hmp-commands-info.hx | 70 +++ hw/virtio/meson.build | 2 + hw/virtio/virtio-stub.c | 42 ++ hw/virtio/virtio.c | 1032 ++++++++++++++++++++++++++++++++++++ include/hw/virtio/vhost.h | 3 + include/hw/virtio/virtio.h | 6 + include/monitor/hmp.h | 5 + monitor/hmp-cmds.c | 310 +++++++++++ qapi/meson.build | 1 + qapi/qapi-schema.json | 1 + qapi/virtio.json | 954 +++++++++++++++++++++++++++++++++ tests/qtest/qmp-cmd-test.c | 1 + 12 files changed, 2427 insertions(+) create mode 100644 hw/virtio/virtio-stub.c create mode 100644 qapi/virtio.json
This series introduces new QMP/HMP commands to dump the status of a
virtio device at different levels.
[Jonah: Rebasing from previous patchset from Apr. 1 (v14). Original patches
are by Laurent Vivier from May 2020.
I sincerely apologize for the *massive* delay in getting this latest v15
series out. It was a perferct storm of other more pressing issues, time off,
and trying to understand why I was seeing this 30 feature bit in my PCI virtio
devices. Please see patch 3/6 for more explanation on this issue.
Rebase from v14 to v15 includes: adding the missing sign-off-by from the poster,
renaming & moving all virtio device feature map definitions to hw/virtio/virtio.c,
including brief descriptions for all status & feature bits, a new virtio device
feature map defined for virtio-rng, and mappings for virtio/vhost-vsock,
virtio-iommu, virtio-mem, and virtio transport features updated with their
newest feature bits. Most of these changes can be found in patch 3/6.
And again, sorry for the long wait on this.]
1. List available virtio devices in the machine
HMP Form:
info virtio
Example:
(qemu) info virtio
/machine/peripheral-anon/device[2]/virtio-backend [virtio-scsi]
/machine/peripheral/vsock0/virtio-backend [vhost-vsock]
/machine/peripheral/crypto0/virtio-backend [virtio-crypto]
/machine/peripheral-anon/device[1]/virtio-backend [virtio-net]
/machine/peripheral-anon/device[0]/virtio-backend [virtio-serial]
QMP Form:
{ 'command': 'x-query-virtio',
'returns': ['VirtioInfo'],
'features': [ 'unstable' ] }
Example:
-> { "execute": "x-query-virtio" }
<- { "return": [
{
"name": "virtio-scsi",
"path": "/machine/peripheral-anon/device[2]/virtio-backend"
},
{
"name": "vhost-vsock",
"path": "/machine/peripheral/vsock0/virtio-backend"
},
{
"name": "virtio-crypto",
"path": "/machine/peripheral/crypto0/virtio-backend"
},
{
"name": "virtio-net",
"path": "/machine/peripheral-anon/device[1]/virtio-backend"
},
{
"name": "virtio-serial",
"path": "/machine/peripheral-anon/device[0]/virtio-backend"
}
]
}
2. Display status of a given virtio device
HMP Form:
info virtio-status <path>
Example:
(qemu) info virtio-status /machine/peripheral/vsock0/virtio-backend
/machine/peripheral/vsock0/virtio-backend:
device_name: vhost-vsock (vhost)
device_id: 19
vhost_started: true
bus_name: (null)
broken: false
disabled: false
disable_legacy_check: false
started: true
use_started: true
start_on_kick: false
use_guest_notifier_mask: true
vm_running: true
num_vqs: 3
queue_sel: 2
isr: 0
endianness: little
status:
VIRTIO_CONFIG_S_ACKNOWLEDGE: Valid virtio device found,
VIRTIO_CONFIG_S_DRIVER: Guest OS compatible with device,
VIRTIO_CONFIG_S_FEATURES_OK: Feature negotiation complete,
VIRTIO_CONFIG_S_DRIVER_OK: Driver setup and ready
Guest features:
VIRTIO_RING_F_EVENT_IDX: Used & avail. event fields enabled,
VIRTIO_RING_F_INDIRECT_DESC: Indirect descriptors supported,
VIRTIO_F_VERSION_1: Device compliant for v1 spec (legacy)
Host features:
VIRTIO_RING_F_EVENT_IDX: Used & avail. event fields enabled,
VIRTIO_RING_F_INDIRECT_DESC: Indirect descriptors supported,
VIRTIO_F_VERSION_1: Device compliant for v1 spec (legacy),
VIRTIO_F_ANY_LAYOUT: Device accepts arbitrary desc. layouts,
VIRTIO_F_NOTIFY_ON_EMPTY: Notify when device runs out of avail. descs. on VQ
VHOST_USER_F_PROTOCOL_FEATURES: Vhost-user protocol features negotiation supported
Backend features:
VHost:
nvqs: 2
vq_index: 0
max_queues: 0
n_mem_sections: 4
n_tmp_sections: 4
backend_cap: 0
log_enabled: false
log_size: 0
Features:
VIRTIO_RING_F_EVENT_IDX: Used & avail. event fields enabled,
VIRTIO_RING_F_INDIRECT_DESC: Indirect descriptors supported,
VIRTIO_F_VERSION_1: Device compliant for v1 spec (legacy),
VIRTIO_F_ANY_LAYOUT: Device accepts arbitrary desc. layouts,
VIRTIO_F_NOTIFY_ON_EMPTY: Notify when device runs out of avail. descs. on VQ
VHOST_F_LOG_ALL: Logging write descriptors supported
Acked features:
VIRTIO_RING_F_EVENT_IDX: Used & avail. event fields enabled,
VIRTIO_RING_F_INDIRECT_DESC: Indirect descriptors supported,
VIRTIO_F_VERSION_1: Device compliant for v1 spec (legacy)
Backend features:
Protocol features:
QMP Form:
{ 'command': 'x-query-virtio-status',
'data': { 'path': 'str' },
'returns': 'VirtioStatus',
'features': [ 'unstable' ] }
Example:
-> { "execute": "x-query-virtio-status",
"arguments": { "path": "/machine/peripheral/vsock0/virtio-backend" }
}
<- { "return": {
"device-endian": "little",
"bus-name": "",
"disable-legacy-check": false,
"name": "vhost-vsock",
"started": true,
"device-id": 19,
"vhost-dev": {
"n-tmp-sections": 4,
"n-mem-sections": 4,
"max-queues": 0,
"backend-cap": 0,
"log-size": 0,
"backend-features": {
"dev-features": [
],
"transports": [
]
},
"nvqs": 2,
"protocol-features": {
"protocols": [
]
},
"vq-index": 0,
"log-enabled": false,
"acked-features": {
"dev-features": [
],
"transports": [
"VIRTIO_RING_F_EVENT_IDX: Used & avail. event fields enabled",
"VIRTIO_RING_F_INDIRECT_DESC: Indirect descriptors supported",
"VIRTIO_F_VERSION_1: Device compliant for v1 spec (legacy)"
]
},
"features": {
"dev-features": [
"VHOST_F_LOG_ALL: Logging write descriptors supported"
],
"transports": [
"VIRTIO_RING_F_EVENT_IDX: Used & avail. event fields enabled",
"VIRTIO_RING_F_INDIRECT_DESC: Indirect descriptors supported",
"VIRTIO_F_VERSION_1: Device compliant for v1 spec (legacy)",
"VIRTIO_F_ANY_LAYOUT: Device accepts arbitrary desc. layouts",
"VIRTIO_F_NOTIFY_ON_EMPTY: Notify when device runs out of avail. descs. on VQ"
]
}
},
"backend-features": {
"dev-features": [
],
"transports": [
]
},
"start-on-kick": false,
"isr": 0,
"broken": false,
"status": {
"statuses": [
"VIRTIO_CONFIG_S_ACKNOWLEDGE: Valid virtio device found",
"VIRTIO_CONFIG_S_DRIVER: Guest OS compatible with device",
"VIRTIO_CONFIG_S_FEATURES_OK: Feature negotiation complete",
"VIRTIO_CONFIG_S_DRIVER_OK: Driver setup and ready"
]
},
"num-vqs": 3,
"guest-features": {
"dev-features": [
],
"transports": [
"VIRTIO_RING_F_EVENT_IDX: Used & avail. event fields enabled",
"VIRTIO_RING_F_INDIRECT_DESC: Indirect descriptors supported",
"VIRTIO_F_VERSION_1: Device compliant for v1 spec (legacy)"
]
},
"host-features": {
"dev-features": [
"VHOST_USER_F_PROTOCOL_FEATURES: Vhost-user protocol features negotiation supported"
],
"transports": [
"VIRTIO_RING_F_EVENT_IDX: Used & avail. event fields enabled",
"VIRTIO_RING_F_INDIRECT_DESC: Indirect descriptors supported",
"VIRTIO_F_VERSION_1: Device compliant for v1 spec (legacy)",
"VIRTIO_F_ANY_LAYOUT: Device accepts arbitrary desc. layouts",
"VIRTIO_F_NOTIFY_ON_EMPTY: Notify when device runs out of avail. descs. on VQ"
]
},
"use-guest-notifier-mask": true,
"vm-running": true,
"queue-sel": 2,
"disabled": false,
"vhost-started": true,
"use-started": true
}
}
3. Display status of a given virtio queue
HMP Form:
info virtio-queue-status <path> <queue>
Example:
(qemu) info virtio-queue-status /machine/peripheral-anon/device[1]/virtio-backend 2
/machine/peripheral-anon/device[1]/virtio-backend:
device_name: virtio-net
queue_index: 2
inuse: 0
used_idx: 15
signalled_used: 15
signalled_used_valid: true
VRing:
num: 64
num_default: 64
align: 4096
desc: 0x000000011e15c000
avail: 0x000000011e15c400
used: 0x000000011e15c4c0
QMP Form:
{ 'command': 'x-query-virtio-queue-status',
'data': { 'path': 'str', 'queue': 'uint16' },
'returns': 'VirtQueueStatus',
'features': [ 'unstable' ] }
Example:
-> { "execute": "x-query-virtio-queue-status",
"arguments": { "path": "/machine/peripheral-anon/device[1]/virtio-backend",
"queue": 2 }
}
<- { "return": {
"signalled-used": 15,
"inuse": 0,
"name": "virtio-net",
"vring-align": 4096,
"vring-desc": 4814004224,
"signalled-used-valid": true,
"vring-num-default": 64,
"vring-avail": 4814005248,
"queue-index": 2,
"vring-used": 4814005440,
"used-idx": 15,
"vring-num": 64
}
}
4. Display status of a given vhost queue
HMP Form:
info virtio-vhost-queue-status <path> <queue>
Example:
(qemu) info virtio-vhost-queue-status /machine/peripheral-anon/device[1]/virtio-backend 1
/machine/peripheral-anon/device[1]/virtio-backend:
device_name: virtio-net (vhost)
kick: 0
call: 0
VRing:
num: 256
desc: 0x00007f804df5a000
desc_phys: 0x000000013315a000
desc_size: 4096
avail: 0x00007f804df5b000
avail_phys: 0x000000013315b000
avail_size: 518
used: 0x00007f804df5b240
used_phys: 0x000000013315b240
used_size: 2054
QMP Form:
{ 'command': 'x-query-virtio-vhost-queue-status',
'data': { 'path': 'str', 'queue': 'uint16' },
'returns': 'VirtVhostQueueStatus',
'features': [ 'unstable' ] }
Example:
-> { "execute": "x-query-virtio-vhost-queue-status",
"arguments": { "path": "/machine/peripheral-anon/device[1]/virtio-backend",
"queue": 1 }
}
<- { "return": {
"avail-phys": 4803538944,
"name": "virtio-net",
"used-phys": 4803539520,
"avail-size": 518,
"desc-size": 4096,
"used-size": 2054,
"desc": 140186477404160,
"num": 256,
"call": 0,
"avail": 140186477408256,
"desc-phys": 4803534848,
"used": 140186477408832,
"kick": 0
}
}
5. Display an element of a given virtio queue
HMP Form:
info virtio-queue-element <path> <queue> [index]
Example:
Dump the information of the head element of the third queue of virtio-scsi:
(qemu) info virtio-queue-element /machine/peripheral-anon/device[2]/virtio-backend 2
/machine/peripheral-anon/device[2]/virtio-backend:
device_name: virtio-scsi
index: 0
desc:
descs:
addr 0x128890000 len 12288 (write),
addr 0x128880000 len 65536 (write, next),
addr 0x128870000 len 65536 (write, next),
addr 0x128860000 len 65536 (write, next),
addr 0x128850000 len 65536 (write, next),
addr 0x128840000 len 65536 (write, next),
addr 0x128830000 len 65536 (write, next),
addr 0x128820000 len 65536 (write, next),
addr 0x128810000 len 65536 (write, next),
addr 0x128800000 len 65536 (write, next),
addr 0x116bf3000 len 53248 (write, next),
addr 0x116be3000 len 65536 (write, next),
addr 0x116bd3000 len 65536 (write, next),
addr 0x116bc3000 len 65536 (write, next),
addr 0x116bb3000 len 65536 (write, next),
addr 0x116ba3000 len 65536 (write, next),
addr 0x116b93000 len 65536 (write, next),
addr 0x116b83000 len 65536 (write, next),
addr 0x116b73000 len 65536 (write, next),
addr 0x116b63000 len 65536 (write, next),
addr 0x116b53000 len 65536 (write, next),
addr 0x126e9763b len 108 (write, next),
addr 0x9af8dae7222881a2 len 51 (next)
avail:
flags: 0
idx: 600
ring: 0
used:
flags: 0
idx: 600
QMP Form:
{ 'command': 'x-query-virtio-queue-element',
'data': { 'path': 'str', 'queue': 'uint16', '*index': 'uint16' },
'returns': 'VirtioQueueElement',
'features': [ 'unstable' ] }
Example:
-> { "execute": "x-query-virtio-queue-element",
"arguments": { "path": "/machine/peripheral-anon/device[2]/virtio-backend",
"queue": 2 }
}
<- { "return": {
"index": 1,
"name": "virtio-scsi",
"descs": [
{
"flags": ["write"],
"len": 65536,
"addr": 4852740096
},
{
"flags": ["write", "next"],
"len": 65536,
"addr": 4852674560
},
{
"flags": ["write", "next"],
"len": 65536,
"addr": 4852609024
},
{
"flags": ["write", "next"],
"len": 65536,
"addr": 4852543488
},
{
"flags": ["write", "next"],
"len": 65536,
"addr": 4852477952
},
{
"flags": ["write", "next"],
"len": 65536,
"addr": 4852412416
},
{
"flags": ["write", "next"],
"len": 65536,
"addr": 4852346880
},
{
"flags": ["write", "next"],
"len": 65536,
"addr": 4852281344
},
{
"flags": ["write", "next"],
"len": 65536,
"addr": 4852215808
},
{
"flags": ["write", "next"],
"len": 65536,
"addr": 4852150272
},
{
"flags": ["write", "next"],
"len": 65536,
"addr": 4852084736
},
{
"flags": ["write", "next"],
"len": 65536,
"addr": 4852019200
},
{
"flags": ["write", "next"],
"len": 65536,
"addr": 4851953664
},
{
"flags": ["write", "next"],
"len": 65536,
"addr": 4851888128
},
{
"flags": ["write", "next"],
"len": 65536,
"addr": 4851822592
},
{
"flags": ["write", "next"],
"len": 65536,
"addr": 4851757056
},
{
"flags": ["write", "next"],
"len": 65536,
"addr": 4851691520
},
{
"flags": ["write", "next"],
"len": 65536,
"addr": 4851625984
},
{
"flags": ["write", "next"],
"len": 65536,
"addr": 4851560448
},
{
"flags": ["write", "next"],
"len": 65536,
"addr": 4851494912
},
{
"flags": ["write", "next"],
"len": 108,
"addr": 4801325435
},
{
"flags": ["next"],
"len": 51,
"addr": 12075758188190198449
}
],
"avail": {
"idx": 922,
"flags": 0,
"ring": 1
},
"used": {
"idx": 922,
"flags": 0
}
}
}
v15: added missing sign-off in patches
rebased for upstream (Qemu 7.1)
minor reformatting to examples in qapi/virtio.json
virtio device feature maps moved to hw/virtio/virtio.c
virtio device feature maps renamed (less generic)
virtio device feature maps now include brief description of feature
virtio device status map now include brief description of status
new feature bits added to mappings for virtio/vhost-vsock,
virtio-iommu, & virtio-mem
new virtio transport feature bit added to mapping
new virtio device feature map defined for virtio-rng
v14: move definition of 'gpu_map' to virtio-gpu.c
add new 'context-init' virtio-gpu feature
created new feature map 'i2c_map' for vhost-user-i2c
fixed casting error for cross-Win64 builds
v13: minor text edit in qapi/virtio.json
v12: minor reformatting for monitor/hmp-cmds.c
use PRI types in virtio hmp monitor printfs
v11: minor reformatting in virtio.json
added synchronicity check when querying virtio devices
v10: rebased for upstream (Qemu 7.0)
reformat virtio.json for better consistency
removed all enums from virtio.json; replaced with string literals
removed @ndescs from VirtioQueueElement
removed examples in hmp-commands-info.hx (to fix 'inconsistent
literal block quoting' error from Sphinx)
v9: rebased for upstream (Qemu 6.3)
change QMP command prefixes from 'x-debug-virtio' to
'x-query-virtio'
add 'unstable' feature to all 'x-query-virtio' prefixed
QMP commands
fix odd indentation for qapi/virtio.json
device features support for virtio-mem
move HMP sub-commands under 'virtio' top level to 'info'
top level
v8: add assert in virtio_id_to_name() to make sure we're
not returning NULL
minor documentation additions to qapi/virtio.json
add virtio introspection support for vhost-user-rng
v7: rebased for upstream (Qemu 6.2)
add ability to map between numberic device ID and
string device ID (name) for virtio devices
add get_vhost() callback function for VirtIODevices
expose more fields of VirtIODevice
expose fields of vhost devices
decode vhost user protocol features
decode VirtIODevice configuration statuses
vhost support for displaying virtio queue statuses
vhost support for displaying vhost queue statuses
expose driver and device areas when introspecting a
virtio queue element
changed patch attribution
v6: rebased for upstream (Qemu 6.1)
add all virtio/vhost types in same order as
include/standard-headers/linux/virtio_ids.h
use QAPI_LIST_PREPEND in qmp_x_debug_query_virtio rather than open
coding
v5: rebased for upstream
add device name, used index, and relative indicies to virtio queue-status
HMP command
add device name to virtio queue-status QMP command
add new virtio device features
v4: re-send series as v3 didn't reach qemu-devel
v3: use qapi_free_VirtioInfoList() on the head of the list, not on the tail.
Prefix the QMP commands with 'x-debug-'
v2: introduce VirtioType enum
use an enum for the endianness
change field names to stick to naming convertions (s/_/-/)
add a patch to decode feature bits
don't check if the queue is empty to allow display of old elements
use enum for desc flags
manage indirect desc
decode device features in the HMP command
Laurent Vivier (6):
qmp: add QMP command x-query-virtio
qmp: add QMP command x-query-virtio-status
qmp: decode feature & status bits in virtio-status
qmp: add QMP commands for virtio/vhost queue-status
qmp: add QMP command x-query-virtio-queue-element
hmp: add virtio commands
hmp-commands-info.hx | 70 +++
hw/virtio/meson.build | 2 +
hw/virtio/virtio-stub.c | 42 ++
hw/virtio/virtio.c | 1032 ++++++++++++++++++++++++++++++++++++
include/hw/virtio/vhost.h | 3 +
include/hw/virtio/virtio.h | 6 +
include/monitor/hmp.h | 5 +
monitor/hmp-cmds.c | 310 +++++++++++
qapi/meson.build | 1 +
qapi/qapi-schema.json | 1 +
qapi/virtio.json | 954 +++++++++++++++++++++++++++++++++
tests/qtest/qmp-cmd-test.c | 1 +
12 files changed, 2427 insertions(+)
create mode 100644 hw/virtio/virtio-stub.c
create mode 100644 qapi/virtio.json
--
2.31.1
Hey Michael, when you get the chance, could you review these patches for me? Specifically patch 3/6, as that has the majority of the changes made for this series. Thanks!
Jonah
________________________________
From: Jonah Palmer
Sent: Thursday, August 11, 2022 8:26 AM
To: qemu-devel@nongnu.org <qemu-devel@nongnu.org>
Cc: mst@redhat.com <mst@redhat.com>; qemu_oss@crudebyte.com <qemu_oss@crudebyte.com>; kraxel@redhat.com <kraxel@redhat.com>; Si-Wei Liu <si-wei.liu@oracle.com>; Joao Martins <joao.m.martins@oracle.com>; eblake@redhat.com <eblake@redhat.com>; qemu-block@nongnu.org <qemu-block@nongnu.org>; david@redhat.com <david@redhat.com>; armbru@redhat.com <armbru@redhat.com>; arei.gonglei@huawei.com <arei.gonglei@huawei.com>; marcandre.lureau@redhat.com <marcandre.lureau@redhat.com>; lvivier@redhat.com <lvivier@redhat.com>; thuth@redhat.com <thuth@redhat.com>; michael.roth@amd.com <michael.roth@amd.com>; groug@kaod.org <groug@kaod.org>; dgilbert@redhat.com <dgilbert@redhat.com>; eric.auger@redhat.com <eric.auger@redhat.com>; stefanha@redhat.com <stefanha@redhat.com>; Boris Ostrovsky <boris.ostrovsky@oracle.com>; kwolf@redhat.com <kwolf@redhat.com>; mathieu.poirier@linaro.org <mathieu.poirier@linaro.org>; raphael.norwitz@nutanix.com <raphael.norwitz@nutanix.com>; pbonzini@redhat.com <pbonzini@redhat.com>
Subject: [PATCH v15 0/6] hmp,qmp: Add commands to introspect virtio devices
This series introduces new QMP/HMP commands to dump the status of a
virtio device at different levels.
[Jonah: Rebasing from previous patchset from Apr. 1 (v14). Original patches
are by Laurent Vivier from May 2020.
I sincerely apologize for the *massive* delay in getting this latest v15
series out. It was a perferct storm of other more pressing issues, time off,
and trying to understand why I was seeing this 30 feature bit in my PCI virtio
devices. Please see patch 3/6 for more explanation on this issue.
Rebase from v14 to v15 includes: adding the missing sign-off-by from the poster,
renaming & moving all virtio device feature map definitions to hw/virtio/virtio.c,
including brief descriptions for all status & feature bits, a new virtio device
feature map defined for virtio-rng, and mappings for virtio/vhost-vsock,
virtio-iommu, virtio-mem, and virtio transport features updated with their
newest feature bits. Most of these changes can be found in patch 3/6.
And again, sorry for the long wait on this.]
1. List available virtio devices in the machine
HMP Form:
info virtio
Example:
(qemu) info virtio
/machine/peripheral-anon/device[2]/virtio-backend [virtio-scsi]
/machine/peripheral/vsock0/virtio-backend [vhost-vsock]
/machine/peripheral/crypto0/virtio-backend [virtio-crypto]
/machine/peripheral-anon/device[1]/virtio-backend [virtio-net]
/machine/peripheral-anon/device[0]/virtio-backend [virtio-serial]
QMP Form:
{ 'command': 'x-query-virtio',
'returns': ['VirtioInfo'],
'features': [ 'unstable' ] }
Example:
-> { "execute": "x-query-virtio" }
<- { "return": [
{
"name": "virtio-scsi",
"path": "/machine/peripheral-anon/device[2]/virtio-backend"
},
{
"name": "vhost-vsock",
"path": "/machine/peripheral/vsock0/virtio-backend"
},
{
"name": "virtio-crypto",
"path": "/machine/peripheral/crypto0/virtio-backend"
},
{
"name": "virtio-net",
"path": "/machine/peripheral-anon/device[1]/virtio-backend"
},
{
"name": "virtio-serial",
"path": "/machine/peripheral-anon/device[0]/virtio-backend"
}
]
}
2. Display status of a given virtio device
HMP Form:
info virtio-status <path>
Example:
(qemu) info virtio-status /machine/peripheral/vsock0/virtio-backend
/machine/peripheral/vsock0/virtio-backend:
device_name: vhost-vsock (vhost)
device_id: 19
vhost_started: true
bus_name: (null)
broken: false
disabled: false
disable_legacy_check: false
started: true
use_started: true
start_on_kick: false
use_guest_notifier_mask: true
vm_running: true
num_vqs: 3
queue_sel: 2
isr: 0
endianness: little
status:
VIRTIO_CONFIG_S_ACKNOWLEDGE: Valid virtio device found,
VIRTIO_CONFIG_S_DRIVER: Guest OS compatible with device,
VIRTIO_CONFIG_S_FEATURES_OK: Feature negotiation complete,
VIRTIO_CONFIG_S_DRIVER_OK: Driver setup and ready
Guest features:
VIRTIO_RING_F_EVENT_IDX: Used & avail. event fields enabled,
VIRTIO_RING_F_INDIRECT_DESC: Indirect descriptors supported,
VIRTIO_F_VERSION_1: Device compliant for v1 spec (legacy)
Host features:
VIRTIO_RING_F_EVENT_IDX: Used & avail. event fields enabled,
VIRTIO_RING_F_INDIRECT_DESC: Indirect descriptors supported,
VIRTIO_F_VERSION_1: Device compliant for v1 spec (legacy),
VIRTIO_F_ANY_LAYOUT: Device accepts arbitrary desc. layouts,
VIRTIO_F_NOTIFY_ON_EMPTY: Notify when device runs out of avail. descs. on VQ
VHOST_USER_F_PROTOCOL_FEATURES: Vhost-user protocol features negotiation supported
Backend features:
VHost:
nvqs: 2
vq_index: 0
max_queues: 0
n_mem_sections: 4
n_tmp_sections: 4
backend_cap: 0
log_enabled: false
log_size: 0
Features:
VIRTIO_RING_F_EVENT_IDX: Used & avail. event fields enabled,
VIRTIO_RING_F_INDIRECT_DESC: Indirect descriptors supported,
VIRTIO_F_VERSION_1: Device compliant for v1 spec (legacy),
VIRTIO_F_ANY_LAYOUT: Device accepts arbitrary desc. layouts,
VIRTIO_F_NOTIFY_ON_EMPTY: Notify when device runs out of avail. descs. on VQ
VHOST_F_LOG_ALL: Logging write descriptors supported
Acked features:
VIRTIO_RING_F_EVENT_IDX: Used & avail. event fields enabled,
VIRTIO_RING_F_INDIRECT_DESC: Indirect descriptors supported,
VIRTIO_F_VERSION_1: Device compliant for v1 spec (legacy)
Backend features:
Protocol features:
QMP Form:
{ 'command': 'x-query-virtio-status',
'data': { 'path': 'str' },
'returns': 'VirtioStatus',
'features': [ 'unstable' ] }
Example:
-> { "execute": "x-query-virtio-status",
"arguments": { "path": "/machine/peripheral/vsock0/virtio-backend" }
}
<- { "return": {
"device-endian": "little",
"bus-name": "",
"disable-legacy-check": false,
"name": "vhost-vsock",
"started": true,
"device-id": 19,
"vhost-dev": {
"n-tmp-sections": 4,
"n-mem-sections": 4,
"max-queues": 0,
"backend-cap": 0,
"log-size": 0,
"backend-features": {
"dev-features": [
],
"transports": [
]
},
"nvqs": 2,
"protocol-features": {
"protocols": [
]
},
"vq-index": 0,
"log-enabled": false,
"acked-features": {
"dev-features": [
],
"transports": [
"VIRTIO_RING_F_EVENT_IDX: Used & avail. event fields enabled",
"VIRTIO_RING_F_INDIRECT_DESC: Indirect descriptors supported",
"VIRTIO_F_VERSION_1: Device compliant for v1 spec (legacy)"
]
},
"features": {
"dev-features": [
"VHOST_F_LOG_ALL: Logging write descriptors supported"
],
"transports": [
"VIRTIO_RING_F_EVENT_IDX: Used & avail. event fields enabled",
"VIRTIO_RING_F_INDIRECT_DESC: Indirect descriptors supported",
"VIRTIO_F_VERSION_1: Device compliant for v1 spec (legacy)",
"VIRTIO_F_ANY_LAYOUT: Device accepts arbitrary desc. layouts",
"VIRTIO_F_NOTIFY_ON_EMPTY: Notify when device runs out of avail. descs. on VQ"
]
}
},
"backend-features": {
"dev-features": [
],
"transports": [
]
},
"start-on-kick": false,
"isr": 0,
"broken": false,
"status": {
"statuses": [
"VIRTIO_CONFIG_S_ACKNOWLEDGE: Valid virtio device found",
"VIRTIO_CONFIG_S_DRIVER: Guest OS compatible with device",
"VIRTIO_CONFIG_S_FEATURES_OK: Feature negotiation complete",
"VIRTIO_CONFIG_S_DRIVER_OK: Driver setup and ready"
]
},
"num-vqs": 3,
"guest-features": {
"dev-features": [
],
"transports": [
"VIRTIO_RING_F_EVENT_IDX: Used & avail. event fields enabled",
"VIRTIO_RING_F_INDIRECT_DESC: Indirect descriptors supported",
"VIRTIO_F_VERSION_1: Device compliant for v1 spec (legacy)"
]
},
"host-features": {
"dev-features": [
"VHOST_USER_F_PROTOCOL_FEATURES: Vhost-user protocol features negotiation supported"
],
"transports": [
"VIRTIO_RING_F_EVENT_IDX: Used & avail. event fields enabled",
"VIRTIO_RING_F_INDIRECT_DESC: Indirect descriptors supported",
"VIRTIO_F_VERSION_1: Device compliant for v1 spec (legacy)",
"VIRTIO_F_ANY_LAYOUT: Device accepts arbitrary desc. layouts",
"VIRTIO_F_NOTIFY_ON_EMPTY: Notify when device runs out of avail. descs. on VQ"
]
},
"use-guest-notifier-mask": true,
"vm-running": true,
"queue-sel": 2,
"disabled": false,
"vhost-started": true,
"use-started": true
}
}
3. Display status of a given virtio queue
HMP Form:
info virtio-queue-status <path> <queue>
Example:
(qemu) info virtio-queue-status /machine/peripheral-anon/device[1]/virtio-backend 2
/machine/peripheral-anon/device[1]/virtio-backend:
device_name: virtio-net
queue_index: 2
inuse: 0
used_idx: 15
signalled_used: 15
signalled_used_valid: true
VRing:
num: 64
num_default: 64
align: 4096
desc: 0x000000011e15c000
avail: 0x000000011e15c400
used: 0x000000011e15c4c0
QMP Form:
{ 'command': 'x-query-virtio-queue-status',
'data': { 'path': 'str', 'queue': 'uint16' },
'returns': 'VirtQueueStatus',
'features': [ 'unstable' ] }
Example:
-> { "execute": "x-query-virtio-queue-status",
"arguments": { "path": "/machine/peripheral-anon/device[1]/virtio-backend",
"queue": 2 }
}
<- { "return": {
"signalled-used": 15,
"inuse": 0,
"name": "virtio-net",
"vring-align": 4096,
"vring-desc": 4814004224,
"signalled-used-valid": true,
"vring-num-default": 64,
"vring-avail": 4814005248,
"queue-index": 2,
"vring-used": 4814005440,
"used-idx": 15,
"vring-num": 64
}
}
4. Display status of a given vhost queue
HMP Form:
info virtio-vhost-queue-status <path> <queue>
Example:
(qemu) info virtio-vhost-queue-status /machine/peripheral-anon/device[1]/virtio-backend 1
/machine/peripheral-anon/device[1]/virtio-backend:
device_name: virtio-net (vhost)
kick: 0
call: 0
VRing:
num: 256
desc: 0x00007f804df5a000
desc_phys: 0x000000013315a000
desc_size: 4096
avail: 0x00007f804df5b000
avail_phys: 0x000000013315b000
avail_size: 518
used: 0x00007f804df5b240
used_phys: 0x000000013315b240
used_size: 2054
QMP Form:
{ 'command': 'x-query-virtio-vhost-queue-status',
'data': { 'path': 'str', 'queue': 'uint16' },
'returns': 'VirtVhostQueueStatus',
'features': [ 'unstable' ] }
Example:
-> { "execute": "x-query-virtio-vhost-queue-status",
"arguments": { "path": "/machine/peripheral-anon/device[1]/virtio-backend",
"queue": 1 }
}
<- { "return": {
"avail-phys": 4803538944,
"name": "virtio-net",
"used-phys": 4803539520,
"avail-size": 518,
"desc-size": 4096,
"used-size": 2054,
"desc": 140186477404160,
"num": 256,
"call": 0,
"avail": 140186477408256,
"desc-phys": 4803534848,
"used": 140186477408832,
"kick": 0
}
}
5. Display an element of a given virtio queue
HMP Form:
info virtio-queue-element <path> <queue> [index]
Example:
Dump the information of the head element of the third queue of virtio-scsi:
(qemu) info virtio-queue-element /machine/peripheral-anon/device[2]/virtio-backend 2
/machine/peripheral-anon/device[2]/virtio-backend:
device_name: virtio-scsi
index: 0
desc:
descs:
addr 0x128890000 len 12288 (write),
addr 0x128880000 len 65536 (write, next),
addr 0x128870000 len 65536 (write, next),
addr 0x128860000 len 65536 (write, next),
addr 0x128850000 len 65536 (write, next),
addr 0x128840000 len 65536 (write, next),
addr 0x128830000 len 65536 (write, next),
addr 0x128820000 len 65536 (write, next),
addr 0x128810000 len 65536 (write, next),
addr 0x128800000 len 65536 (write, next),
addr 0x116bf3000 len 53248 (write, next),
addr 0x116be3000 len 65536 (write, next),
addr 0x116bd3000 len 65536 (write, next),
addr 0x116bc3000 len 65536 (write, next),
addr 0x116bb3000 len 65536 (write, next),
addr 0x116ba3000 len 65536 (write, next),
addr 0x116b93000 len 65536 (write, next),
addr 0x116b83000 len 65536 (write, next),
addr 0x116b73000 len 65536 (write, next),
addr 0x116b63000 len 65536 (write, next),
addr 0x116b53000 len 65536 (write, next),
addr 0x126e9763b len 108 (write, next),
addr 0x9af8dae7222881a2 len 51 (next)
avail:
flags: 0
idx: 600
ring: 0
used:
flags: 0
idx: 600
QMP Form:
{ 'command': 'x-query-virtio-queue-element',
'data': { 'path': 'str', 'queue': 'uint16', '*index': 'uint16' },
'returns': 'VirtioQueueElement',
'features': [ 'unstable' ] }
Example:
-> { "execute": "x-query-virtio-queue-element",
"arguments": { "path": "/machine/peripheral-anon/device[2]/virtio-backend",
"queue": 2 }
}
<- { "return": {
"index": 1,
"name": "virtio-scsi",
"descs": [
{
"flags": ["write"],
"len": 65536,
"addr": 4852740096
},
{
"flags": ["write", "next"],
"len": 65536,
"addr": 4852674560
},
{
"flags": ["write", "next"],
"len": 65536,
"addr": 4852609024
},
{
"flags": ["write", "next"],
"len": 65536,
"addr": 4852543488
},
{
"flags": ["write", "next"],
"len": 65536,
"addr": 4852477952
},
{
"flags": ["write", "next"],
"len": 65536,
"addr": 4852412416
},
{
"flags": ["write", "next"],
"len": 65536,
"addr": 4852346880
},
{
"flags": ["write", "next"],
"len": 65536,
"addr": 4852281344
},
{
"flags": ["write", "next"],
"len": 65536,
"addr": 4852215808
},
{
"flags": ["write", "next"],
"len": 65536,
"addr": 4852150272
},
{
"flags": ["write", "next"],
"len": 65536,
"addr": 4852084736
},
{
"flags": ["write", "next"],
"len": 65536,
"addr": 4852019200
},
{
"flags": ["write", "next"],
"len": 65536,
"addr": 4851953664
},
{
"flags": ["write", "next"],
"len": 65536,
"addr": 4851888128
},
{
"flags": ["write", "next"],
"len": 65536,
"addr": 4851822592
},
{
"flags": ["write", "next"],
"len": 65536,
"addr": 4851757056
},
{
"flags": ["write", "next"],
"len": 65536,
"addr": 4851691520
},
{
"flags": ["write", "next"],
"len": 65536,
"addr": 4851625984
},
{
"flags": ["write", "next"],
"len": 65536,
"addr": 4851560448
},
{
"flags": ["write", "next"],
"len": 65536,
"addr": 4851494912
},
{
"flags": ["write", "next"],
"len": 108,
"addr": 4801325435
},
{
"flags": ["next"],
"len": 51,
"addr": 12075758188190198449
}
],
"avail": {
"idx": 922,
"flags": 0,
"ring": 1
},
"used": {
"idx": 922,
"flags": 0
}
}
}
v15: added missing sign-off in patches
rebased for upstream (Qemu 7.1)
minor reformatting to examples in qapi/virtio.json
virtio device feature maps moved to hw/virtio/virtio.c
virtio device feature maps renamed (less generic)
virtio device feature maps now include brief description of feature
virtio device status map now include brief description of status
new feature bits added to mappings for virtio/vhost-vsock,
virtio-iommu, & virtio-mem
new virtio transport feature bit added to mapping
new virtio device feature map defined for virtio-rng
v14: move definition of 'gpu_map' to virtio-gpu.c
add new 'context-init' virtio-gpu feature
created new feature map 'i2c_map' for vhost-user-i2c
fixed casting error for cross-Win64 builds
v13: minor text edit in qapi/virtio.json
v12: minor reformatting for monitor/hmp-cmds.c
use PRI types in virtio hmp monitor printfs
v11: minor reformatting in virtio.json
added synchronicity check when querying virtio devices
v10: rebased for upstream (Qemu 7.0)
reformat virtio.json for better consistency
removed all enums from virtio.json; replaced with string literals
removed @ndescs from VirtioQueueElement
removed examples in hmp-commands-info.hx (to fix 'inconsistent
literal block quoting' error from Sphinx)
v9: rebased for upstream (Qemu 6.3)
change QMP command prefixes from 'x-debug-virtio' to
'x-query-virtio'
add 'unstable' feature to all 'x-query-virtio' prefixed
QMP commands
fix odd indentation for qapi/virtio.json
device features support for virtio-mem
move HMP sub-commands under 'virtio' top level to 'info'
top level
v8: add assert in virtio_id_to_name() to make sure we're
not returning NULL
minor documentation additions to qapi/virtio.json
add virtio introspection support for vhost-user-rng
v7: rebased for upstream (Qemu 6.2)
add ability to map between numberic device ID and
string device ID (name) for virtio devices
add get_vhost() callback function for VirtIODevices
expose more fields of VirtIODevice
expose fields of vhost devices
decode vhost user protocol features
decode VirtIODevice configuration statuses
vhost support for displaying virtio queue statuses
vhost support for displaying vhost queue statuses
expose driver and device areas when introspecting a
virtio queue element
changed patch attribution
v6: rebased for upstream (Qemu 6.1)
add all virtio/vhost types in same order as
include/standard-headers/linux/virtio_ids.h
use QAPI_LIST_PREPEND in qmp_x_debug_query_virtio rather than open
coding
v5: rebased for upstream
add device name, used index, and relative indicies to virtio queue-status
HMP command
add device name to virtio queue-status QMP command
add new virtio device features
v4: re-send series as v3 didn't reach qemu-devel
v3: use qapi_free_VirtioInfoList() on the head of the list, not on the tail.
Prefix the QMP commands with 'x-debug-'
v2: introduce VirtioType enum
use an enum for the endianness
change field names to stick to naming convertions (s/_/-/)
add a patch to decode feature bits
don't check if the queue is empty to allow display of old elements
use enum for desc flags
manage indirect desc
decode device features in the HMP command
Laurent Vivier (6):
qmp: add QMP command x-query-virtio
qmp: add QMP command x-query-virtio-status
qmp: decode feature & status bits in virtio-status
qmp: add QMP commands for virtio/vhost queue-status
qmp: add QMP command x-query-virtio-queue-element
hmp: add virtio commands
hmp-commands-info.hx | 70 +++
hw/virtio/meson.build | 2 +
hw/virtio/virtio-stub.c | 42 ++
hw/virtio/virtio.c | 1032 ++++++++++++++++++++++++++++++++++++
include/hw/virtio/vhost.h | 3 +
include/hw/virtio/virtio.h | 6 +
include/monitor/hmp.h | 5 +
monitor/hmp-cmds.c | 310 +++++++++++
qapi/meson.build | 1 +
qapi/qapi-schema.json | 1 +
qapi/virtio.json | 954 +++++++++++++++++++++++++++++++++
tests/qtest/qmp-cmd-test.c | 1 +
12 files changed, 2427 insertions(+)
create mode 100644 hw/virtio/virtio-stub.c
create mode 100644 qapi/virtio.json
--
2.31.1
© 2016 - 2024 Red Hat, Inc.