On Wed, Aug 21, 2024 at 01:54:57PM +0530, Imran Shaik wrote:
>
>
> On 8/20/2024 7:36 PM, Andrew Lunn wrote:
> > > +static int gcc_qcs8300_probe(struct platform_device *pdev)
> > > +{
> > > + struct regmap *regmap;
> > > + int ret;
> > > +
> > > + regmap = qcom_cc_map(pdev, &gcc_qcs8300_desc);
> > > + if (IS_ERR(regmap))
> > > + return PTR_ERR(regmap);
> > > +
> > > + ret = qcom_cc_register_rcg_dfs(regmap, gcc_dfs_clocks,
> > > + ARRAY_SIZE(gcc_dfs_clocks));
> > > + if (ret)
> > > + return ret;
> > > +
> > > + /* Keep some clocks always enabled */
> > > + qcom_branch_set_clk_en(regmap, 0x32004); /* GCC_CAMERA_AHB_CLK */
> > > + qcom_branch_set_clk_en(regmap, 0x32020); /* GCC_CAMERA_XO_CLK */
> >
> > It would be good to document why. Why does the camera driver not
> > enable the clock when it loads?
> >
> > > + qcom_branch_set_clk_en(regmap, 0x33004); /* GCC_DISP_AHB_CLK */
> > > + qcom_branch_set_clk_en(regmap, 0x33018); /* GCC_DISP_XO_CLK */
> > > + qcom_branch_set_clk_en(regmap, 0x7d004); /* GCC_GPU_CFG_AHB_CLK */
> > > + qcom_branch_set_clk_en(regmap, 0x34004); /* GCC_VIDEO_AHB_CLK */
> > > + qcom_branch_set_clk_en(regmap, 0x34024); /* GCC_VIDEO_XO_CLK */
> >
> > Why cannot the display driver enable the clock when it loads?
> >
>
> These clocks require for DISPCC and CAMCC drivers for register access, hence
> kept enabled at GCC driver probe. The same approach is followed for all the
> targets.
No, the DISPCC and CAMCC driver should get and enable these clocks
before accessing the hardware. They should be turned off until then,
and maybe they will never be used if it is a headless box without a
camera.
Andrew
