[meta-freescale] linux-congatec (fido) silent after Starting kernel ...

[Nikolay Dimitrov]

Hi Leon,

On 05/21/2015 05:20 PM, Leon Woestenberg wrote:
> I could start the built u-boot from (u)SD card using the
> not-really-public registers.
>
> # setting for BOOT_CFG1, this select USDHC2 (Congatec on-module uSD)
> mw.l 0x020d8040 0x2850
> mw.l 0x020d8044 0x10000000
> reset
>
> # setting for BOOT_CFG1, this select USDHC4 (Congatec carrier board SD slot)
> mw.l 0x020d8040 0x3850
> mw.l 0x020d8044 0x10000000
> reset
>
> As 'bmode' was not available on my pre-installed u-boot, I used this
> convenience command to boot u-boot from the SD card:
>
> setenv sd 'mw.l 0x020d8040 0x3850; mw.l 0x020d8044 0x10000000; reset'
>
>
> However, although the newer u-boot 2013.04 supports the DTB now, I'm
> still getting stuck after the kernel starts.
> Any further ideas?

Here's a simple checklist:

1. Verify that DT configures both the serial port AND the IOMUXC,
please replace port numbers with your own:

&uart3 {                                <<= ttymxc2
	pinctrl-names = "default";
	pinctrl-0 = <&pinctrl_uart3_2>; <<= use pinctrl uart3_2
	status = "okay";                <<= port is now enabled
};

If your UART pins are not available as ready-made pinctrl
configuration, you need to define new pinctrl and pass it here.

2. Verify that u-boot sets bootargs:

setenv bootargs ... console=ttymxcX,115200 ...


I followed these on a customer board and all worked fine without issues.

> Thanks,
>
> Leon.
>
>
> conga-QMX6 U-Boot > run sd
> resetting ...
>
>
> U-Boot 2013.04 QMX6Rx20 (May 19 2015 - 13:34:17)
>
> CPU:   Freescale i.MX6Q rev1.2 at 792 MHz
> CPU:   Temperature 32 C, calibration data: 0x56b4d469
> Reset cause: POR
> Board: conga-QMX6
> I2C:   ready
> DRAM:  1 GiB
> MMC:   FSL_SDHC: 0, FSL_SDHC: 1, FSL_SDHC: 2
> SF: Detected SST25VF032B with page size 4 KiB, total 4 MiB
> No panel detected: default to Hannstar-XGA
> Display: Hannstar-XGA (1024x768)
> In:    serial
> Out:   serial
> Err:   serial
> PFUZE100 Rev. [10/21] detected
> Net:   configure Atheros AR8035 Ethernet Phy at address 6
> FEC
> Warning: FEC using MAC address from net device
>
> Normal Boot
> Hit any key to stop autoboot:  0
> U-Boot > run bootcmd_net
> Using FEC device
> TFTP from server 192.168.1.9; our IP address is 192.168.1.103
> Filename 'cgtqmx6/uImage.dtb'.
> Load address: 0x10780000
> Loading: ####
>       6.6 MiB/s
> done
> Bytes transferred = 48083 (bbd3 hex)
> Using FEC device
> TFTP from server 192.168.1.9; our IP address is 192.168.1.103
> Filename 'cgtqmx6/uImage'.
> Load address: 0x10800000
> Loading: #################################################################
>       #################################################################
>       #################################################################
>       #################################################################
>       #################################################################
>       ##############################################
>       6.9 MiB/s
> done
> Bytes transferred = 5432656 (52e550 hex)
> ## Booting kernel from Legacy Image at 10800000 ...
>     Image Name:   Linux-3.10.53-1.1.1_qmx6+g9f56bf
>     Image Type:   ARM Linux Kernel Image (uncompressed)
>     Data Size:    5432592 Bytes = 5.2 MiB
>     Load Address: 10008000
>     Entry Point:  10008000
>     Verifying Checksum ... OK
> ## Flattened Device Tree blob at 10780000
>     Booting using the fdt blob at 0x10780000
>     Loading Kernel Image ... OK
> OK
> switch to ldo_bypass mode!
>     Loading Device Tree to 4f543000, end 4f551bd2 ... OK
>
> Starting kernel ...
>
>
>
>
>
>
>
>
>
> On Wed, May 20, 2015 at 10:47 PM, Leon Woestenberg
> <leon.woestenberg at gmail.com <mailto:leon.woestenberg at gmail.com>> wrote:
>
>     Reading that patch comment more clearly, I now understand that both
>     u-boot and the SoC do in fact interpret the registers, because the
>     SoC ignores it for SPL type boot.
>
>     With that, I will try to boot the .sdcard image generated by the
>     fsl-community-bsp-platform tomorrow.
>
>     Thanks,
>
>     Leon.
>
>     On Wed, May 20, 2015 at 10:43 PM, Leon Woestenberg
>     <leon.woestenberg at gmail.com <mailto:leon.woestenberg at gmail.com>> wrote:
>
>         Hi Otavio,
>
>
>         On Wed, May 20, 2015 at 10:32 PM, Otavio Salvador
>         <otavio at ossystems.com.br <mailto:otavio at ossystems.com.br>> wrote:
>
>             On Wed, May 20, 2015 at 5:27 PM, Leon Woestenberg
>             <leon.woestenberg at gmail.com
>             <mailto:leon.woestenberg at gmail.com>> wrote:
>             ...
>             > Can anybody provide more insight here?
>             ...
>
>             yes. This is mostly what bmode in new U-Boot versions does. The
>             0x10000000 can be seen at:
>
>             https://github.com/Freescale/u-boot-fslc/blob/patches-2015.04/arch/arm/cpu/armv7/mx6/soc.c#L478
>
>
>         Thanks. So these registers are interpreted by the SoC itself and
>         not by u-boot on subsequent warm-reset?
>
>         I found some explanation of these "secret" registers here:
>
>         https://patchwork.ozlabs.org/patch/404723/

Regards,
Nikolay