Commit 115c9b81 authored by Greg Rose's avatar Greg Rose Committed by David S. Miller

rtnetlink: Fix problem with buffer allocation

Implement a new netlink attribute type IFLA_EXT_MASK.  The mask
is a 32 bit value that can be used to indicate to the kernel that
certain extended ifinfo values are requested by the user application.
At this time the only mask value defined is RTEXT_FILTER_VF to
indicate that the user wants the ifinfo dump to send information
about the VFs belonging to the interface.

This patch fixes a bug in which certain applications do not have
large enough buffers to accommodate the extra information returned
by the kernel with large numbers of SR-IOV virtual functions.
Those applications will not send the new netlink attribute with
the interface info dump request netlink messages so they will
not get unexpectedly large request buffers returned by the kernel.

Modifies the rtnl_calcit function to traverse the list of net
devices and compute the minimum buffer size that can hold the
info dumps of all matching devices based upon the filter passed
in via the new netlink attribute filter mask.  If no filter
mask is sent then the buffer allocation defaults to NLMSG_GOODSIZE.

With this change it is possible to add yet to be defined netlink
attributes to the dump request which should make it fairly extensible
in the future.
Signed-off-by: default avatarGreg Rose <gregory.v.rose@intel.com>
Signed-off-by: default avatarDavid S. Miller <davem@davemloft.net>
parent 84338a6c
...@@ -137,6 +137,7 @@ enum { ...@@ -137,6 +137,7 @@ enum {
IFLA_AF_SPEC, IFLA_AF_SPEC,
IFLA_GROUP, /* Group the device belongs to */ IFLA_GROUP, /* Group the device belongs to */
IFLA_NET_NS_FD, IFLA_NET_NS_FD,
IFLA_EXT_MASK, /* Extended info mask, VFs, etc */
__IFLA_MAX __IFLA_MAX
}; };
......
...@@ -602,6 +602,9 @@ struct tcamsg { ...@@ -602,6 +602,9 @@ struct tcamsg {
#define TCA_ACT_TAB 1 /* attr type must be >=1 */ #define TCA_ACT_TAB 1 /* attr type must be >=1 */
#define TCAA_MAX 1 #define TCAA_MAX 1
/* New extended info filters for IFLA_EXT_MASK */
#define RTEXT_FILTER_VF (1 << 0)
/* End of information exported to user level */ /* End of information exported to user level */
#ifdef __KERNEL__ #ifdef __KERNEL__
......
...@@ -6,7 +6,7 @@ ...@@ -6,7 +6,7 @@
typedef int (*rtnl_doit_func)(struct sk_buff *, struct nlmsghdr *, void *); typedef int (*rtnl_doit_func)(struct sk_buff *, struct nlmsghdr *, void *);
typedef int (*rtnl_dumpit_func)(struct sk_buff *, struct netlink_callback *); typedef int (*rtnl_dumpit_func)(struct sk_buff *, struct netlink_callback *);
typedef u16 (*rtnl_calcit_func)(struct sk_buff *); typedef u16 (*rtnl_calcit_func)(struct sk_buff *, struct nlmsghdr *);
extern int __rtnl_register(int protocol, int msgtype, extern int __rtnl_register(int protocol, int msgtype,
rtnl_doit_func, rtnl_dumpit_func, rtnl_doit_func, rtnl_dumpit_func,
......
...@@ -60,7 +60,6 @@ struct rtnl_link { ...@@ -60,7 +60,6 @@ struct rtnl_link {
}; };
static DEFINE_MUTEX(rtnl_mutex); static DEFINE_MUTEX(rtnl_mutex);
static u16 min_ifinfo_dump_size;
void rtnl_lock(void) void rtnl_lock(void)
{ {
...@@ -724,10 +723,11 @@ static void copy_rtnl_link_stats64(void *v, const struct rtnl_link_stats64 *b) ...@@ -724,10 +723,11 @@ static void copy_rtnl_link_stats64(void *v, const struct rtnl_link_stats64 *b)
} }
/* All VF info */ /* All VF info */
static inline int rtnl_vfinfo_size(const struct net_device *dev) static inline int rtnl_vfinfo_size(const struct net_device *dev,
u32 ext_filter_mask)
{ {
if (dev->dev.parent && dev_is_pci(dev->dev.parent)) { if (dev->dev.parent && dev_is_pci(dev->dev.parent) &&
(ext_filter_mask & RTEXT_FILTER_VF)) {
int num_vfs = dev_num_vf(dev->dev.parent); int num_vfs = dev_num_vf(dev->dev.parent);
size_t size = nla_total_size(sizeof(struct nlattr)); size_t size = nla_total_size(sizeof(struct nlattr));
size += nla_total_size(num_vfs * sizeof(struct nlattr)); size += nla_total_size(num_vfs * sizeof(struct nlattr));
...@@ -766,7 +766,8 @@ static size_t rtnl_port_size(const struct net_device *dev) ...@@ -766,7 +766,8 @@ static size_t rtnl_port_size(const struct net_device *dev)
return port_self_size; return port_self_size;
} }
static noinline size_t if_nlmsg_size(const struct net_device *dev) static noinline size_t if_nlmsg_size(const struct net_device *dev,
u32 ext_filter_mask)
{ {
return NLMSG_ALIGN(sizeof(struct ifinfomsg)) return NLMSG_ALIGN(sizeof(struct ifinfomsg))
+ nla_total_size(IFNAMSIZ) /* IFLA_IFNAME */ + nla_total_size(IFNAMSIZ) /* IFLA_IFNAME */
...@@ -784,8 +785,9 @@ static noinline size_t if_nlmsg_size(const struct net_device *dev) ...@@ -784,8 +785,9 @@ static noinline size_t if_nlmsg_size(const struct net_device *dev)
+ nla_total_size(4) /* IFLA_MASTER */ + nla_total_size(4) /* IFLA_MASTER */
+ nla_total_size(1) /* IFLA_OPERSTATE */ + nla_total_size(1) /* IFLA_OPERSTATE */
+ nla_total_size(1) /* IFLA_LINKMODE */ + nla_total_size(1) /* IFLA_LINKMODE */
+ nla_total_size(4) /* IFLA_NUM_VF */ + nla_total_size(ext_filter_mask
+ rtnl_vfinfo_size(dev) /* IFLA_VFINFO_LIST */ & RTEXT_FILTER_VF ? 4 : 0) /* IFLA_NUM_VF */
+ rtnl_vfinfo_size(dev, ext_filter_mask) /* IFLA_VFINFO_LIST */
+ rtnl_port_size(dev) /* IFLA_VF_PORTS + IFLA_PORT_SELF */ + rtnl_port_size(dev) /* IFLA_VF_PORTS + IFLA_PORT_SELF */
+ rtnl_link_get_size(dev) /* IFLA_LINKINFO */ + rtnl_link_get_size(dev) /* IFLA_LINKINFO */
+ rtnl_link_get_af_size(dev); /* IFLA_AF_SPEC */ + rtnl_link_get_af_size(dev); /* IFLA_AF_SPEC */
...@@ -868,7 +870,7 @@ static int rtnl_port_fill(struct sk_buff *skb, struct net_device *dev) ...@@ -868,7 +870,7 @@ static int rtnl_port_fill(struct sk_buff *skb, struct net_device *dev)
static int rtnl_fill_ifinfo(struct sk_buff *skb, struct net_device *dev, static int rtnl_fill_ifinfo(struct sk_buff *skb, struct net_device *dev,
int type, u32 pid, u32 seq, u32 change, int type, u32 pid, u32 seq, u32 change,
unsigned int flags) unsigned int flags, u32 ext_filter_mask)
{ {
struct ifinfomsg *ifm; struct ifinfomsg *ifm;
struct nlmsghdr *nlh; struct nlmsghdr *nlh;
...@@ -941,10 +943,11 @@ static int rtnl_fill_ifinfo(struct sk_buff *skb, struct net_device *dev, ...@@ -941,10 +943,11 @@ static int rtnl_fill_ifinfo(struct sk_buff *skb, struct net_device *dev,
goto nla_put_failure; goto nla_put_failure;
copy_rtnl_link_stats64(nla_data(attr), stats); copy_rtnl_link_stats64(nla_data(attr), stats);
if (dev->dev.parent) if (dev->dev.parent && (ext_filter_mask & RTEXT_FILTER_VF))
NLA_PUT_U32(skb, IFLA_NUM_VF, dev_num_vf(dev->dev.parent)); NLA_PUT_U32(skb, IFLA_NUM_VF, dev_num_vf(dev->dev.parent));
if (dev->netdev_ops->ndo_get_vf_config && dev->dev.parent) { if (dev->netdev_ops->ndo_get_vf_config && dev->dev.parent
&& (ext_filter_mask & RTEXT_FILTER_VF)) {
int i; int i;
struct nlattr *vfinfo, *vf; struct nlattr *vfinfo, *vf;
...@@ -1048,6 +1051,8 @@ static int rtnl_dump_ifinfo(struct sk_buff *skb, struct netlink_callback *cb) ...@@ -1048,6 +1051,8 @@ static int rtnl_dump_ifinfo(struct sk_buff *skb, struct netlink_callback *cb)
struct net_device *dev; struct net_device *dev;
struct hlist_head *head; struct hlist_head *head;
struct hlist_node *node; struct hlist_node *node;
struct nlattr *tb[IFLA_MAX+1];
u32 ext_filter_mask = 0;
s_h = cb->args[0]; s_h = cb->args[0];
s_idx = cb->args[1]; s_idx = cb->args[1];
...@@ -1055,6 +1060,12 @@ static int rtnl_dump_ifinfo(struct sk_buff *skb, struct netlink_callback *cb) ...@@ -1055,6 +1060,12 @@ static int rtnl_dump_ifinfo(struct sk_buff *skb, struct netlink_callback *cb)
rcu_read_lock(); rcu_read_lock();
cb->seq = net->dev_base_seq; cb->seq = net->dev_base_seq;
nlmsg_parse(cb->nlh, sizeof(struct rtgenmsg), tb, IFLA_MAX,
ifla_policy);
if (tb[IFLA_EXT_MASK])
ext_filter_mask = nla_get_u32(tb[IFLA_EXT_MASK]);
for (h = s_h; h < NETDEV_HASHENTRIES; h++, s_idx = 0) { for (h = s_h; h < NETDEV_HASHENTRIES; h++, s_idx = 0) {
idx = 0; idx = 0;
head = &net->dev_index_head[h]; head = &net->dev_index_head[h];
...@@ -1064,7 +1075,8 @@ static int rtnl_dump_ifinfo(struct sk_buff *skb, struct netlink_callback *cb) ...@@ -1064,7 +1075,8 @@ static int rtnl_dump_ifinfo(struct sk_buff *skb, struct netlink_callback *cb)
if (rtnl_fill_ifinfo(skb, dev, RTM_NEWLINK, if (rtnl_fill_ifinfo(skb, dev, RTM_NEWLINK,
NETLINK_CB(cb->skb).pid, NETLINK_CB(cb->skb).pid,
cb->nlh->nlmsg_seq, 0, cb->nlh->nlmsg_seq, 0,
NLM_F_MULTI) <= 0) NLM_F_MULTI,
ext_filter_mask) <= 0)
goto out; goto out;
nl_dump_check_consistent(cb, nlmsg_hdr(skb)); nl_dump_check_consistent(cb, nlmsg_hdr(skb));
...@@ -1100,6 +1112,7 @@ const struct nla_policy ifla_policy[IFLA_MAX+1] = { ...@@ -1100,6 +1112,7 @@ const struct nla_policy ifla_policy[IFLA_MAX+1] = {
[IFLA_VF_PORTS] = { .type = NLA_NESTED }, [IFLA_VF_PORTS] = { .type = NLA_NESTED },
[IFLA_PORT_SELF] = { .type = NLA_NESTED }, [IFLA_PORT_SELF] = { .type = NLA_NESTED },
[IFLA_AF_SPEC] = { .type = NLA_NESTED }, [IFLA_AF_SPEC] = { .type = NLA_NESTED },
[IFLA_EXT_MASK] = { .type = NLA_U32 },
}; };
EXPORT_SYMBOL(ifla_policy); EXPORT_SYMBOL(ifla_policy);
...@@ -1509,8 +1522,6 @@ static int do_setlink(struct net_device *dev, struct ifinfomsg *ifm, ...@@ -1509,8 +1522,6 @@ static int do_setlink(struct net_device *dev, struct ifinfomsg *ifm,
if (send_addr_notify) if (send_addr_notify)
call_netdevice_notifiers(NETDEV_CHANGEADDR, dev); call_netdevice_notifiers(NETDEV_CHANGEADDR, dev);
min_ifinfo_dump_size = max_t(u16, if_nlmsg_size(dev),
min_ifinfo_dump_size);
return err; return err;
} }
...@@ -1842,6 +1853,7 @@ static int rtnl_getlink(struct sk_buff *skb, struct nlmsghdr* nlh, void *arg) ...@@ -1842,6 +1853,7 @@ static int rtnl_getlink(struct sk_buff *skb, struct nlmsghdr* nlh, void *arg)
struct net_device *dev = NULL; struct net_device *dev = NULL;
struct sk_buff *nskb; struct sk_buff *nskb;
int err; int err;
u32 ext_filter_mask = 0;
err = nlmsg_parse(nlh, sizeof(*ifm), tb, IFLA_MAX, ifla_policy); err = nlmsg_parse(nlh, sizeof(*ifm), tb, IFLA_MAX, ifla_policy);
if (err < 0) if (err < 0)
...@@ -1850,6 +1862,9 @@ static int rtnl_getlink(struct sk_buff *skb, struct nlmsghdr* nlh, void *arg) ...@@ -1850,6 +1862,9 @@ static int rtnl_getlink(struct sk_buff *skb, struct nlmsghdr* nlh, void *arg)
if (tb[IFLA_IFNAME]) if (tb[IFLA_IFNAME])
nla_strlcpy(ifname, tb[IFLA_IFNAME], IFNAMSIZ); nla_strlcpy(ifname, tb[IFLA_IFNAME], IFNAMSIZ);
if (tb[IFLA_EXT_MASK])
ext_filter_mask = nla_get_u32(tb[IFLA_EXT_MASK]);
ifm = nlmsg_data(nlh); ifm = nlmsg_data(nlh);
if (ifm->ifi_index > 0) if (ifm->ifi_index > 0)
dev = __dev_get_by_index(net, ifm->ifi_index); dev = __dev_get_by_index(net, ifm->ifi_index);
...@@ -1861,12 +1876,12 @@ static int rtnl_getlink(struct sk_buff *skb, struct nlmsghdr* nlh, void *arg) ...@@ -1861,12 +1876,12 @@ static int rtnl_getlink(struct sk_buff *skb, struct nlmsghdr* nlh, void *arg)
if (dev == NULL) if (dev == NULL)
return -ENODEV; return -ENODEV;
nskb = nlmsg_new(if_nlmsg_size(dev), GFP_KERNEL); nskb = nlmsg_new(if_nlmsg_size(dev, ext_filter_mask), GFP_KERNEL);
if (nskb == NULL) if (nskb == NULL)
return -ENOBUFS; return -ENOBUFS;
err = rtnl_fill_ifinfo(nskb, dev, RTM_NEWLINK, NETLINK_CB(skb).pid, err = rtnl_fill_ifinfo(nskb, dev, RTM_NEWLINK, NETLINK_CB(skb).pid,
nlh->nlmsg_seq, 0, 0); nlh->nlmsg_seq, 0, 0, ext_filter_mask);
if (err < 0) { if (err < 0) {
/* -EMSGSIZE implies BUG in if_nlmsg_size */ /* -EMSGSIZE implies BUG in if_nlmsg_size */
WARN_ON(err == -EMSGSIZE); WARN_ON(err == -EMSGSIZE);
...@@ -1877,8 +1892,31 @@ static int rtnl_getlink(struct sk_buff *skb, struct nlmsghdr* nlh, void *arg) ...@@ -1877,8 +1892,31 @@ static int rtnl_getlink(struct sk_buff *skb, struct nlmsghdr* nlh, void *arg)
return err; return err;
} }
static u16 rtnl_calcit(struct sk_buff *skb) static u16 rtnl_calcit(struct sk_buff *skb, struct nlmsghdr *nlh)
{ {
struct net *net = sock_net(skb->sk);
struct net_device *dev;
struct nlattr *tb[IFLA_MAX+1];
u32 ext_filter_mask = 0;
u16 min_ifinfo_dump_size = 0;
nlmsg_parse(nlh, sizeof(struct rtgenmsg), tb, IFLA_MAX, ifla_policy);
if (tb[IFLA_EXT_MASK])
ext_filter_mask = nla_get_u32(tb[IFLA_EXT_MASK]);
if (!ext_filter_mask)
return NLMSG_GOODSIZE;
/*
* traverse the list of net devices and compute the minimum
* buffer size based upon the filter mask.
*/
list_for_each_entry(dev, &net->dev_base_head, dev_list) {
min_ifinfo_dump_size = max_t(u16, min_ifinfo_dump_size,
if_nlmsg_size(dev,
ext_filter_mask));
}
return min_ifinfo_dump_size; return min_ifinfo_dump_size;
} }
...@@ -1913,13 +1951,11 @@ void rtmsg_ifinfo(int type, struct net_device *dev, unsigned change) ...@@ -1913,13 +1951,11 @@ void rtmsg_ifinfo(int type, struct net_device *dev, unsigned change)
int err = -ENOBUFS; int err = -ENOBUFS;
size_t if_info_size; size_t if_info_size;
skb = nlmsg_new((if_info_size = if_nlmsg_size(dev)), GFP_KERNEL); skb = nlmsg_new((if_info_size = if_nlmsg_size(dev, 0)), GFP_KERNEL);
if (skb == NULL) if (skb == NULL)
goto errout; goto errout;
min_ifinfo_dump_size = max_t(u16, if_info_size, min_ifinfo_dump_size); err = rtnl_fill_ifinfo(skb, dev, type, 0, 0, change, 0, 0);
err = rtnl_fill_ifinfo(skb, dev, type, 0, 0, change, 0);
if (err < 0) { if (err < 0) {
/* -EMSGSIZE implies BUG in if_nlmsg_size() */ /* -EMSGSIZE implies BUG in if_nlmsg_size() */
WARN_ON(err == -EMSGSIZE); WARN_ON(err == -EMSGSIZE);
...@@ -1977,7 +2013,7 @@ static int rtnetlink_rcv_msg(struct sk_buff *skb, struct nlmsghdr *nlh) ...@@ -1977,7 +2013,7 @@ static int rtnetlink_rcv_msg(struct sk_buff *skb, struct nlmsghdr *nlh)
return -EOPNOTSUPP; return -EOPNOTSUPP;
calcit = rtnl_get_calcit(family, type); calcit = rtnl_get_calcit(family, type);
if (calcit) if (calcit)
min_dump_alloc = calcit(skb); min_dump_alloc = calcit(skb, nlh);
__rtnl_unlock(); __rtnl_unlock();
rtnl = net->rtnl; rtnl = net->rtnl;
......
Markdown is supported
0%
or
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment