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, 301 insertions, 0 deletions

diff --git a/tools/bootconfig/scripts/bconf2ftrace.sh b/tools/bootconfig/scripts/bconf2ftrace.sh
new file mode 100755
index 000000000..850c20734
--- /dev/null
+++ b/tools/bootconfig/scripts/bconf2ftrace.sh
@@ -0,0 +1,301 @@
+#!/bin/sh
+# SPDX-License-Identifier: GPL-2.0-only
+
+usage() {
+	echo "Ftrace boottime trace test tool"
+	echo "Usage: $0 [--apply|--init] [--debug] BOOTCONFIG-FILE"
+	echo "    --apply: Test actual apply to tracefs (need sudo)"
+	echo "    --init:  Initialize ftrace before applying (imply --apply)"
+	exit 1
+}
+
+[ $# -eq 0 ] && usage
+
+BCONF=
+DEBUG=
+APPLY=
+INIT=
+while [ x"$1" != x ]; do
+	case "$1" in
+	"--debug")
+		DEBUG=$1;;
+	"--apply")
+		APPLY=$1;;
+	"--init")
+		APPLY=$1
+		INIT=$1;;
+	*)
+		[ ! -f $1 ] && usage
+		BCONF=$1;;
+	esac
+	shift 1
+done
+
+if [ x"$APPLY" != x ]; then
+	if [ `id -u` -ne 0 ]; then
+		echo "This must be run by root user. Try sudo." 1>&2
+		exec sudo $0 $DEBUG $APPLY $BCONF
+	fi
+fi
+
+run_cmd() { # command
+	echo "$*"
+	if [ x"$APPLY" != x ]; then # apply command
+		eval $*
+	fi
+}
+
+if [ x"$DEBUG" != x ]; then
+	set -x
+fi
+
+TRACEFS=`grep -m 1 -w tracefs /proc/mounts | cut -f 2 -d " "`
+if [ -z "$TRACEFS" ]; then
+	if ! grep -wq debugfs /proc/mounts; then
+		echo "Error: No tracefs/debugfs was mounted." 1>&2
+		exit 1
+	fi
+	TRACEFS=`grep -m 1 -w debugfs /proc/mounts | cut -f 2 -d " "`/tracing
+	if [ ! -d $TRACEFS ]; then
+		echo "Error: ftrace is not enabled on this kernel." 1>&2
+		exit 1
+	fi
+fi
+
+if [ x"$INIT" != x ]; then
+	. `dirname $0`/ftrace.sh
+	(cd $TRACEFS; initialize_ftrace)
+fi
+
+. `dirname $0`/xbc.sh
+
+######## main #########
+set -e
+
+xbc_init $BCONF
+
+set_value_of() { # key file
+	if xbc_has_key $1; then
+		val=`xbc_get_val $1 1`
+		run_cmd "echo '$val' >> $2"
+	fi
+}
+
+set_array_of() { # key file
+	if xbc_has_key $1; then
+		xbc_get_val $1 | while read line; do
+			run_cmd "echo '$line' >> $2"
+		done
+	fi
+}
+
+compose_synth() { # event_name branch
+	echo -n "$1 "
+	xbc_get_val $2 | while read field; do echo -n "$field; "; done
+}
+
+print_hist_array() { # prefix key
+	__sep="="
+	if xbc_has_key ${1}.${2}; then
+		echo -n ":$2"
+		xbc_get_val ${1}.${2} | while read field; do
+			echo -n "$__sep$field"; __sep=","
+		done
+	fi
+}
+
+print_hist_action_array() { # prefix key
+	__sep="("
+	echo -n ".$2"
+	xbc_get_val ${1}.${2} | while read field; do
+		echo -n "$__sep$field"; __sep=","
+	done
+	echo -n ")"
+}
+
+print_hist_one_action() { # prefix handler param
+	echo -n ":${2}("`xbc_get_val ${1}.${3}`")"
+	if xbc_has_key "${1}.trace"; then
+		print_hist_action_array ${1} "trace"
+	elif xbc_has_key "${1}.save"; then
+		print_hist_action_array ${1} "save"
+	elif xbc_has_key "${1}.snapshot"; then
+		echo -n ".snapshot()"
+	fi
+}
+
+print_hist_actions() { # prefix handler param
+	for __hdr in `xbc_subkeys ${1}.${2} 1 ".[0-9]"`; do
+		print_hist_one_action ${1}.${2}.$__hdr ${2} ${3}
+	done
+	if xbc_has_key ${1}.${2}.${3} ; then
+		print_hist_one_action ${1}.${2} ${2} ${3}
+	fi
+}
+
+print_hist_var() { # prefix varname
+	echo -n ":${2}="`xbc_get_val ${1}.var.${2} | tr -d [:space:]`
+}
+
+print_one_histogram() { # prefix
+	echo -n "hist"
+	print_hist_array $1 "keys"
+	print_hist_array $1 "values"
+	print_hist_array $1 "sort"
+	if xbc_has_key "${1}.size"; then
+		echo -n ":size="`xbc_get_val ${1}.size`
+	fi
+	if xbc_has_key "${1}.name"; then
+		echo -n ":name="`xbc_get_val ${1}.name`
+	fi
+	for __var in `xbc_subkeys "${1}.var" 1`; do
+		print_hist_var ${1} ${__var}
+	done
+	if xbc_has_key "${1}.pause"; then
+		echo -n ":pause"
+	elif xbc_has_key "${1}.continue"; then
+		echo -n ":continue"
+	elif xbc_has_key "${1}.clear"; then
+		echo -n ":clear"
+	fi
+	print_hist_actions ${1} "onmax" "var"
+	print_hist_actions ${1} "onchange" "var"
+	print_hist_actions ${1} "onmatch" "event"
+
+	if xbc_has_key "${1}.filter"; then
+		echo -n " if "`xbc_get_val ${1}.filter`
+	fi
+}
+
+setup_one_histogram() { # prefix trigger-file
+	run_cmd "echo '`print_one_histogram ${1}`' >> ${2}"
+}
+
+setup_histograms() { # prefix trigger-file
+	for __hist in `xbc_subkeys ${1} 1 ".[0-9]"`; do
+		setup_one_histogram ${1}.$__hist ${2}
+	done
+	if xbc_has_key ${1}.keys; then
+		setup_one_histogram ${1} ${2}
+	fi
+}
+
+setup_event() { # prefix group event [instance]
+	branch=$1.$2.$3
+	if [ "$4" ]; then
+		eventdir="$TRACEFS/instances/$4/events/$2/$3"
+	else
+		eventdir="$TRACEFS/events/$2/$3"
+	fi
+	# group enable
+	if [ "$3" = "enable" ]; then
+		run_cmd "echo 1 > ${eventdir}"
+		return
+	fi
+
+	case $2 in
+	kprobes)
+		xbc_get_val ${branch}.probes | while read line; do
+			run_cmd "echo 'p:kprobes/$3 $line' >> $TRACEFS/kprobe_events"
+		done
+		;;
+	synthetic)
+		run_cmd "echo '`compose_synth $3 ${branch}.fields`' >> $TRACEFS/synthetic_events"
+		;;
+	esac
+
+	set_value_of ${branch}.filter ${eventdir}/filter
+	set_array_of ${branch}.actions ${eventdir}/trigger
+
+	setup_histograms ${branch}.hist ${eventdir}/trigger
+
+	if xbc_has_key ${branch}.enable; then
+		run_cmd "echo 1 > ${eventdir}/enable"
+	fi
+}
+
+setup_events() { # prefix("ftrace" or "ftrace.instance.INSTANCE") [instance]
+	prefix="${1}.event"
+	if xbc_has_branch ${1}.event; then
+		for grpev in `xbc_subkeys ${1}.event 2`; do
+			setup_event $prefix ${grpev%.*} ${grpev#*.} $2
+		done
+	fi
+	if xbc_has_branch ${1}.event.enable; then
+		if [ "$2" ]; then
+			run_cmd "echo 1 > $TRACEFS/instances/$2/events/enable"
+		else
+			run_cmd "echo 1 > $TRACEFS/events/enable"
+		fi
+	fi
+}
+
+size2kb() { # size[KB|MB]
+	case $1 in
+	*KB)
+		echo ${1%KB};;
+	*MB)
+		expr ${1%MB} \* 1024;;
+	*)
+		expr $1 / 1024 ;;
+	esac
+}
+
+setup_instance() { # [instance]
+	if [ "$1" ]; then
+		instance="ftrace.instance.${1}"
+		instancedir=$TRACEFS/instances/$1
+	else
+		instance="ftrace"
+		instancedir=$TRACEFS
+	fi
+
+	set_array_of ${instance}.options ${instancedir}/trace_options
+	set_value_of ${instance}.trace_clock ${instancedir}/trace_clock
+	set_value_of ${instance}.cpumask ${instancedir}/tracing_cpumask
+	set_value_of ${instance}.tracing_on ${instancedir}/tracing_on
+	set_value_of ${instance}.tracer ${instancedir}/current_tracer
+	set_array_of ${instance}.ftrace.filters \
+		${instancedir}/set_ftrace_filter
+	set_array_of ${instance}.ftrace.notrace \
+		${instancedir}/set_ftrace_notrace
+
+	if xbc_has_key ${instance}.alloc_snapshot; then
+		run_cmd "echo 1 > ${instancedir}/snapshot"
+	fi
+
+	if xbc_has_key ${instance}.buffer_size; then
+		size=`xbc_get_val ${instance}.buffer_size 1`
+		size=`eval size2kb $size`
+		run_cmd "echo $size >> ${instancedir}/buffer_size_kb"
+	fi
+
+	setup_events ${instance} $1
+	set_array_of ${instance}.events ${instancedir}/set_event
+}
+
+# ftrace global configs (kernel.*)
+if xbc_has_key "kernel.dump_on_oops"; then
+	dump_mode=`xbc_get_val "kernel.dump_on_oops" 1`
+	[ "$dump_mode" ] && dump_mode=`eval echo $dump_mode` || dump_mode=1
+	run_cmd "echo \"$dump_mode\" > /proc/sys/kernel/ftrace_dump_on_oops"
+fi
+
+set_value_of kernel.fgraph_max_depth $TRACEFS/max_graph_depth
+set_array_of kernel.fgraph_filters $TRACEFS/set_graph_function
+set_array_of kernel.fgraph_notraces $TRACEFS/set_graph_notrace
+
+# Per-instance/per-event configs
+if ! xbc_has_branch "ftrace" ; then
+	exit 0
+fi
+
+setup_instance # root instance
+
+if xbc_has_branch "ftrace.instance"; then
+	for i in `xbc_subkeys "ftrace.instance" 1`; do
+		run_cmd "mkdir -p $TRACEFS/instances/$i"
+		setup_instance $i
+	done
+fi
+