From: Daode Huang <huangdaode@hisilicon.com>
To: <davem@davemloft.net>
Cc: <liguozhu@hisilicon.com>, <Yisen.Zhuang@huawei.com>,
<linux-kernel@vger.kernel.org>,
<linux-arm-kernel@lists.infradead.org>, <netdev@vger.kernel.org>,
<linuxarm@huawei.com>, <salil.mehta@huawei.com>,
<huangdaode@hisilicon.com>, <kenneth-lee-2012@foxmail.com>,
<xuwei5@hisilicon.com>, <lisheng011@huawei.com>,
<yankejian@huawei.com>
Subject: [PATCH net 1/6] net: hns: bug fix about the overflow of mss
Date: Fri, 4 Mar 2016 09:09:53 +0800 [thread overview]
Message-ID: <1457053798-93513-2-git-send-email-huangdaode@hisilicon.com> (raw)
In-Reply-To: <1457053798-93513-1-git-send-email-huangdaode@hisilicon.com>
When set MTU to the minimum value 68, there are increasing number
of error packets occur, which is caused by the overflowed value of
mss. This patch fix the bug.
This patch is just for huawei internal review, if there is no
comment for three days, I will send out to the mainline.
thanks
Signed-off-by: Daode Huang <huangdaode@hisilicon.com>
---
drivers/net/ethernet/hisilicon/hns/hns_enet.c | 17 ++++++++---------
1 file changed, 8 insertions(+), 9 deletions(-)
diff --git a/drivers/net/ethernet/hisilicon/hns/hns_enet.c b/drivers/net/ethernet/hisilicon/hns/hns_enet.c
index 3f77ff7..9d46d57 100644
--- a/drivers/net/ethernet/hisilicon/hns/hns_enet.c
+++ b/drivers/net/ethernet/hisilicon/hns/hns_enet.c
@@ -48,7 +48,6 @@ static void fill_v2_desc(struct hnae_ring *ring, void *priv,
struct iphdr *iphdr;
struct ipv6hdr *ipv6hdr;
struct sk_buff *skb;
- int skb_tmp_len;
__be16 protocol;
u8 bn_pid = 0;
u8 rrcfv = 0;
@@ -90,13 +89,13 @@ static void fill_v2_desc(struct hnae_ring *ring, void *priv,
hnae_set_bit(rrcfv, HNSV2_TXD_L4CS_B, 1);
/* check for tcp/udp header */
- if (iphdr->protocol == IPPROTO_TCP) {
+ if (iphdr->protocol == IPPROTO_TCP &&
+ skb_is_gso(skb)) {
hnae_set_bit(tvsvsn,
HNSV2_TXD_TSE_B, 1);
- skb_tmp_len = SKB_TMP_LEN(skb);
l4_len = tcp_hdrlen(skb);
- mss = mtu - skb_tmp_len - ETH_FCS_LEN;
- paylen = skb->len - skb_tmp_len;
+ mss = skb_shinfo(skb)->gso_size;
+ paylen = skb->len - SKB_TMP_LEN(skb);
}
} else if (skb->protocol == htons(ETH_P_IPV6)) {
hnae_set_bit(tvsvsn, HNSV2_TXD_IPV6_B, 1);
@@ -104,13 +103,13 @@ static void fill_v2_desc(struct hnae_ring *ring, void *priv,
hnae_set_bit(rrcfv, HNSV2_TXD_L4CS_B, 1);
/* check for tcp/udp header */
- if (ipv6hdr->nexthdr == IPPROTO_TCP) {
+ if (ipv6hdr->nexthdr == IPPROTO_TCP &&
+ skb_is_gso(skb) && skb_is_gso_v6(skb)) {
hnae_set_bit(tvsvsn,
HNSV2_TXD_TSE_B, 1);
- skb_tmp_len = SKB_TMP_LEN(skb);
l4_len = tcp_hdrlen(skb);
- mss = mtu - skb_tmp_len - ETH_FCS_LEN;
- paylen = skb->len - skb_tmp_len;
+ mss = skb_shinfo(skb)->gso_size;
+ paylen = skb->len - SKB_TMP_LEN(skb);
}
}
desc->tx.ip_offset = ip_offset;
--
1.9.1
next prev parent reply other threads:[~2016-03-04 0:51 UTC|newest]
Thread overview: 15+ messages / expand[flat|nested] mbox.gz Atom feed top
2016-03-04 1:09 [PATCH net 0/6]net: hns: hns driver updates Daode Huang
2016-03-04 0:55 ` Daode Huang
2016-03-04 1:09 ` Daode Huang [this message]
2016-03-04 1:09 ` [PATCH net 2/6] net: hns: fixes the hw interrupt bug in using napi Daode Huang
2016-03-04 1:09 ` [PATCH net 3/6] net: hns: fixed portid bug in sending manage pkt Daode Huang
2016-03-04 13:37 ` Sergei Shtylyov
2016-03-11 1:53 ` Daode Huang
2016-03-11 13:29 ` Sergei Shtylyov
2016-03-14 0:45 ` Daode Huang
2016-03-04 1:09 ` [PATCH net 4/6] net: hns: adds uc match for debug port Daode Huang
2016-03-04 13:39 ` Sergei Shtylyov
2016-03-11 1:58 ` Daode Huang
2016-03-04 1:09 ` [PATCH net 5/6] net: hns: fixed service-ges setting MAC-addr bug Daode Huang
2016-03-04 1:09 ` [PATCH net 6/6] net: hns: bug fix of getting hilink status Daode Huang
-- strict thread matches above, loose matches on Subject: below --
2016-03-04 3:05 [PATCH net 0/6]net: hns: hns driver updates Daode Huang
2016-03-04 3:05 ` [PATCH net 1/6] net: hns: bug fix about the overflow of mss Daode Huang
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=1457053798-93513-2-git-send-email-huangdaode@hisilicon.com \
--to=huangdaode@hisilicon.com \
--cc=Yisen.Zhuang@huawei.com \
--cc=davem@davemloft.net \
--cc=kenneth-lee-2012@foxmail.com \
--cc=liguozhu@hisilicon.com \
--cc=linux-arm-kernel@lists.infradead.org \
--cc=linux-kernel@vger.kernel.org \
--cc=linuxarm@huawei.com \
--cc=lisheng011@huawei.com \
--cc=netdev@vger.kernel.org \
--cc=salil.mehta@huawei.com \
--cc=xuwei5@hisilicon.com \
--cc=yankejian@huawei.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).