From: Jakub Jelinek <jakub@redhat.com>
To: Richard Biener <rguenther@suse.de>
Cc: gcc-patches@gcc.gnu.org
Subject: [PATCH] match.pd: Add (-(X < 0) | 1) * X -> abs (X) etc. simplifications [PR94785]
Date: Thu, 31 Dec 2020 10:34:34 +0100 [thread overview]
Message-ID: <20201231093434.GD3788@tucnak> (raw)
Hi!
The following patch adds two simplifications to recognize idioms
for ABS_EXPR resp. ABSU_EXPR.
Bootstrapped/regtested on x86_64-linux and i686-linux, ok for trunk?
2020-12-31 Jakub Jelinek <jakub@redhat.com>
PR tree-optimization/94785
* match.pd ((-(X < 0) | 1) * X -> abs (X)): New simplification.
((-(X < 0) | 1U) * X -> absu (X)): Likewise.
* gcc.dg/tree-ssa/pr94785.c: New test.
--- gcc/match.pd.jj 2020-12-15 22:43:39.254276743 +0100
+++ gcc/match.pd 2020-12-30 12:00:00.924301784 +0100
@@ -1374,6 +1374,22 @@ (define_operator_list COND_TERNARY
(absu tree_expr_nonnegative_p@0)
(convert @0))
+/* Simplify (-(X < 0) | 1) * X into abs (X). */
+(simplify
+ (mult:c (bit_ior (negate (convert? (lt @0 integer_zerop))) integer_onep) @0)
+ (if (INTEGRAL_TYPE_P (type) && !TYPE_UNSIGNED (type))
+ (abs @0)))
+
+/* Similarly (-(X < 0) | 1U) * X into absu (X). */
+(simplify
+ (mult:c (bit_ior (nop_convert (negate (convert? (lt @0 integer_zerop))))
+ integer_onep) (nop_convert @0))
+ (if (INTEGRAL_TYPE_P (type)
+ && TYPE_UNSIGNED (type)
+ && INTEGRAL_TYPE_P (TREE_TYPE (@0))
+ && !TYPE_UNSIGNED (TREE_TYPE (@0)))
+ (absu @0)))
+
/* A few cases of fold-const.c negate_expr_p predicate. */
(match negate_expr_p
INTEGER_CST
--- gcc/testsuite/gcc.dg/tree-ssa/pr94785.c.jj 2020-12-30 12:03:45.693776141 +0100
+++ gcc/testsuite/gcc.dg/tree-ssa/pr94785.c 2020-12-30 12:04:26.895313170 +0100
@@ -0,0 +1,36 @@
+/* PR tree-optimization/94785 */
+/* { dg-do compile } */
+/* { dg-options "-O2 -fdump-tree-optimized" } */
+/* { dg-final { scan-tree-dump-times " = ABS_EXPR <v_\[0-9]*\\\(D\\\)>;" 2 "optimized" } } */
+/* { dg-final { scan-tree-dump-times " = ABSU_EXPR <v_\[0-9]*\\\(D\\\)>;" 2 "optimized" } } */
+
+int
+f1 (int v)
+{
+ return (1 | -(v < 0)) * v;
+}
+
+unsigned
+f2 (int v)
+{
+ return (1U | -(v < 0)) * v;
+}
+
+int
+f3 (int v)
+{
+ int a = v < 0;
+ int b = -a;
+ int c = 1 | b;
+ return c * v;
+}
+
+unsigned
+f4 (int v)
+{
+ int a = v < 0;
+ int b = -a;
+ unsigned c = b;
+ unsigned d = c | 1;
+ return d * v;
+}
Jakub
next reply other threads:[~2020-12-31 9:34 UTC|newest]
Thread overview: 2+ messages / expand[flat|nested] mbox.gz Atom feed top
2020-12-31 9:34 Jakub Jelinek [this message]
2020-12-31 9:56 ` Richard Biener
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=20201231093434.GD3788@tucnak \
--to=jakub@redhat.com \
--cc=gcc-patches@gcc.gnu.org \
--cc=rguenther@suse.de \
/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).