On Sun, Apr 13, 2025 at 09:00:23PM +0100, Marc Zyngier wrote: > On Sun, 13 Apr 2025 17:57:35 +0100, > Manivannan Sadhasivam <manivannan.sadhasivam@xxxxxxxxxx> wrote: > > > > On Tue, Apr 01, 2025 at 10:17:01AM +0100, Marc Zyngier wrote: > > > From: Janne Grunau <j@xxxxxxxxxx> > > > > > > Iterating over disabled ports results in of_irq_parse_raw() parsing > > > the wrong "interrupt-map" entries, as it takes the status of the node > > > > 'as it doesn't take account'? > > > > > into account. > > No, I really mean it in the positive form. of_irq_parse_raw() checks > of_device_is_available(), and gets really confused if walking from a > disabled port. You end up with the interrupt for the next *available* > port, and everything goes pear shaped from then onwards. > Ah okay. > So IMO "as it takes into account" describes pretty accurately the > situation. > Thanks for the clarification. - Mani -- மணிவண்ணன் சதாசிவம்
