From mboxrd@z Thu Jan 1 00:00:00 1970 From: Don Slutz Subject: Re: [RFC PATCH 0/3] Xen hvm saving improvements Date: Mon, 16 Dec 2013 12:50:58 -0500 Message-ID: <52AF3D82.9010104@terremark.com> References: <1387160259-9638-1-git-send-email-andrew.cooper3@citrix.com> Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii"; Format="flowed" Content-Transfer-Encoding: 7bit Return-path: In-Reply-To: <1387160259-9638-1-git-send-email-andrew.cooper3@citrix.com> List-Unsubscribe: , List-Post: List-Help: List-Subscribe: , Sender: xen-devel-bounces@lists.xen.org Errors-To: xen-devel-bounces@lists.xen.org To: Andrew Cooper , Xen-devel Cc: George Dunlap , Keir Fraser , Don Slutz , Jan Beulich List-Id: xen-devel@lists.xenproject.org On 12/15/13 21:17, Andrew Cooper wrote: > This series is *very* RFC at the moment - it is only compile tested. I would > very much appreciate review on the design. (and I shall see about sorting out > some dev testing tomorrow) > > The ultimate purpose is to fix that the fact that > XEN_DOMCTL_gethvmcontext_partial is very unreliable at providing the correct > subset of information. > > All the changes are supposed to be internal to Xen. There should be no > difference at all to the result of XEN_DOMCTL_gethvmcontext (which can be used > as a useful dev test for the validity of the changes). > > George: As for whether this is intended for 4.5, I have no idea. On the one > hand, it is fixing a hypercall which doesn't function correctly, but on the > other hand, it is a very large and complicated set of changes relevant to > migration. Then again, the majority of the changes were largely mechanical, > and there is a simple test to prove whether there are bugs or not. I shall > defer to others for comments on the risk/rewards of this series. > > Andrew Cooper (3): > xen/hvm-save: Refactor HVM_REGISTER_SAVE_RESTORE > xen/hvm-save: Extend hvm_save_handler to take an instance parameter > xen/hvm-save: Adjust calling of multi-instance save handlers. > > xen/arch/x86/cpu/mcheck/vmce.c | 34 +++--- > xen/arch/x86/hvm/hpet.c | 8 +- > xen/arch/x86/hvm/hvm.c | 249 ++++++++++++++++++++-------------------- > xen/arch/x86/hvm/i8254.c | 8 +- > xen/arch/x86/hvm/irq.c | 24 ++-- > xen/arch/x86/hvm/mtrr.c | 61 +++++----- > xen/arch/x86/hvm/pmtimer.c | 8 +- > xen/arch/x86/hvm/rtc.c | 8 +- > xen/arch/x86/hvm/vioapic.c | 7 +- > xen/arch/x86/hvm/viridian.c | 31 ++--- > xen/arch/x86/hvm/vlapic.c | 38 ++---- > xen/arch/x86/hvm/vpic.c | 18 +-- > xen/common/hvm/save.c | 108 ++++++++++------- > xen/include/xen/hvm/save.h | 65 +++++++---- > 14 files changed, 337 insertions(+), 330 deletions(-) > > Signed-off-by: Andrew Cooper > CC: Keir Fraser > CC: Jan Beulich > CC: Don Slutz > CC: George Dunlap > > -- > 1.7.10.4 > I think that with this size of a change, the issue that Jan talked about in the thread: http://lists.xen.org/archives/html/xen-devel/2013-12/msg02246.html about XEN_DOMCTL_gethvmcontext_partial not taking in a size, nor returning a size should also be fixed. -Don Slutz