Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]


Groups > linux.kernel > #1461103 > unrolled thread

[PATCH v2 18/21] net: thunderx: Use napi_consume_skb for bulk free

Started bysunil.kovvuri@gmail.com
First post2016-08-12 13:30 +0200
Last post2016-08-12 13:30 +0200
Articles 1 — 1 participant

Back to article view | Back to linux.kernel

This discussion starts older than the indexed window; earlier articles aren't shown. The article labeled Started by below is the oldest one visible, not the original post.


Contents

  [PATCH v2 18/21] net: thunderx: Use napi_consume_skb for bulk free sunil.kovvuri@gmail.com - 2016-08-12 13:30 +0200

#1461103 — [PATCH v2 18/21] net: thunderx: Use napi_consume_skb for bulk free

Fromsunil.kovvuri@gmail.com
Date2016-08-12 13:30 +0200
Subject[PATCH v2 18/21] net: thunderx: Use napi_consume_skb for bulk free
Message-ID<s5iP0-Yf-25@gated-at.bofh.it>
From: Sunil Goutham <sgoutham@cavium.com>

This patch enables bulk freeing on the Tx side.

Signed-off-by: Sunil Goutham <sgoutham@cavium.com>
---
 drivers/net/ethernet/cavium/thunder/nicvf_main.c | 8 +++++---
 1 file changed, 5 insertions(+), 3 deletions(-)

diff --git a/drivers/net/ethernet/cavium/thunder/nicvf_main.c b/drivers/net/ethernet/cavium/thunder/nicvf_main.c
index 0b68f2b..eb48d33 100644
--- a/drivers/net/ethernet/cavium/thunder/nicvf_main.c
+++ b/drivers/net/ethernet/cavium/thunder/nicvf_main.c
@@ -521,7 +521,8 @@ static int nicvf_init_resources(struct nicvf *nic)
 
 static void nicvf_snd_pkt_handler(struct net_device *netdev,
 				  struct cmp_queue *cq,
-				  struct cqe_send_t *cqe_tx, int cqe_type)
+				  struct cqe_send_t *cqe_tx,
+				  int cqe_type, int budget)
 {
 	struct sk_buff *skb = NULL;
 	struct nicvf *nic = netdev_priv(netdev);
@@ -545,7 +546,7 @@ static void nicvf_snd_pkt_handler(struct net_device *netdev,
 	if (skb) {
 		nicvf_put_sq_desc(sq, hdr->subdesc_cnt + 1);
 		prefetch(skb);
-		dev_consume_skb_any(skb);
+		napi_consume_skb(skb, budget);
 		sq->skbuff[cqe_tx->sqe_ptr] = (u64)NULL;
 	} else {
 		/* In case of HW TSO, HW sends a CQE for each segment of a TSO
@@ -700,7 +701,8 @@ loop:
 		break;
 		case CQE_TYPE_SEND:
 			nicvf_snd_pkt_handler(netdev, cq,
-					      (void *)cq_desc, CQE_TYPE_SEND);
+					      (void *)cq_desc, CQE_TYPE_SEND,
+					      budget);
 			tx_done++;
 		break;
 		case CQE_TYPE_INVALID:
-- 
2.7.4

[toc] | [standalone]


Back to top | Article view | linux.kernel


csiph-web