Return-Path: Received: from userp1040.oracle.com ([156.151.31.81]:39316 "EHLO userp1040.oracle.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1755182AbcCBSgr convert rfc822-to-8bit (ORCPT ); Wed, 2 Mar 2016 13:36:47 -0500 Content-Type: text/plain; charset=us-ascii Mime-Version: 1.0 (1.0) Subject: Re: [PATCH v5 00/14] NFS/RDMA server patches for v4.6 From: Chuck Lever In-Reply-To: <20160301220052.GF23792@fieldses.org> Date: Wed, 2 Mar 2016 10:36:42 -0800 Cc: linux-rdma@vger.kernel.org, linux-nfs@vger.kernel.org Message-Id: References: <20160301180337.2492.4255.stgit@klimt.1015granger.net> <20160301220052.GF23792@fieldses.org> To: "J. Bruce Fields" Sender: linux-nfs-owner@vger.kernel.org List-ID: > On Mar 1, 2016, at 2:00 PM, J. Bruce Fields wrote: > >> On Tue, Mar 01, 2016 at 01:05:14PM -0500, Chuck Lever wrote: >> Hi Bruce- >> >> I think these are ready for linux-next. > > OK, applying for 4.6, thanks. Thanks! > I still wonder whether the code makes incorrect assumptions about how > the server xdr encoder uses the reply xdr_buf. Assumptions, probably. I believe these assumptions are the same as are made on the client side. It would be awesome to build an API contract for struct xdr_buf and have it written down somewhere, then we would know for sure. > That might only affect > clients which constructed rather strange compounds. And shouldn't be > made any worse by these patches. You're thinking of clients that might send more than one READ or WRITE in and NFSv4 COMPOUND? The thought for the current implementation is that the first READ or WRITE operation would be allowed to use the xdr_buf page list, and arguments/results for the remaining READ/WRITE operations would go inline (ie: a Long Call or Reply with data copy on both ends). > --b. > >> >> Two major changes: Finish server-side support for RDMA_ERROR type >> messages, and convert the server's completion handlers to the new >> core CQ API. >> >> Also available in the "nfsd-rdma-for-4.6" topic branch of this git >> repo: >> >> git://git.linux-nfs.org/projects/cel/cel-2.6.git >> >> Or for browsing: >> >> http://git.linux-nfs.org/?p=cel/cel-2.6.git;a=log;h=refs/heads/nfsd-rdma-for-4.6 >> >> >> Changes since v4: >> - Rebased on v4.5-rc6 >> - Replace confusing comments about NFSv4.1 over RDMA support >> >> >> Changes since v3: >> - Rebased on v4.5-rc5 >> - Completed fix for Write chunk XDR padding >> - On error, svc_rdma_sendto() should close connection >> >> >> Changes since v2: >> - Rebased on v4.5-rc4 >> - checkpatch.pl passes >> - Fix iWARP RDMA Read path regression >> - Reduce NFSv4.1 CB message size maximum in CREATE_SESSION >> - Another pass at fixing Write chunk XDR padding >> - Various clean ups >> >> >> Changes since v1: >> - Rebased on v4.5-rc3 >> - Patch 02/10 (pre-allocate frmrs) dropped >> - Patch 04/10 replaced with more surgical patch >> - Refined reporting of flushed completions >> - Function names of completion methods changed to match client's >> new completion function naming convention >> >> --- >> >> Chuck Lever (14): >> nfsd: Update NFS server comments related to RDMA support >> svcrdma: Find client-provided write and reply chunks once per reply >> svcrdma: Do not write xdr_buf::tail in a Write chunk >> svcrdma: Do not send Write chunk XDR pad with inline content >> nfsd: Lower NFSv4.1 callback message size limit >> svcrdma: Close connection when a send error occurs >> svcrdma: svc_rdma_post_recv() should close connection on error >> rpcrdma: Add RPCRDMA_HDRLEN_ERR >> svcrdma: Make RDMA_ERROR messages work >> svcrdma: Use correct XID in error replies >> svcrdma: Hook up the logic to return ERR_CHUNK >> svcrdma: Remove close_out exit path >> svcrdma: Use new CQ API for RPC-over-RDMA server receive CQs >> svcrdma: Use new CQ API for RPC-over-RDMA server send CQs >> >> >> fs/nfsd/nfs4state.c | 26 +- >> fs/nfsd/nfs4xdr.c | 2 >> include/linux/sunrpc/auth.h | 7 >> include/linux/sunrpc/rpc_rdma.h | 1 >> include/linux/sunrpc/svc_rdma.h | 20 + >> net/sunrpc/auth_null.c | 4 >> net/sunrpc/auth_unix.c | 6 >> net/sunrpc/xprtrdma/svc_rdma_backchannel.c | 17 - >> net/sunrpc/xprtrdma/svc_rdma_marshal.c | 64 +++- >> net/sunrpc/xprtrdma/svc_rdma_recvfrom.c | 60 ++-- >> net/sunrpc/xprtrdma/svc_rdma_sendto.c | 196 ++++++++---- >> net/sunrpc/xprtrdma/svc_rdma_transport.c | 445 +++++++++------------------- >> 12 files changed, 393 insertions(+), 455 deletions(-) >> >> -- >> Chuck Lever > -- > To unsubscribe from this list: send the line "unsubscribe linux-rdma" in > the body of a message to majordomo@vger.kernel.org > More majordomo info at http://vger.kernel.org/majordomo-info.html