public inbox for gcc-patches@gcc.gnu.org
 help / color / mirror / Atom feed
From: Alexandre Oliva <aoliva@redhat.com>
To: gcc-patches@gcc.gnu.org
Cc: jason@redhat.com, nathan@acm.org
Subject: [PATCH] [PR86747] tsubst friend tpl ctxt before looking it up for dupes
Date: Fri, 16 Nov 2018 22:33:00 -0000	[thread overview]
Message-ID: <or5zwwk60e.fsf@lxoliva.fsfla.org> (raw)

When a member template is redeclared as a friend, we enter the context
of the member before looking it up, and then we check that the decls
are compatible.  However, when the member template references template
types of the enclosing context, say an enclosing template class, the
compare fails because the friend decl is already tsubsted, whereas the
looked up name isn't.

The problem is that the enclosing context is taken from the friend
declaration before tsubsting it, so we look up in the context of the
generic template instead of that of the tsubsted one we're
specializing.  The solution is to tsubst the enclosing context when
it's a non-namespace scope.

Regstrapped on i686- and x86_64-linux-gnu.  Ok to install?


for  gcc/cp/ChangeLog

	PR c++/86747
	* pt.c (tsubst_friend_class): Enter tsubsted class context.

for  gcc/testsuite/ChangeLog

	PR c++/86747
	* g++.dg/pr86747.C: New.
---
 gcc/cp/pt.c                    |    5 ++++-
 gcc/testsuite/g++.dg/pr86747.C |    8 ++++++++
 2 files changed, 12 insertions(+), 1 deletion(-)
 create mode 100644 gcc/testsuite/g++.dg/pr86747.C

diff --git a/gcc/cp/pt.c b/gcc/cp/pt.c
index 83d0a74b209f..82c8019431b8 100644
--- a/gcc/cp/pt.c
+++ b/gcc/cp/pt.c
@@ -10568,7 +10568,10 @@ tsubst_friend_class (tree friend_tmpl, tree args)
   if (TREE_CODE (context) == NAMESPACE_DECL)
     push_nested_namespace (context);
   else
-    push_nested_class (context);
+    {
+      context = tsubst (context, args, tf_error, NULL_TREE);
+      push_nested_class (context);
+    }
 
   tmpl = lookup_name_real (DECL_NAME (friend_tmpl), /*prefer_type=*/false,
 			   /*non_class=*/false, /*block_p=*/false,
diff --git a/gcc/testsuite/g++.dg/pr86747.C b/gcc/testsuite/g++.dg/pr86747.C
new file mode 100644
index 000000000000..5b0a0bb95146
--- /dev/null
+++ b/gcc/testsuite/g++.dg/pr86747.C
@@ -0,0 +1,8 @@
+// { dg-do compile }
+
+template <typename T> class A {
+  template <void (A::*p)()> class C; // #1
+  template <void (A::*q)()> friend class C; // #2
+};
+
+A<double> a;

-- 
Alexandre Oliva, freedom fighter   https://FSFLA.org/blogs/lxo
Be the change, be Free!         FSF Latin America board member
GNU Toolchain Engineer                Free Software Evangelist
Hay que enGNUrecerse, pero sin perder la terGNUra jamás-GNUChe

             reply	other threads:[~2018-11-16 22:33 UTC|newest]

Thread overview: 2+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2018-11-16 22:33 Alexandre Oliva [this message]
2018-12-06  1:17 ` Jason Merrill

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=or5zwwk60e.fsf@lxoliva.fsfla.org \
    --to=aoliva@redhat.com \
    --cc=gcc-patches@gcc.gnu.org \
    --cc=jason@redhat.com \
    --cc=nathan@acm.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).