Hi Ben, Thanks for reviewing our repos.
The user space code that you are seeing is based on our port (the one we submitted some months ago to this ML, before Guru's Windows patching work started on master), that's why you see all those commits. The good news is that thanks to the way we approached the Windows Netlink replacement, rebasing the porting on the current master tip will be very fast, max one week as written in a previous email, resulting in a way smaller diff since master contains already other patches for Windows. Before getting there, do you think it could be possible to agree on at least some of the general design ideas of the porting by looking at both the Cloudbase and VMWare efforts? The user space integration (netlink replacement) in particular is one area where the two approaches differ quite a bit, please see the comparison that I sent earlier to this ML [1]. I'd be happy to participate in a IRC chat or hangout to discuss the general Windows kernel port design principles and implementation details. Thanks, Alessandro [1] http://openvswitch.org/pipermail/dev/2014-June/042118.html > On 26/giu/2014, at 03:21, "Ben Pfaff" <b...@nicira.com> wrote: > >> On Wed, Jun 25, 2014 at 5:04 PM, Ben Pfaff <b...@nicira.com> wrote: >> On Tue, Jun 10, 2014 at 2:06 PM, Alessandro Pilotti >> <apilo...@cloudbasesolutions.com> wrote: >>> https://github.com/cloudbase/openvswitch-hyperv >>> https://github.com/cloudbase/openvswitch-hyperv-kernel >> >> Thanks a lot for sending this out. I've just now started taking a >> first look at it. >> >> I see that there are 216 commits beyond the base OVS code. I hope that you >> are planning to reduce the number. It is not really practical to review them >> individually when they are organized that way. (It's not going to stop me >> from >> an initial review--I'll just read the overall diff.) > > Actually it's very difficult to read the overall diff, which randomly > changes lots of > files: > 509 files changed, 109551 insertions(+), 47618 deletions(-) > > It also adds all of the headers from openssl and a windows pthread library, > plus binaries for both libraries. > > I think you're going to need to clean this up a lot to make it reviewable. _______________________________________________ dev mailing list dev@openvswitch.org http://openvswitch.org/mailman/listinfo/dev
