From: Dang Huynh <dang.huynh@mainlining.org>

The register doesn't like to be read and would cause all the input
to be switched to output.

This causes the SD Card Detect GPIO to misbehaves in the OS and/or
may cause hardware to malfunction.

Signed-off-by: Dang Huynh <dang.huynh@mainlining.org>
---
 drivers/gpio/gpio-rda.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/gpio/gpio-rda.c b/drivers/gpio/gpio-rda.c
index 7d20dbbb6ec41149a1dbf4d9ef9ac86449773c35..ae2efa22755986b5445fdb8d4f004d595c189590 100644
--- a/drivers/gpio/gpio-rda.c
+++ b/drivers/gpio/gpio-rda.c
@@ -245,7 +245,7 @@ static int rda_gpio_probe(struct platform_device *pdev)
 		.clr = rda_gpio->base + RDA_GPIO_CLR,
 		.dirout = rda_gpio->base + RDA_GPIO_OEN_SET_OUT,
 		.dirin = rda_gpio->base + RDA_GPIO_OEN_SET_IN,
-		.flags = BGPIOF_READ_OUTPUT_REG_SET,
+		.flags = BGPIOF_READ_OUTPUT_REG_SET | BGPIOF_UNREADABLE_REG_DIR,
 	};
 
 	ret = gpio_generic_chip_init(&rda_gpio->chip, &config);

-- 
2.51.0

On Thu, Sep 18, 2025 at 8:49 PM Dang Huynh via B4 Relay
<devnull+dang.huynh.mainlining.org@kernel.org> wrote:
>
> From: Dang Huynh <dang.huynh@mainlining.org>
>
> The register doesn't like to be read and would cause all the input
> to be switched to output.
>
> This causes the SD Card Detect GPIO to misbehaves in the OS and/or
> may cause hardware to malfunction.
>
> Signed-off-by: Dang Huynh <dang.huynh@mainlining.org>
> ---

I have responded to you already[1]. The commit message is still
sub-par. Don't say "the register doesn't like", say: "Reading the X
register causes Y to happen. In order to remedy it, do Z." You haven't
answered my question: what will happen to existing users of this
driver?

Also: you have not bumped the series version to v2, that makes
tracking the changes hard.

Bartosz

[1] https://lore.kernel.org/all/CAMRc=MeHQf_Oa2DRR0T7tum-Tuk3qPh5r5gimxGY3EXTyvoKZQ@mail.gmail.com/