On Jan 24, 2014, at 4:00 PM, Wendy Cheng <s.wendy.cheng@xxxxxxxxx> wrote: > On Fri, Jan 24, 2014 at 10:34 AM, Chuck Lever <chuck.lever@xxxxxxxxxx> wrote: > > On Jan 24, 2014, at 1:26 PM, Wendy Cheng <s.wendy.cheng@xxxxxxxxx> wrote: > > > On Fri, Jan 17, 2014 at 11:37 AM, Chuck Lever <chuck.lever@xxxxxxxxxx> wrote: > > This brief series fixes some long-standing regressions in our > > NFS/RDMA client implementation. Comments / review? > > > > --- > > > > Chuck Lever (3): > > SUNRPC: remove KERN_INFO from dprintk() call sites > > SUNRPC: Fix large reads on NFS/RDMA > > NFS: Fix READDIR oops with NFSv4 on RDMA > > > > > > fs/nfs/nfs4xdr.c | 3 +-- > > net/sunrpc/xprtrdma/rpc_rdma.c | 4 +--- > > net/sunrpc/xprtrdma/transport.c | 10 +++++----- > > 3 files changed, 7 insertions(+), 10 deletions(-) > > > > > > We have an embedded system that is currently set at NFS V3 with kernel 2.6.38 (as NFS client). Could folks suggest a good kernel version (close to 2.6.38 or RHEL based) to pick up NFS v4 (in order to pick up NFS over RDMA) ? > > If NFS/RDMA is working in your kernel, I believe NFSv3 and NFSv4 should both work over RDMA. > > > I did have V3 running (with NFS over RDMA). The issue here is how to move it from an experimental project into a release candidate. I think NFS V4 would be a better platform to start with (?) ... So let me re-phrase the question .. > > Regardless RDMA support, will (client side) NFS V4 work well with 2.6.38 kernel? If not, what would be a good version (close to 2.6.38) to go to for a stable NFS V4 client ? Client support for NFSv4 in recent updates of RHEL6 is stable. -- Chuck Lever chuck[dot]lever[at]oracle[dot]com -- To unsubscribe from this list: send the line "unsubscribe linux-nfs" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
