[multi_index] spurious error in CW 9
This test occasionally fails: http://tinyurl.com/72xcb but there's no error output or any failure indication, nothing. The problem shows only at times, so I guess it'll have to do with the compiler being low on resources, but couldn't say for sure. Any idea? Joaquín M López Muñoz Telefónica, Investigación y Desarrollo
Joaquín Mª López Muñoz wrote:
This test occasionally fails:
but there's no error output or any failure indication, nothing. The problem shows only at times, so I guess it'll have to do with the compiler being low on resources, but couldn't say for sure. Any idea?
There is one testcase in the complete run where CW 9 says 'goodbye' with a GPF. I've never investigated which one it is but this one seems to be a good candidate... BTW, is there any official support for boost from Metrowerks? (for compiler bugs etc, and: my trial licence expires this week!) Stefan
Stefan Slapeta <stefan_nospam_@slapeta.com> writes:
Joaquín Mª López Muñoz wrote:
This test occasionally fails: http://tinyurl.com/72xcb but there's no error output or any failure indication, nothing. The problem shows only at times, so I guess it'll have to do with the compiler being low on resources, but couldn't say for sure. Any idea?
There is one testcase in the complete run where CW 9 says 'goodbye' with a GPF. I've never investigated which one it is but this one seems to be a good candidate...
BTW, is there any official support for boost from Metrowerks? (for compiler bugs etc, and: my trial licence expires this week!)
Howard Hinnant is their library author, participates in this list, and has a direct line to their compiler author. They do care about supporting Boost. Please make a compressed preprocessed source file for any TUs that crash the compiler, and get in touch with Howard. -- Dave Abrahams Boost Consulting http://www.boost-consulting.com
On Aug 3, 2004, at 10:01 AM, Stefan Slapeta wrote:
Joaquín Mª López Muñoz wrote:
This test occasionally fails: http://tinyurl.com/72xcb but there's no error output or any failure indication, nothing. The problem shows only at times, so I guess it'll have to do with the compiler being low on resources, but couldn't say for sure. Any idea?
There is one testcase in the complete run where CW 9 says 'goodbye' with a GPF. I've never investigated which one it is but this one seems to be a good candidate...
BTW, is there any official support for boost from Metrowerks? (for compiler bugs etc, and: my trial licence expires this week!)
There is no official support for boost from Metrowerks. However I'm the C++ lib guy at Metrowerks and I sometimes hang out and see what I can do. I don't usually go poking into the cvs/sandbox areas, but I do usually keep a copy of the latest released version of boost installed. Please feel free to email me a demo case of any problems you find. If it isn't a library issue, I'll forward it on to the compiler team. The more self contained a demo is, the more quickly it will be looked at. If it is both self contained, and already stripped down to a minimal translation unit, it will be looked at very quickly. Preprocessing the troublesome source, and then paring that down to the minimum which still exhibits problems is often the easiest way of creating a self-contained, minimal demo. -Howard
participants (9)
-
David Abrahams -
Howard Hinnant -
Joaquín Mª López Muñoz -
Stefan Slapeta