> From: "Theo de Raadt"
> Date: Tue, 06 Aug 2024 23:04:02 -0600
>
> nvme resume is really crazy, since it does not believe the device is
> stopped, tries to use high-level operations to stop it and then restart
> it, but it ends up reusing queue structures from before.
Well, in sys/dev/ic/nvmere
Here is dmesg:
Il Mar 6 Ago 2024, 11:13 Mark Kettenis ha scritto:
> > From: Stefano Dalla Valentina
> > Date: Tue, 6 Aug 2024 07:43:40 +0200
> >
> > During the boot process with xenodm disabled even if inteldrm0 has
> > resolution of 1440x900 and wsdisplay0: screen 1-5 was created the screen
>
Mark Kettenis writes:
> Anyway, I think you're right in thinking that nvme_intr() needs some
> belt and suspenders. In nvme_shutdown() we delete the "normal"
> command queue, but nvme_intr() inconditionally looks at both of them.
> Now nvme_shutdown() masks the interrupt and nvme_resume() unmask
Greg Steuck wrote:
> Mark Kettenis writes:
>
> > Anyway, I think you're right in thinking that nvme_intr() needs some
> > belt and suspenders. In nvme_shutdown() we delete the "normal"
> > command queue, but nvme_intr() inconditionally looks at both of them.
> > Now nvme_shutdown() masks the i
"Theo de Raadt" writes:
> Greg Steuck wrote:
>
>> Mark Kettenis writes:
>>
>> > Anyway, I think you're right in thinking that nvme_intr() needs some
>> > belt and suspenders. In nvme_shutdown() we delete the "normal"
>> > command queue, but nvme_intr() inconditionally looks at both of them.
>
Hi Mark,
* Mark Kettenis wrote:
> > Date: Mon, 5 Aug 2024 22:23:17 +0200
> > From: Matthias Schmidt
>
> Hi Matthias,
>
> Can you try to narrow this down a bit further?
>
> You can either bisect by building kernels from intermediate dates
> yourself or use kernels from the snapshot archive at:
