

-0.0 and 0.0 compare equals, so a < b ? a : b for min would pick b
incorrectly. We need to use JS semantics here, which returns -0.0.

Bug: v8:8425
Change-Id: I8ab094b566ece9c586de86aad4594dfdf8da930b
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/1724802Reviewed-by: Deepti Gandluri <gdeepti@chromium.org>
Commit-Queue: Zhi An Ng <zhin@chromium.org>
Cr-Commit-Position: refs/heads/master@{#62969}

src/wasm/wasm-interpreter.cc

@@ -2247,8 +2247,8 @@ class ThreadImpl {
       BINOP_CASE(F32x4Add, f32x4, float4, 4, a + b)
       BINOP_CASE(F32x4Sub, f32x4, float4, 4, a - b)
       BINOP_CASE(F32x4Mul, f32x4, float4, 4, a * b)
-      BINOP_CASE(F32x4Min, f32x4, float4, 4, a < b ? a : b)
-      BINOP_CASE(F32x4Max, f32x4, float4, 4, a > b ? a : b)
+      BINOP_CASE(F32x4Min, f32x4, float4, 4, JSMin(a, b))
+      BINOP_CASE(F32x4Max, f32x4, float4, 4, JSMax(a, b))
       BINOP_CASE(I64x2Add, i64x2, int2, 2, base::AddWithWraparound(a, b))
       BINOP_CASE(I64x2Sub, i64x2, int2, 2, base::SubWithWraparound(a, b))
       BINOP_CASE(I64x2Mul, i64x2, int2, 2, base::MulWithWraparound(a, b))

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

@@ -654,11 +654,11 @@ WASM_SIMD_TEST(F32x4Mul) {
   RunF32x4BinOpTest(execution_tier, lower_simd, kExprF32x4Mul, Mul);
 }
 // v8:8425 tracks this test being enabled in the interpreter.
-WASM_SIMD_COMPILED_TEST(F32x4Min) {
+WASM_SIMD_TEST(F32x4Min) {
   RunF32x4BinOpTest(execution_tier, lower_simd, kExprF32x4Min, JSMin);
 }
 // v8:8425 tracks this test being enabled in the interpreter.
-WASM_SIMD_COMPILED_TEST(F32x4Max) {
+WASM_SIMD_TEST(F32x4Max) {
   RunF32x4BinOpTest(execution_tier, lower_simd, kExprF32x4Max, JSMax);
 }