public inbox for gcc-patches@gcc.gnu.org
 help / color / mirror / Atom feed
From: Hongyu Wang <wwwhhhyyy333@gmail.com>
To: Hongtao Liu <crazylht@gmail.com>
Cc: Hongyu Wang <hongyu.wang@intel.com>,
	"Liu, Hongtao" <hongtao.liu@intel.com>,
	GCC Patches <gcc-patches@gcc.gnu.org>
Subject: Re: [PATCH] AVX512F: Add missing macro for mask(z?)_scalf_s[sd] [PR 105339]
Date: Fri, 22 Apr 2022 20:38:18 +0800	[thread overview]
Message-ID: <CA+OydWnytyszAZevYtfNYOxvcR202xk+Pduo9+btNvpmZ_=5=A@mail.gmail.com> (raw)
In-Reply-To: <CAMZc-by1Ss6XtDB6w9PReJfRYvhB8ybmHh4so+p9yoqso4TPyw@mail.gmail.com>

[-- Attachment #1: Type: text/plain, Size: 7068 bytes --]

> Please add the corresponding intrinsic test in sse-14.c

Sorry for forgetting this part. Updated patch. Thanks.

Hongtao Liu via Gcc-patches <gcc-patches@gcc.gnu.org> 于2022年4月22日周五 16:49写道:
>
> On Fri, Apr 22, 2022 at 4:12 PM Hongyu Wang via Gcc-patches
> <gcc-patches@gcc.gnu.org> wrote:
> >
> > Hi,
> >
> > Add missing macro under O0 and adjust macro format for scalf
> > intrinsics.
> >
> Please add the corresponding intrinsic test in sse-14.c.
> > Bootstrapped/regtested on x86_64-pc-linux-gnu{-m32,}.
> >
> > Ok for master and backport to GCC 9/10/11?
> >
> > gcc/ChangeLog:
> >
> >         PR target/105339
> >         * config/i386/avx512fintrin.h (_mm512_scalef_round_pd):
> >         Add parentheses for parameters and djust format.
> >         (_mm512_mask_scalef_round_pd): Ditto.
> >         (_mm512_maskz_scalef_round_pd): Ditto.
> >         (_mm512_scalef_round_ps): Ditto.
> >         (_mm512_mask_scalef_round_ps): Ditto.
> >         (_mm512_maskz_scalef_round_ps): Ditto.
> >         (_mm_scalef_round_sd): Use _mm_undefined_pd.
> >         (_mm_scalef_round_ss): Use _mm_undefined_ps.
> >         (_mm_mask_scalef_round_sd): New macro.
> >         (_mm_mask_scalef_round_ss): Ditto.
> >         (_mm_maskz_scalef_round_sd): Ditto.
> >         (_mm_maskz_scalef_round_ss): Ditto.
> > ---
> >  gcc/config/i386/avx512fintrin.h | 76 ++++++++++++++++++++++++---------
> >  1 file changed, 56 insertions(+), 20 deletions(-)
> >
> > diff --git a/gcc/config/i386/avx512fintrin.h b/gcc/config/i386/avx512fintrin.h
> > index 29511fd2831..6dc69ff0234 100644
> > --- a/gcc/config/i386/avx512fintrin.h
> > +++ b/gcc/config/i386/avx512fintrin.h
> > @@ -3286,31 +3286,67 @@ _mm_maskz_scalef_round_ss (__mmask8 __U, __m128 __A, __m128 __B, const int __R)
> >                                                       (__mmask8) __U, __R);
> >  }
> >  #else
> > -#define _mm512_scalef_round_pd(A, B, C)            \
> > -    (__m512d)__builtin_ia32_scalefpd512_mask(A, B, (__v8df)_mm512_undefined_pd(), -1, C)
> > -
> > -#define _mm512_mask_scalef_round_pd(W, U, A, B, C) \
> > -    (__m512d)__builtin_ia32_scalefpd512_mask(A, B, W, U, C)
> > -
> > -#define _mm512_maskz_scalef_round_pd(U, A, B, C)   \
> > -    (__m512d)__builtin_ia32_scalefpd512_mask(A, B, (__v8df)_mm512_setzero_pd(), U, C)
> > +#define _mm512_scalef_round_pd(A, B, C)                                        \
> > +  ((__m512d)                                                           \
> > +   __builtin_ia32_scalefpd512_mask((A), (B),                           \
> > +                                  (__v8df) _mm512_undefined_pd(),      \
> > +                                  -1, (C)))
> > +
> > +#define _mm512_mask_scalef_round_pd(W, U, A, B, C)                     \
> > +  ((__m512d) __builtin_ia32_scalefpd512_mask((A), (B), (W), (U), (C)))
> > +
> > +#define _mm512_maskz_scalef_round_pd(U, A, B, C)                       \
> > +  ((__m512d)                                                           \
> > +   __builtin_ia32_scalefpd512_mask((A), (B),                           \
> > +                                  (__v8df) _mm512_setzero_pd(),        \
> > +                                  (U), (C)))
> > +
> > +#define _mm512_scalef_round_ps(A, B, C)                                        \
> > +  ((__m512)                                                            \
> > +   __builtin_ia32_scalefps512_mask((A), (B),                           \
> > +                                  (__v16sf) _mm512_undefined_ps(),     \
> > +                                  -1, (C)))
> > +
> > +#define _mm512_mask_scalef_round_ps(W, U, A, B, C)                     \
> > +  ((__m512) __builtin_ia32_scalefps512_mask((A), (B), (W), (U), (C)))
> > +
> > +#define _mm512_maskz_scalef_round_ps(U, A, B, C)                       \
> > +  ((__m512)                                                            \
> > +   __builtin_ia32_scalefps512_mask((A), (B),                           \
> > +                                  (__v16sf) _mm512_setzero_ps(),       \
> > +                                  (U), (C)))
> > +
> > +#define _mm_scalef_round_sd(A, B, C)                                   \
> > +  ((__m128d)                                                           \
> > +   __builtin_ia32_scalefsd_mask_round ((A), (B),                       \
> > +                                      (__v2df) _mm_undefined_pd (),    \
> > +                                      -1, (C)))
> >
> > -#define _mm512_scalef_round_ps(A, B, C)            \
> > -    (__m512)__builtin_ia32_scalefps512_mask(A, B, (__v16sf)_mm512_undefined_ps(), -1, C)
> > +#define _mm_scalef_round_ss(A, B, C)                                   \
> > +  ((__m128)                                                            \
> > +   __builtin_ia32_scalefss_mask_round ((A), (B),                       \
> > +                                      (__v4sf) _mm_undefined_ps (),    \
> > +                                      -1, (C)))
> >
> > -#define _mm512_mask_scalef_round_ps(W, U, A, B, C) \
> > -    (__m512)__builtin_ia32_scalefps512_mask(A, B, W, U, C)
> > +#define _mm_mask_scalef_round_sd(W, U, A, B, C)                                \
> > +  ((__m128d)                                                           \
> > +   __builtin_ia32_scalefsd_mask_round ((A), (B), (W), (U), (C)))
> >
> > -#define _mm512_maskz_scalef_round_ps(U, A, B, C)   \
> > -    (__m512)__builtin_ia32_scalefps512_mask(A, B, (__v16sf)_mm512_setzero_ps(), U, C)
> > +#define _mm_mask_scalef_round_ss(W, U, A, B, C)                                \
> > +  ((__m128)                                                            \
> > +   __builtin_ia32_scalefss_mask_round ((A), (B), (W), (U), (C)))
> >
> > -#define _mm_scalef_round_sd(A, B, C)            \
> > -    (__m128d)__builtin_ia32_scalefsd_mask_round (A, B, \
> > -       (__v2df)_mm_setzero_pd (), -1, C)
> > +#define _mm_maskz_scalef_round_sd(U, A, B, C)                          \
> > +  ((__m128d)                                                           \
> > +   __builtin_ia32_scalefsd_mask_round ((A), (B),                       \
> > +                                      (__v2df) _mm_setzero_pd (),      \
> > +                                      (U), (C)))
> >
> > -#define _mm_scalef_round_ss(A, B, C)            \
> > -    (__m128)__builtin_ia32_scalefss_mask_round (A, B, \
> > -       (__v4sf)_mm_setzero_ps (), -1, C)
> > +#define _mm_maskz_scalef_round_ss(U, A, B, C)                          \
> > +  ((__m128)                                                            \
> > +   __builtin_ia32_scalefss_mask_round ((A), (B),                       \
> > +                                      (__v4sf) _mm_setzero_ps (),      \
> > +                                      (W), (U), (C)))
> >  #endif
> >
> >  #define _mm_mask_scalef_sd(W, U, A, B) \
> > --
> > 2.18.1
> >
>
>
> --
> BR,
> Hongtao

[-- Attachment #2: 0001-AVX512F-Add-missing-macro-for-mask-z-_scalf_s-sd-PR-.patch --]
[-- Type: text/x-patch, Size: 6553 bytes --]

From 93be6b95b3237e1bff61001031ed6ad733c5de3f Mon Sep 17 00:00:00 2001
From: Hongyu Wang <hongyu.wang@intel.com>
Date: Fri, 22 Apr 2022 14:42:30 +0800
Subject: [PATCH] AVX512F: Add missing macro for mask(z?)_scalf_s[sd] [PR
 105339]

Add missing macro under O0 and adjust macro format for scalf
intrinsics.

gcc/ChangeLog:

	PR target/105339
	* config/i386/avx512fintrin.h (_mm512_scalef_round_pd):
	Add parentheses for parameters and djust format.
	(_mm512_mask_scalef_round_pd): Ditto.
	(_mm512_maskz_scalef_round_pd): Ditto.
	(_mm512_scalef_round_ps): Ditto.
	(_mm512_mask_scalef_round_ps): Ditto.
	(_mm512_maskz_scalef_round_ps): Ditto.
	(_mm_scalef_round_sd): Use _mm_undefined_pd.
	(_mm_scalef_round_ss): Use _mm_undefined_ps.
	(_mm_mask_scalef_round_sd): New macro.
	(_mm_mask_scalef_round_ss): Ditto.
	(_mm_maskz_scalef_round_sd): Ditto.
	(_mm_maskz_scalef_round_ss): Ditto.

gcc/testsuite/ChangeLog:

	PR target/105339
	* gcc.target/i386/sse-14.c: Add tests for new macro.
---
 gcc/config/i386/avx512fintrin.h        | 76 +++++++++++++++++++-------
 gcc/testsuite/gcc.target/i386/sse-14.c |  4 ++
 2 files changed, 60 insertions(+), 20 deletions(-)

diff --git a/gcc/config/i386/avx512fintrin.h b/gcc/config/i386/avx512fintrin.h
index 29511fd2831..77d6249c2bc 100644
--- a/gcc/config/i386/avx512fintrin.h
+++ b/gcc/config/i386/avx512fintrin.h
@@ -3286,31 +3286,67 @@ _mm_maskz_scalef_round_ss (__mmask8 __U, __m128 __A, __m128 __B, const int __R)
 						      (__mmask8) __U, __R);
 }
 #else
-#define _mm512_scalef_round_pd(A, B, C)            \
-    (__m512d)__builtin_ia32_scalefpd512_mask(A, B, (__v8df)_mm512_undefined_pd(), -1, C)
-
-#define _mm512_mask_scalef_round_pd(W, U, A, B, C) \
-    (__m512d)__builtin_ia32_scalefpd512_mask(A, B, W, U, C)
-
-#define _mm512_maskz_scalef_round_pd(U, A, B, C)   \
-    (__m512d)__builtin_ia32_scalefpd512_mask(A, B, (__v8df)_mm512_setzero_pd(), U, C)
+#define _mm512_scalef_round_pd(A, B, C)					\
+  ((__m512d)								\
+   __builtin_ia32_scalefpd512_mask((A), (B),				\
+				   (__v8df) _mm512_undefined_pd(),	\
+				   -1, (C)))
+
+#define _mm512_mask_scalef_round_pd(W, U, A, B, C)			\
+  ((__m512d) __builtin_ia32_scalefpd512_mask((A), (B), (W), (U), (C)))
+
+#define _mm512_maskz_scalef_round_pd(U, A, B, C)			\
+  ((__m512d)								\
+   __builtin_ia32_scalefpd512_mask((A), (B),				\
+				   (__v8df) _mm512_setzero_pd(),	\
+				   (U), (C)))
+
+#define _mm512_scalef_round_ps(A, B, C)					\
+  ((__m512)								\
+   __builtin_ia32_scalefps512_mask((A), (B),				\
+				   (__v16sf) _mm512_undefined_ps(),	\
+				   -1, (C)))
+
+#define _mm512_mask_scalef_round_ps(W, U, A, B, C)			\
+  ((__m512) __builtin_ia32_scalefps512_mask((A), (B), (W), (U), (C)))
+
+#define _mm512_maskz_scalef_round_ps(U, A, B, C)			\
+  ((__m512)								\
+   __builtin_ia32_scalefps512_mask((A), (B),				\
+				   (__v16sf) _mm512_setzero_ps(),	\
+				   (U), (C)))
+
+#define _mm_scalef_round_sd(A, B, C)					\
+  ((__m128d)								\
+   __builtin_ia32_scalefsd_mask_round ((A), (B),			\
+				       (__v2df) _mm_undefined_pd (),	\
+				       -1, (C)))
 
-#define _mm512_scalef_round_ps(A, B, C)            \
-    (__m512)__builtin_ia32_scalefps512_mask(A, B, (__v16sf)_mm512_undefined_ps(), -1, C)
+#define _mm_scalef_round_ss(A, B, C)					\
+  ((__m128)								\
+   __builtin_ia32_scalefss_mask_round ((A), (B),			\
+				       (__v4sf) _mm_undefined_ps (),	\
+				       -1, (C)))
 
-#define _mm512_mask_scalef_round_ps(W, U, A, B, C) \
-    (__m512)__builtin_ia32_scalefps512_mask(A, B, W, U, C)
+#define _mm_mask_scalef_round_sd(W, U, A, B, C)				\
+  ((__m128d)								\
+   __builtin_ia32_scalefsd_mask_round ((A), (B), (W), (U), (C)))
 
-#define _mm512_maskz_scalef_round_ps(U, A, B, C)   \
-    (__m512)__builtin_ia32_scalefps512_mask(A, B, (__v16sf)_mm512_setzero_ps(), U, C)
+#define _mm_mask_scalef_round_ss(W, U, A, B, C)				\
+  ((__m128)								\
+   __builtin_ia32_scalefss_mask_round ((A), (B), (W), (U), (C)))
 
-#define _mm_scalef_round_sd(A, B, C)            \
-    (__m128d)__builtin_ia32_scalefsd_mask_round (A, B, \
-	(__v2df)_mm_setzero_pd (), -1, C)
+#define _mm_maskz_scalef_round_sd(U, A, B, C)				\
+  ((__m128d)								\
+   __builtin_ia32_scalefsd_mask_round ((A), (B),			\
+				       (__v2df) _mm_setzero_pd (),	\
+				       (U), (C)))
 
-#define _mm_scalef_round_ss(A, B, C)            \
-    (__m128)__builtin_ia32_scalefss_mask_round (A, B, \
-	(__v4sf)_mm_setzero_ps (), -1, C)
+#define _mm_maskz_scalef_round_ss(U, A, B, C)				\
+  ((__m128)								\
+   __builtin_ia32_scalefss_mask_round ((A), (B),			\
+				       (__v4sf) _mm_setzero_ps (),	\
+				       (U), (C)))
 #endif
 
 #define _mm_mask_scalef_sd(W, U, A, B) \
diff --git a/gcc/testsuite/gcc.target/i386/sse-14.c b/gcc/testsuite/gcc.target/i386/sse-14.c
index 956a9d16f84..f41493b93f3 100644
--- a/gcc/testsuite/gcc.target/i386/sse-14.c
+++ b/gcc/testsuite/gcc.target/i386/sse-14.c
@@ -430,7 +430,9 @@ test_3 (_mm_maskz_mul_round_sd, __m128d, __mmask8, __m128d, __m128d, 9)
 test_3 (_mm512_maskz_mul_round_ps, __m512, __mmask16, __m512, __m512, 9)
 test_3 (_mm_maskz_mul_round_ss, __m128, __mmask8, __m128, __m128, 9)
 test_3 (_mm512_maskz_scalef_round_pd, __m512d, __mmask8, __m512d, __m512d, 9)
+test_3 (_mm_maskz_scalef_round_sd, __m128d, __mmask8, __m128d, __m128d, 9)
 test_3 (_mm512_maskz_scalef_round_ps, __m512, __mmask16, __m512, __m512, 9)
+test_3 (_mm_maskz_scalef_round_ss, __m128, __mmask8, __m128, __m128, 9)
 test_3 (_mm512_maskz_shuffle_f32x4, __m512, __mmask16, __m512, __m512, 1)
 test_3 (_mm512_maskz_shuffle_f64x2, __m512d, __mmask8, __m512d, __m512d, 1)
 test_3 (_mm512_maskz_shuffle_i32x4, __m512i, __mmask16, __m512i, __m512i, 1)
@@ -545,7 +547,9 @@ test_4 (_mm_mask_mul_round_sd, __m128d, __m128d, __mmask8, __m128d, __m128d, 9)
 test_4 (_mm512_mask_mul_round_ps, __m512, __m512, __mmask16, __m512, __m512, 9)
 test_4 (_mm_mask_mul_round_ss, __m128, __m128, __mmask8, __m128, __m128, 9)
 test_4 (_mm512_mask_scalef_round_pd, __m512d, __m512d, __mmask8, __m512d, __m512d, 9)
+test_4 (_mm_mask_scalef_round_sd, __m128d, __m128d, __mmask8, __m128d, __m128d, 9)
 test_4 (_mm512_mask_scalef_round_ps, __m512, __m512, __mmask16, __m512, __m512, 9)
+test_4 (_mm_mask_scalef_round_ss, __m128, __m128, __mmask8, __m128, __m128, 9)
 test_4 (_mm512_mask_shuffle_f32x4, __m512, __m512, __mmask16, __m512, __m512, 1)
 test_4 (_mm512_mask_shuffle_f64x2, __m512d, __m512d, __mmask8, __m512d, __m512d, 1)
 test_4 (_mm512_mask_shuffle_i32x4, __m512i, __m512i, __mmask16, __m512i, __m512i, 1)
-- 
2.18.1


  reply	other threads:[~2022-04-22 12:43 UTC|newest]

Thread overview: 4+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2022-04-22  8:10 Hongyu Wang
2022-04-22  8:49 ` Hongtao Liu
2022-04-22 12:38   ` Hongyu Wang [this message]
2022-04-24  2:35     ` Hongtao Liu

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to='CA+OydWnytyszAZevYtfNYOxvcR202xk+Pduo9+btNvpmZ_=5=A@mail.gmail.com' \
    --to=wwwhhhyyy333@gmail.com \
    --cc=crazylht@gmail.com \
    --cc=gcc-patches@gcc.gnu.org \
    --cc=hongtao.liu@intel.com \
    --cc=hongyu.wang@intel.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for read-only IMAP folder(s) and NNTP newsgroup(s).