[
Date Prev][
Date Next][
Thread Prev][
Thread Next][
Date Index][
Thread Index]
[
List Home]
[cdt-dev] RE: Jump in number of unresolved symbols
|
Hi Sergey,
an
improvement in the name-resolution may also cause an increase in problems. Here
is an example:
I have added the argument dependent lookup for overloaded operators. Due
to a yet unknown bug in the
overload resolution this may
yield additional ambiguities. However, I
am just guessing.
When I look at your numbers, I see that you pick up
more headers than before. This is probably
because
I have fixed the include_next. This could also be
causing the additional problems.
All my changes are rectified by testcases and I have
also tested with my usual test-projects (tao, ace, linux,
boost, firefox) without an increase in problem
bindings. I recommend to make a diff in the problems reported
and
look at the first new
one.
Markus.
I'm seeing a 2x increase in number of unresolved symbols since
Tuesday:
!ENTRY org.eclipse.cdt.core 1 0 2009-10-27 10:56:01.305
!MESSAGE Indexed 'src2' (2,198 sources, 3,460 headers) in 1,695.23 sec:
1,068,881 declarations; 4,002,500 references; 0 unresolved inclusions; 0
syntax errors; 5,085 unresolved names (0.10%)
!ENTRY org.eclipse.cdt.core 1 0 2009-10-31 14:38:11.198
!MESSAGE Indexed 'src2' (2,198 sources, 3,469 headers) in 1,644.21 sec:
1,068,877 declarations; 3,997,458 references; 0 unresolved inclusions; 0
syntax errors; 10,713 unresolved names (0.21%)
-sergey