bpf: Add fd-based tcx multi-prog infra with link support

This work refactors and adds a lightweight extension ("tcx") to the tc BPF
ingress and egress data path side for allowing BPF program management based
on fds via bpf() syscall through the newly added generic multi-prog API.
The main goal behind this work which we also presented at LPC [0] last year
and a recent update at LSF/MM/BPF this year [3] is to support long-awaited
BPF link functionality for tc BPF programs, which allows for a model of safe
ownership and program detachment.

Given the rise in tc BPF users in cloud native environments, this becomes
necessary to avoid hard to debug incidents either through stale leftover
programs or 3rd party applications accidentally stepping on each others toes.
As a recap, a BPF link represents the attachment of a BPF program to a BPF
hook point. The BPF link holds a single reference to keep BPF program alive.
Moreover, hook points do not reference a BPF link, only the application's
fd or pinning does. A BPF link holds meta-data specific to attachment and
implements operations for link creation, (atomic) BPF program update,
detachment and introspection. The motivation for BPF links for tc BPF programs
is multi-fold, for example:

  - From Meta: "It's especially important for applications that are deployed
    fleet-wide and that don't "control" hosts they are deployed to. If such
    application crashes and no one notices and does anything about that, BPF
    program will keep running draining resources or even just, say, dropping
    packets. We at FB had outages due to such permanent BPF attachment
    semantics. With fd-based BPF link we are getting a framework, which allows
    safe, auto-detachable behavior by default, unless application explicitly
    opts in by pinning the BPF link." [1]

  - From Cilium-side the tc BPF programs we attach to host-facing veth devices
    and phys devices build the core datapath for Kubernetes Pods, and they
    implement forwarding, load-balancing, policy, EDT-management, etc, within
    BPF. Currently there is no concept of 'safe' ownership, e.g. we've recently
    experienced hard-to-debug issues in a user's staging environment where
    another Kubernetes application using tc BPF attached to the same prio/handle
    of cls_bpf, accidentally wiping all Cilium-based BPF programs from underneath
    it. The goal is to establish a clear/safe ownership model via links which
    cannot accidentally be overridden. [0,2]

BPF links for tc can co-exist with non-link attachments, and the semantics are
in line also with XDP links: BPF links cannot replace other BPF links, BPF
links cannot replace non-BPF links, non-BPF links cannot replace BPF links and
lastly only non-BPF links can replace non-BPF links. In case of Cilium, this
would solve mentioned issue of safe ownership model as 3rd party applications
would not be able to accidentally wipe Cilium programs, even if they are not
BPF link aware.

Earlier attempts [4] have tried to integrate BPF links into core tc machinery
to solve cls_bpf, which has been intrusive to the generic tc kernel API with
extensions only specific to cls_bpf and suboptimal/complex since cls_bpf could
be wiped from the qdisc also. Locking a tc BPF program in place this way, is
getting into layering hacks given the two object models are vastly different.

We instead implemented the tcx (tc 'express') layer which is an fd-based tc BPF
attach API, so that the BPF link implementation blends in naturally similar to
other link types which are fd-based and without the need for changing core tc
internal APIs. BPF programs for tc can then be successively migrated from classic
cls_bpf to the new tc BPF link without needing to change the program's source
code, just the BPF loader mechanics for attaching is sufficient.

For the current tc framework, there is no change in behavior with this change
and neither does this change touch on tc core kernel APIs. The gist of this
patch is that the ingress and egress hook have a lightweight, qdisc-less
extension for BPF to attach its tc BPF programs, in other words, a minimal
entry point for tc BPF. The name tcx has been suggested from discussion of
earlier revisions of this work as a good fit, and to more easily differ between
the classic cls_bpf attachment and the fd-based one.

For the ingress and egress tcx points, the device holds a cache-friendly array
with program pointers which is separated from control plane (slow-path) data.
Earlier versions of this work used priority to determine ordering and expression
of dependencies similar as with classic tc, but it was challenged that for
something more future-proof a better user experience is required. Hence this
resulted in the design and development of the generic attach/detach/query API
for multi-progs. See prior patch with its discussion on the API design. tcx is
the first user and later we plan to integrate also others, for example, one
candidate is multi-prog support for XDP which would benefit and have the same
'look and feel' from API perspective.

The goal with tcx is to have maximum compatibility to existing tc BPF programs,
so they don't need to be rewritten specifically. Compatibility to call into
classic tcf_classify() is also provided in order to allow successive migration
or both to cleanly co-exist where needed given its all one logical tc layer and
the tcx plus classic tc cls/act build one logical overall processing pipeline.

tcx supports the simplified return codes TCX_NEXT which is non-terminating (go
to next program) and terminating ones with TCX_PASS, TCX_DROP, TCX_REDIRECT.
The fd-based API is behind a static key, so that when unused the code is also
not entered. The struct tcx_entry's program array is currently static, but
could be made dynamic if necessary at a point in future. The a/b pair swap
design has been chosen so that for detachment there are no allocations which
otherwise could fail.

The work has been tested with tc-testing selftest suite which all passes, as
well as the tc BPF tests from the BPF CI, and also with Cilium's L4LB.

Thanks also to Nikolay Aleksandrov and Martin Lau for in-depth early reviews
of this work.

  [0] https://lpc.events/event/16/contributions/1353/
  [1] https://lore.kernel.org/bpf/CAEf4BzbokCJN33Nw_kg82sO=xppXnKWEncGTWCTB9vGCmLB6pw@mail.gmail.com
  [2] https://colocatedeventseu2023.sched.com/event/1Jo6O/tales-from-an-ebpf-programs-murder-mystery-hemanth-malla-guillaume-fournier-datadog
  [3] http://vger.kernel.org/bpfconf2023_material/tcx_meta_netdev_borkmann.pdf
  [4] https://lore.kernel.org/bpf/20210604063116.234316-1-memxor@gmail.com

Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Acked-by: Jakub Kicinski <kuba@kernel.org>
Link: https://lore.kernel.org/r/20230719140858.13224-3-daniel@iogearbox.net
Signed-off-by: Alexei Starovoitov <ast@kernel.org>

1 files changed, 163 insertions, 102 deletions

diff --git a/net/core/dev.c b/net/core/dev.c
index dd4f114a7cbf..8e7d0cb540cd 100644
--- a/net/core/dev.c
+++ b/net/core/dev.c
@@ -107,6 +107,7 @@
 #include <net/pkt_cls.h>
 #include <net/checksum.h>
 #include <net/xfrm.h>
+#include <net/tcx.h>
 #include <linux/highmem.h>
 #include <linux/init.h>
 #include <linux/module.h>
@@ -154,7 +155,6 @@
 #include "dev.h"
 #include "net-sysfs.h"
 
-
 static DEFINE_SPINLOCK(ptype_lock);
 struct list_head ptype_base[PTYPE_HASH_SIZE] __read_mostly;
 struct list_head ptype_all __read_mostly;	/* Taps */
@@ -3882,69 +3882,198 @@ int dev_loopback_xmit(struct net *net, struct sock *sk, struct sk_buff *skb)
 EXPORT_SYMBOL(dev_loopback_xmit);
 
 #ifdef CONFIG_NET_EGRESS
-static struct sk_buff *
-sch_handle_egress(struct sk_buff *skb, int *ret, struct net_device *dev)
+static struct netdev_queue *
+netdev_tx_queue_mapping(struct net_device *dev, struct sk_buff *skb)
+{
+	int qm = skb_get_queue_mapping(skb);
+
+	return netdev_get_tx_queue(dev, netdev_cap_txqueue(dev, qm));
+}
+
+static bool netdev_xmit_txqueue_skipped(void)
 {
+	return __this_cpu_read(softnet_data.xmit.skip_txqueue);
+}
+
+void netdev_xmit_skip_txqueue(bool skip)
+{
+	__this_cpu_write(softnet_data.xmit.skip_txqueue, skip);
+}
+EXPORT_SYMBOL_GPL(netdev_xmit_skip_txqueue);
+#endif /* CONFIG_NET_EGRESS */
+
+#ifdef CONFIG_NET_XGRESS
+static int tc_run(struct tcx_entry *entry, struct sk_buff *skb)
+{
+	int ret = TC_ACT_UNSPEC;
 #ifdef CONFIG_NET_CLS_ACT
-	struct mini_Qdisc *miniq = rcu_dereference_bh(dev->miniq_egress);
-	struct tcf_result cl_res;
+	struct mini_Qdisc *miniq = rcu_dereference_bh(entry->miniq);
+	struct tcf_result res;
 
 	if (!miniq)
-		return skb;
+		return ret;
 
-	/* qdisc_skb_cb(skb)->pkt_len was already set by the caller. */
 	tc_skb_cb(skb)->mru = 0;
 	tc_skb_cb(skb)->post_ct = false;
-	mini_qdisc_bstats_cpu_update(miniq, skb);
 
-	switch (tcf_classify(skb, miniq->block, miniq->filter_list, &cl_res, false)) {
+	mini_qdisc_bstats_cpu_update(miniq, skb);
+	ret = tcf_classify(skb, miniq->block, miniq->filter_list, &res, false);
+	/* Only tcf related quirks below. */
+	switch (ret) {
+	case TC_ACT_SHOT:
+		mini_qdisc_qstats_cpu_drop(miniq);
+		break;
 	case TC_ACT_OK:
 	case TC_ACT_RECLASSIFY:
-		skb->tc_index = TC_H_MIN(cl_res.classid);
+		skb->tc_index = TC_H_MIN(res.classid);
 		break;
+	}
+#endif /* CONFIG_NET_CLS_ACT */
+	return ret;
+}
+
+static DEFINE_STATIC_KEY_FALSE(tcx_needed_key);
+
+void tcx_inc(void)
+{
+	static_branch_inc(&tcx_needed_key);
+}
+
+void tcx_dec(void)
+{
+	static_branch_dec(&tcx_needed_key);
+}
+
+static __always_inline enum tcx_action_base
+tcx_run(const struct bpf_mprog_entry *entry, struct sk_buff *skb,
+	const bool needs_mac)
+{
+	const struct bpf_mprog_fp *fp;
+	const struct bpf_prog *prog;
+	int ret = TCX_NEXT;
+
+	if (needs_mac)
+		__skb_push(skb, skb->mac_len);
+	bpf_mprog_foreach_prog(entry, fp, prog) {
+		bpf_compute_data_pointers(skb);
+		ret = bpf_prog_run(prog, skb);
+		if (ret != TCX_NEXT)
+			break;
+	}
+	if (needs_mac)
+		__skb_pull(skb, skb->mac_len);
+	return tcx_action_code(skb, ret);
+}
+
+static __always_inline struct sk_buff *
+sch_handle_ingress(struct sk_buff *skb, struct packet_type **pt_prev, int *ret,
+		   struct net_device *orig_dev, bool *another)
+{
+	struct bpf_mprog_entry *entry = rcu_dereference_bh(skb->dev->tcx_ingress);
+	int sch_ret;
+
+	if (!entry)
+		return skb;
+	if (*pt_prev) {
+		*ret = deliver_skb(skb, *pt_prev, orig_dev);
+		*pt_prev = NULL;
+	}
+
+	qdisc_skb_cb(skb)->pkt_len = skb->len;
+	tcx_set_ingress(skb, true);
+
+	if (static_branch_unlikely(&tcx_needed_key)) {
+		sch_ret = tcx_run(entry, skb, true);
+		if (sch_ret != TC_ACT_UNSPEC)
+			goto ingress_verdict;
+	}
+	sch_ret = tc_run(tcx_entry(entry), skb);
+ingress_verdict:
+	switch (sch_ret) {
+	case TC_ACT_REDIRECT:
+		/* skb_mac_header check was done by BPF, so we can safely
+		 * push the L2 header back before redirecting to another
+		 * netdev.
+		 */
+		__skb_push(skb, skb->mac_len);
+		if (skb_do_redirect(skb) == -EAGAIN) {
+			__skb_pull(skb, skb->mac_len);
+			*another = true;
+			break;
+		}
+		*ret = NET_RX_SUCCESS;
+		return NULL;
 	case TC_ACT_SHOT:
-		mini_qdisc_qstats_cpu_drop(miniq);
-		*ret = NET_XMIT_DROP;
-		kfree_skb_reason(skb, SKB_DROP_REASON_TC_EGRESS);
+		kfree_skb_reason(skb, SKB_DROP_REASON_TC_INGRESS);
+		*ret = NET_RX_DROP;
 		return NULL;
+	/* used by tc_run */
 	case TC_ACT_STOLEN:
 	case TC_ACT_QUEUED:
 	case TC_ACT_TRAP:
-		*ret = NET_XMIT_SUCCESS;
 		consume_skb(skb);
+		fallthrough;
+	case TC_ACT_CONSUMED:
+		*ret = NET_RX_SUCCESS;
 		return NULL;
+	}
+
+	return skb;
+}
+
+static __always_inline struct sk_buff *
+sch_handle_egress(struct sk_buff *skb, int *ret, struct net_device *dev)
+{
+	struct bpf_mprog_entry *entry = rcu_dereference_bh(dev->tcx_egress);
+	int sch_ret;
+
+	if (!entry)
+		return skb;
+
+	/* qdisc_skb_cb(skb)->pkt_len & tcx_set_ingress() was
+	 * already set by the caller.
+	 */
+	if (static_branch_unlikely(&tcx_needed_key)) {
+		sch_ret = tcx_run(entry, skb, false);
+		if (sch_ret != TC_ACT_UNSPEC)
+			goto egress_verdict;
+	}
+	sch_ret = tc_run(tcx_entry(entry), skb);
+egress_verdict:
+	switch (sch_ret) {
 	case TC_ACT_REDIRECT:
 		/* No need to push/pop skb's mac_header here on egress! */
 		skb_do_redirect(skb);
 		*ret = NET_XMIT_SUCCESS;
 		return NULL;
-	default:
-		break;
+	case TC_ACT_SHOT:
+		kfree_skb_reason(skb, SKB_DROP_REASON_TC_EGRESS);
+		*ret = NET_XMIT_DROP;
+		return NULL;
+	/* used by tc_run */
+	case TC_ACT_STOLEN:
+	case TC_ACT_QUEUED:
+	case TC_ACT_TRAP:
+		*ret = NET_XMIT_SUCCESS;
+		return NULL;
 	}
-#endif /* CONFIG_NET_CLS_ACT */
 
 	return skb;
 }
-
-static struct netdev_queue *
-netdev_tx_queue_mapping(struct net_device *dev, struct sk_buff *skb)
-{
-	int qm = skb_get_queue_mapping(skb);
-
-	return netdev_get_tx_queue(dev, netdev_cap_txqueue(dev, qm));
-}
-
-static bool netdev_xmit_txqueue_skipped(void)
+#else
+static __always_inline struct sk_buff *
+sch_handle_ingress(struct sk_buff *skb, struct packet_type **pt_prev, int *ret,
+		   struct net_device *orig_dev, bool *another)
 {
-	return __this_cpu_read(softnet_data.xmit.skip_txqueue);
+	return skb;
 }
 
-void netdev_xmit_skip_txqueue(bool skip)
+static __always_inline struct sk_buff *
+sch_handle_egress(struct sk_buff *skb, int *ret, struct net_device *dev)
 {
-	__this_cpu_write(softnet_data.xmit.skip_txqueue, skip);
+	return skb;
 }
-EXPORT_SYMBOL_GPL(netdev_xmit_skip_txqueue);
-#endif /* CONFIG_NET_EGRESS */
+#endif /* CONFIG_NET_XGRESS */
 
 #ifdef CONFIG_XPS
 static int __get_xps_queue_idx(struct net_device *dev, struct sk_buff *skb,
@@ -4128,9 +4257,7 @@ int __dev_queue_xmit(struct sk_buff *skb, struct net_device *sb_dev)
 	skb_update_prio(skb);
 
 	qdisc_pkt_len_init(skb);
-#ifdef CONFIG_NET_CLS_ACT
-	skb->tc_at_ingress = 0;
-#endif
+	tcx_set_ingress(skb, false);
 #ifdef CONFIG_NET_EGRESS
 	if (static_branch_unlikely(&egress_needed_key)) {
 		if (nf_hook_egress_active()) {
@@ -5064,72 +5191,6 @@ int (*br_fdb_test_addr_hook)(struct net_device *dev,
 EXPORT_SYMBOL_GPL(br_fdb_test_addr_hook);
 #endif
 
-static inline struct sk_buff *
-sch_handle_ingress(struct sk_buff *skb, struct packet_type **pt_prev, int *ret,
-		   struct net_device *orig_dev, bool *another)
-{
-#ifdef CONFIG_NET_CLS_ACT
-	struct mini_Qdisc *miniq = rcu_dereference_bh(skb->dev->miniq_ingress);
-	struct tcf_result cl_res;
-
-	/* If there's at least one ingress present somewhere (so
-	 * we get here via enabled static key), remaining devices
-	 * that are not configured with an ingress qdisc will bail
-	 * out here.
-	 */
-	if (!miniq)
-		return skb;
-
-	if (*pt_prev) {
-		*ret = deliver_skb(skb, *pt_prev, orig_dev);
-		*pt_prev = NULL;
-	}
-
-	qdisc_skb_cb(skb)->pkt_len = skb->len;
-	tc_skb_cb(skb)->mru = 0;
-	tc_skb_cb(skb)->post_ct = false;
-	skb->tc_at_ingress = 1;
-	mini_qdisc_bstats_cpu_update(miniq, skb);
-
-	switch (tcf_classify(skb, miniq->block, miniq->filter_list, &cl_res, false)) {
-	case TC_ACT_OK:
-	case TC_ACT_RECLASSIFY:
-		skb->tc_index = TC_H_MIN(cl_res.classid);
-		break;
-	case TC_ACT_SHOT:
-		mini_qdisc_qstats_cpu_drop(miniq);
-		kfree_skb_reason(skb, SKB_DROP_REASON_TC_INGRESS);
-		*ret = NET_RX_DROP;
-		return NULL;
-	case TC_ACT_STOLEN:
-	case TC_ACT_QUEUED:
-	case TC_ACT_TRAP:
-		consume_skb(skb);
-		*ret = NET_RX_SUCCESS;
-		return NULL;
-	case TC_ACT_REDIRECT:
-		/* skb_mac_header check was done by cls/act_bpf, so
-		 * we can safely push the L2 header back before
-		 * redirecting to another netdev
-		 */
-		__skb_push(skb, skb->mac_len);
-		if (skb_do_redirect(skb) == -EAGAIN) {
-			__skb_pull(skb, skb->mac_len);
-			*another = true;
-			break;
-		}
-		*ret = NET_RX_SUCCESS;
-		return NULL;
-	case TC_ACT_CONSUMED:
-		*ret = NET_RX_SUCCESS;
-		return NULL;
-	default:
-		break;
-	}
-#endif /* CONFIG_NET_CLS_ACT */
-	return skb;
-}
-
 /**
  *	netdev_is_rx_handler_busy - check if receive handler is registered
  *	@dev: device to check
@@ -10835,7 +10896,7 @@ void unregister_netdevice_many_notify(struct list_head *head,
 
 		/* Shutdown queueing discipline. */
 		dev_shutdown(dev);
-
+		dev_tcx_uninstall(dev);
 		dev_xdp_uninstall(dev);
 		bpf_dev_bound_netdev_unregister(dev);