Re: [Qemu-devel] [PATCH 00/12] Patch Round-up for stable 2.10.1, freeze on 2017-09-27

[Eric Blake]

On 09/19/2017 07:45 PM, Michael Roth wrote:
> Hi everyone,
> 
> The following new patches are queued for QEMU stable v2.10.1:
> 
>   https://github.com/mdroth/qemu/commits/stable-2.10-staging
> 
> The release is planned for 2017-10-02:
> 
>   https://wiki.qemu.org/Planning/2.10
> 
> Please respond here or CC qemu-sta...@nongnu.org on any patches you
> think should be included in the release.
> 

I just cc'd qemu-stable for 3c2d5183f9f, nbd-client: avoid read_reply_co
entry if send failed

-- 
Eric Blake, Principal Software Engineer
Red Hat, Inc.   +1-919-301-3266
Virtualization:  qemu.org | libvirt.org



signature.asc
Description: OpenPGP digital signature

Re: [Qemu-devel] [PATCH 00/12] Patch Round-up for stable 2.10.1, freeze on 2017-09-27

[Anthony PERARD]

On Tue, Sep 19, 2017 at 07:45:09PM -0500, Michael Roth wrote:
> Hi everyone,
> 
> The following new patches are queued for QEMU stable v2.10.1:
> 
>   https://github.com/mdroth/qemu/commits/stable-2.10-staging
> 
> The release is planned for 2017-10-02:
> 
>   https://wiki.qemu.org/Planning/2.10
> 
> Please respond here or CC qemu-sta...@nongnu.org on any patches you
> think should be included in the release.

Hi,

I'd like to add this patch:

a3fd781f653db4dc76635715207c9029208634bd
"xen-disk: use g_new0 to fix build"


And maybe these three patches to fix a regression compared to 2.8:

f5855994fee2f8815dc86b8453e4a63e290aea05
"hw/acpi: Limit hotplug to root bus on legacy mode"

ab938ae43f8a3a71a3525566edf586081b7a7452
"hw/acpi: Move acpi_set_pci_info to pcihp"

2bed1ba77fae50bc8b5e68ede2d80b652b30c3b8
"Revert "ACPI: don't call acpi_pcihp_device_plug_cb on xen""


There is also two more patches that are not commited yet, so I don't
known if they can be candidate for the stable release:

"xen: add a global indicator for grant copy being available"
"xen: dont try setting max grants multiple times"


Thanks,

-- 
Anthony PERARD

Re: [Qemu-devel] [PATCH 00/12] Patch Round-up for stable 2.10.1, freeze on 2017-09-27

[Michael Roth]

Quoting Michael Roth (2017-09-19 19:45:09)
> Hi everyone,
> 
> The following new patches are queued for QEMU stable v2.10.1:
> 
>   https://github.com/mdroth/qemu/commits/stable-2.10-staging
> 
> The release is planned for 2017-10-02:
> 
>   https://wiki.qemu.org/Planning/2.10

Thank you for the suggestions. The following additional patches are now queued:

  s390x/ais: for 2.10 stable: disable ais facility (Christian Borntraeger)
  9pfs: check the size of transport buffer before marshaling (Jan Dakinevich)
  9pfs: fix name_to_path assertion in v9fs_complete_rename() (Jan Dakinevich)
  9pfs: fix readdir() for 9p2000.u (Jan Dakinevich)
  console: fix dpy_gfx_replace_surface assert (Gerd Hoffmann)
  ide: ahci: unparent children buses before freeing their memory (Igor Mammedov)
  hw/ide/microdrive: Mark the dscm1 device with user_creatable = false 
(Thomas Huth)
  hw/arm/aspeed_soc: Mark devices as user_creatable = false (Thomas Huth)
  hw/arm/digic: Mark device with user_creatable = false (Thomas Huth)
  s390x/ipl: The s390-ipl device is not hot-pluggable (Thomas Huth)
  watchdog/wdt_diag288: Mark diag288 watchdog as non-hotpluggable (Thomas Huth)
  multiboot: validate multiboot header address values (Prasad J Pandit)
  vga: stop passing pointers to vga_draw_line* functions (Gerd Hoffmann)
  vga: fix display update region calculation (split screen) (Gerd Hoffmann)

The following patches are tagged for inclusion, but still awaiting pull
requests:

  block/qcow2-bitmap: fix use of uninitialized pointer
  migration: disable auto-converge during bulk block migration
  accel/tcg/cputlb: avoid recursive BQL (fixes #1706296)
  virtio/vhost: reset dev->log after syncing
  block/throttle-groups.c: allocate RestartData on the heap
  throttle-groups: update tg->any_timer_armed[] on detach

> 
> Please respond here or CC qemu-sta...@nongnu.org on any patches you
> think should be included in the release.
> 
> Testing/feedback is greatly appreciated.
> 
> Thanks!
> 
> 
> Alex Williamson (1):
>   vhost: Release memory references on cleanup
> 
> Farhan Ali (1):
>   s390-ccw: Fix alignment for CCW1
> 
> Greg Kurz (1):
>   virtfs: error out gracefully when mandatory suboptions are missing
> 
> Hannes Reinecke (1):
>   scsi-bus: correct responses for INQUIRY and REQUEST SENSE
> 
> Marc-André Lureau (2):
>   libvhost-user: support resuming vq->last_avail_idx based on used_idx
>   vhost-user-bridge: fix resume regression (since 2.9)
> 
> Pavel Butsykin (1):
>   qcow2: move qcow2_store_persistent_dirty_bitmaps() before cache flushing
> 
> Peter Maydell (1):
>   mps2-an511: Fix wiring of UART overflow interrupt lines
> 
> Pranith Kumar (1):
>   arm_gicv3_kvm: Fix compile warning
> 
> Richard Henderson (1):
>   target/arm: Fix aa64 ldp register writeback
> 
> Samuel Thibault (1):
>   slirp: fix clearing ifq_so from pending packets
> 
> Thomas Huth (1):
>   hw/arm/allwinner-a10: Mark the allwinner-a10 device with user_creatable 
> = false
> 
>  block/qcow2.c | 16 +++---
>  contrib/libvhost-user/libvhost-user.c | 13 
>  contrib/libvhost-user/libvhost-user.h |  7 +++
>  hw/arm/allwinner-a10.c|  2 ++
>  hw/arm/mps2.c |  4 ++--
>  hw/intc/arm_gicv3_kvm.c   |  2 +-
>  hw/scsi/scsi-bus.c| 29 ++
>  hw/virtio/vhost.c |  4 
>  pc-bios/s390-ccw/cio.h|  2 +-
>  scripts/device-crash-test |  1 -
>  slirp/socket.c| 39 
> +--
>  target/arm/translate-a64.c| 29 +++---
>  tests/vhost-user-bridge.c |  7 +++
>  vl.c  | 16 --
>  14 files changed, 120 insertions(+), 51 deletions(-)
> 
> 
> 

Re: [Qemu-devel] [PATCH 00/12] Patch Round-up for stable 2.10.1, freeze on 2017-09-27

[Greg Kurz]

On Tue, 19 Sep 2017 19:45:09 -0500
Michael Roth  wrote:

> Hi everyone,
> 
> The following new patches are queued for QEMU stable v2.10.1:
> 
>   https://github.com/mdroth/qemu/commits/stable-2.10-staging
> 
> The release is planned for 2017-10-02:
> 
>   https://wiki.qemu.org/Planning/2.10
> 
> Please respond here or CC qemu-sta...@nongnu.org on any patches you
> think should be included in the release.
> 

Hi Michael,

I'd like to suggest these patches that fix regressions introduced in 2.10:

6069537f4336a59054afda91a6545d3648c64619
"9pfs: fix readdir() for 9p2000.u"

4d8bc7334b06ef01a21cad3d1eb8dc183037a06b
"9pfs: fix name_to_path assertion in v9fs_complete_rename()"

772a73692ecb52bace0cff6f95df62f59b8cabe0
"9pfs: check the size of transport buffer before marshaling"

Cheers,

--
Greg

> Testing/feedback is greatly appreciated.
> 
> Thanks!
> 
> 
> Alex Williamson (1):
>   vhost: Release memory references on cleanup
> 
> Farhan Ali (1):
>   s390-ccw: Fix alignment for CCW1
> 
> Greg Kurz (1):
>   virtfs: error out gracefully when mandatory suboptions are missing
> 
> Hannes Reinecke (1):
>   scsi-bus: correct responses for INQUIRY and REQUEST SENSE
> 
> Marc-André Lureau (2):
>   libvhost-user: support resuming vq->last_avail_idx based on used_idx
>   vhost-user-bridge: fix resume regression (since 2.9)
> 
> Pavel Butsykin (1):
>   qcow2: move qcow2_store_persistent_dirty_bitmaps() before cache flushing
> 
> Peter Maydell (1):
>   mps2-an511: Fix wiring of UART overflow interrupt lines
> 
> Pranith Kumar (1):
>   arm_gicv3_kvm: Fix compile warning
> 
> Richard Henderson (1):
>   target/arm: Fix aa64 ldp register writeback
> 
> Samuel Thibault (1):
>   slirp: fix clearing ifq_so from pending packets
> 
> Thomas Huth (1):
>   hw/arm/allwinner-a10: Mark the allwinner-a10 device with user_creatable 
> = false
> 
>  block/qcow2.c | 16 +++---
>  contrib/libvhost-user/libvhost-user.c | 13 
>  contrib/libvhost-user/libvhost-user.h |  7 +++
>  hw/arm/allwinner-a10.c|  2 ++
>  hw/arm/mps2.c |  4 ++--
>  hw/intc/arm_gicv3_kvm.c   |  2 +-
>  hw/scsi/scsi-bus.c| 29 ++
>  hw/virtio/vhost.c |  4 
>  pc-bios/s390-ccw/cio.h|  2 +-
>  scripts/device-crash-test |  1 -
>  slirp/socket.c| 39 
> +--
>  target/arm/translate-a64.c| 29 +++---
>  tests/vhost-user-bridge.c |  7 +++
>  vl.c  | 16 --
>  14 files changed, 120 insertions(+), 51 deletions(-)
> 
> 
> 



-- 
Gregory Kurz kurzg...@fr.ibm.com
 gk...@linux.vnet.ibm.com
Software Engineer @ IBM/LTC  http://www.ibm.com
Tel 33-5-6218-1607

"Anarchy is about taking complete responsibility for yourself."
Alan Moore.

Re: [Qemu-devel] [PATCH 00/12] Patch Round-up for stable 2.10.1, freeze on 2017-09-27

[Thomas Huth]

On 20.09.2017 02:45, Michael Roth wrote:
> Hi everyone,
> 
> The following new patches are queued for QEMU stable v2.10.1:
> 
>   https://github.com/mdroth/qemu/commits/stable-2.10-staging
> 
> The release is planned for 2017-10-02:
> 
>   https://wiki.qemu.org/Planning/2.10
> 
> Please respond here or CC qemu-sta...@nongnu.org on any patches you
> think should be included in the release.

Looks like there have been some patches that fix CVEs, I think they
should be included, too:

e65294157d4b69393b3f819c99f4f647452b48e3
"vga: fix display update region calculation (split screen)"

3d90c6254863693a6b13d918d2b8682e08bbc681
"vga: stop passing pointers to vga_draw_line* functions"

ed4f86e8b6eff8e600c69adee68c7cd34dd2cccb
"multiboot: validate multiboot header address values"

?


Apart from that, I'd like to suggest the following commits, since
without these patches, QEMU can crash during runtime when the user
accidentally issues the wrong device_add/del commands:

84ebd3e8c7d4fe955b359b9aac84395907b0412e
"watchdog/wdt_diag288: Mark diag288 watchdog as non-hotpluggable"

0d4fa4996fc5ee56ea7d072e272b8e69948460a5
"s390x/ipl: The s390-ipl device is not hot-pluggable"

f58f25599b72c7479e6a1ff67c7f671823aa14da
"hw/arm/digic: Mark device with user_creatable = false"

469f3da42ef4af347fa7831e1cc0bd35d17f5b83
"hw/arm/aspeed_soc: Mark devices as user_creatable = false"

4c93950659487c7ad4f85571ee78524c1e3a94b3
"hw/ide/microdrive: Mark the dscm1 device with user_creatable = false"

955f5c7ba127746345a3d43b4d7c885ca159ae6b
"ide: ahci: unparent children buses before freeing their memory"


And searching the history for "assert", I think these are also good
candidates:

32b6943699948f7adc35ada233fbd25daffad5e9
"virtfs: error out gracefully when mandatory suboptions are missing"

1540008629bbb6a9c0826582d94ecf7a559f784c
"console: fix dpy_gfx_replace_surface assert"

 Thomas


> 
> 
> Alex Williamson (1):
>   vhost: Release memory references on cleanup
> 
> Farhan Ali (1):
>   s390-ccw: Fix alignment for CCW1
> 
> Greg Kurz (1):
>   virtfs: error out gracefully when mandatory suboptions are missing
> 
> Hannes Reinecke (1):
>   scsi-bus: correct responses for INQUIRY and REQUEST SENSE
> 
> Marc-André Lureau (2):
>   libvhost-user: support resuming vq->last_avail_idx based on used_idx
>   vhost-user-bridge: fix resume regression (since 2.9)
> 
> Pavel Butsykin (1):
>   qcow2: move qcow2_store_persistent_dirty_bitmaps() before cache flushing
> 
> Peter Maydell (1):
>   mps2-an511: Fix wiring of UART overflow interrupt lines
> 
> Pranith Kumar (1):
>   arm_gicv3_kvm: Fix compile warning
> 
> Richard Henderson (1):
>   target/arm: Fix aa64 ldp register writeback
> 
> Samuel Thibault (1):
>   slirp: fix clearing ifq_so from pending packets
> 
> Thomas Huth (1):
>   hw/arm/allwinner-a10: Mark the allwinner-a10 device with user_creatable 
> = false
> 
>  block/qcow2.c | 16 +++---
>  contrib/libvhost-user/libvhost-user.c | 13 
>  contrib/libvhost-user/libvhost-user.h |  7 +++
>  hw/arm/allwinner-a10.c|  2 ++
>  hw/arm/mps2.c |  4 ++--
>  hw/intc/arm_gicv3_kvm.c   |  2 +-
>  hw/scsi/scsi-bus.c| 29 ++
>  hw/virtio/vhost.c |  4 
>  pc-bios/s390-ccw/cio.h|  2 +-
>  scripts/device-crash-test |  1 -
>  slirp/socket.c| 39 
> +--
>  target/arm/translate-a64.c| 29 +++---
>  tests/vhost-user-bridge.c |  7 +++
>  vl.c  | 16 --
>  14 files changed, 120 insertions(+), 51 deletions(-)
> 
> 
>