On Mon, Feb 13, 2017 at 8:35 PM, Jakub Jelinek <ja...@redhat.com> wrote:
> Hi!
>
> As mentioned in the PR, ICC as well as clang have these non-masked
> gather prefetch intrinsics in addition to masked (and for scatter
> even GCC has both masked and non-masked), but GCC does not (the
> SDM actually doesn't mention those, only those for scatters).
>
> The following patch implements those, I think it is useful to have
> them for compatibility with the other compilers as well for consistency.
>
> Bootstrapped/regtested on x86_64-linux and i686-linux, ok for trunk?
>
> 2017-02-13 Jakub Jelinek <ja...@redhat.com>
>
> PR target/79481
> * config/i386/avx512pfintrin.h (_mm512_prefetch_i32gather_pd,
> _mm512_prefetch_i32gather_ps, _mm512_prefetch_i64gather_pd,
> _mm512_prefetch_i64gather_ps): New inline functions and macros.
>
> * gcc.target/i386/sse-14.c (test_2vx): Add void return type.
> (test_3vx): Change return type from int to void.
> (_mm512_prefetch_i32gather_ps, _mm512_prefetch_i32scatter_ps,
> _mm512_prefetch_i64gather_ps, _mm512_prefetch_i64scatter_ps,
> _mm512_prefetch_i32gather_pd, _mm512_prefetch_i32scatter_pd,
> _mm512_prefetch_i64gather_pd, _mm512_prefetch_i64scatter_pd): New
> tests.
> * gcc.target/i386/sse-22.c (test_2vx): Add void return type.
> (test_3vx): Change return type from int to void.
> (_mm512_prefetch_i32gather_ps, _mm512_prefetch_i32scatter_ps,
> _mm512_prefetch_i64gather_ps, _mm512_prefetch_i64scatter_ps,
> _mm512_prefetch_i32gather_pd, _mm512_prefetch_i32scatter_pd,
> _mm512_prefetch_i64gather_pd, _mm512_prefetch_i64scatter_pd): New
> tests.
> * gcc.target/i386/avx512pf-vgatherpf0dpd-1.c: Add non-masked
> intrinsic. Change scan-assembler-times number from 1 to 2.
> * gcc.target/i386/avx512pf-vgatherpf0dps-1.c: Likewise.
> * gcc.target/i386/avx512pf-vgatherpf0qpd-1.c: Likewise.
> * gcc.target/i386/avx512pf-vgatherpf0qps-1.c: Likewise.
> * gcc.target/i386/avx512pf-vgatherpf1dpd-1.c: Likewise.
> * gcc.target/i386/avx512pf-vgatherpf1dps-1.c: Likewise.
> * gcc.target/i386/avx512pf-vgatherpf1qpd-1.c: Likewise.
> * gcc.target/i386/avx512pf-vgatherpf1qps-1.c: Likewise.
OK.
Thanks,
Uros.
> --- gcc/config/i386/avx512pfintrin.h.jj 2017-01-17 18:40:59.000000000 +0100
> +++ gcc/config/i386/avx512pfintrin.h 2017-02-13 09:56:21.333303124 +0100
> @@ -48,6 +48,24 @@ typedef unsigned short __mmask16;
> #ifdef __OPTIMIZE__
> extern __inline void
> __attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
> +_mm512_prefetch_i32gather_pd (__m256i __index, void const *__addr,
> + int __scale, int __hint)
> +{
> + __builtin_ia32_gatherpfdpd ((__mmask8) 0xFF, (__v8si) __index, __addr,
> + __scale, __hint);
> +}
> +
> +extern __inline void
> +__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
> +_mm512_prefetch_i32gather_ps (__m512i __index, void const *__addr,
> + int __scale, int __hint)
> +{
> + __builtin_ia32_gatherpfdps ((__mmask16) 0xFFFF, (__v16si) __index, __addr,
> + __scale, __hint);
> +}
> +
> +extern __inline void
> +__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
> _mm512_mask_prefetch_i32gather_pd (__m256i __index, __mmask8 __mask,
> void const *__addr, int __scale, int
> __hint)
> {
> @@ -66,6 +84,24 @@ _mm512_mask_prefetch_i32gather_ps (__m51
>
> extern __inline void
> __attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
> +_mm512_prefetch_i64gather_pd (__m512i __index, void const *__addr,
> + int __scale, int __hint)
> +{
> + __builtin_ia32_gatherpfqpd ((__mmask8) 0xFF, (__v8di) __index, __addr,
> + __scale, __hint);
> +}
> +
> +extern __inline void
> +__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
> +_mm512_prefetch_i64gather_ps (__m512i __index, void const *__addr,
> + int __scale, int __hint)
> +{
> + __builtin_ia32_gatherpfqps ((__mmask8) 0xFF, (__v8di) __index, __addr,
> + __scale, __hint);
> +}
> +
> +extern __inline void
> +__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
> _mm512_mask_prefetch_i64gather_pd (__m512i __index, __mmask8 __mask,
> void const *__addr, int __scale, int
> __hint)
> {
> @@ -155,6 +191,14 @@ _mm512_mask_prefetch_i64scatter_ps (void
> }
>
> #else
> +#define _mm512_prefetch_i32gather_pd(INDEX, ADDR, SCALE, HINT) \
> + __builtin_ia32_gatherpfdpd ((__mmask8)0xFF, (__v8si)(__m256i)INDEX, \
> + (void const *)ADDR, (int)SCALE, (int)HINT)
> +
> +#define _mm512_prefetch_i32gather_ps(INDEX, ADDR, SCALE, HINT) \
> + __builtin_ia32_gatherpfdps ((__mmask16)0xFFFF, (__v16si)(__m512i)INDEX,
> \
> + (void const *)ADDR, (int)SCALE, (int)HINT)
> +
> #define _mm512_mask_prefetch_i32gather_pd(INDEX, MASK, ADDR, SCALE, HINT)
> \
> __builtin_ia32_gatherpfdpd ((__mmask8)MASK, (__v8si)(__m256i)INDEX, \
> (void const *)ADDR, (int)SCALE, (int)HINT)
> @@ -163,6 +207,14 @@ _mm512_mask_prefetch_i64scatter_ps (void
> __builtin_ia32_gatherpfdps ((__mmask16)MASK, (__v16si)(__m512i)INDEX,
> \
> (void const *)ADDR, (int)SCALE, (int)HINT)
>
> +#define _mm512_prefetch_i64gather_pd(INDEX, ADDR, SCALE, HINT) \
> + __builtin_ia32_gatherpfqpd ((__mmask8)0xFF, (__v8di)(__m512i)INDEX, \
> + (void *)ADDR, (int)SCALE, (int)HINT)
> +
> +#define _mm512_prefetch_i64gather_ps(INDEX, ADDR, SCALE, HINT) \
> + __builtin_ia32_gatherpfqps ((__mmask8)0xFF, (__v8di)(__m512i)INDEX, \
> + (void *)ADDR, (int)SCALE, (int)HINT)
> +
> #define _mm512_mask_prefetch_i64gather_pd(INDEX, MASK, ADDR, SCALE, HINT)
> \
> __builtin_ia32_gatherpfqpd ((__mmask8)MASK, (__v8di)(__m512i)INDEX, \
> (void *)ADDR, (int)SCALE, (int)HINT)
> --- gcc/testsuite/gcc.target/i386/sse-14.c.jj 2017-01-12 22:28:47.000000000
> +0100
> +++ gcc/testsuite/gcc.target/i386/sse-14.c 2017-02-13 10:15:36.815163082
> +0100
> @@ -50,7 +50,7 @@
> { return func (A, B, imm1, imm2, imm3); }
>
> #define test_2vx(func, op1_type, op2_type, imm1, imm2) \
> - _CONCAT(_,func) (op1_type A, op2_type B, int const I, int const L) \
> + void _CONCAT(_,func) (op1_type A, op2_type B, int const I, int const L) \
> { func (A, B, imm1, imm2); }
>
> #define test_3(func, type, op1_type, op2_type, op3_type, imm) \
> @@ -74,7 +74,7 @@
> { func (A, B, C, imm); }
>
> #define test_3vx(func, op1_type, op2_type, op3_type, imm1, imm2) \
> - int _CONCAT(_,func) (op1_type A, op2_type B, \
> + void _CONCAT(_,func) (op1_type A, op2_type B, \
> op3_type C, int const I, int const L) \
> { func (A, B, C, imm1, imm2); }
>
> @@ -520,6 +520,14 @@ test_4x (_mm_maskz_fixupimm_round_sd, __
> test_4x (_mm_maskz_fixupimm_round_ss, __m128, __mmask8, __m128, __m128,
> __m128i, 1, 8)
>
> /* avx512pfintrin.h */
> +test_2vx (_mm512_prefetch_i32gather_ps, __m512i, void const *, 1,
> _MM_HINT_T0)
> +test_2vx (_mm512_prefetch_i32scatter_ps, void const *, __m512i, 1,
> _MM_HINT_T0)
> +test_2vx (_mm512_prefetch_i64gather_ps, __m512i, void const *, 1,
> _MM_HINT_T0)
> +test_2vx (_mm512_prefetch_i64scatter_ps, void const *, __m512i, 1,
> _MM_HINT_T0)
> +test_2vx (_mm512_prefetch_i32gather_pd, __m256i, void const *, 1,
> _MM_HINT_T0)
> +test_2vx (_mm512_prefetch_i32scatter_pd, void const *, __m256i, 1,
> _MM_HINT_T0)
> +test_2vx (_mm512_prefetch_i64gather_pd, __m512i, void const *, 1,
> _MM_HINT_T0)
> +test_2vx (_mm512_prefetch_i64scatter_pd, void const *, __m512i, 1,
> _MM_HINT_T0)
> test_3vx (_mm512_mask_prefetch_i32gather_ps, __m512i, __mmask16, void const
> *, 1, _MM_HINT_T0)
> test_3vx (_mm512_mask_prefetch_i32scatter_ps, void const *, __mmask16,
> __m512i, 1, _MM_HINT_T0)
> test_3vx (_mm512_mask_prefetch_i64gather_ps, __m512i, __mmask8, void const
> *, 1, _MM_HINT_T0)
> --- gcc/testsuite/gcc.target/i386/sse-22.c.jj 2017-01-12 22:28:47.000000000
> +0100
> +++ gcc/testsuite/gcc.target/i386/sse-22.c 2017-02-13 10:15:54.781926974
> +0100
> @@ -50,7 +50,7 @@
> { return func (A, B, imm1, imm2, imm3); }
>
> #define test_2vx(func, op1_type, op2_type, imm1, imm2) \
> - _CONCAT(_,func) (op1_type A, op2_type B, int const I, int const L) \
> + void _CONCAT(_,func) (op1_type A, op2_type B, int const I, int const L) \
> { func (A, B, imm1, imm2); }
>
> #define test_3(func, type, op1_type, op2_type, op3_type, imm) \
> @@ -74,7 +74,7 @@
> { func (A, B, C, imm); }
>
> #define test_3vx(func, op1_type, op2_type, op3_type, imm1, imm2) \
> - int _CONCAT(_,func) (op1_type A, op2_type B, \
> + void _CONCAT(_,func) (op1_type A, op2_type B, \
> op3_type C, int const I, int const L) \
> { func (A, B, C, imm1, imm2); }
>
> @@ -647,11 +647,18 @@ test_4x (_mm_maskz_fixupimm_round_sd, __
> test_4x (_mm_maskz_fixupimm_round_ss, __m128, __mmask8, __m128, __m128,
> __m128i, 1, 8)
>
> /* avx512pfintrin.h */
> +test_2vx (_mm512_prefetch_i32gather_ps, __m512i, void const *, 1,
> _MM_HINT_T0)
> +test_2vx (_mm512_prefetch_i32scatter_ps, void const *, __m512i, 1,
> _MM_HINT_T0)
> +test_2vx (_mm512_prefetch_i64gather_ps, __m512i, void const *, 1,
> _MM_HINT_T0)
> +test_2vx (_mm512_prefetch_i64scatter_ps, void const *, __m512i, 1,
> _MM_HINT_T0)
> +test_2vx (_mm512_prefetch_i32gather_pd, __m256i, void const *, 1,
> _MM_HINT_T0)
> +test_2vx (_mm512_prefetch_i32scatter_pd, void const *, __m256i, 1,
> _MM_HINT_T0)
> +test_2vx (_mm512_prefetch_i64gather_pd, __m512i, long long *, 1, _MM_HINT_T0)
> +test_2vx (_mm512_prefetch_i64scatter_pd, void const *, __m512i, 1,
> _MM_HINT_T0)
> test_3vx (_mm512_mask_prefetch_i32gather_ps, __m512i, __mmask16, void const
> *, 1, _MM_HINT_T0)
> test_3vx (_mm512_mask_prefetch_i32scatter_ps, void const *, __mmask16,
> __m512i, 1, _MM_HINT_T0)
> test_3vx (_mm512_mask_prefetch_i64gather_ps, __m512i, __mmask8, void const
> *, 1, _MM_HINT_T0)
> test_3vx (_mm512_mask_prefetch_i64scatter_ps, void const *, __mmask8,
> __m512i, 1, _MM_HINT_T0)
> -
> test_3vx (_mm512_mask_prefetch_i32gather_pd, __m256i, __mmask8, void const
> *, 1, _MM_HINT_T0)
> test_3vx (_mm512_mask_prefetch_i32scatter_pd, void const *, __mmask8,
> __m256i, 1, _MM_HINT_T0)
> test_3vx (_mm512_mask_prefetch_i64gather_pd, __m512i, __mmask8, long long *,
> 1, _MM_HINT_T0)
> --- gcc/testsuite/gcc.target/i386/avx512pf-vgatherpf0dpd-1.c.jj 2016-05-22
> 12:20:09.000000000 +0200
> +++ gcc/testsuite/gcc.target/i386/avx512pf-vgatherpf0dpd-1.c 2017-02-13
> 10:22:18.154888926 +0100
> @@ -1,6 +1,6 @@
> /* { dg-do compile } */
> /* { dg-options "-mavx512pf -O2" } */
> -/* { dg-final { scan-assembler-times "vgatherpf0dpd\[
> \\t\]+\[^\{\n\]*\{%k\[1-7\]\}(?:\n|\[ \\t\]+#)" 1 } } */
> +/* { dg-final { scan-assembler-times "vgatherpf0dpd\[
> \\t\]+\[^\{\n\]*\{%k\[1-7\]\}(?:\n|\[ \\t\]+#)" 2 } } */
>
> #include <immintrin.h>
>
> @@ -11,5 +11,6 @@ void *base;
> void extern
> avx512pf_test (void)
> {
> + _mm512_prefetch_i32gather_pd (idx, base, 8, _MM_HINT_T0);
> _mm512_mask_prefetch_i32gather_pd (idx, m8, base, 8, _MM_HINT_T0);
> }
> --- gcc/testsuite/gcc.target/i386/avx512pf-vgatherpf0dps-1.c.jj 2016-05-22
> 12:20:23.000000000 +0200
> +++ gcc/testsuite/gcc.target/i386/avx512pf-vgatherpf0dps-1.c 2017-02-13
> 10:22:21.992838490 +0100
> @@ -1,6 +1,6 @@
> /* { dg-do compile } */
> /* { dg-options "-mavx512pf -O2" } */
> -/* { dg-final { scan-assembler-times "vgatherpf0dps\[
> \\t\]+\[^\{\n\]*\{%k\[1-7\]\}(?:\n|\[ \\t\]+#)" 1 } } */
> +/* { dg-final { scan-assembler-times "vgatherpf0dps\[
> \\t\]+\[^\{\n\]*\{%k\[1-7\]\}(?:\n|\[ \\t\]+#)" 2 } } */
>
> #include <immintrin.h>
>
> @@ -11,5 +11,6 @@ int *base;
> void extern
> avx512pf_test (void)
> {
> + _mm512_prefetch_i32gather_ps (idx, base, 8, _MM_HINT_T0);
> _mm512_mask_prefetch_i32gather_ps (idx, m16, base, 8, _MM_HINT_T0);
> }
> --- gcc/testsuite/gcc.target/i386/avx512pf-vgatherpf0qpd-1.c.jj 2016-05-22
> 12:20:32.000000000 +0200
> +++ gcc/testsuite/gcc.target/i386/avx512pf-vgatherpf0qpd-1.c 2017-02-13
> 10:22:26.097784546 +0100
> @@ -1,6 +1,6 @@
> /* { dg-do compile } */
> /* { dg-options "-mavx512pf -O2" } */
> -/* { dg-final { scan-assembler-times "vgatherpf0qpd\[
> \\t\]+\[^\{\n\]*\{%k\[1-7\]\}(?:\n|\[ \\t\]+#)" 1 } } */
> +/* { dg-final { scan-assembler-times "vgatherpf0qpd\[
> \\t\]+\[^\{\n\]*\{%k\[1-7\]\}(?:\n|\[ \\t\]+#)" 2 } } */
>
> #include <immintrin.h>
>
> @@ -11,5 +11,6 @@ int *base;
> void extern
> avx512pf_test (void)
> {
> + _mm512_prefetch_i64gather_pd (idx, base, 8, _MM_HINT_T0);
> _mm512_mask_prefetch_i64gather_pd (idx, m8, base, 8, _MM_HINT_T0);
> }
> --- gcc/testsuite/gcc.target/i386/avx512pf-vgatherpf0qps-1.c.jj 2016-05-22
> 12:20:13.000000000 +0200
> +++ gcc/testsuite/gcc.target/i386/avx512pf-vgatherpf0qps-1.c 2017-02-13
> 10:22:29.952733887 +0100
> @@ -1,6 +1,6 @@
> /* { dg-do compile } */
> /* { dg-options "-mavx512pf -O2" } */
> -/* { dg-final { scan-assembler-times "vgatherpf0qps\[
> \\t\]+\[^\{\n\]*\{%k\[1-7\]\}(?:\n|\[ \\t\]+#)" 1 } } */
> +/* { dg-final { scan-assembler-times "vgatherpf0qps\[
> \\t\]+\[^\{\n\]*\{%k\[1-7\]\}(?:\n|\[ \\t\]+#)" 2 } } */
>
> #include <immintrin.h>
>
> @@ -11,5 +11,6 @@ int *base;
> void extern
> avx512pf_test (void)
> {
> + _mm512_prefetch_i64gather_ps (idx, base, 8, _MM_HINT_T0);
> _mm512_mask_prefetch_i64gather_ps (idx, m8, base, 8, _MM_HINT_T0);
> }
> --- gcc/testsuite/gcc.target/i386/avx512pf-vgatherpf1dpd-1.c.jj 2016-05-22
> 12:20:31.000000000 +0200
> +++ gcc/testsuite/gcc.target/i386/avx512pf-vgatherpf1dpd-1.c 2017-02-13
> 10:22:33.866682452 +0100
> @@ -1,6 +1,6 @@
> /* { dg-do compile } */
> /* { dg-options "-mavx512pf -O2" } */
> -/* { dg-final { scan-assembler-times "vgatherpf1dpd\[
> \\t\]+\[^\{\n\]*\{%k\[1-7\]\}(?:\n|\[ \\t\]+#)" 1 } } */
> +/* { dg-final { scan-assembler-times "vgatherpf1dpd\[
> \\t\]+\[^\{\n\]*\{%k\[1-7\]\}(?:\n|\[ \\t\]+#)" 2 } } */
>
> #include <immintrin.h>
>
> @@ -11,5 +11,6 @@ int *base;
> void extern
> avx512pf_test (void)
> {
> + _mm512_prefetch_i32gather_pd (idx, base, 8, _MM_HINT_T1);
> _mm512_mask_prefetch_i32gather_pd (idx, m8, base, 8, _MM_HINT_T1);
> }
> --- gcc/testsuite/gcc.target/i386/avx512pf-vgatherpf1dps-1.c.jj 2016-05-22
> 12:20:13.000000000 +0200
> +++ gcc/testsuite/gcc.target/i386/avx512pf-vgatherpf1dps-1.c 2017-02-13
> 10:22:37.749631425 +0100
> @@ -1,6 +1,6 @@
> /* { dg-do compile } */
> /* { dg-options "-mavx512pf -O2" } */
> -/* { dg-final { scan-assembler-times "vgatherpf1dps\[
> \\t\]+\[^\{\n\]*\{%k\[1-7\]\}(?:\n|\[ \\t\]+#)" 1 } } */
> +/* { dg-final { scan-assembler-times "vgatherpf1dps\[
> \\t\]+\[^\{\n\]*\{%k\[1-7\]\}(?:\n|\[ \\t\]+#)" 2 } } */
>
> #include <immintrin.h>
>
> @@ -11,5 +11,6 @@ int *base;
> void extern
> avx512pf_test (void)
> {
> + _mm512_prefetch_i32gather_ps (idx, base, 8, _MM_HINT_T1);
> _mm512_mask_prefetch_i32gather_ps (idx, m16, base, 8, _MM_HINT_T1);
> }
> --- gcc/testsuite/gcc.target/i386/avx512pf-vgatherpf1qpd-1.c.jj 2016-05-22
> 12:20:18.000000000 +0200
> +++ gcc/testsuite/gcc.target/i386/avx512pf-vgatherpf1qpd-1.c 2017-02-13
> 10:22:41.703579464 +0100
> @@ -1,6 +1,6 @@
> /* { dg-do compile } */
> /* { dg-options "-mavx512pf -O2" } */
> -/* { dg-final { scan-assembler-times "vgatherpf1qpd\[
> \\t\]+\[^\{\n\]*\{%k\[1-7\]\}(?:\n|\[ \\t\]+#)" 1 } } */
> +/* { dg-final { scan-assembler-times "vgatherpf1qpd\[
> \\t\]+\[^\{\n\]*\{%k\[1-7\]\}(?:\n|\[ \\t\]+#)" 2 } } */
>
> #include <immintrin.h>
>
> @@ -11,5 +11,6 @@ int *base;
> void extern
> avx512pf_test (void)
> {
> + _mm512_prefetch_i64gather_pd (idx, base, 8, _MM_HINT_T1);
> _mm512_mask_prefetch_i64gather_pd (idx, m8, base, 8, _MM_HINT_T1);
> }
> --- gcc/testsuite/gcc.target/i386/avx512pf-vgatherpf1qps-1.c.jj 2016-05-22
> 12:20:02.000000000 +0200
> +++ gcc/testsuite/gcc.target/i386/avx512pf-vgatherpf1qps-1.c 2017-02-13
> 10:22:46.684514008 +0100
> @@ -1,6 +1,6 @@
> /* { dg-do compile } */
> /* { dg-options "-mavx512pf -O2" } */
> -/* { dg-final { scan-assembler-times "vgatherpf1qps\[
> \\t\]+\[^\{\n\]*\{%k\[1-7\]\}(?:\n|\[ \\t\]+#)" 1 } } */
> +/* { dg-final { scan-assembler-times "vgatherpf1qps\[
> \\t\]+\[^\{\n\]*\{%k\[1-7\]\}(?:\n|\[ \\t\]+#)" 2 } } */
>
> #include <immintrin.h>
>
> @@ -11,5 +11,6 @@ int *base;
> void extern
> avx512pf_test (void)
> {
> + _mm512_prefetch_i64gather_ps (idx, base, 8, _MM_HINT_T1);
> _mm512_mask_prefetch_i64gather_ps (idx, m8, base, 8, _MM_HINT_T1);
> }
>
> Jakub
