I hear where you are coming from and I think you have a valid argument here. First, I think this mailing list is the wrong forum to discuss this issue and second, I’m not sure which driver needs to be updated to implement this feature. I think from our last discussion gpiolib seemed like the right place to implement this feature, but pinctl-single could also be the right place. I haven’t looked into this for a while.
I would suggest posting a RFC to the Linux or Linux-ARM mailing list and see what feedback you get. Any changes here will affect all architectures so best to iron this out with all the stakeholders. Regards, John > On Jul 21, 2016, at 8:39 PM, Jelena Freiherr <[email protected]> wrote: > > Am Donnerstag, 21. Juli 2016 07:35:23 UTC+2 schrieb john3909: >> >> Why not use the A & B suffix? Surely this just needs modification to your >> regular expressions? >> > > Just renaming doesn't solve the issue. There's just one P9_42 and the > features of P9_92 should get merged in to this pin. Check the libpruio > overlay on how to handle this in a save manner. > > But the point is: both (cape_universal and libpruio overlays) are just > bad workarounds. They load a lot of pin configurations in kernel space > memory, while just a few are used. A full featured dtbo file (67 header > pins) has 135140 bytes. Handling all that pre-configurations slows down > boot time and blocks a lot of kernel memory. And finally the > configuration necessary for a special usecase is still missing. > > From my point of view the kernel driver pinctrl-single should get patched: > > 1. It should protect the system (when one CPU ball gets set, the > other should automatically get in input state - for CPU balls > connected to P9_41 and P9_42). > 2. It should get an API in /sys/devices/ocp to set any pin > configuration (as stated in our March discussion > <https://groups.google.com/d/msg/beagleboard/ZTKntXOXGyc/GLBOQ2r5BQAJ>). > > This API would be the correct basis for Williams web interface (and for > config-pin or libpruio pinmuxing). > > BR > > -- > For more options, visit http://beagleboard.org/discuss > --- You received this message because you are subscribed to a topic in > the Google Groups "BeagleBoard" group. > To unsubscribe from this topic, visit > https://groups.google.com/d/topic/beagleboard/2D5Pz3r7ZZ8/unsubscribe. > To unsubscribe from this group and all its topics, send an email to > [email protected]. > To view this discussion on the web visit > https://groups.google.com/d/msgid/beagleboard/9ff475ab-0f97-4fc0-906c-49d989b8ef8a%40googlegroups.com. > For more options, visit https://groups.google.com/d/optout. -- For more options, visit http://beagleboard.org/discuss --- You received this message because you are subscribed to the Google Groups "BeagleBoard" group. To unsubscribe from this group and stop receiving emails from it, send an email to [email protected]. To view this discussion on the web visit https://groups.google.com/d/msgid/beagleboard/BF76F439-CD96-4E99-A35E-E5531C07E881%40gmail.com. For more options, visit https://groups.google.com/d/optout.
