Re: [PATCH 0/3] bonding: 3 fixes for 2.6.24

[Jay Vosburgh <fubar@us.ibm.com>]

Herbert Xu <herbert@gondor.apana.org.au> wrote:

>On Thu, Jan 10, 2008 at 04:03:53PM -0500, Andy Gospodarek wrote:
>>
>> > >Sure, but where do you call that function while holding the bond lock?
>> > 
>> > 	If I recall correctly, the problem was that tg3, et al, did
>> > things that might sleep, and bonding was calling from a timer context,
>> > which couldn't sleep.  It wasn't about the lock.
>> 
>> Exactly, I was just about to post the same.
>
>In other words, changing read_lock on bond->lock to read_lock_bh doesn't
>affect this one bit.

	For the case of the bond_set_multicast_list function, changing
the existing write_lock to a read_lock_bh doesn't affect any calls to
dev_set_mac_address (since it isn't called from there) and, apparently,
also resolves the lockdep warning.  I'm still trying to get lockdep to
generate the warning for me locally; I'm not sure which magic thing I'm
missing.

	-J

---
	-Jay Vosburgh, IBM Linux Technology Center, fubar@us.ibm.com