Kernel 3.0.15 on i.MX53.

[ricardo_ioct]

Hello,

I'm trying to run the kernel 3.0.15 on a imx53-based board (it's a NOVPECK). I successfully added the kernel that I catch from freescale's git on LTIB and compile it very well, but, when I try to boot it, I get this log on console:

U-Boot 2009.08-svn226 (Feb 15 2012 - 08:19:29)

CPU:   Freescale i.MX53 family 1.0V at 800 MHz

mx53 pll1: 800MHz

mx53 pll2: 800MHz

mx53 pll3: 216MHz

mx53 pll4: 595MHz

ipg clock     : 80000000Hz

ipg per clock : 66666666Hz

uart clock    : 21600000Hz

cspi clock    : 54000000Hz

ahb clock     : 160000000Hz

axi_a clock   : 400000000Hz

axi_b clock   : 266666666Hz

emi_slow clock: 266666666Hz

ddr clock     : 400000000Hz

esdhc1 clock  : 160000000Hz

esdhc2 clock  : 160000000Hz

esdhc3 clock  : 160000000Hz

esdhc4 clock  : 160000000Hz

nfc clock     : 33333333Hz

Board: NOVPEK i.MX53 1.0 [Boot Reason: [POR]

Boot Device: SD

I2C:   ready

DRAM:  512 MB

NAND:  No NAND device found!!!

0 MiB

MMC:   FSL_ESDHC: 0

In:    serial

Out:   serial

Err:   serial

Hit any key to stop autoboot:  0

MMC read: dev # 0, block # 2048, count 8192 partition # 0 ...

8192 blocks read: OK

## Booting kernel from Legacy Image at 70800000 ...

   Image Name:   Linux-3.0.15-01244-g34621b4-dirt

   Image Type:   ARM Linux Kernel Image (uncompressed)

   Data Size:    2621556 Bytes =  2.5 MB

   Load Address: 70008000

   Entry Point:  70008000

   Verifying Checksum ... OK

   Loading Kernel Image ... OK

OK

Starting kernel ...

Uncompressing Linux... done, booting the kernel.

�8<�<��8�88���8�8��8��8<�<��8��8�����8���8�<�8�8�<��8��������8�<���8�����������8��8��8�<� �< 8��8�8�8�8���8��8��8�8�

So, after uncompress the kernel, nothing but trash pops up on console. I'm just tired and ran out of ideas. Anybody already had this problem?

My printenv:

NOVPEK_i.MX53_U-Boot> printenv

bootdelay=3

baudrate=115200

ethaddr=88:77:66:55:34:12

ipaddr=192.168.1.201

gatewayip=192.168.1.1

netmask=255.255.255.0

hostname=NOVPEK

loadaddr=0x70800000

netdev=eth0

ethprime=FEC0

uboot=u-boot.bin

kernel=uImage

bootcmd_net=run bootargs_base bootargs_nfs; tftpboot ${loadaddr} ${kernel}; bootm

loadk=mmc read 0 ${loadaddr} 0x800 0x1800

bootcmd_mmc=run bootargs_base bootargs_mmc; mmc read 0 ${loadaddr} 0x800 0x2000;bootm

bootcmd_nfs=run bootargs_base bootargs_nfs; mmc read 0 ${loadaddr} 0x800 0x2000; bootm

serverip=192.168.15.132

nfsroot=/opt/freescale/rootfs

bootargs_nfs=setenv bootargs ${bootargs} root=/dev/nfs ip=dhcp video=mxcdi0fb:RGB565,240x320M-16@15 nfsroot=${serverip}:${nfsroot},v3,tcp

bootcmd=run bootcmd_mmc

bootargs_mmc=setenv bootargs ${bootargs} noinitrd console=tty1 root=/dev/mmcblk0p1 ro rootwait ip=none

bootargs_base=setenv bootargs console=ttymxc0,115200 jtag=on

stdin=serial

stdout=serial

stderr=serial

Environment size: 930/131068 bytes

Regards,

Ricardo

[rogerio_silva]

I suggest you to use either 2.6.35 from Freescale or the Kernel Mainline (kernel.org).

Kernel 3.0.15 from Freescale was made to be used i.MX6 platforms and wasn't tested on i.MX5x.

If you prefer to use Kernel Mainline, check if all drivers you need are currently there.

Rgds

Rogerio

[hubertchaumette · Accepted solution]

Hello,

I have encountered the same problem with kernel 2.6.35 and solved it by editing arch/arm/mach-mx5/clock.c :

line 2129

static struct clk uart_main_clk = {

     ...

     .parent = &pll3_sw_clk, // originally &pll2_sw_clk

     ...

}

I've tried it on Linux 3.0.15, it worked too, now I can enjoy my kernel panic.

Solution found here (patch included) :

https://community.freescale.com/thread/259422

Regards,

Hubert

[ricardo_ioct]

Oh, I see. Indeed, it make sense.

By now, I'm putting my efforts back to kernel 2.6.35, but as soon as I can, I'll follow your idea. Thank you very much for your help!

Regards,

Ricardo.

[weidong_sun]

Hello,ricardo,

     if Hubert's answer is OK, please press "Correct answer" Button !

Thanks !

Weidong

[LeonardoSandova]

I am not sure the kernel param 'jtag=on' affects something on a normal boot so try to remove it.

Leo

[ricardo_ioct]

Hello Leonardo,

I've tried your tip, stil the same. One thing: this same u-boot env. works perfectly with kernel 2.6.35. Just config issues or implementation bugs?