Skip to contentSkip to conversation

Veyron Speedy / ASUS C201 fixes

PATCH New Urja Rannikko 6 participants 31 messages Mar 22, 2019 – Jul 30, 2019

Subsystems (5)
lore
Share & tools
lore
#0 lore raw

cover Veyron Speedy / ASUS C201 fixes

Hi,

Just saying: I'm not familiar with this stuff so i might've misunderstood something in my commit messages, but anyways i managed to make it boot.

There's more stuff than just these, but i decided I'd keep it short... If you want to know my full setup, you can check out my gh repo for now:
https://github.com/urjaman/u-boot/commits/c201-stuff

Notably I did include http://patchwork.ozlabs.org/patch/1040541/ in my tree when testing (rk3288 pinctrl patch that was linked to on the previous thread i found about the veyron speedy being broken).

The one thing that I havent yet fixed is that the USB host port on the otg-capable controller (near screen) doesn't work - one guess i have is that the dwc2 host driver doesn't force it into host mode... (It doesn't even read dr_mode ...)

Urja Rannikko (4):
pinctrl: exit pinconfig_post_bind if there are no subnodes
rk3288-board: remove pinctrl call for debug uart
rk3288-board: cosmetic: document selecting RK PWM
dw_mmc: turn on the IO supply

 arch/arm/mach-rockchip/rk3288-board.c | 13 +------------
 drivers/mmc/dw_mmc.c                  | 15 +++++++++++++++
 drivers/pinctrl/pinctrl-uclass.c      |  3 +++
 3 files changed, 19 insertions(+), 12 deletions(-)
Signature · 1 line
2.21.0
#1 lore raw

PATCH 1/4 pinctrl: exit pinconfig_post_bind if there are no subnodes

This fixes RK3288 SPL hanging or hitting this assert:
drivers/core/ofnode.c:183: ofnode_first_subnode: Assertion `ofnode_valid(node)' failed.

---
 drivers/pinctrl/pinctrl-uclass.c | 3 +++
 1 file changed, 3 insertions(+)
drivers/pinctrl/pinctrl-uclass.c +3 -0
index 0e3260afd1..bda951b07f 100644
@@ -138,6 +138,9 @@ static int pinconfig_post_bind(struct udevice *dev)
 	ofnode node;
 	int ret;
 
+	if (!dev_of_valid(dev))
+		return 0;
+
 	dev_for_each_subnode(node, dev) {
 		if (pre_reloc_only &&
 		    !ofnode_pre_reloc(node))
Signature · 1 line
2.21.0
#2 lore raw

PATCH 1/4 pinctrl: exit pinconfig_post_bind if there are no subnodes

Quoted 8 lines · Urja Rannikko

This fixes RK3288 SPL hanging or hitting this assert:
drivers/core/ofnode.c:183: ofnode_first_subnode: Assertion `ofnode_valid(node)' failed.

---
 drivers/pinctrl/pinctrl-uclass.c | 3 +++
 1 file changed, 3 insertions(+)
#3 lore raw

PATCH 1/4 pinctrl: exit pinconfig_post_bind if there are no subnodes

In reply to Urja Rannikko
Quoted 9 lines

This fixes RK3288 SPL hanging or hitting this assert:
drivers/core/ofnode.c:183: ofnode_first_subnode: Assertion `ofnode_valid(node)' failed.

---
 drivers/pinctrl/pinctrl-uclass.c | 3 +++
 1 file changed, 3 insertions(+)
#4 lore raw

PATCH 1/4 pinctrl: exit pinconfig_post_bind if there are no subnodes

In reply to Urja Rannikko
Quoted 10 lines

This fixes RK3288 SPL hanging or hitting this assert:
drivers/core/ofnode.c:183: ofnode_first_subnode: Assertion `ofnode_valid(node)' failed.

---
 drivers/pinctrl/pinctrl-uclass.c | 3 +++
 1 file changed, 3 insertions(+)

Applied to u-boot-rockchip, thanks!

#5 lore raw

PATCH 2/4 rk3288-board: remove pinctrl call for debug uart

In reply to Urja Rannikko

This failed and caused a boot failure on c201, and afaik the pins should be setup by the new pinctrl driver.

---
 arch/arm/mach-rockchip/rk3288-board.c | 12 ------------
 1 file changed, 12 deletions(-)
arch/arm/mach-rockchip/rk3288-board.c +0 -12
index 9c4f7f219f..f5df44dab1 100644
@@ -321,7 +321,6 @@ int board_early_init_f(void)
 {
 	const uintptr_t GRF_SOC_CON0 = 0xff770244;
 	const uintptr_t GRF_SOC_CON2 = 0xff77024c;
-	struct udevice *pinctrl;
 	struct udevice *dev;
 	int ret;
 
@@ -335,18 +334,7 @@ int board_early_init_f(void)
 		debug("CLK init failed: %d\n", ret);
 		return ret;
 	}
-	ret = uclass_get_device(UCLASS_PINCTRL, 0, &pinctrl);
-	if (ret) {
-		debug("%s: Cannot find pinctrl device\n", __func__);
-		return ret;
-	}
 
-	/* Enable debug UART */
-	ret = pinctrl_request_noflags(pinctrl, PERIPH_ID_UART_DBG);
-	if (ret) {
-		debug("%s: Failed to set up console UART\n", __func__);
-		return ret;
-	}
 	rk_setreg(GRF_SOC_CON2, 1 << 0);
 
 	/*
Signature · 1 line
2.21.0
#6 lore raw

PATCH 2/4 rk3288-board: remove pinctrl call for debug uart

Hi Urja,

Quoted 3 lines · Urja Rannikko

This failed and caused a boot failure on c201, and afaik the pins should be setup by the new pinctrl driver.

It should be set up in SPL, if enabled.

I wonder when this code is actually used?

Quoted 5 lines
---
 arch/arm/mach-rockchip/rk3288-board.c | 12 ------------
 1 file changed, 12 deletions(-)
#7 lore raw

PATCH 2/4 rk3288-board: remove pinctrl call for debug uart

Hi,

Quoted 11 lines · Simon Glass

Hi Urja,

Quoted 3 lines · Urja Rannikko

This failed and caused a boot failure on c201, and afaik the pins should be setup by the new pinctrl driver.

It should be set up in SPL, if enabled.

I wonder when this code is actually used?

The original code mentioned chainloading so i thought eg. from coreboot, but i mean i thought you should know better than me, see commit fe974716326ce2eee4bcac710ce7ec007919845e.
It seemed to also have something to do with the clock driver (which is a thing that has changed in the last 2 years too...).

Quoted 7 lines
Quoted 5 lines
---
 arch/arm/mach-rockchip/rk3288-board.c | 12 ------------
 1 file changed, 12 deletions(-)
Signature · 1 line
Urja Rannikko
#9 lore raw

PATCH 2/4 rk3288-board: remove pinctrl call for debug uart

In reply to Urja Rannikko
Quoted 10 lines

This failed and caused a boot failure on c201, and afaik the pins should be setup by the new pinctrl driver.

---
 arch/arm/mach-rockchip/rk3288-board.c | 12 ------------
 1 file changed, 12 deletions(-)

Applied to u-boot-rockchip, thanks!

#10 lore raw

PATCH 3/4 rk3288-board: cosmetic: document selecting RK PWM

In reply to Urja Rannikko

Atleast hat is what it says in the TRM.

---
 arch/arm/mach-rockchip/rk3288-board.c | 1 +
 1 file changed, 1 insertion(+)
arch/arm/mach-rockchip/rk3288-board.c +1 -0
index f5df44dab1..9c05149610 100644
@@ -335,6 +335,7 @@ int board_early_init_f(void)
 		return ret;
 	}
 
+	/* Select RK PWM (as opposed to old PWM solution) */
 	rk_setreg(GRF_SOC_CON2, 1 << 0);
 
 	/*
Signature · 1 line
2.21.0
#12 lore raw

PATCH 3/4 rk3288-board: cosmetic: document selecting RK PWM

Hi,

Quoted 6 lines · Simon Glass
Quoted 2 lines · Urja Rannikko

Atleast hat is what it says in the TRM.

At least that

Whoops.
But actually i found out (by chance from an another patch on the list just now) that instead of documenting this bit set, we could instead remove it since it is already set in arch/arm/mach-rockchip/rk3288/rk3288.c.
I just added the comment because i thought from code proximity that this code had something to do with the debug uart thing that i was cleaning out and upon checking it had nothing to do with the uart..

Quoted 8 lines
Quoted 6 lines
---
 arch/arm/mach-rockchip/rk3288-board.c | 1 +
 1 file changed, 1 insertion(+)
Signature · 1 line
Urja Rannikko
#13 lore raw

PATCH 4/4 dw_mmc: turn on the IO supply

In reply to Urja Rannikko

Fixes the microSD slot on the ASUS C201.

---
 drivers/mmc/dw_mmc.c | 15 +++++++++++++++
 1 file changed, 15 insertions(+)
drivers/mmc/dw_mmc.c +15 -0
index 93a836eac3..e1960b213a 100644
@@ -13,6 +13,7 @@
 #include <mmc.h>
 #include <dwmmc.h>
 #include <wait_bit.h>
+#include <power/regulator.h>
 
 #define PAGE_SIZE 4096
 
@@ -440,6 +441,7 @@ static int dwmci_set_ios(struct mmc *mmc)
 #endif
 	struct dwmci_host *host = (struct dwmci_host *)mmc->priv;
 	u32 ctype, regs;
+	int ret;
 
 	debug("Buswidth = %d, clock: %d\n", mmc->bus_width, mmc->clock);
 
@@ -469,6 +471,19 @@ static int dwmci_set_ios(struct mmc *mmc)
 	if (host->clksel)
 		host->clksel(host);
 
+#ifdef CONFIG_DM_REGULATOR
+	if (mmc->vqmmc_supply) {
+		if (mmc->signal_voltage == MMC_SIGNAL_VOLTAGE_180)
+			regulator_set_value(mmc->vqmmc_supply, 1800000);
+		else
+			regulator_set_value(mmc->vqmmc_supply, 3300000);
+
+		ret = regulator_set_enable_if_allowed(mmc->vqmmc_supply, true);
+		if (ret)
+			return ret;
+	}
+#endif
+
 	return 0;
 }
 
Signature · 1 line
2.21.0
maintainer
#14 lore raw

PATCH 4/4 dw_mmc: turn on the IO supply

Am Freitag, 22. März 2019, 20:14:36 CET schrieb Urja Rannikko:

Quoted 28 lines

Fixes the microSD slot on the ASUS C201.

---
 drivers/mmc/dw_mmc.c | 15 +++++++++++++++
 1 file changed, 15 insertions(+)
drivers/mmc/dw_mmc.c +2 -0
index 93a836eac3..e1960b213a 100644
@@ -13,6 +13,7 @@
 #include <mmc.h>
 #include <dwmmc.h>
 #include <wait_bit.h>
+#include <power/regulator.h>
 
 #define PAGE_SIZE 4096
 
@@ -440,6 +441,7 @@ static int dwmci_set_ios(struct mmc *mmc)
 #endif
 	struct dwmci_host *host = (struct dwmci_host *)mmc->priv;
 	u32 ctype, regs;
+	int ret;
 
 	debug("Buswidth = %d, clock: %d\n", mmc->bus_width, mmc->clock);
 

drivers/mmc/dw_mmc.c static int dwmci_set_ios(struct mmc *mmc) · line 474

Quoted 4 lines
drivers/mmc/dw_mmc.c +1 -0
@@ -469,6 +471,19 @@ static int dwmci_set_ios(struct mmc *mmc)
 	if (host->clksel)
 		host->clksel(host);
 
+#ifdef CONFIG_DM_REGULATOR

this should be

	#if CONFIG_IS_ENABLED(DM_REGULATOR)

because otherwise an SPL-build using the mmc driver may fail with

drivers/mmc/dw_mmc.c: In function ‘dwmci_set_ios’:
drivers/mmc/dw_mmc.c:475:9: error: ‘struct mmc’ has no member named ‘vqmmc_supply’

  if (mmc->vqmmc_supply) {
         ^~
drivers/mmc/dw_mmc.c:477:27: error: ‘struct mmc’ has no member named ‘vqmmc_supply’
    regulator_set_value(mmc->vqmmc_supply, 1800000);
                           ^~
drivers/mmc/dw_mmc.c:479:27: error: ‘struct mmc’ has no member named ‘vqmmc_supply’
    regulator_set_value(mmc->vqmmc_supply, 3300000);
                           ^~

drivers/mmc/dw_mmc.c:481:44: error: ‘struct mmc’ has no member named ‘vqmmc_supply’
ret = regulator_set_enable_if_allowed(mmc->vqmmc_supply, true);

Quoted 16 lines
drivers/mmc/dw_mmc.c +12 -0
+	if (mmc->vqmmc_supply) {
+		if (mmc->signal_voltage == MMC_SIGNAL_VOLTAGE_180)
+			regulator_set_value(mmc->vqmmc_supply, 1800000);
+		else
+			regulator_set_value(mmc->vqmmc_supply, 3300000);
+
+		ret = regulator_set_enable_if_allowed(mmc->vqmmc_supply, true);
+		if (ret)
+			return ret;
+	}
+#endif
+
 	return 0;
 }
 
#15 lore raw

PATCH 4/4 dw_mmc: turn on the IO supply

Hi,

Quoted 4 lines · Heiko Stübner
Quoted 1 line

+#ifdef CONFIG_DM_REGULATOR

this should be

        #if CONFIG_IS_ENABLED(DM_REGULATOR)

Ok, just FYI that fragment was based on the one in tmio-common.c, so maybe that should also be fixed(?).

Quoted 32 lines

because otherwise an SPL-build using the mmc driver may fail with

drivers/mmc/dw_mmc.c: In function ‘dwmci_set_ios’:
drivers/mmc/dw_mmc.c:475:9: error: ‘struct mmc’ has no member named ‘vqmmc_supply’

  if (mmc->vqmmc_supply) {
         ^~
drivers/mmc/dw_mmc.c:477:27: error: ‘struct mmc’ has no member named ‘vqmmc_supply’
    regulator_set_value(mmc->vqmmc_supply, 1800000);
                           ^~
drivers/mmc/dw_mmc.c:479:27: error: ‘struct mmc’ has no member named ‘vqmmc_supply’
    regulator_set_value(mmc->vqmmc_supply, 3300000);
                           ^~

drivers/mmc/dw_mmc.c:481:44: error: ‘struct mmc’ has no member named ‘vqmmc_supply’
ret = regulator_set_enable_if_allowed(mmc->vqmmc_supply, true);

Quoted 15 lines
drivers/mmc/dw_mmc.c +12 -0
+     if (mmc->vqmmc_supply) {
+             if (mmc->signal_voltage == MMC_SIGNAL_VOLTAGE_180)
+                     regulator_set_value(mmc->vqmmc_supply, 1800000);
+             else
+                     regulator_set_value(mmc->vqmmc_supply, 3300000);
+
+             ret = regulator_set_enable_if_allowed(mmc->vqmmc_supply, true);
+             if (ret)
+                     return ret;
+     }
+#endif
+
      return 0;
 }
Signature · 1 line
Urja Rannikko
maintainer
#16 lore raw

PATCH 4/4 dw_mmc: turn on the IO supply

Am Sonntag, 31. März 2019, 21:45:15 CEST schrieb Urja Rannikko:

Quoted 10 lines

Hi,

Quoted 4 lines · Heiko Stübner
Quoted 1 line

+#ifdef CONFIG_DM_REGULATOR

this should be

        #if CONFIG_IS_ENABLED(DM_REGULATOR)

Ok, just FYI that fragment was based on the one in tmio-common.c, so maybe that should also be fixed(?).

yep, looks like it. While it doesn't cause compile errors right now, it might after all create runtime errors otherwise.

Heiko

Quoted 34 lines
Quoted 32 lines

because otherwise an SPL-build using the mmc driver may fail with

drivers/mmc/dw_mmc.c: In function ‘dwmci_set_ios’:
drivers/mmc/dw_mmc.c:475:9: error: ‘struct mmc’ has no member named ‘vqmmc_supply’

  if (mmc->vqmmc_supply) {
         ^~
drivers/mmc/dw_mmc.c:477:27: error: ‘struct mmc’ has no member named ‘vqmmc_supply’
    regulator_set_value(mmc->vqmmc_supply, 1800000);
                           ^~
drivers/mmc/dw_mmc.c:479:27: error: ‘struct mmc’ has no member named ‘vqmmc_supply’
    regulator_set_value(mmc->vqmmc_supply, 3300000);
                           ^~

drivers/mmc/dw_mmc.c:481:44: error: ‘struct mmc’ has no member named ‘vqmmc_supply’
ret = regulator_set_enable_if_allowed(mmc->vqmmc_supply, true);

Quoted 15 lines
drivers/mmc/dw_mmc.c +12 -0
+     if (mmc->vqmmc_supply) {
+             if (mmc->signal_voltage == MMC_SIGNAL_VOLTAGE_180)
+                     regulator_set_value(mmc->vqmmc_supply, 1800000);
+             else
+                     regulator_set_value(mmc->vqmmc_supply, 3300000);
+
+             ret = regulator_set_enable_if_allowed(mmc->vqmmc_supply, true);
+             if (ret)
+                     return ret;
+     }
+#endif
+
      return 0;
 }
#17 lore raw

PATCH v2 1/1 dw_mmc: turn on the IO supply

In reply to Urja Rannikko

Fixes the microSD slot on the ASUS C201.

---
v2: use #if CONFIG_IS_ENABLED(DM_REGULATOR)
-

 drivers/mmc/dw_mmc.c | 15 +++++++++++++++
 1 file changed, 15 insertions(+)
drivers/mmc/dw_mmc.c +15 -0
index 93a836eac3..b04c1f9f41 100644
@@ -13,6 +13,7 @@
 #include <mmc.h>
 #include <dwmmc.h>
 #include <wait_bit.h>
+#include <power/regulator.h>
 
 #define PAGE_SIZE 4096
 
@@ -440,6 +441,7 @@ static int dwmci_set_ios(struct mmc *mmc)
 #endif
 	struct dwmci_host *host = (struct dwmci_host *)mmc->priv;
 	u32 ctype, regs;
+	int ret;
 
 	debug("Buswidth = %d, clock: %d\n", mmc->bus_width, mmc->clock);
 
@@ -469,6 +471,19 @@ static int dwmci_set_ios(struct mmc *mmc)
 	if (host->clksel)
 		host->clksel(host);
 
+#if CONFIG_IS_ENABLED(DM_REGULATOR)
+	if (mmc->vqmmc_supply) {
+		if (mmc->signal_voltage == MMC_SIGNAL_VOLTAGE_180)
+			regulator_set_value(mmc->vqmmc_supply, 1800000);
+		else
+			regulator_set_value(mmc->vqmmc_supply, 3300000);
+
+		ret = regulator_set_enable_if_allowed(mmc->vqmmc_supply, true);
+		if (ret)
+			return ret;
+	}
+#endif
+
 	return 0;
 }
 
Signature · 1 line
2.21.0
#18 lore raw

PATCH v2 1/1 dw_mmc: turn on the IO supply

Hi,

Quoted 6 lines · Urja Rannikko

Fixes the microSD slot on the ASUS C201.

---
v2: use #if CONFIG_IS_ENABLED(DM_REGULATOR)

Ping? Anything I should do wrt this?

Signature · 1 line
Urja Rannikko
#19 lore raw

PATCH v2 1/1 dw_mmc: turn on the IO supply

Hiya,

Quoted 5 lines · Urja Rannikko

Hi,

Quoted 1 line · Urja Rannikko
Quoted 7 lines
Quoted 5 lines

Fixes the microSD slot on the ASUS C201.

---
v2: use #if CONFIG_IS_ENABLED(DM_REGULATOR)

Ping? Anything I should do wrt this?

Ping number two. I think i'll resend this the next time so that the actual patch isnt lost in the depths of mailboxes...

Signature · 1 line
Urja Rannikko
maintainer
#20 lore raw

PATCH v2 1/1 dw_mmc: turn on the IO supply

In reply to Urja Rannikko

Am Montag, 1. April 2019, 19:44:22 CEST schrieb Urja Rannikko:

Quoted 22 lines

Fixes the microSD slot on the ASUS C201.

---
v2: use #if CONFIG_IS_ENABLED(DM_REGULATOR)

(quoted diff) +0 -1
-
 drivers/mmc/dw_mmc.c | 15 +++++++++++++++
 1 file changed, 15 insertions(+)
drivers/mmc/dw_mmc.c +1 -0
index 93a836eac3..b04c1f9f41 100644
@@ -13,6 +13,7 @@
 #include <mmc.h>
 #include <dwmmc.h>
 #include <wait_bit.h>
+#include <power/regulator.h>
 
 #define PAGE_SIZE 4096
 

drivers/mmc/dw_mmc.c static int dwmci_set_ios(struct mmc *mmc) · line 444

Quoted 4 lines
drivers/mmc/dw_mmc.c +1 -0
@@ -440,6 +441,7 @@ static int dwmci_set_ios(struct mmc *mmc)
 #endif
 	struct dwmci_host *host = (struct dwmci_host *)mmc->priv;
 	u32 ctype, regs;
+	int ret;

This produces a new warning about an unused variable in the !DM_REGULATOR case. I guess either move the "ret" below or add a " __maybe_unused" if that is possible for variables?

Quoted 24 lines
drivers/mmc/dw_mmc.c +13 -0
 	debug("Buswidth = %d, clock: %d\n", mmc->bus_width, mmc->clock);
 
@@ -469,6 +471,19 @@ static int dwmci_set_ios(struct mmc *mmc)
 	if (host->clksel)
 		host->clksel(host);
 
+#if CONFIG_IS_ENABLED(DM_REGULATOR)
+	if (mmc->vqmmc_supply) {
+		if (mmc->signal_voltage == MMC_SIGNAL_VOLTAGE_180)
+			regulator_set_value(mmc->vqmmc_supply, 1800000);
+		else
+			regulator_set_value(mmc->vqmmc_supply, 3300000);
+
+		ret = regulator_set_enable_if_allowed(mmc->vqmmc_supply, true);
+		if (ret)
+			return ret;
+	}
+#endif
+
 	return 0;
 }
 
#21 lore raw

PATCH v3 1/1 dw_mmc: turn on the IO supply

Fixes the microSD slot on the ASUS C201.

---
v2: use #if CONFIG_IS_ENABLED(DM_REGULATOR)
v3: Move ret variable inside the if block (only used there)

---
 drivers/mmc/dw_mmc.c | 16 ++++++++++++++++
 1 file changed, 16 insertions(+)
drivers/mmc/dw_mmc.c +16 -0
index 1992d61182..22f6c7eefd 100644
@@ -13,6 +13,7 @@
 #include <mmc.h>
 #include <dwmmc.h>
 #include <wait_bit.h>
+#include <power/regulator.h>
 
 #define PAGE_SIZE 4096
 
@@ -493,6 +494,21 @@ static int dwmci_set_ios(struct mmc *mmc)
 	if (host->clksel)
 		host->clksel(host);
 
+#if CONFIG_IS_ENABLED(DM_REGULATOR)
+	if (mmc->vqmmc_supply) {
+		int ret;
+
+		if (mmc->signal_voltage == MMC_SIGNAL_VOLTAGE_180)
+			regulator_set_value(mmc->vqmmc_supply, 1800000);
+		else
+			regulator_set_value(mmc->vqmmc_supply, 3300000);
+
+		ret = regulator_set_enable_if_allowed(mmc->vqmmc_supply, true);
+		if (ret)
+			return ret;
+	}
+#endif
+
 	return 0;
 }
 
Signature · 1 line
2.21.0
#22 lore raw

PATCH v3 1/1 dw_mmc: turn on the IO supply

Quoted 10 lines · Urja Rannikko

Fixes the microSD slot on the ASUS C201.

---
v2: use #if CONFIG_IS_ENABLED(DM_REGULATOR)
v3: Move ret variable inside the if block (only used there)

---
 drivers/mmc/dw_mmc.c | 16 ++++++++++++++++
 1 file changed, 16 insertions(+)
Quoted 14 lines
drivers/mmc/dw_mmc.c +1 -0
index 1992d61182..22f6c7eefd 100644
@@ -13,6 +13,7 @@
 #include <mmc.h>
 #include <dwmmc.h>
 #include <wait_bit.h>
+#include <power/regulator.h>
 
 #define PAGE_SIZE 4096
 

drivers/mmc/dw_mmc.c static int dwmci_set_ios(struct mmc *mmc) · line 497

Quoted 4 lines
drivers/mmc/dw_mmc.c +1 -0
@@ -493,6 +494,21 @@ static int dwmci_set_ios(struct mmc *mmc)
        if (host->clksel)
                host->clksel(host);
 
+#if CONFIG_IS_ENABLED(DM_REGULATOR)

I wonder if we could use if() here instead?

Quoted 20 lines
drivers/mmc/dw_mmc.c +14 -1
+       if (mmc->vqmmc_supply) {
+               int ret;
+
+               if (mmc->signal_voltage == MMC_SIGNAL_VOLTAGE_180)
+                       regulator_set_value(mmc->vqmmc_supply, 1800000);
+               else
+                       regulator_set_value(mmc->vqmmc_supply, 3300000);
+
+               ret = regulator_set_enable_if_allowed(mmc->vqmmc_supply, true);
+               if (ret)
+                       return ret;
+       }
+#endif
+
        return 0;
 }
 
--

2.21.0

#23 lore raw

PATCH v3 1/1 dw_mmc: turn on the IO supply

In reply to Urja Rannikko
Quoted 33 lines

Subject: [U-Boot] [PATCH v3] dw_mmc: turn on the IO supply

Fixes the microSD slot on the ASUS C201.

---
v2: use #if CONFIG_IS_ENABLED(DM_REGULATOR)
v3: Move ret variable inside the if block (only used there)

---
 drivers/mmc/dw_mmc.c | 16 ++++++++++++++++
 1 file changed, 16 insertions(+)
drivers/mmc/dw_mmc.c index renamed from drivers/mmc/dw_mmc.c +0 -0

1992d61182..22f6c7eefd 100644

drivers/mmc/dw_mmc.c +6 -0
@@ -13,6 +13,7 @@
 #include <mmc.h>
 #include <dwmmc.h>
 #include <wait_bit.h>
+#include <power/regulator.h>
 
 #define PAGE_SIZE 4096
 
@@ -493,6 +494,21 @@ static int dwmci_set_ios(struct mmc *mmc)
 	if (host->clksel)
 		host->clksel(host);
 
+#if CONFIG_IS_ENABLED(DM_REGULATOR)
+	if (mmc->vqmmc_supply) {
+		int ret;
+
+		if (mmc->signal_voltage == MMC_SIGNAL_VOLTAGE_180)

drivers/mmc/dw_mmc.c static int dwmci_set_ios(struct mmc *mmc) · line 511

Quoted 10 lines
drivers/mmc/dw_mmc.c +10 -0
+			regulator_set_value(mmc->vqmmc_supply, 1800000);
+		else
+			regulator_set_value(mmc->vqmmc_supply, 3300000);
+
+		ret = regulator_set_enable_if_allowed(mmc->vqmmc_supply, true);
+		if (ret)
+			return ret;
+	}
+#endif
+
Quoted 14 lines
drivers/mmc/dw_mmc.c +0 -1
 	return 0;
 }
 
--

2.21.0

_______________________________________________
U-Boot mailing list
U-Boot at lists.denx.de
https://eur01.safelinks.protection.outlook.com/?url=https%3A%2F%2Flists.d
enx.de%2Flistinfo%2Fu-boot&amp;data=02%7C01%7CPeng.Fan%40nxp.com %7Ce49ea3fa171f4230a40308d6d7a680fc%7C686ea1d3bc2b4c6fa92cd99c5 c301635%7C0%7C0%7C636933507511914198&amp;sdata=Hyq5z5xI5xYUPA tcKJU%2FmdM3PLgCdHQcIr4qkSbz4Fc%3D&amp;reserved=0

#24 lore raw

PATCH v3 1/1 dw_mmc: turn on the IO supply

Hi Peng, Simon,

This patch is delegate to Philipp now, does this patch should go to Rockchip branch first?
I can do it if yes, but double with you in case this patch should go to mmc branch first.

Thanks,
- Kever

Quoted 67 lines · Peng Fan <[email protected]> 于2019年5月20日周一 下午3:04写道:
Quoted 28 lines

Subject: [U-Boot] [PATCH v3] dw_mmc: turn on the IO supply

Fixes the microSD slot on the ASUS C201.

---
v2: use #if CONFIG_IS_ENABLED(DM_REGULATOR)
v3: Move ret variable inside the if block (only used there)

---
 drivers/mmc/dw_mmc.c | 16 ++++++++++++++++
 1 file changed, 16 insertions(+)
drivers/mmc/dw_mmc.c index renamed from drivers/mmc/dw_mmc.c +0 -0

1992d61182..22f6c7eefd 100644

drivers/mmc/dw_mmc.c +1 -0
@@ -13,6 +13,7 @@
 #include <mmc.h>
 #include <dwmmc.h>
 #include <wait_bit.h>
+#include <power/regulator.h>
 
 #define PAGE_SIZE 4096
 
@@ -493,6 +494,21 @@ static int dwmci_set_ios(struct mmc *mmc)
      if (host->clksel)
              host->clksel(host);
 

drivers/mmc/dw_mmc.c static int dwmci_set_ios(struct mmc *mmc) · line 506

Quoted 10 lines
drivers/mmc/dw_mmc.c +10 -0
+#if CONFIG_IS_ENABLED(DM_REGULATOR)
+     if (mmc->vqmmc_supply) {
+             int ret;
+
+             if (mmc->signal_voltage == MMC_SIGNAL_VOLTAGE_180)
+                     regulator_set_value(mmc->vqmmc_supply, 1800000);
+             else
+                     regulator_set_value(mmc->vqmmc_supply, 3300000);
+
+             ret = regulator_set_enable_if_allowed(mmc->vqmmc_supply,

true);

drivers/mmc/dw_mmc.c static int dwmci_set_ios(struct mmc *mmc) · line 511

Quoted 5 lines
drivers/mmc/dw_mmc.c +5 -0
+             if (ret)
+                     return ret;
+     }
+#endif
+
Quoted 10 lines
drivers/mmc/dw_mmc.c +0 -1
      return 0;
 }
 
--

2.21.0

_______________________________________________
U-Boot mailing list
U-Boot at lists.denx.de

https://eur01.safelinks.protection.outlook.com/?url=https%3A%2F%2Flists.d

Quoted 4 lines

enx.de%2Flistinfo%2Fu-boot&amp;data=02%7C01%7CPeng.Fan%40nxp.com %7Ce49ea3fa171f4230a40308d6d7a680fc%7C686ea1d3bc2b4c6fa92cd99c5 c301635%7C0%7C0%7C636933507511914198&amp;sdata=Hyq5z5xI5xYUPA tcKJU%2FmdM3PLgCdHQcIr4qkSbz4Fc%3D&amp;reserved=0

_______________________________________________
U-Boot mailing list
U-Boot at lists.denx.de
https://lists.denx.de/listinfo/u-boot

#25 lore raw

PATCH v3 1/1 dw_mmc: turn on the IO supply

Hi Kever,

You could pick it up, not depends on mmc branch.

Thanks,
Peng.

From: Kever Yang [mailto:kever.yang at rock-chips.com]
Sent: 2019年7月26日 15:06
To: Peng Fan <[email protected]>
Cc: Urja Rannikko <[email protected]>; u-boot at lists.denx.de
Subject: Re: [U-Boot] [PATCH v3] dw_mmc: turn on the IO supply

Hi Peng, Simon,

This patch is delegate to Philipp now, does this patch should go to Rockchip branch first? I can do it if yes, but double with you in case this patch should go to mmc branch first.

Thanks,
- Kever

Quoted 33 lines · Peng Fan <peng.fan at nxp.com<mailto:[email protected]>> 于2019年5月20日周一 下午3:04写道:

Subject: [U-Boot] [PATCH v3] dw_mmc: turn on the IO supply

Fixes the microSD slot on the ASUS C201.

Signed-off-by Urja Rannikko <urjaman at gmail.com<mailto:[email protected]>>

---
v2: use #if CONFIG_IS_ENABLED(DM_REGULATOR)
v3: Move ret variable inside the if block (only used there)

---
 drivers/mmc/dw_mmc.c | 16 ++++++++++++++++
 1 file changed, 16 insertions(+)
drivers/mmc/dw_mmc.c index renamed from drivers/mmc/dw_mmc.c +0 -0

1992d61182..22f6c7eefd 100644

drivers/mmc/dw_mmc.c +6 -0
@@ -13,6 +13,7 @@
 #include <mmc.h>
 #include <dwmmc.h>
 #include <wait_bit.h>
+#include <power/regulator.h>
 
 #define PAGE_SIZE 4096
 
@@ -493,6 +494,21 @@ static int dwmci_set_ios(struct mmc *mmc)
      if (host->clksel)
              host->clksel(host);
 
+#if CONFIG_IS_ENABLED(DM_REGULATOR)
+     if (mmc->vqmmc_supply) {
+             int ret;
+
+             if (mmc->signal_voltage == MMC_SIGNAL_VOLTAGE_180)

drivers/mmc/dw_mmc.c static int dwmci_set_ios(struct mmc *mmc) · line 511

Quoted 10 lines
drivers/mmc/dw_mmc.c +10 -0
+                     regulator_set_value(mmc->vqmmc_supply, 1800000);
+             else
+                     regulator_set_value(mmc->vqmmc_supply, 3300000);
+
+             ret = regulator_set_enable_if_allowed(mmc->vqmmc_supply, true);
+             if (ret)
+                     return ret;
+     }
+#endif
+
Reviewed-by Peng Fan <peng.fan at nxp.com<mailto:[email protected]>>
Quoted 6 lines
drivers/mmc/dw_mmc.c +0 -1
      return 0;
 }
 
--

2.21.0

Quoted 8 lines

_______________________________________________
U-Boot mailing list
U-Boot at lists.denx.de<mailto:[email protected]>
https://eur01.safelinks.protection.outlook.com/?url=https%3A%2F%2Flists.d<https://lists.d>
enx.de<https://eur01.safelinks.protection.outlook.com/?url=http%3A%2F%2Fenx.de&data=02%7C01%7Cpeng.fan%40nxp.com%7C11ccf6dc22aa4b87faa308d71197bfa7%7C686ea1d3bc2b4c6fa92cd99c5c301635%7C0%7C0%7C636997215762062849&sdata=ZRlAdJM9%2BafKEW2CIgJ9AKbFsgYpZ2q61m%2Fxh5ktdkU%3D&reserved=0>%2Flistinfo%2Fu-boot&amp;data=02%7C01%7CPeng.Fan%40nxp.com<https://eur01.safelinks.protection.outlook.com/?url=http%3A%2F%2F40nxp.com&data=02%7C01%7Cpeng.fan%40nxp.com%7C11ccf6dc22aa4b87faa308d71197bfa7%7C686ea1d3bc2b4c6fa92cd99c5c301635%7C0%7C0%7C636997215762062849&sdata=km5AEShdXwRM2MtHYLm2662Sz9Z1b2C7RiZgMeygUsc%3D&reserved=0> %7Ce49ea3fa171f4230a40308d6d7a680fc%7C686ea1d3bc2b4c6fa92cd99c5
c301635%7C0%7C0%7C636933507511914198&amp;sdata=Hyq5z5xI5xYUPA
tcKJU%2FmdM3PLgCdHQcIr4qkSbz4Fc%3D&amp;reserved=0

_______________________________________________
U-Boot mailing list
U-Boot at lists.denx.de<mailto:[email protected]>
https://lists.denx.de/listinfo/u-boot<https://eur01.safelinks.protection.outlook.com/?url=https%3A%2F%2Flists.denx.de%2Flistinfo%2Fu-boot&data=02%7C01%7Cpeng.fan%40nxp.com%7C11ccf6dc22aa4b87faa308d71197bfa7%7C686ea1d3bc2b4c6fa92cd99c5c301635%7C0%7C0%7C636997215762072839&sdata=kbZSSFXgebiyhUrLvOjydXNYlfgIMltU3fXhmNUAmxc%3D&reserved=0>

maintainer
#26 lore raw

cover Veyron Speedy / ASUS C201 fixes

In reply to Urja Rannikko

Hi Urja,

Am Freitag, 22. März 2019, 20:14:32 CET schrieb Urja Rannikko:

Quoted 16 lines

Just saying: I'm not familiar with this stuff so i might've misunderstood something in my commit messages, but anyways i managed to make it boot.

There's more stuff than just these, but i decided I'd keep it short... If you want to know my full setup, you can check out my gh repo for now:
https://github.com/urjaman/u-boot/commits/c201-stuff

Notably I did include http://patchwork.ozlabs.org/patch/1040541/ in my tree when testing (rk3288 pinctrl patch that was linked to on the previous thread i found about the veyron speedy being broken).

The one thing that I havent yet fixed is that the USB host port on the otg-capable controller (near screen) doesn't work - one guess i have is that the dwc2 host driver doesn't force it into host mode... (It doesn't even read dr_mode ...)

Quoted 5 lines

Urja Rannikko (4):
pinctrl: exit pinconfig_post_bind if there are no subnodes
rk3288-board: remove pinctrl call for debug uart
rk3288-board: cosmetic: document selecting RK PWM
dw_mmc: turn on the IO supply

In my u-boot list archive I only see the list itself als recipient. You may want to resend your series with actual maintainers in Cc so that they actually see your patches :-) .

Similar to the kernel, uboot also has a scripts/get_maintainer.pl script to give you a list of applicable maintainers (Philipp and Simon in this case)

Heiko

#27 lore raw

cover Veyron Speedy / ASUS C201 fixes

Hi,

Quoted 8 lines · Heiko Stübner

In my u-boot list archive I only see the list itself als recipient. You may want to resend your series with actual maintainers in Cc so that they actually see your patches :-) .

Similar to the kernel, uboot also has a scripts/get_maintainer.pl script to give you a list of applicable maintainers (Philipp and Simon in this case)

I did CC these to 3 maintainers (who you mentioned plus Jaehoon Chung for MMC ...), as given by that script - weird that you dont see those. Gmail is showing them in the one i sent so atleast i didnt forget but maybe the list filtered them (?).

--
Urja Rannikko

maintainer
#28 lore raw

cover Veyron Speedy / ASUS C201 fixes

Am Sonntag, 31. März 2019, 21:34:00 CEST schrieb Urja Rannikko:

Quoted 8 lines

Hi,

Quoted 5 lines · Heiko Stübner

In my u-boot list archive I only see the list itself als recipient. You may want to resend your series with actual maintainers in Cc so that they actually see your patches :-) .

Quoted 8 lines
Quoted 3 lines

Similar to the kernel, uboot also has a scripts/get_maintainer.pl script to give you a list of applicable maintainers (Philipp and Simon in this case)

I did CC these to 3 maintainers (who you mentioned plus Jaehoon Chung for MMC ...), as given by that script - weird that you dont see those. Gmail is showing them in the one i sent so atleast i didnt forget but maybe the list filtered them (?).

yeah, that is possible ... for whatever reason
and I could only reply to the list mail itself ... so all good then :-)

Heiko

#29 lore raw

cover Veyron Speedy / ASUS C201 fixes

In reply to Urja Rannikko

Hi all,

Quoted 6 lines · Urja Rannikko

Urja Rannikko (4):
pinctrl: exit pinconfig_post_bind if there are no subnodes
rk3288-board: remove pinctrl call for debug uart
rk3288-board: cosmetic: document selecting RK PWM
dw_mmc: turn on the IO supply

I've now replied to what needs replying (and i'll send a fix to the dw_mmc one tomorrow),
but I have a question about fixing the defconfig(s) so that the default configuration is bootable (and maybe also sensible). I made the changes in a very "dont touch anything but speedy"-way in my git tree, but I was thinking
it would make sense to fix all the veyrons at once (maybe all rk3288 boards, but I know even less about them to make the changes)?

Anyways the changes for veyrons would mostly be enable pinctrl for SPI in SPL, slim down the SPL config (drop eMMC from SPL)
and turn off OTG support since they dont have an OTG port.

There is also currently a forced #define CONFIG_CMD_USB_MASS_STORAGE* in rk3288_common.h that seems not very sensible
for devices without an OTG port, i think that would be better moved into defconfig of non-veyron rk3288 boards.
(My patch just does an #ifdef otg..)

(Oh i also have a patch to set the correct spi flash bus for veyrons so that environment in SPI flash works...)

Anyways, comments?

-- Urja Rannikko

(* This feature is a command to turn the device into an usb mass storage device (default block device is mmc, so into a card reader of sorts), not usb stick support just in case you were wondering.)

#30 lore raw

cover Veyron Speedy / ASUS C201 fixes

In reply to Urja Rannikko

Hi,

Quoted 4 lines · Urja Rannikko

Urja Rannikko (4):
pinctrl: exit pinconfig_post_bind if there are no subnodes
rk3288-board: remove pinctrl call for debug uart

AFAIK these first two from this series would be "ready" so to say, anything i should do?
(Resend only them as a 2-patch series?)

Signature · 1 line
Urja Rannikko