public inbox for gcc-cvs@sourceware.org
help / color / mirror / Atom feed
* [gcc r13-1224] c++: context completion in lookup_template_class [PR105982]
@ 2022-06-23 20:19 Patrick Palka
0 siblings, 0 replies; only message in thread
From: Patrick Palka @ 2022-06-23 20:19 UTC (permalink / raw)
To: gcc-cvs
https://gcc.gnu.org/g:01aff2ba18a654324957af446a3065db489c0f14
commit r13-1224-g01aff2ba18a654324957af446a3065db489c0f14
Author: Patrick Palka <ppalka@redhat.com>
Date: Thu Jun 23 16:18:55 2022 -0400
c++: context completion in lookup_template_class [PR105982]
The below testcase demonstrates that completion of the substituted
context during lookup_template_class can end up registering the desired
specialization for us in more cases than r13-1045-gcb7fd1ea85feea
anticipated. In particular this can happen for a non-dependent
specialization of a nested class as well.
For this testcase, during overload resolution with A's guides, we
substitute the deduced argument T=int into the TYPENAME_TYPE B::C,
during which we call lookup_template_class for A<T>::B with T=int,
which completes A<int> for the first time, which recursively registers
the desired specialization of B already. The parent call to
lookup_template_class then tries to register the same specialization,
triggering an ICE.
This patch fixes this by making lookup_template_class determine more
directly whether we need to recheck the specializations table after
completion of the context -- when and only when the call to complete_type
had an effect.
PR c++/105982
gcc/cp/ChangeLog:
* pt.cc (lookup_template_class): After calling complete_type for
the substituted context, check the table again iff the type was
previously incomplete and complete_type made it complete.
gcc/testsuite/ChangeLog:
* g++.dg/cpp1z/class-deduction111.C: New test.
Diff:
---
gcc/cp/pt.cc | 21 ++++++++++++---------
gcc/testsuite/g++.dg/cpp1z/class-deduction111.C | 10 ++++++++++
2 files changed, 22 insertions(+), 9 deletions(-)
diff --git a/gcc/cp/pt.cc b/gcc/cp/pt.cc
index 12a2b57d9e3..8672da123f4 100644
--- a/gcc/cp/pt.cc
+++ b/gcc/cp/pt.cc
@@ -10089,16 +10089,19 @@ lookup_template_class (tree d1, tree arglist, tree in_decl, tree context,
{
context = tsubst_aggr_type (context, arglist,
complain, in_decl, true);
- context = complete_type (context);
- if (is_dependent_type && arg_depth > 1)
+ /* Try completing the enclosing context if it's not already so. */
+ if (context != error_mark_node
+ && !COMPLETE_TYPE_P (context))
{
- /* If this is a dependent nested specialization such as
- A<T>::B<U> [with T=int, U=U], then completion of A<int>
- could have caused to register the desired specialization
- of B already, so check the table again (33959). */
- entry = type_specializations->find_with_hash (&elt, hash);
- if (entry)
- return entry->spec;
+ context = complete_type (context);
+ if (COMPLETE_TYPE_P (context))
+ {
+ /* Completion could have caused us to register the desired
+ specialization already, so check the table again. */
+ entry = type_specializations->find_with_hash (&elt, hash);
+ if (entry)
+ return entry->spec;
+ }
}
}
}
diff --git a/gcc/testsuite/g++.dg/cpp1z/class-deduction111.C b/gcc/testsuite/g++.dg/cpp1z/class-deduction111.C
new file mode 100644
index 00000000000..2406529ea5a
--- /dev/null
+++ b/gcc/testsuite/g++.dg/cpp1z/class-deduction111.C
@@ -0,0 +1,10 @@
+// PR c++/105982
+// { dg-do compile { target c++17 } }
+
+template<class T>
+struct A {
+ struct B { struct C { }; };
+ A(T, typename B::C);
+};
+
+A a(0, {});
^ permalink raw reply [flat|nested] only message in thread
only message in thread, other threads:[~2022-06-23 20:19 UTC | newest]
Thread overview: (only message) (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2022-06-23 20:19 [gcc r13-1224] c++: context completion in lookup_template_class [PR105982] Patrick Palka
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).