| Message ID | 20220224155101.1991626-3-xuemingl@nvidia.com (mailing list archive) |
|---|---|
| State | Superseded |
| Delegated to: | Maxime Coquelin |
| Headers | show |
| Series | vdpa/mlx5: improve device shutdown time | expand |
| Context | Check | Description |
|---|---|---|
| ci/checkpatch | success | coding style OK |
On 2/24/22 16:50, Xueming Li wrote: > In Ctrl+C handling, sometimes kick handling thread gets endless EGAIN > error and fall into dead lock. > > Kick happens frequently in real system due to busy traffic or retry > mechanism. This patch simplifies kick firmware anyway and skip setting > hardware notifier due to potential device error, notifier could be set > in next successful kick request. > > Fixes: 62c813706e41 ("vdpa/mlx5: map doorbell") > Cc: stable@dpdk.org > > Signed-off-by: Xueming Li <xuemingl@nvidia.com> > --- > drivers/vdpa/mlx5/mlx5_vdpa_virtq.c | 8 +++++--- > 1 file changed, 5 insertions(+), 3 deletions(-) > > diff --git a/drivers/vdpa/mlx5/mlx5_vdpa_virtq.c b/drivers/vdpa/mlx5/mlx5_vdpa_virtq.c > index de324506cb9..e1e05924a40 100644 > --- a/drivers/vdpa/mlx5/mlx5_vdpa_virtq.c > +++ b/drivers/vdpa/mlx5/mlx5_vdpa_virtq.c > @@ -23,11 +23,11 @@ mlx5_vdpa_virtq_kick_handler(void *cb_arg) > struct mlx5_vdpa_priv *priv = virtq->priv; > uint64_t buf; > int nbytes; > + int retry; > > if (rte_intr_fd_get(virtq->intr_handle) < 0) > return; > - > - do { > + for (retry = 0; retry < 3; ++retry) { > nbytes = read(rte_intr_fd_get(virtq->intr_handle), &buf, > 8); > if (nbytes < 0) { > @@ -39,7 +39,9 @@ mlx5_vdpa_virtq_kick_handler(void *cb_arg) > virtq->index, strerror(errno)); > } > break; > - } while (1); > + } > + if (nbytes < 0) > + return; > rte_write32(virtq->index, priv->virtq_db_addr); > if (virtq->notifier_state == MLX5_VDPA_NOTIFIER_STATE_DISABLED) { > if (rte_vhost_host_notifier_ctrl(priv->vid, virtq->index, true)) Reviewed-by: Maxime Coquelin <maxime.coquelin@redhat.com> Thanks, Maxime
diff --git a/drivers/vdpa/mlx5/mlx5_vdpa_virtq.c b/drivers/vdpa/mlx5/mlx5_vdpa_virtq.c index de324506cb9..e1e05924a40 100644 --- a/drivers/vdpa/mlx5/mlx5_vdpa_virtq.c +++ b/drivers/vdpa/mlx5/mlx5_vdpa_virtq.c @@ -23,11 +23,11 @@ mlx5_vdpa_virtq_kick_handler(void *cb_arg) struct mlx5_vdpa_priv *priv = virtq->priv; uint64_t buf; int nbytes; + int retry; if (rte_intr_fd_get(virtq->intr_handle) < 0) return; - - do { + for (retry = 0; retry < 3; ++retry) { nbytes = read(rte_intr_fd_get(virtq->intr_handle), &buf, 8); if (nbytes < 0) { @@ -39,7 +39,9 @@ mlx5_vdpa_virtq_kick_handler(void *cb_arg) virtq->index, strerror(errno)); } break; - } while (1); + } + if (nbytes < 0) + return; rte_write32(virtq->index, priv->virtq_db_addr); if (virtq->notifier_state == MLX5_VDPA_NOTIFIER_STATE_DISABLED) { if (rte_vhost_host_notifier_ctrl(priv->vid, virtq->index, true))
In Ctrl+C handling, sometimes kick handling thread gets endless EGAIN error and fall into dead lock. Kick happens frequently in real system due to busy traffic or retry mechanism. This patch simplifies kick firmware anyway and skip setting hardware notifier due to potential device error, notifier could be set in next successful kick request. Fixes: 62c813706e41 ("vdpa/mlx5: map doorbell") Cc: stable@dpdk.org Signed-off-by: Xueming Li <xuemingl@nvidia.com> --- drivers/vdpa/mlx5/mlx5_vdpa_virtq.c | 8 +++++--- 1 file changed, 5 insertions(+), 3 deletions(-)