DE1_SOC Booting

Linux Kernel
1

HI,
I am trying to recompile linux kernel for DE1 SOC board. I am using the latest kernel source from Rocketboard:git.rocketboards.org/linux-socfpga.git.
The Kernel source is created for the SOCFPGA kit.Everything seems fine
when it boots, until the end of kernels booting (login part), I am
getting some strange characters and booting stops here.
I am also using preloader, u-boot.img, .rbf, .dtb files from de1_soc_GHDR project.
Does anyone have any idea what is wrong here?
I do not have much experience in these stuffs.
Thanks a lot,
Sveto

Putty output:
Hit any key to stop autoboot: 0
reading u-boot.scr
177 bytes read in 4 ms (43 KiB/s)

Executing script at 02000000

reading soc_system.rbf
7007184 bytes read in 640 ms (10.4 MiB/s)

Starting application at 0x3FF7D52C …

Application terminated, rc = 0x0

syntax error
reading zImage
3604624 bytes read in 334 ms (10.3 MiB/s)
reading socfpga.dtb
17170 bytes read in 7 ms (2.3 MiB/s)

Flattened Device Tree blob at 00000100

Booting using the fdt blob at 0x00000100
Loading Device Tree to 03ff8000, end 03fff311 … OK

Starting kernel …
…
…
Starting Bootlog daemon: bootlogd.
usb 1-1: New USB device found, idVendor=0424, idProduct=2512
usb 1-1: New USB device strings: Mfr=0, Product=0, SerialNumber=0
hub 1-1:1.0: USB hub found
hub 1-1:1.0: 2 ports detected
Configuring network interfaces… eth0: device MAC address 86:07:03:b6:30:2e
udhcpc (v1.20.2) started
Sending discover…
Sending discover…
Sending discover…
No lease, failing
Starting portmap daemon…
Sun Jul 26 13:00:00 UTC 2015
INIT: Entering runlevel: 5
Starting OpenBSD Secure Shell server: sshd
done.
Starting syslogd/klogd: done
Starting Lighttpd Web Server: lighttpd.
Stopping Bootlog daemon: bootlogd.
怘▒▒▒▒▒▒▒▒
▒?▒0▒x3▒<?▒▒▒3▒枘▒f<<?▒▒?▒x
▒▒fÆx3x▒▒fxÞ<f<
3▒▒▒▒▒枆▒▒▒?▒x

2

It can be if someone change baud rate at that point.

3

Who could change it @Victor?
I recompiled preloader, uboot, dtb, and the kernel, and I made sure that all of these use baudrate 115200.
If I change my terminal at that point to 57600 I can see the prompt and I can interact with it.
I kept the filesystem untouched. Can it have a dependent baudrate? Or what is that thing changing it? does it do so in when rootfs is loaded?

4

Hi @Sveto,
I found a workaround for this problem.
First you need to determine the baudrate that is been changed to, most probably it will be 57600.
So if you are using baudrate 115200 when you power up, close your terminal and run it again with baud rate 57600… make sure that you do not turn off your board, and do not press any key when you see that strange characters.
Now, when you open your terminal (with baud rate 57600) type: root and press Enter. Or type in whatever login you have.
You should see the prompt now. To change this and solve the problem:

  1. Open /etc/inittab to edit it: vi /etc/inittab
  2. Look for this line:
    S:2345:respawn:/sbin/getty 57600 ttyS0
    You will find it under # Normally not reached, but fallthrough in case of emergency.
  3. Change the baud rate 57600 to 115200 or whatever value matches the bootloader baud rate.
  4. Save and reboot using the new baud rate, 115200.

You should be able to boot seamlessly without problems.

5

@sirajmuhammad
So looks like you have solved problem.
As I understand baud rate was changed by inittab due to emergency.

6

@Victor
Actually, baud rate has not changed. But because we recompile preloader, bootloader, and DTS, all with baud rate 115200, and copy them to the factory image which has inittab with baud rate 57600, we face this problem.
It can be resolved by recompiling all the above with the appropriate baud rate.

7

This works for me. Thanks a Lot @sirajmuhammad.

Regards,
Sveto