If the adp5588_read function returns 0, then there will be an
overflow of the kpad->keycode buffer.

If the adp5588_read function returns a negative value, then the
logic is broken - the wrong value is used as an index of
the kpad->keycode array.

Found by Linux Verification Center (linuxtesting.org) with SVACE.

Cc: stable@vger.kernel.org # v5.10+
Signed-off-by: Denis Arefev <arefev@swemel.ru>
---
 drivers/input/keyboard/adp5588-keys.c | 11 +++++++----
 1 file changed, 7 insertions(+), 4 deletions(-)

diff --git a/drivers/input/keyboard/adp5588-keys.c b/drivers/input/keyboard/adp5588-keys.c
index 1b0279393df4..d05387f9c11f 100644
--- a/drivers/input/keyboard/adp5588-keys.c
+++ b/drivers/input/keyboard/adp5588-keys.c
@@ -526,14 +526,17 @@ static void adp5588_report_events(struct adp5588_kpad *kpad, int ev_cnt)
 	int i;
 
 	for (i = 0; i < ev_cnt; i++) {
-		int key = adp5588_read(kpad->client, KEY_EVENTA + i);
-		int key_val = key & KEY_EV_MASK;
-		int key_press = key & KEY_EV_PRESSED;
+		int key, key_val, key_press;
 
+		key = adp5588_read(kpad->client, KEY_EVENTA + i);
+		if (key < 0)
+			continue;
+		key_val = key & KEY_EV_MASK;
+		key_press = key & KEY_EV_PRESSED;
 		if (key_val >= GPI_PIN_BASE && key_val <= GPI_PIN_END) {
 			/* gpio line used as IRQ source */
 			adp5588_gpio_irq_handle(kpad, key_val, key_press);
-		} else {
+		} else if (key_val > 0) {
 			int row = (key_val - 1) / ADP5588_COLS_MAX;
 			int col =  (key_val - 1) % ADP5588_COLS_MAX;
 			int code = MATRIX_SCAN_CODE(row, col, kpad->row_shift);
-- 
2.25.1

On Mon, 2024-09-30 at 11:36 +0300, Denis Arefev wrote:
> If the adp5588_read function returns 0, then there will be an
> overflow of the kpad->keycode buffer.
> 
> If the adp5588_read function returns a negative value, then the
> logic is broken - the wrong value is used as an index of
> the kpad->keycode array.
> 
> Found by Linux Verification Center (linuxtesting.org) with SVACE.
> 
> Cc: stable@vger.kernel.org # v5.10+
> Signed-off-by: Denis Arefev <arefev@swemel.ru>
> ---

Hi Denis,

Thanks for the patch. However, I'm working on a more complete rework of this as
suggested in [1]. I should be sending patches for it today or tomorrow.

[1]: https://lore.kernel.org/linux-input/Zu0vq0ogr2HzXWv7@google.com/
- Nuno Sá