https://gcc.gnu.org/bugzilla/show_bug.cgi?id=55361
Patrick Palka changed:
What|Removed |Added
CC||zhonghao at pku dot org.cn
--- Comment #
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=55361
Eric Gallager changed:
What|Removed |Added
Status|UNCONFIRMED |NEW
Last reconfirmed|
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=55361
Jonathan Wakely changed:
What|Removed |Added
Keywords||diagnostic
Severity|
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=55361
--- Comment #4 from Jonathan Wakely 2012-11-17
00:19:18 UTC ---
unless this is recategorised as a diagnostic enhancement request this is
clearly invalid, the compiler is not required to give an error for this code
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=55361
--- Comment #3 from Paolo Carlini 2012-11-16
23:06:09 UTC ---
Likewise EDG as used by Intel. clang errors out. From the user point of view,
to make debugging easier, the earlier errors are produced the better, but this
is just a general pr
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=55361
--- Comment #2 from Andrew Pinski 2012-11-16
22:38:47 UTC ---
I don't think this is a gcc bug as access control is always done only at
template-instantiation time.
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=55361
--- Comment #1 from Andrew Pinski 2012-11-16
22:32:03 UTC ---
The Comeau C/C++ online tester does not error out on this testcase.
