On 2/8/24 22:29, Linus Walleij wrote: > On Thu, Feb 8, 2024 at 5:19 PM Thomas Richard > <thomas.richard@xxxxxxxxxxx> wrote: >> On 1/28/24 01:12, Linus Walleij wrote: > >>> I guess you could define both pca953x_suspend() and >>> pca953x_suspend_noirq() and selectively bail out on one >>> path on some systems? >> >> Yes. >> >> What do you think if I use a property like for example "ti,pm-noirq" to >> select the right path ? >> Is a property relevant for this use case ? > > That's a Linux-specific property and that's useless for other operating > systems and not normally allowed. PM noirq is just some Linux thing. > > *FIRST* we should check if putting the callbacks to noirq is fine with > other systems too, and I don't see why not. Perhaps we need to even > merge it if we don't get any test results. > > If it doesn't work we can think of other options. I think all systems using a i2c controller which uses autosuspend should be impacted. I guess a patch (like I did in this series for i2c-omap [1]) should be applied for all i2c controller which use autosuspend. [1] https://lore.kernel.org/all/hqnxyffdsiqz5t43bexcqrwmynpjubxbzjchjaagxecso75dc7@y7lznovxg3go/ Regards, -- Thomas Richard, Bootlin Embedded Linux and Kernel engineering https://bootlin.com
