diff options
Diffstat (limited to 'fs')
46 files changed, 1017 insertions, 503 deletions
diff --git a/fs/affs/namei.c b/fs/affs/namei.c index d8aa0ae3d037..41c5749f4db7 100644 --- a/fs/affs/namei.c +++ b/fs/affs/namei.c @@ -201,14 +201,16 @@ affs_lookup(struct inode *dir, struct dentry *dentry, unsigned int flags) struct super_block *sb = dir->i_sb; struct buffer_head *bh; struct inode *inode = NULL; + struct dentry *res; pr_debug("%s(\"%pd\")\n", __func__, dentry); affs_lock_dir(dir); bh = affs_find_entry(dir, dentry); - affs_unlock_dir(dir); - if (IS_ERR(bh)) + if (IS_ERR(bh)) { + affs_unlock_dir(dir); return ERR_CAST(bh); + } if (bh) { u32 ino = bh->b_blocknr; @@ -222,11 +224,12 @@ affs_lookup(struct inode *dir, struct dentry *dentry, unsigned int flags) } affs_brelse(bh); inode = affs_iget(sb, ino); - if (IS_ERR(inode)) - return ERR_CAST(inode); } - d_add(dentry, inode); - return NULL; + res = d_splice_alias(inode, dentry); + if (!IS_ERR_OR_NULL(res)) + res->d_fsdata = dentry->d_fsdata; + affs_unlock_dir(dir); + return res; } int @@ -1078,8 +1078,8 @@ static struct kioctx *lookup_ioctx(unsigned long ctx_id) ctx = rcu_dereference(table->table[id]); if (ctx && ctx->user_id == ctx_id) { - percpu_ref_get(&ctx->users); - ret = ctx; + if (percpu_ref_tryget_live(&ctx->users)) + ret = ctx; } out: rcu_read_unlock(); diff --git a/fs/befs/linuxvfs.c b/fs/befs/linuxvfs.c index af2832aaeec5..4700b4534439 100644 --- a/fs/befs/linuxvfs.c +++ b/fs/befs/linuxvfs.c @@ -198,23 +198,16 @@ befs_lookup(struct inode *dir, struct dentry *dentry, unsigned int flags) if (ret == BEFS_BT_NOT_FOUND) { befs_debug(sb, "<--- %s %pd not found", __func__, dentry); - d_add(dentry, NULL); - return ERR_PTR(-ENOENT); - + inode = NULL; } else if (ret != BEFS_OK || offset == 0) { befs_error(sb, "<--- %s Error", __func__); - return ERR_PTR(-ENODATA); + inode = ERR_PTR(-ENODATA); + } else { + inode = befs_iget(dir->i_sb, (ino_t) offset); } - - inode = befs_iget(dir->i_sb, (ino_t) offset); - if (IS_ERR(inode)) - return ERR_CAST(inode); - - d_add(dentry, inode); - befs_debug(sb, "<--- %s", __func__); - return NULL; + return d_splice_alias(inode, dentry); } static int diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c index 233a2bf70351..035a529084b9 100644 --- a/fs/btrfs/inode.c +++ b/fs/btrfs/inode.c @@ -6586,8 +6586,7 @@ static int btrfs_mknod(struct inode *dir, struct dentry *dentry, goto out_unlock_inode; } else { btrfs_update_inode(trans, root, inode); - unlock_new_inode(inode); - d_instantiate(dentry, inode); + d_instantiate_new(dentry, inode); } out_unlock: @@ -6663,8 +6662,7 @@ static int btrfs_create(struct inode *dir, struct dentry *dentry, goto out_unlock_inode; BTRFS_I(inode)->io_tree.ops = &btrfs_extent_io_ops; - unlock_new_inode(inode); - d_instantiate(dentry, inode); + d_instantiate_new(dentry, inode); out_unlock: btrfs_end_transaction(trans); @@ -6809,12 +6807,7 @@ static int btrfs_mkdir(struct inode *dir, struct dentry *dentry, umode_t mode) if (err) goto out_fail_inode; - d_instantiate(dentry, inode); - /* - * mkdir is special. We're unlocking after we call d_instantiate - * to avoid a race with nfsd calling d_instantiate. - */ - unlock_new_inode(inode); + d_instantiate_new(dentry, inode); drop_on_err = 0; out_fail: @@ -9124,7 +9117,8 @@ static int btrfs_truncate(struct inode *inode, bool skip_writeback) BTRFS_EXTENT_DATA_KEY); trans->block_rsv = &fs_info->trans_block_rsv; if (ret != -ENOSPC && ret != -EAGAIN) { - err = ret; + if (ret < 0) + err = ret; break; } @@ -10257,8 +10251,7 @@ static int btrfs_symlink(struct inode *dir, struct dentry *dentry, goto out_unlock_inode; } - unlock_new_inode(inode); - d_instantiate(dentry, inode); + d_instantiate_new(dentry, inode); out_unlock: btrfs_end_transaction(trans); diff --git a/fs/cachefiles/namei.c b/fs/cachefiles/namei.c index 0daa1e3fe0df..ab0bbe93b398 100644 --- a/fs/cachefiles/namei.c +++ b/fs/cachefiles/namei.c @@ -572,6 +572,11 @@ lookup_again: if (ret < 0) goto create_error; + if (unlikely(d_unhashed(next))) { + dput(next); + inode_unlock(d_inode(dir)); + goto lookup_again; + } ASSERT(d_backing_inode(next)); _debug("mkdir -> %p{%p{ino=%lu}}", @@ -764,6 +769,7 @@ struct dentry *cachefiles_get_directory(struct cachefiles_cache *cache, /* search the current directory for the element name */ inode_lock(d_inode(dir)); +retry: start = jiffies; subdir = lookup_one_len(dirname, dir, strlen(dirname)); cachefiles_hist(cachefiles_lookup_histogram, start); @@ -793,6 +799,10 @@ struct dentry *cachefiles_get_directory(struct cachefiles_cache *cache, if (ret < 0) goto mkdir_error; + if (unlikely(d_unhashed(subdir))) { + dput(subdir); + goto retry; + } ASSERT(d_backing_inode(subdir)); _debug("mkdir -> %p{%p{ino=%lu}}", diff --git a/fs/cifs/Kconfig b/fs/cifs/Kconfig index 5f132d59dfc2..d61e2de8d0eb 100644 --- a/fs/cifs/Kconfig +++ b/fs/cifs/Kconfig @@ -197,7 +197,7 @@ config CIFS_SMB311 config CIFS_SMB_DIRECT bool "SMB Direct support (Experimental)" - depends on CIFS=m && INFINIBAND && INFINIBAND_ADDR_TRANS || CIFS=y && INFINIBAND=y && INFINIBAND_ADDR_TRANS=y + depends on CIFS=m && INFINIBAND_ADDR_TRANS || CIFS=y && INFINIBAND_ADDR_TRANS=y help Enables SMB Direct experimental support for SMB 3.0, 3.02 and 3.1.1. SMB Direct allows transferring SMB packets over RDMA. If unsure, diff --git a/fs/cramfs/inode.c b/fs/cramfs/inode.c index d7a600e00f28..b926de280de0 100644 --- a/fs/cramfs/inode.c +++ b/fs/cramfs/inode.c @@ -492,7 +492,7 @@ static void cramfs_kill_sb(struct super_block *sb) { struct cramfs_sb_info *sbi = CRAMFS_SB(sb); - if (IS_ENABLED(CCONFIG_CRAMFS_MTD) && sb->s_mtd) { + if (IS_ENABLED(CONFIG_CRAMFS_MTD) && sb->s_mtd) { if (sbi && sbi->mtd_point_size) mtd_unpoint(sb->s_mtd, 0, sbi->mtd_point_size); kill_mtd_super(sb); diff --git a/fs/dcache.c b/fs/dcache.c index 86d2de63461e..2acfc69878f5 100644 --- a/fs/dcache.c +++ b/fs/dcache.c @@ -1899,6 +1899,28 @@ void d_instantiate(struct dentry *entry, struct inode * inode) } EXPORT_SYMBOL(d_instantiate); +/* + * This should be equivalent to d_instantiate() + unlock_new_inode(), + * with lockdep-related part of unlock_new_inode() done before + * anything else. Use that instead of open-coding d_instantiate()/ + * unlock_new_inode() combinations. + */ +void d_instantiate_new(struct dentry *entry, struct inode *inode) +{ + BUG_ON(!hlist_unhashed(&entry->d_u.d_alias)); + BUG_ON(!inode); + lockdep_annotate_inode_mutex_key(inode); + security_d_instantiate(entry, inode); + spin_lock(&inode->i_lock); + __d_instantiate(entry, inode); + WARN_ON(!(inode->i_state & I_NEW)); + inode->i_state &= ~I_NEW; + smp_mb(); + wake_up_bit(&inode->i_state, __I_NEW); + spin_unlock(&inode->i_lock); +} +EXPORT_SYMBOL(d_instantiate_new); + /** * d_instantiate_no_diralias - instantiate a non-aliased dentry * @entry: dentry to complete diff --git a/fs/ecryptfs/inode.c b/fs/ecryptfs/inode.c index 97d17eaeba07..49121e5a8de2 100644 --- a/fs/ecryptfs/inode.c +++ b/fs/ecryptfs/inode.c @@ -283,8 +283,7 @@ ecryptfs_create(struct inode *directory_inode, struct dentry *ecryptfs_dentry, iget_failed(ecryptfs_inode); goto out; } - unlock_new_inode(ecryptfs_inode); - d_instantiate(ecryptfs_dentry, ecryptfs_inode); + d_instantiate_new(ecryptfs_dentry, ecryptfs_inode); out: return rc; } diff --git a/fs/ext2/inode.c b/fs/ext2/inode.c index 1e01fabef130..71635909df3b 100644 --- a/fs/ext2/inode.c +++ b/fs/ext2/inode.c @@ -1264,21 +1264,11 @@ do_indirects: static void ext2_truncate_blocks(struct inode *inode, loff_t offset) { - /* - * XXX: it seems like a bug here that we don't allow - * IS_APPEND inode to have blocks-past-i_size trimmed off. - * review and fix this. - * - * Also would be nice to be able to handle IO errors and such, - * but that's probably too much to ask. - */ if (!(S_ISREG(inode->i_mode) || S_ISDIR(inode->i_mode) || S_ISLNK(inode->i_mode))) return; if (ext2_inode_is_fast_symlink(inode)) return; - if (IS_APPEND(inode) || IS_IMMUTABLE(inode)) - return; dax_sem_down_write(EXT2_I(inode)); __ext2_truncate_blocks(inode, offset); diff --git a/fs/ext2/namei.c b/fs/ext2/namei.c index 55f7caadb093..152453a91877 100644 --- a/fs/ext2/namei.c +++ b/fs/ext2/namei.c @@ -41,8 +41,7 @@ static inline int ext2_add_nondir(struct dentry *dentry, struct inode *inode) { int err = ext2_add_link(dentry, inode); if (!err) { - unlock_new_inode(inode); - d_instantiate(dentry, inode); + d_instantiate_new(dentry, inode); return 0; } inode_dec_link_count(inode); @@ -255,8 +254,7 @@ static int ext2_mkdir(struct inode * dir, struct dentry * dentry, umode_t mode) if (err) goto out_fail; - unlock_new_inode(inode); - d_instantiate(dentry, inode); + d_instantiate_new(dentry, inode); out: return err; diff --git a/fs/ext4/namei.c b/fs/ext4/namei.c index 2feb546dfcaf..2a4c25c4681d 100644 --- a/fs/ext4/namei.c +++ b/fs/ext4/namei.c @@ -2411,8 +2411,7 @@ static int ext4_add_nondir(handle_t *handle, int err = ext4_add_entry(handle, dentry, inode); if (!err) { ext4_mark_inode_dirty(handle, inode); - unlock_new_inode(inode); - d_instantiate(dentry, inode); + d_instantiate_new(dentry, inode); return 0; } drop_nlink(inode); @@ -2651,8 +2650,7 @@ out_clear_inode: err = ext4_mark_inode_dirty(handle, dir); if (err) goto out_clear_inode; - unlock_new_inode(inode); - d_instantiate(dentry, inode); + d_instantiate_new(dentry, inode); if (IS_DIRSYNC(dir)) ext4_handle_sync(handle); diff --git a/fs/f2fs/namei.c b/fs/f2fs/namei.c index ae10fe34f63d..4455d6c82808 100644 --- a/fs/f2fs/namei.c +++ b/fs/f2fs/namei.c @@ -294,8 +294,7 @@ static int f2fs_create(struct inode *dir, struct dentry *dentry, umode_t mode, alloc_nid_done(sbi, ino); - d_instantiate(dentry, inode); - unlock_new_inode(inode); + d_instantiate_new(dentry, inode); if (IS_DIRSYNC(dir)) f2fs_sync_fs(sbi->sb, 1); @@ -597,8 +596,7 @@ static int f2fs_symlink(struct inode *dir, struct dentry *dentry, err = page_symlink(inode, disk_link.name, disk_link.len); err_out: - d_instantiate(dentry, inode); - unlock_new_inode(inode); + d_instantiate_new(dentry, inode); /* * Let's flush symlink data in order to avoid broken symlink as much as @@ -661,8 +659,7 @@ static int f2fs_mkdir(struct inode *dir, struct dentry *dentry, umode_t mode) alloc_nid_done(sbi, inode->i_ino); - d_instantiate(dentry, inode); - unlock_new_inode(inode); + d_instantiate_new(dentry, inode); if (IS_DIRSYNC(dir)) f2fs_sync_fs(sbi->sb, 1); @@ -713,8 +710,7 @@ static int f2fs_mknod(struct inode *dir, struct dentry *dentry, alloc_nid_done(sbi, inode->i_ino); - d_instantiate(dentry, inode); - unlock_new_inode(inode); + d_instantiate_new(dentry, inode); if (IS_DIRSYNC(dir)) f2fs_sync_fs(sbi->sb, 1); diff --git a/fs/jffs2/dir.c b/fs/jffs2/dir.c index 4ef97805eb27..b2944f9218f7 100644 --- a/fs/jffs2/dir.c +++ b/fs/jffs2/dir.c @@ -209,8 +209,7 @@ static int jffs2_create(struct inode *dir_i, struct dentry *dentry, __func__, inode->i_ino, inode->i_mode, inode->i_nlink, f->inocache->pino_nlink, inode->i_mapping->nrpages); - unlock_new_inode(inode); - d_instantiate(dentry, inode); + d_instantiate_new(dentry, inode); return 0; fail: @@ -430,8 +429,7 @@ static int jffs2_symlink (struct inode *dir_i, struct dentry *dentry, const char mutex_unlock(&dir_f->sem); jffs2_complete_reservation(c); - unlock_new_inode(inode); - d_instantiate(dentry, inode); + d_instantiate_new(dentry, inode); return 0; fail: @@ -575,8 +573,7 @@ static int jffs2_mkdir (struct inode *dir_i, struct dentry *dentry, umode_t mode mutex_unlock(&dir_f->sem); jffs2_complete_reservation(c); - unlock_new_inode(inode); - d_instantiate(dentry, inode); + d_instantiate_new(dentry, inode); return 0; fail: @@ -747,8 +744,7 @@ static int jffs2_mknod (struct inode *dir_i, struct dentry *dentry, umode_t mode mutex_unlock(&dir_f->sem); jffs2_complete_reservation(c); - unlock_new_inode(inode); - d_instantiate(dentry, inode); + d_instantiate_new(dentry, inode); return 0; fail: diff --git a/fs/jfs/namei.c b/fs/jfs/namei.c index b41596d71858..56c3fcbfe80e 100644 --- a/fs/jfs/namei.c +++ b/fs/jfs/namei.c @@ -178,8 +178,7 @@ static int jfs_create(struct inode *dip, struct dentry *dentry, umode_t mode, unlock_new_inode(ip); iput(ip); } else { - unlock_new_inode(ip); - d_instantiate(dentry, ip); + d_instantiate_new(dentry, ip); } out2: @@ -313,8 +312,7 @@ static int jfs_mkdir(struct inode *dip, struct dentry *dentry, umode_t mode) unlock_new_inode(ip); iput(ip); } else { - unlock_new_inode(ip); - d_instantiate(dentry, ip); + d_instantiate_new(dentry, ip); } out2: @@ -1059,8 +1057,7 @@ static int jfs_symlink(struct inode *dip, struct dentry *dentry, unlock_new_inode(ip); iput(ip); } else { - unlock_new_inode(ip); - d_instantiate(dentry, ip); + d_instantiate_new(dentry, ip); } out2: @@ -1447,8 +1444,7 @@ static int jfs_mknod(struct inode *dir, struct dentry *dentry, unlock_new_inode(ip); iput(ip); } else { - unlock_new_inode(ip); - d_instantiate(dentry, ip); + d_instantiate_new(dentry, ip); } out1: diff --git a/fs/kernfs/mount.c b/fs/kernfs/mount.c index 26dd9a50f383..ff2716f9322e 100644 --- a/fs/kernfs/mount.c +++ b/fs/kernfs/mount.c @@ -316,6 +316,7 @@ struct dentry *kernfs_mount_ns(struct file_system_type *fs_type, int flags, info->root = root; info->ns = ns; + INIT_LIST_HEAD(&info->node); sb = sget_userns(fs_type, kernfs_test_super, kernfs_set_super, flags, &init_user_ns, info); diff --git a/fs/nfs/callback_proc.c b/fs/nfs/callback_proc.c index 9d24aabcd8bb..64c214fb9da6 100644 --- a/fs/nfs/callback_proc.c +++ b/fs/nfs/callback_proc.c @@ -40,7 +40,9 @@ __be32 nfs4_callback_getattr(void *argp, void *resp, rpc_peeraddr2str(cps->clp->cl_rpcclient, RPC_DISPLAY_ADDR)); inode = nfs_delegation_find_inode(cps->clp, &args->fh); - if (inode == NULL) { + if (IS_ERR(inode)) { + if (inode == ERR_PTR(-EAGAIN)) + res->status = htonl(NFS4ERR_DELAY); trace_nfs4_cb_getattr(cps->clp, &args->fh, NULL, -ntohl(res->status)); goto out; @@ -86,7 +88,9 @@ __be32 nfs4_callback_recall(void *argp, void *resp, res = htonl(NFS4ERR_BADHANDLE); inode = nfs_delegation_find_inode(cps->clp, &args->fh); - if (inode == NULL) { + if (IS_ERR(inode)) { + if (inode == ERR_PTR(-EAGAIN)) + res = htonl(NFS4ERR_DELAY); trace_nfs4_cb_recall(cps->clp, &args->fh, NULL, &args->stateid, -ntohl(res)); goto out; @@ -124,7 +128,6 @@ static struct inode *nfs_layout_find_inode_by_stateid(struct nfs_client *clp, struct inode *inode; struct pnfs_layout_hdr *lo; -restart: list_for_each_entry_rcu(server, &clp->cl_superblocks, client_link) { list_for_each_entry(lo, &server->layouts, plh_layouts) { if (stateid != NULL && @@ -132,20 +135,20 @@ restart: continue; inode = igrab(lo->plh_inode); if (!inode) - continue; + return ERR_PTR(-EAGAIN); if (!nfs_sb_active(inode->i_sb)) { rcu_read_unlock(); spin_unlock(&clp->cl_lock); iput(inode); spin_lock(&clp->cl_lock); rcu_read_lock(); - goto restart; + return ERR_PTR(-EAGAIN); } return inode; } } - return NULL; + return ERR_PTR(-ENOENT); } /* @@ -162,7 +165,6 @@ static struct inode *nfs_layout_find_inode_by_fh(struct nfs_client *clp, struct inode *inode; struct pnfs_layout_hdr *lo; -restart: list_for_each_entry_rcu(server, &clp->cl_superblocks, client_link) { list_for_each_entry(lo, &server->layouts, plh_layouts) { nfsi = NFS_I(lo->plh_inode); @@ -172,20 +174,20 @@ restart: continue; inode = igrab(lo->plh_inode); if (!inode) - continue; + return ERR_PTR(-EAGAIN); if (!nfs_sb_active(inode->i_sb)) { rcu_read_unlock(); spin_unlock(&clp->cl_lock); iput(inode); spin_lock(&clp->cl_lock); rcu_read_lock(); - goto restart; + return ERR_PTR(-EAGAIN); } return inode; } } - return NULL; + return ERR_PTR(-ENOENT); } static struct inode *nfs_layout_find_inode(struct nfs_client *clp, @@ -197,7 +199,7 @@ static struct inode *nfs_layout_find_inode(struct nfs_client *clp, spin_lock(&clp->cl_lock); rcu_read_lock(); inode = nfs_layout_find_inode_by_stateid(clp, stateid); - if (!inode) + if (inode == ERR_PTR(-ENOENT)) inode = nfs_layout_find_inode_by_fh(clp, fh); rcu_read_unlock(); spin_unlock(&clp->cl_lock); @@ -252,8 +254,11 @@ static u32 initiate_file_draining(struct nfs_client *clp, LIST_HEAD(free_me_list); ino = nfs_layout_find_inode(clp, &args->cbl_fh, &args->cbl_stateid); - if (!ino) - goto out; + if (IS_ERR(ino)) { + if (ino == ERR_PTR(-EAGAIN)) + rv = NFS4ERR_DELAY; + goto out_noput; + } pnfs_layoutcommit_inode(ino, false); @@ -299,9 +304,10 @@ unlock: nfs_commit_inode(ino, 0); pnfs_put_layout_hdr(lo); out: + nfs_iput_and_deactive(ino); +out_noput: trace_nfs4_cb_layoutrecall_file(clp, &args->cbl_fh, ino, &args->cbl_stateid, -rv); - nfs_iput_and_deactive(ino); return rv; } @@ -322,6 +328,8 @@ static u32 initiate_bulk_draining(struct nfs_client *clp, static u32 do_callback_layoutrecall(struct nfs_client *clp, struct cb_layoutrecallargs *args) { + write_seqcount_begin(&clp->cl_callback_count); + write_seqcount_end(&clp->cl_callback_count); if (args->cbl_recall_type == RETURN_FILE) return initiate_file_draining(clp, args); return initiate_bulk_draining(clp, args); @@ -420,11 +428,8 @@ validate_seqid(const struct nfs4_slot_table *tbl, const struct nfs4_slot *slot, return htonl(NFS4ERR_SEQ_FALSE_RETRY); } - /* Wraparound */ - if (unlikely(slot->seq_nr == 0xFFFFFFFFU)) { - if (args->csa_sequenceid == 1) - return htonl(NFS4_OK); - } else if (likely(args->csa_sequenceid == slot->seq_nr + 1)) + /* Note: wraparound relies on seq_nr being of type u32 */ + if (likely(args->csa_sequenceid == slot->seq_nr + 1)) return htonl(NFS4_OK); /* Misordered request */ diff --git a/fs/nfs/client.c b/fs/nfs/client.c index b9129e2befea..02e97c29af0c 100644 --- a/fs/nfs/client.c +++ b/fs/nfs/client.c @@ -969,7 +969,8 @@ struct nfs_server *nfs_create_server(struct nfs_mount_info *mount_info, } if (!(fattr->valid & NFS_ATTR_FATTR)) { - error = nfs_mod->rpc_ops->getattr(server, mount_info->mntfh, fattr, NULL); + error = nfs_mod->rpc_ops->getattr(server, mount_info->mntfh, + fattr, NULL, NULL); if (error < 0) { dprintk("nfs_create_server: getattr error = %d\n", -error); goto error; diff --git a/fs/nfs/delegation.c b/fs/nfs/delegation.c index 1819d0d0ba4b..bbd0465535eb 100644 --- a/fs/nfs/delegation.c +++ b/fs/nfs/delegation.c @@ -404,6 +404,10 @@ int nfs_inode_set_delegation(struct inode *inode, struct rpc_cred *cred, trace_nfs4_set_delegation(inode, type); + spin_lock(&inode->i_lock); + if (NFS_I(inode)->cache_validity & (NFS_INO_INVALID_ATTR|NFS_INO_INVALID_ATIME)) + NFS_I(inode)->cache_validity |= NFS_INO_REVAL_FORCED; + spin_unlock(&inode->i_lock); out: spin_unlock(&clp->cl_lock); if (delegation != NULL) @@ -483,38 +487,88 @@ out: int nfs_client_return_marked_delegations(struct nfs_client *clp) { struct nfs_delegation *delegation; + struct nfs_delegation *prev; struct nfs_server *server; struct inode *inode; + struct inode *place_holder = NULL; + struct nfs_delegation *place_holder_deleg = NULL; int err = 0; restart: + /* + * To avoid quadratic looping we hold a reference + * to an inode place_holder. Each time we restart, we + * list nfs_servers from the server of that inode, and + * delegation in the server from the delegations of that + * inode. + * prev is an RCU-protected pointer to a delegation which + * wasn't marked for return and might be a good choice for + * the next place_holder. + */ rcu_read_lock(); - list_for_each_entry_rcu(server, &clp->cl_superblocks, client_link) { - list_for_each_entry_rcu(delegation, &server->delegations, - super_list) { - if (!nfs_delegation_need_return(delegation)) + prev = NULL; + if (place_holder) + server = NFS_SERVER(place_holder); + else + server = list_entry_rcu(clp->cl_superblocks.next, + struct nfs_server, client_link); + list_for_each_entry_from_rcu(server, &clp->cl_superblocks, client_link) { + delegation = NULL; + if (place_holder && server == NFS_SERVER(place_holder)) + delegation = rcu_dereference(NFS_I(place_holder)->delegation); + if (!delegation || delegation != place_holder_deleg) + delegation = list_entry_rcu(server->delegations.next, + struct nfs_delegation, super_list); + list_for_each_entry_from_rcu(delegation, &server->delegations, super_list) { + struct inode *to_put = NULL; + + if (!nfs_delegation_need_return(delegation)) { + prev = delegation; continue; + } if (!nfs_sb_active(server->super)) - continue; + break; /* continue in outer loop */ + + if (prev) { + struct inode *tmp; + + tmp = nfs_delegation_grab_inode(prev); + if (tmp) { + to_put = place_holder; + place_holder = tmp; + place_holder_deleg = prev; + } + } + inode = nfs_delegation_grab_inode(delegation); if (inode == NULL) { rcu_read_unlock(); + if (to_put) + iput(to_put); nfs_sb_deactive(server->super); goto restart; } delegation = nfs_start_delegation_return_locked(NFS_I(inode)); rcu_read_unlock(); + if (to_put) + iput(to_put); + err = nfs_end_delegation_return(inode, delegation, 0); iput(inode); nfs_sb_deactive(server->super); + cond_resched(); if (!err) goto restart; set_bit(NFS4CLNT_DELEGRETURN, &clp->cl_state); + if (place_holder) + iput(place_holder); return err; } } rcu_read_unlock(); + if (place_holder) + iput(place_holder); return 0; } @@ -802,12 +856,14 @@ nfs_delegation_find_inode_server(struct nfs_server *server, if (delegation->inode != NULL && nfs_compare_fh(fhandle, &NFS_I(delegation->inode)->fh) == 0) { res = igrab(delegation->inode); + spin_unlock(&delegation->lock); + if (res != NULL) + return res; + return ERR_PTR(-EAGAIN); } spin_unlock(&delegation->lock); - if (res != NULL) - break; } - return res; + return ERR_PTR(-ENOENT); } /** @@ -822,16 +878,16 @@ struct inode *nfs_delegation_find_inode(struct nfs_client *clp, const struct nfs_fh *fhandle) { struct nfs_server *server; - struct inode *res = NULL; + struct inode *res; rcu_read_lock(); list_for_each_entry_rcu(server, &clp->cl_superblocks, client_link) { res = nfs_delegation_find_inode_server(server, fhandle); - if (res != NULL) - break; + if (res != ERR_PTR(-ENOENT)) + return res; } rcu_read_unlock(); - return res; + return ERR_PTR(-ENOENT); } static void nfs_delegation_mark_reclaim_server(struct nfs_server *server) @@ -887,7 +943,7 @@ restart: &delegation->flags) == 0) continue; if (!nfs_sb_active(server->super)) - continue; + break; /* continue in outer loop */ inode = nfs_delegation_grab_inode(delegation); if (inode == NULL) { rcu_read_unlock(); @@ -904,6 +960,7 @@ restart: } iput(inode); nfs_sb_deactive(server->super); + cond_resched(); goto restart; } } @@ -995,7 +1052,7 @@ restart: &delegation->flags) == 0) continue; if (!nfs_sb_active(server->super)) - continue; + break; /* continue in outer loop */ inode = nfs_delegation_grab_inode(delegation); if (inode == NULL) { rcu_read_unlock(); @@ -1020,6 +1077,7 @@ restart: } iput(inode); nfs_sb_deactive(server->super); + cond_resched(); goto restart; } } diff --git a/fs/nfs/dir.c b/fs/nfs/dir.c index 73f8b43d988c..7a9c14426855 100644 --- a/fs/nfs/dir.c +++ b/fs/nfs/dir.c @@ -1012,13 +1012,25 @@ int nfs_lookup_verify_inode(struct inode *inode, unsigned int flags) if (IS_AUTOMOUNT(inode)) return 0; + + if (flags & LOOKUP_OPEN) { + switch (inode->i_mode & S_IFMT) { + case S_IFREG: + /* A NFSv4 OPEN will revalidate later */ + if (server->caps & NFS_CAP_ATOMIC_OPEN) + goto out; + /* Fallthrough */ + case S_IFDIR: + if (server->flags & NFS_MOUNT_NOCTO) + break; + /* NFS close-to-open cache consistency validation */ + goto out_force; + } + } + /* VFS wants an on-the-wire revalidation */ if (flags & LOOKUP_REVAL) goto out_force; - /* This is an open(2) */ - if ((flags & LOOKUP_OPEN) && !(server->flags & NFS_MOUNT_NOCTO) && - (S_ISREG(inode->i_mode) || S_ISDIR(inode->i_mode))) - goto out_force; out: return (inode->i_nlink == 0) ? -ENOENT : 0; out_force: @@ -1039,13 +1051,15 @@ out_force: * * If LOOKUP_RCU prevents us from performing a full check, return 1 * suggesting a reval is needed. + * + * Note that when creating a new file, or looking up a rename target, + * then it shouldn't be necessary to revalidate a negative dentry. */ static inline int nfs_neg_need_reval(struct inode *dir, struct dentry *dentry, unsigned int flags) { - /* Don't revalidate a negative dentry if we're creating a new file */ - if (flags & LOOKUP_CREATE) + if (flags & (LOOKUP_CREATE | LOOKUP_RENAME_TARGET)) return 0; if (NFS_SERVER(dir)->flags & NFS_MOUNT_LOOKUP_CACHE_NONEG) return 1; @@ -1106,7 +1120,7 @@ static int nfs_lookup_revalidate(struct dentry *dentry, unsigned int flags) goto out_set_verifier; /* Force a full look up iff the parent directory has changed */ - if (!nfs_is_exclusive_create(dir, flags) && + if (!(flags & (LOOKUP_EXCL | LOOKUP_REVAL)) && nfs_check_verifier(dir, dentry, flags & LOOKUP_RCU)) { error = nfs_lookup_verify_inode(inode, flags); if (error) { @@ -1270,11 +1284,13 @@ static void nfs_drop_nlink(struct inode *inode) { spin_lock(&inode->i_lock); /* drop the inode if we're reasonably sure this is the last link */ - if (inode->i_nlink == 1) - clear_nlink(inode); + if (inode->i_nlink > 0) + drop_nlink(inode); + NFS_I(inode)->attr_gencount = nfs_inc_attr_generation_counter(); NFS_I(inode)->cache_validity |= NFS_INO_INVALID_CHANGE | NFS_INO_INVALID_CTIME - | NFS_INO_INVALID_OTHER; + | NFS_INO_INVALID_OTHER + | NFS_INO_REVAL_FORCED; spin_unlock(&inode->i_lock); } @@ -1335,7 +1351,7 @@ struct dentry *nfs_lookup(struct inode *dir, struct dentry * dentry, unsigned in * If we're doing an exclusive create, optimize away the lookup * but don't hash the dentry. */ - if (nfs_is_exclusive_create(dir, flags)) + if (nfs_is_exclusive_create(dir, flags) || flags & LOOKUP_RENAME_TARGET) return NULL; res = ERR_PTR(-ENOMEM); @@ -1640,7 +1656,8 @@ int nfs_instantiate(struct dentry *dentry, struct nfs_fh *fhandle, nfs_set_verifier(dentry, nfs_save_change_attribute(dir)); if (!(fattr->valid & NFS_ATTR_FATTR)) { struct nfs_server *server = NFS_SB(dentry->d_sb); - error = server->nfs_client->rpc_ops->getattr(server, fhandle, fattr, NULL); + error = server->nfs_client->rpc_ops->getattr(server, fhandle, + fattr, NULL, NULL); if (error < 0) goto out_error; } @@ -2036,7 +2053,15 @@ int nfs_rename(struct inode *old_dir, struct dentry *old_dentry, } else error = task->tk_status; rpc_put_task(task); - nfs_mark_for_revalidate(old_inode); + /* Ensure the inode attributes are revalidated */ + if (error == 0) { + spin_lock(&old_inode->i_lock); + NFS_I(old_inode)->attr_gencount = nfs_inc_attr_generation_counter(); + NFS_I(old_inode)->cache_validity |= NFS_INO_INVALID_CHANGE + | NFS_INO_INVALID_CTIME + | NFS_INO_REVAL_FORCED; + spin_unlock(&old_inode->i_lock); + } out: if (rehash) d_rehash(rehash); diff --git a/fs/nfs/export.c b/fs/nfs/export.c index ab5de3246c5c..deecb67638aa 100644 --- a/fs/nfs/export.c +++ b/fs/nfs/export.c @@ -102,7 +102,7 @@ nfs_fh_to_dentry(struct super_block *sb, struct fid *fid, } rpc_ops = NFS_SB(sb)->nfs_client->rpc_ops; - ret = rpc_ops->getattr(NFS_SB(sb), server_fh, fattr, label); + ret = rpc_ops->getattr(NFS_SB(sb), server_fh, fattr, label, NULL); if (ret) { dprintk("%s: getattr failed %d\n", __func__, ret); dentry = ERR_PTR(ret); diff --git a/fs/nfs/flexfilelayout/flexfilelayout.c b/fs/nfs/flexfilelayout/flexfilelayout.c index c75ad982bcfc..3ae038d9c292 100644 --- a/fs/nfs/flexfilelayout/flexfilelayout.c +++ b/fs/nfs/flexfilelayout/flexfilelayout.c @@ -2347,6 +2347,7 @@ static struct pnfs_layoutdriver_type flexfilelayout_type = { .id = LAYOUT_FLEX_FILES, .name = "LAYOUT_FLEX_FILES", .owner = THIS_MODULE, + .flags = PNFS_LAYOUTGET_ON_OPEN, .set_layoutdriver = ff_layout_set_layoutdriver, .alloc_layout_hdr = ff_layout_alloc_layout_hdr, .free_layout_hdr = ff_layout_free_layout_hdr, diff --git a/fs/nfs/inode.c b/fs/nfs/inode.c index 138941a97327..b65aee481d13 100644 --- a/fs/nfs/inode.c +++ b/fs/nfs/inode.c @@ -195,10 +195,16 @@ bool nfs_check_cache_invalid(struct inode *inode, unsigned long flags) static void nfs_set_cache_invalid(struct inode *inode, unsigned long flags) { struct nfs_inode *nfsi = NFS_I(inode); - bool have_delegation = nfs_have_delegated_attributes(inode); + bool have_delegation = NFS_PROTO(inode)->have_delegation(inode, FMODE_READ); + + if (have_delegation) { + if (!(flags & NFS_INO_REVAL_FORCED)) + flags &= ~NFS_INO_INVALID_OTHER; + flags &= ~(NFS_INO_INVALID_CHANGE + | NFS_INO_INVALID_SIZE + | NFS_INO_REVAL_PAGECACHE); + } - if (have_delegation) - flags &= ~(NFS_INO_INVALID_CHANGE|NFS_INO_REVAL_PAGECACHE); if (inode->i_mapping->nrpages == 0) flags &= ~NFS_INO_INVALID_DATA; nfsi->cache_validity |= flags; @@ -448,6 +454,7 @@ nfs_fhget(struct super_block *sb, struct nfs_fh *fh, struct nfs_fattr *fattr, st /* We can't support update_atime(), since the server will reset it */ inode->i_flags |= S_NOATIME|S_NOCMTIME; inode->i_mode = fattr->mode; + nfsi->cache_validity = 0; if ((fattr->valid & NFS_ATTR_FATTR_MODE) == 0 && nfs_server_capable(inode, NFS_CAP_MODE)) nfs_set_cache_invalid(inode, NFS_INO_INVALID_OTHER); @@ -534,6 +541,9 @@ nfs_fhget(struct super_block *sb, struct nfs_fh *fh, struct nfs_fattr *fattr, st inode->i_blocks = nfs_calc_block_size(fattr->du.nfs3.used); } + if (nfsi->cache_validity != 0) + nfsi->cache_validity |= NFS_INO_REVAL_FORCED; + nfs_setsecurity(inode, fattr, label); nfsi->attrtimeo = NFS_MINATTRTIMEO(inode); @@ -667,9 +677,13 @@ void nfs_setattr_update_inode(struct inode *inode, struct iattr *attr, spin_lock(&inode->i_lock); NFS_I(inode)->attr_gencount = fattr->gencount; - nfs_set_cache_invalid(inode, NFS_INO_INVALID_CHANGE - | NFS_INO_INVALID_CTIME); + if ((attr->ia_valid & ATTR_SIZE) != 0) { + nfs_set_cache_invalid(inode, NFS_INO_INVALID_MTIME); + nfs_inc_stats(inode, NFSIOS_SETATTRTRUNC); + nfs_vmtruncate(inode, attr->ia_size); + } if ((attr->ia_valid & (ATTR_MODE|ATTR_UID|ATTR_GID)) != 0) { + NFS_I(inode)->cache_validity &= ~NFS_INO_INVALID_CTIME; if ((attr->ia_valid & ATTR_MODE) != 0) { int mode = attr->ia_mode & S_IALLUGO; mode |= inode->i_mode & ~S_IALLUGO; @@ -679,13 +693,45 @@ void nfs_setattr_update_inode(struct inode *inode, struct iattr *attr, inode->i_uid = attr->ia_uid; if ((attr->ia_valid & ATTR_GID) != 0) inode->i_gid = attr->ia_gid; + if (fattr->valid & NFS_ATTR_FATTR_CTIME) + inode->i_ctime = timespec_to_timespec64(fattr->ctime); + else + nfs_set_cache_invalid(inode, NFS_INO_INVALID_CHANGE + | NFS_INO_INVALID_CTIME); nfs_set_cache_invalid(inode, NFS_INO_INVALID_ACCESS | NFS_INO_INVALID_ACL); } - if ((attr->ia_valid & ATTR_SIZE) != 0) { - nfs_set_cache_invalid(inode, NFS_INO_INVALID_MTIME); - nfs_inc_stats(inode, NFSIOS_SETATTRTRUNC); - nfs_vmtruncate(inode, attr->ia_size); + if (attr->ia_valid & (ATTR_ATIME_SET|ATTR_ATIME)) { + NFS_I(inode)->cache_validity &= ~(NFS_INO_INVALID_ATIME + | NFS_INO_INVALID_CTIME); + if (fattr->valid & NFS_ATTR_FATTR_ATIME) + inode->i_atime = timespec_to_timespec64(fattr->atime); + else if (attr->ia_valid & ATTR_ATIME_SET) + inode->i_atime = attr->ia_atime; + else + nfs_set_cache_invalid(inode, NFS_INO_INVALID_ATIME); + + if (fattr->valid & NFS_ATTR_FATTR_CTIME) + inode->i_ctime = timespec_to_timespec64(fattr->ctime); + else + nfs_set_cache_invalid(inode, NFS_INO_INVALID_CHANGE + | NFS_INO_INVALID_CTIME); + } + if (attr->ia_valid & (ATTR_MTIME_SET|ATTR_MTIME)) { + NFS_I(inode)->cache_validity &= ~(NFS_INO_INVALID_MTIME + | NFS_INO_INVALID_CTIME); + if (fattr->valid & NFS_ATTR_FATTR_MTIME) + inode->i_mtime = timespec_to_timespec64(fattr->mtime); + else if (attr->ia_valid & ATTR_MTIME_SET) + inode->i_mtime = attr->ia_mtime; + else + nfs_set_cache_invalid(inode, NFS_INO_INVALID_MTIME); + + if (fattr->valid & NFS_ATTR_FATTR_CTIME) + inode->i_ctime = timespec_to_timespec64(fattr->ctime); + else + nfs_set_cache_invalid(inode, NFS_INO_INVALID_CHANGE + | NFS_INO_INVALID_CTIME); } if (fattr->valid) nfs_update_inode(inode, fattr); @@ -1097,7 +1143,8 @@ __nfs_revalidate_inode(struct nfs_server *server, struct inode *inode) goto out; } - status = NFS_PROTO(inode)->getattr(server, NFS_FH(inode), fattr, label); + status = NFS_PROTO(inode)->getattr(server, NFS_FH(inode), fattr, + label, inode); if (status != 0) { dfprintk(PAGECACHE, "nfs_revalidate_inode: (%s/%Lu) getattr failed, error=%d\n", inode->i_sb->s_id, @@ -1353,8 +1400,9 @@ static int nfs_check_inode_attributes(struct inode *inode, struct nfs_fattr *fat unsigned long invalid = 0; struct timespec ts; - if (nfs_have_delegated_attributes(inode)) + if (NFS_PROTO(inode)->have_delegation(inode, FMODE_READ)) return 0; + /* Has the inode gone and changed behind our back? */ if ((fattr->valid & NFS_ATTR_FATTR_FILEID) && nfsi->fileid != fattr->fileid) return -ESTALE; @@ -1407,7 +1455,7 @@ static int nfs_check_inode_attributes(struct inode *inode, struct nfs_fattr *fat invalid |= NFS_INO_INVALID_ATIME; if (invalid != 0) - nfs_set_cache_invalid(inode, invalid | NFS_INO_REVAL_FORCED); + nfs_set_cache_invalid(inode, invalid); nfsi->read_cache_jiffies = fattr->time_start; return 0; @@ -1636,7 +1684,8 @@ int nfs_post_op_update_inode(struct inode *inode, struct nfs_fattr *fattr) nfs_fattr_set_barrier(fattr); status = nfs_post_op_update_inode_locked(inode, fattr, NFS_INO_INVALID_CHANGE - | NFS_INO_INVALID_CTIME); + | NFS_INO_INVALID_CTIME + | NFS_INO_REVAL_FORCED); spin_unlock(&inode->i_lock); return status; @@ -1753,6 +1802,8 @@ static int nfs_update_inode(struct inode *inode, struct nfs_fattr *fattr) unsigned long save_cache_validity; bool have_writers = nfs_file_has_buffered_writers(nfsi); bool cache_revalidated = true; + bool attr_changed = false; + bool have_delegation; dfprintk(VFS, "NFS: %s(%s/%lu fh_crc=0x%08x ct=%d info=0x%x)\n", __func__, inode->i_sb->s_id, inode->i_ino, @@ -1787,6 +1838,9 @@ static int nfs_update_inode(struct inode *inode, struct nfs_fattr *fattr) !IS_AUTOMOUNT(inode)) server->fsid = fattr->fsid; + /* Save the delegation state before clearing cache_validity */ + have_delegation = nfs_have_delegated_attributes(inode); + /* * Update the read time so we don't revalidate too often. */ @@ -1809,12 +1863,9 @@ static int nfs_update_inode(struct inode *inode, struct nfs_fattr *fattr) /* More cache consistency checks */ if (fattr->valid & NFS_ATTR_FATTR_CHANGE) { if (!inode_eq_iversion_raw(inode, fattr->change_attr)) { - dprintk("NFS: change_attr change on server for file %s/%ld\n", - inode->i_sb->s_id, inode->i_ino); /* Could it be a race with writeback? */ - if (!have_writers) { - invalid |= NFS_INO_INVALID_CHANGE - | NFS_INO_INVALID_DATA + if (!(have_writers || have_delegation)) { + invalid |= NFS_INO_INVALID_DATA | NFS_INO_INVALID_ACCESS | NFS_INO_INVALID_ACL; /* Force revalidate of all attributes */ @@ -1824,8 +1875,12 @@ static int nfs_update_inode(struct inode *inode, struct nfs_fattr *fattr) | NFS_INO_INVALID_OTHER; if (S_ISDIR(inode->i_mode)) nfs_force_lookup_revalidate(inode); + dprintk("NFS: change_attr change on server for file %s/%ld\n", + inode->i_sb->s_id, + inode->i_ino); } inode_set_iversion_raw(inode, fattr->change_attr); + attr_changed = true; } } else { nfsi->cache_validity |= save_cache_validity & @@ -1857,13 +1912,14 @@ static int nfs_update_inode(struct inode *inode, struct nfs_fattr *fattr) if (fattr->valid & NFS_ATTR_FATTR_SIZE) { new_isize = nfs_size_to_loff_t(fattr->size); cur_isize = i_size_read(inode); - if (new_isize != cur_isize) { + if (new_isize != cur_isize && !have_delegation) { /* Do we perhaps have any outstanding writes, or has * the file grown beyond our last write? */ if (!nfs_have_writebacks(inode) || new_isize > cur_isize) { i_size_write(inode, new_isize); if (!have_writers) invalid |= NFS_INO_INVALID_DATA; + attr_changed = true; } dprintk("NFS: isize change on server for file %s/%ld " "(%Ld to %Ld)\n", @@ -1896,14 +1952,12 @@ static int nfs_update_inode(struct inode *inode, struct nfs_fattr *fattr) newmode |= fattr->mode & S_IALLUGO; inode->i_mode = newmode; invalid |= NFS_INO_INVALID_ACCESS - | NFS_INO_INVALID_ACL - | NFS_INO_INVALID_OTHER; + | NFS_INO_INVALID_ACL; + attr_changed = true; } } else if (server->caps & NFS_CAP_MODE) { nfsi->cache_validity |= save_cache_validity & - (NFS_INO_INVALID_ACCESS - | NFS_INO_INVALID_ACL - | NFS_INO_INVALID_OTHER + (NFS_INO_INVALID_OTHER | NFS_INO_REVAL_FORCED); cache_revalidated = false; } @@ -1911,15 +1965,13 @@ static int nfs_update_inode(struct inode *inode, struct nfs_fattr *fattr) if (fattr->valid & NFS_ATTR_FATTR_OWNER) { if (!uid_eq(inode->i_uid, fattr->uid)) { invalid |= NFS_INO_INVALID_ACCESS - | NFS_INO_INVALID_ACL - | NFS_INO_INVALID_OTHER; + | NFS_INO_INVALID_ACL; inode->i_uid = fattr->uid; + attr_changed = true; } } else if (server->caps & NFS_CAP_OWNER) { nfsi->cache_validity |= save_cache_validity & - (NFS_INO_INVALID_ACCESS - | NFS_INO_INVALID_ACL - | NFS_INO_INVALID_OTHER + (NFS_INO_INVALID_OTHER | NFS_INO_REVAL_FORCED); cache_revalidated = false; } @@ -1927,25 +1979,23 @@ static int nfs_update_inode(struct inode *inode, struct nfs_fattr *fattr) if (fattr->valid & NFS_ATTR_FATTR_GROUP) { if (!gid_eq(inode->i_gid, fattr->gid)) { invalid |= NFS_INO_INVALID_ACCESS - | NFS_INO_INVALID_ACL - | NFS_INO_INVALID_OTHER; + | NFS_INO_INVALID_ACL; inode->i_gid = fattr->gid; + attr_changed = true; } } else if (server->caps & NFS_CAP_OWNER_GROUP) { nfsi->cache_validity |= save_cache_validity & - (NFS_INO_INVALID_ACCESS - | NFS_INO_INVALID_ACL - | NFS_INO_INVALID_OTHER + (NFS_INO_INVALID_OTHER | NFS_INO_REVAL_FORCED); cache_revalidated = false; } if (fattr->valid & NFS_ATTR_FATTR_NLINK) { if (inode->i_nlink != fattr->nlink) { - invalid |= NFS_INO_INVALID_OTHER; if (S_ISDIR(inode->i_mode)) invalid |= NFS_INO_INVALID_DATA; set_nlink(inode, fattr->nlink); + attr_changed = true; } } else if (server->caps & NFS_CAP_NLINK) { nfsi->cache_validity |= save_cache_validity & @@ -1965,7 +2015,7 @@ static int nfs_update_inode(struct inode *inode, struct nfs_fattr *fattr) cache_revalidated = false; /* Update attrtimeo value if we're out of the unstable period */ - if (invalid & NFS_INO_INVALID_ATTR) { + if (attr_changed) { invalid &= ~NFS_INO_INVALID_ATTR; nfs_inc_stats(inode, NFSIOS_ATTRINVALIDATE); nfsi->attrtimeo = NFS_MINATTRTIMEO(inode); @@ -1991,9 +2041,7 @@ static int nfs_update_inode(struct inode *inode, struct nfs_fattr *fattr) if (!(S_ISREG(inode->i_mode) || S_ISDIR(inode->i_mode) || S_ISLNK(inode->i_mode))) invalid &= ~NFS_INO_INVALID_DATA; - if (!NFS_PROTO(inode)->have_delegation(inode, FMODE_READ) || - (save_cache_validity & NFS_INO_REVAL_FORCED)) - nfs_set_cache_invalid(inode, invalid); + nfs_set_cache_invalid(inode, invalid); return 0; out_err: diff --git a/fs/nfs/nfs3proc.c b/fs/nfs/nfs3proc.c index eadf1ab31d16..ec8a9efa268f 100644 --- a/fs/nfs/nfs3proc.c +++ b/fs/nfs/nfs3proc.c @@ -101,7 +101,8 @@ nfs3_proc_get_root(struct nfs_server *server, struct nfs_fh *fhandle, */ static int nfs3_proc_getattr(struct nfs_server *server, struct nfs_fh *fhandle, - struct nfs_fattr *fattr, struct nfs4_label *label) + struct nfs_fattr *fattr, struct nfs4_label *label, + struct inode *inode) { struct rpc_message msg = { .rpc_proc = &nfs3_procedures[NFS3PROC_GETATTR], @@ -414,7 +415,9 @@ out: } static void -nfs3_proc_unlink_setup(struct rpc_message *msg, struct dentry *dentry) +nfs3_proc_unlink_setup(struct rpc_message *msg, + struct dentry *dentry, + struct inode *inode) { msg->rpc_proc = &nfs3_procedures[NFS3PROC_REMOVE]; } @@ -823,7 +826,8 @@ static int nfs3_write_done(struct rpc_task *task, struct nfs_pgio_header *hdr) } static void nfs3_proc_write_setup(struct nfs_pgio_header *hdr, - struct rpc_message *msg) + struct rpc_message *msg, + struct rpc_clnt **clnt) { msg->rpc_proc = &nfs3_procedures[NFS3PROC_WRITE]; } @@ -844,7 +848,8 @@ static int nfs3_commit_done(struct rpc_task *task, struct nfs_commit_data *data) return 0; } -static void nfs3_proc_commit_setup(struct nfs_commit_data *data, struct rpc_message *msg) +static void nfs3_proc_commit_setup(struct nfs_commit_data *data, struct rpc_message *msg, + struct rpc_clnt **clnt) { msg->rpc_proc = &nfs3_procedures[NFS3PROC_COMMIT]; } diff --git a/fs/nfs/nfs42proc.c b/fs/nfs/nfs42proc.c index 9c374441f660..5f59b6f65a42 100644 --- a/fs/nfs/nfs42proc.c +++ b/fs/nfs/nfs42proc.c @@ -370,6 +370,10 @@ nfs42_layoutstat_done(struct rpc_task *task, void *calldata) switch (task->tk_status) { case 0: break; + case -NFS4ERR_BADHANDLE: + case -ESTALE: + pnfs_destroy_layout(NFS_I(inode)); + break; case -NFS4ERR_EXPIRED: case -NFS4ERR_ADMIN_REVOKED: case -NFS4ERR_DELEG_REVOKED: @@ -462,7 +466,7 @@ int nfs42_proc_layoutstats_generic(struct nfs_server *server, nfs42_layoutstat_release(data); return -EAGAIN; } - nfs4_init_sequence(&data->args.seq_args, &data->res.seq_res, 0); + nfs4_init_sequence(&data->args.seq_args, &data->res.seq_res, 0, 0); task = rpc_run_task(&task_setup); if (IS_ERR(task)) return PTR_ERR(task); diff --git a/fs/nfs/nfs4_fs.h b/fs/nfs/nfs4_fs.h index b374f680830c..137e18abb7e7 100644 --- a/fs/nfs/nfs4_fs.h +++ b/fs/nfs/nfs4_fs.h @@ -212,6 +212,31 @@ struct nfs4_state_recovery_ops { struct rpc_cred *); }; +struct nfs4_opendata { + struct kref kref; + struct nfs_openargs o_arg; + struct nfs_openres o_res; + struct nfs_open_confirmargs c_arg; + struct nfs_open_confirmres c_res; + struct nfs4_string owner_name; + struct nfs4_string group_name; + struct nfs4_label *a_label; + struct nfs_fattr f_attr; + struct nfs4_label *f_label; + struct dentry *dir; + struct dentry *dentry; + struct nfs4_state_owner *owner; + struct nfs4_state *state; + struct iattr attrs; + struct nfs4_layoutget *lgp; + unsigned long timestamp; + bool rpc_done; + bool file_created; + bool is_recover; + bool cancelled; + int rpc_status; +}; + struct nfs4_add_xprt_data { struct nfs_client *clp; struct rpc_cred *cred; @@ -251,7 +276,7 @@ extern int nfs4_handle_exception(struct nfs_server *, int, struct nfs4_exception extern int nfs4_call_sync(struct rpc_clnt *, struct nfs_server *, struct rpc_message *, struct nfs4_sequence_args *, struct nfs4_sequence_res *, int); -extern void nfs4_init_sequence(struct nfs4_sequence_args *, struct nfs4_sequence_res *, int); +extern void nfs4_init_sequence(struct nfs4_sequence_args *, struct nfs4_sequence_res *, int, int); extern int nfs4_proc_setclientid(struct nfs_client *, u32, unsigned short, struct rpc_cred *, struct nfs4_setclientid_res *); extern int nfs4_proc_setclientid_confirm(struct nfs_client *, struct nfs4_setclientid_res *arg, struct rpc_cred *); extern int nfs4_proc_get_rootfh(struct nfs_server *, struct nfs_fh *, struct nfs_fsinfo *, bool); diff --git a/fs/nfs/nfs4idmap.c b/fs/nfs/nfs4idmap.c index 22dc30a679a0..b6f9d84ba19b 100644 --- a/fs/nfs/nfs4idmap.c +++ b/fs/nfs/nfs4idmap.c @@ -343,7 +343,7 @@ static ssize_t nfs_idmap_lookup_name(__u32 id, const char *type, char *buf, int id_len; ssize_t ret; - id_len = snprintf(id_str, sizeof(id_str), "%u", id); + id_len = nfs_map_numeric_to_string(id, id_str, sizeof(id_str)); ret = nfs_idmap_get_key(id_str, id_len, type, buf, buflen, idmap); if (ret < 0) return -EINVAL; @@ -627,7 +627,8 @@ static int nfs_idmap_read_and_verify_message(struct idmap_msg *im, if (strcmp(upcall->im_name, im->im_name) != 0) break; /* Note: here we store the NUL terminator too */ - len = sprintf(id_str, "%d", im->im_id) + 1; + len = 1 + nfs_map_numeric_to_string(im->im_id, id_str, + sizeof(id_str)); ret = nfs_idmap_instantiate(key, authkey, id_str, len); break; case IDMAP_CONV_IDTONAME: diff --git a/fs/nfs/nfs4proc.c b/fs/nfs/nfs4proc.c index b71757e85066..ed45090e4df6 100644 --- a/fs/nfs/nfs4proc.c +++ b/fs/nfs/nfs4proc.c @@ -71,6 +71,8 @@ #define NFSDBG_FACILITY NFSDBG_PROC +#define NFS4_BITMASK_SZ 3 + #define NFS4_POLL_RETRY_MIN (HZ/10) #define NFS4_POLL_RETRY_MAX (15*HZ) @@ -86,12 +88,11 @@ | ATTR_MTIME_SET) struct nfs4_opendata; -static int _nfs4_proc_open(struct nfs4_opendata *data); 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); -static int nfs4_proc_getattr(struct nfs_server *, struct nfs_fh *, struct nfs_fattr *, struct nfs4_label *label); -static int _nfs4_proc_getattr(struct nfs_server *server, struct nfs_fh *fhandle, struct nfs_fattr *fattr, struct nfs4_label *label); +static int nfs4_proc_getattr(struct nfs_server *, struct nfs_fh *, struct nfs_fattr *, struct nfs4_label *label, struct inode *inode); +static int _nfs4_proc_getattr(struct nfs_server *server, struct nfs_fh *fhandle, struct nfs_fattr *fattr, struct nfs4_label *label, struct inode *inode); static int nfs4_do_setattr(struct inode *inode, struct rpc_cred *cred, struct nfs_fattr *fattr, struct iattr *sattr, struct nfs_open_context *ctx, struct nfs4_label *ilabel, @@ -274,6 +275,33 @@ const u32 nfs4_fs_locations_bitmap[3] = { | FATTR4_WORD1_MOUNTED_ON_FILEID, }; +static void nfs4_bitmap_copy_adjust(__u32 *dst, const __u32 *src, + struct inode *inode) +{ + unsigned long cache_validity; + + memcpy(dst, src, NFS4_BITMASK_SZ*sizeof(*dst)); + if (!inode || !nfs4_have_delegation(inode, FMODE_READ)) + return; + + cache_validity = READ_ONCE(NFS_I(inode)->cache_validity); + if (!(cache_validity & NFS_INO_REVAL_FORCED)) + cache_validity &= ~(NFS_INO_INVALID_CHANGE + | NFS_INO_INVALID_SIZE); + + if (!(cache_validity & NFS_INO_INVALID_SIZE)) + dst[0] &= ~FATTR4_WORD0_SIZE; + + if (!(cache_validity & NFS_INO_INVALID_CHANGE)) + dst[0] &= ~FATTR4_WORD0_CHANGE; +} + +static void nfs4_bitmap_copy_adjust_setattr(__u32 *dst, + const __u32 *src, struct inode *inode) +{ + nfs4_bitmap_copy_adjust(dst, src, inode); +} + static void nfs4_setup_readdir(u64 cookie, __be32 *verifier, struct dentry *dentry, struct nfs4_readdir_arg *readdir) { @@ -407,6 +435,11 @@ static int nfs4_do_handle_exception(struct nfs_server *server, switch(errorcode) { case 0: return 0; + case -NFS4ERR_BADHANDLE: + case -ESTALE: + if (inode != NULL && S_ISREG(inode->i_mode)) + pnfs_destroy_layout(NFS_I(inode)); + break; case -NFS4ERR_DELEG_REVOKED: case -NFS4ERR_ADMIN_REVOKED: case -NFS4ERR_EXPIRED: @@ -608,20 +641,16 @@ struct nfs4_call_sync_data { }; void nfs4_init_sequence(struct nfs4_sequence_args *args, - struct nfs4_sequence_res *res, int cache_reply) + struct nfs4_sequence_res *res, int cache_reply, + int privileged) { args->sa_slot = NULL; args->sa_cache_this = cache_reply; - args->sa_privileged = 0; + args->sa_privileged = privileged; res->sr_slot = NULL; } -static void nfs4_set_sequence_privileged(struct nfs4_sequence_args *args) -{ - args->sa_privileged = 1; -} - static void nfs40_sequence_free_slot(struct nfs4_sequence_res *res) { struct nfs4_slot *slot = res->sr_slot; @@ -746,12 +775,19 @@ static int nfs41_sequence_process(struct rpc_task *task, slot->slot_nr, slot->seq_nr); goto out_retry; + case -NFS4ERR_RETRY_UNCACHED_REP: + case -NFS4ERR_SEQ_FALSE_RETRY: + /* + * The server thinks we tried to replay a request. + * Retry the call after bumping the sequence ID. + */ + goto retry_new_seq; case -NFS4ERR_BADSLOT: /* * The slot id we used was probably retired. Try again * using a different slot id. */ - if (slot->seq_nr < slot->table->target_highest_slotid) + if (slot->slot_nr < slot->table->target_highest_slotid) goto session_recover; goto retry_nowait; case -NFS4ERR_SEQ_MISORDERED: @@ -770,10 +806,6 @@ static int nfs41_sequence_process(struct rpc_task *task, goto retry_nowait; } goto session_recover; - case -NFS4ERR_SEQ_FALSE_RETRY: - if (interrupted) - goto retry_new_seq; - goto session_recover; default: /* Just update the slot sequence no. */ slot->seq_done = 1; @@ -1035,7 +1067,7 @@ int nfs4_call_sync(struct rpc_clnt *clnt, struct nfs4_sequence_res *res, int cache_reply) { - nfs4_init_sequence(args, res, cache_reply); + nfs4_init_sequence(args, res, cache_reply, 0); return nfs4_call_sync_sequence(clnt, server, msg, args, res); } @@ -1064,30 +1096,6 @@ static void update_changeattr(struct inode *dir, struct nfs4_change_info *cinfo, spin_unlock(&dir->i_lock); } -struct nfs4_opendata { - struct kref kref; - struct nfs_openargs o_arg; - struct nfs_openres o_res; - struct nfs_open_confirmargs c_arg; - struct nfs_open_confirmres c_res; - struct nfs4_string owner_name; - struct nfs4_string group_name; - struct nfs4_label *a_label; - struct nfs_fattr f_attr; - struct nfs4_label *f_label; - struct dentry *dir; - struct dentry *dentry; - struct nfs4_state_owner *owner; - struct nfs4_state *state; - struct iattr attrs; - unsigned long timestamp; - bool rpc_done; - bool file_created; - bool is_recover; - bool cancelled; - int rpc_status; -}; - struct nfs4_open_createattrs { struct nfs4_label *label; struct iattr *sattr; @@ -1268,6 +1276,7 @@ static void nfs4_opendata_free(struct kref *kref) struct nfs4_opendata, kref); struct super_block *sb = p->dentry->d_sb; + nfs4_lgopen_release(p->lgp); nfs_free_seqid(p->o_arg.seqid); nfs4_sequence_free_slot(&p->o_res.seq_res); if (p->state != NULL) @@ -2187,13 +2196,12 @@ static int _nfs4_proc_open_confirm(struct nfs4_opendata *data) }; int status; - nfs4_init_sequence(&data->c_arg.seq_args, &data->c_res.seq_res, 1); + nfs4_init_sequence(&data->c_arg.seq_args, &data->c_res.seq_res, 1, + data->is_recover); kref_get(&data->kref); data->rpc_done = false; data->rpc_status = 0; data->timestamp = jiffies; - if (data->is_recover) - nfs4_set_sequence_privileged(&data->c_arg.seq_args); task = rpc_run_task(&task_setup_data); if (IS_ERR(task)) return PTR_ERR(task); @@ -2327,7 +2335,8 @@ static const struct rpc_call_ops nfs4_open_ops = { .rpc_release = nfs4_open_release, }; -static int nfs4_run_open_task(struct nfs4_opendata *data, int isrecover) +static int nfs4_run_open_task(struct nfs4_opendata *data, + struct nfs_open_context *ctx) { struct inode *dir = d_inode(data->dir); struct nfs_server *server = NFS_SERVER(dir); @@ -2350,15 +2359,17 @@ static int nfs4_run_open_task(struct nfs4_opendata *data, int isrecover) }; int status; - nfs4_init_sequence(&o_arg->seq_args, &o_res->seq_res, 1); kref_get(&data->kref); data->rpc_done = false; data->rpc_status = 0; data->cancelled = false; data->is_recover = false; - if (isrecover) { - nfs4_set_sequence_privileged(&o_arg->seq_args); + if (!ctx) { + nfs4_init_sequence(&o_arg->seq_args, &o_res->seq_res, 1, 1); data->is_recover = true; + } 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)) @@ -2380,7 +2391,7 @@ static int _nfs4_recover_proc_open(struct nfs4_opendata *data) struct nfs_openres *o_res = &data->o_res; int status; - status = nfs4_run_open_task(data, 1); + status = nfs4_run_open_task(data, NULL); if (status != 0 || !data->rpc_done) return status; @@ -2441,7 +2452,8 @@ static int nfs4_opendata_access(struct rpc_cred *cred, /* * Note: On error, nfs4_proc_open will free the struct nfs4_opendata */ -static int _nfs4_proc_open(struct nfs4_opendata *data) +static int _nfs4_proc_open(struct nfs4_opendata *data, + struct nfs_open_context *ctx) { struct inode *dir = d_inode(data->dir); struct nfs_server *server = NFS_SERVER(dir); @@ -2449,7 +2461,7 @@ static int _nfs4_proc_open(struct nfs4_opendata *data) struct nfs_openres *o_res = &data->o_res; int status; - status = nfs4_run_open_task(data, 0); + status = nfs4_run_open_task(data, ctx); if (!data->rpc_done) return status; if (status != 0) { @@ -2480,7 +2492,8 @@ static int _nfs4_proc_open(struct nfs4_opendata *data) } if (!(o_res->f_attr->valid & NFS_ATTR_FATTR)) { nfs4_sequence_free_slot(&o_res->seq_res); - nfs4_proc_getattr(server, &o_res->fh, o_res->f_attr, o_res->f_label); + nfs4_proc_getattr(server, &o_res->fh, o_res->f_attr, + o_res->f_label, NULL); } return 0; } @@ -2800,11 +2813,11 @@ static int _nfs4_open_and_get_state(struct nfs4_opendata *opendata, seq = raw_seqcount_begin(&sp->so_reclaim_seqcount); - ret = _nfs4_proc_open(opendata); + ret = _nfs4_proc_open(opendata, ctx); if (ret != 0) goto out; - state = nfs4_opendata_to_nfs4_state(opendata); + state = _nfs4_opendata_to_nfs4_state(opendata); ret = PTR_ERR(state); if (IS_ERR(state)) goto out; @@ -2838,8 +2851,12 @@ 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: + nfs4_sequence_free_slot(&opendata->o_res.seq_res); return ret; } @@ -3039,7 +3056,6 @@ static int _nfs4_do_setattr(struct inode *inode, }; struct rpc_cred *delegation_cred = NULL; unsigned long timestamp = jiffies; - fmode_t fmode; bool truncate; int status; @@ -3047,11 +3063,12 @@ static int _nfs4_do_setattr(struct inode *inode, /* Servers should only apply open mode checks for file size changes */ truncate = (arg->iap->ia_valid & ATTR_SIZE) ? true : false; - fmode = truncate ? FMODE_WRITE : FMODE_READ; + if (!truncate) + goto zero_stateid; - if (nfs4_copy_delegation_stateid(inode, fmode, &arg->stateid, &delegation_cred)) { + if (nfs4_copy_delegation_stateid(inode, FMODE_WRITE, &arg->stateid, &delegation_cred)) { /* Use that stateid */ - } else if (truncate && ctx != NULL) { + } else if (ctx != NULL) { struct nfs_lock_context *l_ctx; if (!nfs4_valid_open_stateid(ctx->state)) return -EBADF; @@ -3063,8 +3080,10 @@ static int _nfs4_do_setattr(struct inode *inode, nfs_put_lock_context(l_ctx); if (status == -EIO) return -EBADF; - } else + } else { +zero_stateid: nfs4_stateid_copy(&arg->stateid, &zero_stateid); + } if (delegation_cred) msg.rpc_cred = delegation_cred; @@ -3083,12 +3102,13 @@ static int nfs4_do_setattr(struct inode *inode, struct rpc_cred *cred, struct nfs4_label *olabel) { struct nfs_server *server = NFS_SERVER(inode); + __u32 bitmask[NFS4_BITMASK_SZ]; struct nfs4_state *state = ctx ? ctx->state : NULL; struct nfs_setattrargs arg = { .fh = NFS_FH(inode), .iap = sattr, .server = server, - .bitmask = server->attr_bitmask, + .bitmask = bitmask, .label = ilabel, }; struct nfs_setattrres res = { @@ -3103,11 +3123,11 @@ static int nfs4_do_setattr(struct inode *inode, struct rpc_cred *cred, }; int err; - arg.bitmask = nfs4_bitmask(server, ilabel); - if (ilabel) - arg.bitmask = nfs4_bitmask(server, olabel); - do { + nfs4_bitmap_copy_adjust_setattr(bitmask, + nfs4_bitmask(server, olabel), + inode); + err = _nfs4_do_setattr(inode, &arg, &res, cred, ctx); switch (err) { case -NFS4ERR_OPENMODE: @@ -3393,7 +3413,7 @@ int nfs4_do_close(struct nfs4_state *state, gfp_t gfp_mask, int wait) calldata = kzalloc(sizeof(*calldata), gfp_mask); if (calldata == NULL) goto out; - nfs4_init_sequence(&calldata->arg.seq_args, &calldata->res.seq_res, 1); + nfs4_init_sequence(&calldata->arg.seq_args, &calldata->res.seq_res, 1, 0); calldata->inode = state->inode; calldata->state = state; calldata->arg.fh = NFS_FH(state->inode); @@ -3742,7 +3762,7 @@ static int nfs4_proc_get_root(struct nfs_server *server, struct nfs_fh *mntfh, if (IS_ERR(label)) return PTR_ERR(label); - error = nfs4_proc_getattr(server, mntfh, fattr, label); + error = nfs4_proc_getattr(server, mntfh, fattr, label, NULL); if (error < 0) { dprintk("nfs4_get_root: getattr error = %d\n", -error); goto err_free_label; @@ -3807,11 +3827,13 @@ out: } static int _nfs4_proc_getattr(struct nfs_server *server, struct nfs_fh *fhandle, - struct nfs_fattr *fattr, struct nfs4_label *label) + struct nfs_fattr *fattr, struct nfs4_label *label, + struct inode *inode) { + __u32 bitmask[NFS4_BITMASK_SZ]; struct nfs4_getattr_arg args = { .fh = fhandle, - .bitmask = server->attr_bitmask, + .bitmask = bitmask, }; struct nfs4_getattr_res res = { .fattr = fattr, @@ -3824,19 +3846,20 @@ static int _nfs4_proc_getattr(struct nfs_server *server, struct nfs_fh *fhandle, .rpc_resp = &res, }; - args.bitmask = nfs4_bitmask(server, label); + nfs4_bitmap_copy_adjust(bitmask, nfs4_bitmask(server, label), inode); nfs_fattr_init(fattr); return nfs4_call_sync(server->client, server, &msg, &args.seq_args, &res.seq_res, 0); } static int nfs4_proc_getattr(struct nfs_server *server, struct nfs_fh *fhandle, - struct nfs_fattr *fattr, struct nfs4_label *label) + struct nfs_fattr *fattr, struct nfs4_label *label, + struct inode *inode) { struct nfs4_exception exception = { }; int err; do { - err = _nfs4_proc_getattr(server, fhandle, fattr, label); + err = _nfs4_proc_getattr(server, fhandle, fattr, label, inode); trace_nfs4_getattr(server, fhandle, fattr, err); err = nfs4_handle_exception(server, err, &exception); @@ -4089,7 +4112,7 @@ static int _nfs4_proc_access(struct inode *inode, struct nfs_access_entry *entry }; int status = 0; - if (!nfs_have_delegated_attributes(inode)) { + if (!nfs4_have_delegation(inode, FMODE_READ)) { res.fattr = nfs_alloc_fattr(); if (res.fattr == NULL) return -ENOMEM; @@ -4265,15 +4288,16 @@ static int nfs4_proc_rmdir(struct inode *dir, const struct qstr *name) return err; } -static void nfs4_proc_unlink_setup(struct rpc_message *msg, struct dentry *dentry) +static void nfs4_proc_unlink_setup(struct rpc_message *msg, + struct dentry *dentry, + struct inode *inode) { struct nfs_removeargs *args = msg->rpc_argp; struct nfs_removeres *res = msg->rpc_resp; - struct inode *inode = d_inode(dentry); res->server = NFS_SB(dentry->d_sb); msg->rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_REMOVE]; - nfs4_init_sequence(&args->seq_args, &res->seq_res, 1); + nfs4_init_sequence(&args->seq_args, &res->seq_res, 1, 0); nfs_fattr_init(res->dir_attr); @@ -4319,7 +4343,7 @@ static void nfs4_proc_rename_setup(struct rpc_message *msg, nfs4_inode_return_delegation(new_inode); msg->rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_RENAME]; res->server = NFS_SB(old_dentry->d_sb); - nfs4_init_sequence(&arg->seq_args, &res->seq_res, 1); + nfs4_init_sequence(&arg->seq_args, &res->seq_res, 1, 0); } static void nfs4_proc_rename_rpc_prepare(struct rpc_task *task, struct nfs_renamedata *data) @@ -4352,11 +4376,12 @@ static int nfs4_proc_rename_done(struct rpc_task *task, struct inode *old_dir, static int _nfs4_proc_link(struct inode *inode, struct inode *dir, const struct qstr *name) { struct nfs_server *server = NFS_SERVER(inode); + __u32 bitmask[NFS4_BITMASK_SZ]; struct nfs4_link_arg arg = { .fh = NFS_FH(inode), .dir_fh = NFS_FH(dir), .name = name, - .bitmask = server->attr_bitmask, + .bitmask = bitmask, }; struct nfs4_link_res res = { .server = server, @@ -4378,9 +4403,9 @@ static int _nfs4_proc_link(struct inode *inode, struct inode *dir, const struct status = PTR_ERR(res.label); goto out; } - arg.bitmask = nfs4_bitmask(server, res.label); nfs4_inode_make_writeable(inode); + nfs4_bitmap_copy_adjust_setattr(bitmask, nfs4_bitmask(server, res.label), inode); status = nfs4_call_sync(server->client, server, &msg, &arg.seq_args, &res.seq_res, 1); if (!status) { @@ -4895,7 +4920,7 @@ static void nfs4_proc_read_setup(struct nfs_pgio_header *hdr, if (!hdr->pgio_done_cb) hdr->pgio_done_cb = nfs4_read_done_cb; msg->rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_READ]; - nfs4_init_sequence(&hdr->args.seq_args, &hdr->res.seq_res, 0); + nfs4_init_sequence(&hdr->args.seq_args, &hdr->res.seq_res, 0, 0); } static int nfs4_proc_pgio_rpc_prepare(struct rpc_task *task, @@ -4979,7 +5004,8 @@ bool nfs4_write_need_cache_consistency_data(struct nfs_pgio_header *hdr) } static void nfs4_proc_write_setup(struct nfs_pgio_header *hdr, - struct rpc_message *msg) + struct rpc_message *msg, + struct rpc_clnt **clnt) { struct nfs_server *server = NFS_SERVER(hdr->inode); @@ -4995,7 +5021,8 @@ static void nfs4_proc_write_setup(struct nfs_pgio_header *hdr, hdr->timestamp = jiffies; msg->rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_WRITE]; - nfs4_init_sequence(&hdr->args.seq_args, &hdr->res.seq_res, 1); + nfs4_init_sequence(&hdr->args.seq_args, &hdr->res.seq_res, 1, 0); + nfs4_state_protect_write(server->nfs_client, clnt, msg, hdr); } static void nfs4_proc_commit_rpc_prepare(struct rpc_task *task, struct nfs_commit_data *data) @@ -5026,7 +5053,8 @@ static int nfs4_commit_done(struct rpc_task *task, struct nfs_commit_data *data) return data->commit_done_cb(task, data); } -static void nfs4_proc_commit_setup(struct nfs_commit_data *data, struct rpc_message *msg) +static void nfs4_proc_commit_setup(struct nfs_commit_data *data, struct rpc_message *msg, + struct rpc_clnt **clnt) { struct nfs_server *server = NFS_SERVER(data->inode); @@ -5034,7 +5062,8 @@ static void nfs4_proc_commit_setup(struct nfs_commit_data *data, struct rpc_mess data->commit_done_cb = nfs4_commit_done_cb; data->res.server = server; msg->rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_COMMIT]; - nfs4_init_sequence(&data->args.seq_args, &data->res.seq_res, 1); + nfs4_init_sequence(&data->args.seq_args, &data->res.seq_res, 1, 0); + nfs4_state_protect(server->nfs_client, NFS_SP4_MACH_CRED_COMMIT, clnt, msg); } struct nfs4_renewdata { @@ -5391,7 +5420,8 @@ static int __nfs4_proc_set_acl(struct inode *inode, const void *buf, size_t bufl */ spin_lock(&inode->i_lock); NFS_I(inode)->cache_validity |= NFS_INO_INVALID_CHANGE - | NFS_INO_INVALID_CTIME; + | NFS_INO_INVALID_CTIME + | NFS_INO_REVAL_FORCED; spin_unlock(&inode->i_lock); nfs_access_zap_cache(inode); nfs_zap_acl_cache(inode); @@ -5591,13 +5621,14 @@ nfs4_init_nonuniform_client_string(struct nfs_client *clp) return 0; rcu_read_lock(); - len = 14 + strlen(clp->cl_ipaddr) + 1 + - strlen(rpc_peeraddr2str(clp->cl_rpcclient, RPC_DISPLAY_ADDR)) + + len = 14 + + strlen(clp->cl_rpcclient->cl_nodename) + 1 + - strlen(rpc_peeraddr2str(clp->cl_rpcclient, RPC_DISPLAY_PROTO)) + + strlen(rpc_peeraddr2str(clp->cl_rpcclient, RPC_DISPLAY_ADDR)) + 1; rcu_read_unlock(); - + if (nfs4_client_id_uniquifier[0] != '\0') + len += strlen(nfs4_client_id_uniquifier) + 1; if (len > NFS4_OPAQUE_LIMIT + 1) return -EINVAL; @@ -5611,10 +5642,17 @@ nfs4_init_nonuniform_client_string(struct nfs_client *clp) return -ENOMEM; rcu_read_lock(); - scnprintf(str, len, "Linux NFSv4.0 %s/%s %s", - clp->cl_ipaddr, - rpc_peeraddr2str(clp->cl_rpcclient, RPC_DISPLAY_ADDR), - rpc_peeraddr2str(clp->cl_rpcclient, RPC_DISPLAY_PROTO)); + if (nfs4_client_id_uniquifier[0] != '\0') + scnprintf(str, len, "Linux NFSv4.0 %s/%s/%s", + clp->cl_rpcclient->cl_nodename, + nfs4_client_id_uniquifier, + rpc_peeraddr2str(clp->cl_rpcclient, + RPC_DISPLAY_ADDR)); + else + scnprintf(str, len, "Linux NFSv4.0 %s/%s", + clp->cl_rpcclient->cl_nodename, + rpc_peeraddr2str(clp->cl_rpcclient, + RPC_DISPLAY_ADDR)); rcu_read_unlock(); clp->cl_owner_id = str; @@ -5972,7 +6010,7 @@ static int _nfs4_proc_delegreturn(struct inode *inode, struct rpc_cred *cred, co data = kzalloc(sizeof(*data), GFP_NOFS); if (data == NULL) return -ENOMEM; - nfs4_init_sequence(&data->args.seq_args, &data->res.seq_res, 1); + nfs4_init_sequence(&data->args.seq_args, &data->res.seq_res, 1, 0); nfs4_state_protect(server->nfs_client, NFS_SP4_MACH_CRED_CLEANUP, @@ -6247,7 +6285,7 @@ static struct rpc_task *nfs4_do_unlck(struct file_lock *fl, return ERR_PTR(-ENOMEM); } - nfs4_init_sequence(&data->arg.seq_args, &data->res.seq_res, 1); + nfs4_init_sequence(&data->arg.seq_args, &data->res.seq_res, 1, 0); msg.rpc_argp = &data->arg; msg.rpc_resp = &data->res; task_setup_data.callback_data = data; @@ -6411,32 +6449,36 @@ static void nfs4_lock_done(struct rpc_task *task, void *calldata) case 0: renew_lease(NFS_SERVER(d_inode(data->ctx->dentry)), data->timestamp); - if (data->arg.new_lock) { + if (data->arg.new_lock && !data->cancelled) { data->fl.fl_flags &= ~(FL_SLEEP | FL_ACCESS); - if (locks_lock_inode_wait(lsp->ls_state->inode, &data->fl) < 0) { - rpc_restart_call_prepare(task); + if (locks_lock_inode_wait(lsp->ls_state->inode, &data->fl) < 0) break; - } } + if (data->arg.new_lock_owner != 0) { nfs_confirm_seqid(&lsp->ls_seqid, 0); nfs4_stateid_copy(&lsp->ls_stateid, &data->res.stateid); set_bit(NFS_LOCK_INITIALIZED, &lsp->ls_flags); - } else if (!nfs4_update_lock_stateid(lsp, &data->res.stateid)) - rpc_restart_call_prepare(task); + goto out_done; + } else if (nfs4_update_lock_stateid(lsp, &data->res.stateid)) + goto out_done; + break; case -NFS4ERR_BAD_STATEID: case -NFS4ERR_OLD_STATEID: case -NFS4ERR_STALE_STATEID: case -NFS4ERR_EXPIRED: if (data->arg.new_lock_owner != 0) { - if (!nfs4_stateid_match(&data->arg.open_stateid, + if (nfs4_stateid_match(&data->arg.open_stateid, &lsp->ls_state->open_stateid)) - rpc_restart_call_prepare(task); - } else if (!nfs4_stateid_match(&data->arg.lock_stateid, + goto out_done; + } else if (nfs4_stateid_match(&data->arg.lock_stateid, &lsp->ls_stateid)) - rpc_restart_call_prepare(task); + goto out_done; } + if (!data->cancelled) + rpc_restart_call_prepare(task); +out_done: dprintk("%s: done, ret = %d!\n", __func__, data->rpc_status); } @@ -6509,14 +6551,14 @@ static int _nfs4_do_setlk(struct nfs4_state *state, int cmd, struct file_lock *f return -ENOMEM; if (IS_SETLKW(cmd)) data->arg.block = 1; - nfs4_init_sequence(&data->arg.seq_args, &data->res.seq_res, 1); + nfs4_init_sequence(&data->arg.seq_args, &data->res.seq_res, 1, + recovery_type > NFS_LOCK_NEW); msg.rpc_argp = &data->arg; msg.rpc_resp = &data->res; task_setup_data.callback_data = data; if (recovery_type > NFS_LOCK_NEW) { if (recovery_type == NFS_LOCK_RECLAIM) data->arg.reclaim = NFS_LOCK_RECLAIM; - nfs4_set_sequence_privileged(&data->arg.seq_args); } else data->arg.new_lock = 1; task = rpc_run_task(&task_setup_data); @@ -6911,7 +6953,7 @@ nfs4_release_lockowner(struct nfs_server *server, struct nfs4_lock_state *lsp) msg.rpc_argp = &data->args; msg.rpc_resp = &data->res; - nfs4_init_sequence(&data->args.seq_args, &data->res.seq_res, 0); + nfs4_init_sequence(&data->args.seq_args, &data->res.seq_res, 0, 0); rpc_call_async(server->client, &msg, 0, &nfs4_release_lockowner_ops, data); } @@ -7107,8 +7149,7 @@ static int _nfs40_proc_get_locations(struct inode *inode, locations->server = server; locations->nlocations = 0; - nfs4_init_sequence(&args.seq_args, &res.seq_res, 0); - nfs4_set_sequence_privileged(&args.seq_args); + nfs4_init_sequence(&args.seq_args, &res.seq_res, 0, 1); status = nfs4_call_sync_sequence(clnt, server, &msg, &args.seq_args, &res.seq_res); if (status) @@ -7161,8 +7202,7 @@ static int _nfs41_proc_get_locations(struct inode *inode, locations->server = server; locations->nlocations = 0; - nfs4_init_sequence(&args.seq_args, &res.seq_res, 0); - nfs4_set_sequence_privileged(&args.seq_args); + nfs4_init_sequence(&args.seq_args, &res.seq_res, 0, 1); status = nfs4_call_sync_sequence(clnt, server, &msg, &args.seq_args, &res.seq_res); if (status == NFS4_OK && @@ -7249,8 +7289,7 @@ static int _nfs40_proc_fsid_present(struct inode *inode, struct rpc_cred *cred) if (res.fh == NULL) return -ENOMEM; - nfs4_init_sequence(&args.seq_args, &res.seq_res, 0); - nfs4_set_sequence_privileged(&args.seq_args); + nfs4_init_sequence(&args.seq_args, &res.seq_res, 0, 1); status = nfs4_call_sync_sequence(clnt, server, &msg, &args.seq_args, &res.seq_res); nfs_free_fhandle(res.fh); @@ -7291,8 +7330,7 @@ static int _nfs41_proc_fsid_present(struct inode *inode, struct rpc_cred *cred) if (res.fh == NULL) return -ENOMEM; - nfs4_init_sequence(&args.seq_args, &res.seq_res, 0); - nfs4_set_sequence_privileged(&args.seq_args); + nfs4_init_sequence(&args.seq_args, &res.seq_res, 0, 1); status = nfs4_call_sync_sequence(clnt, server, &msg, &args.seq_args, &res.seq_res); nfs_free_fhandle(res.fh); @@ -8070,8 +8108,7 @@ int nfs4_proc_get_lease_time(struct nfs_client *clp, struct nfs_fsinfo *fsinfo) }; int status; - nfs4_init_sequence(&args.la_seq_args, &res.lr_seq_res, 0); - nfs4_set_sequence_privileged(&args.la_seq_args); + nfs4_init_sequence(&args.la_seq_args, &res.lr_seq_res, 0, 1); task = rpc_run_task(&task_setup); if (IS_ERR(task)) @@ -8408,10 +8445,8 @@ static struct rpc_task *_nfs41_proc_sequence(struct nfs_client *clp, calldata = kzalloc(sizeof(*calldata), GFP_NOFS); if (calldata == NULL) goto out_put_clp; - nfs4_init_sequence(&calldata->args, &calldata->res, 0); + nfs4_init_sequence(&calldata->args, &calldata->res, 0, is_privileged); nfs4_sequence_attach_slot(&calldata->args, &calldata->res, slot); - if (is_privileged) - nfs4_set_sequence_privileged(&calldata->args); msg.rpc_argp = &calldata->args; msg.rpc_resp = &calldata->res; calldata->clp = clp; @@ -8563,8 +8598,7 @@ static int nfs41_proc_reclaim_complete(struct nfs_client *clp, calldata->clp = clp; calldata->arg.one_fs = 0; - nfs4_init_sequence(&calldata->arg.seq_args, &calldata->res.seq_res, 0); - nfs4_set_sequence_privileged(&calldata->arg.seq_args); + nfs4_init_sequence(&calldata->arg.seq_args, &calldata->res.seq_res, 0, 1); msg.rpc_argp = &calldata->arg; msg.rpc_resp = &calldata->res; task_setup_data.callback_data = calldata; @@ -8693,63 +8727,19 @@ out: return status; } -static size_t max_response_pages(struct nfs_server *server) +size_t max_response_pages(struct nfs_server *server) { u32 max_resp_sz = server->nfs_client->cl_session->fc_attrs.max_resp_sz; return nfs_page_array_len(0, max_resp_sz); } -static void nfs4_free_pages(struct page **pages, size_t size) -{ - int i; - - if (!pages) - return; - - for (i = 0; i < size; i++) { - if (!pages[i]) - break; - __free_page(pages[i]); - } - kfree(pages); -} - -static struct page **nfs4_alloc_pages(size_t size, gfp_t gfp_flags) -{ - struct page **pages; - int i; - - pages = kcalloc(size, sizeof(struct page *), gfp_flags); - if (!pages) { - dprintk("%s: can't alloc array of %zu pages\n", __func__, size); - return NULL; - } - - for (i = 0; i < size; i++) { - pages[i] = alloc_page(gfp_flags); - if (!pages[i]) { - dprintk("%s: failed to allocate page\n", __func__); - nfs4_free_pages(pages, size); - return NULL; - } - } - - return pages; -} - 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); 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); - put_nfs_open_context(lgp->args.ctx); - kfree(calldata); + pnfs_layoutget_free(lgp); dprintk("<-- %s\n", __func__); } @@ -8760,11 +8750,10 @@ static const struct rpc_call_ops nfs4_layoutget_call_ops = { }; struct pnfs_layout_segment * -nfs4_proc_layoutget(struct nfs4_layoutget *lgp, long *timeout, gfp_t gfp_flags) +nfs4_proc_layoutget(struct nfs4_layoutget *lgp, long *timeout) { struct inode *inode = lgp->args.inode; struct nfs_server *server = NFS_SERVER(inode); - size_t max_pages = max_response_pages(server); struct rpc_task *task; struct rpc_message msg = { .rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_LAYOUTGET], @@ -8791,16 +8780,7 @@ nfs4_proc_layoutget(struct nfs4_layoutget *lgp, long *timeout, gfp_t gfp_flags) /* nfs4_layoutget_release calls pnfs_put_layout_hdr */ pnfs_get_layout_hdr(NFS_I(inode)->layout); - lgp->args.layout.pages = nfs4_alloc_pages(max_pages, gfp_flags); - if (!lgp->args.layout.pages) { - nfs4_layoutget_release(lgp); - return ERR_PTR(-ENOMEM); - } - lgp->args.layout.pglen = max_pages * PAGE_SIZE; - - lgp->res.layoutp = &lgp->args.layout; - lgp->res.seq_res.sr_slot = NULL; - nfs4_init_sequence(&lgp->args.seq_args, &lgp->res.seq_res, 0); + nfs4_init_sequence(&lgp->args.seq_args, &lgp->res.seq_res, 0, 0); task = rpc_run_task(&task_setup_data); if (IS_ERR(task)) @@ -8927,7 +8907,7 @@ int nfs4_proc_layoutreturn(struct nfs4_layoutreturn *lrp, bool sync) } task_setup_data.flags |= RPC_TASK_ASYNC; } - nfs4_init_sequence(&lrp->args.seq_args, &lrp->res.seq_res, 1); + nfs4_init_sequence(&lrp->args.seq_args, &lrp->res.seq_res, 1, 0); task = rpc_run_task(&task_setup_data); if (IS_ERR(task)) return PTR_ERR(task); @@ -9074,7 +9054,7 @@ nfs4_proc_layoutcommit(struct nfs4_layoutcommit_data *data, bool sync) } task_setup_data.flags = RPC_TASK_ASYNC; } - nfs4_init_sequence(&data->args.seq_args, &data->res.seq_res, 1); + nfs4_init_sequence(&data->args.seq_args, &data->res.seq_res, 1, 0); task = rpc_run_task(&task_setup_data); if (IS_ERR(task)) return PTR_ERR(task); @@ -9254,8 +9234,7 @@ static int _nfs41_test_stateid(struct nfs_server *server, &rpc_client, &msg); dprintk("NFS call test_stateid %p\n", stateid); - nfs4_init_sequence(&args.seq_args, &res.seq_res, 0); - nfs4_set_sequence_privileged(&args.seq_args); + nfs4_init_sequence(&args.seq_args, &res.seq_res, 0, 1); status = nfs4_call_sync_sequence(rpc_client, server, &msg, &args.seq_args, &res.seq_res); if (status != NFS_OK) { @@ -9347,7 +9326,17 @@ static const struct rpc_call_ops nfs41_free_stateid_ops = { .rpc_release = nfs41_free_stateid_release, }; -static struct rpc_task *_nfs41_free_stateid(struct nfs_server *server, +/** + * nfs41_free_stateid - perform a FREE_STATEID operation + * + * @server: server / transport on which to perform the operation + * @stateid: state ID to release + * @cred: credential + * @is_recovery: set to true if this call needs to be privileged + * + * Note: this function is always asynchronous. + */ +static int nfs41_free_stateid(struct nfs_server *server, const nfs4_stateid *stateid, struct rpc_cred *cred, bool privileged) @@ -9363,6 +9352,7 @@ static struct rpc_task *_nfs41_free_stateid(struct nfs_server *server, .flags = RPC_TASK_ASYNC, }; struct nfs_free_stateid_data *data; + struct rpc_task *task; nfs4_state_protect(server->nfs_client, NFS_SP4_MACH_CRED_STATEID, &task_setup.rpc_client, &msg); @@ -9370,7 +9360,7 @@ static struct rpc_task *_nfs41_free_stateid(struct nfs_server *server, dprintk("NFS call free_stateid %p\n", stateid); data = kmalloc(sizeof(*data), GFP_NOFS); if (!data) - return ERR_PTR(-ENOMEM); + return -ENOMEM; data->server = server; nfs4_stateid_copy(&data->args.stateid, stateid); @@ -9378,31 +9368,8 @@ static struct rpc_task *_nfs41_free_stateid(struct nfs_server *server, msg.rpc_argp = &data->args; msg.rpc_resp = &data->res; - nfs4_init_sequence(&data->args.seq_args, &data->res.seq_res, 1); - if (privileged) - nfs4_set_sequence_privileged(&data->args.seq_args); - - return rpc_run_task(&task_setup); -} - -/** - * nfs41_free_stateid - perform a FREE_STATEID operation - * - * @server: server / transport on which to perform the operation - * @stateid: state ID to release - * @cred: credential - * @is_recovery: set to true if this call needs to be privileged - * - * Note: this function is always asynchronous. - */ -static int nfs41_free_stateid(struct nfs_server *server, - const nfs4_stateid *stateid, - struct rpc_cred *cred, - bool is_recovery) -{ - struct rpc_task *task; - - task = _nfs41_free_stateid(server, stateid, cred, is_recovery); + nfs4_init_sequence(&data->args.seq_args, &data->res.seq_res, 1, privileged); + task = rpc_run_task(&task_setup); if (IS_ERR(task)) return PTR_ERR(task); rpc_put_task(task); @@ -9539,7 +9506,8 @@ static const struct nfs4_minor_version_ops nfs_v4_1_minor_ops = { | NFS_CAP_ATOMIC_OPEN | NFS_CAP_POSIX_LOCK | NFS_CAP_STATEID_NFSV41 - | NFS_CAP_ATOMIC_OPEN_V1, + | NFS_CAP_ATOMIC_OPEN_V1 + | NFS_CAP_LGOPEN, .init_client = nfs41_init_client, .shutdown_client = nfs41_shutdown_client, .match_stateid = nfs41_match_stateid, @@ -9564,6 +9532,7 @@ static const struct nfs4_minor_version_ops nfs_v4_2_minor_ops = { | NFS_CAP_POSIX_LOCK | NFS_CAP_STATEID_NFSV41 | NFS_CAP_ATOMIC_OPEN_V1 + | NFS_CAP_LGOPEN | NFS_CAP_ALLOCATE | NFS_CAP_COPY | NFS_CAP_DEALLOCATE 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/nfs4xdr.c b/fs/nfs/nfs4xdr.c index 481426e9e6f0..cd41d2577a04 100644 --- a/fs/nfs/nfs4xdr.c +++ b/fs/nfs/nfs4xdr.c @@ -65,7 +65,13 @@ /* Mapping from NFS error code to "errno" error code. */ #define errno_NFSERR_IO EIO +struct compound_hdr; static int nfs4_stat_to_errno(int); +static void encode_layoutget(struct xdr_stream *xdr, + const struct nfs4_layoutget_args *args, + struct compound_hdr *hdr); +static int decode_layoutget(struct xdr_stream *xdr, struct rpc_rqst *req, + struct nfs4_layoutget_res *res); /* NFSv4 COMPOUND tags are only wanted for debugging purposes */ #ifdef DEBUG @@ -424,6 +430,8 @@ static int nfs4_stat_to_errno(int); #define decode_sequence_maxsz 0 #define encode_layoutreturn_maxsz 0 #define decode_layoutreturn_maxsz 0 +#define encode_layoutget_maxsz 0 +#define decode_layoutget_maxsz 0 #endif /* CONFIG_NFS_V4_1 */ #define NFS4_enc_compound_sz (1024) /* XXX: large enough? */ @@ -476,14 +484,16 @@ static int nfs4_stat_to_errno(int); encode_open_maxsz + \ encode_access_maxsz + \ encode_getfh_maxsz + \ - encode_getattr_maxsz) + encode_getattr_maxsz + \ + encode_layoutget_maxsz) #define NFS4_dec_open_sz (compound_decode_hdr_maxsz + \ decode_sequence_maxsz + \ decode_putfh_maxsz + \ decode_open_maxsz + \ decode_access_maxsz + \ decode_getfh_maxsz + \ - decode_getattr_maxsz) + decode_getattr_maxsz + \ + decode_layoutget_maxsz) #define NFS4_enc_open_confirm_sz \ (compound_encode_hdr_maxsz + \ encode_putfh_maxsz + \ @@ -497,13 +507,15 @@ static int nfs4_stat_to_errno(int); encode_putfh_maxsz + \ encode_open_maxsz + \ encode_access_maxsz + \ - encode_getattr_maxsz) + encode_getattr_maxsz + \ + encode_layoutget_maxsz) #define NFS4_dec_open_noattr_sz (compound_decode_hdr_maxsz + \ decode_sequence_maxsz + \ decode_putfh_maxsz + \ decode_open_maxsz + \ decode_access_maxsz + \ - decode_getattr_maxsz) + decode_getattr_maxsz + \ + decode_layoutget_maxsz) #define NFS4_enc_open_downgrade_sz \ (compound_encode_hdr_maxsz + \ encode_sequence_maxsz + \ @@ -2073,6 +2085,13 @@ encode_layoutreturn(struct xdr_stream *xdr, struct compound_hdr *hdr) { } + +static void +encode_layoutget(struct xdr_stream *xdr, + const struct nfs4_layoutget_args *args, + struct compound_hdr *hdr) +{ +} #endif /* CONFIG_NFS_V4_1 */ /* @@ -2319,6 +2338,12 @@ static void nfs4_xdr_enc_open(struct rpc_rqst *req, struct xdr_stream *xdr, if (args->access) encode_access(xdr, args->access, &hdr); encode_getfattr_open(xdr, args->bitmask, args->open_bitmap, &hdr); + if (args->lg_args) { + encode_layoutget(xdr, args->lg_args, &hdr); + xdr_inline_pages(&req->rq_rcv_buf, hdr.replen << 2, + args->lg_args->layout.pages, + 0, args->lg_args->layout.pglen); + } encode_nops(&hdr); } @@ -2359,6 +2384,12 @@ static void nfs4_xdr_enc_open_noattr(struct rpc_rqst *req, if (args->access) encode_access(xdr, args->access, &hdr); encode_getfattr_open(xdr, args->bitmask, args->open_bitmap, &hdr); + if (args->lg_args) { + encode_layoutget(xdr, args->lg_args, &hdr); + xdr_inline_pages(&req->rq_rcv_buf, hdr.replen << 2, + args->lg_args->layout.pages, + 0, args->lg_args->layout.pglen); + } encode_nops(&hdr); } @@ -6027,7 +6058,7 @@ static int decode_layoutget(struct xdr_stream *xdr, struct rpc_rqst *req, status = decode_op_hdr(xdr, OP_LAYOUTGET); if (status) - return status; + goto out; p = xdr_inline_decode(xdr, 4); if (unlikely(!p)) goto out_overflow; @@ -6040,7 +6071,8 @@ static int decode_layoutget(struct xdr_stream *xdr, struct rpc_rqst *req, if (!layout_count) { dprintk("%s: server responded with empty layout array\n", __func__); - return -EINVAL; + status = -EINVAL; + goto out; } p = xdr_inline_decode(xdr, 28); @@ -6065,7 +6097,8 @@ static int decode_layoutget(struct xdr_stream *xdr, struct rpc_rqst *req, dprintk("NFS: server cheating in layoutget reply: " "layout len %u > recvd %u\n", res->layoutp->len, recvd); - return -EINVAL; + status = -EINVAL; + goto out; } if (layout_count > 1) { @@ -6078,10 +6111,13 @@ static int decode_layoutget(struct xdr_stream *xdr, struct rpc_rqst *req, __func__, layout_count); } - return 0; +out: + res->status = status; + return status; out_overflow: print_overflow_msg(__func__, xdr); - return -EIO; + status = -EIO; + goto out; } static int decode_layoutreturn(struct xdr_stream *xdr, @@ -6180,6 +6216,13 @@ int decode_layoutreturn(struct xdr_stream *xdr, { return 0; } + +static int decode_layoutget(struct xdr_stream *xdr, struct rpc_rqst *req, + struct nfs4_layoutget_res *res) +{ + return 0; +} + #endif /* CONFIG_NFS_V4_1 */ /* @@ -6626,6 +6669,8 @@ static int nfs4_xdr_dec_open(struct rpc_rqst *rqstp, struct xdr_stream *xdr, if (res->access_request) decode_access(xdr, &res->access_supported, &res->access_result); decode_getfattr_label(xdr, res->f_attr, res->f_label, res->server); + if (res->lg_res) + decode_layoutget(xdr, rqstp, res->lg_res); out: return status; } @@ -6678,6 +6723,8 @@ static int nfs4_xdr_dec_open_noattr(struct rpc_rqst *rqstp, if (res->access_request) decode_access(xdr, &res->access_supported, &res->access_result); decode_getfattr(xdr, res->f_attr, res->server); + if (res->lg_res) + decode_layoutget(xdr, rqstp, res->lg_res); out: return status; } diff --git a/fs/nfs/pnfs.c b/fs/nfs/pnfs.c index ee723aa153a3..bcc3addec3c5 100644 --- a/fs/nfs/pnfs.c +++ b/fs/nfs/pnfs.c @@ -37,6 +37,7 @@ #include "nfs4trace.h" #include "delegation.h" #include "nfs42.h" +#include "nfs4_fs.h" #define NFSDBG_FACILITY NFSDBG_PNFS #define PNFS_LAYOUTGET_RETRY_TIMEOUT (120*HZ) @@ -915,45 +916,99 @@ pnfs_layoutgets_blocked(const struct pnfs_layout_hdr *lo) test_bit(NFS_LAYOUT_BULK_RECALL, &lo->plh_flags); } -/* - * Get layout from server. - * for now, assume that whole file layouts are requested. - * arg->offset: 0 - * arg->length: all ones - */ -static struct pnfs_layout_segment * -send_layoutget(struct pnfs_layout_hdr *lo, +static struct nfs_server * +pnfs_find_server(struct inode *inode, struct nfs_open_context *ctx) +{ + struct nfs_server *server; + + if (inode) { + server = NFS_SERVER(inode); + } else { + struct dentry *parent_dir = dget_parent(ctx->dentry); + server = NFS_SERVER(parent_dir->d_inode); + dput(parent_dir); + } + return server; +} + +static void nfs4_free_pages(struct page **pages, size_t size) +{ + int i; + + if (!pages) + return; + + for (i = 0; i < size; i++) { + if (!pages[i]) + break; + __free_page(pages[i]); + } + kfree(pages); +} + +static struct page **nfs4_alloc_pages(size_t size, gfp_t gfp_flags) +{ + struct page **pages; + int i; + + pages = kcalloc(size, sizeof(struct page *), gfp_flags); + if (!pages) { + dprintk("%s: can't alloc array of %zu pages\n", __func__, size); + return NULL; + } + + for (i = 0; i < size; i++) { + pages[i] = alloc_page(gfp_flags); + if (!pages[i]) { + dprintk("%s: failed to allocate page\n", __func__); + nfs4_free_pages(pages, size); + return NULL; + } + } + + return pages; +} + +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, - long *timeout, gfp_t gfp_flags) + gfp_t gfp_flags) { - struct inode *ino = lo->plh_inode; - struct nfs_server *server = NFS_SERVER(ino); + struct nfs_server *server = pnfs_find_server(ino, ctx); + size_t max_pages = max_response_pages(server); struct nfs4_layoutget *lgp; - loff_t i_size; dprintk("--> %s\n", __func__); - /* - * Synchronously retrieve layout information from server and - * store in lseg. If we race with a concurrent seqid morphing - * op, then re-send the LAYOUTGET. - */ lgp = kzalloc(sizeof(*lgp), gfp_flags); if (lgp == NULL) - return ERR_PTR(-ENOMEM); + return NULL; + + lgp->args.layout.pages = nfs4_alloc_pages(max_pages, gfp_flags); + if (!lgp->args.layout.pages) { + kfree(lgp); + return NULL; + } + lgp->args.layout.pglen = max_pages * PAGE_SIZE; + lgp->res.layoutp = &lgp->args.layout; - i_size = i_size_read(ino); + /* Don't confuse uninitialised result and success */ + lgp->res.status = -NFS4ERR_DELAY; lgp->args.minlength = PAGE_SIZE; if (lgp->args.minlength > range->length) lgp->args.minlength = range->length; - if (range->iomode == IOMODE_READ) { - if (range->offset >= i_size) - lgp->args.minlength = 0; - else if (i_size - range->offset < lgp->args.minlength) - lgp->args.minlength = i_size - range->offset; + if (ino) { + loff_t i_size = i_size_read(ino); + + if (range->iomode == IOMODE_READ) { + if (range->offset >= i_size) + lgp->args.minlength = 0; + else if (i_size - range->offset < lgp->args.minlength) + lgp->args.minlength = i_size - range->offset; + } } lgp->args.maxcount = PNFS_LAYOUT_MAXSIZE; pnfs_copy_range(&lgp->args.range, range); @@ -962,9 +1017,21 @@ send_layoutget(struct pnfs_layout_hdr *lo, lgp->args.ctx = get_nfs_open_context(ctx); nfs4_stateid_copy(&lgp->args.stateid, stateid); lgp->gfp_flags = gfp_flags; - lgp->cred = lo->plh_lc_cred; + lgp->cred = get_rpccred(ctx->cred); + lgp->callback_count = raw_seqcount_begin(&server->nfs_client->cl_callback_count); + return lgp; +} - return nfs4_proc_layoutget(lgp, timeout, gfp_flags); +void pnfs_layoutget_free(struct nfs4_layoutget *lgp) +{ + size_t max_pages = lgp->args.layout.pglen / PAGE_SIZE; + + nfs4_free_pages(lgp->args.layout.pages, max_pages); + if (lgp->args.inode) + pnfs_put_layout_hdr(NFS_I(lgp->args.inode)->layout); + put_rpccred(lgp->cred); + put_nfs_open_context(lgp->args.ctx); + kfree(lgp); } static void pnfs_clear_layoutcommit(struct inode *inode, @@ -1144,7 +1211,7 @@ _pnfs_return_layout(struct inode *ino) LIST_HEAD(tmp_list); nfs4_stateid stateid; int status = 0; - bool send; + bool send, valid_layout; dprintk("NFS: %s for inode %lu\n", __func__, ino->i_ino); @@ -1165,6 +1232,7 @@ _pnfs_return_layout(struct inode *ino) goto out_put_layout_hdr; spin_lock(&ino->i_lock); } + valid_layout = pnfs_layout_is_valid(lo); pnfs_clear_layoutcommit(ino, &tmp_list); pnfs_mark_matching_lsegs_invalid(lo, &tmp_list, NULL, 0); @@ -1178,7 +1246,8 @@ _pnfs_return_layout(struct inode *ino) } /* Don't send a LAYOUTRETURN if list was initially empty */ - if (!test_bit(NFS_LAYOUT_RETURN_REQUESTED, &lo->plh_flags)) { + if (!test_bit(NFS_LAYOUT_RETURN_REQUESTED, &lo->plh_flags) || + !valid_layout) { spin_unlock(&ino->i_lock); dprintk("NFS: %s no layout segments to return\n", __func__); goto out_put_layout_hdr; @@ -1671,6 +1740,22 @@ static void pnfs_clear_first_layoutget(struct pnfs_layout_hdr *lo) wake_up_bit(bitlock, NFS_LAYOUT_FIRST_LAYOUTGET); } +static void _add_to_server_list(struct pnfs_layout_hdr *lo, + struct nfs_server *server) +{ + if (list_empty(&lo->plh_layouts)) { + struct nfs_client *clp = server->nfs_client; + + /* The lo must be on the clp list if there is any + * chance of a CB_LAYOUTRECALL(FILE) coming in. + */ + spin_lock(&clp->cl_lock); + if (list_empty(&lo->plh_layouts)) + list_add_tail(&lo->plh_layouts, &server->layouts); + spin_unlock(&clp->cl_lock); + } +} + /* * Layout segment is retreived from the server if not cached. * The appropriate layout segment is referenced and returned to the caller. @@ -1694,6 +1779,7 @@ pnfs_update_layout(struct inode *ino, struct nfs_client *clp = server->nfs_client; struct pnfs_layout_hdr *lo = NULL; struct pnfs_layout_segment *lseg = NULL; + struct nfs4_layoutget *lgp; nfs4_stateid stateid; long timeout = 0; unsigned long giveup = jiffies + (clp->cl_lease_time << 1); @@ -1820,15 +1906,7 @@ lookup_again: atomic_inc(&lo->plh_outstanding); spin_unlock(&ino->i_lock); - if (list_empty(&lo->plh_layouts)) { - /* The lo must be on the clp list if there is any - * chance of a CB_LAYOUTRECALL(FILE) coming in. - */ - spin_lock(&clp->cl_lock); - if (list_empty(&lo->plh_layouts)) - list_add_tail(&lo->plh_layouts, &server->layouts); - spin_unlock(&clp->cl_lock); - } + _add_to_server_list(lo, server); pg_offset = arg.offset & ~PAGE_MASK; if (pg_offset) { @@ -1838,7 +1916,15 @@ lookup_again: if (arg.length != NFS4_MAX_UINT64) arg.length = PAGE_ALIGN(arg.length); - lseg = send_layoutget(lo, ctx, &stateid, &arg, &timeout, gfp_flags); + lgp = pnfs_alloc_init_layoutget_args(ino, ctx, &stateid, &arg, gfp_flags); + if (!lgp) { + trace_pnfs_update_layout(ino, pos, count, iomode, lo, NULL, + PNFS_UPDATE_LAYOUT_NOMEM); + atomic_dec(&lo->plh_outstanding); + goto out_put_layout_hdr; + } + + lseg = nfs4_proc_layoutget(lgp, &timeout); trace_pnfs_update_layout(ino, pos, count, iomode, lo, lseg, PNFS_UPDATE_LAYOUT_SEND_LAYOUTGET); atomic_dec(&lo->plh_outstanding); @@ -1919,6 +2005,171 @@ pnfs_sanity_check_layout_range(struct pnfs_layout_range *range) return true; } +static struct pnfs_layout_hdr * +_pnfs_grab_empty_layout(struct inode *ino, struct nfs_open_context *ctx) +{ + struct pnfs_layout_hdr *lo; + + spin_lock(&ino->i_lock); + lo = pnfs_find_alloc_layout(ino, ctx, GFP_KERNEL); + if (!lo) + goto out_unlock; + if (!test_bit(NFS_LAYOUT_INVALID_STID, &lo->plh_flags)) + goto out_unlock; + if (test_bit(NFS_LAYOUT_RETURN, &lo->plh_flags)) + goto out_unlock; + if (pnfs_layoutgets_blocked(lo)) + goto out_unlock; + if (test_and_set_bit(NFS_LAYOUT_FIRST_LAYOUTGET, &lo->plh_flags)) + goto out_unlock; + atomic_inc(&lo->plh_outstanding); + spin_unlock(&ino->i_lock); + _add_to_server_list(lo, NFS_SERVER(ino)); + return lo; + +out_unlock: + spin_unlock(&ino->i_lock); + pnfs_put_layout_hdr(lo); + return NULL; +} + +extern const nfs4_stateid current_stateid; + +static void _lgopen_prepare_attached(struct nfs4_opendata *data, + struct nfs_open_context *ctx) +{ + struct inode *ino = data->dentry->d_inode; + 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; + struct pnfs_layout_hdr *lo; + + /* Heuristic: don't send layoutget if we have cached data */ + if (rng.iomode == IOMODE_READ && + (i_size_read(ino) == 0 || ino->i_mapping->nrpages != 0)) + return; + + lo = _pnfs_grab_empty_layout(ino, ctx); + if (!lo) + return; + lgp = pnfs_alloc_init_layoutget_args(ino, ctx, ¤t_stateid, + &rng, GFP_KERNEL); + if (!lgp) { + pnfs_clear_first_layoutget(lo); + pnfs_put_layout_hdr(lo); + return; + } + data->lgp = lgp; + data->o_arg.lg_args = &lgp->args; + data->o_res.lg_res = &lgp->res; +} + +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 (!nfs_server_capable(data->dir->d_inode, NFS_CAP_LGOPEN)) + return; + 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; + struct nfs_server *srv = NFS_SERVER(ino); + u32 iomode; + + if (!lgp) + return; + dprintk("%s: entered with status %i\n", __func__, lgp->res.status); + if (lgp->res.status) { + switch (lgp->res.status) { + default: + break; + /* + * Halt lgopen attempts if the server doesn't recognise + * the "current stateid" value, the layout type, or the + * layoutget operation as being valid. + * Also if it complains about too many ops in the compound + * or of the request/reply being too big. + */ + case -NFS4ERR_BAD_STATEID: + case -NFS4ERR_NOTSUPP: + case -NFS4ERR_REP_TOO_BIG: + case -NFS4ERR_REP_TOO_BIG_TO_CACHE: + case -NFS4ERR_REQ_TOO_BIG: + case -NFS4ERR_TOO_MANY_OPS: + case -NFS4ERR_UNKNOWN_LAYOUTTYPE: + srv->caps &= ~NFS_CAP_LGOPEN; + } + return; + } + if (!lgp->args.inode) { + lo = _pnfs_grab_empty_layout(ino, ctx); + if (!lo) + return; + lgp->args.inode = ino; + } else + lo = NFS_I(lgp->args.inode)->layout; + + if (read_seqcount_retry(&srv->nfs_client->cl_callback_count, + lgp->callback_count)) + return; + lseg = pnfs_layout_process(lgp); + if (!IS_ERR(lseg)) { + iomode = lgp->args.range.iomode; + pnfs_layout_clear_fail_bit(lo, pnfs_iomode_to_fail_bit(iomode)); + pnfs_put_lseg(lseg); + } +} + +void nfs4_lgopen_release(struct nfs4_layoutget *lgp) +{ + if (lgp != NULL) { + struct inode *inode = lgp->args.inode; + if (inode) { + struct pnfs_layout_hdr *lo = NFS_I(inode)->layout; + atomic_dec(&lo->plh_outstanding); + pnfs_clear_first_layoutget(lo); + } + pnfs_layoutget_free(lgp); + } +} + struct pnfs_layout_segment * pnfs_layout_process(struct nfs4_layoutget *lgp) { @@ -1984,8 +2235,6 @@ out_forget: spin_unlock(&ino->i_lock); lseg->pls_layout = lo; NFS_SERVER(ino)->pnfs_curr_ld->free_lseg(lseg); - if (!pnfs_layout_is_valid(lo)) - nfs_commit_inode(ino, 0); return ERR_PTR(-EAGAIN); } diff --git a/fs/nfs/pnfs.h b/fs/nfs/pnfs.h index daf6cbf5c15f..a8f5e6b16749 100644 --- a/fs/nfs/pnfs.h +++ b/fs/nfs/pnfs.h @@ -35,6 +35,8 @@ #include <linux/nfs_page.h> #include <linux/workqueue.h> +struct nfs4_opendata; + enum { NFS_LSEG_VALID = 0, /* cleared when lseg is recalled/returned */ NFS_LSEG_ROC, /* roc bit received from server */ @@ -110,6 +112,7 @@ enum layoutdriver_policy_flags { PNFS_LAYOUTRET_ON_SETATTR = 1 << 0, PNFS_LAYOUTRET_ON_ERROR = 1 << 1, PNFS_READ_WHOLE_PAGE = 1 << 2, + PNFS_LAYOUTGET_ON_OPEN = 1 << 3, }; struct nfs4_deviceid_node; @@ -223,10 +226,11 @@ extern int pnfs_register_layoutdriver(struct pnfs_layoutdriver_type *); extern void pnfs_unregister_layoutdriver(struct pnfs_layoutdriver_type *); /* nfs4proc.c */ +extern size_t max_response_pages(struct nfs_server *server); extern int nfs4_proc_getdeviceinfo(struct nfs_server *server, struct pnfs_device *dev, struct rpc_cred *cred); -extern struct pnfs_layout_segment* nfs4_proc_layoutget(struct nfs4_layoutget *lgp, long *timeout, gfp_t gfp_flags); +extern struct pnfs_layout_segment* nfs4_proc_layoutget(struct nfs4_layoutget *lgp, long *timeout); extern int nfs4_proc_layoutreturn(struct nfs4_layoutreturn *lrp, bool sync); /* pnfs.c */ @@ -246,6 +250,7 @@ size_t pnfs_generic_pg_test(struct nfs_pageio_descriptor *pgio, struct nfs_page *prev, struct nfs_page *req); void pnfs_set_lo_fail(struct pnfs_layout_segment *lseg); struct pnfs_layout_segment *pnfs_layout_process(struct nfs4_layoutget *lgp); +void pnfs_layoutget_free(struct nfs4_layoutget *lgp); void pnfs_free_lseg_list(struct list_head *tmp_list); void pnfs_destroy_layout(struct nfs_inode *); void pnfs_destroy_all_layouts(struct nfs_client *); @@ -375,6 +380,11 @@ 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); +void nfs4_lgopen_release(struct nfs4_layoutget *lgp); static inline bool nfs_have_layout(struct inode *inode) { @@ -775,6 +785,22 @@ 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) +{ +} + +static inline void nfs4_lgopen_release(struct nfs4_layoutget *lgp) +{ +} + #endif /* CONFIG_NFS_V4_1 */ #if IS_ENABLED(CONFIG_NFS_V4_2) diff --git a/fs/nfs/proc.c b/fs/nfs/proc.c index 4e93d6308733..e0c257bd62b9 100644 --- a/fs/nfs/proc.c +++ b/fs/nfs/proc.c @@ -99,7 +99,8 @@ nfs_proc_get_root(struct nfs_server *server, struct nfs_fh *fhandle, */ static int nfs_proc_getattr(struct nfs_server *server, struct nfs_fh *fhandle, - struct nfs_fattr *fattr, struct nfs4_label *label) + struct nfs_fattr *fattr, struct nfs4_label *label, + struct inode *inode) { struct rpc_message msg = { .rpc_proc = &nfs_procedures[NFSPROC_GETATTR], @@ -321,7 +322,9 @@ nfs_proc_remove(struct inode *dir, struct dentry *dentry) } static void -nfs_proc_unlink_setup(struct rpc_message *msg, struct dentry *dentry) +nfs_proc_unlink_setup(struct rpc_message *msg, + struct dentry *dentry, + struct inode *inode) { msg->rpc_proc = &nfs_procedures[NFSPROC_REMOVE]; } @@ -618,7 +621,8 @@ static int nfs_write_done(struct rpc_task *task, struct nfs_pgio_header *hdr) } static void nfs_proc_write_setup(struct nfs_pgio_header *hdr, - struct rpc_message *msg) + struct rpc_message *msg, + struct rpc_clnt **clnt) { /* Note: NFSv2 ignores @stable and always uses NFS_FILE_SYNC */ hdr->args.stable = NFS_FILE_SYNC; @@ -631,7 +635,8 @@ static void nfs_proc_commit_rpc_prepare(struct rpc_task *task, struct nfs_commit } static void -nfs_proc_commit_setup(struct nfs_commit_data *data, struct rpc_message *msg) +nfs_proc_commit_setup(struct nfs_commit_data *data, struct rpc_message *msg, + struct rpc_clnt **clnt) { BUG(); } diff --git a/fs/nfs/unlink.c b/fs/nfs/unlink.c index bf54fc9ae135..fd61bf0fce63 100644 --- a/fs/nfs/unlink.c +++ b/fs/nfs/unlink.c @@ -85,7 +85,7 @@ static const struct rpc_call_ops nfs_unlink_ops = { .rpc_call_prepare = nfs_unlink_prepare, }; -static void nfs_do_call_unlink(struct nfs_unlinkdata *data) +static void nfs_do_call_unlink(struct inode *inode, struct nfs_unlinkdata *data) { struct rpc_message msg = { .rpc_argp = &data->args, @@ -105,7 +105,7 @@ static void nfs_do_call_unlink(struct nfs_unlinkdata *data) data->args.fh = NFS_FH(dir); nfs_fattr_init(data->res.dir_attr); - NFS_PROTO(dir)->unlink_setup(&msg, data->dentry); + NFS_PROTO(dir)->unlink_setup(&msg, data->dentry, inode); task_setup_data.rpc_client = NFS_CLIENT(dir); task = rpc_run_task(&task_setup_data); @@ -113,7 +113,7 @@ static void nfs_do_call_unlink(struct nfs_unlinkdata *data) rpc_put_task_async(task); } -static int nfs_call_unlink(struct dentry *dentry, struct nfs_unlinkdata *data) +static int nfs_call_unlink(struct dentry *dentry, struct inode *inode, struct nfs_unlinkdata *data) { struct inode *dir = d_inode(dentry->d_parent); struct dentry *alias; @@ -153,7 +153,7 @@ static int nfs_call_unlink(struct dentry *dentry, struct nfs_unlinkdata *data) return ret; } data->dentry = alias; - nfs_do_call_unlink(data); + nfs_do_call_unlink(inode, data); return 1; } @@ -231,7 +231,7 @@ nfs_complete_unlink(struct dentry *dentry, struct inode *inode) dentry->d_fsdata = NULL; spin_unlock(&dentry->d_lock); - if (NFS_STALE(inode) || !nfs_call_unlink(dentry, data)) + if (NFS_STALE(inode) || !nfs_call_unlink(dentry, inode, data)) nfs_free_unlinkdata(data); } @@ -448,6 +448,7 @@ nfs_sillyrename(struct inode *dir, struct dentry *dentry) unsigned char silly[SILLYNAME_LEN + 1]; unsigned long long fileid; struct dentry *sdentry; + struct inode *inode = d_inode(dentry); struct rpc_task *task; int error = -EBUSY; @@ -485,6 +486,8 @@ nfs_sillyrename(struct inode *dir, struct dentry *dentry) goto out; } while (d_inode(sdentry) != NULL); /* need negative lookup */ + ihold(inode); + /* queue unlink first. Can't do this from rpc_release as it * has to allocate memory */ @@ -509,6 +512,12 @@ nfs_sillyrename(struct inode *dir, struct dentry *dentry) case 0: /* The rename succeeded */ nfs_set_verifier(dentry, nfs_save_change_attribute(dir)); + spin_lock(&inode->i_lock); + NFS_I(inode)->attr_gencount = nfs_inc_attr_generation_counter(); + NFS_I(inode)->cache_validity |= NFS_INO_INVALID_CHANGE + | NFS_INO_INVALID_CTIME + | NFS_INO_REVAL_FORCED; + spin_unlock(&inode->i_lock); d_move(dentry, sdentry); break; case -ERESTARTSYS: @@ -519,6 +528,7 @@ nfs_sillyrename(struct inode *dir, struct dentry *dentry) } rpc_put_task(task); out_dput: + iput(inode); dput(sdentry); out: return error; diff --git a/fs/nfs/write.c b/fs/nfs/write.c index 0193053bc139..a057b4f45a46 100644 --- a/fs/nfs/write.c +++ b/fs/nfs/write.c @@ -1375,12 +1375,9 @@ static void nfs_initiate_write(struct nfs_pgio_header *hdr, int priority = flush_task_priority(how); task_setup_data->priority = priority; - rpc_ops->write_setup(hdr, msg); + rpc_ops->write_setup(hdr, msg, &task_setup_data->rpc_client); trace_nfs_initiate_write(hdr->inode, hdr->io_start, hdr->good_bytes, hdr->args.stable); - - nfs4_state_protect_write(NFS_SERVER(hdr->inode)->nfs_client, - &task_setup_data->rpc_client, msg, hdr); } /* If a nfs_flush_* function fails, it should remove reqs from @head and @@ -1669,14 +1666,11 @@ int nfs_initiate_commit(struct rpc_clnt *clnt, struct nfs_commit_data *data, .priority = priority, }; /* Set up the initial task struct. */ - nfs_ops->commit_setup(data, &msg); + nfs_ops->commit_setup(data, &msg, &task_setup_data.rpc_client); trace_nfs_initiate_commit(data); dprintk("NFS: initiated commit call\n"); - nfs4_state_protect(NFS_SERVER(data->inode)->nfs_client, - NFS_SP4_MACH_CRED_COMMIT, &task_setup_data.rpc_client, &msg); - task = rpc_run_task(&task_setup_data); if (IS_ERR(task)) return PTR_ERR(task); diff --git a/fs/nfsd/vfs.c b/fs/nfsd/vfs.c index 2410b093a2e6..b0555d7d8200 100644 --- a/fs/nfsd/vfs.c +++ b/fs/nfsd/vfs.c @@ -1201,6 +1201,28 @@ nfsd_create_locked(struct svc_rqst *rqstp, struct svc_fh *fhp, break; case S_IFDIR: host_err = vfs_mkdir(dirp, dchild, iap->ia_mode); + if (!host_err && unlikely(d_unhashed(dchild))) { + struct dentry *d; + d = lookup_one_len(dchild->d_name.name, + dchild->d_parent, + dchild->d_name.len); + if (IS_ERR(d)) { + host_err = PTR_ERR(d); + break; + } + if (unlikely(d_is_negative(d))) { + dput(d); + err = nfserr_serverfault; + goto out; + } + dput(resfhp->fh_dentry); + resfhp->fh_dentry = dget(d); + err = fh_update(resfhp); + dput(dchild); + dchild = d; + if (err) + goto out; + } break; case S_IFCHR: case S_IFBLK: diff --git a/fs/nilfs2/namei.c b/fs/nilfs2/namei.c index 1a2894aa0194..dd52d3f82e8d 100644 --- a/fs/nilfs2/namei.c +++ b/fs/nilfs2/namei.c @@ -46,8 +46,7 @@ static inline int nilfs_add_nondir(struct dentry *dentry, struct inode *inode) int err = nilfs_add_link(dentry, inode); if (!err) { - d_instantiate(dentry, inode); - unlock_new_inode(inode); + d_instantiate_new(dentry, inode); return 0; } inode_dec_link_count(inode); @@ -243,8 +242,7 @@ static int nilfs_mkdir(struct inode *dir, struct dentry *dentry, umode_t mode) goto out_fail; nilfs_mark_inode_dirty(inode); - d_instantiate(dentry, inode); - unlock_new_inode(inode); + d_instantiate_new(dentry, inode); out: if (!err) err = nilfs_transaction_commit(dir->i_sb); diff --git a/fs/ocfs2/cluster/heartbeat.c b/fs/ocfs2/cluster/heartbeat.c index 91a8889abf9b..ea8c551bcd7e 100644 --- a/fs/ocfs2/cluster/heartbeat.c +++ b/fs/ocfs2/cluster/heartbeat.c @@ -570,16 +570,7 @@ static struct bio *o2hb_setup_one_bio(struct o2hb_region *reg, current_page, vec_len, vec_start); len = bio_add_page(bio, page, vec_len, vec_start); - if (len != vec_len) { - mlog(ML_ERROR, "Adding page[%d] to bio failed, " - "page %p, len %d, vec_len %u, vec_start %u, " - "bi_sector %llu\n", current_page, page, len, - vec_len, vec_start, - (unsigned long long)bio->bi_iter.bi_sector); - bio_put(bio); - bio = ERR_PTR(-EIO); - return bio; - } + if (len != vec_len) break; cs += vec_len / (PAGE_SIZE/spp); vec_start = 0; diff --git a/fs/orangefs/namei.c b/fs/orangefs/namei.c index 6e3134e6d98a..1b5707c44c3f 100644 --- a/fs/orangefs/namei.c +++ b/fs/orangefs/namei.c @@ -75,8 +75,7 @@ static int orangefs_create(struct inode *dir, get_khandle_from_ino(inode), dentry); - d_instantiate(dentry, inode); - unlock_new_inode(inode); + d_instantiate_new(dentry, inode); orangefs_set_timeout(dentry); ORANGEFS_I(inode)->getattr_time = jiffies - 1; ORANGEFS_I(inode)->getattr_mask = STATX_BASIC_STATS; @@ -332,8 +331,7 @@ static int orangefs_symlink(struct inode *dir, "Assigned symlink inode new number of %pU\n", get_khandle_from_ino(inode)); - d_instantiate(dentry, inode); - unlock_new_inode(inode); + d_instantiate_new(dentry, inode); orangefs_set_timeout(dentry); ORANGEFS_I(inode)->getattr_time = jiffies - 1; ORANGEFS_I(inode)->getattr_mask = STATX_BASIC_STATS; @@ -402,8 +400,7 @@ static int orangefs_mkdir(struct inode *dir, struct dentry *dentry, umode_t mode "Assigned dir inode new number of %pU\n", get_khandle_from_ino(inode)); - d_instantiate(dentry, inode); - unlock_new_inode(inode); + d_instantiate_new(dentry, inode); orangefs_set_timeout(dentry); ORANGEFS_I(inode)->getattr_time = jiffies - 1; ORANGEFS_I(inode)->getattr_mask = STATX_BASIC_STATS; diff --git a/fs/proc/array.c b/fs/proc/array.c index ae2c807fd719..72391b3f6927 100644 --- a/fs/proc/array.c +++ b/fs/proc/array.c @@ -85,6 +85,7 @@ #include <linux/delayacct.h> #include <linux/seq_file.h> #include <linux/pid_namespace.h> +#include <linux/prctl.h> #include <linux/ptrace.h> #include <linux/tracehook.h> #include <linux/string_helpers.h> @@ -335,6 +336,30 @@ static inline void task_seccomp(struct seq_file *m, struct task_struct *p) #ifdef CONFIG_SECCOMP seq_put_decimal_ull(m, "\nSeccomp:\t", p->seccomp.mode); #endif + seq_printf(m, "\nSpeculation_Store_Bypass:\t"); + switch (arch_prctl_spec_ctrl_get(p, PR_SPEC_STORE_BYPASS)) { + case -EINVAL: + seq_printf(m, "unknown"); + break; + case PR_SPEC_NOT_AFFECTED: + seq_printf(m, "not vulnerable"); + break; + case PR_SPEC_PRCTL | PR_SPEC_FORCE_DISABLE: + seq_printf(m, "thread force mitigated"); + break; + case PR_SPEC_PRCTL | PR_SPEC_DISABLE: + seq_printf(m, "thread mitigated"); + break; + case PR_SPEC_PRCTL | PR_SPEC_ENABLE: + seq_printf(m, "thread vulnerable"); + break; + case PR_SPEC_DISABLE: + seq_printf(m, "globally mitigated"); + break; + default: + seq_printf(m, "vulnerable"); + break; + } seq_putc(m, '\n'); } diff --git a/fs/reiserfs/namei.c b/fs/reiserfs/namei.c index d2869ced6d39..97f3fc4fdd79 100644 --- a/fs/reiserfs/namei.c +++ b/fs/reiserfs/namei.c @@ -687,8 +687,7 @@ static int reiserfs_create(struct inode *dir, struct dentry *dentry, umode_t mod reiserfs_update_inode_transaction(inode); reiserfs_update_inode_transaction(dir); - unlock_new_inode(inode); - d_instantiate(dentry, inode); + d_instantiate_new(dentry, inode); retval = journal_end(&th); out_failed: @@ -771,8 +770,7 @@ static int reiserfs_mknod(struct inode *dir, struct dentry *dentry, umode_t mode goto out_failed; } - unlock_new_inode(inode); - d_instantiate(dentry, inode); + d_instantiate_new(dentry, inode); retval = journal_end(&th); out_failed: @@ -871,8 +869,7 @@ static int reiserfs_mkdir(struct inode *dir, struct dentry *dentry, umode_t mode /* the above add_entry did not update dir's stat data */ reiserfs_update_sd(&th, dir); - unlock_new_inode(inode); - d_instantiate(dentry, inode); + d_instantiate_new(dentry, inode); retval = journal_end(&th); out_failed: reiserfs_write_unlock(dir->i_sb); @@ -1187,8 +1184,7 @@ static int reiserfs_symlink(struct inode *parent_dir, goto out_failed; } - unlock_new_inode(inode); - d_instantiate(dentry, inode); + d_instantiate_new(dentry, inode); retval = journal_end(&th); out_failed: reiserfs_write_unlock(parent_dir->i_sb); diff --git a/fs/seq_file.c b/fs/seq_file.c index c6c27f1f9c98..4cc090b50cc5 100644 --- a/fs/seq_file.c +++ b/fs/seq_file.c @@ -709,11 +709,6 @@ void seq_put_decimal_ull_width(struct seq_file *m, const char *delimiter, if (m->count + width >= m->size) goto overflow; - if (num < 10) { - m->buf[m->count++] = num + '0'; - return; - } - len = num_to_str(m->buf + m->count, m->size - m->count, num, width); if (!len) goto overflow; diff --git a/fs/super.c b/fs/super.c index 122c402049a2..4b5b562176d0 100644 --- a/fs/super.c +++ b/fs/super.c @@ -121,13 +121,23 @@ static unsigned long super_cache_count(struct shrinker *shrink, sb = container_of(shrink, struct super_block, s_shrink); /* - * Don't call trylock_super as it is a potential - * scalability bottleneck. The counts could get updated - * between super_cache_count and super_cache_scan anyway. - * Call to super_cache_count with shrinker_rwsem held - * ensures the safety of call to list_lru_shrink_count() and - * s_op->nr_cached_objects(). + * We don't call trylock_super() here as it is a scalability bottleneck, + * so we're exposed to partial setup state. The shrinker rwsem does not + * protect filesystem operations backing list_lru_shrink_count() or + * s_op->nr_cached_objects(). Counts can change between + * super_cache_count and super_cache_scan, so we really don't need locks + * here. + * + * However, if we are currently mounting the superblock, the underlying + * filesystem might be in a state of partial construction and hence it + * is dangerous to access it. trylock_super() uses a SB_BORN check to + * avoid this situation, so do the same here. The memory barrier is + * matched with the one in mount_fs() as we don't hold locks here. */ + if (!(sb->s_flags & SB_BORN)) + return 0; + smp_rmb(); + if (sb->s_op && sb->s_op->nr_cached_objects) total_objects = sb->s_op->nr_cached_objects(sb, sc); @@ -1272,6 +1282,14 @@ mount_fs(struct file_system_type *type, int flags, const char *name, void *data) sb = root->d_sb; BUG_ON(!sb); WARN_ON(!sb->s_bdi); + + /* + * Write barrier is for super_cache_count(). We place it before setting + * SB_BORN as the data dependency between the two functions is the + * superblock structure contents that we just set up, not the SB_BORN + * flag. + */ + smp_wmb(); sb->s_flags |= SB_BORN; error = security_sb_kern_mount(sb, flags, secdata); diff --git a/fs/sysfs/mount.c b/fs/sysfs/mount.c index b428d317ae92..92682fcc41f6 100644 --- a/fs/sysfs/mount.c +++ b/fs/sysfs/mount.c @@ -25,7 +25,7 @@ static struct dentry *sysfs_mount(struct file_system_type *fs_type, { struct dentry *root; void *ns; - bool new_sb; + bool new_sb = false; if (!(flags & SB_KERNMOUNT)) { if (!kobj_ns_current_may_mount(KOBJ_NS_TYPE_NET)) @@ -35,9 +35,9 @@ static struct dentry *sysfs_mount(struct file_system_type *fs_type, ns = kobj_ns_grab_current(KOBJ_NS_TYPE_NET); root = kernfs_mount_ns(fs_type, flags, sysfs_root, SYSFS_MAGIC, &new_sb, ns); - if (IS_ERR(root) || !new_sb) + if (!new_sb) kobj_ns_drop(KOBJ_NS_TYPE_NET, ns); - else if (new_sb) + else if (!IS_ERR(root)) root->d_sb->s_iflags |= SB_I_USERNS_VISIBLE; return root; diff --git a/fs/udf/namei.c b/fs/udf/namei.c index 0458dd47e105..c586026508db 100644 --- a/fs/udf/namei.c +++ b/fs/udf/namei.c @@ -622,8 +622,7 @@ static int udf_add_nondir(struct dentry *dentry, struct inode *inode) if (fibh.sbh != fibh.ebh) brelse(fibh.ebh); brelse(fibh.sbh); - unlock_new_inode(inode); - d_instantiate(dentry, inode); + d_instantiate_new(dentry, inode); return 0; } @@ -733,8 +732,7 @@ static int udf_mkdir(struct inode *dir, struct dentry *dentry, umode_t mode) inc_nlink(dir); dir->i_ctime = dir->i_mtime = current_time(dir); mark_inode_dirty(dir); - unlock_new_inode(inode); - d_instantiate(dentry, inode); + d_instantiate_new(dentry, inode); if (fibh.sbh != fibh.ebh) brelse(fibh.ebh); brelse(fibh.sbh); diff --git a/fs/ufs/namei.c b/fs/ufs/namei.c index 32545cd00ceb..d5f43ba76c59 100644 --- a/fs/ufs/namei.c +++ b/fs/ufs/namei.c @@ -39,8 +39,7 @@ static inline int ufs_add_nondir(struct dentry *dentry, struct inode *inode) { int err = ufs_add_link(dentry, inode); if (!err) { - unlock_new_inode(inode); - d_instantiate(dentry, inode); + d_instantiate_new(dentry, inode); return 0; } inode_dec_link_count(inode); @@ -193,8 +192,7 @@ static int ufs_mkdir(struct inode * dir, struct dentry * dentry, umode_t mode) if (err) goto out_fail; - unlock_new_inode(inode); - d_instantiate(dentry, inode); + d_instantiate_new(dentry, inode); return 0; out_fail: |