Re: [PATCH] bonding: documentation and code cleanup for resend_igmp

[Flavio Leitner <fbl@redhat.com>]

On 05/25/2011 02:19 PM, Rick Jones wrote:
> On Wed, 2011-05-25 at 11:44 -0300, Flavio Leitner wrote:
>> Improves the documentation about how IGMP resend parameter
>> works, fix two missing checks and coding style issues.
>>
>> Signed-off-by: Flavio Leitner <fbl@redhat.com>
>> ---
>>  Documentation/networking/bonding.txt |   13 +++++++++++--
>>  drivers/net/bonding/bond_main.c      |   12 +++++++-----
>>  drivers/net/bonding/bond_sysfs.c     |   10 +++++-----
>>  3 files changed, 23 insertions(+), 12 deletions(-)
>>
>> diff --git a/Documentation/networking/bonding.txt b/Documentation/networking/bonding.txt
>> index 1f45bd8..683ef76 100644
>> --- a/Documentation/networking/bonding.txt
>> +++ b/Documentation/networking/bonding.txt
>> @@ -770,8 +770,17 @@ resend_igmp
>>  	a failover event. One membership report is issued immediately after
>>  	the failover, subsequent packets are sent in each 200ms interval.
>>  
>> -	The valid range is 0 - 255; the default value is 1. This option
>> -	was added for bonding version 3.7.0.
>> +	The valid range is 0 - 255; the default value is 1. A value of 0
>> +	prevents the IGMP membership report to be issued due the failover
>> +	event.
> 
> Grammar nit - "prevents the ICMP membership report from being issued in
> response to the failover event."  Or "prevents issuing of the IGMP
> membership report in response to a failover event."

I like the first suggestion. I'll update the patch fixing the ICMP
typo.

>> +
>> +	This option is useful for bonding modes balance-rr or 0,
>> +	active-backup or 1, balance-tlb or 5, balance-alb or 6, which a
>> +	failover can move the IGMP traffic from one slave to another.
>> +	Therefore, the switch must be notified with an extra IGMP report
>> +	to start forwarding the IGMP traffic over the new selected slave.
> 
> More nits.  How about (with some added guesses on my part about
> direction)
> 
> This option is useful for bonding modes balance-rr (0), active-backup
> (1), balance-tlb (5) and balance-alb (6), in which a failover can switch
> the outgoing IGMP traffic from one slave to another.  Therefore a fresh
> IGMP report must be issued to cause the switch to forward the incoming
> IGMP traffic over the newly selected slave.

It's also better. thanks. However, the failover can switch both
incoming and outgoing IGMP traffic so I'd leave it generic as before.
For instance:

This option is useful for bonding modes balance-rr (0), active-backup
(1), balance-tlb (5) and balance-alb (6), in which a failover can switch
the IGMP traffic from one slave to another.  Therefore a fresh IGMP
report must be issued to cause the switch to forward the incoming
IGMP traffic over the newly selected slave.

What do you think?

thanks,
fbl