On Tue, May 23, 2017 at 8:37 PM, jmondi <jacopo@xxxxxxxxxx> wrote: >> I did not follow too much. >> But it seems IMX7ULP/Vybrid to be also a fan of generic >> output-enable/input-enable >> property. >> >> See: >> Figure 5-2. GPIO PAD in Page 241 >> http://www.nxp.com/assets/documents/data/en/reference-manuals/VFXXXRM.pdf >> >> It has separate register bits to control input buffer enable and >> output buffer enable >> and we need set it property for GPIO function. > > As it seems we have another user for 'output-enable' here, what if we just > add that one to the generic bindings properties list, and we keep > 'bi-directional' (which seems to be the most debated property we have > added) out of generic properties? > > We can handle 'bi-directional' pins with static tables in our pin > controller driver and not have it anywhere in DT. This sounds like a viable approach. I just want to know if "output-enable" is the right name? "output-buffer-enable"? > I see commit 42d5a11200d0[1] has not been reverted yet as Andy asked > in some previous email. I'm just overloaded. I sent that revert to Torvalds today. > I can send another version of that patch with > only 'output-enable' if you wish. That's what we want. > Once we reach consesus, I can then send v6 of our pin controller driver > based on that. OK sounds like a plan. Sorry for the mess, I'm just trying to get this right :/ Yours, Linus Walleij -- To unsubscribe from this list: send the line "unsubscribe devicetree" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
