On 27.01.2020 07:00, Peng Fan wrote:

quoted

Subject: Re: [PATCH V3 2/4] clk: imx: imx8mq: use
imx8m_clk_hw_composite_core

On 16.01.2020 04:15, Peng Fan wrote:

quoted

From: Peng Fan <peng.fan@nxp.com>

Use imx8m_clk_hw_composite_core to simplify code.

Reviewed-by: Abel Vesa <redacted>
Signed-off-by: Peng Fan <peng.fan@nxp.com>
---
   drivers/clk/imx/clk-imx8mq.c | 22 ++++++++--------------
   1 file changed, 8 insertions(+), 14 deletions(-)

diff --git a/drivers/clk/imx/clk-imx8mq.c

b/drivers/clk/imx/clk-imx8mq.c index 4c0edca1a6d0..e928c1355ad8

100644

quoted

--- a/drivers/clk/imx/clk-imx8mq.c
+++ b/drivers/clk/imx/clk-imx8mq.c

@@ -403,22 +403,16 @@ static int imx8mq_clocks_probe(struct

platform_device *pdev)

   	/* CORE */
   	hws[IMX8MQ_CLK_A53_SRC] = imx_clk_hw_mux2("arm_a53_src",

base + 0x8000, 24, 3, imx8mq_a53_sels, ARRAY_SIZE(imx8mq_a53_sels));

quoted

-	hws[IMX8MQ_CLK_M4_SRC] = imx_clk_hw_mux2("arm_m4_src", base

+ 0x8080, 24, 3, imx8mq_arm_m4_sels, ARRAY_SIZE(imx8mq_arm_m4_sels));

quoted

-	hws[IMX8MQ_CLK_VPU_SRC] = imx_clk_hw_mux2("vpu_src", base +

0x8100, 24, 3, imx8mq_vpu_sels, ARRAY_SIZE(imx8mq_vpu_sels));

quoted

-	hws[IMX8MQ_CLK_GPU_CORE_SRC] =

imx_clk_hw_mux2("gpu_core_src", base + 0x8180, 24, 3,
imx8mq_gpu_core_sels, ARRAY_SIZE(imx8mq_gpu_core_sels));

quoted

-	hws[IMX8MQ_CLK_GPU_SHADER_SRC] =

imx_clk_hw_mux2("gpu_shader_src", base + 0x8200, 24, 3,
imx8mq_gpu_shader_sels,  ARRAY_SIZE(imx8mq_gpu_shader_sels));

quoted

-
   	hws[IMX8MQ_CLK_A53_CG] =

imx_clk_hw_gate3_flags("arm_a53_cg", "arm_a53_src", base + 0x8000, 28,
CLK_IS_CRITICAL);

quoted

-	hws[IMX8MQ_CLK_M4_CG] = imx_clk_hw_gate3("arm_m4_cg",

"arm_m4_src", base + 0x8080, 28);

quoted

-	hws[IMX8MQ_CLK_VPU_CG] = imx_clk_hw_gate3("vpu_cg", "vpu_src",

base + 0x8100, 28);

quoted

-	hws[IMX8MQ_CLK_GPU_CORE_CG] = imx_clk_hw_gate3("gpu_core_cg",

"gpu_core_src", base + 0x8180, 28);

quoted

-	hws[IMX8MQ_CLK_GPU_SHADER_CG] =

imx_clk_hw_gate3("gpu_shader_cg", "gpu_shader_src", base + 0x8200, 28);

quoted

-
   	hws[IMX8MQ_CLK_A53_DIV] =

imx_clk_hw_divider2("arm_a53_div", "arm_a53_cg", base + 0x8000, 0, 3);

quoted

-	hws[IMX8MQ_CLK_M4_DIV] = imx_clk_hw_divider2("arm_m4_div",

"arm_m4_cg", base + 0x8080, 0, 3);

quoted

-	hws[IMX8MQ_CLK_VPU_DIV] = imx_clk_hw_divider2("vpu_div",

"vpu_cg", base + 0x8100, 0, 3);

quoted

-	hws[IMX8MQ_CLK_GPU_CORE_DIV] =

imx_clk_hw_divider2("gpu_core_div", "gpu_core_cg", base + 0x8180, 0, 3);

quoted

-	hws[IMX8MQ_CLK_GPU_SHADER_DIV] =

imx_clk_hw_divider2("gpu_shader_div", "gpu_shader_cg", base + 0x8200, 0,
3);

quoted

+
+	hws[IMX8MQ_CLK_M4_DIV] =

imx8m_clk_hw_composite_core("arm_m4_div", imx8mq_arm_m4_sels, base
+ 0x8080);

quoted

+	hws[IMX8MQ_CLK_VPU_DIV] =

imx8m_clk_hw_composite_core("vpu_div", imx8mq_vpu_sels, base +
0x8100);

quoted

+	hws[IMX8MQ_CLK_GPU_CORE_DIV] =

imx8m_clk_hw_composite_core("gpu_core_div", imx8mq_gpu_core_sels,
base + 0x8180);

quoted

+	hws[IMX8MQ_CLK_GPU_SHADER_DIV] =
+imx8m_clk_hw_composite("gpu_shader_div", imx8mq_gpu_shader_sels,

base

quoted

++ 0x8200);

quoted

+	/* For DTS which still assign parents for gpu core src clk */
+	hws[IMX8MQ_CLK_GPU_CORE_SRC] =

hws[IMX8MQ_CLK_GPU_CORE_DIV];

quoted

+	hws[IMX8MQ_CLK_GPU_SHADER_SRC] =

hws[IMX8MQ_CLK_GPU_SHADER_DIV];

Why not assign to all the old clocks?

Are those clocks expect the GPU ones needed?

Currently only the gpu clocks are needed, others are not used in dts.

For dts update to use the SRC clocks should be avoided in future for Linux,
DIV clocks should be used.

How do you think?

In theory backwards compatibility should be supported at the level of 
"DT bindings", not just hacked for a particular DT. In this case it's 
easy to add aliases for everything, just slightly verbose.

It might also make sense to add new defines for the composite clock so 
that IMX8MQ_CLK_GPU_CORE looks like IMX8MQ_CLK_DSI_CORE and all the _SRC 
_DIV _CG stuff is just aliases to the unsuffixed composite.

--
Regards,
Leonard

_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel