Opened 7 years ago

Closed 5 years ago

Last modified 4 years ago

#9701 closed defect (worksforme)

vmlinux kernel image broken, bcm63xx

Reported by: danitool <dgcbueu@…> Owned by: florian
Priority: high Milestone: Barrier Breaker 14.07
Component: kernel Version: Trunk
Keywords: kernel image Cc:


This happens in my board livebox with a redboot loader.

Previous kernel version (2.6.37) was working with the same boot script in redboot. The strange is vmlinux doesn't do anything when Redboot tries to execute it, but the elf kernel image works totally fine..

This is what happens in a normal boot:

ESA: 00:07:3a:ff:ff:ff

Auto-negotiation timed-out
10 MB Half-Duplex (assumed)

Ethernet eth0: MAC address 00:07:3a:ff:ff:ff
IP:, Default server:
Hardware version 0x0A BLUE5G.6 WITHOUT OPTION
Factory Settings Recovery Switch OFF

RedBoot(tm) bootstrap and debug environment [ROM]
Non-certified release, version release-wanadoo-fr - built 11:30:26, Jun 22 2011

Platform: Blue_5g (MIPS32 4Kc) 
Copyright (C) 2000, 2001, 2002, Red Hat, Inc.

RAM: 0x80000000-0x81000000, [0x80435d10-0x80fe2000] available
FLASH: 0xbe400000 - 0xbec00000, 128 blocks of 0x00010000 bytes each.
== Executing boot script in 20.000 seconds - enter ^C twice to abort
hit '\0x03' (0x03)
RedBoot> fis load -b 0x80010000 -d kernel

Image loaded from 0x80010000-0x8028fd44
RedBoot> exec -c "noinitrd mem=16M" 0x80010000
Now booting linux kernel:
 Base address 0x8000fc00 Entry 0x80010000
 Cmdline : noinitrd mem=16M

It stops there, and the device doesn't return anything else.

But when I stop into Redboot and I load the vmlinux elf image manually, all goes fine..

RedBoot> load -v openwrt-kernel.elf
Using default protocol (TFTP)
Entry point: 0x80206ee0, address range: 0x80010000-0x8028fd44
RedBoot> go
Linux version (dani@tool) (gcc version 4.5.4 20110526 (prerelease) (Linaro GCC 4.5-2011.06-0) ) #1 Wed Jul 13 12:33:33 CEST 2011
Detected Broadcom 0x6348 CPU revision b0
CPU frequency is 256 MHz
16MB of RAM installed
registering 37 GPIOs
board_livebox: enabling icache and dcache...
board_livebox: resetting gpio6 for eth1...
bootconsole [early0] enabled
CPU revision is: 00029107 (Broadcom BMIPS3300)
board_livebox: board name: Livebox
Determined physical RAM map:
 memory: 01000000 @ 00000000 (usable)
Initrd not found or empty - disabling initrd
Zone PFN ranges:
  Normal   0x00000000 -> 0x00001000

until it boots, i can telnet openwrt, etc, etc. All fine

If I manually load the vmlinux regular image without flashing anything, the result is the same as when it loads itself from partitions, the boot process stops, and executes nothing.

The elf image works but vmlinux regular image doesn't.

Then, there is something broken in the vmlinux image itself but not the configuration?

Btw this is my kernel config:

Attachments (1)

019-fix-boot.patch (794 bytes) - added by danitool <dgcbueu@…> 7 years ago.
boot fix

Download all attachments as: .zip

Change History (6)

comment:1 Changed 7 years ago by danitool <dgcbueu@…>

I solved it with a patch already present in brcm47xx platforms.


But adapted for the bcm63xx platform

Changed 7 years ago by danitool <dgcbueu@…>

boot fix

comment:2 Changed 5 years ago by florian

  • Owner changed from developers to florian
  • Status changed from new to assigned

Is this fix still required?

comment:3 Changed 5 years ago by danitool <dgcbueu@…>

No, close the ticket.

comment:4 Changed 5 years ago by florian

  • Resolution set to worksforme
  • Status changed from assigned to closed

comment:5 Changed 4 years ago by jow

  • Milestone changed from Attitude Adjustment 12.09 to Barrier Breaker 14.07

Milestone Attitude Adjustment 12.09 deleted

Add Comment

Modify Ticket

as closed .
The resolution will be deleted. Next status will be 'reopened'.

E-mail address and user name can be saved in the Preferences.

Note: See TracTickets for help on using tickets.