From patchwork Wed May 30 18:05:44 2018 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Trond Myklebust X-Patchwork-Id: 10439527 Return-Path: Received: from mail.wl.linuxfoundation.org (pdx-wl-mail.web.codeaurora.org [172.30.200.125]) by pdx-korg-patchwork.web.codeaurora.org (Postfix) with ESMTP id 98BFB602CC for ; Wed, 30 May 2018 18:06:31 +0000 (UTC) Received: from mail.wl.linuxfoundation.org (localhost [127.0.0.1]) by mail.wl.linuxfoundation.org (Postfix) with ESMTP id 862AA290F0 for ; Wed, 30 May 2018 18:06:31 +0000 (UTC) Received: by mail.wl.linuxfoundation.org (Postfix, from userid 486) id 7A9AC291B0; Wed, 30 May 2018 18:06:31 +0000 (UTC) X-Spam-Checker-Version: SpamAssassin 3.3.1 (2010-03-16) on pdx-wl-mail.web.codeaurora.org X-Spam-Level: X-Spam-Status: No, score=-7.8 required=2.0 tests=BAYES_00, DKIM_ADSP_CUSTOM_MED, DKIM_SIGNED, FREEMAIL_FROM, MAILING_LIST_MULTI, RCVD_IN_DNSWL_HI, T_DKIM_INVALID autolearn=ham version=3.3.1 Received: from vger.kernel.org (vger.kernel.org [209.132.180.67]) by mail.wl.linuxfoundation.org (Postfix) with ESMTP id B651C290F0 for ; Wed, 30 May 2018 18:06:30 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S932106AbeE3SG3 (ORCPT ); Wed, 30 May 2018 14:06:29 -0400 Received: from mail-io0-f193.google.com ([209.85.223.193]:43962 "EHLO mail-io0-f193.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1753641AbeE3SG0 (ORCPT ); Wed, 30 May 2018 14:06:26 -0400 Received: by mail-io0-f193.google.com with SMTP id t6-v6so4130796iob.10 for ; Wed, 30 May 2018 11:06:26 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20161025; h=from:to:subject:date:message-id:in-reply-to:references:mime-version :content-transfer-encoding; bh=RMO8/nohqsJk5JQ3WnBtWr4z+8NvCNVZdptt0y5S3Wc=; b=m76Oaht5zMqdLUMkj+jnCS90fXGl9jzn94ZUEVfzdPwzW+SHW+AF2ZTeYbMnBwxWGy EVOLHx6nuVtMPiAT1jqYuKvMkxrhZBcathlPQWgZoiLJEw1ccxylswSLymCDtyiPMZwN F6ndn6/SJizk7XnT69dLyQE/DyR9+o4qvZSlIurmuyMFwI+Q+xr9gbf3spYXj0Hf3SeK 4ex2NqNiVfvm+TiI1Hh4uPpRoH2Oeo7vwbiR5Id75FpjeRXpUzZNetzuqRfB1eIyOveH iZBck/hkfYMJRtcRnzpv/fMvegbmxvhFHAuNZZzuvj2bY9uFtzlNEtRBlvbcYV2UFx/q NhMQ== X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20161025; h=x-gm-message-state:from:to:subject:date:message-id:in-reply-to :references:mime-version:content-transfer-encoding; bh=RMO8/nohqsJk5JQ3WnBtWr4z+8NvCNVZdptt0y5S3Wc=; b=NIP5Q2fhZMqcA3FLBRDHNV2+VW0lqf3/l8kFgK8nzIex0XsKbqMTJqBw0oYZM+TLmf 0KPYalxZOC7/++w93INzdU6jstrJZnM9hRChpkSX9m8iO85ZXNoXoyc4ywZeTjryl3Bj Z0KmrZU5+dWir+3w5qkxE74UnTcHP+hegcKMLrgMv7/He7AQK9UwW2QexGDRZgYrQU5W OnbCNfXe0xhLc7E4uVzwxQdG1Nx3BHdTChC6Mpo/e9KmuhrvDA+DyN13gR5uF5eiTMn0 /NTaqHHJeZ7V0jX9fY+j54gJk+S4ndEPIvx2vjKbw9hmItnftB9N/mMDeyzYSyIbDlAi L+Uw== X-Gm-Message-State: APt69E2ASjNEUylLZ9ECv8EHUOMeZ15txOc0sD4U6HkpAfhrq15tvf1r xs1EoqigVF+woUkPQn00eiUmk7E= X-Google-Smtp-Source: ADUXVKLQRR3bod0YpkBev0IWgK+Nn29P0OG7+7WI5wYdd/Ejkl6oP6zo3Tr2a/gJO+/VrgGpIsch8Q== X-Received: by 2002:a6b:954c:: with SMTP id x73-v6mr3106536iod.225.1527703585436; Wed, 30 May 2018 11:06:25 -0700 (PDT) Received: from leira.trondhjem.org.localdomain (c-68-49-162-121.hsd1.mi.comcast.net. [68.49.162.121]) by smtp.gmail.com with ESMTPSA id i201-v6sm8589368ite.33.2018.05.30.11.06.24 for (version=TLS1_2 cipher=ECDHE-RSA-CHACHA20-POLY1305 bits=256/256); Wed, 30 May 2018 11:06:24 -0700 (PDT) From: Trond Myklebust X-Google-Original-From: Trond Myklebust To: linux-nfs@vger.kernel.org Subject: [PATCH 10/19] pnfs: Add LAYOUTGET to OPEN of a new file Date: Wed, 30 May 2018 14:05:44 -0400 Message-Id: <20180530180553.38769-11-trond.myklebust@hammerspace.com> X-Mailer: git-send-email 2.17.0 In-Reply-To: <20180530180553.38769-10-trond.myklebust@hammerspace.com> References: <20180530180553.38769-1-trond.myklebust@hammerspace.com> <20180530180553.38769-2-trond.myklebust@hammerspace.com> <20180530180553.38769-3-trond.myklebust@hammerspace.com> <20180530180553.38769-4-trond.myklebust@hammerspace.com> <20180530180553.38769-5-trond.myklebust@hammerspace.com> <20180530180553.38769-6-trond.myklebust@hammerspace.com> <20180530180553.38769-7-trond.myklebust@hammerspace.com> <20180530180553.38769-8-trond.myklebust@hammerspace.com> <20180530180553.38769-9-trond.myklebust@hammerspace.com> <20180530180553.38769-10-trond.myklebust@hammerspace.com> MIME-Version: 1.0 Sender: linux-nfs-owner@vger.kernel.org Precedence: bulk List-ID: X-Mailing-List: linux-nfs@vger.kernel.org X-Virus-Scanned: ClamAV using ClamSMTP From: Fred Isaman This triggers when have no pre-existing inode to attach to. The preexisting case is saved for later. Signed-off-by: Fred Isaman Signed-off-by: Trond Myklebust --- fs/nfs/nfs4proc.c | 20 ++++++++++-- fs/nfs/nfs4state.c | 8 +++++ fs/nfs/pnfs.c | 81 ++++++++++++++++++++++++++++++++++++++++++++-- fs/nfs/pnfs.h | 17 ++++++++++ 4 files changed, 121 insertions(+), 5 deletions(-) diff --git a/fs/nfs/nfs4proc.c b/fs/nfs/nfs4proc.c index 771439971759..a73c83f17a86 100644 --- a/fs/nfs/nfs4proc.c +++ b/fs/nfs/nfs4proc.c @@ -86,6 +86,7 @@ | ATTR_MTIME_SET) struct nfs4_opendata; +static void nfs4_layoutget_release(void *calldata); static int _nfs4_recover_proc_open(struct nfs4_opendata *data); static int nfs4_do_fsinfo(struct nfs_server *, struct nfs_fh *, struct nfs_fsinfo *); static void nfs_fixup_referral_attributes(struct nfs_fattr *fattr); @@ -878,6 +879,10 @@ nfs4_sequence_process_interrupted(struct nfs_client *client, #else /* !CONFIG_NFS_V4_1 */ +static void nfs4_layoutget_release(void *calldata) +{ +} + static int nfs4_sequence_process(struct rpc_task *task, struct nfs4_sequence_res *res) { return nfs40_sequence_done(task, res); @@ -1245,6 +1250,8 @@ static void nfs4_opendata_free(struct kref *kref) struct nfs4_opendata, kref); struct super_block *sb = p->dentry->d_sb; + if (p->lgp) + nfs4_layoutget_release(p->lgp); nfs_free_seqid(p->o_arg.seqid); nfs4_sequence_free_slot(&p->o_res.seq_res); if (p->state != NULL) @@ -2335,8 +2342,10 @@ static int nfs4_run_open_task(struct nfs4_opendata *data, if (!ctx) { nfs4_init_sequence(&o_arg->seq_args, &o_res->seq_res, 1, 1); data->is_recover = true; - } else + } else { nfs4_init_sequence(&o_arg->seq_args, &o_res->seq_res, 1, 0); + pnfs_lgopen_prepare(data, ctx); + } task = rpc_run_task(&task_setup_data); if (IS_ERR(task)) return PTR_ERR(task); @@ -2816,7 +2825,10 @@ static int _nfs4_open_and_get_state(struct nfs4_opendata *opendata, nfs_inode_attach_open_context(ctx); if (read_seqcount_retry(&sp->so_reclaim_seqcount, seq)) nfs4_schedule_stateid_recovery(server, state); + else + pnfs_parse_lgopen(state->inode, opendata->lgp, ctx); } + out: return ret; } @@ -8718,13 +8730,15 @@ static void nfs4_layoutget_release(void *calldata) { struct nfs4_layoutget *lgp = calldata; struct inode *inode = lgp->args.inode; - struct nfs_server *server = NFS_SERVER(inode); - size_t max_pages = max_response_pages(server); + size_t max_pages = lgp->args.layout.pglen / PAGE_SIZE; dprintk("--> %s\n", __func__); nfs4_sequence_free_slot(&lgp->res.seq_res); nfs4_free_pages(lgp->args.layout.pages, max_pages); pnfs_put_layout_hdr(NFS_I(inode)->layout); + if (inode) + pnfs_put_layout_hdr(NFS_I(inode)->layout); + put_rpccred(lgp->cred); put_nfs_open_context(lgp->args.ctx); kfree(calldata); dprintk("<-- %s\n", __func__); diff --git a/fs/nfs/nfs4state.c b/fs/nfs/nfs4state.c index c10a422efe6f..2bf2eaa08ca7 100644 --- a/fs/nfs/nfs4state.c +++ b/fs/nfs/nfs4state.c @@ -77,6 +77,14 @@ const nfs4_stateid invalid_stateid = { .type = NFS4_INVALID_STATEID_TYPE, }; +const nfs4_stateid current_stateid = { + { + /* Funky initialiser keeps older gcc versions happy */ + .data = { 0x0, 0x0, 0x0, 0x1, 0 }, + }, + .type = NFS4_SPECIAL_STATEID_TYPE, +}; + static DEFINE_MUTEX(nfs_clid_init_mutex); int nfs4_init_clientid(struct nfs_client *clp, struct rpc_cred *cred) diff --git a/fs/nfs/pnfs.c b/fs/nfs/pnfs.c index 07f1bbd9100c..a0a2484c3aed 100644 --- a/fs/nfs/pnfs.c +++ b/fs/nfs/pnfs.c @@ -934,7 +934,7 @@ pnfs_find_server(struct inode *inode, struct nfs_open_context *ctx) static struct nfs4_layoutget * pnfs_alloc_init_layoutget_args(struct inode *ino, struct nfs_open_context *ctx, - nfs4_stateid *stateid, + const nfs4_stateid *stateid, const struct pnfs_layout_range *range, gfp_t gfp_flags) { @@ -978,7 +978,7 @@ pnfs_alloc_init_layoutget_args(struct inode *ino, lgp->args.ctx = get_nfs_open_context(ctx); nfs4_stateid_copy(&lgp->args.stateid, stateid); lgp->gfp_flags = gfp_flags; - lgp->cred = ctx->cred; + lgp->cred = get_rpccred(ctx->cred); return lgp; } @@ -1943,6 +1943,83 @@ pnfs_sanity_check_layout_range(struct pnfs_layout_range *range) return true; } +extern const nfs4_stateid current_stateid; + +static void _lgopen_prepare_attached(struct nfs4_opendata *data, + struct nfs_open_context *ctx) +{ + /* STUB */ +} + +static void _lgopen_prepare_floating(struct nfs4_opendata *data, + struct nfs_open_context *ctx) +{ + struct pnfs_layout_range rng = { + .iomode = (data->o_arg.fmode & FMODE_WRITE) ? + IOMODE_RW: IOMODE_READ, + .offset = 0, + .length = NFS4_MAX_UINT64, + }; + struct nfs4_layoutget *lgp; + + lgp = pnfs_alloc_init_layoutget_args(NULL, ctx, ¤t_stateid, + &rng, GFP_KERNEL); + if (!lgp) + return; + data->lgp = lgp; + data->o_arg.lg_args = &lgp->args; + data->o_res.lg_res = &lgp->res; +} + +void pnfs_lgopen_prepare(struct nfs4_opendata *data, + struct nfs_open_context *ctx) +{ + struct nfs_server *server = NFS_SERVER(data->dir->d_inode); + + if (!(pnfs_enabled_sb(server) && + server->pnfs_curr_ld->flags & PNFS_LAYOUTGET_ON_OPEN)) + return; + /* Could check on max_ops, but currently hardcoded high enough */ + if (data->state) + _lgopen_prepare_attached(data, ctx); + else + _lgopen_prepare_floating(data, ctx); +} + +void pnfs_parse_lgopen(struct inode *ino, struct nfs4_layoutget *lgp, + struct nfs_open_context *ctx) +{ + struct pnfs_layout_hdr *lo; + struct pnfs_layout_segment *lseg; + u32 iomode; + + if (!lgp || lgp->res.layoutp->len == 0) + return; + if (!lgp->args.inode) { + /* Need to grab lo */ + spin_lock(&ino->i_lock); + lo = pnfs_find_alloc_layout(ino, ctx, GFP_KERNEL); + atomic_inc(&lo->plh_outstanding); + spin_unlock(&ino->i_lock); + lgp->args.inode = ino; + } else + lo = NFS_I(lgp->args.inode)->layout; + pnfs_get_layout_hdr(lo); + + lseg = pnfs_layout_process(lgp); + atomic_dec(&lo->plh_outstanding); + if (IS_ERR(lseg)) { + /* ignore lseg, but would like to mark not to try lgopen */ + /* clear some lo flags - first and fail ???? */ + } else { + iomode = lgp->args.range.iomode; + pnfs_layout_clear_fail_bit(lo, pnfs_iomode_to_fail_bit(iomode)); + pnfs_put_lseg(lseg); + } + pnfs_clear_first_layoutget(lo); + pnfs_put_layout_hdr(lo); +} + struct pnfs_layout_segment * pnfs_layout_process(struct nfs4_layoutget *lgp) { diff --git a/fs/nfs/pnfs.h b/fs/nfs/pnfs.h index 964a7227ea97..b110c09ea89c 100644 --- a/fs/nfs/pnfs.h +++ b/fs/nfs/pnfs.h @@ -35,6 +35,8 @@ #include #include +struct nfs4_opendata; + enum { NFS_LSEG_VALID = 0, /* cleared when lseg is recalled/returned */ NFS_LSEG_ROC, /* roc bit received from server */ @@ -378,6 +380,10 @@ void pnfs_layout_mark_request_commit(struct nfs_page *req, struct pnfs_layout_segment *lseg, struct nfs_commit_info *cinfo, u32 ds_commit_idx); +void pnfs_lgopen_prepare(struct nfs4_opendata *data, + struct nfs_open_context *ctx); +void pnfs_parse_lgopen(struct inode *ino, struct nfs4_layoutget *lgp, + struct nfs_open_context *ctx); static inline bool nfs_have_layout(struct inode *inode) { @@ -778,6 +784,17 @@ static inline bool nfs4_refresh_layout_stateid(nfs4_stateid *dst, { return false; } + +static inline void pnfs_lgopen_prepare(struct nfs4_opendata *data, + struct nfs_open_context *ctx) +{ +} + +static inline void pnfs_parse_lgopen(struct inode *ino, + struct nfs4_layoutget *lgp, + struct nfs_open_context *ctx) +{ +} #endif /* CONFIG_NFS_V4_1 */ #if IS_ENABLED(CONFIG_NFS_V4_2)