README.md (46361B)
1 # cpuid 2 Package cpuid provides information about the CPU running the current program. 3 4 CPU features are detected on startup, and kept for fast access through the life of the application. 5 Currently x86 / x64 (AMD64/i386) and ARM (ARM64) is supported, and no external C (cgo) code is used, which should make the library very easy to use. 6 7 You can access the CPU information by accessing the shared CPU variable of the cpuid library. 8 9 Package home: https://github.com/klauspost/cpuid 10 11 [](https://pkg.go.dev/github.com/klauspost/cpuid/v2) 12 [![Build Status][3]][4] 13 14 [3]: https://travis-ci.org/klauspost/cpuid.svg?branch=master 15 [4]: https://travis-ci.org/klauspost/cpuid 16 17 ## installing 18 19 `go get -u github.com/klauspost/cpuid/v2` using modules. 20 Drop `v2` for others. 21 22 Installing binary: 23 24 `go install github.com/klauspost/cpuid/v2/cmd/cpuid@latest` 25 26 Or download binaries from release page: https://github.com/klauspost/cpuid/releases 27 28 ### Homebrew 29 30 For macOS/Linux users, you can install via [brew](https://brew.sh/) 31 32 ```sh 33 $ brew install cpuid 34 ``` 35 36 ## example 37 38 ```Go 39 package main 40 41 import ( 42 "fmt" 43 "strings" 44 45 . "github.com/klauspost/cpuid/v2" 46 ) 47 48 func main() { 49 // Print basic CPU information: 50 fmt.Println("Name:", CPU.BrandName) 51 fmt.Println("PhysicalCores:", CPU.PhysicalCores) 52 fmt.Println("ThreadsPerCore:", CPU.ThreadsPerCore) 53 fmt.Println("LogicalCores:", CPU.LogicalCores) 54 fmt.Println("Family", CPU.Family, "Model:", CPU.Model, "Vendor ID:", CPU.VendorID) 55 fmt.Println("Features:", strings.Join(CPU.FeatureSet(), ",")) 56 fmt.Println("Cacheline bytes:", CPU.CacheLine) 57 fmt.Println("L1 Data Cache:", CPU.Cache.L1D, "bytes") 58 fmt.Println("L1 Instruction Cache:", CPU.Cache.L1I, "bytes") 59 fmt.Println("L2 Cache:", CPU.Cache.L2, "bytes") 60 fmt.Println("L3 Cache:", CPU.Cache.L3, "bytes") 61 fmt.Println("Frequency", CPU.Hz, "hz") 62 63 // Test if we have these specific features: 64 if CPU.Supports(SSE, SSE2) { 65 fmt.Println("We have Streaming SIMD 2 Extensions") 66 } 67 } 68 ``` 69 70 Sample output: 71 ``` 72 >go run main.go 73 Name: AMD Ryzen 9 3950X 16-Core Processor 74 PhysicalCores: 16 75 ThreadsPerCore: 2 76 LogicalCores: 32 77 Family 23 Model: 113 Vendor ID: AMD 78 Features: ADX,AESNI,AVX,AVX2,BMI1,BMI2,CLMUL,CMOV,CX16,F16C,FMA3,HTT,HYPERVISOR,LZCNT,MMX,MMXEXT,NX,POPCNT,RDRAND,RDSEED,RDTSCP,SHA,SSE,SSE2,SSE3,SSE4,SSE42,SSE4A,SSSE3 79 Cacheline bytes: 64 80 L1 Data Cache: 32768 bytes 81 L1 Instruction Cache: 32768 bytes 82 L2 Cache: 524288 bytes 83 L3 Cache: 16777216 bytes 84 Frequency 0 hz 85 We have Streaming SIMD 2 Extensions 86 ``` 87 88 # usage 89 90 The `cpuid.CPU` provides access to CPU features. Use `cpuid.CPU.Supports()` to check for CPU features. 91 A faster `cpuid.CPU.Has()` is provided which will usually be inlined by the gc compiler. 92 93 To test a larger number of features, they can be combined using `f := CombineFeatures(CMOV, CMPXCHG8, X87, FXSR, MMX, SYSCALL, SSE, SSE2)`, etc. 94 This can be using with `cpuid.CPU.HasAll(f)` to quickly test if all features are supported. 95 96 Note that for some cpu/os combinations some features will not be detected. 97 `amd64` has rather good support and should work reliably on all platforms. 98 99 Note that hypervisors may not pass through all CPU features through to the guest OS, 100 so even if your host supports a feature it may not be visible on guests. 101 102 ## arm64 feature detection 103 104 Not all operating systems provide ARM features directly 105 and there is no safe way to do so for the rest. 106 107 Currently `arm64/linux` and `arm64/freebsd` should be quite reliable. 108 `arm64/darwin` adds features expected from the M1 processor, but a lot remains undetected. 109 110 A `DetectARM()` can be used if you are able to control your deployment, 111 it will detect CPU features, but may crash if the OS doesn't intercept the calls. 112 A `-cpu.arm` flag for detecting unsafe ARM features can be added. See below. 113 114 Note that currently only features are detected on ARM, 115 no additional information is currently available. 116 117 ## flags 118 119 It is possible to add flags that affects cpu detection. 120 121 For this the `Flags()` command is provided. 122 123 This must be called *before* `flag.Parse()` AND after the flags have been parsed `Detect()` must be called. 124 125 This means that any detection used in `init()` functions will not contain these flags. 126 127 Example: 128 129 ```Go 130 package main 131 132 import ( 133 "flag" 134 "fmt" 135 "strings" 136 137 "github.com/klauspost/cpuid/v2" 138 ) 139 140 func main() { 141 cpuid.Flags() 142 flag.Parse() 143 cpuid.Detect() 144 145 // Test if we have these specific features: 146 if cpuid.CPU.Supports(cpuid.SSE, cpuid.SSE2) { 147 fmt.Println("We have Streaming SIMD 2 Extensions") 148 } 149 } 150 ``` 151 152 ## commandline 153 154 Download as binary from: https://github.com/klauspost/cpuid/releases 155 156 Install from source: 157 158 `go install github.com/klauspost/cpuid/v2/cmd/cpuid@latest` 159 160 ### Example 161 162 ``` 163 λ cpuid 164 Name: AMD Ryzen 9 3950X 16-Core Processor 165 Vendor String: AuthenticAMD 166 Vendor ID: AMD 167 PhysicalCores: 16 168 Threads Per Core: 2 169 Logical Cores: 32 170 CPU Family 23 Model: 113 171 Features: ADX,AESNI,AVX,AVX2,BMI1,BMI2,CLMUL,CLZERO,CMOV,CMPXCHG8,CPBOOST,CX16,F16C,FMA3,FXSR,FXSROPT,HTT,HYPERVISOR,LAHF,LZCNT,MCAOVERFLOW,MMX,MMXEXT,MOVBE,NX,OSXSAVE,POPCNT,RDRAND,RDSEED,RDTSCP,SCE,SHA,SSE,SSE2,SSE3,SSE4,SSE42,SSE4A,SSSE3,SUCCOR,X87,XSAVE 172 Microarchitecture level: 3 173 Cacheline bytes: 64 174 L1 Instruction Cache: 32768 bytes 175 L1 Data Cache: 32768 bytes 176 L2 Cache: 524288 bytes 177 L3 Cache: 16777216 bytes 178 179 ``` 180 ### JSON Output: 181 182 ``` 183 λ cpuid --json 184 { 185 "BrandName": "AMD Ryzen 9 3950X 16-Core Processor", 186 "VendorID": 2, 187 "VendorString": "AuthenticAMD", 188 "PhysicalCores": 16, 189 "ThreadsPerCore": 2, 190 "LogicalCores": 32, 191 "Family": 23, 192 "Model": 113, 193 "CacheLine": 64, 194 "Hz": 0, 195 "BoostFreq": 0, 196 "Cache": { 197 "L1I": 32768, 198 "L1D": 32768, 199 "L2": 524288, 200 "L3": 16777216 201 }, 202 "SGX": { 203 "Available": false, 204 "LaunchControl": false, 205 "SGX1Supported": false, 206 "SGX2Supported": false, 207 "MaxEnclaveSizeNot64": 0, 208 "MaxEnclaveSize64": 0, 209 "EPCSections": null 210 }, 211 "Features": [ 212 "ADX", 213 "AESNI", 214 "AVX", 215 "AVX2", 216 "BMI1", 217 "BMI2", 218 "CLMUL", 219 "CLZERO", 220 "CMOV", 221 "CMPXCHG8", 222 "CPBOOST", 223 "CX16", 224 "F16C", 225 "FMA3", 226 "FXSR", 227 "FXSROPT", 228 "HTT", 229 "HYPERVISOR", 230 "LAHF", 231 "LZCNT", 232 "MCAOVERFLOW", 233 "MMX", 234 "MMXEXT", 235 "MOVBE", 236 "NX", 237 "OSXSAVE", 238 "POPCNT", 239 "RDRAND", 240 "RDSEED", 241 "RDTSCP", 242 "SCE", 243 "SHA", 244 "SSE", 245 "SSE2", 246 "SSE3", 247 "SSE4", 248 "SSE42", 249 "SSE4A", 250 "SSSE3", 251 "SUCCOR", 252 "X87", 253 "XSAVE" 254 ], 255 "X64Level": 3 256 } 257 ``` 258 259 ### Check CPU microarch level 260 261 ``` 262 λ cpuid --check-level=3 263 2022/03/18 17:04:40 AMD Ryzen 9 3950X 16-Core Processor 264 2022/03/18 17:04:40 Microarchitecture level 3 is supported. Max level is 3. 265 Exit Code 0 266 267 λ cpuid --check-level=4 268 2022/03/18 17:06:18 AMD Ryzen 9 3950X 16-Core Processor 269 2022/03/18 17:06:18 Microarchitecture level 4 not supported. Max level is 3. 270 Exit Code 1 271 ``` 272 273 274 ## Available flags 275 276 ### x86 & amd64 277 278 | Feature Flag | Description | 279 |--------------------|------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------| 280 | ADX | Intel ADX (Multi-Precision Add-Carry Instruction Extensions) | 281 | AESNI | Advanced Encryption Standard New Instructions | 282 | AMD3DNOW | AMD 3DNOW | 283 | AMD3DNOWEXT | AMD 3DNowExt | 284 | AMXBF16 | Tile computational operations on BFLOAT16 numbers | 285 | AMXINT8 | Tile computational operations on 8-bit integers | 286 | AMXFP16 | Tile computational operations on FP16 numbers | 287 | AMXTILE | Tile architecture | 288 | AVX | AVX functions | 289 | AVX2 | AVX2 functions | 290 | AVX512BF16 | AVX-512 BFLOAT16 Instructions | 291 | AVX512BITALG | AVX-512 Bit Algorithms | 292 | AVX512BW | AVX-512 Byte and Word Instructions | 293 | AVX512CD | AVX-512 Conflict Detection Instructions | 294 | AVX512DQ | AVX-512 Doubleword and Quadword Instructions | 295 | AVX512ER | AVX-512 Exponential and Reciprocal Instructions | 296 | AVX512F | AVX-512 Foundation | 297 | AVX512FP16 | AVX-512 FP16 Instructions | 298 | AVX512IFMA | AVX-512 Integer Fused Multiply-Add Instructions | 299 | AVX512PF | AVX-512 Prefetch Instructions | 300 | AVX512VBMI | AVX-512 Vector Bit Manipulation Instructions | 301 | AVX512VBMI2 | AVX-512 Vector Bit Manipulation Instructions, Version 2 | 302 | AVX512VL | AVX-512 Vector Length Extensions | 303 | AVX512VNNI | AVX-512 Vector Neural Network Instructions | 304 | AVX512VP2INTERSECT | AVX-512 Intersect for D/Q | 305 | AVX512VPOPCNTDQ | AVX-512 Vector Population Count Doubleword and Quadword | 306 | AVXIFMA | AVX-IFMA instructions | 307 | AVXNECONVERT | AVX-NE-CONVERT instructions | 308 | AVXSLOW | Indicates the CPU performs 2 128 bit operations instead of one | 309 | AVXVNNI | AVX (VEX encoded) VNNI neural network instructions | 310 | AVXVNNIINT8 | AVX-VNNI-INT8 instructions | 311 | BHI_CTRL | Branch History Injection and Intra-mode Branch Target Injection / CVE-2022-0001, CVE-2022-0002 / INTEL-SA-00598 | 312 | BMI1 | Bit Manipulation Instruction Set 1 | 313 | BMI2 | Bit Manipulation Instruction Set 2 | 314 | CETIBT | Intel CET Indirect Branch Tracking | 315 | CETSS | Intel CET Shadow Stack | 316 | CLDEMOTE | Cache Line Demote | 317 | CLMUL | Carry-less Multiplication | 318 | CLZERO | CLZERO instruction supported | 319 | CMOV | i686 CMOV | 320 | CMPCCXADD | CMPCCXADD instructions | 321 | CMPSB_SCADBS_SHORT | Fast short CMPSB and SCASB | 322 | CMPXCHG8 | CMPXCHG8 instruction | 323 | CPBOOST | Core Performance Boost | 324 | CPPC | AMD: Collaborative Processor Performance Control | 325 | CX16 | CMPXCHG16B Instruction | 326 | EFER_LMSLE_UNS | AMD: =Core::X86::Msr::EFER[LMSLE] is not supported, and MBZ | 327 | ENQCMD | Enqueue Command | 328 | ERMS | Enhanced REP MOVSB/STOSB | 329 | F16C | Half-precision floating-point conversion | 330 | FLUSH_L1D | Flush L1D cache | 331 | FMA3 | Intel FMA 3. Does not imply AVX. | 332 | FMA4 | Bulldozer FMA4 functions | 333 | FP128 | AMD: When set, the internal FP/SIMD execution datapath is 128-bits wide | 334 | FP256 | AMD: When set, the internal FP/SIMD execution datapath is 256-bits wide | 335 | FSRM | Fast Short Rep Mov | 336 | FXSR | FXSAVE, FXRESTOR instructions, CR4 bit 9 | 337 | FXSROPT | FXSAVE/FXRSTOR optimizations | 338 | GFNI | Galois Field New Instructions. May require other features (AVX, AVX512VL,AVX512F) based on usage. | 339 | HLE | Hardware Lock Elision | 340 | HRESET | If set CPU supports history reset and the IA32_HRESET_ENABLE MSR | 341 | HTT | Hyperthreading (enabled) | 342 | HWA | Hardware assert supported. Indicates support for MSRC001_10 | 343 | HYBRID_CPU | This part has CPUs of more than one type. | 344 | HYPERVISOR | This bit has been reserved by Intel & AMD for use by hypervisors | 345 | IA32_ARCH_CAP | IA32_ARCH_CAPABILITIES MSR (Intel) | 346 | IA32_CORE_CAP | IA32_CORE_CAPABILITIES MSR | 347 | IBPB | Indirect Branch Restricted Speculation (IBRS) and Indirect Branch Predictor Barrier (IBPB) | 348 | IBRS | AMD: Indirect Branch Restricted Speculation | 349 | IBRS_PREFERRED | AMD: IBRS is preferred over software solution | 350 | IBRS_PROVIDES_SMP | AMD: IBRS provides Same Mode Protection | 351 | IBS | Instruction Based Sampling (AMD) | 352 | IBSBRNTRGT | Instruction Based Sampling Feature (AMD) | 353 | IBSFETCHSAM | Instruction Based Sampling Feature (AMD) | 354 | IBSFFV | Instruction Based Sampling Feature (AMD) | 355 | IBSOPCNT | Instruction Based Sampling Feature (AMD) | 356 | IBSOPCNTEXT | Instruction Based Sampling Feature (AMD) | 357 | IBSOPSAM | Instruction Based Sampling Feature (AMD) | 358 | IBSRDWROPCNT | Instruction Based Sampling Feature (AMD) | 359 | IBSRIPINVALIDCHK | Instruction Based Sampling Feature (AMD) | 360 | IBS_FETCH_CTLX | AMD: IBS fetch control extended MSR supported | 361 | IBS_OPDATA4 | AMD: IBS op data 4 MSR supported | 362 | IBS_OPFUSE | AMD: Indicates support for IbsOpFuse | 363 | IBS_PREVENTHOST | Disallowing IBS use by the host supported | 364 | IBS_ZEN4 | Fetch and Op IBS support IBS extensions added with Zen4 | 365 | IDPRED_CTRL | IPRED_DIS | 366 | INT_WBINVD | WBINVD/WBNOINVD are interruptible. | 367 | INVLPGB | NVLPGB and TLBSYNC instruction supported | 368 | LAHF | LAHF/SAHF in long mode | 369 | LAM | If set, CPU supports Linear Address Masking | 370 | LBRVIRT | LBR virtualization | 371 | LZCNT | LZCNT instruction | 372 | MCAOVERFLOW | MCA overflow recovery support. | 373 | MCDT_NO | Processor do not exhibit MXCSR Configuration Dependent Timing behavior and do not need to mitigate it. | 374 | MCOMMIT | MCOMMIT instruction supported | 375 | MD_CLEAR | VERW clears CPU buffers | 376 | MMX | standard MMX | 377 | MMXEXT | SSE integer functions or AMD MMX ext | 378 | MOVBE | MOVBE instruction (big-endian) | 379 | MOVDIR64B | Move 64 Bytes as Direct Store | 380 | MOVDIRI | Move Doubleword as Direct Store | 381 | MOVSB_ZL | Fast Zero-Length MOVSB | 382 | MPX | Intel MPX (Memory Protection Extensions) | 383 | MOVU | MOVU SSE instructions are more efficient and should be preferred to SSE MOVL/MOVH. MOVUPS is more efficient than MOVLPS/MOVHPS. MOVUPD is more efficient than MOVLPD/MOVHPD | 384 | MSRIRC | Instruction Retired Counter MSR available | 385 | MSRLIST | Read/Write List of Model Specific Registers | 386 | MSR_PAGEFLUSH | Page Flush MSR available | 387 | NRIPS | Indicates support for NRIP save on VMEXIT | 388 | NX | NX (No-Execute) bit | 389 | OSXSAVE | XSAVE enabled by OS | 390 | PCONFIG | PCONFIG for Intel Multi-Key Total Memory Encryption | 391 | POPCNT | POPCNT instruction | 392 | PPIN | AMD: Protected Processor Inventory Number support. Indicates that Protected Processor Inventory Number (PPIN) capability can be enabled | 393 | PREFETCHI | PREFETCHIT0/1 instructions | 394 | PSFD | Predictive Store Forward Disable | 395 | RDPRU | RDPRU instruction supported | 396 | RDRAND | RDRAND instruction is available | 397 | RDSEED | RDSEED instruction is available | 398 | RDTSCP | RDTSCP Instruction | 399 | RRSBA_CTRL | Restricted RSB Alternate | 400 | RTM | Restricted Transactional Memory | 401 | RTM_ALWAYS_ABORT | Indicates that the loaded microcode is forcing RTM abort. | 402 | SERIALIZE | Serialize Instruction Execution | 403 | SEV | AMD Secure Encrypted Virtualization supported | 404 | SEV_64BIT | AMD SEV guest execution only allowed from a 64-bit host | 405 | SEV_ALTERNATIVE | AMD SEV Alternate Injection supported | 406 | SEV_DEBUGSWAP | Full debug state swap supported for SEV-ES guests | 407 | SEV_ES | AMD SEV Encrypted State supported | 408 | SEV_RESTRICTED | AMD SEV Restricted Injection supported | 409 | SEV_SNP | AMD SEV Secure Nested Paging supported | 410 | SGX | Software Guard Extensions | 411 | SGXLC | Software Guard Extensions Launch Control | 412 | SHA | Intel SHA Extensions | 413 | SME | AMD Secure Memory Encryption supported | 414 | SME_COHERENT | AMD Hardware cache coherency across encryption domains enforced | 415 | SPEC_CTRL_SSBD | Speculative Store Bypass Disable | 416 | SRBDS_CTRL | SRBDS mitigation MSR available | 417 | SSE | SSE functions | 418 | SSE2 | P4 SSE functions | 419 | SSE3 | Prescott SSE3 functions | 420 | SSE4 | Penryn SSE4.1 functions | 421 | SSE42 | Nehalem SSE4.2 functions | 422 | SSE4A | AMD Barcelona microarchitecture SSE4a instructions | 423 | SSSE3 | Conroe SSSE3 functions | 424 | STIBP | Single Thread Indirect Branch Predictors | 425 | STIBP_ALWAYSON | AMD: Single Thread Indirect Branch Prediction Mode has Enhanced Performance and may be left Always On | 426 | STOSB_SHORT | Fast short STOSB | 427 | SUCCOR | Software uncorrectable error containment and recovery capability. | 428 | SVM | AMD Secure Virtual Machine | 429 | SVMDA | Indicates support for the SVM decode assists. | 430 | SVMFBASID | SVM, Indicates that TLB flush events, including CR3 writes and CR4.PGE toggles, flush only the current ASID's TLB entries. Also indicates support for the extended VMCBTLB_Control | 431 | SVML | AMD SVM lock. Indicates support for SVM-Lock. | 432 | SVMNP | AMD SVM nested paging | 433 | SVMPF | SVM pause intercept filter. Indicates support for the pause intercept filter | 434 | SVMPFT | SVM PAUSE filter threshold. Indicates support for the PAUSE filter cycle count threshold | 435 | SYSCALL | System-Call Extension (SCE): SYSCALL and SYSRET instructions. | 436 | SYSEE | SYSENTER and SYSEXIT instructions | 437 | TBM | AMD Trailing Bit Manipulation | 438 | TLB_FLUSH_NESTED | AMD: Flushing includes all the nested translations for guest translations | 439 | TME | Intel Total Memory Encryption. The following MSRs are supported: IA32_TME_CAPABILITY, IA32_TME_ACTIVATE, IA32_TME_EXCLUDE_MASK, and IA32_TME_EXCLUDE_BASE. | 440 | TOPEXT | TopologyExtensions: topology extensions support. Indicates support for CPUID Fn8000_001D_EAX_x[N:0]-CPUID Fn8000_001E_EDX. | 441 | TSCRATEMSR | MSR based TSC rate control. Indicates support for MSR TSC ratio MSRC000_0104 | 442 | TSXLDTRK | Intel TSX Suspend Load Address Tracking | 443 | VAES | Vector AES. AVX(512) versions requires additional checks. | 444 | VMCBCLEAN | VMCB clean bits. Indicates support for VMCB clean bits. | 445 | VMPL | AMD VM Permission Levels supported | 446 | VMSA_REGPROT | AMD VMSA Register Protection supported | 447 | VMX | Virtual Machine Extensions | 448 | VPCLMULQDQ | Carry-Less Multiplication Quadword. Requires AVX for 3 register versions. | 449 | VTE | AMD Virtual Transparent Encryption supported | 450 | WAITPKG | TPAUSE, UMONITOR, UMWAIT | 451 | WBNOINVD | Write Back and Do Not Invalidate Cache | 452 | WRMSRNS | Non-Serializing Write to Model Specific Register | 453 | X87 | FPU | 454 | XGETBV1 | Supports XGETBV with ECX = 1 | 455 | XOP | Bulldozer XOP functions | 456 | XSAVE | XSAVE, XRESTOR, XSETBV, XGETBV | 457 | XSAVEC | Supports XSAVEC and the compacted form of XRSTOR. | 458 | XSAVEOPT | XSAVEOPT available | 459 | XSAVES | Supports XSAVES/XRSTORS and IA32_XSS | 460 461 # ARM features: 462 463 | Feature Flag | Description | 464 |--------------|------------------------------------------------------------------| 465 | AESARM | AES instructions | 466 | ARMCPUID | Some CPU ID registers readable at user-level | 467 | ASIMD | Advanced SIMD | 468 | ASIMDDP | SIMD Dot Product | 469 | ASIMDHP | Advanced SIMD half-precision floating point | 470 | ASIMDRDM | Rounding Double Multiply Accumulate/Subtract (SQRDMLAH/SQRDMLSH) | 471 | ATOMICS | Large System Extensions (LSE) | 472 | CRC32 | CRC32/CRC32C instructions | 473 | DCPOP | Data cache clean to Point of Persistence (DC CVAP) | 474 | EVTSTRM | Generic timer | 475 | FCMA | Floatin point complex number addition and multiplication | 476 | FP | Single-precision and double-precision floating point | 477 | FPHP | Half-precision floating point | 478 | GPA | Generic Pointer Authentication | 479 | JSCVT | Javascript-style double->int convert (FJCVTZS) | 480 | LRCPC | Weaker release consistency (LDAPR, etc) | 481 | PMULL | Polynomial Multiply instructions (PMULL/PMULL2) | 482 | SHA1 | SHA-1 instructions (SHA1C, etc) | 483 | SHA2 | SHA-2 instructions (SHA256H, etc) | 484 | SHA3 | SHA-3 instructions (EOR3, RAXI, XAR, BCAX) | 485 | SHA512 | SHA512 instructions | 486 | SM3 | SM3 instructions | 487 | SM4 | SM4 instructions | 488 | SVE | Scalable Vector Extension | 489 490 # license 491 492 This code is published under an MIT license. See LICENSE file for more information.