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, 507 insertions, 0 deletions

diff --git a/net/can/proc.c b/net/can/proc.c
new file mode 100644
index 000000000..de4d05ae3
--- /dev/null
+++ b/net/can/proc.c
@@ -0,0 +1,507 @@
+// SPDX-License-Identifier: (GPL-2.0 OR BSD-3-Clause)
+/*
+ * proc.c - procfs support for Protocol family CAN core module
+ *
+ * Copyright (c) 2002-2007 Volkswagen Group Electronic Research
+ * All rights reserved.
+ *
+ * Redistribution and use in source and binary forms, with or without
+ * modification, are permitted provided that the following conditions
+ * are met:
+ * 1. Redistributions of source code must retain the above copyright
+ *    notice, this list of conditions and the following disclaimer.
+ * 2. Redistributions in binary form must reproduce the above copyright
+ *    notice, this list of conditions and the following disclaimer in the
+ *    documentation and/or other materials provided with the distribution.
+ * 3. Neither the name of Volkswagen nor the names of its contributors
+ *    may be used to endorse or promote products derived from this software
+ *    without specific prior written permission.
+ *
+ * Alternatively, provided that this notice is retained in full, this
+ * software may be distributed under the terms of the GNU General
+ * Public License ("GPL") version 2, in which case the provisions of the
+ * GPL apply INSTEAD OF those given above.
+ *
+ * The provided data structures and external interfaces from this code
+ * are not restricted to be used by modules with a GPL compatible license.
+ *
+ * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
+ * "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
+ * LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
+ * A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT
+ * OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
+ * SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
+ * LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
+ * DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
+ * THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
+ * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
+ * OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH
+ * DAMAGE.
+ *
+ */
+
+#include <linux/module.h>
+#include <linux/proc_fs.h>
+#include <linux/list.h>
+#include <linux/rcupdate.h>
+#include <linux/if_arp.h>
+#include <linux/can/can-ml.h>
+#include <linux/can/core.h>
+
+#include "af_can.h"
+
+/*
+ * proc filenames for the PF_CAN core
+ */
+
+#define CAN_PROC_STATS       "stats"
+#define CAN_PROC_RESET_STATS "reset_stats"
+#define CAN_PROC_RCVLIST_ALL "rcvlist_all"
+#define CAN_PROC_RCVLIST_FIL "rcvlist_fil"
+#define CAN_PROC_RCVLIST_INV "rcvlist_inv"
+#define CAN_PROC_RCVLIST_SFF "rcvlist_sff"
+#define CAN_PROC_RCVLIST_EFF "rcvlist_eff"
+#define CAN_PROC_RCVLIST_ERR "rcvlist_err"
+
+static int user_reset;
+
+static const char rx_list_name[][8] = {
+	[RX_ERR] = "rx_err",
+	[RX_ALL] = "rx_all",
+	[RX_FIL] = "rx_fil",
+	[RX_INV] = "rx_inv",
+};
+
+/*
+ * af_can statistics stuff
+ */
+
+static void can_init_stats(struct net *net)
+{
+	struct can_pkg_stats *pkg_stats = net->can.pkg_stats;
+	struct can_rcv_lists_stats *rcv_lists_stats = net->can.rcv_lists_stats;
+	/*
+	 * This memset function is called from a timer context (when
+	 * can_stattimer is active which is the default) OR in a process
+	 * context (reading the proc_fs when can_stattimer is disabled).
+	 */
+	memset(pkg_stats, 0, sizeof(struct can_pkg_stats));
+	pkg_stats->jiffies_init = jiffies;
+
+	rcv_lists_stats->stats_reset++;
+
+	if (user_reset) {
+		user_reset = 0;
+		rcv_lists_stats->user_reset++;
+	}
+}
+
+static unsigned long calc_rate(unsigned long oldjif, unsigned long newjif,
+			       unsigned long count)
+{
+	if (oldjif == newjif)
+		return 0;
+
+	/* see can_stat_update() - this should NEVER happen! */
+	if (count > (ULONG_MAX / HZ)) {
+		printk(KERN_ERR "can: calc_rate: count exceeded! %ld\n",
+		       count);
+		return 99999999;
+	}
+
+	return (count * HZ) / (newjif - oldjif);
+}
+
+void can_stat_update(struct timer_list *t)
+{
+	struct net *net = timer_container_of(net, t, can.stattimer);
+	struct can_pkg_stats *pkg_stats = net->can.pkg_stats;
+	unsigned long j = jiffies; /* snapshot */
+
+	long rx_frames = atomic_long_read(&pkg_stats->rx_frames);
+	long tx_frames = atomic_long_read(&pkg_stats->tx_frames);
+	long matches = atomic_long_read(&pkg_stats->matches);
+	long rx_frames_delta = atomic_long_read(&pkg_stats->rx_frames_delta);
+	long tx_frames_delta = atomic_long_read(&pkg_stats->tx_frames_delta);
+	long matches_delta = atomic_long_read(&pkg_stats->matches_delta);
+
+	/* restart counting in timer context on user request */
+	if (user_reset)
+		can_init_stats(net);
+
+	/* restart counting on jiffies overflow */
+	if (j < pkg_stats->jiffies_init)
+		can_init_stats(net);
+
+	/* prevent overflow in calc_rate() */
+	if (rx_frames > (LONG_MAX / HZ))
+		can_init_stats(net);
+
+	/* prevent overflow in calc_rate() */
+	if (tx_frames > (LONG_MAX / HZ))
+		can_init_stats(net);
+
+	/* matches overflow - very improbable */
+	if (matches > (LONG_MAX / 100))
+		can_init_stats(net);
+
+	/* calc total values */
+	if (rx_frames)
+		pkg_stats->total_rx_match_ratio = (matches * 100) / rx_frames;
+
+	pkg_stats->total_tx_rate = calc_rate(pkg_stats->jiffies_init, j,
+					    tx_frames);
+	pkg_stats->total_rx_rate = calc_rate(pkg_stats->jiffies_init, j,
+					    rx_frames);
+
+	/* calc current values */
+	if (rx_frames_delta)
+		pkg_stats->current_rx_match_ratio =
+			(matches_delta * 100) /	rx_frames_delta;
+
+	pkg_stats->current_tx_rate = calc_rate(0, HZ, tx_frames_delta);
+	pkg_stats->current_rx_rate = calc_rate(0, HZ, rx_frames_delta);
+
+	/* check / update maximum values */
+	if (pkg_stats->max_tx_rate < pkg_stats->current_tx_rate)
+		pkg_stats->max_tx_rate = pkg_stats->current_tx_rate;
+
+	if (pkg_stats->max_rx_rate < pkg_stats->current_rx_rate)
+		pkg_stats->max_rx_rate = pkg_stats->current_rx_rate;
+
+	if (pkg_stats->max_rx_match_ratio < pkg_stats->current_rx_match_ratio)
+		pkg_stats->max_rx_match_ratio = pkg_stats->current_rx_match_ratio;
+
+	/* clear values for 'current rate' calculation */
+	atomic_long_set(&pkg_stats->tx_frames_delta, 0);
+	atomic_long_set(&pkg_stats->rx_frames_delta, 0);
+	atomic_long_set(&pkg_stats->matches_delta, 0);
+
+	/* restart timer (one second) */
+	mod_timer(&net->can.stattimer, round_jiffies(jiffies + HZ));
+}
+
+/*
+ * proc read functions
+ */
+
+static void can_print_rcvlist(struct seq_file *m, struct hlist_head *rx_list,
+			      struct net_device *dev)
+{
+	struct receiver *r;
+
+	hlist_for_each_entry_rcu(r, rx_list, list) {
+		char *fmt = (r->can_id & CAN_EFF_FLAG)?
+			"   %-5s  %08x  %08x  %pK  %pK  %8ld  %s\n" :
+			"   %-5s     %03x    %08x  %pK  %pK  %8ld  %s\n";
+
+		seq_printf(m, fmt, DNAME(dev), r->can_id, r->mask,
+			   r->func, r->data, atomic_long_read(&r->matches),
+			   r->ident);
+	}
+}
+
+static void can_print_recv_banner(struct seq_file *m)
+{
+	/*
+	 *                  can1.  00000000  00000000  00000000
+	 *                 .......          0  tp20
+	 */
+	if (IS_ENABLED(CONFIG_64BIT))
+		seq_puts(m, "  device   can_id   can_mask      function          userdata       matches  ident\n");
+	else
+		seq_puts(m, "  device   can_id   can_mask  function  userdata   matches  ident\n");
+}
+
+static int can_stats_proc_show(struct seq_file *m, void *v)
+{
+	struct net *net = m->private;
+	struct can_pkg_stats *pkg_stats = net->can.pkg_stats;
+	struct can_rcv_lists_stats *rcv_lists_stats = net->can.rcv_lists_stats;
+
+	seq_putc(m, '\n');
+	seq_printf(m, " %8ld transmitted frames (TXF)\n",
+		   atomic_long_read(&pkg_stats->tx_frames));
+	seq_printf(m, " %8ld received frames (RXF)\n",
+		   atomic_long_read(&pkg_stats->rx_frames));
+	seq_printf(m, " %8ld matched frames (RXMF)\n",
+		   atomic_long_read(&pkg_stats->matches));
+
+	seq_putc(m, '\n');
+
+	if (net->can.stattimer.function == can_stat_update) {
+		seq_printf(m, " %8ld %% total match ratio (RXMR)\n",
+				pkg_stats->total_rx_match_ratio);
+
+		seq_printf(m, " %8ld frames/s total tx rate (TXR)\n",
+				pkg_stats->total_tx_rate);
+		seq_printf(m, " %8ld frames/s total rx rate (RXR)\n",
+				pkg_stats->total_rx_rate);
+
+		seq_putc(m, '\n');
+
+		seq_printf(m, " %8ld %% current match ratio (CRXMR)\n",
+				pkg_stats->current_rx_match_ratio);
+
+		seq_printf(m, " %8ld frames/s current tx rate (CTXR)\n",
+				pkg_stats->current_tx_rate);
+		seq_printf(m, " %8ld frames/s current rx rate (CRXR)\n",
+				pkg_stats->current_rx_rate);
+
+		seq_putc(m, '\n');
+
+		seq_printf(m, " %8ld %% max match ratio (MRXMR)\n",
+				pkg_stats->max_rx_match_ratio);
+
+		seq_printf(m, " %8ld frames/s max tx rate (MTXR)\n",
+				pkg_stats->max_tx_rate);
+		seq_printf(m, " %8ld frames/s max rx rate (MRXR)\n",
+				pkg_stats->max_rx_rate);
+
+		seq_putc(m, '\n');
+	}
+
+	seq_printf(m, " %8ld current receive list entries (CRCV)\n",
+			rcv_lists_stats->rcv_entries);
+	seq_printf(m, " %8ld maximum receive list entries (MRCV)\n",
+			rcv_lists_stats->rcv_entries_max);
+
+	if (rcv_lists_stats->stats_reset)
+		seq_printf(m, "\n %8ld statistic resets (STR)\n",
+				rcv_lists_stats->stats_reset);
+
+	if (rcv_lists_stats->user_reset)
+		seq_printf(m, " %8ld user statistic resets (USTR)\n",
+				rcv_lists_stats->user_reset);
+
+	seq_putc(m, '\n');
+	return 0;
+}
+
+static int can_reset_stats_proc_show(struct seq_file *m, void *v)
+{
+	struct net *net = m->private;
+	struct can_rcv_lists_stats *rcv_lists_stats = net->can.rcv_lists_stats;
+	struct can_pkg_stats *pkg_stats = net->can.pkg_stats;
+
+	user_reset = 1;
+
+	if (net->can.stattimer.function == can_stat_update) {
+		seq_printf(m, "Scheduled statistic reset #%ld.\n",
+				rcv_lists_stats->stats_reset + 1);
+	} else {
+		if (pkg_stats->jiffies_init != jiffies)
+			can_init_stats(net);
+
+		seq_printf(m, "Performed statistic reset #%ld.\n",
+				rcv_lists_stats->stats_reset);
+	}
+	return 0;
+}
+
+static inline void can_rcvlist_proc_show_one(struct seq_file *m, int idx,
+					     struct net_device *dev,
+					     struct can_dev_rcv_lists *dev_rcv_lists)
+{
+	if (!hlist_empty(&dev_rcv_lists->rx[idx])) {
+		can_print_recv_banner(m);
+		can_print_rcvlist(m, &dev_rcv_lists->rx[idx], dev);
+	} else
+		seq_printf(m, "  (%s: no entry)\n", DNAME(dev));
+
+}
+
+static int can_rcvlist_proc_show(struct seq_file *m, void *v)
+{
+	/* double cast to prevent GCC warning */
+	int idx = (int)(long)pde_data(m->file->f_inode);
+	struct net_device *dev;
+	struct can_dev_rcv_lists *dev_rcv_lists;
+	struct net *net = m->private;
+
+	seq_printf(m, "\nreceive list '%s':\n", rx_list_name[idx]);
+
+	rcu_read_lock();
+
+	/* receive list for 'all' CAN devices (dev == NULL) */
+	dev_rcv_lists = net->can.rx_alldev_list;
+	can_rcvlist_proc_show_one(m, idx, NULL, dev_rcv_lists);
+
+	/* receive list for registered CAN devices */
+	for_each_netdev_rcu(net, dev) {
+		struct can_ml_priv *can_ml = can_get_ml_priv(dev);
+
+		if (can_ml)
+			can_rcvlist_proc_show_one(m, idx, dev,
+						  &can_ml->dev_rcv_lists);
+	}
+
+	rcu_read_unlock();
+
+	seq_putc(m, '\n');
+	return 0;
+}
+
+static inline void can_rcvlist_proc_show_array(struct seq_file *m,
+					       struct net_device *dev,
+					       struct hlist_head *rcv_array,
+					       unsigned int rcv_array_sz)
+{
+	unsigned int i;
+	int all_empty = 1;
+
+	/* check whether at least one list is non-empty */
+	for (i = 0; i < rcv_array_sz; i++)
+		if (!hlist_empty(&rcv_array[i])) {
+			all_empty = 0;
+			break;
+		}
+
+	if (!all_empty) {
+		can_print_recv_banner(m);
+		for (i = 0; i < rcv_array_sz; i++) {
+			if (!hlist_empty(&rcv_array[i]))
+				can_print_rcvlist(m, &rcv_array[i], dev);
+		}
+	} else
+		seq_printf(m, "  (%s: no entry)\n", DNAME(dev));
+}
+
+static int can_rcvlist_sff_proc_show(struct seq_file *m, void *v)
+{
+	struct net_device *dev;
+	struct can_dev_rcv_lists *dev_rcv_lists;
+	struct net *net = m->private;
+
+	/* RX_SFF */
+	seq_puts(m, "\nreceive list 'rx_sff':\n");
+
+	rcu_read_lock();
+
+	/* sff receive list for 'all' CAN devices (dev == NULL) */
+	dev_rcv_lists = net->can.rx_alldev_list;
+	can_rcvlist_proc_show_array(m, NULL, dev_rcv_lists->rx_sff,
+				    ARRAY_SIZE(dev_rcv_lists->rx_sff));
+
+	/* sff receive list for registered CAN devices */
+	for_each_netdev_rcu(net, dev) {
+		struct can_ml_priv *can_ml = can_get_ml_priv(dev);
+
+		if (can_ml) {
+			dev_rcv_lists = &can_ml->dev_rcv_lists;
+			can_rcvlist_proc_show_array(m, dev, dev_rcv_lists->rx_sff,
+						    ARRAY_SIZE(dev_rcv_lists->rx_sff));
+		}
+	}
+
+	rcu_read_unlock();
+
+	seq_putc(m, '\n');
+	return 0;
+}
+
+static int can_rcvlist_eff_proc_show(struct seq_file *m, void *v)
+{
+	struct net_device *dev;
+	struct can_dev_rcv_lists *dev_rcv_lists;
+	struct net *net = m->private;
+
+	/* RX_EFF */
+	seq_puts(m, "\nreceive list 'rx_eff':\n");
+
+	rcu_read_lock();
+
+	/* eff receive list for 'all' CAN devices (dev == NULL) */
+	dev_rcv_lists = net->can.rx_alldev_list;
+	can_rcvlist_proc_show_array(m, NULL, dev_rcv_lists->rx_eff,
+				    ARRAY_SIZE(dev_rcv_lists->rx_eff));
+
+	/* eff receive list for registered CAN devices */
+	for_each_netdev_rcu(net, dev) {
+		struct can_ml_priv *can_ml = can_get_ml_priv(dev);
+
+		if (can_ml) {
+			dev_rcv_lists = &can_ml->dev_rcv_lists;
+			can_rcvlist_proc_show_array(m, dev, dev_rcv_lists->rx_eff,
+						    ARRAY_SIZE(dev_rcv_lists->rx_eff));
+		}
+	}
+
+	rcu_read_unlock();
+
+	seq_putc(m, '\n');
+	return 0;
+}
+
+/*
+ * can_init_proc - create main CAN proc directory and procfs entries
+ */
+void can_init_proc(struct net *net)
+{
+	/* create /proc/net/can directory */
+	net->can.proc_dir = proc_net_mkdir(net, "can", net->proc_net);
+
+	if (!net->can.proc_dir) {
+		printk(KERN_INFO "can: failed to create /proc/net/can . "
+			   "CONFIG_PROC_FS missing?\n");
+		return;
+	}
+
+	/* own procfs entries from the AF_CAN core */
+	net->can.pde_stats = proc_create_net_single(CAN_PROC_STATS, 0644,
+			net->can.proc_dir, can_stats_proc_show, NULL);
+	net->can.pde_reset_stats = proc_create_net_single(CAN_PROC_RESET_STATS,
+			0644, net->can.proc_dir, can_reset_stats_proc_show,
+			NULL);
+	net->can.pde_rcvlist_err = proc_create_net_single(CAN_PROC_RCVLIST_ERR,
+			0644, net->can.proc_dir, can_rcvlist_proc_show,
+			(void *)RX_ERR);
+	net->can.pde_rcvlist_all = proc_create_net_single(CAN_PROC_RCVLIST_ALL,
+			0644, net->can.proc_dir, can_rcvlist_proc_show,
+			(void *)RX_ALL);
+	net->can.pde_rcvlist_fil = proc_create_net_single(CAN_PROC_RCVLIST_FIL,
+			0644, net->can.proc_dir, can_rcvlist_proc_show,
+			(void *)RX_FIL);
+	net->can.pde_rcvlist_inv = proc_create_net_single(CAN_PROC_RCVLIST_INV,
+			0644, net->can.proc_dir, can_rcvlist_proc_show,
+			(void *)RX_INV);
+	net->can.pde_rcvlist_eff = proc_create_net_single(CAN_PROC_RCVLIST_EFF,
+			0644, net->can.proc_dir, can_rcvlist_eff_proc_show, NULL);
+	net->can.pde_rcvlist_sff = proc_create_net_single(CAN_PROC_RCVLIST_SFF,
+			0644, net->can.proc_dir, can_rcvlist_sff_proc_show, NULL);
+}
+
+/*
+ * can_remove_proc - remove procfs entries and main CAN proc directory
+ */
+void can_remove_proc(struct net *net)
+{
+	if (!net->can.proc_dir)
+		return;
+
+	if (net->can.pde_stats)
+		remove_proc_entry(CAN_PROC_STATS, net->can.proc_dir);
+
+	if (net->can.pde_reset_stats)
+		remove_proc_entry(CAN_PROC_RESET_STATS, net->can.proc_dir);
+
+	if (net->can.pde_rcvlist_err)
+		remove_proc_entry(CAN_PROC_RCVLIST_ERR, net->can.proc_dir);
+
+	if (net->can.pde_rcvlist_all)
+		remove_proc_entry(CAN_PROC_RCVLIST_ALL, net->can.proc_dir);
+
+	if (net->can.pde_rcvlist_fil)
+		remove_proc_entry(CAN_PROC_RCVLIST_FIL, net->can.proc_dir);
+
+	if (net->can.pde_rcvlist_inv)
+		remove_proc_entry(CAN_PROC_RCVLIST_INV, net->can.proc_dir);
+
+	if (net->can.pde_rcvlist_eff)
+		remove_proc_entry(CAN_PROC_RCVLIST_EFF, net->can.proc_dir);
+
+	if (net->can.pde_rcvlist_sff)
+		remove_proc_entry(CAN_PROC_RCVLIST_SFF, net->can.proc_dir);
+
+	remove_proc_entry("can", net->proc_net);
+}