public inbox for gcc-patches@gcc.gnu.org
 help / color / mirror / Atom feed
From: "Martin Liška" <mliska@suse.cz>
To: "gcc-pa >> GCC Patches" <gcc-patches@gcc.gnu.org>
Subject: [PATCH] IPA ICF: handle IMAGPART_EXPR and REALPART_EXPR
Date: Tue, 13 Jan 2015 21:02:00 -0000	[thread overview]
Message-ID: <54B5844D.8060602@suse.cz> (raw)

[-- Attachment #1: Type: text/plain, Size: 273 bytes --]

Hello.

As pointed out in the following PR64307, IPA ICF is missing support for 
aforementioned TREE types. Apart from that, the patch also fixes 
BIT_FIELD_REF, which was broken.

Tested on x86_64-linux-pc and no regression has been seen.

Ready for trunk?
Thanks,
Martin

[-- Attachment #2: 0001-IPA-ICF-handle-IMAGPART_EXPR-and-REALPART_EXPR.patch --]
[-- Type: text/x-patch, Size: 2521 bytes --]

From 036a27ec12ad556160cb92a86fd99ed02b66ed1a Mon Sep 17 00:00:00 2001
From: mliska <mliska@suse.cz>
Date: Tue, 13 Jan 2015 18:52:22 +0100
Subject: [PATCH] IPA ICF: handle IMAGPART_EXPR and REALPART_EXPR.

gcc/testsuite/ChangeLog:

2015-01-13  Martin Liska  <mliska@suse.cz>

	* gcc.dg/ipa/pr64307.c: New test.

gcc/ChangeLog:

2015-01-13  Martin Liska  <mliska@suse.cz>

	* ipa-icf-gimple.c (func_checker::compare_operand): Add support for
	IMAGPART_EXPR and REALPART_EXPR and fix BIT_FIELD_REF comparison.
---
 gcc/ipa-icf-gimple.c               | 14 +++++++++++++-
 gcc/testsuite/gcc.dg/ipa/pr64307.c | 32 ++++++++++++++++++++++++++++++++
 2 files changed, 45 insertions(+), 1 deletion(-)
 create mode 100644 gcc/testsuite/gcc.dg/ipa/pr64307.c

diff --git a/gcc/ipa-icf-gimple.c b/gcc/ipa-icf-gimple.c
index ed3cdf5..05c2a23 100644
--- a/gcc/ipa-icf-gimple.c
+++ b/gcc/ipa-icf-gimple.c
@@ -448,6 +448,8 @@ func_checker::compare_operand (tree t1, tree t2)
 
 	return return_with_debug (ret);
       }
+    case IMAGPART_EXPR:
+    case REALPART_EXPR:
     case ADDR_EXPR:
       {
 	x1 = TREE_OPERAND (t1, 0);
@@ -458,7 +460,17 @@ func_checker::compare_operand (tree t1, tree t2)
       }
     case BIT_FIELD_REF:
       {
-	ret = compare_decl (t1, t2);
+	x1 = TREE_OPERAND (t1, 0);
+	x2 = TREE_OPERAND (t2, 0);
+	y1 = TREE_OPERAND (t1, 1);
+	y2 = TREE_OPERAND (t2, 1);
+	z1 = TREE_OPERAND (t1, 2);
+	z2 = TREE_OPERAND (t2, 2);
+
+	ret = compare_operand (x1, x2)
+	      && compare_cst_or_decl (y1, y2)
+	      && compare_cst_or_decl (z1, z2);
+
 	return return_with_debug (ret);
       }
     case SSA_NAME:
diff --git a/gcc/testsuite/gcc.dg/ipa/pr64307.c b/gcc/testsuite/gcc.dg/ipa/pr64307.c
new file mode 100644
index 0000000..e8c1b02
--- /dev/null
+++ b/gcc/testsuite/gcc.dg/ipa/pr64307.c
@@ -0,0 +1,32 @@
+/* { dg-do compile } */
+/* { dg-options "-O0 -fipa-icf -fdump-ipa-icf"  } */
+
+#include <complex.h>
+
+typedef _Complex float COMPLEX_FLOAT;
+
+__attribute__ ((noinline))
+static float real_part(COMPLEX_FLOAT a)
+{
+  return *(float*)(&a);
+}
+
+__attribute__ ((noinline))
+static float real_part_2(COMPLEX_FLOAT a)
+{
+  return ((float*)(&a))[0];
+}
+
+int main()
+{
+  COMPLEX_FLOAT f = 1.0f + _Complex_I;
+
+  float r1 = real_part(f);
+  float r2 = real_part(f);
+
+  return r1 - r2;
+}
+
+/* { dg-final { scan-ipa-dump "Semantic equality hit:real_part_2->real_part" "icf"  } } */
+/* { dg-final { scan-ipa-dump "Equal symbols: 1" "icf"  } } */
+/* { dg-final { cleanup-ipa-dump "icf" } } */
-- 
2.1.2


             reply	other threads:[~2015-01-13 20:47 UTC|newest]

Thread overview: 2+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2015-01-13 21:02 Martin Liška [this message]
2015-01-13 21:18 ` Jeff Law

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=54B5844D.8060602@suse.cz \
    --to=mliska@suse.cz \
    --cc=gcc-patches@gcc.gnu.org \
    /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).