x86/kprobes: Fix crash when probing CS CALL instructionsgrafted

When using eBPF to probe CS CALL instructions within a function,
a crash can be triggered.

The eBPF tool probes offset 257 of the __hrtimer_run_queues()
function:

<__hrtimer_run_queues+249>:  nopl   0x0(%rax,%rax,1)
<__hrtimer_run_queues+254>:  mov    %r14,%rdi
<__hrtimer_run_queues+257>:  cs call <__x86_indirect_thunk_r12>
<__hrtimer_run_queues+263>:  mov    %eax,%r12d
<__hrtimer_run_queues+266>:  xchg   %ax,%ax
<__hrtimer_run_queues+268>:  mov    %r13,%rdi

Which triggers this crash:

  BUG: unable to handle page fault for address: 00000000000f41c9
  #PF: supervisor write access in kernel mode
  #PF: error_code(0x0002) - not-present page
  PGD 0 P4D 0
  Oops: 0002 [#1] SMP NOPTI
  CPU: 1 PID: 0 Comm: swapper/1 Kdump: loaded Tainted: P
  RIP: 0010:__hrtimer_run_queues+0x106/0x230

Note that __hrtimer_run_queues+0x106 is __hrtimer_run_queues+262, which is
at the 6th byte of the above CS CALL instruction. Since the CS CALL
instruction occupies 6 bytes, the exception occurred in the middle of that
call instruction.

The root cause is that when using eBPF tools to probe in the middle of a
function, a kprobe with INT3 is used as the underlying implementation.

During single-step emulation of the original CALL instruction,
int3_emulate_call() assumes that the probed CALL instruction is 5 bytes
long. However, the actual CS-prefixed CALL instruction occupies 6 bytes,
so it constructs an incorrect exception return address. When the CPU
returns from the kprobe handler, the next instruction to be executed is at
the address of the last byte of that CS CALL instruction. Coincidentally,
starting from that address, the CPU fetches and decodes a completely
different instruction, which ultimately triggers a kernel crash.

Fix the issue by using the actual instruction length obtained from
the instruction decoder when constructing the exception return
address, rather than relying on the hardcoded CALL_INSN_SIZE macro.

[ mingo: Refined the changelog ]

Fixes: 6256e668b7af ("x86/kprobes: Use int3 instead of debug trap for single-step")
Suggested-by: Masami Hiramatsu (Google) <mhiramat@kernel.org>
Signed-off-by: Jinke Han <jinkehan@didiglobal.com>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Reviewed-by: Masami Hiramatsu (Google) <mhiramat@kernel.org>
Acked-by: Yafang Shao <laoar.shao@gmail.com>
Acked-by: Borislav Petkov <bp@alien8.de>
Cc: Peter Zijlstra <peterz@infradead.org>
Link: https://patch.msgid.link/20260908073742.GA10517@didi-ThinkCentre-M920t-N000

1 files changed, 134 insertions, 0 deletions

diff --git a/net/sched/em_ipset.c b/net/sched/em_ipset.c
new file mode 100644
index 000000000..c95cf86fb
--- /dev/null
+++ b/net/sched/em_ipset.c
@@ -0,0 +1,134 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/*
+ * net/sched/em_ipset.c	ipset ematch
+ *
+ * Copyright (c) 2012 Florian Westphal <fw@strlen.de>
+ */
+
+#include <linux/gfp.h>
+#include <linux/module.h>
+#include <linux/types.h>
+#include <linux/kernel.h>
+#include <linux/string.h>
+#include <linux/skbuff.h>
+#include <linux/netfilter/xt_set.h>
+#include <linux/ipv6.h>
+#include <net/ip.h>
+#include <net/pkt_cls.h>
+
+static int em_ipset_change(struct net *net, void *data, int data_len,
+			   struct tcf_ematch *em)
+{
+	struct xt_set_info *set = data;
+	ip_set_id_t index;
+
+	if (data_len != sizeof(*set))
+		return -EINVAL;
+
+	index = ip_set_nfnl_get_byindex(net, set->index);
+	if (index == IPSET_INVALID_ID)
+		return -ENOENT;
+
+	em->datalen = sizeof(*set);
+	em->data = (unsigned long)kmemdup(data, em->datalen, GFP_KERNEL);
+	if (em->data)
+		return 0;
+
+	ip_set_nfnl_put(net, index);
+	return -ENOMEM;
+}
+
+static void em_ipset_destroy(struct tcf_ematch *em)
+{
+	const struct xt_set_info *set = (const void *) em->data;
+	if (set) {
+		ip_set_nfnl_put(em->net, set->index);
+		kfree((void *) em->data);
+	}
+}
+
+static int em_ipset_match(struct sk_buff *skb, struct tcf_ematch *em,
+			  struct tcf_pkt_info *info)
+{
+	struct ip_set_adt_opt opt;
+	struct xt_action_param acpar;
+	const struct xt_set_info *set = (const void *) em->data;
+	struct net_device *dev, *indev = NULL;
+	struct nf_hook_state state = {
+		.net	= em->net,
+	};
+	int ret, network_offset;
+
+	switch (skb_protocol(skb, true)) {
+	case htons(ETH_P_IP):
+		state.pf = NFPROTO_IPV4;
+		if (!pskb_network_may_pull(skb, sizeof(struct iphdr)))
+			return 0;
+		acpar.thoff = ip_hdrlen(skb);
+		break;
+	case htons(ETH_P_IPV6):
+		state.pf = NFPROTO_IPV6;
+		if (!pskb_network_may_pull(skb, sizeof(struct ipv6hdr)))
+			return 0;
+		/* doesn't call ipv6_find_hdr() because ipset doesn't use thoff, yet */
+		acpar.thoff = sizeof(struct ipv6hdr);
+		break;
+	default:
+		return 0;
+	}
+
+	opt.family = state.pf;
+	opt.dim = set->dim;
+	opt.flags = set->flags;
+	opt.cmdflags = 0;
+	opt.ext.timeout = ~0u;
+
+	network_offset = skb_network_offset(skb);
+	skb_pull(skb, network_offset);
+
+	dev = skb->dev;
+
+	rcu_read_lock();
+
+	if (skb->skb_iif)
+		indev = dev_get_by_index_rcu(em->net, skb->skb_iif);
+
+	state.in      = indev ? indev : dev;
+	state.out     = dev;
+	acpar.state   = &state;
+
+	ret = ip_set_test(set->index, skb, &acpar, &opt);
+
+	rcu_read_unlock();
+
+	skb_push(skb, network_offset);
+	return ret;
+}
+
+static struct tcf_ematch_ops em_ipset_ops = {
+	.kind	  = TCF_EM_IPSET,
+	.change	  = em_ipset_change,
+	.destroy  = em_ipset_destroy,
+	.match	  = em_ipset_match,
+	.owner	  = THIS_MODULE,
+	.link	  = LIST_HEAD_INIT(em_ipset_ops.link)
+};
+
+static int __init init_em_ipset(void)
+{
+	return tcf_em_register(&em_ipset_ops);
+}
+
+static void __exit exit_em_ipset(void)
+{
+	tcf_em_unregister(&em_ipset_ops);
+}
+
+MODULE_LICENSE("GPL");
+MODULE_AUTHOR("Florian Westphal <fw@strlen.de>");
+MODULE_DESCRIPTION("TC extended match for IP sets");
+
+module_init(init_em_ipset);
+module_exit(exit_em_ipset);
+
+MODULE_ALIAS_TCF_EMATCH(TCF_EM_IPSET);