From: Brice Goglin <Brice.Goglin@inria.fr>
To: "David S. Miller" <davem@davemloft.net>
Cc: "Andrew J. Gallatin" <gallatin@myri.com>,
Jan-Bernd Themann <themann@de.ibm.com>,
Linus Torvalds <torvalds@osdl.org>,
LKML <linux-kernel@vger.kernel.org>,
netdev@vger.kernel.org
Subject: [PATCH][LRO] Fix lro_mgr->features checks
Date: Mon, 07 Jan 2008 14:33:32 +0100 [thread overview]
Message-ID: <47822A2C.4020303@inria.fr> (raw)
Hi Dave,
The LRO_F_* checks in inet_lro.c are buggy, the following patch
is needed to check lro_mgr->features correctly.
I decided to follow the NETIF_F_* convention and thus removed
test_bit. Some people might like it better if we keep using
test_bit and just set lro_mgr->features to 1<<LRO_F_NAPI instead
of just LRO_F_NAPI. I don't know what's the best.
Without this patch, we noticed several problems with myri10ge,
at least when using non-IP packets. I guess the ehea driver did
not expose any big problem because test_bit(NAPI) returns true
thanks to lro_mgr->features containing LRO_F_EXTRACT_VLAN_ID.
Please apply for 2.6.24.
Brice
[PATCH][LRO] Fix lro_mgr->features checks
lro_mgr->features contains a bitmask of LRO_F_* values which are
defined as power of two, not as bit indexes.
They must be checked with x&LRO_F_FOO, not with test_bit(LRO_F_FOO,&x).
Signed-off-by: Brice Goglin <Brice.Goglin@inria.fr>
Acked-by: Andrew Gallatin <gallatin@myri.com>
---
diff --git a/net/ipv4/inet_lro.c b/net/ipv4/inet_lro.c
index 9a96c27..4a4d49f 100644
--- a/net/ipv4/inet_lro.c
+++ b/net/ipv4/inet_lro.c
@@ -310,7 +310,7 @@ static void lro_flush(struct net_lro_mgr *lro_mgr,
skb_shinfo(lro_desc->parent)->gso_size = lro_desc->mss;
if (lro_desc->vgrp) {
- if (test_bit(LRO_F_NAPI, &lro_mgr->features))
+ if (lro_mgr->features & LRO_F_NAPI)
vlan_hwaccel_receive_skb(lro_desc->parent,
lro_desc->vgrp,
lro_desc->vlan_tag);
@@ -320,7 +320,7 @@ static void lro_flush(struct net_lro_mgr *lro_mgr,
lro_desc->vlan_tag);
} else {
- if (test_bit(LRO_F_NAPI, &lro_mgr->features))
+ if (lro_mgr->features & LRO_F_NAPI)
netif_receive_skb(lro_desc->parent);
else
netif_rx(lro_desc->parent);
@@ -352,7 +352,7 @@ static int __lro_proc_skb(struct net_lro_mgr *lro_mgr, struct sk_buff *skb,
goto out;
if ((skb->protocol == htons(ETH_P_8021Q))
- && !test_bit(LRO_F_EXTRACT_VLAN_ID, &lro_mgr->features))
+ && !(lro_mgr->features & LRO_F_EXTRACT_VLAN_ID))
vlan_hdr_len = VLAN_HLEN;
if (!lro_desc->active) { /* start new lro session */
@@ -474,7 +474,7 @@ static struct sk_buff *__lro_proc_segment(struct net_lro_mgr *lro_mgr,
goto out;
if ((skb->protocol == htons(ETH_P_8021Q))
- && !test_bit(LRO_F_EXTRACT_VLAN_ID, &lro_mgr->features))
+ && !(lro_mgr->features & LRO_F_EXTRACT_VLAN_ID))
vlan_hdr_len = VLAN_HLEN;
iph = (void *)(skb->data + vlan_hdr_len);
@@ -516,7 +516,7 @@ void lro_receive_skb(struct net_lro_mgr *lro_mgr,
void *priv)
{
if (__lro_proc_skb(lro_mgr, skb, NULL, 0, priv)) {
- if (test_bit(LRO_F_NAPI, &lro_mgr->features))
+ if (lro_mgr->features & LRO_F_NAPI)
netif_receive_skb(skb);
else
netif_rx(skb);
@@ -531,7 +531,7 @@ void lro_vlan_hwaccel_receive_skb(struct net_lro_mgr *lro_mgr,
void *priv)
{
if (__lro_proc_skb(lro_mgr, skb, vgrp, vlan_tag, priv)) {
- if (test_bit(LRO_F_NAPI, &lro_mgr->features))
+ if (lro_mgr->features & LRO_F_NAPI)
vlan_hwaccel_receive_skb(skb, vgrp, vlan_tag);
else
vlan_hwaccel_rx(skb, vgrp, vlan_tag);
@@ -550,7 +550,7 @@ void lro_receive_frags(struct net_lro_mgr *lro_mgr,
if (!skb)
return;
- if (test_bit(LRO_F_NAPI, &lro_mgr->features))
+ if (lro_mgr->features & LRO_F_NAPI)
netif_receive_skb(skb);
else
netif_rx(skb);
@@ -570,7 +570,7 @@ void lro_vlan_hwaccel_receive_frags(struct net_lro_mgr *lro_mgr,
if (!skb)
return;
- if (test_bit(LRO_F_NAPI, &lro_mgr->features))
+ if (lro_mgr->features & LRO_F_NAPI)
vlan_hwaccel_receive_skb(skb, vgrp, vlan_tag);
else
vlan_hwaccel_rx(skb, vgrp, vlan_tag);
next reply other threads:[~2008-01-07 14:00 UTC|newest]
Thread overview: 2+ messages / expand[flat|nested] mbox.gz Atom feed top
2008-01-07 13:33 Brice Goglin [this message]
2008-01-08 6:09 ` [PATCH][LRO] Fix lro_mgr->features checks David Miller
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=47822A2C.4020303@inria.fr \
--to=brice.goglin@inria.fr \
--cc=davem@davemloft.net \
--cc=gallatin@myri.com \
--cc=linux-kernel@vger.kernel.org \
--cc=netdev@vger.kernel.org \
--cc=themann@de.ibm.com \
--cc=torvalds@osdl.org \
/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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.