Re: [PATCH] net: usb: lan78xx: Disable interrupts before calling generic_handle_irq()

[Stefan Wahren]

Hi Daniel,

Am 25.10.19 um 10:04 schrieb Daniel Wagner:
> ...
>
> Fixes: ed194d136769 ("usb: core: remove local_irq_save() around ->complete() handler")
> Cc: Woojung Huh <woojung.huh@xxxxxxxxxxxxx>
> Cc: Marc Zyngier <maz@xxxxxxxxxx>
> Cc: Andrew Lunn <andrew@xxxxxxx>
> Cc: Stefan Wahren <wahrenst@xxxxxxx>
> Cc: Jisheng Zhang <Jisheng.Zhang@xxxxxxxxxxxxx>
> Cc: Sebastian Andrzej Siewior <bigeasy@xxxxxxxxxxxxx>
> Cc: Thomas Gleixner <tglx@xxxxxxxxxxxxx>
> Cc: David Miller <davem@xxxxxxxxxxxxx>
> Signed-off-by: Daniel Wagner <dwagner@xxxxxxx>
> ---
>
> Hi,
>
> This patch just fixes the warning. There are still problems left (the
> unstable NFS report from me) but I suggest to look at this
> separately. The initial patch to revert all the irqdomain code might
> just hide the problem. At this point I don't know what's going on so I
> rather go baby steps. The revert is still possible if nothing else
> works.

did you ever see this pseudo lan78xx-irqs fire? I examined
/proc/interrupts on RPi 3B+ and always saw a 0.

FWIW you can have:

Tested-by: Stefan Wahren <wahrenst@xxxxxxx>

for this patch.

Regards
Stefan