On Wed, 2013-01-23 at 17:55 -0500, J. Bruce Fields wrote: > From: "J. Bruce Fields" <bfields@xxxxxxxxxx> > > Write xdr encoding primitives that can handle crossing page boundaries, > and use them in nfsd4_encode_fattr. > > The main practical advantage for now is that we can return arbitrarily > large ACLs (well, up to our maximum rpc size). However, compounds with > other operations following such a getattr may fail. > > Eventually we plan to use the same xdr code through v4 at least. > > Signed-off-by: J. Bruce Fields <bfields@xxxxxxxxxx> > --- > fs/nfsd/nfs4proc.c | 6 +- > fs/nfsd/nfs4xdr.c | 705 +++++++++++++++++++++++++++++++++++------------------ > fs/nfsd/xdr4.h | 18 +- > 3 files changed, 483 insertions(+), 246 deletions(-) > Why do these belong in the NFS server layer, and not the SUNRPC layer? Also, why are you inventing an entirely new server-specific structure instead of just reusing the existing struct xdr_stream? The xdr_stream already has support for bounce buffers to deal with crossing page boundaries (although I've only implemented the decode side of the equation so far)... -- Trond Myklebust Linux NFS client maintainer NetApp Trond.Myklebust@xxxxxxxxxx www.netapp.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
