Changes to CodeCompletionWorker regarding multithreading

[Aleix Pol]

On Mon, Jan 17, 2011 at 6:45 PM, Andreas Pakulat <apaku at gmx.de> wrote:

> On 17.01.11 17:19:21, David Nolden wrote:
> > 2011/1/17 Milian Wolff <mail at milianw.de>:
> > > David Nolden, 17.01.2011:
> > >> Hi, it seems like when CodeCompletionWorker etc. were ported to
> > >> kdevplatform, someone completely broke the multi-threading. The result
> > >> was that for a year now, we are computing all completion items in the
> > >> foreground thread (ouch).
> > >
> > > How come this was not noticed before?
> > >
> > >> I have fixed this now in kdevplatform and cpp, the other languages
> > >> might need a slight adaption. In case you're overriding
> > >> CodeCompletionWorker::computeCompletions, then you have to make sure
> > >> to match the changed signature of the function you're overriding. The
> > >> few other changes should be self-explanatory from the compiler errors.
> > >>
> > >> Now the question is, if we want to backport this to KDevelop 4.2. IMO
> > >> the broken multi-threading was a quite serious regression which is
> > >> fixed now. The problem is that we're using multi-threading now where
> > >> it wasn't used before, which might expose multi-threading bugs we
> > >> created within the last year, and which weren't visible without
> > >> multi-threading.
> > >
> > > It seems this requires API changes, hence I'm not in favor of
> backporting this
> > > now that we are in RC state.
> > >
> > > I know, this sucks, but I don't see any other way.
> >
> > I don't think the binary compatibility is a useful metric for this
> > decision, it doesn't make any sense (Unlike kdelibs, we're not a
> > library). We should only worry about the stability..
>
> Thats exactly the attitude that makes the situation stay as it is and
> why the libraries in kdevplatform are still not used by any project
> except kdevelop.
>
> Apart from that, a change that has unknown side-effects in a stable
> release is IMHO just a no-go. Nobody forces us to wait 6 months or more
> for the next release if that is your concern. Its also not a regression,
> there's been no release with multi-threaded code-completion.
>
> And apparently for the last year or so, nobody took too much notice of
> this to actually try to find out whats wrong.
>
> Andreas
>
> --
> You will experience a strong urge to do good; but it will pass.
>
> --
> KDevelop-devel mailing list
> KDevelop-devel at kdevelop.org
> https://barney.cs.uni-potsdam.de/mailman/listinfo/kdevelop-devel
>

Yes, let's just try to release some kdev4.3 soon :)

Aleix
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.kde.org/pipermail/kdevelop-devel/attachments/20110117/5725c889/attachment.html>