On Mon, Apr 30, 2018 at 05:10:52PM +0530, Jagan Teki wrote:
From: Icenowy Zheng icenowy@aosc.io
Allwinner SoCs with DWC HDMI controller have a "HVCC" power pin for the HDMI part, and on some boards it's connected to a dedicated regulator rather than the main 3.3v.
Add support for optional HVCC regulator. For boards that doesn't use a dedicated regulator to power it, the default dummy regulator is used.
Signed-off-by: Icenowy Zheng icenowy@aosc.io Signed-off-by: Jagan Teki jagan@amarulasolutions.com
drivers/gpu/drm/sun4i/sun8i_dw_hdmi.c | 14 ++++++++++++++ drivers/gpu/drm/sun4i/sun8i_dw_hdmi.h | 2 ++ 2 files changed, 16 insertions(+)
diff --git a/drivers/gpu/drm/sun4i/sun8i_dw_hdmi.c b/drivers/gpu/drm/sun4i/sun8i_dw_hdmi.c index 9f40a44b456b..7c33faff7ad4 100644 --- a/drivers/gpu/drm/sun4i/sun8i_dw_hdmi.c +++ b/drivers/gpu/drm/sun4i/sun8i_dw_hdmi.c @@ -73,6 +73,12 @@ static int sun8i_dw_hdmi_bind(struct device *dev, struct device *master, if (encoder->possible_crtcs == 0) return -EPROBE_DEFER;
- hdmi->vcc_hdmi = devm_regulator_get(dev, "hvcc");
- if (IS_ERR(hdmi->vcc_hdmi)) {
dev_err(dev, "Could not get HDMI power supply\n");return PTR_ERR(hdmi->vcc_hdmi);- }
You're not making it optional here, unlike what your commit log says. Not that I'm fine with both, but we need to be consistent :)
hdmi->rst_ctrl = devm_reset_control_get(dev, "ctrl"); if (IS_ERR(hdmi->rst_ctrl)) { dev_err(dev, "Could not get ctrl reset control\n"); @@ -91,6 +97,12 @@ static int sun8i_dw_hdmi_bind(struct device *dev, struct device *master, return ret; }
- ret = regulator_enable(hdmi->vcc_hdmi);
- if (ret) {
dev_err(dev, "Cannot enable HDMI power supply\n");goto err_disable_vcc;- }
- ret = clk_prepare_enable(hdmi->clk_tmds); if (ret) { dev_err(dev, "Could not enable tmds clock\n");
@@ -143,6 +155,8 @@ static int sun8i_dw_hdmi_bind(struct device *dev, struct device *master, clk_disable_unprepare(hdmi->clk_tmds); err_assert_ctrl_reset: reset_control_assert(hdmi->rst_ctrl); +err_disable_vcc:
- regulator_disable(hdmi->vcc_hdmi);
The err_disable_vcc label will disable the regulator, but if there's an error in regulator_enable, it will disable it while it wasn't enabled in the first place, which is not really something we want to do.
Maxime