Genode Fiasco.OC kernel changes

Sebastian Sumpf Sebastian.Sumpf at ...1...
Thu Jul 18 22:03:56 CEST 2013


Hi Daniel,

On 07/17/2013 11:49 PM, Daniel Waddington wrote:

> On 07/17/2013 01:05 PM, Sebastian Sumpf wrote:
>> Hi Daniel,
>>
>> On 07/17/2013 09:41 PM, Daniel Waddington wrote:
>>
>>> On 07/17/2013 12:13 PM, Sebastian Sumpf wrote:
>>>>> Hi,
>>>>> Can someone point me to the relevant Fiasco.OC kernel changes made by
>>>>> Genode that might prevent the following error...
>>>>>
>>>>> Could not assign parent interface to new PD
>>>>> unknown exception?
>>>>>
>>>>>
>>>>> This is from vanilla r56 of Fiasco.OC.  I'd like to see if I can get r56
>>>>> running on Genode (to help me fix other issues).
>>>>>
>>>>
>>>> I am already at it, I can push a preliminary branch of r56 tomorrow, if
>>>> you like, so you can check if the error still occurs.
>>
>>>>> Yes, this would be a fantastic help.
>>
>> Guess I'm quicker than expected .-) The relevant branch is:
>>
>> https://github.com/ssumpf/genode/tree/r56
>>
>> Make sure to run 'make clean' and 'make prepare' in 'base-foc'. If the
>> error still occurs (I guess it's on x86_64), I would appreciate any help
>> on how to reproduce it.
>>
> Thanks for the code. I can confirm that it builds and runs fine (x86_64).
> 
> Unfortunately I still have my problem, but now I think it may have 
> something to do with the kernel's internal Ram_quota for the process. 
> Do someone at Genode know how this quota gets established?
> 

Have a look at 'base-foc/contrib/kernel/fiasco/src/kern/config.cpp'.
There is an enum value called 'kernel_mem_max' as well as constant
called 'kernel_mem_per_cent'. Both should have an influence on how much
memory the in kernel buddy allocator can use.

Greetings,

Sebastian

-- 
Sebastian Sumpf
Genode Labs

http://www.genode-labs.com · http://genode.org

Genode Labs GmbH · Amtsgericht Dresden · HRB 28424 · Sitz Dresden
Geschäftsführer: Dr.-Ing. Norman Feske, Christian Helmuth







More information about the users mailing list