[2/2] net/mlx5: fix RxQ reference counting for indirect RSS

Message ID 20211123223159.3324247-3-dkozlyuk@nvidia.com (mailing list archive)
State Superseded, archived
Delegated to: Raslan Darawsheh
Headers
Series next/mlx5: fix indirect RSS reference counting |

Checks

Context Check Description
ci/checkpatch success coding style OK
ci/Intel-compilation success Compilation OK
ci/github-robot: build success github build: passed
ci/intel-Testing success Testing PASS
ci/iol-broadcom-Performance success Performance Testing PASS
ci/iol-mellanox-Performance success Performance Testing PASS
ci/iol-broadcom-Functional success Functional Testing PASS
ci/iol-intel-Functional success Functional Testing PASS
ci/iol-intel-Performance success Performance Testing PASS
ci/iol-x86_64-unit-testing fail Testing issues
ci/iol-aarch64-unit-testing success Testing PASS
ci/iol-x86_64-compile-testing fail Testing issues
ci/iol-aarch64-compile-testing success Testing PASS

Commit Message

Dmitry Kozlyuk Nov. 23, 2021, 10:31 p.m. UTC
  mlx5_ind_table_obj_modify() was not changing the reference counters
of neither the new set of RxQs, nor the old set of RxQs.
On the other hand, creation of the RSS incremented the RxQ refcnt.
If an RxQ was present in both the initial and the modified set,
its reference counter was incremented one extra time
compared to the queues that were only present in the new set.
This prevented releasing said RxQ resources on port stop:

    flow indirect_action 0 create action_id 1 \
        action rss queues 0 1 end / end
    flow indirect_action 0 update 1 \
        action rss queues 2 3 end / end
    quit
    ...
    mlx5_net: mlx5.c:1622: mlx5_dev_close():
        port 0 some Rx queue objects still remain
    mlx5_net: mlx5.c:1626: mlx5_dev_close():
        port 0 some Rx queues still remain

Increment reference counters for the new set of RxQs
and decrement them for the old set of RxQs.
Only do this when the port is started when the port is started.
Remove explicit referencing of RxQ from mlx5_ind_table_obj_attach()
because it reuses mlx5_ind_table_obj_modify() code doing this.

Fixes: ec4e11d41d12 ("net/mlx5: preserve indirect actions on restart")

Signed-off-by: Dmitry Kozlyuk <dkozlyuk@nvidia.com>
---
 drivers/net/mlx5/mlx5_rxq.c | 34 +++++++++++++++++++---------------
 1 file changed, 19 insertions(+), 15 deletions(-)
  

Patch

diff --git a/drivers/net/mlx5/mlx5_rxq.c b/drivers/net/mlx5/mlx5_rxq.c
index 8f9a94572f..46d6536be5 100644
--- a/drivers/net/mlx5/mlx5_rxq.c
+++ b/drivers/net/mlx5/mlx5_rxq.c
@@ -2392,7 +2392,8 @@  mlx5_ind_table_obj_modify(struct rte_eth_dev *dev,
 			  bool standalone)
 {
 	struct mlx5_priv *priv = dev->data->dev_private;
-	unsigned int i;
+	bool dev_started = priv->dev_data->dev_started;
+	unsigned int i, j;
 	int ret = 0, err;
 	const unsigned int n = rte_is_power_of_2(queues_n) ?
 			       log2above(queues_n) :
@@ -2402,22 +2403,30 @@  mlx5_ind_table_obj_modify(struct rte_eth_dev *dev,
 	RTE_SET_USED(standalone);
 	if (mlx5_ind_table_obj_check_standalone(dev, ind_tbl) < 0)
 		return -rte_errno;
-	for (i = 0; i != queues_n; ++i) {
-		if (!mlx5_rxq_get(dev, queues[i])) {
-			ret = -rte_errno;
-			goto error;
+	if (dev_started)
+		for (i = 0; i != queues_n; ++i) {
+			if (!mlx5_rxq_ref(dev, queues[i])) {
+				ret = -rte_errno;
+				goto error;
+			}
 		}
-	}
 	MLX5_ASSERT(priv->obj_ops.ind_table_modify);
 	ret = priv->obj_ops.ind_table_modify(dev, n, queues, queues_n, ind_tbl);
 	if (ret)
 		goto error;
+	if (dev_started)
+		for (i = 0; i < ind_tbl->queues_n; i++)
+			claim_nonzero(mlx5_rxq_deref(dev, ind_tbl->queues[i]));
 	ind_tbl->queues_n = queues_n;
 	ind_tbl->queues = queues;
 	return 0;
 error:
-	err = rte_errno;
-	rte_errno = err;
+	if (dev_started) {
+		err = rte_errno;
+		for (j = 0; j < i; j++)
+			mlx5_rxq_deref(dev, queues[j]);
+		rte_errno = err;
+	}
 	DRV_LOG(DEBUG, "Port %u cannot setup indirection table.",
 		dev->data->port_id);
 	return ret;
@@ -2438,19 +2447,14 @@  int
 mlx5_ind_table_obj_attach(struct rte_eth_dev *dev,
 			  struct mlx5_ind_table_obj *ind_tbl)
 {
-	unsigned int i;
 	int ret;
 
 	ret = mlx5_ind_table_obj_modify(dev, ind_tbl, ind_tbl->queues,
 					ind_tbl->queues_n, true);
-	if (ret != 0) {
+	if (ret != 0)
 		DRV_LOG(ERR, "Port %u could not modify indirect table obj %p",
 			dev->data->port_id, (void *)ind_tbl);
-		return ret;
-	}
-	for (i = 0; i < ind_tbl->queues_n; i++)
-		mlx5_rxq_ref(dev, ind_tbl->queues[i]);
-	return 0;
+	return ret;
 }
 
 /**