Re: [Xenomai] Xenomai 2.6.1 x86 Boot Hanging

[Gilles Chanteperdrix <gilles.chanteperdrix@xenomai.org>]

On 07/25/2012 11:35 PM, Kyle Maroney wrote:

> On 07/25/2012 05:20 PM, Gilles Chanteperdrix wrote:
>> On 07/25/2012 11:18 PM, Kyle Maroney wrote:
>>
>>> On 07/24/2012 04:55 PM, Gilles Chanteperdrix wrote:
>>>> On 07/24/2012 10:31 PM, Kyle Maroney wrote:
>>>>
>>>>> On 07/24/2012 04:10 PM, Gilles Chanteperdrix wrote:
>>>>>> On 07/19/2012 03:33 PM, Jan Kiszka wrote:
>>>>>>
>>>>>>> On 2012-07-19 15:17, Kyle Maroney wrote:
>>>>>>>> Congratulations on the recent release of Xenomai 2.6.1. We are excited
>>>>>>>> about the recent move to the ipipe core series and 3.x kernel patches
>>>>>>>> becoming available for the x86 architecture.
>>>>>>>>
>>>>>>>> I have been able to successfully patch the vanilla kernel (3.2.19) with
>>>>>>>> a slightly modified ipipe-core-3.2.21-x86-1 patch, and tested that
>>>>>>>> working with a generic configuration. I also successfully patched
>>>>>>>> vanilla + ubuntu diff + adeos, and tested this working as well.
>>>>>>>>
>>>>>>>> My problem is when I apply the xenomai configuration options after
>>>>>>>> running the prepare kernel script.  The kernel compiles successfully
>>>>>>>> with no problems, however upon reboot I am met with the boot hanging at
>>>>>>>> the purple boot screen of Ubuntu (no images).
>>>>>>>>
>>>>>>>> I have experimented with many config options for the past few days, and
>>>>>>>> come to you for guidance. Please find attached my .config file.
>>>>>>>>
>>>>>>> There are obviously several issues remaining in the new x86 3.2 patch. I
>>>>>>> just recently ran into some of them as well while updating RTnet.
>>>>>> Any chance for a patch? I am currently running 3.4 on x86, so, it would
>>>>>> be nice to merge any fix you made.
>>>>>>
>>>>> Still no luck creating a configuration allowing me to boot on x86 w/
>>>>> Precise. If anyone else has been successful on x86 we would love to see
>>>>> patches.
>>>> For the screen not to remain black, remove the "quiet" on the kernel
>>>> command line in the bootloader configuration.
>>>>
>>>> You can try earlyprintk, if the failure happens early.
>>>>
>>>> Also, in the kernel, try disabling any feature you do not use, such as
>>>> profiling, tracing, kgdb, perf system, syscall auditing, as much as
>>>> possible.
>>> Thank you all for your continued support (esp. Jan and Gilles). I am
>>> happy to report that I was successful in my compilation and testing!
>>
>> Ok, it would be nice to find which options caused the issue, so that we
>> can investigate and fix if possible.
>>
> Changes from the previously unsuccessful config:
> Disabled KGDB
> Enabled ACPI
> Disabled SPARSE_IRQ
> Disabled DEBUG_KERNEL
> 
> I hope this provides insight.


Well, it would be nice to know which one has an issue (the "options" was
a mistake, I meant "option"...)


-- 
                                                                Gilles.