[PATCH 3/3] C/R: Basic support for network namespaces and devices

Oren Laadan orenl at cs.columbia.edu
Thu Jan 21 08:08:17 PST 2010


On Thu, 21 Jan 2010, Dan Smith wrote:

> BH> What about:
> 
> BH>     1. Multicast
> BH>     2. Device config info (ipv4_devconf)
> 
> <snip>
> 
> BH> What about addr->ifa_flags and all the other elements like
> BH> prefixlen, scope and label?
> 
> I thought I was covered by calling it "Basic support..." :)
> 
> I've added these to the list of TODOs and will cook those up once the
> basic bits are in.
> 
> >> +	if ((ret >= 0) && (ret != h->inet4_addrs)) {
> >> +		ret = -EBUSY;
> >> +		ckpt_err(ctx, ret,
> >> +			 "Addresses on interface %s changed\n", dev->name);
> >> +		goto out;
> >> +	}
> 
> BH> This isn't guaranteed to catch every change to the address list,
> BH> just that the number of addresses is the same, is there no way to
> BH> hold a lock the whole time?
> 
> Yeah, there probably is.  I'll take a look.

Qouting from your reply to my email:

> OL> Is there a reason not to collect all addresses into one buffer
> OL> (can there be more than a page worth of them ?) and write in one
> OL> go ?
>
> I don't really see anywhere that the list is bounded.  I'd say that in
> most cases each interface will only have one address anyway.

So collecting them in one buffer (if one page isn't enough then realloc
and retry) would solve this too.

Oren


More information about the Containers mailing list