Re: [Xen-devel] [PATCH v4 7/8] xen/arm: introduce nr_spis

[Julien Grall <julien.grall@xxxxxxx>]

Hi,

On 25/09/2019 17:16, Stefano Stabellini wrote:
> On Wed, 25 Sep 2019, Julien Grall wrote:
> > Hi,
> >
> > On 24/09/2019 18:56, Stefano Stabellini wrote:
> > > On Wed, 11 Sep 2019, Julien Grall wrote:
> > > > Hi Stefano,
> > > >
> > > > On 8/21/19 4:53 AM, Stefano Stabellini wrote:
> > > > > We don't have a clear way to know how many virtual SPIs we need for the
> > > > > dom0-less domains. Introduce a new option under xen,domain to specify
> > > > > the number of SPIs to allocate for a domain.
> > > > >
> > > > > The property is optional. When absent, we'll use the physical number of
> > > > > GIC lines for dom0-less domains, just like for dom0. Given that
> > > > > dom0-less VMs are meant for static partitioning scenarios where the
> > > > > number of VMs is very low, increased memory overhead should not be a
> > > > > problem, and it is possible to minimize it using "nr_spis".
> > > > >
> > > > > Remove the old setting of nr_spis based on the presence of the vpl011.
> > > > I am afraid this still does not explain the implications of this patch to
> > > > current setup (with and without VPL011).
> > > >
> > > > For instance, with your change, VPL011 may not work anymore. Imagine we
> > > > decide
> > > > to push the vpl011 interrupt towards the end of the Interrupt ID space
> > > > (i.e.
> > > > 1019).
> > > >
> > > > I don't think we want the user to have to select nr_spis by himself for
> > > > this
> > > > case.
> > > >
> > > > Regarding the change without vpl011, this is not explained why all the
> > > > domains
> > > > (even the one without SPIs routed) will have SPIs exposed. For instance,
> > > > if
> > > > you were to expose 256 interrupts for 4 domains, this will roughly use
> > > > 80KB of
> > > > memory. I don't think this is what you had in mind as "low footprint".
> > >    What do you think of the following:
> > >
> > > The implication of this change is that without nr_spis dom0less domains
> > > get the same amount of SPI allocated as dom0, regardless of how many
> > > physical devices they have assigned, and regardless of whether they have
> > > a virtual pl011 (which also needs an emulated SPI).
> > >
> > > When nr_spis is present, the domain gets exactly nr_spis allocated SPIs.
> > > If the number is too low, it might not be enough for the devices
> > > assigned it to it. If the number is less than GUEST_VPL011_SPI, the
> > > virtual pl011 won't work.
> > This is good to address my first remark. How about the others?
>   
> For your point about "VPL011 may not work anymore", are you suggesting
> we print a warning or that we always allocate at least
> GUEST_VPL011_SPI+1 SPIs if vpl011 is requested?
My preference is to do the later as this match the behavior when guest are 
created by libxl. This is also the current behavior.
But if you want to change this behavior, then you need to document it as this is 
a breakage between the previous interface.
> For your point about "it is not explained why all the domains will have
> SPIs exposed" would you like me to add a sentence to the commit message
> to make it clearer or do you have something else in mind?
A sentence in the commit message is be good enough.

Cheers,

--
Julien Grall

_______________________________________________
Xen-devel mailing list
Xen-devel@xxxxxxxxxxxxxxxxxxxx
https://lists.xenproject.org/mailman/listinfo/xen-devel