Skip to contentSkip to conversation

gpiolib: cdev: fix uninitialised kfifo

PATCH Applied Kent Gibson 3 participants 5 messages May 10, 2024 – May 14, 2024

Subsystems (2)
lore
Share & tools
lore

Kent Gibson

author reviewer
#0 lore raw

PATCH 1/1 gpiolib: cdev: fix uninitialised kfifo

If a line is requested with debounce, and that results in debouncing in software, and the line is subsequently reconfigured to enable edge detection then the allocation of the kfifo to contain edge events is overlooked. This results in events being written to and read from an unitialised kfifo. Read events are returned to userspace.

Initialise the kfifo in the case where the software debounce is already active.

Fixes 65cff7046406 ("gpiolib: cdev: support setting debounce")
---
 drivers/gpio/gpiolib-cdev.c | 14 ++++++++++++++
 1 file changed, 14 insertions(+)
drivers/gpio/gpiolib-cdev.c +14 -0
index d09c7d728365..57c92395219e 100644
@@ -1193,6 +1193,8 @@ static int edge_detector_update(struct line *line,
 				struct gpio_v2_line_config *lc,
 				unsigned int line_idx, u64 edflags)
 {
+	u64 eflags;
+	int ret;
 	u64 active_edflags = READ_ONCE(line->edflags);
 	unsigned int debounce_period_us =
 			gpio_v2_line_config_debounce_period(lc, line_idx);
@@ -1204,6 +1206,18 @@ static int edge_detector_update(struct line *line,
 	/* sw debounced and still will be...*/
 	if (debounce_period_us && READ_ONCE(line->sw_debounced)) {
 		line_set_debounce_period(line, debounce_period_us);
+		/*
+		 * ensure event fifo is initialised if edge detection
+		 * is now enabled.
+		 */
+		eflags = edflags & GPIO_V2_LINE_EDGE_FLAGS;
+		if (eflags && !kfifo_initialized(&line->req->events)) {
+			ret = kfifo_alloc(&line->req->events,
+					  line->req->event_buffer_size,
+					  GFP_KERNEL);
+			if (ret)
+				return ret;
+		}
 		return 0;
 	}
 
Signature · 1 line
2.39.2
#1 lore raw

From: Bartosz Golaszewski <[email protected]>

Quoted 10 lines · Kent Gibson

If a line is requested with debounce, and that results in debouncing in software, and the line is subsequently reconfigured to enable edge detection then the allocation of the kfifo to contain edge events is overlooked. This results in events being written to and read from an unitialised kfifo. Read events are returned to userspace.

Initialise the kfifo in the case where the software debounce is already active.

[...]

Applied, thanks!

[1/1] gpiolib: cdev: fix uninitialised kfifo

      commit: 3c1625fe5a2e0d68cd7b68156f02c1b5de09a161

Best regards,

Signature · 1 line
Bartosz Golaszewski <[email protected]>
author reviewer
#2 lore raw
Quoted 13 lines · Bartosz Golaszewski

From: Bartosz Golaszewski <[email protected]>

Quoted 9 lines · Kent Gibson

If a line is requested with debounce, and that results in debouncing in software, and the line is subsequently reconfigured to enable edge detection then the allocation of the kfifo to contain edge events is overlooked. This results in events being written to and read from an unitialised kfifo. Read events are returned to userspace.

Initialise the kfifo in the case where the software debounce is already active.

Quoted 7 lines
Quoted 1 line

[...]

Applied, thanks!

[1/1] gpiolib: cdev: fix uninitialised kfifo

      commit: 3c1625fe5a2e0d68cd7b68156f02c1b5de09a161

I've got a patch series to tidy this up and catch any similar errors earlier going forward.
It is of course based on this patch, but that isn't in gpio/for-next yet. How should I proceed?

Cheers,
Kent.

#3 lore raw
Quoted 23 lines · Kent Gibson
Quoted 13 lines · Bartosz Golaszewski

From: Bartosz Golaszewski <[email protected]>

Quoted 9 lines · Kent Gibson

If a line is requested with debounce, and that results in debouncing in software, and the line is subsequently reconfigured to enable edge detection then the allocation of the kfifo to contain edge events is overlooked. This results in events being written to and read from an unitialised kfifo. Read events are returned to userspace.

Initialise the kfifo in the case where the software debounce is already active.

Quoted 7 lines
Quoted 1 line

[...]

Applied, thanks!

[1/1] gpiolib: cdev: fix uninitialised kfifo

      commit: 3c1625fe5a2e0d68cd7b68156f02c1b5de09a161
Quoted 7 lines

I've got a patch series to tidy this up and catch any similar errors earlier going forward.
It is of course based on this patch, but that isn't in gpio/for-next yet. How should I proceed?

Cheers,
Kent.

Pull in current master into gpio/for-next, apply your series, make sure it works and then wait until v6.10-rc1 is tagged because I won't be picking up anything during merge window.

Bart

author reviewer
#4 lore raw
Quoted 29 lines · Bartosz Golaszewski
Quoted 28 lines · Kent Gibson
Quoted 13 lines · Bartosz Golaszewski

From: Bartosz Golaszewski <[email protected]>

Quoted 9 lines · Kent Gibson

If a line is requested with debounce, and that results in debouncing in software, and the line is subsequently reconfigured to enable edge detection then the allocation of the kfifo to contain edge events is overlooked. This results in events being written to and read from an unitialised kfifo. Read events are returned to userspace.

Initialise the kfifo in the case where the software debounce is already active.

Quoted 7 lines
Quoted 1 line

[...]

Applied, thanks!

[1/1] gpiolib: cdev: fix uninitialised kfifo

      commit: 3c1625fe5a2e0d68cd7b68156f02c1b5de09a161

I've got a patch series to tidy this up and catch any similar errors earlier going forward.
It is of course based on this patch, but that isn't in gpio/for-next yet. How should I proceed?

Quoted 7 lines
Quoted 2 lines

Cheers,
Kent.

Pull in current master into gpio/for-next, apply your series, make sure it works and then wait until v6.10-rc1 is tagged because I won't be picking up anything during merge window.

Will do.

Thanks,
Kent.