From 17620d18db8d4a67e4b917596c760107d26fadc5 Mon Sep 17 00:00:00 2001 From: Wunk Date: Mon, 20 Mar 2023 12:09:24 -0700 Subject: ARMeilleure: Add initial support for AVX512 (EVEX encoding) (cont) (#4147) * ARMeilleure: Add AVX512{F,VL,DQ,BW} detection Add `UseAvx512Ortho` and `UseAvx512OrthoFloat` optimization flags as short-hands for `F+VL` and `F+VL+DQ`. * ARMeilleure: Add initial support for EVEX instruction encoding Does not implement rounding, or exception controls. * ARMeilleure: Add `X86Vpternlogd` Accelerates the vector-`Not` instruction. * ARMeilleure: Add check for `OSXSAVE` for AVX{2,512} * ARMeilleure: Add check for `XCR0` flags Add XCR0 register checks for AVX and AVX512F, following the guidelines from section 14.3 and 15.2 from the Intel Architecture Software Developer's Manual. * ARMeilleure: Remove redundant `ReProtect` and `Dispose`, formatting * ARMeilleure: Move XCR0 procedure to GetXcr0Eax * ARMeilleure: Add `XCR0` to `FeatureInfo` structure * ARMeilleure: Utilize `ReadOnlySpan` for Xcr0 assembly Avoids an additional allocation * ARMeilleure: Formatting fixes * ARMeilleure: Fix EVEX encoding src2 register index > Just like in VEX prefix, vvvv is provided in inverted form. * ARMeilleure: Add `X86Vpternlogd` acceleration to `Vmvn_I` Passes unit tests, verified instruction utilization * ARMeilleure: Fix EVEX register operand designations Operand 2 was being sourced improperly. EVEX encoded instructions source their operands like so: Operand 1: ModRM:reg Operand 2: EVEX.vvvvv Operand 3: ModRM:r/m Operand 4: Imm This fixes the improper register designations when emitting vpternlog. Now "dest", "src1", "src2" arguments emit in the proper order in EVEX instructions. * ARMeilleure: Add `X86Vpternlogd` acceleration to `Orn_V` * ARMeilleure: PTC version bump * ARMeilleure: Update EVEX encoding Debug.Assert to Debug.Fail * ARMeilleure: Update EVEX encoding comment capitalization --- ARMeilleure/Instructions/InstEmitSimdLogical.cs | 33 ++++++++++++++++++++++++- 1 file changed, 32 insertions(+), 1 deletion(-) (limited to 'ARMeilleure/Instructions/InstEmitSimdLogical.cs') diff --git a/ARMeilleure/Instructions/InstEmitSimdLogical.cs b/ARMeilleure/Instructions/InstEmitSimdLogical.cs index 8ca81580..2bf531e6 100644 --- a/ARMeilleure/Instructions/InstEmitSimdLogical.cs +++ b/ARMeilleure/Instructions/InstEmitSimdLogical.cs @@ -254,7 +254,22 @@ namespace ARMeilleure.Instructions public static void Not_V(ArmEmitterContext context) { - if (Optimizations.UseSse2) + if (Optimizations.UseAvx512Ortho) + { + OpCodeSimd op = (OpCodeSimd)context.CurrOp; + + Operand n = GetVec(op.Rn); + + Operand res = context.AddIntrinsic(Intrinsic.X86Vpternlogd, n, n, Const(~0b10101010)); + + if (op.RegisterSize == RegisterSize.Simd64) + { + res = context.VectorZeroUpper64(res); + } + + context.Copy(GetVec(op.Rd), res); + } + else if (Optimizations.UseSse2) { OpCodeSimd op = (OpCodeSimd)context.CurrOp; @@ -283,6 +298,22 @@ namespace ARMeilleure.Instructions { InstEmitSimdHelperArm64.EmitVectorBinaryOp(context, Intrinsic.Arm64OrnV); } + else if (Optimizations.UseAvx512Ortho) + { + OpCodeSimdReg op = (OpCodeSimdReg)context.CurrOp; + + Operand n = GetVec(op.Rn); + Operand m = GetVec(op.Rm); + + Operand res = context.AddIntrinsic(Intrinsic.X86Vpternlogd, n, m, Const(0b11001100 | ~0b10101010)); + + if (op.RegisterSize == RegisterSize.Simd64) + { + res = context.VectorZeroUpper64(res); + } + + context.Copy(GetVec(op.Rd), res); + } else if (Optimizations.UseSse2) { OpCodeSimdReg op = (OpCodeSimdReg)context.CurrOp; -- cgit v1.2.3