On Sat, Sep 10, 2022 at 9:50 PM Hans de Goede <hdegoede@xxxxxxxxxx> wrote: > > Hi Patrik, > > On 9/9/22 10:45, Hans de Goede wrote: > > Hi, > > > > On 9/9/22 09:34, Patrik Jakobsson wrote: > >> On Thu, Sep 8, 2022 at 3:39 PM Hans de Goede > >> <hdegoede@xxxxxxxxxx> wrote: > >>> > >>> Hi, > >>> > >>> On 9/8/22 15:26, Patrik Jakobsson wrote: > > <snip> > > >>>> On Poulsbo I can see > >>>> interrupts not getting handled during suspend/resume even with this > >>>> patch applied. > >>> > >>> "during" ? I guess you mean _after_ a suspend/resume ? > >> > >> Yes. I get: irq 16: nobody cared (try booting with the "irqpoll" option). > >> But perhaps the system is just too slow to respond. > > So I've just tested on a Sony Vaio VPCX11S1E (Atom Z540 with PSB graphics) > and with my entire patch-set (did not test without) suspend/resume works > fine there without any "irq xx: nobody cared" messages. Your patch fixes the problem on my FitPC (Atom Z530) as well. Great! I have the Acer AOD270 so I can look further into the MSI problem. But as you say, the current solution might be what we want. -Patrik > > Note that on the Vaio VPCX11S1E the gma500 is using irq 18 rather then > 16 and that irq is not shared with anything. So I wonder if this is > related to the irq being shared? > > Regards, > > Hans >