On 2018-09-29, Vasily Khoruzhick anarsoul@gmail.com wrote:
On Sat, Sep 29, 2018 at 4:46 PM Vagrant Cascadian vagrant@debian.org wrote:
Sleep gpio is optional, so it's possible to have reset gpio, but no sleep gpio. We shouldn't fail early in case of missing sleep gpio, otherwise we won't deassert reset.
...
diff --git a/drivers/video/bridge/video-bridge-uclass.c b/drivers/video/bridge/video-bridge-uclass.c index cd4959cc71..46936a0626 100644 --- a/drivers/video/bridge/video-bridge-uclass.c +++ b/drivers/video/bridge/video-bridge-uclass.c @@ -110,7 +110,7 @@ int video_bridge_set_active(struct udevice *dev, bool active)
debug("%s: %d\n", __func__, active); ret = dm_gpio_set_value(&uc_priv->sleep, !active);
if (ret)
if (ret != -ENOENT)It should be 'if (ret && ret != -ENOENT)'. Btw, I fixed it in pinebook-wip-20180909 branch.
That's where I pulled the patch from; it's present in the patch to anx6345.c, but apparently unpatched in this patch against video-bridge-uclass.c.
I'll submit the fixed version in a new patch series after collecting more comments...
Thanks for all your work on it!
live well, vagrant