Merge tag 'net-5.18-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/netdev/net

Pull more networking updates from Jakub Kicinski:
 "Networking fixes and rethook patches.

  Features:

   - kprobes: rethook: x86: replace kretprobe trampoline with rethook

  Current release - regressions:

   - sfc: avoid null-deref on systems without NUMA awareness in the new
     queue sizing code

  Current release - new code bugs:

   - vxlan: do not feed vxlan_vnifilter_dump_dev with non-vxlan devices

   - eth: lan966x: fix null-deref on PHY pointer in timestamp ioctl when
     interface is down

  Previous releases - always broken:

   - openvswitch: correct neighbor discovery target mask field in the
     flow dump

   - wireguard: ignore v6 endpoints when ipv6 is disabled and fix a leak

   - rxrpc: fix call timer start racing with call destruction

   - rxrpc: fix null-deref when security type is rxrpc_no_security

   - can: fix UAF bugs around echo skbs in multiple drivers

  Misc:

   - docs: move netdev-FAQ to the 'process' section of the
     documentation"

* tag 'net-5.18-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/netdev/net: (57 commits)
  vxlan: do not feed vxlan_vnifilter_dump_dev with non vxlan devices
  openvswitch: Add recirc_id to recirc warning
  rxrpc: fix some null-ptr-deref bugs in server_key.c
  rxrpc: Fix call timer start racing with call destruction
  net: hns3: fix software vlan talbe of vlan 0 inconsistent with hardware
  net: hns3: fix the concurrency between functions reading debugfs
  docs: netdev: move the netdev-FAQ to the process pages
  docs: netdev: broaden the new vs old code formatting guidelines
  docs: netdev: call out the merge window in tag checking
  docs: netdev: add missing back ticks
  docs: netdev: make the testing requirement more stringent
  docs: netdev: add a question about re-posting frequency
  docs: netdev: rephrase the 'should I update patchwork' question
  docs: netdev: rephrase the 'Under review' question
  docs: netdev: shorten the name and mention msgid for patch status
  docs: netdev: note that RFC postings are allowed any time
  docs: netdev: turn the net-next closed into a Warning
  docs: netdev: move the patch marking section up
  docs: netdev: minor reword
  docs: netdev: replace references to old archives
  ...

5 files changed, 112 insertions, 27 deletions

diff --git a/kernel/Makefile b/kernel/Makefile
index 56f4ee97f328..471d71935e90 100644
--- a/kernel/Makefile
+++ b/kernel/Makefile
@@ -108,6 +108,7 @@ obj-$(CONFIG_TRACING) += trace/
 obj-$(CONFIG_TRACE_CLOCK) += trace/
 obj-$(CONFIG_RING_BUFFER) += trace/
 obj-$(CONFIG_TRACEPOINTS) += trace/
+obj-$(CONFIG_RETHOOK) += trace/
 obj-$(CONFIG_IRQ_WORK) += irq_work.o
 obj-$(CONFIG_CPU_PM) += cpu_pm.o
 obj-$(CONFIG_BPF) += bpf/
diff --git a/kernel/bpf/btf.c b/kernel/bpf/btf.c
index 24788ce564a0..0918a39279f6 100644
--- a/kernel/bpf/btf.c
+++ b/kernel/bpf/btf.c
@@ -5507,7 +5507,7 @@ int btf_distill_func_proto(struct bpf_verifier_log *log,
 	}
 	args = (const struct btf_param *)(func + 1);
 	nargs = btf_type_vlen(func);
-	if (nargs >= MAX_BPF_FUNC_ARGS) {
+	if (nargs > MAX_BPF_FUNC_ARGS) {
 		bpf_log(log,
 			"The function %s has %d arguments. Too many.\n",
 			tname, nargs);
diff --git a/kernel/kprobes.c b/kernel/kprobes.c
index 185badc780b7..dbe57df2e199 100644
--- a/kernel/kprobes.c
+++ b/kernel/kprobes.c
@@ -1237,6 +1237,27 @@ void kprobes_inc_nmissed_count(struct kprobe *p)
 }
 NOKPROBE_SYMBOL(kprobes_inc_nmissed_count);
 
+static struct kprobe kprobe_busy = {
+	.addr = (void *) get_kprobe,
+};
+
+void kprobe_busy_begin(void)
+{
+	struct kprobe_ctlblk *kcb;
+
+	preempt_disable();
+	__this_cpu_write(current_kprobe, &kprobe_busy);
+	kcb = get_kprobe_ctlblk();
+	kcb->kprobe_status = KPROBE_HIT_ACTIVE;
+}
+
+void kprobe_busy_end(void)
+{
+	__this_cpu_write(current_kprobe, NULL);
+	preempt_enable();
+}
+
+#if !defined(CONFIG_KRETPROBE_ON_RETHOOK)
 static void free_rp_inst_rcu(struct rcu_head *head)
 {
 	struct kretprobe_instance *ri = container_of(head, struct kretprobe_instance, rcu);
@@ -1258,26 +1279,6 @@ static void recycle_rp_inst(struct kretprobe_instance *ri)
 }
 NOKPROBE_SYMBOL(recycle_rp_inst);
 
-static struct kprobe kprobe_busy = {
-	.addr = (void *) get_kprobe,
-};
-
-void kprobe_busy_begin(void)
-{
-	struct kprobe_ctlblk *kcb;
-
-	preempt_disable();
-	__this_cpu_write(current_kprobe, &kprobe_busy);
-	kcb = get_kprobe_ctlblk();
-	kcb->kprobe_status = KPROBE_HIT_ACTIVE;
-}
-
-void kprobe_busy_end(void)
-{
-	__this_cpu_write(current_kprobe, NULL);
-	preempt_enable();
-}
-
 /*
  * This function is called from delayed_put_task_struct() when a task is
  * dead and cleaned up to recycle any kretprobe instances associated with
@@ -1327,6 +1328,7 @@ static inline void free_rp_inst(struct kretprobe *rp)
 		rp->rph = NULL;
 	}
 }
+#endif	/* !CONFIG_KRETPROBE_ON_RETHOOK */
 
 /* Add the new probe to 'ap->list'. */
 static int add_new_kprobe(struct kprobe *ap, struct kprobe *p)
@@ -1925,6 +1927,7 @@ static struct notifier_block kprobe_exceptions_nb = {
 
 #ifdef CONFIG_KRETPROBES
 
+#if !defined(CONFIG_KRETPROBE_ON_RETHOOK)
 /* This assumes the 'tsk' is the current task or the is not running. */
 static kprobe_opcode_t *__kretprobe_find_ret_addr(struct task_struct *tsk,
 						  struct llist_node **cur)
@@ -2087,6 +2090,57 @@ static int pre_handler_kretprobe(struct kprobe *p, struct pt_regs *regs)
 	return 0;
 }
 NOKPROBE_SYMBOL(pre_handler_kretprobe);
+#else /* CONFIG_KRETPROBE_ON_RETHOOK */
+/*
+ * This kprobe pre_handler is registered with every kretprobe. When probe
+ * hits it will set up the return probe.
+ */
+static int pre_handler_kretprobe(struct kprobe *p, struct pt_regs *regs)
+{
+	struct kretprobe *rp = container_of(p, struct kretprobe, kp);
+	struct kretprobe_instance *ri;
+	struct rethook_node *rhn;
+
+	rhn = rethook_try_get(rp->rh);
+	if (!rhn) {
+		rp->nmissed++;
+		return 0;
+	}
+
+	ri = container_of(rhn, struct kretprobe_instance, node);
+
+	if (rp->entry_handler && rp->entry_handler(ri, regs))
+		rethook_recycle(rhn);
+	else
+		rethook_hook(rhn, regs, kprobe_ftrace(p));
+
+	return 0;
+}
+NOKPROBE_SYMBOL(pre_handler_kretprobe);
+
+static void kretprobe_rethook_handler(struct rethook_node *rh, void *data,
+				      struct pt_regs *regs)
+{
+	struct kretprobe *rp = (struct kretprobe *)data;
+	struct kretprobe_instance *ri;
+	struct kprobe_ctlblk *kcb;
+
+	/* The data must NOT be null. This means rethook data structure is broken. */
+	if (WARN_ON_ONCE(!data))
+		return;
+
+	__this_cpu_write(current_kprobe, &rp->kp);
+	kcb = get_kprobe_ctlblk();
+	kcb->kprobe_status = KPROBE_HIT_ACTIVE;
+
+	ri = container_of(rh, struct kretprobe_instance, node);
+	rp->handler(ri, regs);
+
+	__this_cpu_write(current_kprobe, NULL);
+}
+NOKPROBE_SYMBOL(kretprobe_rethook_handler);
+
+#endif /* !CONFIG_KRETPROBE_ON_RETHOOK */
 
 /**
  * kprobe_on_func_entry() -- check whether given address is function entry
@@ -2155,6 +2209,29 @@ int register_kretprobe(struct kretprobe *rp)
 		rp->maxactive = num_possible_cpus();
 #endif
 	}
+#ifdef CONFIG_KRETPROBE_ON_RETHOOK
+	rp->rh = rethook_alloc((void *)rp, kretprobe_rethook_handler);
+	if (!rp->rh)
+		return -ENOMEM;
+
+	for (i = 0; i < rp->maxactive; i++) {
+		inst = kzalloc(sizeof(struct kretprobe_instance) +
+			       rp->data_size, GFP_KERNEL);
+		if (inst == NULL) {
+			rethook_free(rp->rh);
+			rp->rh = NULL;
+			return -ENOMEM;
+		}
+		rethook_add_node(rp->rh, &inst->node);
+	}
+	rp->nmissed = 0;
+	/* Establish function entry probe point */
+	ret = register_kprobe(&rp->kp);
+	if (ret != 0) {
+		rethook_free(rp->rh);
+		rp->rh = NULL;
+	}
+#else	/* !CONFIG_KRETPROBE_ON_RETHOOK */
 	rp->freelist.head = NULL;
 	rp->rph = kzalloc(sizeof(struct kretprobe_holder), GFP_KERNEL);
 	if (!rp->rph)
@@ -2179,6 +2256,7 @@ int register_kretprobe(struct kretprobe *rp)
 	ret = register_kprobe(&rp->kp);
 	if (ret != 0)
 		free_rp_inst(rp);
+#endif
 	return ret;
 }
 EXPORT_SYMBOL_GPL(register_kretprobe);
@@ -2217,7 +2295,11 @@ void unregister_kretprobes(struct kretprobe **rps, int num)
 	for (i = 0; i < num; i++) {
 		if (__unregister_kprobe_top(&rps[i]->kp) < 0)
 			rps[i]->kp.addr = NULL;
+#ifdef CONFIG_KRETPROBE_ON_RETHOOK
+		rethook_free(rps[i]->rh);
+#else
 		rps[i]->rph->rp = NULL;
+#endif
 	}
 	mutex_unlock(&kprobe_mutex);
 
@@ -2225,7 +2307,9 @@ void unregister_kretprobes(struct kretprobe **rps, int num)
 	for (i = 0; i < num; i++) {
 		if (rps[i]->kp.addr) {
 			__unregister_kprobe_bottom(&rps[i]->kp);
+#ifndef CONFIG_KRETPROBE_ON_RETHOOK
 			free_rp_inst(rps[i]);
+#endif
 		}
 	}
 }
diff --git a/kernel/trace/fprobe.c b/kernel/trace/fprobe.c
index 8b2dd5b9dcd1..89d9f994ebb0 100644
--- a/kernel/trace/fprobe.c
+++ b/kernel/trace/fprobe.c
@@ -150,15 +150,15 @@ static int fprobe_init_rethook(struct fprobe *fp, int num)
 
 	fp->rethook = rethook_alloc((void *)fp, fprobe_exit_handler);
 	for (i = 0; i < size; i++) {
-		struct rethook_node *node;
+		struct fprobe_rethook_node *node;
 
-		node = kzalloc(sizeof(struct fprobe_rethook_node), GFP_KERNEL);
+		node = kzalloc(sizeof(*node), GFP_KERNEL);
 		if (!node) {
 			rethook_free(fp->rethook);
 			fp->rethook = NULL;
 			return -ENOMEM;
 		}
-		rethook_add_node(fp->rethook, node);
+		rethook_add_node(fp->rethook, &node->node);
 	}
 	return 0;
 }
@@ -215,7 +215,7 @@ int register_fprobe(struct fprobe *fp, const char *filter, const char *notfilter
 	 * correctly calculate the total number of filtered symbols
 	 * from both filter and notfilter.
 	 */
-	hash = fp->ops.local_hash.filter_hash;
+	hash = rcu_access_pointer(fp->ops.local_hash.filter_hash);
 	if (WARN_ON_ONCE(!hash))
 		goto out;
 
diff --git a/kernel/trace/trace_kprobe.c b/kernel/trace/trace_kprobe.c
index b62fd785b599..47cebef78532 100644
--- a/kernel/trace/trace_kprobe.c
+++ b/kernel/trace/trace_kprobe.c
@@ -1433,7 +1433,7 @@ __kretprobe_trace_func(struct trace_kprobe *tk, struct kretprobe_instance *ri,
 	fbuffer.regs = regs;
 	entry = fbuffer.entry = ring_buffer_event_data(fbuffer.event);
 	entry->func = (unsigned long)tk->rp.kp.addr;
-	entry->ret_ip = (unsigned long)ri->ret_addr;
+	entry->ret_ip = get_kretprobe_retaddr(ri);
 	store_trace_args(&entry[1], &tk->tp, regs, sizeof(*entry), dsize);
 
 	trace_event_buffer_commit(&fbuffer);
@@ -1628,7 +1628,7 @@ kretprobe_perf_func(struct trace_kprobe *tk, struct kretprobe_instance *ri,
 		return;
 
 	entry->func = (unsigned long)tk->rp.kp.addr;
-	entry->ret_ip = (unsigned long)ri->ret_addr;
+	entry->ret_ip = get_kretprobe_retaddr(ri);
 	store_trace_args(&entry[1], &tk->tp, regs, sizeof(*entry), dsize);
 	perf_trace_buf_submit(entry, size, rctx, call->event.type, 1, regs,
 			      head, NULL);