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, 277 insertions, 0 deletions

diff --git a/sound/ppc/daca.c b/sound/ppc/daca.c
new file mode 100644
index 000000000..f35cb870d
--- /dev/null
+++ b/sound/ppc/daca.c
@@ -0,0 +1,277 @@
+// SPDX-License-Identifier: GPL-2.0-or-later
+/*
+ * PMac DACA lowlevel functions
+ *
+ * Copyright (c) by Takashi Iwai <tiwai@suse.de>
+ */
+
+
+#include <linux/init.h>
+#include <linux/i2c.h>
+#include <linux/kmod.h>
+#include <linux/slab.h>
+#include <sound/core.h>
+#include "pmac.h"
+
+/* i2c address */
+#define DACA_I2C_ADDR	0x4d
+
+/* registers */
+#define DACA_REG_SR	0x01
+#define DACA_REG_AVOL	0x02
+#define DACA_REG_GCFG	0x03
+
+/* maximum volume value */
+#define DACA_VOL_MAX	0x38
+
+
+struct pmac_daca {
+	struct pmac_keywest i2c;
+	int left_vol, right_vol;
+	unsigned int deemphasis : 1;
+	unsigned int amp_on : 1;
+};
+
+
+/*
+ * initialize / detect DACA
+ */
+static int daca_init_client(struct pmac_keywest *i2c)
+{
+	unsigned short wdata = 0x00;
+	/* SR: no swap, 1bit delay, 32-48kHz */
+	/* GCFG: power amp inverted, DAC on */
+	if (i2c_smbus_write_byte_data(i2c->client, DACA_REG_SR, 0x08) < 0 ||
+	    i2c_smbus_write_byte_data(i2c->client, DACA_REG_GCFG, 0x05) < 0)
+		return -EINVAL;
+	return i2c_smbus_write_block_data(i2c->client, DACA_REG_AVOL,
+					  2, (unsigned char*)&wdata);
+}
+
+/*
+ * update volume
+ */
+static int daca_set_volume(struct pmac_daca *mix)
+{
+	unsigned char data[2];
+  
+	if (! mix->i2c.client)
+		return -ENODEV;
+  
+	if (mix->left_vol > DACA_VOL_MAX)
+		data[0] = DACA_VOL_MAX;
+	else
+		data[0] = mix->left_vol;
+	if (mix->right_vol > DACA_VOL_MAX)
+		data[1] = DACA_VOL_MAX;
+	else
+		data[1] = mix->right_vol;
+	data[1] |= mix->deemphasis ? 0x40 : 0;
+	if (i2c_smbus_write_block_data(mix->i2c.client, DACA_REG_AVOL,
+				       2, data) < 0) {
+		dev_err(&mix->i2c.client->dev, "failed to set volume\n");
+		return -EINVAL;
+	}
+	return 0;
+}
+
+
+/* deemphasis switch */
+#define daca_info_deemphasis		snd_ctl_boolean_mono_info
+
+static int daca_get_deemphasis(struct snd_kcontrol *kcontrol,
+			       struct snd_ctl_elem_value *ucontrol)
+{
+	struct snd_pmac *chip = snd_kcontrol_chip(kcontrol);
+	struct pmac_daca *mix;
+	mix = chip->mixer_data;
+	if (!mix)
+		return -ENODEV;
+	ucontrol->value.integer.value[0] = mix->deemphasis ? 1 : 0;
+	return 0;
+}
+
+static int daca_put_deemphasis(struct snd_kcontrol *kcontrol,
+			       struct snd_ctl_elem_value *ucontrol)
+{
+	struct snd_pmac *chip = snd_kcontrol_chip(kcontrol);
+	struct pmac_daca *mix;
+	int change;
+
+	mix = chip->mixer_data;
+	if (!mix)
+		return -ENODEV;
+	change = mix->deemphasis != ucontrol->value.integer.value[0];
+	if (change) {
+		mix->deemphasis = !!ucontrol->value.integer.value[0];
+		daca_set_volume(mix);
+	}
+	return change;
+}
+
+/* output volume */
+static int daca_info_volume(struct snd_kcontrol *kcontrol,
+			    struct snd_ctl_elem_info *uinfo)
+{
+	uinfo->type = SNDRV_CTL_ELEM_TYPE_INTEGER;
+	uinfo->count = 2;
+	uinfo->value.integer.min = 0;
+	uinfo->value.integer.max = DACA_VOL_MAX;
+	return 0;
+}
+
+static int daca_get_volume(struct snd_kcontrol *kcontrol,
+			   struct snd_ctl_elem_value *ucontrol)
+{
+	struct snd_pmac *chip = snd_kcontrol_chip(kcontrol);
+	struct pmac_daca *mix;
+	mix = chip->mixer_data;
+	if (!mix)
+		return -ENODEV;
+	ucontrol->value.integer.value[0] = mix->left_vol;
+	ucontrol->value.integer.value[1] = mix->right_vol;
+	return 0;
+}
+
+static int daca_put_volume(struct snd_kcontrol *kcontrol,
+			   struct snd_ctl_elem_value *ucontrol)
+{
+	struct snd_pmac *chip = snd_kcontrol_chip(kcontrol);
+	struct pmac_daca *mix;
+	unsigned int vol[2];
+	int change;
+
+	mix = chip->mixer_data;
+	if (!mix)
+		return -ENODEV;
+	vol[0] = ucontrol->value.integer.value[0];
+	vol[1] = ucontrol->value.integer.value[1];
+	if (vol[0] > DACA_VOL_MAX || vol[1] > DACA_VOL_MAX)
+		return -EINVAL;
+	change = mix->left_vol != vol[0] ||
+		mix->right_vol != vol[1];
+	if (change) {
+		mix->left_vol = vol[0];
+		mix->right_vol = vol[1];
+		daca_set_volume(mix);
+	}
+	return change;
+}
+
+/* amplifier switch */
+#define daca_info_amp	daca_info_deemphasis
+
+static int daca_get_amp(struct snd_kcontrol *kcontrol,
+			struct snd_ctl_elem_value *ucontrol)
+{
+	struct snd_pmac *chip = snd_kcontrol_chip(kcontrol);
+	struct pmac_daca *mix;
+	mix = chip->mixer_data;
+	if (!mix)
+		return -ENODEV;
+	ucontrol->value.integer.value[0] = mix->amp_on ? 1 : 0;
+	return 0;
+}
+
+static int daca_put_amp(struct snd_kcontrol *kcontrol,
+			struct snd_ctl_elem_value *ucontrol)
+{
+	struct snd_pmac *chip = snd_kcontrol_chip(kcontrol);
+	struct pmac_daca *mix;
+	int change;
+
+	mix = chip->mixer_data;
+	if (!mix)
+		return -ENODEV;
+	change = mix->amp_on != ucontrol->value.integer.value[0];
+	if (change) {
+		mix->amp_on = !!ucontrol->value.integer.value[0];
+		i2c_smbus_write_byte_data(mix->i2c.client, DACA_REG_GCFG,
+					  mix->amp_on ? 0x05 : 0x04);
+	}
+	return change;
+}
+
+static const struct snd_kcontrol_new daca_mixers[] = {
+	{ .iface = SNDRV_CTL_ELEM_IFACE_MIXER,
+	  .name = "Deemphasis Switch",
+	  .info = daca_info_deemphasis,
+	  .get = daca_get_deemphasis,
+	  .put = daca_put_deemphasis
+	},
+	{ .iface = SNDRV_CTL_ELEM_IFACE_MIXER,
+	  .name = "Master Playback Volume",
+	  .info = daca_info_volume,
+	  .get = daca_get_volume,
+	  .put = daca_put_volume
+	},
+	{ .iface = SNDRV_CTL_ELEM_IFACE_MIXER,
+	  .name = "Power Amplifier Switch",
+	  .info = daca_info_amp,
+	  .get = daca_get_amp,
+	  .put = daca_put_amp
+	},
+};
+
+
+#ifdef CONFIG_PM
+static void daca_resume(struct snd_pmac *chip)
+{
+	struct pmac_daca *mix = chip->mixer_data;
+	i2c_smbus_write_byte_data(mix->i2c.client, DACA_REG_SR, 0x08);
+	i2c_smbus_write_byte_data(mix->i2c.client, DACA_REG_GCFG,
+				  mix->amp_on ? 0x05 : 0x04);
+	daca_set_volume(mix);
+}
+#endif /* CONFIG_PM */
+
+
+static void daca_cleanup(struct snd_pmac *chip)
+{
+	struct pmac_daca *mix = chip->mixer_data;
+	if (! mix)
+		return;
+	snd_pmac_keywest_cleanup(&mix->i2c);
+	kfree(mix);
+	chip->mixer_data = NULL;
+}
+
+/* exported */
+int snd_pmac_daca_init(struct snd_pmac *chip)
+{
+	int i, err;
+	struct pmac_daca *mix;
+
+	request_module("i2c-powermac");
+
+	mix = kzalloc_obj(*mix);
+	if (! mix)
+		return -ENOMEM;
+	chip->mixer_data = mix;
+	chip->mixer_free = daca_cleanup;
+	mix->amp_on = 1; /* default on */
+
+	mix->i2c.addr = DACA_I2C_ADDR;
+	mix->i2c.init_client = daca_init_client;
+	mix->i2c.name = "DACA";
+	err = snd_pmac_keywest_init(&mix->i2c);
+	if (err < 0)
+		return err;
+
+	/*
+	 * build mixers
+	 */
+	strscpy(chip->card->mixername, "PowerMac DACA");
+
+	for (i = 0; i < ARRAY_SIZE(daca_mixers); i++) {
+		err = snd_ctl_add(chip->card, snd_ctl_new1(&daca_mixers[i], chip));
+		if (err < 0)
+			return err;
+	}
+
+#ifdef CONFIG_PM
+	chip->resume = daca_resume;
+#endif
+
+	return 0;
+}