public inbox for gcc-cvs@sourceware.org
help / color / mirror / Atom feed
From: Jason Merrill <jason@gcc.gnu.org>
To: gcc-cvs@gcc.gnu.org
Subject: [gcc r12-10109] c++: NRV and goto [PR92407]
Date: Wed, 24 Jan 2024 19:40:38 +0000 (GMT)	[thread overview]
Message-ID: <20240124194039.0B8783858C20@sourceware.org> (raw)

https://gcc.gnu.org/g:193321d2649a818c3b322c8db79a2805cfdd7095

commit r12-10109-g193321d2649a818c3b322c8db79a2805cfdd7095
Author: Jason Merrill <jason@redhat.com>
Date:   Sun Jun 4 12:00:55 2023 -0400

    c++: NRV and goto [PR92407]
    
    Here our named return value optimization was breaking the required
    destructor when the goto takes 'a' out of scope.  A simple fix for the
    release branches is to disable the optimization in the presence of backward
    goto.
    
    We could do better by disabling the optimization only if there is a backward
    goto across the variable declaration, but we don't track that, and in GCC 14
    we instead make the goto work with NRV.
    
            PR c++/92407
    
    gcc/cp/ChangeLog:
    
            * cp-tree.h (struct language_function): Add backward_goto.
            * decl.cc (check_goto): Set it.
            * typeck.cc (check_return_expr): Prevent NRV if set.
    
    gcc/testsuite/ChangeLog:
    
            * g++.dg/opt/nrv22.C: New test.
    
    (cherry picked from commit a645347c19b07cc7abd7bf276c6769fc41afc932)

Diff:
---
 gcc/cp/cp-tree.h                 |  1 +
 gcc/cp/decl.cc                   |  2 ++
 gcc/cp/typeck.cc                 |  3 +++
 gcc/testsuite/g++.dg/opt/nrv22.C | 30 ++++++++++++++++++++++++++++++
 4 files changed, 36 insertions(+)

diff --git a/gcc/cp/cp-tree.h b/gcc/cp/cp-tree.h
index 5b3836422d6..132929375ea 100644
--- a/gcc/cp/cp-tree.h
+++ b/gcc/cp/cp-tree.h
@@ -2077,6 +2077,7 @@ struct GTY(()) language_function {
 
   BOOL_BITFIELD invalid_constexpr : 1;
   BOOL_BITFIELD throwing_cleanup : 1;
+  BOOL_BITFIELD backward_goto : 1;
 
   hash_table<named_label_hash> *x_named_labels;
 
diff --git a/gcc/cp/decl.cc b/gcc/cp/decl.cc
index 22060b11176..878b14fe771 100644
--- a/gcc/cp/decl.cc
+++ b/gcc/cp/decl.cc
@@ -3650,6 +3650,8 @@ check_goto (tree decl)
       return;
     }
 
+  cp_function_chain->backward_goto = true;
+
   bool saw_catch = false, complained = false;
   int identified = 0;
   tree bad;
diff --git a/gcc/cp/typeck.cc b/gcc/cp/typeck.cc
index 3be59f47179..6ca240fe40a 100644
--- a/gcc/cp/typeck.cc
+++ b/gcc/cp/typeck.cc
@@ -10678,6 +10678,9 @@ check_return_expr (tree retval, bool *no_warning)
   if (fn_returns_value_p && flag_elide_constructors)
     {
       if (named_return_value_okay_p
+	  /* The current NRV implementation breaks if a backward goto needs to
+	     destroy the object (PR92407).  */
+	  && !cp_function_chain->backward_goto
           && (current_function_return_value == NULL_TREE
 	      || current_function_return_value == bare_retval))
 	current_function_return_value = bare_retval;
diff --git a/gcc/testsuite/g++.dg/opt/nrv22.C b/gcc/testsuite/g++.dg/opt/nrv22.C
new file mode 100644
index 00000000000..eb889fa615b
--- /dev/null
+++ b/gcc/testsuite/g++.dg/opt/nrv22.C
@@ -0,0 +1,30 @@
+// PR c++/92407
+// { dg-do run }
+
+struct A
+{
+  A () { a++; }
+  A (const A &) { a++; }
+  ~A () { a--; }
+  static int a;
+};
+int A::a = 0;
+
+A
+foo ()
+{
+  int cnt = 10;
+lab:
+  A a;
+  if (cnt--)
+    goto lab;
+  return a;
+}
+
+int
+main ()
+{
+  foo ();
+  if (A::a)
+    __builtin_abort ();
+}

                 reply	other threads:[~2024-01-24 19:40 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=20240124194039.0B8783858C20@sourceware.org \
    --to=jason@gcc.gnu.org \
    --cc=gcc-cvs@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).