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, 96 insertions, 0 deletions

diff --git a/Documentation/i2c/i2c-protocol.rst b/Documentation/i2c/i2c-protocol.rst
new file mode 100644
index 000000000..df0febfe6
--- /dev/null
+++ b/Documentation/i2c/i2c-protocol.rst
@@ -0,0 +1,96 @@
+================
+The I2C Protocol
+================
+
+This document is an overview of the basic I2C transactions and the kernel
+APIs to perform them.
+
+Key to symbols
+==============
+
+=============== =============================================================
+S               Start condition
+P               Stop condition
+Rd/Wr (1 bit)   Read/Write bit. Rd equals 1, Wr equals 0.
+A, NA (1 bit)   Acknowledge (ACK) and Not Acknowledge (NACK) bit
+Addr  (7 bits)  I2C 7 bit address. Note that this can be expanded to
+                get a 10 bit I2C address.
+Data  (8 bits)  A plain data byte.
+
+[..]            Data sent by I2C device, as opposed to data sent by the
+                host adapter.
+=============== =============================================================
+
+
+Simple send transaction
+=======================
+
+Implemented by i2c_master_send()::
+
+  S Addr Wr [A] Data [A] Data [A] ... [A] Data [A] P
+
+
+Simple receive transaction
+==========================
+
+Implemented by i2c_master_recv()::
+
+  S Addr Rd [A] [Data] A [Data] A ... A [Data] NA P
+
+
+Combined transactions
+=====================
+
+Implemented by i2c_transfer().
+
+They are just like the above transactions, but instead of a stop
+condition P a start condition S is sent and the transaction continues.
+An example of a byte read, followed by a byte write::
+
+  S Addr Rd [A] [Data] NA S Addr Wr [A] Data [A] P
+
+
+Modified transactions
+=====================
+
+The following modifications to the I2C protocol can also be generated by
+setting these flags for I2C messages. With the exception of I2C_M_NOSTART, they
+are usually only needed to work around device issues:
+
+I2C_M_IGNORE_NAK:
+    Normally message is interrupted immediately if there is [NA] from the
+    client. Setting this flag treats any [NA] as [A], and all of
+    message is sent.
+    These messages may still fail to SCL lo->hi timeout.
+
+I2C_M_NO_RD_ACK:
+    In a read message, master A/NA bit is skipped.
+
+I2C_M_NOSTART:
+    In a combined transaction, no 'S Addr Wr/Rd [A]' is generated at some
+    point. For example, setting I2C_M_NOSTART on the second partial message
+    generates something like::
+
+      S Addr Rd [A] [Data] NA Data [A] P
+
+    If you set the I2C_M_NOSTART variable for the first partial message,
+    we do not generate Addr, but we do generate the start condition S.
+    This will probably confuse all other clients on your bus, so don't
+    try this.
+
+    This is often used to gather transmits from multiple data buffers in
+    system memory into something that appears as a single transfer to the
+    I2C device but may also be used between direction changes by some
+    rare devices.
+
+I2C_M_REV_DIR_ADDR:
+    This toggles the Rd/Wr flag. That is, if you want to do a write, but
+    need to emit an Rd instead of a Wr, or vice versa, you set this
+    flag. For example::
+
+      S Addr Rd [A] Data [A] Data [A] ... [A] Data [A] P
+
+I2C_M_STOP:
+    Force a stop condition (P) after the message. Some I2C related protocols
+    like SCCB require that. Normally, you really don't want to get interrupted
+    between the messages of one transfer.