Hi Sage, Check this[1], we should I do next, entity_addrvec_t ? [1] https://github.com/zhjwpku/ceph/commits/wip-addr-work On Tue, May 17, 2016 at 12:29 AM, Sage Weil <sage@xxxxxxxxxxxx> wrote: > On Mon, 16 May 2016, Sage Weil wrote: >> Hi Junwang, >> >> On Mon, 16 May 2016, Junwang Zhao wrote: >> > Hi Sage, Haomai, >> > >> > I am working on the require-features-patch, I use 'make' to see the conflicts, >> > and change the code to fit the required-feature-patch, [1] is a huge patch >> > that I am still working on, it has not been finished. I really need to check >> > with you whether I am doing it right, since it seems the errors are endless. >> > >> > There are some comments where I am not sure in the patch, like 'not sure'. >> >> I skimmed through this and it looks mostly right, but I see a few cases >> where features aren't needed, e.g. cls_refcount_get in >> cls_refcount_client.cc (there's no addr being encoded in >> cls_refcount_get_op, so no need to make the encoding featureful). >> >> > I didn't split this huge patch into small ones, I am not sure is that a >> > must, if yes, I will split it into small ones. >> >> The end result needs to be a series of small patches, but that doesn't >> have to happen right away. I think it might be useful to do a few small >> sets of changes first just to show what the goal is, though. >> >> I will take the patch below and pull a few sample changes out so you can >> see. Traveling at the moment, but I'll have something pushed today (that >> is also rebased on top of the latest wip-addr-cleanup branch). > > I went on a bit of a binge and have a wip-addr-work branch that is mostly > there. The last thing (I think) is updating msg/simple and msg/async. > These ones should explicitly encode with features 0 since the protocol is > defined in terms of the legacy encoding. > > sage -- To unsubscribe from this list: send the line "unsubscribe ceph-devel" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
