[dpdk-stable] [PATCH v1] net/iavf: fix wrong bandwidth unit in TM capability querying
    Ting Xu 
    ting.xu at intel.com
       
    Thu Jul 15 04:08:11 CEST 2021
    
    
  
In IAVF node TM capability querying, the unit of bandwidth is Kbps,
which is not correct according to TM specification. Change the unit to
Byte per second. Refine some unclear comments as well.
Fixes: 44d0a720a538 ("net/iavf: query QoS capabilities and set queue TC mapping")
Cc: stable at dpdk.org
Signed-off-by: Ting Xu <ting.xu at intel.com>
---
 drivers/net/iavf/iavf_tm.c | 11 +++++++----
 1 file changed, 7 insertions(+), 4 deletions(-)
diff --git a/drivers/net/iavf/iavf_tm.c b/drivers/net/iavf/iavf_tm.c
index 558021014e..c2908986b9 100644
--- a/drivers/net/iavf/iavf_tm.c
+++ b/drivers/net/iavf/iavf_tm.c
@@ -469,7 +469,7 @@ iavf_tm_capabilities_get(struct rte_eth_dev *dev,
 	cap->shaper_private_n_max = cap->n_nodes_max;
 	cap->shaper_private_dual_rate_n_max = 0;
 	cap->shaper_private_rate_min = 0;
-	/* GBps */
+	/* Bytes per second */
 	cap->shaper_private_rate_max =
 		vf->link_speed * 1000 / IAVF_BITS_PER_BYTE;
 	cap->shaper_private_packet_mode_supported = 0;
@@ -544,7 +544,7 @@ iavf_level_capabilities_get(struct rte_eth_dev *dev,
 		cap->nonleaf.shaper_private_supported = true;
 		cap->nonleaf.shaper_private_dual_rate_supported = false;
 		cap->nonleaf.shaper_private_rate_min = 0;
-		/* GBps */
+		/* Bytes per second */
 		cap->nonleaf.shaper_private_rate_max =
 			vf->link_speed * 1000 / IAVF_BITS_PER_BYTE;
 		cap->nonleaf.shaper_private_packet_mode_supported = 0;
@@ -632,8 +632,11 @@ iavf_node_capabilities_get(struct rte_eth_dev *dev,
 
 	cap->shaper_private_supported = true;
 	cap->shaper_private_dual_rate_supported = false;
-	cap->shaper_private_rate_min = tc_cap.shaper.committed;
-	cap->shaper_private_rate_max = tc_cap.shaper.peak;
+	/* Bytes per second */
+	cap->shaper_private_rate_min =
+		tc_cap.shaper.committed * 1000 / IAVF_BITS_PER_BYTE;
+	cap->shaper_private_rate_max =
+		tc_cap.shaper.peak * 1000 / IAVF_BITS_PER_BYTE;
 	cap->shaper_shared_n_max = 0;
 	cap->nonleaf.sched_n_children_max = vf->num_queue_pairs;
 	cap->nonleaf.sched_sp_n_priorities_max = 1;
-- 
2.17.1
    
    
More information about the stable
mailing list