pwm: lp3943: validate DT output indices in lp3943_pwm_parse_dt()
Subsystems (2)
PATCH 1/1 pwm: lp3943: validate DT output indices in lp3943_pwm_parse_dt()
The ti,pwm0/ti,pwm1 devicetree properties are raw u32 arrays naming which of the 16 physical LP3943 outputs (0-15, per the documented binding and the 16-entry enum lp3943_pwm_output) each PWM channel drives. lp3943_pwm_parse_dt() reads these values straight from the devicetree with of_property_read_u32_array() and stores them in pwm_map->output[] without checking they are actually in range.
Both consumers of pwm_map->output[] then use an out-of-range value directly as an index/bit-offset with no bounds check of their own:
- lp3943_pwm_request_map() does
test_and_set_bit(offset, &lp3943->pin_used)
with offset taken straight from output[i], corrupting memory beyond the single 'unsigned long pin_used' field for offset >= BITS_PER_LONG.
- lp3943_pwm_set_mode() does
mux[index].reg / .mask / .shift
with index taken straight from output[i], reading past the 16-entry lp3943_mux_cfg[] array and then issuing a regmap write built from whatever garbage register/mask/shift values were read out of bounds.
A devicetree (malformed board file or a loaded overlay) with an out-of-range ti,pwm0/ti,pwm1 value is enough to reach either path; nothing upstream of these two functions validates the value.
Add a range check in lp3943_pwm_parse_dt() right after the values are read, rejecting the devicetree with -EINVAL before any out-of-range value can reach pwm_map->output[]. Add the LP3943_NUM_OUTPUTS(16) constant next to the existing enum it bounds, alongside the existing LP3943_NUM_PWMS(2) constant.
--- drivers/pwm/pwm-lp3943.c | 7 ++++++- include/linux/mfd/lp3943.h | 2 ++ 2 files changed, 8 insertions(+), 1 deletion(-)
drivers/pwm/pwm-lp3943.c +6 -1
@@ -218,7 +218,7 @@ static int lp3943_pwm_parse_dt(struct device *dev,struct lp3943_platform_data *pdata;struct lp3943_pwm_map *pwm_map;enum lp3943_pwm_output *output; -int i, err, num_outputs, count =0 ; +int i, j, err, num_outputs, count =0 ;if (!node)return -EINVAL ;
@@ -247,6 +247,11 @@ static int lp3943_pwm_parse_dt(struct device *dev,if (err)return err; +for (j =0 ; j < num_outputs; j++) { +if (output[j] >=LP3943_NUM_OUTPUTS ) +return -EINVAL ; + } + pwm_map =devm_kzalloc (dev,sizeof (*pwm_map),GFP_KERNEL );if (!pwm_map)return -ENOMEM ;
include/linux/mfd/lp3943.h +2 -0
@@ -53,6 +53,8 @@ enum lp3943_pwm_output { LP3943_PWM_OUT15, }; +#define LP3943_NUM_OUTPUTS 16 +/* * struct lp3943_pwm_map * @output: Output pins which are mapped to each PWM channel
Signature · 1 line
2.46.2.windows.1
[email protected]
AI reviewThank you for your contribution! Sashiko AI review found 2 potential issue(s) to consider:
Thank you for your contribution! Sashiko AI review found 2 potential issue(s) to consider:
New issues:
- [High] The newly added bounds check for device tree output indices is incomplete because it performs a signed comparison. Negative values can bypass the check, leading to out-of-bounds memory accesses.
Pre-existing issues:
- [High] Missing NULL check on pdata->pwms[hwpwm] leads to a crash if a partially configured PWM chip is requested. --
commit 6bf884f857d8ed9b87f551fe2b3bdb8a985eae4f
Author: Manush Prajwal <[email protected]>
pwm: lp3943: validate DT output indices in lp3943_pwm_parse_dt()
This commit adds a bounds check in lp3943_pwm_parse_dt() to validate PWM output indices read from the devicetree. It ensures that the parsed output values are within the valid range before storing them.
Quoted 4 lines
drivers/pwm/pwm-lp3943.c +0
-0
[ ... ]
drivers/pwm/pwm-lp3943.c static int lp3943_pwm_parse_dt(struct device *dev, · line 254
Quoted 9 lines
drivers/pwm/pwm-lp3943.c +5
-0@@ -247,6 +247,11 @@ static int lp3943_pwm_parse_dt(struct device *dev,if (err)return err; +for (j =0 ; j < num_outputs; j++) { +if (output[j] >=LP3943_NUM_OUTPUTS ) +return -EINVAL ; + } +
[Severity: High]
Does this bounds check fail to reject negative values resulting from type conversion, leaving the driver vulnerable to out-of-bounds memory accesses?
A malformed devicetree (e.g., loaded via DT overlays) can provide a large 32-bit value with the MSB set (like 0xFFFFFFFF) for a PWM output index.
Because enum lp3943_pwm_output is treated as a signed integer type under standard ABIs, the value is interpreted as negative.
The newly added check if (output[j] >= LP3943_NUM_OUTPUTS) performs a signed comparison because LP3943_NUM_OUTPUTS (16) is a signed literal. The negative value evaluates as less than 16, incorrectly bypassing the bounds check.
The negative value can then be used as an out-of-bounds array index:
drivers/pwm/pwm-lp3943.c:lp3943_pwm_set_mode() {
...
index = pwm_map->output[i];
err = lp3943_update_bits(lp3943, mux[index].reg,
mux[index].mask,
val << mux[index].shift);
...
}drivers/pwm/pwm-lp3943.c static int lp3943_pwm_parse_dt(struct device *dev, · line 257
Quoted 3 lines
drivers/pwm/pwm-lp3943.c +0
-0pwm_map =devm_kzalloc (dev,sizeof (*pwm_map),GFP_KERNEL );if (!pwm_map)return -ENOMEM ;
[Severity: High]
This isn't a bug introduced by this patch, but is it possible for a missing NULL check on pdata->pwms[hwpwm] to lead to a crash if a partially configured PWM chip is requested?
During probe, lp3943_pwm_parse_dt() iterates over the ti,pwm0 and ti,pwm1 properties. If a property is missing, the loop skips initializing that channel, leaving pdata->pwms[i] as NULL.
However, the probe function registers the PWM chip unconditionally for both channels. If a sysfs user requests the missing channel, the PWM core invokes lp3943_pwm_request_map(), which blindly dereferences pdata->pwms[hwpwm]:
drivers/pwm/pwm-lp3943.c:lp3943_pwm_request_map() {
...
pwm_map->output = pdata->pwms[hwpwm]->output;
pwm_map->num_outputs = pdata->pwms[hwpwm]->num_outputs;
...
}Signature · 1 line
Sashiko AI review · https://sashiko.dev/#/patchset/[email protected]?part=1