Date: 2001-03-15 13:21:10
--- In boost_at_y..., Beman Dawes <bdawes_at_a...> wrote:
> At 12:12 PM 3/15/2001, Dan Nuffer wrote:
> >... Now if the Xerces
> >library had written those two functions in the header files, they
> >would've been inlined and the app would've gained at least 15%
> >improvement. The extra layers effectively doubled the time spent
> >locking/unlocking mutexes.
> >I can understand your not wanting to include windows.h in the
> >file, but I think you're overstating the negative impact. If
> >writing a program to run on windows, odds are you've already
> >windows.h and then there's pre-compiled headers with MSVC. For
> >platforms, including pthreads.h is trivial.
> >Maybe you could take the approach of using a .ipp file, where if
> >user wants a minimal header they just include the .hpp, but if
> >speed, to have things inlined they include the .hpp and
the .ipp. Thus
> >the user has the option to choose.
> There may come a time when 15% difference is a big number. But
> we need to focus first on the interface issues and functionality.
> an interface we feel comfortable with? Ditto functionality. Have
> safety issues been addressed? Then there are issues of correct
> Eventually, after all those hurdles are passed, we can bother the
> developers about 15% performance gains. But for now let's try to
> focused on the larger questions.
> Just my opinion, of course, and not meant to be critical of Dan's
Good point on top of those I tried to make myself. Considering this
point, debate about the implementation detail of the pimpl idiom will
no longer be considered by myself at this time. It's easily enough
addressed later if found to be important. Let's focus on the design
and the implementation bugs.
Boost list run by bdawes at acm.org, gregod at cs.rpi.edu, cpdaniel at pacbell.net, john at johnmaddock.co.uk