[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: [PATCH v5 00/18] SMMUv3 nested translation support
From: |
Peter Maydell |
Subject: |
Re: [PATCH v5 00/18] SMMUv3 nested translation support |
Date: |
Fri, 19 Jul 2024 16:57:18 +0100 |
On Fri, 19 Jul 2024 at 16:36, Julien Grall <julien@xen.org> wrote:
>
> Hi,
>
> On 18/07/2024 10:43, Julien Grall wrote:
> > Hi Eric,
> >
> > On 17/07/2024 18:43, Eric Auger wrote:
> >> Hi Peter, Richard,
> >>
> >> On 7/17/24 17:09, Jean-Philippe Brucker wrote:
> >>> On Mon, Jul 15, 2024 at 08:45:00AM +0000, Mostafa Saleh wrote:
> >>>> Currently, QEMU supports emulating either stage-1 or stage-2 SMMUs
> >>>> but not nested instances.
> >>>> This patch series adds support for nested translation in SMMUv3,
> >>>> this is controlled by property “arm-smmuv3.stage=nested”, and
> >>>> advertised to guests as (IDR0.S1P == 1 && IDR0.S2P == 2)
> >>> For the whole series (3-9, 11, 12, 15, 16, 18):
> >>>
> >>> Reviewed-by: Jean-Philippe Brucker <jean-philippe@linaro.org>
> >>>
> >>> (and I think patch 16 is missing Eric's R-b)
> >>
> >> Jean-Philippe and I have followed up the progress of this series,
> >> Mostafa took into account all our comments and all the patches were
> >> reviewed. It seems to be in a pretty decent state so if you don't have
> >> any objection, please consider pulling it for 9.1.
> >>
> >> On my end I did some testing in non nesting mode with virtio-net/vhost
> >> and I have not noticed any regression.
> >> Would be nice if someone could send his T-b for the nested part though
> >> (Julien?).
> >
> > I haven't yet tried the latest version. I will do that in the next
> > couple of days.
> I see this is already merged. If this still matters:
>
> Tested-by: Julien Grall <jgrall@amazon.com>
We can't retrospectively add the tag, but the testing itself
is still important -- thanks for doing it.
Q: is there any reason not to:
(a) change the default to "nested" rather than "1"
(b) make the virt board (for new virt machine versions) use
"nested"?
AIUI "nested" should be a superset of "stage-1 only", the guest
can just ignore stage-2 if it doesn't care about it. Or is
there a performance hit from having stage-2 around even if the
guest doesn't enable it?
thanks
-- PMM
- [PATCH v5 14/18] hw/arm/smmu: Support nesting in the rest of commands, (continued)
- [PATCH v5 14/18] hw/arm/smmu: Support nesting in the rest of commands, Mostafa Saleh, 2024/07/15
- [PATCH v5 17/18] hw/arm/smmuv3: Support and advertise nesting, Mostafa Saleh, 2024/07/15
- [PATCH v5 15/18] hw/arm/smmuv3: Support nested SMMUs in smmuv3_notify_iova(), Mostafa Saleh, 2024/07/15
- [PATCH v5 18/18] hw/arm/smmu: Refactor SMMU OAS, Mostafa Saleh, 2024/07/15
- Re: [PATCH v5 00/18] SMMUv3 nested translation support, Jean-Philippe Brucker, 2024/07/17
- Re: [PATCH v5 00/18] SMMUv3 nested translation support, Eric Auger, 2024/07/17
- Re: [PATCH v5 00/18] SMMUv3 nested translation support, Peter Maydell, 2024/07/17
- Re: [PATCH v5 00/18] SMMUv3 nested translation support, Julien Grall, 2024/07/18
- Re: [PATCH v5 00/18] SMMUv3 nested translation support, Julien Grall, 2024/07/19
- Re: [PATCH v5 00/18] SMMUv3 nested translation support,
Peter Maydell <=
- Re: [PATCH v5 00/18] SMMUv3 nested translation support, Mostafa Saleh, 2024/07/20
- Re: [PATCH v5 00/18] SMMUv3 nested translation support, Peter Maydell, 2024/07/22