[Buildroot] Problem with an MPC8272 build

[Buildroot] Problem with an MPC8272 build

[Jason Rennie]

Hi,

I'm trying to get buildroot to work with an MPC8272 based device that we 
have. I've been trying for a while and using google to see if it can 
help but i'm getting nowhere so I thought I would try going to the source.

I think i've set up buildroot correctly and have it configured to be an 
mpc8272ads as the kernel defconfig. I know the kernel can be built (so 
it seems) because the device does have an old 2.6.18 kernel running on 
it at the moment that boots successfully, using U-Boot and an older 
version of buildroot to make the file system.

But now I can't get the kernel to boot at all. Or at least nothing shows 
when it comes to booting the kernel.

Buildroot does run successfully, and I have put the rootfs that is made 
on the device and it does boot successfully with the older kernel (well 
mostly, some networking stuff appears to be missing). But when I put the 
new kernel on the device I get nothing. I've tried enabling early kernel 
debugging but with no success (setting the CPM to the addresses 
suggested by the old kernels output). I also don't seem to be able to 
get U-Boot to work and I need to end up breaking out the BDI2000 anytime 
I try to update U-Boot with the newer version as I brick the device.

I have tried with the most basic possible build, not tweaking anything 
and basically just doing the stock buildroot build for the MPC8272 
through to poking and tweaking all sorts of thing.

Any tips? I've attached the old dmesg output if tht helps. I'm 
hopelessly confused and been beating my head against this for a month so 
any help would be greatly appreciated. I'm pretty much out of ideas.

Jason
-------------- next part --------------
Linux version 2.6.18 (dmytro at woomera) (gcc version 4.1.1) #51 PREEMPT Thu Dec 31 15:22:09 EST 2009
Motorola PQ2 ADS PowerPC port
On node 0 totalpages: 16384
  DMA zone: 16384 pages, LIFO batch:3
Built 1 zonelists.  Total pages: 16384
Kernel command line: console=ttyCPM0,115200 root=31:02 rw rootfstype=jffs2 mtdparts=physmap-flash.0:512K(U-Boot),1024k(Linux),-(root) ip=192.168.1.201:192.168.1.165::255.255.255.0:eclipse:eth0:on panic=1
PID hash table entries: 512 (order: 9, 2048 bytes)
Warning: real time clock seems stuck!
cpm_uart: console: compat mode
Dentry cache hash table entries: 8192 (order: 3, 32768 bytes)
Inode-cache hash table entries: 4096 (order: 2, 16384 bytes)
Memory: 62720k available (1708k kernel code, 436k data, 92k init, 0k highmem)
Calibrating delay loop... 175.10 BogoMIPS (lpj=87552)
Mount-cache hash table entries: 512
NET: Registered protocol family 16
mpc8272ads: Init
NET: Registered protocol family 2
IP route cache hash table entries: 512 (order: -1, 2048 bytes)
TCP established hash table entries: 2048 (order: 1, 8192 bytes)
TCP bind hash table entries: 1024 (order: 0, 4096 bytes)
TCP: Hash tables configured (established 2048 bind 1024)
TCP reno registered
JFFS2 version 2.2. (C) 2001-2006 Red Hat, Inc.
fuse init (API version 7.7)
io scheduler noop registered
io scheduler deadline registered (default)
Generic RTC Driver v1.07
Software Watchdog Timer: 0.07 initialized. soft_noboot=0 soft_margin=60 sec (nowayout= 1)
Serial: CPM driver $Revision: 0.02 $
ttyCPM0 at MMIO 0xf0011a00 (irq = 40) is a CPM UART
ttyCPM1 at MMIO 0xf0011a60 (irq = 43) is a CPM UART
RAMDISK driver initialized: 16 RAM disks of 32768K size 1024 blocksize
physmap platform flash device: 01000000 at ff000000
physmap-flash.0: Found 1 x16 devices at 0x0 in 16-bit bank
 Intel/Sharp Extended Query Table at 0x010A
 Intel/Sharp Extended Query Table at 0x010A
 Intel/Sharp Extended Query Table at 0x010A
 Intel/Sharp Extended Query Table at 0x010A
 Intel/Sharp Extended Query Table at 0x010A
Using buffer write method
cfi_cmdset_0001: Erase suspend on write enabled
erase region 0: offset=0x0,size=0x8000,blocks=4
erase region 1: offset=0x20000,size=0x20000,blocks=127
3 cmdlinepart partitions found on MTD device physmap-flash.0
Creating 3 MTD partitions on "physmap-flash.0":
0x00000000-0x00080000 : "U-Boot"
0x00080000-0x00180000 : "Linux"
0x00180000-0x01000000 : "root"
eth0: FCC ENET Version 0.3, de:ad:be:ef:12:34
ip_tables: (C) 2000-2006 Netfilter Core Team
TCP bic registered
NET: Registered protocol family 1
NET: Registered protocol family 17
IP-Config: Complete:
      device=eth0, addr=192.168.1.201, mask=255.255.255.0, gw=255.255.255.255,
     host=eclipse, domain=, nis-domain=(none),
     bootserver=192.168.1.165, rootserver=192.168.1.165, rootpath=
VFS: Mounted root (jffs2 filesystem).
Freeing unused kernel memory: 92k init

[Buildroot] Problem with an MPC8272 build

[Charles Krinke]

One thing to check is that you have defined a compatible root console. Your
printout indicates ttyCPM0 and in buildroot config, one can set this to
ttyCPM0 as it normally defaults to ttyS0

Charles
On Aug 1, 2013 4:49 PM, "Jason Rennie" <jason.rennie@rftechnology.com.au>
wrote:

> Hi,
>
> I'm trying to get buildroot to work with an MPC8272 based device that we
> have. I've been trying for a while and using google to see if it can help
> but i'm getting nowhere so I thought I would try going to the source.
>
> I think i've set up buildroot correctly and have it configured to be an
> mpc8272ads as the kernel defconfig. I know the kernel can be built (so it
> seems) because the device does have an old 2.6.18 kernel running on it at
> the moment that boots successfully, using U-Boot and an older version of
> buildroot to make the file system.
>
> But now I can't get the kernel to boot at all. Or at least nothing shows
> when it comes to booting the kernel.
>
> Buildroot does run successfully, and I have put the rootfs that is made on
> the device and it does boot successfully with the older kernel (well
> mostly, some networking stuff appears to be missing). But when I put the
> new kernel on the device I get nothing. I've tried enabling early kernel
> debugging but with no success (setting the CPM to the addresses suggested
> by the old kernels output). I also don't seem to be able to get U-Boot to
> work and I need to end up breaking out the BDI2000 anytime I try to update
> U-Boot with the newer version as I brick the device.
>
> I have tried with the most basic possible build, not tweaking anything and
> basically just doing the stock buildroot build for the MPC8272 through to
> poking and tweaking all sorts of thing.
>
> Any tips? I've attached the old dmesg output if tht helps. I'm hopelessly
> confused and been beating my head against this for a month so any help
> would be greatly appreciated. I'm pretty much out of ideas.
>
> Jason
>
> _______________________________________________
> buildroot mailing list
> buildroot at busybox.net
> http://lists.busybox.net/mailman/listinfo/buildroot
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.busybox.net/pipermail/buildroot/attachments/20130801/d70eb412/attachment.html>

[Buildroot] Problem with an MPC8272 build

[Jason Rennie]

Thanks for the tip Charles. I have made that change (after I booted the 
image and it got really upset that ttyS0 didn't exist) as well as using 
that memory address (and the one for CPM1) in the Early Kernel debugging 
option. All sadly give no output.

Is there anyway to tell if the system has booted apart from the boot 
output? I'm guessing the answer is very hardware dependent, but thought 
I would ask anyway.

Jason

On 2/08/2013 1:46 PM, Charles Krinke wrote:
>
> One thing to check is that you have defined a compatible root console. 
> Your printout indicates ttyCPM0 and in buildroot config, one can set 
> this to ttyCPM0 as it normally defaults to ttyS0
>
> Charles
>
> On Aug 1, 2013 4:49 PM, "Jason Rennie" 
> <jason.rennie@rftechnology.com.au 
> <mailto:jason.rennie@rftechnology.com.au>> wrote:
>
>     Hi,
>
>     I'm trying to get buildroot to work with an MPC8272 based device
>     that we have. I've been trying for a while and using google to see
>     if it can help but i'm getting nowhere so I thought I would try
>     going to the source.
>
>     I think i've set up buildroot correctly and have it configured to
>     be an mpc8272ads as the kernel defconfig. I know the kernel can be
>     built (so it seems) because the device does have an old 2.6.18
>     kernel running on it at the moment that boots successfully, using
>     U-Boot and an older version of buildroot to make the file system.
>
>     But now I can't get the kernel to boot at all. Or at least nothing
>     shows when it comes to booting the kernel.
>
>     Buildroot does run successfully, and I have put the rootfs that is
>     made on the device and it does boot successfully with the older
>     kernel (well mostly, some networking stuff appears to be missing).
>     But when I put the new kernel on the device I get nothing. I've
>     tried enabling early kernel debugging but with no success (setting
>     the CPM to the addresses suggested by the old kernels output). I
>     also don't seem to be able to get U-Boot to work and I need to end
>     up breaking out the BDI2000 anytime I try to update U-Boot with
>     the newer version as I brick the device.
>
>     I have tried with the most basic possible build, not tweaking
>     anything and basically just doing the stock buildroot build for
>     the MPC8272 through to poking and tweaking all sorts of thing.
>
>     Any tips? I've attached the old dmesg output if tht helps. I'm
>     hopelessly confused and been beating my head against this for a
>     month so any help would be greatly appreciated. I'm pretty much
>     out of ideas.
>
>     Jason
>
>     _______________________________________________
>     buildroot mailing list
>     buildroot at busybox.net <mailto:buildroot@busybox.net>
>     http://lists.busybox.net/mailman/listinfo/buildroot
>

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.busybox.net/pipermail/buildroot/attachments/20130802/38c5a17a/attachment-0001.html>

[Buildroot] Problem with an MPC8272 build

[Gustavo Zacarias]

On 08/01/2013 08:49 PM, Jason Rennie wrote:
> Buildroot does run successfully, and I have put the rootfs that is made
> on the device and it does boot successfully with the older kernel (well
> mostly, some networking stuff appears to be missing). But when I put the
> new kernel on the device I get nothing. I've tried enabling early kernel
> debugging but with no success (setting the CPM to the addresses
> suggested by the old kernels output). I also don't seem to be able to
> get U-Boot to work and I need to end up breaking out the BDI2000 anytime
> I try to update U-Boot with the newer version as I brick the device.
> 
> I have tried with the most basic possible build, not tweaking anything
> and basically just doing the stock buildroot build for the MPC8272
> through to poking and tweaking all sorts of thing.
> 
> Any tips? I've attached the old dmesg output if tht helps. I'm
> hopelessly confused and been beating my head against this for a month so
> any help would be greatly appreciated. I'm pretty much out of ideas.

A buildroot .config would help so we can see if there's anything that
could go wrong.
On the toolchain side a 603e should be the target, if that's wrong then
anything you build might fail.
On the kernel side are you building a 2.6.18ish kernel with a new
buildroot or a newer kernel? An updated kernel will very likely need an
updated DTB as well.
Regards.

[Buildroot] Problem with an MPC8272 build

[Charles Krinke]

Good point Gustavo. If I recall, if the dtb is wrong, one will not get
as far in the kernel boot process.

In looking at the output, it looks like it is all the way to running
the init process. Usually when that happens in the systems I work
with, it is a matter of not satisfying the init process properly. Most
usually that is a mismatch of the name of the root console.

For whatever reason, in recent years, all the folks making SOC designs
have gone away from the old, original ttyS0 and every vendor seems to
have a special name for their serial port, so I thought that would be
the most fruitful place to look.

Charles

On 8/2/13, Gustavo Zacarias <gustavo@zacarias.com.ar> wrote:
> On 08/01/2013 08:49 PM, Jason Rennie wrote:
>> Buildroot does run successfully, and I have put the rootfs that is made
>> on the device and it does boot successfully with the older kernel (well
>> mostly, some networking stuff appears to be missing). But when I put the
>> new kernel on the device I get nothing. I've tried enabling early kernel
>> debugging but with no success (setting the CPM to the addresses
>> suggested by the old kernels output). I also don't seem to be able to
>> get U-Boot to work and I need to end up breaking out the BDI2000 anytime
>> I try to update U-Boot with the newer version as I brick the device.
>>
>> I have tried with the most basic possible build, not tweaking anything
>> and basically just doing the stock buildroot build for the MPC8272
>> through to poking and tweaking all sorts of thing.
>>
>> Any tips? I've attached the old dmesg output if tht helps. I'm
>> hopelessly confused and been beating my head against this for a month so
>> any help would be greatly appreciated. I'm pretty much out of ideas.
>
> A buildroot .config would help so we can see if there's anything that
> could go wrong.
> On the toolchain side a 603e should be the target, if that's wrong then
> anything you build might fail.
> On the kernel side are you building a 2.6.18ish kernel with a new
> buildroot or a newer kernel? An updated kernel will very likely need an
> updated DTB as well.
> Regards.
> _______________________________________________
> buildroot mailing list
> buildroot at busybox.net
> http://lists.busybox.net/mailman/listinfo/buildroot
>


-- 
Charles Krinke

[Buildroot] Problem with an MPC8272 build

[Jason Rennie]

Thanks for the reply Charles.

I do know I need to use ttyCPM0/1 (made that mistake already). I agree, 
what is wrong with the old S0?

The kernel output I included was from the 2.6.18 build made long ago by 
someone who has long since moved on. I figured it would be worth 
including as a "last known good" configuration.

Jason

On 3/08/2013 4:18 AM, Charles Krinke wrote:
> Good point Gustavo. If I recall, if the dtb is wrong, one will not get
> as far in the kernel boot process.
>
> In looking at the output, it looks like it is all the way to running
> the init process. Usually when that happens in the systems I work
> with, it is a matter of not satisfying the init process properly. Most
> usually that is a mismatch of the name of the root console.
>
> For whatever reason, in recent years, all the folks making SOC designs
> have gone away from the old, original ttyS0 and every vendor seems to
> have a special name for their serial port, so I thought that would be
> the most fruitful place to look.
>
> Charles
>
> On 8/2/13, Gustavo Zacarias <gustavo@zacarias.com.ar> wrote:
>> On 08/01/2013 08:49 PM, Jason Rennie wrote:
>>> Buildroot does run successfully, and I have put the rootfs that is made
>>> on the device and it does boot successfully with the older kernel (well
>>> mostly, some networking stuff appears to be missing). But when I put the
>>> new kernel on the device I get nothing. I've tried enabling early kernel
>>> debugging but with no success (setting the CPM to the addresses
>>> suggested by the old kernels output). I also don't seem to be able to
>>> get U-Boot to work and I need to end up breaking out the BDI2000 anytime
>>> I try to update U-Boot with the newer version as I brick the device.
>>>
>>> I have tried with the most basic possible build, not tweaking anything
>>> and basically just doing the stock buildroot build for the MPC8272
>>> through to poking and tweaking all sorts of thing.
>>>
>>> Any tips? I've attached the old dmesg output if tht helps. I'm
>>> hopelessly confused and been beating my head against this for a month so
>>> any help would be greatly appreciated. I'm pretty much out of ideas.
>> A buildroot .config would help so we can see if there's anything that
>> could go wrong.
>> On the toolchain side a 603e should be the target, if that's wrong then
>> anything you build might fail.
>> On the kernel side are you building a 2.6.18ish kernel with a new
>> buildroot or a newer kernel? An updated kernel will very likely need an
>> updated DTB as well.
>> Regards.
>> _______________________________________________
>> buildroot mailing list
>> buildroot at busybox.net
>> http://lists.busybox.net/mailman/listinfo/buildroot
>>
>

[Buildroot] Problem with an MPC8272 build

[Jason Rennie]

Thank you so much everyone for the help. I got it to boot. It paniced 
near the end but i'll figure that out. It booted :D

The problem appeared to be related to lacking the device table and 
making a cuImage seems to have done the trick.

Thanks again.

Jason

On 3/08/2013 4:18 AM, Charles Krinke wrote:
> Good point Gustavo. If I recall, if the dtb is wrong, one will not get
> as far in the kernel boot process.
>
> In looking at the output, it looks like it is all the way to running
> the init process. Usually when that happens in the systems I work
> with, it is a matter of not satisfying the init process properly. Most
> usually that is a mismatch of the name of the root console.
>
> For whatever reason, in recent years, all the folks making SOC designs
> have gone away from the old, original ttyS0 and every vendor seems to
> have a special name for their serial port, so I thought that would be
> the most fruitful place to look.
>
> Charles
>
> On 8/2/13, Gustavo Zacarias <gustavo@zacarias.com.ar> wrote:
>> On 08/01/2013 08:49 PM, Jason Rennie wrote:
>>> Buildroot does run successfully, and I have put the rootfs that is made
>>> on the device and it does boot successfully with the older kernel (well
>>> mostly, some networking stuff appears to be missing). But when I put the
>>> new kernel on the device I get nothing. I've tried enabling early kernel
>>> debugging but with no success (setting the CPM to the addresses
>>> suggested by the old kernels output). I also don't seem to be able to
>>> get U-Boot to work and I need to end up breaking out the BDI2000 anytime
>>> I try to update U-Boot with the newer version as I brick the device.
>>>
>>> I have tried with the most basic possible build, not tweaking anything
>>> and basically just doing the stock buildroot build for the MPC8272
>>> through to poking and tweaking all sorts of thing.
>>>
>>> Any tips? I've attached the old dmesg output if tht helps. I'm
>>> hopelessly confused and been beating my head against this for a month so
>>> any help would be greatly appreciated. I'm pretty much out of ideas.
>> A buildroot .config would help so we can see if there's anything that
>> could go wrong.
>> On the toolchain side a 603e should be the target, if that's wrong then
>> anything you build might fail.
>> On the kernel side are you building a 2.6.18ish kernel with a new
>> buildroot or a newer kernel? An updated kernel will very likely need an
>> updated DTB as well.
>> Regards.
>> _______________________________________________
>> buildroot mailing list
>> buildroot at busybox.net
>> http://lists.busybox.net/mailman/listinfo/buildroot
>>
>