Re: bug #44412

[David Riley <fraveydank%gmail.com@localhost>]

I have the original patch I made against 5.1, if that would help; I don't think 
it was really different from -current (I don't think the file changed at all).

At some point (my wife is likely to go into labor in the next few hours/days, 
so probably not right now), it would be good to try it on your non-working i386 
with some debug printfs thrown in.

The problem I had seen was that broadcast requests weren't getting 
automatically routed to the loopback interface as well, so the machine couldn't 
see its own broadcasts (necessary when trying to lookup an address, though I 
think netatalk does it the "routed" way all the time, which is maybe not 
ideal).  What version of Netatalk (user-land) are you using?


- Dave


On Aug 3, 2011, at 3:21 PM, Hauke Fath wrote:

> At 16:48 Uhr +0200 3.8.2011, Ignatios Souvatzis wrote:
>> On Wed, Aug 03, 2011 at 04:41:01PM +0200, Thomas Klausner wrote:
>>> On Fri, Jul 22, 2011 at 02:55:21PM -0400, David Riley wrote:
>>>> I filed this report a while back.  Someone else has tested my fix on
>>> non-PPC systems (x86, x86_64) and reported that it seems to work as well.
>>> I'm attaching the patch against -current here; could someone give it a
>>> look and include it if it seems to be OK?  Or provide feedback if it
>>> doesn't.
>>> 
>>> Thanks for the bug report and patch.
>>> I'm not using netatalk and can't judge the patch; until someone comes
>>> along that can, please add the patch to the bug report as well, so
>>> it's not lost.
>> 
>> I can look at this; I have an Appletalk printer in occasional
>> production us.
> 
> Please do.
> 
> I applied the patch a few days ago to netbsd-5 sources (i386) and -current
> sources (sparc). It didn't fix the problem on i386, and didn't break
> anything on sparc (where things worked before with Netatalk 2.0).
> 
> From what I have seen, AppleTalk worked fine up to netbsd-4, and broke,
> depending on the network interface, on netbsd-5. My test matrix is
> 
>                 sparc          i386
>               le   hme        wm   re
> netbsd-4        +     +         +    ?
> netbsd-5        +     -         -    -
> netbsd-HEAD     +     +         ?    ?
> 
> -- guess I should try a patched -current kernel on i386, and a patched
> netbsd-5 kernel on sparc.
> 
>       hauke
> 
> --
> "It's never straight up and down"     (DEVO)
> 
>