Messages in this thread | | | Date | Thu, 18 Jan 2018 10:06:14 +0100 (CET) | From | Thomas Gleixner <> | Subject | Re: [BUG 4.15-rc7] IRQ matrix management errors |
| |
On Thu, 18 Jan 2018, Keith Busch wrote:
> On Thu, Jan 18, 2018 at 09:10:43AM +0100, Thomas Gleixner wrote: > > Can you please provide the output of > > > > # cat /sys/kernel/debug/irq/irqs/$ONE_I40_IRQ > > # cat /sys/kernel/debug/irq/irqs/48 > handler: handle_edge_irq > device: 0000:1a:00.0 > status: 0x00000000 > istate: 0x00000000 > ddepth: 0 > wdepth: 0 > dstate: 0x05401200 > IRQD_ACTIVATED > IRQD_IRQ_STARTED > IRQD_SINGLE_TARGET > IRQD_AFFINITY_SET > IRQD_CAN_RESERVE > node: 0 > affinity: 0-27,56-83 > effectiv: 0 > pending: > domain: PCI-MSI-2 > hwirq: 0xd00002 > chip: PCI-MSI > flags: 0x10 > IRQCHIP_SKIP_SET_WAKE
Bah. That's MSI where we can't use interrupt reservation because we cant mask MSI vectors. I really admire these designed by committee hardware bogosities.
Thanks,
tglx
| |