Return-Path: Received: from mail-vk0-f65.google.com ([209.85.213.65]:52907 "EHLO mail-vk0-f65.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1752513AbdJSRVr (ORCPT ); Thu, 19 Oct 2017 13:21:47 -0400 Received: by mail-vk0-f65.google.com with SMTP id 126so5809253vkj.9 for ; Thu, 19 Oct 2017 10:21:47 -0700 (PDT) MIME-Version: 1.0 In-Reply-To: <1508361919-30495-1-git-send-email-bfields@redhat.com> References: <1508361919-30495-1-git-send-email-bfields@redhat.com> From: Olga Kornievskaia Date: Thu, 19 Oct 2017 13:21:46 -0400 Message-ID: Subject: Re: [PATCH 1/2] nfsd4: fix cached replies to solo SEQUENCE compounds To: "J. Bruce Fields" Cc: Trond Myklebust , linux-nfs Content-Type: text/plain; charset="UTF-8" Sender: linux-nfs-owner@vger.kernel.org List-ID: On Wed, Oct 18, 2017 at 5:25 PM, J. Bruce Fields wrote: > From: "J. Bruce Fields" > > Currently our handling of 4.1+ requests without "cachethis" set is > confusing and not quite correct. > > Suppose a client sends a compound consisting of only a single SEQUENCE > op, and it matches the seqid in a session slot (so it's a retry), but > the previous request with that seqid did not have "cachethis" set. > > The obvious thing to do might be to return NFS4ERR_RETRY_UNCACHED_REP, > but the protocol only allows that to be returned on the op following the > SEQUENCE, and there is no such op in this case. > > The protocol permits us to cache replies even if the client didn't ask > us to. And it's easy to do so in the case of solo SEQUENCE compounds. > > So, when we get a solo SEQUENCE, we can either return the previously > cached reply or NFSERR_SEQ_FALSE_RETRY if we notice it differs in some > way from the original call. I'm confused in my testing the error was SEQ_MISORDERED and not SEQ_FALSE_RETRY error? > Currently, we're returning a corrupt reply in the case a solo SEQUENCE > matches a previous compound with more ops. This actually matters > because the Linux client recently started doing this as a way to recover > from lost replies to idempotent operations in the case the process doing > the original reply was killed: in that case it's difficult to keep the > original arguments around to do a real retry, and the client no longer > cares what the result is anyway, but it would like to make sure that the > slot's sequence id has been incremented, and the solo SEQUENCE assures > that: if the server never got the original reply, it will increment the > sequence id. If it did get the original reply, it won't increment, and > nothing else that about the reply really matters much. But we can at > least attempt to return valid xdr! > > Signed-off-by: J. Bruce Fields > --- > fs/nfsd/nfs4state.c | 23 ++++++++++++++++------- > fs/nfsd/state.h | 1 + > fs/nfsd/xdr4.h | 13 +++++++++++-- > 3 files changed, 28 insertions(+), 9 deletions(-) > > diff --git a/fs/nfsd/nfs4state.c b/fs/nfsd/nfs4state.c > index 9db8a19cceaa..7bd3ad88b85c 100644 > --- a/fs/nfsd/nfs4state.c > +++ b/fs/nfsd/nfs4state.c > @@ -2292,14 +2292,15 @@ nfsd4_store_cache_entry(struct nfsd4_compoundres *resp) > > dprintk("--> %s slot %p\n", __func__, slot); > > - slot->sl_opcnt = resp->opcnt; > - slot->sl_status = resp->cstate.status; > - > slot->sl_flags |= NFSD4_SLOT_INITIALIZED; > - if (nfsd4_not_cached(resp)) { > - slot->sl_datalen = 0; > + if (!nfsd4_cache_this(resp)) { > + slot->sl_flags &= !NFSD4_SLOT_CACHED; > return; > } > + slot->sl_flags |= NFSD4_SLOT_CACHED; > + slot->sl_opcnt = resp->opcnt; > + slot->sl_status = resp->cstate.status; > + > base = resp->cstate.data_offset; > slot->sl_datalen = buf->len - base; > if (read_bytes_from_xdr_buf(buf, base, slot->sl_data, slot->sl_datalen)) > @@ -2326,8 +2327,16 @@ nfsd4_enc_sequence_replay(struct nfsd4_compoundargs *args, > op = &args->ops[resp->opcnt - 1]; > nfsd4_encode_operation(resp, op); > > - /* Return nfserr_retry_uncached_rep in next operation. */ > - if (args->opcnt > 1 && !(slot->sl_flags & NFSD4_SLOT_CACHETHIS)) { > + if (slot->sl_flags & NFSD4_SLOT_CACHED) > + return op->status; > + if (args->opcnt == 1) { > + /* > + * The original operation wasn't a solo sequence--we > + * always cache those--so this retry must not match the > + * original: > + */ > + op->status = nfserr_seq_false_retry; > + } else { > op = &args->ops[resp->opcnt++]; > op->status = nfserr_retry_uncached_rep; > nfsd4_encode_operation(resp, op); > diff --git a/fs/nfsd/state.h b/fs/nfsd/state.h > index 005c911b34ac..2488b7df1b35 100644 > --- a/fs/nfsd/state.h > +++ b/fs/nfsd/state.h > @@ -174,6 +174,7 @@ struct nfsd4_slot { > #define NFSD4_SLOT_INUSE (1 << 0) > #define NFSD4_SLOT_CACHETHIS (1 << 1) > #define NFSD4_SLOT_INITIALIZED (1 << 2) > +#define NFSD4_SLOT_CACHED (1 << 3) > u8 sl_flags; > char sl_data[]; > }; > diff --git a/fs/nfsd/xdr4.h b/fs/nfsd/xdr4.h > index 1e4edbf70052..bc29511b6405 100644 > --- a/fs/nfsd/xdr4.h > +++ b/fs/nfsd/xdr4.h > @@ -649,9 +649,18 @@ static inline bool nfsd4_is_solo_sequence(struct nfsd4_compoundres *resp) > return resp->opcnt == 1 && args->ops[0].opnum == OP_SEQUENCE; > } > > -static inline bool nfsd4_not_cached(struct nfsd4_compoundres *resp) > +/* > + * The session reply cache only needs to cache replies that the client > + * actually asked us to. But it's almost free for us to cache compounds > + * consisting of only a SEQUENCE op, so we may as well cache those too. > + * Also, the protocol doesn't give us a convenient response in the case > + * of a replay of a solo SEQUENCE op that wasn't cached > + * (RETRY_UNCACHED_REP can only be returned in the second op of a > + * compound). > + */ > +static inline bool nfsd4_cache_this(struct nfsd4_compoundres *resp) > { > - return !(resp->cstate.slot->sl_flags & NFSD4_SLOT_CACHETHIS) > + return (resp->cstate.slot->sl_flags & NFSD4_SLOT_CACHETHIS) > || nfsd4_is_solo_sequence(resp); > } > > -- > 2.13.5 > > -- > To unsubscribe from this list: send the line "unsubscribe linux-nfs" in > the body of a message to majordomo@vger.kernel.org > More majordomo info at http://vger.kernel.org/majordomo-info.html