Skip to contentSkip to conversation

Restore imx8mn_bsh_smm_s2 properly booting

RFC PATCH Superseded Dario Binacchi 5 participants 17 messages May 15, 2025 – May 17, 2025 Arm i.MX and Layerscape

v1 of 3 Superseded Latest is v3 →

lore
Share & tools
lore
#0 lore raw

cover Restore imx8mn_bsh_smm_s2 properly booting

Commits [1] and [2] broke the booting of the BSH SMM S2 board. The patches in this series restore proper booting.

[1] 2a00d73d081a ("spl: mmc: Try to clean up raw-mode options")
[2] dda454e933c6 ("serial: mxc: Support bulk enabling clocks")

Dario Binacchi (3):
spl: Kconfig: allow to enable SYS_MMCSD_RAW_MODE_U_BOOT_USE_SECTOR for

    i.MX8M

configs: imx8mn_bsh_smm_s2: restore
CONFIG_SYS_MMCSD_RAW_MODE_U_BOOT_SECTOR
serial: mxc: restore booting for imx8mn_bsh_smm_s2

 common/spl/Kconfig                  |  2 +-
 configs/imx8mn_bsh_smm_s2_defconfig |  1 +
 drivers/serial/serial_mxc.c         | 23 ++++++++++++++++-------
 3 files changed, 18 insertions(+), 8 deletions(-)
Signature · 4 lines
2.43.0

base-commit: 1b5e435102aa29a665119430196cb366ce36a01b
branch: bsh-s2-break
#1 lore raw

RFC PATCH 1/3 spl: Kconfig: allow to enable SYS_MMCSD_RAW_MODE_U_BOOT_USE_SECTOR for i.MX8M

Commit 2a00d73d081a ("spl: mmc: Try to clean up raw-mode options") breaks the boot of the BSH SMM S2 board. As stated in the commit itself, "Some boards use this value even though MMC is not enabled in SPL, for example imx8mn_bsh_smm_s2". The same commit makes SPL_SYS_MMCSD_RAW_MODE depend on SPL_DM_MMC || SPL_MMC. With SPL_SYS_MMCSD_RAW_MODE unset, it is not possible to enable SYS_MMCSD_RAW_MODE_U_BOOT_USE_SECTOR, thus breaking the board's boot process.

The patch once again allows SYS_MMCSD_RAW_MODE_U_BOOT_USE_SECTOR to be selected also for boards that do not use MMC, in this case on the i.MX8M platform.

Fixes 2a00d73d081a ("spl: mmc: Try to clean up raw-mode options")

---

 common/spl/Kconfig | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
common/spl/Kconfig +1 -1
index aa3a85eea54d..77bb21e74cd6 100644
@@ -542,7 +542,7 @@ config SPL_SYS_MMCSD_RAW_MODE
 	help
 	  Support booting from an MMC without a filesystem.
 
-if SPL_SYS_MMCSD_RAW_MODE
+if SPL_SYS_MMCSD_RAW_MODE || ARCH_IMX8M
 
 choice
 	prompt "Method for locating next phase of boot (e.g. U-Boot)"
Signature · 1 line
2.43.0
#2 lore raw

Re: [RFC PATCH 1/3] spl: Kconfig: allow to enable SYS_MMCSD_RAW_MODE_U_BOOT_USE_SECTOR for i.MX8M

Hi Dario,

Quoted 8 lines · Dario Binacchi

Commit 2a00d73d081a ("spl: mmc: Try to clean up raw-mode options") breaks the boot of the BSH SMM S2 board. As stated in the commit itself, "Some boards use this value even though MMC is not enabled in SPL, for example imx8mn_bsh_smm_s2". The same commit makes SPL_SYS_MMCSD_RAW_MODE depend on SPL_DM_MMC || SPL_MMC. With SPL_SYS_MMCSD_RAW_MODE unset, it is not possible to enable SYS_MMCSD_RAW_MODE_U_BOOT_USE_SECTOR, thus breaking the board's boot process.

Quoted 4 lines

The patch once again allows SYS_MMCSD_RAW_MODE_U_BOOT_USE_SECTOR to be selected also for boards that do not use MMC, in this case on the i.MX8M platform.

Why would you use SYS_MMCSD_RAW_MODE_U_BOOT_USE_SECTOR if you don't have any MMC? Shouldn't we rather fix the incorrect usage of this variable to be something else (or better guarded?).

Cheers,
Quentin

#3 lore raw

Re: [RFC PATCH 1/3] spl: Kconfig: allow to enable SYS_MMCSD_RAW_MODE_U_BOOT_USE_SECTOR for i.MX8M

Quoted 11 lines · Quentin Schulz

Hi Dario,

Quoted 8 lines · Dario Binacchi

Commit 2a00d73d081a ("spl: mmc: Try to clean up raw-mode options") breaks the boot of the BSH SMM S2 board. As stated in the commit itself, "Some boards use this value even though MMC is not enabled in SPL, for example imx8mn_bsh_smm_s2". The same commit makes SPL_SYS_MMCSD_RAW_MODE depend on SPL_DM_MMC || SPL_MMC. With SPL_SYS_MMCSD_RAW_MODE unset, it is not possible to enable SYS_MMCSD_RAW_MODE_U_BOOT_USE_SECTOR, thus breaking the board's boot process.

Quoted 8 lines
Quoted 4 lines

The patch once again allows SYS_MMCSD_RAW_MODE_U_BOOT_USE_SECTOR to be selected also for boards that do not use MMC, in this case on the i.MX8M platform.

Why would you use SYS_MMCSD_RAW_MODE_U_BOOT_USE_SECTOR if you don't have any MMC? Shouldn't we rather fix the incorrect usage of this variable to be something else (or better guarded?).

It appears that arch/arm/mach-imx/spl_imx_romapi.c::spl_romapi_get_uboot_base() will use the value so that we (apparently) go ROM->U-Boot, or at least ROM->some series of prior stage loaders->ROM->U-Boot. I'm not sure I love overloading the value like this, but I can't say it's wrong. Perhaps the help needs to be updated too?

Signature · 1 line
Tom
#4 lore raw

Re: [RFC PATCH 1/3] spl: Kconfig: allow to enable SYS_MMCSD_RAW_MODE_U_BOOT_USE_SECTOR for i.MX8M

Quoted 21 lines · Tom Rini
Quoted 11 lines · Quentin Schulz

Hi Dario,

Quoted 8 lines · Dario Binacchi

Commit 2a00d73d081a ("spl: mmc: Try to clean up raw-mode options") breaks the boot of the BSH SMM S2 board. As stated in the commit itself, "Some boards use this value even though MMC is not enabled in SPL, for example imx8mn_bsh_smm_s2". The same commit makes SPL_SYS_MMCSD_RAW_MODE depend on SPL_DM_MMC || SPL_MMC. With SPL_SYS_MMCSD_RAW_MODE unset, it is not possible to enable SYS_MMCSD_RAW_MODE_U_BOOT_USE_SECTOR, thus breaking the board's boot process.

Quoted 8 lines
Quoted 4 lines

The patch once again allows SYS_MMCSD_RAW_MODE_U_BOOT_USE_SECTOR to be selected also for boards that do not use MMC, in this case on the i.MX8M platform.

Why would you use SYS_MMCSD_RAW_MODE_U_BOOT_USE_SECTOR if you don't have any MMC? Shouldn't we rather fix the incorrect usage of this variable to be something else (or better guarded?).

Quoted 6 lines

It appears that arch/arm/mach-imx/spl_imx_romapi.c::spl_romapi_get_uboot_base() will use the value so that we (apparently) go ROM->U-Boot, or at least ROM->some series of prior stage loaders->ROM->U-Boot. I'm not sure I love overloading the value like this, but I can't say it's wrong. Perhaps the help needs to be updated too?

Isn't it rather U-Boot SPL asking the BootROM to load the image for it instead of doing it by itself? Some MMC is technically present then, just that because the BootROM handles it and not U-Boot, we need to specify the offset in the MMC even though we only return that value to the BootROM.

If that is correct, then I'm not sure we want to use SYS_MMCSD_RAW_MODE_U_BOOT_SECTOR considering it selects SPL_LOAD_BLOCK which I would assume to be useless if the BootROM is doing all the work for the SPL?

Cheers,
Quentin

#5 lore raw

RFC PATCH 2/3 configs: imx8mn_bsh_smm_s2: restore CONFIG_SYS_MMCSD_RAW_MODE_U_BOOT_SECTOR

In reply to Dario Binacchi

The commit d892702080d4 ("configs: Resync with savedefconfig") drop the CONFIG_SYS_MMCSD_RAW_MODE_U_BOOT_SECTOR=0x300 option due to the previous commit 2a00d73d081a ("spl: mmc: Try to clean up raw-mode options").

Restore the option, which is necessary for the board to boot correctly.

Fixes 2a00d73d081a ("spl: mmc: Try to clean up raw-mode options")

---

 configs/imx8mn_bsh_smm_s2_defconfig | 1 +
 1 file changed, 1 insertion(+)
configs/imx8mn_bsh_smm_s2_defconfig +1 -0
index 9ffcd3b221c3..61aa0c6c0607 100644
@@ -43,6 +43,7 @@ CONFIG_SPL_SYS_MALLOC=y
 CONFIG_SPL_HAS_CUSTOM_MALLOC_START=y
 CONFIG_SPL_CUSTOM_SYS_MALLOC_ADDR=0x42200000
 CONFIG_SPL_SYS_MALLOC_SIZE=0x80000
+CONFIG_SYS_MMCSD_RAW_MODE_U_BOOT_SECTOR=0x300
 CONFIG_SPL_DMA=y
 CONFIG_SPL_I2C=y
 CONFIG_SPL_MTD=y
Signature · 1 line
2.43.0
#6 lore raw

RFC PATCH 3/3 serial: mxc: restore booting for imx8mn_bsh_smm_s2

In reply to Dario Binacchi

The commit dda454e933c6 ("serial: mxc: Support bulk enabling clocks") breaks the booting of the BSH SMM S2 board.
Restore proper booting of the board even in case of failure of either clk_get_bulk() or clk_enable_bulk().

Fixes dda454e933c6 ("serial: mxc: Support bulk enabling clocks")

---

 drivers/serial/serial_mxc.c | 23 ++++++++++++++++-------
 1 file changed, 16 insertions(+), 7 deletions(-)
drivers/serial/serial_mxc.c +16 -7
index 28f4435d01dd..b09a5fe0f7a4 100644
@@ -310,22 +310,31 @@ int mxc_serial_setbrg(struct udevice *dev, int baudrate)
 	return 0;
 }
 
-static int mxc_serial_probe(struct udevice *dev)
+#if CONFIG_IS_ENABLED(CLK_CCF)
+static void mxc_serial_enable_clk_bulk(struct udevice *dev)
 {
 	struct mxc_serial_plat *plat = dev_get_plat(dev);
-#if CONFIG_IS_ENABLED(CLK_CCF)
 	int ret;
 
 	ret = clk_get_bulk(dev, &plat->clks);
 	if (ret)
-		return ret;
+		return;
 
-	ret = clk_enable_bulk(&plat->clks);
-	if (ret)
-		return ret;
+	clk_enable_bulk(&plat->clks);
+}
+#else
+static mxc_serial_enable_clk_bulk(struct udevice *dev)
+{
+}
 #endif
-	_mxc_serial_init(plat->reg, plat->use_dte);
 
+static int mxc_serial_probe(struct udevice *dev)
+{
+	struct mxc_serial_plat *plat = dev_get_plat(dev);
+
+	mxc_serial_enable_clk_bulk(dev);
+
+	_mxc_serial_init(plat->reg, plat->use_dte);
 	return 0;
 }
 
Signature · 1 line
2.43.0
reviewer
#7 lore raw

Re: [RFC PATCH 3/3] serial: mxc: restore booting for imx8mn_bsh_smm_s2

Quoted 5 lines · Dario Binacchi

The commit dda454e933c6 ("serial: mxc: Support bulk enabling clocks") breaks the booting of the BSH SMM S2 board.
Restore proper booting of the board even in case of failure of either clk_get_bulk() or clk_enable_bulk().

We need a better understanding of why the boot fails.

Does the failure come from clk_get_bulk() or clk_enable_bulk()?

Please investigate.

#8 lore raw

Re: [RFC PATCH 3/3] serial: mxc: restore booting for imx8mn_bsh_smm_s2

Hi Fabio

Quoted 11 lines · Fabio Estevam
Quoted 5 lines · Dario Binacchi

The commit dda454e933c6 ("serial: mxc: Support bulk enabling clocks") breaks the booting of the BSH SMM S2 board.
Restore proper booting of the board even in case of failure of either clk_get_bulk() or clk_enable_bulk().

We need a better understanding of why the boot fails.

The patch keeps the same functionality for the working board and fixes the no booting one. Agree
that we need to understand but even we need working boards

Quoted 2 lines

Does the failure come from clk_get_bulk() or clk_enable_bulk()?

The serial is not up at that time so we need to buffer the error and print later. Right now
we are working on other fixes

Michael

Quoted 3 lines

Please investigate.

Signature · 11 lines
Michael Nazzareno Trimarchi
Co-Founder & Chief Executive Officer
M. +39 347 913 2170
[email protected]
__________________________________

Amarula Solutions BV
Joop Geesinkweg 125, 1114 AB, Amsterdam, NL
T. +31 (0)85 111 9172
[email protected]
www.amarulasolutions.com
reviewer
#9 lore raw

Re: [RFC PATCH 3/3] serial: mxc: restore booting for imx8mn_bsh_smm_s2

Hi Michael,

Quoted 2 lines · Michael Nazzareno Trimarchi

The serial is not up at that time so we need to buffer the error and print later. Right now we are working on other fixes

Please share more details about the boot problem on the imx8mn_bsh_smm_s2 board.

Does a hang occur at SPL? Do you get anything in the console?

The mx8mn_bsh_smm_s2 board uses UART4 as the console.

The other i.MX8MN boards where we tested this patch used UART2.

UART4 does not have U-Boot/Linux permission by default.

TF-A enables the other UART port permission access at:
https://github.com/ARM-software/arm-trusted-firmware/blob/v2.12/plat/imx/imx8m/imx8mn/imx8mn_bl31_setup.c#L124-L143

Is the UART clock access falling in your case?

Is clk_enable_bulk() the one that fails?

#10 lore raw

Re: [RFC PATCH 3/3] serial: mxc: restore booting for imx8mn_bsh_smm_s2

Hi Fabio

Quoted 9 lines · Fabio Estevam

Hi Michael,

Quoted 2 lines · Michael Nazzareno Trimarchi

The serial is not up at that time so we need to buffer the error and print later. Right now we are working on other fixes

Quoted 8 lines

Please share more details about the boot problem on the imx8mn_bsh_smm_s2 board.

Does a hang occur at SPL? Do you get anything in the console?

The mx8mn_bsh_smm_s2 board uses UART4 as the console.

The other i.MX8MN boards where we tested this patch used UART2.

Can you please point me to an example of a tested board?

Quoted 2 lines

UART4 does not have U-Boot/Linux permission by default.

TF-A is not in the game, the board is working fine.

Quoted 2 lines

TF-A enables the other UART port permission access at:
https://github.com/ARM-software/arm-trusted-firmware/blob/v2.12/plat/imx/imx8m/imx8mn/imx8mn_bl31_setup.c#L124-L143

I think we have upstream the fix there. Anyway...

Quoted 3 lines

Is the UART clock access falling in your case?

If the uart is failing you don't have the console

Quoted 1 line

Is clk_enable_bulk() the one that fails?

I think that something connected to -u-boot.dtsi if you have tested with some imx8mn

Michael

Signature · 11 lines
Michael Nazzareno Trimarchi
Co-Founder & Chief Executive Officer
M. +39 347 913 2170
[email protected]
__________________________________

Amarula Solutions BV
Joop Geesinkweg 125, 1114 AB, Amsterdam, NL
T. +31 (0)85 111 9172
[email protected]
www.amarulasolutions.com
reviewer
#11 lore raw

Re: [RFC PATCH 3/3] serial: mxc: restore booting for imx8mn_bsh_smm_s2

Quoted 1 line · Michael Nazzareno Trimarchi

Can you please point me to an example of a tested board?

I have just tested the top-of-tree U-Boot on an imx8mn evk board:

U-Boot SPL 2025.07-rc2-00018-g126a88d49bca (May 17 2025 - 17:21:14 -0300)

WDT:   Started watchdog@30280000 with servicing every 1000ms (60s timeout)
SEC0:  RNG instantiated

Normal Boot
Trying to boot from BOOTROM
Boot Stage: Primary boot
image offset 0x8000, pagesize 0x200, ivt offset 0x0

NOTICE:  Do not release JR0 to NS as it can be used by HAB
NOTICE:  BL31: v2.12.0(release):v2.12.0
NOTICE:  BL31: Built : 17:16:49, May 17 2025

U-Boot 2025.07-rc2-00018-g126a88d49bca (May 17 2025 - 17:21:14 -0300)

CPU:   NXP i.MX8MNano Quad Rev1.0 A53 at 1200 MHz
CPU:   Consumer temperature grade (0C to 95C) at 32C
Model: NXP i.MX8MNano DDR4 EVK board
DRAM:  2 GiB
Core:  180 devices, 23 uclasses, devicetree: separate
WDT:   Started watchdog@30280000 with servicing every 1000ms (60s timeout)
MMC:   FSL_SDHC: 1, FSL_SDHC: 2
Loading Environment from MMC... Reading from MMC(1)... OK
In:    serial@30890000
Out:   serial@30890000
Err:   serial@30890000
SEC0:  RNG instantiated
Net:   eth0: ethernet@30be0000
Hit any key to stop autoboot:  0

u-boot=>

Make sure you are using the most recent U-Boot. We had some imx8m clock drive hickups after 2025.04.

#12 lore raw

Re: [RFC PATCH 3/3] serial: mxc: restore booting for imx8mn_bsh_smm_s2

Hi

Quoted 11 lines · Fabio Estevam
Quoted 1 line · Michael Nazzareno Trimarchi

Can you please point me to an example of a tested board?

I have just tested the top-of-tree U-Boot on an imx8mn evk board:

U-Boot SPL 2025.07-rc2-00018-g126a88d49bca (May 17 2025 - 17:21:14 -0300)

WDT:   Started watchdog@30280000 with servicing every 1000ms (60s timeout)
SEC0:  RNG instantiated
Quoted 8 lines

Normal Boot
Trying to boot from BOOTROM
Boot Stage: Primary boot
image offset 0x8000, pagesize 0x200, ivt offset 0x0

NOTICE:  Do not release JR0 to NS as it can be used by HAB
NOTICE:  BL31: v2.12.0(release):v2.12.0
NOTICE:  BL31: Built : 17:16:49, May 17 2025

I was on master except this commit

commit 128d997a8772cc174f38d529d8b25f90b3aa8ad8

Author: Jonas Karlman <[email protected]>
Date:   Sat May 10 15:32:01 2025 +0000

clk: Fix clk_set_parent() regression

The commit ac30d90f3367 ("clk: Ensure the parent clocks are enabled while reparenting") add a call to clk_enable() for the parent clock.

For clock drivers that do not implement the enable() ops, like most Rockchip clock drivers, this now cause the set_parent() ops to never be called when CLK_CCF=n (default for Rockchip).

I can not try today but if you can, is this the commit that could break my boot?

Michael

Quoted 21 lines

U-Boot 2025.07-rc2-00018-g126a88d49bca (May 17 2025 - 17:21:14 -0300)

CPU:   NXP i.MX8MNano Quad Rev1.0 A53 at 1200 MHz
CPU:   Consumer temperature grade (0C to 95C) at 32C
Model: NXP i.MX8MNano DDR4 EVK board
DRAM:  2 GiB
Core:  180 devices, 23 uclasses, devicetree: separate
WDT:   Started watchdog@30280000 with servicing every 1000ms (60s timeout)
MMC:   FSL_SDHC: 1, FSL_SDHC: 2
Loading Environment from MMC... Reading from MMC(1)... OK
In:    serial@30890000
Out:   serial@30890000
Err:   serial@30890000
SEC0:  RNG instantiated
Net:   eth0: ethernet@30be0000
Hit any key to stop autoboot:  0

u-boot=>

Make sure you are using the most recent U-Boot. We had some imx8m clock drive hickups after 2025.04.

Signature · 11 lines
Michael Nazzareno Trimarchi
Co-Founder & Chief Executive Officer
M. +39 347 913 2170
[email protected]
__________________________________

Amarula Solutions BV
Joop Geesinkweg 125, 1114 AB, Amsterdam, NL
T. +31 (0)85 111 9172
[email protected]
www.amarulasolutions.com
#13 lore raw

Re: [RFC PATCH 3/3] serial: mxc: restore booting for imx8mn_bsh_smm_s2

Hi Fabio

Quoted 30 lines · Michael Nazzareno Trimarchi

Hi

Quoted 11 lines · Fabio Estevam
Quoted 1 line · Michael Nazzareno Trimarchi

Can you please point me to an example of a tested board?

I have just tested the top-of-tree U-Boot on an imx8mn evk board:

U-Boot SPL 2025.07-rc2-00018-g126a88d49bca (May 17 2025 - 17:21:14 -0300)

WDT:   Started watchdog@30280000 with servicing every 1000ms (60s timeout)
SEC0:  RNG instantiated
Quoted 8 lines

Normal Boot
Trying to boot from BOOTROM
Boot Stage: Primary boot
image offset 0x8000, pagesize 0x200, ivt offset 0x0

NOTICE:  Do not release JR0 to NS as it can be used by HAB
NOTICE:  BL31: v2.12.0(release):v2.12.0
NOTICE:  BL31: Built : 17:16:49, May 17 2025

I was on master except this commit

commit 128d997a8772cc174f38d529d8b25f90b3aa8ad8

Author: Jonas Karlman <[email protected]>
Date:   Sat May 10 15:32:01 2025 +0000
Quoted 10 lines

clk: Fix clk_set_parent() regression

The commit ac30d90f3367 ("clk: Ensure the parent clocks are enabled while reparenting") add a call to clk_enable() for the parent clock.

For clock drivers that do not implement the enable() ops, like most Rockchip clock drivers, this now cause the set_parent() ops to never be called when CLK_CCF=n (default for Rockchip).

BTW is really needed now?

git grep init_uart_clk board/freescale/imx8mn_evk/
board/freescale/imx8mn_evk/spl.c: init_uart_clk(1);

Michael

Quoted 39 lines

I can not try today but if you can, is this the commit that could break my boot?

Michael

Quoted 17 lines

U-Boot 2025.07-rc2-00018-g126a88d49bca (May 17 2025 - 17:21:14 -0300)

CPU:   NXP i.MX8MNano Quad Rev1.0 A53 at 1200 MHz
CPU:   Consumer temperature grade (0C to 95C) at 32C
Model: NXP i.MX8MNano DDR4 EVK board
DRAM:  2 GiB
Core:  180 devices, 23 uclasses, devicetree: separate
WDT:   Started watchdog@30280000 with servicing every 1000ms (60s timeout)
MMC:   FSL_SDHC: 1, FSL_SDHC: 2
Loading Environment from MMC... Reading from MMC(1)... OK
In:    serial@30890000
Out:   serial@30890000
Err:   serial@30890000
SEC0:  RNG instantiated
Net:   eth0: ethernet@30be0000
Hit any key to stop autoboot:  0
Quoted 4 lines

u-boot=>

Make sure you are using the most recent U-Boot. We had some imx8m clock drive hickups after 2025.04.

--
Michael Nazzareno Trimarchi
Co-Founder & Chief Executive Officer
M. +39 347 913 2170
[email protected]
__________________________________

Amarula Solutions BV
Joop Geesinkweg 125, 1114 AB, Amsterdam, NL
T. +31 (0)85 111 9172
[email protected]
www.amarulasolutions.com

Signature · 11 lines
Michael Nazzareno Trimarchi
Co-Founder & Chief Executive Officer
M. +39 347 913 2170
[email protected]
__________________________________

Amarula Solutions BV
Joop Geesinkweg 125, 1114 AB, Amsterdam, NL
T. +31 (0)85 111 9172
[email protected]
www.amarulasolutions.com
reviewer
#14 lore raw

Re: [RFC PATCH 3/3] serial: mxc: restore booting for imx8mn_bsh_smm_s2

Quoted 6 lines · Michael Nazzareno Trimarchi
Quoted 6 lines

I was on master except this commit

commit 128d997a8772cc174f38d529d8b25f90b3aa8ad8

Author: Jonas Karlman <[email protected]>
Date:   Sat May 10 15:32:01 2025 +0000
Quoted 8 lines
Quoted 8 lines

clk: Fix clk_set_parent() regression

The commit ac30d90f3367 ("clk: Ensure the parent clocks are enabled while reparenting") add a call to clk_enable() for the parent clock.

For clock drivers that do not implement the enable() ops, like most Rockchip clock drivers, this now cause the set_parent() ops to never be called when CLK_CCF=n (default for Rockchip).

CLK_CCF=y in your case, so this does not help for you.

Quoted 5 lines

BTW is really needed now?

git grep init_uart_clk board/freescale/imx8mn_evk/
board/freescale/imx8mn_evk/spl.c: init_uart_clk(1);

This can be removed. I tested without this line, and it boots fine.

#15 lore raw

Re: [RFC PATCH 3/3] serial: mxc: restore booting for imx8mn_bsh_smm_s2

Hi

Quoted 19 lines · Fabio Estevam
Quoted 6 lines · Michael Nazzareno Trimarchi
Quoted 6 lines

I was on master except this commit

commit 128d997a8772cc174f38d529d8b25f90b3aa8ad8

Author: Jonas Karlman <[email protected]>
Date:   Sat May 10 15:32:01 2025 +0000
Quoted 8 lines
Quoted 8 lines

clk: Fix clk_set_parent() regression

The commit ac30d90f3367 ("clk: Ensure the parent clocks are enabled while reparenting") add a call to clk_enable() for the parent clock.

For clock drivers that do not implement the enable() ops, like most Rockchip clock drivers, this now cause the set_parent() ops to never be called when CLK_CCF=n (default for Rockchip).

Quoted 8 lines

CLK_CCF=y in your case, so this does not help for you.

Quoted 5 lines

BTW is really needed now?

git grep init_uart_clk board/freescale/imx8mn_evk/
board/freescale/imx8mn_evk/spl.c: init_uart_clk(1);

This can be removed. I tested without this line, and it boots fine.

Are you sending a patch? or I will send it

Michael

Signature · 11 lines
Michael Nazzareno Trimarchi
Co-Founder & Chief Executive Officer
M. +39 347 913 2170
[email protected]
__________________________________

Amarula Solutions BV
Joop Geesinkweg 125, 1114 AB, Amsterdam, NL
T. +31 (0)85 111 9172
[email protected]
www.amarulasolutions.com
reviewer
#16 lore raw

Re: [RFC PATCH 3/3] serial: mxc: restore booting for imx8mn_bsh_smm_s2

Quoted 1 line · Michael Nazzareno Trimarchi

Are you sending a patch? or I will send it

Feel free to send it when you have a chance. Thanks!