Hi, Chun-Kuang:
On Fri, Dec 31, 2021 at 12:31:33AM +0800, Chun-Kuang Hu wrote:
diff --git a/drivers/phy/mediatek/phy-mtk-mipi-dsi.c b/drivers/phy/mediatek/phy-mtk-mipi-dsi.c index 28ad9403c441..67b005d5b9e3 100644 --- a/drivers/phy/mediatek/phy-mtk-mipi-dsi.c +++ b/drivers/phy/mediatek/phy-mtk-mipi-dsi.c @@ -146,6 +146,8 @@ static int mtk_mipi_tx_probe(struct platform_device *pdev) return -ENOMEM;
mipi_tx->driver_data = of_device_get_match_data(dev);
if (!mipi_tx->driver_data)
I'm confused. mtk_mipi_tx_probe() is called because this device node's compatible match one in mtk_mipi_tx_match[]. So I think the return value of of_device_get_match_data(dev) would not be NULL. If this is true, this checking is redundant.
I think your are right, this checking is redundant. Thanks for your reply.
Regards, Miaoqian.