public inbox for gcc-patches@gcc.gnu.org
 help / color / mirror / Atom feed
From: liuhongt <hongtao.liu@intel.com>
To: gcc-patches@gcc.gnu.org
Cc: crazylht@gmail.com, hjl.tools@gmail.com
Subject: [PATCH] [vect]Use intermiediate integer type for float_expr/fix_trunc_expr when direct optab is not existed.
Date: Fri,  2 Jun 2023 09:00:15 +0800	[thread overview]
Message-ID: <20230602010015.2571612-1-hongtao.liu@intel.com> (raw)

We have already use intermidate type in case WIDEN, but not for NONE,
this patch extended that.

I didn't do that in pattern recog since we need to know whether the
stmt belongs to any slp_node to decide the vectype, the related optabs
are checked according to vectype_in and vectype_out. For non-slp case,
vec_pack/unpack are always used when lhs has different size from rhs,
for slp case, sometimes vec_pack/unpack is used, somethings
direct conversion is used.

Bootstrapped and regtested on x86_64-pc-linux-gnu{-m32,}.
Ok for trunk?

gcc/ChangeLog:

	PR target/110018
	* tree-vect-stmts.cc (vectorizable_conversion): Use
	intermiediate integer type for float_expr/fix_trunc_expr when
	direct optab is not existed.

gcc/testsuite/ChangeLog:

	* gcc.target/i386/pr110018-1.c: New test.
---
 gcc/testsuite/gcc.target/i386/pr110018-1.c | 94 ++++++++++++++++++++++
 gcc/tree-vect-stmts.cc                     | 56 ++++++++++++-
 2 files changed, 149 insertions(+), 1 deletion(-)
 create mode 100644 gcc/testsuite/gcc.target/i386/pr110018-1.c

diff --git a/gcc/testsuite/gcc.target/i386/pr110018-1.c b/gcc/testsuite/gcc.target/i386/pr110018-1.c
new file mode 100644
index 00000000000..b1baffd7af1
--- /dev/null
+++ b/gcc/testsuite/gcc.target/i386/pr110018-1.c
@@ -0,0 +1,94 @@
+/* { dg-do compile } */
+/* { dg-options "-mavx512fp16 -mavx512vl -O2 -mavx512dq" } */
+/* { dg-final { scan-assembler-times {(?n)vcvttp[dsh]2[dqw]} 5 } } */
+/* { dg-final { scan-assembler-times {(?n)vcvt[dqw]*2p[dsh]} 5 } } */
+
+void
+foo (double* __restrict a, char* b)
+{
+  a[0] = b[0];
+  a[1] = b[1];
+}
+
+void
+foo1 (float* __restrict a, char* b)
+{
+  a[0] = b[0];
+  a[1] = b[1];
+  a[2] = b[2];
+  a[3] = b[3];
+}
+
+void
+foo2 (_Float16* __restrict a, char* b)
+{
+  a[0] = b[0];
+  a[1] = b[1];
+  a[2] = b[2];
+  a[3] = b[3];
+  a[4] = b[4];
+  a[5] = b[5];
+  a[6] = b[6];
+  a[7] = b[7];
+}
+
+void
+foo3 (double* __restrict a, short* b)
+{
+  a[0] = b[0];
+  a[1] = b[1];
+}
+
+void
+foo4 (float* __restrict a, char* b)
+{
+  a[0] = b[0];
+  a[1] = b[1];
+  a[2] = b[2];
+  a[3] = b[3];
+}
+
+void
+foo5 (double* __restrict b, char* a)
+{
+  a[0] = b[0];
+  a[1] = b[1];
+}
+
+void
+foo6 (float* __restrict b, char* a)
+{
+  a[0] = b[0];
+  a[1] = b[1];
+  a[2] = b[2];
+  a[3] = b[3];
+}
+
+void
+foo7 (_Float16* __restrict b, char* a)
+{
+  a[0] = b[0];
+  a[1] = b[1];
+  a[2] = b[2];
+  a[3] = b[3];
+  a[4] = b[4];
+  a[5] = b[5];
+  a[6] = b[6];
+  a[7] = b[7];
+}
+
+void
+foo8 (double* __restrict b, short* a)
+{
+  a[0] = b[0];
+  a[1] = b[1];
+}
+
+void
+foo9 (float* __restrict b, char* a)
+{
+  a[0] = b[0];
+  a[1] = b[1];
+  a[2] = b[2];
+  a[3] = b[3];
+}
diff --git a/gcc/tree-vect-stmts.cc b/gcc/tree-vect-stmts.cc
index bd3b07a3aa1..1118c89686d 100644
--- a/gcc/tree-vect-stmts.cc
+++ b/gcc/tree-vect-stmts.cc
@@ -5162,6 +5162,49 @@ vectorizable_conversion (vec_info *vinfo,
 	return false;
       if (supportable_convert_operation (code, vectype_out, vectype_in, &code1))
 	break;
+      if ((code == FLOAT_EXPR
+	   && GET_MODE_SIZE (lhs_mode) > GET_MODE_SIZE (rhs_mode))
+	  || (code == FIX_TRUNC_EXPR
+	      && GET_MODE_SIZE (rhs_mode) > GET_MODE_SIZE (lhs_mode)))
+	{
+	  bool float_expr_p = code == FLOAT_EXPR;
+	  scalar_mode imode = float_expr_p ? rhs_mode : lhs_mode;
+	  fltsz = GET_MODE_SIZE (float_expr_p ? lhs_mode : rhs_mode);
+	  code1 = float_expr_p ? code : NOP_EXPR;
+	  codecvt1 = float_expr_p ? NOP_EXPR : code;
+	  FOR_EACH_2XWIDER_MODE (rhs_mode_iter, imode)
+	    {
+	      imode = rhs_mode_iter.require ();
+	      if (GET_MODE_SIZE (imode) > fltsz)
+		break;
+
+	      cvt_type
+		= build_nonstandard_integer_type (GET_MODE_BITSIZE (imode),
+						  0);
+	      cvt_type = get_vectype_for_scalar_type (vinfo, cvt_type,
+						      slp_node);
+	      /* This should only happened for SLP as long as loop vectorizer
+		 only supports same-sized vector.  */
+	      if (cvt_type == NULL_TREE
+		  || maybe_ne (TYPE_VECTOR_SUBPARTS (cvt_type), nunits_in)
+		  || !supportable_convert_operation (code1, vectype_out,
+						     cvt_type, &code1)
+		  || !supportable_convert_operation (codecvt1, cvt_type,
+						     vectype_in, &codecvt1))
+		continue;
+
+	      found_mode = true;
+	      break;
+	    }
+
+	  if (found_mode)
+	    {
+	      multi_step_cvt++;
+	      interm_types.safe_push (cvt_type);
+	      cvt_type = NULL_TREE;
+	      break;
+	    }
+	}
       /* FALLTHRU */
     unsupported:
       if (dump_enabled_p ())
@@ -5381,7 +5424,18 @@ vectorizable_conversion (vec_info *vinfo,
 	{
 	  /* Arguments are ready, create the new vector stmt.  */
 	  gcc_assert (TREE_CODE_LENGTH (code1) == unary_op);
-	  gassign *new_stmt = gimple_build_assign (vec_dest, code1, vop0);
+	  gassign* new_stmt;
+	  if (multi_step_cvt)
+	    {
+	      gcc_assert (multi_step_cvt == 1);
+	      new_stmt = gimple_build_assign (vec_dest, codecvt1, vop0);
+	      new_temp = make_ssa_name (vec_dest, new_stmt);
+	      gimple_assign_set_lhs (new_stmt, new_temp);
+	      vect_finish_stmt_generation (vinfo, stmt_info, new_stmt, gsi);
+	      vop0 = new_temp;
+	      vec_dest = vec_dsts[0];
+	    }
+	  new_stmt = gimple_build_assign (vec_dest, code1, vop0);
 	  new_temp = make_ssa_name (vec_dest, new_stmt);
 	  gimple_assign_set_lhs (new_stmt, new_temp);
 	  vect_finish_stmt_generation (vinfo, stmt_info, new_stmt, gsi);
-- 
2.39.1.388.g2fc9e9ca3c


             reply	other threads:[~2023-06-02  1:00 UTC|newest]

Thread overview: 16+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2023-06-02  1:00 liuhongt [this message]
2023-06-20  8:38 ` Richard Biener
2023-06-20  9:09   ` Hongtao Liu
2023-06-20  9:22     ` Richard Biener
2023-06-20 16:11       ` liuhongt
2023-06-21  7:49         ` Uros Bizjak
2023-06-21  8:19           ` Richard Biener
2023-06-21 14:37             ` Uros Bizjak
2023-06-22 20:36         ` Thiago Jung Bauermann
2023-06-21  9:10   ` Richard Sandiford
2023-06-21  9:32     ` Richard Sandiford
2023-06-21 11:04       ` Richard Biener
2023-06-21 11:20         ` Richard Sandiford
2023-06-21 20:39         ` Joseph Myers
2023-06-22  7:32           ` Richard Biener
2023-07-12  9:19             ` Robin Dapp

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=20230602010015.2571612-1-hongtao.liu@intel.com \
    --to=hongtao.liu@intel.com \
    --cc=crazylht@gmail.com \
    --cc=gcc-patches@gcc.gnu.org \
    --cc=hjl.tools@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).