Support AES accelerating instructions (#107)
Add support for four Neon AES accelerating instructions.
diff --git a/src/aarch64/assembler-aarch64.cc b/src/aarch64/assembler-aarch64.cc
index f423a23..de2dbc9 100644
--- a/src/aarch64/assembler-aarch64.cc
+++ b/src/aarch64/assembler-aarch64.cc
@@ -6021,6 +6021,38 @@
Emit(0xce608800 | Rd(vd) | Rn(vn) | Rm(vm));
}
+void Assembler::aesd(const VRegister& vd, const VRegister& vn) {
+ VIXL_ASSERT(CPUHas(CPUFeatures::kNEON));
+ VIXL_ASSERT(CPUHas(CPUFeatures::kAES));
+ VIXL_ASSERT(vd.Is16B() && vn.Is16B());
+
+ Emit(0x4e285800 | Rd(vd) | Rn(vn));
+}
+
+void Assembler::aese(const VRegister& vd, const VRegister& vn) {
+ VIXL_ASSERT(CPUHas(CPUFeatures::kNEON));
+ VIXL_ASSERT(CPUHas(CPUFeatures::kAES));
+ VIXL_ASSERT(vd.Is16B() && vn.Is16B());
+
+ Emit(0x4e284800 | Rd(vd) | Rn(vn));
+}
+
+void Assembler::aesimc(const VRegister& vd, const VRegister& vn) {
+ VIXL_ASSERT(CPUHas(CPUFeatures::kNEON));
+ VIXL_ASSERT(CPUHas(CPUFeatures::kAES));
+ VIXL_ASSERT(vd.Is16B() && vn.Is16B());
+
+ Emit(0x4e287800 | Rd(vd) | Rn(vn));
+}
+
+void Assembler::aesmc(const VRegister& vd, const VRegister& vn) {
+ VIXL_ASSERT(CPUHas(CPUFeatures::kNEON));
+ VIXL_ASSERT(CPUHas(CPUFeatures::kAES));
+ VIXL_ASSERT(vd.Is16B() && vn.Is16B());
+
+ Emit(0x4e286800 | Rd(vd) | Rn(vn));
+}
+
// Note:
// For all ToImm instructions below, a difference in case
// for the same letter indicates a negated bit.
diff --git a/src/aarch64/assembler-aarch64.h b/src/aarch64/assembler-aarch64.h
index 9887295..cc4b346 100644
--- a/src/aarch64/assembler-aarch64.h
+++ b/src/aarch64/assembler-aarch64.h
@@ -3684,6 +3684,18 @@
// SHA512 schedule Update 1.
void sha512su1(const VRegister& vd, const VRegister& vn, const VRegister& vm);
+ // AES single round decryption.
+ void aesd(const VRegister& vd, const VRegister& vn);
+
+ // AES single round encryption.
+ void aese(const VRegister& vd, const VRegister& vn);
+
+ // AES inverse mix columns.
+ void aesimc(const VRegister& vd, const VRegister& vn);
+
+ // AES mix columns.
+ void aesmc(const VRegister& vd, const VRegister& vn);
+
// Scalable Vector Extensions.
// Absolute value (predicated).
diff --git a/src/aarch64/cpu-features-auditor-aarch64.cc b/src/aarch64/cpu-features-auditor-aarch64.cc
index 1c2d20b..d815924 100644
--- a/src/aarch64/cpu-features-auditor-aarch64.cc
+++ b/src/aarch64/cpu-features-auditor-aarch64.cc
@@ -275,6 +275,7 @@
void CPUFeaturesAuditor::VisitCryptoAES(const Instruction* instr) {
RecordInstructionFeaturesScope scope(this);
+ scope.Record(CPUFeatures::kNEON, CPUFeatures::kAES);
USE(instr);
}
diff --git a/src/aarch64/disasm-aarch64.cc b/src/aarch64/disasm-aarch64.cc
index 87cc62d..f03b257 100644
--- a/src/aarch64/disasm-aarch64.cc
+++ b/src/aarch64/disasm-aarch64.cc
@@ -2201,7 +2201,7 @@
void Disassembler::VisitCryptoAES(const Instruction *instr) {
- VisitUnimplemented(instr);
+ FormatWithDecodedMnemonic(instr, "'Vd.16b, 'Vn.16b");
}
void Disassembler::DisassembleSHA512(const Instruction *instr) {
diff --git a/src/aarch64/logic-aarch64.cc b/src/aarch64/logic-aarch64.cc
index a1870e1..9a81e49 100644
--- a/src/aarch64/logic-aarch64.cc
+++ b/src/aarch64/logic-aarch64.cc
@@ -8071,6 +8071,286 @@
return srcdst;
}
+static uint8_t GalMul(int table, uint64_t x) {
+ // Galois multiplication lookup tables.
+ static const uint8_t ffmul02[256] = {
+ 0x00, 0x02, 0x04, 0x06, 0x08, 0x0a, 0x0c, 0x0e, 0x10, 0x12, 0x14, 0x16,
+ 0x18, 0x1a, 0x1c, 0x1e, 0x20, 0x22, 0x24, 0x26, 0x28, 0x2a, 0x2c, 0x2e,
+ 0x30, 0x32, 0x34, 0x36, 0x38, 0x3a, 0x3c, 0x3e, 0x40, 0x42, 0x44, 0x46,
+ 0x48, 0x4a, 0x4c, 0x4e, 0x50, 0x52, 0x54, 0x56, 0x58, 0x5a, 0x5c, 0x5e,
+ 0x60, 0x62, 0x64, 0x66, 0x68, 0x6a, 0x6c, 0x6e, 0x70, 0x72, 0x74, 0x76,
+ 0x78, 0x7a, 0x7c, 0x7e, 0x80, 0x82, 0x84, 0x86, 0x88, 0x8a, 0x8c, 0x8e,
+ 0x90, 0x92, 0x94, 0x96, 0x98, 0x9a, 0x9c, 0x9e, 0xa0, 0xa2, 0xa4, 0xa6,
+ 0xa8, 0xaa, 0xac, 0xae, 0xb0, 0xb2, 0xb4, 0xb6, 0xb8, 0xba, 0xbc, 0xbe,
+ 0xc0, 0xc2, 0xc4, 0xc6, 0xc8, 0xca, 0xcc, 0xce, 0xd0, 0xd2, 0xd4, 0xd6,
+ 0xd8, 0xda, 0xdc, 0xde, 0xe0, 0xe2, 0xe4, 0xe6, 0xe8, 0xea, 0xec, 0xee,
+ 0xf0, 0xf2, 0xf4, 0xf6, 0xf8, 0xfa, 0xfc, 0xfe, 0x1b, 0x19, 0x1f, 0x1d,
+ 0x13, 0x11, 0x17, 0x15, 0x0b, 0x09, 0x0f, 0x0d, 0x03, 0x01, 0x07, 0x05,
+ 0x3b, 0x39, 0x3f, 0x3d, 0x33, 0x31, 0x37, 0x35, 0x2b, 0x29, 0x2f, 0x2d,
+ 0x23, 0x21, 0x27, 0x25, 0x5b, 0x59, 0x5f, 0x5d, 0x53, 0x51, 0x57, 0x55,
+ 0x4b, 0x49, 0x4f, 0x4d, 0x43, 0x41, 0x47, 0x45, 0x7b, 0x79, 0x7f, 0x7d,
+ 0x73, 0x71, 0x77, 0x75, 0x6b, 0x69, 0x6f, 0x6d, 0x63, 0x61, 0x67, 0x65,
+ 0x9b, 0x99, 0x9f, 0x9d, 0x93, 0x91, 0x97, 0x95, 0x8b, 0x89, 0x8f, 0x8d,
+ 0x83, 0x81, 0x87, 0x85, 0xbb, 0xb9, 0xbf, 0xbd, 0xb3, 0xb1, 0xb7, 0xb5,
+ 0xab, 0xa9, 0xaf, 0xad, 0xa3, 0xa1, 0xa7, 0xa5, 0xdb, 0xd9, 0xdf, 0xdd,
+ 0xd3, 0xd1, 0xd7, 0xd5, 0xcb, 0xc9, 0xcf, 0xcd, 0xc3, 0xc1, 0xc7, 0xc5,
+ 0xfb, 0xf9, 0xff, 0xfd, 0xf3, 0xf1, 0xf7, 0xf5, 0xeb, 0xe9, 0xef, 0xed,
+ 0xe3, 0xe1, 0xe7, 0xe5,
+ };
+
+ static const uint8_t ffmul03[256] = {
+ 0x00, 0x03, 0x06, 0x05, 0x0c, 0x0f, 0x0a, 0x09, 0x18, 0x1b, 0x1e, 0x1d,
+ 0x14, 0x17, 0x12, 0x11, 0x30, 0x33, 0x36, 0x35, 0x3c, 0x3f, 0x3a, 0x39,
+ 0x28, 0x2b, 0x2e, 0x2d, 0x24, 0x27, 0x22, 0x21, 0x60, 0x63, 0x66, 0x65,
+ 0x6c, 0x6f, 0x6a, 0x69, 0x78, 0x7b, 0x7e, 0x7d, 0x74, 0x77, 0x72, 0x71,
+ 0x50, 0x53, 0x56, 0x55, 0x5c, 0x5f, 0x5a, 0x59, 0x48, 0x4b, 0x4e, 0x4d,
+ 0x44, 0x47, 0x42, 0x41, 0xc0, 0xc3, 0xc6, 0xc5, 0xcc, 0xcf, 0xca, 0xc9,
+ 0xd8, 0xdb, 0xde, 0xdd, 0xd4, 0xd7, 0xd2, 0xd1, 0xf0, 0xf3, 0xf6, 0xf5,
+ 0xfc, 0xff, 0xfa, 0xf9, 0xe8, 0xeb, 0xee, 0xed, 0xe4, 0xe7, 0xe2, 0xe1,
+ 0xa0, 0xa3, 0xa6, 0xa5, 0xac, 0xaf, 0xaa, 0xa9, 0xb8, 0xbb, 0xbe, 0xbd,
+ 0xb4, 0xb7, 0xb2, 0xb1, 0x90, 0x93, 0x96, 0x95, 0x9c, 0x9f, 0x9a, 0x99,
+ 0x88, 0x8b, 0x8e, 0x8d, 0x84, 0x87, 0x82, 0x81, 0x9b, 0x98, 0x9d, 0x9e,
+ 0x97, 0x94, 0x91, 0x92, 0x83, 0x80, 0x85, 0x86, 0x8f, 0x8c, 0x89, 0x8a,
+ 0xab, 0xa8, 0xad, 0xae, 0xa7, 0xa4, 0xa1, 0xa2, 0xb3, 0xb0, 0xb5, 0xb6,
+ 0xbf, 0xbc, 0xb9, 0xba, 0xfb, 0xf8, 0xfd, 0xfe, 0xf7, 0xf4, 0xf1, 0xf2,
+ 0xe3, 0xe0, 0xe5, 0xe6, 0xef, 0xec, 0xe9, 0xea, 0xcb, 0xc8, 0xcd, 0xce,
+ 0xc7, 0xc4, 0xc1, 0xc2, 0xd3, 0xd0, 0xd5, 0xd6, 0xdf, 0xdc, 0xd9, 0xda,
+ 0x5b, 0x58, 0x5d, 0x5e, 0x57, 0x54, 0x51, 0x52, 0x43, 0x40, 0x45, 0x46,
+ 0x4f, 0x4c, 0x49, 0x4a, 0x6b, 0x68, 0x6d, 0x6e, 0x67, 0x64, 0x61, 0x62,
+ 0x73, 0x70, 0x75, 0x76, 0x7f, 0x7c, 0x79, 0x7a, 0x3b, 0x38, 0x3d, 0x3e,
+ 0x37, 0x34, 0x31, 0x32, 0x23, 0x20, 0x25, 0x26, 0x2f, 0x2c, 0x29, 0x2a,
+ 0x0b, 0x08, 0x0d, 0x0e, 0x07, 0x04, 0x01, 0x02, 0x13, 0x10, 0x15, 0x16,
+ 0x1f, 0x1c, 0x19, 0x1a,
+ };
+
+ static const uint8_t ffmul09[256] = {
+ 0x00, 0x09, 0x12, 0x1b, 0x24, 0x2d, 0x36, 0x3f, 0x48, 0x41, 0x5a, 0x53,
+ 0x6c, 0x65, 0x7e, 0x77, 0x90, 0x99, 0x82, 0x8b, 0xb4, 0xbd, 0xa6, 0xaf,
+ 0xd8, 0xd1, 0xca, 0xc3, 0xfc, 0xf5, 0xee, 0xe7, 0x3b, 0x32, 0x29, 0x20,
+ 0x1f, 0x16, 0x0d, 0x04, 0x73, 0x7a, 0x61, 0x68, 0x57, 0x5e, 0x45, 0x4c,
+ 0xab, 0xa2, 0xb9, 0xb0, 0x8f, 0x86, 0x9d, 0x94, 0xe3, 0xea, 0xf1, 0xf8,
+ 0xc7, 0xce, 0xd5, 0xdc, 0x76, 0x7f, 0x64, 0x6d, 0x52, 0x5b, 0x40, 0x49,
+ 0x3e, 0x37, 0x2c, 0x25, 0x1a, 0x13, 0x08, 0x01, 0xe6, 0xef, 0xf4, 0xfd,
+ 0xc2, 0xcb, 0xd0, 0xd9, 0xae, 0xa7, 0xbc, 0xb5, 0x8a, 0x83, 0x98, 0x91,
+ 0x4d, 0x44, 0x5f, 0x56, 0x69, 0x60, 0x7b, 0x72, 0x05, 0x0c, 0x17, 0x1e,
+ 0x21, 0x28, 0x33, 0x3a, 0xdd, 0xd4, 0xcf, 0xc6, 0xf9, 0xf0, 0xeb, 0xe2,
+ 0x95, 0x9c, 0x87, 0x8e, 0xb1, 0xb8, 0xa3, 0xaa, 0xec, 0xe5, 0xfe, 0xf7,
+ 0xc8, 0xc1, 0xda, 0xd3, 0xa4, 0xad, 0xb6, 0xbf, 0x80, 0x89, 0x92, 0x9b,
+ 0x7c, 0x75, 0x6e, 0x67, 0x58, 0x51, 0x4a, 0x43, 0x34, 0x3d, 0x26, 0x2f,
+ 0x10, 0x19, 0x02, 0x0b, 0xd7, 0xde, 0xc5, 0xcc, 0xf3, 0xfa, 0xe1, 0xe8,
+ 0x9f, 0x96, 0x8d, 0x84, 0xbb, 0xb2, 0xa9, 0xa0, 0x47, 0x4e, 0x55, 0x5c,
+ 0x63, 0x6a, 0x71, 0x78, 0x0f, 0x06, 0x1d, 0x14, 0x2b, 0x22, 0x39, 0x30,
+ 0x9a, 0x93, 0x88, 0x81, 0xbe, 0xb7, 0xac, 0xa5, 0xd2, 0xdb, 0xc0, 0xc9,
+ 0xf6, 0xff, 0xe4, 0xed, 0x0a, 0x03, 0x18, 0x11, 0x2e, 0x27, 0x3c, 0x35,
+ 0x42, 0x4b, 0x50, 0x59, 0x66, 0x6f, 0x74, 0x7d, 0xa1, 0xa8, 0xb3, 0xba,
+ 0x85, 0x8c, 0x97, 0x9e, 0xe9, 0xe0, 0xfb, 0xf2, 0xcd, 0xc4, 0xdf, 0xd6,
+ 0x31, 0x38, 0x23, 0x2a, 0x15, 0x1c, 0x07, 0x0e, 0x79, 0x70, 0x6b, 0x62,
+ 0x5d, 0x54, 0x4f, 0x46,
+ };
+
+ static const uint8_t ffmul0b[256] = {
+ 0x00, 0x0b, 0x16, 0x1d, 0x2c, 0x27, 0x3a, 0x31, 0x58, 0x53, 0x4e, 0x45,
+ 0x74, 0x7f, 0x62, 0x69, 0xb0, 0xbb, 0xa6, 0xad, 0x9c, 0x97, 0x8a, 0x81,
+ 0xe8, 0xe3, 0xfe, 0xf5, 0xc4, 0xcf, 0xd2, 0xd9, 0x7b, 0x70, 0x6d, 0x66,
+ 0x57, 0x5c, 0x41, 0x4a, 0x23, 0x28, 0x35, 0x3e, 0x0f, 0x04, 0x19, 0x12,
+ 0xcb, 0xc0, 0xdd, 0xd6, 0xe7, 0xec, 0xf1, 0xfa, 0x93, 0x98, 0x85, 0x8e,
+ 0xbf, 0xb4, 0xa9, 0xa2, 0xf6, 0xfd, 0xe0, 0xeb, 0xda, 0xd1, 0xcc, 0xc7,
+ 0xae, 0xa5, 0xb8, 0xb3, 0x82, 0x89, 0x94, 0x9f, 0x46, 0x4d, 0x50, 0x5b,
+ 0x6a, 0x61, 0x7c, 0x77, 0x1e, 0x15, 0x08, 0x03, 0x32, 0x39, 0x24, 0x2f,
+ 0x8d, 0x86, 0x9b, 0x90, 0xa1, 0xaa, 0xb7, 0xbc, 0xd5, 0xde, 0xc3, 0xc8,
+ 0xf9, 0xf2, 0xef, 0xe4, 0x3d, 0x36, 0x2b, 0x20, 0x11, 0x1a, 0x07, 0x0c,
+ 0x65, 0x6e, 0x73, 0x78, 0x49, 0x42, 0x5f, 0x54, 0xf7, 0xfc, 0xe1, 0xea,
+ 0xdb, 0xd0, 0xcd, 0xc6, 0xaf, 0xa4, 0xb9, 0xb2, 0x83, 0x88, 0x95, 0x9e,
+ 0x47, 0x4c, 0x51, 0x5a, 0x6b, 0x60, 0x7d, 0x76, 0x1f, 0x14, 0x09, 0x02,
+ 0x33, 0x38, 0x25, 0x2e, 0x8c, 0x87, 0x9a, 0x91, 0xa0, 0xab, 0xb6, 0xbd,
+ 0xd4, 0xdf, 0xc2, 0xc9, 0xf8, 0xf3, 0xee, 0xe5, 0x3c, 0x37, 0x2a, 0x21,
+ 0x10, 0x1b, 0x06, 0x0d, 0x64, 0x6f, 0x72, 0x79, 0x48, 0x43, 0x5e, 0x55,
+ 0x01, 0x0a, 0x17, 0x1c, 0x2d, 0x26, 0x3b, 0x30, 0x59, 0x52, 0x4f, 0x44,
+ 0x75, 0x7e, 0x63, 0x68, 0xb1, 0xba, 0xa7, 0xac, 0x9d, 0x96, 0x8b, 0x80,
+ 0xe9, 0xe2, 0xff, 0xf4, 0xc5, 0xce, 0xd3, 0xd8, 0x7a, 0x71, 0x6c, 0x67,
+ 0x56, 0x5d, 0x40, 0x4b, 0x22, 0x29, 0x34, 0x3f, 0x0e, 0x05, 0x18, 0x13,
+ 0xca, 0xc1, 0xdc, 0xd7, 0xe6, 0xed, 0xf0, 0xfb, 0x92, 0x99, 0x84, 0x8f,
+ 0xbe, 0xb5, 0xa8, 0xa3,
+ };
+
+ static const uint8_t ffmul0d[256] = {
+ 0x00, 0x0d, 0x1a, 0x17, 0x34, 0x39, 0x2e, 0x23, 0x68, 0x65, 0x72, 0x7f,
+ 0x5c, 0x51, 0x46, 0x4b, 0xd0, 0xdd, 0xca, 0xc7, 0xe4, 0xe9, 0xfe, 0xf3,
+ 0xb8, 0xb5, 0xa2, 0xaf, 0x8c, 0x81, 0x96, 0x9b, 0xbb, 0xb6, 0xa1, 0xac,
+ 0x8f, 0x82, 0x95, 0x98, 0xd3, 0xde, 0xc9, 0xc4, 0xe7, 0xea, 0xfd, 0xf0,
+ 0x6b, 0x66, 0x71, 0x7c, 0x5f, 0x52, 0x45, 0x48, 0x03, 0x0e, 0x19, 0x14,
+ 0x37, 0x3a, 0x2d, 0x20, 0x6d, 0x60, 0x77, 0x7a, 0x59, 0x54, 0x43, 0x4e,
+ 0x05, 0x08, 0x1f, 0x12, 0x31, 0x3c, 0x2b, 0x26, 0xbd, 0xb0, 0xa7, 0xaa,
+ 0x89, 0x84, 0x93, 0x9e, 0xd5, 0xd8, 0xcf, 0xc2, 0xe1, 0xec, 0xfb, 0xf6,
+ 0xd6, 0xdb, 0xcc, 0xc1, 0xe2, 0xef, 0xf8, 0xf5, 0xbe, 0xb3, 0xa4, 0xa9,
+ 0x8a, 0x87, 0x90, 0x9d, 0x06, 0x0b, 0x1c, 0x11, 0x32, 0x3f, 0x28, 0x25,
+ 0x6e, 0x63, 0x74, 0x79, 0x5a, 0x57, 0x40, 0x4d, 0xda, 0xd7, 0xc0, 0xcd,
+ 0xee, 0xe3, 0xf4, 0xf9, 0xb2, 0xbf, 0xa8, 0xa5, 0x86, 0x8b, 0x9c, 0x91,
+ 0x0a, 0x07, 0x10, 0x1d, 0x3e, 0x33, 0x24, 0x29, 0x62, 0x6f, 0x78, 0x75,
+ 0x56, 0x5b, 0x4c, 0x41, 0x61, 0x6c, 0x7b, 0x76, 0x55, 0x58, 0x4f, 0x42,
+ 0x09, 0x04, 0x13, 0x1e, 0x3d, 0x30, 0x27, 0x2a, 0xb1, 0xbc, 0xab, 0xa6,
+ 0x85, 0x88, 0x9f, 0x92, 0xd9, 0xd4, 0xc3, 0xce, 0xed, 0xe0, 0xf7, 0xfa,
+ 0xb7, 0xba, 0xad, 0xa0, 0x83, 0x8e, 0x99, 0x94, 0xdf, 0xd2, 0xc5, 0xc8,
+ 0xeb, 0xe6, 0xf1, 0xfc, 0x67, 0x6a, 0x7d, 0x70, 0x53, 0x5e, 0x49, 0x44,
+ 0x0f, 0x02, 0x15, 0x18, 0x3b, 0x36, 0x21, 0x2c, 0x0c, 0x01, 0x16, 0x1b,
+ 0x38, 0x35, 0x22, 0x2f, 0x64, 0x69, 0x7e, 0x73, 0x50, 0x5d, 0x4a, 0x47,
+ 0xdc, 0xd1, 0xc6, 0xcb, 0xe8, 0xe5, 0xf2, 0xff, 0xb4, 0xb9, 0xae, 0xa3,
+ 0x80, 0x8d, 0x9a, 0x97,
+ };
+
+ static const uint8_t ffmul0e[256] = {
+ 0x00, 0x0e, 0x1c, 0x12, 0x38, 0x36, 0x24, 0x2a, 0x70, 0x7e, 0x6c, 0x62,
+ 0x48, 0x46, 0x54, 0x5a, 0xe0, 0xee, 0xfc, 0xf2, 0xd8, 0xd6, 0xc4, 0xca,
+ 0x90, 0x9e, 0x8c, 0x82, 0xa8, 0xa6, 0xb4, 0xba, 0xdb, 0xd5, 0xc7, 0xc9,
+ 0xe3, 0xed, 0xff, 0xf1, 0xab, 0xa5, 0xb7, 0xb9, 0x93, 0x9d, 0x8f, 0x81,
+ 0x3b, 0x35, 0x27, 0x29, 0x03, 0x0d, 0x1f, 0x11, 0x4b, 0x45, 0x57, 0x59,
+ 0x73, 0x7d, 0x6f, 0x61, 0xad, 0xa3, 0xb1, 0xbf, 0x95, 0x9b, 0x89, 0x87,
+ 0xdd, 0xd3, 0xc1, 0xcf, 0xe5, 0xeb, 0xf9, 0xf7, 0x4d, 0x43, 0x51, 0x5f,
+ 0x75, 0x7b, 0x69, 0x67, 0x3d, 0x33, 0x21, 0x2f, 0x05, 0x0b, 0x19, 0x17,
+ 0x76, 0x78, 0x6a, 0x64, 0x4e, 0x40, 0x52, 0x5c, 0x06, 0x08, 0x1a, 0x14,
+ 0x3e, 0x30, 0x22, 0x2c, 0x96, 0x98, 0x8a, 0x84, 0xae, 0xa0, 0xb2, 0xbc,
+ 0xe6, 0xe8, 0xfa, 0xf4, 0xde, 0xd0, 0xc2, 0xcc, 0x41, 0x4f, 0x5d, 0x53,
+ 0x79, 0x77, 0x65, 0x6b, 0x31, 0x3f, 0x2d, 0x23, 0x09, 0x07, 0x15, 0x1b,
+ 0xa1, 0xaf, 0xbd, 0xb3, 0x99, 0x97, 0x85, 0x8b, 0xd1, 0xdf, 0xcd, 0xc3,
+ 0xe9, 0xe7, 0xf5, 0xfb, 0x9a, 0x94, 0x86, 0x88, 0xa2, 0xac, 0xbe, 0xb0,
+ 0xea, 0xe4, 0xf6, 0xf8, 0xd2, 0xdc, 0xce, 0xc0, 0x7a, 0x74, 0x66, 0x68,
+ 0x42, 0x4c, 0x5e, 0x50, 0x0a, 0x04, 0x16, 0x18, 0x32, 0x3c, 0x2e, 0x20,
+ 0xec, 0xe2, 0xf0, 0xfe, 0xd4, 0xda, 0xc8, 0xc6, 0x9c, 0x92, 0x80, 0x8e,
+ 0xa4, 0xaa, 0xb8, 0xb6, 0x0c, 0x02, 0x10, 0x1e, 0x34, 0x3a, 0x28, 0x26,
+ 0x7c, 0x72, 0x60, 0x6e, 0x44, 0x4a, 0x58, 0x56, 0x37, 0x39, 0x2b, 0x25,
+ 0x0f, 0x01, 0x13, 0x1d, 0x47, 0x49, 0x5b, 0x55, 0x7f, 0x71, 0x63, 0x6d,
+ 0xd7, 0xd9, 0xcb, 0xc5, 0xef, 0xe1, 0xf3, 0xfd, 0xa7, 0xa9, 0xbb, 0xb5,
+ 0x9f, 0x91, 0x83, 0x8d,
+ };
+
+ x &= 255;
+ switch (table) {
+ case 0x2:
+ return ffmul02[x];
+ case 0x3:
+ return ffmul03[x];
+ case 0x9:
+ return ffmul09[x];
+ case 0xb:
+ return ffmul0b[x];
+ case 0xd:
+ return ffmul0d[x];
+ case 0xe:
+ return ffmul0e[x];
+ case 0:
+ // Case 0 indicates no table lookup, used for some forward mix stages.
+ return x;
+ default:
+ VIXL_UNREACHABLE();
+ return x;
+ }
+}
+
+
+static uint8_t AESMixInner(uint64_t* x, int stage, bool inverse) {
+ VIXL_ASSERT(IsUint2(stage));
+
+ int imc_gm[7] = {0xb, 0xd, 0x9, 0xe};
+ int mc_gm[7] = {0x3, 0x0, 0x0, 0x2};
+
+ int* gm = inverse ? imc_gm : mc_gm;
+ int index = 3 - stage;
+
+ uint8_t result = 0;
+ for (int i = 0; i < 4; i++) {
+ result ^= GalMul(gm[(index + i) % 4], x[i]);
+ }
+ return result;
+}
+
+
+LogicVRegister Simulator::aesmix(LogicVRegister dst,
+ const LogicVRegister& src,
+ bool inverse) {
+ uint64_t in[16] = {};
+ src.UintArray(kFormat16B, in);
+ dst.ClearForWrite(kFormat16B);
+
+ for (int c = 0; c < 16; c++) {
+ int cmod4 = c % 4;
+ int d = c - cmod4;
+ VIXL_ASSERT((d == 0) || (d == 4) || (d == 8) || (d == 12));
+ dst.SetUint(kFormat16B, c, AESMixInner(&in[d], cmod4, inverse));
+ }
+
+ return dst;
+}
+
+LogicVRegister Simulator::aes(LogicVRegister dst,
+ const LogicVRegister& src,
+ bool decrypt) {
+ dst.ClearForWrite(kFormat16B);
+
+ // (Inverse) shift rows.
+ uint8_t shift[] = {0, 5, 10, 15, 4, 9, 14, 3, 8, 13, 2, 7, 12, 1, 6, 11};
+ uint8_t shift_inv[] = {0, 13, 10, 7, 4, 1, 14, 11, 8, 5, 2, 15, 12, 9, 6, 3};
+ for (int i = 0; i < LaneCountFromFormat(kFormat16B); i++) {
+ uint8_t index = decrypt ? shift_inv[i] : shift[i];
+ dst.SetUint(kFormat16B, i, src.Uint(kFormat16B, index));
+ }
+
+ // (Inverse) substitute bytes.
+ static const uint8_t gf2[256] = {
+ 0x63, 0x7c, 0x77, 0x7b, 0xf2, 0x6b, 0x6f, 0xc5, 0x30, 0x01, 0x67, 0x2b,
+ 0xfe, 0xd7, 0xab, 0x76, 0xca, 0x82, 0xc9, 0x7d, 0xfa, 0x59, 0x47, 0xf0,
+ 0xad, 0xd4, 0xa2, 0xaf, 0x9c, 0xa4, 0x72, 0xc0, 0xb7, 0xfd, 0x93, 0x26,
+ 0x36, 0x3f, 0xf7, 0xcc, 0x34, 0xa5, 0xe5, 0xf1, 0x71, 0xd8, 0x31, 0x15,
+ 0x04, 0xc7, 0x23, 0xc3, 0x18, 0x96, 0x05, 0x9a, 0x07, 0x12, 0x80, 0xe2,
+ 0xeb, 0x27, 0xb2, 0x75, 0x09, 0x83, 0x2c, 0x1a, 0x1b, 0x6e, 0x5a, 0xa0,
+ 0x52, 0x3b, 0xd6, 0xb3, 0x29, 0xe3, 0x2f, 0x84, 0x53, 0xd1, 0x00, 0xed,
+ 0x20, 0xfc, 0xb1, 0x5b, 0x6a, 0xcb, 0xbe, 0x39, 0x4a, 0x4c, 0x58, 0xcf,
+ 0xd0, 0xef, 0xaa, 0xfb, 0x43, 0x4d, 0x33, 0x85, 0x45, 0xf9, 0x02, 0x7f,
+ 0x50, 0x3c, 0x9f, 0xa8, 0x51, 0xa3, 0x40, 0x8f, 0x92, 0x9d, 0x38, 0xf5,
+ 0xbc, 0xb6, 0xda, 0x21, 0x10, 0xff, 0xf3, 0xd2, 0xcd, 0x0c, 0x13, 0xec,
+ 0x5f, 0x97, 0x44, 0x17, 0xc4, 0xa7, 0x7e, 0x3d, 0x64, 0x5d, 0x19, 0x73,
+ 0x60, 0x81, 0x4f, 0xdc, 0x22, 0x2a, 0x90, 0x88, 0x46, 0xee, 0xb8, 0x14,
+ 0xde, 0x5e, 0x0b, 0xdb, 0xe0, 0x32, 0x3a, 0x0a, 0x49, 0x06, 0x24, 0x5c,
+ 0xc2, 0xd3, 0xac, 0x62, 0x91, 0x95, 0xe4, 0x79, 0xe7, 0xc8, 0x37, 0x6d,
+ 0x8d, 0xd5, 0x4e, 0xa9, 0x6c, 0x56, 0xf4, 0xea, 0x65, 0x7a, 0xae, 0x08,
+ 0xba, 0x78, 0x25, 0x2e, 0x1c, 0xa6, 0xb4, 0xc6, 0xe8, 0xdd, 0x74, 0x1f,
+ 0x4b, 0xbd, 0x8b, 0x8a, 0x70, 0x3e, 0xb5, 0x66, 0x48, 0x03, 0xf6, 0x0e,
+ 0x61, 0x35, 0x57, 0xb9, 0x86, 0xc1, 0x1d, 0x9e, 0xe1, 0xf8, 0x98, 0x11,
+ 0x69, 0xd9, 0x8e, 0x94, 0x9b, 0x1e, 0x87, 0xe9, 0xce, 0x55, 0x28, 0xdf,
+ 0x8c, 0xa1, 0x89, 0x0d, 0xbf, 0xe6, 0x42, 0x68, 0x41, 0x99, 0x2d, 0x0f,
+ 0xb0, 0x54, 0xbb, 0x16,
+ };
+ static const uint8_t gf2_inv[256] = {
+ 0x52, 0x09, 0x6a, 0xd5, 0x30, 0x36, 0xa5, 0x38, 0xbf, 0x40, 0xa3, 0x9e,
+ 0x81, 0xf3, 0xd7, 0xfb, 0x7c, 0xe3, 0x39, 0x82, 0x9b, 0x2f, 0xff, 0x87,
+ 0x34, 0x8e, 0x43, 0x44, 0xc4, 0xde, 0xe9, 0xcb, 0x54, 0x7b, 0x94, 0x32,
+ 0xa6, 0xc2, 0x23, 0x3d, 0xee, 0x4c, 0x95, 0x0b, 0x42, 0xfa, 0xc3, 0x4e,
+ 0x08, 0x2e, 0xa1, 0x66, 0x28, 0xd9, 0x24, 0xb2, 0x76, 0x5b, 0xa2, 0x49,
+ 0x6d, 0x8b, 0xd1, 0x25, 0x72, 0xf8, 0xf6, 0x64, 0x86, 0x68, 0x98, 0x16,
+ 0xd4, 0xa4, 0x5c, 0xcc, 0x5d, 0x65, 0xb6, 0x92, 0x6c, 0x70, 0x48, 0x50,
+ 0xfd, 0xed, 0xb9, 0xda, 0x5e, 0x15, 0x46, 0x57, 0xa7, 0x8d, 0x9d, 0x84,
+ 0x90, 0xd8, 0xab, 0x00, 0x8c, 0xbc, 0xd3, 0x0a, 0xf7, 0xe4, 0x58, 0x05,
+ 0xb8, 0xb3, 0x45, 0x06, 0xd0, 0x2c, 0x1e, 0x8f, 0xca, 0x3f, 0x0f, 0x02,
+ 0xc1, 0xaf, 0xbd, 0x03, 0x01, 0x13, 0x8a, 0x6b, 0x3a, 0x91, 0x11, 0x41,
+ 0x4f, 0x67, 0xdc, 0xea, 0x97, 0xf2, 0xcf, 0xce, 0xf0, 0xb4, 0xe6, 0x73,
+ 0x96, 0xac, 0x74, 0x22, 0xe7, 0xad, 0x35, 0x85, 0xe2, 0xf9, 0x37, 0xe8,
+ 0x1c, 0x75, 0xdf, 0x6e, 0x47, 0xf1, 0x1a, 0x71, 0x1d, 0x29, 0xc5, 0x89,
+ 0x6f, 0xb7, 0x62, 0x0e, 0xaa, 0x18, 0xbe, 0x1b, 0xfc, 0x56, 0x3e, 0x4b,
+ 0xc6, 0xd2, 0x79, 0x20, 0x9a, 0xdb, 0xc0, 0xfe, 0x78, 0xcd, 0x5a, 0xf4,
+ 0x1f, 0xdd, 0xa8, 0x33, 0x88, 0x07, 0xc7, 0x31, 0xb1, 0x12, 0x10, 0x59,
+ 0x27, 0x80, 0xec, 0x5f, 0x60, 0x51, 0x7f, 0xa9, 0x19, 0xb5, 0x4a, 0x0d,
+ 0x2d, 0xe5, 0x7a, 0x9f, 0x93, 0xc9, 0x9c, 0xef, 0xa0, 0xe0, 0x3b, 0x4d,
+ 0xae, 0x2a, 0xf5, 0xb0, 0xc8, 0xeb, 0xbb, 0x3c, 0x83, 0x53, 0x99, 0x61,
+ 0x17, 0x2b, 0x04, 0x7e, 0xba, 0x77, 0xd6, 0x26, 0xe1, 0x69, 0x14, 0x63,
+ 0x55, 0x21, 0x0c, 0x7d,
+ };
+
+ for (int i = 0; i < LaneCountFromFormat(kFormat16B); i++) {
+ const uint8_t* table = decrypt ? gf2_inv : gf2;
+ dst.SetUint(kFormat16B, i, table[dst.Uint(kFormat16B, i)]);
+ }
+ return dst;
+}
+
} // namespace aarch64
} // namespace vixl
diff --git a/src/aarch64/macro-assembler-aarch64.h b/src/aarch64/macro-assembler-aarch64.h
index eead245..a989ddd 100644
--- a/src/aarch64/macro-assembler-aarch64.h
+++ b/src/aarch64/macro-assembler-aarch64.h
@@ -2906,6 +2906,10 @@
V(abs, Abs) \
V(addp, Addp) \
V(addv, Addv) \
+ V(aesd, Aesd) \
+ V(aese, Aese) \
+ V(aesimc, Aesimc) \
+ V(aesmc, Aesmc) \
V(cls, Cls) \
V(clz, Clz) \
V(cnt, Cnt) \
diff --git a/src/aarch64/simulator-aarch64.cc b/src/aarch64/simulator-aarch64.cc
index 134362c..e63715c 100644
--- a/src/aarch64/simulator-aarch64.cc
+++ b/src/aarch64/simulator-aarch64.cc
@@ -7239,7 +7239,26 @@
void Simulator::VisitCryptoAES(const Instruction* instr) {
- VisitUnimplemented(instr);
+ SimVRegister& rd = ReadVRegister(instr->GetRd());
+ SimVRegister& rn = ReadVRegister(instr->GetRn());
+ SimVRegister temp;
+
+ switch (form_hash_) {
+ case "aesd_b_cryptoaes"_h:
+ eor(kFormat16B, temp, rd, rn);
+ aes(rd, temp, /* decrypt = */ true);
+ break;
+ case "aese_b_cryptoaes"_h:
+ eor(kFormat16B, temp, rd, rn);
+ aes(rd, temp, /* decrypt = */ false);
+ break;
+ case "aesimc_b_cryptoaes"_h:
+ aesmix(rd, rn, /* inverse = */ true);
+ break;
+ case "aesmc_b_cryptoaes"_h:
+ aesmix(rd, rn, /* inverse = */ false);
+ break;
+ }
}
void Simulator::SimulateSHA512(const Instruction* instr) {
diff --git a/src/aarch64/simulator-aarch64.h b/src/aarch64/simulator-aarch64.h
index 19032a6..c20ec3c 100644
--- a/src/aarch64/simulator-aarch64.h
+++ b/src/aarch64/simulator-aarch64.h
@@ -4553,6 +4553,14 @@
const LogicVRegister& src1,
const LogicVRegister& src2);
+
+ LogicVRegister aes(LogicVRegister srcdst,
+ const LogicVRegister& src1,
+ bool decrypt);
+ LogicVRegister aesmix(LogicVRegister srcdst,
+ const LogicVRegister& src1,
+ bool inverse);
+
#define NEON_3VREG_LOGIC_LIST(V) \
V(addhn) \
V(addhn2) \
diff --git a/test/aarch64/test-cpu-features-aarch64.cc b/test/aarch64/test-cpu-features-aarch64.cc
index a6b68fd..a14c524 100644
--- a/test/aarch64/test-cpu-features-aarch64.cc
+++ b/test/aarch64/test-cpu-features-aarch64.cc
@@ -3825,5 +3825,15 @@
TEST_FEAT(sha512su1_0, sha512su1(v19.V2D(), v9.V2D(), v27.V2D()))
#undef TEST_FEAT
+#define TEST_FEAT(NAME, ASM) \
+ TEST_TEMPLATE(CPUFeatures(CPUFeatures::kNEON, CPUFeatures::kAES), \
+ NEON_AES_##NAME, \
+ ASM)
+TEST_FEAT(aesd_0, aesd(v0.V16B(), v29.V16B()))
+TEST_FEAT(aese_0, aese(v0.V16B(), v29.V16B()))
+TEST_FEAT(aesimc_0, aesimc(v0.V16B(), v29.V16B()))
+TEST_FEAT(aesmc_0, aesmc(v0.V16B(), v29.V16B()))
+#undef TEST_FEAT
+
} // namespace aarch64
} // namespace vixl
diff --git a/test/aarch64/test-disasm-neon-aarch64.cc b/test/aarch64/test-disasm-neon-aarch64.cc
index 5bf58b3..5bdded9 100644
--- a/test/aarch64/test-disasm-neon-aarch64.cc
+++ b/test/aarch64/test-disasm-neon-aarch64.cc
@@ -4568,6 +4568,17 @@
CLEANUP();
}
+TEST(neon_aes) {
+ SETUP();
+
+ COMPARE_MACRO(Aesd(v0.V16B(), v29.V16B()), "aesd v0.16b, v29.16b");
+ COMPARE_MACRO(Aese(v0.V16B(), v29.V16B()), "aese v0.16b, v29.16b");
+ COMPARE_MACRO(Aesimc(v0.V16B(), v29.V16B()), "aesimc v0.16b, v29.16b");
+ COMPARE_MACRO(Aesmc(v0.V16B(), v29.V16B()), "aesmc v0.16b, v29.16b");
+
+ CLEANUP();
+}
+
TEST(neon_unallocated_regression_test) {
SETUP();
diff --git a/test/aarch64/test-simulator-sve-aarch64.cc b/test/aarch64/test-simulator-sve-aarch64.cc
index 2961b6a..e62fea2 100644
--- a/test/aarch64/test-simulator-sve-aarch64.cc
+++ b/test/aarch64/test-simulator-sve-aarch64.cc
@@ -1326,5 +1326,213 @@
}
}
+TEST_SVE(neon_aes) {
+ SVE_SETUP_WITH_FEATURES(CPUFeatures::kSVE,
+ CPUFeatures::kNEON,
+ CPUFeatures::kCRC32,
+ CPUFeatures::kAES);
+ START();
+
+ SetInitialMachineState(&masm);
+ // state = 0xe2bd2480
+
+ {
+ ExactAssemblyScope scope(&masm, 30 * kInstructionSize);
+ __ dci(0x4e285a86); // aesd v6.16b, v20.16b
+ // vl128 state = 0x801bfc08
+ __ dci(0x4e2858ae); // aesd v14.16b, v5.16b
+ // vl128 state = 0xbd83a757
+ __ dci(0x4e2858ac); // aesd v12.16b, v5.16b
+ // vl128 state = 0x9fb1dc6b
+ __ dci(0x4e2858ae); // aesd v14.16b, v5.16b
+ // vl128 state = 0xfa1fa7e4
+ __ dci(0x4e28482a); // aese v10.16b, v1.16b
+ // vl128 state = 0xecfcfe2d
+ __ dci(0x4e28483a); // aese v26.16b, v1.16b
+ // vl128 state = 0x05e22f07
+ __ dci(0x4e28488a); // aese v10.16b, v4.16b
+ // vl128 state = 0xdd53df5f
+ __ dci(0x4e28488e); // aese v14.16b, v4.16b
+ // vl128 state = 0x9d2ac50f
+ __ dci(0x4e28484f); // aese v15.16b, v2.16b
+ // vl128 state = 0xf45146ab
+ __ dci(0x4e28484b); // aese v11.16b, v2.16b
+ // vl128 state = 0xf1260a7c
+ __ dci(0x4e28485b); // aese v27.16b, v2.16b
+ // vl128 state = 0x3a0844da
+ __ dci(0x4e285819); // aesd v25.16b, v0.16b
+ // vl128 state = 0xaca89993
+ __ dci(0x4e284a09); // aese v9.16b, v16.16b
+ // vl128 state = 0xef4e9a5f
+ __ dci(0x4e285a4b); // aesd v11.16b, v18.16b
+ // vl128 state = 0x209a44bc
+ __ dci(0x4e285a4f); // aesd v15.16b, v18.16b
+ // vl128 state = 0xc6d2d718
+ __ dci(0x4e285a4d); // aesd v13.16b, v18.16b
+ // vl128 state = 0x1aceef8f
+ __ dci(0x4e285a45); // aesd v5.16b, v18.16b
+ // vl128 state = 0x7ed056c6
+ __ dci(0x4e285af5); // aesd v21.16b, v23.16b
+ // vl128 state = 0x429ed71e
+ __ dci(0x4e285a91); // aesd v17.16b, v20.16b
+ // vl128 state = 0xd7a1f687
+ __ dci(0x4e284ad9); // aese v25.16b, v22.16b
+ // vl128 state = 0x8fa44574
+ __ dci(0x4e284adb); // aese v27.16b, v22.16b
+ // vl128 state = 0xd2792169
+ __ dci(0x4e285afa); // aesd v26.16b, v23.16b
+ // vl128 state = 0xe502f095
+ __ dci(0x4e285bbb); // aesd v27.16b, v29.16b
+ // vl128 state = 0x0e3d3238
+ __ dci(0x4e285bbf); // aesd v31.16b, v29.16b
+ // vl128 state = 0x0ad06592
+ __ dci(0x4e285baf); // aesd v15.16b, v29.16b
+ // vl128 state = 0xb94f3c19
+ __ dci(0x4e284b3f); // aese v31.16b, v25.16b
+ // vl128 state = 0xf31a0da1
+ __ dci(0x4e284917); // aese v23.16b, v8.16b
+ // vl128 state = 0x7d2d7811
+ __ dci(0x4e284913); // aese v19.16b, v8.16b
+ // vl128 state = 0x41b7b854
+ __ dci(0x4e284911); // aese v17.16b, v8.16b
+ // vl128 state = 0x60600536
+ __ dci(0x4e2849d5); // aese v21.16b, v14.16b
+ // vl128 state = 0x3e0cc74f
+ }
+
+ uint32_t state;
+ ComputeMachineStateHash(&masm, &state);
+ __ Mov(x0, reinterpret_cast<uint64_t>(&state));
+ __ Ldr(w0, MemOperand(x0));
+
+ END();
+ if (CAN_RUN()) {
+ RUN();
+ uint32_t expected_hashes[] = {
+ 0x3e0cc74f,
+ 0x7f17ba2e,
+ 0xd59f8e91,
+ 0x9f15a51b,
+ 0x11d92e66,
+ 0xcd53d015,
+ 0xbc652785,
+ 0x6974fa54,
+ 0x953d342e,
+ 0xf1aa56b3,
+ 0xde8ca1d3,
+ 0xba408b82,
+ 0x48094fa4,
+ 0xb757bcf1,
+ 0x2cc5be58,
+ 0x6e7a0f58,
+ };
+ ASSERT_EQUAL_64(expected_hashes[core.GetSVELaneCount(kQRegSize) - 1], x0);
+ }
+}
+
+TEST_SVE(neon_aesmc) {
+ SVE_SETUP_WITH_FEATURES(CPUFeatures::kSVE,
+ CPUFeatures::kNEON,
+ CPUFeatures::kCRC32,
+ CPUFeatures::kAES);
+ START();
+
+ SetInitialMachineState(&masm);
+ // state = 0xe2bd2480
+
+ {
+ ExactAssemblyScope scope(&masm, 30 * kInstructionSize);
+ __ dci(0x4e287800); // aesimc v0.16b, v0.16b
+ // vl128 state = 0x03554749
+ __ dci(0x4e287a28); // aesimc v8.16b, v17.16b
+ // vl128 state = 0x59d5fedd
+ __ dci(0x4e287a2a); // aesimc v10.16b, v17.16b
+ // vl128 state = 0xcda29514
+ __ dci(0x4e286aae); // aesmc v14.16b, v21.16b
+ // vl128 state = 0xae8f019a
+ __ dci(0x4e286abe); // aesmc v30.16b, v21.16b
+ // vl128 state = 0x7b04c6c0
+ __ dci(0x4e286a0e); // aesmc v14.16b, v16.16b
+ // vl128 state = 0xaf6c5ce6
+ __ dci(0x4e286a0a); // aesmc v10.16b, v16.16b
+ // vl128 state = 0xf1d7fd2b
+ __ dci(0x4e286acb); // aesmc v11.16b, v22.16b
+ // vl128 state = 0x5d693c63
+ __ dci(0x4e286acf); // aesmc v15.16b, v22.16b
+ // vl128 state = 0xec8971ad
+ __ dci(0x4e286adf); // aesmc v31.16b, v22.16b
+ // vl128 state = 0x6389b200
+ __ dci(0x4e287a9d); // aesimc v29.16b, v20.16b
+ // vl128 state = 0xd69341fb
+ __ dci(0x4e28688d); // aesmc v13.16b, v4.16b
+ // vl128 state = 0x6344af95
+ __ dci(0x4e2878cf); // aesimc v15.16b, v6.16b
+ // vl128 state = 0x5c58dfac
+ __ dci(0x4e2878cb); // aesimc v11.16b, v6.16b
+ // vl128 state = 0x7dc9cf34
+ __ dci(0x4e2878c9); // aesimc v9.16b, v6.16b
+ // vl128 state = 0xff4b3544
+ __ dci(0x4e2878c1); // aesimc v1.16b, v6.16b
+ // vl128 state = 0xd1937de2
+ __ dci(0x4e287871); // aesimc v17.16b, v3.16b
+ // vl128 state = 0x7cabd208
+ __ dci(0x4e287815); // aesimc v21.16b, v0.16b
+ // vl128 state = 0xbc06df94
+ __ dci(0x4e28685d); // aesmc v29.16b, v2.16b
+ // vl128 state = 0xfc4478bb
+ __ dci(0x4e28685f); // aesmc v31.16b, v2.16b
+ // vl128 state = 0x0c72c200
+ __ dci(0x4e28787e); // aesimc v30.16b, v3.16b
+ // vl128 state = 0xdd822b9d
+ __ dci(0x4e28793f); // aesimc v31.16b, v9.16b
+ // vl128 state = 0x1397dcc6
+ __ dci(0x4e28793b); // aesimc v27.16b, v9.16b
+ // vl128 state = 0x43f3abd6
+ __ dci(0x4e28792b); // aesimc v11.16b, v9.16b
+ // vl128 state = 0xeb8ca365
+ __ dci(0x4e2869bb); // aesmc v27.16b, v13.16b
+ // vl128 state = 0x0a957f4f
+ __ dci(0x4e286b93); // aesmc v19.16b, v28.16b
+ // vl128 state = 0xbc5da8bd
+ __ dci(0x4e286b97); // aesmc v23.16b, v28.16b
+ // vl128 state = 0xc49343cc
+ __ dci(0x4e286b95); // aesmc v21.16b, v28.16b
+ // vl128 state = 0x8c80c144
+ __ dci(0x4e286b51); // aesmc v17.16b, v26.16b
+ // vl128 state = 0xeda3255d
+ __ dci(0x4e2869d3); // aesmc v19.16b, v14.16b
+ // vl128 state = 0x8db8a9d0
+ }
+
+ uint32_t state;
+ ComputeMachineStateHash(&masm, &state);
+ __ Mov(x0, reinterpret_cast<uint64_t>(&state));
+ __ Ldr(w0, MemOperand(x0));
+
+ END();
+ if (CAN_RUN()) {
+ RUN();
+ uint32_t expected_hashes[] = {
+ 0x8db8a9d0,
+ 0xb13d8e1e,
+ 0x9f33ca70,
+ 0x38f7ef7a,
+ 0x65352b29,
+ 0xc4257260,
+ 0xf49587c2,
+ 0xb3f61256,
+ 0x8ef4a534,
+ 0x6e061aa9,
+ 0x7270527d,
+ 0x3e1f82f9,
+ 0x1fe79e60,
+ 0x985cab68,
+ 0xe77b4484,
+ 0xe3817f4e,
+ };
+ ASSERT_EQUAL_64(expected_hashes[core.GetSVELaneCount(kQRegSize) - 1], x0);
+ }
+}
+
} // namespace aarch64
} // namespace vixl