On Wed, Dec 20, 2023 at 05:09:46PM +0100, Thierry Reding wrote: > On Wed, Dec 20, 2023 at 05:06:06PM +0100, Thierry Reding wrote: > > On Tue, 19 Dec 2023 16:30:23 +0000, Sean Young wrote: > > > The pwm-ir-tx driver has to turn the pwm signal on and off, and suffers > > > from delays as this is done in process context. Make this work in atomic > > > context. > > > > > > changes: > > > > > > v10: > > > - Uwe Kleine-König review comments/fixes > > > > > > [...] > > > > Applied, thanks! > > > > [6/6] media: pwm-ir-tx: Trigger edges from hrtimer interrupt context > > commit: 363d0e56285e80cda997d41d94c22313b673557d > > Hi Sean, > > I actually applied all of them (except I took patch 5 from v11), not > sure why b4 only kept track of patch 6. Great, thank you. I'm very pleased about this: the pwm-ir-tx driver now produces an prefect signal on rpi. The alternative driver, gpio-ir-tx, bitbangs and holds the cpu under spinlock for 100ms or however long the signal is, which is just a disaster (for e.g. real time linux). This gives us a way out. > Mauro didn't reply when I asked > whether he's okay with me picking this up into the PWM tree, but given > that you're the pwm-ir-tx maintainer I think that'll be fine. If not I > can back this out again. Mauro is not very responsive nowadays, but I'm sure it's fine. Sean
