From patchwork Sun May 29 18:24:22 2011 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Boaz Harrosh X-Patchwork-Id: 828222 Received: from vger.kernel.org (vger.kernel.org [209.132.180.67]) by demeter1.kernel.org (8.14.4/8.14.3) with ESMTP id p4TIOnLu029744 for ; Sun, 29 May 2011 18:24:49 GMT Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1755465Ab1E2SYf (ORCPT ); Sun, 29 May 2011 14:24:35 -0400 Received: from daytona.panasas.com ([67.152.220.89]:50902 "EHLO daytona.panasas.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1755863Ab1E2SYe (ORCPT ); Sun, 29 May 2011 14:24:34 -0400 Received: from fs2.bhalevy.com ([172.17.33.72]) by daytona.panasas.com with Microsoft SMTPSVC(6.0.3790.4675); Sun, 29 May 2011 14:24:32 -0400 From: Boaz Harrosh To: Trond Myklebust , Benny Halevy , NFS list , open-osd Cc: Benny Halevy Subject: [PATCH 20/32] pnfs: support for non-rpc layout drivers Date: Sun, 29 May 2011 21:24:22 +0300 Message-Id: <1306693462-8268-1-git-send-email-bharrosh@panasas.com> X-Mailer: git-send-email 1.7.2.3 In-Reply-To: <4DE28CFB.4060608@panasas.com> References: <4DE28CFB.4060608@panasas.com> X-OriginalArrivalTime: 29 May 2011 18:24:33.0166 (UTC) FILETIME=[A83A5AE0:01CC1E2D] Sender: linux-nfs-owner@vger.kernel.org Precedence: bulk List-ID: X-Mailing-List: linux-nfs@vger.kernel.org X-Greylist: IP, sender and recipient auto-whitelisted, not delayed by milter-greylist-4.2.6 (demeter1.kernel.org [140.211.167.41]); Sun, 29 May 2011 18:24:49 +0000 (UTC) From: Benny Halevy Non-rpc layout driver such as for objects and blocks implement their own I/O path and error handling logic. Therefore bypass NFS-based error handling for these layout drivers. [fix lseg ref-count bugs, and null de-refs] [Fall out from: non-rpc layout drivers] Signed-off-by: Boaz Harrosh [get rid of PNFS_USE_RPC_CODE] [get rid of __nfs4_write_done_cb] [revert useless change in nfs4_write_done_cb] Signed-off-by: Benny Halevy --- fs/nfs/internal.h | 1 + fs/nfs/nfs4proc.c | 13 +++++++++-- fs/nfs/pnfs.c | 48 ++++++++++++++++++++++++++++++++++++++++++++++- fs/nfs/pnfs.h | 2 + include/linux/nfs_xdr.h | 2 + 5 files changed, 62 insertions(+), 4 deletions(-) diff --git a/fs/nfs/internal.h b/fs/nfs/internal.h index ce118ce..bcf0f0f 100644 --- a/fs/nfs/internal.h +++ b/fs/nfs/internal.h @@ -310,6 +310,7 @@ extern int nfs_migrate_page(struct address_space *, #endif /* nfs4proc.c */ +extern void __nfs4_read_done_cb(struct nfs_read_data *); extern void nfs4_reset_read(struct rpc_task *task, struct nfs_read_data *data); extern int nfs4_init_client(struct nfs_client *clp, const struct rpc_timeout *timeparms, diff --git a/fs/nfs/nfs4proc.c b/fs/nfs/nfs4proc.c index cf1b339..92c8bc4 100644 --- a/fs/nfs/nfs4proc.c +++ b/fs/nfs/nfs4proc.c @@ -3175,6 +3175,11 @@ static int nfs4_proc_pathconf(struct nfs_server *server, struct nfs_fh *fhandle, return err; } +void __nfs4_read_done_cb(struct nfs_read_data *data) +{ + nfs_invalidate_atime(data->inode); +} + static int nfs4_read_done_cb(struct rpc_task *task, struct nfs_read_data *data) { struct nfs_server *server = NFS_SERVER(data->inode); @@ -3184,7 +3189,7 @@ static int nfs4_read_done_cb(struct rpc_task *task, struct nfs_read_data *data) return -EAGAIN; } - nfs_invalidate_atime(data->inode); + __nfs4_read_done_cb(data); if (task->tk_status > 0) renew_lease(server, data->timestamp); return 0; @@ -3198,7 +3203,8 @@ static int nfs4_read_done(struct rpc_task *task, struct nfs_read_data *data) if (!nfs4_sequence_done(task, &data->res.seq_res)) return -EAGAIN; - return data->read_done_cb(task, data); + return data->read_done_cb ? data->read_done_cb(task, data) : + nfs4_read_done_cb(task, data); } static void nfs4_proc_read_setup(struct nfs_read_data *data, struct rpc_message *msg) @@ -3243,7 +3249,8 @@ static int nfs4_write_done(struct rpc_task *task, struct nfs_write_data *data) { if (!nfs4_sequence_done(task, &data->res.seq_res)) return -EAGAIN; - return data->write_done_cb(task, data); + return data->write_done_cb ? data->write_done_cb(task, data) : + nfs4_write_done_cb(task, data); } /* Reset the the nfs_write_data to send the write to the MDS. */ diff --git a/fs/nfs/pnfs.c b/fs/nfs/pnfs.c index ef535f2..1716621 100644 --- a/fs/nfs/pnfs.c +++ b/fs/nfs/pnfs.c @@ -243,7 +243,7 @@ put_lseg_common(struct pnfs_layout_segment *lseg) { struct inode *inode = lseg->pls_layout->plh_inode; - BUG_ON(test_bit(NFS_LSEG_VALID, &lseg->pls_flags)); + WARN_ON(test_bit(NFS_LSEG_VALID, &lseg->pls_flags)); list_del_init(&lseg->pls_list); if (list_empty(&lseg->pls_layout->plh_segs)) { set_bit(NFS_LAYOUT_DESTROYED, &lseg->pls_layout->plh_flags); @@ -1054,6 +1054,29 @@ pnfs_pageio_init_write(struct nfs_pageio_descriptor *pgio, struct inode *inode) pgio->pg_test = (ld && ld->pg_test) ? pnfs_write_pg_test : NULL; } +/* + * Called by non rpc-based layout drivers + */ +int +pnfs_ld_write_done(struct nfs_write_data *data) +{ + int status; + + if (!data->pnfs_error) { + pnfs_set_layoutcommit(data); + data->mds_ops->rpc_call_done(&data->task, data); + data->mds_ops->rpc_release(data); + return 0; + } + + dprintk("%s: pnfs_error=%d, retry via MDS\n", __func__, + data->pnfs_error); + status = nfs_initiate_write(data, NFS_CLIENT(data->inode), + data->mds_ops, NFS_FILE_SYNC); + return status ? : -EAGAIN; +} +EXPORT_SYMBOL_GPL(pnfs_ld_write_done); + enum pnfs_try_status pnfs_try_to_write_data(struct nfs_write_data *wdata, const struct rpc_call_ops *call_ops, int how) @@ -1079,6 +1102,29 @@ pnfs_try_to_write_data(struct nfs_write_data *wdata, } /* + * Called by non rpc-based layout drivers + */ +int +pnfs_ld_read_done(struct nfs_read_data *data) +{ + int status; + + if (!data->pnfs_error) { + __nfs4_read_done_cb(data); + data->mds_ops->rpc_call_done(&data->task, data); + data->mds_ops->rpc_release(data); + return 0; + } + + dprintk("%s: pnfs_error=%d, retry via MDS\n", __func__, + data->pnfs_error); + status = nfs_initiate_read(data, NFS_CLIENT(data->inode), + data->mds_ops); + return status ? : -EAGAIN; +} +EXPORT_SYMBOL_GPL(pnfs_ld_read_done); + +/* * Call the appropriate parallel I/O subsystem read function. */ enum pnfs_try_status diff --git a/fs/nfs/pnfs.h b/fs/nfs/pnfs.h index 925d6ef..0383e66 100644 --- a/fs/nfs/pnfs.h +++ b/fs/nfs/pnfs.h @@ -165,6 +165,8 @@ void pnfs_roc_set_barrier(struct inode *ino, u32 barrier); bool pnfs_roc_drain(struct inode *ino, u32 *barrier); void pnfs_set_layoutcommit(struct nfs_write_data *wdata); int pnfs_layoutcommit_inode(struct inode *inode, bool sync); +int pnfs_ld_write_done(struct nfs_write_data *); +int pnfs_ld_read_done(struct nfs_read_data *); /* pnfs_dev.c */ struct nfs4_deviceid_node { diff --git a/include/linux/nfs_xdr.h b/include/linux/nfs_xdr.h index 7e371f7..7c8ff09 100644 --- a/include/linux/nfs_xdr.h +++ b/include/linux/nfs_xdr.h @@ -1087,6 +1087,7 @@ struct nfs_read_data { const struct rpc_call_ops *mds_ops; int (*read_done_cb) (struct rpc_task *task, struct nfs_read_data *data); __u64 mds_offset; + int pnfs_error; struct page *page_array[NFS_PAGEVEC_SIZE]; }; @@ -1112,6 +1113,7 @@ struct nfs_write_data { unsigned long timestamp; /* For lease renewal */ #endif __u64 mds_offset; /* Filelayout dense stripe */ + int pnfs_error; struct page *page_array[NFS_PAGEVEC_SIZE]; };