Messages in this thread | | | Date | Thu, 06 Aug 2009 14:55:51 +0900 | From | Tejun Heo <> | Subject | Re: [PATCH 2/2] sound: make OSS device number claiming optional |
| |
Hello, Alan.
Alan Cox wrote: >> 1. Merge the weird switch thing and the extra standard chrdev module >> alias patch > > Might as well fake the normal aliases in soundcore if loaded rather than > pollute chardev with it but otherwise I think we agree.
Yeap, that was what I meant by the second part of the sentence.
>> 2. Add to feature-removal that snd-slot/service-* are going away in a >> year along with the weird switches. This allows people who wish to >> try or switch in the meantime to do so. > > Yep > >> 3. After a year, drop module loading related code from sound_core >> along with the weird config option and kernel parameter. > > Do we need soundcore at all at that point ? It seems phase 3 is "move any > needed logic to ALSA oss emulation and kill it off"
Yeah, sure. There's nothing preventing ALSA from directly registering chrdevs.
>> In the end, the only choice we have to make is whether to keep >> snd-slot/service-* aliases. If we're gonna (I don't see why tho), the > > I think we need to for a year or so - and its trivial to do so. > >> cleanest way would be to teach chrdev about aliases. If not, the best >> way is to add a switch so that it can be phased out gradually. > > In the mean time if you are grabbing just some minors sound_core.c:chains > also needs keeping in sync with direct character range grabs because your > CUSE based device might grab some minors before sound_core, and then it > appears tears will result ?
Hmmm... okay, will take a deeper look at that.
Thanks.
-- tejun
| |