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, 99 insertions, 0 deletions

diff --git a/net/sched/em_cmp.c b/net/sched/em_cmp.c
new file mode 100644
index 000000000..48c1bce74
--- /dev/null
+++ b/net/sched/em_cmp.c
@@ -0,0 +1,99 @@
+// SPDX-License-Identifier: GPL-2.0-or-later
+/*
+ * net/sched/em_cmp.c	Simple packet data comparison ematch
+ *
+ * Authors:	Thomas Graf <tgraf@suug.ch>
+ */
+
+#include <linux/module.h>
+#include <linux/types.h>
+#include <linux/kernel.h>
+#include <linux/skbuff.h>
+#include <linux/tc_ematch/tc_em_cmp.h>
+#include <linux/unaligned.h>
+#include <net/pkt_cls.h>
+
+static inline int cmp_needs_transformation(struct tcf_em_cmp *cmp)
+{
+	return unlikely(cmp->flags & TCF_EM_CMP_TRANS);
+}
+
+static int em_cmp_match(struct sk_buff *skb, struct tcf_ematch *em,
+			struct tcf_pkt_info *info)
+{
+	struct tcf_em_cmp *cmp = (struct tcf_em_cmp *) em->data;
+	unsigned char *ptr = tcf_get_base_ptr(skb, cmp->layer);
+	u32 val = 0;
+
+	if (!ptr)
+		return 0;
+	ptr += cmp->off;
+	if (!tcf_valid_offset(skb, ptr, cmp->align))
+		return 0;
+
+	switch (cmp->align) {
+	case TCF_EM_ALIGN_U8:
+		val = *ptr;
+		break;
+
+	case TCF_EM_ALIGN_U16:
+		val = get_unaligned_be16(ptr);
+
+		if (cmp_needs_transformation(cmp))
+			val = be16_to_cpu(val);
+		break;
+
+	case TCF_EM_ALIGN_U32:
+		/* Worth checking boundaries? The branching seems
+		 * to get worse. Visit again.
+		 */
+		val = get_unaligned_be32(ptr);
+
+		if (cmp_needs_transformation(cmp))
+			val = be32_to_cpu(val);
+		break;
+
+	default:
+		return 0;
+	}
+
+	if (cmp->mask)
+		val &= cmp->mask;
+
+	switch (cmp->opnd) {
+	case TCF_EM_OPND_EQ:
+		return val == cmp->val;
+	case TCF_EM_OPND_LT:
+		return val < cmp->val;
+	case TCF_EM_OPND_GT:
+		return val > cmp->val;
+	}
+
+	return 0;
+}
+
+static struct tcf_ematch_ops em_cmp_ops = {
+	.kind	  = TCF_EM_CMP,
+	.datalen  = sizeof(struct tcf_em_cmp),
+	.match	  = em_cmp_match,
+	.owner	  = THIS_MODULE,
+	.link	  = LIST_HEAD_INIT(em_cmp_ops.link)
+};
+
+static int __init init_em_cmp(void)
+{
+	return tcf_em_register(&em_cmp_ops);
+}
+
+static void __exit exit_em_cmp(void)
+{
+	tcf_em_unregister(&em_cmp_ops);
+}
+
+MODULE_DESCRIPTION("ematch classifier for basic data types(8/16/32 bit) against skb data");
+MODULE_LICENSE("GPL");
+
+module_init(init_em_cmp);
+module_exit(exit_em_cmp);
+
+MODULE_ALIAS_TCF_EMATCH(TCF_EM_CMP);