Skip to content

Commit

Permalink
sched, bpf: add helper for retrieving routing realms
Browse files Browse the repository at this point in the history
Using routing realms as part of the classifier is quite useful, it
can be viewed as a tag for one or multiple routing entries (think of
an analogy to net_cls cgroup for processes), set by user space routing
daemons or via iproute2 as an indicator for traffic classifiers and
later on processed in the eBPF program.

Unlike actions, the classifier can inspect device flags and enable
netif_keep_dst() if necessary. tc actions don't have that possibility,
but in case people know what they are doing, it can be used from there
as well (e.g. via devs that must keep dsts by design anyway).

If a realm is set, the handler returns the non-zero realm. User space
can set the full 32bit realm for the dst.

Signed-off-by: Daniel Borkmann <[email protected]>
Acked-by: Alexei Starovoitov <[email protected]>
Signed-off-by: David S. Miller <[email protected]>
  • Loading branch information
borkmann authored and davem330 committed Oct 3, 2015
1 parent a91263d commit c46646d
Show file tree
Hide file tree
Showing 5 changed files with 39 additions and 3 deletions.
3 changes: 2 additions & 1 deletion include/linux/filter.h
Original file line number Diff line number Diff line change
Expand Up @@ -328,7 +328,8 @@ struct bpf_prog {
u16 pages; /* Number of allocated pages */
kmemcheck_bitfield_begin(meta);
u16 jited:1, /* Is our filter JIT'ed? */
gpl_compatible:1; /* Is filter GPL compatible? */
gpl_compatible:1, /* Is filter GPL compatible? */
dst_needed:1; /* Do we need dst entry? */
kmemcheck_bitfield_end(meta);
u32 len; /* Number of filter blocks */
enum bpf_prog_type type; /* Type of BPF program */
Expand Down
7 changes: 7 additions & 0 deletions include/uapi/linux/bpf.h
Original file line number Diff line number Diff line change
Expand Up @@ -280,6 +280,13 @@ enum bpf_func_id {
* Return: TC_ACT_REDIRECT
*/
BPF_FUNC_redirect,

/**
* bpf_get_route_realm(skb) - retrieve a dst's tclassid
* @skb: pointer to skb
* Return: realm if != 0
*/
BPF_FUNC_get_route_realm,
__BPF_FUNC_MAX_ID,
};

Expand Down
2 changes: 2 additions & 0 deletions kernel/bpf/syscall.c
Original file line number Diff line number Diff line change
Expand Up @@ -402,6 +402,8 @@ static void fixup_bpf_calls(struct bpf_prog *prog)
*/
BUG_ON(!prog->aux->ops->get_func_proto);

if (insn->imm == BPF_FUNC_get_route_realm)
prog->dst_needed = 1;
if (insn->imm == BPF_FUNC_tail_call) {
/* mark bpf_tail_call as different opcode
* to avoid conditional branch in
Expand Down
22 changes: 22 additions & 0 deletions net/core/filter.c
Original file line number Diff line number Diff line change
Expand Up @@ -49,6 +49,7 @@
#include <net/sch_generic.h>
#include <net/cls_cgroup.h>
#include <net/dst_metadata.h>
#include <net/dst.h>

/**
* sk_filter - run a packet through a socket filter
Expand Down Expand Up @@ -1478,6 +1479,25 @@ static const struct bpf_func_proto bpf_get_cgroup_classid_proto = {
.arg1_type = ARG_PTR_TO_CTX,
};

static u64 bpf_get_route_realm(u64 r1, u64 r2, u64 r3, u64 r4, u64 r5)
{
#ifdef CONFIG_IP_ROUTE_CLASSID
const struct dst_entry *dst;

dst = skb_dst((struct sk_buff *) (unsigned long) r1);
if (dst)
return dst->tclassid;
#endif
return 0;
}

static const struct bpf_func_proto bpf_get_route_realm_proto = {
.func = bpf_get_route_realm,
.gpl_only = false,
.ret_type = RET_INTEGER,
.arg1_type = ARG_PTR_TO_CTX,
};

static u64 bpf_skb_vlan_push(u64 r1, u64 r2, u64 vlan_tci, u64 r4, u64 r5)
{
struct sk_buff *skb = (struct sk_buff *) (long) r1;
Expand Down Expand Up @@ -1648,6 +1668,8 @@ tc_cls_act_func_proto(enum bpf_func_id func_id)
return bpf_get_skb_set_tunnel_key_proto();
case BPF_FUNC_redirect:
return &bpf_redirect_proto;
case BPF_FUNC_get_route_realm:
return &bpf_get_route_realm_proto;
default:
return sk_filter_func_proto(func_id);
}
Expand Down
8 changes: 6 additions & 2 deletions net/sched/cls_bpf.c
Original file line number Diff line number Diff line change
Expand Up @@ -262,7 +262,8 @@ static int cls_bpf_prog_from_ops(struct nlattr **tb, struct cls_bpf_prog *prog)
return 0;
}

static int cls_bpf_prog_from_efd(struct nlattr **tb, struct cls_bpf_prog *prog)
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;
Expand Down Expand Up @@ -294,6 +295,9 @@ static int cls_bpf_prog_from_efd(struct nlattr **tb, struct cls_bpf_prog *prog)
prog->bpf_name = name;
prog->filter = fp;

if (fp->dst_needed)
netif_keep_dst(qdisc_dev(tp->q));

return 0;
}

Expand Down Expand Up @@ -330,7 +334,7 @@ static int cls_bpf_modify_existing(struct net *net, struct tcf_proto *tp,
prog->exts_integrated = have_exts;

ret = is_bpf ? cls_bpf_prog_from_ops(tb, prog) :
cls_bpf_prog_from_efd(tb, prog);
cls_bpf_prog_from_efd(tb, prog, tp);
if (ret < 0) {
tcf_exts_destroy(&exts);
return ret;
Expand Down

0 comments on commit c46646d

Please sign in to comment.