* [1/2] [NET]: Add NETIF_F_GEN_CSUM and NETIF_F_ALL_CSUM
@ 2006-06-13 11:25 Herbert Xu
2006-06-13 11:25 ` [2/2] [BRIDGE]: Add support for NETIF_F_HW_CSUM devices Herbert Xu
2006-06-18 5:06 ` [1/2] [NET]: Add NETIF_F_GEN_CSUM and NETIF_F_ALL_CSUM David Miller
0 siblings, 2 replies; 5+ messages in thread
From: Herbert Xu @ 2006-06-13 11:25 UTC (permalink / raw)
To: David S. Miller, Stephen Hemminger, netdev
[-- Attachment #1: Type: text/plain, Size: 755 bytes --]
Hi:
These patches are based on your net-2.6.18 tree. They add support for
NETIF_F_HW_CSUM on bridges.
[NET]: Add NETIF_F_GEN_CSUM and NETIF_F_ALL_CSUM
The current stack treats NETIF_F_HW_CSUM and NETIF_F_NO_CSUM identically
so we test for them in quite a few places. For the sake of brevity, I'm
adding the macro NETIF_F_GEN_CSUM for these two. We also test the disjunct
of NETIF_F_IP_CSUM and the other two in various places, for that purpose
I've added NETIF_F_ALL_CSUM.
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Cheers,
--
Visit Openswan at http://www.openswan.org/
Email: Herbert Xu ~{PmV>HI~} <herbert@gondor.apana.org.au>
Home Page: http://gondor.apana.org.au/~herbert/
PGP Key: http://gondor.apana.org.au/~herbert/pubkey.txt
[-- Attachment #2: p1.patch --]
[-- Type: text/plain, Size: 4741 bytes --]
diff --git a/drivers/net/bonding/bond_main.c b/drivers/net/bonding/bond_main.c
--- a/drivers/net/bonding/bond_main.c
+++ b/drivers/net/bonding/bond_main.c
@@ -1199,8 +1199,7 @@ int bond_sethwaddr(struct net_device *bo
}
#define BOND_INTERSECT_FEATURES \
- (NETIF_F_SG|NETIF_F_IP_CSUM|NETIF_F_NO_CSUM|NETIF_F_HW_CSUM|\
- NETIF_F_TSO|NETIF_F_UFO)
+ (NETIF_F_SG | NETIF_F_ALL_CSUM | NETIF_F_TSO | NETIF_F_UFO)
/*
* Compute the common dev->feature set available to all slaves. Some
@@ -1218,9 +1217,7 @@ static int bond_compute_features(struct
features &= (slave->dev->features & BOND_INTERSECT_FEATURES);
if ((features & NETIF_F_SG) &&
- !(features & (NETIF_F_IP_CSUM |
- NETIF_F_NO_CSUM |
- NETIF_F_HW_CSUM)))
+ !(features & NETIF_F_ALL_CSUM))
features &= ~NETIF_F_SG;
/*
diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h
--- a/include/linux/netdevice.h
+++ b/include/linux/netdevice.h
@@ -312,6 +312,9 @@ struct net_device
#define NETIF_F_LLTX 4096 /* LockLess TX */
#define NETIF_F_UFO 8192 /* Can offload UDP Large Send*/
+#define NETIF_F_GEN_CSUM (NETIF_F_NO_CSUM | NETIF_F_HW_CSUM)
+#define NETIF_F_ALL_CSUM (NETIF_F_IP_CSUM | NETIF_F_GEN_CSUM)
+
struct net_device *next_sched;
/* Interface index. Unique device identifier */
diff --git a/net/bridge/br_if.c b/net/bridge/br_if.c
--- a/net/bridge/br_if.c
+++ b/net/bridge/br_if.c
@@ -381,8 +381,7 @@ void br_features_recompute(struct net_br
checksum = br->feature_mask & NETIF_F_IP_CSUM;
list_for_each_entry(p, &br->port_list, list) {
- if (!(p->dev->features
- & (NETIF_F_IP_CSUM|NETIF_F_NO_CSUM|NETIF_F_HW_CSUM)))
+ if (!(p->dev->features & NETIF_F_ALL_CSUM))
checksum = 0;
features &= p->dev->features;
}
diff --git a/net/core/dev.c b/net/core/dev.c
--- a/net/core/dev.c
+++ b/net/core/dev.c
@@ -1284,7 +1284,7 @@ int dev_queue_xmit(struct sk_buff *skb)
* checksumming for this protocol, complete checksumming here.
*/
if (skb->ip_summed == CHECKSUM_HW &&
- (!(dev->features & (NETIF_F_HW_CSUM | NETIF_F_NO_CSUM)) &&
+ (!(dev->features & NETIF_F_GEN_CSUM) &&
(!(dev->features & NETIF_F_IP_CSUM) ||
skb->protocol != htons(ETH_P_IP))))
if (skb_checksum_help(skb, 0))
@@ -2789,9 +2789,7 @@ int register_netdevice(struct net_device
/* Fix illegal SG+CSUM combinations. */
if ((dev->features & NETIF_F_SG) &&
- !(dev->features & (NETIF_F_IP_CSUM |
- NETIF_F_NO_CSUM |
- NETIF_F_HW_CSUM))) {
+ !(dev->features & NETIF_F_ALL_CSUM)) {
printk("%s: Dropping NETIF_F_SG since no checksum feature.\n",
dev->name);
dev->features &= ~NETIF_F_SG;
diff --git a/net/core/ethtool.c b/net/core/ethtool.c
--- a/net/core/ethtool.c
+++ b/net/core/ethtool.c
@@ -30,7 +30,7 @@ u32 ethtool_op_get_link(struct net_devic
u32 ethtool_op_get_tx_csum(struct net_device *dev)
{
- return (dev->features & (NETIF_F_IP_CSUM | NETIF_F_HW_CSUM)) != 0;
+ return (dev->features & NETIF_F_ALL_CSUM) != 0;
}
int ethtool_op_set_tx_csum(struct net_device *dev, u32 data)
@@ -551,9 +551,7 @@ static int ethtool_set_sg(struct net_dev
return -EFAULT;
if (edata.data &&
- !(dev->features & (NETIF_F_IP_CSUM |
- NETIF_F_NO_CSUM |
- NETIF_F_HW_CSUM)))
+ !(dev->features & NETIF_F_ALL_CSUM))
return -EINVAL;
return __ethtool_set_sg(dev, edata.data);
diff --git a/net/ipv4/ip_output.c b/net/ipv4/ip_output.c
--- a/net/ipv4/ip_output.c
+++ b/net/ipv4/ip_output.c
@@ -839,7 +839,7 @@ int ip_append_data(struct sock *sk,
*/
if (transhdrlen &&
length + fragheaderlen <= mtu &&
- rt->u.dst.dev->features&(NETIF_F_IP_CSUM|NETIF_F_NO_CSUM|NETIF_F_HW_CSUM) &&
+ rt->u.dst.dev->features & NETIF_F_ALL_CSUM &&
!exthdrlen)
csummode = CHECKSUM_HW;
diff --git a/net/ipv4/tcp.c b/net/ipv4/tcp.c
--- a/net/ipv4/tcp.c
+++ b/net/ipv4/tcp.c
@@ -622,14 +622,10 @@ ssize_t tcp_sendpage(struct socket *sock
ssize_t res;
struct sock *sk = sock->sk;
-#define TCP_ZC_CSUM_FLAGS (NETIF_F_IP_CSUM | NETIF_F_NO_CSUM | NETIF_F_HW_CSUM)
-
if (!(sk->sk_route_caps & NETIF_F_SG) ||
- !(sk->sk_route_caps & TCP_ZC_CSUM_FLAGS))
+ !(sk->sk_route_caps & NETIF_F_ALL_CSUM))
return sock_no_sendpage(sock, page, offset, size, flags);
-#undef TCP_ZC_CSUM_FLAGS
-
lock_sock(sk);
TCP_CHECK_TIMER(sk);
res = do_tcp_sendpages(sk, &page, offset, size, flags);
@@ -726,9 +722,7 @@ new_segment:
/*
* Check whether we can use HW checksum.
*/
- if (sk->sk_route_caps &
- (NETIF_F_IP_CSUM | NETIF_F_NO_CSUM |
- NETIF_F_HW_CSUM))
+ if (sk->sk_route_caps & NETIF_F_ALL_CSUM)
skb->ip_summed = CHECKSUM_HW;
skb_entail(sk, tp, skb);
^ permalink raw reply [flat|nested] 5+ messages in thread* [2/2] [BRIDGE]: Add support for NETIF_F_HW_CSUM devices
2006-06-13 11:25 [1/2] [NET]: Add NETIF_F_GEN_CSUM and NETIF_F_ALL_CSUM Herbert Xu
@ 2006-06-13 11:25 ` Herbert Xu
2006-06-15 20:55 ` Stephen Hemminger
2006-06-18 5:06 ` David Miller
2006-06-18 5:06 ` [1/2] [NET]: Add NETIF_F_GEN_CSUM and NETIF_F_ALL_CSUM David Miller
1 sibling, 2 replies; 5+ messages in thread
From: Herbert Xu @ 2006-06-13 11:25 UTC (permalink / raw)
To: David S. Miller, Stephen Hemminger, netdev
[-- Attachment #1: Type: text/plain, Size: 610 bytes --]
Hi:
[BRIDGE]: Add support for NETIF_F_HW_CSUM devices
As it is the bridge will only ever declare NETIF_F_IP_CSUM even if all
its constituent devices support NETIF_F_HW_CSUM. This patch fixes this
by supporting the first one out of NETIF_F_NO_CSUM, NETIF_F_HW_CSUM,
and NETIF_F_IP_CSUM that is supported by all constituent devices.
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Cheers,
--
Visit Openswan at http://www.openswan.org/
Email: Herbert Xu ~{PmV>HI~} <herbert@gondor.apana.org.au>
Home Page: http://gondor.apana.org.au/~herbert/
PGP Key: http://gondor.apana.org.au/~herbert/pubkey.txt
[-- Attachment #2: p2.patch --]
[-- Type: text/plain, Size: 1631 bytes --]
diff --git a/net/bridge/br_device.c b/net/bridge/br_device.c
--- a/net/bridge/br_device.c
+++ b/net/bridge/br_device.c
@@ -145,9 +145,9 @@ static int br_set_tx_csum(struct net_dev
struct net_bridge *br = netdev_priv(dev);
if (data)
- br->feature_mask |= NETIF_F_IP_CSUM;
+ br->feature_mask |= NETIF_F_NO_CSUM;
else
- br->feature_mask &= ~NETIF_F_IP_CSUM;
+ br->feature_mask &= ~NETIF_F_ALL_CSUM;
br_features_recompute(br);
return 0;
@@ -185,5 +185,5 @@ void br_dev_setup(struct net_device *dev
dev->priv_flags = IFF_EBRIDGE;
dev->features = NETIF_F_SG | NETIF_F_FRAGLIST
- | NETIF_F_HIGHDMA | NETIF_F_TSO | NETIF_F_IP_CSUM;
+ | NETIF_F_HIGHDMA | NETIF_F_TSO | NETIF_F_NO_CSUM;
}
diff --git a/net/bridge/br_if.c b/net/bridge/br_if.c
--- a/net/bridge/br_if.c
+++ b/net/bridge/br_if.c
@@ -377,11 +377,17 @@ void br_features_recompute(struct net_br
struct net_bridge_port *p;
unsigned long features, checksum;
- features = br->feature_mask &~ NETIF_F_IP_CSUM;
- checksum = br->feature_mask & NETIF_F_IP_CSUM;
+ checksum = br->feature_mask & NETIF_F_ALL_CSUM ? NETIF_F_NO_CSUM : 0;
+ features = br->feature_mask & ~NETIF_F_ALL_CSUM;
list_for_each_entry(p, &br->port_list, list) {
- if (!(p->dev->features & NETIF_F_ALL_CSUM))
+ if (checksum & NETIF_F_NO_CSUM &&
+ !(p->dev->features & NETIF_F_NO_CSUM))
+ checksum ^= NETIF_F_NO_CSUM | NETIF_F_HW_CSUM;
+ if (checksum & NETIF_F_HW_CSUM &&
+ !(p->dev->features & NETIF_F_HW_CSUM))
+ checksum ^= NETIF_F_HW_CSUM | NETIF_F_IP_CSUM;
+ if (!(p->dev->features & NETIF_F_IP_CSUM))
checksum = 0;
features &= p->dev->features;
}
^ permalink raw reply [flat|nested] 5+ messages in thread* Re: [2/2] [BRIDGE]: Add support for NETIF_F_HW_CSUM devices
2006-06-13 11:25 ` [2/2] [BRIDGE]: Add support for NETIF_F_HW_CSUM devices Herbert Xu
@ 2006-06-15 20:55 ` Stephen Hemminger
2006-06-18 5:06 ` David Miller
1 sibling, 0 replies; 5+ messages in thread
From: Stephen Hemminger @ 2006-06-15 20:55 UTC (permalink / raw)
To: Herbert Xu; +Cc: David S. Miller, netdev
On Tue, 13 Jun 2006 21:25:43 +1000
Herbert Xu <herbert@gondor.apana.org.au> wrote:
> Hi:
>
> [BRIDGE]: Add support for NETIF_F_HW_CSUM devices
>
> As it is the bridge will only ever declare NETIF_F_IP_CSUM even if all
> its constituent devices support NETIF_F_HW_CSUM. This patch fixes this
> by supporting the first one out of NETIF_F_NO_CSUM, NETIF_F_HW_CSUM,
> and NETIF_F_IP_CSUM that is supported by all constituent devices.
>
> Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
>
> Cheers,
Looks fine.
^ permalink raw reply [flat|nested] 5+ messages in thread
* Re: [2/2] [BRIDGE]: Add support for NETIF_F_HW_CSUM devices
2006-06-13 11:25 ` [2/2] [BRIDGE]: Add support for NETIF_F_HW_CSUM devices Herbert Xu
2006-06-15 20:55 ` Stephen Hemminger
@ 2006-06-18 5:06 ` David Miller
1 sibling, 0 replies; 5+ messages in thread
From: David Miller @ 2006-06-18 5:06 UTC (permalink / raw)
To: herbert; +Cc: shemminger, netdev
From: Herbert Xu <herbert@gondor.apana.org.au>
Date: Tue, 13 Jun 2006 21:25:43 +1000
> [BRIDGE]: Add support for NETIF_F_HW_CSUM devices
>
> As it is the bridge will only ever declare NETIF_F_IP_CSUM even if all
> its constituent devices support NETIF_F_HW_CSUM. This patch fixes this
> by supporting the first one out of NETIF_F_NO_CSUM, NETIF_F_HW_CSUM,
> and NETIF_F_IP_CSUM that is supported by all constituent devices.
>
> Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Also applied, thanks a lot!
^ permalink raw reply [flat|nested] 5+ messages in thread
* Re: [1/2] [NET]: Add NETIF_F_GEN_CSUM and NETIF_F_ALL_CSUM
2006-06-13 11:25 [1/2] [NET]: Add NETIF_F_GEN_CSUM and NETIF_F_ALL_CSUM Herbert Xu
2006-06-13 11:25 ` [2/2] [BRIDGE]: Add support for NETIF_F_HW_CSUM devices Herbert Xu
@ 2006-06-18 5:06 ` David Miller
1 sibling, 0 replies; 5+ messages in thread
From: David Miller @ 2006-06-18 5:06 UTC (permalink / raw)
To: herbert; +Cc: shemminger, netdev
From: Herbert Xu <herbert@gondor.apana.org.au>
Date: Tue, 13 Jun 2006 21:25:11 +1000
> [NET]: Add NETIF_F_GEN_CSUM and NETIF_F_ALL_CSUM
>
> The current stack treats NETIF_F_HW_CSUM and NETIF_F_NO_CSUM identically
> so we test for them in quite a few places. For the sake of brevity, I'm
> adding the macro NETIF_F_GEN_CSUM for these two. We also test the disjunct
> of NETIF_F_IP_CSUM and the other two in various places, for that purpose
> I've added NETIF_F_ALL_CSUM.
>
> Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Applied, thanks Herbert.
^ permalink raw reply [flat|nested] 5+ messages in thread
end of thread, other threads:[~2006-06-18 5:06 UTC | newest]
Thread overview: 5+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2006-06-13 11:25 [1/2] [NET]: Add NETIF_F_GEN_CSUM and NETIF_F_ALL_CSUM Herbert Xu
2006-06-13 11:25 ` [2/2] [BRIDGE]: Add support for NETIF_F_HW_CSUM devices Herbert Xu
2006-06-15 20:55 ` Stephen Hemminger
2006-06-18 5:06 ` David Miller
2006-06-18 5:06 ` [1/2] [NET]: Add NETIF_F_GEN_CSUM and NETIF_F_ALL_CSUM David Miller
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox