[PATCH v2 0/2] add support holding 24Mhz osc during suspend

[PATCH v2 0/2] add support holding 24Mhz osc during suspend

[Chris Zhong]

Since the usb-phy clocks are not connected to the 32khz clock-in, they
only have 24Mhz as the clock source. If we want to wake up system via
usb device, we should hold the 24Mhz osc during suspend.
This Patch has tested on chromebook jerry board, with another patch[0],
we can wakeup system via usb keyboard.

[0] https://chromium-review.googlesource.com/#/c/277811/3


Chris Zhong (2):
  ARM: rockchip: add support holding 24Mhz osc during suspend
  ARM: rockchip: remove some useless macro in pm.h

 arch/arm/mach-rockchip/pm.c | 38 ++++++++++++++++++++++++++++++++++++--
 arch/arm/mach-rockchip/pm.h |  7 -------
 2 files changed, 36 insertions(+), 9 deletions(-)

-- 
1.9.1

--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
Please read the FAQ at  http://www.tux.org/lkml/

[PATCH v2 0/2] add support holding 24Mhz osc during suspend

[Chris Zhong]

Since the usb-phy clocks are not connected to the 32khz clock-in, they
only have 24Mhz as the clock source. If we want to wake up system via
usb device, we should hold the 24Mhz osc during suspend.
This Patch has tested on chromebook jerry board, with another patch[0],
we can wakeup system via usb keyboard.

[0] https://chromium-review.googlesource.com/#/c/277811/3


Chris Zhong (2):
  ARM: rockchip: add support holding 24Mhz osc during suspend
  ARM: rockchip: remove some useless macro in pm.h

 arch/arm/mach-rockchip/pm.c | 38 ++++++++++++++++++++++++++++++++++++--
 arch/arm/mach-rockchip/pm.h |  7 -------
 2 files changed, 36 insertions(+), 9 deletions(-)

-- 
1.9.1

[PATCH v2 0/2] add support holding 24Mhz osc during suspend

[Chris Zhong]

Since the usb-phy clocks are not connected to the 32khz clock-in, they
only have 24Mhz as the clock source. If we want to wake up system via
usb device, we should hold the 24Mhz osc during suspend.
This Patch has tested on chromebook jerry board, with another patch[0],
we can wakeup system via usb keyboard.

[0] https://chromium-review.googlesource.com/#/c/277811/3


Chris Zhong (2):
  ARM: rockchip: add support holding 24Mhz osc during suspend
  ARM: rockchip: remove some useless macro in pm.h

 arch/arm/mach-rockchip/pm.c | 38 ++++++++++++++++++++++++++++++++++++--
 arch/arm/mach-rockchip/pm.h |  7 -------
 2 files changed, 36 insertions(+), 9 deletions(-)

-- 
1.9.1

[PATCH v2 1/2] ARM: rockchip: add support holding 24Mhz osc during suspend

[Chris Zhong]

If we want to wake up system via usb, the 24Mhz osc could not be
disabled during suspend, read the usb phy SIDDQ bit to decide whether
to switch to 32khz clock-in.

Signed-off-by: Chris Zhong <zyw@rock-chips.com>
---

 arch/arm/mach-rockchip/pm.c | 38 ++++++++++++++++++++++++++++++++++++--
 1 file changed, 36 insertions(+), 2 deletions(-)

diff --git a/arch/arm/mach-rockchip/pm.c b/arch/arm/mach-rockchip/pm.c
index b0dcbe2..fda81d3 100644
--- a/arch/arm/mach-rockchip/pm.c
+++ b/arch/arm/mach-rockchip/pm.c
@@ -29,6 +29,11 @@
 
 #include "pm.h"
 
+#define GRF_UOC0_CON0			0x320
+#define GRF_UOC1_CON0			0x334
+#define GRF_UOC2_CON0			0x348
+#define GRF_SIDDQ			BIT(13)
+
 /* These enum are option of low power mode */
 enum {
 	ROCKCHIP_ARM_OFF_LOGIC_NORMAL = 0,
@@ -45,6 +50,7 @@ static phys_addr_t rk3288_bootram_phy;
 
 static struct regmap *pmu_regmap;
 static struct regmap *sgrf_regmap;
+static struct regmap *grf_regmap;
 
 static u32 rk3288_pmu_pwr_mode_con;
 static u32 rk3288_sgrf_soc_con0;
@@ -66,9 +72,28 @@ static void rk3288_config_bootdata(void)
 	rkpm_bootdata_l2ctlr = rk3288_l2_config();
 }
 
+static bool rk3288_slp_disable_osc(void)
+{
+	static const u32 reg_offset[] = { GRF_UOC0_CON0, GRF_UOC1_CON0,
+					  GRF_UOC2_CON0 };
+	u32 reg, i;
+
+	/* if any usb phy is still on(GRF_SIDDQ==0), that means we need the
+	 * function of usb wakeup, so do not switch to 32khz, since the usb phy
+	 * clk does not connect to 32khz osc*/
+	for (i = 0; i < ARRAY_SIZE(reg_offset); i++) {
+		regmap_read(grf_regmap, reg_offset[i], &reg);
+		if (!(reg & GRF_SIDDQ))
+			return false;
+	}
+
+	return true;
+}
+
 static void rk3288_slp_mode_set(int level)
 {
 	u32 mode_set, mode_set1;
+	bool osc_switch_to_32k = rk3288_slp_disable_osc();
 
 	regmap_read(sgrf_regmap, RK3288_SGRF_SOC_CON0, &rk3288_sgrf_soc_con0);
 
@@ -107,11 +132,13 @@ static void rk3288_slp_mode_set(int level)
 
 	if (level == ROCKCHIP_ARM_OFF_LOGIC_DEEP) {
 		/* arm off, logic deep sleep */
-		mode_set |= BIT(PMU_BUS_PD_EN) |
+		mode_set |= BIT(PMU_BUS_PD_EN) | BIT(PMU_PMU_USE_LF) |
 			    BIT(PMU_DDR1IO_RET_EN) | BIT(PMU_DDR0IO_RET_EN) |
-			    BIT(PMU_OSC_24M_DIS) | BIT(PMU_PMU_USE_LF) |
 			    BIT(PMU_ALIVE_USE_LF) | BIT(PMU_PLL_PD_EN);
 
+		if (osc_switch_to_32k)
+			mode_set |= BIT(PMU_OSC_24M_DIS);
+
 		mode_set1 |= BIT(PMU_CLR_ALIVE) | BIT(PMU_CLR_BUS) |
 			     BIT(PMU_CLR_PERI) | BIT(PMU_CLR_DMA);
 	} else {
@@ -193,6 +220,13 @@ static int rk3288_suspend_init(struct device_node *np)
 		return PTR_ERR(pmu_regmap);
 	}
 
+	grf_regmap = syscon_regmap_lookup_by_compatible(
+				"rockchip,rk3288-grf");
+	if (IS_ERR(grf_regmap)) {
+		pr_err("%s: could not find grf regmap\n", __func__);
+		return PTR_ERR(pmu_regmap);
+	}
+
 	sram_np = of_find_compatible_node(NULL, NULL,
 					  "rockchip,rk3288-pmu-sram");
 	if (!sram_np) {
-- 
1.9.1

--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
Please read the FAQ at  http://www.tux.org/lkml/

[PATCH v2 1/2] ARM: rockchip: add support holding 24Mhz osc during suspend

[Chris Zhong]

If we want to wake up system via usb, the 24Mhz osc could not be
disabled during suspend, read the usb phy SIDDQ bit to decide whether
to switch to 32khz clock-in.

Signed-off-by: Chris Zhong <zyw@rock-chips.com>
---

 arch/arm/mach-rockchip/pm.c | 38 ++++++++++++++++++++++++++++++++++++--
 1 file changed, 36 insertions(+), 2 deletions(-)

diff --git a/arch/arm/mach-rockchip/pm.c b/arch/arm/mach-rockchip/pm.c
index b0dcbe2..fda81d3 100644
--- a/arch/arm/mach-rockchip/pm.c
+++ b/arch/arm/mach-rockchip/pm.c
@@ -29,6 +29,11 @@
 
 #include "pm.h"
 
+#define GRF_UOC0_CON0			0x320
+#define GRF_UOC1_CON0			0x334
+#define GRF_UOC2_CON0			0x348
+#define GRF_SIDDQ			BIT(13)
+
 /* These enum are option of low power mode */
 enum {
 	ROCKCHIP_ARM_OFF_LOGIC_NORMAL = 0,
@@ -45,6 +50,7 @@ static phys_addr_t rk3288_bootram_phy;
 
 static struct regmap *pmu_regmap;
 static struct regmap *sgrf_regmap;
+static struct regmap *grf_regmap;
 
 static u32 rk3288_pmu_pwr_mode_con;
 static u32 rk3288_sgrf_soc_con0;
@@ -66,9 +72,28 @@ static void rk3288_config_bootdata(void)
 	rkpm_bootdata_l2ctlr = rk3288_l2_config();
 }
 
+static bool rk3288_slp_disable_osc(void)
+{
+	static const u32 reg_offset[] = { GRF_UOC0_CON0, GRF_UOC1_CON0,
+					  GRF_UOC2_CON0 };
+	u32 reg, i;
+
+	/* if any usb phy is still on(GRF_SIDDQ==0), that means we need the
+	 * function of usb wakeup, so do not switch to 32khz, since the usb phy
+	 * clk does not connect to 32khz osc*/
+	for (i = 0; i < ARRAY_SIZE(reg_offset); i++) {
+		regmap_read(grf_regmap, reg_offset[i], &reg);
+		if (!(reg & GRF_SIDDQ))
+			return false;
+	}
+
+	return true;
+}
+
 static void rk3288_slp_mode_set(int level)
 {
 	u32 mode_set, mode_set1;
+	bool osc_switch_to_32k = rk3288_slp_disable_osc();
 
 	regmap_read(sgrf_regmap, RK3288_SGRF_SOC_CON0, &rk3288_sgrf_soc_con0);
 
@@ -107,11 +132,13 @@ static void rk3288_slp_mode_set(int level)
 
 	if (level == ROCKCHIP_ARM_OFF_LOGIC_DEEP) {
 		/* arm off, logic deep sleep */
-		mode_set |= BIT(PMU_BUS_PD_EN) |
+		mode_set |= BIT(PMU_BUS_PD_EN) | BIT(PMU_PMU_USE_LF) |
 			    BIT(PMU_DDR1IO_RET_EN) | BIT(PMU_DDR0IO_RET_EN) |
-			    BIT(PMU_OSC_24M_DIS) | BIT(PMU_PMU_USE_LF) |
 			    BIT(PMU_ALIVE_USE_LF) | BIT(PMU_PLL_PD_EN);
 
+		if (osc_switch_to_32k)
+			mode_set |= BIT(PMU_OSC_24M_DIS);
+
 		mode_set1 |= BIT(PMU_CLR_ALIVE) | BIT(PMU_CLR_BUS) |
 			     BIT(PMU_CLR_PERI) | BIT(PMU_CLR_DMA);
 	} else {
@@ -193,6 +220,13 @@ static int rk3288_suspend_init(struct device_node *np)
 		return PTR_ERR(pmu_regmap);
 	}
 
+	grf_regmap = syscon_regmap_lookup_by_compatible(
+				"rockchip,rk3288-grf");
+	if (IS_ERR(grf_regmap)) {
+		pr_err("%s: could not find grf regmap\n", __func__);
+		return PTR_ERR(pmu_regmap);
+	}
+
 	sram_np = of_find_compatible_node(NULL, NULL,
 					  "rockchip,rk3288-pmu-sram");
 	if (!sram_np) {
-- 
1.9.1

[PATCH v2 1/2] ARM: rockchip: add support holding 24Mhz osc during suspend

[Chris Zhong]

If we want to wake up system via usb, the 24Mhz osc could not be
disabled during suspend, read the usb phy SIDDQ bit to decide whether
to switch to 32khz clock-in.

Signed-off-by: Chris Zhong <zyw@rock-chips.com>
---

 arch/arm/mach-rockchip/pm.c | 38 ++++++++++++++++++++++++++++++++++++--
 1 file changed, 36 insertions(+), 2 deletions(-)

diff --git a/arch/arm/mach-rockchip/pm.c b/arch/arm/mach-rockchip/pm.c
index b0dcbe2..fda81d3 100644
--- a/arch/arm/mach-rockchip/pm.c
+++ b/arch/arm/mach-rockchip/pm.c
@@ -29,6 +29,11 @@
 
 #include "pm.h"
 
+#define GRF_UOC0_CON0			0x320
+#define GRF_UOC1_CON0			0x334
+#define GRF_UOC2_CON0			0x348
+#define GRF_SIDDQ			BIT(13)
+
 /* These enum are option of low power mode */
 enum {
 	ROCKCHIP_ARM_OFF_LOGIC_NORMAL = 0,
@@ -45,6 +50,7 @@ static phys_addr_t rk3288_bootram_phy;
 
 static struct regmap *pmu_regmap;
 static struct regmap *sgrf_regmap;
+static struct regmap *grf_regmap;
 
 static u32 rk3288_pmu_pwr_mode_con;
 static u32 rk3288_sgrf_soc_con0;
@@ -66,9 +72,28 @@ static void rk3288_config_bootdata(void)
 	rkpm_bootdata_l2ctlr = rk3288_l2_config();
 }
 
+static bool rk3288_slp_disable_osc(void)
+{
+	static const u32 reg_offset[] = { GRF_UOC0_CON0, GRF_UOC1_CON0,
+					  GRF_UOC2_CON0 };
+	u32 reg, i;
+
+	/* if any usb phy is still on(GRF_SIDDQ==0), that means we need the
+	 * function of usb wakeup, so do not switch to 32khz, since the usb phy
+	 * clk does not connect to 32khz osc*/
+	for (i = 0; i < ARRAY_SIZE(reg_offset); i++) {
+		regmap_read(grf_regmap, reg_offset[i], &reg);
+		if (!(reg & GRF_SIDDQ))
+			return false;
+	}
+
+	return true;
+}
+
 static void rk3288_slp_mode_set(int level)
 {
 	u32 mode_set, mode_set1;
+	bool osc_switch_to_32k = rk3288_slp_disable_osc();
 
 	regmap_read(sgrf_regmap, RK3288_SGRF_SOC_CON0, &rk3288_sgrf_soc_con0);
 
@@ -107,11 +132,13 @@ static void rk3288_slp_mode_set(int level)
 
 	if (level == ROCKCHIP_ARM_OFF_LOGIC_DEEP) {
 		/* arm off, logic deep sleep */
-		mode_set |= BIT(PMU_BUS_PD_EN) |
+		mode_set |= BIT(PMU_BUS_PD_EN) | BIT(PMU_PMU_USE_LF) |
 			    BIT(PMU_DDR1IO_RET_EN) | BIT(PMU_DDR0IO_RET_EN) |
-			    BIT(PMU_OSC_24M_DIS) | BIT(PMU_PMU_USE_LF) |
 			    BIT(PMU_ALIVE_USE_LF) | BIT(PMU_PLL_PD_EN);
 
+		if (osc_switch_to_32k)
+			mode_set |= BIT(PMU_OSC_24M_DIS);
+
 		mode_set1 |= BIT(PMU_CLR_ALIVE) | BIT(PMU_CLR_BUS) |
 			     BIT(PMU_CLR_PERI) | BIT(PMU_CLR_DMA);
 	} else {
@@ -193,6 +220,13 @@ static int rk3288_suspend_init(struct device_node *np)
 		return PTR_ERR(pmu_regmap);
 	}
 
+	grf_regmap = syscon_regmap_lookup_by_compatible(
+				"rockchip,rk3288-grf");
+	if (IS_ERR(grf_regmap)) {
+		pr_err("%s: could not find grf regmap\n", __func__);
+		return PTR_ERR(pmu_regmap);
+	}
+
 	sram_np = of_find_compatible_node(NULL, NULL,
 					  "rockchip,rk3288-pmu-sram");
 	if (!sram_np) {
-- 
1.9.1

[PATCH v2 2/2] ARM: rockchip: remove some useless macro in pm.h

[Chris Zhong]

These are actually not used in the pm code, as we moved suspend handling
to the clock driver, remove them here.

Signed-off-by: Chris Zhong <zyw@rock-chips.com>

---

 arch/arm/mach-rockchip/pm.h | 7 -------
 1 file changed, 7 deletions(-)

diff --git a/arch/arm/mach-rockchip/pm.h b/arch/arm/mach-rockchip/pm.h
index 3e8d39c..b6494c2 100644
--- a/arch/arm/mach-rockchip/pm.h
+++ b/arch/arm/mach-rockchip/pm.h
@@ -59,13 +59,6 @@ static inline void rockchip_suspend_init(void)
 #define SGRF_DAPDEVICEEN		BIT(0)
 #define SGRF_DAPDEVICEEN_WRITE		BIT(16)
 
-#define RK3288_CRU_MODE_CON		0x50
-#define RK3288_CRU_SEL0_CON		0x60
-#define RK3288_CRU_SEL1_CON		0x64
-#define RK3288_CRU_SEL10_CON		0x88
-#define RK3288_CRU_SEL33_CON		0xe4
-#define RK3288_CRU_SEL37_CON		0xf4
-
 /* PMU_WAKEUP_CFG1 bits */
 #define PMU_ARMINT_WAKEUP_EN		BIT(0)
 
-- 
1.9.1

--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
Please read the FAQ at  http://www.tux.org/lkml/

[PATCH v2 2/2] ARM: rockchip: remove some useless macro in pm.h

[Chris Zhong]

These are actually not used in the pm code, as we moved suspend handling
to the clock driver, remove them here.

Signed-off-by: Chris Zhong <zyw@rock-chips.com>

---

 arch/arm/mach-rockchip/pm.h | 7 -------
 1 file changed, 7 deletions(-)

diff --git a/arch/arm/mach-rockchip/pm.h b/arch/arm/mach-rockchip/pm.h
index 3e8d39c..b6494c2 100644
--- a/arch/arm/mach-rockchip/pm.h
+++ b/arch/arm/mach-rockchip/pm.h
@@ -59,13 +59,6 @@ static inline void rockchip_suspend_init(void)
 #define SGRF_DAPDEVICEEN		BIT(0)
 #define SGRF_DAPDEVICEEN_WRITE		BIT(16)
 
-#define RK3288_CRU_MODE_CON		0x50
-#define RK3288_CRU_SEL0_CON		0x60
-#define RK3288_CRU_SEL1_CON		0x64
-#define RK3288_CRU_SEL10_CON		0x88
-#define RK3288_CRU_SEL33_CON		0xe4
-#define RK3288_CRU_SEL37_CON		0xf4
-
 /* PMU_WAKEUP_CFG1 bits */
 #define PMU_ARMINT_WAKEUP_EN		BIT(0)
 
-- 
1.9.1

[PATCH v2 2/2] ARM: rockchip: remove some useless macro in pm.h

[Chris Zhong]

These are actually not used in the pm code, as we moved suspend handling
to the clock driver, remove them here.

Signed-off-by: Chris Zhong <zyw@rock-chips.com>

---

 arch/arm/mach-rockchip/pm.h | 7 -------
 1 file changed, 7 deletions(-)

diff --git a/arch/arm/mach-rockchip/pm.h b/arch/arm/mach-rockchip/pm.h
index 3e8d39c..b6494c2 100644
--- a/arch/arm/mach-rockchip/pm.h
+++ b/arch/arm/mach-rockchip/pm.h
@@ -59,13 +59,6 @@ static inline void rockchip_suspend_init(void)
 #define SGRF_DAPDEVICEEN		BIT(0)
 #define SGRF_DAPDEVICEEN_WRITE		BIT(16)
 
-#define RK3288_CRU_MODE_CON		0x50
-#define RK3288_CRU_SEL0_CON		0x60
-#define RK3288_CRU_SEL1_CON		0x64
-#define RK3288_CRU_SEL10_CON		0x88
-#define RK3288_CRU_SEL33_CON		0xe4
-#define RK3288_CRU_SEL37_CON		0xf4
-
 /* PMU_WAKEUP_CFG1 bits */
 #define PMU_ARMINT_WAKEUP_EN		BIT(0)
 
-- 
1.9.1

Re: [PATCH v2 2/2] ARM: rockchip: remove some useless macro in pm.h

[Doug Anderson]

Chris,

On Sun, Jun 21, 2015 at 3:00 AM, Chris Zhong <zyw@rock-chips.com> wrote:
> These are actually not used in the pm code, as we moved suspend handling
> to the clock driver, remove them here.
>
> Signed-off-by: Chris Zhong <zyw@rock-chips.com>
>
> ---
>
>  arch/arm/mach-rockchip/pm.h | 7 -------
>  1 file changed, 7 deletions(-)

Reviewed-by: Douglas Anderson <dianders@chromium.org>
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
Please read the FAQ at  http://www.tux.org/lkml/

Re: [PATCH v2 2/2] ARM: rockchip: remove some useless macro in pm.h

[Doug Anderson]

Chris,

On Sun, Jun 21, 2015 at 3:00 AM, Chris Zhong <zyw@rock-chips.com> wrote:
> These are actually not used in the pm code, as we moved suspend handling
> to the clock driver, remove them here.
>
> Signed-off-by: Chris Zhong <zyw@rock-chips.com>
>
> ---
>
>  arch/arm/mach-rockchip/pm.h | 7 -------
>  1 file changed, 7 deletions(-)

Reviewed-by: Douglas Anderson <dianders@chromium.org>

[PATCH v2 2/2] ARM: rockchip: remove some useless macro in pm.h

[Doug Anderson]

Chris,

On Sun, Jun 21, 2015 at 3:00 AM, Chris Zhong <zyw@rock-chips.com> wrote:
> These are actually not used in the pm code, as we moved suspend handling
> to the clock driver, remove them here.
>
> Signed-off-by: Chris Zhong <zyw@rock-chips.com>
>
> ---
>
>  arch/arm/mach-rockchip/pm.h | 7 -------
>  1 file changed, 7 deletions(-)

Reviewed-by: Douglas Anderson <dianders@chromium.org>

Re: [PATCH v2 1/2] ARM: rockchip: add support holding 24Mhz osc during suspend

[Doug Anderson]

Chris,

On Sun, Jun 21, 2015 at 3:00 AM, Chris Zhong <zyw@rock-chips.com> wrote:
> If we want to wake up system via usb, the 24Mhz osc could not be
> disabled during suspend, read the usb phy SIDDQ bit to decide whether
> to switch to 32khz clock-in.
>
> Signed-off-by: Chris Zhong <zyw@rock-chips.com>
> ---
>
>  arch/arm/mach-rockchip/pm.c | 38 ++++++++++++++++++++++++++++++++++++--
>  1 file changed, 36 insertions(+), 2 deletions(-)

As talked about privately, this patch is ugly.  ...but I don't see a
solution that is less ugly.  This method has the advantages that it is
"automatic"--the system figures out whether it needs the 24MHz clock
automatically based on whether the PHYs were left on.  No other
communication / device tree stuff is needed.


> +       /* if any usb phy is still on(GRF_SIDDQ==0), that means we need the
> +        * function of usb wakeup, so do not switch to 32khz, since the usb phy
> +        * clk does not connect to 32khz osc*/

This is not _quite_ the block commenting style used in this file.  It
would be a bit nicer if it matched.

Other than that:

Reviewed-by: Douglas Anderson <dianders@chromium.org>

On the chromeos-3.14 kernel (with the WIP dwc2 patches):
Tested-by: Douglas Anderson <dianders@chromium.org>
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
Please read the FAQ at  http://www.tux.org/lkml/

Re: [PATCH v2 1/2] ARM: rockchip: add support holding 24Mhz osc during suspend

[Doug Anderson]

Chris,

On Sun, Jun 21, 2015 at 3:00 AM, Chris Zhong <zyw@rock-chips.com> wrote:
> If we want to wake up system via usb, the 24Mhz osc could not be
> disabled during suspend, read the usb phy SIDDQ bit to decide whether
> to switch to 32khz clock-in.
>
> Signed-off-by: Chris Zhong <zyw@rock-chips.com>
> ---
>
>  arch/arm/mach-rockchip/pm.c | 38 ++++++++++++++++++++++++++++++++++++--
>  1 file changed, 36 insertions(+), 2 deletions(-)

As talked about privately, this patch is ugly.  ...but I don't see a
solution that is less ugly.  This method has the advantages that it is
"automatic"--the system figures out whether it needs the 24MHz clock
automatically based on whether the PHYs were left on.  No other
communication / device tree stuff is needed.


> +       /* if any usb phy is still on(GRF_SIDDQ==0), that means we need the
> +        * function of usb wakeup, so do not switch to 32khz, since the usb phy
> +        * clk does not connect to 32khz osc*/

This is not _quite_ the block commenting style used in this file.  It
would be a bit nicer if it matched.

Other than that:

Reviewed-by: Douglas Anderson <dianders@chromium.org>

On the chromeos-3.14 kernel (with the WIP dwc2 patches):
Tested-by: Douglas Anderson <dianders@chromium.org>

[PATCH v2 1/2] ARM: rockchip: add support holding 24Mhz osc during suspend

[Doug Anderson]

Chris,

On Sun, Jun 21, 2015 at 3:00 AM, Chris Zhong <zyw@rock-chips.com> wrote:
> If we want to wake up system via usb, the 24Mhz osc could not be
> disabled during suspend, read the usb phy SIDDQ bit to decide whether
> to switch to 32khz clock-in.
>
> Signed-off-by: Chris Zhong <zyw@rock-chips.com>
> ---
>
>  arch/arm/mach-rockchip/pm.c | 38 ++++++++++++++++++++++++++++++++++++--
>  1 file changed, 36 insertions(+), 2 deletions(-)

As talked about privately, this patch is ugly.  ...but I don't see a
solution that is less ugly.  This method has the advantages that it is
"automatic"--the system figures out whether it needs the 24MHz clock
automatically based on whether the PHYs were left on.  No other
communication / device tree stuff is needed.


> +       /* if any usb phy is still on(GRF_SIDDQ==0), that means we need the
> +        * function of usb wakeup, so do not switch to 32khz, since the usb phy
> +        * clk does not connect to 32khz osc*/

This is not _quite_ the block commenting style used in this file.  It
would be a bit nicer if it matched.

Other than that:

Reviewed-by: Douglas Anderson <dianders@chromium.org>

On the chromeos-3.14 kernel (with the WIP dwc2 patches):
Tested-by: Douglas Anderson <dianders@chromium.org>

Re: [PATCH v2 1/2] ARM: rockchip: add support holding 24Mhz osc during suspend

[Heiko Stübner]

Am Montag, 22. Juni 2015, 13:04:02 schrieb Doug Anderson:
> Chris,
> 
> On Sun, Jun 21, 2015 at 3:00 AM, Chris Zhong <zyw@rock-chips.com> wrote:
> > If we want to wake up system via usb, the 24Mhz osc could not be
> > disabled during suspend, read the usb phy SIDDQ bit to decide whether
> > to switch to 32khz clock-in.
> > 
> > Signed-off-by: Chris Zhong <zyw@rock-chips.com>
> > ---
> > 
> >  arch/arm/mach-rockchip/pm.c | 38 ++++++++++++++++++++++++++++++++++++--
> >  1 file changed, 36 insertions(+), 2 deletions(-)
> 
> As talked about privately, this patch is ugly.  ...but I don't see a
> solution that is less ugly.  This method has the advantages that it is
> "automatic"--the system figures out whether it needs the 24MHz clock
> automatically based on whether the PHYs were left on.  No other
> communication / device tree stuff is needed.

Yep, fully agreed :-) .

As I said before, another possible solution would probably involve checking 
the phy clocks themselfs, but getting these out of the devicetree would make 
this even more ugly.


So I'll let this sit for some more days in case somebody wants to complain 
(middle of the merge window anyway) and apply these two after.


> 
> > +       /* if any usb phy is still on(GRF_SIDDQ==0), that means we need
> > the
> > +        * function of usb wakeup, so do not switch to 32khz, since the
> > usb phy +        * clk does not connect to 32khz osc*/
> 
> This is not _quite_ the block commenting style used in this file.  It
> would be a bit nicer if it matched.
> 
> Other than that:
> 
> Reviewed-by: Douglas Anderson <dianders@chromium.org>
> 
> On the chromeos-3.14 kernel (with the WIP dwc2 patches):
> Tested-by: Douglas Anderson <dianders@chromium.org>

--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
Please read the FAQ at  http://www.tux.org/lkml/

Re: [PATCH v2 1/2] ARM: rockchip: add support holding 24Mhz osc during suspend

[Heiko Stübner]

Am Montag, 22. Juni 2015, 13:04:02 schrieb Doug Anderson:
> Chris,
> 
> On Sun, Jun 21, 2015 at 3:00 AM, Chris Zhong <zyw@rock-chips.com> wrote:
> > If we want to wake up system via usb, the 24Mhz osc could not be
> > disabled during suspend, read the usb phy SIDDQ bit to decide whether
> > to switch to 32khz clock-in.
> > 
> > Signed-off-by: Chris Zhong <zyw@rock-chips.com>
> > ---
> > 
> >  arch/arm/mach-rockchip/pm.c | 38 ++++++++++++++++++++++++++++++++++++--
> >  1 file changed, 36 insertions(+), 2 deletions(-)
> 
> As talked about privately, this patch is ugly.  ...but I don't see a
> solution that is less ugly.  This method has the advantages that it is
> "automatic"--the system figures out whether it needs the 24MHz clock
> automatically based on whether the PHYs were left on.  No other
> communication / device tree stuff is needed.

Yep, fully agreed :-) .

As I said before, another possible solution would probably involve checking 
the phy clocks themselfs, but getting these out of the devicetree would make 
this even more ugly.


So I'll let this sit for some more days in case somebody wants to complain 
(middle of the merge window anyway) and apply these two after.


> 
> > +       /* if any usb phy is still on(GRF_SIDDQ==0), that means we need
> > the
> > +        * function of usb wakeup, so do not switch to 32khz, since the
> > usb phy +        * clk does not connect to 32khz osc*/
> 
> This is not _quite_ the block commenting style used in this file.  It
> would be a bit nicer if it matched.
> 
> Other than that:
> 
> Reviewed-by: Douglas Anderson <dianders@chromium.org>
> 
> On the chromeos-3.14 kernel (with the WIP dwc2 patches):
> Tested-by: Douglas Anderson <dianders@chromium.org>

[PATCH v2 1/2] ARM: rockchip: add support holding 24Mhz osc during suspend

[Heiko Stübner]

Am Montag, 22. Juni 2015, 13:04:02 schrieb Doug Anderson:
> Chris,
> 
> On Sun, Jun 21, 2015 at 3:00 AM, Chris Zhong <zyw@rock-chips.com> wrote:
> > If we want to wake up system via usb, the 24Mhz osc could not be
> > disabled during suspend, read the usb phy SIDDQ bit to decide whether
> > to switch to 32khz clock-in.
> > 
> > Signed-off-by: Chris Zhong <zyw@rock-chips.com>
> > ---
> > 
> >  arch/arm/mach-rockchip/pm.c | 38 ++++++++++++++++++++++++++++++++++++--
> >  1 file changed, 36 insertions(+), 2 deletions(-)
> 
> As talked about privately, this patch is ugly.  ...but I don't see a
> solution that is less ugly.  This method has the advantages that it is
> "automatic"--the system figures out whether it needs the 24MHz clock
> automatically based on whether the PHYs were left on.  No other
> communication / device tree stuff is needed.

Yep, fully agreed :-) .

As I said before, another possible solution would probably involve checking 
the phy clocks themselfs, but getting these out of the devicetree would make 
this even more ugly.


So I'll let this sit for some more days in case somebody wants to complain 
(middle of the merge window anyway) and apply these two after.


> 
> > +       /* if any usb phy is still on(GRF_SIDDQ==0), that means we need
> > the
> > +        * function of usb wakeup, so do not switch to 32khz, since the
> > usb phy +        * clk does not connect to 32khz osc*/
> 
> This is not _quite_ the block commenting style used in this file.  It
> would be a bit nicer if it matched.
> 
> Other than that:
> 
> Reviewed-by: Douglas Anderson <dianders@chromium.org>
> 
> On the chromeos-3.14 kernel (with the WIP dwc2 patches):
> Tested-by: Douglas Anderson <dianders@chromium.org>