* Re: [RFC PATCH] vsprintf: Add %p*D extension for 80211 SSIDs
From: Joe Perches @ 2013-01-07 6:37 UTC (permalink / raw)
To: Chen Gang
Cc: John W. Linville, stas.yakovlev-Re5JQEeQqe8AvxtiuMwx3w,
linux-wireless-u79uwXL29TY76Z2rM5mHXA,
netdev-u79uwXL29TY76Z2rM5mHXA
In-Reply-To: <50EA6612.6010506-bOixZGp5f+dBDgjK7y7TUQ@public.gmane.org>
On Mon, 2013-01-07 at 14:07 +0800, Chen Gang wrote:
> 于 2013年01月07日 12:49, Joe Perches 写道:
> > On Sun, 2013-01-06 at 19:19 -0800, Joe Perches wrote:
> >> Maybe these days this should be another vsprintf %p extension
> >> like %pM when the DECLARE_MAC_BUF/print_mac uses were converted.
> >>
> >> (or maybe extend %ph for ssids with %*phs, length, array)
> >
> excuse me:
> I do not quite know how to reply the [RFC PATCH].
> it would be better if you can tell me how to do for [RFC PATCH], thanks.
You did fine except you unnecessarily quoted the entire original email.
Remember to trim your replies please. _lots_ of people read these
mailing lists and unnecessary quoting wastes all of their times.
> at least for me:
> although it is good idea to add common, widely used features in public tools function.
It's akin to most of the minor uuid/bluetooth mac, etc codes
in vsprintf I've added.
cheers, Joe
--
To unsubscribe from this list: send the line "unsubscribe linux-wireless" in
the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
^ permalink raw reply
* Re: [RFC PATCH] vsprintf: Add %p*D extension for 80211 SSIDs
From: Chen Gang @ 2013-01-07 6:58 UTC (permalink / raw)
To: Joe Perches
Cc: John W. Linville, stas.yakovlev-Re5JQEeQqe8AvxtiuMwx3w,
linux-wireless-u79uwXL29TY76Z2rM5mHXA,
netdev-u79uwXL29TY76Z2rM5mHXA
In-Reply-To: <1357540670.21481.38.camel@joe-AO722>
于 2013年01月07日 14:37, Joe Perches 写道:
> You did fine except you unnecessarily quoted the entire original email.
> Remember to trim your replies please. _lots_ of people read these
> mailing lists and unnecessary quoting wastes all of their times.
>
ok, thank you.
I should notice, next time. :-)
>> > at least for me:
>> > although it is good idea to add common, widely used features in public tools function.
> It's akin to most of the minor uuid/bluetooth mac, etc codes
after see the comments for function pointer in lib/vsprintf.c
sorry for I do not think it is suitable to add ssid in vsprintf
all formate features in vsprintf are face to kernel wide, not for one sub system
one exception is for mac and ip:
they have already been well known in kernel wide
every one knows about mac and ip.
I feel, our ssid is not quite well known in kernel wide
maybe current place (net/wireless/) is not quite suitable for print_ssid:
for bluetooth, it is not belong to net/wireless.
if necessary, I suggest to reconstruct it within net sub system wide, not in kernel wide.
(it is also my fault for my original reply, it is not quite clear)
Regards
--
Chen Gang
Asianux Corporation
--
To unsubscribe from this list: send the line "unsubscribe linux-wireless" in
the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
^ permalink raw reply
* Re: [RFC PATCH 0/3] Make rfc3686 template work with asynchronous block ciphers
From: Steffen Klassert @ 2013-01-07 7:00 UTC (permalink / raw)
To: Herbert Xu; +Cc: David Miller, jussi.kivilinna, linux-crypto, netdev
In-Reply-To: <20121229000812.GA29866@gondor.apana.org.au>
On Sat, Dec 29, 2012 at 11:08:12AM +1100, Herbert Xu wrote:
> On Fri, Dec 28, 2012 at 03:23:11PM -0800, David Miller wrote:
> > From: Herbert Xu <herbert@gondor.apana.org.au>
> > Date: Sat, 29 Dec 2012 09:42:49 +1100
> >
> > > On Fri, Dec 28, 2012 at 12:04:48PM +0200, Jussi Kivilinna wrote:
> > >> I'm not sure how this patchset should be dealt with (should 1st patch go
> > >> through different tree than 2nd and 3rd?), so therefore it's RFC.
> > >>
> > >> Second patch makes rfc3686 template work with asynchronous block ciphers and
> > >> third patch changes aesni-intel to use this template. First patch fixed problem
> > >> in xfrm_algo found with help of 2nd and 3rd patches and without 1st patch
> > >> 2nd patch breaks aes-ctr with IPSEC.
> > >
> > > I think the easiest way is to have these patches go through Dave's
> > > tree.
> >
> > Better yet, Steffen's IPSEC tree :-)
>
> Of course :)
Herbert, can I add your Ack to the crypto patches?
^ permalink raw reply
* [PATCH V2 1/2] virtio-net: fix the set affinity bug when CPU IDs are not consecutive
From: Wanlong Gao @ 2013-01-07 7:14 UTC (permalink / raw)
To: linux-kernel; +Cc: Michael S. Tsirkin, netdev, virtualization, Eric Dumazet
As M.S.T mentioned, set affinity will not work very well when
CPU IDs are not consecutive, this can happen with hot unplug.
Fix this bug by traversal the online CPUs.
Cc: Rusty Russell <rusty@rustcorp.com.au>
Cc: "Michael S. Tsirkin" <mst@redhat.com>
Cc: Jason Wang <jasowang@redhat.com>
Cc: Eric Dumazet <erdnetdev@gmail.com>
Cc: virtualization@lists.linux-foundation.org
Cc: netdev@vger.kernel.org
Signed-off-by: Wanlong Gao <gaowanlong@cn.fujitsu.com>
---
drivers/net/virtio_net.c | 24 +++++++++++++++++-------
1 file changed, 17 insertions(+), 7 deletions(-)
diff --git a/drivers/net/virtio_net.c b/drivers/net/virtio_net.c
index a6fcf15..b483fb5 100644
--- a/drivers/net/virtio_net.c
+++ b/drivers/net/virtio_net.c
@@ -1016,6 +1016,7 @@ static int virtnet_vlan_rx_kill_vid(struct net_device *dev, u16 vid)
static void virtnet_set_affinity(struct virtnet_info *vi, bool set)
{
int i;
+ int cpu;
/* In multiqueue mode, when the number of cpu is equal to the number of
* queue pairs, we let the queue pairs to be private to one cpu by
@@ -1029,16 +1030,25 @@ static void virtnet_set_affinity(struct virtnet_info *vi, bool set)
return;
}
- for (i = 0; i < vi->max_queue_pairs; i++) {
- int cpu = set ? i : -1;
- virtqueue_set_affinity(vi->rq[i].vq, cpu);
- virtqueue_set_affinity(vi->sq[i].vq, cpu);
- }
+ if (set) {
+ i = 0;
+ for_each_online_cpu(cpu) {
+ virtqueue_set_affinity(vi->rq[i].vq, cpu);
+ virtqueue_set_affinity(vi->sq[i].vq, cpu);
+ i++;
+ if (i >= vi->max_queue_pairs)
+ break;
+ }
- if (set)
vi->affinity_hint_set = true;
- else
+ } else {
+ for(i = 0; i < vi->max_queue_pairs; i++) {
+ virtqueue_set_affinity(vi->rq[i].vq, -1);
+ virtqueue_set_affinity(vi->sq[i].vq, -1);
+ }
+
vi->affinity_hint_set = false;
+ }
}
static void virtnet_get_ringparam(struct net_device *dev,
--
1.8.1
^ permalink raw reply related
* [PATCH V2 2/2] virtio-net: reset virtqueue affinity when doing cpu hotplug
From: Wanlong Gao @ 2013-01-07 7:15 UTC (permalink / raw)
To: linux-kernel; +Cc: Michael S. Tsirkin, netdev, virtualization, Eric Dumazet
In-Reply-To: <1357542900-15557-1-git-send-email-gaowanlong@cn.fujitsu.com>
Add a cpu notifier to virtio-net, so that we can reset the
virtqueue affinity if the cpu hotplug happens. It improve
the performance through enabling or disabling the virtqueue
affinity after doing cpu hotplug.
Adding the notifier block to virtnet_info is suggested by
Jason, thank you.
Cc: Rusty Russell <rusty@rustcorp.com.au>
Cc: "Michael S. Tsirkin" <mst@redhat.com>
Cc: Jason Wang <jasowang@redhat.com>
Cc: Eric Dumazet <erdnetdev@gmail.com>
Cc: virtualization@lists.linux-foundation.org
Cc: netdev@vger.kernel.org
Signed-off-by: Wanlong Gao <gaowanlong@cn.fujitsu.com>
---
drivers/net/virtio_net.c | 30 ++++++++++++++++++++++++++++++
1 file changed, 30 insertions(+)
diff --git a/drivers/net/virtio_net.c b/drivers/net/virtio_net.c
index b483fb5..9547b4c 100644
--- a/drivers/net/virtio_net.c
+++ b/drivers/net/virtio_net.c
@@ -26,6 +26,7 @@
#include <linux/scatterlist.h>
#include <linux/if_vlan.h>
#include <linux/slab.h>
+#include <linux/cpu.h>
static int napi_weight = 128;
module_param(napi_weight, int, 0444);
@@ -123,6 +124,9 @@ struct virtnet_info {
/* Does the affinity hint is set for virtqueues? */
bool affinity_hint_set;
+
+ /* CPU hot plug notifier */
+ struct notifier_block nb;
};
struct skb_vnet_hdr {
@@ -1051,6 +1055,23 @@ static void virtnet_set_affinity(struct virtnet_info *vi, bool set)
}
}
+static int virtnet_cpu_callback(struct notifier_block *nfb,
+ unsigned long action, void *hcpu)
+{
+ struct virtnet_info *vi = container_of(nfb, struct virtnet_info, nb);
+ switch(action) {
+ case CPU_ONLINE:
+ case CPU_ONLINE_FROZEN:
+ case CPU_DEAD:
+ case CPU_DEAD_FROZEN:
+ virtnet_set_affinity(vi, true);
+ break;
+ default:
+ break;
+ }
+ return NOTIFY_OK;
+}
+
static void virtnet_get_ringparam(struct net_device *dev,
struct ethtool_ringparam *ring)
{
@@ -1509,6 +1530,13 @@ static int virtnet_probe(struct virtio_device *vdev)
}
}
+ vi->nb.notifier_call = &virtnet_cpu_callback;
+ err = register_hotcpu_notifier(&vi->nb);
+ if (err) {
+ pr_debug("virtio_net: registering cpu notifier failed\n");
+ goto free_recv_bufs;
+ }
+
/* Assume link up if device can't report link status,
otherwise get link status from config. */
if (virtio_has_feature(vi->vdev, VIRTIO_NET_F_STATUS)) {
@@ -1553,6 +1581,8 @@ static void virtnet_remove(struct virtio_device *vdev)
{
struct virtnet_info *vi = vdev->priv;
+ unregister_hotcpu_notifier(&vi->nb);
+
/* Prevent config work handler from accessing the device. */
mutex_lock(&vi->config_lock);
vi->config_enable = false;
--
1.8.1
^ permalink raw reply related
* [PATCH net-next v4 0/3] netpoll: add IPv6 support
From: Cong Wang @ 2013-01-07 7:17 UTC (permalink / raw)
To: netdev; +Cc: Cong Wang
v4: remove ndisc_parse_options(), ignore ICMP options there
make udp6_csum_init() inline so that we don't rely on IPV6=y
fix a conflict with Jiri's patchset
v3: fix CONFIG_IPV6=m case
v2: use bool for ipv6 flag
fix !CONFIG_IPV6 case
This patchset implements netpoll over IPv6.
Cong Wang (3):
netpoll: prepare for ipv6
netpoll: add IPv6 support
netconsole: add IPv6 example in doc
Documentation/networking/netconsole.txt | 7 +-
drivers/net/netconsole.c | 46 ++-
include/linux/netpoll.h | 13 +-
include/net/udp.h | 34 ++
net/core/netpoll.c | 617 ++++++++++++++++++++++---------
net/ipv6/udp.c | 34 --
6 files changed, 534 insertions(+), 217 deletions(-)
--
1.7.7.6
^ permalink raw reply
* [PATCH net-next v4 1/3] netpoll: prepare for ipv6
From: Cong Wang @ 2013-01-07 7:17 UTC (permalink / raw)
To: netdev; +Cc: David S. Miller, Cong Wang
In-Reply-To: <1357543036-15492-1-git-send-email-amwang@redhat.com>
From: Cong Wang <amwang@redhat.com>
This patch adjusts some struct and functions, to prepare
for supporting IPv6.
Cc: David S. Miller <davem@davemloft.net>
Signed-off-by: Cong Wang <amwang@redhat.com>
---
drivers/net/netconsole.c | 12 +-
include/linux/netpoll.h | 13 ++-
net/core/netpoll.c | 402 ++++++++++++++++++++++++++--------------------
3 files changed, 243 insertions(+), 184 deletions(-)
diff --git a/drivers/net/netconsole.c b/drivers/net/netconsole.c
index 6989ebe..998fa02 100644
--- a/drivers/net/netconsole.c
+++ b/drivers/net/netconsole.c
@@ -269,12 +269,14 @@ static ssize_t show_remote_port(struct netconsole_target *nt, char *buf)
static ssize_t show_local_ip(struct netconsole_target *nt, char *buf)
{
- return snprintf(buf, PAGE_SIZE, "%pI4\n", &nt->np.local_ip);
+ if (!nt->np.ipv6)
+ return snprintf(buf, PAGE_SIZE, "%pI4\n", &nt->np.local_ip);
}
static ssize_t show_remote_ip(struct netconsole_target *nt, char *buf)
{
- return snprintf(buf, PAGE_SIZE, "%pI4\n", &nt->np.remote_ip);
+ if (!nt->np.ipv6)
+ return snprintf(buf, PAGE_SIZE, "%pI4\n", &nt->np.remote_ip);
}
static ssize_t show_local_mac(struct netconsole_target *nt, char *buf)
@@ -410,7 +412,8 @@ static ssize_t store_local_ip(struct netconsole_target *nt,
return -EINVAL;
}
- nt->np.local_ip = in_aton(buf);
+ if (!strnchr(buf, count, ':'))
+ nt->np.local_ip.ip = in_aton(buf);
return strnlen(buf, count);
}
@@ -426,7 +429,8 @@ static ssize_t store_remote_ip(struct netconsole_target *nt,
return -EINVAL;
}
- nt->np.remote_ip = in_aton(buf);
+ if (!strnchr(buf, count, ':'))
+ nt->np.remote_ip.ip = in_aton(buf);
return strnlen(buf, count);
}
diff --git a/include/linux/netpoll.h b/include/linux/netpoll.h
index 66d5379..f54c3bb 100644
--- a/include/linux/netpoll.h
+++ b/include/linux/netpoll.h
@@ -12,13 +12,22 @@
#include <linux/rcupdate.h>
#include <linux/list.h>
+union inet_addr {
+ __u32 all[4];
+ __be32 ip;
+ __be32 ip6[4];
+ struct in_addr in;
+ struct in6_addr in6;
+};
+
struct netpoll {
struct net_device *dev;
char dev_name[IFNAMSIZ];
const char *name;
void (*rx_hook)(struct netpoll *, int, char *, int);
- __be32 local_ip, remote_ip;
+ union inet_addr local_ip, remote_ip;
+ bool ipv6;
u16 local_port, remote_port;
u8 remote_mac[ETH_ALEN];
@@ -33,7 +42,7 @@ struct netpoll_info {
spinlock_t rx_lock;
struct list_head rx_np; /* netpolls that registered an rx_hook */
- struct sk_buff_head arp_tx; /* list of arp requests to reply to */
+ struct sk_buff_head neigh_tx; /* list of neigh requests to reply to */
struct sk_buff_head txq;
struct delayed_work tx_work;
diff --git a/net/core/netpoll.c b/net/core/netpoll.c
index d2bda8e..6bd0736 100644
--- a/net/core/netpoll.c
+++ b/net/core/netpoll.c
@@ -55,7 +55,7 @@ static atomic_t trapped;
MAX_UDP_CHUNK)
static void zap_completion_queue(void);
-static void netpoll_arp_reply(struct sk_buff *skb, struct netpoll_info *npinfo);
+static void netpoll_neigh_reply(struct sk_buff *skb, struct netpoll_info *npinfo);
static unsigned int carrier_timeout = 4;
module_param(carrier_timeout, uint, 0644);
@@ -181,13 +181,13 @@ static void poll_napi(struct net_device *dev)
}
}
-static void service_arp_queue(struct netpoll_info *npi)
+static void service_neigh_queue(struct netpoll_info *npi)
{
if (npi) {
struct sk_buff *skb;
- while ((skb = skb_dequeue(&npi->arp_tx)))
- netpoll_arp_reply(skb, npi);
+ while ((skb = skb_dequeue(&npi->neigh_tx)))
+ netpoll_neigh_reply(skb, npi);
}
}
@@ -216,14 +216,14 @@ static void netpoll_poll_dev(struct net_device *dev)
bond_dev = netdev_master_upper_dev_get_rcu(dev);
bond_ni = rcu_dereference_bh(bond_dev->npinfo);
- while ((skb = skb_dequeue(&ni->arp_tx))) {
+ while ((skb = skb_dequeue(&ni->neigh_tx))) {
skb->dev = bond_dev;
- skb_queue_tail(&bond_ni->arp_tx, skb);
+ skb_queue_tail(&bond_ni->neigh_tx, skb);
}
}
}
- service_arp_queue(ni);
+ service_neigh_queue(ni);
zap_completion_queue();
}
@@ -386,7 +386,9 @@ void netpoll_send_udp(struct netpoll *np, const char *msg, int len)
static atomic_t ip_ident;
udp_len = len + sizeof(*udph);
- ip_len = udp_len + sizeof(*iph);
+ if (!np->ipv6)
+ ip_len = udp_len + sizeof(*iph);
+
total_len = ip_len + LL_RESERVED_SPACE(np->dev);
skb = find_skb(np, total_len + np->dev->needed_tailroom,
@@ -403,34 +405,38 @@ void netpoll_send_udp(struct netpoll *np, const char *msg, int len)
udph->source = htons(np->local_port);
udph->dest = htons(np->remote_port);
udph->len = htons(udp_len);
- udph->check = 0;
- udph->check = csum_tcpudp_magic(np->local_ip,
- np->remote_ip,
- udp_len, IPPROTO_UDP,
- csum_partial(udph, udp_len, 0));
- if (udph->check == 0)
- udph->check = CSUM_MANGLED_0;
-
- skb_push(skb, sizeof(*iph));
- skb_reset_network_header(skb);
- iph = ip_hdr(skb);
-
- /* iph->version = 4; iph->ihl = 5; */
- put_unaligned(0x45, (unsigned char *)iph);
- iph->tos = 0;
- put_unaligned(htons(ip_len), &(iph->tot_len));
- iph->id = htons(atomic_inc_return(&ip_ident));
- iph->frag_off = 0;
- iph->ttl = 64;
- iph->protocol = IPPROTO_UDP;
- iph->check = 0;
- put_unaligned(np->local_ip, &(iph->saddr));
- put_unaligned(np->remote_ip, &(iph->daddr));
- iph->check = ip_fast_csum((unsigned char *)iph, iph->ihl);
-
- eth = (struct ethhdr *) skb_push(skb, ETH_HLEN);
- skb_reset_mac_header(skb);
- skb->protocol = eth->h_proto = htons(ETH_P_IP);
+
+ if (!np->ipv6) {
+ udph->check = 0;
+ udph->check = csum_tcpudp_magic(np->local_ip.ip,
+ np->remote_ip.ip,
+ udp_len, IPPROTO_UDP,
+ csum_partial(udph, udp_len, 0));
+ if (udph->check == 0)
+ udph->check = CSUM_MANGLED_0;
+
+ skb_push(skb, sizeof(*iph));
+ skb_reset_network_header(skb);
+ iph = ip_hdr(skb);
+
+ /* iph->version = 4; iph->ihl = 5; */
+ put_unaligned(0x45, (unsigned char *)iph);
+ iph->tos = 0;
+ put_unaligned(htons(ip_len), &(iph->tot_len));
+ iph->id = htons(atomic_inc_return(&ip_ident));
+ iph->frag_off = 0;
+ iph->ttl = 64;
+ iph->protocol = IPPROTO_UDP;
+ iph->check = 0;
+ put_unaligned(np->local_ip.ip, &(iph->saddr));
+ put_unaligned(np->remote_ip.ip, &(iph->daddr));
+ iph->check = ip_fast_csum((unsigned char *)iph, iph->ihl);
+
+ eth = (struct ethhdr *) skb_push(skb, ETH_HLEN);
+ skb_reset_mac_header(skb);
+ skb->protocol = eth->h_proto = htons(ETH_P_IP);
+ }
+
memcpy(eth->h_source, np->dev->dev_addr, ETH_ALEN);
memcpy(eth->h_dest, np->remote_mac, ETH_ALEN);
@@ -440,7 +446,7 @@ void netpoll_send_udp(struct netpoll *np, const char *msg, int len)
}
EXPORT_SYMBOL(netpoll_send_udp);
-static void netpoll_arp_reply(struct sk_buff *skb, struct netpoll_info *npinfo)
+static void netpoll_neigh_reply(struct sk_buff *skb, struct netpoll_info *npinfo)
{
struct arphdr *arp;
unsigned char *arp_ptr;
@@ -451,7 +457,7 @@ static void netpoll_arp_reply(struct sk_buff *skb, struct netpoll_info *npinfo)
struct netpoll *np, *tmp;
unsigned long flags;
int hlen, tlen;
- int hits = 0;
+ int hits = 0, proto;
if (list_empty(&npinfo->rx_np))
return;
@@ -469,94 +475,97 @@ static void netpoll_arp_reply(struct sk_buff *skb, struct netpoll_info *npinfo)
if (!hits)
return;
- /* No arp on this interface */
- if (skb->dev->flags & IFF_NOARP)
- return;
-
- if (!pskb_may_pull(skb, arp_hdr_len(skb->dev)))
- return;
+ proto = ntohs(eth_hdr(skb)->h_proto);
+ if (proto == ETH_P_IP) {
+ /* No arp on this interface */
+ if (skb->dev->flags & IFF_NOARP)
+ return;
- skb_reset_network_header(skb);
- skb_reset_transport_header(skb);
- arp = arp_hdr(skb);
+ if (!pskb_may_pull(skb, arp_hdr_len(skb->dev)))
+ return;
- if ((arp->ar_hrd != htons(ARPHRD_ETHER) &&
- arp->ar_hrd != htons(ARPHRD_IEEE802)) ||
- arp->ar_pro != htons(ETH_P_IP) ||
- arp->ar_op != htons(ARPOP_REQUEST))
- return;
+ skb_reset_network_header(skb);
+ skb_reset_transport_header(skb);
+ arp = arp_hdr(skb);
- arp_ptr = (unsigned char *)(arp+1);
- /* save the location of the src hw addr */
- sha = arp_ptr;
- arp_ptr += skb->dev->addr_len;
- memcpy(&sip, arp_ptr, 4);
- arp_ptr += 4;
- /* If we actually cared about dst hw addr,
- it would get copied here */
- arp_ptr += skb->dev->addr_len;
- memcpy(&tip, arp_ptr, 4);
-
- /* Should we ignore arp? */
- if (ipv4_is_loopback(tip) || ipv4_is_multicast(tip))
- return;
+ if ((arp->ar_hrd != htons(ARPHRD_ETHER) &&
+ arp->ar_hrd != htons(ARPHRD_IEEE802)) ||
+ arp->ar_pro != htons(ETH_P_IP) ||
+ arp->ar_op != htons(ARPOP_REQUEST))
+ return;
- size = arp_hdr_len(skb->dev);
+ arp_ptr = (unsigned char *)(arp+1);
+ /* save the location of the src hw addr */
+ sha = arp_ptr;
+ arp_ptr += skb->dev->addr_len;
+ memcpy(&sip, arp_ptr, 4);
+ arp_ptr += 4;
+ /* If we actually cared about dst hw addr,
+ it would get copied here */
+ arp_ptr += skb->dev->addr_len;
+ memcpy(&tip, arp_ptr, 4);
- spin_lock_irqsave(&npinfo->rx_lock, flags);
- list_for_each_entry_safe(np, tmp, &npinfo->rx_np, rx) {
- if (tip != np->local_ip)
- continue;
+ /* Should we ignore arp? */
+ if (ipv4_is_loopback(tip) || ipv4_is_multicast(tip))
+ return;
- hlen = LL_RESERVED_SPACE(np->dev);
- tlen = np->dev->needed_tailroom;
- send_skb = find_skb(np, size + hlen + tlen, hlen);
- if (!send_skb)
- continue;
+ size = arp_hdr_len(skb->dev);
- skb_reset_network_header(send_skb);
- arp = (struct arphdr *) skb_put(send_skb, size);
- send_skb->dev = skb->dev;
- send_skb->protocol = htons(ETH_P_ARP);
+ spin_lock_irqsave(&npinfo->rx_lock, flags);
+ list_for_each_entry_safe(np, tmp, &npinfo->rx_np, rx) {
+ if (tip != np->local_ip.ip)
+ continue;
+
+ hlen = LL_RESERVED_SPACE(np->dev);
+ tlen = np->dev->needed_tailroom;
+ send_skb = find_skb(np, size + hlen + tlen, hlen);
+ if (!send_skb)
+ continue;
+
+ skb_reset_network_header(send_skb);
+ arp = (struct arphdr *) skb_put(send_skb, size);
+ send_skb->dev = skb->dev;
+ send_skb->protocol = htons(ETH_P_ARP);
+
+ /* Fill the device header for the ARP frame */
+ if (dev_hard_header(send_skb, skb->dev, ptype,
+ sha, np->dev->dev_addr,
+ send_skb->len) < 0) {
+ kfree_skb(send_skb);
+ continue;
+ }
- /* Fill the device header for the ARP frame */
- if (dev_hard_header(send_skb, skb->dev, ptype,
- sha, np->dev->dev_addr,
- send_skb->len) < 0) {
- kfree_skb(send_skb);
- continue;
+ /*
+ * Fill out the arp protocol part.
+ *
+ * we only support ethernet device type,
+ * which (according to RFC 1390) should
+ * always equal 1 (Ethernet).
+ */
+
+ arp->ar_hrd = htons(np->dev->type);
+ arp->ar_pro = htons(ETH_P_IP);
+ arp->ar_hln = np->dev->addr_len;
+ arp->ar_pln = 4;
+ arp->ar_op = htons(type);
+
+ arp_ptr = (unsigned char *)(arp + 1);
+ memcpy(arp_ptr, np->dev->dev_addr, np->dev->addr_len);
+ arp_ptr += np->dev->addr_len;
+ memcpy(arp_ptr, &tip, 4);
+ arp_ptr += 4;
+ memcpy(arp_ptr, sha, np->dev->addr_len);
+ arp_ptr += np->dev->addr_len;
+ memcpy(arp_ptr, &sip, 4);
+
+ netpoll_send_skb(np, send_skb);
+
+ /* If there are several rx_hooks for the same address,
+ we're fine by sending a single reply */
+ break;
}
-
- /*
- * Fill out the arp protocol part.
- *
- * we only support ethernet device type,
- * which (according to RFC 1390) should
- * always equal 1 (Ethernet).
- */
-
- arp->ar_hrd = htons(np->dev->type);
- arp->ar_pro = htons(ETH_P_IP);
- arp->ar_hln = np->dev->addr_len;
- arp->ar_pln = 4;
- arp->ar_op = htons(type);
-
- arp_ptr = (unsigned char *)(arp + 1);
- memcpy(arp_ptr, np->dev->dev_addr, np->dev->addr_len);
- arp_ptr += np->dev->addr_len;
- memcpy(arp_ptr, &tip, 4);
- arp_ptr += 4;
- memcpy(arp_ptr, sha, np->dev->addr_len);
- arp_ptr += np->dev->addr_len;
- memcpy(arp_ptr, &sip, 4);
-
- netpoll_send_skb(np, send_skb);
-
- /* If there are several rx_hooks for the same address,
- we're fine by sending a single reply */
- break;
+ spin_unlock_irqrestore(&npinfo->rx_lock, flags);
}
- spin_unlock_irqrestore(&npinfo->rx_lock, flags);
}
int __netpoll_rx(struct sk_buff *skb, struct netpoll_info *npinfo)
@@ -576,7 +585,7 @@ int __netpoll_rx(struct sk_buff *skb, struct netpoll_info *npinfo)
/* check if netpoll clients need ARP */
if (skb->protocol == htons(ETH_P_ARP) &&
atomic_read(&trapped)) {
- skb_queue_tail(&npinfo->arp_tx, skb);
+ skb_queue_tail(&npinfo->neigh_tx, skb);
return 1;
}
@@ -587,60 +596,61 @@ int __netpoll_rx(struct sk_buff *skb, struct netpoll_info *npinfo)
}
proto = ntohs(eth_hdr(skb)->h_proto);
- if (proto != ETH_P_IP)
+ if (proto != ETH_P_IP && proto != ETH_P_IPV6)
goto out;
if (skb->pkt_type == PACKET_OTHERHOST)
goto out;
if (skb_shared(skb))
goto out;
- if (!pskb_may_pull(skb, sizeof(struct iphdr)))
- goto out;
- iph = (struct iphdr *)skb->data;
- if (iph->ihl < 5 || iph->version != 4)
- goto out;
- if (!pskb_may_pull(skb, iph->ihl*4))
- goto out;
- iph = (struct iphdr *)skb->data;
- if (ip_fast_csum((u8 *)iph, iph->ihl) != 0)
- goto out;
-
- len = ntohs(iph->tot_len);
- if (skb->len < len || len < iph->ihl*4)
- goto out;
-
- /*
- * Our transport medium may have padded the buffer out.
- * Now We trim to the true length of the frame.
- */
- if (pskb_trim_rcsum(skb, len))
- goto out;
+ if (proto == ETH_P_IP) {
+ if (!pskb_may_pull(skb, sizeof(struct iphdr)))
+ goto out;
+ iph = (struct iphdr *)skb->data;
+ if (iph->ihl < 5 || iph->version != 4)
+ goto out;
+ if (!pskb_may_pull(skb, iph->ihl*4))
+ goto out;
+ iph = (struct iphdr *)skb->data;
+ if (ip_fast_csum((u8 *)iph, iph->ihl) != 0)
+ goto out;
- iph = (struct iphdr *)skb->data;
- if (iph->protocol != IPPROTO_UDP)
- goto out;
+ len = ntohs(iph->tot_len);
+ if (skb->len < len || len < iph->ihl*4)
+ goto out;
- len -= iph->ihl*4;
- uh = (struct udphdr *)(((char *)iph) + iph->ihl*4);
- ulen = ntohs(uh->len);
+ /*
+ * Our transport medium may have padded the buffer out.
+ * Now We trim to the true length of the frame.
+ */
+ if (pskb_trim_rcsum(skb, len))
+ goto out;
- if (ulen != len)
- goto out;
- if (checksum_udp(skb, uh, ulen, iph->saddr, iph->daddr))
- goto out;
+ iph = (struct iphdr *)skb->data;
+ if (iph->protocol != IPPROTO_UDP)
+ goto out;
- list_for_each_entry_safe(np, tmp, &npinfo->rx_np, rx) {
- if (np->local_ip && np->local_ip != iph->daddr)
- continue;
- if (np->remote_ip && np->remote_ip != iph->saddr)
- continue;
- if (np->local_port && np->local_port != ntohs(uh->dest))
- continue;
+ len -= iph->ihl*4;
+ uh = (struct udphdr *)(((char *)iph) + iph->ihl*4);
+ ulen = ntohs(uh->len);
- np->rx_hook(np, ntohs(uh->source),
- (char *)(uh+1),
- ulen - sizeof(struct udphdr));
- hits++;
+ if (ulen != len)
+ goto out;
+ if (checksum_udp(skb, uh, ulen, iph->saddr, iph->daddr))
+ goto out;
+ list_for_each_entry_safe(np, tmp, &npinfo->rx_np, rx) {
+ if (np->local_ip.ip && np->local_ip.ip != iph->daddr)
+ continue;
+ if (np->remote_ip.ip && np->remote_ip.ip != iph->saddr)
+ continue;
+ if (np->local_port && np->local_port != ntohs(uh->dest))
+ continue;
+
+ np->rx_hook(np, ntohs(uh->source),
+ (char *)(uh+1),
+ ulen - sizeof(struct udphdr));
+ hits++;
+ }
}
if (!hits)
@@ -661,17 +671,40 @@ out:
void netpoll_print_options(struct netpoll *np)
{
np_info(np, "local port %d\n", np->local_port);
- np_info(np, "local IP %pI4\n", &np->local_ip);
+ if (!np->ipv6)
+ np_info(np, "local IPv4 address %pI4\n", &np->local_ip.ip);
np_info(np, "interface '%s'\n", np->dev_name);
np_info(np, "remote port %d\n", np->remote_port);
- np_info(np, "remote IP %pI4\n", &np->remote_ip);
+ if (!np->ipv6)
+ np_info(np, "remote IPv4 address %pI4\n", &np->remote_ip.ip);
np_info(np, "remote ethernet address %pM\n", np->remote_mac);
}
EXPORT_SYMBOL(netpoll_print_options);
+static int netpoll_parse_ip_addr(const char *str, union inet_addr *addr)
+{
+ const char *end;
+
+ if (!strchr(str, ':') &&
+ in4_pton(str, -1, (void *)addr, -1, &end) > 0) {
+ if (!*end)
+ return 0;
+ }
+ if (in6_pton(str, -1, addr->in6.s6_addr, -1, &end) > 0) {
+#if IS_ENABLED(CONFIG_IPV6)
+ if (!*end)
+ return 1;
+#else
+ return -1;
+#endif
+ }
+ return -1;
+}
+
int netpoll_parse_options(struct netpoll *np, char *opt)
{
char *cur=opt, *delim;
+ int ipv6;
if (*cur != '@') {
if ((delim = strchr(cur, '@')) == NULL)
@@ -687,7 +720,11 @@ int netpoll_parse_options(struct netpoll *np, char *opt)
if ((delim = strchr(cur, '/')) == NULL)
goto parse_failed;
*delim = 0;
- np->local_ip = in_aton(cur);
+ ipv6 = netpoll_parse_ip_addr(cur, &np->local_ip);
+ if (ipv6 < 0)
+ goto parse_failed;
+ else
+ np->ipv6 = (bool)ipv6;
cur = delim;
}
cur++;
@@ -719,7 +756,13 @@ int netpoll_parse_options(struct netpoll *np, char *opt)
if ((delim = strchr(cur, '/')) == NULL)
goto parse_failed;
*delim = 0;
- np->remote_ip = in_aton(cur);
+ ipv6 = netpoll_parse_ip_addr(cur, &np->remote_ip);
+ if (ipv6 < 0)
+ goto parse_failed;
+ else if (np->ipv6 != (bool)ipv6)
+ goto parse_failed;
+ else
+ np->ipv6 = (bool)ipv6;
cur = delim + 1;
if (*cur != 0) {
@@ -767,7 +810,7 @@ int __netpoll_setup(struct netpoll *np, struct net_device *ndev, gfp_t gfp)
INIT_LIST_HEAD(&npinfo->rx_np);
spin_lock_init(&npinfo->rx_lock);
- skb_queue_head_init(&npinfo->arp_tx);
+ skb_queue_head_init(&npinfo->neigh_tx);
skb_queue_head_init(&npinfo->txq);
INIT_DELAYED_WORK(&npinfo->tx_work, queue_process);
@@ -859,21 +902,24 @@ int netpoll_setup(struct netpoll *np)
}
}
- if (!np->local_ip) {
- rcu_read_lock();
- in_dev = __in_dev_get_rcu(ndev);
+ if (!np->local_ip.ip) {
+ if (!np->ipv6) {
+ rcu_read_lock();
+ in_dev = __in_dev_get_rcu(ndev);
- if (!in_dev || !in_dev->ifa_list) {
+
+ if (!in_dev || !in_dev->ifa_list) {
+ rcu_read_unlock();
+ np_err(np, "no IP address for %s, aborting\n",
+ np->dev_name);
+ err = -EDESTADDRREQ;
+ goto put;
+ }
+
+ np->local_ip.ip = in_dev->ifa_list->ifa_local;
rcu_read_unlock();
- np_err(np, "no IP address for %s, aborting\n",
- np->dev_name);
- err = -EDESTADDRREQ;
- goto put;
+ np_info(np, "local IP %pI4\n", &np->local_ip.ip);
}
-
- np->local_ip = in_dev->ifa_list->ifa_local;
- rcu_read_unlock();
- np_info(np, "local IP %pI4\n", &np->local_ip);
}
/* fill up the skb queue */
@@ -906,7 +952,7 @@ 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->neigh_tx);
skb_queue_purge(&npinfo->txq);
/* we can't call cancel_delayed_work_sync here, as we are in softirq */
--
1.7.7.6
^ permalink raw reply related
* [PATCH net-next v4 2/3] netpoll: add IPv6 support
From: Cong Wang @ 2013-01-07 7:17 UTC (permalink / raw)
To: netdev; +Cc: David S. Miller, Cong Wang
In-Reply-To: <1357543036-15492-1-git-send-email-amwang@redhat.com>
From: Cong Wang <amwang@redhat.com>
Currently, netpoll only supports IPv4. This patch adds IPv6
support to netpoll so that we can run netconsole over IPv6 network.
Cc: David S. Miller <davem@davemloft.net>
Signed-off-by: Cong Wang <amwang@redhat.com>
---
drivers/net/netconsole.c | 46 ++++++++-
include/net/udp.h | 34 +++++++
net/core/netpoll.c | 239 ++++++++++++++++++++++++++++++++++++++++++++--
net/ipv6/udp.c | 34 -------
4 files changed, 303 insertions(+), 50 deletions(-)
diff --git a/drivers/net/netconsole.c b/drivers/net/netconsole.c
index 998fa02..62f494b 100644
--- a/drivers/net/netconsole.c
+++ b/drivers/net/netconsole.c
@@ -269,13 +269,17 @@ static ssize_t show_remote_port(struct netconsole_target *nt, char *buf)
static ssize_t show_local_ip(struct netconsole_target *nt, char *buf)
{
- if (!nt->np.ipv6)
+ if (nt->np.ipv6)
+ return snprintf(buf, PAGE_SIZE, "%pI6c\n", &nt->np.local_ip.in6);
+ else
return snprintf(buf, PAGE_SIZE, "%pI4\n", &nt->np.local_ip);
}
static ssize_t show_remote_ip(struct netconsole_target *nt, char *buf)
{
- if (!nt->np.ipv6)
+ if (nt->np.ipv6)
+ return snprintf(buf, PAGE_SIZE, "%pI6c\n", &nt->np.remote_ip.in6);
+ else
return snprintf(buf, PAGE_SIZE, "%pI4\n", &nt->np.remote_ip);
}
@@ -412,8 +416,23 @@ static ssize_t store_local_ip(struct netconsole_target *nt,
return -EINVAL;
}
- if (!strnchr(buf, count, ':'))
- nt->np.local_ip.ip = in_aton(buf);
+ if (strnchr(buf, count, ':')) {
+ const char *end;
+ if (in6_pton(buf, count, nt->np.local_ip.in6.s6_addr, -1, &end) > 0) {
+ if (*end && *end != '\n') {
+ printk(KERN_ERR "netconsole: invalid IPv6 address at: <%c>\n", *end);
+ return -EINVAL;
+ }
+ nt->np.ipv6 = true;
+ } else
+ return -EINVAL;
+ } else {
+ if (!nt->np.ipv6) {
+ nt->np.local_ip.ip = in_aton(buf);
+ nt->np.ipv6 = false;
+ } else
+ return -EINVAL;
+ }
return strnlen(buf, count);
}
@@ -429,8 +448,23 @@ static ssize_t store_remote_ip(struct netconsole_target *nt,
return -EINVAL;
}
- if (!strnchr(buf, count, ':'))
- nt->np.remote_ip.ip = in_aton(buf);
+ if (strnchr(buf, count, ':')) {
+ const char *end;
+ if (in6_pton(buf, count, nt->np.remote_ip.in6.s6_addr, -1, &end) > 0) {
+ if (*end && *end != '\n') {
+ printk(KERN_ERR "netconsole: invalid IPv6 address at: <%c>\n", *end);
+ return -EINVAL;
+ }
+ nt->np.ipv6 = true;
+ } else
+ return -EINVAL;
+ } else {
+ if (!nt->np.ipv6) {
+ nt->np.remote_ip.ip = in_aton(buf);
+ nt->np.ipv6 = false;
+ } else
+ return -EINVAL;
+ }
return strnlen(buf, count);
}
diff --git a/include/net/udp.h b/include/net/udp.h
index 065f379..73e712a 100644
--- a/include/net/udp.h
+++ b/include/net/udp.h
@@ -156,6 +156,40 @@ static inline __wsum udp_csum(struct sk_buff *skb)
return csum;
}
+#include <net/udplite.h>
+static inline int udp6_csum_init(struct sk_buff *skb, struct udphdr *uh, int proto)
+{
+ int err;
+
+ UDP_SKB_CB(skb)->partial_cov = 0;
+ UDP_SKB_CB(skb)->cscov = skb->len;
+
+ if (proto == IPPROTO_UDPLITE) {
+ err = udplite_checksum_init(skb, uh);
+ if (err)
+ return err;
+ }
+
+ if (uh->check == 0) {
+ /* RFC 2460 section 8.1 says that we SHOULD log
+ this error. Well, it is reasonable.
+ */
+ LIMIT_NETDEBUG(KERN_INFO "IPv6: udp checksum is 0\n");
+ return 1;
+ }
+ if (skb->ip_summed == CHECKSUM_COMPLETE &&
+ !csum_ipv6_magic(&ipv6_hdr(skb)->saddr, &ipv6_hdr(skb)->daddr,
+ skb->len, proto, skb->csum))
+ skb->ip_summed = CHECKSUM_UNNECESSARY;
+
+ if (!skb_csum_unnecessary(skb))
+ skb->csum = ~csum_unfold(csum_ipv6_magic(&ipv6_hdr(skb)->saddr,
+ &ipv6_hdr(skb)->daddr,
+ skb->len, proto, 0));
+
+ return 0;
+}
+
/* hash routines shared between UDPv4/6 and UDP-Litev4/6 */
static inline void udp_lib_hash(struct sock *sk)
{
diff --git a/net/core/netpoll.c b/net/core/netpoll.c
index 6bd0736..cd9c121 100644
--- a/net/core/netpoll.c
+++ b/net/core/netpoll.c
@@ -29,6 +29,8 @@
#include <linux/if_vlan.h>
#include <net/tcp.h>
#include <net/udp.h>
+#include <net/addrconf.h>
+#include <net/ndisc.h>
#include <asm/unaligned.h>
#include <trace/events/napi.h>
@@ -384,9 +386,12 @@ void netpoll_send_udp(struct netpoll *np, const char *msg, int len)
struct iphdr *iph;
struct ethhdr *eth;
static atomic_t ip_ident;
+ struct ipv6hdr *ip6h;
udp_len = len + sizeof(*udph);
- if (!np->ipv6)
+ if (np->ipv6)
+ ip_len = udp_len + sizeof(*ip6h);
+ else
ip_len = udp_len + sizeof(*iph);
total_len = ip_len + LL_RESERVED_SPACE(np->dev);
@@ -406,7 +411,35 @@ void netpoll_send_udp(struct netpoll *np, const char *msg, int len)
udph->dest = htons(np->remote_port);
udph->len = htons(udp_len);
- if (!np->ipv6) {
+ if (np->ipv6) {
+ udph->check = 0;
+ udph->check = csum_ipv6_magic(&np->local_ip.in6,
+ &np->remote_ip.in6,
+ udp_len, IPPROTO_UDP,
+ csum_partial(udph, udp_len, 0));
+ if (udph->check == 0)
+ udph->check = CSUM_MANGLED_0;
+
+ skb_push(skb, sizeof(*ip6h));
+ skb_reset_network_header(skb);
+ ip6h = ipv6_hdr(skb);
+
+ /* ip6h->version = 6; ip6h->priority = 0; */
+ put_unaligned(0x60, (unsigned char *)ip6h);
+ ip6h->flow_lbl[0] = 0;
+ ip6h->flow_lbl[1] = 0;
+ ip6h->flow_lbl[2] = 0;
+
+ ip6h->payload_len = htons(sizeof(struct udphdr) + len);
+ ip6h->nexthdr = IPPROTO_UDP;
+ ip6h->hop_limit = 32;
+ ip6h->saddr = np->local_ip.in6;
+ ip6h->daddr = np->remote_ip.in6;
+
+ eth = (struct ethhdr *) skb_push(skb, ETH_HLEN);
+ skb_reset_mac_header(skb);
+ skb->protocol = eth->h_proto = htons(ETH_P_IPV6);
+ } else {
udph->check = 0;
udph->check = csum_tcpudp_magic(np->local_ip.ip,
np->remote_ip.ip,
@@ -448,9 +481,7 @@ EXPORT_SYMBOL(netpoll_send_udp);
static void netpoll_neigh_reply(struct sk_buff *skb, struct netpoll_info *npinfo)
{
- struct arphdr *arp;
- unsigned char *arp_ptr;
- int size, type = ARPOP_REPLY, ptype = ETH_P_ARP;
+ int size, type = ARPOP_REPLY;
__be32 sip, tip;
unsigned char *sha;
struct sk_buff *send_skb;
@@ -477,6 +508,8 @@ static void netpoll_neigh_reply(struct sk_buff *skb, struct netpoll_info *npinfo
proto = ntohs(eth_hdr(skb)->h_proto);
if (proto == ETH_P_IP) {
+ struct arphdr *arp;
+ unsigned char *arp_ptr;
/* No arp on this interface */
if (skb->dev->flags & IFF_NOARP)
return;
@@ -528,7 +561,7 @@ static void netpoll_neigh_reply(struct sk_buff *skb, struct netpoll_info *npinfo
send_skb->protocol = htons(ETH_P_ARP);
/* Fill the device header for the ARP frame */
- if (dev_hard_header(send_skb, skb->dev, ptype,
+ if (dev_hard_header(send_skb, skb->dev, ETH_P_ARP,
sha, np->dev->dev_addr,
send_skb->len) < 0) {
kfree_skb(send_skb);
@@ -565,9 +598,124 @@ static void netpoll_neigh_reply(struct sk_buff *skb, struct netpoll_info *npinfo
break;
}
spin_unlock_irqrestore(&npinfo->rx_lock, flags);
+ } else if( proto == ETH_P_IPV6) {
+#if IS_ENABLED(CONFIG_IPV6)
+ struct nd_msg *msg;
+ u8 *lladdr = NULL;
+ struct ipv6hdr *hdr;
+ struct icmp6hdr *icmp6h;
+ const struct in6_addr *saddr;
+ const struct in6_addr *daddr;
+ struct inet6_dev *in6_dev = NULL;
+ struct in6_addr *target;
+
+ in6_dev = in6_dev_get(skb->dev);
+ if (!in6_dev || !in6_dev->cnf.accept_ra)
+ return;
+
+ if (!pskb_may_pull(skb, skb->len))
+ return;
+
+ msg = (struct nd_msg *)skb_transport_header(skb);
+
+ __skb_push(skb, skb->data - skb_transport_header(skb));
+
+ if (ipv6_hdr(skb)->hop_limit != 255)
+ return;
+ if (msg->icmph.icmp6_code != 0)
+ return;
+ if (msg->icmph.icmp6_type != NDISC_NEIGHBOUR_SOLICITATION)
+ return;
+
+ saddr = &ipv6_hdr(skb)->saddr;
+ daddr = &ipv6_hdr(skb)->daddr;
+
+ size = sizeof(struct icmp6hdr) + sizeof(struct in6_addr);
+
+ spin_lock_irqsave(&npinfo->rx_lock, flags);
+ list_for_each_entry_safe(np, tmp, &npinfo->rx_np, rx) {
+ if (memcmp(daddr, &np->local_ip, sizeof(*daddr)))
+ continue;
+
+ hlen = LL_RESERVED_SPACE(np->dev);
+ tlen = np->dev->needed_tailroom;
+ send_skb = find_skb(np, size + hlen + tlen, hlen);
+ if (!send_skb)
+ continue;
+
+ send_skb->protocol = htons(ETH_P_IPV6);
+ send_skb->dev = skb->dev;
+
+ skb_reset_network_header(send_skb);
+ skb_put(send_skb, sizeof(struct ipv6hdr));
+ hdr = ipv6_hdr(send_skb);
+
+ *(__be32*)hdr = htonl(0x60000000);
+
+ hdr->payload_len = htons(size);
+ hdr->nexthdr = IPPROTO_ICMPV6;
+ hdr->hop_limit = 255;
+ hdr->saddr = *saddr;
+ hdr->daddr = *daddr;
+
+ send_skb->transport_header = send_skb->tail;
+ skb_put(send_skb, size);
+
+ icmp6h = (struct icmp6hdr *)skb_transport_header(skb);
+ icmp6h->icmp6_type = NDISC_NEIGHBOUR_ADVERTISEMENT;
+ icmp6h->icmp6_router = 0;
+ icmp6h->icmp6_solicited = 1;
+ target = (struct in6_addr *)skb_transport_header(send_skb) + sizeof(struct icmp6hdr);
+ *target = msg->target;
+ icmp6h->icmp6_cksum = csum_ipv6_magic(saddr, daddr, size,
+ IPPROTO_ICMPV6,
+ csum_partial(icmp6h,
+ size, 0));
+
+ if (dev_hard_header(send_skb, skb->dev, ETH_P_IPV6,
+ lladdr, np->dev->dev_addr,
+ send_skb->len) < 0) {
+ kfree_skb(send_skb);
+ continue;
+ }
+
+ netpoll_send_skb(np, send_skb);
+
+ /* If there are several rx_hooks for the same address,
+ we're fine by sending a single reply */
+ break;
+ }
+ spin_unlock_irqrestore(&npinfo->rx_lock, flags);
+#endif
}
}
+static bool pkt_is_ns(struct sk_buff *skb)
+{
+ struct nd_msg *msg;
+ struct ipv6hdr *hdr;
+
+ if (skb->protocol != htons(ETH_P_ARP))
+ return false;
+ if (!pskb_may_pull(skb, sizeof(struct ipv6hdr) + sizeof(struct nd_msg)))
+ return false;
+
+ msg = (struct nd_msg *)skb_transport_header(skb);
+ __skb_push(skb, skb->data - skb_transport_header(skb));
+ hdr = ipv6_hdr(skb);
+
+ if (hdr->nexthdr != IPPROTO_ICMPV6)
+ return false;
+ if (hdr->hop_limit != 255)
+ return false;
+ if (msg->icmph.icmp6_code != 0)
+ return false;
+ if (msg->icmph.icmp6_type != NDISC_NEIGHBOUR_SOLICITATION)
+ return false;
+
+ return true;
+}
+
int __netpoll_rx(struct sk_buff *skb, struct netpoll_info *npinfo)
{
int proto, len, ulen;
@@ -583,8 +731,10 @@ int __netpoll_rx(struct sk_buff *skb, struct netpoll_info *npinfo)
goto out;
/* check if netpoll clients need ARP */
- if (skb->protocol == htons(ETH_P_ARP) &&
- atomic_read(&trapped)) {
+ if (skb->protocol == htons(ETH_P_ARP) && atomic_read(&trapped)) {
+ skb_queue_tail(&npinfo->neigh_tx, skb);
+ return 1;
+ } else if (pkt_is_ns(skb) && atomic_read(&trapped)) {
skb_queue_tail(&npinfo->neigh_tx, skb);
return 1;
}
@@ -651,6 +801,45 @@ int __netpoll_rx(struct sk_buff *skb, struct netpoll_info *npinfo)
ulen - sizeof(struct udphdr));
hits++;
}
+ } else {
+#if IS_ENABLED(CONFIG_IPV6)
+ const struct ipv6hdr *ip6h;
+
+ if (!pskb_may_pull(skb, sizeof(struct ipv6hdr)))
+ goto out;
+ ip6h = (struct ipv6hdr *)skb->data;
+ if (ip6h->version != 6)
+ goto out;
+ len = ntohs(ip6h->payload_len);
+ if (!len)
+ goto out;
+ if (len + sizeof(struct ipv6hdr) > skb->len)
+ goto out;
+ if (pskb_trim_rcsum(skb, len + sizeof(struct ipv6hdr)))
+ goto out;
+ ip6h = ipv6_hdr(skb);
+ if (!pskb_may_pull(skb, sizeof(struct udphdr)))
+ goto out;
+ uh = udp_hdr(skb);
+ ulen = ntohs(uh->len);
+ if (ulen != skb->len)
+ goto out;
+ if (udp6_csum_init(skb, uh, IPPROTO_UDP))
+ goto out;
+ list_for_each_entry_safe(np, tmp, &npinfo->rx_np, rx) {
+ if (memcmp(&np->local_ip.in6, &ip6h->daddr, sizeof(struct in6_addr)) != 0)
+ continue;
+ if (memcmp(&np->remote_ip.in6, &ip6h->saddr, sizeof(struct in6_addr)) != 0)
+ continue;
+ if (np->local_port && np->local_port != ntohs(uh->dest))
+ continue;
+
+ np->rx_hook(np, ntohs(uh->source),
+ (char *)(uh+1),
+ ulen - sizeof(struct udphdr));
+ hits++;
+ }
+#endif
}
if (!hits)
@@ -671,11 +860,15 @@ out:
void netpoll_print_options(struct netpoll *np)
{
np_info(np, "local port %d\n", np->local_port);
- if (!np->ipv6)
+ if (np->ipv6)
+ np_info(np, "local IPv6 address %pI6c\n", &np->local_ip.in6);
+ else
np_info(np, "local IPv4 address %pI4\n", &np->local_ip.ip);
np_info(np, "interface '%s'\n", np->dev_name);
np_info(np, "remote port %d\n", np->remote_port);
- if (!np->ipv6)
+ if (np->ipv6)
+ np_info(np, "remote IPv6 address %pI6c\n", &np->remote_ip.in6);
+ else
np_info(np, "remote IPv4 address %pI4\n", &np->remote_ip.ip);
np_info(np, "remote ethernet address %pM\n", np->remote_mac);
}
@@ -919,6 +1112,32 @@ int netpoll_setup(struct netpoll *np)
np->local_ip.ip = in_dev->ifa_list->ifa_local;
rcu_read_unlock();
np_info(np, "local IP %pI4\n", &np->local_ip.ip);
+ } else {
+ struct inet6_dev *idev;
+
+ err = -EDESTADDRREQ;
+ rcu_read_lock();
+ idev = __in6_dev_get(ndev);
+ if (idev) {
+ struct inet6_ifaddr *ifp;
+
+ read_lock_bh(&idev->lock);
+ list_for_each_entry(ifp, &idev->addr_list, if_list) {
+ if (ipv6_addr_type(&ifp->addr) & IPV6_ADDR_LINKLOCAL)
+ continue;
+ np->local_ip.in6 = ifp->addr;
+ err = 0;
+ break;
+ }
+ read_unlock_bh(&idev->lock);
+ }
+ rcu_read_unlock();
+ if (err) {
+ np_err(np, "no IPv6 address for %s, aborting\n",
+ np->dev_name);
+ goto put;
+ } else
+ np_info(np, "local IPv6 %pI6c\n", &np->local_ip.in6);
}
}
diff --git a/net/ipv6/udp.c b/net/ipv6/udp.c
index dfaa29b..1afb635 100644
--- a/net/ipv6/udp.c
+++ b/net/ipv6/udp.c
@@ -752,40 +752,6 @@ static int __udp6_lib_mcast_deliver(struct net *net, struct sk_buff *skb,
return 0;
}
-static inline int udp6_csum_init(struct sk_buff *skb, struct udphdr *uh,
- int proto)
-{
- int err;
-
- UDP_SKB_CB(skb)->partial_cov = 0;
- UDP_SKB_CB(skb)->cscov = skb->len;
-
- if (proto == IPPROTO_UDPLITE) {
- err = udplite_checksum_init(skb, uh);
- if (err)
- return err;
- }
-
- if (uh->check == 0) {
- /* RFC 2460 section 8.1 says that we SHOULD log
- this error. Well, it is reasonable.
- */
- LIMIT_NETDEBUG(KERN_INFO "IPv6: udp checksum is 0\n");
- return 1;
- }
- if (skb->ip_summed == CHECKSUM_COMPLETE &&
- !csum_ipv6_magic(&ipv6_hdr(skb)->saddr, &ipv6_hdr(skb)->daddr,
- skb->len, proto, skb->csum))
- skb->ip_summed = CHECKSUM_UNNECESSARY;
-
- if (!skb_csum_unnecessary(skb))
- skb->csum = ~csum_unfold(csum_ipv6_magic(&ipv6_hdr(skb)->saddr,
- &ipv6_hdr(skb)->daddr,
- skb->len, proto, 0));
-
- return 0;
-}
-
int __udp6_lib_rcv(struct sk_buff *skb, struct udp_table *udptable,
int proto)
{
--
1.7.7.6
^ permalink raw reply related
* [PATCH net-next v4 3/3] netconsole: add IPv6 example in doc
From: Cong Wang @ 2013-01-07 7:17 UTC (permalink / raw)
To: netdev; +Cc: David S. Miller, Cong Wang
In-Reply-To: <1357543036-15492-1-git-send-email-amwang@redhat.com>
From: Cong Wang <amwang@redhat.com>
Update the netconsole document as well.
Cc: David S. Miller <davem@davemloft.net>
Signed-off-by: Cong Wang <amwang@redhat.com>
---
Documentation/networking/netconsole.txt | 7 ++++++-
1 files changed, 6 insertions(+), 1 deletions(-)
diff --git a/Documentation/networking/netconsole.txt b/Documentation/networking/netconsole.txt
index 2e9e0ae2..a5d574a 100644
--- a/Documentation/networking/netconsole.txt
+++ b/Documentation/networking/netconsole.txt
@@ -1,9 +1,10 @@
started by Ingo Molnar <mingo@redhat.com>, 2001.09.17
2.6 port and netpoll api by Matt Mackall <mpm@selenic.com>, Sep 9 2003
+IPv6 support by Cong Wang <xiyou.wangcong@gmail.com>, Jan 1 2013
Please send bug reports to Matt Mackall <mpm@selenic.com>
-and Satyam Sharma <satyam.sharma@gmail.com>
+Satyam Sharma <satyam.sharma@gmail.com>, and Cong Wang <xiyou.wangcong@gmail.com>
Introduction:
=============
@@ -41,6 +42,10 @@ Examples:
insmod netconsole netconsole=@/,@10.0.0.2/
+ or using IPv6
+
+ insmod netconsole netconsole=@/,@fd00:1:2:3::1/
+
It also supports logging to multiple remote agents by specifying
parameters for the multiple agents separated by semicolons and the
complete string enclosed in "quotes", thusly:
--
1.7.7.6
^ permalink raw reply related
* [RFC PATCH v2 net-next 0/2] Remove NETIF_F_NETNS_LOCAL from software devices.
From: Rami Rosen @ 2013-01-07 7:19 UTC (permalink / raw)
To: davem; +Cc: netdev, shemminger, ebiederm, Rami Rosen
When NETIF_F_NETNS_LOCAL feature is set in some network device,
we do not allow this device to be moved from one network namespace
to another (see dev_change_net_namespace()). Vlan device is not considered
a local device and does not have the NETIF_F_NETNS_LOCAL feature. There is no
reason that software devices like ppp and bridge will set the
NETIF_F_NETNS_LOCAL feature. This patchset removes NETIF_F_NETNS_LOCAL
from ppp and bridge devices.
* Changes from v1
Do not remove NETIF_F_NETNS_LOCAL from vxlan devices. Moving vxlan device
to a different namesapce does not move the state of the UDP socket related
to this vxlan device (Following Stephen Hemminger comment).
Rami Rosen (2):
[RFC PATCH v2 net-next 1/3] ppp: remove NETIF_F_NETNS_LOCAL from ppp
device features.
[RFC PATCH v2 net-next 2/3] bridge: remove NETIF_F_NETNS_LOCAL from
bridge device features.
drivers/net/ppp/ppp_generic.c | 1 -
net/bridge/br_device.c | 2 +-
2 files changed, 1 insertion(+), 2 deletions(-)
--
1.7.11.7
^ permalink raw reply
* [RFC PATCH v2 net-next 1/2] ppp: remove NETIF_F_NETNS_LOCAL from ppp device features.
From: Rami Rosen @ 2013-01-07 7:19 UTC (permalink / raw)
To: davem; +Cc: netdev, shemminger, ebiederm, Rami Rosen
In-Reply-To: <1357543176-15352-1-git-send-email-ramirose@gmail.com>
There is no need for NETIF_F_NETNS_LOCAL for ppp device; this patch removes it.
Signed-off-by: Rami Rosen <ramirose@gmail.com>
---
drivers/net/ppp/ppp_generic.c | 1 -
1 file changed, 1 deletion(-)
diff --git a/drivers/net/ppp/ppp_generic.c b/drivers/net/ppp/ppp_generic.c
index 0b2706a..89b6fc5 100644
--- a/drivers/net/ppp/ppp_generic.c
+++ b/drivers/net/ppp/ppp_generic.c
@@ -1073,7 +1073,6 @@ static void ppp_setup(struct net_device *dev)
dev->tx_queue_len = 3;
dev->type = ARPHRD_PPP;
dev->flags = IFF_POINTOPOINT | IFF_NOARP | IFF_MULTICAST;
- dev->features |= NETIF_F_NETNS_LOCAL;
dev->priv_flags &= ~IFF_XMIT_DST_RELEASE;
}
--
1.7.11.7
^ permalink raw reply related
* [RFC PATCH v2 net-next 2/2] bridge: remove NETIF_F_NETNS_LOCAL from bridge device features.
From: Rami Rosen @ 2013-01-07 7:19 UTC (permalink / raw)
To: davem; +Cc: netdev, shemminger, ebiederm, Rami Rosen
In-Reply-To: <1357543176-15352-1-git-send-email-ramirose@gmail.com>
There is no need for NETIF_F_NETNS_LOCAL for bridge device; this patch removes it.
Signed-off-by: Rami Rosen <ramirose@gmail.com>
---
net/bridge/br_device.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/net/bridge/br_device.c b/net/bridge/br_device.c
index e1bc090..0f88d21 100644
--- a/net/bridge/br_device.c
+++ b/net/bridge/br_device.c
@@ -344,7 +344,7 @@ void br_dev_setup(struct net_device *dev)
dev->features = NETIF_F_SG | NETIF_F_FRAGLIST | NETIF_F_HIGHDMA |
NETIF_F_GSO_MASK | NETIF_F_HW_CSUM | NETIF_F_LLTX |
- NETIF_F_NETNS_LOCAL | NETIF_F_HW_VLAN_TX;
+ NETIF_F_HW_VLAN_TX;
dev->hw_features = NETIF_F_SG | NETIF_F_FRAGLIST | NETIF_F_HIGHDMA |
NETIF_F_GSO_MASK | NETIF_F_HW_CSUM |
NETIF_F_HW_VLAN_TX;
--
1.7.11.7
^ permalink raw reply related
* Re: [PATCH V2 2/2] virtio-net: reset virtqueue affinity when doing cpu hotplug
From: Jason Wang @ 2013-01-07 7:28 UTC (permalink / raw)
To: Wanlong Gao
Cc: Michael S. Tsirkin, netdev, linux-kernel, virtualization,
Eric Dumazet
In-Reply-To: <1357542900-15557-2-git-send-email-gaowanlong@cn.fujitsu.com>
On 01/07/2013 03:15 PM, Wanlong Gao wrote:
> Add a cpu notifier to virtio-net, so that we can reset the
> virtqueue affinity if the cpu hotplug happens. It improve
> the performance through enabling or disabling the virtqueue
> affinity after doing cpu hotplug.
> Adding the notifier block to virtnet_info is suggested by
> Jason, thank you.
>
> Cc: Rusty Russell <rusty@rustcorp.com.au>
> Cc: "Michael S. Tsirkin" <mst@redhat.com>
> Cc: Jason Wang <jasowang@redhat.com>
> Cc: Eric Dumazet <erdnetdev@gmail.com>
> Cc: virtualization@lists.linux-foundation.org
> Cc: netdev@vger.kernel.org
> Signed-off-by: Wanlong Gao <gaowanlong@cn.fujitsu.com>
> ---
> drivers/net/virtio_net.c | 30 ++++++++++++++++++++++++++++++
> 1 file changed, 30 insertions(+)
>
> diff --git a/drivers/net/virtio_net.c b/drivers/net/virtio_net.c
> index b483fb5..9547b4c 100644
> --- a/drivers/net/virtio_net.c
> +++ b/drivers/net/virtio_net.c
> @@ -26,6 +26,7 @@
> #include <linux/scatterlist.h>
> #include <linux/if_vlan.h>
> #include <linux/slab.h>
> +#include <linux/cpu.h>
>
> static int napi_weight = 128;
> module_param(napi_weight, int, 0444);
> @@ -123,6 +124,9 @@ struct virtnet_info {
>
> /* Does the affinity hint is set for virtqueues? */
> bool affinity_hint_set;
> +
> + /* CPU hot plug notifier */
> + struct notifier_block nb;
> };
>
> struct skb_vnet_hdr {
> @@ -1051,6 +1055,23 @@ static void virtnet_set_affinity(struct virtnet_info *vi, bool set)
> }
> }
>
> +static int virtnet_cpu_callback(struct notifier_block *nfb,
> + unsigned long action, void *hcpu)
> +{
> + struct virtnet_info *vi = container_of(nfb, struct virtnet_info, nb);
> + switch(action) {
> + case CPU_ONLINE:
> + case CPU_ONLINE_FROZEN:
> + case CPU_DEAD:
> + case CPU_DEAD_FROZEN:
> + virtnet_set_affinity(vi, true);
> + break;
> + default:
> + break;
> + }
> + return NOTIFY_OK;
> +}
> +
I think you'd better fix the .ndo_select_queue() as well (as Michael
said in your V1) since it currently uses smp processor id which may not
work very well in this case also.
Thanks
> static void virtnet_get_ringparam(struct net_device *dev,
> struct ethtool_ringparam *ring)
> {
> @@ -1509,6 +1530,13 @@ static int virtnet_probe(struct virtio_device *vdev)
> }
> }
>
> + vi->nb.notifier_call = &virtnet_cpu_callback;
> + err = register_hotcpu_notifier(&vi->nb);
> + if (err) {
> + pr_debug("virtio_net: registering cpu notifier failed\n");
> + goto free_recv_bufs;
> + }
> +
> /* Assume link up if device can't report link status,
> otherwise get link status from config. */
> if (virtio_has_feature(vi->vdev, VIRTIO_NET_F_STATUS)) {
> @@ -1553,6 +1581,8 @@ static void virtnet_remove(struct virtio_device *vdev)
> {
> struct virtnet_info *vi = vdev->priv;
>
> + unregister_hotcpu_notifier(&vi->nb);
> +
> /* Prevent config work handler from accessing the device. */
> mutex_lock(&vi->config_lock);
> vi->config_enable = false;
^ permalink raw reply
* Re: [PATCH 1/2] usbnet: allow status interrupt URB to always be active
From: Oliver Neukum @ 2013-01-07 7:40 UTC (permalink / raw)
To: Bjørn Mork
Cc: Dan Williams, Elina Pasheva, netdev-u79uwXL29TY76Z2rM5mHXA,
linux-usb-u79uwXL29TY76Z2rM5mHXA, Rory Filer, Phil Sutter
In-Reply-To: <87y5g7nbej.fsf-lbf33ChDnrE/G1V5fR+Y7Q@public.gmane.org>
On Saturday 05 January 2013 11:59:16 Bjørn Mork wrote:
Hi,
> You can still use a trick similar to what qmi_wwan and cdc_mbim does to
> take over the status endpoint from usbnet: By not implementing .status,
> and possibly setting dev->status to NULL in .bind, you are free to
> handle the status endpoint entirely inside the minidriver. Not sure if
> that is smart though. You would have to reimplement init_status and
> intr_complete from usbnet, and kill or resubmit the interrupt urb on
> suspend/resume/disconnect yourself.
Please no. We don't look for ways to circumvent frameworks. We alter
frameworks to act as we need them to act.
> The new usbnet flag is probably a better solution.
Indeed.
> FWIW, I agree with Oliver that always submitting the interrupt URB is
> both risky and will cause too much unnecessary USB activity for most
> usbnet devices.
Good.
Regards
Oliver
--
To unsubscribe from this list: send the line "unsubscribe linux-usb" in
the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
^ permalink raw reply
* Re: [RFC PATCH] vsprintf: Add %p*D extension for 80211 SSIDs
From: Johannes Berg @ 2013-01-07 7:47 UTC (permalink / raw)
To: Joe Perches
Cc: John W. Linville, Chen Gang, stas.yakovlev, linux-wireless,
netdev
In-Reply-To: <1357534195.21481.31.camel@joe-AO722>
On Sun, 2013-01-06 at 20:49 -0800, Joe Perches wrote:
> On Sun, 2013-01-06 at 19:19 -0800, Joe Perches wrote:
> > Maybe these days this should be another vsprintf %p extension
> > like %pM when the DECLARE_MAC_BUF/print_mac uses were converted.
> >
> > (or maybe extend %ph for ssids with %*phs, length, array)
>
> Maybe like this:
> + * - 'D' For a 80211 SSID, it prints the SSID escaping any non-printable
> + characters with a leading \ and octal or [0nrt\]
Honestly, I'm not sure it's worth it. print_ssid() is used in two or
three legacy drivers only, not in any modern driver, and is unlikely to
be used in the more modern drivers due to tracing etc.
johannes
^ permalink raw reply
* Re: [PATCH V2 2/2] virtio-net: reset virtqueue affinity when doing cpu hotplug
From: Wanlong Gao @ 2013-01-07 7:48 UTC (permalink / raw)
To: Jason Wang
Cc: Michael S. Tsirkin, netdev, linux-kernel, virtualization,
Eric Dumazet
In-Reply-To: <50EA792F.1010408@redhat.com>
On 01/07/2013 03:28 PM, Jason Wang wrote:
> On 01/07/2013 03:15 PM, Wanlong Gao wrote:
>> Add a cpu notifier to virtio-net, so that we can reset the
>> virtqueue affinity if the cpu hotplug happens. It improve
>> the performance through enabling or disabling the virtqueue
>> affinity after doing cpu hotplug.
>> Adding the notifier block to virtnet_info is suggested by
>> Jason, thank you.
>>
>> Cc: Rusty Russell <rusty@rustcorp.com.au>
>> Cc: "Michael S. Tsirkin" <mst@redhat.com>
>> Cc: Jason Wang <jasowang@redhat.com>
>> Cc: Eric Dumazet <erdnetdev@gmail.com>
>> Cc: virtualization@lists.linux-foundation.org
>> Cc: netdev@vger.kernel.org
>> Signed-off-by: Wanlong Gao <gaowanlong@cn.fujitsu.com>
>> ---
>> drivers/net/virtio_net.c | 30 ++++++++++++++++++++++++++++++
>> 1 file changed, 30 insertions(+)
>>
>> diff --git a/drivers/net/virtio_net.c b/drivers/net/virtio_net.c
>> index b483fb5..9547b4c 100644
>> --- a/drivers/net/virtio_net.c
>> +++ b/drivers/net/virtio_net.c
>> @@ -26,6 +26,7 @@
>> #include <linux/scatterlist.h>
>> #include <linux/if_vlan.h>
>> #include <linux/slab.h>
>> +#include <linux/cpu.h>
>>
>> static int napi_weight = 128;
>> module_param(napi_weight, int, 0444);
>> @@ -123,6 +124,9 @@ struct virtnet_info {
>>
>> /* Does the affinity hint is set for virtqueues? */
>> bool affinity_hint_set;
>> +
>> + /* CPU hot plug notifier */
>> + struct notifier_block nb;
>> };
>>
>> struct skb_vnet_hdr {
>> @@ -1051,6 +1055,23 @@ static void virtnet_set_affinity(struct virtnet_info *vi, bool set)
>> }
>> }
>>
>> +static int virtnet_cpu_callback(struct notifier_block *nfb,
>> + unsigned long action, void *hcpu)
>> +{
>> + struct virtnet_info *vi = container_of(nfb, struct virtnet_info, nb);
>> + switch(action) {
>> + case CPU_ONLINE:
>> + case CPU_ONLINE_FROZEN:
>> + case CPU_DEAD:
>> + case CPU_DEAD_FROZEN:
>> + virtnet_set_affinity(vi, true);
>> + break;
>> + default:
>> + break;
>> + }
>> + return NOTIFY_OK;
>> +}
>> +
>
> I think you'd better fix the .ndo_select_queue() as well (as Michael
> said in your V1) since it currently uses smp processor id which may not
> work very well in this case also.
The bug is we can't get the right txq if the CPU IDs are not consecutive,
right? Do you have any good idea about fixing this?
Thanks,
Wanlong Gao
>
> Thanks
>> static void virtnet_get_ringparam(struct net_device *dev,
>> struct ethtool_ringparam *ring)
>> {
>> @@ -1509,6 +1530,13 @@ static int virtnet_probe(struct virtio_device *vdev)
>> }
>> }
>>
>> + vi->nb.notifier_call = &virtnet_cpu_callback;
>> + err = register_hotcpu_notifier(&vi->nb);
>> + if (err) {
>> + pr_debug("virtio_net: registering cpu notifier failed\n");
>> + goto free_recv_bufs;
>> + }
>> +
>> /* Assume link up if device can't report link status,
>> otherwise get link status from config. */
>> if (virtio_has_feature(vi->vdev, VIRTIO_NET_F_STATUS)) {
>> @@ -1553,6 +1581,8 @@ static void virtnet_remove(struct virtio_device *vdev)
>> {
>> struct virtnet_info *vi = vdev->priv;
>>
>> + unregister_hotcpu_notifier(&vi->nb);
>> +
>> /* Prevent config work handler from accessing the device. */
>> mutex_lock(&vi->config_lock);
>> vi->config_enable = false;
>
>
^ permalink raw reply
* Re: [PATCH V2 2/2] virtio-net: reset virtqueue affinity when doing cpu hotplug
From: Jason Wang @ 2013-01-07 7:55 UTC (permalink / raw)
To: gaowanlong
Cc: Michael S. Tsirkin, netdev, linux-kernel, virtualization,
Eric Dumazet
In-Reply-To: <50EA7DC0.20605@cn.fujitsu.com>
On 01/07/2013 03:48 PM, Wanlong Gao wrote:
> On 01/07/2013 03:28 PM, Jason Wang wrote:
>> On 01/07/2013 03:15 PM, Wanlong Gao wrote:
>>> Add a cpu notifier to virtio-net, so that we can reset the
>>> virtqueue affinity if the cpu hotplug happens. It improve
>>> the performance through enabling or disabling the virtqueue
>>> affinity after doing cpu hotplug.
>>> Adding the notifier block to virtnet_info is suggested by
>>> Jason, thank you.
>>>
>>> Cc: Rusty Russell <rusty@rustcorp.com.au>
>>> Cc: "Michael S. Tsirkin" <mst@redhat.com>
>>> Cc: Jason Wang <jasowang@redhat.com>
>>> Cc: Eric Dumazet <erdnetdev@gmail.com>
>>> Cc: virtualization@lists.linux-foundation.org
>>> Cc: netdev@vger.kernel.org
>>> Signed-off-by: Wanlong Gao <gaowanlong@cn.fujitsu.com>
>>> ---
>>> drivers/net/virtio_net.c | 30 ++++++++++++++++++++++++++++++
>>> 1 file changed, 30 insertions(+)
>>>
>>> diff --git a/drivers/net/virtio_net.c b/drivers/net/virtio_net.c
>>> index b483fb5..9547b4c 100644
>>> --- a/drivers/net/virtio_net.c
>>> +++ b/drivers/net/virtio_net.c
>>> @@ -26,6 +26,7 @@
>>> #include <linux/scatterlist.h>
>>> #include <linux/if_vlan.h>
>>> #include <linux/slab.h>
>>> +#include <linux/cpu.h>
>>>
>>> static int napi_weight = 128;
>>> module_param(napi_weight, int, 0444);
>>> @@ -123,6 +124,9 @@ struct virtnet_info {
>>>
>>> /* Does the affinity hint is set for virtqueues? */
>>> bool affinity_hint_set;
>>> +
>>> + /* CPU hot plug notifier */
>>> + struct notifier_block nb;
>>> };
>>>
>>> struct skb_vnet_hdr {
>>> @@ -1051,6 +1055,23 @@ static void virtnet_set_affinity(struct virtnet_info *vi, bool set)
>>> }
>>> }
>>>
>>> +static int virtnet_cpu_callback(struct notifier_block *nfb,
>>> + unsigned long action, void *hcpu)
>>> +{
>>> + struct virtnet_info *vi = container_of(nfb, struct virtnet_info, nb);
>>> + switch(action) {
>>> + case CPU_ONLINE:
>>> + case CPU_ONLINE_FROZEN:
>>> + case CPU_DEAD:
>>> + case CPU_DEAD_FROZEN:
>>> + virtnet_set_affinity(vi, true);
>>> + break;
>>> + default:
>>> + break;
>>> + }
>>> + return NOTIFY_OK;
>>> +}
>>> +
>> I think you'd better fix the .ndo_select_queue() as well (as Michael
>> said in your V1) since it currently uses smp processor id which may not
>> work very well in this case also.
> The bug is we can't get the right txq if the CPU IDs are not consecutive,
> right? Do you have any good idea about fixing this?
>
> Thanks,
> Wanlong Gao
The point is make the virtqueue private to a specific cpu when the
number of queue pairs is equal to the number of cpus. So after you bind
the vq affinity to a specific cpu, you'd better use the reverse mapping
of this affinity to do .ndo_select_queue(). One possible idea, as
Michael suggested, is a per-cpu structure to record the preferable
virtqueue and do both .ndo_select_queue() and affinity hint setting
based on this.
>
>> Thanks
>>> static void virtnet_get_ringparam(struct net_device *dev,
>>> struct ethtool_ringparam *ring)
>>> {
>>> @@ -1509,6 +1530,13 @@ static int virtnet_probe(struct virtio_device *vdev)
>>> }
>>> }
>>>
>>> + vi->nb.notifier_call = &virtnet_cpu_callback;
>>> + err = register_hotcpu_notifier(&vi->nb);
>>> + if (err) {
>>> + pr_debug("virtio_net: registering cpu notifier failed\n");
>>> + goto free_recv_bufs;
>>> + }
>>> +
>>> /* Assume link up if device can't report link status,
>>> otherwise get link status from config. */
>>> if (virtio_has_feature(vi->vdev, VIRTIO_NET_F_STATUS)) {
>>> @@ -1553,6 +1581,8 @@ static void virtnet_remove(struct virtio_device *vdev)
>>> {
>>> struct virtnet_info *vi = vdev->priv;
>>>
>>> + unregister_hotcpu_notifier(&vi->nb);
>>> +
>>> /* Prevent config work handler from accessing the device. */
>>> mutex_lock(&vi->config_lock);
>>> vi->config_enable = false;
>>
^ permalink raw reply
* re: bnx2x: Prepare device and initialize VF database
From: Dan Carpenter @ 2013-01-07 8:18 UTC (permalink / raw)
To: ariele; +Cc: netdev
Hello Ariel Elior,
The patch b56e9670ffa4: "bnx2x: Prepare device and initialize VF
database" from Jan 1, 2013, leads to the following warning:
drivers/net/ethernet/broadcom/bnx2x/bnx2x_sriov.c:1346
bnx2x_pretend_func()
warn: should this be '>= E1H_FUNC_MAX'
drivers/net/ethernet/broadcom/bnx2x/bnx2x_sriov.c
1342 int bnx2x_pretend_func(struct bnx2x *bp, u16 pretend_func_val)
1343 {
1344 u32 pretend_reg;
1345
1346 if (CHIP_IS_E1H(bp) && pretend_func_val > E1H_FUNC_MAX)
1347 return -1;
1348
This is static checker test has too many false positives so I haven't
released it. Basically, every other place that uses E1H_FUNC_MAX
assumes we start counting from zero except this one. I don't know the
code enough to say if it's a problem or intended.
regards,
dan carpenter
^ permalink raw reply
* Re: [PATCH v2 1/1 net-next] net: fec: report correct hardware stamping info to ethtool
From: Sascha Hauer @ 2013-01-07 8:34 UTC (permalink / raw)
To: Frank Li; +Cc: lznuaa, shawn.guo, richardcochran, davem, linux-arm-kernel,
netdev
In-Reply-To: <1357525507-16148-1-git-send-email-Frank.Li@freescale.com>
On Mon, Jan 07, 2013 at 10:25:07AM +0800, Frank Li wrote:
> Report correct hardware stamping capability by ethtool interface.
> The v1.0 ptp4l check it.
>
> Signed-off-by: Frank Li <Frank.Li@freescale.com>
Acked-by: Sascha Hauer <s.hauer@pengutronix.de>
Sascha
> ---
> Change v1->v2
> 1. s/capibilty/capability/
> 2. apply Sascha's feedback
> avoid return the return value of ethtool_op_get_ts_info()
> when bufdesc_ex == 1
>
> drivers/net/ethernet/freescale/fec.c | 31 ++++++++++++++++++++++++++++++-
> 1 files changed, 30 insertions(+), 1 deletions(-)
>
> diff --git a/drivers/net/ethernet/freescale/fec.c b/drivers/net/ethernet/freescale/fec.c
> index a379319..1c117e5 100644
> --- a/drivers/net/ethernet/freescale/fec.c
> +++ b/drivers/net/ethernet/freescale/fec.c
> @@ -1173,12 +1173,41 @@ static void fec_enet_get_drvinfo(struct net_device *ndev,
> strcpy(info->bus_info, dev_name(&ndev->dev));
> }
>
> +static int fec_enet_get_ts_info(struct net_device *ndev,
> + struct ethtool_ts_info *info)
> +{
> + struct fec_enet_private *fep = netdev_priv(ndev);
> +
> + if (fep->bufdesc_ex) {
> +
> + info->so_timestamping = SOF_TIMESTAMPING_TX_SOFTWARE |
> + SOF_TIMESTAMPING_RX_SOFTWARE |
> + SOF_TIMESTAMPING_SOFTWARE |
> + SOF_TIMESTAMPING_TX_HARDWARE |
> + SOF_TIMESTAMPING_RX_HARDWARE |
> + SOF_TIMESTAMPING_RAW_HARDWARE;
> + if (fep->ptp_clock)
> + info->phc_index = ptp_clock_index(fep->ptp_clock);
> + else
> + info->phc_index = -1;
> +
> + info->tx_types = (1 << HWTSTAMP_TX_OFF) |
> + (1 << HWTSTAMP_TX_ON);
> +
> + info->rx_filters = (1 << HWTSTAMP_FILTER_NONE) |
> + (1 << HWTSTAMP_FILTER_ALL);
> + return 0;
> + } else {
> + return ethtool_op_get_ts_info(ndev, info);
> + }
> +}
> +
> static const struct ethtool_ops fec_enet_ethtool_ops = {
> .get_settings = fec_enet_get_settings,
> .set_settings = fec_enet_set_settings,
> .get_drvinfo = fec_enet_get_drvinfo,
> .get_link = ethtool_op_get_link,
> - .get_ts_info = ethtool_op_get_ts_info,
> + .get_ts_info = fec_enet_get_ts_info,
> };
>
> static int fec_enet_ioctl(struct net_device *ndev, struct ifreq *rq, int cmd)
> --
> 1.7.1
>
>
>
--
Pengutronix e.K. | |
Industrial Linux Solutions | http://www.pengutronix.de/ |
Peiner Str. 6-8, 31137 Hildesheim, Germany | Phone: +49-5121-206917-0 |
Amtsgericht Hildesheim, HRA 2686 | Fax: +49-5121-206917-5555 |
^ permalink raw reply
* Re: [PATCH V2 2/2] virtio-net: reset virtqueue affinity when doing cpu hotplug
From: Wanlong Gao @ 2013-01-07 8:46 UTC (permalink / raw)
To: Jason Wang
Cc: Michael S. Tsirkin, netdev, linux-kernel, virtualization,
Eric Dumazet
In-Reply-To: <50EA7F5D.7050907@redhat.com>
On 01/07/2013 03:55 PM, Jason Wang wrote:
> On 01/07/2013 03:48 PM, Wanlong Gao wrote:
>> On 01/07/2013 03:28 PM, Jason Wang wrote:
>>> On 01/07/2013 03:15 PM, Wanlong Gao wrote:
>>>> Add a cpu notifier to virtio-net, so that we can reset the
>>>> virtqueue affinity if the cpu hotplug happens. It improve
>>>> the performance through enabling or disabling the virtqueue
>>>> affinity after doing cpu hotplug.
>>>> Adding the notifier block to virtnet_info is suggested by
>>>> Jason, thank you.
>>>>
>>>> Cc: Rusty Russell <rusty@rustcorp.com.au>
>>>> Cc: "Michael S. Tsirkin" <mst@redhat.com>
>>>> Cc: Jason Wang <jasowang@redhat.com>
>>>> Cc: Eric Dumazet <erdnetdev@gmail.com>
>>>> Cc: virtualization@lists.linux-foundation.org
>>>> Cc: netdev@vger.kernel.org
>>>> Signed-off-by: Wanlong Gao <gaowanlong@cn.fujitsu.com>
>>>> ---
>>>> drivers/net/virtio_net.c | 30 ++++++++++++++++++++++++++++++
>>>> 1 file changed, 30 insertions(+)
>>>>
>>>> diff --git a/drivers/net/virtio_net.c b/drivers/net/virtio_net.c
>>>> index b483fb5..9547b4c 100644
>>>> --- a/drivers/net/virtio_net.c
>>>> +++ b/drivers/net/virtio_net.c
>>>> @@ -26,6 +26,7 @@
>>>> #include <linux/scatterlist.h>
>>>> #include <linux/if_vlan.h>
>>>> #include <linux/slab.h>
>>>> +#include <linux/cpu.h>
>>>>
>>>> static int napi_weight = 128;
>>>> module_param(napi_weight, int, 0444);
>>>> @@ -123,6 +124,9 @@ struct virtnet_info {
>>>>
>>>> /* Does the affinity hint is set for virtqueues? */
>>>> bool affinity_hint_set;
>>>> +
>>>> + /* CPU hot plug notifier */
>>>> + struct notifier_block nb;
>>>> };
>>>>
>>>> struct skb_vnet_hdr {
>>>> @@ -1051,6 +1055,23 @@ static void virtnet_set_affinity(struct virtnet_info *vi, bool set)
>>>> }
>>>> }
>>>>
>>>> +static int virtnet_cpu_callback(struct notifier_block *nfb,
>>>> + unsigned long action, void *hcpu)
>>>> +{
>>>> + struct virtnet_info *vi = container_of(nfb, struct virtnet_info, nb);
>>>> + switch(action) {
>>>> + case CPU_ONLINE:
>>>> + case CPU_ONLINE_FROZEN:
>>>> + case CPU_DEAD:
>>>> + case CPU_DEAD_FROZEN:
>>>> + virtnet_set_affinity(vi, true);
>>>> + break;
>>>> + default:
>>>> + break;
>>>> + }
>>>> + return NOTIFY_OK;
>>>> +}
>>>> +
>>> I think you'd better fix the .ndo_select_queue() as well (as Michael
>>> said in your V1) since it currently uses smp processor id which may not
>>> work very well in this case also.
>> The bug is we can't get the right txq if the CPU IDs are not consecutive,
>> right? Do you have any good idea about fixing this?
>>
>> Thanks,
>> Wanlong Gao
>
> The point is make the virtqueue private to a specific cpu when the
> number of queue pairs is equal to the number of cpus. So after you bind
> the vq affinity to a specific cpu, you'd better use the reverse mapping
> of this affinity to do .ndo_select_queue(). One possible idea, as
> Michael suggested, is a per-cpu structure to record the preferable
> virtqueue and do both .ndo_select_queue() and affinity hint setting
> based on this.
Yeah, I think I got it now, will address it in V3. thank you. ;)
Regards,
Wanlong Gao
>>
>>> Thanks
>>>> static void virtnet_get_ringparam(struct net_device *dev,
>>>> struct ethtool_ringparam *ring)
>>>> {
>>>> @@ -1509,6 +1530,13 @@ static int virtnet_probe(struct virtio_device *vdev)
>>>> }
>>>> }
>>>>
>>>> + vi->nb.notifier_call = &virtnet_cpu_callback;
>>>> + err = register_hotcpu_notifier(&vi->nb);
>>>> + if (err) {
>>>> + pr_debug("virtio_net: registering cpu notifier failed\n");
>>>> + goto free_recv_bufs;
>>>> + }
>>>> +
>>>> /* Assume link up if device can't report link status,
>>>> otherwise get link status from config. */
>>>> if (virtio_has_feature(vi->vdev, VIRTIO_NET_F_STATUS)) {
>>>> @@ -1553,6 +1581,8 @@ static void virtnet_remove(struct virtio_device *vdev)
>>>> {
>>>> struct virtnet_info *vi = vdev->priv;
>>>>
>>>> + unregister_hotcpu_notifier(&vi->nb);
>>>> +
>>>> /* Prevent config work handler from accessing the device. */
>>>> mutex_lock(&vi->config_lock);
>>>> vi->config_enable = false;
>>>
>
>
^ permalink raw reply
* Re: [RFC PATCH v2 net-next 1/2] ppp: remove NETIF_F_NETNS_LOCAL from ppp device features.
From: Eric W. Biederman @ 2013-01-07 9:16 UTC (permalink / raw)
To: Rami Rosen; +Cc: davem, netdev, shemminger
In-Reply-To: <1357543176-15352-2-git-send-email-ramirose@gmail.com>
Rami Rosen <ramirose@gmail.com> writes:
> There is no need for NETIF_F_NETNS_LOCAL for ppp device; this patch
> removes it.
At least pppoe_rcv_core assumes that the ppp socket and the network
devices are all in the same network namespace. So it looks like
NETIF_F_NETNS_LOCAL is needed.
Eric
> Signed-off-by: Rami Rosen <ramirose@gmail.com>
> ---
> drivers/net/ppp/ppp_generic.c | 1 -
> 1 file changed, 1 deletion(-)
>
> diff --git a/drivers/net/ppp/ppp_generic.c b/drivers/net/ppp/ppp_generic.c
> index 0b2706a..89b6fc5 100644
> --- a/drivers/net/ppp/ppp_generic.c
> +++ b/drivers/net/ppp/ppp_generic.c
> @@ -1073,7 +1073,6 @@ static void ppp_setup(struct net_device *dev)
> dev->tx_queue_len = 3;
> dev->type = ARPHRD_PPP;
> dev->flags = IFF_POINTOPOINT | IFF_NOARP | IFF_MULTICAST;
> - dev->features |= NETIF_F_NETNS_LOCAL;
> dev->priv_flags &= ~IFF_XMIT_DST_RELEASE;
> }
^ permalink raw reply
* Re: [RFC PATCH v2 net-next 2/2] bridge: remove NETIF_F_NETNS_LOCAL from bridge device features.
From: Eric W. Biederman @ 2013-01-07 9:20 UTC (permalink / raw)
To: Rami Rosen; +Cc: davem, netdev, shemminger
In-Reply-To: <1357543176-15352-3-git-send-email-ramirose@gmail.com>
Rami Rosen <ramirose@gmail.com> writes:
> There is no need for NETIF_F_NETNS_LOCAL for bridge device; this patch
> removes it.
At the very least add_del_if assumes are network devices are in the same
network namespace as the bridge. So removing NETIF_F_NETNS_LOCAL is not
safe.
Eric
^ permalink raw reply
* Re: [RFC PATCH v2 net-next 0/2] Remove NETIF_F_NETNS_LOCAL from software devices.
From: Eric W. Biederman @ 2013-01-07 9:25 UTC (permalink / raw)
To: Rami Rosen; +Cc: davem, netdev, shemminger
In-Reply-To: <1357543176-15352-1-git-send-email-ramirose@gmail.com>
Rami Rosen <ramirose@gmail.com> writes:
> When NETIF_F_NETNS_LOCAL feature is set in some network device,
> we do not allow this device to be moved from one network namespace
> to another (see dev_change_net_namespace()). Vlan device is not considered
> a local device and does not have the NETIF_F_NETNS_LOCAL feature. There is no
> reason that software devices like ppp and bridge will set the
> NETIF_F_NETNS_LOCAL feature. This patchset removes NETIF_F_NETNS_LOCAL
> from ppp and bridge devices.
After you saw the case for vxlan I am disappointed you respun this
patchset without asking the question: Could other devices suffer
from the same class of problem.
It didn't take me more than a few minutes of looking to see that the
bridge and the ppp network device have the same class of issue as
vxlan.
> * Changes from v1
> Do not remove NETIF_F_NETNS_LOCAL from vxlan devices. Moving vxlan device
> to a different namesapce does not move the state of the UDP socket related
> to this vxlan device (Following Stephen Hemminger comment).
>
> Rami Rosen (2):
> [RFC PATCH v2 net-next 1/3] ppp: remove NETIF_F_NETNS_LOCAL from ppp
> device features.
> [RFC PATCH v2 net-next 2/3] bridge: remove NETIF_F_NETNS_LOCAL from
> bridge device features.
>
> drivers/net/ppp/ppp_generic.c | 1 -
> net/bridge/br_device.c | 2 +-
> 2 files changed, 1 insertion(+), 2 deletions(-)
^ permalink raw reply
* [PATCH repost] net,wireless: check against default_ethtool_ops
From: Stanislaw Gruszka @ 2013-01-07 9:55 UTC (permalink / raw)
To: netdev, David S. Miller
Cc: Eric Dumazet, Ben Greear, Bjørn Mork, linux-wireless,
Ben Hutchings
Since:
commit 2c60db037034d27f8c636403355d52872da92f81
Author: Eric Dumazet <edumazet@google.com>
Date: Sun Sep 16 09:17:26 2012 +0000
net: provide a default dev->ethtool_ops
wireless core does not correctly assign ethtool_ops. In order to fix
the problem, and avoid assigning ethtool_ops on each individual cfg80211
drivers, we check against default_ethool_ops pointer instead of NULL in
wireless core.
Signed-off-by: Stanislaw Gruszka <sgruszka@redhat.com>
Acked-by: Ben Hutchings <bhutchings@solarflare.com>
Cc: stable@vger.kernel.org # 3.7+
---
This should go directly through net tree since patch include core
net specific changes.
include/linux/netdevice.h | 2 ++
net/core/dev.c | 3 ++-
net/wireless/core.c | 2 +-
3 files changed, 5 insertions(+), 2 deletions(-)
diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h
index f8eda02..c98e1c3 100644
--- a/include/linux/netdevice.h
+++ b/include/linux/netdevice.h
@@ -60,6 +60,8 @@ struct wireless_dev;
#define SET_ETHTOOL_OPS(netdev,ops) \
( (netdev)->ethtool_ops = (ops) )
+extern const struct ethtool_ops default_ethtool_ops;
+
/* hardware address assignment types */
#define NET_ADDR_PERM 0 /* address is permanent (default) */
#define NET_ADDR_RANDOM 1 /* address is generated randomly */
diff --git a/net/core/dev.c b/net/core/dev.c
index c0946cb..4cd2168 100644
--- a/net/core/dev.c
+++ b/net/core/dev.c
@@ -6008,7 +6008,8 @@ struct netdev_queue *dev_ingress_queue_create(struct net_device *dev)
return queue;
}
-static const struct ethtool_ops default_ethtool_ops;
+const struct ethtool_ops default_ethtool_ops;
+EXPORT_SYMBOL_GPL(default_ethtool_ops);
/**
* alloc_netdev_mqs - allocate network device
diff --git a/net/wireless/core.c b/net/wireless/core.c
index 14d9904..90915d4 100644
--- a/net/wireless/core.c
+++ b/net/wireless/core.c
@@ -866,7 +866,7 @@ static int cfg80211_netdev_notifier_call(struct notifier_block *nb,
/* allow mac80211 to determine the timeout */
wdev->ps_timeout = -1;
- if (!dev->ethtool_ops)
+ if (dev->ethtool_ops == &default_ethtool_ops)
dev->ethtool_ops = &cfg80211_ethtool_ops;
if ((wdev->iftype == NL80211_IFTYPE_STATION ||
--
1.7.1
^ permalink raw reply related
* Re: [PATCH repost] net,wireless: check against default_ethtool_ops
From: Jiri Pirko @ 2013-01-07 10:23 UTC (permalink / raw)
To: Stanislaw Gruszka
Cc: netdev-u79uwXL29TY76Z2rM5mHXA, David S. Miller, Eric Dumazet,
Ben Greear, Bjørn Mork,
linux-wireless-u79uwXL29TY76Z2rM5mHXA, Ben Hutchings
In-Reply-To: <20130107095548.GA6931-H+wXaHxf7aLQT0dZR+AlfA@public.gmane.org>
Mon, Jan 07, 2013 at 10:55:49AM CET, sgruszka-H+wXaHxf7aLQT0dZR+AlfA@public.gmane.org wrote:
>Since:
>
>commit 2c60db037034d27f8c636403355d52872da92f81
>Author: Eric Dumazet <edumazet-hpIqsD4AKlfQT0dZR+AlfA@public.gmane.org>
>Date: Sun Sep 16 09:17:26 2012 +0000
>
> net: provide a default dev->ethtool_ops
>
>wireless core does not correctly assign ethtool_ops. In order to fix
>the problem, and avoid assigning ethtool_ops on each individual cfg80211
>drivers, we check against default_ethool_ops pointer instead of NULL in
>wireless core.
>
>Signed-off-by: Stanislaw Gruszka <sgruszka-H+wXaHxf7aLQT0dZR+AlfA@public.gmane.org>
>Acked-by: Ben Hutchings <bhutchings-s/n/eUQHGBpZroRs9YW3xA@public.gmane.org>
>Cc: stable-u79uwXL29TY76Z2rM5mHXA@public.gmane.org # 3.7+
>---
>This should go directly through net tree since patch include core
>net specific changes.
>
> include/linux/netdevice.h | 2 ++
> net/core/dev.c | 3 ++-
> net/wireless/core.c | 2 +-
> 3 files changed, 5 insertions(+), 2 deletions(-)
>
>diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h
>index f8eda02..c98e1c3 100644
>--- a/include/linux/netdevice.h
>+++ b/include/linux/netdevice.h
>@@ -60,6 +60,8 @@ struct wireless_dev;
> #define SET_ETHTOOL_OPS(netdev,ops) \
> ( (netdev)->ethtool_ops = (ops) )
>
>+extern const struct ethtool_ops default_ethtool_ops;
>+
> /* hardware address assignment types */
> #define NET_ADDR_PERM 0 /* address is permanent (default) */
> #define NET_ADDR_RANDOM 1 /* address is generated randomly */
>diff --git a/net/core/dev.c b/net/core/dev.c
>index c0946cb..4cd2168 100644
>--- a/net/core/dev.c
>+++ b/net/core/dev.c
>@@ -6008,7 +6008,8 @@ struct netdev_queue *dev_ingress_queue_create(struct net_device *dev)
> return queue;
> }
>
>-static const struct ethtool_ops default_ethtool_ops;
>+const struct ethtool_ops default_ethtool_ops;
>+EXPORT_SYMBOL_GPL(default_ethtool_ops);
I think that default_ethtool_ops should stay static. Wouldn't it be
nicer to introduce a helper like:
bool dev_has_default_ethtool_ops(struct net_device *dev)
{
return dev->ethtool_ops == &default_ethtool_ops;
}
>
> /**
> * alloc_netdev_mqs - allocate network device
>diff --git a/net/wireless/core.c b/net/wireless/core.c
>index 14d9904..90915d4 100644
>--- a/net/wireless/core.c
>+++ b/net/wireless/core.c
>@@ -866,7 +866,7 @@ static int cfg80211_netdev_notifier_call(struct notifier_block *nb,
> /* allow mac80211 to determine the timeout */
> wdev->ps_timeout = -1;
>
>- if (!dev->ethtool_ops)
>+ if (dev->ethtool_ops == &default_ethtool_ops)
> dev->ethtool_ops = &cfg80211_ethtool_ops;
>
> if ((wdev->iftype == NL80211_IFTYPE_STATION ||
>--
>1.7.1
>
>--
>To unsubscribe from this list: send the line "unsubscribe netdev" in
>the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
>More majordomo info at http://vger.kernel.org/majordomo-info.html
--
To unsubscribe from this list: send the line "unsubscribe linux-wireless" in
the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
^ permalink raw reply
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