Hi,
> -----Original Message-----
> From: dev <dev-boun...@dpdk.org> On Behalf Of Xueming Li
> Sent: Wednesday, October 21, 2020 2:15 PM
> To: Matan Azrad <ma...@nvidia.com>; Slava Ovsiienko
> <viachesl...@nvidia.com>
> Cc: dev@dpdk.org; Xueming(Steven) Li <xuemi...@nvidia.com>; Asaf Penso
> <as...@nvidia.com>; Jack Min <jack...@nvidia.com>; sta...@dpdk.org
> Subject: [dpdk-dev] [PATCH v1] net/mlx5: fix port shared data reference
> count
>
> When probe a representor, tag cache hash table and modification cache
> hash table allocated memory upon each port, overwrote previous existing
> cache in shared context data.
>
> This patch moves reference check of shared data prior to hash table
> allocation to avoid such issue.
>
> Fixes: 6801116688fe ("net/mlx5: fix multiple flow table hash list")
> Cc: jack...@nvidia.com
> Fixes: 1ef4cdef2682 ("net/mlx5: fix flow tag hash list conversion")
> Cc: ma...@nvidia.com
>
> Cc: sta...@dpdk.org
> Acked-by: Matan Azrad <ma...@nvidia.com>
> Signed-off-by: Xueming Li <xuemi...@nvidia.com>
> ---
> drivers/net/mlx5/linux/mlx5_os.c | 30 ++++++++----------------------
> drivers/net/mlx5/mlx5.h | 2 --
> drivers/net/mlx5/mlx5_flow_dv.c | 4 ++--
> 3 files changed, 10 insertions(+), 26 deletions(-)
>
> diff --git a/drivers/net/mlx5/linux/mlx5_os.c
> b/drivers/net/mlx5/linux/mlx5_os.c
> index 457008ef9e..40f9446d43 100644
> --- a/drivers/net/mlx5/linux/mlx5_os.c
> +++ b/drivers/net/mlx5/linux/mlx5_os.c
> @@ -226,13 +226,12 @@ mlx5_alloc_shared_dr(struct mlx5_priv *priv)
> {
> struct mlx5_dev_ctx_shared *sh = priv->sh;
> char s[MLX5_HLIST_NAMESIZE];
> - int err = 0;
> + int err;
>
> - if (!sh->flow_tbls)
> - err = mlx5_alloc_table_hash_list(priv);
> - else
> - DRV_LOG(DEBUG, "sh->flow_tbls[%p] already created,
> reuse\n",
> - (void *)sh->flow_tbls);
> + MLX5_ASSERT(sh && sh->refcnt);
> + if (sh->refcnt > 1)
> + return 0;
> + err = mlx5_alloc_table_hash_list(priv);
> if (err)
> return err;
> /* Create tags hash list table. */
> @@ -261,12 +260,6 @@ mlx5_alloc_shared_dr(struct mlx5_priv *priv)
> #ifdef HAVE_MLX5DV_DR
> void *domain;
>
> - if (sh->dv_refcnt) {
> - /* Shared DV/DR structures is already initialized. */
> - sh->dv_refcnt++;
> - priv->dr_shared = 1;
> - return 0;
> - }
> /* Reference counter is zero, we should initialize structures. */
> domain = mlx5_glue->dr_create_domain(sh->ctx,
>
> MLX5DV_DR_DOMAIN_TYPE_NIC_RX);
> @@ -306,8 +299,6 @@ mlx5_alloc_shared_dr(struct mlx5_priv *priv)
> }
> sh->pop_vlan_action = mlx5_glue-
> >dr_create_flow_action_pop_vlan();
> #endif /* HAVE_MLX5DV_DR */
> - sh->dv_refcnt++;
> - priv->dr_shared = 1;
> return 0;
> error:
> /* Rollback the created objects. */
> @@ -357,17 +348,12 @@ mlx5_alloc_shared_dr(struct mlx5_priv *priv)
> void
> mlx5_os_free_shared_dr(struct mlx5_priv *priv)
> {
> - struct mlx5_dev_ctx_shared *sh;
> + struct mlx5_dev_ctx_shared *sh = priv->sh;
>
> - if (!priv->dr_shared)
> + MLX5_ASSERT(sh && sh->refcnt);
> + if (sh->refcnt > 1)
> return;
> - priv->dr_shared = 0;
> - sh = priv->sh;
> - MLX5_ASSERT(sh);
> #ifdef HAVE_MLX5DV_DR
> - MLX5_ASSERT(sh->dv_refcnt);
> - if (sh->dv_refcnt && --sh->dv_refcnt)
> - return;
> if (sh->rx_domain) {
> mlx5_glue->dr_destroy_domain(sh->rx_domain);
> sh->rx_domain = NULL;
> diff --git a/drivers/net/mlx5/mlx5.h b/drivers/net/mlx5/mlx5.h
> index fa69c66137..c9d5d71630 100644
> --- a/drivers/net/mlx5/mlx5.h
> +++ b/drivers/net/mlx5/mlx5.h
> @@ -635,7 +635,6 @@ struct mlx5_dev_ctx_shared {
> uint32_t dv_meta_mask; /* flow META metadata supported mask.
> */
> uint32_t dv_mark_mask; /* flow MARK metadata supported mask.
> */
> uint32_t dv_regc0_mask; /* available bits of metatada reg_c[0]. */
> - uint32_t dv_refcnt; /* DV/DR data reference counter. */
> void *fdb_domain; /* FDB Direct Rules name space handle. */
> void *rx_domain; /* RX Direct Rules name space handle. */
> void *tx_domain; /* TX Direct Rules name space handle. */
> @@ -823,7 +822,6 @@ struct mlx5_priv {
> unsigned int isolated:1; /* Whether isolated mode is enabled. */
> unsigned int representor:1; /* Device is a port representor. */
> unsigned int master:1; /* Device is a E-Switch master. */
> - unsigned int dr_shared:1; /* DV/DR data is shared. */
> unsigned int txpp_en:1; /* Tx packet pacing enabled. */
> unsigned int mtr_en:1; /* Whether support meter. */
> unsigned int mtr_reg_share:1; /* Whether support meter REG_C
> share. */
> diff --git a/drivers/net/mlx5/mlx5_flow_dv.c
> b/drivers/net/mlx5/mlx5_flow_dv.c
> index 49d96362f8..15cd34e331 100644
> --- a/drivers/net/mlx5/mlx5_flow_dv.c
> +++ b/drivers/net/mlx5/mlx5_flow_dv.c
> @@ -293,7 +293,7 @@ flow_dv_shared_lock(struct rte_eth_dev *dev)
> struct mlx5_priv *priv = dev->data->dev_private;
> struct mlx5_dev_ctx_shared *sh = priv->sh;
>
> - if (sh->dv_refcnt > 1) {
> + if (sh->refcnt > 1) {
> int ret;
>
> ret = pthread_mutex_lock(&sh->dv_mutex);
> @@ -308,7 +308,7 @@ flow_dv_shared_unlock(struct rte_eth_dev *dev)
> struct mlx5_priv *priv = dev->data->dev_private;
> struct mlx5_dev_ctx_shared *sh = priv->sh;
>
> - if (sh->dv_refcnt > 1) {
> + if (sh->refcnt > 1) {
> int ret;
>
> ret = pthread_mutex_unlock(&sh->dv_mutex);
> --
> 2.25.1
Patch applied to next-net-mlx,
Kindest regards,
Raslan Darawsheh
