public inbox for gcc-patches@gcc.gnu.org
 help / color / mirror / Atom feed
From: Ilya Enkovich <enkovich.gnu@gmail.com>
To: gcc-patches@gcc.gnu.org
Cc: hubicka@ucw.cz
Subject: [PATCH] Walk through thunks when propagating comdat group
Date: Tue, 07 Apr 2015 14:35:00 -0000	[thread overview]
Message-ID: <20150407143501.GD11622@msticlxl57.ims.intel.com> (raw)

Hi,

Currently enqueue_references in comdats pass enqueue target function instead of thunks.  But propagate_comdat_group doesn't walk through thunks and therefore comdat group of thunk's caller is not propagated into thunk's target function.  This patch tries to fix it.  Testing is in progress.  Does it look OK?

Thanks,
Ilya
--
gcc/

2015-04-07  Ilya Enkovich  <ilya.enkovich@intel.com>

	* ipa-comdats.c (propagate_comdat_group): Walk through thunks.

gcc/testsuite/

2015-04-07  Ilya Enkovich  <ilya.enkovich@intel.com>

	* gcc.target/i386/mpx/chkp-thunk-comdat-3.c: New.


diff --git a/gcc/ipa-comdats.c b/gcc/ipa-comdats.c
index f349f9f..74088a9 100644
--- a/gcc/ipa-comdats.c
+++ b/gcc/ipa-comdats.c
@@ -144,10 +144,12 @@ propagate_comdat_group (struct symtab_node *symbol,
 
 	/* If we see inline clone, its comdat group actually
 	   corresponds to the comdat group of the function it is inlined
-	   to.  */
+	   to.  Dive into thunks similar to aliases. */
 
 	if (cgraph_node * cn = dyn_cast <cgraph_node *> (symbol2))
 	  {
+	    if (cn->thunk.thunk_p)
+	      newgroup = propagate_comdat_group (symbol2, newgroup, map);
 	    if (cn->global.inlined_to)
 	      symbol2 = cn->global.inlined_to;
 	  }
diff --git a/gcc/testsuite/gcc.target/i386/mpx/chkp-thunk-comdat-3.c b/gcc/testsuite/gcc.target/i386/mpx/chkp-thunk-comdat-3.c
new file mode 100644
index 0000000..dd0057e
--- /dev/null
+++ b/gcc/testsuite/gcc.target/i386/mpx/chkp-thunk-comdat-3.c
@@ -0,0 +1,23 @@
+/* { dg-do compile } */
+/* { dg-options "-fcheck-pointer-bounds -mmpx -O -fvisibility=hidden" } */
+
+int val;
+
+static int __attribute__((noinline))
+test1 ()
+{
+  return val;
+}
+
+static int __attribute__((bnd_legacy,noinline))
+test2 ()
+{
+  return test1 ();
+}
+
+int
+test3 (void)
+{
+  return test2 ();
+}
+

             reply	other threads:[~2015-04-07 14:35 UTC|newest]

Thread overview: 3+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2015-04-07 14:35 Ilya Enkovich [this message]
2015-04-07 20:37 ` Jan Hubicka
2015-04-08 13:48   ` Ilya Enkovich

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=20150407143501.GD11622@msticlxl57.ims.intel.com \
    --to=enkovich.gnu@gmail.com \
    --cc=gcc-patches@gcc.gnu.org \
    --cc=hubicka@ucw.cz \
    /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).