Skip to contentSkip to conversation

rockchip: chromebook_minnie: Enable sound

PATCH New Simon Glass 4 participants 27 messages Dec 29, 2018 – Mar 18, 2019

Subsystems (3)
lore
Share & tools
lore
#0 lore raw

PATCH 1/2 rockchip: chromebook_minnie: Enable sound

Enable sound for this board, which has the same codec as jerry.

---

 arch/arm/dts/rk3288-veyron-minnie.dts | 12 ++++++++++++
 configs/chromebook_minnie_defconfig   |  5 +++++
 2 files changed, 17 insertions(+)
arch/arm/dts/rk3288-veyron-minnie.dts +12 -0
index 0a2915fba09..646f6ae7424 100644
@@ -85,6 +85,18 @@
 		regulator-boot-on;
 		vin-supply = <&vcc18_wl>;
 	};
+
+	sound {
+		compatible = "rockchip,audio-max98090-jerry";
+
+		cpu {
+			sound-dai = <&i2s 0>;
+		};
+
+		codec {
+			sound-dai = <&max98090 0>;
+		};
+	};
 };
 
 &backlight {
configs/chromebook_minnie_defconfig +5 -0
index b0428740736..93e40c02101 100644
@@ -31,6 +31,7 @@ CONFIG_CMD_USB=y
 # CONFIG_CMD_SETEXPR is not set
 CONFIG_CMD_CACHE=y
 CONFIG_CMD_TIME=y
+CONFIG_CMD_SOUND=y
 CONFIG_CMD_PMIC=y
 CONFIG_CMD_REGULATOR=y
 # CONFIG_SPL_DOS_PARTITION is not set
@@ -75,6 +76,10 @@ CONFIG_RAM=y
 CONFIG_SPL_RAM=y
 CONFIG_DEBUG_UART_SHIFT=2
 CONFIG_ROCKCHIP_SERIAL=y
+CONFIG_SOUND=y
+CONFIG_I2S=y
+CONFIG_I2S_ROCKCHIP=y
+CONFIG_SOUND_MAX98090=y
 CONFIG_ROCKCHIP_SPI=y
 CONFIG_SYSRESET=y
 CONFIG_USB=y
Signature · 1 line
2.20.1.415.g653613c723-goog
#1 lore raw

PATCH 2/2 rockchip: Drop call to rockchip_dnl_mode_check() for now

This function causes a 5-second delay and stops the display working on minnie. This code should be in a driver and should only be enabled by a device-tree property, so that it does not affect devices which do not have this feature.

---

 arch/arm/mach-rockchip/boot_mode.c | 8 +++++++-
 1 file changed, 7 insertions(+), 1 deletion(-)
arch/arm/mach-rockchip/boot_mode.c +7 -1
index f32b3c4ce56..d7997d71e30 100644
@@ -61,7 +61,13 @@ int setup_boot_mode(void)
 	void *reg = (void *)CONFIG_ROCKCHIP_BOOT_MODE_REG;
 	int boot_mode = readl(reg);
 
-	rockchip_dnl_mode_check();
+	/*
+	 * This should be handled using a driver-tree property and a suitable
+	 * driver which can read the appropriate settings. As it is, this
+	 * breaks chromebook_minnie.\
+	 *
+	 * rockchip_dnl_mode_check();
+	 */
 
 	boot_mode = readl(reg);
 	debug("%s: boot mode 0x%08x\n", __func__, boot_mode);
Signature · 1 line
2.20.1.415.g653613c723-goog
#2 lore raw

PATCH 2/2 rockchip: Drop call to rockchip_dnl_mode_check() for now

Quoted 10 lines · Simon Glass

This function causes a 5-second delay and stops the display working on minnie. This code should be in a driver and should only be enabled by a device-tree property, so that it does not affect devices which do not have this feature.

#3 lore raw

PATCH 2/2 rockchip: Drop call to rockchip_dnl_mode_check() for now

Simon,

As an alternative, you can turn this off by setting CONFIG_BOOT_MODE_REG=0 in your defconfig.
We’ve done the same for our RK3399-Q7 module:

	configs/puma-rk3399_defconfig:CONFIG_ROCKCHIP_BOOT_MODE_REG=0x0

If this works for you, we can avoid removing this feature this late in the cycle and defeature/remove it in the next release, unless moved into a DTS-controlled driver.

Thanks,
Philipp.

Quoted 19 lines · Philipp Tomsich
Quoted 11 lines · Simon Glass

This function causes a 5-second delay and stops the display working on minnie. This code should be in a driver and should only be enabled by a device-tree property, so that it does not affect devices which do not have this feature.

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

#4 lore raw

PATCH 2/2 rockchip: Drop call to rockchip_dnl_mode_check() for now

Hi Philipp,

Quoted 10 lines · Philipp Tomsich

Simon,

As an alternative, you can turn this off by setting CONFIG_BOOT_MODE_REG=0 in your defconfig.
We’ve done the same for our RK3399-Q7 module:

        configs/puma-rk3399_defconfig:CONFIG_ROCKCHIP_BOOT_MODE_REG=0x0

If this works for you, we can avoid removing this feature this late in the cycle and defeature/remove it in the next release, unless moved into a DTS-controlled driver.

That's OK with me. Since the display was broken in the last release too, I'll just wait for the fix.

Regards,
Simon

#5 lore raw

PATCH 2/2 rockchip: Drop call to rockchip_dnl_mode_check() for now[Please note, mail behalf by 3vnmnxammciszqnjoyvtpbt.vynrlcly.fhunyvjr-jopwz.jvt@chromium-sendgmr.bounces.google.com]

In reply to Simon Glass

Hi Simon,

Quoted 2 lines · Simon Glass

This function causes a 5-second delay and stops the display working on minnie.

I don't understand why need 5-second, the logic is to check the adc key press to decide the mode, maybe because the different usage of the adc in minnie?

Most of boards(which do not have ec like chromebook) need this check to get into download mode.
I would prefer chrombook to implement a function to replace the week rockchip_dnl_mode_check();

Thanks,
- Kever

Quoted 29 lines

This code should be in a driver and should only be enabled by a device-tree property, so that it does not affect devices which do not have this feature.

---

 arch/arm/mach-rockchip/boot_mode.c | 8 +++++++-
 1 file changed, 7 insertions(+), 1 deletion(-)
arch/arm/mach-rockchip/boot_mode.c +7 -1
index f32b3c4ce56..d7997d71e30 100644
@@ -61,7 +61,13 @@ int setup_boot_mode(void)
 	void *reg = (void *)CONFIG_ROCKCHIP_BOOT_MODE_REG;
 	int boot_mode = readl(reg);
 
-	rockchip_dnl_mode_check();
+	/*
+	 * This should be handled using a driver-tree property and a suitable
+	 * driver which can read the appropriate settings. As it is, this
+	 * breaks chromebook_minnie.\
+	 *
+	 * rockchip_dnl_mode_check();
+	 */
 
 	boot_mode = readl(reg);
 	debug("%s: boot mode 0x%08x\n", __func__, boot_mode);
#6 lore raw

PATCH 2/2 rockchip: Drop call to rockchip_dnl_mode_check() for now[Please note, mail behalf by 3vnmnxammciszqnjoyvtpbt.vynrlcly.fhunyvjr-jopwz.jvt@chromium-sendgmr.bounces.google.com]

Hi,

Quoted 8 lines · Kever Yang

Hi Simon,

Quoted 2 lines · Simon Glass

This function causes a 5-second delay and stops the display working on minnie.

Quoted 8 lines

I don't understand why need 5-second, the logic is to check the adc key press to decide the mode, maybe because the different usage of the adc in minnie?

Most of boards(which do not have ec like chromebook) need this check to get into download mode.
I would prefer chrombook to implement a function to replace the week rockchip_dnl_mode_check();

But where is it specified in the DT that it uses the ADC for this feature? It really should have a driver...

Regards,
Simon

Quoted 35 lines

Thanks,
- Kever

Quoted 29 lines

This code should be in a driver and should only be enabled by a device-tree property, so that it does not affect devices which do not have this feature.

---

 arch/arm/mach-rockchip/boot_mode.c | 8 +++++++-
 1 file changed, 7 insertions(+), 1 deletion(-)
arch/arm/mach-rockchip/boot_mode.c +7 -1
index f32b3c4ce56..d7997d71e30 100644
@@ -61,7 +61,13 @@ int setup_boot_mode(void)
      void *reg = (void *)CONFIG_ROCKCHIP_BOOT_MODE_REG;
      int boot_mode = readl(reg);
 
-     rockchip_dnl_mode_check();
+     /*
+      * This should be handled using a driver-tree property and a suitable
+      * driver which can read the appropriate settings. As it is, this
+      * breaks chromebook_minnie.\
+      *
+      * rockchip_dnl_mode_check();
+      */
 
      boot_mode = readl(reg);
      debug("%s: boot mode 0x%08x\n", __func__, boot_mode);
#7 lore raw

PATCH 2/2 rockchip: Drop call to rockchip_dnl_mode_check() for now[Please note, mail behalf by 3vnmnxammciszqnjoyvtpbt.vynrlcly.fhunyvjr-jopwz.jvt@chromium-sendgmr.bounces.google.com]

Quoted 19 lines · Simon Glass

Hi,

Quoted 12 lines · On Thu, 3 Jan 2019 at 00:32, Kever Yang <kever.yang at rock-chips.com…

Hi Simon,

Quoted 2 lines · Simon Glass

This function causes a 5-second delay and stops the display working on minnie.

I don't understand why need 5-second, the logic is to check the adc key press to decide the mode, maybe because the different usage of the adc in minnie?

Quoted 7 lines
Quoted 4 lines

Most of boards(which do not have ec like chromebook) need this check to get into download mode.
I would prefer chrombook to implement a function to replace the week rockchip_dnl_mode_check();

But where is it specified in the DT that it uses the ADC for this feature? It really should have a driver…

3 things:
a. It really should have a driver.
b. One does not have to replace rockchip_dnl_mode_check(), as there
already is logic to remove it if CONFIG_ROCKCHIP_BOOT_MODE_REG is set to 0 in the config.
c. We really need to get away from configuring magic addresses via Kconfig.

#8 lore raw

PATCH 2/2 rockchip: Drop call to rockchip_dnl_mode_check() for now

In reply to Simon Glass
Quoted 5 lines

This function causes a 5-second delay and stops the display working on minnie. This code should be in a driver and should only be enabled by a device-tree property, so that it does not affect devices which do not have this feature.

Quoted 7 lines

---

 arch/arm/mach-rockchip/boot_mode.c | 8 +++++++-
 1 file changed, 7 insertions(+), 1 deletion(-)

Applied to u-boot-rockchip, thanks!

#9 lore raw

PATCH 2/2 rockchip: Drop call to rockchip_dnl_mode_check() for now

Hi:

Quoted 5 lines

This function causes a 5-second delay and stops the display working on minnie. This code should be in a driver and should only be enabled by a device-tree property, so that it does not affect devices which do not have this feature.

Quoted 7 lines

---

 arch/arm/mach-rockchip/boot_mode.c | 8 +++++++-
 1 file changed, 7 insertions(+), 1 deletion(-)

》Applied to u-boot-rockchip, thanks!

According to the previous discussing, shouldn‘t this be fixed by by setting CONFIG_BOOT_MODE_REG=0 ? All rockchip products need this function,we really don't want it be turned off.

#10 lore raw

PATCH 2/2 rockchip: Drop call to rockchip_dnl_mode_check() for now

Andy,

Quoted 17 lines · andy.yan at rock-chips.com

Hi:

Quoted 5 lines

This function causes a 5-second delay and stops the display working on minnie. This code should be in a driver and should only be enabled by a device-tree property, so that it does not affect devices which do not have this feature.

Quoted 7 lines

---

 arch/arm/mach-rockchip/boot_mode.c | 8 +++++++-
 1 file changed, 7 insertions(+), 1 deletion(-)
Quoted 5 lines

》Applied to u-boot-rockchip, thanks!

According to the previous discussing, shouldn‘t this be fixed by by setting CONFIG_BOOT_MODE_REG=0 ? All rockchip products need this function,we really don't want it be turned off.

The actual decision was that this needed to be converted to the driver-model. Setting CONFIG_BOOT_MODE_REG=0 was a workaround that we used only for 2019.1, so I could delay applying Simon’s patch until this cycle.

Please convert this into a DM driver, so we can add the driver for 2019.4.

Thanks,
Philipp.

#11 lore raw

PATCH 2/2 rockchip: Drop call to rockchip_dnl_mode_check() for now

Quoted 26 lines · Philipp Tomsich

Andy,

Quoted 20 lines · andy.yan at rock-chips.com

Hi:

Quoted 5 lines

This function causes a 5-second delay and stops the display working on minnie. This code should be in a driver and should only be enabled by a device-tree property, so that it does not affect devices which do not have this feature.

Quoted 7 lines

---

arch/arm/mach-rockchip/boot_mode.c | 8 +++++++-

1 file changed, 7 insertions(+), 1 deletion(-)

》Applied to u-boot-rockchip, thanks!

Quoted 6 lines
Quoted 2 lines

According to the previous discussing, shouldn‘t this be fixed by by setting CONFIG_BOOT_MODE_REG=0 ? All rockchip products need this function,we really don't want it be turned off.

The actual decision was that this needed to be converted to the driver-model. Setting CONFIG_BOOT_MODE_REG=0 was a workaround that we used only for 2019.1, so I could delay applying Simon’s patch until this cycle.

Just for the record, this is a verbatim quote from the original review:

Quoted 13 lines

If this works for you, we can avoid removing this feature this late in the cycle and defeature/remove it in the next release, unless moved into a DTS-controlled driver.

Please convert this into a DM driver, so we can add the driver for 2019.4.

Thanks,
Philipp.

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

#12 lore raw

PATCH 2/2 rockchip: Drop call to rockchip_dnl_mode_check() for now【请注意,邮件由[email protected]代发】

In reply to Philipp Tomsich

Hi Philipp,

    This is not right,  this patch should not merged like this!!!

    I have give my review comment in previous mail, and this will break many boards.

    My another patch do not break anything, but you insist NAK it without acceptable reason;
    This patch definitely break other board and I have comment it, but you just ignore other people's review and merge it, good job!

Thanks,
- Kever

Quoted 17 lines · Philipp Tomsich
Quoted 5 lines

This function causes a 5-second delay and stops the display working on minnie. This code should be in a driver and should only be enabled by a device-tree property, so that it does not affect devices which do not have this feature.

Quoted 7 lines

---

 arch/arm/mach-rockchip/boot_mode.c | 8 +++++++-
 1 file changed, 7 insertions(+), 1 deletion(-)

Applied to u-boot-rockchip, thanks!
_______________________________________________
U-Boot mailing list
U-Boot at lists.denx.de
https://lists.denx.de/listinfo/u-boot

#13 lore raw

PATCH 2/2 rockchip: Drop call to rockchip_dnl_mode_check() for now【请注意,邮件由[email protected]代发】

Quoted 13 lines · Kever Yang

Hi Philipp,

This is not right, this patch should not merged like this!!!

I have give my review comment in previous mail, and this will break
many boards.

My another patch do not break anything, but you insist NAK it
without acceptable reason;

What other patch?
I don’t remember seeing that one...

Quoted 26 lines

This patch definitely break other board and I have comment it, but
you just ignore other people's review and merge it, good job!

Thanks,
- Kever

Quoted 17 lines · Philipp Tomsich
Quoted 5 lines

This function causes a 5-second delay and stops the display working on minnie. This code should be in a driver and should only be enabled by a device-tree property, so that it does not affect devices which do not have this feature.

Quoted 7 lines

---

arch/arm/mach-rockchip/boot_mode.c | 8 +++++++-

1 file changed, 7 insertions(+), 1 deletion(-)

Applied to u-boot-rockchip, thanks!
_______________________________________________
U-Boot mailing list
U-Boot at lists.denx.de
https://lists.denx.de/listinfo/u-boot

#14 lore raw

PATCH 2/2 rockchip: Drop call to rockchip_dnl_mode_check() for now【请注意,邮件由[email protected]代发】

Kever,

Independent of whether we revert this for the current cycle (and also independent of if I ever find the other patch you had been referring to — I couldn’t find it in my local mailing list archive) and then deprecate it for the next release (unless converted to DM), we still have a number of architectural issues that need to be addressed:
1. This really should be a driver under DTS control.
2. We need to not get away from configuring SOM-specific addresses via Kconfig

Both these issues are technical debt that we’ve accumulated over the last 18 months and need to address for the sake of future maintainability.
E.g. ‘setting an address to 0x0 via Kconfig to disable a driver/feature’ really isn’t in line with the architectural direction of U-Boot.

I don’t have my own house completely in order (I’ve been talking for a year now about finally wrapping the RGMII/GMII selection into an ioctl-call to a driver) yet, but that doesn’t mean that we we should delay this clean-up more than absolutely necessary.

Thanks,
Philipp.

Quoted 50 lines · Philipp Tomsich
Quoted 14 lines · On 01.02.2019, at 10:32, Kever Yang <kever.yang at rock-chips.com…

Hi Philipp,

This is not right, this patch should not merged like this!!!

I have give my review comment in previous mail, and this will break
many boards.

My another patch do not break anything, but you insist NAK it
without acceptable reason;

What other patch?
I don’t remember seeing that one...

Quoted 26 lines

This patch definitely break other board and I have comment it, but
you just ignore other people's review and merge it, good job!

Thanks,
- Kever

Quoted 17 lines · Philipp Tomsich
Quoted 5 lines

This function causes a 5-second delay and stops the display working on minnie. This code should be in a driver and should only be enabled by a device-tree property, so that it does not affect devices which do not have this feature.

Quoted 7 lines

---

arch/arm/mach-rockchip/boot_mode.c | 8 +++++++-

1 file changed, 7 insertions(+), 1 deletion(-)

Applied to u-boot-rockchip, thanks!
_______________________________________________
U-Boot mailing list
U-Boot at lists.denx.de
https://lists.denx.de/listinfo/u-boot

_______________________________________________
U-Boot mailing list
U-Boot at lists.denx.de <mailto:[email protected]>
https://lists.denx.de/listinfo/u-boot <https://lists.denx.de/listinfo/u-boot>

#15 lore raw

PATCH 2/2 rockchip: Drop call to rockchip_dnl_mode_check() for now【请注意,邮件由[email protected]代发】

Hi Kever,

Quoted 9 lines · Philipp Tomsich

Kever,

Independent of whether we revert this for the current cycle (and also independent of if I ever find the other patch you had been referring to — I couldn’t find it in my local mailing list archive) and then deprecate it for the next release (unless converted to DM), we still have a number of architectural issues that need to be addressed:
1. This really should be a driver under DTS control.
2. We need to not get away from configuring SOM-specific addresses via Kconfig

Let me know if I can help with this. Philipp I am find with any short-term solution that comes up, but agree that it should be tidied up.

- Simon

Quoted 67 lines

Both these issues are technical debt that we’ve accumulated over the last 18 months and need to address for the sake of future maintainability.
E.g. ‘setting an address to 0x0 via Kconfig to disable a driver/feature’ really isn’t in line with the architectural direction of U-Boot.

I don’t have my own house completely in order (I’ve been talking for a year now about finally wrapping the RGMII/GMII selection into an ioctl-call to a driver) yet, but that doesn’t mean that we we should delay this clean-up more than absolutely necessary.

Thanks,
Philipp.

Quoted 46 lines · Philipp Tomsich
Quoted 14 lines · On 01.02.2019, at 10:32, Kever Yang <kever.yang at rock-chips.com…

Hi Philipp,

This is not right, this patch should not merged like this!!!

I have give my review comment in previous mail, and this will break
many boards.

My another patch do not break anything, but you insist NAK it
without acceptable reason;

What other patch?
I don’t remember seeing that one...

Quoted 26 lines

This patch definitely break other board and I have comment it, but
you just ignore other people's review and merge it, good job!

Thanks,
- Kever

Quoted 17 lines · Philipp Tomsich
Quoted 5 lines

This function causes a 5-second delay and stops the display working on minnie. This code should be in a driver and should only be enabled by a device-tree property, so that it does not affect devices which do not have this feature.

Quoted 7 lines

---

arch/arm/mach-rockchip/boot_mode.c | 8 +++++++-

1 file changed, 7 insertions(+), 1 deletion(-)

Applied to u-boot-rockchip, thanks!
_______________________________________________
U-Boot mailing list
U-Boot at lists.denx.de
https://lists.denx.de/listinfo/u-boot

Quoted 4 lines

_______________________________________________
U-Boot mailing list
U-Boot at lists.denx.de <mailto:[email protected]>
https://lists.denx.de/listinfo/u-boot <https://lists.denx.de/listinfo/u-boot>

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

#16 lore raw

PATCH 2/2 rockchip: Drop call to rockchip_dnl_mode_check() for now【请注意,邮件由[email protected]代发】

In reply to Philipp Tomsich

Hi Philipp:

    Sorry for the late reply, we just come back from the Chinese Spring
Festival.

Quoted 10 lines · Philipp Tomsich

Kever,

Independent of whether we revert this for the current cycle (and also independent of
if I ever find the other patch you had been referring to — I couldn’t find it in my local
mailing list archive) and then deprecate it for the next release (unless converted to
DM), we still have a number of architectural issues that need to be addressed:

I still doubt  is this a right  work-flow for patch apply. Before we apply  a patch  which will break many other boards , should we  make sure there is a solution patch applied for these boars first?

Quoted 11 lines

1.This really should be a driver under DTS control.
2.We need to not get away from configuring SOM-specific addresses via Kconfig

Both these issues are technical debt that we’ve accumulated over the last 18 months
and need to address for the sake of future maintainability. E.g. ‘setting an address to 0x0 via Kconfig to disable a driver/feature’ really isn’t in line
with the architectural direction of U-Boot.

For technical side, I think CONFIG_ROCKCHIP_BOOT_MODE_REG is necessary here, we will read this register from save_boot_params when we get out from bootrom,  the dtb is not available at this point.

On the other hand, almost rockchip based products use a recovery key to enter download(upgrade)mode, this is a muti-funtion key, most of them reuse with vol+- key,  we would like the u-boot share

dtb with linux kernel. To keep the linux kernel dts as clean as possible ,we don't want to add another dts property to describe this key either. This is why I implement function rockchip_dnl_key_pressed as __weak.

Quoted 67 lines

I don’t have my own house completely in order (I’ve been talking for a year now about
finally wrapping the RGMII/GMII selection into an ioctl-call to a driver) yet, but that doesn’t
mean that we we should delay this clean-up more than absolutely necessary.

Thanks,
Philipp.

Quoted 57 lines
Quoted 5 lines · On 01.02.2019, at 10:34, Philipp Tomsich…

On 01.02.2019, at 10:32, Kever Yang <[email protected] <mailto:[email protected]>> wrote:

Hi Philipp,

Quoted 7 lines

   This is not right,  this patch should not merged like this!!!

   I have give my review comment in previous mail, and this will break many boards.

   My another patch do not break anything, but you insist NAK it without acceptable reason;

What other patch?
I don’t remember seeing that one...

Quoted 30 lines

   This patch definitely break other board and I have comment it, but you just ignore other people's review and merge it, good job!

Thanks,
- Kever

Quoted 21 lines · Philipp Tomsich
Quoted 7 lines

This function causes a 5-second delay and stops the display working on minnie. This code should be in a driver and should only be enabled by a device-tree property, so that it does not affect devices which do not
have this feature.

Quoted 9 lines

<mailto:[email protected]>>
Reviewed-by: Philipp Tomsich
<[email protected]
<mailto:[email protected]>>
---

arch/arm/mach-rockchip/boot_mode.c | 8 +++++++-

1 file changed, 7 insertions(+), 1 deletion(-)

Applied to u-boot-rockchip, thanks!
_______________________________________________
U-Boot mailing list
U-Boot at lists.denx.de <mailto:[email protected]>
https://lists.denx.de/listinfo/u-boot

_______________________________________________
U-Boot mailing list
U-Boot at lists.denx.de <mailto:[email protected]>
https://lists.denx.de/listinfo/u-boot

#17 lore raw

PATCH 2/2 rockchip: Drop call to rockchip_dnl_mode_check() for now【请注意,邮件由[email protected]代发】

HI Andy,

Quoted 34 lines · Andy Yan

Hi Philipp:

Sorry for the late reply, we just come back from the Chinese Spring
Festival.

Quoted 10 lines · Philipp Tomsich

Kever,

Independent of whether we revert this for the current cycle (and also independent of
if I ever find the other patch you had been referring to — I couldn’t find it in my local
mailing list archive) and then deprecate it for the next release (unless converted to
DM), we still have a number of architectural issues that need to be addressed:

I still doubt is this a right work-flow for patch apply. Before we apply a patch which will break many other boards , should we make sure there is a solution patch applied for these boars first?

Quoted 11 lines

1.This really should be a driver under DTS control.
2.We need to not get away from configuring SOM-specific addresses via Kconfig

Both these issues are technical debt that we’ve accumulated over the last 18 months
and need to address for the sake of future maintainability. E.g. ‘setting an address to 0x0 via Kconfig to disable a driver/feature’ really isn’t in line
with the architectural direction of U-Boot.

Quoted 3 lines

For technical side, I think CONFIG_ROCKCHIP_BOOT_MODE_REG is necessary here, we will read this register from save_boot_params when we get out from bootrom, the dtb is not available at this point.

Yes this is happening very early, but I wonder why this is necessary? Can it be moved to later?

The call to check_back_to_brom_dnl_flag() happens much later so there should be no problem to convert that.

Quoted 9 lines

On the other hand, almost rockchip based products use a recovery key to enter download(upgrade)mode, this is a muti-funtion key, most of them reuse with vol+- key, we would like the u-boot share

dtb with linux kernel. To keep the linux kernel dts as clean as possible ,we don't want to add another dts property to describe this key either. This is why I implement function rockchip_dnl_key_pressed as __weak.

OK, but given that it is called later, it seems to me that it could use a driver.

Regards,
Simon

Quoted 72 lines
Quoted 68 lines

I don’t have my own house completely in order (I’ve been talking for a year now about
finally wrapping the RGMII/GMII selection into an ioctl-call to a driver) yet, but that doesn’t
mean that we we should delay this clean-up more than absolutely necessary.

Thanks,
Philipp.

Quoted 57 lines
Quoted 5 lines · On 01.02.2019, at 10:34, Philipp Tomsich…

On 01.02.2019, at 10:32, Kever Yang <[email protected] <mailto:[email protected]>> wrote:

Hi Philipp,

Quoted 7 lines

This is not right, this patch should not merged like this!!!

I have give my review comment in previous mail, and this will break
many boards.

My another patch do not break anything, but you insist NAK it
without acceptable reason;

What other patch?
I don’t remember seeing that one...

Quoted 30 lines

This patch definitely break other board and I have comment it, but
you just ignore other people's review and merge it, good job!

Thanks,
- Kever

Quoted 21 lines · Philipp Tomsich
Quoted 7 lines

This function causes a 5-second delay and stops the display working on minnie. This code should be in a driver and should only be enabled by a device-tree property, so that it does not affect devices which do not
have this feature.

Quoted 9 lines

<mailto:[email protected]>>
Reviewed-by: Philipp Tomsich
<[email protected]
<mailto:[email protected]>>
---

arch/arm/mach-rockchip/boot_mode.c | 8 +++++++-

1 file changed, 7 insertions(+), 1 deletion(-)

Applied to u-boot-rockchip, thanks!
_______________________________________________
U-Boot mailing list
U-Boot at lists.denx.de <mailto:[email protected]>
https://lists.denx.de/listinfo/u-boot

_______________________________________________
U-Boot mailing list
U-Boot at lists.denx.de <mailto:[email protected]>
https://lists.denx.de/listinfo/u-boot

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

#18 lore raw

PATCH 2/2 rockchip: Drop call to rockchip_dnl_mode_check() for now【请注意,邮件由[email protected]代发】【请注意,邮件由[email protected]代发】

Hi Simon:

Quoted 48 lines · Simon Glass

HI Andy,

Quoted 32 lines · Andy Yan

Hi Philipp:

Sorry for the late reply, we just come back from the Chinese Spring
Festival.

Quoted 10 lines · Philipp Tomsich

Kever,

Independent of whether we revert this for the current cycle (and also independent of
if I ever find the other patch you had been referring to — I couldn’t find it in my local
mailing list archive) and then deprecate it for the next release (unless converted to
DM), we still have a number of architectural issues that need to be addressed:

I still doubt is this a right work-flow for patch apply. Before we apply a patch which will break many other boards , should we make sure there is a solution patch applied for these boars first?

Quoted 11 lines

1.This really should be a driver under DTS control.
2.We need to not get away from configuring SOM-specific addresses via Kconfig

Both these issues are technical debt that we’ve accumulated over the last 18 months
and need to address for the sake of future maintainability. E.g. ‘setting an address to 0x0 via Kconfig to disable a driver/feature’ really isn’t in line
with the architectural direction of U-Boot.

Quoted 3 lines

For technical side, I think CONFIG_ROCKCHIP_BOOT_MODE_REG is necessary here, we will read this register from save_boot_params when we get out from bootrom, the dtb is not available at this point.

Yes this is happening very early, but I wonder why this is necessary? Can it be moved to later?

The call to check_back_to_brom_dnl_flag() happens much later so there should be no problem to convert that.

Quoted 4 lines

On the other hand, almost rockchip based products use a recovery key to enter download(upgrade)mode, this is a muti-funtion key, most of them reuse with vol+- key, we would like the u-boot share

Quoted 6 lines
Quoted 4 lines

dtb with linux kernel. To keep the linux kernel dts as clean as possible ,we don't want to add another dts property to describe this key either. This is why I implement function rockchip_dnl_key_pressed as __weak.

OK, but given that it is called later, it seems to me that it could use a driver.

We can't let it called later. See the discuss here :
http://patchwork.ozlabs.org/patch/812228/

Quoted 73 lines

Regards,
Simon

Quoted 67 lines
Quoted 6 lines

I don’t have my own house completely in order (I’ve been talking for a year now about
finally wrapping the RGMII/GMII selection into an ioctl-call to a driver) yet, but that doesn’t
mean that we we should delay this clean-up more than absolutely necessary.

Quoted 57 lines

Thanks,
Philipp.

Quoted 54 lines
Quoted 5 lines · On 01.02.2019, at 10:34, Philipp Tomsich…

On 01.02.2019, at 10:32, Kever Yang <[email protected] <mailto:[email protected]>> wrote:

Hi Philipp,

Quoted 7 lines

This is not right, this patch should not merged like this!!!

I have give my review comment in previous mail, and this will break
many boards.

My another patch do not break anything, but you insist NAK it
without acceptable reason;

What other patch?
I don’t remember seeing that one...

Quoted 29 lines

This patch definitely break other board and I have comment it, but
you just ignore other people's review and merge it, good job!

Thanks,
- Kever

Quoted 21 lines · Philipp Tomsich
Quoted 7 lines

This function causes a 5-second delay and stops the display working on minnie. This code should be in a driver and should only be enabled by a device-tree property, so that it does not affect devices which do not
have this feature.

Quoted 9 lines

<mailto:[email protected]>>
Reviewed-by: Philipp Tomsich
<[email protected]
<mailto:[email protected]>>
---

arch/arm/mach-rockchip/boot_mode.c | 8 +++++++-

1 file changed, 7 insertions(+), 1 deletion(-)

Applied to u-boot-rockchip, thanks!
_______________________________________________
U-Boot mailing list
U-Boot at lists.denx.de <mailto:[email protected]>
https://lists.denx.de/listinfo/u-boot

_______________________________________________
U-Boot mailing list
U-Boot at lists.denx.de <mailto:[email protected]>
https://lists.denx.de/listinfo/u-boot

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

#19 lore raw

PATCH 2/2 rockchip: Drop call to rockchip_dnl_mode_check() for now【请注意,邮件由[email protected]代发】【请注意,邮件由[email protected]代发】

Hi Andy,

Quoted 56 lines · Andy Yan

Hi Simon:

Quoted 52 lines · Simon Glass

HI Andy,

Quoted 32 lines · Andy Yan

Hi Philipp:

Sorry for the late reply, we just come back from the Chinese Spring
Festival.

Quoted 10 lines · Philipp Tomsich

Kever,

Independent of whether we revert this for the current cycle (and also independent of
if I ever find the other patch you had been referring to — I couldn’t find it in my local
mailing list archive) and then deprecate it for the next release (unless converted to
DM), we still have a number of architectural issues that need to be addressed:

I still doubt is this a right work-flow for patch apply. Before we apply a patch which will break many other boards , should we make sure there is a solution patch applied for these boars first?

Quoted 11 lines

1.This really should be a driver under DTS control.
2.We need to not get away from configuring SOM-specific addresses via Kconfig

Both these issues are technical debt that we’ve accumulated over the last 18 months
and need to address for the sake of future maintainability. E.g. ‘setting an address to 0x0 via Kconfig to disable a driver/feature’ really isn’t in line
with the architectural direction of U-Boot.

Quoted 3 lines

For technical side, I think CONFIG_ROCKCHIP_BOOT_MODE_REG is necessary here, we will read this register from save_boot_params when we get out from bootrom, the dtb is not available at this point.

Yes this is happening very early, but I wonder why this is necessary? Can it be moved to later?

The call to check_back_to_brom_dnl_flag() happens much later so there should be no problem to convert that.

Quoted 8 lines

On the other hand, almost rockchip based products use a recovery key to enter download(upgrade)mode, this is a muti-funtion key, most of them reuse with vol+- key, we would like the u-boot share

dtb with linux kernel. To keep the linux kernel dts as clean as possible ,we don't want to add another dts property to describe this key either. This is why I implement function rockchip_dnl_key_pressed as __weak.

Quoted 7 lines
Quoted 2 lines

OK, but given that it is called later, it seems to me that it could use a driver.

We can't let it called later. See the discuss here :
http://patchwork.ozlabs.org/patch/812228/

Yes I read that, and took a look myself, and certainly it does not look very easy and I take your point about potential bugs being introduced.

It seems like you have tried quite hard, so I'm not going to object if you can't find a way. My main objection was that it broke other boards.

Is there any way the check could be delayed enough to actually be able to read the device tree? It could call spl_early_init() quite early and then do that.

Regards,
Simon

#20 lore raw

PATCH 2/2 rockchip: Drop call to rockchip_dnl_mode_check() for now【请注意,邮件由[email protected]代发】【请注意,邮件由[email protected]代发】

Quoted 74 lines · Simon Glass

Hi Andy,

Quoted 56 lines · Andy Yan

Hi Simon:

Quoted 52 lines · Simon Glass

HI Andy,

Quoted 32 lines · Andy Yan

Hi Philipp:

Sorry for the late reply, we just come back from the Chinese Spring
Festival.

Quoted 10 lines · Philipp Tomsich

Kever,

Independent of whether we revert this for the current cycle (and also independent of
if I ever find the other patch you had been referring to — I couldn’t find it in my local
mailing list archive) and then deprecate it for the next release (unless converted to
DM), we still have a number of architectural issues that need to be addressed:

I still doubt is this a right work-flow for patch apply. Before we apply a patch which will break many other boards , should we make sure there is a solution patch applied for these boars first?

Quoted 11 lines

1.This really should be a driver under DTS control.
2.We need to not get away from configuring SOM-specific addresses via Kconfig

Both these issues are technical debt that we’ve accumulated over the last 18 months
and need to address for the sake of future maintainability. E.g. ‘setting an address to 0x0 via Kconfig to disable a driver/feature’ really isn’t in line
with the architectural direction of U-Boot.

Quoted 3 lines

For technical side, I think CONFIG_ROCKCHIP_BOOT_MODE_REG is necessary here, we will read this register from save_boot_params when we get out from bootrom, the dtb is not available at this point.

Yes this is happening very early, but I wonder why this is necessary? Can it be moved to later?

The call to check_back_to_brom_dnl_flag() happens much later so there should be no problem to convert that.

Quoted 8 lines

On the other hand, almost rockchip based products use a recovery key to enter download(upgrade)mode, this is a muti-funtion key, most of them reuse with vol+- key, we would like the u-boot share

dtb with linux kernel. To keep the linux kernel dts as clean as possible ,we don't want to add another dts property to describe this key either. This is why I implement function rockchip_dnl_key_pressed as __weak.

Quoted 7 lines
Quoted 2 lines

OK, but given that it is called later, it seems to me that it could use a driver.

We can't let it called later. See the discuss here :
http://patchwork.ozlabs.org/patch/812228/

Yes I read that, and took a look myself, and certainly it does not look very easy and I take your point about potential bugs being introduced.

Quoted 7 lines

It seems like you have tried quite hard, so I'm not going to object if you can't find a way. My main objection was that it broke other boards.

Is there any way the check could be delayed enough to actually be able to read the device tree? It could call spl_early_init() quite early and then do that.

Note that the minimum improvement that I’d expect to get this fully enabled again would be a clean-up of the Kconfig options, so it is easy and (from the documentation) predictable for boards to turn this off.

I still want this to be turned back on in the current cycle.

Thanks,
Philipp.

#21 lore raw

PATCH 2/2 rockchip: Drop call to rockchip_dnl_mode_check() for now

Quoted 6 lines · Philipp Tomsich

Note that the minimum improvement that I’d expect to get this fully enabled again would be a clean-up of the Kconfig options, so it is easy and (from the documentation) predictable for boards to turn this off.

I still want this to be turned back on in the current cycle.

Great maintainer!!!!

You objection for other patch is that it break other boards(which is non exist actually),
and you merge this patch(break more than 10 boards very clearly) with objection comment.
And then you require for all the Kconfig options update before the feature back.

WoW!!!

Thanks,
- Kever

#22 lore raw

PATCH 2/2 rockchip: Drop call to rockchip_dnl_mode_check() for now

Hi,

Quoted 13 lines · Kever Yang
Quoted 6 lines · Philipp Tomsich

Note that the minimum improvement that I’d expect to get this fully enabled again would be a clean-up of the Kconfig options, so it is easy and (from the documentation) predictable for boards to turn this off.

I still want this to be turned back on in the current cycle.

Great maintainer!!!!

Quoted 8 lines

You objection for other patch is that it break other boards(which is non exist actually),
and you merge this patch(break more than 10 boards very clearly) with objection comment.
And then you require for all the Kconfig options update before the feature back.

WoW!!!

How do we resolve this? I have not had time to dig into this myself, and there seem to be big problems with doing this 'properly'.

Should we bring back the Kconfig for this release? I may be able to look at it for the next release, although time is not my friend at present.

Regards,
Simon

#23 lore raw

PATCH 2/2 rockchip: Drop call to rockchip_dnl_mode_check() for now

Simon,

Quoted 25 lines · Simon Glass

Hi,

Quoted 20 lines · On Tue, 12 Mar 2019 at 09:26, Kever Yang <kever.yang at rock-chips.com…
Quoted 6 lines · Philipp Tomsich

Note that the minimum improvement that I’d expect to get this fully enabled again would be a clean-up of the Kconfig options, so it is easy and (from the documentation) predictable for boards to turn this off.

I still want this to be turned back on in the current cycle.

Great maintainer!!!!

You objection for other patch is that it break other boards(which is non exist actually),
and you merge this patch(break more than 10 boards very clearly) with objection comment.
And then you require for all the Kconfig options update before the feature back.

Quoted 8 lines
Quoted 1 line

WoW!!!

How do we resolve this? I have not had time to dig into this myself, and there seem to be big problems with doing this 'properly'.

Should we bring back the Kconfig for this release? I may be able to look at it for the next release, although time is not my friend at present.

I may have the time to update the Kconfig this week. If not, I’ll revert the deprecation-patch for this release.

Philipp.

#24 lore raw

PATCH 2/2 rockchip: Drop call to rockchip_dnl_mode_check() for now【请注意,邮件由[email protected]代发】【请注意,邮件由[email protected]代发】

In reply to Simon Glass

Hi Simon:

Quoted 64 lines · Simon Glass

Hi Andy,

Quoted 55 lines · Andy Yan

Hi Simon:

Quoted 52 lines · Simon Glass

HI Andy,

Quoted 32 lines · Andy Yan

Hi Philipp:

Sorry for the late reply, we just come back from the Chinese Spring
Festival.

Quoted 10 lines · Philipp Tomsich

Kever,

Independent of whether we revert this for the current cycle (and also independent of
if I ever find the other patch you had been referring to — I couldn’t find it in my local
mailing list archive) and then deprecate it for the next release (unless converted to
DM), we still have a number of architectural issues that need to be addressed:

I still doubt is this a right work-flow for patch apply. Before we apply a patch which will break many other boards , should we make sure there is a solution patch applied for these boars first?

Quoted 11 lines

1.This really should be a driver under DTS control.
2.We need to not get away from configuring SOM-specific addresses via Kconfig

Both these issues are technical debt that we’ve accumulated over the last 18 months
and need to address for the sake of future maintainability. E.g. ‘setting an address to 0x0 via Kconfig to disable a driver/feature’ really isn’t in line
with the architectural direction of U-Boot.

Quoted 3 lines

For technical side, I think CONFIG_ROCKCHIP_BOOT_MODE_REG is necessary here, we will read this register from save_boot_params when we get out from bootrom, the dtb is not available at this point.

Yes this is happening very early, but I wonder why this is necessary? Can it be moved to later?

The call to check_back_to_brom_dnl_flag() happens much later so there should be no problem to convert that.

Quoted 8 lines

On the other hand, almost rockchip based products use a recovery key to enter download(upgrade)mode, this is a muti-funtion key, most of them reuse with vol+- key, we would like the u-boot share

dtb with linux kernel. To keep the linux kernel dts as clean as possible ,we don't want to add another dts property to describe this key either. This is why I implement function rockchip_dnl_key_pressed as __weak.

Quoted 6 lines
Quoted 2 lines

OK, but given that it is called later, it seems to me that it could use a driver.

We can't let it called later. See the discuss here :
http://patchwork.ozlabs.org/patch/812228/

Quoted 7 lines

Yes I read that, and took a look myself, and certainly it does not look very easy and I take your point about potential bugs being introduced.

It seems like you have tried quite hard, so I'm not going to object if you can't find a way. My main objection was that it broke other boards.

I took the code many times,  what this code done is : call adc_channel_single_shot to check if a download key is pressed, so it is very strange it broke the display on minnie. Would you please show the boot log when the display broke.

Quoted 4 lines

Is there any way the check could be delayed enough to actually be able to read the device tree? It could call spl_early_init() quite early and then do that.

Actually,  the interrupt overwrite happens in start.s,  so this don't do help.

Quoted 6 lines

Regards,
Simon