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, 104 insertions, 0 deletions

diff --git a/Documentation/arch/powerpc/htm.rst b/Documentation/arch/powerpc/htm.rst
new file mode 100644
index 000000000..fcb4eb630
--- /dev/null
+++ b/Documentation/arch/powerpc/htm.rst
@@ -0,0 +1,104 @@
+.. SPDX-License-Identifier: GPL-2.0
+.. _htm:
+
+===================================
+HTM (Hardware Trace Macro)
+===================================
+
+Athira Rajeev, 2 Mar 2025
+
+.. contents::
+    :depth: 3
+
+
+Basic overview
+==============
+
+H_HTM is used as an interface for executing Hardware Trace Macro (HTM)
+functions, including setup, configuration, control and dumping of the HTM data.
+For using HTM, it is required to setup HTM buffers and HTM operations can
+be controlled using the H_HTM hcall. The hcall can be invoked for any core/chip
+of the system from within a partition itself. To use this feature, a debugfs
+folder called "htmdump" is present under /sys/kernel/debug/powerpc.
+
+
+HTM debugfs example usage
+=========================
+
+.. code-block:: sh
+
+  #  ls /sys/kernel/debug/powerpc/htmdump/
+  coreindexonchip  htmcaps  htmconfigure  htmflags  htminfo  htmsetup
+  htmstart  htmstatus  htmtype  nodalchipindex  nodeindex  trace
+
+Details on each file:
+
+* nodeindex, nodalchipindex, coreindexonchip specifies which partition to configure the HTM for.
+* htmtype: specifies the type of HTM. Supported target is hardwareTarget.
+* trace: is to read the HTM data.
+* htmconfigure: Configure/Deconfigure the HTM. Writing 1 to the file will configure the trace, writing 0 to the file will do deconfigure.
+* htmstart: start/Stop the HTM. Writing 1 to the file will start the tracing, writing 0 to the file will stop the tracing.
+* htmstatus: get the status of HTM. This is needed to understand the HTM state after each operation.
+* htmsetup: set the HTM buffer size. Size of HTM buffer is in power of 2
+* htminfo: provides the system processor configuration details. This is needed to understand the appropriate values for nodeindex, nodalchipindex, coreindexonchip.
+* htmcaps : provides the HTM capabilities like minimum/maximum buffer size, what kind of tracing the HTM supports etc.
+* htmflags : allows to pass flags to hcall. Currently supports controlling the wrapping of HTM buffer.
+
+To see the system processor configuration details:
+
+.. code-block:: sh
+
+  # cat /sys/kernel/debug/powerpc/htmdump/htminfo > htminfo_file
+
+The result can be interpreted using hexdump.
+
+To collect HTM traces for a partition represented by nodeindex as
+zero, nodalchipindex as 1 and coreindexonchip as 12
+
+.. code-block:: sh
+
+  # cd /sys/kernel/debug/powerpc/htmdump/
+  # echo 2 > htmtype
+  # echo 33 > htmsetup ( sets 8GB memory for HTM buffer, number is size in power of 2 )
+
+This requires a CEC reboot to get the HTM buffers allocated.
+
+.. code-block:: sh
+
+  # cd /sys/kernel/debug/powerpc/htmdump/
+  # echo 2 > htmtype
+  # echo 0 > nodeindex
+  # echo 1 > nodalchipindex
+  # echo 12 > coreindexonchip
+  # echo 1 > htmflags     # to set noWrap for HTM buffers
+  # echo 1 > htmconfigure # Configure the HTM
+  # echo 1 > htmstart     # Start the HTM
+  # echo 0 > htmstart     # Stop the HTM
+  # echo 0 > htmconfigure # Deconfigure the HTM
+  # cat htmstatus         # Dump the status of HTM entries as data
+
+Above will set the htmtype and core details, followed by executing respective HTM operation.
+
+Read the HTM trace data
+========================
+
+After starting the trace collection, run the workload
+of interest. Stop the trace collection after required period
+of time, and read the trace file.
+
+.. code-block:: sh
+
+  # cat /sys/kernel/debug/powerpc/htmdump/trace > trace_file
+
+This trace file will contain the relevant instruction traces
+collected during the workload execution. And can be used as
+input file for trace decoders to understand data.
+
+Benefits of using HTM debugfs interface
+=======================================
+
+It is now possible to collect traces for a particular core/chip
+from within any partition of the system and decode it. Through
+this enablement, a small partition can be dedicated to collect the
+trace data and analyze to provide important information for Performance
+analysis, Software tuning, or Hardware debug.