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, 92 insertions, 0 deletions

diff --git a/arch/mips/dec/wbflush.c b/arch/mips/dec/wbflush.c
new file mode 100644
index 000000000..dad64d178
--- /dev/null
+++ b/arch/mips/dec/wbflush.c
@@ -0,0 +1,92 @@
+/*
+ * Setup the right wbflush routine for the different DECstations.
+ *
+ * Created with information from:
+ *	DECstation 3100 Desktop Workstation Functional Specification
+ *	DECstation 5000/200 KN02 System Module Functional Specification
+ *	mipsel-linux-objdump --disassemble vmunix | grep "wbflush" :-)
+ *
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
+ *
+ * Copyright (C) 1998 Harald Koerfgen
+ * Copyright (C) 2002 Maciej W. Rozycki
+ */
+
+#include <linux/export.h>
+#include <linux/init.h>
+
+#include <asm/bootinfo.h>
+#include <asm/wbflush.h>
+#include <asm/barrier.h>
+
+static void wbflush_kn01(void);
+static void wbflush_kn210(void);
+static void wbflush_mips(void);
+
+void (*__wbflush) (void);
+
+void __init wbflush_setup(void)
+{
+	switch (mips_machtype) {
+	case MACH_DS23100:
+	case MACH_DS5000_200:	/* DS5000 3max */
+		__wbflush = wbflush_kn01;
+		break;
+	case MACH_DS5100:	/* DS5100 MIPSMATE */
+		__wbflush = wbflush_kn210;
+		break;
+	case MACH_DS5000_1XX:	/* DS5000/100 3min */
+	case MACH_DS5000_XX:	/* Personal DS5000/2x */
+	case MACH_DS5000_2X0:	/* DS5000/240 3max+ */
+	case MACH_DS5900:	/* DS5900 bigmax */
+	default:
+		__wbflush = wbflush_mips;
+		break;
+	}
+}
+
+/*
+ * For the DS3100 and DS5000/200 the R2020/R3220 writeback buffer functions
+ * as part of Coprocessor 0.
+ */
+static void wbflush_kn01(void)
+{
+    asm(".set\tpush\n\t"
+	".set\tnoreorder\n\t"
+	"1:\tbc0f\t1b\n\t"
+	"nop\n\t"
+	".set\tpop");
+}
+
+/*
+ * For the DS5100 the writeback buffer seems to be a part of Coprocessor 3.
+ * But CP3 has to enabled first.
+ */
+static void wbflush_kn210(void)
+{
+    asm(".set\tpush\n\t"
+	".set\tnoreorder\n\t"
+	"mfc0\t$2,$12\n\t"
+	"lui\t$3,0x8000\n\t"
+	"or\t$3,$2,$3\n\t"
+	"mtc0\t$3,$12\n\t"
+	"nop\n"
+	"1:\tbc3f\t1b\n\t"
+	"nop\n\t"
+	"mtc0\t$2,$12\n\t"
+	"nop\n\t"
+	".set\tpop"
+	: : : "$2", "$3");
+}
+
+/*
+ * I/O ASIC systems use a standard writeback buffer that gets flushed
+ * upon an uncached read.
+ */
+static void wbflush_mips(void)
+{
+	__fast_iob();
+}
+EXPORT_SYMBOL(__wbflush);