On Wed, 2014-01-22 at 10:01 +0100, Johannes Berg wrote: > On Wed, 2014-01-22 at 09:54 +0100, Michal Kazior wrote: > > > I'd say Luca's approach has more future, as it makes channel context > > > support native. With this, I don't even see how you *could* later add > > > real channel context support. A function like this, that essentially > > > assumes exactly a single chanctx that might be doing CSA and similar > > > seems to make that rather difficult. > > > > I'm not really aware of Luca's work. > > I guess he should post it :) Yes, I have an internal patch that I wrote last year as a proof of concept. I have cut some corners (eg. the non-chanctx case Johannes mentioned), so I need to rework it a bit before posting here. > > > In any case, unless I'm missing some bigger picture this seems like a > > > really hacky way to add things, basically ignoring all the channel > > > context work. Since I care more about channel contexts, maybe we should > > > merge support for channel context CSA before we try to do the multi > > > thing. > > > > This is, I assume, Luca's work you're talking about here again. What's > > the progress on this? Any time estimates? > > I think he has patches but they break the non-chanctx case right now... I'll post the patch (as an early RFC) and explain the concept as soon as I manage to catch up with Michal's patches and all the discussions. ;) -- Cheers, Luca. -- To unsubscribe from this list: send the line "unsubscribe linux-wireless" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
