gpio: pca953x: add a comment explaining the need for a lockdep subclass

This is a follow-up to commit 559b46990e76 ("gpio: pca953x: fix an
incorrect lockdep warning"). The reason for calling
lockdep_set_subclass() in pca953x_probe() is not explained in
the code.

Add a comment describing the problem, partial solution and required
future extensions.

Signed-off-by: Bartosz Golaszewski <bgolaszewski@baylibre.com>
Acked-by: Linus Walleij <linus.walleij@linaro.org>
Signed-off-by: Wolfram Sang <wsa@the-dreams.de>

diff --git a/drivers/gpio/gpio-pca953x.c b/drivers/gpio/gpio-pca953x.c
index 45c8817..e422568 100644 (file)
--- a/drivers/gpio/gpio-pca953x.c
+++ b/drivers/gpio/gpio-pca953x.c
@@ -794,6 +794,22 @@ static int pca953x_probe(struct i2c_client *client,
        }
 
        mutex_init(&chip->i2c_lock);
        }
 
        mutex_init(&chip->i2c_lock);

+       /*
+        * In case we have an i2c-mux controlled by a GPIO provided by an
+        * expander using the same driver higher on the device tree, read the
+        * i2c adapter nesting depth and use the retrieved value as lockdep
+        * subclass for chip->i2c_lock.
+        *
+        * REVISIT: This solution is not complete. It protects us from lockdep
+        * false positives when the expander controlling the i2c-mux is on
+        * a different level on the device tree, but not when it's on the same
+        * level on a different branch (in which case the subclass number
+        * would be the same).
+        *
+        * TODO: Once a correct solution is developed, a similar fix should be
+        * applied to all other i2c-controlled GPIO expanders (and potentially
+        * regmap-i2c).
+        */

        lockdep_set_subclass(&chip->i2c_lock,
                             i2c_adapter_depth(client->adapter));
 
        lockdep_set_subclass(&chip->i2c_lock,
                             i2c_adapter_depth(client->adapter));