Re: vector_capacity:test03() failing on mips-irix6

["Kaveh R. Ghazi" <ghazi at caip dot rutgers dot edu>]

 > From: Hans-Peter Nilsson <hp@bitrange.com>
 > 
 > On Fri, 13 Dec 2002, Kaveh R. Ghazi wrote:
 > > When I ran vector_capacity.exe under gdb, sure enough it dies in the
 > > new test03.  Here's a backtrace, any thoughts?
 > 
 > Yes, I think "thank you!" :-)
 > This is PR 8884.  See also the thread on gcc@ and libstdc++@
 > named "Regression(?) on the 3.2 branch:
 > 23_containers/vector_capacity.cc execution test"
 > brgds, H-P

Thanks.  

http://gcc.gnu.org/ml/gcc/2002-12/msg00541.html

BTW I completely agree with you in that thread that the libstdc++-v3
testcase file bundling is a problem, and I've suggested changing it in
the past to no avail.  Besides the fact that tracking which particular
test#() function failed is harder, this setup also obscures things
when the failing test function changes (e.g. from 03 to 06) and it
hides multiple failures since everything after the first crash doesn't
get run.

The only time its as good as separate files (i.e. clear reporting and
complete testing) is unfortunately when everything passes.

		--Kaveh
--
Kaveh R. Ghazi			ghazi@caip.rutgers.edu