* [PATCH net-next v2 1/2] net: AF-specific RTM_GETSTATS attributes
From: Robert Shearman @ 2017-01-16 14:16 UTC (permalink / raw)
To: davem; +Cc: netdev, David Ahern, ebiederm, Roopa Prabhu, Robert Shearman
In-Reply-To: <1484576197-19442-1-git-send-email-rshearma@brocade.com>
Add the functionality for including address-family-specific per-link
stats in RTM_GETSTATS messages. This is done through adding a new
IFLA_STATS_AF_SPEC attribute under which address family attributes are
nested and then the AF-specific attributes can be further nested. This
follows the model of IFLA_AF_SPEC on RTM_*LINK messages and it has the
advantage of presenting an easily extended hierarchy. The rtnl_af_ops
structure is extended to provide AFs with the opportunity to fill and
provide the size of their stats attributes.
One alternative would have been to provide AFs with the ability to add
attributes directly into the RTM_GETSTATS message without a nested
hierarchy. I discounted this approach as it increases the rate at
which the 32 attribute number space is used up and it makes
implementation a little more tricky for stats dump resuming (at the
moment the order in which attributes are added to the message has to
match the numeric order of the attributes).
Another alternative would have been to register per-AF RTM_GETSTATS
handlers. I discounted this approach as I perceived a common use-case
to be getting all the stats for an interface and this approach would
necessitate multiple requests/dumps to retrieve them all.
Signed-off-by: Robert Shearman <rshearma@brocade.com>
Acked-by: Roopa Prabhu <roopa@cumulusnetworks.com>
---
include/net/rtnetlink.h | 4 ++++
include/uapi/linux/if_link.h | 1 +
net/core/rtnetlink.c | 50 ++++++++++++++++++++++++++++++++++++++++++++
3 files changed, 55 insertions(+)
diff --git a/include/net/rtnetlink.h b/include/net/rtnetlink.h
index 4113916cc1bb..106de5f7bf06 100644
--- a/include/net/rtnetlink.h
+++ b/include/net/rtnetlink.h
@@ -139,6 +139,10 @@ struct rtnl_af_ops {
const struct nlattr *attr);
int (*set_link_af)(struct net_device *dev,
const struct nlattr *attr);
+
+ int (*fill_stats_af)(struct sk_buff *skb,
+ const struct net_device *dev);
+ size_t (*get_stats_af_size)(const struct net_device *dev);
};
void __rtnl_af_unregister(struct rtnl_af_ops *ops);
diff --git a/include/uapi/linux/if_link.h b/include/uapi/linux/if_link.h
index 6b13e591abc9..184b16ed2b84 100644
--- a/include/uapi/linux/if_link.h
+++ b/include/uapi/linux/if_link.h
@@ -847,6 +847,7 @@ enum {
IFLA_STATS_LINK_XSTATS,
IFLA_STATS_LINK_XSTATS_SLAVE,
IFLA_STATS_LINK_OFFLOAD_XSTATS,
+ IFLA_STATS_AF_SPEC,
__IFLA_STATS_MAX,
};
diff --git a/net/core/rtnetlink.c b/net/core/rtnetlink.c
index 18b5aae99bec..4edc1bd7a735 100644
--- a/net/core/rtnetlink.c
+++ b/net/core/rtnetlink.c
@@ -3829,6 +3829,39 @@ static int rtnl_fill_statsinfo(struct sk_buff *skb, struct net_device *dev,
*idxattr = 0;
}
+ if (stats_attr_valid(filter_mask, IFLA_STATS_AF_SPEC, *idxattr)) {
+ struct rtnl_af_ops *af_ops;
+
+ *idxattr = IFLA_STATS_AF_SPEC;
+ attr = nla_nest_start(skb, IFLA_STATS_AF_SPEC);
+ if (!attr)
+ goto nla_put_failure;
+
+ list_for_each_entry(af_ops, &rtnl_af_ops, list) {
+ if (af_ops->fill_stats_af) {
+ struct nlattr *af;
+ int err;
+
+ af = nla_nest_start(skb, af_ops->family);
+ if (!af)
+ goto nla_put_failure;
+
+ err = af_ops->fill_stats_af(skb, dev);
+
+ if (err == -ENODATA)
+ nla_nest_cancel(skb, af);
+ else if (err < 0)
+ goto nla_put_failure;
+
+ nla_nest_end(skb, af);
+ }
+ }
+
+ nla_nest_end(skb, attr);
+
+ *idxattr = 0;
+ }
+
nlmsg_end(skb, nlh);
return 0;
@@ -3885,6 +3918,23 @@ static size_t if_nlmsg_stats_size(const struct net_device *dev,
if (stats_attr_valid(filter_mask, IFLA_STATS_LINK_OFFLOAD_XSTATS, 0))
size += rtnl_get_offload_stats_size(dev);
+ if (stats_attr_valid(filter_mask, IFLA_STATS_AF_SPEC, 0)) {
+ struct rtnl_af_ops *af_ops;
+
+ /* for IFLA_STATS_AF_SPEC */
+ size += nla_total_size(0);
+
+ list_for_each_entry(af_ops, &rtnl_af_ops, list) {
+ if (af_ops->get_stats_af_size) {
+ size += nla_total_size(
+ af_ops->get_stats_af_size(dev));
+
+ /* for AF_* */
+ size += nla_total_size(0);
+ }
+ }
+ }
+
return size;
}
--
2.1.4
^ permalink raw reply related
* [PATCH net-next v2 2/2] mpls: Packet stats
From: Robert Shearman @ 2017-01-16 14:16 UTC (permalink / raw)
To: davem; +Cc: netdev, David Ahern, ebiederm, Roopa Prabhu, Robert Shearman
In-Reply-To: <1484576197-19442-1-git-send-email-rshearma@brocade.com>
Having MPLS packet stats is useful for observing network operation and
for diagnosing network problems. In the absence of anything better,
RFC2863 and RFC3813 are used for guidance for which stats to expose
and the semantics of them. In particular rx_noroutes maps to in
unknown protos in RFC2863. The stats are exposed to userspace via
AF_MPLS attributes embedded in the IFLA_STATS_AF_SPEC attribute of
RTM_GETSTATS messages.
All the introduced fields are 64-bit, even error ones, to ensure no
overflow with long uptimes. Per-CPU counters are used to avoid
cache-line contention on the commonly used fields. The other fields
have also been made per-CPU for code to avoid performance problems in
error conditions on the assumption that on some platforms the cost of
atomic operations could be more expensive than sending the packet
(which is what would be done in the success case). If that's not the
case, we could instead not use per-CPU counters for these fields.
Only unicast and non-fragment are exposed at the moment, but other
counters can be exposed in the future either by adding to the end of
struct mpls_link_stats or by additional netlink attributes in the
AF_MPLS IFLA_STATS_AF_SPEC nested attribute.
Signed-off-by: Robert Shearman <rshearma@brocade.com>
---
include/uapi/linux/mpls.h | 30 ++++++++
net/mpls/af_mpls.c | 181 ++++++++++++++++++++++++++++++++++++++++------
net/mpls/internal.h | 58 ++++++++++++++-
net/mpls/mpls_iptunnel.c | 11 ++-
4 files changed, 252 insertions(+), 28 deletions(-)
diff --git a/include/uapi/linux/mpls.h b/include/uapi/linux/mpls.h
index 24a6cb1aec86..77a19dfe3990 100644
--- a/include/uapi/linux/mpls.h
+++ b/include/uapi/linux/mpls.h
@@ -43,4 +43,34 @@ struct mpls_label {
#define MPLS_LABEL_FIRST_UNRESERVED 16 /* RFC3032 */
+/* These are embedded into IFLA_STATS_AF_SPEC:
+ * [IFLA_STATS_AF_SPEC]
+ * -> [AF_MPLS]
+ * -> [MPLS_STATS_xxx]
+ *
+ * Attributes:
+ * [MPLS_STATS_LINK] = {
+ * struct mpls_link_stats
+ * }
+ */
+enum {
+ MPLS_STATS_UNSPEC, /* also used as 64bit pad attribute */
+ MPLS_STATS_LINK,
+ __MPLS_STATS_MAX,
+};
+
+#define MPLS_STATS_MAX (__MPLS_STATS_MAX - 1)
+
+struct mpls_link_stats {
+ __u64 rx_packets; /* total packets received */
+ __u64 tx_packets; /* total packets transmitted */
+ __u64 rx_bytes; /* total bytes received */
+ __u64 tx_bytes; /* total bytes transmitted */
+ __u64 rx_errors; /* bad packets received */
+ __u64 tx_errors; /* packet transmit problems */
+ __u64 rx_dropped; /* packet dropped on receive */
+ __u64 tx_dropped; /* packet dropped on transmit */
+ __u64 rx_noroute; /* no route for packet dest */
+};
+
#endif /* _UAPI_MPLS_H */
diff --git a/net/mpls/af_mpls.c b/net/mpls/af_mpls.c
index 15fe97644ffe..4dc81963af8f 100644
--- a/net/mpls/af_mpls.c
+++ b/net/mpls/af_mpls.c
@@ -8,6 +8,7 @@
#include <linux/ipv6.h>
#include <linux/mpls.h>
#include <linux/vmalloc.h>
+#include <linux/percpu.h>
#include <net/ip.h>
#include <net/dst.h>
#include <net/sock.h>
@@ -17,8 +18,8 @@
#include <net/netns/generic.h>
#if IS_ENABLED(CONFIG_IPV6)
#include <net/ipv6.h>
-#include <net/addrconf.h>
#endif
+#include <net/addrconf.h>
#include <net/nexthop.h>
#include "internal.h"
@@ -48,11 +49,6 @@ static struct mpls_route *mpls_route_input_rcu(struct net *net, unsigned index)
return rt;
}
-static inline struct mpls_dev *mpls_dev_get(const struct net_device *dev)
-{
- return rcu_dereference_rtnl(dev->mpls_ptr);
-}
-
bool mpls_output_possible(const struct net_device *dev)
{
return dev && (dev->flags & IFF_UP) && netif_carrier_ok(dev);
@@ -98,6 +94,31 @@ bool mpls_pkt_too_big(const struct sk_buff *skb, unsigned int mtu)
}
EXPORT_SYMBOL_GPL(mpls_pkt_too_big);
+void mpls_stats_inc_outucastpkts(struct net_device *dev,
+ const struct sk_buff *skb)
+{
+ struct mpls_dev *mdev;
+
+ if (skb->protocol == htons(ETH_P_MPLS_UC)) {
+ mdev = mpls_dev_get(dev);
+ if (mdev)
+ MPLS_INC_STATS_LEN(mdev, skb->len,
+ tx_packets,
+ tx_bytes);
+ } else if (skb->protocol == htons(ETH_P_IP)) {
+ IP_UPD_PO_STATS(dev_net(dev), IPSTATS_MIB_OUT, skb->len);
+#if IS_ENABLED(CONFIG_IPV6)
+ } else if (skb->protocol == htons(ETH_P_IPV6)) {
+ struct inet6_dev *in6dev = __in6_dev_get(dev);
+
+ if (in6dev)
+ IP6_UPD_PO_STATS(dev_net(dev), in6dev,
+ IPSTATS_MIB_OUT, skb->len);
+#endif
+ }
+}
+EXPORT_SYMBOL_GPL(mpls_stats_inc_outucastpkts);
+
static u32 mpls_multipath_hash(struct mpls_route *rt,
struct sk_buff *skb, bool bos)
{
@@ -253,6 +274,7 @@ static int mpls_forward(struct sk_buff *skb, struct net_device *dev,
struct mpls_nh *nh;
struct mpls_entry_decoded dec;
struct net_device *out_dev;
+ struct mpls_dev *out_mdev;
struct mpls_dev *mdev;
unsigned int hh_len;
unsigned int new_header_size;
@@ -262,17 +284,25 @@ static int mpls_forward(struct sk_buff *skb, struct net_device *dev,
/* Careful this entire function runs inside of an rcu critical section */
mdev = mpls_dev_get(dev);
- if (!mdev || !mdev->input_enabled)
+ if (!mdev)
goto drop;
- if (skb->pkt_type != PACKET_HOST)
+ MPLS_INC_STATS_LEN(mdev, skb->len, rx_packets,
+ rx_bytes);
+
+ if (!mdev->input_enabled) {
+ MPLS_INC_STATS(mdev, rx_dropped);
goto drop;
+ }
+
+ if (skb->pkt_type != PACKET_HOST)
+ goto err;
if ((skb = skb_share_check(skb, GFP_ATOMIC)) == NULL)
- goto drop;
+ goto err;
if (!pskb_may_pull(skb, sizeof(*hdr)))
- goto drop;
+ goto err;
/* Read and decode the label */
hdr = mpls_hdr(skb);
@@ -285,33 +315,35 @@ static int mpls_forward(struct sk_buff *skb, struct net_device *dev,
skb_orphan(skb);
rt = mpls_route_input_rcu(net, dec.label);
- if (!rt)
+ if (!rt) {
+ MPLS_INC_STATS(mdev, rx_noroute);
goto drop;
+ }
nh = mpls_select_multipath(rt, skb, dec.bos);
if (!nh)
- goto drop;
-
- /* Find the output device */
- out_dev = rcu_dereference(nh->nh_dev);
- if (!mpls_output_possible(out_dev))
- goto drop;
+ goto err;
if (skb_warn_if_lro(skb))
- goto drop;
+ goto err;
skb_forward_csum(skb);
/* Verify ttl is valid */
if (dec.ttl <= 1)
- goto drop;
+ goto err;
dec.ttl -= 1;
+ /* Find the output device */
+ out_dev = rcu_dereference(nh->nh_dev);
+ if (!mpls_output_possible(out_dev))
+ goto tx_err;
+
/* Verify the destination can hold the packet */
new_header_size = mpls_nh_header_size(nh);
mtu = mpls_dev_mtu(out_dev);
if (mpls_pkt_too_big(skb, mtu - new_header_size))
- goto drop;
+ goto tx_err;
hh_len = LL_RESERVED_SPACE(out_dev);
if (!out_dev->header_ops)
@@ -319,7 +351,7 @@ static int mpls_forward(struct sk_buff *skb, struct net_device *dev,
/* Ensure there is enough space for the headers in the skb */
if (skb_cow(skb, hh_len + new_header_size))
- goto drop;
+ goto tx_err;
skb->dev = out_dev;
skb->protocol = htons(ETH_P_MPLS_UC);
@@ -327,7 +359,7 @@ static int mpls_forward(struct sk_buff *skb, struct net_device *dev,
if (unlikely(!new_header_size && dec.bos)) {
/* Penultimate hop popping */
if (!mpls_egress(rt, skb, dec))
- goto drop;
+ goto err;
} else {
bool bos;
int i;
@@ -343,6 +375,8 @@ static int mpls_forward(struct sk_buff *skb, struct net_device *dev,
}
}
+ mpls_stats_inc_outucastpkts(out_dev, skb);
+
/* If via wasn't specified then send out using device address */
if (nh->nh_via_table == MPLS_NEIGH_TABLE_UNSPEC)
err = neigh_xmit(NEIGH_LINK_TABLE, out_dev,
@@ -355,6 +389,13 @@ static int mpls_forward(struct sk_buff *skb, struct net_device *dev,
__func__, err);
return 0;
+tx_err:
+ out_mdev = out_dev ? mpls_dev_get(out_dev) : NULL;
+ if (out_mdev)
+ MPLS_INC_STATS(out_mdev, tx_errors);
+ goto drop;
+err:
+ MPLS_INC_STATS(mdev, rx_errors);
drop:
kfree_skb(skb);
return NET_RX_DROP;
@@ -853,6 +894,70 @@ static int mpls_route_del(struct mpls_route_config *cfg)
return err;
}
+static void mpls_get_stats(struct mpls_dev *mdev,
+ struct mpls_link_stats *stats)
+{
+ struct mpls_pcpu_stats *p;
+ int i;
+
+ memset(stats, 0, sizeof(*stats));
+
+ for_each_possible_cpu(i) {
+ struct mpls_link_stats local;
+ unsigned int start;
+
+ p = per_cpu_ptr(mdev->stats, i);
+ do {
+ start = u64_stats_fetch_begin(&p->syncp);
+ local = p->stats;
+ } while (u64_stats_fetch_retry(&p->syncp, start));
+
+ stats->rx_packets += local.rx_packets;
+ stats->rx_bytes += local.rx_bytes;
+ stats->tx_packets += local.tx_packets;
+ stats->tx_bytes += local.tx_bytes;
+ stats->rx_errors += local.rx_errors;
+ stats->tx_errors += local.tx_errors;
+ stats->rx_dropped += local.rx_dropped;
+ stats->tx_dropped += local.tx_dropped;
+ stats->rx_noroute += local.rx_noroute;
+ }
+}
+
+static int mpls_fill_stats_af(struct sk_buff *skb,
+ const struct net_device *dev)
+{
+ struct mpls_link_stats *stats;
+ struct mpls_dev *mdev;
+ struct nlattr *nla;
+
+ mdev = mpls_dev_get(dev);
+ if (!mdev)
+ return -ENODATA;
+
+ nla = nla_reserve_64bit(skb, MPLS_STATS_LINK,
+ sizeof(struct mpls_link_stats),
+ MPLS_STATS_UNSPEC);
+ if (!nla)
+ return -EMSGSIZE;
+
+ stats = nla_data(nla);
+ mpls_get_stats(mdev, stats);
+
+ return 0;
+}
+
+static size_t mpls_get_stats_af_size(const struct net_device *dev)
+{
+ struct mpls_dev *mdev;
+
+ mdev = mpls_dev_get(dev);
+ if (!mdev)
+ return 0;
+
+ return nla_total_size_64bit(sizeof(struct mpls_link_stats));
+}
+
#define MPLS_PERDEV_SYSCTL_OFFSET(field) \
(&((struct mpls_dev *)0)->field)
@@ -911,6 +1016,7 @@ static struct mpls_dev *mpls_add_dev(struct net_device *dev)
{
struct mpls_dev *mdev;
int err = -ENOMEM;
+ int i;
ASSERT_RTNL();
@@ -918,6 +1024,17 @@ static struct mpls_dev *mpls_add_dev(struct net_device *dev)
if (!mdev)
return ERR_PTR(err);
+ mdev->stats = alloc_percpu(struct mpls_pcpu_stats);
+ if (!mdev->stats)
+ goto free;
+
+ for_each_possible_cpu(i) {
+ struct mpls_pcpu_stats *mpls_stats;
+
+ mpls_stats = per_cpu_ptr(mdev->stats, i);
+ u64_stats_init(&mpls_stats->syncp);
+ }
+
err = mpls_dev_sysctl_register(dev, mdev);
if (err)
goto free;
@@ -927,10 +1044,19 @@ static struct mpls_dev *mpls_add_dev(struct net_device *dev)
return mdev;
free:
+ free_percpu(mdev->stats);
kfree(mdev);
return ERR_PTR(err);
}
+static void mpls_dev_destroy_rcu(struct rcu_head *head)
+{
+ struct mpls_dev *mdev = container_of(head, struct mpls_dev, rcu);
+
+ free_percpu(mdev->stats);
+ kfree(mdev);
+}
+
static void mpls_ifdown(struct net_device *dev, int event)
{
struct mpls_route __rcu **platform_label;
@@ -1045,7 +1171,7 @@ static int mpls_dev_notify(struct notifier_block *this, unsigned long event,
if (mdev) {
mpls_dev_sysctl_unregister(mdev);
RCU_INIT_POINTER(dev->mpls_ptr, NULL);
- kfree_rcu(mdev, rcu);
+ call_rcu(&mdev->rcu, mpls_dev_destroy_rcu);
}
break;
case NETDEV_CHANGENAME:
@@ -1706,6 +1832,12 @@ static struct pernet_operations mpls_net_ops = {
.exit = mpls_net_exit,
};
+static struct rtnl_af_ops mpls_af_ops __read_mostly = {
+ .family = AF_MPLS,
+ .fill_stats_af = mpls_fill_stats_af,
+ .get_stats_af_size = mpls_get_stats_af_size,
+};
+
static int __init mpls_init(void)
{
int err;
@@ -1722,6 +1854,8 @@ static int __init mpls_init(void)
dev_add_pack(&mpls_packet_type);
+ rtnl_af_register(&mpls_af_ops);
+
rtnl_register(PF_MPLS, RTM_NEWROUTE, mpls_rtm_newroute, NULL, NULL);
rtnl_register(PF_MPLS, RTM_DELROUTE, mpls_rtm_delroute, NULL, NULL);
rtnl_register(PF_MPLS, RTM_GETROUTE, NULL, mpls_dump_routes, NULL);
@@ -1738,6 +1872,7 @@ module_init(mpls_init);
static void __exit mpls_exit(void)
{
rtnl_unregister_all(PF_MPLS);
+ rtnl_af_unregister(&mpls_af_ops);
dev_remove_pack(&mpls_packet_type);
unregister_netdevice_notifier(&mpls_dev_notifier);
unregister_pernet_subsys(&mpls_net_ops);
diff --git a/net/mpls/internal.h b/net/mpls/internal.h
index bdfef6c3271a..d97243034605 100644
--- a/net/mpls/internal.h
+++ b/net/mpls/internal.h
@@ -9,13 +9,58 @@ struct mpls_entry_decoded {
u8 bos;
};
+struct mpls_pcpu_stats {
+ struct mpls_link_stats stats;
+ struct u64_stats_sync syncp;
+};
+
struct mpls_dev {
- int input_enabled;
+ int input_enabled;
- struct ctl_table_header *sysctl;
- struct rcu_head rcu;
+ struct mpls_pcpu_stats __percpu *stats;
+
+ struct ctl_table_header *sysctl;
+ struct rcu_head rcu;
};
+#if BITS_PER_LONG == 32
+
+#define MPLS_INC_STATS_LEN(mdev, len, pkts_field, bytes_field) \
+ do { \
+ __typeof__(*(mdev)->stats) *ptr = \
+ raw_cpu_ptr((mdev)->stats); \
+ local_bh_disable(); \
+ u64_stats_update_begin(&ptr->syncp); \
+ ptr->stats.pkts_field++; \
+ ptr->stats.bytes_field += (len); \
+ u64_stats_update_end(&ptr->syncp); \
+ local_bh_enable(); \
+ } while (0)
+
+#define MPLS_INC_STATS(mdev, field) \
+ do { \
+ __typeof__(*(mdev)->stats) *ptr = \
+ raw_cpu_ptr((mdev)->stats); \
+ local_bh_disable(); \
+ u64_stats_update_begin(&ptr->syncp); \
+ ptr->stats.field++; \
+ u64_stats_update_end(&ptr->syncp); \
+ local_bh_enable(); \
+ } while (0)
+
+#else
+
+#define MPLS_INC_STATS_LEN(mdev, len, pkts_field, bytes_field) \
+ do { \
+ this_cpu_inc((mdev)->stats->stats.pkts_field); \
+ this_cpu_add((mdev)->stats->stats.bytes_field, (len)); \
+ } while (0)
+
+#define MPLS_INC_STATS(mdev, field) \
+ this_cpu_inc((mdev)->stats->stats.field)
+
+#endif
+
struct sk_buff;
#define LABEL_NOT_SPECIFIED (1 << 20)
@@ -114,6 +159,11 @@ static inline struct mpls_entry_decoded mpls_entry_decode(struct mpls_shim_hdr *
return result;
}
+static inline struct mpls_dev *mpls_dev_get(const struct net_device *dev)
+{
+ return rcu_dereference_rtnl(dev->mpls_ptr);
+}
+
int nla_put_labels(struct sk_buff *skb, int attrtype, u8 labels,
const u32 label[]);
int nla_get_labels(const struct nlattr *nla, u32 max_labels, u8 *labels,
@@ -123,5 +173,7 @@ int nla_get_via(const struct nlattr *nla, u8 *via_alen, u8 *via_table,
bool mpls_output_possible(const struct net_device *dev);
unsigned int mpls_dev_mtu(const struct net_device *dev);
bool mpls_pkt_too_big(const struct sk_buff *skb, unsigned int mtu);
+void mpls_stats_inc_outucastpkts(struct net_device *dev,
+ const struct sk_buff *skb);
#endif /* MPLS_INTERNAL_H */
diff --git a/net/mpls/mpls_iptunnel.c b/net/mpls/mpls_iptunnel.c
index 2f7ccd934416..02531284bc49 100644
--- a/net/mpls/mpls_iptunnel.c
+++ b/net/mpls/mpls_iptunnel.c
@@ -48,11 +48,15 @@ static int mpls_xmit(struct sk_buff *skb)
struct dst_entry *dst = skb_dst(skb);
struct rtable *rt = NULL;
struct rt6_info *rt6 = NULL;
+ struct mpls_dev *out_mdev;
int err = 0;
bool bos;
int i;
unsigned int ttl;
+ /* Find the output device */
+ out_dev = dst->dev;
+
/* Obtain the ttl */
if (dst->ops->family == AF_INET) {
ttl = ip_hdr(skb)->ttl;
@@ -66,8 +70,6 @@ static int mpls_xmit(struct sk_buff *skb)
skb_orphan(skb);
- /* Find the output device */
- out_dev = dst->dev;
if (!mpls_output_possible(out_dev) ||
!dst->lwtstate || skb_warn_if_lro(skb))
goto drop;
@@ -109,6 +111,8 @@ static int mpls_xmit(struct sk_buff *skb)
bos = false;
}
+ mpls_stats_inc_outucastpkts(out_dev, skb);
+
if (rt)
err = neigh_xmit(NEIGH_ARP_TABLE, out_dev, &rt->rt_gateway,
skb);
@@ -122,6 +126,9 @@ static int mpls_xmit(struct sk_buff *skb)
return LWTUNNEL_XMIT_DONE;
drop:
+ out_mdev = out_dev ? mpls_dev_get(out_dev) : NULL;
+ if (out_mdev)
+ MPLS_INC_STATS(out_mdev, tx_errors);
kfree_skb(skb);
return -EINVAL;
}
--
2.1.4
^ permalink raw reply related
* Re: [PATCH net-next] sctp: remove unused var from sctp_process_asconf
From: Xin Long @ 2017-01-16 14:40 UTC (permalink / raw)
To: Marcelo Ricardo Leitner
Cc: network dev, linux-sctp, Vlad Yasevich, Neil Horman
In-Reply-To: <5da2cda90a7489b22ae98703d7543c5316bae375.1484336552.git.marcelo.leitner@gmail.com>
On Sat, Jan 14, 2017 at 4:27 AM, Marcelo Ricardo Leitner
<marcelo.leitner@gmail.com> wrote:
> Assigned but not used.
>
> Signed-off-by: Marcelo Ricardo Leitner <marcelo.leitner@gmail.com>
Reviewed-by: Xin Long <lucien.xin@gmail.com>
> ---
> net/sctp/sm_make_chunk.c | 2 --
> 1 file changed, 2 deletions(-)
>
> diff --git a/net/sctp/sm_make_chunk.c b/net/sctp/sm_make_chunk.c
> index a15d824a313d310ed03ba77055d22b1c7c9d0662..80a9088084ac0d1116bf983fad9a0cdfddeefe44 100644
> --- a/net/sctp/sm_make_chunk.c
> +++ b/net/sctp/sm_make_chunk.c
> @@ -3210,7 +3210,6 @@ struct sctp_chunk *sctp_process_asconf(struct sctp_association *asoc,
> union sctp_params param;
> sctp_addiphdr_t *hdr;
> union sctp_addr_param *addr_param;
> - sctp_addip_param_t *asconf_param;
> struct sctp_chunk *asconf_ack;
> __be16 err_code;
> int length = 0;
> @@ -3230,7 +3229,6 @@ struct sctp_chunk *sctp_process_asconf(struct sctp_association *asoc,
> * asconf parameter.
> */
> length = ntohs(addr_param->p.length);
> - asconf_param = (void *)addr_param + length;
> chunk_len -= length;
>
> /* create an ASCONF_ACK chunk.
> --
> 2.9.3
>
> --
> To unsubscribe from this list: send the line "unsubscribe linux-sctp" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at http://vger.kernel.org/majordomo-info.html
^ permalink raw reply
* RE: sctp: kernel memory overwrite attempt detected in sctp_getsockopt_assoc_stats
From: David Laight @ 2017-01-16 14:50 UTC (permalink / raw)
To: 'Dmitry Vyukov', Neil Horman
Cc: Vladislav Yasevich, David Miller, linux-sctp@vger.kernel.org,
netdev, LKML, Kees Cook, syzkaller
In-Reply-To: <CACT4Y+aoRnZyPYM=Bz3RLWhK+MYrQJBZ22TijJd=fzHp9KG+9A@mail.gmail.com>
From: Dmitry Vyukov
> Sent: 16 January 2017 14:04
> >> >> I've enabled CONFIG_HARDENED_USERCOPY_PAGESPAN on syzkaller fuzzer and
...
> >> The code also takes into account compound pages. As far as I
> >> understand the intention of the check is to effectively find
> >> out-of-bounds copies (e.g. goes beyond the current heap allocation). I
> >> would expect that stacks are allocated as compound pages and don't
> >> trigger this check. I don't see it is firing in other similar places.
> >>
> > Honestly, I'm not overly familiar with stack page allocation, at least not so
> > far as compound vs. single page allocation is concerned. I suppose the question
> > your really asking here is: Have you found a case in which the syscall fuzzer
> > has forced the allocation of an insecure non-compound page on the stack, or is
> > this a false positive warning. I can't provide the answer to that.
>
> Yes. I added Kees, author of CONFIG_HARDENED_USERCOPY_PAGESPAN, to To line.
> Kees, is this a false positive?
I'd guess that the kernel stack is (somehow) allocated page by page
rather than by a single multi-page allocate.
Or maybe vmalloc() isn't setting the required flag??
David
^ permalink raw reply
* [PATCH v2] net/irda: fix lockdep annotation
From: Dmitry Vyukov @ 2017-01-16 14:51 UTC (permalink / raw)
To: davej, samuel, davem; +Cc: glider, andreyknvl, Dmitry Vyukov, netdev
The current annotation uses a global variable as recursion counter.
The variable is not atomic nor protected with a mutex, but mutated
by multiple threads. This causes lockdep bug reports episodically:
BUG: looking up invalid subclass: 4294967295
...
_raw_spin_lock_irqsave_nested+0x120/0x180
hashbin_delete+0x4fe/0x750
__irias_delete_object+0xab/0x170
irias_delete_object+0x5f/0xc0
ircomm_tty_detach_cable+0x1d5/0x3f0
...
Make the hashbin_lock_depth variable atomic to prevent bug reports.
As is this causes "unused variable 'depth'" warning without LOCKDEP.
So also change raw_spin_lock_irqsave_nested() macro to not cause
the warning without LOCKDEP. Similar to what raw_spin_lock_nested()
already does.
Signed-off-by: Dmitry Vyukov <dvyukov@google.com>
Cc: Dave Jones <davej@redhat.com>
Cc: Samuel Ortiz <samuel@sortiz.org>
Cc: David S. Miller <davem@davemloft.net>
Cc: netdev@vger.kernel.org
Fixes: c7630a4b932af ("[IrDA]: irda lockdep annotation")
---
Changes since v1:
- Added raw_spin_lock_irqsave_nested() change
as 0-DAY bot reported compiler warning without LOCKDEP.
---
include/linux/spinlock.h | 1 +
net/irda/irqueue.c | 12 +++++++-----
2 files changed, 8 insertions(+), 5 deletions(-)
diff --git a/include/linux/spinlock.h b/include/linux/spinlock.h
index 47dd0cebd204..27aca8c1b129 100644
--- a/include/linux/spinlock.h
+++ b/include/linux/spinlock.h
@@ -217,6 +217,7 @@ static inline void do_raw_spin_unlock(raw_spinlock_t *lock) __releases(lock)
#else
#define raw_spin_lock_irqsave_nested(lock, flags, subclass) \
do { \
+ (void)subclass; \
typecheck(unsigned long, flags); \
flags = _raw_spin_lock_irqsave(lock); \
} while (0)
diff --git a/net/irda/irqueue.c b/net/irda/irqueue.c
index acbe61c7e683..b9fd74e6ca99 100644
--- a/net/irda/irqueue.c
+++ b/net/irda/irqueue.c
@@ -384,21 +384,23 @@ EXPORT_SYMBOL(hashbin_new);
* just supply kfree, which should take care of the job.
*/
#ifdef CONFIG_LOCKDEP
-static int hashbin_lock_depth = 0;
+static atomic_t hashbin_lock_depth = ATOMIC_INIT(0);
#endif
int hashbin_delete( hashbin_t* hashbin, FREE_FUNC free_func)
{
irda_queue_t* queue;
unsigned long flags = 0;
- int i;
+ int i, depth = 0;
IRDA_ASSERT(hashbin != NULL, return -1;);
IRDA_ASSERT(hashbin->magic == HB_MAGIC, return -1;);
/* Synchronize */
if ( hashbin->hb_type & HB_LOCK ) {
- spin_lock_irqsave_nested(&hashbin->hb_spinlock, flags,
- hashbin_lock_depth++);
+#ifdef CONFIG_LOCKDEP
+ depth = atomic_inc_return(&hashbin_lock_depth) - 1;
+#endif
+ spin_lock_irqsave_nested(&hashbin->hb_spinlock, flags, depth);
}
/*
@@ -423,7 +425,7 @@ int hashbin_delete( hashbin_t* hashbin, FREE_FUNC free_func)
if ( hashbin->hb_type & HB_LOCK) {
spin_unlock_irqrestore(&hashbin->hb_spinlock, flags);
#ifdef CONFIG_LOCKDEP
- hashbin_lock_depth--;
+ atomic_dec(&hashbin_lock_depth);
#endif
}
--
2.11.0.483.g087da7b7c-goog
^ permalink raw reply related
* Re: [PATCH] net/irda: fix lockdep annotation
From: Dmitry Vyukov @ 2017-01-16 14:52 UTC (permalink / raw)
To: kbuild test robot
Cc: kbuild-all, Dave Jones, Samuel Ortiz, David Miller,
Alexander Potapenko, andreyknvl, netdev
In-Reply-To: <201701162214.BqECKe5z%fengguang.wu@intel.com>
On Mon, Jan 16, 2017 at 3:11 PM, kbuild test robot <lkp@intel.com> wrote:
> Hi Dmitry,
>
> [auto build test WARNING on net-next/master]
> [also build test WARNING on v4.10-rc4 next-20170116]
> [if your patch is applied to the wrong git tree, please drop us a note to help improve the system]
>
> url: https://github.com/0day-ci/linux/commits/Dmitry-Vyukov/net-irda-fix-lockdep-annotation/20170116-183405
> config: x86_64-randconfig-s2-01161850 (attached as .config)
> compiler: gcc-4.4 (Debian 4.4.7-8) 4.4.7
> reproduce:
> # save the attached .config to linux build tree
> make ARCH=x86_64
>
> All warnings (new ones prefixed by >>):
>
> net/irda/irqueue.c: In function 'hashbin_delete':
>>> net/irda/irqueue.c:393: warning: unused variable 'depth'
Mailed v2 with a fix for the warning.
Thanks, 0-DAY!
> At top level:
> cc1: warning: unrecognized command line option "-Wno-maybe-uninitialized"
>
> vim +/depth +393 net/irda/irqueue.c
>
> 377
> 378
> 379 /*
> 380 * Function hashbin_delete (hashbin, free_func)
> 381 *
> 382 * Destroy hashbin, the free_func can be a user supplied special routine
> 383 * for deallocating this structure if it's complex. If not the user can
> 384 * just supply kfree, which should take care of the job.
> 385 */
> 386 #ifdef CONFIG_LOCKDEP
> 387 static atomic_t hashbin_lock_depth = ATOMIC_INIT(0);
> 388 #endif
> 389 int hashbin_delete( hashbin_t* hashbin, FREE_FUNC free_func)
> 390 {
> 391 irda_queue_t* queue;
> 392 unsigned long flags = 0;
> > 393 int i, depth = 0;
> 394
> 395 IRDA_ASSERT(hashbin != NULL, return -1;);
> 396 IRDA_ASSERT(hashbin->magic == HB_MAGIC, return -1;);
> 397
> 398 /* Synchronize */
> 399 if ( hashbin->hb_type & HB_LOCK ) {
> 400 #ifdef CONFIG_LOCKDEP
> 401 depth = atomic_inc_return(&hashbin_lock_depth) - 1;
>
> ---
> 0-DAY kernel test infrastructure Open Source Technology Center
> https://lists.01.org/pipermail/kbuild-all Intel Corporation
^ permalink raw reply
* Re: sctp: kernel memory overwrite attempt detected in sctp_getsockopt_assoc_stats
From: Dmitry Vyukov @ 2017-01-16 14:56 UTC (permalink / raw)
To: David Laight
Cc: Neil Horman, Vladislav Yasevich, David Miller,
linux-sctp@vger.kernel.org, netdev, LKML, Kees Cook, syzkaller
In-Reply-To: <063D6719AE5E284EB5DD2968C1650D6DB0264AD2@AcuExch.aculab.com>
On Mon, Jan 16, 2017 at 3:50 PM, David Laight <David.Laight@aculab.com> wrote:
> From: Dmitry Vyukov
>> Sent: 16 January 2017 14:04
>> >> >> I've enabled CONFIG_HARDENED_USERCOPY_PAGESPAN on syzkaller fuzzer and
> ...
>> >> The code also takes into account compound pages. As far as I
>> >> understand the intention of the check is to effectively find
>> >> out-of-bounds copies (e.g. goes beyond the current heap allocation). I
>> >> would expect that stacks are allocated as compound pages and don't
>> >> trigger this check. I don't see it is firing in other similar places.
>> >>
>> > Honestly, I'm not overly familiar with stack page allocation, at least not so
>> > far as compound vs. single page allocation is concerned. I suppose the question
>> > your really asking here is: Have you found a case in which the syscall fuzzer
>> > has forced the allocation of an insecure non-compound page on the stack, or is
>> > this a false positive warning. I can't provide the answer to that.
>>
>> Yes. I added Kees, author of CONFIG_HARDENED_USERCOPY_PAGESPAN, to To line.
>> Kees, is this a false positive?
>
> I'd guess that the kernel stack is (somehow) allocated page by page
> rather than by a single multi-page allocate.
> Or maybe vmalloc() isn't setting the required flag??
Just in case, I don't have CONFIG_VMAP_STACK selected.
If it is a generic issue, then CONFIG_HARDENED_USERCOPY_PAGESPAN looks
considerably broken as there are tons of copies onto stack. I don't
see what's special in this particular case.
^ permalink raw reply
* Re: 4.9.0-rc8: tg3 dead after resume
From: Billy Shuman @ 2017-01-16 14:59 UTC (permalink / raw)
To: Siva Reddy Kallam; +Cc: Michael Chan, Netdev
In-Reply-To: <9ac07689abefd319a2baae2860fd8247@mail.gmail.com>
On Mon, Jan 16, 2017 at 2:15 AM, Siva Reddy Kallam
<siva.kallam@broadcom.com> wrote:
> Hi,
>
> Sorry for the late reply.
> We don't have exact model.
> We tried to reproduce with HP elitebook folio laptop & HP elite thunderbolt
> 3 dock.
> We observed issue with Thunder bolt bridge. All the connected devices on the
> TH3 dock don't function at all.
> Are you observing same issue? Did you get a chance to check with HP on this?
No the only device that does not function is the nic and only after resume, it
functions prior to that. I have a usb keyboard and three monitors
attached to the
dock that function properly before and after resume. Did you turn off
authentication for thunderbolt in bios?
> As per our discussion with HP, Thunder bolt bridge driver is provided for
> windows OS only.
>
> Regards,
> Siva
> ----Original Message-----
> From: Billy Shuman [mailto:wshuman3@gmail.com]
> Sent: Wednesday, December 28, 2016 7:33 PM
> To: Siva Reddy Kallam <siva.kallam@broadcom.com>
> Subject: Re: 4.9.0-rc8: tg3 dead after resume
>
> On Mon, Dec 26, 2016 at 1:27 AM, Siva Reddy Kallam
> <siva.kallam@broadcom.com> wrote:
>> On Mon, Dec 12, 2016 at 3:53 PM, Siva Reddy Kallam
>> <siva.kallam@broadcom.com> wrote:
>>> On Fri, Dec 9, 2016 at 7:59 PM, Billy Shuman <wshuman3@gmail.com> wrote:
>>>> On Thu, Dec 8, 2016 at 4:03 AM, Siva Reddy Kallam
>>>> <siva.kallam@broadcom.com> wrote:
>>>>> On Thu, Dec 8, 2016 at 12:14 AM, Billy Shuman <wshuman3@gmail.com>
>>>>> wrote:
>>>>>> On Wed, Dec 7, 2016 at 12:37 PM, Michael Chan
>>>>>> <michael.chan@broadcom.com> wrote:
>>>>>>> On Wed, Dec 7, 2016 at 7:20 AM, Billy Shuman <wshuman3@gmail.com>
>>>>>>> wrote:
>>>>>>>> After resume on 4.9.0-rc8 tg3 is dead.
>>>>>>>>
>>>>>>>> In logs I see:
>>>>>>>> kernel: tg3 0000:44:00.0: phy probe failed, err -19
>>>>>>>> kernel: tg3 0000:44:00.0: Problem fetching invariants of chip,
>>>>>>>> aborting
>>>>>>>
>>>>>>> -19 is -ENODEV which means tg3 cannot read the PHY ID.
>>>>>>>
>>>>>>> If it's a true suspend/resume operation, the driver does not have
>>>>>>> to go through probe during resume. Please explain how you do
>>>>>>> suspend/resume.
>>>>>>>
>>>>>>
>>>>>> Sorry my previous message was accidentally sent to early.
>>>>>>
>>>>>> I used systemd (systemctl suspend) to suspend.
>>>>>>
>>>>> We need more information to proceed further.
>>>>> Without suspend, Are you able to use the tg3 port?
>>>>
>>>> Yes the port works fine without suspend.
>>> OK
>>>>
>>>>> Which Broadcom card are you having in laptop?
>>>>
>>>> The nic is a NetXtreme BCM57762 Gigabit Ethernet PCIe in a thunderbolt3
>>>> dock.
>>>>
>>> OK
>>>>> Please provide complete tg3 specific logs in dmesg.
>>>>>
>>>>
>>>> [ 32.084010] tg3.c:v3.137 (May 11, 2014)
>>>> [ 32.124695] tg3 0000:44:00.0 eth0: Tigon3 [partno(BCM957762) rev
>>>> 57766001] (PCI Express) MAC address 98:e7:f4:8b:13:19
>>>> [ 32.124698] tg3 0000:44:00.0 eth0: attached PHY is 57765
>>>> (10/100/1000Base-T Ethernet) (WireSpeed[1], EEE[1])
>>>> [ 32.124699] tg3 0000:44:00.0 eth0: RXcsums[1] LinkChgREG[0]
>>>> MIirq[0] ASF[0] TSOcap[1]
>>>> [ 32.124700] tg3 0000:44:00.0 eth0: dma_rwctrl[00000001]
>>>> dma_mask[64-bit]
>>>> [ 32.219764] tg3 0000:44:00.0 enp68s0: renamed from eth0
>>>> [ 36.219245] tg3 0000:44:00.0 enp68s0: Link is up at 1000 Mbps, full
>>>> duplex
>>>> [ 36.219250] tg3 0000:44:00.0 enp68s0: Flow control is on for TX and
>>>> on for RX
>>>> [ 36.219251] tg3 0000:44:00.0 enp68s0: EEE is disabled
>>>>
>>>> after resume
>>>> [ 92.292838] tg3 0000:44:00.0 enp68s0: No firmware running
>>>> [ 93.521744] tg3 0000:44:00.0: tg3_abort_hw timed out,
>>>> TX_MODE_ENABLE will not clear MAC_TX_MODE=ffffffff [ 106.704655]
>>>> tg3 0000:44:00.0 enp68s0: Link is down [ 108.370356] tg3
>>>> 0000:44:00.0: tg3_abort_hw timed out, TX_MODE_ENABLE will not clear
>>>> MAC_TX_MODE=ffffffff
>>>>
>>>> after rmmod, modprobe
>>>> [ 570.933636] tg3 0000:44:00.0: tg3_abort_hw timed out,
>>>> TX_MODE_ENABLE will not clear MAC_TX_MODE=ffffffff [ 604.847215]
>>>> tg3.c:v3.137 (May 11, 2014) [ 605.010075] tg3 0000:44:00.0: phy
>>>> probe failed, err -19 [ 605.010077] tg3 0000:44:00.0: Problem
>>>> fetching invariants of chip, aborting
>>>>
>>>>
>>>>
>>>>
>>> We will try to reproduce and update you on this.
>> We are unable to reproduce this issue with Ubuntu 16.10 (K4.8.0-22)
>> kernel.
>> We are in the process of verifying with 4.9.0-rc8 kernel and let you
>> know the feedback.
>> Can you please let us know the make/model of your laptop and procedure
>> followed to enable tg3 driver?
>
> The model is an HP Zbook Studio G3.
> The NIC is in the dock which is an HP Zbook Thunderbolt 3 dock.
>
> The module is loaded via udev.
> $ cat config | grep CONFIG_TIGON3
> CONFIG_TIGON3=m
>
>>>>>>> Did this work before? There has been very few changes to tg3
>>>>>>> recently.
>>>>>>>
>>>>>>
>>>>>> This is a new laptop for me, but the same behavior is seen on 4.4.36
>>>>>> and 4.8.12.
>>>>>>
>>>>>>>>
>>>>>>>> rmmod and modprobe does not fix the problem only a reboot resolves
>>>>>>>> the issue.
>>>>>>>>
>>>>>>>> Billy
^ permalink raw reply
* ANNOUNCE: Netdev 2.1 Call For Proposals Opened!
From: Jamal Hadi Salim @ 2017-01-16 15:00 UTC (permalink / raw)
To: netdev
Cc: netfilter-devel, netfilter, info, linux-wireless, tech-committee,
David Miller, Stephen Jaworski, Rob Echlin, lwn,
info@netdev01.org, tech-committee
Folks,
We are pleased to announce that the CFP for netdev 2.1 is now open.
Netdev 2.1 is a community-driven conference geared towards Linux
netheads. Linux kernel networking and user space utilization of the
interfaces to the Linux kernel networking subsystem are the focus.
If you are using Linux as a boot system for proprietary networking, then
this conference _may not be for you_.
Netdev2.1 will be held in Montreal April 6-8
For more details visit: www.netdevconf.org
Netdev 2.1 has a theme: IOT networking and DDOS.
Submissions in that area are highly encouraged.
Current topics include:
IOT networking.
DDOS.
Wireless.
Performance analysis and improvement.
Networking hardware and offload.
Netfilter.
Traffic control.
Different networking layers (L2/3, etc.).
Security.
Additional topics can be suggested.
Session submissions
--------------------
We encourage submission of paper and tutorial proposals, though we
*highly highly discourage* submission of recycled talks. Your proposals
will be evaluated by the technical committee that will provide prompt
feedback to you. If your proposal is approved, you will be requested to
submit your Camera-ready paper.
Unlike other conferences where only the schedule is announced,
proposals will be regularly announced as they get accepted on our
website and on relevant lists and twitter.
Please stay within the relevant topic focus and tie to FOSS Linux
networking to make it easier for the technical committee to provide
quick feedback. In order to give a talk you must be registered.
If your proposal is accepted you will not be charged a conference fee
or your conference fee will be refunded to you when your talk gets
accepted.
Types of proposals
-------------------
1. BoF
This serves to carry out initial discussions on some topic without any
pre-planned agenda. At least one person is responsible for introducing
the topic, chairing and moderating it. You have to list a few people
that will be planning to participate in the discussion. Expect people
from the public to participate actively in the discussions.
The lifetime of a BOF may be only one or two Netdev conference
gatherings. We discourage perpetual BOFs. BoFs don't need to have an
existing networking subsystem or mailing list. BOFs also don't need to
strive to be upgraded to be a Workshop in the future. Their longevity
could only be one conference. The sitting format could vary and be
either lecture or round table format depending on the proposal.
2. Talk
At least one person (of possibly multiple authors) is responsible for
the presentation. Details proposal/development plans to the public for
45 minutes.
A paper submission of the talk is expected (please use the provided
template). The paper has to be a minimum of two pages and not to exceed
10 pages (not including the references).
3. Tutorial
This implies planned agenda too.
Introduces audience to a topic.
Includes practical examples on using the functionality.
Functionality should already be in mainline or at least have been posted
already.
4. Workshops
Workshops are centered around existing networking subsystems.
Workshops are intended to be an extension of the mailing list in the
sense that many times previous discussions from the mailing list (or
that could otherwise have happened there) are taken to the round-table
to simplify the decision-making process.
The networking subsystem maintainer(s) should at least prepare a
list of agenda items well before the workshop takes place to allow
participants to come prepared; this makes the discussions most
productive.
Sometimes brain-storming sessions will also be appropriate where
being prepared is less important, for example for discussions
around new user requirements this can be very valuable.
At the workshop meeting itself discussions prevail and notes are
later sent back to the mailing list; presentations are typically
- at the discretion of the chairs - only used where needed to
clarify a problem statement for discussion.
Submiting Proposals
-------------------
Your proposal must be sent to tech-committee@netdevconf.org before
February 20, 2017 with the following format:
Title: Title of the proposal
Type: BoF, Talk, Tutorial.
Description: (350 words max).
Submitter names:
Estimated time: (talk = 45 minutes max., tutorial = 90 minutes max.).
Once your proposal is accepted, please prepare your paper based on the
conference template (odt or latex).
How to submit your paper
------------------------
Once your proposal is accepted and you have been invited to submit a
paper, please send the paper to both info@netdevconf.org and
tech-committee@netdevconf.org.
Deadlines
---------
February 20, 2017 Call for proposals close
Paper submission for review March 27, 2017
Slides submission April 3, 2017
How to submit your slides
-------------------------
If your slides of your talk are ready, please send the PDF to:
info@netdevconf.org.
cheers,
jamal
^ permalink raw reply
* Re: [PATCH net-next] sctp: remove useless code from sctp_apply_peer_addr_params
From: Xin Long @ 2017-01-16 15:10 UTC (permalink / raw)
To: Marcelo Ricardo Leitner
Cc: network dev, linux-sctp, Vlad Yasevich, Neil Horman
In-Reply-To: <99bc9f227c4c857d2bb258391231231d46c1899d.1484336556.git.marcelo.leitner@gmail.com>
On Sat, Jan 14, 2017 at 4:31 AM, Marcelo Ricardo Leitner
<marcelo.leitner@gmail.com> wrote:
> sctp_frag_point() doesn't store anything, and thus just calling it
> cannot do anything useful.
>
> sctp_apply_peer_addr_params is only called by
> sctp_setsockopt_peer_addr_params. When operating on an asoc,
> sctp_setsockopt_peer_addr_params will call sctp_apply_peer_addr_params
> once for the asoc, and then once for each transport this asoc has,
> meaning that the frag_point will be recomputed when updating the
> transports and calling it when updating the asoc is not necessary.
> IOW, no action is needed here and we can remove this call.
>
> Signed-off-by: Marcelo Ricardo Leitner <marcelo.leitner@gmail.com>
Reviewed-by: Xin Long <lucien.xin@gmail.com>
> ---
> net/sctp/socket.c | 1 -
> 1 file changed, 1 deletion(-)
>
> diff --git a/net/sctp/socket.c b/net/sctp/socket.c
> index 318c6786d6539a301ac7b76d82a49a1af3818d10..635e0341269330187c78ba93a35689f5c5d6be02 100644
> --- a/net/sctp/socket.c
> +++ b/net/sctp/socket.c
> @@ -2430,7 +2430,6 @@ static int sctp_apply_peer_addr_params(struct sctp_paddrparams *params,
> sctp_assoc_sync_pmtu(sctp_opt2sk(sp), asoc);
> } else if (asoc) {
> asoc->pathmtu = params->spp_pathmtu;
> - sctp_frag_point(asoc, params->spp_pathmtu);
> } else {
> sp->pathmtu = params->spp_pathmtu;
> }
> --
> 2.9.3
>
> --
> To unsubscribe from this list: send the line "unsubscribe linux-sctp" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at http://vger.kernel.org/majordomo-info.html
^ permalink raw reply
* Re: [PATCH net-next v2 1/3] net: ipv6: Allow shorthand delete of all nexthops in multipath route
From: Roopa Prabhu @ 2017-01-16 15:48 UTC (permalink / raw)
To: David Ahern; +Cc: netdev, ddutt
In-Reply-To: <1484510826-2723-2-git-send-email-dsa@cumulusnetworks.com>
On 1/15/17, 12:07 PM, David Ahern wrote:
> IPv4 allows multipath routes to be deleted using just the prefix and
> length. For example:
> $ ip ro ls vrf red
> unreachable default metric 8192
> 1.1.1.0/24
> nexthop via 10.100.1.254 dev eth1 weight 1
> nexthop via 10.11.200.2 dev eth11.200 weight 1
> 10.11.200.0/24 dev eth11.200 proto kernel scope link src 10.11.200.3
> 10.100.1.0/24 dev eth1 proto kernel scope link src 10.100.1.3
>
> $ ip ro del 1.1.1.0/24 vrf red
>
> $ ip ro ls vrf red
> unreachable default metric 8192
> 10.11.200.0/24 dev eth11.200 proto kernel scope link src 10.11.200.3
> 10.100.1.0/24 dev eth1 proto kernel scope link src 10.100.1.3
>
> The same notation does not work with IPv6 because of how multipath routes
> are implemented for IPv6. For IPv6 only the first nexthop of a multipath
> route is deleted if the request contains only a prefix and length. This
> leads to unnecessary complexity in userspace dealing with IPv6 multipath
> routes.
>
> This patch allows all nexthops to be deleted without specifying each one
> in the delete request by passing a new flag, RTM_F_ALL_NEXTHOPS, in
> rtm_flags. Internally, this is done by walking the sibling list of the
> route matching the specifications given (prefix, length, metric, protocol,
> etc).
>
> With this patch (and an updated iproute2 command):
> $ ip -6 ro ls vrf red
> 2001:db8::/120 via 2001:db8:1::62 dev eth1 metric 256 pref medium
> 2001:db8::/120 via 2001:db8:1::61 dev eth1 metric 256 pref medium
> 2001:db8::/120 via 2001:db8:1::60 dev eth1 metric 256 pref medium
> 2001:db8:1::/120 dev eth1 proto kernel metric 256 pref medium
> ...
>
> $ ip -6 ro del vrf red 1111::1/120
> $ ip -6 ro ls vrf red
> 2001:db8:1::/120 dev eth1 proto kernel metric 256 pref medium
> ...
>
> The flag is added to fib6_config by converting fc_type to a u16 (as
> noted fc_type only uses 8 bits). The new u16 hole is a bitmap with
> fc_delete_all_nexthop as the first bit.
>
> Suggested-by: Dinesh Dutt <ddutt@cumulusnetworks.com>
> Signed-off-by: David Ahern <dsa@cumulusnetworks.com>
> ---
> v2
> - switched example to rfc 3849 documentation address per request
> - changed delete loop to explicitly look at siblings list for
> first route matching specs given (metric, protocol, etc)
>
> include/net/ip6_fib.h | 4 +++-
> include/uapi/linux/rtnetlink.h | 1 +
> net/ipv6/route.c | 28 +++++++++++++++++++++++++---
> 3 files changed, 29 insertions(+), 4 deletions(-)
>
> diff --git a/include/net/ip6_fib.h b/include/net/ip6_fib.h
> index a74e2aa40ef4..11ab99e87c5f 100644
> --- a/include/net/ip6_fib.h
> +++ b/include/net/ip6_fib.h
> @@ -37,7 +37,9 @@ struct fib6_config {
> int fc_ifindex;
> u32 fc_flags;
> u32 fc_protocol;
> - u32 fc_type; /* only 8 bits are used */
> + u16 fc_type; /* only 8 bits are used */
> + u16 fc_delete_all_nexthop : 1,
> + __unused : 15;
>
> struct in6_addr fc_dst;
> struct in6_addr fc_src;
> diff --git a/include/uapi/linux/rtnetlink.h b/include/uapi/linux/rtnetlink.h
> index 8c93ad1ef9ab..7fb206bc42f9 100644
> --- a/include/uapi/linux/rtnetlink.h
> +++ b/include/uapi/linux/rtnetlink.h
> @@ -276,6 +276,7 @@ enum rt_scope_t {
> #define RTM_F_EQUALIZE 0x400 /* Multipath equalizer: NI */
> #define RTM_F_PREFIX 0x800 /* Prefix addresses */
> #define RTM_F_LOOKUP_TABLE 0x1000 /* set rtm_table to FIB lookup result */
> +#define RTM_F_ALL_NEXTHOPS 0x2000 /* delete all nexthops (IPv6) */
>
Do we really need the flag ?. It seems like delete with just prefix should delete all the routes in a multipath
route by default... (understand that you have it there to preserve existing behavior...for people who maybe relying on it. But this seems more like a bug fix. route replace went through a few such bug fixes "ipv6: fix ECMP route replacement"). ok with either approach.
^ permalink raw reply
* Re: [PATCH net-next v2 1/3] net: ipv6: Allow shorthand delete of all nexthops in multipath route
From: David Ahern @ 2017-01-16 15:58 UTC (permalink / raw)
To: Roopa Prabhu; +Cc: netdev, ddutt
In-Reply-To: <587CEB67.3040807@cumulusnetworks.com>
On 1/16/17 8:48 AM, Roopa Prabhu wrote:
> Do we really need the flag ?. It seems like delete with just prefix should delete all the routes in a multipath
> route by default... (understand that you have it there to preserve existing behavior...for people who maybe relying on it. But this seems more like a bug fix. route replace went through a few such bug fixes "ipv6: fix ECMP route replacement"). ok with either approach.
I'm fine dropping the flag for this patch and just deleting the entire route. The flag only exists because this patch changes behavior visible to the user.
^ permalink raw reply
* [PATCH net 1/3] net/mlx4_core: Fix racy CQ (Completion Queue) free
From: Tariq Toukan @ 2017-01-16 16:31 UTC (permalink / raw)
To: David S. Miller
Cc: netdev, Eran Ben Elisha, Jack Morgenstein, Matan Barak,
Tariq Toukan
In-Reply-To: <1484584299-3950-1-git-send-email-tariqt@mellanox.com>
From: Jack Morgenstein <jackm@dev.mellanox.co.il>
In function mlx4_cq_completion() and mlx4_cq_event(), the
radix_tree_lookup requires a rcu_read_lock.
This is mandatory: if another core frees the CQ, it could
run the radix_tree_node_rcu_free() call_rcu() callback while
its being used by the radix tree lookup function.
Additionally, in function mlx4_cq_event(), since we are adding
the rcu lock around the radix-tree lookup, we no longer need to take
the spinlock. Also, the synchronize_irq() call for the async event
eliminates the need for incrementing the cq reference count in
mlx4_cq_event().
Other changes:
1. In function mlx4_cq_free(), replace spin_lock_irq with spin_lock:
we no longer take this spinlock in the interrupt context.
The spinlock here, therefore, simply protects against different
threads simultaneously invoking mlx4_cq_free() for different cq's.
2. In function mlx4_cq_free(), we move the radix tree delete to before
the synchronize_irq() calls. This guarantees that we will not
access this cq during any subsequent interrupts, and therefore can
safely free the CQ after the synchronize_irq calls. The rcu_read_lock
in the interrupt handlers only needs to protect against corrupting the
radix tree; the interrupt handlers may access the cq outside the
rcu_read_lock due to the synchronize_irq calls which protect against
premature freeing of the cq.
3. In function mlx4_cq_event(), we change the mlx_warn message to mlx4_dbg.
4. We leave the cq reference count mechanism in place, because it is
still needed for the cq completion tasklet mechanism.
Fixes: 6d90aa5cf17b ("net/mlx4_core: Make sure there are no pending async events when freeing CQ")
Fixes: 225c7b1feef1 ("IB/mlx4: Add a driver Mellanox ConnectX InfiniBand adapters")
Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il>
Signed-off-by: Matan Barak <matanb@mellanox.com>
Signed-off-by: Tariq Toukan <tariqt@mellanox.com>
---
drivers/net/ethernet/mellanox/mlx4/cq.c | 38 +++++++++++++++++----------------
1 file changed, 20 insertions(+), 18 deletions(-)
diff --git a/drivers/net/ethernet/mellanox/mlx4/cq.c b/drivers/net/ethernet/mellanox/mlx4/cq.c
index a849da92f857..6b8635378f1f 100644
--- a/drivers/net/ethernet/mellanox/mlx4/cq.c
+++ b/drivers/net/ethernet/mellanox/mlx4/cq.c
@@ -101,13 +101,19 @@ void mlx4_cq_completion(struct mlx4_dev *dev, u32 cqn)
{
struct mlx4_cq *cq;
+ rcu_read_lock();
cq = radix_tree_lookup(&mlx4_priv(dev)->cq_table.tree,
cqn & (dev->caps.num_cqs - 1));
+ rcu_read_unlock();
+
if (!cq) {
mlx4_dbg(dev, "Completion event for bogus CQ %08x\n", cqn);
return;
}
+ /* Acessing the CQ outside of rcu_read_lock is safe, because
+ * the CQ is freed only after interrupt handling is completed.
+ */
++cq->arm_sn;
cq->comp(cq);
@@ -118,23 +124,19 @@ void mlx4_cq_event(struct mlx4_dev *dev, u32 cqn, int event_type)
struct mlx4_cq_table *cq_table = &mlx4_priv(dev)->cq_table;
struct mlx4_cq *cq;
- spin_lock(&cq_table->lock);
-
+ rcu_read_lock();
cq = radix_tree_lookup(&cq_table->tree, cqn & (dev->caps.num_cqs - 1));
- if (cq)
- atomic_inc(&cq->refcount);
-
- spin_unlock(&cq_table->lock);
+ rcu_read_unlock();
if (!cq) {
- mlx4_warn(dev, "Async event for bogus CQ %08x\n", cqn);
+ mlx4_dbg(dev, "Async event for bogus CQ %08x\n", cqn);
return;
}
+ /* Acessing the CQ outside of rcu_read_lock is safe, because
+ * the CQ is freed only after interrupt handling is completed.
+ */
cq->event(cq, event_type);
-
- if (atomic_dec_and_test(&cq->refcount))
- complete(&cq->free);
}
static int mlx4_SW2HW_CQ(struct mlx4_dev *dev, struct mlx4_cmd_mailbox *mailbox,
@@ -301,9 +303,9 @@ int mlx4_cq_alloc(struct mlx4_dev *dev, int nent,
if (err)
return err;
- spin_lock_irq(&cq_table->lock);
+ spin_lock(&cq_table->lock);
err = radix_tree_insert(&cq_table->tree, cq->cqn, cq);
- spin_unlock_irq(&cq_table->lock);
+ spin_unlock(&cq_table->lock);
if (err)
goto err_icm;
@@ -349,9 +351,9 @@ int mlx4_cq_alloc(struct mlx4_dev *dev, int nent,
return 0;
err_radix:
- spin_lock_irq(&cq_table->lock);
+ spin_lock(&cq_table->lock);
radix_tree_delete(&cq_table->tree, cq->cqn);
- spin_unlock_irq(&cq_table->lock);
+ spin_unlock(&cq_table->lock);
err_icm:
mlx4_cq_free_icm(dev, cq->cqn);
@@ -370,15 +372,15 @@ void mlx4_cq_free(struct mlx4_dev *dev, struct mlx4_cq *cq)
if (err)
mlx4_warn(dev, "HW2SW_CQ failed (%d) for CQN %06x\n", err, cq->cqn);
+ spin_lock(&cq_table->lock);
+ radix_tree_delete(&cq_table->tree, cq->cqn);
+ spin_unlock(&cq_table->lock);
+
synchronize_irq(priv->eq_table.eq[MLX4_CQ_TO_EQ_VECTOR(cq->vector)].irq);
if (priv->eq_table.eq[MLX4_CQ_TO_EQ_VECTOR(cq->vector)].irq !=
priv->eq_table.eq[MLX4_EQ_ASYNC].irq)
synchronize_irq(priv->eq_table.eq[MLX4_EQ_ASYNC].irq);
- spin_lock_irq(&cq_table->lock);
- radix_tree_delete(&cq_table->tree, cq->cqn);
- spin_unlock_irq(&cq_table->lock);
-
if (atomic_dec_and_test(&cq->refcount))
complete(&cq->free);
wait_for_completion(&cq->free);
--
1.8.3.1
^ permalink raw reply related
* [PATCH net 3/3] net/mlx4_core: Eliminate warning messages for SRQ_LIMIT under SRIOV
From: Tariq Toukan @ 2017-01-16 16:31 UTC (permalink / raw)
To: David S. Miller; +Cc: netdev, Eran Ben Elisha, Jack Morgenstein, Tariq Toukan
In-Reply-To: <1484584299-3950-1-git-send-email-tariqt@mellanox.com>
From: Jack Morgenstein <jackm@dev.mellanox.co.il>
When running SRIOV, warnings for SRQ LIMIT events flood the Hypervisor's
message log when (correct, normally operating) apps use SRQ LIMIT events
as a trigger to post WQEs to SRQs.
Add more information to the existing debug printout for SRQ_LIMIT, and
output the warning messages only for the SRQ CATAS ERROR event.
Fixes: acba2420f9d2 ("mlx4_core: Add wrapper functions and comm channel and slave event support to EQs")
Fixes: e0debf9cb50d ("mlx4_core: Reduce warning message for SRQ_LIMIT event to debug level")
Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il>
Signed-off-by: Tariq Toukan <tariqt@mellanox.com>
---
drivers/net/ethernet/mellanox/mlx4/eq.c | 23 ++++++++++++++---------
1 file changed, 14 insertions(+), 9 deletions(-)
diff --git a/drivers/net/ethernet/mellanox/mlx4/eq.c b/drivers/net/ethernet/mellanox/mlx4/eq.c
index cd3638e6fe25..0509996957d9 100644
--- a/drivers/net/ethernet/mellanox/mlx4/eq.c
+++ b/drivers/net/ethernet/mellanox/mlx4/eq.c
@@ -554,8 +554,9 @@ static int mlx4_eq_int(struct mlx4_dev *dev, struct mlx4_eq *eq)
break;
case MLX4_EVENT_TYPE_SRQ_LIMIT:
- mlx4_dbg(dev, "%s: MLX4_EVENT_TYPE_SRQ_LIMIT\n",
- __func__);
+ mlx4_dbg(dev, "%s: MLX4_EVENT_TYPE_SRQ_LIMIT. srq_no=0x%x, eq 0x%x\n",
+ __func__, be32_to_cpu(eqe->event.srq.srqn),
+ eq->eqn);
case MLX4_EVENT_TYPE_SRQ_CATAS_ERROR:
if (mlx4_is_master(dev)) {
/* forward only to slave owning the SRQ */
@@ -570,15 +571,19 @@ static int mlx4_eq_int(struct mlx4_dev *dev, struct mlx4_eq *eq)
eq->eqn, eq->cons_index, ret);
break;
}
- mlx4_warn(dev, "%s: slave:%d, srq_no:0x%x, event: %02x(%02x)\n",
- __func__, slave,
- be32_to_cpu(eqe->event.srq.srqn),
- eqe->type, eqe->subtype);
+ if (eqe->type ==
+ MLX4_EVENT_TYPE_SRQ_CATAS_ERROR)
+ mlx4_warn(dev, "%s: slave:%d, srq_no:0x%x, event: %02x(%02x)\n",
+ __func__, slave,
+ be32_to_cpu(eqe->event.srq.srqn),
+ eqe->type, eqe->subtype);
if (!ret && slave != dev->caps.function) {
- mlx4_warn(dev, "%s: sending event %02x(%02x) to slave:%d\n",
- __func__, eqe->type,
- eqe->subtype, slave);
+ if (eqe->type ==
+ MLX4_EVENT_TYPE_SRQ_CATAS_ERROR)
+ mlx4_warn(dev, "%s: sending event %02x(%02x) to slave:%d\n",
+ __func__, eqe->type,
+ eqe->subtype, slave);
mlx4_slave_event(dev, slave, eqe);
break;
}
--
1.8.3.1
^ permalink raw reply related
* [PATCH net 2/3] net/mlx4_core: Fix when to save some qp context flags for dynamic VST to VGT transitions
From: Tariq Toukan @ 2017-01-16 16:31 UTC (permalink / raw)
To: David S. Miller; +Cc: netdev, Eran Ben Elisha, Jack Morgenstein, Tariq Toukan
In-Reply-To: <1484584299-3950-1-git-send-email-tariqt@mellanox.com>
From: Jack Morgenstein <jackm@dev.mellanox.co.il>
Save the qp context flags byte containing the flag disabling vlan stripping
in the RESET to INIT qp transition, rather than in the INIT to RTR
transition. Per the firmware spec, the flags in this byte are active
in the RESET to INIT transition.
As a result of saving the flags in the incorrect qp transition, when
switching dynamically from VGT to VST and back to VGT, the vlan
remained stripped (as is required for VST) and did not return to
not-stripped (as is required for VGT).
Fixes: f0f829bf42cd ("net/mlx4_core: Add immediate activate for VGT->VST->VGT")
Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il>
Signed-off-by: Tariq Toukan <tariqt@mellanox.com>
---
drivers/net/ethernet/mellanox/mlx4/resource_tracker.c | 5 +++--
1 file changed, 3 insertions(+), 2 deletions(-)
diff --git a/drivers/net/ethernet/mellanox/mlx4/resource_tracker.c b/drivers/net/ethernet/mellanox/mlx4/resource_tracker.c
index 56185a0b827d..1822382212ee 100644
--- a/drivers/net/ethernet/mellanox/mlx4/resource_tracker.c
+++ b/drivers/net/ethernet/mellanox/mlx4/resource_tracker.c
@@ -2980,6 +2980,9 @@ int mlx4_RST2INIT_QP_wrapper(struct mlx4_dev *dev, int slave,
put_res(dev, slave, srqn, RES_SRQ);
qp->srq = srq;
}
+
+ /* Save param3 for dynamic changes from VST back to VGT */
+ qp->param3 = qpc->param3;
put_res(dev, slave, rcqn, RES_CQ);
put_res(dev, slave, mtt_base, RES_MTT);
res_end_move(dev, slave, RES_QP, qpn);
@@ -3772,7 +3775,6 @@ int mlx4_INIT2RTR_QP_wrapper(struct mlx4_dev *dev, int slave,
int qpn = vhcr->in_modifier & 0x7fffff;
struct res_qp *qp;
u8 orig_sched_queue;
- __be32 orig_param3 = qpc->param3;
u8 orig_vlan_control = qpc->pri_path.vlan_control;
u8 orig_fvl_rx = qpc->pri_path.fvl_rx;
u8 orig_pri_path_fl = qpc->pri_path.fl;
@@ -3814,7 +3816,6 @@ int mlx4_INIT2RTR_QP_wrapper(struct mlx4_dev *dev, int slave,
*/
if (!err) {
qp->sched_queue = orig_sched_queue;
- qp->param3 = orig_param3;
qp->vlan_control = orig_vlan_control;
qp->fvl_rx = orig_fvl_rx;
qp->pri_path_fl = orig_pri_path_fl;
--
1.8.3.1
^ permalink raw reply related
* [PATCH net 0/3] mlx4 core fixes
From: Tariq Toukan @ 2017-01-16 16:31 UTC (permalink / raw)
To: David S. Miller; +Cc: netdev, Eran Ben Elisha, Jack Morgenstein, Tariq Toukan
Hi Dave,
This patchset contains bug fixes from Jack to the mlx4 Core driver.
Patch 1 solves a race in the flow of CQ free.
Patch 2 moves some qp context flags update to the correct qp transition.
Patch 3 eliminates warnings from the path of SRQ_LIMIT that flood the message log,
and keeps them only in the path of SRQ_CATAS_ERROR.
Series generated against net commit:
1a717fcf8bbe Merge tag 'mac80211-for-davem-2017-01-13' of git://git.kernel.org/pub/scm/linux/kernel/git/jberg/mac80211
Thanks,
Tariq.
Jack Morgenstein (3):
net/mlx4_core: Fix racy CQ (Completion Queue) free
net/mlx4_core: Fix when to save some qp context flags for dynamic VST
to VGT transitions
net/mlx4_core: Eliminate warning messages for SRQ_LIMIT under SRIOV
drivers/net/ethernet/mellanox/mlx4/cq.c | 38 ++++++++++++----------
drivers/net/ethernet/mellanox/mlx4/eq.c | 23 ++++++++-----
.../net/ethernet/mellanox/mlx4/resource_tracker.c | 5 +--
3 files changed, 37 insertions(+), 29 deletions(-)
--
1.8.3.1
^ permalink raw reply
* Re: pull request: bluetooth 2017-01-16
From: David Miller @ 2017-01-16 16:37 UTC (permalink / raw)
To: johan.hedberg; +Cc: linux-bluetooth, netdev
In-Reply-To: <20170116075831.GA7190@x1c>
From: Johan Hedberg <johan.hedberg@gmail.com>
Date: Mon, 16 Jan 2017 09:58:31 +0200
> Here are a couple of important 802.15.4 driver fixes for the 4.10
> kernel.
>
> Please let me know if there are any issues pulling. Thanks.
Pulled, thank you.
^ permalink raw reply
* Re: [PATCH net] mlx4: do not call napi_schedule() without care
From: David Miller @ 2017-01-16 16:45 UTC (permalink / raw)
To: ttoukan.linux; +Cc: eric.dumazet, netdev, erezsh, eugenia, tariqt
In-Reply-To: <352911f4-7073-ad7f-83d9-dbcc6e07d553@gmail.com>
From: Tariq Toukan <ttoukan.linux@gmail.com>
Date: Mon, 16 Jan 2017 11:02:44 +0200
> Thanks Eric.
>
> On 13/01/2017 6:39 PM, Eric Dumazet wrote:
>> From: Eric Dumazet <edumazet@google.com>
>>
>> Disable BH around the call to napi_schedule() to avoid following
>> warning
>>
>> [ 52.095499] NOHZ: local_softirq_pending 08
>> [ 52.421291] NOHZ: local_softirq_pending 08
>> [ 52.608313] NOHZ: local_softirq_pending 08
>>
>> Fixes: 8d59de8f7bb3 ("net/mlx4_en: Process all completions in RX rings
>> after port goes up")
>> Signed-off-by: Eric Dumazet <edumazet@google.com>
...
> Acked-by: Tariq Toukan <tariqt@mellanox.com>
Applied and queued up for -stable, thanks.
^ permalink raw reply
* Re: [PATCH 1/2] qed: Replace memset with eth_zero_addr
From: David Miller @ 2017-01-16 16:46 UTC (permalink / raw)
To: mayhs11saini; +Cc: Yuval.Mintz, Ariel.Elior, everest-linux-l2, netdev
In-Reply-To: <20170116092435.GA2970@mystictot>
From: Shyam Saini <mayhs11saini@gmail.com>
Date: Mon, 16 Jan 2017 14:54:35 +0530
> On Sun, Jan 15, 2017 at 11:38:30PM -0500, David Miller wrote:
>>
>> Please do not ever submit two patches which have the same exact commit
>> header line, as these two patches do.
>>
>> When someone looks into the shortlog of GIT history all they will see
>> is "qed: Replace memset with eth_zero_addr" twice.
>>
>> This gives the reader no idea what might be different between those
>> two changes.
>>
>> Therefore you must give unique a commit header text for each change,
>> which communicates sufficiently what is different in each change.
>
> Thanks a lot for correcting me. I'll take care of this thing.
>
> I'm resending these two patches as
> 1). qed: Replace memset with eth_zero_addr
> 2). qed: Use eth_zero_addr
>
> I hope it resolves same commit header line conflict.
You aren't understanding the point.
Those two lines still say exactly the same thing.
What is different about these two changes? The answer to that question
must propagate into those lines of text.
^ permalink raw reply
* Re: [PATCH] sfc: Replace memset with eth_zero_addr
From: David Miller @ 2017-01-16 16:47 UTC (permalink / raw)
To: ecree; +Cc: mayhs11saini, bkenward, netdev, linux-net-drivers
In-Reply-To: <7f1c75ec-3cad-0071-9f41-86773361281e@solarflare.com>
From: Edward Cree <ecree@solarflare.com>
Date: Mon, 16 Jan 2017 09:35:38 +0000
> On 16/01/17 03:56, Shyam Saini wrote:
>> Use eth_zero_addr to assign zero address to the given address array
>> instead of memset when the second argument in memset is address
>> of zero which makes the code clearer and also add header
>> file linux/etherdevice.h
>>
>> Signed-off-by: Shyam Saini <mayhs11saini@gmail.com>
>> ---
> Acked-by: Edward Cree <ecree@solarflare.com>
Applied to net-next.
^ permalink raw reply
* Re: [PATCH net v3] ravb: do not use zero-length alignment DMA descriptor
From: David Miller @ 2017-01-16 16:53 UTC (permalink / raw)
To: sergei.shtylyov
Cc: horms+renesas, magnus.damm, netdev, linux-renesas-soc,
masaru.nagai.vx
In-Reply-To: <96396803-caac-13cd-3e17-f9d10fc608ad@cogentembedded.com>
From: Sergei Shtylyov <sergei.shtylyov@cogentembedded.com>
Date: Mon, 16 Jan 2017 16:01:49 +0300
> On 01/16/2017 01:45 PM, Simon Horman wrote:
>
>> From: Masaru Nagai <masaru.nagai.vx@renesas.com>
>>
>> Due to alignment requirements of the hardware transmissions are split
>> into
>> two DMA descriptors, a small padding descriptor of 0 - 3 bytes in
>> length
>> followed by a descriptor for rest of the packet.
>>
>> In the case of IP packets the first descriptor will never be zero due
>> to
>> the way that the stack aligns buffers for IP packets. However, for
>> non-IP
>> packets it may be zero.
>>
>> In that case it has been reported that timeouts occur, presumably
>> because
>> transmission stops at the first zero-length DMA descriptor and thus
>> the
>> packet is not transmitted. However, in my environment a BUG is
>> triggered as
>> follows:
> [...]
>
>> Fixes: 2f45d1902acf ("ravb: minimize TX data copying")
>> Signed-off-by: Masaru Nagai <masaru.nagai.vx@renesas.com
>> Signed-off-by: Simon Horman <horms+renesas@verge.net.au>
>
> Acked-by: Sergei Shtylyov <sergei.shtylyov@cogentembedded.com>
Applied and queued up for -stable, thanks.
^ permalink raw reply
* Re: [PATCH] cpmac: remove hopeless #warning
From: David Miller @ 2017-01-16 16:54 UTC (permalink / raw)
To: arnd; +Cc: f.fainelli, tremyfr, netdev, linux-kernel
In-Reply-To: <20170116132113.2056012-1-arnd@arndb.de>
From: Arnd Bergmann <arnd@arndb.de>
Date: Mon, 16 Jan 2017 14:20:54 +0100
> The #warning was present 10 years ago when the driver first got merged.
> As the platform is rather obsolete by now, it seems very unlikely that
> the warning will cause anyone to fix the code properly.
>
> kernelci.org reports the warning for every build in the meantime, so
> I think it's better to just turn it into a code comment to reduce
> noise.
>
> Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Agreed, applied, thanks Arnd.
^ permalink raw reply
* RE: [PATCH 1/2] qed: Replace memset with eth_zero_addr
From: Mintz, Yuval @ 2017-01-16 17:05 UTC (permalink / raw)
To: mayhs11saini@gmail.com; +Cc: netdev@vger.kernel.org, David Miller
In-Reply-To: <20170116.114606.1692928906878877115.davem@davemloft.net>
> > On Sun, Jan 15, 2017 at 11:38:30PM -0500, David Miller wrote:
> >>
> >> Please do not ever submit two patches which have the same exact
> >> commit header line, as these two patches do.
> >>
> >> When someone looks into the shortlog of GIT history all they will see
> >> is "qed: Replace memset with eth_zero_addr" twice.
> >>
> >> This gives the reader no idea what might be different between those
> >> two changes.
> >>
> >> Therefore you must give unique a commit header text for each change,
> >> which communicates sufficiently what is different in each change.
> >
> > Thanks a lot for correcting me. I'll take care of this thing.
> >
> > I'm resending these two patches as
> > 1). qed: Replace memset with eth_zero_addr
> > 2). qed: Use eth_zero_addr
> >
> > I hope it resolves same commit header line conflict.
>
> You aren't understanding the point.
>
> Those two lines still say exactly the same thing.
>
> What is different about these two changes? The answer to that question
> must propagate into those lines of text.
Other than the fact these 2 patches change 2 different qed files,
is there any significant difference between what each does?
If not, why not simply do both in a single patch?
^ permalink raw reply
* [PATCH v2 net-next 0/2] mvneta xmit_more and bql support
From: Marcin Wojtas @ 2017-01-16 17:08 UTC (permalink / raw)
To: linux-kernel, linux-arm-kernel, netdev
Cc: davem, linux, sebastian.hesselbarth, andrew, jason,
thomas.petazzoni, gregory.clement, mw, jaz
Hi,
This is a delayed v2 of short patchset, which introduces xmit_more and BQL
to mvneta driver. The only one change was added in xmit_more support -
condition check preventing excessive descriptors concatenation before
flushing in HW.
Any comments or feedback would be welcome.
Best regards,
Marcin
Changelog:
v1 -> v2:
* Add checking condition that ensures too much descriptors are not
concatenated before flushing in HW.
Marcin Wojtas (1):
net: mvneta: add BQL support
Simon Guinot (1):
net: mvneta: add xmit_more support
drivers/net/ethernet/marvell/mvneta.c | 35 +++++++++++++++++++++++++++++------
1 file changed, 29 insertions(+), 6 deletions(-)
--
1.8.3.1
^ permalink raw reply
* [PATCH v2 net-next 1/2] net: mvneta: add xmit_more support
From: Marcin Wojtas @ 2017-01-16 17:08 UTC (permalink / raw)
To: linux-kernel, linux-arm-kernel, netdev
Cc: davem, linux, sebastian.hesselbarth, andrew, jason,
thomas.petazzoni, gregory.clement, mw, jaz, Simon Guinot
In-Reply-To: <1484586512-16412-1-git-send-email-mw@semihalf.com>
From: Simon Guinot <simon.guinot@sequanux.org>
Basing on xmit_more flag of the skb, TX descriptors can be concatenated
before flushing. This commit delay Tx descriptor flush if the queue is
running and if there is more skb's to send.
A maximum allowed number of descriptors for flushing at once due to
MVNETA_TXQ_UPDATE_REG(q) reqisters limitation, is 255. Because of that
a new macro was added (MVNETA_TXQ_DEC_SENT_MASK) in order to ensure that
concatenated amount of descriptor does not exceed that value.
Signed-off-by: Simon Guinot <simon.guinot@sequanux.org>
Signed-off-by: Marcin Wojtas <mw@semihalf.com>
---
drivers/net/ethernet/marvell/mvneta.c | 13 ++++++++++---
1 file changed, 10 insertions(+), 3 deletions(-)
diff --git a/drivers/net/ethernet/marvell/mvneta.c b/drivers/net/ethernet/marvell/mvneta.c
index 3607d8f..9624537 100644
--- a/drivers/net/ethernet/marvell/mvneta.c
+++ b/drivers/net/ethernet/marvell/mvneta.c
@@ -224,6 +224,7 @@
#define MVNETA_TXQ_SENT_THRESH_MASK(coal) ((coal) << 16)
#define MVNETA_TXQ_UPDATE_REG(q) (0x3c60 + ((q) << 2))
#define MVNETA_TXQ_DEC_SENT_SHIFT 16
+#define MVNETA_TXQ_DEC_SENT_MASK 0xff
#define MVNETA_TXQ_STATUS_REG(q) (0x3c40 + ((q) << 2))
#define MVNETA_TXQ_SENT_DESC_SHIFT 16
#define MVNETA_TXQ_SENT_DESC_MASK 0x3fff0000
@@ -525,6 +526,7 @@ struct mvneta_tx_queue {
* descriptor ring
*/
int count;
+ int pending;
int tx_stop_threshold;
int tx_wake_threshold;
@@ -818,8 +820,9 @@ static void mvneta_txq_pend_desc_add(struct mvneta_port *pp,
/* Only 255 descriptors can be added at once ; Assume caller
* process TX desriptors in quanta less than 256
*/
- val = pend_desc;
+ val = pend_desc + txq->pending;
mvreg_write(pp, MVNETA_TXQ_UPDATE_REG(txq->id), val);
+ txq->pending = 0;
}
/* Get pointer to next TX descriptor to be processed (send) by HW */
@@ -2399,11 +2402,15 @@ static int mvneta_tx(struct sk_buff *skb, struct net_device *dev)
struct netdev_queue *nq = netdev_get_tx_queue(dev, txq_id);
txq->count += frags;
- mvneta_txq_pend_desc_add(pp, txq, frags);
-
if (txq->count >= txq->tx_stop_threshold)
netif_tx_stop_queue(nq);
+ if (!skb->xmit_more || netif_xmit_stopped(nq) ||
+ txq->pending + frags > MVNETA_TXQ_DEC_SENT_MASK)
+ mvneta_txq_pend_desc_add(pp, txq, frags);
+ else
+ txq->pending += frags;
+
u64_stats_update_begin(&stats->syncp);
stats->tx_packets++;
stats->tx_bytes += len;
--
1.8.3.1
^ 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