From 7414539c5f2e43bad67ae88a3612455d01583429 Mon Sep 17 00:00:00 2001 From: "Michael S. Tsirkin" Date: Wed, 30 Mar 2022 02:19:16 -0400 Subject: [PATCH 1/4] Revert "virtio: use virtio_device_ready() in virtio_device_restore()" This reverts commit 8d65bc9a5be3f23c5e2ab36b6b8ef40095165b18. We reverted the problematic changes, no more need for work arounds on restore. Signed-off-by: Michael S. Tsirkin Acked-by: Jason Wang --- drivers/virtio/virtio.c | 5 ++--- 1 file changed, 2 insertions(+), 3 deletions(-) diff --git a/drivers/virtio/virtio.c b/drivers/virtio/virtio.c index 75c8d560bbd3..22f15f444f75 100644 --- a/drivers/virtio/virtio.c +++ b/drivers/virtio/virtio.c @@ -526,9 +526,8 @@ int virtio_device_restore(struct virtio_device *dev) goto err; } - /* If restore didn't do it, mark device DRIVER_OK ourselves. */ - if (!(dev->config->get_status(dev) & VIRTIO_CONFIG_S_DRIVER_OK)) - virtio_device_ready(dev); + /* Finally, tell the device we're all set */ + virtio_add_status(dev, VIRTIO_CONFIG_S_DRIVER_OK); virtio_config_enable(dev); From c18c86808b78c4c2dc69f27f37c57abab14ee387 Mon Sep 17 00:00:00 2001 From: "Michael S. Tsirkin" Date: Wed, 30 Mar 2022 02:22:17 -0400 Subject: [PATCH 2/4] Revert "virtio_config: introduce a new .enable_cbs method" This reverts commit d50497eb4e554e1f0351e1836ee7241c059592e6. The new callback ended up not being used, and it's asymmetrical: just enable, no disable. Signed-off-by: Michael S. Tsirkin Acked-by: Jason Wang --- include/linux/virtio_config.h | 6 ------ 1 file changed, 6 deletions(-) diff --git a/include/linux/virtio_config.h b/include/linux/virtio_config.h index dafdc7f48c01..b341dd62aa4d 100644 --- a/include/linux/virtio_config.h +++ b/include/linux/virtio_config.h @@ -23,8 +23,6 @@ struct virtio_shm_region { * any of @get/@set, @get_status/@set_status, or @get_features/ * @finalize_features are NOT safe to be called from an atomic * context. - * @enable_cbs: enable the callbacks - * vdev: the virtio_device * @get: read the value of a configuration field * vdev: the virtio_device * offset: the offset of the configuration field @@ -78,7 +76,6 @@ struct virtio_shm_region { */ typedef void vq_callback_t(struct virtqueue *); struct virtio_config_ops { - void (*enable_cbs)(struct virtio_device *vdev); void (*get)(struct virtio_device *vdev, unsigned offset, void *buf, unsigned len); void (*set)(struct virtio_device *vdev, unsigned offset, @@ -233,9 +230,6 @@ void virtio_device_ready(struct virtio_device *dev) { unsigned status = dev->config->get_status(dev); - if (dev->config->enable_cbs) - dev->config->enable_cbs(dev); - BUG_ON(status & VIRTIO_CONFIG_S_DRIVER_OK); dev->config->set_status(dev, status | VIRTIO_CONFIG_S_DRIVER_OK); } From 55ebf0d60e3cc6c9e8593399e185842c00e12f36 Mon Sep 17 00:00:00 2001 From: Jason Wang Date: Tue, 29 Mar 2022 12:21:07 +0800 Subject: [PATCH 3/4] vdpa: mlx5: prevent cvq work from hogging CPU A userspace triggerable infinite loop could happen in mlx5_cvq_kick_handler() if userspace keeps sending a huge amount of cvq requests. Fixing this by introducing a quota and re-queue the work if we're out of the budget (currently the implicit budget is one) . While at it, using a per device work struct to avoid on demand memory allocation for cvq. Fixes: 5262912ef3cfc ("vdpa/mlx5: Add support for control VQ and MAC setting") Signed-off-by: Jason Wang Link: https://lore.kernel.org/r/20220329042109.4029-1-jasowang@redhat.com Signed-off-by: Michael S. Tsirkin Acked-by: Eli Cohen --- drivers/vdpa/mlx5/net/mlx5_vnet.c | 21 +++++++++------------ 1 file changed, 9 insertions(+), 12 deletions(-) diff --git a/drivers/vdpa/mlx5/net/mlx5_vnet.c b/drivers/vdpa/mlx5/net/mlx5_vnet.c index 2f4fb09f1e89..da21e5430434 100644 --- a/drivers/vdpa/mlx5/net/mlx5_vnet.c +++ b/drivers/vdpa/mlx5/net/mlx5_vnet.c @@ -163,6 +163,7 @@ struct mlx5_vdpa_net { u32 cur_num_vqs; struct notifier_block nb; struct vdpa_callback config_cb; + struct mlx5_vdpa_wq_ent cvq_ent; }; static void free_resources(struct mlx5_vdpa_net *ndev); @@ -1659,10 +1660,10 @@ static void mlx5_cvq_kick_handler(struct work_struct *work) ndev = to_mlx5_vdpa_ndev(mvdev); cvq = &mvdev->cvq; if (!(ndev->mvdev.actual_features & BIT_ULL(VIRTIO_NET_F_CTRL_VQ))) - goto out; + return; if (!cvq->ready) - goto out; + return; while (true) { err = vringh_getdesc_iotlb(&cvq->vring, &cvq->riov, &cvq->wiov, &cvq->head, @@ -1696,9 +1697,10 @@ static void mlx5_cvq_kick_handler(struct work_struct *work) if (vringh_need_notify_iotlb(&cvq->vring)) vringh_notify(&cvq->vring); + + queue_work(mvdev->wq, &wqent->work); + break; } -out: - kfree(wqent); } static void mlx5_vdpa_kick_vq(struct vdpa_device *vdev, u16 idx) @@ -1706,7 +1708,6 @@ static void mlx5_vdpa_kick_vq(struct vdpa_device *vdev, u16 idx) struct mlx5_vdpa_dev *mvdev = to_mvdev(vdev); struct mlx5_vdpa_net *ndev = to_mlx5_vdpa_ndev(mvdev); struct mlx5_vdpa_virtqueue *mvq; - struct mlx5_vdpa_wq_ent *wqent; if (!is_index_valid(mvdev, idx)) return; @@ -1715,13 +1716,7 @@ static void mlx5_vdpa_kick_vq(struct vdpa_device *vdev, u16 idx) if (!mvdev->wq || !mvdev->cvq.ready) return; - wqent = kzalloc(sizeof(*wqent), GFP_ATOMIC); - if (!wqent) - return; - - wqent->mvdev = mvdev; - INIT_WORK(&wqent->work, mlx5_cvq_kick_handler); - queue_work(mvdev->wq, &wqent->work); + queue_work(mvdev->wq, &ndev->cvq_ent.work); return; } @@ -2740,6 +2735,8 @@ static int mlx5_vdpa_dev_add(struct vdpa_mgmt_dev *v_mdev, const char *name, if (err) goto err_mr; + ndev->cvq_ent.mvdev = mvdev; + INIT_WORK(&ndev->cvq_ent.work, mlx5_cvq_kick_handler); mvdev->wq = create_singlethread_workqueue("mlx5_vdpa_wq"); if (!mvdev->wq) { err = -ENOMEM; From 1c80cf031e0204fde471558ee40183695773ce13 Mon Sep 17 00:00:00 2001 From: Jason Wang Date: Tue, 29 Mar 2022 12:21:08 +0800 Subject: [PATCH 4/4] vdpa: mlx5: synchronize driver status with CVQ Currently, CVQ doesn't have any synchronization with the driver status. Then CVQ emulation code run in the middle of: 1) device reset 2) device status changed 3) map updating The will lead several unexpected issue like trying to execute CVQ command after the driver has been teared down. Fixing this by using reslock to synchronize CVQ emulation code with the driver status changing: - protect the whole device reset, status changing and set_map() updating with reslock - protect the CVQ handler with the reslock and check VIRTIO_CONFIG_S_DRIVER_OK in the CVQ handler This will guarantee that: 1) CVQ handler won't work if VIRTIO_CONFIG_S_DRIVER_OK is not set 2) CVQ handler will see a consistent state of the driver instead of the partial one when it is running in the middle of the teardown_driver() or setup_driver(). Cc: 5262912ef3cfc ("vdpa/mlx5: Add support for control VQ and MAC setting") Signed-off-by: Jason Wang Link: https://lore.kernel.org/r/20220329042109.4029-2-jasowang@redhat.com Signed-off-by: Michael S. Tsirkin Acked-by: Eli Cohen --- drivers/vdpa/mlx5/net/mlx5_vnet.c | 51 ++++++++++++++++++++++--------- 1 file changed, 37 insertions(+), 14 deletions(-) diff --git a/drivers/vdpa/mlx5/net/mlx5_vnet.c b/drivers/vdpa/mlx5/net/mlx5_vnet.c index da21e5430434..79001301b383 100644 --- a/drivers/vdpa/mlx5/net/mlx5_vnet.c +++ b/drivers/vdpa/mlx5/net/mlx5_vnet.c @@ -1659,11 +1659,17 @@ static void mlx5_cvq_kick_handler(struct work_struct *work) mvdev = wqent->mvdev; ndev = to_mlx5_vdpa_ndev(mvdev); cvq = &mvdev->cvq; + + mutex_lock(&ndev->reslock); + + if (!(mvdev->status & VIRTIO_CONFIG_S_DRIVER_OK)) + goto out; + if (!(ndev->mvdev.actual_features & BIT_ULL(VIRTIO_NET_F_CTRL_VQ))) - return; + goto out; if (!cvq->ready) - return; + goto out; while (true) { err = vringh_getdesc_iotlb(&cvq->vring, &cvq->riov, &cvq->wiov, &cvq->head, @@ -1701,6 +1707,9 @@ static void mlx5_cvq_kick_handler(struct work_struct *work) queue_work(mvdev->wq, &wqent->work); break; } + +out: + mutex_unlock(&ndev->reslock); } static void mlx5_vdpa_kick_vq(struct vdpa_device *vdev, u16 idx) @@ -2175,7 +2184,7 @@ static int mlx5_vdpa_change_map(struct mlx5_vdpa_dev *mvdev, struct vhost_iotlb goto err_mr; if (!(mvdev->status & VIRTIO_CONFIG_S_DRIVER_OK)) - return 0; + goto err_mr; restore_channels_info(ndev); err = setup_driver(mvdev); @@ -2190,12 +2199,14 @@ err_mr: return err; } +/* reslock must be held for this function */ static int setup_driver(struct mlx5_vdpa_dev *mvdev) { struct mlx5_vdpa_net *ndev = to_mlx5_vdpa_ndev(mvdev); int err; - mutex_lock(&ndev->reslock); + WARN_ON(!mutex_is_locked(&ndev->reslock)); + if (ndev->setup) { mlx5_vdpa_warn(mvdev, "setup driver called for already setup driver\n"); err = 0; @@ -2225,7 +2236,6 @@ static int setup_driver(struct mlx5_vdpa_dev *mvdev) goto err_fwd; } ndev->setup = true; - mutex_unlock(&ndev->reslock); return 0; @@ -2236,23 +2246,23 @@ err_tir: err_rqt: teardown_virtqueues(ndev); out: - mutex_unlock(&ndev->reslock); return err; } +/* reslock must be held for this function */ static void teardown_driver(struct mlx5_vdpa_net *ndev) { - mutex_lock(&ndev->reslock); + + WARN_ON(!mutex_is_locked(&ndev->reslock)); + if (!ndev->setup) - goto out; + return; remove_fwd_to_tir(ndev); destroy_tir(ndev); destroy_rqt(ndev); teardown_virtqueues(ndev); ndev->setup = false; -out: - mutex_unlock(&ndev->reslock); } static void clear_vqs_ready(struct mlx5_vdpa_net *ndev) @@ -2273,6 +2283,8 @@ static void mlx5_vdpa_set_status(struct vdpa_device *vdev, u8 status) print_status(mvdev, status, true); + mutex_lock(&ndev->reslock); + if ((status ^ ndev->mvdev.status) & VIRTIO_CONFIG_S_DRIVER_OK) { if (status & VIRTIO_CONFIG_S_DRIVER_OK) { err = setup_driver(mvdev); @@ -2282,16 +2294,19 @@ static void mlx5_vdpa_set_status(struct vdpa_device *vdev, u8 status) } } else { mlx5_vdpa_warn(mvdev, "did not expect DRIVER_OK to be cleared\n"); - return; + goto err_clear; } } ndev->mvdev.status = status; + mutex_unlock(&ndev->reslock); return; err_setup: mlx5_vdpa_destroy_mr(&ndev->mvdev); ndev->mvdev.status |= VIRTIO_CONFIG_S_FAILED; +err_clear: + mutex_unlock(&ndev->reslock); } static int mlx5_vdpa_reset(struct vdpa_device *vdev) @@ -2301,6 +2316,8 @@ static int mlx5_vdpa_reset(struct vdpa_device *vdev) print_status(mvdev, 0, true); mlx5_vdpa_info(mvdev, "performing device reset\n"); + + mutex_lock(&ndev->reslock); teardown_driver(ndev); clear_vqs_ready(ndev); mlx5_vdpa_destroy_mr(&ndev->mvdev); @@ -2313,6 +2330,7 @@ static int mlx5_vdpa_reset(struct vdpa_device *vdev) if (mlx5_vdpa_create_mr(mvdev, NULL)) mlx5_vdpa_warn(mvdev, "create MR failed\n"); } + mutex_unlock(&ndev->reslock); return 0; } @@ -2348,19 +2366,24 @@ static u32 mlx5_vdpa_get_generation(struct vdpa_device *vdev) static int mlx5_vdpa_set_map(struct vdpa_device *vdev, struct vhost_iotlb *iotlb) { struct mlx5_vdpa_dev *mvdev = to_mvdev(vdev); + struct mlx5_vdpa_net *ndev = to_mlx5_vdpa_ndev(mvdev); bool change_map; int err; + mutex_lock(&ndev->reslock); + err = mlx5_vdpa_handle_set_map(mvdev, iotlb, &change_map); if (err) { mlx5_vdpa_warn(mvdev, "set map failed(%d)\n", err); - return err; + goto err; } if (change_map) - return mlx5_vdpa_change_map(mvdev, iotlb); + err = mlx5_vdpa_change_map(mvdev, iotlb); - return 0; +err: + mutex_unlock(&ndev->reslock); + return err; } static void mlx5_vdpa_free(struct vdpa_device *vdev)