From b17621fed6aa039387e35f9b4d34d98f213e5673 Mon Sep 17 00:00:00 2001 From: Wu Fengguang Date: Thu, 3 Dec 2009 13:54:25 +0100 Subject: writeback: introduce wbc.for_background It will lower the flush priority for NFS, and maybe more in future. Signed-off-by: Wu Fengguang Cc: Trond Myklebust Cc: Jens Axboe Cc: Christoph Hellwig Signed-off-by: Jens Axboe --- fs/nfs/write.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'fs/nfs/write.c') diff --git a/fs/nfs/write.c b/fs/nfs/write.c index 53eb26c16b5..c84b5cc1a94 100644 --- a/fs/nfs/write.c +++ b/fs/nfs/write.c @@ -178,7 +178,7 @@ static int wb_priority(struct writeback_control *wbc) { if (wbc->for_reclaim) return FLUSH_HIGHPRI | FLUSH_STABLE; - if (wbc->for_kupdate) + if (wbc->for_kupdate || wbc->for_background) return FLUSH_LOWPRI; return 0; } -- cgit v1.2.3-70-g09d2 From e608e79f1bf4b967afcf57777e63b5f0939b00e8 Mon Sep 17 00:00:00 2001 From: Andy Adamson Date: Fri, 4 Dec 2009 15:55:29 -0500 Subject: nfs41: call free slot from nfs4_restart_rpc nfs41_sequence_free_slot can be called multiple times on SEQUENCE operation errors. No reason to inline nfs4_restart_rpc Reported-by: Trond Myklebust nfs_writeback_done and nfs_readpage_retry call nfs4_restart_rpc outside the error handler, and the slot is not freed prior to restarting in the rpc_prepare state during session reset. Fix this by moving the call to nfs41_sequence_free_slot from the error path of nfs41_sequence_done into nfs4_restart_rpc, and by removing the test for NFS4CLNT_SESSION_SETUP. Always free slot and goto the rpc prepare state on async errors. Signed-off-by: Andy Adamson Signed-off-by: Trond Myklebust --- fs/nfs/internal.h | 16 ++-------------- fs/nfs/nfs4proc.c | 29 +++++++++++++++++++++++------ fs/nfs/read.c | 3 ++- fs/nfs/unlink.c | 4 +++- fs/nfs/write.c | 3 ++- 5 files changed, 32 insertions(+), 23 deletions(-) (limited to 'fs/nfs/write.c') diff --git a/fs/nfs/internal.h b/fs/nfs/internal.h index ebcd3795389..a6f7b6cbfd0 100644 --- a/fs/nfs/internal.h +++ b/fs/nfs/internal.h @@ -177,26 +177,14 @@ extern __be32 * nfs_decode_dirent(__be32 *, struct nfs_entry *, int); extern struct rpc_procinfo nfs3_procedures[]; extern __be32 *nfs3_decode_dirent(__be32 *, struct nfs_entry *, int); -/* nfs4proc.c */ -static inline void nfs4_restart_rpc(struct rpc_task *task, - const struct nfs_client *clp) -{ -#ifdef CONFIG_NFS_V4_1 - if (nfs4_has_session(clp) && - test_bit(NFS4CLNT_SESSION_RESET, &clp->cl_state)) { - rpc_restart_call_prepare(task); - return; - } -#endif /* CONFIG_NFS_V4_1 */ - rpc_restart_call(task); -} - /* nfs4xdr.c */ #ifdef CONFIG_NFS_V4 extern __be32 *nfs4_decode_dirent(__be32 *p, struct nfs_entry *entry, int plus); #endif /* nfs4proc.c */ +extern void nfs4_restart_rpc(struct rpc_task *, const struct nfs_client *, + struct nfs4_sequence_res *); #ifdef CONFIG_NFS_V4 extern struct rpc_procinfo nfs4_procedures[]; #endif diff --git a/fs/nfs/nfs4proc.c b/fs/nfs/nfs4proc.c index a23110d1d2a..6ef50aa785d 100644 --- a/fs/nfs/nfs4proc.c +++ b/fs/nfs/nfs4proc.c @@ -644,6 +644,19 @@ static void nfs4_sequence_done(const struct nfs_server *server, #endif /* CONFIG_NFS_V4_1 */ } +void nfs4_restart_rpc(struct rpc_task *task, const struct nfs_client *clp, + struct nfs4_sequence_res *res) +{ +#ifdef CONFIG_NFS_V4_1 + if (nfs4_has_session(clp)) { + nfs41_sequence_free_slot(clp, res); + rpc_restart_call_prepare(task); + return; + } +#endif /* CONFIG_NFS_V4_1 */ + rpc_restart_call(task); +} + /* no restart, therefore free slot here */ static void nfs4_sequence_done_free_slot(const struct nfs_server *server, struct nfs4_sequence_res *res, @@ -1750,7 +1763,8 @@ static void nfs4_close_done(struct rpc_task *task, void *data) break; default: if (nfs4_async_handle_error(task, server, state) == -EAGAIN) { - nfs4_restart_rpc(task, server->nfs_client); + nfs4_restart_rpc(task, server->nfs_client, + &calldata->res.seq_res); return; } } @@ -2988,7 +3002,7 @@ static int nfs4_read_done(struct rpc_task *task, struct nfs_read_data *data) nfs4_sequence_done(server, &data->res.seq_res, task->tk_status); if (nfs4_async_handle_error(task, server, data->args.context->state) == -EAGAIN) { - nfs4_restart_rpc(task, server->nfs_client); + nfs4_restart_rpc(task, server->nfs_client, &data->res.seq_res); return -EAGAIN; } @@ -3013,7 +3027,8 @@ static int nfs4_write_done(struct rpc_task *task, struct nfs_write_data *data) task->tk_status); if (nfs4_async_handle_error(task, NFS_SERVER(inode), data->args.context->state) == -EAGAIN) { - nfs4_restart_rpc(task, NFS_SERVER(inode)->nfs_client); + nfs4_restart_rpc(task, NFS_SERVER(inode)->nfs_client, + &data->res.seq_res); return -EAGAIN; } if (task->tk_status >= 0) { @@ -3041,7 +3056,8 @@ static int nfs4_commit_done(struct rpc_task *task, struct nfs_write_data *data) nfs4_sequence_done(NFS_SERVER(inode), &data->res.seq_res, task->tk_status); if (nfs4_async_handle_error(task, NFS_SERVER(inode), NULL) == -EAGAIN) { - nfs4_restart_rpc(task, NFS_SERVER(inode)->nfs_client); + nfs4_restart_rpc(task, NFS_SERVER(inode)->nfs_client, + &data->res.seq_res); return -EAGAIN; } nfs4_sequence_free_slot(NFS_SERVER(inode)->nfs_client, @@ -3755,7 +3771,8 @@ static void nfs4_locku_done(struct rpc_task *task, void *data) default: if (nfs4_async_handle_error(task, calldata->server, NULL) == -EAGAIN) nfs4_restart_rpc(task, - calldata->server->nfs_client); + calldata->server->nfs_client, + &calldata->res.seq_res); } nfs4_sequence_free_slot(calldata->server->nfs_client, &calldata->res.seq_res); @@ -4890,7 +4907,7 @@ void nfs41_sequence_call_done(struct rpc_task *task, void *data) if (_nfs4_async_handle_error(task, NULL, clp, NULL) == -EAGAIN) { - nfs4_restart_rpc(task, clp); + nfs4_restart_rpc(task, clp, task->tk_msg.rpc_resp); return; } } diff --git a/fs/nfs/read.c b/fs/nfs/read.c index 12c9e66d3f1..3e04fb9ea64 100644 --- a/fs/nfs/read.c +++ b/fs/nfs/read.c @@ -368,7 +368,8 @@ static void nfs_readpage_retry(struct rpc_task *task, struct nfs_read_data *data argp->offset += resp->count; argp->pgbase += resp->count; argp->count -= resp->count; - nfs4_restart_rpc(task, NFS_SERVER(data->inode)->nfs_client); + nfs4_restart_rpc(task, NFS_SERVER(data->inode)->nfs_client, + &data->res.seq_res); return; out: nfs4_sequence_free_slot(NFS_SERVER(data->inode)->nfs_client, diff --git a/fs/nfs/unlink.c b/fs/nfs/unlink.c index 1064c91ae81..52f7bdb12c8 100644 --- a/fs/nfs/unlink.c +++ b/fs/nfs/unlink.c @@ -81,9 +81,11 @@ static void nfs_async_unlink_done(struct rpc_task *task, void *calldata) { struct nfs_unlinkdata *data = calldata; struct inode *dir = data->dir; + struct nfs_removeres *res = task->tk_msg.rpc_resp; if (!NFS_PROTO(dir)->unlink_done(task, dir)) - nfs4_restart_rpc(task, NFS_SERVER(dir)->nfs_client); + nfs4_restart_rpc(task, NFS_SERVER(dir)->nfs_client, + &res->seq_res); } /** diff --git a/fs/nfs/write.c b/fs/nfs/write.c index 53eb26c16b5..556668ff022 100644 --- a/fs/nfs/write.c +++ b/fs/nfs/write.c @@ -1216,7 +1216,8 @@ int nfs_writeback_done(struct rpc_task *task, struct nfs_write_data *data) */ argp->stable = NFS_FILE_SYNC; } - nfs4_restart_rpc(task, server->nfs_client); + nfs4_restart_rpc(task, server->nfs_client, + &data->res.seq_res); return -EAGAIN; } if (time_before(complain, jiffies)) { -- cgit v1.2.3-70-g09d2 From d61e612a728fb9bf848c4383f8f6645e822d5b57 Mon Sep 17 00:00:00 2001 From: Trond Myklebust Date: Sat, 5 Dec 2009 19:32:19 -0500 Subject: NFSv41: Clean up slot table management We no longer need to maintain a distinction between nfs41_sequence_done and nfs41_sequence_free_slot. This fixes a number of slot table leakages in the NFSv4.1 code. Signed-off-by: Trond Myklebust --- fs/nfs/internal.h | 17 +-------------- fs/nfs/nfs4proc.c | 65 +++++++++++++------------------------------------------ fs/nfs/read.c | 13 +++-------- fs/nfs/unlink.c | 4 +--- fs/nfs/write.c | 4 +--- 5 files changed, 21 insertions(+), 82 deletions(-) (limited to 'fs/nfs/write.c') diff --git a/fs/nfs/internal.h b/fs/nfs/internal.h index 83a9284b83c..b1a020c1172 100644 --- a/fs/nfs/internal.h +++ b/fs/nfs/internal.h @@ -197,8 +197,7 @@ extern const u32 nfs41_maxwrite_overhead; #endif /* nfs4proc.c */ -extern void nfs4_restart_rpc(struct rpc_task *, const struct nfs_client *, - struct nfs4_sequence_res *); +extern void nfs4_restart_rpc(struct rpc_task *, const struct nfs_client *); #ifdef CONFIG_NFS_V4 extern struct rpc_procinfo nfs4_procedures[]; #endif @@ -275,20 +274,6 @@ extern int _nfs4_call_sync_session(struct nfs_server *server, struct nfs4_sequence_res *res, int cache_reply); -#ifdef CONFIG_NFS_V4_1 -extern void nfs41_sequence_free_slot(const struct nfs_client *, - struct nfs4_sequence_res *res); -#endif /* CONFIG_NFS_V4_1 */ - -static inline void nfs4_sequence_free_slot(const struct nfs_client *clp, - struct nfs4_sequence_res *res) -{ -#ifdef CONFIG_NFS_V4_1 - if (nfs4_has_session(clp)) - nfs41_sequence_free_slot(clp, res); -#endif /* CONFIG_NFS_V4_1 */ -} - /* * Determine the device name as a string */ diff --git a/fs/nfs/nfs4proc.c b/fs/nfs/nfs4proc.c index be96d28bacc..c06a2bade59 100644 --- a/fs/nfs/nfs4proc.c +++ b/fs/nfs/nfs4proc.c @@ -341,15 +341,11 @@ nfs4_free_slot(struct nfs4_slot_table *tbl, u8 free_slotid) free_slotid, tbl->highest_used_slotid); } -void nfs41_sequence_free_slot(const struct nfs_client *clp, +static void nfs41_sequence_free_slot(const struct nfs_client *clp, struct nfs4_sequence_res *res) { struct nfs4_slot_table *tbl; - if (!nfs4_has_session(clp)) { - dprintk("%s: No session\n", __func__); - return; - } tbl = &clp->cl_session->fc_slot_table; if (res->sr_slotid == NFS4_MAX_SLOT_TABLE) { /* just wake up the next guy waiting since @@ -407,7 +403,6 @@ static void nfs41_sequence_done(struct nfs_client *clp, spin_unlock(&clp->cl_lock); /* Check sequence flags */ nfs41_handle_sequence_flag_errors(clp, res->sr_status_flags); - return; } out: /* The session may be reset by one of the error handlers. */ @@ -556,7 +551,6 @@ static void nfs41_call_sync_done(struct rpc_task *task, void *calldata) struct nfs41_call_sync_data *data = calldata; nfs41_sequence_done(data->clp, data->seq_res, task->tk_status); - nfs41_sequence_free_slot(data->clp, data->seq_res); } struct rpc_call_ops nfs41_call_sync_ops = { @@ -632,12 +626,10 @@ static void nfs4_sequence_done(const struct nfs_server *server, #endif /* CONFIG_NFS_V4_1 */ } -void nfs4_restart_rpc(struct rpc_task *task, const struct nfs_client *clp, - struct nfs4_sequence_res *res) +void nfs4_restart_rpc(struct rpc_task *task, const struct nfs_client *clp) { #ifdef CONFIG_NFS_V4_1 if (nfs4_has_session(clp)) { - nfs41_sequence_free_slot(clp, res); rpc_restart_call_prepare(task); return; } @@ -645,15 +637,6 @@ void nfs4_restart_rpc(struct rpc_task *task, const struct nfs_client *clp, rpc_restart_call(task); } -/* no restart, therefore free slot here */ -static void nfs4_sequence_done_free_slot(const struct nfs_server *server, - struct nfs4_sequence_res *res, - int rpc_status) -{ - nfs4_sequence_done(server, res, rpc_status); - nfs4_sequence_free_slot(server->nfs_client, res); -} - static void update_changeattr(struct inode *dir, struct nfs4_change_info *cinfo) { struct nfs_inode *nfsi = NFS_I(dir); @@ -1350,8 +1333,8 @@ static void nfs4_open_done(struct rpc_task *task, void *calldata) data->rpc_status = task->tk_status; - nfs4_sequence_done_free_slot(data->o_arg.server, &data->o_res.seq_res, - task->tk_status); + nfs4_sequence_done(data->o_arg.server, &data->o_res.seq_res, + task->tk_status); if (RPC_ASSASSINATED(task)) return; @@ -1757,12 +1740,10 @@ static void nfs4_close_done(struct rpc_task *task, void *data) break; default: if (nfs4_async_handle_error(task, server, state) == -EAGAIN) { - nfs4_restart_rpc(task, server->nfs_client, - &calldata->res.seq_res); + nfs4_restart_rpc(task, server->nfs_client); return; } } - nfs4_sequence_free_slot(server->nfs_client, &calldata->res.seq_res); nfs_refresh_inode(calldata->inode, calldata->res.fattr); } @@ -2553,7 +2534,6 @@ static int nfs4_proc_unlink_done(struct rpc_task *task, struct inode *dir) nfs4_sequence_done(res->server, &res->seq_res, task->tk_status); if (nfs4_async_handle_error(task, res->server, NULL) == -EAGAIN) return 0; - nfs4_sequence_free_slot(res->server->nfs_client, &res->seq_res); update_changeattr(dir, &res->cinfo); nfs_post_op_update_inode(dir, &res->dir_attr); return 1; @@ -2992,20 +2972,16 @@ static int nfs4_read_done(struct rpc_task *task, struct nfs_read_data *data) dprintk("--> %s\n", __func__); - /* nfs4_sequence_free_slot called in the read rpc_call_done */ nfs4_sequence_done(server, &data->res.seq_res, task->tk_status); if (nfs4_async_handle_error(task, server, data->args.context->state) == -EAGAIN) { - nfs4_restart_rpc(task, server->nfs_client, &data->res.seq_res); + nfs4_restart_rpc(task, server->nfs_client); return -EAGAIN; } nfs_invalidate_atime(data->inode); if (task->tk_status > 0) renew_lease(server, data->timestamp); - else if (task->tk_status < 0) - nfs4_sequence_free_slot(server->nfs_client, &data->res.seq_res); - return 0; } @@ -3019,13 +2995,11 @@ static int nfs4_write_done(struct rpc_task *task, struct nfs_write_data *data) { struct inode *inode = data->inode; - /* slot is freed in nfs_writeback_done */ nfs4_sequence_done(NFS_SERVER(inode), &data->res.seq_res, task->tk_status); if (nfs4_async_handle_error(task, NFS_SERVER(inode), data->args.context->state) == -EAGAIN) { - nfs4_restart_rpc(task, NFS_SERVER(inode)->nfs_client, - &data->res.seq_res); + nfs4_restart_rpc(task, NFS_SERVER(inode)->nfs_client); return -EAGAIN; } if (task->tk_status >= 0) { @@ -3053,12 +3027,9 @@ static int nfs4_commit_done(struct rpc_task *task, struct nfs_write_data *data) nfs4_sequence_done(NFS_SERVER(inode), &data->res.seq_res, task->tk_status); if (nfs4_async_handle_error(task, NFS_SERVER(inode), NULL) == -EAGAIN) { - nfs4_restart_rpc(task, NFS_SERVER(inode)->nfs_client, - &data->res.seq_res); + nfs4_restart_rpc(task, NFS_SERVER(inode)->nfs_client); return -EAGAIN; } - nfs4_sequence_free_slot(NFS_SERVER(inode)->nfs_client, - &data->res.seq_res); nfs_refresh_inode(inode, data->res.fattr); return 0; } @@ -3509,8 +3480,8 @@ static void nfs4_delegreturn_done(struct rpc_task *task, void *calldata) { struct nfs4_delegreturndata *data = calldata; - nfs4_sequence_done_free_slot(data->res.server, &data->res.seq_res, - task->tk_status); + nfs4_sequence_done(data->res.server, &data->res.seq_res, + task->tk_status); data->rpc_status = task->tk_status; if (data->rpc_status == 0) @@ -3768,11 +3739,8 @@ static void nfs4_locku_done(struct rpc_task *task, void *data) default: if (nfs4_async_handle_error(task, calldata->server, NULL) == -EAGAIN) nfs4_restart_rpc(task, - calldata->server->nfs_client, - &calldata->res.seq_res); + calldata->server->nfs_client); } - nfs4_sequence_free_slot(calldata->server->nfs_client, - &calldata->res.seq_res); } static void nfs4_locku_prepare(struct rpc_task *task, void *data) @@ -3954,8 +3922,8 @@ static void nfs4_lock_done(struct rpc_task *task, void *calldata) dprintk("%s: begin!\n", __func__); - nfs4_sequence_done_free_slot(data->server, &data->res.seq_res, - task->tk_status); + nfs4_sequence_done(data->server, &data->res.seq_res, + task->tk_status); data->rpc_status = task->tk_status; if (RPC_ASSASSINATED(task)) @@ -4425,10 +4393,9 @@ static void nfs4_get_lease_time_done(struct rpc_task *task, void *calldata) dprintk("%s Retry: tk_status %d\n", __func__, task->tk_status); rpc_delay(task, NFS4_POLL_RETRY_MIN); task->tk_status = 0; - rpc_restart_call(task); + nfs4_restart_rpc(task, data->clp); return; } - nfs41_sequence_free_slot(data->clp, &data->res->lr_seq_res); dprintk("<-- %s\n", __func__); } @@ -4900,11 +4867,10 @@ void nfs41_sequence_call_done(struct rpc_task *task, void *data) if (_nfs4_async_handle_error(task, NULL, clp, NULL) == -EAGAIN) { - nfs4_restart_rpc(task, clp, task->tk_msg.rpc_resp); + nfs4_restart_rpc(task, clp); return; } } - nfs41_sequence_free_slot(clp, task->tk_msg.rpc_resp); dprintk("%s rpc_cred %p\n", __func__, task->tk_msg.rpc_cred); kfree(task->tk_msg.rpc_argp); @@ -5008,7 +4974,6 @@ static void nfs4_reclaim_complete_done(struct rpc_task *task, void *data) return; } } - nfs41_sequence_free_slot(clp, res); dprintk("<-- %s\n", __func__); } diff --git a/fs/nfs/read.c b/fs/nfs/read.c index 3e04fb9ea64..d319bfbe513 100644 --- a/fs/nfs/read.c +++ b/fs/nfs/read.c @@ -356,26 +356,19 @@ static void nfs_readpage_retry(struct rpc_task *task, struct nfs_read_data *data struct nfs_readres *resp = &data->res; if (resp->eof || resp->count == argp->count) - goto out; + return; /* This is a short read! */ nfs_inc_stats(data->inode, NFSIOS_SHORTREAD); /* Has the server at least made some progress? */ if (resp->count == 0) - goto out; + return; /* Yes, so retry the read at the end of the data */ argp->offset += resp->count; argp->pgbase += resp->count; argp->count -= resp->count; - nfs4_restart_rpc(task, NFS_SERVER(data->inode)->nfs_client, - &data->res.seq_res); - return; -out: - nfs4_sequence_free_slot(NFS_SERVER(data->inode)->nfs_client, - &data->res.seq_res); - return; - + nfs4_restart_rpc(task, NFS_SERVER(data->inode)->nfs_client); } /* diff --git a/fs/nfs/unlink.c b/fs/nfs/unlink.c index 52f7bdb12c8..1064c91ae81 100644 --- a/fs/nfs/unlink.c +++ b/fs/nfs/unlink.c @@ -81,11 +81,9 @@ static void nfs_async_unlink_done(struct rpc_task *task, void *calldata) { struct nfs_unlinkdata *data = calldata; struct inode *dir = data->dir; - struct nfs_removeres *res = task->tk_msg.rpc_resp; if (!NFS_PROTO(dir)->unlink_done(task, dir)) - nfs4_restart_rpc(task, NFS_SERVER(dir)->nfs_client, - &res->seq_res); + nfs4_restart_rpc(task, NFS_SERVER(dir)->nfs_client); } /** diff --git a/fs/nfs/write.c b/fs/nfs/write.c index 556668ff022..d546c607de0 100644 --- a/fs/nfs/write.c +++ b/fs/nfs/write.c @@ -1216,8 +1216,7 @@ int nfs_writeback_done(struct rpc_task *task, struct nfs_write_data *data) */ argp->stable = NFS_FILE_SYNC; } - nfs4_restart_rpc(task, server->nfs_client, - &data->res.seq_res); + nfs4_restart_rpc(task, server->nfs_client); return -EAGAIN; } if (time_before(complain, jiffies)) { @@ -1229,7 +1228,6 @@ int nfs_writeback_done(struct rpc_task *task, struct nfs_write_data *data) /* Can't do anything about it except throw an error. */ task->tk_status = -EIO; } - nfs4_sequence_free_slot(server->nfs_client, &data->res.seq_res); return 0; } -- cgit v1.2.3-70-g09d2 From 0110ee152b69f8cbde19d8bc1dd59e197e419d76 Mon Sep 17 00:00:00 2001 From: Trond Myklebust Date: Mon, 7 Dec 2009 09:00:24 -0500 Subject: NFS: Fix up the declaration of nfs4_restart_rpc when NFSv4 not configured Also rename it: it is used in generic code, and so should not have a 'nfs4' prefix. Reported-by: Stephen Rothwell Signed-off-by: Trond Myklebust --- fs/nfs/internal.h | 13 ++++++++++++- fs/nfs/nfs4proc.c | 25 +++++++------------------ fs/nfs/read.c | 2 +- fs/nfs/unlink.c | 2 +- fs/nfs/write.c | 2 +- 5 files changed, 22 insertions(+), 22 deletions(-) (limited to 'fs/nfs/write.c') diff --git a/fs/nfs/internal.h b/fs/nfs/internal.h index b1a020c1172..29e464d23b3 100644 --- a/fs/nfs/internal.h +++ b/fs/nfs/internal.h @@ -197,7 +197,6 @@ extern const u32 nfs41_maxwrite_overhead; #endif /* nfs4proc.c */ -extern void nfs4_restart_rpc(struct rpc_task *, const struct nfs_client *); #ifdef CONFIG_NFS_V4 extern struct rpc_procinfo nfs4_procedures[]; #endif @@ -367,3 +366,15 @@ unsigned int nfs_page_array_len(unsigned int base, size_t len) return ((unsigned long)len + (unsigned long)base + PAGE_SIZE - 1) >> PAGE_SHIFT; } + +/* + * Helper for restarting RPC calls in the possible presence of NFSv4.1 + * sessions. + */ +static inline void nfs_restart_rpc(struct rpc_task *task, const struct nfs_client *clp) +{ + if (nfs4_has_session(clp)) + rpc_restart_call_prepare(task); + else + rpc_restart_call(task); +} diff --git a/fs/nfs/nfs4proc.c b/fs/nfs/nfs4proc.c index fbae2c94dbc..acde7765452 100644 --- a/fs/nfs/nfs4proc.c +++ b/fs/nfs/nfs4proc.c @@ -625,17 +625,6 @@ static void nfs4_sequence_done(const struct nfs_server *server, #endif /* CONFIG_NFS_V4_1 */ } -void nfs4_restart_rpc(struct rpc_task *task, const struct nfs_client *clp) -{ -#ifdef CONFIG_NFS_V4_1 - if (nfs4_has_session(clp)) { - rpc_restart_call_prepare(task); - return; - } -#endif /* CONFIG_NFS_V4_1 */ - rpc_restart_call(task); -} - static void update_changeattr(struct inode *dir, struct nfs4_change_info *cinfo) { struct nfs_inode *nfsi = NFS_I(dir); @@ -1739,7 +1728,7 @@ static void nfs4_close_done(struct rpc_task *task, void *data) break; default: if (nfs4_async_handle_error(task, server, state) == -EAGAIN) { - nfs4_restart_rpc(task, server->nfs_client); + nfs_restart_rpc(task, server->nfs_client); return; } } @@ -2974,7 +2963,7 @@ static int nfs4_read_done(struct rpc_task *task, struct nfs_read_data *data) nfs4_sequence_done(server, &data->res.seq_res, task->tk_status); if (nfs4_async_handle_error(task, server, data->args.context->state) == -EAGAIN) { - nfs4_restart_rpc(task, server->nfs_client); + nfs_restart_rpc(task, server->nfs_client); return -EAGAIN; } @@ -2998,7 +2987,7 @@ static int nfs4_write_done(struct rpc_task *task, struct nfs_write_data *data) task->tk_status); if (nfs4_async_handle_error(task, NFS_SERVER(inode), data->args.context->state) == -EAGAIN) { - nfs4_restart_rpc(task, NFS_SERVER(inode)->nfs_client); + nfs_restart_rpc(task, NFS_SERVER(inode)->nfs_client); return -EAGAIN; } if (task->tk_status >= 0) { @@ -3026,7 +3015,7 @@ static int nfs4_commit_done(struct rpc_task *task, struct nfs_write_data *data) nfs4_sequence_done(NFS_SERVER(inode), &data->res.seq_res, task->tk_status); if (nfs4_async_handle_error(task, NFS_SERVER(inode), NULL) == -EAGAIN) { - nfs4_restart_rpc(task, NFS_SERVER(inode)->nfs_client); + nfs_restart_rpc(task, NFS_SERVER(inode)->nfs_client); return -EAGAIN; } nfs_refresh_inode(inode, data->res.fattr); @@ -3737,7 +3726,7 @@ static void nfs4_locku_done(struct rpc_task *task, void *data) break; default: if (nfs4_async_handle_error(task, calldata->server, NULL) == -EAGAIN) - nfs4_restart_rpc(task, + nfs_restart_rpc(task, calldata->server->nfs_client); } } @@ -4392,7 +4381,7 @@ static void nfs4_get_lease_time_done(struct rpc_task *task, void *calldata) dprintk("%s Retry: tk_status %d\n", __func__, task->tk_status); rpc_delay(task, NFS4_POLL_RETRY_MIN); task->tk_status = 0; - nfs4_restart_rpc(task, data->clp); + nfs_restart_rpc(task, data->clp); return; } dprintk("<-- %s\n", __func__); @@ -4854,7 +4843,7 @@ void nfs41_sequence_call_done(struct rpc_task *task, void *data) if (_nfs4_async_handle_error(task, NULL, clp, NULL) == -EAGAIN) { - nfs4_restart_rpc(task, clp); + nfs_restart_rpc(task, clp); return; } } diff --git a/fs/nfs/read.c b/fs/nfs/read.c index d319bfbe513..db9b360ae19 100644 --- a/fs/nfs/read.c +++ b/fs/nfs/read.c @@ -368,7 +368,7 @@ static void nfs_readpage_retry(struct rpc_task *task, struct nfs_read_data *data argp->offset += resp->count; argp->pgbase += resp->count; argp->count -= resp->count; - nfs4_restart_rpc(task, NFS_SERVER(data->inode)->nfs_client); + nfs_restart_rpc(task, NFS_SERVER(data->inode)->nfs_client); } /* diff --git a/fs/nfs/unlink.c b/fs/nfs/unlink.c index 1064c91ae81..6da3d3ff6ed 100644 --- a/fs/nfs/unlink.c +++ b/fs/nfs/unlink.c @@ -83,7 +83,7 @@ static void nfs_async_unlink_done(struct rpc_task *task, void *calldata) struct inode *dir = data->dir; if (!NFS_PROTO(dir)->unlink_done(task, dir)) - nfs4_restart_rpc(task, NFS_SERVER(dir)->nfs_client); + nfs_restart_rpc(task, NFS_SERVER(dir)->nfs_client); } /** diff --git a/fs/nfs/write.c b/fs/nfs/write.c index d546c607de0..a28123be08a 100644 --- a/fs/nfs/write.c +++ b/fs/nfs/write.c @@ -1216,7 +1216,7 @@ int nfs_writeback_done(struct rpc_task *task, struct nfs_write_data *data) */ argp->stable = NFS_FILE_SYNC; } - nfs4_restart_rpc(task, server->nfs_client); + nfs_restart_rpc(task, server->nfs_client); return -EAGAIN; } if (time_before(complain, jiffies)) { -- cgit v1.2.3-70-g09d2 From 6b2f3d1f769be5779b479c37800229d9a4809fc3 Mon Sep 17 00:00:00 2001 From: Christoph Hellwig Date: Tue, 27 Oct 2009 11:05:28 +0100 Subject: vfs: Implement proper O_SYNC semantics While Linux provided an O_SYNC flag basically since day 1, it took until Linux 2.4.0-test12pre2 to actually get it implemented for filesystems, since that day we had generic_osync_around with only minor changes and the great "For now, when the user asks for O_SYNC, we'll actually give O_DSYNC" comment. This patch intends to actually give us real O_SYNC semantics in addition to the O_DSYNC semantics. After Jan's O_SYNC patches which are required before this patch it's actually surprisingly simple, we just need to figure out when to set the datasync flag to vfs_fsync_range and when not. This patch renames the existing O_SYNC flag to O_DSYNC while keeping it's numerical value to keep binary compatibility, and adds a new real O_SYNC flag. To guarantee backwards compatiblity it is defined as expanding to both the O_DSYNC and the new additional binary flag (__O_SYNC) to make sure we are backwards-compatible when compiled against the new headers. This also means that all places that don't care about the differences can just check O_DSYNC and get the right behaviour for O_SYNC, too - only places that actuall care need to check __O_SYNC in addition. Drivers and network filesystems have been updated in a fail safe way to always do the full sync magic if O_DSYNC is set. The few places setting O_SYNC for lower layers are kept that way for now to stay failsafe. We enforce that O_DSYNC is set when __O_SYNC is set early in the open path to make sure we always get these sane options. Note that parisc really screwed up their headers as they already define a O_DSYNC that has always been a no-op. We try to repair it by using it for the new O_DSYNC and redefinining O_SYNC to send both the traditional O_SYNC numerical value _and_ the O_DSYNC one. Cc: Richard Henderson Cc: Ivan Kokshaysky Cc: Grant Grundler Cc: "David S. Miller" Cc: Ingo Molnar Cc: "H. Peter Anvin" Cc: Thomas Gleixner Cc: Al Viro Cc: Andreas Dilger Acked-by: Trond Myklebust Acked-by: Kyle McMartin Acked-by: Ulrich Drepper Signed-off-by: Christoph Hellwig Signed-off-by: Andrew Morton Signed-off-by: Jan Kara --- arch/alpha/include/asm/fcntl.h | 19 ++++++++++++++++--- arch/blackfin/include/asm/fcntl.h | 2 -- arch/mips/include/asm/fcntl.h | 17 ++++++++++++++++- arch/mips/kernel/kspd.c | 1 + arch/mips/loongson/common/mem.c | 2 +- arch/mips/mm/cache.c | 2 +- arch/parisc/include/asm/fcntl.h | 5 ++--- arch/sparc/include/asm/fcntl.h | 19 ++++++++++++++++--- arch/x86/mm/pat.c | 3 +-- drivers/char/mem.c | 6 +++--- drivers/usb/gadget/file_storage.c | 2 +- fs/afs/write.c | 5 +++-- fs/btrfs/file.c | 4 ++-- fs/cifs/dir.c | 3 ++- fs/cifs/file.c | 6 ++++-- fs/namei.c | 9 +++++++++ fs/nfs/file.c | 4 ++-- fs/nfs/write.c | 2 +- fs/ocfs2/file.c | 2 +- fs/sync.c | 5 +++-- fs/ubifs/file.c | 2 +- fs/xfs/linux-2.6/xfs_lrw.c | 2 +- include/asm-generic/fcntl.h | 25 +++++++++++++++++++++---- sound/core/rawmidi.c | 2 +- 24 files changed, 109 insertions(+), 40 deletions(-) (limited to 'fs/nfs/write.c') diff --git a/arch/alpha/include/asm/fcntl.h b/arch/alpha/include/asm/fcntl.h index 25da0017ec8..21b1117a0c6 100644 --- a/arch/alpha/include/asm/fcntl.h +++ b/arch/alpha/include/asm/fcntl.h @@ -1,8 +1,6 @@ #ifndef _ALPHA_FCNTL_H #define _ALPHA_FCNTL_H -/* open/fcntl - O_SYNC is only implemented on blocks devices and on files - located on an ext2 file system */ #define O_CREAT 01000 /* not fcntl */ #define O_TRUNC 02000 /* not fcntl */ #define O_EXCL 04000 /* not fcntl */ @@ -10,13 +8,28 @@ #define O_NONBLOCK 00004 #define O_APPEND 00010 -#define O_SYNC 040000 +#define O_DSYNC 040000 /* used to be O_SYNC, see below */ #define O_DIRECTORY 0100000 /* must be a directory */ #define O_NOFOLLOW 0200000 /* don't follow links */ #define O_LARGEFILE 0400000 /* will be set by the kernel on every open */ #define O_DIRECT 02000000 /* direct disk access - should check with OSF/1 */ #define O_NOATIME 04000000 #define O_CLOEXEC 010000000 /* set close_on_exec */ +/* + * Before Linux 2.6.32 only O_DSYNC semantics were implemented, but using + * the O_SYNC flag. We continue to use the existing numerical value + * for O_DSYNC semantics now, but using the correct symbolic name for it. + * This new value is used to request true Posix O_SYNC semantics. It is + * defined in this strange way to make sure applications compiled against + * new headers get at least O_DSYNC semantics on older kernels. + * + * This has the nice side-effect that we can simply test for O_DSYNC + * wherever we do not care if O_DSYNC or O_SYNC is used. + * + * Note: __O_SYNC must never be used directly. + */ +#define __O_SYNC 020000000 +#define O_SYNC (__O_SYNC|O_DSYNC) #define F_GETLK 7 #define F_SETLK 8 diff --git a/arch/blackfin/include/asm/fcntl.h b/arch/blackfin/include/asm/fcntl.h index 8727b2b382f..251c911d59c 100644 --- a/arch/blackfin/include/asm/fcntl.h +++ b/arch/blackfin/include/asm/fcntl.h @@ -7,8 +7,6 @@ #ifndef _BFIN_FCNTL_H #define _BFIN_FCNTL_H -/* open/fcntl - O_SYNC is only implemented on blocks devices and on files - located on an ext2 file system */ #define O_DIRECTORY 040000 /* must be a directory */ #define O_NOFOLLOW 0100000 /* don't follow links */ #define O_DIRECT 0200000 /* direct disk access hint - currently ignored */ diff --git a/arch/mips/include/asm/fcntl.h b/arch/mips/include/asm/fcntl.h index 2a52333a062..7c6681aa2ab 100644 --- a/arch/mips/include/asm/fcntl.h +++ b/arch/mips/include/asm/fcntl.h @@ -10,7 +10,7 @@ #define O_APPEND 0x0008 -#define O_SYNC 0x0010 +#define O_DSYNC 0x0010 /* used to be O_SYNC, see below */ #define O_NONBLOCK 0x0080 #define O_CREAT 0x0100 /* not fcntl */ #define O_TRUNC 0x0200 /* not fcntl */ @@ -18,6 +18,21 @@ #define O_NOCTTY 0x0800 /* not fcntl */ #define FASYNC 0x1000 /* fcntl, for BSD compatibility */ #define O_LARGEFILE 0x2000 /* allow large file opens */ +/* + * Before Linux 2.6.32 only O_DSYNC semantics were implemented, but using + * the O_SYNC flag. We continue to use the existing numerical value + * for O_DSYNC semantics now, but using the correct symbolic name for it. + * This new value is used to request true Posix O_SYNC semantics. It is + * defined in this strange way to make sure applications compiled against + * new headers get at least O_DSYNC semantics on older kernels. + * + * This has the nice side-effect that we can simply test for O_DSYNC + * wherever we do not care if O_DSYNC or O_SYNC is used. + * + * Note: __O_SYNC must never be used directly. + */ +#define __O_SYNC 0x4000 +#define O_SYNC (__O_SYNC|O_DSYNC) #define O_DIRECT 0x8000 /* direct disk access hint */ #define F_GETLK 14 diff --git a/arch/mips/kernel/kspd.c b/arch/mips/kernel/kspd.c index ad4e017ed2f..80e2ba694ba 100644 --- a/arch/mips/kernel/kspd.c +++ b/arch/mips/kernel/kspd.c @@ -82,6 +82,7 @@ static int sp_stopping; #define MTSP_O_SHLOCK 0x0010 #define MTSP_O_EXLOCK 0x0020 #define MTSP_O_ASYNC 0x0040 +/* XXX: check which of these is actually O_SYNC vs O_DSYNC */ #define MTSP_O_FSYNC O_SYNC #define MTSP_O_NOFOLLOW 0x0100 #define MTSP_O_SYNC 0x0080 diff --git a/arch/mips/loongson/common/mem.c b/arch/mips/loongson/common/mem.c index 7c92f79b648..e94ef158f98 100644 --- a/arch/mips/loongson/common/mem.c +++ b/arch/mips/loongson/common/mem.c @@ -26,7 +26,7 @@ void __init prom_init_memory(void) /* override of arch/mips/mm/cache.c: __uncached_access */ int __uncached_access(struct file *file, unsigned long addr) { - if (file->f_flags & O_SYNC) + if (file->f_flags & O_DSYNC) return 1; return addr >= __pa(high_memory) || diff --git a/arch/mips/mm/cache.c b/arch/mips/mm/cache.c index 694d51f523d..102b2dfa542 100644 --- a/arch/mips/mm/cache.c +++ b/arch/mips/mm/cache.c @@ -194,7 +194,7 @@ void __devinit cpu_cache_init(void) int __weak __uncached_access(struct file *file, unsigned long addr) { - if (file->f_flags & O_SYNC) + if (file->f_flags & O_DSYNC) return 1; return addr >= __pa(high_memory); diff --git a/arch/parisc/include/asm/fcntl.h b/arch/parisc/include/asm/fcntl.h index 1e1c824764e..f357fc693c8 100644 --- a/arch/parisc/include/asm/fcntl.h +++ b/arch/parisc/include/asm/fcntl.h @@ -1,14 +1,13 @@ #ifndef _PARISC_FCNTL_H #define _PARISC_FCNTL_H -/* open/fcntl - O_SYNC is only implemented on blocks devices and on files - located on an ext2 file system */ #define O_APPEND 000000010 #define O_BLKSEEK 000000100 /* HPUX only */ #define O_CREAT 000000400 /* not fcntl */ #define O_EXCL 000002000 /* not fcntl */ #define O_LARGEFILE 000004000 -#define O_SYNC 000100000 +#define __O_SYNC 000100000 +#define O_SYNC (__O_SYNC|O_DSYNC) #define O_NONBLOCK 000200004 /* HPUX has separate NDELAY & NONBLOCK */ #define O_NOCTTY 000400000 /* not fcntl */ #define O_DSYNC 001000000 /* HPUX only */ diff --git a/arch/sparc/include/asm/fcntl.h b/arch/sparc/include/asm/fcntl.h index d4d9c9d852c..3b9cfb39175 100644 --- a/arch/sparc/include/asm/fcntl.h +++ b/arch/sparc/include/asm/fcntl.h @@ -1,14 +1,12 @@ #ifndef _SPARC_FCNTL_H #define _SPARC_FCNTL_H -/* open/fcntl - O_SYNC is only implemented on blocks devices and on files - located on an ext2 file system */ #define O_APPEND 0x0008 #define FASYNC 0x0040 /* fcntl, for BSD compatibility */ #define O_CREAT 0x0200 /* not fcntl */ #define O_TRUNC 0x0400 /* not fcntl */ #define O_EXCL 0x0800 /* not fcntl */ -#define O_SYNC 0x2000 +#define O_DSYNC 0x2000 /* used to be O_SYNC, see below */ #define O_NONBLOCK 0x4000 #if defined(__sparc__) && defined(__arch64__) #define O_NDELAY 0x0004 @@ -20,6 +18,21 @@ #define O_DIRECT 0x100000 /* direct disk access hint */ #define O_NOATIME 0x200000 #define O_CLOEXEC 0x400000 +/* + * Before Linux 2.6.32 only O_DSYNC semantics were implemented, but using + * the O_SYNC flag. We continue to use the existing numerical value + * for O_DSYNC semantics now, but using the correct symbolic name for it. + * This new value is used to request true Posix O_SYNC semantics. It is + * defined in this strange way to make sure applications compiled against + * new headers get at least O_DSYNC semantics on older kernels. + * + * This has the nice side-effect that we can simply test for O_DSYNC + * wherever we do not care if O_DSYNC or O_SYNC is used. + * + * Note: __O_SYNC must never be used directly. + */ +#define __O_SYNC 0x800000 +#define O_SYNC (__O_SYNC|O_DSYNC) #define F_GETOWN 5 /* for sockets. */ #define F_SETOWN 6 /* for sockets. */ diff --git a/arch/x86/mm/pat.c b/arch/x86/mm/pat.c index 66b55d6e69e..ae9648eb1c7 100644 --- a/arch/x86/mm/pat.c +++ b/arch/x86/mm/pat.c @@ -704,9 +704,8 @@ int phys_mem_access_prot_allowed(struct file *file, unsigned long pfn, if (!range_is_allowed(pfn, size)) return 0; - if (file->f_flags & O_SYNC) { + if (file->f_flags & O_DSYNC) flags = _PAGE_CACHE_UC_MINUS; - } #ifdef CONFIG_X86_32 /* diff --git a/drivers/char/mem.c b/drivers/char/mem.c index 30eff80fed6..fba76fb55ab 100644 --- a/drivers/char/mem.c +++ b/drivers/char/mem.c @@ -43,7 +43,7 @@ static inline int uncached_access(struct file *file, unsigned long addr) { #if defined(CONFIG_IA64) /* - * On ia64, we ignore O_SYNC because we cannot tolerate memory attribute aliases. + * On ia64, we ignore O_DSYNC because we cannot tolerate memory attribute aliases. */ return !(efi_mem_attributes(addr) & EFI_MEMORY_WB); #elif defined(CONFIG_MIPS) @@ -56,9 +56,9 @@ static inline int uncached_access(struct file *file, unsigned long addr) #else /* * Accessing memory above the top the kernel knows about or through a file pointer - * that was marked O_SYNC will be done non-cached. + * that was marked O_DSYNC will be done non-cached. */ - if (file->f_flags & O_SYNC) + if (file->f_flags & O_DSYNC) return 1; return addr >= __pa(high_memory); #endif diff --git a/drivers/usb/gadget/file_storage.c b/drivers/usb/gadget/file_storage.c index 1e6aa504d58..5e14dbaf65b 100644 --- a/drivers/usb/gadget/file_storage.c +++ b/drivers/usb/gadget/file_storage.c @@ -1713,7 +1713,7 @@ static int do_write(struct fsg_dev *fsg) } if (fsg->cmnd[1] & 0x08) { // FUA spin_lock(&curlun->filp->f_lock); - curlun->filp->f_flags |= O_SYNC; + curlun->filp->f_flags |= O_DSYNC; spin_unlock(&curlun->filp->f_lock); } } diff --git a/fs/afs/write.c b/fs/afs/write.c index c63a3c8beb7..6be1bc31616 100644 --- a/fs/afs/write.c +++ b/fs/afs/write.c @@ -692,8 +692,9 @@ ssize_t afs_file_write(struct kiocb *iocb, const struct iovec *iov, } /* return error values for O_SYNC and IS_SYNC() */ - if (IS_SYNC(&vnode->vfs_inode) || iocb->ki_filp->f_flags & O_SYNC) { - ret = afs_fsync(iocb->ki_filp, dentry, 1); + if (IS_SYNC(&vnode->vfs_inode) || iocb->ki_filp->f_flags & O_DSYNC) { + ret = afs_fsync(iocb->ki_filp, dentry, + (iocb->ki_filp->f_flags & __O_SYNC) ? 0 : 1); if (ret < 0) result = ret; } diff --git a/fs/btrfs/file.c b/fs/btrfs/file.c index 06550affbd2..77f759302e1 100644 --- a/fs/btrfs/file.c +++ b/fs/btrfs/file.c @@ -909,7 +909,7 @@ static ssize_t btrfs_file_write(struct file *file, const char __user *buf, unsigned long last_index; int will_write; - will_write = ((file->f_flags & O_SYNC) || IS_SYNC(inode) || + will_write = ((file->f_flags & O_DSYNC) || IS_SYNC(inode) || (file->f_flags & O_DIRECT)); nrptrs = min((count + PAGE_CACHE_SIZE - 1) / PAGE_CACHE_SIZE, @@ -1076,7 +1076,7 @@ out_nolock: if (err) num_written = err; - if ((file->f_flags & O_SYNC) || IS_SYNC(inode)) { + if ((file->f_flags & O_DSYNC) || IS_SYNC(inode)) { trans = btrfs_start_transaction(root, 1); ret = btrfs_log_dentry_safe(trans, root, file->f_dentry); diff --git a/fs/cifs/dir.c b/fs/cifs/dir.c index 1f42f772865..6ccf7262d1b 100644 --- a/fs/cifs/dir.c +++ b/fs/cifs/dir.c @@ -214,7 +214,8 @@ int cifs_posix_open(char *full_path, struct inode **pinode, posix_flags |= SMB_O_EXCL; if (oflags & O_TRUNC) posix_flags |= SMB_O_TRUNC; - if (oflags & O_SYNC) + /* be safe and imply O_SYNC for O_DSYNC */ + if (oflags & O_DSYNC) posix_flags |= SMB_O_SYNC; if (oflags & O_DIRECTORY) posix_flags |= SMB_O_DIRECTORY; diff --git a/fs/cifs/file.c b/fs/cifs/file.c index 429337eb7af..057e1dae12a 100644 --- a/fs/cifs/file.c +++ b/fs/cifs/file.c @@ -76,8 +76,10 @@ static inline fmode_t cifs_posix_convert_flags(unsigned int flags) reopening a file. They had their effect on the original open */ if (flags & O_APPEND) posix_flags |= (fmode_t)O_APPEND; - if (flags & O_SYNC) - posix_flags |= (fmode_t)O_SYNC; + if (flags & O_DSYNC) + posix_flags |= (fmode_t)O_DSYNC; + if (flags & __O_SYNC) + posix_flags |= (fmode_t)__O_SYNC; if (flags & O_DIRECTORY) posix_flags |= (fmode_t)O_DIRECTORY; if (flags & O_NOFOLLOW) diff --git a/fs/namei.c b/fs/namei.c index d11f404667e..b83d38f614f 100644 --- a/fs/namei.c +++ b/fs/namei.c @@ -1678,6 +1678,15 @@ struct file *do_filp_open(int dfd, const char *pathname, int will_write; int flag = open_to_namei_flags(open_flag); + /* + * O_SYNC is implemented as __O_SYNC|O_DSYNC. As many places only + * check for O_DSYNC if the need any syncing at all we enforce it's + * always set instead of having to deal with possibly weird behaviour + * for malicious applications setting only __O_SYNC. + */ + if (open_flag & __O_SYNC) + open_flag |= O_DSYNC; + if (!acc_mode) acc_mode = MAY_OPEN | ACC_MODE(flag); diff --git a/fs/nfs/file.c b/fs/nfs/file.c index f5fdd39e037..6b891328f33 100644 --- a/fs/nfs/file.c +++ b/fs/nfs/file.c @@ -581,7 +581,7 @@ static int nfs_need_sync_write(struct file *filp, struct inode *inode) { struct nfs_open_context *ctx; - if (IS_SYNC(inode) || (filp->f_flags & O_SYNC)) + if (IS_SYNC(inode) || (filp->f_flags & O_DSYNC)) return 1; ctx = nfs_file_open_context(filp); if (test_bit(NFS_CONTEXT_ERROR_WRITE, &ctx->flags)) @@ -622,7 +622,7 @@ static ssize_t nfs_file_write(struct kiocb *iocb, const struct iovec *iov, nfs_add_stats(inode, NFSIOS_NORMALWRITTENBYTES, count); result = generic_file_aio_write(iocb, iov, nr_segs, pos); - /* Return error values for O_SYNC and IS_SYNC() */ + /* Return error values for O_DSYNC and IS_SYNC() */ if (result >= 0 && nfs_need_sync_write(iocb->ki_filp, inode)) { int err = nfs_do_fsync(nfs_file_open_context(iocb->ki_filp), inode); if (err < 0) diff --git a/fs/nfs/write.c b/fs/nfs/write.c index c84b5cc1a94..b1ce2ea9b93 100644 --- a/fs/nfs/write.c +++ b/fs/nfs/write.c @@ -774,7 +774,7 @@ int nfs_updatepage(struct file *file, struct page *page, */ if (nfs_write_pageuptodate(page, inode) && inode->i_flock == NULL && - !(file->f_flags & O_SYNC)) { + !(file->f_flags & O_DSYNC)) { count = max(count + offset, nfs_page_length(page)); offset = 0; } diff --git a/fs/ocfs2/file.c b/fs/ocfs2/file.c index de059f49058..3d30a1c974a 100644 --- a/fs/ocfs2/file.c +++ b/fs/ocfs2/file.c @@ -2006,7 +2006,7 @@ out_dio: /* buffered aio wouldn't have proper lock coverage today */ BUG_ON(ret == -EIOCBQUEUED && !(file->f_flags & O_DIRECT)); - if ((file->f_flags & O_SYNC && !direct_io) || IS_SYNC(inode)) { + if ((file->f_flags & O_DSYNC && !direct_io) || IS_SYNC(inode)) { ret = filemap_fdatawrite_range(file->f_mapping, pos, pos + count - 1); if (ret < 0) diff --git a/fs/sync.c b/fs/sync.c index d104591b066..b75ca68dc08 100644 --- a/fs/sync.c +++ b/fs/sync.c @@ -295,10 +295,11 @@ SYSCALL_DEFINE1(fdatasync, unsigned int, fd) */ int generic_write_sync(struct file *file, loff_t pos, loff_t count) { - if (!(file->f_flags & O_SYNC) && !IS_SYNC(file->f_mapping->host)) + if (!(file->f_flags & O_DSYNC) && !IS_SYNC(file->f_mapping->host)) return 0; return vfs_fsync_range(file, file->f_path.dentry, pos, - pos + count - 1, 1); + pos + count - 1, + (file->f_flags & __O_SYNC) ? 0 : 1); } EXPORT_SYMBOL(generic_write_sync); diff --git a/fs/ubifs/file.c b/fs/ubifs/file.c index 1009adc8d60..eaa3d480bc2 100644 --- a/fs/ubifs/file.c +++ b/fs/ubifs/file.c @@ -1401,7 +1401,7 @@ static ssize_t ubifs_aio_write(struct kiocb *iocb, const struct iovec *iov, if (ret < 0) return ret; - if (ret > 0 && (IS_SYNC(inode) || iocb->ki_filp->f_flags & O_SYNC)) { + if (ret > 0 && (IS_SYNC(inode) || iocb->ki_filp->f_flags & O_DSYNC)) { err = ubifs_sync_wbufs_by_inode(c, inode); if (err) return err; diff --git a/fs/xfs/linux-2.6/xfs_lrw.c b/fs/xfs/linux-2.6/xfs_lrw.c index 072050f8d34..339c52b1a43 100644 --- a/fs/xfs/linux-2.6/xfs_lrw.c +++ b/fs/xfs/linux-2.6/xfs_lrw.c @@ -811,7 +811,7 @@ write_retry: XFS_STATS_ADD(xs_write_bytes, ret); /* Handle various SYNC-type writes */ - if ((file->f_flags & O_SYNC) || IS_SYNC(inode)) { + if ((file->f_flags & O_DSYNC) || IS_SYNC(inode)) { loff_t end = pos + ret - 1; int error2; diff --git a/include/asm-generic/fcntl.h b/include/asm-generic/fcntl.h index 495dc8af404..681ddf3e844 100644 --- a/include/asm-generic/fcntl.h +++ b/include/asm-generic/fcntl.h @@ -3,8 +3,6 @@ #include -/* open/fcntl - O_SYNC is only implemented on blocks devices and on files - located on an ext2 file system */ #define O_ACCMODE 00000003 #define O_RDONLY 00000000 #define O_WRONLY 00000001 @@ -27,8 +25,8 @@ #ifndef O_NONBLOCK #define O_NONBLOCK 00004000 #endif -#ifndef O_SYNC -#define O_SYNC 00010000 +#ifndef O_DSYNC +#define O_DSYNC 00010000 /* used to be O_SYNC, see below */ #endif #ifndef FASYNC #define FASYNC 00020000 /* fcntl, for BSD compatibility */ @@ -51,6 +49,25 @@ #ifndef O_CLOEXEC #define O_CLOEXEC 02000000 /* set close_on_exec */ #endif + +/* + * Before Linux 2.6.32 only O_DSYNC semantics were implemented, but using + * the O_SYNC flag. We continue to use the existing numerical value + * for O_DSYNC semantics now, but using the correct symbolic name for it. + * This new value is used to request true Posix O_SYNC semantics. It is + * defined in this strange way to make sure applications compiled against + * new headers get at least O_DSYNC semantics on older kernels. + * + * This has the nice side-effect that we can simply test for O_DSYNC + * wherever we do not care if O_DSYNC or O_SYNC is used. + * + * Note: __O_SYNC must never be used directly. + */ +#ifndef O_SYNC +#define __O_SYNC 04000000 +#define O_SYNC (__O_SYNC|O_DSYNC) +#endif + #ifndef O_NDELAY #define O_NDELAY O_NONBLOCK #endif diff --git a/sound/core/rawmidi.c b/sound/core/rawmidi.c index 2f766123b15..0f5a194695d 100644 --- a/sound/core/rawmidi.c +++ b/sound/core/rawmidi.c @@ -1257,7 +1257,7 @@ static ssize_t snd_rawmidi_write(struct file *file, const char __user *buf, break; count -= count1; } - if (file->f_flags & O_SYNC) { + if (file->f_flags & O_DSYNC) { spin_lock_irq(&runtime->lock); while (runtime->avail != runtime->buffer_size) { wait_queue_t wait; -- cgit v1.2.3-70-g09d2 From 190f38e5cedc910940b1da9015f00458c18f97b4 Mon Sep 17 00:00:00 2001 From: Trond Myklebust Date: Thu, 10 Dec 2009 09:05:55 -0500 Subject: NFS: Fix nfs_migrate_page() The call to migrate_page() will cause the page->private field to be cleared. Also fix up the locking around the page->private transfer, so that we ensure that calls to nfs_page_find_request() don't end up racing. Finally, fix up a double free bug: nfs_unlock_request() already calls nfs_release_request() for us... Reported-by: Wu Fengguang Tested-by: Andi Kleen Cc: stable@kernel.org Signed-off-by: Trond Myklebust --- fs/nfs/write.c | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-) (limited to 'fs/nfs/write.c') diff --git a/fs/nfs/write.c b/fs/nfs/write.c index a28123be08a..6d40a538e3d 100644 --- a/fs/nfs/write.c +++ b/fs/nfs/write.c @@ -1611,15 +1611,16 @@ int nfs_migrate_page(struct address_space *mapping, struct page *newpage, if (ret) goto out_unlock; page_cache_get(newpage); + spin_lock(&mapping->host->i_lock); req->wb_page = newpage; SetPagePrivate(newpage); - set_page_private(newpage, page_private(page)); + set_page_private(newpage, (unsigned long)req); ClearPagePrivate(page); set_page_private(page, 0); + spin_unlock(&mapping->host->i_lock); page_cache_release(page); out_unlock: nfs_clear_page_tag_locked(req); - nfs_release_request(req); out: return ret; } -- cgit v1.2.3-70-g09d2