Skip to contentSkip to conversation

mdev based hardware virtio offloading support

PATCH Superseded Jason Wang 7 participants 75 messages Sep 24, 2019 – Oct 10, 2019

Subsystems (17)

v2 of 13 Superseded Latest is v13 →

lore
Share & tools
lore

Jason Wang

author
#0 lore raw

v2 cover mdev based hardware virtio offloading support

Hi all:

There are hardware that can do virtio datapath offloading while having its own control path. This path tries to implement a mdev based unified API to support using kernel virtio driver to drive those devices. This is done by introducing a new mdev transport for virtio (virtio_mdev) and register itself as a new kind of mdev driver. Then it provides a unified way for kernel virtio driver to talk with mdev device implementation.

Though the series only contains kernel driver support, the goal is to make the transport generic enough to support userspace drivers. This means vhost-mdev[1] could be built on top as well by resuing the transport.

A sample driver is also implemented which simulate a virito-net loopback ethernet device on top of vringh + workqueue. This could be used as a reference implementation for real hardware driver.

Consider mdev framework only support VFIO device and driver right now, this series also extend it to support other types. This is done through introducing class id to the device and pairing it with id_talbe claimed by the driver. On top, this seris also decouple device specific parents ops out of the common ones.

Pktgen test was done with virito-net + mvnet loop back device.

Please review.

[1] https://lkml.org/lkml/2019/9/16/869

Changes from V1:

- move virtio_mdev.c to drivers/virtio
- store class_id in mdev_device instead of mdev_parent
- store device_ops in mdev_device instead of mdev_parent
- reorder the patch, vringh fix comes first
- really silent compiling warnings
- really switch to use u16 for class_id
- uevent and modpost support for mdev class_id
- vraious tweaks per comments from Parav

Changes from RFC-V2:

- silent compile warnings on some specific configuration
- use u16 instead u8 for class id
- reseve MDEV_ID_VHOST for future vhost-mdev work
- introduce "virtio" type for mvnet and make "vhost" type for future work
- add entries in MAINTAINER
- tweak and typos fixes in commit log

Changes from RFC-V1:

- rename device id to class id
- add docs for class id and device specific ops (device_ops)
- split device_ops into seperate headers
- drop the mdev_set_dma_ops()
- use device_ops to implement the transport API, then it's not a part of UAPI any more
- use GFP_ATOMIC in mvnet sample device and other tweaks
- set_vring_base/get_vring_base support for mvnet device

Jason Wang (8):
vringh: fix copy direction of vringh_iov_push_kern()
mdev: class id support
mdev: bus uevent support
modpost: add support for mdev class id
mdev: introduce device specific ops
mdev: introduce virtio device and its device ops
virtio: introduce a mdev based transport
docs: sample driver to demonstrate how to implement virtio-mdev
framework

 .../driver-api/vfio-mediated-device.rst       |   7 +-
 MAINTAINERS                                   |   2 +
 drivers/gpu/drm/i915/gvt/kvmgt.c              |  18 +-
 drivers/s390/cio/vfio_ccw_ops.c               |  18 +-
 drivers/s390/crypto/vfio_ap_ops.c             |  14 +-
 drivers/vfio/mdev/mdev_core.c                 |  19 +
 drivers/vfio/mdev/mdev_driver.c               |  22 +
 drivers/vfio/mdev/mdev_private.h              |   2 +
 drivers/vfio/mdev/vfio_mdev.c                 |  45 +-
 drivers/vhost/vringh.c                        |   8 +-
 drivers/virtio/Kconfig                        |   7 +
 drivers/virtio/Makefile                       |   1 +
 drivers/virtio/virtio_mdev.c                  | 417 +++++++++++
 include/linux/mdev.h                          |  52 +-
 include/linux/mod_devicetable.h               |   8 +
 include/linux/vfio_mdev.h                     |  52 ++
 include/linux/virtio_mdev.h                   | 145 ++++
 samples/Kconfig                               |   7 +
 samples/vfio-mdev/Makefile                    |   1 +
 samples/vfio-mdev/mbochs.c                    |  20 +-
 samples/vfio-mdev/mdpy.c                      |  20 +-
 samples/vfio-mdev/mtty.c                      |  18 +-
 samples/vfio-mdev/mvnet.c                     | 692 ++++++++++++++++++
 scripts/mod/devicetable-offsets.c             |   3 +
 scripts/mod/file2alias.c                      |  10 +
 25 files changed, 1524 insertions(+), 84 deletions(-)
 create mode 100644 drivers/virtio/virtio_mdev.c
 create mode 100644 include/linux/vfio_mdev.h
 create mode 100644 include/linux/virtio_mdev.h
 create mode 100644 samples/vfio-mdev/mvnet.c
Signature · 1 line
2.19.1
author
#1 lore raw

PATCH v2 1/8 vringh: fix copy direction of vringh_iov_push_kern()

We want to copy from iov to buf, so the direction was wrong.

---
 drivers/vhost/vringh.c | 8 +++++++-
 1 file changed, 7 insertions(+), 1 deletion(-)
drivers/vhost/vringh.c +7 -1
index 08ad0d1f0476..a0a2d74967ef 100644
@@ -852,6 +852,12 @@ static inline int xfer_kern(void *src, void *dst, size_t len)
 	return 0;
 }
 
+static inline int kern_xfer(void *dst, void *src, size_t len)
+{
+	memcpy(dst, src, len);
+	return 0;
+}
+
 /**
  * vringh_init_kern - initialize a vringh for a kernelspace vring.
  * @vrh: the vringh to initialize.
@@ -958,7 +964,7 @@ EXPORT_SYMBOL(vringh_iov_pull_kern);
 ssize_t vringh_iov_push_kern(struct vringh_kiov *wiov,
 			     const void *src, size_t len)
 {
-	return vringh_iov_xfer(wiov, (void *)src, len, xfer_kern);
+	return vringh_iov_xfer(wiov, (void *)src, len, kern_xfer);
 }
 EXPORT_SYMBOL(vringh_iov_push_kern);
 
Signature · 1 line
2.19.1
author
#2 lore raw

PATCH v2 1/8 vringh: fix copy direction of vringh_iov_push_kern()

In reply to Jason Wang

We want to copy from iov to buf, so the direction was wrong.

---
 drivers/vhost/vringh.c | 8 +++++++-
 1 file changed, 7 insertions(+), 1 deletion(-)
drivers/vhost/vringh.c +7 -1
index 08ad0d1f0476..a0a2d74967ef 100644
@@ -852,6 +852,12 @@ static inline int xfer_kern(void *src, void *dst, size_t len)
 	return 0;
 }
 
+static inline int kern_xfer(void *dst, void *src, size_t len)
+{
+	memcpy(dst, src, len);
+	return 0;
+}
+
 /**
  * vringh_init_kern - initialize a vringh for a kernelspace vring.
  * @vrh: the vringh to initialize.
@@ -958,7 +964,7 @@ EXPORT_SYMBOL(vringh_iov_pull_kern);
 ssize_t vringh_iov_push_kern(struct vringh_kiov *wiov,
 			     const void *src, size_t len)
 {
-	return vringh_iov_xfer(wiov, (void *)src, len, xfer_kern);
+	return vringh_iov_xfer(wiov, (void *)src, len, kern_xfer);
 }
 EXPORT_SYMBOL(vringh_iov_push_kern);
 
Signature · 1 line
2.19.1
author
#3 lore raw

PATCH v2 2/8 mdev: class id support

In reply to Jason Wang

Mdev bus only supports vfio driver right now, so it doesn't implement match method. But in the future, we may add drivers other than vfio, the first driver could be virtio-mdev. This means we need to add device class id support in bus match method to pair the mdev device and mdev driver correctly.

So this patch adds id_table to mdev_driver and class_id for mdev parent with the match method for mdev bus.

---
 Documentation/driver-api/vfio-mediated-device.rst |  3 +++
 drivers/gpu/drm/i915/gvt/kvmgt.c                  |  1 +
 drivers/s390/cio/vfio_ccw_ops.c                   |  1 +
 drivers/s390/crypto/vfio_ap_ops.c                 |  1 +
 drivers/vfio/mdev/mdev_core.c                     |  7 +++++++
 drivers/vfio/mdev/mdev_driver.c                   | 14 ++++++++++++++
 drivers/vfio/mdev/mdev_private.h                  |  1 +
 drivers/vfio/mdev/vfio_mdev.c                     |  6 ++++++
 include/linux/mdev.h                              |  8 ++++++++
 include/linux/mod_devicetable.h                   |  8 ++++++++
 samples/vfio-mdev/mbochs.c                        |  1 +
 samples/vfio-mdev/mdpy.c                          |  1 +
 samples/vfio-mdev/mtty.c                          |  1 +
 13 files changed, 53 insertions(+)
Documentation/driver-api/vfio-mediated-device.rst +3 -0
index 25eb7d5b834b..a5bdc60d62a1 100644
@@ -102,12 +102,14 @@ structure to represent a mediated device's driver::
       * @probe: called when new device created
       * @remove: called when device removed
       * @driver: device driver structure
+      * @id_table: the ids serviced by this driver
       */
      struct mdev_driver {
 	     const char *name;
 	     int  (*probe)  (struct device *dev);
 	     void (*remove) (struct device *dev);
 	     struct device_driver    driver;
+	     const struct mdev_class_id *id_table;
      };
 
 A mediated bus driver for mdev should use this structure in the function calls
@@ -165,6 +167,7 @@ register itself with the mdev core driver::
 	extern int  mdev_register_device(struct device *dev,
 	                                 const struct mdev_parent_ops *ops);
 
+
 However, the mdev_parent_ops structure is not required in the function call
 that a driver should use to unregister itself with the mdev core driver::
 
drivers/gpu/drm/i915/gvt/kvmgt.c +1 -0
index 23aa3e50cbf8..f793252a3d2a 100644
@@ -678,6 +678,7 @@ static int intel_vgpu_create(struct kobject *kobj, struct mdev_device *mdev)
 		     dev_name(mdev_dev(mdev)));
 	ret = 0;
 
+	mdev_set_class_id(mdev, MDEV_ID_VFIO);
 out:
 	return ret;
 }
drivers/s390/cio/vfio_ccw_ops.c +1 -0
index f0d71ab77c50..d258ef1fedb9 100644
@@ -129,6 +129,7 @@ static int vfio_ccw_mdev_create(struct kobject *kobj, struct mdev_device *mdev)
 			   private->sch->schid.ssid,
 			   private->sch->schid.sch_no);
 
+	mdev_set_class_id(mdev, MDEV_ID_VFIO);
 	return 0;
 }
 
drivers/s390/crypto/vfio_ap_ops.c +1 -0
index 5c0f53c6dde7..2cfd96112aa0 100644
@@ -343,6 +343,7 @@ static int vfio_ap_mdev_create(struct kobject *kobj, struct mdev_device *mdev)
 	list_add(&matrix_mdev->node, &matrix_dev->mdev_list);
 	mutex_unlock(&matrix_dev->lock);
 
+	mdev_set_class_id(mdev, MDEV_ID_VFIO);
 	return 0;
 }
 
drivers/vfio/mdev/mdev_core.c +7 -0
index b558d4cfd082..8764cf4a276d 100644
@@ -45,6 +45,12 @@ void mdev_set_drvdata(struct mdev_device *mdev, void *data)
 }
 EXPORT_SYMBOL(mdev_set_drvdata);
 
+void mdev_set_class_id(struct mdev_device *mdev, u16 id)
+{
+	mdev->class_id = id;
+}
+EXPORT_SYMBOL(mdev_set_class_id);
+
 struct device *mdev_dev(struct mdev_device *mdev)
 {
 	return &mdev->dev;
@@ -135,6 +141,7 @@ static int mdev_device_remove_cb(struct device *dev, void *data)
  * mdev_register_device : Register a device
  * @dev: device structure representing parent device.
  * @ops: Parent device operation structure to be registered.
+ * @id: device id.
  *
  * Add device to list of registered parent devices.
  * Returns a negative value on error, otherwise 0.
drivers/vfio/mdev/mdev_driver.c +14 -0
index 0d3223aee20b..b7c40ce86ee3 100644
@@ -69,8 +69,22 @@ static int mdev_remove(struct device *dev)
 	return 0;
 }
 
+static int mdev_match(struct device *dev, struct device_driver *drv)
+{
+	unsigned int i;
+	struct mdev_device *mdev = to_mdev_device(dev);
+	struct mdev_driver *mdrv = to_mdev_driver(drv);
+	const struct mdev_class_id *ids = mdrv->id_table;
+
+	for (i = 0; ids[i].id; i++)
+		if (ids[i].id == mdev->class_id)
+			return 1;
+	return 0;
+}
+
 struct bus_type mdev_bus_type = {
 	.name		= "mdev",
+	.match		= mdev_match,
 	.probe		= mdev_probe,
 	.remove		= mdev_remove,
 };
drivers/vfio/mdev/mdev_private.h +1 -0
index 7d922950caaf..c65f436c1869 100644
@@ -33,6 +33,7 @@ struct mdev_device {
 	struct kobject *type_kobj;
 	struct device *iommu_device;
 	bool active;
+	u16 class_id;
 };
 
 #define to_mdev_device(dev)	container_of(dev, struct mdev_device, dev)
drivers/vfio/mdev/vfio_mdev.c +6 -0
index 30964a4e0a28..fd2a4d9a3f26 100644
@@ -120,10 +120,16 @@ static void vfio_mdev_remove(struct device *dev)
 	vfio_del_group_dev(dev);
 }
 
+static struct mdev_class_id id_table[] = {
+	{ MDEV_ID_VFIO },
+	{ 0 },
+};
+
 static struct mdev_driver vfio_mdev_driver = {
 	.name	= "vfio_mdev",
 	.probe	= vfio_mdev_probe,
 	.remove	= vfio_mdev_remove,
+	.id_table = id_table,
 };
 
 static int __init vfio_mdev_init(void)
include/linux/mdev.h +8 -0
index 0ce30ca78db0..3974650c074f 100644
@@ -118,6 +118,7 @@ struct mdev_type_attribute mdev_type_attr_##_name =		\
  * @probe: called when new device created
  * @remove: called when device removed
  * @driver: device driver structure
+ * @id_table: the ids serviced by this driver.
  *
  **/
 struct mdev_driver {
@@ -125,12 +126,14 @@ struct mdev_driver {
 	int  (*probe)(struct device *dev);
 	void (*remove)(struct device *dev);
 	struct device_driver driver;
+	const struct mdev_class_id *id_table;
 };
 
 #define to_mdev_driver(drv)	container_of(drv, struct mdev_driver, driver)
 
 void *mdev_get_drvdata(struct mdev_device *mdev);
 void mdev_set_drvdata(struct mdev_device *mdev, void *data);
+void mdev_set_class_id(struct mdev_device *mdev, u16 id);
 const guid_t *mdev_uuid(struct mdev_device *mdev);
 
 extern struct bus_type mdev_bus_type;
@@ -145,4 +148,9 @@ struct device *mdev_parent_dev(struct mdev_device *mdev);
 struct device *mdev_dev(struct mdev_device *mdev);
 struct mdev_device *mdev_from_dev(struct device *dev);
 
+enum {
+	MDEV_ID_VFIO = 1,
+	/* New entries must be added here */
+};
+
 #endif /* MDEV_H */
include/linux/mod_devicetable.h +8 -0
index 5714fd35a83c..f32c6e44fb1a 100644
@@ -821,4 +821,12 @@ struct wmi_device_id {
 	const void *context;
 };
 
+/**
+ * struct mdev_class_id - MDEV device class identifier
+ * @id: Used to identify a specific class of device, e.g vfio-mdev device.
+ */
+struct mdev_class_id {
+	__u16 id;
+};
+
 #endif /* LINUX_MOD_DEVICETABLE_H */
samples/vfio-mdev/mbochs.c +1 -0
index ac5c8c17b1ff..8a8583c892b2 100644
@@ -561,6 +561,7 @@ static int mbochs_create(struct kobject *kobj, struct mdev_device *mdev)
 	mbochs_reset(mdev);
 
 	mbochs_used_mbytes += type->mbytes;
+	mdev_set_class_id(mdev, MDEV_ID_VFIO);
 	return 0;
 
 err_mem:
samples/vfio-mdev/mdpy.c +1 -0
index cc86bf6566e4..88d7e76f3836 100644
@@ -269,6 +269,7 @@ static int mdpy_create(struct kobject *kobj, struct mdev_device *mdev)
 	mdpy_reset(mdev);
 
 	mdpy_count++;
+	mdev_set_class_id(mdev, MDEV_ID_VFIO);
 	return 0;
 }
 
samples/vfio-mdev/mtty.c +1 -0
index 92e770a06ea2..4e0735143b69 100644
@@ -770,6 +770,7 @@ static int mtty_create(struct kobject *kobj, struct mdev_device *mdev)
 	list_add(&mdev_state->next, &mdev_devices_list);
 	mutex_unlock(&mdev_list_lock);
 
+	mdev_set_class_id(mdev, MDEV_ID_VFIO);
 	return 0;
 }
 
Signature · 6 lines
2.19.1

_______________________________________________
dri-devel mailing list
[email protected]
https://lists.freedesktop.org/mailman/listinfo/dri-devel
#4 lore raw

Re: [PATCH V2 2/8] mdev: class id support

Quoted 8 lines · On Tue, 24 Sep 2019 21:53:26 +0800 Jason Wang <[email protected]> wrote:

Mdev bus only supports vfio driver right now, so it doesn't implement match method. But in the future, we may add drivers other than vfio, the first driver could be virtio-mdev. This means we need to add device class id support in bus match method to pair the mdev device and mdev driver correctly.

So this patch adds id_table to mdev_driver and class_id for mdev parent with the match method for mdev bus.

Description needs to be revised from v1, class_id is no longer on the parent.

Quoted 37 lines
---
 Documentation/driver-api/vfio-mediated-device.rst |  3 +++
 drivers/gpu/drm/i915/gvt/kvmgt.c                  |  1 +
 drivers/s390/cio/vfio_ccw_ops.c                   |  1 +
 drivers/s390/crypto/vfio_ap_ops.c                 |  1 +
 drivers/vfio/mdev/mdev_core.c                     |  7 +++++++
 drivers/vfio/mdev/mdev_driver.c                   | 14 ++++++++++++++
 drivers/vfio/mdev/mdev_private.h                  |  1 +
 drivers/vfio/mdev/vfio_mdev.c                     |  6 ++++++
 include/linux/mdev.h                              |  8 ++++++++
 include/linux/mod_devicetable.h                   |  8 ++++++++
 samples/vfio-mdev/mbochs.c                        |  1 +
 samples/vfio-mdev/mdpy.c                          |  1 +
 samples/vfio-mdev/mtty.c                          |  1 +
 13 files changed, 53 insertions(+)
Documentation/driver-api/vfio-mediated-device.rst +2 -0
index 25eb7d5b834b..a5bdc60d62a1 100644
@@ -102,12 +102,14 @@ structure to represent a mediated device's driver::
       * @probe: called when new device created
       * @remove: called when device removed
       * @driver: device driver structure
+      * @id_table: the ids serviced by this driver
       */
      struct mdev_driver {
 	     const char *name;
 	     int  (*probe)  (struct device *dev);
 	     void (*remove) (struct device *dev);
 	     struct device_driver    driver;
+	     const struct mdev_class_id *id_table;
      };
 
 A mediated bus driver for mdev should use this structure in the function calls

Documentation/driver-api/vfio-mediated-device.rst register itself with the mdev core driver:: · line 172

Quoted 6 lines
Documentation/driver-api/vfio-mediated-device.rst +1 -0
@@ -165,6 +167,7 @@ register itself with the mdev core driver::
 	extern int  mdev_register_device(struct device *dev,
 	                                 const struct mdev_parent_ops *ops);
 
+
 However, the mdev_parent_ops structure is not required in the function call
 that a driver should use to unregister itself with the mdev core driver::

Unintended extra line? Doesn't seem to match surrounding formatting.

Calling mdev_set_class_id() as part of create seems relatively fundamental to the vendor driver with this change, it should be added to the documentation.

Quoted 68 lines
drivers/gpu/drm/i915/gvt/kvmgt.c +1 -0
index 23aa3e50cbf8..f793252a3d2a 100644
@@ -678,6 +678,7 @@ static int intel_vgpu_create(struct kobject *kobj, struct mdev_device *mdev)
 		     dev_name(mdev_dev(mdev)));
 	ret = 0;
 
+	mdev_set_class_id(mdev, MDEV_ID_VFIO);
 out:
 	return ret;
 }
drivers/s390/cio/vfio_ccw_ops.c +1 -0
index f0d71ab77c50..d258ef1fedb9 100644
@@ -129,6 +129,7 @@ static int vfio_ccw_mdev_create(struct kobject *kobj, struct mdev_device *mdev)
 			   private->sch->schid.ssid,
 			   private->sch->schid.sch_no);
 
+	mdev_set_class_id(mdev, MDEV_ID_VFIO);
 	return 0;
 }
 
drivers/s390/crypto/vfio_ap_ops.c +1 -0
index 5c0f53c6dde7..2cfd96112aa0 100644
@@ -343,6 +343,7 @@ static int vfio_ap_mdev_create(struct kobject *kobj, struct mdev_device *mdev)
 	list_add(&matrix_mdev->node, &matrix_dev->mdev_list);
 	mutex_unlock(&matrix_dev->lock);
 
+	mdev_set_class_id(mdev, MDEV_ID_VFIO);
 	return 0;
 }
 
drivers/vfio/mdev/mdev_core.c +7 -0
index b558d4cfd082..8764cf4a276d 100644
@@ -45,6 +45,12 @@ void mdev_set_drvdata(struct mdev_device *mdev, void *data)
 }
 EXPORT_SYMBOL(mdev_set_drvdata);
 
+void mdev_set_class_id(struct mdev_device *mdev, u16 id)
+{
+	mdev->class_id = id;
+}
+EXPORT_SYMBOL(mdev_set_class_id);
+
 struct device *mdev_dev(struct mdev_device *mdev)
 {
 	return &mdev->dev;
@@ -135,6 +141,7 @@ static int mdev_device_remove_cb(struct device *dev, void *data)
  * mdev_register_device : Register a device
  * @dev: device structure representing parent device.
  * @ops: Parent device operation structure to be registered.
+ * @id: device id.
  *
  * Add device to list of registered parent devices.
  * Returns a negative value on error, otherwise 0.
drivers/vfio/mdev/mdev_driver.c +0 -0
index 0d3223aee20b..b7c40ce86ee3 100644
@@ -69,8 +69,22 @@ static int mdev_remove(struct device *dev)
 	return 0;
 }

drivers/vfio/mdev/mdev_driver.c static int mdev_remove(struct device *dev) · line 80

Quoted 10 lines
drivers/vfio/mdev/mdev_driver.c +9 -0
 
+static int mdev_match(struct device *dev, struct device_driver *drv)
+{
+	unsigned int i;
+	struct mdev_device *mdev = to_mdev_device(dev);
+	struct mdev_driver *mdrv = to_mdev_driver(drv);
+	const struct mdev_class_id *ids = mdrv->id_table;
+
+	for (i = 0; ids[i].id; i++)
+		if (ids[i].id == mdev->class_id)

With the proposed API here, mdev->class_id can be NULL, do we allow devices to exist in that state or should mdev_device_create() generate an error if the .create() callback doesn't register a type? Or a warn_once and assume MDEV_ID_VFIO? The latter seems pretty sketchy when we get to patch 5/8. Thanks,

Alex

Quoted 133 lines
drivers/vfio/mdev/mdev_driver.c +5 -0
+			return 1;
+	return 0;
+}
+
 struct bus_type mdev_bus_type = {
 	.name		= "mdev",
+	.match		= mdev_match,
 	.probe		= mdev_probe,
 	.remove		= mdev_remove,
 };
drivers/vfio/mdev/mdev_private.h +1 -0
index 7d922950caaf..c65f436c1869 100644
@@ -33,6 +33,7 @@ struct mdev_device {
 	struct kobject *type_kobj;
 	struct device *iommu_device;
 	bool active;
+	u16 class_id;
 };
 
 #define to_mdev_device(dev)	container_of(dev, struct mdev_device, dev)
drivers/vfio/mdev/vfio_mdev.c +6 -0
index 30964a4e0a28..fd2a4d9a3f26 100644
@@ -120,10 +120,16 @@ static void vfio_mdev_remove(struct device *dev)
 	vfio_del_group_dev(dev);
 }
 
+static struct mdev_class_id id_table[] = {
+	{ MDEV_ID_VFIO },
+	{ 0 },
+};
+
 static struct mdev_driver vfio_mdev_driver = {
 	.name	= "vfio_mdev",
 	.probe	= vfio_mdev_probe,
 	.remove	= vfio_mdev_remove,
+	.id_table = id_table,
 };
 
 static int __init vfio_mdev_init(void)
include/linux/mdev.h +8 -0
index 0ce30ca78db0..3974650c074f 100644
@@ -118,6 +118,7 @@ struct mdev_type_attribute mdev_type_attr_##_name =		\
  * @probe: called when new device created
  * @remove: called when device removed
  * @driver: device driver structure
+ * @id_table: the ids serviced by this driver.
  *
  **/
 struct mdev_driver {
@@ -125,12 +126,14 @@ struct mdev_driver {
 	int  (*probe)(struct device *dev);
 	void (*remove)(struct device *dev);
 	struct device_driver driver;
+	const struct mdev_class_id *id_table;
 };
 
 #define to_mdev_driver(drv)	container_of(drv, struct mdev_driver, driver)
 
 void *mdev_get_drvdata(struct mdev_device *mdev);
 void mdev_set_drvdata(struct mdev_device *mdev, void *data);
+void mdev_set_class_id(struct mdev_device *mdev, u16 id);
 const guid_t *mdev_uuid(struct mdev_device *mdev);
 
 extern struct bus_type mdev_bus_type;
@@ -145,4 +148,9 @@ struct device *mdev_parent_dev(struct mdev_device *mdev);
 struct device *mdev_dev(struct mdev_device *mdev);
 struct mdev_device *mdev_from_dev(struct device *dev);
 
+enum {
+	MDEV_ID_VFIO = 1,
+	/* New entries must be added here */
+};
+
 #endif /* MDEV_H */
include/linux/mod_devicetable.h +8 -0
index 5714fd35a83c..f32c6e44fb1a 100644
@@ -821,4 +821,12 @@ struct wmi_device_id {
 	const void *context;
 };
 
+/**
+ * struct mdev_class_id - MDEV device class identifier
+ * @id: Used to identify a specific class of device, e.g vfio-mdev device.
+ */
+struct mdev_class_id {
+	__u16 id;
+};
+
 #endif /* LINUX_MOD_DEVICETABLE_H */
samples/vfio-mdev/mbochs.c +1 -0
index ac5c8c17b1ff..8a8583c892b2 100644
@@ -561,6 +561,7 @@ static int mbochs_create(struct kobject *kobj, struct mdev_device *mdev)
 	mbochs_reset(mdev);
 
 	mbochs_used_mbytes += type->mbytes;
+	mdev_set_class_id(mdev, MDEV_ID_VFIO);
 	return 0;
 
 err_mem:
samples/vfio-mdev/mdpy.c +1 -0
index cc86bf6566e4..88d7e76f3836 100644
@@ -269,6 +269,7 @@ static int mdpy_create(struct kobject *kobj, struct mdev_device *mdev)
 	mdpy_reset(mdev);
 
 	mdpy_count++;
+	mdev_set_class_id(mdev, MDEV_ID_VFIO);
 	return 0;
 }
 
samples/vfio-mdev/mtty.c +1 -0
index 92e770a06ea2..4e0735143b69 100644
@@ -770,6 +770,7 @@ static int mtty_create(struct kobject *kobj, struct mdev_device *mdev)
 	list_add(&mdev_state->next, &mdev_devices_list);
 	mutex_unlock(&mdev_list_lock);
 
+	mdev_set_class_id(mdev, MDEV_ID_VFIO);
 	return 0;
 }
 
author
#5 lore raw

Re: [PATCH V2 2/8] mdev: class id support

Quoted 9 lines · Alex Williamson
Quoted 6 lines · On Tue, 24 Sep 2019 21:53:26 +0800 Jason Wang <[email protected]> wrote:

Mdev bus only supports vfio driver right now, so it doesn't implement match method. But in the future, we may add drivers other than vfio, the first driver could be virtio-mdev. This means we need to add device class id support in bus match method to pair the mdev device and mdev driver correctly.

Quoted 4 lines
Quoted 2 lines

So this patch adds id_table to mdev_driver and class_id for mdev parent with the match method for mdev bus.

Description needs to be revised from v1, class_id is no longer on the parent.

Yes, will fix.

Quoted 40 lines
Quoted 40 lines
---
 Documentation/driver-api/vfio-mediated-device.rst |  3 +++
 drivers/gpu/drm/i915/gvt/kvmgt.c                  |  1 +
 drivers/s390/cio/vfio_ccw_ops.c                   |  1 +
 drivers/s390/crypto/vfio_ap_ops.c                 |  1 +
 drivers/vfio/mdev/mdev_core.c                     |  7 +++++++
 drivers/vfio/mdev/mdev_driver.c                   | 14 ++++++++++++++
 drivers/vfio/mdev/mdev_private.h                  |  1 +
 drivers/vfio/mdev/vfio_mdev.c                     |  6 ++++++
 include/linux/mdev.h                              |  8 ++++++++
 include/linux/mod_devicetable.h                   |  8 ++++++++
 samples/vfio-mdev/mbochs.c                        |  1 +
 samples/vfio-mdev/mdpy.c                          |  1 +
 samples/vfio-mdev/mtty.c                          |  1 +
 13 files changed, 53 insertions(+)
Documentation/driver-api/vfio-mediated-device.rst +2 -0
index 25eb7d5b834b..a5bdc60d62a1 100644
@@ -102,12 +102,14 @@ structure to represent a mediated device's driver::
       * @probe: called when new device created
       * @remove: called when device removed
       * @driver: device driver structure
+      * @id_table: the ids serviced by this driver
       */
      struct mdev_driver {
 	     const char *name;
 	     int  (*probe)  (struct device *dev);
 	     void (*remove) (struct device *dev);
 	     struct device_driver    driver;
+	     const struct mdev_class_id *id_table;
      };
 
 A mediated bus driver for mdev should use this structure in the function calls
@@ -165,6 +167,7 @@ register itself with the mdev core driver::
 	extern int  mdev_register_device(struct device *dev,
 	                                 const struct mdev_parent_ops *ops);
 
Quoted 8 lines

Documentation/driver-api/vfio-mediated-device.rst register itself with the mdev core driver:: · line 172

Quoted 3 lines
Documentation/driver-api/vfio-mediated-device.rst +1 -0
+
 However, the mdev_parent_ops structure is not required in the function call
 that a driver should use to unregister itself with the mdev core driver::

Unintended extra line? Doesn't seem to match surrounding formatting.

Calling mdev_set_class_id() as part of create seems relatively fundamental to the vendor driver with this change, it should be added to the documentation.

Right.

Quoted 70 lines
Quoted 70 lines
drivers/gpu/drm/i915/gvt/kvmgt.c +1 -0
index 23aa3e50cbf8..f793252a3d2a 100644
@@ -678,6 +678,7 @@ static int intel_vgpu_create(struct kobject *kobj, struct mdev_device *mdev)
 		     dev_name(mdev_dev(mdev)));
 	ret = 0;
 
+	mdev_set_class_id(mdev, MDEV_ID_VFIO);
 out:
 	return ret;
 }
drivers/s390/cio/vfio_ccw_ops.c +1 -0
index f0d71ab77c50..d258ef1fedb9 100644
@@ -129,6 +129,7 @@ static int vfio_ccw_mdev_create(struct kobject *kobj, struct mdev_device *mdev)
 			   private->sch->schid.ssid,
 			   private->sch->schid.sch_no);
 
+	mdev_set_class_id(mdev, MDEV_ID_VFIO);
 	return 0;
 }
 
drivers/s390/crypto/vfio_ap_ops.c +1 -0
index 5c0f53c6dde7..2cfd96112aa0 100644
@@ -343,6 +343,7 @@ static int vfio_ap_mdev_create(struct kobject *kobj, struct mdev_device *mdev)
 	list_add(&matrix_mdev->node, &matrix_dev->mdev_list);
 	mutex_unlock(&matrix_dev->lock);
 
+	mdev_set_class_id(mdev, MDEV_ID_VFIO);
 	return 0;
 }
 
drivers/vfio/mdev/mdev_core.c +7 -0
index b558d4cfd082..8764cf4a276d 100644
@@ -45,6 +45,12 @@ void mdev_set_drvdata(struct mdev_device *mdev, void *data)
 }
 EXPORT_SYMBOL(mdev_set_drvdata);
 
+void mdev_set_class_id(struct mdev_device *mdev, u16 id)
+{
+	mdev->class_id = id;
+}
+EXPORT_SYMBOL(mdev_set_class_id);
+
 struct device *mdev_dev(struct mdev_device *mdev)
 {
 	return &mdev->dev;
@@ -135,6 +141,7 @@ static int mdev_device_remove_cb(struct device *dev, void *data)
  * mdev_register_device : Register a device
  * @dev: device structure representing parent device.
  * @ops: Parent device operation structure to be registered.
+ * @id: device id.
  *
  * Add device to list of registered parent devices.
  * Returns a negative value on error, otherwise 0.
drivers/vfio/mdev/mdev_driver.c +0 -0
index 0d3223aee20b..b7c40ce86ee3 100644
@@ -69,8 +69,22 @@ static int mdev_remove(struct device *dev)
 	return 0;
 }
 
Quoted 14 lines

drivers/vfio/mdev/mdev_driver.c static int mdev_remove(struct device *dev) · line 80

Quoted 9 lines
drivers/vfio/mdev/mdev_driver.c +9 -0
+static int mdev_match(struct device *dev, struct device_driver *drv)
+{
+	unsigned int i;
+	struct mdev_device *mdev = to_mdev_device(dev);
+	struct mdev_driver *mdrv = to_mdev_driver(drv);
+	const struct mdev_class_id *ids = mdrv->id_table;
+
+	for (i = 0; ids[i].id; i++)
+		if (ids[i].id == mdev->class_id)

With the proposed API here, mdev->class_id can be NULL, do we allow devices to exist in that state or should mdev_device_create() generate an error if the .create() callback doesn't register a type? Or a warn_once and assume MDEV_ID_VFIO? The latter seems pretty sketchy when we get to patch 5/8. Thanks,

It looks to me the error from mdev_device_create() is better.

Thanks

Quoted 132 lines

Alex

Quoted 129 lines
drivers/vfio/mdev/mdev_driver.c +5 -0
+			return 1;
+	return 0;
+}
+
 struct bus_type mdev_bus_type = {
 	.name		= "mdev",
+	.match		= mdev_match,
 	.probe		= mdev_probe,
 	.remove		= mdev_remove,
 };
drivers/vfio/mdev/mdev_private.h +1 -0
index 7d922950caaf..c65f436c1869 100644
@@ -33,6 +33,7 @@ struct mdev_device {
 	struct kobject *type_kobj;
 	struct device *iommu_device;
 	bool active;
+	u16 class_id;
 };
 
 #define to_mdev_device(dev)	container_of(dev, struct mdev_device, dev)
drivers/vfio/mdev/vfio_mdev.c +6 -0
index 30964a4e0a28..fd2a4d9a3f26 100644
@@ -120,10 +120,16 @@ static void vfio_mdev_remove(struct device *dev)
 	vfio_del_group_dev(dev);
 }
 
+static struct mdev_class_id id_table[] = {
+	{ MDEV_ID_VFIO },
+	{ 0 },
+};
+
 static struct mdev_driver vfio_mdev_driver = {
 	.name	= "vfio_mdev",
 	.probe	= vfio_mdev_probe,
 	.remove	= vfio_mdev_remove,
+	.id_table = id_table,
 };
 
 static int __init vfio_mdev_init(void)
include/linux/mdev.h +8 -0
index 0ce30ca78db0..3974650c074f 100644
@@ -118,6 +118,7 @@ struct mdev_type_attribute mdev_type_attr_##_name =		\
  * @probe: called when new device created
  * @remove: called when device removed
  * @driver: device driver structure
+ * @id_table: the ids serviced by this driver.
  *
  **/
 struct mdev_driver {
@@ -125,12 +126,14 @@ struct mdev_driver {
 	int  (*probe)(struct device *dev);
 	void (*remove)(struct device *dev);
 	struct device_driver driver;
+	const struct mdev_class_id *id_table;
 };
 
 #define to_mdev_driver(drv)	container_of(drv, struct mdev_driver, driver)
 
 void *mdev_get_drvdata(struct mdev_device *mdev);
 void mdev_set_drvdata(struct mdev_device *mdev, void *data);
+void mdev_set_class_id(struct mdev_device *mdev, u16 id);
 const guid_t *mdev_uuid(struct mdev_device *mdev);
 
 extern struct bus_type mdev_bus_type;
@@ -145,4 +148,9 @@ struct device *mdev_parent_dev(struct mdev_device *mdev);
 struct device *mdev_dev(struct mdev_device *mdev);
 struct mdev_device *mdev_from_dev(struct device *dev);
 
+enum {
+	MDEV_ID_VFIO = 1,
+	/* New entries must be added here */
+};
+
 #endif /* MDEV_H */
include/linux/mod_devicetable.h +8 -0
index 5714fd35a83c..f32c6e44fb1a 100644
@@ -821,4 +821,12 @@ struct wmi_device_id {
 	const void *context;
 };
 
+/**
+ * struct mdev_class_id - MDEV device class identifier
+ * @id: Used to identify a specific class of device, e.g vfio-mdev device.
+ */
+struct mdev_class_id {
+	__u16 id;
+};
+
 #endif /* LINUX_MOD_DEVICETABLE_H */
samples/vfio-mdev/mbochs.c +1 -0
index ac5c8c17b1ff..8a8583c892b2 100644
@@ -561,6 +561,7 @@ static int mbochs_create(struct kobject *kobj, struct mdev_device *mdev)
 	mbochs_reset(mdev);
 
 	mbochs_used_mbytes += type->mbytes;
+	mdev_set_class_id(mdev, MDEV_ID_VFIO);
 	return 0;
 
 err_mem:
samples/vfio-mdev/mdpy.c +1 -0
index cc86bf6566e4..88d7e76f3836 100644
@@ -269,6 +269,7 @@ static int mdpy_create(struct kobject *kobj, struct mdev_device *mdev)
 	mdpy_reset(mdev);
 
 	mdpy_count++;
+	mdev_set_class_id(mdev, MDEV_ID_VFIO);
 	return 0;
 }
 
samples/vfio-mdev/mtty.c +0 -0
index 92e770a06ea2..4e0735143b69 100644
@@ -770,6 +770,7 @@ static int mtty_create(struct kobject *kobj, struct mdev_device *mdev)
 	list_add(&mdev_state->next, &mdev_devices_list);
 	mutex_unlock(&mdev_list_lock);
 
Quoted 4 lines

samples/vfio-mdev/mtty.c static int mtty_create(struct kobject *kobj, struct mdev_device *mdev) · line 776

Quoted 4 lines
samples/vfio-mdev/mtty.c +1 -0
+	mdev_set_class_id(mdev, MDEV_ID_VFIO);
 	return 0;
 }
 

_______________________________________________
Intel-gfx mailing list
[email protected]
https://lists.freedesktop.org/mailman/listinfo/intel-gfx

author
#6 lore raw

Re: [PATCH V2 2/8] mdev: class id support

In reply to Alex Williamson
Quoted 9 lines · Alex Williamson
Quoted 6 lines · On Tue, 24 Sep 2019 21:53:26 +0800 Jason Wang <[email protected]> wrote:

Mdev bus only supports vfio driver right now, so it doesn't implement match method. But in the future, we may add drivers other than vfio, the first driver could be virtio-mdev. This means we need to add device class id support in bus match method to pair the mdev device and mdev driver correctly.

Quoted 4 lines
Quoted 2 lines

So this patch adds id_table to mdev_driver and class_id for mdev parent with the match method for mdev bus.

Description needs to be revised from v1, class_id is no longer on the parent.

Yes, will fix.

Quoted 40 lines
Quoted 40 lines
---
 Documentation/driver-api/vfio-mediated-device.rst |  3 +++
 drivers/gpu/drm/i915/gvt/kvmgt.c                  |  1 +
 drivers/s390/cio/vfio_ccw_ops.c                   |  1 +
 drivers/s390/crypto/vfio_ap_ops.c                 |  1 +
 drivers/vfio/mdev/mdev_core.c                     |  7 +++++++
 drivers/vfio/mdev/mdev_driver.c                   | 14 ++++++++++++++
 drivers/vfio/mdev/mdev_private.h                  |  1 +
 drivers/vfio/mdev/vfio_mdev.c                     |  6 ++++++
 include/linux/mdev.h                              |  8 ++++++++
 include/linux/mod_devicetable.h                   |  8 ++++++++
 samples/vfio-mdev/mbochs.c                        |  1 +
 samples/vfio-mdev/mdpy.c                          |  1 +
 samples/vfio-mdev/mtty.c                          |  1 +
 13 files changed, 53 insertions(+)
Documentation/driver-api/vfio-mediated-device.rst +2 -0
index 25eb7d5b834b..a5bdc60d62a1 100644
@@ -102,12 +102,14 @@ structure to represent a mediated device's driver::
       * @probe: called when new device created
       * @remove: called when device removed
       * @driver: device driver structure
+      * @id_table: the ids serviced by this driver
       */
      struct mdev_driver {
 	     const char *name;
 	     int  (*probe)  (struct device *dev);
 	     void (*remove) (struct device *dev);
 	     struct device_driver    driver;
+	     const struct mdev_class_id *id_table;
      };
 
 A mediated bus driver for mdev should use this structure in the function calls
@@ -165,6 +167,7 @@ register itself with the mdev core driver::
 	extern int  mdev_register_device(struct device *dev,
 	                                 const struct mdev_parent_ops *ops);
 
Quoted 8 lines

Documentation/driver-api/vfio-mediated-device.rst register itself with the mdev core driver:: · line 172

Quoted 3 lines
Documentation/driver-api/vfio-mediated-device.rst +1 -0
+
 However, the mdev_parent_ops structure is not required in the function call
 that a driver should use to unregister itself with the mdev core driver::

Unintended extra line? Doesn't seem to match surrounding formatting.

Calling mdev_set_class_id() as part of create seems relatively fundamental to the vendor driver with this change, it should be added to the documentation.

Right.

Quoted 70 lines
Quoted 70 lines
drivers/gpu/drm/i915/gvt/kvmgt.c +1 -0
index 23aa3e50cbf8..f793252a3d2a 100644
@@ -678,6 +678,7 @@ static int intel_vgpu_create(struct kobject *kobj, struct mdev_device *mdev)
 		     dev_name(mdev_dev(mdev)));
 	ret = 0;
 
+	mdev_set_class_id(mdev, MDEV_ID_VFIO);
 out:
 	return ret;
 }
drivers/s390/cio/vfio_ccw_ops.c +1 -0
index f0d71ab77c50..d258ef1fedb9 100644
@@ -129,6 +129,7 @@ static int vfio_ccw_mdev_create(struct kobject *kobj, struct mdev_device *mdev)
 			   private->sch->schid.ssid,
 			   private->sch->schid.sch_no);
 
+	mdev_set_class_id(mdev, MDEV_ID_VFIO);
 	return 0;
 }
 
drivers/s390/crypto/vfio_ap_ops.c +1 -0
index 5c0f53c6dde7..2cfd96112aa0 100644
@@ -343,6 +343,7 @@ static int vfio_ap_mdev_create(struct kobject *kobj, struct mdev_device *mdev)
 	list_add(&matrix_mdev->node, &matrix_dev->mdev_list);
 	mutex_unlock(&matrix_dev->lock);
 
+	mdev_set_class_id(mdev, MDEV_ID_VFIO);
 	return 0;
 }
 
drivers/vfio/mdev/mdev_core.c +7 -0
index b558d4cfd082..8764cf4a276d 100644
@@ -45,6 +45,12 @@ void mdev_set_drvdata(struct mdev_device *mdev, void *data)
 }
 EXPORT_SYMBOL(mdev_set_drvdata);
 
+void mdev_set_class_id(struct mdev_device *mdev, u16 id)
+{
+	mdev->class_id = id;
+}
+EXPORT_SYMBOL(mdev_set_class_id);
+
 struct device *mdev_dev(struct mdev_device *mdev)
 {
 	return &mdev->dev;
@@ -135,6 +141,7 @@ static int mdev_device_remove_cb(struct device *dev, void *data)
  * mdev_register_device : Register a device
  * @dev: device structure representing parent device.
  * @ops: Parent device operation structure to be registered.
+ * @id: device id.
  *
  * Add device to list of registered parent devices.
  * Returns a negative value on error, otherwise 0.
drivers/vfio/mdev/mdev_driver.c +0 -0
index 0d3223aee20b..b7c40ce86ee3 100644
@@ -69,8 +69,22 @@ static int mdev_remove(struct device *dev)
 	return 0;
 }
 
Quoted 14 lines

drivers/vfio/mdev/mdev_driver.c static int mdev_remove(struct device *dev) · line 80

Quoted 9 lines
drivers/vfio/mdev/mdev_driver.c +9 -0
+static int mdev_match(struct device *dev, struct device_driver *drv)
+{
+	unsigned int i;
+	struct mdev_device *mdev = to_mdev_device(dev);
+	struct mdev_driver *mdrv = to_mdev_driver(drv);
+	const struct mdev_class_id *ids = mdrv->id_table;
+
+	for (i = 0; ids[i].id; i++)
+		if (ids[i].id == mdev->class_id)

With the proposed API here, mdev->class_id can be NULL, do we allow devices to exist in that state or should mdev_device_create() generate an error if the .create() callback doesn't register a type? Or a warn_once and assume MDEV_ID_VFIO? The latter seems pretty sketchy when we get to patch 5/8. Thanks,

It looks to me the error from mdev_device_create() is better.

Thanks

Quoted 132 lines

Alex

Quoted 129 lines
drivers/vfio/mdev/mdev_driver.c +5 -0
+			return 1;
+	return 0;
+}
+
 struct bus_type mdev_bus_type = {
 	.name		= "mdev",
+	.match		= mdev_match,
 	.probe		= mdev_probe,
 	.remove		= mdev_remove,
 };
drivers/vfio/mdev/mdev_private.h +1 -0
index 7d922950caaf..c65f436c1869 100644
@@ -33,6 +33,7 @@ struct mdev_device {
 	struct kobject *type_kobj;
 	struct device *iommu_device;
 	bool active;
+	u16 class_id;
 };
 
 #define to_mdev_device(dev)	container_of(dev, struct mdev_device, dev)
drivers/vfio/mdev/vfio_mdev.c +6 -0
index 30964a4e0a28..fd2a4d9a3f26 100644
@@ -120,10 +120,16 @@ static void vfio_mdev_remove(struct device *dev)
 	vfio_del_group_dev(dev);
 }
 
+static struct mdev_class_id id_table[] = {
+	{ MDEV_ID_VFIO },
+	{ 0 },
+};
+
 static struct mdev_driver vfio_mdev_driver = {
 	.name	= "vfio_mdev",
 	.probe	= vfio_mdev_probe,
 	.remove	= vfio_mdev_remove,
+	.id_table = id_table,
 };
 
 static int __init vfio_mdev_init(void)
include/linux/mdev.h +8 -0
index 0ce30ca78db0..3974650c074f 100644
@@ -118,6 +118,7 @@ struct mdev_type_attribute mdev_type_attr_##_name =		\
  * @probe: called when new device created
  * @remove: called when device removed
  * @driver: device driver structure
+ * @id_table: the ids serviced by this driver.
  *
  **/
 struct mdev_driver {
@@ -125,12 +126,14 @@ struct mdev_driver {
 	int  (*probe)(struct device *dev);
 	void (*remove)(struct device *dev);
 	struct device_driver driver;
+	const struct mdev_class_id *id_table;
 };
 
 #define to_mdev_driver(drv)	container_of(drv, struct mdev_driver, driver)
 
 void *mdev_get_drvdata(struct mdev_device *mdev);
 void mdev_set_drvdata(struct mdev_device *mdev, void *data);
+void mdev_set_class_id(struct mdev_device *mdev, u16 id);
 const guid_t *mdev_uuid(struct mdev_device *mdev);
 
 extern struct bus_type mdev_bus_type;
@@ -145,4 +148,9 @@ struct device *mdev_parent_dev(struct mdev_device *mdev);
 struct device *mdev_dev(struct mdev_device *mdev);
 struct mdev_device *mdev_from_dev(struct device *dev);
 
+enum {
+	MDEV_ID_VFIO = 1,
+	/* New entries must be added here */
+};
+
 #endif /* MDEV_H */
include/linux/mod_devicetable.h +8 -0
index 5714fd35a83c..f32c6e44fb1a 100644
@@ -821,4 +821,12 @@ struct wmi_device_id {
 	const void *context;
 };
 
+/**
+ * struct mdev_class_id - MDEV device class identifier
+ * @id: Used to identify a specific class of device, e.g vfio-mdev device.
+ */
+struct mdev_class_id {
+	__u16 id;
+};
+
 #endif /* LINUX_MOD_DEVICETABLE_H */
samples/vfio-mdev/mbochs.c +1 -0
index ac5c8c17b1ff..8a8583c892b2 100644
@@ -561,6 +561,7 @@ static int mbochs_create(struct kobject *kobj, struct mdev_device *mdev)
 	mbochs_reset(mdev);
 
 	mbochs_used_mbytes += type->mbytes;
+	mdev_set_class_id(mdev, MDEV_ID_VFIO);
 	return 0;
 
 err_mem:
samples/vfio-mdev/mdpy.c +1 -0
index cc86bf6566e4..88d7e76f3836 100644
@@ -269,6 +269,7 @@ static int mdpy_create(struct kobject *kobj, struct mdev_device *mdev)
 	mdpy_reset(mdev);
 
 	mdpy_count++;
+	mdev_set_class_id(mdev, MDEV_ID_VFIO);
 	return 0;
 }
 
samples/vfio-mdev/mtty.c +0 -0
index 92e770a06ea2..4e0735143b69 100644
@@ -770,6 +770,7 @@ static int mtty_create(struct kobject *kobj, struct mdev_device *mdev)
 	list_add(&mdev_state->next, &mdev_devices_list);
 	mutex_unlock(&mdev_list_lock);
 
Quoted 4 lines

samples/vfio-mdev/mtty.c static int mtty_create(struct kobject *kobj, struct mdev_device *mdev) · line 776

Quoted 4 lines
samples/vfio-mdev/mtty.c +1 -0
+	mdev_set_class_id(mdev, MDEV_ID_VFIO);
 	return 0;
 }
 

_______________________________________________
Virtualization mailing list
[email protected]
https://lists.linuxfoundation.org/mailman/listinfo/virtualization

#7 lore raw

Re: [PATCH V2 2/8] mdev: class id support

In reply to Jason Wang
Quoted 8 lines · On Tue, 24 Sep 2019 21:53:26 +0800 Jason Wang <[email protected]> wrote:

Mdev bus only supports vfio driver right now, so it doesn't implement match method. But in the future, we may add drivers other than vfio, the first driver could be virtio-mdev. This means we need to add device class id support in bus match method to pair the mdev device and mdev driver correctly.

So this patch adds id_table to mdev_driver and class_id for mdev parent with the match method for mdev bus.

Description needs to be revised from v1, class_id is no longer on the parent.

Quoted 37 lines
---
 Documentation/driver-api/vfio-mediated-device.rst |  3 +++
 drivers/gpu/drm/i915/gvt/kvmgt.c                  |  1 +
 drivers/s390/cio/vfio_ccw_ops.c                   |  1 +
 drivers/s390/crypto/vfio_ap_ops.c                 |  1 +
 drivers/vfio/mdev/mdev_core.c                     |  7 +++++++
 drivers/vfio/mdev/mdev_driver.c                   | 14 ++++++++++++++
 drivers/vfio/mdev/mdev_private.h                  |  1 +
 drivers/vfio/mdev/vfio_mdev.c                     |  6 ++++++
 include/linux/mdev.h                              |  8 ++++++++
 include/linux/mod_devicetable.h                   |  8 ++++++++
 samples/vfio-mdev/mbochs.c                        |  1 +
 samples/vfio-mdev/mdpy.c                          |  1 +
 samples/vfio-mdev/mtty.c                          |  1 +
 13 files changed, 53 insertions(+)
Documentation/driver-api/vfio-mediated-device.rst +2 -0
index 25eb7d5b834b..a5bdc60d62a1 100644
@@ -102,12 +102,14 @@ structure to represent a mediated device's driver::
       * @probe: called when new device created
       * @remove: called when device removed
       * @driver: device driver structure
+      * @id_table: the ids serviced by this driver
       */
      struct mdev_driver {
 	     const char *name;
 	     int  (*probe)  (struct device *dev);
 	     void (*remove) (struct device *dev);
 	     struct device_driver    driver;
+	     const struct mdev_class_id *id_table;
      };
 
 A mediated bus driver for mdev should use this structure in the function calls

Documentation/driver-api/vfio-mediated-device.rst register itself with the mdev core driver:: · line 172

Quoted 6 lines
Documentation/driver-api/vfio-mediated-device.rst +1 -0
@@ -165,6 +167,7 @@ register itself with the mdev core driver::
 	extern int  mdev_register_device(struct device *dev,
 	                                 const struct mdev_parent_ops *ops);
 
+
 However, the mdev_parent_ops structure is not required in the function call
 that a driver should use to unregister itself with the mdev core driver::

Unintended extra line? Doesn't seem to match surrounding formatting.

Calling mdev_set_class_id() as part of create seems relatively fundamental to the vendor driver with this change, it should be added to the documentation.

Quoted 68 lines
drivers/gpu/drm/i915/gvt/kvmgt.c +1 -0
index 23aa3e50cbf8..f793252a3d2a 100644
@@ -678,6 +678,7 @@ static int intel_vgpu_create(struct kobject *kobj, struct mdev_device *mdev)
 		     dev_name(mdev_dev(mdev)));
 	ret = 0;
 
+	mdev_set_class_id(mdev, MDEV_ID_VFIO);
 out:
 	return ret;
 }
drivers/s390/cio/vfio_ccw_ops.c +1 -0
index f0d71ab77c50..d258ef1fedb9 100644
@@ -129,6 +129,7 @@ static int vfio_ccw_mdev_create(struct kobject *kobj, struct mdev_device *mdev)
 			   private->sch->schid.ssid,
 			   private->sch->schid.sch_no);
 
+	mdev_set_class_id(mdev, MDEV_ID_VFIO);
 	return 0;
 }
 
drivers/s390/crypto/vfio_ap_ops.c +1 -0
index 5c0f53c6dde7..2cfd96112aa0 100644
@@ -343,6 +343,7 @@ static int vfio_ap_mdev_create(struct kobject *kobj, struct mdev_device *mdev)
 	list_add(&matrix_mdev->node, &matrix_dev->mdev_list);
 	mutex_unlock(&matrix_dev->lock);
 
+	mdev_set_class_id(mdev, MDEV_ID_VFIO);
 	return 0;
 }
 
drivers/vfio/mdev/mdev_core.c +7 -0
index b558d4cfd082..8764cf4a276d 100644
@@ -45,6 +45,12 @@ void mdev_set_drvdata(struct mdev_device *mdev, void *data)
 }
 EXPORT_SYMBOL(mdev_set_drvdata);
 
+void mdev_set_class_id(struct mdev_device *mdev, u16 id)
+{
+	mdev->class_id = id;
+}
+EXPORT_SYMBOL(mdev_set_class_id);
+
 struct device *mdev_dev(struct mdev_device *mdev)
 {
 	return &mdev->dev;
@@ -135,6 +141,7 @@ static int mdev_device_remove_cb(struct device *dev, void *data)
  * mdev_register_device : Register a device
  * @dev: device structure representing parent device.
  * @ops: Parent device operation structure to be registered.
+ * @id: device id.
  *
  * Add device to list of registered parent devices.
  * Returns a negative value on error, otherwise 0.
drivers/vfio/mdev/mdev_driver.c +0 -0
index 0d3223aee20b..b7c40ce86ee3 100644
@@ -69,8 +69,22 @@ static int mdev_remove(struct device *dev)
 	return 0;
 }

drivers/vfio/mdev/mdev_driver.c static int mdev_remove(struct device *dev) · line 80

Quoted 10 lines
drivers/vfio/mdev/mdev_driver.c +9 -0
 
+static int mdev_match(struct device *dev, struct device_driver *drv)
+{
+	unsigned int i;
+	struct mdev_device *mdev = to_mdev_device(dev);
+	struct mdev_driver *mdrv = to_mdev_driver(drv);
+	const struct mdev_class_id *ids = mdrv->id_table;
+
+	for (i = 0; ids[i].id; i++)
+		if (ids[i].id == mdev->class_id)

With the proposed API here, mdev->class_id can be NULL, do we allow devices to exist in that state or should mdev_device_create() generate an error if the .create() callback doesn't register a type? Or a warn_once and assume MDEV_ID_VFIO? The latter seems pretty sketchy when we get to patch 5/8. Thanks,

Alex

Quoted 133 lines
drivers/vfio/mdev/mdev_driver.c +5 -0
+			return 1;
+	return 0;
+}
+
 struct bus_type mdev_bus_type = {
 	.name		= "mdev",
+	.match		= mdev_match,
 	.probe		= mdev_probe,
 	.remove		= mdev_remove,
 };
drivers/vfio/mdev/mdev_private.h +1 -0
index 7d922950caaf..c65f436c1869 100644
@@ -33,6 +33,7 @@ struct mdev_device {
 	struct kobject *type_kobj;
 	struct device *iommu_device;
 	bool active;
+	u16 class_id;
 };
 
 #define to_mdev_device(dev)	container_of(dev, struct mdev_device, dev)
drivers/vfio/mdev/vfio_mdev.c +6 -0
index 30964a4e0a28..fd2a4d9a3f26 100644
@@ -120,10 +120,16 @@ static void vfio_mdev_remove(struct device *dev)
 	vfio_del_group_dev(dev);
 }
 
+static struct mdev_class_id id_table[] = {
+	{ MDEV_ID_VFIO },
+	{ 0 },
+};
+
 static struct mdev_driver vfio_mdev_driver = {
 	.name	= "vfio_mdev",
 	.probe	= vfio_mdev_probe,
 	.remove	= vfio_mdev_remove,
+	.id_table = id_table,
 };
 
 static int __init vfio_mdev_init(void)
include/linux/mdev.h +8 -0
index 0ce30ca78db0..3974650c074f 100644
@@ -118,6 +118,7 @@ struct mdev_type_attribute mdev_type_attr_##_name =		\
  * @probe: called when new device created
  * @remove: called when device removed
  * @driver: device driver structure
+ * @id_table: the ids serviced by this driver.
  *
  **/
 struct mdev_driver {
@@ -125,12 +126,14 @@ struct mdev_driver {
 	int  (*probe)(struct device *dev);
 	void (*remove)(struct device *dev);
 	struct device_driver driver;
+	const struct mdev_class_id *id_table;
 };
 
 #define to_mdev_driver(drv)	container_of(drv, struct mdev_driver, driver)
 
 void *mdev_get_drvdata(struct mdev_device *mdev);
 void mdev_set_drvdata(struct mdev_device *mdev, void *data);
+void mdev_set_class_id(struct mdev_device *mdev, u16 id);
 const guid_t *mdev_uuid(struct mdev_device *mdev);
 
 extern struct bus_type mdev_bus_type;
@@ -145,4 +148,9 @@ struct device *mdev_parent_dev(struct mdev_device *mdev);
 struct device *mdev_dev(struct mdev_device *mdev);
 struct mdev_device *mdev_from_dev(struct device *dev);
 
+enum {
+	MDEV_ID_VFIO = 1,
+	/* New entries must be added here */
+};
+
 #endif /* MDEV_H */
include/linux/mod_devicetable.h +8 -0
index 5714fd35a83c..f32c6e44fb1a 100644
@@ -821,4 +821,12 @@ struct wmi_device_id {
 	const void *context;
 };
 
+/**
+ * struct mdev_class_id - MDEV device class identifier
+ * @id: Used to identify a specific class of device, e.g vfio-mdev device.
+ */
+struct mdev_class_id {
+	__u16 id;
+};
+
 #endif /* LINUX_MOD_DEVICETABLE_H */
samples/vfio-mdev/mbochs.c +1 -0
index ac5c8c17b1ff..8a8583c892b2 100644
@@ -561,6 +561,7 @@ static int mbochs_create(struct kobject *kobj, struct mdev_device *mdev)
 	mbochs_reset(mdev);
 
 	mbochs_used_mbytes += type->mbytes;
+	mdev_set_class_id(mdev, MDEV_ID_VFIO);
 	return 0;
 
 err_mem:
samples/vfio-mdev/mdpy.c +1 -0
index cc86bf6566e4..88d7e76f3836 100644
@@ -269,6 +269,7 @@ static int mdpy_create(struct kobject *kobj, struct mdev_device *mdev)
 	mdpy_reset(mdev);
 
 	mdpy_count++;
+	mdev_set_class_id(mdev, MDEV_ID_VFIO);
 	return 0;
 }
 
samples/vfio-mdev/mtty.c +1 -0
index 92e770a06ea2..4e0735143b69 100644
@@ -770,6 +770,7 @@ static int mtty_create(struct kobject *kobj, struct mdev_device *mdev)
 	list_add(&mdev_state->next, &mdev_devices_list);
 	mutex_unlock(&mdev_list_lock);
 
+	mdev_set_class_id(mdev, MDEV_ID_VFIO);
 	return 0;
 }
 
#8 lore raw

RE: [PATCH V2 2/8] mdev: class id support

In reply to Jason Wang
Quoted 122 lines

From: Jason Wang
Sent: Tuesday, September 24, 2019 9:53 PM

Mdev bus only supports vfio driver right now, so it doesn't implement match method. But in the future, we may add drivers other than vfio, the first driver could be virtio-mdev. This means we need to add device class id support in bus match method to pair the mdev device and mdev driver correctly.

So this patch adds id_table to mdev_driver and class_id for mdev parent with the match method for mdev bus.

---
 Documentation/driver-api/vfio-mediated-device.rst |  3 +++
 drivers/gpu/drm/i915/gvt/kvmgt.c                  |  1 +
 drivers/s390/cio/vfio_ccw_ops.c                   |  1 +
 drivers/s390/crypto/vfio_ap_ops.c                 |  1 +
 drivers/vfio/mdev/mdev_core.c                     |  7 +++++++
 drivers/vfio/mdev/mdev_driver.c                   | 14 ++++++++++++++
 drivers/vfio/mdev/mdev_private.h                  |  1 +
 drivers/vfio/mdev/vfio_mdev.c                     |  6 ++++++
 include/linux/mdev.h                              |  8 ++++++++
 include/linux/mod_devicetable.h                   |  8 ++++++++
 samples/vfio-mdev/mbochs.c                        |  1 +
 samples/vfio-mdev/mdpy.c                          |  1 +
 samples/vfio-mdev/mtty.c                          |  1 +
 13 files changed, 53 insertions(+)
Documentation/driver-api/vfio-mediated-device.rst +0 -0
index 25eb7d5b834b..a5bdc60d62a1 100644
@@ -102,12 +102,14 @@ structure to represent a mediated device's

driver::
* @probe: called when new device created
* @remove: called when device removed
* @driver: device driver structure

Documentation/driver-api/vfio-mediated-device.rst +2 -0
+      * @id_table: the ids serviced by this driver
       */
      struct mdev_driver {
 	     const char *name;
 	     int  (*probe)  (struct device *dev);
 	     void (*remove) (struct device *dev);
 	     struct device_driver    driver;
+	     const struct mdev_class_id *id_table;
      };
 
 A mediated bus driver for mdev should use this structure in the function

calls

Documentation/driver-api/vfio-mediated-device.rst +1 -0
@@ -165,6 +167,7 @@ register itself with the mdev core driver::
 	extern int  mdev_register_device(struct device *dev,
 	                                 const struct mdev_parent_ops *ops);
 
+
 However, the mdev_parent_ops structure is not required in the function

call
that a driver should use to unregister itself with the mdev core driver::

drivers/gpu/drm/i915/gvt/kvmgt.c +1 -0
index 23aa3e50cbf8..f793252a3d2a 100644
@@ -678,6 +678,7 @@ static int intel_vgpu_create(struct kobject *kobj,
 struct mdev_device *mdev)
 		     dev_name(mdev_dev(mdev)));
 	ret = 0;
 
+	mdev_set_class_id(mdev, MDEV_ID_VFIO);
 out:
 	return ret;

}

drivers/s390/cio/vfio_ccw_ops.c +1 -0
index f0d71ab77c50..d258ef1fedb9 100644
@@ -129,6 +129,7 @@ static int vfio_ccw_mdev_create(struct kobject
 *kobj, struct mdev_device *mdev)
 			   private->sch->schid.ssid,
 			   private->sch->schid.sch_no);
 
+	mdev_set_class_id(mdev, MDEV_ID_VFIO);
 	return 0;
 }
drivers/s390/crypto/vfio_ap_ops.c +1 -0
index 5c0f53c6dde7..2cfd96112aa0 100644
@@ -343,6 +343,7 @@ static int vfio_ap_mdev_create(struct kobject *kobj,
 struct mdev_device *mdev)
 	list_add(&matrix_mdev->node, &matrix_dev->mdev_list);
 	mutex_unlock(&matrix_dev->lock);
 
+	mdev_set_class_id(mdev, MDEV_ID_VFIO);
 	return 0;
 }
drivers/vfio/mdev/mdev_core.c +6 -0
index b558d4cfd082..8764cf4a276d 100644
@@ -45,6 +45,12 @@ void mdev_set_drvdata(struct mdev_device *mdev,
 void *data)
 }
 EXPORT_SYMBOL(mdev_set_drvdata);
 
+void mdev_set_class_id(struct mdev_device *mdev, u16 id)
+{
+	mdev->class_id = id;
+}
+EXPORT_SYMBOL(mdev_set_class_id);
+
 struct device *mdev_dev(struct mdev_device *mdev)
 {
drivers/vfio/mdev/mdev_core.c +0 -0
 	return &mdev->dev;

drivers/vfio/mdev/mdev_core.c static int mdev_device_remove_cb(struct device · line 145

Quoted 5 lines
drivers/vfio/mdev/mdev_core.c +1 -0
@@ -135,6 +141,7 @@ static int mdev_device_remove_cb(struct device
 *dev, void *data)
  * mdev_register_device : Register a device
  * @dev: device structure representing parent device.
  * @ops: Parent device operation structure to be registered.
+ * @id: device id.

class id.

Quoted 163 lines
drivers/vfio/mdev/mdev_core.c +0 -0
  *
  * Add device to list of registered parent devices.

* Returns a negative value on error, otherwise 0.

drivers/vfio/mdev/mdev_driver.c +14 -0
index 0d3223aee20b..b7c40ce86ee3 100644
@@ -69,8 +69,22 @@ static int mdev_remove(struct device *dev)
 	return 0;
 }
 
+static int mdev_match(struct device *dev, struct device_driver *drv)
+{
+	unsigned int i;
+	struct mdev_device *mdev = to_mdev_device(dev);
+	struct mdev_driver *mdrv = to_mdev_driver(drv);
+	const struct mdev_class_id *ids = mdrv->id_table;
+
+	for (i = 0; ids[i].id; i++)
+		if (ids[i].id == mdev->class_id)
+			return 1;
+	return 0;
+}
+
 struct bus_type mdev_bus_type = {
 	.name		= "mdev",
+	.match		= mdev_match,
 	.probe		= mdev_probe,
 	.remove		= mdev_remove,
 };
drivers/vfio/mdev/mdev_private.h +1 -0
index 7d922950caaf..c65f436c1869 100644
@@ -33,6 +33,7 @@ struct mdev_device {
 	struct kobject *type_kobj;
 	struct device *iommu_device;
 	bool active;
+	u16 class_id;
 };
 
 #define to_mdev_device(dev)	container_of(dev, struct mdev_device, dev)
drivers/vfio/mdev/vfio_mdev.c +6 -0
index 30964a4e0a28..fd2a4d9a3f26 100644
@@ -120,10 +120,16 @@ static void vfio_mdev_remove(struct device *dev)
 	vfio_del_group_dev(dev);
 }
 
+static struct mdev_class_id id_table[] = {
+	{ MDEV_ID_VFIO },
+	{ 0 },
+};
+
 static struct mdev_driver vfio_mdev_driver = {
 	.name	= "vfio_mdev",
 	.probe	= vfio_mdev_probe,
 	.remove	= vfio_mdev_remove,
+	.id_table = id_table,
 };
 
 static int __init vfio_mdev_init(void)
include/linux/mdev.h +1 -0
index 0ce30ca78db0..3974650c074f 100644
@@ -118,6 +118,7 @@ struct mdev_type_attribute
 mdev_type_attr_##_name =		\
  * @probe: called when new device created
  * @remove: called when device removed
  * @driver: device driver structure
+ * @id_table: the ids serviced by this driver.
  *
  **/

struct mdev_driver {

include/linux/mdev.h +7 -0
@@ -125,12 +126,14 @@ struct mdev_driver {
 	int  (*probe)(struct device *dev);
 	void (*remove)(struct device *dev);
 	struct device_driver driver;
+	const struct mdev_class_id *id_table;
 };
 
 #define to_mdev_driver(drv)	container_of(drv, struct mdev_driver, driver)
 
 void *mdev_get_drvdata(struct mdev_device *mdev);
 void mdev_set_drvdata(struct mdev_device *mdev, void *data);
+void mdev_set_class_id(struct mdev_device *mdev, u16 id);
 const guid_t *mdev_uuid(struct mdev_device *mdev);
 
 extern struct bus_type mdev_bus_type;
@@ -145,4 +148,9 @@ struct device *mdev_parent_dev(struct
 mdev_device *mdev);
 struct device *mdev_dev(struct mdev_device *mdev);
 struct mdev_device *mdev_from_dev(struct device *dev);
 
+enum {
+	MDEV_ID_VFIO = 1,
+	/* New entries must be added here */
+};
+

#endif /* MDEV_H */

include/linux/mod_devicetable.h +8 -0
index 5714fd35a83c..f32c6e44fb1a 100644
@@ -821,4 +821,12 @@ struct wmi_device_id {
 	const void *context;
 };
 
+/**
+ * struct mdev_class_id - MDEV device class identifier
+ * @id: Used to identify a specific class of device, e.g vfio-mdev device.
+ */
+struct mdev_class_id {
+	__u16 id;
+};
+
 #endif /* LINUX_MOD_DEVICETABLE_H */
samples/vfio-mdev/mbochs.c +1 -0
index ac5c8c17b1ff..8a8583c892b2 100644
@@ -561,6 +561,7 @@ static int mbochs_create(struct kobject *kobj, struct
 mdev_device *mdev)
 	mbochs_reset(mdev);
 
 	mbochs_used_mbytes += type->mbytes;
+	mdev_set_class_id(mdev, MDEV_ID_VFIO);
 	return 0;
 

err_mem:

samples/vfio-mdev/mdpy.c +1 -0
index cc86bf6566e4..88d7e76f3836 100644
@@ -269,6 +269,7 @@ static int mdpy_create(struct kobject *kobj, struct
 mdev_device *mdev)
 	mdpy_reset(mdev);
 
 	mdpy_count++;
+	mdev_set_class_id(mdev, MDEV_ID_VFIO);
 	return 0;
 }
samples/vfio-mdev/mtty.c +1 -1
index 92e770a06ea2..4e0735143b69 100644
@@ -770,6 +770,7 @@ static int mtty_create(struct kobject *kobj, struct
 mdev_device *mdev)
 	list_add(&mdev_state->next, &mdev_devices_list);
 	mutex_unlock(&mdev_list_lock);
 
+	mdev_set_class_id(mdev, MDEV_ID_VFIO);
 	return 0;
 }
 
--
Quoted 1 line

2.19.1

_______________________________________________
Intel-gfx mailing list
[email protected]
https://lists.freedesktop.org/mailman/listinfo/intel-gfx

author
#9 lore raw

Re: [PATCH V2 2/8] mdev: class id support

Quoted 112 lines · Kevin
Quoted 112 lines

From: Jason Wang
Sent: Tuesday, September 24, 2019 9:53 PM

Mdev bus only supports vfio driver right now, so it doesn't implement match method. But in the future, we may add drivers other than vfio, the first driver could be virtio-mdev. This means we need to add device class id support in bus match method to pair the mdev device and mdev driver correctly.

So this patch adds id_table to mdev_driver and class_id for mdev parent with the match method for mdev bus.

---
 Documentation/driver-api/vfio-mediated-device.rst |  3 +++
 drivers/gpu/drm/i915/gvt/kvmgt.c                  |  1 +
 drivers/s390/cio/vfio_ccw_ops.c                   |  1 +
 drivers/s390/crypto/vfio_ap_ops.c                 |  1 +
 drivers/vfio/mdev/mdev_core.c                     |  7 +++++++
 drivers/vfio/mdev/mdev_driver.c                   | 14 ++++++++++++++
 drivers/vfio/mdev/mdev_private.h                  |  1 +
 drivers/vfio/mdev/vfio_mdev.c                     |  6 ++++++
 include/linux/mdev.h                              |  8 ++++++++
 include/linux/mod_devicetable.h                   |  8 ++++++++
 samples/vfio-mdev/mbochs.c                        |  1 +
 samples/vfio-mdev/mdpy.c                          |  1 +
 samples/vfio-mdev/mtty.c                          |  1 +
 13 files changed, 53 insertions(+)
Documentation/driver-api/vfio-mediated-device.rst +0 -0
index 25eb7d5b834b..a5bdc60d62a1 100644
@@ -102,12 +102,14 @@ structure to represent a mediated device's

driver::
* @probe: called when new device created
* @remove: called when device removed
* @driver: device driver structure

Documentation/driver-api/vfio-mediated-device.rst +2 -0
+      * @id_table: the ids serviced by this driver
       */
      struct mdev_driver {
 	     const char *name;
 	     int  (*probe)  (struct device *dev);
 	     void (*remove) (struct device *dev);
 	     struct device_driver    driver;
+	     const struct mdev_class_id *id_table;
      };
 
 A mediated bus driver for mdev should use this structure in the function

calls

Documentation/driver-api/vfio-mediated-device.rst +1 -0
@@ -165,6 +167,7 @@ register itself with the mdev core driver::
 	extern int  mdev_register_device(struct device *dev,
 	                                 const struct mdev_parent_ops *ops);
 
+
 However, the mdev_parent_ops structure is not required in the function

call
that a driver should use to unregister itself with the mdev core driver::

drivers/gpu/drm/i915/gvt/kvmgt.c +1 -0
index 23aa3e50cbf8..f793252a3d2a 100644
@@ -678,6 +678,7 @@ static int intel_vgpu_create(struct kobject *kobj,
 struct mdev_device *mdev)
 		     dev_name(mdev_dev(mdev)));
 	ret = 0;
 
+	mdev_set_class_id(mdev, MDEV_ID_VFIO);
 out:
 	return ret;

}

drivers/s390/cio/vfio_ccw_ops.c +1 -0
index f0d71ab77c50..d258ef1fedb9 100644
@@ -129,6 +129,7 @@ static int vfio_ccw_mdev_create(struct kobject
 *kobj, struct mdev_device *mdev)
 			   private->sch->schid.ssid,
 			   private->sch->schid.sch_no);
 
+	mdev_set_class_id(mdev, MDEV_ID_VFIO);
 	return 0;
 }
drivers/s390/crypto/vfio_ap_ops.c +1 -0
index 5c0f53c6dde7..2cfd96112aa0 100644
@@ -343,6 +343,7 @@ static int vfio_ap_mdev_create(struct kobject *kobj,
 struct mdev_device *mdev)
 	list_add(&matrix_mdev->node, &matrix_dev->mdev_list);
 	mutex_unlock(&matrix_dev->lock);
 
+	mdev_set_class_id(mdev, MDEV_ID_VFIO);
 	return 0;
 }
drivers/vfio/mdev/mdev_core.c +0 -0
index b558d4cfd082..8764cf4a276d 100644
@@ -45,6 +45,12 @@ void mdev_set_drvdata(struct mdev_device *mdev,
 void *data)
 }
 EXPORT_SYMBOL(mdev_set_drvdata);
 
Quoted 16 lines
Quoted 10 lines
drivers/vfio/mdev/mdev_core.c +6 -0
+void mdev_set_class_id(struct mdev_device *mdev, u16 id)
+{
+	mdev->class_id = id;
+}
+EXPORT_SYMBOL(mdev_set_class_id);
+
 struct device *mdev_dev(struct mdev_device *mdev)
 {
drivers/vfio/mdev/mdev_core.c +0 -0
 	return &mdev->dev;

drivers/vfio/mdev/mdev_core.c static int mdev_device_remove_cb(struct device · line 145

Quoted 5 lines
drivers/vfio/mdev/mdev_core.c +1 -0
@@ -135,6 +141,7 @@ static int mdev_device_remove_cb(struct device
 *dev, void *data)
  * mdev_register_device : Register a device
  * @dev: device structure representing parent device.
  * @ops: Parent device operation structure to be registered.
+ * @id: device id.

class id.

Right, will fix.

Thanks

_______________________________________________
dri-devel mailing list
[email protected]
https://lists.freedesktop.org/mailman/listinfo/dri-devel

author
#10 lore raw

Re: [PATCH V2 2/8] mdev: class id support

In reply to Tian, Kevin
Quoted 112 lines · Kevin
Quoted 112 lines

From: Jason Wang
Sent: Tuesday, September 24, 2019 9:53 PM

Mdev bus only supports vfio driver right now, so it doesn't implement match method. But in the future, we may add drivers other than vfio, the first driver could be virtio-mdev. This means we need to add device class id support in bus match method to pair the mdev device and mdev driver correctly.

So this patch adds id_table to mdev_driver and class_id for mdev parent with the match method for mdev bus.

---
 Documentation/driver-api/vfio-mediated-device.rst |  3 +++
 drivers/gpu/drm/i915/gvt/kvmgt.c                  |  1 +
 drivers/s390/cio/vfio_ccw_ops.c                   |  1 +
 drivers/s390/crypto/vfio_ap_ops.c                 |  1 +
 drivers/vfio/mdev/mdev_core.c                     |  7 +++++++
 drivers/vfio/mdev/mdev_driver.c                   | 14 ++++++++++++++
 drivers/vfio/mdev/mdev_private.h                  |  1 +
 drivers/vfio/mdev/vfio_mdev.c                     |  6 ++++++
 include/linux/mdev.h                              |  8 ++++++++
 include/linux/mod_devicetable.h                   |  8 ++++++++
 samples/vfio-mdev/mbochs.c                        |  1 +
 samples/vfio-mdev/mdpy.c                          |  1 +
 samples/vfio-mdev/mtty.c                          |  1 +
 13 files changed, 53 insertions(+)
Documentation/driver-api/vfio-mediated-device.rst +0 -0
index 25eb7d5b834b..a5bdc60d62a1 100644
@@ -102,12 +102,14 @@ structure to represent a mediated device's

driver::
* @probe: called when new device created
* @remove: called when device removed
* @driver: device driver structure

Documentation/driver-api/vfio-mediated-device.rst +2 -0
+      * @id_table: the ids serviced by this driver
       */
      struct mdev_driver {
 	     const char *name;
 	     int  (*probe)  (struct device *dev);
 	     void (*remove) (struct device *dev);
 	     struct device_driver    driver;
+	     const struct mdev_class_id *id_table;
      };
 
 A mediated bus driver for mdev should use this structure in the function

calls

Documentation/driver-api/vfio-mediated-device.rst +1 -0
@@ -165,6 +167,7 @@ register itself with the mdev core driver::
 	extern int  mdev_register_device(struct device *dev,
 	                                 const struct mdev_parent_ops *ops);
 
+
 However, the mdev_parent_ops structure is not required in the function

call
that a driver should use to unregister itself with the mdev core driver::

drivers/gpu/drm/i915/gvt/kvmgt.c +1 -0
index 23aa3e50cbf8..f793252a3d2a 100644
@@ -678,6 +678,7 @@ static int intel_vgpu_create(struct kobject *kobj,
 struct mdev_device *mdev)
 		     dev_name(mdev_dev(mdev)));
 	ret = 0;
 
+	mdev_set_class_id(mdev, MDEV_ID_VFIO);
 out:
 	return ret;

}

drivers/s390/cio/vfio_ccw_ops.c +1 -0
index f0d71ab77c50..d258ef1fedb9 100644
@@ -129,6 +129,7 @@ static int vfio_ccw_mdev_create(struct kobject
 *kobj, struct mdev_device *mdev)
 			   private->sch->schid.ssid,
 			   private->sch->schid.sch_no);
 
+	mdev_set_class_id(mdev, MDEV_ID_VFIO);
 	return 0;
 }
drivers/s390/crypto/vfio_ap_ops.c +1 -0
index 5c0f53c6dde7..2cfd96112aa0 100644
@@ -343,6 +343,7 @@ static int vfio_ap_mdev_create(struct kobject *kobj,
 struct mdev_device *mdev)
 	list_add(&matrix_mdev->node, &matrix_dev->mdev_list);
 	mutex_unlock(&matrix_dev->lock);
 
+	mdev_set_class_id(mdev, MDEV_ID_VFIO);
 	return 0;
 }
drivers/vfio/mdev/mdev_core.c +0 -0
index b558d4cfd082..8764cf4a276d 100644
@@ -45,6 +45,12 @@ void mdev_set_drvdata(struct mdev_device *mdev,
 void *data)
 }
 EXPORT_SYMBOL(mdev_set_drvdata);
 
Quoted 16 lines
Quoted 10 lines
drivers/vfio/mdev/mdev_core.c +6 -0
+void mdev_set_class_id(struct mdev_device *mdev, u16 id)
+{
+	mdev->class_id = id;
+}
+EXPORT_SYMBOL(mdev_set_class_id);
+
 struct device *mdev_dev(struct mdev_device *mdev)
 {
drivers/vfio/mdev/mdev_core.c +0 -0
 	return &mdev->dev;

drivers/vfio/mdev/mdev_core.c static int mdev_device_remove_cb(struct device · line 145

Quoted 5 lines
drivers/vfio/mdev/mdev_core.c +1 -0
@@ -135,6 +141,7 @@ static int mdev_device_remove_cb(struct device
 *dev, void *data)
  * mdev_register_device : Register a device
  * @dev: device structure representing parent device.
  * @ops: Parent device operation structure to be registered.
+ * @id: device id.

class id.

Right, will fix.

Thanks

_______________________________________________
Virtualization mailing list
[email protected]
https://lists.linuxfoundation.org/mailman/listinfo/virtualization

#11 lore raw

RE: [PATCH V2 2/8] mdev: class id support

In reply to Jason Wang
Quoted 122 lines

From: Jason Wang
Sent: Tuesday, September 24, 2019 9:53 PM

Mdev bus only supports vfio driver right now, so it doesn't implement match method. But in the future, we may add drivers other than vfio, the first driver could be virtio-mdev. This means we need to add device class id support in bus match method to pair the mdev device and mdev driver correctly.

So this patch adds id_table to mdev_driver and class_id for mdev parent with the match method for mdev bus.

---
 Documentation/driver-api/vfio-mediated-device.rst |  3 +++
 drivers/gpu/drm/i915/gvt/kvmgt.c                  |  1 +
 drivers/s390/cio/vfio_ccw_ops.c                   |  1 +
 drivers/s390/crypto/vfio_ap_ops.c                 |  1 +
 drivers/vfio/mdev/mdev_core.c                     |  7 +++++++
 drivers/vfio/mdev/mdev_driver.c                   | 14 ++++++++++++++
 drivers/vfio/mdev/mdev_private.h                  |  1 +
 drivers/vfio/mdev/vfio_mdev.c                     |  6 ++++++
 include/linux/mdev.h                              |  8 ++++++++
 include/linux/mod_devicetable.h                   |  8 ++++++++
 samples/vfio-mdev/mbochs.c                        |  1 +
 samples/vfio-mdev/mdpy.c                          |  1 +
 samples/vfio-mdev/mtty.c                          |  1 +
 13 files changed, 53 insertions(+)
Documentation/driver-api/vfio-mediated-device.rst +0 -0
index 25eb7d5b834b..a5bdc60d62a1 100644
@@ -102,12 +102,14 @@ structure to represent a mediated device's

driver::
* @probe: called when new device created
* @remove: called when device removed
* @driver: device driver structure

Documentation/driver-api/vfio-mediated-device.rst +2 -0
+      * @id_table: the ids serviced by this driver
       */
      struct mdev_driver {
 	     const char *name;
 	     int  (*probe)  (struct device *dev);
 	     void (*remove) (struct device *dev);
 	     struct device_driver    driver;
+	     const struct mdev_class_id *id_table;
      };
 
 A mediated bus driver for mdev should use this structure in the function

calls

Documentation/driver-api/vfio-mediated-device.rst +1 -0
@@ -165,6 +167,7 @@ register itself with the mdev core driver::
 	extern int  mdev_register_device(struct device *dev,
 	                                 const struct mdev_parent_ops *ops);
 
+
 However, the mdev_parent_ops structure is not required in the function

call
that a driver should use to unregister itself with the mdev core driver::

drivers/gpu/drm/i915/gvt/kvmgt.c +1 -0
index 23aa3e50cbf8..f793252a3d2a 100644
@@ -678,6 +678,7 @@ static int intel_vgpu_create(struct kobject *kobj,
 struct mdev_device *mdev)
 		     dev_name(mdev_dev(mdev)));
 	ret = 0;
 
+	mdev_set_class_id(mdev, MDEV_ID_VFIO);
 out:
 	return ret;

}

drivers/s390/cio/vfio_ccw_ops.c +1 -0
index f0d71ab77c50..d258ef1fedb9 100644
@@ -129,6 +129,7 @@ static int vfio_ccw_mdev_create(struct kobject
 *kobj, struct mdev_device *mdev)
 			   private->sch->schid.ssid,
 			   private->sch->schid.sch_no);
 
+	mdev_set_class_id(mdev, MDEV_ID_VFIO);
 	return 0;
 }
drivers/s390/crypto/vfio_ap_ops.c +1 -0
index 5c0f53c6dde7..2cfd96112aa0 100644
@@ -343,6 +343,7 @@ static int vfio_ap_mdev_create(struct kobject *kobj,
 struct mdev_device *mdev)
 	list_add(&matrix_mdev->node, &matrix_dev->mdev_list);
 	mutex_unlock(&matrix_dev->lock);
 
+	mdev_set_class_id(mdev, MDEV_ID_VFIO);
 	return 0;
 }
drivers/vfio/mdev/mdev_core.c +6 -0
index b558d4cfd082..8764cf4a276d 100644
@@ -45,6 +45,12 @@ void mdev_set_drvdata(struct mdev_device *mdev,
 void *data)
 }
 EXPORT_SYMBOL(mdev_set_drvdata);
 
+void mdev_set_class_id(struct mdev_device *mdev, u16 id)
+{
+	mdev->class_id = id;
+}
+EXPORT_SYMBOL(mdev_set_class_id);
+
 struct device *mdev_dev(struct mdev_device *mdev)
 {
drivers/vfio/mdev/mdev_core.c +0 -0
 	return &mdev->dev;

drivers/vfio/mdev/mdev_core.c static int mdev_device_remove_cb(struct device · line 145

Quoted 5 lines
drivers/vfio/mdev/mdev_core.c +1 -0
@@ -135,6 +141,7 @@ static int mdev_device_remove_cb(struct device
 *dev, void *data)
  * mdev_register_device : Register a device
  * @dev: device structure representing parent device.
  * @ops: Parent device operation structure to be registered.
+ * @id: device id.

class id.

Quoted 164 lines
drivers/vfio/mdev/mdev_core.c +0 -0
  *
  * Add device to list of registered parent devices.

* Returns a negative value on error, otherwise 0.

drivers/vfio/mdev/mdev_driver.c +14 -0
index 0d3223aee20b..b7c40ce86ee3 100644
@@ -69,8 +69,22 @@ static int mdev_remove(struct device *dev)
 	return 0;
 }
 
+static int mdev_match(struct device *dev, struct device_driver *drv)
+{
+	unsigned int i;
+	struct mdev_device *mdev = to_mdev_device(dev);
+	struct mdev_driver *mdrv = to_mdev_driver(drv);
+	const struct mdev_class_id *ids = mdrv->id_table;
+
+	for (i = 0; ids[i].id; i++)
+		if (ids[i].id == mdev->class_id)
+			return 1;
+	return 0;
+}
+
 struct bus_type mdev_bus_type = {
 	.name		= "mdev",
+	.match		= mdev_match,
 	.probe		= mdev_probe,
 	.remove		= mdev_remove,
 };
drivers/vfio/mdev/mdev_private.h +1 -0
index 7d922950caaf..c65f436c1869 100644
@@ -33,6 +33,7 @@ struct mdev_device {
 	struct kobject *type_kobj;
 	struct device *iommu_device;
 	bool active;
+	u16 class_id;
 };
 
 #define to_mdev_device(dev)	container_of(dev, struct mdev_device, dev)
drivers/vfio/mdev/vfio_mdev.c +6 -0
index 30964a4e0a28..fd2a4d9a3f26 100644
@@ -120,10 +120,16 @@ static void vfio_mdev_remove(struct device *dev)
 	vfio_del_group_dev(dev);
 }
 
+static struct mdev_class_id id_table[] = {
+	{ MDEV_ID_VFIO },
+	{ 0 },
+};
+
 static struct mdev_driver vfio_mdev_driver = {
 	.name	= "vfio_mdev",
 	.probe	= vfio_mdev_probe,
 	.remove	= vfio_mdev_remove,
+	.id_table = id_table,
 };
 
 static int __init vfio_mdev_init(void)
include/linux/mdev.h +1 -0
index 0ce30ca78db0..3974650c074f 100644
@@ -118,6 +118,7 @@ struct mdev_type_attribute
 mdev_type_attr_##_name =		\
  * @probe: called when new device created
  * @remove: called when device removed
  * @driver: device driver structure
+ * @id_table: the ids serviced by this driver.
  *
  **/

struct mdev_driver {

include/linux/mdev.h +7 -0
@@ -125,12 +126,14 @@ struct mdev_driver {
 	int  (*probe)(struct device *dev);
 	void (*remove)(struct device *dev);
 	struct device_driver driver;
+	const struct mdev_class_id *id_table;
 };
 
 #define to_mdev_driver(drv)	container_of(drv, struct mdev_driver, driver)
 
 void *mdev_get_drvdata(struct mdev_device *mdev);
 void mdev_set_drvdata(struct mdev_device *mdev, void *data);
+void mdev_set_class_id(struct mdev_device *mdev, u16 id);
 const guid_t *mdev_uuid(struct mdev_device *mdev);
 
 extern struct bus_type mdev_bus_type;
@@ -145,4 +148,9 @@ struct device *mdev_parent_dev(struct
 mdev_device *mdev);
 struct device *mdev_dev(struct mdev_device *mdev);
 struct mdev_device *mdev_from_dev(struct device *dev);
 
+enum {
+	MDEV_ID_VFIO = 1,
+	/* New entries must be added here */
+};
+

#endif /* MDEV_H */

include/linux/mod_devicetable.h +8 -0
index 5714fd35a83c..f32c6e44fb1a 100644
@@ -821,4 +821,12 @@ struct wmi_device_id {
 	const void *context;
 };
 
+/**
+ * struct mdev_class_id - MDEV device class identifier
+ * @id: Used to identify a specific class of device, e.g vfio-mdev device.
+ */
+struct mdev_class_id {
+	__u16 id;
+};
+
 #endif /* LINUX_MOD_DEVICETABLE_H */
samples/vfio-mdev/mbochs.c +1 -0
index ac5c8c17b1ff..8a8583c892b2 100644
@@ -561,6 +561,7 @@ static int mbochs_create(struct kobject *kobj, struct
 mdev_device *mdev)
 	mbochs_reset(mdev);
 
 	mbochs_used_mbytes += type->mbytes;
+	mdev_set_class_id(mdev, MDEV_ID_VFIO);
 	return 0;
 

err_mem:

samples/vfio-mdev/mdpy.c +1 -0
index cc86bf6566e4..88d7e76f3836 100644
@@ -269,6 +269,7 @@ static int mdpy_create(struct kobject *kobj, struct
 mdev_device *mdev)
 	mdpy_reset(mdev);
 
 	mdpy_count++;
+	mdev_set_class_id(mdev, MDEV_ID_VFIO);
 	return 0;
 }
samples/vfio-mdev/mtty.c +1 -1
index 92e770a06ea2..4e0735143b69 100644
@@ -770,6 +770,7 @@ static int mtty_create(struct kobject *kobj, struct
 mdev_device *mdev)
 	list_add(&mdev_state->next, &mdev_devices_list);
 	mutex_unlock(&mdev_list_lock);
 
+	mdev_set_class_id(mdev, MDEV_ID_VFIO);
 	return 0;
 }
 
--

2.19.1

author
#12 lore raw

PATCH v2 2/8 mdev: class id support

In reply to Jason Wang

Mdev bus only supports vfio driver right now, so it doesn't implement match method. But in the future, we may add drivers other than vfio, the first driver could be virtio-mdev. This means we need to add device class id support in bus match method to pair the mdev device and mdev driver correctly.

So this patch adds id_table to mdev_driver and class_id for mdev parent with the match method for mdev bus.

---
 Documentation/driver-api/vfio-mediated-device.rst |  3 +++
 drivers/gpu/drm/i915/gvt/kvmgt.c                  |  1 +
 drivers/s390/cio/vfio_ccw_ops.c                   |  1 +
 drivers/s390/crypto/vfio_ap_ops.c                 |  1 +
 drivers/vfio/mdev/mdev_core.c                     |  7 +++++++
 drivers/vfio/mdev/mdev_driver.c                   | 14 ++++++++++++++
 drivers/vfio/mdev/mdev_private.h                  |  1 +
 drivers/vfio/mdev/vfio_mdev.c                     |  6 ++++++
 include/linux/mdev.h                              |  8 ++++++++
 include/linux/mod_devicetable.h                   |  8 ++++++++
 samples/vfio-mdev/mbochs.c                        |  1 +
 samples/vfio-mdev/mdpy.c                          |  1 +
 samples/vfio-mdev/mtty.c                          |  1 +
 13 files changed, 53 insertions(+)
Documentation/driver-api/vfio-mediated-device.rst +3 -0
index 25eb7d5b834b..a5bdc60d62a1 100644
@@ -102,12 +102,14 @@ structure to represent a mediated device's driver::
       * @probe: called when new device created
       * @remove: called when device removed
       * @driver: device driver structure
+      * @id_table: the ids serviced by this driver
       */
      struct mdev_driver {
 	     const char *name;
 	     int  (*probe)  (struct device *dev);
 	     void (*remove) (struct device *dev);
 	     struct device_driver    driver;
+	     const struct mdev_class_id *id_table;
      };
 
 A mediated bus driver for mdev should use this structure in the function calls
@@ -165,6 +167,7 @@ register itself with the mdev core driver::
 	extern int  mdev_register_device(struct device *dev,
 	                                 const struct mdev_parent_ops *ops);
 
+
 However, the mdev_parent_ops structure is not required in the function call
 that a driver should use to unregister itself with the mdev core driver::
 
drivers/gpu/drm/i915/gvt/kvmgt.c +1 -0
index 23aa3e50cbf8..f793252a3d2a 100644
@@ -678,6 +678,7 @@ static int intel_vgpu_create(struct kobject *kobj, struct mdev_device *mdev)
 		     dev_name(mdev_dev(mdev)));
 	ret = 0;
 
+	mdev_set_class_id(mdev, MDEV_ID_VFIO);
 out:
 	return ret;
 }
drivers/s390/cio/vfio_ccw_ops.c +1 -0
index f0d71ab77c50..d258ef1fedb9 100644
@@ -129,6 +129,7 @@ static int vfio_ccw_mdev_create(struct kobject *kobj, struct mdev_device *mdev)
 			   private->sch->schid.ssid,
 			   private->sch->schid.sch_no);
 
+	mdev_set_class_id(mdev, MDEV_ID_VFIO);
 	return 0;
 }
 
drivers/s390/crypto/vfio_ap_ops.c +1 -0
index 5c0f53c6dde7..2cfd96112aa0 100644
@@ -343,6 +343,7 @@ static int vfio_ap_mdev_create(struct kobject *kobj, struct mdev_device *mdev)
 	list_add(&matrix_mdev->node, &matrix_dev->mdev_list);
 	mutex_unlock(&matrix_dev->lock);
 
+	mdev_set_class_id(mdev, MDEV_ID_VFIO);
 	return 0;
 }
 
drivers/vfio/mdev/mdev_core.c +7 -0
index b558d4cfd082..8764cf4a276d 100644
@@ -45,6 +45,12 @@ void mdev_set_drvdata(struct mdev_device *mdev, void *data)
 }
 EXPORT_SYMBOL(mdev_set_drvdata);
 
+void mdev_set_class_id(struct mdev_device *mdev, u16 id)
+{
+	mdev->class_id = id;
+}
+EXPORT_SYMBOL(mdev_set_class_id);
+
 struct device *mdev_dev(struct mdev_device *mdev)
 {
 	return &mdev->dev;
@@ -135,6 +141,7 @@ static int mdev_device_remove_cb(struct device *dev, void *data)
  * mdev_register_device : Register a device
  * @dev: device structure representing parent device.
  * @ops: Parent device operation structure to be registered.
+ * @id: device id.
  *
  * Add device to list of registered parent devices.
  * Returns a negative value on error, otherwise 0.
drivers/vfio/mdev/mdev_driver.c +14 -0
index 0d3223aee20b..b7c40ce86ee3 100644
@@ -69,8 +69,22 @@ static int mdev_remove(struct device *dev)
 	return 0;
 }
 
+static int mdev_match(struct device *dev, struct device_driver *drv)
+{
+	unsigned int i;
+	struct mdev_device *mdev = to_mdev_device(dev);
+	struct mdev_driver *mdrv = to_mdev_driver(drv);
+	const struct mdev_class_id *ids = mdrv->id_table;
+
+	for (i = 0; ids[i].id; i++)
+		if (ids[i].id == mdev->class_id)
+			return 1;
+	return 0;
+}
+
 struct bus_type mdev_bus_type = {
 	.name		= "mdev",
+	.match		= mdev_match,
 	.probe		= mdev_probe,
 	.remove		= mdev_remove,
 };
drivers/vfio/mdev/mdev_private.h +1 -0
index 7d922950caaf..c65f436c1869 100644
@@ -33,6 +33,7 @@ struct mdev_device {
 	struct kobject *type_kobj;
 	struct device *iommu_device;
 	bool active;
+	u16 class_id;
 };
 
 #define to_mdev_device(dev)	container_of(dev, struct mdev_device, dev)
drivers/vfio/mdev/vfio_mdev.c +6 -0
index 30964a4e0a28..fd2a4d9a3f26 100644
@@ -120,10 +120,16 @@ static void vfio_mdev_remove(struct device *dev)
 	vfio_del_group_dev(dev);
 }
 
+static struct mdev_class_id id_table[] = {
+	{ MDEV_ID_VFIO },
+	{ 0 },
+};
+
 static struct mdev_driver vfio_mdev_driver = {
 	.name	= "vfio_mdev",
 	.probe	= vfio_mdev_probe,
 	.remove	= vfio_mdev_remove,
+	.id_table = id_table,
 };
 
 static int __init vfio_mdev_init(void)
include/linux/mdev.h +8 -0
index 0ce30ca78db0..3974650c074f 100644
@@ -118,6 +118,7 @@ struct mdev_type_attribute mdev_type_attr_##_name =		\
  * @probe: called when new device created
  * @remove: called when device removed
  * @driver: device driver structure
+ * @id_table: the ids serviced by this driver.
  *
  **/
 struct mdev_driver {
@@ -125,12 +126,14 @@ struct mdev_driver {
 	int  (*probe)(struct device *dev);
 	void (*remove)(struct device *dev);
 	struct device_driver driver;
+	const struct mdev_class_id *id_table;
 };
 
 #define to_mdev_driver(drv)	container_of(drv, struct mdev_driver, driver)
 
 void *mdev_get_drvdata(struct mdev_device *mdev);
 void mdev_set_drvdata(struct mdev_device *mdev, void *data);
+void mdev_set_class_id(struct mdev_device *mdev, u16 id);
 const guid_t *mdev_uuid(struct mdev_device *mdev);
 
 extern struct bus_type mdev_bus_type;
@@ -145,4 +148,9 @@ struct device *mdev_parent_dev(struct mdev_device *mdev);
 struct device *mdev_dev(struct mdev_device *mdev);
 struct mdev_device *mdev_from_dev(struct device *dev);
 
+enum {
+	MDEV_ID_VFIO = 1,
+	/* New entries must be added here */
+};
+
 #endif /* MDEV_H */
include/linux/mod_devicetable.h +8 -0
index 5714fd35a83c..f32c6e44fb1a 100644
@@ -821,4 +821,12 @@ struct wmi_device_id {
 	const void *context;
 };
 
+/**
+ * struct mdev_class_id - MDEV device class identifier
+ * @id: Used to identify a specific class of device, e.g vfio-mdev device.
+ */
+struct mdev_class_id {
+	__u16 id;
+};
+
 #endif /* LINUX_MOD_DEVICETABLE_H */
samples/vfio-mdev/mbochs.c +1 -0
index ac5c8c17b1ff..8a8583c892b2 100644
@@ -561,6 +561,7 @@ static int mbochs_create(struct kobject *kobj, struct mdev_device *mdev)
 	mbochs_reset(mdev);
 
 	mbochs_used_mbytes += type->mbytes;
+	mdev_set_class_id(mdev, MDEV_ID_VFIO);
 	return 0;
 
 err_mem:
samples/vfio-mdev/mdpy.c +1 -0
index cc86bf6566e4..88d7e76f3836 100644
@@ -269,6 +269,7 @@ static int mdpy_create(struct kobject *kobj, struct mdev_device *mdev)
 	mdpy_reset(mdev);
 
 	mdpy_count++;
+	mdev_set_class_id(mdev, MDEV_ID_VFIO);
 	return 0;
 }
 
samples/vfio-mdev/mtty.c +1 -0
index 92e770a06ea2..4e0735143b69 100644
@@ -770,6 +770,7 @@ static int mtty_create(struct kobject *kobj, struct mdev_device *mdev)
 	list_add(&mdev_state->next, &mdev_devices_list);
 	mutex_unlock(&mdev_list_lock);
 
+	mdev_set_class_id(mdev, MDEV_ID_VFIO);
 	return 0;
 }
 
Signature · 1 line
2.19.1
author
#13 lore raw

PATCH v2 3/8 mdev: bus uevent support

In reply to Jason Wang

This patch adds bus uevent support for mdev bus in order to allow cooperation with userspace.

---
 drivers/vfio/mdev/mdev_driver.c | 8 ++++++++
 1 file changed, 8 insertions(+)
drivers/vfio/mdev/mdev_driver.c +8 -0
index b7c40ce86ee3..319d886ffaf7 100644
@@ -82,9 +82,17 @@ static int mdev_match(struct device *dev, struct device_driver *drv)
 	return 0;
 }
 
+static int mdev_uevent(struct device *dev, struct kobj_uevent_env *env)
+{
+	struct mdev_device *mdev = to_mdev_device(dev);
+
+	return add_uevent_var(env, "MODALIAS=mdev:c%02X", mdev->class_id);
+}
+
 struct bus_type mdev_bus_type = {
 	.name		= "mdev",
 	.match		= mdev_match,
+	.uevent		= mdev_uevent,
 	.probe		= mdev_probe,
 	.remove		= mdev_remove,
 };
Signature · 6 lines
2.19.1

_______________________________________________
Intel-gfx mailing list
[email protected]
https://lists.freedesktop.org/mailman/listinfo/intel-gfx
author
#14 lore raw

PATCH v2 3/8 mdev: bus uevent support

In reply to Jason Wang

This patch adds bus uevent support for mdev bus in order to allow cooperation with userspace.

---
 drivers/vfio/mdev/mdev_driver.c | 8 ++++++++
 1 file changed, 8 insertions(+)
drivers/vfio/mdev/mdev_driver.c +8 -0
index b7c40ce86ee3..319d886ffaf7 100644
@@ -82,9 +82,17 @@ static int mdev_match(struct device *dev, struct device_driver *drv)
 	return 0;
 }
 
+static int mdev_uevent(struct device *dev, struct kobj_uevent_env *env)
+{
+	struct mdev_device *mdev = to_mdev_device(dev);
+
+	return add_uevent_var(env, "MODALIAS=mdev:c%02X", mdev->class_id);
+}
+
 struct bus_type mdev_bus_type = {
 	.name		= "mdev",
 	.match		= mdev_match,
+	.uevent		= mdev_uevent,
 	.probe		= mdev_probe,
 	.remove		= mdev_remove,
 };
Signature · 1 line
2.19.1
author
#15 lore raw

PATCH v2 4/8 modpost: add support for mdev class id

In reply to Jason Wang

Add support to parse mdev class id table.

---
 drivers/vfio/mdev/vfio_mdev.c     |  2 ++
 scripts/mod/devicetable-offsets.c |  3 +++
 scripts/mod/file2alias.c          | 10 ++++++++++
 3 files changed, 15 insertions(+)
drivers/vfio/mdev/vfio_mdev.c +2 -0
index fd2a4d9a3f26..891cf83a2d9a 100644
@@ -125,6 +125,8 @@ static struct mdev_class_id id_table[] = {
 	{ 0 },
 };
 
+MODULE_DEVICE_TABLE(mdev, id_table);
+
 static struct mdev_driver vfio_mdev_driver = {
 	.name	= "vfio_mdev",
 	.probe	= vfio_mdev_probe,
scripts/mod/devicetable-offsets.c +3 -0
index 054405b90ba4..6cbb1062488a 100644
@@ -231,5 +231,8 @@ int main(void)
 	DEVID(wmi_device_id);
 	DEVID_FIELD(wmi_device_id, guid_string);
 
+	DEVID(mdev_class_id);
+	DEVID_FIELD(mdev_class_id, id);
+
 	return 0;
 }
scripts/mod/file2alias.c +10 -0
index e17a29ae2e97..6805f87a1149 100644
@@ -1330,6 +1330,15 @@ static int do_wmi_entry(const char *filename, void *symval, char *alias)
 	return 1;
 }
 
+/* looks like: "mdev:cN" */
+static int do_mdev_entry(const char *filename, void *symval, char *alias)
+{
+	DEF_FIELD(symval, mdev_class_id, id);
+
+	sprintf(alias, "mdev:c%02X", id);
+	return 1;
+}
+
 /* Does namelen bytes of name exactly match the symbol? */
 static bool sym_is(const char *name, unsigned namelen, const char *symbol)
 {
@@ -1402,6 +1411,7 @@ static const struct devtable devtable[] = {
 	{"typec", SIZE_typec_device_id, do_typec_entry},
 	{"tee", SIZE_tee_client_device_id, do_tee_entry},
 	{"wmi", SIZE_wmi_device_id, do_wmi_entry},
+	{"mdev", SIZE_mdev_class_id, do_mdev_entry},
 };
 
 /* Create MODULE_ALIAS() statements.
Signature · 1 line
2.19.1
author
#16 lore raw

PATCH v2 4/8 modpost: add support for mdev class id

In reply to Jason Wang

Add support to parse mdev class id table.

---
 drivers/vfio/mdev/vfio_mdev.c     |  2 ++
 scripts/mod/devicetable-offsets.c |  3 +++
 scripts/mod/file2alias.c          | 10 ++++++++++
 3 files changed, 15 insertions(+)
drivers/vfio/mdev/vfio_mdev.c +2 -0
index fd2a4d9a3f26..891cf83a2d9a 100644
@@ -125,6 +125,8 @@ static struct mdev_class_id id_table[] = {
 	{ 0 },
 };
 
+MODULE_DEVICE_TABLE(mdev, id_table);
+
 static struct mdev_driver vfio_mdev_driver = {
 	.name	= "vfio_mdev",
 	.probe	= vfio_mdev_probe,
scripts/mod/devicetable-offsets.c +3 -0
index 054405b90ba4..6cbb1062488a 100644
@@ -231,5 +231,8 @@ int main(void)
 	DEVID(wmi_device_id);
 	DEVID_FIELD(wmi_device_id, guid_string);
 
+	DEVID(mdev_class_id);
+	DEVID_FIELD(mdev_class_id, id);
+
 	return 0;
 }
scripts/mod/file2alias.c +10 -0
index e17a29ae2e97..6805f87a1149 100644
@@ -1330,6 +1330,15 @@ static int do_wmi_entry(const char *filename, void *symval, char *alias)
 	return 1;
 }
 
+/* looks like: "mdev:cN" */
+static int do_mdev_entry(const char *filename, void *symval, char *alias)
+{
+	DEF_FIELD(symval, mdev_class_id, id);
+
+	sprintf(alias, "mdev:c%02X", id);
+	return 1;
+}
+
 /* Does namelen bytes of name exactly match the symbol? */
 static bool sym_is(const char *name, unsigned namelen, const char *symbol)
 {
@@ -1402,6 +1411,7 @@ static const struct devtable devtable[] = {
 	{"typec", SIZE_typec_device_id, do_typec_entry},
 	{"tee", SIZE_tee_client_device_id, do_tee_entry},
 	{"wmi", SIZE_wmi_device_id, do_wmi_entry},
+	{"mdev", SIZE_mdev_class_id, do_mdev_entry},
 };
 
 /* Create MODULE_ALIAS() statements.
Signature · 1 line
2.19.1
author
#17 lore raw

PATCH v2 5/8 mdev: introduce device specific ops

In reply to Jason Wang

Currently, except for the create and remove, the rest of mdev_parent_ops is designed for vfio-mdev driver only and may not help for kernel mdev driver. With the help of class id, this patch introduces device specific callbacks inside mdev_device structure. This allows different set of callback to be used by vfio-mdev and virtio-mdev.

---
 .../driver-api/vfio-mediated-device.rst       |  4 +-
 MAINTAINERS                                   |  1 +
 drivers/gpu/drm/i915/gvt/kvmgt.c              | 17 +++---
 drivers/s390/cio/vfio_ccw_ops.c               | 17 ++++--
 drivers/s390/crypto/vfio_ap_ops.c             | 13 +++--
 drivers/vfio/mdev/mdev_core.c                 | 12 +++++
 drivers/vfio/mdev/mdev_private.h              |  1 +
 drivers/vfio/mdev/vfio_mdev.c                 | 37 ++++++-------
 include/linux/mdev.h                          | 42 ++++-----------
 include/linux/vfio_mdev.h                     | 52 +++++++++++++++++++
 samples/vfio-mdev/mbochs.c                    | 19 ++++---
 samples/vfio-mdev/mdpy.c                      | 19 ++++---
 samples/vfio-mdev/mtty.c                      | 17 ++++--
 13 files changed, 168 insertions(+), 83 deletions(-)
 create mode 100644 include/linux/vfio_mdev.h
Documentation/driver-api/vfio-mediated-device.rst +3 -1
index a5bdc60d62a1..d50425b368bb 100644
@@ -152,7 +152,9 @@ callbacks per mdev parent device, per mdev type, or any other categorization.
 Vendor drivers are expected to be fully asynchronous in this respect or
 provide their own internal resource protection.)
 
-The callbacks in the mdev_parent_ops structure are as follows:
+The device specific callbacks are referred through device_ops pointer
+in mdev_parent_ops. For vfio-mdev device, its callbacks in device_ops
+are as follows:
 
 * open: open callback of mediated device
 * close: close callback of mediated device
MAINTAINERS +1 -0
index b2326dece28e..89832b316500 100644
@@ -17075,6 +17075,7 @@ S:	Maintained
 F:	Documentation/driver-api/vfio-mediated-device.rst
 F:	drivers/vfio/mdev/
 F:	include/linux/mdev.h
+F:	include/linux/vfio_mdev.h
 F:	samples/vfio-mdev/
 
 VFIO PLATFORM DRIVER
drivers/gpu/drm/i915/gvt/kvmgt.c +11 -6
index f793252a3d2a..b274f5ee481f 100644
@@ -42,6 +42,7 @@
 #include <linux/kvm_host.h>
 #include <linux/vfio.h>
 #include <linux/mdev.h>
+#include <linux/vfio_mdev.h>
 #include <linux/debugfs.h>
 
 #include <linux/nospec.h>
@@ -643,6 +644,8 @@ static void kvmgt_put_vfio_device(void *vgpu)
 	vfio_device_put(((struct intel_vgpu *)vgpu)->vdev.vfio_device);
 }
 
+static struct vfio_mdev_device_ops intel_vfio_vgpu_dev_ops;
+
 static int intel_vgpu_create(struct kobject *kobj, struct mdev_device *mdev)
 {
 	struct intel_vgpu *vgpu = NULL;
@@ -679,6 +682,7 @@ static int intel_vgpu_create(struct kobject *kobj, struct mdev_device *mdev)
 	ret = 0;
 
 	mdev_set_class_id(mdev, MDEV_ID_VFIO);
+	mdev_set_dev_ops(mdev, &intel_vfio_vgpu_dev_ops);
 out:
 	return ret;
 }
@@ -1601,20 +1605,21 @@ static const struct attribute_group *intel_vgpu_groups[] = {
 	NULL,
 };
 
-static struct mdev_parent_ops intel_vgpu_ops = {
-	.mdev_attr_groups       = intel_vgpu_groups,
-	.create			= intel_vgpu_create,
-	.remove			= intel_vgpu_remove,
-
+static struct vfio_mdev_device_ops intel_vfio_vgpu_dev_ops = {
 	.open			= intel_vgpu_open,
 	.release		= intel_vgpu_release,
-
 	.read			= intel_vgpu_read,
 	.write			= intel_vgpu_write,
 	.mmap			= intel_vgpu_mmap,
 	.ioctl			= intel_vgpu_ioctl,
 };
 
+static struct mdev_parent_ops intel_vgpu_ops = {
+	.mdev_attr_groups       = intel_vgpu_groups,
+	.create			= intel_vgpu_create,
+	.remove			= intel_vgpu_remove,
+};
+
 static int kvmgt_host_init(struct device *dev, void *gvt, const void *ops)
 {
 	struct attribute **kvm_type_attrs;
drivers/s390/cio/vfio_ccw_ops.c +12 -5
index d258ef1fedb9..329d53c1f46b 100644
@@ -12,6 +12,7 @@
 
 #include <linux/vfio.h>
 #include <linux/mdev.h>
+#include <linux/vfio_mdev.h>
 #include <linux/nospec.h>
 #include <linux/slab.h>
 
@@ -110,6 +111,8 @@ static struct attribute_group *mdev_type_groups[] = {
 	NULL,
 };
 
+static const struct vfio_mdev_device_ops vfio_mdev_ops;
+
 static int vfio_ccw_mdev_create(struct kobject *kobj, struct mdev_device *mdev)
 {
 	struct vfio_ccw_private *private =
@@ -130,6 +133,7 @@ static int vfio_ccw_mdev_create(struct kobject *kobj, struct mdev_device *mdev)
 			   private->sch->schid.sch_no);
 
 	mdev_set_class_id(mdev, MDEV_ID_VFIO);
+	mdev_set_dev_ops(mdev, &vfio_mdev_ops);
 	return 0;
 }
 
@@ -575,11 +579,7 @@ static ssize_t vfio_ccw_mdev_ioctl(struct mdev_device *mdev,
 	}
 }
 
-static const struct mdev_parent_ops vfio_ccw_mdev_ops = {
-	.owner			= THIS_MODULE,
-	.supported_type_groups  = mdev_type_groups,
-	.create			= vfio_ccw_mdev_create,
-	.remove			= vfio_ccw_mdev_remove,
+static const struct vfio_mdev_device_ops vfio_mdev_ops = {
 	.open			= vfio_ccw_mdev_open,
 	.release		= vfio_ccw_mdev_release,
 	.read			= vfio_ccw_mdev_read,
@@ -587,6 +587,13 @@ static const struct mdev_parent_ops vfio_ccw_mdev_ops = {
 	.ioctl			= vfio_ccw_mdev_ioctl,
 };
 
+static const struct mdev_parent_ops vfio_ccw_mdev_ops = {
+	.owner			= THIS_MODULE,
+	.supported_type_groups  = mdev_type_groups,
+	.create			= vfio_ccw_mdev_create,
+	.remove			= vfio_ccw_mdev_remove,
+};
+
 int vfio_ccw_mdev_reg(struct subchannel *sch)
 {
 	return mdev_register_device(&sch->dev, &vfio_ccw_mdev_ops);
drivers/s390/crypto/vfio_ap_ops.c +10 -3
index 2cfd96112aa0..3a89933f0d3e 100644
@@ -16,6 +16,7 @@
 #include <linux/bitops.h>
 #include <linux/kvm_host.h>
 #include <linux/module.h>
+#include <linux/vfio_mdev.h>
 #include <asm/kvm.h>
 #include <asm/zcrypt.h>
 
@@ -321,6 +322,8 @@ static void vfio_ap_matrix_init(struct ap_config_info *info,
 	matrix->adm_max = info->apxa ? info->Nd : 15;
 }
 
+static const struct vfio_mdev_device_ops vfio_mdev_ops;
+
 static int vfio_ap_mdev_create(struct kobject *kobj, struct mdev_device *mdev)
 {
 	struct ap_matrix_mdev *matrix_mdev;
@@ -344,6 +347,7 @@ static int vfio_ap_mdev_create(struct kobject *kobj, struct mdev_device *mdev)
 	mutex_unlock(&matrix_dev->lock);
 
 	mdev_set_class_id(mdev, MDEV_ID_VFIO);
+	mdev_set_dev_ops(mdev, &vfio_mdev_ops);
 	return 0;
 }
 
@@ -1281,15 +1285,18 @@ static ssize_t vfio_ap_mdev_ioctl(struct mdev_device *mdev,
 	return ret;
 }
 
+static const struct vfio_mdev_device_ops vfio_mdev_ops = {
+	.open			= vfio_ap_mdev_open,
+	.release		= vfio_ap_mdev_release,
+	.ioctl			= vfio_ap_mdev_ioctl,
+};
+
 static const struct mdev_parent_ops vfio_ap_matrix_ops = {
 	.owner			= THIS_MODULE,
 	.supported_type_groups	= vfio_ap_mdev_type_groups,
 	.mdev_attr_groups	= vfio_ap_mdev_attr_groups,
 	.create			= vfio_ap_mdev_create,
 	.remove			= vfio_ap_mdev_remove,
-	.open			= vfio_ap_mdev_open,
-	.release		= vfio_ap_mdev_release,
-	.ioctl			= vfio_ap_mdev_ioctl,
 };
 
 int vfio_ap_mdev_register(void)
drivers/vfio/mdev/mdev_core.c +12 -0
index 8764cf4a276d..6f35f2ced2c9 100644
@@ -51,6 +51,18 @@ void mdev_set_class_id(struct mdev_device *mdev, u16 id)
 }
 EXPORT_SYMBOL(mdev_set_class_id);
 
+const void *mdev_get_dev_ops(struct mdev_device *mdev)
+{
+	return mdev->device_ops;
+}
+EXPORT_SYMBOL(mdev_get_dev_ops);
+
+void mdev_set_dev_ops(struct mdev_device *mdev, const void *ops)
+{
+	mdev->device_ops = ops;
+}
+EXPORT_SYMBOL(mdev_set_dev_ops);
+
 struct device *mdev_dev(struct mdev_device *mdev)
 {
 	return &mdev->dev;
drivers/vfio/mdev/mdev_private.h +1 -0
index c65f436c1869..b666805f0b1f 100644
@@ -34,6 +34,7 @@ struct mdev_device {
 	struct device *iommu_device;
 	bool active;
 	u16 class_id;
+	const void *device_ops;
 };
 
 #define to_mdev_device(dev)	container_of(dev, struct mdev_device, dev)
drivers/vfio/mdev/vfio_mdev.c +19 -18
index 891cf83a2d9a..95efa054442f 100644
@@ -14,6 +14,7 @@
 #include <linux/slab.h>
 #include <linux/vfio.h>
 #include <linux/mdev.h>
+#include <linux/vfio_mdev.h>
 
 #include "mdev_private.h"
 
@@ -24,16 +25,16 @@
 static int vfio_mdev_open(void *device_data)
 {
 	struct mdev_device *mdev = device_data;
-	struct mdev_parent *parent = mdev->parent;
+	const struct vfio_mdev_device_ops *ops = mdev_get_dev_ops(mdev);
 	int ret;
 
-	if (unlikely(!parent->ops->open))
+	if (unlikely(!ops->open))
 		return -EINVAL;
 
 	if (!try_module_get(THIS_MODULE))
 		return -ENODEV;
 
-	ret = parent->ops->open(mdev);
+	ret = ops->open(mdev);
 	if (ret)
 		module_put(THIS_MODULE);
 
@@ -43,10 +44,10 @@ static int vfio_mdev_open(void *device_data)
 static void vfio_mdev_release(void *device_data)
 {
 	struct mdev_device *mdev = device_data;
-	struct mdev_parent *parent = mdev->parent;
+	const struct vfio_mdev_device_ops *ops = mdev_get_dev_ops(mdev);
 
-	if (likely(parent->ops->release))
-		parent->ops->release(mdev);
+	if (likely(ops->release))
+		ops->release(mdev);
 
 	module_put(THIS_MODULE);
 }
@@ -55,47 +56,47 @@ static long vfio_mdev_unlocked_ioctl(void *device_data,
 				     unsigned int cmd, unsigned long arg)
 {
 	struct mdev_device *mdev = device_data;
-	struct mdev_parent *parent = mdev->parent;
+	const struct vfio_mdev_device_ops *ops = mdev_get_dev_ops(mdev);
 
-	if (unlikely(!parent->ops->ioctl))
+	if (unlikely(!ops->ioctl))
 		return -EINVAL;
 
-	return parent->ops->ioctl(mdev, cmd, arg);
+	return ops->ioctl(mdev, cmd, arg);
 }
 
 static ssize_t vfio_mdev_read(void *device_data, char __user *buf,
 			      size_t count, loff_t *ppos)
 {
 	struct mdev_device *mdev = device_data;
-	struct mdev_parent *parent = mdev->parent;
+	const struct vfio_mdev_device_ops *ops = mdev_get_dev_ops(mdev);
 
-	if (unlikely(!parent->ops->read))
+	if (unlikely(!ops->read))
 		return -EINVAL;
 
-	return parent->ops->read(mdev, buf, count, ppos);
+	return ops->read(mdev, buf, count, ppos);
 }
 
 static ssize_t vfio_mdev_write(void *device_data, const char __user *buf,
 			       size_t count, loff_t *ppos)
 {
 	struct mdev_device *mdev = device_data;
-	struct mdev_parent *parent = mdev->parent;
+	const struct vfio_mdev_device_ops *ops = mdev_get_dev_ops(mdev);
 
-	if (unlikely(!parent->ops->write))
+	if (unlikely(!ops->write))
 		return -EINVAL;
 
-	return parent->ops->write(mdev, buf, count, ppos);
+	return ops->write(mdev, buf, count, ppos);
 }
 
 static int vfio_mdev_mmap(void *device_data, struct vm_area_struct *vma)
 {
 	struct mdev_device *mdev = device_data;
-	struct mdev_parent *parent = mdev->parent;
+	const struct vfio_mdev_device_ops *ops = mdev_get_dev_ops(mdev);
 
-	if (unlikely(!parent->ops->mmap))
+	if (unlikely(!ops->mmap))
 		return -EINVAL;
 
-	return parent->ops->mmap(mdev, vma);
+	return ops->mmap(mdev, vma);
 }
 
 static const struct vfio_device_ops vfio_mdev_dev_ops = {
include/linux/mdev.h +9 -33

66 lines not on this page · raw message

include/linux/vfio_mdev.h new file +52 -0

52 lines not on this page · raw message

samples/vfio-mdev/mbochs.c +13 -6
index 8a8583c892b2..b4bf29c6136c 100644
@@ -30,6 +30,7 @@
 #include <linux/iommu.h>
 #include <linux/sysfs.h>
 #include <linux/mdev.h>
+#include <linux/vfio_mdev.h>
 #include <linux/pci.h>
 #include <linux/dma-buf.h>
 #include <linux/highmem.h>
@@ -516,6 +517,8 @@ static int mbochs_reset(struct mdev_device *mdev)
 	return 0;
 }
 
+static const struct vfio_mdev_device_ops vfio_mdev_ops;
+
 static int mbochs_create(struct kobject *kobj, struct mdev_device *mdev)
 {
 	const struct mbochs_type *type = mbochs_find_type(kobj);
@@ -562,6 +565,7 @@ static int mbochs_create(struct kobject *kobj, struct mdev_device *mdev)
 
 	mbochs_used_mbytes += type->mbytes;
 	mdev_set_class_id(mdev, MDEV_ID_VFIO);
+	mdev_set_dev_ops(mdev, &vfio_mdev_ops);
 	return 0;
 
 err_mem:
@@ -1419,12 +1423,7 @@ static struct attribute_group *mdev_type_groups[] = {
 	NULL,
 };
 
-static const struct mdev_parent_ops mdev_fops = {
-	.owner			= THIS_MODULE,
-	.mdev_attr_groups	= mdev_dev_groups,
-	.supported_type_groups	= mdev_type_groups,
-	.create			= mbochs_create,
-	.remove			= mbochs_remove,
+static const struct vfio_mdev_device_ops vfio_mdev_ops = {
 	.open			= mbochs_open,
 	.release		= mbochs_close,
 	.read			= mbochs_read,
@@ -1433,6 +1432,14 @@ static const struct mdev_parent_ops mdev_fops = {
 	.mmap			= mbochs_mmap,
 };
 
+static const struct mdev_parent_ops mdev_fops = {
+	.owner			= THIS_MODULE,
+	.mdev_attr_groups	= mdev_dev_groups,
+	.supported_type_groups	= mdev_type_groups,
+	.create			= mbochs_create,
+	.remove			= mbochs_remove,
+};
+
 static const struct file_operations vd_fops = {
 	.owner		= THIS_MODULE,
 };
samples/vfio-mdev/mdpy.c +13 -6
index 88d7e76f3836..80c2df531326 100644
@@ -26,6 +26,7 @@
 #include <linux/iommu.h>
 #include <linux/sysfs.h>
 #include <linux/mdev.h>
+#include <linux/vfio_mdev.h>
 #include <linux/pci.h>
 #include <drm/drm_fourcc.h>
 #include "mdpy-defs.h"
@@ -226,6 +227,8 @@ static int mdpy_reset(struct mdev_device *mdev)
 	return 0;
 }
 
+static const struct vfio_mdev_device_ops vfio_mdev_ops;
+
 static int mdpy_create(struct kobject *kobj, struct mdev_device *mdev)
 {
 	const struct mdpy_type *type = mdpy_find_type(kobj);
@@ -270,6 +273,7 @@ static int mdpy_create(struct kobject *kobj, struct mdev_device *mdev)
 
 	mdpy_count++;
 	mdev_set_class_id(mdev, MDEV_ID_VFIO);
+	mdev_set_dev_ops(mdev, &vfio_mdev_ops);
 	return 0;
 }
 
@@ -726,12 +730,7 @@ static struct attribute_group *mdev_type_groups[] = {
 	NULL,
 };
 
-static const struct mdev_parent_ops mdev_fops = {
-	.owner			= THIS_MODULE,
-	.mdev_attr_groups	= mdev_dev_groups,
-	.supported_type_groups	= mdev_type_groups,
-	.create			= mdpy_create,
-	.remove			= mdpy_remove,
+static const struct vfio_mdev_device_ops vfio_mdev_ops = {
 	.open			= mdpy_open,
 	.release		= mdpy_close,
 	.read			= mdpy_read,
@@ -740,6 +739,14 @@ static const struct mdev_parent_ops mdev_fops = {
 	.mmap			= mdpy_mmap,
 };
 
+static const struct mdev_parent_ops mdev_fops = {
+	.owner			= THIS_MODULE,
+	.mdev_attr_groups	= mdev_dev_groups,
+	.supported_type_groups	= mdev_type_groups,
+	.create			= mdpy_create,
+	.remove			= mdpy_remove,
+};
+
 static const struct file_operations vd_fops = {
 	.owner		= THIS_MODULE,
 };
samples/vfio-mdev/mtty.c +12 -5
index 4e0735143b69..2db860ccb02c 100644
@@ -27,6 +27,7 @@
 #include <linux/ctype.h>
 #include <linux/file.h>
 #include <linux/mdev.h>
+#include <linux/vfio_mdev.h>
 #include <linux/pci.h>
 #include <linux/serial.h>
 #include <uapi/linux/serial_reg.h>
@@ -723,6 +724,8 @@ static ssize_t mdev_access(struct mdev_device *mdev, u8 *buf, size_t count,
 	return ret;
 }
 
+static const struct vfio_mdev_device_ops vfio_dev_ops;
+
 static int mtty_create(struct kobject *kobj, struct mdev_device *mdev)
 {
 	struct mdev_state *mdev_state;
@@ -771,6 +774,7 @@ static int mtty_create(struct kobject *kobj, struct mdev_device *mdev)
 	mutex_unlock(&mdev_list_lock);
 
 	mdev_set_class_id(mdev, MDEV_ID_VFIO);
+	mdev_set_dev_ops(mdev, &vfio_dev_ops);
 	return 0;
 }
 
@@ -1411,6 +1415,14 @@ static struct attribute_group *mdev_type_groups[] = {
 	NULL,
 };
 
+static const struct vfio_mdev_device_ops vfio_dev_ops = {
+	.open                   = mtty_open,
+	.release                = mtty_close,
+	.read                   = mtty_read,
+	.write                  = mtty_write,
+	.ioctl		        = mtty_ioctl,
+};
+
 static const struct mdev_parent_ops mdev_fops = {
 	.owner                  = THIS_MODULE,
 	.dev_attr_groups        = mtty_dev_groups,
@@ -1418,11 +1430,6 @@ static const struct mdev_parent_ops mdev_fops = {
 	.supported_type_groups  = mdev_type_groups,
 	.create                 = mtty_create,
 	.remove			= mtty_remove,
-	.open                   = mtty_open,
-	.release                = mtty_close,
-	.read                   = mtty_read,
-	.write                  = mtty_write,
-	.ioctl		        = mtty_ioctl,
 };
 
 static void mtty_device_release(struct device *dev)
Signature · 1 line
2.19.1
#18 lore raw

Re: [PATCH V2 5/8] mdev: introduce device specific ops

Quoted 30 lines · On Tue, 24 Sep 2019 21:53:29 +0800 Jason Wang <[email protected]> wrote:

Currently, except for the create and remove, the rest of mdev_parent_ops is designed for vfio-mdev driver only and may not help for kernel mdev driver. With the help of class id, this patch introduces device specific callbacks inside mdev_device structure. This allows different set of callback to be used by vfio-mdev and virtio-mdev.

---
 .../driver-api/vfio-mediated-device.rst       |  4 +-
 MAINTAINERS                                   |  1 +
 drivers/gpu/drm/i915/gvt/kvmgt.c              | 17 +++---
 drivers/s390/cio/vfio_ccw_ops.c               | 17 ++++--
 drivers/s390/crypto/vfio_ap_ops.c             | 13 +++--
 drivers/vfio/mdev/mdev_core.c                 | 12 +++++
 drivers/vfio/mdev/mdev_private.h              |  1 +
 drivers/vfio/mdev/vfio_mdev.c                 | 37 ++++++-------
 include/linux/mdev.h                          | 42 ++++-----------
 include/linux/vfio_mdev.h                     | 52 +++++++++++++++++++
 samples/vfio-mdev/mbochs.c                    | 19 ++++---
 samples/vfio-mdev/mdpy.c                      | 19 ++++---
 samples/vfio-mdev/mtty.c                      | 17 ++++--
 13 files changed, 168 insertions(+), 83 deletions(-)
 create mode 100644 include/linux/vfio_mdev.h
Documentation/driver-api/vfio-mediated-device.rst +0 -0
index a5bdc60d62a1..d50425b368bb 100644

Documentation/driver-api/vfio-mediated-device.rst callbacks per mdev parent device, per mdev type, or any other categorization. · line 157

Quoted 7 lines
Documentation/driver-api/vfio-mediated-device.rst +3 -1
@@ -152,7 +152,9 @@ callbacks per mdev parent device, per mdev type, or any other categorization.
 Vendor drivers are expected to be fully asynchronous in this respect or
 provide their own internal resource protection.)
 
-The callbacks in the mdev_parent_ops structure are as follows:
+The device specific callbacks are referred through device_ops pointer
+in mdev_parent_ops. For vfio-mdev device, its callbacks in device_ops
+are as follows:

This is not accurate. device_ops is now on the mdev_device and is an mdev bus driver specific structure of callbacks that must be registered for each mdev device by the parent driver during the create callback. There's a one to one mapping of class_id to mdev_device_ops callbacks.

That also suggests to me that we could be more clever in registering both of these with mdev-core. Can we embed the class_id in the ops structure in a common way so that the core can extract it and the bus drivers can access their specific callbacks?

Quoted 36 lines
Documentation/driver-api/vfio-mediated-device.rst +0 -0
 * open: open callback of mediated device
 * close: close callback of mediated device
MAINTAINERS +1 -0
index b2326dece28e..89832b316500 100644
@@ -17075,6 +17075,7 @@ S:	Maintained
 F:	Documentation/driver-api/vfio-mediated-device.rst
 F:	drivers/vfio/mdev/
 F:	include/linux/mdev.h
+F:	include/linux/vfio_mdev.h
 F:	samples/vfio-mdev/
 
 VFIO PLATFORM DRIVER
drivers/gpu/drm/i915/gvt/kvmgt.c +3 -0
index f793252a3d2a..b274f5ee481f 100644
@@ -42,6 +42,7 @@
 #include <linux/kvm_host.h>
 #include <linux/vfio.h>
 #include <linux/mdev.h>
+#include <linux/vfio_mdev.h>
 #include <linux/debugfs.h>
 
 #include <linux/nospec.h>
@@ -643,6 +644,8 @@ static void kvmgt_put_vfio_device(void *vgpu)
 	vfio_device_put(((struct intel_vgpu *)vgpu)->vdev.vfio_device);
 }
 
+static struct vfio_mdev_device_ops intel_vfio_vgpu_dev_ops;
+
 static int intel_vgpu_create(struct kobject *kobj, struct mdev_device *mdev)
 {
 	struct intel_vgpu *vgpu = NULL;

drivers/gpu/drm/i915/gvt/kvmgt.c static int intel_vgpu_create(struct kobject *kobj, struct mdev_device *mdev) · line 685

Quoted 4 lines
drivers/gpu/drm/i915/gvt/kvmgt.c +1 -0
@@ -679,6 +682,7 @@ static int intel_vgpu_create(struct kobject *kobj, struct mdev_device *mdev)
 	ret = 0;
 
 	mdev_set_class_id(mdev, MDEV_ID_VFIO);
+	mdev_set_dev_ops(mdev, &intel_vfio_vgpu_dev_ops);

This seems rather unrefined. We're registering interdependent data in separate calls. All drivers need to make both of these calls. I'm not sure if this is a good idea, but what if we had:

static const struct vfio_mdev_device_ops intel_vfio_vgpu_dev_ops = {
	.id			= MDEV_ID_VFIO,
 	.open			= intel_vgpu_open,
 	.release		= intel_vgpu_release,
        ...

And the set function passed &intel_vfio_vgpu_dev_ops.id and the mdev bus drivers used container_of to get to their callbacks?

Quoted 13 lines
drivers/gpu/drm/i915/gvt/kvmgt.c +1 -5
 out:
 	return ret;
 }
@@ -1601,20 +1605,21 @@ static const struct attribute_group *intel_vgpu_groups[] = {
 	NULL,
 };
 
-static struct mdev_parent_ops intel_vgpu_ops = {
-	.mdev_attr_groups       = intel_vgpu_groups,
-	.create			= intel_vgpu_create,
-	.remove			= intel_vgpu_remove,
-
+static struct vfio_mdev_device_ops intel_vfio_vgpu_dev_ops = {

drivers/gpu/drm/i915/gvt/kvmgt.c static const struct attribute_group *intel_vgpu_groups[] = { · line 1617

Quoted 10 lines
drivers/gpu/drm/i915/gvt/kvmgt.c +1 -1
 	.open			= intel_vgpu_open,
 	.release		= intel_vgpu_release,
-
 	.read			= intel_vgpu_read,
 	.write			= intel_vgpu_write,
 	.mmap			= intel_vgpu_mmap,
 	.ioctl			= intel_vgpu_ioctl,
 };
 
+static struct mdev_parent_ops intel_vgpu_ops = {

These could maybe be made const at the same time. Thanks,

Alex

Quoted 560 lines
drivers/gpu/drm/i915/gvt/kvmgt.c +5 -0
+	.mdev_attr_groups       = intel_vgpu_groups,
+	.create			= intel_vgpu_create,
+	.remove			= intel_vgpu_remove,
+};
+
 static int kvmgt_host_init(struct device *dev, void *gvt, const void *ops)
 {
 	struct attribute **kvm_type_attrs;
drivers/s390/cio/vfio_ccw_ops.c +12 -5
index d258ef1fedb9..329d53c1f46b 100644
@@ -12,6 +12,7 @@
 
 #include <linux/vfio.h>
 #include <linux/mdev.h>
+#include <linux/vfio_mdev.h>
 #include <linux/nospec.h>
 #include <linux/slab.h>
 
@@ -110,6 +111,8 @@ static struct attribute_group *mdev_type_groups[] = {
 	NULL,
 };
 
+static const struct vfio_mdev_device_ops vfio_mdev_ops;
+
 static int vfio_ccw_mdev_create(struct kobject *kobj, struct mdev_device *mdev)
 {
 	struct vfio_ccw_private *private =
@@ -130,6 +133,7 @@ static int vfio_ccw_mdev_create(struct kobject *kobj, struct mdev_device *mdev)
 			   private->sch->schid.sch_no);
 
 	mdev_set_class_id(mdev, MDEV_ID_VFIO);
+	mdev_set_dev_ops(mdev, &vfio_mdev_ops);
 	return 0;
 }
 
@@ -575,11 +579,7 @@ static ssize_t vfio_ccw_mdev_ioctl(struct mdev_device *mdev,
 	}
 }
 
-static const struct mdev_parent_ops vfio_ccw_mdev_ops = {
-	.owner			= THIS_MODULE,
-	.supported_type_groups  = mdev_type_groups,
-	.create			= vfio_ccw_mdev_create,
-	.remove			= vfio_ccw_mdev_remove,
+static const struct vfio_mdev_device_ops vfio_mdev_ops = {
 	.open			= vfio_ccw_mdev_open,
 	.release		= vfio_ccw_mdev_release,
 	.read			= vfio_ccw_mdev_read,
@@ -587,6 +587,13 @@ static const struct mdev_parent_ops vfio_ccw_mdev_ops = {
 	.ioctl			= vfio_ccw_mdev_ioctl,
 };
 
+static const struct mdev_parent_ops vfio_ccw_mdev_ops = {
+	.owner			= THIS_MODULE,
+	.supported_type_groups  = mdev_type_groups,
+	.create			= vfio_ccw_mdev_create,
+	.remove			= vfio_ccw_mdev_remove,
+};
+
 int vfio_ccw_mdev_reg(struct subchannel *sch)
 {
 	return mdev_register_device(&sch->dev, &vfio_ccw_mdev_ops);
drivers/s390/crypto/vfio_ap_ops.c +10 -3
index 2cfd96112aa0..3a89933f0d3e 100644
@@ -16,6 +16,7 @@
 #include <linux/bitops.h>
 #include <linux/kvm_host.h>
 #include <linux/module.h>
+#include <linux/vfio_mdev.h>
 #include <asm/kvm.h>
 #include <asm/zcrypt.h>
 
@@ -321,6 +322,8 @@ static void vfio_ap_matrix_init(struct ap_config_info *info,
 	matrix->adm_max = info->apxa ? info->Nd : 15;
 }
 
+static const struct vfio_mdev_device_ops vfio_mdev_ops;
+
 static int vfio_ap_mdev_create(struct kobject *kobj, struct mdev_device *mdev)
 {
 	struct ap_matrix_mdev *matrix_mdev;
@@ -344,6 +347,7 @@ static int vfio_ap_mdev_create(struct kobject *kobj, struct mdev_device *mdev)
 	mutex_unlock(&matrix_dev->lock);
 
 	mdev_set_class_id(mdev, MDEV_ID_VFIO);
+	mdev_set_dev_ops(mdev, &vfio_mdev_ops);
 	return 0;
 }
 
@@ -1281,15 +1285,18 @@ static ssize_t vfio_ap_mdev_ioctl(struct mdev_device *mdev,
 	return ret;
 }
 
+static const struct vfio_mdev_device_ops vfio_mdev_ops = {
+	.open			= vfio_ap_mdev_open,
+	.release		= vfio_ap_mdev_release,
+	.ioctl			= vfio_ap_mdev_ioctl,
+};
+
 static const struct mdev_parent_ops vfio_ap_matrix_ops = {
 	.owner			= THIS_MODULE,
 	.supported_type_groups	= vfio_ap_mdev_type_groups,
 	.mdev_attr_groups	= vfio_ap_mdev_attr_groups,
 	.create			= vfio_ap_mdev_create,
 	.remove			= vfio_ap_mdev_remove,
-	.open			= vfio_ap_mdev_open,
-	.release		= vfio_ap_mdev_release,
-	.ioctl			= vfio_ap_mdev_ioctl,
 };
 
 int vfio_ap_mdev_register(void)
drivers/vfio/mdev/mdev_core.c +12 -0
index 8764cf4a276d..6f35f2ced2c9 100644
@@ -51,6 +51,18 @@ void mdev_set_class_id(struct mdev_device *mdev, u16 id)
 }
 EXPORT_SYMBOL(mdev_set_class_id);
 
+const void *mdev_get_dev_ops(struct mdev_device *mdev)
+{
+	return mdev->device_ops;
+}
+EXPORT_SYMBOL(mdev_get_dev_ops);
+
+void mdev_set_dev_ops(struct mdev_device *mdev, const void *ops)
+{
+	mdev->device_ops = ops;
+}
+EXPORT_SYMBOL(mdev_set_dev_ops);
+
 struct device *mdev_dev(struct mdev_device *mdev)
 {
 	return &mdev->dev;
drivers/vfio/mdev/mdev_private.h +1 -0
index c65f436c1869..b666805f0b1f 100644
@@ -34,6 +34,7 @@ struct mdev_device {
 	struct device *iommu_device;
 	bool active;
 	u16 class_id;
+	const void *device_ops;
 };
 
 #define to_mdev_device(dev)	container_of(dev, struct mdev_device, dev)
drivers/vfio/mdev/vfio_mdev.c +19 -18
index 891cf83a2d9a..95efa054442f 100644
@@ -14,6 +14,7 @@
 #include <linux/slab.h>
 #include <linux/vfio.h>
 #include <linux/mdev.h>
+#include <linux/vfio_mdev.h>
 
 #include "mdev_private.h"
 
@@ -24,16 +25,16 @@
 static int vfio_mdev_open(void *device_data)
 {
 	struct mdev_device *mdev = device_data;
-	struct mdev_parent *parent = mdev->parent;
+	const struct vfio_mdev_device_ops *ops = mdev_get_dev_ops(mdev);
 	int ret;
 
-	if (unlikely(!parent->ops->open))
+	if (unlikely(!ops->open))
 		return -EINVAL;
 
 	if (!try_module_get(THIS_MODULE))
 		return -ENODEV;
 
-	ret = parent->ops->open(mdev);
+	ret = ops->open(mdev);
 	if (ret)
 		module_put(THIS_MODULE);
 
@@ -43,10 +44,10 @@ static int vfio_mdev_open(void *device_data)
 static void vfio_mdev_release(void *device_data)
 {
 	struct mdev_device *mdev = device_data;
-	struct mdev_parent *parent = mdev->parent;
+	const struct vfio_mdev_device_ops *ops = mdev_get_dev_ops(mdev);
 
-	if (likely(parent->ops->release))
-		parent->ops->release(mdev);
+	if (likely(ops->release))
+		ops->release(mdev);
 
 	module_put(THIS_MODULE);
 }
@@ -55,47 +56,47 @@ static long vfio_mdev_unlocked_ioctl(void *device_data,
 				     unsigned int cmd, unsigned long arg)
 {
 	struct mdev_device *mdev = device_data;
-	struct mdev_parent *parent = mdev->parent;
+	const struct vfio_mdev_device_ops *ops = mdev_get_dev_ops(mdev);
 
-	if (unlikely(!parent->ops->ioctl))
+	if (unlikely(!ops->ioctl))
 		return -EINVAL;
 
-	return parent->ops->ioctl(mdev, cmd, arg);
+	return ops->ioctl(mdev, cmd, arg);
 }
 
 static ssize_t vfio_mdev_read(void *device_data, char __user *buf,
 			      size_t count, loff_t *ppos)
 {
 	struct mdev_device *mdev = device_data;
-	struct mdev_parent *parent = mdev->parent;
+	const struct vfio_mdev_device_ops *ops = mdev_get_dev_ops(mdev);
 
-	if (unlikely(!parent->ops->read))
+	if (unlikely(!ops->read))
 		return -EINVAL;
 
-	return parent->ops->read(mdev, buf, count, ppos);
+	return ops->read(mdev, buf, count, ppos);
 }
 
 static ssize_t vfio_mdev_write(void *device_data, const char __user *buf,
 			       size_t count, loff_t *ppos)
 {
 	struct mdev_device *mdev = device_data;
-	struct mdev_parent *parent = mdev->parent;
+	const struct vfio_mdev_device_ops *ops = mdev_get_dev_ops(mdev);
 
-	if (unlikely(!parent->ops->write))
+	if (unlikely(!ops->write))
 		return -EINVAL;
 
-	return parent->ops->write(mdev, buf, count, ppos);
+	return ops->write(mdev, buf, count, ppos);
 }
 
 static int vfio_mdev_mmap(void *device_data, struct vm_area_struct *vma)
 {
 	struct mdev_device *mdev = device_data;
-	struct mdev_parent *parent = mdev->parent;
+	const struct vfio_mdev_device_ops *ops = mdev_get_dev_ops(mdev);
 
-	if (unlikely(!parent->ops->mmap))
+	if (unlikely(!ops->mmap))
 		return -EINVAL;
 
-	return parent->ops->mmap(mdev, vma);
+	return ops->mmap(mdev, vma);
 }
 
 static const struct vfio_device_ops vfio_mdev_dev_ops = {
include/linux/mdev.h +9 -33
index 3974650c074f..3414307311f1 100644
@@ -10,6 +10,11 @@
 #ifndef MDEV_H
 #define MDEV_H
 
+#include <linux/types.h>
+#include <linux/device.h>
+#include <linux/mdev.h>
+#include <uapi/linux/uuid.h>
+
 struct mdev_device;
 
 /*
@@ -48,30 +53,8 @@ struct device *mdev_get_iommu_device(struct device *dev);
  *			@mdev: mdev_device device structure which is being
  *			       destroyed
  *			Returns integer: success (0) or error (< 0)
- * @open:		Open mediated device.
- *			@mdev: mediated device.
- *			Returns integer: success (0) or error (< 0)
- * @release:		release mediated device
- *			@mdev: mediated device.
- * @read:		Read emulation callback
- *			@mdev: mediated device structure
- *			@buf: read buffer
- *			@count: number of bytes to read
- *			@ppos: address.
- *			Retuns number on bytes read on success or error.
- * @write:		Write emulation callback
- *			@mdev: mediated device structure
- *			@buf: write buffer
- *			@count: number of bytes to be written
- *			@ppos: address.
- *			Retuns number on bytes written on success or error.
- * @ioctl:		IOCTL callback
- *			@mdev: mediated device structure
- *			@cmd: ioctl command
- *			@arg: arguments to ioctl
- * @mmap:		mmap callback
- *			@mdev: mediated device structure
- *			@vma: vma structure
+ * @device_ops:         Device specific emulation callback.
+ *
  * Parent device that support mediated device should be registered with mdev
  * module with mdev_parent_ops structure.
  **/
@@ -83,15 +66,6 @@ struct mdev_parent_ops {
 
 	int     (*create)(struct kobject *kobj, struct mdev_device *mdev);
 	int     (*remove)(struct mdev_device *mdev);
-	int     (*open)(struct mdev_device *mdev);
-	void    (*release)(struct mdev_device *mdev);
-	ssize_t (*read)(struct mdev_device *mdev, char __user *buf,
-			size_t count, loff_t *ppos);
-	ssize_t (*write)(struct mdev_device *mdev, const char __user *buf,
-			 size_t count, loff_t *ppos);
-	long	(*ioctl)(struct mdev_device *mdev, unsigned int cmd,
-			 unsigned long arg);
-	int	(*mmap)(struct mdev_device *mdev, struct vm_area_struct *vma);
 };
 
 /* interface for exporting mdev supported type attributes */
@@ -133,6 +107,8 @@ struct mdev_driver {
 
 void *mdev_get_drvdata(struct mdev_device *mdev);
 void mdev_set_drvdata(struct mdev_device *mdev, void *data);
+const void *mdev_get_dev_ops(struct mdev_device *mdev);
+void mdev_set_dev_ops(struct mdev_device *mdev, const void *ops);
 void mdev_set_class_id(struct mdev_device *mdev, u16 id);
 const guid_t *mdev_uuid(struct mdev_device *mdev);
 
include/linux/vfio_mdev.h new file +52 -0
new file mode 100644
index 000000000000..3907c5371c2b
@@ -0,0 +1,52 @@
+/* SPDX-License-Identifier: GPL-2.0-only */
+/*
+ * VFIO Mediated device definition
+ */
+
+#ifndef VFIO_MDEV_H
+#define VFIO_MDEV_H
+
+#include <linux/mdev.h>
+
+/**
+ * struct vfio_mdev_device_ops - Structure to be registered for each
+ * mdev device to register the device to vfio-mdev module.
+ *
+ * @open:		Open mediated device.
+ *			@mdev: mediated device.
+ *			Returns integer: success (0) or error (< 0)
+ * @release:		release mediated device
+ *			@mdev: mediated device.
+ * @read:		Read emulation callback
+ *			@mdev: mediated device structure
+ *			@buf: read buffer
+ *			@count: number of bytes to read
+ *			@ppos: address.
+ *			Retuns number on bytes read on success or error.
+ * @write:		Write emulation callback
+ *			@mdev: mediated device structure
+ *			@buf: write buffer
+ *			@count: number of bytes to be written
+ *			@ppos: address.
+ *			Retuns number on bytes written on success or error.
+ * @ioctl:		IOCTL callback
+ *			@mdev: mediated device structure
+ *			@cmd: ioctl command
+ *			@arg: arguments to ioctl
+ * @mmap:		mmap callback
+ *			@mdev: mediated device structure
+ *			@vma: vma structure
+ */
+struct vfio_mdev_device_ops {
+	int     (*open)(struct mdev_device *mdev);
+	void    (*release)(struct mdev_device *mdev);
+	ssize_t (*read)(struct mdev_device *mdev, char __user *buf,
+			size_t count, loff_t *ppos);
+	ssize_t (*write)(struct mdev_device *mdev, const char __user *buf,
+			 size_t count, loff_t *ppos);
+	long	(*ioctl)(struct mdev_device *mdev, unsigned int cmd,
+			 unsigned long arg);
+	int	(*mmap)(struct mdev_device *mdev, struct vm_area_struct *vma);
+};
+
+#endif
samples/vfio-mdev/mbochs.c +13 -6
index 8a8583c892b2..b4bf29c6136c 100644
@@ -30,6 +30,7 @@
 #include <linux/iommu.h>
 #include <linux/sysfs.h>
 #include <linux/mdev.h>
+#include <linux/vfio_mdev.h>
 #include <linux/pci.h>
 #include <linux/dma-buf.h>
 #include <linux/highmem.h>
@@ -516,6 +517,8 @@ static int mbochs_reset(struct mdev_device *mdev)
 	return 0;
 }
 
+static const struct vfio_mdev_device_ops vfio_mdev_ops;
+
 static int mbochs_create(struct kobject *kobj, struct mdev_device *mdev)
 {
 	const struct mbochs_type *type = mbochs_find_type(kobj);
@@ -562,6 +565,7 @@ static int mbochs_create(struct kobject *kobj, struct mdev_device *mdev)
 
 	mbochs_used_mbytes += type->mbytes;
 	mdev_set_class_id(mdev, MDEV_ID_VFIO);
+	mdev_set_dev_ops(mdev, &vfio_mdev_ops);
 	return 0;
 
 err_mem:
@@ -1419,12 +1423,7 @@ static struct attribute_group *mdev_type_groups[] = {
 	NULL,
 };
 
-static const struct mdev_parent_ops mdev_fops = {
-	.owner			= THIS_MODULE,
-	.mdev_attr_groups	= mdev_dev_groups,
-	.supported_type_groups	= mdev_type_groups,
-	.create			= mbochs_create,
-	.remove			= mbochs_remove,
+static const struct vfio_mdev_device_ops vfio_mdev_ops = {
 	.open			= mbochs_open,
 	.release		= mbochs_close,
 	.read			= mbochs_read,
@@ -1433,6 +1432,14 @@ static const struct mdev_parent_ops mdev_fops = {
 	.mmap			= mbochs_mmap,
 };
 
+static const struct mdev_parent_ops mdev_fops = {
+	.owner			= THIS_MODULE,
+	.mdev_attr_groups	= mdev_dev_groups,
+	.supported_type_groups	= mdev_type_groups,
+	.create			= mbochs_create,
+	.remove			= mbochs_remove,
+};
+
 static const struct file_operations vd_fops = {
 	.owner		= THIS_MODULE,
 };
samples/vfio-mdev/mdpy.c +13 -6
index 88d7e76f3836..80c2df531326 100644
@@ -26,6 +26,7 @@
 #include <linux/iommu.h>
 #include <linux/sysfs.h>
 #include <linux/mdev.h>
+#include <linux/vfio_mdev.h>
 #include <linux/pci.h>
 #include <drm/drm_fourcc.h>
 #include "mdpy-defs.h"
@@ -226,6 +227,8 @@ static int mdpy_reset(struct mdev_device *mdev)
 	return 0;
 }
 
+static const struct vfio_mdev_device_ops vfio_mdev_ops;
+
 static int mdpy_create(struct kobject *kobj, struct mdev_device *mdev)
 {
 	const struct mdpy_type *type = mdpy_find_type(kobj);
@@ -270,6 +273,7 @@ static int mdpy_create(struct kobject *kobj, struct mdev_device *mdev)
 
 	mdpy_count++;
 	mdev_set_class_id(mdev, MDEV_ID_VFIO);
+	mdev_set_dev_ops(mdev, &vfio_mdev_ops);
 	return 0;
 }
 
@@ -726,12 +730,7 @@ static struct attribute_group *mdev_type_groups[] = {
 	NULL,
 };
 
-static const struct mdev_parent_ops mdev_fops = {
-	.owner			= THIS_MODULE,
-	.mdev_attr_groups	= mdev_dev_groups,
-	.supported_type_groups	= mdev_type_groups,
-	.create			= mdpy_create,
-	.remove			= mdpy_remove,
+static const struct vfio_mdev_device_ops vfio_mdev_ops = {
 	.open			= mdpy_open,
 	.release		= mdpy_close,
 	.read			= mdpy_read,
@@ -740,6 +739,14 @@ static const struct mdev_parent_ops mdev_fops = {
 	.mmap			= mdpy_mmap,
 };
 
+static const struct mdev_parent_ops mdev_fops = {
+	.owner			= THIS_MODULE,
+	.mdev_attr_groups	= mdev_dev_groups,
+	.supported_type_groups	= mdev_type_groups,
+	.create			= mdpy_create,
+	.remove			= mdpy_remove,
+};
+
 static const struct file_operations vd_fops = {
 	.owner		= THIS_MODULE,
 };
samples/vfio-mdev/mtty.c +12 -5
index 4e0735143b69..2db860ccb02c 100644
@@ -27,6 +27,7 @@
 #include <linux/ctype.h>
 #include <linux/file.h>
 #include <linux/mdev.h>
+#include <linux/vfio_mdev.h>
 #include <linux/pci.h>
 #include <linux/serial.h>
 #include <uapi/linux/serial_reg.h>
@@ -723,6 +724,8 @@ static ssize_t mdev_access(struct mdev_device *mdev, u8 *buf, size_t count,
 	return ret;
 }
 
+static const struct vfio_mdev_device_ops vfio_dev_ops;
+
 static int mtty_create(struct kobject *kobj, struct mdev_device *mdev)
 {
 	struct mdev_state *mdev_state;
@@ -771,6 +774,7 @@ static int mtty_create(struct kobject *kobj, struct mdev_device *mdev)
 	mutex_unlock(&mdev_list_lock);
 
 	mdev_set_class_id(mdev, MDEV_ID_VFIO);
+	mdev_set_dev_ops(mdev, &vfio_dev_ops);
 	return 0;
 }
 
@@ -1411,6 +1415,14 @@ static struct attribute_group *mdev_type_groups[] = {
 	NULL,
 };
 
+static const struct vfio_mdev_device_ops vfio_dev_ops = {
+	.open                   = mtty_open,
+	.release                = mtty_close,
+	.read                   = mtty_read,
+	.write                  = mtty_write,
+	.ioctl		        = mtty_ioctl,
+};
+
 static const struct mdev_parent_ops mdev_fops = {
 	.owner                  = THIS_MODULE,
 	.dev_attr_groups        = mtty_dev_groups,
@@ -1418,11 +1430,6 @@ static const struct mdev_parent_ops mdev_fops = {
 	.supported_type_groups  = mdev_type_groups,
 	.create                 = mtty_create,
 	.remove			= mtty_remove,
-	.open                   = mtty_open,
-	.release                = mtty_close,
-	.read                   = mtty_read,
-	.write                  = mtty_write,
-	.ioctl		        = mtty_ioctl,
 };
 
 static void mtty_device_release(struct device *dev)
#19 lore raw

Re: [PATCH V2 5/8] mdev: introduce device specific ops

Quoted 41 lines

From: Alex Williamson
Sent: Wednesday, September 25, 2019 7:07 AM

Quoted 25 lines · On Tue, 24 Sep 2019 21:53:29 +0800 Jason Wang <[email protected]> wrote:

Currently, except for the create and remove, the rest of mdev_parent_ops is designed for vfio-mdev driver only and may not help for kernel mdev driver. With the help of class id, this patch introduces device specific callbacks inside mdev_device structure. This allows different set of callback to be used by vfio-mdev and virtio-mdev.

---
 .../driver-api/vfio-mediated-device.rst       |  4 +-
 MAINTAINERS                                   |  1 +
 drivers/gpu/drm/i915/gvt/kvmgt.c              | 17 +++---
 drivers/s390/cio/vfio_ccw_ops.c               | 17 ++++--
 drivers/s390/crypto/vfio_ap_ops.c             | 13 +++--
 drivers/vfio/mdev/mdev_core.c                 | 12 +++++
 drivers/vfio/mdev/mdev_private.h              |  1 +
 drivers/vfio/mdev/vfio_mdev.c                 | 37 ++++++-------
 include/linux/mdev.h                          | 42 ++++-----------
 include/linux/vfio_mdev.h                     | 52 +++++++++++++++++++
 samples/vfio-mdev/mbochs.c                    | 19 ++++---
 samples/vfio-mdev/mdpy.c                      | 19 ++++---
 samples/vfio-mdev/mtty.c                      | 17 ++++--
 13 files changed, 168 insertions(+), 83 deletions(-)
 create mode 100644 include/linux/vfio_mdev.h

Review

Quoted 1 line
(quoted diff) +0 -0

b/Documentation/driver-api/vfio-mediated-device.rst

Documentation/driver-api/vfio-mediated-device.rst callbacks per mdev parent device, per mdev type,

Quoted 4 lines

index a5bdc60d62a1..d50425b368bb 100644

Documentation/driver-api/vfio-mediated-device.rst +0 -0
@@ -152,7 +152,9 @@ callbacks per mdev parent device, per mdev type,

or any other categorization.

Quoted 3 lines
Documentation/driver-api/vfio-mediated-device.rst +0 -0
 Vendor drivers are expected to be fully asynchronous in this respect or
 provide their own internal resource protection.)
 
Quoted 9 lines

Documentation/driver-api/vfio-mediated-device.rst callbacks per mdev parent device, per mdev type, or any other categorization. · line 157

Quoted 4 lines
Documentation/driver-api/vfio-mediated-device.rst +3 -1
-The callbacks in the mdev_parent_ops structure are as follows:
+The device specific callbacks are referred through device_ops pointer
+in mdev_parent_ops. For vfio-mdev device, its callbacks in device_ops
+are as follows:

This is not accurate. device_ops is now on the mdev_device and is an mdev bus driver specific structure of callbacks that must be registered for each mdev device by the parent driver during the create callback. There's a one to one mapping of class_id to mdev_device_ops callbacks.

there is also a mistake in include/Linux/mdev.h, where device_ops is still part of mdev_parent_ops in the comment line.

Quoted 60 lines

That also suggests to me that we could be more clever in registering both of these with mdev-core. Can we embed the class_id in the ops structure in a common way so that the core can extract it and the bus drivers can access their specific callbacks?

Review

Quoted 15 lines
Documentation/driver-api/vfio-mediated-device.rst +0 -0
 * open: open callback of mediated device
 * close: close callback of mediated device
MAINTAINERS +1 -0
index b2326dece28e..89832b316500 100644
@@ -17075,6 +17075,7 @@ S:	Maintained
 F:	Documentation/driver-api/vfio-mediated-device.rst
 F:	drivers/vfio/mdev/
 F:	include/linux/mdev.h
+F:	include/linux/vfio_mdev.h
 F:	samples/vfio-mdev/
 
 VFIO PLATFORM DRIVER
(quoted diff) +0 -0

b/drivers/gpu/drm/i915/gvt/kvmgt.c

Quoted 12 lines

index f793252a3d2a..b274f5ee481f 100644

drivers/gpu/drm/i915/gvt/kvmgt.c +1 -0
@@ -42,6 +42,7 @@
 #include <linux/kvm_host.h>
 #include <linux/vfio.h>
 #include <linux/mdev.h>
+#include <linux/vfio_mdev.h>
 #include <linux/debugfs.h>
 
 #include <linux/nospec.h>

drivers/gpu/drm/i915/gvt/kvmgt.c static void kvmgt_put_vfio_device(void *vgpu) · line 649

Quoted 6 lines
drivers/gpu/drm/i915/gvt/kvmgt.c +2 -0
@@ -643,6 +644,8 @@ static void kvmgt_put_vfio_device(void *vgpu)
 	vfio_device_put(((struct intel_vgpu *)vgpu)->vdev.vfio_device);
 }
 
+static struct vfio_mdev_device_ops intel_vfio_vgpu_dev_ops;
+
 static int intel_vgpu_create(struct kobject *kobj, struct mdev_device

*mdev)

drivers/gpu/drm/i915/gvt/kvmgt.c static int intel_vgpu_create(struct kobject *kobj,

Quoted 3 lines
drivers/gpu/drm/i915/gvt/kvmgt.c +0 -0
 {
 	struct intel_vgpu *vgpu = NULL;
@@ -679,6 +682,7 @@ static int intel_vgpu_create(struct kobject *kobj,

struct mdev_device *mdev)

drivers/gpu/drm/i915/gvt/kvmgt.c static int intel_vgpu_create(struct kobject *kobj, struct mdev_device *mdev) · line 685

Quoted 4 lines
drivers/gpu/drm/i915/gvt/kvmgt.c +1 -0
 	ret = 0;
 
 	mdev_set_class_id(mdev, MDEV_ID_VFIO);
+	mdev_set_dev_ops(mdev, &intel_vfio_vgpu_dev_ops);

This seems rather unrefined. We're registering interdependent data in separate calls. All drivers need to make both of these calls. I'm not sure if this is a good idea, but what if we had:

static const struct vfio_mdev_device_ops intel_vfio_vgpu_dev_ops = {
	.id			= MDEV_ID_VFIO,
drivers/gpu/drm/i915/gvt/kvmgt.c +0 -0
 	.open			= intel_vgpu_open,
 	.release		= intel_vgpu_release,
...
Quoted 2 lines

And the set function passed &intel_vfio_vgpu_dev_ops.id and the mdev bus drivers used container_of to get to their callbacks?

or just make it explicit? e.g.

mdev_set_class(mdev, MDEV_ID_VFIO, &intel_vfio_vgpu_dev_ops);

Quoted 629 lines

drivers/gpu/drm/i915/gvt/kvmgt.c static const struct attribute_group

Quoted 5 lines
drivers/gpu/drm/i915/gvt/kvmgt.c +0 -0
 out:
 	return ret;
 }
@@ -1601,20 +1605,21 @@ static const struct attribute_group

*intel_vgpu_groups[] = {

Quoted 9 lines
drivers/gpu/drm/i915/gvt/kvmgt.c +1 -5
 	NULL,
 };
 
-static struct mdev_parent_ops intel_vgpu_ops = {
-	.mdev_attr_groups       = intel_vgpu_groups,
-	.create			= intel_vgpu_create,
-	.remove			= intel_vgpu_remove,
-
+static struct vfio_mdev_device_ops intel_vfio_vgpu_dev_ops = {

drivers/gpu/drm/i915/gvt/kvmgt.c static const struct attribute_group *intel_vgpu_groups[] = { · line 1617

Quoted 10 lines
drivers/gpu/drm/i915/gvt/kvmgt.c +1 -1
 	.open			= intel_vgpu_open,
 	.release		= intel_vgpu_release,
-
 	.read			= intel_vgpu_read,
 	.write			= intel_vgpu_write,
 	.mmap			= intel_vgpu_mmap,
 	.ioctl			= intel_vgpu_ioctl,
 };
 
+static struct mdev_parent_ops intel_vgpu_ops = {

These could maybe be made const at the same time. Thanks,

Alex

Review

Quoted 9 lines
drivers/gpu/drm/i915/gvt/kvmgt.c +5 -0
+	.mdev_attr_groups       = intel_vgpu_groups,
+	.create			= intel_vgpu_create,
+	.remove			= intel_vgpu_remove,
+};
+
 static int kvmgt_host_init(struct device *dev, void *gvt, const void *ops)
 {
 	struct attribute **kvm_type_attrs;
(quoted diff) +0 -0

b/drivers/s390/cio/vfio_ccw_ops.c

drivers/s390/cio/vfio_ccw_ops.c static struct attribute_group *mdev_type_groups[]

Quoted 12 lines

index d258ef1fedb9..329d53c1f46b 100644

drivers/s390/cio/vfio_ccw_ops.c +1 -0
@@ -12,6 +12,7 @@
 
 #include <linux/vfio.h>
 #include <linux/mdev.h>
+#include <linux/vfio_mdev.h>
 #include <linux/nospec.h>
 #include <linux/slab.h>
 
@@ -110,6 +111,8 @@ static struct attribute_group *mdev_type_groups[]

= {

drivers/s390/cio/vfio_ccw_ops.c static struct attribute_group *mdev_type_groups[] = {

Quoted 6 lines
drivers/s390/cio/vfio_ccw_ops.c +2 -0
 	NULL,
 };
 
+static const struct vfio_mdev_device_ops vfio_mdev_ops;
+
 static int vfio_ccw_mdev_create(struct kobject *kobj, struct mdev_device

*mdev)

drivers/s390/cio/vfio_ccw_ops.c static int vfio_ccw_mdev_create(struct kobject

Quoted 3 lines
drivers/s390/cio/vfio_ccw_ops.c +0 -0
 {
 	struct vfio_ccw_private *private =
@@ -130,6 +133,7 @@ static int vfio_ccw_mdev_create(struct kobject

*kobj, struct mdev_device *mdev)

drivers/s390/cio/vfio_ccw_ops.c static ssize_t vfio_ccw_mdev_ioctl(struct

Quoted 8 lines
drivers/s390/cio/vfio_ccw_ops.c +1 -0
 			   private->sch->schid.sch_no);
 
 	mdev_set_class_id(mdev, MDEV_ID_VFIO);
+	mdev_set_dev_ops(mdev, &vfio_mdev_ops);
 	return 0;
 }
 
@@ -575,11 +579,7 @@ static ssize_t vfio_ccw_mdev_ioctl(struct

mdev_device *mdev,

drivers/s390/cio/vfio_ccw_ops.c static const struct mdev_parent_ops

Quoted 13 lines
drivers/s390/cio/vfio_ccw_ops.c +1 -5
 	}
 }
 
-static const struct mdev_parent_ops vfio_ccw_mdev_ops = {
-	.owner			= THIS_MODULE,
-	.supported_type_groups  = mdev_type_groups,
-	.create			= vfio_ccw_mdev_create,
-	.remove			= vfio_ccw_mdev_remove,
+static const struct vfio_mdev_device_ops vfio_mdev_ops = {
 	.open			= vfio_ccw_mdev_open,
 	.release		= vfio_ccw_mdev_release,
 	.read			= vfio_ccw_mdev_read,
@@ -587,6 +587,13 @@ static const struct mdev_parent_ops

vfio_ccw_mdev_ops = {

Review

Quoted 14 lines
drivers/s390/cio/vfio_ccw_ops.c +7 -0
 	.ioctl			= vfio_ccw_mdev_ioctl,
 };
 
+static const struct mdev_parent_ops vfio_ccw_mdev_ops = {
+	.owner			= THIS_MODULE,
+	.supported_type_groups  = mdev_type_groups,
+	.create			= vfio_ccw_mdev_create,
+	.remove			= vfio_ccw_mdev_remove,
+};
+
 int vfio_ccw_mdev_reg(struct subchannel *sch)
 {
 	return mdev_register_device(&sch->dev, &vfio_ccw_mdev_ops);
(quoted diff) +0 -0

b/drivers/s390/crypto/vfio_ap_ops.c

drivers/s390/crypto/vfio_ap_ops.c static void vfio_ap_matrix_init(struct

Quoted 12 lines

index 2cfd96112aa0..3a89933f0d3e 100644

drivers/s390/crypto/vfio_ap_ops.c +1 -0
@@ -16,6 +16,7 @@
 #include <linux/bitops.h>
 #include <linux/kvm_host.h>
 #include <linux/module.h>
+#include <linux/vfio_mdev.h>
 #include <asm/kvm.h>
 #include <asm/zcrypt.h>
 
@@ -321,6 +322,8 @@ static void vfio_ap_matrix_init(struct

ap_config_info *info,

drivers/s390/crypto/vfio_ap_ops.c static void vfio_ap_matrix_init(struct ap_config_info *info,

Quoted 6 lines
drivers/s390/crypto/vfio_ap_ops.c +2 -0
 	matrix->adm_max = info->apxa ? info->Nd : 15;
 }
 
+static const struct vfio_mdev_device_ops vfio_mdev_ops;
+
 static int vfio_ap_mdev_create(struct kobject *kobj, struct mdev_device

*mdev)

drivers/s390/crypto/vfio_ap_ops.c static int vfio_ap_mdev_create(struct kobject

Quoted 3 lines
drivers/s390/crypto/vfio_ap_ops.c +0 -0
 {
 	struct ap_matrix_mdev *matrix_mdev;
@@ -344,6 +347,7 @@ static int vfio_ap_mdev_create(struct kobject

*kobj, struct mdev_device *mdev)

drivers/s390/crypto/vfio_ap_ops.c static ssize_t vfio_ap_mdev_ioctl(struct

Quoted 8 lines
drivers/s390/crypto/vfio_ap_ops.c +1 -0
 	mutex_unlock(&matrix_dev->lock);
 
 	mdev_set_class_id(mdev, MDEV_ID_VFIO);
+	mdev_set_dev_ops(mdev, &vfio_mdev_ops);
 	return 0;
 }
 
@@ -1281,15 +1285,18 @@ static ssize_t vfio_ap_mdev_ioctl(struct

mdev_device *mdev,

Review

Quoted 22 lines
drivers/s390/crypto/vfio_ap_ops.c +6 -3
 	return ret;
 }
 
+static const struct vfio_mdev_device_ops vfio_mdev_ops = {
+	.open			= vfio_ap_mdev_open,
+	.release		= vfio_ap_mdev_release,
+	.ioctl			= vfio_ap_mdev_ioctl,
+};
+
 static const struct mdev_parent_ops vfio_ap_matrix_ops = {
 	.owner			= THIS_MODULE,
 	.supported_type_groups	= vfio_ap_mdev_type_groups,
 	.mdev_attr_groups	= vfio_ap_mdev_attr_groups,
 	.create			= vfio_ap_mdev_create,
 	.remove			= vfio_ap_mdev_remove,
-	.open			= vfio_ap_mdev_open,
-	.release		= vfio_ap_mdev_release,
-	.ioctl			= vfio_ap_mdev_ioctl,
 };
 
 int vfio_ap_mdev_register(void)
(quoted diff) +0 -0

b/drivers/vfio/mdev/mdev_core.c

drivers/vfio/mdev/mdev_core.c void mdev_set_class_id(struct mdev_device *mdev,

Quoted 4 lines

index 8764cf4a276d..6f35f2ced2c9 100644

drivers/vfio/mdev/mdev_core.c +0 -0
@@ -51,6 +51,18 @@ void mdev_set_class_id(struct mdev_device *mdev,

u16 id)

Review

Quoted 19 lines
drivers/vfio/mdev/mdev_core.c +12 -0
 }
 EXPORT_SYMBOL(mdev_set_class_id);
 
+const void *mdev_get_dev_ops(struct mdev_device *mdev)
+{
+	return mdev->device_ops;
+}
+EXPORT_SYMBOL(mdev_get_dev_ops);
+
+void mdev_set_dev_ops(struct mdev_device *mdev, const void *ops)
+{
+	mdev->device_ops = ops;
+}
+EXPORT_SYMBOL(mdev_set_dev_ops);
+
 struct device *mdev_dev(struct mdev_device *mdev)
 {
 	return &mdev->dev;
(quoted diff) +0 -0

b/drivers/vfio/mdev/mdev_private.h

drivers/vfio/mdev/mdev_private.h struct mdev_device { · line 40

Quoted 11 lines

index c65f436c1869..b666805f0b1f 100644

drivers/vfio/mdev/mdev_private.h +1 -0
@@ -34,6 +34,7 @@ struct mdev_device {
 	struct device *iommu_device;
 	bool active;
 	u16 class_id;
+	const void *device_ops;
 };
 
 #define to_mdev_device(dev)	container_of(dev, struct

mdev_device, dev)

Review

Quoted 1 line
(quoted diff) +0 -0

b/drivers/vfio/mdev/vfio_mdev.c

Quoted 12 lines

index 891cf83a2d9a..95efa054442f 100644

drivers/vfio/mdev/vfio_mdev.c +1 -0
@@ -14,6 +14,7 @@
 #include <linux/slab.h>
 #include <linux/vfio.h>
 #include <linux/mdev.h>
+#include <linux/vfio_mdev.h>
 
 #include "mdev_private.h"
 

drivers/vfio/mdev/vfio_mdev.c line 28

Quoted 5 lines
drivers/vfio/mdev/vfio_mdev.c +1 -1
@@ -24,16 +25,16 @@
 static int vfio_mdev_open(void *device_data)
 {
 	struct mdev_device *mdev = device_data;
-	struct mdev_parent *parent = mdev->parent;
+	const struct vfio_mdev_device_ops *ops =

mdev_get_dev_ops(mdev);

Quoted 15 lines
drivers/vfio/mdev/vfio_mdev.c +2 -2
 	int ret;
 
-	if (unlikely(!parent->ops->open))
+	if (unlikely(!ops->open))
 		return -EINVAL;
 
 	if (!try_module_get(THIS_MODULE))
 		return -ENODEV;
 
-	ret = parent->ops->open(mdev);
+	ret = ops->open(mdev);
 	if (ret)
 		module_put(THIS_MODULE);
 

drivers/vfio/mdev/vfio_mdev.c static int vfio_mdev_open(void *device_data) · line 47

Quoted 5 lines
drivers/vfio/mdev/vfio_mdev.c +1 -1
@@ -43,10 +44,10 @@ static int vfio_mdev_open(void *device_data)
 static void vfio_mdev_release(void *device_data)
 {
 	struct mdev_device *mdev = device_data;
-	struct mdev_parent *parent = mdev->parent;
+	const struct vfio_mdev_device_ops *ops =

mdev_get_dev_ops(mdev);

drivers/vfio/mdev/vfio_mdev.c static long vfio_mdev_unlocked_ioctl(void

Quoted 9 lines
drivers/vfio/mdev/vfio_mdev.c +2 -2
-	if (likely(parent->ops->release))
-		parent->ops->release(mdev);
+	if (likely(ops->release))
+		ops->release(mdev);
 
 	module_put(THIS_MODULE);
 }
@@ -55,47 +56,47 @@ static long vfio_mdev_unlocked_ioctl(void

*device_data,

drivers/vfio/mdev/vfio_mdev.c static long vfio_mdev_unlocked_ioctl(void *device_data,

Quoted 5 lines
drivers/vfio/mdev/vfio_mdev.c +1 -1
 				     unsigned int cmd, unsigned long arg)
 {
 	struct mdev_device *mdev = device_data;
-	struct mdev_parent *parent = mdev->parent;
+	const struct vfio_mdev_device_ops *ops =

mdev_get_dev_ops(mdev);

Quoted 5 lines
drivers/vfio/mdev/vfio_mdev.c +1 -1
-	if (unlikely(!parent->ops->ioctl))
+	if (unlikely(!ops->ioctl))
 		return -EINVAL;
 

drivers/vfio/mdev/vfio_mdev.c static long vfio_mdev_unlocked_ioctl(void *device_data,

Quoted 10 lines
drivers/vfio/mdev/vfio_mdev.c +2 -2
-	return parent->ops->ioctl(mdev, cmd, arg);
+	return ops->ioctl(mdev, cmd, arg);
 }
 
 static ssize_t vfio_mdev_read(void *device_data, char __user *buf,
 			      size_t count, loff_t *ppos)
 {
 	struct mdev_device *mdev = device_data;
-	struct mdev_parent *parent = mdev->parent;
+	const struct vfio_mdev_device_ops *ops =

mdev_get_dev_ops(mdev);

Quoted 5 lines
drivers/vfio/mdev/vfio_mdev.c +1 -1
-	if (unlikely(!parent->ops->read))
+	if (unlikely(!ops->read))
 		return -EINVAL;
 

drivers/vfio/mdev/vfio_mdev.c static long vfio_mdev_unlocked_ioctl(void *device_data,

Quoted 10 lines
drivers/vfio/mdev/vfio_mdev.c +2 -2
-	return parent->ops->read(mdev, buf, count, ppos);
+	return ops->read(mdev, buf, count, ppos);
 }
 
 static ssize_t vfio_mdev_write(void *device_data, const char __user *buf,
 			       size_t count, loff_t *ppos)
 {
 	struct mdev_device *mdev = device_data;
-	struct mdev_parent *parent = mdev->parent;
+	const struct vfio_mdev_device_ops *ops =

mdev_get_dev_ops(mdev);

drivers/vfio/mdev/vfio_mdev.c static long vfio_mdev_unlocked_ioctl(void *device_data,

Quoted 10 lines
drivers/vfio/mdev/vfio_mdev.c +2 -2
-	if (unlikely(!parent->ops->write))
+	if (unlikely(!ops->write))
 		return -EINVAL;
 
-	return parent->ops->write(mdev, buf, count, ppos);
+	return ops->write(mdev, buf, count, ppos);
 }
 
 static int vfio_mdev_mmap(void *device_data, struct vm_area_struct

*vma)

drivers/vfio/mdev/vfio_mdev.c

Quoted 4 lines
drivers/vfio/mdev/vfio_mdev.c +1 -1
 {
 	struct mdev_device *mdev = device_data;
-	struct mdev_parent *parent = mdev->parent;
+	const struct vfio_mdev_device_ops *ops =

mdev_get_dev_ops(mdev);

include/linux/mdev.h struct device *mdev_get_iommu_device(struct

Quoted 27 lines
drivers/vfio/mdev/vfio_mdev.c +2 -2
-	if (unlikely(!parent->ops->mmap))
+	if (unlikely(!ops->mmap))
 		return -EINVAL;
 
-	return parent->ops->mmap(mdev, vma);
+	return ops->mmap(mdev, vma);
 }
 
 static const struct vfio_device_ops vfio_mdev_dev_ops = {
include/linux/mdev.h +5 -0
index 3974650c074f..3414307311f1 100644
@@ -10,6 +10,11 @@
 #ifndef MDEV_H
 #define MDEV_H
 
+#include <linux/types.h>
+#include <linux/device.h>
+#include <linux/mdev.h>
+#include <uapi/linux/uuid.h>
+
 struct mdev_device;
 
 /*
@@ -48,30 +53,8 @@ struct device *mdev_get_iommu_device(struct

device *dev);

include/linux/mdev.h struct device *mdev_get_iommu_device(struct · line 53

Quoted 1 line
include/linux/mdev.h +0 -0
  *			@mdev: mdev_device device structure which is

being

Quoted 19 lines
include/linux/mdev.h +0 -17
  *			       destroyed
  *			Returns integer: success (0) or error (< 0)
- * @open:		Open mediated device.
- *			@mdev: mediated device.
- *			Returns integer: success (0) or error (< 0)
- * @release:		release mediated device
- *			@mdev: mediated device.
- * @read:		Read emulation callback
- *			@mdev: mediated device structure
- *			@buf: read buffer
- *			@count: number of bytes to read
- *			@ppos: address.
- *			Retuns number on bytes read on success or error.
- * @write:		Write emulation callback
- *			@mdev: mediated device structure
- *			@buf: write buffer
- *			@count: number of bytes to be written
- *			@ppos: address.
- *			Retuns number on bytes written on success or error.

include/linux/mdev.h struct device *mdev_get_iommu_device(struct device *dev);

Quoted 10 lines
include/linux/mdev.h +2 -7
- * @ioctl:		IOCTL callback
- *			@mdev: mediated device structure
- *			@cmd: ioctl command
- *			@arg: arguments to ioctl
- * @mmap:		mmap callback
- *			@mdev: mediated device structure
- *			@vma: vma structure
+ * @device_ops:         Device specific emulation callback.
+ *
  * Parent device that support mediated device should be registered with

mdev

Quoted 5 lines
include/linux/mdev.h +0 -0
  * module with mdev_parent_ops structure.
  **/
@@ -83,15 +66,6 @@ struct mdev_parent_ops {
 
 	int     (*create)(struct kobject *kobj, struct mdev_device *mdev);

include/linux/mdev.h struct mdev_parent_ops { · line 94

Quoted 10 lines
include/linux/mdev.h +0 -9
 	int     (*remove)(struct mdev_device *mdev);
-	int     (*open)(struct mdev_device *mdev);
-	void    (*release)(struct mdev_device *mdev);
-	ssize_t (*read)(struct mdev_device *mdev, char __user *buf,
-			size_t count, loff_t *ppos);
-	ssize_t (*write)(struct mdev_device *mdev, const char __user *buf,
-			 size_t count, loff_t *ppos);
-	long	(*ioctl)(struct mdev_device *mdev, unsigned int cmd,
-			 unsigned long arg);
-	int	(*mmap)(struct mdev_device *mdev, struct vm_area_struct

*vma);

Quoted 57 lines
include/linux/mdev.h +2 -0
 };
 
 /* interface for exporting mdev supported type attributes */
@@ -133,6 +107,8 @@ struct mdev_driver {
 
 void *mdev_get_drvdata(struct mdev_device *mdev);
 void mdev_set_drvdata(struct mdev_device *mdev, void *data);
+const void *mdev_get_dev_ops(struct mdev_device *mdev);
+void mdev_set_dev_ops(struct mdev_device *mdev, const void *ops);
 void mdev_set_class_id(struct mdev_device *mdev, u16 id);
 const guid_t *mdev_uuid(struct mdev_device *mdev);
 
include/linux/vfio_mdev.h new file +39 -0
new file mode 100644
index 000000000000..3907c5371c2b
@@ -0,0 +1,52 @@
+/* SPDX-License-Identifier: GPL-2.0-only */
+/*
+ * VFIO Mediated device definition
+ */
+
+#ifndef VFIO_MDEV_H
+#define VFIO_MDEV_H
+
+#include <linux/mdev.h>
+
+/**
+ * struct vfio_mdev_device_ops - Structure to be registered for each
+ * mdev device to register the device to vfio-mdev module.
+ *
+ * @open:		Open mediated device.
+ *			@mdev: mediated device.
+ *			Returns integer: success (0) or error (< 0)
+ * @release:		release mediated device
+ *			@mdev: mediated device.
+ * @read:		Read emulation callback
+ *			@mdev: mediated device structure
+ *			@buf: read buffer
+ *			@count: number of bytes to read
+ *			@ppos: address.
+ *			Retuns number on bytes read on success or error.
+ * @write:		Write emulation callback
+ *			@mdev: mediated device structure
+ *			@buf: write buffer
+ *			@count: number of bytes to be written
+ *			@ppos: address.
+ *			Retuns number on bytes written on success or error.
+ * @ioctl:		IOCTL callback
+ *			@mdev: mediated device structure
+ *			@cmd: ioctl command
+ *			@arg: arguments to ioctl
+ * @mmap:		mmap callback
+ *			@mdev: mediated device structure
+ *			@vma: vma structure
+ */

include/linux/vfio_mdev.h line 49

Quoted 10 lines
include/linux/vfio_mdev.h new file +10 -0
+struct vfio_mdev_device_ops {
+	int     (*open)(struct mdev_device *mdev);
+	void    (*release)(struct mdev_device *mdev);
+	ssize_t (*read)(struct mdev_device *mdev, char __user *buf,
+			size_t count, loff_t *ppos);
+	ssize_t (*write)(struct mdev_device *mdev, const char __user *buf,
+			 size_t count, loff_t *ppos);
+	long	(*ioctl)(struct mdev_device *mdev, unsigned int cmd,
+			 unsigned long arg);
+	int	(*mmap)(struct mdev_device *mdev, struct vm_area_struct

*vma);

samples/vfio-mdev/mbochs.c static int mbochs_reset(struct mdev_device

Quoted 16 lines
include/linux/vfio_mdev.h new file +3 -0
+};
+
+#endif
samples/vfio-mdev/mbochs.c +1 -0
index 8a8583c892b2..b4bf29c6136c 100644
@@ -30,6 +30,7 @@
 #include <linux/iommu.h>
 #include <linux/sysfs.h>
 #include <linux/mdev.h>
+#include <linux/vfio_mdev.h>
 #include <linux/pci.h>
 #include <linux/dma-buf.h>
 #include <linux/highmem.h>
@@ -516,6 +517,8 @@ static int mbochs_reset(struct mdev_device

*mdev)

samples/vfio-mdev/mbochs.c static int mbochs_create(struct kobject *kobj,

Quoted 9 lines
samples/vfio-mdev/mbochs.c +2 -0
 	return 0;
 }
 
+static const struct vfio_mdev_device_ops vfio_mdev_ops;
+
 static int mbochs_create(struct kobject *kobj, struct mdev_device *mdev)
 {
 	const struct mbochs_type *type = mbochs_find_type(kobj);
@@ -562,6 +565,7 @@ static int mbochs_create(struct kobject *kobj,

struct mdev_device *mdev)

samples/vfio-mdev/mbochs.c static struct attribute_group

Quoted 8 lines
samples/vfio-mdev/mbochs.c +1 -0
 	mbochs_used_mbytes += type->mbytes;
 	mdev_set_class_id(mdev, MDEV_ID_VFIO);
+	mdev_set_dev_ops(mdev, &vfio_mdev_ops);
 	return 0;
 
 err_mem:
@@ -1419,12 +1423,7 @@ static struct attribute_group

*mdev_type_groups[] = {

samples/vfio-mdev/mbochs.c static const struct mdev_parent_ops

Quoted 14 lines
samples/vfio-mdev/mbochs.c +1 -6
 	NULL,
 };
 
-static const struct mdev_parent_ops mdev_fops = {
-	.owner			= THIS_MODULE,
-	.mdev_attr_groups	= mdev_dev_groups,
-	.supported_type_groups	= mdev_type_groups,
-	.create			= mbochs_create,
-	.remove			= mbochs_remove,
+static const struct vfio_mdev_device_ops vfio_mdev_ops = {
 	.open			= mbochs_open,
 	.release		= mbochs_close,
 	.read			= mbochs_read,
@@ -1433,6 +1432,14 @@ static const struct mdev_parent_ops

mdev_fops = {

samples/vfio-mdev/mdpy.c static int mdpy_create(struct kobject *kobj, struct

Quoted 36 lines
samples/vfio-mdev/mbochs.c +8 -0
 	.mmap			= mbochs_mmap,
 };
 
+static const struct mdev_parent_ops mdev_fops = {
+	.owner			= THIS_MODULE,
+	.mdev_attr_groups	= mdev_dev_groups,
+	.supported_type_groups	= mdev_type_groups,
+	.create			= mbochs_create,
+	.remove			= mbochs_remove,
+};
+
 static const struct file_operations vd_fops = {
 	.owner		= THIS_MODULE,
 };
samples/vfio-mdev/mdpy.c +3 -0
index 88d7e76f3836..80c2df531326 100644
@@ -26,6 +26,7 @@
 #include <linux/iommu.h>
 #include <linux/sysfs.h>
 #include <linux/mdev.h>
+#include <linux/vfio_mdev.h>
 #include <linux/pci.h>
 #include <drm/drm_fourcc.h>
 #include "mdpy-defs.h"
@@ -226,6 +227,8 @@ static int mdpy_reset(struct mdev_device *mdev)
 	return 0;
 }
 
+static const struct vfio_mdev_device_ops vfio_mdev_ops;
+
 static int mdpy_create(struct kobject *kobj, struct mdev_device *mdev)
 {
 	const struct mdpy_type *type = mdpy_find_type(kobj);
@@ -270,6 +273,7 @@ static int mdpy_create(struct kobject *kobj, struct

mdev_device *mdev)

samples/vfio-mdev/mdpy.c static struct attribute_group

Quoted 8 lines
samples/vfio-mdev/mdpy.c +1 -0
 	mdpy_count++;
 	mdev_set_class_id(mdev, MDEV_ID_VFIO);
+	mdev_set_dev_ops(mdev, &vfio_mdev_ops);
 	return 0;
 }
 
@@ -726,12 +730,7 @@ static struct attribute_group

*mdev_type_groups[] = {

samples/vfio-mdev/mdpy.c static const struct mdev_parent_ops mdev_fops

Quoted 14 lines
samples/vfio-mdev/mdpy.c +1 -6
 	NULL,
 };
 
-static const struct mdev_parent_ops mdev_fops = {
-	.owner			= THIS_MODULE,
-	.mdev_attr_groups	= mdev_dev_groups,
-	.supported_type_groups	= mdev_type_groups,
-	.create			= mdpy_create,
-	.remove			= mdpy_remove,
+static const struct vfio_mdev_device_ops vfio_mdev_ops = {
 	.open			= mdpy_open,
 	.release		= mdpy_close,
 	.read			= mdpy_read,
@@ -740,6 +739,14 @@ static const struct mdev_parent_ops mdev_fops

= {

samples/vfio-mdev/mtty.c static ssize_t mdev_access(struct mdev_device

Quoted 27 lines
samples/vfio-mdev/mdpy.c +8 -0
 	.mmap			= mdpy_mmap,
 };
 
+static const struct mdev_parent_ops mdev_fops = {
+	.owner			= THIS_MODULE,
+	.mdev_attr_groups	= mdev_dev_groups,
+	.supported_type_groups	= mdev_type_groups,
+	.create			= mdpy_create,
+	.remove			= mdpy_remove,
+};
+
 static const struct file_operations vd_fops = {
 	.owner		= THIS_MODULE,
 };
samples/vfio-mdev/mtty.c +1 -0
index 4e0735143b69..2db860ccb02c 100644
@@ -27,6 +27,7 @@
 #include <linux/ctype.h>
 #include <linux/file.h>
 #include <linux/mdev.h>
+#include <linux/vfio_mdev.h>
 #include <linux/pci.h>
 #include <linux/serial.h>
 #include <uapi/linux/serial_reg.h>
@@ -723,6 +724,8 @@ static ssize_t mdev_access(struct mdev_device

*mdev, u8 *buf, size_t count,

samples/vfio-mdev/mtty.c static int mtty_create(struct kobject *kobj, struct

Quoted 9 lines
samples/vfio-mdev/mtty.c +2 -0
 	return ret;
 }
 
+static const struct vfio_mdev_device_ops vfio_dev_ops;
+
 static int mtty_create(struct kobject *kobj, struct mdev_device *mdev)
 {
 	struct mdev_state *mdev_state;
@@ -771,6 +774,7 @@ static int mtty_create(struct kobject *kobj, struct

mdev_device *mdev)

samples/vfio-mdev/mtty.c static struct attribute_group

Quoted 8 lines
samples/vfio-mdev/mtty.c +1 -0
 	mutex_unlock(&mdev_list_lock);
 
 	mdev_set_class_id(mdev, MDEV_ID_VFIO);
+	mdev_set_dev_ops(mdev, &vfio_dev_ops);
 	return 0;
 }
 
@@ -1411,6 +1415,14 @@ static struct attribute_group

*mdev_type_groups[] = {

samples/vfio-mdev/mtty.c static const struct mdev_parent_ops

Quoted 15 lines
samples/vfio-mdev/mtty.c +8 -0
 	NULL,
 };
 
+static const struct vfio_mdev_device_ops vfio_dev_ops = {
+	.open                   = mtty_open,
+	.release                = mtty_close,
+	.read                   = mtty_read,
+	.write                  = mtty_write,
+	.ioctl		        = mtty_ioctl,
+};
+
 static const struct mdev_parent_ops mdev_fops = {
 	.owner                  = THIS_MODULE,
 	.dev_attr_groups        = mtty_dev_groups,
@@ -1418,11 +1430,6 @@ static const struct mdev_parent_ops

mdev_fops = {

Quoted 10 lines
samples/vfio-mdev/mtty.c +0 -5
 	.supported_type_groups  = mdev_type_groups,
 	.create                 = mtty_create,
 	.remove			= mtty_remove,
-	.open                   = mtty_open,
-	.release                = mtty_close,
-	.read                   = mtty_read,
-	.write                  = mtty_write,
-	.ioctl		        = mtty_ioctl,
 };
 
Quoted 6 lines

samples/vfio-mdev/mtty.c static const struct mdev_parent_ops mdev_fops = { · line 1435

Quoted 1 line
samples/vfio-mdev/mtty.c +0 -0
 static void mtty_device_release(struct device *dev)

_______________________________________________
intel-gvt-dev mailing list
[email protected]
https://lists.freedesktop.org/mailman/listinfo/intel-gvt-dev

_______________________________________________
Intel-gfx mailing list
[email protected]
https://lists.freedesktop.org/mailman/listinfo/intel-gfx

#20 lore raw

Re: [PATCH V2 5/8] mdev: introduce device specific ops

Quoted 239 lines · Kevin
Quoted 31 lines

From: Alex Williamson
Sent: Wednesday, September 25, 2019 7:07 AM

Quoted 25 lines · On Tue, 24 Sep 2019 21:53:29 +0800 Jason Wang <[email protected]> wrote:

Currently, except for the create and remove, the rest of mdev_parent_ops is designed for vfio-mdev driver only and may not help for kernel mdev driver. With the help of class id, this patch introduces device specific callbacks inside mdev_device structure. This allows different set of callback to be used by vfio-mdev and virtio-mdev.

---
 .../driver-api/vfio-mediated-device.rst       |  4 +-
 MAINTAINERS                                   |  1 +
 drivers/gpu/drm/i915/gvt/kvmgt.c              | 17 +++---
 drivers/s390/cio/vfio_ccw_ops.c               | 17 ++++--
 drivers/s390/crypto/vfio_ap_ops.c             | 13 +++--
 drivers/vfio/mdev/mdev_core.c                 | 12 +++++
 drivers/vfio/mdev/mdev_private.h              |  1 +
 drivers/vfio/mdev/vfio_mdev.c                 | 37 ++++++-------
 include/linux/mdev.h                          | 42 ++++-----------
 include/linux/vfio_mdev.h                     | 52 +++++++++++++++++++
 samples/vfio-mdev/mbochs.c                    | 19 ++++---
 samples/vfio-mdev/mdpy.c                      | 19 ++++---
 samples/vfio-mdev/mtty.c                      | 17 ++++--
 13 files changed, 168 insertions(+), 83 deletions(-)
 create mode 100644 include/linux/vfio_mdev.h
Quoted 8 lines

Review

Quoted 1 line
(quoted diff) +0 -0

b/Documentation/driver-api/vfio-mediated-device.rst

Documentation/driver-api/vfio-mediated-device.rst callbacks per mdev parent device, per mdev type,

Quoted 4 lines

index a5bdc60d62a1..d50425b368bb 100644

Documentation/driver-api/vfio-mediated-device.rst +0 -0
@@ -152,7 +152,9 @@ callbacks per mdev parent device, per mdev type,

or any other categorization.

Quoted 1 line
Documentation/driver-api/vfio-mediated-device.rst +0 -0
 Vendor drivers are expected to be fully asynchronous in this respect

or

Quoted 11 lines

Documentation/driver-api/vfio-mediated-device.rst callbacks per mdev parent device, per mdev type, or any other categorization. · line 157

Quoted 6 lines
Documentation/driver-api/vfio-mediated-device.rst +3 -1
 provide their own internal resource protection.)
 
-The callbacks in the mdev_parent_ops structure are as follows:
+The device specific callbacks are referred through device_ops pointer
+in mdev_parent_ops. For vfio-mdev device, its callbacks in device_ops
+are as follows:

This is not accurate. device_ops is now on the mdev_device and is an mdev bus driver specific structure of callbacks that must be registered for each mdev device by the parent driver during the create callback. There's a one to one mapping of class_id to mdev_device_ops callbacks.

there is also a mistake in include/Linux/mdev.h, where device_ops is still part of mdev_parent_ops in the comment line.

Quoted 54 lines

That also suggests to me that we could be more clever in registering both of these with mdev-core. Can we embed the class_id in the ops structure in a common way so that the core can extract it and the bus drivers can access their specific callbacks?

Review

Quoted 15 lines
Documentation/driver-api/vfio-mediated-device.rst +0 -0
 * open: open callback of mediated device
 * close: close callback of mediated device
MAINTAINERS +1 -0
index b2326dece28e..89832b316500 100644
@@ -17075,6 +17075,7 @@ S:  Maintained
 F: Documentation/driver-api/vfio-mediated-device.rst
 F: drivers/vfio/mdev/
 F: include/linux/mdev.h
+F: include/linux/vfio_mdev.h
 F: samples/vfio-mdev/
 
 VFIO PLATFORM DRIVER
(quoted diff) +0 -0

b/drivers/gpu/drm/i915/gvt/kvmgt.c

Quoted 12 lines

index f793252a3d2a..b274f5ee481f 100644

drivers/gpu/drm/i915/gvt/kvmgt.c +1 -0
@@ -42,6 +42,7 @@
 #include <linux/kvm_host.h>
 #include <linux/vfio.h>
 #include <linux/mdev.h>
+#include <linux/vfio_mdev.h>
 #include <linux/debugfs.h>
 
 #include <linux/nospec.h>

drivers/gpu/drm/i915/gvt/kvmgt.c static void kvmgt_put_vfio_device(void *vgpu) · line 649

Quoted 6 lines
drivers/gpu/drm/i915/gvt/kvmgt.c +2 -0
@@ -643,6 +644,8 @@ static void kvmgt_put_vfio_device(void *vgpu)
    vfio_device_put(((struct intel_vgpu *)vgpu)->vdev.vfio_device);
 }
 
+static struct vfio_mdev_device_ops intel_vfio_vgpu_dev_ops;
+
 static int intel_vgpu_create(struct kobject *kobj, struct mdev_device

*mdev)

drivers/gpu/drm/i915/gvt/kvmgt.c static int intel_vgpu_create(struct kobject *kobj,

Quoted 3 lines
drivers/gpu/drm/i915/gvt/kvmgt.c +0 -0
 {
    struct intel_vgpu *vgpu = NULL;
@@ -679,6 +682,7 @@ static int intel_vgpu_create(struct kobject *kobj,

struct mdev_device *mdev)

drivers/gpu/drm/i915/gvt/kvmgt.c static int intel_vgpu_create(struct kobject *kobj, struct mdev_device *mdev) · line 685

Quoted 4 lines
drivers/gpu/drm/i915/gvt/kvmgt.c +1 -0
    ret = 0;
 
    mdev_set_class_id(mdev, MDEV_ID_VFIO);
+   mdev_set_dev_ops(mdev, &intel_vfio_vgpu_dev_ops);

This seems rather unrefined. We're registering interdependent data in separate calls. All drivers need to make both of these calls. I'm not sure if this is a good idea, but what if we had:

Quoted 8 lines

static const struct vfio_mdev_device_ops intel_vfio_vgpu_dev_ops = {

      .id                     = MDEV_ID_VFIO,
      .open                   = intel_vgpu_open,
      .release                = intel_vgpu_release,
        ...

And the set function passed &intel_vfio_vgpu_dev_ops.id and the mdev bus drivers used container_of to get to their callbacks?

or just make it explicit? e.g.

mdev_set_class(mdev, MDEV_ID_VFIO, &intel_vfio_vgpu_dev_ops);

Quoted 28 lines

drivers/gpu/drm/i915/gvt/kvmgt.c static const struct attribute_group

Quoted 5 lines
drivers/gpu/drm/i915/gvt/kvmgt.c +0 -0
 out:
    return ret;
 }
@@ -1601,20 +1605,21 @@ static const struct attribute_group

*intel_vgpu_groups[] = {

Quoted 9 lines
drivers/gpu/drm/i915/gvt/kvmgt.c +1 -5
    NULL,
 };
 
-static struct mdev_parent_ops intel_vgpu_ops = {
-   .mdev_attr_groups       = intel_vgpu_groups,
-   .create                 = intel_vgpu_create,
-   .remove                 = intel_vgpu_remove,
-
+static struct vfio_mdev_device_ops intel_vfio_vgpu_dev_ops = {

drivers/gpu/drm/i915/gvt/kvmgt.c static const struct attribute_group *intel_vgpu_groups[] = { · line 1617

Quoted 10 lines
drivers/gpu/drm/i915/gvt/kvmgt.c +1 -1
    .open                   = intel_vgpu_open,
    .release                = intel_vgpu_release,
-
    .read                   = intel_vgpu_read,
    .write                  = intel_vgpu_write,
    .mmap                   = intel_vgpu_mmap,
    .ioctl                  = intel_vgpu_ioctl,
 };
 
+static struct mdev_parent_ops intel_vgpu_ops = {

These could maybe be made const at the same time. Thanks,

Quoted 8 lines

Alex

drivers/gpu/drm/i915/gvt/kvmgt.c static const struct attribute_group *intel_vgpu_groups[] = {

Quoted 6 lines
drivers/gpu/drm/i915/gvt/kvmgt.c +5 -0
+   .mdev_attr_groups       = intel_vgpu_groups,
+   .create                 = intel_vgpu_create,
+   .remove                 = intel_vgpu_remove,
+};
+
 static int kvmgt_host_init(struct device *dev, void *gvt, const void

*ops)

Quoted 15 lines

Review

Quoted 3 lines
drivers/gpu/drm/i915/gvt/kvmgt.c +0 -0
 {
    struct attribute **kvm_type_attrs;
(quoted diff) +0 -0

b/drivers/s390/cio/vfio_ccw_ops.c

Quoted 11 lines

index d258ef1fedb9..329d53c1f46b 100644

drivers/s390/cio/vfio_ccw_ops.c +1 -0
@@ -12,6 +12,7 @@
 
 #include <linux/vfio.h>
 #include <linux/mdev.h>
+#include <linux/vfio_mdev.h>
 #include <linux/nospec.h>
 #include <linux/slab.h>
 
Quoted 8 lines

drivers/s390/cio/vfio_ccw_ops.c static struct attribute_group *mdev_type_groups[]

Quoted 1 line
drivers/s390/cio/vfio_ccw_ops.c +0 -0
@@ -110,6 +111,8 @@ static struct attribute_group *mdev_type_groups[]

= {

drivers/s390/cio/vfio_ccw_ops.c static struct attribute_group *mdev_type_groups[] = {

Quoted 6 lines
drivers/s390/cio/vfio_ccw_ops.c +2 -0
    NULL,
 };
 
+static const struct vfio_mdev_device_ops vfio_mdev_ops;
+
 static int vfio_ccw_mdev_create(struct kobject *kobj, struct

mdev_device

Quoted 56 lines

*mdev)

drivers/s390/cio/vfio_ccw_ops.c static int vfio_ccw_mdev_create(struct kobject

Quoted 3 lines
drivers/s390/cio/vfio_ccw_ops.c +0 -0
 {
    struct vfio_ccw_private *private =
@@ -130,6 +133,7 @@ static int vfio_ccw_mdev_create(struct kobject

*kobj, struct mdev_device *mdev)

drivers/s390/cio/vfio_ccw_ops.c static ssize_t vfio_ccw_mdev_ioctl(struct

Quoted 8 lines
drivers/s390/cio/vfio_ccw_ops.c +1 -0
                       private->sch->schid.sch_no);
 
    mdev_set_class_id(mdev, MDEV_ID_VFIO);
+   mdev_set_dev_ops(mdev, &vfio_mdev_ops);
    return 0;
 }
 
@@ -575,11 +579,7 @@ static ssize_t vfio_ccw_mdev_ioctl(struct

mdev_device *mdev,

drivers/s390/cio/vfio_ccw_ops.c static const struct mdev_parent_ops

Quoted 13 lines
drivers/s390/cio/vfio_ccw_ops.c +1 -5
    }
 }
 
-static const struct mdev_parent_ops vfio_ccw_mdev_ops = {
-   .owner                  = THIS_MODULE,
-   .supported_type_groups  = mdev_type_groups,
-   .create                 = vfio_ccw_mdev_create,
-   .remove                 = vfio_ccw_mdev_remove,
+static const struct vfio_mdev_device_ops vfio_mdev_ops = {
    .open                   = vfio_ccw_mdev_open,
    .release                = vfio_ccw_mdev_release,
    .read                   = vfio_ccw_mdev_read,
@@ -587,6 +587,13 @@ static const struct mdev_parent_ops

vfio_ccw_mdev_ops = {

Review

Quoted 14 lines
drivers/s390/cio/vfio_ccw_ops.c +7 -0
    .ioctl                  = vfio_ccw_mdev_ioctl,
 };
 
+static const struct mdev_parent_ops vfio_ccw_mdev_ops = {
+   .owner                  = THIS_MODULE,
+   .supported_type_groups  = mdev_type_groups,
+   .create                 = vfio_ccw_mdev_create,
+   .remove                 = vfio_ccw_mdev_remove,
+};
+
 int vfio_ccw_mdev_reg(struct subchannel *sch)
 {
    return mdev_register_device(&sch->dev, &vfio_ccw_mdev_ops);
(quoted diff) +0 -0

b/drivers/s390/crypto/vfio_ap_ops.c

drivers/s390/crypto/vfio_ap_ops.c static void vfio_ap_matrix_init(struct

Quoted 12 lines

index 2cfd96112aa0..3a89933f0d3e 100644

drivers/s390/crypto/vfio_ap_ops.c +1 -0
@@ -16,6 +16,7 @@
 #include <linux/bitops.h>
 #include <linux/kvm_host.h>
 #include <linux/module.h>
+#include <linux/vfio_mdev.h>
 #include <asm/kvm.h>
 #include <asm/zcrypt.h>
 
@@ -321,6 +322,8 @@ static void vfio_ap_matrix_init(struct

ap_config_info *info,

Quoted 110 lines
Quoted 6 lines

drivers/s390/crypto/vfio_ap_ops.c static void vfio_ap_matrix_init(struct ap_config_info *info,

Quoted 6 lines
drivers/s390/crypto/vfio_ap_ops.c +2 -0
    matrix->adm_max = info->apxa ? info->Nd : 15;
 }
 
+static const struct vfio_mdev_device_ops vfio_mdev_ops;
+
 static int vfio_ap_mdev_create(struct kobject *kobj, struct

mdev_device

Quoted 102 lines

*mdev)

drivers/s390/crypto/vfio_ap_ops.c static int vfio_ap_mdev_create(struct kobject

Quoted 3 lines
drivers/s390/crypto/vfio_ap_ops.c +0 -0
 {
    struct ap_matrix_mdev *matrix_mdev;
@@ -344,6 +347,7 @@ static int vfio_ap_mdev_create(struct kobject

*kobj, struct mdev_device *mdev)

drivers/s390/crypto/vfio_ap_ops.c static ssize_t vfio_ap_mdev_ioctl(struct

Quoted 8 lines
drivers/s390/crypto/vfio_ap_ops.c +1 -0
    mutex_unlock(&matrix_dev->lock);
 
    mdev_set_class_id(mdev, MDEV_ID_VFIO);
+   mdev_set_dev_ops(mdev, &vfio_mdev_ops);
    return 0;
 }
 
@@ -1281,15 +1285,18 @@ static ssize_t vfio_ap_mdev_ioctl(struct

mdev_device *mdev,

Review

Quoted 22 lines
drivers/s390/crypto/vfio_ap_ops.c +6 -3
    return ret;
 }
 
+static const struct vfio_mdev_device_ops vfio_mdev_ops = {
+   .open                   = vfio_ap_mdev_open,
+   .release                = vfio_ap_mdev_release,
+   .ioctl                  = vfio_ap_mdev_ioctl,
+};
+
 static const struct mdev_parent_ops vfio_ap_matrix_ops = {
    .owner                  = THIS_MODULE,
    .supported_type_groups  = vfio_ap_mdev_type_groups,
    .mdev_attr_groups       = vfio_ap_mdev_attr_groups,
    .create                 = vfio_ap_mdev_create,
    .remove                 = vfio_ap_mdev_remove,
-   .open                   = vfio_ap_mdev_open,
-   .release                = vfio_ap_mdev_release,
-   .ioctl                  = vfio_ap_mdev_ioctl,
 };
 
 int vfio_ap_mdev_register(void)
(quoted diff) +0 -0

b/drivers/vfio/mdev/mdev_core.c

drivers/vfio/mdev/mdev_core.c void mdev_set_class_id(struct mdev_device *mdev,

Quoted 4 lines

index 8764cf4a276d..6f35f2ced2c9 100644

drivers/vfio/mdev/mdev_core.c +0 -0
@@ -51,6 +51,18 @@ void mdev_set_class_id(struct mdev_device *mdev,

u16 id)

Review

Quoted 19 lines
drivers/vfio/mdev/mdev_core.c +12 -0
 }
 EXPORT_SYMBOL(mdev_set_class_id);
 
+const void *mdev_get_dev_ops(struct mdev_device *mdev)
+{
+   return mdev->device_ops;
+}
+EXPORT_SYMBOL(mdev_get_dev_ops);
+
+void mdev_set_dev_ops(struct mdev_device *mdev, const void *ops)
+{
+   mdev->device_ops = ops;
+}
+EXPORT_SYMBOL(mdev_set_dev_ops);
+
 struct device *mdev_dev(struct mdev_device *mdev)
 {
    return &mdev->dev;
(quoted diff) +0 -0

b/drivers/vfio/mdev/mdev_private.h

drivers/vfio/mdev/mdev_private.h struct mdev_device { · line 40

Quoted 11 lines

index c65f436c1869..b666805f0b1f 100644

drivers/vfio/mdev/mdev_private.h +1 -0
@@ -34,6 +34,7 @@ struct mdev_device {
    struct device *iommu_device;
    bool active;
    u16 class_id;
+   const void *device_ops;
 };
 
 #define to_mdev_device(dev)        container_of(dev, struct

mdev_device, dev)

Review

Quoted 1 line
(quoted diff) +0 -0

b/drivers/vfio/mdev/vfio_mdev.c

Quoted 12 lines

index 891cf83a2d9a..95efa054442f 100644

drivers/vfio/mdev/vfio_mdev.c +1 -0
@@ -14,6 +14,7 @@
 #include <linux/slab.h>
 #include <linux/vfio.h>
 #include <linux/mdev.h>
+#include <linux/vfio_mdev.h>
 
 #include "mdev_private.h"
 

drivers/vfio/mdev/vfio_mdev.c line 28

Quoted 5 lines
drivers/vfio/mdev/vfio_mdev.c +1 -1
@@ -24,16 +25,16 @@
 static int vfio_mdev_open(void *device_data)
 {
    struct mdev_device *mdev = device_data;
-   struct mdev_parent *parent = mdev->parent;
+   const struct vfio_mdev_device_ops *ops =

mdev_get_dev_ops(mdev);

Quoted 8 lines
drivers/vfio/mdev/vfio_mdev.c +1 -1
    int ret;
 
-   if (unlikely(!parent->ops->open))
+   if (unlikely(!ops->open))
            return -EINVAL;
 
    if (!try_module_get(THIS_MODULE))
            return -ENODEV;

RJM>] My understanding lately is that this call to try_module_get(THIS_MODULE) is no longer needed as is considered as a latent bug.
Quote from
https://stackoverflow.com/questions/1741415/linux-kernel-modules-when-to-use-try-module-get-module-put
:
There are a number of uses of try_module_get(THIS_MODULE) in the kernel source but most if not all of them are latent bugs that should be cleaned up.

Quoted 240 lines
Quoted 38 lines

drivers/vfio/mdev/vfio_mdev.c static int vfio_mdev_open(void *device_data) · line 47

Quoted 12 lines
drivers/vfio/mdev/vfio_mdev.c +2 -2
-   ret = parent->ops->open(mdev);
+   ret = ops->open(mdev);
    if (ret)
            module_put(THIS_MODULE);
 
@@ -43,10 +44,10 @@ static int vfio_mdev_open(void *device_data)
 static void vfio_mdev_release(void *device_data)
 {
    struct mdev_device *mdev = device_data;
-   struct mdev_parent *parent = mdev->parent;
+   const struct vfio_mdev_device_ops *ops =

mdev_get_dev_ops(mdev);

drivers/vfio/mdev/vfio_mdev.c static long vfio_mdev_unlocked_ioctl(void

Quoted 9 lines
drivers/vfio/mdev/vfio_mdev.c +2 -2
-   if (likely(parent->ops->release))
-           parent->ops->release(mdev);
+   if (likely(ops->release))
+           ops->release(mdev);
 
    module_put(THIS_MODULE);
 }
@@ -55,47 +56,47 @@ static long vfio_mdev_unlocked_ioctl(void

*device_data,

drivers/vfio/mdev/vfio_mdev.c static long vfio_mdev_unlocked_ioctl(void *device_data,

Quoted 5 lines
drivers/vfio/mdev/vfio_mdev.c +1 -1
                                 unsigned int cmd, unsigned long arg)
 {
    struct mdev_device *mdev = device_data;
-   struct mdev_parent *parent = mdev->parent;
+   const struct vfio_mdev_device_ops *ops =

mdev_get_dev_ops(mdev);

Quoted 9 lines
drivers/vfio/mdev/vfio_mdev.c +2 -2
-   if (unlikely(!parent->ops->ioctl))
+   if (unlikely(!ops->ioctl))
            return -EINVAL;
 
-   return parent->ops->ioctl(mdev, cmd, arg);
+   return ops->ioctl(mdev, cmd, arg);
 }
 
Quoted 17 lines

drivers/vfio/mdev/vfio_mdev.c static long vfio_mdev_unlocked_ioctl(void *device_data,

Quoted 6 lines
drivers/vfio/mdev/vfio_mdev.c +1 -1
 static ssize_t vfio_mdev_read(void *device_data, char __user *buf,
                          size_t count, loff_t *ppos)
 {
    struct mdev_device *mdev = device_data;
-   struct mdev_parent *parent = mdev->parent;
+   const struct vfio_mdev_device_ops *ops =

mdev_get_dev_ops(mdev);

Quoted 10 lines
drivers/vfio/mdev/vfio_mdev.c +2 -2
-   if (unlikely(!parent->ops->read))
+   if (unlikely(!ops->read))
            return -EINVAL;
 
-   return parent->ops->read(mdev, buf, count, ppos);
+   return ops->read(mdev, buf, count, ppos);
 }
 
 static ssize_t vfio_mdev_write(void *device_data, const char __user

*buf,

Quoted 47 lines

drivers/vfio/mdev/vfio_mdev.c static long vfio_mdev_unlocked_ioctl(void *device_data,

Quoted 5 lines
drivers/vfio/mdev/vfio_mdev.c +1 -1
                           size_t count, loff_t *ppos)
 {
    struct mdev_device *mdev = device_data;
-   struct mdev_parent *parent = mdev->parent;
+   const struct vfio_mdev_device_ops *ops =

mdev_get_dev_ops(mdev);

drivers/vfio/mdev/vfio_mdev.c static long vfio_mdev_unlocked_ioctl(void *device_data,

Quoted 10 lines
drivers/vfio/mdev/vfio_mdev.c +2 -2
-   if (unlikely(!parent->ops->write))
+   if (unlikely(!ops->write))
            return -EINVAL;
 
-   return parent->ops->write(mdev, buf, count, ppos);
+   return ops->write(mdev, buf, count, ppos);
 }
 
 static int vfio_mdev_mmap(void *device_data, struct vm_area_struct

*vma)

drivers/vfio/mdev/vfio_mdev.c

Quoted 4 lines
drivers/vfio/mdev/vfio_mdev.c +1 -1
 {
    struct mdev_device *mdev = device_data;
-   struct mdev_parent *parent = mdev->parent;
+   const struct vfio_mdev_device_ops *ops =

mdev_get_dev_ops(mdev);

Quoted 25 lines
drivers/vfio/mdev/vfio_mdev.c +2 -2
-   if (unlikely(!parent->ops->mmap))
+   if (unlikely(!ops->mmap))
            return -EINVAL;
 
-   return parent->ops->mmap(mdev, vma);
+   return ops->mmap(mdev, vma);
 }
 
 static const struct vfio_device_ops vfio_mdev_dev_ops = {
include/linux/mdev.h +5 -0
index 3974650c074f..3414307311f1 100644
@@ -10,6 +10,11 @@
 #ifndef MDEV_H
 #define MDEV_H
 
+#include <linux/types.h>
+#include <linux/device.h>
+#include <linux/mdev.h>
+#include <uapi/linux/uuid.h>
+
 struct mdev_device;
 
Quoted 34 lines

include/linux/mdev.h struct device *mdev_get_iommu_device(struct

Quoted 2 lines
include/linux/mdev.h +0 -0
 /*
@@ -48,30 +53,8 @@ struct device *mdev_get_iommu_device(struct

device *dev);

include/linux/mdev.h struct device *mdev_get_iommu_device(struct · line 53

Quoted 1 line
include/linux/mdev.h +0 -0
  *                 @mdev: mdev_device device structure which is

being

Quoted 29 lines
include/linux/mdev.h +2 -24
  *                        destroyed
  *                 Returns integer: success (0) or error (< 0)
- * @open:          Open mediated device.
- *                 @mdev: mediated device.
- *                 Returns integer: success (0) or error (< 0)
- * @release:               release mediated device
- *                 @mdev: mediated device.
- * @read:          Read emulation callback
- *                 @mdev: mediated device structure
- *                 @buf: read buffer
- *                 @count: number of bytes to read
- *                 @ppos: address.
- *                 Retuns number on bytes read on success or error.
- * @write:         Write emulation callback
- *                 @mdev: mediated device structure
- *                 @buf: write buffer
- *                 @count: number of bytes to be written
- *                 @ppos: address.
- *                 Retuns number on bytes written on success or error.
- * @ioctl:         IOCTL callback
- *                 @mdev: mediated device structure
- *                 @cmd: ioctl command
- *                 @arg: arguments to ioctl
- * @mmap:          mmap callback
- *                 @mdev: mediated device structure
- *                 @vma: vma structure
+ * @device_ops:         Device specific emulation callback.
+ *
  * Parent device that support mediated device should be registered

with

Quoted 102 lines

mdev

Quoted 5 lines
include/linux/mdev.h +0 -0
  * module with mdev_parent_ops structure.
  **/
@@ -83,15 +66,6 @@ struct mdev_parent_ops {
 
    int     (*create)(struct kobject *kobj, struct mdev_device *mdev);

include/linux/mdev.h struct mdev_parent_ops { · line 94

Quoted 10 lines
include/linux/mdev.h +0 -9
    int     (*remove)(struct mdev_device *mdev);
-   int     (*open)(struct mdev_device *mdev);
-   void    (*release)(struct mdev_device *mdev);
-   ssize_t (*read)(struct mdev_device *mdev, char __user *buf,
-                   size_t count, loff_t *ppos);
-   ssize_t (*write)(struct mdev_device *mdev, const char __user *buf,
-                    size_t count, loff_t *ppos);
-   long    (*ioctl)(struct mdev_device *mdev, unsigned int cmd,
-                    unsigned long arg);
-   int     (*mmap)(struct mdev_device *mdev, struct vm_area_struct

*vma);

Quoted 57 lines
include/linux/mdev.h +2 -0
 };
 
 /* interface for exporting mdev supported type attributes */
@@ -133,6 +107,8 @@ struct mdev_driver {
 
 void *mdev_get_drvdata(struct mdev_device *mdev);
 void mdev_set_drvdata(struct mdev_device *mdev, void *data);
+const void *mdev_get_dev_ops(struct mdev_device *mdev);
+void mdev_set_dev_ops(struct mdev_device *mdev, const void *ops);
 void mdev_set_class_id(struct mdev_device *mdev, u16 id);
 const guid_t *mdev_uuid(struct mdev_device *mdev);
 
include/linux/vfio_mdev.h new file +39 -0
new file mode 100644
index 000000000000..3907c5371c2b
@@ -0,0 +1,52 @@
+/* SPDX-License-Identifier: GPL-2.0-only */
+/*
+ * VFIO Mediated device definition
+ */
+
+#ifndef VFIO_MDEV_H
+#define VFIO_MDEV_H
+
+#include <linux/mdev.h>
+
+/**
+ * struct vfio_mdev_device_ops - Structure to be registered for each
+ * mdev device to register the device to vfio-mdev module.
+ *
+ * @open:          Open mediated device.
+ *                 @mdev: mediated device.
+ *                 Returns integer: success (0) or error (< 0)
+ * @release:               release mediated device
+ *                 @mdev: mediated device.
+ * @read:          Read emulation callback
+ *                 @mdev: mediated device structure
+ *                 @buf: read buffer
+ *                 @count: number of bytes to read
+ *                 @ppos: address.
+ *                 Retuns number on bytes read on success or error.
+ * @write:         Write emulation callback
+ *                 @mdev: mediated device structure
+ *                 @buf: write buffer
+ *                 @count: number of bytes to be written
+ *                 @ppos: address.
+ *                 Retuns number on bytes written on success or error.
+ * @ioctl:         IOCTL callback
+ *                 @mdev: mediated device structure
+ *                 @cmd: ioctl command
+ *                 @arg: arguments to ioctl
+ * @mmap:          mmap callback
+ *                 @mdev: mediated device structure
+ *                 @vma: vma structure
+ */

include/linux/vfio_mdev.h line 49

Quoted 10 lines
include/linux/vfio_mdev.h new file +10 -0
+struct vfio_mdev_device_ops {
+   int     (*open)(struct mdev_device *mdev);
+   void    (*release)(struct mdev_device *mdev);
+   ssize_t (*read)(struct mdev_device *mdev, char __user *buf,
+                   size_t count, loff_t *ppos);
+   ssize_t (*write)(struct mdev_device *mdev, const char __user *buf,
+                    size_t count, loff_t *ppos);
+   long    (*ioctl)(struct mdev_device *mdev, unsigned int cmd,
+                    unsigned long arg);
+   int     (*mmap)(struct mdev_device *mdev, struct vm_area_struct

*vma);

samples/vfio-mdev/mbochs.c static int mbochs_reset(struct mdev_device

Quoted 16 lines
include/linux/vfio_mdev.h new file +3 -0
+};
+
+#endif
samples/vfio-mdev/mbochs.c +1 -0
index 8a8583c892b2..b4bf29c6136c 100644
@@ -30,6 +30,7 @@
 #include <linux/iommu.h>
 #include <linux/sysfs.h>
 #include <linux/mdev.h>
+#include <linux/vfio_mdev.h>
 #include <linux/pci.h>
 #include <linux/dma-buf.h>
 #include <linux/highmem.h>
@@ -516,6 +517,8 @@ static int mbochs_reset(struct mdev_device

*mdev)

Quoted 176 lines
Quoted 6 lines

samples/vfio-mdev/mbochs.c static int mbochs_reset(struct mdev_device *mdev)

Quoted 6 lines
samples/vfio-mdev/mbochs.c +2 -0
    return 0;
 }
 
+static const struct vfio_mdev_device_ops vfio_mdev_ops;
+
 static int mbochs_create(struct kobject *kobj, struct mdev_device

*mdev)

Quoted 168 lines

samples/vfio-mdev/mbochs.c static int mbochs_create(struct kobject *kobj,

Quoted 3 lines
samples/vfio-mdev/mbochs.c +0 -0
 {
    const struct mbochs_type *type = mbochs_find_type(kobj);
@@ -562,6 +565,7 @@ static int mbochs_create(struct kobject *kobj,

struct mdev_device *mdev)

samples/vfio-mdev/mbochs.c static struct attribute_group

Quoted 8 lines
samples/vfio-mdev/mbochs.c +1 -0
    mbochs_used_mbytes += type->mbytes;
    mdev_set_class_id(mdev, MDEV_ID_VFIO);
+   mdev_set_dev_ops(mdev, &vfio_mdev_ops);
    return 0;
 
 err_mem:
@@ -1419,12 +1423,7 @@ static struct attribute_group

*mdev_type_groups[] = {

samples/vfio-mdev/mbochs.c static const struct mdev_parent_ops

Quoted 14 lines
samples/vfio-mdev/mbochs.c +1 -6
    NULL,
 };
 
-static const struct mdev_parent_ops mdev_fops = {
-   .owner                  = THIS_MODULE,
-   .mdev_attr_groups       = mdev_dev_groups,
-   .supported_type_groups  = mdev_type_groups,
-   .create                 = mbochs_create,
-   .remove                 = mbochs_remove,
+static const struct vfio_mdev_device_ops vfio_mdev_ops = {
    .open                   = mbochs_open,
    .release                = mbochs_close,
    .read                   = mbochs_read,
@@ -1433,6 +1432,14 @@ static const struct mdev_parent_ops

mdev_fops = {

samples/vfio-mdev/mdpy.c static int mdpy_create(struct kobject *kobj, struct

Quoted 36 lines
samples/vfio-mdev/mbochs.c +8 -0
    .mmap                   = mbochs_mmap,
 };
 
+static const struct mdev_parent_ops mdev_fops = {
+   .owner                  = THIS_MODULE,
+   .mdev_attr_groups       = mdev_dev_groups,
+   .supported_type_groups  = mdev_type_groups,
+   .create                 = mbochs_create,
+   .remove                 = mbochs_remove,
+};
+
 static const struct file_operations vd_fops = {
    .owner          = THIS_MODULE,
 };
samples/vfio-mdev/mdpy.c +3 -0
index 88d7e76f3836..80c2df531326 100644
@@ -26,6 +26,7 @@
 #include <linux/iommu.h>
 #include <linux/sysfs.h>
 #include <linux/mdev.h>
+#include <linux/vfio_mdev.h>
 #include <linux/pci.h>
 #include <drm/drm_fourcc.h>
 #include "mdpy-defs.h"
@@ -226,6 +227,8 @@ static int mdpy_reset(struct mdev_device *mdev)
    return 0;
 }
 
+static const struct vfio_mdev_device_ops vfio_mdev_ops;
+
 static int mdpy_create(struct kobject *kobj, struct mdev_device *mdev)
 {
    const struct mdpy_type *type = mdpy_find_type(kobj);
@@ -270,6 +273,7 @@ static int mdpy_create(struct kobject *kobj, struct

mdev_device *mdev)

samples/vfio-mdev/mdpy.c static struct attribute_group

Quoted 8 lines
samples/vfio-mdev/mdpy.c +1 -0
    mdpy_count++;
    mdev_set_class_id(mdev, MDEV_ID_VFIO);
+   mdev_set_dev_ops(mdev, &vfio_mdev_ops);
    return 0;
 }
 
@@ -726,12 +730,7 @@ static struct attribute_group

*mdev_type_groups[] = {

samples/vfio-mdev/mdpy.c static const struct mdev_parent_ops mdev_fops

Quoted 14 lines
samples/vfio-mdev/mdpy.c +1 -6
    NULL,
 };
 
-static const struct mdev_parent_ops mdev_fops = {
-   .owner                  = THIS_MODULE,
-   .mdev_attr_groups       = mdev_dev_groups,
-   .supported_type_groups  = mdev_type_groups,
-   .create                 = mdpy_create,
-   .remove                 = mdpy_remove,
+static const struct vfio_mdev_device_ops vfio_mdev_ops = {
    .open                   = mdpy_open,
    .release                = mdpy_close,
    .read                   = mdpy_read,
@@ -740,6 +739,14 @@ static const struct mdev_parent_ops mdev_fops

= {

samples/vfio-mdev/mtty.c static ssize_t mdev_access(struct mdev_device

Quoted 27 lines
samples/vfio-mdev/mdpy.c +8 -0
    .mmap                   = mdpy_mmap,
 };
 
+static const struct mdev_parent_ops mdev_fops = {
+   .owner                  = THIS_MODULE,
+   .mdev_attr_groups       = mdev_dev_groups,
+   .supported_type_groups  = mdev_type_groups,
+   .create                 = mdpy_create,
+   .remove                 = mdpy_remove,
+};
+
 static const struct file_operations vd_fops = {
    .owner          = THIS_MODULE,
 };
samples/vfio-mdev/mtty.c +1 -0
index 4e0735143b69..2db860ccb02c 100644
@@ -27,6 +27,7 @@
 #include <linux/ctype.h>
 #include <linux/file.h>
 #include <linux/mdev.h>
+#include <linux/vfio_mdev.h>
 #include <linux/pci.h>
 #include <linux/serial.h>
 #include <uapi/linux/serial_reg.h>
@@ -723,6 +724,8 @@ static ssize_t mdev_access(struct mdev_device

*mdev, u8 *buf, size_t count,

samples/vfio-mdev/mtty.c static int mtty_create(struct kobject *kobj, struct

Quoted 9 lines
samples/vfio-mdev/mtty.c +2 -0
    return ret;
 }
 
+static const struct vfio_mdev_device_ops vfio_dev_ops;
+
 static int mtty_create(struct kobject *kobj, struct mdev_device *mdev)
 {
    struct mdev_state *mdev_state;
@@ -771,6 +774,7 @@ static int mtty_create(struct kobject *kobj, struct

mdev_device *mdev)

samples/vfio-mdev/mtty.c static struct attribute_group

Quoted 8 lines
samples/vfio-mdev/mtty.c +1 -0
    mutex_unlock(&mdev_list_lock);
 
    mdev_set_class_id(mdev, MDEV_ID_VFIO);
+   mdev_set_dev_ops(mdev, &vfio_dev_ops);
    return 0;
 }
 
@@ -1411,6 +1415,14 @@ static struct attribute_group

*mdev_type_groups[] = {

samples/vfio-mdev/mtty.c static const struct mdev_parent_ops

Quoted 15 lines
samples/vfio-mdev/mtty.c +8 -0
    NULL,
 };
 
+static const struct vfio_mdev_device_ops vfio_dev_ops = {
+   .open                   = mtty_open,
+   .release                = mtty_close,
+   .read                   = mtty_read,
+   .write                  = mtty_write,
+   .ioctl                  = mtty_ioctl,
+};
+
 static const struct mdev_parent_ops mdev_fops = {
    .owner                  = THIS_MODULE,
    .dev_attr_groups        = mtty_dev_groups,
@@ -1418,11 +1430,6 @@ static const struct mdev_parent_ops

mdev_fops = {

samples/vfio-mdev/mtty.c static const struct mdev_parent_ops mdev_fops = { · line 1435

Quoted 11 lines
samples/vfio-mdev/mtty.c +0 -5
    .supported_type_groups  = mdev_type_groups,
    .create                 = mtty_create,
    .remove                 = mtty_remove,
-   .open                   = mtty_open,
-   .release                = mtty_close,
-   .read                   = mtty_read,
-   .write                  = mtty_write,
-   .ioctl                  = mtty_ioctl,
 };
 
 static void mtty_device_release(struct device *dev)

_______________________________________________
intel-gvt-dev mailing list
[email protected]
https://lists.freedesktop.org/mailman/listinfo/intel-gvt-dev

_______________________________________________
Virtualization mailing list
[email protected]
https://lists.linuxfoundation.org/mailman/listinfo/virtualization

#21 lore raw

Re: [PATCH V2 5/8] mdev: introduce device specific ops

Quoted 32 lines · On Wed, 25 Sep 2019 10:11:00 -0400 Rob Miller <[email protected]> wrote:
Quoted 22 lines
Quoted 22 lines

Review

Quoted 1 line · On Tue, 24 Sep 2019 21:53:29 +0800 Jason Wang <[email protected]> wrote:
(quoted diff) +0 -0

b/drivers/vfio/mdev/vfio_mdev.c

Quoted 12 lines

index 891cf83a2d9a..95efa054442f 100644

drivers/vfio/mdev/vfio_mdev.c +1 -0
@@ -14,6 +14,7 @@
 #include <linux/slab.h>
 #include <linux/vfio.h>
 #include <linux/mdev.h>
+#include <linux/vfio_mdev.h>
 
 #include "mdev_private.h"
 

drivers/vfio/mdev/vfio_mdev.c line 28

Quoted 5 lines
drivers/vfio/mdev/vfio_mdev.c +1 -1
@@ -24,16 +25,16 @@
 static int vfio_mdev_open(void *device_data)
 {
    struct mdev_device *mdev = device_data;
-   struct mdev_parent *parent = mdev->parent;
+   const struct vfio_mdev_device_ops *ops =  

mdev_get_dev_ops(mdev);

Quoted 9 lines
Quoted 8 lines

drivers/vfio/mdev/vfio_mdev.c line 35

Quoted 8 lines
drivers/vfio/mdev/vfio_mdev.c +1 -1
    int ret;
 
-   if (unlikely(!parent->ops->open))
+   if (unlikely(!ops->open))
            return -EINVAL;
 
    if (!try_module_get(THIS_MODULE))
            return -ENODEV;  
Quoted 9 lines

RJM>] My understanding lately is that this call to try_module_get(THIS_MODULE) is no longer needed as is considered as a latent bug.
Quote from
https://stackoverflow.com/questions/1741415/linux-kernel-modules-when-to-use-try-module-get-module-put
:
There are a number of uses of try_module_get(THIS_MODULE) in the kernel source but most if not all of them are latent bugs that should be cleaned up.

This use seems to fall exactly into the case where it is necessary, the open here is not a direct VFS call, it's an internal interface between modules. The user is interacting with filesystem objects from the vfio module and the module reference we're trying to acquire here is to the vfio-mdev module. Thanks,

Alex
_______________________________________________
dri-devel mailing list
[email protected]
https://lists.freedesktop.org/mailman/listinfo/dri-devel

#22 lore raw

Re: [PATCH V2 5/8] mdev: introduce device specific ops

Quoted 44 lines · Alex Williamson
Quoted 32 lines · On Wed, 25 Sep 2019 10:11:00 -0400 Rob Miller <[email protected]> wrote:
Quoted 22 lines
Quoted 22 lines

Review

Quoted 1 line · On Tue, 24 Sep 2019 21:53:29 +0800 Jason Wang <[email protected]> wrote:
(quoted diff) +0 -0

b/drivers/vfio/mdev/vfio_mdev.c

Quoted 12 lines

index 891cf83a2d9a..95efa054442f 100644

drivers/vfio/mdev/vfio_mdev.c +1 -0
@@ -14,6 +14,7 @@
 #include <linux/slab.h>
 #include <linux/vfio.h>
 #include <linux/mdev.h>
+#include <linux/vfio_mdev.h>
 
 #include "mdev_private.h"
 

drivers/vfio/mdev/vfio_mdev.c line 28

Quoted 5 lines
drivers/vfio/mdev/vfio_mdev.c +1 -1
@@ -24,16 +25,16 @@
 static int vfio_mdev_open(void *device_data)
 {
    struct mdev_device *mdev = device_data;
-   struct mdev_parent *parent = mdev->parent;
+   const struct vfio_mdev_device_ops *ops =  

mdev_get_dev_ops(mdev);

Quoted 9 lines
Quoted 8 lines

drivers/vfio/mdev/vfio_mdev.c line 35

Quoted 8 lines
drivers/vfio/mdev/vfio_mdev.c +1 -1
    int ret;
 
-   if (unlikely(!parent->ops->open))
+   if (unlikely(!ops->open))
            return -EINVAL;
 
    if (!try_module_get(THIS_MODULE))
            return -ENODEV;  
Quoted 9 lines

RJM>] My understanding lately is that this call to try_module_get(THIS_MODULE) is no longer needed as is considered as a latent bug.
Quote from
https://stackoverflow.com/questions/1741415/linux-kernel-modules-when-to-use-try-module-get-module-put
:
There are a number of uses of try_module_get(THIS_MODULE) in the kernel source but most if not all of them are latent bugs that should be cleaned up.

Quoted 7 lines

This use seems to fall exactly into the case where it is necessary, the open here is not a direct VFS call, it's an internal interface between modules. The user is interacting with filesystem objects from the vfio module and the module reference we're trying to acquire here is to the vfio-mdev module. Thanks,

Alex

I think the latent bug refers not to module get per se, but to the module_put tied to it. E.g.:

static void vfio_mdev_release(void *device_data)

 {
        struct mdev_device *mdev = device_data;
        struct mdev_parent *parent = mdev->parent;

        if (likely(parent->ops->release))
                parent->ops->release(mdev);

        module_put(THIS_MODULE);

Does anything prevent the module from unloading at this point? if not then ...

}

it looks like the implicit return (with instructions for argument pop and functuon return) here can get overwritten on module unload, causing a crash when executed.

IOW there's generally no way for module to keep a reference to itself: it can take a reference but it needs someone else to keep it and put.

Signature · 1 line
MST
#23 lore raw

Re: [PATCH V2 5/8] mdev: introduce device specific ops

Quoted 73 lines · On Thu, 26 Sep 2019 11:46:55 -0400 "Michael S. Tsirkin" <[email protected]> wrote:
Quoted 44 lines · Alex Williamson
Quoted 32 lines · On Wed, 25 Sep 2019 10:11:00 -0400 Rob Miller <[email protected]> wrote:
Quoted 22 lines
Quoted 22 lines

Review

Quoted 1 line · On Tue, 24 Sep 2019 21:53:29 +0800 Jason Wang <[email protected]> wrote:
(quoted diff) +0 -0

b/drivers/vfio/mdev/vfio_mdev.c

Quoted 12 lines

index 891cf83a2d9a..95efa054442f 100644

drivers/vfio/mdev/vfio_mdev.c +1 -0
@@ -14,6 +14,7 @@
 #include <linux/slab.h>
 #include <linux/vfio.h>
 #include <linux/mdev.h>
+#include <linux/vfio_mdev.h>
 
 #include "mdev_private.h"
 

drivers/vfio/mdev/vfio_mdev.c line 28

Quoted 5 lines
drivers/vfio/mdev/vfio_mdev.c +1 -1
@@ -24,16 +25,16 @@
 static int vfio_mdev_open(void *device_data)
 {
    struct mdev_device *mdev = device_data;
-   struct mdev_parent *parent = mdev->parent;
+   const struct vfio_mdev_device_ops *ops =    

mdev_get_dev_ops(mdev);

Quoted 9 lines
Quoted 8 lines

drivers/vfio/mdev/vfio_mdev.c line 35

Quoted 8 lines
drivers/vfio/mdev/vfio_mdev.c +1 -1
    int ret;
 
-   if (unlikely(!parent->ops->open))
+   if (unlikely(!ops->open))
            return -EINVAL;
 
    if (!try_module_get(THIS_MODULE))
            return -ENODEV;    
Quoted 9 lines

RJM>] My understanding lately is that this call to try_module_get(THIS_MODULE) is no longer needed as is considered as a latent bug.
Quote from
https://stackoverflow.com/questions/1741415/linux-kernel-modules-when-to-use-try-module-get-module-put
:
There are a number of uses of try_module_get(THIS_MODULE) in the kernel source but most if not all of them are latent bugs that should be cleaned up.

Quoted 7 lines

This use seems to fall exactly into the case where it is necessary, the open here is not a direct VFS call, it's an internal interface between modules. The user is interacting with filesystem objects from the vfio module and the module reference we're trying to acquire here is to the vfio-mdev module. Thanks,

Alex

I think the latent bug refers not to module get per se, but to the module_put tied to it. E.g.:

static void vfio_mdev_release(void *device_data)

 {
        struct mdev_device *mdev = device_data;
        struct mdev_parent *parent = mdev->parent;

        if (likely(parent->ops->release))
                parent->ops->release(mdev);

        module_put(THIS_MODULE);

Does anything prevent the module from unloading at this point? if not then ...

}

Quoted 7 lines

it looks like the implicit return (with instructions for argument pop and functuon return) here can get overwritten on module unload, causing a crash when executed.

IOW there's generally no way for module to keep a reference to itself: it can take a reference but it needs someone else to keep it and put.

I'd always assumed this would exit cleanly, but perhaps there is a latent race there. In any case, taking a module reference within the module in this case is better than not doing so, as the latter would potentially allow the module to be removed at any point in time, while the former only seems to expose acquire and release gaps. Add it to the todo list. Thanks,

Alex

#24 lore raw

Re: [PATCH V2 5/8] mdev: introduce device specific ops

Quoted 84 lines · Alex Williamson
Quoted 73 lines · On Thu, 26 Sep 2019 11:46:55 -0400 "Michael S. Tsirkin" <[email protected]> wrote:
Quoted 44 lines · Alex Williamson
Quoted 32 lines · On Wed, 25 Sep 2019 10:11:00 -0400 Rob Miller <[email protected]> wrote:
Quoted 22 lines
Quoted 22 lines

Review

Quoted 1 line · On Tue, 24 Sep 2019 21:53:29 +0800 Jason Wang <[email protected]> wrote:
(quoted diff) +0 -0

b/drivers/vfio/mdev/vfio_mdev.c

Quoted 12 lines

index 891cf83a2d9a..95efa054442f 100644

drivers/vfio/mdev/vfio_mdev.c +1 -0
@@ -14,6 +14,7 @@
 #include <linux/slab.h>
 #include <linux/vfio.h>
 #include <linux/mdev.h>
+#include <linux/vfio_mdev.h>
 
 #include "mdev_private.h"
 

drivers/vfio/mdev/vfio_mdev.c line 28

Quoted 5 lines
drivers/vfio/mdev/vfio_mdev.c +1 -1
@@ -24,16 +25,16 @@
 static int vfio_mdev_open(void *device_data)
 {
    struct mdev_device *mdev = device_data;
-   struct mdev_parent *parent = mdev->parent;
+   const struct vfio_mdev_device_ops *ops =    

mdev_get_dev_ops(mdev);

Quoted 9 lines
Quoted 8 lines

drivers/vfio/mdev/vfio_mdev.c line 35

Quoted 8 lines
drivers/vfio/mdev/vfio_mdev.c +1 -1
    int ret;
 
-   if (unlikely(!parent->ops->open))
+   if (unlikely(!ops->open))
            return -EINVAL;
 
    if (!try_module_get(THIS_MODULE))
            return -ENODEV;    
Quoted 9 lines

RJM>] My understanding lately is that this call to try_module_get(THIS_MODULE) is no longer needed as is considered as a latent bug.
Quote from
https://stackoverflow.com/questions/1741415/linux-kernel-modules-when-to-use-try-module-get-module-put
:
There are a number of uses of try_module_get(THIS_MODULE) in the kernel source but most if not all of them are latent bugs that should be cleaned up.

Quoted 7 lines

This use seems to fall exactly into the case where it is necessary, the open here is not a direct VFS call, it's an internal interface between modules. The user is interacting with filesystem objects from the vfio module and the module reference we're trying to acquire here is to the vfio-mdev module. Thanks,

Alex

I think the latent bug refers not to module get per se, but to the module_put tied to it. E.g.:

static void vfio_mdev_release(void *device_data)

 {
        struct mdev_device *mdev = device_data;
        struct mdev_parent *parent = mdev->parent;

        if (likely(parent->ops->release))
                parent->ops->release(mdev);

        module_put(THIS_MODULE);

Does anything prevent the module from unloading at this point? if not then ...

}

Quoted 7 lines

it looks like the implicit return (with instructions for argument pop and functuon return) here can get overwritten on module unload, causing a crash when executed.

IOW there's generally no way for module to keep a reference to itself: it can take a reference but it needs someone else to keep it and put.

Quoted 8 lines

I'd always assumed this would exit cleanly, but perhaps there is a latent race there. In any case, taking a module reference within the module in this case is better than not doing so, as the latter would potentially allow the module to be removed at any point in time, while the former only seems to expose acquire and release gaps. Add it to the todo list. Thanks,

Alex

Right. I agree with the stack overflow quote: as this example seems to show this is a latent bug.

But I also agree that just removing the reference isn't the right way to clean it up.

Signature · 5 lines
MST
_______________________________________________
dri-devel mailing list
[email protected]
https://lists.freedesktop.org/mailman/listinfo/dri-devel
#25 lore raw

Re: [PATCH V2 5/8] mdev: introduce device specific ops

In reply to Alex Williamson
Quoted 84 lines · Alex Williamson
Quoted 73 lines · On Thu, 26 Sep 2019 11:46:55 -0400 "Michael S. Tsirkin" <[email protected]> wrote:
Quoted 44 lines · Alex Williamson
Quoted 32 lines · On Wed, 25 Sep 2019 10:11:00 -0400 Rob Miller <[email protected]> wrote:
Quoted 22 lines
Quoted 22 lines

Review

Quoted 1 line · On Tue, 24 Sep 2019 21:53:29 +0800 Jason Wang <[email protected]> wrote:
(quoted diff) +0 -0

b/drivers/vfio/mdev/vfio_mdev.c

Quoted 12 lines

index 891cf83a2d9a..95efa054442f 100644

drivers/vfio/mdev/vfio_mdev.c +1 -0
@@ -14,6 +14,7 @@
 #include <linux/slab.h>
 #include <linux/vfio.h>
 #include <linux/mdev.h>
+#include <linux/vfio_mdev.h>
 
 #include "mdev_private.h"
 

drivers/vfio/mdev/vfio_mdev.c line 28

Quoted 5 lines
drivers/vfio/mdev/vfio_mdev.c +1 -1
@@ -24,16 +25,16 @@
 static int vfio_mdev_open(void *device_data)
 {
    struct mdev_device *mdev = device_data;
-   struct mdev_parent *parent = mdev->parent;
+   const struct vfio_mdev_device_ops *ops =    

mdev_get_dev_ops(mdev);

Quoted 9 lines
Quoted 8 lines

drivers/vfio/mdev/vfio_mdev.c line 35

Quoted 8 lines
drivers/vfio/mdev/vfio_mdev.c +1 -1
    int ret;
 
-   if (unlikely(!parent->ops->open))
+   if (unlikely(!ops->open))
            return -EINVAL;
 
    if (!try_module_get(THIS_MODULE))
            return -ENODEV;    
Quoted 9 lines

RJM>] My understanding lately is that this call to try_module_get(THIS_MODULE) is no longer needed as is considered as a latent bug.
Quote from
https://stackoverflow.com/questions/1741415/linux-kernel-modules-when-to-use-try-module-get-module-put
:
There are a number of uses of try_module_get(THIS_MODULE) in the kernel source but most if not all of them are latent bugs that should be cleaned up.

Quoted 7 lines

This use seems to fall exactly into the case where it is necessary, the open here is not a direct VFS call, it's an internal interface between modules. The user is interacting with filesystem objects from the vfio module and the module reference we're trying to acquire here is to the vfio-mdev module. Thanks,

Alex

I think the latent bug refers not to module get per se, but to the module_put tied to it. E.g.:

static void vfio_mdev_release(void *device_data)

 {
        struct mdev_device *mdev = device_data;
        struct mdev_parent *parent = mdev->parent;

        if (likely(parent->ops->release))
                parent->ops->release(mdev);

        module_put(THIS_MODULE);

Does anything prevent the module from unloading at this point? if not then ...

}

Quoted 7 lines

it looks like the implicit return (with instructions for argument pop and functuon return) here can get overwritten on module unload, causing a crash when executed.

IOW there's generally no way for module to keep a reference to itself: it can take a reference but it needs someone else to keep it and put.

Quoted 8 lines

I'd always assumed this would exit cleanly, but perhaps there is a latent race there. In any case, taking a module reference within the module in this case is better than not doing so, as the latter would potentially allow the module to be removed at any point in time, while the former only seems to expose acquire and release gaps. Add it to the todo list. Thanks,

Alex

Right. I agree with the stack overflow quote: as this example seems to show this is a latent bug.

But I also agree that just removing the reference isn't the right way to clean it up.

Signature · 1 line
MST
#26 lore raw

Re: [PATCH V2 5/8] mdev: introduce device specific ops

In reply to Rob Miller via Virtualization
Quoted 32 lines · On Wed, 25 Sep 2019 10:11:00 -0400 Rob Miller <[email protected]> wrote:
Quoted 22 lines
Quoted 22 lines

Review

Quoted 1 line · On Tue, 24 Sep 2019 21:53:29 +0800 Jason Wang <[email protected]> wrote:
(quoted diff) +0 -0

b/drivers/vfio/mdev/vfio_mdev.c

Quoted 12 lines

index 891cf83a2d9a..95efa054442f 100644

drivers/vfio/mdev/vfio_mdev.c +1 -0
@@ -14,6 +14,7 @@
 #include <linux/slab.h>
 #include <linux/vfio.h>
 #include <linux/mdev.h>
+#include <linux/vfio_mdev.h>
 
 #include "mdev_private.h"
 

drivers/vfio/mdev/vfio_mdev.c line 28

Quoted 5 lines
drivers/vfio/mdev/vfio_mdev.c +1 -1
@@ -24,16 +25,16 @@
 static int vfio_mdev_open(void *device_data)
 {
    struct mdev_device *mdev = device_data;
-   struct mdev_parent *parent = mdev->parent;
+   const struct vfio_mdev_device_ops *ops =  

mdev_get_dev_ops(mdev);

Quoted 9 lines
Quoted 8 lines

drivers/vfio/mdev/vfio_mdev.c line 35

Quoted 8 lines
drivers/vfio/mdev/vfio_mdev.c +1 -1
    int ret;
 
-   if (unlikely(!parent->ops->open))
+   if (unlikely(!ops->open))
            return -EINVAL;
 
    if (!try_module_get(THIS_MODULE))
            return -ENODEV;  
Quoted 9 lines

RJM>] My understanding lately is that this call to try_module_get(THIS_MODULE) is no longer needed as is considered as a latent bug.
Quote from
https://stackoverflow.com/questions/1741415/linux-kernel-modules-when-to-use-try-module-get-module-put
:
There are a number of uses of try_module_get(THIS_MODULE) in the kernel source but most if not all of them are latent bugs that should be cleaned up.

This use seems to fall exactly into the case where it is necessary, the open here is not a direct VFS call, it's an internal interface between modules. The user is interacting with filesystem objects from the vfio module and the module reference we're trying to acquire here is to the vfio-mdev module. Thanks,

Alex

author
#27 lore raw

Re: [PATCH V2 5/8] mdev: introduce device specific ops

In reply to Rob Miller via Virtualization
Quoted 14 lines · Rob Miller
Quoted 14 lines
Quoted 3 lines

     mdev_set_class_id(mdev, MDEV_ID_VFIO);
+   mdev_set_dev_ops(mdev, &intel_vfio_vgpu_dev_ops);

This seems rather unrefined.  We're registering interdependent data in separate calls.  All drivers need to make both of these calls.  I'm not sure if this is a good idea, but what if we had:

static const struct vfio_mdev_device_ops intel_vfio_vgpu_dev_ops = { .id                     = MDEV_ID_VFIO,
       .open                   = intel_vgpu_open, .release                = intel_vgpu_release,
         ...

Quoted 7 lines
Quoted 1 line

And the set function passed &intel_vfio_vgpu_dev_ops.id

<http://intel_vfio_vgpu_dev_ops.id> and the mdev

Quoted 1 line

bus drivers used container_of to get to their callbacks?

or just make it explicit? e.g.

mdev_set_class(mdev, MDEV_ID_VFIO, &intel_vfio_vgpu_dev_ops);

Yes, this seems even better.

Will do this in V3.

Thanks

_______________________________________________
Virtualization mailing list
[email protected]
https://lists.linuxfoundation.org/mailman/listinfo/virtualization

#28 lore raw

RE: [PATCH V2 5/8] mdev: introduce device specific ops

In reply to Alex Williamson
Quoted 41 lines

From: Alex Williamson
Sent: Wednesday, September 25, 2019 7:07 AM

Quoted 25 lines · On Tue, 24 Sep 2019 21:53:29 +0800 Jason Wang <[email protected]> wrote:

Currently, except for the create and remove, the rest of mdev_parent_ops is designed for vfio-mdev driver only and may not help for kernel mdev driver. With the help of class id, this patch introduces device specific callbacks inside mdev_device structure. This allows different set of callback to be used by vfio-mdev and virtio-mdev.

---
 .../driver-api/vfio-mediated-device.rst       |  4 +-
 MAINTAINERS                                   |  1 +
 drivers/gpu/drm/i915/gvt/kvmgt.c              | 17 +++---
 drivers/s390/cio/vfio_ccw_ops.c               | 17 ++++--
 drivers/s390/crypto/vfio_ap_ops.c             | 13 +++--
 drivers/vfio/mdev/mdev_core.c                 | 12 +++++
 drivers/vfio/mdev/mdev_private.h              |  1 +
 drivers/vfio/mdev/vfio_mdev.c                 | 37 ++++++-------
 include/linux/mdev.h                          | 42 ++++-----------
 include/linux/vfio_mdev.h                     | 52 +++++++++++++++++++
 samples/vfio-mdev/mbochs.c                    | 19 ++++---
 samples/vfio-mdev/mdpy.c                      | 19 ++++---
 samples/vfio-mdev/mtty.c                      | 17 ++++--
 13 files changed, 168 insertions(+), 83 deletions(-)
 create mode 100644 include/linux/vfio_mdev.h

Review

Quoted 1 line
(quoted diff) +0 -0

b/Documentation/driver-api/vfio-mediated-device.rst

Documentation/driver-api/vfio-mediated-device.rst callbacks per mdev parent device, per mdev type,

Quoted 4 lines

index a5bdc60d62a1..d50425b368bb 100644

Documentation/driver-api/vfio-mediated-device.rst +0 -0
@@ -152,7 +152,9 @@ callbacks per mdev parent device, per mdev type,

or any other categorization.

Quoted 3 lines
Documentation/driver-api/vfio-mediated-device.rst +0 -0
 Vendor drivers are expected to be fully asynchronous in this respect or
 provide their own internal resource protection.)
 
Quoted 9 lines

Documentation/driver-api/vfio-mediated-device.rst callbacks per mdev parent device, per mdev type, or any other categorization. · line 157

Quoted 4 lines
Documentation/driver-api/vfio-mediated-device.rst +3 -1
-The callbacks in the mdev_parent_ops structure are as follows:
+The device specific callbacks are referred through device_ops pointer
+in mdev_parent_ops. For vfio-mdev device, its callbacks in device_ops
+are as follows:

This is not accurate. device_ops is now on the mdev_device and is an mdev bus driver specific structure of callbacks that must be registered for each mdev device by the parent driver during the create callback. There's a one to one mapping of class_id to mdev_device_ops callbacks.

there is also a mistake in include/Linux/mdev.h, where device_ops is still part of mdev_parent_ops in the comment line.

Quoted 60 lines

That also suggests to me that we could be more clever in registering both of these with mdev-core. Can we embed the class_id in the ops structure in a common way so that the core can extract it and the bus drivers can access their specific callbacks?

Review

Quoted 15 lines
Documentation/driver-api/vfio-mediated-device.rst +0 -0
 * open: open callback of mediated device
 * close: close callback of mediated device
MAINTAINERS +1 -0
index b2326dece28e..89832b316500 100644
@@ -17075,6 +17075,7 @@ S:	Maintained
 F:	Documentation/driver-api/vfio-mediated-device.rst
 F:	drivers/vfio/mdev/
 F:	include/linux/mdev.h
+F:	include/linux/vfio_mdev.h
 F:	samples/vfio-mdev/
 
 VFIO PLATFORM DRIVER
(quoted diff) +0 -0

b/drivers/gpu/drm/i915/gvt/kvmgt.c

Quoted 12 lines

index f793252a3d2a..b274f5ee481f 100644

drivers/gpu/drm/i915/gvt/kvmgt.c +1 -0
@@ -42,6 +42,7 @@
 #include <linux/kvm_host.h>
 #include <linux/vfio.h>
 #include <linux/mdev.h>
+#include <linux/vfio_mdev.h>
 #include <linux/debugfs.h>
 
 #include <linux/nospec.h>

drivers/gpu/drm/i915/gvt/kvmgt.c static void kvmgt_put_vfio_device(void *vgpu) · line 649

Quoted 6 lines
drivers/gpu/drm/i915/gvt/kvmgt.c +2 -0
@@ -643,6 +644,8 @@ static void kvmgt_put_vfio_device(void *vgpu)
 	vfio_device_put(((struct intel_vgpu *)vgpu)->vdev.vfio_device);
 }
 
+static struct vfio_mdev_device_ops intel_vfio_vgpu_dev_ops;
+
 static int intel_vgpu_create(struct kobject *kobj, struct mdev_device

*mdev)

drivers/gpu/drm/i915/gvt/kvmgt.c static int intel_vgpu_create(struct kobject *kobj,

Quoted 3 lines
drivers/gpu/drm/i915/gvt/kvmgt.c +0 -0
 {
 	struct intel_vgpu *vgpu = NULL;
@@ -679,6 +682,7 @@ static int intel_vgpu_create(struct kobject *kobj,

struct mdev_device *mdev)

drivers/gpu/drm/i915/gvt/kvmgt.c static int intel_vgpu_create(struct kobject *kobj, struct mdev_device *mdev) · line 685

Quoted 4 lines
drivers/gpu/drm/i915/gvt/kvmgt.c +1 -0
 	ret = 0;
 
 	mdev_set_class_id(mdev, MDEV_ID_VFIO);
+	mdev_set_dev_ops(mdev, &intel_vfio_vgpu_dev_ops);

This seems rather unrefined. We're registering interdependent data in separate calls. All drivers need to make both of these calls. I'm not sure if this is a good idea, but what if we had:

static const struct vfio_mdev_device_ops intel_vfio_vgpu_dev_ops = {
	.id			= MDEV_ID_VFIO,
drivers/gpu/drm/i915/gvt/kvmgt.c +0 -0
 	.open			= intel_vgpu_open,
 	.release		= intel_vgpu_release,
...
Quoted 2 lines

And the set function passed &intel_vfio_vgpu_dev_ops.id and the mdev bus drivers used container_of to get to their callbacks?

or just make it explicit? e.g.

mdev_set_class(mdev, MDEV_ID_VFIO, &intel_vfio_vgpu_dev_ops);

Quoted 635 lines

drivers/gpu/drm/i915/gvt/kvmgt.c static const struct attribute_group

Quoted 5 lines
drivers/gpu/drm/i915/gvt/kvmgt.c +0 -0
 out:
 	return ret;
 }
@@ -1601,20 +1605,21 @@ static const struct attribute_group

*intel_vgpu_groups[] = {

Quoted 9 lines
drivers/gpu/drm/i915/gvt/kvmgt.c +1 -5
 	NULL,
 };
 
-static struct mdev_parent_ops intel_vgpu_ops = {
-	.mdev_attr_groups       = intel_vgpu_groups,
-	.create			= intel_vgpu_create,
-	.remove			= intel_vgpu_remove,
-
+static struct vfio_mdev_device_ops intel_vfio_vgpu_dev_ops = {

drivers/gpu/drm/i915/gvt/kvmgt.c static const struct attribute_group *intel_vgpu_groups[] = { · line 1617

Quoted 10 lines
drivers/gpu/drm/i915/gvt/kvmgt.c +1 -1
 	.open			= intel_vgpu_open,
 	.release		= intel_vgpu_release,
-
 	.read			= intel_vgpu_read,
 	.write			= intel_vgpu_write,
 	.mmap			= intel_vgpu_mmap,
 	.ioctl			= intel_vgpu_ioctl,
 };
 
+static struct mdev_parent_ops intel_vgpu_ops = {

These could maybe be made const at the same time. Thanks,

Alex

Review

Quoted 9 lines
drivers/gpu/drm/i915/gvt/kvmgt.c +5 -0
+	.mdev_attr_groups       = intel_vgpu_groups,
+	.create			= intel_vgpu_create,
+	.remove			= intel_vgpu_remove,
+};
+
 static int kvmgt_host_init(struct device *dev, void *gvt, const void *ops)
 {
 	struct attribute **kvm_type_attrs;
(quoted diff) +0 -0

b/drivers/s390/cio/vfio_ccw_ops.c

drivers/s390/cio/vfio_ccw_ops.c static struct attribute_group *mdev_type_groups[]

Quoted 12 lines

index d258ef1fedb9..329d53c1f46b 100644

drivers/s390/cio/vfio_ccw_ops.c +1 -0
@@ -12,6 +12,7 @@
 
 #include <linux/vfio.h>
 #include <linux/mdev.h>
+#include <linux/vfio_mdev.h>
 #include <linux/nospec.h>
 #include <linux/slab.h>
 
@@ -110,6 +111,8 @@ static struct attribute_group *mdev_type_groups[]

= {

drivers/s390/cio/vfio_ccw_ops.c static struct attribute_group *mdev_type_groups[] = {

Quoted 6 lines
drivers/s390/cio/vfio_ccw_ops.c +2 -0
 	NULL,
 };
 
+static const struct vfio_mdev_device_ops vfio_mdev_ops;
+
 static int vfio_ccw_mdev_create(struct kobject *kobj, struct mdev_device

*mdev)

drivers/s390/cio/vfio_ccw_ops.c static int vfio_ccw_mdev_create(struct kobject

Quoted 3 lines
drivers/s390/cio/vfio_ccw_ops.c +0 -0
 {
 	struct vfio_ccw_private *private =
@@ -130,6 +133,7 @@ static int vfio_ccw_mdev_create(struct kobject

*kobj, struct mdev_device *mdev)

drivers/s390/cio/vfio_ccw_ops.c static ssize_t vfio_ccw_mdev_ioctl(struct

Quoted 8 lines
drivers/s390/cio/vfio_ccw_ops.c +1 -0
 			   private->sch->schid.sch_no);
 
 	mdev_set_class_id(mdev, MDEV_ID_VFIO);
+	mdev_set_dev_ops(mdev, &vfio_mdev_ops);
 	return 0;
 }
 
@@ -575,11 +579,7 @@ static ssize_t vfio_ccw_mdev_ioctl(struct

mdev_device *mdev,

drivers/s390/cio/vfio_ccw_ops.c static const struct mdev_parent_ops

Quoted 13 lines
drivers/s390/cio/vfio_ccw_ops.c +1 -5
 	}
 }
 
-static const struct mdev_parent_ops vfio_ccw_mdev_ops = {
-	.owner			= THIS_MODULE,
-	.supported_type_groups  = mdev_type_groups,
-	.create			= vfio_ccw_mdev_create,
-	.remove			= vfio_ccw_mdev_remove,
+static const struct vfio_mdev_device_ops vfio_mdev_ops = {
 	.open			= vfio_ccw_mdev_open,
 	.release		= vfio_ccw_mdev_release,
 	.read			= vfio_ccw_mdev_read,
@@ -587,6 +587,13 @@ static const struct mdev_parent_ops

vfio_ccw_mdev_ops = {

Review

Quoted 14 lines
drivers/s390/cio/vfio_ccw_ops.c +7 -0
 	.ioctl			= vfio_ccw_mdev_ioctl,
 };
 
+static const struct mdev_parent_ops vfio_ccw_mdev_ops = {
+	.owner			= THIS_MODULE,
+	.supported_type_groups  = mdev_type_groups,
+	.create			= vfio_ccw_mdev_create,
+	.remove			= vfio_ccw_mdev_remove,
+};
+
 int vfio_ccw_mdev_reg(struct subchannel *sch)
 {
 	return mdev_register_device(&sch->dev, &vfio_ccw_mdev_ops);
(quoted diff) +0 -0

b/drivers/s390/crypto/vfio_ap_ops.c

drivers/s390/crypto/vfio_ap_ops.c static void vfio_ap_matrix_init(struct

Quoted 12 lines

index 2cfd96112aa0..3a89933f0d3e 100644

drivers/s390/crypto/vfio_ap_ops.c +1 -0
@@ -16,6 +16,7 @@
 #include <linux/bitops.h>
 #include <linux/kvm_host.h>
 #include <linux/module.h>
+#include <linux/vfio_mdev.h>
 #include <asm/kvm.h>
 #include <asm/zcrypt.h>
 
@@ -321,6 +322,8 @@ static void vfio_ap_matrix_init(struct

ap_config_info *info,

drivers/s390/crypto/vfio_ap_ops.c static void vfio_ap_matrix_init(struct ap_config_info *info,

Quoted 6 lines
drivers/s390/crypto/vfio_ap_ops.c +2 -0
 	matrix->adm_max = info->apxa ? info->Nd : 15;
 }
 
+static const struct vfio_mdev_device_ops vfio_mdev_ops;
+
 static int vfio_ap_mdev_create(struct kobject *kobj, struct mdev_device

*mdev)

drivers/s390/crypto/vfio_ap_ops.c static int vfio_ap_mdev_create(struct kobject

Quoted 3 lines
drivers/s390/crypto/vfio_ap_ops.c +0 -0
 {
 	struct ap_matrix_mdev *matrix_mdev;
@@ -344,6 +347,7 @@ static int vfio_ap_mdev_create(struct kobject

*kobj, struct mdev_device *mdev)

drivers/s390/crypto/vfio_ap_ops.c static ssize_t vfio_ap_mdev_ioctl(struct

Quoted 8 lines
drivers/s390/crypto/vfio_ap_ops.c +1 -0
 	mutex_unlock(&matrix_dev->lock);
 
 	mdev_set_class_id(mdev, MDEV_ID_VFIO);
+	mdev_set_dev_ops(mdev, &vfio_mdev_ops);
 	return 0;
 }
 
@@ -1281,15 +1285,18 @@ static ssize_t vfio_ap_mdev_ioctl(struct

mdev_device *mdev,

Review

Quoted 22 lines
drivers/s390/crypto/vfio_ap_ops.c +6 -3
 	return ret;
 }
 
+static const struct vfio_mdev_device_ops vfio_mdev_ops = {
+	.open			= vfio_ap_mdev_open,
+	.release		= vfio_ap_mdev_release,
+	.ioctl			= vfio_ap_mdev_ioctl,
+};
+
 static const struct mdev_parent_ops vfio_ap_matrix_ops = {
 	.owner			= THIS_MODULE,
 	.supported_type_groups	= vfio_ap_mdev_type_groups,
 	.mdev_attr_groups	= vfio_ap_mdev_attr_groups,
 	.create			= vfio_ap_mdev_create,
 	.remove			= vfio_ap_mdev_remove,
-	.open			= vfio_ap_mdev_open,
-	.release		= vfio_ap_mdev_release,
-	.ioctl			= vfio_ap_mdev_ioctl,
 };
 
 int vfio_ap_mdev_register(void)
(quoted diff) +0 -0

b/drivers/vfio/mdev/mdev_core.c

drivers/vfio/mdev/mdev_core.c void mdev_set_class_id(struct mdev_device *mdev,

Quoted 4 lines

index 8764cf4a276d..6f35f2ced2c9 100644

drivers/vfio/mdev/mdev_core.c +0 -0
@@ -51,6 +51,18 @@ void mdev_set_class_id(struct mdev_device *mdev,

u16 id)

Review

Quoted 19 lines
drivers/vfio/mdev/mdev_core.c +12 -0
 }
 EXPORT_SYMBOL(mdev_set_class_id);
 
+const void *mdev_get_dev_ops(struct mdev_device *mdev)
+{
+	return mdev->device_ops;
+}
+EXPORT_SYMBOL(mdev_get_dev_ops);
+
+void mdev_set_dev_ops(struct mdev_device *mdev, const void *ops)
+{
+	mdev->device_ops = ops;
+}
+EXPORT_SYMBOL(mdev_set_dev_ops);
+
 struct device *mdev_dev(struct mdev_device *mdev)
 {
 	return &mdev->dev;
(quoted diff) +0 -0

b/drivers/vfio/mdev/mdev_private.h

drivers/vfio/mdev/mdev_private.h struct mdev_device { · line 40

Quoted 11 lines

index c65f436c1869..b666805f0b1f 100644

drivers/vfio/mdev/mdev_private.h +1 -0
@@ -34,6 +34,7 @@ struct mdev_device {
 	struct device *iommu_device;
 	bool active;
 	u16 class_id;
+	const void *device_ops;
 };
 
 #define to_mdev_device(dev)	container_of(dev, struct

mdev_device, dev)

Review

Quoted 1 line
(quoted diff) +0 -0

b/drivers/vfio/mdev/vfio_mdev.c

Quoted 12 lines

index 891cf83a2d9a..95efa054442f 100644

drivers/vfio/mdev/vfio_mdev.c +1 -0
@@ -14,6 +14,7 @@
 #include <linux/slab.h>
 #include <linux/vfio.h>
 #include <linux/mdev.h>
+#include <linux/vfio_mdev.h>
 
 #include "mdev_private.h"
 

drivers/vfio/mdev/vfio_mdev.c line 28

Quoted 5 lines
drivers/vfio/mdev/vfio_mdev.c +1 -1
@@ -24,16 +25,16 @@
 static int vfio_mdev_open(void *device_data)
 {
 	struct mdev_device *mdev = device_data;
-	struct mdev_parent *parent = mdev->parent;
+	const struct vfio_mdev_device_ops *ops =

mdev_get_dev_ops(mdev);

Quoted 15 lines
drivers/vfio/mdev/vfio_mdev.c +2 -2
 	int ret;
 
-	if (unlikely(!parent->ops->open))
+	if (unlikely(!ops->open))
 		return -EINVAL;
 
 	if (!try_module_get(THIS_MODULE))
 		return -ENODEV;
 
-	ret = parent->ops->open(mdev);
+	ret = ops->open(mdev);
 	if (ret)
 		module_put(THIS_MODULE);
 

drivers/vfio/mdev/vfio_mdev.c static int vfio_mdev_open(void *device_data) · line 47

Quoted 5 lines
drivers/vfio/mdev/vfio_mdev.c +1 -1
@@ -43,10 +44,10 @@ static int vfio_mdev_open(void *device_data)
 static void vfio_mdev_release(void *device_data)
 {
 	struct mdev_device *mdev = device_data;
-	struct mdev_parent *parent = mdev->parent;
+	const struct vfio_mdev_device_ops *ops =

mdev_get_dev_ops(mdev);

drivers/vfio/mdev/vfio_mdev.c static long vfio_mdev_unlocked_ioctl(void

Quoted 9 lines
drivers/vfio/mdev/vfio_mdev.c +2 -2
-	if (likely(parent->ops->release))
-		parent->ops->release(mdev);
+	if (likely(ops->release))
+		ops->release(mdev);
 
 	module_put(THIS_MODULE);
 }
@@ -55,47 +56,47 @@ static long vfio_mdev_unlocked_ioctl(void

*device_data,

drivers/vfio/mdev/vfio_mdev.c static long vfio_mdev_unlocked_ioctl(void *device_data,

Quoted 5 lines
drivers/vfio/mdev/vfio_mdev.c +1 -1
 				     unsigned int cmd, unsigned long arg)
 {
 	struct mdev_device *mdev = device_data;
-	struct mdev_parent *parent = mdev->parent;
+	const struct vfio_mdev_device_ops *ops =

mdev_get_dev_ops(mdev);

Quoted 5 lines
drivers/vfio/mdev/vfio_mdev.c +1 -1
-	if (unlikely(!parent->ops->ioctl))
+	if (unlikely(!ops->ioctl))
 		return -EINVAL;
 

drivers/vfio/mdev/vfio_mdev.c static long vfio_mdev_unlocked_ioctl(void *device_data,

Quoted 10 lines
drivers/vfio/mdev/vfio_mdev.c +2 -2
-	return parent->ops->ioctl(mdev, cmd, arg);
+	return ops->ioctl(mdev, cmd, arg);
 }
 
 static ssize_t vfio_mdev_read(void *device_data, char __user *buf,
 			      size_t count, loff_t *ppos)
 {
 	struct mdev_device *mdev = device_data;
-	struct mdev_parent *parent = mdev->parent;
+	const struct vfio_mdev_device_ops *ops =

mdev_get_dev_ops(mdev);

Quoted 5 lines
drivers/vfio/mdev/vfio_mdev.c +1 -1
-	if (unlikely(!parent->ops->read))
+	if (unlikely(!ops->read))
 		return -EINVAL;
 

drivers/vfio/mdev/vfio_mdev.c static long vfio_mdev_unlocked_ioctl(void *device_data,

Quoted 10 lines
drivers/vfio/mdev/vfio_mdev.c +2 -2
-	return parent->ops->read(mdev, buf, count, ppos);
+	return ops->read(mdev, buf, count, ppos);
 }
 
 static ssize_t vfio_mdev_write(void *device_data, const char __user *buf,
 			       size_t count, loff_t *ppos)
 {
 	struct mdev_device *mdev = device_data;
-	struct mdev_parent *parent = mdev->parent;
+	const struct vfio_mdev_device_ops *ops =

mdev_get_dev_ops(mdev);

drivers/vfio/mdev/vfio_mdev.c static long vfio_mdev_unlocked_ioctl(void *device_data,

Quoted 10 lines
drivers/vfio/mdev/vfio_mdev.c +2 -2
-	if (unlikely(!parent->ops->write))
+	if (unlikely(!ops->write))
 		return -EINVAL;
 
-	return parent->ops->write(mdev, buf, count, ppos);
+	return ops->write(mdev, buf, count, ppos);
 }
 
 static int vfio_mdev_mmap(void *device_data, struct vm_area_struct

*vma)

drivers/vfio/mdev/vfio_mdev.c

Quoted 4 lines
drivers/vfio/mdev/vfio_mdev.c +1 -1
 {
 	struct mdev_device *mdev = device_data;
-	struct mdev_parent *parent = mdev->parent;
+	const struct vfio_mdev_device_ops *ops =

mdev_get_dev_ops(mdev);

include/linux/mdev.h struct device *mdev_get_iommu_device(struct

Quoted 27 lines
drivers/vfio/mdev/vfio_mdev.c +2 -2
-	if (unlikely(!parent->ops->mmap))
+	if (unlikely(!ops->mmap))
 		return -EINVAL;
 
-	return parent->ops->mmap(mdev, vma);
+	return ops->mmap(mdev, vma);
 }
 
 static const struct vfio_device_ops vfio_mdev_dev_ops = {
include/linux/mdev.h +5 -0
index 3974650c074f..3414307311f1 100644
@@ -10,6 +10,11 @@
 #ifndef MDEV_H
 #define MDEV_H
 
+#include <linux/types.h>
+#include <linux/device.h>
+#include <linux/mdev.h>
+#include <uapi/linux/uuid.h>
+
 struct mdev_device;
 
 /*
@@ -48,30 +53,8 @@ struct device *mdev_get_iommu_device(struct

device *dev);

include/linux/mdev.h struct device *mdev_get_iommu_device(struct · line 53

Quoted 1 line
include/linux/mdev.h +0 -0
  *			@mdev: mdev_device device structure which is

being

Quoted 19 lines
include/linux/mdev.h +0 -17
  *			       destroyed
  *			Returns integer: success (0) or error (< 0)
- * @open:		Open mediated device.
- *			@mdev: mediated device.
- *			Returns integer: success (0) or error (< 0)
- * @release:		release mediated device
- *			@mdev: mediated device.
- * @read:		Read emulation callback
- *			@mdev: mediated device structure
- *			@buf: read buffer
- *			@count: number of bytes to read
- *			@ppos: address.
- *			Retuns number on bytes read on success or error.
- * @write:		Write emulation callback
- *			@mdev: mediated device structure
- *			@buf: write buffer
- *			@count: number of bytes to be written
- *			@ppos: address.
- *			Retuns number on bytes written on success or error.

include/linux/mdev.h struct device *mdev_get_iommu_device(struct device *dev);

Quoted 10 lines
include/linux/mdev.h +2 -7
- * @ioctl:		IOCTL callback
- *			@mdev: mediated device structure
- *			@cmd: ioctl command
- *			@arg: arguments to ioctl
- * @mmap:		mmap callback
- *			@mdev: mediated device structure
- *			@vma: vma structure
+ * @device_ops:         Device specific emulation callback.
+ *
  * Parent device that support mediated device should be registered with

mdev

Quoted 5 lines
include/linux/mdev.h +0 -0
  * module with mdev_parent_ops structure.
  **/
@@ -83,15 +66,6 @@ struct mdev_parent_ops {
 
 	int     (*create)(struct kobject *kobj, struct mdev_device *mdev);

include/linux/mdev.h struct mdev_parent_ops { · line 94

Quoted 10 lines
include/linux/mdev.h +0 -9
 	int     (*remove)(struct mdev_device *mdev);
-	int     (*open)(struct mdev_device *mdev);
-	void    (*release)(struct mdev_device *mdev);
-	ssize_t (*read)(struct mdev_device *mdev, char __user *buf,
-			size_t count, loff_t *ppos);
-	ssize_t (*write)(struct mdev_device *mdev, const char __user *buf,
-			 size_t count, loff_t *ppos);
-	long	(*ioctl)(struct mdev_device *mdev, unsigned int cmd,
-			 unsigned long arg);
-	int	(*mmap)(struct mdev_device *mdev, struct vm_area_struct

*vma);

Quoted 57 lines
include/linux/mdev.h +2 -0
 };
 
 /* interface for exporting mdev supported type attributes */
@@ -133,6 +107,8 @@ struct mdev_driver {
 
 void *mdev_get_drvdata(struct mdev_device *mdev);
 void mdev_set_drvdata(struct mdev_device *mdev, void *data);
+const void *mdev_get_dev_ops(struct mdev_device *mdev);
+void mdev_set_dev_ops(struct mdev_device *mdev, const void *ops);
 void mdev_set_class_id(struct mdev_device *mdev, u16 id);
 const guid_t *mdev_uuid(struct mdev_device *mdev);
 
include/linux/vfio_mdev.h new file +39 -0
new file mode 100644
index 000000000000..3907c5371c2b
@@ -0,0 +1,52 @@
+/* SPDX-License-Identifier: GPL-2.0-only */
+/*
+ * VFIO Mediated device definition
+ */
+
+#ifndef VFIO_MDEV_H
+#define VFIO_MDEV_H
+
+#include <linux/mdev.h>
+
+/**
+ * struct vfio_mdev_device_ops - Structure to be registered for each
+ * mdev device to register the device to vfio-mdev module.
+ *
+ * @open:		Open mediated device.
+ *			@mdev: mediated device.
+ *			Returns integer: success (0) or error (< 0)
+ * @release:		release mediated device
+ *			@mdev: mediated device.
+ * @read:		Read emulation callback
+ *			@mdev: mediated device structure
+ *			@buf: read buffer
+ *			@count: number of bytes to read
+ *			@ppos: address.
+ *			Retuns number on bytes read on success or error.
+ * @write:		Write emulation callback
+ *			@mdev: mediated device structure
+ *			@buf: write buffer
+ *			@count: number of bytes to be written
+ *			@ppos: address.
+ *			Retuns number on bytes written on success or error.
+ * @ioctl:		IOCTL callback
+ *			@mdev: mediated device structure
+ *			@cmd: ioctl command
+ *			@arg: arguments to ioctl
+ * @mmap:		mmap callback
+ *			@mdev: mediated device structure
+ *			@vma: vma structure
+ */

include/linux/vfio_mdev.h line 49

Quoted 10 lines
include/linux/vfio_mdev.h new file +10 -0
+struct vfio_mdev_device_ops {
+	int     (*open)(struct mdev_device *mdev);
+	void    (*release)(struct mdev_device *mdev);
+	ssize_t (*read)(struct mdev_device *mdev, char __user *buf,
+			size_t count, loff_t *ppos);
+	ssize_t (*write)(struct mdev_device *mdev, const char __user *buf,
+			 size_t count, loff_t *ppos);
+	long	(*ioctl)(struct mdev_device *mdev, unsigned int cmd,
+			 unsigned long arg);
+	int	(*mmap)(struct mdev_device *mdev, struct vm_area_struct

*vma);

samples/vfio-mdev/mbochs.c static int mbochs_reset(struct mdev_device

Quoted 16 lines
include/linux/vfio_mdev.h new file +3 -0
+};
+
+#endif
samples/vfio-mdev/mbochs.c +1 -0
index 8a8583c892b2..b4bf29c6136c 100644
@@ -30,6 +30,7 @@
 #include <linux/iommu.h>
 #include <linux/sysfs.h>
 #include <linux/mdev.h>
+#include <linux/vfio_mdev.h>
 #include <linux/pci.h>
 #include <linux/dma-buf.h>
 #include <linux/highmem.h>
@@ -516,6 +517,8 @@ static int mbochs_reset(struct mdev_device

*mdev)

samples/vfio-mdev/mbochs.c static int mbochs_create(struct kobject *kobj,

Quoted 9 lines
samples/vfio-mdev/mbochs.c +2 -0
 	return 0;
 }
 
+static const struct vfio_mdev_device_ops vfio_mdev_ops;
+
 static int mbochs_create(struct kobject *kobj, struct mdev_device *mdev)
 {
 	const struct mbochs_type *type = mbochs_find_type(kobj);
@@ -562,6 +565,7 @@ static int mbochs_create(struct kobject *kobj,

struct mdev_device *mdev)

samples/vfio-mdev/mbochs.c static struct attribute_group

Quoted 8 lines
samples/vfio-mdev/mbochs.c +1 -0
 	mbochs_used_mbytes += type->mbytes;
 	mdev_set_class_id(mdev, MDEV_ID_VFIO);
+	mdev_set_dev_ops(mdev, &vfio_mdev_ops);
 	return 0;
 
 err_mem:
@@ -1419,12 +1423,7 @@ static struct attribute_group

*mdev_type_groups[] = {

samples/vfio-mdev/mbochs.c static const struct mdev_parent_ops

Quoted 14 lines
samples/vfio-mdev/mbochs.c +1 -6
 	NULL,
 };
 
-static const struct mdev_parent_ops mdev_fops = {
-	.owner			= THIS_MODULE,
-	.mdev_attr_groups	= mdev_dev_groups,
-	.supported_type_groups	= mdev_type_groups,
-	.create			= mbochs_create,
-	.remove			= mbochs_remove,
+static const struct vfio_mdev_device_ops vfio_mdev_ops = {
 	.open			= mbochs_open,
 	.release		= mbochs_close,
 	.read			= mbochs_read,
@@ -1433,6 +1432,14 @@ static const struct mdev_parent_ops

mdev_fops = {

samples/vfio-mdev/mdpy.c static int mdpy_create(struct kobject *kobj, struct

Quoted 36 lines
samples/vfio-mdev/mbochs.c +8 -0
 	.mmap			= mbochs_mmap,
 };
 
+static const struct mdev_parent_ops mdev_fops = {
+	.owner			= THIS_MODULE,
+	.mdev_attr_groups	= mdev_dev_groups,
+	.supported_type_groups	= mdev_type_groups,
+	.create			= mbochs_create,
+	.remove			= mbochs_remove,
+};
+
 static const struct file_operations vd_fops = {
 	.owner		= THIS_MODULE,
 };
samples/vfio-mdev/mdpy.c +3 -0
index 88d7e76f3836..80c2df531326 100644
@@ -26,6 +26,7 @@
 #include <linux/iommu.h>
 #include <linux/sysfs.h>
 #include <linux/mdev.h>
+#include <linux/vfio_mdev.h>
 #include <linux/pci.h>
 #include <drm/drm_fourcc.h>
 #include "mdpy-defs.h"
@@ -226,6 +227,8 @@ static int mdpy_reset(struct mdev_device *mdev)
 	return 0;
 }
 
+static const struct vfio_mdev_device_ops vfio_mdev_ops;
+
 static int mdpy_create(struct kobject *kobj, struct mdev_device *mdev)
 {
 	const struct mdpy_type *type = mdpy_find_type(kobj);
@@ -270,6 +273,7 @@ static int mdpy_create(struct kobject *kobj, struct

mdev_device *mdev)

samples/vfio-mdev/mdpy.c static struct attribute_group

Quoted 8 lines
samples/vfio-mdev/mdpy.c +1 -0
 	mdpy_count++;
 	mdev_set_class_id(mdev, MDEV_ID_VFIO);
+	mdev_set_dev_ops(mdev, &vfio_mdev_ops);
 	return 0;
 }
 
@@ -726,12 +730,7 @@ static struct attribute_group

*mdev_type_groups[] = {

samples/vfio-mdev/mdpy.c static const struct mdev_parent_ops mdev_fops

Quoted 14 lines
samples/vfio-mdev/mdpy.c +1 -6
 	NULL,
 };
 
-static const struct mdev_parent_ops mdev_fops = {
-	.owner			= THIS_MODULE,
-	.mdev_attr_groups	= mdev_dev_groups,
-	.supported_type_groups	= mdev_type_groups,
-	.create			= mdpy_create,
-	.remove			= mdpy_remove,
+static const struct vfio_mdev_device_ops vfio_mdev_ops = {
 	.open			= mdpy_open,
 	.release		= mdpy_close,
 	.read			= mdpy_read,
@@ -740,6 +739,14 @@ static const struct mdev_parent_ops mdev_fops

= {

samples/vfio-mdev/mtty.c static ssize_t mdev_access(struct mdev_device

Quoted 27 lines
samples/vfio-mdev/mdpy.c +8 -0
 	.mmap			= mdpy_mmap,
 };
 
+static const struct mdev_parent_ops mdev_fops = {
+	.owner			= THIS_MODULE,
+	.mdev_attr_groups	= mdev_dev_groups,
+	.supported_type_groups	= mdev_type_groups,
+	.create			= mdpy_create,
+	.remove			= mdpy_remove,
+};
+
 static const struct file_operations vd_fops = {
 	.owner		= THIS_MODULE,
 };
samples/vfio-mdev/mtty.c +1 -0
index 4e0735143b69..2db860ccb02c 100644
@@ -27,6 +27,7 @@
 #include <linux/ctype.h>
 #include <linux/file.h>
 #include <linux/mdev.h>
+#include <linux/vfio_mdev.h>
 #include <linux/pci.h>
 #include <linux/serial.h>
 #include <uapi/linux/serial_reg.h>
@@ -723,6 +724,8 @@ static ssize_t mdev_access(struct mdev_device

*mdev, u8 *buf, size_t count,

samples/vfio-mdev/mtty.c static int mtty_create(struct kobject *kobj, struct

Quoted 9 lines
samples/vfio-mdev/mtty.c +2 -0
 	return ret;
 }
 
+static const struct vfio_mdev_device_ops vfio_dev_ops;
+
 static int mtty_create(struct kobject *kobj, struct mdev_device *mdev)
 {
 	struct mdev_state *mdev_state;
@@ -771,6 +774,7 @@ static int mtty_create(struct kobject *kobj, struct

mdev_device *mdev)

samples/vfio-mdev/mtty.c static struct attribute_group

Quoted 8 lines
samples/vfio-mdev/mtty.c +1 -0
 	mutex_unlock(&mdev_list_lock);
 
 	mdev_set_class_id(mdev, MDEV_ID_VFIO);
+	mdev_set_dev_ops(mdev, &vfio_dev_ops);
 	return 0;
 }
 
@@ -1411,6 +1415,14 @@ static struct attribute_group

*mdev_type_groups[] = {

samples/vfio-mdev/mtty.c static const struct mdev_parent_ops

Quoted 15 lines
samples/vfio-mdev/mtty.c +8 -0
 	NULL,
 };
 
+static const struct vfio_mdev_device_ops vfio_dev_ops = {
+	.open                   = mtty_open,
+	.release                = mtty_close,
+	.read                   = mtty_read,
+	.write                  = mtty_write,
+	.ioctl		        = mtty_ioctl,
+};
+
 static const struct mdev_parent_ops mdev_fops = {
 	.owner                  = THIS_MODULE,
 	.dev_attr_groups        = mtty_dev_groups,
@@ -1418,11 +1430,6 @@ static const struct mdev_parent_ops

mdev_fops = {

samples/vfio-mdev/mtty.c static const struct mdev_parent_ops mdev_fops = { · line 1435

Quoted 11 lines
samples/vfio-mdev/mtty.c +0 -5
 	.supported_type_groups  = mdev_type_groups,
 	.create                 = mtty_create,
 	.remove			= mtty_remove,
-	.open                   = mtty_open,
-	.release                = mtty_close,
-	.read                   = mtty_read,
-	.write                  = mtty_write,
-	.ioctl		        = mtty_ioctl,
 };
 
 static void mtty_device_release(struct device *dev)

_______________________________________________
intel-gvt-dev mailing list
[email protected]
https://lists.freedesktop.org/mailman/listinfo/intel-gvt-dev

author
#29 lore raw

Re: [PATCH V2 5/8] mdev: introduce device specific ops

In reply to Alex Williamson
Quoted 36 lines · Alex Williamson
Quoted 33 lines · On Tue, 24 Sep 2019 21:53:29 +0800 Jason Wang <[email protected]> wrote:

Currently, except for the create and remove, the rest of mdev_parent_ops is designed for vfio-mdev driver only and may not help for kernel mdev driver. With the help of class id, this patch introduces device specific callbacks inside mdev_device structure. This allows different set of callback to be used by vfio-mdev and virtio-mdev.

---
 .../driver-api/vfio-mediated-device.rst       |  4 +-
 MAINTAINERS                                   |  1 +
 drivers/gpu/drm/i915/gvt/kvmgt.c              | 17 +++---
 drivers/s390/cio/vfio_ccw_ops.c               | 17 ++++--
 drivers/s390/crypto/vfio_ap_ops.c             | 13 +++--
 drivers/vfio/mdev/mdev_core.c                 | 12 +++++
 drivers/vfio/mdev/mdev_private.h              |  1 +
 drivers/vfio/mdev/vfio_mdev.c                 | 37 ++++++-------
 include/linux/mdev.h                          | 42 ++++-----------
 include/linux/vfio_mdev.h                     | 52 +++++++++++++++++++
 samples/vfio-mdev/mbochs.c                    | 19 ++++---
 samples/vfio-mdev/mdpy.c                      | 19 ++++---
 samples/vfio-mdev/mtty.c                      | 17 ++++--
 13 files changed, 168 insertions(+), 83 deletions(-)
 create mode 100644 include/linux/vfio_mdev.h
Documentation/driver-api/vfio-mediated-device.rst +0 -0
index a5bdc60d62a1..d50425b368bb 100644
@@ -152,7 +152,9 @@ callbacks per mdev parent device, per mdev type, or any other categorization.
 Vendor drivers are expected to be fully asynchronous in this respect or
 provide their own internal resource protection.)
 
Quoted 8 lines

Documentation/driver-api/vfio-mediated-device.rst callbacks per mdev parent device, per mdev type, or any other categorization. · line 157

Quoted 4 lines
Documentation/driver-api/vfio-mediated-device.rst +3 -1
-The callbacks in the mdev_parent_ops structure are as follows:
+The device specific callbacks are referred through device_ops pointer
+in mdev_parent_ops. For vfio-mdev device, its callbacks in device_ops
+are as follows:

This is not accurate. device_ops is now on the mdev_device and is an mdev bus driver specific structure of callbacks that must be registered for each mdev device by the parent driver during the create callback. There's a one to one mapping of class_id to mdev_device_ops callbacks.

Yes.

Quoted 5 lines

That also suggests to me that we could be more clever in registering both of these with mdev-core. Can we embed the class_id in the ops structure in a common way so that the core can extract it and the bus drivers can access their specific callbacks?

That seems much cleaner, let me try to do that in V3.

Quoted 51 lines
Quoted 37 lines
Documentation/driver-api/vfio-mediated-device.rst +0 -0
 * open: open callback of mediated device
 * close: close callback of mediated device
MAINTAINERS +1 -0
index b2326dece28e..89832b316500 100644
@@ -17075,6 +17075,7 @@ S:	Maintained
 F:	Documentation/driver-api/vfio-mediated-device.rst
 F:	drivers/vfio/mdev/
 F:	include/linux/mdev.h
+F:	include/linux/vfio_mdev.h
 F:	samples/vfio-mdev/
 
 VFIO PLATFORM DRIVER
drivers/gpu/drm/i915/gvt/kvmgt.c +3 -0
index f793252a3d2a..b274f5ee481f 100644
@@ -42,6 +42,7 @@
 #include <linux/kvm_host.h>
 #include <linux/vfio.h>
 #include <linux/mdev.h>
+#include <linux/vfio_mdev.h>
 #include <linux/debugfs.h>
 
 #include <linux/nospec.h>
@@ -643,6 +644,8 @@ static void kvmgt_put_vfio_device(void *vgpu)
 	vfio_device_put(((struct intel_vgpu *)vgpu)->vdev.vfio_device);
 }
 
+static struct vfio_mdev_device_ops intel_vfio_vgpu_dev_ops;
+
 static int intel_vgpu_create(struct kobject *kobj, struct mdev_device *mdev)
 {
 	struct intel_vgpu *vgpu = NULL;

drivers/gpu/drm/i915/gvt/kvmgt.c static int intel_vgpu_create(struct kobject *kobj, struct mdev_device *mdev) · line 685

Quoted 4 lines
drivers/gpu/drm/i915/gvt/kvmgt.c +1 -0
@@ -679,6 +682,7 @@ static int intel_vgpu_create(struct kobject *kobj, struct mdev_device *mdev)
 	ret = 0;
 
 	mdev_set_class_id(mdev, MDEV_ID_VFIO);
+	mdev_set_dev_ops(mdev, &intel_vfio_vgpu_dev_ops);

This seems rather unrefined. We're registering interdependent data in separate calls. All drivers need to make both of these calls. I'm not sure if this is a good idea, but what if we had:

static const struct vfio_mdev_device_ops intel_vfio_vgpu_dev_ops = {
	.id			= MDEV_ID_VFIO,
drivers/gpu/drm/i915/gvt/kvmgt.c +0 -0
 	.open			= intel_vgpu_open,
 	.release		= intel_vgpu_release,
...
Quoted 2 lines

And the set function passed &intel_vfio_vgpu_dev_ops.id and the mdev bus drivers used container_of to get to their callbacks?

Yes, with the check of mdev_device_create() if nothing is set by the device.

Quoted 23 lines
Quoted 23 lines
drivers/gpu/drm/i915/gvt/kvmgt.c +1 -6
 out:
 	return ret;
 }
@@ -1601,20 +1605,21 @@ static const struct attribute_group *intel_vgpu_groups[] = {
 	NULL,
 };
 
-static struct mdev_parent_ops intel_vgpu_ops = {
-	.mdev_attr_groups       = intel_vgpu_groups,
-	.create			= intel_vgpu_create,
-	.remove			= intel_vgpu_remove,
-
+static struct vfio_mdev_device_ops intel_vfio_vgpu_dev_ops = {
 	.open			= intel_vgpu_open,
 	.release		= intel_vgpu_release,
-
 	.read			= intel_vgpu_read,
 	.write			= intel_vgpu_write,
 	.mmap			= intel_vgpu_mmap,
 	.ioctl			= intel_vgpu_ioctl,
 };
 
Quoted 4 lines

drivers/gpu/drm/i915/gvt/kvmgt.c static const struct attribute_group *intel_vgpu_groups[] = { · line 1617

Quoted 1 line
drivers/gpu/drm/i915/gvt/kvmgt.c +1 -0
+static struct mdev_parent_ops intel_vgpu_ops = {

These could maybe be made const at the same time. Thanks,

Alex

Ok, let me fix.

Thanks

author
#30 lore raw

Re: [PATCH V2 5/8] mdev: introduce device specific ops

In reply to Alex Williamson
Quoted 36 lines · Alex Williamson
Quoted 33 lines · On Tue, 24 Sep 2019 21:53:29 +0800 Jason Wang <[email protected]> wrote:

Currently, except for the create and remove, the rest of mdev_parent_ops is designed for vfio-mdev driver only and may not help for kernel mdev driver. With the help of class id, this patch introduces device specific callbacks inside mdev_device structure. This allows different set of callback to be used by vfio-mdev and virtio-mdev.

---
 .../driver-api/vfio-mediated-device.rst       |  4 +-
 MAINTAINERS                                   |  1 +
 drivers/gpu/drm/i915/gvt/kvmgt.c              | 17 +++---
 drivers/s390/cio/vfio_ccw_ops.c               | 17 ++++--
 drivers/s390/crypto/vfio_ap_ops.c             | 13 +++--
 drivers/vfio/mdev/mdev_core.c                 | 12 +++++
 drivers/vfio/mdev/mdev_private.h              |  1 +
 drivers/vfio/mdev/vfio_mdev.c                 | 37 ++++++-------
 include/linux/mdev.h                          | 42 ++++-----------
 include/linux/vfio_mdev.h                     | 52 +++++++++++++++++++
 samples/vfio-mdev/mbochs.c                    | 19 ++++---
 samples/vfio-mdev/mdpy.c                      | 19 ++++---
 samples/vfio-mdev/mtty.c                      | 17 ++++--
 13 files changed, 168 insertions(+), 83 deletions(-)
 create mode 100644 include/linux/vfio_mdev.h
Documentation/driver-api/vfio-mediated-device.rst +0 -0
index a5bdc60d62a1..d50425b368bb 100644
@@ -152,7 +152,9 @@ callbacks per mdev parent device, per mdev type, or any other categorization.
 Vendor drivers are expected to be fully asynchronous in this respect or
 provide their own internal resource protection.)
 
Quoted 8 lines

Documentation/driver-api/vfio-mediated-device.rst callbacks per mdev parent device, per mdev type, or any other categorization. · line 157

Quoted 4 lines
Documentation/driver-api/vfio-mediated-device.rst +3 -1
-The callbacks in the mdev_parent_ops structure are as follows:
+The device specific callbacks are referred through device_ops pointer
+in mdev_parent_ops. For vfio-mdev device, its callbacks in device_ops
+are as follows:

This is not accurate. device_ops is now on the mdev_device and is an mdev bus driver specific structure of callbacks that must be registered for each mdev device by the parent driver during the create callback. There's a one to one mapping of class_id to mdev_device_ops callbacks.

Yes.

Quoted 5 lines

That also suggests to me that we could be more clever in registering both of these with mdev-core. Can we embed the class_id in the ops structure in a common way so that the core can extract it and the bus drivers can access their specific callbacks?

That seems much cleaner, let me try to do that in V3.

Quoted 51 lines
Quoted 37 lines
Documentation/driver-api/vfio-mediated-device.rst +0 -0
 * open: open callback of mediated device
 * close: close callback of mediated device
MAINTAINERS +1 -0
index b2326dece28e..89832b316500 100644
@@ -17075,6 +17075,7 @@ S:	Maintained
 F:	Documentation/driver-api/vfio-mediated-device.rst
 F:	drivers/vfio/mdev/
 F:	include/linux/mdev.h
+F:	include/linux/vfio_mdev.h
 F:	samples/vfio-mdev/
 
 VFIO PLATFORM DRIVER
drivers/gpu/drm/i915/gvt/kvmgt.c +3 -0
index f793252a3d2a..b274f5ee481f 100644
@@ -42,6 +42,7 @@
 #include <linux/kvm_host.h>
 #include <linux/vfio.h>
 #include <linux/mdev.h>
+#include <linux/vfio_mdev.h>
 #include <linux/debugfs.h>
 
 #include <linux/nospec.h>
@@ -643,6 +644,8 @@ static void kvmgt_put_vfio_device(void *vgpu)
 	vfio_device_put(((struct intel_vgpu *)vgpu)->vdev.vfio_device);
 }
 
+static struct vfio_mdev_device_ops intel_vfio_vgpu_dev_ops;
+
 static int intel_vgpu_create(struct kobject *kobj, struct mdev_device *mdev)
 {
 	struct intel_vgpu *vgpu = NULL;

drivers/gpu/drm/i915/gvt/kvmgt.c static int intel_vgpu_create(struct kobject *kobj, struct mdev_device *mdev) · line 685

Quoted 4 lines
drivers/gpu/drm/i915/gvt/kvmgt.c +1 -0
@@ -679,6 +682,7 @@ static int intel_vgpu_create(struct kobject *kobj, struct mdev_device *mdev)
 	ret = 0;
 
 	mdev_set_class_id(mdev, MDEV_ID_VFIO);
+	mdev_set_dev_ops(mdev, &intel_vfio_vgpu_dev_ops);

This seems rather unrefined. We're registering interdependent data in separate calls. All drivers need to make both of these calls. I'm not sure if this is a good idea, but what if we had:

static const struct vfio_mdev_device_ops intel_vfio_vgpu_dev_ops = {
	.id			= MDEV_ID_VFIO,
drivers/gpu/drm/i915/gvt/kvmgt.c +0 -0
 	.open			= intel_vgpu_open,
 	.release		= intel_vgpu_release,
...
Quoted 2 lines

And the set function passed &intel_vfio_vgpu_dev_ops.id and the mdev bus drivers used container_of to get to their callbacks?

Yes, with the check of mdev_device_create() if nothing is set by the device.

Quoted 23 lines
Quoted 23 lines
drivers/gpu/drm/i915/gvt/kvmgt.c +1 -6
 out:
 	return ret;
 }
@@ -1601,20 +1605,21 @@ static const struct attribute_group *intel_vgpu_groups[] = {
 	NULL,
 };
 
-static struct mdev_parent_ops intel_vgpu_ops = {
-	.mdev_attr_groups       = intel_vgpu_groups,
-	.create			= intel_vgpu_create,
-	.remove			= intel_vgpu_remove,
-
+static struct vfio_mdev_device_ops intel_vfio_vgpu_dev_ops = {
 	.open			= intel_vgpu_open,
 	.release		= intel_vgpu_release,
-
 	.read			= intel_vgpu_read,
 	.write			= intel_vgpu_write,
 	.mmap			= intel_vgpu_mmap,
 	.ioctl			= intel_vgpu_ioctl,
 };
 
Quoted 4 lines

drivers/gpu/drm/i915/gvt/kvmgt.c static const struct attribute_group *intel_vgpu_groups[] = { · line 1617

Quoted 1 line
drivers/gpu/drm/i915/gvt/kvmgt.c +1 -0
+static struct mdev_parent_ops intel_vgpu_ops = {

These could maybe be made const at the same time. Thanks,

Alex

Ok, let me fix.

Thanks

_______________________________________________
Virtualization mailing list
[email protected]
https://lists.linuxfoundation.org/mailman/listinfo/virtualization

#31 lore raw

Re: [PATCH V2 5/8] mdev: introduce device specific ops

In reply to Jason Wang
Quoted 30 lines · On Tue, 24 Sep 2019 21:53:29 +0800 Jason Wang <[email protected]> wrote:

Currently, except for the create and remove, the rest of mdev_parent_ops is designed for vfio-mdev driver only and may not help for kernel mdev driver. With the help of class id, this patch introduces device specific callbacks inside mdev_device structure. This allows different set of callback to be used by vfio-mdev and virtio-mdev.

---
 .../driver-api/vfio-mediated-device.rst       |  4 +-
 MAINTAINERS                                   |  1 +
 drivers/gpu/drm/i915/gvt/kvmgt.c              | 17 +++---
 drivers/s390/cio/vfio_ccw_ops.c               | 17 ++++--
 drivers/s390/crypto/vfio_ap_ops.c             | 13 +++--
 drivers/vfio/mdev/mdev_core.c                 | 12 +++++
 drivers/vfio/mdev/mdev_private.h              |  1 +
 drivers/vfio/mdev/vfio_mdev.c                 | 37 ++++++-------
 include/linux/mdev.h                          | 42 ++++-----------
 include/linux/vfio_mdev.h                     | 52 +++++++++++++++++++
 samples/vfio-mdev/mbochs.c                    | 19 ++++---
 samples/vfio-mdev/mdpy.c                      | 19 ++++---
 samples/vfio-mdev/mtty.c                      | 17 ++++--
 13 files changed, 168 insertions(+), 83 deletions(-)
 create mode 100644 include/linux/vfio_mdev.h
Documentation/driver-api/vfio-mediated-device.rst +0 -0
index a5bdc60d62a1..d50425b368bb 100644

Documentation/driver-api/vfio-mediated-device.rst callbacks per mdev parent device, per mdev type, or any other categorization. · line 157

Quoted 7 lines
Documentation/driver-api/vfio-mediated-device.rst +3 -1
@@ -152,7 +152,9 @@ callbacks per mdev parent device, per mdev type, or any other categorization.
 Vendor drivers are expected to be fully asynchronous in this respect or
 provide their own internal resource protection.)
 
-The callbacks in the mdev_parent_ops structure are as follows:
+The device specific callbacks are referred through device_ops pointer
+in mdev_parent_ops. For vfio-mdev device, its callbacks in device_ops
+are as follows:

This is not accurate. device_ops is now on the mdev_device and is an mdev bus driver specific structure of callbacks that must be registered for each mdev device by the parent driver during the create callback. There's a one to one mapping of class_id to mdev_device_ops callbacks.

That also suggests to me that we could be more clever in registering both of these with mdev-core. Can we embed the class_id in the ops structure in a common way so that the core can extract it and the bus drivers can access their specific callbacks?

Quoted 36 lines
Documentation/driver-api/vfio-mediated-device.rst +0 -0
 * open: open callback of mediated device
 * close: close callback of mediated device
MAINTAINERS +1 -0
index b2326dece28e..89832b316500 100644
@@ -17075,6 +17075,7 @@ S:	Maintained
 F:	Documentation/driver-api/vfio-mediated-device.rst
 F:	drivers/vfio/mdev/
 F:	include/linux/mdev.h
+F:	include/linux/vfio_mdev.h
 F:	samples/vfio-mdev/
 
 VFIO PLATFORM DRIVER
drivers/gpu/drm/i915/gvt/kvmgt.c +3 -0
index f793252a3d2a..b274f5ee481f 100644
@@ -42,6 +42,7 @@
 #include <linux/kvm_host.h>
 #include <linux/vfio.h>
 #include <linux/mdev.h>
+#include <linux/vfio_mdev.h>
 #include <linux/debugfs.h>
 
 #include <linux/nospec.h>
@@ -643,6 +644,8 @@ static void kvmgt_put_vfio_device(void *vgpu)
 	vfio_device_put(((struct intel_vgpu *)vgpu)->vdev.vfio_device);
 }
 
+static struct vfio_mdev_device_ops intel_vfio_vgpu_dev_ops;
+
 static int intel_vgpu_create(struct kobject *kobj, struct mdev_device *mdev)
 {
 	struct intel_vgpu *vgpu = NULL;

drivers/gpu/drm/i915/gvt/kvmgt.c static int intel_vgpu_create(struct kobject *kobj, struct mdev_device *mdev) · line 685

Quoted 4 lines
drivers/gpu/drm/i915/gvt/kvmgt.c +1 -0
@@ -679,6 +682,7 @@ static int intel_vgpu_create(struct kobject *kobj, struct mdev_device *mdev)
 	ret = 0;
 
 	mdev_set_class_id(mdev, MDEV_ID_VFIO);
+	mdev_set_dev_ops(mdev, &intel_vfio_vgpu_dev_ops);

This seems rather unrefined. We're registering interdependent data in separate calls. All drivers need to make both of these calls. I'm not sure if this is a good idea, but what if we had:

static const struct vfio_mdev_device_ops intel_vfio_vgpu_dev_ops = {
	.id			= MDEV_ID_VFIO,
 	.open			= intel_vgpu_open,
 	.release		= intel_vgpu_release,
        ...

And the set function passed &intel_vfio_vgpu_dev_ops.id and the mdev bus drivers used container_of to get to their callbacks?

Quoted 13 lines
drivers/gpu/drm/i915/gvt/kvmgt.c +1 -5
 out:
 	return ret;
 }
@@ -1601,20 +1605,21 @@ static const struct attribute_group *intel_vgpu_groups[] = {
 	NULL,
 };
 
-static struct mdev_parent_ops intel_vgpu_ops = {
-	.mdev_attr_groups       = intel_vgpu_groups,
-	.create			= intel_vgpu_create,
-	.remove			= intel_vgpu_remove,
-
+static struct vfio_mdev_device_ops intel_vfio_vgpu_dev_ops = {

drivers/gpu/drm/i915/gvt/kvmgt.c static const struct attribute_group *intel_vgpu_groups[] = { · line 1617

Quoted 10 lines
drivers/gpu/drm/i915/gvt/kvmgt.c +1 -1
 	.open			= intel_vgpu_open,
 	.release		= intel_vgpu_release,
-
 	.read			= intel_vgpu_read,
 	.write			= intel_vgpu_write,
 	.mmap			= intel_vgpu_mmap,
 	.ioctl			= intel_vgpu_ioctl,
 };
 
+static struct mdev_parent_ops intel_vgpu_ops = {

These could maybe be made const at the same time. Thanks,

Alex

Quoted 560 lines
drivers/gpu/drm/i915/gvt/kvmgt.c +5 -0
+	.mdev_attr_groups       = intel_vgpu_groups,
+	.create			= intel_vgpu_create,
+	.remove			= intel_vgpu_remove,
+};
+
 static int kvmgt_host_init(struct device *dev, void *gvt, const void *ops)
 {
 	struct attribute **kvm_type_attrs;
drivers/s390/cio/vfio_ccw_ops.c +12 -5
index d258ef1fedb9..329d53c1f46b 100644
@@ -12,6 +12,7 @@
 
 #include <linux/vfio.h>
 #include <linux/mdev.h>
+#include <linux/vfio_mdev.h>
 #include <linux/nospec.h>
 #include <linux/slab.h>
 
@@ -110,6 +111,8 @@ static struct attribute_group *mdev_type_groups[] = {
 	NULL,
 };
 
+static const struct vfio_mdev_device_ops vfio_mdev_ops;
+
 static int vfio_ccw_mdev_create(struct kobject *kobj, struct mdev_device *mdev)
 {
 	struct vfio_ccw_private *private =
@@ -130,6 +133,7 @@ static int vfio_ccw_mdev_create(struct kobject *kobj, struct mdev_device *mdev)
 			   private->sch->schid.sch_no);
 
 	mdev_set_class_id(mdev, MDEV_ID_VFIO);
+	mdev_set_dev_ops(mdev, &vfio_mdev_ops);
 	return 0;
 }
 
@@ -575,11 +579,7 @@ static ssize_t vfio_ccw_mdev_ioctl(struct mdev_device *mdev,
 	}
 }
 
-static const struct mdev_parent_ops vfio_ccw_mdev_ops = {
-	.owner			= THIS_MODULE,
-	.supported_type_groups  = mdev_type_groups,
-	.create			= vfio_ccw_mdev_create,
-	.remove			= vfio_ccw_mdev_remove,
+static const struct vfio_mdev_device_ops vfio_mdev_ops = {
 	.open			= vfio_ccw_mdev_open,
 	.release		= vfio_ccw_mdev_release,
 	.read			= vfio_ccw_mdev_read,
@@ -587,6 +587,13 @@ static const struct mdev_parent_ops vfio_ccw_mdev_ops = {
 	.ioctl			= vfio_ccw_mdev_ioctl,
 };
 
+static const struct mdev_parent_ops vfio_ccw_mdev_ops = {
+	.owner			= THIS_MODULE,
+	.supported_type_groups  = mdev_type_groups,
+	.create			= vfio_ccw_mdev_create,
+	.remove			= vfio_ccw_mdev_remove,
+};
+
 int vfio_ccw_mdev_reg(struct subchannel *sch)
 {
 	return mdev_register_device(&sch->dev, &vfio_ccw_mdev_ops);
drivers/s390/crypto/vfio_ap_ops.c +10 -3
index 2cfd96112aa0..3a89933f0d3e 100644
@@ -16,6 +16,7 @@
 #include <linux/bitops.h>
 #include <linux/kvm_host.h>
 #include <linux/module.h>
+#include <linux/vfio_mdev.h>
 #include <asm/kvm.h>
 #include <asm/zcrypt.h>
 
@@ -321,6 +322,8 @@ static void vfio_ap_matrix_init(struct ap_config_info *info,
 	matrix->adm_max = info->apxa ? info->Nd : 15;
 }
 
+static const struct vfio_mdev_device_ops vfio_mdev_ops;
+
 static int vfio_ap_mdev_create(struct kobject *kobj, struct mdev_device *mdev)
 {
 	struct ap_matrix_mdev *matrix_mdev;
@@ -344,6 +347,7 @@ static int vfio_ap_mdev_create(struct kobject *kobj, struct mdev_device *mdev)
 	mutex_unlock(&matrix_dev->lock);
 
 	mdev_set_class_id(mdev, MDEV_ID_VFIO);
+	mdev_set_dev_ops(mdev, &vfio_mdev_ops);
 	return 0;
 }
 
@@ -1281,15 +1285,18 @@ static ssize_t vfio_ap_mdev_ioctl(struct mdev_device *mdev,
 	return ret;
 }
 
+static const struct vfio_mdev_device_ops vfio_mdev_ops = {
+	.open			= vfio_ap_mdev_open,
+	.release		= vfio_ap_mdev_release,
+	.ioctl			= vfio_ap_mdev_ioctl,
+};
+
 static const struct mdev_parent_ops vfio_ap_matrix_ops = {
 	.owner			= THIS_MODULE,
 	.supported_type_groups	= vfio_ap_mdev_type_groups,
 	.mdev_attr_groups	= vfio_ap_mdev_attr_groups,
 	.create			= vfio_ap_mdev_create,
 	.remove			= vfio_ap_mdev_remove,
-	.open			= vfio_ap_mdev_open,
-	.release		= vfio_ap_mdev_release,
-	.ioctl			= vfio_ap_mdev_ioctl,
 };
 
 int vfio_ap_mdev_register(void)
drivers/vfio/mdev/mdev_core.c +12 -0
index 8764cf4a276d..6f35f2ced2c9 100644
@@ -51,6 +51,18 @@ void mdev_set_class_id(struct mdev_device *mdev, u16 id)
 }
 EXPORT_SYMBOL(mdev_set_class_id);
 
+const void *mdev_get_dev_ops(struct mdev_device *mdev)
+{
+	return mdev->device_ops;
+}
+EXPORT_SYMBOL(mdev_get_dev_ops);
+
+void mdev_set_dev_ops(struct mdev_device *mdev, const void *ops)
+{
+	mdev->device_ops = ops;
+}
+EXPORT_SYMBOL(mdev_set_dev_ops);
+
 struct device *mdev_dev(struct mdev_device *mdev)
 {
 	return &mdev->dev;
drivers/vfio/mdev/mdev_private.h +1 -0
index c65f436c1869..b666805f0b1f 100644
@@ -34,6 +34,7 @@ struct mdev_device {
 	struct device *iommu_device;
 	bool active;
 	u16 class_id;
+	const void *device_ops;
 };
 
 #define to_mdev_device(dev)	container_of(dev, struct mdev_device, dev)
drivers/vfio/mdev/vfio_mdev.c +19 -18
index 891cf83a2d9a..95efa054442f 100644
@@ -14,6 +14,7 @@
 #include <linux/slab.h>
 #include <linux/vfio.h>
 #include <linux/mdev.h>
+#include <linux/vfio_mdev.h>
 
 #include "mdev_private.h"
 
@@ -24,16 +25,16 @@
 static int vfio_mdev_open(void *device_data)
 {
 	struct mdev_device *mdev = device_data;
-	struct mdev_parent *parent = mdev->parent;
+	const struct vfio_mdev_device_ops *ops = mdev_get_dev_ops(mdev);
 	int ret;
 
-	if (unlikely(!parent->ops->open))
+	if (unlikely(!ops->open))
 		return -EINVAL;
 
 	if (!try_module_get(THIS_MODULE))
 		return -ENODEV;
 
-	ret = parent->ops->open(mdev);
+	ret = ops->open(mdev);
 	if (ret)
 		module_put(THIS_MODULE);
 
@@ -43,10 +44,10 @@ static int vfio_mdev_open(void *device_data)
 static void vfio_mdev_release(void *device_data)
 {
 	struct mdev_device *mdev = device_data;
-	struct mdev_parent *parent = mdev->parent;
+	const struct vfio_mdev_device_ops *ops = mdev_get_dev_ops(mdev);
 
-	if (likely(parent->ops->release))
-		parent->ops->release(mdev);
+	if (likely(ops->release))
+		ops->release(mdev);
 
 	module_put(THIS_MODULE);
 }
@@ -55,47 +56,47 @@ static long vfio_mdev_unlocked_ioctl(void *device_data,
 				     unsigned int cmd, unsigned long arg)
 {
 	struct mdev_device *mdev = device_data;
-	struct mdev_parent *parent = mdev->parent;
+	const struct vfio_mdev_device_ops *ops = mdev_get_dev_ops(mdev);
 
-	if (unlikely(!parent->ops->ioctl))
+	if (unlikely(!ops->ioctl))
 		return -EINVAL;
 
-	return parent->ops->ioctl(mdev, cmd, arg);
+	return ops->ioctl(mdev, cmd, arg);
 }
 
 static ssize_t vfio_mdev_read(void *device_data, char __user *buf,
 			      size_t count, loff_t *ppos)
 {
 	struct mdev_device *mdev = device_data;
-	struct mdev_parent *parent = mdev->parent;
+	const struct vfio_mdev_device_ops *ops = mdev_get_dev_ops(mdev);
 
-	if (unlikely(!parent->ops->read))
+	if (unlikely(!ops->read))
 		return -EINVAL;
 
-	return parent->ops->read(mdev, buf, count, ppos);
+	return ops->read(mdev, buf, count, ppos);
 }
 
 static ssize_t vfio_mdev_write(void *device_data, const char __user *buf,
 			       size_t count, loff_t *ppos)
 {
 	struct mdev_device *mdev = device_data;
-	struct mdev_parent *parent = mdev->parent;
+	const struct vfio_mdev_device_ops *ops = mdev_get_dev_ops(mdev);
 
-	if (unlikely(!parent->ops->write))
+	if (unlikely(!ops->write))
 		return -EINVAL;
 
-	return parent->ops->write(mdev, buf, count, ppos);
+	return ops->write(mdev, buf, count, ppos);
 }
 
 static int vfio_mdev_mmap(void *device_data, struct vm_area_struct *vma)
 {
 	struct mdev_device *mdev = device_data;
-	struct mdev_parent *parent = mdev->parent;
+	const struct vfio_mdev_device_ops *ops = mdev_get_dev_ops(mdev);
 
-	if (unlikely(!parent->ops->mmap))
+	if (unlikely(!ops->mmap))
 		return -EINVAL;
 
-	return parent->ops->mmap(mdev, vma);
+	return ops->mmap(mdev, vma);
 }
 
 static const struct vfio_device_ops vfio_mdev_dev_ops = {
include/linux/mdev.h +9 -33
index 3974650c074f..3414307311f1 100644
@@ -10,6 +10,11 @@
 #ifndef MDEV_H
 #define MDEV_H
 
+#include <linux/types.h>
+#include <linux/device.h>
+#include <linux/mdev.h>
+#include <uapi/linux/uuid.h>
+
 struct mdev_device;
 
 /*
@@ -48,30 +53,8 @@ struct device *mdev_get_iommu_device(struct device *dev);
  *			@mdev: mdev_device device structure which is being
  *			       destroyed
  *			Returns integer: success (0) or error (< 0)
- * @open:		Open mediated device.
- *			@mdev: mediated device.
- *			Returns integer: success (0) or error (< 0)
- * @release:		release mediated device
- *			@mdev: mediated device.
- * @read:		Read emulation callback
- *			@mdev: mediated device structure
- *			@buf: read buffer
- *			@count: number of bytes to read
- *			@ppos: address.
- *			Retuns number on bytes read on success or error.
- * @write:		Write emulation callback
- *			@mdev: mediated device structure
- *			@buf: write buffer
- *			@count: number of bytes to be written
- *			@ppos: address.
- *			Retuns number on bytes written on success or error.
- * @ioctl:		IOCTL callback
- *			@mdev: mediated device structure
- *			@cmd: ioctl command
- *			@arg: arguments to ioctl
- * @mmap:		mmap callback
- *			@mdev: mediated device structure
- *			@vma: vma structure
+ * @device_ops:         Device specific emulation callback.
+ *
  * Parent device that support mediated device should be registered with mdev
  * module with mdev_parent_ops structure.
  **/
@@ -83,15 +66,6 @@ struct mdev_parent_ops {
 
 	int     (*create)(struct kobject *kobj, struct mdev_device *mdev);
 	int     (*remove)(struct mdev_device *mdev);
-	int     (*open)(struct mdev_device *mdev);
-	void    (*release)(struct mdev_device *mdev);
-	ssize_t (*read)(struct mdev_device *mdev, char __user *buf,
-			size_t count, loff_t *ppos);
-	ssize_t (*write)(struct mdev_device *mdev, const char __user *buf,
-			 size_t count, loff_t *ppos);
-	long	(*ioctl)(struct mdev_device *mdev, unsigned int cmd,
-			 unsigned long arg);
-	int	(*mmap)(struct mdev_device *mdev, struct vm_area_struct *vma);
 };
 
 /* interface for exporting mdev supported type attributes */
@@ -133,6 +107,8 @@ struct mdev_driver {
 
 void *mdev_get_drvdata(struct mdev_device *mdev);
 void mdev_set_drvdata(struct mdev_device *mdev, void *data);
+const void *mdev_get_dev_ops(struct mdev_device *mdev);
+void mdev_set_dev_ops(struct mdev_device *mdev, const void *ops);
 void mdev_set_class_id(struct mdev_device *mdev, u16 id);
 const guid_t *mdev_uuid(struct mdev_device *mdev);
 
include/linux/vfio_mdev.h new file +52 -0
new file mode 100644
index 000000000000..3907c5371c2b
@@ -0,0 +1,52 @@
+/* SPDX-License-Identifier: GPL-2.0-only */
+/*
+ * VFIO Mediated device definition
+ */
+
+#ifndef VFIO_MDEV_H
+#define VFIO_MDEV_H
+
+#include <linux/mdev.h>
+
+/**
+ * struct vfio_mdev_device_ops - Structure to be registered for each
+ * mdev device to register the device to vfio-mdev module.
+ *
+ * @open:		Open mediated device.
+ *			@mdev: mediated device.
+ *			Returns integer: success (0) or error (< 0)
+ * @release:		release mediated device
+ *			@mdev: mediated device.
+ * @read:		Read emulation callback
+ *			@mdev: mediated device structure
+ *			@buf: read buffer
+ *			@count: number of bytes to read
+ *			@ppos: address.
+ *			Retuns number on bytes read on success or error.
+ * @write:		Write emulation callback
+ *			@mdev: mediated device structure
+ *			@buf: write buffer
+ *			@count: number of bytes to be written
+ *			@ppos: address.
+ *			Retuns number on bytes written on success or error.
+ * @ioctl:		IOCTL callback
+ *			@mdev: mediated device structure
+ *			@cmd: ioctl command
+ *			@arg: arguments to ioctl
+ * @mmap:		mmap callback
+ *			@mdev: mediated device structure
+ *			@vma: vma structure
+ */
+struct vfio_mdev_device_ops {
+	int     (*open)(struct mdev_device *mdev);
+	void    (*release)(struct mdev_device *mdev);
+	ssize_t (*read)(struct mdev_device *mdev, char __user *buf,
+			size_t count, loff_t *ppos);
+	ssize_t (*write)(struct mdev_device *mdev, const char __user *buf,
+			 size_t count, loff_t *ppos);
+	long	(*ioctl)(struct mdev_device *mdev, unsigned int cmd,
+			 unsigned long arg);
+	int	(*mmap)(struct mdev_device *mdev, struct vm_area_struct *vma);
+};
+
+#endif
samples/vfio-mdev/mbochs.c +13 -6
index 8a8583c892b2..b4bf29c6136c 100644
@@ -30,6 +30,7 @@
 #include <linux/iommu.h>
 #include <linux/sysfs.h>
 #include <linux/mdev.h>
+#include <linux/vfio_mdev.h>
 #include <linux/pci.h>
 #include <linux/dma-buf.h>
 #include <linux/highmem.h>
@@ -516,6 +517,8 @@ static int mbochs_reset(struct mdev_device *mdev)
 	return 0;
 }
 
+static const struct vfio_mdev_device_ops vfio_mdev_ops;
+
 static int mbochs_create(struct kobject *kobj, struct mdev_device *mdev)
 {
 	const struct mbochs_type *type = mbochs_find_type(kobj);
@@ -562,6 +565,7 @@ static int mbochs_create(struct kobject *kobj, struct mdev_device *mdev)
 
 	mbochs_used_mbytes += type->mbytes;
 	mdev_set_class_id(mdev, MDEV_ID_VFIO);
+	mdev_set_dev_ops(mdev, &vfio_mdev_ops);
 	return 0;
 
 err_mem:
@@ -1419,12 +1423,7 @@ static struct attribute_group *mdev_type_groups[] = {
 	NULL,
 };
 
-static const struct mdev_parent_ops mdev_fops = {
-	.owner			= THIS_MODULE,
-	.mdev_attr_groups	= mdev_dev_groups,
-	.supported_type_groups	= mdev_type_groups,
-	.create			= mbochs_create,
-	.remove			= mbochs_remove,
+static const struct vfio_mdev_device_ops vfio_mdev_ops = {
 	.open			= mbochs_open,
 	.release		= mbochs_close,
 	.read			= mbochs_read,
@@ -1433,6 +1432,14 @@ static const struct mdev_parent_ops mdev_fops = {
 	.mmap			= mbochs_mmap,
 };
 
+static const struct mdev_parent_ops mdev_fops = {
+	.owner			= THIS_MODULE,
+	.mdev_attr_groups	= mdev_dev_groups,
+	.supported_type_groups	= mdev_type_groups,
+	.create			= mbochs_create,
+	.remove			= mbochs_remove,
+};
+
 static const struct file_operations vd_fops = {
 	.owner		= THIS_MODULE,
 };
samples/vfio-mdev/mdpy.c +13 -6
index 88d7e76f3836..80c2df531326 100644
@@ -26,6 +26,7 @@
 #include <linux/iommu.h>
 #include <linux/sysfs.h>
 #include <linux/mdev.h>
+#include <linux/vfio_mdev.h>
 #include <linux/pci.h>
 #include <drm/drm_fourcc.h>
 #include "mdpy-defs.h"
@@ -226,6 +227,8 @@ static int mdpy_reset(struct mdev_device *mdev)
 	return 0;
 }
 
+static const struct vfio_mdev_device_ops vfio_mdev_ops;
+
 static int mdpy_create(struct kobject *kobj, struct mdev_device *mdev)
 {
 	const struct mdpy_type *type = mdpy_find_type(kobj);
@@ -270,6 +273,7 @@ static int mdpy_create(struct kobject *kobj, struct mdev_device *mdev)
 
 	mdpy_count++;
 	mdev_set_class_id(mdev, MDEV_ID_VFIO);
+	mdev_set_dev_ops(mdev, &vfio_mdev_ops);
 	return 0;
 }
 
@@ -726,12 +730,7 @@ static struct attribute_group *mdev_type_groups[] = {
 	NULL,
 };
 
-static const struct mdev_parent_ops mdev_fops = {
-	.owner			= THIS_MODULE,
-	.mdev_attr_groups	= mdev_dev_groups,
-	.supported_type_groups	= mdev_type_groups,
-	.create			= mdpy_create,
-	.remove			= mdpy_remove,
+static const struct vfio_mdev_device_ops vfio_mdev_ops = {
 	.open			= mdpy_open,
 	.release		= mdpy_close,
 	.read			= mdpy_read,
@@ -740,6 +739,14 @@ static const struct mdev_parent_ops mdev_fops = {
 	.mmap			= mdpy_mmap,
 };
 
+static const struct mdev_parent_ops mdev_fops = {
+	.owner			= THIS_MODULE,
+	.mdev_attr_groups	= mdev_dev_groups,
+	.supported_type_groups	= mdev_type_groups,
+	.create			= mdpy_create,
+	.remove			= mdpy_remove,
+};
+
 static const struct file_operations vd_fops = {
 	.owner		= THIS_MODULE,
 };
samples/vfio-mdev/mtty.c +12 -5
index 4e0735143b69..2db860ccb02c 100644
@@ -27,6 +27,7 @@
 #include <linux/ctype.h>
 #include <linux/file.h>
 #include <linux/mdev.h>
+#include <linux/vfio_mdev.h>
 #include <linux/pci.h>
 #include <linux/serial.h>
 #include <uapi/linux/serial_reg.h>
@@ -723,6 +724,8 @@ static ssize_t mdev_access(struct mdev_device *mdev, u8 *buf, size_t count,
 	return ret;
 }
 
+static const struct vfio_mdev_device_ops vfio_dev_ops;
+
 static int mtty_create(struct kobject *kobj, struct mdev_device *mdev)
 {
 	struct mdev_state *mdev_state;
@@ -771,6 +774,7 @@ static int mtty_create(struct kobject *kobj, struct mdev_device *mdev)
 	mutex_unlock(&mdev_list_lock);
 
 	mdev_set_class_id(mdev, MDEV_ID_VFIO);
+	mdev_set_dev_ops(mdev, &vfio_dev_ops);
 	return 0;
 }
 
@@ -1411,6 +1415,14 @@ static struct attribute_group *mdev_type_groups[] = {
 	NULL,
 };
 
+static const struct vfio_mdev_device_ops vfio_dev_ops = {
+	.open                   = mtty_open,
+	.release                = mtty_close,
+	.read                   = mtty_read,
+	.write                  = mtty_write,
+	.ioctl		        = mtty_ioctl,
+};
+
 static const struct mdev_parent_ops mdev_fops = {
 	.owner                  = THIS_MODULE,
 	.dev_attr_groups        = mtty_dev_groups,
@@ -1418,11 +1430,6 @@ static const struct mdev_parent_ops mdev_fops = {
 	.supported_type_groups  = mdev_type_groups,
 	.create                 = mtty_create,
 	.remove			= mtty_remove,
-	.open                   = mtty_open,
-	.release                = mtty_close,
-	.read                   = mtty_read,
-	.write                  = mtty_write,
-	.ioctl		        = mtty_ioctl,
 };
 
 static void mtty_device_release(struct device *dev)
author
#32 lore raw

PATCH v2 5/8 mdev: introduce device specific ops

In reply to Jason Wang

Currently, except for the create and remove, the rest of mdev_parent_ops is designed for vfio-mdev driver only and may not help for kernel mdev driver. With the help of class id, this patch introduces device specific callbacks inside mdev_device structure. This allows different set of callback to be used by vfio-mdev and virtio-mdev.

---
 .../driver-api/vfio-mediated-device.rst       |  4 +-
 MAINTAINERS                                   |  1 +
 drivers/gpu/drm/i915/gvt/kvmgt.c              | 17 +++---
 drivers/s390/cio/vfio_ccw_ops.c               | 17 ++++--
 drivers/s390/crypto/vfio_ap_ops.c             | 13 +++--
 drivers/vfio/mdev/mdev_core.c                 | 12 +++++
 drivers/vfio/mdev/mdev_private.h              |  1 +
 drivers/vfio/mdev/vfio_mdev.c                 | 37 ++++++-------
 include/linux/mdev.h                          | 42 ++++-----------
 include/linux/vfio_mdev.h                     | 52 +++++++++++++++++++
 samples/vfio-mdev/mbochs.c                    | 19 ++++---
 samples/vfio-mdev/mdpy.c                      | 19 ++++---
 samples/vfio-mdev/mtty.c                      | 17 ++++--
 13 files changed, 168 insertions(+), 83 deletions(-)
 create mode 100644 include/linux/vfio_mdev.h
Documentation/driver-api/vfio-mediated-device.rst +3 -1
index a5bdc60d62a1..d50425b368bb 100644
@@ -152,7 +152,9 @@ callbacks per mdev parent device, per mdev type, or any other categorization.
 Vendor drivers are expected to be fully asynchronous in this respect or
 provide their own internal resource protection.)
 
-The callbacks in the mdev_parent_ops structure are as follows:
+The device specific callbacks are referred through device_ops pointer
+in mdev_parent_ops. For vfio-mdev device, its callbacks in device_ops
+are as follows:
 
 * open: open callback of mediated device
 * close: close callback of mediated device
MAINTAINERS +1 -0
index b2326dece28e..89832b316500 100644
@@ -17075,6 +17075,7 @@ S:	Maintained
 F:	Documentation/driver-api/vfio-mediated-device.rst
 F:	drivers/vfio/mdev/
 F:	include/linux/mdev.h
+F:	include/linux/vfio_mdev.h
 F:	samples/vfio-mdev/
 
 VFIO PLATFORM DRIVER
drivers/gpu/drm/i915/gvt/kvmgt.c +11 -6
index f793252a3d2a..b274f5ee481f 100644
@@ -42,6 +42,7 @@
 #include <linux/kvm_host.h>
 #include <linux/vfio.h>
 #include <linux/mdev.h>
+#include <linux/vfio_mdev.h>
 #include <linux/debugfs.h>
 
 #include <linux/nospec.h>
@@ -643,6 +644,8 @@ static void kvmgt_put_vfio_device(void *vgpu)
 	vfio_device_put(((struct intel_vgpu *)vgpu)->vdev.vfio_device);
 }
 
+static struct vfio_mdev_device_ops intel_vfio_vgpu_dev_ops;
+
 static int intel_vgpu_create(struct kobject *kobj, struct mdev_device *mdev)
 {
 	struct intel_vgpu *vgpu = NULL;
@@ -679,6 +682,7 @@ static int intel_vgpu_create(struct kobject *kobj, struct mdev_device *mdev)
 	ret = 0;
 
 	mdev_set_class_id(mdev, MDEV_ID_VFIO);
+	mdev_set_dev_ops(mdev, &intel_vfio_vgpu_dev_ops);
 out:
 	return ret;
 }
@@ -1601,20 +1605,21 @@ static const struct attribute_group *intel_vgpu_groups[] = {
 	NULL,
 };
 
-static struct mdev_parent_ops intel_vgpu_ops = {
-	.mdev_attr_groups       = intel_vgpu_groups,
-	.create			= intel_vgpu_create,
-	.remove			= intel_vgpu_remove,
-
+static struct vfio_mdev_device_ops intel_vfio_vgpu_dev_ops = {
 	.open			= intel_vgpu_open,
 	.release		= intel_vgpu_release,
-
 	.read			= intel_vgpu_read,
 	.write			= intel_vgpu_write,
 	.mmap			= intel_vgpu_mmap,
 	.ioctl			= intel_vgpu_ioctl,
 };
 
+static struct mdev_parent_ops intel_vgpu_ops = {
+	.mdev_attr_groups       = intel_vgpu_groups,
+	.create			= intel_vgpu_create,
+	.remove			= intel_vgpu_remove,
+};
+
 static int kvmgt_host_init(struct device *dev, void *gvt, const void *ops)
 {
 	struct attribute **kvm_type_attrs;
drivers/s390/cio/vfio_ccw_ops.c +12 -5
index d258ef1fedb9..329d53c1f46b 100644
@@ -12,6 +12,7 @@
 
 #include <linux/vfio.h>
 #include <linux/mdev.h>
+#include <linux/vfio_mdev.h>
 #include <linux/nospec.h>
 #include <linux/slab.h>
 
@@ -110,6 +111,8 @@ static struct attribute_group *mdev_type_groups[] = {
 	NULL,
 };
 
+static const struct vfio_mdev_device_ops vfio_mdev_ops;
+
 static int vfio_ccw_mdev_create(struct kobject *kobj, struct mdev_device *mdev)
 {
 	struct vfio_ccw_private *private =
@@ -130,6 +133,7 @@ static int vfio_ccw_mdev_create(struct kobject *kobj, struct mdev_device *mdev)
 			   private->sch->schid.sch_no);
 
 	mdev_set_class_id(mdev, MDEV_ID_VFIO);
+	mdev_set_dev_ops(mdev, &vfio_mdev_ops);
 	return 0;
 }
 
@@ -575,11 +579,7 @@ static ssize_t vfio_ccw_mdev_ioctl(struct mdev_device *mdev,
 	}
 }
 
-static const struct mdev_parent_ops vfio_ccw_mdev_ops = {
-	.owner			= THIS_MODULE,
-	.supported_type_groups  = mdev_type_groups,
-	.create			= vfio_ccw_mdev_create,
-	.remove			= vfio_ccw_mdev_remove,
+static const struct vfio_mdev_device_ops vfio_mdev_ops = {
 	.open			= vfio_ccw_mdev_open,
 	.release		= vfio_ccw_mdev_release,
 	.read			= vfio_ccw_mdev_read,
@@ -587,6 +587,13 @@ static const struct mdev_parent_ops vfio_ccw_mdev_ops = {
 	.ioctl			= vfio_ccw_mdev_ioctl,
 };
 
+static const struct mdev_parent_ops vfio_ccw_mdev_ops = {
+	.owner			= THIS_MODULE,
+	.supported_type_groups  = mdev_type_groups,
+	.create			= vfio_ccw_mdev_create,
+	.remove			= vfio_ccw_mdev_remove,
+};
+
 int vfio_ccw_mdev_reg(struct subchannel *sch)
 {
 	return mdev_register_device(&sch->dev, &vfio_ccw_mdev_ops);
drivers/s390/crypto/vfio_ap_ops.c +10 -3
index 2cfd96112aa0..3a89933f0d3e 100644
@@ -16,6 +16,7 @@
 #include <linux/bitops.h>
 #include <linux/kvm_host.h>
 #include <linux/module.h>
+#include <linux/vfio_mdev.h>
 #include <asm/kvm.h>
 #include <asm/zcrypt.h>
 
@@ -321,6 +322,8 @@ static void vfio_ap_matrix_init(struct ap_config_info *info,
 	matrix->adm_max = info->apxa ? info->Nd : 15;
 }
 
+static const struct vfio_mdev_device_ops vfio_mdev_ops;
+
 static int vfio_ap_mdev_create(struct kobject *kobj, struct mdev_device *mdev)
 {
 	struct ap_matrix_mdev *matrix_mdev;
@@ -344,6 +347,7 @@ static int vfio_ap_mdev_create(struct kobject *kobj, struct mdev_device *mdev)
 	mutex_unlock(&matrix_dev->lock);
 
 	mdev_set_class_id(mdev, MDEV_ID_VFIO);
+	mdev_set_dev_ops(mdev, &vfio_mdev_ops);
 	return 0;
 }
 
@@ -1281,15 +1285,18 @@ static ssize_t vfio_ap_mdev_ioctl(struct mdev_device *mdev,
 	return ret;
 }
 
+static const struct vfio_mdev_device_ops vfio_mdev_ops = {
+	.open			= vfio_ap_mdev_open,
+	.release		= vfio_ap_mdev_release,
+	.ioctl			= vfio_ap_mdev_ioctl,
+};
+
 static const struct mdev_parent_ops vfio_ap_matrix_ops = {
 	.owner			= THIS_MODULE,
 	.supported_type_groups	= vfio_ap_mdev_type_groups,
 	.mdev_attr_groups	= vfio_ap_mdev_attr_groups,
 	.create			= vfio_ap_mdev_create,
 	.remove			= vfio_ap_mdev_remove,
-	.open			= vfio_ap_mdev_open,
-	.release		= vfio_ap_mdev_release,
-	.ioctl			= vfio_ap_mdev_ioctl,
 };
 
 int vfio_ap_mdev_register(void)
drivers/vfio/mdev/mdev_core.c +12 -0
index 8764cf4a276d..6f35f2ced2c9 100644
@@ -51,6 +51,18 @@ void mdev_set_class_id(struct mdev_device *mdev, u16 id)
 }
 EXPORT_SYMBOL(mdev_set_class_id);
 
+const void *mdev_get_dev_ops(struct mdev_device *mdev)
+{
+	return mdev->device_ops;
+}
+EXPORT_SYMBOL(mdev_get_dev_ops);
+
+void mdev_set_dev_ops(struct mdev_device *mdev, const void *ops)
+{
+	mdev->device_ops = ops;
+}
+EXPORT_SYMBOL(mdev_set_dev_ops);
+
 struct device *mdev_dev(struct mdev_device *mdev)
 {
 	return &mdev->dev;
drivers/vfio/mdev/mdev_private.h +1 -0
index c65f436c1869..b666805f0b1f 100644
@@ -34,6 +34,7 @@ struct mdev_device {
 	struct device *iommu_device;
 	bool active;
 	u16 class_id;
+	const void *device_ops;
 };
 
 #define to_mdev_device(dev)	container_of(dev, struct mdev_device, dev)
drivers/vfio/mdev/vfio_mdev.c +19 -18
index 891cf83a2d9a..95efa054442f 100644
@@ -14,6 +14,7 @@
 #include <linux/slab.h>
 #include <linux/vfio.h>
 #include <linux/mdev.h>
+#include <linux/vfio_mdev.h>
 
 #include "mdev_private.h"
 
@@ -24,16 +25,16 @@
 static int vfio_mdev_open(void *device_data)
 {
 	struct mdev_device *mdev = device_data;
-	struct mdev_parent *parent = mdev->parent;
+	const struct vfio_mdev_device_ops *ops = mdev_get_dev_ops(mdev);
 	int ret;
 
-	if (unlikely(!parent->ops->open))
+	if (unlikely(!ops->open))
 		return -EINVAL;
 
 	if (!try_module_get(THIS_MODULE))
 		return -ENODEV;
 
-	ret = parent->ops->open(mdev);
+	ret = ops->open(mdev);
 	if (ret)
 		module_put(THIS_MODULE);
 
@@ -43,10 +44,10 @@ static int vfio_mdev_open(void *device_data)
 static void vfio_mdev_release(void *device_data)
 {
 	struct mdev_device *mdev = device_data;
-	struct mdev_parent *parent = mdev->parent;
+	const struct vfio_mdev_device_ops *ops = mdev_get_dev_ops(mdev);
 
-	if (likely(parent->ops->release))
-		parent->ops->release(mdev);
+	if (likely(ops->release))
+		ops->release(mdev);
 
 	module_put(THIS_MODULE);
 }
@@ -55,47 +56,47 @@ static long vfio_mdev_unlocked_ioctl(void *device_data,
 				     unsigned int cmd, unsigned long arg)
 {
 	struct mdev_device *mdev = device_data;
-	struct mdev_parent *parent = mdev->parent;
+	const struct vfio_mdev_device_ops *ops = mdev_get_dev_ops(mdev);
 
-	if (unlikely(!parent->ops->ioctl))
+	if (unlikely(!ops->ioctl))
 		return -EINVAL;
 
-	return parent->ops->ioctl(mdev, cmd, arg);
+	return ops->ioctl(mdev, cmd, arg);
 }
 
 static ssize_t vfio_mdev_read(void *device_data, char __user *buf,
 			      size_t count, loff_t *ppos)
 {
 	struct mdev_device *mdev = device_data;
-	struct mdev_parent *parent = mdev->parent;
+	const struct vfio_mdev_device_ops *ops = mdev_get_dev_ops(mdev);
 
-	if (unlikely(!parent->ops->read))
+	if (unlikely(!ops->read))
 		return -EINVAL;
 
-	return parent->ops->read(mdev, buf, count, ppos);
+	return ops->read(mdev, buf, count, ppos);
 }
 
 static ssize_t vfio_mdev_write(void *device_data, const char __user *buf,
 			       size_t count, loff_t *ppos)
 {
 	struct mdev_device *mdev = device_data;
-	struct mdev_parent *parent = mdev->parent;
+	const struct vfio_mdev_device_ops *ops = mdev_get_dev_ops(mdev);
 
-	if (unlikely(!parent->ops->write))
+	if (unlikely(!ops->write))
 		return -EINVAL;
 
-	return parent->ops->write(mdev, buf, count, ppos);
+	return ops->write(mdev, buf, count, ppos);
 }
 
 static int vfio_mdev_mmap(void *device_data, struct vm_area_struct *vma)
 {
 	struct mdev_device *mdev = device_data;
-	struct mdev_parent *parent = mdev->parent;
+	const struct vfio_mdev_device_ops *ops = mdev_get_dev_ops(mdev);
 
-	if (unlikely(!parent->ops->mmap))
+	if (unlikely(!ops->mmap))
 		return -EINVAL;
 
-	return parent->ops->mmap(mdev, vma);
+	return ops->mmap(mdev, vma);
 }
 
 static const struct vfio_device_ops vfio_mdev_dev_ops = {
include/linux/mdev.h +9 -33

66 lines not on this page · raw message

include/linux/vfio_mdev.h new file +52 -0

52 lines not on this page · raw message

samples/vfio-mdev/mbochs.c +13 -6
index 8a8583c892b2..b4bf29c6136c 100644
@@ -30,6 +30,7 @@
 #include <linux/iommu.h>
 #include <linux/sysfs.h>
 #include <linux/mdev.h>
+#include <linux/vfio_mdev.h>
 #include <linux/pci.h>
 #include <linux/dma-buf.h>
 #include <linux/highmem.h>
@@ -516,6 +517,8 @@ static int mbochs_reset(struct mdev_device *mdev)
 	return 0;
 }
 
+static const struct vfio_mdev_device_ops vfio_mdev_ops;
+
 static int mbochs_create(struct kobject *kobj, struct mdev_device *mdev)
 {
 	const struct mbochs_type *type = mbochs_find_type(kobj);
@@ -562,6 +565,7 @@ static int mbochs_create(struct kobject *kobj, struct mdev_device *mdev)
 
 	mbochs_used_mbytes += type->mbytes;
 	mdev_set_class_id(mdev, MDEV_ID_VFIO);
+	mdev_set_dev_ops(mdev, &vfio_mdev_ops);
 	return 0;
 
 err_mem:
@@ -1419,12 +1423,7 @@ static struct attribute_group *mdev_type_groups[] = {
 	NULL,
 };
 
-static const struct mdev_parent_ops mdev_fops = {
-	.owner			= THIS_MODULE,
-	.mdev_attr_groups	= mdev_dev_groups,
-	.supported_type_groups	= mdev_type_groups,
-	.create			= mbochs_create,
-	.remove			= mbochs_remove,
+static const struct vfio_mdev_device_ops vfio_mdev_ops = {
 	.open			= mbochs_open,
 	.release		= mbochs_close,
 	.read			= mbochs_read,
@@ -1433,6 +1432,14 @@ static const struct mdev_parent_ops mdev_fops = {
 	.mmap			= mbochs_mmap,
 };
 
+static const struct mdev_parent_ops mdev_fops = {
+	.owner			= THIS_MODULE,
+	.mdev_attr_groups	= mdev_dev_groups,
+	.supported_type_groups	= mdev_type_groups,
+	.create			= mbochs_create,
+	.remove			= mbochs_remove,
+};
+
 static const struct file_operations vd_fops = {
 	.owner		= THIS_MODULE,
 };
samples/vfio-mdev/mdpy.c +13 -6
index 88d7e76f3836..80c2df531326 100644
@@ -26,6 +26,7 @@
 #include <linux/iommu.h>
 #include <linux/sysfs.h>
 #include <linux/mdev.h>
+#include <linux/vfio_mdev.h>
 #include <linux/pci.h>
 #include <drm/drm_fourcc.h>
 #include "mdpy-defs.h"
@@ -226,6 +227,8 @@ static int mdpy_reset(struct mdev_device *mdev)
 	return 0;
 }
 
+static const struct vfio_mdev_device_ops vfio_mdev_ops;
+
 static int mdpy_create(struct kobject *kobj, struct mdev_device *mdev)
 {
 	const struct mdpy_type *type = mdpy_find_type(kobj);
@@ -270,6 +273,7 @@ static int mdpy_create(struct kobject *kobj, struct mdev_device *mdev)
 
 	mdpy_count++;
 	mdev_set_class_id(mdev, MDEV_ID_VFIO);
+	mdev_set_dev_ops(mdev, &vfio_mdev_ops);
 	return 0;
 }
 
@@ -726,12 +730,7 @@ static struct attribute_group *mdev_type_groups[] = {
 	NULL,
 };
 
-static const struct mdev_parent_ops mdev_fops = {
-	.owner			= THIS_MODULE,
-	.mdev_attr_groups	= mdev_dev_groups,
-	.supported_type_groups	= mdev_type_groups,
-	.create			= mdpy_create,
-	.remove			= mdpy_remove,
+static const struct vfio_mdev_device_ops vfio_mdev_ops = {
 	.open			= mdpy_open,
 	.release		= mdpy_close,
 	.read			= mdpy_read,
@@ -740,6 +739,14 @@ static const struct mdev_parent_ops mdev_fops = {
 	.mmap			= mdpy_mmap,
 };
 
+static const struct mdev_parent_ops mdev_fops = {
+	.owner			= THIS_MODULE,
+	.mdev_attr_groups	= mdev_dev_groups,
+	.supported_type_groups	= mdev_type_groups,
+	.create			= mdpy_create,
+	.remove			= mdpy_remove,
+};
+
 static const struct file_operations vd_fops = {
 	.owner		= THIS_MODULE,
 };
samples/vfio-mdev/mtty.c +12 -5
index 4e0735143b69..2db860ccb02c 100644
@@ -27,6 +27,7 @@
 #include <linux/ctype.h>
 #include <linux/file.h>
 #include <linux/mdev.h>
+#include <linux/vfio_mdev.h>
 #include <linux/pci.h>
 #include <linux/serial.h>
 #include <uapi/linux/serial_reg.h>
@@ -723,6 +724,8 @@ static ssize_t mdev_access(struct mdev_device *mdev, u8 *buf, size_t count,
 	return ret;
 }
 
+static const struct vfio_mdev_device_ops vfio_dev_ops;
+
 static int mtty_create(struct kobject *kobj, struct mdev_device *mdev)
 {
 	struct mdev_state *mdev_state;
@@ -771,6 +774,7 @@ static int mtty_create(struct kobject *kobj, struct mdev_device *mdev)
 	mutex_unlock(&mdev_list_lock);
 
 	mdev_set_class_id(mdev, MDEV_ID_VFIO);
+	mdev_set_dev_ops(mdev, &vfio_dev_ops);
 	return 0;
 }
 
@@ -1411,6 +1415,14 @@ static struct attribute_group *mdev_type_groups[] = {
 	NULL,
 };
 
+static const struct vfio_mdev_device_ops vfio_dev_ops = {
+	.open                   = mtty_open,
+	.release                = mtty_close,
+	.read                   = mtty_read,
+	.write                  = mtty_write,
+	.ioctl		        = mtty_ioctl,
+};
+
 static const struct mdev_parent_ops mdev_fops = {
 	.owner                  = THIS_MODULE,
 	.dev_attr_groups        = mtty_dev_groups,
@@ -1418,11 +1430,6 @@ static const struct mdev_parent_ops mdev_fops = {
 	.supported_type_groups  = mdev_type_groups,
 	.create                 = mtty_create,
 	.remove			= mtty_remove,
-	.open                   = mtty_open,
-	.release                = mtty_close,
-	.read                   = mtty_read,
-	.write                  = mtty_write,
-	.ioctl		        = mtty_ioctl,
 };
 
 static void mtty_device_release(struct device *dev)
Signature · 1 line
2.19.1
author
#33 lore raw

PATCH v2 6/8 mdev: introduce virtio device and its device ops

In reply to Jason Wang

This patch implements basic support for mdev driver that supports virtio transport for kernel virtio driver.

---
 include/linux/mdev.h        |   2 +
 include/linux/virtio_mdev.h | 145 ++++++++++++++++++++++++++++++++++++
 2 files changed, 147 insertions(+)
 create mode 100644 include/linux/virtio_mdev.h
include/linux/mdev.h +2 -0
index 3414307311f1..73ac27b3b868 100644
@@ -126,6 +126,8 @@ struct mdev_device *mdev_from_dev(struct device *dev);
 
 enum {
 	MDEV_ID_VFIO = 1,
+	MDEV_ID_VIRTIO = 2,
+	MDEV_ID_VHOST = 3,
 	/* New entries must be added here */
 };
 
include/linux/virtio_mdev.h new file +145 -0

145 lines not on this page · raw message

Signature · 1 line
2.19.1
#34 lore raw

Re: [PATCH V2 6/8] mdev: introduce virtio device and its device ops

Quoted 15 lines · On Tue, 24 Sep 2019 21:53:30 +0800 Jason Wang <[email protected]> wrote:

This patch implements basic support for mdev driver that supports virtio transport for kernel virtio driver.

---
 include/linux/mdev.h        |   2 +
 include/linux/virtio_mdev.h | 145 ++++++++++++++++++++++++++++++++++++
 2 files changed, 147 insertions(+)
 create mode 100644 include/linux/virtio_mdev.h
include/linux/mdev.h +0 -0
index 3414307311f1..73ac27b3b868 100644

include/linux/mdev.h struct mdev_device *mdev_from_dev(struct device *dev); · line 130

Quoted 5 lines
include/linux/mdev.h +2 -0
@@ -126,6 +126,8 @@ struct mdev_device *mdev_from_dev(struct device *dev);
 
 enum {
 	MDEV_ID_VFIO = 1,
+	MDEV_ID_VIRTIO = 2,
+	MDEV_ID_VHOST = 3,

MDEV_ID_VHOST isn't used yet here. Also, given the strong interdependence between the class_id and the ops structure, we might wand to define them in the same place. Thanks,

Alex

Quoted 154 lines
include/linux/mdev.h +0 -0
 	/* New entries must be added here */
 };
 
include/linux/virtio_mdev.h new file +145 -0
new file mode 100644
index 000000000000..d1a40a739266
@@ -0,0 +1,145 @@
+/* SPDX-License-Identifier: GPL-2.0-only */
+/*
+ * Virtio mediated device driver
+ *
+ * Copyright 2019, Red Hat Corp.
+ *     Author: Jason Wang <[email protected]>
+ */
+#ifndef _LINUX_VIRTIO_MDEV_H
+#define _LINUX_VIRTIO_MDEV_H
+
+#include <linux/interrupt.h>
+#include <linux/mdev.h>
+#include <uapi/linux/vhost.h>
+
+#define VIRTIO_MDEV_DEVICE_API_STRING		"virtio-mdev"
+#define VIRTIO_MDEV_VERSION 0x1
+
+struct virtio_mdev_callback {
+	irqreturn_t (*callback)(void *data);
+	void *private;
+};
+
+/**
+ * struct vfio_mdev_device_ops - Structure to be registered for each
+ * mdev device to register the device to virtio-mdev module.
+ *
+ * @set_vq_address:		Set the address of virtqueue
+ *				@mdev: mediated device
+ *				@idx: virtqueue index
+ *				@desc_area: address of desc area
+ *				@driver_area: address of driver area
+ *				@device_area: address of device area
+ *				Returns integer: success (0) or error (< 0)
+ * @set_vq_num:		Set the size of virtqueue
+ *				@mdev: mediated device
+ *				@idx: virtqueue index
+ *				@num: the size of virtqueue
+ * @kick_vq:			Kick the virtqueue
+ *				@mdev: mediated device
+ *				@idx: virtqueue index
+ * @set_vq_cb:			Set the interrut calback function for
+ *				a virtqueue
+ *				@mdev: mediated device
+ *				@idx: virtqueue index
+ *				@cb: virtio-mdev interrupt callback structure
+ * @set_vq_ready:		Set ready status for a virtqueue
+ *				@mdev: mediated device
+ *				@idx: virtqueue index
+ *				@ready: ready (true) not ready(false)
+ * @get_vq_ready:		Get ready status for a virtqueue
+ *				@mdev: mediated device
+ *				@idx: virtqueue index
+ *				Returns boolean: ready (true) or not (false)
+ * @set_vq_state:		Set the state for a virtqueue
+ *				@mdev: mediated device
+ *				@idx: virtqueue index
+ *				@state: virtqueue state (last_avail_idx)
+ *				Returns integer: success (0) or error (< 0)
+ * @get_vq_state:		Get the state for a virtqueue
+ *				@mdev: mediated device
+ *				@idx: virtqueue index
+ *				Returns virtqueue state (last_avail_idx)
+ * @get_vq_align:		Get the virtqueue align requirement
+ *				for the device
+ *				@mdev: mediated device
+ *				Returns virtqueue algin requirement
+ * @get_features:		Get virtio features supported by the device
+ *				@mdev: mediated device
+ *				Returns the features support by the
+ *				device
+ * @get_features:		Set virtio features supported by the driver
+ *				@mdev: mediated device
+ *				@features: feature support by the driver
+ *				Returns integer: success (0) or error (< 0)
+ * @set_config_cb:		Set the config interrupt callback
+ *				@mdev: mediated device
+ *				@cb: virtio-mdev interrupt callback structure
+ * @get_device_id:		Get virtio device id
+ *				@mdev: mediated device
+ *				Returns u32: virtio device id
+ * @get_vendor_id:		Get virtio vendor id
+ *				@mdev: mediated device
+ *				Returns u32: virtio vendor id
+ * @get_status:		Get the device status
+ *				@mdev: mediated device
+ *				Returns u8: virtio device status
+ * @set_status:		Set the device status
+ *				@mdev: mediated device
+ *				@status: virtio device status
+ * @get_config:		Read from device specific confiugration space
+ *				@mdev: mediated device
+ *				@offset: offset from the beginning of
+ *				configuration space
+ *				@buf: buffer used to read to
+ *				@len: the length to read from
+ *				configration space
+ * @set_config:		Write to device specific confiugration space
+ *				@mdev: mediated device
+ *				@offset: offset from the beginning of
+ *				configuration space
+ *				@buf: buffer used to write from
+ *				@len: the length to write to
+ *				configration space
+ * @get_version:		Get the version of virtio mdev device
+ *				@mdev: mediated device
+ *				Returns integer: version of the device
+ * @get_generation:		Get device generaton
+ *				@mdev: mediated device
+ *				Returns u32: device generation
+ */
+struct virtio_mdev_device_ops {
+	/* Virtqueue ops */
+	int (*set_vq_address)(struct mdev_device *mdev,
+			      u16 idx, u64 desc_area, u64 driver_area,
+			      u64 device_area);
+	void (*set_vq_num)(struct mdev_device *mdev, u16 idx, u32 num);
+	void (*kick_vq)(struct mdev_device *mdev, u16 idx);
+	void (*set_vq_cb)(struct mdev_device *mdev, u16 idx,
+			  struct virtio_mdev_callback *cb);
+	void (*set_vq_ready)(struct mdev_device *mdev, u16 idx, bool ready);
+	bool (*get_vq_ready)(struct mdev_device *mdev, u16 idx);
+	int (*set_vq_state)(struct mdev_device *mdev, u16 idx, u64 state);
+	u64 (*get_vq_state)(struct mdev_device *mdev, u16 idx);
+
+	/* Device ops */
+	u16 (*get_vq_align)(struct mdev_device *mdev);
+	u64 (*get_features)(struct mdev_device *mdev);
+	int (*set_features)(struct mdev_device *mdev, u64 features);
+	void (*set_config_cb)(struct mdev_device *mdev,
+			      struct virtio_mdev_callback *cb);
+	u16 (*get_queue_max)(struct mdev_device *mdev);
+	u32 (*get_device_id)(struct mdev_device *mdev);
+	u32 (*get_vendor_id)(struct mdev_device *mdev);
+	u8 (*get_status)(struct mdev_device *mdev);
+	void (*set_status)(struct mdev_device *mdev, u8 status);
+	void (*get_config)(struct mdev_device *mdev, unsigned int offset,
+			   void *buf, unsigned int len);
+	void (*set_config)(struct mdev_device *mdev, unsigned int offset,
+			   const void *buf, unsigned int len);
+	int (*get_version)(struct mdev_device *mdev);
+	u32 (*get_generation)(struct mdev_device *mdev);
+};
+
+#endif
+
author
#35 lore raw

Re: [PATCH V2 6/8] mdev: introduce virtio device and its device ops

Quoted 19 lines · Alex Williamson
Quoted 16 lines · On Tue, 24 Sep 2019 21:53:30 +0800 Jason Wang <[email protected]> wrote:

This patch implements basic support for mdev driver that supports virtio transport for kernel virtio driver.

---
 include/linux/mdev.h        |   2 +
 include/linux/virtio_mdev.h | 145 ++++++++++++++++++++++++++++++++++++
 2 files changed, 147 insertions(+)
 create mode 100644 include/linux/virtio_mdev.h
include/linux/mdev.h +0 -0
index 3414307311f1..73ac27b3b868 100644
@@ -126,6 +126,8 @@ struct mdev_device *mdev_from_dev(struct device *dev);
 
Quoted 5 lines

include/linux/mdev.h struct mdev_device *mdev_from_dev(struct device *dev); · line 130

Quoted 4 lines
include/linux/mdev.h +2 -0
 enum {
 	MDEV_ID_VFIO = 1,
+	MDEV_ID_VIRTIO = 2,
+	MDEV_ID_VHOST = 3,

MDEV_ID_VHOST isn't used yet here.

Yes, just want to reserve one but it look too early to do that.

Quoted 5 lines

Also, given the strong
interdependence between the class_id and the ops structure, we might wand to define them in the same place. Thanks,

Alex

Yes, as you suggest in patch 5/8.

Thanks

Quoted 155 lines
Quoted 155 lines
include/linux/mdev.h +0 -0
 	/* New entries must be added here */
 };
 
include/linux/virtio_mdev.h new file +145 -0
new file mode 100644
index 000000000000..d1a40a739266
@@ -0,0 +1,145 @@
+/* SPDX-License-Identifier: GPL-2.0-only */
+/*
+ * Virtio mediated device driver
+ *
+ * Copyright 2019, Red Hat Corp.
+ *     Author: Jason Wang <[email protected]>
+ */
+#ifndef _LINUX_VIRTIO_MDEV_H
+#define _LINUX_VIRTIO_MDEV_H
+
+#include <linux/interrupt.h>
+#include <linux/mdev.h>
+#include <uapi/linux/vhost.h>
+
+#define VIRTIO_MDEV_DEVICE_API_STRING		"virtio-mdev"
+#define VIRTIO_MDEV_VERSION 0x1
+
+struct virtio_mdev_callback {
+	irqreturn_t (*callback)(void *data);
+	void *private;
+};
+
+/**
+ * struct vfio_mdev_device_ops - Structure to be registered for each
+ * mdev device to register the device to virtio-mdev module.
+ *
+ * @set_vq_address:		Set the address of virtqueue
+ *				@mdev: mediated device
+ *				@idx: virtqueue index
+ *				@desc_area: address of desc area
+ *				@driver_area: address of driver area
+ *				@device_area: address of device area
+ *				Returns integer: success (0) or error (< 0)
+ * @set_vq_num:		Set the size of virtqueue
+ *				@mdev: mediated device
+ *				@idx: virtqueue index
+ *				@num: the size of virtqueue
+ * @kick_vq:			Kick the virtqueue
+ *				@mdev: mediated device
+ *				@idx: virtqueue index
+ * @set_vq_cb:			Set the interrut calback function for
+ *				a virtqueue
+ *				@mdev: mediated device
+ *				@idx: virtqueue index
+ *				@cb: virtio-mdev interrupt callback structure
+ * @set_vq_ready:		Set ready status for a virtqueue
+ *				@mdev: mediated device
+ *				@idx: virtqueue index
+ *				@ready: ready (true) not ready(false)
+ * @get_vq_ready:		Get ready status for a virtqueue
+ *				@mdev: mediated device
+ *				@idx: virtqueue index
+ *				Returns boolean: ready (true) or not (false)
+ * @set_vq_state:		Set the state for a virtqueue
+ *				@mdev: mediated device
+ *				@idx: virtqueue index
+ *				@state: virtqueue state (last_avail_idx)
+ *				Returns integer: success (0) or error (< 0)
+ * @get_vq_state:		Get the state for a virtqueue
+ *				@mdev: mediated device
+ *				@idx: virtqueue index
+ *				Returns virtqueue state (last_avail_idx)
+ * @get_vq_align:		Get the virtqueue align requirement
+ *				for the device
+ *				@mdev: mediated device
+ *				Returns virtqueue algin requirement
+ * @get_features:		Get virtio features supported by the device
+ *				@mdev: mediated device
+ *				Returns the features support by the
+ *				device
+ * @get_features:		Set virtio features supported by the driver
+ *				@mdev: mediated device
+ *				@features: feature support by the driver
+ *				Returns integer: success (0) or error (< 0)
+ * @set_config_cb:		Set the config interrupt callback
+ *				@mdev: mediated device
+ *				@cb: virtio-mdev interrupt callback structure
+ * @get_device_id:		Get virtio device id
+ *				@mdev: mediated device
+ *				Returns u32: virtio device id
+ * @get_vendor_id:		Get virtio vendor id
+ *				@mdev: mediated device
+ *				Returns u32: virtio vendor id
+ * @get_status:		Get the device status
+ *				@mdev: mediated device
+ *				Returns u8: virtio device status
+ * @set_status:		Set the device status
+ *				@mdev: mediated device
+ *				@status: virtio device status
+ * @get_config:		Read from device specific confiugration space
+ *				@mdev: mediated device
+ *				@offset: offset from the beginning of
+ *				configuration space
+ *				@buf: buffer used to read to
+ *				@len: the length to read from
+ *				configration space
+ * @set_config:		Write to device specific confiugration space
+ *				@mdev: mediated device
+ *				@offset: offset from the beginning of
+ *				configuration space
+ *				@buf: buffer used to write from
+ *				@len: the length to write to
+ *				configration space
+ * @get_version:		Get the version of virtio mdev device
+ *				@mdev: mediated device
+ *				Returns integer: version of the device
+ * @get_generation:		Get device generaton
+ *				@mdev: mediated device
+ *				Returns u32: device generation
+ */
+struct virtio_mdev_device_ops {
+	/* Virtqueue ops */
+	int (*set_vq_address)(struct mdev_device *mdev,
+			      u16 idx, u64 desc_area, u64 driver_area,
+			      u64 device_area);
+	void (*set_vq_num)(struct mdev_device *mdev, u16 idx, u32 num);
+	void (*kick_vq)(struct mdev_device *mdev, u16 idx);
+	void (*set_vq_cb)(struct mdev_device *mdev, u16 idx,
+			  struct virtio_mdev_callback *cb);
+	void (*set_vq_ready)(struct mdev_device *mdev, u16 idx, bool ready);
+	bool (*get_vq_ready)(struct mdev_device *mdev, u16 idx);
+	int (*set_vq_state)(struct mdev_device *mdev, u16 idx, u64 state);
+	u64 (*get_vq_state)(struct mdev_device *mdev, u16 idx);
+
+	/* Device ops */
+	u16 (*get_vq_align)(struct mdev_device *mdev);
+	u64 (*get_features)(struct mdev_device *mdev);
+	int (*set_features)(struct mdev_device *mdev, u64 features);
+	void (*set_config_cb)(struct mdev_device *mdev,
+			      struct virtio_mdev_callback *cb);
+	u16 (*get_queue_max)(struct mdev_device *mdev);
+	u32 (*get_device_id)(struct mdev_device *mdev);
+	u32 (*get_vendor_id)(struct mdev_device *mdev);
+	u8 (*get_status)(struct mdev_device *mdev);
+	void (*set_status)(struct mdev_device *mdev, u8 status);
+	void (*get_config)(struct mdev_device *mdev, unsigned int offset,
+			   void *buf, unsigned int len);
+	void (*set_config)(struct mdev_device *mdev, unsigned int offset,
+			   const void *buf, unsigned int len);
+	int (*get_version)(struct mdev_device *mdev);
+	u32 (*get_generation)(struct mdev_device *mdev);
+};
+
+#endif
+
author
#36 lore raw

Re: [PATCH V2 6/8] mdev: introduce virtio device and its device ops

In reply to Alex Williamson
Quoted 19 lines · Alex Williamson
Quoted 16 lines · On Tue, 24 Sep 2019 21:53:30 +0800 Jason Wang <[email protected]> wrote:

This patch implements basic support for mdev driver that supports virtio transport for kernel virtio driver.

---
 include/linux/mdev.h        |   2 +
 include/linux/virtio_mdev.h | 145 ++++++++++++++++++++++++++++++++++++
 2 files changed, 147 insertions(+)
 create mode 100644 include/linux/virtio_mdev.h
include/linux/mdev.h +0 -0
index 3414307311f1..73ac27b3b868 100644
@@ -126,6 +126,8 @@ struct mdev_device *mdev_from_dev(struct device *dev);
 
Quoted 5 lines

include/linux/mdev.h struct mdev_device *mdev_from_dev(struct device *dev); · line 130

Quoted 4 lines
include/linux/mdev.h +2 -0
 enum {
 	MDEV_ID_VFIO = 1,
+	MDEV_ID_VIRTIO = 2,
+	MDEV_ID_VHOST = 3,

MDEV_ID_VHOST isn't used yet here.

Yes, just want to reserve one but it look too early to do that.

Quoted 5 lines

Also, given the strong
interdependence between the class_id and the ops structure, we might wand to define them in the same place. Thanks,

Alex

Yes, as you suggest in patch 5/8.

Thanks

Quoted 147 lines
Quoted 147 lines
include/linux/mdev.h +0 -0
 	/* New entries must be added here */
 };
 
include/linux/virtio_mdev.h new file +137 -0
new file mode 100644
index 000000000000..d1a40a739266
@@ -0,0 +1,145 @@
+/* SPDX-License-Identifier: GPL-2.0-only */
+/*
+ * Virtio mediated device driver
+ *
+ * Copyright 2019, Red Hat Corp.
+ *     Author: Jason Wang <[email protected]>
+ */
+#ifndef _LINUX_VIRTIO_MDEV_H
+#define _LINUX_VIRTIO_MDEV_H
+
+#include <linux/interrupt.h>
+#include <linux/mdev.h>
+#include <uapi/linux/vhost.h>
+
+#define VIRTIO_MDEV_DEVICE_API_STRING		"virtio-mdev"
+#define VIRTIO_MDEV_VERSION 0x1
+
+struct virtio_mdev_callback {
+	irqreturn_t (*callback)(void *data);
+	void *private;
+};
+
+/**
+ * struct vfio_mdev_device_ops - Structure to be registered for each
+ * mdev device to register the device to virtio-mdev module.
+ *
+ * @set_vq_address:		Set the address of virtqueue
+ *				@mdev: mediated device
+ *				@idx: virtqueue index
+ *				@desc_area: address of desc area
+ *				@driver_area: address of driver area
+ *				@device_area: address of device area
+ *				Returns integer: success (0) or error (< 0)
+ * @set_vq_num:		Set the size of virtqueue
+ *				@mdev: mediated device
+ *				@idx: virtqueue index
+ *				@num: the size of virtqueue
+ * @kick_vq:			Kick the virtqueue
+ *				@mdev: mediated device
+ *				@idx: virtqueue index
+ * @set_vq_cb:			Set the interrut calback function for
+ *				a virtqueue
+ *				@mdev: mediated device
+ *				@idx: virtqueue index
+ *				@cb: virtio-mdev interrupt callback structure
+ * @set_vq_ready:		Set ready status for a virtqueue
+ *				@mdev: mediated device
+ *				@idx: virtqueue index
+ *				@ready: ready (true) not ready(false)
+ * @get_vq_ready:		Get ready status for a virtqueue
+ *				@mdev: mediated device
+ *				@idx: virtqueue index
+ *				Returns boolean: ready (true) or not (false)
+ * @set_vq_state:		Set the state for a virtqueue
+ *				@mdev: mediated device
+ *				@idx: virtqueue index
+ *				@state: virtqueue state (last_avail_idx)
+ *				Returns integer: success (0) or error (< 0)
+ * @get_vq_state:		Get the state for a virtqueue
+ *				@mdev: mediated device
+ *				@idx: virtqueue index
+ *				Returns virtqueue state (last_avail_idx)
+ * @get_vq_align:		Get the virtqueue align requirement
+ *				for the device
+ *				@mdev: mediated device
+ *				Returns virtqueue algin requirement
+ * @get_features:		Get virtio features supported by the device
+ *				@mdev: mediated device
+ *				Returns the features support by the
+ *				device
+ * @get_features:		Set virtio features supported by the driver
+ *				@mdev: mediated device
+ *				@features: feature support by the driver
+ *				Returns integer: success (0) or error (< 0)
+ * @set_config_cb:		Set the config interrupt callback
+ *				@mdev: mediated device
+ *				@cb: virtio-mdev interrupt callback structure
+ * @get_device_id:		Get virtio device id
+ *				@mdev: mediated device
+ *				Returns u32: virtio device id
+ * @get_vendor_id:		Get virtio vendor id
+ *				@mdev: mediated device
+ *				Returns u32: virtio vendor id
+ * @get_status:		Get the device status
+ *				@mdev: mediated device
+ *				Returns u8: virtio device status
+ * @set_status:		Set the device status
+ *				@mdev: mediated device
+ *				@status: virtio device status
+ * @get_config:		Read from device specific confiugration space
+ *				@mdev: mediated device
+ *				@offset: offset from the beginning of
+ *				configuration space
+ *				@buf: buffer used to read to
+ *				@len: the length to read from
+ *				configration space
+ * @set_config:		Write to device specific confiugration space
+ *				@mdev: mediated device
+ *				@offset: offset from the beginning of
+ *				configuration space
+ *				@buf: buffer used to write from
+ *				@len: the length to write to
+ *				configration space
+ * @get_version:		Get the version of virtio mdev device
+ *				@mdev: mediated device
+ *				Returns integer: version of the device
+ * @get_generation:		Get device generaton
+ *				@mdev: mediated device
+ *				Returns u32: device generation
+ */
+struct virtio_mdev_device_ops {
+	/* Virtqueue ops */
+	int (*set_vq_address)(struct mdev_device *mdev,
+			      u16 idx, u64 desc_area, u64 driver_area,
+			      u64 device_area);
+	void (*set_vq_num)(struct mdev_device *mdev, u16 idx, u32 num);
+	void (*kick_vq)(struct mdev_device *mdev, u16 idx);
+	void (*set_vq_cb)(struct mdev_device *mdev, u16 idx,
+			  struct virtio_mdev_callback *cb);
+	void (*set_vq_ready)(struct mdev_device *mdev, u16 idx, bool ready);
+	bool (*get_vq_ready)(struct mdev_device *mdev, u16 idx);
+	int (*set_vq_state)(struct mdev_device *mdev, u16 idx, u64 state);
+	u64 (*get_vq_state)(struct mdev_device *mdev, u16 idx);
+
+	/* Device ops */
+	u16 (*get_vq_align)(struct mdev_device *mdev);
+	u64 (*get_features)(struct mdev_device *mdev);
+	int (*set_features)(struct mdev_device *mdev, u64 features);
+	void (*set_config_cb)(struct mdev_device *mdev,
+			      struct virtio_mdev_callback *cb);
+	u16 (*get_queue_max)(struct mdev_device *mdev);
+	u32 (*get_device_id)(struct mdev_device *mdev);
+	u32 (*get_vendor_id)(struct mdev_device *mdev);
+	u8 (*get_status)(struct mdev_device *mdev);
+	void (*set_status)(struct mdev_device *mdev, u8 status);
+	void (*get_config)(struct mdev_device *mdev, unsigned int offset,
+			   void *buf, unsigned int len);
Quoted 8 lines

include/linux/virtio_mdev.h line 145

Quoted 8 lines
include/linux/virtio_mdev.h new file +8 -0
+	void (*set_config)(struct mdev_device *mdev, unsigned int offset,
+			   const void *buf, unsigned int len);
+	int (*get_version)(struct mdev_device *mdev);
+	u32 (*get_generation)(struct mdev_device *mdev);
+};
+
+#endif
+

_______________________________________________
Virtualization mailing list
[email protected]
https://lists.linuxfoundation.org/mailman/listinfo/virtualization

author
#37 lore raw

Re: [PATCH V2 6/8] mdev: introduce virtio device and its device ops

In reply to Alex Williamson
Quoted 27 lines · Alex Williamson
Quoted 15 lines · On Tue, 24 Sep 2019 21:53:30 +0800 Jason Wang<[email protected]> wrote:

This patch implements basic support for mdev driver that supports virtio transport for kernel virtio driver.

---
  include/linux/mdev.h        |   2 +
  include/linux/virtio_mdev.h | 145 ++++++++++++++++++++++++++++++++++++
  2 files changed, 147 insertions(+)

create mode 100644 include/linux/virtio_mdev.h

include/linux/mdev.h +0 -0
index 3414307311f1..73ac27b3b868 100644

include/linux/mdev.h struct mdev_device *mdev_from_dev(struct device *dev); · line 130

Quoted 5 lines
include/linux/mdev.h +2 -0
@@ -126,6 +126,8 @@ struct mdev_device *mdev_from_dev(struct device *dev);
 
  enum {
  	MDEV_ID_VFIO = 1,
+	MDEV_ID_VIRTIO = 2,
+	MDEV_ID_VHOST = 3,

MDEV_ID_VHOST isn't used yet here. Also, given the strong interdependence between the class_id and the ops structure, we might wand to define them in the same place. Thanks,

Quoted 2 lines

Alex

Rethink about this,  I believe it's better to define the device ops in their own headers, and one set of device ops could be used for two classes (e.g both virtio and vhost). And to avoid a duplicated ID definition. I tend to keep this in the common mdev.h header.

Thanks

author
#38 lore raw

Re: [PATCH V2 6/8] mdev: introduce virtio device and its device ops

In reply to Alex Williamson
Quoted 27 lines · Alex Williamson
Quoted 15 lines · On Tue, 24 Sep 2019 21:53:30 +0800 Jason Wang<[email protected]> wrote:

This patch implements basic support for mdev driver that supports virtio transport for kernel virtio driver.

---
  include/linux/mdev.h        |   2 +
  include/linux/virtio_mdev.h | 145 ++++++++++++++++++++++++++++++++++++
  2 files changed, 147 insertions(+)

create mode 100644 include/linux/virtio_mdev.h

include/linux/mdev.h +0 -0
index 3414307311f1..73ac27b3b868 100644

include/linux/mdev.h struct mdev_device *mdev_from_dev(struct device *dev); · line 130

Quoted 5 lines
include/linux/mdev.h +2 -0
@@ -126,6 +126,8 @@ struct mdev_device *mdev_from_dev(struct device *dev);
 
  enum {
  	MDEV_ID_VFIO = 1,
+	MDEV_ID_VIRTIO = 2,
+	MDEV_ID_VHOST = 3,

MDEV_ID_VHOST isn't used yet here. Also, given the strong interdependence between the class_id and the ops structure, we might wand to define them in the same place. Thanks,

Quoted 2 lines

Alex

Rethink about this,  I believe it's better to define the device ops in their own headers, and one set of device ops could be used for two classes (e.g both virtio and vhost). And to avoid a duplicated ID definition. I tend to keep this in the common mdev.h header.

Thanks

_______________________________________________
Virtualization mailing list
[email protected]
https://lists.linuxfoundation.org/mailman/listinfo/virtualization

#39 lore raw

Re: [PATCH V2 6/8] mdev: introduce virtio device and its device ops

In reply to Alex Williamson

Hi Alex,

Quoted 45 lines

-----Original Message-----
From: Alex Williamson <[email protected]>
Sent: Tuesday, September 24, 2019 6:07 PM
To: Jason Wang <[email protected]>
Cc: [email protected]; [email protected]; linux- [email protected]; [email protected]; intel- [email protected]; [email protected]; [email protected]; [email protected]; [email protected]; [email protected]; [email protected]; [email protected]; [email protected]; [email protected]; [email protected]; [email protected]; [email protected]; [email protected]; [email protected]; [email protected]; [email protected]; [email protected]; [email protected]; [email protected]; [email protected]; [email protected]; [email protected]; [email protected]; [email protected]; [email protected]; [email protected]; [email protected]; [email protected]; [email protected]; [email protected]; Ido Shamay <[email protected]>; [email protected]; [email protected]; Parav Pandit <[email protected]>; [email protected]; [email protected]
Subject: Re: [PATCH V2 6/8] mdev: introduce virtio device and its device ops

Quoted 18 lines · On Tue, 24 Sep 2019 21:53:30 +0800 Jason Wang <[email protected]> wrote:

This patch implements basic support for mdev driver that supports virtio transport for kernel virtio driver.

---
 include/linux/mdev.h        |   2 +
 include/linux/virtio_mdev.h | 145
++++++++++++++++++++++++++++++++++++
 2 files changed, 147 insertions(+)
 create mode 100644 include/linux/virtio_mdev.h
include/linux/mdev.h index renamed from include/linux/mdev.h +0 -0

3414307311f1..73ac27b3b868 100644

include/linux/mdev.h +0 -0
@@ -126,6 +126,8 @@ struct mdev_device *mdev_from_dev(struct device
 *dev);
 
Quoted 9 lines

include/linux/mdev.h struct mdev_device *mdev_from_dev(struct device *dev); · line 130

Quoted 4 lines
include/linux/mdev.h +2 -0
 enum {
 	MDEV_ID_VFIO = 1,
+	MDEV_ID_VIRTIO = 2,
+	MDEV_ID_VHOST = 3,

MDEV_ID_VHOST isn't used yet here. Also, given the strong interdependence between the class_id and the ops structure, we might wand to define them in the same place. Thanks,

When mlx5_core creates mdevs (parent->ops->create() and it wants to bind to mlx5 mdev driver (which does mdev_register_driver()), mlx5 core driver will publish MDEV_ID_MLX5_NET defined in central place as include/linux/mdev.h without any ops structure. Because such ops are not relevant. It uses usual, standard ops probe() remove() on the mdev (similar to a regular PCI device). So for VHOST case ops may be closely related to ID, but not for other type of ID.

Just want to make sure, that scope of ID covers this case.
_______________________________________________
Intel-gfx mailing list [email protected]
https://lists.freedesktop.org/mailman/listinfo/intel-gfx

#40 lore raw

Re: [PATCH V2 6/8] mdev: introduce virtio device and its device ops

Quoted 66 lines · On Fri, 27 Sep 2019 16:25:13 +0000 Parav Pandit <[email protected]> wrote:

Hi Alex,

Quoted 45 lines

-----Original Message-----
From: Alex Williamson <[email protected]>
Sent: Tuesday, September 24, 2019 6:07 PM
To: Jason Wang <[email protected]>
Cc: [email protected]; [email protected]; linux- [email protected]; [email protected]; intel- [email protected]; [email protected]; [email protected]; [email protected]; [email protected]; [email protected]; [email protected]; [email protected]; [email protected]; [email protected]; [email protected]; [email protected]; [email protected]; [email protected]; [email protected]; [email protected]; [email protected]; [email protected]; [email protected]; [email protected]; [email protected]; [email protected]; [email protected]; [email protected]; [email protected]; [email protected]; [email protected]; [email protected]; [email protected]; [email protected]; [email protected]; Ido Shamay <[email protected]>; [email protected]; [email protected]; Parav Pandit <[email protected]>; [email protected]; [email protected]
Subject: Re: [PATCH V2 6/8] mdev: introduce virtio device and its device ops

Quoted 18 lines · On Tue, 24 Sep 2019 21:53:30 +0800 Jason Wang <[email protected]> wrote:

This patch implements basic support for mdev driver that supports virtio transport for kernel virtio driver.

---
 include/linux/mdev.h        |   2 +
 include/linux/virtio_mdev.h | 145
++++++++++++++++++++++++++++++++++++
 2 files changed, 147 insertions(+)
 create mode 100644 include/linux/virtio_mdev.h
include/linux/mdev.h index renamed from include/linux/mdev.h +0 -0

3414307311f1..73ac27b3b868 100644

include/linux/mdev.h +0 -0
@@ -126,6 +126,8 @@ struct mdev_device *mdev_from_dev(struct device
 *dev);
 
Quoted 9 lines

include/linux/mdev.h struct mdev_device *mdev_from_dev(struct device *dev); · line 130

Quoted 4 lines
include/linux/mdev.h +2 -0
 enum {
 	MDEV_ID_VFIO = 1,
+	MDEV_ID_VIRTIO = 2,
+	MDEV_ID_VHOST = 3,  

MDEV_ID_VHOST isn't used yet here. Also, given the strong interdependence between the class_id and the ops structure, we might wand to define them in the same place. Thanks,

When mlx5_core creates mdevs (parent->ops->create() and it wants to bind to mlx5 mdev driver (which does mdev_register_driver()), mlx5 core driver will publish MDEV_ID_MLX5_NET defined in central place as include/linux/mdev.h without any ops structure. Because such ops are not relevant. It uses usual, standard ops probe() remove() on the mdev (similar to a regular PCI device). So for VHOST case ops may be closely related to ID, but not for other type of ID.

Quoted 1 line

Just want to make sure, that scope of ID covers this case.

AIUI, these device-ops are primarily meant to have 1:N multiplexing of the mdev bus driver. One mdev bus driver supports N vendor drivers via a common "protocol" defined by this structure. vfio-mdev supports GVT-g, GRID, and several sample drivers. I think Jason and Tiwei are attempting something similar if we have multiple vendors that may provide virtio/vhost parent drivers. If you have a 1:1 model with mlx5 where you're not trying to abstract a common channel between the mdev bus driver and the mdev vendor driver, then I suppose you might not use the device-ops capabilities of the mdev-core. Did I interpret the question correctly? I think that's probably fine, mdev-core shouldn't have any dependencies on the device-ops and we shouldn't really be dictating the bus/vendor link through mdev. Thanks,

Alex
_______________________________________________
Intel-gfx mailing list
[email protected]
https://lists.freedesktop.org/mailman/listinfo/intel-gfx

author
#41 lore raw

Re: [PATCH V2 6/8] mdev: introduce virtio device and its device ops

Quoted 67 lines · Alex Williamson
Quoted 64 lines · On Fri, 27 Sep 2019 16:25:13 +0000 Parav Pandit <[email protected]> wrote:

Hi Alex,

Quoted 45 lines

-----Original Message-----
From: Alex Williamson <[email protected]>
Sent: Tuesday, September 24, 2019 6:07 PM
To: Jason Wang <[email protected]>
Cc: [email protected]; [email protected]; linux- [email protected]; [email protected]; intel- [email protected]; [email protected]; [email protected]; [email protected]; [email protected]; [email protected]; [email protected]; [email protected]; [email protected]; [email protected]; [email protected]; [email protected]; [email protected]; [email protected]; [email protected]; [email protected]; [email protected]; [email protected]; [email protected]; [email protected]; [email protected]; [email protected]; [email protected]; [email protected]; [email protected]; [email protected]; [email protected]; [email protected]; [email protected]; [email protected]; [email protected]; Ido Shamay <[email protected]>; [email protected]; [email protected]; Parav Pandit <[email protected]>; [email protected]; [email protected]
Subject: Re: [PATCH V2 6/8] mdev: introduce virtio device and its device ops

Quoted 18 lines · On Tue, 24 Sep 2019 21:53:30 +0800 Jason Wang <[email protected]> wrote:

This patch implements basic support for mdev driver that supports virtio transport for kernel virtio driver.

---
  include/linux/mdev.h        |   2 +
  include/linux/virtio_mdev.h | 145
++++++++++++++++++++++++++++++++++++
  2 files changed, 147 insertions(+)

create mode 100644 include/linux/virtio_mdev.h

include/linux/mdev.h index renamed from include/linux/mdev.h +0 -0

3414307311f1..73ac27b3b868 100644

include/linux/mdev.h +0 -0
@@ -126,6 +126,8 @@ struct mdev_device *mdev_from_dev(struct device
 *dev);
 
Quoted 8 lines

include/linux/mdev.h struct mdev_device *mdev_from_dev(struct device *dev); · line 130

Quoted 4 lines
include/linux/mdev.h +2 -0
  enum {
  	MDEV_ID_VFIO = 1,
+	MDEV_ID_VIRTIO = 2,
+	MDEV_ID_VHOST = 3,

MDEV_ID_VHOST isn't used yet here. Also, given the strong interdependence between the class_id and the ops structure, we might wand to define them in the same place. Thanks,

When mlx5_core creates mdevs (parent->ops->create() and it wants to bind to mlx5 mdev driver (which does mdev_register_driver()), mlx5 core driver will publish MDEV_ID_MLX5_NET defined in central place as include/linux/mdev.h without any ops structure. Because such ops are not relevant. It uses usual, standard ops probe() remove() on the mdev (similar to a regular PCI device). So for VHOST case ops may be closely related to ID, but not for other type of ID.

Quoted 7 lines
Quoted 1 line

Just want to make sure, that scope of ID covers this case.

AIUI, these device-ops are primarily meant to have 1:N multiplexing of the mdev bus driver. One mdev bus driver supports N vendor drivers via a common "protocol" defined by this structure. vfio-mdev supports GVT-g, GRID, and several sample drivers. I think Jason and Tiwei are attempting something similar if we have multiple vendors that may provide virtio/vhost parent drivers.

Exactly.

Quoted 4 lines

If you have a 1:1 model with
mlx5 where you're not trying to abstract a common channel between the mdev bus driver and the mdev vendor driver, then I suppose you might not use the device-ops capabilities of the mdev-core.

Yes, current proposed API allows NULL to be passed as device ops.

Thanks

Quoted 6 lines

Did I interpret
the question correctly? I think that's probably fine, mdev-core shouldn't have any dependencies on the device-ops and we shouldn't really be dictating the bus/vendor link through mdev. Thanks,

Alex

_______________________________________________
Virtualization mailing list
[email protected]
https://lists.linuxfoundation.org/mailman/listinfo/virtualization

#42 lore raw

Re: [PATCH V2 6/8] mdev: introduce virtio device and its device ops

In reply to Parav Pandit
Quoted 66 lines · On Fri, 27 Sep 2019 16:25:13 +0000 Parav Pandit <[email protected]> wrote:

Hi Alex,

Quoted 45 lines

-----Original Message-----
From: Alex Williamson <[email protected]>
Sent: Tuesday, September 24, 2019 6:07 PM
To: Jason Wang <[email protected]>
Cc: [email protected]; [email protected]; linux- [email protected]; [email protected]; intel- [email protected]; [email protected]; [email protected]; [email protected]; [email protected]; [email protected]; [email protected]; [email protected]; [email protected]; [email protected]; [email protected]; [email protected]; [email protected]; [email protected]; [email protected]; [email protected]; [email protected]; [email protected]; [email protected]; [email protected]; [email protected]; [email protected]; [email protected]; [email protected]; [email protected]; [email protected]; [email protected]; [email protected]; [email protected]; [email protected]; [email protected]; Ido Shamay <[email protected]>; [email protected]; [email protected]; Parav Pandit <[email protected]>; [email protected]; [email protected]
Subject: Re: [PATCH V2 6/8] mdev: introduce virtio device and its device ops

Quoted 18 lines · On Tue, 24 Sep 2019 21:53:30 +0800 Jason Wang <[email protected]> wrote:

This patch implements basic support for mdev driver that supports virtio transport for kernel virtio driver.

---
 include/linux/mdev.h        |   2 +
 include/linux/virtio_mdev.h | 145
++++++++++++++++++++++++++++++++++++
 2 files changed, 147 insertions(+)
 create mode 100644 include/linux/virtio_mdev.h
include/linux/mdev.h index renamed from include/linux/mdev.h +0 -0

3414307311f1..73ac27b3b868 100644

include/linux/mdev.h +0 -0
@@ -126,6 +126,8 @@ struct mdev_device *mdev_from_dev(struct device
 *dev);
 
Quoted 9 lines

include/linux/mdev.h struct mdev_device *mdev_from_dev(struct device *dev); · line 130

Quoted 4 lines
include/linux/mdev.h +2 -0
 enum {
 	MDEV_ID_VFIO = 1,
+	MDEV_ID_VIRTIO = 2,
+	MDEV_ID_VHOST = 3,  

MDEV_ID_VHOST isn't used yet here. Also, given the strong interdependence between the class_id and the ops structure, we might wand to define them in the same place. Thanks,

When mlx5_core creates mdevs (parent->ops->create() and it wants to bind to mlx5 mdev driver (which does mdev_register_driver()), mlx5 core driver will publish MDEV_ID_MLX5_NET defined in central place as include/linux/mdev.h without any ops structure. Because such ops are not relevant. It uses usual, standard ops probe() remove() on the mdev (similar to a regular PCI device). So for VHOST case ops may be closely related to ID, but not for other type of ID.

Quoted 1 line

Just want to make sure, that scope of ID covers this case.

AIUI, these device-ops are primarily meant to have 1:N multiplexing of the mdev bus driver. One mdev bus driver supports N vendor drivers via a common "protocol" defined by this structure. vfio-mdev supports GVT-g, GRID, and several sample drivers. I think Jason and Tiwei are attempting something similar if we have multiple vendors that may provide virtio/vhost parent drivers. If you have a 1:1 model with mlx5 where you're not trying to abstract a common channel between the mdev bus driver and the mdev vendor driver, then I suppose you might not use the device-ops capabilities of the mdev-core. Did I interpret the question correctly? I think that's probably fine, mdev-core shouldn't have any dependencies on the device-ops and we shouldn't really be dictating the bus/vendor link through mdev. Thanks,

Alex

author
#43 lore raw

Re: [PATCH V2 6/8] mdev: introduce virtio device and its device ops

In reply to Alex Williamson
Quoted 27 lines · Alex Williamson
Quoted 15 lines · On Tue, 24 Sep 2019 21:53:30 +0800 Jason Wang<[email protected]> wrote:

This patch implements basic support for mdev driver that supports virtio transport for kernel virtio driver.

---
  include/linux/mdev.h        |   2 +
  include/linux/virtio_mdev.h | 145 ++++++++++++++++++++++++++++++++++++
  2 files changed, 147 insertions(+)

create mode 100644 include/linux/virtio_mdev.h

include/linux/mdev.h +0 -0
index 3414307311f1..73ac27b3b868 100644

include/linux/mdev.h struct mdev_device *mdev_from_dev(struct device *dev); · line 130

Quoted 5 lines
include/linux/mdev.h +2 -0
@@ -126,6 +126,8 @@ struct mdev_device *mdev_from_dev(struct device *dev);
 
  enum {
  	MDEV_ID_VFIO = 1,
+	MDEV_ID_VIRTIO = 2,
+	MDEV_ID_VHOST = 3,

MDEV_ID_VHOST isn't used yet here. Also, given the strong interdependence between the class_id and the ops structure, we might wand to define them in the same place. Thanks,

Quoted 2 lines

Alex

Rethink about this, consider we may have more types of devices supported in the future, moving all device_ops to mdev.h seems unnecessary. I prefer to keep the device_ops into their own headers.

Thanks

author
#44 lore raw

Re: [PATCH V2 6/8] mdev: introduce virtio device and its device ops

In reply to Alex Williamson
Quoted 27 lines · Alex Williamson
Quoted 15 lines · On Tue, 24 Sep 2019 21:53:30 +0800 Jason Wang<[email protected]> wrote:

This patch implements basic support for mdev driver that supports virtio transport for kernel virtio driver.

---
  include/linux/mdev.h        |   2 +
  include/linux/virtio_mdev.h | 145 ++++++++++++++++++++++++++++++++++++
  2 files changed, 147 insertions(+)

create mode 100644 include/linux/virtio_mdev.h

include/linux/mdev.h +0 -0
index 3414307311f1..73ac27b3b868 100644

include/linux/mdev.h struct mdev_device *mdev_from_dev(struct device *dev); · line 130

Quoted 5 lines
include/linux/mdev.h +2 -0
@@ -126,6 +126,8 @@ struct mdev_device *mdev_from_dev(struct device *dev);
 
  enum {
  	MDEV_ID_VFIO = 1,
+	MDEV_ID_VIRTIO = 2,
+	MDEV_ID_VHOST = 3,

MDEV_ID_VHOST isn't used yet here. Also, given the strong interdependence between the class_id and the ops structure, we might wand to define them in the same place. Thanks,

Quoted 2 lines

Alex

Rethink about this, consider we may have more types of devices supported in the future, moving all device_ops to mdev.h seems unnecessary. I prefer to keep the device_ops into their own headers.

Thanks

_______________________________________________
Virtualization mailing list
[email protected]
https://lists.linuxfoundation.org/mailman/listinfo/virtualization

#45 lore raw

Re: [PATCH V2 6/8] mdev: introduce virtio device and its device ops

In reply to Jason Wang
Quoted 15 lines · On Tue, 24 Sep 2019 21:53:30 +0800 Jason Wang <[email protected]> wrote:

This patch implements basic support for mdev driver that supports virtio transport for kernel virtio driver.

---
 include/linux/mdev.h        |   2 +
 include/linux/virtio_mdev.h | 145 ++++++++++++++++++++++++++++++++++++
 2 files changed, 147 insertions(+)
 create mode 100644 include/linux/virtio_mdev.h
include/linux/mdev.h +0 -0
index 3414307311f1..73ac27b3b868 100644

include/linux/mdev.h struct mdev_device *mdev_from_dev(struct device *dev); · line 130

Quoted 5 lines
include/linux/mdev.h +2 -0
@@ -126,6 +126,8 @@ struct mdev_device *mdev_from_dev(struct device *dev);
 
 enum {
 	MDEV_ID_VFIO = 1,
+	MDEV_ID_VIRTIO = 2,
+	MDEV_ID_VHOST = 3,

MDEV_ID_VHOST isn't used yet here. Also, given the strong interdependence between the class_id and the ops structure, we might wand to define them in the same place. Thanks,

Alex

Quoted 154 lines
include/linux/mdev.h +0 -0
 	/* New entries must be added here */
 };
 
include/linux/virtio_mdev.h new file +145 -0
new file mode 100644
index 000000000000..d1a40a739266
@@ -0,0 +1,145 @@
+/* SPDX-License-Identifier: GPL-2.0-only */
+/*
+ * Virtio mediated device driver
+ *
+ * Copyright 2019, Red Hat Corp.
+ *     Author: Jason Wang <[email protected]>
+ */
+#ifndef _LINUX_VIRTIO_MDEV_H
+#define _LINUX_VIRTIO_MDEV_H
+
+#include <linux/interrupt.h>
+#include <linux/mdev.h>
+#include <uapi/linux/vhost.h>
+
+#define VIRTIO_MDEV_DEVICE_API_STRING		"virtio-mdev"
+#define VIRTIO_MDEV_VERSION 0x1
+
+struct virtio_mdev_callback {
+	irqreturn_t (*callback)(void *data);
+	void *private;
+};
+
+/**
+ * struct vfio_mdev_device_ops - Structure to be registered for each
+ * mdev device to register the device to virtio-mdev module.
+ *
+ * @set_vq_address:		Set the address of virtqueue
+ *				@mdev: mediated device
+ *				@idx: virtqueue index
+ *				@desc_area: address of desc area
+ *				@driver_area: address of driver area
+ *				@device_area: address of device area
+ *				Returns integer: success (0) or error (< 0)
+ * @set_vq_num:		Set the size of virtqueue
+ *				@mdev: mediated device
+ *				@idx: virtqueue index
+ *				@num: the size of virtqueue
+ * @kick_vq:			Kick the virtqueue
+ *				@mdev: mediated device
+ *				@idx: virtqueue index
+ * @set_vq_cb:			Set the interrut calback function for
+ *				a virtqueue
+ *				@mdev: mediated device
+ *				@idx: virtqueue index
+ *				@cb: virtio-mdev interrupt callback structure
+ * @set_vq_ready:		Set ready status for a virtqueue
+ *				@mdev: mediated device
+ *				@idx: virtqueue index
+ *				@ready: ready (true) not ready(false)
+ * @get_vq_ready:		Get ready status for a virtqueue
+ *				@mdev: mediated device
+ *				@idx: virtqueue index
+ *				Returns boolean: ready (true) or not (false)
+ * @set_vq_state:		Set the state for a virtqueue
+ *				@mdev: mediated device
+ *				@idx: virtqueue index
+ *				@state: virtqueue state (last_avail_idx)
+ *				Returns integer: success (0) or error (< 0)
+ * @get_vq_state:		Get the state for a virtqueue
+ *				@mdev: mediated device
+ *				@idx: virtqueue index
+ *				Returns virtqueue state (last_avail_idx)
+ * @get_vq_align:		Get the virtqueue align requirement
+ *				for the device
+ *				@mdev: mediated device
+ *				Returns virtqueue algin requirement
+ * @get_features:		Get virtio features supported by the device
+ *				@mdev: mediated device
+ *				Returns the features support by the
+ *				device
+ * @get_features:		Set virtio features supported by the driver
+ *				@mdev: mediated device
+ *				@features: feature support by the driver
+ *				Returns integer: success (0) or error (< 0)
+ * @set_config_cb:		Set the config interrupt callback
+ *				@mdev: mediated device
+ *				@cb: virtio-mdev interrupt callback structure
+ * @get_device_id:		Get virtio device id
+ *				@mdev: mediated device
+ *				Returns u32: virtio device id
+ * @get_vendor_id:		Get virtio vendor id
+ *				@mdev: mediated device
+ *				Returns u32: virtio vendor id
+ * @get_status:		Get the device status
+ *				@mdev: mediated device
+ *				Returns u8: virtio device status
+ * @set_status:		Set the device status
+ *				@mdev: mediated device
+ *				@status: virtio device status
+ * @get_config:		Read from device specific confiugration space
+ *				@mdev: mediated device
+ *				@offset: offset from the beginning of
+ *				configuration space
+ *				@buf: buffer used to read to
+ *				@len: the length to read from
+ *				configration space
+ * @set_config:		Write to device specific confiugration space
+ *				@mdev: mediated device
+ *				@offset: offset from the beginning of
+ *				configuration space
+ *				@buf: buffer used to write from
+ *				@len: the length to write to
+ *				configration space
+ * @get_version:		Get the version of virtio mdev device
+ *				@mdev: mediated device
+ *				Returns integer: version of the device
+ * @get_generation:		Get device generaton
+ *				@mdev: mediated device
+ *				Returns u32: device generation
+ */
+struct virtio_mdev_device_ops {
+	/* Virtqueue ops */
+	int (*set_vq_address)(struct mdev_device *mdev,
+			      u16 idx, u64 desc_area, u64 driver_area,
+			      u64 device_area);
+	void (*set_vq_num)(struct mdev_device *mdev, u16 idx, u32 num);
+	void (*kick_vq)(struct mdev_device *mdev, u16 idx);
+	void (*set_vq_cb)(struct mdev_device *mdev, u16 idx,
+			  struct virtio_mdev_callback *cb);
+	void (*set_vq_ready)(struct mdev_device *mdev, u16 idx, bool ready);
+	bool (*get_vq_ready)(struct mdev_device *mdev, u16 idx);
+	int (*set_vq_state)(struct mdev_device *mdev, u16 idx, u64 state);
+	u64 (*get_vq_state)(struct mdev_device *mdev, u16 idx);
+
+	/* Device ops */
+	u16 (*get_vq_align)(struct mdev_device *mdev);
+	u64 (*get_features)(struct mdev_device *mdev);
+	int (*set_features)(struct mdev_device *mdev, u64 features);
+	void (*set_config_cb)(struct mdev_device *mdev,
+			      struct virtio_mdev_callback *cb);
+	u16 (*get_queue_max)(struct mdev_device *mdev);
+	u32 (*get_device_id)(struct mdev_device *mdev);
+	u32 (*get_vendor_id)(struct mdev_device *mdev);
+	u8 (*get_status)(struct mdev_device *mdev);
+	void (*set_status)(struct mdev_device *mdev, u8 status);
+	void (*get_config)(struct mdev_device *mdev, unsigned int offset,
+			   void *buf, unsigned int len);
+	void (*set_config)(struct mdev_device *mdev, unsigned int offset,
+			   const void *buf, unsigned int len);
+	int (*get_version)(struct mdev_device *mdev);
+	u32 (*get_generation)(struct mdev_device *mdev);
+};
+
+#endif
+
#46 lore raw

RE: [PATCH V2 6/8] mdev: introduce virtio device and its device ops

In reply to Jason Wang
Quoted 40 lines

From: Jason Wang [mailto:[email protected]]
Sent: Tuesday, September 24, 2019 9:54 PM

This patch implements basic support for mdev driver that supports virtio transport for kernel virtio driver.

---
 include/linux/mdev.h        |   2 +
 include/linux/virtio_mdev.h | 145
++++++++++++++++++++++++++++++++++++
 2 files changed, 147 insertions(+)
 create mode 100644 include/linux/virtio_mdev.h
include/linux/mdev.h +2 -0
index 3414307311f1..73ac27b3b868 100644
@@ -126,6 +126,8 @@ struct mdev_device *mdev_from_dev(struct device
 *dev);
 
 enum {
 	MDEV_ID_VFIO = 1,
+	MDEV_ID_VIRTIO = 2,
+	MDEV_ID_VHOST = 3,
 	/* New entries must be added here */
 };
include/linux/virtio_mdev.h new file +6 -0
new file mode 100644
index 000000000000..d1a40a739266
@@ -0,0 +1,145 @@
+/* SPDX-License-Identifier: GPL-2.0-only */
+/*
+ * Virtio mediated device driver
+ *
+ * Copyright 2019, Red Hat Corp.
+ *     Author: Jason Wang <[email protected]>

include/linux/virtio_mdev.h line 16

Quoted 10 lines
include/linux/virtio_mdev.h new file +10 -0
+ */
+#ifndef _LINUX_VIRTIO_MDEV_H
+#define _LINUX_VIRTIO_MDEV_H
+
+#include <linux/interrupt.h>
+#include <linux/mdev.h>
+#include <uapi/linux/vhost.h>
+
+#define VIRTIO_MDEV_DEVICE_API_STRING		"virtio-mdev"
+#define VIRTIO_MDEV_VERSION 0x1

Just be curious. is this version identical to virtio spec version that below callbacks are created for, or just irrelevant?

Quoted 69 lines
include/linux/virtio_mdev.h new file +29 -0
+
+struct virtio_mdev_callback {
+	irqreturn_t (*callback)(void *data);
+	void *private;
+};
+
+/**
+ * struct vfio_mdev_device_ops - Structure to be registered for each
+ * mdev device to register the device to virtio-mdev module.
+ *
+ * @set_vq_address:		Set the address of virtqueue
+ *				@mdev: mediated device
+ *				@idx: virtqueue index
+ *				@desc_area: address of desc area
+ *				@driver_area: address of driver area
+ *				@device_area: address of device area
+ *				Returns integer: success (0) or error (< 0)
+ * @set_vq_num:		Set the size of virtqueue
+ *				@mdev: mediated device
+ *				@idx: virtqueue index
+ *				@num: the size of virtqueue
+ * @kick_vq:			Kick the virtqueue
+ *				@mdev: mediated device
+ *				@idx: virtqueue index
+ * @set_vq_cb:			Set the interrut calback function for
+ *				a virtqueue
+ *				@mdev: mediated device
+ *				@idx: virtqueue index
+ *				@cb: virtio-mdev interrupt callback

structure
+ * @set_vq_ready: Set ready status for a virtqueue
+ * @mdev: mediated device
+ * @idx: virtqueue index
+ * @ready: ready (true) not ready(false)
+ * @get_vq_ready: Get ready status for a virtqueue
+ * @mdev: mediated device
+ * @idx: virtqueue index
+ * Returns boolean: ready (true) or not (false)
+ * @set_vq_state: Set the state for a virtqueue
+ * @mdev: mediated device
+ * @idx: virtqueue index
+ * @state: virtqueue state (last_avail_idx)
+ * Returns integer: success (0) or error (< 0)
+ * @get_vq_state: Get the state for a virtqueue
+ * @mdev: mediated device
+ * @idx: virtqueue index
+ * Returns virtqueue state (last_avail_idx)
+ * @get_vq_align: Get the virtqueue align requirement
+ * for the device
+ * @mdev: mediated device
+ * Returns virtqueue algin requirement
+ * @get_features: Get virtio features supported by the device
+ * @mdev: mediated device
+ * Returns the features support by the
+ * device
+ * @get_features: Set virtio features supported by the driver
+ * @mdev: mediated device
+ * @features: feature support by the driver
+ * Returns integer: success (0) or error (< 0)
+ * @set_config_cb: Set the config interrupt callback
+ * @mdev: mediated device
+ * @cb: virtio-mdev interrupt callback
structure
+ * @get_device_id: Get virtio device id
+ * @mdev: mediated device
+ * Returns u32: virtio device id
+ * @get_vendor_id: Get virtio vendor id
+ * @mdev: mediated device
+ * Returns u32: virtio vendor id

Quoted 8 lines

+ * @get_status: Get the device status
+ * @mdev: mediated device
+ * Returns u8: virtio device status
+ * @set_status: Set the device status
+ * @mdev: mediated device
+ * @status: virtio device status
+ * @get_config: Read from device specific confiugration
space

configuration (and similar typos downward)

Quoted 43 lines

+ * @mdev: mediated device
+ * @offset: offset from the beginning of
+ * configuration space
+ * @buf: buffer used to read to
+ * @len: the length to read from
+ * configration space
+ * @set_config: Write to device specific confiugration space
+ * @mdev: mediated device
+ * @offset: offset from the beginning of
+ * configuration space
+ * @buf: buffer used to write from
+ * @len: the length to write to
+ * configration space
+ * @get_version: Get the version of virtio mdev device
+ * @mdev: mediated device
+ * Returns integer: version of the device
+ * @get_generation: Get device generaton
+ * @mdev: mediated device
+ * Returns u32: device generation
+ */

include/linux/virtio_mdev.h new file +10 -0
+struct virtio_mdev_device_ops {
+	/* Virtqueue ops */
+	int (*set_vq_address)(struct mdev_device *mdev,
+			      u16 idx, u64 desc_area, u64 driver_area,
+			      u64 device_area);
+	void (*set_vq_num)(struct mdev_device *mdev, u16 idx, u32 num);
+	void (*kick_vq)(struct mdev_device *mdev, u16 idx);
+	void (*set_vq_cb)(struct mdev_device *mdev, u16 idx,
+			  struct virtio_mdev_callback *cb);
+	void (*set_vq_ready)(struct mdev_device *mdev, u16 idx, bool

ready);

include/linux/virtio_mdev.h new file +12 -0
+	bool (*get_vq_ready)(struct mdev_device *mdev, u16 idx);
+	int (*set_vq_state)(struct mdev_device *mdev, u16 idx, u64 state);
+	u64 (*get_vq_state)(struct mdev_device *mdev, u16 idx);
+
+	/* Device ops */
+	u16 (*get_vq_align)(struct mdev_device *mdev);
+	u64 (*get_features)(struct mdev_device *mdev);
+	int (*set_features)(struct mdev_device *mdev, u64 features);
+	void (*set_config_cb)(struct mdev_device *mdev,
+			      struct virtio_mdev_callback *cb);
+	u16 (*get_queue_max)(struct mdev_device *mdev);
+	u32 (*get_device_id)(struct mdev_device *mdev);

include/linux/virtio_mdev.h line 142

Quoted 10 lines
include/linux/virtio_mdev.h new file +10 -0
+	u32 (*get_vendor_id)(struct mdev_device *mdev);
+	u8 (*get_status)(struct mdev_device *mdev);
+	void (*set_status)(struct mdev_device *mdev, u8 status);
+	void (*get_config)(struct mdev_device *mdev, unsigned int offset,
+			   void *buf, unsigned int len);
+	void (*set_config)(struct mdev_device *mdev, unsigned int offset,
+			   const void *buf, unsigned int len);
+	int (*get_version)(struct mdev_device *mdev);
+	u32 (*get_generation)(struct mdev_device *mdev);
+};

I'm not sure how stable above ops are. Does it make sense if defining just two callbacks here, e.g. vq_ctrl and device_ctrl, and then let the vendor driver to handle specific ops in each category (similar to how ioctl works)?

Thanks
Kevin

author
#47 lore raw

Re: [PATCH V2 6/8] mdev: introduce virtio device and its device ops

Quoted 44 lines · Kevin
Quoted 44 lines

From: Jason Wang [mailto:[email protected]]
Sent: Tuesday, September 24, 2019 9:54 PM

This patch implements basic support for mdev driver that supports virtio transport for kernel virtio driver.

---
  include/linux/mdev.h        |   2 +
  include/linux/virtio_mdev.h | 145
++++++++++++++++++++++++++++++++++++
  2 files changed, 147 insertions(+)

create mode 100644 include/linux/virtio_mdev.h

include/linux/mdev.h +2 -0
index 3414307311f1..73ac27b3b868 100644
@@ -126,6 +126,8 @@ struct mdev_device *mdev_from_dev(struct device
 *dev);
 
  enum {
  	MDEV_ID_VFIO = 1,
+	MDEV_ID_VIRTIO = 2,
+	MDEV_ID_VHOST = 3,
  	/* New entries must be added here */
  };
include/linux/virtio_mdev.h new file +10 -0
new file mode 100644
index 000000000000..d1a40a739266
@@ -0,0 +1,145 @@
+/* SPDX-License-Identifier: GPL-2.0-only */
+/*
+ * Virtio mediated device driver
+ *
+ * Copyright 2019, Red Hat Corp.
+ *     Author: Jason Wang <[email protected]>
+ */
+#ifndef _LINUX_VIRTIO_MDEV_H
+#define _LINUX_VIRTIO_MDEV_H
+
Quoted 8 lines

include/linux/virtio_mdev.h line 16

Quoted 6 lines
include/linux/virtio_mdev.h new file +6 -0
+#include <linux/interrupt.h>
+#include <linux/mdev.h>
+#include <uapi/linux/vhost.h>
+
+#define VIRTIO_MDEV_DEVICE_API_STRING		"virtio-mdev"
+#define VIRTIO_MDEV_VERSION 0x1

Just be curious. is this version identical to virtio spec version that below callbacks are created for, or just irrelevant?

It could be a hint but basically it's a way for userspace driver compatibility. For kernel we don't need this.

Quoted 71 lines
Quoted 71 lines
include/linux/virtio_mdev.h new file +29 -0
+
+struct virtio_mdev_callback {
+	irqreturn_t (*callback)(void *data);
+	void *private;
+};
+
+/**
+ * struct vfio_mdev_device_ops - Structure to be registered for each
+ * mdev device to register the device to virtio-mdev module.
+ *
+ * @set_vq_address:		Set the address of virtqueue
+ *				@mdev: mediated device
+ *				@idx: virtqueue index
+ *				@desc_area: address of desc area
+ *				@driver_area: address of driver area
+ *				@device_area: address of device area
+ *				Returns integer: success (0) or error (< 0)
+ * @set_vq_num:		Set the size of virtqueue
+ *				@mdev: mediated device
+ *				@idx: virtqueue index
+ *				@num: the size of virtqueue
+ * @kick_vq:			Kick the virtqueue
+ *				@mdev: mediated device
+ *				@idx: virtqueue index
+ * @set_vq_cb:			Set the interrut calback function for
+ *				a virtqueue
+ *				@mdev: mediated device
+ *				@idx: virtqueue index
+ *				@cb: virtio-mdev interrupt callback

structure
+ * @set_vq_ready: Set ready status for a virtqueue
+ * @mdev: mediated device
+ * @idx: virtqueue index
+ * @ready: ready (true) not ready(false)
+ * @get_vq_ready: Get ready status for a virtqueue
+ * @mdev: mediated device
+ * @idx: virtqueue index
+ * Returns boolean: ready (true) or not (false)
+ * @set_vq_state: Set the state for a virtqueue
+ * @mdev: mediated device
+ * @idx: virtqueue index
+ * @state: virtqueue state (last_avail_idx)
+ * Returns integer: success (0) or error (< 0)
+ * @get_vq_state: Get the state for a virtqueue
+ * @mdev: mediated device
+ * @idx: virtqueue index
+ * Returns virtqueue state (last_avail_idx)
+ * @get_vq_align: Get the virtqueue align requirement
+ * for the device
+ * @mdev: mediated device
+ * Returns virtqueue algin requirement
+ * @get_features: Get virtio features supported by the device
+ * @mdev: mediated device
+ * Returns the features support by the
+ * device
+ * @get_features: Set virtio features supported by the driver
+ * @mdev: mediated device
+ * @features: feature support by the driver
+ * Returns integer: success (0) or error (< 0)
+ * @set_config_cb: Set the config interrupt callback
+ * @mdev: mediated device
+ * @cb: virtio-mdev interrupt callback
structure
+ * @get_device_id: Get virtio device id
+ * @mdev: mediated device
+ * Returns u32: virtio device id
+ * @get_vendor_id: Get virtio vendor id
+ * @mdev: mediated device
+ * Returns u32: virtio vendor id
+ * @get_status: Get the device status

Quoted 8 lines
Quoted 7 lines

+ * @mdev: mediated device
+ * Returns u8: virtio device status
+ * @set_status: Set the device status
+ * @mdev: mediated device
+ * @status: virtio device status
+ * @get_config: Read from device specific confiugration
space

configuration (and similar typos downward)

Let me fix.

Quoted 47 lines
Quoted 47 lines

+ * @mdev: mediated device
+ * @offset: offset from the beginning of
+ * configuration space
+ * @buf: buffer used to read to
+ * @len: the length to read from
+ * configration space
+ * @set_config: Write to device specific confiugration space
+ * @mdev: mediated device
+ * @offset: offset from the beginning of
+ * configuration space
+ * @buf: buffer used to write from
+ * @len: the length to write to
+ * configration space
+ * @get_version: Get the version of virtio mdev device
+ * @mdev: mediated device
+ * Returns integer: version of the device
+ * @get_generation: Get device generaton
+ * @mdev: mediated device
+ * Returns u32: device generation
+ */

include/linux/virtio_mdev.h new file +10 -0
+struct virtio_mdev_device_ops {
+	/* Virtqueue ops */
+	int (*set_vq_address)(struct mdev_device *mdev,
+			      u16 idx, u64 desc_area, u64 driver_area,
+			      u64 device_area);
+	void (*set_vq_num)(struct mdev_device *mdev, u16 idx, u32 num);
+	void (*kick_vq)(struct mdev_device *mdev, u16 idx);
+	void (*set_vq_cb)(struct mdev_device *mdev, u16 idx,
+			  struct virtio_mdev_callback *cb);
+	void (*set_vq_ready)(struct mdev_device *mdev, u16 idx, bool

ready);

include/linux/virtio_mdev.h new file +15 -0
+	bool (*get_vq_ready)(struct mdev_device *mdev, u16 idx);
+	int (*set_vq_state)(struct mdev_device *mdev, u16 idx, u64 state);
+	u64 (*get_vq_state)(struct mdev_device *mdev, u16 idx);
+
+	/* Device ops */
+	u16 (*get_vq_align)(struct mdev_device *mdev);
+	u64 (*get_features)(struct mdev_device *mdev);
+	int (*set_features)(struct mdev_device *mdev, u64 features);
+	void (*set_config_cb)(struct mdev_device *mdev,
+			      struct virtio_mdev_callback *cb);
+	u16 (*get_queue_max)(struct mdev_device *mdev);
+	u32 (*get_device_id)(struct mdev_device *mdev);
+	u32 (*get_vendor_id)(struct mdev_device *mdev);
+	u8 (*get_status)(struct mdev_device *mdev);
+	void (*set_status)(struct mdev_device *mdev, u8 status);
Quoted 8 lines

include/linux/virtio_mdev.h line 142

Quoted 7 lines
include/linux/virtio_mdev.h new file +7 -0
+	void (*get_config)(struct mdev_device *mdev, unsigned int offset,
+			   void *buf, unsigned int len);
+	void (*set_config)(struct mdev_device *mdev, unsigned int offset,
+			   const void *buf, unsigned int len);
+	int (*get_version)(struct mdev_device *mdev);
+	u32 (*get_generation)(struct mdev_device *mdev);
+};

I'm not sure how stable above ops are.

It's the kernel internal API, so there's no strict requirement for this. We will export a version value for userspace for compatibility.

Quoted 4 lines

Does it make sense if defining
just two callbacks here, e.g. vq_ctrl and device_ctrl, and then let the vendor driver to handle specific ops in each category (similar to how ioctl works)?

My understanding is that it introduce another indirection, you still need to differ from different command, and it's less flexible than direct callback.

What's the value of doing this?

Thanks

Quoted 4 lines

Thanks
Kevin

#48 lore raw

Re: [PATCH V2 6/8] mdev: introduce virtio device and its device ops

Quoted 201 lines · Jason Wang
Quoted 44 lines · Kevin
Quoted 44 lines

From: Jason Wang [mailto:[email protected]]
Sent: Tuesday, September 24, 2019 9:54 PM

This patch implements basic support for mdev driver that supports virtio transport for kernel virtio driver.

---
  include/linux/mdev.h        |   2 +
  include/linux/virtio_mdev.h | 145
++++++++++++++++++++++++++++++++++++
  2 files changed, 147 insertions(+)

create mode 100644 include/linux/virtio_mdev.h

include/linux/mdev.h +2 -0
index 3414307311f1..73ac27b3b868 100644
@@ -126,6 +126,8 @@ struct mdev_device *mdev_from_dev(struct device
 *dev);
 
  enum {
  	MDEV_ID_VFIO = 1,
+	MDEV_ID_VIRTIO = 2,
+	MDEV_ID_VHOST = 3,
  	/* New entries must be added here */
  };
include/linux/virtio_mdev.h new file +10 -0
new file mode 100644
index 000000000000..d1a40a739266
@@ -0,0 +1,145 @@
+/* SPDX-License-Identifier: GPL-2.0-only */
+/*
+ * Virtio mediated device driver
+ *
+ * Copyright 2019, Red Hat Corp.
+ *     Author: Jason Wang <[email protected]>
+ */
+#ifndef _LINUX_VIRTIO_MDEV_H
+#define _LINUX_VIRTIO_MDEV_H
+
Quoted 8 lines

include/linux/virtio_mdev.h line 16

Quoted 6 lines
include/linux/virtio_mdev.h new file +6 -0
+#include <linux/interrupt.h>
+#include <linux/mdev.h>
+#include <uapi/linux/vhost.h>
+
+#define VIRTIO_MDEV_DEVICE_API_STRING		"virtio-mdev"
+#define VIRTIO_MDEV_VERSION 0x1

Just be curious. is this version identical to virtio spec version that below callbacks are created for, or just irrelevant?

It could be a hint but basically it's a way for userspace driver compatibility. For kernel we don't need this.

Quoted 71 lines
Quoted 71 lines
include/linux/virtio_mdev.h new file +29 -0
+
+struct virtio_mdev_callback {
+	irqreturn_t (*callback)(void *data);
+	void *private;
+};
+
+/**
+ * struct vfio_mdev_device_ops - Structure to be registered for each
+ * mdev device to register the device to virtio-mdev module.
+ *
+ * @set_vq_address:		Set the address of virtqueue
+ *				@mdev: mediated device
+ *				@idx: virtqueue index
+ *				@desc_area: address of desc area
+ *				@driver_area: address of driver area
+ *				@device_area: address of device area
+ *				Returns integer: success (0) or error (< 0)
+ * @set_vq_num:		Set the size of virtqueue
+ *				@mdev: mediated device
+ *				@idx: virtqueue index
+ *				@num: the size of virtqueue
+ * @kick_vq:			Kick the virtqueue
+ *				@mdev: mediated device
+ *				@idx: virtqueue index
+ * @set_vq_cb:			Set the interrut calback function for
+ *				a virtqueue
+ *				@mdev: mediated device
+ *				@idx: virtqueue index
+ *				@cb: virtio-mdev interrupt callback

structure
+ * @set_vq_ready: Set ready status for a virtqueue
+ * @mdev: mediated device
+ * @idx: virtqueue index
+ * @ready: ready (true) not ready(false)
+ * @get_vq_ready: Get ready status for a virtqueue
+ * @mdev: mediated device
+ * @idx: virtqueue index
+ * Returns boolean: ready (true) or not (false)
+ * @set_vq_state: Set the state for a virtqueue
+ * @mdev: mediated device
+ * @idx: virtqueue index
+ * @state: virtqueue state (last_avail_idx)
+ * Returns integer: success (0) or error (< 0)
+ * @get_vq_state: Get the state for a virtqueue
+ * @mdev: mediated device
+ * @idx: virtqueue index
+ * Returns virtqueue state (last_avail_idx)
+ * @get_vq_align: Get the virtqueue align requirement
+ * for the device
+ * @mdev: mediated device
+ * Returns virtqueue algin requirement
+ * @get_features: Get virtio features supported by the device
+ * @mdev: mediated device
+ * Returns the features support by the
+ * device
+ * @get_features: Set virtio features supported by the driver
+ * @mdev: mediated device
+ * @features: feature support by the driver
+ * Returns integer: success (0) or error (< 0)
+ * @set_config_cb: Set the config interrupt callback
+ * @mdev: mediated device
+ * @cb: virtio-mdev interrupt callback
structure
+ * @get_device_id: Get virtio device id
+ * @mdev: mediated device
+ * Returns u32: virtio device id
+ * @get_vendor_id: Get virtio vendor id
+ * @mdev: mediated device
+ * Returns u32: virtio vendor id
+ * @get_status: Get the device status

Quoted 8 lines
Quoted 7 lines

+ * @mdev: mediated device
+ * Returns u8: virtio device status
+ * @set_status: Set the device status
+ * @mdev: mediated device
+ * @status: virtio device status
+ * @get_config: Read from device specific confiugration
space

configuration (and similar typos downward)

Let me fix.

Quoted 47 lines
Quoted 47 lines

+ * @mdev: mediated device
+ * @offset: offset from the beginning of
+ * configuration space
+ * @buf: buffer used to read to
+ * @len: the length to read from
+ * configration space
+ * @set_config: Write to device specific confiugration space
+ * @mdev: mediated device
+ * @offset: offset from the beginning of
+ * configuration space
+ * @buf: buffer used to write from
+ * @len: the length to write to
+ * configration space
+ * @get_version: Get the version of virtio mdev device
+ * @mdev: mediated device
+ * Returns integer: version of the device
+ * @get_generation: Get device generaton
+ * @mdev: mediated device
+ * Returns u32: device generation
+ */

include/linux/virtio_mdev.h new file +10 -0
+struct virtio_mdev_device_ops {
+	/* Virtqueue ops */
+	int (*set_vq_address)(struct mdev_device *mdev,
+			      u16 idx, u64 desc_area, u64 driver_area,
+			      u64 device_area);
+	void (*set_vq_num)(struct mdev_device *mdev, u16 idx, u32 num);
+	void (*kick_vq)(struct mdev_device *mdev, u16 idx);
+	void (*set_vq_cb)(struct mdev_device *mdev, u16 idx,
+			  struct virtio_mdev_callback *cb);
+	void (*set_vq_ready)(struct mdev_device *mdev, u16 idx, bool

ready);

include/linux/virtio_mdev.h new file +15 -0
+	bool (*get_vq_ready)(struct mdev_device *mdev, u16 idx);
+	int (*set_vq_state)(struct mdev_device *mdev, u16 idx, u64 state);
+	u64 (*get_vq_state)(struct mdev_device *mdev, u16 idx);
+
+	/* Device ops */
+	u16 (*get_vq_align)(struct mdev_device *mdev);
+	u64 (*get_features)(struct mdev_device *mdev);
+	int (*set_features)(struct mdev_device *mdev, u64 features);
+	void (*set_config_cb)(struct mdev_device *mdev,
+			      struct virtio_mdev_callback *cb);
+	u16 (*get_queue_max)(struct mdev_device *mdev);
+	u32 (*get_device_id)(struct mdev_device *mdev);
+	u32 (*get_vendor_id)(struct mdev_device *mdev);
+	u8 (*get_status)(struct mdev_device *mdev);
+	void (*set_status)(struct mdev_device *mdev, u8 status);
Quoted 8 lines

include/linux/virtio_mdev.h line 142

Quoted 7 lines
include/linux/virtio_mdev.h new file +7 -0
+	void (*get_config)(struct mdev_device *mdev, unsigned int offset,
+			   void *buf, unsigned int len);
+	void (*set_config)(struct mdev_device *mdev, unsigned int offset,
+			   const void *buf, unsigned int len);
+	int (*get_version)(struct mdev_device *mdev);
+	u32 (*get_generation)(struct mdev_device *mdev);
+};

I'm not sure how stable above ops are.

Quoted 2 lines

It's the kernel internal API, so there's no strict requirement for this. We will export a version value for userspace for compatibility.

Given it's tied to virtio we probably want kernel+userspace feature bits?

Quoted 7 lines
Quoted 5 lines

Does it make sense if defining
just two callbacks here, e.g. vq_ctrl and device_ctrl, and then let the vendor driver to handle specific ops in each category (similar to how ioctl works)?

Quoted 12 lines

My understanding is that it introduce another indirection, you still need to differ from different command, and it's less flexible than direct callback.

What's the value of doing this?

Thanks

Quoted 4 lines

Thanks
Kevin

_______________________________________________
dri-devel mailing list
[email protected]
https://lists.freedesktop.org/mailman/listinfo/dri-devel

author
#49 lore raw

Re: [PATCH V2 6/8] mdev: introduce virtio device and its device ops

Quoted 193 lines · Michael S. Tsirkin
Quoted 192 lines · Jason Wang
Quoted 44 lines · Kevin
Quoted 44 lines

From: Jason Wang [mailto:[email protected]]
Sent: Tuesday, September 24, 2019 9:54 PM

This patch implements basic support for mdev driver that supports virtio transport for kernel virtio driver.

---
   include/linux/mdev.h        |   2 +
   include/linux/virtio_mdev.h | 145
++++++++++++++++++++++++++++++++++++
   2 files changed, 147 insertions(+)

create mode 100644 include/linux/virtio_mdev.h

include/linux/mdev.h +2 -0
index 3414307311f1..73ac27b3b868 100644
@@ -126,6 +126,8 @@ struct mdev_device *mdev_from_dev(struct device
 *dev);
 
   enum {
   	MDEV_ID_VFIO = 1,
+	MDEV_ID_VIRTIO = 2,
+	MDEV_ID_VHOST = 3,
   	/* New entries must be added here */
   };
include/linux/virtio_mdev.h new file +10 -0
new file mode 100644
index 000000000000..d1a40a739266
@@ -0,0 +1,145 @@
+/* SPDX-License-Identifier: GPL-2.0-only */
+/*
+ * Virtio mediated device driver
+ *
+ * Copyright 2019, Red Hat Corp.
+ *     Author: Jason Wang<[email protected]>
+ */
+#ifndef _LINUX_VIRTIO_MDEV_H
+#define _LINUX_VIRTIO_MDEV_H
+
Quoted 8 lines

include/linux/virtio_mdev.h line 16

Quoted 6 lines
include/linux/virtio_mdev.h new file +6 -0
+#include <linux/interrupt.h>
+#include <linux/mdev.h>
+#include <uapi/linux/vhost.h>
+
+#define VIRTIO_MDEV_DEVICE_API_STRING		"virtio-mdev"
+#define VIRTIO_MDEV_VERSION 0x1

Just be curious. is this version identical to virtio spec version that below callbacks are created for, or just irrelevant?

It could be a hint but basically it's a way for userspace driver compatibility. For kernel we don't need this.

Quoted 70 lines
Quoted 70 lines
include/linux/virtio_mdev.h new file +29 -0
+
+struct virtio_mdev_callback {
+	irqreturn_t (*callback)(void *data);
+	void *private;
+};
+
+/**
+ * struct vfio_mdev_device_ops - Structure to be registered for each
+ * mdev device to register the device to virtio-mdev module.
+ *
+ * @set_vq_address:		Set the address of virtqueue
+ *				@mdev: mediated device
+ *				@idx: virtqueue index
+ *				@desc_area: address of desc area
+ *				@driver_area: address of driver area
+ *				@device_area: address of device area
+ *				Returns integer: success (0) or error (< 0)
+ * @set_vq_num:		Set the size of virtqueue
+ *				@mdev: mediated device
+ *				@idx: virtqueue index
+ *				@num: the size of virtqueue
+ * @kick_vq:			Kick the virtqueue
+ *				@mdev: mediated device
+ *				@idx: virtqueue index
+ * @set_vq_cb:			Set the interrut calback function for
+ *				a virtqueue
+ *				@mdev: mediated device
+ *				@idx: virtqueue index
+ *				@cb: virtio-mdev interrupt callback

structure
+ * @set_vq_ready: Set ready status for a virtqueue
+ * @mdev: mediated device
+ * @idx: virtqueue index
+ * @ready: ready (true) not ready(false)
+ * @get_vq_ready: Get ready status for a virtqueue
+ * @mdev: mediated device
+ * @idx: virtqueue index
+ * Returns boolean: ready (true) or not (false)
+ * @set_vq_state: Set the state for a virtqueue
+ * @mdev: mediated device
+ * @idx: virtqueue index
+ * @state: virtqueue state (last_avail_idx)
+ * Returns integer: success (0) or error (< 0)
+ * @get_vq_state: Get the state for a virtqueue
+ * @mdev: mediated device
+ * @idx: virtqueue index
+ * Returns virtqueue state (last_avail_idx)
+ * @get_vq_align: Get the virtqueue align requirement
+ * for the device
+ * @mdev: mediated device
+ * Returns virtqueue algin requirement
+ * @get_features: Get virtio features supported by the device
+ * @mdev: mediated device
+ * Returns the features support by the
+ * device
+ * @get_features: Set virtio features supported by the driver
+ * @mdev: mediated device
+ * @features: feature support by the driver
+ * Returns integer: success (0) or error (< 0)
+ * @set_config_cb: Set the config interrupt callback
+ * @mdev: mediated device
+ * @cb: virtio-mdev interrupt callback
structure
+ * @get_device_id: Get virtio device id
+ * @mdev: mediated device
+ * Returns u32: virtio device id
+ * @get_vendor_id: Get virtio vendor id
+ * @mdev: mediated device
+ * Returns u32: virtio vendor id
+ * @get_status: Get the device status

Quoted 8 lines
Quoted 7 lines

+ * @mdev: mediated device
+ * Returns u8: virtio device status
+ * @set_status: Set the device status
+ * @mdev: mediated device
+ * @status: virtio device status
+ * @get_config: Read from device specific confiugration
space

configuration (and similar typos downward)

Let me fix.

Quoted 54 lines
Quoted 43 lines

+ * @mdev: mediated device
+ * @offset: offset from the beginning of
+ * configuration space
+ * @buf: buffer used to read to
+ * @len: the length to read from
+ * configration space
+ * @set_config: Write to device specific confiugration space
+ * @mdev: mediated device
+ * @offset: offset from the beginning of
+ * configuration space
+ * @buf: buffer used to write from
+ * @len: the length to write to
+ * configration space
+ * @get_version: Get the version of virtio mdev device
+ * @mdev: mediated device
+ * Returns integer: version of the device
+ * @get_generation: Get device generaton
+ * @mdev: mediated device
+ * Returns u32: device generation
+ */

include/linux/virtio_mdev.h new file +10 -0
+struct virtio_mdev_device_ops {
+	/* Virtqueue ops */
+	int (*set_vq_address)(struct mdev_device *mdev,
+			      u16 idx, u64 desc_area, u64 driver_area,
+			      u64 device_area);
+	void (*set_vq_num)(struct mdev_device *mdev, u16 idx, u32 num);
+	void (*kick_vq)(struct mdev_device *mdev, u16 idx);
+	void (*set_vq_cb)(struct mdev_device *mdev, u16 idx,
+			  struct virtio_mdev_callback *cb);
+	void (*set_vq_ready)(struct mdev_device *mdev, u16 idx, bool

ready);

include/linux/virtio_mdev.h new file +12 -0
+	bool (*get_vq_ready)(struct mdev_device *mdev, u16 idx);
+	int (*set_vq_state)(struct mdev_device *mdev, u16 idx, u64 state);
+	u64 (*get_vq_state)(struct mdev_device *mdev, u16 idx);
+
+	/* Device ops */
+	u16 (*get_vq_align)(struct mdev_device *mdev);
+	u64 (*get_features)(struct mdev_device *mdev);
+	int (*set_features)(struct mdev_device *mdev, u64 features);
+	void (*set_config_cb)(struct mdev_device *mdev,
+			      struct virtio_mdev_callback *cb);
+	u16 (*get_queue_max)(struct mdev_device *mdev);
+	u32 (*get_device_id)(struct mdev_device *mdev);

include/linux/virtio_mdev.h line 142

Quoted 10 lines
include/linux/virtio_mdev.h new file +10 -0
+	u32 (*get_vendor_id)(struct mdev_device *mdev);
+	u8 (*get_status)(struct mdev_device *mdev);
+	void (*set_status)(struct mdev_device *mdev, u8 status);
+	void (*get_config)(struct mdev_device *mdev, unsigned int offset,
+			   void *buf, unsigned int len);
+	void (*set_config)(struct mdev_device *mdev, unsigned int offset,
+			   const void *buf, unsigned int len);
+	int (*get_version)(struct mdev_device *mdev);
+	u32 (*get_generation)(struct mdev_device *mdev);
+};

I'm not sure how stable above ops are.

Quoted 4 lines
Quoted 2 lines

It's the kernel internal API, so there's no strict requirement for this. We will export a version value for userspace for compatibility.

Given it's tied to virtio we probably want kernel+userspace feature bits?

Yes, then I think we could probably have a version field inside e.g device_ops structure. Then it could be fetched from both kernel and userspace driver.

Thanks

Quoted 1 line

_______________________________________________
Intel-gfx mailing list
[email protected]
https://lists.freedesktop.org/mailman/listinfo/intel-gfx

#50 lore raw

Re: [PATCH V2 6/8] mdev: introduce virtio device and its device ops

Quoted 7 lines · Jason Wang
Quoted 5 lines
Quoted 3 lines
Quoted 1 line

I'm not sure how stable above ops are.

It's the kernel internal API, so there's no strict requirement for this. We will export a version value for userspace for compatibility.

Given it's tied to virtio we probably want kernel+userspace feature bits?

Quoted 6 lines

Yes, then I think we could probably have a version field inside e.g device_ops structure. Then it could be fetched from both kernel and userspace driver.

Thanks

my point was feature bits not a version number.

Quoted 1 line
Quoted 1 line

_______________________________________________
dri-devel mailing list
[email protected]
https://lists.freedesktop.org/mailman/listinfo/dri-devel

author
#51 lore raw

Re: [PATCH V2 6/8] mdev: introduce virtio device and its device ops

Quoted 7 lines · Michael S. Tsirkin
Quoted 6 lines · Jason Wang
Quoted 5 lines
Quoted 3 lines
Quoted 1 line

I'm not sure how stable above ops are.

It's the kernel internal API, so there's no strict requirement for this. We will export a version value for userspace for compatibility.

Given it's tied to virtio we probably want kernel+userspace feature bits?

Quoted 8 lines
Quoted 6 lines

Yes, then I think we could probably have a version field inside e.g device_ops structure. Then it could be fetched from both kernel and userspace driver.

Thanks

my point was feature bits not a version number.

Something like backend_feature that current vhost_net did?

Thanks

Quoted 1 line

_______________________________________________
dri-devel mailing list
[email protected]
https://lists.freedesktop.org/mailman/listinfo/dri-devel

#52 lore raw

Re: [PATCH V2 6/8] mdev: introduce virtio device and its device ops

Quoted 16 lines · Jason Wang
Quoted 14 lines · Michael S. Tsirkin
Quoted 12 lines · Jason Wang
Quoted 5 lines
Quoted 3 lines
Quoted 1 line

I'm not sure how stable above ops are.

It's the kernel internal API, so there's no strict requirement for this. We will export a version value for userspace for compatibility.

Given it's tied to virtio we probably want kernel+userspace feature bits?

Yes, then I think we could probably have a version field inside e.g device_ops structure. Then it could be fetched from both kernel and userspace driver.

Thanks

Quoted 9 lines
Quoted 1 line

my point was feature bits not a version number.

Something like backend_feature that current vhost_net did?

Thanks

Quoted 1 line

right
_______________________________________________
dri-devel mailing list
[email protected]
https://lists.freedesktop.org/mailman/listinfo/dri-devel

#53 lore raw

Re: [PATCH V2 6/8] mdev: introduce virtio device and its device ops

In reply to Jason Wang
Quoted 16 lines · Jason Wang
Quoted 14 lines · Michael S. Tsirkin
Quoted 12 lines · Jason Wang
Quoted 5 lines
Quoted 3 lines
Quoted 1 line

I'm not sure how stable above ops are.

It's the kernel internal API, so there's no strict requirement for this. We will export a version value for userspace for compatibility.

Given it's tied to virtio we probably want kernel+userspace feature bits?

Yes, then I think we could probably have a version field inside e.g device_ops structure. Then it could be fetched from both kernel and userspace driver.

Thanks

Quoted 9 lines
Quoted 1 line

my point was feature bits not a version number.

Something like backend_feature that current vhost_net did?

Thanks

Quoted 1 line

right
_______________________________________________
Virtualization mailing list
[email protected]
https://lists.linuxfoundation.org/mailman/listinfo/virtualization

author
#54 lore raw

Re: [PATCH V2 6/8] mdev: introduce virtio device and its device ops

In reply to Michael S. Tsirkin
Quoted 7 lines · Michael S. Tsirkin
Quoted 6 lines · Jason Wang
Quoted 5 lines
Quoted 3 lines
Quoted 1 line

I'm not sure how stable above ops are.

It's the kernel internal API, so there's no strict requirement for this. We will export a version value for userspace for compatibility.

Given it's tied to virtio we probably want kernel+userspace feature bits?

Quoted 8 lines
Quoted 6 lines

Yes, then I think we could probably have a version field inside e.g device_ops structure. Then it could be fetched from both kernel and userspace driver.

Thanks

my point was feature bits not a version number.

Something like backend_feature that current vhost_net did?

Thanks

Quoted 1 line

_______________________________________________
Virtualization mailing list
[email protected]
https://lists.linuxfoundation.org/mailman/listinfo/virtualization

#55 lore raw

Re: [PATCH V2 6/8] mdev: introduce virtio device and its device ops

In reply to Jason Wang
Quoted 7 lines · Jason Wang
Quoted 5 lines
Quoted 3 lines
Quoted 1 line

I'm not sure how stable above ops are.

It's the kernel internal API, so there's no strict requirement for this. We will export a version value for userspace for compatibility.

Given it's tied to virtio we probably want kernel+userspace feature bits?

Quoted 6 lines

Yes, then I think we could probably have a version field inside e.g device_ops structure. Then it could be fetched from both kernel and userspace driver.

Thanks

my point was feature bits not a version number.

Quoted 1 line
Quoted 1 line
author
#56 lore raw

Re: [PATCH V2 6/8] mdev: introduce virtio device and its device ops

In reply to Michael S. Tsirkin
Quoted 193 lines · Michael S. Tsirkin
Quoted 192 lines · Jason Wang
Quoted 44 lines · Kevin
Quoted 44 lines

From: Jason Wang [mailto:[email protected]]
Sent: Tuesday, September 24, 2019 9:54 PM

This patch implements basic support for mdev driver that supports virtio transport for kernel virtio driver.

---
   include/linux/mdev.h        |   2 +
   include/linux/virtio_mdev.h | 145
++++++++++++++++++++++++++++++++++++
   2 files changed, 147 insertions(+)

create mode 100644 include/linux/virtio_mdev.h

include/linux/mdev.h +2 -0
index 3414307311f1..73ac27b3b868 100644
@@ -126,6 +126,8 @@ struct mdev_device *mdev_from_dev(struct device
 *dev);
 
   enum {
   	MDEV_ID_VFIO = 1,
+	MDEV_ID_VIRTIO = 2,
+	MDEV_ID_VHOST = 3,
   	/* New entries must be added here */
   };
include/linux/virtio_mdev.h new file +10 -0
new file mode 100644
index 000000000000..d1a40a739266
@@ -0,0 +1,145 @@
+/* SPDX-License-Identifier: GPL-2.0-only */
+/*
+ * Virtio mediated device driver
+ *
+ * Copyright 2019, Red Hat Corp.
+ *     Author: Jason Wang<[email protected]>
+ */
+#ifndef _LINUX_VIRTIO_MDEV_H
+#define _LINUX_VIRTIO_MDEV_H
+
Quoted 8 lines

include/linux/virtio_mdev.h line 16

Quoted 6 lines
include/linux/virtio_mdev.h new file +6 -0
+#include <linux/interrupt.h>
+#include <linux/mdev.h>
+#include <uapi/linux/vhost.h>
+
+#define VIRTIO_MDEV_DEVICE_API_STRING		"virtio-mdev"
+#define VIRTIO_MDEV_VERSION 0x1

Just be curious. is this version identical to virtio spec version that below callbacks are created for, or just irrelevant?

It could be a hint but basically it's a way for userspace driver compatibility. For kernel we don't need this.

Quoted 70 lines
Quoted 70 lines
include/linux/virtio_mdev.h new file +29 -0
+
+struct virtio_mdev_callback {
+	irqreturn_t (*callback)(void *data);
+	void *private;
+};
+
+/**
+ * struct vfio_mdev_device_ops - Structure to be registered for each
+ * mdev device to register the device to virtio-mdev module.
+ *
+ * @set_vq_address:		Set the address of virtqueue
+ *				@mdev: mediated device
+ *				@idx: virtqueue index
+ *				@desc_area: address of desc area
+ *				@driver_area: address of driver area
+ *				@device_area: address of device area
+ *				Returns integer: success (0) or error (< 0)
+ * @set_vq_num:		Set the size of virtqueue
+ *				@mdev: mediated device
+ *				@idx: virtqueue index
+ *				@num: the size of virtqueue
+ * @kick_vq:			Kick the virtqueue
+ *				@mdev: mediated device
+ *				@idx: virtqueue index
+ * @set_vq_cb:			Set the interrut calback function for
+ *				a virtqueue
+ *				@mdev: mediated device
+ *				@idx: virtqueue index
+ *				@cb: virtio-mdev interrupt callback

structure
+ * @set_vq_ready: Set ready status for a virtqueue
+ * @mdev: mediated device
+ * @idx: virtqueue index
+ * @ready: ready (true) not ready(false)
+ * @get_vq_ready: Get ready status for a virtqueue
+ * @mdev: mediated device
+ * @idx: virtqueue index
+ * Returns boolean: ready (true) or not (false)
+ * @set_vq_state: Set the state for a virtqueue
+ * @mdev: mediated device
+ * @idx: virtqueue index
+ * @state: virtqueue state (last_avail_idx)
+ * Returns integer: success (0) or error (< 0)
+ * @get_vq_state: Get the state for a virtqueue
+ * @mdev: mediated device
+ * @idx: virtqueue index
+ * Returns virtqueue state (last_avail_idx)
+ * @get_vq_align: Get the virtqueue align requirement
+ * for the device
+ * @mdev: mediated device
+ * Returns virtqueue algin requirement
+ * @get_features: Get virtio features supported by the device
+ * @mdev: mediated device
+ * Returns the features support by the
+ * device
+ * @get_features: Set virtio features supported by the driver
+ * @mdev: mediated device
+ * @features: feature support by the driver
+ * Returns integer: success (0) or error (< 0)
+ * @set_config_cb: Set the config interrupt callback
+ * @mdev: mediated device
+ * @cb: virtio-mdev interrupt callback
structure
+ * @get_device_id: Get virtio device id
+ * @mdev: mediated device
+ * Returns u32: virtio device id
+ * @get_vendor_id: Get virtio vendor id
+ * @mdev: mediated device
+ * Returns u32: virtio vendor id
+ * @get_status: Get the device status

Quoted 8 lines
Quoted 7 lines

+ * @mdev: mediated device
+ * Returns u8: virtio device status
+ * @set_status: Set the device status
+ * @mdev: mediated device
+ * @status: virtio device status
+ * @get_config: Read from device specific confiugration
space

configuration (and similar typos downward)

Let me fix.

Quoted 54 lines
Quoted 43 lines

+ * @mdev: mediated device
+ * @offset: offset from the beginning of
+ * configuration space
+ * @buf: buffer used to read to
+ * @len: the length to read from
+ * configration space
+ * @set_config: Write to device specific confiugration space
+ * @mdev: mediated device
+ * @offset: offset from the beginning of
+ * configuration space
+ * @buf: buffer used to write from
+ * @len: the length to write to
+ * configration space
+ * @get_version: Get the version of virtio mdev device
+ * @mdev: mediated device
+ * Returns integer: version of the device
+ * @get_generation: Get device generaton
+ * @mdev: mediated device
+ * Returns u32: device generation
+ */

include/linux/virtio_mdev.h new file +10 -0
+struct virtio_mdev_device_ops {
+	/* Virtqueue ops */
+	int (*set_vq_address)(struct mdev_device *mdev,
+			      u16 idx, u64 desc_area, u64 driver_area,
+			      u64 device_area);
+	void (*set_vq_num)(struct mdev_device *mdev, u16 idx, u32 num);
+	void (*kick_vq)(struct mdev_device *mdev, u16 idx);
+	void (*set_vq_cb)(struct mdev_device *mdev, u16 idx,
+			  struct virtio_mdev_callback *cb);
+	void (*set_vq_ready)(struct mdev_device *mdev, u16 idx, bool

ready);

include/linux/virtio_mdev.h new file +12 -0
+	bool (*get_vq_ready)(struct mdev_device *mdev, u16 idx);
+	int (*set_vq_state)(struct mdev_device *mdev, u16 idx, u64 state);
+	u64 (*get_vq_state)(struct mdev_device *mdev, u16 idx);
+
+	/* Device ops */
+	u16 (*get_vq_align)(struct mdev_device *mdev);
+	u64 (*get_features)(struct mdev_device *mdev);
+	int (*set_features)(struct mdev_device *mdev, u64 features);
+	void (*set_config_cb)(struct mdev_device *mdev,
+			      struct virtio_mdev_callback *cb);
+	u16 (*get_queue_max)(struct mdev_device *mdev);
+	u32 (*get_device_id)(struct mdev_device *mdev);

include/linux/virtio_mdev.h line 142

Quoted 10 lines
include/linux/virtio_mdev.h new file +10 -0
+	u32 (*get_vendor_id)(struct mdev_device *mdev);
+	u8 (*get_status)(struct mdev_device *mdev);
+	void (*set_status)(struct mdev_device *mdev, u8 status);
+	void (*get_config)(struct mdev_device *mdev, unsigned int offset,
+			   void *buf, unsigned int len);
+	void (*set_config)(struct mdev_device *mdev, unsigned int offset,
+			   const void *buf, unsigned int len);
+	int (*get_version)(struct mdev_device *mdev);
+	u32 (*get_generation)(struct mdev_device *mdev);
+};

I'm not sure how stable above ops are.

Quoted 4 lines
Quoted 2 lines

It's the kernel internal API, so there's no strict requirement for this. We will export a version value for userspace for compatibility.

Given it's tied to virtio we probably want kernel+userspace feature bits?

Yes, then I think we could probably have a version field inside e.g device_ops structure. Then it could be fetched from both kernel and userspace driver.

Thanks

Quoted 1 line

_______________________________________________
Virtualization mailing list
[email protected]
https://lists.linuxfoundation.org/mailman/listinfo/virtualization

#57 lore raw

Re: [PATCH V2 6/8] mdev: introduce virtio device and its device ops

In reply to Jason Wang
Quoted 201 lines · Jason Wang
Quoted 44 lines · Kevin
Quoted 44 lines

From: Jason Wang [mailto:[email protected]]
Sent: Tuesday, September 24, 2019 9:54 PM

This patch implements basic support for mdev driver that supports virtio transport for kernel virtio driver.

---
  include/linux/mdev.h        |   2 +
  include/linux/virtio_mdev.h | 145
++++++++++++++++++++++++++++++++++++
  2 files changed, 147 insertions(+)

create mode 100644 include/linux/virtio_mdev.h

include/linux/mdev.h +2 -0
index 3414307311f1..73ac27b3b868 100644
@@ -126,6 +126,8 @@ struct mdev_device *mdev_from_dev(struct device
 *dev);
 
  enum {
  	MDEV_ID_VFIO = 1,
+	MDEV_ID_VIRTIO = 2,
+	MDEV_ID_VHOST = 3,
  	/* New entries must be added here */
  };
include/linux/virtio_mdev.h new file +10 -0
new file mode 100644
index 000000000000..d1a40a739266
@@ -0,0 +1,145 @@
+/* SPDX-License-Identifier: GPL-2.0-only */
+/*
+ * Virtio mediated device driver
+ *
+ * Copyright 2019, Red Hat Corp.
+ *     Author: Jason Wang <[email protected]>
+ */
+#ifndef _LINUX_VIRTIO_MDEV_H
+#define _LINUX_VIRTIO_MDEV_H
+
Quoted 8 lines

include/linux/virtio_mdev.h line 16

Quoted 6 lines
include/linux/virtio_mdev.h new file +6 -0
+#include <linux/interrupt.h>
+#include <linux/mdev.h>
+#include <uapi/linux/vhost.h>
+
+#define VIRTIO_MDEV_DEVICE_API_STRING		"virtio-mdev"
+#define VIRTIO_MDEV_VERSION 0x1

Just be curious. is this version identical to virtio spec version that below callbacks are created for, or just irrelevant?

It could be a hint but basically it's a way for userspace driver compatibility. For kernel we don't need this.

Quoted 71 lines
Quoted 71 lines
include/linux/virtio_mdev.h new file +29 -0
+
+struct virtio_mdev_callback {
+	irqreturn_t (*callback)(void *data);
+	void *private;
+};
+
+/**
+ * struct vfio_mdev_device_ops - Structure to be registered for each
+ * mdev device to register the device to virtio-mdev module.
+ *
+ * @set_vq_address:		Set the address of virtqueue
+ *				@mdev: mediated device
+ *				@idx: virtqueue index
+ *				@desc_area: address of desc area
+ *				@driver_area: address of driver area
+ *				@device_area: address of device area
+ *				Returns integer: success (0) or error (< 0)
+ * @set_vq_num:		Set the size of virtqueue
+ *				@mdev: mediated device
+ *				@idx: virtqueue index
+ *				@num: the size of virtqueue
+ * @kick_vq:			Kick the virtqueue
+ *				@mdev: mediated device
+ *				@idx: virtqueue index
+ * @set_vq_cb:			Set the interrut calback function for
+ *				a virtqueue
+ *				@mdev: mediated device
+ *				@idx: virtqueue index
+ *				@cb: virtio-mdev interrupt callback

structure
+ * @set_vq_ready: Set ready status for a virtqueue
+ * @mdev: mediated device
+ * @idx: virtqueue index
+ * @ready: ready (true) not ready(false)
+ * @get_vq_ready: Get ready status for a virtqueue
+ * @mdev: mediated device
+ * @idx: virtqueue index
+ * Returns boolean: ready (true) or not (false)
+ * @set_vq_state: Set the state for a virtqueue
+ * @mdev: mediated device
+ * @idx: virtqueue index
+ * @state: virtqueue state (last_avail_idx)
+ * Returns integer: success (0) or error (< 0)
+ * @get_vq_state: Get the state for a virtqueue
+ * @mdev: mediated device
+ * @idx: virtqueue index
+ * Returns virtqueue state (last_avail_idx)
+ * @get_vq_align: Get the virtqueue align requirement
+ * for the device
+ * @mdev: mediated device
+ * Returns virtqueue algin requirement
+ * @get_features: Get virtio features supported by the device
+ * @mdev: mediated device
+ * Returns the features support by the
+ * device
+ * @get_features: Set virtio features supported by the driver
+ * @mdev: mediated device
+ * @features: feature support by the driver
+ * Returns integer: success (0) or error (< 0)
+ * @set_config_cb: Set the config interrupt callback
+ * @mdev: mediated device
+ * @cb: virtio-mdev interrupt callback
structure
+ * @get_device_id: Get virtio device id
+ * @mdev: mediated device
+ * Returns u32: virtio device id
+ * @get_vendor_id: Get virtio vendor id
+ * @mdev: mediated device
+ * Returns u32: virtio vendor id
+ * @get_status: Get the device status

Quoted 8 lines
Quoted 7 lines

+ * @mdev: mediated device
+ * Returns u8: virtio device status
+ * @set_status: Set the device status
+ * @mdev: mediated device
+ * @status: virtio device status
+ * @get_config: Read from device specific confiugration
space

configuration (and similar typos downward)

Let me fix.

Quoted 47 lines
Quoted 47 lines

+ * @mdev: mediated device
+ * @offset: offset from the beginning of
+ * configuration space
+ * @buf: buffer used to read to
+ * @len: the length to read from
+ * configration space
+ * @set_config: Write to device specific confiugration space
+ * @mdev: mediated device
+ * @offset: offset from the beginning of
+ * configuration space
+ * @buf: buffer used to write from
+ * @len: the length to write to
+ * configration space
+ * @get_version: Get the version of virtio mdev device
+ * @mdev: mediated device
+ * Returns integer: version of the device
+ * @get_generation: Get device generaton
+ * @mdev: mediated device
+ * Returns u32: device generation
+ */

include/linux/virtio_mdev.h new file +10 -0
+struct virtio_mdev_device_ops {
+	/* Virtqueue ops */
+	int (*set_vq_address)(struct mdev_device *mdev,
+			      u16 idx, u64 desc_area, u64 driver_area,
+			      u64 device_area);
+	void (*set_vq_num)(struct mdev_device *mdev, u16 idx, u32 num);
+	void (*kick_vq)(struct mdev_device *mdev, u16 idx);
+	void (*set_vq_cb)(struct mdev_device *mdev, u16 idx,
+			  struct virtio_mdev_callback *cb);
+	void (*set_vq_ready)(struct mdev_device *mdev, u16 idx, bool

ready);

include/linux/virtio_mdev.h new file +15 -0
+	bool (*get_vq_ready)(struct mdev_device *mdev, u16 idx);
+	int (*set_vq_state)(struct mdev_device *mdev, u16 idx, u64 state);
+	u64 (*get_vq_state)(struct mdev_device *mdev, u16 idx);
+
+	/* Device ops */
+	u16 (*get_vq_align)(struct mdev_device *mdev);
+	u64 (*get_features)(struct mdev_device *mdev);
+	int (*set_features)(struct mdev_device *mdev, u64 features);
+	void (*set_config_cb)(struct mdev_device *mdev,
+			      struct virtio_mdev_callback *cb);
+	u16 (*get_queue_max)(struct mdev_device *mdev);
+	u32 (*get_device_id)(struct mdev_device *mdev);
+	u32 (*get_vendor_id)(struct mdev_device *mdev);
+	u8 (*get_status)(struct mdev_device *mdev);
+	void (*set_status)(struct mdev_device *mdev, u8 status);
Quoted 8 lines

include/linux/virtio_mdev.h line 142

Quoted 7 lines
include/linux/virtio_mdev.h new file +7 -0
+	void (*get_config)(struct mdev_device *mdev, unsigned int offset,
+			   void *buf, unsigned int len);
+	void (*set_config)(struct mdev_device *mdev, unsigned int offset,
+			   const void *buf, unsigned int len);
+	int (*get_version)(struct mdev_device *mdev);
+	u32 (*get_generation)(struct mdev_device *mdev);
+};

I'm not sure how stable above ops are.

Quoted 2 lines

It's the kernel internal API, so there's no strict requirement for this. We will export a version value for userspace for compatibility.

Given it's tied to virtio we probably want kernel+userspace feature bits?

Quoted 7 lines
Quoted 5 lines

Does it make sense if defining
just two callbacks here, e.g. vq_ctrl and device_ctrl, and then let the vendor driver to handle specific ops in each category (similar to how ioctl works)?

Quoted 12 lines

My understanding is that it introduce another indirection, you still need to differ from different command, and it's less flexible than direct callback.

What's the value of doing this?

Thanks

Quoted 4 lines

Thanks
Kevin

_______________________________________________
Virtualization mailing list
[email protected]
https://lists.linuxfoundation.org/mailman/listinfo/virtualization

#58 lore raw

RE: [PATCH V2 6/8] mdev: introduce virtio device and its device ops

In reply to Jason Wang
Quoted 216 lines

From: Jason Wang
Sent: Wednesday, September 25, 2019 8:45 PM

Quoted 14 lines · Kevin
Quoted 14 lines

From: Jason Wang [mailto:[email protected]]
Sent: Tuesday, September 24, 2019 9:54 PM

This patch implements basic support for mdev driver that supports virtio transport for kernel virtio driver.

---
  include/linux/mdev.h        |   2 +
  include/linux/virtio_mdev.h | 145
++++++++++++++++++++++++++++++++++++
  2 files changed, 147 insertions(+)

create mode 100644 include/linux/virtio_mdev.h

Quoted 5 lines

include/linux/mdev.h struct mdev_device *mdev_from_dev(struct

Quoted 5 lines
include/linux/mdev.h +0 -0
index 3414307311f1..73ac27b3b868 100644
@@ -126,6 +126,8 @@ struct mdev_device *mdev_from_dev(struct

device

Quoted 22 lines
Quoted 22 lines
include/linux/mdev.h +2 -0
 *dev);
 
  enum {
  	MDEV_ID_VFIO = 1,
+	MDEV_ID_VIRTIO = 2,
+	MDEV_ID_VHOST = 3,
  	/* New entries must be added here */
  };
include/linux/virtio_mdev.h new file +7 -0
new file mode 100644
index 000000000000..d1a40a739266
@@ -0,0 +1,145 @@
+/* SPDX-License-Identifier: GPL-2.0-only */
+/*
+ * Virtio mediated device driver
+ *
+ * Copyright 2019, Red Hat Corp.
+ *     Author: Jason Wang <[email protected]>
+ */
Quoted 8 lines

include/linux/virtio_mdev.h

Quoted 8 lines
include/linux/virtio_mdev.h new file +8 -0
+#ifndef _LINUX_VIRTIO_MDEV_H
+#define _LINUX_VIRTIO_MDEV_H
+
+#include <linux/interrupt.h>
+#include <linux/mdev.h>
+#include <uapi/linux/vhost.h>
+
+#define VIRTIO_MDEV_DEVICE_API_STRING		"virtio-

mdev"

Quoted 3 lines

include/linux/virtio_mdev.h line 16

Quoted 1 line
include/linux/virtio_mdev.h new file +1 -0
+#define VIRTIO_MDEV_VERSION 0x1

Just be curious. is this version identical to virtio spec version that below callbacks are created for, or just irrelevant?

It could be a hint but basically it's a way for userspace driver compatibility. For kernel we don't need this.

Quoted 71 lines
Quoted 71 lines
include/linux/virtio_mdev.h new file +29 -0
+
+struct virtio_mdev_callback {
+	irqreturn_t (*callback)(void *data);
+	void *private;
+};
+
+/**
+ * struct vfio_mdev_device_ops - Structure to be registered for each
+ * mdev device to register the device to virtio-mdev module.
+ *
+ * @set_vq_address:		Set the address of virtqueue
+ *				@mdev: mediated device
+ *				@idx: virtqueue index
+ *				@desc_area: address of desc area
+ *				@driver_area: address of driver area
+ *				@device_area: address of device area
+ *				Returns integer: success (0) or error (< 0)
+ * @set_vq_num:		Set the size of virtqueue
+ *				@mdev: mediated device
+ *				@idx: virtqueue index
+ *				@num: the size of virtqueue
+ * @kick_vq:			Kick the virtqueue
+ *				@mdev: mediated device
+ *				@idx: virtqueue index
+ * @set_vq_cb:			Set the interrut calback function for
+ *				a virtqueue
+ *				@mdev: mediated device
+ *				@idx: virtqueue index
+ *				@cb: virtio-mdev interrupt callback

structure
+ * @set_vq_ready: Set ready status for a virtqueue
+ * @mdev: mediated device
+ * @idx: virtqueue index
+ * @ready: ready (true) not ready(false)
+ * @get_vq_ready: Get ready status for a virtqueue
+ * @mdev: mediated device
+ * @idx: virtqueue index
+ * Returns boolean: ready (true) or not (false)
+ * @set_vq_state: Set the state for a virtqueue
+ * @mdev: mediated device
+ * @idx: virtqueue index
+ * @state: virtqueue state (last_avail_idx)
+ * Returns integer: success (0) or error (< 0)
+ * @get_vq_state: Get the state for a virtqueue
+ * @mdev: mediated device
+ * @idx: virtqueue index
+ * Returns virtqueue state (last_avail_idx)
+ * @get_vq_align: Get the virtqueue align requirement
+ * for the device
+ * @mdev: mediated device
+ * Returns virtqueue algin requirement
+ * @get_features: Get virtio features supported by the device
+ * @mdev: mediated device
+ * Returns the features support by the
+ * device
+ * @get_features: Set virtio features supported by the driver
+ * @mdev: mediated device
+ * @features: feature support by the driver
+ * Returns integer: success (0) or error (< 0)
+ * @set_config_cb: Set the config interrupt callback
+ * @mdev: mediated device
+ * @cb: virtio-mdev interrupt callback
structure
+ * @get_device_id: Get virtio device id
+ * @mdev: mediated device
+ * Returns u32: virtio device id
+ * @get_vendor_id: Get virtio vendor id
+ * @mdev: mediated device
+ * Returns u32: virtio vendor id
+ * @get_status: Get the device status

Quoted 8 lines
Quoted 7 lines

+ * @mdev: mediated device
+ * Returns u8: virtio device status
+ * @set_status: Set the device status
+ * @mdev: mediated device
+ * @status: virtio device status
+ * @get_config: Read from device specific confiugration
space

configuration (and similar typos downward)

Let me fix.

Quoted 47 lines
Quoted 47 lines

+ * @mdev: mediated device
+ * @offset: offset from the beginning of
+ * configuration space
+ * @buf: buffer used to read to
+ * @len: the length to read from
+ * configration space
+ * @set_config: Write to device specific confiugration space
+ * @mdev: mediated device
+ * @offset: offset from the beginning of
+ * configuration space
+ * @buf: buffer used to write from
+ * @len: the length to write to
+ * configration space
+ * @get_version: Get the version of virtio mdev device
+ * @mdev: mediated device
+ * Returns integer: version of the device
+ * @get_generation: Get device generaton
+ * @mdev: mediated device
+ * Returns u32: device generation
+ */

include/linux/virtio_mdev.h new file +10 -0
+struct virtio_mdev_device_ops {
+	/* Virtqueue ops */
+	int (*set_vq_address)(struct mdev_device *mdev,
+			      u16 idx, u64 desc_area, u64 driver_area,
+			      u64 device_area);
+	void (*set_vq_num)(struct mdev_device *mdev, u16 idx, u32 num);
+	void (*kick_vq)(struct mdev_device *mdev, u16 idx);
+	void (*set_vq_cb)(struct mdev_device *mdev, u16 idx,
+			  struct virtio_mdev_callback *cb);
+	void (*set_vq_ready)(struct mdev_device *mdev, u16 idx, bool

ready);

include/linux/virtio_mdev.h new file +15 -0
+	bool (*get_vq_ready)(struct mdev_device *mdev, u16 idx);
+	int (*set_vq_state)(struct mdev_device *mdev, u16 idx, u64 state);
+	u64 (*get_vq_state)(struct mdev_device *mdev, u16 idx);
+
+	/* Device ops */
+	u16 (*get_vq_align)(struct mdev_device *mdev);
+	u64 (*get_features)(struct mdev_device *mdev);
+	int (*set_features)(struct mdev_device *mdev, u64 features);
+	void (*set_config_cb)(struct mdev_device *mdev,
+			      struct virtio_mdev_callback *cb);
+	u16 (*get_queue_max)(struct mdev_device *mdev);
+	u32 (*get_device_id)(struct mdev_device *mdev);
+	u32 (*get_vendor_id)(struct mdev_device *mdev);
+	u8 (*get_status)(struct mdev_device *mdev);
+	void (*set_status)(struct mdev_device *mdev, u8 status);
Quoted 8 lines

include/linux/virtio_mdev.h line 142

Quoted 7 lines
include/linux/virtio_mdev.h new file +7 -0
+	void (*get_config)(struct mdev_device *mdev, unsigned int offset,
+			   void *buf, unsigned int len);
+	void (*set_config)(struct mdev_device *mdev, unsigned int offset,
+			   const void *buf, unsigned int len);
+	int (*get_version)(struct mdev_device *mdev);
+	u32 (*get_generation)(struct mdev_device *mdev);
+};

I'm not sure how stable above ops are.

It's the kernel internal API, so there's no strict requirement for this. We will export a version value for userspace for compatibility.

Quoted 4 lines

Does it make sense if defining
just two callbacks here, e.g. vq_ctrl and device_ctrl, and then let the vendor driver to handle specific ops in each category (similar to how ioctl works)?

Quoted 6 lines

My understanding is that it introduce another indirection, you still need to differ from different command, and it's less flexible than direct callback.

What's the value of doing this?

I just thought doing so may provide better compatibility to the parent driver. Even when new op is introduced, a parent driver that was developed against the old set can still be loaded in the new kernel. It just returns error when unrecognized ops are routed through vq_ctrl and device_ctrl, if the userspace doesn't favor the exposed version value. But if above ops set is pretty stable, then this comment can be ignored.

Thanks
Kevin
_______________________________________________
Virtualization mailing list
[email protected]
https://lists.linuxfoundation.org/mailman/listinfo/virtualization

author
#59 lore raw

Re: [PATCH V2 6/8] mdev: introduce virtio device and its device ops

Quoted 16 lines · Kevin
Quoted 10 lines
Quoted 2 lines
Quoted 1 line

+};

I'm not sure how stable above ops are.

It's the kernel internal API, so there's no strict requirement for this. We will export a version value for userspace for compatibility.

Quoted 4 lines

Does it make sense if defining
just two callbacks here, e.g. vq_ctrl and device_ctrl, and then let the vendor driver to handle specific ops in each category (similar to how ioctl works)?

Quoted 6 lines

My understanding is that it introduce another indirection, you still need to differ from different command, and it's less flexible than direct callback.

What's the value of doing this?

Quoted 7 lines

I just thought doing so may provide better compatibility to the parent driver. Even when new op is introduced, a parent driver that was developed against the old set can still be loaded in the new kernel. It just returns error when unrecognized ops are routed through vq_ctrl and device_ctrl, if the userspace doesn't favor the exposed version value. But if above ops set is pretty stable, then this comment can be ignored.

This is really good point, we should keep it stable as a real transport. And when there's major changes, we should advertise through version then we can provide a new set of functions.

Thanks

Quoted 3 lines

Thanks
Kevin

author
#60 lore raw

Re: [PATCH V2 6/8] mdev: introduce virtio device and its device ops

In reply to Tian, Kevin
Quoted 16 lines · Kevin
Quoted 10 lines
Quoted 2 lines
Quoted 1 line

+};

I'm not sure how stable above ops are.

It's the kernel internal API, so there's no strict requirement for this. We will export a version value for userspace for compatibility.

Quoted 4 lines

Does it make sense if defining
just two callbacks here, e.g. vq_ctrl and device_ctrl, and then let the vendor driver to handle specific ops in each category (similar to how ioctl works)?

Quoted 6 lines

My understanding is that it introduce another indirection, you still need to differ from different command, and it's less flexible than direct callback.

What's the value of doing this?

Quoted 7 lines

I just thought doing so may provide better compatibility to the parent driver. Even when new op is introduced, a parent driver that was developed against the old set can still be loaded in the new kernel. It just returns error when unrecognized ops are routed through vq_ctrl and device_ctrl, if the userspace doesn't favor the exposed version value. But if above ops set is pretty stable, then this comment can be ignored.

This is really good point, we should keep it stable as a real transport. And when there's major changes, we should advertise through version then we can provide a new set of functions.

Thanks

Quoted 3 lines

Thanks
Kevin

_______________________________________________
Virtualization mailing list
[email protected]
https://lists.linuxfoundation.org/mailman/listinfo/virtualization

author
#61 lore raw

Re: [PATCH V2 6/8] mdev: introduce virtio device and its device ops

In reply to Tian, Kevin
Quoted 44 lines · Kevin
Quoted 44 lines

From: Jason Wang [mailto:[email protected]]
Sent: Tuesday, September 24, 2019 9:54 PM

This patch implements basic support for mdev driver that supports virtio transport for kernel virtio driver.

---
  include/linux/mdev.h        |   2 +
  include/linux/virtio_mdev.h | 145
++++++++++++++++++++++++++++++++++++
  2 files changed, 147 insertions(+)

create mode 100644 include/linux/virtio_mdev.h

include/linux/mdev.h +2 -0
index 3414307311f1..73ac27b3b868 100644
@@ -126,6 +126,8 @@ struct mdev_device *mdev_from_dev(struct device
 *dev);
 
  enum {
  	MDEV_ID_VFIO = 1,
+	MDEV_ID_VIRTIO = 2,
+	MDEV_ID_VHOST = 3,
  	/* New entries must be added here */
  };
include/linux/virtio_mdev.h new file +10 -0
new file mode 100644
index 000000000000..d1a40a739266
@@ -0,0 +1,145 @@
+/* SPDX-License-Identifier: GPL-2.0-only */
+/*
+ * Virtio mediated device driver
+ *
+ * Copyright 2019, Red Hat Corp.
+ *     Author: Jason Wang <[email protected]>
+ */
+#ifndef _LINUX_VIRTIO_MDEV_H
+#define _LINUX_VIRTIO_MDEV_H
+
Quoted 8 lines

include/linux/virtio_mdev.h line 16

Quoted 6 lines
include/linux/virtio_mdev.h new file +6 -0
+#include <linux/interrupt.h>
+#include <linux/mdev.h>
+#include <uapi/linux/vhost.h>
+
+#define VIRTIO_MDEV_DEVICE_API_STRING		"virtio-mdev"
+#define VIRTIO_MDEV_VERSION 0x1

Just be curious. is this version identical to virtio spec version that below callbacks are created for, or just irrelevant?

It could be a hint but basically it's a way for userspace driver compatibility. For kernel we don't need this.

Quoted 71 lines
Quoted 71 lines
include/linux/virtio_mdev.h new file +29 -0
+
+struct virtio_mdev_callback {
+	irqreturn_t (*callback)(void *data);
+	void *private;
+};
+
+/**
+ * struct vfio_mdev_device_ops - Structure to be registered for each
+ * mdev device to register the device to virtio-mdev module.
+ *
+ * @set_vq_address:		Set the address of virtqueue
+ *				@mdev: mediated device
+ *				@idx: virtqueue index
+ *				@desc_area: address of desc area
+ *				@driver_area: address of driver area
+ *				@device_area: address of device area
+ *				Returns integer: success (0) or error (< 0)
+ * @set_vq_num:		Set the size of virtqueue
+ *				@mdev: mediated device
+ *				@idx: virtqueue index
+ *				@num: the size of virtqueue
+ * @kick_vq:			Kick the virtqueue
+ *				@mdev: mediated device
+ *				@idx: virtqueue index
+ * @set_vq_cb:			Set the interrut calback function for
+ *				a virtqueue
+ *				@mdev: mediated device
+ *				@idx: virtqueue index
+ *				@cb: virtio-mdev interrupt callback

structure
+ * @set_vq_ready: Set ready status for a virtqueue
+ * @mdev: mediated device
+ * @idx: virtqueue index
+ * @ready: ready (true) not ready(false)
+ * @get_vq_ready: Get ready status for a virtqueue
+ * @mdev: mediated device
+ * @idx: virtqueue index
+ * Returns boolean: ready (true) or not (false)
+ * @set_vq_state: Set the state for a virtqueue
+ * @mdev: mediated device
+ * @idx: virtqueue index
+ * @state: virtqueue state (last_avail_idx)
+ * Returns integer: success (0) or error (< 0)
+ * @get_vq_state: Get the state for a virtqueue
+ * @mdev: mediated device
+ * @idx: virtqueue index
+ * Returns virtqueue state (last_avail_idx)
+ * @get_vq_align: Get the virtqueue align requirement
+ * for the device
+ * @mdev: mediated device
+ * Returns virtqueue algin requirement
+ * @get_features: Get virtio features supported by the device
+ * @mdev: mediated device
+ * Returns the features support by the
+ * device
+ * @get_features: Set virtio features supported by the driver
+ * @mdev: mediated device
+ * @features: feature support by the driver
+ * Returns integer: success (0) or error (< 0)
+ * @set_config_cb: Set the config interrupt callback
+ * @mdev: mediated device
+ * @cb: virtio-mdev interrupt callback
structure
+ * @get_device_id: Get virtio device id
+ * @mdev: mediated device
+ * Returns u32: virtio device id
+ * @get_vendor_id: Get virtio vendor id
+ * @mdev: mediated device
+ * Returns u32: virtio vendor id
+ * @get_status: Get the device status

Quoted 8 lines
Quoted 7 lines

+ * @mdev: mediated device
+ * Returns u8: virtio device status
+ * @set_status: Set the device status
+ * @mdev: mediated device
+ * @status: virtio device status
+ * @get_config: Read from device specific confiugration
space

configuration (and similar typos downward)

Let me fix.

Quoted 47 lines
Quoted 47 lines

+ * @mdev: mediated device
+ * @offset: offset from the beginning of
+ * configuration space
+ * @buf: buffer used to read to
+ * @len: the length to read from
+ * configration space
+ * @set_config: Write to device specific confiugration space
+ * @mdev: mediated device
+ * @offset: offset from the beginning of
+ * configuration space
+ * @buf: buffer used to write from
+ * @len: the length to write to
+ * configration space
+ * @get_version: Get the version of virtio mdev device
+ * @mdev: mediated device
+ * Returns integer: version of the device
+ * @get_generation: Get device generaton
+ * @mdev: mediated device
+ * Returns u32: device generation
+ */

include/linux/virtio_mdev.h new file +10 -0
+struct virtio_mdev_device_ops {
+	/* Virtqueue ops */
+	int (*set_vq_address)(struct mdev_device *mdev,
+			      u16 idx, u64 desc_area, u64 driver_area,
+			      u64 device_area);
+	void (*set_vq_num)(struct mdev_device *mdev, u16 idx, u32 num);
+	void (*kick_vq)(struct mdev_device *mdev, u16 idx);
+	void (*set_vq_cb)(struct mdev_device *mdev, u16 idx,
+			  struct virtio_mdev_callback *cb);
+	void (*set_vq_ready)(struct mdev_device *mdev, u16 idx, bool

ready);

include/linux/virtio_mdev.h new file +15 -0
+	bool (*get_vq_ready)(struct mdev_device *mdev, u16 idx);
+	int (*set_vq_state)(struct mdev_device *mdev, u16 idx, u64 state);
+	u64 (*get_vq_state)(struct mdev_device *mdev, u16 idx);
+
+	/* Device ops */
+	u16 (*get_vq_align)(struct mdev_device *mdev);
+	u64 (*get_features)(struct mdev_device *mdev);
+	int (*set_features)(struct mdev_device *mdev, u64 features);
+	void (*set_config_cb)(struct mdev_device *mdev,
+			      struct virtio_mdev_callback *cb);
+	u16 (*get_queue_max)(struct mdev_device *mdev);
+	u32 (*get_device_id)(struct mdev_device *mdev);
+	u32 (*get_vendor_id)(struct mdev_device *mdev);
+	u8 (*get_status)(struct mdev_device *mdev);
+	void (*set_status)(struct mdev_device *mdev, u8 status);
Quoted 8 lines

include/linux/virtio_mdev.h line 142

Quoted 7 lines
include/linux/virtio_mdev.h new file +7 -0
+	void (*get_config)(struct mdev_device *mdev, unsigned int offset,
+			   void *buf, unsigned int len);
+	void (*set_config)(struct mdev_device *mdev, unsigned int offset,
+			   const void *buf, unsigned int len);
+	int (*get_version)(struct mdev_device *mdev);
+	u32 (*get_generation)(struct mdev_device *mdev);
+};

I'm not sure how stable above ops are.

It's the kernel internal API, so there's no strict requirement for this. We will export a version value for userspace for compatibility.

Quoted 4 lines

Does it make sense if defining
just two callbacks here, e.g. vq_ctrl and device_ctrl, and then let the vendor driver to handle specific ops in each category (similar to how ioctl works)?

My understanding is that it introduce another indirection, you still need to differ from different command, and it's less flexible than direct callback.

What's the value of doing this?

Thanks

Quoted 4 lines

Thanks
Kevin

_______________________________________________
Virtualization mailing list
[email protected]
https://lists.linuxfoundation.org/mailman/listinfo/virtualization

#62 lore raw

RE: [PATCH V2 6/8] mdev: introduce virtio device and its device ops

In reply to Jason Wang
Quoted 40 lines

From: Jason Wang [mailto:[email protected]]
Sent: Tuesday, September 24, 2019 9:54 PM

This patch implements basic support for mdev driver that supports virtio transport for kernel virtio driver.

---
 include/linux/mdev.h        |   2 +
 include/linux/virtio_mdev.h | 145
++++++++++++++++++++++++++++++++++++
 2 files changed, 147 insertions(+)
 create mode 100644 include/linux/virtio_mdev.h
include/linux/mdev.h +2 -0
index 3414307311f1..73ac27b3b868 100644
@@ -126,6 +126,8 @@ struct mdev_device *mdev_from_dev(struct device
 *dev);
 
 enum {
 	MDEV_ID_VFIO = 1,
+	MDEV_ID_VIRTIO = 2,
+	MDEV_ID_VHOST = 3,
 	/* New entries must be added here */
 };
include/linux/virtio_mdev.h new file +6 -0
new file mode 100644
index 000000000000..d1a40a739266
@@ -0,0 +1,145 @@
+/* SPDX-License-Identifier: GPL-2.0-only */
+/*
+ * Virtio mediated device driver
+ *
+ * Copyright 2019, Red Hat Corp.
+ *     Author: Jason Wang <[email protected]>

include/linux/virtio_mdev.h line 16

Quoted 10 lines
include/linux/virtio_mdev.h new file +10 -0
+ */
+#ifndef _LINUX_VIRTIO_MDEV_H
+#define _LINUX_VIRTIO_MDEV_H
+
+#include <linux/interrupt.h>
+#include <linux/mdev.h>
+#include <uapi/linux/vhost.h>
+
+#define VIRTIO_MDEV_DEVICE_API_STRING		"virtio-mdev"
+#define VIRTIO_MDEV_VERSION 0x1

Just be curious. is this version identical to virtio spec version that below callbacks are created for, or just irrelevant?

Quoted 69 lines
include/linux/virtio_mdev.h new file +29 -0
+
+struct virtio_mdev_callback {
+	irqreturn_t (*callback)(void *data);
+	void *private;
+};
+
+/**
+ * struct vfio_mdev_device_ops - Structure to be registered for each
+ * mdev device to register the device to virtio-mdev module.
+ *
+ * @set_vq_address:		Set the address of virtqueue
+ *				@mdev: mediated device
+ *				@idx: virtqueue index
+ *				@desc_area: address of desc area
+ *				@driver_area: address of driver area
+ *				@device_area: address of device area
+ *				Returns integer: success (0) or error (< 0)
+ * @set_vq_num:		Set the size of virtqueue
+ *				@mdev: mediated device
+ *				@idx: virtqueue index
+ *				@num: the size of virtqueue
+ * @kick_vq:			Kick the virtqueue
+ *				@mdev: mediated device
+ *				@idx: virtqueue index
+ * @set_vq_cb:			Set the interrut calback function for
+ *				a virtqueue
+ *				@mdev: mediated device
+ *				@idx: virtqueue index
+ *				@cb: virtio-mdev interrupt callback

structure
+ * @set_vq_ready: Set ready status for a virtqueue
+ * @mdev: mediated device
+ * @idx: virtqueue index
+ * @ready: ready (true) not ready(false)
+ * @get_vq_ready: Get ready status for a virtqueue
+ * @mdev: mediated device
+ * @idx: virtqueue index
+ * Returns boolean: ready (true) or not (false)
+ * @set_vq_state: Set the state for a virtqueue
+ * @mdev: mediated device
+ * @idx: virtqueue index
+ * @state: virtqueue state (last_avail_idx)
+ * Returns integer: success (0) or error (< 0)
+ * @get_vq_state: Get the state for a virtqueue
+ * @mdev: mediated device
+ * @idx: virtqueue index
+ * Returns virtqueue state (last_avail_idx)
+ * @get_vq_align: Get the virtqueue align requirement
+ * for the device
+ * @mdev: mediated device
+ * Returns virtqueue algin requirement
+ * @get_features: Get virtio features supported by the device
+ * @mdev: mediated device
+ * Returns the features support by the
+ * device
+ * @get_features: Set virtio features supported by the driver
+ * @mdev: mediated device
+ * @features: feature support by the driver
+ * Returns integer: success (0) or error (< 0)
+ * @set_config_cb: Set the config interrupt callback
+ * @mdev: mediated device
+ * @cb: virtio-mdev interrupt callback
structure
+ * @get_device_id: Get virtio device id
+ * @mdev: mediated device
+ * Returns u32: virtio device id
+ * @get_vendor_id: Get virtio vendor id
+ * @mdev: mediated device
+ * Returns u32: virtio vendor id

Quoted 8 lines

+ * @get_status: Get the device status
+ * @mdev: mediated device
+ * Returns u8: virtio device status
+ * @set_status: Set the device status
+ * @mdev: mediated device
+ * @status: virtio device status
+ * @get_config: Read from device specific confiugration
space

configuration (and similar typos downward)

Quoted 43 lines

+ * @mdev: mediated device
+ * @offset: offset from the beginning of
+ * configuration space
+ * @buf: buffer used to read to
+ * @len: the length to read from
+ * configration space
+ * @set_config: Write to device specific confiugration space
+ * @mdev: mediated device
+ * @offset: offset from the beginning of
+ * configuration space
+ * @buf: buffer used to write from
+ * @len: the length to write to
+ * configration space
+ * @get_version: Get the version of virtio mdev device
+ * @mdev: mediated device
+ * Returns integer: version of the device
+ * @get_generation: Get device generaton
+ * @mdev: mediated device
+ * Returns u32: device generation
+ */

include/linux/virtio_mdev.h new file +10 -0
+struct virtio_mdev_device_ops {
+	/* Virtqueue ops */
+	int (*set_vq_address)(struct mdev_device *mdev,
+			      u16 idx, u64 desc_area, u64 driver_area,
+			      u64 device_area);
+	void (*set_vq_num)(struct mdev_device *mdev, u16 idx, u32 num);
+	void (*kick_vq)(struct mdev_device *mdev, u16 idx);
+	void (*set_vq_cb)(struct mdev_device *mdev, u16 idx,
+			  struct virtio_mdev_callback *cb);
+	void (*set_vq_ready)(struct mdev_device *mdev, u16 idx, bool

ready);

include/linux/virtio_mdev.h new file +12 -0
+	bool (*get_vq_ready)(struct mdev_device *mdev, u16 idx);
+	int (*set_vq_state)(struct mdev_device *mdev, u16 idx, u64 state);
+	u64 (*get_vq_state)(struct mdev_device *mdev, u16 idx);
+
+	/* Device ops */
+	u16 (*get_vq_align)(struct mdev_device *mdev);
+	u64 (*get_features)(struct mdev_device *mdev);
+	int (*set_features)(struct mdev_device *mdev, u64 features);
+	void (*set_config_cb)(struct mdev_device *mdev,
+			      struct virtio_mdev_callback *cb);
+	u16 (*get_queue_max)(struct mdev_device *mdev);
+	u32 (*get_device_id)(struct mdev_device *mdev);

include/linux/virtio_mdev.h line 142

Quoted 10 lines
include/linux/virtio_mdev.h new file +10 -0
+	u32 (*get_vendor_id)(struct mdev_device *mdev);
+	u8 (*get_status)(struct mdev_device *mdev);
+	void (*set_status)(struct mdev_device *mdev, u8 status);
+	void (*get_config)(struct mdev_device *mdev, unsigned int offset,
+			   void *buf, unsigned int len);
+	void (*set_config)(struct mdev_device *mdev, unsigned int offset,
+			   const void *buf, unsigned int len);
+	int (*get_version)(struct mdev_device *mdev);
+	u32 (*get_generation)(struct mdev_device *mdev);
+};

I'm not sure how stable above ops are. Does it make sense if defining just two callbacks here, e.g. vq_ctrl and device_ctrl, and then let the vendor driver to handle specific ops in each category (similar to how ioctl works)?

Thanks
Kevin

author
#63 lore raw

PATCH v2 6/8 mdev: introduce virtio device and its device ops

In reply to Jason Wang

This patch implements basic support for mdev driver that supports virtio transport for kernel virtio driver.

---
 include/linux/mdev.h        |   2 +
 include/linux/virtio_mdev.h | 145 ++++++++++++++++++++++++++++++++++++
 2 files changed, 147 insertions(+)
 create mode 100644 include/linux/virtio_mdev.h
include/linux/mdev.h +2 -0
index 3414307311f1..73ac27b3b868 100644
@@ -126,6 +126,8 @@ struct mdev_device *mdev_from_dev(struct device *dev);
 
 enum {
 	MDEV_ID_VFIO = 1,
+	MDEV_ID_VIRTIO = 2,
+	MDEV_ID_VHOST = 3,
 	/* New entries must be added here */
 };
 
include/linux/virtio_mdev.h new file +145 -0

145 lines not on this page · raw message

Signature · 1 line
2.19.1
author
#64 lore raw

PATCH v2 7/8 virtio: introduce a mdev based transport

In reply to Jason Wang

This patch introduces a new mdev transport for virtio. This is used to use kernel virtio driver to drive the mediated device that is capable of populating virtqueue directly.

A new virtio-mdev driver will be registered to the mdev bus, when a new virtio-mdev device is probed, it will register the device with mdev based config ops. This means it is a software transport between mdev driver and mdev device. The transport was implemented through device specific opswhich is a part of mdev_parent_ops now.

---
 drivers/virtio/Kconfig       |   7 +
 drivers/virtio/Makefile      |   1 +
 drivers/virtio/virtio_mdev.c | 417 +++++++++++++++++++++++++++++++++++
 3 files changed, 425 insertions(+)
 create mode 100644 drivers/virtio/virtio_mdev.c
drivers/virtio/Kconfig +7 -0
index 078615cf2afc..8d18722ab572 100644
@@ -43,6 +43,13 @@ config VIRTIO_PCI_LEGACY
 
 	  If unsure, say Y.
 
+config VIRTIO_MDEV_DEVICE
+	tristate "VIRTIO driver for Mediated devices"
+	depends on VFIO_MDEV && VIRTIO
+	default n
+	help
+	  VIRTIO based driver for Mediated devices.
+
 config VIRTIO_PMEM
 	tristate "Support for virtio pmem driver"
 	depends on VIRTIO
drivers/virtio/Makefile +1 -0
index 3a2b5c5dcf46..ebc7fa15ae82 100644
@@ -6,3 +6,4 @@ virtio_pci-y := virtio_pci_modern.o virtio_pci_common.o
 virtio_pci-$(CONFIG_VIRTIO_PCI_LEGACY) += virtio_pci_legacy.o
 obj-$(CONFIG_VIRTIO_BALLOON) += virtio_balloon.o
 obj-$(CONFIG_VIRTIO_INPUT) += virtio_input.o
+obj-$(CONFIG_VIRTIO_MDEV_DEVICE) += virtio_mdev.o
drivers/virtio/virtio_mdev.c new file +417 -0

417 lines not on this page · raw message

Signature · 6 lines
2.19.1

_______________________________________________
Intel-gfx mailing list
[email protected]
https://lists.freedesktop.org/mailman/listinfo/intel-gfx
author
#65 lore raw

PATCH v2 7/8 virtio: introduce a mdev based transport

In reply to Jason Wang

This patch introduces a new mdev transport for virtio. This is used to use kernel virtio driver to drive the mediated device that is capable of populating virtqueue directly.

A new virtio-mdev driver will be registered to the mdev bus, when a new virtio-mdev device is probed, it will register the device with mdev based config ops. This means it is a software transport between mdev driver and mdev device. The transport was implemented through device specific opswhich is a part of mdev_parent_ops now.

---
 drivers/virtio/Kconfig       |   7 +
 drivers/virtio/Makefile      |   1 +
 drivers/virtio/virtio_mdev.c | 417 +++++++++++++++++++++++++++++++++++
 3 files changed, 425 insertions(+)
 create mode 100644 drivers/virtio/virtio_mdev.c
drivers/virtio/Kconfig +7 -0
index 078615cf2afc..8d18722ab572 100644
@@ -43,6 +43,13 @@ config VIRTIO_PCI_LEGACY
 
 	  If unsure, say Y.
 
+config VIRTIO_MDEV_DEVICE
+	tristate "VIRTIO driver for Mediated devices"
+	depends on VFIO_MDEV && VIRTIO
+	default n
+	help
+	  VIRTIO based driver for Mediated devices.
+
 config VIRTIO_PMEM
 	tristate "Support for virtio pmem driver"
 	depends on VIRTIO
drivers/virtio/Makefile +1 -0
index 3a2b5c5dcf46..ebc7fa15ae82 100644
@@ -6,3 +6,4 @@ virtio_pci-y := virtio_pci_modern.o virtio_pci_common.o
 virtio_pci-$(CONFIG_VIRTIO_PCI_LEGACY) += virtio_pci_legacy.o
 obj-$(CONFIG_VIRTIO_BALLOON) += virtio_balloon.o
 obj-$(CONFIG_VIRTIO_INPUT) += virtio_input.o
+obj-$(CONFIG_VIRTIO_MDEV_DEVICE) += virtio_mdev.o
drivers/virtio/virtio_mdev.c new file +417 -0

417 lines not on this page · raw message

Signature · 1 line
2.19.1
author
#66 lore raw

PATCH v2 8/8 docs: sample driver to demonstrate how to implement virtio-mdev framework

In reply to Jason Wang

This sample driver creates mdev device that simulate virtio net device over virtio mdev transport. The device is implemented through vringh and workqueue. A device specific dma ops is to make sure HVA is used directly as the IOVA. This should be sufficient for kernel virtio driver to work.

Only 'virtio' type is supported right now. I plan to add 'vhost' type on top which requires some virtual IOMMU implemented in this sample driver.

---
 MAINTAINERS                |   1 +
 samples/Kconfig            |   7 +
 samples/vfio-mdev/Makefile |   1 +
 samples/vfio-mdev/mvnet.c  | 692 +++++++++++++++++++++++++++++++++++++
 4 files changed, 701 insertions(+)
 create mode 100644 samples/vfio-mdev/mvnet.c
MAINTAINERS +1 -0
index 89832b316500..347dc64cdc23 100644
@@ -17202,6 +17202,7 @@ F:	include/linux/virtio*.h
 F:	include/uapi/linux/virtio_*.h
 F:	drivers/crypto/virtio/
 F:	mm/balloon_compaction.c
+F:	samples/vfio-mdev/mvnet.c
 
 VIRTIO BLOCK AND SCSI DRIVERS
 M:	"Michael S. Tsirkin" <[email protected]>
samples/Kconfig +7 -0
index c8dacb4dda80..a1a1ca2c00b7 100644
@@ -131,6 +131,13 @@ config SAMPLE_VFIO_MDEV_MDPY
 	  mediated device.  It is a simple framebuffer and supports
 	  the region display interface (VFIO_GFX_PLANE_TYPE_REGION).
 
+config SAMPLE_VIRTIO_MDEV_NET
+        tristate "Build virtio mdev net example mediated device sample code -- loadable modules only"
+	depends on VIRTIO_MDEV_DEVICE && VHOST_RING && m
+	help
+	  Build a networking sample device for use as a virtio
+	  mediated device.
+
 config SAMPLE_VFIO_MDEV_MDPY_FB
 	tristate "Build VFIO mdpy example guest fbdev driver -- loadable module only"
 	depends on FB && m
samples/vfio-mdev/Makefile +1 -0
index 10d179c4fdeb..f34af90ed0a0 100644
@@ -3,3 +3,4 @@ obj-$(CONFIG_SAMPLE_VFIO_MDEV_MTTY) += mtty.o
 obj-$(CONFIG_SAMPLE_VFIO_MDEV_MDPY) += mdpy.o
 obj-$(CONFIG_SAMPLE_VFIO_MDEV_MDPY_FB) += mdpy-fb.o
 obj-$(CONFIG_SAMPLE_VFIO_MDEV_MBOCHS) += mbochs.o
+obj-$(CONFIG_SAMPLE_VIRTIO_MDEV_NET) += mvnet.o
samples/vfio-mdev/mvnet.c new file +692 -0

692 lines not on this page · raw message

Signature · 1 line
2.19.1
author
#67 lore raw

PATCH v2 8/8 docs: sample driver to demonstrate how to implement virtio-mdev framework

In reply to Jason Wang

This sample driver creates mdev device that simulate virtio net device over virtio mdev transport. The device is implemented through vringh and workqueue. A device specific dma ops is to make sure HVA is used directly as the IOVA. This should be sufficient for kernel virtio driver to work.

Only 'virtio' type is supported right now. I plan to add 'vhost' type on top which requires some virtual IOMMU implemented in this sample driver.

---
 MAINTAINERS                |   1 +
 samples/Kconfig            |   7 +
 samples/vfio-mdev/Makefile |   1 +
 samples/vfio-mdev/mvnet.c  | 692 +++++++++++++++++++++++++++++++++++++
 4 files changed, 701 insertions(+)
 create mode 100644 samples/vfio-mdev/mvnet.c
MAINTAINERS +1 -0
index 89832b316500..347dc64cdc23 100644
@@ -17202,6 +17202,7 @@ F:	include/linux/virtio*.h
 F:	include/uapi/linux/virtio_*.h
 F:	drivers/crypto/virtio/
 F:	mm/balloon_compaction.c
+F:	samples/vfio-mdev/mvnet.c
 
 VIRTIO BLOCK AND SCSI DRIVERS
 M:	"Michael S. Tsirkin" <[email protected]>
samples/Kconfig +7 -0
index c8dacb4dda80..a1a1ca2c00b7 100644
@@ -131,6 +131,13 @@ config SAMPLE_VFIO_MDEV_MDPY
 	  mediated device.  It is a simple framebuffer and supports
 	  the region display interface (VFIO_GFX_PLANE_TYPE_REGION).
 
+config SAMPLE_VIRTIO_MDEV_NET
+        tristate "Build virtio mdev net example mediated device sample code -- loadable modules only"
+	depends on VIRTIO_MDEV_DEVICE && VHOST_RING && m
+	help
+	  Build a networking sample device for use as a virtio
+	  mediated device.
+
 config SAMPLE_VFIO_MDEV_MDPY_FB
 	tristate "Build VFIO mdpy example guest fbdev driver -- loadable module only"
 	depends on FB && m
samples/vfio-mdev/Makefile +1 -0
index 10d179c4fdeb..f34af90ed0a0 100644
@@ -3,3 +3,4 @@ obj-$(CONFIG_SAMPLE_VFIO_MDEV_MTTY) += mtty.o
 obj-$(CONFIG_SAMPLE_VFIO_MDEV_MDPY) += mdpy.o
 obj-$(CONFIG_SAMPLE_VFIO_MDEV_MDPY_FB) += mdpy-fb.o
 obj-$(CONFIG_SAMPLE_VFIO_MDEV_MBOCHS) += mbochs.o
+obj-$(CONFIG_SAMPLE_VIRTIO_MDEV_NET) += mvnet.o
samples/vfio-mdev/mvnet.c new file +692 -0

692 lines not on this page · raw message

Signature · 1 line
2.19.1
#68 lore raw

✗ Fi.CI.CHECKPATCH: warning for mdev based hardware virtio offloading support (rev3)

In reply to Jason Wang
== Series Details ==

== Series Details ==

Series: mdev based hardware virtio offloading support (rev3)
URL : https://patchwork.freedesktop.org/series/66989/
State : warning

== Summary ==

$ dim checkpatch origin/drm-tip
6346e474d4a5 vringh: fix copy direction of vringh_iov_push_kern()
1f34e53afc28 mdev: class id support
3c4ddc86a519 mdev: bus uevent support
eb320bc12b94 modpost: add support for mdev class id
96bc50618f39 mdev: introduce device specific ops
-:422: WARNING:FILE_PATH_CHANGES: added, moved or deleted file(s), does MAINTAINERS need updating?
#422: 
new file mode 100644

total: 0 errors, 1 warnings, 0 checks, 542 lines checked
7645552e98b2 mdev: introduce virtio device and its device ops
-:25: WARNING:FILE_PATH_CHANGES: added, moved or deleted file(s), does MAINTAINERS need updating? #25:
new file mode 100644

total: 0 errors, 1 warnings, 0 checks, 153 lines checked
6a3638fcef65 virtio: introduce a mdev based transport
-:26: WARNING:CONFIG_DESCRIPTION: please write a paragraph that describes the config symbol fully #26: FILE: drivers/virtio/Kconfig:46:
+config VIRTIO_MDEV_DEVICE

-:46: WARNING:FILE_PATH_CHANGES: added, moved or deleted file(s), does MAINTAINERS need updating? #46:
new file mode 100644

-:106: WARNING:UNSPECIFIED_INT: Prefer 'unsigned int' to bare use of 'unsigned' #106: FILE: drivers/virtio/virtio_mdev.c:56:
+static void virtio_mdev_get(struct virtio_device *vdev, unsigned offset,

-:107: WARNING:UNSPECIFIED_INT: Prefer 'unsigned int' to bare use of 'unsigned' #107: FILE: drivers/virtio/virtio_mdev.c:57:
+ void *buf, unsigned len)

-:115: WARNING:UNSPECIFIED_INT: Prefer 'unsigned int' to bare use of 'unsigned' #115: FILE: drivers/virtio/virtio_mdev.c:65:
+static void virtio_mdev_set(struct virtio_device *vdev, unsigned offset,

-:116: WARNING:UNSPECIFIED_INT: Prefer 'unsigned int' to bare use of 'unsigned' #116: FILE: drivers/virtio/virtio_mdev.c:66:
+ const void *buf, unsigned len)

-:183: WARNING:UNSPECIFIED_INT: Prefer 'unsigned int' to bare use of 'unsigned' #183: FILE: drivers/virtio/virtio_mdev.c:133:
+virtio_mdev_setup_vq(struct virtio_device *vdev, unsigned index,

-:262: CHECK:BRACES: Blank lines aren't necessary before a close brace '}' #262: FILE: drivers/virtio/virtio_mdev.c:212:
+
+}

-:294: WARNING:UNSPECIFIED_INT: Prefer 'unsigned int' to bare use of 'unsigned' #294: FILE: drivers/virtio/virtio_mdev.c:244:
+static int virtio_mdev_find_vqs(struct virtio_device *vdev, unsigned nvqs,

total: 0 errors, 8 warnings, 1 checks, 434 lines checked
d4e1d237c373 docs: sample driver to demonstrate how to implement virtio-mdev framework -:39: WARNING:CONFIG_DESCRIPTION: please write a paragraph that describes the config symbol fully #39: FILE: samples/Kconfig:134:
+config SAMPLE_VIRTIO_MDEV_NET

-:59: WARNING:FILE_PATH_CHANGES: added, moved or deleted file(s), does MAINTAINERS need updating? #59:
new file mode 100644

-:147: CHECK:UNCOMMENTED_DEFINITION: spinlock_t definition without comment #147: FILE: samples/vfio-mdev/mvnet.c:84:
+ spinlock_t lock;

-:180: CHECK:PARENTHESIS_ALIGNMENT: Alignment should match open parenthesis #180: FILE: samples/vfio-mdev/mvnet.c:117:
+ vringh_init_kern(&vq->vring, mvnet_features, MVNET_QUEUE_MAX,
+ false, 0, 0, 0);

-:285: CHECK:COMPARISON_TO_NULL: Comparison to NULL could be written "!addr" #285: FILE: samples/vfio-mdev/mvnet.c:222:
+ if (addr == NULL)

-:288: CHECK:SPACING: No space is necessary after a cast #288: FILE: samples/vfio-mdev/mvnet.c:225:
+ *dma_addr = (dma_addr_t) addr;

-:318: CHECK:ALLOC_SIZEOF_STRUCT: Prefer kzalloc(sizeof(*mvnet)...) over kzalloc(sizeof(struct mvnet_state)...) #318: FILE: samples/vfio-mdev/mvnet.c:255:
+ mvnet = kzalloc(sizeof(struct mvnet_state), GFP_KERNEL);

-:319: CHECK:COMPARISON_TO_NULL: Comparison to NULL could be written "!mvnet" #319: FILE: samples/vfio-mdev/mvnet.c:256:
+ if (mvnet == NULL)

-:376: CHECK:PARENTHESIS_ALIGNMENT: Alignment should match open parenthesis #376: FILE: samples/vfio-mdev/mvnet.c:313:
+sample_mvnet_dev_show(struct device *dev, struct device_attribute *attr,
+ char *buf)

total: 0 errors, 2 warnings, 7 checks, 716 lines checked

_______________________________________________
Intel-gfx mailing list
[email protected]
https://lists.freedesktop.org/mailman/listinfo/intel-gfx

#69 lore raw

✓ Fi.CI.BAT: success for mdev based hardware virtio offloading support (rev3)

In reply to Jason Wang
== Series Details ==

== Series Details ==

Series: mdev based hardware virtio offloading support (rev3)
URL : https://patchwork.freedesktop.org/series/66989/
State : success

== Summary ==

CI Bug Log - changes from CI_DRM_6948 -> Patchwork_14516
====================================================

Summary
-------

**SUCCESS**

No regressions found.

External URL: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_14516/

Known issues
------------

Here are the changes found in Patchwork_14516 that come from known issues:

### IGT changes ###

#### Issues hit ####

* igt@kms_chamelium@hdmi-hpd-fast:
- fi-kbl-7500u: [PASS][1] -> [FAIL][2] ([fdo#111407])
[1]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_6948/fi-kbl-7500u/igt@[email protected]
[2]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_14516/fi-kbl-7500u/igt@[email protected]

#### Possible fixes ####

* igt@gem_exec_suspend@basic-s4-devices:
- fi-blb-e6850: [INCOMPLETE][3] ([fdo#107718]) -> [PASS][4]
[3]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_6948/fi-blb-e6850/igt@[email protected]
[4]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_14516/fi-blb-e6850/igt@[email protected]

{name}: This element is suppressed. This means it is ignored when computing
the status of the difference (SUCCESS, WARNING, or FAILURE).

[fdo#107718]: https://bugs.freedesktop.org/show_bug.cgi?id=107718
[fdo#110566]: https://bugs.freedesktop.org/show_bug.cgi?id=110566
[fdo#111407]: https://bugs.freedesktop.org/show_bug.cgi?id=111407
[fdo#111735]: https://bugs.freedesktop.org/show_bug.cgi?id=111735

Participating hosts (52 -> 46)
------------------------------

Additional (1): fi-icl-dsi
Missing (7): fi-ilk-m540 fi-hsw-4200u fi-byt-squawks fi-bsw-cyan fi-icl-y fi-byt-clapper fi-bdw-samus

Build changes
-------------

* CI: CI-20190529 -> None
* Linux: CI_DRM_6948 -> Patchwork_14516

CI-20190529: 20190529
CI_DRM_6948: 485ca160d8ffac7ffb5be5e76fa12ad46a7e5a19 @ git://anongit.freedesktop.org/gfx-ci/linux
IGT_5201: 3c1633abec14679300d52eeaf9fb7b63e435e51e @ git://anongit.freedesktop.org/xorg/app/intel-gpu-tools
Patchwork_14516: d4e1d237c373a6058e1094b6edd2095e78c01c6c @ git://anongit.freedesktop.org/gfx-ci/linux

== Linux commits ==

d4e1d237c373 docs: sample driver to demonstrate how to implement virtio-mdev framework 6a3638fcef65 virtio: introduce a mdev based transport
7645552e98b2 mdev: introduce virtio device and its device ops 96bc50618f39 mdev: introduce device specific ops
eb320bc12b94 modpost: add support for mdev class id
3c4ddc86a519 mdev: bus uevent support
1f34e53afc28 mdev: class id support
6346e474d4a5 vringh: fix copy direction of vringh_iov_push_kern()

== Logs ==

For more details see: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_14516/index.html
_______________________________________________
Intel-gfx mailing list
[email protected]
https://lists.freedesktop.org/mailman/listinfo/intel-gfx

#70 lore raw

✓ Fi.CI.IGT: success for mdev based hardware virtio offloading support (rev3)

In reply to Jason Wang
== Series Details ==

== Series Details ==

Series: mdev based hardware virtio offloading support (rev3)
URL : https://patchwork.freedesktop.org/series/66989/
State : success

== Summary ==

CI Bug Log - changes from CI_DRM_6948_full -> Patchwork_14516_full
====================================================

Summary
-------

**SUCCESS**

No regressions found.

Known issues
------------

Here are the changes found in Patchwork_14516_full that come from known issues:

### IGT changes ###

#### Issues hit ####

* igt@gem_ctx_isolation@vcs1-clean:
- shard-iclb: [PASS][1] -> [SKIP][2] ([fdo#109276]) +7 similar issues
[1]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_6948/shard-iclb1/igt@[email protected]
[2]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_14516/shard-iclb3/igt@[email protected]

* igt@gem_ctx_shared@exec-single-timeline-bsd:
- shard-iclb: [PASS][3] -> [SKIP][4] ([fdo#110841])
[3]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_6948/shard-iclb8/igt@[email protected]
[4]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_14516/shard-iclb4/igt@[email protected]

* igt@gem_exec_schedule@preemptive-hang-bsd:
- shard-iclb: [PASS][5] -> [SKIP][6] ([fdo#111325]) +3 similar issues
[5]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_6948/shard-iclb8/igt@[email protected]
[6]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_14516/shard-iclb4/igt@[email protected]

* igt@i915_pm_rps@waitboost:
- shard-glk: [PASS][7] -> [FAIL][8] ([fdo#102250])
[7]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_6948/shard-glk3/igt@[email protected]
[8]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_14516/shard-glk9/igt@[email protected]

* igt@i915_selftest@live_hangcheck:
- shard-iclb: [PASS][9] -> [INCOMPLETE][10] ([fdo#107713] / [fdo#108569])
[9]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_6948/shard-iclb5/igt@i915_selftest@live_hangcheck.html
[10]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_14516/shard-iclb2/igt@i915_selftest@live_hangcheck.html

* igt@i915_suspend@forcewake:
- shard-skl: [PASS][11] -> [INCOMPLETE][12] ([fdo#104108])
[11]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_6948/shard-skl9/igt@[email protected]
[12]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_14516/shard-skl9/igt@[email protected]

* igt@kms_cursor_crc@pipe-a-cursor-suspend:
- shard-skl: [PASS][13] -> [INCOMPLETE][14] ([fdo#110741])
[13]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_6948/shard-skl9/igt@[email protected]
[14]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_14516/shard-skl4/igt@[email protected]

* igt@kms_flip@flip-vs-expired-vblank-interruptible:
- shard-glk: [PASS][15] -> [FAIL][16] ([fdo#105363])
[15]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_6948/shard-glk5/igt@[email protected]
[16]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_14516/shard-glk5/igt@[email protected]

* igt@kms_flip@flip-vs-suspend-interruptible:
- shard-skl: [PASS][17] -> [INCOMPLETE][18] ([fdo#109507])
[17]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_6948/shard-skl6/igt@[email protected]
[18]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_14516/shard-skl7/igt@[email protected]

* igt@kms_frontbuffer_tracking@fbc-1p-primscrn-pri-shrfb-draw-mmap-gtt:
- shard-iclb: [PASS][19] -> [INCOMPLETE][20] ([fdo#107713])
[19]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_6948/shard-iclb4/igt@kms_frontbuffer_tracking@fbc-1p-primscrn-pri-shrfb-draw-mmap-gtt.html
[20]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_14516/shard-iclb7/igt@kms_frontbuffer_tracking@fbc-1p-primscrn-pri-shrfb-draw-mmap-gtt.html

* igt@kms_frontbuffer_tracking@fbc-1p-primscrn-spr-indfb-draw-blt:
- shard-iclb: [PASS][21] -> [FAIL][22] ([fdo#103167]) +5 similar issues
[21]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_6948/shard-iclb3/igt@[email protected]
[22]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_14516/shard-iclb8/igt@[email protected]

* igt@kms_plane@plane-panning-bottom-right-suspend-pipe-b-planes:
- shard-apl: [PASS][23] -> [DMESG-WARN][24] ([fdo#108566]) +4 similar issues
[23]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_6948/shard-apl1/igt@[email protected]
[24]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_14516/shard-apl2/igt@[email protected]
- shard-kbl: [PASS][25] -> [INCOMPLETE][26] ([fdo#103665])
[25]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_6948/shard-kbl7/igt@[email protected]
[26]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_14516/shard-kbl1/igt@[email protected]

* igt@kms_plane_alpha_blend@pipe-a-constant-alpha-min:
- shard-skl: [PASS][27] -> [FAIL][28] ([fdo#108145])
[27]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_6948/shard-skl4/igt@[email protected]
[28]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_14516/shard-skl8/igt@[email protected]

* igt@kms_psr@psr2_cursor_plane_move:
- shard-iclb: [PASS][29] -> [SKIP][30] ([fdo#109441]) +1 similar issue
[29]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_6948/shard-iclb2/igt@kms_psr@psr2_cursor_plane_move.html
[30]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_14516/shard-iclb4/igt@kms_psr@psr2_cursor_plane_move.html

* igt@kms_setmode@basic:
- shard-apl: [PASS][31] -> [FAIL][32] ([fdo#99912])
[31]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_6948/shard-apl1/igt@[email protected]
[32]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_14516/shard-apl1/igt@[email protected]

#### Possible fixes ####

* igt@gem_ctx_isolation@rcs0-s3:
- shard-apl: [DMESG-WARN][33] ([fdo#108566]) -> [PASS][34] +3 similar issues
[33]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_6948/shard-apl2/igt@[email protected]
[34]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_14516/shard-apl3/igt@[email protected]

* igt@gem_ctx_isolation@vecs0-s3:
- shard-skl: [INCOMPLETE][35] ([fdo#104108]) -> [PASS][36]
[35]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_6948/shard-skl9/igt@[email protected]
[36]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_14516/shard-skl9/igt@[email protected]

* igt@gem_ctx_switch@rcs0-heavy-queue:
- shard-iclb: [INCOMPLETE][37] ([fdo#107713]) -> [PASS][38]
[37]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_6948/shard-iclb7/igt@[email protected]
[38]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_14516/shard-iclb6/igt@[email protected]

* igt@gem_eio@reset-stress:
- shard-snb: [FAIL][39] ([fdo#109661]) -> [PASS][40]
[39]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_6948/shard-snb1/igt@[email protected]
[40]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_14516/shard-snb2/igt@[email protected]

* igt@gem_exec_schedule@wide-bsd:
- shard-iclb: [SKIP][41] ([fdo#111325]) -> [PASS][42] +4 similar issues
[41]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_6948/shard-iclb1/igt@[email protected]
[42]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_14516/shard-iclb3/igt@[email protected]

* igt@gem_pipe_control_store_loop@reused-buffer:
- shard-apl: [INCOMPLETE][43] ([fdo#103927]) -> [PASS][44] +1 similar issue
[43]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_6948/shard-apl6/igt@[email protected]
[44]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_14516/shard-apl3/igt@[email protected]

* igt@gem_tiled_swapping@non-threaded:
- shard-apl: [DMESG-WARN][45] ([fdo#108686]) -> [PASS][46]
[45]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_6948/shard-apl8/igt@[email protected]
[46]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_14516/shard-apl4/igt@[email protected]

* {igt@i915_pm_dc@dc5-dpms}:
- shard-iclb: [FAIL][47] ([fdo#111795 ]) -> [PASS][48]
[47]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_6948/shard-iclb3/igt@[email protected]
[48]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_14516/shard-iclb8/igt@[email protected]

* igt@kms_flip@flip-vs-suspend-interruptible:
- shard-hsw: [INCOMPLETE][49] ([fdo#103540]) -> [PASS][50]
[49]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_6948/shard-hsw5/igt@[email protected]
[50]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_14516/shard-hsw5/igt@[email protected]

* igt@kms_frontbuffer_tracking@fbc-1p-primscrn-spr-indfb-draw-render:
- shard-iclb: [FAIL][51] ([fdo#103167]) -> [PASS][52]
[51]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_6948/shard-iclb2/igt@kms_frontbuffer_tracking@fbc-1p-primscrn-spr-indfb-draw-render.html
[52]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_14516/shard-iclb4/igt@kms_frontbuffer_tracking@fbc-1p-primscrn-spr-indfb-draw-render.html

* igt@kms_frontbuffer_tracking@fbcpsr-slowdraw:
- shard-iclb: [INCOMPLETE][53] ([fdo#106978] / [fdo#107713]) -> [PASS][54]
[53]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_6948/shard-iclb7/igt@[email protected]
[54]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_14516/shard-iclb5/igt@[email protected]

* igt@kms_psr2_su@frontbuffer:
- shard-iclb: [SKIP][55] ([fdo#109642] / [fdo#111068]) -> [PASS][56]
[55]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_6948/shard-iclb4/igt@[email protected]
[56]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_14516/shard-iclb2/igt@[email protected]

* igt@kms_psr@psr2_cursor_plane_onoff:
- shard-iclb: [SKIP][57] ([fdo#109441]) -> [PASS][58]
[57]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_6948/shard-iclb6/igt@kms_psr@psr2_cursor_plane_onoff.html
[58]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_14516/shard-iclb2/igt@kms_psr@psr2_cursor_plane_onoff.html

* igt@kms_setmode@basic:
- shard-glk: [FAIL][59] ([fdo#99912]) -> [PASS][60]
[59]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_6948/shard-glk4/igt@[email protected]
[60]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_14516/shard-glk6/igt@[email protected]

* igt@kms_vblank@pipe-a-ts-continuation-dpms-suspend:
- shard-kbl: [INCOMPLETE][61] ([fdo#103665]) -> [PASS][62]
[61]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_6948/shard-kbl6/igt@[email protected]
[62]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_14516/shard-kbl7/igt@[email protected]

* igt@perf@polling:
- shard-skl: [FAIL][63] ([fdo#110728]) -> [PASS][64]
[63]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_6948/shard-skl6/igt@[email protected]
[64]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_14516/shard-skl10/igt@[email protected]

* igt@prime_vgem@fence-wait-bsd2:
- shard-iclb: [SKIP][65] ([fdo#109276]) -> [PASS][66] +15 similar issues
[65]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_6948/shard-iclb8/igt@[email protected]
[66]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_14516/shard-iclb4/igt@[email protected]

#### Warnings ####

* igt@gem_mocs_settings@mocs-reset-bsd2:
- shard-iclb: [SKIP][67] ([fdo#109276]) -> [FAIL][68] ([fdo#111330])
[67]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_6948/shard-iclb6/igt@[email protected]
[68]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_14516/shard-iclb4/igt@[email protected]

* igt@kms_cursor_crc@pipe-a-cursor-suspend:
- shard-apl: [INCOMPLETE][69] ([fdo#103927]) -> [DMESG-WARN][70] ([fdo#108566])
[69]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_6948/shard-apl1/igt@[email protected]
[70]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_14516/shard-apl7/igt@[email protected]

{name}: This element is suppressed. This means it is ignored when computing
the status of the difference (SUCCESS, WARNING, or FAILURE).

[fdo#102250]: https://bugs.freedesktop.org/show_bug.cgi?id=102250
[fdo#103167]: https://bugs.freedesktop.org/show_bug.cgi?id=103167
[fdo#103540]: https://bugs.freedesktop.org/show_bug.cgi?id=103540
[fdo#103665]: https://bugs.freedesktop.org/show_bug.cgi?id=103665
[fdo#103927]: https://bugs.freedesktop.org/show_bug.cgi?id=103927
[fdo#104108]: https://bugs.freedesktop.org/show_bug.cgi?id=104108
[fdo#105363]: https://bugs.freedesktop.org/show_bug.cgi?id=105363
[fdo#106978]: https://bugs.freedesktop.org/show_bug.cgi?id=106978
[fdo#107713]: https://bugs.freedesktop.org/show_bug.cgi?id=107713
[fdo#108145]: https://bugs.freedesktop.org/show_bug.cgi?id=108145
[fdo#108566]: https://bugs.freedesktop.org/show_bug.cgi?id=108566
[fdo#108569]: https://bugs.freedesktop.org/show_bug.cgi?id=108569
[fdo#108686]: https://bugs.freedesktop.org/show_bug.cgi?id=108686
[fdo#109276]: https://bugs.freedesktop.org/show_bug.cgi?id=109276
[fdo#109441]: https://bugs.freedesktop.org/show_bug.cgi?id=109441
[fdo#109507]: https://bugs.freedesktop.org/show_bug.cgi?id=109507
[fdo#109642]: https://bugs.freedesktop.org/show_bug.cgi?id=109642
[fdo#109661]: https://bugs.freedesktop.org/show_bug.cgi?id=109661
[fdo#110548]: https://bugs.freedesktop.org/show_bug.cgi?id=110548
[fdo#110728]: https://bugs.freedesktop.org/show_bug.cgi?id=110728
[fdo#110741]: https://bugs.freedesktop.org/show_bug.cgi?id=110741
[fdo#110841]: https://bugs.freedesktop.org/show_bug.cgi?id=110841
[fdo#111068]: https://bugs.freedesktop.org/show_bug.cgi?id=111068
[fdo#111325]: https://bugs.freedesktop.org/show_bug.cgi?id=111325
[fdo#111330]: https://bugs.freedesktop.org/show_bug.cgi?id=111330
[fdo#111795 ]: https://bugs.freedesktop.org/show_bug.cgi?id=111795
[fdo#99912]: https://bugs.freedesktop.org/show_bug.cgi?id=99912

Participating hosts (9 -> 9)
------------------------------

No changes in participating hosts

Build changes
-------------

* CI: CI-20190529 -> None
* Linux: CI_DRM_6948 -> Patchwork_14516

CI-20190529: 20190529
CI_DRM_6948: 485ca160d8ffac7ffb5be5e76fa12ad46a7e5a19 @ git://anongit.freedesktop.org/gfx-ci/linux
IGT_5201: 3c1633abec14679300d52eeaf9fb7b63e435e51e @ git://anongit.freedesktop.org/xorg/app/intel-gpu-tools
Patchwork_14516: d4e1d237c373a6058e1094b6edd2095e78c01c6c @ git://anongit.freedesktop.org/gfx-ci/linux
piglit_4509: fdc5a4ca11124ab8413c7988896eec4c97336694 @ git://anongit.freedesktop.org/piglit

== Logs ==

For more details see: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_14516/
_______________________________________________
Intel-gfx mailing list
[email protected]
https://lists.freedesktop.org/mailman/listinfo/intel-gfx

#71 lore raw
In reply to Jason Wang
Quoted 18 lines

From: Jason Wang [mailto:[email protected]]
Sent: Tuesday, September 24, 2019 9:53 PM

Hi all:

There are hardware that can do virtio datapath offloading while having its own control path. This path tries to implement a mdev based unified API to support using kernel virtio driver to drive those devices. This is done by introducing a new mdev transport for virtio (virtio_mdev) and register itself as a new kind of mdev driver. Then it provides a unified way for kernel virtio driver to talk with mdev device implementation.

Though the series only contains kernel driver support, the goal is to make the transport generic enough to support userspace drivers. This means vhost-mdev[1] could be built on top as well by resuing the transport.

Quoted 8 lines

A sample driver is also implemented which simulate a virito-net loopback ethernet device on top of vringh + workqueue. This could be used as a reference implementation for real hardware driver.

Consider mdev framework only support VFIO device and driver right now, this series also extend it to support other types. This is done through introducing class id to the device and pairing it with id_talbe claimed by the driver. On top, this seris also decouple

id_table claimed ... this series ...

Quoted 84 lines

device specific parents ops out of the common ones.

Pktgen test was done with virito-net + mvnet loop back device.

Please review.

[1] https://lkml.org/lkml/2019/9/16/869

Changes from V1:

- move virtio_mdev.c to drivers/virtio
- store class_id in mdev_device instead of mdev_parent
- store device_ops in mdev_device instead of mdev_parent
- reorder the patch, vringh fix comes first
- really silent compiling warnings
- really switch to use u16 for class_id
- uevent and modpost support for mdev class_id
- vraious tweaks per comments from Parav

Changes from RFC-V2:

- silent compile warnings on some specific configuration
- use u16 instead u8 for class id
- reseve MDEV_ID_VHOST for future vhost-mdev work
- introduce "virtio" type for mvnet and make "vhost" type for future work
- add entries in MAINTAINER
- tweak and typos fixes in commit log

Changes from RFC-V1:

- rename device id to class id
- add docs for class id and device specific ops (device_ops)
- split device_ops into seperate headers
- drop the mdev_set_dma_ops()
- use device_ops to implement the transport API, then it's not a part of UAPI any more
- use GFP_ATOMIC in mvnet sample device and other tweaks
- set_vring_base/get_vring_base support for mvnet device

Jason Wang (8):
vringh: fix copy direction of vringh_iov_push_kern()
mdev: class id support
mdev: bus uevent support
modpost: add support for mdev class id
mdev: introduce device specific ops
mdev: introduce virtio device and its device ops
virtio: introduce a mdev based transport
docs: sample driver to demonstrate how to implement virtio-mdev
framework

 .../driver-api/vfio-mediated-device.rst       |   7 +-
 MAINTAINERS                                   |   2 +
 drivers/gpu/drm/i915/gvt/kvmgt.c              |  18 +-
 drivers/s390/cio/vfio_ccw_ops.c               |  18 +-
 drivers/s390/crypto/vfio_ap_ops.c             |  14 +-
 drivers/vfio/mdev/mdev_core.c                 |  19 +
 drivers/vfio/mdev/mdev_driver.c               |  22 +
 drivers/vfio/mdev/mdev_private.h              |   2 +
 drivers/vfio/mdev/vfio_mdev.c                 |  45 +-
 drivers/vhost/vringh.c                        |   8 +-
 drivers/virtio/Kconfig                        |   7 +
 drivers/virtio/Makefile                       |   1 +
 drivers/virtio/virtio_mdev.c                  | 417 +++++++++++
 include/linux/mdev.h                          |  52 +-
 include/linux/mod_devicetable.h               |   8 +
 include/linux/vfio_mdev.h                     |  52 ++
 include/linux/virtio_mdev.h                   | 145 ++++
 samples/Kconfig                               |   7 +
 samples/vfio-mdev/Makefile                    |   1 +
 samples/vfio-mdev/mbochs.c                    |  20 +-
 samples/vfio-mdev/mdpy.c                      |  20 +-
 samples/vfio-mdev/mtty.c                      |  18 +-
 samples/vfio-mdev/mvnet.c                     | 692 ++++++++++++++++++
 scripts/mod/devicetable-offsets.c             |   3 +
 scripts/mod/file2alias.c                      |  10 +
 25 files changed, 1524 insertions(+), 84 deletions(-)
 create mode 100644 drivers/virtio/virtio_mdev.c
 create mode 100644 include/linux/vfio_mdev.h
 create mode 100644 include/linux/virtio_mdev.h
 create mode 100644 samples/vfio-mdev/mvnet.c

--
2.19.1

author
#72 lore raw
Quoted 22 lines · Kevin
Quoted 22 lines

From: Jason Wang [mailto:[email protected]]
Sent: Tuesday, September 24, 2019 9:53 PM

Hi all:

There are hardware that can do virtio datapath offloading while having its own control path. This path tries to implement a mdev based unified API to support using kernel virtio driver to drive those devices. This is done by introducing a new mdev transport for virtio (virtio_mdev) and register itself as a new kind of mdev driver. Then it provides a unified way for kernel virtio driver to talk with mdev device implementation.

Though the series only contains kernel driver support, the goal is to make the transport generic enough to support userspace drivers. This means vhost-mdev[1] could be built on top as well by resuing the transport.

A sample driver is also implemented which simulate a virito-net loopback ethernet device on top of vringh + workqueue. This could be used as a reference implementation for real hardware driver.

Quoted 5 lines
Quoted 4 lines

Consider mdev framework only support VFIO device and driver right now, this series also extend it to support other types. This is done through introducing class id to the device and pairing it with id_talbe claimed by the driver. On top, this seris also decouple

id_table claimed ... this series ...

Let me fix in V3.

Thanks

Quoted 85 lines
Quoted 85 lines

device specific parents ops out of the common ones.

Pktgen test was done with virito-net + mvnet loop back device.

Please review.

[1] https://lkml.org/lkml/2019/9/16/869

Changes from V1:

- move virtio_mdev.c to drivers/virtio
- store class_id in mdev_device instead of mdev_parent
- store device_ops in mdev_device instead of mdev_parent
- reorder the patch, vringh fix comes first
- really silent compiling warnings
- really switch to use u16 for class_id
- uevent and modpost support for mdev class_id
- vraious tweaks per comments from Parav

Changes from RFC-V2:

- silent compile warnings on some specific configuration
- use u16 instead u8 for class id
- reseve MDEV_ID_VHOST for future vhost-mdev work
- introduce "virtio" type for mvnet and make "vhost" type for future work
- add entries in MAINTAINER
- tweak and typos fixes in commit log

Changes from RFC-V1:

- rename device id to class id
- add docs for class id and device specific ops (device_ops)
- split device_ops into seperate headers
- drop the mdev_set_dma_ops()
- use device_ops to implement the transport API, then it's not a part of UAPI any more
- use GFP_ATOMIC in mvnet sample device and other tweaks
- set_vring_base/get_vring_base support for mvnet device

Jason Wang (8):
vringh: fix copy direction of vringh_iov_push_kern()
mdev: class id support
mdev: bus uevent support
modpost: add support for mdev class id
mdev: introduce device specific ops
mdev: introduce virtio device and its device ops
virtio: introduce a mdev based transport
docs: sample driver to demonstrate how to implement virtio-mdev
framework

 .../driver-api/vfio-mediated-device.rst       |   7 +-
 MAINTAINERS                                   |   2 +
 drivers/gpu/drm/i915/gvt/kvmgt.c              |  18 +-
 drivers/s390/cio/vfio_ccw_ops.c               |  18 +-
 drivers/s390/crypto/vfio_ap_ops.c             |  14 +-
 drivers/vfio/mdev/mdev_core.c                 |  19 +
 drivers/vfio/mdev/mdev_driver.c               |  22 +
 drivers/vfio/mdev/mdev_private.h              |   2 +
 drivers/vfio/mdev/vfio_mdev.c                 |  45 +-
 drivers/vhost/vringh.c                        |   8 +-
 drivers/virtio/Kconfig                        |   7 +
 drivers/virtio/Makefile                       |   1 +
 drivers/virtio/virtio_mdev.c                  | 417 +++++++++++
 include/linux/mdev.h                          |  52 +-
 include/linux/mod_devicetable.h               |   8 +
 include/linux/vfio_mdev.h                     |  52 ++
 include/linux/virtio_mdev.h                   | 145 ++++
 samples/Kconfig                               |   7 +
 samples/vfio-mdev/Makefile                    |   1 +
 samples/vfio-mdev/mbochs.c                    |  20 +-
 samples/vfio-mdev/mdpy.c                      |  20 +-
 samples/vfio-mdev/mtty.c                      |  18 +-
 samples/vfio-mdev/mvnet.c                     | 692 ++++++++++++++++++
 scripts/mod/devicetable-offsets.c             |   3 +
 scripts/mod/file2alias.c                      |  10 +
 25 files changed, 1524 insertions(+), 84 deletions(-)
 create mode 100644 drivers/virtio/virtio_mdev.c
 create mode 100644 include/linux/vfio_mdev.h
 create mode 100644 include/linux/virtio_mdev.h
 create mode 100644 samples/vfio-mdev/mvnet.c

--
2.19.1

author
#73 lore raw
In reply to Tian, Kevin
Quoted 22 lines · Kevin
Quoted 22 lines

From: Jason Wang [mailto:[email protected]]
Sent: Tuesday, September 24, 2019 9:53 PM

Hi all:

There are hardware that can do virtio datapath offloading while having its own control path. This path tries to implement a mdev based unified API to support using kernel virtio driver to drive those devices. This is done by introducing a new mdev transport for virtio (virtio_mdev) and register itself as a new kind of mdev driver. Then it provides a unified way for kernel virtio driver to talk with mdev device implementation.

Though the series only contains kernel driver support, the goal is to make the transport generic enough to support userspace drivers. This means vhost-mdev[1] could be built on top as well by resuing the transport.

A sample driver is also implemented which simulate a virito-net loopback ethernet device on top of vringh + workqueue. This could be used as a reference implementation for real hardware driver.

Quoted 5 lines
Quoted 4 lines

Consider mdev framework only support VFIO device and driver right now, this series also extend it to support other types. This is done through introducing class id to the device and pairing it with id_talbe claimed by the driver. On top, this seris also decouple

id_table claimed ... this series ...

Let me fix in V3.

Thanks

Quoted 83 lines
Quoted 83 lines

device specific parents ops out of the common ones.

Pktgen test was done with virito-net + mvnet loop back device.

Please review.

[1] https://lkml.org/lkml/2019/9/16/869

Changes from V1:

- move virtio_mdev.c to drivers/virtio
- store class_id in mdev_device instead of mdev_parent
- store device_ops in mdev_device instead of mdev_parent
- reorder the patch, vringh fix comes first
- really silent compiling warnings
- really switch to use u16 for class_id
- uevent and modpost support for mdev class_id
- vraious tweaks per comments from Parav

Changes from RFC-V2:

- silent compile warnings on some specific configuration
- use u16 instead u8 for class id
- reseve MDEV_ID_VHOST for future vhost-mdev work
- introduce "virtio" type for mvnet and make "vhost" type for future work
- add entries in MAINTAINER
- tweak and typos fixes in commit log

Changes from RFC-V1:

- rename device id to class id
- add docs for class id and device specific ops (device_ops)
- split device_ops into seperate headers
- drop the mdev_set_dma_ops()
- use device_ops to implement the transport API, then it's not a part of UAPI any more
- use GFP_ATOMIC in mvnet sample device and other tweaks
- set_vring_base/get_vring_base support for mvnet device

Jason Wang (8):
vringh: fix copy direction of vringh_iov_push_kern()
mdev: class id support
mdev: bus uevent support
modpost: add support for mdev class id
mdev: introduce device specific ops
mdev: introduce virtio device and its device ops
virtio: introduce a mdev based transport
docs: sample driver to demonstrate how to implement virtio-mdev
framework

 .../driver-api/vfio-mediated-device.rst       |   7 +-
 MAINTAINERS                                   |   2 +
 drivers/gpu/drm/i915/gvt/kvmgt.c              |  18 +-
 drivers/s390/cio/vfio_ccw_ops.c               |  18 +-
 drivers/s390/crypto/vfio_ap_ops.c             |  14 +-
 drivers/vfio/mdev/mdev_core.c                 |  19 +
 drivers/vfio/mdev/mdev_driver.c               |  22 +
 drivers/vfio/mdev/mdev_private.h              |   2 +
 drivers/vfio/mdev/vfio_mdev.c                 |  45 +-
 drivers/vhost/vringh.c                        |   8 +-
 drivers/virtio/Kconfig                        |   7 +
 drivers/virtio/Makefile                       |   1 +
 drivers/virtio/virtio_mdev.c                  | 417 +++++++++++
 include/linux/mdev.h                          |  52 +-
 include/linux/mod_devicetable.h               |   8 +
 include/linux/vfio_mdev.h                     |  52 ++
 include/linux/virtio_mdev.h                   | 145 ++++
 samples/Kconfig                               |   7 +
 samples/vfio-mdev/Makefile                    |   1 +
 samples/vfio-mdev/mbochs.c                    |  20 +-
 samples/vfio-mdev/mdpy.c                      |  20 +-
 samples/vfio-mdev/mtty.c                      |  18 +-
 samples/vfio-mdev/mvnet.c                     | 692 ++++++++++++++++++
 scripts/mod/devicetable-offsets.c             |   3 +
 scripts/mod/file2alias.c                      |  10 +
 25 files changed, 1524 insertions(+), 84 deletions(-)
 create mode 100644 drivers/virtio/virtio_mdev.c
 create mode 100644 include/linux/vfio_mdev.h
 create mode 100644 include/linux/virtio_mdev.h
 create mode 100644 samples/vfio-mdev/mvnet.c
Quoted 2 lines
Quoted 2 lines

--
2.19.1

_______________________________________________
Virtualization mailing list
[email protected]
https://lists.linuxfoundation.org/mailman/listinfo/virtualization

#74 lore raw
In reply to Jason Wang
Quoted 18 lines

From: Jason Wang [mailto:[email protected]]
Sent: Tuesday, September 24, 2019 9:53 PM

Hi all:

There are hardware that can do virtio datapath offloading while having its own control path. This path tries to implement a mdev based unified API to support using kernel virtio driver to drive those devices. This is done by introducing a new mdev transport for virtio (virtio_mdev) and register itself as a new kind of mdev driver. Then it provides a unified way for kernel virtio driver to talk with mdev device implementation.

Though the series only contains kernel driver support, the goal is to make the transport generic enough to support userspace drivers. This means vhost-mdev[1] could be built on top as well by resuing the transport.

Quoted 8 lines

A sample driver is also implemented which simulate a virito-net loopback ethernet device on top of vringh + workqueue. This could be used as a reference implementation for real hardware driver.

Consider mdev framework only support VFIO device and driver right now, this series also extend it to support other types. This is done through introducing class id to the device and pairing it with id_talbe claimed by the driver. On top, this seris also decouple

id_table claimed ... this series ...

Quoted 84 lines

device specific parents ops out of the common ones.

Pktgen test was done with virito-net + mvnet loop back device.

Please review.

[1] https://lkml.org/lkml/2019/9/16/869

Changes from V1:

- move virtio_mdev.c to drivers/virtio
- store class_id in mdev_device instead of mdev_parent
- store device_ops in mdev_device instead of mdev_parent
- reorder the patch, vringh fix comes first
- really silent compiling warnings
- really switch to use u16 for class_id
- uevent and modpost support for mdev class_id
- vraious tweaks per comments from Parav

Changes from RFC-V2:

- silent compile warnings on some specific configuration
- use u16 instead u8 for class id
- reseve MDEV_ID_VHOST for future vhost-mdev work
- introduce "virtio" type for mvnet and make "vhost" type for future work
- add entries in MAINTAINER
- tweak and typos fixes in commit log

Changes from RFC-V1:

- rename device id to class id
- add docs for class id and device specific ops (device_ops)
- split device_ops into seperate headers
- drop the mdev_set_dma_ops()
- use device_ops to implement the transport API, then it's not a part of UAPI any more
- use GFP_ATOMIC in mvnet sample device and other tweaks
- set_vring_base/get_vring_base support for mvnet device

Jason Wang (8):
vringh: fix copy direction of vringh_iov_push_kern()
mdev: class id support
mdev: bus uevent support
modpost: add support for mdev class id
mdev: introduce device specific ops
mdev: introduce virtio device and its device ops
virtio: introduce a mdev based transport
docs: sample driver to demonstrate how to implement virtio-mdev
framework

 .../driver-api/vfio-mediated-device.rst       |   7 +-
 MAINTAINERS                                   |   2 +
 drivers/gpu/drm/i915/gvt/kvmgt.c              |  18 +-
 drivers/s390/cio/vfio_ccw_ops.c               |  18 +-
 drivers/s390/crypto/vfio_ap_ops.c             |  14 +-
 drivers/vfio/mdev/mdev_core.c                 |  19 +
 drivers/vfio/mdev/mdev_driver.c               |  22 +
 drivers/vfio/mdev/mdev_private.h              |   2 +
 drivers/vfio/mdev/vfio_mdev.c                 |  45 +-
 drivers/vhost/vringh.c                        |   8 +-
 drivers/virtio/Kconfig                        |   7 +
 drivers/virtio/Makefile                       |   1 +
 drivers/virtio/virtio_mdev.c                  | 417 +++++++++++
 include/linux/mdev.h                          |  52 +-
 include/linux/mod_devicetable.h               |   8 +
 include/linux/vfio_mdev.h                     |  52 ++
 include/linux/virtio_mdev.h                   | 145 ++++
 samples/Kconfig                               |   7 +
 samples/vfio-mdev/Makefile                    |   1 +
 samples/vfio-mdev/mbochs.c                    |  20 +-
 samples/vfio-mdev/mdpy.c                      |  20 +-
 samples/vfio-mdev/mtty.c                      |  18 +-
 samples/vfio-mdev/mvnet.c                     | 692 ++++++++++++++++++
 scripts/mod/devicetable-offsets.c             |   3 +
 scripts/mod/file2alias.c                      |  10 +
 25 files changed, 1524 insertions(+), 84 deletions(-)
 create mode 100644 drivers/virtio/virtio_mdev.c
 create mode 100644 include/linux/vfio_mdev.h
 create mode 100644 include/linux/virtio_mdev.h
 create mode 100644 samples/vfio-mdev/mvnet.c

--
2.19.1