* [patch][i386, AVX] Adding missing mask[z]_sqrt_round_s[d,s] intrinsics
[not found] <DC8A343E1B673F4AA224650A1F844CCD26F41F@IRSMSX103.ger.corp.intel.com>
@ 2017-11-21 13:48 ` Makhotina, Olga
2018-02-12 5:46 ` Kirill Yukhin
0 siblings, 1 reply; 2+ messages in thread
From: Makhotina, Olga @ 2017-11-21 13:48 UTC (permalink / raw)
To: 'gcc-patches@gcc.gnu.org'
Cc: 'Kirill Yukhin', Makhotina, Olga, Peryt, Sebastian
[-- Attachment #1: Type: text/plain, Size: 2719 bytes --]
Hi,
This patch adds missing intrinsics for _mm_mask[z]_sqrt_round_[sd,ss].
21.11.2017 Olga Makhotina <olga.makhotina@intel.com>
gcc/
* config/i386/avx512fintrin.h (_mm_mask_sqrt_round_sd,
_mm_maskz_sqrt_round_sd, _mm_mask_sqrt_round_ss,
_mm_maskz_sqrt_round_ss): New intrinsics.
(__builtin_ia32_sqrtsd_round, __builtin_ia32_sqrtss_round): Remove.
(__builtin_ia32_sqrtsd_mask_round,
__builtin_ia32_sqrtss_mask_round): New builtins.
* config/i386/i386-builtin.def (__builtin_ia32_sqrtsd_round,
__builtin_ia32_sqrtss_round): Remove.
(__builtin_ia32_sqrtsd_mask_round,
__builtin_ia32_sqrtss_mask_round): New builtins.
* config/i386/sse.md (vmsqrt<mode>2<round_name>): Renamed to ...
(vmsqrt<mode>2<mask_scalar_name><round_scalar_name>): ... this.
((match_operand:VF_128 1 "vector_operand"
"xBm,<round_constraint>")): Changed to ...
((match_operand:VF_128 1 "vector_operand"
"xBm,<round_scalar_constraint>")): ... this.
(vsqrt<ssescalarmodesuffix>\t{<round_op3>%1, %2, %0|
%0, %2, %<iptr>1<round_op3>}): Changed to ...
(vsqrt<ssescalarmodesuffix>\t{<round_scalar_mask_op3>%1, %2,
%0<mask_scalar_operand3>|%0<mask_scalar_operand3>, %2,
%<iptr>1<round_scalar_mask_op3>}): ... this.
((set_attr "prefix" "<round_prefix>")): Changed to ...
((set_attr "prefix" "<round_scalar_prefix>")): ... this.
21.11.2017 Olga Makhotina <olga.makhotina@intel.com>
gcc/testsuite/
* gcc.target/i386/avx512f-vsqrtsd-1.c (_mm_mask_sqrt_round_sd,
_mm_maskz_sqrt_round_sd): Test new intrinsics.
* gcc.target/i386/avx512f-vsqrtsd-2.c (_mm_sqrt_round_sd,
_mm_mask_sqrt_round_sd, _mm_maskz_sqrt_round_sd): Test new intrinsics.
* gcc.target/i386/avx512f-vsqrtss-1.c (_mm_mask_sqrt_round_ss,
_mm_maskz_sqrt_round_ss): Test new intrinsics.
* gcc.target/i386/avx512f-vsqrtss-2.c (_mm_sqrt_round_ss,
_mm_mask_sqrt_round_ss, _mm_maskz_sqrt_round_ss): Test new intrinsics.
* gcc.target/i386/avx-1.c (__builtin_ia32_sqrtsd_round,
__builtin_ia32_sqrtss_round): Remove builtins.
(__builtin_ia32_sqrtsd_mask_round,
__builtin_ia32_sqrtss_mask_round): Test new builtins.
* gcc.target/i386/sse-13.c: Ditto.
* gcc.target/i386/sse-23.c: Ditto.
Is it ok for trunk?
Thanks,
Olga
[-- Attachment #2: 0001-sqrt.patch --]
[-- Type: application/octet-stream, Size: 17307 bytes --]
From 77d2762f269acd976e9e96a19d78b9083b6ddfa2 Mon Sep 17 00:00:00 2001
From: Olga Makhotina <olga.makhotina@intel.com>
Date: Mon, 20 Nov 2017 17:08:02 +0300
Subject: [PATCH] sqrt
---
gcc/config/i386/avx512fintrin.h | 84 ++++++++++++++++++++---
gcc/config/i386/i386-builtin.def | 4 +-
gcc/config/i386/sse.md | 8 +--
gcc/testsuite/gcc.target/i386/avx-1.c | 4 +-
gcc/testsuite/gcc.target/i386/avx512f-vsqrtsd-1.c | 5 ++
gcc/testsuite/gcc.target/i386/avx512f-vsqrtsd-2.c | 62 +++++++++++++++++
gcc/testsuite/gcc.target/i386/avx512f-vsqrtss-1.c | 6 ++
gcc/testsuite/gcc.target/i386/avx512f-vsqrtss-2.c | 63 +++++++++++++++++
gcc/testsuite/gcc.target/i386/sse-13.c | 4 +-
gcc/testsuite/gcc.target/i386/sse-23.c | 4 +-
10 files changed, 222 insertions(+), 22 deletions(-)
create mode 100644 gcc/testsuite/gcc.target/i386/avx512f-vsqrtsd-2.c
create mode 100644 gcc/testsuite/gcc.target/i386/avx512f-vsqrtss-2.c
diff --git a/gcc/config/i386/avx512fintrin.h b/gcc/config/i386/avx512fintrin.h
index 5dc5fae..1bd35bf 100644
--- a/gcc/config/i386/avx512fintrin.h
+++ b/gcc/config/i386/avx512fintrin.h
@@ -1955,18 +1955,66 @@ extern __inline __m128d
__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
_mm_sqrt_round_sd (__m128d __A, __m128d __B, const int __R)
{
- return (__m128d) __builtin_ia32_sqrtsd_round ((__v2df) __B,
- (__v2df) __A,
- __R);
+ return (__m128d) __builtin_ia32_sqrtsd_mask_round ((__v2df) __B,
+ (__v2df) __A,
+ (__v2df)
+ _mm_setzero_pd (),
+ (__mmask8) -1, __R);
+}
+
+extern __inline __m128d
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_mask_sqrt_round_sd (__m128d __W, __mmask8 __U, __m128d __A, __m128d __B,
+ const int __R)
+{
+ return (__m128d) __builtin_ia32_sqrtsd_mask_round ((__v2df) __B,
+ (__v2df) __A,
+ (__v2df) __W,
+ (__mmask8) __U, __R);
+}
+
+extern __inline __m128d
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_maskz_sqrt_round_sd (__mmask8 __U, __m128d __A, __m128d __B, const int __R)
+{
+ return (__m128d) __builtin_ia32_sqrtsd_mask_round ((__v2df) __B,
+ (__v2df) __A,
+ (__v2df)
+ _mm_setzero_pd (),
+ (__mmask8) __U, __R);
}
extern __inline __m128
__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
_mm_sqrt_round_ss (__m128 __A, __m128 __B, const int __R)
{
- return (__m128) __builtin_ia32_sqrtss_round ((__v4sf) __B,
- (__v4sf) __A,
- __R);
+ return (__m128) __builtin_ia32_sqrtss_mask_round ((__v4sf) __B,
+ (__v4sf) __A,
+ (__v4sf)
+ _mm_setzero_ps (),
+ (__mmask8) -1, __R);
+}
+
+extern __inline __m128
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_mask_sqrt_round_ss (__m128 __W, __mmask8 __U, __m128 __A, __m128 __B,
+ const int __R)
+{
+ return (__m128) __builtin_ia32_sqrtss_mask_round ((__v4sf) __B,
+ (__v4sf) __A,
+ (__v4sf) __W,
+ (__mmask8) __U, __R);
+}
+
+extern __inline __m128
+__attribute__ ((__gnu_inline__, __always_inline__, __artificial__))
+_mm_maskz_sqrt_round_ss (__mmask8 __U, __m128 __A, __m128 __B, const int __R)
+{
+ return (__m128) __builtin_ia32_sqrtss_mask_round ((__v4sf) __B,
+ (__v4sf) __A,
+ (__v4sf)
+ _mm_setzero_ps (),
+ (__mmask8) __U, __R);
}
#else
#define _mm512_sqrt_round_pd(A, C) \
@@ -1987,11 +2035,27 @@ _mm_sqrt_round_ss (__m128 __A, __m128 __B, const int __R)
#define _mm512_maskz_sqrt_round_ps(U, A, C) \
(__m512)__builtin_ia32_sqrtps512_mask(A, (__v16sf)_mm512_setzero_ps(), U, C)
-#define _mm_sqrt_round_sd(A, B, C) \
- (__m128d)__builtin_ia32_sqrtsd_round(A, B, C)
+#define _mm_sqrt_round_sd(A, B, C) \
+ (__m128d)__builtin_ia32_sqrtsd_mask_round (B, A, \
+ (__v2df) _mm_setzero_pd (), -1, C)
+
+#define _mm_mask_sqrt_round_sd(W, U, A, B, C) \
+ (__m128d)__builtin_ia32_sqrtsd_mask_round (B, A, W, U, C)
+
+#define _mm_maskz_sqrt_round_sd(U, A, B, C) \
+ (__m128d)__builtin_ia32_sqrtsd_mask_round (B, A, \
+ (__v2df) _mm_setzero_pd (), U, C)
+
+#define _mm_sqrt_round_ss(A, B, C) \
+ (__m128)__builtin_ia32_sqrtss_mask_round (B, A, \
+ (__v4sf) _mm_setzero_ps (), -1, C)
+
+#define _mm_mask_sqrt_round_ss(W, U, A, B, C) \
+ (__m128)__builtin_ia32_sqrtss_mask_round (B, A, W, U, C)
-#define _mm_sqrt_round_ss(A, B, C) \
- (__m128)__builtin_ia32_sqrtss_round(A, B, C)
+#define _mm_maskz_sqrt_round_ss(U, A, B, C) \
+ (__m128)__builtin_ia32_sqrtss_mask_round (B, A, \
+ (__v4sf) _mm_setzero_ps (), U, C)
#endif
extern __inline __m512i
diff --git a/gcc/config/i386/i386-builtin.def b/gcc/config/i386/i386-builtin.def
index e46a6ab..f7eb6e7 100644
--- a/gcc/config/i386/i386-builtin.def
+++ b/gcc/config/i386/i386-builtin.def
@@ -2500,8 +2500,8 @@ BDESC (OPTION_MASK_ISA_AVX512F, CODE_FOR_avx512f_vmscalefv2df_round, "__builtin_
BDESC (OPTION_MASK_ISA_AVX512F, CODE_FOR_avx512f_vmscalefv4sf_round, "__builtin_ia32_scalefss_round", IX86_BUILTIN_SCALEFSS, UNKNOWN, (int) V4SF_FTYPE_V4SF_V4SF_INT)
BDESC (OPTION_MASK_ISA_AVX512F, CODE_FOR_avx512f_sqrtv8df2_mask_round, "__builtin_ia32_sqrtpd512_mask", IX86_BUILTIN_SQRTPD512_MASK, UNKNOWN, (int) V8DF_FTYPE_V8DF_V8DF_QI_INT)
BDESC (OPTION_MASK_ISA_AVX512F, CODE_FOR_avx512f_sqrtv16sf2_mask_round, "__builtin_ia32_sqrtps512_mask", IX86_BUILTIN_SQRTPS512_MASK, UNKNOWN, (int) V16SF_FTYPE_V16SF_V16SF_HI_INT)
-BDESC (OPTION_MASK_ISA_AVX512F, CODE_FOR_sse2_vmsqrtv2df2_round, "__builtin_ia32_sqrtsd_round", IX86_BUILTIN_SQRTSD_ROUND, UNKNOWN, (int) V2DF_FTYPE_V2DF_V2DF_INT)
-BDESC (OPTION_MASK_ISA_AVX512F, CODE_FOR_sse_vmsqrtv4sf2_round, "__builtin_ia32_sqrtss_round", IX86_BUILTIN_SQRTSS_ROUND, UNKNOWN, (int) V4SF_FTYPE_V4SF_V4SF_INT)
+BDESC (OPTION_MASK_ISA_AVX512F, CODE_FOR_sse2_vmsqrtv2df2_mask_round, "__builtin_ia32_sqrtsd_mask_round", IX86_BUILTIN_SQRTSD_ROUND, UNKNOWN, (int) V2DF_FTYPE_V2DF_V2DF_V2DF_UQI_INT)
+BDESC (OPTION_MASK_ISA_AVX512F, CODE_FOR_sse_vmsqrtv4sf2_mask_round, "__builtin_ia32_sqrtss_mask_round", IX86_BUILTIN_SQRTSS_ROUND, UNKNOWN, (int) V4SF_FTYPE_V4SF_V4SF_V4SF_UQI_INT)
BDESC (OPTION_MASK_ISA_AVX512F, CODE_FOR_subv8df3_mask_round, "__builtin_ia32_subpd512_mask", IX86_BUILTIN_SUBPD512, UNKNOWN, (int) V8DF_FTYPE_V8DF_V8DF_V8DF_UQI_INT)
BDESC (OPTION_MASK_ISA_AVX512F, CODE_FOR_subv16sf3_mask_round, "__builtin_ia32_subps512_mask", IX86_BUILTIN_SUBPS512, UNKNOWN, (int) V16SF_FTYPE_V16SF_V16SF_V16SF_HI_INT)
BDESC (OPTION_MASK_ISA_AVX512F, CODE_FOR_sse2_vmsubv2df3_round, "__builtin_ia32_subsd_round", IX86_BUILTIN_SUBSD_ROUND, UNKNOWN, (int) V2DF_FTYPE_V2DF_V2DF_INT)
diff --git a/gcc/config/i386/sse.md b/gcc/config/i386/sse.md
index 7f17231..5d2b83c 100644
--- a/gcc/config/i386/sse.md
+++ b/gcc/config/i386/sse.md
@@ -1820,21 +1820,21 @@
(set_attr "prefix" "maybe_vex")
(set_attr "mode" "<MODE>")])
-(define_insn "<sse>_vmsqrt<mode>2<round_name>"
+(define_insn "<sse>_vmsqrt<mode>2<mask_scalar_name><round_scalar_name>"
[(set (match_operand:VF_128 0 "register_operand" "=x,v")
(vec_merge:VF_128
(sqrt:VF_128
- (match_operand:VF_128 1 "vector_operand" "xBm,<round_constraint>"))
+ (match_operand:VF_128 1 "vector_operand" "xBm,<round_scalar_constraint>"))
(match_operand:VF_128 2 "register_operand" "0,v")
(const_int 1)))]
"TARGET_SSE"
"@
sqrt<ssescalarmodesuffix>\t{%1, %0|%0, %<iptr>1}
- vsqrt<ssescalarmodesuffix>\t{<round_op3>%1, %2, %0|%0, %2, %<iptr>1<round_op3>}"
+ vsqrt<ssescalarmodesuffix>\t{<round_scalar_mask_op3>%1, %2, %0<mask_scalar_operand3>|%0<mask_scalar_operand3>, %2, %<iptr>1<round_scalar_mask_op3>}"
[(set_attr "isa" "noavx,avx")
(set_attr "type" "sse")
(set_attr "atom_sse_attr" "sqrt")
- (set_attr "prefix" "<round_prefix>")
+ (set_attr "prefix" "<round_scalar_prefix>")
(set_attr "btver2_sse_attr" "sqrt")
(set_attr "mode" "<ssescalarmode>")])
diff --git a/gcc/testsuite/gcc.target/i386/avx-1.c b/gcc/testsuite/gcc.target/i386/avx-1.c
index 1133a83..c27e482 100644
--- a/gcc/testsuite/gcc.target/i386/avx-1.c
+++ b/gcc/testsuite/gcc.target/i386/avx-1.c
@@ -305,8 +305,8 @@
#define __builtin_ia32_shufps512_mask(A, B, F, D, E) __builtin_ia32_shufps512_mask(A, B, 1, D, E)
#define __builtin_ia32_sqrtpd512_mask(A, B, C, D) __builtin_ia32_sqrtpd512_mask(A, B, C, 8)
#define __builtin_ia32_sqrtps512_mask(A, B, C, D) __builtin_ia32_sqrtps512_mask(A, B, C, 8)
-#define __builtin_ia32_sqrtss_round(A, B, C) __builtin_ia32_sqrtss_round(A, B, 8)
-#define __builtin_ia32_sqrtsd_round(A, B, C) __builtin_ia32_sqrtsd_round(A, B, 8)
+#define __builtin_ia32_sqrtss_mask_round(A, B, C, D, E) __builtin_ia32_sqrtss_mask_round(A, B, C, D, 8)
+#define __builtin_ia32_sqrtsd_mask_round(A, B, C, D, E) __builtin_ia32_sqrtsd_mask_round(A, B, C, D, 8)
#define __builtin_ia32_subpd512_mask(A, B, C, D, E) __builtin_ia32_subpd512_mask(A, B, C, D, 8)
#define __builtin_ia32_subps512_mask(A, B, C, D, E) __builtin_ia32_subps512_mask(A, B, C, D, 8)
#define __builtin_ia32_subsd_round(A, B, C) __builtin_ia32_subsd_round(A, B, 8)
diff --git a/gcc/testsuite/gcc.target/i386/avx512f-vsqrtsd-1.c b/gcc/testsuite/gcc.target/i386/avx512f-vsqrtsd-1.c
index c0559c0..a7d7af9 100644
--- a/gcc/testsuite/gcc.target/i386/avx512f-vsqrtsd-1.c
+++ b/gcc/testsuite/gcc.target/i386/avx512f-vsqrtsd-1.c
@@ -1,13 +1,18 @@
/* { dg-do compile } */
/* { dg-options "-mavx512f -O2" } */
/* { dg-final { scan-assembler-times "vsqrtsd\[ \\t\]+\[^\n\]*\{rn-sae\}\[^\{\n\]*%xmm\[0-9\]+(?:\n|\[ \\t\]+#)" 1 } } */
+/* { dg-final { scan-assembler-times "vsqrtsd\[ \\t\]+\[^\n\]*\{rd-sae\}\[^\{\n\]*%xmm\[0-9\]+\[^\n\]*%xmm\[0-9\]+\[^\n\]*%xmm\[0-9\]+\{%k\[1-7\]\}(?:\n|\[ \\t\]+#)" 1 } } */
+/* { dg-final { scan-assembler-times "vsqrtsd\[ \\t\]+\[^\n\]*\{rz-sae\}\[^\{\n\]*%xmm\[0-9\]+\[^\n\]*%xmm\[0-9\]+\[^\n\]*%xmm\[0-9\]+\{%k\[1-7\]\}\{z\}(?:\n|\[ \\t\]+#)" 1 } } */
#include <immintrin.h>
volatile __m128d x1, x2;
+volatile __mmask8 m;
void extern
avx512f_test (void)
{
x1 = _mm_sqrt_round_sd (x1, x2, _MM_FROUND_TO_NEAREST_INT | _MM_FROUND_NO_EXC);
+ x1 = _mm_mask_sqrt_round_sd (x1, m, x1, x2, _MM_FROUND_TO_NEG_INF | _MM_FROUND_NO_EXC);
+ x1 = _mm_maskz_sqrt_round_sd (m, x1, x2, _MM_FROUND_TO_ZERO | _MM_FROUND_NO_EXC);
}
diff --git a/gcc/testsuite/gcc.target/i386/avx512f-vsqrtsd-2.c b/gcc/testsuite/gcc.target/i386/avx512f-vsqrtsd-2.c
new file mode 100644
index 0000000..49ca7ee
--- /dev/null
+++ b/gcc/testsuite/gcc.target/i386/avx512f-vsqrtsd-2.c
@@ -0,0 +1,62 @@
+/* { dg-do run } */
+/* { dg-options "-mavx512f -O2" } */
+/* { dg-require-effective-target avx512f } */
+
+#include <math.h>
+#include "avx512f-check.h"
+
+#define SIZE (128 / 64)
+#include "avx512f-mask-type.h"
+
+static void
+compute_sqrtsd (double *s1, double *s2, double *r)
+{
+ r[0] = sqrt(s2[0]);
+ r[1] = s1[1];
+}
+
+void static
+avx512f_test (void)
+{
+ union128d res1, res2, res3;
+ union128d s1, s2;
+ double res_ref[SIZE];
+ MASK_TYPE mask = MASK_VALUE;
+ int i;
+
+ for (i = 0; i < SIZE; i++)
+ {
+ s1.a[i] = 11.5 * (i + 1);
+ s2.a[i] = 10.5 * (i + 1);
+ res_ref[i] = 9.5 * (i + 1);
+ res1.a[i] = DEFAULT_VALUE;
+ res2.a[i] = DEFAULT_VALUE;
+ res3.a[i] = DEFAULT_VALUE;
+ }
+
+ res1.x = _mm_sqrt_round_sd (s1.x, s2.x,
+ _MM_FROUND_TO_NEAREST_INT | _MM_FROUND_NO_EXC);
+ res2.x = _mm_mask_sqrt_round_sd (s1.x, mask, s1.x, s2.x,
+ _MM_FROUND_TO_NEAREST_INT | _MM_FROUND_NO_EXC);
+ res3.x = _mm_maskz_sqrt_round_sd (mask, s1.x, s2.x,
+ _MM_FROUND_TO_NEAREST_INT | _MM_FROUND_NO_EXC);
+
+ compute_sqrtsd (s1.a, s2.a, res_ref);
+
+ if (check_union128d (res1, res_ref))
+ abort ();
+
+ MASK_MERGE (d) (res_ref, mask, 1);
+
+ if (check_union128d (res2, res_ref))
+ abort ();
+
+ MASK_ZERO (d) (res_ref, mask, 1);
+
+ if (check_union128d (res3, res_ref))
+ abort ();
+}
+
+
+
+
diff --git a/gcc/testsuite/gcc.target/i386/avx512f-vsqrtss-1.c b/gcc/testsuite/gcc.target/i386/avx512f-vsqrtss-1.c
index e43b4a1..103ff30 100644
--- a/gcc/testsuite/gcc.target/i386/avx512f-vsqrtss-1.c
+++ b/gcc/testsuite/gcc.target/i386/avx512f-vsqrtss-1.c
@@ -1,13 +1,19 @@
/* { dg-do compile } */
/* { dg-options "-mavx512f -O2" } */
/* { dg-final { scan-assembler-times "vsqrtss\[ \\t\]+\[^\n\]*\{rn-sae\}\[^\{\n\]*%xmm\[0-9\]+(?:\n|\[ \\t\]+#)" 1 } } */
+/* { dg-final { scan-assembler-times "vsqrtss\[ \\t\]+\[^\n\]*\{rd-sae\}\[^\{\n\]*%xmm\[0-9\]+\[^\n\]*%xmm\[0-9\]+\[^\n\]*%xmm\[0-9\]+\{%k\[1-7\]\}(?:\n|\[ \\t\]+#)" 1 } } */
+/* { dg-final { scan-assembler-times "vsqrtss\[ \\t\]+\[^\n\]*\{rz-sae\}\[^\{\n\]*%xmm\[0-9\]+\[^\n\]*%xmm\[0-9\]+\[^\n\]*%xmm\[0-9\]+\{%k\[1-7\]\}\{z\}(?:\n|\[ \\t\]+#)" 1 } } */
+
#include <immintrin.h>
volatile __m128 x1, x2;
+volatile __mmask8 m;
void extern
avx512f_test (void)
{
x1 = _mm_sqrt_round_ss (x1, x2, _MM_FROUND_TO_NEAREST_INT | _MM_FROUND_NO_EXC);
+ x1 = _mm_mask_sqrt_round_ss (x1, m, x1, x2, _MM_FROUND_TO_NEG_INF | _MM_FROUND_NO_EXC);
+ x1 = _mm_maskz_sqrt_round_ss (m, x1, x2, _MM_FROUND_TO_ZERO | _MM_FROUND_NO_EXC);
}
diff --git a/gcc/testsuite/gcc.target/i386/avx512f-vsqrtss-2.c b/gcc/testsuite/gcc.target/i386/avx512f-vsqrtss-2.c
new file mode 100644
index 0000000..90f88be
--- /dev/null
+++ b/gcc/testsuite/gcc.target/i386/avx512f-vsqrtss-2.c
@@ -0,0 +1,63 @@
+/* { dg-do run } */
+/* { dg-options "-mavx512f -O2" } */
+/* { dg-require-effective-target avx512f } */
+
+#include <math.h>
+#include "avx512f-check.h"
+
+#define SIZE (128 / 32)
+#include "avx512f-mask-type.h"
+
+static void
+compute_sqrtss (float *s1, float *s2, float *r)
+{
+ r[0] = sqrt(s2[0]);
+ int i;
+ for (i = 1; i < SIZE; i++)
+ {
+ r[i] = s1[i];
+ }
+}
+
+static void
+avx512f_test (void)
+{
+ union128 res1, res2, res3;
+ union128 s1, s2;
+ float res_ref[SIZE];
+ MASK_TYPE mask = MASK_VALUE;
+ int i;
+
+ for (i = 0; i < SIZE; i++)
+ {
+ s1.a[i] = 11.5 * (i + 1);
+ s2.a[i] = 10.5 * (i + 1);
+ res_ref[i] = 9.5 * (i + 1);
+ res1.a[i] = DEFAULT_VALUE;
+ res2.a[i] = DEFAULT_VALUE;
+ res3.a[i] = DEFAULT_VALUE;
+ }
+
+ res1.x = _mm_sqrt_round_ss (s1.x, s2.x,
+ _MM_FROUND_TO_NEAREST_INT | _MM_FROUND_NO_EXC);
+ res2.x = _mm_mask_sqrt_round_ss (s1.x, mask, s1.x, s2.x,
+ _MM_FROUND_TO_NEAREST_INT | _MM_FROUND_NO_EXC);
+ res3.x = _mm_maskz_sqrt_round_ss (mask, s1.x, s2.x,
+ _MM_FROUND_TO_NEAREST_INT | _MM_FROUND_NO_EXC);
+
+ compute_sqrtss (s1.a, s2.a, res_ref);
+
+ if (check_union128 (res1, res_ref))
+ abort ();
+
+ MASK_MERGE () (res_ref, mask, 1);
+
+ if (check_union128 (res2, res_ref))
+ abort ();
+
+ MASK_ZERO () (res_ref, mask, 1);
+
+ if (check_union128 (res3, res_ref))
+ abort ();
+}
+
diff --git a/gcc/testsuite/gcc.target/i386/sse-13.c b/gcc/testsuite/gcc.target/i386/sse-13.c
index 9bdc73f..3740bc0 100644
--- a/gcc/testsuite/gcc.target/i386/sse-13.c
+++ b/gcc/testsuite/gcc.target/i386/sse-13.c
@@ -322,8 +322,8 @@
#define __builtin_ia32_shufps512_mask(A, B, F, D, E) __builtin_ia32_shufps512_mask(A, B, 1, D, E)
#define __builtin_ia32_sqrtpd512_mask(A, B, C, D) __builtin_ia32_sqrtpd512_mask(A, B, C, 8)
#define __builtin_ia32_sqrtps512_mask(A, B, C, D) __builtin_ia32_sqrtps512_mask(A, B, C, 8)
-#define __builtin_ia32_sqrtss_round(A, B, C) __builtin_ia32_sqrtss_round(A, B, 8)
-#define __builtin_ia32_sqrtsd_round(A, B, C) __builtin_ia32_sqrtsd_round(A, B, 8)
+#define __builtin_ia32_sqrtss_mask_round(A, B, C, D, E) __builtin_ia32_sqrtss_mask_round(A, B, C, E, 8)
+#define __builtin_ia32_sqrtsd_mask_round(A, B, C, D, E) __builtin_ia32_sqrtsd_mask_round(A, B, C, E, 8)
#define __builtin_ia32_subpd512_mask(A, B, C, D, E) __builtin_ia32_subpd512_mask(A, B, C, D, 8)
#define __builtin_ia32_subps512_mask(A, B, C, D, E) __builtin_ia32_subps512_mask(A, B, C, D, 8)
#define __builtin_ia32_subsd_round(A, B, C) __builtin_ia32_subsd_round(A, B, 8)
diff --git a/gcc/testsuite/gcc.target/i386/sse-23.c b/gcc/testsuite/gcc.target/i386/sse-23.c
index 66c25c7..36116ae 100644
--- a/gcc/testsuite/gcc.target/i386/sse-23.c
+++ b/gcc/testsuite/gcc.target/i386/sse-23.c
@@ -323,8 +323,8 @@
#define __builtin_ia32_shufps512_mask(A, B, F, D, E) __builtin_ia32_shufps512_mask(A, B, 1, D, E)
#define __builtin_ia32_sqrtpd512_mask(A, B, C, D) __builtin_ia32_sqrtpd512_mask(A, B, C, 8)
#define __builtin_ia32_sqrtps512_mask(A, B, C, D) __builtin_ia32_sqrtps512_mask(A, B, C, 8)
-#define __builtin_ia32_sqrtss_round(A, B, C) __builtin_ia32_sqrtss_round(A, B, 8)
-#define __builtin_ia32_sqrtsd_round(A, B, C) __builtin_ia32_sqrtsd_round(A, B, 8)
+#define __builtin_ia32_sqrtss_mask_round(A, B, C, D, E) __builtin_ia32_sqrtss_mask_round(A, B, C, D, 8)
+#define __builtin_ia32_sqrtsd_mask_round(A, B, C, D, E) __builtin_ia32_sqrtsd_mask_round(A, B, C, D, 8)
#define __builtin_ia32_subpd512_mask(A, B, C, D, E) __builtin_ia32_subpd512_mask(A, B, C, D, 8)
#define __builtin_ia32_subps512_mask(A, B, C, D, E) __builtin_ia32_subps512_mask(A, B, C, D, 8)
#define __builtin_ia32_subsd_round(A, B, C) __builtin_ia32_subsd_round(A, B, 8)
--
2.5.5
^ permalink raw reply [flat|nested] 2+ messages in thread