Re: [PATCH v7 2/7] Revert "clk: imx: clk-imx8mp: Allow media_disp pixel clock reconfigure parent rate"

[Marek Vasut]

On 11/14/24 7:57 AM, Liu Ying wrote:
> This reverts commit ff06ea04e4cf3ba2f025024776e83bfbdfa05155.
>
> media_disp1_pix clock is the pixel clock of the first i.MX8MP LCDIFv3
> display controller, while media_disp2_pix clock is the pixel clock of
> the second i.MX8MP LCDIFv3 display controller.  The two display
> controllers connect with Samsung MIPI DSI controller and LVDS Display
> Bridge(LDB) respectively.  Since the two display controllers are driven
> by separate DRM driver instances and the two pixel clocks may be derived
> from the same video_pll1_out clock(sys_pll3_out clock could be already
> used to derive audio_axi clock), there is no way to negotiate a dynamically
> changeable video_pll1_out clock rate to satisfy both of the two display
> controllers.  In this case, the only solution to drive them with the
> single video_pll1_out clock is to assign a sensible/unchangeable clock
> rate for video_pll1_out clock.  Thus, there is no need to set the
> CLK_SET_RATE_PARENT flag for media_disp{1,2}_pix clocks, drop it then.
>
> Fixes: ff06ea04e4cf ("clk: imx: clk-imx8mp: Allow media_disp pixel clock reconfigure parent rate")
> Signed-off-by: Liu Ying <victor.liu@xxxxxxx>
Uh, I almost missed this revert between all the LDB patches.

This revert will break my usecase on MX8MP where I need to operate two 
disparate panels attached to LVDS and TC358767 DSI-to-DP bridge and I 
need accurate pixel clock for both. Not being able to configure accurate 
pixel clock will make the displays not work, so from my side, this is a 
NAK, sorry.

There has to be some better solution which still allows the PLL 
reconfiguration to achieve accurate pixel clock.