Re: suspend2 merge (was Re: [Suspend2-devel] Re: CFS and suspend2: hang in atomic copy)

[David Lang <david.lang@digitalinsight.com>]

On Thu, 26 Apr 2007, Rafael J. Wysocki wrote:

> > The general scheme has been working for four or five years - if there
> > was a fundamental issue, we would have found it by now.
> >
> > The scheme isn't complicated.
>
> Conceptually, it is complicated just because you're using the LRU.

I know that I've seen many projects that are working on or claim to have 
suceeded in being able to do live migration of processes from one system to 
another. has anyone looked at useing any of these mechanisms for snapshoting the 
user processes for the std situation? if you can do this a process at a time you 
may be able to avoid the massive blob of a write

instead of what linus was saying

buff = snapshot()
write(buff)

it would be
start_snapshot()   /* stops all userspace schedulers except for this process */
foreach(pid) {
  buff = snapshotpid(pid)
  write(buff)
}

David Lang
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/