| Message ID | 20220505001656.395419-7-sstabellini@kernel.org (mailing list archive) |
|---|---|
| State | Superseded |
| Headers | show |
| Series | dom0less PV drivers | expand |
Hi, On 05/05/2022 01:16, Stefano Stabellini wrote: > From: Stefano Stabellini <stefano.stabellini@xilinx.com> > > Document how to use the feature and how the implementation works. > > Signed-off-by: Stefano Stabellini <stefano.stabellini@xilinx.com> > Reviewed-by: Luca Fancellu <luca.fancellu@arm.com> AFAICT, this match the code in Xen. So: Acked-by: Julien Grall <jgrall@amazon.com> > --- > docs/features/dom0less.pandoc | 43 ++++++++++++++++++++++++++++++++--- > 1 file changed, 40 insertions(+), 3 deletions(-) > > diff --git a/docs/features/dom0less.pandoc b/docs/features/dom0less.pandoc > index c9edb529e1..725afa0558 100644 > --- a/docs/features/dom0less.pandoc > +++ b/docs/features/dom0less.pandoc > @@ -90,6 +90,46 @@ Otherwise, they may be unusable in Xen (for instance if they are compressed). > > See docs/misc/arm/device-tree/booting.txt for more information. > > +PV Drivers > +---------- > + > +It is possible to use PV drivers with dom0less guests with some > +restrictions: > + > +- dom0less domUs that want to use PV drivers support should have the > + "xen,enhanced" property set under their device tree nodes (see > + docs/misc/arm/device-tree/booting.txt) > +- a dom0 must be present (or another domain with enough privileges to > + run the toolstack) > +- after dom0 is booted, the utility "init-dom0less" must be run > +- do not run "init-dom0less" while creating other guests with xl > + > +After the execution of init-dom0less, it is possible to use "xl" to > +hotplug PV drivers to dom0less guests. E.g. xl network-attach domU. > + > +The implementation works as follows: > +- Xen allocates the xenstore event channel for each dom0less domU that > + has the "xen,enhanced" property, and sets HVM_PARAM_STORE_EVTCHN > +- Xen does *not* allocate the xenstore page and sets HVM_PARAM_STORE_PFN > + to ~0ULL (invalid) > +- Dom0less domU kernels check that HVM_PARAM_STORE_PFN is set to invalid > + - Old kernels will continue without xenstore support (Note: some old > + buggy kernels might crash because they don't check the validity of > + HVM_PARAM_STORE_PFN before using it! Disable "xen,enhanced" in > + those cases) > + - New kernels will wait for a notification on the xenstore event > + channel (HVM_PARAM_STORE_EVTCHN) before continuing with the > + initialization > +- Once dom0 is booted, init-dom0less is executed: > + - it allocates the xenstore shared page and sets HVM_PARAM_STORE_PFN > + - it calls xs_introduce_domain > +- Xenstored notices the new domain, initializes interfaces as usual, and > + sends an event channel notification to the domain using the xenstore > + event channel (HVM_PARAM_STORE_EVTCHN) > +- The Linux domU kernel receives the event channel notification, checks > + HVM_PARAM_STORE_PFN again and continue with the initialization > + > + > Limitations > ----------- > > @@ -107,9 +147,6 @@ limitations: > information, the GIC version exposed to the domains started by Xen at > boot is the same as the native GIC version. > > -- No PV drivers. There is no support for PV devices at the moment. All > - devices need to be statically assigned to guests. > - > - Pinning vCPUs of domains started by Xen at boot can be > done from the control domain, using `xl vcpu-pin` as usual. It is not > currently possible to configure vCPU pinning without a control domain.
diff --git a/docs/features/dom0less.pandoc b/docs/features/dom0less.pandoc index c9edb529e1..725afa0558 100644 --- a/docs/features/dom0less.pandoc +++ b/docs/features/dom0less.pandoc @@ -90,6 +90,46 @@ Otherwise, they may be unusable in Xen (for instance if they are compressed). See docs/misc/arm/device-tree/booting.txt for more information. +PV Drivers +---------- + +It is possible to use PV drivers with dom0less guests with some +restrictions: + +- dom0less domUs that want to use PV drivers support should have the + "xen,enhanced" property set under their device tree nodes (see + docs/misc/arm/device-tree/booting.txt) +- a dom0 must be present (or another domain with enough privileges to + run the toolstack) +- after dom0 is booted, the utility "init-dom0less" must be run +- do not run "init-dom0less" while creating other guests with xl + +After the execution of init-dom0less, it is possible to use "xl" to +hotplug PV drivers to dom0less guests. E.g. xl network-attach domU. + +The implementation works as follows: +- Xen allocates the xenstore event channel for each dom0less domU that + has the "xen,enhanced" property, and sets HVM_PARAM_STORE_EVTCHN +- Xen does *not* allocate the xenstore page and sets HVM_PARAM_STORE_PFN + to ~0ULL (invalid) +- Dom0less domU kernels check that HVM_PARAM_STORE_PFN is set to invalid + - Old kernels will continue without xenstore support (Note: some old + buggy kernels might crash because they don't check the validity of + HVM_PARAM_STORE_PFN before using it! Disable "xen,enhanced" in + those cases) + - New kernels will wait for a notification on the xenstore event + channel (HVM_PARAM_STORE_EVTCHN) before continuing with the + initialization +- Once dom0 is booted, init-dom0less is executed: + - it allocates the xenstore shared page and sets HVM_PARAM_STORE_PFN + - it calls xs_introduce_domain +- Xenstored notices the new domain, initializes interfaces as usual, and + sends an event channel notification to the domain using the xenstore + event channel (HVM_PARAM_STORE_EVTCHN) +- The Linux domU kernel receives the event channel notification, checks + HVM_PARAM_STORE_PFN again and continue with the initialization + + Limitations ----------- @@ -107,9 +147,6 @@ limitations: information, the GIC version exposed to the domains started by Xen at boot is the same as the native GIC version. -- No PV drivers. There is no support for PV devices at the moment. All - devices need to be statically assigned to guests. - - Pinning vCPUs of domains started by Xen at boot can be done from the control domain, using `xl vcpu-pin` as usual. It is not currently possible to configure vCPU pinning without a control domain.