Skip to contentSkip to conversation

vduse: Add suspend

PATCH Superseded Eugenio Pérez 2 participants 4 messages Mar 10 – Mar 13 virtio core

v2 of 5 Superseded Latest is v5 →

lore
Share & tools
lore

Eugenio Pérez

author maintainer
#0 lore raw

PATCH v2 1/1 vduse: Add suspend

Implement suspend operation for vduse devices, so vhost-vdpa will offer that backend feature and userspace can effectively suspend the device.

This is a must before get virtqueue indexes (base) for live migration, since the device could modify them after userland gets them.

---
This patch depends on
https://lore.kernel.org/lkml/[email protected]

v2:
* Take the rwsem only before the actual kick, not in vduse_vdpa_kick_vq. This assures that we're not in a critical section.

---
 drivers/vdpa/vdpa_user/vduse_dev.c | 86 +++++++++++++++++++++++++++++-
 include/uapi/linux/vduse.h         |  4 ++
 2 files changed, 88 insertions(+), 2 deletions(-)
drivers/vdpa/vdpa_user/vduse_dev.c +84 -2
index 4f642b95a7cb..f56b1e3eb82d 100644
@@ -54,7 +54,8 @@
 #define IRQ_UNBOUND -1
 
 /* Supported VDUSE features */
-static const uint64_t vduse_features = BIT_U64(VDUSE_F_QUEUE_READY);
+static const uint64_t vduse_features = BIT_U64(VDUSE_F_QUEUE_READY) |
+				       BIT_U64(VDUSE_F_SUSPEND);
 
 /*
  * VDUSE instance have not asked the vduse API version, so assume 0.
@@ -85,6 +86,7 @@ struct vduse_virtqueue {
 	int irq_effective_cpu;
 	struct cpumask irq_affinity;
 	struct kobject kobj;
+	struct vduse_dev *dev;
 };
 
 struct vduse_dev;
@@ -134,6 +136,7 @@ struct vduse_dev {
 	int minor;
 	bool broken;
 	bool connected;
+	bool suspended;
 	u64 api_version;
 	u64 device_features;
 	u64 driver_features;
@@ -480,6 +483,7 @@ static void vduse_dev_reset(struct vduse_dev *dev)
 
 	down_write(&dev->rwsem);
 
+	dev->suspended = false;
 	dev->status = 0;
 	dev->driver_features = 0;
 	dev->generation++;
@@ -538,6 +542,10 @@ static void vduse_vq_kick(struct vduse_virtqueue *vq)
 	if (!vq->ready)
 		goto unlock;
 
+	guard(rwsem_read)(&vq->dev->rwsem);
+	if (vq->dev->suspended)
+		return;
+
 	if (vq->kickfd)
 		eventfd_signal(vq->kickfd);
 	else
@@ -896,6 +904,27 @@ static int vduse_vdpa_set_map(struct vdpa_device *vdpa,
 	return 0;
 }
 
+static int vduse_vdpa_suspend(struct vdpa_device *vdpa)
+{
+	struct vduse_dev *dev = vdpa_to_vduse(vdpa);
+	struct vduse_dev_msg msg = { 0 };
+	int ret;
+
+	msg.req.type = VDUSE_SUSPEND;
+
+	ret = vduse_dev_msg_sync(dev, &msg);
+	if (ret == 0) {
+		scoped_guard(rwsem_write, &dev->rwsem)
+			dev->suspended = true;
+
+		cancel_work_sync(&dev->inject);
+		for (u32 i = 0; i < dev->vq_num; i++)
+			cancel_work_sync(&dev->vqs[i]->inject);
+	}
+
+	return ret;
+}
+
 static void vduse_vdpa_free(struct vdpa_device *vdpa)
 {
 	struct vduse_dev *dev = vdpa_to_vduse(vdpa);
@@ -937,6 +966,41 @@ static const struct vdpa_config_ops vduse_vdpa_config_ops = {
 	.free			= vduse_vdpa_free,
 };
 
+static const struct vdpa_config_ops vduse_vdpa_config_ops_with_suspend = {
+	.set_vq_address		= vduse_vdpa_set_vq_address,
+	.kick_vq		= vduse_vdpa_kick_vq,
+	.set_vq_cb		= vduse_vdpa_set_vq_cb,
+	.set_vq_num             = vduse_vdpa_set_vq_num,
+	.get_vq_size		= vduse_vdpa_get_vq_size,
+	.get_vq_group		= vduse_get_vq_group,
+	.set_vq_ready		= vduse_vdpa_set_vq_ready,
+	.get_vq_ready		= vduse_vdpa_get_vq_ready,
+	.set_vq_state		= vduse_vdpa_set_vq_state,
+	.get_vq_state		= vduse_vdpa_get_vq_state,
+	.get_vq_align		= vduse_vdpa_get_vq_align,
+	.get_device_features	= vduse_vdpa_get_device_features,
+	.set_driver_features	= vduse_vdpa_set_driver_features,
+	.get_driver_features	= vduse_vdpa_get_driver_features,
+	.set_config_cb		= vduse_vdpa_set_config_cb,
+	.get_vq_num_max		= vduse_vdpa_get_vq_num_max,
+	.get_device_id		= vduse_vdpa_get_device_id,
+	.get_vendor_id		= vduse_vdpa_get_vendor_id,
+	.get_status		= vduse_vdpa_get_status,
+	.set_status		= vduse_vdpa_set_status,
+	.get_config_size	= vduse_vdpa_get_config_size,
+	.get_config		= vduse_vdpa_get_config,
+	.set_config		= vduse_vdpa_set_config,
+	.get_generation		= vduse_vdpa_get_generation,
+	.set_vq_affinity	= vduse_vdpa_set_vq_affinity,
+	.get_vq_affinity	= vduse_vdpa_get_vq_affinity,
+	.reset			= vduse_vdpa_reset,
+	.set_map		= vduse_vdpa_set_map,
+	.set_group_asid		= vduse_set_group_asid,
+	.get_vq_map		= vduse_get_vq_map,
+	.suspend		= vduse_vdpa_suspend,
+	.free			= vduse_vdpa_free,
+};
+
 static void vduse_dev_sync_single_for_device(union virtio_map token,
 					     dma_addr_t dma_addr, size_t size,
 					     enum dma_data_direction dir)
@@ -1148,6 +1212,10 @@ static void vduse_dev_irq_inject(struct work_struct *work)
 {
 	struct vduse_dev *dev = container_of(work, struct vduse_dev, inject);
 
+	guard(rwsem_read)(&dev->rwsem);
+	if (dev->suspended)
+		return;
+
 	spin_lock_bh(&dev->irq_lock);
 	if (dev->config_cb.callback)
 		dev->config_cb.callback(dev->config_cb.private);
@@ -1159,6 +1227,10 @@ static void vduse_vq_irq_inject(struct work_struct *work)
 	struct vduse_virtqueue *vq = container_of(work,
 					struct vduse_virtqueue, inject);
 
+	guard(rwsem_read)(&vq->dev->rwsem);
+	if (vq->dev->suspended)
+		return;
+
 	spin_lock_bh(&vq->irq_lock);
 	if (vq->ready && vq->cb.callback)
 		vq->cb.callback(vq->cb.private);
@@ -1189,6 +1261,9 @@ static int vduse_dev_queue_irq_work(struct vduse_dev *dev,
 	int ret = -EINVAL;
 
 	down_read(&dev->rwsem);
+	if (dev->suspended)
+		return ret;
+
 	if (!(dev->status & VIRTIO_CONFIG_S_DRIVER_OK))
 		goto unlock;
 
@@ -1839,6 +1914,7 @@ static int vduse_dev_init_vqs(struct vduse_dev *dev, u32 vq_align, u32 vq_num)
 		}
 
 		dev->vqs[i]->index = i;
+		dev->vqs[i]->dev = dev;
 		dev->vqs[i]->irq_effective_cpu = IRQ_UNBOUND;
 		INIT_WORK(&dev->vqs[i]->inject, vduse_vq_irq_inject);
 		INIT_WORK(&dev->vqs[i]->kick, vduse_vq_kick_work);
@@ -2290,12 +2366,18 @@ static struct vduse_mgmt_dev *vduse_mgmt;
 static int vduse_dev_init_vdpa(struct vduse_dev *dev, const char *name)
 {
 	struct vduse_vdpa *vdev;
+	const struct vdpa_config_ops *ops;
 
 	if (dev->vdev)
 		return -EEXIST;
 
+	if (dev->vduse_features & BIT_U64(VDUSE_F_SUSPEND))
+		ops = &vduse_vdpa_config_ops_with_suspend;
+	else
+		ops = &vduse_vdpa_config_ops;
+
 	vdev = vdpa_alloc_device(struct vduse_vdpa, vdpa, dev->dev,
-				 &vduse_vdpa_config_ops, &vduse_map_ops,
+				 ops, &vduse_map_ops,
 				 dev->ngroups, dev->nas, name, true);
 	if (IS_ERR(vdev))
 		return PTR_ERR(vdev);
include/uapi/linux/vduse.h +4 -0
index 7324faea5df4..8c616895c511 100644
@@ -17,6 +17,9 @@
 /* The VDUSE instance expects a request for vq ready */
 #define VDUSE_F_QUEUE_READY	0
 
+/* The VDUSE instance expects a request for suspend */
+#define VDUSE_F_SUSPEND		1
+
 /*
  * Get the version of VDUSE API that kernel supported (VDUSE_API_VERSION).
  * This is used for future extension.
@@ -334,6 +337,7 @@ enum vduse_req_type {
 	VDUSE_UPDATE_IOTLB,
 	VDUSE_SET_VQ_GROUP_ASID,
 	VDUSE_SET_VQ_READY,
+	VDUSE_SUSPEND,
 };
 
 /**
Signature · 1 line
2.53.0
#1 lore raw
Quoted 6 lines · Eugenio Pérez

Implement suspend operation for vduse devices, so vhost-vdpa will offer that backend feature and userspace can effectively suspend the device.

This is a must before get virtqueue indexes (base) for live migration, since the device could modify them after userland gets them.

As discussed in the pervious version, let's explain why and the plan to support resume.

Quoted 53 lines

---
This patch depends on
https://lore.kernel.org/lkml/[email protected]

v2:
* Take the rwsem only before the actual kick, not in vduse_vdpa_kick_vq. This assures that we're not in a critical section.

---
 drivers/vdpa/vdpa_user/vduse_dev.c | 86 +++++++++++++++++++++++++++++-
 include/uapi/linux/vduse.h         |  4 ++
 2 files changed, 88 insertions(+), 2 deletions(-)
drivers/vdpa/vdpa_user/vduse_dev.c +5 -1
index 4f642b95a7cb..f56b1e3eb82d 100644
@@ -54,7 +54,8 @@
 #define IRQ_UNBOUND -1
 
 /* Supported VDUSE features */
-static const uint64_t vduse_features = BIT_U64(VDUSE_F_QUEUE_READY);
+static const uint64_t vduse_features = BIT_U64(VDUSE_F_QUEUE_READY) |
+                                      BIT_U64(VDUSE_F_SUSPEND);
 
 /*
  * VDUSE instance have not asked the vduse API version, so assume 0.
@@ -85,6 +86,7 @@ struct vduse_virtqueue {
        int irq_effective_cpu;
        struct cpumask irq_affinity;
        struct kobject kobj;
+       struct vduse_dev *dev;
 };
 
 struct vduse_dev;
@@ -134,6 +136,7 @@ struct vduse_dev {
        int minor;
        bool broken;
        bool connected;
+       bool suspended;
        u64 api_version;
        u64 device_features;
        u64 driver_features;
@@ -480,6 +483,7 @@ static void vduse_dev_reset(struct vduse_dev *dev)
 
        down_write(&dev->rwsem);
 
+       dev->suspended = false;
        dev->status = 0;
        dev->driver_features = 0;
        dev->generation++;

drivers/vdpa/vdpa_user/vduse_dev.c static void vduse_vq_kick(struct vduse_virtqueue *vq) · line 547

Quoted 6 lines
drivers/vdpa/vdpa_user/vduse_dev.c +3 -0
@@ -538,6 +542,10 @@ static void vduse_vq_kick(struct vduse_virtqueue *vq)
        if (!vq->ready)
                goto unlock;
 
+       guard(rwsem_read)(&vq->dev->rwsem);
+       if (vq->dev->suspended)
+               return;

Any reason for this? E.g We don't do this for other transports.

Everything else looks good.

Thanks

author maintainer
#2 lore raw
Quoted 12 lines · Jason Wang
Quoted 6 lines · Eugenio Pérez

Implement suspend operation for vduse devices, so vhost-vdpa will offer that backend feature and userspace can effectively suspend the device.

This is a must before get virtqueue indexes (base) for live migration, since the device could modify them after userland gets them.

As discussed in the pervious version, let's explain why and the plan to support resume.

Is it ok to explain it just in the patch message or do you prefer something like a /* TODO: resume op */ before the suspend callback?

Quoted 56 lines
Quoted 56 lines

---
This patch depends on
https://lore.kernel.org/lkml/[email protected]

v2:
* Take the rwsem only before the actual kick, not in vduse_vdpa_kick_vq. This assures that we're not in a critical section.

---
 drivers/vdpa/vdpa_user/vduse_dev.c | 86 +++++++++++++++++++++++++++++-
 include/uapi/linux/vduse.h         |  4 ++
 2 files changed, 88 insertions(+), 2 deletions(-)
drivers/vdpa/vdpa_user/vduse_dev.c +5 -1
index 4f642b95a7cb..f56b1e3eb82d 100644
@@ -54,7 +54,8 @@
 #define IRQ_UNBOUND -1
 
 /* Supported VDUSE features */
-static const uint64_t vduse_features = BIT_U64(VDUSE_F_QUEUE_READY);
+static const uint64_t vduse_features = BIT_U64(VDUSE_F_QUEUE_READY) |
+                                      BIT_U64(VDUSE_F_SUSPEND);
 
 /*
  * VDUSE instance have not asked the vduse API version, so assume 0.
@@ -85,6 +86,7 @@ struct vduse_virtqueue {
        int irq_effective_cpu;
        struct cpumask irq_affinity;
        struct kobject kobj;
+       struct vduse_dev *dev;
 };
 
 struct vduse_dev;
@@ -134,6 +136,7 @@ struct vduse_dev {
        int minor;
        bool broken;
        bool connected;
+       bool suspended;
        u64 api_version;
        u64 device_features;
        u64 driver_features;
@@ -480,6 +483,7 @@ static void vduse_dev_reset(struct vduse_dev *dev)
 
        down_write(&dev->rwsem);
 
+       dev->suspended = false;
        dev->status = 0;
        dev->driver_features = 0;
        dev->generation++;
@@ -538,6 +542,10 @@ static void vduse_vq_kick(struct vduse_virtqueue *vq)
        if (!vq->ready)
                goto unlock;
 
Quoted 6 lines

drivers/vdpa/vdpa_user/vduse_dev.c static void vduse_vq_kick(struct vduse_virtqueue *vq) · line 547

Quoted 3 lines
drivers/vdpa/vdpa_user/vduse_dev.c +3 -0
+       guard(rwsem_read)(&vq->dev->rwsem);
+       if (vq->dev->suspended)
+               return;

Any reason for this? E.g We don't do this for other transports.

It's just a hardening but I'm happy to remove it.

Quoted 4 lines

Everything else looks good.

Thanks

#3 lore raw
Quoted 11 lines · Eugenio Perez Martin
Quoted 9 lines · Jason Wang
Quoted 6 lines · Eugenio Pérez

Implement suspend operation for vduse devices, so vhost-vdpa will offer that backend feature and userspace can effectively suspend the device.

This is a must before get virtqueue indexes (base) for live migration, since the device could modify them after userland gets them.

Quoted 6 lines
Quoted 3 lines

As discussed in the pervious version, let's explain why and the plan to support resume.

Is it ok to explain it just in the patch message or do you prefer something like a /* TODO: resume op */ before the suspend callback?

Maybe, but I wonder why lacking resume can make thing like migration work (e.g when migration fails, we should resume the device, or it's workaround by reset + DRIVER_OK?).

Quoted 57 lines
Quoted 57 lines
Quoted 57 lines

---
This patch depends on
https://lore.kernel.org/lkml/[email protected]

v2:
* Take the rwsem only before the actual kick, not in vduse_vdpa_kick_vq. This assures that we're not in a critical section.

---
 drivers/vdpa/vdpa_user/vduse_dev.c | 86 +++++++++++++++++++++++++++++-
 include/uapi/linux/vduse.h         |  4 ++
 2 files changed, 88 insertions(+), 2 deletions(-)
drivers/vdpa/vdpa_user/vduse_dev.c +5 -1
index 4f642b95a7cb..f56b1e3eb82d 100644
@@ -54,7 +54,8 @@
 #define IRQ_UNBOUND -1
 
 /* Supported VDUSE features */
-static const uint64_t vduse_features = BIT_U64(VDUSE_F_QUEUE_READY);
+static const uint64_t vduse_features = BIT_U64(VDUSE_F_QUEUE_READY) |
+                                      BIT_U64(VDUSE_F_SUSPEND);
 
 /*
  * VDUSE instance have not asked the vduse API version, so assume 0.
@@ -85,6 +86,7 @@ struct vduse_virtqueue {
        int irq_effective_cpu;
        struct cpumask irq_affinity;
        struct kobject kobj;
+       struct vduse_dev *dev;
 };
 
 struct vduse_dev;
@@ -134,6 +136,7 @@ struct vduse_dev {
        int minor;
        bool broken;
        bool connected;
+       bool suspended;
        u64 api_version;
        u64 device_features;
        u64 driver_features;
@@ -480,6 +483,7 @@ static void vduse_dev_reset(struct vduse_dev *dev)
 
        down_write(&dev->rwsem);
 
+       dev->suspended = false;
        dev->status = 0;
        dev->driver_features = 0;
        dev->generation++;
@@ -538,6 +542,10 @@ static void vduse_vq_kick(struct vduse_virtqueue *vq)
        if (!vq->ready)
                goto unlock;
 
Quoted 8 lines
Quoted 6 lines

drivers/vdpa/vdpa_user/vduse_dev.c static void vduse_vq_kick(struct vduse_virtqueue *vq) · line 547

Quoted 3 lines
drivers/vdpa/vdpa_user/vduse_dev.c +3 -0
+       guard(rwsem_read)(&vq->dev->rwsem);
+       if (vq->dev->suspended)
+               return;

Any reason for this? E.g We don't do this for other transports.

It's just a hardening but I'm happy to remove it.

Either should be fine.

Quoted 6 lines
Quoted 5 lines

Everything else looks good.

Thanks

Thanks