Dawid Ciezarkiewicz <[EMAIL PROTECTED]> wrote:
[...]
>Oh. I'm quite puzzled here. What is current policy? I'd like sysfs interfaces
>better than ioctl - they are much cleaner etc. - but I thought ioctl will be
>better here because current bonding control uses ioctl and extending it is
>much simpler.
The bonding control stuff is (slowly but surely) moving to
sysfs. There really isn't anything you can't do to bonding via sysfs
now; the only real issue is that it's difficult to get success / failure
status from a sysfs request.
>More generally: should I use sysfs always when adding anything to kernel and
>use it even when extending older functionality?
Well, I can't speak for the kernel as a whole, but for bonding,
I see no reason to add any new functionality to ifenslave at this point
in time, including new ioctl paths into the driver.
>If wrr bonding have chances for inclusion into mainline I'll do any necessary
>changes to let it meet requirements. I'd like to know if functionality from
>Andy's patch is desired to be a part of such round-robin bonding extension. I
>think I could easily integrate it with wrr.
I'm unfamiliar with Andy's patch (other than his description).
If there's interest in, and justification for, functionality to
manipulate the way the round-robin mode doles out packets it would be
best to have a single, generic dingus rather that two special-purpose
thingies.
-J
---
-Jay Vosburgh, IBM Linux Technology Center, [EMAIL PROTECTED]
-
To unsubscribe from this list: send the line "unsubscribe netdev" in
the body of a message to [EMAIL PROTECTED]
More majordomo info at http://vger.kernel.org/majordomo-info.html
