l4Linux problems.

[Stefan Kalkowski]

Hi Daniel,

thank you for providing the image! When investigating the error I could
see that an exception is raised, when the first VCPU gets constructed
probably due to a memory region conflict in its thread context area.

Currently, I'm producing an improved memory model for Genode/L4Linux,
because the current subversion version implies some shortcomings that
might cause region-conflicts between L4Linux memory management and the
Genode framework, although in theory the thread context area should be
exempted hereby.

Nevertheless, I would like to share the improved version with you, as
soon as its available (I assume to be ready tomorrow). Probably this
will fix the issue.

Regards
Stefan

On 05/12/2011 06:19 PM, Daniel Waddington wrote:
> Hi Stefan,
> Do you have a google account?  - I have just shared the files on google 
> docs.
> 
> Thanks
> Daniel
> 
> On 05/12/2011 02:49 AM, Stefan Kalkowski wrote:
>> Hi Daniel,
>>
>> that's strange, I use the same run-script with the same initrd.gz. Could
>> you please somehow provide your resulting iso-image and the unstripped
>> linux-binary for debugging purposes?
>>
>> the image is located at:
>> [build-dir]/var/run/l4linux.iso
>>
>> the linux-binary at:
>> [build-dir]/bin/vmlinux
>>
>> Regards
>> Stefan
>>
>> On 05/11/2011 07:55 PM, Daniel Waddington wrote:
>>> Hi,
>>> I am testing the new L4Linux on Fiasco.OC capabilities with the
>>> initrd.gz from genode.org/files.
>>>
>>> Everything builds OK (32bit target on a 64bit host) but I run it (via
>>> 'make run/l4linux')  I get the below output.
>>>
>>> Any ideas?
>>> Daniel
>>>
>>> int main(): --- create local services ---
>>> int main(): --- start init ---
>>> int main(): transferred 450 MB to init
>>> int main(): --- init created, waiting for exit condition ---
>>> [init] Could not open file "ldso"
>>> [init ->  pci_drv] PCI driver started
>>> [init ->  ps2_drv] Detected ExPS/2 mouse - activating scroll-wheel and
>>> 5-button support.
>>> [init ->  ps2_drv] Using keyboard with scan code set 1 (xlate).
>>> [init ->  vmlinux] Region map:
>>> [init ->  vmlinux]      0x00000000 - 0x00001000 [reserved]
>>> [init ->  vmlinux]      0x00001000 - 0x00006000 [genode-heap]
>>> [init ->  vmlinux]      0x00006000 - 0x00400000 [unused]
>>> [init ->  vmlinux]      0x00400000 - 0x00754000 [Binary]
>>> [init ->  vmlinux]      0x00754000 - 0x40000000 [unused]
>>> [init ->  vmlinux]      0x40000000 - 0x50000000 [Context area]
>>> [init ->  vmlinux]      0x50000000 - 0x50100000 [l4lx-heap]
>>> [init ->  vmlinux]      0x50100000 - 0xffffffff [unused]
>>> [init ->  vmlinux] ======>  L4Linux 2.6 starting...<========
>>> [init ->  vmlinux] Linux version 2.6.37-l4+ (dwaddington at ...70...) (gcc
>>> version 4.4.5 (GCC) ) #11 Wed May 11 10:47:10 PDT 2011
>>> [init ->  vmlinux] Binary name: vmlinux
>>> [init ->  vmlinux] Linux kernel command line (3 args): mem=128M
>>> l4x_rd=initrd.gz text
>>> [init ->  vmlinux] CPU mapping (l:p)[1]: 0:0
>>> [init ->  vmlinux] Image: 00401000 - 00754000 [3404 KiB].
>>> [init ->  vmlinux] Areas: Text:     00401000 - 00680000 [2556kB] (a bit
>>> longer)
>>> [init ->  vmlinux]        Data:     00680000 - 006a01cc [128kB]
>>> [init ->  vmlinux]        Initdata: 006a8000 - 006d2000 [168kB]
>>> [init ->  vmlinux]        BSS:      006d2000 - 00743604 [453kB]
>>> [init ->  vmlinux] l4util_cpu_capabilities: Not implemented yet!
>>> [init ->  vmlinux] gs=43   fs=43
>>> [init ->  vmlinux] l4io_get_root_device: Not implemented yet!
>>> [init ->  vmlinux] Device scan:
>>> [init ->  vmlinux] l4io_iterate_devices: Not implemented yet!
>>> [init ->  vmlinux] l4x_fiasco_gdt_entry_offset = 9
>>> [init ->  vmlinux] l4lx_thread_name_set: Not implemented yet!
>>> no RM attachment (READ pf_addr=20 pf_ip=50abca from 508000)
>>> invalid signal-context capability
>>>
>>>
>>>
>>> ------------------------------------------------------------------------------
>>> Achieve unprecedented app performance and reliability
>>> What every C/C++ and Fortran developer should know.
>>> Learn how Intel has extended the reach of its next-generation tools
>>> to help boost performance applications - inlcuding clusters.
>>> http://p.sf.net/sfu/intel-dev2devmay
>>>
>>>
>>>
>>> _______________________________________________
>>> Genode-main mailing list
>>> Genode-main at lists.sourceforge.net
>>> https://lists.sourceforge.net/lists/listinfo/genode-main
> 
> ------------------------------------------------------------------------------
> Achieve unprecedented app performance and reliability
> What every C/C++ and Fortran developer should know.
> Learn how Intel has extended the reach of its next-generation tools
> to help boost performance applications - inlcuding clusters.
> http://p.sf.net/sfu/intel-dev2devmay
> _______________________________________________
> Genode-main mailing list
> Genode-main at lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/genode-main
> 

-- 
Stefan Kalkowski
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