Skip to contentSkip to conversation

remove the "cpu_clk" from the GXBB/GXL/GXM driver

PATCH Superseded Martin Blumenstingl 2 participants 8 messages Apr 1, 2017 – May 15, 2017

Subsystems (4)
Device tree bindings Arm/Amlogic Meson SoC support Common clk Arm/Amlogic Meson SoC clock framework
lore
Share & tools
lore

Martin Blumenstingl

author reviewer
#0 lore raw

cover remove the "cpu_clk" from the GXBB/GXL/GXM driver

This is the updated version of my RFC patch from [0].

The quick summary for this patch is that the "cpu_clk" seems to have been copied from the Meson8b clock driver when the GXBB clock driver was initially added. However, on GXBB (and the other GX SoCs) the actual CPU clock is provided by a SCPI DVFS clock.
More details can be found in the patch description itself.

This was tested on a Khadas VIM board (GXL S905X).

Changes since the RFC version:
- rebased to the "clk-meson" branch (e65ae3fb97b4 "dt-bindings: clock:
gxbb-clkc: Add GXL compatible variant") and Jerome's audio clock patches (in version 2: [1])
- remove the now unused cpu_div_table (which was left over in the RFC version)
- slightly updated the comment for the now unused clock ID 1 in drivers/clk/meson/gxbb.h

[0] https://patchwork.kernel.org/patch/9644993/
[1] http://lists.infradead.org/pipermail/linux-amlogic/2017-March/003200.html

Martin Blumenstingl (1):
clk: meson: gxbb: remove the "cpu_clk" from the GXBB and GXL driver

 drivers/clk/meson/gxbb.c              | 64 ++---------------------------------
 drivers/clk/meson/gxbb.h              |  2 +-
 include/dt-bindings/clock/gxbb-clkc.h |  1 -
 3 files changed, 4 insertions(+), 63 deletions(-)
Signature · 1 line
2.12.1
author reviewer
#1 lore raw

PATCH 1/1 clk: meson: gxbb: remove the "cpu_clk" from the GXBB and GXL driver

It seems that the "cpu_clk" was carried over from the meson8b clock controller driver. On Meson GX (GXBB/GXL/GXM) the registers which are used by the cpu_clk have a different purpose (in other words: they don't control the CPU clock anymore). HHI_SYS_CPU_CLK_CNTL1 bits 31:24 are reserved according to the public S905 datasheet, while bit 23 is the "A53_trace_clk_DIS" gate (which according to the datasheet should only be used in case a silicon bug is discovered) and bits 22:20 are a divider (A53_trace_clk). The meson clk-cpu code however expects that bits 28:20 are reserved for a divider (according to the public S805 datasheet this "SCALE_DIV: This value represents an N+1 divider of the input clock.").

The CPU clock on Meson GX SoCs is provided by the SCPI DVFS clock driver instead. Two examples from a Meson GXL S905X SoC:
- vcpu (SCPI DVFS clock 0) rate: 1000000000 / cpu_clk rate: 708000000
- vcpu (SCPI DVFS clock 0) rate: 1512000000 / cpu_clk rate: 708000000

Unfortunately the CLKID_CPUCLK was already exported (but is currently not used) to DT. Due to the removal of this clock definition there is now a hole in the clk_hw_onecell_data (which is not a problem because this case is already handled in gxbb_clkc_probe).

Signed-off-by Martin Blumenstingl <[email protected]>
---
 drivers/clk/meson/gxbb.c              | 64 ++---------------------------------
 drivers/clk/meson/gxbb.h              |  2 +-
 include/dt-bindings/clock/gxbb-clkc.h |  1 -
 3 files changed, 4 insertions(+), 63 deletions(-)
drivers/clk/meson/gxbb.c +3 -61
index ad5f027af1a2..7cf88ca9bdce 100644
@@ -278,20 +278,6 @@ static const struct pll_rate_table gxl_gp0_pll_rate_table[] = {
 	{ /* sentinel */ },
 };
 
-static const struct clk_div_table cpu_div_table[] = {
-	{ .val = 1, .div = 1 },
-	{ .val = 2, .div = 2 },
-	{ .val = 3, .div = 3 },
-	{ .val = 2, .div = 4 },
-	{ .val = 3, .div = 6 },
-	{ .val = 4, .div = 8 },
-	{ .val = 5, .div = 10 },
-	{ .val = 6, .div = 12 },
-	{ .val = 7, .div = 14 },
-	{ .val = 8, .div = 16 },
-	{ /* sentinel */ },
-};
-
 static struct meson_clk_pll gxbb_fixed_pll = {
 	.m = {
 		.reg_off = HHI_MPLL_CNTL,

@@ -612,21 +598,10 @@ static struct meson_clk_mpll gxbb_mpll2 = {
 };
 
 /*
- * FIXME cpu clocks and the legacy composite clocks (e.g. clk81) are both PLL
- * post-dividers and should be modeled with their respective PLLs via the
- * forthcoming coordinated clock rates feature
+ * FIXME The legacy composite clocks (e.g. clk81) are both PLL post-dividers
+ * and should be modeled with their respective PLLs via the forthcoming
+ * coordinated clock rates feature
  */
-static struct meson_clk_cpu gxbb_cpu_clk = {
-	.reg_off = HHI_SYS_CPU_CLK_CNTL1,
-	.div_table = cpu_div_table,
-	.clk_nb.notifier_call = meson_clk_cpu_notifier_cb,
-	.hw.init = &(struct clk_init_data){
-		.name = "cpu_clk",
-		.ops = &meson_clk_cpu_ops,
-		.parent_names = (const char *[]){ "sys_pll" },
-		.num_parents = 1,
-	},
-};
 
 static u32 mux_table_clk81[]	= { 6, 5, 7 };
 

@@ -1045,7 +1020,6 @@ static MESON_GATE(gxbb_ao_i2c, HHI_GCLK_AO, 4);
 static struct clk_hw_onecell_data gxbb_hw_onecell_data = {
 	.hws = {
 		[CLKID_SYS_PLL]		    = &gxbb_sys_pll.hw,
-		[CLKID_CPUCLK]		    = &gxbb_cpu_clk.hw,
 		[CLKID_HDMI_PLL]	    = &gxbb_hdmi_pll.hw,
 		[CLKID_FIXED_PLL]	    = &gxbb_fixed_pll.hw,
 		[CLKID_FCLK_DIV2]	    = &gxbb_fclk_div2.hw,

@@ -1165,7 +1139,6 @@ static struct clk_hw_onecell_data gxbb_hw_onecell_data = {
 static struct clk_hw_onecell_data gxl_hw_onecell_data = {
 	.hws = {
 		[CLKID_SYS_PLL]		    = &gxbb_sys_pll.hw,
-		[CLKID_CPUCLK]		    = &gxbb_cpu_clk.hw,
 		[CLKID_HDMI_PLL]	    = &gxbb_hdmi_pll.hw,
 		[CLKID_FIXED_PLL]	    = &gxbb_fixed_pll.hw,
 		[CLKID_FCLK_DIV2]	    = &gxbb_fclk_div2.hw,

@@ -1430,7 +1403,6 @@ struct clkc_data {
 	unsigned int clk_dividers_count;
 	struct meson_clk_audio_divider *const *clk_audio_dividers;
 	unsigned int clk_audio_dividers_count;
-	struct meson_clk_cpu *cpu_clk;
 	struct clk_hw_onecell_data *hw_onecell_data;
 };
 

@@ -1447,7 +1419,6 @@ static const struct clkc_data gxbb_clkc_data = {
 	.clk_dividers_count = ARRAY_SIZE(gxbb_clk_dividers),
 	.clk_audio_dividers = gxbb_audio_dividers,
 	.clk_audio_dividers_count = ARRAY_SIZE(gxbb_audio_dividers),
-	.cpu_clk = &gxbb_cpu_clk,
 	.hw_onecell_data = &gxbb_hw_onecell_data,
 };
 

@@ -1464,7 +1435,6 @@ static const struct clkc_data gxl_clkc_data = {
 	.clk_dividers_count = ARRAY_SIZE(gxbb_clk_dividers),
 	.clk_audio_dividers = gxbb_audio_dividers,
 	.clk_audio_dividers_count = ARRAY_SIZE(gxbb_audio_dividers),
-	.cpu_clk = &gxbb_cpu_clk,
 	.hw_onecell_data = &gxl_hw_onecell_data,
 };
 

@@ -1479,8 +1449,6 @@ static int gxbb_clkc_probe(struct platform_device *pdev)
 	const struct clkc_data *clkc_data;
 	void __iomem *clk_base;
 	int ret, clkid, i;
-	struct clk_hw *parent_hw;
-	struct clk *parent_clk;
 	struct device *dev = &pdev->dev;
 
 	clkc_data = of_device_get_match_data(&pdev->dev);

@@ -1502,9 +1470,6 @@ static int gxbb_clkc_probe(struct platform_device *pdev)
 	for (i = 0; i < clkc_data->clk_mplls_count; i++)
 		clkc_data->clk_mplls[i]->base = clk_base;
 
-	/* Populate the base address for CPU clk */
-	clkc_data->cpu_clk->base = clk_base;
-
 	/* Populate base address for gates */
 	for (i = 0; i < clkc_data->clk_gates_count; i++)
 		clkc_data->clk_gates[i]->reg = clk_base +

@@ -1538,29 +1503,6 @@ static int gxbb_clkc_probe(struct platform_device *pdev)
 			goto iounmap;
 	}
 
-	/*
-	 * Register CPU clk notifier
-	 *
-	 * FIXME this is wrong for a lot of reasons. First, the muxes should be
-	 * struct clk_hw objects. Second, we shouldn't program the muxes in
-	 * notifier handlers. The tricky programming sequence will be handled
-	 * by the forthcoming coordinated clock rates mechanism once that
-	 * feature is released.
-	 *
-	 * Furthermore, looking up the parent this way is terrible. At some
-	 * point we will stop allocating a default struct clk when registering
-	 * a new clk_hw, and this hack will no longer work. Releasing the ccr
-	 * feature before that time solves the problem :-)
-	 */
-	parent_hw = clk_hw_get_parent(&clkc_data->cpu_clk->hw);
-	parent_clk = parent_hw->clk;
-	ret = clk_notifier_register(parent_clk, &clkc_data->cpu_clk->clk_nb);
-	if (ret) {
-		pr_err("%s: failed to register clock notifier for cpu_clk\n",
-				__func__);
-		goto iounmap;
-	}
-
 	return of_clk_add_hw_provider(dev->of_node, of_clk_hw_onecell_get,
 			clkc_data->hw_onecell_data);
drivers/clk/meson/gxbb.h +1 -1
index 17c6aef033ff..36330c2d4433 100644
@@ -171,7 +171,7 @@
  * to be exposed to client nodes in DT: include/dt-bindings/clock/gxbb-clkc.h
  */
 #define CLKID_SYS_PLL		  0
-/* CLKID_CPUCLK */
+/* ID 1 is unused (it was used by the non-existing CLKID_CPUCLK before) */
 /* CLKID_HDMI_PLL */
 #define CLKID_FIXED_PLL		  3
 /* CLKID_FCLK_DIV2 */
include/dt-bindings/clock/gxbb-clkc.h +0 -1
index 4516bc4253b5..54faf83a4851 100644
@@ -5,7 +5,6 @@
 #ifndef __GXBB_CLKC_H
 #define __GXBB_CLKC_H
 
-#define CLKID_CPUCLK		1
 #define CLKID_HDMI_PLL		2
 #define CLKID_FCLK_DIV2		4
 #define CLKID_FCLK_DIV3		5
Signature · 1 line
2.12.1
maintainer
#2 lore raw

PATCH 1/1 clk: meson: gxbb: remove the "cpu_clk" from the GXBB and GXL driver

Quoted 17 lines · Martin Blumenstingl

It seems that the "cpu_clk" was carried over from the meson8b clock controller driver. On Meson GX (GXBB/GXL/GXM) the registers which are used by the cpu_clk have a different purpose (in other words: they don't control the CPU clock anymore). HHI_SYS_CPU_CLK_CNTL1 bits 31:24 are reserved according to the public S905 datasheet, while bit 23 is the "A53_trace_clk_DIS" gate (which according to the datasheet should only be used in case a silicon bug is discovered) and bits 22:20 are a divider (A53_trace_clk). The meson clk-cpu code however expects that bits 28:20 are reserved for a divider (according to the public S805 datasheet this "SCALE_DIV: This value represents an N+1 divider of the input clock.").

The CPU clock on Meson GX SoCs is provided by the SCPI DVFS clock driver instead. Two examples from a Meson GXL S905X SoC:
- vcpu (SCPI DVFS clock 0) rate: 1000000000 / cpu_clk rate: 708000000
- vcpu (SCPI DVFS clock 0) rate: 1512000000 / cpu_clk rate: 708000000

Quoted 6 lines

Unfortunately the CLKID_CPUCLK was already exported (but is currently not used) to DT. Due to the removal of this clock definition there is now a hole in the clk_hw_onecell_data (which is not a problem because this case is already handled in gxbb_clkc_probe).

Signed-off-by Martin Blumenstingl <[email protected]>

Looks good to me.
I'll wait for the Ack of one of the DT maintainers to apply it. Thx Martin

Cheers

Quoted 177 lines

---
?drivers/clk/meson/gxbb.c??????????????| 64 ++------------------------------
---
?drivers/clk/meson/gxbb.h??????????????|??2 +- ?include/dt-bindings/clock/gxbb-clkc.h |??1 -
?3 files changed, 4 insertions(+), 63 deletions(-)

drivers/clk/meson/gxbb.c +0 -0
index ad5f027af1a2..7cf88ca9bdce 100644
@@ -278,20 +278,6 @@ static const struct pll_rate_table
 gxl_gp0_pll_rate_table[] = {
 ?	{ /* sentinel */ },
 ?};

?

drivers/clk/meson/gxbb.c +0 -14
-static const struct clk_div_table cpu_div_table[] = {
-	{ .val = 1, .div = 1 },
-	{ .val = 2, .div = 2 },
-	{ .val = 3, .div = 3 },
-	{ .val = 2, .div = 4 },
-	{ .val = 3, .div = 6 },
-	{ .val = 4, .div = 8 },
-	{ .val = 5, .div = 10 },
-	{ .val = 6, .div = 12 },
-	{ .val = 7, .div = 14 },
-	{ .val = 8, .div = 16 },
-	{ /* sentinel */ },
-};
-

?static struct meson_clk_pll gxbb_fixed_pll = {
? .m = {
? .reg_off = HHI_MPLL_CNTL,

drivers/clk/meson/gxbb.c +0 -0
@@ -612,21 +598,10 @@ static struct meson_clk_mpll gxbb_mpll2 = {
 ?};

?
?/*
- * FIXME cpu clocks and the legacy composite clocks (e.g. clk81) are both PLL
- * post-dividers and should be modeled with their respective PLLs via the
- * forthcoming coordinated clock rates feature
+ * FIXME The legacy composite clocks (e.g. clk81) are both PLL post-dividers
+ * and should be modeled with their respective PLLs via the forthcoming
+ * coordinated clock rates feature
? */

drivers/clk/meson/gxbb.c +0 -11
-static struct meson_clk_cpu gxbb_cpu_clk = {
-	.reg_off = HHI_SYS_CPU_CLK_CNTL1,
-	.div_table = cpu_div_table,
-	.clk_nb.notifier_call = meson_clk_cpu_notifier_cb,
-	.hw.init = &(struct clk_init_data){
-		.name = "cpu_clk",
-		.ops = &meson_clk_cpu_ops,
-		.parent_names = (const char *[]){ "sys_pll" },
-		.num_parents = 1,
-	},
-};

?
?static u32 mux_table_clk81[] = { 6, 5, 7 };
?

drivers/clk/meson/gxbb.c +0 -2
@@ -1045,7 +1020,6 @@ static MESON_GATE(gxbb_ao_i2c, HHI_GCLK_AO, 4);
 ?static struct clk_hw_onecell_data gxbb_hw_onecell_data = {
 ?	.hws = {
 ?		[CLKID_SYS_PLL]		????= &gxbb_sys_pll.hw,
-		[CLKID_CPUCLK]		????= &gxbb_cpu_clk.hw,
 ?		[CLKID_HDMI_PLL]	????= &gxbb_hdmi_pll.hw,
 ?		[CLKID_FIXED_PLL]	????= &gxbb_fixed_pll.hw,
 ?		[CLKID_FCLK_DIV2]	????= &gxbb_fclk_div2.hw,

@@ -1165,7 +1139,6 @@ static struct clk_hw_onecell_data gxbb_hw_onecell_data =
 {
 ?static struct clk_hw_onecell_data gxl_hw_onecell_data = {
 ?	.hws = {
 ?		[CLKID_SYS_PLL]		????= &gxbb_sys_pll.hw,
-		[CLKID_CPUCLK]		????= &gxbb_cpu_clk.hw,
 ?		[CLKID_HDMI_PLL]	????= &gxbb_hdmi_pll.hw,
 ?		[CLKID_FIXED_PLL]	????= &gxbb_fixed_pll.hw,

? [CLKID_FCLK_DIV2] ????= &gxbb_fclk_div2.hw,

drivers/clk/meson/gxbb.c +0 -1
@@ -1430,7 +1403,6 @@ struct clkc_data {
 ?	unsigned int clk_dividers_count;
 ?	struct meson_clk_audio_divider *const *clk_audio_dividers;
 ?	unsigned int clk_audio_dividers_count;
-	struct meson_clk_cpu *cpu_clk;
 ?	struct clk_hw_onecell_data *hw_onecell_data;
 ?};

?

drivers/clk/meson/gxbb.c +0 -1
@@ -1447,7 +1419,6 @@ static const struct clkc_data gxbb_clkc_data = {
 ?	.clk_dividers_count = ARRAY_SIZE(gxbb_clk_dividers),
 ?	.clk_audio_dividers = gxbb_audio_dividers,
 ?	.clk_audio_dividers_count = ARRAY_SIZE(gxbb_audio_dividers),
-	.cpu_clk = &gxbb_cpu_clk,
 ?	.hw_onecell_data = &gxbb_hw_onecell_data,
 ?};

?

drivers/clk/meson/gxbb.c +0 -1
@@ -1464,7 +1435,6 @@ static const struct clkc_data gxl_clkc_data = {
 ?	.clk_dividers_count = ARRAY_SIZE(gxbb_clk_dividers),
 ?	.clk_audio_dividers = gxbb_audio_dividers,
 ?	.clk_audio_dividers_count = ARRAY_SIZE(gxbb_audio_dividers),
-	.cpu_clk = &gxbb_cpu_clk,
 ?	.hw_onecell_data = &gxl_hw_onecell_data,
 ?};

?

drivers/clk/meson/gxbb.c +0 -2
@@ -1479,8 +1449,6 @@ static int gxbb_clkc_probe(struct platform_device *pdev)
 ?	const struct clkc_data *clkc_data;
 ?	void __iomem *clk_base;
 ?	int ret, clkid, i;
-	struct clk_hw *parent_hw;
-	struct clk *parent_clk;
 ?	struct device *dev = &pdev->dev;

?
? clkc_data = of_device_get_match_data(&pdev->dev);

drivers/clk/meson/gxbb.c +0 -0
@@ -1502,9 +1470,6 @@ static int gxbb_clkc_probe(struct platform_device *pdev)
 ?	for (i = 0; i < clkc_data->clk_mplls_count; i++)
 ?		clkc_data->clk_mplls[i]->base = clk_base;

?

drivers/clk/meson/gxbb.c +0 -3
-	/* Populate the base address for CPU clk */
-	clkc_data->cpu_clk->base = clk_base;
-

? /* Populate base address for gates */
? for (i = 0; i < clkc_data->clk_gates_count; i++)
? clkc_data->clk_gates[i]->reg = clk_base +

drivers/clk/meson/gxbb.c +0 -0
@@ -1538,29 +1503,6 @@ static int gxbb_clkc_probe(struct platform_device
 *pdev)
 ?			goto iounmap;
 ?	}

?

drivers/clk/meson/gxbb.c +0 -4
-	/*
-	?* Register CPU clk notifier
-	?*
-	?* FIXME this is wrong for a lot of reasons. First, the muxes should

be

drivers/clk/meson/gxbb.c +0 -7
-	?* struct clk_hw objects. Second, we shouldn't program the muxes in
-	?* notifier handlers. The tricky programming sequence will be handled
-	?* by the forthcoming coordinated clock rates mechanism once that
-	?* feature is released.
-	?*
-	?* Furthermore, looking up the parent this way is terrible. At some
-	?* point we will stop allocating a default struct clk when

registering

drivers/clk/meson/gxbb.c +0 -12
-	?* a new clk_hw, and this hack will no longer work. Releasing the ccr
-	?* feature before that time solves the problem :-)
-	?*/
-	parent_hw = clk_hw_get_parent(&clkc_data->cpu_clk->hw);
-	parent_clk = parent_hw->clk;
-	ret = clk_notifier_register(parent_clk, &clkc_data->cpu_clk->clk_nb);
-	if (ret) {
-		pr_err("%s: failed to register clock notifier for cpu_clk\n",
-				__func__);
-		goto iounmap;
-	}
-

? return of_clk_add_hw_provider(dev->of_node, of_clk_hw_onecell_get,
? clkc_data->hw_onecell_data);
?

drivers/clk/meson/gxbb.h +0 -0
index 17c6aef033ff..36330c2d4433 100644
@@ -171,7 +171,7 @@

? * to be exposed to client nodes in DT: include/dt-bindings/clock/gxbb-clkc.h ? */
?#define CLKID_SYS_PLL ??0

drivers/clk/meson/gxbb.h +1 -1
-/* CLKID_CPUCLK */
+/* ID 1 is unused (it was used by the non-existing CLKID_CPUCLK before) */

?/* CLKID_HDMI_PLL */
?#define CLKID_FIXED_PLL ??3
?/* CLKID_FCLK_DIV2 */

include/dt- renamed from include/dt-bindings/clock/gxbb-clkc.h +0 -0

bindings/clock/gxbb-clkc.h
index 4516bc4253b5..54faf83a4851 100644

include/dt-bindings/clock/gxbb-clkc.h +0 -0
@@ -5,7 +5,6 @@

?#ifndef __GXBB_CLKC_H
?#define __GXBB_CLKC_H
?
-#define CLKID_CPUCLK 1
?#define CLKID_HDMI_PLL 2
?#define CLKID_FCLK_DIV2 4
?#define CLKID_FCLK_DIV3 5

author reviewer
#3 lore raw

v2 cover remove the "cpu_clk" from the GXBB/GXL/GXM driver

In reply to Martin Blumenstingl

The quick summary for this patch is that the "cpu_clk" seems to have been copied from the Meson8b clock driver when the GXBB clock driver was initially added. However, on GXBB (and the other GX SoCs) the actual CPU clock is provided by a SCPI DVFS clock.
More details can be found in the description of the actual patch (#2).

This was tested on a Khadas VIM board (GXL S905X).

Changes since v1:
- split the patch into two patches: one which un-exports the clock ID first and a second which does the actual cleanup. This was done based on the recent discussion that new clocks should be added first and then exported in a second patch (this patch does the same, just the other way round: un-exporting first, then removing the actual clock)

Changes since the RFC version from [0]:
- rebased to the "clk-meson" branch (e65ae3fb97b4 "dt-bindings: clock:
gxbb-clkc: Add GXL compatible variant") and Jerome's audio clock patches (in version 2: [1])
- remove the now unused cpu_div_table (which was left over in the RFC version)
- slightly updated the comment for the now unused clock ID 1 in drivers/clk/meson/gxbb.h

[0] https://patchwork.kernel.org/patch/9644993/
[1] http://lists.infradead.org/pipermail/linux-amlogic/2017-March/003200.html

Martin Blumenstingl (2):
clk: meson-gxbb: un-export the CPU clock
clk: meson: gxbb: remove the "cpu_clk" from the GXBB and GXL driver

 drivers/clk/meson/gxbb.c              | 64 ++---------------------------------
 drivers/clk/meson/gxbb.h              |  2 +-
 include/dt-bindings/clock/gxbb-clkc.h |  1 -
 3 files changed, 4 insertions(+), 63 deletions(-)
Signature · 1 line
2.12.2
author reviewer
#4 lore raw

PATCH v2 1/2 clk: meson-gxbb: un-export the CPU clock

The CPU clock defined in the Meson GX clock driver is actually a left-over from the Meson8b clock controller. Un-export the clock so we can remove it from the driver.

Signed-off-by Martin Blumenstingl <[email protected]>
---
 drivers/clk/meson/gxbb.h              | 2 +-
 include/dt-bindings/clock/gxbb-clkc.h | 1 -
 2 files changed, 1 insertion(+), 2 deletions(-)
drivers/clk/meson/gxbb.h +1 -1
index 3e5f375af7b6..c15bbce8c585 100644
@@ -171,7 +171,7 @@
  * to be exposed to client nodes in DT: include/dt-bindings/clock/gxbb-clkc.h
  */
 #define CLKID_SYS_PLL		  0
-/* CLKID_CPUCLK */
+#define CLKID_CPUCLK		  1
 /* CLKID_HDMI_PLL */
 #define CLKID_FIXED_PLL		  3
 /* CLKID_FCLK_DIV2 */
include/dt-bindings/clock/gxbb-clkc.h +0 -1
index 3190e30b9398..46ff81fbc02e 100644
@@ -5,7 +5,6 @@
 #ifndef __GXBB_CLKC_H
 #define __GXBB_CLKC_H
 
-#define CLKID_CPUCLK		1
 #define CLKID_HDMI_PLL		2
 #define CLKID_FCLK_DIV2		4
 #define CLKID_FCLK_DIV3		5
Signature · 1 line
2.12.2
maintainer Applied
#5 lore raw

PATCH v2 1/2 clk: meson-gxbb: un-export the CPU clock

Quoted 29 lines · Martin Blumenstingl

The CPU clock defined in the Meson GX clock driver is actually a left-over from the Meson8b clock controller. Un-export the clock so we can remove it from the driver.

Signed-off-by Martin Blumenstingl <[email protected]>

---
?drivers/clk/meson/gxbb.h??????????????| 2 +-
?include/dt-bindings/clock/gxbb-clkc.h | 1 -
?2 files changed, 1 insertion(+), 2 deletions(-)

drivers/clk/meson/gxbb.h +0 -0
index 3e5f375af7b6..c15bbce8c585 100644
@@ -171,7 +171,7 @@

? * to be exposed to client nodes in DT: include/dt-bindings/clock/gxbb-clkc.h ? */
?#define CLKID_SYS_PLL ??0

drivers/clk/meson/gxbb.h +1 -1
-/* CLKID_CPUCLK */
+#define CLKID_CPUCLK		??1

?/* CLKID_HDMI_PLL */
?#define CLKID_FIXED_PLL ??3
?/* CLKID_FCLK_DIV2 */

include/dt- renamed from include/dt-bindings/clock/gxbb-clkc.h +0 -0

bindings/clock/gxbb-clkc.h
index 3190e30b9398..46ff81fbc02e 100644

include/dt-bindings/clock/gxbb-clkc.h +0 -0
@@ -5,7 +5,6 @@
Quoted 7 lines

?#ifndef __GXBB_CLKC_H
?#define __GXBB_CLKC_H
?
-#define CLKID_CPUCLK 1
?#define CLKID_HDMI_PLL 2
?#define CLKID_FCLK_DIV2 4
?#define CLKID_FCLK_DIV3 5

Applied. Thanks
Jerome

author reviewer
#6 lore raw

PATCH v2 2/2 clk: meson: gxbb: remove the "cpu_clk" from the GXBB and GXL driver

In reply to Martin Blumenstingl

It seems that the "cpu_clk" was carried over from the meson8b clock controller driver. On Meson GX (GXBB/GXL/GXM) the registers which are used by the cpu_clk have a different purpose (in other words: they don't control the CPU clock anymore). HHI_SYS_CPU_CLK_CNTL1 bits 31:24 are reserved according to the public S905 datasheet, while bit 23 is the "A53_trace_clk_DIS" gate (which according to the datasheet should only be used in case a silicon bug is discovered) and bits 22:20 are a divider (A53_trace_clk). The meson clk-cpu code however expects that bits 28:20 are reserved for a divider (according to the public S805 datasheet this "SCALE_DIV: This value represents an N+1 divider of the input clock.").

The CPU clock on Meson GX SoCs is provided by the SCPI DVFS clock driver instead. Two examples from a Meson GXL S905X SoC:
- vcpu (SCPI DVFS clock 0) rate: 1000000000 / cpu_clk rate: 708000000
- vcpu (SCPI DVFS clock 0) rate: 1512000000 / cpu_clk rate: 708000000

Unfortunately the CLKID_CPUCLK was already exported (but is currently not used) to DT. Due to the removal of this clock definition there is now a hole in the clk_hw_onecell_data (which is not a problem because this case is already handled in gxbb_clkc_probe).

Signed-off-by Martin Blumenstingl <[email protected]>
---
 drivers/clk/meson/gxbb.c | 64 +++---------------------------------------------
 drivers/clk/meson/gxbb.h |  2 +-
 2 files changed, 4 insertions(+), 62 deletions(-)
drivers/clk/meson/gxbb.c +3 -61
index 75197664a7ee..25d05fc2d045 100644
@@ -278,20 +278,6 @@ static const struct pll_rate_table gxl_gp0_pll_rate_table[] = {
 	{ /* sentinel */ },
 };
 
-static const struct clk_div_table cpu_div_table[] = {
-	{ .val = 1, .div = 1 },
-	{ .val = 2, .div = 2 },
-	{ .val = 3, .div = 3 },
-	{ .val = 2, .div = 4 },
-	{ .val = 3, .div = 6 },
-	{ .val = 4, .div = 8 },
-	{ .val = 5, .div = 10 },
-	{ .val = 6, .div = 12 },
-	{ .val = 7, .div = 14 },
-	{ .val = 8, .div = 16 },
-	{ /* sentinel */ },
-};
-
 static struct meson_clk_pll gxbb_fixed_pll = {
 	.m = {
 		.reg_off = HHI_MPLL_CNTL,

@@ -612,21 +598,10 @@ static struct meson_clk_mpll gxbb_mpll2 = {
 };
 
 /*
- * FIXME cpu clocks and the legacy composite clocks (e.g. clk81) are both PLL
- * post-dividers and should be modeled with their respective PLLs via the
- * forthcoming coordinated clock rates feature
+ * FIXME The legacy composite clocks (e.g. clk81) are both PLL post-dividers
+ * and should be modeled with their respective PLLs via the forthcoming
+ * coordinated clock rates feature
  */
-static struct meson_clk_cpu gxbb_cpu_clk = {
-	.reg_off = HHI_SYS_CPU_CLK_CNTL1,
-	.div_table = cpu_div_table,
-	.clk_nb.notifier_call = meson_clk_cpu_notifier_cb,
-	.hw.init = &(struct clk_init_data){
-		.name = "cpu_clk",
-		.ops = &meson_clk_cpu_ops,
-		.parent_names = (const char *[]){ "sys_pll" },
-		.num_parents = 1,
-	},
-};
 
 static u32 mux_table_clk81[]	= { 6, 5, 7 };
 

@@ -939,7 +914,6 @@ static MESON_GATE(gxbb_ao_i2c, HHI_GCLK_AO, 4);
 static struct clk_hw_onecell_data gxbb_hw_onecell_data = {
 	.hws = {
 		[CLKID_SYS_PLL]		    = &gxbb_sys_pll.hw,
-		[CLKID_CPUCLK]		    = &gxbb_cpu_clk.hw,
 		[CLKID_HDMI_PLL]	    = &gxbb_hdmi_pll.hw,
 		[CLKID_FIXED_PLL]	    = &gxbb_fixed_pll.hw,
 		[CLKID_FCLK_DIV2]	    = &gxbb_fclk_div2.hw,

@@ -1052,7 +1026,6 @@ static struct clk_hw_onecell_data gxbb_hw_onecell_data = {
 static struct clk_hw_onecell_data gxl_hw_onecell_data = {
 	.hws = {
 		[CLKID_SYS_PLL]		    = &gxbb_sys_pll.hw,
-		[CLKID_CPUCLK]		    = &gxbb_cpu_clk.hw,
 		[CLKID_HDMI_PLL]	    = &gxbb_hdmi_pll.hw,
 		[CLKID_FIXED_PLL]	    = &gxbb_fixed_pll.hw,
 		[CLKID_FCLK_DIV2]	    = &gxbb_fclk_div2.hw,

@@ -1298,7 +1271,6 @@ struct clkc_data {
 	unsigned int clk_muxes_count;
 	struct clk_divider *const *clk_dividers;
 	unsigned int clk_dividers_count;
-	struct meson_clk_cpu *cpu_clk;
 	struct clk_hw_onecell_data *hw_onecell_data;
 };
 

@@ -1313,7 +1285,6 @@ static const struct clkc_data gxbb_clkc_data = {
 	.clk_muxes_count = ARRAY_SIZE(gxbb_clk_muxes),
 	.clk_dividers = gxbb_clk_dividers,
 	.clk_dividers_count = ARRAY_SIZE(gxbb_clk_dividers),
-	.cpu_clk = &gxbb_cpu_clk,
 	.hw_onecell_data = &gxbb_hw_onecell_data,
 };
 

@@ -1328,7 +1299,6 @@ static const struct clkc_data gxl_clkc_data = {
 	.clk_muxes_count = ARRAY_SIZE(gxbb_clk_muxes),
 	.clk_dividers = gxbb_clk_dividers,
 	.clk_dividers_count = ARRAY_SIZE(gxbb_clk_dividers),
-	.cpu_clk = &gxbb_cpu_clk,
 	.hw_onecell_data = &gxl_hw_onecell_data,
 };
 

@@ -1343,8 +1313,6 @@ static int gxbb_clkc_probe(struct platform_device *pdev)
 	const struct clkc_data *clkc_data;
 	void __iomem *clk_base;
 	int ret, clkid, i;
-	struct clk_hw *parent_hw;
-	struct clk *parent_clk;
 	struct device *dev = &pdev->dev;
 
 	clkc_data = of_device_get_match_data(&pdev->dev);

@@ -1366,9 +1334,6 @@ static int gxbb_clkc_probe(struct platform_device *pdev)
 	for (i = 0; i < clkc_data->clk_mplls_count; i++)
 		clkc_data->clk_mplls[i]->base = clk_base;
 
-	/* Populate the base address for CPU clk */
-	clkc_data->cpu_clk->base = clk_base;
-
 	/* Populate base address for gates */
 	for (i = 0; i < clkc_data->clk_gates_count; i++)
 		clkc_data->clk_gates[i]->reg = clk_base +

@@ -1394,29 +1359,6 @@ static int gxbb_clkc_probe(struct platform_device *pdev)
 			goto iounmap;
 	}
 
-	/*
-	 * Register CPU clk notifier
-	 *
-	 * FIXME this is wrong for a lot of reasons. First, the muxes should be
-	 * struct clk_hw objects. Second, we shouldn't program the muxes in
-	 * notifier handlers. The tricky programming sequence will be handled
-	 * by the forthcoming coordinated clock rates mechanism once that
-	 * feature is released.
-	 *
-	 * Furthermore, looking up the parent this way is terrible. At some
-	 * point we will stop allocating a default struct clk when registering
-	 * a new clk_hw, and this hack will no longer work. Releasing the ccr
-	 * feature before that time solves the problem :-)
-	 */
-	parent_hw = clk_hw_get_parent(&clkc_data->cpu_clk->hw);
-	parent_clk = parent_hw->clk;
-	ret = clk_notifier_register(parent_clk, &clkc_data->cpu_clk->clk_nb);
-	if (ret) {
-		pr_err("%s: failed to register clock notifier for cpu_clk\n",
-				__func__);
-		goto iounmap;
-	}
-
 	return of_clk_add_hw_provider(dev->of_node, of_clk_hw_onecell_get,
 			clkc_data->hw_onecell_data);
drivers/clk/meson/gxbb.h +1 -1
index c15bbce8c585..e5ed71e6353a 100644
@@ -171,7 +171,7 @@
  * to be exposed to client nodes in DT: include/dt-bindings/clock/gxbb-clkc.h
  */
 #define CLKID_SYS_PLL		  0
-#define CLKID_CPUCLK		  1
+/* ID 1 is unused (it was used by the non-existing CLKID_CPUCLK before) */
 /* CLKID_HDMI_PLL */
 #define CLKID_FIXED_PLL		  3
 /* CLKID_FCLK_DIV2 */
Signature · 1 line
2.12.2
maintainer Applied
#7 lore raw

PATCH v2 2/2 clk: meson: gxbb: remove the "cpu_clk" from the GXBB and GXL driver

Quoted 17 lines · Martin Blumenstingl

It seems that the "cpu_clk" was carried over from the meson8b clock controller driver. On Meson GX (GXBB/GXL/GXM) the registers which are used by the cpu_clk have a different purpose (in other words: they don't control the CPU clock anymore). HHI_SYS_CPU_CLK_CNTL1 bits 31:24 are reserved according to the public S905 datasheet, while bit 23 is the "A53_trace_clk_DIS" gate (which according to the datasheet should only be used in case a silicon bug is discovered) and bits 22:20 are a divider (A53_trace_clk). The meson clk-cpu code however expects that bits 28:20 are reserved for a divider (according to the public S805 datasheet this "SCALE_DIV: This value represents an N+1 divider of the input clock.").

The CPU clock on Meson GX SoCs is provided by the SCPI DVFS clock driver instead. Two examples from a Meson GXL S905X SoC:
- vcpu (SCPI DVFS clock 0) rate: 1000000000 / cpu_clk rate: 708000000
- vcpu (SCPI DVFS clock 0) rate: 1512000000 / cpu_clk rate: 708000000

Quoted 6 lines

Unfortunately the CLKID_CPUCLK was already exported (but is currently not used) to DT. Due to the removal of this clock definition there is now a hole in the clk_hw_onecell_data (which is not a problem because this case is already handled in gxbb_clkc_probe).

Signed-off-by Martin Blumenstingl <[email protected]>

Applied, Thanks
Jerome