instrset_detect.cpp File Reference

#include "instrset.h"

Include dependency graph for instrset_detect.cpp:

This graph shows which files directly or indirectly include this file:

Go to the source code of this file.

Functions
int	instrset_detect (void)
bool	hasFMA3 (void)
bool	hasFMA4 (void)
bool	hasXOP (void)
bool	hasF16C (void)
bool	hasAVX512ER (void)

Function Documentation

bool hasAVX512ER

(

void

)

Definition at line 178 of file instrset_detect.cpp.

                           {
    if ( instrset_detect() < 9 ) return false; // must have AVX512F
    int abcd[4];                               // cpuid results
    cpuid( abcd, 7 );                          // call cpuid function 7
    return ( ( abcd[1] & ( 1 << 27 ) ) != 0 ); // ebx bit 27 indicates AVX512ER
  }

bool hasF16C

(

void

)

Definition at line 170 of file instrset_detect.cpp.

                       {
    if ( instrset_detect() < 7 ) return false; // must have AVX
    int abcd[4];                               // cpuid results
    cpuid( abcd, 1 );                          // call cpuid function 1
    return ( ( abcd[2] & ( 1 << 29 ) ) != 0 ); // ecx bit 29 indicates F16C
  }

bool hasFMA3

(

void

)

Definition at line 146 of file instrset_detect.cpp.

                       {
    if ( instrset_detect() < 7 ) return false; // must have AVX
    int abcd[4];                               // cpuid results
    cpuid( abcd, 1 );                          // call cpuid function 1
    return ( ( abcd[2] & ( 1 << 12 ) ) != 0 ); // ecx bit 12 indicates FMA3
  }

bool hasFMA4

(

void

)

Definition at line 154 of file instrset_detect.cpp.

                       {
    if ( instrset_detect() < 7 ) return false; // must have AVX
    int abcd[4];                               // cpuid results
    cpuid( abcd, 0x80000001 );                 // call cpuid function 0x80000001
    return ( ( abcd[2] & ( 1 << 16 ) ) != 0 ); // ecx bit 16 indicates FMA4
  }

bool hasXOP

(

void

)

Definition at line 162 of file instrset_detect.cpp.

                      {
    if ( instrset_detect() < 7 ) return false; // must have AVX
    int abcd[4];                               // cpuid results
    cpuid( abcd, 0x80000001 );                 // call cpuid function 0x80000001
    return ( ( abcd[2] & ( 1 << 11 ) ) != 0 ); // ecx bit 11 indicates XOP
  }

int instrset_detect

(

void

)

Definition at line 98 of file instrset_detect.cpp.

                              {

    static int iset = -1; // remember value for next call
    if ( iset >= 0 ) {
      return iset; // called before
    }
    iset        = 0;                                   // default value
    int abcd[4] = {0, 0, 0, 0};                        // cpuid results
    cpuid( abcd, 0 );                                  // call cpuid function 0
    if ( abcd[0] == 0 ) return iset;                   // no further cpuid function supported
    cpuid( abcd, 1 );                                  // call cpuid function 1 for feature flags
    if ( ( abcd[3] & ( 1 << 0 ) ) == 0 ) return iset;  // no floating point
    if ( ( abcd[3] & ( 1 << 23 ) ) == 0 ) return iset; // no MMX
    if ( ( abcd[3] & ( 1 << 15 ) ) == 0 ) return iset; // no conditional move
    if ( ( abcd[3] & ( 1 << 24 ) ) == 0 ) return iset; // no FXSAVE
    if ( ( abcd[3] & ( 1 << 25 ) ) == 0 ) return iset; // no SSE
    iset = 1;                                          // 1: SSE supported
    if ( ( abcd[3] & ( 1 << 26 ) ) == 0 ) return iset; // no SSE2
    iset = 2;                                          // 2: SSE2 supported
    if ( ( abcd[2] & ( 1 << 0 ) ) == 0 ) return iset;  // no SSE3
    iset = 3;                                          // 3: SSE3 supported
    if ( ( abcd[2] & ( 1 << 9 ) ) == 0 ) return iset;  // no SSSE3
    iset = 4;                                          // 4: SSSE3 supported
    if ( ( abcd[2] & ( 1 << 19 ) ) == 0 ) return iset; // no SSE4.1
    iset = 5;                                          // 5: SSE4.1 supported
    if ( ( abcd[2] & ( 1 << 23 ) ) == 0 ) return iset; // no POPCNT
    if ( ( abcd[2] & ( 1 << 20 ) ) == 0 ) return iset; // no SSE4.2
    iset = 6;                                          // 6: SSE4.2 supported
    if ( ( abcd[2] & ( 1 << 27 ) ) == 0 ) return iset; // no OSXSAVE
    if ( ( xgetbv( 0 ) & 6 ) != 6 ) return iset;       // AVX not enabled in O.S.
    if ( ( abcd[2] & ( 1 << 28 ) ) == 0 ) return iset; // no AVX
    iset = 7;                                          // 7: AVX supported
    cpuid( abcd, 7 );                                  // call cpuid leaf 7 for feature flags
    if ( ( abcd[1] & ( 1 << 5 ) ) == 0 ) return iset;  // no AVX2
    iset = 8;
    if ( ( abcd[1] & ( 1 << 16 ) ) == 0 ) return iset; // no AVX512
    cpuid( abcd, 0xD );                                // call cpuid leaf 0xD for feature flags
    if ( ( abcd[0] & 0x60 ) != 0x60 ) return iset;     // no AVX512
    iset = 9;
    cpuid( abcd, 7 );                                  // call cpuid leaf 7 for feature flags
    if ( ( abcd[1] & ( 1 << 31 ) ) == 0 ) return iset; // no AVX512VL
    iset = 10;
    if ( ( abcd[1] & 0x40020000 ) != 0x40020000 ) return iset; // no AVX512BW, AVX512DQ
    iset = 11;
    return iset;
  }