Hi,

Am Fr., 10. Juni 2022 um 15:00 Uhr schrieb Fabio Estevam festevam@gmail.com:

Hi Heiko,

On Fri, Jun 10, 2022 at 7:29 AM Heiko Thiery heiko.thiery@gmail.com wrote:

...

Hi,

I just rebased my pending patch for adding support for the NXP imx8mn ddr3l evk board. When applying on the current master I see some garbage on the serial input.

Can anyone confirm that behavior or is it just here?

Just tried booting on a imx8mn ddr4 evk and such a problem does not happen:

U-Boot SPL 2022.07-rc4-00007-g57bd363de7 (Jun 10 2022 - 09:42:49 -0300) SEC0: RNG instantiated Normal Boot WDT: Started watchdog@30280000 with servicing (60s timeout) Trying to boot from BOOTROM image offset 0x8000, pagesize 0x200, ivt offset 0x0 NOTICE: BL31: v2.7(release):v2.7.0 NOTICE: BL31: Built : 09:42:28, Jun 10 2022

U-Boot 2022.07-rc4-00007-g57bd363de7 (Jun 10 2022 - 09:42:49 -0300)

CPU: Freescale i.MX8MNano Quad rev1.0 at 1200 MHz Reset cause: POR Model: NXP i.MX8MNano DDR4 EVK board DRAM: 2 GiB Core: 139 devices, 22 uclasses, devicetree: separate WDT: Started watchdog@30280000 with servicing (60s timeout) MMC: FSL_SDHC: 1, FSL_SDHC: 2 Loading Environment from MMC... OK In: serial@30890000 Out: serial@30890000 Err: serial@30890000 Net: eth0: ethernet@30be0000 Hit any key to stop autoboot: 0

Also tried booting on a imx8mm evk and did not see the serial garbage either.

Not sure why this happens on the imx8mn ddr3 evk and on the kontron_sl_imx8mm though.

Yes, this is strange.

None of the boards I have physical access show the problem. The Kontron board have remote access only, so not very easy to run git bisect.

Are you able to bisect it with the kontron_sl_imx8mm board?

Unfortunately only next week again. But I will try on my imx8mq later.

-- Heiko

Hi Heiko,

On Sat, Jun 11, 2022 at 3:12 AM Heiko Thiery heiko.thiery@gmail.com wrote:

It took me some effort to get the current u-boot running on my imx8mq. But finally it works and I do not see this garbage on the imx8mq. Next week I will check on the kontron_sl_imx8mm.

Thanks, appreciated it!

Hi Heiko,

On Mon, Jun 13, 2022 at 10:54 AM Heiko Thiery heiko.thiery@gmail.com wrote:

This is the output on the kontron bl-imx8m board:

U-Boot SPL 2022.07-rc4-00007-g8c3f838be2 (Jun 13 2022 - 15:36:08 +0200) Kontron SL i.MX8MM (N801X) module, 1 GB RAM detected Touch controller detected, assuming LVDS panel... Normal Boot WDT: Not starting watchdog@30280000 Trying to boot from MMC2 NOTICE: BL31: v2.4(release):v2.4

Just tried with the same v2.4 TF-A version:

U-Boot SPL 2022.07-rc4-00007-g57bd363de7 (Jun 13 2022 - 11:18:10 -0300) Kontron SL i.MX8MM (N801X) module, 4 GB RAM detected Normal Boot WDT: Not starting watchdog@30280000 Trying to boot from MMC1 NOTICE: BL31: v2.4(release):v2.4 NOTICE: BL31: Built : 11:17:34, Jun 13 2022

U-Boot 2022.07-rc4-00007-g57bd363de7 (Jun 13 2022 - 11:18:10 -0300)

CPU: Freescale i.MX8MMQ rev1.0 1600 MHz (running at 1200 MHz) CPU: Industrial temperature grade (-40C to 105C) at 60C Reset cause: POR Model: Kontron i.MX8MM N801X S DRAM: 4 GiB Core: 175 devices, 25 uclasses, devicetree: separate WDT: Not starting watchdog@30280000 MMC: FSL_SDHC: 0, FSL_SDHC: 1 Loading Environment from SPIFlash... SF: Detected mx25r1635f with page size 256 Bytes, erase size 64 KiB, total 2 MiB OK In: serial Out: serial Err: serial Net: Could not get PHY for FEC0: addr 0 No ethernet found.

Hit any key to stop autoboot: 0 => [55;203R Unknown command '[55' - try 'help' Unknown command '203R' - try 'help'

and the serial garbage is seen.

My version has 4GiB of RAM, yours has only 1GiB.

Could this make a difference?

Thanks

On Mon, Jun 13, 2022 at 11:27 AM Fabio Estevam festevam@gmail.com wrote:

U-Boot 2022.07-rc4-00007-g57bd363de7 (Jun 13 2022 - 11:18:10 -0300)

CPU: Freescale i.MX8MMQ rev1.0 1600 MHz (running at 1200 MHz) CPU: Industrial temperature grade (-40C to 105C) at 60C Reset cause: POR Model: Kontron i.MX8MM N801X S DRAM: 4 GiB Core: 175 devices, 25 uclasses, devicetree: separate WDT: Not starting watchdog@30280000 MMC: FSL_SDHC: 0, FSL_SDHC: 1 Loading Environment from SPIFlash... SF: Detected mx25r1635f with page size 256 Bytes, erase size 64 KiB, total 2 MiB OK In: serial Out: serial Err: serial Net: Could not get PHY for FEC0: addr 0 No ethernet found.

Hit any key to stop autoboot: 0 => [55;203R Unknown command '[55' - try 'help' Unknown command '203R' - try 'help'

and the serial garbage is seen.

I managed to bisect it, and the serial console garbage is caused by:

commit a9bf024b2933bba0e23038892970a18b72dfaeb4 Author: AKASHI Takahiro takahiro.akashi@linaro.org Date: Tue Apr 19 10:05:12 2022 +0900

efi_loader: disk: a helper function to create efi_disk objects from udevice

Add efi_disk_probe() function. This function creates an efi_disk object for a raw disk device (UCLASS_BLK) and additional objects for related partitions (UCLASS_PARTITION).

So this function is expected to be called through driver model's "probe" interface every time one raw disk device is detected and activated. We assume that partition devices (UCLASS_PARTITION) have been created when this function is invoked.

Signed-off-by: AKASHI Takahiro takahiro.akashi@linaro.org

Akashi-san, any ideas?

Hi Fabio,

Am Mo., 13. Juni 2022 um 19:01 Uhr schrieb Fabio Estevam festevam@gmail.com:

Hi Heiko,

On Fri, Jun 10, 2022 at 7:29 AM Heiko Thiery heiko.thiery@gmail.com wrote:

...

Hit any key to stop autoboot: 0 u-boot=> [25;88R Unknown command '[25' - try 'help' Unknown command '88R' - try 'help'

git bisect points me to an efi loader commit as the guilty one.

Can you please pass the line below into your imx8mn ddr3l evk board defconfig?

# CONFIG_EFI_LOADER is not set

Do you see serial console garbage in this case?

Just wanted to confirm if the error we are seeing is due to the efi loader breakage.

I can confirm that by disabling EFI loader the error is gone.

-- Heiko