On Tue, Oct 22, 2024 at 6:28 PM Conor Dooley <conor@xxxxxxxxxx> wrote: > On Wed, Oct 16, 2024 at 08:42:51PM +0100, Conor Dooley wrote: > > On Wed, Oct 16, 2024 at 09:26:13PM +0200, Linus Walleij wrote: > > > On Wed, Oct 16, 2024 at 12:29 PM Conor Dooley <conor@xxxxxxxxxx> wrote: > > > > > > > What does bring a nice simplification though, IMO, is regmap. I am > > > > pretty sure that using it was one of the suggestions made last time > > > > Lewis submitted this - so I think I'm going to do that instead. > > > > > > If you have the time. Using GPIO_REGMAP for MMIO is not that > > > common and I think the driver is pretty neat as it stands. > > > > As with using the common MMIO stuff, I don't think GPIO_REGMAP provides > > that much value as I cannot use the direction stuff from it. I was > > thinking of using regmap directly, like: > > https://git.kernel.org/pub/scm/linux/kernel/git/conor/linux.git/commit/?h=gpio-no-irq&id=c8933e1e3600e3fa29efe28fbb2e343e133f9d67 > > which I think reduces how ugly the two direction functions look. > > Sorry to bother you Linus, but I was hoping to see some sort of comment > here before I squash this stuff and submit a new version. Is something > like what I linked above acceptable? I pretty much trust your judgement on this, I'm fine with either solution, the patch is also perfectly fine already as it is unless you want to polish it further. Yours, Linus Walleij