Re: [Boost-users] boost 1.40 beta, serialization via base pointer crashes
Is the "latest AIX v10.1 compiler" the same as the IBM VaCPP used in the trunk tests? These tests show that compiler failing a number of tests which use export. This has been traced to the compiler optimizing out code not explicitly referred to.
Robert Ramey
I am not familiar with the trunk tests. However XL 10.1.0.3 AIX is the
latest C++ compiler. This
use vacpp as the bjam toolset argument. But internally I gather it uses xlC.
When I ran the sample in the Totalview(debugger) it shows the following.
Can't read type info for "basic_pointer_oserializer": Expecting mangled
name, and didn't find one.
Can't read type info for "basic_pointer_iserializer": Expecting mangled
name, and didn't find one.
Can't read type info for
"singleton ": Expecting ":", and didn't find one.
Can't read type info for
"singleton From your reply your saying that this is a compiler bug
Do you know if any defect report has been raised ?
Is there any workaround that stops the compiler doing these optimisation. If not is there possibility of changing the serialization code ?
Best regards,
Ta,
Avi
"Avi Bahra"
Is the "latest AIX v10.1 compiler" the same as the IBM VaCPP used in the trunk tests? These tests show that compiler failing a number of tests which use export. This has been traced to the compiler optimizing out code not explicitly referred to.
Robert Ramey
I am not familiar with the trunk tests. However XL 10.1.0.3 AIX is the latest C++ compiler. http://www.boost.org/development/tests/trunk/developer/serialization_.html From your reply your saying that this is a compiler bug Export facility depends upon non-portable C++ features. Not all compilers support these features. Do you know if any defect report has been raised ? I believe that the development team for this compiler is aware of this situation. Is there any workaround that stops the compiler doing these optimisation. Consider using explicity type registration in the archive. This is totally portable to all C++ compilers. If not is there possibility of changing the serialization code ? I don't believe that it's possible. I don't have the compiler. And I couldn't justifty the time in any case Robert Ramey Best regards, Ta, Avi ------------------------------------------------------------------------------ _______________________________________________ Boost-users mailing list Boost-users@lists.boost.org http://lists.boost.org/mailman/listinfo.cgi/boost-users
participants (2)
-
Avi Bahra -
Robert Ramey