> > You can replace cape_universal by the libpruio universal overlay. That > doesn't enable drivers/subsystems (= saves power and resources > consumptions), but has the same pinmuxing capability. It's even more safe, > since it seems that cape_universal can damage your CPU by a sequence like > > config-pin P9_42 gpio high > config-pin P9_92 gpio low > > (I didn't test it, but if you do so, please report.) >
Ok, maybe, but any smart engineer should have pin isolation built into their circuitry. Here, we were using buffers, but now we're going to try bi powered FET's( sorry I'm not an EE so not sure that's the proper term ). But basically a MOSFET that has to be powered from both sides of the connection before the given "buffered" IO can complete it's circuit. Regarding other capes, libpruio ships with a tool to adapt the universal device tree overlay. It can generate overlays that do not claim a specified set of pins. Instead of fiddling with device tree entries, you just list the pins you want to get freed and let the tool deal with the low-level stuff. Such an overlay can get loaded before or after any other cape overlay. In order to replace the config_pin tool, you can write small programs (compiled against libpruio), which do the pinmuxing and enable the subsystems in use (only that ones). BR Here's the deal. I plan on creating a web interface for universal-io + config-pin. So a user can eventually open up the web page that comes with the beaglebone, and configure their IO / peripherals from a web front end. No idea if that is possible with your stuff, but more importantly, I've spent a good amount of my spare time looking into doing this with universal IO. Which my time is much more finite lately than in the past. So I can not afford to go around and research every possible way to do a thing, under the sun. I know universal IO well enough now to make this happen once I get the time to createthe web front end stuff. But I already have the back-end written. Well, I have the Bonejs wrapper library which took me only a few days a couple hours here and there . . .But the rest will take some time as I learn how to get data from the Nodejs backend, to a web front end, such as Angular, and I do not know what else right now . . . Also for what it's worth. You do not need cape_universal=enable does not need to be enabled in order to use config-pin, and universal IO. On Wed, Jul 20, 2016 at 11:06 AM, William Hermans <[email protected]> wrote: > If we create gpio/pinmux group, I think we could keep from burdening users > while moving Cloud9 IDE to the 'debian' user. I believe we also have a > sudoers for the 'debian' user, meaning we could probably at that point > prevent direct root login unless someone does something to disable the root > password. I'd worry about that breaking things like LabVIEW, etc., but if > we can at least try out some minor steps towards security, it will at least > make everyone more aware of the holes and challenges. > > I kind of roughly describe that here: > https://github.com/wphermans/Bonejs/blob/master/documentation/permissions.md > Although there is much mroe to consider than just the little bit I covered > there. But that should be a good start. > > On Wed, Jul 20, 2016 at 11:00 AM, Jason Kridner < > [email protected]> wrote: > >> >> >> On Fri, Jun 24, 2016 at 8:10 PM Charles Steinkuehler < >> [email protected]> wrote: >> >>> On 6/24/2016 5:52 PM, William Hermans wrote: >>> > /Note the security 'bar' is not set particularly high, given the/ >>> > /default BBB images have no root password. :)/ >>> > >>> > >>> > Thats been changed, since at least the last couple of images. >>> >> >> I don't think that's the case---we still have no root password, though >> one can be set. The Cloud9 IDE further doesn't require a login. >> >> If we create gpio/pinmux group, I think we could keep from burdening >> users while moving Cloud9 IDE to the 'debian' user. I believe we also have >> a sudoers for the 'debian' user, meaning we could probably at that point >> prevent direct root login unless someone does something to disable the root >> password. I'd worry about that breaking things like LabVIEW, etc., but if >> we can at least try out some minor steps towards security, it will at least >> make everyone more aware of the holes and challenges. >> >> >>> >>> Ahh...that's probably why you were getting the "askpass" errors. >>> >>> I haven't tried anything more recent than a few months ago. >>> >>> -- >>> Charles Steinkuehler >>> [email protected] >>> >>> -- >>> 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/d19cdd9b-9ae5-08a8-6028-cecbbea7d4f8%40steinkuehler.net >>> . >>> 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/CA%2BT6QPmBBQk0hy%2Bwz3ktPnaUYNoKC9aSvNK7y-xGDNc7gKaqUg%40mail.gmail.com >> <https://groups.google.com/d/msgid/beagleboard/CA%2BT6QPmBBQk0hy%2Bwz3ktPnaUYNoKC9aSvNK7y-xGDNc7gKaqUg%40mail.gmail.com?utm_medium=email&utm_source=footer> >> . >> >> 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/CALHSORp%3Dp5bycRkzmjcRzRAsPnE4%2B%3DDSZoPFFCjLN-bhxCEZGw%40mail.gmail.com. For more options, visit https://groups.google.com/d/optout.
