public inbox for gcc-patches@gcc.gnu.org
 help / color / mirror / Atom feed
From: Jason Merrill <jason@redhat.com>
To: gcc-patches List <gcc-patches@gcc.gnu.org>
Subject: C++/gimplify PATCH for c++/45307 (store of empty class not removed in gimplification)
Date: Thu, 19 Aug 2010 17:23:00 -0000	[thread overview]
Message-ID: <4C6D62E4.4030207@redhat.com> (raw)

[-- Attachment #1: Type: text/plain, Size: 217 bytes --]

The code in cp-gimplify for eliminating empty stores didn't get a chance 
to remove this one because gimplify_init_constructor moved it into the 
pre-queue.  Fixed thus.

Tested x86_64-pc-linux-gnu, applied to trunk.

[-- Attachment #2: 45307.patch --]
[-- Type: text/x-patch, Size: 2682 bytes --]

commit 6edaaca64b5498d9c3b267cb0d60d70d801b5dd5
Author: Jason Merrill <jason@redhat.com>
Date:   Tue Aug 17 14:51:04 2010 -0400

    	PR c++/45307
    	* gimplify.c (gimplify_init_constructor): Just return GS_UNHANDLED
    	if ctor is empty.
    	(gimplify_modify_expr_rhs): Adjust.

diff --git a/gcc/cp/cp-gimplify.c b/gcc/cp/cp-gimplify.c
index abd5bf3..e5a7f26 100644
--- a/gcc/cp/cp-gimplify.c
+++ b/gcc/cp/cp-gimplify.c
@@ -575,15 +575,18 @@ cp_gimplify_expr (tree *expr_p, gimple_seq *pre_p, gimple_seq *post_p)
 	  TREE_OPERAND (*expr_p, 1) = build1 (VIEW_CONVERT_EXPR,
 					      TREE_TYPE (op0), op1);
 
-	else if ((is_gimple_lvalue (op1) || INDIRECT_REF_P (op1))
-		 && !(TREE_CODE (op1) == CALL_EXPR
-		      && CALL_EXPR_RETURN_SLOT_OPT (op1))
+	else if ((is_gimple_lvalue (op1) || INDIRECT_REF_P (op1)
+		  || (TREE_CODE (op1) == CONSTRUCTOR
+		      && CONSTRUCTOR_NELTS (op1) == 0)
+		  || (TREE_CODE (op1) == CALL_EXPR
+		      && !CALL_EXPR_RETURN_SLOT_OPT (op1)))
 		 && is_really_empty_class (TREE_TYPE (op0)))
 	  {
 	    /* Remove any copies of empty classes.  We check that the RHS
-	       has a simple form so that TARGET_EXPRs and CONSTRUCTORs get
-	       reduced properly, and we leave the return slot optimization
-	       alone because it isn't a copy.
+	       has a simple form so that TARGET_EXPRs and non-empty
+	       CONSTRUCTORs get reduced properly, and we leave the return
+	       slot optimization alone because it isn't a copy (FIXME so it
+	       shouldn't be represented as one).
 
 	       Also drop volatile variables on the RHS to avoid infinite
 	       recursion from gimplify_expr trying to load the value.  */
diff --git a/gcc/gimplify.c b/gcc/gimplify.c
index 8b97ee3..7e33666 100644
--- a/gcc/gimplify.c
+++ b/gcc/gimplify.c
@@ -4237,6 +4237,10 @@ gimplify_modify_expr_rhs (tree *expr_p, tree *from_p, tree *to_p,
 	  break;
 
 	case CONSTRUCTOR:
+	  /* If we already made some changes, let the front end have a
+	     crack at this before we break it down.  */
+	  if (ret != GS_UNHANDLED)
+	    break;
 	  /* If we're initializing from a CONSTRUCTOR, break this into
 	     individual MODIFY_EXPRs.  */
 	  return gimplify_init_constructor (expr_p, pre_p, post_p, want_value,
diff --git a/gcc/testsuite/g++.dg/tree-ssa/empty-2.C b/gcc/testsuite/g++.dg/tree-ssa/empty-2.C
new file mode 100644
index 0000000..728678a
--- /dev/null
+++ b/gcc/testsuite/g++.dg/tree-ssa/empty-2.C
@@ -0,0 +1,8 @@
+// PR c++/45307
+// { dg-options -fdump-tree-gimple }
+
+struct fallible_t { };
+const fallible_t fallible = fallible_t();
+
+// { dg-final { scan-tree-dump-not "fallible" "gimple" } }
+// { dg-final { cleanup-tree-dump "gimple" } }

                 reply	other threads:[~2010-08-19 16:59 UTC|newest]

Thread overview: [no followups] expand[flat|nested]  mbox.gz  Atom feed

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=4C6D62E4.4030207@redhat.com \
    --to=jason@redhat.com \
    --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).