Skip to contentSkip to conversation

clk: ti: Fix missing omap4 mcbsp functional clock and aliases

PATCH In review Tony Lindgren 2 participants 5 messages Oct 11, 2023 – Oct 13, 2023

Subsystems (4)
lore
Share & tools
lore

Tony Lindgren

author maintainer
#0 lore raw

PATCH 1/2 clk: ti: Fix missing omap4 mcbsp functional clock and aliases

We are using a wrong mcbsp functional clock. The interconnect target module driver provided clock for mcbsp is not same as the mcbsp functional clock known as the gfclk main_clk. The mcbsp functional clocks for mcbsp should have been added before we dropped the legacy platform data.

Additionally we are also missing the clock aliases for the clocks used by the audio driver if reparenting is needed. This causes audio driver errors like "CLKS: could not clk_get() prcm_fck" for mcbsp as reported by Andreas. The mcbsp clock aliases too should have been added before we dropped the legacy platform data.

Let's add the clocks and aliases with a single patch to fix the issue.

Fixes 349355ce3a05 ("ARM: OMAP2+: Drop legacy platform data for omap4 mcbsp")
---
 arch/arm/boot/dts/ti/omap/omap4-l4-abe.dtsi | 6 ++++++
 arch/arm/boot/dts/ti/omap/omap4-l4.dtsi     | 2 ++
 drivers/clk/ti/clk-44xx.c                   | 5 +++++
 3 files changed, 13 insertions(+)
arch/arm/boot/dts/ti/omap/omap4-l4-abe.dtsi +6 -0
@@ -109,6 +109,8 @@ mcbsp1: mcbsp@0 {
 				reg = <0x0 0xff>, /* MPU private access */
 				      <0x49022000 0xff>; /* L3 Interconnect */
 				reg-names = "mpu", "dma";
+				clocks = <&abe_clkctrl OMAP4_MCBSP1_CLKCTRL 24>;
+				clock-names = "fck";
 				interrupts = <GIC_SPI 17 IRQ_TYPE_LEVEL_HIGH>;
 				interrupt-names = "common";
 				ti,buffer-size = <128>;
@@ -142,6 +144,8 @@ mcbsp2: mcbsp@0 {
 				reg = <0x0 0xff>, /* MPU private access */
 				      <0x49024000 0xff>; /* L3 Interconnect */
 				reg-names = "mpu", "dma";
+				clocks = <&abe_clkctrl OMAP4_MCBSP2_CLKCTRL 24>;
+				clock-names = "fck";
 				interrupts = <GIC_SPI 22 IRQ_TYPE_LEVEL_HIGH>;
 				interrupt-names = "common";
 				ti,buffer-size = <128>;
@@ -175,6 +179,8 @@ mcbsp3: mcbsp@0 {
 				reg = <0x0 0xff>, /* MPU private access */
 				      <0x49026000 0xff>; /* L3 Interconnect */
 				reg-names = "mpu", "dma";
+				clocks = <&abe_clkctrl OMAP4_MCBSP3_CLKCTRL 24>;
+				clock-names = "fck";
 				interrupts = <GIC_SPI 23 IRQ_TYPE_LEVEL_HIGH>;
 				interrupt-names = "common";
 				ti,buffer-size = <128>;
arch/arm/boot/dts/ti/omap/omap4-l4.dtsi +2 -0
@@ -2043,6 +2043,8 @@ mcbsp4: mcbsp@0 {
 				compatible = "ti,omap4-mcbsp";
 				reg = <0x0 0xff>; /* L4 Interconnect */
 				reg-names = "mpu";
+				clocks = <&l4_per_clkctrl OMAP4_MCBSP4_CLKCTRL 24>;
+				clock-names = "fck";
 				interrupts = <GIC_SPI 16 IRQ_TYPE_LEVEL_HIGH>;
 				interrupt-names = "common";
 				ti,buffer-size = <128>;
drivers/clk/ti/clk-44xx.c +5 -0
@@ -749,9 +749,14 @@ static struct ti_dt_clk omap44xx_clks[] = {
 	DT_CLK(NULL, "mcbsp1_sync_mux_ck", "abe-clkctrl:0028:26"),
 	DT_CLK(NULL, "mcbsp2_sync_mux_ck", "abe-clkctrl:0030:26"),
 	DT_CLK(NULL, "mcbsp3_sync_mux_ck", "abe-clkctrl:0038:26"),
+	DT_CLK("40122000.mcbsp", "prcm_fck", "abe-clkctrl:0028:26"),
+	DT_CLK("40124000.mcbsp", "prcm_fck", "abe-clkctrl:0030:26"),
+	DT_CLK("40126000.mcbsp", "prcm_fck", "abe-clkctrl:0038:26"),
 	DT_CLK(NULL, "mcbsp4_sync_mux_ck", "l4-per-clkctrl:00c0:26"),
+	DT_CLK("48096000.mcbsp", "prcm_fck", "l4-per-clkctrl:00c0:26"),
 	DT_CLK(NULL, "ocp2scp_usb_phy_phy_48m", "l3-init-clkctrl:00c0:8"),
 	DT_CLK(NULL, "otg_60m_gfclk", "l3-init-clkctrl:0040:24"),
+	DT_CLK(NULL, "pad_fck", "pad_clks_ck"),
 	DT_CLK(NULL, "per_mcbsp4_gfclk", "l4-per-clkctrl:00c0:24"),
 	DT_CLK(NULL, "pmd_stm_clock_mux_ck", "emu-sys-clkctrl:0000:20"),
 	DT_CLK(NULL, "pmd_trace_clk_mux_ck", "emu-sys-clkctrl:0000:22"),
Signature · 1 line
2.42.0
author maintainer
#1 lore raw

PATCH 2/2 clk: ti: Fix missing omap5 mcbsp functional clock and aliases

We are using a wrong mcbsp functional clock. The interconnect target module driver provided clock for mcbsp is not same as the mcbsp functional clock known as the gfclk main_clk. The mcbsp functional clocks for mcbsp should have been added before we dropped the legacy platform data.

Additionally we are also missing the clock aliases for the clocks used by the audio driver if reparenting is needed. This causes audio driver errors like "CLKS: could not clk_get() prcm_fck" for mcbsp as reported by Andreas. The mcbsp clock aliases too should have been added before we dropped the legacy platform data.

Let's add the clocks and aliases with a single patch to fix the issue similar to omap4. On omap5, there is no mcbsp4 instance on the l4_per interconnect.

Fixes b1da0fa21bd1 ("ARM: OMAP2+: Drop legacy platform data for omap5 mcbsp")
Cc H. Nikolaus Schaller <[email protected]>
---
 arch/arm/boot/dts/ti/omap/omap5-l4-abe.dtsi | 6 ++++++
 drivers/clk/ti/clk-54xx.c                   | 4 ++++
 2 files changed, 10 insertions(+)
arch/arm/boot/dts/ti/omap/omap5-l4-abe.dtsi +6 -0
@@ -109,6 +109,8 @@ mcbsp1: mcbsp@0 {
 				reg = <0x0 0xff>, /* MPU private access */
 				      <0x49022000 0xff>; /* L3 Interconnect */
 				reg-names = "mpu", "dma";
+				clocks = <&abe_clkctrl OMAP5_MCBSP1_CLKCTRL 24>;
+				clock-names = "fck";
 				interrupts = <GIC_SPI 17 IRQ_TYPE_LEVEL_HIGH>;
 				interrupt-names = "common";
 				ti,buffer-size = <128>;
@@ -142,6 +144,8 @@ mcbsp2: mcbsp@0 {
 				reg = <0x0 0xff>, /* MPU private access */
 				      <0x49024000 0xff>; /* L3 Interconnect */
 				reg-names = "mpu", "dma";
+				clocks = <&abe_clkctrl OMAP5_MCBSP2_CLKCTRL 24>;
+				clock-names = "fck";
 				interrupts = <GIC_SPI 22 IRQ_TYPE_LEVEL_HIGH>;
 				interrupt-names = "common";
 				ti,buffer-size = <128>;
@@ -175,6 +179,8 @@ mcbsp3: mcbsp@0 {
 				reg = <0x0 0xff>, /* MPU private access */
 				      <0x49026000 0xff>; /* L3 Interconnect */
 				reg-names = "mpu", "dma";
+				clocks = <&abe_clkctrl OMAP5_MCBSP3_CLKCTRL 24>;
+				clock-names = "fck";
 				interrupts = <GIC_SPI 23 IRQ_TYPE_LEVEL_HIGH>;
 				interrupt-names = "common";
 				ti,buffer-size = <128>;
drivers/clk/ti/clk-54xx.c +4 -0
@@ -565,15 +565,19 @@ static struct ti_dt_clk omap54xx_clks[] = {
 	DT_CLK(NULL, "gpio8_dbclk", "l4per-clkctrl:00f8:8"),
 	DT_CLK(NULL, "mcbsp1_gfclk", "abe-clkctrl:0028:24"),
 	DT_CLK(NULL, "mcbsp1_sync_mux_ck", "abe-clkctrl:0028:26"),
+	DT_CLK("40122000.mcbsp", "prcm_fck", "abe-clkctrl:0028:26"),
 	DT_CLK(NULL, "mcbsp2_gfclk", "abe-clkctrl:0030:24"),
 	DT_CLK(NULL, "mcbsp2_sync_mux_ck", "abe-clkctrl:0030:26"),
+	DT_CLK("40124000.mcbsp", "prcm_fck", "abe-clkctrl:0030:26"),
 	DT_CLK(NULL, "mcbsp3_gfclk", "abe-clkctrl:0038:24"),
 	DT_CLK(NULL, "mcbsp3_sync_mux_ck", "abe-clkctrl:0038:26"),
+	DT_CLK("40126000.mcbsp", "prcm_fck", "abe-clkctrl:0038:26"),
 	DT_CLK(NULL, "mmc1_32khz_clk", "l3init-clkctrl:0008:8"),
 	DT_CLK(NULL, "mmc1_fclk", "l3init-clkctrl:0008:25"),
 	DT_CLK(NULL, "mmc1_fclk_mux", "l3init-clkctrl:0008:24"),
 	DT_CLK(NULL, "mmc2_fclk", "l3init-clkctrl:0010:25"),
 	DT_CLK(NULL, "mmc2_fclk_mux", "l3init-clkctrl:0010:24"),
+	DT_CLK(NULL, "pad_fck", "pad_clks_ck"),
 	DT_CLK(NULL, "sata_ref_clk", "l3init-clkctrl:0068:8"),
 	DT_CLK(NULL, "timer10_gfclk_mux", "l4per-clkctrl:0008:24"),
 	DT_CLK(NULL, "timer11_gfclk_mux", "l4per-clkctrl:0010:24"),
Signature · 1 line
2.42.0
maintainer Acked-by on v1 2/2
#2 lore raw

Re: [PATCH 2/2] clk: ti: Fix missing omap5 mcbsp functional clock and aliases

Quoting Tony Lindgren (2023-10-11 00:18:56)

Quoted 15 lines

We are using a wrong mcbsp functional clock. The interconnect target module driver provided clock for mcbsp is not same as the mcbsp functional clock known as the gfclk main_clk. The mcbsp functional clocks for mcbsp should have been added before we dropped the legacy platform data.

Additionally we are also missing the clock aliases for the clocks used by the audio driver if reparenting is needed. This causes audio driver errors like "CLKS: could not clk_get() prcm_fck" for mcbsp as reported by Andreas. The mcbsp clock aliases too should have been added before we dropped the legacy platform data.

Let's add the clocks and aliases with a single patch to fix the issue similar to omap4. On omap5, there is no mcbsp4 instance on the l4_per interconnect.

Quoted 6 lines
Fixes b1da0fa21bd1 ("ARM: OMAP2+: Drop legacy platform data for omap5 mcbsp")
Cc H. Nikolaus Schaller <[email protected]>

---

maintainer Acked-by on v1 1/2
#3 lore raw
In reply to Tony Lindgren

Quoting Tony Lindgren (2023-10-11 00:18:55)

Quoted 11 lines

We are using a wrong mcbsp functional clock. The interconnect target module driver provided clock for mcbsp is not same as the mcbsp functional clock known as the gfclk main_clk. The mcbsp functional clocks for mcbsp should have been added before we dropped the legacy platform data.

Additionally we are also missing the clock aliases for the clocks used by the audio driver if reparenting is needed. This causes audio driver errors like "CLKS: could not clk_get() prcm_fck" for mcbsp as reported by Andreas. The mcbsp clock aliases too should have been added before we dropped the legacy platform data.

Quoted 7 lines

Let's add the clocks and aliases with a single patch to fix the issue.

Fixes 349355ce3a05 ("ARM: OMAP2+: Drop legacy platform data for omap4 mcbsp")

---

author maintainer
#4 lore raw
Quoted 14 lines · Stephen Boyd

Quoting Tony Lindgren (2023-10-11 00:18:55)

Quoted 13 lines

We are using a wrong mcbsp functional clock. The interconnect target module driver provided clock for mcbsp is not same as the mcbsp functional clock known as the gfclk main_clk. The mcbsp functional clocks for mcbsp should have been added before we dropped the legacy platform data.

Additionally we are also missing the clock aliases for the clocks used by the audio driver if reparenting is needed. This causes audio driver errors like "CLKS: could not clk_get() prcm_fck" for mcbsp as reported by Andreas. The mcbsp clock aliases too should have been added before we dropped the legacy platform data.

Let's add the clocks and aliases with a single patch to fix the issue.

Quoted 7 lines
Quoted 5 lines
Fixes 349355ce3a05 ("ARM: OMAP2+: Drop legacy platform data for omap4 mcbsp")

---

Thanks applying both patches into my fixes branch.

Regards,

Tony