On 3/3/21 6:06 AM, Álvaro Fernández Rojas wrote: > Hi Philipp, > >> El 3 mar 2021, a las 14:52, Philipp Zabel <p.zabel@xxxxxxxxxxxxxx> escribió: >> >> Hi Álvaro, >> >> On Wed, 2021-02-24 at 09:22 +0100, Álvaro Fernández Rojas wrote: >> [...] >>> @@ -115,6 +121,8 @@ static void bcm2835_rng_cleanup(struct hwrng *rng) >>> /* disable rng hardware */ >>> rng_writel(priv, 0, RNG_CTRL); >>> >>> + reset_control_rearm(priv->reset); >>> + >>> if (!IS_ERR(priv->clk)) >>> clk_disable_unprepare(priv->clk); >>> } >>> @@ -159,6 +167,10 @@ static int bcm2835_rng_probe(struct platform_device *pdev) >>> if (PTR_ERR(priv->clk) == -EPROBE_DEFER) >>> return -EPROBE_DEFER; >>> >>> + priv->reset = devm_reset_control_get_optional_exclusive(dev, NULL); >>> + if (IS_ERR(priv->reset)) >>> + return PTR_ERR(priv->reset); >>> + >>> priv->rng.name = pdev->name; >>> priv->rng.init = bcm2835_rng_init; >>> priv->rng.read = bcm2835_rng_read; >> >> That doesn't seem right. reset_control_rearm() doesn't do anything if >> the reset control is exclusive. Either the reset control should be >> requested as shared, or the _rearm should be removed. > > In only added reset_control_rearm() because Florian requested it… > I think it’s not needed, so we can use v3, since it was the only change between v3 and v4... Not the first time I am confused by the reset API not sure if I will ever get it one day, so apologies for suggesting something incorrect here. -- Florian
