6.6.145-rc1 review

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Biren Pandya <[email protected]>

commit 44a5fd874bb6873bdaec59f722c1d57832fbc9df upstream.

kxsd9_write_raw() takes a runtime PM reference with pm_runtime_get_sync() but returns -EINVAL directly when a scale with a non-zero integer part is requested, skipping the matching pm_runtime_put_autosuspend(). This leaks a runtime PM usage-counter reference on every such write, after which the device can no longer autosuspend.

Set the error code and fall through to the existing put instead of returning early.

---
 drivers/iio/accel/kxsd9.c |    5 +++--
 1 file changed, 3 insertions(+), 2 deletions(-)

@@ -146,8 +146,9 @@ static int kxsd9_write_raw(struct iio_de
 	if (mask == IIO_CHAN_INFO_SCALE) {
 		/* Check no integer component */
 		if (val)
-			return -EINVAL;
-		ret = kxsd9_write_scale(indio_dev, val2);
+			ret = -EINVAL;
+		else
+			ret = kxsd9_write_scale(indio_dev, val2);
 	}
 
 	pm_runtime_mark_last_busy(st->dev);

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Maxwell Doose <[email protected]>

commit e561b35633f450ee607e87a6401d97f156a0cd54 upstream.

In the report from Jaeyoung Chung:

"lpc32xx_adc_probe() in drivers/iio/adc/lpc32xx_adc.c registers its interrupt handler with devm_request_irq() before it initializes st->completion with init_completion(). If an interrupt arrives after devm_request_irq() and before init_completion(), the handler calls complete() on an uninitialized completion, causing a kernel panic.

The probe path, in lpc32xx_adc_probe():

    iodev = devm_iio_device_alloc(&pdev->dev, sizeof(*st)); /* st kzalloc-zeroed */
    ...
    retval = devm_request_irq(&pdev->dev, irq, lpc32xx_adc_isr, 0,
                              LPC32XXAD_NAME, st);           /* register handler */
    ...
    init_completion(&st->completion);                       /* initialize completion */

lpc32xx_adc_isr() calls complete():

    complete(&st->completion);

If the device raises an interrupt before init_completion() runs, complete() acquires the uninitialized wait.lock and walks the zeroed task_list in swake_up_locked(). The zeroed task_list makes list_empty() return false, so swake_up_locked() dereferences a NULL list entry, triggering a KASAN wild-memory-access."

Fix the chance of a spurious IRQ causing an uninitialized pointer dereference by moving init_completion() above devm_request_irq().

---
 drivers/iio/adc/lpc32xx_adc.c |    4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

@@ -179,6 +179,8 @@ static int lpc32xx_adc_probe(struct plat
 	if (irq < 0)
 		return irq;
 
+	init_completion(&st->completion);
+
 	retval = devm_request_irq(&pdev->dev, irq, lpc32xx_adc_isr, 0,
 				  LPC32XXAD_NAME, st);
 	if (retval < 0) {

@@ -197,8 +199,6 @@ static int lpc32xx_adc_probe(struct plat
 
 	platform_set_drvdata(pdev, iodev);
 
-	init_completion(&st->completion);
-
 	iodev->name = LPC32XXAD_NAME;
 	iodev->info = &lpc32xx_adc_iio_info;
 	iodev->modes = INDIO_DIRECT_MODE;

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Maxwell Doose <[email protected]>

commit 3ee2128b6f0eb0be7b6cb8f6e0f1f113a65201a0 upstream.

In the report from Jaeyoung Chung:

"spear_adc_probe() in drivers/iio/adc/spear_adc.c registers its interrupt handler with devm_request_irq() before it initializes st->completion with init_completion(). If an interrupt arrives after devm_request_irq() and before init_completion(), the handler calls complete() on an uninitialized completion, causing a kernel panic.

The probe path, in spear_adc_probe():

    iodev = devm_iio_device_alloc(&pdev->dev, sizeof(*st)); /* st kzalloc-zeroed */
    ...
    retval = devm_request_irq(&pdev->dev, irq, spear_adc_isr, 0,
                              LPC32XXAD_NAME, st);           /* register handler */
    ...
    init_completion(&st->completion);                       /* initialize completion */

spear_adc_isr() calls complete():

    complete(&st->completion);

If the device raises an interrupt before init_completion() runs, complete() acquires the uninitialized wait.lock and walks the zeroed task_list in swake_up_locked(). The zeroed task_list makes list_empty() return false, so swake_up_locked() dereferences a NULL list entry, triggering a KASAN wild-memory-access."

Fix the chance of a spurious IRQ causing an uninitialized pointer dereference by moving init_completion() above devm_request_irq().

---
 drivers/iio/adc/spear_adc.c |    3 +--
 1 file changed, 1 insertion(+), 2 deletions(-)

@@ -282,6 +282,7 @@ static int spear_adc_probe(struct platfo
 
 	st = iio_priv(indio_dev);
 
+	init_completion(&st->completion);
 	mutex_init(&st->lock);
 
 	st->np = np;

@@ -346,8 +347,6 @@ static int spear_adc_probe(struct platfo
 
 	platform_set_drvdata(pdev, indio_dev);
 
-	init_completion(&st->completion);
-
 	indio_dev->name = SPEAR_ADC_MOD_NAME;
 	indio_dev->info = &spear_adc_info;
 	indio_dev->modes = INDIO_DIRECT_MODE;

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Pengpeng Hou <[email protected]>

commit 7dc4de2aa6316f1d044cde21f5acfec5f3ec6b47 upstream.

devm_gpiod_get_optional() returns NULL when the optional GPIO is absent, but returns an ERR_PTR when the GPIO provider lookup fails, including probe deferral.

Probe currently logs the ERR_PTR case as if the reset GPIO were simply absent and keeps the error pointer in reset_gpio. Later ads124s_reset() treats any non-NULL reset_gpio as a valid descriptor and passes it to gpiod_set_value_cansleep().

Return the lookup error instead of retaining the ERR_PTR.

---
 drivers/iio/adc/ti-ads124s08.c |    3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

@@ -322,7 +322,8 @@ static int ads124s_probe(struct spi_devi
 	ads124s_priv->reset_gpio = devm_gpiod_get_optional(&spi->dev,
 						   "reset", GPIOD_OUT_LOW);
 	if (IS_ERR(ads124s_priv->reset_gpio))
-		dev_info(&spi->dev, "Reset GPIO not defined\n");
+		return dev_err_probe(&spi->dev, PTR_ERR(ads124s_priv->reset_gpio),
+				     "Failed to get reset GPIO\n");
 
 	ads124s_priv->chip_info = &ads124s_chip_info_tbl[spi_id->driver_data];
 

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Maxwell Doose <[email protected]>

commit 60d877910a43c305b5165131b258a17b1d772d57 upstream.

Include linux/bitfield.h for FIELD_GET().

Create new macros for bit manipulation in combination with manual bit manipulation being replaced with FIELD_GET().

The current variable declaration and initializations are barely readable and use comma separations across multiple lines. Refactor the initializations so that mantissa and exp have separate declarations and sign gets initialized later.

In addition (and due to the nature of the cleanup), fix a sign-extension bug where, float32 would get bitwise anded with ~BIT(31) (which is 0xFFFFFFFF7FFFFFFF) which corrupted the exponent.

---
 drivers/iio/chemical/scd30_core.c |   19 +++++++++++++++----
 1 file changed, 15 insertions(+), 4 deletions(-)

@@ -4,6 +4,8 @@
  *
  * Copyright (c) 2020 Tomasz Duszynski <[email protected]>
  */
+
+#include <linux/bitfield.h>
 #include <linux/bits.h>
 #include <linux/cleanup.h>
 #include <linux/completion.h>

@@ -43,6 +45,11 @@
 #define SCD30_TEMP_OFFSET_MAX 655360
 #define SCD30_EXTRA_TIMEOUT_PER_S 250
 
+/* Floating point arithmetic macros */
+#define SCD30_FLOAT_MANTISSA_MSK GENMASK(22, 0)
+#define SCD30_FLOAT_EXP_MSK GENMASK(30, 23)
+#define SCD30_FLOAT_SIGN_MSK BIT(31)
+
 enum {
 	SCD30_CONC,
 	SCD30_TEMP,

@@ -89,10 +96,14 @@ static int scd30_reset(struct scd30_stat
 /* simplified float to fixed point conversion with a scaling factor of 0.01 */
 static int scd30_float_to_fp(int float32)
 {
-	int fraction, shift,
-	    mantissa = float32 & GENMASK(22, 0),
-	    sign = (float32 & BIT(31)) ? -1 : 1,
-	    exp = (float32 & ~BIT(31)) >> 23;
+	int fraction, shift, sign;
+	int mantissa = FIELD_GET(SCD30_FLOAT_MANTISSA_MSK, float32);
+	int exp = FIELD_GET(SCD30_FLOAT_EXP_MSK, float32);
+
+	if (float32 & SCD30_FLOAT_SIGN_MSK)
+		sign = -1;
+	else
+		sign = 1;
 
 	/* special case 0 */
 	if (!exp && !mantissa)

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Lars-Peter Clausen <[email protected]>

commit af791d295737ea6b6ff2c8d8488462a49c14af01 upstream.

iio_event_getfd() creates the event file descriptor with anon_inode_getfd(), which allocates a new fd, creates the anonymous file and installs it in the process fd table before returning to the caller.

The IIO code resets the event FIFO after anon_inode_getfd() has returned, but before IIO_GET_EVENT_FD_IOCTL has copied the fd number to userspace. But since fd tables are shared between threads, another thread can guess the newly allocated fd number and issue a read() on it as soon as the fd has been installed.

This means the kfifo_to_user() in iio_event_chrdev_read() can run in parallel with the kfifo_reset_out() in iio_event_getfd().

The kfifo documentation says that kfifo_reset_out() is only safe when it is called from the reader thread and there is only one concurrent reader. Otherwise it is dangerous and must be handled in the same way as kfifo_reset().

If that happens, kfifo_to_user() can advance the FIFO out index based on state from before the reset, after the reset has already moved the out index to the current in index. That can leave the FIFO with an out index past the in index. A later read() can then see an underflowed FIFO length and copy more data than the event FIFO buffer contains. This can result in an out-of-bounds read and leak adjacent kernel memory to userspace.

Move the FIFO reset before anon_inode_getfd(). At that point the event fd is marked busy, but the new fd has not been installed yet, so userspace cannot access it while the FIFO is reset.

---
 drivers/iio/industrialio-event.c |    5 ++---
 1 file changed, 2 insertions(+), 3 deletions(-)

@@ -207,6 +207,8 @@ static int iio_event_getfd(struct iio_de
 		goto unlock;
 	}
 
+	kfifo_reset_out(&ev_int->det_events);
+
 	iio_device_get(indio_dev);
 
 	fd = anon_inode_getfd("iio:event", &iio_event_chrdev_fileops,

@@ -214,10 +216,7 @@ static int iio_event_getfd(struct iio_de
 	if (fd < 0) {
 		clear_bit(IIO_BUSY_BIT_POS, &ev_int->flags);
 		iio_device_put(indio_dev);
-	} else {
-		kfifo_reset_out(&ev_int->det_events);
 	}
-
 unlock:
 	mutex_unlock(&iio_dev_opaque->mlock);
 	return fd;

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Stepan Ionichev <[email protected]>

commit 8320c77e67382d5d55d77043a5f60a867d408a2b upstream.

bmg160_get_filter() walks bmg160_samp_freq_table[] looking for the entry matching the bw_bits value read from the chip:

	for (i = 0; i < ARRAY_SIZE(bmg160_samp_freq_table); ++i) {
		if (bmg160_samp_freq_table[i].bw_bits == bw_bits)
			break;
	}
	*val = bmg160_samp_freq_table[i].filter;

If no entry matches, i ends up equal to the array size and the next line reads one slot past the end. bmg160_set_filter() has the same shape, driven by 'val' instead of bw_bits.

smatch flags both:

drivers/iio/gyro/bmg160_core.c:204 bmg160_get_filter() error:
buffer overflow 'bmg160_samp_freq_table' 7 <= 7 drivers/iio/gyro/bmg160_core.c:222 bmg160_set_filter() error:
buffer overflow 'bmg160_samp_freq_table' 7 <= 7

Return -EINVAL when no entry matches.

The set_filter() path is reachable from userspace via the sysfs in_anglvel_filter_low_pass_3db_frequency interface, so userspace can trivially trigger the out-of-bounds read with a value that is not in bmg160_samp_freq_table[].filter.

---
 drivers/iio/gyro/bmg160_core.c |    6 ++++++
 1 file changed, 6 insertions(+)

@@ -204,6 +204,9 @@ static int bmg160_get_filter(struct bmg1
 			break;
 	}
 
+	if (i == ARRAY_SIZE(bmg160_samp_freq_table))
+		return -EINVAL;
+
 	*val = bmg160_samp_freq_table[i].filter;
 
 	return ret ? ret : IIO_VAL_INT;

@@ -221,6 +224,9 @@ static int bmg160_set_filter(struct bmg1
 			break;
 	}
 
+	if (i == ARRAY_SIZE(bmg160_samp_freq_table))
+		return -EINVAL;
+
 	ret = regmap_write(data->regmap, BMG160_REG_PMU_BW,
 			   bmg160_samp_freq_table[i].bw_bits);
 	if (ret < 0) {

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Stepan Ionichev <[email protected]>

commit 088fcb9b567f8723074ad9eb1bf5cb46f8a0096b upstream.

bmg160_chip_init() calls bmg160_set_mode(BMG160_MODE_NORMAL) and then waits only 500-1000 us. Per the BMG160 datasheet (BST-BMG160-DS000-07 Rev. 1.0, May 2013), the start-up and wake-up times (tsu, twusm) are 30 ms.

The same file already waits BMG160_MAX_STARTUP_TIME_MS (80 ms) in bmg160_runtime_resume() after the same set_mode(NORMAL) operation. The 500 us value at probe was likely a unit mix-up; the old comment said "500 ms" while the code used microseconds.

Reuse the same constant via msleep() and add a code comment explaining the datasheet basis for the wait. Without this, register writes that follow the mode change can hit the chip before it is ready.

---
 drivers/iio/gyro/bmg160_core.c |   10 ++++++++--
 1 file changed, 8 insertions(+), 2 deletions(-)

@@ -267,8 +267,14 @@ static int bmg160_chip_init(struct bmg16
 	if (ret < 0)
 		return ret;
 
-	/* Wait upto 500 ms to be ready after changing mode */
-	usleep_range(500, 1000);
+	/*
+	 * Wait for the chip to be ready after switching to normal mode.
+	 * The BMG160 datasheet (BST-BMG160-DS000-07 Rev. 1.0, May 2013)
+	 * specifies a start-up / wake-up time (tsu, twusm) of 30 ms; use
+	 * BMG160_MAX_STARTUP_TIME_MS (80 ms) as a safety margin, matching
+	 * what bmg160_runtime_resume() already does.
+	 */
+	msleep(BMG160_MAX_STARTUP_TIME_MS);
 
 	/* Set Bandwidth */
 	ret = bmg160_set_bw(data, BMG160_DEF_BW);

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Runyu Xiao <[email protected]>

commit cd5a6a5096b246e10600da3ac47a1274ce9573c8 upstream.

bmi160_probe_trigger() registers iio_trigger_generic_data_rdy_poll() through devm_request_irq(), but it passes only irq_type and does not add IRQF_NO_THREAD.

When the kernel is booted with forced IRQ threading, the parent IRQ can otherwise be threaded by the IRQ core and the subsequent IIO trigger child IRQ is dispatched from irq/... thread context instead of hardirq context. Because the handler immediately pushes the event into iio_trigger_poll(), this violates the hardirq-only IIO trigger helper contract and can drive downstream trigger consumers through the wrong execution context.

Add IRQF_NO_THREAD on top of irq_type when registering the BMI160 data- ready trigger handler.

---
 drivers/iio/imu/bmi160/bmi160_core.c |    3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

@@ -784,7 +784,8 @@ int bmi160_probe_trigger(struct iio_dev
 
 	ret = devm_request_irq(&indio_dev->dev, irq,
 			       &iio_trigger_generic_data_rdy_poll,
-			       irq_type, "bmi160", data->trig);
+			       irq_type | IRQF_NO_THREAD,
+			       "bmi160", data->trig);
 	if (ret)
 		return ret;
 

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Jean-Baptiste Maneyrol <[email protected]>

commit 8b0b864c11a2e2ada470f9d5010e1c2bf1eceef2 upstream.

Clock period value is used for computing periods of sampling. There is no need for it to be higher than the maximum odr, otherwise we are losing precision in the computation for nothing.

Switch clock period value to maximum odr period (8kHz).

---
 drivers/iio/imu/inv_icm42600/inv_icm42600_accel.c |    4 ++--
 drivers/iio/imu/inv_icm42600/inv_icm42600_gyro.c  |    4 ++--
 2 files changed, 4 insertions(+), 4 deletions(-)

@@ -718,10 +718,10 @@ struct iio_dev *inv_icm42600_accel_init(
 		return ERR_PTR(-ENOMEM);
 
 	/*
-	 * clock period is 32kHz (31250ns)
+	 * clock period is 8kHz (125000ns)
 	 * jitter is +/- 2% (20 per mille)
 	 */
-	ts_chip.clock_period = 31250;
+	ts_chip.clock_period = 125000;
 	ts_chip.jitter = 20;
 	ts_chip.init_period = inv_icm42600_odr_to_period(st->conf.accel.odr);
 	ts = iio_priv(indio_dev);

@@ -729,10 +729,10 @@ struct iio_dev *inv_icm42600_gyro_init(s
 		return ERR_PTR(-ENOMEM);
 
 	/*
-	 * clock period is 32kHz (31250ns)
+	 * clock period is 8kHz (125000ns)
 	 * jitter is +/- 2% (20 per mille)
 	 */
-	ts_chip.clock_period = 31250;
+	ts_chip.clock_period = 125000;
 	ts_chip.jitter = 20;
 	ts_chip.init_period = inv_icm42600_odr_to_period(st->conf.accel.odr);
 	ts = iio_priv(indio_dev);

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Andreas Kempe <[email protected]>

commit aede83625ff5d9539508582036df30c809d51058 upstream.

As part of driver initialization, e.g. st_lsm6dsx_init_shub() selects the shub register page using st_lsm6dsx_set_page(). Selecting the shub register page shadows the regular register space so whoami, among other registers, is no longer accessible.

In applications where the IMU is permanently powered separately from the processor, there is a window where a reset of the CPU leaves the IMU in the shub register page. Once this occurs, any subsequent probe attempt fails because of the register shadowing.

Using the ism330dlc, the error typically looks like

    st_lsm6dsx_i2c 3-006a: unsupported whoami [10]

with the unknown whoami read from a reserved register in the shub page.

The reset register is also shadowed by the page select, preventing a reset from recovering the chip.

Unconditionally clear the shub page before the whoami readout to ensure normal register access and allow the initialization to proceed.

Place the fix in st_lsm6dsx_check_whoami() before the whoami check because hw->settings, which st_lsm6dsx_set_page() relies on, is first assigned in that function.

Placing the fix in a more logical place than the whoami check would require a bigger restructuring of the code.

---
 drivers/iio/imu/st_lsm6dsx/st_lsm6dsx_core.c |   21 ++++++++++++++++++++-
 1 file changed, 20 insertions(+), 1 deletion(-)

@@ -1576,6 +1576,26 @@ static int st_lsm6dsx_check_whoami(struc
 		return -ENODEV;
 	}
 
+	hw->settings = &st_lsm6dsx_sensor_settings[i];
+
+	if (hw->settings->shub_settings.page_mux.addr) {
+		/*
+		 * If the IMU has the shub page selected on init, for example
+		 * after a CPU watchdog reset while the page is selected, the
+		 * regular register space is shadowed. While the regular
+		 * register space is shadowed, the registers needed for
+		 * initializing the IMU are not available.
+		 *
+		 * Unconditionally clear the shub page selection to ensure
+		 * normal register access.
+		 */
+		err = st_lsm6dsx_set_page(hw, false);
+		if (err < 0) {
+			dev_err(hw->dev, "failed to clear shub page\n");
+			return err;
+		}
+	}
+
 	err = regmap_read(hw->regmap, ST_LSM6DSX_REG_WHOAMI_ADDR, &data);
 	if (err < 0) {
 		dev_err(hw->dev, "failed to read whoami register\n");

@@ -1588,7 +1608,6 @@ static int st_lsm6dsx_check_whoami(struc
 	}
 
 	*name = st_lsm6dsx_sensor_settings[i].id[j].name;
-	hw->settings = &st_lsm6dsx_sensor_settings[i];
 
 	return 0;
 }

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Vidhu Sarwal <[email protected]>

commit aa411adc6ce40ad1a55ebc965f255a4cfc0005f8 upstream.

al3010_scales[] encodes the highest gain range as {0, 1187200}. For IIO_VAL_INT_PLUS_MICRO, the fractional part must be less than 1000000, so the scale 1.1872 should instead be represented as { 1, 187200 }.

Since write_raw() compares the value from userspace against this table, writing the advertised 1.1872 scale never matches the malformed entry and returns -EINVAL. As a result, the highest gain range cannot be selected. Reading the scale in that state also reports the malformed value.

---
 drivers/iio/light/al3010.c |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

@@ -43,7 +43,7 @@ enum al3xxxx_range {
 };
 
 static const int al3010_scales[][2] = {
-	{0, 1187200}, {0, 296800}, {0, 74200}, {0, 18600}
+	{ 1, 187200 }, { 0, 296800 }, { 0, 74200 }, { 0, 18600 },
 };
 
 struct al3010_data {

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Biren Pandya <[email protected]>

commit 38b72267b7e22768a1f26d9935de4e1752a1dc85 upstream.

gp2ap002_read_raw() calls pm_runtime_get_sync() before reading the lux value, but if gp2ap002_get_lux() fails, it returns directly. This skips the pm_runtime_put_autosuspend() call at the "out" label, permanently leaking a runtime PM reference and preventing the device from autosuspending.

Replace the direct return with a "goto out" to ensure the reference is properly dropped on the error path.

---
 drivers/iio/light/gp2ap002.c |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

@@ -258,7 +258,7 @@ static int gp2ap002_read_raw(struct iio_
 		case IIO_LIGHT:
 			ret = gp2ap002_get_lux(gp2ap002);
 			if (ret < 0)
-				return ret;
+				goto out;
 			*val = ret;
 			ret = IIO_VAL_INT;
 			goto out;

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Joshua Crofts <[email protected]>

commit c123ca6ee26ad98f70a866ff428b08145c5a24fe upstream.

Currently in the function opt3001_get_processed(), there is a check that directly returns -ETIMEDOUT if the conversion IRQ times out, completely bypassing the err label, leaving ok_to_ignore_lock permanently true, potentially breaking the device's falling threshold interrupt detection.

Assign -ETIMEDOUT to the return variable and jump to the error label to ensure ok_to_ignore_lock is properly reset.

---
 drivers/iio/light/opt3001.c |    6 ++++--
 1 file changed, 4 insertions(+), 2 deletions(-)

@@ -280,8 +280,10 @@ static int opt3001_get_lux(struct opt300
 		ret = wait_event_timeout(opt->result_ready_queue,
 				opt->result_ready,
 				msecs_to_jiffies(OPT3001_RESULT_READY_LONG));
-		if (ret == 0)
-			return -ETIMEDOUT;
+		if (ret == 0) {
+			ret = -ETIMEDOUT;
+			goto err;
+		}
 	} else {
 		/* Sleep for result ready time */
 		timeout = (opt->int_time == OPT3001_INT_TIME_SHORT) ?

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Stepan Ionichev <[email protected]>

commit a00ffd15674bfaf8b906503c1600e3d8709af56c upstream.

When devm_request_threaded_irq() fails, probe logs the error and then returns -EINVAL, dropping the real error code and breaking the deferred-probe flow for -EPROBE_DEFER.

Return ret directly; the IRQ subsystem already prints on failure.

---
 drivers/iio/light/tsl2591.c |    6 ++----
 1 file changed, 2 insertions(+), 4 deletions(-)

@@ -1139,10 +1139,8 @@ static int tsl2591_probe(struct i2c_clie
 						NULL, tsl2591_event_handler,
 						IRQF_TRIGGER_FALLING | IRQF_ONESHOT,
 						"tsl2591_irq", indio_dev);
-		if (ret) {
-			dev_err_probe(&client->dev, ret, "IRQ request error\n");
-			return -EINVAL;
-		}
+		if (ret)
+			return ret;
 		indio_dev->info = &tsl2591_info;
 	} else {
 		indio_dev->info = &tsl2591_info_no_irq;

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Javier Carrasco <[email protected]>

commit c52bb33b641ebaae3e209f97714cb1758206f7d9 upstream.

The events are registered for IIO_LIGHT and not for IIO_INTENSITY. Use the correct channel type.

When at it, fix minor checkpatch code style warning (alignment).

---
 drivers/iio/light/veml6030.c |    8 +++++---
 1 file changed, 5 insertions(+), 3 deletions(-)

@@ -710,9 +710,11 @@ static irqreturn_t veml6030_event_handle
 	else
 		evtdir = IIO_EV_DIR_FALLING;
 
-	iio_push_event(indio_dev, IIO_UNMOD_EVENT_CODE(IIO_INTENSITY,
-					0, IIO_EV_TYPE_THRESH, evtdir),
-					iio_get_time_ns(indio_dev));
+	iio_push_event(indio_dev, IIO_UNMOD_EVENT_CODE(IIO_LIGHT,
+						       0,
+						       IIO_EV_TYPE_THRESH,
+						       evtdir),
+		       iio_get_time_ns(indio_dev));
 
 	return IRQ_HANDLED;
 }

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Andy Shevchenko <[email protected]>

commit e94944d7364d3ddb273539492f9bd9c9a622549a upstream.

On the failure in the ak8975_read_axis() the PM runtime gets unbalanced. Balance it by calling pm_runtime_put_autosuspend() on error path as well.

---
 drivers/iio/magnetometer/ak8975.c |    1 +
 1 file changed, 1 insertion(+)

@@ -746,6 +746,7 @@ static int ak8975_read_axis(struct iio_d
 
 exit:
 	mutex_unlock(&data->lock);
+	pm_runtime_put_autosuspend(&data->client->dev);
 	dev_err(&client->dev, "Error in reading axis\n");
 	return ret;
 }

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Liviu Stan <[email protected]>

commit 5cb9fdb446bfc3ae0524496f53fb68e67051701b upstream.

reinit_completion() was called after regmap_write() initiated the hardware conversion, creating a race window where the interrupt could fire and call complete() before reinit_completion() reset the completion.

Move reinit_completion() before the regmap_write() to close the race. ltc2983_eeprom_cmd() already does it in the correct order.

---
 drivers/iio/temperature/ltc2983.c |    3 +--
 1 file changed, 1 insertion(+), 2 deletions(-)

@@ -1231,12 +1231,11 @@ static int ltc2983_chan_read(struct ltc2
 	start_conversion |= LTC2983_STATUS_CHAN_SEL(sensor->chan);
 	dev_dbg(&st->spi->dev, "Start conversion on chan:%d, status:%02X\n",
 		sensor->chan, start_conversion);
+	reinit_completion(&st->completion);
 	/* start conversion */
 	ret = regmap_write(st->regmap, LTC2983_STATUS_REG, start_conversion);
 	if (ret)
 		return ret;
-
-	reinit_completion(&st->completion);
 	/*
 	 * wait for conversion to complete.
 	 * 300 ms should be more than enough to complete the conversion.

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Cássio Gabriel <[email protected]>

commit 83fbbcb7935ec6d2c8ba3bc133e8a0ead2ab0b2d upstream.

The VirtIO sound UAPI defines VIRTIO_SND_PCM_RATE_384000, and ALSA has SNDRV_PCM_RATE_384000. However, virtio-snd's rate conversion tables stop at 192 kHz.

A device advertising only 384 kHz is rejected as having no supported PCM frame rates. A device advertising 384 kHz together with lower rates does not expose 384 kHz through the ALSA hardware constraints. The selected ALSA rate also needs a reverse mapping for SET_PARAMS.

Add the missing 384 kHz entries to both conversion tables.

---
 sound/virtio/virtio_pcm.c     |    3 ++-
 sound/virtio/virtio_pcm_ops.c |    3 ++-
 2 files changed, 4 insertions(+), 2 deletions(-)

@@ -77,7 +77,8 @@ static const struct virtsnd_v2a_rate g_v
 	[VIRTIO_SND_PCM_RATE_88200] = { SNDRV_PCM_RATE_88200, 88200 },
 	[VIRTIO_SND_PCM_RATE_96000] = { SNDRV_PCM_RATE_96000, 96000 },
 	[VIRTIO_SND_PCM_RATE_176400] = { SNDRV_PCM_RATE_176400, 176400 },
-	[VIRTIO_SND_PCM_RATE_192000] = { SNDRV_PCM_RATE_192000, 192000 }
+	[VIRTIO_SND_PCM_RATE_192000] = { SNDRV_PCM_RATE_192000, 192000 },
+	[VIRTIO_SND_PCM_RATE_384000] = { SNDRV_PCM_RATE_384000, 384000 }
 };
 
 /**

@@ -90,7 +90,8 @@ static const struct virtsnd_a2v_rate g_a
 	{ 88200, VIRTIO_SND_PCM_RATE_88200 },
 	{ 96000, VIRTIO_SND_PCM_RATE_96000 },
 	{ 176400, VIRTIO_SND_PCM_RATE_176400 },
-	{ 192000, VIRTIO_SND_PCM_RATE_192000 }
+	{ 192000, VIRTIO_SND_PCM_RATE_192000 },
+	{ 384000, VIRTIO_SND_PCM_RATE_384000 }
 };
 
 static int virtsnd_pcm_sync_stop(struct snd_pcm_substream *substream);

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Zhao Dongdong <[email protected]>

commit e64d170346d00b580c0043de3e5ccb3e331c47d4 upstream.

snd_ctl_new1() can return NULL when memory allocation fails. snd_ymfpci_create_spdif_controls() does not check the return value before dereferencing kctl->id.device, which can lead to a NULL pointer dereference.

Add NULL checks after snd_ctl_new1() calls and return -ENOMEM if any fails.

---
 sound/pci/ymfpci/ymfpci_main.c |    6 ++++++
 1 file changed, 6 insertions(+)

@@ -1822,16 +1822,22 @@ int snd_ymfpci_mixer(struct snd_ymfpci *
 	if (snd_BUG_ON(!chip->pcm_spdif))
 		return -ENXIO;
 	kctl = snd_ctl_new1(&snd_ymfpci_spdif_default, chip);
+	if (!kctl)
+		return -ENOMEM;
 	kctl->id.device = chip->pcm_spdif->device;
 	err = snd_ctl_add(chip->card, kctl);
 	if (err < 0)
 		return err;
 	kctl = snd_ctl_new1(&snd_ymfpci_spdif_mask, chip);
+	if (!kctl)
+		return -ENOMEM;
 	kctl->id.device = chip->pcm_spdif->device;
 	err = snd_ctl_add(chip->card, kctl);
 	if (err < 0)
 		return err;
 	kctl = snd_ctl_new1(&snd_ymfpci_spdif_stream, chip);
+	if (!kctl)
+		return -ENOMEM;
 	kctl->id.device = chip->pcm_spdif->device;
 	err = snd_ctl_add(chip->card, kctl);
 	if (err < 0)

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Maoyi Xie <[email protected]>

commit f7f3f9fd81e7adbaa12c2e62ee07f0e094a543fd upstream.

snd_usb_caiaq_tks4_dispatch() decodes the Traktor Kontrol S4 input stream in fixed 16-byte (TKS4_MSGBLOCK_SIZE) message blocks. On every iteration it advances buf and subtracts the block size while looping on "while (len)".

len is urb->actual_length. That value is supplied by the device and is not guaranteed to be a multiple of 16. When a final short block leaves len between 1 and 15, the loop runs once more, reads up to buf[15], and then does "len -= TKS4_MSGBLOCK_SIZE". As len is unsigned this underflows to a huge value. The loop then keeps iterating and walking buf far past the end of the 512-byte ep4_in_buf, reading out of bounds until a bogus block id happens to be hit.

Iterate only while a full message block is available. This stops the unsigned underflow and silently drops any trailing partial block, which carries no complete control value anyway.

The sibling endpoint-4 parsers are not affected. The Traktor Kontrol X1 and Maschine arms in snd_usb_caiaq_ep4_reply_dispatch() floor urb->actual_length before dispatching.

---
 sound/usb/caiaq/input.c |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

@@ -330,7 +330,7 @@ static void snd_usb_caiaq_tks4_dispatch(
 {
 	struct device *dev = caiaqdev_to_dev(cdev);
 
-	while (len) {
+	while (len >= TKS4_MSGBLOCK_SIZE) {
 		unsigned int i, block_id = (buf[0] << 8) | buf[1];
 
 		switch (block_id) {

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Zhao Dongdong <[email protected]>

commit c205bd1b28fb7e5f1061a4e78813fad7d315cb3e upstream.

snd_ctl_new1() can return NULL when memory allocation fails. snd_cmipci_spdif_controls() does not check the return value before dereferencing kctl->id.device, which can lead to a NULL pointer dereference.

Add NULL checks after snd_ctl_new1() calls and return -ENOMEM if any fails.

---
 sound/pci/cmipci.c |    6 ++++++
 1 file changed, 6 insertions(+)

@@ -2688,16 +2688,22 @@ static int snd_cmipci_mixer_new(struct c
 		}
 		if (cm->can_ac3_hw) {
 			kctl = snd_ctl_new1(&snd_cmipci_spdif_default, cm);
+			if (!kctl)
+				return -ENOMEM;
 			kctl->id.device = pcm_spdif_device;
 			err = snd_ctl_add(card, kctl);
 			if (err < 0)
 				return err;
 			kctl = snd_ctl_new1(&snd_cmipci_spdif_mask, cm);
+			if (!kctl)
+				return -ENOMEM;
 			kctl->id.device = pcm_spdif_device;
 			err = snd_ctl_add(card, kctl);
 			if (err < 0)
 				return err;
 			kctl = snd_ctl_new1(&snd_cmipci_spdif_stream, cm);
+			if (!kctl)
+				return -ENOMEM;
 			kctl->id.device = pcm_spdif_device;
 			err = snd_ctl_add(card, kctl);
 			if (err < 0)

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Zhao Dongdong <[email protected]>

commit 1edd1f02dddd20aeb6066ded41017615766ea42f upstream.

snd_ctl_new1() can return NULL when memory allocation fails. snd_es1938_mixer() does not check the return value before dereferencing the pointer, which can lead to a NULL pointer dereference.

Add a NULL check after snd_ctl_new1() and return -ENOMEM if it fails.

---
 sound/pci/es1938.c |    2 ++
 1 file changed, 2 insertions(+)

@@ -1670,6 +1670,8 @@ static int snd_es1938_mixer(struct es193
 	for (idx = 0; idx < ARRAY_SIZE(snd_es1938_controls); idx++) {
 		struct snd_kcontrol *kctl;
 		kctl = snd_ctl_new1(&snd_es1938_controls[idx], chip);
+		if (!kctl)
+			return -ENOMEM;
 		switch (idx) {
 			case 0:
 				chip->master_volume = kctl;

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Maoyi Xie <[email protected]>

commit 29b9667982e4df2ed7744f86b1144f8bb58eb698 upstream.

isight_packet() takes the frame count from the device iso packet and checks it only against the device claimed iso length.

	count = be32_to_cpu(payload->sample_count);
	if (likely(count <= (length - 16) / 4))
		isight_samples(isight, payload->samples, count);

length is the iso header data_length. It can be up to 0xffff. So the gate allows a count up to about 16379. isight_samples() then copies count frames out of payload->samples into the PCM DMA buffer.

payload->samples holds only 2 * MAX_FRAMES_PER_PACKET values. The device multiplexes two samples per frame. A count past MAX_FRAMES_PER_PACKET reads past the payload. A count past the buffer size writes past runtime->dma_area. The smallest PCM buffer is larger than MAX_FRAMES_PER_PACKET. Bounding the count to MAX_FRAMES_PER_PACKET keeps both the read and the write in range.

A malicious or faulty Apple iSight on the FireWire bus reaches this during a normal capture.

Add the MAX_FRAMES_PER_PACKET bound to the gate.

---
 sound/firewire/isight.c |    3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

@@ -179,7 +179,8 @@ static void isight_packet(struct fw_iso_
 	if (likely(length >= 16 &&
 		   payload->signature == cpu_to_be32(0x73676874/*"sght"*/))) {
 		count = be32_to_cpu(payload->sample_count);
-		if (likely(count <= (length - 16) / 4)) {
+		if (likely(count <= (length - 16) / 4 &&
+			   count <= MAX_FRAMES_PER_PACKET)) {
 			total = be32_to_cpu(payload->sample_total);
 			if (unlikely(total != isight->total_samples)) {
 				if (!isight->first_packet)

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Zhao Dongdong <[email protected]>

commit c7fa99d30c7a166a5e5db5a585ce7501ff68326b upstream.

snd_ctl_new1() can return NULL when memory allocation fails. snd_gf1_pcm_volume_control() does not check the return value before dereferencing kctl->id.index, which can lead to a NULL pointer dereference.

Add a NULL check after snd_ctl_new1() and return -ENOMEM if it fails.

---
 sound/isa/gus/gus_pcm.c |    2 ++
 1 file changed, 2 insertions(+)

@@ -875,6 +875,8 @@ int snd_gf1_pcm_new(struct snd_gus_card
 		kctl = snd_ctl_new1(&snd_gf1_pcm_volume_control1, gus);
 	else
 		kctl = snd_ctl_new1(&snd_gf1_pcm_volume_control, gus);
+	if (!kctl)
+		return -ENOMEM;
 	kctl->id.index = control_index;
 	err = snd_ctl_add(card, kctl);
 	if (err < 0)

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Zhao Dongdong <[email protected]>

commit 2b929b91b0f3bc6de8a844370049cd99ee8e31ff upstream.

snd_ctl_new1() can return NULL when memory allocation fails. The ice1712 driver calls snd_ctl_new1() without checking the return value before dereferencing the pointer in multiple places (ice1712.c, ice1724.c, aureon.c), which can lead to NULL pointer dereferences.

Add NULL checks after snd_ctl_new1() calls and return -ENOMEM if any fails.

---
 sound/pci/ice1712/aureon.c  |    2 ++
 sound/pci/ice1712/ice1712.c |    8 ++++++++
 sound/pci/ice1712/ice1724.c |    6 ++++++
 3 files changed, 16 insertions(+)

@@ -1900,6 +1900,8 @@ static int aureon_add_controls(struct sn
 			for (i = 0; i < ARRAY_SIZE(cs8415_controls); i++) {
 				struct snd_kcontrol *kctl;
 				kctl = snd_ctl_new1(&cs8415_controls[i], ice);
+				if (!kctl)
+					return -ENOMEM;
 				if (i > 1)
 					kctl->id.device = ice->pcm->device;
 				err = snd_ctl_add(ice->card, kctl);

@@ -2372,21 +2372,29 @@ int snd_ice1712_spdif_build_controls(str
 	if (snd_BUG_ON(!ice->pcm_pro))
 		return -EIO;
 	kctl = snd_ctl_new1(&snd_ice1712_spdif_default, ice);
+	if (!kctl)
+		return -ENOMEM;
 	kctl->id.device = ice->pcm_pro->device;
 	err = snd_ctl_add(ice->card, kctl);
 	if (err < 0)
 		return err;
 	kctl = snd_ctl_new1(&snd_ice1712_spdif_maskc, ice);
+	if (!kctl)
+		return -ENOMEM;
 	kctl->id.device = ice->pcm_pro->device;
 	err = snd_ctl_add(ice->card, kctl);
 	if (err < 0)
 		return err;
 	kctl = snd_ctl_new1(&snd_ice1712_spdif_maskp, ice);
+	if (!kctl)
+		return -ENOMEM;
 	kctl->id.device = ice->pcm_pro->device;
 	err = snd_ctl_add(ice->card, kctl);
 	if (err < 0)
 		return err;
 	kctl = snd_ctl_new1(&snd_ice1712_spdif_stream, ice);
+	if (!kctl)
+		return -ENOMEM;
 	kctl->id.device = ice->pcm_pro->device;
 	err = snd_ctl_add(ice->card, kctl);
 	if (err < 0)

@@ -2393,16 +2393,22 @@ static int snd_vt1724_spdif_build_contro
 		return err;
 
 	kctl = snd_ctl_new1(&snd_vt1724_spdif_default, ice);
+	if (!kctl)
+		return -ENOMEM;
 	kctl->id.device = ice->pcm->device;
 	err = snd_ctl_add(ice->card, kctl);
 	if (err < 0)
 		return err;
 	kctl = snd_ctl_new1(&snd_vt1724_spdif_maskc, ice);
+	if (!kctl)
+		return -ENOMEM;
 	kctl->id.device = ice->pcm->device;
 	err = snd_ctl_add(ice->card, kctl);
 	if (err < 0)
 		return err;
 	kctl = snd_ctl_new1(&snd_vt1724_spdif_maskp, ice);
+	if (!kctl)
+		return -ENOMEM;
 	kctl->id.device = ice->pcm->device;
 	err = snd_ctl_add(ice->card, kctl);
 	if (err < 0)

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: HyeongJun An <[email protected]>

commit 435990e25bf1f4af3e6df12a6fbfd1f7ba4a97d4 upstream.

snd_seq_event_dup() copies an incoming event into a pool cell and, in the UMP-enabled build, clears the trailing cell->ump.raw.extra word that the memcpy() did not cover. The guard deciding whether to clear it compares the copied size against sizeof(cell->event):

	memcpy(&cell->ump, event, size);
	if (size < sizeof(cell->event))
		cell->ump.raw.extra = 0;

For a legacy (non-UMP) event, size == sizeof(struct snd_seq_event) == sizeof(cell->event), so the condition is false and the extra word keeps stale data. The cell pool is allocated with kvmalloc() (not zeroed) and cells are reused via a free list, so that word holds uninitialised heap or leftover event data.

When such a cell is delivered to a UMP client (client->midi_version > 0) that set SNDRV_SEQ_FILTER_NO_CONVERT -- so the legacy event reaches it unconverted -- snd_seq_read() reads it out as the larger struct snd_seq_ump_event and copies the stale word to user space, a 4-byte kernel heap infoleak to an unprivileged /dev/snd/seq client.

Compare against sizeof(cell->ump) instead, so the trailing word is zeroed for every event shorter than the UMP cell.

---
 sound/core/seq/seq_memory.c |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

@@ -366,7 +366,7 @@ int snd_seq_event_dup(struct snd_seq_poo
 	size = snd_seq_event_packet_size(event);
 	memcpy(&cell->ump, event, size);
 #if IS_ENABLED(CONFIG_SND_SEQ_UMP)
-	if (size < sizeof(cell->event))
+	if (size < sizeof(cell->ump))
 		cell->ump.raw.extra = 0;
 #endif
 

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Darvell Long <[email protected]>

commit 7693c0cc415f3a16a7a3355f245474a5e661be4e upstream.

The DualSense jack-detection input handler verifies that a matching input device belongs to the same physical controller by building kobject path strings for both the input device and the USB audio device, then comparing the path prefix.

This was observed when a weak physical connection caused the controller to rapidly disconnect and reconnect. During that repeated hotplug, snd_dualsense_ih_match() can run while the controller's USB device is being disconnected. kobject_get_path() walks ancestor kobjects and dereferences their names; if the USB device kobject name is no longer valid, this can fault in strlen():

  RIP: 0010:strlen+0x10/0x30
  Call Trace:
   kobject_get_path+0x34/0x150
   snd_dualsense_ih_match+0x49/0xd0 [snd_usb_audio]
   input_register_device+0x566/0x6a0
   ps_probe+0xb89/0x1590 [hid_playstation]

The same ownership check can be done without building kobject path strings. The input device is parented below the HID device, USB interface and USB device, so walking the input device parent chain and comparing against the mixer USB device preserves the check without dereferencing kobject names during disconnect.

---
 sound/usb/mixer_quirks.c |   40 ++++++++++++----------------------------
 1 file changed, 12 insertions(+), 28 deletions(-)

@@ -573,46 +573,30 @@ static bool snd_dualsense_ih_match(struc
 {
 	struct dualsense_mixer_elem_info *mei;
 	struct usb_device *snd_dev;
-	char *input_dev_path, *usb_dev_path;
-	size_t usb_dev_path_len;
-	bool match = false;
+	struct device *parent;
 
 	mei = container_of(handler, struct dualsense_mixer_elem_info, ih);
 	snd_dev = mei->info.head.mixer->chip->dev;
 
-	input_dev_path = kobject_get_path(&dev->dev.kobj, GFP_KERNEL);
-	if (!input_dev_path) {
-		dev_warn(&snd_dev->dev, "Failed to get input dev path\n");
-		return false;
-	}
-
-	usb_dev_path = kobject_get_path(&snd_dev->dev.kobj, GFP_KERNEL);
-	if (!usb_dev_path) {
-		dev_warn(&snd_dev->dev, "Failed to get USB dev path\n");
-		goto free_paths;
-	}
-
 	/*
 	 * Ensure the VID:PID matched input device supposedly owned by the
 	 * hid-playstation driver belongs to the actual hardware handled by
-	 * the current USB audio device, which implies input_dev_path being
-	 * a subpath of usb_dev_path.
+	 * the current USB audio device.
 	 *
 	 * This verification is necessary when there is more than one identical
 	 * controller attached to the host system.
+	 *
+	 * The input device is registered below the HID device, USB interface and
+	 * USB device, so compare the parent chain directly instead of building
+	 * kobject path strings. This avoids dereferencing kobject names while the
+	 * USB device hierarchy is being torn down during disconnect.
 	 */
-	usb_dev_path_len = strlen(usb_dev_path);
-	if (usb_dev_path_len >= strlen(input_dev_path))
-		goto free_paths;
-
-	usb_dev_path[usb_dev_path_len] = '/';
-	match = !memcmp(input_dev_path, usb_dev_path, usb_dev_path_len + 1);
-
-free_paths:
-	kfree(input_dev_path);
-	kfree(usb_dev_path);
+	for (parent = dev->dev.parent; parent; parent = parent->parent) {
+		if (parent == &snd_dev->dev)
+			return true;
+	}
 
-	return match;
+	return false;
 }
 
 static int snd_dualsense_ih_connect(struct input_handler *handler,

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Cássio Gabriel <[email protected]>

commit 0f25cf1f02e3dba626791d949c759a48c0a44996 upstream.

scarlett_ctl_enum_put() ignores the return value from snd_usb_set_cur_mix_value() and reports success whenever the requested enum value differs from the current one.

If the SET_CUR request fails, the callback still returns success even though neither the hardware state nor the cached mixer value changed.

---
 sound/usb/mixer_scarlett.c |    4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)

@@ -438,7 +438,9 @@ static int scarlett_ctl_enum_put(struct
 	val = ucontrol->value.integer.value[0];
 	val = val + opt->start;
 	if (val != oval) {
-		snd_usb_set_cur_mix_value(elem, 0, 0, val);
+		err = snd_usb_set_cur_mix_value(elem, 0, 0, val);
+		if (err < 0)
+			return err;
 		return 1;
 	}
 	return 0;

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Cássio Gabriel <[email protected]>

commit 3c06aec8abda6ba068b58a8b7119cdb2a48456b1 upstream.

Several US-16x08 mixer put callbacks log failed control URBs but still return success to userspace. That hides device write failures even though the requested value was not applied.

Return the negative write error instead in the route, master, bus, channel, and EQ switch put callbacks.

---
 sound/usb/mixer_us16x08.c |   49 ++++++++++++++++++++++------------------------
 1 file changed, 24 insertions(+), 25 deletions(-)

@@ -225,14 +225,14 @@ static int snd_us16x08_route_put(struct
 
 	err = snd_us16x08_send_urb(chip, buf, sizeof(route_msg));
 
-	if (err > 0) {
-		elem->cached |= 1 << index;
-		elem->cache_val[index] = val;
-	} else {
+	if (err < 0) {
 		usb_audio_dbg(chip, "Failed to set routing, err:%d\n", err);
+		return err;
 	}
 
-	return err > 0 ? 1 : 0;
+	elem->cached |= 1 << index;
+	elem->cache_val[index] = val;
+	return 1;
 }
 
 static int snd_us16x08_master_info(struct snd_kcontrol *kcontrol,

@@ -284,14 +284,14 @@ static int snd_us16x08_master_put(struct
 	buf[5] = index + 1;
 	err = snd_us16x08_send_urb(chip, buf, sizeof(mix_msg_out));
 
-	if (err > 0) {
-		elem->cached |= 1 << index;
-		elem->cache_val[index] = val;
-	} else {
+	if (err < 0) {
 		usb_audio_dbg(chip, "Failed to set master, err:%d\n", err);
+		return err;
 	}
 
-	return err > 0 ? 1 : 0;
+	elem->cached |= 1 << index;
+	elem->cache_val[index] = val;
+	return 1;
 }
 
 static int snd_us16x08_bus_put(struct snd_kcontrol *kcontrol,

@@ -325,14 +325,14 @@ static int snd_us16x08_bus_put(struct sn
 		break;
 	}
 
-	if (err > 0) {
-		elem->cached |= 1;
-		elem->cache_val[0] = val;
-	} else {
+	if (err < 0) {
 		usb_audio_dbg(chip, "Failed to set bus parameter, err:%d\n", err);
+		return err;
 	}
 
-	return err > 0 ? 1 : 0;
+	elem->cached |= 1;
+	elem->cache_val[0] = val;
+	return 1;
 }
 
 static int snd_us16x08_bus_get(struct snd_kcontrol *kcontrol,

@@ -393,14 +393,14 @@ static int snd_us16x08_channel_put(struc
 
 	err = snd_us16x08_send_urb(chip, buf, sizeof(mix_msg_in));
 
-	if (err > 0) {
-		elem->cached |= 1 << index;
-		elem->cache_val[index] = val;
-	} else {
+	if (err < 0) {
 		usb_audio_dbg(chip, "Failed to set channel, err:%d\n", err);
+		return err;
 	}
 
-	return err > 0 ? 1 : 0;
+	elem->cached |= 1 << index;
+	elem->cache_val[index] = val;
+	return 1;
 }
 
 static int snd_us16x08_mix_info(struct snd_kcontrol *kcontrol,

@@ -530,13 +530,13 @@ static int snd_us16x08_eqswitch_put(stru
 		msleep(15);
 	}
 
-	if (err > 0) {
-		elem->cached |= 1 << index;
-		elem->cache_val[index] = val;
-	} else {
+	if (err < 0) {
 		usb_audio_dbg(chip, "Failed to set eq switch, err:%d\n", err);
+		return err;
 	}
 
+	elem->cached |= 1 << index;
+	elem->cache_val[index] = val;
 	return 1;
 }
 

@@ -1419,4 +1419,3 @@ int snd_us16x08_controls_create(struct u
 
 	return 0;
 }
-

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Cássio Gabriel <[email protected]>

commit 6380957fa24251856a532e48a46a4dc3d1ae26b6 upstream.

Several mixer quirk callbacks cache the requested
control value in kcontrol->private_value before
issuing a single vendor or class write.

Their paired get and resume paths consume that cache directly, so a failed write currently leaves software state changed even though the update did not succeed. That can make later reads report a value the device never accepted and can replay the stale cache on resume.

Restore the previous cached value on failure in
the Audigy2NX LED, Emu0204 channel switch,
Xonar U1 output switch, Native Instruments controls,
FTU effect program switch, and Sound Blaster E1 input source switch.

---
 sound/usb/mixer_quirks.c |   45 +++++++++++++++++++++++++++++++++++++--------
 1 file changed, 37 insertions(+), 8 deletions(-)

@@ -335,6 +335,7 @@ static int snd_audigy2nx_led_put(struct
 	int index = kcontrol->private_value & 0xff;
 	unsigned int value = ucontrol->value.integer.value[0];
 	int old_value = kcontrol->private_value >> 8;
+	unsigned long old_pval = kcontrol->private_value;
 	int err;
 
 	if (value > 1)

@@ -343,7 +344,11 @@ static int snd_audigy2nx_led_put(struct
 		return 0;
 	kcontrol->private_value = (value << 8) | index;
 	err = snd_audigy2nx_led_update(mixer, value, index);
-	return err < 0 ? err : 1;
+	if (err < 0) {
+		kcontrol->private_value = old_pval;
+		return err;
+	}
+	return 1;
 }
 
 static int snd_audigy2nx_led_resume(struct usb_mixer_elem_list *list)

@@ -493,6 +498,7 @@ static int snd_emu0204_ch_switch_put(str
 	struct usb_mixer_elem_list *list = snd_kcontrol_chip(kcontrol);
 	struct usb_mixer_interface *mixer = list->mixer;
 	unsigned int value = ucontrol->value.enumerated.item[0];
+	unsigned long old_pval = kcontrol->private_value;
 	int err;
 
 	if (value > 1)

@@ -503,7 +509,11 @@ static int snd_emu0204_ch_switch_put(str
 
 	kcontrol->private_value = value;
 	err = snd_emu0204_ch_switch_update(mixer, value);
-	return err < 0 ? err : 1;
+	if (err < 0) {
+		kcontrol->private_value = old_pval;
+		return err;
+	}
+	return 1;
 }
 
 static int snd_emu0204_ch_switch_resume(struct usb_mixer_elem_list *list)

@@ -814,7 +824,11 @@ static int snd_xonar_u1_switch_put(struc
 
 	kcontrol->private_value = new_status;
 	err = snd_xonar_u1_switch_update(list->mixer, new_status);
-	return err < 0 ? err : 1;
+	if (err < 0) {
+		kcontrol->private_value = old_status;
+		return err;
+	}
+	return 1;
 }
 
 static int snd_xonar_u1_switch_resume(struct usb_mixer_elem_list *list)

@@ -1163,7 +1177,8 @@ static int snd_nativeinstruments_control
 					     struct snd_ctl_elem_value *ucontrol)
 {
 	struct usb_mixer_elem_list *list = snd_kcontrol_chip(kcontrol);
-	u8 oldval = (kcontrol->private_value >> 24) & 0xff;
+	unsigned long old_pval = kcontrol->private_value;
+	u8 oldval = (old_pval >> 24) & 0xff;
 	u8 newval = ucontrol->value.integer.value[0];
 	int err;
 

@@ -1173,7 +1188,11 @@ static int snd_nativeinstruments_control
 	kcontrol->private_value &= ~(0xff << 24);
 	kcontrol->private_value |= (unsigned int)newval << 24;
 	err = snd_ni_update_cur_val(list);
-	return err < 0 ? err : 1;
+	if (err < 0) {
+		kcontrol->private_value = old_pval;
+		return err;
+	}
+	return 1;
 }
 
 static const struct snd_kcontrol_new snd_nativeinstruments_ta6_mixers[] = {

@@ -1331,7 +1350,8 @@ static int snd_ftu_eff_switch_put(struct
 					struct snd_ctl_elem_value *ucontrol)
 {
 	struct usb_mixer_elem_list *list = snd_kcontrol_chip(kctl);
-	unsigned int pval = list->kctl->private_value;
+	unsigned long old_pval = list->kctl->private_value;
+	unsigned int pval = old_pval;
 	int cur_val, err, new_val;
 
 	cur_val = pval >> 24;

@@ -1342,7 +1362,11 @@ static int snd_ftu_eff_switch_put(struct
 	kctl->private_value &= ~(0xff << 24);
 	kctl->private_value |= new_val << 24;
 	err = snd_ftu_eff_switch_update(list);
-	return err < 0 ? err : 1;
+	if (err < 0) {
+		kctl->private_value = old_pval;
+		return err;
+	}
+	return 1;
 }
 
 static int snd_ftu_create_effect_switch(struct usb_mixer_interface *mixer,

@@ -2140,13 +2164,18 @@ static int snd_soundblaster_e1_switch_pu
 {
 	struct usb_mixer_elem_list *list = snd_kcontrol_chip(kcontrol);
 	unsigned char value = !!ucontrol->value.integer.value[0];
+	unsigned long old_pval = kcontrol->private_value;
 	int err;
 
 	if (kcontrol->private_value == value)
 		return 0;
 	kcontrol->private_value = value;
 	err = snd_soundblaster_e1_switch_update(list->mixer, value);
-	return err < 0 ? err : 1;
+	if (err < 0) {
+		kcontrol->private_value = old_pval;
+		return err;
+	}
+	return 1;
 }
 
 static int snd_soundblaster_e1_switch_resume(struct usb_mixer_elem_list *list)

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Cássio Gabriel <[email protected]>

commit d8f802ccf1fdbeb89d62748d6a0d0fbd442c8127 upstream.

snd_bbfpro_ctl_put() and snd_bbfpro_vol_put() cache the requested packed control state in kcontrol->private_value before issuing the USB write.

Their get and resume paths use that cached value directly, so a failed write can leave the driver reporting and later replaying a setting the hardware never accepted.

Update the cached state only after a successful USB write.

---
 sound/usb/mixer_quirks.c |   16 ++++++++++------
 1 file changed, 10 insertions(+), 6 deletions(-)

@@ -2950,12 +2950,14 @@ static int snd_bbfpro_ctl_put(struct snd
 	if (val == old_value)
 		return 0;
 
+	err = snd_bbfpro_ctl_update(mixer, reg, idx, val);
+	if (err < 0)
+		return err;
+
 	kcontrol->private_value = reg
 		| ((idx & SND_BBFPRO_CTL_IDX_MASK) << SND_BBFPRO_CTL_IDX_SHIFT)
 		| ((val & SND_BBFPRO_CTL_VAL_MASK) << SND_BBFPRO_CTL_VAL_SHIFT);
-
-	err = snd_bbfpro_ctl_update(mixer, reg, idx, val);
-	return err < 0 ? err : 1;
+	return 1;
 }
 
 static int snd_bbfpro_ctl_resume(struct usb_mixer_elem_list *list)

@@ -3040,11 +3042,13 @@ static int snd_bbfpro_vol_put(struct snd
 
 	new_val = uvalue & SND_BBFPRO_MIXER_VAL_MASK;
 
+	err = snd_bbfpro_vol_update(mixer, idx, new_val);
+	if (err < 0)
+		return err;
+
 	kcontrol->private_value = idx
 		| (new_val << SND_BBFPRO_MIXER_VAL_SHIFT);
-
-	err = snd_bbfpro_vol_update(mixer, idx, new_val);
-	return err < 0 ? err : 1;
+	return 1;
 }
 
 static int snd_bbfpro_vol_resume(struct usb_mixer_elem_list *list)

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Cássio Gabriel <[email protected]>

commit a440c17869ecd71da0f295b62868fc742d09a8ba upstream.

snd_us16x08_comp_put() and snd_us16x08_eq_put() update their software stores before sending the USB write. If the transfer fails, later get callbacks report a value the hardware never accepted.

Build the outgoing message from the current store plus the pending value, then commit the store only after a successful write.

---
 sound/usb/mixer_us16x08.c |   78 ++++++++++++++++++++++++++++++----------------
 1 file changed, 52 insertions(+), 26 deletions(-)

@@ -436,6 +436,7 @@ static int snd_us16x08_comp_put(struct s
 	int index = ucontrol->id.index;
 	char buf[sizeof(comp_msg)];
 	int val_idx, val;
+	int threshold, ratio, attack, release, gain, switch_on;
 	int err;
 
 	val = ucontrol->value.integer.value[0];

@@ -448,36 +449,61 @@ static int snd_us16x08_comp_put(struct s
 	/* new control value incl. bias*/
 	val_idx = elem->head.id - SND_US16X08_ID_COMP_BASE;
 
-	store->val[val_idx][index] = ucontrol->value.integer.value[0];
+	threshold = store->val[COMP_STORE_IDX(SND_US16X08_ID_COMP_THRESHOLD)]
+		[index];
+	ratio = store->val[COMP_STORE_IDX(SND_US16X08_ID_COMP_RATIO)][index];
+	attack = store->val[COMP_STORE_IDX(SND_US16X08_ID_COMP_ATTACK)][index];
+	release = store->val[COMP_STORE_IDX(SND_US16X08_ID_COMP_RELEASE)]
+		[index];
+	gain = store->val[COMP_STORE_IDX(SND_US16X08_ID_COMP_GAIN)][index];
+	switch_on = store->val[COMP_STORE_IDX(SND_US16X08_ID_COMP_SWITCH)]
+		[index];
+
+	switch (val_idx) {
+	case COMP_STORE_IDX(SND_US16X08_ID_COMP_THRESHOLD):
+		threshold = val;
+		break;
+	case COMP_STORE_IDX(SND_US16X08_ID_COMP_RATIO):
+		ratio = val;
+		break;
+	case COMP_STORE_IDX(SND_US16X08_ID_COMP_ATTACK):
+		attack = val;
+		break;
+	case COMP_STORE_IDX(SND_US16X08_ID_COMP_RELEASE):
+		release = val;
+		break;
+	case COMP_STORE_IDX(SND_US16X08_ID_COMP_GAIN):
+		gain = val;
+		break;
+	case COMP_STORE_IDX(SND_US16X08_ID_COMP_SWITCH):
+		switch_on = val;
+		break;
+	}
 
 	/* prepare compressor URB message from template  */
 	memcpy(buf, comp_msg, sizeof(comp_msg));
 
 	/* place comp values in message buffer watch bias! */
-	buf[8] = store->val[
-		COMP_STORE_IDX(SND_US16X08_ID_COMP_THRESHOLD)][index]
-		- SND_US16X08_COMP_THRESHOLD_BIAS;
-	buf[11] = ratio_map[store->val[
-		COMP_STORE_IDX(SND_US16X08_ID_COMP_RATIO)][index]];
-	buf[14] = store->val[COMP_STORE_IDX(SND_US16X08_ID_COMP_ATTACK)][index]
-		+ SND_US16X08_COMP_ATTACK_BIAS;
-	buf[17] = store->val[COMP_STORE_IDX(SND_US16X08_ID_COMP_RELEASE)][index]
-		+ SND_US16X08_COMP_RELEASE_BIAS;
-	buf[20] = store->val[COMP_STORE_IDX(SND_US16X08_ID_COMP_GAIN)][index];
-	buf[26] = store->val[COMP_STORE_IDX(SND_US16X08_ID_COMP_SWITCH)][index];
+	buf[8] = threshold - SND_US16X08_COMP_THRESHOLD_BIAS;
+	buf[11] = ratio_map[ratio];
+	buf[14] = attack + SND_US16X08_COMP_ATTACK_BIAS;
+	buf[17] = release + SND_US16X08_COMP_RELEASE_BIAS;
+	buf[20] = gain;
+	buf[26] = switch_on;
 
 	/* place channel selector in message buffer */
 	buf[5] = index + 1;
 
 	err = snd_us16x08_send_urb(chip, buf, sizeof(comp_msg));
 
-	if (err > 0) {
-		elem->cached |= 1 << index;
-		elem->cache_val[index] = val;
-	} else {
+	if (err < 0) {
 		usb_audio_dbg(chip, "Failed to set compressor, err:%d\n", err);
+		return err;
 	}
 
+	store->val[val_idx][index] = val;
+	elem->cached |= 1 << index;
+	elem->cache_val[index] = val;
 	return 1;
 }
 

@@ -579,11 +605,10 @@ static int snd_us16x08_eq_put(struct snd
 	/* copy URB buffer from EQ template */
 	memcpy(buf, eqs_msq, sizeof(eqs_msq));
 
-	store->val[b_idx][p_idx][index] = val;
-	buf[20] = store->val[b_idx][3][index];
-	buf[17] = store->val[b_idx][2][index];
-	buf[14] = store->val[b_idx][1][index];
-	buf[11] = store->val[b_idx][0][index];
+	buf[20] = p_idx == 3 ? val : store->val[b_idx][3][index];
+	buf[17] = p_idx == 2 ? val : store->val[b_idx][2][index];
+	buf[14] = p_idx == 1 ? val : store->val[b_idx][1][index];
+	buf[11] = p_idx == 0 ? val : store->val[b_idx][0][index];
 
 	/* place channel index in URB buffer */
 	buf[5] = index + 1;

@@ -593,14 +618,15 @@ static int snd_us16x08_eq_put(struct snd
 
 	err = snd_us16x08_send_urb(chip, buf, sizeof(eqs_msq));
 
-	if (err > 0) {
-		/* store new value in EQ band cache */
-		elem->cached |= 1 << index;
-		elem->cache_val[index] = val;
-	} else {
+	if (err < 0) {
 		usb_audio_dbg(chip, "Failed to set eq param, err:%d\n", err);
+		return err;
 	}
 
+	store->val[b_idx][p_idx][index] = val;
+	/* store new value in EQ band cache */
+	elem->cached |= 1 << index;
+	elem->cache_val[index] = val;
 	return 1;
 }
 

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Raghavendra Rao Ananta <[email protected]>

commit 40ef3edf151e184d021917a5c4c771cc0870844a upstream.

The current implementation uses pci_num_vf() while holding the memory_lock to prevent changing the power state of a PF when VFs are enabled. This creates a lockdep circular dependency warning because memory_lock is held during device probing.

[  286.997167] ======================================================
[  287.003363] WARNING: possible circular locking dependency detected
[  287.009562] 7.0.0-dbg-DEV #3 Tainted: G S
[  287.015074] ------------------------------------------------------
[  287.021270] vfio_pci_sriov_/18636 is trying to acquire lock:
[  287.026942] ff45bea2294d4968 (&vdev->memory_lock){+.+.}-{4:4}, at:
vfio_pci_core_runtime_resume+0x1f/0xa0
[  287.036530]
[  287.036530] but task is already holding lock:
[  287.042383] ff45bea3a96b8230 (&new_dev_set->lock){+.+.}-{4:4}, at:
vfio_group_fops_unl_ioctl+0x44d/0x7b0
[  287.051879]
[  287.051879] which lock already depends on the new lock.
[  287.051879]
[  287.060070]
[  287.060070] the existing dependency chain (in reverse order) is:
[  287.067568]
[  287.067568] -> #2 (&new_dev_set->lock){+.+.}-{4:4}:
[  287.073941]        __mutex_lock+0x92/0xb80
[  287.078058]        vfio_assign_device_set+0x66/0x1b0
[  287.083042]        vfio_pci_core_register_device+0xd1/0x2a0
[  287.088638]        vfio_pci_probe+0xd2/0x100
[  287.092933]        local_pci_probe_callback+0x4d/0xa0
[  287.098001]        process_scheduled_works+0x2ca/0x680
[  287.103158]        worker_thread+0x1e8/0x2f0
[  287.107452]        kthread+0x10c/0x140
[  287.111230]        ret_from_fork+0x18e/0x360
[  287.115519]        ret_from_fork_asm+0x1a/0x30
[  287.119983]
[  287.119983] -> #1 ((work_completion)(&arg.work)){+.+.}-{0:0}:
[  287.127219]        __flush_work+0x345/0x490
[  287.131429]        pci_device_probe+0x2e3/0x490
[  287.135979]        really_probe+0x1f9/0x4e0
[  287.140180]        __driver_probe_device+0x77/0x100
[  287.145079]        driver_probe_device+0x1e/0x110
[  287.149803]        __device_attach_driver+0xe3/0x170
[  287.154789]        bus_for_each_drv+0x125/0x150
[  287.159346]        __device_attach+0xca/0x1a0
[  287.163720]        device_initial_probe+0x34/0x50
[  287.168445]        pci_bus_add_device+0x6e/0x90
[  287.172995]        pci_iov_add_virtfn+0x3c9/0x3e0
[  287.177719]        sriov_add_vfs+0x2c/0x60
[  287.181838]        sriov_enable+0x306/0x4a0
[  287.186038]        vfio_pci_core_sriov_configure+0x184/0x220
[  287.191715]        sriov_numvfs_store+0xd9/0x1c0
[  287.196351]        kernfs_fop_write_iter+0x13f/0x1d0
[  287.201338]        vfs_write+0x2be/0x3b0
[  287.205286]        ksys_write+0x73/0x100
[  287.209233]        do_syscall_64+0x14d/0x750
[  287.213529]        entry_SYSCALL_64_after_hwframe+0x77/0x7f
[  287.219120]
[  287.219120] -> #0 (&vdev->memory_lock){+.+.}-{4:4}:
[  287.225491]        __lock_acquire+0x14c6/0x2800
[  287.230048]        lock_acquire+0xd3/0x2f0
[  287.234168]        down_write+0x3a/0xc0
[  287.238019]        vfio_pci_core_runtime_resume+0x1f/0xa0
[  287.243436]        __rpm_callback+0x8c/0x310
[  287.247730]        rpm_resume+0x529/0x6f0
[  287.251765]        __pm_runtime_resume+0x68/0x90
[  287.256402]        vfio_pci_core_enable+0x44/0x310
[  287.261216]        vfio_pci_open_device+0x1c/0x80
[  287.265947]        vfio_df_open+0x10f/0x150
[  287.270148]        vfio_group_fops_unl_ioctl+0x4a4/0x7b0
[  287.275476]        __se_sys_ioctl+0x71/0xc0
[  287.279679]        do_syscall_64+0x14d/0x750
[  287.283975]        entry_SYSCALL_64_after_hwframe+0x77/0x7f
[  287.289559]
[  287.289559] other info that might help us debug this:
[  287.289559]
[  287.297582] Chain exists of:
[  287.297582]   &vdev->memory_lock --> (work_completion)(&arg.work)
--> &new_dev_set->lock
[  287.297582]
[  287.310023]  Possible unsafe locking scenario:
[  287.310023]
[  287.315961]        CPU0                    CPU1
[  287.320510]        ----                    ----
[  287.325059]   lock(&new_dev_set->lock);
[  287.328917]
lock((work_completion)(&arg.work));
[  287.336153]                                lock(&new_dev_set->lock);
[  287.342523]   lock(&vdev->memory_lock);
[  287.346382]
[  287.346382]  *** DEADLOCK ***
[  287.346382]
[  287.352315] 2 locks held by vfio_pci_sriov_/18636:
[  287.357125]  #0: ff45bea208ed3e18 (&group->group_lock){+.+.}-{4:4},
at: vfio_group_fops_unl_ioctl+0x3e3/0x7b0
[  287.367048]  #1: ff45bea3a96b8230 (&new_dev_set->lock){+.+.}-{4:4},
at: vfio_group_fops_unl_ioctl+0x44d/0x7b0
[  287.376976]
[  287.376976] stack backtrace:
[  287.381353] CPU: 191 UID: 0 PID: 18636 Comm: vfio_pci_sriov_
Tainted: G S                  7.0.0-dbg-DEV #3 PREEMPTLAZY
[  287.381355] Tainted: [S]=CPU_OUT_OF_SPEC
[  287.381356] Call Trace:
[  287.381357]  <TASK>
[  287.381358]  dump_stack_lvl+0x54/0x70
[  287.381361]  print_circular_bug+0x2e1/0x300
[  287.381363]  check_noncircular+0xf9/0x120
[  287.381364]  ? __lock_acquire+0x5b4/0x2800
[  287.381366]  __lock_acquire+0x14c6/0x2800
[  287.381368]  ? pci_mmcfg_read+0x4f/0x220
[  287.381370]  ? pci_mmcfg_write+0x57/0x220
[  287.381371]  ? lock_acquire+0xd3/0x2f0
[  287.381373]  ? pci_mmcfg_write+0x57/0x220
[  287.381374]  ? lock_release+0xef/0x360
[  287.381376]  ? vfio_pci_core_runtime_resume+0x1f/0xa0
[  287.381377]  lock_acquire+0xd3/0x2f0
[  287.381378]  ? vfio_pci_core_runtime_resume+0x1f/0xa0
[  287.381379]  ? lock_is_held_type+0x76/0x100
[  287.381382]  down_write+0x3a/0xc0
[  287.381382]  ? vfio_pci_core_runtime_resume+0x1f/0xa0
[  287.381383]  vfio_pci_core_runtime_resume+0x1f/0xa0
[  287.381384]  ? __pfx_pci_pm_runtime_resume+0x10/0x10
[  287.381385]  __rpm_callback+0x8c/0x310
[  287.381386]  ? ktime_get_mono_fast_ns+0x3d/0xb0
[  287.381389]  ? __pfx_pci_pm_runtime_resume+0x10/0x10
[  287.381390]  rpm_resume+0x529/0x6f0
[  287.381392]  ? lock_is_held_type+0x76/0x100
[  287.381394]  __pm_runtime_resume+0x68/0x90
[  287.381396]  vfio_pci_core_enable+0x44/0x310
[  287.381398]  vfio_pci_open_device+0x1c/0x80
[  287.381399]  vfio_df_open+0x10f/0x150
[  287.381401]  vfio_group_fops_unl_ioctl+0x4a4/0x7b0
[  287.381402]  __se_sys_ioctl+0x71/0xc0
[  287.381404]  do_syscall_64+0x14d/0x750
[  287.381405]  ? entry_SYSCALL_64_after_hwframe+0x77/0x7f
[  287.381406]  ? trace_irq_disable+0x25/0xd0
[  287.381409]  entry_SYSCALL_64_after_hwframe+0x77/0x7f

Introduce a private flag 'sriov_active' in the vfio_pci_core_device struct. This allows the driver to track the SR-IOV power state requirement without relying on pci_num_vf() while holding the memory_lock. The lock is now only held to set the flag and ensure the device is in D0, after which pci_enable_sriov() can be called without the lock.

---
 drivers/vfio/pci/vfio_pci_core.c |   17 ++++++++++++++---
 include/linux/vfio_pci_core.h    |    1 +
 2 files changed, 15 insertions(+), 3 deletions(-)

@@ -242,8 +242,11 @@ int vfio_pci_set_power_state(struct vfio
 	int ret;
 
 	/* Prevent changing power state for PFs with VFs enabled */
-	if (pci_num_vf(pdev) && state > PCI_D0)
-		return -EBUSY;
+	if (state > PCI_D0) {
+		lockdep_assert_held_write(&vdev->memory_lock);
+		if (vdev->sriov_active)
+			return -EBUSY;
+	}
 
 	if (vdev->needs_pm_restore) {
 		if (pdev->current_state < PCI_D3hot && state >= PCI_D3hot) {

@@ -2236,8 +2239,9 @@ int vfio_pci_core_sriov_configure(struct
 
 		down_write(&vdev->memory_lock);
 		vfio_pci_set_power_state(vdev, PCI_D0);
-		ret = pci_enable_sriov(pdev, nr_virtfn);
+		vdev->sriov_active = true;
 		up_write(&vdev->memory_lock);
+		ret = pci_enable_sriov(pdev, nr_virtfn);
 		if (ret) {
 			pm_runtime_put(&pdev->dev);
 			goto out_del;

@@ -2251,6 +2255,13 @@ int vfio_pci_core_sriov_configure(struct
 	}
 
 out_del:
+	/*
+	 * Avoid taking the memory_lock intentionally. A race with a power
+	 * state transition would at most result in an -EBUSY, leaving the
+	 * device in PCI_D0.
+	 */
+	vdev->sriov_active = false;
+
 	mutex_lock(&vfio_pci_sriov_pfs_mutex);
 	list_del_init(&vdev->sriov_pfs_item);
 out_unlock:

@@ -80,6 +80,7 @@ struct vfio_pci_core_device {
 	bool			needs_pm_restore:1;
 	bool			pm_intx_masked:1;
 	bool			pm_runtime_engaged:1;
+	bool			sriov_active;
 	struct pci_saved_state	*pci_saved_state;
 	struct pci_saved_state	*pm_save;
 	int			ioeventfds_nr;

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Alex Williamson <[email protected]>

commit 4575e9aac5336d1365138c0284773bf8da4b1fa3 upstream.

When disable_idle_d3 was introduced in vfio-pci, it directly manipulated the device power state with pci_set_power_state(). There were no refcounts to maintain or balanced operations, we could unconditionally bring the device to D0 and conditionally move it to D3hot. Therefore the module parameter was made writable.

Later, in commit c61302aa48f7 ("vfio/pci: Move module parameters to vfio_pci.c"), as part of the vfio-pci-core split, the writable aspect of the module parameter was nullified. The parameter value could still be changed through sysfs, but the vfio-pci driver latched the values into vfio-pci-core globals at module init. Loading the vfio-pci module, or unloading and reloading, with non-default or different values could change the globals relative to existing devices bound to vfio-pci variant drivers.

Runtime PM was introduced in commit 7ab5e10eda02 ("vfio/pci: Move the unused device into low power state with runtime PM"), which marks the point where power states became refcounted. PM get and put operations need to be balanced, but the same module operations noted above can change the global variables relative to those devices already bound to vfio-pci variant drivers. This introduces a window where PM operations can now become unbalanced.

To resolve this with a narrow footprint for stable backports, the disable_idle_d3 flag is latched into the vfio_pci_core_device at the time of initialization, such that the device always operates with a consistent value.

NB. vfio_pci_dev_set_try_reset() now unconditionally raises the runtime PM usage count around bus reset to account for disable_idle_d3 becoming a per-device rather than global flag. When this flag is set, the additional get/put pair is harmless and allows continued use of the shared vfio_pci_dev_set_pm_runtime_get() helper.

---
 drivers/vfio/pci/vfio_pci_core.c |   19 ++++++++++---------
 include/linux/vfio_pci_core.h    |    1 +
 2 files changed, 11 insertions(+), 9 deletions(-)

@@ -477,7 +477,7 @@ int vfio_pci_core_enable(struct vfio_pci
 	u16 cmd;
 	u8 msix_pos;
 
-	if (!disable_idle_d3) {
+	if (!vdev->disable_idle_d3) {
 		ret = pm_runtime_resume_and_get(&pdev->dev);
 		if (ret < 0)
 			return ret;

@@ -555,7 +555,7 @@ out_free_state:
 out_disable_device:
 	pci_disable_device(pdev);
 out_power:
-	if (!disable_idle_d3)
+	if (!vdev->disable_idle_d3)
 		pm_runtime_put(&pdev->dev);
 	return ret;
 }

@@ -682,7 +682,7 @@ out:
 	vfio_pci_dev_set_try_reset(vdev->vdev.dev_set);
 
 	/* Put the pm-runtime usage counter acquired during enable */
-	if (!disable_idle_d3)
+	if (!vdev->disable_idle_d3)
 		pm_runtime_put(&pdev->dev);
 }
 EXPORT_SYMBOL_GPL(vfio_pci_core_disable);

@@ -2058,6 +2058,8 @@ int vfio_pci_core_init_dev(struct vfio_d
 	init_rwsem(&vdev->memory_lock);
 	xa_init(&vdev->ctx);
 
+	vdev->disable_idle_d3 = disable_idle_d3;
+
 	return 0;
 }
 EXPORT_SYMBOL_GPL(vfio_pci_core_init_dev);

@@ -2149,7 +2151,7 @@ int vfio_pci_core_register_device(struct
 
 	dev->driver->pm = &vfio_pci_core_pm_ops;
 	pm_runtime_allow(dev);
-	if (!disable_idle_d3)
+	if (!vdev->disable_idle_d3)
 		pm_runtime_put(dev);
 
 	ret = vfio_register_group_dev(&vdev->vdev);

@@ -2158,7 +2160,7 @@ int vfio_pci_core_register_device(struct
 	return 0;
 
 out_power:
-	if (!disable_idle_d3)
+	if (!vdev->disable_idle_d3)
 		pm_runtime_get_noresume(dev);
 
 	pm_runtime_forbid(dev);

@@ -2177,7 +2179,7 @@ void vfio_pci_core_unregister_device(str
 	vfio_pci_vf_uninit(vdev);
 	vfio_pci_vga_uninit(vdev);
 
-	if (!disable_idle_d3)
+	if (!vdev->disable_idle_d3)
 		pm_runtime_get_noresume(&vdev->pdev->dev);
 
 	pm_runtime_forbid(&vdev->pdev->dev);

@@ -2505,7 +2507,7 @@ static void vfio_pci_dev_set_try_reset(s
 	 * state. Increment the usage count for all the devices in the dev_set
 	 * before reset and decrement the same after reset.
 	 */
-	if (!disable_idle_d3 && vfio_pci_dev_set_pm_runtime_get(dev_set))
+	if (vfio_pci_dev_set_pm_runtime_get(dev_set))
 		return;
 
 	if (!pci_reset_bus(pdev))

@@ -2515,8 +2517,7 @@ static void vfio_pci_dev_set_try_reset(s
 		if (reset_done)
 			cur->needs_reset = false;
 
-		if (!disable_idle_d3)
-			pm_runtime_put(&cur->pdev->dev);
+		pm_runtime_put(&cur->pdev->dev);
 	}
 }
 

@@ -80,6 +80,7 @@ struct vfio_pci_core_device {
 	bool			needs_pm_restore:1;
 	bool			pm_intx_masked:1;
 	bool			pm_runtime_engaged:1;
+	bool			disable_idle_d3:1;
 	bool			sriov_active;
 	struct pci_saved_state	*pci_saved_state;
 	struct pci_saved_state	*pm_save;

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Alex Williamson <[email protected]>

commit daedde7f024ecf88bc8e832ed40cf2c795f0796a upstream.

The re-order in the Fixes commit below displaced vfio_pci_vga_init() as the last failure point of what is now vfio_pci_core_register_device() without introducing an unwind for the VGA arbiter registration.

In current kernels this is mostly benign because vfio_pci_set_decode() only uses pci_dev state, but the original failure path could leave a callback with a freed vdev cookie. The stale registration also becomes unsafe again once the callback follows drvdata to the vfio device.

Add the required VGA unwind callout.

---
 drivers/vfio/pci/vfio_pci_core.c |    1 +
 1 file changed, 1 insertion(+)

@@ -2164,6 +2164,7 @@ out_power:
 		pm_runtime_get_noresume(dev);
 
 	pm_runtime_forbid(dev);
+	vfio_pci_vga_uninit(vdev);
 out_vf:
 	vfio_pci_vf_uninit(vdev);
 	return ret;

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Alex Williamson <[email protected]>

commit e73638e55f861758d49f14d7bb5dba3035981cd7 upstream.

Bitfield operations are not atomic, they use a read-modify-write pattern, therefore we should be careful not to pack bitfields that can be concurrently updated into the same storage unit.

This split takes a binary approach: flags that are only modified pre/post open/close remain bitfields, flags modified from user action, including actions that reach across to another device (ex. reset) use dedicated storage units.

Note that the virq_disabled and bardirty flags are relocated to fill an existing hole in the structure.

Bitfield justifications:

has_dyn_msix: written only in vfio_pci_core_enable()
pci_2_3: written only in vfio_pci_core_enable()
reset_works: written only in vfio_pci_core_enable()
extended_caps: written only in vfio_cap_len() under vfio_config_init()
has_vga: written only in vfio_pci_core_enable()
nointx: written only in vfio_pci_core_enable()
needs_pm_restore: written only in vfio_pci_probe_power_state()
disable_idle_d3: written only at .init in vfio_pci_core_init_dev()

Dedicated storage units:

virq_disabled: written by guest INTx command writes in
vfio_basic_config_write() while the device is open
bardirty: written by guest BAR writes in vfio_basic_config_write()
while the device is open
pm_intx_masked: written in the runtime-PM suspend path.
pm_runtime_engaged: written by low-power feature entry/exit paths
needs_reset: set in vfio_pci_core_disable() and cleared for devices in
the set by vfio_pci_dev_set_try_reset()
sriov_active: written by vfio_pci_core_sriov_configure() via sysfs
sriov_numvfs while bound.

---
 include/linux/vfio_pci_core.h |   13 ++++++++-----
 1 file changed, 8 insertions(+), 5 deletions(-)

@@ -54,6 +54,9 @@ struct vfio_pci_core_device {
 	struct pci_dev		*pdev;
 	void __iomem		*barmap[PCI_STD_NUM_BARS];
 	bool			bar_mmap_supported[PCI_STD_NUM_BARS];
+	/* Flags modified at runtime - dedicated storage unit */
+	bool			virq_disabled;
+	bool			bardirty;
 	u8			*pci_config_map;
 	u8			*vconfig;
 	struct perm_bits	*msi_perm;

@@ -68,19 +71,19 @@ struct vfio_pci_core_device {
 	u16			msix_size;
 	u32			msix_offset;
 	u32			rbar[7];
+	/* Flags only modified on setup/release - bitfield ok */
 	bool			has_dyn_msix:1;
 	bool			pci_2_3:1;
-	bool			virq_disabled:1;
 	bool			reset_works:1;
 	bool			extended_caps:1;
-	bool			bardirty:1;
 	bool			has_vga:1;
-	bool			needs_reset:1;
 	bool			nointx:1;
 	bool			needs_pm_restore:1;
-	bool			pm_intx_masked:1;
-	bool			pm_runtime_engaged:1;
 	bool			disable_idle_d3:1;
+	/* Flags modified at runtime - dedicated storage unit */
+	bool			needs_reset;
+	bool			pm_intx_masked;
+	bool			pm_runtime_engaged;
 	bool			sriov_active;
 	struct pci_saved_state	*pci_saved_state;
 	struct pci_saved_state	*pm_save;

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Junrui Luo <[email protected]>

commit a26b499b757cfc8bbff1088bb1b844639e250893 upstream.

vfio_mig_get_next_state() walks vfio_from_fsm_table[] one step at a time, looping to skip optional states the device does not support until *next_fsm is supported. A blocked transition is encoded as VFIO_DEVICE_STATE_ERROR, which the trailing return reports as -EINVAL.

The skip loop does not account for the ERROR sentinel. state_flags_table[ERROR] is ~0U and vfio_from_fsm_table[ERROR][*] is ERROR, so once *next_fsm becomes ERROR the loop condition stays true and *next_fsm never changes. The blocked arcs STOP_COPY -> PRE_COPY and STOP_COPY -> PRE_COPY_P2P map to ERROR yet pass the support check on a precopy-capable device, causing the loop to spin forever while holding the driver state mutex. This can result in a soft lockup, and a panic with softlockup_panic set.

Terminate the skip loop on the ERROR sentinel so a blocked transition falls through to the existing return and reports -EINVAL.

---
 drivers/vfio/vfio_main.c |    3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

@@ -843,7 +843,8 @@ int vfio_mig_get_next_state(struct vfio_
 	 * logical state, as per the above comment.
 	 */
 	*next_fsm = vfio_from_fsm_table[cur_fsm][new_fsm];
-	while ((state_flags_table[*next_fsm] & device->migration_flags) !=
+	while (*next_fsm != VFIO_DEVICE_STATE_ERROR &&
+	       (state_flags_table[*next_fsm] & device->migration_flags) !=
 			state_flags_table[*next_fsm])
 		*next_fsm = vfio_from_fsm_table[*next_fsm][new_fsm];
 

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Zenm Chen <[email protected]>

commit 88b4d528eda4ac71c2952b3458f2abbc80a91cd2 upstream.

Add USB ID 2c4e:0128 for Mercusys MA60XNB, an RTL8851BU-based Wi-Fi + Bluetooth adapter.

The information in /sys/kernel/debug/usb/devices about the Bluetooth device is listed as the below:

T:  Bus=03 Lev=01 Prnt=01 Port=04 Cnt=01 Dev#=  3 Spd=480  MxCh= 0
D:  Ver= 2.00 Cls=ef(misc ) Sub=02 Prot=01 MxPS=64 #Cfgs=  1
P:  Vendor=2c4e ProdID=0128 Rev= 0.00
S:  Manufacturer=Realtek
S:  Product=802.11ax WLAN Adapter
S:  SerialNumber=00e04c000001
C:* #Ifs= 3 Cfg#= 1 Atr=e0 MxPwr=500mA
A:  FirstIf#= 0 IfCount= 2 Cls=e0(wlcon) Sub=01 Prot=01
I:* If#= 0 Alt= 0 #EPs= 3 Cls=e0(wlcon) Sub=01 Prot=01 Driver=btusb
E:  Ad=81(I) Atr=03(Int.) MxPS=  16 Ivl=1ms
E:  Ad=02(O) Atr=02(Bulk) MxPS= 512 Ivl=0ms
E:  Ad=82(I) Atr=02(Bulk) MxPS= 512 Ivl=0ms
I:* If#= 1 Alt= 0 #EPs= 2 Cls=e0(wlcon) Sub=01 Prot=01 Driver=btusb
E:  Ad=03(O) Atr=01(Isoc) MxPS=   0 Ivl=1ms
E:  Ad=83(I) Atr=01(Isoc) MxPS=   0 Ivl=1ms
I:  If#= 1 Alt= 1 #EPs= 2 Cls=e0(wlcon) Sub=01 Prot=01 Driver=btusb
E:  Ad=03(O) Atr=01(Isoc) MxPS=   9 Ivl=1ms
E:  Ad=83(I) Atr=01(Isoc) MxPS=   9 Ivl=1ms
I:  If#= 1 Alt= 2 #EPs= 2 Cls=e0(wlcon) Sub=01 Prot=01 Driver=btusb
E:  Ad=03(O) Atr=01(Isoc) MxPS=  17 Ivl=1ms
E:  Ad=83(I) Atr=01(Isoc) MxPS=  17 Ivl=1ms
I:  If#= 1 Alt= 3 #EPs= 2 Cls=e0(wlcon) Sub=01 Prot=01 Driver=btusb
E:  Ad=03(O) Atr=01(Isoc) MxPS=  25 Ivl=1ms
E:  Ad=83(I) Atr=01(Isoc) MxPS=  25 Ivl=1ms
I:  If#= 1 Alt= 4 #EPs= 2 Cls=e0(wlcon) Sub=01 Prot=01 Driver=btusb
E:  Ad=03(O) Atr=01(Isoc) MxPS=  33 Ivl=1ms
E:  Ad=83(I) Atr=01(Isoc) MxPS=  33 Ivl=1ms
I:  If#= 1 Alt= 5 #EPs= 2 Cls=e0(wlcon) Sub=01 Prot=01 Driver=btusb
E:  Ad=03(O) Atr=01(Isoc) MxPS=  49 Ivl=1ms
E:  Ad=83(I) Atr=01(Isoc) MxPS=  49 Ivl=1ms
I:  If#= 1 Alt= 6 #EPs= 2 Cls=e0(wlcon) Sub=01 Prot=01 Driver=btusb
E:  Ad=03(O) Atr=01(Isoc) MxPS=  63 Ivl=1ms
E:  Ad=83(I) Atr=01(Isoc) MxPS=  63 Ivl=1ms
I:* If#= 2 Alt= 0 #EPs= 8 Cls=ff(vend.) Sub=ff Prot=ff Driver=rtw89_8851bu
E:  Ad=84(I) Atr=02(Bulk) MxPS= 512 Ivl=0ms
E:  Ad=05(O) Atr=02(Bulk) MxPS= 512 Ivl=0ms
E:  Ad=06(O) Atr=02(Bulk) MxPS= 512 Ivl=0ms
E:  Ad=07(O) Atr=02(Bulk) MxPS= 512 Ivl=0ms
E:  Ad=09(O) Atr=02(Bulk) MxPS= 512 Ivl=0ms
E:  Ad=0a(O) Atr=02(Bulk) MxPS= 512 Ivl=0ms
E:  Ad=0b(O) Atr=02(Bulk) MxPS= 512 Ivl=0ms
E:  Ad=0c(O) Atr=02(Bulk) MxPS= 512 Ivl=0ms

---
 drivers/bluetooth/btusb.c |    2 ++
 1 file changed, 2 insertions(+)

@@ -518,6 +518,8 @@ static const struct usb_device_id quirks
 						     BTUSB_WIDEBAND_SPEECH },
 	{ USB_DEVICE(0x7392, 0xe611), .driver_info = BTUSB_REALTEK |
 						     BTUSB_WIDEBAND_SPEECH },
+	{ USB_DEVICE(0x2c4e, 0x0128), .driver_info = BTUSB_REALTEK |
+						     BTUSB_WIDEBAND_SPEECH },
 
 	/* Realtek 8852AE Bluetooth devices */
 	{ USB_DEVICE(0x0bda, 0x2852), .driver_info = BTUSB_REALTEK |

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Johan Hovold <[email protected]>

commit eedc6867ebad73edbfaf9a0a65fbef7115cc4753 upstream.

Make sure to release the sibling interfaces in case controller registration fails to avoid use-after-free and double-free when they are eventually disconnected.

This issue was reported by Sashiko while reviewing a fix for a wakeup source leak in the btusb probe errors paths.

---
 drivers/bluetooth/btusb.c |   11 ++++++++++-
 1 file changed, 10 insertions(+), 1 deletion(-)

@@ -4078,7 +4078,7 @@ static int btusb_probe(struct usb_interf
 
 	err = hci_register_dev(hdev);
 	if (err < 0)
-		goto out_free_dev;
+		goto err_release_siblings;
 
 	usb_set_intfdata(intf, data);
 

@@ -4087,6 +4087,15 @@ static int btusb_probe(struct usb_interf
 
 	return 0;
 
+err_release_siblings:
+	if (data->diag) {
+		usb_set_intfdata(data->diag, NULL);
+		usb_driver_release_interface(&btusb_driver, data->diag);
+	}
+	if (data->isoc) {
+		usb_set_intfdata(data->isoc, NULL);
+		usb_driver_release_interface(&btusb_driver, data->isoc);
+	}
 out_free_dev:
 	if (data->reset_gpio)
 		gpiod_put(data->reset_gpio);

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Johan Hovold <[email protected]>

commit c5b600a3c05b1a7a110d558df935a8fc8a471c79 upstream.

Make sure to stop any TX URBs submitted during Marvell OOB wakeup configuration on later probe failures to avoid use-after-free in the completion callback.

This issue was reported by Sashiko while reviewing a fix for a wakeup source leak in the btusb probe errors paths.

---
 drivers/bluetooth/btusb.c |    8 +++++---
 1 file changed, 5 insertions(+), 3 deletions(-)

@@ -3894,7 +3894,7 @@ static int btusb_probe(struct usb_interf
 	if (id->driver_info & BTUSB_INTEL_COMBINED) {
 		err = btintel_configure_setup(hdev, btusb_driver.name);
 		if (err)
-			goto out_free_dev;
+			goto err_kill_tx_urbs;
 
 		/* Transport specific configuration */
 		hdev->send = btusb_send_frame_intel;

@@ -4052,7 +4052,7 @@ static int btusb_probe(struct usb_interf
 		err = usb_set_interface(data->udev, 0, 0);
 		if (err < 0) {
 			BT_ERR("failed to set interface 0, alt 0 %d", err);
-			goto out_free_dev;
+			goto err_kill_tx_urbs;
 		}
 	}
 

@@ -4060,7 +4060,7 @@ static int btusb_probe(struct usb_interf
 		err = usb_driver_claim_interface(&btusb_driver,
 						 data->isoc, data);
 		if (err < 0)
-			goto out_free_dev;
+			goto err_kill_tx_urbs;
 	}
 
 	if (IS_ENABLED(CONFIG_BT_HCIBTUSB_BCM) && data->diag) {

@@ -4096,6 +4096,8 @@ err_release_siblings:
 		usb_set_intfdata(data->isoc, NULL);
 		usb_driver_release_interface(&btusb_driver, data->isoc);
 	}
+err_kill_tx_urbs:
+	usb_kill_anchored_urbs(&data->tx_anchor);
 out_free_dev:
 	if (data->reset_gpio)
 		gpiod_put(data->reset_gpio);

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Johan Hovold <[email protected]>

commit 3d93e1bb0fb881fe3ef961d1120556658e9cac4d upstream.

Make sure to disable wakeup on probe failure to avoid leaking the wakeup source.

---
 drivers/bluetooth/btusb.c |   18 +++++++++++++++---
 1 file changed, 15 insertions(+), 3 deletions(-)

@@ -2811,6 +2811,11 @@ static int marvell_config_oob_wake(struc
 
 	return 0;
 }
+#else
+static inline int marvell_config_oob_wake(struct hci_dev *hdev)
+{
+	return 0;
+}
 #endif
 
 static int btusb_set_bdaddr_marvell(struct hci_dev *hdev,

@@ -3625,6 +3630,11 @@ static int btusb_config_oob_wake(struct
 	bt_dev_info(hdev, "OOB Wake-on-BT configured at IRQ %u", irq);
 	return 0;
 }
+#else
+static inline int btusb_config_oob_wake(struct hci_dev *hdev)
+{
+	return 0;
+}
 #endif
 
 static void btusb_check_needs_reset_resume(struct usb_interface *intf)

@@ -3848,7 +3858,6 @@ static int btusb_probe(struct usb_interf
 	hdev->notify = btusb_notify;
 	hdev->wakeup = btusb_wakeup;
 
-#ifdef CONFIG_PM
 	err = btusb_config_oob_wake(hdev);
 	if (err)
 		goto out_free_dev;

@@ -3857,9 +3866,9 @@ static int btusb_probe(struct usb_interf
 	if (id->driver_info & BTUSB_MARVELL && data->oob_wake_irq) {
 		err = marvell_config_oob_wake(hdev);
 		if (err)
-			goto out_free_dev;
+			goto err_disable_wakeup;
 	}
-#endif
+
 	if (id->driver_info & BTUSB_CW6622)
 		set_bit(HCI_QUIRK_BROKEN_STORED_LINK_KEY, &hdev->quirks);
 

@@ -4098,6 +4107,9 @@ err_release_siblings:
 	}
 err_kill_tx_urbs:
 	usb_kill_anchored_urbs(&data->tx_anchor);
+err_disable_wakeup:
+	if (data->oob_wake_irq)
+		device_init_wakeup(&data->udev->dev, false);
 out_free_dev:
 	if (data->reset_gpio)
 		gpiod_put(data->reset_gpio);

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Carlos Llamas <[email protected]>

commit 114a116aaa5f0295376cdf12da743c5bce3b20ce upstream.

When a thread exits, binder_thread_release() walks its transaction stack to clear the t->from and t->to_proc that correspond with the exiting thread. However, a process dying in parallel might attempt to kfree some of these transactions. And if one of them has no associated t->to_proc, the t->to_proc->inner_lock will not be acquired.

This means that transaction accesses in binder_thread_release() after t->to_proc has been cleared might race with binder_free_transaction() and cause a use-after-free error as reported by KASAN:

==================================================================

  BUG: KASAN: slab-use-after-free in binder_thread_release+0x5d0/0x798
  Write of size 8 at addr ffff000016627500 by task X/715

  CPU: 17 UID: 0 PID: 715 Comm: X Not tainted 7.1.0-rc5-00149-g8fde5d1d47f6 #30 PREEMPT
  Hardware name: linux,dummy-virt (DT)
  Call trace:
   binder_thread_release+0x5d0/0x798
   binder_ioctl+0x12c0/0x299c
   [...]

  Allocated by task 717 on cpu 18 at 67.267803s:
   __kasan_kmalloc+0xa0/0xbc
   __kmalloc_cache_noprof+0x174/0x444
   binder_transaction+0x554/0x8150
   binder_thread_write+0xa30/0x4354
   binder_ioctl+0x20f0/0x299c
   [...]

  Freed by task 202 on cpu 18 at 90.416221s:
   __kasan_slab_free+0x58/0x80
   kfree+0x1a0/0x4a4
   binder_free_transaction+0x150/0x294
   binder_send_failed_reply+0x398/0x6d8
   binder_release_work+0x3e4/0x4ec
   binder_deferred_func+0xbd8/0x104c
   [...]
  ==================================================================

In order to avoid this, make sure that binder_free_transaction() reads the t->to_proc under the transaction lock. This will serialize the transaction release with the accesses in binder_thread_release(). Plus, it matches the documented locking rules for @to_proc.

---
 drivers/android/binder.c |    6 +++++-
 1 file changed, 5 insertions(+), 1 deletion(-)

@@ -1567,7 +1567,11 @@ static void binder_txn_latency_free(stru
 
 static void binder_free_transaction(struct binder_transaction *t)
 {
-	struct binder_proc *target_proc = t->to_proc;
+	struct binder_proc *target_proc;
+
+	spin_lock(&t->lock);
+	target_proc = t->to_proc;
+	spin_unlock(&t->lock);
 
 	if (target_proc) {
 		binder_inner_proc_lock(target_proc);

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Carlos Llamas <[email protected]>

commit f223d27a546c1e1f48d38fd67760e78f068fe8c4 upstream.

In binder_free_transaction(), the t->to_proc is read under the t->lock. However, once the t->lock is dropped, the to_proc can die in parallel. This leads to a use-after-free error when we attempt to acquire its inner lock right afterwards:

==================================================================

  BUG: KASAN: slab-use-after-free in _raw_spin_lock+0xe4/0x1a0
  Write of size 4 at addr ffff00001125da70 by task B/672

  CPU: 20 UID: 0 PID: 672 Comm: B Not tainted 7.1.0-rc6-00284-g8e65320d91cd #4 PREEMPT
  Hardware name: linux,dummy-virt (DT)
  Call trace:
   _raw_spin_lock+0xe4/0x1a0
   binder_free_transaction+0x8c/0x320
   binder_send_failed_reply+0x21c/0x2f8
   binder_thread_release+0x488/0x7e0
   binder_ioctl+0x12c0/0x29a0
  [...]

  Allocated by task 675:
   __kmalloc_cache_noprof+0x174/0x444
   binder_open+0x118/0xb70
   do_dentry_open+0x374/0x1040
   vfs_open+0x58/0x3bc
  [...]

  Freed by task 212:
   __kasan_slab_free+0x58/0x80
   kfree+0x1a0/0x4a4
   binder_proc_dec_tmpref+0x32c/0x5e0
   binder_deferred_func+0xc48/0x104c
   process_one_work+0x53c/0xbc0
  [...]
  ==================================================================

To prevent this, pin the target thread (t->to_thread) to guarantee the target process remains alive. Undelivered transactions without a target thread are already safe, as the target process can only be the current context in those paths.

---
 drivers/android/binder.c |   13 +++++++++++++
 1 file changed, 13 insertions(+)

@@ -1567,10 +1567,19 @@ static void binder_txn_latency_free(stru
 
 static void binder_free_transaction(struct binder_transaction *t)
 {
+	struct binder_thread *target_thread;
 	struct binder_proc *target_proc;
 
 	spin_lock(&t->lock);
 	target_proc = t->to_proc;
+	target_thread = t->to_thread;
+	/*
+	 * Pin target_thread to keep target_proc alive. Undelivered
+	 * transactions with !target_thread are safe, as target_proc
+	 * can only be the current context there.
+	 */
+	if (target_thread)
+		atomic_inc(&target_thread->tmp_ref);
 	spin_unlock(&t->lock);
 
 	if (target_proc) {

@@ -1585,6 +1594,10 @@ static void binder_free_transaction(stru
 			t->buffer->transaction = NULL;
 		binder_inner_proc_unlock(target_proc);
 	}
+
+	if (target_thread)
+		binder_thread_dec_tmpref(target_thread);
+
 	if (trace_binder_txn_latency_free_enabled())
 		binder_txn_latency_free(t);
 	/*

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: 胡连勤 <[email protected]>

commit 42c37c4b75d38b51d84f31a8e29427f5e06a7c2a upstream.

When a USB device with active stream endpoints is disconnected, xhci_free_streams() is called from the hub_event workqueue to free the stream resources. It calls xhci_free_stream_info() while holding xhci->lock with irqs disabled.

xhci_free_stream_info() invokes xhci_free_stream_ctx(), which calls dma_free_coherent() for large stream context arrays.

dma_free_coherent() can sleep (e.g. via vunmap), triggering a BUG when called from atomic context.

Call trace:
 dma_free_attrs+0x174/0x220
 xhci_free_stream_info+0xd0/0x11c
 xhci_free_streams+0x278/0x37c
 usb_free_streams+0x98/0xc0
 usb_unbind_interface+0x1b8/0x2f8
 device_release_driver_internal+0x1d4/0x2cc
 device_release_driver+0x18/0x28
 bus_remove_device+0x160/0x1a4
 device_del+0x1ec/0x350
 usb_disable_device+0x98/0x214
 usb_disconnect+0xf0/0x35c
 hub_event+0xab4/0x19ec
 process_one_work+0x278/0x63c

Fix this by saving the stream_info pointers and clearing the ep references under the lock, then calling xhci_free_stream_info() outside the lock where sleeping is allowed.

---
 drivers/usb/host/xhci.c |   11 ++++++++++-
 1 file changed, 10 insertions(+), 1 deletion(-)

@@ -3577,6 +3577,7 @@ static int xhci_free_streams(struct usb_
 	struct xhci_virt_device *vdev;
 	struct xhci_command *command;
 	struct xhci_input_control_ctx *ctrl_ctx;
+	struct xhci_stream_info *stream_info[EP_CTX_PER_DEV];
 	unsigned int ep_index;
 	unsigned long flags;
 	u32 changed_ep_bitmask;

@@ -3637,10 +3638,15 @@ static int xhci_free_streams(struct usb_
 	if (ret < 0)
 		return ret;
 
+	/*
+	 * dma_free_coherent() called by xhci_free_stream_info() may sleep,
+	 * so save stream_info pointers and clear references under lock,
+	 * then free the memory outside lock.
+	 */
 	spin_lock_irqsave(&xhci->lock, flags);
 	for (i = 0; i < num_eps; i++) {
 		ep_index = xhci_get_endpoint_index(&eps[i]->desc);
-		xhci_free_stream_info(xhci, vdev->eps[ep_index].stream_info);
+		stream_info[i] = vdev->eps[ep_index].stream_info;
 		vdev->eps[ep_index].stream_info = NULL;
 		/* FIXME Unset maxPstreams in endpoint context and
 		 * update deq ptr to point to normal string ring.

@@ -3650,6 +3656,9 @@ static int xhci_free_streams(struct usb_
 	}
 	spin_unlock_irqrestore(&xhci->lock, flags);
 
+	for (i = 0; i < num_eps; i++)
+		xhci_free_stream_info(xhci, stream_info[i]);
+
 	return 0;
 }
 

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Myeonghun Pak <[email protected]>

commit e8da46d99d3710106e7c44db14566bf9b57386b5 upstream.

rt1711h_probe() registers the TCPCI port before requesting the interrupt and enabling alert interrupts. If either of those later steps fails, the probe function returns without unregistering the TCPCI port. The explicit unregister currently only happens from the remove callback.

Register a devres action immediately after tcpci_register_port() succeeds, so tcpci_unregister_port() runs on later probe failures and on driver detach. Drop the remove callback to avoid unregistering the same port twice.

This issue was identified during our ongoing static-analysis research while reviewing kernel code.

---
 drivers/usb/typec/tcpm/tcpci_rt1711h.c |   13 ++++++++-----
 1 file changed, 8 insertions(+), 5 deletions(-)

@@ -327,6 +327,8 @@ static int rt1711h_check_revision(struct
 	return ret;
 }
 
+static void rt1711h_unregister_tcpci_port(void *tcpci);
+
 static int rt1711h_probe(struct i2c_client *client)
 {
 	int ret;

@@ -378,6 +380,10 @@ static int rt1711h_probe(struct i2c_clie
 	if (IS_ERR_OR_NULL(chip->tcpci))
 		return PTR_ERR(chip->tcpci);
 
+	ret = devm_add_action_or_reset(chip->dev, rt1711h_unregister_tcpci_port, chip->tcpci);
+	if (ret)
+		return ret;
+
 	ret = devm_request_threaded_irq(chip->dev, client->irq, NULL,
 					rt1711h_irq,
 					IRQF_ONESHOT | IRQF_TRIGGER_LOW,

@@ -395,11 +401,9 @@ static int rt1711h_probe(struct i2c_clie
 	return 0;
 }
 
-static void rt1711h_remove(struct i2c_client *client)
+static void rt1711h_unregister_tcpci_port(void *tcpci)
 {
-	struct rt1711h_chip *chip = i2c_get_clientdata(client);
-
-	tcpci_unregister_port(chip->tcpci);
+	tcpci_unregister_port(tcpci);
 }
 
 static const struct i2c_device_id rt1711h_id[] = {

@@ -424,7 +428,6 @@ static struct i2c_driver rt1711h_i2c_dri
 		.of_match_table = of_match_ptr(rt1711h_of_match),
 	},
 	.probe = rt1711h_probe,
-	.remove = rt1711h_remove,
 	.id_table = rt1711h_id,
 };
 module_i2c_driver(rt1711h_i2c_driver);

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Mahesh Vaidya <[email protected]>

commit 5ef4bac02189bee0b7c170e352d7a38e13fe9678 upstream.

altera_pcie_irq_teardown() calls irq_dispose_mapping() on pcie->irq. However, pcie->irq is the parent IRQ returned by platform_get_irq(), not the mapping created by Altera INTx irq_domain.

The Altera driver only sets the chained handler on the parent IRQ. It should detach that handler during teardown, but it should not dispose the parent IRQ mapping, which belongs to the parent interrupt controller's irq_domain.

Drop irq_dispose_mapping(pcie->irq) from the teardown path.

Note that during irqchip remove(), the child IRQs should've disposed. But since the chained handler itself is removed, there is no way the stale child IRQs (if exists) could fire. So it is safe here.

---
 drivers/pci/controller/pcie-altera.c |    1 -
 1 file changed, 1 deletion(-)

@@ -684,7 +684,6 @@ static void altera_pcie_irq_teardown(str
 {
 	irq_set_chained_handler_and_data(pcie->irq, NULL, NULL);
 	irq_domain_remove(pcie->irq_domain);
-	irq_dispose_mapping(pcie->irq);
 }
 
 static int altera_pcie_parse_dt(struct altera_pcie *pcie)

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Ratheesh Kannoth <[email protected]>

commit fda8749ba73638f5bbca3ffb39bc6861eb3b23fa upstream.

pci_host_common_init() is used by several generic ECAM host drivers. After PCI core changes around pci_flags and preserve_config, these hosts no longer opted into full bus number reassignment the way they did before, which broke enumeration of devices on a Marvell CN106XX board.

When PCI_PROBE_ONLY is not set, add PCI_REASSIGN_ALL_BUS so pci_scan_bridge_extend() takes the reassignment path: bus numbers can be assigned from firmware EA data (e.g. pci_ea_fixed_busnrs()). Skip the flag in probe-only mode so existing assignments are not overridden.

---
 drivers/pci/controller/pci-host-common.c |    4 ++++
 1 file changed, 4 insertions(+)

@@ -77,6 +77,10 @@ int pci_host_common_probe(struct platfor
 	if (!pci_has_flag(PCI_PROBE_ONLY))
 		pci_add_flags(PCI_REASSIGN_ALL_BUS);
 
+	/* Do not reassign bus numbers if probe only */
+	if (!pci_has_flag(PCI_PROBE_ONLY))
+		pci_add_flags(PCI_REASSIGN_ALL_BUS);
+
 	bridge->sysdata = cfg;
 	bridge->ops = (struct pci_ops *)&ops->pci_ops;
 	bridge->msi_domain = true;

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: SeongJae Park <[email protected]>

commit 35d4a3cf70a855b50e53189ac2f8463e20a02046 upstream.

Fix three issues in damon_hot_score() that comes from wrong handling of extreme (zero or too high) monitoring intervals user setup.

When the user sets sampling interval zero, damon_max_nr_accesses(), which is called from damon_hot_score(), causes a divide-by-zero. Needless to say, it is a problem.

When the user sets the aggregation interval zero, the function returns zero. It is wrong, since the real maximum nr_acceses in the setup should be one. Worse yet, it can cause another divide-by-zero from its caller, damon_hot_score(), since it uses damon_max_nr_accesses() return value as a denominator.

When the user sets the aggregation interval very high, damon_hot_score() could return a value out of [0, DAMOS_MAX_SCORE] range. Since the return value is used as an index to the regions_score_histogram array, which is DAMOS_MAX_SCORE+1 size, it causes out of bounds array access.

The issues can be relatively easily reproduced like below. The sysfs write permission is required, though.

    # ./damo start --damos_action lru_prio --damos_quota_space 100M \
            --damos_quota_interval 1s
    # cd /sys/kernel/mm/damon/admin/kdamonds/0
    # echo 0 > contexts/0/monitoring_attrs/intervals/sample_us
    # echo 0 > contexts/0/monitoring_attrs/intervals/aggr_us
    # echo commit > state
    # dmesg
    [...]
    [  131.329762] Oops: divide error: 0000 [#1] SMP NOPTI
    [...]
    [  131.336089] RIP: 0010:damon_hot_score+0x27/0xd0
    [...]

Fix the divide-by-zero intervals problems by explicitly handling the zero intervals in damon_max_nr_accesses(). Fix the out-of-bound array access by applying [0, DAMOS_MAX_SCORE] bounds before returning from damon_hot_score().

The issue was discovered [1] by Sashiko.

---
 include/linux/damon.h |    8 ++++++--
 mm/damon/ops-common.c |    1 +
 2 files changed, 7 insertions(+), 2 deletions(-)

@@ -669,9 +669,13 @@ static inline bool damon_target_has_pid(
 
 static inline unsigned int damon_max_nr_accesses(const struct damon_attrs *attrs)
 {
-	/* {aggr,sample}_interval are unsigned long, hence could overflow */
-	return min(attrs->aggr_interval / attrs->sample_interval,
+	unsigned long sample_interval;
+	unsigned long max_nr_accesses;
+
+	sample_interval = attrs->sample_interval ? : 1;
+	max_nr_accesses = min(attrs->aggr_interval / sample_interval,
 			(unsigned long)UINT_MAX);
+	return max_nr_accesses ? : 1;
 }
 
 

@@ -107,6 +107,7 @@ int damon_hot_score(struct damon_ctx *c,
 	 * Transform it to fit in [0, DAMOS_MAX_SCORE]
 	 */
 	hotness = hotness * DAMOS_MAX_SCORE / DAMON_MAX_SUBSCORE;
+	hotness = max(min(hotness, DAMOS_MAX_SCORE), 0);
 
 	return hotness;
 }

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Xiang Mei <[email protected]>

commit 7cd9103283b26b917360ec99d7d2f2d761bcf1ab upstream.

The release path of ip_set_dump_do() and ip_set_dump_done() read inst->ip_set_list via ip_set_ref_netlink(), a plain rcu_dereference_raw() of the array pointer. These run from netlink_recvmsg() without the nfnl mutex and without an RCU read-side critical section.

A concurrent ip_set_create() can grow the array: it publishes the new array, calls synchronize_net() and then kvfree()s the old one. Since the dump paths read the array outside any RCU reader, synchronize_net() does not wait for them and the old array can be freed while they still index into it, causing a use-after-free.

The dumped set itself stays pinned via set->ref_netlink, so only the array load needs protecting. Take rcu_read_lock() around it, matching ip_set_get_byname() and __ip_set_put_byindex().

  BUG: KASAN: slab-use-after-free in ip_set_dump_do (net/netfilter/ipset/ip_set_core.c:1697)
  Read of size 8 at addr ffff88800b5c4018 by task exploit/150
  Call Trace:
   ...
   kasan_report (mm/kasan/report.c:595)
   ip_set_dump_do (net/netfilter/ipset/ip_set_core.c:1697)
   netlink_dump (net/netlink/af_netlink.c:2325)
   netlink_recvmsg (net/netlink/af_netlink.c:1976)
   sock_recvmsg (net/socket.c:1159)
   __sys_recvfrom (net/socket.c:2315)
   ...
  Oops: general protection fault, probably for non-canonical address ... KASAN NOPTI
  KASAN: maybe wild-memory-access in range [0x02d6...d0-0x02d6...d7]
  RIP: 0010:ip_set_dump_do (net/netfilter/ipset/ip_set_core.c:1698)
  Kernel panic - not syncing: Fatal exception

---
 net/netfilter/ipset/ip_set_core.c |    8 +++++++-
 1 file changed, 7 insertions(+), 1 deletion(-)

@@ -1480,7 +1480,11 @@ ip_set_dump_done(struct netlink_callback
 		struct ip_set_net *inst =
 			(struct ip_set_net *)cb->args[IPSET_CB_NET];
 		ip_set_id_t index = (ip_set_id_t)cb->args[IPSET_CB_INDEX];
-		struct ip_set *set = ip_set_ref_netlink(inst, index);
+		struct ip_set *set;
+
+		rcu_read_lock();
+		set = ip_set_ref_netlink(inst, index);
+		rcu_read_unlock();
 
 		if (set->variant->uref)
 			set->variant->uref(set, cb, false);

@@ -1685,7 +1689,9 @@ next_set:
 release_refcount:
 	/* If there was an error or set is done, release set */
 	if (ret || !cb->args[IPSET_CB_ARG0]) {
+		rcu_read_lock();
 		set = ip_set_ref_netlink(inst, index);
+		rcu_read_unlock();
 		if (set->variant->uref)
 			set->variant->uref(set, cb, false);
 		pr_debug("release set %s\n", set->name);

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Johan Hovold <[email protected]>

commit c687bc35694698ec4c7f92bf929c3d659f0cecb8 upstream.

Driver core expects devices to be allocated dynamically and complains loudly when a device that lacks a release function is freed.

Use __root_device_register() to allocate and register the root device instead of open coding using a static device.

Note that root_device_register(), which also creates a link to the module, cannot be used as the device is registered when parsing the module parameters which happens before the module kobject has been set up.

Message-ID: <[email protected]>

---
 drivers/virtio/virtio_mmio.c |   26 +++++++++++++-------------
 1 file changed, 13 insertions(+), 13 deletions(-)

@@ -710,9 +710,7 @@ static int virtio_mmio_remove(struct pla
 
 #if defined(CONFIG_VIRTIO_MMIO_CMDLINE_DEVICES)
 
-static struct device vm_cmdline_parent = {
-	.init_name = "virtio-mmio-cmdline",
-};
+static struct device *vm_cmdline_parent;
 
 static int vm_cmdline_parent_registered;
 static int vm_cmdline_id;

@@ -720,7 +718,6 @@ static int vm_cmdline_id;
 static int vm_cmdline_set(const char *device,
 		const struct kernel_param *kp)
 {
-	int err;
 	struct resource resources[2] = {};
 	char *str;
 	long long base, size;

@@ -752,11 +749,10 @@ static int vm_cmdline_set(const char *de
 	resources[1].start = resources[1].end = irq;
 
 	if (!vm_cmdline_parent_registered) {
-		err = device_register(&vm_cmdline_parent);
-		if (err) {
-			put_device(&vm_cmdline_parent);
+		vm_cmdline_parent = __root_device_register("virtio-mmio-cmdline", NULL);
+		if (IS_ERR(vm_cmdline_parent)) {
 			pr_err("Failed to register parent device!\n");
-			return err;
+			return PTR_ERR(vm_cmdline_parent);
 		}
 		vm_cmdline_parent_registered = 1;
 	}

@@ -767,7 +763,7 @@ static int vm_cmdline_set(const char *de
 		       (unsigned long long)resources[0].end,
 		       (int)resources[1].start);
 
-	pdev = platform_device_register_resndata(&vm_cmdline_parent,
+	pdev = platform_device_register_resndata(vm_cmdline_parent,
 			"virtio-mmio", vm_cmdline_id++,
 			resources, ARRAY_SIZE(resources), NULL, 0);
 

@@ -791,8 +787,12 @@ static int vm_cmdline_get_device(struct
 static int vm_cmdline_get(char *buffer, const struct kernel_param *kp)
 {
 	buffer[0] = '\0';
-	device_for_each_child(&vm_cmdline_parent, buffer,
-			vm_cmdline_get_device);
+
+	if (vm_cmdline_parent_registered) {
+		device_for_each_child(vm_cmdline_parent, buffer,
+				vm_cmdline_get_device);
+	}
+
 	return strlen(buffer) + 1;
 }
 

@@ -814,9 +814,9 @@ static int vm_unregister_cmdline_device(
 static void vm_unregister_cmdline_devices(void)
 {
 	if (vm_cmdline_parent_registered) {
-		device_for_each_child(&vm_cmdline_parent, NULL,
+		device_for_each_child(vm_cmdline_parent, NULL,
 				vm_unregister_cmdline_device);
-		device_unregister(&vm_cmdline_parent);
+		root_device_unregister(vm_cmdline_parent);
 		vm_cmdline_parent_registered = 0;
 	}
 }

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Michael Bommarito <[email protected]>

commit e3046eeada299f917a8ad883af4434bfb86556b1 upstream.

random_recv_done() stores the device-reported used.len directly into vi->data_avail. copy_data() then indexes vi->data[] using vi->data_idx (advanced by previous copy_data() calls) and issues a memcpy() without re-validating either value against the posted buffer size sizeof(vi->data) (SMP_CACHE_BYTES bytes, typically 32 or 64).

A malicious or buggy virtio-rng backend can set used.len beyond sizeof(vi->data), steering the memcpy() past the end of the inline array into adjacent kmalloc-1k slab bytes. hwrng_fillfn() mixes those bytes into the guest RNG, and guest root can also observe them directly via /dev/hwrng.

Concrete impact is inside the guest:

- Memory-safety / hardening: any virtio-rng backend that over-reports used.len causes the driver to read past vi->data into unrelated slab contents. hwrng_fillfn() is a kernel thread that runs as soon as the device is probed; no guest userspace interaction is required to first-trigger the OOB.

- Cross-boundary leak (confidential-compute threat model): a malicious hypervisor cooperating with a malicious or compromised guest root userspace can use /dev/hwrng as a leak channel for guest-kernel heap data. The host sets a large used.len, guest root reads /dev/hwrng, and the returned bytes contain guest kernel slab contents that were adjacent to vi->data. In practice, confidential-compute guests (SEV-SNP, TDX) usually disable virtio-rng entirely, so this path is narrow, but the fix is still worth carrying because the underlying memory-safety bug contaminates the guest RNG on any host.

KASAN confirms the OOB on a 7.1-rc4 guest whose virtio-rng backend has been patched to report used.len = 0x10000:

  BUG: KASAN: slab-out-of-bounds in virtio_read+0x394/0x5d0
  Read of size 64 at addr ffff88800ae0ba20 by task hwrng/52
  Call Trace:
   __asan_memcpy+0x23/0x60
   virtio_read+0x394/0x5d0
   hwrng_fillfn+0xb2/0x470
   kthread+0x2cc/0x3a0
  Allocated by task 1:
   probe_common+0xa5/0x660
   virtio_dev_probe+0x549/0xbc0
  The buggy address belongs to the object at ffff88800ae0b800
   which belongs to the cache kmalloc-1k of size 1024
  The buggy address is located 0 bytes to the right of
   allocated 544-byte region [ffff88800ae0b800, ffff88800ae0ba20)

Same class of bug as commit c04db81cd028 ("net/9p: Fix buffer overflow in USB transport layer"), which hardened usb9pfs_rx_complete() against unchecked device-reported length in the USB 9p transport.

With the clamp at point of use and array_index_nospec() in place, the same harness boots cleanly: copy_data() returns zero for the bogus report, the device-supplied bytes after data_idx are discarded, and the driver issues a fresh request.

Message-ID: <[email protected]>

---
 drivers/char/hw_random/virtio-rng.c |   23 +++++++++++++++++++++--
 1 file changed, 21 insertions(+), 2 deletions(-)

@@ -7,6 +7,7 @@
 #include <asm/barrier.h>
 #include <linux/err.h>
 #include <linux/hw_random.h>
+#include <linux/nospec.h>
 #include <linux/scatterlist.h>
 #include <linux/spinlock.h>
 #include <linux/virtio.h>

@@ -66,8 +67,26 @@ static void request_entropy(struct virtr
 static unsigned int copy_data(struct virtrng_info *vi, void *buf,
 			      unsigned int size)
 {
-	size = min_t(unsigned int, size, vi->data_avail);
-	memcpy(buf, vi->data + vi->data_idx, size);
+	unsigned int idx, avail;
+
+	/*
+	 * vi->data_avail was set from the device-reported used.len and
+	 * vi->data_idx was advanced by previous copy_data() calls.  A
+	 * malicious or buggy virtio-rng backend can drive either past
+	 * sizeof(vi->data).  Clamp at point of use and harden the index
+	 * with array_index_nospec() so the memcpy() below cannot be
+	 * steered into adjacent slab memory, including under
+	 * speculation.
+	 */
+	avail = min_t(unsigned int, vi->data_avail, sizeof(vi->data));
+	if (vi->data_idx >= avail) {
+		vi->data_avail = 0;
+		request_entropy(vi);
+		return 0;
+	}
+	size = min_t(unsigned int, size, avail - vi->data_idx);
+	idx = array_index_nospec(vi->data_idx, sizeof(vi->data));
+	memcpy(buf, vi->data + idx, size);
 	vi->data_idx += size;
 	vi->data_avail -= size;
 	if (vi->data_avail == 0)

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Alan Stern <[email protected]>

commit abf76d3239dee97b66e7241ad04811f1ce562e28 upstream.

The chaoskey driver has a use-after-free bug in its release routine. If the user closes the device file after the USB device has been unplugged, a debugging log statement will try to access the usb_interface structure after it has been deallocated:

	BUG: KASAN: slab-use-after-free in dev_driver_string (drivers/base/core.c:2406)
	Read of size 8 at addr ffff888168e8a0b8 by task chaoskey_raw_re/10106

	Hardware name: QEMU Ubuntu 24.04 PC v2 (i440FX + PIIX, arch_caps fix, 1996), BIOS 1.16.3-debian-1.16.3-2 04/01/2014
	Call Trace:
	 <TASK>
	 dump_stack_lvl (lib/dump_stack.c:94 lib/dump_stack.c:120)
	 print_report (mm/kasan/report.c:378 mm/kasan/report.c:482)
	 kasan_report (mm/kasan/report.c:595)
	 dev_driver_string (drivers/base/core.c:2406)
	 __dynamic_dev_dbg (lib/dynamic_debug.c:906)
	 chaoskey_release (drivers/usb/misc/chaoskey.c:323)
	 __fput (fs/file_table.c:510)
	 fput_close_sync (fs/file_table.c:615)
	 __x64_sys_close (fs/open.c:1507 fs/open.c:1492 fs/open.c:1492)
	 do_syscall_64 (arch/x86/entry/syscall_64.c:63 arch/x86/entry/syscall_64.c:94)
	 entry_SYSCALL_64_after_hwframe (arch/x86/entry/entry_64.S:121)

The driver's last reference to the interface structure is dropped in the chaoskey_free() routine, so the code must not use the interface -- even in a debugging statement -- after that routine returns. (Exception: If we know that another reference is held by someone else, such as the device core while the disconnect routine runs, there's no problem. Thanks to Johan Hovold for pointing this out.)

Since the bad access is part of an unimportant debugging statement, we can fix the problem simply by removing the whole statement.

---
 drivers/usb/misc/chaoskey.c |    1 -
 1 file changed, 1 deletion(-)

@@ -320,7 +320,6 @@ bail:
 	mutex_unlock(&dev->lock);
 destruction:
 	mutex_unlock(&chaoskey_list_lock);
-	usb_dbg(interface, "release success");
 	return rv;
 }
 

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Runyu Xiao <[email protected]>

commit 010382937fb69892b3469ac4d30af072262f59e8 upstream.

dwc3_gadget_suspend() takes dwc->lock with IRQs disabled and then calls dwc3_disconnect_gadget(). For async callbacks that helper only uses plain spin_unlock()/spin_lock(), so the gadget ->disconnect() callback still runs with IRQs disabled and any sleepable callback trips Lockdep.

This issue was found by our static analysis tool and then manually reviewed against the current tree.

The grounded PoC kept the dwc3_gadget_suspend() -> dwc3_disconnect_gadget() -> gadget_driver->disconnect() chain, and Lockdep reported:

  BUG: sleeping function called from invalid context
  gadget_disconnect+0x21/0x39 [vuln_msv]
  dwc3_gadget_suspend.constprop.0+0x2b/0x42 [vuln_msv]

Keep the disconnect callback selection in one common helper, but add a sleepable suspend-side wrapper which snapshots the callback under dwc->lock and then runs it after spin_unlock_irqrestore(). The regular event path still uses the existing spin_unlock()/spin_lock() window.

---
 drivers/usb/dwc3/gadget.c |   43 ++++++++++++++++++++++++++++++++++++-------
 1 file changed, 36 insertions(+), 7 deletions(-)

@@ -3930,15 +3930,48 @@ static void dwc3_endpoint_interrupt(stru
 	}
 }
 
+static bool dwc3_prepare_disconnect_gadget(struct dwc3 *dwc,
+					   struct usb_gadget_driver **driver,
+					   struct usb_gadget **gadget)
+{
+	if (!dwc->async_callbacks || !dwc->gadget_driver ||
+	    !dwc->gadget_driver->disconnect)
+		return false;
+
+	*driver = dwc->gadget_driver;
+	*gadget = dwc->gadget;
+
+	return true;
+}
+
 static void dwc3_disconnect_gadget(struct dwc3 *dwc)
 {
-	if (dwc->async_callbacks && dwc->gadget_driver->disconnect) {
+	struct usb_gadget_driver *driver;
+	struct usb_gadget *gadget;
+
+	if (dwc3_prepare_disconnect_gadget(dwc, &driver, &gadget)) {
 		spin_unlock(&dwc->lock);
-		dwc->gadget_driver->disconnect(dwc->gadget);
+		driver->disconnect(gadget);
 		spin_lock(&dwc->lock);
 	}
 }
 
+static void dwc3_disconnect_gadget_sleepable(struct dwc3 *dwc)
+{
+	struct usb_gadget_driver *driver;
+	struct usb_gadget *gadget;
+	unsigned long flags;
+
+	spin_lock_irqsave(&dwc->lock, flags);
+	if (!dwc3_prepare_disconnect_gadget(dwc, &driver, &gadget)) {
+		spin_unlock_irqrestore(&dwc->lock, flags);
+		return;
+	}
+
+	spin_unlock_irqrestore(&dwc->lock, flags);
+	driver->disconnect(gadget);
+}
+
 static void dwc3_suspend_gadget(struct dwc3 *dwc)
 {
 	if (dwc->async_callbacks && dwc->gadget_driver->suspend) {

@@ -4818,7 +4851,6 @@ void dwc3_gadget_exit(struct dwc3 *dwc)
 
 int dwc3_gadget_suspend(struct dwc3 *dwc)
 {
-	unsigned long flags;
 	int ret;
 
 	ret = dwc3_gadget_soft_disconnect(dwc);

@@ -4832,10 +4864,7 @@ int dwc3_gadget_suspend(struct dwc3 *dwc
 		return -EAGAIN;
 	}
 
-	spin_lock_irqsave(&dwc->lock, flags);
-	if (dwc->gadget_driver)
-		dwc3_disconnect_gadget(dwc);
-	spin_unlock_irqrestore(&dwc->lock, flags);
+	dwc3_disconnect_gadget_sleepable(dwc);
 
 	return 0;
 }

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Yizhou Zhao <[email protected]>

commit 1720db928e5a58ca7d75ac1d514c3b73fd7061a7 upstream.

lowpan_nhc_do_uncompression() looks up an NHC descriptor while holding lowpan_nhc_lock. If the descriptor has no uncompress callback, the error path drops the lock before printing nhc->name.

lowpan_nhc_del() removes descriptors under the same lock and then relies on synchronize_net() before the owning module can be unloaded. That only waits for net RX RCU readers. lowpan_header_decompress() is also exported and can be reached from callers that are not necessarily covered by the net core RX critical section, for example the Bluetooth 6LoWPAN L2CAP receive path.

This leaves a race where one task drops lowpan_nhc_lock in the error path, another task unregisters and frees the matching descriptor after synchronize_net() returns, and the first task then dereferences nhc->name for the warning.

With the post-unlock window widened, KASAN reports:

  BUG: KASAN: slab-use-after-free in lowpan_nhc_do_uncompression+0x1f4/0x220
  Read of size 8
  lowpan_nhc_do_uncompression
  lowpan_header_decompress

Fix this by printing the warning before dropping lowpan_nhc_lock, so the descriptor name is read while unregister is still excluded. The malformed packet is still rejected with -ENOTSUPP.

---
 net/6lowpan/nhc.c |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

@@ -117,9 +117,9 @@ int lowpan_nhc_do_uncompression(struct s
 				return ret;
 			}
 		} else {
-			spin_unlock_bh(&lowpan_nhc_lock);
 			netdev_warn(dev, "received nhc id for %s which is not implemented.\n",
 				    nhc->name);
+			spin_unlock_bh(&lowpan_nhc_lock);
 			return -ENOTSUPP;
 		}
 	} else {

6.6-stable review patch. If anyone has any objections, please let me know.

------------------

From: Samuel Page <[email protected]>

commit 2b66974a1b6134a4bbc3bfed181f7418f688eb54 upstream.

A broadcast PROTOCOL/STATE_MSG can carry a Gap ACK blocks record in its data area. tipc_get_gap_ack_blks() only verifies that the record's len field is self-consistent with its ugack_cnt/bgack_cnt counts (sz == struct_size(p, gacks, ugack_cnt + bgack_cnt)); it does not check that the record actually fits in the message data area, msg_data_sz().

The unicast caller tipc_link_proto_rcv() bounds it ("if (glen > dlen) break;"), but the broadcast caller tipc_bcast_sync_rcv() discards the returned size, so tipc_link_advance_transmq() copies the record off the receive skb with an attacker-controlled count:

	this_ga = kmemdup(ga, struct_size(ga, gacks, ga->bgack_cnt),
			  GFP_ATOMIC);

A TIPC neighbour that negotiated TIPC_GAP_ACK_BLOCK triggers it with one ordinary broadcast STATE_MSG (msg_bc_ack_invalid() clear), sized so its data area is short, carrying a Gap ACK record with len = 0x400, bgack_cnt = 0xff and ugack_cnt = 0. len then equals struct_size(p, gacks, 255), so the consistency check passes and ga is non-NULL; kmemdup() reads struct_size(ga, gacks, 255) = 1024 bytes out of the much smaller skb:

  BUG: KASAN: slab-out-of-bounds in kmemdup_noprof+0x48/0x60
  Read of size 1024 at addr ffff0000c7030d38 by task poc864/69
  Call trace:
   kmemdup_noprof+0x48/0x60
   tipc_link_advance_transmq+0x86c/0xb80
   tipc_link_bc_ack_rcv+0x19c/0x1e0
   tipc_bcast_sync_rcv+0x1c4/0x2c4
   tipc_rcv+0x85c/0x1340
   tipc_l2_rcv_msg+0xac/0x104
  The buggy address belongs to the object at ffff0000c7030d00
   which belongs to the cache skbuff_small_head of size 704
  The buggy address is located 56 bytes inside of
   allocated 704-byte region [ffff0000c7030d00, ffff0000c7030fc0)

The copied-out bytes are subsequently consumed as gap/ack values, but the read is already out of bounds at the kmemdup() regardless of how they are used.

The unicast STATE path drops such a message: "if (glen > dlen) break;" skips the rest of STATE_MSG handling and the skb is freed. Make the broadcast path drop it too. tipc_bcast_sync_rcv() now bounds the record against msg_data_sz() and, when it does not fit, reports it back through tipc_node_bc_sync_rcv() to tipc_rcv() so the skb is discarded rather than processed. ga is not cleared on this path: ga == NULL already means "legacy peer without Selective ACK", a distinct legitimate state.

---
 net/tipc/bcast.c |   22 ++++++++++++++--------
 net/tipc/bcast.h |    2 +-
 net/tipc/node.c  |   15 ++++++++++++---
 3 files changed, 27 insertions(+), 12 deletions(-)

@@ -497,12 +497,13 @@ void tipc_bcast_ack_rcv(struct net *net,
  */
 int tipc_bcast_sync_rcv(struct net *net, struct tipc_link *l,
 			struct tipc_msg *hdr,
-			struct sk_buff_head *retrq)
+			struct sk_buff_head *retrq, bool *valid)
 {
 	struct sk_buff_head *inputq = &tipc_bc_base(net)->inputq;
 	struct tipc_gap_ack_blks *ga;
 	struct sk_buff_head xmitq;
 	int rc = 0;
+	u16 glen;
 
 	__skb_queue_head_init(&xmitq);
 

@@ -510,13 +511,18 @@ int tipc_bcast_sync_rcv(struct net *net,
 	if (msg_type(hdr) != STATE_MSG) {
 		tipc_link_bc_init_rcv(l, hdr);
 	} else if (!msg_bc_ack_invalid(hdr)) {
-		tipc_get_gap_ack_blks(&ga, l, hdr, false);
-		if (!sysctl_tipc_bc_retruni)
-			retrq = &xmitq;
-		rc = tipc_link_bc_ack_rcv(l, msg_bcast_ack(hdr),
-					  msg_bc_gap(hdr), ga, &xmitq,
-					  retrq);
-		rc |= tipc_link_bc_sync_rcv(l, hdr, &xmitq);
+		glen = tipc_get_gap_ack_blks(&ga, l, hdr, false);
+		if (glen > msg_data_sz(hdr)) {
+			/* Malformed Gap ACK blocks; caller drops the msg */
+			*valid = false;
+		} else {
+			if (!sysctl_tipc_bc_retruni)
+				retrq = &xmitq;
+			rc = tipc_link_bc_ack_rcv(l, msg_bcast_ack(hdr),
+						  msg_bc_gap(hdr), ga, &xmitq,
+						  retrq);
+			rc |= tipc_link_bc_sync_rcv(l, hdr, &xmitq);
+		}
 	}
 	tipc_bcast_unlock(net);
 

@@ -97,7 +97,7 @@ void tipc_bcast_ack_rcv(struct net *net,
 			struct tipc_msg *hdr);
 int tipc_bcast_sync_rcv(struct net *net, struct tipc_link *l,
 			struct tipc_msg *hdr,
-			struct sk_buff_head *retrq);
+			struct sk_buff_head *retrq, bool *valid);
 int tipc_nl_add_bc_link(struct net *net, struct tipc_nl_msg *msg,
 			struct tipc_link *bcl);
 int tipc_nl_bc_link_set(struct net *net, struct nlattr *attrs[]);

@@ -1833,12 +1833,15 @@ static void tipc_node_mcast_rcv(struct t
 }
 
 static void tipc_node_bc_sync_rcv(struct tipc_node *n, struct tipc_msg *hdr,
-				  int bearer_id, struct sk_buff_head *xmitq)
+				  int bearer_id, struct sk_buff_head *xmitq,
+				  bool *valid)
 {
 	struct tipc_link *ucl;
 	int rc;
 
-	rc = tipc_bcast_sync_rcv(n->net, n->bc_entry.link, hdr, xmitq);
+	rc = tipc_bcast_sync_rcv(n->net, n->bc_entry.link, hdr, xmitq, valid);
+	if (!*valid)
+		return;
 
 	if (rc & TIPC_LINK_DOWN_EVT) {
 		tipc_node_reset_links(n);

@@ -2142,12 +2145,18 @@ rcv:
 
 	/* Ensure broadcast reception is in synch with peer's send state */
 	if (unlikely(usr == LINK_PROTOCOL)) {
+		bool valid = true;
+
 		if (unlikely(skb_linearize(skb))) {
 			tipc_node_put(n);
 			goto discard;
 		}
 		hdr = buf_msg(skb);
-		tipc_node_bc_sync_rcv(n, hdr, bearer_id, &xmitq);
+		tipc_node_bc_sync_rcv(n, hdr, bearer_id, &xmitq, &valid);
+		if (!valid) {
+			tipc_node_put(n);
+			goto discard;
+		}
 	} else if (unlikely(tipc_link_acked(n->bc_entry.link) != bc_ack)) {
 		tipc_bcast_ack_rcv(net, n->bc_entry.link, hdr);
 	}