[dev.simd] cmd/compile, simd: change AES op names and add missing size

This CL changed AESEncryptRound and AESDecryptRound to
AESEncryptOneRound and AESDecryptOneRound.

This CL also adds the 512-bit version of some AES instructions.

Change-Id: Ia851a008cce2145b1ff193a89e172862060a725d
Reviewed-on: https://go-review.googlesource.com/c/go/+/721280
LUCI-TryBot-Result: Go LUCI <golang-scoped@luci-project-accounts.iam.gserviceaccount.com>
Reviewed-by: David Chase <drchase@google.com>

src/cmd/compile/internal/amd64/simdssa.go

@@ -178,12 +178,16 @@ func ssaGenSIMDValue(s *ssagen.State, v *ssa.Value) bool {
 
 	case ssa.OpAMD64VAESDECLAST128,
 		ssa.OpAMD64VAESDECLAST256,
+		ssa.OpAMD64VAESDECLAST512,
 		ssa.OpAMD64VAESDEC128,
 		ssa.OpAMD64VAESDEC256,
+		ssa.OpAMD64VAESDEC512,
 		ssa.OpAMD64VAESENCLAST128,
 		ssa.OpAMD64VAESENCLAST256,
+		ssa.OpAMD64VAESENCLAST512,
 		ssa.OpAMD64VAESENC128,
 		ssa.OpAMD64VAESENC256,
+		ssa.OpAMD64VAESENC512,
 		ssa.OpAMD64VADDPS128,
 		ssa.OpAMD64VADDPS256,
 		ssa.OpAMD64VADDPS512,

src/cmd/compile/internal/ssa/_gen/simdAMD64.rules

@@ -2,12 +2,16 @@
 
 (AESDecryptLastRoundUint8x16 ...) => (VAESDECLAST128 ...)
 (AESDecryptLastRoundUint8x32 ...) => (VAESDECLAST256 ...)
-(AESDecryptRoundUint8x16 ...) => (VAESDEC128 ...)
+(AESDecryptLastRoundUint8x64 ...) => (VAESDECLAST512 ...)
-(AESDecryptRoundUint8x32 ...) => (VAESDEC256 ...)
+(AESDecryptOneRoundUint8x16 ...) => (VAESDEC128 ...)
+(AESDecryptOneRoundUint8x32 ...) => (VAESDEC256 ...)
+(AESDecryptOneRoundUint8x64 ...) => (VAESDEC512 ...)
 (AESEncryptLastRoundUint8x16 ...) => (VAESENCLAST128 ...)
 (AESEncryptLastRoundUint8x32 ...) => (VAESENCLAST256 ...)
-(AESEncryptRoundUint8x16 ...) => (VAESENC128 ...)
+(AESEncryptLastRoundUint8x64 ...) => (VAESENCLAST512 ...)
-(AESEncryptRoundUint8x32 ...) => (VAESENC256 ...)
+(AESEncryptOneRoundUint8x16 ...) => (VAESENC128 ...)
+(AESEncryptOneRoundUint8x32 ...) => (VAESENC256 ...)
+(AESEncryptOneRoundUint8x64 ...) => (VAESENC512 ...)
 (AESInvMixColumnsUint32x4 ...) => (VAESIMC128 ...)
 (AESRoundKeyGenAssistUint32x4 ...) => (VAESKEYGENASSIST128 ...)
 (AbsInt8x16 ...) => (VPABSB128 ...)

src/cmd/compile/internal/ssa/_gen/simdAMD64ops.go

@@ -28,12 +28,16 @@ func simdAMD64Ops(v11, v21, v2k, vkv, v2kv, v2kk, v31, v3kv, vgpv, vgp, vfpv, vf
 		{name: "VADDSUBPS256", argLength: 2, reg: v21, asm: "VADDSUBPS", commutative: false, typ: "Vec256", resultInArg0: false},
 		{name: "VAESDEC128", argLength: 2, reg: v21, asm: "VAESDEC", commutative: false, typ: "Vec128", resultInArg0: false},
 		{name: "VAESDEC256", argLength: 2, reg: w21, asm: "VAESDEC", commutative: false, typ: "Vec256", resultInArg0: false},
+		{name: "VAESDEC512", argLength: 2, reg: w21, asm: "VAESDEC", commutative: false, typ: "Vec512", resultInArg0: false},
 		{name: "VAESDECLAST128", argLength: 2, reg: v21, asm: "VAESDECLAST", commutative: false, typ: "Vec128", resultInArg0: false},
 		{name: "VAESDECLAST256", argLength: 2, reg: w21, asm: "VAESDECLAST", commutative: false, typ: "Vec256", resultInArg0: false},
+		{name: "VAESDECLAST512", argLength: 2, reg: w21, asm: "VAESDECLAST", commutative: false, typ: "Vec512", resultInArg0: false},
 		{name: "VAESENC128", argLength: 2, reg: v21, asm: "VAESENC", commutative: false, typ: "Vec128", resultInArg0: false},
 		{name: "VAESENC256", argLength: 2, reg: w21, asm: "VAESENC", commutative: false, typ: "Vec256", resultInArg0: false},
+		{name: "VAESENC512", argLength: 2, reg: w21, asm: "VAESENC", commutative: false, typ: "Vec512", resultInArg0: false},
 		{name: "VAESENCLAST128", argLength: 2, reg: v21, asm: "VAESENCLAST", commutative: false, typ: "Vec128", resultInArg0: false},
 		{name: "VAESENCLAST256", argLength: 2, reg: w21, asm: "VAESENCLAST", commutative: false, typ: "Vec256", resultInArg0: false},
+		{name: "VAESENCLAST512", argLength: 2, reg: w21, asm: "VAESENCLAST", commutative: false, typ: "Vec512", resultInArg0: false},
 		{name: "VAESIMC128", argLength: 1, reg: v11, asm: "VAESIMC", commutative: false, typ: "Vec128", resultInArg0: false},
 		{name: "VBROADCASTSD256", argLength: 1, reg: v11, asm: "VBROADCASTSD", commutative: false, typ: "Vec256", resultInArg0: false},
 		{name: "VBROADCASTSD512", argLength: 1, reg: w11, asm: "VBROADCASTSD", commutative: false, typ: "Vec512", resultInArg0: false},

src/cmd/compile/internal/ssa/_gen/simdgenericOps.go

@@ -6,12 +6,16 @@ func simdGenericOps() []opData {
 	return []opData{
 		{name: "AESDecryptLastRoundUint8x16", argLength: 2, commutative: false},
 		{name: "AESDecryptLastRoundUint8x32", argLength: 2, commutative: false},
-		{name: "AESDecryptRoundUint8x16", argLength: 2, commutative: false},
+		{name: "AESDecryptLastRoundUint8x64", argLength: 2, commutative: false},
-		{name: "AESDecryptRoundUint8x32", argLength: 2, commutative: false},
+		{name: "AESDecryptOneRoundUint8x16", argLength: 2, commutative: false},
+		{name: "AESDecryptOneRoundUint8x32", argLength: 2, commutative: false},
+		{name: "AESDecryptOneRoundUint8x64", argLength: 2, commutative: false},
 		{name: "AESEncryptLastRoundUint8x16", argLength: 2, commutative: false},
 		{name: "AESEncryptLastRoundUint8x32", argLength: 2, commutative: false},
-		{name: "AESEncryptRoundUint8x16", argLength: 2, commutative: false},
+		{name: "AESEncryptLastRoundUint8x64", argLength: 2, commutative: false},
-		{name: "AESEncryptRoundUint8x32", argLength: 2, commutative: false},
+		{name: "AESEncryptOneRoundUint8x16", argLength: 2, commutative: false},
+		{name: "AESEncryptOneRoundUint8x32", argLength: 2, commutative: false},
+		{name: "AESEncryptOneRoundUint8x64", argLength: 2, commutative: false},
 		{name: "AESInvMixColumnsUint32x4", argLength: 1, commutative: false},
 		{name: "AbsInt8x16", argLength: 1, commutative: false},
 		{name: "AbsInt8x32", argLength: 1, commutative: false},

src/cmd/compile/internal/ssa/opGen.go

@@ -1269,12 +1269,16 @@ const (
 	OpAMD64VADDSUBPS256
 	OpAMD64VAESDEC128
 	OpAMD64VAESDEC256
+	OpAMD64VAESDEC512
 	OpAMD64VAESDECLAST128
 	OpAMD64VAESDECLAST256
+	OpAMD64VAESDECLAST512
 	OpAMD64VAESENC128
 	OpAMD64VAESENC256
+	OpAMD64VAESENC512
 	OpAMD64VAESENCLAST128
 	OpAMD64VAESENCLAST256
+	OpAMD64VAESENCLAST512
 	OpAMD64VAESIMC128
 	OpAMD64VBROADCASTSD256
 	OpAMD64VBROADCASTSD512
@@ -5950,12 +5954,16 @@ const (
 	OpIsZeroVec
 	OpAESDecryptLastRoundUint8x16
 	OpAESDecryptLastRoundUint8x32
-	OpAESDecryptRoundUint8x16
+	OpAESDecryptLastRoundUint8x64
-	OpAESDecryptRoundUint8x32
+	OpAESDecryptOneRoundUint8x16
+	OpAESDecryptOneRoundUint8x32
+	OpAESDecryptOneRoundUint8x64
 	OpAESEncryptLastRoundUint8x16
 	OpAESEncryptLastRoundUint8x32
-	OpAESEncryptRoundUint8x16
+	OpAESEncryptLastRoundUint8x64
-	OpAESEncryptRoundUint8x32
+	OpAESEncryptOneRoundUint8x16
+	OpAESEncryptOneRoundUint8x32
+	OpAESEncryptOneRoundUint8x64
 	OpAESInvMixColumnsUint32x4
 	OpAbsInt8x16
 	OpAbsInt8x32
@@ -20873,6 +20881,20 @@ var opcodeTable = [...]opInfo{
 			},
 		},
 	},
+	{
+		name:   "VAESDEC512",
+		argLen: 2,
+		asm:    x86.AVAESDEC,
+		reg: regInfo{
+			inputs: []inputInfo{
+				{0, 281474976645120}, // X0 X1 X2 X3 X4 X5 X6 X7 X8 X9 X10 X11 X12 X13 X14 X15 X16 X17 X18 X19 X20 X21 X22 X23 X24 X25 X26 X27 X28 X29 X30 X31
+				{1, 281474976645120}, // X0 X1 X2 X3 X4 X5 X6 X7 X8 X9 X10 X11 X12 X13 X14 X15 X16 X17 X18 X19 X20 X21 X22 X23 X24 X25 X26 X27 X28 X29 X30 X31
+			},
+			outputs: []outputInfo{
+				{0, 281472829161472}, // X0 X1 X2 X3 X4 X5 X6 X7 X8 X9 X10 X11 X12 X13 X14 X16 X17 X18 X19 X20 X21 X22 X23 X24 X25 X26 X27 X28 X29 X30 X31
+			},
+		},
+	},
 	{
 		name:   "VAESDECLAST128",
 		argLen: 2,
@@ -20901,6 +20923,20 @@ var opcodeTable = [...]opInfo{
 			},
 		},
 	},
+	{
+		name:   "VAESDECLAST512",
+		argLen: 2,
+		asm:    x86.AVAESDECLAST,
+		reg: regInfo{
+			inputs: []inputInfo{
+				{0, 281474976645120}, // X0 X1 X2 X3 X4 X5 X6 X7 X8 X9 X10 X11 X12 X13 X14 X15 X16 X17 X18 X19 X20 X21 X22 X23 X24 X25 X26 X27 X28 X29 X30 X31
+				{1, 281474976645120}, // X0 X1 X2 X3 X4 X5 X6 X7 X8 X9 X10 X11 X12 X13 X14 X15 X16 X17 X18 X19 X20 X21 X22 X23 X24 X25 X26 X27 X28 X29 X30 X31
+			},
+			outputs: []outputInfo{
+				{0, 281472829161472}, // X0 X1 X2 X3 X4 X5 X6 X7 X8 X9 X10 X11 X12 X13 X14 X16 X17 X18 X19 X20 X21 X22 X23 X24 X25 X26 X27 X28 X29 X30 X31
+			},
+		},
+	},
 	{
 		name:   "VAESENC128",
 		argLen: 2,
@@ -20929,6 +20965,20 @@ var opcodeTable = [...]opInfo{
 			},
 		},
 	},
+	{
+		name:   "VAESENC512",
+		argLen: 2,
+		asm:    x86.AVAESENC,
+		reg: regInfo{
+			inputs: []inputInfo{
+				{0, 281474976645120}, // X0 X1 X2 X3 X4 X5 X6 X7 X8 X9 X10 X11 X12 X13 X14 X15 X16 X17 X18 X19 X20 X21 X22 X23 X24 X25 X26 X27 X28 X29 X30 X31
+				{1, 281474976645120}, // X0 X1 X2 X3 X4 X5 X6 X7 X8 X9 X10 X11 X12 X13 X14 X15 X16 X17 X18 X19 X20 X21 X22 X23 X24 X25 X26 X27 X28 X29 X30 X31
+			},
+			outputs: []outputInfo{
+				{0, 281472829161472}, // X0 X1 X2 X3 X4 X5 X6 X7 X8 X9 X10 X11 X12 X13 X14 X16 X17 X18 X19 X20 X21 X22 X23 X24 X25 X26 X27 X28 X29 X30 X31
+			},
+		},
+	},
 	{
 		name:   "VAESENCLAST128",
 		argLen: 2,
@@ -20957,6 +21007,20 @@ var opcodeTable = [...]opInfo{
 			},
 		},
 	},
+	{
+		name:   "VAESENCLAST512",
+		argLen: 2,
+		asm:    x86.AVAESENCLAST,
+		reg: regInfo{
+			inputs: []inputInfo{
+				{0, 281474976645120}, // X0 X1 X2 X3 X4 X5 X6 X7 X8 X9 X10 X11 X12 X13 X14 X15 X16 X17 X18 X19 X20 X21 X22 X23 X24 X25 X26 X27 X28 X29 X30 X31
+				{1, 281474976645120}, // X0 X1 X2 X3 X4 X5 X6 X7 X8 X9 X10 X11 X12 X13 X14 X15 X16 X17 X18 X19 X20 X21 X22 X23 X24 X25 X26 X27 X28 X29 X30 X31
+			},
+			outputs: []outputInfo{
+				{0, 281472829161472}, // X0 X1 X2 X3 X4 X5 X6 X7 X8 X9 X10 X11 X12 X13 X14 X16 X17 X18 X19 X20 X21 X22 X23 X24 X25 X26 X27 X28 X29 X30 X31
+			},
+		},
+	},
 	{
 		name:   "VAESIMC128",
 		argLen: 1,
@@ -85740,12 +85804,22 @@ var opcodeTable = [...]opInfo{
 		generic: true,
 	},
 	{
-		name:    "AESDecryptRoundUint8x16",
+		name:    "AESDecryptLastRoundUint8x64",
 		argLen:  2,
 		generic: true,
 	},
 	{
-		name:    "AESDecryptRoundUint8x32",
+		name:    "AESDecryptOneRoundUint8x16",
+		argLen:  2,
+		generic: true,
+	},
+	{
+		name:    "AESDecryptOneRoundUint8x32",
+		argLen:  2,
+		generic: true,
+	},
+	{
+		name:    "AESDecryptOneRoundUint8x64",
 		argLen:  2,
 		generic: true,
 	},
@@ -85760,12 +85834,22 @@ var opcodeTable = [...]opInfo{
 		generic: true,
 	},
 	{
-		name:    "AESEncryptRoundUint8x16",
+		name:    "AESEncryptLastRoundUint8x64",
 		argLen:  2,
 		generic: true,
 	},
 	{
-		name:    "AESEncryptRoundUint8x32",
+		name:    "AESEncryptOneRoundUint8x16",
+		argLen:  2,
+		generic: true,
+	},
+	{
+		name:    "AESEncryptOneRoundUint8x32",
+		argLen:  2,
+		generic: true,
+	},
+	{
+		name:    "AESEncryptOneRoundUint8x64",
 		argLen:  2,
 		generic: true,
 	},

src/cmd/compile/internal/ssa/rewriteAMD64.go

@@ -16,24 +16,36 @@ func rewriteValueAMD64(v *Value) bool {
 	case OpAESDecryptLastRoundUint8x32:
 		v.Op = OpAMD64VAESDECLAST256
 		return true
-	case OpAESDecryptRoundUint8x16:
+	case OpAESDecryptLastRoundUint8x64:
+		v.Op = OpAMD64VAESDECLAST512
+		return true
+	case OpAESDecryptOneRoundUint8x16:
 		v.Op = OpAMD64VAESDEC128
 		return true
-	case OpAESDecryptRoundUint8x32:
+	case OpAESDecryptOneRoundUint8x32:
 		v.Op = OpAMD64VAESDEC256
 		return true
+	case OpAESDecryptOneRoundUint8x64:
+		v.Op = OpAMD64VAESDEC512
+		return true
 	case OpAESEncryptLastRoundUint8x16:
 		v.Op = OpAMD64VAESENCLAST128
 		return true
 	case OpAESEncryptLastRoundUint8x32:
 		v.Op = OpAMD64VAESENCLAST256
 		return true
-	case OpAESEncryptRoundUint8x16:
+	case OpAESEncryptLastRoundUint8x64:
+		v.Op = OpAMD64VAESENCLAST512
+		return true
+	case OpAESEncryptOneRoundUint8x16:
 		v.Op = OpAMD64VAESENC128
 		return true
-	case OpAESEncryptRoundUint8x32:
+	case OpAESEncryptOneRoundUint8x32:
 		v.Op = OpAMD64VAESENC256
 		return true
+	case OpAESEncryptOneRoundUint8x64:
+		v.Op = OpAMD64VAESENC512
+		return true
 	case OpAESInvMixColumnsUint32x4:
 		v.Op = OpAMD64VAESIMC128
 		return true

src/cmd/compile/internal/ssagen/simdintrinsics.go

@@ -14,12 +14,16 @@ const simdPackage = "simd"
 func simdIntrinsics(addF func(pkg, fn string, b intrinsicBuilder, archFamilies ...sys.ArchFamily)) {
 	addF(simdPackage, "Uint8x16.AESDecryptLastRound", opLen2(ssa.OpAESDecryptLastRoundUint8x16, types.TypeVec128), sys.AMD64)
 	addF(simdPackage, "Uint8x32.AESDecryptLastRound", opLen2(ssa.OpAESDecryptLastRoundUint8x32, types.TypeVec256), sys.AMD64)
-	addF(simdPackage, "Uint8x16.AESDecryptRound", opLen2(ssa.OpAESDecryptRoundUint8x16, types.TypeVec128), sys.AMD64)
+	addF(simdPackage, "Uint8x64.AESDecryptLastRound", opLen2(ssa.OpAESDecryptLastRoundUint8x64, types.TypeVec512), sys.AMD64)
-	addF(simdPackage, "Uint8x32.AESDecryptRound", opLen2(ssa.OpAESDecryptRoundUint8x32, types.TypeVec256), sys.AMD64)
+	addF(simdPackage, "Uint8x16.AESDecryptOneRound", opLen2(ssa.OpAESDecryptOneRoundUint8x16, types.TypeVec128), sys.AMD64)
+	addF(simdPackage, "Uint8x32.AESDecryptOneRound", opLen2(ssa.OpAESDecryptOneRoundUint8x32, types.TypeVec256), sys.AMD64)
+	addF(simdPackage, "Uint8x64.AESDecryptOneRound", opLen2(ssa.OpAESDecryptOneRoundUint8x64, types.TypeVec512), sys.AMD64)
 	addF(simdPackage, "Uint8x16.AESEncryptLastRound", opLen2(ssa.OpAESEncryptLastRoundUint8x16, types.TypeVec128), sys.AMD64)
 	addF(simdPackage, "Uint8x32.AESEncryptLastRound", opLen2(ssa.OpAESEncryptLastRoundUint8x32, types.TypeVec256), sys.AMD64)
-	addF(simdPackage, "Uint8x16.AESEncryptRound", opLen2(ssa.OpAESEncryptRoundUint8x16, types.TypeVec128), sys.AMD64)
+	addF(simdPackage, "Uint8x64.AESEncryptLastRound", opLen2(ssa.OpAESEncryptLastRoundUint8x64, types.TypeVec512), sys.AMD64)
-	addF(simdPackage, "Uint8x32.AESEncryptRound", opLen2(ssa.OpAESEncryptRoundUint8x32, types.TypeVec256), sys.AMD64)
+	addF(simdPackage, "Uint8x16.AESEncryptOneRound", opLen2(ssa.OpAESEncryptOneRoundUint8x16, types.TypeVec128), sys.AMD64)
+	addF(simdPackage, "Uint8x32.AESEncryptOneRound", opLen2(ssa.OpAESEncryptOneRoundUint8x32, types.TypeVec256), sys.AMD64)
+	addF(simdPackage, "Uint8x64.AESEncryptOneRound", opLen2(ssa.OpAESEncryptOneRoundUint8x64, types.TypeVec512), sys.AMD64)
 	addF(simdPackage, "Uint32x4.AESInvMixColumns", opLen1(ssa.OpAESInvMixColumnsUint32x4, types.TypeVec128), sys.AMD64)
 	addF(simdPackage, "Uint32x4.AESRoundKeyGenAssist", opLen1Imm8(ssa.OpAESRoundKeyGenAssistUint32x4, types.TypeVec128, 0), sys.AMD64)
 	addF(simdPackage, "Int8x16.Abs", opLen1(ssa.OpAbsInt8x16, types.TypeVec128), sys.AMD64)

src/simd/_gen/simdgen/ops/Others/categories.yaml

@@ -3,7 +3,7 @@
   commutative: false
   documentation: !string |-
     // NAME counts the leading zeros of each element in x.
-- go: AESEncryptRound
+- go: AESEncryptOneRound
   commutative: false
   documentation: !string |-
     // NAME performs a series of operations in AES cipher algorithm defined in FIPS 197.
@@ -27,7 +27,7 @@
     // result[1] = SubWord(x[1])
     // result[2] = XOR(SubWord(RotWord(x[2])), r)
     // result[3] = SubWord(x[3])
-- go: AESDecryptRound
+- go: AESDecryptOneRound
   commutative: false
   documentation: !string |-
     // NAME performs a series of operations in AES cipher algorithm defined in FIPS 197.

src/simd/_gen/simdgen/ops/Others/go.yaml

@@ -6,7 +6,7 @@
     go: $t
   out:
   - *any
-- go: AESEncryptRound
+- go: AESEncryptOneRound
   asm: VAESENC
   in:
   - &uint8s
@@ -33,7 +33,7 @@
     name: rconVal
   out:
   - *uint32s
-- go: AESDecryptRound
+- go: AESDecryptOneRound
   asm: VAESDEC
   in:
   - *uint8s

src/simd/_gen/simdgen/types.yaml

@@ -83,6 +83,17 @@ in: !repeat
   - {class: vreg, go: Int64x4,    base: "int",   elemBits: 128, bits: 256, lanes: 4}
   - {class: vreg, go: Uint64x4,   base: "uint",  elemBits: 128, bits: 256, lanes: 4}
 
+# Special shapes just to make VAES(ENC|DEC)(LAST)?512 work.
+# The elemBits field of these shapes are wrong, it would be overwritten by overwriteElemBits.
+  - {class: vreg, go: Int8x32,    base: "int",   elemBits: 128, bits: 512, lanes: 32}
+  - {class: vreg, go: Uint8x32,   base: "uint",  elemBits: 128, bits: 512, lanes: 32}
+  - {class: vreg, go: Int16x16,   base: "int",   elemBits: 128, bits: 512, lanes: 16}
+  - {class: vreg, go: Uint16x16,  base: "uint",  elemBits: 128, bits: 512, lanes: 16}
+  - {class: vreg, go: Int32x8,    base: "int",   elemBits: 128, bits: 512, lanes: 8}
+  - {class: vreg, go: Uint32x8,   base: "uint",  elemBits: 128, bits: 512, lanes: 8}
+  - {class: vreg, go: Int64x4,    base: "int",   elemBits: 128, bits: 512, lanes: 4}
+  - {class: vreg, go: Uint64x4,   base: "uint",  elemBits: 128, bits: 512, lanes: 4}
+
   - {class: immediate, go: Immediate} # TODO: we only support imms that are not used as value -- usually as instruction semantic predicate like VPCMP as of now.
 inVariant: !repeat
 - *types

src/simd/ops_amd64.go

@@ -22,23 +22,39 @@ func (x Uint8x16) AESDecryptLastRound(y Uint32x4) Uint8x16
 // Asm: VAESDECLAST, CPU Feature: AVX512VAES
 func (x Uint8x32) AESDecryptLastRound(y Uint32x8) Uint8x32
 
-/* AESDecryptRound */
+// AESDecryptLastRound performs a series of operations in AES cipher algorithm defined in FIPS 197.
+// x is the state array, starting from low index to high are s00, s10, s20, s30, s01, ..., s33.
+// y is the chunk of dw array in use.
+// result = AddRoundKey(InvShiftRows(InvSubBytes(x)), y)
+//
+// Asm: VAESDECLAST, CPU Feature: AVX512VAES
+func (x Uint8x64) AESDecryptLastRound(y Uint32x16) Uint8x64
 
-// AESDecryptRound performs a series of operations in AES cipher algorithm defined in FIPS 197.
+/* AESDecryptOneRound */
+
+// AESDecryptOneRound performs a series of operations in AES cipher algorithm defined in FIPS 197.
 // x is the state array, starting from low index to high are s00, s10, s20, s30, s01, ..., s33.
 // y is the chunk of dw array in use.
 // result = AddRoundKey(InvMixColumns(InvShiftRows(InvSubBytes(x))), y)
 //
 // Asm: VAESDEC, CPU Feature: AVX, AES
-func (x Uint8x16) AESDecryptRound(y Uint32x4) Uint8x16
+func (x Uint8x16) AESDecryptOneRound(y Uint32x4) Uint8x16
 
-// AESDecryptRound performs a series of operations in AES cipher algorithm defined in FIPS 197.
+// AESDecryptOneRound performs a series of operations in AES cipher algorithm defined in FIPS 197.
 // x is the state array, starting from low index to high are s00, s10, s20, s30, s01, ..., s33.
 // y is the chunk of dw array in use.
 // result = AddRoundKey(InvMixColumns(InvShiftRows(InvSubBytes(x))), y)
 //
 // Asm: VAESDEC, CPU Feature: AVX512VAES
-func (x Uint8x32) AESDecryptRound(y Uint32x8) Uint8x32
+func (x Uint8x32) AESDecryptOneRound(y Uint32x8) Uint8x32
+
+// AESDecryptOneRound performs a series of operations in AES cipher algorithm defined in FIPS 197.
+// x is the state array, starting from low index to high are s00, s10, s20, s30, s01, ..., s33.
+// y is the chunk of dw array in use.
+// result = AddRoundKey(InvMixColumns(InvShiftRows(InvSubBytes(x))), y)
+//
+// Asm: VAESDEC, CPU Feature: AVX512VAES
+func (x Uint8x64) AESDecryptOneRound(y Uint32x16) Uint8x64
 
 /* AESEncryptLastRound */
 
@@ -58,23 +74,39 @@ func (x Uint8x16) AESEncryptLastRound(y Uint32x4) Uint8x16
 // Asm: VAESENCLAST, CPU Feature: AVX512VAES
 func (x Uint8x32) AESEncryptLastRound(y Uint32x8) Uint8x32
 
-/* AESEncryptRound */
+// AESEncryptLastRound performs a series of operations in AES cipher algorithm defined in FIPS 197.
+// x is the state array, starting from low index to high are s00, s10, s20, s30, s01, ..., s33.
+// y is the chunk of w array in use.
+// result = AddRoundKey((ShiftRows(SubBytes(x))), y)
+//
+// Asm: VAESENCLAST, CPU Feature: AVX512VAES
+func (x Uint8x64) AESEncryptLastRound(y Uint32x16) Uint8x64
 
-// AESEncryptRound performs a series of operations in AES cipher algorithm defined in FIPS 197.
+/* AESEncryptOneRound */
+
+// AESEncryptOneRound performs a series of operations in AES cipher algorithm defined in FIPS 197.
 // x is the state array, starting from low index to high are s00, s10, s20, s30, s01, ..., s33.
 // y is the chunk of w array in use.
 // result = AddRoundKey(MixColumns(ShiftRows(SubBytes(x))), y)
 //
 // Asm: VAESENC, CPU Feature: AVX, AES
-func (x Uint8x16) AESEncryptRound(y Uint32x4) Uint8x16
+func (x Uint8x16) AESEncryptOneRound(y Uint32x4) Uint8x16
 
-// AESEncryptRound performs a series of operations in AES cipher algorithm defined in FIPS 197.
+// AESEncryptOneRound performs a series of operations in AES cipher algorithm defined in FIPS 197.
 // x is the state array, starting from low index to high are s00, s10, s20, s30, s01, ..., s33.
 // y is the chunk of w array in use.
 // result = AddRoundKey(MixColumns(ShiftRows(SubBytes(x))), y)
 //
 // Asm: VAESENC, CPU Feature: AVX512VAES
-func (x Uint8x32) AESEncryptRound(y Uint32x8) Uint8x32
+func (x Uint8x32) AESEncryptOneRound(y Uint32x8) Uint8x32
+
+// AESEncryptOneRound performs a series of operations in AES cipher algorithm defined in FIPS 197.
+// x is the state array, starting from low index to high are s00, s10, s20, s30, s01, ..., s33.
+// y is the chunk of w array in use.
+// result = AddRoundKey(MixColumns(ShiftRows(SubBytes(x))), y)
+//
+// Asm: VAESENC, CPU Feature: AVX512VAES
+func (x Uint8x64) AESEncryptOneRound(y Uint32x16) Uint8x64
 
 /* AESInvMixColumns */