Latest cvs causes segfault on gcc version 3.4.2 20041003 (Red Hat 3.4.2-4)

This is current gcc for Linux/FC3T2. (Fedora core 3 test 2) /usr/local/src/boost.cvs/libs/python/build/../src/tuple.cpp: In constructor `boost::python::detail::tuple_base::tuple_base(const boost::python::api::object&)': /usr/local/src/boost.cvs/libs/python/build/../src/tuple.cpp:21: internal compiler error: Segmentation fault Bug report filed.

"Neal D. Becker" <ndbecker2@verizon.net> writes:
This is current gcc for Linux/FC3T2. (Fedora core 3 test 2)
/usr/local/src/boost.cvs/libs/python/build/../src/tuple.cpp: In constructor `boost::python::detail::tuple_base::tuple_base(const boost::python::api::object&)': /usr/local/src/boost.cvs/libs/python/build/../src/tuple.cpp:21: internal compiler error: Segmentation fault
Bug report filed.
Any idea what we might do to work around it? What happens if you add a typedef object base; to tuple_base in tuple.hpp and replace object(...) in the offending line with base(...)? -- Dave Abrahams Boost Consulting http://www.boost-consulting.com

David Abrahams wrote:
"Neal D. Becker" <ndbecker2@verizon.net> writes:
This is current gcc for Linux/FC3T2. (Fedora core 3 test 2)
/usr/local/src/boost.cvs/libs/python/build/../src/tuple.cpp: In constructor `boost::python::detail::tuple_base::tuple_base(const boost::python::api::object&)': /usr/local/src/boost.cvs/libs/python/build/../src/tuple.cpp:21: internal compiler error: Segmentation fault
Bug report filed.
Any idea what we might do to work around it?
What happens if you add a
typedef object base;
to tuple_base in tuple.hpp and replace object(...) in the offending line with base(...)?
No help. Apparantly a duplicate of: http://gcc.gnu.org/bugzilla/show_bug.cgi?id=17846 This is supposedly resolved in gcc cvs, but I don't know how to workaround. I hope it will be fixed in gcc before FC3 is released though.

Neal Becker wrote:
This is supposedly resolved in gcc cvs, but I don't know how to workaround. I hope it will be fixed in gcc before FC3 is released though.
I doubt it. FC3 is due in early november. No time for another gcc release. Too bad. Regards, Stefan

Stefan Seefeld wrote:
Neal Becker wrote:
This is supposedly resolved in gcc cvs, but I don't know how to workaround. I hope it will be fixed in gcc before FC3 is released though.
I doubt it. FC3 is due in early november. No time for another gcc release. Too bad.
The gcc test versions from Redhat are not releases. They change every few days. We'll see.

The problem will be fixed in 3.4.2-5, so I think there is no reason to try to find a workaround.

"Neal D. Becker" <ndbecker2@verizon.net> writes:
The problem will be fixed in 3.4.2-5, so I think there is no reason to try to find a workaround.
Thanks for looking into it. -- Dave Abrahams Boost Consulting http://www.boost-consulting.com
participants (4)
-
David Abrahams
-
Neal Becker
-
Neal D. Becker
-
Stefan Seefeld