This is the mail archive of the
gcc-bugs@gcc.gnu.org
mailing list for the GCC project.
[Bug c++/12573] [3.4 Regression ] ICE (segfault) with Boost.Python
- From: "giovannibajo at libero dot it" <gcc-bugzilla at gcc dot gnu dot org>
- To: gcc-bugs at gcc dot gnu dot org
- Date: 22 Nov 2003 13:52:23 -0000
- Subject: [Bug c++/12573] [3.4 Regression ] ICE (segfault) with Boost.Python
- References: <20031011014942.12573.jbrandmeyer@users.sourceforge.net>
- Reply-to: gcc-bugzilla at gcc dot gnu dot org
------- Additional Comments From giovannibajo at libero dot it 2003-11-22 13:52 -------
Subject: Re: [3.4 Regression ] ICE (segfault) with Boost.Python
On Saturday, November 22, 2003 6:50 AM [GMT+1=CET],
gdr at integrable-solutions dot net <gcc-bugzilla@gcc.gnu.org> wrote:
>> + if (TREE_CODE (expression) == COMPONENT_REF)
>> + return value_dependent_expression_p (TREE_OPERAND (expression, 0));
>
> Don't you want to test both arms of COMPONENT_REF?
Yup, I was misguided by the internal documentation, which (wrongly) says that
COMPONENT_REF is supposed to always have a FIELD_DECL as second operand. Of
course, it could easily be a SCOPE_REF. I'll update the patch and try to prepare
a testcase, thanks for the quick comment.
Giovanni Bajo
--
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=12573