* [PATCH] iio: adc: ad_sigma_delta: fix use-after-free on unbind
@ 2026-09-23 9:48 Fan Wu
2026-09-24 8:45 ` Nuno Sá
0 siblings, 1 reply; 2+ messages in thread
From: Fan Wu @ 2026-09-23 9:48 UTC (permalink / raw)
To: Jonathan Cameron, David Lechner, Nuno Sá, Andy Shevchenko
Cc: linux-iio, linux-kernel, stable, Song Li, Fan Wu
ad_sd_buffer_postenable() allocates sigma_delta->samples_buf with
devm_krealloc() at runtime, so its devres entry sits after all
probe-time entries of the driver. devm resources are released in
reverse allocation order, which means unbind frees samples_buf before
iio_device_unregister() disables the buffers and detaches the trigger
pollfunc. The data ready IRQ is still enabled at that point, so
ad_sd_trigger_handler() can still run and memcpy() incoming samples
into the freed samples_buf.
Fix this by preallocating the buffer in
devm_ad_sd_setup_buffer_and_trigger(), before the triggered buffer and
the IRQ are set up, so it is freed only after iio_device_unregister()
has drained the trigger handler via free_irq(). Size it for the worst
case of all sequencer slots being active; ad_sd_validate_scan_mask()
already caps the number of active channels at num_slots.
This issue was found by an in-house static analysis tool.
Fixes: 8bea9af887de ("iio: adc: ad_sigma_delta: Add sequencer support")
Cc: stable@vger.kernel.org
Co-developed-by: Song Li <songl@zju.edu.cn>
Signed-off-by: Song Li <songl@zju.edu.cn>
Signed-off-by: Fan Wu <fanwu01@zju.edu.cn>
---
drivers/iio/adc/ad_sigma_delta.c | 32 +++++++++++++++++++-------------
1 file changed, 19 insertions(+), 13 deletions(-)
diff --git a/drivers/iio/adc/ad_sigma_delta.c b/drivers/iio/adc/ad_sigma_delta.c
index 1b41029..4f982c6 100644
--- a/drivers/iio/adc/ad_sigma_delta.c
+++ b/drivers/iio/adc/ad_sigma_delta.c
@@ -498,7 +498,6 @@ static int ad_sd_buffer_postenable(struct iio_dev *indio_dev)
const struct iio_scan_type *scan_type = &indio_dev->channels[0].scan_type;
struct spi_transfer *xfer = sigma_delta->sample_xfer;
unsigned int i, slot, channel;
- u8 *samples_buf;
int ret;
if (sigma_delta->num_slots == 1) {
@@ -530,7 +529,7 @@ static int ad_sd_buffer_postenable(struct iio_dev *indio_dev)
xfer[1].bits_per_word = scan_type->realbits;
xfer[1].len = spi_bpw_to_bytes(scan_type->realbits);
} else {
- unsigned int samples_buf_size, scan_size;
+ unsigned int scan_size;
if (sigma_delta->active_slots > 1) {
ret = ad_sigma_delta_append_status(sigma_delta, true);
@@ -538,17 +537,6 @@ static int ad_sd_buffer_postenable(struct iio_dev *indio_dev)
return ret;
}
- samples_buf_size =
- ALIGN(slot * BITS_TO_BYTES(scan_type->storagebits),
- sizeof(s64));
- samples_buf_size += sizeof(s64);
- samples_buf = devm_krealloc(&sigma_delta->spi->dev,
- sigma_delta->samples_buf,
- samples_buf_size, GFP_KERNEL);
- if (!samples_buf)
- return -ENOMEM;
-
- sigma_delta->samples_buf = samples_buf;
scan_size = BITS_TO_BYTES(scan_type->realbits + scan_type->shift);
/* For 24-bit data, there is an extra byte of padding. */
xfer[1].rx_buf = &sigma_delta->rx_buf[scan_size == 3 ? 1 : 0];
@@ -855,6 +843,24 @@ int devm_ad_sd_setup_buffer_and_trigger(struct device *dev, struct iio_dev *indi
indio_dev->setup_ops = &ad_sd_buffer_setup_ops;
} else {
+ const struct iio_scan_type *scan_type =
+ &indio_dev->channels[0].scan_type;
+ unsigned int samples_buf_size;
+
+ /*
+ * Worst-case size: all sequencer slots can be active, capped
+ * at num_slots by ad_sd_validate_scan_mask().
+ */
+ samples_buf_size =
+ ALIGN(sigma_delta->num_slots *
+ BITS_TO_BYTES(scan_type->storagebits),
+ sizeof(s64));
+ samples_buf_size += sizeof(s64);
+ sigma_delta->samples_buf =
+ devm_kzalloc(dev, samples_buf_size, GFP_KERNEL);
+ if (!sigma_delta->samples_buf)
+ return -ENOMEM;
+
ret = devm_iio_triggered_buffer_setup(dev, indio_dev,
&iio_pollfunc_store_time,
&ad_sd_trigger_handler,
^ permalink raw reply [flat|nested] 2+ messages in thread
* Re: [PATCH] iio: adc: ad_sigma_delta: fix use-after-free on unbind
2026-09-23 9:48 [PATCH] iio: adc: ad_sigma_delta: fix use-after-free on unbind Fan Wu
@ 2026-09-24 8:45 ` Nuno Sá
0 siblings, 0 replies; 2+ messages in thread
From: Nuno Sá @ 2026-09-24 8:45 UTC (permalink / raw)
To: Fan Wu
Cc: Jonathan Cameron, David Lechner, Andy Shevchenko, linux-iio,
linux-kernel, stable, Song Li
On Wed, Sep 23, 2026 at 09:48:07AM +0000, Fan Wu wrote:
> ad_sd_buffer_postenable() allocates sigma_delta->samples_buf with
> devm_krealloc() at runtime, so its devres entry sits after all
> probe-time entries of the driver. devm resources are released in
> reverse allocation order, which means unbind frees samples_buf before
> iio_device_unregister() disables the buffers and detaches the trigger
> pollfunc. The data ready IRQ is still enabled at that point, so
> ad_sd_trigger_handler() can still run and memcpy() incoming samples
> into the freed samples_buf.
>
> Fix this by preallocating the buffer in
> devm_ad_sd_setup_buffer_and_trigger(), before the triggered buffer and
> the IRQ are set up, so it is freed only after iio_device_unregister()
> has drained the trigger handler via free_irq(). Size it for the worst
> case of all sequencer slots being active; ad_sd_validate_scan_mask()
> already caps the number of active channels at num_slots.
>
> This issue was found by an in-house static analysis tool.
>
> Fixes: 8bea9af887de ("iio: adc: ad_sigma_delta: Add sequencer support")
> Cc: stable@vger.kernel.org
> Co-developed-by: Song Li <songl@zju.edu.cn>
> Signed-off-by: Song Li <songl@zju.edu.cn>
> Signed-off-by: Fan Wu <fanwu01@zju.edu.cn>
> ---
Makes sense to me! One minor nit Jonathan might be ale to tweak when
applying. With that:
Reviewed-by: Nuno Sá <nuno.sa@analog.com>
> drivers/iio/adc/ad_sigma_delta.c | 32 +++++++++++++++++++-------------
> 1 file changed, 19 insertions(+), 13 deletions(-)
>
> diff --git a/drivers/iio/adc/ad_sigma_delta.c b/drivers/iio/adc/ad_sigma_delta.c
> index 1b41029..4f982c6 100644
> --- a/drivers/iio/adc/ad_sigma_delta.c
> +++ b/drivers/iio/adc/ad_sigma_delta.c
> @@ -498,7 +498,6 @@ static int ad_sd_buffer_postenable(struct iio_dev *indio_dev)
> const struct iio_scan_type *scan_type = &indio_dev->channels[0].scan_type;
> struct spi_transfer *xfer = sigma_delta->sample_xfer;
> unsigned int i, slot, channel;
> - u8 *samples_buf;
> int ret;
>
> if (sigma_delta->num_slots == 1) {
> @@ -530,7 +529,7 @@ static int ad_sd_buffer_postenable(struct iio_dev *indio_dev)
> xfer[1].bits_per_word = scan_type->realbits;
> xfer[1].len = spi_bpw_to_bytes(scan_type->realbits);
> } else {
> - unsigned int samples_buf_size, scan_size;
> + unsigned int scan_size;
>
> if (sigma_delta->active_slots > 1) {
> ret = ad_sigma_delta_append_status(sigma_delta, true);
> @@ -538,17 +537,6 @@ static int ad_sd_buffer_postenable(struct iio_dev *indio_dev)
> return ret;
> }
>
> - samples_buf_size =
> - ALIGN(slot * BITS_TO_BYTES(scan_type->storagebits),
> - sizeof(s64));
> - samples_buf_size += sizeof(s64);
> - samples_buf = devm_krealloc(&sigma_delta->spi->dev,
> - sigma_delta->samples_buf,
> - samples_buf_size, GFP_KERNEL);
> - if (!samples_buf)
> - return -ENOMEM;
> -
> - sigma_delta->samples_buf = samples_buf;
> scan_size = BITS_TO_BYTES(scan_type->realbits + scan_type->shift);
> /* For 24-bit data, there is an extra byte of padding. */
> xfer[1].rx_buf = &sigma_delta->rx_buf[scan_size == 3 ? 1 : 0];
> @@ -855,6 +843,24 @@ int devm_ad_sd_setup_buffer_and_trigger(struct device *dev, struct iio_dev *indi
>
> indio_dev->setup_ops = &ad_sd_buffer_setup_ops;
> } else {
> + const struct iio_scan_type *scan_type =
> + &indio_dev->channels[0].scan_type;
> + unsigned int samples_buf_size;
> +
> + /*
> + * Worst-case size: all sequencer slots can be active, capped
> + * at num_slots by ad_sd_validate_scan_mask().
> + */
> + samples_buf_size =
> + ALIGN(sigma_delta->num_slots *
> + BITS_TO_BYTES(scan_type->storagebits),
> + sizeof(s64));
> + samples_buf_size += sizeof(s64);
> + sigma_delta->samples_buf =
> + devm_kzalloc(dev, samples_buf_size, GFP_KERNEL);
Why the line break for devm_kzalloc(). Keep it in the same line please.
Tbh same for for the ALIGN() call but at least that one is coherent with
the original code.
- Nuno Sá
> + if (!sigma_delta->samples_buf)
> + return -ENOMEM;
> +
> ret = devm_iio_triggered_buffer_setup(dev, indio_dev,
> &iio_pollfunc_store_time,
> &ad_sd_trigger_handler,
>
^ permalink raw reply [flat|nested] 2+ messages in thread
end of thread, other threads:[~2026-09-24 8:44 UTC | newest]
Thread overview: 2+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2026-09-23 9:48 [PATCH] iio: adc: ad_sigma_delta: fix use-after-free on unbind Fan Wu
2026-09-24 8:45 ` Nuno Sá
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox
all inboxes | Powered by JetHome®