svn commit: r318160 - in head: sbin/ifconfig sys/net sys/sys

[Ravi Pokala]

Fixed in r318176. Sorry for the breakage everyone. :-p

-Ravi (rpokala@)

-----Original Message-----
From: Ravi Pokala <rpokala at mac.com>
Date: 2017-05-10, Wednesday at 22:47
To: Kristof Provost <kristof at sigsegv.be>
Cc: <src-committers at freebsd.org>, <svn-src-all at freebsd.org>, <svn-src-head at freebsd.org>
Subject: Re: svn commit: r318160 - in head: sbin/ifconfig sys/net sys/sys

-----Original Message-----
> From: <owner-src-committers at freebsd.org> on behalf of Kristof Provost <kristof at sigsegv.be>
> Date: 2017-05-10, Wednesday at 22:38
> To: Ravi Pokala <rpokala at FreeBSD.org>
> Cc: <src-committers at freebsd.org>, <svn-src-all at freebsd.org>, <svn-src-head at freebsd.org>
> Subject: Re: svn commit: r318160 - in head: sbin/ifconfig sys/net sys/sys
> 
> On 11 May 2017, at 11:05, Kristof Provost wrote:
>> On 11 May 2017, at 3:43, Ravi Pokala wrote:
>>> Author: rpokala
>>> Date: Wed May 10 22:13:47 2017
>>> New Revision: 318160
>>> URL: https://svnweb.freebsd.org/changeset/base/318160
>>>
>>> Log:
>>>   Persistently store NIC's hardware MAC address, and add a way to 
>>> retrive it
>>>
>>
>> This seems to cause panics when I create a bridge interface:
>>
>> #10 0xffffffff80ef9304 in bcopy () at 
>> /usr/src/sys/amd64/amd64/support.S:139
>> #11 0xffffffff80b57a80 in ether_ifattach (ifp=0xfffff80035663000, 
>> lla=0xfffff800090e13f8 "\002k") at /usr/src/sys/net/if_ethersubr.c:919
>> #12 0xffffffff8325dc67 in bridge_clone_create (ifc=<value optimized 
>> out>, unit=<value optimized out>, params=<value optimized out>) at 
>> /usr/src/sys/net/if_bridge.c:704
>> ...
>>
>> ifp->if_hw_addr is NULL here:
>> (kgdb) fr 11
>> #11 0xffffffff80b57a80 in ether_ifattach (ifp=0xfffff80035663000, 
>> lla=0xfffff800090e13f8 "\002k") at /usr/src/sys/net/if_ethersubr.c:919
>> 919             bcopy(lla, ifp->if_hw_addr, ifp->if_addrlen);
>> (kgdb) p ifp->if_hw_addr
>> $1 = (void *) 0x0
>>
> 
> This may be because I’ve not yet updated world, just the kernel, but 
> clearly that still shouldn’t cause panics.
> 
> Regards,
> Kristof

No, this is purely a kernel mistake -- I forgot to add a NULL-check in ether_ifattach(). :-p

Fixing now...

-Ravi (rpokala@)