+2017-05-31 Sebastian Peryt <sebastian.peryt@intel.com>
+
+ * config/i386/avx512fintrin.h (_mm_mask_max_sd)
+ (_mm_maskz_max_sd, _mm_mask_max_ss, _mm_maskz_max_ss)
+ (_mm_mask_min_sd, _mm_maskz_min_sd, _mm_mask_min_ss)
+ (_mm_maskz_min_ss): New intrinsics.
+
2017-05-31 Martin Liska <mliska@suse.cz>
* tree-vect-loop.c (vect_create_epilog_for_reduction):
m_format_postprocessor.
(pretty_printer::~pretty_printer): Delete any
m_format_postprocessor.
- * pretty-print.h (printer_fn): Add bool and const char **
- parameters.
+ * pretty-print.h (printer_fn): Add bool and const char ** parameters.
(class format_postprocessor): New class.
- (struct pretty_printer::format_decoder): Document the new
- parameters.
+ (struct pretty_printer::format_decoder): Document the new parameters.
(struct pretty_printer::m_format_postprocessor): New field.
* tree-diagnostic.c (default_tree_printer): Update for new
bool and const char ** params.
2017-05-26 Richard Biener <rguenther@suse.de>
PR tree-optimization/80844
- * tree-vectorizer.c (adjust_simduid_builtins): Propagate
- results.
+ * tree-vectorizer.c (adjust_simduid_builtins): Propagate results.
2017-05-25 Sebastian Peryt <sebastian.peryt@intel.com>
_MM_FROUND_CUR_DIRECTION);
}
+extern __inline __m128d
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_mask_max_sd (__m128d __W, __mmask8 __U, __m128d __A, __m128d __B)
+{
+ return (__m128d) __builtin_ia32_maxsd_mask_round ((__v2df) __A,
+ (__v2df) __B,
+ (__v2df) __W,
+ (__mmask8) __U,
+ _MM_FROUND_CUR_DIRECTION);
+}
+
+extern __inline __m128d
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_maskz_max_sd (__mmask8 __U, __m128d __A, __m128d __B)
+{
+ return (__m128d) __builtin_ia32_maxsd_mask_round ((__v2df) __A,
+ (__v2df) __B,
+ (__v2df)
+ _mm_setzero_pd (),
+ (__mmask8) __U,
+ _MM_FROUND_CUR_DIRECTION);
+}
+
+extern __inline __m128
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_mask_max_ss (__m128 __W, __mmask8 __U, __m128 __A, __m128 __B)
+{
+ return (__m128) __builtin_ia32_maxss_mask_round ((__v4sf) __A,
+ (__v4sf) __B,
+ (__v4sf) __W,
+ (__mmask8) __U,
+ _MM_FROUND_CUR_DIRECTION);
+}
+
+extern __inline __m128
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_maskz_max_ss (__mmask8 __U, __m128 __A, __m128 __B)
+{
+ return (__m128) __builtin_ia32_maxss_mask_round ((__v4sf) __A,
+ (__v4sf) __B,
+ (__v4sf)
+ _mm_setzero_ps (),
+ (__mmask8) __U,
+ _MM_FROUND_CUR_DIRECTION);
+}
+
extern __inline __m512d
__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
_mm512_min_pd (__m512d __A, __m512d __B)
_MM_FROUND_CUR_DIRECTION);
}
+extern __inline __m128d
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_mask_min_sd (__m128d __W, __mmask8 __U, __m128d __A, __m128d __B)
+{
+ return (__m128d) __builtin_ia32_minsd_mask_round ((__v2df) __A,
+ (__v2df) __B,
+ (__v2df) __W,
+ (__mmask8) __U,
+ _MM_FROUND_CUR_DIRECTION);
+}
+
+extern __inline __m128d
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_maskz_min_sd (__mmask8 __U, __m128d __A, __m128d __B)
+{
+ return (__m128d) __builtin_ia32_minsd_mask_round ((__v2df) __A,
+ (__v2df) __B,
+ (__v2df)
+ _mm_setzero_pd (),
+ (__mmask8) __U,
+ _MM_FROUND_CUR_DIRECTION);
+}
+
+extern __inline __m128
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_mask_min_ss (__m128 __W, __mmask8 __U, __m128 __A, __m128 __B)
+{
+ return (__m128) __builtin_ia32_minss_mask_round ((__v4sf) __A,
+ (__v4sf) __B,
+ (__v4sf) __W,
+ (__mmask8) __U,
+ _MM_FROUND_CUR_DIRECTION);
+}
+
+extern __inline __m128
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_maskz_min_ss (__mmask8 __U, __m128 __A, __m128 __B)
+{
+ return (__m128) __builtin_ia32_minss_mask_round ((__v4sf) __A,
+ (__v4sf) __B,
+ (__v4sf)
+ _mm_setzero_ps (),
+ (__mmask8) __U,
+ _MM_FROUND_CUR_DIRECTION);
+}
+
extern __inline __m512d
__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
_mm512_scalef_pd (__m512d __A, __m512d __B)
+2017-05-31 Sebastian Peryt <sebastian.peryt@intel.com>
+
+ * gcc.target/i386/avx512f-vmaxsd-1.c (_mm_mask_max_sd)
+ (_mm_maskz_max_sd): Test new intrinsics.
+ * gcc.target/i386/avx512f-vmaxsd-2.c (_mm_mask_max_sd)
+ (_mm_maskz_max_sd): Test new intrinsics.
+ * gcc.target/i386/avx512f-vmaxss-1.c (_mm_mask_max_ss)
+ (_mm_maskz_max_ss): Test new intrinsics.
+ * gcc.target/i386/avx512f-vmaxss-2.c (_mm_mask_max_ss)
+ (_mm_maskz_max_ss): Test new intrinsics.
+ * gcc.target/i386/avx512f-vminsd-1.c (_mm_mask_min_sd)
+ (_mm_maskz_min_sd): Test new intrinsics.
+ * gcc.target/i386/avx512f-vminsd-2.c (_mm_mask_min_sd)
+ (_mm_maskz_min_sd): Test new intrinsics.
+ * gcc.target/i386/avx512f-vminss-1.c (_mm_mask_min_ss)
+ (_mm_maskz_min_ss): Test new intrinsics.
+ * gcc.target/i386/avx512f-vminss-2.c (_mm_mask_min_ss)
+ (_mm_maskz_min_ss): Test new intrinsics.
+
2017-05-31 Nathan Sidwell <nathan@acm.org>
* g++.dg/lookup/lambda1.C New.
/* { dg-do compile } */
/* { dg-options "-mavx512f -O2" } */
+/* { dg-final { scan-assembler-times "vmaxsd\[ \\t\]+\[^\{\n\]*%xmm\[0-9\]+\{%k\[1-7\]\}(?:\n|\[ \\t\]+#)" 1 } } */
+/* { dg-final { scan-assembler-times "vmaxsd\[ \\t\]+\[^\{\n\]*%xmm\[0-9\]+\{%k\[1-7\]\}\{z\}(?:\n|\[ \\t\]+#)" 1 } } */
/* { dg-final { scan-assembler-times "vmaxsd\[ \\t\]+\[^\{\n\]*\{sae\}\[^\n\]*%xmm\[0-9\]+(?:\n|\[ \\t\]+#)" 1 } } */
/* { dg-final { scan-assembler-times "vmaxsd\[ \\t\]+\[^\n\]*\{sae\}\[^\{\n\]*%xmm\[0-9\]+\{%k\[1-7\]\}(?:\n|\[ \\t\]+#)" 1 } } */
/* { dg-final { scan-assembler-times "vmaxsd\[ \\t\]+\[^\n\]*\{sae\}\[^\{\n\]*%xmm\[0-9\]+\{%k\[1-7\]\}\{z\}(?:\n|\[ \\t\]+#)" 1 } } */
void extern
avx512f_test (void)
{
+ x1 = _mm_mask_max_sd (x1, m, x2, x3);
+ x1 = _mm_maskz_max_sd (m, x1, x2);
x1 = _mm_max_round_sd (x1, x2, _MM_FROUND_NO_EXC);
x1 = _mm_mask_max_round_sd (x1, m, x2, x3, _MM_FROUND_NO_EXC);
x1 = _mm_maskz_max_round_sd (m, x1, x2, _MM_FROUND_NO_EXC);
avx512f_test (void)
{
int i, sign;
- union128d res1, res2, res3, src1, src2;
+ union128d res1, res2, res3, res4, res5, src1, src2;
MASK_TYPE mask = MASK_VALUE;
double res_ref[SIZE];
{
src1.a[i] = 1.5 + 34.67 * i * sign;
src2.a[i] = -22.17 * i * sign + 1.0;
+ res1.a[i] = DEFAULT_VALUE;
+ res4.a[i] = DEFAULT_VALUE;
sign = sign * -1;
}
- for (i = 0; i < SIZE; i++)
- res2.a[i] = DEFAULT_VALUE;
- res1.x = _mm_max_round_sd (src1.x, src2.x, _MM_FROUND_NO_EXC);
- res2.x = _mm_mask_max_round_sd (res2.x, mask, src1.x, src2.x, _MM_FROUND_NO_EXC);
- res3.x = _mm_maskz_max_round_sd (mask, src1.x, src2.x, _MM_FROUND_NO_EXC);
+ res1.x = _mm_mask_max_sd (res1.x, mask, src1.x, src2.x);
+ res2.x = _mm_maskz_max_sd (mask, src1.x, src2.x);
+ res3.x = _mm_max_round_sd (src1.x, src2.x, _MM_FROUND_NO_EXC);
+ res4.x = _mm_mask_max_round_sd (res4.x, mask, src1.x, src2.x, _MM_FROUND_NO_EXC);
+ res5.x = _mm_maskz_max_round_sd (mask, src1.x, src2.x, _MM_FROUND_NO_EXC);
calc_max (res_ref, src1.a, src2.a);
+ MASK_MERGE (d) (res_ref, mask, 1);
if (check_union128d (res1, res_ref))
+ abort ();
+
+ MASK_ZERO (d) (res_ref, mask, 1);
+ if (check_union128d (res2, res_ref))
+ abort ();
+
+ calc_max (res_ref, src1.a, src2.a);
+
+ if (check_union128d (res3, res_ref))
abort();
MASK_MERGE (d) (res_ref, mask, 1);
- if (check_union128d (res2, res_ref))
+ if (check_union128d (res4, res_ref))
abort ();
MASK_ZERO (d) (res_ref, mask, 1);
- if (check_union128d (res3, res_ref))
+ if (check_union128d (res5, res_ref))
abort ();
}
/* { dg-do compile } */
/* { dg-options "-mavx512f -O2" } */
+/* { dg-final { scan-assembler-times "vmaxss\[ \\t\]+\[^\{\n\]*%xmm\[0-9\]+\{%k\[1-7\]\}(?:\n|\[ \\t\]+#)" 1 } } */
+/* { dg-final { scan-assembler-times "vmaxss\[ \\t\]+\[^\{\n\]*%xmm\[0-9\]+\{%k\[1-7\]\}\{z\}(?:\n|\[ \\t\]+#)" 1 } } */
/* { dg-final { scan-assembler-times "vmaxss\[ \\t\]+\[^\{\n\]*\{sae\}\[^\n\]*%xmm\[0-9\]+(?:\n|\[ \\t\]+#)" 1 } } */
/* { dg-final { scan-assembler-times "vmaxss\[ \\t\]+\[^\n\]*\{sae\}\[^\{\n\]*%xmm\[0-9\]+\{%k\[1-7\]\}(?:\n|\[ \\t\]+#)" 1 } } */
/* { dg-final { scan-assembler-times "vmaxss\[ \\t\]+\[^\n\]*\{sae\}\[^\{\n\]*%xmm\[0-9\]+\{%k\[1-7\]\}\{z\}(?:\n|\[ \\t\]+#)" 1 } } */
void extern
avx512f_test (void)
{
+ x1 = _mm_mask_max_ss (x1, m, x2, x3);
+ x1 = _mm_maskz_max_ss (m, x1, x2);
x1 = _mm_max_round_ss (x1, x2, _MM_FROUND_NO_EXC);
x1 = _mm_mask_max_round_ss (x1, m, x2, x3, _MM_FROUND_NO_EXC);
x1 = _mm_maskz_max_round_ss (m, x1, x2, _MM_FROUND_NO_EXC);
avx512f_test (void)
{
int i, sign;
- union128 res1, res2, res3, src1, src2;
+ union128 res1, res2, res3, res4, res5, src1, src2;
MASK_TYPE mask = MASK_VALUE;
float res_ref[SIZE];
{
src1.a[i] = 1.5 + 34.67 * i * sign;
src2.a[i] = -22.17 * i * sign + 1.0;
+ res1.a[i] = DEFAULT_VALUE;
+ res4.a[i] = DEFAULT_VALUE;
sign = sign * -1;
}
- for (i = 0; i < SIZE; i++)
- res2.a[i] = DEFAULT_VALUE;
- res1.x = _mm_max_round_ss (src1.x, src2.x, _MM_FROUND_NO_EXC);
- res2.x = _mm_mask_max_round_ss (res2.x, mask, src1.x, src2.x, _MM_FROUND_NO_EXC);
- res3.x = _mm_maskz_max_round_ss (mask, src1.x, src2.x, _MM_FROUND_NO_EXC);
+ res1.x = _mm_mask_max_ss (res1.x, mask, src1.x, src2.x);
+ res2.x = _mm_maskz_max_ss (mask, src1.x, src2.x);
+ res3.x = _mm_max_round_ss (src1.x, src2.x, _MM_FROUND_NO_EXC);
+ res4.x = _mm_mask_max_round_ss (res4.x, mask, src1.x, src2.x, _MM_FROUND_NO_EXC);
+ res5.x = _mm_maskz_max_round_ss (mask, src1.x, src2.x, _MM_FROUND_NO_EXC);
calc_max (res_ref, src1.a, src2.a);
+ MASK_MERGE () (res_ref, mask, 1);
if (check_union128 (res1, res_ref))
+ abort ();
+
+ MASK_ZERO () (res_ref, mask, 1);
+ if (check_union128 (res2, res_ref))
+ abort ();
+
+ calc_max (res_ref, src1.a, src2.a);
+
+ if (check_union128 (res3, res_ref))
abort();
MASK_MERGE () (res_ref, mask, 1);
- if (check_union128 (res2, res_ref))
+ if (check_union128 (res4, res_ref))
abort ();
MASK_ZERO () (res_ref, mask, 1);
- if (check_union128 (res3, res_ref))
+ if (check_union128 (res5, res_ref))
abort ();
}
/* { dg-do compile } */
/* { dg-options "-mavx512f -O2" } */
+/* { dg-final { scan-assembler-times "vminsd\[ \\t\]+\[^\{\n\]*%xmm\[0-9\]+\{%k\[1-7\]\}(?:\n|\[ \\t\]+#)" 1 } } */
+/* { dg-final { scan-assembler-times "vminsd\[ \\t\]+\[^\{\n\]*%xmm\[0-9\]+\{%k\[1-7\]\}\{z\}(?:\n|\[ \\t\]+#)" 1 } } */
/* { dg-final { scan-assembler-times "vminsd\[ \\t\]+\[^\{\n\]*\{sae\}\[^\n\]*%xmm\[0-9\]+(?:\n|\[ \\t\]+#)" 1 } } */
/* { dg-final { scan-assembler-times "vminsd\[ \\t\]+\[^\n\]*\{sae\}\[^\{\n\]*%xmm\[0-9\]+\{%k\[1-7\]\}(?:\n|\[ \\t\]+#)" 1 } } */
/* { dg-final { scan-assembler-times "vminsd\[ \\t\]+\[^\n\]*\{sae\}\[^\{\n\]*%xmm\[0-9\]+\{%k\[1-7\]\}\{z\}(?:\n|\[ \\t\]+#)" 1 } } */
void extern
avx512f_test (void)
{
+ x1 = _mm_mask_min_sd (x1, m, x2, x3);
+ x1 = _mm_maskz_min_sd (m, x1, x2);
x1 = _mm_min_round_sd (x1, x2, _MM_FROUND_NO_EXC);
x1 = _mm_mask_min_round_sd (x1, m, x2, x3, _MM_FROUND_NO_EXC);
x1 = _mm_maskz_min_round_sd (m, x1, x2, _MM_FROUND_NO_EXC);
avx512f_test (void)
{
int i, sign;
- union128d res1, res2, res3, src1, src2;
+ union128d res1, res2, res3, res4, res5, src1, src2;
MASK_TYPE mask = MASK_VALUE;
double res_ref[SIZE];
{
src1.a[i] = 1.5 + 34.67 * i * sign;
src2.a[i] = -22.17 * i * sign + 1.0;
+ res1.a[i] = DEFAULT_VALUE;
+ res4.a[i] = DEFAULT_VALUE;
sign = sign * -1;
}
- for (i = 0; i < SIZE; i++)
- res2.a[i] = DEFAULT_VALUE;
-
- res1.x = _mm_min_round_sd (src1.x, src2.x, _MM_FROUND_NO_EXC);
- res2.x = _mm_mask_min_round_sd (res2.x, mask, src1.x, src2.x, _MM_FROUND_NO_EXC);
- res3.x = _mm_maskz_min_round_sd (mask, src1.x, src2.x, _MM_FROUND_NO_EXC);
+
+ res1.x = _mm_mask_min_sd (res1.x, mask, src1.x, src2.x);
+ res2.x = _mm_maskz_min_sd (mask, src1.x, src2.x);
+ res3.x = _mm_min_round_sd (src1.x, src2.x, _MM_FROUND_NO_EXC);
+ res4.x = _mm_mask_min_round_sd (res4.x, mask, src1.x, src2.x, _MM_FROUND_NO_EXC);
+ res5.x = _mm_maskz_min_round_sd (mask, src1.x, src2.x, _MM_FROUND_NO_EXC);
calc_min (res_ref, src1.a, src2.a);
+ MASK_MERGE (d) (res_ref, mask, 1);
if (check_union128d (res1, res_ref))
+ abort ();
+
+ MASK_ZERO (d) (res_ref, mask, 1);
+ if (check_union128d (res2, res_ref))
+ abort ();
+
+ calc_min (res_ref, src1.a, src2.a);
+
+ if (check_union128d (res3, res_ref))
abort();
MASK_MERGE (d) (res_ref, mask, 1);
- if (check_union128d (res2, res_ref))
+ if (check_union128d (res4, res_ref))
abort ();
MASK_ZERO (d) (res_ref, mask, 1);
- if (check_union128d (res3, res_ref))
+ if (check_union128d (res5, res_ref))
abort ();
}
/* { dg-do compile } */
/* { dg-options "-mavx512f -O2" } */
+/* { dg-final { scan-assembler-times "vminss\[ \\t\]+\[^\{\n\]*%xmm\[0-9\]+\{%k\[1-7\]\}(?:\n|\[ \\t\]+#)" 1 } } */
+/* { dg-final { scan-assembler-times "vminss\[ \\t\]+\[^\{\n\]*%xmm\[0-9\]+\{%k\[1-7\]\}\{z\}(?:\n|\[ \\t\]+#)" 1 } } */
/* { dg-final { scan-assembler-times "vminss\[ \\t\]+\[^\{\n\]*\{sae\}\[^\n\]*%xmm\[0-9\]+(?:\n|\[ \\t\]+#)" 1 } } */
/* { dg-final { scan-assembler-times "vminss\[ \\t\]+\[^\n\]*\{sae\}\[^\{\n\]*%xmm\[0-9\]+\{%k\[1-7\]\}(?:\n|\[ \\t\]+#)" 1 } } */
/* { dg-final { scan-assembler-times "vminss\[ \\t\]+\[^\n\]*\{sae\}\[^\{\n\]*%xmm\[0-9\]+\{%k\[1-7\]\}\{z\}(?:\n|\[ \\t\]+#)" 1 } } */
void extern
avx512f_test (void)
{
+ x1 = _mm_mask_min_ss (x1, m, x2, x3);
+ x1 = _mm_maskz_min_ss (m, x1, x2);
x1 = _mm_min_round_ss (x1, x2, _MM_FROUND_NO_EXC);
x1 = _mm_mask_min_round_ss (x1, m, x2, x3, _MM_FROUND_NO_EXC);
x1 = _mm_maskz_min_round_ss (m, x1, x2, _MM_FROUND_NO_EXC);
avx512f_test (void)
{
int i, sign;
- union128 res1, res2, res3, src1, src2;
+ union128 res1, res2, res3, res4, res5, src1, src2;
MASK_TYPE mask = MASK_VALUE;
float res_ref[SIZE];
{
src1.a[i] = 1.5 + 34.67 * i * sign;
src2.a[i] = -22.17 * i * sign + 1.0;
+ res1.a[i] = DEFAULT_VALUE;
+ res4.a[i] = DEFAULT_VALUE;
sign = sign * -1;
}
- for (i = 0; i < SIZE; i++)
- res2.a[i] = DEFAULT_VALUE;
- res1.x = _mm_min_round_ss (src1.x, src2.x, _MM_FROUND_NO_EXC);
- res2.x = _mm_mask_min_round_ss (res2.x, mask, src1.x, src2.x, _MM_FROUND_NO_EXC);
- res3.x = _mm_maskz_min_round_ss (mask, src1.x, src2.x, _MM_FROUND_NO_EXC);
+ res1.x = _mm_mask_min_ss (res1.x, mask, src1.x, src2.x);
+ res2.x = _mm_maskz_min_ss (mask, src1.x, src2.x);
+ res3.x = _mm_min_round_ss (src1.x, src2.x, _MM_FROUND_NO_EXC);
+ res4.x = _mm_mask_min_round_ss (res4.x, mask, src1.x, src2.x, _MM_FROUND_NO_EXC);
+ res5.x = _mm_maskz_min_round_ss (mask, src1.x, src2.x, _MM_FROUND_NO_EXC);
calc_min (res_ref, src1.a, src2.a);
-
+
+ MASK_MERGE () (res_ref, mask, 1);
if (check_union128 (res1, res_ref))
+ abort ();
+
+ MASK_ZERO () (res_ref, mask, 1);
+ if (check_union128 (res2, res_ref))
+ abort ();
+
+ calc_min (res_ref, src1.a, src2.a);
+
+ if (check_union128 (res3, res_ref))
abort();
MASK_MERGE () (res_ref, mask, 1);
- if (check_union128 (res2, res_ref))
+ if (check_union128 (res4, res_ref))
abort ();
MASK_ZERO () (res_ref, mask, 1);
- if (check_union128 (res3, res_ref))
+ if (check_union128 (res5, res_ref))
abort ();
}
projects / gcc.git / commitdiff