OSDN Git Service

qmiga/qemu.git
3 years agovfio: Add ioctl to get dirty pages bitmap during dma unmap
Kirti Wankhede [Mon, 26 Oct 2020 09:36:25 +0000 (15:06 +0530)]
vfio: Add ioctl to get dirty pages bitmap during dma unmap

With vIOMMU, IO virtual address range can get unmapped while in pre-copy
phase of migration. In that case, unmap ioctl should return pages pinned
in that range and QEMU should find its correcponding guest physical
addresses and report those dirty.

Suggested-by: Alex Williamson <alex.williamson@redhat.com>
Signed-off-by: Kirti Wankhede <kwankhede@nvidia.com>
Reviewed-by: Neo Jia <cjia@nvidia.com>
[aw: fix error_report types, fix cpu_physical_memory_set_dirty_lebitmap() cast]
Signed-off-by: Alex Williamson <alex.williamson@redhat.com>
3 years agovfio: Dirty page tracking when vIOMMU is enabled
Kirti Wankhede [Mon, 26 Oct 2020 09:36:24 +0000 (15:06 +0530)]
vfio: Dirty page tracking when vIOMMU is enabled

When vIOMMU is enabled, register MAP notifier from log_sync when all
devices in container are in stop and copy phase of migration. Call replay
and get dirty pages from notifier callback.

Suggested-by: Alex Williamson <alex.williamson@redhat.com>
Signed-off-by: Kirti Wankhede <kwankhede@nvidia.com>
Reviewed-by: Yan Zhao <yan.y.zhao@intel.com>
Signed-off-by: Alex Williamson <alex.williamson@redhat.com>
3 years agovfio: Add vfio_listener_log_sync to mark dirty pages
Kirti Wankhede [Mon, 26 Oct 2020 09:36:23 +0000 (15:06 +0530)]
vfio: Add vfio_listener_log_sync to mark dirty pages

vfio_listener_log_sync gets list of dirty pages from container using
VFIO_IOMMU_GET_DIRTY_BITMAP ioctl and mark those pages dirty when all
devices are stopped and saving state.
Return early for the RAM block section of mapped MMIO region.

Signed-off-by: Kirti Wankhede <kwankhede@nvidia.com>
Reviewed-by: Neo Jia <cjia@nvidia.com>
[aw: fix error_report types, fix cpu_physical_memory_set_dirty_lebitmap() cast]
Signed-off-by: Alex Williamson <alex.williamson@redhat.com>
3 years agovfio: Add function to start and stop dirty pages tracking
Kirti Wankhede [Mon, 26 Oct 2020 09:36:22 +0000 (15:06 +0530)]
vfio: Add function to start and stop dirty pages tracking

Call VFIO_IOMMU_DIRTY_PAGES ioctl to start and stop dirty pages tracking
for VFIO devices.

Signed-off-by: Kirti Wankhede <kwankhede@nvidia.com>
Reviewed-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
Signed-off-by: Alex Williamson <alex.williamson@redhat.com>
3 years agovfio: Get migration capability flags for container
Kirti Wankhede [Mon, 26 Oct 2020 09:36:21 +0000 (15:06 +0530)]
vfio: Get migration capability flags for container

Added helper functions to get IOMMU info capability chain.
Added function to get migration capability information from that
capability chain for IOMMU container.

Similar change was proposed earlier:
https://lists.gnu.org/archive/html/qemu-devel/2018-05/msg03759.html

Disable migration for devices if IOMMU module doesn't support migration
capability.

Signed-off-by: Kirti Wankhede <kwankhede@nvidia.com>
Cc: Shameer Kolothum <shameerali.kolothum.thodi@huawei.com>
Cc: Eric Auger <eric.auger@redhat.com>
Signed-off-by: Alex Williamson <alex.williamson@redhat.com>
3 years agomemory: Set DIRTY_MEMORY_MIGRATION when IOMMU is enabled
Kirti Wankhede [Mon, 26 Oct 2020 09:36:20 +0000 (15:06 +0530)]
memory: Set DIRTY_MEMORY_MIGRATION when IOMMU is enabled

mr->ram_block is NULL when mr->is_iommu is true, then fr.dirty_log_mask
wasn't set correctly due to which memory listener's log_sync doesn't
get called.
This patch returns log_mask with DIRTY_MEMORY_MIGRATION set when
IOMMU is enabled.

Signed-off-by: Kirti Wankhede <kwankhede@nvidia.com>
Reviewed-by: Yan Zhao <yan.y.zhao@intel.com>
Acked-by: Paolo Bonzini <pbonzini@redhat.com>
Signed-off-by: Alex Williamson <alex.williamson@redhat.com>
3 years agovfio: Add load state functions to SaveVMHandlers
Kirti Wankhede [Mon, 26 Oct 2020 09:36:19 +0000 (15:06 +0530)]
vfio: Add load state functions to SaveVMHandlers

Sequence  during _RESUMING device state:
While data for this device is available, repeat below steps:
a. read data_offset from where user application should write data.
b. write data of data_size to migration region from data_offset.
c. write data_size which indicates vendor driver that data is written in
   staging buffer.

For user, data is opaque. User should write data in the same order as
received.

Signed-off-by: Kirti Wankhede <kwankhede@nvidia.com>
Reviewed-by: Neo Jia <cjia@nvidia.com>
Reviewed-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
Reviewed-by: Yan Zhao <yan.y.zhao@intel.com>
Signed-off-by: Alex Williamson <alex.williamson@redhat.com>
3 years agovfio: Add save state functions to SaveVMHandlers
Kirti Wankhede [Mon, 26 Oct 2020 09:36:18 +0000 (15:06 +0530)]
vfio: Add save state functions to SaveVMHandlers

Added .save_live_pending, .save_live_iterate and .save_live_complete_precopy
functions. These functions handles pre-copy and stop-and-copy phase.

In _SAVING|_RUNNING device state or pre-copy phase:
- read pending_bytes. If pending_bytes > 0, go through below steps.
- read data_offset - indicates kernel driver to write data to staging
  buffer.
- read data_size - amount of data in bytes written by vendor driver in
  migration region.
- read data_size bytes of data from data_offset in the migration region.
- Write data packet to file stream as below:
{VFIO_MIG_FLAG_DEV_DATA_STATE, data_size, actual data,
VFIO_MIG_FLAG_END_OF_STATE }

In _SAVING device state or stop-and-copy phase
a. read config space of device and save to migration file stream. This
   doesn't need to be from vendor driver. Any other special config state
   from driver can be saved as data in following iteration.
b. read pending_bytes. If pending_bytes > 0, go through below steps.
c. read data_offset - indicates kernel driver to write data to staging
   buffer.
d. read data_size - amount of data in bytes written by vendor driver in
   migration region.
e. read data_size bytes of data from data_offset in the migration region.
f. Write data packet as below:
   {VFIO_MIG_FLAG_DEV_DATA_STATE, data_size, actual data}
g. iterate through steps b to f while (pending_bytes > 0)
h. Write {VFIO_MIG_FLAG_END_OF_STATE}

When data region is mapped, its user's responsibility to read data from
data_offset of data_size before moving to next steps.

Added fix suggested by Artem Polyakov to reset pending_bytes in
vfio_save_iterate().
Added fix suggested by Zhi Wang to add 0 as data size in migration stream and
add END_OF_STATE delimiter to indicate phase complete.

Suggested-by: Artem Polyakov <artemp@nvidia.com>
Suggested-by: Zhi Wang <zhi.wang.linux@gmail.com>
Signed-off-by: Kirti Wankhede <kwankhede@nvidia.com>
Reviewed-by: Neo Jia <cjia@nvidia.com>
Reviewed-by: Yan Zhao <yan.y.zhao@intel.com>
Signed-off-by: Alex Williamson <alex.williamson@redhat.com>
3 years agovfio: Register SaveVMHandlers for VFIO device
Kirti Wankhede [Mon, 26 Oct 2020 09:36:17 +0000 (15:06 +0530)]
vfio: Register SaveVMHandlers for VFIO device

Define flags to be used as delimiter in migration stream for VFIO devices.
Added .save_setup and .save_cleanup functions. Map & unmap migration
region from these functions at source during saving or pre-copy phase.

Set VFIO device state depending on VM's state. During live migration, VM is
running when .save_setup is called, _SAVING | _RUNNING state is set for VFIO
device. During save-restore, VM is paused, _SAVING state is set for VFIO device.

Signed-off-by: Kirti Wankhede <kwankhede@nvidia.com>
Reviewed-by: Neo Jia <cjia@nvidia.com>
Reviewed-by: Cornelia Huck <cohuck@redhat.com>
Reviewed-by: Yan Zhao <yan.y.zhao@intel.com>
Signed-off-by: Alex Williamson <alex.williamson@redhat.com>
3 years agovfio: Add migration state change notifier
Kirti Wankhede [Mon, 26 Oct 2020 09:36:16 +0000 (15:06 +0530)]
vfio: Add migration state change notifier

Added migration state change notifier to get notification on migration state
change. These states are translated to VFIO device state and conveyed to
vendor driver.

Signed-off-by: Kirti Wankhede <kwankhede@nvidia.com>
Reviewed-by: Neo Jia <cjia@nvidia.com>
Reviewed-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
Reviewed-by: Cornelia Huck <cohuck@redhat.com>
Signed-off-by: Alex Williamson <alex.williamson@redhat.com>
3 years agovfio: Add VM state change handler to know state of VM
Kirti Wankhede [Mon, 26 Oct 2020 09:36:15 +0000 (15:06 +0530)]
vfio: Add VM state change handler to know state of VM

VM state change handler is called on change in VM's state. Based on
VM state, VFIO device state should be changed.
Added read/write helper functions for migration region.
Added function to set device_state.

Signed-off-by: Kirti Wankhede <kwankhede@nvidia.com>
Reviewed-by: Neo Jia <cjia@nvidia.com>
Reviewed-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
Reviewed-by: Cornelia Huck <cohuck@redhat.com>
[aw: lx -> HWADDR_PRIx, remove redundant parens]
Signed-off-by: Alex Williamson <alex.williamson@redhat.com>
3 years agovfio: Add migration region initialization and finalize function
Kirti Wankhede [Mon, 26 Oct 2020 09:36:14 +0000 (15:06 +0530)]
vfio: Add migration region initialization and finalize function

Whether the VFIO device supports migration or not is decided based of
migration region query. If migration region query is successful and migration
region initialization is successful then migration is supported else
migration is blocked.

Signed-off-by: Kirti Wankhede <kwankhede@nvidia.com>
Reviewed-by: Neo Jia <cjia@nvidia.com>
Acked-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
Reviewed-by: Cornelia Huck <cohuck@redhat.com>
Signed-off-by: Alex Williamson <alex.williamson@redhat.com>
3 years agovfio: Add save and load functions for VFIO PCI devices
Kirti Wankhede [Mon, 26 Oct 2020 09:36:13 +0000 (15:06 +0530)]
vfio: Add save and load functions for VFIO PCI devices

Added functions to save and restore PCI device specific data,
specifically config space of PCI device.

Signed-off-by: Kirti Wankhede <kwankhede@nvidia.com>
Reviewed-by: Neo Jia <cjia@nvidia.com>
Signed-off-by: Alex Williamson <alex.williamson@redhat.com>
3 years agovfio: Add vfio_get_object callback to VFIODeviceOps
Kirti Wankhede [Mon, 26 Oct 2020 09:36:12 +0000 (15:06 +0530)]
vfio: Add vfio_get_object callback to VFIODeviceOps

Hook vfio_get_object callback for PCI devices.

Signed-off-by: Kirti Wankhede <kwankhede@nvidia.com>
Reviewed-by: Neo Jia <cjia@nvidia.com>
Suggested-by: Cornelia Huck <cohuck@redhat.com>
Reviewed-by: Cornelia Huck <cohuck@redhat.com>
Signed-off-by: Alex Williamson <alex.williamson@redhat.com>
3 years agovfio: Add function to unmap VFIO region
Kirti Wankhede [Mon, 26 Oct 2020 09:36:11 +0000 (15:06 +0530)]
vfio: Add function to unmap VFIO region

This function will be used for migration region.
Migration region is mmaped when migration starts and will be unmapped when
migration is complete.

Signed-off-by: Kirti Wankhede <kwankhede@nvidia.com>
Reviewed-by: Neo Jia <cjia@nvidia.com>
Reviewed-by: Cornelia Huck <cohuck@redhat.com>
Signed-off-by: Alex Williamson <alex.williamson@redhat.com>
3 years agoMerge remote-tracking branch 'remotes/mst/tags/for_upstream' into staging
Peter Maydell [Sun, 1 Nov 2020 14:02:19 +0000 (14:02 +0000)]
Merge remote-tracking branch 'remotes/mst/tags/for_upstream' into staging

pc,pci,vhost,virtio: misc fixes

Just a bunch of bugfixes all over the place.

Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
# gpg: Signature made Fri 30 Oct 2020 12:44:31 GMT
# gpg:                using RSA key 5D09FD0871C8F85B94CA8A0D281F0DB8D28D5469
# gpg:                issuer "mst@redhat.com"
# gpg: Good signature from "Michael S. Tsirkin <mst@kernel.org>" [full]
# gpg:                 aka "Michael S. Tsirkin <mst@redhat.com>" [full]
# Primary key fingerprint: 0270 606B 6F3C DF3D 0B17  0970 C350 3912 AFBE 8E67
#      Subkey fingerprint: 5D09 FD08 71C8 F85B 94CA  8A0D 281F 0DB8 D28D 5469

* remotes/mst/tags/for_upstream:
  intel_iommu: Fix two misuse of "0x%u" prints
  virtio: skip guest index check on device load
  vhost-blk: set features before setting inflight feature
  pci: Disallow improper BAR registration for type 1
  pci: Change error_report to assert(3)
  pci: advertise a page aligned ATS
  pc: Implement -no-hpet as sugar for -machine hpet=on
  vhost: Don't special case vq->used_phys in vhost_get_log_size()
  pci: Assert irqnum is between 0 and bus->nirqs in pci_bus_change_irq_level
  hw/pci: Extract pci_bus_change_irq_level() from pci_change_irq_level()
  hw/virtio/vhost-vdpa: Fix Coverity CID 1432864
  acpi/crs: Support ranges > 32b for hosts
  acpi/crs: Prevent bad ranges for host bridges
  vhost-vsock: set vhostfd to non-blocking mode
  vhost-vdpa: negotiate VIRTIO_NET_F_STATUS with driver

Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
3 years agoMerge remote-tracking branch 'remotes/kraxel/tags/modules-20201029-pull-request'...
Peter Maydell [Sat, 31 Oct 2020 20:32:56 +0000 (20:32 +0000)]
Merge remote-tracking branch 'remotes/kraxel/tags/modules-20201029-pull-request' into staging

modules: build virtio-gpu-pci & virtio-vga modular.
modules: various bugfixes, mostly for macos.

# gpg: Signature made Thu 29 Oct 2020 11:09:41 GMT
# gpg:                using RSA key 4CB6D8EED3E87138
# gpg: Good signature from "Gerd Hoffmann (work) <kraxel@redhat.com>" [full]
# gpg:                 aka "Gerd Hoffmann <gerd@kraxel.org>" [full]
# gpg:                 aka "Gerd Hoffmann (private) <kraxel@gmail.com>" [full]
# Primary key fingerprint: A032 8CFF B93A 17A7 9901  FE7D 4CB6 D8EE D3E8 7138

* remotes/kraxel/tags/modules-20201029-pull-request:
  modules: turn off lazy binding
  modules: unbreak them on macos
  virtio-gpu: only compile virtio-gpu-3d.c for CONFIG_VIRGL=y
  virtio-gpu: add virtio-vga module
  virtio-gpu: add virtio-gpu-pci module

Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
3 years agoMerge remote-tracking branch 'remotes/berrange-gitlab/tags/misc-next-pull-request...
Peter Maydell [Sat, 31 Oct 2020 14:55:10 +0000 (14:55 +0000)]
Merge remote-tracking branch 'remotes/berrange-gitlab/tags/misc-next-pull-request' into staging

Misc fixes

 * Improve socket cnnection failure error reporting
 * Fix LGPL version number

# gpg: Signature made Thu 29 Oct 2020 10:00:47 GMT
# gpg:                using RSA key DAF3A6FDB26B62912D0E8E3FBE86EBB415104FDF
# gpg: Good signature from "Daniel P. Berrange <dan@berrange.com>" [full]
# gpg:                 aka "Daniel P. Berrange <berrange@redhat.com>" [full]
# Primary key fingerprint: DAF3 A6FD B26B 6291 2D0E  8E3F BE86 EBB4 1510 4FDF

* remotes/berrange-gitlab/tags/misc-next-pull-request:
  util: include the target address in socket connect failures
  io: Don't use '#' flag of printf format
  authz: Fix Lesser GPL version number
  crypt: Fix Lesser GPL version number
  io: Fix Lesser GPL version number

Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
3 years agoMerge remote-tracking branch 'remotes/mcayland/tags/qemu-sparc-20201028' into staging
Peter Maydell [Sat, 31 Oct 2020 09:48:20 +0000 (09:48 +0000)]
Merge remote-tracking branch 'remotes/mcayland/tags/qemu-sparc-20201028' into staging

qemu-sparc queue

# gpg: Signature made Wed 28 Oct 2020 08:17:23 GMT
# gpg:                using RSA key CC621AB98E82200D915CC9C45BC2C56FAE0F321F
# gpg:                issuer "mark.cave-ayland@ilande.co.uk"
# gpg: Good signature from "Mark Cave-Ayland <mark.cave-ayland@ilande.co.uk>" [full]
# Primary key fingerprint: CC62 1AB9 8E82 200D 915C  C9C4 5BC2 C56F AE0F 321F

* remotes/mcayland/tags/qemu-sparc-20201028:
  hw/pci-host/sabre: Simplify code initializing variable once
  hw/pci-host/sabre: Remove superfluous address range check
  hw/pci-host/sabre: Update documentation link
  sabre: increase number of PCI bus IRQs from 32 to 64
  hw/display/tcx: Allow 64-bit accesses to framebuffer stippler and blitter
  sabre: don't call sysbus_mmio_map() in sabre_realize()
  sparc32-ledma: don't reference nd_table directly within the device
  sparc32-espdma: use object_initialize_child() for esp child object
  sparc32-ledma: use object_initialize_child() for lance child object
  sparc32-dma: use object_initialize_child() for espdma and ledma child objects

Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
3 years agoMerge remote-tracking branch 'remotes/rth-gitlab/tags/pull-tcg-20201027' into staging
Peter Maydell [Fri, 30 Oct 2020 19:47:54 +0000 (19:47 +0000)]
Merge remote-tracking branch 'remotes/rth-gitlab/tags/pull-tcg-20201027' into staging

Optimize across branches.
Add logging for cpu_io_recompile.

# gpg: Signature made Tue 27 Oct 2020 16:48:50 GMT
# gpg:                using RSA key 7A481E78868B4DB6A85A05C064DF38E8AF7E215F
# gpg:                issuer "richard.henderson@linaro.org"
# gpg: Good signature from "Richard Henderson <richard.henderson@linaro.org>" [full]
# Primary key fingerprint: 7A48 1E78 868B 4DB6 A85A  05C0 64DF 38E8 AF7E 215F

* remotes/rth-gitlab/tags/pull-tcg-20201027:
  accel/tcg: Add CPU_LOG_EXEC tracing for cpu_io_recompile()
  tcg/optimize: Flush data at labels not TCG_OPF_BB_END
  tcg: Do not kill globals at conditional branches

Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
3 years agoMerge remote-tracking branch 'remotes/vivier2/tags/trivial-branch-for-5.2-pull-reques...
Peter Maydell [Fri, 30 Oct 2020 15:49:35 +0000 (15:49 +0000)]
Merge remote-tracking branch 'remotes/vivier2/tags/trivial-branch-for-5.2-pull-request' into staging

Pull request trivial branch 20201027

# gpg: Signature made Tue 27 Oct 2020 16:29:47 GMT
# gpg:                using RSA key CD2F75DDC8E3A4DC2E4F5173F30C38BD3F2FBE3C
# gpg:                issuer "laurent@vivier.eu"
# gpg: Good signature from "Laurent Vivier <lvivier@redhat.com>" [full]
# gpg:                 aka "Laurent Vivier <laurent@vivier.eu>" [full]
# gpg:                 aka "Laurent Vivier (Red Hat) <lvivier@redhat.com>" [full]
# Primary key fingerprint: CD2F 75DD C8E3 A4DC 2E4F  5173 F30C 38BD 3F2F BE3C

* remotes/vivier2/tags/trivial-branch-for-5.2-pull-request:
  CHANGELOG: remove disused file
  qdev: Fix two typos
  scripts/qmp: delete 'qmp' script
  cryptodev: Fix Lesser GPL version number
  io: Fix Lesser GPL version number
  cpus: Drop declaration of cpu_remove()
  Makefile: Add *.[ch].inc files to cscope/ctags/TAGS
  elf2dmp: Fix memory leak on main() error paths

Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
3 years agoMerge remote-tracking branch 'remotes/kevin/tags/for-upstream' into staging
Peter Maydell [Fri, 30 Oct 2020 14:36:52 +0000 (14:36 +0000)]
Merge remote-tracking branch 'remotes/kevin/tags/for-upstream' into staging

Block layer patches:

- qcow2: Skip copy-on-write when allocating a zero cluster
- qemu-img: add support for rate limit in qemu-img convert/commit
- Fix deadlock when deleting a block node during drain_all

# gpg: Signature made Tue 27 Oct 2020 15:14:07 GMT
# gpg:                using RSA key DC3DEB159A9AF95D3D7456FE7F09B272C88F2FD6
# gpg:                issuer "kwolf@redhat.com"
# gpg: Good signature from "Kevin Wolf <kwolf@redhat.com>" [full]
# Primary key fingerprint: DC3D EB15 9A9A F95D 3D74  56FE 7F09 B272 C88F 2FD6

* remotes/kevin/tags/for-upstream:
  block: End quiescent sections when a BDS is deleted
  qcow2: Skip copy-on-write when allocating a zero cluster
  qcow2: Report BDRV_BLOCK_ZERO more accurately in bdrv_co_block_status()
  qemu-img: add support for rate limit in qemu-img convert
  qemu-img: add support for rate limit in qemu-img commit

Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
3 years agoMerge remote-tracking branch 'remotes/jsnow-gitlab/tags/ide-pull-request' into staging
Peter Maydell [Fri, 30 Oct 2020 11:56:07 +0000 (11:56 +0000)]
Merge remote-tracking branch 'remotes/jsnow-gitlab/tags/ide-pull-request' into staging

IDE Pull request

# gpg: Signature made Tue 27 Oct 2020 14:41:59 GMT
# gpg:                using RSA key F9B7ABDBBCACDF95BE76CBD07DEF8106AAFC390E
# gpg: Good signature from "John Snow (John Huston) <jsnow@redhat.com>" [full]
# Primary key fingerprint: FAEB 9711 A12C F475 812F  18F2 88A9 064D 1835 61EB
#      Subkey fingerprint: F9B7 ABDB BCAC DF95 BE76  CBD0 7DEF 8106 AAFC 390E

* remotes/jsnow-gitlab/tags/ide-pull-request:
  ide: clear SRST after SRST finishes
  ide: perform SRST as early as possible
  ide: run diagnostic after SRST

Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
3 years agointel_iommu: Fix two misuse of "0x%u" prints
Peter Xu [Mon, 19 Oct 2020 17:39:22 +0000 (13:39 -0400)]
intel_iommu: Fix two misuse of "0x%u" prints

Dave magically found this.  Fix them with "0x%x".

Reported-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
Signed-off-by: Peter Xu <peterx@redhat.com>
Message-Id: <20201019173922.100270-1-peterx@redhat.com>
Reviewed-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
Reviewed-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Acked-by: Jason Wang <jasowang@redhat.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
3 years agovirtio: skip guest index check on device load
Felipe Franciosi [Wed, 28 Oct 2020 13:47:03 +0000 (13:47 +0000)]
virtio: skip guest index check on device load

QEMU must be careful when loading device state off migration streams to
prevent a malicious source from exploiting the emulator. Overdoing these
checks has the side effect of allowing a guest to "pin itself" in cloud
environments by messing with state which is entirely in its control.

Similarly to what f3081539 achieved in usb_device_post_load(), this
commit removes such a check from virtio_load(). Worth noting, the result
of a load without this check is the same as if a guest enables a VQ with
invalid indexes to begin with. That is, the virtual device is set in a
broken state (by the datapath handler) and must be reset.

Signed-off-by: Felipe Franciosi <felipe@nutanix.com>
Message-Id: <20201028134643.110698-1-felipe@nutanix.com>
Acked-by: Jason Wang <jasowang@redhat.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
3 years agovhost-blk: set features before setting inflight feature
Jin Yu [Thu, 10 Sep 2020 13:48:51 +0000 (21:48 +0800)]
vhost-blk: set features before setting inflight feature

Virtqueue has split and packed, so before setting inflight,
you need to inform the back-end virtqueue format.

Signed-off-by: Jin Yu <jin.yu@intel.com>
Message-Id: <20200910134851.7817-1-jin.yu@intel.com>
Acked-by: Raphael Norwitz <raphael.norwitz@nutanix.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
3 years agopci: Disallow improper BAR registration for type 1
Ben Widawsky [Thu, 15 Oct 2020 18:14:11 +0000 (11:14 -0700)]
pci: Disallow improper BAR registration for type 1

Prevent future developers working on root complexes, root ports, or
bridges that also wish to implement a BAR for those, from shooting
themselves in the foot. PCI type 1 headers only support 2 base address
registers. It is incorrect and difficult to figure out what is wrong
with the device when this mistake is made. With this, it is immediate
and obvious what has gone wrong.

Signed-off-by: Ben Widawsky <ben.widawsky@intel.com>
Message-Id: <20201015181411.89104-2-ben.widawsky@intel.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
3 years agopci: Change error_report to assert(3)
Ben Widawsky [Thu, 15 Oct 2020 18:14:10 +0000 (11:14 -0700)]
pci: Change error_report to assert(3)

Asserts are used for developer bugs. As registering a bar of the wrong
size is not something that should be possible for a user to achieve,
this is a developer bug.

While here, use the more obvious helper function.

Signed-off-by: Ben Widawsky <ben.widawsky@intel.com>
Message-Id: <20201015181411.89104-1-ben.widawsky@intel.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Reviewed-by: Philippe Mathieu-Daudé <philmd@redhat.com>
3 years agopci: advertise a page aligned ATS
Jason Wang [Wed, 9 Sep 2020 08:17:31 +0000 (16:17 +0800)]
pci: advertise a page aligned ATS

After Linux kernel commit 61363c1474b1 ("iommu/vt-d: Enable ATS only
if the device uses page aligned address."), ATS will be only enabled
if device advertises a page aligned request.

Unfortunately, vhost-net is the only user and we don't advertise the
aligned request capability in the past since both vhost IOTLB and
address_space_get_iotlb_entry() can support non page aligned request.

Though it's not clear that if the above kernel commit makes
sense. Let's advertise a page aligned ATS here to make vhost device
IOTLB work with Intel IOMMU again.

Note that in the future we may extend pcie_ats_init() to accept
parameters like queue depth and page alignment.

Cc: qemu-stable@nongnu.org
Signed-off-by: Jason Wang <jasowang@redhat.com>
Message-Id: <20200909081731.24688-1-jasowang@redhat.com>
Reviewed-by: Peter Xu <peterx@redhat.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
3 years agopc: Implement -no-hpet as sugar for -machine hpet=on
Eduardo Habkost [Wed, 21 Oct 2020 14:47:16 +0000 (10:47 -0400)]
pc: Implement -no-hpet as sugar for -machine hpet=on

Get rid of yet another global variable.

The default will be hpet=on only if CONFIG_HPET=y.

Signed-off-by: Eduardo Habkost <ehabkost@redhat.com>
Message-Id: <20201021144716.1536388-1-ehabkost@redhat.com>
Acked-by: Paolo Bonzini <pbonzini@redhat.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
3 years agovhost: Don't special case vq->used_phys in vhost_get_log_size()
Greg Kurz [Wed, 7 Oct 2020 16:30:34 +0000 (18:30 +0200)]
vhost: Don't special case vq->used_phys in vhost_get_log_size()

The first loop in vhost_get_log_size() computes the size of the dirty log
bitmap so that it allows to track changes in the entire guest memory, in
terms of GPA.

When not using a vIOMMU, the address of the vring's used structure,
vq->used_phys, is a GPA. It is thus already covered by the first loop.

When using a vIOMMU, vq->used_phys is a GIOVA that will be translated
to an HVA when the vhost backend needs to update the used structure. It
will log the corresponding GPAs into the bitmap but it certainly won't
log the GIOVA.

So in any case, vq->used_phys shouldn't be explicitly used to size the
bitmap. Drop the second loop.

This fixes a crash of the source when migrating a guest using in-kernel
vhost-net and iommu_platform=on on POWER, because DMA regions are put
over 0x800000000000000ULL. The resulting insanely huge log size causes
g_malloc0() to abort.

BugLink: https://bugzilla.redhat.com/show_bug.cgi?id=1879349
Signed-off-by: Greg Kurz <groug@kaod.org>
Message-Id: <160208823418.29027.15172801181796272300.stgit@bahia.lan>
Acked-by: Jason Wang <jasowang@redhat.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
3 years agopci: Assert irqnum is between 0 and bus->nirqs in pci_bus_change_irq_level
Mark Cave-Ayland [Sat, 24 Oct 2020 20:39:00 +0000 (22:39 +0200)]
pci: Assert irqnum is between 0 and bus->nirqs in pci_bus_change_irq_level

These assertions similar to those in the adjacent pci_bus_get_irq_level()
function ensure that irqnum lies within the valid PCI bus IRQ range.

Signed-off-by: Mark Cave-Ayland <mark.cave-ayland@ilande.co.uk>
Message-Id: <20201011082022.3016-1-mark.cave-ayland@ilande.co.uk>
Signed-off-by: Philippe Mathieu-Daudé <f4bug@amsat.org>
Message-Id: <20201024203900.3619498-3-f4bug@amsat.org>
Reviewed-by: Mark Cave-Ayland <mark.cave-ayland@ilande.co.uk>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
3 years agohw/pci: Extract pci_bus_change_irq_level() from pci_change_irq_level()
Philippe Mathieu-Daudé [Sat, 24 Oct 2020 20:38:59 +0000 (22:38 +0200)]
hw/pci: Extract pci_bus_change_irq_level() from pci_change_irq_level()

Extract pci_bus_change_irq_level() from pci_change_irq_level() to
make it clearer it operates on the bus.

Reported-by: Mark Cave-Ayland <mark.cave-ayland@ilande.co.uk>
Signed-off-by: Philippe Mathieu-Daudé <f4bug@amsat.org>
Message-Id: <20201024203900.3619498-2-f4bug@amsat.org>
Reviewed-by: Mark Cave-Ayland <mark.cave-ayland@ilande.co.uk>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
3 years agohw/virtio/vhost-vdpa: Fix Coverity CID 1432864
Philippe Mathieu-Daudé [Wed, 28 Oct 2020 15:40:04 +0000 (16:40 +0100)]
hw/virtio/vhost-vdpa: Fix Coverity CID 1432864

Fix uninitialized value issues reported by Coverity:

  Field 'msg.reserved' is uninitialized when calling write().

Fixes: a5bd05800f8 ("vhost-vdpa: batch updating IOTLB mappings")
Reported-by: Coverity (CID 1432864: UNINIT)
Signed-off-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Message-Id: <20201028154004.776760-1-philmd@redhat.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
3 years agoacpi/crs: Support ranges > 32b for hosts
Ben Widawsky [Mon, 26 Oct 2020 19:39:24 +0000 (12:39 -0700)]
acpi/crs: Support ranges > 32b for hosts

According to PCIe spec 5.0 Type 1 header space Base Address Registers
are defined by 7.5.1.2.1 Base Address Registers (same as Type 0). The
_CRS region should allow for the same range (up to 64b). Prior to this
change, any host bridge utilizing more than 32b for the BAR would have
the address truncated and likely lead to conflicts when the operating
systems reads the _CRS object.

Signed-off-by: Ben Widawsky <ben.widawsky@intel.com>
Message-Id: <20201026193924.985014-2-ben.widawsky@intel.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Reviewed-by: Igor Mammedov <imammedo@redhat.com>
3 years agoacpi/crs: Prevent bad ranges for host bridges
Ben Widawsky [Mon, 26 Oct 2020 19:39:23 +0000 (12:39 -0700)]
acpi/crs: Prevent bad ranges for host bridges

Prevent _CRS resources being quietly chopped off and instead throw an
assertion. _CRS is used by host bridges to declare regions of io and/or
memory that they consume. On some (all?) platforms the host bridge
doesn't have PCI header space and so they need some way to convey the
information.

Signed-off-by: Ben Widawsky <ben.widawsky@intel.com>
Message-Id: <20201026193924.985014-1-ben.widawsky@intel.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Acked-by: Igor Mammedov <imammedo@redhat.com>
3 years agovhost-vsock: set vhostfd to non-blocking mode
Stefano Garzarella [Thu, 29 Oct 2020 14:48:49 +0000 (15:48 +0100)]
vhost-vsock: set vhostfd to non-blocking mode

vhost IOTLB API uses read()/write() to exchange iotlb messages with
the kernel module.
The QEMU implementation expects a non-blocking fd, indeed commit
c471ad0e9b ("vhost_net: device IOTLB support") set it for vhost-net.

Without this patch, if we enable iommu for the vhost-vsock device,
QEMU can hang when exchanging IOTLB messages.

As commit 894022e616 ("net: check if the file descriptor is valid
before using it") did for tap, let's use qemu_try_set_nonblock()
when fd is provided by the user.

Signed-off-by: Stefano Garzarella <sgarzare@redhat.com>
Message-Id: <20201029144849.70958-1-sgarzare@redhat.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
3 years agovhost-vdpa: negotiate VIRTIO_NET_F_STATUS with driver
Si-Wei Liu [Thu, 1 Oct 2020 20:09:45 +0000 (16:09 -0400)]
vhost-vdpa: negotiate VIRTIO_NET_F_STATUS with driver

Vendor driver may not support or implement config
interrupt delivery for link status notifications.
In this event, vendor driver is expected to NACK
the feature, but guest will keep link always up.

Signed-off-by: Si-Wei Liu <si-wei.liu@oracle.com>
Message-Id: <1601582985-14944-1-git-send-email-si-wei.liu@oracle.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
3 years agoMerge remote-tracking branch 'remotes/dgibson/tags/ppc-for-5.2-20201028' into staging
Peter Maydell [Thu, 29 Oct 2020 14:30:58 +0000 (14:30 +0000)]
Merge remote-tracking branch 'remotes/dgibson/tags/ppc-for-5.2-20201028' into staging

ppc patch queue 2020-10-28

Here's the next pull request for ppc and spapr related patches, which
should be the last things for soft freeze.  Includes:

 * Numerous error handling cleanups from Greg Kurz
 * Cleanups to cpu realization and hotplug handling from Greg Kurz
 * A handful of other small fixes and cleanups

This does include a change to pc_dimm_plug() that isn't in my normal
areas of concern.  That's there as a a prerequisite for ppc specific
changes, and has an ack from Igor.

# gpg: Signature made Tue 27 Oct 2020 14:13:21 GMT
# gpg:                using RSA key 75F46586AE61A66CC44E87DC6C38CACA20D9B392
# gpg: Good signature from "David Gibson <david@gibson.dropbear.id.au>" [full]
# gpg:                 aka "David Gibson (Red Hat) <dgibson@redhat.com>" [full]
# gpg:                 aka "David Gibson (ozlabs.org) <dgibson@ozlabs.org>" [full]
# gpg:                 aka "David Gibson (kernel.org) <dwg@kernel.org>" [unknown]
# Primary key fingerprint: 75F4 6586 AE61 A66C C44E  87DC 6C38 CACA 20D9 B392

* remotes/dgibson/tags/ppc-for-5.2-20201028:
  ppc/: fix some comment spelling errors
  spapr: Improve spapr_reallocate_hpt() error reporting
  target/ppc: Fix kvmppc_load_htab_chunk() error reporting
  spapr: Use error_append_hint() in spapr_reallocate_hpt()
  spapr: Simplify error handling in spapr_memory_plug()
  spapr: Pass &error_abort when getting some PC DIMM properties
  spapr: Use appropriate getter for PC_DIMM_SLOT_PROP
  spapr: Use appropriate getter for PC_DIMM_ADDR_PROP
  pc-dimm: Drop @errp argument of pc_dimm_plug()
  spapr: Simplify spapr_cpu_core_realize() and spapr_cpu_core_unrealize()
  spapr: Make spapr_cpu_core_unrealize() idempotent
  spapr: Drop spapr_delete_vcpu() unused argument
  spapr: Unrealize vCPUs with qdev_unrealize()
  spapr: Fix leak of CPU machine specific data
  spapr: Move spapr_create_nvdimm_dr_connectors() to core machine code
  hw/net: move allocation to the heap due to very large stack frame
  ppc/spapr: re-assert IRQs during event-scan if there are pending
  spapr: Clarify why DR connectors aren't user creatable

Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
3 years agoMerge remote-tracking branch 'remotes/pmaydell/tags/pull-target-arm-20201027-1' into...
Peter Maydell [Thu, 29 Oct 2020 11:40:04 +0000 (11:40 +0000)]
Merge remote-tracking branch 'remotes/pmaydell/tags/pull-target-arm-20201027-1' into staging

target-arm queue:
 * raspi: add model of cprman clock manager
 * sbsa-ref: add an SBSA generic watchdog device
 * arm/trace: Fix hex printing
 * raspi: Add models of Pi 3 model A+, Pi Zero and Pi A+
 * hw/arm/smmuv3: Set the restoration priority of the vSMMUv3 explicitly
 * Nuvoton NPCM7xx: Add USB, RNG, GPIO and watchdog support
 * hw/arm: fix min_cpus for xlnx-versal-virt platform
 * hw/arm/highbank: Silence warnings about missing fallthrough statements
 * linux-user: Support Aarch64 BTI
 * Armv7M systick: fix corner case bugs by rewriting to use ptimer

# gpg: Signature made Tue 27 Oct 2020 11:27:10 GMT
# gpg:                using RSA key E1A5C593CD419DE28E8315CF3C2525ED14360CDE
# gpg:                issuer "peter.maydell@linaro.org"
# gpg: Good signature from "Peter Maydell <peter.maydell@linaro.org>" [ultimate]
# gpg:                 aka "Peter Maydell <pmaydell@gmail.com>" [ultimate]
# gpg:                 aka "Peter Maydell <pmaydell@chiark.greenend.org.uk>" [ultimate]
# Primary key fingerprint: E1A5 C593 CD41 9DE2 8E83  15CF 3C25 25ED 1436 0CDE

* remotes/pmaydell/tags/pull-target-arm-20201027-1: (48 commits)
  hw/timer/armv7m_systick: Rewrite to use ptimers
  hw/core/ptimer: Support ptimer being disabled by timer callback
  hw/arm/sbsa-ref: add SBSA watchdog device
  hw/watchdog: Implement SBSA watchdog device
  hw/arm/bcm2835_peripherals: connect the UART clock
  hw/char/pl011: add a clock input
  hw/misc/bcm2835_cprman: add sane reset values to the registers
  hw/misc/bcm2835_cprman: add the DSI0HSCK multiplexer
  hw/misc/bcm2835_cprman: implement clock mux behaviour
  hw/misc/bcm2835_cprman: add a clock mux skeleton implementation
  hw/misc/bcm2835_cprman: implement PLL channels behaviour
  hw/misc/bcm2835_cprman: add a PLL channel skeleton implementation
  hw/misc/bcm2835_cprman: implement PLLs behaviour
  hw/misc/bcm2835_cprman: add a PLL skeleton implementation
  hw/arm/raspi: add a skeleton implementation of the CPRMAN
  hw/arm/raspi: fix CPRMAN base address
  hw/core/clock: trace clock values in Hz instead of ns
  hw/core/clock: provide the VMSTATE_ARRAY_CLOCK macro
  arm/trace: Fix hex printing
  hw/arm/raspi: Add the Raspberry Pi 3 model A+
  ...

Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
3 years agoMerge remote-tracking branch 'remotes/armbru/tags/pull-qmp-2020-10-27' into staging
Peter Maydell [Thu, 29 Oct 2020 10:03:32 +0000 (10:03 +0000)]
Merge remote-tracking branch 'remotes/armbru/tags/pull-qmp-2020-10-27' into staging

QMP patches patches for 2020-10-27

# gpg: Signature made Tue 27 Oct 2020 10:13:00 GMT
# gpg:                using RSA key 354BC8B3D7EB2A6B68674E5F3870B400EB918653
# gpg:                issuer "armbru@redhat.com"
# gpg: Good signature from "Markus Armbruster <armbru@redhat.com>" [full]
# gpg:                 aka "Markus Armbruster <armbru@pond.sub.org>" [full]
# Primary key fingerprint: 354B C8B3 D7EB 2A6B 6867  4E5F 3870 B400 EB91 8653

* remotes/armbru/tags/pull-qmp-2020-10-27:
  scripts/qmp: delete 'qmp' script
  qmp-shell: Sort by key when pretty-printing

Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
3 years agoutil: include the target address in socket connect failures
Daniel P. Berrangé [Fri, 18 Sep 2020 09:31:45 +0000 (10:31 +0100)]
util: include the target address in socket connect failures

Reporting "Failed to connect socket" is essentially useless for a user
attempting to diagnose failure. It needs to include the target address
details. Similarly when failing to create a socket we should include the
socket family info, so the user understands what particular feature was
missing in their kernel build (IPv6, VSock in particular).

Reviewed-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Signed-off-by: Daniel P. Berrangé <berrange@redhat.com>
3 years agoio: Don't use '#' flag of printf format
AlexChen [Mon, 19 Oct 2020 12:12:02 +0000 (20:12 +0800)]
io: Don't use '#' flag of printf format

Reviewed-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Signed-off-by: AlexChen <alex.chen@huawei.com>
Signed-off-by: Daniel P. Berrangé <berrange@redhat.com>
3 years agoauthz: Fix Lesser GPL version number
Chetan Pant [Wed, 14 Oct 2020 13:49:02 +0000 (13:49 +0000)]
authz: Fix Lesser GPL version number

There is no "version 2" of the "Lesser" General Public License.
It is either "GPL version 2.0" or "Lesser GPL version 2.1".
This patch replaces all occurrences of "Lesser GPL version 2" with
"Lesser GPL version 2.1" in comment section.

Signed-off-by: Chetan Pant <chetan4windows@gmail.com>
Signed-off-by: Daniel P. Berrangé <berrange@redhat.com>
3 years agocrypt: Fix Lesser GPL version number
Chetan Pant [Wed, 14 Oct 2020 13:42:48 +0000 (13:42 +0000)]
crypt: Fix Lesser GPL version number

There is no "version 2" of the "Lesser" General Public License.
It is either "GPL version 2.0" or "Lesser GPL version 2.1".
This patch replaces all occurrences of "Lesser GPL version 2" with
"Lesser GPL version 2.1" in comment section.

Signed-off-by: Chetan Pant <chetan4windows@gmail.com>
Signed-off-by: Daniel P. Berrangé <berrange@redhat.com>
3 years agoio: Fix Lesser GPL version number
Chetan Pant [Wed, 14 Oct 2020 13:40:33 +0000 (13:40 +0000)]
io: Fix Lesser GPL version number

There is no "version 2" of the "Lesser" General Public License.
It is either "GPL version 2.0" or "Lesser GPL version 2.1".
This patch replaces all occurrences of "Lesser GPL version 2" with
"Lesser GPL version 2.1" in comment section.

Signed-off-by: Chetan Pant <chetan4windows@gmail.com>
Signed-off-by: Daniel P. Berrangé <berrange@redhat.com>
3 years agomodules: turn off lazy binding
Gerd Hoffmann [Wed, 28 Oct 2020 05:49:44 +0000 (06:49 +0100)]
modules: turn off lazy binding

We want missing symbols fail module load right away instead of having
qemu abort later on in case lazy binding fails.  Can happen -- for
example -- when trying to load a module for a pci device
(virtio-gpu-pci) into a qemu without pci support (qemu-system-avr).

Signed-off-by: Gerd Hoffmann <kraxel@redhat.com>
Message-id: 20201028054944.5772-1-kraxel@redhat.com

3 years agomodules: unbreak them on macos
Gerd Hoffmann [Tue, 27 Oct 2020 12:06:03 +0000 (13:06 +0100)]
modules: unbreak them on macos

Using the correct shared library suffix helps ;)

Signed-off-by: Gerd Hoffmann <kraxel@redhat.com>
Message-id: 20201027120603.3625-1-kraxel@redhat.com

3 years agovirtio-gpu: only compile virtio-gpu-3d.c for CONFIG_VIRGL=y
Gerd Hoffmann [Mon, 26 Oct 2020 14:28:51 +0000 (15:28 +0100)]
virtio-gpu: only compile virtio-gpu-3d.c for CONFIG_VIRGL=y

There is no actual code in the CONFIG_VIRGL=n case.  So building is
(a) pointless and (b) makes macos ranlib complain.

Reported-by: Paolo Bonzini <pbonzini@redhat.com>
Signed-off-by: Gerd Hoffmann <kraxel@redhat.com>
Reviewed-by: Marc-André Lureau <marcandre.lureau@redhat.com>
Message-id: 20201026142851.28735-1-kraxel@redhat.com

3 years agovirtio-gpu: add virtio-vga module
Gerd Hoffmann [Fri, 23 Oct 2020 06:46:18 +0000 (08:46 +0200)]
virtio-gpu: add virtio-vga module

Build virtio-gpu vga devices modular.  Must be a separate module because
not all qemu softmmu variants come with VGA support.

Signed-off-by: Gerd Hoffmann <kraxel@redhat.com>
Reviewed-by: Marc-André Lureau <marcandre.lureau@redhat.com>
Message-id: 20201023064618.21409-3-kraxel@redhat.com

3 years agovirtio-gpu: add virtio-gpu-pci module
Gerd Hoffmann [Fri, 23 Oct 2020 06:46:17 +0000 (08:46 +0200)]
virtio-gpu: add virtio-gpu-pci module

Build virtio-gpu pci devices modular.  Must be a separate module because
not all qemu softmmu variants come with PCI support.

Signed-off-by: Gerd Hoffmann <kraxel@redhat.com>
Message-id: 20201023064618.21409-2-kraxel@redhat.com

3 years agoMerge remote-tracking branch 'remotes/stsquad/tags/pull-testing-and-misc-271020-1...
Peter Maydell [Wed, 28 Oct 2020 20:40:08 +0000 (20:40 +0000)]
Merge remote-tracking branch 'remotes/stsquad/tags/pull-testing-and-misc-271020-1' into staging

Testing and gitdm updates

  - add some more individual contributors
  - include SDL2 in centos images
  - skip checkpatch check when no commits found
  - use random port for gdb reverse debugging
  - make gitlab use it's own mirrors to clone
  - fix detection of make -nqp

# gpg: Signature made Tue 27 Oct 2020 09:55:55 GMT
# gpg:                using RSA key 6685AE99E75167BCAFC8DF35FBD0DB095A9E2A44
# gpg: Good signature from "Alex Bennée (Master Work Key) <alex.bennee@linaro.org>" [full]
# Primary key fingerprint: 6685 AE99 E751 67BC AFC8  DF35 FBD0 DB09 5A9E 2A44

* remotes/stsquad/tags/pull-testing-and-misc-271020-1:
  makefile: handle -n / -k / -q correctly
  gitlab-ci: Clone from GitLab itself
  tests/acceptance: pick a random gdb port for reverse debugging
  scripts: fix error from checkpatch.pl when no commits are found
  gitlab: skip checkpatch.pl checks if no commit delta on branch
  tests/docker/dockerfiles/centos: Use SDL2 instead of SDL1
  contrib/gitdm: Add more individual contributors
  Adding ani's email as an individual contributor

Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
3 years agoMerge remote-tracking branch 'remotes/philmd-gitlab/tags/renesas-20201027' into staging
Peter Maydell [Wed, 28 Oct 2020 16:25:31 +0000 (16:25 +0000)]
Merge remote-tracking branch 'remotes/philmd-gitlab/tags/renesas-20201027' into staging

Renesas patches (SH4 and RX)

- Fix few warnings (Thomas Huth)
- Fix typos (Lichang Zhao, Chetan Pant)

CI jobs results:
. https://cirrus-ci.com/build/6368903343374336
. https://gitlab.com/philmd/qemu/-/pipelines/207919103
. https://travis-ci.org/github/philmd/qemu/builds/739133105

# gpg: Signature made Mon 26 Oct 2020 23:27:39 GMT
# gpg:                using RSA key FAABE75E12917221DCFD6BB2E3E32C2CDEADC0DE
# gpg: Good signature from "Philippe Mathieu-Daudé (F4BUG) <f4bug@amsat.org>" [full]
# Primary key fingerprint: FAAB E75E 1291 7221 DCFD  6BB2 E3E3 2C2C DEAD C0DE

* remotes/philmd-gitlab/tags/renesas-20201027:
  target/rx: Fix Lesser GPL version number
  target/rx: Fix some comment spelling errors
  target/sh4: fix some comment spelling errors
  target/sh4: Update coding style to make checkpatch.pl happy
  hw/timer/sh_timer: Remove superfluous "break" statements
  hw/timer/sh_timer: Silence warnings about missing fallthrough statements
  hw/timer/sh_timer: Coding style clean-up
  elf: Add EM_RX definition

Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
3 years agoMerge remote-tracking branch 'remotes/vivier2/tags/linux-user-for-5.2-pull-request...
Peter Maydell [Wed, 28 Oct 2020 15:08:35 +0000 (15:08 +0000)]
Merge remote-tracking branch 'remotes/vivier2/tags/linux-user-for-5.2-pull-request' into staging

Update syscall numbers to 5.9-rc7
Fixes for prctl(), accept4() and xtensa

# gpg: Signature made Mon 26 Oct 2020 22:02:56 GMT
# gpg:                using RSA key CD2F75DDC8E3A4DC2E4F5173F30C38BD3F2FBE3C
# gpg:                issuer "laurent@vivier.eu"
# gpg: Good signature from "Laurent Vivier <lvivier@redhat.com>" [full]
# gpg:                 aka "Laurent Vivier <laurent@vivier.eu>" [full]
# gpg:                 aka "Laurent Vivier (Red Hat) <lvivier@redhat.com>" [full]
# Primary key fingerprint: CD2F 75DD C8E3 A4DC 2E4F  5173 F30C 38BD 3F2F BE3C

* remotes/vivier2/tags/linux-user-for-5.2-pull-request:
  target/xtensa: enable all coprocessors for linux-user
  linux-user: correct errno returned from accept4() syscall
  linux-user: remove _sysctl
  linux-user: update syscall.tbl to Linux 5.9-rc7
  linux-user: update mips/syscall-args-o32.c.inc to Linux 5.9-rc7
  linux-user: update syscall_nr.h to Linux 5.9-rc7
  linux-user: Support f_flags in statfs64 when available.
  Fix stack smashing when handling PR_GET_PDEATHSIG

Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
3 years agoRevert series: virtiofsd: Announce submounts to the guest
Alex Williamson [Wed, 28 Oct 2020 02:13:28 +0000 (20:13 -0600)]
Revert series: virtiofsd: Announce submounts to the guest

This reverts the following commits due to their basis on a bogus
linux kernel header update:

c93a656f7b65 ("tests/acceptance: Add virtiofs_submounts.py")
45ced7ca2f27 ("tests/acceptance/boot_linux: Accept SSH pubkey")
08dce386e77e ("virtiofsd: Announce sub-mount points")
eba8b096c17c ("virtiofsd: Store every lo_inode's parent_dev")
ede24b6be798 ("virtiofsd: Add fuse_reply_attr_with_flags()")
e2577435d343 ("virtiofsd: Add attr_flags to fuse_entry_param")
2f10415abfc5 ("virtiofsd: Announce FUSE_ATTR_FLAGS")
97d741cc96dd ("linux/fuse.h: Pull in from Linux")

Cc: Max Reitz <mreitz@redhat.com>
Cc: Stefan Hajnoczi <stefanha@redhat.com>
Cc: Dr. David Alan Gilbert <dgilbert@redhat.com>
Signed-off-by: Alex Williamson <alex.williamson@redhat.com>
Reviewed-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
Reviewed-by: Max Reitz <mreitz@redhat.com>
Message-id: 160385090886.20017.13382256442750027666.stgit@gimli.home
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
3 years agohw/pci-host/sabre: Simplify code initializing variable once
Philippe Mathieu-Daudé [Mon, 12 Oct 2020 17:09:47 +0000 (19:09 +0200)]
hw/pci-host/sabre: Simplify code initializing variable once

We only need to zero-initialize 'val' once.

Signed-off-by: Philippe Mathieu-Daudé <f4bug@amsat.org>
Reviewed-by: Mark Cave-Ayland <mark.cave-ayland@ilande.co.uk>
Message-Id: <20201012170950.3491912-4-f4bug@amsat.org>
Signed-off-by: Mark Cave-Ayland <mark.cave-ayland@ilande.co.uk>
3 years agohw/pci-host/sabre: Remove superfluous address range check
Philippe Mathieu-Daudé [Mon, 12 Oct 2020 17:09:46 +0000 (19:09 +0200)]
hw/pci-host/sabre: Remove superfluous address range check

The region is registered as 64KiB in sabre_init():

    memory_region_init_io(&s->sabre_config, OBJECT(s), &sabre_config_ops, s,
                          "sabre-config", 0x10000);

Remove the superfluous check.

Signed-off-by: Philippe Mathieu-Daudé <f4bug@amsat.org>
Reviewed-by: Mark Cave-Ayland <mark.cave-ayland@ilande.co.uk>
Message-Id: <20201012170950.3491912-3-f4bug@amsat.org>
Signed-off-by: Mark Cave-Ayland <mark.cave-ayland@ilande.co.uk>
3 years agohw/pci-host/sabre: Update documentation link
Philippe Mathieu-Daudé [Mon, 12 Oct 2020 17:09:45 +0000 (19:09 +0200)]
hw/pci-host/sabre: Update documentation link

The current link redirects to https://www.oracle.com/sun/
announcing "Oracle acquired Sun Microsystems in 2010, ..."
but does not give hint where to find the datasheet.

Use the archived PDF on the Wayback Machine, which works.

Signed-off-by: Philippe Mathieu-Daudé <f4bug@amsat.org>
Reviewed-by: Mark Cave-Ayland <mark.cave-ayland@ilande.co.uk>
Message-Id: <20201012170950.3491912-2-f4bug@amsat.org>
Signed-off-by: Mark Cave-Ayland <mark.cave-ayland@ilande.co.uk>
3 years agosabre: increase number of PCI bus IRQs from 32 to 64
Mark Cave-Ayland [Sun, 11 Oct 2020 08:13:47 +0000 (09:13 +0100)]
sabre: increase number of PCI bus IRQs from 32 to 64

The rework of the sabre IRQs in commit 6864fa3897 "sun4u: update PCI topology to
include simba PCI bridges" changed the IRQ routing so that both PCI and legacy
OBIO IRQs are routed through the sabre PCI host bridge to the CPU.

Unfortunately this commit failed to increase the number of PCI bus IRQs
accordingly meaning that access to the legacy IRQs OBIO (irqnum >= 0x20) would
overflow the PCI bus IRQ array causing strange failures running qemu-system-sparc64
in NetBSD.

Cc: qemu-stable@nongnu.org
Reported-by: Harold Gutch <logix@foobar.franken.de>
Fixes: https://bugs.launchpad.net/qemu/+bug/1838658
Fixes: 6864fa3897 ("sun4u: update PCI topology to include simba PCI bridges")
Signed-off-by: Mark Cave-Ayland <mark.cave-ayland@ilande.co.uk>
Reviewed-by: Philippe Mathieu-Daudé <f4bug@amsat.org>
Message-Id: <20201011081347.2146-1-mark.cave-ayland@ilande.co.uk>
Signed-off-by: Mark Cave-Ayland <mark.cave-ayland@ilande.co.uk>
3 years agohw/display/tcx: Allow 64-bit accesses to framebuffer stippler and blitter
Philippe Mathieu-Daudé [Sat, 24 Oct 2020 20:51:00 +0000 (20:51 +0000)]
hw/display/tcx: Allow 64-bit accesses to framebuffer stippler and blitter

The S24/TCX datasheet is listed as "Unable to locate" on [1].

However the NetBSD revision 1.32 of the driver introduced
64-bit accesses to the stippler and blitter [2]. It is safe
to assume these memory regions are 64-bit accessible.
QEMU implementation is 32-bit, so fill the 'impl' fields.

Michael Lorenz (author of the NetBSD code [2]) provided us with more
information in [3]:

> IIRC the real hardware *requires* 64bit accesses for stipple and
> blitter operations to work. For stipples you write a 64bit word into
> STIP space, the address defines where in the framebuffer you want to
> draw, the data contain a 32bit bitmask, foreground colour and a ROP.
> BLIT space works similarly, the 64bit word contains an offset were to
> read pixels from, and how many you want to copy.
>
> One more thing since there seems to be some confusion - 64bit accesses
> on the framebuffer are fine as well. TCX/S24 is *not* an SBus device,
> even though its node says it is.
> S24 is a card that plugs into a special slot on the SS5 mainboard,
> which is shared with an SBus slot and looks a lot like a horizontal
> UPA slot. Both S24 and TCX are accessed through the Micro/TurboSPARC's
> AFX bus which is 64bit wide and intended for graphics.
> Early FFB docs even mentioned connecting to both AFX and UPA,
> no idea if that was ever realized in hardware though.

[1] http://web.archive.org/web/20111209011516/http://wikis.sun.com/display/FOSSdocs/Home
[2] http://cvsweb.netbsd.org/bsdweb.cgi/src/sys/dev/sbus/tcx.c.diff?r1=1.31&r2=1.32
[3] https://www.mail-archive.com/qemu-devel@nongnu.org/msg734928.html

Cc: qemu-stable@nongnu.org
Reported-by: Andreas Gustafsson <gson@gson.org>
Buglink: https://bugs.launchpad.net/bugs/1892540
Fixes: 55d7bfe2293 ("tcx: Implement hardware acceleration")
Tested-by: Michael S. Tsirkin <mst@redhat.com>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Tested-by: Andreas Gustafsson <gson@gson.org>
Signed-off-by: Philippe Mathieu-Daudé <f4bug@amsat.org>
Message-Id: <20201024205100.3623006-1-f4bug@amsat.org>
Signed-off-by: Mark Cave-Ayland <mark.cave-ayland@ilande.co.uk>
3 years agosabre: don't call sysbus_mmio_map() in sabre_realize()
Mark Cave-Ayland [Sat, 26 Sep 2020 14:02:16 +0000 (15:02 +0100)]
sabre: don't call sysbus_mmio_map() in sabre_realize()

The device should not map itself but instead should be mapped to sysbus by the
sun4u machine.

Signed-off-by: Mark Cave-Ayland <mark.cave-ayland@ilande.co.uk>
Reviewed-by: Philippe Mathieu-Daudé <f4bug@amsat.org>
Message-Id: <20200926140216.7368-7-mark.cave-ayland@ilande.co.uk>
Signed-off-by: Mark Cave-Ayland <mark.cave-ayland@ilande.co.uk>
3 years agosparc32-ledma: don't reference nd_table directly within the device
Mark Cave-Ayland [Sat, 26 Sep 2020 14:02:14 +0000 (15:02 +0100)]
sparc32-ledma: don't reference nd_table directly within the device

Instead use qdev_set_nic_properties() to configure the on-board NIC at the
sun4m machine level.

Signed-off-by: Mark Cave-Ayland <mark.cave-ayland@ilande.co.uk>
Reviewed-by: Philippe Mathieu-Daudé <f4bug@amsat.org>
Message-Id: <20200926140216.7368-5-mark.cave-ayland@ilande.co.uk>
Signed-off-by: Mark Cave-Ayland <mark.cave-ayland@ilande.co.uk>
3 years agosparc32-espdma: use object_initialize_child() for esp child object
Mark Cave-Ayland [Sat, 26 Sep 2020 14:02:13 +0000 (15:02 +0100)]
sparc32-espdma: use object_initialize_child() for esp child object

Store the child object directly within the sparc32-espdma object rather than
using link properties.

Signed-off-by: Mark Cave-Ayland <mark.cave-ayland@ilande.co.uk>
Reviewed-by: Philippe Mathieu-Daudé <f4bug@amsat.org>
Message-Id: <20200926140216.7368-4-mark.cave-ayland@ilande.co.uk>
Signed-off-by: Mark Cave-Ayland <mark.cave-ayland@ilande.co.uk>
3 years agosparc32-ledma: use object_initialize_child() for lance child object
Mark Cave-Ayland [Sat, 26 Sep 2020 14:02:12 +0000 (15:02 +0100)]
sparc32-ledma: use object_initialize_child() for lance child object

Store the child object directly within the sparc32-ledma object rather than
using link properties.

Signed-off-by: Mark Cave-Ayland <mark.cave-ayland@ilande.co.uk>
Reviewed-by: Philippe Mathieu-Daudé <f4bug@amsat.org>
Message-Id: <20200926140216.7368-3-mark.cave-ayland@ilande.co.uk>
Signed-off-by: Mark Cave-Ayland <mark.cave-ayland@ilande.co.uk>
3 years agosparc32-dma: use object_initialize_child() for espdma and ledma child objects
Mark Cave-Ayland [Sat, 26 Sep 2020 14:02:11 +0000 (15:02 +0100)]
sparc32-dma: use object_initialize_child() for espdma and ledma child objects

Store the child objects directly within the sparc32-dma object rather than using
link properties.

Signed-off-by: Mark Cave-Ayland <mark.cave-ayland@ilande.co.uk>
Reviewed-by: Philippe Mathieu-Daudé <f4bug@amsat.org>
Message-Id: <20200926140216.7368-2-mark.cave-ayland@ilande.co.uk>
Signed-off-by: Mark Cave-Ayland <mark.cave-ayland@ilande.co.uk>
3 years agoMerge remote-tracking branch 'remotes/philmd-gitlab/tags/led-api-20201026' into staging
Peter Maydell [Tue, 27 Oct 2020 23:43:53 +0000 (23:43 +0000)]
Merge remote-tracking branch 'remotes/philmd-gitlab/tags/led-api-20201026' into staging

API to model LED.

CI jobs results:
. https://cirrus-ci.com/build/4879251751043072
. https://gitlab.com/philmd/qemu/-/pipelines/207661784
. https://travis-ci.org/github/philmd/qemu/builds/738958191
. https://app.shippable.com/github/philmd/qemu/runs/891/summary/console

# gpg: Signature made Mon 26 Oct 2020 22:03:59 GMT
# gpg:                using RSA key FAABE75E12917221DCFD6BB2E3E32C2CDEADC0DE
# gpg: Good signature from "Philippe Mathieu-Daudé (F4BUG) <f4bug@amsat.org>" [full]
# Primary key fingerprint: FAAB E75E 1291 7221 DCFD  6BB2 E3E3 2C2C DEAD C0DE

* remotes/philmd-gitlab/tags/led-api-20201026:
  hw/arm/tosa: Replace fprintf() calls by LED devices
  hw/misc/mps2-scc: Use the LED device
  hw/misc/mps2-fpgaio: Use the LED device
  hw/arm/aspeed: Add the 3 front LEDs drived by the PCA9552 #1
  hw/misc/led: Emit a trace event when LED intensity has changed
  hw/misc/led: Allow connecting from GPIO output
  hw/misc/led: Add a LED device

Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
3 years agoMerge remote-tracking branch 'remotes/philmd-gitlab/tags/acceptance-testing-20201026...
Peter Maydell [Tue, 27 Oct 2020 16:58:39 +0000 (16:58 +0000)]
Merge remote-tracking branch 'remotes/philmd-gitlab/tags/acceptance-testing-20201026' into staging

Acceptance testing patches

- More ARM tests
- Documentation update

CI jobs results:
. https://cirrus-ci.com/build/5754555684093952
. https://gitlab.com/philmd/qemu/-/pipelines/207807648

# gpg: Signature made Mon 26 Oct 2020 19:06:52 GMT
# gpg:                using RSA key FAABE75E12917221DCFD6BB2E3E32C2CDEADC0DE
# gpg: Good signature from "Philippe Mathieu-Daudé (F4BUG) <f4bug@amsat.org>" [full]
# Primary key fingerprint: FAAB E75E 1291 7221 DCFD  6BB2 E3E3 2C2C DEAD C0DE

* remotes/philmd-gitlab/tags/acceptance-testing-20201026:
  docs/devel/testing.rst: Update outdated Avocado URLs
  tests/acceptance: Allow running Orange Pi test using cached artifacts
  tests/acceptance: Add a 'virt_kvm' test using the GICv3
  tests/boot_linux_console: Boot Trusted Firmware-A on the Raspberry Pi 3

Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
3 years agoaccel/tcg: Add CPU_LOG_EXEC tracing for cpu_io_recompile()
Peter Maydell [Tue, 13 Oct 2020 12:26:58 +0000 (13:26 +0100)]
accel/tcg: Add CPU_LOG_EXEC tracing for cpu_io_recompile()

When using -icount, it's useful for the CPU_LOG_EXEC logging
to include information about when cpu_io_recompile() was
called, because it alerts the reader of the log that the
tracing of a previous TB execution may not actually
correspond to an actually executed instruction. For instance
if you're using -icount and also -singlestep then a guest
instruction that makes an IO access appears in two
"Trace" lines, once in a TB that triggers the cpu_io_recompile()
and then again in the TB that actually executes.

(This is a similar reason to why the "Stopped execution of
TB chain before..." logging in cpu_tb_exec() is helpful
when trying to track execution flow in the logs.)

Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
Message-Id: <20201013122658.4620-1-peter.maydell@linaro.org>
Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
3 years agotcg/optimize: Flush data at labels not TCG_OPF_BB_END
Richard Henderson [Fri, 9 Oct 2020 13:23:52 +0000 (08:23 -0500)]
tcg/optimize: Flush data at labels not TCG_OPF_BB_END

We can easily propagate temp values through the entire extended
basic block (in this case, the set of blocks connected by fallthru),
simply by not discarding the register state at the branch.

Reviewed-by: Alex Bennée <alex.bennee@linaro.org>
Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
3 years agotcg: Do not kill globals at conditional branches
Richard Henderson [Thu, 8 Oct 2020 20:21:43 +0000 (15:21 -0500)]
tcg: Do not kill globals at conditional branches

We can easily register allocate the entire extended basic block
(in this case, the set of blocks connected by fallthru), simply
by not discarding the register state at the branch.

This does not help blocks starting with a label, as they are
reached via a taken branch, and that would require saving the
complete register state at the branch.

Reviewed-by: Alex Bennée <alex.bennee@linaro.org>
Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
3 years agoCHANGELOG: remove disused file
John Snow [Mon, 26 Oct 2020 19:51:31 +0000 (15:51 -0400)]
CHANGELOG: remove disused file

There's no reason to keep this here; the versions described are
ancient. Everything here is still mirrored on
https://wiki.qemu.org/ChangeLog/old if anyone is curious; otherwise, use
the git history.

Signed-off-by: John Snow <jsnow@redhat.com>
Reviewed-by: Thomas Huth <thuth@redhat.com>
Reviewed-by: Alex Bennée <alex.bennee@linaro.org>
Message-Id: <20201026195131.13848-1-jsnow@redhat.com>
Signed-off-by: Laurent Vivier <laurent@vivier.eu>
3 years agoqdev: Fix two typos
Maxim Levitsky [Mon, 19 Oct 2020 16:36:59 +0000 (19:36 +0300)]
qdev: Fix two typos

Signed-off-by: Maxim Levitsky <mlevitsk@redhat.com>
Reviewed-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Reviewed-by: Thomas Huth <thuth@redhat.com>
Message-Id: <20201019163702.471239-2-mlevitsk@redhat.com>
Signed-off-by: Laurent Vivier <laurent@vivier.eu>
3 years agoscripts/qmp: delete 'qmp' script
John Snow [Mon, 19 Oct 2020 21:04:30 +0000 (17:04 -0400)]
scripts/qmp: delete 'qmp' script

This script has not seen a patch that was specifically for this script
since it was moved to this location in 2013, and I doubt it is used. It
uses "man qmp" for its help message, which does not exist. It also
presumes there is a manual page for qmp-XXX, for each defined qmp
command XXX. I don't think that's true.

The format it expects arguments in is something like:

block-dirty-bitmap-add --node=foo --name=bar

and has no capacity to support nested JSON arguments, either.

Most developers use either qmp-shell or socat (or pasting JSON directly
into qmp stdio), so this duplication and additional alternate syntax is
not helpful.

Remove it. Leave a breadcrumb script just in case, to be removed next
release cycle.

Signed-off-by: John Snow <jsnow@redhat.com>
Reviewed-by: Eric Blake <eblake@redhat.com>
Reviewed-by: Thomas Huth <thuth@redhat.com>
Message-Id: <20201019210430.1063390-1-jsnow@redhat.com>
Signed-off-by: Laurent Vivier <laurent@vivier.eu>
3 years agocryptodev: Fix Lesser GPL version number
Chetan Pant [Wed, 14 Oct 2020 13:37:22 +0000 (13:37 +0000)]
cryptodev: Fix Lesser GPL version number

There is no "version 2" of the "Lesser" General Public License.
It is either "GPL version 2.0" or "Lesser GPL version 2.1".
This patch replaces all occurrences of "Lesser GPL version 2" with
"Lesser GPL version 2.1" in comment section.

Signed-off-by: Chetan Pant <chetan4windows@gmail.com>
Reviewed-by: Thomas Huth <thuth@redhat.com>
Message-Id: <20201014133722.14041-1-chetan4windows@gmail.com>
Signed-off-by: Laurent Vivier <laurent@vivier.eu>
3 years agoio: Fix Lesser GPL version number
Chetan Pant [Wed, 14 Oct 2020 13:40:33 +0000 (13:40 +0000)]
io: Fix Lesser GPL version number

There is no "version 2" of the "Lesser" General Public License.
It is either "GPL version 2.0" or "Lesser GPL version 2.1".
This patch replaces all occurrences of "Lesser GPL version 2" with
"Lesser GPL version 2.1" in comment section.

Signed-off-by: Chetan Pant <chetan4windows@gmail.com>
Acked-by: Daniel P. Berrangé <berrange@redhat.com>
Message-Id: <20201014134033.14095-1-chetan4windows@gmail.com>
Signed-off-by: Laurent Vivier <laurent@vivier.eu>
3 years agocpus: Drop declaration of cpu_remove()
Greg Kurz [Wed, 14 Oct 2020 13:40:57 +0000 (15:40 +0200)]
cpus: Drop declaration of cpu_remove()

cpu_remove() got superseded by cpu_remove_sync() in commit dbadee4ff4a0,
but its prototype wasn't removed. We could possibly keep the shorter
cpu_remove() naming but it seems better to highligth that this blocks
until the CPU thread is joined.

Fixes: dbadee4ff4a0 ("cpus: join thread when removing a vCPU")
Cc: pbonzini@redhat.com
Signed-off-by: Greg Kurz <groug@kaod.org>
Reviewed-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Message-Id: <160268285707.1107461.15035929822602623985.stgit@bahia.lan>
Signed-off-by: Laurent Vivier <laurent@vivier.eu>
3 years agoMakefile: Add *.[ch].inc files to cscope/ctags/TAGS
Greg Kurz [Tue, 13 Oct 2020 09:18:13 +0000 (11:18 +0200)]
Makefile: Add *.[ch].inc files to cscope/ctags/TAGS

The code base has some C source and header files that don't get indexed
because their name ends with .inc:

$ git ls-files "*.[ch].inc" | wc -l
66

Add them to the list.

Signed-off-by: Greg Kurz <groug@kaod.org>
Message-Id: <160258069310.900922.1495166540282536628.stgit@bahia.lan>
Signed-off-by: Laurent Vivier <laurent@vivier.eu>
3 years agoelf2dmp: Fix memory leak on main() error paths
AlexChen [Wed, 26 Aug 2020 10:15:53 +0000 (18:15 +0800)]
elf2dmp: Fix memory leak on main() error paths

The 'kdgb' is allocating memory in get_kdbg(), but it is not freed
in both fill_header() and fill_context() failed branches, fix it.

Signed-off-by: AlexChen <alex.chen@huawei.com>
Reviewed-by: Li Qiang <liq3ea@gmail.com>
Reviewed-by: Viktor Prutyanov <viktor.prutyanov@phystech.edu>
Reviewed-by: Thomas Huth <thuth@redhat.com>
Message-Id: <5F463659.8080101@huawei.com>
Signed-off-by: Laurent Vivier <laurent@vivier.eu>
3 years agoide: clear SRST after SRST finishes
John Snow [Tue, 20 Oct 2020 20:02:42 +0000 (16:02 -0400)]
ide: clear SRST after SRST finishes

The SRST protocol states that after diagnostics are complete and the
status is posted, we should clear the SRST bit if it should so happen to
be set.

The reset method itself should handle this, but just in case -- make our
intention explicit here.

Signed-off-by: John Snow <jsnow@redhat.com>
Tested-by: Mark Cave-Ayland <mark.cave-ayland@ilande.co.uk>
Message-id: 20201020200242.1497705-4-jsnow@redhat.com
Signed-off-by: John Snow <jsnow@redhat.com>
3 years agoide: perform SRST as early as possible
John Snow [Tue, 20 Oct 2020 20:02:41 +0000 (16:02 -0400)]
ide: perform SRST as early as possible

We don't need to wait for the falling edge. We can set BSY as
soon as possible and begin immediately resetting the drive. Devices
don't appear to need to take any specific action on the falling edge.

Signed-off-by: John Snow <jsnow@redhat.com>
Tested-by: Mark Cave-Ayland <mark.cave-ayland@ilande.co.uk>
Message-id: 20201020200242.1497705-3-jsnow@redhat.com
Signed-off-by: John Snow <jsnow@redhat.com>
3 years agoide: run diagnostic after SRST
John Snow [Tue, 20 Oct 2020 20:02:40 +0000 (16:02 -0400)]
ide: run diagnostic after SRST

Software reset (SRST) should cause the diagnostic command to be run. Make an
explicit call to that routine.

Reported-by: Mark Cave-Ayland <mark.cave-ayland@ilande.co.uk>
Tested-by: Mark Cave-Ayland <mark.cave-ayland@ilande.co.uk>
Signed-off-by: John Snow <jsnow@redhat.com>
Message-id: 20201020200242.1497705-2-jsnow@redhat.com
Fixes: 55adb3c45620c31f29978f209e2a44a08d34e2da
Fixes: https://bugs.launchpad.net/bugs/1900155
Tested-by: Mark Cave-Ayland <mark.cave-ayland@ilande.co.uk>
Signed-off-by: John Snow <jsnow@redhat.com>
3 years agoMerge remote-tracking branch 'remotes/dgilbert-gitlab/tags/pull-virtiofs-20201026...
Peter Maydell [Tue, 27 Oct 2020 14:29:52 +0000 (14:29 +0000)]
Merge remote-tracking branch 'remotes/dgilbert-gitlab/tags/pull-virtiofs-20201026' into staging

virtiofsd pull 2020-10-26

Misono
   Set default log level to info
   Explicit build option for virtiofsd

Me
   xattr name mapping

Stefan
  Alternative chroot sandbox method

Max
  Submount mechanism

Signed-off-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
# gpg: Signature made Mon 26 Oct 2020 18:41:36 GMT
# gpg:                using RSA key 45F5C71B4A0CB7FB977A9FA90516331EBC5BFDE7
# gpg: Good signature from "Dr. David Alan Gilbert (RH2) <dgilbert@redhat.com>" [full]
# Primary key fingerprint: 45F5 C71B 4A0C B7FB 977A  9FA9 0516 331E BC5B FDE7

* remotes/dgilbert-gitlab/tags/pull-virtiofs-20201026:
  tests/acceptance: Add virtiofs_submounts.py
  tests/acceptance/boot_linux: Accept SSH pubkey
  virtiofsd: Announce sub-mount points
  virtiofsd: Store every lo_inode's parent_dev
  virtiofsd: Add fuse_reply_attr_with_flags()
  virtiofsd: Add attr_flags to fuse_entry_param
  virtiofsd: Announce FUSE_ATTR_FLAGS
  linux/fuse.h: Pull in from Linux
  tools/virtiofsd: xattr name mappings: Simple 'map'
  tools/virtiofsd: xattr name mapping examples
  tools/virtiofsd: xattr name mappings: Map server xattr names
  tools/virtiofsd: xattr name mappings: Map client xattr names
  tools/virtiofsd: xattr name mappings: Add option
  virtiofsd: add container-friendly -o sandbox=chroot option
  virtiofsd: passthrough_ll: set FUSE_LOG_INFO as default log_level
  configure: add option for virtiofsd

Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
3 years agoblock: End quiescent sections when a BDS is deleted
Greg Kurz [Fri, 23 Oct 2020 15:01:10 +0000 (17:01 +0200)]
block: End quiescent sections when a BDS is deleted

If a BDS gets deleted during blk_drain_all(), it might miss a
call to bdrv_do_drained_end(). This means missing a call to
aio_enable_external() and the AIO context remains disabled for
ever. This can cause a device to become irresponsive and to
disrupt the guest execution, ie. hang, loop forever or worse.

This scenario is quite easy to encounter with virtio-scsi
on POWER when punching multiple blockdev-create QMP commands
while the guest is booting and it is still running the SLOF
firmware. This happens because SLOF disables/re-enables PCI
devices multiple times via IO/MEM/MASTER bits of PCI_COMMAND
register after the initial probe/feature negotiation, as it
tends to work with a single device at a time at various stages
like probing and running block/network bootloaders without
doing a full reset in-between. This naturally generates many
dataplane stops and starts, and thus many drain sections that
can race with blockdev_create_run(). In the end, SLOF bails
out.

It is somehow reproducible on x86 but it requires to generate
articial dataplane start/stop activity with stop/cont QMP
commands. In this case, seabios ends up looping for ever,
waiting for the virtio-scsi device to send a response to
a command it never received.

Add a helper that pairs all previously called bdrv_do_drained_begin()
with a bdrv_do_drained_end() and call it from bdrv_close().
While at it, update the "/bdrv-drain/graph-change/drain_all"
test in test-bdrv-drain so that it can catch the issue.

BugId: https://bugzilla.redhat.com/show_bug.cgi?id=1874441
Signed-off-by: Greg Kurz <groug@kaod.org>
Message-Id: <160346526998.272601.9045392804399803158.stgit@bahia.lan>
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
3 years agoqcow2: Skip copy-on-write when allocating a zero cluster
Alberto Garcia [Mon, 26 Oct 2020 16:58:27 +0000 (17:58 +0100)]
qcow2: Skip copy-on-write when allocating a zero cluster

Since commit c8bb23cbdbe32f5c326365e0a82e1b0e68cdcd8a when a write
request results in a new allocation QEMU first tries to see if the
rest of the cluster outside the written area contains only zeroes.

In that case, instead of doing a normal copy-on-write operation and
writing explicit zero buffers to disk, the code zeroes the whole
cluster efficiently using pwrite_zeroes() with BDRV_REQ_NO_FALLBACK.

This improves performance very significantly but it only happens when
we are writing to an area that was completely unallocated before. Zero
clusters (QCOW2_CLUSTER_ZERO_*) are treated like normal clusters and
are therefore slower to allocate.

This happens because the code uses bdrv_is_allocated_above() rather
bdrv_block_status_above(). The former is not as accurate for this
purpose but it is faster. However in the case of qcow2 the underlying
call does already report zero clusters just fine so there is no reason
why we cannot use that information.

After testing 4KB writes on an image that only contains zero clusters
this patch results in almost five times more IOPS.

Signed-off-by: Alberto Garcia <berto@igalia.com>
Reviewed-by: Vladimir Sementsov-Ogievskiy <vsementsov@virtuozzo.com>
Message-Id: <6d77cab968c501c44d6e1089b9bc91b04170b49e.1603731354.git.berto@igalia.com>
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
3 years agoqcow2: Report BDRV_BLOCK_ZERO more accurately in bdrv_co_block_status()
Alberto Garcia [Mon, 26 Oct 2020 16:58:26 +0000 (17:58 +0100)]
qcow2: Report BDRV_BLOCK_ZERO more accurately in bdrv_co_block_status()

If a BlockDriverState supports backing files but has none then any
unallocated area reads back as zeroes.

bdrv_co_block_status() is only reporting this is if want_zero is true,
but this is an inexpensive test and there is no reason not to do it in
all cases.

Suggested-by: Vladimir Sementsov-Ogievskiy <vsementsov@virtuozzo.com>
Signed-off-by: Alberto Garcia <berto@igalia.com>
Reviewed-by: Vladimir Sementsov-Ogievskiy <vsementsov@virtuozzo.com>
Message-Id: <66fa0914a0e2b727ab6d1b63ca773d7cd29a9a9e.1603731354.git.berto@igalia.com>
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
3 years agoqemu-img: add support for rate limit in qemu-img convert
Zhengui [Tue, 20 Oct 2020 14:47:44 +0000 (14:47 +0000)]
qemu-img: add support for rate limit in qemu-img convert

add support for rate limit in qemu-img convert.

Signed-off-by: Zhengui <lizhengui@huawei.com>
Message-Id: <1603205264-17424-3-git-send-email-lizhengui@huawei.com>
Reviewed-by: Alberto Garcia <berto@igalia.com>
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
3 years agoqemu-img: add support for rate limit in qemu-img commit
Zhengui [Tue, 20 Oct 2020 14:47:43 +0000 (14:47 +0000)]
qemu-img: add support for rate limit in qemu-img commit

add support for rate limit in qemu-img commit.

Signed-off-by: Zhengui <lizhengui@huawei.com>
Message-Id: <1603205264-17424-2-git-send-email-lizhengui@huawei.com>
Reviewed-by: Alberto Garcia <berto@igalia.com>
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
3 years agoppc/: fix some comment spelling errors
zhaolichang [Fri, 9 Oct 2020 06:44:37 +0000 (14:44 +0800)]
ppc/: fix some comment spelling errors

I found that there are many spelling errors in the comments of qemu/target/ppc.
I used spellcheck to check the spelling errors and found some errors in the folder.

Signed-off-by: zhaolichang <zhaolichang@huawei.com>
Reviewed-by: David Edmondson <david.edmondson@oracle.com>
Message-Id: <20201009064449.2336-3-zhaolichang@huawei.com>
Reviewed-by: Thomas Huth <thuth@redhat.com>
Reviewed-by: Greg Kurz <groug@kaod.org>
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
3 years agospapr: Improve spapr_reallocate_hpt() error reporting
Greg Kurz [Mon, 26 Oct 2020 12:40:54 +0000 (13:40 +0100)]
spapr: Improve spapr_reallocate_hpt() error reporting

spapr_reallocate_hpt() has three users, two of which pass &error_fatal
and the third one, htab_load(), passes &local_err, uses it to detect
failures and simply propagates -EINVAL up to vmstate_load(), which will
cause QEMU to exit. It is thus confusing that spapr_reallocate_hpt()
doesn't return right away when an error is detected in some cases. Also,
the comment suggesting that the caller is welcome to try to carry on
seems like a remnant in this respect.

This can be improved:
- change spapr_reallocate_hpt() to always report a negative errno on
  failure, either as reported by KVM or -ENOSPC if the HPT is smaller
  than what was asked,
- use that to detect failures in htab_load() which is preferred over
  checking &local_err,
- propagate this negative errno to vmstate_load() because it is more
  accurate than propagating -EINVAL for all possible errors.

[dwg: Fix compile error due to omitted prelim patch]
Signed-off-by: Greg Kurz <groug@kaod.org>
Message-Id: <160371605460.305923.5890143959901241157.stgit@bahia.lan>
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
3 years agotarget/ppc: Fix kvmppc_load_htab_chunk() error reporting
Greg Kurz [Mon, 26 Oct 2020 12:40:47 +0000 (13:40 +0100)]
target/ppc: Fix kvmppc_load_htab_chunk() error reporting

If kvmppc_load_htab_chunk() fails, its return value is propagated up
to vmstate_load(). It should thus be a negative errno, not -1 (which
maps to EPERM and would lure the user into thinking that the problem
is necessarily related to a lack of privilege).

Return the error reported by KVM or ENOSPC in case of short write.
While here, propagate the error message through an @errp argument
and have the caller to print it with error_report_err() instead
of relying on fprintf().

Signed-off-by: Greg Kurz <groug@kaod.org>
Message-Id: <160371604713.305923.5264900354159029580.stgit@bahia.lan>
Reviewed-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
3 years agospapr: Use error_append_hint() in spapr_reallocate_hpt()
Greg Kurz [Mon, 26 Oct 2020 12:40:40 +0000 (13:40 +0100)]
spapr: Use error_append_hint() in spapr_reallocate_hpt()

Hints should be added with the dedicated error_append_hint() API
because we don't want to print them when using QMP. This requires
to insert ERRP_GUARD as explained in "qapi/error.h".

Signed-off-by: Greg Kurz <groug@kaod.org>
Message-Id: <160371604030.305923.17464161378167312662.stgit@bahia.lan>
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
3 years agospapr: Simplify error handling in spapr_memory_plug()
Greg Kurz [Mon, 19 Oct 2020 08:49:01 +0000 (10:49 +0200)]
spapr: Simplify error handling in spapr_memory_plug()

As recommended in "qapi/error.h", add a bool return value to
spapr_add_lmbs() and spapr_add_nvdimm(), and use them instead
of local_err in spapr_memory_plug().

This allows to get rid of the error propagation overhead.

Signed-off-by: Greg Kurz <groug@kaod.org>
Message-Id: <160309734178.2739814.3488437759887793902.stgit@bahia.lan>
Reviewed-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
3 years agospapr: Pass &error_abort when getting some PC DIMM properties
Greg Kurz [Mon, 19 Oct 2020 08:48:41 +0000 (10:48 +0200)]
spapr: Pass &error_abort when getting some PC DIMM properties

Both PC_DIMM_SLOT_PROP and PC_DIMM_ADDR_PROP are defined in the
default property list of the PC DIMM device class:

    DEFINE_PROP_UINT64(PC_DIMM_ADDR_PROP, PCDIMMDevice, addr, 0),

    DEFINE_PROP_INT32(PC_DIMM_SLOT_PROP, PCDIMMDevice, slot,
                      PC_DIMM_UNASSIGNED_SLOT),

They should thus be always gettable for both PC DIMMs and NVDIMMs.
An error in getting them can only be the result of a programming
error. It doesn't make much sense to propagate the error in this
case. Abort instead.

Signed-off-by: Greg Kurz <groug@kaod.org>
Message-Id: <160309732180.2739814.7243774674998010907.stgit@bahia.lan>
Reviewed-by: Igor Mammedov <imammedo@redhat.com>
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
3 years agospapr: Use appropriate getter for PC_DIMM_SLOT_PROP
Greg Kurz [Mon, 19 Oct 2020 08:48:27 +0000 (10:48 +0200)]
spapr: Use appropriate getter for PC_DIMM_SLOT_PROP

The PC_DIMM_SLOT_PROP property is defined as:

    DEFINE_PROP_INT32(PC_DIMM_SLOT_PROP, PCDIMMDevice, slot,
                      PC_DIMM_UNASSIGNED_SLOT),

Use object_property_get_int() instead of object_property_get_uint().
Since spapr_memory_plug() only gets called if pc_dimm_pre_plug()
succeeded, we expect to have a valid >= 0 slot number, either because
the user passed a valid slot number or because pc_dimm_get_free_slot()
picked one up for us.

Signed-off-by: Greg Kurz <groug@kaod.org>
Message-Id: <160309730758.2739814.15821922745424652642.stgit@bahia.lan>
Reviewed-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
3 years agospapr: Use appropriate getter for PC_DIMM_ADDR_PROP
Greg Kurz [Mon, 19 Oct 2020 08:48:16 +0000 (10:48 +0200)]
spapr: Use appropriate getter for PC_DIMM_ADDR_PROP

The PC_DIMM_ADDR_PROP property is defined as:

    DEFINE_PROP_UINT64(PC_DIMM_ADDR_PROP, PCDIMMDevice, addr, 0),

Use object_property_get_uint() instead of object_property_get_int().

Signed-off-by: Greg Kurz <groug@kaod.org>
Message-Id: <160309729609.2739814.4996614957953215591.stgit@bahia.lan>
Reviewed-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
3 years agopc-dimm: Drop @errp argument of pc_dimm_plug()
Greg Kurz [Mon, 19 Oct 2020 08:48:04 +0000 (10:48 +0200)]
pc-dimm: Drop @errp argument of pc_dimm_plug()

pc_dimm_plug() doesn't use it. It only aborts on error.

Drop @errp and adapt the callers accordingly.

[dwg: Removed unused label to fix compile]
Signed-off-by: Greg Kurz <groug@kaod.org>
Message-Id: <160309728447.2739814.12831204841251148202.stgit@bahia.lan>
Reviewed-by: Vladimir Sementsov-Ogievskiy <vsementsov@virtuozzo.com>
Reviewed-by: Igor Mammedov <imammedo@redhat.com>
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
3 years agospapr: Simplify spapr_cpu_core_realize() and spapr_cpu_core_unrealize()
Greg Kurz [Thu, 15 Oct 2020 21:18:53 +0000 (23:18 +0200)]
spapr: Simplify spapr_cpu_core_realize() and spapr_cpu_core_unrealize()

Now that the error path of spapr_cpu_core_realize() is just to call
idempotent spapr_cpu_core_unrealize() for rollback, no need to create
and realize the vCPUs in two separate loops.

Merge them and do them same in spapr_cpu_core_unrealize() for symmetry.

Signed-off-by: Greg Kurz <groug@kaod.org>
Message-Id: <160279673321.1808373.2248221100790367912.stgit@bahia.lan>
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
3 years agospapr: Make spapr_cpu_core_unrealize() idempotent
Greg Kurz [Thu, 15 Oct 2020 21:18:46 +0000 (23:18 +0200)]
spapr: Make spapr_cpu_core_unrealize() idempotent

spapr_cpu_core_realize() has a rollback path which partially duplicates
the code of spapr_cpu_core_unrealize().

Let's make spapr_cpu_core_unrealize() idempotent and call it instead. This
requires to:
- move the registration and unregistration of the reset handler around
  but it is harmless,
- allocate the array of vCPUs with g_new0() to be able to filter out
  unused slots,
- make sure to only unrealize vCPUs that have been already realized.

Signed-off-by: Greg Kurz <groug@kaod.org>
Message-Id: <160279672626.1808373.14142129300586424514.stgit@bahia.lan>
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
3 years agospapr: Drop spapr_delete_vcpu() unused argument
Greg Kurz [Thu, 15 Oct 2020 21:18:39 +0000 (23:18 +0200)]
spapr: Drop spapr_delete_vcpu() unused argument

The 'sc' argument is unused. Drop it.

Signed-off-by: Greg Kurz <groug@kaod.org>
Message-Id: <160279671929.1808373.10333672533575251075.stgit@bahia.lan>
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
3 years agospapr: Unrealize vCPUs with qdev_unrealize()
Greg Kurz [Thu, 15 Oct 2020 21:18:32 +0000 (23:18 +0200)]
spapr: Unrealize vCPUs with qdev_unrealize()

Since we introduced CPU hot-unplug in sPAPR, we don't unrealize the
vCPU objects explicitly. Instead, we let QOM handle that for us under
object_property_del_all() when the CPU core object is finalized. The
only thing we do is calling cpu_remove_sync() to tear the vCPU thread
down.

This happens to work but it is ugly because:
- we call qdev_realize() but the corresponding qdev_unrealize() is
  buried deep in the QOM code
- we call cpu_remove_sync() to undo qemu_init_vcpu() called by
  ppc_cpu_realize() in target/ppc/translate_init.c.inc
- the CPU init and teardown paths aren't really symmetrical

The latter didn't bite us so far but a future patch that greatly
simplifies the CPU core realize path needs it to avoid a crash
in QOM.

For all these reasons, have ppc_cpu_unrealize() to undo the changes
of ppc_cpu_realize() by calling cpu_remove_sync() at the right place,
and have the sPAPR CPU core code to call qdev_unrealize().

This requires to add a missing stub because translate_init.c.inc is
also compiled for user mode.

Signed-off-by: Greg Kurz <groug@kaod.org>
Message-Id: <160279671236.1808373.14732005038172874990.stgit@bahia.lan>
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>