On Mon, 2016-10-24 at 16:28 +0000, Sage Weil wrote: > On Mon, 24 Oct 2016, Jeff Layton wrote: > > On Mon, 2016-10-24 at 12:19 -0400, Jeff Layton wrote: > > > On Mon, 2016-10-24 at 16:16 +0000, Sage Weil wrote: > > > > > > > > Hi, > > > > > > > > This PR that changes the entity_addr encoding is finally ready to merge > > > > > > > > https://github.com/ceph/ceph/pull/9825 > > > > > > > > but it defines an MSG_ADDR2 feature bit indicating whether we support the > > > > new addr encoding. Right now it's defined with the same value as the > > > > FS_CHANGE_ATTR and FS_BTIME feature bit. Are those upstream yet? If not, > > > > we should get the addr2 support in at the same time so that we can > > > > conserve a feature bit. If not, we'll have to consume another bit. :( > > > > > > > > sage > > > > > > > > > > That support is merged into master -- is that what you mean by > > > "upstream"? > > > > > > > Sorry, I misread the subject... > > > > No, those are not yet supported or recognized in kclient yet. In which > > kernel are you planning to merge the addr2 patches? > > That's good news! :) The kclient patches for addr2 haven't been written > yet, though. There is no particular rush, but we should plan to merge > all three together so that we can use the same feature bit. > > Thanks! > sage Makes sense to me and I agree that there's no rush: The kernel doesn't care about the btime. I think we'll probably just discard it until we have a way to expose it to userland. The change_attr could be useful when exporting via knfsd, but we've lived without it so far. -- Jeff Layton <jlayton@xxxxxxxxxx> -- 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
