vfio/mlx5: Introduce device transitions of PRE_COPY

In order to support PRE_COPY, mlx5 driver is transferring multiple
states (images) of the device. e.g.: the source VF can save and transfer
multiple states, and the target VF will load them by that order.

The device is saving three kinds of states:
1) Initial state - when the device moves to PRE_COPY state.
2) Middle state - during PRE_COPY phase via VFIO_MIG_GET_PRECOPY_INFO.
   There can be multiple states of this type.
3) Final state - when the device moves to STOP_COPY state.

After moving to PRE_COPY state, user is holding the saving migf FD and
can use it. For example: user can start transferring data via read()
callback. Also, user can switch from PRE_COPY to STOP_COPY whenever he
sees it fits. This will invoke saving of final state.

This means that mlx5 VFIO device can be switched to STOP_COPY without
transferring any data in PRE_COPY state. Therefore, when the device
moves to STOP_COPY, mlx5 will store the final state on a dedicated queue
entry on the list.

Co-developed-by: Shay Drory <shayd@nvidia.com>
Signed-off-by: Shay Drory <shayd@nvidia.com>
Reviewed-by: Jason Gunthorpe <jgg@nvidia.com>
Signed-off-by: Yishai Hadas <yishaih@nvidia.com>
Link: https://lore.kernel.org/r/20221206083438.37807-9-yishaih@nvidia.com
Signed-off-by: Alex Williamson <alex.williamson@redhat.com>

1 files changed, 82 insertions, 8 deletions

diff --git a/drivers/vfio/pci/mlx5/main.c b/drivers/vfio/pci/mlx5/main.c
index ca16425811c4..9cabba456044 100644
--- a/drivers/vfio/pci/mlx5/main.c
+++ b/drivers/vfio/pci/mlx5/main.c
@@ -195,6 +195,7 @@ static ssize_t mlx5vf_buf_read(struct mlx5_vhca_data_buffer *vhca_buf,
 	if (*pos >= vhca_buf->start_pos + vhca_buf->length) {
 		spin_lock_irq(&vhca_buf->migf->list_lock);
 		list_del_init(&vhca_buf->buf_elm);
+		list_add_tail(&vhca_buf->buf_elm, &vhca_buf->migf->avail_list);
 		spin_unlock_irq(&vhca_buf->migf->list_lock);
 	}
 
@@ -283,6 +284,16 @@ static __poll_t mlx5vf_save_poll(struct file *filp,
 	return pollflags;
 }
 
+/*
+ * FD is exposed and user can use it after receiving an error.
+ * Mark migf in error, and wake the user.
+ */
+static void mlx5vf_mark_err(struct mlx5_vf_migration_file *migf)
+{
+	migf->state = MLX5_MIGF_STATE_ERROR;
+	wake_up_interruptible(&migf->poll_wait);
+}
+
 static const struct file_operations mlx5vf_save_fops = {
 	.owner = THIS_MODULE,
 	.read = mlx5vf_save_read,
@@ -291,8 +302,42 @@ static const struct file_operations mlx5vf_save_fops = {
 	.llseek = no_llseek,
 };
 
+static int mlx5vf_pci_save_device_inc_data(struct mlx5vf_pci_core_device *mvdev)
+{
+	struct mlx5_vf_migration_file *migf = mvdev->saving_migf;
+	struct mlx5_vhca_data_buffer *buf;
+	size_t length;
+	int ret;
+
+	if (migf->state == MLX5_MIGF_STATE_ERROR)
+		return -ENODEV;
+
+	ret = mlx5vf_cmd_query_vhca_migration_state(mvdev, &length,
+						    MLX5VF_QUERY_INC);
+	if (ret)
+		goto err;
+
+	buf = mlx5vf_get_data_buffer(migf, length, DMA_FROM_DEVICE);
+	if (IS_ERR(buf)) {
+		ret = PTR_ERR(buf);
+		goto err;
+	}
+
+	ret = mlx5vf_cmd_save_vhca_state(mvdev, migf, buf, true, false);
+	if (ret)
+		goto err_save;
+
+	return 0;
+
+err_save:
+	mlx5vf_put_data_buffer(buf);
+err:
+	mlx5vf_mark_err(migf);
+	return ret;
+}
+
 static struct mlx5_vf_migration_file *
-mlx5vf_pci_save_device_data(struct mlx5vf_pci_core_device *mvdev)
+mlx5vf_pci_save_device_data(struct mlx5vf_pci_core_device *mvdev, bool track)
 {
 	struct mlx5_vf_migration_file *migf;
 	struct mlx5_vhca_data_buffer *buf;
@@ -328,8 +373,9 @@ mlx5vf_pci_save_device_data(struct mlx5vf_pci_core_device *mvdev)
 	mlx5_cmd_init_async_ctx(mvdev->mdev, &migf->async_ctx);
 	INIT_WORK(&migf->async_data.work, mlx5vf_mig_file_cleanup_cb);
 	INIT_LIST_HEAD(&migf->buf_list);
+	INIT_LIST_HEAD(&migf->avail_list);
 	spin_lock_init(&migf->list_lock);
-	ret = mlx5vf_cmd_query_vhca_migration_state(mvdev, &length);
+	ret = mlx5vf_cmd_query_vhca_migration_state(mvdev, &length, 0);
 	if (ret)
 		goto out_pd;
 
@@ -339,7 +385,7 @@ mlx5vf_pci_save_device_data(struct mlx5vf_pci_core_device *mvdev)
 		goto out_pd;
 	}
 
-	ret = mlx5vf_cmd_save_vhca_state(mvdev, migf, buf);
+	ret = mlx5vf_cmd_save_vhca_state(mvdev, migf, buf, false, track);
 	if (ret)
 		goto out_save;
 	return migf;
@@ -462,6 +508,7 @@ mlx5vf_pci_resume_device_data(struct mlx5vf_pci_core_device *mvdev)
 	stream_open(migf->filp->f_inode, migf->filp);
 	mutex_init(&migf->lock);
 	INIT_LIST_HEAD(&migf->buf_list);
+	INIT_LIST_HEAD(&migf->avail_list);
 	spin_lock_init(&migf->list_lock);
 	return migf;
 out_pd:
@@ -514,7 +561,8 @@ mlx5vf_pci_step_device_state_locked(struct mlx5vf_pci_core_device *mvdev,
 		return NULL;
 	}
 
-	if (cur == VFIO_DEVICE_STATE_RUNNING && new == VFIO_DEVICE_STATE_RUNNING_P2P) {
+	if ((cur == VFIO_DEVICE_STATE_RUNNING && new == VFIO_DEVICE_STATE_RUNNING_P2P) ||
+	    (cur == VFIO_DEVICE_STATE_PRE_COPY && new == VFIO_DEVICE_STATE_PRE_COPY_P2P)) {
 		ret = mlx5vf_cmd_suspend_vhca(mvdev,
 			MLX5_SUSPEND_VHCA_IN_OP_MOD_SUSPEND_INITIATOR);
 		if (ret)
@@ -522,7 +570,8 @@ mlx5vf_pci_step_device_state_locked(struct mlx5vf_pci_core_device *mvdev,
 		return NULL;
 	}
 
-	if (cur == VFIO_DEVICE_STATE_RUNNING_P2P && new == VFIO_DEVICE_STATE_RUNNING) {
+	if ((cur == VFIO_DEVICE_STATE_RUNNING_P2P && new == VFIO_DEVICE_STATE_RUNNING) ||
+	    (cur == VFIO_DEVICE_STATE_PRE_COPY_P2P && new == VFIO_DEVICE_STATE_PRE_COPY)) {
 		ret = mlx5vf_cmd_resume_vhca(mvdev,
 			MLX5_RESUME_VHCA_IN_OP_MOD_RESUME_INITIATOR);
 		if (ret)
@@ -533,7 +582,7 @@ mlx5vf_pci_step_device_state_locked(struct mlx5vf_pci_core_device *mvdev,
 	if (cur == VFIO_DEVICE_STATE_STOP && new == VFIO_DEVICE_STATE_STOP_COPY) {
 		struct mlx5_vf_migration_file *migf;
 
-		migf = mlx5vf_pci_save_device_data(mvdev);
+		migf = mlx5vf_pci_save_device_data(mvdev, false);
 		if (IS_ERR(migf))
 			return ERR_CAST(migf);
 		get_file(migf->filp);
@@ -541,7 +590,10 @@ mlx5vf_pci_step_device_state_locked(struct mlx5vf_pci_core_device *mvdev,
 		return migf->filp;
 	}
 
-	if ((cur == VFIO_DEVICE_STATE_STOP_COPY && new == VFIO_DEVICE_STATE_STOP)) {
+	if ((cur == VFIO_DEVICE_STATE_STOP_COPY && new == VFIO_DEVICE_STATE_STOP) ||
+	    (cur == VFIO_DEVICE_STATE_PRE_COPY && new == VFIO_DEVICE_STATE_RUNNING) ||
+	    (cur == VFIO_DEVICE_STATE_PRE_COPY_P2P &&
+	     new == VFIO_DEVICE_STATE_RUNNING_P2P)) {
 		mlx5vf_disable_fds(mvdev);
 		return NULL;
 	}
@@ -567,6 +619,28 @@ mlx5vf_pci_step_device_state_locked(struct mlx5vf_pci_core_device *mvdev,
 		return NULL;
 	}
 
+	if ((cur == VFIO_DEVICE_STATE_RUNNING && new == VFIO_DEVICE_STATE_PRE_COPY) ||
+	    (cur == VFIO_DEVICE_STATE_RUNNING_P2P &&
+	     new == VFIO_DEVICE_STATE_PRE_COPY_P2P)) {
+		struct mlx5_vf_migration_file *migf;
+
+		migf = mlx5vf_pci_save_device_data(mvdev, true);
+		if (IS_ERR(migf))
+			return ERR_CAST(migf);
+		get_file(migf->filp);
+		mvdev->saving_migf = migf;
+		return migf->filp;
+	}
+
+	if (cur == VFIO_DEVICE_STATE_PRE_COPY_P2P && new == VFIO_DEVICE_STATE_STOP_COPY) {
+		ret = mlx5vf_cmd_suspend_vhca(mvdev,
+			MLX5_SUSPEND_VHCA_IN_OP_MOD_SUSPEND_RESPONDER);
+		if (ret)
+			return ERR_PTR(ret);
+		ret = mlx5vf_pci_save_device_inc_data(mvdev);
+		return ret ? ERR_PTR(ret) : NULL;
+	}
+
 	/*
 	 * vfio_mig_get_next_state() does not use arcs other than the above
 	 */
@@ -635,7 +709,7 @@ static int mlx5vf_pci_get_data_size(struct vfio_device *vdev,
 
 	mutex_lock(&mvdev->state_mutex);
 	ret = mlx5vf_cmd_query_vhca_migration_state(mvdev,
-						    &state_size);
+						    &state_size, 0);
 	if (!ret)
 		*stop_copy_length = state_size;
 	mlx5vf_state_mutex_unlock(mvdev);