Re: [PATCH] iio: adc: exynos: do not rely on 'users' counter in ISR

2020-10-05 Thread dmitry . torokhov
On Mon, Oct 05, 2020 at 09:00:10PM +0200, Michał Mirosław wrote: > On Mon, Oct 05, 2020 at 10:36:36AM -0700, dmitry.torok...@gmail.com wrote: > > Hi Michał, > > > > On Mon, Oct 05, 2020 at 01:09:08PM +0200, Michał Mirosław wrote: > > > and breaking the loop will desync touch > > > state (I would g

Re: [PATCH] iio: adc: exynos: do not rely on 'users' counter in ISR

2020-10-05 Thread Michał Mirosław
On Mon, Oct 05, 2020 at 10:36:36AM -0700, dmitry.torok...@gmail.com wrote: > Hi Michał, > > On Mon, Oct 05, 2020 at 01:09:08PM +0200, Michał Mirosław wrote: > > On Sun, Oct 04, 2020 at 10:24:20PM -0700, dmitry.torok...@gmail.com wrote: > > > The order in which 'users' counter is decremented vs cal

Re: [PATCH] iio: adc: exynos: do not rely on 'users' counter in ISR

2020-10-05 Thread dmitry . torokhov
Hi Michał, On Mon, Oct 05, 2020 at 01:09:08PM +0200, Michał Mirosław wrote: > On Sun, Oct 04, 2020 at 10:24:20PM -0700, dmitry.torok...@gmail.com wrote: > > The order in which 'users' counter is decremented vs calling drivers' > > close() method is implementation specific, and we should not rely o

Re: [PATCH] iio: adc: exynos: do not rely on 'users' counter in ISR

2020-10-05 Thread Michał Mirosław
On Sun, Oct 04, 2020 at 10:24:20PM -0700, dmitry.torok...@gmail.com wrote: > The order in which 'users' counter is decremented vs calling drivers' > close() method is implementation specific, and we should not rely on > it. Let's introduce driver private flag and use it to signal ISR > to exit when

Re: [PATCH] iio: adc: exynos: do not rely on 'users' counter in ISR

2020-10-05 Thread Krzysztof Kozlowski
On Sun, Oct 04, 2020 at 10:24:20PM -0700, dmitry.torok...@gmail.com wrote: > The order in which 'users' counter is decremented vs calling drivers' > close() method is implementation specific, and we should not rely on > it. Let's introduce driver private flag and use it to signal ISR > to exit when