Re: [Xen-devel] [PATCH v12 6/6] introduce a 'passthrough' configuration option to xl.cfg...

[Paul Durrant <Paul.Durrant@xxxxxxxxxx>]

> -----Original Message-----
> From: George Dunlap <george.dunlap@xxxxxxxxxx>
> Sent: 16 September 2019 11:15
> To: Paul Durrant <Paul.Durrant@xxxxxxxxxx>; xen-devel@xxxxxxxxxxxxxxxxxxxx
> Cc: Jan Beulich <jbeulich@xxxxxxxx>; Christian Lindig 
> <christian.lindig@xxxxxxxxxx>; Ian Jackson
> <Ian.Jackson@xxxxxxxxxx>; Wei Liu <wl@xxxxxxx>; Andrew Cooper 
> <Andrew.Cooper3@xxxxxxxxxx>; George
> Dunlap <George.Dunlap@xxxxxxxxxx>; Julien Grall <julien.grall@xxxxxxx>; 
> Konrad Rzeszutek Wilk
> <konrad.wilk@xxxxxxxxxx>; Stefano Stabellini <sstabellini@xxxxxxxxxx>; Tim 
> (Xen.org) <tim@xxxxxxx>;
> Anthony Perard <anthony.perard@xxxxxxxxxx>; David Scott <dave@xxxxxxxxxx>; 
> Volodymyr Babchuk
> <Volodymyr_Babchuk@xxxxxxxx>; Roger Pau Monne <roger.pau@xxxxxxxxxx>
> Subject: Re: [PATCH v12 6/6] introduce a 'passthrough' configuration option 
> to xl.cfg...
> 
> On 9/16/19 10:27 AM, Paul Durrant wrote:
> > ...and hence the ability to disable IOMMU mappings, and control EPT
> > sharing.
> >
> > This patch introduces a new 'libxl_passthrough' enumeration into
> > libxl_domain_create_info. The value will be set by xl either when it parses
> > a new 'passthrough' option in xl.cfg, or implicitly if there is passthrough
> > hardware specified for the domain.
> >
> > If the value of the passthrough configuration option is 'disabled' then
> > the XEN_DOMCTL_CDF_iommu flag will be clear in the xen_domctl_createdomain
> > flags, thus allowing the toolstack to control whether the domain gets
> > IOMMU mappings or not (where previously they were globally set).
> >
> > If the value of the passthrough configuration option is 'sync_pt' then
> > a new 'iommu_opts' field in xen_domctl_createdomain will be set with the
> > value XEN_DOMCTL_IOMMU_no_sharept. This will override the global default
> > set in iommu_hap_pt_share, thus allowing the toolstack to control whether
> > EPT sharing is used for the domain.
> >
> > If the value of passthrough is 'enabled' then xl will choose an appropriate
> > default according to the type of domain and hardware support.
> 
> Minor suggestion: I prefer using a word like "auto" when you're letting
> the computer decide something.

Not sure I like that. I'll wait for a second opinion on that one.

> 
> I'd also...
> 
> > diff --git a/docs/man/xl.cfg.5.pod.in b/docs/man/xl.cfg.5.pod.in
> > index c99d40307e..154d847fb9 100644
> > --- a/docs/man/xl.cfg.5.pod.in
> > +++ b/docs/man/xl.cfg.5.pod.in
> > @@ -605,6 +605,62 @@ option should only be used with a trusted device tree.
> >  Note that the partial device tree should avoid using the phandle 65000
> >  which is reserved by the toolstack.
> >
> > +=item B<passthrough="STRING">
> > +
> > +Specify whether IOMMU mappings are enabled for the domain and hence whether
> > +it will be enabled for passthrough hardware. Valid values for this option
> > +are:
> > +
> > +=over 4
> > +
> > +=item B<disabled>
> > +
> > +IOMMU mappings are disabled for the domain and so hardware may not be
> > +passed through.
> > +
> > +This option is the default if no passthrough hardware is specified in the
> > +domain's configuration.
> > +
> > +=item B<sync_pt>
> > +
> > +This option means that IOMMU mappings will be synchronized with the
> > +domain's P2M table as follows:
> > +
> > +For a PV domain, all writable pages assigned to the domain are identity
> > +mapped by MFN in the IOMMU page table. Thus a device driver running in the
> > +domain may program passthrough hardware for DMA using MFN values
> > +(i.e. host/machine frame numbers) looked up in its P2M.
> > +
> > +For an HVM domain, all non-foreign RAM pages present in its P2M will be
> > +mapped by GFN in the IOMMU page table. Thus a device driver running in the
> > +domain may program passthrough hardware using GFN values (i.e. guest
> > +physical frame numbers) without any further translation.
> > +
> > +This option is not currently available on Arm.
> > +
> > +=item B<share_pt>
> > +
> > +This option is unavailable for a PV domain. For an HVM domain, this option
> > +means that the IOMMU will be programmed to directly reference the domain's
> > +P2M table as its page table. From the point of view of a device driver
> > +running in the domain this is functionally equivalent to B<sync_pt> but
> > +places less load on the hypervisor and so should generally be selected in
> > +preference. However, the availability of this option is hardware specific.
> > +If B<xl info> reports B<virt_caps> containing B<iommu_hap_pt_share> then
> > +this option may be used.
> > +
> > +=item B<enabled>
> > +
> > +This option enables IOMMU mappings and selects an appropriate default
> > +operating mode. For HVM domains running on platforms where the option is
> > +available, this is equivalent to B<share_pt>. Otherwise, and also for PV
> > +domains, this options is equivalent to B<sync_pt>.
> 
> ...put the option we want / expect people nearer the top (either first
> or second).
> 

Ok, I'll do that.

  Paul

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