On Thu, Jan 29, 2026 at 1:36 PM Hongyang Zhao
<[email protected]> wrote:
>
> The lt9611_sleep_setup() modifies the chip's I2C address via registers
> 0x8023/0x8024 and reconfigures GPIO0 via 0x8157/0x8149. Since there is
> no corresponding restore in the wake-up path, I2C communication fails
> after resume.
>
> Remove the I2C address and GPIO reconfiguration from sleep setup. The
> MIPI Rx power-down and TX PHY disable sequences are sufficient.
>
> Signed-off-by: Hongyang Zhao <[email protected]>
> ---
> The lt9611_sleep_setup() function modifies the chip's I2C address via
> registers 0x8023/0x8024 and reconfigures GPIO0. However, there is no
> corresponding restore in the wake-up path, which causes I2C communication
> to fail after resume.
>
> This patch removes the I2C address and GPIO reconfiguration from sleep
> setup. The MIPI Rx power-down and TX PHY disable sequences are sufficient
> for proper sleep operation.
>
> Tested on RubikPi3 (QCS6490) platform with the LT9611 HDMI bridge.
> ---
>  drivers/gpu/drm/bridge/lontium-lt9611.c | 5 -----
>  1 file changed, 5 deletions(-)
>
> diff --git a/drivers/gpu/drm/bridge/lontium-lt9611.c 
> b/drivers/gpu/drm/bridge/lontium-lt9611.c
> index a2d032ee4744..ae47efe9d77f 100644
> --- a/drivers/gpu/drm/bridge/lontium-lt9611.c
> +++ b/drivers/gpu/drm/bridge/lontium-lt9611.c
> @@ -424,11 +424,6 @@ static void lt9611_enable_hpd_interrupts(struct lt9611 
> *lt9611)
>  static void lt9611_sleep_setup(struct lt9611 *lt9611)
>  {
>         const struct reg_sequence sleep_setup[] = {
> -               { 0x8024, 0x76 },
> -               { 0x8023, 0x01 },
> -               { 0x8157, 0x03 }, /* set addr pin as output */
> -               { 0x8149, 0x0b },
> -
>                 { 0x8102, 0x48 }, /* MIPI Rx power down */
>                 { 0x8123, 0x80 },
>                 { 0x8130, 0x00 },
>
> ---

Tested-by: Roger Shimizu <[email protected]> # Rubik Pi 3 (QCOM QCS6490 platform)

Before this patch, HDMI output can be used just after booting / 1st plugging in.
HDMI gets lost after the disconnecting, and plugging in again.

-Roger

Reply via email to