usb start in uEnv.txt fails with test - minimal test >>> like /bin/sh

[mahmoh]

Build Version/Date: u-boot-linaro-omap4-panda_2011.07.2-0ubuntu1_armel.deb
Environment used for testing: pandaboard A2

Summary: usb start in uEnv.txt fails with test - minimal test >>> like /bin/sh

Steps to Reproduce: boot with no boot.scr, and uEnv.txt:
    autoload=no
    uenvcmd=usb start; pxe get; pxe boot; mmcinfo; fatload ...

Expected result: u-boot initializes usb devices and continues on

Actual result: u-boot fails to initialize usb devices and skips instructions

<START: no boot.scr, uEnv.txt: uenvcmd=usb start; pxe get; pxe boot; mmcinfo; fatload ... >
...
reading boot.scr

** Unable to read "boot.scr" from mmc 0:1 **
reading uEnv.txt

323 bytes read
Importing environment from mmc ...
test - minimal test like /bin/sh

Usage:
test [args..]
reading uImage

4174740 bytes read
Booting from mmc0 ...
## Booting kernel from Legacy Image at 82000000 ...
   Image Name: Ubuntu Kernel
   Image Type: ARM Linux Kernel Image (uncompressed)
   Data Size: 4174676 Bytes = 4 MiB
   Load Address: 80008000
   Entry Point: 80008000
   Verifying Checksum ... OK
   Loading Kernel Image ... OK
OK
...
<END>

[John Rigby (jcrigby)]

with this uEnv.txt:
autoload=no
uenvcmd=usb start; bootp; pxe get; pxe boot

I get this:
Texas Instruments X-Loader 1.5.1 (Jul 26 2011 - 00:40:14)
Reading boot sector
Loading u-boot.bin from mmc

U-Boot 2011.06 (Jul 27 2011 - 00:58:33)

CPU : OMAP4430
Board: OMAP4 Panda
I2C: ready
DRAM: 1 GiB
MMC: OMAP SD/MMC: 0
Using default environment

In: serial
Out: serial
Err: serial
Net: No ethernet found.
Hit any key to stop autoboot: 0
reading boot.scr

** Unable to read "boot.scr" from mmc 0:1 **
reading uEnv.txt

78 bytes read
Importing environment from mmc ...
Running uenvcmd ...
(Re)start USB...
USB: Register 1313 NbrPorts 3
USB EHCI 1.00
scanning bus for devices... The request port(2) is not configured
The request port(2) is not configured
3 USB Device(s) found
       scanning bus for storage devices... 0 Storage Device(s) found
       scanning bus for ethernet devices... 1 Ethernet Device(s) found
Waiting for Ethernet connection... done.
BOOTP broadcast 1
EHCI timed out on TD - token=0x8008d80
BOOTP broadcast 2
*** Unhandled DHCP Option in OFFER/ACK: 28
*** Unhandled DHCP Option in OFFER/ACK: 28
DHCP client bound to address 10.42.43.50
missing environment variable: pxeuuid
missing environment variable: ethaddr
Retreiving file: pxelinux.cfg/2E-40-4F-F6-6C-0A
Waiting for Ethernet connection... done.
Using sms0 device
TFTP from server 10.42.43.1; our IP address is 10.42.43.50
Filename 'pxelinux.cfg/2E-40-4F-F6-6C-0A'.
Load address: 0x88000000
Loading: #
done
Bytes transferred = 508 (1fc hex)
Config file found
Ignoring malformed menu command: label
Ignoring malformed menu command: label
Hit any key to stop autoboot: 0
Label: linux-2.6.38
        kernel: kernels/linux-2.6.38.bin
        append: root=/dev/sdb1
Retreiving file: kernels/linux-2.6.38.bin
Waiting for Ethernet connection... done.
Using sms0 device
TFTP from server 10.42.43.1; our IP address is 10.42.43.50
Filename 'kernels/linux-2.6.38.bin'.
Load address: 0x80200000
Loading: *
TFTP error: 'File not found' (1)
Not retrying...
File not found
Skipping label
Label: install
        kernel: kernels/install.bin
        append: console=ttyAMA0,38400 debug earlyprintk
        initrd: initrds/uzInitrdDebInstall
Retreiving file: initrds/uzInitrdDebInstall
Waiting for Ethernet connection... done.
Using sms0 device
TFTP from server 10.42.43.1; our IP address is 10.42.43.50
Filename 'initrds/uzInitrdDebInstall'.
Load address: 0x81600000
Loading: *
TFTP error: 'File not found' (1)
Not retrying...
File not found
Skipping label
Label: local
        append: root=/dev/sdb1
missing environment variable: localcmd
reading uImage

3610768 bytes read
Booting from mmc0 ...
## Booting kernel from Legacy Image at 82000000 ...
   Image Name: Ubuntu Kernel
   Image Type: ARM Linux Kernel Image (uncompressed)
   Data Size: 3610704 Bytes = 3.4 MiB
   Load Address: 80008000
   Entry Point: 80008000
   Verifying Checksum ... OK
   Loading Kernel Image ... OK
OK

Starting kernel ...

Uncompressing Linux... done, booting the kernel.

[John Rigby (jcrigby)]

mahmoh, can you pastebin your uEnv.txt file?

[mahmoh (mahmoh)]

#pxecfg_ram=0x82000000
#ethaddr=00:02:03:04:05:06
#usbethaddr=00:02:03:04:05:06
autoload=no
bootargs=ro root=/dev/sda2 console=ttyO2,115200 debug earlyprintk elevator=deadline
uenvcmd=usb start; pxe get; pxe boot; mmcinfo; fatload mmc 0:1 0x80000000 uimage; fatload mmc 0:1 0x81600000 uinitrd; bootm 0x80000000 0x81600000

[John Rigby (jcrigby)]

mahmoh, ok weird problems in u-boot, the uenvcmd variable does not get expanded properly. Working on that.

Another issue, your uenvcmd is missing a "bootp" after "usb start".

[John Rigby (jcrigby)]

U-Boot is getting an error parsing the uenvcmd env setting. The current limit of number of tokens in a command line is 16, normally you can get around this by quoting things to put them in a single token but for some reason the import env doesn't grok that correctly so as an alternate work around you can split the command. Looks like you want to do a normal mmc boot if the pxe boot does not work so this should do the trick (works for me at least):

#pxecfg_ram=0x82000000
#ethaddr=00:02:03:04:05:06
#usbethaddr=00:02:03:04:05:06
autoload=no
bootargs=ro root=/dev/sda2 console=ttyO2,115200 debug earlyprintk elevator=deadline
#uenvcmd=usb start; bootp; pxe get; pxe boot; mmcinfo; fatload mmc 0:1 0x80000000 uimage; fatload mmc 0:1 0x81600000 uinitrd; bootm 0x80000000 0x81600000
mmcbootcmd=mmcinfo; fatload mmc 0:1 0x80000000 uimage; fatload mmc 0:1 0x81600000 uinitrd; bootm 0x80000000 0x81600000
uenvcmd=usb start; bootp; pxe get; pxe boot; run mmcbootcmd

The next release of u-boot-linaro will change the max args from 16 to 32 so this will be fixed but the work around is still a good idea because separating env commands into named subcommands serves to document what you are doing.

[mahmoh (mahmoh)]

Unsure if you meant that you were going to add it to the documentation but that may be a good idea too? I'll try the above out, Thank you.

[John Rigby (jcrigby)]

Not sure what you want documented here? The bogus handling of long env commands?

[John Rigby (jcrigby)]

The MAX_ARGS change from 16 to 32 was committed 19Sep2011 so marking this fix released.