Report a bug If you spot a problem with this page, click here to create a Bugzilla issue. Improve this page Quickly fork, edit online, and submit a pull request for this page. Requires a signed-in GitHub account. This works well for small changes. If you'd like to make larger changes you may want to consider using local clone.


Identify the characteristics of the host CPU, providing information about cache sizes and assembly optimisation hints. This module is provided primarily for assembly language programmers.

References: Some of this information was extremely difficult to track down. Some of the documents below were found only in cached versions stored by search engines! This code relies on information found in:

Currently only works on x86 and Itanium CPUs. Many processors have bugs in their microcode for the CPUID instruction, so sometimes the cache information may be incorrect.
Don Clugston, Tomas Lindquist Olsen <>

Source: core/cpuid.d

struct CacheInfo;
Cache size and behaviour
size_t size;
Size of the cache, in kilobytes, per CPU.
For L1 unified (data + code) caches, this size is half the physical size.

(we don't halve it for larger sizes, since normally

data size is much greater than code size for critical loops).
ubyte associativity;
Number of ways of associativity, eg:
1 = direct mapped

2 = 2-way set associative

3 = 3-way set associative

ubyte.max = fully associative
uint lineSize;
Number of bytes read into the cache when a cache miss occurs.
CacheInfo[5] datacache;
Scheduled for deprecation. Please use dataCaches instead.
pure nothrow @nogc @property @trusted const(CacheInfo)[5] dataCaches();
The data caches. If there are fewer than 5 physical caches levels,
the remaining levels are set to size_t.max (== entire memory space)
pure nothrow @nogc @property @trusted string vendor();
Returns vendor string, for display purposes only.
Do NOT use this to determine features!

Note that some CPUs have programmable vendorIDs.
pure nothrow @nogc @property @trusted string processor();
Returns processor string, for display purposes only
pure nothrow @nogc @property @trusted bool x87onChip();
Does it have an x87 FPU on-chip?
pure nothrow @nogc @property @trusted bool mmx();
Is MMX supported?
pure nothrow @nogc @property @trusted bool sse();
Is SSE supported?
pure nothrow @nogc @property @trusted bool sse2();
Is SSE2 supported?
pure nothrow @nogc @property @trusted bool sse3();
Is SSE3 supported?
pure nothrow @nogc @property @trusted bool ssse3();
Is SSSE3 supported?
pure nothrow @nogc @property @trusted bool sse41();
Is SSE4.1 supported?
pure nothrow @nogc @property @trusted bool sse42();
Is SSE4.2 supported?
pure nothrow @nogc @property @trusted bool sse4a();
Is SSE4a supported?
pure nothrow @nogc @property @trusted bool aes();
Is AES supported
pure nothrow @nogc @property @trusted bool hasPclmulqdq();
Is pclmulqdq supported
pure nothrow @nogc @property @trusted bool hasRdrand();
Is rdrand supported
pure nothrow @nogc @property @trusted bool avx();
Is AVX supported
pure nothrow @nogc @property @trusted bool vaes();
Is VEX-Encoded AES supported
pure nothrow @nogc @property @trusted bool hasVpclmulqdq();
Is vpclmulqdq supported
pure nothrow @nogc @property @trusted bool fma();
Is FMA supported
pure nothrow @nogc @property @trusted bool fp16c();
Is FP16C supported
pure nothrow @nogc @property @trusted bool avx2();
Is AVX2 supported
pure nothrow @nogc @property @trusted bool hle();
Is HLE (hardware lock elision) supported
pure nothrow @nogc @property @trusted bool rtm();
Is RTM (restricted transactional memory) supported
pure nothrow @nogc @property @trusted bool hasRdseed();
Is rdseed supported
pure nothrow @nogc @property @trusted bool hasSha();
Is SHA supported
pure nothrow @nogc @property @trusted bool amd3dnow();
Is AMD 3DNOW supported?
pure nothrow @nogc @property @trusted bool amd3dnowExt();
Is AMD 3DNOW Ext supported?
pure nothrow @nogc @property @trusted bool amdMmx();
Are AMD extensions to MMX supported?
pure nothrow @nogc @property @trusted bool hasFxsr();
Is fxsave/fxrstor supported?
pure nothrow @nogc @property @trusted bool hasCmov();
Is cmov supported?
pure nothrow @nogc @property @trusted bool hasRdtsc();
Is rdtsc supported?
pure nothrow @nogc @property @trusted bool hasCmpxchg8b();
Is cmpxchg8b supported?
pure nothrow @nogc @property @trusted bool hasCmpxchg16b();
Is cmpxchg8b supported?
pure nothrow @nogc @property @trusted bool hasSysEnterSysExit();
pure nothrow @nogc @property @trusted bool has3dnowPrefetch();
Is 3DNow prefetch supported?
pure nothrow @nogc @property @trusted bool hasLahfSahf();
Are LAHF and SAHF supported in 64-bit mode?
pure nothrow @nogc @property @trusted bool hasPopcnt();
Is POPCNT supported?
pure nothrow @nogc @property @trusted bool hasLzcnt();
Is LZCNT supported?
pure nothrow @nogc @property @trusted bool isX86_64();
Is this an Intel64 or AMD 64?
pure nothrow @nogc @property @trusted bool isItanium();
Is this an IA64 (Itanium) processor?
pure nothrow @nogc @property @trusted bool hyperThreading();
Is hyperthreading supported?
pure nothrow @nogc @property @trusted uint threadsPerCPU();
Returns number of threads per CPU
pure nothrow @nogc @property @trusted uint coresPerCPU();
Returns number of cores in CPU
pure nothrow @nogc @property @trusted bool preferAthlon();
Optimisation hints for assembly code.
For forward compatibility, the CPU is compared against different

microarchitectures. For 32-bit x86, comparisons are made against

the Intel PPro/PII/PIII/PM family.

The major 32-bit x86 microarchitecture 'dynasties' have been:

* Intel P6 (PentiumPro, PII, PIII, PM, Core, Core2).

* AMD Athlon (K7, K8, K10).

* Intel NetBurst (Pentium 4, Pentium D).

* In-order Pentium (Pentium1, PMMX, Atom)

Other early CPUs (Nx586, AMD K5, K6, Centaur C3, Transmeta,

Cyrix, Rise) were mostly in-order.

Some new processors do not fit into the existing categories:

* Intel Atom 230/330 (family 6, model 0x1C) is an in-order core.

* Centaur Isiah = VIA Nano (family 6, model F) is an out-of-order core.

Within each dynasty, the optimisation techniques are largely

identical (eg, use instruction pairing for group 4). Major

instruction set improvements occur within each dynasty.

Does this CPU perform better on AMD K7 code than PentiumPro..Core2 code?
pure nothrow @nogc @property @trusted bool preferPentium4();
Does this CPU perform better on Pentium4 code than PentiumPro..Core2 code?
pure nothrow @nogc @property @trusted bool preferPentium1();
Does this CPU perform better on Pentium I code than Pentium Pro code?
uint stepping;
Warning: This field will be turned into a property in a future release.
Processor type (vendor-dependent).

This should be visible ONLY for display purposes.
uint model;
Warning: This field will be turned into a property in a future release.
Processor type (vendor-dependent).

This should be visible ONLY for display purposes.
uint family;
Warning: This field will be turned into a property in a future release.
Processor type (vendor-dependent).

This should be visible ONLY for display purposes.
uint numCacheLevels;
This field has been deprecated. Please use cacheLevels instead.
nothrow @nogc @property @trusted uint cacheLevels();
The number of cache levels in the CPU.