All of lore.kernel.org
 help / color / mirror / Atom feed
* [Q] No message from Kernel (Howto start debug?)
@ 2011-07-28 13:15 Arno Steffen
  2011-07-28 13:25 ` Gary Thomas
                   ` (2 more replies)
  0 siblings, 3 replies; 10+ messages in thread
From: Arno Steffen @ 2011-07-28 13:15 UTC (permalink / raw)
  To: linux-omap

Maybe you are so kind to give me a hint:

Try to migrate from Kernel 33 to 37. Currently I don't see any message after

  Verifying Checksum ... OK
   Loading Kernel Image ... OK
OK

Starting kernel ...

>From what I understand kernel starts with start_kernel()).
What is best way to find out which part is failing? At which point I
do a printk to see what's going on, or at least set a GPIO Pin of the
processor?
How to debug without any external tools like JTAG.

^ permalink raw reply	[flat|nested] 10+ messages in thread

* Re: [Q] No message from Kernel (Howto start debug?)
  2011-07-28 13:15 [Q] No message from Kernel (Howto start debug?) Arno Steffen
@ 2011-07-28 13:25 ` Gary Thomas
  2011-07-28 13:26 ` Felipe Balbi
  2011-07-28 14:56 ` Sricharan R
  2 siblings, 0 replies; 10+ messages in thread
From: Gary Thomas @ 2011-07-28 13:25 UTC (permalink / raw)
  To: Arno Steffen; +Cc: linux-omap

On 2011-07-28 07:15, Arno Steffen wrote:
> Maybe you are so kind to give me a hint:
>
> Try to migrate from Kernel 33 to 37. Currently I don't see any message after
>
>    Verifying Checksum ... OK
>     Loading Kernel Image ... OK
> OK
>
> Starting kernel ...
>
>> From what I understand kernel starts with start_kernel()).
> What is best way to find out which part is failing? At which point I
> do a printk to see what's going on, or at least set a GPIO Pin of the
> processor?
> How to debug without any external tools like JTAG.

Start with your bootargs - likely ttyS2 became ttyO2

-- 
------------------------------------------------------------
Gary Thomas                 |  Consulting for the
MLB Associates              |    Embedded world
------------------------------------------------------------

^ permalink raw reply	[flat|nested] 10+ messages in thread

* Re: [Q] No message from Kernel (Howto start debug?)
  2011-07-28 13:15 [Q] No message from Kernel (Howto start debug?) Arno Steffen
  2011-07-28 13:25 ` Gary Thomas
@ 2011-07-28 13:26 ` Felipe Balbi
  2011-07-28 14:18   ` Arno Steffen
  2011-07-28 14:56 ` Sricharan R
  2 siblings, 1 reply; 10+ messages in thread
From: Felipe Balbi @ 2011-07-28 13:26 UTC (permalink / raw)
  To: Arno Steffen; +Cc: linux-omap

[-- Attachment #1: Type: text/plain, Size: 744 bytes --]

On Thu, Jul 28, 2011 at 03:15:20PM +0200, Arno Steffen wrote:
> Maybe you are so kind to give me a hint:
> 
> Try to migrate from Kernel 33 to 37. Currently I don't see any message after
> 
>   Verifying Checksum ... OK
>    Loading Kernel Image ... OK
> OK
> 
> Starting kernel ...
> 
> From what I understand kernel starts with start_kernel()).
> What is best way to find out which part is failing? At which point I
> do a printk to see what's going on, or at least set a GPIO Pin of the
> processor?
> How to debug without any external tools like JTAG.

try to enable CONFIG_DEBUG_LL and CONFIG_EARLY_PRINTK, then add "debug
earlyprintk" to your command line. See if that helps showing more debug
messages.

-- 
balbi

[-- Attachment #2: Digital signature --]
[-- Type: application/pgp-signature, Size: 490 bytes --]

^ permalink raw reply	[flat|nested] 10+ messages in thread

* Re: [Q] No message from Kernel (Howto start debug?)
  2011-07-28 13:26 ` Felipe Balbi
@ 2011-07-28 14:18   ` Arno Steffen
  2011-07-28 14:33     ` Felipe Balbi
  2011-07-29  1:44     ` Tapani Utriainen
  0 siblings, 2 replies; 10+ messages in thread
From: Arno Steffen @ 2011-07-28 14:18 UTC (permalink / raw)
  To: balbi, gary; +Cc: linux-omap

That has been good points. I've tried both, but with no result so far.

boot_nfs_nand=tftpboot 0x80000000 uImage; set bootargs
${bootargs_fs1}; run addip; bootm 0x80000000 ethaddr=00:11:22:33:44:55
bootcmd=run boot_nfs_nand
bootargs=mem=128M console=ttyO2,115200 noinitrd root=/dev/mtdblock6
rootfstype=jffs2
bootargs_fs1=mem=128M console=ttyO2,115200 noinitrd
root=/dev/mtdblock6 rootfstype=jffs2 debug earlyprintk
- Arno


2011/7/28 Felipe Balbi <balbi@ti.com>:
> On Thu, Jul 28, 2011 at 03:15:20PM +0200, Arno Steffen wrote:
>> Maybe you are so kind to give me a hint:
>>
>> Try to migrate from Kernel 33 to 37. Currently I don't see any message after
>>
>>   Verifying Checksum ... OK
>>    Loading Kernel Image ... OK
>> OK
>>
>> Starting kernel ...
>>
>> From what I understand kernel starts with start_kernel()).
>> What is best way to find out which part is failing? At which point I
>> do a printk to see what's going on, or at least set a GPIO Pin of the
>> processor?
>> How to debug without any external tools like JTAG.
>
> try to enable CONFIG_DEBUG_LL and CONFIG_EARLY_PRINTK, then add "debug
> earlyprintk" to your command line. See if that helps showing more debug
> messages.
>
> --
> balbi
>
--
To unsubscribe from this list: send the line "unsubscribe linux-omap" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

^ permalink raw reply	[flat|nested] 10+ messages in thread

* Re: [Q] No message from Kernel (Howto start debug?)
  2011-07-28 14:18   ` Arno Steffen
@ 2011-07-28 14:33     ` Felipe Balbi
  2011-07-29  1:44     ` Tapani Utriainen
  1 sibling, 0 replies; 10+ messages in thread
From: Felipe Balbi @ 2011-07-28 14:33 UTC (permalink / raw)
  To: Arno Steffen; +Cc: balbi, gary, linux-omap

[-- Attachment #1: Type: text/plain, Size: 865 bytes --]

Hi,

(please don't top-post)

On Thu, Jul 28, 2011 at 04:18:51PM +0200, Arno Steffen wrote:
> That has been good points. I've tried both, but with no result so far.
> 
> boot_nfs_nand=tftpboot 0x80000000 uImage; set bootargs
> ${bootargs_fs1}; run addip; bootm 0x80000000 ethaddr=00:11:22:33:44:55
> bootcmd=run boot_nfs_nand
> bootargs=mem=128M console=ttyO2,115200 noinitrd root=/dev/mtdblock6
> rootfstype=jffs2
> bootargs_fs1=mem=128M console=ttyO2,115200 noinitrd
> root=/dev/mtdblock6 rootfstype=jffs2 debug earlyprintk

Dumb question, did you recompile your kernel with those enabled ?

If nothing comes out, then you might need the help of a JTAG debugger.
You might want to try openocd though as it's open source and there are
many cheap JTAG debuggers which work with that (Flyswatter,
arm-usb-ocd, arm-usb-ocd-h, etc).

-- 
balbi

[-- Attachment #2: Digital signature --]
[-- Type: application/pgp-signature, Size: 490 bytes --]

^ permalink raw reply	[flat|nested] 10+ messages in thread

* RE: [Q] No message from Kernel (Howto start debug?)
  2011-07-28 13:15 [Q] No message from Kernel (Howto start debug?) Arno Steffen
  2011-07-28 13:25 ` Gary Thomas
  2011-07-28 13:26 ` Felipe Balbi
@ 2011-07-28 14:56 ` Sricharan R
       [not found]   ` <CAACX+R37F4G9To3QT97-jODcdmgTiE-9_cTvoS1Z+89n-N_icg@mail.gmail.com>
  2 siblings, 1 reply; 10+ messages in thread
From: Sricharan R @ 2011-07-28 14:56 UTC (permalink / raw)
  To: Arno Steffen, linux-omap

>-----Original Message-----
>From: linux-omap-owner@vger.kernel.org [mailto:linux-omap-
>owner@vger.kernel.org] On Behalf Of Arno Steffen
>Sent: Thursday, July 28, 2011 6:45 PM
>To: linux-omap@vger.kernel.org
>Subject: [Q] No message from Kernel (Howto start debug?)
>
>Maybe you are so kind to give me a hint:
>
>Try to migrate from Kernel 33 to 37. Currently I don't see any message
>after
>
>  Verifying Checksum ... OK
>   Loading Kernel Image ... OK
>OK
>
>Starting kernel ...
>
 If you are stuck with Starting kernel ... and do not see
   "Uncompressing Linux... done, booting the kernel" means could have
crashed before your kernel in case of uImage.
 Check the arch_id that you are passing from the bootloader.

Also enable early printk.

>From what I understand kernel starts with start_kernel()).
>What is best way to find out which part is failing? At which point I
>do a printk to see what's going on, or at least set a GPIO Pin of the
>processor?
>How to debug without any external tools like JTAG.
>--
>To unsubscribe from this list: send the line "unsubscribe linux-omap" in
>the body of a message to majordomo@vger.kernel.org
>More majordomo info at  http://vger.kernel.org/majordomo-info.html

Thanks,
 Sricharan

^ permalink raw reply	[flat|nested] 10+ messages in thread

* Re: [Q] No message from Kernel (Howto start debug?)
       [not found]       ` <accf999739ee7850d7e468c522a2342f@mail.gmail.com>
@ 2011-07-28 16:13         ` Arno Steffen
  0 siblings, 0 replies; 10+ messages in thread
From: Arno Steffen @ 2011-07-28 16:13 UTC (permalink / raw)
  To: Sricharan R, linux-omap

2011/7/28 Sricharan R <r.sricharan@ti.com>:
>>-----Original Message-----
>>From: Arno Steffen [mailto:arno.steffen@googlemail.com]
>>Sent: Thursday, July 28, 2011 9:13 PM
>>To: Sricharan R
>>Subject: Re: [Q] No message from Kernel (Howto start debug?)
>>
>>2011/7/28 Arno Steffen <arno.steffen@googlemail.com>:
>>> 2011/7/28 Sricharan R <r.sricharan@ti.com>:
>>>>  If you are stuck with Starting kernel ... and do not see
>>>>   "Uncompressing Linux... done, booting the kernel" means could have
>>>> crashed before your kernel in case of uImage.
>>>>  Check the arch_id that you are passing from the bootloader.
>>>>
>>>> Also enable early printk.
>>>>
>>>> >From what I understand kernel starts with start_kernel()).
>>>>>What is best way to find out which part is failing? At which point I
>>>>>do a printk to see what's going on, or at least set a GPIO Pin of the
>>>>>processor?
>>>>>How to debug without any external tools like JTAG.
>>>>>--
>>>>>To unsubscribe from this list: send the line "unsubscribe linux-omap"
> in
>>>>>the body of a message to majordomo@vger.kernel.org
>>>>>More majordomo info at  http://vger.kernel.org/majordomo-info.html
>>>>
>>>> Thanks,
>>>>  Sricharan
>>>>
>>>
>>> You are absolutly right! The strange thing is, uboot can boot my old
>>> kernel out of NAND flash.
>>> And with the old kernel I can also boot via ethernet. Yepp, probably
>>> the arch-id. I will check this.
>>> Thanks
>>> Arno
>>>
>>
>>To optimistic. :(
>>It seems that I have set the MACHINE_START in a proper way.
>>What can prevent this message in uboot? Why it can uncompressing the
>>old uImages, but not the new one?
>
> Is the uncompressing done by u-boot in your case?
>
> I have seen that the uncompress algorithm is built on
> Top the Image. U-boot simply jumps to that location and
> Rest is taken care by the uncompresser. It also sets up the
> Serial port according to the arch
> and prints " Uncompressing Linux... done, booting the kernel".
>
> The serial port number is used for early prints by kernel.
>
> Also check if your load address passed from u-boot is correct?
>
>>Seems I have to dig into uboot.
>>Thanks
>>-Arno
>

Mh, so it's not a uboot issue. Checking the outputs (see below) are
identical, except Linux version and the new image is a bit smaller.
Where does the uncompressor knows about which serial port to use?

Bytes transferred = 1587592 (183988 hex)
## Booting kernel from Legacy Image at 80000000 ...
   Image Name:   Linux-2.6.37
   Image Type:   ARM Linux Kernel Image (uncompressed)
   Data Size:    1587528 Bytes =  1.5 MB
   Load Address: 80008000
   Entry Point:  80008000
   Verifying Checksum ... OK
   Loading Kernel Image ... OK
OK

Thanks
-Arno
--
To unsubscribe from this list: send the line "unsubscribe linux-omap" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

^ permalink raw reply	[flat|nested] 10+ messages in thread

* Re: [Q] No message from Kernel (Howto start debug?)
  2011-07-28 14:18   ` Arno Steffen
  2011-07-28 14:33     ` Felipe Balbi
@ 2011-07-29  1:44     ` Tapani Utriainen
  2011-07-29 10:20       ` Arno Steffen
  1 sibling, 1 reply; 10+ messages in thread
From: Tapani Utriainen @ 2011-07-29  1:44 UTC (permalink / raw)
  To: Arno Steffen; +Cc: linux-omap

On Thu, 28 Jul 2011 16:18:51 +0200
Arno Steffen <arno.steffen@googlemail.com> wrote:

> That has been good points. I've tried both, but with no result so far.
> 

Try appending earlyprintk=${console} to your bootargs (where ${console}
is your console string, e.g. ttyO0,115200n8 )

---
Tapani Utriainen, PhD
Senior Software Engineer
TechNexion Ltd, www.technexion.com


^ permalink raw reply	[flat|nested] 10+ messages in thread

* Re: [Q] No message from Kernel (Howto start debug?)
  2011-07-29  1:44     ` Tapani Utriainen
@ 2011-07-29 10:20       ` Arno Steffen
  2011-08-01  8:15         ` Arno Steffen
  0 siblings, 1 reply; 10+ messages in thread
From: Arno Steffen @ 2011-07-29 10:20 UTC (permalink / raw)
  To: Tapani Utriainen, Sricharan R; +Cc: linux-omap

2011/7/29 Tapani Utriainen <tapani@technexion.com>:
> On Thu, 28 Jul 2011 16:18:51 +0200
> Arno Steffen <arno.steffen@googlemail.com> wrote:
>
>> That has been good points. I've tried both, but with no result so far.
>>
>
> Try appending earlyprintk=${console} to your bootargs (where ${console}
> is your console string, e.g. ttyO0,115200n8 )
>
> ---
> Tapani Utriainen, PhD
> Senior Software Engineer
> TechNexion Ltd, www.technexion.com
>
>

Tried this without a change. I really think this is a machine-id problem.
So although I have done same changes for my board as in my previous 33
kernel, it might be not enough to cover the machine-id. Maybe I can
overule this for test to find out the source of this misbehaviour.
The problem is already before the first debug messages are coming.
Sricharan is right, first the message uncompress has to be printed.
Thanks
- Arno

^ permalink raw reply	[flat|nested] 10+ messages in thread

* Re: [Q] No message from Kernel (Howto start debug?)
  2011-07-29 10:20       ` Arno Steffen
@ 2011-08-01  8:15         ` Arno Steffen
  0 siblings, 0 replies; 10+ messages in thread
From: Arno Steffen @ 2011-08-01  8:15 UTC (permalink / raw)
  To: Tapani Utriainen, Sricharan R; +Cc: linux-omap

2011/7/29 Arno Steffen <arno.steffen@googlemail.com>:
> 2011/7/29 Tapani Utriainen <tapani@technexion.com>:
>> On Thu, 28 Jul 2011 16:18:51 +0200
>> Arno Steffen <arno.steffen@googlemail.com> wrote:
>>
>>> That has been good points. I've tried both, but with no result so far.
>>>
>>
>> Try appending earlyprintk=${console} to your bootargs (where ${console}
>> is your console string, e.g. ttyO0,115200n8 )
>>
>> ---
>> Tapani Utriainen, PhD
>> Senior Software Engineer
>> TechNexion Ltd, www.technexion.com
>>
>>
>
> Tried this without a change. I really think this is a machine-id problem.
> So although I have done same changes for my board as in my previous 33
> kernel, it might be not enough to cover the machine-id. Maybe I can
> overule this for test to find out the source of this misbehaviour.
> The problem is already before the first debug messages are coming.
> Sricharan is right, first the message uncompress has to be printed.
> Thanks
> - Arno
>

Did some test. So with old I can always see the message "Uncompressing
Linux... done, booting the kernel."
Even if machine-id is not match or if bootargs are wrong regarding ttyS/ttyO.
With new kernel I can't see this message. So the question: Is
something changes there in uncompressing?
I am also a bit confused that I get this:

## Booting kernel from Legacy Image at 80000000 ...
   Image Name:   Linux-2.6.33
   Image Type:   ARM Linux Kernel Image (uncompressed)
   Data Size:    1505256 Bytes = 1.4 MiB
   Load Address: 80008000
   Entry Point:  80008000
   Verifying Checksum ... OK
   Loading Kernel Image ... OK
OK

and the kernel still has to be uncompressed?
- Arno

^ permalink raw reply	[flat|nested] 10+ messages in thread

end of thread, other threads:[~2011-08-01  8:15 UTC | newest]

Thread overview: 10+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2011-07-28 13:15 [Q] No message from Kernel (Howto start debug?) Arno Steffen
2011-07-28 13:25 ` Gary Thomas
2011-07-28 13:26 ` Felipe Balbi
2011-07-28 14:18   ` Arno Steffen
2011-07-28 14:33     ` Felipe Balbi
2011-07-29  1:44     ` Tapani Utriainen
2011-07-29 10:20       ` Arno Steffen
2011-08-01  8:15         ` Arno Steffen
2011-07-28 14:56 ` Sricharan R
     [not found]   ` <CAACX+R37F4G9To3QT97-jODcdmgTiE-9_cTvoS1Z+89n-N_icg@mail.gmail.com>
     [not found]     ` <CAACX+R1Qtzn1uTX4aHg1cSRuBR+29_Tek08RhTU2DVmiH7UP3Q@mail.gmail.com>
     [not found]       ` <accf999739ee7850d7e468c522a2342f@mail.gmail.com>
2011-07-28 16:13         ` Arno Steffen

This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.