[PATCH v4] um: Enable preemption in UML

[Anton Ivanov]

On 21/09/2023 22:55, Richard Weinberger wrote:
> ----- Ursprüngliche Mail -----
>> Von: "anton ivanov" <anton.ivanov at cambridgegreys.com>
>> An: "richard" <richard at nod.at>
>> CC: "linux-um" <linux-um at lists.infradead.org>, "Johannes Berg" <johannes at sipsolutions.net>
>> Gesendet: Donnerstag, 21. September 2023 23:41:16
>> Betreff: Re: [PATCH v4] um: Enable preemption in UML
>> On 21/09/2023 22:02, Richard Weinberger wrote:
>>> ----- Ursprüngliche Mail -----
>>>> Von: "anton ivanov" <anton.ivanov at cambridgegreys.com>
>>>> An: "linux-um" <linux-um at lists.infradead.org>
>>>> CC: "Johannes Berg" <johannes at sipsolutions.net>, "richard" <richard at nod.at>,
>>>> "anton ivanov"
>>>> <anton.ivanov at cambridgegreys.com>
>>>> Gesendet: Donnerstag, 21. September 2023 17:55:22
>>>> Betreff: [PATCH v4] um: Enable preemption in UML
>>>> From: Anton Ivanov <anton.ivanov at cambridgegreys.com>
>>>>
>>>> Preemption requires saving/restoring FPU state. This patch
>>>> adds support for it using GCC intrinsics.
>>> This patch triggers here the following splat:
>> Ack. I will look into it tomorrow.
> Thx!
> I had a brief look, UML's init_IRQ() seems to return now with interrupts enabled.
> Dunno why, I need some sleep now.

Tlb. Looking at other architectures (and specifically x86) it needs preempt_disable/enable around critical sections.

 From there you may end up in the wrong state across the board.

>
>> There will be places where it will bomb out. F.E. I am chasing one is
>> triggered by the NFS server (it may be the same as in this trace - it
>> happens around the end of the initial setup).
>>
>> All in all it looks better than I expected.
> True that! :-)
>
> Thanks,
> //richard
>
> _______________________________________________
> linux-um mailing list
> linux-um at lists.infradead.org
> http://lists.infradead.org/mailman/listinfo/linux-um

-- 
Anton R. Ivanov
https://www.kot-begemot.co.uk/