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, 118 insertions, 0 deletions

diff --git a/net/ipv6/rpl.c b/net/ipv6/rpl.c
new file mode 100644
index 000000000..e186998bf
--- /dev/null
+++ b/net/ipv6/rpl.c
@@ -0,0 +1,118 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/*
+ * Authors:
+ * (C) 2020 Alexander Aring <alex.aring@gmail.com>
+ */
+
+#include <net/ipv6.h>
+#include <net/rpl.h>
+
+#define IPV6_PFXTAIL_LEN(x) (sizeof(struct in6_addr) - (x))
+#define IPV6_RPL_BEST_ADDR_COMPRESSION 15
+
+static void ipv6_rpl_addr_decompress(struct in6_addr *dst,
+				     const struct in6_addr *daddr,
+				     const void *post, unsigned char pfx)
+{
+	memcpy(dst, daddr, pfx);
+	memcpy(&dst->s6_addr[pfx], post, IPV6_PFXTAIL_LEN(pfx));
+}
+
+static void ipv6_rpl_addr_compress(void *dst, const struct in6_addr *addr,
+				   unsigned char pfx)
+{
+	memcpy(dst, &addr->s6_addr[pfx], IPV6_PFXTAIL_LEN(pfx));
+}
+
+static void *ipv6_rpl_segdata_pos(const struct ipv6_rpl_sr_hdr *hdr, int i)
+{
+	return (void *)&hdr->rpl_segdata[i * IPV6_PFXTAIL_LEN(hdr->cmpri)];
+}
+
+void ipv6_rpl_srh_decompress(struct ipv6_rpl_sr_hdr *outhdr,
+			     const struct ipv6_rpl_sr_hdr *inhdr,
+			     const struct in6_addr *daddr, unsigned char n)
+{
+	int i;
+
+	outhdr->nexthdr = inhdr->nexthdr;
+	outhdr->hdrlen = (((n + 1) * sizeof(struct in6_addr)) >> 3);
+	outhdr->pad = 0;
+	outhdr->type = inhdr->type;
+	outhdr->segments_left = inhdr->segments_left;
+	outhdr->cmpri = 0;
+	outhdr->cmpre = 0;
+
+	for (i = 0; i < n; i++)
+		ipv6_rpl_addr_decompress(&outhdr->rpl_segaddr[i], daddr,
+					 ipv6_rpl_segdata_pos(inhdr, i),
+					 inhdr->cmpri);
+
+	ipv6_rpl_addr_decompress(&outhdr->rpl_segaddr[n], daddr,
+				 ipv6_rpl_segdata_pos(inhdr, n),
+				 inhdr->cmpre);
+}
+
+static unsigned char ipv6_rpl_srh_calc_cmpri(const struct ipv6_rpl_sr_hdr *inhdr,
+					     const struct in6_addr *daddr,
+					     unsigned char n)
+{
+	unsigned char plen;
+	int i;
+
+	for (plen = 0; plen < sizeof(*daddr); plen++) {
+		for (i = 0; i < n; i++) {
+			if (daddr->s6_addr[plen] !=
+			    inhdr->rpl_segaddr[i].s6_addr[plen])
+				return plen;
+		}
+	}
+
+	return IPV6_RPL_BEST_ADDR_COMPRESSION;
+}
+
+static unsigned char ipv6_rpl_srh_calc_cmpre(const struct in6_addr *daddr,
+					     const struct in6_addr *last_segment)
+{
+	unsigned int plen;
+
+	for (plen = 0; plen < sizeof(*daddr); plen++) {
+		if (daddr->s6_addr[plen] != last_segment->s6_addr[plen])
+			return plen;
+	}
+
+	return IPV6_RPL_BEST_ADDR_COMPRESSION;
+}
+
+void ipv6_rpl_srh_compress(struct ipv6_rpl_sr_hdr *outhdr,
+			   const struct ipv6_rpl_sr_hdr *inhdr,
+			   const struct in6_addr *daddr, unsigned char n)
+{
+	unsigned char cmpri, cmpre;
+	size_t seglen;
+	int i;
+
+	cmpri = ipv6_rpl_srh_calc_cmpri(inhdr, daddr, n);
+	cmpre = ipv6_rpl_srh_calc_cmpre(daddr, &inhdr->rpl_segaddr[n]);
+
+	outhdr->nexthdr = inhdr->nexthdr;
+	seglen = (n * IPV6_PFXTAIL_LEN(cmpri)) + IPV6_PFXTAIL_LEN(cmpre);
+	outhdr->hdrlen = seglen >> 3;
+	if (seglen & 0x7) {
+		outhdr->hdrlen++;
+		outhdr->pad = 8 - (seglen & 0x7);
+	} else {
+		outhdr->pad = 0;
+	}
+	outhdr->type = inhdr->type;
+	outhdr->segments_left = inhdr->segments_left;
+	outhdr->cmpri = cmpri;
+	outhdr->cmpre = cmpre;
+
+	for (i = 0; i < n; i++)
+		ipv6_rpl_addr_compress(ipv6_rpl_segdata_pos(outhdr, i),
+				       &inhdr->rpl_segaddr[i], cmpri);
+
+	ipv6_rpl_addr_compress(ipv6_rpl_segdata_pos(outhdr, n),
+			       &inhdr->rpl_segaddr[n], cmpre);
+}