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, 185 insertions, 0 deletions

diff --git a/arch/powerpc/boot/libfdt-wrapper.c b/arch/powerpc/boot/libfdt-wrapper.c
new file mode 100644
index 000000000..217d0d7a6
--- /dev/null
+++ b/arch/powerpc/boot/libfdt-wrapper.c
@@ -0,0 +1,185 @@
+// SPDX-License-Identifier: GPL-2.0-or-later
+/*
+ * This file does the necessary interface mapping between the bootwrapper
+ * device tree operations and the interface provided by shared source
+ * files flatdevicetree.[ch].
+ *
+ * Copyright 2007 David Gibson, IBM Corporation.
+ */
+
+#include <stddef.h>
+#include <stdio.h>
+#include <page.h>
+#include <libfdt.h>
+#include "ops.h"
+
+#define DEBUG	0
+#define BAD_ERROR(err)	(((err) < 0) \
+			 && ((err) != -FDT_ERR_NOTFOUND) \
+			 && ((err) != -FDT_ERR_EXISTS))
+
+#define check_err(err) \
+	({ \
+		if (BAD_ERROR(err) || ((err < 0) && DEBUG)) \
+			printf("%s():%d  %s\n\r", __func__, __LINE__, \
+			       fdt_strerror(err)); \
+		if (BAD_ERROR(err)) \
+			exit(); \
+		(err < 0) ? -1 : 0; \
+	})
+
+#define offset_devp(off)	\
+	({ \
+		unsigned long _offset = (off); \
+		check_err(_offset) ? NULL : (void *)(_offset+1); \
+	})
+
+#define devp_offset_find(devp)	(((unsigned long)(devp))-1)
+#define devp_offset(devp)	(devp ? ((unsigned long)(devp))-1 : 0)
+
+static void *fdt;
+static void *buf; /* = NULL */
+
+#define EXPAND_GRANULARITY	1024
+
+static void expand_buf(int minexpand)
+{
+	int size = fdt_totalsize(fdt);
+	int rc;
+
+	size = _ALIGN(size + minexpand, EXPAND_GRANULARITY);
+	buf = platform_ops.realloc(buf, size);
+	if (!buf)
+		fatal("Couldn't find %d bytes to expand device tree\n\r", size);
+	rc = fdt_open_into(fdt, buf, size);
+	if (rc != 0)
+		fatal("Couldn't expand fdt into new buffer: %s\n\r",
+		      fdt_strerror(rc));
+
+	fdt = buf;
+}
+
+static void *fdt_wrapper_finddevice(const char *path)
+{
+	return offset_devp(fdt_path_offset(fdt, path));
+}
+
+static int fdt_wrapper_getprop(const void *devp, const char *name,
+			       void *buf, const int buflen)
+{
+	const void *p;
+	int len;
+
+	p = fdt_getprop(fdt, devp_offset(devp), name, &len);
+	if (!p)
+		return check_err(len);
+	memcpy(buf, p, min(len, buflen));
+	return len;
+}
+
+static int fdt_wrapper_setprop(const void *devp, const char *name,
+			       const void *buf, const int len)
+{
+	int rc;
+
+	rc = fdt_setprop(fdt, devp_offset(devp), name, buf, len);
+	if (rc == -FDT_ERR_NOSPACE) {
+		expand_buf(len + 16);
+		rc = fdt_setprop(fdt, devp_offset(devp), name, buf, len);
+	}
+
+	return check_err(rc);
+}
+
+static int fdt_wrapper_del_node(const void *devp)
+{
+	return fdt_del_node(fdt, devp_offset(devp));
+}
+
+static void *fdt_wrapper_get_parent(const void *devp)
+{
+	return offset_devp(fdt_parent_offset(fdt, devp_offset(devp)));
+}
+
+static void *fdt_wrapper_create_node(const void *devp, const char *name)
+{
+	int offset;
+
+	offset = fdt_add_subnode(fdt, devp_offset(devp), name);
+	if (offset == -FDT_ERR_NOSPACE) {
+		expand_buf(strlen(name) + 16);
+		offset = fdt_add_subnode(fdt, devp_offset(devp), name);
+	}
+
+	return offset_devp(offset);
+}
+
+static void *fdt_wrapper_find_node_by_prop_value(const void *prev,
+						 const char *name,
+						 const char *val,
+						 int len)
+{
+	int offset = fdt_node_offset_by_prop_value(fdt, devp_offset_find(prev),
+	                                           name, val, len);
+	return offset_devp(offset);
+}
+
+static void *fdt_wrapper_find_node_by_compatible(const void *prev,
+						 const char *val)
+{
+	int offset = fdt_node_offset_by_compatible(fdt, devp_offset_find(prev),
+	                                           val);
+	return offset_devp(offset);
+}
+
+static char *fdt_wrapper_get_path(const void *devp, char *buf, int len)
+{
+	int rc;
+
+	rc = fdt_get_path(fdt, devp_offset(devp), buf, len);
+	if (check_err(rc))
+		return NULL;
+	return buf;
+}
+
+static unsigned long fdt_wrapper_finalize(void)
+{
+	int rc;
+
+	rc = fdt_pack(fdt);
+	if (rc != 0)
+		fatal("Couldn't pack flat tree: %s\n\r",
+		      fdt_strerror(rc));
+	return (unsigned long)fdt;
+}
+
+void fdt_init(void *blob)
+{
+	int err;
+	int bufsize;
+
+	dt_ops.finddevice = fdt_wrapper_finddevice;
+	dt_ops.getprop = fdt_wrapper_getprop;
+	dt_ops.setprop = fdt_wrapper_setprop;
+	dt_ops.get_parent = fdt_wrapper_get_parent;
+	dt_ops.create_node = fdt_wrapper_create_node;
+	dt_ops.find_node_by_prop_value = fdt_wrapper_find_node_by_prop_value;
+	dt_ops.find_node_by_compatible = fdt_wrapper_find_node_by_compatible;
+	dt_ops.del_node = fdt_wrapper_del_node;
+	dt_ops.get_path = fdt_wrapper_get_path;
+	dt_ops.finalize = fdt_wrapper_finalize;
+
+	/* Make sure the dt blob is the right version and so forth */
+	fdt = blob;
+	bufsize = fdt_totalsize(fdt) + EXPAND_GRANULARITY;
+	buf = malloc(bufsize);
+	if(!buf)
+		fatal("malloc failed. can't relocate the device tree\n\r");
+
+	err = fdt_open_into(fdt, buf, bufsize);
+
+	if (err != 0)
+		fatal("fdt_init(): %s\n\r", fdt_strerror(err));
+
+	fdt = buf;
+}