Re: Xen 4.5-rc1 update (RC1 is out 2014-Oct-24th)

[Andrew Cooper <andrew.cooper3@citrix.com>]

On 24/10/2014 19:08, konrad.wilk@oracle.com wrote:
> Feature patchsets that did not make it in by today have been put
> on the deferred list. If you think your feature should make it in Xen 4.5-rc2
> please make your case.
>
> Xen 4.5-rc1 is out today. There are some issues (see 'Known Issues' below)
> which are to be fixed in RC2. The official test-day is on Wednesday (Oct 29th)
> but if you want to start testing it today - please do!
>
> Details for the test-day are at
>
> http://wiki.xen.org/wiki/Xen_4.5_RC1_test_instructions
>
> In terms of bugs, we have:
>
> #6 linux: pv drivers miss shutdown command if issued too early
> #8 "linux, xenbus mutex hangs when rebooting dom0 and guests hung."
> #11 qxl hypervisor support
> #13 Re: [Xen-devel] man page example: xm block-attach
> #18 xl improve support for migration over non-sshlike tunnels
> #19 xl migrate transport improvements
> #22 xl does not support specifying virtual function for passthrough device
> #23 Remove arbitrary LIBXL_MAXMEM_CONSTANT from libxl, see what breaks
> #24 xl missing support for encrypted VNC
> #27 Re: [Xen-devel] xend vs xl with pci=['<bdf'] wherein the '<bdf>' are not owned by pciback or pcistub will still launch.
> #28 support PCI hole resize in qemu-xen
> #30 libxl should implement non-suspend-cancel based resume path
> #36 credit2 only uses one runqueue instead of one runq per socket
> #38 Implement VT-d large pages so we can avoid sharing between EPT
> #40 linux pvops: fpu corruption due to incorrect assumptions
> #42 "linux, S3 resume of PVHVM fails - missing call to xen_arch_post_suspend?"
> #43 "30s delay loading xenfb driver on some systems"
> #44 Security policy ambiguities - XSA-108 process post-mortem
>
> And we should start closing them if they are fixed.
>
>
> = Timeline =
>
> We are planning on a 9-month release cycle.  Based on that, below are
> our estimated dates:
>
>
> * Feature Freeze: 24th September 2014
> * First RC: 24th October [Friday!] <==== <WE ARE HERE>
> * RC2: Unknown. Need to talk to maintainers on scheduling.
> * RC2: Unknown. Need to talk to maintainers on scheduling.
> * Release: 10th December 2014
>
> The RCs and release will of course depend on stability and bugs, and
> will therefore be fairly unpredictable.  The feature freeze may be
> slipped for especially important features which are near completion.
>
> Bug-fixes, if Acked-by by maintainer, can go anytime before the First
> RC. Later on we will need to figure out the risk of regression/reward
> to eliminate the possiblity of a bug introducing another bug.
>
> = Prognosis =
>
> The states are: none -> fair -> ok -> good -> done
>
> none - nothing yet
> fair - still working on it, patches are prototypes or RFC
> ok   - patches posted, acting on review
> good - some last minute pieces
> done - all done, might have bugs
>
> = Feature freeze exception =
>
> Remember our goal for the release:
>   1. A bug-free release
>   2. An awesome release
>   3. An on-time release
>
> Accepting a new feature may make Xen more awesome; but it also
> introduces a risk that it will introduce more bugs.  That bug may be
> found before the release (threatening #3), or it may not be found
> until after the release (threatening #1).  Each freeze exception
> request will attempt to balance the benefits (how awesome the
> exception is) vs the risks (will it cause the release to slip, or
> worse, cause a bug which goes un-noticed into the final release).
>
> The idea is that today we will be pretty permissive, but that we will
> become progressively more conservative until the first RC, which is
> scheduled for 3 weeks' time (October 25).  After that, we will only
> accept bug fixes.
>
> Bug fixes can be checked in without a freeze exception throughout the
> code freeze, unless the maintainer thinks they are particularly high
> risk.  In later RC's, we may even begin rejecting bug fixes if the
> broken functionality is small and the risk to other functionality is
> high.
>
> Features which are currently marked "experimental" or do not at the
> moment work at all cannot be broken really; so changes to code only
> used by those features should be able to get a freeze exception
> easily.
>
> Features which change or add new interfaces which will need to be
> supported in a backwards-compatible way (for instance, vNUMA) will
> need freeze exceptions to make sure that the interface itself has
> enough time to be considered stable.
>
> These are guidelines and principles to give you an idea where we're
> coming from; if you think there's a good reason why making an
> exception for you will help us achieve goals 1-3 above better than not
> doing so, feel free to make your case.
>
> = Open =
>
> == Known issues == 
>
> *  Systemd integration
>    Affects CentOS7, SLES12, Fedora Core 21 and Debian Jessie. Xen source contains systemd files that can be used to configure the various run-time services. In the past the distributions would carry their own version of it - but now we host them. This is not yet complete - [[http://lists.xenproject.org/archives/html/xen-devel/2014-10/msg03064.html patches]] for this are being worked on for RC2.
>   -  Wei and Olaf
>
> *  Stubdomains build issues
>    stubdomains will not build. Fix is in staging (and will make RC2) or [[http://lists.xen.org/archives/html/xen-devel/2014-10/msg02925.html stubdom/Makefile should use QEMU_TRADITIONAL_LOC]]
>   -  Michael Young
>
> *  Building against libxl (outside code)
>    If you are building against libxl for any APIs before Xen 4.5 you will encounter building errors.
>   -  Andrew Cooper
>
> *  Migrating large Windows guests can cause WMI service to hang
>    Patch is in 'staging' and will be in RC2.

What is this in reference to?  I can't see any patches relevant to
windows in staging.

On the other hand, there is a different issue not listed here.

* Migrating VMs with viridian enabled from older versions of Xen will fail.
http://lists.xen.org/archives/html/xen-devel/2014-10/msg02816.html


>
> *  pygrub parsing grub1 problems ==
>    pygrub (Python code) has problems parsing grub.cfg file and hence cannot boot PV guests.

http://lists.xen.org/archives/html/xen-devel/2014-09/msg01053.html

>
> *  xc_reserved_device_memory_map in hvmloader to avoid conflicting MMIO/RAM (good)
>    v6
>    Treating pieces as bug-fixes only.
>   -  Tiejun Chen
>
> == Linux == 
>
> *  Linux block multiqueue (ok)
>    v3 posted.
>   -  Arianna Avanzini
>
> *  VPMU - 'perf' support in Linux (ok)
>    Depends on Xen patches
>    Acked by David Vrabel
>   -  Boris Ostrovsky
>
> *  vNUMA in Linux (ok)
>    v6 posted
>    git://gitorious.org/vnuma/linux_vnuma.git
>   -  Elena Ufimtseva
>
> *  vsyscall in Linux (fair)
>   -  Konrad Rzeszutek Wilk
>
> *  COLO Agent in Linux (fair)
>   -  Gui Jianfeng
>   -  Yang Hongyang
>   -  Dong, Eddie
>
> *  vAPIC in PVHVM guests (Linux side) (none)
>   -  Boris Ostrovsky
>
> *  Fix PAT in Linux kernel (aka Full support for PAT) (good)
>    Acked and reposted for v3.18. Waiting for x86 maintainers.
>   -  Juergen Gross
>
> == FreeBSD == 
>
> *  PVH FreeBSD dom0 (ok)
>    FreeBSD 11 goal. Toolstack side done in Xen 4.5
>   -  Roger Pau Monné
>
> == Other OSes (MiniOS, QNX) == 
>
> *  PV drivers for automotive kernels (fair)
>   -  Artem Mygaiev
>
> *  mini-os: xenbus changes for rump kernels (ok)
>    git://xenbits.xen.org/people/iwj/rumpuser-xen.git
>    branch: base.dev-xen-xenbus.v1..dev-xen-xenbus.v1
>    v2 posted
>   -  Ian Jackson
>
> == GRUB2 == 
>
> *  GRUB2 multiboot2 (fair)
>   -  Daniel Kiper
>
> == OSSTEST == 
>
> *  OSSTest: libvirt (good)
>   -  Ian Campbell
>
> == Deferred to QEMU v2.next == 
>
> *  Using qemu-upstream in a stubdomain (fair)
>    Will use rump kernels.
>   -  Ian Jackson
>
> *  AMD Radeon PCI GPU passthrough (none)
>    Focusing on Xen 4.2 and qemu-traditional
>   -  Kelly Zytaruk
>
> *  Intel IGD PCI GPU passthrough (ok)
>    v5 posted
>   -  Chen, Tiejun
>
> *  Xen PV block driver in OVMF (UEFI in guest) (ok)
>    v1
>   -  Anthony PERARD
>
> == Deferred to Xen hypervisor 4.6 == 
>
> *  Xen Boot Information (xbi) (ok)
>    Dependency for GRUB2 + EFI work
>    http://lists.xen.org/archives/html/xen-devel/2014-10/msg02068.html
>    v4, No go for full patchset. Only some of the patches.
>    No ARM EFI hardware (yet) available to test them.
>   -  Daniel Kiper
>
> *  PVH - AMD hardware support. (fair)
>    Posted.
>   -  Mukesh Rathor
>
> *  VMware backdoor (hypercall) (ok)
>    v5 posted.
>   -  Don Slutz
>
> *  extending mem_access support to PV domain (fair)
>    RFC v2
>   -  Aravindh Puthiyaparambil (aravindp)
>
> *  Repurpose SEDF Scheduler for Real-time (fair)
>    RFC patch posted (v2)
>   -  Joshua Whitehead, Robert VanVossen
>
> *  ARM remote processor iommu module (GPUs + IPUs) (fair)
>    v3 posted
>   -  Andrii Tseglytskyi
>
> *  dirty vram / IOMMU bug (fair)
>    http://bugs.xenproject.org/xen/bug/38
>   -  Zhang, Yang Z
>
> *  Xen multiboot2-EFI support (fair)
>    Needed for GrUB2
>    Depends on Xen Boot info (rework multiboot and other structs)
>    See http://lists.xen.org/archives/html/xen-devel/2013-05/msg02281.html
>    RFC posted
>   -  Daniel Kiper
>
> *  Support controlling the max C-state sub-state (ok)
>    v3 posted
>   -  Ross Lagerwall
>
> *  IOMMU ABI for guests to map their DMA regions (fair)
>   -  Malcolm Crossley
>
> *  Default to credit2 (none)
>    cpu pinning, numa affinity and cpu reservation
>   -  George Dunlap
>
> *  Convert tasklet to per-cpu tasklets (fair)
>    RFC posted
>   -  Konrad Rzeszutek Wilk
>
> *  Further tmem cleanups/fixes (16TB etc) (fair)
>   -  Bob Liu
>
> *  1TB slow destruction (ok)
>   -  Bob Liu
>
> *  ARM VM save/restore/live migration (none)
>    Need to rebased against migrationv2 - no code posted.
>   -  Junghyun Yoo
>
> *  ARM GICv2m support (none)
>   -  Linaro (unknown)
>
> == Deferred to Xen toolstack 4.6 == 
>
> *  vNUMA in Xen toolstack (ok)
>    v11 posted
>    Hypervisor part in
>    git://gitorious.org/vnuma/xen_vnuma.git:v11
>   -  Elena Ufimtseva
>
> *  pvscsi in libxl (fair)
>   -  Juergen Gross and Olaf
>
> *  COarse-grain LOck-stepping Virtual Machines in Xen (fair)
>    RFC v3 posted, based on remus-v19
>   -  Wen Congyang
>   -  Gui Jianfeng
>   -  Yang Hongyang
>   -  Dong, Eddie
>
> *  extend the xenstore ring with a 'closing' signal (fair)
>    RFC patch posted
>   -  David Scott

Already committed, therefore complete.
674ad2be409da77ebda3616085a2f67c72556c11

>
> *  New Migration (v2). (good)
>    v7 (libxc and libxl)
>    git://xenbits.xen.org/people/andrewcoop/xen.git
>    Seems that it might need to slip or we run v1 alongside v2.
>   -  Andrew Cooper & David Vrabel
>
> *  libxl migrationv2 patches. (none)
>   -  Andrew Cooper & David Vrabel
>
> *  tmem migrationv2 patches. (none)
>   -  Bob Liu & Andrew Cooper & David Vrabel
>
> *  Remus using migration-v2 (fair)
>    RFC posted - depends on v6 of 'New Migration'
>   -  Yang Hongyang
>
> *  snapshot API extension (checkpointing disk) (ok)
>    v5
>    His email bounces.
>   -  Bamvor Jian Zhang
>
> *  Rearrange and cleanup installation destination directories (/var -> var/lib/xen) (fair)
>   -  Daniel Kiper
>
> *  libxl/xl - xm compatibility mode for mem-max and mem-set; (ok)
>   -  Daniel Kiper
>
> *  xl list --long (and some related xl commands) have some bugs (none)
>   -  Zhigang Wang
>
> *  Xen HPET interrupt fixes (fair)
>    behind migration v2
>   -  Andrew Cooper
>
> *  cpuid leveling (none)
>    http://xenbits.xen.org/people/andrewcoop/feature-levelling/feature-levelling-D.pdf
>   -  Andrew Cooper
>
> *  live migration knobs, there is no suitable code yet, just ideas (none)
>     http://lists.xenproject.org/archives/html/xen-devel/2014-03/msg00785.html
>   -  Olaf Hering
>
> *  xl does not handle migrate interruption gracefully (none)
>    If you start a localhost migrate, and press "Ctrl-C" in the middle, you get two hung domains
>   -  Ian Jackson
>
> *  IO-NUMA - hwloc and xl (none)
>    Andrew Cooper had an RFC patch for hwloc
>    add restrictions  as to which devices cannot safely/functionally be split apart.
>   -  Boris Ostrovsky
>
> *  HVM guest NUMA (none)
>   -  Matt Wilson
>
> *  PVH - Migration of PVH DomUs. (none)
>    Depends on migration2 code
>   -  Roger Pau Monné
>
> *  PVH - Migration of guests from a PVH dom0  (none)
>    Depends on migration2 code
>   -  Roger Pau Monné
>
> *  "Short" grant copy (just header) of packets. (none)
>   -  Zoltan Kiss
>
> == Deferred to Linux's after Xen 4.6 == 
>
> *  ARM - Device assigment usage in Linux code (arch/arm) (none)
>    Depends on Xen pieces which are on the Xen 4.6 list.
>   -  Julien Grall
>
> *  Linux ARM - Device assigment (fair)
>   -  Julien Grall
>
> == Up for grabs == 
>
> *  OSSTest - also test Linux PVH guests
>
> *  PoD fixes
>    if you boot with memory <= maxmem we have a size estimation bug
>
> *  TLB flushing without locks in Xen
>
> *  xl does not support specifying virtual function for passthrough device
>    http://bugs.xenproject.org/xen/bug/22
>
> *  PCI hole resize support hvmloader/qemu-traditional/qemu-upstream with PCI/GPU passthrough
>    http://bugs.xenproject.org/xen/bug/28
>
> *  libx{c,l} error handling cleanup 
>
> *  Adding missing 'xend' features in libxl
>
> *  xl list -l on a dom0-only system
>
> *  xl list -l doesn't contain tty console port
>
> *  xl: passing more defaults in configuration in xl.conf
>    There are a number of options for which it might be useful to pass a default in xl.conf.  For example, if we could have a default "backend" parameter for vifs, then it would be easy to switch back and forth between a backend in a driver domain and a backend in dom0.
>
> *  PVH - PVH working with shadow.
>    Based on Tim's work
>
> *  PVH - PCI passthrough for DomU.
>
> *  AMD performance regressions
>
> *  Performance due to hypercall preemption. More preemptions - slower. (none)
>
> == Completed == 
>
> == Hypervisor == 
>
> *  ARM - introduce GNTTABOP_cache_flush (ok)
>    v11
>   -  Stefano Stabellini
>
> *  ARM - VGIC emulation (done)
>    Reposted as gic and vgic fixes and improvements
>    v12
>   -  Stefano Stabellini
>
> *  ARM implement mem_access (done)
>    v12, two patches for Xen 4.6
>    https://github.com/tklengyel/xen/tree arm_memaccess_12-for-4.5
>   -  Tamas K Lengyel
>
> *  ARM - Add Odroid-XU (Exynos5410) support (done)
>    v6
>   -  Suriyan Ramasami
>
> *  ARM GICv3 support (done)
>    v11 posted
>   -  Vijay Kilari
>
> *  ARM implement mem_access (done)
>    v12, two patches for Xen 4.6
>    https://github.com/tklengyel/xen/tree arm_memaccess_12-for-4.5
>   -  Tamas K Lengyel
>
> *  ARM - MiniOS (done)
>    v7 posted
>   -  Thomas Leonard
>
> *  ARM XEN_DOMCTL_memory_mapping hypercall for ARM (done)
>    v12 posted.
>   -  Arianna Avanzini
>
> *  ARM - XENFEAT_grant_map_11 (aka map grants refs at pfn = mfn) (done)
>    Provide kernels an grant->MFN lookup
>    v4
>   -  Stefano Stabellini
>
> *  ARM PSCI v0.2 (done)
>    v11 posted
>   -  Parth Dixit
>
> *  ARM  - IOMMU support (done)
>   -  Julien Grall
>
> *  ARM Interrupt latency reduction (no maintenance interrupts) (good)
>   -  Stefano Stabellini
>
> *  ARM DRA7 support (done)
>    v3 posted
>    v3 with comments applied
>   -  Andrii Tseglytskyi
>
> *  ARM: Use super pages in p2m (done)
>    v5 posted
>   -  Ian Campbell
>
> *  ARM Xen UEFI booting on ARM (done)
>    v5
>   -  Roy Franz
>
> *  VPMU - 'perf' support in Xen (good)
>    v14 posted
>    Need reviews/final ack.
>   -  Boris Ostrovsky
>
> *  Cache QoS Monitoring - hypercalls (done)
>    Just hypercalls - no toolstack changes.
>    v15
>    Hit a snag with rdmsr/IPI/wrmsr/IPI, possible redesign
>   -  Chao Peng, Dongxiao Xu, and Shantong Kang
>
> *  XenRT (Preemptive Global Earliest Deadline First) (done)
>    v3
>   -  Meng Xu
>
> *  Introspection of HVM guests (done)
>    v10, split out in for 4.5 (smaller subset)
>   -  Razvan Cojocaru
>
> *  alternative_asm in Xen (done)
>   -  Feng Wu
>
> *  SMAP (done)
>   -  Feng Wu
>
> *  Re-write of vHPET (done)
>    aka hvm/hpet: Detect comparator values in the past
>   -  Don Slutz
>
> *  vAPIC in PVHVM guests (Xen side) (done)
>   -  Boris Ostrovsky
>
> *  Xen PVH dom0 (done)
>   -  Mukesh Rathor
>
> *  amd_ucode cleanups, verify patch size(enhancement) (mostly in master except one patch)
>
> *  Data breakpoint Extension support (new-feat) (in master)
>
> *  Feature masking MSR support (enhancement) (in master)
>
> *  Support BRCM TruManage chip (Serial over LAN support) (new-feat) (in master)
>
> *  fix vmce_amd* functions, unify mce_amd mcheck initialization (fixes/cleanups)
>
> *  multiple AMD container files appended together in initrd (early initramfs)
>   -  Aravind and Suravee
>
> *  NUMA memory scrubbing (done)
>   -  Konrad Rzeszutek Wilk
>
> *  ioreq-server, aka secondary emulators (done)
>   -  Paul Durrant
>
> *  Soft affinity for vcpus (was NUMA affinity for vcpus) (good)
>    v11 posted
>   -  Dario Faggioli
>
> *  HT enabled, virtualization overhead is high (Xen 4.4) (done)
>    kernbench demonstrated it
>    Looking and tracing
>    http://www.gossamer-threads.com/lists/xen/devel/339409
>    False alarm.
>   -  Dario Faggioli
>
> == lib{xc,xl} and toolstack == 
>
> *  Build systems fixes/improvements (ok)
>   -  Andrew Cooper

Not complete.

Patch 4 is still pending review with regards to a release-ack or not. 

http://lists.xen.org/archives/html/xen-devel/2014-10/msg01817.html

~Andrew

>
> *  libxl work - JSON to keep track of guest configs (done)
>    Some patches merged, need to post more.
>   -  Wei Liu
>
> *  Remus in Xen (libxl) (done)
>    v19
>    url:  https://github.com/macrosheep/xen/tree/remus-v19
>   -  Gui Jianfeng
>   -  Yang Hongyang
>   -  Dong, Eddie
>
> *  libvirt and xl discard support, so that libvirt can start using it (done)
>   -  Olaf Hering
>
> *  OSSTest: upstream QEMU (done)
>   -  Ian Campbell
>
> *  rework VM Generation ID (done)
>    v7 posted
>   -  David Vrabel
>
> *  systemd support (done)
>    v11
>   -  Luis R. Rodriguez
>
> *  Soft affinity for vcpus libxl/xl changes (done)
>    v13 posted
>   -  Dario Faggioli
>
> == QEMU == 
>
> *  Bigger PCI hole in QEMU (done)
>    Needs to be rebased
>   -  Don Slutz
>
> *  QEMU 2.0 branch for qemu-upstream (done)
>    It is v2.0 with 2.1 Xen backports.
>   -  Stefano Stabellini
>
> == Linux 3.18 and earlier == 
>
> *  pvSCSI in Linux (fronted and backend) (done)
>    v6
>   -  Juergen Gross
>
> *  Linux PVH dom0 (done)
>   -  Mukesh Rathor
>
> *  Netback multiqueue (good)
>   -  Wei Liu
>
> *  Linux pvops of Xen EFI hypercall support (done)
>   -  Daniel Kiper
>