Re: [PATCH] genirq: Deduplicate WARN_ON_ONCE() in generic_handle_domain_irq()

[Lukas Wunner <lukas@xxxxxxxxx>]

On Mon, May 16, 2022 at 08:29:50AM +0200, Sebastian Andrzej Siewior wrote:
> From looking at gpio-dln2
> this is called from USB URB's callback which is softirq. In the end
> dln2_gpio_event() is invoked while dln2_dev::event_cb_lock is acquired.
> That lock is acquired by disabling interrupts which is what gets the
> locking right for generic_handle_domain_irq(). If that lock lifted to
> spin_lock_bh() (because it is always in urb's calback context and all
> HCDs complete in one context unlike now) then this breaks.

I think you want WARN_ON(!irqs_disabled) to catch that,
not WARN_ON_ONCE(!in_hardirq()).


> and sadly I missed dln2. Please let me know if you have more users
> similar to dln2. I will add those to my list once upstream buys that
> interface.

There's another USB GPIO irqchip queued up for 5.19 now:
https://git.kernel.org/netdev/net-next/c/1ce8b37241ed

And there's one more in drivers/hid/hid-cp2112.c, but that calls
handle_nested_irq(), which disables hardirqs.

Thanks,

Lukas