vdpa/mlx5: add device configuration lock
Checks
Commit Message
Under active traffic, qemu switch vq enable state, there will be segment
fault.
The reason is that timer polling thread or channel event thread still
referencing resources that being destroyed or partially initialized.
This patch add per device vq configuration lock and test device ready
flag in event thread to make sure vq being accessed fully initialized.
Redmine: 2239647
Signed-off-by: Xueming Li <xuemingl@mellanox.com>
---
drivers/vdpa/mlx5/mlx5_vdpa.c | 8 +++++++-
drivers/vdpa/mlx5/mlx5_vdpa.h | 1 +
drivers/vdpa/mlx5/mlx5_vdpa_event.c | 8 ++++++++
3 files changed, 16 insertions(+), 1 deletion(-)
@@ -133,6 +133,7 @@ mlx5_vdpa_set_vring_state(int vid, int vring, int state)
struct rte_vdpa_device *vdev = rte_vhost_get_vdpa_device(vid);
struct mlx5_vdpa_priv *priv =
mlx5_vdpa_find_priv_resource_by_vdev(vdev);
+ int ret;
if (priv == NULL) {
DRV_LOG(ERR, "Invalid vDPA device: %s.", vdev->device->name);
@@ -142,7 +143,10 @@ mlx5_vdpa_set_vring_state(int vid, int vring, int state)
DRV_LOG(ERR, "Too big vring id: %d.", vring);
return -E2BIG;
}
- return mlx5_vdpa_virtq_enable(priv, vring, state);
+ pthread_mutex_lock(&priv->vq_config_lock);
+ ret = mlx5_vdpa_virtq_enable(priv, vring, state);
+ pthread_mutex_unlock(&priv->vq_config_lock);
+ return ret;
}
static int
@@ -742,6 +746,7 @@ mlx5_vdpa_pci_probe(struct rte_pci_driver *pci_drv __rte_unused,
}
mlx5_vdpa_config_get(pci_dev->device.devargs, priv);
SLIST_INIT(&priv->mr_list);
+ pthread_mutex_init(&priv->vq_config_lock, NULL);
pthread_mutex_lock(&priv_list_lock);
TAILQ_INSERT_TAIL(&priv_list, priv, next);
pthread_mutex_unlock(&priv_list_lock);
@@ -793,6 +798,7 @@ mlx5_vdpa_pci_remove(struct rte_pci_device *pci_dev)
priv->var = NULL;
}
mlx5_glue->close_device(priv->ctx);
+ pthread_mutex_destroy(&priv->vq_config_lock);
rte_free(priv);
}
return 0;
@@ -120,6 +120,7 @@ enum {
struct mlx5_vdpa_priv {
TAILQ_ENTRY(mlx5_vdpa_priv) next;
uint8_t configured;
+ pthread_mutex_t vq_config_lock;
uint64_t last_traffic_tic;
pthread_t timer_tid;
pthread_mutex_t timer_lock;
@@ -274,6 +274,7 @@ mlx5_vdpa_poll_handle(void *arg)
priv->event_us;
while (1) {
max = 0;
+ pthread_mutex_lock(&priv->vq_config_lock);
for (i = 0; i < priv->nr_virtqs; i++) {
cq = &priv->virtqs[i].eqp.cq;
if (cq->cq && !cq->armed) {
@@ -298,6 +299,7 @@ mlx5_vdpa_poll_handle(void *arg)
priv->vdev->device->name);
mlx5_vdpa_arm_all_cqs(priv);
pthread_mutex_lock(&priv->timer_lock);
+ pthread_mutex_unlock(&priv->vq_config_lock);
priv->timer_on = 0;
while (!priv->timer_on)
pthread_cond_wait(&priv->timer_cond,
@@ -312,6 +314,7 @@ mlx5_vdpa_poll_handle(void *arg)
} else {
priv->last_traffic_tic = current_tic;
}
+ pthread_mutex_unlock(&priv->vq_config_lock);
mlx5_vdpa_timer_sleep(priv, max);
}
return NULL;
@@ -327,6 +330,7 @@ mlx5_vdpa_interrupt_handler(void *cb_arg)
uint8_t buf[sizeof(struct mlx5dv_devx_async_event_hdr) + 128];
} out;
+ pthread_mutex_lock(&priv->vq_config_lock);
while (mlx5_glue->devx_get_event(priv->eventc, &out.event_resp,
sizeof(out.buf)) >=
(ssize_t)sizeof(out.event_resp.cookie)) {
@@ -337,12 +341,15 @@ mlx5_vdpa_interrupt_handler(void *cb_arg)
struct mlx5_vdpa_virtq *virtq = container_of(eqp,
struct mlx5_vdpa_virtq, eqp);
+ if (!virtq->enable)
+ continue;
mlx5_vdpa_cq_poll(cq);
/* Notify guest for descs consuming. */
if (cq->callfd != -1)
eventfd_write(cq->callfd, (eventfd_t)1);
if (priv->event_mode == MLX5_VDPA_EVENT_MODE_ONLY_INTERRUPT) {
mlx5_vdpa_cq_arm(priv, cq);
+ pthread_mutex_unlock(&priv->vq_config_lock);
return;
}
/* Don't arm again - timer will take control. */
@@ -363,6 +370,7 @@ mlx5_vdpa_interrupt_handler(void *cb_arg)
pthread_cond_signal(&priv->timer_cond);
}
pthread_mutex_unlock(&priv->timer_lock);
+ pthread_mutex_unlock(&priv->vq_config_lock);
}
int