2019-02-14 08:58:58

by Anand Moon

Subject: [RFC 0/2] Add support for Suspend and Resume for Exynos4412 on

Add support to set the Odroid U3+ device into DEEP sleep mode
and is able to resume from suspend state.

Changes base on Krxk/next/dt and worked on 5.0.0-rc6+
TODO: fix the eMMC failuer.

Any input are welcome.

Best Regards
-Anand

Anand Moon (2):
ARM: dts: exynos: Add proper regulator states for suspend-to-mem for
odroid-u3
soc: samsung: pmu: Add the PMU data of exynos4412 to support low-power
state

.../boot/dts/exynos4412-odroid-common.dtsi | 63 ++++++++++++++
arch/arm/boot/dts/exynos4412-odroidu3.dts | 3 +
drivers/soc/samsung/exynos4-pmu.c | 83 +++++++++++++++++++
include/linux/soc/samsung/exynos-regs-pmu.h | 21 +++++
4 files changed, 170 insertions(+)

--
2.20.1

2019-02-14 09:00:01

by Anand Moon

Subject: [RFC 2/2] soc: samsung: pmu: Add the PMU data of exynos4412 to support low-power state

This patch adds configration for PMU (Power Management Unit) state
tuning for exynos4412 SoC in order to enter low-power mode during
suspend power modes and help resume from suspend state.

Fixes: bfce552d0b1 ("drivers: soc: Add support for Exynos PMU driver")
Cc: Marek Szyprowski <[email protected]>
Cc: Krzysztof Kozlowski <[email protected]>
Cc: Chanwoo Choi <[email protected]>
Signed-off-by: Anand Moon <[email protected]>
---

Changes from previous patch.
New patch to this series to support suspend and resume state

Changes have been tested on microSD card but fails to resume on cMMC.
It need to be investigated and more debuging
---
drivers/soc/samsung/exynos4-pmu.c | 83 +++++++++++++++++++++
include/linux/soc/samsung/exynos-regs-pmu.h | 21 ++++++
2 files changed, 104 insertions(+)

diff --git a/drivers/soc/samsung/exynos4-pmu.c b/drivers/soc/samsung/exynos4-pmu.c
index a7cdbf1aac0c..d261a0d2371e 100644
--- a/drivers/soc/samsung/exynos4-pmu.c
+++ b/drivers/soc/samsung/exynos4-pmu.c
@@ -200,10 +200,93 @@ static const struct exynos_pmu_conf exynos4412_pmu_config[] = {
{ PMU_TABLE_END,},
};

+static unsigned int const exynos4412_list_feed[] = {
+ EXYNOS4_ARM_CORE0_OPTION,
+ EXYNOS4_ARM_CORE1_OPTION,
+ EXYNOS4_ARM_CORE2_OPTION,
+ EXYNOS4_ARM_CORE3_OPTION,
+ EXYNOS4_ARM_COMMON_OPTION,
+ EXYNOS4_CAM_OPTION,
+ EXYNOS4_TV_OPTION,
+ EXYNOS4_MFC_OPTION,
+ EXYNOS4_G3D_OPTION,
+ EXYNOS4_LCD0_OPTION,
+ EXYNOS4_ISP_OPTION,
+ EXYNOS4_MAUDIO_OPTION,
+ EXYNOS4_GPS_OPTION,
+ EXYNOS4_GPS_ALIVE_OPTION,
+};
+
+static void exynos4412_pmu_central_seq(bool enable)
+{
+ unsigned int value;
+
+ value = pmu_raw_readl(S5P_CENTRAL_SEQ_CONFIGURATION);
+ if (enable)
+ value &= ~S5P_CENTRAL_LOWPWR_CFG;
+ else
+ value |= S5P_CENTRAL_LOWPWR_CFG;
+ pmu_raw_writel(value, S5P_CENTRAL_SEQ_CONFIGURATION);
+
+ value = pmu_raw_readl(S5P_CENTRAL_SEQ_CONFIGURATION_COREBLK);
+ if (enable)
+ value &= ~S5P_CENTRAL_LOWPWR_CFG;
+ else
+ value |= S5P_CENTRAL_LOWPWR_CFG;
+ pmu_raw_writel(value, S5P_CENTRAL_SEQ_CONFIGURATION_COREBLK);
+}
+
+static void exynos4412_pmu_init(void)
+{
+ unsigned int value;
+ int i;
+
+ /* Enable USE_STANDBY_WFI for all CORE */
+ pmu_raw_writel(S5P_USE_STANDBY_WFI_ALL, S5P_CENTRAL_SEQ_OPTION);
+
+ /* Decides whether to use retention capability */
+ value = pmu_raw_readl(S5P_ARM_L2_0_OPTION);
+ value &= ~EXYNOS_L2_USE_RETENTION;
+ pmu_raw_writel(value, S5P_ARM_L2_0_OPTION);
+
+ value = pmu_raw_readl(S5P_ARM_L2_1_OPTION);
+ value &= ~EXYNOS_L2_USE_RETENTION;
+ pmu_raw_writel(value, S5P_ARM_L2_1_OPTION);
+
+ /* Set PSHOLD port for output high */
+ value = pmu_raw_readl(S5P_PS_HOLD_CONTROL);
+ value |= S5P_PS_HOLD_OUTPUT_HIGH;
+ pmu_raw_writel(value, S5P_PS_HOLD_CONTROL);
+
+ /* Enable signal for PSHOLD port */
+ value = pmu_raw_readl(S5P_PS_HOLD_CONTROL);
+ value |= S5P_PS_HOLD_EN;
+ pmu_raw_writel(value, S5P_PS_HOLD_CONTROL);
+
+ /* Enable only SC_FEEDBACK */
+ for (i = 0; i < ARRAY_SIZE(exynos4412_list_feed); i++) {
+ value = pmu_raw_readl(exynos4412_list_feed[i]);
+ value &= ~(EXYNOS_USE_SC_COUNTER);
+ value |= EXYNOS_USE_SC_FEEDBACK;
+ pmu_raw_writel(value, exynos4412_list_feed[i]);
+ }
+
+ exynos4412_pmu_central_seq(false);
+
+ pr_info("EXYNOS4x12 PMU Initialize\n");
+}
+
+static void exynos4412_powerdown_conf(enum sys_powerdown mode)
+{
+ exynos4412_pmu_central_seq(true);
+}
+
const struct exynos_pmu_data exynos4210_pmu_data = {
.pmu_config = exynos4210_pmu_config,
};

const struct exynos_pmu_data exynos4412_pmu_data = {
.pmu_config = exynos4412_pmu_config,
+ .pmu_init = exynos4412_pmu_init,
+ .powerdown_conf = exynos4412_powerdown_conf,
};
diff --git a/include/linux/soc/samsung/exynos-regs-pmu.h b/include/linux/soc/samsung/exynos-regs-pmu.h
index 5addaf5ccbce..6beed3e669d2 100644
--- a/include/linux/soc/samsung/exynos-regs-pmu.h
+++ b/include/linux/soc/samsung/exynos-regs-pmu.h
@@ -16,6 +16,8 @@

#define S5P_CENTRAL_SEQ_CONFIGURATION 0x0200

+#define S5P_CENTRAL_SEQ_CONFIGURATION_COREBLK 0x0240
+
#define S5P_CENTRAL_LOWPWR_CFG (1 << 16)

#define S5P_CENTRAL_SEQ_OPTION 0x0208
@@ -347,6 +349,25 @@
#define EXYNOS3_OPTION_USE_SC_FEEDBACK (1 << 1)
#define EXYNOS3_OPTION_SKIP_DEACTIVATE_ACEACP_IN_PWDN (1 << 7)

+/* EXYNOS4 */
+#define EXYNOS_USE_SC_FEEDBACK BIT(1)
+#define EXYNOS_USE_SC_COUNTER BIT(0)
+
+#define EXYNOS4_ARM_CORE0_OPTION 0x2008
+#define EXYNOS4_ARM_CORE1_OPTION 0x2088
+#define EXYNOS4_ARM_CORE2_OPTION 0x2108
+#define EXYNOS4_ARM_CORE3_OPTION 0x2188
+#define EXYNOS4_ARM_COMMON_OPTION 0x2408
+#define EXYNOS4_CAM_OPTION 0x3C08
+#define EXYNOS4_TV_OPTION 0x3C28
+#define EXYNOS4_MFC_OPTION 0x3C48
+#define EXYNOS4_G3D_OPTION 0x3C68
+#define EXYNOS4_LCD0_OPTION 0x3C88
+#define EXYNOS4_ISP_OPTION 0x3CA8
+#define EXYNOS4_MAUDIO_OPTION 0x3CC8
+#define EXYNOS4_GPS_OPTION 0x3CE8
+#define EXYNOS4_GPS_ALIVE_OPTION 0x3D08
+
/* For EXYNOS5 */

#define EXYNOS5_AUTO_WDTRESET_DISABLE 0x0408
--
2.20.1

2019-02-14 09:01:06

by Anand Moon

Subject: [RFC 1/2] ARM: dts: exynos: Add proper regulator states for suspend-to-mem for odroid-u3

Add suspend-to-mem node to regulator core to be enabled or disabled
during system suspend and also support changing the regulator operating
mode during runtime and when the system enter sleep mode.

Cc: Marek Szyprowski <[email protected]>
Cc: Krzysztof Kozlowski <[email protected]>
Cc: Chanwoo Choi <[email protected]>
Signed-off-by: Anand Moon <[email protected]>
---

Changes from previos patch
[0] https://patchwork.kernel.org/patch/10712549/

Set all the WAKEUP source regulator in suspend-on state.
LD04, LD012, LD015, LD020, LD022
Set all the non used regulator in suspend-odd state
LD02, LD03, LD05, LD06, LD07, LD011, LD013, LDO14, LD016

BUCK5, BUCK6, BUCK7 and not confirable as per driver max77686-regulator

Tested on microSD card and it resumes correcly after suspend.
eMMC is not able to resume after entering into suspend state,
which need to be investigated and how to debug more.
---
.../boot/dts/exynos4412-odroid-common.dtsi | 63 +++++++++++++++++++
arch/arm/boot/dts/exynos4412-odroidu3.dts | 3 +
2 files changed, 66 insertions(+)

diff --git a/arch/arm/boot/dts/exynos4412-odroid-common.dtsi b/arch/arm/boot/dts/exynos4412-odroid-common.dtsi
index 08d3a0a7b4eb..e984461c37d9 100644
--- a/arch/arm/boot/dts/exynos4412-odroid-common.dtsi
+++ b/arch/arm/boot/dts/exynos4412-odroid-common.dtsi
@@ -288,6 +288,9 @@
regulator-min-microvolt = <1800000>;
regulator-max-microvolt = <1800000>;
regulator-always-on;
+ regulator-state-mem {
+ regulator-off-in-suspend;
+ };
};

ldo3_reg: LDO3 {
@@ -295,6 +298,9 @@
regulator-min-microvolt = <1800000>;
regulator-max-microvolt = <1800000>;
regulator-always-on;
+ regulator-state-mem {
+ regulator-off-in-suspend;
+ };
};

ldo4_reg: LDO4 {
@@ -302,6 +308,9 @@
regulator-min-microvolt = <2800000>;
regulator-max-microvolt = <2800000>;
regulator-boot-on;
+ regulator-state-mem {
+ regulator-on-in-suspend;
+ };
};

ldo5_reg: LDO5 {
@@ -310,6 +319,9 @@
regulator-max-microvolt = <1800000>;
regulator-always-on;
regulator-boot-on;
+ regulator-state-mem {
+ regulator-off-in-suspend;
+ };
};

ldo6_reg: LDO6 {
@@ -317,6 +329,9 @@
regulator-min-microvolt = <1000000>;
regulator-max-microvolt = <1000000>;
regulator-always-on;
+ regulator-state-mem {
+ regulator-off-in-suspend;
+ };
};

ldo7_reg: LDO7 {
@@ -324,18 +339,27 @@
regulator-min-microvolt = <1000000>;
regulator-max-microvolt = <1000000>;
regulator-always-on;
+ regulator-state-mem {
+ regulator-off-in-suspend;
+ };
};

ldo8_reg: LDO8 {
regulator-name = "VDD10_HDMI_1.0V";
regulator-min-microvolt = <1000000>;
regulator-max-microvolt = <1000000>;
+ regulator-state-mem {
+ regulator-on-in-suspend;
+ };
};

ldo10_reg: LDO10 {
regulator-name = "VDDQ_MIPIHSI_1.8V";
regulator-min-microvolt = <1800000>;
regulator-max-microvolt = <1800000>;
+ regulator-state-mem {
+ regulator-on-in-suspend;
+ };
};

ldo11_reg: LDO11 {
@@ -343,6 +367,9 @@
regulator-min-microvolt = <1800000>;
regulator-max-microvolt = <1800000>;
regulator-always-on;
+ regulator-state-mem {
+ regulator-off-in-suspend;
+ };
};

ldo12_reg: LDO12 {
@@ -351,6 +378,9 @@
regulator-max-microvolt = <3300000>;
regulator-always-on;
regulator-boot-on;
+ regulator-state-mem {
+ regulator-off-in-suspend;
+ };
};

ldo13_reg: LDO13 {
@@ -359,6 +389,9 @@
regulator-max-microvolt = <1800000>;
regulator-always-on;
regulator-boot-on;
+ regulator-state-mem {
+ regulator-off-in-suspend;
+ };
};

ldo14_reg: LDO14 {
@@ -367,6 +400,9 @@
regulator-max-microvolt = <1800000>;
regulator-always-on;
regulator-boot-on;
+ regulator-state-mem {
+ regulator-off-in-suspend;
+ };
};

ldo15_reg: LDO15 {
@@ -375,6 +411,9 @@
regulator-max-microvolt = <1000000>;
regulator-always-on;
regulator-boot-on;
+ regulator-state-mem {
+ regulator-off-in-suspend;
+ };
};

ldo16_reg: LDO16 {
@@ -383,6 +422,9 @@
regulator-max-microvolt = <1800000>;
regulator-always-on;
regulator-boot-on;
+ regulator-state-mem {
+ regulator-off-in-suspend;
+ };
};

ldo20_reg: LDO20 {
@@ -396,6 +438,9 @@
regulator-min-microvolt = <2800000>;
regulator-max-microvolt = <2800000>;
regulator-boot-on;
+ regulator-state-mem {
+ regulator-on-in-suspend;
+ };
};

ldo22_reg: LDO22 {
@@ -405,6 +450,9 @@
*/
regulator-name = "LDO22";
regulator-boot-on;
+ regulator-state-mem {
+ regulator-on-in-suspend;
+ };
};

ldo25_reg: LDO25 {
@@ -413,6 +461,9 @@
regulator-max-microvolt = <1800000>;
regulator-always-on;
regulator-boot-on;
+ regulator-state-mem {
+ regulator-off-in-suspend;
+ };
};

buck1_reg: BUCK1 {
@@ -421,6 +472,9 @@
regulator-max-microvolt = <1100000>;
regulator-always-on;
regulator-boot-on;
+ regulator-state-mem {
+ regulator-off-in-suspend;
+ };
};

buck2_reg: BUCK2 {
@@ -429,6 +483,9 @@
regulator-max-microvolt = <1350000>;
regulator-always-on;
regulator-boot-on;
+ regulator-state-mem {
+ regulator-on-in-suspend;
+ };
};

buck3_reg: BUCK3 {
@@ -437,6 +494,9 @@
regulator-max-microvolt = <1050000>;
regulator-always-on;
regulator-boot-on;
+ regulator-state-mem {
+ regulator-off-in-suspend;
+ };
};

buck4_reg: BUCK4 {
@@ -444,6 +504,9 @@
regulator-min-microvolt = <900000>;
regulator-max-microvolt = <1100000>;
regulator-microvolt-offset = <50000>;
+ regulator-state-mem {
+ regulator-off-in-suspend;
+ };
};

buck5_reg: BUCK5 {
diff --git a/arch/arm/boot/dts/exynos4412-odroidu3.dts b/arch/arm/boot/dts/exynos4412-odroidu3.dts
index 2bdf899df436..4ebde09fc51d 100644
--- a/arch/arm/boot/dts/exynos4412-odroidu3.dts
+++ b/arch/arm/boot/dts/exynos4412-odroidu3.dts
@@ -82,6 +82,9 @@
regulator-name = "LDO22_VDDQ_MMC4_2.8V";
regulator-min-microvolt = <2800000>;
regulator-max-microvolt = <2800000>;
+ regulator-state-mem {
+ regulator-on-in-suspend;
+ };
};

&mshc_0 {
--
2.20.1

2019-02-14 23:17:34

by Krzysztof Kozlowski

Subject: Re: [RFC 1/2] ARM: dts: exynos: Add proper regulator states for suspend-to-mem for odroid-u3

Hi Anand,

Thanks for the patch. See comments below.

On Wed, 13 Feb 2019 at 22:41, Anand Moon <[email protected]> wrote:
>
> Add suspend-to-mem node to regulator core to be enabled or disabled
> during system suspend and also support changing the regulator operating
> mode during runtime and when the system enter sleep mode.
>
> Cc: Marek Szyprowski <[email protected]>
> Cc: Krzysztof Kozlowski <[email protected]>
> Cc: Chanwoo Choi <[email protected]>
> Signed-off-by: Anand Moon <[email protected]>
> ---
>
> Changes from previos patch
> [0] https://patchwork.kernel.org/patch/10712549/
>
> Set all the WAKEUP source regulator in suspend-on state.
> LD04, LD012, LD015, LD020, LD022
> Set all the non used regulator in suspend-odd state
> LD02, LD03, LD05, LD06, LD07, LD011, LD013, LDO14, LD016
>
> BUCK5, BUCK6, BUCK7 and not confirable as per driver max77686-regulator
>
> Tested on microSD card and it resumes correcly after suspend.
> eMMC is not able to resume after entering into suspend state,
> which need to be investigated and how to debug more.
> ---
> .../boot/dts/exynos4412-odroid-common.dtsi | 63 +++++++++++++++++++
> arch/arm/boot/dts/exynos4412-odroidu3.dts | 3 +
> 2 files changed, 66 insertions(+)
>
> diff --git a/arch/arm/boot/dts/exynos4412-odroid-common.dtsi b/arch/arm/boot/dts/exynos4412-odroid-common.dtsi
> index 08d3a0a7b4eb..e984461c37d9 100644
> --- a/arch/arm/boot/dts/exynos4412-odroid-common.dtsi
> +++ b/arch/arm/boot/dts/exynos4412-odroid-common.dtsi
> @@ -288,6 +288,9 @@
> regulator-min-microvolt = <1800000>;
> regulator-max-microvolt = <1800000>;
> regulator-always-on;
> + regulator-state-mem {
> + regulator-off-in-suspend;
> + };

I see my comment from previous patch was not addressed.

> };
>
> ldo3_reg: LDO3 {
> @@ -295,6 +298,9 @@
> regulator-min-microvolt = <1800000>;
> regulator-max-microvolt = <1800000>;
> regulator-always-on;
> + regulator-state-mem {
> + regulator-off-in-suspend;
> + };

The same...

> };
>
> ldo4_reg: LDO4 {
> @@ -302,6 +308,9 @@
> regulator-min-microvolt = <2800000>;
> regulator-max-microvolt = <2800000>;
> regulator-boot-on;
> + regulator-state-mem {
> + regulator-on-in-suspend;
> + };
> };
>
> ldo5_reg: LDO5 {
> @@ -310,6 +319,9 @@
> regulator-max-microvolt = <1800000>;
> regulator-always-on;
> regulator-boot-on;
> + regulator-state-mem {
> + regulator-off-in-suspend;
> + };
> };
>
> ldo6_reg: LDO6 {
> @@ -317,6 +329,9 @@
> regulator-min-microvolt = <1000000>;
> regulator-max-microvolt = <1000000>;
> regulator-always-on;
> + regulator-state-mem {
> + regulator-off-in-suspend;
> + };
> };
>
> ldo7_reg: LDO7 {
> @@ -324,18 +339,27 @@
> regulator-min-microvolt = <1000000>;
> regulator-max-microvolt = <1000000>;
> regulator-always-on;
> + regulator-state-mem {
> + regulator-off-in-suspend;
> + };
> };
>
> ldo8_reg: LDO8 {
> regulator-name = "VDD10_HDMI_1.0V";
> regulator-min-microvolt = <1000000>;
> regulator-max-microvolt = <1000000>;
> + regulator-state-mem {
> + regulator-on-in-suspend;
> + };
> };
>
> ldo10_reg: LDO10 {
> regulator-name = "VDDQ_MIPIHSI_1.8V";
> regulator-min-microvolt = <1800000>;
> regulator-max-microvolt = <1800000>;
> + regulator-state-mem {
> + regulator-on-in-suspend;
> + };
> };
>
> ldo11_reg: LDO11 {
> @@ -343,6 +367,9 @@
> regulator-min-microvolt = <1800000>;
> regulator-max-microvolt = <1800000>;
> regulator-always-on;
> + regulator-state-mem {
> + regulator-off-in-suspend;
> + };
> };
>
> ldo12_reg: LDO12 {
> @@ -351,6 +378,9 @@
> regulator-max-microvolt = <3300000>;
> regulator-always-on;
> regulator-boot-on;
> + regulator-state-mem {
> + regulator-off-in-suspend;
> + };
> };
>
> ldo13_reg: LDO13 {
> @@ -359,6 +389,9 @@
> regulator-max-microvolt = <1800000>;
> regulator-always-on;
> regulator-boot-on;
> + regulator-state-mem {
> + regulator-off-in-suspend;
> + };
> };
>
> ldo14_reg: LDO14 {
> @@ -367,6 +400,9 @@
> regulator-max-microvolt = <1800000>;
> regulator-always-on;
> regulator-boot-on;
> + regulator-state-mem {
> + regulator-off-in-suspend;
> + };
> };
>
> ldo15_reg: LDO15 {
> @@ -375,6 +411,9 @@
> regulator-max-microvolt = <1000000>;
> regulator-always-on;
> regulator-boot-on;
> + regulator-state-mem {
> + regulator-off-in-suspend;
> + };
> };
>
> ldo16_reg: LDO16 {
> @@ -383,6 +422,9 @@
> regulator-max-microvolt = <1800000>;
> regulator-always-on;
> regulator-boot-on;
> + regulator-state-mem {
> + regulator-off-in-suspend;
> + };
> };
>
> ldo20_reg: LDO20 {
> @@ -396,6 +438,9 @@
> regulator-min-microvolt = <2800000>;
> regulator-max-microvolt = <2800000>;
> regulator-boot-on;
> + regulator-state-mem {
> + regulator-on-in-suspend;
> + };


The same... any comments?

> };
>
> ldo22_reg: LDO22 {
> @@ -405,6 +450,9 @@
> */
> regulator-name = "LDO22";
> regulator-boot-on;
> + regulator-state-mem {
> + regulator-on-in-suspend;
> + };
> };
>
> ldo25_reg: LDO25 {
> @@ -413,6 +461,9 @@
> regulator-max-microvolt = <1800000>;
> regulator-always-on;
> regulator-boot-on;
> + regulator-state-mem {
> + regulator-off-in-suspend;
> + };
> };
>
> buck1_reg: BUCK1 {
> @@ -421,6 +472,9 @@
> regulator-max-microvolt = <1100000>;
> regulator-always-on;
> regulator-boot-on;
> + regulator-state-mem {
> + regulator-off-in-suspend;

Again, you did not address my comments.

> + };
> };
>
> buck2_reg: BUCK2 {
> @@ -429,6 +483,9 @@
> regulator-max-microvolt = <1350000>;
> regulator-always-on;
> regulator-boot-on;
> + regulator-state-mem {
> + regulator-on-in-suspend;
> + };
> };
>
> buck3_reg: BUCK3 {
> @@ -437,6 +494,9 @@
> regulator-max-microvolt = <1050000>;
> regulator-always-on;
> regulator-boot-on;
> + regulator-state-mem {
> + regulator-off-in-suspend;
> + };

The same...

> };
>
> buck4_reg: BUCK4 {
> @@ -444,6 +504,9 @@
> regulator-min-microvolt = <900000>;
> regulator-max-microvolt = <1100000>;
> regulator-microvolt-offset = <50000>;
> + regulator-state-mem {
> + regulator-off-in-suspend;
> + };
> };
>
> buck5_reg: BUCK5 {
> diff --git a/arch/arm/boot/dts/exynos4412-odroidu3.dts b/arch/arm/boot/dts/exynos4412-odroidu3.dts
> index 2bdf899df436..4ebde09fc51d 100644
> --- a/arch/arm/boot/dts/exynos4412-odroidu3.dts
> +++ b/arch/arm/boot/dts/exynos4412-odroidu3.dts
> @@ -82,6 +82,9 @@
> regulator-name = "LDO22_VDDQ_MMC4_2.8V";
> regulator-min-microvolt = <2800000>;
> regulator-max-microvolt = <2800000>;
> + regulator-state-mem {
> + regulator-on-in-suspend;

Why?

Best regards,
Krzysztof

2019-02-14 23:20:33

by Krzysztof Kozlowski

Subject: Re: [RFC 2/2] soc: samsung: pmu: Add the PMU data of exynos4412 to support low-power state

On Wed, 13 Feb 2019 at 22:41, Anand Moon <[email protected]> wrote:
>
> This patch adds configration for PMU (Power Management Unit) state
> tuning for exynos4412 SoC in order to enter low-power mode during
> suspend power modes and help resume from suspend state.

The U3 and Trats2 already enter STOP/S2R so please describe what
exactly you change.

> Fixes: bfce552d0b1 ("drivers: soc: Add support for Exynos PMU driver")

How it fixes it? What was broken in that commit?

> Cc: Marek Szyprowski <[email protected]>
> Cc: Krzysztof Kozlowski <[email protected]>
> Cc: Chanwoo Choi <[email protected]>
> Signed-off-by: Anand Moon <[email protected]>
> ---
>
> Changes from previous patch.
> New patch to this series to support suspend and resume state
>
> Changes have been tested on microSD card but fails to resume on cMMC.
> It need to be investigated and more debuging
> ---
> drivers/soc/samsung/exynos4-pmu.c | 83 +++++++++++++++++++++
> include/linux/soc/samsung/exynos-regs-pmu.h | 21 ++++++
> 2 files changed, 104 insertions(+)
>
> diff --git a/drivers/soc/samsung/exynos4-pmu.c b/drivers/soc/samsung/exynos4-pmu.c
> index a7cdbf1aac0c..d261a0d2371e 100644
> --- a/drivers/soc/samsung/exynos4-pmu.c
> +++ b/drivers/soc/samsung/exynos4-pmu.c
> @@ -200,10 +200,93 @@ static const struct exynos_pmu_conf exynos4412_pmu_config[] = {
> { PMU_TABLE_END,},
> };
>
> +static unsigned int const exynos4412_list_feed[] = {
> + EXYNOS4_ARM_CORE0_OPTION,
> + EXYNOS4_ARM_CORE1_OPTION,
> + EXYNOS4_ARM_CORE2_OPTION,
> + EXYNOS4_ARM_CORE3_OPTION,
> + EXYNOS4_ARM_COMMON_OPTION,
> + EXYNOS4_CAM_OPTION,
> + EXYNOS4_TV_OPTION,
> + EXYNOS4_MFC_OPTION,
> + EXYNOS4_G3D_OPTION,
> + EXYNOS4_LCD0_OPTION,
> + EXYNOS4_ISP_OPTION,
> + EXYNOS4_MAUDIO_OPTION,
> + EXYNOS4_GPS_OPTION,
> + EXYNOS4_GPS_ALIVE_OPTION,
> +};
> +
> +static void exynos4412_pmu_central_seq(bool enable)

You name the argument as "enable" but during initialization and
system running you pass here false. It confuses me. What do you enable
here?

> +{
> + unsigned int value;
> +
> + value = pmu_raw_readl(S5P_CENTRAL_SEQ_CONFIGURATION);
> + if (enable)
> + value &= ~S5P_CENTRAL_LOWPWR_CFG;
> + else
> + value |= S5P_CENTRAL_LOWPWR_CFG;
> + pmu_raw_writel(value, S5P_CENTRAL_SEQ_CONFIGURATION);

You duplicate exynos_pm_central_suspend() without removing the original code.

> +
> + value = pmu_raw_readl(S5P_CENTRAL_SEQ_CONFIGURATION_COREBLK);
> + if (enable)
> + value &= ~S5P_CENTRAL_LOWPWR_CFG;
> + else
> + value |= S5P_CENTRAL_LOWPWR_CFG;

As manual says - set this register only if you disable C2C. Our entire
low power configuration for STOP mode is for C2C enabled case so you
add inconsistent configuration.

> + pmu_raw_writel(value, S5P_CENTRAL_SEQ_CONFIGURATION_COREBLK);
> +}
> +
> +static void exynos4412_pmu_init(void)
> +{
> + unsigned int value;
> + int i;
> +
> + /* Enable USE_STANDBY_WFI for all CORE */
> + pmu_raw_writel(S5P_USE_STANDBY_WFI_ALL, S5P_CENTRAL_SEQ_OPTION);

This does not look related to improving suspend... it looks unrelated.

> +
> + /* Decides whether to use retention capability */
> + value = pmu_raw_readl(S5P_ARM_L2_0_OPTION);
> + value &= ~EXYNOS_L2_USE_RETENTION;
> + pmu_raw_writel(value, S5P_ARM_L2_0_OPTION);
> +
> + value = pmu_raw_readl(S5P_ARM_L2_1_OPTION);
> + value &= ~EXYNOS_L2_USE_RETENTION;
> + pmu_raw_writel(value, S5P_ARM_L2_1_OPTION);
> +
> + /* Set PSHOLD port for output high */
> + value = pmu_raw_readl(S5P_PS_HOLD_CONTROL);
> + value |= S5P_PS_HOLD_OUTPUT_HIGH;
> + pmu_raw_writel(value, S5P_PS_HOLD_CONTROL);
> +
> + /* Enable signal for PSHOLD port */
> + value = pmu_raw_readl(S5P_PS_HOLD_CONTROL);
> + value |= S5P_PS_HOLD_EN;
> + pmu_raw_writel(value, S5P_PS_HOLD_CONTROL);
> +

The same - not related.

> + /* Enable only SC_FEEDBACK */
> + for (i = 0; i < ARRAY_SIZE(exynos4412_list_feed); i++) {
> + value = pmu_raw_readl(exynos4412_list_feed[i]);
> + value &= ~(EXYNOS_USE_SC_COUNTER);
> + value |= EXYNOS_USE_SC_FEEDBACK;
> + pmu_raw_writel(value, exynos4412_list_feed[i]);

Why do you prefer to use feedback instead of counter?

> + }
> +
> + exynos4412_pmu_central_seq(false);
> +
> + pr_info("EXYNOS4x12 PMU Initialize\n");
> +}
> +
> +static void exynos4412_powerdown_conf(enum sys_powerdown mode)
> +{
> + exynos4412_pmu_central_seq(true);
> +}
> +
> const struct exynos_pmu_data exynos4210_pmu_data = {
> .pmu_config = exynos4210_pmu_config,
> };
>
> const struct exynos_pmu_data exynos4412_pmu_data = {
> .pmu_config = exynos4412_pmu_config,
> + .pmu_init = exynos4412_pmu_init,
> + .powerdown_conf = exynos4412_powerdown_conf,
> };
> diff --git a/include/linux/soc/samsung/exynos-regs-pmu.h b/include/linux/soc/samsung/exynos-regs-pmu.h
> index 5addaf5ccbce..6beed3e669d2 100644
> --- a/include/linux/soc/samsung/exynos-regs-pmu.h
> +++ b/include/linux/soc/samsung/exynos-regs-pmu.h
> @@ -16,6 +16,8 @@
>
> #define S5P_CENTRAL_SEQ_CONFIGURATION 0x0200
>
> +#define S5P_CENTRAL_SEQ_CONFIGURATION_COREBLK 0x0240
> +
> #define S5P_CENTRAL_LOWPWR_CFG (1 << 16)
>
> #define S5P_CENTRAL_SEQ_OPTION 0x0208
> @@ -347,6 +349,25 @@
> #define EXYNOS3_OPTION_USE_SC_FEEDBACK (1 << 1)
> #define EXYNOS3_OPTION_SKIP_DEACTIVATE_ACEACP_IN_PWDN (1 << 7)
>
> +/* EXYNOS4 */
> +#define EXYNOS_USE_SC_FEEDBACK BIT(1)
> +#define EXYNOS_USE_SC_COUNTER BIT(0)

Instead of adding third set of these defines, I would prefer to
cleanup existing and squash all into one (exynos3 and exynos5).

> +
> +#define EXYNOS4_ARM_CORE0_OPTION 0x2008
> +#define EXYNOS4_ARM_CORE1_OPTION 0x2088
> +#define EXYNOS4_ARM_CORE2_OPTION 0x2108
> +#define EXYNOS4_ARM_CORE3_OPTION 0x2188

The same.

> +#define EXYNOS4_ARM_COMMON_OPTION 0x2408
> +#define EXYNOS4_CAM_OPTION 0x3C08
> +#define EXYNOS4_TV_OPTION 0x3C28
> +#define EXYNOS4_MFC_OPTION 0x3C48
> +#define EXYNOS4_G3D_OPTION 0x3C68
> +#define EXYNOS4_LCD0_OPTION 0x3C88
> +#define EXYNOS4_ISP_OPTION 0x3CA8
> +#define EXYNOS4_MAUDIO_OPTION 0x3CC8
> +#define EXYNOS4_GPS_OPTION 0x3CE8
> +#define EXYNOS4_GPS_ALIVE_OPTION 0x3D08

Do you need them? I think the are already defined and used by Exynos3.

Best regards,
Krzysztof

> +
> /* For EXYNOS5 */
>
> #define EXYNOS5_AUTO_WDTRESET_DISABLE 0x0408
> --
> 2.20.1
>

2019-02-15 01:49:14

by Anand Moon

Subject: Re: [RFC 1/2] ARM: dts: exynos: Add proper regulator states for suspend-to-mem for odroid-u3

hi Krzysztof,

Thanks fro your review comments.

On Thu, 14 Feb 2019 at 18:11, Krzysztof Kozlowski <[email protected]> wrote:
>
> Hi Anand,
>
> Thanks for the patch. See comments below.
>
> On Wed, 13 Feb 2019 at 22:41, Anand Moon <[email protected]> wrote:
> >
> > Add suspend-to-mem node to regulator core to be enabled or disabled
> > during system suspend and also support changing the regulator operating
> > mode during runtime and when the system enter sleep mode.
> >
> > Cc: Marek Szyprowski <[email protected]>
> > Cc: Krzysztof Kozlowski <[email protected]>
> > Cc: Chanwoo Choi <[email protected]>
> > Signed-off-by: Anand Moon <[email protected]>
> > ---
> >
> > Changes from previos patch
> > [0] https://patchwork.kernel.org/patch/10712549/
> >
> > Set all the WAKEUP source regulator in suspend-on state.
> > LD04, LD012, LD015, LD020, LD022
> > Set all the non used regulator in suspend-odd state
> > LD02, LD03, LD05, LD06, LD07, LD011, LD013, LDO14, LD016
> >
> > BUCK5, BUCK6, BUCK7 and not confirable as per driver max77686-regulator
> >
> > Tested on microSD card and it resumes correcly after suspend.
> > eMMC is not able to resume after entering into suspend state,
> > which need to be investigated and how to debug more.
> > ---
> > .../boot/dts/exynos4412-odroid-common.dtsi | 63 +++++++++++++++++++
> > arch/arm/boot/dts/exynos4412-odroidu3.dts | 3 +
> > 2 files changed, 66 insertions(+)
> >
> > diff --git a/arch/arm/boot/dts/exynos4412-odroid-common.dtsi b/arch/arm/boot/dts/exynos4412-odroid-common.dtsi
> > index 08d3a0a7b4eb..e984461c37d9 100644
> > --- a/arch/arm/boot/dts/exynos4412-odroid-common.dtsi
> > +++ b/arch/arm/boot/dts/exynos4412-odroid-common.dtsi
> > @@ -288,6 +288,9 @@
> > regulator-min-microvolt = <1800000>;
> > regulator-max-microvolt = <1800000>;
> > regulator-always-on;
> > + regulator-state-mem {
> > + regulator-off-in-suspend;
> > + };
>
> I see my comment from previous patch was not addressed.

I left this unchanged since this regulator is not active linked and used.
Ok I will change this to regulator-on-in-suspend,

From documentation device tree binding regulator.txt
"- regulator-always-on: boolean, regulator should never be disabled"

But some regulator switches to a low power mode under light current loads
and the device automatically switches back to a fast response mode as the
output load increases.

>
> > };
> >
> > ldo3_reg: LDO3 {
> > @@ -295,6 +298,9 @@
> > regulator-min-microvolt = <1800000>;
> > regulator-max-microvolt = <1800000>;
> > regulator-always-on;
> > + regulator-state-mem {
> > + regulator-off-in-suspend;
> > + };
>
> The same...

Ok I will change this to regulator-on-in-suspend,

>
> > };
> >
> > ldo4_reg: LDO4 {
> > @@ -302,6 +308,9 @@
> > regulator-min-microvolt = <2800000>;
> > regulator-max-microvolt = <2800000>;
> > regulator-boot-on;
> > + regulator-state-mem {
> > + regulator-on-in-suspend;
> > + };
> > };
> >
> > ldo5_reg: LDO5 {
> > @@ -310,6 +319,9 @@
> > regulator-max-microvolt = <1800000>;
> > regulator-always-on;
> > regulator-boot-on;
> > + regulator-state-mem {
> > + regulator-off-in-suspend;
> > + };
> > };
> >
> > ldo6_reg: LDO6 {
> > @@ -317,6 +329,9 @@
> > regulator-min-microvolt = <1000000>;
> > regulator-max-microvolt = <1000000>;
> > regulator-always-on;
> > + regulator-state-mem {
> > + regulator-off-in-suspend;
> > + };
> > };
> >
> > ldo7_reg: LDO7 {
> > @@ -324,18 +339,27 @@
> > regulator-min-microvolt = <1000000>;
> > regulator-max-microvolt = <1000000>;
> > regulator-always-on;
> > + regulator-state-mem {
> > + regulator-off-in-suspend;
> > + };
> > };
> >
> > ldo8_reg: LDO8 {
> > regulator-name = "VDD10_HDMI_1.0V";
> > regulator-min-microvolt = <1000000>;
> > regulator-max-microvolt = <1000000>;
> > + regulator-state-mem {
> > + regulator-on-in-suspend;
> > + };
> > };
> >
> > ldo10_reg: LDO10 {
> > regulator-name = "VDDQ_MIPIHSI_1.8V";
> > regulator-min-microvolt = <1800000>;
> > regulator-max-microvolt = <1800000>;
> > + regulator-state-mem {
> > + regulator-on-in-suspend;
> > + };
> > };
> >
> > ldo11_reg: LDO11 {
> > @@ -343,6 +367,9 @@
> > regulator-min-microvolt = <1800000>;
> > regulator-max-microvolt = <1800000>;
> > regulator-always-on;
> > + regulator-state-mem {
> > + regulator-off-in-suspend;
> > + };
> > };
> >
> > ldo12_reg: LDO12 {
> > @@ -351,6 +378,9 @@
> > regulator-max-microvolt = <3300000>;
> > regulator-always-on;
> > regulator-boot-on;
> > + regulator-state-mem {
> > + regulator-off-in-suspend;
> > + };
> > };
> >
> > ldo13_reg: LDO13 {
> > @@ -359,6 +389,9 @@
> > regulator-max-microvolt = <1800000>;
> > regulator-always-on;
> > regulator-boot-on;
> > + regulator-state-mem {
> > + regulator-off-in-suspend;
> > + };
> > };
> >
> > ldo14_reg: LDO14 {
> > @@ -367,6 +400,9 @@
> > regulator-max-microvolt = <1800000>;
> > regulator-always-on;
> > regulator-boot-on;
> > + regulator-state-mem {
> > + regulator-off-in-suspend;
> > + };
> > };
> >
> > ldo15_reg: LDO15 {
> > @@ -375,6 +411,9 @@
> > regulator-max-microvolt = <1000000>;
> > regulator-always-on;
> > regulator-boot-on;
> > + regulator-state-mem {
> > + regulator-off-in-suspend;
> > + };
> > };
> >
> > ldo16_reg: LDO16 {
> > @@ -383,6 +422,9 @@
> > regulator-max-microvolt = <1800000>;
> > regulator-always-on;
> > regulator-boot-on;
> > + regulator-state-mem {
> > + regulator-off-in-suspend;
> > + };
> > };
> >
> > ldo20_reg: LDO20 {
> > @@ -396,6 +438,9 @@
> > regulator-min-microvolt = <2800000>;
> > regulator-max-microvolt = <2800000>;
> > regulator-boot-on;
> > + regulator-state-mem {
> > + regulator-on-in-suspend;
> > + };
>
>
> The same... any comments?

I left this on for mshc_0 (emmc) regulator,
since I could not get suspend resume working on eMMc

>
> > };
> >
> > ldo22_reg: LDO22 {
> > @@ -405,6 +450,9 @@
> > */
> > regulator-name = "LDO22";
> > regulator-boot-on;
> > + regulator-state-mem {
> > + regulator-on-in-suspend;
> > + };
> > };
> >
> > ldo25_reg: LDO25 {
> > @@ -413,6 +461,9 @@
> > regulator-max-microvolt = <1800000>;
> > regulator-always-on;
> > regulator-boot-on;
> > + regulator-state-mem {
> > + regulator-off-in-suspend;
> > + };
> > };
> >
> > buck1_reg: BUCK1 {
> > @@ -421,6 +472,9 @@
> > regulator-max-microvolt = <1100000>;
> > regulator-always-on;
> > regulator-boot-on;
> > + regulator-state-mem {
> > + regulator-off-in-suspend;
>
> Again, you did not address my comments.
>

Buck1 support entering into LPM setting,
But I will set this to regulator-on-in-suspend; in next patch.

> > + };
> > };
> >
> > buck2_reg: BUCK2 {
> > @@ -429,6 +483,9 @@
> > regulator-max-microvolt = <1350000>;
> > regulator-always-on;
> > regulator-boot-on;
> > + regulator-state-mem {
> > + regulator-on-in-suspend;
> > + };
> > };
> >
> > buck3_reg: BUCK3 {
> > @@ -437,6 +494,9 @@
> > regulator-max-microvolt = <1050000>;
> > regulator-always-on;
> > regulator-boot-on;
> > + regulator-state-mem {
> > + regulator-off-in-suspend;
> > + };
>
> The same...

Buck3 support entering into LPM setting,
But I will set this to regulator-on-in-suspend; in next patch.

>
> > };
> >
> > buck4_reg: BUCK4 {
> > @@ -444,6 +504,9 @@
> > regulator-min-microvolt = <900000>;
> > regulator-max-microvolt = <1100000>;
> > regulator-microvolt-offset = <50000>;
> > + regulator-state-mem {
> > + regulator-off-in-suspend;
> > + };
> > };
> >
> > buck5_reg: BUCK5 {
> > diff --git a/arch/arm/boot/dts/exynos4412-odroidu3.dts b/arch/arm/boot/dts/exynos4412-odroidu3.dts
> > index 2bdf899df436..4ebde09fc51d 100644
> > --- a/arch/arm/boot/dts/exynos4412-odroidu3.dts
> > +++ b/arch/arm/boot/dts/exynos4412-odroidu3.dts
> > @@ -82,6 +82,9 @@
> > regulator-name = "LDO22_VDDQ_MMC4_2.8V";
> > regulator-min-microvolt = <2800000>;
> > regulator-max-microvolt = <2800000>;
> > + regulator-state-mem {
> > + regulator-on-in-suspend;
>
> Why?
>

I chose not to disable mshc_0 (emmc) regulator,
since I could not get suspend resume working on eMMC.

> Best regards,
> Krzysztof

Best Regards

-Anand

2019-02-15 01:50:44

by Anand Moon

Subject: Re: [RFC 2/2] soc: samsung: pmu: Add the PMU data of exynos4412 to support low-power state

Hi Krzysztof,

Thanks for your review comments.

On Thu, 14 Feb 2019 at 18:29, Krzysztof Kozlowski <[email protected]> wrote:
>
> On Wed, 13 Feb 2019 at 22:41, Anand Moon <[email protected]> wrote:
> >
> > This patch adds configration for PMU (Power Management Unit) state
> > tuning for exynos4412 SoC in order to enter low-power mode during
> > suspend power modes and help resume from suspend state.
>
> The U3 and Trats2 already enter STOP/S2R so please describe what
> exactly you change.
>
> > Fixes: bfce552d0b1 ("drivers: soc: Add support for Exynos PMU driver")
>
> How it fixes it? What was broken in that commit?

* I was not aware on their is common framework for suspend and resume
other than setting this here.I only look in to some the other exynos
pmu architecture
and referring 3.10.x kernel to model my changes.*

>
> > Cc: Marek Szyprowski <[email protected]>
> > Cc: Krzysztof Kozlowski <[email protected]>
> > Cc: Chanwoo Choi <[email protected]>
> > Signed-off-by: Anand Moon <[email protected]>
> > ---
> >
> > Changes from previous patch.
> > New patch to this series to support suspend and resume state
> >
> > Changes have been tested on microSD card but fails to resume on cMMC.
> > It need to be investigated and more debuging
> > ---
> > drivers/soc/samsung/exynos4-pmu.c | 83 +++++++++++++++++++++
> > include/linux/soc/samsung/exynos-regs-pmu.h | 21 ++++++
> > 2 files changed, 104 insertions(+)
> >
> > diff --git a/drivers/soc/samsung/exynos4-pmu.c b/drivers/soc/samsung/exynos4-pmu.c
> > index a7cdbf1aac0c..d261a0d2371e 100644
> > --- a/drivers/soc/samsung/exynos4-pmu.c
> > +++ b/drivers/soc/samsung/exynos4-pmu.c
> > @@ -200,10 +200,93 @@ static const struct exynos_pmu_conf exynos4412_pmu_config[] = {
> > { PMU_TABLE_END,},
> > };
> >
> > +static unsigned int const exynos4412_list_feed[] = {
> > + EXYNOS4_ARM_CORE0_OPTION,
> > + EXYNOS4_ARM_CORE1_OPTION,
> > + EXYNOS4_ARM_CORE2_OPTION,
> > + EXYNOS4_ARM_CORE3_OPTION,
> > + EXYNOS4_ARM_COMMON_OPTION,
> > + EXYNOS4_CAM_OPTION,
> > + EXYNOS4_TV_OPTION,
> > + EXYNOS4_MFC_OPTION,
> > + EXYNOS4_G3D_OPTION,
> > + EXYNOS4_LCD0_OPTION,
> > + EXYNOS4_ISP_OPTION,
> > + EXYNOS4_MAUDIO_OPTION,
> > + EXYNOS4_GPS_OPTION,
> > + EXYNOS4_GPS_ALIVE_OPTION,
> > +};
> > +
> > +static void exynos4412_pmu_central_seq(bool enable)
>
> You name the argument as "enable" but during initialization and
> system running you pass here false. It confuses me. What do you enable
> here?
>

Yep your are correct need to drop this function as already done in
common frame work.

> > +{
> > + unsigned int value;
> > +
> > + value = pmu_raw_readl(S5P_CENTRAL_SEQ_CONFIGURATION);
> > + if (enable)
> > + value &= ~S5P_CENTRAL_LOWPWR_CFG;
> > + else
> > + value |= S5P_CENTRAL_LOWPWR_CFG;
> > + pmu_raw_writel(value, S5P_CENTRAL_SEQ_CONFIGURATION);
>
> You duplicate exynos_pm_central_suspend() without removing the original code.
>
> > +
> > + value = pmu_raw_readl(S5P_CENTRAL_SEQ_CONFIGURATION_COREBLK);
> > + if (enable)
> > + value &= ~S5P_CENTRAL_LOWPWR_CFG;
> > + else
> > + value |= S5P_CENTRAL_LOWPWR_CFG;
>
> As manual says - set this register only if you disable C2C. Our entire
> low power configuration for STOP mode is for C2C enabled case so you
> add inconsistent configuration.

Ok Sorry I overlook this code change.

Enable system power down. Set only CENTRAL_SEQ_CONFIGURATION register
if you disable C2C.
Set both CENTRAL_SEQ_CONFIGURATION and
CENTRAL_SEQ_CONFIGURATION_COREBLK registers if you enable C2C.

> > + pmu_raw_writel(value, S5P_CENTRAL_SEQ_CONFIGURATION_COREBLK);
> > +}
> > +
> > +static void exynos4412_pmu_init(void)
> > +{
> > + unsigned int value;
> > + int i;
> > +
> > + /* Enable USE_STANDBY_WFI for all CORE */
> > + pmu_raw_writel(S5P_USE_STANDBY_WFI_ALL, S5P_CENTRAL_SEQ_OPTION);
>
> This does not look related to improving suspend... it looks unrelated.
>

Ok yes your are correct all ready done in comment frame work.

"Execute WFI/WFE for all CPU cores. As soon as all the CPU cores in
Exynos 4412 SCP enter STANDBY mode"
> > +
> > + /* Decides whether to use retention capability */
> > + value = pmu_raw_readl(S5P_ARM_L2_0_OPTION);
> > + value &= ~EXYNOS_L2_USE_RETENTION;
> > + pmu_raw_writel(value, S5P_ARM_L2_0_OPTION);
> > +
> > + value = pmu_raw_readl(S5P_ARM_L2_1_OPTION);
> > + value &= ~EXYNOS_L2_USE_RETENTION;
> > + pmu_raw_writel(value, S5P_ARM_L2_1_OPTION);
> > +
> > + /* Set PSHOLD port for output high */
> > + value = pmu_raw_readl(S5P_PS_HOLD_CONTROL);
> > + value |= S5P_PS_HOLD_OUTPUT_HIGH;
> > + pmu_raw_writel(value, S5P_PS_HOLD_CONTROL);
> > +
> > + /* Enable signal for PSHOLD port */
> > + value = pmu_raw_readl(S5P_PS_HOLD_CONTROL);
> > + value |= S5P_PS_HOLD_EN;
> > + pmu_raw_writel(value, S5P_PS_HOLD_CONTROL);
> > +
>
> The same - not related.

PS_HOLD_CONTROL need to be set under following set
Hardware reset Reset
Watchdog timer reset Keep its value
Software reset Keep its value
Warm reset Keep its value
Wakeup reset Keep its value

>
> > + /* Enable only SC_FEEDBACK */
> > + for (i = 0; i < ARRAY_SIZE(exynos4412_list_feed); i++) {
> > + value = pmu_raw_readl(exynos4412_list_feed[i]);
> > + value &= ~(EXYNOS_USE_SC_COUNTER);
> > + value |= EXYNOS_USE_SC_FEEDBACK;
> > + pmu_raw_writel(value, exynos4412_list_feed[i]);
>
> Why do you prefer to use feedback instead of counter?
>

Their is a note below if we use this setting.
NOTE: Either one of USE_SC_FEEDBACK and USE_SC_COUNTER should be activated.

> > + }
> > +
> > + exynos4412_pmu_central_seq(false);
> > +
> > + pr_info("EXYNOS4x12 PMU Initialize\n");
> > +}
> > +
> > +static void exynos4412_powerdown_conf(enum sys_powerdown mode)
> > +{
> > + exynos4412_pmu_central_seq(true);
> > +}
> > +
> > const struct exynos_pmu_data exynos4210_pmu_data = {
> > .pmu_config = exynos4210_pmu_config,
> > };
> >
> > const struct exynos_pmu_data exynos4412_pmu_data = {
> > .pmu_config = exynos4412_pmu_config,
> > + .pmu_init = exynos4412_pmu_init,
> > + .powerdown_conf = exynos4412_powerdown_conf,
> > };
> > diff --git a/include/linux/soc/samsung/exynos-regs-pmu.h b/include/linux/soc/samsung/exynos-regs-pmu.h
> > index 5addaf5ccbce..6beed3e669d2 100644
> > --- a/include/linux/soc/samsung/exynos-regs-pmu.h
> > +++ b/include/linux/soc/samsung/exynos-regs-pmu.h
> > @@ -16,6 +16,8 @@
> >
> > #define S5P_CENTRAL_SEQ_CONFIGURATION 0x0200
> >
> > +#define S5P_CENTRAL_SEQ_CONFIGURATION_COREBLK 0x0240
> > +
> > #define S5P_CENTRAL_LOWPWR_CFG (1 << 16)
> >
> > #define S5P_CENTRAL_SEQ_OPTION 0x0208
> > @@ -347,6 +349,25 @@
> > #define EXYNOS3_OPTION_USE_SC_FEEDBACK (1 << 1)
> > #define EXYNOS3_OPTION_SKIP_DEACTIVATE_ACEACP_IN_PWDN (1 << 7)
> >
> > +/* EXYNOS4 */
> > +#define EXYNOS_USE_SC_FEEDBACK BIT(1)
> > +#define EXYNOS_USE_SC_COUNTER BIT(0)
>
> Instead of adding third set of these defines, I would prefer to
> cleanup existing and squash all into one (exynos3 and exynos5).
>

Ok I will try to do this changes or use the existing one.

> > +
> > +#define EXYNOS4_ARM_CORE0_OPTION 0x2008
> > +#define EXYNOS4_ARM_CORE1_OPTION 0x2088
> > +#define EXYNOS4_ARM_CORE2_OPTION 0x2108
> > +#define EXYNOS4_ARM_CORE3_OPTION 0x2188
>
> The same.
>
> > +#define EXYNOS4_ARM_COMMON_OPTION 0x2408
> > +#define EXYNOS4_CAM_OPTION 0x3C08
> > +#define EXYNOS4_TV_OPTION 0x3C28
> > +#define EXYNOS4_MFC_OPTION 0x3C48
> > +#define EXYNOS4_G3D_OPTION 0x3C68
> > +#define EXYNOS4_LCD0_OPTION 0x3C88
> > +#define EXYNOS4_ISP_OPTION 0x3CA8
> > +#define EXYNOS4_MAUDIO_OPTION 0x3CC8
> > +#define EXYNOS4_GPS_OPTION 0x3CE8
> > +#define EXYNOS4_GPS_ALIVE_OPTION 0x3D08
>
> Do you need them? I think the are already defined and used by Exynos3.

Ok I will uses these form Exynos3.

>
> Best regards,
> Krzysztof
>
> > +
> > /* For EXYNOS5 */
> >
> > #define EXYNOS5_AUTO_WDTRESET_DISABLE 0x0408
> > --
> > 2.20.1
> >

Best Regards


-Anand

2019-02-15 15:27:13

by Krzysztof Kozlowski

Subject: Re: [RFC 1/2] ARM: dts: exynos: Add proper regulator states for suspend-to-mem for odroid-u3

On Thu, 14 Feb 2019 at 19:35, Anand Moon <[email protected]> wrote:
>
> hi Krzysztof,
>
> Thanks fro your review comments.
>
> On Thu, 14 Feb 2019 at 18:11, Krzysztof Kozlowski <[email protected]> wrote:
> >
> > Hi Anand,
> >
> > Thanks for the patch. See comments below.
> >
> > On Wed, 13 Feb 2019 at 22:41, Anand Moon <[email protected]> wrote:
> > >
> > > Add suspend-to-mem node to regulator core to be enabled or disabled
> > > during system suspend and also support changing the regulator operating
> > > mode during runtime and when the system enter sleep mode.
> > >
> > > Cc: Marek Szyprowski <[email protected]>
> > > Cc: Krzysztof Kozlowski <[email protected]>
> > > Cc: Chanwoo Choi <[email protected]>
> > > Signed-off-by: Anand Moon <[email protected]>
> > > ---
> > >
> > > Changes from previos patch
> > > [0] https://patchwork.kernel.org/patch/10712549/
> > >
> > > Set all the WAKEUP source regulator in suspend-on state.
> > > LD04, LD012, LD015, LD020, LD022
> > > Set all the non used regulator in suspend-odd state
> > > LD02, LD03, LD05, LD06, LD07, LD011, LD013, LDO14, LD016
> > >
> > > BUCK5, BUCK6, BUCK7 and not confirable as per driver max77686-regulator
> > >
> > > Tested on microSD card and it resumes correcly after suspend.
> > > eMMC is not able to resume after entering into suspend state,
> > > which need to be investigated and how to debug more.
> > > ---
> > > .../boot/dts/exynos4412-odroid-common.dtsi | 63 +++++++++++++++++++
> > > arch/arm/boot/dts/exynos4412-odroidu3.dts | 3 +
> > > 2 files changed, 66 insertions(+)
> > >
> > > diff --git a/arch/arm/boot/dts/exynos4412-odroid-common.dtsi b/arch/arm/boot/dts/exynos4412-odroid-common.dtsi
> > > index 08d3a0a7b4eb..e984461c37d9 100644
> > > --- a/arch/arm/boot/dts/exynos4412-odroid-common.dtsi
> > > +++ b/arch/arm/boot/dts/exynos4412-odroid-common.dtsi
> > > @@ -288,6 +288,9 @@
> > > regulator-min-microvolt = <1800000>;
> > > regulator-max-microvolt = <1800000>;
> > > regulator-always-on;
> > > + regulator-state-mem {
> > > + regulator-off-in-suspend;
> > > + };
> >
> > I see my comment from previous patch was not addressed.
>
> I left this unchanged since this regulator is not active linked and used.
> Ok I will change this to regulator-on-in-suspend,

Why?

Previous patch was setting this to "on". I said that this is noop and
if you want to add it, I need explanation why this regulator has to
stay on during suspend.

So you changed to "off"... which is still noop... and you did not
provide explanation. Now you replied that you will change to "on"...
so this will be circle. Still without explanation.

>
> From documentation device tree binding regulator.txt
> "- regulator-always-on: boolean, regulator should never be disabled"
>
> But some regulator switches to a low power mode under light current loads
> and the device automatically switches back to a fast response mode as the
> output load increases.
>
> >
> > > };
> > >
> > > ldo3_reg: LDO3 {
> > > @@ -295,6 +298,9 @@
> > > regulator-min-microvolt = <1800000>;
> > > regulator-max-microvolt = <1800000>;
> > > regulator-always-on;
> > > + regulator-state-mem {
> > > + regulator-off-in-suspend;
> > > + };
> >
> > The same...
>
> Ok I will change this to regulator-on-in-suspend,

The same - why changing this to on or off? I need explanation why this
should be in first place.

> >
> > > };
> > >
> > > ldo4_reg: LDO4 {
> > > @@ -302,6 +308,9 @@
> > > regulator-min-microvolt = <2800000>;
> > > regulator-max-microvolt = <2800000>;
> > > regulator-boot-on;
> > > + regulator-state-mem {
> > > + regulator-on-in-suspend;
> > > + };
> > > };
> > >
> > > ldo5_reg: LDO5 {
> > > @@ -310,6 +319,9 @@
> > > regulator-max-microvolt = <1800000>;
> > > regulator-always-on;
> > > regulator-boot-on;
> > > + regulator-state-mem {
> > > + regulator-off-in-suspend;
> > > + };
> > > };
> > >
> > > ldo6_reg: LDO6 {
> > > @@ -317,6 +329,9 @@
> > > regulator-min-microvolt = <1000000>;
> > > regulator-max-microvolt = <1000000>;
> > > regulator-always-on;
> > > + regulator-state-mem {
> > > + regulator-off-in-suspend;
> > > + };
> > > };
> > >
> > > ldo7_reg: LDO7 {
> > > @@ -324,18 +339,27 @@
> > > regulator-min-microvolt = <1000000>;
> > > regulator-max-microvolt = <1000000>;
> > > regulator-always-on;
> > > + regulator-state-mem {
> > > + regulator-off-in-suspend;
> > > + };
> > > };
> > >
> > > ldo8_reg: LDO8 {
> > > regulator-name = "VDD10_HDMI_1.0V";
> > > regulator-min-microvolt = <1000000>;
> > > regulator-max-microvolt = <1000000>;
> > > + regulator-state-mem {
> > > + regulator-on-in-suspend;
> > > + };
> > > };
> > >
> > > ldo10_reg: LDO10 {
> > > regulator-name = "VDDQ_MIPIHSI_1.8V";
> > > regulator-min-microvolt = <1800000>;
> > > regulator-max-microvolt = <1800000>;
> > > + regulator-state-mem {
> > > + regulator-on-in-suspend;
> > > + };
> > > };
> > >
> > > ldo11_reg: LDO11 {
> > > @@ -343,6 +367,9 @@
> > > regulator-min-microvolt = <1800000>;
> > > regulator-max-microvolt = <1800000>;
> > > regulator-always-on;
> > > + regulator-state-mem {
> > > + regulator-off-in-suspend;
> > > + };
> > > };
> > >
> > > ldo12_reg: LDO12 {
> > > @@ -351,6 +378,9 @@
> > > regulator-max-microvolt = <3300000>;
> > > regulator-always-on;
> > > regulator-boot-on;
> > > + regulator-state-mem {
> > > + regulator-off-in-suspend;
> > > + };
> > > };
> > >
> > > ldo13_reg: LDO13 {
> > > @@ -359,6 +389,9 @@
> > > regulator-max-microvolt = <1800000>;
> > > regulator-always-on;
> > > regulator-boot-on;
> > > + regulator-state-mem {
> > > + regulator-off-in-suspend;
> > > + };
> > > };
> > >
> > > ldo14_reg: LDO14 {
> > > @@ -367,6 +400,9 @@
> > > regulator-max-microvolt = <1800000>;
> > > regulator-always-on;
> > > regulator-boot-on;
> > > + regulator-state-mem {
> > > + regulator-off-in-suspend;
> > > + };
> > > };
> > >
> > > ldo15_reg: LDO15 {
> > > @@ -375,6 +411,9 @@
> > > regulator-max-microvolt = <1000000>;
> > > regulator-always-on;
> > > regulator-boot-on;
> > > + regulator-state-mem {
> > > + regulator-off-in-suspend;
> > > + };
> > > };
> > >
> > > ldo16_reg: LDO16 {
> > > @@ -383,6 +422,9 @@
> > > regulator-max-microvolt = <1800000>;
> > > regulator-always-on;
> > > regulator-boot-on;
> > > + regulator-state-mem {
> > > + regulator-off-in-suspend;
> > > + };
> > > };
> > >
> > > ldo20_reg: LDO20 {
> > > @@ -396,6 +438,9 @@
> > > regulator-min-microvolt = <2800000>;
> > > regulator-max-microvolt = <2800000>;
> > > regulator-boot-on;
> > > + regulator-state-mem {
> > > + regulator-on-in-suspend;
> > > + };
> >
> >
> > The same... any comments?
>
> I left this on for mshc_0 (emmc) regulator,
> since I could not get suspend resume working on eMMc

In general eMMC can be turned off during suspend. However eMMC layer
may be handling this (manually turning off/on). So after checking and
confirming it, please document it.

> >
> > > };
> > >
> > > ldo22_reg: LDO22 {
> > > @@ -405,6 +450,9 @@
> > > */
> > > regulator-name = "LDO22";
> > > regulator-boot-on;
> > > + regulator-state-mem {
> > > + regulator-on-in-suspend;
> > > + };
> > > };
> > >
> > > ldo25_reg: LDO25 {
> > > @@ -413,6 +461,9 @@
> > > regulator-max-microvolt = <1800000>;
> > > regulator-always-on;
> > > regulator-boot-on;
> > > + regulator-state-mem {
> > > + regulator-off-in-suspend;
> > > + };
> > > };
> > >
> > > buck1_reg: BUCK1 {
> > > @@ -421,6 +472,9 @@
> > > regulator-max-microvolt = <1100000>;
> > > regulator-always-on;
> > > regulator-boot-on;
> > > + regulator-state-mem {
> > > + regulator-off-in-suspend;
> >
> > Again, you did not address my comments.
> >
>
> Buck1 support entering into LPM setting,
> But I will set this to regulator-on-in-suspend; in next patch.

Why?

> > > + };
> > > };
> > >
> > > buck2_reg: BUCK2 {
> > > @@ -429,6 +483,9 @@
> > > regulator-max-microvolt = <1350000>;
> > > regulator-always-on;
> > > regulator-boot-on;
> > > + regulator-state-mem {
> > > + regulator-on-in-suspend;
> > > + };
> > > };
> > >
> > > buck3_reg: BUCK3 {
> > > @@ -437,6 +494,9 @@
> > > regulator-max-microvolt = <1050000>;
> > > regulator-always-on;
> > > regulator-boot-on;
> > > + regulator-state-mem {
> > > + regulator-off-in-suspend;
> > > + };
> >
> > The same...
>
> Buck3 support entering into LPM setting,
> But I will set this to regulator-on-in-suspend; in next patch.

Again - why this should be on or off?

> >
> > > };
> > >
> > > buck4_reg: BUCK4 {
> > > @@ -444,6 +504,9 @@
> > > regulator-min-microvolt = <900000>;
> > > regulator-max-microvolt = <1100000>;
> > > regulator-microvolt-offset = <50000>;
> > > + regulator-state-mem {
> > > + regulator-off-in-suspend;
> > > + };
> > > };
> > >
> > > buck5_reg: BUCK5 {
> > > diff --git a/arch/arm/boot/dts/exynos4412-odroidu3.dts b/arch/arm/boot/dts/exynos4412-odroidu3.dts
> > > index 2bdf899df436..4ebde09fc51d 100644
> > > --- a/arch/arm/boot/dts/exynos4412-odroidu3.dts
> > > +++ b/arch/arm/boot/dts/exynos4412-odroidu3.dts
> > > @@ -82,6 +82,9 @@
> > > regulator-name = "LDO22_VDDQ_MMC4_2.8V";
> > > regulator-min-microvolt = <2800000>;
> > > regulator-max-microvolt = <2800000>;
> > > + regulator-state-mem {
> > > + regulator-on-in-suspend;
> >
> > Why?
> >
>
> I chose not to disable mshc_0 (emmc) regulator,
> since I could not get suspend resume working on eMMC.

Having "regulator-on-in-suspend" is not the same as not disabling
regulator during suspend. This property means you will explicitly
enable it during suspend. To me it looks wrong so I would be happy to
see explanations.

Best regards,
Krzysztof

2019-02-15 15:29:17

by Krzysztof Kozlowski

Subject: Re: [RFC 2/2] soc: samsung: pmu: Add the PMU data of exynos4412 to support low-power state

On Thu, 14 Feb 2019 at 19:37, Anand Moon <[email protected]> wrote:
>
> Hi Krzysztof,
>
> Thanks for your review comments.
>
> On Thu, 14 Feb 2019 at 18:29, Krzysztof Kozlowski <[email protected]> wrote:
> >
> > On Wed, 13 Feb 2019 at 22:41, Anand Moon <[email protected]> wrote:
> > >
> > > This patch adds configration for PMU (Power Management Unit) state
> > > tuning for exynos4412 SoC in order to enter low-power mode during
> > > suspend power modes and help resume from suspend state.
> >
> > The U3 and Trats2 already enter STOP/S2R so please describe what
> > exactly you change.
> >
> > > Fixes: bfce552d0b1 ("drivers: soc: Add support for Exynos PMU driver")
> >
> > How it fixes it? What was broken in that commit?
>
> * I was not aware on their is common framework for suspend and resume
> other than setting this here.I only look in to some the other exynos
> pmu architecture
> and referring 3.10.x kernel to model my changes.*

Suspend in general should be working already so adding this code just
because is not a valid reason. Please specify what is not working
first.

> >
> > > Cc: Marek Szyprowski <[email protected]>
> > > Cc: Krzysztof Kozlowski <[email protected]>
> > > Cc: Chanwoo Choi <[email protected]>
> > > Signed-off-by: Anand Moon <[email protected]>
> > > ---
> > >
> > > Changes from previous patch.
> > > New patch to this series to support suspend and resume state
> > >
> > > Changes have been tested on microSD card but fails to resume on cMMC.
> > > It need to be investigated and more debuging
> > > ---
> > > drivers/soc/samsung/exynos4-pmu.c | 83 +++++++++++++++++++++
> > > include/linux/soc/samsung/exynos-regs-pmu.h | 21 ++++++
> > > 2 files changed, 104 insertions(+)
> > >
> > > diff --git a/drivers/soc/samsung/exynos4-pmu.c b/drivers/soc/samsung/exynos4-pmu.c
> > > index a7cdbf1aac0c..d261a0d2371e 100644
> > > --- a/drivers/soc/samsung/exynos4-pmu.c
> > > +++ b/drivers/soc/samsung/exynos4-pmu.c
> > > @@ -200,10 +200,93 @@ static const struct exynos_pmu_conf exynos4412_pmu_config[] = {
> > > { PMU_TABLE_END,},
> > > };
> > >
> > > +static unsigned int const exynos4412_list_feed[] = {
> > > + EXYNOS4_ARM_CORE0_OPTION,
> > > + EXYNOS4_ARM_CORE1_OPTION,
> > > + EXYNOS4_ARM_CORE2_OPTION,
> > > + EXYNOS4_ARM_CORE3_OPTION,
> > > + EXYNOS4_ARM_COMMON_OPTION,
> > > + EXYNOS4_CAM_OPTION,
> > > + EXYNOS4_TV_OPTION,
> > > + EXYNOS4_MFC_OPTION,
> > > + EXYNOS4_G3D_OPTION,
> > > + EXYNOS4_LCD0_OPTION,
> > > + EXYNOS4_ISP_OPTION,
> > > + EXYNOS4_MAUDIO_OPTION,
> > > + EXYNOS4_GPS_OPTION,
> > > + EXYNOS4_GPS_ALIVE_OPTION,
> > > +};
> > > +
> > > +static void exynos4412_pmu_central_seq(bool enable)
> >
> > You name the argument as "enable" but during initialization and
> > system running you pass here false. It confuses me. What do you enable
> > here?
> >
>
> Yep your are correct need to drop this function as already done in
> common frame work.
>
> > > +{
> > > + unsigned int value;
> > > +
> > > + value = pmu_raw_readl(S5P_CENTRAL_SEQ_CONFIGURATION);
> > > + if (enable)
> > > + value &= ~S5P_CENTRAL_LOWPWR_CFG;
> > > + else
> > > + value |= S5P_CENTRAL_LOWPWR_CFG;
> > > + pmu_raw_writel(value, S5P_CENTRAL_SEQ_CONFIGURATION);
> >
> > You duplicate exynos_pm_central_suspend() without removing the original code.
> >
> > > +
> > > + value = pmu_raw_readl(S5P_CENTRAL_SEQ_CONFIGURATION_COREBLK);
> > > + if (enable)
> > > + value &= ~S5P_CENTRAL_LOWPWR_CFG;
> > > + else
> > > + value |= S5P_CENTRAL_LOWPWR_CFG;
> >
> > As manual says - set this register only if you disable C2C. Our entire
> > low power configuration for STOP mode is for C2C enabled case so you
> > add inconsistent configuration.
>
> Ok Sorry I overlook this code change.
>
> Enable system power down. Set only CENTRAL_SEQ_CONFIGURATION register
> if you disable C2C.
> Set both CENTRAL_SEQ_CONFIGURATION and
> CENTRAL_SEQ_CONFIGURATION_COREBLK registers if you enable C2C.
>
> > > + pmu_raw_writel(value, S5P_CENTRAL_SEQ_CONFIGURATION_COREBLK);
> > > +}
> > > +
> > > +static void exynos4412_pmu_init(void)
> > > +{
> > > + unsigned int value;
> > > + int i;
> > > +
> > > + /* Enable USE_STANDBY_WFI for all CORE */
> > > + pmu_raw_writel(S5P_USE_STANDBY_WFI_ALL, S5P_CENTRAL_SEQ_OPTION);
> >
> > This does not look related to improving suspend... it looks unrelated.
> >
>
> Ok yes your are correct all ready done in comment frame work.
>
> "Execute WFI/WFE for all CPU cores. As soon as all the CPU cores in
> Exynos 4412 SCP enter STANDBY mode"
> > > +
> > > + /* Decides whether to use retention capability */
> > > + value = pmu_raw_readl(S5P_ARM_L2_0_OPTION);
> > > + value &= ~EXYNOS_L2_USE_RETENTION;
> > > + pmu_raw_writel(value, S5P_ARM_L2_0_OPTION);
> > > +
> > > + value = pmu_raw_readl(S5P_ARM_L2_1_OPTION);
> > > + value &= ~EXYNOS_L2_USE_RETENTION;
> > > + pmu_raw_writel(value, S5P_ARM_L2_1_OPTION);
> > > +
> > > + /* Set PSHOLD port for output high */
> > > + value = pmu_raw_readl(S5P_PS_HOLD_CONTROL);
> > > + value |= S5P_PS_HOLD_OUTPUT_HIGH;
> > > + pmu_raw_writel(value, S5P_PS_HOLD_CONTROL);
> > > +
> > > + /* Enable signal for PSHOLD port */
> > > + value = pmu_raw_readl(S5P_PS_HOLD_CONTROL);
> > > + value |= S5P_PS_HOLD_EN;
> > > + pmu_raw_writel(value, S5P_PS_HOLD_CONTROL);
> > > +
> >
> > The same - not related.
>
> PS_HOLD_CONTROL need to be set under following set
> Hardware reset Reset
> Watchdog timer reset Keep its value
> Software reset Keep its value
> Warm reset Keep its value
> Wakeup reset Keep its value

Which is not related to suspend...

>
> >
> > > + /* Enable only SC_FEEDBACK */
> > > + for (i = 0; i < ARRAY_SIZE(exynos4412_list_feed); i++) {
> > > + value = pmu_raw_readl(exynos4412_list_feed[i]);
> > > + value &= ~(EXYNOS_USE_SC_COUNTER);
> > > + value |= EXYNOS_USE_SC_FEEDBACK;
> > > + pmu_raw_writel(value, exynos4412_list_feed[i]);
> >
> > Why do you prefer to use feedback instead of counter?
> >
>
> Their is a note below if we use this setting.
> NOTE: Either one of USE_SC_FEEDBACK and USE_SC_COUNTER should be activated.

This is not the answer to my question at all.

Best regards,
Krzysztof

2019-02-15 15:37:38

by Anand Moon

Subject: Re: [RFC 2/2] soc: samsung: pmu: Add the PMU data of exynos4412 to support low-power state

Hi Krzysztof,

On Fri, 15 Feb 2019 at 13:03, Krzysztof Kozlowski <[email protected]> wrote:
>
> On Thu, 14 Feb 2019 at 19:37, Anand Moon <[email protected]> wrote:
> >
> > Hi Krzysztof,
> >
> > Thanks for your review comments.
> >
> > On Thu, 14 Feb 2019 at 18:29, Krzysztof Kozlowski <[email protected]> wrote:
> > >
> > > On Wed, 13 Feb 2019 at 22:41, Anand Moon <[email protected]> wrote:
> > > >
> > > > This patch adds configration for PMU (Power Management Unit) state
> > > > tuning for exynos4412 SoC in order to enter low-power mode during
> > > > suspend power modes and help resume from suspend state.
> > >
> > > The U3 and Trats2 already enter STOP/S2R so please describe what
> > > exactly you change.
> > >
> > > > Fixes: bfce552d0b1 ("drivers: soc: Add support for Exynos PMU driver")
> > >
> > > How it fixes it? What was broken in that commit?
> >
> > * I was not aware on their is common framework for suspend and resume
> > other than setting this here.I only look in to some the other exynos
> > pmu architecture
> > and referring 3.10.x kernel to model my changes.*
>
> Suspend in general should be working already so adding this code just
> because is not a valid reason. Please specify what is not working
> first.
>

Please accept *my apology* for not studying the code changes required.
Sorry for wasting your precious time in review.

My only intention with this patch was to correctly initialize some PMU
parameters.
which might be wrong. so lets drop this.

> > >
> > > > Cc: Marek Szyprowski <[email protected]>
> > > > Cc: Krzysztof Kozlowski <[email protected]>
> > > > Cc: Chanwoo Choi <[email protected]>
> > > > Signed-off-by: Anand Moon <[email protected]>
> > > > ---
> > > >
> > > > Changes from previous patch.
> > > > New patch to this series to support suspend and resume state
> > > >
> > > > Changes have been tested on microSD card but fails to resume on cMMC.
> > > > It need to be investigated and more debuging
> > > > ---
> > > > drivers/soc/samsung/exynos4-pmu.c | 83 +++++++++++++++++++++
> > > > include/linux/soc/samsung/exynos-regs-pmu.h | 21 ++++++
> > > > 2 files changed, 104 insertions(+)
> > > >
> > > > diff --git a/drivers/soc/samsung/exynos4-pmu.c b/drivers/soc/samsung/exynos4-pmu.c
> > > > index a7cdbf1aac0c..d261a0d2371e 100644
> > > > --- a/drivers/soc/samsung/exynos4-pmu.c
> > > > +++ b/drivers/soc/samsung/exynos4-pmu.c
> > > > @@ -200,10 +200,93 @@ static const struct exynos_pmu_conf exynos4412_pmu_config[] = {
> > > > { PMU_TABLE_END,},
> > > > };
> > > >
> > > > +static unsigned int const exynos4412_list_feed[] = {
> > > > + EXYNOS4_ARM_CORE0_OPTION,
> > > > + EXYNOS4_ARM_CORE1_OPTION,
> > > > + EXYNOS4_ARM_CORE2_OPTION,
> > > > + EXYNOS4_ARM_CORE3_OPTION,
> > > > + EXYNOS4_ARM_COMMON_OPTION,
> > > > + EXYNOS4_CAM_OPTION,
> > > > + EXYNOS4_TV_OPTION,
> > > > + EXYNOS4_MFC_OPTION,
> > > > + EXYNOS4_G3D_OPTION,
> > > > + EXYNOS4_LCD0_OPTION,
> > > > + EXYNOS4_ISP_OPTION,
> > > > + EXYNOS4_MAUDIO_OPTION,
> > > > + EXYNOS4_GPS_OPTION,
> > > > + EXYNOS4_GPS_ALIVE_OPTION,
> > > > +};
> > > > +
> > > > +static void exynos4412_pmu_central_seq(bool enable)
> > >
> > > You name the argument as "enable" but during initialization and
> > > system running you pass here false. It confuses me. What do you enable
> > > here?
> > >
> >
> > Yep your are correct need to drop this function as already done in
> > common frame work.
> >
> > > > +{
> > > > + unsigned int value;
> > > > +
> > > > + value = pmu_raw_readl(S5P_CENTRAL_SEQ_CONFIGURATION);
> > > > + if (enable)
> > > > + value &= ~S5P_CENTRAL_LOWPWR_CFG;
> > > > + else
> > > > + value |= S5P_CENTRAL_LOWPWR_CFG;
> > > > + pmu_raw_writel(value, S5P_CENTRAL_SEQ_CONFIGURATION);
> > >
> > > You duplicate exynos_pm_central_suspend() without removing the original code.
> > >
> > > > +
> > > > + value = pmu_raw_readl(S5P_CENTRAL_SEQ_CONFIGURATION_COREBLK);
> > > > + if (enable)
> > > > + value &= ~S5P_CENTRAL_LOWPWR_CFG;
> > > > + else
> > > > + value |= S5P_CENTRAL_LOWPWR_CFG;
> > >
> > > As manual says - set this register only if you disable C2C. Our entire
> > > low power configuration for STOP mode is for C2C enabled case so you
> > > add inconsistent configuration.
> >
> > Ok Sorry I overlook this code change.
> >
> > Enable system power down. Set only CENTRAL_SEQ_CONFIGURATION register
> > if you disable C2C.
> > Set both CENTRAL_SEQ_CONFIGURATION and
> > CENTRAL_SEQ_CONFIGURATION_COREBLK registers if you enable C2C.
> >
> > > > + pmu_raw_writel(value, S5P_CENTRAL_SEQ_CONFIGURATION_COREBLK);
> > > > +}
> > > > +
> > > > +static void exynos4412_pmu_init(void)
> > > > +{
> > > > + unsigned int value;
> > > > + int i;
> > > > +
> > > > + /* Enable USE_STANDBY_WFI for all CORE */
> > > > + pmu_raw_writel(S5P_USE_STANDBY_WFI_ALL, S5P_CENTRAL_SEQ_OPTION);
> > >
> > > This does not look related to improving suspend... it looks unrelated.
> > >
> >
> > Ok yes your are correct all ready done in comment frame work.
> >
> > "Execute WFI/WFE for all CPU cores. As soon as all the CPU cores in
> > Exynos 4412 SCP enter STANDBY mode"
> > > > +
> > > > + /* Decides whether to use retention capability */
> > > > + value = pmu_raw_readl(S5P_ARM_L2_0_OPTION);
> > > > + value &= ~EXYNOS_L2_USE_RETENTION;
> > > > + pmu_raw_writel(value, S5P_ARM_L2_0_OPTION);
> > > > +
> > > > + value = pmu_raw_readl(S5P_ARM_L2_1_OPTION);
> > > > + value &= ~EXYNOS_L2_USE_RETENTION;
> > > > + pmu_raw_writel(value, S5P_ARM_L2_1_OPTION);
> > > > +
> > > > + /* Set PSHOLD port for output high */
> > > > + value = pmu_raw_readl(S5P_PS_HOLD_CONTROL);
> > > > + value |= S5P_PS_HOLD_OUTPUT_HIGH;
> > > > + pmu_raw_writel(value, S5P_PS_HOLD_CONTROL);
> > > > +
> > > > + /* Enable signal for PSHOLD port */
> > > > + value = pmu_raw_readl(S5P_PS_HOLD_CONTROL);
> > > > + value |= S5P_PS_HOLD_EN;
> > > > + pmu_raw_writel(value, S5P_PS_HOLD_CONTROL);
> > > > +
> > >
> > > The same - not related.
> >
> > PS_HOLD_CONTROL need to be set under following set
> > Hardware reset Reset
> > Watchdog timer reset Keep its value
> > Software reset Keep its value
> > Warm reset Keep its value
> > Wakeup reset Keep its value
>
> Which is not related to suspend...
>
> >
> > >
> > > > + /* Enable only SC_FEEDBACK */
> > > > + for (i = 0; i < ARRAY_SIZE(exynos4412_list_feed); i++) {
> > > > + value = pmu_raw_readl(exynos4412_list_feed[i]);
> > > > + value &= ~(EXYNOS_USE_SC_COUNTER);
> > > > + value |= EXYNOS_USE_SC_FEEDBACK;
> > > > + pmu_raw_writel(value, exynos4412_list_feed[i]);
> > >
> > > Why do you prefer to use feedback instead of counter?
> > >
> >
> > Their is a note below if we use this setting.
> > NOTE: Either one of USE_SC_FEEDBACK and USE_SC_COUNTER should be activated.
>
> This is not the answer to my question at all.
>
> Best regards,
> Krzysztof

Once again sorry for wasting your time.
I know I have the improve a lot of things.

Best Regards

-Anand

2019-02-15 15:38:58

by Anand Moon

Subject: Re: [RFC 1/2] ARM: dts: exynos: Add proper regulator states for suspend-to-mem for odroid-u3

Hi Krzysztof,

On Fri, 15 Feb 2019 at 13:01, Krzysztof Kozlowski <[email protected]> wrote:
>
> On Thu, 14 Feb 2019 at 19:35, Anand Moon <[email protected]> wrote:
> >
> > hi Krzysztof,
> >
> > Thanks fro your review comments.
> >
> > On Thu, 14 Feb 2019 at 18:11, Krzysztof Kozlowski <[email protected]> wrote:
> > >
> > > Hi Anand,
> > >
> > > Thanks for the patch. See comments below.
> > >
> > > On Wed, 13 Feb 2019 at 22:41, Anand Moon <[email protected]> wrote:
> > > >
> > > > Add suspend-to-mem node to regulator core to be enabled or disabled
> > > > during system suspend and also support changing the regulator operating
> > > > mode during runtime and when the system enter sleep mode.
> > > >
> > > > Cc: Marek Szyprowski <[email protected]>
> > > > Cc: Krzysztof Kozlowski <[email protected]>
> > > > Cc: Chanwoo Choi <[email protected]>
> > > > Signed-off-by: Anand Moon <[email protected]>
> > > > ---
> > > >
> > > > Changes from previos patch
> > > > [0] https://patchwork.kernel.org/patch/10712549/
> > > >
> > > > Set all the WAKEUP source regulator in suspend-on state.
> > > > LD04, LD012, LD015, LD020, LD022
> > > > Set all the non used regulator in suspend-odd state
> > > > LD02, LD03, LD05, LD06, LD07, LD011, LD013, LDO14, LD016
> > > >
> > > > BUCK5, BUCK6, BUCK7 and not confirable as per driver max77686-regulator
> > > >
> > > > Tested on microSD card and it resumes correcly after suspend.
> > > > eMMC is not able to resume after entering into suspend state,
> > > > which need to be investigated and how to debug more.
> > > > ---
> > > > .../boot/dts/exynos4412-odroid-common.dtsi | 63 +++++++++++++++++++
> > > > arch/arm/boot/dts/exynos4412-odroidu3.dts | 3 +
> > > > 2 files changed, 66 insertions(+)
> > > >
> > > > diff --git a/arch/arm/boot/dts/exynos4412-odroid-common.dtsi b/arch/arm/boot/dts/exynos4412-odroid-common.dtsi
> > > > index 08d3a0a7b4eb..e984461c37d9 100644
> > > > --- a/arch/arm/boot/dts/exynos4412-odroid-common.dtsi
> > > > +++ b/arch/arm/boot/dts/exynos4412-odroid-common.dtsi
> > > > @@ -288,6 +288,9 @@
> > > > regulator-min-microvolt = <1800000>;
> > > > regulator-max-microvolt = <1800000>;
> > > > regulator-always-on;
> > > > + regulator-state-mem {
> > > > + regulator-off-in-suspend;
> > > > + };
> > >
> > > I see my comment from previous patch was not addressed.
> >
> > I left this unchanged since this regulator is not active linked and used.
> > Ok I will change this to regulator-on-in-suspend,
>
> Why?
>
> Previous patch was setting this to "on". I said that this is noop and
> if you want to add it, I need explanation why this regulator has to
> stay on during suspend.
>
> So you changed to "off"... which is still noop... and you did not
> provide explanation. Now you replied that you will change to "on"...
> so this will be circle. Still without explanation.
>

I will re check with my previous patch and the driver code which
support suspend state.
Ok check the regulator driver max77686-regulator and enable the logs their.

> >
> > From documentation device tree binding regulator.txt
> > "- regulator-always-on: boolean, regulator should never be disabled"
> >
> > But some regulator switches to a low power mode under light current loads
> > and the device automatically switches back to a fast response mode as the
> > output load increases.
> >
> > >
> > > > };
> > > >
> > > > ldo3_reg: LDO3 {
> > > > @@ -295,6 +298,9 @@
> > > > regulator-min-microvolt = <1800000>;
> > > > regulator-max-microvolt = <1800000>;
> > > > regulator-always-on;
> > > > + regulator-state-mem {
> > > > + regulator-off-in-suspend;
> > > > + };
> > >
> > > The same...
> >
> > Ok I will change this to regulator-on-in-suspend,
>
> The same - why changing this to on or off? I need explanation why this
> should be in first place.

I will check in the driver code.

>
> > >
> > > > };
> > > >
> > > > ldo4_reg: LDO4 {
> > > > @@ -302,6 +308,9 @@
> > > > regulator-min-microvolt = <2800000>;
> > > > regulator-max-microvolt = <2800000>;
> > > > regulator-boot-on;
> > > > + regulator-state-mem {
> > > > + regulator-on-in-suspend;
> > > > + };
> > > > };
> > > >
> > > > ldo5_reg: LDO5 {
> > > > @@ -310,6 +319,9 @@
> > > > regulator-max-microvolt = <1800000>;
> > > > regulator-always-on;
> > > > regulator-boot-on;
> > > > + regulator-state-mem {
> > > > + regulator-off-in-suspend;
> > > > + };
> > > > };
> > > >
> > > > ldo6_reg: LDO6 {
> > > > @@ -317,6 +329,9 @@
> > > > regulator-min-microvolt = <1000000>;
> > > > regulator-max-microvolt = <1000000>;
> > > > regulator-always-on;
> > > > + regulator-state-mem {
> > > > + regulator-off-in-suspend;
> > > > + };
> > > > };
> > > >
> > > > ldo7_reg: LDO7 {
> > > > @@ -324,18 +339,27 @@
> > > > regulator-min-microvolt = <1000000>;
> > > > regulator-max-microvolt = <1000000>;
> > > > regulator-always-on;
> > > > + regulator-state-mem {
> > > > + regulator-off-in-suspend;
> > > > + };
> > > > };
> > > >
> > > > ldo8_reg: LDO8 {
> > > > regulator-name = "VDD10_HDMI_1.0V";
> > > > regulator-min-microvolt = <1000000>;
> > > > regulator-max-microvolt = <1000000>;
> > > > + regulator-state-mem {
> > > > + regulator-on-in-suspend;
> > > > + };
> > > > };
> > > >
> > > > ldo10_reg: LDO10 {
> > > > regulator-name = "VDDQ_MIPIHSI_1.8V";
> > > > regulator-min-microvolt = <1800000>;
> > > > regulator-max-microvolt = <1800000>;
> > > > + regulator-state-mem {
> > > > + regulator-on-in-suspend;
> > > > + };
> > > > };
> > > >
> > > > ldo11_reg: LDO11 {
> > > > @@ -343,6 +367,9 @@
> > > > regulator-min-microvolt = <1800000>;
> > > > regulator-max-microvolt = <1800000>;
> > > > regulator-always-on;
> > > > + regulator-state-mem {
> > > > + regulator-off-in-suspend;
> > > > + };
> > > > };
> > > >
> > > > ldo12_reg: LDO12 {
> > > > @@ -351,6 +378,9 @@
> > > > regulator-max-microvolt = <3300000>;
> > > > regulator-always-on;
> > > > regulator-boot-on;
> > > > + regulator-state-mem {
> > > > + regulator-off-in-suspend;
> > > > + };
> > > > };
> > > >
> > > > ldo13_reg: LDO13 {
> > > > @@ -359,6 +389,9 @@
> > > > regulator-max-microvolt = <1800000>;
> > > > regulator-always-on;
> > > > regulator-boot-on;
> > > > + regulator-state-mem {
> > > > + regulator-off-in-suspend;
> > > > + };
> > > > };
> > > >
> > > > ldo14_reg: LDO14 {
> > > > @@ -367,6 +400,9 @@
> > > > regulator-max-microvolt = <1800000>;
> > > > regulator-always-on;
> > > > regulator-boot-on;
> > > > + regulator-state-mem {
> > > > + regulator-off-in-suspend;
> > > > + };
> > > > };
> > > >
> > > > ldo15_reg: LDO15 {
> > > > @@ -375,6 +411,9 @@
> > > > regulator-max-microvolt = <1000000>;
> > > > regulator-always-on;
> > > > regulator-boot-on;
> > > > + regulator-state-mem {
> > > > + regulator-off-in-suspend;
> > > > + };
> > > > };
> > > >
> > > > ldo16_reg: LDO16 {
> > > > @@ -383,6 +422,9 @@
> > > > regulator-max-microvolt = <1800000>;
> > > > regulator-always-on;
> > > > regulator-boot-on;
> > > > + regulator-state-mem {
> > > > + regulator-off-in-suspend;
> > > > + };
> > > > };
> > > >
> > > > ldo20_reg: LDO20 {
> > > > @@ -396,6 +438,9 @@
> > > > regulator-min-microvolt = <2800000>;
> > > > regulator-max-microvolt = <2800000>;
> > > > regulator-boot-on;
> > > > + regulator-state-mem {
> > > > + regulator-on-in-suspend;
> > > > + };
> > >
> > >
> > > The same... any comments?
> >
> > I left this on for mshc_0 (emmc) regulator,
> > since I could not get suspend resume working on eMMc
>
> In general eMMC can be turned off during suspend. However eMMC layer
> may be handling this (manually turning off/on). So after checking and
> confirming it, please document it.

I will check in the driver code.

>
> > >
> > > > };
> > > >
> > > > ldo22_reg: LDO22 {
> > > > @@ -405,6 +450,9 @@
> > > > */
> > > > regulator-name = "LDO22";
> > > > regulator-boot-on;
> > > > + regulator-state-mem {
> > > > + regulator-on-in-suspend;
> > > > + };
> > > > };
> > > >
> > > > ldo25_reg: LDO25 {
> > > > @@ -413,6 +461,9 @@
> > > > regulator-max-microvolt = <1800000>;
> > > > regulator-always-on;
> > > > regulator-boot-on;
> > > > + regulator-state-mem {
> > > > + regulator-off-in-suspend;
> > > > + };
> > > > };
> > > >
> > > > buck1_reg: BUCK1 {
> > > > @@ -421,6 +472,9 @@
> > > > regulator-max-microvolt = <1100000>;
> > > > regulator-always-on;
> > > > regulator-boot-on;
> > > > + regulator-state-mem {
> > > > + regulator-off-in-suspend;
> > >
> > > Again, you did not address my comments.
> > >
> >
> > Buck1 support entering into LPM setting,
> > But I will set this to regulator-on-in-suspend; in next patch.
>
> Why?

I will check in the driver code.

>
> > > > + };
> > > > };
> > > >
> > > > buck2_reg: BUCK2 {
> > > > @@ -429,6 +483,9 @@
> > > > regulator-max-microvolt = <1350000>;
> > > > regulator-always-on;
> > > > regulator-boot-on;
> > > > + regulator-state-mem {
> > > > + regulator-on-in-suspend;
> > > > + };
> > > > };
> > > >
> > > > buck3_reg: BUCK3 {
> > > > @@ -437,6 +494,9 @@
> > > > regulator-max-microvolt = <1050000>;
> > > > regulator-always-on;
> > > > regulator-boot-on;
> > > > + regulator-state-mem {
> > > > + regulator-off-in-suspend;
> > > > + };
> > >
> > > The same...
> >
> > Buck3 support entering into LPM setting,
> > But I will set this to regulator-on-in-suspend; in next patch.
>
> Again - why this should be on or off?
>
> > >
> > > > };
> > > >
> > > > buck4_reg: BUCK4 {
> > > > @@ -444,6 +504,9 @@
> > > > regulator-min-microvolt = <900000>;
> > > > regulator-max-microvolt = <1100000>;
> > > > regulator-microvolt-offset = <50000>;
> > > > + regulator-state-mem {
> > > > + regulator-off-in-suspend;
> > > > + };
> > > > };
> > > >
> > > > buck5_reg: BUCK5 {
> > > > diff --git a/arch/arm/boot/dts/exynos4412-odroidu3.dts b/arch/arm/boot/dts/exynos4412-odroidu3.dts
> > > > index 2bdf899df436..4ebde09fc51d 100644
> > > > --- a/arch/arm/boot/dts/exynos4412-odroidu3.dts
> > > > +++ b/arch/arm/boot/dts/exynos4412-odroidu3.dts
> > > > @@ -82,6 +82,9 @@
> > > > regulator-name = "LDO22_VDDQ_MMC4_2.8V";
> > > > regulator-min-microvolt = <2800000>;
> > > > regulator-max-microvolt = <2800000>;
> > > > + regulator-state-mem {
> > > > + regulator-on-in-suspend;
> > >
> > > Why?
> > >
> >
> > I chose not to disable mshc_0 (emmc) regulator,
> > since I could not get suspend resume working on eMMC.
>
> Having "regulator-on-in-suspend" is not the same as not disabling
> regulator during suspend. This property means you will explicitly
> enable it during suspend. To me it looks wrong so I would be happy to
> see explanations.
>

Ok I will disable this and check this help me fix the issue I am observing.

> Best regards,
> Krzysztof

*Sorry for wasting your time*
I know I have the improve a lot of things.

Best regards

-Anand

2019-02-15 16:19:57

by Anand Moon

Subject: Re: [RFC 1/2] ARM: dts: exynos: Add proper regulator states for suspend-to-mem for odroid-u3

Hi Krzysztof,

On Fri, 15 Feb 2019 at 13:01, Krzysztof Kozlowski <[email protected]> wrote:
>
> On Thu, 14 Feb 2019 at 19:35, Anand Moon <[email protected]> wrote:
> >
> > hi Krzysztof,
> >
> > Thanks fro your review comments.
> >
> > On Thu, 14 Feb 2019 at 18:11, Krzysztof Kozlowski <[email protected]> wrote:
> > >
> > > Hi Anand,
> > >
> > > Thanks for the patch. See comments below.
> > >
> > > On Wed, 13 Feb 2019 at 22:41, Anand Moon <[email protected]> wrote:
> > > >
> > > > Add suspend-to-mem node to regulator core to be enabled or disabled
> > > > during system suspend and also support changing the regulator operating
> > > > mode during runtime and when the system enter sleep mode.
> > > >
> > > > Cc: Marek Szyprowski <[email protected]>
> > > > Cc: Krzysztof Kozlowski <[email protected]>
> > > > Cc: Chanwoo Choi <[email protected]>
> > > > Signed-off-by: Anand Moon <[email protected]>
> > > > ---
> > > >
> > > > Changes from previos patch
> > > > [0] https://patchwork.kernel.org/patch/10712549/
> > > >
> > > > Set all the WAKEUP source regulator in suspend-on state.
> > > > LD04, LD012, LD015, LD020, LD022
> > > > Set all the non used regulator in suspend-odd state
> > > > LD02, LD03, LD05, LD06, LD07, LD011, LD013, LDO14, LD016
> > > >
> > > > BUCK5, BUCK6, BUCK7 and not confirable as per driver max77686-regulator
> > > >
> > > > Tested on microSD card and it resumes correcly after suspend.
> > > > eMMC is not able to resume after entering into suspend state,
> > > > which need to be investigated and how to debug more.
> > > > ---
> > > > .../boot/dts/exynos4412-odroid-common.dtsi | 63 +++++++++++++++++++
> > > > arch/arm/boot/dts/exynos4412-odroidu3.dts | 3 +
> > > > 2 files changed, 66 insertions(+)
> > > >
> > > > diff --git a/arch/arm/boot/dts/exynos4412-odroid-common.dtsi b/arch/arm/boot/dts/exynos4412-odroid-common.dtsi
> > > > index 08d3a0a7b4eb..e984461c37d9 100644
> > > > --- a/arch/arm/boot/dts/exynos4412-odroid-common.dtsi
> > > > +++ b/arch/arm/boot/dts/exynos4412-odroid-common.dtsi
> > > > @@ -288,6 +288,9 @@
> > > > regulator-min-microvolt = <1800000>;
> > > > regulator-max-microvolt = <1800000>;
> > > > regulator-always-on;
> > > > + regulator-state-mem {
> > > > + regulator-off-in-suspend;
> > > > + };
> > >
> > > I see my comment from previous patch was not addressed.
> >
> > I left this unchanged since this regulator is not active linked and used.
> > Ok I will change this to regulator-on-in-suspend,
>
> Why?
>
> Previous patch was setting this to "on". I said that this is noop and
> if you want to add it, I need explanation why this regulator has to
> stay on during suspend.
>
> So you changed to "off"... which is still noop... and you did not
> provide explanation. Now you replied that you will change to "on"...
> so this will be circle. Still without explanation.
>
------------------------------------------------------------------------------------------------
I have added some debug prints to get more inputs.

Here is the boot logs for sdcard suspend resume

[1] https://pastebin.com/wvJvJidp

[root@archl-u3m ~]# rtcwake -d /dev/rtc0 -m mem -s 10
rtcwake: assuming RTC uses UTC ...
rtcwake: wakeup from "mem" using /dev/rtc0 at Fri Feb 15 13:28:06 2019
[ 97.165349] PM: suspend entry (deep)
[ 97.165846] PM: Syncing filesystems ... done.
[ 97.344436] Freezing user space processes ... (elapsed 0.006 seconds) done.
[ 97.354738] OOM killer disabled.
[ 97.356028] Freezing remaining freezable tasks ... (elapsed 0.004
seconds) done.
[ 97.413453] smsc95xx 1-2:1.0 eth0: entering SUSPEND2 mode
[ 97.423227] sd 0:0:0:0: [sda] Synchronizing SCSI cache
[ 97.444346] dwc2 12480000.hsotg: suspending usb gadget g_ether
[ 97.446728] dwc2 12480000.hsotg: dwc2_hsotg_ep_disable: called for ep0
[ 97.451287] dwc2 12480000.hsotg: dwc2_hsotg_ep_disable: called for ep0
[ 97.457742] dwc2 12480000.hsotg: new device is full-speed
[ 97.465766] wake enabled for irq 119
[ 97.466979] BUCK9: No configuration
[ 97.470417] BUCK8_P3V3: No configuration
[ 97.474131] BUCK7_2.0V: No configuration
[ 97.477954] BUCK6_1.35V: No configuration
[ 97.481946] VDDQ_CKEM1_2_1.2V: No configuration
[ 97.487930] BUCK4: regulator suspend disable supported
[ 97.491964] BUCK3: regulator suspend disable supported
[ 97.497184] BUCK1: regulator suspend disable supported
[ 97.501838] LDO26: No configuration
[ 97.505383] LDO24: No configuration
[ 97.508761] LDO23: No configuration
[ 97.512368] LDO19: No configuration
[ 97.515706] LDO18: No configuration
[ 97.519178] LDO17: No configuration
[ 97.523076] LDO16: regulator suspend disable supported
[ 97.528154] LDO15: regulator suspend disable supported
[ 97.533367] LDO14: regulator suspend disable supported
[ 97.538447] LDO12: regulator suspend disable supported
[ 97.543603] LDO11: regulator suspend disable supported
[ 97.548308] LDO9: No configuration
[ 97.552107] LDO7: regulator suspend disable supported
[ 97.557064] LDO6: regulator suspend disable supported
[ 97.562251] LDO2: regulator suspend disable supported
[ 97.566769] VDD_ALIVE_1.0V: No configuration
[ 97.581717] usb3503 0-0008: switched to STANDBY mode
[ 97.582496] wake enabled for irq 123
[ 97.602770] samsung-pinctrl 11000000.pinctrl: Setting external
wakeup interrupt mask: 0xfbfff7ff
[ 97.623616] Disabling non-boot CPUs ...
[ 97.681453] s3c2410-wdt 10060000.watchdog: watchdog disabled
[ 97.681743] wake disabled for irq 123
[ 97.692569] usb3503 0-0008: switched to HUB mode
[ 97.801553] wake disabled for irq 119
[ 97.802087] dwc2 12480000.hsotg: resuming usb gadget g_ether
[ 98.080458] usb 1-2: reset high-speed USB device number 2 using exynos-ehci
[ 98.432321] usb 1-3: reset high-speed USB device number 3 using exynos-ehci
[ 98.970441] usb 1-3.3: reset high-speed USB device number 4 using exynos-ehci
[ 99.135835] OOM killer enabled.
[ 99.135940] Restarting tasks ... done.
[ 99.140760] PM: suspend exit
[root@archl-u3m ~]# [ 99.877843] smsc95xx 1-2:1.0 eth0: link up,
100Mbps, full-duplex, lpa 0xC5E1

------------------------------------------------
Here is the boot log for emmc suspend resume. after disable of emmc regulator.

[2] https://pastebin.com/Z05K3kj2

rtcwake: wakeup from "mem" using /dev/rtc0 at Fri Feb 15 13:27:26 2019
[ 79.775108] PM: suspend entry (deep)
[ 79.775462] PM: Syncing filesystems ... done.
[ 79.819009] Freezing user space processes ... (elapsed 0.004 seconds) done.
[ 79.826300] OOM killer disabled.
[ 79.828603] Freezing remaining freezable tasks ... (elapsed 0.001
seconds) done.
[ 79.875967] smsc95xx 1-2:1.0 eth0: entering SUSPEND2 mode
[ 79.959495] dwc2 12480000.hsotg: suspending usb gadget g_ether
[ 79.960480] dwc2 12480000.hsotg: dwc2_hsotg_ep_disable: called for ep0
[ 79.966431] dwc2 12480000.hsotg: new device is full-speed
[ 79.971805] dwc2 12480000.hsotg: dwc2_hsotg_ep_disable: called for ep0
[ 79.980568] wake enabled for irq 119
[ 79.982195] BUCK9: No configuration
[ 79.985540] BUCK8_P3V3: No configuration
[ 79.989201] BUCK7_2.0V: No configuration
[ 79.993175] BUCK6_1.35V: No configuration
[ 79.997099] VDDQ_CKEM1_2_1.2V: No configuration
[ 80.002765] BUCK4: regulator suspend disable supported
[ 80.007115] BUCK3: regulator suspend disable supported
[ 80.012346] BUCK1: regulator suspend disable supported
[ 80.016991] LDO26: No configuration
[ 80.020467] LDO24: No configuration
[ 80.023991] LDO23: No configuration
[ 80.027456] LDO19: No configuration
[ 80.030858] LDO18: No configuration
[ 80.034388] LDO17: No configuration
[ 80.038172] LDO16: regulator suspend disable supported
[ 80.043410] LDO15: regulator suspend disable supported
[ 80.048429] LDO14: regulator suspend disable supported
[ 80.053658] LDO12: regulator suspend disable supported
[ 80.058670] LDO11: regulator suspend disable supported
[ 80.063522] LDO9: No configuration
[ 80.067187] LDO7: regulator suspend disable supported
[ 80.072290] LDO6: regulator suspend disable supported
[ 80.077321] LDO2: regulator suspend disable supported
[ 80.081995] VDD_ALIVE_1.0V: No configuration
[ 80.096885] usb3503 0-0008: switched to STANDBY mode
[ 80.097568] wake enabled for irq 123
[ 80.116002] samsung-pinctrl 11000000.pinctrl: Setting external
wakeup interrupt mask: 0xfbfff7ff
[ 80.134695] Disabling non-boot CPUs ...

[snip]

> > > > diff --git a/arch/arm/boot/dts/exynos4412-odroidu3.dts b/arch/arm/boot/dts/exynos4412-odroidu3.dts
> > > > index 2bdf899df436..4ebde09fc51d 100644
> > > > --- a/arch/arm/boot/dts/exynos4412-odroidu3.dts
> > > > +++ b/arch/arm/boot/dts/exynos4412-odroidu3.dts
> > > > @@ -82,6 +82,9 @@
> > > > regulator-name = "LDO22_VDDQ_MMC4_2.8V";
> > > > regulator-min-microvolt = <2800000>;
> > > > regulator-max-microvolt = <2800000>;
> > > > + regulator-state-mem {
> > > > + regulator-on-in-suspend;
> > >
> > > Why?
> > >
> >
> > I chose not to disable mshc_0 (emmc) regulator,
> > since I could not get suspend resume working on eMMC.
>
> Having "regulator-on-in-suspend" is not the same as not disabling
> regulator during suspend. This property means you will explicitly
> enable it during suspend. To me it looks wrong so I would be happy to
> see explanations.
>

During suspend the driver set the regulator in standby mode until resume.
Opps I understood the regulator in suspend mode enter into

REGULATOR_MODE_STANDBY : MAX77686_OFF_PWRREQ;

[ 79.997099] VDDQ_CKEM1_2_1.2V: No configuration
[ 80.002765] BUCK4: regulator suspend disable supported
[ 80.007115] BUCK3: regulator suspend disable supported
[ 80.012346] BUCK1: regulator suspend disable supported
[ 80.016991] LDO26: No configuration
[ 80.020467] LDO24: No configuration
[ 80.023991] LDO23: No configuration
[ 80.027456] LDO19: No configuration
[ 80.030858] LDO18: No configuration
[ 80.034388] LDO17: No configuration
[ 80.038172] LDO16: regulator suspend disable supported
[ 80.043410] LDO15: regulator suspend disable supported
[ 80.048429] LDO14: regulator suspend disable supported
[ 80.053658] LDO12: regulator suspend disable supported
[ 80.058670] LDO11: regulator suspend disable supported
[ 80.063522] LDO9: No configuration
[ 80.067187] LDO7: regulator suspend disable supported
[ 80.072290] LDO6: regulator suspend disable supported
[ 80.077321] LDO2: regulator suspend disable supported

So as per the logs BUCK1 / BUCK3 / BUCK4 successfully entered into suspend mode.

Where as suspend is disabled in groups as per the documentation.
(Group 1)
Buck6-->Buck7-->LD01-->LD011-->LD014
Group 2)
Buck2-->Buck1-->Buck3-->Buck4
Group 3>
LD06-->LD013-->LD08-->LD015-->LD010-->LD016-->LD05-->LD012

Best Regards
-Anand