[wasm-simd][liftoff][ia32] Implement v64x2.alltrue and i64x2.ne

Bug: v8:11347,v8:11348
Change-Id: Ib9eb0c8d03af9e0d9171ba668e8720332659e187
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/2686010
Commit-Queue: Zhi An Ng <zhin@chromium.org>
Reviewed-by: Deepti Gandluri <gdeepti@chromium.org>
Reviewed-by: Clemens Backes <clemensb@chromium.org>
Cr-Commit-Position: refs/heads/master@{#72680}

src/codegen/ia32/macro-assembler-ia32.h

@@ -553,7 +553,8 @@ class V8_EXPORT_PRIVATE TurboAssembler : public TurboAssemblerBase {
 #undef AVX_OP3_WITH_TYPE_SCOPE
 
   void Haddps(XMMRegister dst, XMMRegister src1, Operand src2);
-  void Pcmpeqq(XMMRegister dst, Operand src2);
+  void Pcmpeqq(XMMRegister dst, Operand src);
+  void Pcmpeqq(XMMRegister dst, XMMRegister src) { Pcmpeqq(dst, Operand(src)); }
   void Pcmpeqq(XMMRegister dst, XMMRegister src1, Operand src2);
   void Pcmpeqq(XMMRegister dst, XMMRegister src1, XMMRegister src2);
   void Pshufb(XMMRegister dst, XMMRegister src) { Pshufb(dst, dst, src); }

src/wasm/baseline/ia32/liftoff-assembler-ia32.h

@@ -2698,7 +2698,11 @@ inline void EmitAnyTrue(LiftoffAssembler* assm, LiftoffRegister dst,
 
 template <void (TurboAssembler::*pcmp)(XMMRegister, XMMRegister)>
 inline void EmitAllTrue(LiftoffAssembler* assm, LiftoffRegister dst,
-                        LiftoffRegister src) {
+                        LiftoffRegister src,
+                        base::Optional<CpuFeature> feature = base::nullopt) {
+  base::Optional<CpuFeatureScope> sse_scope;
+  if (feature.has_value()) sse_scope.emplace(assm, *feature);
+
   Register tmp =
       assm->GetUnusedRegister(kGpReg, LiftoffRegList::ForRegs(dst)).gp();
   XMMRegister tmp_simd = liftoff::kScratchDoubleReg;
@@ -3113,7 +3117,10 @@ void LiftoffAssembler::emit_i64x2_eq(LiftoffRegister dst, LiftoffRegister lhs,
 
 void LiftoffAssembler::emit_i64x2_ne(LiftoffRegister dst, LiftoffRegister lhs,
                                      LiftoffRegister rhs) {
-  bailout(kSimd, "i64x2_ne");
+  liftoff::EmitSimdCommutativeBinOp<&Assembler::vpcmpeqq, &Assembler::pcmpeqq>(
+      this, dst, lhs, rhs, SSE4_1);
+  Pcmpeqq(liftoff::kScratchDoubleReg, liftoff::kScratchDoubleReg);
+  Pxor(dst.fp(), liftoff::kScratchDoubleReg);
 }
 
 void LiftoffAssembler::emit_f32x4_eq(LiftoffRegister dst, LiftoffRegister lhs,
@@ -3826,7 +3833,7 @@ void LiftoffAssembler::emit_i64x2_neg(LiftoffRegister dst,
 
 void LiftoffAssembler::emit_v64x2_alltrue(LiftoffRegister dst,
                                           LiftoffRegister src) {
-  bailout(kSimd, "v64x2_alltrue");
+  liftoff::EmitAllTrue<&TurboAssembler::Pcmpeqq>(this, dst, src, SSE4_1);
 }
 
 void LiftoffAssembler::emit_i64x2_shl(LiftoffRegister dst, LiftoffRegister lhs,