[Pixman] [PATCH] Adding infrastructure to permit future AVX2 implementations

Matt Turner mattst88 at gmail.com
Wed Aug 29 18:43:23 UTC 2018


Thank you for the patches! Some comments inline.

On Wed, Aug 22, 2018 at 10:03 AM raghuveer devulapalli
<raghuveer.devulapalli at intel.com> wrote:
>
> ---
>  configure.ac            | 44 ++++++++++++++++++++++++++++++++++++++++++++
>  pixman/Makefile.am      | 12 ++++++++++++
>  pixman/pixman-avx2.c    | 32 ++++++++++++++++++++++++++++++++
>  pixman/pixman-private.h |  5 +++++
>  pixman/pixman-x86.c     | 15 +++++++++++++--
>  5 files changed, 106 insertions(+), 2 deletions(-)
>  create mode 100644 pixman/pixman-avx2.c
>
> diff --git a/configure.ac b/configure.ac
> index e833e45..27f4305 100644
> --- a/configure.ac
> +++ b/configure.ac
> @@ -503,6 +503,48 @@ fi
>  AM_CONDITIONAL(USE_SSSE3, test $have_ssse3_intrinsics = yes)
>
>  dnl ===========================================================================
> +dnl Check for AVX2

Trailing whitespace

> +
> +if test "x$AVX2_CFLAGS" = "x" ; then
> +    AVX2_CFLAGS="-mavx2 -Winline"
> +fi
> +
> +have_avx2_intrinsics=no
> +AC_MSG_CHECKING(whether to use AVX2 intrinsics)
> +xserver_save_CFLAGS=$CFLAGS
> +CFLAGS="$AVX2_CFLAGS $CFLAGS"
> +
> +AC_COMPILE_IFELSE([AC_LANG_SOURCE([[
> +#include <immintrin.h>
> +int param;
> +int main () {
> +    __m256i a = _mm256_set1_epi32 (param), b = _mm256_set1_epi32 (param + 1), c;
> +    c = _mm256_maddubs_epi16 (a, b);
> +    return _mm256_cvtsi256_si32(c);
> +}]])], have_avx2_intrinsics=yes)
> +CFLAGS=$xserver_save_CFLAGS
> +
> +AC_ARG_ENABLE(avx2,
> +   [AC_HELP_STRING([--disable-avx2],
> +                   [disable AVX2 fast paths])],
> +   [enable_avx2=$enableval], [enable_avx2=auto])
> +
> +if test $enable_avx2 = no ; then
> +   have_avx2_intrinsics=disabled
> +fi
> +
> +if test $have_avx2_intrinsics = yes ; then
> +   AC_DEFINE(USE_AVX2, 1, [use AVX2 compiler intrinsics])
> +fi
> +
> +AC_MSG_RESULT($have_avx2_intrinsics)
> +if test $enable_avx2 = yes && test $have_avx2_intrinsics = no ; then
> +   AC_MSG_ERROR([AVX2 intrinsics not detected])
> +fi
> +
> +AM_CONDITIONAL(USE_AVX2, test $have_avx2_intrinsics = yes)
> +
> +dnl ===========================================================================
>  dnl Other special flags needed when building code using MMX or SSE instructions
>  case $host_os in
>     solaris*)
> @@ -538,6 +580,8 @@ AC_SUBST(MMX_LDFLAGS)
>  AC_SUBST(SSE2_CFLAGS)
>  AC_SUBST(SSE2_LDFLAGS)
>  AC_SUBST(SSSE3_CFLAGS)
> +AC_SUBST(AVX2_CFLAGS)
> +AC_SUBST(AVX2_LDFLAGS)
>
>  dnl ===========================================================================
>  dnl Check for VMX/Altivec
> diff --git a/pixman/Makefile.am b/pixman/Makefile.am
> index 581b6f6..7204621 100644
> --- a/pixman/Makefile.am
> +++ b/pixman/Makefile.am
> @@ -64,6 +64,18 @@ libpixman_1_la_LIBADD += libpixman-ssse3.la
>  ASM_CFLAGS_ssse3=$(SSSE3_CFLAGS)
>  endif
>
> +# avx2 code
> +if USE_AVX2
> +noinst_LTLIBRARIES += libpixman-avx2.la
> +libpixman_avx2_la_SOURCES = \
> +       pixman-avx2.c
> +libpixman_avx2_la_CFLAGS = $(AVX2_CFLAGS)
> +libpixman_1_la_LDFLAGS += $(AVX2_LDFLAGS)
> +libpixman_1_la_LIBADD += libpixman-avx2.la
> +
> +ASM_CFLAGS_avx2=$(AVX2_CFLAGS)
> +endif
> +
>  # arm simd code
>  if USE_ARM_SIMD
>  noinst_LTLIBRARIES += libpixman-arm-simd.la
> diff --git a/pixman/pixman-avx2.c b/pixman/pixman-avx2.c
> new file mode 100644
> index 0000000..d860d67
> --- /dev/null
> +++ b/pixman/pixman-avx2.c
> @@ -0,0 +1,32 @@
> +#ifdef HAVE_CONFIG_H
> +#include <config.h>
> +#endif
> +
> +#include <immintrin.h> /* for AVX2 intrinsics */
> +#include "pixman-private.h"
> +#include "pixman-combine32.h"
> +#include "pixman-inlines.h"
> +
> +static const pixman_fast_path_t avx2_fast_paths[] =
> +{
> +    { PIXMAN_OP_NONE },
> +};
> +
> +static const pixman_iter_info_t avx2_iters[] =

Trailing whitespace

> +{
> +    { PIXMAN_null },
> +};
> +
> +#if defined(__GNUC__) && !defined(__x86_64__) && !defined(__amd64__)
> +__attribute__((__force_align_arg_pointer__))
> +#endif
> +pixman_implementation_t *
> +_pixman_implementation_create_avx2 (pixman_implementation_t *fallback)
> +{
> +    pixman_implementation_t *imp = _pixman_implementation_create (fallback, avx2_fast_paths);
> +
> +    /* Set up function pointers */
> +    imp->iter_info = avx2_iters;
> +
> +    return imp;
> +}
> diff --git a/pixman/pixman-private.h b/pixman/pixman-private.h
> index 73a5414..b6b15df 100644
> --- a/pixman/pixman-private.h
> +++ b/pixman/pixman-private.h
> @@ -597,6 +597,11 @@ pixman_implementation_t *
>  _pixman_implementation_create_ssse3 (pixman_implementation_t *fallback);
>  #endif
>
> +#ifdef USE_AVX2
> +pixman_implementation_t *
> +_pixman_implementation_create_avx2 (pixman_implementation_t *fallback);
> +#endif
> +
>  #ifdef USE_ARM_SIMD
>  pixman_implementation_t *
>  _pixman_implementation_create_arm_simd (pixman_implementation_t *fallback);
> diff --git a/pixman/pixman-x86.c b/pixman/pixman-x86.c
> index 05297c4..687c83b 100644
> --- a/pixman/pixman-x86.c
> +++ b/pixman/pixman-x86.c

At the top of this file there is a preprocessor check:

#if defined(USE_X86_MMX) || defined (USE_SSE2) || defined (USE_SSSE3)

I think || defined (USE_AVX2) should be added here.

> @@ -40,7 +40,8 @@ typedef enum
>      X86_SSE                    = (1 << 2) | X86_MMX_EXTENSIONS,
>      X86_SSE2                   = (1 << 3),
>      X86_CMOV                   = (1 << 4),
> -    X86_SSSE3                  = (1 << 5)
> +    X86_SSSE3                  = (1 << 5),
> +    X86_AVX2                   = (1 << 6),

I'm not 100% we can use trailing commas in pixman due to MSVC.
Probably safer just to leave it off.

>  } cpu_features_t;
>
>  #ifdef HAVE_GETISAX
> @@ -119,7 +120,7 @@ pixman_cpuid (uint32_t feature,
>      __asm__ volatile (
>          "cpuid"                                "\n\t"
>         : "=a" (*a), "=b" (*b), "=c" (*c), "=d" (*d)
> -       : "a" (feature));
> +       : "a" (feature), "c" (0));

Just to make sure I'm understanding: cpuid returns AVX2 presence in
bit 5 of ebx when it is executed with eax=7 and ecx=0, so we need to
ensure ecx is set to 0?

I think that's fine. It seems like ecx isn't required to be any
particular value for the other cases. Perhaps a comment would help
future readers understand.

>  #else
>      /* On x86-32 we need to be careful about the handling of %ebx
>       * and %esp. We can't declare either one as clobbered
> @@ -172,6 +173,10 @@ detect_cpu_features (void)
>         features |= X86_SSE2;
>      if (c & (1 << 9))
>         features |= X86_SSSE3;
> +

Spurious whitespace

> +    pixman_cpuid (0x07, &a, &b, &c, &d);
> +    if (b & (1 << 5))
> +       features |= X86_AVX2;
>
>      /* Check for AMD specific features */
>      if ((features & X86_MMX) && !(features & X86_SSE))
> @@ -228,6 +233,7 @@ _pixman_x86_get_implementations (pixman_implementation_t *imp)
>  #define MMX_BITS  (X86_MMX | X86_MMX_EXTENSIONS)
>  #define SSE2_BITS (X86_MMX | X86_MMX_EXTENSIONS | X86_SSE | X86_SSE2)
>  #define SSSE3_BITS (X86_SSE | X86_SSE2 | X86_SSSE3)
> +#define AVX2_BITS (X86_AVX2)
>
>  #ifdef USE_X86_MMX
>      if (!_pixman_disabled ("mmx") && have_feature (MMX_BITS))
> @@ -244,5 +250,10 @@ _pixman_x86_get_implementations (pixman_implementation_t *imp)
>         imp = _pixman_implementation_create_ssse3 (imp);
>  #endif
>
> +#if (defined USE_AVX2 && defined USE_SSE2)
> +    if (!_pixman_disabled ("avx2") && have_feature (AVX2_BITS))
> +       imp = _pixman_implementation_create_avx2(imp);
> +#endif
> +
>      return imp;
>  }
> --
> 2.7.4


More information about the Pixman mailing list