* [PATCH 2/3] Input: atmel_mxt_ts - check mem_size before calculating config memory size
2026-05-04 18:54 [PATCH 1/3] Input: atmel_mxt_ts - fix boundary check in mxt_prepare_cfg_mem Dmitry Torokhov
@ 2026-05-04 18:54 ` Dmitry Torokhov
2026-05-04 22:59 ` Dmitry Torokhov
2026-05-04 18:54 ` [PATCH 3/3] Input: atmel_mxt_ts - use __free() for obuf in mxt_object_show Dmitry Torokhov
2026-05-05 9:08 ` [PATCH 1/3] Input: atmel_mxt_ts - fix boundary check in mxt_prepare_cfg_mem Ricardo Ribalda
2 siblings, 1 reply; 6+ messages in thread
From: Dmitry Torokhov @ 2026-05-04 18:54 UTC (permalink / raw)
To: Nick Dyer, linux-input; +Cc: Ricardo Ribalda, linux-kernel
In mxt_update_cfg(), the driver calculates the memory size needed to store
the configuration as data->mem_size - cfg.start_ofs. If data->mem_size is
less than or equal to cfg.start_ofs, this calculation will underflow or
result in a zero-size buffer, neither of which is valid for a configuration
update.
Add a check to return -EINVAL if data->mem_size is too small. While at it,
change the types of start_ofs and mem_size in struct mxt_cfg to u16 to
match the device address space.
Assisted-by: Gemini:gemini-3.1-pro
Signed-off-by: Dmitry Torokhov <dmitry.torokhov@gmail.com>
---
drivers/input/touchscreen/atmel_mxt_ts.c | 11 +++++++++--
1 file changed, 9 insertions(+), 2 deletions(-)
diff --git a/drivers/input/touchscreen/atmel_mxt_ts.c b/drivers/input/touchscreen/atmel_mxt_ts.c
index 28b2bd889c70..d660cc5b5fe3 100644
--- a/drivers/input/touchscreen/atmel_mxt_ts.c
+++ b/drivers/input/touchscreen/atmel_mxt_ts.c
@@ -275,8 +275,8 @@ struct mxt_cfg {
off_t raw_pos;
u8 *mem;
- size_t mem_size;
- int start_ofs;
+ u16 mem_size;
+ u16 start_ofs;
struct mxt_info info;
};
@@ -1657,6 +1657,13 @@ static int mxt_update_cfg(struct mxt_data *data, const struct firmware *fw)
cfg.start_ofs = MXT_OBJECT_START +
data->info->object_num * sizeof(struct mxt_object) +
MXT_INFO_CHECKSUM_SIZE;
+
+ if (data->mem_size < cfg.start_ofs) {
+ dev_err(dev, "Memory size too small: %u < %u\n",
+ data->mem_size, cfg.start_ofs);
+ return -EINVAL;
+ }
+
cfg.mem_size = data->mem_size - cfg.start_ofs;
u8 *mem_buf __free(kfree) = cfg.mem = kzalloc(cfg.mem_size, GFP_KERNEL);
--
2.54.0.545.g6539524ca2-goog
^ permalink raw reply related [flat|nested] 6+ messages in thread* Re: [PATCH 2/3] Input: atmel_mxt_ts - check mem_size before calculating config memory size
2026-05-04 18:54 ` [PATCH 2/3] Input: atmel_mxt_ts - check mem_size before calculating config memory size Dmitry Torokhov
@ 2026-05-04 22:59 ` Dmitry Torokhov
0 siblings, 0 replies; 6+ messages in thread
From: Dmitry Torokhov @ 2026-05-04 22:59 UTC (permalink / raw)
To: Nick Dyer, linux-input; +Cc: Ricardo Ribalda, linux-kernel
On Mon, May 04, 2026 at 11:54:46AM -0700, Dmitry Torokhov wrote:
> In mxt_update_cfg(), the driver calculates the memory size needed to store
> the configuration as data->mem_size - cfg.start_ofs. If data->mem_size is
> less than or equal to cfg.start_ofs, this calculation will underflow or
> result in a zero-size buffer, neither of which is valid for a configuration
> update.
>
> Add a check to return -EINVAL if data->mem_size is too small. While at it,
> change the types of start_ofs and mem_size in struct mxt_cfg to u16 to
> match the device address space.
>
> Assisted-by: Gemini:gemini-3.1-pro
> Signed-off-by: Dmitry Torokhov <dmitry.torokhov@gmail.com>
> ---
> drivers/input/touchscreen/atmel_mxt_ts.c | 11 +++++++++--
> 1 file changed, 9 insertions(+), 2 deletions(-)
>
> diff --git a/drivers/input/touchscreen/atmel_mxt_ts.c b/drivers/input/touchscreen/atmel_mxt_ts.c
> index 28b2bd889c70..d660cc5b5fe3 100644
> --- a/drivers/input/touchscreen/atmel_mxt_ts.c
> +++ b/drivers/input/touchscreen/atmel_mxt_ts.c
> @@ -275,8 +275,8 @@ struct mxt_cfg {
> off_t raw_pos;
>
> u8 *mem;
> - size_t mem_size;
> - int start_ofs;
> + u16 mem_size;
> + u16 start_ofs;
>
> struct mxt_info info;
> };
> @@ -1657,6 +1657,13 @@ static int mxt_update_cfg(struct mxt_data *data, const struct firmware *fw)
> cfg.start_ofs = MXT_OBJECT_START +
> data->info->object_num * sizeof(struct mxt_object) +
> MXT_INFO_CHECKSUM_SIZE;
> +
> + if (data->mem_size < cfg.start_ofs) {
This is supposed to be "<=", like the commit message says.
> + dev_err(dev, "Memory size too small: %u < %u\n",
> + data->mem_size, cfg.start_ofs);
> + return -EINVAL;
> + }
> +
> cfg.mem_size = data->mem_size - cfg.start_ofs;
>
> u8 *mem_buf __free(kfree) = cfg.mem = kzalloc(cfg.mem_size, GFP_KERNEL);
--
Dmitry
^ permalink raw reply [flat|nested] 6+ messages in thread
* [PATCH 3/3] Input: atmel_mxt_ts - use __free() for obuf in mxt_object_show
2026-05-04 18:54 [PATCH 1/3] Input: atmel_mxt_ts - fix boundary check in mxt_prepare_cfg_mem Dmitry Torokhov
2026-05-04 18:54 ` [PATCH 2/3] Input: atmel_mxt_ts - check mem_size before calculating config memory size Dmitry Torokhov
@ 2026-05-04 18:54 ` Dmitry Torokhov
2026-05-05 9:10 ` Ricardo Ribalda
2026-05-05 9:08 ` [PATCH 1/3] Input: atmel_mxt_ts - fix boundary check in mxt_prepare_cfg_mem Ricardo Ribalda
2 siblings, 1 reply; 6+ messages in thread
From: Dmitry Torokhov @ 2026-05-04 18:54 UTC (permalink / raw)
To: Nick Dyer, linux-input; +Cc: Ricardo Ribalda, linux-kernel
Use the __free(kfree) macro for the obuf allocation in mxt_object_show()
to simplify the code.
Assisted-by: Gemini:gemini-3.1-pro
Signed-off-by: Dmitry Torokhov <dmitry.torokhov@gmail.com>
---
drivers/input/touchscreen/atmel_mxt_ts.c | 10 +++-------
1 file changed, 3 insertions(+), 7 deletions(-)
diff --git a/drivers/input/touchscreen/atmel_mxt_ts.c b/drivers/input/touchscreen/atmel_mxt_ts.c
index d660cc5b5fe3..6af4db5ed117 100644
--- a/drivers/input/touchscreen/atmel_mxt_ts.c
+++ b/drivers/input/touchscreen/atmel_mxt_ts.c
@@ -2870,14 +2870,12 @@ static ssize_t mxt_object_show(struct device *dev,
int count = 0;
int i, j;
int error;
- u8 *obuf;
/* Pre-allocate buffer large enough to hold max sized object. */
- obuf = kmalloc(256, GFP_KERNEL);
+ u8 *obuf __free(kfree) = kmalloc(256, GFP_KERNEL);
if (!obuf)
return -ENOMEM;
- error = 0;
for (i = 0; i < data->info->object_num; i++) {
object = data->object_table + i;
@@ -2892,15 +2890,13 @@ static ssize_t mxt_object_show(struct device *dev,
error = __mxt_read_reg(data->client, addr, size, obuf);
if (error)
- goto done;
+ return error;
count = mxt_show_instance(buf, count, object, j, obuf);
}
}
-done:
- kfree(obuf);
- return error ?: count;
+ return count;
}
static int mxt_check_firmware_format(struct device *dev,
--
2.54.0.545.g6539524ca2-goog
^ permalink raw reply related [flat|nested] 6+ messages in thread* Re: [PATCH 3/3] Input: atmel_mxt_ts - use __free() for obuf in mxt_object_show
2026-05-04 18:54 ` [PATCH 3/3] Input: atmel_mxt_ts - use __free() for obuf in mxt_object_show Dmitry Torokhov
@ 2026-05-05 9:10 ` Ricardo Ribalda
0 siblings, 0 replies; 6+ messages in thread
From: Ricardo Ribalda @ 2026-05-05 9:10 UTC (permalink / raw)
To: Dmitry Torokhov; +Cc: Nick Dyer, linux-input, linux-kernel
On Mon, 4 May 2026 at 20:54, Dmitry Torokhov <dmitry.torokhov@gmail.com> wrote:
>
> Use the __free(kfree) macro for the obuf allocation in mxt_object_show()
> to simplify the code.
>
> Assisted-by: Gemini:gemini-3.1-pro
> Signed-off-by: Dmitry Torokhov <dmitry.torokhov@gmail.com>
Reviewed-by: Ricardo Ribalda <ribalda@chromium.org>
> ---
> drivers/input/touchscreen/atmel_mxt_ts.c | 10 +++-------
> 1 file changed, 3 insertions(+), 7 deletions(-)
>
> diff --git a/drivers/input/touchscreen/atmel_mxt_ts.c b/drivers/input/touchscreen/atmel_mxt_ts.c
> index d660cc5b5fe3..6af4db5ed117 100644
> --- a/drivers/input/touchscreen/atmel_mxt_ts.c
> +++ b/drivers/input/touchscreen/atmel_mxt_ts.c
> @@ -2870,14 +2870,12 @@ static ssize_t mxt_object_show(struct device *dev,
> int count = 0;
> int i, j;
> int error;
> - u8 *obuf;
>
> /* Pre-allocate buffer large enough to hold max sized object. */
> - obuf = kmalloc(256, GFP_KERNEL);
> + u8 *obuf __free(kfree) = kmalloc(256, GFP_KERNEL);
> if (!obuf)
> return -ENOMEM;
>
> - error = 0;
> for (i = 0; i < data->info->object_num; i++) {
> object = data->object_table + i;
>
> @@ -2892,15 +2890,13 @@ static ssize_t mxt_object_show(struct device *dev,
>
> error = __mxt_read_reg(data->client, addr, size, obuf);
> if (error)
> - goto done;
> + return error;
>
> count = mxt_show_instance(buf, count, object, j, obuf);
> }
> }
>
> -done:
> - kfree(obuf);
> - return error ?: count;
> + return count;
> }
>
> static int mxt_check_firmware_format(struct device *dev,
> --
> 2.54.0.545.g6539524ca2-goog
>
--
Ricardo Ribalda
^ permalink raw reply [flat|nested] 6+ messages in thread
* Re: [PATCH 1/3] Input: atmel_mxt_ts - fix boundary check in mxt_prepare_cfg_mem
2026-05-04 18:54 [PATCH 1/3] Input: atmel_mxt_ts - fix boundary check in mxt_prepare_cfg_mem Dmitry Torokhov
2026-05-04 18:54 ` [PATCH 2/3] Input: atmel_mxt_ts - check mem_size before calculating config memory size Dmitry Torokhov
2026-05-04 18:54 ` [PATCH 3/3] Input: atmel_mxt_ts - use __free() for obuf in mxt_object_show Dmitry Torokhov
@ 2026-05-05 9:08 ` Ricardo Ribalda
2 siblings, 0 replies; 6+ messages in thread
From: Ricardo Ribalda @ 2026-05-05 9:08 UTC (permalink / raw)
To: Dmitry Torokhov; +Cc: Nick Dyer, linux-input, linux-kernel, stable
HI Dmitry
FWIW this patch looks correct to me...
Reviewed-by: Ricardo Ribalda <ribalda@chromium.org>
But there are a couple of things that look weird.
1) The patch line (1503) does not seem to match your tree
https://git.kernel.org/pub/scm/linux/kernel/git/dtor/input.git/tree/drivers/input/touchscreen/atmel_mxt_ts.c#n1503
2) The sscanf just before this check has two conversions (val and
offset), but you only check for ret != 1. Should't it be ret !=2? or I
am missing something?
On Mon, 4 May 2026 at 20:54, Dmitry Torokhov <dmitry.torokhov@gmail.com> wrote:
>
> When a configuration file provides an object size that is larger than the
> driver's known mxt_obj_size(object), the driver intends to discard the
> extra bytes.
>
> The loop iterates using for (i = 0; i < size; i++). Inside the loop, the
> condition to skip processing extra bytes is:
>
> if (i > mxt_obj_size(object))
> continue;
>
> Since i is a 0-based index, the valid indices for the object are 0 through
> mxt_obj_size(object) - 1.
>
> When i == mxt_obj_size(object), the condition evaluates to false, and the
> code processes the byte instead of discarding it.
>
> This causes the code to calculate byte_offset = reg + i - cfg->start_ofs
> and writes the byte there, overwriting exactly one byte of the adjacent
> instance or object.
>
> Update the boundary check to skip extra bytes correctly by using >=.
>
> Fixes: 50a77c658b80 ("Input: atmel_mxt_ts - download device config using firmware loader")
> Cc: stable@vger.kernel.org
> Assisted-by: Gemini:gemini-3.1-pro
> Signed-off-by: Dmitry Torokhov <dmitry.torokhov@gmail.com>
> ---
> drivers/input/touchscreen/atmel_mxt_ts.c | 2 +-
> 1 file changed, 1 insertion(+), 1 deletion(-)
>
> diff --git a/drivers/input/touchscreen/atmel_mxt_ts.c b/drivers/input/touchscreen/atmel_mxt_ts.c
> index d62bf2c95578..28b2bd889c70 100644
> --- a/drivers/input/touchscreen/atmel_mxt_ts.c
> +++ b/drivers/input/touchscreen/atmel_mxt_ts.c
> @@ -1503,7 +1503,7 @@ static int mxt_prepare_cfg_mem(struct mxt_data *data, struct mxt_cfg *cfg)
> }
> cfg->raw_pos += offset;
>
> - if (i > mxt_obj_size(object))
> + if (i >= mxt_obj_size(object))
> continue;
>
> byte_offset = reg + i - cfg->start_ofs;
> --
> 2.54.0.545.g6539524ca2-goog
>
--
Ricardo Ribalda
^ permalink raw reply [flat|nested] 6+ messages in thread