[PATCH RFC] genirq/kexec: Do not take the bus lock of an interrupt chip on the crash path

[Igor Velkov]

On Mon, 05 Oct 2026 01:41:55 +0000, sashiko-bot at kernel.org wrote:
> If an irqchip implements irq_bus_sync_unlock but omits irq_bus_lock, this
> condition evaluates to true and irq_set_irqchip_state() is called.
[...]
> Should this check also ensure !chip->irq_bus_sync_unlock?

Yes. The next version will test both callbacks, as irq_supports_nmi()
does. No chip in tip/irq/core sets only one of them, as far as I can
grep, so nothing changes for current users.

Igor