diff options
Diffstat (limited to 'fs')
96 files changed, 1463 insertions, 875 deletions
diff --git a/fs/9p/cache.c b/fs/9p/cache.c index 1c8dc696d516..cebba4eaa0b5 100644 --- a/fs/9p/cache.c +++ b/fs/9p/cache.c @@ -62,12 +62,12 @@ void v9fs_cache_inode_get_cookie(struct inode *inode) version = cpu_to_le32(v9inode->qid.version); path = cpu_to_le64(v9inode->qid.path); v9ses = v9fs_inode2v9ses(inode); - v9inode->netfs_ctx.cache = + v9inode->netfs.cache = fscache_acquire_cookie(v9fs_session_cache(v9ses), 0, &path, sizeof(path), &version, sizeof(version), - i_size_read(&v9inode->vfs_inode)); + i_size_read(&v9inode->netfs.inode)); p9_debug(P9_DEBUG_FSC, "inode %p get cookie %p\n", inode, v9fs_inode_cookie(v9inode)); diff --git a/fs/9p/fid.c b/fs/9p/fid.c index 79df61fe0e59..baf2b152229e 100644 --- a/fs/9p/fid.c +++ b/fs/9p/fid.c @@ -152,7 +152,7 @@ static struct p9_fid *v9fs_fid_lookup_with_uid(struct dentry *dentry, const unsigned char **wnames, *uname; int i, n, l, clone, access; struct v9fs_session_info *v9ses; - struct p9_fid *fid, *old_fid = NULL; + struct p9_fid *fid, *old_fid; v9ses = v9fs_dentry2v9ses(dentry); access = v9ses->flags & V9FS_ACCESS_MASK; @@ -194,13 +194,12 @@ static struct p9_fid *v9fs_fid_lookup_with_uid(struct dentry *dentry, if (IS_ERR(fid)) return fid; + refcount_inc(&fid->count); v9fs_fid_add(dentry->d_sb->s_root, fid); } /* If we are root ourself just return that */ - if (dentry->d_sb->s_root == dentry) { - refcount_inc(&fid->count); + if (dentry->d_sb->s_root == dentry) return fid; - } /* * Do a multipath walk with attached root. * When walking parent we need to make sure we @@ -212,6 +211,7 @@ static struct p9_fid *v9fs_fid_lookup_with_uid(struct dentry *dentry, fid = ERR_PTR(n); goto err_out; } + old_fid = fid; clone = 1; i = 0; while (i < n) { @@ -221,19 +221,15 @@ static struct p9_fid *v9fs_fid_lookup_with_uid(struct dentry *dentry, * walk to ensure none of the patch component change */ fid = p9_client_walk(fid, l, &wnames[i], clone); + /* non-cloning walk will return the same fid */ + if (fid != old_fid) { + p9_client_clunk(old_fid); + old_fid = fid; + } if (IS_ERR(fid)) { - if (old_fid) { - /* - * If we fail, clunk fid which are mapping - * to path component and not the last component - * of the path. - */ - p9_client_clunk(old_fid); - } kfree(wnames); goto err_out; } - old_fid = fid; i += l; clone = 0; } diff --git a/fs/9p/v9fs.c b/fs/9p/v9fs.c index e28ddf763b3b..0129de2ea31a 100644 --- a/fs/9p/v9fs.c +++ b/fs/9p/v9fs.c @@ -625,7 +625,7 @@ static void v9fs_inode_init_once(void *foo) struct v9fs_inode *v9inode = (struct v9fs_inode *)foo; memset(&v9inode->qid, 0, sizeof(v9inode->qid)); - inode_init_once(&v9inode->vfs_inode); + inode_init_once(&v9inode->netfs.inode); } /** diff --git a/fs/9p/v9fs.h b/fs/9p/v9fs.h index ec0e8df3b2eb..6acabc2e7dc9 100644 --- a/fs/9p/v9fs.h +++ b/fs/9p/v9fs.h @@ -109,11 +109,7 @@ struct v9fs_session_info { #define V9FS_INO_INVALID_ATTR 0x01 struct v9fs_inode { - struct { - /* These must be contiguous */ - struct inode vfs_inode; /* the VFS's inode record */ - struct netfs_i_context netfs_ctx; /* Netfslib context */ - }; + struct netfs_inode netfs; /* Netfslib context and vfs inode */ struct p9_qid qid; unsigned int cache_validity; struct p9_fid *writeback_fid; @@ -122,13 +118,13 @@ struct v9fs_inode { static inline struct v9fs_inode *V9FS_I(const struct inode *inode) { - return container_of(inode, struct v9fs_inode, vfs_inode); + return container_of(inode, struct v9fs_inode, netfs.inode); } static inline struct fscache_cookie *v9fs_inode_cookie(struct v9fs_inode *v9inode) { #ifdef CONFIG_9P_FSCACHE - return netfs_i_cookie(&v9inode->vfs_inode); + return netfs_i_cookie(&v9inode->netfs); #else return NULL; #endif diff --git a/fs/9p/vfs_addr.c b/fs/9p/vfs_addr.c index 8ce82ff1e40a..d0833fa69faf 100644 --- a/fs/9p/vfs_addr.c +++ b/fs/9p/vfs_addr.c @@ -58,21 +58,33 @@ static void v9fs_issue_read(struct netfs_io_subrequest *subreq) */ static int v9fs_init_request(struct netfs_io_request *rreq, struct file *file) { + struct inode *inode = file_inode(file); + struct v9fs_inode *v9inode = V9FS_I(inode); struct p9_fid *fid = file->private_data; + BUG_ON(!fid); + + /* we might need to read from a fid that was opened write-only + * for read-modify-write of page cache, use the writeback fid + * for that */ + if (rreq->origin == NETFS_READ_FOR_WRITE && + (fid->mode & O_ACCMODE) == O_WRONLY) { + fid = v9inode->writeback_fid; + BUG_ON(!fid); + } + refcount_inc(&fid->count); rreq->netfs_priv = fid; return 0; } /** - * v9fs_req_cleanup - Cleanup request initialized by v9fs_init_request - * @mapping: unused mapping of request to cleanup - * @priv: private data to cleanup, a fid, guaranted non-null. + * v9fs_free_request - Cleanup request initialized by v9fs_init_rreq + * @rreq: The I/O request to clean up */ -static void v9fs_req_cleanup(struct address_space *mapping, void *priv) +static void v9fs_free_request(struct netfs_io_request *rreq) { - struct p9_fid *fid = priv; + struct p9_fid *fid = rreq->netfs_priv; p9_client_clunk(fid); } @@ -94,9 +106,9 @@ static int v9fs_begin_cache_operation(struct netfs_io_request *rreq) const struct netfs_request_ops v9fs_req_ops = { .init_request = v9fs_init_request, + .free_request = v9fs_free_request, .begin_cache_operation = v9fs_begin_cache_operation, .issue_read = v9fs_issue_read, - .cleanup = v9fs_req_cleanup, }; /** @@ -140,7 +152,7 @@ static void v9fs_write_to_cache_done(void *priv, ssize_t transferred_or_error, transferred_or_error != -ENOBUFS) { version = cpu_to_le32(v9inode->qid.version); fscache_invalidate(v9fs_inode_cookie(v9inode), &version, - i_size_read(&v9inode->vfs_inode), 0); + i_size_read(&v9inode->netfs.inode), 0); } } @@ -274,7 +286,7 @@ static int v9fs_write_begin(struct file *filp, struct address_space *mapping, * file. We need to do this before we get a lock on the page in case * there's more than one writer competing for the same cache block. */ - retval = netfs_write_begin(filp, mapping, pos, len, &folio, fsdata); + retval = netfs_write_begin(&v9inode->netfs, filp, mapping, pos, len, &folio, fsdata); if (retval < 0) return retval; diff --git a/fs/9p/vfs_inode.c b/fs/9p/vfs_inode.c index 55367ecb9442..3d8297714772 100644 --- a/fs/9p/vfs_inode.c +++ b/fs/9p/vfs_inode.c @@ -234,7 +234,7 @@ struct inode *v9fs_alloc_inode(struct super_block *sb) v9inode->writeback_fid = NULL; v9inode->cache_validity = 0; mutex_init(&v9inode->v_mutex); - return &v9inode->vfs_inode; + return &v9inode->netfs.inode; } /** @@ -252,7 +252,8 @@ void v9fs_free_inode(struct inode *inode) */ static void v9fs_set_netfs_context(struct inode *inode) { - netfs_i_context_init(inode, &v9fs_req_ops); + struct v9fs_inode *v9inode = V9FS_I(inode); + netfs_inode_init(&v9inode->netfs, &v9fs_req_ops); } int v9fs_init_inode(struct v9fs_session_info *v9ses, @@ -1250,15 +1251,15 @@ static const char *v9fs_vfs_get_link(struct dentry *dentry, return ERR_PTR(-ECHILD); v9ses = v9fs_dentry2v9ses(dentry); - fid = v9fs_fid_lookup(dentry); + if (!v9fs_proto_dotu(v9ses)) + return ERR_PTR(-EBADF); + p9_debug(P9_DEBUG_VFS, "%pd\n", dentry); + fid = v9fs_fid_lookup(dentry); if (IS_ERR(fid)) return ERR_CAST(fid); - if (!v9fs_proto_dotu(v9ses)) - return ERR_PTR(-EBADF); - st = p9_client_stat(fid); p9_client_clunk(fid); if (IS_ERR(st)) diff --git a/fs/9p/vfs_inode_dotl.c b/fs/9p/vfs_inode_dotl.c index d17502a738a9..b6eb1160296c 100644 --- a/fs/9p/vfs_inode_dotl.c +++ b/fs/9p/vfs_inode_dotl.c @@ -274,6 +274,7 @@ v9fs_vfs_atomic_open_dotl(struct inode *dir, struct dentry *dentry, if (IS_ERR(ofid)) { err = PTR_ERR(ofid); p9_debug(P9_DEBUG_VFS, "p9_client_walk failed %d\n", err); + p9_client_clunk(dfid); goto out; } @@ -285,6 +286,7 @@ v9fs_vfs_atomic_open_dotl(struct inode *dir, struct dentry *dentry, if (err) { p9_debug(P9_DEBUG_VFS, "Failed to get acl values in creat %d\n", err); + p9_client_clunk(dfid); goto error; } err = p9_client_create_dotl(ofid, name, v9fs_open_to_dotl_flags(flags), @@ -292,6 +294,7 @@ v9fs_vfs_atomic_open_dotl(struct inode *dir, struct dentry *dentry, if (err < 0) { p9_debug(P9_DEBUG_VFS, "p9_client_open_dotl failed in creat %d\n", err); + p9_client_clunk(dfid); goto error; } v9fs_invalidate_inode_attr(dir); diff --git a/fs/afs/callback.c b/fs/afs/callback.c index 1b4d5809808d..a484fa642808 100644 --- a/fs/afs/callback.c +++ b/fs/afs/callback.c @@ -30,7 +30,7 @@ void afs_invalidate_mmap_work(struct work_struct *work) { struct afs_vnode *vnode = container_of(work, struct afs_vnode, cb_work); - unmap_mapping_pages(vnode->vfs_inode.i_mapping, 0, 0, false); + unmap_mapping_pages(vnode->netfs.inode.i_mapping, 0, 0, false); } void afs_server_init_callback_work(struct work_struct *work) diff --git a/fs/afs/dir.c b/fs/afs/dir.c index 79f6b74336d2..56ae5cd5184f 100644 --- a/fs/afs/dir.c +++ b/fs/afs/dir.c @@ -109,7 +109,7 @@ struct afs_lookup_cookie { */ static void afs_dir_read_cleanup(struct afs_read *req) { - struct address_space *mapping = req->vnode->vfs_inode.i_mapping; + struct address_space *mapping = req->vnode->netfs.inode.i_mapping; struct folio *folio; pgoff_t last = req->nr_pages - 1; @@ -153,7 +153,7 @@ static bool afs_dir_check_folio(struct afs_vnode *dvnode, struct folio *folio, block = kmap_local_folio(folio, offset); if (block->hdr.magic != AFS_DIR_MAGIC) { printk("kAFS: %s(%lx): [%llx] bad magic %zx/%zx is %04hx\n", - __func__, dvnode->vfs_inode.i_ino, + __func__, dvnode->netfs.inode.i_ino, pos, offset, size, ntohs(block->hdr.magic)); trace_afs_dir_check_failed(dvnode, pos + offset, i_size); kunmap_local(block); @@ -183,7 +183,7 @@ error: static void afs_dir_dump(struct afs_vnode *dvnode, struct afs_read *req) { union afs_xdr_dir_block *block; - struct address_space *mapping = dvnode->vfs_inode.i_mapping; + struct address_space *mapping = dvnode->netfs.inode.i_mapping; struct folio *folio; pgoff_t last = req->nr_pages - 1; size_t offset, size; @@ -217,7 +217,7 @@ static void afs_dir_dump(struct afs_vnode *dvnode, struct afs_read *req) */ static int afs_dir_check(struct afs_vnode *dvnode, struct afs_read *req) { - struct address_space *mapping = dvnode->vfs_inode.i_mapping; + struct address_space *mapping = dvnode->netfs.inode.i_mapping; struct folio *folio; pgoff_t last = req->nr_pages - 1; int ret = 0; @@ -269,7 +269,7 @@ static int afs_dir_open(struct inode *inode, struct file *file) static struct afs_read *afs_read_dir(struct afs_vnode *dvnode, struct key *key) __acquires(&dvnode->validate_lock) { - struct address_space *mapping = dvnode->vfs_inode.i_mapping; + struct address_space *mapping = dvnode->netfs.inode.i_mapping; struct afs_read *req; loff_t i_size; int nr_pages, i; @@ -287,7 +287,7 @@ static struct afs_read *afs_read_dir(struct afs_vnode *dvnode, struct key *key) req->cleanup = afs_dir_read_cleanup; expand: - i_size = i_size_read(&dvnode->vfs_inode); + i_size = i_size_read(&dvnode->netfs.inode); if (i_size < 2048) { ret = afs_bad(dvnode, afs_file_error_dir_small); goto error; @@ -305,7 +305,7 @@ expand: req->actual_len = i_size; /* May change */ req->len = nr_pages * PAGE_SIZE; /* We can ask for more than there is */ req->data_version = dvnode->status.data_version; /* May change */ - iov_iter_xarray(&req->def_iter, READ, &dvnode->vfs_inode.i_mapping->i_pages, + iov_iter_xarray(&req->def_iter, READ, &dvnode->netfs.inode.i_mapping->i_pages, 0, i_size); req->iter = &req->def_iter; @@ -897,7 +897,7 @@ static struct inode *afs_do_lookup(struct inode *dir, struct dentry *dentry, out_op: if (op->error == 0) { - inode = &op->file[1].vnode->vfs_inode; + inode = &op->file[1].vnode->netfs.inode; op->file[1].vnode = NULL; } @@ -1139,7 +1139,7 @@ static int afs_d_revalidate(struct dentry *dentry, unsigned int flags) afs_stat_v(dir, n_reval); /* search the directory for this vnode */ - ret = afs_do_lookup_one(&dir->vfs_inode, dentry, &fid, key, &dir_version); + ret = afs_do_lookup_one(&dir->netfs.inode, dentry, &fid, key, &dir_version); switch (ret) { case 0: /* the filename maps to something */ @@ -1170,7 +1170,7 @@ static int afs_d_revalidate(struct dentry *dentry, unsigned int flags) _debug("%pd: file deleted (uq %u -> %u I:%u)", dentry, fid.unique, vnode->fid.unique, - vnode->vfs_inode.i_generation); + vnode->netfs.inode.i_generation); goto not_found; } goto out_valid; @@ -1368,7 +1368,7 @@ static void afs_dir_remove_subdir(struct dentry *dentry) if (d_really_is_positive(dentry)) { struct afs_vnode *vnode = AFS_FS_I(d_inode(dentry)); - clear_nlink(&vnode->vfs_inode); + clear_nlink(&vnode->netfs.inode); set_bit(AFS_VNODE_DELETED, &vnode->flags); clear_bit(AFS_VNODE_CB_PROMISED, &vnode->flags); clear_bit(AFS_VNODE_DIR_VALID, &vnode->flags); @@ -1487,8 +1487,8 @@ static void afs_dir_remove_link(struct afs_operation *op) /* Already done */ } else if (test_bit(AFS_VNODE_DIR_VALID, &dvnode->flags)) { write_seqlock(&vnode->cb_lock); - drop_nlink(&vnode->vfs_inode); - if (vnode->vfs_inode.i_nlink == 0) { + drop_nlink(&vnode->netfs.inode); + if (vnode->netfs.inode.i_nlink == 0) { set_bit(AFS_VNODE_DELETED, &vnode->flags); __afs_break_callback(vnode, afs_cb_break_for_unlink); } @@ -1504,7 +1504,7 @@ static void afs_dir_remove_link(struct afs_operation *op) op->error = ret; } - _debug("nlink %d [val %d]", vnode->vfs_inode.i_nlink, op->error); + _debug("nlink %d [val %d]", vnode->netfs.inode.i_nlink, op->error); } static void afs_unlink_success(struct afs_operation *op) @@ -1680,8 +1680,8 @@ static void afs_link_success(struct afs_operation *op) afs_update_dentry_version(op, dvp, op->dentry); if (op->dentry_2->d_parent == op->dentry->d_parent) afs_update_dentry_version(op, dvp, op->dentry_2); - ihold(&vp->vnode->vfs_inode); - d_instantiate(op->dentry, &vp->vnode->vfs_inode); + ihold(&vp->vnode->netfs.inode); + d_instantiate(op->dentry, &vp->vnode->netfs.inode); } static void afs_link_put(struct afs_operation *op) diff --git a/fs/afs/dir_edit.c b/fs/afs/dir_edit.c index d98e109ecee9..0ab7752d1b75 100644 --- a/fs/afs/dir_edit.c +++ b/fs/afs/dir_edit.c @@ -109,7 +109,7 @@ static void afs_clear_contig_bits(union afs_xdr_dir_block *block, */ static struct folio *afs_dir_get_folio(struct afs_vnode *vnode, pgoff_t index) { - struct address_space *mapping = vnode->vfs_inode.i_mapping; + struct address_space *mapping = vnode->netfs.inode.i_mapping; struct folio *folio; folio = __filemap_get_folio(mapping, index, @@ -216,7 +216,7 @@ void afs_edit_dir_add(struct afs_vnode *vnode, _enter(",,{%d,%s},", name->len, name->name); - i_size = i_size_read(&vnode->vfs_inode); + i_size = i_size_read(&vnode->netfs.inode); if (i_size > AFS_DIR_BLOCK_SIZE * AFS_DIR_MAX_BLOCKS || (i_size & (AFS_DIR_BLOCK_SIZE - 1))) { clear_bit(AFS_VNODE_DIR_VALID, &vnode->flags); @@ -336,7 +336,7 @@ found_space: if (b < AFS_DIR_BLOCKS_WITH_CTR) meta->meta.alloc_ctrs[b] -= need_slots; - inode_inc_iversion_raw(&vnode->vfs_inode); + inode_inc_iversion_raw(&vnode->netfs.inode); afs_stat_v(vnode, n_dir_cr); _debug("Insert %s in %u[%u]", name->name, b, slot); @@ -383,7 +383,7 @@ void afs_edit_dir_remove(struct afs_vnode *vnode, _enter(",,{%d,%s},", name->len, name->name); - i_size = i_size_read(&vnode->vfs_inode); + i_size = i_size_read(&vnode->netfs.inode); if (i_size < AFS_DIR_BLOCK_SIZE || i_size > AFS_DIR_BLOCK_SIZE * AFS_DIR_MAX_BLOCKS || (i_size & (AFS_DIR_BLOCK_SIZE - 1))) { @@ -463,7 +463,7 @@ found_dirent: if (b < AFS_DIR_BLOCKS_WITH_CTR) meta->meta.alloc_ctrs[b] += need_slots; - inode_set_iversion_raw(&vnode->vfs_inode, vnode->status.data_version); + inode_set_iversion_raw(&vnode->netfs.inode, vnode->status.data_version); afs_stat_v(vnode, n_dir_rm); _debug("Remove %s from %u[%u]", name->name, b, slot); diff --git a/fs/afs/dir_silly.c b/fs/afs/dir_silly.c index 45cfd50a9521..bb5807e87fa4 100644 --- a/fs/afs/dir_silly.c +++ b/fs/afs/dir_silly.c @@ -131,7 +131,7 @@ int afs_sillyrename(struct afs_vnode *dvnode, struct afs_vnode *vnode, goto out; } while (!d_is_negative(sdentry)); - ihold(&vnode->vfs_inode); + ihold(&vnode->netfs.inode); ret = afs_do_silly_rename(dvnode, vnode, dentry, sdentry, key); switch (ret) { @@ -148,7 +148,7 @@ int afs_sillyrename(struct afs_vnode *dvnode, struct afs_vnode *vnode, d_drop(sdentry); } - iput(&vnode->vfs_inode); + iput(&vnode->netfs.inode); dput(sdentry); out: _leave(" = %d", ret); diff --git a/fs/afs/dynroot.c b/fs/afs/dynroot.c index f120bcb8bf73..d7d9402ff718 100644 --- a/fs/afs/dynroot.c +++ b/fs/afs/dynroot.c @@ -76,7 +76,7 @@ struct inode *afs_iget_pseudo_dir(struct super_block *sb, bool root) /* there shouldn't be an existing inode */ BUG_ON(!(inode->i_state & I_NEW)); - netfs_i_context_init(inode, NULL); + netfs_inode_init(&vnode->netfs, NULL); inode->i_size = 0; inode->i_mode = S_IFDIR | S_IRUGO | S_IXUGO; if (root) { diff --git a/fs/afs/file.c b/fs/afs/file.c index a8e8832179e4..42118a4f3383 100644 --- a/fs/afs/file.c +++ b/fs/afs/file.c @@ -194,7 +194,7 @@ int afs_release(struct inode *inode, struct file *file) afs_put_wb_key(af->wb); if ((file->f_mode & FMODE_WRITE)) { - i_size = i_size_read(&vnode->vfs_inode); + i_size = i_size_read(&vnode->netfs.inode); afs_set_cache_aux(vnode, &aux); fscache_unuse_cookie(afs_vnode_cache(vnode), &aux, &i_size); } else { @@ -325,7 +325,7 @@ static void afs_issue_read(struct netfs_io_subrequest *subreq) fsreq->iter = &fsreq->def_iter; iov_iter_xarray(&fsreq->def_iter, READ, - &fsreq->vnode->vfs_inode.i_mapping->i_pages, + &fsreq->vnode->netfs.inode.i_mapping->i_pages, fsreq->pos, fsreq->len); afs_fetch_data(fsreq->vnode, fsreq); @@ -382,17 +382,17 @@ static int afs_check_write_begin(struct file *file, loff_t pos, unsigned len, return test_bit(AFS_VNODE_DELETED, &vnode->flags) ? -ESTALE : 0; } -static void afs_priv_cleanup(struct address_space *mapping, void *netfs_priv) +static void afs_free_request(struct netfs_io_request *rreq) { - key_put(netfs_priv); + key_put(rreq->netfs_priv); } const struct netfs_request_ops afs_req_ops = { .init_request = afs_init_request, + .free_request = afs_free_request, .begin_cache_operation = afs_begin_cache_operation, .check_write_begin = afs_check_write_begin, .issue_read = afs_issue_read, - .cleanup = afs_priv_cleanup, }; int afs_write_inode(struct inode *inode, struct writeback_control *wbc) diff --git a/fs/afs/fs_operation.c b/fs/afs/fs_operation.c index d222dfbe976b..7a3803ce3a22 100644 --- a/fs/afs/fs_operation.c +++ b/fs/afs/fs_operation.c @@ -232,14 +232,14 @@ int afs_put_operation(struct afs_operation *op) if (op->file[1].modification && op->file[1].vnode != op->file[0].vnode) clear_bit(AFS_VNODE_MODIFYING, &op->file[1].vnode->flags); if (op->file[0].put_vnode) - iput(&op->file[0].vnode->vfs_inode); + iput(&op->file[0].vnode->netfs.inode); if (op->file[1].put_vnode) - iput(&op->file[1].vnode->vfs_inode); + iput(&op->file[1].vnode->netfs.inode); if (op->more_files) { for (i = 0; i < op->nr_files - 2; i++) if (op->more_files[i].put_vnode) - iput(&op->more_files[i].vnode->vfs_inode); + iput(&op->more_files[i].vnode->netfs.inode); kfree(op->more_files); } diff --git a/fs/afs/inode.c b/fs/afs/inode.c index 30b066299d39..64dab70d4a4f 100644 --- a/fs/afs/inode.c +++ b/fs/afs/inode.c @@ -58,7 +58,7 @@ static noinline void dump_vnode(struct afs_vnode *vnode, struct afs_vnode *paren */ static void afs_set_netfs_context(struct afs_vnode *vnode) { - netfs_i_context_init(&vnode->vfs_inode, &afs_req_ops); + netfs_inode_init(&vnode->netfs, &afs_req_ops); } /* @@ -96,7 +96,7 @@ static int afs_inode_init_from_status(struct afs_operation *op, inode->i_flags |= S_NOATIME; inode->i_uid = make_kuid(&init_user_ns, status->owner); inode->i_gid = make_kgid(&init_user_ns, status->group); - set_nlink(&vnode->vfs_inode, status->nlink); + set_nlink(&vnode->netfs.inode, status->nlink); switch (status->type) { case AFS_FTYPE_FILE: @@ -139,7 +139,7 @@ static int afs_inode_init_from_status(struct afs_operation *op, afs_set_netfs_context(vnode); vnode->invalid_before = status->data_version; - inode_set_iversion_raw(&vnode->vfs_inode, status->data_version); + inode_set_iversion_raw(&vnode->netfs.inode, status->data_version); if (!vp->scb.have_cb) { /* it's a symlink we just created (the fileserver @@ -163,7 +163,7 @@ static void afs_apply_status(struct afs_operation *op, { struct afs_file_status *status = &vp->scb.status; struct afs_vnode *vnode = vp->vnode; - struct inode *inode = &vnode->vfs_inode; + struct inode *inode = &vnode->netfs.inode; struct timespec64 t; umode_t mode; bool data_changed = false; @@ -246,7 +246,7 @@ static void afs_apply_status(struct afs_operation *op, * idea of what the size should be that's not the same as * what's on the server. */ - vnode->netfs_ctx.remote_i_size = status->size; + vnode->netfs.remote_i_size = status->size; if (change_size) { afs_set_i_size(vnode, status->size); inode->i_ctime = t; @@ -289,7 +289,7 @@ void afs_vnode_commit_status(struct afs_operation *op, struct afs_vnode_param *v */ if (vp->scb.status.abort_code == VNOVNODE) { set_bit(AFS_VNODE_DELETED, &vnode->flags); - clear_nlink(&vnode->vfs_inode); + clear_nlink(&vnode->netfs.inode); __afs_break_callback(vnode, afs_cb_break_for_deleted); op->flags &= ~AFS_OPERATION_DIR_CONFLICT; } @@ -306,8 +306,8 @@ void afs_vnode_commit_status(struct afs_operation *op, struct afs_vnode_param *v if (vp->scb.have_cb) afs_apply_callback(op, vp); } else if (vp->op_unlinked && !(op->flags & AFS_OPERATION_DIR_CONFLICT)) { - drop_nlink(&vnode->vfs_inode); - if (vnode->vfs_inode.i_nlink == 0) { + drop_nlink(&vnode->netfs.inode); + if (vnode->netfs.inode.i_nlink == 0) { set_bit(AFS_VNODE_DELETED, &vnode->flags); __afs_break_callback(vnode, afs_cb_break_for_deleted); } @@ -326,7 +326,7 @@ static void afs_fetch_status_success(struct afs_operation *op) struct afs_vnode *vnode = vp->vnode; int ret; - if (vnode->vfs_inode.i_state & I_NEW) { + if (vnode->netfs.inode.i_state & I_NEW) { ret = afs_inode_init_from_status(op, vp, vnode); op->error = ret; if (ret == 0) @@ -430,7 +430,7 @@ static void afs_get_inode_cache(struct afs_vnode *vnode) struct afs_vnode_cache_aux aux; if (vnode->status.type != AFS_FTYPE_FILE) { - vnode->netfs_ctx.cache = NULL; + vnode->netfs.cache = NULL; return; } @@ -457,7 +457,7 @@ static void afs_get_inode_cache(struct afs_vnode *vnode) struct inode *afs_iget(struct afs_operation *op, struct afs_vnode_param *vp) { struct afs_vnode_param *dvp = &op->file[0]; - struct super_block *sb = dvp->vnode->vfs_inode.i_sb; + struct super_block *sb = dvp->vnode->netfs.inode.i_sb; struct afs_vnode *vnode; struct inode *inode; int ret; @@ -582,10 +582,10 @@ static void afs_zap_data(struct afs_vnode *vnode) /* nuke all the non-dirty pages that aren't locked, mapped or being * written back in a regular file and completely discard the pages in a * directory or symlink */ - if (S_ISREG(vnode->vfs_inode.i_mode)) - invalidate_remote_inode(&vnode->vfs_inode); + if (S_ISREG(vnode->netfs.inode.i_mode)) + invalidate_remote_inode(&vnode->netfs.inode); else - invalidate_inode_pages2(vnode->vfs_inode.i_mapping); + invalidate_inode_pages2(vnode->netfs.inode.i_mapping); } /* @@ -683,8 +683,8 @@ int afs_validate(struct afs_vnode *vnode, struct key *key) key_serial(key)); if (unlikely(test_bit(AFS_VNODE_DELETED, &vnode->flags))) { - if (vnode->vfs_inode.i_nlink) - clear_nlink(&vnode->vfs_inode); + if (vnode->netfs.inode.i_nlink) + clear_nlink(&vnode->netfs.inode); goto valid; } @@ -745,7 +745,8 @@ int afs_getattr(struct user_namespace *mnt_userns, const struct path *path, _enter("{ ino=%lu v=%u }", inode->i_ino, inode->i_generation); - if (!(query_flags & AT_STATX_DONT_SYNC) && + if (vnode->volume && + !(query_flags & AT_STATX_DONT_SYNC) && !test_bit(AFS_VNODE_CB_PROMISED, &vnode->flags)) { key = afs_request_key(vnode->volume->cell); if (IS_ERR(key)) @@ -826,7 +827,7 @@ void afs_evict_inode(struct inode *inode) static void afs_setattr_success(struct afs_operation *op) { struct afs_vnode_param *vp = &op->file[0]; - struct inode *inode = &vp->vnode->vfs_inode; + struct inode *inode = &vp->vnode->netfs.inode; loff_t old_i_size = i_size_read(inode); op->setattr.old_i_size = old_i_size; @@ -843,7 +844,7 @@ static void afs_setattr_success(struct afs_operation *op) static void afs_setattr_edit_file(struct afs_operation *op) { struct afs_vnode_param *vp = &op->file[0]; - struct inode *inode = &vp->vnode->vfs_inode; + struct inode *inode = &vp->vnode->netfs.inode; if (op->setattr.attr->ia_valid & ATTR_SIZE) { loff_t size = op->setattr.attr->ia_size; @@ -875,7 +876,7 @@ int afs_setattr(struct user_namespace *mnt_userns, struct dentry *dentry, ATTR_MTIME | ATTR_MTIME_SET | ATTR_TIMES_SET | ATTR_TOUCH; struct afs_operation *op; struct afs_vnode *vnode = AFS_FS_I(d_inode(dentry)); - struct inode *inode = &vnode->vfs_inode; + struct inode *inode = &vnode->netfs.inode; loff_t i_size; int ret; diff --git a/fs/afs/internal.h b/fs/afs/internal.h index a30995901266..a6f25d9e75b5 100644 --- a/fs/afs/internal.h +++ b/fs/afs/internal.h @@ -619,12 +619,7 @@ enum afs_lock_state { * leak from one inode to another. */ struct afs_vnode { - struct { - /* These must be contiguous */ - struct inode vfs_inode; /* the VFS's inode record */ - struct netfs_i_context netfs_ctx; /* Netfslib context */ - }; - + struct netfs_inode netfs; /* Netfslib context and vfs inode */ struct afs_volume *volume; /* volume on which vnode resides */ struct afs_fid fid; /* the file identifier for this inode */ struct afs_file_status status; /* AFS status info for this file */ @@ -675,7 +670,7 @@ struct afs_vnode { static inline struct fscache_cookie *afs_vnode_cache(struct afs_vnode *vnode) { #ifdef CONFIG_AFS_FSCACHE - return netfs_i_cookie(&vnode->vfs_inode); + return netfs_i_cookie(&vnode->netfs); #else return NULL; #endif @@ -685,7 +680,7 @@ static inline void afs_vnode_set_cache(struct afs_vnode *vnode, struct fscache_cookie *cookie) { #ifdef CONFIG_AFS_FSCACHE - vnode->netfs_ctx.cache = cookie; + vnode->netfs.cache = cookie; #endif } @@ -892,7 +887,7 @@ static inline void afs_invalidate_cache(struct afs_vnode *vnode, unsigned int fl afs_set_cache_aux(vnode, &aux); fscache_invalidate(afs_vnode_cache(vnode), &aux, - i_size_read(&vnode->vfs_inode), flags); + i_size_read(&vnode->netfs.inode), flags); } /* @@ -1217,7 +1212,7 @@ static inline struct afs_net *afs_i2net(struct inode *inode) static inline struct afs_net *afs_v2net(struct afs_vnode *vnode) { - return afs_i2net(&vnode->vfs_inode); + return afs_i2net(&vnode->netfs.inode); } static inline struct afs_net *afs_sock2net(struct sock *sk) @@ -1593,12 +1588,12 @@ extern void yfs_fs_store_opaque_acl2(struct afs_operation *); */ static inline struct afs_vnode *AFS_FS_I(struct inode *inode) { - return container_of(inode, struct afs_vnode, vfs_inode); + return container_of(inode, struct afs_vnode, netfs.inode); } static inline struct inode *AFS_VNODE_TO_I(struct afs_vnode *vnode) { - return &vnode->vfs_inode; + return &vnode->netfs.inode; } /* @@ -1621,8 +1616,8 @@ static inline void afs_update_dentry_version(struct afs_operation *op, */ static inline void afs_set_i_size(struct afs_vnode *vnode, u64 size) { - i_size_write(&vnode->vfs_inode, size); - vnode->vfs_inode.i_blocks = ((size + 1023) >> 10) << 1; + i_size_write(&vnode->netfs.inode, size); + vnode->netfs.inode.i_blocks = ((size + 1023) >> 10) << 1; } /* diff --git a/fs/afs/super.c b/fs/afs/super.c index 1fea195b0b27..95d713074dc8 100644 --- a/fs/afs/super.c +++ b/fs/afs/super.c @@ -659,7 +659,7 @@ static void afs_i_init_once(void *_vnode) struct afs_vnode *vnode = _vnode; memset(vnode, 0, sizeof(*vnode)); - inode_init_once(&vnode->vfs_inode); + inode_init_once(&vnode->netfs.inode); mutex_init(&vnode->io_lock); init_rwsem(&vnode->validate_lock); spin_lock_init(&vnode->wb_lock); @@ -700,8 +700,8 @@ static struct inode *afs_alloc_inode(struct super_block *sb) init_rwsem(&vnode->rmdir_lock); INIT_WORK(&vnode->cb_work, afs_invalidate_mmap_work); - _leave(" = %p", &vnode->vfs_inode); - return &vnode->vfs_inode; + _leave(" = %p", &vnode->netfs.inode); + return &vnode->netfs.inode; } static void afs_free_inode(struct inode *inode) diff --git a/fs/afs/volume.c b/fs/afs/volume.c index 94a3d247924b..cc665cef0abe 100644 --- a/fs/afs/volume.c +++ b/fs/afs/volume.c @@ -9,8 +9,7 @@ #include <linux/slab.h> #include "internal.h" -unsigned __read_mostly afs_volume_gc_delay = 10; -unsigned __read_mostly afs_volume_record_life = 60 * 60; +static unsigned __read_mostly afs_volume_record_life = 60 * 60; /* * Insert a volume into a cell. If there's an existing volume record, that is diff --git a/fs/afs/write.c b/fs/afs/write.c index 2236b2165e37..2c885b22de34 100644 --- a/fs/afs/write.c +++ b/fs/afs/write.c @@ -60,7 +60,7 @@ int afs_write_begin(struct file *file, struct address_space *mapping, * file. We need to do this before we get a lock on the page in case * there's more than one writer competing for the same cache block. */ - ret = netfs_write_begin(file, mapping, pos, len, &folio, fsdata); + ret = netfs_write_begin(&vnode->netfs, file, mapping, pos, len, &folio, fsdata); if (ret < 0) return ret; @@ -146,10 +146,10 @@ int afs_write_end(struct file *file, struct address_space *mapping, write_end_pos = pos + copied; - i_size = i_size_read(&vnode->vfs_inode); + i_size = i_size_read(&vnode->netfs.inode); if (write_end_pos > i_size) { write_seqlock(&vnode->cb_lock); - i_size = i_size_read(&vnode->vfs_inode); + i_size = i_size_read(&vnode->netfs.inode); if (write_end_pos > i_size) afs_set_i_size(vnode, write_end_pos); write_sequnlock(&vnode->cb_lock); @@ -257,7 +257,7 @@ static void afs_redirty_pages(struct writeback_control *wbc, */ static void afs_pages_written_back(struct afs_vnode *vnode, loff_t start, unsigned int len) { - struct address_space *mapping = vnode->vfs_inode.i_mapping; + struct address_space *mapping = vnode->netfs.inode.i_mapping; struct folio *folio; pgoff_t end; @@ -354,7 +354,6 @@ static const struct afs_operation_ops afs_store_data_operation = { static int afs_store_data(struct afs_vnode *vnode, struct iov_iter *iter, loff_t pos, bool laundering) { - struct netfs_i_context *ictx = &vnode->netfs_ctx; struct afs_operation *op; struct afs_wb_key *wbk = NULL; loff_t size = iov_iter_count(iter); @@ -385,9 +384,9 @@ static int afs_store_data(struct afs_vnode *vnode, struct iov_iter *iter, loff_t op->store.write_iter = iter; op->store.pos = pos; op->store.size = size; - op->store.i_size = max(pos + size, ictx->remote_i_size); + op->store.i_size = max(pos + size, vnode->netfs.remote_i_size); op->store.laundering = laundering; - op->mtime = vnode->vfs_inode.i_mtime; + op->mtime = vnode->netfs.inode.i_mtime; op->flags |= AFS_OPERATION_UNINTR; op->ops = &afs_store_data_operation; @@ -554,7 +553,7 @@ static ssize_t afs_write_back_from_locked_folio(struct address_space *mapping, struct iov_iter iter; unsigned long priv; unsigned int offset, to, len, max_len; - loff_t i_size = i_size_read(&vnode->vfs_inode); + loff_t i_size = i_size_read(&vnode->netfs.inode); bool new_content = test_bit(AFS_VNODE_NEW_CONTENT, &vnode->flags); bool caching = fscache_cookie_enabled(afs_vnode_cache(vnode)); long count = wbc->nr_to_write; @@ -845,7 +844,7 @@ ssize_t afs_file_write(struct kiocb *iocb, struct iov_iter *from) _enter("{%llx:%llu},{%zu},", vnode->fid.vid, vnode->fid.vnode, count); - if (IS_SWAPFILE(&vnode->vfs_inode)) { + if (IS_SWAPFILE(&vnode->netfs.inode)) { printk(KERN_INFO "AFS: Attempt to write to active swap file!\n"); return -EBUSY; @@ -958,8 +957,8 @@ void afs_prune_wb_keys(struct afs_vnode *vnode) /* Discard unused keys */ spin_lock(&vnode->wb_lock); - if (!mapping_tagged(&vnode->vfs_inode.i_data, PAGECACHE_TAG_WRITEBACK) && - !mapping_tagged(&vnode->vfs_inode.i_data, PAGECACHE_TAG_DIRTY)) { + if (!mapping_tagged(&vnode->netfs.inode.i_data, PAGECACHE_TAG_WRITEBACK) && + !mapping_tagged(&vnode->netfs.inode.i_data, PAGECACHE_TAG_DIRTY)) { list_for_each_entry_safe(wbk, tmp, &vnode->wb_keys, vnode_link) { if (refcount_read(&wbk->usage) == 1) list_move(&wbk->vnode_link, &graveyard); @@ -1034,6 +1033,6 @@ static void afs_write_to_cache(struct afs_vnode *vnode, bool caching) { fscache_write_to_cache(afs_vnode_cache(vnode), - vnode->vfs_inode.i_mapping, start, len, i_size, + vnode->netfs.inode.i_mapping, start, len, i_size, afs_write_to_cache_done, vnode, caching); } diff --git a/fs/attr.c b/fs/attr.c index 66899b6e9bd8..dbe996b0dedf 100644 --- a/fs/attr.c +++ b/fs/attr.c @@ -61,9 +61,15 @@ static bool chgrp_ok(struct user_namespace *mnt_userns, const struct inode *inode, kgid_t gid) { kgid_t kgid = i_gid_into_mnt(mnt_userns, inode); - if (uid_eq(current_fsuid(), i_uid_into_mnt(mnt_userns, inode)) && - (in_group_p(gid) || gid_eq(gid, inode->i_gid))) - return true; + if (uid_eq(current_fsuid(), i_uid_into_mnt(mnt_userns, inode))) { + kgid_t mapped_gid; + + if (gid_eq(gid, inode->i_gid)) + return true; + mapped_gid = mapped_kgid_fs(mnt_userns, i_user_ns(inode), gid); + if (in_group_p(mapped_gid)) + return true; + } if (capable_wrt_inode_uidgid(mnt_userns, inode, CAP_CHOWN)) return true; if (gid_eq(kgid, INVALID_GID) && @@ -123,12 +129,20 @@ int setattr_prepare(struct user_namespace *mnt_userns, struct dentry *dentry, /* Make sure a caller can chmod. */ if (ia_valid & ATTR_MODE) { + kgid_t mapped_gid; + if (!inode_owner_or_capable(mnt_userns, inode)) return -EPERM; + + if (ia_valid & ATTR_GID) + mapped_gid = mapped_kgid_fs(mnt_userns, + i_user_ns(inode), attr->ia_gid); + else + mapped_gid = i_gid_into_mnt(mnt_userns, inode); + /* Also check the setgid bit! */ - if (!in_group_p((ia_valid & ATTR_GID) ? attr->ia_gid : - i_gid_into_mnt(mnt_userns, inode)) && - !capable_wrt_inode_uidgid(mnt_userns, inode, CAP_FSETID)) + if (!in_group_p(mapped_gid) && + !capable_wrt_inode_uidgid(mnt_userns, inode, CAP_FSETID)) attr->ia_mode &= ~S_ISGID; } diff --git a/fs/btrfs/block-group.h b/fs/btrfs/block-group.h index 3ac668ace50a..35e0e860cc0b 100644 --- a/fs/btrfs/block-group.h +++ b/fs/btrfs/block-group.h @@ -104,6 +104,7 @@ struct btrfs_block_group { unsigned int relocating_repair:1; unsigned int chunk_item_inserted:1; unsigned int zone_is_active:1; + unsigned int zoned_data_reloc_ongoing:1; int disk_cache_state; diff --git a/fs/btrfs/ctree.h b/fs/btrfs/ctree.h index 0e49b1a0c071..415bf1823fb3 100644 --- a/fs/btrfs/ctree.h +++ b/fs/btrfs/ctree.h @@ -1330,6 +1330,8 @@ struct btrfs_replace_extent_info { * existing extent into a file range. */ bool is_new_extent; + /* Indicate if we should update the inode's mtime and ctime. */ + bool update_times; /* Meaningful only if is_new_extent is true. */ int qgroup_reserved; /* diff --git a/fs/btrfs/disk-io.c b/fs/btrfs/disk-io.c index 89e94ea2fef5..4ba005c41983 100644 --- a/fs/btrfs/disk-io.c +++ b/fs/btrfs/disk-io.c @@ -4632,6 +4632,17 @@ void __cold close_ctree(struct btrfs_fs_info *fs_info) int ret; set_bit(BTRFS_FS_CLOSING_START, &fs_info->flags); + + /* + * We may have the reclaim task running and relocating a data block group, + * in which case it may create delayed iputs. So stop it before we park + * the cleaner kthread otherwise we can get new delayed iputs after + * parking the cleaner, and that can make the async reclaim task to hang + * if it's waiting for delayed iputs to complete, since the cleaner is + * parked and can not run delayed iputs - this will make us hang when + * trying to stop the async reclaim task. + */ + cancel_work_sync(&fs_info->reclaim_bgs_work); /* * We don't want the cleaner to start new transactions, add more delayed * iputs, etc. while we're closing. We can't use kthread_stop() yet @@ -4672,8 +4683,6 @@ void __cold close_ctree(struct btrfs_fs_info *fs_info) cancel_work_sync(&fs_info->async_data_reclaim_work); cancel_work_sync(&fs_info->preempt_reclaim_work); - cancel_work_sync(&fs_info->reclaim_bgs_work); - /* Cancel or finish ongoing discard work */ btrfs_discard_cleanup(fs_info); diff --git a/fs/btrfs/extent-tree.c b/fs/btrfs/extent-tree.c index 0867c5cd6e01..4157ecc27d4b 100644 --- a/fs/btrfs/extent-tree.c +++ b/fs/btrfs/extent-tree.c @@ -3832,7 +3832,7 @@ static int do_allocation_zoned(struct btrfs_block_group *block_group, block_group->start == fs_info->data_reloc_bg || fs_info->data_reloc_bg == 0); - if (block_group->ro) { + if (block_group->ro || block_group->zoned_data_reloc_ongoing) { ret = 1; goto out; } @@ -3894,8 +3894,24 @@ static int do_allocation_zoned(struct btrfs_block_group *block_group, out: if (ret && ffe_ctl->for_treelog) fs_info->treelog_bg = 0; - if (ret && ffe_ctl->for_data_reloc) + if (ret && ffe_ctl->for_data_reloc && + fs_info->data_reloc_bg == block_group->start) { + /* + * Do not allow further allocations from this block group. + * Compared to increasing the ->ro, setting the + * ->zoned_data_reloc_ongoing flag still allows nocow + * writers to come in. See btrfs_inc_nocow_writers(). + * + * We need to disable an allocation to avoid an allocation of + * regular (non-relocation data) extent. With mix of relocation + * extents and regular extents, we can dispatch WRITE commands + * (for relocation extents) and ZONE APPEND commands (for + * regular extents) at the same time to the same zone, which + * easily break the write pointer. + */ + block_group->zoned_data_reloc_ongoing = 1; fs_info->data_reloc_bg = 0; + } spin_unlock(&fs_info->relocation_bg_lock); spin_unlock(&fs_info->treelog_bg_lock); spin_unlock(&block_group->lock); diff --git a/fs/btrfs/extent_io.c b/fs/btrfs/extent_io.c index 8f6b544ae616..04e36343da3a 100644 --- a/fs/btrfs/extent_io.c +++ b/fs/btrfs/extent_io.c @@ -5241,13 +5241,14 @@ int extent_writepages(struct address_space *mapping, */ btrfs_zoned_data_reloc_lock(BTRFS_I(inode)); ret = extent_write_cache_pages(mapping, wbc, &epd); - btrfs_zoned_data_reloc_unlock(BTRFS_I(inode)); ASSERT(ret <= 0); if (ret < 0) { + btrfs_zoned_data_reloc_unlock(BTRFS_I(inode)); end_write_bio(&epd, ret); return ret; } flush_write_bio(&epd); + btrfs_zoned_data_reloc_unlock(BTRFS_I(inode)); return ret; } diff --git a/fs/btrfs/file.c b/fs/btrfs/file.c index 1fd827b99c1b..9dfde1af8a64 100644 --- a/fs/btrfs/file.c +++ b/fs/btrfs/file.c @@ -2323,25 +2323,62 @@ int btrfs_sync_file(struct file *file, loff_t start, loff_t end, int datasync) */ btrfs_inode_unlock(inode, BTRFS_ILOCK_MMAP); - if (ret != BTRFS_NO_LOG_SYNC) { + if (ret == BTRFS_NO_LOG_SYNC) { + ret = btrfs_end_transaction(trans); + goto out; + } + + /* We successfully logged the inode, attempt to sync the log. */ + if (!ret) { + ret = btrfs_sync_log(trans, root, &ctx); if (!ret) { - ret = btrfs_sync_log(trans, root, &ctx); - if (!ret) { - ret = btrfs_end_transaction(trans); - goto out; - } - } - if (!full_sync) { - ret = btrfs_wait_ordered_range(inode, start, len); - if (ret) { - btrfs_end_transaction(trans); - goto out; - } + ret = btrfs_end_transaction(trans); + goto out; } - ret = btrfs_commit_transaction(trans); - } else { + } + + /* + * At this point we need to commit the transaction because we had + * btrfs_need_log_full_commit() or some other error. + * + * If we didn't do a full sync we have to stop the trans handle, wait on + * the ordered extents, start it again and commit the transaction. If + * we attempt to wait on the ordered extents here we could deadlock with + * something like fallocate() that is holding the extent lock trying to + * start a transaction while some other thread is trying to commit the + * transaction while we (fsync) are currently holding the transaction + * open. + */ + if (!full_sync) { ret = btrfs_end_transaction(trans); + if (ret) + goto out; + ret = btrfs_wait_ordered_range(inode, start, len); + if (ret) + goto out; + + /* + * This is safe to use here because we're only interested in + * making sure the transaction that had the ordered extents is + * committed. We aren't waiting on anything past this point, + * we're purely getting the transaction and committing it. + */ + trans = btrfs_attach_transaction_barrier(root); + if (IS_ERR(trans)) { + ret = PTR_ERR(trans); + + /* + * We committed the transaction and there's no currently + * running transaction, this means everything we care + * about made it to disk and we are done. + */ + if (ret == -ENOENT) + ret = 0; + goto out; + } } + + ret = btrfs_commit_transaction(trans); out: ASSERT(list_empty(&ctx.list)); err = file_check_and_advance_wb_err(file); @@ -2719,7 +2756,8 @@ int btrfs_replace_file_extents(struct btrfs_inode *inode, ret = btrfs_block_rsv_migrate(&fs_info->trans_block_rsv, rsv, min_size, false); - BUG_ON(ret); + if (WARN_ON(ret)) + goto out_trans; trans->block_rsv = rsv; cur_offset = start; @@ -2803,6 +2841,25 @@ int btrfs_replace_file_extents(struct btrfs_inode *inode, extent_info->file_offset += replace_len; } + /* + * We are releasing our handle on the transaction, balance the + * dirty pages of the btree inode and flush delayed items, and + * then get a new transaction handle, which may now point to a + * new transaction in case someone else may have committed the + * transaction we used to replace/drop file extent items. So + * bump the inode's iversion and update mtime and ctime except + * if we are called from a dedupe context. This is because a + * power failure/crash may happen after the transaction is + * committed and before we finish replacing/dropping all the + * file extent items we need. + */ + inode_inc_iversion(&inode->vfs_inode); + + if (!extent_info || extent_info->update_times) { + inode->vfs_inode.i_mtime = current_time(&inode->vfs_inode); + inode->vfs_inode.i_ctime = inode->vfs_inode.i_mtime; + } + ret = btrfs_update_inode(trans, root, inode); if (ret) break; @@ -2819,7 +2876,8 @@ int btrfs_replace_file_extents(struct btrfs_inode *inode, ret = btrfs_block_rsv_migrate(&fs_info->trans_block_rsv, rsv, min_size, false); - BUG_ON(ret); /* shouldn't happen */ + if (WARN_ON(ret)) + break; trans->block_rsv = rsv; cur_offset = drop_args.drop_end; diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c index 81737eff92f3..05e0c4a5affd 100644 --- a/fs/btrfs/inode.c +++ b/fs/btrfs/inode.c @@ -3195,6 +3195,8 @@ static int btrfs_finish_ordered_io(struct btrfs_ordered_extent *ordered_extent) ordered_extent->file_offset, ordered_extent->file_offset + logical_len); + btrfs_zoned_release_data_reloc_bg(fs_info, ordered_extent->disk_bytenr, + ordered_extent->disk_num_bytes); } else { BUG_ON(root == fs_info->tree_root); ret = insert_ordered_extent_file_extent(trans, ordered_extent); @@ -9897,6 +9899,7 @@ static struct btrfs_trans_handle *insert_prealloc_file_extent( extent_info.file_offset = file_offset; extent_info.extent_buf = (char *)&stack_fi; extent_info.is_new_extent = true; + extent_info.update_times = true; extent_info.qgroup_reserved = qgroup_released; extent_info.insertions = 0; diff --git a/fs/btrfs/locking.c b/fs/btrfs/locking.c index 313d9d685adb..33461b4f9c8b 100644 --- a/fs/btrfs/locking.c +++ b/fs/btrfs/locking.c @@ -45,7 +45,6 @@ void __btrfs_tree_read_lock(struct extent_buffer *eb, enum btrfs_lock_nesting ne start_ns = ktime_get_ns(); down_read_nested(&eb->lock, nest); - eb->lock_owner = current->pid; trace_btrfs_tree_read_lock(eb, start_ns); } @@ -62,7 +61,6 @@ void btrfs_tree_read_lock(struct extent_buffer *eb) int btrfs_try_tree_read_lock(struct extent_buffer *eb) { if (down_read_trylock(&eb->lock)) { - eb->lock_owner = current->pid; trace_btrfs_try_tree_read_lock(eb); return 1; } @@ -90,7 +88,6 @@ int btrfs_try_tree_write_lock(struct extent_buffer *eb) void btrfs_tree_read_unlock(struct extent_buffer *eb) { trace_btrfs_tree_read_unlock(eb); - eb->lock_owner = 0; up_read(&eb->lock); } diff --git a/fs/btrfs/reflink.c b/fs/btrfs/reflink.c index c39f8b3a5a4a..a3549d587464 100644 --- a/fs/btrfs/reflink.c +++ b/fs/btrfs/reflink.c @@ -344,6 +344,7 @@ static int btrfs_clone(struct inode *src, struct inode *inode, int ret; const u64 len = olen_aligned; u64 last_dest_end = destoff; + u64 prev_extent_end = off; ret = -ENOMEM; buf = kvmalloc(fs_info->nodesize, GFP_KERNEL); @@ -363,7 +364,6 @@ static int btrfs_clone(struct inode *src, struct inode *inode, key.offset = off; while (1) { - u64 next_key_min_offset = key.offset + 1; struct btrfs_file_extent_item *extent; u64 extent_gen; int type; @@ -431,14 +431,21 @@ process_slot: * The first search might have left us at an extent item that * ends before our target range's start, can happen if we have * holes and NO_HOLES feature enabled. + * + * Subsequent searches may leave us on a file range we have + * processed before - this happens due to a race with ordered + * extent completion for a file range that is outside our source + * range, but that range was part of a file extent item that + * also covered a leading part of our source range. */ - if (key.offset + datal <= off) { + if (key.offset + datal <= prev_extent_end) { path->slots[0]++; goto process_slot; } else if (key.offset >= off + len) { break; } - next_key_min_offset = key.offset + datal; + + prev_extent_end = key.offset + datal; size = btrfs_item_size(leaf, slot); read_extent_buffer(leaf, buf, btrfs_item_ptr_offset(leaf, slot), size); @@ -489,6 +496,7 @@ process_slot: clone_info.file_offset = new_key.offset; clone_info.extent_buf = buf; clone_info.is_new_extent = false; + clone_info.update_times = !no_time_update; ret = btrfs_replace_file_extents(BTRFS_I(inode), path, drop_start, new_key.offset + datal - 1, &clone_info, &trans); @@ -550,7 +558,7 @@ process_slot: break; btrfs_release_path(path); - key.offset = next_key_min_offset; + key.offset = prev_extent_end; if (fatal_signal_pending(current)) { ret = -EINTR; diff --git a/fs/btrfs/super.c b/fs/btrfs/super.c index b1fdc6a26c76..6627dd7875ee 100644 --- a/fs/btrfs/super.c +++ b/fs/btrfs/super.c @@ -763,6 +763,8 @@ int btrfs_parse_options(struct btrfs_fs_info *info, char *options, compress_force = false; no_compress++; } else { + btrfs_err(info, "unrecognized compression value %s", + args[0].from); ret = -EINVAL; goto out; } @@ -821,8 +823,11 @@ int btrfs_parse_options(struct btrfs_fs_info *info, char *options, case Opt_thread_pool: ret = match_int(&args[0], &intarg); if (ret) { + btrfs_err(info, "unrecognized thread_pool value %s", + args[0].from); goto out; } else if (intarg == 0) { + btrfs_err(info, "invalid value 0 for thread_pool"); ret = -EINVAL; goto out; } @@ -883,8 +888,11 @@ int btrfs_parse_options(struct btrfs_fs_info *info, char *options, break; case Opt_ratio: ret = match_int(&args[0], &intarg); - if (ret) + if (ret) { + btrfs_err(info, "unrecognized metadata_ratio value %s", + args[0].from); goto out; + } info->metadata_ratio = intarg; btrfs_info(info, "metadata ratio %u", info->metadata_ratio); @@ -901,6 +909,8 @@ int btrfs_parse_options(struct btrfs_fs_info *info, char *options, btrfs_set_and_info(info, DISCARD_ASYNC, "turning on async discard"); } else { + btrfs_err(info, "unrecognized discard mode value %s", + args[0].from); ret = -EINVAL; goto out; } @@ -933,6 +943,8 @@ int btrfs_parse_options(struct btrfs_fs_info *info, char *options, btrfs_set_and_info(info, FREE_SPACE_TREE, "enabling free space tree"); } else { + btrfs_err(info, "unrecognized space_cache value %s", + args[0].from); ret = -EINVAL; goto out; } @@ -1014,8 +1026,12 @@ int btrfs_parse_options(struct btrfs_fs_info *info, char *options, break; case Opt_check_integrity_print_mask: ret = match_int(&args[0], &intarg); - if (ret) + if (ret) { + btrfs_err(info, + "unrecognized check_integrity_print_mask value %s", + args[0].from); goto out; + } info->check_integrity_print_mask = intarg; btrfs_info(info, "check_integrity_print_mask 0x%x", info->check_integrity_print_mask); @@ -1030,13 +1046,15 @@ int btrfs_parse_options(struct btrfs_fs_info *info, char *options, goto out; #endif case Opt_fatal_errors: - if (strcmp(args[0].from, "panic") == 0) + if (strcmp(args[0].from, "panic") == 0) { btrfs_set_opt(info->mount_opt, PANIC_ON_FATAL_ERROR); - else if (strcmp(args[0].from, "bug") == 0) + } else if (strcmp(args[0].from, "bug") == 0) { btrfs_clear_opt(info->mount_opt, PANIC_ON_FATAL_ERROR); - else { + } else { + btrfs_err(info, "unrecognized fatal_errors value %s", + args[0].from); ret = -EINVAL; goto out; } @@ -1044,8 +1062,12 @@ int btrfs_parse_options(struct btrfs_fs_info *info, char *options, case Opt_commit_interval: intarg = 0; ret = match_int(&args[0], &intarg); - if (ret) + if (ret) { + btrfs_err(info, "unrecognized commit_interval value %s", + args[0].from); + ret = -EINVAL; goto out; + } if (intarg == 0) { btrfs_info(info, "using default commit interval %us", @@ -1059,8 +1081,11 @@ int btrfs_parse_options(struct btrfs_fs_info *info, char *options, break; case Opt_rescue: ret = parse_rescue_options(info, args[0].from); - if (ret < 0) + if (ret < 0) { + btrfs_err(info, "unrecognized rescue value %s", + args[0].from); goto out; + } break; #ifdef CONFIG_BTRFS_DEBUG case Opt_fragment_all: @@ -1985,6 +2010,14 @@ static int btrfs_remount(struct super_block *sb, int *flags, char *data) if (ret) goto restore; + /* V1 cache is not supported for subpage mount. */ + if (fs_info->sectorsize < PAGE_SIZE && btrfs_test_opt(fs_info, SPACE_CACHE)) { + btrfs_warn(fs_info, + "v1 space cache is not supported for page size %lu with sectorsize %u", + PAGE_SIZE, fs_info->sectorsize); + ret = -EINVAL; + goto restore; + } btrfs_remount_begin(fs_info, old_opts, *flags); btrfs_resize_thread_pool(fs_info, fs_info->thread_pool_size, old_thread_pool_size); diff --git a/fs/btrfs/zoned.c b/fs/btrfs/zoned.c index 11237a913bee..79e8c8cd75ed 100644 --- a/fs/btrfs/zoned.c +++ b/fs/btrfs/zoned.c @@ -2139,3 +2139,30 @@ bool btrfs_zoned_should_reclaim(struct btrfs_fs_info *fs_info) factor = div64_u64(used * 100, total); return factor >= fs_info->bg_reclaim_threshold; } + +void btrfs_zoned_release_data_reloc_bg(struct btrfs_fs_info *fs_info, u64 logical, + u64 length) +{ + struct btrfs_block_group *block_group; + + if (!btrfs_is_zoned(fs_info)) + return; + + block_group = btrfs_lookup_block_group(fs_info, logical); + /* It should be called on a previous data relocation block group. */ + ASSERT(block_group && (block_group->flags & BTRFS_BLOCK_GROUP_DATA)); + + spin_lock(&block_group->lock); + if (!block_group->zoned_data_reloc_ongoing) + goto out; + + /* All relocation extents are written. */ + if (block_group->start + block_group->alloc_offset == logical + length) { + /* Now, release this block group for further allocations. */ + block_group->zoned_data_reloc_ongoing = 0; + } + +out: + spin_unlock(&block_group->lock); + btrfs_put_block_group(block_group); +} diff --git a/fs/btrfs/zoned.h b/fs/btrfs/zoned.h index bb1a189e11f9..6b2eec99162b 100644 --- a/fs/btrfs/zoned.h +++ b/fs/btrfs/zoned.h @@ -77,6 +77,8 @@ void btrfs_schedule_zone_finish_bg(struct btrfs_block_group *bg, void btrfs_clear_data_reloc_bg(struct btrfs_block_group *bg); void btrfs_free_zone_cache(struct btrfs_fs_info *fs_info); bool btrfs_zoned_should_reclaim(struct btrfs_fs_info *fs_info); +void btrfs_zoned_release_data_reloc_bg(struct btrfs_fs_info *fs_info, u64 logical, + u64 length); #else /* CONFIG_BLK_DEV_ZONED */ static inline int btrfs_get_dev_zone(struct btrfs_device *device, u64 pos, struct blk_zone *zone) @@ -243,6 +245,9 @@ static inline bool btrfs_zoned_should_reclaim(struct btrfs_fs_info *fs_info) { return false; } + +static inline void btrfs_zoned_release_data_reloc_bg(struct btrfs_fs_info *fs_info, + u64 logical, u64 length) { } #endif static inline bool btrfs_dev_is_sequential(struct btrfs_device *device, u64 pos) diff --git a/fs/ceph/addr.c b/fs/ceph/addr.c index e5221be6eb55..6dee88815491 100644 --- a/fs/ceph/addr.c +++ b/fs/ceph/addr.c @@ -394,11 +394,10 @@ static int ceph_init_request(struct netfs_io_request *rreq, struct file *file) return 0; } -static void ceph_readahead_cleanup(struct address_space *mapping, void *priv) +static void ceph_netfs_free_request(struct netfs_io_request *rreq) { - struct inode *inode = mapping->host; - struct ceph_inode_info *ci = ceph_inode(inode); - int got = (uintptr_t)priv; + struct ceph_inode_info *ci = ceph_inode(rreq->inode); + int got = (uintptr_t)rreq->netfs_priv; if (got) ceph_put_cap_refs(ci, got); @@ -406,12 +405,12 @@ static void ceph_readahead_cleanup(struct address_space *mapping, void *priv) const struct netfs_request_ops ceph_netfs_ops = { .init_request = ceph_init_request, + .free_request = ceph_netfs_free_request, .begin_cache_operation = ceph_begin_cache_operation, .issue_read = ceph_netfs_issue_read, .expand_readahead = ceph_netfs_expand_readahead, .clamp_length = ceph_netfs_clamp_length, .check_write_begin = ceph_netfs_check_write_begin, - .cleanup = ceph_readahead_cleanup, }; #ifdef CONFIG_CEPH_FSCACHE @@ -1322,10 +1321,11 @@ static int ceph_write_begin(struct file *file, struct address_space *mapping, struct page **pagep, void **fsdata) { struct inode *inode = file_inode(file); + struct ceph_inode_info *ci = ceph_inode(inode); struct folio *folio = NULL; int r; - r = netfs_write_begin(file, inode->i_mapping, pos, len, &folio, NULL); + r = netfs_write_begin(&ci->netfs, file, inode->i_mapping, pos, len, &folio, NULL); if (r == 0) folio_wait_fscache(folio); if (r < 0) { @@ -1798,7 +1798,7 @@ enum { static int __ceph_pool_perm_get(struct ceph_inode_info *ci, s64 pool, struct ceph_string *pool_ns) { - struct ceph_fs_client *fsc = ceph_inode_to_client(&ci->vfs_inode); + struct ceph_fs_client *fsc = ceph_inode_to_client(&ci->netfs.inode); struct ceph_mds_client *mdsc = fsc->mdsc; struct ceph_osd_request *rd_req = NULL, *wr_req = NULL; struct rb_node **p, *parent; @@ -1913,7 +1913,7 @@ static int __ceph_pool_perm_get(struct ceph_inode_info *ci, 0, false, true); err = ceph_osdc_start_request(&fsc->client->osdc, rd_req, false); - wr_req->r_mtime = ci->vfs_inode.i_mtime; + wr_req->r_mtime = ci->netfs.inode.i_mtime; err2 = ceph_osdc_start_request(&fsc->client->osdc, wr_req, false); if (!err) diff --git a/fs/ceph/cache.c b/fs/ceph/cache.c index ddea99922073..177d8e8d73fe 100644 --- a/fs/ceph/cache.c +++ b/fs/ceph/cache.c @@ -29,9 +29,9 @@ void ceph_fscache_register_inode_cookie(struct inode *inode) if (!(inode->i_state & I_NEW)) return; - WARN_ON_ONCE(ci->netfs_ctx.cache); + WARN_ON_ONCE(ci->netfs.cache); - ci->netfs_ctx.cache = + ci->netfs.cache = fscache_acquire_cookie(fsc->fscache, 0, &ci->i_vino, sizeof(ci->i_vino), &ci->i_version, sizeof(ci->i_version), diff --git a/fs/ceph/cache.h b/fs/ceph/cache.h index 7255b790a4c1..dc502daac49a 100644 --- a/fs/ceph/cache.h +++ b/fs/ceph/cache.h @@ -28,7 +28,7 @@ void ceph_fscache_invalidate(struct inode *inode, bool dio_write); static inline struct fscache_cookie *ceph_fscache_cookie(struct ceph_inode_info *ci) { - return netfs_i_cookie(&ci->vfs_inode); + return netfs_i_cookie(&ci->netfs); } static inline void ceph_fscache_resize(struct inode *inode, loff_t to) diff --git a/fs/ceph/caps.c b/fs/ceph/caps.c index bf2e94005598..38c930384d41 100644 --- a/fs/ceph/caps.c +++ b/fs/ceph/caps.c @@ -492,7 +492,7 @@ static void __cap_set_timeouts(struct ceph_mds_client *mdsc, struct ceph_mount_options *opt = mdsc->fsc->mount_options; ci->i_hold_caps_max = round_jiffies(jiffies + opt->caps_wanted_delay_max * HZ); - dout("__cap_set_timeouts %p %lu\n", &ci->vfs_inode, + dout("__cap_set_timeouts %p %lu\n", &ci->netfs.inode, ci->i_hold_caps_max - jiffies); } @@ -507,7 +507,7 @@ static void __cap_set_timeouts(struct ceph_mds_client *mdsc, static void __cap_delay_requeue(struct ceph_mds_client *mdsc, struct ceph_inode_info *ci) { - dout("__cap_delay_requeue %p flags 0x%lx at %lu\n", &ci->vfs_inode, + dout("__cap_delay_requeue %p flags 0x%lx at %lu\n", &ci->netfs.inode, ci->i_ceph_flags, ci->i_hold_caps_max); if (!mdsc->stopping) { spin_lock(&mdsc->cap_delay_lock); @@ -531,7 +531,7 @@ no_change: static void __cap_delay_requeue_front(struct ceph_mds_client *mdsc, struct ceph_inode_info *ci) { - dout("__cap_delay_requeue_front %p\n", &ci->vfs_inode); + dout("__cap_delay_requeue_front %p\n", &ci->netfs.inode); spin_lock(&mdsc->cap_delay_lock); ci->i_ceph_flags |= CEPH_I_FLUSH; if (!list_empty(&ci->i_cap_delay_list)) @@ -548,7 +548,7 @@ static void __cap_delay_requeue_front(struct ceph_mds_client *mdsc, static void __cap_delay_cancel(struct ceph_mds_client *mdsc, struct ceph_inode_info *ci) { - dout("__cap_delay_cancel %p\n", &ci->vfs_inode); + dout("__cap_delay_cancel %p\n", &ci->netfs.inode); if (list_empty(&ci->i_cap_delay_list)) return; spin_lock(&mdsc->cap_delay_lock); @@ -568,7 +568,7 @@ static void __check_cap_issue(struct ceph_inode_info *ci, struct ceph_cap *cap, * Each time we receive FILE_CACHE anew, we increment * i_rdcache_gen. */ - if (S_ISREG(ci->vfs_inode.i_mode) && + if (S_ISREG(ci->netfs.inode.i_mode) && (issued & (CEPH_CAP_FILE_CACHE|CEPH_CAP_FILE_LAZYIO)) && (had & (CEPH_CAP_FILE_CACHE|CEPH_CAP_FILE_LAZYIO)) == 0) { ci->i_rdcache_gen++; @@ -583,14 +583,14 @@ static void __check_cap_issue(struct ceph_inode_info *ci, struct ceph_cap *cap, if ((issued & CEPH_CAP_FILE_SHARED) != (had & CEPH_CAP_FILE_SHARED)) { if (issued & CEPH_CAP_FILE_SHARED) atomic_inc(&ci->i_shared_gen); - if (S_ISDIR(ci->vfs_inode.i_mode)) { - dout(" marking %p NOT complete\n", &ci->vfs_inode); + if (S_ISDIR(ci->netfs.inode.i_mode)) { + dout(" marking %p NOT complete\n", &ci->netfs.inode); __ceph_dir_clear_complete(ci); } } /* Wipe saved layout if we're losing DIR_CREATE caps */ - if (S_ISDIR(ci->vfs_inode.i_mode) && (had & CEPH_CAP_DIR_CREATE) && + if (S_ISDIR(ci->netfs.inode.i_mode) && (had & CEPH_CAP_DIR_CREATE) && !(issued & CEPH_CAP_DIR_CREATE)) { ceph_put_string(rcu_dereference_raw(ci->i_cached_layout.pool_ns)); memset(&ci->i_cached_layout, 0, sizeof(ci->i_cached_layout)); @@ -771,7 +771,7 @@ static int __cap_is_valid(struct ceph_cap *cap) if (cap->cap_gen < gen || time_after_eq(jiffies, ttl)) { dout("__cap_is_valid %p cap %p issued %s " - "but STALE (gen %u vs %u)\n", &cap->ci->vfs_inode, + "but STALE (gen %u vs %u)\n", &cap->ci->netfs.inode, cap, ceph_cap_string(cap->issued), cap->cap_gen, gen); return 0; } @@ -797,7 +797,7 @@ int __ceph_caps_issued(struct ceph_inode_info *ci, int *implemented) if (!__cap_is_valid(cap)) continue; dout("__ceph_caps_issued %p cap %p issued %s\n", - &ci->vfs_inode, cap, ceph_cap_string(cap->issued)); + &ci->netfs.inode, cap, ceph_cap_string(cap->issued)); have |= cap->issued; if (implemented) *implemented |= cap->implemented; @@ -844,12 +844,12 @@ static void __touch_cap(struct ceph_cap *cap) spin_lock(&s->s_cap_lock); if (!s->s_cap_iterator) { - dout("__touch_cap %p cap %p mds%d\n", &cap->ci->vfs_inode, cap, + dout("__touch_cap %p cap %p mds%d\n", &cap->ci->netfs.inode, cap, s->s_mds); list_move_tail(&cap->session_caps, &s->s_caps); } else { dout("__touch_cap %p cap %p mds%d NOP, iterating over caps\n", - &cap->ci->vfs_inode, cap, s->s_mds); + &cap->ci->netfs.inode, cap, s->s_mds); } spin_unlock(&s->s_cap_lock); } @@ -867,7 +867,7 @@ int __ceph_caps_issued_mask(struct ceph_inode_info *ci, int mask, int touch) if ((have & mask) == mask) { dout("__ceph_caps_issued_mask ino 0x%llx snap issued %s" - " (mask %s)\n", ceph_ino(&ci->vfs_inode), + " (mask %s)\n", ceph_ino(&ci->netfs.inode), ceph_cap_string(have), ceph_cap_string(mask)); return 1; @@ -879,7 +879,7 @@ int __ceph_caps_issued_mask(struct ceph_inode_info *ci, int mask, int touch) continue; if ((cap->issued & mask) == mask) { dout("__ceph_caps_issued_mask ino 0x%llx cap %p issued %s" - " (mask %s)\n", ceph_ino(&ci->vfs_inode), cap, + " (mask %s)\n", ceph_ino(&ci->netfs.inode), cap, ceph_cap_string(cap->issued), ceph_cap_string(mask)); if (touch) @@ -891,7 +891,7 @@ int __ceph_caps_issued_mask(struct ceph_inode_info *ci, int mask, int touch) have |= cap->issued; if ((have & mask) == mask) { dout("__ceph_caps_issued_mask ino 0x%llx combo issued %s" - " (mask %s)\n", ceph_ino(&ci->vfs_inode), + " (mask %s)\n", ceph_ino(&ci->netfs.inode), ceph_cap_string(cap->issued), ceph_cap_string(mask)); if (touch) { @@ -919,7 +919,7 @@ int __ceph_caps_issued_mask(struct ceph_inode_info *ci, int mask, int touch) int __ceph_caps_issued_mask_metric(struct ceph_inode_info *ci, int mask, int touch) { - struct ceph_fs_client *fsc = ceph_sb_to_client(ci->vfs_inode.i_sb); + struct ceph_fs_client *fsc = ceph_sb_to_client(ci->netfs.inode.i_sb); int r; r = __ceph_caps_issued_mask(ci, mask, touch); @@ -950,7 +950,7 @@ int __ceph_caps_revoking_other(struct ceph_inode_info *ci, int ceph_caps_revoking(struct ceph_inode_info *ci, int mask) { - struct inode *inode = &ci->vfs_inode; + struct inode *inode = &ci->netfs.inode; int ret; spin_lock(&ci->i_ceph_lock); @@ -969,8 +969,8 @@ int __ceph_caps_used(struct ceph_inode_info *ci) if (ci->i_rd_ref) used |= CEPH_CAP_FILE_RD; if (ci->i_rdcache_ref || - (S_ISREG(ci->vfs_inode.i_mode) && - ci->vfs_inode.i_data.nrpages)) + (S_ISREG(ci->netfs.inode.i_mode) && + ci->netfs.inode.i_data.nrpages)) used |= CEPH_CAP_FILE_CACHE; if (ci->i_wr_ref) used |= CEPH_CAP_FILE_WR; @@ -993,11 +993,11 @@ int __ceph_caps_file_wanted(struct ceph_inode_info *ci) const int WR_SHIFT = ffs(CEPH_FILE_MODE_WR); const int LAZY_SHIFT = ffs(CEPH_FILE_MODE_LAZY); struct ceph_mount_options *opt = - ceph_inode_to_client(&ci->vfs_inode)->mount_options; + ceph_inode_to_client(&ci->netfs.inode)->mount_options; unsigned long used_cutoff = jiffies - opt->caps_wanted_delay_max * HZ; unsigned long idle_cutoff = jiffies - opt->caps_wanted_delay_min * HZ; - if (S_ISDIR(ci->vfs_inode.i_mode)) { + if (S_ISDIR(ci->netfs.inode.i_mode)) { int want = 0; /* use used_cutoff here, to keep dir's wanted caps longer */ @@ -1050,7 +1050,7 @@ int __ceph_caps_file_wanted(struct ceph_inode_info *ci) int __ceph_caps_wanted(struct ceph_inode_info *ci) { int w = __ceph_caps_file_wanted(ci) | __ceph_caps_used(ci); - if (S_ISDIR(ci->vfs_inode.i_mode)) { + if (S_ISDIR(ci->netfs.inode.i_mode)) { /* we want EXCL if holding caps of dir ops */ if (w & CEPH_CAP_ANY_DIR_OPS) w |= CEPH_CAP_FILE_EXCL; @@ -1116,9 +1116,9 @@ void __ceph_remove_cap(struct ceph_cap *cap, bool queue_release) lockdep_assert_held(&ci->i_ceph_lock); - dout("__ceph_remove_cap %p from %p\n", cap, &ci->vfs_inode); + dout("__ceph_remove_cap %p from %p\n", cap, &ci->netfs.inode); - mdsc = ceph_inode_to_client(&ci->vfs_inode)->mdsc; + mdsc = ceph_inode_to_client(&ci->netfs.inode)->mdsc; /* remove from inode's cap rbtree, and clear auth cap */ rb_erase(&cap->ci_node, &ci->i_caps); @@ -1169,7 +1169,7 @@ void __ceph_remove_cap(struct ceph_cap *cap, bool queue_release) * keep i_snap_realm. */ if (ci->i_wr_ref == 0 && ci->i_snap_realm) - ceph_change_snap_realm(&ci->vfs_inode, NULL); + ceph_change_snap_realm(&ci->netfs.inode, NULL); __cap_delay_cancel(mdsc, ci); } @@ -1188,11 +1188,11 @@ void ceph_remove_cap(struct ceph_cap *cap, bool queue_release) lockdep_assert_held(&ci->i_ceph_lock); - fsc = ceph_inode_to_client(&ci->vfs_inode); + fsc = ceph_inode_to_client(&ci->netfs.inode); WARN_ON_ONCE(ci->i_auth_cap == cap && !list_empty(&ci->i_dirty_item) && !fsc->blocklisted && - !ceph_inode_is_shutdown(&ci->vfs_inode)); + !ceph_inode_is_shutdown(&ci->netfs.inode)); __ceph_remove_cap(cap, queue_release); } @@ -1343,7 +1343,7 @@ static void __prep_cap(struct cap_msg_args *arg, struct ceph_cap *cap, int flushing, u64 flush_tid, u64 oldest_flush_tid) { struct ceph_inode_info *ci = cap->ci; - struct inode *inode = &ci->vfs_inode; + struct inode *inode = &ci->netfs.inode; int held, revoking; lockdep_assert_held(&ci->i_ceph_lock); @@ -1440,7 +1440,7 @@ static void __prep_cap(struct cap_msg_args *arg, struct ceph_cap *cap, static void __send_cap(struct cap_msg_args *arg, struct ceph_inode_info *ci) { struct ceph_msg *msg; - struct inode *inode = &ci->vfs_inode; + struct inode *inode = &ci->netfs.inode; msg = ceph_msg_new(CEPH_MSG_CLIENT_CAPS, CAP_MSG_SIZE, GFP_NOFS, false); if (!msg) { @@ -1528,7 +1528,7 @@ static void __ceph_flush_snaps(struct ceph_inode_info *ci, __releases(ci->i_ceph_lock) __acquires(ci->i_ceph_lock) { - struct inode *inode = &ci->vfs_inode; + struct inode *inode = &ci->netfs.inode; struct ceph_mds_client *mdsc = session->s_mdsc; struct ceph_cap_snap *capsnap; u64 oldest_flush_tid = 0; @@ -1622,7 +1622,7 @@ static void __ceph_flush_snaps(struct ceph_inode_info *ci, void ceph_flush_snaps(struct ceph_inode_info *ci, struct ceph_mds_session **psession) { - struct inode *inode = &ci->vfs_inode; + struct inode *inode = &ci->netfs.inode; struct ceph_mds_client *mdsc = ceph_inode_to_client(inode)->mdsc; struct ceph_mds_session *session = NULL; int mds; @@ -1682,8 +1682,8 @@ int __ceph_mark_dirty_caps(struct ceph_inode_info *ci, int mask, struct ceph_cap_flush **pcf) { struct ceph_mds_client *mdsc = - ceph_sb_to_client(ci->vfs_inode.i_sb)->mdsc; - struct inode *inode = &ci->vfs_inode; + ceph_sb_to_client(ci->netfs.inode.i_sb)->mdsc; + struct inode *inode = &ci->netfs.inode; int was = ci->i_dirty_caps; int dirty = 0; @@ -1696,7 +1696,7 @@ int __ceph_mark_dirty_caps(struct ceph_inode_info *ci, int mask, return 0; } - dout("__mark_dirty_caps %p %s dirty %s -> %s\n", &ci->vfs_inode, + dout("__mark_dirty_caps %p %s dirty %s -> %s\n", &ci->netfs.inode, ceph_cap_string(mask), ceph_cap_string(was), ceph_cap_string(was | mask)); ci->i_dirty_caps |= mask; @@ -1712,7 +1712,7 @@ int __ceph_mark_dirty_caps(struct ceph_inode_info *ci, int mask, ci->i_snap_realm->cached_context); } dout(" inode %p now dirty snapc %p auth cap %p\n", - &ci->vfs_inode, ci->i_head_snapc, ci->i_auth_cap); + &ci->netfs.inode, ci->i_head_snapc, ci->i_auth_cap); BUG_ON(!list_empty(&ci->i_dirty_item)); spin_lock(&mdsc->cap_dirty_lock); list_add(&ci->i_dirty_item, &session->s_cap_dirty); @@ -1875,7 +1875,7 @@ static int try_nonblocking_invalidate(struct inode *inode) bool __ceph_should_report_size(struct ceph_inode_info *ci) { - loff_t size = i_size_read(&ci->vfs_inode); + loff_t size = i_size_read(&ci->netfs.inode); /* mds will adjust max size according to the reported size */ if (ci->i_flushing_caps & CEPH_CAP_FILE_WR) return false; @@ -1900,7 +1900,7 @@ bool __ceph_should_report_size(struct ceph_inode_info *ci) void ceph_check_caps(struct ceph_inode_info *ci, int flags, struct ceph_mds_session *session) { - struct inode *inode = &ci->vfs_inode; + struct inode *inode = &ci->netfs.inode; struct ceph_mds_client *mdsc = ceph_sb_to_mdsc(inode->i_sb); struct ceph_cap *cap; u64 flush_tid, oldest_flush_tid; @@ -2467,7 +2467,7 @@ static void __kick_flushing_caps(struct ceph_mds_client *mdsc, __releases(ci->i_ceph_lock) __acquires(ci->i_ceph_lock) { - struct inode *inode = &ci->vfs_inode; + struct inode *inode = &ci->netfs.inode; struct ceph_cap *cap; struct ceph_cap_flush *cf; int ret; @@ -2560,7 +2560,7 @@ void ceph_early_kick_flushing_caps(struct ceph_mds_client *mdsc, cap = ci->i_auth_cap; if (!(cap && cap->session == session)) { pr_err("%p auth cap %p not mds%d ???\n", - &ci->vfs_inode, cap, session->s_mds); + &ci->netfs.inode, cap, session->s_mds); spin_unlock(&ci->i_ceph_lock); continue; } @@ -2610,7 +2610,7 @@ void ceph_kick_flushing_caps(struct ceph_mds_client *mdsc, cap = ci->i_auth_cap; if (!(cap && cap->session == session)) { pr_err("%p auth cap %p not mds%d ???\n", - &ci->vfs_inode, cap, session->s_mds); + &ci->netfs.inode, cap, session->s_mds); spin_unlock(&ci->i_ceph_lock); continue; } @@ -2630,7 +2630,7 @@ void ceph_kick_flushing_inode_caps(struct ceph_mds_session *session, lockdep_assert_held(&ci->i_ceph_lock); - dout("%s %p flushing %s\n", __func__, &ci->vfs_inode, + dout("%s %p flushing %s\n", __func__, &ci->netfs.inode, ceph_cap_string(ci->i_flushing_caps)); if (!list_empty(&ci->i_cap_flush_list)) { @@ -2673,10 +2673,10 @@ void ceph_take_cap_refs(struct ceph_inode_info *ci, int got, } if (got & CEPH_CAP_FILE_BUFFER) { if (ci->i_wb_ref == 0) - ihold(&ci->vfs_inode); + ihold(&ci->netfs.inode); ci->i_wb_ref++; dout("%s %p wb %d -> %d (?)\n", __func__, - &ci->vfs_inode, ci->i_wb_ref-1, ci->i_wb_ref); + &ci->netfs.inode, ci->i_wb_ref-1, ci->i_wb_ref); } } @@ -3004,7 +3004,7 @@ int ceph_get_caps(struct file *filp, int need, int want, loff_t endoff, int *got return ret; } - if (S_ISREG(ci->vfs_inode.i_mode) && + if (S_ISREG(ci->netfs.inode.i_mode) && ci->i_inline_version != CEPH_INLINE_NONE && (_got & (CEPH_CAP_FILE_CACHE|CEPH_CAP_FILE_LAZYIO)) && i_size_read(inode) > 0) { @@ -3094,7 +3094,7 @@ enum put_cap_refs_mode { static void __ceph_put_cap_refs(struct ceph_inode_info *ci, int had, enum put_cap_refs_mode mode) { - struct inode *inode = &ci->vfs_inode; + struct inode *inode = &ci->netfs.inode; int last = 0, put = 0, flushsnaps = 0, wake = 0; bool check_flushsnaps = false; @@ -3202,7 +3202,7 @@ void ceph_put_cap_refs_no_check_caps(struct ceph_inode_info *ci, int had) void ceph_put_wrbuffer_cap_refs(struct ceph_inode_info *ci, int nr, struct ceph_snap_context *snapc) { - struct inode *inode = &ci->vfs_inode; + struct inode *inode = &ci->netfs.inode; struct ceph_cap_snap *capsnap = NULL, *iter; int put = 0; bool last = false; @@ -3698,7 +3698,7 @@ static void handle_cap_flush_ack(struct inode *inode, u64 flush_tid, session->s_mds, &list_first_entry(&session->s_cap_flushing, struct ceph_inode_info, - i_flushing_item)->vfs_inode); + i_flushing_item)->netfs.inode); } } mdsc->num_cap_flushing--; @@ -4345,7 +4345,7 @@ unsigned long ceph_check_delayed_caps(struct ceph_mds_client *mdsc) break; list_del_init(&ci->i_cap_delay_list); - inode = igrab(&ci->vfs_inode); + inode = igrab(&ci->netfs.inode); if (inode) { spin_unlock(&mdsc->cap_delay_lock); dout("check_delayed_caps on %p\n", inode); @@ -4373,7 +4373,7 @@ static void flush_dirty_session_caps(struct ceph_mds_session *s) while (!list_empty(&s->s_cap_dirty)) { ci = list_first_entry(&s->s_cap_dirty, struct ceph_inode_info, i_dirty_item); - inode = &ci->vfs_inode; + inode = &ci->netfs.inode; ihold(inode); dout("flush_dirty_caps %llx.%llx\n", ceph_vinop(inode)); spin_unlock(&mdsc->cap_dirty_lock); @@ -4407,7 +4407,7 @@ void __ceph_touch_fmode(struct ceph_inode_info *ci, void ceph_get_fmode(struct ceph_inode_info *ci, int fmode, int count) { - struct ceph_mds_client *mdsc = ceph_sb_to_mdsc(ci->vfs_inode.i_sb); + struct ceph_mds_client *mdsc = ceph_sb_to_mdsc(ci->netfs.inode.i_sb); int bits = (fmode << 1) | 1; bool already_opened = false; int i; @@ -4441,7 +4441,7 @@ void ceph_get_fmode(struct ceph_inode_info *ci, int fmode, int count) */ void ceph_put_fmode(struct ceph_inode_info *ci, int fmode, int count) { - struct ceph_mds_client *mdsc = ceph_sb_to_mdsc(ci->vfs_inode.i_sb); + struct ceph_mds_client *mdsc = ceph_sb_to_mdsc(ci->netfs.inode.i_sb); int bits = (fmode << 1) | 1; bool is_closed = true; int i; @@ -4656,7 +4656,7 @@ int ceph_purge_inode_cap(struct inode *inode, struct ceph_cap *cap, bool *invali lockdep_assert_held(&ci->i_ceph_lock); dout("removing cap %p, ci is %p, inode is %p\n", - cap, ci, &ci->vfs_inode); + cap, ci, &ci->netfs.inode); is_auth = (cap == ci->i_auth_cap); __ceph_remove_cap(cap, false); diff --git a/fs/ceph/file.c b/fs/ceph/file.c index 8c8226c0feac..da59e836a06e 100644 --- a/fs/ceph/file.c +++ b/fs/ceph/file.c @@ -205,7 +205,7 @@ static int ceph_init_file_info(struct inode *inode, struct file *file, { struct ceph_inode_info *ci = ceph_inode(inode); struct ceph_mount_options *opt = - ceph_inode_to_client(&ci->vfs_inode)->mount_options; + ceph_inode_to_client(&ci->netfs.inode)->mount_options; struct ceph_file_info *fi; int ret; diff --git a/fs/ceph/inode.c b/fs/ceph/inode.c index b7e9cac3aeef..56c53ab3618e 100644 --- a/fs/ceph/inode.c +++ b/fs/ceph/inode.c @@ -176,7 +176,7 @@ static struct ceph_inode_frag *__get_or_create_frag(struct ceph_inode_info *ci, rb_insert_color(&frag->node, &ci->i_fragtree); dout("get_or_create_frag added %llx.%llx frag %x\n", - ceph_vinop(&ci->vfs_inode), f); + ceph_vinop(&ci->netfs.inode), f); return frag; } @@ -457,10 +457,10 @@ struct inode *ceph_alloc_inode(struct super_block *sb) if (!ci) return NULL; - dout("alloc_inode %p\n", &ci->vfs_inode); + dout("alloc_inode %p\n", &ci->netfs.inode); /* Set parameters for the netfs library */ - netfs_i_context_init(&ci->vfs_inode, &ceph_netfs_ops); + netfs_inode_init(&ci->netfs, &ceph_netfs_ops); spin_lock_init(&ci->i_ceph_lock); @@ -547,7 +547,7 @@ struct inode *ceph_alloc_inode(struct super_block *sb) INIT_WORK(&ci->i_work, ceph_inode_work); ci->i_work_mask = 0; memset(&ci->i_btime, '\0', sizeof(ci->i_btime)); - return &ci->vfs_inode; + return &ci->netfs.inode; } void ceph_free_inode(struct inode *inode) @@ -1978,7 +1978,7 @@ static void ceph_inode_work(struct work_struct *work) { struct ceph_inode_info *ci = container_of(work, struct ceph_inode_info, i_work); - struct inode *inode = &ci->vfs_inode; + struct inode *inode = &ci->netfs.inode; if (test_and_clear_bit(CEPH_I_WORK_WRITEBACK, &ci->i_work_mask)) { dout("writeback %p\n", inode); diff --git a/fs/ceph/mds_client.c b/fs/ceph/mds_client.c index f5d110d90b77..33f517d549ce 100644 --- a/fs/ceph/mds_client.c +++ b/fs/ceph/mds_client.c @@ -1564,7 +1564,7 @@ int ceph_iterate_session_caps(struct ceph_mds_session *session, p = session->s_caps.next; while (p != &session->s_caps) { cap = list_entry(p, struct ceph_cap, session_caps); - inode = igrab(&cap->ci->vfs_inode); + inode = igrab(&cap->ci->netfs.inode); if (!inode) { p = p->next; continue; @@ -1622,7 +1622,7 @@ static int remove_session_caps_cb(struct inode *inode, struct ceph_cap *cap, int iputs; dout("removing cap %p, ci is %p, inode is %p\n", - cap, ci, &ci->vfs_inode); + cap, ci, &ci->netfs.inode); spin_lock(&ci->i_ceph_lock); iputs = ceph_purge_inode_cap(inode, cap, &invalidate); spin_unlock(&ci->i_ceph_lock); diff --git a/fs/ceph/snap.c b/fs/ceph/snap.c index 322ee5add942..864cdaa0d2bd 100644 --- a/fs/ceph/snap.c +++ b/fs/ceph/snap.c @@ -521,7 +521,7 @@ static bool has_new_snaps(struct ceph_snap_context *o, static void ceph_queue_cap_snap(struct ceph_inode_info *ci, struct ceph_cap_snap **pcapsnap) { - struct inode *inode = &ci->vfs_inode; + struct inode *inode = &ci->netfs.inode; struct ceph_snap_context *old_snapc, *new_snapc; struct ceph_cap_snap *capsnap = *pcapsnap; struct ceph_buffer *old_blob = NULL; @@ -652,7 +652,7 @@ update_snapc: int __ceph_finish_cap_snap(struct ceph_inode_info *ci, struct ceph_cap_snap *capsnap) { - struct inode *inode = &ci->vfs_inode; + struct inode *inode = &ci->netfs.inode; struct ceph_mds_client *mdsc = ceph_sb_to_mdsc(inode->i_sb); BUG_ON(capsnap->writing); @@ -712,7 +712,7 @@ static void queue_realm_cap_snaps(struct ceph_snap_realm *realm) spin_lock(&realm->inodes_with_caps_lock); list_for_each_entry(ci, &realm->inodes_with_caps, i_snap_realm_item) { - struct inode *inode = igrab(&ci->vfs_inode); + struct inode *inode = igrab(&ci->netfs.inode); if (!inode) continue; spin_unlock(&realm->inodes_with_caps_lock); @@ -904,7 +904,7 @@ static void flush_snaps(struct ceph_mds_client *mdsc) while (!list_empty(&mdsc->snap_flush_list)) { ci = list_first_entry(&mdsc->snap_flush_list, struct ceph_inode_info, i_snap_flush_item); - inode = &ci->vfs_inode; + inode = &ci->netfs.inode; ihold(inode); spin_unlock(&mdsc->snap_flush_lock); ceph_flush_snaps(ci, &session); diff --git a/fs/ceph/super.c b/fs/ceph/super.c index b73b4f75462c..40140805bdcf 100644 --- a/fs/ceph/super.c +++ b/fs/ceph/super.c @@ -876,7 +876,7 @@ mempool_t *ceph_wb_pagevec_pool; static void ceph_inode_init_once(void *foo) { struct ceph_inode_info *ci = foo; - inode_init_once(&ci->vfs_inode); + inode_init_once(&ci->netfs.inode); } static int __init init_caches(void) diff --git a/fs/ceph/super.h b/fs/ceph/super.h index dd7dac0f984a..f59dac66955b 100644 --- a/fs/ceph/super.h +++ b/fs/ceph/super.h @@ -316,11 +316,7 @@ struct ceph_inode_xattrs_info { * Ceph inode. */ struct ceph_inode_info { - struct { - /* These must be contiguous */ - struct inode vfs_inode; - struct netfs_i_context netfs_ctx; /* Netfslib context */ - }; + struct netfs_inode netfs; /* Netfslib context and vfs inode */ struct ceph_vino i_vino; /* ceph ino + snap */ spinlock_t i_ceph_lock; @@ -436,7 +432,7 @@ struct ceph_inode_info { static inline struct ceph_inode_info * ceph_inode(const struct inode *inode) { - return container_of(inode, struct ceph_inode_info, vfs_inode); + return container_of(inode, struct ceph_inode_info, netfs.inode); } static inline struct ceph_fs_client * @@ -1316,7 +1312,7 @@ static inline void __ceph_update_quota(struct ceph_inode_info *ci, has_quota = __ceph_has_quota(ci, QUOTA_GET_ANY); if (had_quota != has_quota) - ceph_adjust_quota_realms_count(&ci->vfs_inode, has_quota); + ceph_adjust_quota_realms_count(&ci->netfs.inode, has_quota); } extern void ceph_handle_quota(struct ceph_mds_client *mdsc, diff --git a/fs/ceph/xattr.c b/fs/ceph/xattr.c index 8c2dc2c762a4..f141f5246163 100644 --- a/fs/ceph/xattr.c +++ b/fs/ceph/xattr.c @@ -57,7 +57,7 @@ static bool ceph_vxattrcb_layout_exists(struct ceph_inode_info *ci) static ssize_t ceph_vxattrcb_layout(struct ceph_inode_info *ci, char *val, size_t size) { - struct ceph_fs_client *fsc = ceph_sb_to_client(ci->vfs_inode.i_sb); + struct ceph_fs_client *fsc = ceph_sb_to_client(ci->netfs.inode.i_sb); struct ceph_osd_client *osdc = &fsc->client->osdc; struct ceph_string *pool_ns; s64 pool = ci->i_layout.pool_id; @@ -69,7 +69,7 @@ static ssize_t ceph_vxattrcb_layout(struct ceph_inode_info *ci, char *val, pool_ns = ceph_try_get_string(ci->i_layout.pool_ns); - dout("ceph_vxattrcb_layout %p\n", &ci->vfs_inode); + dout("ceph_vxattrcb_layout %p\n", &ci->netfs.inode); down_read(&osdc->lock); pool_name = ceph_pg_pool_name_by_id(osdc->osdmap, pool); if (pool_name) { @@ -161,7 +161,7 @@ static ssize_t ceph_vxattrcb_layout_pool(struct ceph_inode_info *ci, char *val, size_t size) { ssize_t ret; - struct ceph_fs_client *fsc = ceph_sb_to_client(ci->vfs_inode.i_sb); + struct ceph_fs_client *fsc = ceph_sb_to_client(ci->netfs.inode.i_sb); struct ceph_osd_client *osdc = &fsc->client->osdc; s64 pool = ci->i_layout.pool_id; const char *pool_name; @@ -313,7 +313,7 @@ static ssize_t ceph_vxattrcb_snap_btime(struct ceph_inode_info *ci, char *val, static ssize_t ceph_vxattrcb_cluster_fsid(struct ceph_inode_info *ci, char *val, size_t size) { - struct ceph_fs_client *fsc = ceph_sb_to_client(ci->vfs_inode.i_sb); + struct ceph_fs_client *fsc = ceph_sb_to_client(ci->netfs.inode.i_sb); return ceph_fmt_xattr(val, size, "%pU", &fsc->client->fsid); } @@ -321,7 +321,7 @@ static ssize_t ceph_vxattrcb_cluster_fsid(struct ceph_inode_info *ci, static ssize_t ceph_vxattrcb_client_id(struct ceph_inode_info *ci, char *val, size_t size) { - struct ceph_fs_client *fsc = ceph_sb_to_client(ci->vfs_inode.i_sb); + struct ceph_fs_client *fsc = ceph_sb_to_client(ci->netfs.inode.i_sb); return ceph_fmt_xattr(val, size, "client%lld", ceph_client_gid(fsc->client)); @@ -629,7 +629,7 @@ static int __set_xattr(struct ceph_inode_info *ci, } dout("__set_xattr_val added %llx.%llx xattr %p %.*s=%.*s\n", - ceph_vinop(&ci->vfs_inode), xattr, name_len, name, val_len, val); + ceph_vinop(&ci->netfs.inode), xattr, name_len, name, val_len, val); return 0; } @@ -871,7 +871,7 @@ struct ceph_buffer *__ceph_build_xattrs_blob(struct ceph_inode_info *ci) struct ceph_buffer *old_blob = NULL; void *dest; - dout("__build_xattrs_blob %p\n", &ci->vfs_inode); + dout("__build_xattrs_blob %p\n", &ci->netfs.inode); if (ci->i_xattrs.dirty) { int need = __get_required_blob_size(ci, 0, 0); diff --git a/fs/cifs/cifs_debug.c b/fs/cifs/cifs_debug.c index 1dd995efd5b8..2cfbac8bb965 100644 --- a/fs/cifs/cifs_debug.c +++ b/fs/cifs/cifs_debug.c @@ -162,6 +162,8 @@ cifs_dump_iface(struct seq_file *m, struct cifs_server_iface *iface) seq_printf(m, "\t\tIPv4: %pI4\n", &ipv4->sin_addr); else if (iface->sockaddr.ss_family == AF_INET6) seq_printf(m, "\t\tIPv6: %pI6\n", &ipv6->sin6_addr); + if (!iface->is_active) + seq_puts(m, "\t\t[for-cleanup]\n"); } static int cifs_debug_files_proc_show(struct seq_file *m, void *v) @@ -221,6 +223,7 @@ static int cifs_debug_data_proc_show(struct seq_file *m, void *v) struct TCP_Server_Info *server; struct cifs_ses *ses; struct cifs_tcon *tcon; + struct cifs_server_iface *iface; int c, i, j; seq_puts(m, @@ -456,11 +459,10 @@ skip_rdma: if (ses->iface_count) seq_printf(m, "\n\n\tServer interfaces: %zu", ses->iface_count); - for (j = 0; j < ses->iface_count; j++) { - struct cifs_server_iface *iface; - - iface = &ses->iface_list[j]; - seq_printf(m, "\n\t%d)", j+1); + j = 0; + list_for_each_entry(iface, &ses->iface_list, + iface_head) { + seq_printf(m, "\n\t%d)", ++j); cifs_dump_iface(m, iface); if (is_ses_using_iface(ses, iface)) seq_puts(m, "\t\t[CONNECTED]\n"); diff --git a/fs/cifs/cifsfs.c b/fs/cifs/cifsfs.c index 12c872800326..8f2e003e0590 100644 --- a/fs/cifs/cifsfs.c +++ b/fs/cifs/cifsfs.c @@ -377,7 +377,7 @@ cifs_alloc_inode(struct super_block *sb) cifs_inode->flags = 0; spin_lock_init(&cifs_inode->writers_lock); cifs_inode->writers = 0; - cifs_inode->vfs_inode.i_blkbits = 14; /* 2**14 = CIFS_MAX_MSGSIZE */ + cifs_inode->netfs.inode.i_blkbits = 14; /* 2**14 = CIFS_MAX_MSGSIZE */ cifs_inode->server_eof = 0; cifs_inode->uniqueid = 0; cifs_inode->createtime = 0; @@ -389,12 +389,12 @@ cifs_alloc_inode(struct super_block *sb) * Can not set i_flags here - they get immediately overwritten to zero * by the VFS. */ - /* cifs_inode->vfs_inode.i_flags = S_NOATIME | S_NOCMTIME; */ + /* cifs_inode->netfs.inode.i_flags = S_NOATIME | S_NOCMTIME; */ INIT_LIST_HEAD(&cifs_inode->openFileList); INIT_LIST_HEAD(&cifs_inode->llist); INIT_LIST_HEAD(&cifs_inode->deferred_closes); spin_lock_init(&cifs_inode->deferred_lock); - return &cifs_inode->vfs_inode; + return &cifs_inode->netfs.inode; } static void @@ -1086,7 +1086,7 @@ struct file_system_type cifs_fs_type = { }; MODULE_ALIAS_FS("cifs"); -static struct file_system_type smb3_fs_type = { +struct file_system_type smb3_fs_type = { .owner = THIS_MODULE, .name = "smb3", .init_fs_context = smb3_init_fs_context, @@ -1418,7 +1418,7 @@ cifs_init_once(void *inode) { struct cifsInodeInfo *cifsi = inode; - inode_init_once(&cifsi->vfs_inode); + inode_init_once(&cifsi->netfs.inode); init_rwsem(&cifsi->lock_sem); } diff --git a/fs/cifs/cifsfs.h b/fs/cifs/cifsfs.h index dd7e070ca243..b17be47a8e59 100644 --- a/fs/cifs/cifsfs.h +++ b/fs/cifs/cifsfs.h @@ -38,7 +38,7 @@ static inline unsigned long cifs_get_time(struct dentry *dentry) return (unsigned long) dentry->d_fsdata; } -extern struct file_system_type cifs_fs_type; +extern struct file_system_type cifs_fs_type, smb3_fs_type; extern const struct address_space_operations cifs_addr_ops; extern const struct address_space_operations cifs_addr_ops_smallbuf; diff --git a/fs/cifs/cifsglob.h b/fs/cifs/cifsglob.h index f873379066c7..a643c84ff1e9 100644 --- a/fs/cifs/cifsglob.h +++ b/fs/cifs/cifsglob.h @@ -80,6 +80,9 @@ #define SMB_DNS_RESOLVE_INTERVAL_MIN 120 #define SMB_DNS_RESOLVE_INTERVAL_DEFAULT 600 +/* smb multichannel query server interfaces interval in seconds */ +#define SMB_INTERFACE_POLL_INTERVAL 600 + /* maximum number of PDUs in one compound */ #define MAX_COMPOUND 5 @@ -933,15 +936,67 @@ static inline void cifs_set_net_ns(struct TCP_Server_Info *srv, struct net *net) #endif struct cifs_server_iface { + struct list_head iface_head; + struct kref refcount; size_t speed; unsigned int rdma_capable : 1; unsigned int rss_capable : 1; + unsigned int is_active : 1; /* unset if non existent */ struct sockaddr_storage sockaddr; }; +/* release iface when last ref is dropped */ +static inline void +release_iface(struct kref *ref) +{ + struct cifs_server_iface *iface = container_of(ref, + struct cifs_server_iface, + refcount); + list_del_init(&iface->iface_head); + kfree(iface); +} + +/* + * compare two interfaces a and b + * return 0 if everything matches. + * return 1 if a has higher link speed, or rdma capable, or rss capable + * return -1 otherwise. + */ +static inline int +iface_cmp(struct cifs_server_iface *a, struct cifs_server_iface *b) +{ + int cmp_ret = 0; + + WARN_ON(!a || !b); + if (a->speed == b->speed) { + if (a->rdma_capable == b->rdma_capable) { + if (a->rss_capable == b->rss_capable) { + cmp_ret = memcmp(&a->sockaddr, &b->sockaddr, + sizeof(a->sockaddr)); + if (!cmp_ret) + return 0; + else if (cmp_ret > 0) + return 1; + else + return -1; + } else if (a->rss_capable > b->rss_capable) + return 1; + else + return -1; + } else if (a->rdma_capable > b->rdma_capable) + return 1; + else + return -1; + } else if (a->speed > b->speed) + return 1; + else + return -1; +} + struct cifs_chan { unsigned int in_reconnect : 1; /* if session setup in progress for this channel */ struct TCP_Server_Info *server; + struct cifs_server_iface *iface; /* interface in use */ __u8 signkey[SMB3_SIGN_KEY_SIZE]; }; @@ -993,7 +1048,7 @@ struct cifs_ses { */ spinlock_t iface_lock; /* ========= begin: protected by iface_lock ======== */ - struct cifs_server_iface *iface_list; + struct list_head iface_list; size_t iface_count; unsigned long iface_last_update; /* jiffies */ /* ========= end: protected by iface_lock ======== */ @@ -1203,6 +1258,7 @@ struct cifs_tcon { #ifdef CONFIG_CIFS_DFS_UPCALL struct list_head ulist; /* cache update list */ #endif + struct delayed_work query_interfaces; /* query interfaces workqueue job */ }; /* @@ -1479,20 +1535,16 @@ void cifsFileInfo_put(struct cifsFileInfo *cifs_file); #define CIFS_CACHE_RW_FLG (CIFS_CACHE_READ_FLG | CIFS_CACHE_WRITE_FLG) #define CIFS_CACHE_RHW_FLG (CIFS_CACHE_RW_FLG | CIFS_CACHE_HANDLE_FLG) -#define CIFS_CACHE_READ(cinode) ((cinode->oplock & CIFS_CACHE_READ_FLG) || (CIFS_SB(cinode->vfs_inode.i_sb)->mnt_cifs_flags & CIFS_MOUNT_RO_CACHE)) +#define CIFS_CACHE_READ(cinode) ((cinode->oplock & CIFS_CACHE_READ_FLG) || (CIFS_SB(cinode->netfs.inode.i_sb)->mnt_cifs_flags & CIFS_MOUNT_RO_CACHE)) #define CIFS_CACHE_HANDLE(cinode) (cinode->oplock & CIFS_CACHE_HANDLE_FLG) -#define CIFS_CACHE_WRITE(cinode) ((cinode->oplock & CIFS_CACHE_WRITE_FLG) || (CIFS_SB(cinode->vfs_inode.i_sb)->mnt_cifs_flags & CIFS_MOUNT_RW_CACHE)) +#define CIFS_CACHE_WRITE(cinode) ((cinode->oplock & CIFS_CACHE_WRITE_FLG) || (CIFS_SB(cinode->netfs.inode.i_sb)->mnt_cifs_flags & CIFS_MOUNT_RW_CACHE)) /* * One of these for each file inode */ struct cifsInodeInfo { - struct { - /* These must be contiguous */ - struct inode vfs_inode; /* the VFS's inode record */ - struct netfs_i_context netfs_ctx; /* Netfslib context */ - }; + struct netfs_inode netfs; /* Netfslib context and vfs inode */ bool can_cache_brlcks; struct list_head llist; /* locks helb by this inode */ /* @@ -1531,7 +1583,7 @@ struct cifsInodeInfo { static inline struct cifsInodeInfo * CIFS_I(struct inode *inode) { - return container_of(inode, struct cifsInodeInfo, vfs_inode); + return container_of(inode, struct cifsInodeInfo, netfs.inode); } static inline struct cifs_sb_info * diff --git a/fs/cifs/cifsproto.h b/fs/cifs/cifsproto.h index 3b7366ec03c7..d59aebefa71c 100644 --- a/fs/cifs/cifsproto.h +++ b/fs/cifs/cifsproto.h @@ -636,6 +636,13 @@ cifs_chan_clear_need_reconnect(struct cifs_ses *ses, bool cifs_chan_needs_reconnect(struct cifs_ses *ses, struct TCP_Server_Info *server); +bool +cifs_chan_is_iface_active(struct cifs_ses *ses, + struct TCP_Server_Info *server); +int +cifs_chan_update_iface(struct cifs_ses *ses, struct TCP_Server_Info *server); +int +SMB3_request_interfaces(const unsigned int xid, struct cifs_tcon *tcon); void extract_unc_hostname(const char *unc, const char **h, size_t *len); int copy_path_name(char *dst, const char *src); diff --git a/fs/cifs/connect.c b/fs/cifs/connect.c index d46702f5a663..fa29c9aae24b 100644 --- a/fs/cifs/connect.c +++ b/fs/cifs/connect.c @@ -97,6 +97,10 @@ static int reconn_set_ipaddr_from_hostname(struct TCP_Server_Info *server) if (!server->hostname) return -EINVAL; + /* if server hostname isn't populated, there's nothing to do here */ + if (server->hostname[0] == '\0') + return 0; + len = strlen(server->hostname) + 3; unc = kmalloc(len, GFP_KERNEL); @@ -141,6 +145,25 @@ requeue_resolve: return rc; } +static void smb2_query_server_interfaces(struct work_struct *work) +{ + int rc; + struct cifs_tcon *tcon = container_of(work, + struct cifs_tcon, + query_interfaces.work); + + /* + * query server network interfaces, in case they change + */ + rc = SMB3_request_interfaces(0, tcon); + if (rc) { + cifs_dbg(FYI, "%s: failed to query server interfaces: %d\n", + __func__, rc); + } + + queue_delayed_work(cifsiod_wq, &tcon->query_interfaces, + (SMB_INTERFACE_POLL_INTERVAL * HZ)); +} static void cifs_resolve_server(struct work_struct *work) { @@ -213,7 +236,7 @@ cifs_mark_tcp_ses_conns_for_reconnect(struct TCP_Server_Info *server, bool mark_smb_session) { struct TCP_Server_Info *pserver; - struct cifs_ses *ses; + struct cifs_ses *ses, *nses; struct cifs_tcon *tcon; /* @@ -227,7 +250,20 @@ cifs_mark_tcp_ses_conns_for_reconnect(struct TCP_Server_Info *server, spin_lock(&cifs_tcp_ses_lock); - list_for_each_entry(ses, &pserver->smb_ses_list, smb_ses_list) { + list_for_each_entry_safe(ses, nses, &pserver->smb_ses_list, smb_ses_list) { + /* check if iface is still active */ + if (!cifs_chan_is_iface_active(ses, server)) { + /* + * HACK: drop the lock before calling + * cifs_chan_update_iface to avoid deadlock + */ + ses->ses_count++; + spin_unlock(&cifs_tcp_ses_lock); + cifs_chan_update_iface(ses, server); + spin_lock(&cifs_tcp_ses_lock); + ses->ses_count--; + } + spin_lock(&ses->chan_lock); if (!mark_smb_session && cifs_chan_needs_reconnect(ses, server)) goto next_session; @@ -1890,9 +1926,11 @@ void cifs_put_smb_ses(struct cifs_ses *ses) int i; for (i = 1; i < chan_count; i++) { - spin_unlock(&ses->chan_lock); + if (ses->chans[i].iface) { + kref_put(&ses->chans[i].iface->refcount, release_iface); + ses->chans[i].iface = NULL; + } cifs_put_tcp_session(ses->chans[i].server, 0); - spin_lock(&ses->chan_lock); ses->chans[i].server = NULL; } } @@ -2266,6 +2304,9 @@ cifs_put_tcon(struct cifs_tcon *tcon) list_del_init(&tcon->tcon_list); spin_unlock(&cifs_tcp_ses_lock); + /* cancel polling of interfaces */ + cancel_delayed_work_sync(&tcon->query_interfaces); + if (tcon->use_witness) { int rc; @@ -2503,6 +2544,12 @@ cifs_get_tcon(struct cifs_ses *ses, struct smb3_fs_context *ctx) tcon->local_lease = ctx->local_lease; INIT_LIST_HEAD(&tcon->pending_opens); + /* schedule query interfaces poll */ + INIT_DELAYED_WORK(&tcon->query_interfaces, + smb2_query_server_interfaces); + queue_delayed_work(cifsiod_wq, &tcon->query_interfaces, + (SMB_INTERFACE_POLL_INTERVAL * HZ)); + spin_lock(&cifs_tcp_ses_lock); list_add(&tcon->tcon_list, &ses->tcon_list); spin_unlock(&cifs_tcp_ses_lock); @@ -3978,10 +4025,16 @@ cifs_setup_session(const unsigned int xid, struct cifs_ses *ses, struct nls_table *nls_info) { int rc = -ENOSYS; + struct sockaddr_in6 *addr6 = (struct sockaddr_in6 *)&server->dstaddr; + struct sockaddr_in *addr = (struct sockaddr_in *)&server->dstaddr; bool is_binding = false; - spin_lock(&cifs_tcp_ses_lock); + if (server->dstaddr.ss_family == AF_INET6) + scnprintf(ses->ip_addr, sizeof(ses->ip_addr), "%pI6", &addr6->sin6_addr); + else + scnprintf(ses->ip_addr, sizeof(ses->ip_addr), "%pI4", &addr->sin_addr); + if (ses->ses_status != SES_GOOD && ses->ses_status != SES_NEW && ses->ses_status != SES_NEED_RECON) { diff --git a/fs/cifs/file.c b/fs/cifs/file.c index 1618e0537d58..e64cda7a7610 100644 --- a/fs/cifs/file.c +++ b/fs/cifs/file.c @@ -2004,7 +2004,7 @@ struct cifsFileInfo *find_readable_file(struct cifsInodeInfo *cifs_inode, bool fsuid_only) { struct cifsFileInfo *open_file = NULL; - struct cifs_sb_info *cifs_sb = CIFS_SB(cifs_inode->vfs_inode.i_sb); + struct cifs_sb_info *cifs_sb = CIFS_SB(cifs_inode->netfs.inode.i_sb); /* only filter by fsuid on multiuser mounts */ if (!(cifs_sb->mnt_cifs_flags & CIFS_MOUNT_MULTIUSER)) @@ -2060,7 +2060,7 @@ cifs_get_writable_file(struct cifsInodeInfo *cifs_inode, int flags, return rc; } - cifs_sb = CIFS_SB(cifs_inode->vfs_inode.i_sb); + cifs_sb = CIFS_SB(cifs_inode->netfs.inode.i_sb); /* only filter by fsuid on multiuser mounts */ if (!(cifs_sb->mnt_cifs_flags & CIFS_MOUNT_MULTIUSER)) @@ -4669,14 +4669,14 @@ bool is_size_safe_to_change(struct cifsInodeInfo *cifsInode, __u64 end_of_file) /* This inode is open for write at least once */ struct cifs_sb_info *cifs_sb; - cifs_sb = CIFS_SB(cifsInode->vfs_inode.i_sb); + cifs_sb = CIFS_SB(cifsInode->netfs.inode.i_sb); if (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_DIRECT_IO) { /* since no page cache to corrupt on directio we can change size safely */ return true; } - if (i_size_read(&cifsInode->vfs_inode) < end_of_file) + if (i_size_read(&cifsInode->netfs.inode) < end_of_file) return true; return false; diff --git a/fs/cifs/fscache.c b/fs/cifs/fscache.c index a638b29e9062..23ef56f55ce5 100644 --- a/fs/cifs/fscache.c +++ b/fs/cifs/fscache.c @@ -101,13 +101,13 @@ void cifs_fscache_get_inode_cookie(struct inode *inode) struct cifs_sb_info *cifs_sb = CIFS_SB(inode->i_sb); struct cifs_tcon *tcon = cifs_sb_master_tcon(cifs_sb); - cifs_fscache_fill_coherency(&cifsi->vfs_inode, &cd); + cifs_fscache_fill_coherency(&cifsi->netfs.inode, &cd); - cifsi->netfs_ctx.cache = + cifsi->netfs.cache = fscache_acquire_cookie(tcon->fscache, 0, &cifsi->uniqueid, sizeof(cifsi->uniqueid), &cd, sizeof(cd), - i_size_read(&cifsi->vfs_inode)); + i_size_read(&cifsi->netfs.inode)); } void cifs_fscache_unuse_inode_cookie(struct inode *inode, bool update) @@ -131,7 +131,7 @@ void cifs_fscache_release_inode_cookie(struct inode *inode) if (cookie) { cifs_dbg(FYI, "%s: (0x%p)\n", __func__, cookie); fscache_relinquish_cookie(cookie, false); - cifsi->netfs_ctx.cache = NULL; + cifsi->netfs.cache = NULL; } } diff --git a/fs/cifs/fscache.h b/fs/cifs/fscache.h index 52355c0912ae..aa3b941a5555 100644 --- a/fs/cifs/fscache.h +++ b/fs/cifs/fscache.h @@ -52,16 +52,16 @@ void cifs_fscache_fill_coherency(struct inode *inode, struct cifsInodeInfo *cifsi = CIFS_I(inode); memset(cd, 0, sizeof(*cd)); - cd->last_write_time_sec = cpu_to_le64(cifsi->vfs_inode.i_mtime.tv_sec); - cd->last_write_time_nsec = cpu_to_le32(cifsi->vfs_inode.i_mtime.tv_nsec); - cd->last_change_time_sec = cpu_to_le64(cifsi->vfs_inode.i_ctime.tv_sec); - cd->last_change_time_nsec = cpu_to_le32(cifsi->vfs_inode.i_ctime.tv_nsec); + cd->last_write_time_sec = cpu_to_le64(cifsi->netfs.inode.i_mtime.tv_sec); + cd->last_write_time_nsec = cpu_to_le32(cifsi->netfs.inode.i_mtime.tv_nsec); + cd->last_change_time_sec = cpu_to_le64(cifsi->netfs.inode.i_ctime.tv_sec); + cd->last_change_time_nsec = cpu_to_le32(cifsi->netfs.inode.i_ctime.tv_nsec); } static inline struct fscache_cookie *cifs_inode_cookie(struct inode *inode) { - return netfs_i_cookie(inode); + return netfs_i_cookie(&CIFS_I(inode)->netfs); } static inline void cifs_invalidate_cache(struct inode *inode, unsigned int flags) diff --git a/fs/cifs/inode.c b/fs/cifs/inode.c index 2f9e7d2f81b6..81da81e18553 100644 --- a/fs/cifs/inode.c +++ b/fs/cifs/inode.c @@ -115,7 +115,7 @@ cifs_revalidate_cache(struct inode *inode, struct cifs_fattr *fattr) __func__, cifs_i->uniqueid); set_bit(CIFS_INO_INVALID_MAPPING, &cifs_i->flags); /* Invalidate fscache cookie */ - cifs_fscache_fill_coherency(&cifs_i->vfs_inode, &cd); + cifs_fscache_fill_coherency(&cifs_i->netfs.inode, &cd); fscache_invalidate(cifs_inode_cookie(inode), &cd, i_size_read(inode), 0); } @@ -2499,7 +2499,7 @@ int cifs_fiemap(struct inode *inode, struct fiemap_extent_info *fei, u64 start, u64 len) { struct cifsInodeInfo *cifs_i = CIFS_I(inode); - struct cifs_sb_info *cifs_sb = CIFS_SB(cifs_i->vfs_inode.i_sb); + struct cifs_sb_info *cifs_sb = CIFS_SB(cifs_i->netfs.inode.i_sb); struct cifs_tcon *tcon = cifs_sb_master_tcon(cifs_sb); struct TCP_Server_Info *server = tcon->ses->server; struct cifsFileInfo *cfile; diff --git a/fs/cifs/misc.c b/fs/cifs/misc.c index 35962a1a23b9..0e84e6fcf8ab 100644 --- a/fs/cifs/misc.c +++ b/fs/cifs/misc.c @@ -75,6 +75,7 @@ sesInfoAlloc(void) INIT_LIST_HEAD(&ret_buf->tcon_list); mutex_init(&ret_buf->session_mutex); spin_lock_init(&ret_buf->iface_lock); + INIT_LIST_HEAD(&ret_buf->iface_list); spin_lock_init(&ret_buf->chan_lock); } return ret_buf; @@ -83,6 +84,8 @@ sesInfoAlloc(void) void sesInfoFree(struct cifs_ses *buf_to_free) { + struct cifs_server_iface *iface = NULL, *niface = NULL; + if (buf_to_free == NULL) { cifs_dbg(FYI, "Null buffer passed to sesInfoFree\n"); return; @@ -96,7 +99,11 @@ sesInfoFree(struct cifs_ses *buf_to_free) kfree(buf_to_free->user_name); kfree(buf_to_free->domainName); kfree_sensitive(buf_to_free->auth_key.response); - kfree(buf_to_free->iface_list); + spin_lock(&buf_to_free->iface_lock); + list_for_each_entry_safe(iface, niface, &buf_to_free->iface_list, + iface_head) + kref_put(&iface->refcount, release_iface); + spin_unlock(&buf_to_free->iface_lock); kfree_sensitive(buf_to_free); } @@ -537,11 +544,11 @@ void cifs_set_oplock_level(struct cifsInodeInfo *cinode, __u32 oplock) if (oplock == OPLOCK_EXCLUSIVE) { cinode->oplock = CIFS_CACHE_WRITE_FLG | CIFS_CACHE_READ_FLG; cifs_dbg(FYI, "Exclusive Oplock granted on inode %p\n", - &cinode->vfs_inode); + &cinode->netfs.inode); } else if (oplock == OPLOCK_READ) { cinode->oplock = CIFS_CACHE_READ_FLG; cifs_dbg(FYI, "Level II Oplock granted on inode %p\n", - &cinode->vfs_inode); + &cinode->netfs.inode); } else cinode->oplock = 0; } @@ -1211,18 +1218,23 @@ static struct super_block *__cifs_get_super(void (*f)(struct super_block *, void .data = data, .sb = NULL, }; + struct file_system_type **fs_type = (struct file_system_type *[]) { + &cifs_fs_type, &smb3_fs_type, NULL, + }; - iterate_supers_type(&cifs_fs_type, f, &sd); - - if (!sd.sb) - return ERR_PTR(-EINVAL); - /* - * Grab an active reference in order to prevent automounts (DFS links) - * of expiring and then freeing up our cifs superblock pointer while - * we're doing failover. - */ - cifs_sb_active(sd.sb); - return sd.sb; + for (; *fs_type; fs_type++) { + iterate_supers_type(*fs_type, f, &sd); + if (sd.sb) { + /* + * Grab an active reference in order to prevent automounts (DFS links) + * of expiring and then freeing up our cifs superblock pointer while + * we're doing failover. + */ + cifs_sb_active(sd.sb); + return sd.sb; + } + } + return ERR_PTR(-EINVAL); } static void __cifs_put_super(struct super_block *sb) diff --git a/fs/cifs/sess.c b/fs/cifs/sess.c index 3b7915af1f62..b85718f32b53 100644 --- a/fs/cifs/sess.c +++ b/fs/cifs/sess.c @@ -58,7 +58,7 @@ bool is_ses_using_iface(struct cifs_ses *ses, struct cifs_server_iface *iface) spin_lock(&ses->chan_lock); for (i = 0; i < ses->chan_count; i++) { - if (is_server_using_iface(ses->chans[i].server, iface)) { + if (ses->chans[i].iface == iface) { spin_unlock(&ses->chan_lock); return true; } @@ -81,6 +81,9 @@ cifs_ses_get_chan_index(struct cifs_ses *ses, } /* If we didn't find the channel, it is likely a bug */ + if (server) + cifs_dbg(VFS, "unable to get chan index for server: 0x%llx", + server->conn_id); WARN_ON(1); return 0; } @@ -143,16 +146,24 @@ cifs_chan_needs_reconnect(struct cifs_ses *ses, return CIFS_CHAN_NEEDS_RECONNECT(ses, chan_index); } +bool +cifs_chan_is_iface_active(struct cifs_ses *ses, + struct TCP_Server_Info *server) +{ + unsigned int chan_index = cifs_ses_get_chan_index(ses, server); + + return ses->chans[chan_index].iface && + ses->chans[chan_index].iface->is_active; +} + /* returns number of channels added */ int cifs_try_adding_channels(struct cifs_sb_info *cifs_sb, struct cifs_ses *ses) { int old_chan_count, new_chan_count; int left; - int i = 0; int rc = 0; int tries = 0; - struct cifs_server_iface *ifaces = NULL; - size_t iface_count; + struct cifs_server_iface *iface = NULL, *niface = NULL; spin_lock(&ses->chan_lock); @@ -182,32 +193,16 @@ int cifs_try_adding_channels(struct cifs_sb_info *cifs_sb, struct cifs_ses *ses) spin_unlock(&ses->chan_lock); /* - * Make a copy of the iface list at the time and use that - * instead so as to not hold the iface spinlock for opening - * channels - */ - spin_lock(&ses->iface_lock); - iface_count = ses->iface_count; - if (iface_count <= 0) { - spin_unlock(&ses->iface_lock); - cifs_dbg(VFS, "no iface list available to open channels\n"); - return 0; - } - ifaces = kmemdup(ses->iface_list, iface_count*sizeof(*ifaces), - GFP_ATOMIC); - if (!ifaces) { - spin_unlock(&ses->iface_lock); - return 0; - } - spin_unlock(&ses->iface_lock); - - /* * Keep connecting to same, fastest, iface for all channels as * long as its RSS. Try next fastest one if not RSS or channel * creation fails. */ + spin_lock(&ses->iface_lock); + iface = list_first_entry(&ses->iface_list, struct cifs_server_iface, + iface_head); + spin_unlock(&ses->iface_lock); + while (left > 0) { - struct cifs_server_iface *iface; tries++; if (tries > 3*ses->chan_max) { @@ -216,31 +211,128 @@ int cifs_try_adding_channels(struct cifs_sb_info *cifs_sb, struct cifs_ses *ses) break; } - iface = &ifaces[i]; - if (is_ses_using_iface(ses, iface) && !iface->rss_capable) { - i = (i+1) % iface_count; - continue; + spin_lock(&ses->iface_lock); + if (!ses->iface_count) { + spin_unlock(&ses->iface_lock); + break; } - rc = cifs_ses_add_channel(cifs_sb, ses, iface); - if (rc) { - cifs_dbg(FYI, "failed to open extra channel on iface#%d rc=%d\n", - i, rc); - i = (i+1) % iface_count; - continue; + list_for_each_entry_safe_from(iface, niface, &ses->iface_list, + iface_head) { + /* skip ifaces that are unusable */ + if (!iface->is_active || + (is_ses_using_iface(ses, iface) && + !iface->rss_capable)) { + continue; + } + + /* take ref before unlock */ + kref_get(&iface->refcount); + + spin_unlock(&ses->iface_lock); + rc = cifs_ses_add_channel(cifs_sb, ses, iface); + spin_lock(&ses->iface_lock); + + if (rc) { + cifs_dbg(VFS, "failed to open extra channel on iface:%pIS rc=%d\n", + &iface->sockaddr, + rc); + kref_put(&iface->refcount, release_iface); + continue; + } + + cifs_dbg(FYI, "successfully opened new channel on iface:%pIS\n", + &iface->sockaddr); + break; } + spin_unlock(&ses->iface_lock); - cifs_dbg(FYI, "successfully opened new channel on iface#%d\n", - i); left--; new_chan_count++; } - kfree(ifaces); return new_chan_count - old_chan_count; } /* + * update the iface for the channel if necessary. + * will return 0 when iface is updated, 1 if removed, 2 otherwise + * Must be called with chan_lock held. + */ +int +cifs_chan_update_iface(struct cifs_ses *ses, struct TCP_Server_Info *server) +{ + unsigned int chan_index; + struct cifs_server_iface *iface = NULL; + struct cifs_server_iface *old_iface = NULL; + int rc = 0; + + spin_lock(&ses->chan_lock); + chan_index = cifs_ses_get_chan_index(ses, server); + if (!chan_index) { + spin_unlock(&ses->chan_lock); + return 0; + } + + if (ses->chans[chan_index].iface) { + old_iface = ses->chans[chan_index].iface; + if (old_iface->is_active) { + spin_unlock(&ses->chan_lock); + return 1; + } + } + spin_unlock(&ses->chan_lock); + + spin_lock(&ses->iface_lock); + /* then look for a new one */ + list_for_each_entry(iface, &ses->iface_list, iface_head) { + if (!iface->is_active || + (is_ses_using_iface(ses, iface) && + !iface->rss_capable)) { + continue; + } + kref_get(&iface->refcount); + } + + if (!list_entry_is_head(iface, &ses->iface_list, iface_head)) { + rc = 1; + iface = NULL; + cifs_dbg(FYI, "unable to find a suitable iface\n"); + } + + /* now drop the ref to the current iface */ + if (old_iface && iface) { + kref_put(&old_iface->refcount, release_iface); + cifs_dbg(FYI, "replacing iface: %pIS with %pIS\n", + &old_iface->sockaddr, + &iface->sockaddr); + } else if (old_iface) { + kref_put(&old_iface->refcount, release_iface); + cifs_dbg(FYI, "releasing ref to iface: %pIS\n", + &old_iface->sockaddr); + } else { + WARN_ON(!iface); + cifs_dbg(FYI, "adding new iface: %pIS\n", &iface->sockaddr); + } + spin_unlock(&ses->iface_lock); + + spin_lock(&ses->chan_lock); + chan_index = cifs_ses_get_chan_index(ses, server); + ses->chans[chan_index].iface = iface; + + /* No iface is found. if secondary chan, drop connection */ + if (!iface && CIFS_SERVER_IS_CHAN(server)) + ses->chans[chan_index].server = NULL; + + spin_unlock(&ses->chan_lock); + + if (!iface && CIFS_SERVER_IS_CHAN(server)) + cifs_put_tcp_session(server, false); + + return rc; +} + +/* * If server is a channel of ses, return the corresponding enclosing * cifs_chan otherwise return NULL. */ @@ -301,7 +393,10 @@ cifs_ses_add_channel(struct cifs_sb_info *cifs_sb, struct cifs_ses *ses, /* Auth */ ctx.domainauto = ses->domainAuto; ctx.domainname = ses->domainName; - ctx.server_hostname = ses->server->hostname; + + /* no hostname for extra channels */ + ctx.server_hostname = ""; + ctx.username = ses->user_name; ctx.password = ses->password; ctx.sectype = ses->sectype; @@ -349,6 +444,7 @@ cifs_ses_add_channel(struct cifs_sb_info *cifs_sb, struct cifs_ses *ses, spin_unlock(&ses->chan_lock); goto out; } + chan->iface = iface; ses->chan_count++; atomic_set(&ses->chan_seq, 0); diff --git a/fs/cifs/smb2ops.c b/fs/cifs/smb2ops.c index 98a76fa791c0..8802995b2d3d 100644 --- a/fs/cifs/smb2ops.c +++ b/fs/cifs/smb2ops.c @@ -512,73 +512,41 @@ smb3_negotiate_rsize(struct cifs_tcon *tcon, struct smb3_fs_context *ctx) static int parse_server_interfaces(struct network_interface_info_ioctl_rsp *buf, size_t buf_len, - struct cifs_server_iface **iface_list, - size_t *iface_count) + struct cifs_ses *ses) { struct network_interface_info_ioctl_rsp *p; struct sockaddr_in *addr4; struct sockaddr_in6 *addr6; struct iface_info_ipv4 *p4; struct iface_info_ipv6 *p6; - struct cifs_server_iface *info; + struct cifs_server_iface *info = NULL, *iface = NULL, *niface = NULL; + struct cifs_server_iface tmp_iface; ssize_t bytes_left; size_t next = 0; int nb_iface = 0; - int rc = 0; - - *iface_list = NULL; - *iface_count = 0; - - /* - * Fist pass: count and sanity check - */ + int rc = 0, ret = 0; bytes_left = buf_len; p = buf; - while (bytes_left >= sizeof(*p)) { - nb_iface++; - next = le32_to_cpu(p->Next); - if (!next) { - bytes_left -= sizeof(*p); - break; - } - p = (struct network_interface_info_ioctl_rsp *)((u8 *)p+next); - bytes_left -= next; - } - - if (!nb_iface) { - cifs_dbg(VFS, "%s: malformed interface info\n", __func__); - rc = -EINVAL; - goto out; - } - - /* Azure rounds the buffer size up 8, to a 16 byte boundary */ - if ((bytes_left > 8) || p->Next) - cifs_dbg(VFS, "%s: incomplete interface info\n", __func__); - + spin_lock(&ses->iface_lock); /* - * Second pass: extract info to internal structure + * Go through iface_list and do kref_put to remove + * any unused ifaces. ifaces in use will be removed + * when the last user calls a kref_put on it */ - - *iface_list = kcalloc(nb_iface, sizeof(**iface_list), GFP_KERNEL); - if (!*iface_list) { - rc = -ENOMEM; - goto out; + list_for_each_entry_safe(iface, niface, &ses->iface_list, + iface_head) { + iface->is_active = 0; + kref_put(&iface->refcount, release_iface); } + spin_unlock(&ses->iface_lock); - info = *iface_list; - bytes_left = buf_len; - p = buf; while (bytes_left >= sizeof(*p)) { - info->speed = le64_to_cpu(p->LinkSpeed); - info->rdma_capable = le32_to_cpu(p->Capability & RDMA_CAPABLE) ? 1 : 0; - info->rss_capable = le32_to_cpu(p->Capability & RSS_CAPABLE) ? 1 : 0; - - cifs_dbg(FYI, "%s: adding iface %zu\n", __func__, *iface_count); - cifs_dbg(FYI, "%s: speed %zu bps\n", __func__, info->speed); - cifs_dbg(FYI, "%s: capabilities 0x%08x\n", __func__, - le32_to_cpu(p->Capability)); + memset(&tmp_iface, 0, sizeof(tmp_iface)); + tmp_iface.speed = le64_to_cpu(p->LinkSpeed); + tmp_iface.rdma_capable = le32_to_cpu(p->Capability & RDMA_CAPABLE) ? 1 : 0; + tmp_iface.rss_capable = le32_to_cpu(p->Capability & RSS_CAPABLE) ? 1 : 0; switch (p->Family) { /* @@ -587,7 +555,7 @@ parse_server_interfaces(struct network_interface_info_ioctl_rsp *buf, * conversion explicit in case either one changes. */ case INTERNETWORK: - addr4 = (struct sockaddr_in *)&info->sockaddr; + addr4 = (struct sockaddr_in *)&tmp_iface.sockaddr; p4 = (struct iface_info_ipv4 *)p->Buffer; addr4->sin_family = AF_INET; memcpy(&addr4->sin_addr, &p4->IPv4Address, 4); @@ -599,7 +567,7 @@ parse_server_interfaces(struct network_interface_info_ioctl_rsp *buf, &addr4->sin_addr); break; case INTERNETWORKV6: - addr6 = (struct sockaddr_in6 *)&info->sockaddr; + addr6 = (struct sockaddr_in6 *)&tmp_iface.sockaddr; p6 = (struct iface_info_ipv6 *)p->Buffer; addr6->sin6_family = AF_INET6; memcpy(&addr6->sin6_addr, &p6->IPv6Address, 16); @@ -619,46 +587,96 @@ parse_server_interfaces(struct network_interface_info_ioctl_rsp *buf, goto next_iface; } - (*iface_count)++; - info++; + /* + * The iface_list is assumed to be sorted by speed. + * Check if the new interface exists in that list. + * NEVER change iface. it could be in use. + * Add a new one instead + */ + spin_lock(&ses->iface_lock); + iface = niface = NULL; + list_for_each_entry_safe(iface, niface, &ses->iface_list, + iface_head) { + ret = iface_cmp(iface, &tmp_iface); + if (!ret) { + /* just get a ref so that it doesn't get picked/freed */ + iface->is_active = 1; + kref_get(&iface->refcount); + spin_unlock(&ses->iface_lock); + goto next_iface; + } else if (ret < 0) { + /* all remaining ifaces are slower */ + kref_get(&iface->refcount); + break; + } + } + spin_unlock(&ses->iface_lock); + + /* no match. insert the entry in the list */ + info = kmalloc(sizeof(struct cifs_server_iface), + GFP_KERNEL); + if (!info) { + rc = -ENOMEM; + goto out; + } + memcpy(info, &tmp_iface, sizeof(tmp_iface)); + + /* add this new entry to the list */ + kref_init(&info->refcount); + info->is_active = 1; + + cifs_dbg(FYI, "%s: adding iface %zu\n", __func__, ses->iface_count); + cifs_dbg(FYI, "%s: speed %zu bps\n", __func__, info->speed); + cifs_dbg(FYI, "%s: capabilities 0x%08x\n", __func__, + le32_to_cpu(p->Capability)); + + spin_lock(&ses->iface_lock); + if (!list_entry_is_head(iface, &ses->iface_list, iface_head)) { + list_add_tail(&info->iface_head, &iface->iface_head); + kref_put(&iface->refcount, release_iface); + } else + list_add_tail(&info->iface_head, &ses->iface_list); + spin_unlock(&ses->iface_lock); + + ses->iface_count++; + ses->iface_last_update = jiffies; next_iface: + nb_iface++; next = le32_to_cpu(p->Next); - if (!next) + if (!next) { + bytes_left -= sizeof(*p); break; + } p = (struct network_interface_info_ioctl_rsp *)((u8 *)p+next); bytes_left -= next; } - if (!*iface_count) { + if (!nb_iface) { + cifs_dbg(VFS, "%s: malformed interface info\n", __func__); rc = -EINVAL; goto out; } -out: - if (rc) { - kfree(*iface_list); - *iface_count = 0; - *iface_list = NULL; - } - return rc; -} + /* Azure rounds the buffer size up 8, to a 16 byte boundary */ + if ((bytes_left > 8) || p->Next) + cifs_dbg(VFS, "%s: incomplete interface info\n", __func__); -static int compare_iface(const void *ia, const void *ib) -{ - const struct cifs_server_iface *a = (struct cifs_server_iface *)ia; - const struct cifs_server_iface *b = (struct cifs_server_iface *)ib; - return a->speed == b->speed ? 0 : (a->speed > b->speed ? -1 : 1); + if (!ses->iface_count) { + rc = -EINVAL; + goto out; + } + +out: + return rc; } -static int +int SMB3_request_interfaces(const unsigned int xid, struct cifs_tcon *tcon) { int rc; unsigned int ret_data_len = 0; struct network_interface_info_ioctl_rsp *out_buf = NULL; - struct cifs_server_iface *iface_list; - size_t iface_count; struct cifs_ses *ses = tcon->ses; rc = SMB2_ioctl(xid, tcon, NO_FILE_ID, NO_FILE_ID, @@ -674,21 +692,10 @@ SMB3_request_interfaces(const unsigned int xid, struct cifs_tcon *tcon) goto out; } - rc = parse_server_interfaces(out_buf, ret_data_len, - &iface_list, &iface_count); + rc = parse_server_interfaces(out_buf, ret_data_len, ses); if (rc) goto out; - /* sort interfaces from fastest to slowest */ - sort(iface_list, iface_count, sizeof(*iface_list), compare_iface, NULL); - - spin_lock(&ses->iface_lock); - kfree(ses->iface_list); - ses->iface_list = iface_list; - ses->iface_count = iface_count; - ses->iface_last_update = jiffies; - spin_unlock(&ses->iface_lock); - out: kfree(out_buf); return rc; @@ -4260,15 +4267,15 @@ smb2_set_oplock_level(struct cifsInodeInfo *cinode, __u32 oplock, if (oplock == SMB2_OPLOCK_LEVEL_BATCH) { cinode->oplock = CIFS_CACHE_RHW_FLG; cifs_dbg(FYI, "Batch Oplock granted on inode %p\n", - &cinode->vfs_inode); + &cinode->netfs.inode); } else if (oplock == SMB2_OPLOCK_LEVEL_EXCLUSIVE) { cinode->oplock = CIFS_CACHE_RW_FLG; cifs_dbg(FYI, "Exclusive Oplock granted on inode %p\n", - &cinode->vfs_inode); + &cinode->netfs.inode); } else if (oplock == SMB2_OPLOCK_LEVEL_II) { cinode->oplock = CIFS_CACHE_READ_FLG; cifs_dbg(FYI, "Level II Oplock granted on inode %p\n", - &cinode->vfs_inode); + &cinode->netfs.inode); } else cinode->oplock = 0; } @@ -4307,7 +4314,7 @@ smb21_set_oplock_level(struct cifsInodeInfo *cinode, __u32 oplock, cinode->oplock = new_oplock; cifs_dbg(FYI, "%s Lease granted on inode %p\n", message, - &cinode->vfs_inode); + &cinode->netfs.inode); } static void diff --git a/fs/cifs/smb2pdu.c b/fs/cifs/smb2pdu.c index 0e8c85249579..12b4dddaedb0 100644 --- a/fs/cifs/smb2pdu.c +++ b/fs/cifs/smb2pdu.c @@ -288,6 +288,9 @@ smb2_reconnect(__le16 smb2_command, struct cifs_tcon *tcon, mutex_unlock(&ses->session_mutex); rc = -EHOSTDOWN; goto failed; + } else if (rc) { + mutex_unlock(&ses->session_mutex); + goto out; } } else { mutex_unlock(&ses->session_mutex); @@ -540,6 +543,7 @@ assemble_neg_contexts(struct smb2_negotiate_req *req, struct TCP_Server_Info *server, unsigned int *total_len) { char *pneg_ctxt; + char *hostname = NULL; unsigned int ctxt_len, neg_context_count; if (*total_len > 200) { @@ -567,16 +571,24 @@ assemble_neg_contexts(struct smb2_negotiate_req *req, *total_len += ctxt_len; pneg_ctxt += ctxt_len; - ctxt_len = build_netname_ctxt((struct smb2_netname_neg_context *)pneg_ctxt, - server->hostname); - *total_len += ctxt_len; - pneg_ctxt += ctxt_len; - build_posix_ctxt((struct smb2_posix_neg_context *)pneg_ctxt); *total_len += sizeof(struct smb2_posix_neg_context); pneg_ctxt += sizeof(struct smb2_posix_neg_context); - neg_context_count = 4; + /* + * secondary channels don't have the hostname field populated + * use the hostname field in the primary channel instead + */ + hostname = CIFS_SERVER_IS_CHAN(server) ? + server->primary_server->hostname : server->hostname; + if (hostname && (hostname[0] != 0)) { + ctxt_len = build_netname_ctxt((struct smb2_netname_neg_context *)pneg_ctxt, + hostname); + *total_len += ctxt_len; + pneg_ctxt += ctxt_len; + neg_context_count = 4; + } else /* second channels do not have a hostname */ + neg_context_count = 3; if (server->compress_algorithm) { build_compression_ctxt((struct smb2_compression_capabilities_context *) @@ -5151,6 +5163,8 @@ SMB2_set_eof(const unsigned int xid, struct cifs_tcon *tcon, u64 persistent_fid, data = &info; size = sizeof(struct smb2_file_eof_info); + trace_smb3_set_eof(xid, persistent_fid, tcon->tid, tcon->ses->Suid, le64_to_cpu(*eof)); + return send_set_info(xid, tcon, persistent_fid, volatile_fid, pid, FILE_END_OF_FILE_INFORMATION, SMB2_O_INFO_FILE, 0, 1, &data, &size); diff --git a/fs/cifs/trace.h b/fs/cifs/trace.h index 2be5e0c8564d..6b88dc2e364f 100644 --- a/fs/cifs/trace.h +++ b/fs/cifs/trace.h @@ -121,6 +121,44 @@ DEFINE_SMB3_RW_DONE_EVENT(query_dir_done); DEFINE_SMB3_RW_DONE_EVENT(zero_done); DEFINE_SMB3_RW_DONE_EVENT(falloc_done); +/* For logging successful set EOF (truncate) */ +DECLARE_EVENT_CLASS(smb3_eof_class, + TP_PROTO(unsigned int xid, + __u64 fid, + __u32 tid, + __u64 sesid, + __u64 offset), + TP_ARGS(xid, fid, tid, sesid, offset), + TP_STRUCT__entry( + __field(unsigned int, xid) + __field(__u64, fid) + __field(__u32, tid) + __field(__u64, sesid) + __field(__u64, offset) + ), + TP_fast_assign( + __entry->xid = xid; + __entry->fid = fid; + __entry->tid = tid; + __entry->sesid = sesid; + __entry->offset = offset; + ), + TP_printk("xid=%u sid=0x%llx tid=0x%x fid=0x%llx offset=0x%llx", + __entry->xid, __entry->sesid, __entry->tid, __entry->fid, + __entry->offset) +) + +#define DEFINE_SMB3_EOF_EVENT(name) \ +DEFINE_EVENT(smb3_eof_class, smb3_##name, \ + TP_PROTO(unsigned int xid, \ + __u64 fid, \ + __u32 tid, \ + __u64 sesid, \ + __u64 offset), \ + TP_ARGS(xid, fid, tid, sesid, offset)) + +DEFINE_SMB3_EOF_EVENT(set_eof); + /* * For handle based calls other than read and write, and get/set info */ diff --git a/fs/exfat/namei.c b/fs/exfat/namei.c index 76acc3721951..c6eaf7e9ea74 100644 --- a/fs/exfat/namei.c +++ b/fs/exfat/namei.c @@ -1198,7 +1198,9 @@ static int __exfat_rename(struct inode *old_parent_inode, return -ENOENT; } - exfat_chain_dup(&olddir, &ei->dir); + exfat_chain_set(&olddir, EXFAT_I(old_parent_inode)->start_clu, + EXFAT_B_TO_CLU_ROUND_UP(i_size_read(old_parent_inode), sbi), + EXFAT_I(old_parent_inode)->flags); dentry = ei->entry; ep = exfat_get_dentry(sb, &olddir, dentry, &old_bh); diff --git a/fs/ext2/dir.c b/fs/ext2/dir.c index 2c2f179b6977..43de293cef56 100644 --- a/fs/ext2/dir.c +++ b/fs/ext2/dir.c @@ -672,17 +672,14 @@ int ext2_empty_dir (struct inode * inode) void *page_addr = NULL; struct page *page = NULL; unsigned long i, npages = dir_pages(inode); - int dir_has_error = 0; for (i = 0; i < npages; i++) { char *kaddr; ext2_dirent * de; - page = ext2_get_page(inode, i, dir_has_error, &page_addr); + page = ext2_get_page(inode, i, 0, &page_addr); - if (IS_ERR(page)) { - dir_has_error = 1; - continue; - } + if (IS_ERR(page)) + goto not_empty; kaddr = page_addr; de = (ext2_dirent *)kaddr; diff --git a/fs/ext2/inode.c b/fs/ext2/inode.c index 360ce3604a2d..e6b932219803 100644 --- a/fs/ext2/inode.c +++ b/fs/ext2/inode.c @@ -1549,7 +1549,7 @@ static int __ext2_write_inode(struct inode *inode, int do_sync) if (IS_ERR(raw_inode)) return -EIO; - /* For fields not not tracking in the in-memory inode, + /* For fields not tracking in the in-memory inode, * initialise them to zero for new inodes. */ if (ei->i_state & EXT2_STATE_NEW) memset(raw_inode, 0, EXT2_SB(sb)->s_inode_size); diff --git a/fs/ext4/inode.c b/fs/ext4/inode.c index 3dce7d058985..84c0eb55071d 100644 --- a/fs/ext4/inode.c +++ b/fs/ext4/inode.c @@ -829,7 +829,7 @@ int ext4_get_block_unwritten(struct inode *inode, sector_t iblock, ext4_debug("ext4_get_block_unwritten: inode %lu, create flag %d\n", inode->i_ino, create); return _ext4_get_block(inode, iblock, bh_result, - EXT4_GET_BLOCKS_IO_CREATE_EXT); + EXT4_GET_BLOCKS_CREATE_UNWRIT_EXT); } /* Maximum number of blocks we map for direct IO at once. */ diff --git a/fs/ext4/mballoc.c b/fs/ext4/mballoc.c index 9f12f29bc346..9e06334771a3 100644 --- a/fs/ext4/mballoc.c +++ b/fs/ext4/mballoc.c @@ -4104,6 +4104,15 @@ ext4_mb_normalize_request(struct ext4_allocation_context *ac, size = size >> bsbits; start = start_off >> bsbits; + /* + * For tiny groups (smaller than 8MB) the chosen allocation + * alignment may be larger than group size. Make sure the + * alignment does not move allocation to a different group which + * makes mballoc fail assertions later. + */ + start = max(start, rounddown(ac->ac_o_ex.fe_logical, + (ext4_lblk_t)EXT4_BLOCKS_PER_GROUP(ac->ac_sb))); + /* don't cover already allocated blocks in selected range */ if (ar->pleft && start <= ar->lleft) { size -= ar->lleft + 1 - start; @@ -4176,7 +4185,22 @@ ext4_mb_normalize_request(struct ext4_allocation_context *ac, } rcu_read_unlock(); - if (start + size <= ac->ac_o_ex.fe_logical && + /* + * In this function "start" and "size" are normalized for better + * alignment and length such that we could preallocate more blocks. + * This normalization is done such that original request of + * ac->ac_o_ex.fe_logical & fe_len should always lie within "start" and + * "size" boundaries. + * (Note fe_len can be relaxed since FS block allocation API does not + * provide gurantee on number of contiguous blocks allocation since that + * depends upon free space left, etc). + * In case of inode pa, later we use the allocated blocks + * [pa_start + fe_logical - pa_lstart, fe_len/size] from the preallocated + * range of goal/best blocks [start, size] to put it at the + * ac_o_ex.fe_logical extent of this inode. + * (See ext4_mb_use_inode_pa() for more details) + */ + if (start + size <= ac->ac_o_ex.fe_logical || start > ac->ac_o_ex.fe_logical) { ext4_msg(ac->ac_sb, KERN_ERR, "start %lu, size %lu, fe_logical %lu", diff --git a/fs/ext4/migrate.c b/fs/ext4/migrate.c index 7a5353a8cfd7..42f590518b4c 100644 --- a/fs/ext4/migrate.c +++ b/fs/ext4/migrate.c @@ -438,7 +438,7 @@ int ext4_ext_migrate(struct inode *inode) /* * Worst case we can touch the allocation bitmaps and a block - * group descriptor block. We do need need to worry about + * group descriptor block. We do need to worry about * credits for modifying the quota inode. */ handle = ext4_journal_start(inode, EXT4_HT_MIGRATE, diff --git a/fs/ext4/namei.c b/fs/ext4/namei.c index 47d0ca4c795b..db4ba99d1ceb 100644 --- a/fs/ext4/namei.c +++ b/fs/ext4/namei.c @@ -1929,7 +1929,8 @@ static struct ext4_dir_entry_2 *do_split(handle_t *handle, struct inode *dir, struct dx_hash_info *hinfo) { unsigned blocksize = dir->i_sb->s_blocksize; - unsigned count, continued; + unsigned continued; + int count; struct buffer_head *bh2; ext4_lblk_t newblock; u32 hash2; diff --git a/fs/ext4/page-io.c b/fs/ext4/page-io.c index 14695e2b5042..97fa7b4c645f 100644 --- a/fs/ext4/page-io.c +++ b/fs/ext4/page-io.c @@ -465,7 +465,7 @@ int ext4_bio_write_page(struct ext4_io_submit *io, /* * In the first loop we prepare and mark buffers to submit. We have to * mark all buffers in the page before submitting so that - * end_page_writeback() cannot be called from ext4_bio_end_io() when IO + * end_page_writeback() cannot be called from ext4_end_bio() when IO * on the first buffer finishes and we are still working on submitting * the second buffer. */ diff --git a/fs/ext4/resize.c b/fs/ext4/resize.c index 90a941d20dff..8b70a4701293 100644 --- a/fs/ext4/resize.c +++ b/fs/ext4/resize.c @@ -54,6 +54,16 @@ int ext4_resize_begin(struct super_block *sb) return -EPERM; /* + * If the reserved GDT blocks is non-zero, the resize_inode feature + * should always be set. + */ + if (EXT4_SB(sb)->s_es->s_reserved_gdt_blocks && + !ext4_has_feature_resize_inode(sb)) { + ext4_error(sb, "resize_inode disabled but reserved GDT blocks non-zero"); + return -EFSCORRUPTED; + } + + /* * If we are not using the primary superblock/GDT copy don't resize, * because the user tools have no way of handling this. Probably a * bad time to do it anyways. diff --git a/fs/ext4/super.c b/fs/ext4/super.c index 450c918d68fc..845f2f8aee5f 100644 --- a/fs/ext4/super.c +++ b/fs/ext4/super.c @@ -87,7 +87,7 @@ static struct inode *ext4_get_journal_inode(struct super_block *sb, static int ext4_validate_options(struct fs_context *fc); static int ext4_check_opt_consistency(struct fs_context *fc, struct super_block *sb); -static int ext4_apply_options(struct fs_context *fc, struct super_block *sb); +static void ext4_apply_options(struct fs_context *fc, struct super_block *sb); static int ext4_parse_param(struct fs_context *fc, struct fs_parameter *param); static int ext4_get_tree(struct fs_context *fc); static int ext4_reconfigure(struct fs_context *fc); @@ -1870,31 +1870,12 @@ ext4_sb_read_encoding(const struct ext4_super_block *es) } #endif -static int ext4_set_test_dummy_encryption(struct super_block *sb, char *arg) -{ -#ifdef CONFIG_FS_ENCRYPTION - struct ext4_sb_info *sbi = EXT4_SB(sb); - int err; - - err = fscrypt_set_test_dummy_encryption(sb, arg, - &sbi->s_dummy_enc_policy); - if (err) { - ext4_msg(sb, KERN_WARNING, - "Error while setting test dummy encryption [%d]", err); - return err; - } - ext4_msg(sb, KERN_WARNING, "Test dummy encryption mode enabled"); -#endif - return 0; -} - #define EXT4_SPEC_JQUOTA (1 << 0) #define EXT4_SPEC_JQFMT (1 << 1) #define EXT4_SPEC_DATAJ (1 << 2) #define EXT4_SPEC_SB_BLOCK (1 << 3) #define EXT4_SPEC_JOURNAL_DEV (1 << 4) #define EXT4_SPEC_JOURNAL_IOPRIO (1 << 5) -#define EXT4_SPEC_DUMMY_ENCRYPTION (1 << 6) #define EXT4_SPEC_s_want_extra_isize (1 << 7) #define EXT4_SPEC_s_max_batch_time (1 << 8) #define EXT4_SPEC_s_min_batch_time (1 << 9) @@ -1911,7 +1892,7 @@ static int ext4_set_test_dummy_encryption(struct super_block *sb, char *arg) struct ext4_fs_context { char *s_qf_names[EXT4_MAXQUOTAS]; - char *test_dummy_enc_arg; + struct fscrypt_dummy_policy dummy_enc_policy; int s_jquota_fmt; /* Format of quota to use */ #ifdef CONFIG_EXT4_DEBUG int s_fc_debug_max_replay; @@ -1953,7 +1934,7 @@ static void ext4_fc_free(struct fs_context *fc) for (i = 0; i < EXT4_MAXQUOTAS; i++) kfree(ctx->s_qf_names[i]); - kfree(ctx->test_dummy_enc_arg); + fscrypt_free_dummy_policy(&ctx->dummy_enc_policy); kfree(ctx); } @@ -2029,6 +2010,29 @@ static int unnote_qf_name(struct fs_context *fc, int qtype) } #endif +static int ext4_parse_test_dummy_encryption(const struct fs_parameter *param, + struct ext4_fs_context *ctx) +{ + int err; + + if (!IS_ENABLED(CONFIG_FS_ENCRYPTION)) { + ext4_msg(NULL, KERN_WARNING, + "test_dummy_encryption option not supported"); + return -EINVAL; + } + err = fscrypt_parse_test_dummy_encryption(param, + &ctx->dummy_enc_policy); + if (err == -EINVAL) { + ext4_msg(NULL, KERN_WARNING, + "Value of option \"%s\" is unrecognized", param->key); + } else if (err == -EEXIST) { + ext4_msg(NULL, KERN_WARNING, + "Conflicting test_dummy_encryption options"); + return -EINVAL; + } + return err; +} + #define EXT4_SET_CTX(name) \ static inline void ctx_set_##name(struct ext4_fs_context *ctx, \ unsigned long flag) \ @@ -2291,29 +2295,7 @@ static int ext4_parse_param(struct fs_context *fc, struct fs_parameter *param) ctx->spec |= EXT4_SPEC_JOURNAL_IOPRIO; return 0; case Opt_test_dummy_encryption: -#ifdef CONFIG_FS_ENCRYPTION - if (param->type == fs_value_is_flag) { - ctx->spec |= EXT4_SPEC_DUMMY_ENCRYPTION; - ctx->test_dummy_enc_arg = NULL; - return 0; - } - if (*param->string && - !(!strcmp(param->string, "v1") || - !strcmp(param->string, "v2"))) { - ext4_msg(NULL, KERN_WARNING, - "Value of option \"%s\" is unrecognized", - param->key); - return -EINVAL; - } - ctx->spec |= EXT4_SPEC_DUMMY_ENCRYPTION; - ctx->test_dummy_enc_arg = kmemdup_nul(param->string, param->size, - GFP_KERNEL); - return 0; -#else - ext4_msg(NULL, KERN_WARNING, - "test_dummy_encryption option not supported"); - return -EINVAL; -#endif + return ext4_parse_test_dummy_encryption(param, ctx); case Opt_dax: case Opt_dax_type: #ifdef CONFIG_FS_DAX @@ -2504,7 +2486,8 @@ parse_failed: if (s_ctx->spec & EXT4_SPEC_JOURNAL_IOPRIO) m_ctx->journal_ioprio = s_ctx->journal_ioprio; - ret = ext4_apply_options(fc, sb); + ext4_apply_options(fc, sb); + ret = 0; out_free: if (fc) { @@ -2673,11 +2656,11 @@ err_jquota_specified: static int ext4_check_test_dummy_encryption(const struct fs_context *fc, struct super_block *sb) { -#ifdef CONFIG_FS_ENCRYPTION const struct ext4_fs_context *ctx = fc->fs_private; const struct ext4_sb_info *sbi = EXT4_SB(sb); + int err; - if (!(ctx->spec & EXT4_SPEC_DUMMY_ENCRYPTION)) + if (!fscrypt_is_dummy_policy_set(&ctx->dummy_enc_policy)) return 0; if (!ext4_has_feature_encrypt(sb)) { @@ -2691,14 +2674,46 @@ static int ext4_check_test_dummy_encryption(const struct fs_context *fc, * needed to allow it to be set or changed during remount. We do allow * it to be specified during remount, but only if there is no change. */ - if (fc->purpose == FS_CONTEXT_FOR_RECONFIGURE && - !sbi->s_dummy_enc_policy.policy) { + if (fc->purpose == FS_CONTEXT_FOR_RECONFIGURE) { + if (fscrypt_dummy_policies_equal(&sbi->s_dummy_enc_policy, + &ctx->dummy_enc_policy)) + return 0; ext4_msg(NULL, KERN_WARNING, - "Can't set test_dummy_encryption on remount"); + "Can't set or change test_dummy_encryption on remount"); return -EINVAL; } -#endif /* CONFIG_FS_ENCRYPTION */ - return 0; + /* Also make sure s_mount_opts didn't contain a conflicting value. */ + if (fscrypt_is_dummy_policy_set(&sbi->s_dummy_enc_policy)) { + if (fscrypt_dummy_policies_equal(&sbi->s_dummy_enc_policy, + &ctx->dummy_enc_policy)) + return 0; + ext4_msg(NULL, KERN_WARNING, + "Conflicting test_dummy_encryption options"); + return -EINVAL; + } + /* + * fscrypt_add_test_dummy_key() technically changes the super_block, so + * technically it should be delayed until ext4_apply_options() like the + * other changes. But since we never get here for remounts (see above), + * and this is the last chance to report errors, we do it here. + */ + err = fscrypt_add_test_dummy_key(sb, &ctx->dummy_enc_policy); + if (err) + ext4_msg(NULL, KERN_WARNING, + "Error adding test dummy encryption key [%d]", err); + return err; +} + +static void ext4_apply_test_dummy_encryption(struct ext4_fs_context *ctx, + struct super_block *sb) +{ + if (!fscrypt_is_dummy_policy_set(&ctx->dummy_enc_policy) || + /* if already set, it was already verified to be the same */ + fscrypt_is_dummy_policy_set(&EXT4_SB(sb)->s_dummy_enc_policy)) + return; + EXT4_SB(sb)->s_dummy_enc_policy = ctx->dummy_enc_policy; + memset(&ctx->dummy_enc_policy, 0, sizeof(ctx->dummy_enc_policy)); + ext4_msg(sb, KERN_WARNING, "Test dummy encryption mode enabled"); } static int ext4_check_opt_consistency(struct fs_context *fc, @@ -2785,11 +2800,10 @@ fail_dax_change_remount: return ext4_check_quota_consistency(fc, sb); } -static int ext4_apply_options(struct fs_context *fc, struct super_block *sb) +static void ext4_apply_options(struct fs_context *fc, struct super_block *sb) { struct ext4_fs_context *ctx = fc->fs_private; struct ext4_sb_info *sbi = fc->s_fs_info; - int ret = 0; sbi->s_mount_opt &= ~ctx->mask_s_mount_opt; sbi->s_mount_opt |= ctx->vals_s_mount_opt; @@ -2825,11 +2839,7 @@ static int ext4_apply_options(struct fs_context *fc, struct super_block *sb) #endif ext4_apply_quota_options(fc, sb); - - if (ctx->spec & EXT4_SPEC_DUMMY_ENCRYPTION) - ret = ext4_set_test_dummy_encryption(sb, ctx->test_dummy_enc_arg); - - return ret; + ext4_apply_test_dummy_encryption(ctx, sb); } @@ -4552,9 +4562,7 @@ static int __ext4_fill_super(struct fs_context *fc, struct super_block *sb) if (err < 0) goto failed_mount; - err = ext4_apply_options(fc, sb); - if (err < 0) - goto failed_mount; + ext4_apply_options(fc, sb); #if IS_ENABLED(CONFIG_UNICODE) if (ext4_has_feature_casefold(sb) && !sb->s_encoding) { @@ -5302,14 +5310,6 @@ no_journal: err = percpu_counter_init(&sbi->s_freeinodes_counter, freei, GFP_KERNEL); } - /* - * Update the checksum after updating free space/inode - * counters. Otherwise the superblock can have an incorrect - * checksum in the buffer cache until it is written out and - * e2fsprogs programs trying to open a file system immediately - * after it is mounted can fail. - */ - ext4_superblock_csum_set(sb); if (!err) err = percpu_counter_init(&sbi->s_dirs_counter, ext4_count_dirs(sb), GFP_KERNEL); @@ -5367,6 +5367,14 @@ no_journal: EXT4_SB(sb)->s_mount_state |= EXT4_ORPHAN_FS; ext4_orphan_cleanup(sb, es); EXT4_SB(sb)->s_mount_state &= ~EXT4_ORPHAN_FS; + /* + * Update the checksum after updating free space/inode counters and + * ext4_orphan_cleanup. Otherwise the superblock can have an incorrect + * checksum in the buffer cache until it is written out and + * e2fsprogs programs trying to open a file system immediately + * after it is mounted can fail. + */ + ext4_superblock_csum_set(sb); if (needs_recovery) { ext4_msg(sb, KERN_INFO, "recovery complete"); err = ext4_mark_recovery_complete(sb, es); @@ -5898,7 +5906,6 @@ static void ext4_update_super(struct super_block *sb) static int ext4_commit_super(struct super_block *sb) { struct buffer_head *sbh = EXT4_SB(sb)->s_sbh; - int error = 0; if (!sbh) return -EINVAL; @@ -5907,6 +5914,13 @@ static int ext4_commit_super(struct super_block *sb) ext4_update_super(sb); + lock_buffer(sbh); + /* Buffer got discarded which means block device got invalidated */ + if (!buffer_mapped(sbh)) { + unlock_buffer(sbh); + return -EIO; + } + if (buffer_write_io_error(sbh) || !buffer_uptodate(sbh)) { /* * Oh, dear. A previous attempt to write the @@ -5921,17 +5935,21 @@ static int ext4_commit_super(struct super_block *sb) clear_buffer_write_io_error(sbh); set_buffer_uptodate(sbh); } - BUFFER_TRACE(sbh, "marking dirty"); - mark_buffer_dirty(sbh); - error = __sync_dirty_buffer(sbh, - REQ_SYNC | (test_opt(sb, BARRIER) ? REQ_FUA : 0)); + get_bh(sbh); + /* Clear potential dirty bit if it was journalled update */ + clear_buffer_dirty(sbh); + sbh->b_end_io = end_buffer_write_sync; + submit_bh(REQ_OP_WRITE, + REQ_SYNC | (test_opt(sb, BARRIER) ? REQ_FUA : 0), sbh); + wait_on_buffer(sbh); if (buffer_write_io_error(sbh)) { ext4_msg(sb, KERN_ERR, "I/O error while writing " "superblock"); clear_buffer_write_io_error(sbh); set_buffer_uptodate(sbh); + return -EIO; } - return error; + return 0; } /* diff --git a/fs/ext4/xattr.c b/fs/ext4/xattr.c index 042325349098..564e28a1aa94 100644 --- a/fs/ext4/xattr.c +++ b/fs/ext4/xattr.c @@ -1895,11 +1895,10 @@ ext4_xattr_block_set(handle_t *handle, struct inode *inode, unlock_buffer(bs->bh); ea_bdebug(bs->bh, "cloning"); - s->base = kmalloc(bs->bh->b_size, GFP_NOFS); + s->base = kmemdup(BHDR(bs->bh), bs->bh->b_size, GFP_NOFS); error = -ENOMEM; if (s->base == NULL) goto cleanup; - memcpy(s->base, BHDR(bs->bh), bs->bh->b_size); s->first = ENTRY(header(s->base)+1); header(s->base)->h_refcount = cpu_to_le32(1); s->here = ENTRY(s->base + offset); diff --git a/fs/f2fs/iostat.c b/fs/f2fs/iostat.c index be599f31d3c4..d84c5f6cc09d 100644 --- a/fs/f2fs/iostat.c +++ b/fs/f2fs/iostat.c @@ -91,8 +91,9 @@ static inline void __record_iostat_latency(struct f2fs_sb_info *sbi) unsigned int cnt; struct f2fs_iostat_latency iostat_lat[MAX_IO_TYPE][NR_PAGE_TYPE]; struct iostat_lat_info *io_lat = sbi->iostat_io_lat; + unsigned long flags; - spin_lock_bh(&sbi->iostat_lat_lock); + spin_lock_irqsave(&sbi->iostat_lat_lock, flags); for (idx = 0; idx < MAX_IO_TYPE; idx++) { for (io = 0; io < NR_PAGE_TYPE; io++) { cnt = io_lat->bio_cnt[idx][io]; @@ -106,7 +107,7 @@ static inline void __record_iostat_latency(struct f2fs_sb_info *sbi) io_lat->bio_cnt[idx][io] = 0; } } - spin_unlock_bh(&sbi->iostat_lat_lock); + spin_unlock_irqrestore(&sbi->iostat_lat_lock, flags); trace_f2fs_iostat_latency(sbi, iostat_lat); } @@ -115,14 +116,15 @@ static inline void f2fs_record_iostat(struct f2fs_sb_info *sbi) { unsigned long long iostat_diff[NR_IO_TYPE]; int i; + unsigned long flags; if (time_is_after_jiffies(sbi->iostat_next_period)) return; /* Need double check under the lock */ - spin_lock_bh(&sbi->iostat_lock); + spin_lock_irqsave(&sbi->iostat_lock, flags); if (time_is_after_jiffies(sbi->iostat_next_period)) { - spin_unlock_bh(&sbi->iostat_lock); + spin_unlock_irqrestore(&sbi->iostat_lock, flags); return; } sbi->iostat_next_period = jiffies + @@ -133,7 +135,7 @@ static inline void f2fs_record_iostat(struct f2fs_sb_info *sbi) sbi->prev_rw_iostat[i]; sbi->prev_rw_iostat[i] = sbi->rw_iostat[i]; } - spin_unlock_bh(&sbi->iostat_lock); + spin_unlock_irqrestore(&sbi->iostat_lock, flags); trace_f2fs_iostat(sbi, iostat_diff); @@ -145,25 +147,27 @@ void f2fs_reset_iostat(struct f2fs_sb_info *sbi) struct iostat_lat_info *io_lat = sbi->iostat_io_lat; int i; - spin_lock_bh(&sbi->iostat_lock); + spin_lock_irq(&sbi->iostat_lock); for (i = 0; i < NR_IO_TYPE; i++) { sbi->rw_iostat[i] = 0; sbi->prev_rw_iostat[i] = 0; } - spin_unlock_bh(&sbi->iostat_lock); + spin_unlock_irq(&sbi->iostat_lock); - spin_lock_bh(&sbi->iostat_lat_lock); + spin_lock_irq(&sbi->iostat_lat_lock); memset(io_lat, 0, sizeof(struct iostat_lat_info)); - spin_unlock_bh(&sbi->iostat_lat_lock); + spin_unlock_irq(&sbi->iostat_lat_lock); } void f2fs_update_iostat(struct f2fs_sb_info *sbi, enum iostat_type type, unsigned long long io_bytes) { + unsigned long flags; + if (!sbi->iostat_enable) return; - spin_lock_bh(&sbi->iostat_lock); + spin_lock_irqsave(&sbi->iostat_lock, flags); sbi->rw_iostat[type] += io_bytes; if (type == APP_BUFFERED_IO || type == APP_DIRECT_IO) @@ -172,7 +176,7 @@ void f2fs_update_iostat(struct f2fs_sb_info *sbi, if (type == APP_BUFFERED_READ_IO || type == APP_DIRECT_READ_IO) sbi->rw_iostat[APP_READ_IO] += io_bytes; - spin_unlock_bh(&sbi->iostat_lock); + spin_unlock_irqrestore(&sbi->iostat_lock, flags); f2fs_record_iostat(sbi); } @@ -185,6 +189,7 @@ static inline void __update_iostat_latency(struct bio_iostat_ctx *iostat_ctx, struct f2fs_sb_info *sbi = iostat_ctx->sbi; struct iostat_lat_info *io_lat = sbi->iostat_io_lat; int idx; + unsigned long flags; if (!sbi->iostat_enable) return; @@ -202,12 +207,12 @@ static inline void __update_iostat_latency(struct bio_iostat_ctx *iostat_ctx, idx = WRITE_ASYNC_IO; } - spin_lock_bh(&sbi->iostat_lat_lock); + spin_lock_irqsave(&sbi->iostat_lat_lock, flags); io_lat->sum_lat[idx][iotype] += ts_diff; io_lat->bio_cnt[idx][iotype]++; if (ts_diff > io_lat->peak_lat[idx][iotype]) io_lat->peak_lat[idx][iotype] = ts_diff; - spin_unlock_bh(&sbi->iostat_lat_lock); + spin_unlock_irqrestore(&sbi->iostat_lat_lock, flags); } void iostat_update_and_unbind_ctx(struct bio *bio, int rw) diff --git a/fs/f2fs/namei.c b/fs/f2fs/namei.c index c549acb52ac4..bf00d5057abb 100644 --- a/fs/f2fs/namei.c +++ b/fs/f2fs/namei.c @@ -89,8 +89,6 @@ static struct inode *f2fs_new_inode(struct user_namespace *mnt_userns, if (test_opt(sbi, INLINE_XATTR)) set_inode_flag(inode, FI_INLINE_XATTR); - if (test_opt(sbi, INLINE_DATA) && f2fs_may_inline_data(inode)) - set_inode_flag(inode, FI_INLINE_DATA); if (f2fs_may_inline_dentry(inode)) set_inode_flag(inode, FI_INLINE_DENTRY); @@ -107,10 +105,6 @@ static struct inode *f2fs_new_inode(struct user_namespace *mnt_userns, f2fs_init_extent_tree(inode, NULL); - stat_inc_inline_xattr(inode); - stat_inc_inline_inode(inode); - stat_inc_inline_dir(inode); - F2FS_I(inode)->i_flags = f2fs_mask_flags(mode, F2FS_I(dir)->i_flags & F2FS_FL_INHERITED); @@ -127,6 +121,14 @@ static struct inode *f2fs_new_inode(struct user_namespace *mnt_userns, set_compress_context(inode); } + /* Should enable inline_data after compression set */ + if (test_opt(sbi, INLINE_DATA) && f2fs_may_inline_data(inode)) + set_inode_flag(inode, FI_INLINE_DATA); + + stat_inc_inline_xattr(inode); + stat_inc_inline_inode(inode); + stat_inc_inline_dir(inode); + f2fs_set_inode_flags(inode); trace_f2fs_new_inode(inode, 0); @@ -325,6 +327,9 @@ static void set_compress_inode(struct f2fs_sb_info *sbi, struct inode *inode, if (!is_extension_exist(name, ext[i], false)) continue; + /* Do not use inline_data with compression */ + stat_dec_inline_inode(inode); + clear_inode_flag(inode, FI_INLINE_DATA); set_compress_context(inode); return; } diff --git a/fs/f2fs/node.c b/fs/f2fs/node.c index 836c79a20afc..cf6f7fc83c08 100644 --- a/fs/f2fs/node.c +++ b/fs/f2fs/node.c @@ -1450,7 +1450,9 @@ page_hit: out_err: ClearPageUptodate(page); out_put_err: - f2fs_handle_page_eio(sbi, page->index, NODE); + /* ENOENT comes from read_node_page which is not an error. */ + if (err != -ENOENT) + f2fs_handle_page_eio(sbi, page->index, NODE); f2fs_put_page(page, 1); return ERR_PTR(err); } diff --git a/fs/fs-writeback.c b/fs/fs-writeback.c index a21d8f1a56d1..05221366a16d 100644 --- a/fs/fs-writeback.c +++ b/fs/fs-writeback.c @@ -120,6 +120,7 @@ static bool inode_io_list_move_locked(struct inode *inode, struct list_head *head) { assert_spin_locked(&wb->list_lock); + assert_spin_locked(&inode->i_lock); list_move(&inode->i_io_list, head); @@ -1365,9 +1366,9 @@ static int move_expired_inodes(struct list_head *delaying_queue, inode = wb_inode(delaying_queue->prev); if (inode_dirtied_after(inode, dirtied_before)) break; + spin_lock(&inode->i_lock); list_move(&inode->i_io_list, &tmp); moved++; - spin_lock(&inode->i_lock); inode->i_state |= I_SYNC_QUEUED; spin_unlock(&inode->i_lock); if (sb_is_blkdev_sb(inode->i_sb)) @@ -1383,7 +1384,12 @@ static int move_expired_inodes(struct list_head *delaying_queue, goto out; } - /* Move inodes from one superblock together */ + /* + * Although inode's i_io_list is moved from 'tmp' to 'dispatch_queue', + * we don't take inode->i_lock here because it is just a pointless overhead. + * Inode is already marked as I_SYNC_QUEUED so writeback list handling is + * fully under our control. + */ while (!list_empty(&tmp)) { sb = wb_inode(tmp.prev)->i_sb; list_for_each_prev_safe(pos, node, &tmp) { @@ -1826,8 +1832,8 @@ static long writeback_sb_inodes(struct super_block *sb, * We'll have another go at writing back this inode * when we completed a full scan of b_io. */ - spin_unlock(&inode->i_lock); requeue_io(inode, wb); + spin_unlock(&inode->i_lock); trace_writeback_sb_inodes_requeue(inode); continue; } @@ -2358,6 +2364,7 @@ void __mark_inode_dirty(struct inode *inode, int flags) { struct super_block *sb = inode->i_sb; int dirtytime = 0; + struct bdi_writeback *wb = NULL; trace_writeback_mark_inode_dirty(inode, flags); @@ -2410,13 +2417,24 @@ void __mark_inode_dirty(struct inode *inode, int flags) inode->i_state |= flags; /* + * Grab inode's wb early because it requires dropping i_lock and we + * need to make sure following checks happen atomically with dirty + * list handling so that we don't move inodes under flush worker's + * hands. + */ + if (!was_dirty) { + wb = locked_inode_to_wb_and_lock_list(inode); + spin_lock(&inode->i_lock); + } + + /* * If the inode is queued for writeback by flush worker, just * update its dirty state. Once the flush worker is done with * the inode it will place it on the appropriate superblock * list, based upon its state. */ if (inode->i_state & I_SYNC_QUEUED) - goto out_unlock_inode; + goto out_unlock; /* * Only add valid (hashed) inodes to the superblock's @@ -2424,22 +2442,19 @@ void __mark_inode_dirty(struct inode *inode, int flags) */ if (!S_ISBLK(inode->i_mode)) { if (inode_unhashed(inode)) - goto out_unlock_inode; + goto out_unlock; } if (inode->i_state & I_FREEING) - goto out_unlock_inode; + goto out_unlock; /* * If the inode was already on b_dirty/b_io/b_more_io, don't * reposition it (that would break b_dirty time-ordering). */ if (!was_dirty) { - struct bdi_writeback *wb; struct list_head *dirty_list; bool wakeup_bdi = false; - wb = locked_inode_to_wb_and_lock_list(inode); - inode->dirtied_when = jiffies; if (dirtytime) inode->dirtied_time_when = jiffies; @@ -2453,6 +2468,7 @@ void __mark_inode_dirty(struct inode *inode, int flags) dirty_list); spin_unlock(&wb->list_lock); + spin_unlock(&inode->i_lock); trace_writeback_dirty_inode_enqueue(inode); /* @@ -2467,6 +2483,9 @@ void __mark_inode_dirty(struct inode *inode, int flags) return; } } +out_unlock: + if (wb) + spin_unlock(&wb->list_lock); out_unlock_inode: spin_unlock(&inode->i_lock); } diff --git a/fs/hugetlbfs/inode.c b/fs/hugetlbfs/inode.c index 62408047e8d7..02eb72351b15 100644 --- a/fs/hugetlbfs/inode.c +++ b/fs/hugetlbfs/inode.c @@ -600,41 +600,79 @@ static void hugetlb_vmtruncate(struct inode *inode, loff_t offset) remove_inode_hugepages(inode, offset, LLONG_MAX); } +static void hugetlbfs_zero_partial_page(struct hstate *h, + struct address_space *mapping, + loff_t start, + loff_t end) +{ + pgoff_t idx = start >> huge_page_shift(h); + struct folio *folio; + + folio = filemap_lock_folio(mapping, idx); + if (!folio) + return; + + start = start & ~huge_page_mask(h); + end = end & ~huge_page_mask(h); + if (!end) + end = huge_page_size(h); + + folio_zero_segment(folio, (size_t)start, (size_t)end); + + folio_unlock(folio); + folio_put(folio); +} + static long hugetlbfs_punch_hole(struct inode *inode, loff_t offset, loff_t len) { + struct hugetlbfs_inode_info *info = HUGETLBFS_I(inode); + struct address_space *mapping = inode->i_mapping; struct hstate *h = hstate_inode(inode); loff_t hpage_size = huge_page_size(h); loff_t hole_start, hole_end; /* - * For hole punch round up the beginning offset of the hole and - * round down the end. + * hole_start and hole_end indicate the full pages within the hole. */ hole_start = round_up(offset, hpage_size); hole_end = round_down(offset + len, hpage_size); - if (hole_end > hole_start) { - struct address_space *mapping = inode->i_mapping; - struct hugetlbfs_inode_info *info = HUGETLBFS_I(inode); + inode_lock(inode); - inode_lock(inode); + /* protected by i_rwsem */ + if (info->seals & (F_SEAL_WRITE | F_SEAL_FUTURE_WRITE)) { + inode_unlock(inode); + return -EPERM; + } - /* protected by i_rwsem */ - if (info->seals & (F_SEAL_WRITE | F_SEAL_FUTURE_WRITE)) { - inode_unlock(inode); - return -EPERM; - } + i_mmap_lock_write(mapping); + + /* If range starts before first full page, zero partial page. */ + if (offset < hole_start) + hugetlbfs_zero_partial_page(h, mapping, + offset, min(offset + len, hole_start)); - i_mmap_lock_write(mapping); + /* Unmap users of full pages in the hole. */ + if (hole_end > hole_start) { if (!RB_EMPTY_ROOT(&mapping->i_mmap.rb_root)) hugetlb_vmdelete_list(&mapping->i_mmap, hole_start >> PAGE_SHIFT, hole_end >> PAGE_SHIFT, 0); - i_mmap_unlock_write(mapping); - remove_inode_hugepages(inode, hole_start, hole_end); - inode_unlock(inode); } + /* If range extends beyond last full page, zero partial page. */ + if ((offset + len) > hole_end && (offset + len) > hole_start) + hugetlbfs_zero_partial_page(h, mapping, + hole_end, offset + len); + + i_mmap_unlock_write(mapping); + + /* Remove full pages from the file. */ + if (hole_end > hole_start) + remove_inode_hugepages(inode, hole_start, hole_end); + + inode_unlock(inode); + return 0; } diff --git a/fs/inode.c b/fs/inode.c index 9d9b422504d1..bd4da9c5207e 100644 --- a/fs/inode.c +++ b/fs/inode.c @@ -27,7 +27,7 @@ * Inode locking rules: * * inode->i_lock protects: - * inode->i_state, inode->i_hash, __iget() + * inode->i_state, inode->i_hash, __iget(), inode->i_io_list * Inode LRU list locks protect: * inode->i_sb->s_inode_lru, inode->i_lru * inode->i_sb->s_inode_list_lock protects: diff --git a/fs/io_uring.c b/fs/io_uring.c index 3aab4182fd89..5ff2cdb425bc 100644 --- a/fs/io_uring.c +++ b/fs/io_uring.c @@ -298,8 +298,8 @@ struct io_buffer_list { /* below is for ring provided buffers */ __u16 buf_nr_pages; __u16 nr_entries; - __u32 head; - __u32 mask; + __u16 head; + __u16 mask; }; struct io_buffer { @@ -576,7 +576,6 @@ struct io_close { struct file *file; int fd; u32 file_slot; - u32 flags; }; struct io_timeout_data { @@ -784,12 +783,6 @@ struct io_msg { u32 len; }; -struct io_nop { - struct file *file; - u64 extra1; - u64 extra2; -}; - struct io_async_connect { struct sockaddr_storage address; }; @@ -851,6 +844,7 @@ enum { REQ_F_SINGLE_POLL_BIT, REQ_F_DOUBLE_POLL_BIT, REQ_F_PARTIAL_IO_BIT, + REQ_F_CQE32_INIT_BIT, REQ_F_APOLL_MULTISHOT_BIT, /* keep async read/write and isreg together and in order */ REQ_F_SUPPORT_NOWAIT_BIT, @@ -920,6 +914,8 @@ enum { REQ_F_PARTIAL_IO = BIT(REQ_F_PARTIAL_IO_BIT), /* fast poll multishot mode */ REQ_F_APOLL_MULTISHOT = BIT(REQ_F_APOLL_MULTISHOT_BIT), + /* ->extra1 and ->extra2 are initialised */ + REQ_F_CQE32_INIT = BIT(REQ_F_CQE32_INIT_BIT), }; struct async_poll { @@ -994,7 +990,6 @@ struct io_kiocb { struct io_msg msg; struct io_xattr xattr; struct io_socket sock; - struct io_nop nop; struct io_uring_cmd uring_cmd; }; @@ -1121,7 +1116,6 @@ static const struct io_op_def io_op_defs[] = { [IORING_OP_NOP] = { .audit_skip = 1, .iopoll = 1, - .buffer_select = 1, }, [IORING_OP_READV] = { .needs_file = 1, @@ -1729,9 +1723,16 @@ static void io_kbuf_recycle(struct io_kiocb *req, unsigned issue_flags) if (!(req->flags & (REQ_F_BUFFER_SELECTED|REQ_F_BUFFER_RING))) return; - /* don't recycle if we already did IO to this buffer */ - if (req->flags & REQ_F_PARTIAL_IO) + /* + * For legacy provided buffer mode, don't recycle if we already did + * IO to this buffer. For ring-mapped provided buffer mode, we should + * increment ring->head to explicitly monopolize the buffer to avoid + * multiple use. + */ + if ((req->flags & REQ_F_BUFFER_SELECTED) && + (req->flags & REQ_F_PARTIAL_IO)) return; + /* * We don't need to recycle for REQ_F_BUFFER_RING, we can just clear * the flag and hence ensure that bl->head doesn't get incremented. @@ -1739,8 +1740,13 @@ static void io_kbuf_recycle(struct io_kiocb *req, unsigned issue_flags) */ if (req->flags & REQ_F_BUFFER_RING) { if (req->buf_list) { - req->buf_index = req->buf_list->bgid; - req->flags &= ~REQ_F_BUFFER_RING; + if (req->flags & REQ_F_PARTIAL_IO) { + req->buf_list->head++; + req->buf_list = NULL; + } else { + req->buf_index = req->buf_list->bgid; + req->flags &= ~REQ_F_BUFFER_RING; + } } return; } @@ -1969,7 +1975,7 @@ static inline void io_req_track_inflight(struct io_kiocb *req) { if (!(req->flags & REQ_F_INFLIGHT)) { req->flags |= REQ_F_INFLIGHT; - atomic_inc(¤t->io_uring->inflight_tracked); + atomic_inc(&req->task->io_uring->inflight_tracked); } } @@ -2441,94 +2447,66 @@ static bool io_cqring_event_overflow(struct io_ring_ctx *ctx, u64 user_data, return true; } -static inline bool __io_fill_cqe(struct io_ring_ctx *ctx, u64 user_data, - s32 res, u32 cflags) +static inline bool __io_fill_cqe_req(struct io_ring_ctx *ctx, + struct io_kiocb *req) { struct io_uring_cqe *cqe; - /* - * If we can't get a cq entry, userspace overflowed the - * submission (by quite a lot). Increment the overflow count in - * the ring. - */ - cqe = io_get_cqe(ctx); - if (likely(cqe)) { - WRITE_ONCE(cqe->user_data, user_data); - WRITE_ONCE(cqe->res, res); - WRITE_ONCE(cqe->flags, cflags); - return true; - } - return io_cqring_event_overflow(ctx, user_data, res, cflags, 0, 0); -} + if (!(ctx->flags & IORING_SETUP_CQE32)) { + trace_io_uring_complete(req->ctx, req, req->cqe.user_data, + req->cqe.res, req->cqe.flags, 0, 0); -static inline bool __io_fill_cqe_req_filled(struct io_ring_ctx *ctx, - struct io_kiocb *req) -{ - struct io_uring_cqe *cqe; + /* + * If we can't get a cq entry, userspace overflowed the + * submission (by quite a lot). Increment the overflow count in + * the ring. + */ + cqe = io_get_cqe(ctx); + if (likely(cqe)) { + memcpy(cqe, &req->cqe, sizeof(*cqe)); + return true; + } - trace_io_uring_complete(req->ctx, req, req->cqe.user_data, - req->cqe.res, req->cqe.flags, 0, 0); + return io_cqring_event_overflow(ctx, req->cqe.user_data, + req->cqe.res, req->cqe.flags, + 0, 0); + } else { + u64 extra1 = 0, extra2 = 0; - /* - * If we can't get a cq entry, userspace overflowed the - * submission (by quite a lot). Increment the overflow count in - * the ring. - */ - cqe = io_get_cqe(ctx); - if (likely(cqe)) { - memcpy(cqe, &req->cqe, sizeof(*cqe)); - return true; - } - return io_cqring_event_overflow(ctx, req->cqe.user_data, - req->cqe.res, req->cqe.flags, 0, 0); -} + if (req->flags & REQ_F_CQE32_INIT) { + extra1 = req->extra1; + extra2 = req->extra2; + } -static inline bool __io_fill_cqe32_req_filled(struct io_ring_ctx *ctx, - struct io_kiocb *req) -{ - struct io_uring_cqe *cqe; - u64 extra1 = req->extra1; - u64 extra2 = req->extra2; + trace_io_uring_complete(req->ctx, req, req->cqe.user_data, + req->cqe.res, req->cqe.flags, extra1, extra2); - trace_io_uring_complete(req->ctx, req, req->cqe.user_data, - req->cqe.res, req->cqe.flags, extra1, extra2); + /* + * If we can't get a cq entry, userspace overflowed the + * submission (by quite a lot). Increment the overflow count in + * the ring. + */ + cqe = io_get_cqe(ctx); + if (likely(cqe)) { + memcpy(cqe, &req->cqe, sizeof(struct io_uring_cqe)); + WRITE_ONCE(cqe->big_cqe[0], extra1); + WRITE_ONCE(cqe->big_cqe[1], extra2); + return true; + } - /* - * If we can't get a cq entry, userspace overflowed the - * submission (by quite a lot). Increment the overflow count in - * the ring. - */ - cqe = io_get_cqe(ctx); - if (likely(cqe)) { - memcpy(cqe, &req->cqe, sizeof(struct io_uring_cqe)); - cqe->big_cqe[0] = extra1; - cqe->big_cqe[1] = extra2; - return true; + return io_cqring_event_overflow(ctx, req->cqe.user_data, + req->cqe.res, req->cqe.flags, + extra1, extra2); } - - return io_cqring_event_overflow(ctx, req->cqe.user_data, req->cqe.res, - req->cqe.flags, extra1, extra2); } -static inline bool __io_fill_cqe_req(struct io_kiocb *req, s32 res, u32 cflags) -{ - trace_io_uring_complete(req->ctx, req, req->cqe.user_data, res, cflags, 0, 0); - return __io_fill_cqe(req->ctx, req->cqe.user_data, res, cflags); -} - -static inline void __io_fill_cqe32_req(struct io_kiocb *req, s32 res, u32 cflags, - u64 extra1, u64 extra2) +static noinline bool io_fill_cqe_aux(struct io_ring_ctx *ctx, u64 user_data, + s32 res, u32 cflags) { - struct io_ring_ctx *ctx = req->ctx; struct io_uring_cqe *cqe; - if (WARN_ON_ONCE(!(ctx->flags & IORING_SETUP_CQE32))) - return; - if (req->flags & REQ_F_CQE_SKIP) - return; - - trace_io_uring_complete(ctx, req, req->cqe.user_data, res, cflags, - extra1, extra2); + ctx->cq_extra++; + trace_io_uring_complete(ctx, NULL, user_data, res, cflags, 0, 0); /* * If we can't get a cq entry, userspace overflowed the @@ -2537,23 +2515,17 @@ static inline void __io_fill_cqe32_req(struct io_kiocb *req, s32 res, u32 cflags */ cqe = io_get_cqe(ctx); if (likely(cqe)) { - WRITE_ONCE(cqe->user_data, req->cqe.user_data); + WRITE_ONCE(cqe->user_data, user_data); WRITE_ONCE(cqe->res, res); WRITE_ONCE(cqe->flags, cflags); - WRITE_ONCE(cqe->big_cqe[0], extra1); - WRITE_ONCE(cqe->big_cqe[1], extra2); - return; - } - io_cqring_event_overflow(ctx, req->cqe.user_data, res, cflags, extra1, extra2); -} - -static noinline bool io_fill_cqe_aux(struct io_ring_ctx *ctx, u64 user_data, - s32 res, u32 cflags) -{ - ctx->cq_extra++; - trace_io_uring_complete(ctx, NULL, user_data, res, cflags, 0, 0); - return __io_fill_cqe(ctx, user_data, res, cflags); + if (ctx->flags & IORING_SETUP_CQE32) { + WRITE_ONCE(cqe->big_cqe[0], 0); + WRITE_ONCE(cqe->big_cqe[1], 0); + } + return true; + } + return io_cqring_event_overflow(ctx, user_data, res, cflags, 0, 0); } static void __io_req_complete_put(struct io_kiocb *req) @@ -2590,16 +2562,11 @@ static void __io_req_complete_put(struct io_kiocb *req) static void __io_req_complete_post(struct io_kiocb *req, s32 res, u32 cflags) { - if (!(req->flags & REQ_F_CQE_SKIP)) - __io_fill_cqe_req(req, res, cflags); - __io_req_complete_put(req); -} - -static void __io_req_complete_post32(struct io_kiocb *req, s32 res, - u32 cflags, u64 extra1, u64 extra2) -{ - if (!(req->flags & REQ_F_CQE_SKIP)) - __io_fill_cqe32_req(req, res, cflags, extra1, extra2); + if (!(req->flags & REQ_F_CQE_SKIP)) { + req->cqe.res = res; + req->cqe.flags = cflags; + __io_fill_cqe_req(req->ctx, req); + } __io_req_complete_put(req); } @@ -2614,18 +2581,6 @@ static void io_req_complete_post(struct io_kiocb *req, s32 res, u32 cflags) io_cqring_ev_posted(ctx); } -static void io_req_complete_post32(struct io_kiocb *req, s32 res, - u32 cflags, u64 extra1, u64 extra2) -{ - struct io_ring_ctx *ctx = req->ctx; - - spin_lock(&ctx->completion_lock); - __io_req_complete_post32(req, res, cflags, extra1, extra2); - io_commit_cqring(ctx); - spin_unlock(&ctx->completion_lock); - io_cqring_ev_posted(ctx); -} - static inline void io_req_complete_state(struct io_kiocb *req, s32 res, u32 cflags) { @@ -2643,19 +2598,6 @@ static inline void __io_req_complete(struct io_kiocb *req, unsigned issue_flags, io_req_complete_post(req, res, cflags); } -static inline void __io_req_complete32(struct io_kiocb *req, - unsigned int issue_flags, s32 res, - u32 cflags, u64 extra1, u64 extra2) -{ - if (issue_flags & IO_URING_F_COMPLETE_DEFER) { - io_req_complete_state(req, res, cflags); - req->extra1 = extra1; - req->extra2 = extra2; - } else { - io_req_complete_post32(req, res, cflags, extra1, extra2); - } -} - static inline void io_req_complete(struct io_kiocb *req, s32 res) { if (res < 0) @@ -3202,12 +3144,8 @@ static void __io_submit_flush_completions(struct io_ring_ctx *ctx) struct io_kiocb *req = container_of(node, struct io_kiocb, comp_list); - if (!(req->flags & REQ_F_CQE_SKIP)) { - if (!(ctx->flags & IORING_SETUP_CQE32)) - __io_fill_cqe_req_filled(ctx, req); - else - __io_fill_cqe32_req_filled(ctx, req); - } + if (!(req->flags & REQ_F_CQE_SKIP)) + __io_fill_cqe_req(ctx, req); } io_commit_cqring(ctx); @@ -3326,7 +3264,9 @@ static int io_do_iopoll(struct io_ring_ctx *ctx, bool force_nonspin) nr_events++; if (unlikely(req->flags & REQ_F_CQE_SKIP)) continue; - __io_fill_cqe_req(req, req->cqe.res, io_put_kbuf(req, 0)); + + req->cqe.flags = io_put_kbuf(req, 0); + __io_fill_cqe_req(req->ctx, req); } if (unlikely(!nr_events)) @@ -3497,7 +3437,7 @@ static bool __io_complete_rw_common(struct io_kiocb *req, long res) if (unlikely(res != req->cqe.res)) { if ((res == -EAGAIN || res == -EOPNOTSUPP) && io_rw_should_reissue(req)) { - req->flags |= REQ_F_REISSUE; + req->flags |= REQ_F_REISSUE | REQ_F_PARTIAL_IO; return true; } req_set_fail(req); @@ -3547,7 +3487,7 @@ static void io_complete_rw_iopoll(struct kiocb *kiocb, long res) kiocb_end_write(req); if (unlikely(res != req->cqe.res)) { if (res == -EAGAIN && io_rw_should_reissue(req)) { - req->flags |= REQ_F_REISSUE; + req->flags |= REQ_F_REISSUE | REQ_F_PARTIAL_IO; return; } req->cqe.res = res; @@ -3677,6 +3617,20 @@ static int io_prep_rw(struct io_kiocb *req, const struct io_uring_sqe *sqe) int ret; kiocb->ki_pos = READ_ONCE(sqe->off); + /* used for fixed read/write too - just read unconditionally */ + req->buf_index = READ_ONCE(sqe->buf_index); + + if (req->opcode == IORING_OP_READ_FIXED || + req->opcode == IORING_OP_WRITE_FIXED) { + struct io_ring_ctx *ctx = req->ctx; + u16 index; + + if (unlikely(req->buf_index >= ctx->nr_user_bufs)) + return -EFAULT; + index = array_index_nospec(req->buf_index, ctx->nr_user_bufs); + req->imu = ctx->user_bufs[index]; + io_req_set_rsrc_node(req, ctx, 0); + } ioprio = READ_ONCE(sqe->ioprio); if (ioprio) { @@ -3689,12 +3643,9 @@ static int io_prep_rw(struct io_kiocb *req, const struct io_uring_sqe *sqe) kiocb->ki_ioprio = get_current_ioprio(); } - req->imu = NULL; req->rw.addr = READ_ONCE(sqe->addr); req->rw.len = READ_ONCE(sqe->len); req->rw.flags = READ_ONCE(sqe->rw_flags); - /* used for fixed read/write too - just read unconditionally */ - req->buf_index = READ_ONCE(sqe->buf_index); return 0; } @@ -3826,20 +3777,9 @@ static int __io_import_fixed(struct io_kiocb *req, int rw, struct iov_iter *iter static int io_import_fixed(struct io_kiocb *req, int rw, struct iov_iter *iter, unsigned int issue_flags) { - struct io_mapped_ubuf *imu = req->imu; - u16 index, buf_index = req->buf_index; - - if (likely(!imu)) { - struct io_ring_ctx *ctx = req->ctx; - - if (unlikely(buf_index >= ctx->nr_user_bufs)) - return -EFAULT; - io_req_set_rsrc_node(req, ctx, issue_flags); - index = array_index_nospec(buf_index, ctx->nr_user_bufs); - imu = READ_ONCE(ctx->user_bufs[index]); - req->imu = imu; - } - return __io_import_fixed(req, rw, iter, imu); + if (WARN_ON_ONCE(!req->imu)) + return -EFAULT; + return __io_import_fixed(req, rw, iter, req->imu); } static int io_buffer_add_list(struct io_ring_ctx *ctx, @@ -3876,19 +3816,17 @@ static void __user *io_ring_buffer_select(struct io_kiocb *req, size_t *len, { struct io_uring_buf_ring *br = bl->buf_ring; struct io_uring_buf *buf; - __u32 head = bl->head; + __u16 head = bl->head; - if (unlikely(smp_load_acquire(&br->tail) == head)) { - io_ring_submit_unlock(req->ctx, issue_flags); + if (unlikely(smp_load_acquire(&br->tail) == head)) return NULL; - } head &= bl->mask; if (head < IO_BUFFER_LIST_BUF_PER_PAGE) { buf = &br->bufs[head]; } else { int off = head & (IO_BUFFER_LIST_BUF_PER_PAGE - 1); - int index = head / IO_BUFFER_LIST_BUF_PER_PAGE - 1; + int index = head / IO_BUFFER_LIST_BUF_PER_PAGE; buf = page_address(bl->buf_pages[index]); buf += off; } @@ -3898,7 +3836,7 @@ static void __user *io_ring_buffer_select(struct io_kiocb *req, size_t *len, req->buf_list = bl; req->buf_index = buf->bid; - if (issue_flags & IO_URING_F_UNLOCKED) { + if (issue_flags & IO_URING_F_UNLOCKED || !file_can_poll(req->file)) { /* * If we came in unlocked, we have no choice but to consume the * buffer here. This does mean it'll be pinned until the IO @@ -5079,10 +5017,18 @@ void io_uring_cmd_complete_in_task(struct io_uring_cmd *ioucmd, req->uring_cmd.task_work_cb = task_work_cb; req->io_task_work.func = io_uring_cmd_work; - io_req_task_prio_work_add(req); + io_req_task_work_add(req); } EXPORT_SYMBOL_GPL(io_uring_cmd_complete_in_task); +static inline void io_req_set_cqe32_extra(struct io_kiocb *req, + u64 extra1, u64 extra2) +{ + req->extra1 = extra1; + req->extra2 = extra2; + req->flags |= REQ_F_CQE32_INIT; +} + /* * Called by consumers of io_uring_cmd, if they originally returned * -EIOCBQUEUED upon receiving the command. @@ -5093,10 +5039,10 @@ void io_uring_cmd_done(struct io_uring_cmd *ioucmd, ssize_t ret, ssize_t res2) if (ret < 0) req_set_fail(req); + if (req->ctx->flags & IORING_SETUP_CQE32) - __io_req_complete32(req, 0, ret, 0, res2, 0); - else - io_req_complete(req, ret); + io_req_set_cqe32_extra(req, res2, 0); + io_req_complete(req, ret); } EXPORT_SYMBOL_GPL(io_uring_cmd_done); @@ -5258,14 +5204,6 @@ done: static int io_nop_prep(struct io_kiocb *req, const struct io_uring_sqe *sqe) { - /* - * If the ring is setup with CQE32, relay back addr/addr - */ - if (req->ctx->flags & IORING_SETUP_CQE32) { - req->nop.extra1 = READ_ONCE(sqe->addr); - req->nop.extra2 = READ_ONCE(sqe->addr2); - } - return 0; } @@ -5274,23 +5212,7 @@ static int io_nop_prep(struct io_kiocb *req, const struct io_uring_sqe *sqe) */ static int io_nop(struct io_kiocb *req, unsigned int issue_flags) { - unsigned int cflags; - void __user *buf; - - if (req->flags & REQ_F_BUFFER_SELECT) { - size_t len = 1; - - buf = io_buffer_select(req, &len, issue_flags); - if (!buf) - return -ENOBUFS; - } - - cflags = io_put_kbuf(req, issue_flags); - if (!(req->ctx->flags & IORING_SETUP_CQE32)) - __io_req_complete(req, issue_flags, 0, cflags); - else - __io_req_complete32(req, issue_flags, 0, cflags, - req->nop.extra1, req->nop.extra2); + __io_req_complete(req, issue_flags, 0, 0); return 0; } @@ -5988,18 +5910,14 @@ static int io_statx(struct io_kiocb *req, unsigned int issue_flags) static int io_close_prep(struct io_kiocb *req, const struct io_uring_sqe *sqe) { - if (sqe->off || sqe->addr || sqe->len || sqe->buf_index) + if (sqe->off || sqe->addr || sqe->len || sqe->rw_flags || sqe->buf_index) return -EINVAL; if (req->flags & REQ_F_FIXED_FILE) return -EBADF; req->close.fd = READ_ONCE(sqe->fd); req->close.file_slot = READ_ONCE(sqe->file_index); - req->close.flags = READ_ONCE(sqe->close_flags); - if (req->close.flags & ~IORING_CLOSE_FD_AND_FILE_SLOT) - return -EINVAL; - if (!(req->close.flags & IORING_CLOSE_FD_AND_FILE_SLOT) && - req->close.file_slot && req->close.fd) + if (req->close.file_slot && req->close.fd) return -EINVAL; return 0; @@ -6015,8 +5933,7 @@ static int io_close(struct io_kiocb *req, unsigned int issue_flags) if (req->close.file_slot) { ret = io_close_fixed(req, issue_flags); - if (ret || !(req->close.flags & IORING_CLOSE_FD_AND_FILE_SLOT)) - goto err; + goto err; } spin_lock(&files->file_lock); @@ -6160,8 +6077,6 @@ static int io_sendmsg_prep(struct io_kiocb *req, const struct io_uring_sqe *sqe) if (unlikely(sqe->file_index)) return -EINVAL; - if (unlikely(sqe->addr2 || sqe->file_index)) - return -EINVAL; sr->umsg = u64_to_user_ptr(READ_ONCE(sqe->addr)); sr->len = READ_ONCE(sqe->len); @@ -6398,8 +6313,6 @@ static int io_recvmsg_prep(struct io_kiocb *req, const struct io_uring_sqe *sqe) if (unlikely(sqe->file_index)) return -EINVAL; - if (unlikely(sqe->addr2 || sqe->file_index)) - return -EINVAL; sr->umsg = u64_to_user_ptr(READ_ONCE(sqe->addr)); sr->len = READ_ONCE(sqe->len); @@ -7037,7 +6950,8 @@ static void io_apoll_task_func(struct io_kiocb *req, bool *locked) io_req_complete_failed(req, ret); } -static void __io_poll_execute(struct io_kiocb *req, int mask, __poll_t events) +static void __io_poll_execute(struct io_kiocb *req, int mask, + __poll_t __maybe_unused events) { req->cqe.res = mask; /* @@ -7046,7 +6960,6 @@ static void __io_poll_execute(struct io_kiocb *req, int mask, __poll_t events) * CPU. We want to avoid pulling in req->apoll->events for that * case. */ - req->apoll_events = events; if (req->opcode == IORING_OP_POLL_ADD) req->io_task_work.func = io_poll_task_func; else @@ -7197,6 +7110,8 @@ static int __io_arm_poll_handler(struct io_kiocb *req, io_init_poll_iocb(poll, mask, io_poll_wake); poll->file = req->file; + req->apoll_events = poll->events; + ipt->pt._key = mask; ipt->req = req; ipt->error = 0; @@ -7227,8 +7142,11 @@ static int __io_arm_poll_handler(struct io_kiocb *req, if (mask) { /* can't multishot if failed, just queue the event we've got */ - if (unlikely(ipt->error || !ipt->nr_entries)) + if (unlikely(ipt->error || !ipt->nr_entries)) { poll->events |= EPOLLONESHOT; + req->apoll_events |= EPOLLONESHOT; + ipt->error = 0; + } __io_poll_execute(req, mask, poll->events); return 0; } @@ -7290,6 +7208,7 @@ static int io_arm_poll_handler(struct io_kiocb *req, unsigned issue_flags) mask |= EPOLLEXCLUSIVE; if (req->flags & REQ_F_POLLED) { apoll = req->apoll; + kfree(apoll->double_poll); } else if (!(issue_flags & IO_URING_F_UNLOCKED) && !list_empty(&ctx->apoll_cache)) { apoll = list_first_entry(&ctx->apoll_cache, struct async_poll, @@ -7475,7 +7394,7 @@ static int io_poll_add_prep(struct io_kiocb *req, const struct io_uring_sqe *sqe return -EINVAL; io_req_set_refcount(req); - req->apoll_events = poll->events = io_poll_parse_events(sqe, flags); + poll->events = io_poll_parse_events(sqe, flags); return 0; } @@ -7488,6 +7407,8 @@ static int io_poll_add(struct io_kiocb *req, unsigned int issue_flags) ipt.pt._qproc = io_poll_queue_proc; ret = __io_arm_poll_handler(req, &req->poll, &ipt, poll->events); + if (!ret && ipt.error) + req_set_fail(req); ret = ret ?: ipt.error; if (ret) __io_req_complete(req, issue_flags, ret, 0); @@ -8063,8 +7984,8 @@ static int io_files_update_with_index_alloc(struct io_kiocb *req, if (ret < 0) break; if (copy_to_user(&fds[done], &ret, sizeof(ret))) { - ret = -EFAULT; __io_close_fixed(req, issue_flags, ret); + ret = -EFAULT; break; } } @@ -8773,6 +8694,7 @@ static void io_queue_async(struct io_kiocb *req, int ret) * Queued up for async execution, worker will release * submit reference when the iocb is actually submitted. */ + io_kbuf_recycle(req, 0); io_queue_iowq(req, NULL); break; case IO_APOLL_OK: @@ -9788,11 +9710,19 @@ static void __io_sqe_files_unregister(struct io_ring_ctx *ctx) static int io_sqe_files_unregister(struct io_ring_ctx *ctx) { + unsigned nr = ctx->nr_user_files; int ret; if (!ctx->file_data) return -ENXIO; + + /* + * Quiesce may unlock ->uring_lock, and while it's not held + * prevent new requests using the table. + */ + ctx->nr_user_files = 0; ret = io_rsrc_ref_quiesce(ctx->file_data, ctx); + ctx->nr_user_files = nr; if (!ret) __io_sqe_files_unregister(ctx); return ret; @@ -10690,12 +10620,19 @@ static void __io_sqe_buffers_unregister(struct io_ring_ctx *ctx) static int io_sqe_buffers_unregister(struct io_ring_ctx *ctx) { + unsigned nr = ctx->nr_user_bufs; int ret; if (!ctx->buf_data) return -ENXIO; + /* + * Quiesce may unlock ->uring_lock, and while it's not held + * prevent new requests using the table. + */ + ctx->nr_user_bufs = 0; ret = io_rsrc_ref_quiesce(ctx->buf_data, ctx); + ctx->nr_user_bufs = nr; if (!ret) __io_sqe_buffers_unregister(ctx); return ret; @@ -13002,6 +12939,10 @@ static int io_register_pbuf_ring(struct io_ring_ctx *ctx, void __user *arg) if (!is_power_of_2(reg.ring_entries)) return -EINVAL; + /* cannot disambiguate full vs empty due to head/tail size */ + if (reg.ring_entries >= 65536) + return -EINVAL; + if (unlikely(reg.bgid < BGID_ARRAY && !ctx->io_bl)) { int ret = io_init_bl_list(ctx); if (ret) diff --git a/fs/jbd2/transaction.c b/fs/jbd2/transaction.c index e49bb0938376..e9c308ae475f 100644 --- a/fs/jbd2/transaction.c +++ b/fs/jbd2/transaction.c @@ -2114,7 +2114,7 @@ out: /** * jbd2_journal_try_to_free_buffers() - try to free page buffers. * @journal: journal for operation - * @page: to try and free + * @folio: Folio to detach data from. * * For all the buffers on this page, * if they are fully written out ordered data, move them onto BUF_CLEAN diff --git a/fs/netfs/buffered_read.c b/fs/netfs/buffered_read.c index 8742d22dfd2b..42f892c5712e 100644 --- a/fs/netfs/buffered_read.c +++ b/fs/netfs/buffered_read.c @@ -155,7 +155,7 @@ static void netfs_rreq_expand(struct netfs_io_request *rreq, void netfs_readahead(struct readahead_control *ractl) { struct netfs_io_request *rreq; - struct netfs_i_context *ctx = netfs_i_context(ractl->mapping->host); + struct netfs_inode *ctx = netfs_inode(ractl->mapping->host); int ret; _enter("%lx,%x", readahead_index(ractl), readahead_count(ractl)); @@ -215,7 +215,7 @@ int netfs_read_folio(struct file *file, struct folio *folio) { struct address_space *mapping = folio_file_mapping(folio); struct netfs_io_request *rreq; - struct netfs_i_context *ctx = netfs_i_context(mapping->host); + struct netfs_inode *ctx = netfs_inode(mapping->host); int ret; _enter("%lx", folio_index(folio)); @@ -297,6 +297,7 @@ zero_out: /** * netfs_write_begin - Helper to prepare for writing + * @ctx: The netfs context * @file: The file to read from * @mapping: The mapping to read from * @pos: File position at which the write will begin @@ -326,12 +327,12 @@ zero_out: * * This is usable whether or not caching is enabled. */ -int netfs_write_begin(struct file *file, struct address_space *mapping, +int netfs_write_begin(struct netfs_inode *ctx, + struct file *file, struct address_space *mapping, loff_t pos, unsigned int len, struct folio **_folio, void **_fsdata) { struct netfs_io_request *rreq; - struct netfs_i_context *ctx = netfs_i_context(file_inode(file )); struct folio *folio; unsigned int fgp_flags = FGP_LOCK | FGP_WRITE | FGP_CREAT | FGP_STABLE; pgoff_t index = pos >> PAGE_SHIFT; diff --git a/fs/netfs/internal.h b/fs/netfs/internal.h index b7b0e3d18d9e..43fac1b14e40 100644 --- a/fs/netfs/internal.h +++ b/fs/netfs/internal.h @@ -91,7 +91,7 @@ static inline void netfs_stat_d(atomic_t *stat) /* * Miscellaneous functions. */ -static inline bool netfs_is_cache_enabled(struct netfs_i_context *ctx) +static inline bool netfs_is_cache_enabled(struct netfs_inode *ctx) { #if IS_ENABLED(CONFIG_FSCACHE) struct fscache_cookie *cookie = ctx->cache; diff --git a/fs/netfs/objects.c b/fs/netfs/objects.c index e86107b30ba4..e17cdf53f6a7 100644 --- a/fs/netfs/objects.c +++ b/fs/netfs/objects.c @@ -18,7 +18,7 @@ struct netfs_io_request *netfs_alloc_request(struct address_space *mapping, { static atomic_t debug_ids; struct inode *inode = file ? file_inode(file) : mapping->host; - struct netfs_i_context *ctx = netfs_i_context(inode); + struct netfs_inode *ctx = netfs_inode(inode); struct netfs_io_request *rreq; int ret; @@ -75,10 +75,10 @@ static void netfs_free_request(struct work_struct *work) struct netfs_io_request *rreq = container_of(work, struct netfs_io_request, work); - netfs_clear_subrequests(rreq, false); - if (rreq->netfs_priv) - rreq->netfs_ops->cleanup(rreq->mapping, rreq->netfs_priv); trace_netfs_rreq(rreq, netfs_rreq_trace_free); + netfs_clear_subrequests(rreq, false); + if (rreq->netfs_ops->free_request) + rreq->netfs_ops->free_request(rreq); if (rreq->cache_resources.ops) rreq->cache_resources.ops->end_operation(&rreq->cache_resources); kfree(rreq); diff --git a/fs/nfs/callback_proc.c b/fs/nfs/callback_proc.c index c8520284dda7..c1eda73254e1 100644 --- a/fs/nfs/callback_proc.c +++ b/fs/nfs/callback_proc.c @@ -288,6 +288,7 @@ static u32 initiate_file_draining(struct nfs_client *clp, rv = NFS4_OK; break; case -ENOENT: + set_bit(NFS_LAYOUT_DRAIN, &lo->plh_flags); /* Embrace your forgetfulness! */ rv = NFS4ERR_NOMATCHING_LAYOUT; diff --git a/fs/nfs/dir.c b/fs/nfs/dir.c index a8ecdd527662..0c4e8dd6aa96 100644 --- a/fs/nfs/dir.c +++ b/fs/nfs/dir.c @@ -2124,6 +2124,7 @@ int nfs_atomic_open(struct inode *dir, struct dentry *dentry, } goto out; } + file->f_mode |= FMODE_CAN_ODIRECT; err = nfs_finish_open(ctx, ctx->dentry, file, open_flags); trace_nfs_atomic_open_exit(dir, ctx, open_flags, err); diff --git a/fs/nfs/nfs4file.c b/fs/nfs/nfs4file.c index 03d3a270eff4..e88f6b18445e 100644 --- a/fs/nfs/nfs4file.c +++ b/fs/nfs/nfs4file.c @@ -93,6 +93,7 @@ nfs4_file_open(struct inode *inode, struct file *filp) nfs_file_set_open_context(filp, ctx); nfs_fscache_open_file(inode, filp); err = 0; + filp->f_mode |= FMODE_CAN_ODIRECT; out_put_ctx: put_nfs_open_context(ctx); diff --git a/fs/nfs/pnfs.c b/fs/nfs/pnfs.c index 68a87be3e6f9..41a9b6b58fb9 100644 --- a/fs/nfs/pnfs.c +++ b/fs/nfs/pnfs.c @@ -469,6 +469,7 @@ pnfs_mark_layout_stateid_invalid(struct pnfs_layout_hdr *lo, pnfs_clear_lseg_state(lseg, lseg_list); pnfs_clear_layoutreturn_info(lo); pnfs_free_returned_lsegs(lo, lseg_list, &range, 0); + set_bit(NFS_LAYOUT_DRAIN, &lo->plh_flags); if (test_bit(NFS_LAYOUT_RETURN, &lo->plh_flags) && !test_and_set_bit(NFS_LAYOUT_RETURN_LOCK, &lo->plh_flags)) pnfs_clear_layoutreturn_waitbit(lo); @@ -1917,8 +1918,9 @@ static void nfs_layoutget_begin(struct pnfs_layout_hdr *lo) static void nfs_layoutget_end(struct pnfs_layout_hdr *lo) { - if (atomic_dec_and_test(&lo->plh_outstanding)) - wake_up_var(&lo->plh_outstanding); + if (atomic_dec_and_test(&lo->plh_outstanding) && + test_and_clear_bit(NFS_LAYOUT_DRAIN, &lo->plh_flags)) + wake_up_bit(&lo->plh_flags, NFS_LAYOUT_DRAIN); } static bool pnfs_is_first_layoutget(struct pnfs_layout_hdr *lo) @@ -2025,11 +2027,11 @@ lookup_again: * If the layout segment list is empty, but there are outstanding * layoutget calls, then they might be subject to a layoutrecall. */ - if ((list_empty(&lo->plh_segs) || !pnfs_layout_is_valid(lo)) && + if (test_bit(NFS_LAYOUT_DRAIN, &lo->plh_flags) && atomic_read(&lo->plh_outstanding) != 0) { spin_unlock(&ino->i_lock); - lseg = ERR_PTR(wait_var_event_killable(&lo->plh_outstanding, - !atomic_read(&lo->plh_outstanding))); + lseg = ERR_PTR(wait_on_bit(&lo->plh_flags, NFS_LAYOUT_DRAIN, + TASK_KILLABLE)); if (IS_ERR(lseg)) goto out_put_layout_hdr; pnfs_put_layout_hdr(lo); @@ -2152,6 +2154,12 @@ lookup_again: case -ERECALLCONFLICT: case -EAGAIN: break; + case -ENODATA: + /* The server returned NFS4ERR_LAYOUTUNAVAILABLE */ + pnfs_layout_set_fail_bit( + lo, pnfs_iomode_to_fail_bit(iomode)); + lseg = NULL; + goto out_put_layout_hdr; default: if (!nfs_error_is_fatal(PTR_ERR(lseg))) { pnfs_layout_clear_fail_bit(lo, pnfs_iomode_to_fail_bit(iomode)); @@ -2407,7 +2415,8 @@ pnfs_layout_process(struct nfs4_layoutget *lgp) goto out_forget; } - if (!pnfs_layout_is_valid(lo) && !pnfs_is_first_layoutget(lo)) + if (test_bit(NFS_LAYOUT_DRAIN, &lo->plh_flags) && + !pnfs_is_first_layoutget(lo)) goto out_forget; if (nfs4_stateid_match_other(&lo->plh_stateid, &res->stateid)) { diff --git a/fs/nfs/pnfs.h b/fs/nfs/pnfs.h index 07f11489e4e9..f331f067691b 100644 --- a/fs/nfs/pnfs.h +++ b/fs/nfs/pnfs.h @@ -105,6 +105,7 @@ enum { NFS_LAYOUT_FIRST_LAYOUTGET, /* Serialize first layoutget */ NFS_LAYOUT_INODE_FREEING, /* The inode is being freed */ NFS_LAYOUT_HASHED, /* The layout visible */ + NFS_LAYOUT_DRAIN, }; enum layoutdriver_policy_flags { diff --git a/fs/nfsd/filecache.c b/fs/nfsd/filecache.c index f172412447f5..9cb2d590c036 100644 --- a/fs/nfsd/filecache.c +++ b/fs/nfsd/filecache.c @@ -309,11 +309,12 @@ nfsd_file_put(struct nfsd_file *nf) if (test_bit(NFSD_FILE_HASHED, &nf->nf_flags) == 0) { nfsd_file_flush(nf); nfsd_file_put_noref(nf); - } else { + } else if (nf->nf_file) { nfsd_file_put_noref(nf); - if (nf->nf_file) - nfsd_file_schedule_laundrette(); - } + nfsd_file_schedule_laundrette(); + } else + nfsd_file_put_noref(nf); + if (atomic_long_read(&nfsd_filecache_count) >= NFSD_FILE_LRU_LIMIT) nfsd_file_gc(); } diff --git a/fs/quota/dquot.c b/fs/quota/dquot.c index a74aef99bd3d..09d1307959d0 100644 --- a/fs/quota/dquot.c +++ b/fs/quota/dquot.c @@ -79,6 +79,7 @@ #include <linux/capability.h> #include <linux/quotaops.h> #include <linux/blkdev.h> +#include <linux/sched/mm.h> #include "../internal.h" /* ugh */ #include <linux/uaccess.h> @@ -425,9 +426,11 @@ EXPORT_SYMBOL(mark_info_dirty); int dquot_acquire(struct dquot *dquot) { int ret = 0, ret2 = 0; + unsigned int memalloc; struct quota_info *dqopt = sb_dqopt(dquot->dq_sb); mutex_lock(&dquot->dq_lock); + memalloc = memalloc_nofs_save(); if (!test_bit(DQ_READ_B, &dquot->dq_flags)) { ret = dqopt->ops[dquot->dq_id.type]->read_dqblk(dquot); if (ret < 0) @@ -458,6 +461,7 @@ int dquot_acquire(struct dquot *dquot) smp_mb__before_atomic(); set_bit(DQ_ACTIVE_B, &dquot->dq_flags); out_iolock: + memalloc_nofs_restore(memalloc); mutex_unlock(&dquot->dq_lock); return ret; } @@ -469,9 +473,11 @@ EXPORT_SYMBOL(dquot_acquire); int dquot_commit(struct dquot *dquot) { int ret = 0; + unsigned int memalloc; struct quota_info *dqopt = sb_dqopt(dquot->dq_sb); mutex_lock(&dquot->dq_lock); + memalloc = memalloc_nofs_save(); if (!clear_dquot_dirty(dquot)) goto out_lock; /* Inactive dquot can be only if there was error during read/init @@ -481,6 +487,7 @@ int dquot_commit(struct dquot *dquot) else ret = -EIO; out_lock: + memalloc_nofs_restore(memalloc); mutex_unlock(&dquot->dq_lock); return ret; } @@ -492,9 +499,11 @@ EXPORT_SYMBOL(dquot_commit); int dquot_release(struct dquot *dquot) { int ret = 0, ret2 = 0; + unsigned int memalloc; struct quota_info *dqopt = sb_dqopt(dquot->dq_sb); mutex_lock(&dquot->dq_lock); + memalloc = memalloc_nofs_save(); /* Check whether we are not racing with some other dqget() */ if (dquot_is_busy(dquot)) goto out_dqlock; @@ -510,6 +519,7 @@ int dquot_release(struct dquot *dquot) } clear_bit(DQ_ACTIVE_B, &dquot->dq_flags); out_dqlock: + memalloc_nofs_restore(memalloc); mutex_unlock(&dquot->dq_lock); return ret; } diff --git a/fs/tracefs/inode.c b/fs/tracefs/inode.c index de7252715b12..81d26abf486f 100644 --- a/fs/tracefs/inode.c +++ b/fs/tracefs/inode.c @@ -553,7 +553,7 @@ struct dentry *tracefs_create_dir(const char *name, struct dentry *parent) * * Only one instances directory is allowed. * - * The instances directory is special as it allows for mkdir and rmdir to + * The instances directory is special as it allows for mkdir and rmdir * to be done by userspace. When a mkdir or rmdir is performed, the inode * locks are released and the methods passed in (@mkdir and @rmdir) are * called without locks and with the name of the directory being created diff --git a/fs/xfs/libxfs/xfs_attr.c b/fs/xfs/libxfs/xfs_attr.c index 836ab1b8ed7b..1824f61621a2 100644 --- a/fs/xfs/libxfs/xfs_attr.c +++ b/fs/xfs/libxfs/xfs_attr.c @@ -997,9 +997,11 @@ xfs_attr_set( /* * We have no control over the attribute names that userspace passes us * to remove, so we have to allow the name lookup prior to attribute - * removal to fail as well. + * removal to fail as well. Preserve the logged flag, since we need + * to pass that through to the logging code. */ - args->op_flags = XFS_DA_OP_OKNOENT; + args->op_flags = XFS_DA_OP_OKNOENT | + (args->op_flags & XFS_DA_OP_LOGGED); if (args->value) { XFS_STATS_INC(mp, xs_attr_set); @@ -1439,12 +1441,11 @@ static int xfs_attr_node_try_addname( struct xfs_attr_intent *attr) { - struct xfs_da_args *args = attr->xattri_da_args; struct xfs_da_state *state = attr->xattri_da_state; struct xfs_da_state_blk *blk; int error; - trace_xfs_attr_node_addname(args); + trace_xfs_attr_node_addname(state->args); blk = &state->path.blk[state->path.active-1]; ASSERT(blk->magic == XFS_ATTR_LEAF_MAGIC); diff --git a/fs/xfs/libxfs/xfs_attr.h b/fs/xfs/libxfs/xfs_attr.h index e329da3e7afa..b4a2fc77017e 100644 --- a/fs/xfs/libxfs/xfs_attr.h +++ b/fs/xfs/libxfs/xfs_attr.h @@ -28,16 +28,6 @@ struct xfs_attr_list_context; */ #define ATTR_MAX_VALUELEN (64*1024) /* max length of a value */ -static inline bool xfs_has_larp(struct xfs_mount *mp) -{ -#ifdef DEBUG - /* Logged xattrs require a V5 super for log_incompat */ - return xfs_has_crc(mp) && xfs_globals.larp; -#else - return false; -#endif -} - /* * Kernel-internal version of the attrlist cursor. */ @@ -624,7 +614,7 @@ static inline enum xfs_delattr_state xfs_attr_init_replace_state(struct xfs_da_args *args) { args->op_flags |= XFS_DA_OP_ADDNAME | XFS_DA_OP_REPLACE; - if (xfs_has_larp(args->dp->i_mount)) + if (args->op_flags & XFS_DA_OP_LOGGED) return xfs_attr_init_remove_state(args); return xfs_attr_init_add_state(args); } diff --git a/fs/xfs/libxfs/xfs_attr_leaf.c b/fs/xfs/libxfs/xfs_attr_leaf.c index 15a990409463..37e7c33f6283 100644 --- a/fs/xfs/libxfs/xfs_attr_leaf.c +++ b/fs/xfs/libxfs/xfs_attr_leaf.c @@ -1530,7 +1530,7 @@ xfs_attr3_leaf_add_work( if (tmp) entry->flags |= XFS_ATTR_LOCAL; if (args->op_flags & XFS_DA_OP_REPLACE) { - if (!xfs_has_larp(mp)) + if (!(args->op_flags & XFS_DA_OP_LOGGED)) entry->flags |= XFS_ATTR_INCOMPLETE; if ((args->blkno2 == args->blkno) && (args->index2 <= args->index)) { diff --git a/fs/xfs/libxfs/xfs_da_btree.h b/fs/xfs/libxfs/xfs_da_btree.h index d33b7686a0b3..ffa3df5b2893 100644 --- a/fs/xfs/libxfs/xfs_da_btree.h +++ b/fs/xfs/libxfs/xfs_da_btree.h @@ -92,6 +92,7 @@ typedef struct xfs_da_args { #define XFS_DA_OP_NOTIME (1u << 5) /* don't update inode timestamps */ #define XFS_DA_OP_REMOVE (1u << 6) /* this is a remove operation */ #define XFS_DA_OP_RECOVERY (1u << 7) /* Log recovery operation */ +#define XFS_DA_OP_LOGGED (1u << 8) /* Use intent items to track op */ #define XFS_DA_OP_FLAGS \ { XFS_DA_OP_JUSTCHECK, "JUSTCHECK" }, \ @@ -101,7 +102,8 @@ typedef struct xfs_da_args { { XFS_DA_OP_CILOOKUP, "CILOOKUP" }, \ { XFS_DA_OP_NOTIME, "NOTIME" }, \ { XFS_DA_OP_REMOVE, "REMOVE" }, \ - { XFS_DA_OP_RECOVERY, "RECOVERY" } + { XFS_DA_OP_RECOVERY, "RECOVERY" }, \ + { XFS_DA_OP_LOGGED, "LOGGED" } /* * Storage for holding state during Btree searches and split/join ops. diff --git a/fs/xfs/xfs_attr_item.c b/fs/xfs/xfs_attr_item.c index 4a28c2d77070..135d44133477 100644 --- a/fs/xfs/xfs_attr_item.c +++ b/fs/xfs/xfs_attr_item.c @@ -413,18 +413,20 @@ xfs_attr_create_intent( struct xfs_mount *mp = tp->t_mountp; struct xfs_attri_log_item *attrip; struct xfs_attr_intent *attr; + struct xfs_da_args *args; ASSERT(count == 1); - if (!xfs_sb_version_haslogxattrs(&mp->m_sb)) - return NULL; - /* * Each attr item only performs one attribute operation at a time, so * this is a list of one */ attr = list_first_entry_or_null(items, struct xfs_attr_intent, xattri_list); + args = attr->xattri_da_args; + + if (!(args->op_flags & XFS_DA_OP_LOGGED)) + return NULL; /* * Create a buffer to store the attribute name and value. This buffer @@ -432,8 +434,6 @@ xfs_attr_create_intent( * and the lower level xattr log items. */ if (!attr->xattri_nameval) { - struct xfs_da_args *args = attr->xattri_da_args; - /* * Transfer our reference to the name/value buffer to the * deferred work state structure. @@ -617,7 +617,10 @@ xfs_attri_item_recover( args->namelen = nv->name.i_len; args->hashval = xfs_da_hashname(args->name, args->namelen); args->attr_filter = attrp->alfi_attr_filter & XFS_ATTRI_FILTER_MASK; - args->op_flags = XFS_DA_OP_RECOVERY | XFS_DA_OP_OKNOENT; + args->op_flags = XFS_DA_OP_RECOVERY | XFS_DA_OP_OKNOENT | + XFS_DA_OP_LOGGED; + + ASSERT(xfs_sb_version_haslogxattrs(&mp->m_sb)); switch (attr->xattri_op_flags) { case XFS_ATTRI_OP_FLAGS_SET: diff --git a/fs/xfs/xfs_ioctl.c b/fs/xfs/xfs_ioctl.c index 5a364a7d58fd..0d67ff8a8961 100644 --- a/fs/xfs/xfs_ioctl.c +++ b/fs/xfs/xfs_ioctl.c @@ -1096,7 +1096,8 @@ xfs_flags2diflags2( { uint64_t di_flags2 = (ip->i_diflags2 & (XFS_DIFLAG2_REFLINK | - XFS_DIFLAG2_BIGTIME)); + XFS_DIFLAG2_BIGTIME | + XFS_DIFLAG2_NREXT64)); if (xflags & FS_XFLAG_DAX) di_flags2 |= XFS_DIFLAG2_DAX; diff --git a/fs/xfs/xfs_xattr.c b/fs/xfs/xfs_xattr.c index 35e13e125ec6..c325a28b89a8 100644 --- a/fs/xfs/xfs_xattr.c +++ b/fs/xfs/xfs_xattr.c @@ -68,6 +68,18 @@ xfs_attr_rele_log_assist( xlog_drop_incompat_feat(mp->m_log); } +static inline bool +xfs_attr_want_log_assist( + struct xfs_mount *mp) +{ +#ifdef DEBUG + /* Logged xattrs require a V5 super for log_incompat */ + return xfs_has_crc(mp) && xfs_globals.larp; +#else + return false; +#endif +} + /* * Set or remove an xattr, having grabbed the appropriate logging resources * prior to calling libxfs. @@ -80,11 +92,14 @@ xfs_attr_change( bool use_logging = false; int error; - if (xfs_has_larp(mp)) { + ASSERT(!(args->op_flags & XFS_DA_OP_LOGGED)); + + if (xfs_attr_want_log_assist(mp)) { error = xfs_attr_grab_log_assist(mp); if (error) return error; + args->op_flags |= XFS_DA_OP_LOGGED; use_logging = true; } diff --git a/fs/zonefs/super.c b/fs/zonefs/super.c index bcb21aea990a..053299758deb 100644 --- a/fs/zonefs/super.c +++ b/fs/zonefs/super.c @@ -110,15 +110,51 @@ static inline void zonefs_i_size_write(struct inode *inode, loff_t isize) } } -static int zonefs_iomap_begin(struct inode *inode, loff_t offset, loff_t length, - unsigned int flags, struct iomap *iomap, - struct iomap *srcmap) +static int zonefs_read_iomap_begin(struct inode *inode, loff_t offset, + loff_t length, unsigned int flags, + struct iomap *iomap, struct iomap *srcmap) { struct zonefs_inode_info *zi = ZONEFS_I(inode); struct super_block *sb = inode->i_sb; loff_t isize; - /* All I/Os should always be within the file maximum size */ + /* + * All blocks are always mapped below EOF. If reading past EOF, + * act as if there is a hole up to the file maximum size. + */ + mutex_lock(&zi->i_truncate_mutex); + iomap->bdev = inode->i_sb->s_bdev; + iomap->offset = ALIGN_DOWN(offset, sb->s_blocksize); + isize = i_size_read(inode); + if (iomap->offset >= isize) { + iomap->type = IOMAP_HOLE; + iomap->addr = IOMAP_NULL_ADDR; + iomap->length = length; + } else { + iomap->type = IOMAP_MAPPED; + iomap->addr = (zi->i_zsector << SECTOR_SHIFT) + iomap->offset; + iomap->length = isize - iomap->offset; + } + mutex_unlock(&zi->i_truncate_mutex); + + trace_zonefs_iomap_begin(inode, iomap); + + return 0; +} + +static const struct iomap_ops zonefs_read_iomap_ops = { + .iomap_begin = zonefs_read_iomap_begin, +}; + +static int zonefs_write_iomap_begin(struct inode *inode, loff_t offset, + loff_t length, unsigned int flags, + struct iomap *iomap, struct iomap *srcmap) +{ + struct zonefs_inode_info *zi = ZONEFS_I(inode); + struct super_block *sb = inode->i_sb; + loff_t isize; + + /* All write I/Os should always be within the file maximum size */ if (WARN_ON_ONCE(offset + length > zi->i_max_size)) return -EIO; @@ -128,7 +164,7 @@ static int zonefs_iomap_begin(struct inode *inode, loff_t offset, loff_t length, * operation. */ if (WARN_ON_ONCE(zi->i_ztype == ZONEFS_ZTYPE_SEQ && - (flags & IOMAP_WRITE) && !(flags & IOMAP_DIRECT))) + !(flags & IOMAP_DIRECT))) return -EIO; /* @@ -137,47 +173,44 @@ static int zonefs_iomap_begin(struct inode *inode, loff_t offset, loff_t length, * write pointer) and unwriten beyond. */ mutex_lock(&zi->i_truncate_mutex); + iomap->bdev = inode->i_sb->s_bdev; + iomap->offset = ALIGN_DOWN(offset, sb->s_blocksize); + iomap->addr = (zi->i_zsector << SECTOR_SHIFT) + iomap->offset; isize = i_size_read(inode); - if (offset >= isize) + if (iomap->offset >= isize) { iomap->type = IOMAP_UNWRITTEN; - else + iomap->length = zi->i_max_size - iomap->offset; + } else { iomap->type = IOMAP_MAPPED; - if (flags & IOMAP_WRITE) - length = zi->i_max_size - offset; - else - length = min(length, isize - offset); + iomap->length = isize - iomap->offset; + } mutex_unlock(&zi->i_truncate_mutex); - iomap->offset = ALIGN_DOWN(offset, sb->s_blocksize); - iomap->length = ALIGN(offset + length, sb->s_blocksize) - iomap->offset; - iomap->bdev = inode->i_sb->s_bdev; - iomap->addr = (zi->i_zsector << SECTOR_SHIFT) + iomap->offset; - trace_zonefs_iomap_begin(inode, iomap); return 0; } -static const struct iomap_ops zonefs_iomap_ops = { - .iomap_begin = zonefs_iomap_begin, +static const struct iomap_ops zonefs_write_iomap_ops = { + .iomap_begin = zonefs_write_iomap_begin, }; static int zonefs_read_folio(struct file *unused, struct folio *folio) { - return iomap_read_folio(folio, &zonefs_iomap_ops); + return iomap_read_folio(folio, &zonefs_read_iomap_ops); } static void zonefs_readahead(struct readahead_control *rac) { - iomap_readahead(rac, &zonefs_iomap_ops); + iomap_readahead(rac, &zonefs_read_iomap_ops); } /* * Map blocks for page writeback. This is used only on conventional zone files, * which implies that the page range can only be within the fixed inode size. */ -static int zonefs_map_blocks(struct iomap_writepage_ctx *wpc, - struct inode *inode, loff_t offset) +static int zonefs_write_map_blocks(struct iomap_writepage_ctx *wpc, + struct inode *inode, loff_t offset) { struct zonefs_inode_info *zi = ZONEFS_I(inode); @@ -191,12 +224,12 @@ static int zonefs_map_blocks(struct iomap_writepage_ctx *wpc, offset < wpc->iomap.offset + wpc->iomap.length) return 0; - return zonefs_iomap_begin(inode, offset, zi->i_max_size - offset, - IOMAP_WRITE, &wpc->iomap, NULL); + return zonefs_write_iomap_begin(inode, offset, zi->i_max_size - offset, + IOMAP_WRITE, &wpc->iomap, NULL); } static const struct iomap_writeback_ops zonefs_writeback_ops = { - .map_blocks = zonefs_map_blocks, + .map_blocks = zonefs_write_map_blocks, }; static int zonefs_writepage(struct page *page, struct writeback_control *wbc) @@ -226,7 +259,8 @@ static int zonefs_swap_activate(struct swap_info_struct *sis, return -EINVAL; } - return iomap_swapfile_activate(sis, swap_file, span, &zonefs_iomap_ops); + return iomap_swapfile_activate(sis, swap_file, span, + &zonefs_read_iomap_ops); } static const struct address_space_operations zonefs_file_aops = { @@ -647,7 +681,7 @@ static vm_fault_t zonefs_filemap_page_mkwrite(struct vm_fault *vmf) /* Serialize against truncates */ filemap_invalidate_lock_shared(inode->i_mapping); - ret = iomap_page_mkwrite(vmf, &zonefs_iomap_ops); + ret = iomap_page_mkwrite(vmf, &zonefs_write_iomap_ops); filemap_invalidate_unlock_shared(inode->i_mapping); sb_end_pagefault(inode->i_sb); @@ -899,7 +933,7 @@ static ssize_t zonefs_file_dio_write(struct kiocb *iocb, struct iov_iter *from) if (append) ret = zonefs_file_dio_append(iocb, from); else - ret = iomap_dio_rw(iocb, from, &zonefs_iomap_ops, + ret = iomap_dio_rw(iocb, from, &zonefs_write_iomap_ops, &zonefs_write_dio_ops, 0, NULL, 0); if (zi->i_ztype == ZONEFS_ZTYPE_SEQ && (ret > 0 || ret == -EIOCBQUEUED)) { @@ -948,7 +982,7 @@ static ssize_t zonefs_file_buffered_write(struct kiocb *iocb, if (ret <= 0) goto inode_unlock; - ret = iomap_file_buffered_write(iocb, from, &zonefs_iomap_ops); + ret = iomap_file_buffered_write(iocb, from, &zonefs_write_iomap_ops); if (ret > 0) iocb->ki_pos += ret; else if (ret == -EIO) @@ -1041,7 +1075,7 @@ static ssize_t zonefs_file_read_iter(struct kiocb *iocb, struct iov_iter *to) goto inode_unlock; } file_accessed(iocb->ki_filp); - ret = iomap_dio_rw(iocb, to, &zonefs_iomap_ops, + ret = iomap_dio_rw(iocb, to, &zonefs_read_iomap_ops, &zonefs_read_dio_ops, 0, NULL, 0); } else { ret = generic_file_read_iter(iocb, to); @@ -1085,7 +1119,8 @@ static int zonefs_seq_file_write_open(struct inode *inode) if (sbi->s_mount_opts & ZONEFS_MNTOPT_EXPLICIT_OPEN) { - if (wro > sbi->s_max_wro_seq_files) { + if (sbi->s_max_wro_seq_files + && wro > sbi->s_max_wro_seq_files) { atomic_dec(&sbi->s_wro_seq_files); ret = -EBUSY; goto unlock; @@ -1760,12 +1795,6 @@ static int zonefs_fill_super(struct super_block *sb, void *data, int silent) atomic_set(&sbi->s_wro_seq_files, 0); sbi->s_max_wro_seq_files = bdev_max_open_zones(sb->s_bdev); - if (!sbi->s_max_wro_seq_files && - sbi->s_mount_opts & ZONEFS_MNTOPT_EXPLICIT_OPEN) { - zonefs_info(sb, "No open zones limit. Ignoring explicit_open mount option\n"); - sbi->s_mount_opts &= ~ZONEFS_MNTOPT_EXPLICIT_OPEN; - } - atomic_set(&sbi->s_active_seq_files, 0); sbi->s_max_active_seq_files = bdev_max_active_zones(sb->s_bdev); @@ -1790,6 +1819,14 @@ static int zonefs_fill_super(struct super_block *sb, void *data, int silent) zonefs_info(sb, "Mounting %u zones", blkdev_nr_zones(sb->s_bdev->bd_disk)); + if (!sbi->s_max_wro_seq_files && + !sbi->s_max_active_seq_files && + sbi->s_mount_opts & ZONEFS_MNTOPT_EXPLICIT_OPEN) { + zonefs_info(sb, + "No open and active zone limits. Ignoring explicit_open mount option\n"); + sbi->s_mount_opts &= ~ZONEFS_MNTOPT_EXPLICIT_OPEN; + } + /* Create root directory inode */ ret = -ENOMEM; inode = new_inode(sb); |