[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [Xen-devel] [PATCH v8 for-4.12 00/17] Argo: hypervisor-mediated interdomain communication



On Feb 6, 2019, at 08:57, Jan Beulich <JBeulich@xxxxxxxx> wrote:

>>>> On 06.02.19 at 14:53, <jgross@xxxxxxxx> wrote:
>> On 06/02/2019 14:45, Jan Beulich wrote:
>>>>>> On 06.02.19 at 09:54, <christopher.w.clark@xxxxxxxxx> wrote:
>>>> Version eight of this series:
>>>> 
>>>> Note: This version may not address the currently open discussion on the
>>>> ARM hypercall argument convention and type selection for hypercall
>>>> parameters.
>>>> 
>>>> * Range check applied to numeric args in native hypercall entry
>>>> (ref: the above open discussion)
>>>> 
>>>> * Revises the compat ABI and implementation
>>>> - avoids duplication of hypercall op implementations via
>>>>   forwarding to native for ops other than sendv
>>>> - register op uses an always-64-bit fixed width pfn type
>>>>   for consistent ABI as well as compat reuse of the native op
>>>> - tested communication between VMs on x86-64 host with:
>>>>   32-bit PV, 32-bit HVM and 64-bit PV guests
>>>> 
>>>> * Applies list_first_entry_or_null macro in multiple loops to
>>>> replace previous use of a list foreach to address review feedback
>>>> 
>>>> * Removed stale comments from the public header
>>>> 
>>>> New to this series:
>>>> 
>>>> * Adds an initial version of a design document for Argo
>>>> - based on work previously sent to the mailing list, covers
>>>>   the implementation's granular locking
>>>> 
>>>> * Adds a SUPPORT.md section for the feature and Experimental statement
>>>> 
>>>> Christopher Clark (17):
>>>> argo: Introduce the Kconfig option to govern inclusion of Argo
>>>> argo: introduce the argo_op hypercall boilerplate
>>>> argo: define argo_dprintk for subsystem debugging
>>>> argo: init, destroy and soft-reset, with enable command line opt
>>>> errno: add POSIX error codes EMSGSIZE, ECONNREFUSED to the ABI
>>>> xen/arm: introduce guest_handle_for_field()
>>>> argo: implement the register op
>>>> argo: implement the unregister op
>>>> argo: implement the sendv op; evtchn: expose send_guest_global_virq
>>>> argo: implement the notify op
>>>> xsm, argo: XSM control for argo register
>>>> xsm, argo: XSM control for argo message send operation
>>>> xsm, argo: XSM control for any access to argo by a domain
>>>> xsm, argo: notify: don't describe rings that cannot be sent to
>>>> MAINTAINERS: add new section for Argo and self as maintainer
>>>> SUPPORT.md : add new entry for the Argo feature
>>>> docs, argo: add design document for Argo
>>> 
>>> Where necessary and not already present
>>> Acked-by: Jan Beulich <jbeulich@xxxxxxxx>
>>> 
>>> Jürgen, for this to be committed, your Rab would be needed, assuming
>>> you're still comfortable with this going in this late.
>> 
>> What about the ARM hypercall parameters? Is this settled?
> 
> My interpretation of Stefano's latest response was "yes, it is".

Julien's latest response said, "None of this is specific to Argo and I would be 
happy to defer this as a follow-up series."

>> If yes or if this question is solved this week:
>> 
>> Release-acked-by: Juergen Gross <jgross@xxxxxxxx>

If necessary to unblock the merge of Argo to Xen 4.12, we can limit/constrain 
Arm support via SUPPORT.md.

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

 


Rackspace

Lists.xenproject.org is hosted with RackSpace, monitoring our
servers 24x7x365 and backed by RackSpace's Fanatical Support®.