[PATCH 02/47] pwm: mxs: Don't modify HW state in .probe() after the PWM chip was registered

["Uwe Kleine-König" <u.kleine-koenig@pengutronix.de>]

This fixes a race condition: After pwmchip_add() is called there might
already be a consumer and then modifying the hardware behind the
consumer's back is bad. So reset before calling pwmchip_add().

Note that reseting the hardware isn't the right thing to do if the PWM
is already running as it might e.g. disable (or even enable) a backlight
that is supposed to be on (or off).

Fixes: 4dce82c1e840 ("pwm: add pwm-mxs support")
Cc: Sascha Hauer <s.hauer@pengutronix.de>
Cc: Shawn Guo <shawnguo@kernel.org>
Signed-off-by: Uwe Kleine-König <u.kleine-koenig@pengutronix.de>
---
 drivers/pwm/pwm-mxs.c | 13 +++++--------
 1 file changed, 5 insertions(+), 8 deletions(-)

diff --git a/drivers/pwm/pwm-mxs.c b/drivers/pwm/pwm-mxs.c
index a22180803bd7..558dc1de8f5d 100644
--- a/drivers/pwm/pwm-mxs.c
+++ b/drivers/pwm/pwm-mxs.c
@@ -145,6 +145,11 @@ static int mxs_pwm_probe(struct platform_device *pdev)
 		return ret;
 	}
 
+	/* FIXME: Only do this if the PWM isn't already running */
+	ret = stmp_reset_block(mxs->base);
+	if (ret)
+		return dev_err_probe(&pdev->dev, ret, "failed to reset PWM\n");
+
 	ret = pwmchip_add(&mxs->chip);
 	if (ret < 0) {
 		dev_err(&pdev->dev, "failed to add pwm chip %d\n", ret);
@@ -153,15 +158,7 @@ static int mxs_pwm_probe(struct platform_device *pdev)
 
 	platform_set_drvdata(pdev, mxs);
 
-	ret = stmp_reset_block(mxs->base);
-	if (ret)
-		goto pwm_remove;
-
 	return 0;
-
-pwm_remove:
-	pwmchip_remove(&mxs->chip);
-	return ret;
 }
 
 static int mxs_pwm_remove(struct platform_device *pdev)
-- 
2.30.2