summaryrefslogtreecommitdiff
path: root/net/sched
diff options
context:
space:
mode:
Diffstat (limited to 'net/sched')
-rw-r--r--net/sched/act_bpf.c1
-rw-r--r--net/sched/act_connmark.c5
-rw-r--r--net/sched/act_ipt.c1
-rw-r--r--net/sched/cls_bpf.c82
-rw-r--r--net/sched/cls_flow.c15
-rw-r--r--net/sched/em_ipset.c1
-rw-r--r--net/sched/em_meta.c138
-rw-r--r--net/sched/sch_api.c2
-rw-r--r--net/sched/sch_blackhole.c15
-rw-r--r--net/sched/sch_choke.c59
-rw-r--r--net/sched/sch_dsmark.c63
-rw-r--r--net/sched/sch_fq.c13
-rw-r--r--net/sched/sch_generic.c4
13 files changed, 220 insertions, 179 deletions
diff --git a/net/sched/act_bpf.c b/net/sched/act_bpf.c
index 559bfa011..0bc6f912f 100644
--- a/net/sched/act_bpf.c
+++ b/net/sched/act_bpf.c
@@ -72,6 +72,7 @@ static int tcf_bpf(struct sk_buff *skb, const struct tc_action *act,
case TC_ACT_PIPE:
case TC_ACT_RECLASSIFY:
case TC_ACT_OK:
+ case TC_ACT_REDIRECT:
action = filter_res;
break;
case TC_ACT_SHOT:
diff --git a/net/sched/act_connmark.c b/net/sched/act_connmark.c
index 5019a47b9..bb41699c6 100644
--- a/net/sched/act_connmark.c
+++ b/net/sched/act_connmark.c
@@ -68,13 +68,13 @@ static int tcf_connmark(struct sk_buff *skb, const struct tc_action *a,
}
if (!nf_ct_get_tuplepr(skb, skb_network_offset(skb),
- proto, &tuple))
+ proto, ca->net, &tuple))
goto out;
zone.id = ca->zone;
zone.dir = NF_CT_DEFAULT_ZONE_DIR;
- thash = nf_conntrack_find_get(dev_net(skb->dev), &zone, &tuple);
+ thash = nf_conntrack_find_get(ca->net, &zone, &tuple);
if (!thash)
goto out;
@@ -119,6 +119,7 @@ static int tcf_connmark_init(struct net *net, struct nlattr *nla,
ci = to_connmark(a);
ci->tcf_action = parm->action;
+ ci->net = net;
ci->zone = parm->zone;
tcf_hash_insert(a);
diff --git a/net/sched/act_ipt.c b/net/sched/act_ipt.c
index 99c9cc1c7..d05869646 100644
--- a/net/sched/act_ipt.c
+++ b/net/sched/act_ipt.c
@@ -189,6 +189,7 @@ static int tcf_ipt(struct sk_buff *skb, const struct tc_action *a,
* worry later - danger - this API seems to have changed
* from earlier kernels
*/
+ par.net = dev_net(skb->dev);
par.in = skb->dev;
par.out = NULL;
par.hooknum = ipt->tcfi_hook;
diff --git a/net/sched/cls_bpf.c b/net/sched/cls_bpf.c
index e5168f8b9..5faaa5425 100644
--- a/net/sched/cls_bpf.c
+++ b/net/sched/cls_bpf.c
@@ -38,6 +38,7 @@ struct cls_bpf_prog {
struct bpf_prog *filter;
struct list_head link;
struct tcf_result res;
+ bool exts_integrated;
struct tcf_exts exts;
u32 handle;
union {
@@ -52,6 +53,7 @@ struct cls_bpf_prog {
static const struct nla_policy bpf_policy[TCA_BPF_MAX + 1] = {
[TCA_BPF_CLASSID] = { .type = NLA_U32 },
+ [TCA_BPF_FLAGS] = { .type = NLA_U32 },
[TCA_BPF_FD] = { .type = NLA_U32 },
[TCA_BPF_NAME] = { .type = NLA_NUL_STRING, .len = CLS_BPF_NAME_LEN },
[TCA_BPF_OPS_LEN] = { .type = NLA_U16 },
@@ -59,6 +61,20 @@ static const struct nla_policy bpf_policy[TCA_BPF_MAX + 1] = {
.len = sizeof(struct sock_filter) * BPF_MAXINSNS },
};
+static int cls_bpf_exec_opcode(int code)
+{
+ switch (code) {
+ case TC_ACT_OK:
+ case TC_ACT_SHOT:
+ case TC_ACT_STOLEN:
+ case TC_ACT_REDIRECT:
+ case TC_ACT_UNSPEC:
+ return code;
+ default:
+ return TC_ACT_UNSPEC;
+ }
+}
+
static int cls_bpf_classify(struct sk_buff *skb, const struct tcf_proto *tp,
struct tcf_result *res)
{
@@ -79,6 +95,8 @@ static int cls_bpf_classify(struct sk_buff *skb, const struct tcf_proto *tp,
list_for_each_entry_rcu(prog, &head->plist, link) {
int filter_res;
+ qdisc_skb_cb(skb)->tc_classid = prog->res.classid;
+
if (at_ingress) {
/* It is safe to push/pull even if skb_shared() */
__skb_push(skb, skb->mac_len);
@@ -88,6 +106,16 @@ static int cls_bpf_classify(struct sk_buff *skb, const struct tcf_proto *tp,
filter_res = BPF_PROG_RUN(prog->filter, skb);
}
+ if (prog->exts_integrated) {
+ res->class = prog->res.class;
+ res->classid = qdisc_skb_cb(skb)->tc_classid;
+
+ ret = cls_bpf_exec_opcode(filter_res);
+ if (ret == TC_ACT_UNSPEC)
+ continue;
+ break;
+ }
+
if (filter_res == 0)
continue;
@@ -195,8 +223,7 @@ static unsigned long cls_bpf_get(struct tcf_proto *tp, u32 handle)
return ret;
}
-static int cls_bpf_prog_from_ops(struct nlattr **tb,
- struct cls_bpf_prog *prog, u32 classid)
+static int cls_bpf_prog_from_ops(struct nlattr **tb, struct cls_bpf_prog *prog)
{
struct sock_filter *bpf_ops;
struct sock_fprog_kern fprog_tmp;
@@ -230,15 +257,13 @@ static int cls_bpf_prog_from_ops(struct nlattr **tb,
prog->bpf_ops = bpf_ops;
prog->bpf_num_ops = bpf_num_ops;
prog->bpf_name = NULL;
-
prog->filter = fp;
- prog->res.classid = classid;
return 0;
}
-static int cls_bpf_prog_from_efd(struct nlattr **tb,
- struct cls_bpf_prog *prog, u32 classid)
+static int cls_bpf_prog_from_efd(struct nlattr **tb, struct cls_bpf_prog *prog,
+ const struct tcf_proto *tp)
{
struct bpf_prog *fp;
char *name = NULL;
@@ -268,9 +293,10 @@ static int cls_bpf_prog_from_efd(struct nlattr **tb,
prog->bpf_ops = NULL;
prog->bpf_fd = bpf_fd;
prog->bpf_name = name;
-
prog->filter = fp;
- prog->res.classid = classid;
+
+ if (fp->dst_needed)
+ netif_keep_dst(qdisc_dev(tp->q));
return 0;
}
@@ -280,16 +306,13 @@ static int cls_bpf_modify_existing(struct net *net, struct tcf_proto *tp,
unsigned long base, struct nlattr **tb,
struct nlattr *est, bool ovr)
{
+ bool is_bpf, is_ebpf, have_exts = false;
struct tcf_exts exts;
- bool is_bpf, is_ebpf;
- u32 classid;
int ret;
is_bpf = tb[TCA_BPF_OPS_LEN] && tb[TCA_BPF_OPS];
is_ebpf = tb[TCA_BPF_FD];
-
- if ((!is_bpf && !is_ebpf) || (is_bpf && is_ebpf) ||
- !tb[TCA_BPF_CLASSID])
+ if ((!is_bpf && !is_ebpf) || (is_bpf && is_ebpf))
return -EINVAL;
tcf_exts_init(&exts, TCA_BPF_ACT, TCA_BPF_POLICE);
@@ -297,18 +320,32 @@ static int cls_bpf_modify_existing(struct net *net, struct tcf_proto *tp,
if (ret < 0)
return ret;
- classid = nla_get_u32(tb[TCA_BPF_CLASSID]);
+ if (tb[TCA_BPF_FLAGS]) {
+ u32 bpf_flags = nla_get_u32(tb[TCA_BPF_FLAGS]);
+
+ if (bpf_flags & ~TCA_BPF_FLAG_ACT_DIRECT) {
+ tcf_exts_destroy(&exts);
+ return -EINVAL;
+ }
+
+ have_exts = bpf_flags & TCA_BPF_FLAG_ACT_DIRECT;
+ }
+
+ prog->exts_integrated = have_exts;
- ret = is_bpf ? cls_bpf_prog_from_ops(tb, prog, classid) :
- cls_bpf_prog_from_efd(tb, prog, classid);
+ ret = is_bpf ? cls_bpf_prog_from_ops(tb, prog) :
+ cls_bpf_prog_from_efd(tb, prog, tp);
if (ret < 0) {
tcf_exts_destroy(&exts);
return ret;
}
- tcf_bind_filter(tp, &prog->res, base);
- tcf_exts_change(tp, &prog->exts, &exts);
+ if (tb[TCA_BPF_CLASSID]) {
+ prog->res.classid = nla_get_u32(tb[TCA_BPF_CLASSID]);
+ tcf_bind_filter(tp, &prog->res, base);
+ }
+ tcf_exts_change(tp, &prog->exts, &exts);
return 0;
}
@@ -429,6 +466,7 @@ static int cls_bpf_dump(struct net *net, struct tcf_proto *tp, unsigned long fh,
{
struct cls_bpf_prog *prog = (struct cls_bpf_prog *) fh;
struct nlattr *nest;
+ u32 bpf_flags = 0;
int ret;
if (prog == NULL)
@@ -440,7 +478,8 @@ static int cls_bpf_dump(struct net *net, struct tcf_proto *tp, unsigned long fh,
if (nest == NULL)
goto nla_put_failure;
- if (nla_put_u32(skb, TCA_BPF_CLASSID, prog->res.classid))
+ if (prog->res.classid &&
+ nla_put_u32(skb, TCA_BPF_CLASSID, prog->res.classid))
goto nla_put_failure;
if (cls_bpf_is_ebpf(prog))
@@ -453,6 +492,11 @@ static int cls_bpf_dump(struct net *net, struct tcf_proto *tp, unsigned long fh,
if (tcf_exts_dump(skb, &prog->exts) < 0)
goto nla_put_failure;
+ if (prog->exts_integrated)
+ bpf_flags |= TCA_BPF_FLAG_ACT_DIRECT;
+ if (bpf_flags && nla_put_u32(skb, TCA_BPF_FLAGS, bpf_flags))
+ goto nla_put_failure;
+
nla_nest_end(skb, nest);
if (tcf_exts_dump_stats(skb, &prog->exts) < 0)
diff --git a/net/sched/cls_flow.c b/net/sched/cls_flow.c
index 536838b65..fbfec6a18 100644
--- a/net/sched/cls_flow.c
+++ b/net/sched/cls_flow.c
@@ -22,6 +22,7 @@
#include <linux/if_vlan.h>
#include <linux/slab.h>
#include <linux/module.h>
+#include <net/inet_sock.h>
#include <net/pkt_cls.h>
#include <net/ip.h>
@@ -197,8 +198,11 @@ static u32 flow_get_rtclassid(const struct sk_buff *skb)
static u32 flow_get_skuid(const struct sk_buff *skb)
{
- if (skb->sk && skb->sk->sk_socket && skb->sk->sk_socket->file) {
- kuid_t skuid = skb->sk->sk_socket->file->f_cred->fsuid;
+ struct sock *sk = skb_to_full_sk(skb);
+
+ if (sk && sk->sk_socket && sk->sk_socket->file) {
+ kuid_t skuid = sk->sk_socket->file->f_cred->fsuid;
+
return from_kuid(&init_user_ns, skuid);
}
return 0;
@@ -206,8 +210,11 @@ static u32 flow_get_skuid(const struct sk_buff *skb)
static u32 flow_get_skgid(const struct sk_buff *skb)
{
- if (skb->sk && skb->sk->sk_socket && skb->sk->sk_socket->file) {
- kgid_t skgid = skb->sk->sk_socket->file->f_cred->fsgid;
+ struct sock *sk = skb_to_full_sk(skb);
+
+ if (sk && sk->sk_socket && sk->sk_socket->file) {
+ kgid_t skgid = sk->sk_socket->file->f_cred->fsgid;
+
return from_kgid(&init_user_ns, skgid);
}
return 0;
diff --git a/net/sched/em_ipset.c b/net/sched/em_ipset.c
index df0328ba6..c66ca9400 100644
--- a/net/sched/em_ipset.c
+++ b/net/sched/em_ipset.c
@@ -95,6 +95,7 @@ static int em_ipset_match(struct sk_buff *skb, struct tcf_ematch *em,
if (skb->skb_iif)
indev = dev_get_by_index_rcu(em->net, skb->skb_iif);
+ acpar.net = em->net;
acpar.in = indev ? indev : dev;
acpar.out = dev;
diff --git a/net/sched/em_meta.c b/net/sched/em_meta.c
index b5294ce20..f2aabc008 100644
--- a/net/sched/em_meta.c
+++ b/net/sched/em_meta.c
@@ -343,119 +343,145 @@ META_COLLECTOR(int_sk_refcnt)
META_COLLECTOR(int_sk_rcvbuf)
{
- if (skip_nonlocal(skb)) {
+ const struct sock *sk = skb_to_full_sk(skb);
+
+ if (!sk) {
*err = -1;
return;
}
- dst->value = skb->sk->sk_rcvbuf;
+ dst->value = sk->sk_rcvbuf;
}
META_COLLECTOR(int_sk_shutdown)
{
- if (skip_nonlocal(skb)) {
+ const struct sock *sk = skb_to_full_sk(skb);
+
+ if (!sk) {
*err = -1;
return;
}
- dst->value = skb->sk->sk_shutdown;
+ dst->value = sk->sk_shutdown;
}
META_COLLECTOR(int_sk_proto)
{
- if (skip_nonlocal(skb)) {
+ const struct sock *sk = skb_to_full_sk(skb);
+
+ if (!sk) {
*err = -1;
return;
}
- dst->value = skb->sk->sk_protocol;
+ dst->value = sk->sk_protocol;
}
META_COLLECTOR(int_sk_type)
{
- if (skip_nonlocal(skb)) {
+ const struct sock *sk = skb_to_full_sk(skb);
+
+ if (!sk) {
*err = -1;
return;
}
- dst->value = skb->sk->sk_type;
+ dst->value = sk->sk_type;
}
META_COLLECTOR(int_sk_rmem_alloc)
{
- if (skip_nonlocal(skb)) {
+ const struct sock *sk = skb_to_full_sk(skb);
+
+ if (!sk) {
*err = -1;
return;
}
- dst->value = sk_rmem_alloc_get(skb->sk);
+ dst->value = sk_rmem_alloc_get(sk);
}
META_COLLECTOR(int_sk_wmem_alloc)
{
- if (skip_nonlocal(skb)) {
+ const struct sock *sk = skb_to_full_sk(skb);
+
+ if (!sk) {
*err = -1;
return;
}
- dst->value = sk_wmem_alloc_get(skb->sk);
+ dst->value = sk_wmem_alloc_get(sk);
}
META_COLLECTOR(int_sk_omem_alloc)
{
- if (skip_nonlocal(skb)) {
+ const struct sock *sk = skb_to_full_sk(skb);
+
+ if (!sk) {
*err = -1;
return;
}
- dst->value = atomic_read(&skb->sk->sk_omem_alloc);
+ dst->value = atomic_read(&sk->sk_omem_alloc);
}
META_COLLECTOR(int_sk_rcv_qlen)
{
- if (skip_nonlocal(skb)) {
+ const struct sock *sk = skb_to_full_sk(skb);
+
+ if (!sk) {
*err = -1;
return;
}
- dst->value = skb->sk->sk_receive_queue.qlen;
+ dst->value = sk->sk_receive_queue.qlen;
}
META_COLLECTOR(int_sk_snd_qlen)
{
- if (skip_nonlocal(skb)) {
+ const struct sock *sk = skb_to_full_sk(skb);
+
+ if (!sk) {
*err = -1;
return;
}
- dst->value = skb->sk->sk_write_queue.qlen;
+ dst->value = sk->sk_write_queue.qlen;
}
META_COLLECTOR(int_sk_wmem_queued)
{
- if (skip_nonlocal(skb)) {
+ const struct sock *sk = skb_to_full_sk(skb);
+
+ if (!sk) {
*err = -1;
return;
}
- dst->value = skb->sk->sk_wmem_queued;
+ dst->value = sk->sk_wmem_queued;
}
META_COLLECTOR(int_sk_fwd_alloc)
{
- if (skip_nonlocal(skb)) {
+ const struct sock *sk = skb_to_full_sk(skb);
+
+ if (!sk) {
*err = -1;
return;
}
- dst->value = skb->sk->sk_forward_alloc;
+ dst->value = sk->sk_forward_alloc;
}
META_COLLECTOR(int_sk_sndbuf)
{
- if (skip_nonlocal(skb)) {
+ const struct sock *sk = skb_to_full_sk(skb);
+
+ if (!sk) {
*err = -1;
return;
}
- dst->value = skb->sk->sk_sndbuf;
+ dst->value = sk->sk_sndbuf;
}
META_COLLECTOR(int_sk_alloc)
{
- if (skip_nonlocal(skb)) {
+ const struct sock *sk = skb_to_full_sk(skb);
+
+ if (!sk) {
*err = -1;
return;
}
- dst->value = (__force int) skb->sk->sk_allocation;
+ dst->value = (__force int) sk->sk_allocation;
}
META_COLLECTOR(int_sk_hash)
@@ -469,92 +495,112 @@ META_COLLECTOR(int_sk_hash)
META_COLLECTOR(int_sk_lingertime)
{
- if (skip_nonlocal(skb)) {
+ const struct sock *sk = skb_to_full_sk(skb);
+
+ if (!sk) {
*err = -1;
return;
}
- dst->value = skb->sk->sk_lingertime / HZ;
+ dst->value = sk->sk_lingertime / HZ;
}
META_COLLECTOR(int_sk_err_qlen)
{
- if (skip_nonlocal(skb)) {
+ const struct sock *sk = skb_to_full_sk(skb);
+
+ if (!sk) {
*err = -1;
return;
}
- dst->value = skb->sk->sk_error_queue.qlen;
+ dst->value = sk->sk_error_queue.qlen;
}
META_COLLECTOR(int_sk_ack_bl)
{
- if (skip_nonlocal(skb)) {
+ const struct sock *sk = skb_to_full_sk(skb);
+
+ if (!sk) {
*err = -1;
return;
}
- dst->value = skb->sk->sk_ack_backlog;
+ dst->value = sk->sk_ack_backlog;
}
META_COLLECTOR(int_sk_max_ack_bl)
{
- if (skip_nonlocal(skb)) {
+ const struct sock *sk = skb_to_full_sk(skb);
+
+ if (!sk) {
*err = -1;
return;
}
- dst->value = skb->sk->sk_max_ack_backlog;
+ dst->value = sk->sk_max_ack_backlog;
}
META_COLLECTOR(int_sk_prio)
{
- if (skip_nonlocal(skb)) {
+ const struct sock *sk = skb_to_full_sk(skb);
+
+ if (!sk) {
*err = -1;
return;
}
- dst->value = skb->sk->sk_priority;
+ dst->value = sk->sk_priority;
}
META_COLLECTOR(int_sk_rcvlowat)
{
- if (skip_nonlocal(skb)) {
+ const struct sock *sk = skb_to_full_sk(skb);
+
+ if (!sk) {
*err = -1;
return;
}
- dst->value = skb->sk->sk_rcvlowat;
+ dst->value = sk->sk_rcvlowat;
}
META_COLLECTOR(int_sk_rcvtimeo)
{
- if (skip_nonlocal(skb)) {
+ const struct sock *sk = skb_to_full_sk(skb);
+
+ if (!sk) {
*err = -1;
return;
}
- dst->value = skb->sk->sk_rcvtimeo / HZ;
+ dst->value = sk->sk_rcvtimeo / HZ;
}
META_COLLECTOR(int_sk_sndtimeo)
{
- if (skip_nonlocal(skb)) {
+ const struct sock *sk = skb_to_full_sk(skb);
+
+ if (!sk) {
*err = -1;
return;
}
- dst->value = skb->sk->sk_sndtimeo / HZ;
+ dst->value = sk->sk_sndtimeo / HZ;
}
META_COLLECTOR(int_sk_sendmsg_off)
{
- if (skip_nonlocal(skb)) {
+ const struct sock *sk = skb_to_full_sk(skb);
+
+ if (!sk) {
*err = -1;
return;
}
- dst->value = skb->sk->sk_frag.offset;
+ dst->value = sk->sk_frag.offset;
}
META_COLLECTOR(int_sk_write_pend)
{
- if (skip_nonlocal(skb)) {
+ const struct sock *sk = skb_to_full_sk(skb);
+
+ if (!sk) {
*err = -1;
return;
}
- dst->value = skb->sk->sk_write_pending;
+ dst->value = sk->sk_write_pending;
}
/**************************************************************************
diff --git a/net/sched/sch_api.c b/net/sched/sch_api.c
index 7ec667dd4..b5c2cf2aa 100644
--- a/net/sched/sch_api.c
+++ b/net/sched/sch_api.c
@@ -950,7 +950,7 @@ qdisc_create(struct net_device *dev, struct netdev_queue *dev_queue,
}
lockdep_set_class(qdisc_lock(sch), &qdisc_tx_lock);
if (!netif_is_multiqueue(dev))
- sch->flags |= TCQ_F_ONETXQUEUE | TCQ_F_NOPARENT;
+ sch->flags |= TCQ_F_ONETXQUEUE;
}
sch->handle = handle;
diff --git a/net/sched/sch_blackhole.c b/net/sched/sch_blackhole.c
index 094a874b4..3fee70d98 100644
--- a/net/sched/sch_blackhole.c
+++ b/net/sched/sch_blackhole.c
@@ -11,7 +11,7 @@
* Note: Quantum tunneling is not supported.
*/
-#include <linux/module.h>
+#include <linux/init.h>
#include <linux/types.h>
#include <linux/kernel.h>
#include <linux/skbuff.h>
@@ -37,17 +37,8 @@ static struct Qdisc_ops blackhole_qdisc_ops __read_mostly = {
.owner = THIS_MODULE,
};
-static int __init blackhole_module_init(void)
+static int __init blackhole_init(void)
{
return register_qdisc(&blackhole_qdisc_ops);
}
-
-static void __exit blackhole_module_exit(void)
-{
- unregister_qdisc(&blackhole_qdisc_ops);
-}
-
-module_init(blackhole_module_init)
-module_exit(blackhole_module_exit)
-
-MODULE_LICENSE("GPL");
+device_initcall(blackhole_init)
diff --git a/net/sched/sch_choke.c b/net/sched/sch_choke.c
index 02bfd3d1c..5ffb8b833 100644
--- a/net/sched/sch_choke.c
+++ b/net/sched/sch_choke.c
@@ -553,65 +553,6 @@ static void choke_destroy(struct Qdisc *sch)
choke_free(q->tab);
}
-static struct Qdisc *choke_leaf(struct Qdisc *sch, unsigned long arg)
-{
- return NULL;
-}
-
-static unsigned long choke_get(struct Qdisc *sch, u32 classid)
-{
- return 0;
-}
-
-static void choke_put(struct Qdisc *q, unsigned long cl)
-{
-}
-
-static unsigned long choke_bind(struct Qdisc *sch, unsigned long parent,
- u32 classid)
-{
- return 0;
-}
-
-static struct tcf_proto __rcu **choke_find_tcf(struct Qdisc *sch,
- unsigned long cl)
-{
- struct choke_sched_data *q = qdisc_priv(sch);
-
- if (cl)
- return NULL;
- return &q->filter_list;
-}
-
-static int choke_dump_class(struct Qdisc *sch, unsigned long cl,
- struct sk_buff *skb, struct tcmsg *tcm)
-{
- tcm->tcm_handle |= TC_H_MIN(cl);
- return 0;
-}
-
-static void choke_walk(struct Qdisc *sch, struct qdisc_walker *arg)
-{
- if (!arg->stop) {
- if (arg->fn(sch, 1, arg) < 0) {
- arg->stop = 1;
- return;
- }
- arg->count++;
- }
-}
-
-static const struct Qdisc_class_ops choke_class_ops = {
- .leaf = choke_leaf,
- .get = choke_get,
- .put = choke_put,
- .tcf_chain = choke_find_tcf,
- .bind_tcf = choke_bind,
- .unbind_tcf = choke_put,
- .dump = choke_dump_class,
- .walk = choke_walk,
-};
-
static struct sk_buff *choke_peek_head(struct Qdisc *sch)
{
struct choke_sched_data *q = qdisc_priv(sch);
diff --git a/net/sched/sch_dsmark.c b/net/sched/sch_dsmark.c
index c4d45fd8c..f357f34d0 100644
--- a/net/sched/sch_dsmark.c
+++ b/net/sched/sch_dsmark.c
@@ -35,14 +35,20 @@
#define NO_DEFAULT_INDEX (1 << 16)
+struct mask_value {
+ u8 mask;
+ u8 value;
+};
+
struct dsmark_qdisc_data {
struct Qdisc *q;
struct tcf_proto __rcu *filter_list;
- u8 *mask; /* "owns" the array */
- u8 *value;
+ struct mask_value *mv;
u16 indices;
+ u8 set_tc_index;
u32 default_index; /* index range is 0...0xffff */
- int set_tc_index;
+#define DSMARK_EMBEDDED_SZ 16
+ struct mask_value embedded[DSMARK_EMBEDDED_SZ];
};
static inline int dsmark_valid_index(struct dsmark_qdisc_data *p, u16 index)
@@ -116,7 +122,6 @@ static int dsmark_change(struct Qdisc *sch, u32 classid, u32 parent,
struct nlattr *opt = tca[TCA_OPTIONS];
struct nlattr *tb[TCA_DSMARK_MAX + 1];
int err = -EINVAL;
- u8 mask = 0;
pr_debug("%s(sch %p,[qdisc %p],classid %x,parent %x), arg 0x%lx\n",
__func__, sch, p, classid, parent, *arg);
@@ -133,14 +138,11 @@ static int dsmark_change(struct Qdisc *sch, u32 classid, u32 parent,
if (err < 0)
goto errout;
- if (tb[TCA_DSMARK_MASK])
- mask = nla_get_u8(tb[TCA_DSMARK_MASK]);
-
if (tb[TCA_DSMARK_VALUE])
- p->value[*arg - 1] = nla_get_u8(tb[TCA_DSMARK_VALUE]);
+ p->mv[*arg - 1].value = nla_get_u8(tb[TCA_DSMARK_VALUE]);
if (tb[TCA_DSMARK_MASK])
- p->mask[*arg - 1] = mask;
+ p->mv[*arg - 1].mask = nla_get_u8(tb[TCA_DSMARK_MASK]);
err = 0;
@@ -155,8 +157,8 @@ static int dsmark_delete(struct Qdisc *sch, unsigned long arg)
if (!dsmark_valid_index(p, arg))
return -EINVAL;
- p->mask[arg - 1] = 0xff;
- p->value[arg - 1] = 0;
+ p->mv[arg - 1].mask = 0xff;
+ p->mv[arg - 1].value = 0;
return 0;
}
@@ -173,7 +175,7 @@ static void dsmark_walk(struct Qdisc *sch, struct qdisc_walker *walker)
return;
for (i = 0; i < p->indices; i++) {
- if (p->mask[i] == 0xff && !p->value[i])
+ if (p->mv[i].mask == 0xff && !p->mv[i].value)
goto ignore;
if (walker->count >= walker->skip) {
if (walker->fn(sch, i + 1, walker) < 0) {
@@ -291,12 +293,12 @@ static struct sk_buff *dsmark_dequeue(struct Qdisc *sch)
switch (tc_skb_protocol(skb)) {
case htons(ETH_P_IP):
- ipv4_change_dsfield(ip_hdr(skb), p->mask[index],
- p->value[index]);
+ ipv4_change_dsfield(ip_hdr(skb), p->mv[index].mask,
+ p->mv[index].value);
break;
case htons(ETH_P_IPV6):
- ipv6_change_dsfield(ipv6_hdr(skb), p->mask[index],
- p->value[index]);
+ ipv6_change_dsfield(ipv6_hdr(skb), p->mv[index].mask,
+ p->mv[index].value);
break;
default:
/*
@@ -304,7 +306,7 @@ static struct sk_buff *dsmark_dequeue(struct Qdisc *sch)
* This way, we can send non-IP traffic through dsmark
* and don't need yet another qdisc as a bypass.
*/
- if (p->mask[index] != 0xff || p->value[index])
+ if (p->mv[index].mask != 0xff || p->mv[index].value)
pr_warn("%s: unsupported protocol %d\n",
__func__, ntohs(tc_skb_protocol(skb)));
break;
@@ -346,7 +348,7 @@ static int dsmark_init(struct Qdisc *sch, struct nlattr *opt)
int err = -EINVAL;
u32 default_index = NO_DEFAULT_INDEX;
u16 indices;
- u8 *mask;
+ int i;
pr_debug("%s(sch %p,[qdisc %p],opt %p)\n", __func__, sch, p, opt);
@@ -366,18 +368,18 @@ static int dsmark_init(struct Qdisc *sch, struct nlattr *opt)
if (tb[TCA_DSMARK_DEFAULT_INDEX])
default_index = nla_get_u16(tb[TCA_DSMARK_DEFAULT_INDEX]);
- mask = kmalloc(indices * 2, GFP_KERNEL);
- if (mask == NULL) {
+ if (indices <= DSMARK_EMBEDDED_SZ)
+ p->mv = p->embedded;
+ else
+ p->mv = kmalloc_array(indices, sizeof(*p->mv), GFP_KERNEL);
+ if (!p->mv) {
err = -ENOMEM;
goto errout;
}
-
- p->mask = mask;
- memset(p->mask, 0xff, indices);
-
- p->value = p->mask + indices;
- memset(p->value, 0, indices);
-
+ for (i = 0; i < indices; i++) {
+ p->mv[i].mask = 0xff;
+ p->mv[i].value = 0;
+ }
p->indices = indices;
p->default_index = default_index;
p->set_tc_index = nla_get_flag(tb[TCA_DSMARK_SET_TC_INDEX]);
@@ -410,7 +412,8 @@ static void dsmark_destroy(struct Qdisc *sch)
tcf_destroy_chain(&p->filter_list);
qdisc_destroy(p->q);
- kfree(p->mask);
+ if (p->mv != p->embedded)
+ kfree(p->mv);
}
static int dsmark_dump_class(struct Qdisc *sch, unsigned long cl,
@@ -430,8 +433,8 @@ static int dsmark_dump_class(struct Qdisc *sch, unsigned long cl,
opts = nla_nest_start(skb, TCA_OPTIONS);
if (opts == NULL)
goto nla_put_failure;
- if (nla_put_u8(skb, TCA_DSMARK_MASK, p->mask[cl - 1]) ||
- nla_put_u8(skb, TCA_DSMARK_VALUE, p->value[cl - 1]))
+ if (nla_put_u8(skb, TCA_DSMARK_MASK, p->mv[cl - 1].mask) ||
+ nla_put_u8(skb, TCA_DSMARK_VALUE, p->mv[cl - 1].value))
goto nla_put_failure;
return nla_nest_end(skb, opts);
diff --git a/net/sched/sch_fq.c b/net/sched/sch_fq.c
index f377702d4..109b23227 100644
--- a/net/sched/sch_fq.c
+++ b/net/sched/sch_fq.c
@@ -224,13 +224,16 @@ static struct fq_flow *fq_classify(struct sk_buff *skb, struct fq_sched_data *q)
if (unlikely((skb->priority & TC_PRIO_MAX) == TC_PRIO_CONTROL))
return &q->internal;
- /* SYNACK messages are attached to a listener socket.
- * 1) They are not part of a 'flow' yet
- * 2) We do not want to rate limit them (eg SYNFLOOD attack),
+ /* SYNACK messages are attached to a TCP_NEW_SYN_RECV request socket
+ * or a listener (SYNCOOKIE mode)
+ * 1) request sockets are not full blown,
+ * they do not contain sk_pacing_rate
+ * 2) They are not part of a 'flow' yet
+ * 3) We do not want to rate limit them (eg SYNFLOOD attack),
* especially if the listener set SO_MAX_PACING_RATE
- * 3) We pretend they are orphaned
+ * 4) We pretend they are orphaned
*/
- if (!sk || sk->sk_state == TCP_LISTEN) {
+ if (!sk || sk_listener(sk)) {
unsigned long hash = skb_get_hash(skb) & q->orphan_mask;
/* By forcing low order bit to 1, we make sure to not
diff --git a/net/sched/sch_generic.c b/net/sched/sch_generic.c
index e82a1ad80..16bc83b28 100644
--- a/net/sched/sch_generic.c
+++ b/net/sched/sch_generic.c
@@ -658,8 +658,10 @@ static void qdisc_rcu_free(struct rcu_head *head)
{
struct Qdisc *qdisc = container_of(head, struct Qdisc, rcu_head);
- if (qdisc_is_percpu_stats(qdisc))
+ if (qdisc_is_percpu_stats(qdisc)) {
free_percpu(qdisc->cpu_bstats);
+ free_percpu(qdisc->cpu_qstats);
+ }
kfree((char *) qdisc - qdisc->padded);
}