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

Re: [Xen-devel] Requesting for freeze exception for altp2m

On Fri, Jul 10, 2015 at 08:58:33PM +0000, Sahita, Ravi wrote:
> Hi Wei,
> I had already requested the freeze exception already for altp2m[0] - I'm 
> reposting that request with updates for v4 of patch series, and other updates:
> 1. clarify the state of patch series / feature.
> Reviewed      Acked   altp2m series v4
> Y     N       [PATCH v3 01/15] common/domain: Helpers to pause a domain while 
> in context
> Status if not acked:          Good? - Jan has commented that he is ok with 
> this patch
> Y     Y       [PATCH v3 02/15] VMX: VMFUNC and #VE definitions and detection.
> Y     Y       [PATCH v3 03/15] VMX: implement suppress #VE.
> Y     N       [PATCH v3 04/15] x86/HVM: Hardware alternate p2m support 
> detection
> Status if not acked:          Good? no pending comments
> Y     N       [PATCH v3 05/15] x86/altp2m: basic data structures and support 
> routines
> Status if not acked:          Working on some remaining comments from Jan. 
> George ok with documentation to be added to mm-locks.h
> Y     Y       [PATCH v3 06/15] VMX/altp2m: add code to support EPTP switching 
> and #VE
> Y     N       [PATCH v3 07/15] VMX: add VMFUNC leaf 0 (EPTP switching) to 
> emulator
> Status if not acked:          Some straightforward comments from Jan to be 
> addressed - ack with those addressed?
> Y     N       [PATCH v3 08/15] x86/altp2m: add control of suppress_ve
> Status if not acked:          George's suggested patch seems acceptable - ack?
> Y     Y       [PATCH v3 09/15] x86/altp2m: alternate p2m memory events
> Y     N       [PATCH v3 10/15] x86/altp2m: add remaining support routines
> Status if not acked:          Some things could be moved around for 'future 
> patch series' - ack with that status?
> Y     N       [PATCH v3 11/15] x86/altp2m: define and implement alternate p2m 
> HVMOP types
> Status if not acked:          main hvm op reword done - other comments being 
> addressed (no major changes expected)
> Y     Y       [PATCH v3 12/15] x86/altp2m: Add altp2mhvm HVM domain parameter
> Y     Y       [PATCH v3 13/15] x86/altp2m: XSM hooks for altp2m HVM ops
> Y     Y       [PATCH v4 14/15] tools/libxc: add support to altp2m hvmops
> Y     N       [PATCH v4 15/15] tools/xen-access: altp2m testcases
> Status if not acked:          Ian's questions seem answered - ack?

More than half of this series is still not acked. If I only look at the
number of patches acked, that would be a flat out rejection from me.

However I've been keeping my eyes on this series long enough, I think
they are getting very close now. So I reserve my judgement at this
point.  On one hand I would very much like this feature to go in 4.6, on
the other hand I need to make sure 4.6 release on schedule. I hope you
understand my dilemma.

So please keep up with the good work, address all comments and repost
early next week.  Then we can decide what to do about this feature,
together with maintainers.  I think we have a very busy week ahead.


> 2. explain why it needs to be in this release (benefits).
> Altp2m is a novel functionality for VM introspection with benefits for many 
> use cases and consumers - see design doc at For further details:
> http://lists.xenproject.org/archives/html/xen-devel/2015-06/msg01319.html
> 3. explain why it doesn't break things (risks).
> Altp2m is default off globally, and additionally can be enabled per domain 
> when opted in.
> It has been tested for inter-domain and intro-domain (on legacy and new Intel 
> CPUs) with support from community.
> Other feedback from maintainers also suggests it is low risk:
> http://lists.xen.org/archives/html/xen-devel/2015-07/msg01793.html
> 4. CC relevant maintainers and release manager.
> Done
> Per the general rules of thumb of granting freeze exception:
> 1. the patch series is in shape, only pending a few minor tweaks.
> 2. the patch series is endorsed by maintainer(s).
> See previous messages from discussions stating good health of the patch 
> series and low impact:
> http://lists.xen.org/archives/html/xen-devel/2015-07/msg01793.html
> http://lists.xen.org/archives/html/xen-devel/2015-07/msg01849.html
> Cheers,
> Ravi
> [0] http://lists.xen.org/archives/html/xen-devel/2015-07/msg01601.html

Xen-devel mailing list



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