[Adeos-main] port to s3c2440 / mini2440

[Adeos-main] port to s3c2440 / mini2440

[Michel He]

hello all,

    Is there anyone who managed to port Adeos I-Pipe to the arm9  
S3C2440 of the Mini2440 board (Friendly-ARM) ?

    The boot is blocked in the __delay function. Something maybe wrong  
with the interrupt scheduling ...

thank you

Re: [Xenomai-help] [Adeos-main] port to s3c2440 / mini2440

[Gilles Chanteperdrix]

Michel He wrote:
> hello all,
> 
>     Is there anyone who managed to port Adeos I-Pipe to the arm9  
> S3C2440 of the Mini2440 board (Friendly-ARM) ?

The s3c24xx support is still in the I-pipe patch, if it was not, the
compilation of the kernel with CONFIG_IPIPE would not even work.

> 
>     The boot is blocked in the __delay function. Something maybe wrong  
> with the interrupt scheduling ...

This port has been contributed a long time ago. Since then, we use an
"optimistic testing strategy", we make sure that the port compiles, but
 we assume people really having that board are in charge of testing our
releases. Not having a board based on this chip, it would be hard for us
to test it. IOW, I have no idea whether it is supposed to work.

I suggest you assume nothing, read:
http://www.xenomai.org/index.php/I-pipe:ArmPorting

And try and debug your port. It even contains a tip for the particular
case you are mentioning.

-- 
					    Gilles.

Re: [Adeos-main] port to s3c2440 / mini2440

[Michel He]

Ok many thanks for the tips.

Refering to the webpage, actually the cause should be :
     *  If the kernel locks at "Calibrating delay loop...", it means  
that the timer interrupt is not ticking and that the delay calibration  
routine is running an infinite loop at while (ticks == jiffies) in  
function calibrate_delay, file init/calibrate.c. To help debugging  
this situation, you can put printks in the while (ticks == jiffies)  
loop, printing any hardware timer registers or status register you want.

It seems that S3C2440 / Mini2440 Timer has changed. There is no more  
register named S3C2410_T...

But I have no experience in debugging ARM cpu hardware timer... Nor  
have time to investigate further how to debug the port...

You can still find the kernel on FriendlyARM.net



Gilles Chanteperdrix <gilles.chanteperdrix@xenomai.org> a écrit :

> Michel He wrote:
>> hello all,
>>
>>     Is there anyone who managed to port Adeos I-Pipe to the arm9
>> S3C2440 of the Mini2440 board (Friendly-ARM) ?
>
> The s3c24xx support is still in the I-pipe patch, if it was not, the
> compilation of the kernel with CONFIG_IPIPE would not even work.
>
>>
>>     The boot is blocked in the __delay function. Something maybe wrong
>> with the interrupt scheduling ...
>
> This port has been contributed a long time ago. Since then, we use an
> "optimistic testing strategy", we make sure that the port compiles, but
>  we assume people really having that board are in charge of testing our
> releases. Not having a board based on this chip, it would be hard for us
> to test it. IOW, I have no idea whether it is supposed to work.
>
> I suggest you assume nothing, read:
> http://www.xenomai.org/index.php/I-pipe:ArmPorting
>
> And try and debug your port. It even contains a tip for the particular
> case you are mentioning.
>
> --
> 					    Gilles.
>

Re: [Adeos-main] port to s3c2440 / mini2440

[Gilles Chanteperdrix]

Michel He wrote:
> Ok many thanks for the tips.
> 
> Refering to the webpage, actually the cause should be :
>      *  If the kernel locks at "Calibrating delay loop...", it means  
> that the timer interrupt is not ticking and that the delay calibration  
> routine is running an infinite loop at while (ticks == jiffies) in  
> function calibrate_delay, file init/calibrate.c. To help debugging  
> this situation, you can put printks in the while (ticks == jiffies)  
> loop, printing any hardware timer registers or status register you want.
> 
> It seems that S3C2440 / Mini2440 Timer has changed. There is no more  
> register named S3C2410_T...

If a define had changed name, the I-pipe support would not even compile.
However, it still compiles. So, the problem is probably not as simple as
this.

> 
> But I have no experience in debugging ARM cpu hardware timer... Nor  
> have time to investigate further how to debug the port...
> 
> You can still find the kernel on FriendlyARM.net

I have neither time, nor the hardware to check. So, if you are
interested in the port, I am afraid you will have to debug it yourself.

-- 
					    Gilles.