Skip to contentSkip to conversation

virtio: Add vhost-user-gpio device's support

PATCH Applied Viresh Kumar 5 participants 24 messages Jan 12, 2022 – Apr 5, 2022 Arm/Amlogic Meson SoC support

lore
Share & tools
lore
#0 lore raw

cover virtio: Add vhost-user-gpio device's support

Hello,

This patchset adds vhost-user-gpio device's support in Qemu. The support for the same has already been added to virtio specification and Linux Kernel.

A Rust based backend is also in progress and is tested against this patchset:

https://github.com/rust-vmm/vhost-device/pull/76

--
Viresh

Viresh Kumar (2):
hw/virtio: add boilerplate for vhost-user-gpio device hw/virtio: add vhost-user-gpio-pci boilerplate

 hw/virtio/Kconfig                   |   5 +
 hw/virtio/meson.build               |   2 +
 hw/virtio/vhost-user-gpio-pci.c     |  69 ++++++
 hw/virtio/vhost-user-gpio.c         | 343 ++++++++++++++++++++++++++++
 include/hw/virtio/vhost-user-gpio.h |  35 +++
 5 files changed, 454 insertions(+)
 create mode 100644 hw/virtio/vhost-user-gpio-pci.c
 create mode 100644 hw/virtio/vhost-user-gpio.c
 create mode 100644 include/hw/virtio/vhost-user-gpio.h
Signature · 1 line
2.31.1.272.g89b43f80a514
#1 lore raw

PATCH 1/2 hw/virtio: add boilerplate for vhost-user-gpio device

This creates the QEMU side of the vhost-user-gpio device which connects to the remote daemon. It is based of vhost-user-i2c code.

---
 hw/virtio/Kconfig                   |   5 +
 hw/virtio/meson.build               |   1 +
 hw/virtio/vhost-user-gpio.c         | 343 ++++++++++++++++++++++++++++
 include/hw/virtio/vhost-user-gpio.h |  35 +++
 4 files changed, 384 insertions(+)
 create mode 100644 hw/virtio/vhost-user-gpio.c
 create mode 100644 include/hw/virtio/vhost-user-gpio.h
hw/virtio/Kconfig +5 -0
index c144d42f9bd0..b16a63609530 100644
@@ -59,6 +59,11 @@ config VIRTIO_MEM
     depends on VIRTIO_MEM_SUPPORTED
     select MEM_DEVICE
 
+config VHOST_USER_GPIO
+    bool
+    default y
+    depends on VIRTIO && VHOST_USER
+
 config VHOST_USER_I2C
     bool
     default y
hw/virtio/meson.build +1 -0
index 521f7d64a86a..c7e67adff337 100644
@@ -29,6 +29,7 @@ virtio_ss.add(when: 'CONFIG_VHOST_USER_I2C', if_true: files('vhost-user-i2c.c'))
 virtio_ss.add(when: ['CONFIG_VIRTIO_PCI', 'CONFIG_VHOST_USER_I2C'], if_true: files('vhost-user-i2c-pci.c'))
 virtio_ss.add(when: 'CONFIG_VHOST_USER_RNG', if_true: files('vhost-user-rng.c'))
 virtio_ss.add(when: ['CONFIG_VHOST_USER_RNG', 'CONFIG_VIRTIO_PCI'], if_true: files('vhost-user-rng-pci.c'))
+virtio_ss.add(when: 'CONFIG_VHOST_USER_GPIO', if_true: files('vhost-user-gpio.c'))
 
 virtio_pci_ss = ss.source_set()
 virtio_pci_ss.add(when: 'CONFIG_VHOST_VSOCK', if_true: files('vhost-vsock-pci.c'))
hw/virtio/vhost-user-gpio.c new file +343 -0

343 lines not on this page · raw message

include/hw/virtio/vhost-user-gpio.h new file +35 -0
new file mode 100644
index 000000000000..afeb56f53e99
@@ -0,0 +1,35 @@
+/*
+ * Vhost-user GPIO virtio device
+ *
+ * Copyright (c) 2021 Viresh Kumar <[email protected]>
+ *
+ * SPDX-License-Identifier: GPL-2.0-or-later
+ */
+
+#ifndef _QEMU_VHOST_USER_GPIO_H
+#define _QEMU_VHOST_USER_GPIO_H
+
+#include "hw/virtio/virtio.h"
+#include "hw/virtio/vhost.h"
+#include "hw/virtio/vhost-user.h"
+#include "standard-headers/linux/virtio_gpio.h"
+#include "chardev/char-fe.h"
+
+#define TYPE_VHOST_USER_GPIO "vhost-user-gpio-device"
+OBJECT_DECLARE_SIMPLE_TYPE(VHostUserGPIO, VHOST_USER_GPIO);
+
+struct VHostUserGPIO {
+    /*< private >*/
+    VirtIODevice parent;
+    CharBackend chardev;
+    struct virtio_gpio_config config;
+    struct vhost_virtqueue *vhost_vq;
+    struct vhost_dev vhost_dev;
+    VhostUserState vhost_user;
+    VirtQueue *command_vq;
+    VirtQueue *interrupt_vq;
+    bool connected;
+    /*< public >*/
+};
+
+#endif /* _QEMU_VHOST_USER_GPIO_H */
Signature · 1 line
2.31.1.272.g89b43f80a514
#3 lore raw

Re: [PATCH 1/2] hw/virtio: add boilerplate for vhost-user-gpio device

In reply to Viresh Kumar
Quoted 4 lines · Viresh Kumar

This creates the QEMU side of the vhost-user-gpio device which connects to the remote daemon. It is based of vhost-user-i2c code.

<snip>

Quoted 7 lines

+++ b/include/hw/virtio/vhost-user-gpio.h

include/hw/virtio/vhost-user-gpio.h new file +5 -0
@@ -0,0 +1,35 @@
+/*
+ * Vhost-user GPIO virtio device
+ *
+ * Copyright (c) 2021 Viresh Kumar <[email protected]>
+ *

include/hw/virtio/vhost-user-gpio.h line 15

Quoted 10 lines
include/hw/virtio/vhost-user-gpio.h new file +10 -0
+ * SPDX-License-Identifier: GPL-2.0-or-later
+ */
+
+#ifndef _QEMU_VHOST_USER_GPIO_H
+#define _QEMU_VHOST_USER_GPIO_H
+
+#include "hw/virtio/virtio.h"
+#include "hw/virtio/vhost.h"
+#include "hw/virtio/vhost-user.h"
+#include "standard-headers/linux/virtio_gpio.h"

Hmm this fails:

In file included from ../../hw/virtio/vhost-user-gpio.c:13:
/home/alex/lsrc/qemu.git/include/hw/virtio/vhost-user-gpio.h:15:10: fatal error: standard-headers/linux/virtio_gpio.h: No such file or directory
15 | #include "standard-headers/linux/virtio_gpio.h"
| ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
compilation terminated.

The usual solution is to create a patch that imports the headers using:

./scripts/update-linux-headers.sh

either from the current mainline (or your own tree if the feature is in flight) and mark the patch clearly as not for merging.

Signature · 1 line
Alex Bennée
#4 lore raw

Re: [PATCH 1/2] hw/virtio: add boilerplate for vhost-user-gpio device

Quoted 34 lines · Alex Bennée
Quoted 4 lines · Viresh Kumar

This creates the QEMU side of the vhost-user-gpio device which connects to the remote daemon. It is based of vhost-user-i2c code.

<snip>

Quoted 7 lines

+++ b/include/hw/virtio/vhost-user-gpio.h

include/hw/virtio/vhost-user-gpio.h new file +5 -0
@@ -0,0 +1,35 @@
+/*
+ * Vhost-user GPIO virtio device
+ *
+ * Copyright (c) 2021 Viresh Kumar <[email protected]>
+ *

include/hw/virtio/vhost-user-gpio.h line 15

Quoted 10 lines
include/hw/virtio/vhost-user-gpio.h new file +10 -0
+ * SPDX-License-Identifier: GPL-2.0-or-later
+ */
+
+#ifndef _QEMU_VHOST_USER_GPIO_H
+#define _QEMU_VHOST_USER_GPIO_H
+
+#include "hw/virtio/virtio.h"
+#include "hw/virtio/vhost.h"
+#include "hw/virtio/vhost-user.h"
+#include "standard-headers/linux/virtio_gpio.h"

Hmm this fails:

In file included from ../../hw/virtio/vhost-user-gpio.c:13:
/home/alex/lsrc/qemu.git/include/hw/virtio/vhost-user-gpio.h:15:10: fatal error: standard-headers/linux/virtio_gpio.h: No such file or directory
15 | #include "standard-headers/linux/virtio_gpio.h"
| ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
compilation terminated.

Quoted 6 lines

The usual solution is to create a patch that imports the headers using:

./scripts/update-linux-headers.sh

either from the current mainline (or your own tree if the feature is in flight) and mark the patch clearly as not for merging.

I have that patch with me and I planned to mention it in the cover-letter, but completely failed while sending it.

Should I resend this with the NOT-TO-BE-APPLIED patch now ? Everything is in mainline, so the committer just needs to run scripts/update-linux-headers.sh.

Signature · 1 line
viresh
#5 lore raw

Re: [PATCH 1/2] hw/virtio: add boilerplate for vhost-user-gpio device

Quoted 42 lines · Viresh Kumar
Quoted 34 lines · Alex Bennée
Quoted 4 lines · Viresh Kumar

This creates the QEMU side of the vhost-user-gpio device which connects to the remote daemon. It is based of vhost-user-i2c code.

<snip>

Quoted 7 lines

+++ b/include/hw/virtio/vhost-user-gpio.h

include/hw/virtio/vhost-user-gpio.h new file +5 -0
@@ -0,0 +1,35 @@
+/*
+ * Vhost-user GPIO virtio device
+ *
+ * Copyright (c) 2021 Viresh Kumar <[email protected]>
+ *

include/hw/virtio/vhost-user-gpio.h line 15

Quoted 10 lines
include/hw/virtio/vhost-user-gpio.h new file +10 -0
+ * SPDX-License-Identifier: GPL-2.0-or-later
+ */
+
+#ifndef _QEMU_VHOST_USER_GPIO_H
+#define _QEMU_VHOST_USER_GPIO_H
+
+#include "hw/virtio/virtio.h"
+#include "hw/virtio/vhost.h"
+#include "hw/virtio/vhost-user.h"
+#include "standard-headers/linux/virtio_gpio.h"

Hmm this fails:

In file included from ../../hw/virtio/vhost-user-gpio.c:13:
/home/alex/lsrc/qemu.git/include/hw/virtio/vhost-user-gpio.h:15:10: fatal error: standard-headers/linux/virtio_gpio.h: No such file or directory
15 | #include "standard-headers/linux/virtio_gpio.h"
| ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
compilation terminated.

Quoted 6 lines

The usual solution is to create a patch that imports the headers using:

./scripts/update-linux-headers.sh

either from the current mainline (or your own tree if the feature is in flight) and mark the patch clearly as not for merging.

Quoted 6 lines

I have that patch with me and I planned to mention it in the cover-letter, but completely failed while sending it.

Should I resend this with the NOT-TO-BE-APPLIED patch now ? Everything is in mainline, so the committer just needs to run scripts/update-linux-headers.sh.

If it's a released version of mainline even better because you can just include the patch generated from a clean release and it can get merged.

Signature · 1 line
Alex Bennée
#6 lore raw

PATCH 2/2 hw/virtio: add vhost-user-gpio-pci boilerplate

In reply to Viresh Kumar

This allows is to instantiate a vhost-user-gpio device as part of a PCI bus. It is mostly boilerplate which looks pretty similar to the vhost-user-fs-pci device.

---
 hw/virtio/meson.build           |  1 +
 hw/virtio/vhost-user-gpio-pci.c | 69 +++++++++++++++++++++++++++++++++
 2 files changed, 70 insertions(+)
 create mode 100644 hw/virtio/vhost-user-gpio-pci.c
hw/virtio/meson.build +1 -0
index c7e67adff337..fe6a4e9acb2a 100644
@@ -30,6 +30,7 @@ virtio_ss.add(when: ['CONFIG_VIRTIO_PCI', 'CONFIG_VHOST_USER_I2C'], if_true: fil
 virtio_ss.add(when: 'CONFIG_VHOST_USER_RNG', if_true: files('vhost-user-rng.c'))
 virtio_ss.add(when: ['CONFIG_VHOST_USER_RNG', 'CONFIG_VIRTIO_PCI'], if_true: files('vhost-user-rng-pci.c'))
 virtio_ss.add(when: 'CONFIG_VHOST_USER_GPIO', if_true: files('vhost-user-gpio.c'))
+virtio_ss.add(when: ['CONFIG_VIRTIO_PCI', 'CONFIG_VHOST_USER_GPIO'], if_true: files('vhost-user-gpio-pci.c'))
 
 virtio_pci_ss = ss.source_set()
 virtio_pci_ss.add(when: 'CONFIG_VHOST_VSOCK', if_true: files('vhost-vsock-pci.c'))
hw/virtio/vhost-user-gpio-pci.c new file +69 -0

69 lines not on this page · raw message

Signature · 1 line
2.31.1.272.g89b43f80a514
Reviewed-by on v1 2/2
#7 lore raw

Re: [PATCH 2/2] hw/virtio: add vhost-user-gpio-pci boilerplate

Quoted 5 lines · Viresh Kumar

This allows is to instantiate a vhost-user-gpio device as part of a PCI bus. It is mostly boilerplate which looks pretty similar to the vhost-user-fs-pci device.

Signature · 1 line
Alex Bennée
#8 lore raw
In reply to Viresh Kumar
Quoted 8 lines · Viresh Kumar

Hello,

This patchset adds vhost-user-gpio device's support in Qemu. The support for the same has already been added to virtio specification and Linux Kernel.

A Rust based backend is also in progress and is tested against this patchset:

https://github.com/rust-vmm/vhost-device/pull/76

I'm reluctant to add this with no tests in tree. Want to write a minimal libhost-user based backend? We also need some maintainers to step up.

Quoted 19 lines

--
Viresh

Viresh Kumar (2):
hw/virtio: add boilerplate for vhost-user-gpio device hw/virtio: add vhost-user-gpio-pci boilerplate

 hw/virtio/Kconfig                   |   5 +
 hw/virtio/meson.build               |   2 +
 hw/virtio/vhost-user-gpio-pci.c     |  69 ++++++
 hw/virtio/vhost-user-gpio.c         | 343 ++++++++++++++++++++++++++++
 include/hw/virtio/vhost-user-gpio.h |  35 +++
 5 files changed, 454 insertions(+)
 create mode 100644 hw/virtio/vhost-user-gpio-pci.c
 create mode 100644 hw/virtio/vhost-user-gpio.c
 create mode 100644 include/hw/virtio/vhost-user-gpio.h
Signature · 1 line
2.31.1.272.g89b43f80a514
#9 lore raw
Quoted 6 lines · Michael S. Tsirkin
Quoted 5 lines · Viresh Kumar

Hello,

This patchset adds vhost-user-gpio device's support in Qemu. The support for the same has already been added to virtio specification and Linux Kernel.

Quoted 7 lines
Quoted 3 lines

A Rust based backend is also in progress and is tested against this patchset:

https://github.com/rust-vmm/vhost-device/pull/76

I'm reluctant to add this with no tests in tree. Want to write a minimal libhost-user based backend?

This is going to be a problem going forward as we have more out-of-tree backends written as a first preference. While the first couple of vhost devices have C implementations in contrib before we worked on the rust version I think we are getting to the point of skipping a first C version for future devices.

However I notice we have qtest/vhost-user-test.c so would that be enough to ensure we can instantiate the device and a basic vhost-user initialisation sequence doesn't cause it to crap out. This obviously won't be exercising the virtq processing itself but does that really exercise any of QEMU's boilerplate anyway?

Quoted 1 line

We also need some maintainers to step up.

You mean more reviewers for the vhost and virtio sections of QEMU's maintainers?

Quoted 21 lines
Quoted 21 lines

--
Viresh

Viresh Kumar (2):
hw/virtio: add boilerplate for vhost-user-gpio device hw/virtio: add vhost-user-gpio-pci boilerplate

 hw/virtio/Kconfig                   |   5 +
 hw/virtio/meson.build               |   2 +
 hw/virtio/vhost-user-gpio-pci.c     |  69 ++++++
 hw/virtio/vhost-user-gpio.c         | 343 ++++++++++++++++++++++++++++
 include/hw/virtio/vhost-user-gpio.h |  35 +++
 5 files changed, 454 insertions(+)
 create mode 100644 hw/virtio/vhost-user-gpio-pci.c
 create mode 100644 hw/virtio/vhost-user-gpio.c
 create mode 100644 include/hw/virtio/vhost-user-gpio.h
Signature · 1 line
2.31.1.272.g89b43f80a514
Signature · 1 line
Alex Bennée
#10 lore raw
Quoted 9 lines · Alex Bennée
Quoted 6 lines · Michael S. Tsirkin
Quoted 5 lines · Viresh Kumar

Hello,

This patchset adds vhost-user-gpio device's support in Qemu. The support for the same has already been added to virtio specification and Linux Kernel.

Quoted 7 lines
Quoted 7 lines
Quoted 3 lines

A Rust based backend is also in progress and is tested against this patchset:

https://github.com/rust-vmm/vhost-device/pull/76

I'm reluctant to add this with no tests in tree. Want to write a minimal libhost-user based backend?

I actually have one already, that I wrote before attempting the Rust counterpart, but never upstreamed as I am not sure if anyone is ever going to use it, as I am not. And I thought what's the point of merging code which I will never end up using.

I am not sure what test I can add here to make sure this doesn't breaks in future though.

Quoted 12 lines

This is going to be a problem going forward as we have more out-of-tree backends written as a first preference. While the first couple of vhost devices have C implementations in contrib before we worked on the rust version I think we are getting to the point of skipping a first C version for future devices.

However I notice we have qtest/vhost-user-test.c so would that be enough to ensure we can instantiate the device and a basic vhost-user initialisation sequence doesn't cause it to crap out. This obviously won't be exercising the virtq processing itself but does that really exercise any of QEMU's boilerplate anyway?

Quoted 4 lines
Quoted 1 line

We also need some maintainers to step up.

You mean more reviewers for the vhost and virtio sections of QEMU's maintainers?

And I too was waiting for replies on these. I can surely write something up if you guys feel there is a need. I just want to understand it better.

Signature · 1 line
viresh
#11 lore raw
Quoted 18 lines · Viresh Kumar
Quoted 9 lines · Alex Bennée
Quoted 6 lines · Michael S. Tsirkin
Quoted 5 lines · Viresh Kumar

Hello,

This patchset adds vhost-user-gpio device's support in Qemu. The support for the same has already been added to virtio specification and Linux Kernel.

Quoted 7 lines
Quoted 7 lines
Quoted 3 lines

A Rust based backend is also in progress and is tested against this patchset:

https://github.com/rust-vmm/vhost-device/pull/76

I'm reluctant to add this with no tests in tree. Want to write a minimal libhost-user based backend?

Quoted 7 lines

I actually have one already, that I wrote before attempting the Rust counterpart, but never upstreamed as I am not sure if anyone is ever going to use it, as I am not. And I thought what's the point of merging code which I will never end up using.

I am not sure what test I can add here to make sure this doesn't breaks in future though.

something that executes with make check.

Quoted 23 lines
Quoted 12 lines

This is going to be a problem going forward as we have more out-of-tree backends written as a first preference. While the first couple of vhost devices have C implementations in contrib before we worked on the rust version I think we are getting to the point of skipping a first C version for future devices.

However I notice we have qtest/vhost-user-test.c so would that be enough to ensure we can instantiate the device and a basic vhost-user initialisation sequence doesn't cause it to crap out. This obviously won't be exercising the virtq processing itself but does that really exercise any of QEMU's boilerplate anyway?

Quoted 4 lines
Quoted 1 line

We also need some maintainers to step up.

You mean more reviewers for the vhost and virtio sections of QEMU's maintainers?

And I too was waiting for replies on these. I can surely write something up if you guys feel there is a need. I just want to understand it better.

Signature · 1 line
viresh
#12 lore raw
Quoted 24 lines · Michael S. Tsirkin
Quoted 23 lines · Viresh Kumar
Quoted 9 lines · Alex Bennée
Quoted 6 lines · Michael S. Tsirkin
Quoted 5 lines · Viresh Kumar

Hello,

This patchset adds vhost-user-gpio device's support in Qemu. The support for the same has already been added to virtio specification and Linux Kernel.

Quoted 7 lines
Quoted 7 lines
Quoted 3 lines

A Rust based backend is also in progress and is tested against this patchset:

https://github.com/rust-vmm/vhost-device/pull/76

I'm reluctant to add this with no tests in tree. Want to write a minimal libhost-user based backend?

I actually have one already, that I wrote before attempting the Rust counterpart, but never upstreamed as I am not sure if anyone is ever going to use it, as I am not. And I thought what's the point of merging code which I will never end up using.

Quoted 4 lines
Quoted 2 lines

I am not sure what test I can add here to make sure this doesn't breaks in future though.

something that executes with make check.

With Eric's latest series ([email protected]) we are already exercising most of the PCI code:

➜ gcov ./libqemu-aarch64-softmmu.fa.p/hw_virtio_vhost-user-gpio.c.gcda File '../../hw/virtio/vhost-user-gpio.c'
Lines executed:10.43% of 163
Creating 'vhost-user-gpio.c.gcov'

File '/home/alex/lsrc/qemu.git/include/qemu/bitops.h' Lines executed:100.00% of 4
Creating 'bitops.h.gcov'

File '/home/alex/lsrc/qemu.git/include/hw/virtio/virtio.h' Lines executed:12.50% of 8
Creating 'virtio.h.gcov'

File '/home/alex/lsrc/qemu.git/include/hw/virtio/vhost-user-gpio.h' Lines executed:0.00% of 1
Creating 'vhost-user-gpio.h.gcov'

File '/home/alex/lsrc/qemu.git/include/hw/virtio/virtio-bus.h' Lines executed:0.00% of 1
Creating 'virtio-bus.h.gcov'

File '/home/alex/lsrc/qemu.git/include/hw/qdev-core.h' Lines executed:50.00% of 2
Creating 'qdev-core.h.gcov'

🕙15:40:35 alex@zen:qemu.git/builds/gprof on  review/virtio-gpio-boilerplate [$?] ➜ gcov ./libqemu-aarch64-softmmu.fa.p/hw_virtio_vhost-user-gpio-pci.c.gcda File '../../hw/virtio/vhost-user-gpio-pci.c'
Lines executed:76.92% of 26
Creating 'vhost-user-gpio-pci.c.gcov'

File '/home/alex/lsrc/qemu.git/include/qemu/bitops.h' Lines executed:100.00% of 4
Creating 'bitops.h.gcov'

File '../../hw/virtio/virtio-pci.h'
Lines executed:100.00% of 1
Creating 'virtio-pci.h.gcov'

File '/home/alex/lsrc/qemu.git/include/hw/pci/pci.h' Lines executed:100.00% of 1
Creating 'pci.h.gcov'

File '/home/alex/lsrc/qemu.git/include/hw/qdev-core.h' Lines executed:50.00% of 2
Creating 'qdev-core.h.gcov'

So I think if we add the runes to ensure we instantiate both types explicitly I think we will have enough coverage.

Signature · 1 line
Alex Bennée
#13 lore raw

RFC PATCH 1/1 tests/qtest: attempt to enable tests for virtio-gpio (!working)

In reply to Viresh Kumar

We don't have a virtio-gpio implementation in QEMU and only support a vhost-user backend. The QEMU side of the code is minimal so it should be enough to instantiate the device and pass some vhost-user messages over the control socket. To do this we hook into the existing vhost-user-test code and just add the bits required for gpio.

Based-on: [email protected]

Cc Viresh Kumar <[email protected]>
Cc Paolo Bonzini <[email protected]>

---

This goes as far as to add things to the QOS tree but so far it's failing to properly start QEMU with the chardev socket needed to communicate between the mock vhost-user daemon and QEMU itself.

---
 tests/qtest/libqos/virtio-gpio.h | 34 +++++++++++
 tests/qtest/libqos/virtio-gpio.c | 98 ++++++++++++++++++++++++++++++++
 tests/qtest/vhost-user-test.c    | 34 +++++++++++
 tests/qtest/libqos/meson.build   |  1 +
 4 files changed, 167 insertions(+)
 create mode 100644 tests/qtest/libqos/virtio-gpio.h
 create mode 100644 tests/qtest/libqos/virtio-gpio.c
tests/qtest/libqos/virtio-gpio.h new file +34 -0
new file mode 100644
index 0000000000..abe6967ae9
@@ -0,0 +1,34 @@
+/*
+ * virtio-gpio structures
+ *
+ * Copyright (c) 2022 Linaro Ltd
+ *
+ * SPDX-License-Identifier: GPL-2.0-or-later
+ */
+
+#ifndef TESTS_LIBQOS_VIRTIO_GPIO_H
+#define TESTS_LIBQOS_VIRTIO_GPIO_H
+
+#include "qgraph.h"
+#include "virtio.h"
+#include "virtio-pci.h"
+
+typedef struct QVhostUserGPIO QVhostUserGPIO;
+typedef struct QVhostUserGPIOPCI QVhostUserGPIOPCI;
+typedef struct QVhostUserGPIODevice QVhostUserGPIODevice;
+
+struct QVhostUserGPIO {
+    QVirtioDevice *vdev;
+};
+
+struct QVhostUserGPIOPCI {
+    QVirtioPCIDevice pci_vdev;
+    QVhostUserGPIO gpio;
+};
+
+struct QVhostUserGPIODevice {
+    QOSGraphObject obj;
+    QVhostUserGPIO gpio;
+};
+
+#endif
tests/qtest/libqos/virtio-gpio.c new file +98 -0

98 lines not on this page · raw message

tests/qtest/vhost-user-test.c +34 -0
index ee30f54796..4f8789424b 100644
@@ -43,6 +43,7 @@
                         " -numa node,memdev=mem"
 #define QEMU_CMD_CHR    " -chardev socket,id=%s,path=%s%s"
 #define QEMU_CMD_NETDEV " -netdev vhost-user,id=hs0,chardev=%s,vhostforce=on"
+#define QEMU_CMD_GPIO   " -device vhost-user-gpio-pci,id=gpio0,chardev=%s,vhostforce=on"
 
 #define HUGETLBFS_MAGIC       0x958458f6
 
@@ -137,6 +138,7 @@ enum {
 
 enum {
     VHOST_USER_NET,
+    VHOST_USER_GPIO,
 };
 
 typedef struct TestServer {
@@ -194,6 +196,14 @@ static void append_vhost_net_opts(TestServer *s, GString *cmd_line,
                            chr_opts, s->chr_name);
 }
 
+static void append_vhost_gpio_opts(TestServer *s, GString *cmd_line,
+                             const char *chr_opts)
+{
+    g_string_append_printf(cmd_line, QEMU_CMD_CHR QEMU_CMD_GPIO,
+                           s->chr_name, s->socket_path,
+                           chr_opts, s->chr_name);
+}
+
 static void append_mem_opts(TestServer *server, GString *cmd_line,
                             int size, enum test_memfd memfd)
 {
@@ -1014,3 +1024,27 @@ static void register_vhost_user_test(void)
                  test_multiqueue, &opts);
 }
 libqos_init(register_vhost_user_test);
+
+static struct vhost_user_ops g_vu_gpio_ops = {
+    .type = VHOST_USER_GPIO,
+
+    .append_opts = append_vhost_gpio_opts,
+
+    .set_features = vu_net_set_features,
+    .get_protocol_features = vu_net_get_protocol_features,
+};
+
+static void register_vhost_gpio_test(void)
+{
+    QOSGraphTestOptions opts = {
+        .before = vhost_user_test_setup,
+        .subprocess = true,
+        .arg = &g_vu_gpio_ops,
+    };
+
+    qemu_add_opts(&qemu_chardev_opts);
+
+    qos_add_test("read-guest-mem/memfile",
+                 "vhost-user-gpio", test_read_guest_mem, &opts);
+}
+libqos_init(register_vhost_gpio_test);
tests/qtest/libqos/meson.build +1 -0
index 8c8ee15553..f581330d8b 100644
@@ -42,6 +42,7 @@ libqos_srcs = files('../libqtest.c',
         'virtio-scsi.c',
         'virtio-serial.c',
         'virtio-iommu.c',
+        'virtio-gpio.c',
         'generic-pcihost.c',
 
         # qgraph machines:
Signature · 1 line
2.30.2
#14 lore raw

Re: [RFC PATCH] tests/qtest: attempt to enable tests for virtio-gpio (!working)

(expanding the CC list for help, anyone have a better idea about how vhost-user qtests should work/see obvious issues with this patch?)

Quoted 244 lines · Alex Bennée

We don't have a virtio-gpio implementation in QEMU and only support a vhost-user backend. The QEMU side of the code is minimal so it should be enough to instantiate the device and pass some vhost-user messages over the control socket. To do this we hook into the existing vhost-user-test code and just add the bits required for gpio.

Based-on: [email protected]

Cc Viresh Kumar <[email protected]>
Cc Paolo Bonzini <[email protected]>

---

This goes as far as to add things to the QOS tree but so far it's failing to properly start QEMU with the chardev socket needed to communicate between the mock vhost-user daemon and QEMU itself.

---
 tests/qtest/libqos/virtio-gpio.h | 34 +++++++++++
 tests/qtest/libqos/virtio-gpio.c | 98 ++++++++++++++++++++++++++++++++
 tests/qtest/vhost-user-test.c    | 34 +++++++++++
 tests/qtest/libqos/meson.build   |  1 +
 4 files changed, 167 insertions(+)
 create mode 100644 tests/qtest/libqos/virtio-gpio.h
 create mode 100644 tests/qtest/libqos/virtio-gpio.c
tests/qtest/libqos/virtio-gpio.h new file +34 -0
new file mode 100644
index 0000000000..abe6967ae9
@@ -0,0 +1,34 @@
+/*
+ * virtio-gpio structures
+ *
+ * Copyright (c) 2022 Linaro Ltd
+ *
+ * SPDX-License-Identifier: GPL-2.0-or-later
+ */
+
+#ifndef TESTS_LIBQOS_VIRTIO_GPIO_H
+#define TESTS_LIBQOS_VIRTIO_GPIO_H
+
+#include "qgraph.h"
+#include "virtio.h"
+#include "virtio-pci.h"
+
+typedef struct QVhostUserGPIO QVhostUserGPIO;
+typedef struct QVhostUserGPIOPCI QVhostUserGPIOPCI;
+typedef struct QVhostUserGPIODevice QVhostUserGPIODevice;
+
+struct QVhostUserGPIO {
+    QVirtioDevice *vdev;
+};
+
+struct QVhostUserGPIOPCI {
+    QVirtioPCIDevice pci_vdev;
+    QVhostUserGPIO gpio;
+};
+
+struct QVhostUserGPIODevice {
+    QOSGraphObject obj;
+    QVhostUserGPIO gpio;
+};
+
+#endif
tests/qtest/libqos/virtio-gpio.c new file +98 -0
new file mode 100644
index 0000000000..62c8074777
@@ -0,0 +1,98 @@
+/*
+ * virtio-gpio nodes for testing
+ *
+ * Copyright (c) 2022 Linaro Ltd
+ *
+ * SPDX-License-Identifier: GPL-2.0-or-later
+ */
+
+#include "qemu/osdep.h"
+#include "libqtest.h"
+#include "qemu/module.h"
+#include "qgraph.h"
+#include "virtio-gpio.h"
+
+static void *qvirtio_gpio_get_driver(QVhostUserGPIO *v_gpio,
+                                       const char *interface)
+{
+    if (!g_strcmp0(interface, "vhost-user-gpio")) {
+        return v_gpio;
+    }
+    if (!g_strcmp0(interface, "virtio")) {
+        return v_gpio->vdev;
+    }
+
+    fprintf(stderr, "%s not present in virtio-gpio-device\n", interface);
+    g_assert_not_reached();
+}
+
+static void *qvirtio_gpio_device_get_driver(void *object,
+                                              const char *interface)
+{
+    QVhostUserGPIODevice *v_gpio = object;
+    return qvirtio_gpio_get_driver(&v_gpio->gpio, interface);
+}
+
+static void *virtio_gpio_device_create(void *virtio_dev,
+                                         QGuestAllocator *t_alloc,
+                                         void *addr)
+{
+    QVhostUserGPIODevice *virtio_device = g_new0(QVhostUserGPIODevice, 1);
+    QVhostUserGPIO *interface = &virtio_device->gpio;
+
+    interface->vdev = virtio_dev;
+
+    virtio_device->obj.get_driver = qvirtio_gpio_device_get_driver;
+
+    return &virtio_device->obj;
+}
+
+/* virtio-gpio-pci */
+static void *qvirtio_gpio_pci_get_driver(void *object, const char *interface)
+{
+    QVhostUserGPIOPCI *v_gpio = object;
+    if (!g_strcmp0(interface, "pci-device")) {
+        return v_gpio->pci_vdev.pdev;
+    }
+    return qvirtio_gpio_get_driver(&v_gpio->gpio, interface);
+}
+
+static void *virtio_gpio_pci_create(void *pci_bus, QGuestAllocator *t_alloc,
+                                      void *addr)
+{
+    QVhostUserGPIOPCI *virtio_spci = g_new0(QVhostUserGPIOPCI, 1);
+    QVhostUserGPIO *interface = &virtio_spci->gpio;
+    QOSGraphObject *obj = &virtio_spci->pci_vdev.obj;
+
+    virtio_pci_init(&virtio_spci->pci_vdev, pci_bus, addr);
+    interface->vdev = &virtio_spci->pci_vdev.vdev;
+
+    obj->get_driver = qvirtio_gpio_pci_get_driver;
+
+    return obj;
+}
+
+static void virtio_gpio_register_nodes(void)
+{
+    QPCIAddress addr = {
+        .devfn = QPCI_DEVFN(4, 0),
+    };
+
+    QOSGraphEdgeOptions edge_opts = { };
+
+    /* vhost-user-gpio-device */
+    edge_opts.extra_device_opts = "id=gpio0,chardev=vhgpio0";
+    qos_node_create_driver("vhost-user-gpio-device",
+                            virtio_gpio_device_create);
+    qos_node_consumes("vhost-user-gpio-device", "virtio-bus", &edge_opts);
+    qos_node_produces("vhost-user-gpio-device", "vhost-user-gpio");
+
+    /* virtio-gpio-pci */
+    edge_opts.extra_device_opts = "id=gpio0,addr=04.0,chardev=vhgpio0";
+    add_qpci_address(&edge_opts, &addr);
+    qos_node_create_driver("vhost-user-gpio-pci", virtio_gpio_pci_create);
+    qos_node_consumes("vhost-user-gpio-pci", "pci-bus", &edge_opts);
+    qos_node_produces("vhost-user-gpio-pci", "vhost-user-gpio");
+}
+
+libqos_init(virtio_gpio_register_nodes);
tests/qtest/vhost-user-test.c +34 -0
index ee30f54796..4f8789424b 100644
@@ -43,6 +43,7 @@
                         " -numa node,memdev=mem"
 #define QEMU_CMD_CHR    " -chardev socket,id=%s,path=%s%s"
 #define QEMU_CMD_NETDEV " -netdev vhost-user,id=hs0,chardev=%s,vhostforce=on"
+#define QEMU_CMD_GPIO   " -device vhost-user-gpio-pci,id=gpio0,chardev=%s,vhostforce=on"
 
 #define HUGETLBFS_MAGIC       0x958458f6
 
@@ -137,6 +138,7 @@ enum {
 
 enum {
     VHOST_USER_NET,
+    VHOST_USER_GPIO,
 };
 
 typedef struct TestServer {
@@ -194,6 +196,14 @@ static void append_vhost_net_opts(TestServer *s, GString *cmd_line,
                            chr_opts, s->chr_name);
 }
 
+static void append_vhost_gpio_opts(TestServer *s, GString *cmd_line,
+                             const char *chr_opts)
+{
+    g_string_append_printf(cmd_line, QEMU_CMD_CHR QEMU_CMD_GPIO,
+                           s->chr_name, s->socket_path,
+                           chr_opts, s->chr_name);
+}
+
 static void append_mem_opts(TestServer *server, GString *cmd_line,
                             int size, enum test_memfd memfd)
 {
@@ -1014,3 +1024,27 @@ static void register_vhost_user_test(void)
                  test_multiqueue, &opts);
 }
 libqos_init(register_vhost_user_test);
+
+static struct vhost_user_ops g_vu_gpio_ops = {
+    .type = VHOST_USER_GPIO,
+
+    .append_opts = append_vhost_gpio_opts,
+
+    .set_features = vu_net_set_features,
+    .get_protocol_features = vu_net_get_protocol_features,
+};
+
+static void register_vhost_gpio_test(void)
+{
+    QOSGraphTestOptions opts = {
+        .before = vhost_user_test_setup,
+        .subprocess = true,
+        .arg = &g_vu_gpio_ops,
+    };
+
+    qemu_add_opts(&qemu_chardev_opts);
+
+    qos_add_test("read-guest-mem/memfile",
+                 "vhost-user-gpio", test_read_guest_mem, &opts);
+}
+libqos_init(register_vhost_gpio_test);
tests/qtest/libqos/meson.build +1 -0
index 8c8ee15553..f581330d8b 100644
@@ -42,6 +42,7 @@ libqos_srcs = files('../libqtest.c',
         'virtio-scsi.c',
         'virtio-serial.c',
         'virtio-iommu.c',
+        'virtio-gpio.c',
         'generic-pcihost.c',
 
         # qgraph machines:
Signature · 1 line
Alex Bennée
#15 lore raw

Re: [RFC PATCH] tests/qtest: attempt to enable tests for virtio-gpio (!working)

Quoted 3 lines · Alex Bennée

(expanding the CC list for help, anyone have a better idea about how vhost-user qtests should work/see obvious issues with this patch?)

This is better than no test :).

Stefan

#16 lore raw

Re: [RFC PATCH] tests/qtest: attempt to enable tests for virtio-gpio (!working)

In reply to Alex Bennée
Quoted 3 lines · * Alex Bennée ([email protected]) wrote:

(expanding the CC list for help, anyone have a better idea about how vhost-user qtests should work/see obvious issues with this patch?)

How exactly does it fail?

DAve

Quoted 251 lines
Quoted 244 lines · Alex Bennée

We don't have a virtio-gpio implementation in QEMU and only support a vhost-user backend. The QEMU side of the code is minimal so it should be enough to instantiate the device and pass some vhost-user messages over the control socket. To do this we hook into the existing vhost-user-test code and just add the bits required for gpio.

Based-on: [email protected]

Cc Viresh Kumar <[email protected]>
Cc Paolo Bonzini <[email protected]>

---

This goes as far as to add things to the QOS tree but so far it's failing to properly start QEMU with the chardev socket needed to communicate between the mock vhost-user daemon and QEMU itself.

---
 tests/qtest/libqos/virtio-gpio.h | 34 +++++++++++
 tests/qtest/libqos/virtio-gpio.c | 98 ++++++++++++++++++++++++++++++++
 tests/qtest/vhost-user-test.c    | 34 +++++++++++
 tests/qtest/libqos/meson.build   |  1 +
 4 files changed, 167 insertions(+)
 create mode 100644 tests/qtest/libqos/virtio-gpio.h
 create mode 100644 tests/qtest/libqos/virtio-gpio.c
tests/qtest/libqos/virtio-gpio.h new file +34 -0
new file mode 100644
index 0000000000..abe6967ae9
@@ -0,0 +1,34 @@
+/*
+ * virtio-gpio structures
+ *
+ * Copyright (c) 2022 Linaro Ltd
+ *
+ * SPDX-License-Identifier: GPL-2.0-or-later
+ */
+
+#ifndef TESTS_LIBQOS_VIRTIO_GPIO_H
+#define TESTS_LIBQOS_VIRTIO_GPIO_H
+
+#include "qgraph.h"
+#include "virtio.h"
+#include "virtio-pci.h"
+
+typedef struct QVhostUserGPIO QVhostUserGPIO;
+typedef struct QVhostUserGPIOPCI QVhostUserGPIOPCI;
+typedef struct QVhostUserGPIODevice QVhostUserGPIODevice;
+
+struct QVhostUserGPIO {
+    QVirtioDevice *vdev;
+};
+
+struct QVhostUserGPIOPCI {
+    QVirtioPCIDevice pci_vdev;
+    QVhostUserGPIO gpio;
+};
+
+struct QVhostUserGPIODevice {
+    QOSGraphObject obj;
+    QVhostUserGPIO gpio;
+};
+
+#endif
tests/qtest/libqos/virtio-gpio.c new file +98 -0
new file mode 100644
index 0000000000..62c8074777
@@ -0,0 +1,98 @@
+/*
+ * virtio-gpio nodes for testing
+ *
+ * Copyright (c) 2022 Linaro Ltd
+ *
+ * SPDX-License-Identifier: GPL-2.0-or-later
+ */
+
+#include "qemu/osdep.h"
+#include "libqtest.h"
+#include "qemu/module.h"
+#include "qgraph.h"
+#include "virtio-gpio.h"
+
+static void *qvirtio_gpio_get_driver(QVhostUserGPIO *v_gpio,
+                                       const char *interface)
+{
+    if (!g_strcmp0(interface, "vhost-user-gpio")) {
+        return v_gpio;
+    }
+    if (!g_strcmp0(interface, "virtio")) {
+        return v_gpio->vdev;
+    }
+
+    fprintf(stderr, "%s not present in virtio-gpio-device\n", interface);
+    g_assert_not_reached();
+}
+
+static void *qvirtio_gpio_device_get_driver(void *object,
+                                              const char *interface)
+{
+    QVhostUserGPIODevice *v_gpio = object;
+    return qvirtio_gpio_get_driver(&v_gpio->gpio, interface);
+}
+
+static void *virtio_gpio_device_create(void *virtio_dev,
+                                         QGuestAllocator *t_alloc,
+                                         void *addr)
+{
+    QVhostUserGPIODevice *virtio_device = g_new0(QVhostUserGPIODevice, 1);
+    QVhostUserGPIO *interface = &virtio_device->gpio;
+
+    interface->vdev = virtio_dev;
+
+    virtio_device->obj.get_driver = qvirtio_gpio_device_get_driver;
+
+    return &virtio_device->obj;
+}
+
+/* virtio-gpio-pci */
+static void *qvirtio_gpio_pci_get_driver(void *object, const char *interface)
+{
+    QVhostUserGPIOPCI *v_gpio = object;
+    if (!g_strcmp0(interface, "pci-device")) {
+        return v_gpio->pci_vdev.pdev;
+    }
+    return qvirtio_gpio_get_driver(&v_gpio->gpio, interface);
+}
+
+static void *virtio_gpio_pci_create(void *pci_bus, QGuestAllocator *t_alloc,
+                                      void *addr)
+{
+    QVhostUserGPIOPCI *virtio_spci = g_new0(QVhostUserGPIOPCI, 1);
+    QVhostUserGPIO *interface = &virtio_spci->gpio;
+    QOSGraphObject *obj = &virtio_spci->pci_vdev.obj;
+
+    virtio_pci_init(&virtio_spci->pci_vdev, pci_bus, addr);
+    interface->vdev = &virtio_spci->pci_vdev.vdev;
+
+    obj->get_driver = qvirtio_gpio_pci_get_driver;
+
+    return obj;
+}
+
+static void virtio_gpio_register_nodes(void)
+{
+    QPCIAddress addr = {
+        .devfn = QPCI_DEVFN(4, 0),
+    };
+
+    QOSGraphEdgeOptions edge_opts = { };
+
+    /* vhost-user-gpio-device */
+    edge_opts.extra_device_opts = "id=gpio0,chardev=vhgpio0";
+    qos_node_create_driver("vhost-user-gpio-device",
+                            virtio_gpio_device_create);
+    qos_node_consumes("vhost-user-gpio-device", "virtio-bus", &edge_opts);
+    qos_node_produces("vhost-user-gpio-device", "vhost-user-gpio");
+
+    /* virtio-gpio-pci */
+    edge_opts.extra_device_opts = "id=gpio0,addr=04.0,chardev=vhgpio0";
+    add_qpci_address(&edge_opts, &addr);
+    qos_node_create_driver("vhost-user-gpio-pci", virtio_gpio_pci_create);
+    qos_node_consumes("vhost-user-gpio-pci", "pci-bus", &edge_opts);
+    qos_node_produces("vhost-user-gpio-pci", "vhost-user-gpio");
+}
+
+libqos_init(virtio_gpio_register_nodes);
tests/qtest/vhost-user-test.c +34 -0
index ee30f54796..4f8789424b 100644
@@ -43,6 +43,7 @@
                         " -numa node,memdev=mem"
 #define QEMU_CMD_CHR    " -chardev socket,id=%s,path=%s%s"
 #define QEMU_CMD_NETDEV " -netdev vhost-user,id=hs0,chardev=%s,vhostforce=on"
+#define QEMU_CMD_GPIO   " -device vhost-user-gpio-pci,id=gpio0,chardev=%s,vhostforce=on"
 
 #define HUGETLBFS_MAGIC       0x958458f6
 
@@ -137,6 +138,7 @@ enum {
 
 enum {
     VHOST_USER_NET,
+    VHOST_USER_GPIO,
 };
 
 typedef struct TestServer {
@@ -194,6 +196,14 @@ static void append_vhost_net_opts(TestServer *s, GString *cmd_line,
                            chr_opts, s->chr_name);
 }
 
+static void append_vhost_gpio_opts(TestServer *s, GString *cmd_line,
+                             const char *chr_opts)
+{
+    g_string_append_printf(cmd_line, QEMU_CMD_CHR QEMU_CMD_GPIO,
+                           s->chr_name, s->socket_path,
+                           chr_opts, s->chr_name);
+}
+
 static void append_mem_opts(TestServer *server, GString *cmd_line,
                             int size, enum test_memfd memfd)
 {
@@ -1014,3 +1024,27 @@ static void register_vhost_user_test(void)
                  test_multiqueue, &opts);
 }
 libqos_init(register_vhost_user_test);
+
+static struct vhost_user_ops g_vu_gpio_ops = {
+    .type = VHOST_USER_GPIO,
+
+    .append_opts = append_vhost_gpio_opts,
+
+    .set_features = vu_net_set_features,
+    .get_protocol_features = vu_net_get_protocol_features,
+};
+
+static void register_vhost_gpio_test(void)
+{
+    QOSGraphTestOptions opts = {
+        .before = vhost_user_test_setup,
+        .subprocess = true,
+        .arg = &g_vu_gpio_ops,
+    };
+
+    qemu_add_opts(&qemu_chardev_opts);
+
+    qos_add_test("read-guest-mem/memfile",
+                 "vhost-user-gpio", test_read_guest_mem, &opts);
+}
+libqos_init(register_vhost_gpio_test);
tests/qtest/libqos/meson.build +1 -0
index 8c8ee15553..f581330d8b 100644
@@ -42,6 +42,7 @@ libqos_srcs = files('../libqtest.c',
         'virtio-scsi.c',
         'virtio-serial.c',
         'virtio-iommu.c',
+        'virtio-gpio.c',
         'generic-pcihost.c',
 
         # qgraph machines:
Signature · 1 line
Alex Bennée
Signature · 1 line
Dr. David Alan Gilbert / [email protected] / Manchester, UK
#17 lore raw

Re: [RFC PATCH] tests/qtest: attempt to enable tests for virtio-gpio (!working)

Quoted 6 lines · Dr. David Alan Gilbert
Quoted 3 lines · * Alex Bennée ([email protected]) wrote:

(expanding the CC list for help, anyone have a better idea about how vhost-user qtests should work/see obvious issues with this patch?)

How exactly does it fail?

➜ env QTEST_QEMU_BINARY=./qemu-system-aarch64 QTEST_QEMU_STORAGE_DAEMON_BINARY=./storage-daemon/qemu-storage-daemon G_TEST_DBUS_DAEMON=/home/alex/lsrc/qemu.git/tests/dbus-v mstate-daemon.sh QTEST_QEMU_IMG=./qemu-img MALLOC_PERTURB_=137 ./tests/qtest/qos-test -p /aarch64/virt/generic-pcihost/pci-bus-generic/pci-bus/vhost-user-gpio-pci/vhost-user-gpio/vhost-user-gpio-tests/read-guest-mem/memfile

# random seed: R02S5d7667675b4f6dd3b8559f8db621296c
# starting QEMU: exec ./qemu-system-aarch64 -qtest unix:/tmp/qtest-1245871.sock -qtest-log /dev/null -chardev socket,path=/tmp/qtest-1245871.qmp,id=char0 -mon chardev=char0,mode=control -display none -machine none -accel qtest
# Start of aarch64 tests
# Start of virt tests
# Start of generic-pcihost tests
# Start of pci-bus-generic tests
# Start of pci-bus tests
# Start of vhost-user-gpio-pci tests
# Start of vhost-user-gpio tests
# Start of vhost-user-gpio-tests tests
# Start of read-guest-mem tests
# child process (/aarch64/virt/generic-pcihost/pci-bus-generic/pci-bus/vhost-user-gpio-pci/vhost-user-gpio/vhost-user-gpio-tests/read-guest-mem/memfile/subprocess [1245877])
 exit status: 1 (error)
# child process (/aarch64/virt/generic-pcihost/pci-bus-generic/pci-bus/vhost-user-gpio-pci/vhost-user-gpio/vhost-user-gpio-tests/read-guest-mem/memfile/subprocess [1245877]) stdout: ""
# child process (/aarch64/virt/generic-pcihost/pci-bus-generic/pci-bus/vhost-user-gpio-pci/vhost-user-gpio/vhost-user-gpio-tests/read-guest-mem/memfile/subprocess [1245877]) stderr: "qemu-system-aarch64: -device vhost-user-gpio-pci,id=gpio0,chardev=chr-vhost-user-test,vhostforce=on: Duplicate ID 'gpio0' for device\nsocket_accept failed: Resource temporarily unavailable\n**\nERROR:../../tests/qtest/libqtest.c:321:qtest_init_without_qmp_handshake: assertion failed: (s->fd >= 0 && s->qmp_fd >= 0)\n"
**
ERROR:../../tests/qtest/qos-test.c:189:subprocess_run_one_test: child process (/aarch64/virt/generic-pcihost/pci-bus-generic/pci-bus/vhost-user-gpio-pci/vhost-user-gpio/vhost-user-gpio-tests/read-guest-mem/memfile/subprocess [1245877]) failed unexpectedly
Bail out! ERROR:../../tests/qtest/qos-test.c:189:subprocess_run_one_test: child process (/aarch64/virt/generic-pcihost/pci-bus-generic/pci-bus/vhost-user-gpio-pci/vhost-user-gpio/vhost-user-gpio-tests/read-guest-mem/memfile/subprocess [1245877]) failed unexpectedly
fish: “env QTEST_QEMU_BINARY=./qemu-sy…” terminated by signal SIGABRT
(Abort)

Although it would be nice if I could individually run qos-tests with all the make machinery setting things up.

Quoted 254 lines

DAve

Quoted 251 lines
Quoted 244 lines · Alex Bennée

We don't have a virtio-gpio implementation in QEMU and only support a vhost-user backend. The QEMU side of the code is minimal so it should be enough to instantiate the device and pass some vhost-user messages over the control socket. To do this we hook into the existing vhost-user-test code and just add the bits required for gpio.

Based-on: [email protected]

Cc Viresh Kumar <[email protected]>
Cc Paolo Bonzini <[email protected]>

---

This goes as far as to add things to the QOS tree but so far it's failing to properly start QEMU with the chardev socket needed to communicate between the mock vhost-user daemon and QEMU itself.

---
 tests/qtest/libqos/virtio-gpio.h | 34 +++++++++++
 tests/qtest/libqos/virtio-gpio.c | 98 ++++++++++++++++++++++++++++++++
 tests/qtest/vhost-user-test.c    | 34 +++++++++++
 tests/qtest/libqos/meson.build   |  1 +
 4 files changed, 167 insertions(+)
 create mode 100644 tests/qtest/libqos/virtio-gpio.h
 create mode 100644 tests/qtest/libqos/virtio-gpio.c
tests/qtest/libqos/virtio-gpio.h new file +34 -0
new file mode 100644
index 0000000000..abe6967ae9
@@ -0,0 +1,34 @@
+/*
+ * virtio-gpio structures
+ *
+ * Copyright (c) 2022 Linaro Ltd
+ *
+ * SPDX-License-Identifier: GPL-2.0-or-later
+ */
+
+#ifndef TESTS_LIBQOS_VIRTIO_GPIO_H
+#define TESTS_LIBQOS_VIRTIO_GPIO_H
+
+#include "qgraph.h"
+#include "virtio.h"
+#include "virtio-pci.h"
+
+typedef struct QVhostUserGPIO QVhostUserGPIO;
+typedef struct QVhostUserGPIOPCI QVhostUserGPIOPCI;
+typedef struct QVhostUserGPIODevice QVhostUserGPIODevice;
+
+struct QVhostUserGPIO {
+    QVirtioDevice *vdev;
+};
+
+struct QVhostUserGPIOPCI {
+    QVirtioPCIDevice pci_vdev;
+    QVhostUserGPIO gpio;
+};
+
+struct QVhostUserGPIODevice {
+    QOSGraphObject obj;
+    QVhostUserGPIO gpio;
+};
+
+#endif
tests/qtest/libqos/virtio-gpio.c new file +98 -0
new file mode 100644
index 0000000000..62c8074777
@@ -0,0 +1,98 @@
+/*
+ * virtio-gpio nodes for testing
+ *
+ * Copyright (c) 2022 Linaro Ltd
+ *
+ * SPDX-License-Identifier: GPL-2.0-or-later
+ */
+
+#include "qemu/osdep.h"
+#include "libqtest.h"
+#include "qemu/module.h"
+#include "qgraph.h"
+#include "virtio-gpio.h"
+
+static void *qvirtio_gpio_get_driver(QVhostUserGPIO *v_gpio,
+                                       const char *interface)
+{
+    if (!g_strcmp0(interface, "vhost-user-gpio")) {
+        return v_gpio;
+    }
+    if (!g_strcmp0(interface, "virtio")) {
+        return v_gpio->vdev;
+    }
+
+    fprintf(stderr, "%s not present in virtio-gpio-device\n", interface);
+    g_assert_not_reached();
+}
+
+static void *qvirtio_gpio_device_get_driver(void *object,
+                                              const char *interface)
+{
+    QVhostUserGPIODevice *v_gpio = object;
+    return qvirtio_gpio_get_driver(&v_gpio->gpio, interface);
+}
+
+static void *virtio_gpio_device_create(void *virtio_dev,
+                                         QGuestAllocator *t_alloc,
+                                         void *addr)
+{
+    QVhostUserGPIODevice *virtio_device = g_new0(QVhostUserGPIODevice, 1);
+    QVhostUserGPIO *interface = &virtio_device->gpio;
+
+    interface->vdev = virtio_dev;
+
+    virtio_device->obj.get_driver = qvirtio_gpio_device_get_driver;
+
+    return &virtio_device->obj;
+}
+
+/* virtio-gpio-pci */
+static void *qvirtio_gpio_pci_get_driver(void *object, const char *interface)
+{
+    QVhostUserGPIOPCI *v_gpio = object;
+    if (!g_strcmp0(interface, "pci-device")) {
+        return v_gpio->pci_vdev.pdev;
+    }
+    return qvirtio_gpio_get_driver(&v_gpio->gpio, interface);
+}
+
+static void *virtio_gpio_pci_create(void *pci_bus, QGuestAllocator *t_alloc,
+                                      void *addr)
+{
+    QVhostUserGPIOPCI *virtio_spci = g_new0(QVhostUserGPIOPCI, 1);
+    QVhostUserGPIO *interface = &virtio_spci->gpio;
+    QOSGraphObject *obj = &virtio_spci->pci_vdev.obj;
+
+    virtio_pci_init(&virtio_spci->pci_vdev, pci_bus, addr);
+    interface->vdev = &virtio_spci->pci_vdev.vdev;
+
+    obj->get_driver = qvirtio_gpio_pci_get_driver;
+
+    return obj;
+}
+
+static void virtio_gpio_register_nodes(void)
+{
+    QPCIAddress addr = {
+        .devfn = QPCI_DEVFN(4, 0),
+    };
+
+    QOSGraphEdgeOptions edge_opts = { };
+
+    /* vhost-user-gpio-device */
+    edge_opts.extra_device_opts = "id=gpio0,chardev=vhgpio0";
+    qos_node_create_driver("vhost-user-gpio-device",
+                            virtio_gpio_device_create);
+    qos_node_consumes("vhost-user-gpio-device", "virtio-bus", &edge_opts);
+    qos_node_produces("vhost-user-gpio-device", "vhost-user-gpio");
+
+    /* virtio-gpio-pci */
+    edge_opts.extra_device_opts = "id=gpio0,addr=04.0,chardev=vhgpio0";
+    add_qpci_address(&edge_opts, &addr);
+    qos_node_create_driver("vhost-user-gpio-pci", virtio_gpio_pci_create);
+    qos_node_consumes("vhost-user-gpio-pci", "pci-bus", &edge_opts);
+    qos_node_produces("vhost-user-gpio-pci", "vhost-user-gpio");
+}
+
+libqos_init(virtio_gpio_register_nodes);
tests/qtest/vhost-user-test.c +34 -0
index ee30f54796..4f8789424b 100644
@@ -43,6 +43,7 @@
                         " -numa node,memdev=mem"
 #define QEMU_CMD_CHR    " -chardev socket,id=%s,path=%s%s"
 #define QEMU_CMD_NETDEV " -netdev vhost-user,id=hs0,chardev=%s,vhostforce=on"
+#define QEMU_CMD_GPIO   " -device vhost-user-gpio-pci,id=gpio0,chardev=%s,vhostforce=on"
 
 #define HUGETLBFS_MAGIC       0x958458f6
 
@@ -137,6 +138,7 @@ enum {
 
 enum {
     VHOST_USER_NET,
+    VHOST_USER_GPIO,
 };
 
 typedef struct TestServer {
@@ -194,6 +196,14 @@ static void append_vhost_net_opts(TestServer *s, GString *cmd_line,
                            chr_opts, s->chr_name);
 }
 
+static void append_vhost_gpio_opts(TestServer *s, GString *cmd_line,
+                             const char *chr_opts)
+{
+    g_string_append_printf(cmd_line, QEMU_CMD_CHR QEMU_CMD_GPIO,
+                           s->chr_name, s->socket_path,
+                           chr_opts, s->chr_name);
+}
+
 static void append_mem_opts(TestServer *server, GString *cmd_line,
                             int size, enum test_memfd memfd)
 {
@@ -1014,3 +1024,27 @@ static void register_vhost_user_test(void)
                  test_multiqueue, &opts);
 }
 libqos_init(register_vhost_user_test);
+
+static struct vhost_user_ops g_vu_gpio_ops = {
+    .type = VHOST_USER_GPIO,
+
+    .append_opts = append_vhost_gpio_opts,
+
+    .set_features = vu_net_set_features,
+    .get_protocol_features = vu_net_get_protocol_features,
+};
+
+static void register_vhost_gpio_test(void)
+{
+    QOSGraphTestOptions opts = {
+        .before = vhost_user_test_setup,
+        .subprocess = true,
+        .arg = &g_vu_gpio_ops,
+    };
+
+    qemu_add_opts(&qemu_chardev_opts);
+
+    qos_add_test("read-guest-mem/memfile",
+                 "vhost-user-gpio", test_read_guest_mem, &opts);
+}
+libqos_init(register_vhost_gpio_test);
tests/qtest/libqos/meson.build +1 -0
index 8c8ee15553..f581330d8b 100644
@@ -42,6 +42,7 @@ libqos_srcs = files('../libqtest.c',
         'virtio-scsi.c',
         'virtio-serial.c',
         'virtio-iommu.c',
+        'virtio-gpio.c',
         'generic-pcihost.c',
 
         # qgraph machines:
Signature · 1 line
Alex Bennée
Signature · 1 line
Alex Bennée
#18 lore raw
In reply to Alex Bennée
Quoted 14 lines · Alex Bennée
Quoted 11 lines · Michael S. Tsirkin
Quoted 8 lines · Viresh Kumar

Hello,

This patchset adds vhost-user-gpio device's support in Qemu. The support for the same has already been added to virtio specification and Linux Kernel.

A Rust based backend is also in progress and is tested against this patchset:

https://github.com/rust-vmm/vhost-device/pull/76

Quoted 8 lines
Quoted 2 lines

I'm reluctant to add this with no tests in tree. Want to write a minimal libhost-user based backend?

This is going to be a problem going forward as we have more out-of-tree backends written as a first preference. While the first couple of vhost devices have C implementations in contrib before we worked on the rust version I think we are getting to the point of skipping a first C version for future devices.

Not sure what to do then. Integrate using submodules somehow?

Quoted 5 lines

However I notice we have qtest/vhost-user-test.c so would that be enough to ensure we can instantiate the device and a basic vhost-user initialisation sequence doesn't cause it to crap out. This obviously won't be exercising the virtq processing itself but does that really exercise any of QEMU's boilerplate anyway?

we do have tests e.g. for network based on vhost-user, yes.

Quoted 30 lines
Quoted 1 line

We also need some maintainers to step up.

You mean more reviewers for the vhost and virtio sections of QEMU's maintainers?

Quoted 21 lines
Quoted 21 lines

--
Viresh

Viresh Kumar (2):
hw/virtio: add boilerplate for vhost-user-gpio device hw/virtio: add vhost-user-gpio-pci boilerplate

 hw/virtio/Kconfig                   |   5 +
 hw/virtio/meson.build               |   2 +
 hw/virtio/vhost-user-gpio-pci.c     |  69 ++++++
 hw/virtio/vhost-user-gpio.c         | 343 ++++++++++++++++++++++++++++
 include/hw/virtio/vhost-user-gpio.h |  35 +++
 5 files changed, 454 insertions(+)
 create mode 100644 hw/virtio/vhost-user-gpio-pci.c
 create mode 100644 hw/virtio/vhost-user-gpio.c
 create mode 100644 include/hw/virtio/vhost-user-gpio.h
Signature · 1 line
2.31.1.272.g89b43f80a514
Signature · 1 line
Alex Bennée
#19 lore raw
In reply to Viresh Kumar
Quoted 8 lines · Viresh Kumar

Hello,

This patchset adds vhost-user-gpio device's support in Qemu. The support for the same has already been added to virtio specification and Linux Kernel.

A Rust based backend is also in progress and is tested against this patchset:

https://github.com/rust-vmm/vhost-device/pull/76

So I'm waiting for v2 of this, right?

Quoted 19 lines

--
Viresh

Viresh Kumar (2):
hw/virtio: add boilerplate for vhost-user-gpio device hw/virtio: add vhost-user-gpio-pci boilerplate

 hw/virtio/Kconfig                   |   5 +
 hw/virtio/meson.build               |   2 +
 hw/virtio/vhost-user-gpio-pci.c     |  69 ++++++
 hw/virtio/vhost-user-gpio.c         | 343 ++++++++++++++++++++++++++++
 include/hw/virtio/vhost-user-gpio.h |  35 +++
 5 files changed, 454 insertions(+)
 create mode 100644 hw/virtio/vhost-user-gpio-pci.c
 create mode 100644 hw/virtio/vhost-user-gpio.c
 create mode 100644 include/hw/virtio/vhost-user-gpio.h
Signature · 1 line
2.31.1.272.g89b43f80a514
#20 lore raw
Quoted 11 lines · Michael S. Tsirkin
Quoted 8 lines · Viresh Kumar

Hello,

This patchset adds vhost-user-gpio device's support in Qemu. The support for the same has already been added to virtio specification and Linux Kernel.

A Rust based backend is also in progress and is tested against this patchset:

https://github.com/rust-vmm/vhost-device/pull/76

So I'm waiting for v2 of this, right?

If you can give any pointers for how to properly instantiate the stub in qtest then we can spin v2 pretty quickly. I tried but it didn't work:

Subject: [RFC PATCH] tests/qtest: attempt to enable tests for virtio-gpio (!working)
Date: Fri, 21 Jan 2022 15:15:34 +0000
Message-Id: <[email protected]>
X-Mailer: git-send-email 2.30.2
In-Reply-To: <20220120040234.ehcjcdr6wrvlz3yc@vireshk-i7>
References: <20220120040234.ehcjcdr6wrvlz3yc@vireshk-i7>

it's all a bit opaque as to why it didn't.

Quoted 20 lines
Quoted 20 lines

--
Viresh

Viresh Kumar (2):
hw/virtio: add boilerplate for vhost-user-gpio device hw/virtio: add vhost-user-gpio-pci boilerplate

 hw/virtio/Kconfig                   |   5 +
 hw/virtio/meson.build               |   2 +
 hw/virtio/vhost-user-gpio-pci.c     |  69 ++++++
 hw/virtio/vhost-user-gpio.c         | 343 ++++++++++++++++++++++++++++
 include/hw/virtio/vhost-user-gpio.h |  35 +++
 5 files changed, 454 insertions(+)
 create mode 100644 hw/virtio/vhost-user-gpio-pci.c
 create mode 100644 hw/virtio/vhost-user-gpio.c
 create mode 100644 include/hw/virtio/vhost-user-gpio.h
Signature · 1 line
2.31.1.272.g89b43f80a514
Signature · 1 line
Alex Bennée
#21 lore raw
Quoted 18 lines · Alex Bennée
Quoted 11 lines · Michael S. Tsirkin
Quoted 8 lines · Viresh Kumar

Hello,

This patchset adds vhost-user-gpio device's support in Qemu. The support for the same has already been added to virtio specification and Linux Kernel.

A Rust based backend is also in progress and is tested against this patchset:

https://github.com/rust-vmm/vhost-device/pull/76

So I'm waiting for v2 of this, right?

If you can give any pointers for how to properly instantiate the stub in qtest then we can spin v2 pretty quickly. I tried but it didn't work:

Quoted 8 lines

Subject: [RFC PATCH] tests/qtest: attempt to enable tests for virtio-gpio (!working)
Date: Fri, 21 Jan 2022 15:15:34 +0000
Message-Id: <[email protected]>
X-Mailer: git-send-email 2.30.2
In-Reply-To: <20220120040234.ehcjcdr6wrvlz3yc@vireshk-i7>
References: <20220120040234.ehcjcdr6wrvlz3yc@vireshk-i7>

it's all a bit opaque as to why it didn't.

A bit rushed short term, it this isn't resolved soon I'll try to help. Maybe try CC other vhost-user maintainers (for blk,gpu etc) meanwhile.

Quoted 24 lines
Quoted 20 lines
Quoted 20 lines

--
Viresh

Viresh Kumar (2):
hw/virtio: add boilerplate for vhost-user-gpio device hw/virtio: add vhost-user-gpio-pci boilerplate

 hw/virtio/Kconfig                   |   5 +
 hw/virtio/meson.build               |   2 +
 hw/virtio/vhost-user-gpio-pci.c     |  69 ++++++
 hw/virtio/vhost-user-gpio.c         | 343 ++++++++++++++++++++++++++++
 include/hw/virtio/vhost-user-gpio.h |  35 +++
 5 files changed, 454 insertions(+)
 create mode 100644 hw/virtio/vhost-user-gpio-pci.c
 create mode 100644 hw/virtio/vhost-user-gpio.c
 create mode 100644 include/hw/virtio/vhost-user-gpio.h
Signature · 1 line
2.31.1.272.g89b43f80a514
Signature · 1 line
Alex Bennée
#22 lore raw
Quoted 26 lines · Michael S. Tsirkin
Quoted 25 lines · Alex Bennée
Quoted 11 lines · Michael S. Tsirkin
Quoted 8 lines · Viresh Kumar

Hello,

This patchset adds vhost-user-gpio device's support in Qemu. The support for the same has already been added to virtio specification and Linux Kernel.

A Rust based backend is also in progress and is tested against this patchset:

https://github.com/rust-vmm/vhost-device/pull/76

So I'm waiting for v2 of this, right?

If you can give any pointers for how to properly instantiate the stub in qtest then we can spin v2 pretty quickly. I tried but it didn't work:

Subject: [RFC PATCH] tests/qtest: attempt to enable tests for virtio-gpio (!working)
Date: Fri, 21 Jan 2022 15:15:34 +0000
Message-Id: <[email protected]>
X-Mailer: git-send-email 2.30.2
In-Reply-To: <20220120040234.ehcjcdr6wrvlz3yc@vireshk-i7>
References: <20220120040234.ehcjcdr6wrvlz3yc@vireshk-i7>

Quoted 4 lines
Quoted 1 line

it's all a bit opaque as to why it didn't.

A bit rushed short term, it this isn't resolved soon I'll try to help. Maybe try CC other vhost-user maintainers (for blk,gpu etc) meanwhile.

I have expanded the CC list for this thread and the RFC patch to see if anyone has pointers. I'll pull this series into my vhost cleanups series for post 7.0.

Signature · 1 line
Alex Bennée
Applied
#23 lore raw
In reply to Viresh Kumar
Quoted 8 lines · Viresh Kumar

Hello,

This patchset adds vhost-user-gpio device's support in Qemu. The support for the same has already been added to virtio specification and Linux Kernel.

A Rust based backend is also in progress and is tested against this patchset:

https://github.com/rust-vmm/vhost-device/pull/76

Queued to virtio/virtio-cleanups-and-gpio-v2, thanks.

Signature · 1 line
Alex Bennée