public inbox for gcc-patches@gcc.gnu.org
 help / color / mirror / Atom feed
From: Alexander Monakov <amonakov@ispras.ru>
To: "Liu, Hongtao" <hongtao.liu@intel.com>
Cc: Yan Simonaytes <simonaytes.yan@ispras.ru>,
	 "gcc-patches@gcc.gnu.org" <gcc-patches@gcc.gnu.org>,
	 Uros Bizjak <ubizjak@gmail.com>
Subject: RE: [PATCH] Replace invariant ternlog operands
Date: Thu, 3 Aug 2023 20:30:39 +0300 (MSK)	[thread overview]
Message-ID: <bd1047e3-5f2b-c5e8-2da0-1423f231ac76@ispras.ru> (raw)
In-Reply-To: <SA1PR11MB6757209037AC39C10C688E42E501A@SA1PR11MB6757.namprd11.prod.outlook.com>


On Thu, 27 Jul 2023, Liu, Hongtao via Gcc-patches wrote:

> > +;; If the first and the second operands of ternlog are invariant and ;;
> > +the third operand is memory ;; then we should add load third operand
> > +from memory to register and ;; replace first and second operands with
> > +this register (define_split
> > +  [(set (match_operand:V 0 "register_operand")
> > +	(unspec:V
> > +	  [(match_operand:V 1 "register_operand")
> > +	   (match_operand:V 2 "register_operand")
> > +	   (match_operand:V 3 "memory_operand")
> > +	   (match_operand:SI 4 "const_0_to_255_operand")]
> > +	  UNSPEC_VTERNLOG))]
> > +  "ternlog_invariant_operand_mask (operands) == 3 && !reload_completed"
> Maybe better with "!reload_completed  && ternlog_invariant_operand_mask (operands) == 3"

I made this change (in both places), plus some style TLC. Ok to apply?

From d24304a9efd049e8db6df5ac78de8ca2d941a3c7 Mon Sep 17 00:00:00 2001
From: Yan Simonaytes <simonaytes.yan@ispras.ru>
Date: Tue, 25 Jul 2023 20:43:19 +0300
Subject: [PATCH] Eliminate irrelevant operands of VPTERNLOG

As mentioned in PR 110202, GCC may be presented with input where control
word of the VPTERNLOG intrinsic implies that some of its operands do not
affect the result.  In that case, we can eliminate irrelevant operands
of the instruction by substituting any other operand in their place.
This removes false dependencies.

For instance, instead of (252 = 0xfc = _MM_TERNLOG_A | _MM_TERNLOG_B)

	vpternlogq	$252, %zmm2, %zmm1, %zmm0

emit

	vpternlogq	$252, %zmm0, %zmm1, %zmm0

When VPTERNLOG is invariant w.r.t first and second operands, and the
third operand is memory, load memory into the output operand first, i.e.
instead of (85 = 0x55 = ~_MM_TERNLOG_C)

	vpternlogq	$85, (%rdi), %zmm1, %zmm0

emit

	vmovdqa64	(%rdi), %zmm0
	vpternlogq	$85, %zmm0, %zmm0, %zmm0

gcc/ChangeLog:

	* config/i386/i386-protos.h (vpternlog_irrelevant_operand_mask):
        Declare.
	(substitute_vpternlog_operands): Declare.
	* config/i386/i386.cc (vpternlog_irrelevant_operand_mask): New
        helper.
	(substitute_vpternlog_operands): New function.  Use them...
	* config/i386/sse.md: ... here in new VPTERNLOG define_splits.

gcc/testsuite/ChangeLog:

	* gcc.target/i386/invariant-ternlog-1.c: New test.
	* gcc.target/i386/invariant-ternlog-2.c: New test.
---
 gcc/config/i386/i386-protos.h                 |  3 ++
 gcc/config/i386/i386.cc                       | 43 +++++++++++++++++++
 gcc/config/i386/sse.md                        | 42 ++++++++++++++++++
 .../gcc.target/i386/invariant-ternlog-1.c     | 21 +++++++++
 .../gcc.target/i386/invariant-ternlog-2.c     | 12 ++++++
 5 files changed, 121 insertions(+)
 create mode 100644 gcc/testsuite/gcc.target/i386/invariant-ternlog-1.c
 create mode 100644 gcc/testsuite/gcc.target/i386/invariant-ternlog-2.c

diff --git a/gcc/config/i386/i386-protos.h b/gcc/config/i386/i386-protos.h
index 27fe73ca65..12e6ff0ebc 100644
--- a/gcc/config/i386/i386-protos.h
+++ b/gcc/config/i386/i386-protos.h
@@ -70,6 +70,9 @@ extern machine_mode ix86_cc_mode (enum rtx_code, rtx, rtx);
 extern int avx_vpermilp_parallel (rtx par, machine_mode mode);
 extern int avx_vperm2f128_parallel (rtx par, machine_mode mode);
 
+extern int vpternlog_irrelevant_operand_mask (rtx[]);
+extern void substitute_vpternlog_operands (rtx[]);
+
 extern bool ix86_expand_strlen (rtx, rtx, rtx, rtx);
 extern bool ix86_expand_set_or_cpymem (rtx, rtx, rtx, rtx, rtx, rtx,
 				       rtx, rtx, rtx, rtx, bool);
diff --git a/gcc/config/i386/i386.cc b/gcc/config/i386/i386.cc
index 32851a514a..9a7c1135a0 100644
--- a/gcc/config/i386/i386.cc
+++ b/gcc/config/i386/i386.cc
@@ -19420,6 +19420,49 @@ avx_vperm2f128_parallel (rtx par, machine_mode mode)
   return mask + 1;
 }
 \f
+/* Return a mask of VPTERNLOG operands that do not affect output.  */
+
+int
+vpternlog_irrelevant_operand_mask (rtx *operands)
+{
+  int mask = 0;
+  int imm8 = XINT (operands[4], 0);
+
+  if (((imm8 >> 4) & 0x0F) == (imm8 & 0x0F))
+    mask |= 1;
+  if (((imm8 >> 2) & 0x33) == (imm8 & 0x33))
+    mask |= 2;
+  if (((imm8 >> 1) & 0x55) == (imm8 & 0x55))
+    mask |= 4;
+
+  return mask;
+}
+
+/* Eliminate false dependencies on operands that do not affect output
+   by substituting other operands of a VPTERNLOG.  */
+
+void
+substitute_vpternlog_operands (rtx *operands)
+{
+  int mask = vpternlog_irrelevant_operand_mask (operands);
+
+  if (mask & 1) /* The first operand is irrelevant.  */
+    operands[1] = operands[2];
+
+  if (mask & 2) /* The second operand is irrelevant.  */
+    operands[2] = operands[1];
+
+  if (mask & 4) /* The third operand is irrelevant.  */
+    operands[3] = operands[1];
+  else if (REG_P (operands[3]))
+    {
+      if (mask & 1)
+	operands[1] = operands[3];
+      if (mask & 2)
+	operands[2] = operands[3];
+    }
+}
+\f
 /* Return a register priority for hard reg REGNO.  */
 static int
 ix86_register_priority (int hard_regno)
diff --git a/gcc/config/i386/sse.md b/gcc/config/i386/sse.md
index f793258b6c..1e2ec4bedc 100644
--- a/gcc/config/i386/sse.md
+++ b/gcc/config/i386/sse.md
@@ -12627,6 +12627,48 @@ (define_insn "*<avx512>_vternlog<mode>_all"
 		      (symbol_ref "<MODE_SIZE> == 64 || TARGET_AVX512VL")
 		      (const_string "*")))])
 
+;; When VPTERNLOG happens to be invariant w.r.t first and second operands,
+;; and the third operand is memory, eliminate false dependencies by loading
+;; memory into the output operand first.
+(define_split
+  [(set (match_operand:V 0 "register_operand")
+	(unspec:V
+	  [(match_operand:V 1 "register_operand")
+	   (match_operand:V 2 "register_operand")
+	   (match_operand:V 3 "memory_operand")
+	   (match_operand:SI 4 "const_0_to_255_operand")]
+	  UNSPEC_VTERNLOG))]
+  "!reload_completed && vpternlog_irrelevant_operand_mask (operands) == 3"
+  [(set (match_dup 0)
+	(match_dup 3))
+   (set (match_dup 0)
+	(unspec:V
+	  [(match_dup 0)
+	   (match_dup 0)
+	   (match_dup 0)
+	   (match_dup 4)]
+	  UNSPEC_VTERNLOG))])
+
+;; Eliminate false dependencies when VPTERNLOG is invariant w.r.t any
+;; of input operands (except the case handled in the above split).
+(define_split
+  [(set (match_operand:V 0 "register_operand")
+	(unspec:V
+	  [(match_operand:V 1 "register_operand")
+	   (match_operand:V 2 "register_operand")
+	   (match_operand:V 3 "nonimmediate_operand")
+	   (match_operand:SI 4 "const_0_to_255_operand")]
+	  UNSPEC_VTERNLOG))]
+  "!reload_completed && vpternlog_irrelevant_operand_mask (operands) != 0"
+  [(set (match_dup 0)
+	(unspec:V
+	  [(match_dup 1)
+	   (match_dup 2)
+	   (match_dup 3)
+	   (match_dup 4)]
+	  UNSPEC_VTERNLOG))]
+  "substitute_vpternlog_operands (operands);")
+
 ;; There must be lots of other combinations like
 ;;
 ;; (any_logic:V
diff --git a/gcc/testsuite/gcc.target/i386/invariant-ternlog-1.c b/gcc/testsuite/gcc.target/i386/invariant-ternlog-1.c
new file mode 100644
index 0000000000..21051c6bba
--- /dev/null
+++ b/gcc/testsuite/gcc.target/i386/invariant-ternlog-1.c
@@ -0,0 +1,21 @@
+/* { dg-do compile } */
+/* { dg-options "-mavx512f -O2" } */
+/* { dg-final { scan-assembler-times "vmovdqa" 4 } } */
+/* { dg-final { scan-assembler-times {vpternlog[^\n\r]*\(%rdx\)} 2 } } */
+
+#include <immintrin.h>
+
+__m512i f(__m512i* a, __m512i* b, __m512i* c)
+{
+	return _mm512_ternarylogic_epi64 (a[0], b[0], c[0], ~_MM_TERNLOG_B | ~_MM_TERNLOG_C);
+}
+
+__m512i g(__m512i* a, __m512i* b, __m512i* c)
+{
+	return _mm512_ternarylogic_epi64 (a[0], b[0], c[0], ~_MM_TERNLOG_A | ~_MM_TERNLOG_C);
+}
+
+__m512i h(__m512i* a, __m512i* b, __m512i* c)
+{
+	return _mm512_ternarylogic_epi64 (a[0], b[0], c[0], ~_MM_TERNLOG_A | ~_MM_TERNLOG_B);
+}
diff --git a/gcc/testsuite/gcc.target/i386/invariant-ternlog-2.c b/gcc/testsuite/gcc.target/i386/invariant-ternlog-2.c
new file mode 100644
index 0000000000..d70bbb0239
--- /dev/null
+++ b/gcc/testsuite/gcc.target/i386/invariant-ternlog-2.c
@@ -0,0 +1,12 @@
+/* { dg-do compile } */
+/* { dg-options "-mavx512f -O2" } */
+/* { dg-final { scan-assembler-times "vmovdqa" 1 } } */
+/* { dg-final { scan-assembler "vpternlog.*zmm0.*zmm0.*zmm0" } } */
+
+#include <immintrin.h>
+
+__m512i f(__m512i* a, __m512i* b, __m512i* c)
+{
+	return _mm512_ternarylogic_epi64 (a[0], b[0], c[0], ~_MM_TERNLOG_C);
+}
+
-- 
2.39.2


  reply	other threads:[~2023-08-03 17:30 UTC|newest]

Thread overview: 4+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2023-07-25 18:11 Yan Simonaytes
2023-07-27  3:00 ` Liu, Hongtao
2023-08-03 17:30   ` Alexander Monakov [this message]
2023-08-04  0:43     ` 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=bd1047e3-5f2b-c5e8-2da0-1423f231ac76@ispras.ru \
    --to=amonakov@ispras.ru \
    --cc=gcc-patches@gcc.gnu.org \
    --cc=hongtao.liu@intel.com \
    --cc=simonaytes.yan@ispras.ru \
    --cc=ubizjak@gmail.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).