Re: [Boost-users] [thread] Keeping leak detectors still... - Results

From: Daniel Krügler (dsp_at_[hidden])
Date: 2006-07-19 09:31:12

Hello,

I apologize for the late response to the original thread

"Re: [thread] Keeping leak detectors still..."

but we had some hard time to bring the boost tests to run
and I hope we did that correctly... Furtheron I would like to thank
my colleague Rüdiger Dreier for his patience for doing all
this testing work.
Please note, that the bjam boost tests where performed twice,
each one with compiler MSVC 8.00.50727.42 on Windows XP SP2 and
the official release of boost 1.33.1. The only difference was that the
original file libs/thread/src/tss_hooks.cpp was replaced by the version
posted by Peter Dimov on 2006-07-15T18:31 (where the leading and
trailing html tags have been removed). All 14 thread-related tests
passed the tests for both situations. If anyone is interested in the
complete test output (which includes the results of all other libs),
please don't resist to ask for them (I didn't want to pollute the news
group space without explicit invitation to do that...)

Additionally to the boost test, we run a homegrewn, simple
threading-mutex-locking test, which also passed successfully for the
old and new tss_hooks.cpp version. We could also observe that the 
standard leak reporter did *not* choke anymore ;-)

Greetings from Bremen,

Daniel Krügler