On 24-01-20, 09:31, Peter Ujfalusi wrote: > Vinod, Geert, > > On 24/01/2020 8.13, Vinod Koul wrote: > > On 22-01-20, 15:10, Vinod Koul wrote: > > > >> I like the idea of adding this in debugfs and giving more info, I would > >> actually love to add bytes_transferred and few more info (descriptors > >> submitted etc) to it... > >> > >>>> This way we will have all the information in one place, easy to look up > >>>> and you don't need to manage symlinks dynamically, just check all > >>>> channels if they have slave_device/name when they are in_use (in_use w/o > >>>> slave_device is 'non slave') > >>>> > >>>> Some drivers are requesting and releasing the DMA channel per transfer > >>>> or when they are opened/closed or other variations. > >>>> > >>>>> What do other people think? > >>> > >>> Vinod: do you have some guidance for your minions? ;-) > >> > >> > >> That said, I am not against merging this patch while we add more > >> (debugfs)... So do my minions agree or they have better ideas :-) > > > > So no new ideas, I am going to apply this and queue for 5.6, something > > is better than nothing. > > My only issue with the symlink is that it is created/removed on some > setups quite frequently as they request/release channel per transfer or > open/close. > It might be a small hit in performance, but it is going to be for them. > > > And I am looking forward for debugfs to give better picture, volunteers? > > Well, I still feel that the debugfs can give better view in one place > and in production it can be disabled to save few bytes per channel and > code is not complied in. > > If we have the debugfs we can remove some of the sysfs devices files > probably. Sure I dont mind if we move to something better :) We went from zero to something and can do better! Thanks -- ~Vinod
