[ECOS] global addressing in ecos

[matthew de veza]

We've applied the patch and re-run all the tests for the stateless address 
autoconfiguration, and we also exempted the tests indicated below (21-30, 
54-56). The tests that passed are those that wait for the DAD to fail (it 
wants the target to basically not send any echo reply/packet) and all the 
other tests failed. These are basically the tests that wants the target to 
echo-reply. We believe that the cause is that the global address cannot be 
set; it would seem that the target does not respond to DAD which are sent 
using a multicast address.

This seems to lead me into thinking that the entire set of stateless 
autoconfiguration is exempted. Can anyone please verify? Or on another train 
of thought, I would like to ask if anyone would know why we cannot set the 
global address, and why the target does not seem to respond to multicast 
queries.

I would like to ask Mr. Hamilton if you could share your test results for 
the ecos target tested by TAHI if this is not confidential so we could more 
or less compare our test results with yours.


Thank you.
Matt



>From: Jonathan Larmour <jifl@eCosCentric.com>
>To: Mark Salter <msalter@redhat.com>
>CC: gary.thomas@mind.be, 
>imdeveza@hotmail.com,ecos-discuss@sources.redhat.com
>Subject: Re: [ECOS] global addressing in ecos
>Date: Thu, 09 Jan 2003 16:52:07 +0000
>
>Mark Salter wrote:
>>>>>>>On Wed, 2003-01-08 at 04:53, matthew de veza wrote:
>>>>>>>
>>>>>>>
>>>>>>>>we've been testing an ecos target with the tahi test suite and the 
>>>>>>>>tests which involves autoconfiguration of global addresses seem to 
>>>>>>>>all be unsuccessful. i would really like to find out if the global 
>>>>>>>>address autoconfiguration feature/stateless address 
>>>>>>>>autoconfiguration feature which was ported from freebsd to ecos has 
>>>>>>>>been fully ported and is robust and functional. can anyone please 
>>>>>>>>help me?
>>>>>>>
>>
>>This patch cleared up a lot of TAHI failures for us:
>>
>>   http://sources.redhat.com/ml/ecos-patches/2002-08/msg00078.html
>>
>>AFAICT, all of our patches are in the public CVS.
>
>Thanks for checking that.
>
>>>From a report on the test run:
>>
>>   ============================================
>>    IPv6 Stateless Address Autoconfiguration
>>   ============================================
>>
>>   Originally failed tests should now pass.  Reason: new
>>   ipv6_routing_thread does not emit unexpected packets.
>>
>>   Exempt tests: #21-#30, #54-#56
>>   Reason:  Not applicable: Infrastructure missing, no 'ifconfig',
>>   'route' or 'setkey' utilities, so even in manual mode, boot-time
>>   config is not possible in eCos.
>
>Ah, I didn't remember/know there were exemptions. At a guess the 
>autoconfiguration stuff must be one of those, but I'm sure Matthew will say 
>if not.
>
>Jifl
>--
>eCosCentric       http://www.eCosCentric.com/       <info@eCosCentric.com>
>--[ "You can complain because roses have thorns, or you ]--
>--[  can rejoice because thorns have roses." -Lincoln   ]-- Opinions==mine


_________________________________________________________________
The new MSN 8 is here: Try it free* for 2 months 
http://join.msn.com/?page=dept/dialup