bpf: add skb->queue_mapping write access from tc clsact

netoptimizer · borkmann · commit 74e31ca850c1 · 2019-02-19T21:56:05.000+01:00
The skb-&gt;queue_mapping already have read access, via __sk_buff-&gt;queue_mapping.

This patch allow BPF tc qdisc clsact write access to the queue_mapping via
tc_cls_act_is_valid_access.  Also handle that the value NO_QUEUE_MAPPING
is not allowed.

It is already possible to change this via TC filter action skbedit
tc-skbedit(8).  Due to the lack of TC examples, lets show one:

  # tc qdisc  add  dev ixgbe1 clsact
  # tc filter add  dev ixgbe1 ingress matchall action skbedit queue_mapping 5
  # tc filter list dev ixgbe1 ingress

The most common mistake is that XPS (Transmit Packet Steering) takes
precedence over setting skb-&gt;queue_mapping. XPS is configured per DEVICE
via /sys/class/net/DEVICE/queues/tx-*/xps_cpus via a CPU hex mask. To
disable set mask=00.

The purpose of changing skb-&gt;queue_mapping is to influence the selection of
the net_device "txq" (struct netdev_queue), which influence selection of
the qdisc "root_lock" (via txq-&gt;qdisc-&gt;q.lock) and txq-&gt;_xmit_lock. When
using the MQ qdisc the txq-&gt;qdisc points to different qdiscs and associated
locks, and HARD_TX_LOCK (txq-&gt;_xmit_lock), allowing for CPU scalability.

Due to lack of TC examples, lets show howto attach clsact BPF programs:

 # tc qdisc  add  dev ixgbe2 clsact
 # tc filter add  dev ixgbe2 egress bpf da obj XXX_kern.o sec tc_qmap2cpu
 # tc filter list dev ixgbe2 egress

Signed-off-by: Jesper Dangaard Brouer &lt;brouer@redhat.com&gt;
Signed-off-by: Daniel Borkmann &lt;daniel@iogearbox.net&gt;
diff --git a/net/core/filter.c b/net/core/filter.c
@@ -6279,6 +6279,7 @@ static bool tc_cls_act_is_valid_access(int off, int size,
 		case bpf_ctx_range(struct __sk_buff, tc_classid):
 		case bpf_ctx_range_till(struct __sk_buff, cb[0], cb[4]):
 		case bpf_ctx_range(struct __sk_buff, tstamp):
+		case bpf_ctx_range(struct __sk_buff, queue_mapping):
 			break;
 		default:
 			return false;
@@ -6683,9 +6684,18 @@ static u32 bpf_convert_ctx_access(enum bpf_access_type type,
 		break;
 
 	case offsetof(struct __sk_buff, queue_mapping):
-		*insn++ = BPF_LDX_MEM(BPF_H, si->dst_reg, si->src_reg,
-				      bpf_target_off(struct sk_buff, queue_mapping, 2,
-						     target_size));
+		if (type == BPF_WRITE) {
+			*insn++ = BPF_JMP_IMM(BPF_JGE, si->src_reg, NO_QUEUE_MAPPING, 1);
+			*insn++ = BPF_STX_MEM(BPF_H, si->dst_reg, si->src_reg,
+					      bpf_target_off(struct sk_buff,
+							     queue_mapping,
+							     2, target_size));
+		} else {
+			*insn++ = BPF_LDX_MEM(BPF_H, si->dst_reg, si->src_reg,
+					      bpf_target_off(struct sk_buff,
+							     queue_mapping,
+							     2, target_size));
+		}
 		break;
 
 	case offsetof(struct __sk_buff, vlan_present):
