Re: [Qemu-devel] [Qemu-stable] [PATCH 00/54] Patch Round-up for stable 2.11.1, freeze on 2018-02-12

2018-02-16 Thread Peter Lieven 
Am 12.02.2018 um 17:13 schrieb Dr. David Alan Gilbert:
> * Peter Lieven (p...@kamp.de) wrote:
>> Am 06.02.2018 um 20:14 schrieb Michael Roth:
>>> Hi everyone,
>>>
>>> The following new patches are queued for QEMU stable v2.11.1:
>>>
>>>https://github.com/mdroth/qemu/commits/stable-2.11-staging
>>>
>>> The release is planned for 2017-02-14:
>>>
>>>https://wiki.qemu.org/Planning/2.11
>>>
>>> Please respond here or CC qemu-sta...@nongnu.org on any patches you
>>> think should be included in the release.
>>>
>>> Of particular importance would be any feedback on the various QEMU
>>> patches relating to Spectre/Meltdown mitigation. The current tree has
>>> what I understand to be the QEMU components required for x86, s390,
>>> and pseries, but feedback/confirmation from the various authors would
>>> be greatly appreciated.
>> Hi,
>>
>> I also found the following patches that affect migration:
>>
>> migration: Don't leak IO channels
>> migration: Recover block devices if failure in device state
>> migration/savevm.c: set MAX_VM_CMD_PACKAGED_SIZE to 1ul << 32
>>
>> In general it seems that migration related patches are often not tagged 
>> qemu-stable.
>> David, can you check if there are patches missing?
> I'm always a bit cautious about forwarding stuff to stable; just because
> it's landed in our main branch it doesn't mean many people have tested
> it, so I'd rather wait unless it was particularly urgent.

Understood. Maybe we need a special mechanism to track such patches.
It would be good to have them in a stable release. Maybe tag them as 
stable-candidate
or sth. Then we can recheck their status when a stable release is going to be 
released.

Peter

Re: [Qemu-devel] [Qemu-stable] [PATCH 00/54] Patch Round-up for stable 2.11.1, freeze on 2018-02-12

2018-02-13 Thread Greg Kurz 
On Mon, 12 Feb 2018 16:13:23 +
"Dr. David Alan Gilbert"  wrote:

> * Peter Lieven (p...@kamp.de) wrote:
> > Am 06.02.2018 um 20:14 schrieb Michael Roth:  
> > > Hi everyone,
> > > 
> > > The following new patches are queued for QEMU stable v2.11.1:
> > > 
> > >https://github.com/mdroth/qemu/commits/stable-2.11-staging
> > > 
> > > The release is planned for 2017-02-14:
> > > 
> > >https://wiki.qemu.org/Planning/2.11
> > > 
> > > Please respond here or CC qemu-sta...@nongnu.org on any patches you
> > > think should be included in the release.
> > > 
> > > Of particular importance would be any feedback on the various QEMU
> > > patches relating to Spectre/Meltdown mitigation. The current tree has
> > > what I understand to be the QEMU components required for x86, s390,
> > > and pseries, but feedback/confirmation from the various authors would
> > > be greatly appreciated.  
> > 
> > Hi,
> > 
> > I also found the following patches that affect migration:
> > 
> > migration: Don't leak IO channels
> > migration: Recover block devices if failure in device state
> > migration/savevm.c: set MAX_VM_CMD_PACKAGED_SIZE to 1ul << 32
> > 
> > In general it seems that migration related patches are often not tagged 
> > qemu-stable.
> > David, can you check if there are patches missing?  
> 
> I'm always a bit cautious about forwarding stuff to stable; just because
> it's landed in our main branch it doesn't mean many people have tested
> it, so I'd rather wait unless it was particularly urgent.
> 
> The 'migration: Recover block devices if failure in device state' is
> probably a good candidate.
> Another one would be Greg Kurz's:
> 875fcd013ab68c64802998b22f54f0184479d21b - migration: incoming postcopy 
> advise sanity checks
> 

Yeah. Mike pushed it to its stable-2.11-staging tree:

https://github.com/mdroth/qemu/commit/a3fd64f2fe9f90c7b4d002910a54049ee6a09b64

Cheers,

--
Greg

> Dave
> 
> > Thank you,
> > Peter
> >   
> --
> Dr. David Alan Gilbert / dgilb...@redhat.com / Manchester, UK
>

Re: [Qemu-devel] [Qemu-stable] [PATCH 00/54] Patch Round-up for stable 2.11.1, freeze on 2018-02-12

2018-02-12 Thread Michael Roth 
Quoting Michael Roth (2018-02-06 13:14:21)
> Hi everyone,  
> 
> 
> The following new patches are queued for QEMU stable v2.11.1:
> 
>   https://github.com/mdroth/qemu/commits/stable-2.11-staging
> 
> The release is planned for 2017-02-14:
> 
>   https://wiki.qemu.org/Planning/2.11
> 
> Please respond here or CC qemu-sta...@nongnu.org on any patches you
> think should be included in the release.
> 
> Of particular importance would be any feedback on the various QEMU
> patches relating to Spectre/Meltdown mitigation. The current tree has
> what I understand to be the QEMU components required for x86, s390,
> and pseries, but feedback/confirmation from the various authors would
> be greatly appreciated.

Thank you for the responses/suggestions. The following additional
patches have been queued for the release and pushed to:

https://github.com/mdroth/qemu/commits/stable-2.11-staging

spapr: add missing break in h_get_cpu_characteristics() (Greg Kurz)
vga: check the validation of memory addr when draw text (linzhecheng)
input: fix memory leak (linzhecheng)
ui: correctly advance output buffer when writing SASL data (Daniel P.  Berrangé)
ui: avoid sign extension using client width/height (Daniel P. Berrange)
ui: mix misleading comments & return types of VNC I/O helper methods (Daniel P. 
Berrange)
ui: add trace events related to VNC client throttling (Daniel P.  Berrange)
ui: place a hard cap on VNC server output buffer size (Daniel P.  Berrange)
ui: fix VNC client throttling when forced update is requested (Daniel P.  
Berrange)
ui: fix VNC client throttling when audio capture is active (Daniel P.  Berrange)
ui: refactor code for determining if an update should be sent to the client 
(Daniel P. Berrange)
ui: correctly reset framebuffer update state after processing dirty regions 
(Daniel P. Berrange)
ui: introduce enum to track VNC client framebuffer update request state (Daniel 
P. Berrange)
ui: track how much decoded data we consumed when doing SASL encoding (Daniel P. 
Berrange)
ui: avoid pointless VNC updates if framebuffer isn't dirty (Daniel P.  Berrange)
ui: remove redundant indentation in vnc_client_update (Daniel P.  Berrange)
ui: remove unreachable code in vnc_update_client (Daniel P. Berrange)
ui: remove 'sync' parameter from vnc_update_client (Daniel P. Berrange)
migration: incoming postcopy advise sanity checks (Greg Kurz)
target/sh4: add missing tcg_temp_free() in _decode_opc() (Philippe 
Mathieu-Daudé)
migration/savevm.c: set MAX_VM_CMD_PACKAGED_SIZE to 1ul << 32 (Daniel Henrique 
Barboza)
migration: Recover block devices if failure in device state (Dr. David Alan 
Gilbert)
migration: Don't leak IO channels (Ross Lagerwall)
s390x/sclp: fix event mask handling (Christian Borntraeger)
memory: set ioeventfd_update_pending after address_space_update_ioeventfds 
(linzhecheng)

> 
> Thanks!
> 
> 
> 
> The following changes since commit 0a0dc59d27527b78a195c2d838d28b7b49e5a639:
> 
>   Update version for v2.11.0 release (2017-12-13 14:31:09 +)
> 
> are available in the git repository at:
> 
>   git://github.com/mdroth/qemu.git stable-2.11-staging
> 
> for you to fetch changes up to ed8b4ecc68d6bfe98000b08d649049d0c1174c11:
> 
>   target/ppc/spapr: Add H-Call H_GET_CPU_CHARACTERISTICS (2018-02-05 19:07:38 
> -0600)
> 
> 
> Alex Bennée (1):
>   target/sh4: fix TCG leak during gusa sequence
> 
> Alex Williamson (1):
>   vfio: Fix vfio-kvm group registration
> 
> Christian Borntraeger (2):
>   s390x/kvm: Handle bpb feature
>   s390x/kvm: provide stfle.81
> 
> Claudio Imbrenda (1):
>   s390x: fix storage attributes migration for non-small guests
> 
> Cornelia Huck (1):
>   linux-headers: update
> 
> Cédric Le Goater (1):
>   target/ppc: introduce the PPC_BIT() macro
> 
> David Gibson (7):
>   spapr: Add pseries-2.12 machine type
>   spapr: Capabilities infrastructure
>   spapr: Treat Hardware Transactional Memory (HTM) as an optional 
> capability
>   spapr: Validate capabilities on migration
>   target/ppc: Clean up probing of VMX, VSX and DFP availability on KVM
>   spapr: Handle VMX/VSX presence as an spapr capability flag
>   spapr: Handle Decimal Floating Point (DFP) as an optional capability
> 
> Eduardo Habkost (5):
>   i386: Change X86CPUDefinition::model_id to const char*
>   i386: Add spec-ctrl CPUID bit
>   i386: Add FEAT_8000_0008_EBX CPUID feature word
>   i386: Add new -IBRS versions of Intel CPU models
>   i386: Add EPYC-IBPB CPU model
> 
> Eric Auger (1):
>   linux-headers: update to 4.15-rc1
> 
> Fam Zheng (3):
>   block: Open backing image in force share mode for size probe
>   osdep: Retry SETLK upon EINTR
>   usb-storage: Fix share-rw option parsing
> 
> Greg Kurz (2):
>

Re: [Qemu-devel] [Qemu-stable] [PATCH 00/54] Patch Round-up for stable 2.11.1, freeze on 2018-02-12

2018-02-12 Thread Dr. David Alan Gilbert 
* Peter Lieven (p...@kamp.de) wrote:
> Am 06.02.2018 um 20:14 schrieb Michael Roth:
> > Hi everyone,
> > 
> > The following new patches are queued for QEMU stable v2.11.1:
> > 
> >https://github.com/mdroth/qemu/commits/stable-2.11-staging
> > 
> > The release is planned for 2017-02-14:
> > 
> >https://wiki.qemu.org/Planning/2.11
> > 
> > Please respond here or CC qemu-sta...@nongnu.org on any patches you
> > think should be included in the release.
> > 
> > Of particular importance would be any feedback on the various QEMU
> > patches relating to Spectre/Meltdown mitigation. The current tree has
> > what I understand to be the QEMU components required for x86, s390,
> > and pseries, but feedback/confirmation from the various authors would
> > be greatly appreciated.
> 
> Hi,
> 
> I also found the following patches that affect migration:
> 
> migration: Don't leak IO channels
> migration: Recover block devices if failure in device state
> migration/savevm.c: set MAX_VM_CMD_PACKAGED_SIZE to 1ul << 32
> 
> In general it seems that migration related patches are often not tagged 
> qemu-stable.
> David, can you check if there are patches missing?

I'm always a bit cautious about forwarding stuff to stable; just because
it's landed in our main branch it doesn't mean many people have tested
it, so I'd rather wait unless it was particularly urgent.

The 'migration: Recover block devices if failure in device state' is
probably a good candidate.
Another one would be Greg Kurz's:
875fcd013ab68c64802998b22f54f0184479d21b - migration: incoming postcopy advise 
sanity checks

Dave

> Thank you,
> Peter
> 
--
Dr. David Alan Gilbert / dgilb...@redhat.com / Manchester, UK

Re: [Qemu-devel] [Qemu-stable] [PATCH 00/54] Patch Round-up for stable 2.11.1, freeze on 2018-02-12

2018-02-08 Thread Peter Lieven 

Am 06.02.2018 um 20:14 schrieb Michael Roth:

Hi everyone,

The following new patches are queued for QEMU stable v2.11.1:

   https://github.com/mdroth/qemu/commits/stable-2.11-staging

The release is planned for 2017-02-14:

   https://wiki.qemu.org/Planning/2.11

Please respond here or CC qemu-sta...@nongnu.org on any patches you
think should be included in the release.

Of particular importance would be any feedback on the various QEMU
patches relating to Spectre/Meltdown mitigation. The current tree has
what I understand to be the QEMU components required for x86, s390,
and pseries, but feedback/confirmation from the various authors would
be greatly appreciated.


Hi,

I also found the following patches that affect migration:

migration: Don't leak IO channels
migration: Recover block devices if failure in device state
migration/savevm.c: set MAX_VM_CMD_PACKAGED_SIZE to 1ul << 32

In general it seems that migration related patches are often not tagged 
qemu-stable.
David, can you check if there are patches missing?

Thank you,
Peter

Re: [Qemu-devel] [Qemu-stable] [PATCH 00/54] Patch Round-up for stable 2.11.1, freeze on 2018-02-12

2018-02-07 Thread Greg Kurz 
Hi Michael,

On Tue,  6 Feb 2018 13:14:21 -0600
Michael Roth  wrote:

> Hi everyone,  
> 
> 
> The following new patches are queued for QEMU stable v2.11.1:
> 
>   https://github.com/mdroth/qemu/commits/stable-2.11-staging
> 
> The release is planned for 2017-02-14:
> 
>   https://wiki.qemu.org/Planning/2.11
> 
> Please respond here or CC qemu-sta...@nongnu.org on any patches you
> think should be included in the release.
> 

Maybe worth to pick the following patch when it gets merged:

http://patchwork.ozlabs.org/patch/869873/

It fixes a 2.11 regression with RAM postcopy.

> Of particular importance would be any feedback on the various QEMU
> patches relating to Spectre/Meltdown mitigation. The current tree has
> what I understand to be the QEMU components required for x86, s390,
> and pseries, but feedback/confirmation from the various authors would
> be greatly appreciated.
> 

Patch 54 "target/ppc/spapr: Add H-Call  H_GET_CPU_CHARACTERISTICS" has a
trivial but potentially embarrassing bug (QEMU aborts if the host supports
KVM_PPC_CPU_CHAR_BCCTRL_SERIALISED and cap-ibs="fixed" is used).

A patch was sent to the list:

http://patchwork.ozlabs.org/patch/868426/

AFAIK David Gibson (Cc'd) is away and I don't know when this patch
will be merged.

Cheers,

--
Greg

> Thanks!
> 
> 
> 
> The following changes since commit 0a0dc59d27527b78a195c2d838d28b7b49e5a639:
> 
>   Update version for v2.11.0 release (2017-12-13 14:31:09 +)
> 
> are available in the git repository at:
> 
>   git://github.com/mdroth/qemu.git stable-2.11-staging
> 
> for you to fetch changes up to ed8b4ecc68d6bfe98000b08d649049d0c1174c11:
> 
>   target/ppc/spapr: Add H-Call H_GET_CPU_CHARACTERISTICS (2018-02-05 19:07:38 
> -0600)
> 
> 
> Alex Bennée (1):
>   target/sh4: fix TCG leak during gusa sequence
> 
> Alex Williamson (1):
>   vfio: Fix vfio-kvm group registration
> 
> Christian Borntraeger (2):
>   s390x/kvm: Handle bpb feature
>   s390x/kvm: provide stfle.81
> 
> Claudio Imbrenda (1):
>   s390x: fix storage attributes migration for non-small guests
> 
> Cornelia Huck (1):
>   linux-headers: update
> 
> Cédric Le Goater (1):
>   target/ppc: introduce the PPC_BIT() macro
> 
> David Gibson (7):
>   spapr: Add pseries-2.12 machine type
>   spapr: Capabilities infrastructure
>   spapr: Treat Hardware Transactional Memory (HTM) as an optional 
> capability
>   spapr: Validate capabilities on migration
>   target/ppc: Clean up probing of VMX, VSX and DFP availability on KVM
>   spapr: Handle VMX/VSX presence as an spapr capability flag
>   spapr: Handle Decimal Floating Point (DFP) as an optional capability
> 
> Eduardo Habkost (5):
>   i386: Change X86CPUDefinition::model_id to const char*
>   i386: Add spec-ctrl CPUID bit
>   i386: Add FEAT_8000_0008_EBX CPUID feature word
>   i386: Add new -IBRS versions of Intel CPU models
>   i386: Add EPYC-IBPB CPU model
> 
> Eric Auger (1):
>   linux-headers: update to 4.15-rc1
> 
> Fam Zheng (3):
>   block: Open backing image in force share mode for size probe
>   osdep: Retry SETLK upon EINTR
>   usb-storage: Fix share-rw option parsing
> 
> Greg Kurz (2):
>   spapr_pci: fix MSI/MSIX selection
>   spapr: fix device tree properties when using compatibility mode
> 
> Jay Zhou (1):
>   vhost: remove assertion to prevent crash
> 
> Jose Ricardo Ziviani (1):
>   ppc: Change Power9 compat table to support at most 8 threads/core
> 
> Kevin Wolf (2):
>   block: Make bdrv_drain_invoke() recursive
>   block: Call .drain_begin only once in bdrv_drain_all_begin()
> 
> Laurent Vivier (1):
>   spapr: don't initialize PATB entry if max-cpu-compat < power9
> 
> Marcel Apfelbaum (1):
>   hw/pci-bridge: fix QEMU crash because of pcie-root-port
> 
> Markus Armbruster (1):
>   qemu-options: Remove stray colons from output of --help
> 
> Michael S. Tsirkin (1):
>   virtio_error: don't invoke status callbacks
> 
> Murilo Opsfelder Araujo (1):
>   block/nbd: fix segmentation fault when .desc is not null-terminated
> 
> Paolo Bonzini (2):
>   qemu-pr-helper: miscellaneous fixes
>   i386: Add support for SPEC_CTRL MSR
> 
> Peter Lieven (1):
>   block/iscsi: dont leave allocmap in an invalid state on UNMAP failure
> 
> Peter Maydell (8):
>   target/i386: Fix handling of VEX prefixes
>   hw/intc/arm_gicv3: Make reserved register addresses RAZ/WI
>   hw/intc/arm_gic: reserved register addresses are RAZ/WI
>   hw/sd/pl181: Reset SD card on controller reset
>   hw/sd/milkymist-memcard: Reset SD card on controller reset
>   hw/sd/ssi-sd: Reset SD card on controller reset
>   linux-user: Fix locking order in