* Re: [PATCH 5/5] net: Loopback ifindex is constant now
From: Eric Dumazet @ 2012-08-06 11:54 UTC (permalink / raw)
To: Pavel Emelyanov; +Cc: David Miller, Eric W. Biederman, Linux Netdev List
In-Reply-To: <501F9D2B.8000504@parallels.com>
On Mon, 2012-08-06 at 14:32 +0400, Pavel Emelyanov wrote:
> As pointed out, there are places, that access net->loopback_dev->ifindex
> and after ifindex generation is made per-net this value becomes constant
> equals 1. So go ahead and introduce the LOOPBACK_IFINDEX constant and use
> it where appropriate.
>
> Signed-off-by: Pavel Emelyanov <xemul@parallels.com>
What guarantee do we have that loopback is the first device per net ?
You should add this to the changelog because its not that obvious.
^ permalink raw reply
* Re: [PATCH 5/5] net: Loopback ifindex is constant now
From: Pavel Emelyanov @ 2012-08-06 12:13 UTC (permalink / raw)
To: Eric Dumazet; +Cc: David Miller, Eric W. Biederman, Linux Netdev List
In-Reply-To: <1344254071.26674.15.camel@edumazet-glaptop>
On 08/06/2012 03:54 PM, Eric Dumazet wrote:
> On Mon, 2012-08-06 at 14:32 +0400, Pavel Emelyanov wrote:
>> As pointed out, there are places, that access net->loopback_dev->ifindex
>> and after ifindex generation is made per-net this value becomes constant
>> equals 1. So go ahead and introduce the LOOPBACK_IFINDEX constant and use
>> it where appropriate.
>>
>> Signed-off-by: Pavel Emelyanov <xemul@parallels.com>
>
> What guarantee do we have that loopback is the first device per net ?
In net_dev_init():
/* The loopback device is special if any other network devices
* is present in a network namespace the loopback device must
* be present. Since we now dynamically allocate and free the
* loopback device ensure this invariant is maintained by
* keeping the loopback device as the first device on the
* list of network devices. Ensuring the loopback devices
* is the first device that appears and the last network device
* that disappears.
*/
if (register_pernet_device(&loopback_net_ops))
goto out;
> You should add this to the changelog because its not that obvious.
OK.
Thanks,
Pavel
^ permalink raw reply
* Re: [PATCH 1/7] netpoll: use GFP_ATOMIC in slave_enable_netpoll() and __netpoll_setup()
From: Cong Wang @ 2012-08-06 12:31 UTC (permalink / raw)
To: Eric Dumazet
Cc: netdev, David S. Miller, Jay Vosburgh, Andy Gospodarek,
Eric Dumazet, Cong Wang, Joe Perches, Neil Horman, linux-kernel
In-Reply-To: <1344246244.26674.0.camel@edumazet-glaptop>
On Mon, 2012-08-06 at 11:44 +0200, Eric Dumazet wrote:
> On Mon, 2012-08-06 at 17:08 +0800, Cong Wang wrote:
> > On Fri, 2012-08-03 at 12:10 +0200, Eric Dumazet wrote:
> > >
> > > I did this , just take it ;)
> >
> > Do we have to pass gfp to ->ndo_netpoll_setup() too? It seems no, so far
> > I don't think we have to do that.
> >
> > Thanks.
> >
>
> It is needed.
>
> ->ndo_netpoll_setup() is called from __netpoll_setup()
>
> So it would make no sense to allow GFP_KERNEL allocations in
> ndo_netpoll_setup() hgandlers if we called __netpoll_setup() with
> GFP_ATOMIC gfp
>
I see your point. Thanks!
^ permalink raw reply
* [PATCH] llc: free the right skb
From: Sorin Dumitru @ 2012-08-06 12:35 UTC (permalink / raw)
To: netdev; +Cc: acme, davem, Sorin Dumitru
We are freeing skb instead of nskb, resulting in a double
free on skb and a leak from nskb.
Signed-off-by: Sorin Dumitru <sdumitru@ixiacom.com>
---
net/llc/llc_station.c | 6 +++---
1 file changed, 3 insertions(+), 3 deletions(-)
diff --git a/net/llc/llc_station.c b/net/llc/llc_station.c
index 39a8d89..6828e39 100644
--- a/net/llc/llc_station.c
+++ b/net/llc/llc_station.c
@@ -268,7 +268,7 @@ static int llc_station_ac_send_null_dsap_xid_c(struct sk_buff *skb)
out:
return rc;
free:
- kfree_skb(skb);
+ kfree_skb(nskb);
goto out;
}
@@ -293,7 +293,7 @@ static int llc_station_ac_send_xid_r(struct sk_buff *skb)
out:
return rc;
free:
- kfree_skb(skb);
+ kfree_skb(nskb);
goto out;
}
@@ -322,7 +322,7 @@ static int llc_station_ac_send_test_r(struct sk_buff *skb)
out:
return rc;
free:
- kfree_skb(skb);
+ kfree_skb(nskb);
goto out;
}
--
1.7.11.4
^ permalink raw reply related
* [QUESTION] Push a driver upstream
From: Lin Ming @ 2012-08-06 12:46 UTC (permalink / raw)
To: netdev
Hi list,
I bought a USB ethernet device and found the Linux driver on CD.
I asked the author if he'd like to push the driver upstream, but no response.
So I wonder if it's legal for me to push the driver upstream.
Do I need the author's agreement?
Below is the driver header.
/*
* QF9700 one chip USB 1.1 ethernet devices
*
* Author : jokeliujl <jokeliu@163.com>
* Date : 2010-10-01
*
* This file is licensed under the terms of the GNU General Public License
* version 2. This program is licensed "as is" without any warranty of any
* kind, whether express or implied.
*/
Thanks,
Lin Ming
^ permalink raw reply
* Re: skb_warn_bad_offload with kernel 3.5 (maybe gso/bridge related ?)
From: Herbert Xu @ 2012-08-06 13:42 UTC (permalink / raw)
To: Eric Dumazet; +Cc: Ben Hutchings, Yann Dupont, netdev@vger.kernel.org
In-Reply-To: <1344250640.26674.9.camel@edumazet-glaptop>
On Mon, Aug 06, 2012 at 12:57:20PM +0200, Eric Dumazet wrote:
> On Mon, 2012-08-06 at 17:18 +0800, Herbert Xu wrote:
> > On Fri, Aug 03, 2012 at 04:40:08PM +0100, Ben Hutchings wrote:
> > >
> > > I think GRO is doing the right thing, and I can't think why we should
> > > see ip_summed = CHECKSUM_PARTIAL if the skb is forwarded by a bridge. I
> >
> > If it is was created by GRO then it should always be CHECKSUM_PARTIAL,
> > even if it came through a bridge.
>
> Not sure where in GRO we set ip_summed to CHECKSUM_PARTIAL ?
tcp_gro_complete
Chers,
--
Email: Herbert Xu <herbert@gondor.apana.org.au>
Home Page: http://gondor.apana.org.au/~herbert/
PGP Key: http://gondor.apana.org.au/~herbert/pubkey.txt
^ permalink raw reply
* Re: skb_warn_bad_offload with kernel 3.5 (maybe gso/bridge related ?)
From: Eric Dumazet @ 2012-08-06 13:54 UTC (permalink / raw)
To: Herbert Xu; +Cc: Ben Hutchings, Yann Dupont, netdev@vger.kernel.org
In-Reply-To: <20120806134232.GA19849@gondor.apana.org.au>
On Mon, 2012-08-06 at 21:42 +0800, Herbert Xu wrote:
> tcp_gro_complete
>
Oops thats right, thanks ;)
^ permalink raw reply
* [PATCH] ip: fix error handling in ip_finish_output2()
From: Vasily Kulikov @ 2012-08-06 13:55 UTC (permalink / raw)
To: David S. Miller
Cc: Alexey Kuznetsov, James Morris, Hideaki YOSHIFUJI,
Patrick McHardy, netdev, linux-kernel
__neigh_create() returns either a pointer to struct neighbour or PTR_ERR().
But the caller expects it to return either a pointer or NULL. Replace
the NULL check with IS_ERR() check.
The bug was introduced in a263b3093641fb1ec377582c90986a7fd0625184.
Signed-off-by: Vasily Kulikov <segoon@openwall.com>
---
Compile tested only.
net/ipv4/ip_output.c | 2 +-
1 files changed, 1 insertions(+), 1 deletions(-)
diff --git a/net/ipv4/ip_output.c b/net/ipv4/ip_output.c
index ba39a52..76dde25 100644
--- a/net/ipv4/ip_output.c
+++ b/net/ipv4/ip_output.c
@@ -197,7 +197,7 @@ static inline int ip_finish_output2(struct sk_buff *skb)
neigh = __ipv4_neigh_lookup_noref(dev, nexthop);
if (unlikely(!neigh))
neigh = __neigh_create(&arp_tbl, &nexthop, dev, false);
- if (neigh) {
+ if (!IS_ERR(neigh)) {
int res = dst_neigh_output(dst, neigh, skb);
rcu_read_unlock_bh();
--
1.7.0.4
^ permalink raw reply related
* [PATCH net-next 0/6] Per-net and on-demand link indices (and related) v2
From: Pavel Emelyanov @ 2012-08-06 14:13 UTC (permalink / raw)
To: David Miller, Eric Dumazet, Eric W. Biederman, Linux Netdev List
Hi!
This set tries to summarize the recent discussion of making ifindices friendly
to checkpoint-restore and consists of:
1. Prepare hash function to non-unique ifindices
2. Allow for specifying the desired ifindex on net link creation
3. Make ifindex generation per-net
4. Simplify loopback device ifindex access
Changes since v1:
* Fixed a stupid mistake with a pointer bits shift
* Turned the netdev_hash_mix routine into the generic ptr_hash_mix one
* Added a comment describing why loopback index is always 1 after the patch
Thanks,
Pavel
^ permalink raw reply
* [PATCH 1/6] hash: Introduce ptr_hash_mix routine
From: Pavel Emelyanov @ 2012-08-06 14:13 UTC (permalink / raw)
To: David Miller, Eric Dumazet, Eric W. Biederman, Linux Netdev List
In-Reply-To: <501FD0F2.4040609@parallels.com>
This one is used to make a salt out of a pointer to be mixed to some
hash function later. Idea and implementation are proposed by Eric Dumazet.
Signed-off-by: Pavel Emelyanov <xemul@parallels.com>
---
include/linux/hash.h | 10 ++++++++++
include/net/netns/hash.h | 9 ++-------
2 files changed, 12 insertions(+), 7 deletions(-)
diff --git a/include/linux/hash.h b/include/linux/hash.h
index b80506b..1bd0ab1 100644
--- a/include/linux/hash.h
+++ b/include/linux/hash.h
@@ -14,6 +14,7 @@
* machines where multiplications are slow.
*/
+#include <linux/cache.h>
#include <asm/types.h>
/* 2^31 + 2^29 - 2^25 + 2^22 - 2^19 - 2^16 + 1 */
@@ -67,4 +68,13 @@ static inline unsigned long hash_ptr(const void *ptr, unsigned int bits)
{
return hash_long((unsigned long)ptr, bits);
}
+
+static inline u32 ptr_hash_mix(const void *ptr)
+{
+#if BITS_PER_LONG == 32
+ return (u32)(unsigned long)ptr;
+#else
+ return (u32)((unsigned long)ptr >> L1_CACHE_SHIFT);
+#endif
+}
#endif /* _LINUX_HASH_H */
diff --git a/include/net/netns/hash.h b/include/net/netns/hash.h
index c06ac58..bcdabe0 100644
--- a/include/net/netns/hash.h
+++ b/include/net/netns/hash.h
@@ -1,19 +1,14 @@
#ifndef __NET_NS_HASH_H__
#define __NET_NS_HASH_H__
-#include <asm/cache.h>
+#include <linux/hash.h>
struct net;
static inline unsigned int net_hash_mix(struct net *net)
{
#ifdef CONFIG_NET_NS
- /*
- * shift this right to eliminate bits, that are
- * always zeroed
- */
-
- return (unsigned)(((unsigned long)net) >> L1_CACHE_SHIFT);
+ return ptr_hash_mix(net);
#else
return 0;
#endif
--
1.7.6.5
^ permalink raw reply related
* [PATCH 2/6] net: Dont use ifindices in hash fns
From: Pavel Emelyanov @ 2012-08-06 14:14 UTC (permalink / raw)
To: David Miller, Eric Dumazet, Eric W. Biederman, Linux Netdev List
In-Reply-To: <501FD0F2.4040609@parallels.com>
Eric noticed, that when there will be devices with equal indices, some
hash functions that use them will become less effective as they could.
Fix this in advance by taking the net_device address into calculations
instead of the device index. Since the net_device is always aligned in
memory, shift the pointer to eliminate always zero bits (like we do it
in net_hash_mix).
This is true for arp and ndisc hash fns. The netlabel, can and llc ones
are also ifindex-based, but that three are init_net-only, thus will not
be affected.
Signed-off-by: Pavel Emelyanov <xemul@parallels.com>
---
include/net/arp.h | 3 ++-
include/net/ndisc.h | 3 ++-
2 files changed, 4 insertions(+), 2 deletions(-)
diff --git a/include/net/arp.h b/include/net/arp.h
index 7f7df93..b9ce3a9 100644
--- a/include/net/arp.h
+++ b/include/net/arp.h
@@ -3,6 +3,7 @@
#define _ARP_H
#include <linux/if_arp.h>
+#include <linux/hash.h>
#include <net/neighbour.h>
@@ -10,7 +11,7 @@ extern struct neigh_table arp_tbl;
static inline u32 arp_hashfn(u32 key, const struct net_device *dev, u32 hash_rnd)
{
- u32 val = key ^ dev->ifindex;
+ u32 val = key ^ ptr_hash_mix(dev);
return val * hash_rnd;
}
diff --git a/include/net/ndisc.h b/include/net/ndisc.h
index 96a3b5c..a9b55b2 100644
--- a/include/net/ndisc.h
+++ b/include/net/ndisc.h
@@ -49,6 +49,7 @@ enum {
#include <linux/types.h>
#include <linux/if_arp.h>
#include <linux/netdevice.h>
+#include <linux/hash.h>
#include <net/neighbour.h>
@@ -134,7 +135,7 @@ static inline u32 ndisc_hashfn(const void *pkey, const struct net_device *dev, _
{
const u32 *p32 = pkey;
- return (((p32[0] ^ dev->ifindex) * hash_rnd[0]) +
+ return (((p32[0] ^ ptr_hash_mix(dev)) * hash_rnd[0]) +
(p32[1] * hash_rnd[1]) +
(p32[2] * hash_rnd[2]) +
(p32[3] * hash_rnd[3]));
--
1.7.6.5
^ permalink raw reply related
* [PATCH 3/6] net: Allow to create links with given ifindex
From: Pavel Emelyanov @ 2012-08-06 14:14 UTC (permalink / raw)
To: David Miller, Eric Dumazet, Eric W. Biederman, Linux Netdev List
In-Reply-To: <501FD0F2.4040609@parallels.com>
Currently the RTM_NEWLINK results in -EOPNOTSUPP if the ifinfomsg->ifi_index
is not zero. I propose to allow requesting ifindices on link creation. This
is required by the checkpoint-restore to correctly restore a net namespace
(i.e. -- a container).
Signed-off-by: Pavel Emelyanov <xemul@parallels.com>
---
net/core/dev.c | 7 ++++++-
net/core/rtnetlink.c | 12 +++++++-----
2 files changed, 13 insertions(+), 6 deletions(-)
diff --git a/net/core/dev.c b/net/core/dev.c
index f91abf8..3ca300d 100644
--- a/net/core/dev.c
+++ b/net/core/dev.c
@@ -5579,7 +5579,12 @@ int register_netdevice(struct net_device *dev)
}
}
- dev->ifindex = dev_new_index(net);
+ ret = -EBUSY;
+ if (!dev->ifindex)
+ dev->ifindex = dev_new_index(net);
+ else if (__dev_get_by_index(net, dev->ifindex))
+ goto err_uninit;
+
if (dev->iflink == -1)
dev->iflink = dev->ifindex;
diff --git a/net/core/rtnetlink.c b/net/core/rtnetlink.c
index 2c5a0a0..1aa1456 100644
--- a/net/core/rtnetlink.c
+++ b/net/core/rtnetlink.c
@@ -1812,8 +1812,6 @@ replay:
return -ENODEV;
}
- if (ifm->ifi_index)
- return -EOPNOTSUPP;
if (tb[IFLA_MAP] || tb[IFLA_MASTER] || tb[IFLA_PROTINFO])
return -EOPNOTSUPP;
@@ -1839,10 +1837,14 @@ replay:
return PTR_ERR(dest_net);
dev = rtnl_create_link(net, dest_net, ifname, ops, tb);
-
- if (IS_ERR(dev))
+ if (IS_ERR(dev)) {
err = PTR_ERR(dev);
- else if (ops->newlink)
+ goto out;
+ }
+
+ dev->ifindex = ifm->ifi_index;
+
+ if (ops->newlink)
err = ops->newlink(net, dev, tb, data);
else
err = register_netdevice(dev);
--
1.7.6.5
^ permalink raw reply related
* [PATCH 4/6] veth: Allow to create peer link with given ifindex
From: Pavel Emelyanov @ 2012-08-06 14:14 UTC (permalink / raw)
To: David Miller, Eric Dumazet, Eric W. Biederman, Linux Netdev List
In-Reply-To: <501FD0F2.4040609@parallels.com>
The ifinfomsg is in there (thanks kaber@ for foreseeing this long time ago),
so take the given ifidex and register netdev with it.
Signed-off-by: Pavel Emelyanov <xemul@parallels.com>
---
drivers/net/veth.c | 3 +++
1 files changed, 3 insertions(+), 0 deletions(-)
diff --git a/drivers/net/veth.c b/drivers/net/veth.c
index 5852361..496c026 100644
--- a/drivers/net/veth.c
+++ b/drivers/net/veth.c
@@ -348,6 +348,9 @@ static int veth_newlink(struct net *src_net, struct net_device *dev,
if (tbp[IFLA_ADDRESS] == NULL)
eth_hw_addr_random(peer);
+ if (ifmp)
+ peer->ifindex = ifmp->ifi_index;
+
err = register_netdevice(peer);
put_net(net);
net = NULL;
--
1.7.6.5
^ permalink raw reply related
* [PATCH 5/6] net: Make ifindex generation per-net namespace
From: Pavel Emelyanov @ 2012-08-06 14:14 UTC (permalink / raw)
To: David Miller, Eric Dumazet, Eric W. Biederman, Linux Netdev List
In-Reply-To: <501FD0F2.4040609@parallels.com>
Strictly speaking this is only _really_ required for checkpoint-restore to
make loopback device always have the same index.
This change appears to be safe wrt "ifindex should be unique per-system"
concept, as all the ifindex usage is either already made per net namespace
of is explicitly limited with init_net only.
There are two cool side effects of this. The first one -- ifindices of
devices in container are always small, regardless of how many containers
we've started (and re-started) so far. The second one is -- we can speed
up the loopback ifidex access as shown in the next patch.
Signed-off-by: Pavel Emelyanov <xemul@parallels.com>
---
include/net/net_namespace.h | 1 +
net/core/dev.c | 4 ++--
2 files changed, 3 insertions(+), 2 deletions(-)
diff --git a/include/net/net_namespace.h b/include/net/net_namespace.h
index ae1cd6c..c5fbebf 100644
--- a/include/net/net_namespace.h
+++ b/include/net/net_namespace.h
@@ -62,6 +62,7 @@ struct net {
struct sock *rtnl; /* rtnetlink socket */
struct sock *genl_sock;
+ int ifindex;
struct list_head dev_base_head;
struct hlist_head *dev_name_head;
struct hlist_head *dev_index_head;
diff --git a/net/core/dev.c b/net/core/dev.c
index 3ca300d..1f06df8 100644
--- a/net/core/dev.c
+++ b/net/core/dev.c
@@ -5221,12 +5221,12 @@ int dev_ioctl(struct net *net, unsigned int cmd, void __user *arg)
*/
static int dev_new_index(struct net *net)
{
- static int ifindex;
+ int ifindex = net->ifindex;
for (;;) {
if (++ifindex <= 0)
ifindex = 1;
if (!__dev_get_by_index(net, ifindex))
- return ifindex;
+ return net->ifindex = ifindex;
}
}
--
1.7.6.5
^ permalink raw reply related
* [PATCH 6/6] net: Loopback ifindex is constant now
From: Pavel Emelyanov @ 2012-08-06 14:15 UTC (permalink / raw)
To: David Miller, Eric Dumazet, Eric W. Biederman, Linux Netdev List
In-Reply-To: <501FD0F2.4040609@parallels.com>
As pointed out, there are places, that access net->loopback_dev->ifindex
and after ifindex generation is made per-net this value becomes constant
equals 1. So go ahead and introduce the LOOPBACK_IFINDEX constant and use
it where appropriate.
Signed-off-by: Pavel Emelyanov <xemul@parallels.com>
---
drivers/net/loopback.c | 1 +
include/net/net_namespace.h | 7 +++++++
net/decnet/dn_route.c | 6 +++---
net/ipv4/fib_frontend.c | 2 +-
net/ipv4/ipmr.c | 2 +-
net/ipv4/netfilter/ipt_rpfilter.c | 2 +-
net/ipv4/route.c | 6 +++---
net/ipv6/route.c | 2 +-
8 files changed, 18 insertions(+), 10 deletions(-)
diff --git a/drivers/net/loopback.c b/drivers/net/loopback.c
index e2a06fd..4a075ba 100644
--- a/drivers/net/loopback.c
+++ b/drivers/net/loopback.c
@@ -197,6 +197,7 @@ static __net_init int loopback_net_init(struct net *net)
if (err)
goto out_free_netdev;
+ BUG_ON(dev->ifindex != LOOPBACK_IFINDEX);
net->loopback_dev = dev;
return 0;
diff --git a/include/net/net_namespace.h b/include/net/net_namespace.h
index c5fbebf..29a2370 100644
--- a/include/net/net_namespace.h
+++ b/include/net/net_namespace.h
@@ -105,6 +105,13 @@ struct net {
struct sock *diag_nlsk;
};
+/*
+ * ifindex generation is per-net namespace, and loopback is
+ * always the 1st device in ns (see net_dev_init), thus any
+ * loopback device should get ifindex 1
+ */
+
+#define LOOPBACK_IFINDEX 1
#include <linux/seq_file_net.h>
diff --git a/net/decnet/dn_route.c b/net/decnet/dn_route.c
index 85a3604..c855e8d 100644
--- a/net/decnet/dn_route.c
+++ b/net/decnet/dn_route.c
@@ -961,7 +961,7 @@ static int dn_route_output_slow(struct dst_entry **pprt, const struct flowidn *o
.saddr = oldflp->saddr,
.flowidn_scope = RT_SCOPE_UNIVERSE,
.flowidn_mark = oldflp->flowidn_mark,
- .flowidn_iif = init_net.loopback_dev->ifindex,
+ .flowidn_iif = LOOPBACK_IFINDEX,
.flowidn_oif = oldflp->flowidn_oif,
};
struct dn_route *rt = NULL;
@@ -979,7 +979,7 @@ static int dn_route_output_slow(struct dst_entry **pprt, const struct flowidn *o
"dn_route_output_slow: dst=%04x src=%04x mark=%d"
" iif=%d oif=%d\n", le16_to_cpu(oldflp->daddr),
le16_to_cpu(oldflp->saddr),
- oldflp->flowidn_mark, init_net.loopback_dev->ifindex,
+ oldflp->flowidn_mark, LOOPBACK_IFINDEX,
oldflp->flowidn_oif);
/* If we have an output interface, verify its a DECnet device */
@@ -1042,7 +1042,7 @@ source_ok:
if (!fld.daddr)
goto out;
}
- fld.flowidn_oif = init_net.loopback_dev->ifindex;
+ fld.flowidn_oif = LOOPBACK_IFINDEX;
res.type = RTN_LOCAL;
goto make_route;
}
diff --git a/net/ipv4/fib_frontend.c b/net/ipv4/fib_frontend.c
index c43ae3f..7f073a3 100644
--- a/net/ipv4/fib_frontend.c
+++ b/net/ipv4/fib_frontend.c
@@ -218,7 +218,7 @@ __be32 fib_compute_spec_dst(struct sk_buff *skb)
scope = RT_SCOPE_UNIVERSE;
if (!ipv4_is_zeronet(ip_hdr(skb)->saddr)) {
fl4.flowi4_oif = 0;
- fl4.flowi4_iif = net->loopback_dev->ifindex;
+ fl4.flowi4_iif = LOOPBACK_IFINDEX;
fl4.daddr = ip_hdr(skb)->saddr;
fl4.saddr = 0;
fl4.flowi4_tos = RT_TOS(ip_hdr(skb)->tos);
diff --git a/net/ipv4/ipmr.c b/net/ipv4/ipmr.c
index 8eec8f4..3a57570 100644
--- a/net/ipv4/ipmr.c
+++ b/net/ipv4/ipmr.c
@@ -1798,7 +1798,7 @@ static struct mr_table *ipmr_rt_fib_lookup(struct net *net, struct sk_buff *skb)
.flowi4_oif = (rt_is_output_route(rt) ?
skb->dev->ifindex : 0),
.flowi4_iif = (rt_is_output_route(rt) ?
- net->loopback_dev->ifindex :
+ LOOPBACK_IFINDEX :
skb->dev->ifindex),
.flowi4_mark = skb->mark,
};
diff --git a/net/ipv4/netfilter/ipt_rpfilter.c b/net/ipv4/netfilter/ipt_rpfilter.c
index 31371be..c301300 100644
--- a/net/ipv4/netfilter/ipt_rpfilter.c
+++ b/net/ipv4/netfilter/ipt_rpfilter.c
@@ -85,7 +85,7 @@ static bool rpfilter_mt(const struct sk_buff *skb, struct xt_action_param *par)
return ipv4_is_local_multicast(iph->daddr) ^ invert;
flow.flowi4_iif = 0;
} else {
- flow.flowi4_iif = dev_net(par->in)->loopback_dev->ifindex;
+ flow.flowi4_iif = LOOPBACK_IFINDEX;
}
flow.daddr = iph->saddr;
diff --git a/net/ipv4/route.c b/net/ipv4/route.c
index 21ad369..c581373 100644
--- a/net/ipv4/route.c
+++ b/net/ipv4/route.c
@@ -1619,7 +1619,7 @@ static int ip_route_input_slow(struct sk_buff *skb, __be32 daddr, __be32 saddr,
if (res.type == RTN_LOCAL) {
err = fib_validate_source(skb, saddr, daddr, tos,
- net->loopback_dev->ifindex,
+ LOOPBACK_IFINDEX,
dev, in_dev, &itag);
if (err < 0)
goto martian_source_keep_err;
@@ -1895,7 +1895,7 @@ struct rtable *__ip_route_output_key(struct net *net, struct flowi4 *fl4)
orig_oif = fl4->flowi4_oif;
- fl4->flowi4_iif = net->loopback_dev->ifindex;
+ fl4->flowi4_iif = LOOPBACK_IFINDEX;
fl4->flowi4_tos = tos & IPTOS_RT_MASK;
fl4->flowi4_scope = ((tos & RTO_ONLINK) ?
RT_SCOPE_LINK : RT_SCOPE_UNIVERSE);
@@ -1984,7 +1984,7 @@ struct rtable *__ip_route_output_key(struct net *net, struct flowi4 *fl4)
if (!fl4->daddr)
fl4->daddr = fl4->saddr = htonl(INADDR_LOOPBACK);
dev_out = net->loopback_dev;
- fl4->flowi4_oif = net->loopback_dev->ifindex;
+ fl4->flowi4_oif = LOOPBACK_IFINDEX;
res.type = RTN_LOCAL;
flags |= RTCF_LOCAL;
goto make_route;
diff --git a/net/ipv6/route.c b/net/ipv6/route.c
index 8e80fd2..0ddf2d1 100644
--- a/net/ipv6/route.c
+++ b/net/ipv6/route.c
@@ -965,7 +965,7 @@ struct dst_entry * ip6_route_output(struct net *net, const struct sock *sk,
{
int flags = 0;
- fl6->flowi6_iif = net->loopback_dev->ifindex;
+ fl6->flowi6_iif = LOOPBACK_IFINDEX;
if ((sk && sk->sk_bound_dev_if) || rt6_need_strict(&fl6->daddr))
flags |= RT6_LOOKUP_F_IFACE;
--
1.7.6.5
^ permalink raw reply related
* RE: pch_gbe: dont-copy-payload (was [PATCH 1/4] ...)
From: Andy Cress @ 2012-08-06 14:19 UTC (permalink / raw)
To: Eric Dumazet; +Cc: netdev
In-Reply-To: <40680C535D6FE6498883F1640FACD44D011BBE1D@ka-exchange-1.kontronamerica.local>
I found out why the proposed dont-copy-payload patch didn't work with this pch_gbe NIC.
This NIC PHY requires 64-byte-aligned DMA, and the transmit buffers won't be transferred if skb->data is not 64-byte-aligned. Apparently the data copy has a by-product of aligning the buffers.
I tried using skb_reserve(skb,64) in pch_gbe_alloc_tx_buffers and pch-gbe_alloc_rx_buffers, but that didn't seem to resolve it.
How can I make sure that the transmit data buffers are 64-byte-aligned?
Andy
-----Original Message-----
From: Andy Cress
Sent: Wednesday, July 25, 2012 4:11 PM
Subject: RE: [PATCH 1/4] pch_gbe: Fix the checksum fill to the error location
[...]
I tried applying the dont-copy-payload patch you outlined below, and for some reason it won't send successfully with that change. I'm not sure why though. This approach seems sound, and should greatly help transmit performance, if I could figure out what else it needs.
Andy
-----Original Message-----
From: Eric Dumazet
Sent: Tuesday, July 17, 2012 3:33 AM
To: Andy Cress
Cc: netdev@vger.kernel.org; Zhong Hongbo
Subject: Re: [PATCH 1/4] pch_gbe: Fix the checksum fill to the error location
On Tue, 2012-07-17 at 09:09 +0200, Eric Dumazet wrote:
[...]
> There _must_ be a way to avoid most of these copies (ie not touching
> payload), only mess with the header to insert these 2 nul bytes ?
>
> /* [Header:14][payload] ---> [Header:14][paddong:2][payload] */
[...]
Something like the following (untested) patch
drivers/net/ethernet/oki-semi/pch_gbe/pch_gbe_main.c | 55 +++++-----
1 file changed, 29 insertions(+), 26 deletions(-)
diff --git a/drivers/net/ethernet/oki-semi/pch_gbe/pch_gbe_main.c b/drivers/net/ethernet/oki-semi/pch_gbe/pch_gbe_main.c
index b100656..2d3d982 100644
--- a/drivers/net/ethernet/oki-semi/pch_gbe/pch_gbe_main.c
+++ b/drivers/net/ethernet/oki-semi/pch_gbe/pch_gbe_main.c
@@ -1163,7 +1163,7 @@ static void pch_gbe_tx_queue(struct pch_gbe_adapter *adapter,
struct pch_gbe_hw *hw = &adapter->hw;
struct pch_gbe_tx_desc *tx_desc;
struct pch_gbe_buffer *buffer_info;
- struct sk_buff *tmp_skb;
+ char *ptr;
unsigned int frame_ctrl;
unsigned int ring_num;
@@ -1221,18 +1221,27 @@ static void pch_gbe_tx_queue(struct pch_gbe_adapter *adapter,
buffer_info = &tx_ring->buffer_info[ring_num];
- tmp_skb = buffer_info->skb;
+ if (skb_headroom(skb) < 2) {
+ struct sk_buff *skb_new;
+
+ skb_new = skb_realloc_headroom(skb, 2);
+ if (!skb_new) {
+ tx_ring->next_to_use = ring_num;
+ dev_kfree_skb_any(skb);
+ return;
+ }
+ consume_skb(skb);
+ skb = skb_new;
+ }
/* [Header:14][payload] ---> [Header:14][paddong:2][payload] */
- memcpy(tmp_skb->data, skb->data, ETH_HLEN);
- tmp_skb->data[ETH_HLEN] = 0x00;
- tmp_skb->data[ETH_HLEN + 1] = 0x00;
- tmp_skb->len = skb->len;
- memcpy(&tmp_skb->data[ETH_HLEN + 2], &skb->data[ETH_HLEN],
- (skb->len - ETH_HLEN));
+ ptr = skb_push(skb, 2);
+ memmove(ptr, ptr + 2, ETH_HLEN);
+ ptr[ETH_HLEN] = 0x00;
+ ptr[ETH_HLEN + 1] = 0x00;
/*-- Set Buffer information --*/
- buffer_info->length = tmp_skb->len;
- buffer_info->dma = dma_map_single(&adapter->pdev->dev, tmp_skb->data,
+ buffer_info->length = skb->len;
+ buffer_info->dma = dma_map_single(&adapter->pdev->dev, skb->data,
buffer_info->length,
DMA_TO_DEVICE);
if (dma_mapping_error(&adapter->pdev->dev, buffer_info->dma)) {
@@ -1240,18 +1249,20 @@ static void pch_gbe_tx_queue(struct pch_gbe_adapter *adapter,
buffer_info->dma = 0;
buffer_info->time_stamp = 0;
tx_ring->next_to_use = ring_num;
+ dev_kfree_skb_any(skb);
return;
}
buffer_info->mapped = true;
buffer_info->time_stamp = jiffies;
+ buffer_info->skb = skb;
/*-- Set Tx descriptor --*/
tx_desc = PCH_GBE_TX_DESC(*tx_ring, ring_num);
- tx_desc->buffer_addr = (buffer_info->dma);
- tx_desc->length = (tmp_skb->len);
- tx_desc->tx_words_eob = ((tmp_skb->len + 3));
+ tx_desc->buffer_addr = buffer_info->dma;
+ tx_desc->length = skb->len;
+ tx_desc->tx_words_eob = skb->len + 3;
tx_desc->tx_frame_ctrl = (frame_ctrl);
- tx_desc->gbec_status = (DSC_INIT16);
+ tx_desc->gbec_status = DSC_INIT16;
if (unlikely(++ring_num == tx_ring->count))
ring_num = 0;
@@ -1265,7 +1276,6 @@ static void pch_gbe_tx_queue(struct pch_gbe_adapter *adapter,
pch_tx_timestamp(adapter, skb);
#endif
- dev_kfree_skb_any(skb);
}
/**
@@ -1543,19 +1553,12 @@ static void pch_gbe_alloc_tx_buffers(struct pch_gbe_adapter *adapter,
struct pch_gbe_tx_ring *tx_ring)
{
struct pch_gbe_buffer *buffer_info;
- struct sk_buff *skb;
unsigned int i;
- unsigned int bufsz;
struct pch_gbe_tx_desc *tx_desc;
- bufsz =
- adapter->hw.mac.max_frame_size + PCH_GBE_DMA_ALIGN + NET_IP_ALIGN;
-
for (i = 0; i < tx_ring->count; i++) {
buffer_info = &tx_ring->buffer_info[i];
- skb = netdev_alloc_skb(adapter->netdev, bufsz);
- skb_reserve(skb, PCH_GBE_DMA_ALIGN);
- buffer_info->skb = skb;
+ buffer_info->skb = NULL;
tx_desc = PCH_GBE_TX_DESC(*tx_ring, i);
tx_desc->gbec_status = (DSC_INIT16);
}
@@ -1622,9 +1625,9 @@ pch_gbe_clean_tx(struct pch_gbe_adapter *adapter,
buffer_info->length, DMA_TO_DEVICE);
buffer_info->mapped = false;
}
- if (buffer_info->skb) {
- pr_debug("trim buffer_info->skb : %d\n", i);
- skb_trim(buffer_info->skb, 0);
+ if (skb) {
+ dev_kfree_skb_any(skb);
+ buffer_info->skb = NULL;
}
tx_desc->gbec_status = DSC_INIT16;
if (unlikely(++i == tx_ring->count))
^ permalink raw reply related
* [PATCH v2 0/8] some netpoll and netconsole fixes
From: Cong Wang @ 2012-08-06 14:23 UTC (permalink / raw)
To: netdev; +Cc: Cong Wang, David S. Miller
V2:
* update patch 1/8 as Eric suggested
* drop the bridge patch, add comments instead
* add patch 8/8
This patch fixes serval problems in netconsole and netpoll.
I ran this patch in my KVM guest with some netpoll test cases,
even covered with some corner cases, everything worked as expected.
BTW, my kernel config enables the following validatation options:
CONFIG_DEBUG_ATOMIC_SLEEP=y
CONFIG_PROVE_LOCKING=y
CONFIG_PROVE_RCU=y
CONFIG_LOCKDEP=y
CONFIG_RCU_CPU_STALL_TIMEOUT=60
CONFIG_LOCKUP_DETECTOR=y
CONFIG_HARDLOCKUP_DETECTOR=y
Cc: "David S. Miller" <davem@davemloft.net>
Signed-off-by: Cong Wang <amwang@redhat.com>
---
Cong Wang (8):
netpoll: use GFP_ATOMIC in slave_enable_netpoll() and __netpoll_setup()
netpoll: make __netpoll_cleanup non-block
netconsole: do not release spin_lock when calling __netpoll_cleanup
netpoll: take rcu_read_lock_bh() in netpoll_rx()
netpoll: use netpoll_rx_on() in netpoll_rx()
netpoll: take rcu_read_lock_bh() in netpoll_send_skb_on_dev()
bridge: add some comments for NETDEV_RELEASE
bridge: use list_for_each_entry() in netpoll functions
drivers/net/bonding/bond_main.c | 10 ++---
drivers/net/netconsole.c | 5 --
drivers/net/team/team.c | 16 ++++---
include/linux/netdevice.h | 3 +-
include/linux/netpoll.h | 27 +++++++------
net/8021q/vlan_dev.c | 13 ++----
net/bridge/br_device.c | 27 +++++-------
net/bridge/br_if.c | 6 ++-
net/bridge/br_private.h | 4 +-
net/core/netpoll.c | 84 +++++++++++++++++++++++++--------------
10 files changed, 107 insertions(+), 88 deletions(-)
--
1.7.7.6
^ permalink raw reply
* [PATCH 1/8] netpoll: use GFP_ATOMIC in slave_enable_netpoll() and __netpoll_setup()
From: Cong Wang @ 2012-08-06 14:23 UTC (permalink / raw)
To: netdev; +Cc: Cong Wang, Eric Dumazet, David S. Miller
In-Reply-To: <1344263012-4031-1-git-send-email-amwang@redhat.com>
slave_enable_netpoll() and __netpoll_setup() may be called
with read_lock() held, so should use GFP_ATOMIC to allocate
memory. Eric suggested to pass gfp flags to __netpoll_setup().
Cc: Eric Dumazet <eric.dumazet@gmail.com>
Cc: "David S. Miller" <davem@davemloft.net>
Reported-by: Dan Carpenter <dan.carpenter@oracle.com>
Signed-off-by: Cong Wang <amwang@redhat.com>
---
drivers/net/bonding/bond_main.c | 6 +++---
drivers/net/team/team.c | 16 +++++++++-------
include/linux/netdevice.h | 3 ++-
include/linux/netpoll.h | 2 +-
net/8021q/vlan_dev.c | 7 ++++---
net/bridge/br_device.c | 12 ++++++------
net/bridge/br_if.c | 2 +-
net/bridge/br_private.h | 4 ++--
net/core/netpoll.c | 8 ++++----
9 files changed, 32 insertions(+), 28 deletions(-)
diff --git a/drivers/net/bonding/bond_main.c b/drivers/net/bonding/bond_main.c
index 6fae5f3..8697136 100644
--- a/drivers/net/bonding/bond_main.c
+++ b/drivers/net/bonding/bond_main.c
@@ -1235,12 +1235,12 @@ static inline int slave_enable_netpoll(struct slave *slave)
struct netpoll *np;
int err = 0;
- np = kzalloc(sizeof(*np), GFP_KERNEL);
+ np = kzalloc(sizeof(*np), GFP_ATOMIC);
err = -ENOMEM;
if (!np)
goto out;
- err = __netpoll_setup(np, slave->dev);
+ err = __netpoll_setup(np, slave->dev, GFP_ATOMIC);
if (err) {
kfree(np);
goto out;
@@ -1292,7 +1292,7 @@ static void bond_netpoll_cleanup(struct net_device *bond_dev)
read_unlock(&bond->lock);
}
-static int bond_netpoll_setup(struct net_device *dev, struct netpoll_info *ni)
+static int bond_netpoll_setup(struct net_device *dev, struct netpoll_info *ni, gfp_t gfp)
{
struct bonding *bond = netdev_priv(dev);
struct slave *slave;
diff --git a/drivers/net/team/team.c b/drivers/net/team/team.c
index ba10c46..bbb8524 100644
--- a/drivers/net/team/team.c
+++ b/drivers/net/team/team.c
@@ -913,16 +913,17 @@ static void team_port_leave(struct team *team, struct team_port *port)
}
#ifdef CONFIG_NET_POLL_CONTROLLER
-static int team_port_enable_netpoll(struct team *team, struct team_port *port)
+static int team_port_enable_netpoll(struct team *team, struct team_port *port,
+ gfp_t gfp)
{
struct netpoll *np;
int err;
- np = kzalloc(sizeof(*np), GFP_KERNEL);
+ np = kzalloc(sizeof(*np), gfp);
if (!np)
return -ENOMEM;
- err = __netpoll_setup(np, port->dev);
+ err = __netpoll_setup(np, port->dev, gfp);
if (err) {
kfree(np);
return err;
@@ -951,7 +952,8 @@ static struct netpoll_info *team_netpoll_info(struct team *team)
}
#else
-static int team_port_enable_netpoll(struct team *team, struct team_port *port)
+static int team_port_enable_netpoll(struct team *team, struct team_port *port,
+ gfp_t gfp)
{
return 0;
}
@@ -1032,7 +1034,7 @@ static int team_port_add(struct team *team, struct net_device *port_dev)
}
if (team_netpoll_info(team)) {
- err = team_port_enable_netpoll(team, port);
+ err = team_port_enable_netpoll(team, port, GFP_KERNEL);
if (err) {
netdev_err(dev, "Failed to enable netpoll on device %s\n",
portname);
@@ -1627,7 +1629,7 @@ static void team_netpoll_cleanup(struct net_device *dev)
}
static int team_netpoll_setup(struct net_device *dev,
- struct netpoll_info *npifo)
+ struct netpoll_info *npifo, gfp_t gfp)
{
struct team *team = netdev_priv(dev);
struct team_port *port;
@@ -1635,7 +1637,7 @@ static int team_netpoll_setup(struct net_device *dev,
mutex_lock(&team->lock);
list_for_each_entry(port, &team->port_list, list) {
- err = team_port_enable_netpoll(team, port);
+ err = team_port_enable_netpoll(team, port, gfp);
if (err) {
__team_netpoll_cleanup(team);
break;
diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h
index a9db4f3..3560d68 100644
--- a/include/linux/netdevice.h
+++ b/include/linux/netdevice.h
@@ -953,7 +953,8 @@ struct net_device_ops {
#ifdef CONFIG_NET_POLL_CONTROLLER
void (*ndo_poll_controller)(struct net_device *dev);
int (*ndo_netpoll_setup)(struct net_device *dev,
- struct netpoll_info *info);
+ struct netpoll_info *info,
+ gfp_t gfp);
void (*ndo_netpoll_cleanup)(struct net_device *dev);
#endif
int (*ndo_set_vf_mac)(struct net_device *dev,
diff --git a/include/linux/netpoll.h b/include/linux/netpoll.h
index 28f5389..bf2d51e 100644
--- a/include/linux/netpoll.h
+++ b/include/linux/netpoll.h
@@ -43,7 +43,7 @@ struct netpoll_info {
void netpoll_send_udp(struct netpoll *np, const char *msg, int len);
void netpoll_print_options(struct netpoll *np);
int netpoll_parse_options(struct netpoll *np, char *opt);
-int __netpoll_setup(struct netpoll *np, struct net_device *ndev);
+int __netpoll_setup(struct netpoll *np, struct net_device *ndev, gfp_t gfp);
int netpoll_setup(struct netpoll *np);
int netpoll_trap(void);
void netpoll_set_trap(int trap);
diff --git a/net/8021q/vlan_dev.c b/net/8021q/vlan_dev.c
index 73a2a83..ee4ae09 100644
--- a/net/8021q/vlan_dev.c
+++ b/net/8021q/vlan_dev.c
@@ -669,19 +669,20 @@ static void vlan_dev_poll_controller(struct net_device *dev)
return;
}
-static int vlan_dev_netpoll_setup(struct net_device *dev, struct netpoll_info *npinfo)
+static int vlan_dev_netpoll_setup(struct net_device *dev, struct netpoll_info *npinfo,
+ gfp_t gfp)
{
struct vlan_dev_priv *info = vlan_dev_priv(dev);
struct net_device *real_dev = info->real_dev;
struct netpoll *netpoll;
int err = 0;
- netpoll = kzalloc(sizeof(*netpoll), GFP_KERNEL);
+ netpoll = kzalloc(sizeof(*netpoll), gfp);
err = -ENOMEM;
if (!netpoll)
goto out;
- err = __netpoll_setup(netpoll, real_dev);
+ err = __netpoll_setup(netpoll, real_dev, gfp);
if (err) {
kfree(netpoll);
goto out;
diff --git a/net/bridge/br_device.c b/net/bridge/br_device.c
index 3334845..ed0e0f9 100644
--- a/net/bridge/br_device.c
+++ b/net/bridge/br_device.c
@@ -213,7 +213,8 @@ static void br_netpoll_cleanup(struct net_device *dev)
}
}
-static int br_netpoll_setup(struct net_device *dev, struct netpoll_info *ni)
+static int br_netpoll_setup(struct net_device *dev, struct netpoll_info *ni,
+ gfp_t gfp)
{
struct net_bridge *br = netdev_priv(dev);
struct net_bridge_port *p, *n;
@@ -222,8 +223,7 @@ static int br_netpoll_setup(struct net_device *dev, struct netpoll_info *ni)
list_for_each_entry_safe(p, n, &br->port_list, list) {
if (!p->dev)
continue;
-
- err = br_netpoll_enable(p);
+ err = br_netpoll_enable(p, gfp);
if (err)
goto fail;
}
@@ -236,17 +236,17 @@ fail:
goto out;
}
-int br_netpoll_enable(struct net_bridge_port *p)
+int br_netpoll_enable(struct net_bridge_port *p, gfp_t gfp)
{
struct netpoll *np;
int err = 0;
- np = kzalloc(sizeof(*p->np), GFP_KERNEL);
+ np = kzalloc(sizeof(*p->np), gfp);
err = -ENOMEM;
if (!np)
goto out;
- err = __netpoll_setup(np, p->dev);
+ err = __netpoll_setup(np, p->dev, gfp);
if (err) {
kfree(np);
goto out;
diff --git a/net/bridge/br_if.c b/net/bridge/br_if.c
index e1144e1..171fd6b 100644
--- a/net/bridge/br_if.c
+++ b/net/bridge/br_if.c
@@ -361,7 +361,7 @@ int br_add_if(struct net_bridge *br, struct net_device *dev)
if (err)
goto err2;
- if (br_netpoll_info(br) && ((err = br_netpoll_enable(p))))
+ if (br_netpoll_info(br) && ((err = br_netpoll_enable(p, GFP_KERNEL))))
goto err3;
err = netdev_set_master(dev, br->dev);
diff --git a/net/bridge/br_private.h b/net/bridge/br_private.h
index a768b24..f507d2a 100644
--- a/net/bridge/br_private.h
+++ b/net/bridge/br_private.h
@@ -316,7 +316,7 @@ static inline void br_netpoll_send_skb(const struct net_bridge_port *p,
netpoll_send_skb(np, skb);
}
-extern int br_netpoll_enable(struct net_bridge_port *p);
+extern int br_netpoll_enable(struct net_bridge_port *p, gfp_t gfp);
extern void br_netpoll_disable(struct net_bridge_port *p);
#else
static inline struct netpoll_info *br_netpoll_info(struct net_bridge *br)
@@ -329,7 +329,7 @@ static inline void br_netpoll_send_skb(const struct net_bridge_port *p,
{
}
-static inline int br_netpoll_enable(struct net_bridge_port *p)
+static inline int br_netpoll_enable(struct net_bridge_port *p, gfp_t gfp)
{
return 0;
}
diff --git a/net/core/netpoll.c b/net/core/netpoll.c
index b4c90e4..37cc854 100644
--- a/net/core/netpoll.c
+++ b/net/core/netpoll.c
@@ -715,7 +715,7 @@ int netpoll_parse_options(struct netpoll *np, char *opt)
}
EXPORT_SYMBOL(netpoll_parse_options);
-int __netpoll_setup(struct netpoll *np, struct net_device *ndev)
+int __netpoll_setup(struct netpoll *np, struct net_device *ndev, gfp_t gfp)
{
struct netpoll_info *npinfo;
const struct net_device_ops *ops;
@@ -734,7 +734,7 @@ int __netpoll_setup(struct netpoll *np, struct net_device *ndev)
}
if (!ndev->npinfo) {
- npinfo = kmalloc(sizeof(*npinfo), GFP_KERNEL);
+ npinfo = kmalloc(sizeof(*npinfo), gfp);
if (!npinfo) {
err = -ENOMEM;
goto out;
@@ -752,7 +752,7 @@ int __netpoll_setup(struct netpoll *np, struct net_device *ndev)
ops = np->dev->netdev_ops;
if (ops->ndo_netpoll_setup) {
- err = ops->ndo_netpoll_setup(ndev, npinfo);
+ err = ops->ndo_netpoll_setup(ndev, npinfo, gfp);
if (err)
goto free_npinfo;
}
@@ -857,7 +857,7 @@ int netpoll_setup(struct netpoll *np)
refill_skbs();
rtnl_lock();
- err = __netpoll_setup(np, ndev);
+ err = __netpoll_setup(np, ndev, GFP_KERNEL);
rtnl_unlock();
if (err)
--
1.7.7.6
^ permalink raw reply related
* [PATCH 2/8] netpoll: make __netpoll_cleanup non-block
From: Cong Wang @ 2012-08-06 14:23 UTC (permalink / raw)
To: netdev; +Cc: Cong Wang, David S. Miller
In-Reply-To: <1344263012-4031-1-git-send-email-amwang@redhat.com>
Like the previous patch, slave_disable_netpoll() and __netpoll_cleanup()
may be called with read_lock() held too, so we should make them
non-block, by moving the cleanup and kfree() to call_rcu_bh() callbacks.
Cc: "David S. Miller" <davem@davemloft.net>
Signed-off-by: Cong Wang <amwang@redhat.com>
---
drivers/net/bonding/bond_main.c | 4 +--
include/linux/netpoll.h | 3 ++
net/8021q/vlan_dev.c | 6 +----
net/bridge/br_device.c | 6 +----
net/core/netpoll.c | 42 +++++++++++++++++++++++++++++---------
5 files changed, 38 insertions(+), 23 deletions(-)
diff --git a/drivers/net/bonding/bond_main.c b/drivers/net/bonding/bond_main.c
index 8697136..e428916 100644
--- a/drivers/net/bonding/bond_main.c
+++ b/drivers/net/bonding/bond_main.c
@@ -1257,9 +1257,7 @@ static inline void slave_disable_netpoll(struct slave *slave)
return;
slave->np = NULL;
- synchronize_rcu_bh();
- __netpoll_cleanup(np);
- kfree(np);
+ __netpoll_free_rcu(np);
}
static inline bool slave_dev_support_netpoll(struct net_device *slave_dev)
{
diff --git a/include/linux/netpoll.h b/include/linux/netpoll.h
index bf2d51e..907812e 100644
--- a/include/linux/netpoll.h
+++ b/include/linux/netpoll.h
@@ -23,6 +23,7 @@ struct netpoll {
u8 remote_mac[ETH_ALEN];
struct list_head rx; /* rx_np list element */
+ struct rcu_head rcu;
};
struct netpoll_info {
@@ -38,6 +39,7 @@ struct netpoll_info {
struct delayed_work tx_work;
struct netpoll *netpoll;
+ struct rcu_head rcu;
};
void netpoll_send_udp(struct netpoll *np, const char *msg, int len);
@@ -48,6 +50,7 @@ int netpoll_setup(struct netpoll *np);
int netpoll_trap(void);
void netpoll_set_trap(int trap);
void __netpoll_cleanup(struct netpoll *np);
+void __netpoll_free_rcu(struct netpoll *np);
void netpoll_cleanup(struct netpoll *np);
int __netpoll_rx(struct sk_buff *skb);
void netpoll_send_skb_on_dev(struct netpoll *np, struct sk_buff *skb,
diff --git a/net/8021q/vlan_dev.c b/net/8021q/vlan_dev.c
index ee4ae09..b65623f 100644
--- a/net/8021q/vlan_dev.c
+++ b/net/8021q/vlan_dev.c
@@ -704,11 +704,7 @@ static void vlan_dev_netpoll_cleanup(struct net_device *dev)
info->netpoll = NULL;
- /* Wait for transmitting packets to finish before freeing. */
- synchronize_rcu_bh();
-
- __netpoll_cleanup(netpoll);
- kfree(netpoll);
+ __netpoll_free_rcu(netpoll);
}
#endif /* CONFIG_NET_POLL_CONTROLLER */
diff --git a/net/bridge/br_device.c b/net/bridge/br_device.c
index ed0e0f9..f41ba40 100644
--- a/net/bridge/br_device.c
+++ b/net/bridge/br_device.c
@@ -267,11 +267,7 @@ void br_netpoll_disable(struct net_bridge_port *p)
p->np = NULL;
- /* Wait for transmitting packets to finish before freeing. */
- synchronize_rcu_bh();
-
- __netpoll_cleanup(np);
- kfree(np);
+ __netpoll_free_rcu(np);
}
#endif
diff --git a/net/core/netpoll.c b/net/core/netpoll.c
index 37cc854..dc17f1d 100644
--- a/net/core/netpoll.c
+++ b/net/core/netpoll.c
@@ -878,6 +878,24 @@ static int __init netpoll_init(void)
}
core_initcall(netpoll_init);
+static void rcu_cleanup_netpoll_info(struct rcu_head *rcu_head)
+{
+ struct netpoll_info *npinfo =
+ container_of(rcu_head, struct netpoll_info, rcu);
+
+ skb_queue_purge(&npinfo->arp_tx);
+ skb_queue_purge(&npinfo->txq);
+
+ /* we can't call cancel_delayed_work_sync here, as we are in softirq */
+ cancel_delayed_work(&npinfo->tx_work);
+
+ /* clean after last, unfinished work */
+ __skb_queue_purge(&npinfo->txq);
+ /* now cancel it again */
+ cancel_delayed_work(&npinfo->tx_work);
+ kfree(npinfo);
+}
+
void __netpoll_cleanup(struct netpoll *np)
{
struct netpoll_info *npinfo;
@@ -903,20 +921,24 @@ void __netpoll_cleanup(struct netpoll *np)
ops->ndo_netpoll_cleanup(np->dev);
RCU_INIT_POINTER(np->dev->npinfo, NULL);
+ call_rcu_bh(&npinfo->rcu, rcu_cleanup_netpoll_info);
+ }
+}
+EXPORT_SYMBOL_GPL(__netpoll_cleanup);
- /* avoid racing with NAPI reading npinfo */
- synchronize_rcu_bh();
+static void rcu_cleanup_netpoll(struct rcu_head *rcu_head)
+{
+ struct netpoll *np = container_of(rcu_head, struct netpoll, rcu);
- skb_queue_purge(&npinfo->arp_tx);
- skb_queue_purge(&npinfo->txq);
- cancel_delayed_work_sync(&npinfo->tx_work);
+ __netpoll_cleanup(np);
+ kfree(np);
+}
- /* clean after last, unfinished work */
- __skb_queue_purge(&npinfo->txq);
- kfree(npinfo);
- }
+void __netpoll_free_rcu(struct netpoll *np)
+{
+ call_rcu_bh(&np->rcu, rcu_cleanup_netpoll);
}
-EXPORT_SYMBOL_GPL(__netpoll_cleanup);
+EXPORT_SYMBOL_GPL(__netpoll_free_rcu);
void netpoll_cleanup(struct netpoll *np)
{
--
1.7.7.6
^ permalink raw reply related
* [PATCH 3/8] netconsole: do not release spin_lock when calling __netpoll_cleanup
From: Cong Wang @ 2012-08-06 14:23 UTC (permalink / raw)
To: netdev; +Cc: Cong Wang, David S. Miller
In-Reply-To: <1344263012-4031-1-git-send-email-amwang@redhat.com>
With the previous patch applied, __netpoll_cleanup() is non-block now,
so we don't need to release the spin_lock before calling it.
Cc: "David S. Miller" <davem@davemloft.net>
Signed-off-by: Cong Wang <amwang@redhat.com>
---
drivers/net/netconsole.c | 5 -----
1 files changed, 0 insertions(+), 5 deletions(-)
diff --git a/drivers/net/netconsole.c b/drivers/net/netconsole.c
index f9347ea..f0ad56c 100644
--- a/drivers/net/netconsole.c
+++ b/drivers/net/netconsole.c
@@ -640,12 +640,7 @@ static int netconsole_netdev_event(struct notifier_block *this,
* rtnl_lock already held
*/
if (nt->np.dev) {
- spin_unlock_irqrestore(
- &target_list_lock,
- flags);
__netpoll_cleanup(&nt->np);
- spin_lock_irqsave(&target_list_lock,
- flags);
dev_put(nt->np.dev);
nt->np.dev = NULL;
netconsole_target_put(nt);
--
1.7.7.6
^ permalink raw reply related
* [PATCH 4/8] netpoll: take rcu_read_lock_bh() in netpoll_rx()
From: Cong Wang @ 2012-08-06 14:23 UTC (permalink / raw)
To: netdev; +Cc: Cong Wang, David S. Miller
In-Reply-To: <1344263012-4031-1-git-send-email-amwang@redhat.com>
In __netpoll_rx(), it dereferences ->npinfo without rcu_dereference_bh(),
this patch fixes it by using the 'npinfo' passed from netpoll_rx()
where it is already dereferenced with rcu_dereference_bh().
Cc: "David S. Miller" <davem@davemloft.net>
Signed-off-by: Cong Wang <amwang@redhat.com>
---
include/linux/netpoll.h | 4 ++--
net/core/netpoll.c | 3 +--
2 files changed, 3 insertions(+), 4 deletions(-)
diff --git a/include/linux/netpoll.h b/include/linux/netpoll.h
index 907812e..5d881c3 100644
--- a/include/linux/netpoll.h
+++ b/include/linux/netpoll.h
@@ -52,7 +52,7 @@ void netpoll_set_trap(int trap);
void __netpoll_cleanup(struct netpoll *np);
void __netpoll_free_rcu(struct netpoll *np);
void netpoll_cleanup(struct netpoll *np);
-int __netpoll_rx(struct sk_buff *skb);
+int __netpoll_rx(struct sk_buff *skb, struct netpoll_info *npinfo);
void netpoll_send_skb_on_dev(struct netpoll *np, struct sk_buff *skb,
struct net_device *dev);
static inline void netpoll_send_skb(struct netpoll *np, struct sk_buff *skb)
@@ -77,7 +77,7 @@ static inline bool netpoll_rx(struct sk_buff *skb)
spin_lock(&npinfo->rx_lock);
/* check rx_flags again with the lock held */
- if (npinfo->rx_flags && __netpoll_rx(skb))
+ if (npinfo->rx_flags && __netpoll_rx(skb, npinfo))
ret = true;
spin_unlock(&npinfo->rx_lock);
diff --git a/net/core/netpoll.c b/net/core/netpoll.c
index dc17f1d..d055bb0 100644
--- a/net/core/netpoll.c
+++ b/net/core/netpoll.c
@@ -543,13 +543,12 @@ static void arp_reply(struct sk_buff *skb)
spin_unlock_irqrestore(&npinfo->rx_lock, flags);
}
-int __netpoll_rx(struct sk_buff *skb)
+int __netpoll_rx(struct sk_buff *skb, struct netpoll_info *npinfo)
{
int proto, len, ulen;
int hits = 0;
const struct iphdr *iph;
struct udphdr *uh;
- struct netpoll_info *npinfo = skb->dev->npinfo;
struct netpoll *np, *tmp;
if (list_empty(&npinfo->rx_np))
--
1.7.7.6
^ permalink raw reply related
* [PATCH 7/8] bridge: add some comments for NETDEV_RELEASE
From: Cong Wang @ 2012-08-06 14:23 UTC (permalink / raw)
To: netdev; +Cc: Cong Wang, Stephen Hemminger
In-Reply-To: <1344263012-4031-1-git-send-email-amwang@redhat.com>
Add comments on why we don't notify NETDEV_RELEASE.
Cc: Stephen Hemminger <shemminger@vyatta.com>
Signed-off-by: Cong Wang <amwang@redhat.com>
---
net/bridge/br_if.c | 4 ++++
1 files changed, 4 insertions(+), 0 deletions(-)
diff --git a/net/bridge/br_if.c b/net/bridge/br_if.c
index 171fd6b..910f4aa 100644
--- a/net/bridge/br_if.c
+++ b/net/bridge/br_if.c
@@ -427,6 +427,10 @@ int br_del_if(struct net_bridge *br, struct net_device *dev)
if (!p || p->br != br)
return -EINVAL;
+ /*
+ * We don't notify NETDEV_RELEASE event, as this will
+ * stop netconsole on the bridge.
+ */
del_nbp(p);
spin_lock_bh(&br->lock);
--
1.7.7.6
^ permalink raw reply related
* [PATCH 6/8] netpoll: take rcu_read_lock_bh() in netpoll_send_skb_on_dev()
From: Cong Wang @ 2012-08-06 14:23 UTC (permalink / raw)
To: netdev; +Cc: Cong Wang, David S. Miller
In-Reply-To: <1344263012-4031-1-git-send-email-amwang@redhat.com>
This patch fixes several problems in the call path of
netpoll_send_skb_on_dev():
1. We already disable IRQ's before calling netpoll_send_skb_on_dev(),
so we don't need to disable IRQ's again.
2. All the callees of netpoll_send_skb_on_dev() should use
rcu_dereference_bh() to dereference ->npinfo.
3. Rename arp_reply() to netpoll_arp_reply(), the former is too generic.
Cc: "David S. Miller" <davem@davemloft.net>
Signed-off-by: Cong Wang <amwang@redhat.com>
---
net/core/netpoll.c | 31 +++++++++++++++++--------------
1 files changed, 17 insertions(+), 14 deletions(-)
diff --git a/net/core/netpoll.c b/net/core/netpoll.c
index d055bb0..174346a 100644
--- a/net/core/netpoll.c
+++ b/net/core/netpoll.c
@@ -54,7 +54,7 @@ static atomic_t trapped;
MAX_UDP_CHUNK)
static void zap_completion_queue(void);
-static void arp_reply(struct sk_buff *skb);
+static void netpoll_arp_reply(struct sk_buff *skb, struct netpoll_info *npinfo);
static unsigned int carrier_timeout = 4;
module_param(carrier_timeout, uint, 0644);
@@ -170,7 +170,8 @@ static void poll_napi(struct net_device *dev)
list_for_each_entry(napi, &dev->napi_list, dev_list) {
if (napi->poll_owner != smp_processor_id() &&
spin_trylock(&napi->poll_lock)) {
- budget = poll_one_napi(dev->npinfo, napi, budget);
+ budget = poll_one_napi(rcu_dereference_bh(dev->npinfo),
+ napi, budget);
spin_unlock(&napi->poll_lock);
if (!budget)
@@ -185,13 +186,14 @@ static void service_arp_queue(struct netpoll_info *npi)
struct sk_buff *skb;
while ((skb = skb_dequeue(&npi->arp_tx)))
- arp_reply(skb);
+ netpoll_arp_reply(skb, npi);
}
}
static void netpoll_poll_dev(struct net_device *dev)
{
const struct net_device_ops *ops;
+ struct netpoll_info *ni = rcu_dereference_bh(dev->npinfo);
if (!dev || !netif_running(dev))
return;
@@ -206,17 +208,18 @@ static void netpoll_poll_dev(struct net_device *dev)
poll_napi(dev);
if (dev->flags & IFF_SLAVE) {
- if (dev->npinfo) {
+ if (ni) {
struct net_device *bond_dev = dev->master;
struct sk_buff *skb;
- while ((skb = skb_dequeue(&dev->npinfo->arp_tx))) {
+ struct netpoll_info *bond_ni = rcu_dereference_bh(bond_dev->npinfo);
+ while ((skb = skb_dequeue(&ni->arp_tx))) {
skb->dev = bond_dev;
- skb_queue_tail(&bond_dev->npinfo->arp_tx, skb);
+ skb_queue_tail(&bond_ni->arp_tx, skb);
}
}
}
- service_arp_queue(dev->npinfo);
+ service_arp_queue(ni);
zap_completion_queue();
}
@@ -302,6 +305,7 @@ static int netpoll_owner_active(struct net_device *dev)
return 0;
}
+/* call with IRQ disabled */
void netpoll_send_skb_on_dev(struct netpoll *np, struct sk_buff *skb,
struct net_device *dev)
{
@@ -309,8 +313,11 @@ void netpoll_send_skb_on_dev(struct netpoll *np, struct sk_buff *skb,
unsigned long tries;
const struct net_device_ops *ops = dev->netdev_ops;
/* It is up to the caller to keep npinfo alive. */
- struct netpoll_info *npinfo = np->dev->npinfo;
+ struct netpoll_info *npinfo;
+
+ WARN_ON_ONCE(!irqs_disabled());
+ npinfo = rcu_dereference_bh(np->dev->npinfo);
if (!npinfo || !netif_running(dev) || !netif_device_present(dev)) {
__kfree_skb(skb);
return;
@@ -319,11 +326,9 @@ void netpoll_send_skb_on_dev(struct netpoll *np, struct sk_buff *skb,
/* don't get messages out of order, and no recursion */
if (skb_queue_len(&npinfo->txq) == 0 && !netpoll_owner_active(dev)) {
struct netdev_queue *txq;
- unsigned long flags;
txq = netdev_get_tx_queue(dev, skb_get_queue_mapping(skb));
- local_irq_save(flags);
/* try until next clock tick */
for (tries = jiffies_to_usecs(1)/USEC_PER_POLL;
tries > 0; --tries) {
@@ -347,10 +352,9 @@ void netpoll_send_skb_on_dev(struct netpoll *np, struct sk_buff *skb,
}
WARN_ONCE(!irqs_disabled(),
- "netpoll_send_skb(): %s enabled interrupts in poll (%pF)\n",
+ "netpoll_send_skb_on_dev(): %s enabled interrupts in poll (%pF)\n",
dev->name, ops->ndo_start_xmit);
- local_irq_restore(flags);
}
if (status != NETDEV_TX_OK) {
@@ -423,9 +427,8 @@ void netpoll_send_udp(struct netpoll *np, const char *msg, int len)
}
EXPORT_SYMBOL(netpoll_send_udp);
-static void arp_reply(struct sk_buff *skb)
+static void netpoll_arp_reply(struct sk_buff *skb, struct netpoll_info *npinfo)
{
- struct netpoll_info *npinfo = skb->dev->npinfo;
struct arphdr *arp;
unsigned char *arp_ptr;
int size, type = ARPOP_REPLY, ptype = ETH_P_ARP;
--
1.7.7.6
^ permalink raw reply related
* [PATCH 5/8] netpoll: use netpoll_rx_on() in netpoll_rx()
From: Cong Wang @ 2012-08-06 14:23 UTC (permalink / raw)
To: netdev; +Cc: Cong Wang, David S. Miller
In-Reply-To: <1344263012-4031-1-git-send-email-amwang@redhat.com>
The logic of the code is same, just call netpoll_rx_on().
Cc: "David S. Miller" <davem@davemloft.net>
Signed-off-by: Cong Wang <amwang@redhat.com>
---
include/linux/netpoll.h | 18 +++++++++---------
1 files changed, 9 insertions(+), 9 deletions(-)
diff --git a/include/linux/netpoll.h b/include/linux/netpoll.h
index 5d881c3..2d178ba 100644
--- a/include/linux/netpoll.h
+++ b/include/linux/netpoll.h
@@ -63,6 +63,13 @@ static inline void netpoll_send_skb(struct netpoll *np, struct sk_buff *skb)
#ifdef CONFIG_NETPOLL
+static inline int netpoll_rx_on(struct sk_buff *skb)
+{
+ struct netpoll_info *npinfo = rcu_dereference_bh(skb->dev->npinfo);
+
+ return npinfo && (!list_empty(&npinfo->rx_np) || npinfo->rx_flags);
+}
+
static inline bool netpoll_rx(struct sk_buff *skb)
{
struct netpoll_info *npinfo;
@@ -70,11 +77,11 @@ static inline bool netpoll_rx(struct sk_buff *skb)
bool ret = false;
local_irq_save(flags);
- npinfo = rcu_dereference_bh(skb->dev->npinfo);
- if (!npinfo || (list_empty(&npinfo->rx_np) && !npinfo->rx_flags))
+ if (!netpoll_rx_on(skb))
goto out;
+ npinfo = rcu_dereference_bh(skb->dev->npinfo);
spin_lock(&npinfo->rx_lock);
/* check rx_flags again with the lock held */
if (npinfo->rx_flags && __netpoll_rx(skb, npinfo))
@@ -86,13 +93,6 @@ out:
return ret;
}
-static inline int netpoll_rx_on(struct sk_buff *skb)
-{
- struct netpoll_info *npinfo = rcu_dereference_bh(skb->dev->npinfo);
-
- return npinfo && (!list_empty(&npinfo->rx_np) || npinfo->rx_flags);
-}
-
static inline int netpoll_receive_skb(struct sk_buff *skb)
{
if (!list_empty(&skb->dev->napi_list))
--
1.7.7.6
^ permalink raw reply related
* [PATCH 8/8] bridge: use list_for_each_entry() in netpoll functions
From: Cong Wang @ 2012-08-06 14:23 UTC (permalink / raw)
To: netdev; +Cc: Cong Wang, Stephen Hemminger
In-Reply-To: <1344263012-4031-1-git-send-email-amwang@redhat.com>
We don't delete 'p' from the list in the loop,
so we can just use list_for_each_entry().
Cc: Stephen Hemminger <shemminger@vyatta.com>
Signed-off-by: Cong Wang <amwang@redhat.com>
---
net/bridge/br_device.c | 9 ++++-----
1 files changed, 4 insertions(+), 5 deletions(-)
diff --git a/net/bridge/br_device.c b/net/bridge/br_device.c
index f41ba40..32211fa 100644
--- a/net/bridge/br_device.c
+++ b/net/bridge/br_device.c
@@ -206,21 +206,20 @@ static void br_poll_controller(struct net_device *br_dev)
static void br_netpoll_cleanup(struct net_device *dev)
{
struct net_bridge *br = netdev_priv(dev);
- struct net_bridge_port *p, *n;
+ struct net_bridge_port *p;
- list_for_each_entry_safe(p, n, &br->port_list, list) {
+ list_for_each_entry(p, &br->port_list, list)
br_netpoll_disable(p);
- }
}
static int br_netpoll_setup(struct net_device *dev, struct netpoll_info *ni,
gfp_t gfp)
{
struct net_bridge *br = netdev_priv(dev);
- struct net_bridge_port *p, *n;
+ struct net_bridge_port *p;
int err = 0;
- list_for_each_entry_safe(p, n, &br->port_list, list) {
+ list_for_each_entry(p, &br->port_list, list) {
if (!p->dev)
continue;
err = br_netpoll_enable(p, gfp);
--
1.7.7.6
^ permalink raw reply related
page: next (older) | prev (newer) | latest
- recent:[subjects (threaded)|topics (new)|topics (active)]
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox