Review Request: Implemented multithreading in KFileItemModelSortAlgorithm

David Faure faure at kde.org
Sat Dec 8 09:09:44 GMT 2012


On Thursday 25 October 2012 13:48:55 Frank Reininghaus wrote:
> I can't really believe that a recursive implementation is slower than the
> loop-based one you posted here... (provided that you make sure that the
> number of invocations of QtConcurrent::run() is equal to the number of CPU
> cores, which means that you always have to do the second recursive
> sort()-call without QtConcurrent, as I said).

Hello from the "this doesn't make sense" police.
The above doesn't make sense :-)

QtConcurrent is perfectly capable of running a runnable in the main thread,
and does so automatically.

-- 
David Faure, faure at kde.org, http://www.davidfaure.fr
Working on KDE, in particular KDE Frameworks 5





More information about the kfm-devel mailing list