Add helper function to generate dsdt aml code for the gpex pci host.
Largely copied from arm/virt. Configuration is handled by passing
a config struct instead of looked up from memory map.
* remotes/kraxel/tags/vga-20200929-pull-request:
ppc/pseries: enable big-endian-framebuffer quirk for bochs-display and virtio-vga
virtio-vga: implement big-endian-framebuffer property
virtio-gpu: set physical dimensions for EDID
spice: get monitors physical dimension
spice: remove the single monitor config logic
ui: add getter for UIInfo
edid: use physical dimensions if available
edid: fix physical display size computation
virtio-gpu-3d: fix abnormal display after a warm reboot
Peter Maydell [Tue, 29 Sep 2020 10:10:29 +0000 (11:10 +0100)]
Merge remote-tracking branch 'remotes/mst/tags/for_upstream' into staging
virtio,pc,acpi: fixes, tests
Fixes and tests all over the place.
Batch iommu updates for vdpa.
Removal of deprecated cpu hotplug commands.
SMBIOS OEM string support.
Signed-off-by: Michael S. Tsirkin <[email protected]>
# gpg: Signature made Tue 29 Sep 2020 08:09:21 BST
# gpg: using RSA key 5D09FD0871C8F85B94CA8A0D281F0DB8D28D5469
# gpg: issuer "[email protected]"
# gpg: Good signature from "Michael S. Tsirkin <[email protected]>" [full]
# gpg: aka "Michael S. Tsirkin <[email protected]>" [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: (48 commits)
libvhost-user: return on error in vu_log_queue_fill()
libvhost-user: return early on virtqueue errors
hw: virtio-pmem: detach the element fromt the virtqueue when error occurs
tests/acpi: update golden master DSDT binary table blobs for q35
piix4: don't reserve hw resources when hotplug is off globally
Add ACPI DSDT tables for q35 that are being updated by the next patch
tests/acpi: add newly added acpi DSDT table blob for pci bridge hotplug flag
tests/acpi: unit test for 'acpi-pci-hotplug-with-bridge-support' bridge flag
tests/acpi: list added acpi table binary file for pci bridge hotplug test
i440fx/acpi: do not add hotplug related amls for cold plugged bridges
Fix a gap where acpi_pcihp_find_hotplug_bus() returns a non-hotpluggable bus
tests/acpi: add a new ACPI table in order to test root pci hotplug on/off
tests/acpi: add new unit test to test hotplug off/on feature on the root pci bus
tests/acpi: mark addition of table DSDT.roothp for unit testing root pci hotplug
vhost-user: save features of multiqueues if chardev is closed
qemu-options: document SMBIOS type 11 settings
hw/smbios: report error if table size is too large
hw/smbios: support loading OEM strings values from a file
tests: acpi: update acpi blobs with new AML
x68: acpi: trigger SMI before sending hotplug Notify event to OSPM
...
ppc/pseries: enable big-endian-framebuffer quirk for bochs-display and virtio-vga
Already done for stdvga and secondary-vga, bochs-display and virtio-vga
support the big-endian-framebuffer property too. Fixes blue console
background at boot (offb firmware console).
Introduced in commit 9c956e646178fee8c14ce7dfae5a9d7cb901876c ("spice:
prepare for upcoming spice-server change"), the new logic never
materialized in the spice server source tree. Let's remove it for now,
until it actually changes in Spice.
Li Qiang [Thu, 13 Aug 2020 16:51:25 +0000 (09:51 -0700)]
hw: virtio-pmem: detach the element fromt the virtqueue when error occurs
If error occurs while processing the virtio request we should call
'virtqueue_detach_element' to detach the element from the virtqueue
before free the elem.
Ani Sinha [Fri, 18 Sep 2020 08:41:09 +0000 (14:11 +0530)]
tests/acpi: update golden master DSDT binary table blobs for q35
In the previously applied commit ("piix4: don't reserve hw resources when
hotplug is off globally"), we make changes to the ACPI DSDT tables
such that some ACPI code are not generated when bsel is absent. Since
as of this point in time, in q35 machines, we do not use bsel for pci
buses, we need to update the DSDT table blobs.
This patch updates the DSDT golden master tables for q35 machines.
At the same time, we clear bios-tables-test-allowed-diff.h for future
changes which update tables.
Following is a typical diff between the q35 acpi DSDT table blobs:
Ani Sinha [Fri, 18 Sep 2020 08:41:08 +0000 (14:11 +0530)]
piix4: don't reserve hw resources when hotplug is off globally
When acpi hotplug is turned off for both root pci bus as well as for pci
bridges, we should not generate the related ACPI code for DSDT table or
initialize related hw ports or reserve hw resources. This change makes
sure all those operations are turned off in the case ACPI pci hotplug is
off globally.
In this change, we also make sure ACPI code for the PCNT method are only
added when bsel is enabled for the corresponding pci bus or bridge hotplug
is turned on.
As q35 machines do not use bsel for it's pci buses at this point in time, this
change affects DSDT acpi table for q35 machines as well. Therefore, we will
also need to commit the updated golden master DSDT table acpi binary blobs as
well. Following is the list of blobs which needs updating:
Ani Sinha [Fri, 18 Sep 2020 08:41:07 +0000 (14:11 +0530)]
Add ACPI DSDT tables for q35 that are being updated by the next patch
The following patch ("piix4: don't reserve hw resources when
hotplug is off globally") modifies certain ACPI tables for q35 machines.
This patch adds those table names to tests/qtest/bios-tables-test-allowed-diff.h
so that unit tests continue to pass and bisection is not broken.
Ani Sinha [Fri, 18 Sep 2020 08:41:06 +0000 (14:11 +0530)]
tests/acpi: add newly added acpi DSDT table blob for pci bridge hotplug flag
This patch adds a binary blob corresponding to the DSDT acpi table. It is used
to unit test the flag 'acpi-pci-hotplug-with-bridge-support' used with pci
bridges.
This change also clears the file tests/qtest/bios-tables-test-allowed-diff.h
so that future changes which affect the table can be caught.
The following is the diff between files tests/data/acpi/pc/DSDT.bridge and
tests/data/acpi/pc/DSDT.hpbridge after disassembly :
Ani Sinha [Fri, 18 Sep 2020 08:41:05 +0000 (14:11 +0530)]
tests/acpi: unit test for 'acpi-pci-hotplug-with-bridge-support' bridge flag
This change adds a new unit test for the global flag
'acpi-pci-hotplug-with-bridge-support' which is available for cold plugged pci
bridges in i440fx. The flag can be used to turn off ACPI based hotplug support
on all pci bridges.
Ani Sinha [Fri, 18 Sep 2020 08:41:04 +0000 (14:11 +0530)]
tests/acpi: list added acpi table binary file for pci bridge hotplug test
The file 'tests/data/acpi/pc/DSDT.hpbridge' is a newly added acpi table file
for testing the pci bridge option 'acpi-pci-hotplug-with-bridge-support' under
i440fx. This change documents this fact.
Ani Sinha [Fri, 18 Sep 2020 08:41:03 +0000 (14:11 +0530)]
i440fx/acpi: do not add hotplug related amls for cold plugged bridges
Cold plugged bridges are not hot unpluggable, even when their hotplug
property (acpi-pci-hotplug-with-bridge-support) is turned off. Please see
the function acpi_pcihp_pc_no_hotplug(). However, with
the current implementaton, Windows would try to hot-unplug a pci bridge when
it's hotplug switch is off. This is regardless of whether there are devices
attached to the bridge. This is because we add ACPI code like _EJ0 etc for the
pci slot where the bridge is cold plugged.
In this fix, we identify a cold plugged bridge and for cold plugged bridges,
we do not add the appropriate ACPI methods that are used by the OS
to identify a hot-pluggable/unpluggable pci device. After this change, Windows
does not detect the cold plugged pci bridge as ejectable.
As a result of the patch, the following are the changes to the DSDT ACPI
table:
Ani Sinha [Fri, 18 Sep 2020 08:41:02 +0000 (14:11 +0530)]
Fix a gap where acpi_pcihp_find_hotplug_bus() returns a non-hotpluggable bus
When ACPI hotplug for the root bus is disabled, the bsel property for that
bus is not set. Please see the following commit:
3d7e78aa7777f ("Introduce a new flag for i440fx to disable PCI hotplug on the
root bus").
As a result, when acpi_pcihp_find_hotplug_bus() is called
with bsel set to 0, it may return the root bus. This can cause devices
attached to the root bus to get hot-unplugged if the user issues the following
set of commmands:
outl 0xae10 0
outl 0xae08 your_slot
Thanks to Julia for pointing this out here:
https://www.mail-archive.com/[email protected]/msg734548.html
In this patch, we fix the issue in this function by checking if the bus which
is returned by the function is actually hotpluggable. If not, we simply return
NULL. This avoids the scenario where we were returning a non-hotpluggable bus.
Ani Sinha [Fri, 18 Sep 2020 08:41:01 +0000 (14:11 +0530)]
tests/acpi: add a new ACPI table in order to test root pci hotplug on/off
A new binary ACPI table tests/data/acpi/pc/DSDT.roothp is added in order
to unit test hotplug on/off capability on the root pci bus for i440fx.
The diff between the table DSDT.bridge and DSDT.roothp is listed below:
Ani Sinha [Fri, 18 Sep 2020 08:41:00 +0000 (14:11 +0530)]
tests/acpi: add new unit test to test hotplug off/on feature on the root pci bus
Ability to turn hotplug off on the pci root bus for i440fx was added in commit: 3d7e78aa7777f0 ("Introduce a new flag for i440fx to disable PCI hotplug on the
root bus")
This change adds a unit test in order to test this feature.
Ani Sinha [Fri, 18 Sep 2020 08:40:59 +0000 (14:10 +0530)]
tests/acpi: mark addition of table DSDT.roothp for unit testing root pci hotplug
A new binary acpi table tests/data/acpi/pc/DSDT.roothp is added in order to
unit test the feature flag that can disable/enable root pci bus hotplug on
i440fx. This feature was added with the commit: 3d7e78aa7777f0 ("Introduce a new flag for i440fx to disable PCI hotplug on
the root bus")
This change documents the fact that this new file addition was made as a part
of the unit test change.
vhost-user: save features of multiqueues if chardev is closed
Fore-commit(c6beefd674) only saves features of queue0,
this makes wrong features of other queues in multiqueues
situation.
For examples:
qemu-system-aarch64 ... \
-chardev socket,id=charnet0,path=/var/run/vhost_sock \
-netdev vhost-user,chardev=charnet0,queues=2,id=hostnet0 \
...
There are two queues in nic assocated with one chardev.
When chardev is reconnected, it is necessary to save and
restore features of all queues.
hw/smbios: report error if table size is too large
The SMBIOS 2.1 entry point uses a uint16 data type for reporting the
total length of the tables. If the user passes -smbios configuration to
QEMU that causes the table size to exceed this limit then various bad
behaviours result, including
- firmware hangs in an infinite loop
- firmware triggers a KVM crash on bad memory access
- firmware silently discards user's SMBIOS data replacing it with
a generic data set.
Limiting the size to 0xffff in QEMU avoids triggering most of these
problems. There is a remaining bug in SeaBIOS which tries to prepend its
own data for table 0, and does not check whether there is sufficient
space before attempting this.
hw/smbios: support loading OEM strings values from a file
Some applications want to pass quite large values for the OEM strings
entries. Rather than having huge strings on the command line, it would
be better to load them from a file, as supported with -fw_cfg.
This introduces the "path" parameter allowing for:
Igor Mammedov [Wed, 23 Sep 2020 09:46:49 +0000 (05:46 -0400)]
tests: acpi: update acpi blobs with new AML
here is diff against tests/data/acpi/q35/DSDT
for currently shipped ovmf binary.
(once firmware blob is updated, it will negotiate CPU hotplug
feature which will ad extra hunk sending SMI and Q35 tests will
need to be updated), but otherwise diff shows new CPU hotplug
AML that is shared between q35 and pc machines.
Method (CSCN, 0, Serialized)
{
Acquire (\_SB.PCI0.PRES.CPLK, 0xFFFF)
- Local0 = One
- While ((Local0 == One))
- {
- Local0 = Zero
- \_SB.PCI0.PRES.CCMD = Zero
- If ((\_SB.PCI0.PRES.CINS == One))
+ Name (CNEW, Package (0xFF){})
+ Local3 = Zero
+ Local4 = One
+ While ((Local4 == One))
+ {
+ Local4 = Zero
+ Local0 = One
+ Local1 = Zero
+ While (((Local0 == One) && (Local3 < One)))
{
- CTFY (\_SB.PCI0.PRES.CDAT, One)
- \_SB.PCI0.PRES.CINS = One
- Local0 = One
+ Local0 = Zero
+ \_SB.PCI0.PRES.CSEL = Local3
+ \_SB.PCI0.PRES.CCMD = Zero
+ If ((\_SB.PCI0.PRES.CDAT < Local3))
+ {
+ Break
+ }
+
+ If ((Local1 == 0xFF))
+ {
+ Local4 = One
+ Break
+ }
+
+ Local3 = \_SB.PCI0.PRES.CDAT
+ If ((\_SB.PCI0.PRES.CINS == One))
+ {
+ CNEW [Local1] = Local3
+ Local1++
+ Local0 = One
+ }
+ ElseIf ((\_SB.PCI0.PRES.CRMV == One))
+ {
+ CTFY (Local3, 0x03)
+ \_SB.PCI0.PRES.CRMV = One
+ Local0 = One
+ }
+
+ Local3++
}
- ElseIf ((\_SB.PCI0.PRES.CRMV == One))
+
+ Local2 = Zero
+ While ((Local2 < Local1))
{
- CTFY (\_SB.PCI0.PRES.CDAT, 0x03)
- \_SB.PCI0.PRES.CRMV = One
- Local0 = One
+ Local3 = DerefOf (CNEW [Local2])
+ CTFY (Local3, One)
+ Debug = Local3
+ \_SB.PCI0.PRES.CSEL = Local3
+ \_SB.PCI0.PRES.CINS = One
+ Local2++
}
}
Igor Mammedov [Wed, 23 Sep 2020 09:46:48 +0000 (05:46 -0400)]
x68: acpi: trigger SMI before sending hotplug Notify event to OSPM
In case firmware has negotiated CPU hotplug SMI feature, generate
AML to describe SMI IO port region and send SMI to firmware
on each CPU hotplug SCI in case new CPUs were hotplugged.
Since new CPUs can be hotplugged while CPU_SCAN_METHOD is running
we can't send SMI before new CPUs are fetched from QEMU as it
could cause sending Notify to a CPU that firmware hasn't seen yet.
So fetch new CPUs into local cache first, then send SMI and
after that send Notify events to cached CPUs. This should ensure
that Notify is sent only to CPUs which were processed by firmware
first.
Any CPUs that were hotplugged after caching will be processed
by the next CPU_SCAN_METHOD, when pending SCI is handled.
Igor Mammedov [Wed, 23 Sep 2020 09:46:47 +0000 (05:46 -0400)]
x86: acpi: introduce the PCI0.SMI0 ACPI device
When CPU hotplug with SMI has been negotiated, describe the SMI
register block in the DSDT. Pass the ACPI name of the SMI control
register to build_cpus_aml(), so that CPU_SCAN_METHOD can access the
register in the next patch.
Igor Mammedov [Wed, 23 Sep 2020 09:46:42 +0000 (05:46 -0400)]
x86: cpuhp: refuse cpu hot-unplug request earlier if not supported
CPU hot-unplug with SMM requires firmware participation to prevent
guest crash (i.e. CPU can be removed only after OS _and_ firmware
were prepared for the action).
Previous patches introduced ICH9_LPC_SMI_F_CPU_HOT_UNPLUG_BIT
feature bit, which is advertised by firmware when it has support
for CPU hot-unplug. Use it to check if guest is able to handle
unplug and make device_del fail gracefully if hot-unplug feature
hasn't been negotiated.
Igor Mammedov [Wed, 23 Sep 2020 09:46:41 +0000 (05:46 -0400)]
x86: cpuhp: prevent guest crash on CPU hotplug when broadcast SMI is in use
There were reports of guest crash on CPU hotplug, when using q35 machine
type and OVMF with SMM, due to hotplugged CPU trying to process SMI at
default SMI handler location without it being relocated by firmware first.
Fix it by refusing hotplug if firmware hasn't negotiated CPU hotplug with
SMI support while SMI broadcast is in use.
Igor Mammedov [Wed, 23 Sep 2020 09:46:40 +0000 (05:46 -0400)]
x86: lpc9: let firmware negotiate 'CPU hotplug with SMI' features
It will allow firmware to notify QEMU that firmware requires SMI
being triggered on CPU hot[un]plug, so that it would be able to account
for hotplugged CPU and relocate it to new SMM base and/or safely remove
CPU on unplug.
Using negotiated features, follow up patches will insert SMI upcall
into AML code, to make sure that firmware processes hotplug before
guest OS would attempt to use new CPU.
Li Qiang [Sat, 19 Sep 2020 08:27:06 +0000 (01:27 -0700)]
virtio: update MemoryRegionCaches when guest set bad features
Current the 'virtio_set_features' only update the 'MemorRegionCaches'
when the 'virtio_set_features_nocheck' return '0' which means it is
not bad features. However the guest can still trigger the access of the
used vring after set bad features. In this situation it will cause assert
failure in 'ADDRESS_SPACE_ST_CACHED'.
Commit 9b3a35ec82 ("virtio: verify that legacy support is not
accidentally on") added a safety check that requires to set
'disable-legacy=on' on vhost-user-vsock-pci device:
$ ./qemu-system-x86_64 ... \
-chardev socket,id=char0,reconnect=0,path=/tmp/vhost4.socket \
-device vhost-user-vsock-pci,chardev=char0
qemu-system-x86_64: -device vhost-user-vsock-pci,chardev=char0:
device is modern-only, use disable-legacy=on
virtio-vsock was introduced after the release of VIRTIO 1.0
specifications, so it should be 'modern-only'.
This patch forces virtio version 1 and removes the 'transitional_name'
property, as done for vhost-vsock-pci, removing the need to specify
'disable-legacy=on' on vhost-user-vsock-pci device.
Commit 9b3a35ec82 ("virtio: verify that legacy support is not
accidentally on") added a safety check that requires to set
'disable-legacy=on' on vhost-vsock-pci device:
$ ./qemu-system-x86_64 ... -device vhost-vsock-pci,guest-cid=5
qemu-system-x86_64: -device vhost-vsock-pci,guest-cid=5:
device is modern-only, use disable-legacy=on
virtio-vsock was introduced after the release of VIRTIO 1.0
specifications, so it should be 'modern-only'.
In addition Cornelia verified that forcing a legacy mode on
vhost-vsock-pci device using x86-64 host and s390x guest, so with
different endianness, produces strange behaviours.
This patch forces virtio version 1 and removes the 'transitional_name'
property removing the need to specify 'disable-legacy=on' on
vhost-vsock-pci device.
To avoid migration issues, we force virtio version 1 only when
legacy check is enabled in the new machine types (>= 5.1).
As the transitional device name is not commonly used, we do not
provide compatibility handling for it.
virtio: skip legacy support check on machine types less than 5.1
Commit 9b3a35ec82 ("virtio: verify that legacy support is not accidentally
on") added a check that returns an error if legacy support is on, but the
device does not support legacy.
Unfortunately some devices were wrongly declared legacy capable even if
they were not (e.g vhost-vsock).
To avoid migration issues, we add a virtio-device property
(x-disable-legacy-check) to skip the legacy error, printing a warning
instead, for machine types < 5.1.
configure: Fix build dependencies with vhost-vdpa.
Following the same logic as for vhost-net-user and vhost-kernel,
enable vhost-net if vhost-net-vdpa is enabled and vhost-net is not
explicitly disabled.
See 299e6f19b3e2 ("vhost-net: revamp configure logic")
Autoselect VHOST if VHOST_VDPA is set
See 21c6b0c87e85 ("configure: simplify vhost condition with Kconfig")
See 2becc36a3e53 ("meson: infrastructure for building emulators"
Problems can be triggered using;
... --enable-vhost-vdpa --disable-vhost-user --disable-vhost-kernel ...
Eric Auger [Tue, 8 Sep 2020 19:33:09 +0000 (21:33 +0200)]
virtio-iommu-pci: force virtio version 1
Commit 9b3a35ec82 ("virtio: verify that legacy support is not
accidentally on") added a safety check that requires to set
'disable-legacy=on' on virtio-iommu-pci:
qemu-system-aarch64: -device virtio-iommu-pci: device is modern-only,
use disable-legacy=on
virtio-iommu was introduced after the release of VIRTIO 1.0
specifications, so it should be 'modern-only'.
This patch forces virtio version 1 and removes the 'transitional_name'
property removing the need to specify 'disable-legacy=on' on
virtio-iommu-pci device.
Igor Mammedov [Tue, 15 Sep 2020 12:04:03 +0000 (08:04 -0400)]
cphp: remove deprecated cpu-add command(s)
These were deprecated since 4.0, remove both HMP and QMP variants.
Users should use device_add command instead. To get list of
possible CPUs and options, use 'info hotpluggable-cpus' HMP
or query-hotpluggable-cpus QMP command.
tests/qtest/vhost-user-test: prepare the tests for adding new dev class
For now only vhost-user-net device is supported by the test. Other
vhost-user devices are not tested. As a first step make source code
refactoring so new devices can reuse the same test routines. To make
this provide a new vhost_user_ops structure with the methods to
initialize device, its command line or make a proper vhost-user
responses.
vhost: check queue state in the vhost_dev_set_log routine
If the vhost-user-blk daemon provides only one virtqueue, but device was
added with several queues, then QEMU will send more VHOST-USER command
than expected by daemon side. The vhost_virtqueue_start() routine
handles such case by checking the return value from the
virtio_queue_get_desc_addr() function call. Add the same check to the
vhost_dev_set_log() routine.
vhost: recheck dev state in the vhost_migration_log routine
vhost-user devices can get a disconnect in the middle of the VHOST-USER
handshake on the migration start. If disconnect event happened right
before sending next VHOST-USER command, then the vhost_dev_set_log()
call in the vhost_migration_log() function will return error. This error
will lead to the assert() and close the QEMU migration source process.
For the vhost-user devices the disconnect event should not break the
migration process, because:
- the device will be in the stopped state, so it will not be changed
during migration
- if reconnect will be made the migration log will be reinitialized as
part of reconnect/init process:
#0 vhost_log_global_start (listener=0x563989cf7be0)
at hw/virtio/vhost.c:920
#1 0x000056398603d8bc in listener_add_address_space (listener=0x563989cf7be0,
as=0x563986ea4340 <address_space_memory>)
at softmmu/memory.c:2664
#2 0x000056398603dd30 in memory_listener_register (listener=0x563989cf7be0,
as=0x563986ea4340 <address_space_memory>)
at softmmu/memory.c:2740
#3 0x0000563985fd6956 in vhost_dev_init (hdev=0x563989cf7bd8,
opaque=0x563989cf7e30, backend_type=VHOST_BACKEND_TYPE_USER,
busyloop_timeout=0)
at hw/virtio/vhost.c:1385
#4 0x0000563985f7d0b8 in vhost_user_blk_connect (dev=0x563989cf7990)
at hw/block/vhost-user-blk.c:315
#5 0x0000563985f7d3f6 in vhost_user_blk_event (opaque=0x563989cf7990,
event=CHR_EVENT_OPENED)
at hw/block/vhost-user-blk.c:379
Update the vhost-user-blk device with the internal started_vu field which
will be used for initialization (vhost_user_blk_start) and clean up
(vhost_user_blk_stop). This additional flag in the VhostUserBlk structure
will be used to track whether the device really needs to be stopped and
cleaned up on a vhost-user level.
The disconnect event will set the overall VHOST device (not vhost-user) to
the stopped state, so it can be used by the general vhost_migration_log
routine.
Such approach could be propogated to the other vhost-user devices, but
better idea is just to make the same connect/disconnect code for all the
vhost-user devices.
This migration issue was slightly discussed earlier:
- https://lists.gnu.org/archive/html/qemu-devel/2020-05/msg01509.html
- https://lists.gnu.org/archive/html/qemu-devel/2020-05/msg05241.html
Li Qiang [Sun, 16 Aug 2020 14:22:45 +0000 (07:22 -0700)]
virtio-mem: detach the element from the virtqueue when error occurs
If error occurs while processing the virtio request we should call
'virtqueue_detach_element' to detach the element from the virtqueue
before free the elem.
Jason Wang [Mon, 7 Sep 2020 10:49:03 +0000 (18:49 +0800)]
vhost-vdpa: batch updating IOTLB mappings
To speed up the memory mapping updating between vhost-vDPA and vDPA
device driver, this patch passes the IOTLB batching flags via IOTLB
API. Two new flags was introduced, VHOST_IOTLB_BATCH_BEGIN is a hint
that a bathced IOTLB updating may be initiated from the
userspace. VHOST_IOTLB_BATCH_END is a hint that userspace has finished
the updating:
Jason Wang [Mon, 7 Sep 2020 10:49:02 +0000 (18:49 +0800)]
vhost: switch to use IOTLB v2 format
This patch tries to switch to use new kernel IOTLB format V2. Previous
version may have inconsistent ABI between 32bit and 64bit machines
because of the hole after type field. Refer kernel commit
("429711aec282 vhost: switch to use new message format") for more
information.
To enable this feature, qemu need to use a new ioctl
VHOST_SET_BACKEND_FEATURE with VHOST_BACKEND_F_IOTLB_MSG_V2 bit. A new
vhost setting backend features ops was introduced. And when we try to
set features for vhost dev, we will examine the support of new IOTLB
format and enable it. This process is total transparent to guest,
which means we can have different IOTLB message type in src and dst
during migration.
The conversion of IOTLB message is straightforward, just check the
type and behave accordingly.
* remotes/kraxel/tags/usb-20200928-pull-request:
hw/usb: Use lock guard macros
usb: hcd-xhci-sysbus: Attach xhci to sysbus device
usb/hcd-xhci: Split pci wrapper for xhci base model
usb/hcd-xhci: Move qemu-xhci device to hcd-xhci-pci.c
usb/hcd-xhci: Make dma read/writes hooks pci free
Peter Maydell [Mon, 28 Sep 2020 15:49:10 +0000 (16:49 +0100)]
Merge remote-tracking branch 'remotes/alistair/tags/pull-register-20200927' into staging
Two small patches. One with a fix for the register API instance_size
and one for removing unused address variables from load_elf.
# gpg: Signature made Sun 27 Sep 2020 14:45:06 BST
# gpg: using RSA key F6C4AC46D4934868D3B8CE8F21E10D29DF977054
# gpg: Good signature from "Alistair Francis <[email protected]>" [full]
# Primary key fingerprint: F6C4 AC46 D493 4868 D3B8 CE8F 21E1 0D29 DF97 7054
* remotes/alistair/tags/pull-register-20200927:
core/register: Specify instance_size in the TypeInfo
load_elf: Remove unused address variables from callers
* remotes/vivier2/tags/trivial-branch-for-5.2-pull-request:
docs/system/deprecated: Move lm32 and unicore32 to the right section
migration/multifd: Remove superfluous semicolons
timer: Fix timer_mod_anticipate() documentation
vhost-vdpa: remove useless variable
Add *.pyc back to the .gitignore file
virtio: vdpa: omit check return of g_malloc
meson: fix static flag summary
vhost-vdpa: fix indentation in vdpa_ops
usb/hcd-xhci: Split pci wrapper for xhci base model
This patch sets the base to use xhci as sysbus model, for which pci
specific hooks are moved to hcd-xhci-pci.c. As a part of this requirment
msi/msix interrupts handling is moved under XHCIPCIState. Made required
changes for qemu-xhci-nec.
load_elf: Remove unused address variables from callers
Several callers of load_elf() pass pointers for lowaddr and highaddr
parameters which are then not used for anything. This may stem from a
misunderstanding that load_elf need a value here but in fact it can
take NULL to ignore these values. Remove such unused variables and
pass NULL instead from callers that don't need these.
Peter Maydell [Fri, 25 Sep 2020 13:46:18 +0000 (14:46 +0100)]
Merge remote-tracking branch 'remotes/dgilbert/tags/pull-migration-20200925a' into staging
Migration and virtiofsd pull
Chuan Zheng's Dirtyrate and TLS changes, with small fixes from Dov and
Luarent.
Small virtiofs changes from Harry, Stefan, Vivek and Jiachen.
One HMP/monitor rework from me.
# gpg: Signature made Fri 25 Sep 2020 13:03:50 BST
# gpg: using RSA key 45F5C71B4A0CB7FB977A9FA90516331EBC5BFDE7
# gpg: Good signature from "Dr. David Alan Gilbert (RH2) <[email protected]>" [full]
# Primary key fingerprint: 45F5 C71B 4A0C B7FB 977A 9FA9 0516 331E BC5B FDE7
* remotes/dgilbert/tags/pull-migration-20200925a: (26 commits)
virtiofsd: Add -o allow_direct_io|no_allow_direct_io options
virtiofsd: Used glib "shared" thread pool
virtiofsd: document cache=auto default
monitor: Use LOCK_GUARD macros
migration/tls: add trace points for multifd-tls
migration/tls: add support for multifd tls-handshake
migration/tls: extract cleanup function for common-use
migration/tls: add tls_hostname into MultiFDSendParams
migration/tls: extract migration_tls_client_create for common-use
migration/tls: save hostname into MigrationState
migration: increase max-bandwidth to 128 MiB/s (1 Gib/s)
migration: Truncate state file in xen-save-devices-state
migration/dirtyrate: Add trace_calls to make it easier to debug
migration/dirtyrate: Implement qmp_cal_dirty_rate()/qmp_get_dirty_rate() function
migration/dirtyrate: Implement calculate_dirtyrate() function
migration/dirtyrate: Implement set_sample_page_period() and is_sample_period_valid()
migration/dirtyrate: skip sampling ramblock with size below MIN_RAMBLOCK_SIZE
migration/dirtyrate: Compare page hash results for recorded sampled page
migration/dirtyrate: Record hash results for each sampled page
migration/dirtyrate: move RAMBLOCK_FOREACH_MIGRATABLE into ram.h
...
Due to the commit 65da4539803373ec4eec97ffc49ee90083e56efd, the O_DIRECT
open flag of guest applications will be discarded by virtiofsd. While
this behavior makes it consistent with the virtio-9p scheme when guest
applications use direct I/O, we no longer have any chance to bypass the
host page cache.
Therefore, we add a flag 'allow_direct_io' to lo_data. If '-o
no_allow_direct_io' option is added, or none of '-o allow_direct_io' or
'-o no_allow_direct_io' is added, the 'allow_direct_io' will be set to
0, and virtiofsd discards O_DIRECT as before. If '-o allow_direct_io'
is added to the starting command-line, 'allow_direct_io' will be set to
1, so that the O_DIRECT flags will be retained and host page cache can
be bypassed.
Currently we use "exlusive" thread pools but its performance seems to be
poor. I tried using "shared" thread pools and performance seems much
better. I posted performance results here.
So lets switch to shared thread pools. We can think of making it optional
once somebody can show in what cases exclusive thread pools offer better
results. For now, my simple performance tests across the board see
better results with shared thread pools.
migration/dirtyrate: Implement set_sample_page_period() and is_sample_period_valid()
Implement is_sample_period_valid() to check if the sample period is vaild and
do set_sample_page_period() to sleep specific time between sample actions.
projects / qemu.git / log