diff options
Diffstat (limited to 'fs/nfs/dir.c')
| -rw-r--r-- | fs/nfs/dir.c | 2299 |
1 files changed, 1555 insertions, 744 deletions
diff --git a/fs/nfs/dir.c b/fs/nfs/dir.c index 3522b1249019..ea9f6ca8f30f 100644 --- a/fs/nfs/dir.c +++ b/fs/nfs/dir.c @@ -1,3 +1,4 @@ +// SPDX-License-Identifier: GPL-2.0-only /* * linux/fs/nfs/dir.c * @@ -17,6 +18,7 @@ * 6 Jun 1999 Cache readdir lookups in the page cache. -DaveM */ +#include <linux/compat.h> #include <linux/module.h> #include <linux/time.h> #include <linux/errno.h> @@ -37,6 +39,7 @@ #include <linux/sched.h> #include <linux/kmemleak.h> #include <linux/xattr.h> +#include <linux/hash.h> #include "delegation.h" #include "iostat.h" @@ -52,34 +55,43 @@ static int nfs_closedir(struct inode *, struct file *); static int nfs_readdir(struct file *, struct dir_context *); static int nfs_fsync_dir(struct file *, loff_t, loff_t, int); static loff_t nfs_llseek_dir(struct file *, loff_t, int); -static void nfs_readdir_clear_array(struct page*); +static void nfs_readdir_clear_array(struct folio *); +static int nfs_do_create(struct inode *dir, struct dentry *dentry, + umode_t mode, int open_flags); const struct file_operations nfs_dir_operations = { .llseek = nfs_llseek_dir, .read = generic_read_dir, - .iterate = nfs_readdir, + .iterate_shared = nfs_readdir, .open = nfs_opendir, .release = nfs_closedir, .fsync = nfs_fsync_dir, }; const struct address_space_operations nfs_dir_aops = { - .freepage = nfs_readdir_clear_array, + .free_folio = nfs_readdir_clear_array, }; -static struct nfs_open_dir_context *alloc_nfs_open_dir_context(struct inode *dir, struct rpc_cred *cred) +#define NFS_INIT_DTSIZE PAGE_SIZE + +static struct nfs_open_dir_context * +alloc_nfs_open_dir_context(struct inode *dir) { struct nfs_inode *nfsi = NFS_I(dir); struct nfs_open_dir_context *ctx; - ctx = kmalloc(sizeof(*ctx), GFP_KERNEL); + + ctx = kzalloc(sizeof(*ctx), GFP_KERNEL_ACCOUNT); if (ctx != NULL) { - ctx->duped = 0; ctx->attr_gencount = nfsi->attr_gencount; - ctx->dir_cookie = 0; - ctx->dup_cookie = 0; - ctx->cred = get_rpccred(cred); + ctx->dtsize = NFS_INIT_DTSIZE; spin_lock(&dir->i_lock); - list_add(&ctx->list, &nfsi->open_files); + if (list_empty(&nfsi->open_files) && + (nfsi->cache_validity & NFS_INO_DATA_INVAL_DEFER)) + nfs_set_cache_invalid(dir, + NFS_INO_INVALID_DATA | + NFS_INO_REVAL_FORCED); + list_add_tail_rcu(&ctx->list, &nfsi->open_files); + memcpy(ctx->verf, nfsi->cookieverf, sizeof(ctx->verf)); spin_unlock(&dir->i_lock); return ctx; } @@ -89,10 +101,9 @@ static struct nfs_open_dir_context *alloc_nfs_open_dir_context(struct inode *dir static void put_nfs_open_dir_context(struct inode *dir, struct nfs_open_dir_context *ctx) { spin_lock(&dir->i_lock); - list_del(&ctx->list); + list_del_rcu(&ctx->list); spin_unlock(&dir->i_lock); - put_rpccred(ctx->cred); - kfree(ctx); + kfree_rcu(ctx, rcu_head); } /* @@ -103,30 +114,18 @@ nfs_opendir(struct inode *inode, struct file *filp) { int res = 0; struct nfs_open_dir_context *ctx; - struct rpc_cred *cred; dfprintk(FILE, "NFS: open dir(%pD2)\n", filp); nfs_inc_stats(inode, NFSIOS_VFSOPEN); - cred = rpc_lookup_cred(); - if (IS_ERR(cred)) - return PTR_ERR(cred); - ctx = alloc_nfs_open_dir_context(inode, cred); + ctx = alloc_nfs_open_dir_context(inode); if (IS_ERR(ctx)) { res = PTR_ERR(ctx); goto out; } filp->private_data = ctx; - if (filp->f_path.dentry == filp->f_path.mnt->mnt_root) { - /* This is a mountpoint, so d_revalidate will never - * have been called, so we need to refresh the - * inode (for close-open consistency) ourselves. - */ - __nfs_revalidate_inode(NFS_SERVER(inode), inode); - } out: - put_rpccred(cred); return res; } @@ -140,48 +139,135 @@ nfs_closedir(struct inode *inode, struct file *filp) struct nfs_cache_array_entry { u64 cookie; u64 ino; - struct qstr string; + const char *name; + unsigned int name_len; unsigned char d_type; }; struct nfs_cache_array { - int size; - int eof_index; + u64 change_attr; u64 last_cookie; - struct nfs_cache_array_entry array[0]; + unsigned int size; + unsigned char folio_full : 1, + folio_is_eof : 1, + cookies_are_ordered : 1; + struct nfs_cache_array_entry array[] __counted_by(size); }; -typedef int (*decode_dirent_t)(struct xdr_stream *, struct nfs_entry *, bool); -typedef struct { +struct nfs_readdir_descriptor { struct file *file; - struct page *page; + struct folio *folio; struct dir_context *ctx; - unsigned long page_index; - u64 *dir_cookie; + pgoff_t folio_index; + pgoff_t folio_index_max; + u64 dir_cookie; u64 last_cookie; loff_t current_index; - decode_dirent_t decode; + __be32 verf[NFS_DIR_VERIFIER_SIZE]; + unsigned long dir_verifier; unsigned long timestamp; unsigned long gencount; + unsigned long attr_gencount; unsigned int cache_entry_index; + unsigned int buffer_fills; + unsigned int dtsize; + bool clear_cache; bool plus; + bool eob; bool eof; -} nfs_readdir_descriptor_t; +}; + +static void nfs_set_dtsize(struct nfs_readdir_descriptor *desc, unsigned int sz) +{ + struct nfs_server *server = NFS_SERVER(file_inode(desc->file)); + unsigned int maxsize = server->dtsize; + + if (sz > maxsize) + sz = maxsize; + if (sz < NFS_MIN_FILE_IO_SIZE) + sz = NFS_MIN_FILE_IO_SIZE; + desc->dtsize = sz; +} + +static void nfs_shrink_dtsize(struct nfs_readdir_descriptor *desc) +{ + nfs_set_dtsize(desc, desc->dtsize >> 1); +} + +static void nfs_grow_dtsize(struct nfs_readdir_descriptor *desc) +{ + nfs_set_dtsize(desc, desc->dtsize << 1); +} + +static void nfs_readdir_folio_init_array(struct folio *folio, u64 last_cookie, + u64 change_attr) +{ + struct nfs_cache_array *array; + + array = kmap_local_folio(folio, 0); + array->change_attr = change_attr; + array->last_cookie = last_cookie; + array->size = 0; + array->folio_full = 0; + array->folio_is_eof = 0; + array->cookies_are_ordered = 1; + kunmap_local(array); +} /* * we are freeing strings created by nfs_add_to_readdir_array() */ -static -void nfs_readdir_clear_array(struct page *page) +static void nfs_readdir_clear_array(struct folio *folio) { struct nfs_cache_array *array; - int i; + unsigned int i; - array = kmap_atomic(page); + array = kmap_local_folio(folio, 0); for (i = 0; i < array->size; i++) - kfree(array->array[i].string.name); - kunmap_atomic(array); + kfree(array->array[i].name); + array->size = 0; + kunmap_local(array); +} + +static void nfs_readdir_folio_reinit_array(struct folio *folio, u64 last_cookie, + u64 change_attr) +{ + nfs_readdir_clear_array(folio); + nfs_readdir_folio_init_array(folio, last_cookie, change_attr); +} + +static struct folio * +nfs_readdir_folio_array_alloc(u64 last_cookie, gfp_t gfp_flags) +{ + struct folio *folio = folio_alloc(gfp_flags, 0); + if (folio) + nfs_readdir_folio_init_array(folio, last_cookie, 0); + return folio; +} + +static void nfs_readdir_folio_array_free(struct folio *folio) +{ + if (folio) { + nfs_readdir_clear_array(folio); + folio_put(folio); + } +} + +static u64 nfs_readdir_array_index_cookie(struct nfs_cache_array *array) +{ + return array->size == 0 ? array->last_cookie : array->array[0].cookie; +} + +static void nfs_readdir_array_set_eof(struct nfs_cache_array *array) +{ + array->folio_is_eof = 1; + array->folio_full = 1; +} + +static bool nfs_readdir_array_is_full(struct nfs_cache_array *array) +{ + return array->folio_full; } /* @@ -189,53 +275,229 @@ void nfs_readdir_clear_array(struct page *page) * when called by nfs_readdir_add_to_array, the strings will be freed in * nfs_clear_readdir_array() */ -static -int nfs_readdir_make_qstr(struct qstr *string, const char *name, unsigned int len) +static const char *nfs_readdir_copy_name(const char *name, unsigned int len) { - string->len = len; - string->name = kmemdup(name, len, GFP_KERNEL); - if (string->name == NULL) - return -ENOMEM; + const char *ret = kmemdup_nul(name, len, GFP_KERNEL); + /* * Avoid a kmemleak false positive. The pointer to the name is stored * in a page cache page which kmemleak does not scan. */ - kmemleak_not_leak(string->name); - string->hash = full_name_hash(NULL, name, len); + if (ret != NULL) + kmemleak_not_leak(ret); + return ret; +} + +static size_t nfs_readdir_array_maxentries(void) +{ + return (PAGE_SIZE - sizeof(struct nfs_cache_array)) / + sizeof(struct nfs_cache_array_entry); +} + +/* + * Check that the next array entry lies entirely within the page bounds + */ +static int nfs_readdir_array_can_expand(struct nfs_cache_array *array) +{ + if (array->folio_full) + return -ENOSPC; + if (array->size == nfs_readdir_array_maxentries()) { + array->folio_full = 1; + return -ENOSPC; + } return 0; } -static -int nfs_readdir_add_to_array(struct nfs_entry *entry, struct page *page) +static int nfs_readdir_folio_array_append(struct folio *folio, + const struct nfs_entry *entry, + u64 *cookie) { - struct nfs_cache_array *array = kmap(page); + struct nfs_cache_array *array; struct nfs_cache_array_entry *cache_entry; - int ret; + const char *name; + int ret = -ENOMEM; - cache_entry = &array->array[array->size]; + name = nfs_readdir_copy_name(entry->name, entry->len); - /* Check that this entry lies within the page bounds */ - ret = -ENOSPC; - if ((char *)&cache_entry[1] - (char *)page_address(page) > PAGE_SIZE) + array = kmap_local_folio(folio, 0); + if (!name) + goto out; + ret = nfs_readdir_array_can_expand(array); + if (ret) { + kfree(name); goto out; + } - cache_entry->cookie = entry->prev_cookie; + array->size++; + cache_entry = &array->array[array->size - 1]; + cache_entry->cookie = array->last_cookie; cache_entry->ino = entry->ino; cache_entry->d_type = entry->d_type; - ret = nfs_readdir_make_qstr(&cache_entry->string, entry->name, entry->len); - if (ret) - goto out; + cache_entry->name_len = entry->len; + cache_entry->name = name; array->last_cookie = entry->cookie; - array->size++; + if (array->last_cookie <= cache_entry->cookie) + array->cookies_are_ordered = 0; if (entry->eof != 0) - array->eof_index = array->size; + nfs_readdir_array_set_eof(array); out: - kunmap(page); + *cookie = array->last_cookie; + kunmap_local(array); return ret; } +#define NFS_READDIR_COOKIE_MASK (U32_MAX >> 14) +/* + * Hash algorithm allowing content addressible access to sequences + * of directory cookies. Content is addressed by the value of the + * cookie index of the first readdir entry in a page. + * + * We select only the first 18 bits to avoid issues with excessive + * memory use for the page cache XArray. 18 bits should allow the caching + * of 262144 pages of sequences of readdir entries. Since each page holds + * 127 readdir entries for a typical 64-bit system, that works out to a + * cache of ~ 33 million entries per directory. + */ +static pgoff_t nfs_readdir_folio_cookie_hash(u64 cookie) +{ + if (cookie == 0) + return 0; + return hash_64(cookie, 18); +} + +static bool nfs_readdir_folio_validate(struct folio *folio, u64 last_cookie, + u64 change_attr) +{ + struct nfs_cache_array *array = kmap_local_folio(folio, 0); + int ret = true; + + if (array->change_attr != change_attr) + ret = false; + if (nfs_readdir_array_index_cookie(array) != last_cookie) + ret = false; + kunmap_local(array); + return ret; +} + +static void nfs_readdir_folio_unlock_and_put(struct folio *folio) +{ + folio_unlock(folio); + folio_put(folio); +} + +static void nfs_readdir_folio_init_and_validate(struct folio *folio, u64 cookie, + u64 change_attr) +{ + if (folio_test_uptodate(folio)) { + if (nfs_readdir_folio_validate(folio, cookie, change_attr)) + return; + nfs_readdir_clear_array(folio); + } + nfs_readdir_folio_init_array(folio, cookie, change_attr); + folio_mark_uptodate(folio); +} + +static struct folio *nfs_readdir_folio_get_locked(struct address_space *mapping, + u64 cookie, u64 change_attr) +{ + pgoff_t index = nfs_readdir_folio_cookie_hash(cookie); + struct folio *folio; + + folio = filemap_grab_folio(mapping, index); + if (IS_ERR(folio)) + return NULL; + nfs_readdir_folio_init_and_validate(folio, cookie, change_attr); + return folio; +} + +static u64 nfs_readdir_folio_last_cookie(struct folio *folio) +{ + struct nfs_cache_array *array; + u64 ret; + + array = kmap_local_folio(folio, 0); + ret = array->last_cookie; + kunmap_local(array); + return ret; +} + +static bool nfs_readdir_folio_needs_filling(struct folio *folio) +{ + struct nfs_cache_array *array; + bool ret; + + array = kmap_local_folio(folio, 0); + ret = !nfs_readdir_array_is_full(array); + kunmap_local(array); + return ret; +} + +static void nfs_readdir_folio_set_eof(struct folio *folio) +{ + struct nfs_cache_array *array; + + array = kmap_local_folio(folio, 0); + nfs_readdir_array_set_eof(array); + kunmap_local(array); +} + +static struct folio *nfs_readdir_folio_get_next(struct address_space *mapping, + u64 cookie, u64 change_attr) +{ + pgoff_t index = nfs_readdir_folio_cookie_hash(cookie); + struct folio *folio; + + folio = __filemap_get_folio(mapping, index, + FGP_LOCK|FGP_CREAT|FGP_NOFS|FGP_NOWAIT, + mapping_gfp_mask(mapping)); + if (IS_ERR(folio)) + return NULL; + nfs_readdir_folio_init_and_validate(folio, cookie, change_attr); + if (nfs_readdir_folio_last_cookie(folio) != cookie) + nfs_readdir_folio_reinit_array(folio, cookie, change_attr); + return folio; +} + +static inline +int is_32bit_api(void) +{ +#ifdef CONFIG_COMPAT + return in_compat_syscall(); +#else + return (BITS_PER_LONG == 32); +#endif +} + static -int nfs_readdir_search_for_pos(struct nfs_cache_array *array, nfs_readdir_descriptor_t *desc) +bool nfs_readdir_use_cookie(const struct file *filp) +{ + if ((filp->f_mode & FMODE_32BITHASH) || + (!(filp->f_mode & FMODE_64BITHASH) && is_32bit_api())) + return false; + return true; +} + +static void nfs_readdir_seek_next_array(struct nfs_cache_array *array, + struct nfs_readdir_descriptor *desc) +{ + if (array->folio_full) { + desc->last_cookie = array->last_cookie; + desc->current_index += array->size; + desc->cache_entry_index = 0; + desc->folio_index++; + } else + desc->last_cookie = nfs_readdir_array_index_cookie(array); +} + +static void nfs_readdir_rewind_search(struct nfs_readdir_descriptor *desc) +{ + desc->current_index = 0; + desc->last_cookie = 0; + desc->folio_index = 0; +} + +static int nfs_readdir_search_for_pos(struct nfs_cache_array *array, + struct nfs_readdir_descriptor *desc) { loff_t diff = desc->ctx->pos - desc->current_index; unsigned int index; @@ -243,119 +505,111 @@ int nfs_readdir_search_for_pos(struct nfs_cache_array *array, nfs_readdir_descri if (diff < 0) goto out_eof; if (diff >= array->size) { - if (array->eof_index >= 0) + if (array->folio_is_eof) goto out_eof; + nfs_readdir_seek_next_array(array, desc); return -EAGAIN; } index = (unsigned int)diff; - *desc->dir_cookie = array->array[index].cookie; + desc->dir_cookie = array->array[index].cookie; desc->cache_entry_index = index; return 0; out_eof: - desc->eof = 1; + desc->eof = true; return -EBADCOOKIE; } -static bool -nfs_readdir_inode_mapping_valid(struct nfs_inode *nfsi) +static bool nfs_readdir_array_cookie_in_range(struct nfs_cache_array *array, + u64 cookie) { - if (nfsi->cache_validity & (NFS_INO_INVALID_ATTR|NFS_INO_INVALID_DATA)) + if (!array->cookies_are_ordered) + return true; + /* Optimisation for monotonically increasing cookies */ + if (cookie >= array->last_cookie) return false; - smp_rmb(); - return !test_bit(NFS_INO_INVALIDATING, &nfsi->flags); + if (array->size && cookie < array->array[0].cookie) + return false; + return true; } -static -int nfs_readdir_search_for_cookie(struct nfs_cache_array *array, nfs_readdir_descriptor_t *desc) +static int nfs_readdir_search_for_cookie(struct nfs_cache_array *array, + struct nfs_readdir_descriptor *desc) { - int i; - loff_t new_pos; + unsigned int i; int status = -EAGAIN; + if (!nfs_readdir_array_cookie_in_range(array, desc->dir_cookie)) + goto check_eof; + for (i = 0; i < array->size; i++) { - if (array->array[i].cookie == *desc->dir_cookie) { - struct nfs_inode *nfsi = NFS_I(file_inode(desc->file)); - struct nfs_open_dir_context *ctx = desc->file->private_data; - - new_pos = desc->current_index + i; - if (ctx->attr_gencount != nfsi->attr_gencount || - !nfs_readdir_inode_mapping_valid(nfsi)) { - ctx->duped = 0; - ctx->attr_gencount = nfsi->attr_gencount; - } else if (new_pos < desc->ctx->pos) { - if (ctx->duped > 0 - && ctx->dup_cookie == *desc->dir_cookie) { - if (printk_ratelimit()) { - pr_notice("NFS: directory %pD2 contains a readdir loop." - "Please contact your server vendor. " - "The file: %.*s has duplicate cookie %llu\n", - desc->file, array->array[i].string.len, - array->array[i].string.name, *desc->dir_cookie); - } - status = -ELOOP; - goto out; - } - ctx->dup_cookie = *desc->dir_cookie; - ctx->duped = -1; - } - desc->ctx->pos = new_pos; + if (array->array[i].cookie == desc->dir_cookie) { + if (nfs_readdir_use_cookie(desc->file)) + desc->ctx->pos = desc->dir_cookie; + else + desc->ctx->pos = desc->current_index + i; desc->cache_entry_index = i; return 0; } } - if (array->eof_index >= 0) { +check_eof: + if (array->folio_is_eof) { status = -EBADCOOKIE; - if (*desc->dir_cookie == array->last_cookie) - desc->eof = 1; - } -out: + if (desc->dir_cookie == array->last_cookie) + desc->eof = true; + } else + nfs_readdir_seek_next_array(array, desc); return status; } -static -int nfs_readdir_search_array(nfs_readdir_descriptor_t *desc) +static int nfs_readdir_search_array(struct nfs_readdir_descriptor *desc) { struct nfs_cache_array *array; int status; - array = kmap(desc->page); + array = kmap_local_folio(desc->folio, 0); - if (*desc->dir_cookie == 0) + if (desc->dir_cookie == 0) status = nfs_readdir_search_for_pos(array, desc); else status = nfs_readdir_search_for_cookie(array, desc); - if (status == -EAGAIN) { - desc->last_cookie = array->last_cookie; - desc->current_index += array->size; - desc->page_index++; - } - kunmap(desc->page); + kunmap_local(array); return status; } /* Fill a page with xdr information before transferring to the cache page */ -static -int nfs_readdir_xdr_filler(struct page **pages, nfs_readdir_descriptor_t *desc, - struct nfs_entry *entry, struct file *file, struct inode *inode) +static int nfs_readdir_xdr_filler(struct nfs_readdir_descriptor *desc, + __be32 *verf, u64 cookie, + struct page **pages, size_t bufsize, + __be32 *verf_res) { - struct nfs_open_dir_context *ctx = file->private_data; - struct rpc_cred *cred = ctx->cred; + struct inode *inode = file_inode(desc->file); + struct nfs_readdir_arg arg = { + .dentry = file_dentry(desc->file), + .cred = desc->file->f_cred, + .verf = verf, + .cookie = cookie, + .pages = pages, + .page_len = bufsize, + .plus = desc->plus, + }; + struct nfs_readdir_res res = { + .verf = verf_res, + }; unsigned long timestamp, gencount; int error; again: timestamp = jiffies; gencount = nfs_inc_attr_generation_counter(); - error = NFS_PROTO(inode)->readdir(file_dentry(file), cred, entry->cookie, pages, - NFS_SERVER(inode)->dtsize, desc->plus); + desc->dir_verifier = nfs_save_change_attribute(inode); + error = NFS_PROTO(inode)->readdir(&arg, &res); if (error < 0) { /* We requested READDIRPLUS, but the server doesn't grok it */ if (error == -ENOTSUPP && desc->plus) { NFS_SERVER(inode)->caps &= ~NFS_CAP_READDIRPLUS; - clear_bit(NFS_INO_ADVISE_RDPLUS, &NFS_I(inode)->flags); - desc->plus = false; + desc->plus = arg.plus = false; goto again; } goto error; @@ -366,12 +620,13 @@ error: return error; } -static int xdr_decode(nfs_readdir_descriptor_t *desc, +static int xdr_decode(struct nfs_readdir_descriptor *desc, struct nfs_entry *entry, struct xdr_stream *xdr) { + struct inode *inode = file_inode(desc->file); int error; - error = desc->decode(xdr, entry, desc->plus); + error = NFS_PROTO(inode)->decode_dirent(xdr, entry, desc->plus); if (error) return error; entry->fattr->time_start = desc->timestamp; @@ -403,59 +658,78 @@ int nfs_same_file(struct dentry *dentry, struct nfs_entry *entry) return 1; } -static -bool nfs_use_readdirplus(struct inode *dir, struct dir_context *ctx) +#define NFS_READDIR_CACHE_USAGE_THRESHOLD (8UL) + +static bool nfs_use_readdirplus(struct inode *dir, struct dir_context *ctx, + unsigned int cache_hits, + unsigned int cache_misses) { if (!nfs_server_capable(dir, NFS_CAP_READDIRPLUS)) return false; - if (test_and_clear_bit(NFS_INO_ADVISE_RDPLUS, &NFS_I(dir)->flags)) + if (NFS_SERVER(dir)->flags & NFS_MOUNT_FORCE_RDIRPLUS) return true; - if (ctx->pos == 0) + if (ctx->pos == 0 || + cache_hits + cache_misses > NFS_READDIR_CACHE_USAGE_THRESHOLD) return true; return false; } /* - * This function is called by the lookup and getattr code to request the + * This function is called by the getattr code to request the * use of readdirplus to accelerate any future lookups in the same * directory. */ -void nfs_advise_use_readdirplus(struct inode *dir) +void nfs_readdir_record_entry_cache_hit(struct inode *dir) { struct nfs_inode *nfsi = NFS_I(dir); + struct nfs_open_dir_context *ctx; if (nfs_server_capable(dir, NFS_CAP_READDIRPLUS) && - !list_empty(&nfsi->open_files)) - set_bit(NFS_INO_ADVISE_RDPLUS, &nfsi->flags); + S_ISDIR(dir->i_mode)) { + rcu_read_lock(); + list_for_each_entry_rcu (ctx, &nfsi->open_files, list) + atomic_inc(&ctx->cache_hits); + rcu_read_unlock(); + } } /* * This function is mainly for use by nfs_getattr(). * * If this is an 'ls -l', we want to force use of readdirplus. - * Do this by checking if there is an active file descriptor - * and calling nfs_advise_use_readdirplus, then forcing a - * cache flush. */ -void nfs_force_use_readdirplus(struct inode *dir) +void nfs_readdir_record_entry_cache_miss(struct inode *dir) { struct nfs_inode *nfsi = NFS_I(dir); + struct nfs_open_dir_context *ctx; if (nfs_server_capable(dir, NFS_CAP_READDIRPLUS) && - !list_empty(&nfsi->open_files)) { - set_bit(NFS_INO_ADVISE_RDPLUS, &nfsi->flags); - invalidate_mapping_pages(dir->i_mapping, 0, -1); + S_ISDIR(dir->i_mode)) { + rcu_read_lock(); + list_for_each_entry_rcu (ctx, &nfsi->open_files, list) + atomic_inc(&ctx->cache_misses); + rcu_read_unlock(); } } +static void nfs_lookup_advise_force_readdirplus(struct inode *dir, + unsigned int flags) +{ + if (nfs_server_capable(dir, NFS_CAP_CASE_INSENSITIVE)) + return; + if (flags & (LOOKUP_EXCL | LOOKUP_PARENT | LOOKUP_REVAL)) + return; + nfs_readdir_record_entry_cache_miss(dir); +} + static -void nfs_prime_dcache(struct dentry *parent, struct nfs_entry *entry) +void nfs_prime_dcache(struct dentry *parent, struct nfs_entry *entry, + unsigned long dir_verifier) { struct qstr filename = QSTR_INIT(entry->name, entry->len); DECLARE_WAIT_QUEUE_HEAD_ONSTACK(wq); struct dentry *dentry; struct dentry *alias; - struct inode *dir = d_inode(parent); struct inode *inode; int status; @@ -494,12 +768,16 @@ again: if (nfs_same_file(dentry, entry)) { if (!entry->fh->size) goto out; - nfs_set_verifier(dentry, nfs_save_change_attribute(dir)); + nfs_set_verifier(dentry, dir_verifier); status = nfs_refresh_inode(d_inode(dentry), entry->fattr); if (!status) - nfs_setsecurity(d_inode(dentry), entry->fattr, entry->label); + nfs_setsecurity(d_inode(dentry), entry->fattr); + trace_nfs_readdir_lookup_revalidate(d_inode(parent), + dentry, 0, status); goto out; } else { + trace_nfs_readdir_lookup_revalidate_failed( + d_inode(parent), dentry, 0); d_invalidate(dentry); dput(dentry); dentry = NULL; @@ -511,7 +789,7 @@ again: goto out; } - inode = nfs_fhget(dentry->d_sb, entry->fh, entry->fattr, entry->label); + inode = nfs_fhget(dentry->d_sb, entry->fh, entry->fattr); alias = d_splice_alias(inode, dentry); d_lookup_done(dentry); if (alias) { @@ -520,266 +798,337 @@ again: dput(dentry); dentry = alias; } - nfs_set_verifier(dentry, nfs_save_change_attribute(dir)); + nfs_set_verifier(dentry, dir_verifier); + trace_nfs_readdir_lookup(d_inode(parent), dentry, 0); out: dput(dentry); } -/* Perform conversion from xdr to cache array */ -static -int nfs_readdir_page_filler(nfs_readdir_descriptor_t *desc, struct nfs_entry *entry, - struct page **xdr_pages, struct page *page, unsigned int buflen) +static int nfs_readdir_entry_decode(struct nfs_readdir_descriptor *desc, + struct nfs_entry *entry, + struct xdr_stream *stream) { + int ret; + + if (entry->fattr->label) + entry->fattr->label->len = NFS4_MAXLABELLEN; + ret = xdr_decode(desc, entry, stream); + if (ret || !desc->plus) + return ret; + nfs_prime_dcache(file_dentry(desc->file), entry, desc->dir_verifier); + return 0; +} + +/* Perform conversion from xdr to cache array */ +static int nfs_readdir_folio_filler(struct nfs_readdir_descriptor *desc, + struct nfs_entry *entry, + struct page **xdr_pages, unsigned int buflen, + struct folio **arrays, size_t narrays, + u64 change_attr) +{ + struct address_space *mapping = desc->file->f_mapping; + struct folio *new, *folio = *arrays; struct xdr_stream stream; + struct folio *scratch; struct xdr_buf buf; - struct page *scratch; - struct nfs_cache_array *array; - unsigned int count = 0; + u64 cookie; int status; - scratch = alloc_page(GFP_KERNEL); + scratch = folio_alloc(GFP_KERNEL, 0); if (scratch == NULL) return -ENOMEM; - if (buflen == 0) - goto out_nopages; - xdr_init_decode_pages(&stream, &buf, xdr_pages, buflen); - xdr_set_scratch_buffer(&stream, page_address(scratch), PAGE_SIZE); + xdr_set_scratch_folio(&stream, scratch); do { - status = xdr_decode(desc, entry, &stream); - if (status != 0) { - if (status == -EAGAIN) - status = 0; + status = nfs_readdir_entry_decode(desc, entry, &stream); + if (status != 0) break; - } - count++; + status = nfs_readdir_folio_array_append(folio, entry, &cookie); + if (status != -ENOSPC) + continue; - if (desc->plus) - nfs_prime_dcache(file_dentry(desc->file), entry); + if (folio->mapping != mapping) { + if (!--narrays) + break; + new = nfs_readdir_folio_array_alloc(cookie, GFP_KERNEL); + if (!new) + break; + arrays++; + *arrays = folio = new; + } else { + new = nfs_readdir_folio_get_next(mapping, cookie, + change_attr); + if (!new) + break; + if (folio != *arrays) + nfs_readdir_folio_unlock_and_put(folio); + folio = new; + } + desc->folio_index_max++; + status = nfs_readdir_folio_array_append(folio, entry, &cookie); + } while (!status && !entry->eof); - status = nfs_readdir_add_to_array(entry, page); - if (status != 0) + switch (status) { + case -EBADCOOKIE: + if (!entry->eof) break; - } while (!entry->eof); - -out_nopages: - if (count == 0 || (status == -EBADCOOKIE && entry->eof != 0)) { - array = kmap(page); - array->eof_index = array->size; + nfs_readdir_folio_set_eof(folio); + fallthrough; + case -EAGAIN: + status = 0; + break; + case -ENOSPC: status = 0; - kunmap(page); + if (!desc->plus) + break; + while (!nfs_readdir_entry_decode(desc, entry, &stream)) + ; } - put_page(scratch); + if (folio != *arrays) + nfs_readdir_folio_unlock_and_put(folio); + + folio_put(scratch); return status; } -static -void nfs_readdir_free_pages(struct page **pages, unsigned int npages) +static void nfs_readdir_free_pages(struct page **pages, size_t npages) { - unsigned int i; - for (i = 0; i < npages; i++) - put_page(pages[i]); + while (npages--) + put_page(pages[npages]); + kfree(pages); } /* - * nfs_readdir_large_page will allocate pages that must be freed with a call - * to nfs_readdir_free_pagearray + * nfs_readdir_alloc_pages() will allocate pages that must be freed with a call + * to nfs_readdir_free_pages() */ -static -int nfs_readdir_alloc_pages(struct page **pages, unsigned int npages) +static struct page **nfs_readdir_alloc_pages(size_t npages) { - unsigned int i; + struct page **pages; + size_t i; + pages = kmalloc_array(npages, sizeof(*pages), GFP_KERNEL); + if (!pages) + return NULL; for (i = 0; i < npages; i++) { struct page *page = alloc_page(GFP_KERNEL); if (page == NULL) goto out_freepages; pages[i] = page; } - return 0; + return pages; out_freepages: nfs_readdir_free_pages(pages, i); - return -ENOMEM; + return NULL; } -static -int nfs_readdir_xdr_to_array(nfs_readdir_descriptor_t *desc, struct page *page, struct inode *inode) +static int nfs_readdir_xdr_to_array(struct nfs_readdir_descriptor *desc, + __be32 *verf_arg, __be32 *verf_res, + struct folio **arrays, size_t narrays) { - struct page *pages[NFS_MAX_READDIR_PAGES]; - struct nfs_entry entry; - struct file *file = desc->file; - struct nfs_cache_array *array; + u64 change_attr; + struct page **pages; + struct folio *folio = *arrays; + struct nfs_entry *entry; + size_t array_size; + struct inode *inode = file_inode(desc->file); + unsigned int dtsize = desc->dtsize; + unsigned int pglen; int status = -ENOMEM; - unsigned int array_size = ARRAY_SIZE(pages); - - entry.prev_cookie = 0; - entry.cookie = desc->last_cookie; - entry.eof = 0; - entry.fh = nfs_alloc_fhandle(); - entry.fattr = nfs_alloc_fattr(); - entry.server = NFS_SERVER(inode); - if (entry.fh == NULL || entry.fattr == NULL) - goto out; - entry.label = nfs4_label_alloc(NFS_SERVER(inode), GFP_NOWAIT); - if (IS_ERR(entry.label)) { - status = PTR_ERR(entry.label); + entry = kzalloc(sizeof(*entry), GFP_KERNEL); + if (!entry) + return -ENOMEM; + entry->cookie = nfs_readdir_folio_last_cookie(folio); + entry->fh = nfs_alloc_fhandle(); + entry->fattr = nfs_alloc_fattr_with_label(NFS_SERVER(inode)); + entry->server = NFS_SERVER(inode); + if (entry->fh == NULL || entry->fattr == NULL) goto out; - } - array = kmap(page); - memset(array, 0, sizeof(struct nfs_cache_array)); - array->eof_index = -1; + array_size = (dtsize + PAGE_SIZE - 1) >> PAGE_SHIFT; + pages = nfs_readdir_alloc_pages(array_size); + if (!pages) + goto out; - status = nfs_readdir_alloc_pages(pages, array_size); + change_attr = inode_peek_iversion_raw(inode); + status = nfs_readdir_xdr_filler(desc, verf_arg, entry->cookie, pages, + dtsize, verf_res); if (status < 0) - goto out_release_array; - do { - unsigned int pglen; - status = nfs_readdir_xdr_filler(pages, desc, &entry, file, inode); + goto free_pages; - if (status < 0) - break; - pglen = status; - status = nfs_readdir_page_filler(desc, &entry, pages, page, pglen); - if (status < 0) { - if (status == -ENOSPC) - status = 0; - break; - } - } while (array->eof_index < 0); + pglen = status; + if (pglen != 0) + status = nfs_readdir_folio_filler(desc, entry, pages, pglen, + arrays, narrays, change_attr); + else + nfs_readdir_folio_set_eof(folio); + desc->buffer_fills++; +free_pages: nfs_readdir_free_pages(pages, array_size); -out_release_array: - kunmap(page); - nfs4_label_free(entry.label); out: - nfs_free_fattr(entry.fattr); - nfs_free_fhandle(entry.fh); + nfs_free_fattr(entry->fattr); + nfs_free_fhandle(entry->fh); + kfree(entry); return status; } -/* - * Now we cache directories properly, by converting xdr information - * to an array that can be used for lookups later. This results in - * fewer cache pages, since we can store more information on each page. - * We only need to convert from xdr once so future lookups are much simpler - */ -static -int nfs_readdir_filler(nfs_readdir_descriptor_t *desc, struct page* page) +static void nfs_readdir_folio_put(struct nfs_readdir_descriptor *desc) { - struct inode *inode = file_inode(desc->file); - int ret; - - ret = nfs_readdir_xdr_to_array(desc, page, inode); - if (ret < 0) - goto error; - SetPageUptodate(page); - - if (invalidate_inode_pages2_range(inode->i_mapping, page->index + 1, -1) < 0) { - /* Should never happen */ - nfs_zap_mapping(inode, inode->i_mapping); - } - unlock_page(page); - return 0; - error: - unlock_page(page); - return ret; + folio_put(desc->folio); + desc->folio = NULL; } -static -void cache_page_release(nfs_readdir_descriptor_t *desc) +static void +nfs_readdir_folio_unlock_and_put_cached(struct nfs_readdir_descriptor *desc) { - if (!desc->page->mapping) - nfs_readdir_clear_array(desc->page); - put_page(desc->page); - desc->page = NULL; + folio_unlock(desc->folio); + nfs_readdir_folio_put(desc); } -static -struct page *get_cache_page(nfs_readdir_descriptor_t *desc) +static struct folio * +nfs_readdir_folio_get_cached(struct nfs_readdir_descriptor *desc) { - return read_cache_page(desc->file->f_mapping, - desc->page_index, (filler_t *)nfs_readdir_filler, desc); + struct address_space *mapping = desc->file->f_mapping; + u64 change_attr = inode_peek_iversion_raw(mapping->host); + u64 cookie = desc->last_cookie; + struct folio *folio; + + folio = nfs_readdir_folio_get_locked(mapping, cookie, change_attr); + if (!folio) + return NULL; + if (desc->clear_cache && !nfs_readdir_folio_needs_filling(folio)) + nfs_readdir_folio_reinit_array(folio, cookie, change_attr); + return folio; } /* * Returns 0 if desc->dir_cookie was found on page desc->page_index + * and locks the page to prevent removal from the page cache. */ -static -int find_cache_page(nfs_readdir_descriptor_t *desc) +static int find_and_lock_cache_page(struct nfs_readdir_descriptor *desc) { + struct inode *inode = file_inode(desc->file); + struct nfs_inode *nfsi = NFS_I(inode); + __be32 verf[NFS_DIR_VERIFIER_SIZE]; int res; - desc->page = get_cache_page(desc); - if (IS_ERR(desc->page)) - return PTR_ERR(desc->page); - + desc->folio = nfs_readdir_folio_get_cached(desc); + if (!desc->folio) + return -ENOMEM; + if (nfs_readdir_folio_needs_filling(desc->folio)) { + /* Grow the dtsize if we had to go back for more pages */ + if (desc->folio_index == desc->folio_index_max) + nfs_grow_dtsize(desc); + desc->folio_index_max = desc->folio_index; + trace_nfs_readdir_cache_fill(desc->file, nfsi->cookieverf, + desc->last_cookie, + desc->folio->index, desc->dtsize); + res = nfs_readdir_xdr_to_array(desc, nfsi->cookieverf, verf, + &desc->folio, 1); + if (res < 0) { + nfs_readdir_folio_unlock_and_put_cached(desc); + trace_nfs_readdir_cache_fill_done(inode, res); + if (res == -EBADCOOKIE || res == -ENOTSYNC) { + invalidate_inode_pages2(desc->file->f_mapping); + nfs_readdir_rewind_search(desc); + trace_nfs_readdir_invalidate_cache_range( + inode, 0, MAX_LFS_FILESIZE); + return -EAGAIN; + } + return res; + } + /* + * Set the cookie verifier if the page cache was empty + */ + if (desc->last_cookie == 0 && + memcmp(nfsi->cookieverf, verf, sizeof(nfsi->cookieverf))) { + memcpy(nfsi->cookieverf, verf, + sizeof(nfsi->cookieverf)); + invalidate_inode_pages2_range(desc->file->f_mapping, 1, + -1); + trace_nfs_readdir_invalidate_cache_range( + inode, 1, MAX_LFS_FILESIZE); + } + desc->clear_cache = false; + } res = nfs_readdir_search_array(desc); - if (res != 0) - cache_page_release(desc); + if (res == 0) + return 0; + nfs_readdir_folio_unlock_and_put_cached(desc); return res; } /* Search for desc->dir_cookie from the beginning of the page cache */ -static inline -int readdir_search_pagecache(nfs_readdir_descriptor_t *desc) +static int readdir_search_pagecache(struct nfs_readdir_descriptor *desc) { int res; - if (desc->page_index == 0) { - desc->current_index = 0; - desc->last_cookie = 0; - } do { - res = find_cache_page(desc); + res = find_and_lock_cache_page(desc); } while (res == -EAGAIN); return res; } +#define NFS_READDIR_CACHE_MISS_THRESHOLD (16UL) + /* * Once we've found the start of the dirent within a page: fill 'er up... */ -static -int nfs_do_filldir(nfs_readdir_descriptor_t *desc) +static void nfs_do_filldir(struct nfs_readdir_descriptor *desc, + const __be32 *verf) { struct file *file = desc->file; - int i = 0; - int res = 0; - struct nfs_cache_array *array = NULL; - struct nfs_open_dir_context *ctx = file->private_data; + struct nfs_cache_array *array; + unsigned int i; + bool first_emit = !desc->dir_cookie; - array = kmap(desc->page); + array = kmap_local_folio(desc->folio, 0); for (i = desc->cache_entry_index; i < array->size; i++) { struct nfs_cache_array_entry *ent; + /* + * nfs_readdir_handle_cache_misses return force clear at + * (cache_misses > NFS_READDIR_CACHE_MISS_THRESHOLD) for + * readdir heuristic, NFS_READDIR_CACHE_MISS_THRESHOLD + 1 + * entries need be emitted here. + */ + if (first_emit && i > NFS_READDIR_CACHE_MISS_THRESHOLD + 2) { + desc->eob = true; + break; + } + ent = &array->array[i]; - if (!dir_emit(desc->ctx, ent->string.name, ent->string.len, + if (!dir_emit(desc->ctx, ent->name, ent->name_len, nfs_compat_user_ino64(ent->ino), ent->d_type)) { - desc->eof = 1; + desc->eob = true; break; } - desc->ctx->pos++; - if (i < (array->size-1)) - *desc->dir_cookie = array->array[i+1].cookie; + memcpy(desc->verf, verf, sizeof(desc->verf)); + if (i == array->size - 1) { + desc->dir_cookie = array->last_cookie; + nfs_readdir_seek_next_array(array, desc); + } else { + desc->dir_cookie = array->array[i + 1].cookie; + desc->last_cookie = array->array[0].cookie; + } + if (nfs_readdir_use_cookie(file)) + desc->ctx->pos = desc->dir_cookie; else - *desc->dir_cookie = array->last_cookie; - if (ctx->duped != 0) - ctx->duped = 1; + desc->ctx->pos++; } - if (array->eof_index >= 0) - desc->eof = 1; + if (array->folio_is_eof) + desc->eof = !desc->eob; - kunmap(desc->page); - cache_page_release(desc); - dfprintk(DIRCACHE, "NFS: nfs_do_filldir() filling ended @ cookie %Lu; returning = %d\n", - (unsigned long long)*desc->dir_cookie, res); - return res; + kunmap_local(array); + dfprintk(DIRCACHE, "NFS: nfs_do_filldir() filling ended @ cookie %llu\n", + (unsigned long long)desc->dir_cookie); } /* @@ -794,41 +1143,77 @@ int nfs_do_filldir(nfs_readdir_descriptor_t *desc) * we should already have a complete representation of the * directory in the page cache by the time we get here. */ -static inline -int uncached_readdir(nfs_readdir_descriptor_t *desc) +static int uncached_readdir(struct nfs_readdir_descriptor *desc) { - struct page *page = NULL; - int status; - struct inode *inode = file_inode(desc->file); - struct nfs_open_dir_context *ctx = desc->file->private_data; + struct folio **arrays; + size_t i, sz = 512; + __be32 verf[NFS_DIR_VERIFIER_SIZE]; + int status = -ENOMEM; - dfprintk(DIRCACHE, "NFS: uncached_readdir() searching for cookie %Lu\n", - (unsigned long long)*desc->dir_cookie); + dfprintk(DIRCACHE, "NFS: uncached_readdir() searching for cookie %llu\n", + (unsigned long long)desc->dir_cookie); - page = alloc_page(GFP_HIGHUSER); - if (!page) { - status = -ENOMEM; + arrays = kcalloc(sz, sizeof(*arrays), GFP_KERNEL); + if (!arrays) + goto out; + arrays[0] = nfs_readdir_folio_array_alloc(desc->dir_cookie, GFP_KERNEL); + if (!arrays[0]) goto out; - } - desc->page_index = 0; - desc->last_cookie = *desc->dir_cookie; - desc->page = page; - ctx->duped = 0; + desc->folio_index = 0; + desc->cache_entry_index = 0; + desc->last_cookie = desc->dir_cookie; + desc->folio_index_max = 0; - status = nfs_readdir_xdr_to_array(desc, page, inode); - if (status < 0) - goto out_release; + trace_nfs_readdir_uncached(desc->file, desc->verf, desc->last_cookie, + -1, desc->dtsize); - status = nfs_do_filldir(desc); + status = nfs_readdir_xdr_to_array(desc, desc->verf, verf, arrays, sz); + if (status < 0) { + trace_nfs_readdir_uncached_done(file_inode(desc->file), status); + goto out_free; + } + + for (i = 0; !desc->eob && i < sz && arrays[i]; i++) { + desc->folio = arrays[i]; + nfs_do_filldir(desc, verf); + } + desc->folio = NULL; - out: - dfprintk(DIRCACHE, "NFS: %s: returns %d\n", - __func__, status); + /* + * Grow the dtsize if we have to go back for more pages, + * or shrink it if we're reading too many. + */ + if (!desc->eof) { + if (!desc->eob) + nfs_grow_dtsize(desc); + else if (desc->buffer_fills == 1 && + i < (desc->folio_index_max >> 1)) + nfs_shrink_dtsize(desc); + } +out_free: + for (i = 0; i < sz && arrays[i]; i++) + nfs_readdir_folio_array_free(arrays[i]); +out: + if (!nfs_readdir_use_cookie(desc->file)) + nfs_readdir_rewind_search(desc); + desc->folio_index_max = -1; + kfree(arrays); + dfprintk(DIRCACHE, "NFS: %s: returns %d\n", __func__, status); return status; - out_release: - cache_page_release(desc); - goto out; +} + +static bool nfs_readdir_handle_cache_misses(struct inode *inode, + struct nfs_readdir_descriptor *desc, + unsigned int cache_misses, + bool force_clear) +{ + if (desc->ctx->pos == 0 || !desc->plus) + return false; + if (cache_misses <= NFS_READDIR_CACHE_MISS_THRESHOLD && !force_clear) + return false; + trace_nfs_readdir_force_readdirplus(inode); + return true; } /* The file offset position represents the dirent entry number. A @@ -839,10 +1224,12 @@ static int nfs_readdir(struct file *file, struct dir_context *ctx) { struct dentry *dentry = file_dentry(file); struct inode *inode = d_inode(dentry); - nfs_readdir_descriptor_t my_desc, - *desc = &my_desc; + struct nfs_inode *nfsi = NFS_I(inode); struct nfs_open_dir_context *dir_ctx = file->private_data; - int res = 0; + struct nfs_readdir_descriptor *desc; + unsigned int cache_hits, cache_misses; + bool force_clear; + int res; dfprintk(FILE, "NFS: readdir(%pD2) starting at cookie %llu\n", file, (long long)ctx->pos); @@ -854,18 +1241,38 @@ static int nfs_readdir(struct file *file, struct dir_context *ctx) * to either find the entry with the appropriate number or * revalidate the cookie. */ - memset(desc, 0, sizeof(*desc)); + nfs_revalidate_mapping(inode, file->f_mapping); + res = -ENOMEM; + desc = kzalloc(sizeof(*desc), GFP_KERNEL); + if (!desc) + goto out; desc->file = file; desc->ctx = ctx; - desc->dir_cookie = &dir_ctx->dir_cookie; - desc->decode = NFS_PROTO(inode)->decode_dirent; - desc->plus = nfs_use_readdirplus(inode, ctx); + desc->folio_index_max = -1; + + spin_lock(&file->f_lock); + desc->dir_cookie = dir_ctx->dir_cookie; + desc->folio_index = dir_ctx->page_index; + desc->last_cookie = dir_ctx->last_cookie; + desc->attr_gencount = dir_ctx->attr_gencount; + desc->eof = dir_ctx->eof; + nfs_set_dtsize(desc, dir_ctx->dtsize); + memcpy(desc->verf, dir_ctx->verf, sizeof(desc->verf)); + cache_hits = atomic_xchg(&dir_ctx->cache_hits, 0); + cache_misses = atomic_xchg(&dir_ctx->cache_misses, 0); + force_clear = dir_ctx->force_clear; + spin_unlock(&file->f_lock); + + if (desc->eof) { + res = 0; + goto out_free; + } - if (ctx->pos == 0 || nfs_attribute_cache_expired(inode)) - res = nfs_revalidate_mapping(inode, file->f_mapping); - if (res < 0) - goto out; + desc->plus = nfs_use_readdirplus(inode, ctx, cache_hits, cache_misses); + force_clear = nfs_readdir_handle_cache_misses(inode, desc, cache_misses, + force_clear); + desc->clear_cache = force_clear; do { res = readdir_search_pagecache(desc); @@ -873,18 +1280,18 @@ static int nfs_readdir(struct file *file, struct dir_context *ctx) if (res == -EBADCOOKIE) { res = 0; /* This means either end of directory */ - if (*desc->dir_cookie && desc->eof == 0) { + if (desc->dir_cookie && !desc->eof) { /* Or that the server has 'lost' a cookie */ res = uncached_readdir(desc); if (res == 0) continue; + if (res == -EBADCOOKIE || res == -ENOTSYNC) + res = 0; } break; } if (res == -ETOOSMALL && desc->plus) { - clear_bit(NFS_INO_ADVISE_RDPLUS, &NFS_I(inode)->flags); nfs_zap_caches(inode); - desc->page_index = 0; desc->plus = false; desc->eof = false; continue; @@ -892,43 +1299,68 @@ static int nfs_readdir(struct file *file, struct dir_context *ctx) if (res < 0) break; - res = nfs_do_filldir(desc); - if (res < 0) - break; - } while (!desc->eof); + nfs_do_filldir(desc, nfsi->cookieverf); + nfs_readdir_folio_unlock_and_put_cached(desc); + if (desc->folio_index == desc->folio_index_max) + desc->clear_cache = force_clear; + } while (!desc->eob && !desc->eof); + + spin_lock(&file->f_lock); + dir_ctx->dir_cookie = desc->dir_cookie; + dir_ctx->last_cookie = desc->last_cookie; + dir_ctx->attr_gencount = desc->attr_gencount; + dir_ctx->page_index = desc->folio_index; + dir_ctx->force_clear = force_clear; + dir_ctx->eof = desc->eof; + dir_ctx->dtsize = desc->dtsize; + memcpy(dir_ctx->verf, desc->verf, sizeof(dir_ctx->verf)); + spin_unlock(&file->f_lock); +out_free: + kfree(desc); + out: - if (res > 0) - res = 0; dfprintk(FILE, "NFS: readdir(%pD2) returns %d\n", file, res); return res; } static loff_t nfs_llseek_dir(struct file *filp, loff_t offset, int whence) { - struct inode *inode = file_inode(filp); struct nfs_open_dir_context *dir_ctx = filp->private_data; dfprintk(FILE, "NFS: llseek dir(%pD2, %lld, %d)\n", filp, offset, whence); - inode_lock(inode); switch (whence) { - case 1: - offset += filp->f_pos; - case 0: - if (offset >= 0) - break; - default: - offset = -EINVAL; - goto out; + default: + return -EINVAL; + case SEEK_SET: + if (offset < 0) + return -EINVAL; + spin_lock(&filp->f_lock); + break; + case SEEK_CUR: + if (offset == 0) + return filp->f_pos; + spin_lock(&filp->f_lock); + offset += filp->f_pos; + if (offset < 0) { + spin_unlock(&filp->f_lock); + return -EINVAL; + } } if (offset != filp->f_pos) { filp->f_pos = offset; - dir_ctx->dir_cookie = 0; - dir_ctx->duped = 0; + dir_ctx->page_index = 0; + if (!nfs_readdir_use_cookie(filp)) { + dir_ctx->dir_cookie = 0; + dir_ctx->last_cookie = 0; + } else { + dir_ctx->dir_cookie = offset; + dir_ctx->last_cookie = offset; + } + dir_ctx->eof = false; } -out: - inode_unlock(inode); + spin_unlock(&filp->f_lock); return offset; } @@ -939,32 +1371,134 @@ out: static int nfs_fsync_dir(struct file *filp, loff_t start, loff_t end, int datasync) { - struct inode *inode = file_inode(filp); - dfprintk(FILE, "NFS: fsync dir(%pD2) datasync %d\n", filp, datasync); - inode_lock(inode); - nfs_inc_stats(inode, NFSIOS_VFSFSYNC); - inode_unlock(inode); + nfs_inc_stats(file_inode(filp), NFSIOS_VFSFSYNC); return 0; } /** * nfs_force_lookup_revalidate - Mark the directory as having changed - * @dir - pointer to directory inode + * @dir: pointer to directory inode * * This forces the revalidation code in nfs_lookup_revalidate() to do a * full lookup on all child dentries of 'dir' whenever a change occurs * on the server that might have invalidated our dcache. * + * Note that we reserve bit '0' as a tag to let us know when a dentry + * was revalidated while holding a delegation on its inode. + * * The caller should be holding dir->i_lock */ void nfs_force_lookup_revalidate(struct inode *dir) { - NFS_I(dir)->cache_change_attribute++; + NFS_I(dir)->cache_change_attribute += 2; } EXPORT_SYMBOL_GPL(nfs_force_lookup_revalidate); +/** + * nfs_verify_change_attribute - Detects NFS remote directory changes + * @dir: pointer to parent directory inode + * @verf: previously saved change attribute + * + * Return "false" if the verifiers doesn't match the change attribute. + * This would usually indicate that the directory contents have changed on + * the server, and that any dentries need revalidating. + */ +static bool nfs_verify_change_attribute(struct inode *dir, unsigned long verf) +{ + return (verf & ~1UL) == nfs_save_change_attribute(dir); +} + +static void nfs_set_verifier_delegated(unsigned long *verf) +{ + *verf |= 1UL; +} + +#if IS_ENABLED(CONFIG_NFS_V4) +static void nfs_unset_verifier_delegated(unsigned long *verf) +{ + *verf &= ~1UL; +} +#endif /* IS_ENABLED(CONFIG_NFS_V4) */ + +static bool nfs_test_verifier_delegated(unsigned long verf) +{ + return verf & 1; +} + +static bool nfs_verifier_is_delegated(struct dentry *dentry) +{ + return nfs_test_verifier_delegated(dentry->d_time); +} + +static void nfs_set_verifier_locked(struct dentry *dentry, unsigned long verf) +{ + struct inode *inode = d_inode(dentry); + struct inode *dir = d_inode_rcu(dentry->d_parent); + + if (!dir || !nfs_verify_change_attribute(dir, verf)) + return; + if (inode && NFS_PROTO(inode)->have_delegation(inode, FMODE_READ, 0)) + nfs_set_verifier_delegated(&verf); + dentry->d_time = verf; +} + +/** + * nfs_set_verifier - save a parent directory verifier in the dentry + * @dentry: pointer to dentry + * @verf: verifier to save + * + * Saves the parent directory verifier in @dentry. If the inode has + * a delegation, we also tag the dentry as having been revalidated + * while holding a delegation so that we know we don't have to + * look it up again after a directory change. + */ +void nfs_set_verifier(struct dentry *dentry, unsigned long verf) +{ + + spin_lock(&dentry->d_lock); + nfs_set_verifier_locked(dentry, verf); + spin_unlock(&dentry->d_lock); +} +EXPORT_SYMBOL_GPL(nfs_set_verifier); + +#if IS_ENABLED(CONFIG_NFS_V4) +/** + * nfs_clear_verifier_delegated - clear the dir verifier delegation tag + * @inode: pointer to inode + * + * Iterates through the dentries in the inode alias list and clears + * the tag used to indicate that the dentry has been revalidated + * while holding a delegation. + * This function is intended for use when the delegation is being + * returned or revoked. + */ +void nfs_clear_verifier_delegated(struct inode *inode) +{ + struct dentry *alias; + + if (!inode) + return; + spin_lock(&inode->i_lock); + hlist_for_each_entry(alias, &inode->i_dentry, d_u.d_alias) { + spin_lock(&alias->d_lock); + nfs_unset_verifier_delegated(&alias->d_time); + spin_unlock(&alias->d_lock); + } + spin_unlock(&inode->i_lock); +} +EXPORT_SYMBOL_GPL(nfs_clear_verifier_delegated); +#endif /* IS_ENABLED(CONFIG_NFS_V4) */ + +static int nfs_dentry_verify_change(struct inode *dir, struct dentry *dentry) +{ + if (nfs_server_capable(dir, NFS_CAP_CASE_INSENSITIVE) && + d_really_is_negative(dentry)) + return dentry->d_time == inode_peek_iversion_raw(dir); + return nfs_verify_change_attribute(dir, dentry->d_time); +} + /* * A check for whether or not the parent directory has changed. * In the case it has, we assume that the dentries are untrustworthy @@ -978,7 +1512,7 @@ static int nfs_check_verifier(struct inode *dir, struct dentry *dentry, return 1; if (NFS_SERVER(dir)->flags & NFS_MOUNT_LOOKUP_CACHE_NONE) return 0; - if (!nfs_verify_change_attribute(dir, dentry->d_time)) + if (!nfs_dentry_verify_change(dir, dentry)) return 0; /* Revalidate nfsi->cache_change_attribute before we declare a match */ if (nfs_mapping_need_revalidate_inode(dir)) { @@ -987,7 +1521,7 @@ static int nfs_check_verifier(struct inode *dir, struct dentry *dentry, if (__nfs_revalidate_inode(NFS_SERVER(dir), dir) < 0) return 0; } - if (!nfs_verify_change_attribute(dir, dentry->d_time)) + if (!nfs_dentry_verify_change(dir, dentry)) return 0; return 1; } @@ -1000,7 +1534,8 @@ static int nfs_is_exclusive_create(struct inode *dir, unsigned int flags) { if (NFS_PROTO(dir)->version == 2) return 0; - return flags & LOOKUP_EXCL; + return (flags & (LOOKUP_CREATE | LOOKUP_EXCL)) == + (LOOKUP_CREATE | LOOKUP_EXCL); } /* @@ -1019,15 +1554,32 @@ int nfs_lookup_verify_inode(struct inode *inode, unsigned int flags) if (IS_AUTOMOUNT(inode)) return 0; + + if (flags & LOOKUP_OPEN) { + switch (inode->i_mode & S_IFMT) { + case S_IFREG: + /* A NFSv4 OPEN will revalidate later */ + if (server->caps & NFS_CAP_ATOMIC_OPEN) + goto out; + fallthrough; + case S_IFDIR: + if (server->flags & NFS_MOUNT_NOCTO) + break; + /* NFS close-to-open cache consistency validation */ + goto out_force; + } + } + /* VFS wants an on-the-wire revalidation */ if (flags & LOOKUP_REVAL) goto out_force; - /* This is an open(2) */ - if ((flags & LOOKUP_OPEN) && !(server->flags & NFS_MOUNT_NOCTO) && - (S_ISREG(inode->i_mode) || S_ISDIR(inode->i_mode))) - goto out_force; out: - return (inode->i_nlink == 0) ? -ENOENT : 0; + if (inode->i_nlink > 0 || + (inode->i_nlink == 0 && + test_bit(NFS_INO_PRESERVE_UNLINKED, &NFS_I(inode)->flags))) + return 0; + else + return -ESTALE; out_force: if (flags & LOOKUP_RCU) return -ECHILD; @@ -1037,6 +1589,13 @@ out_force: goto out; } +static void nfs_mark_dir_for_revalidate(struct inode *inode) +{ + spin_lock(&inode->i_lock); + nfs_set_cache_invalid(inode, NFS_INO_INVALID_CHANGE); + spin_unlock(&inode->i_lock); +} + /* * We judge how long we want to trust negative * dentries by looking at the parent inode mtime. @@ -1046,19 +1605,124 @@ out_force: * * If LOOKUP_RCU prevents us from performing a full check, return 1 * suggesting a reval is needed. + * + * Note that when creating a new file, or looking up a rename target, + * then it shouldn't be necessary to revalidate a negative dentry. */ static inline int nfs_neg_need_reval(struct inode *dir, struct dentry *dentry, unsigned int flags) { - /* Don't revalidate a negative dentry if we're creating a new file */ - if (flags & LOOKUP_CREATE) + if (flags & (LOOKUP_CREATE | LOOKUP_RENAME_TARGET)) return 0; if (NFS_SERVER(dir)->flags & NFS_MOUNT_LOOKUP_CACHE_NONEG) return 1; + /* Case insensitive server? Revalidate negative dentries */ + if (nfs_server_capable(dir, NFS_CAP_CASE_INSENSITIVE)) + return 1; return !nfs_check_verifier(dir, dentry, flags & LOOKUP_RCU); } +static int +nfs_lookup_revalidate_done(struct inode *dir, struct dentry *dentry, + struct inode *inode, int error) +{ + switch (error) { + case 1: + break; + case -ETIMEDOUT: + if (inode && (IS_ROOT(dentry) || + NFS_SERVER(inode)->flags & NFS_MOUNT_SOFTREVAL)) + error = 1; + break; + case -ESTALE: + case -ENOENT: + error = 0; + fallthrough; + default: + /* + * We can't d_drop the root of a disconnected tree: + * its d_hash is on the s_anon list and d_drop() would hide + * it from shrink_dcache_for_unmount(), leading to busy + * inodes on unmount and further oopses. + */ + if (inode && IS_ROOT(dentry)) + error = 1; + break; + } + trace_nfs_lookup_revalidate_exit(dir, dentry, 0, error); + return error; +} + +static int +nfs_lookup_revalidate_negative(struct inode *dir, struct dentry *dentry, + unsigned int flags) +{ + int ret = 1; + if (nfs_neg_need_reval(dir, dentry, flags)) { + if (flags & LOOKUP_RCU) + return -ECHILD; + ret = 0; + } + return nfs_lookup_revalidate_done(dir, dentry, NULL, ret); +} + +static int +nfs_lookup_revalidate_delegated(struct inode *dir, struct dentry *dentry, + struct inode *inode) +{ + nfs_set_verifier(dentry, nfs_save_change_attribute(dir)); + return nfs_lookup_revalidate_done(dir, dentry, inode, 1); +} + +static int nfs_lookup_revalidate_dentry(struct inode *dir, const struct qstr *name, + struct dentry *dentry, + struct inode *inode, unsigned int flags) +{ + struct nfs_fh *fhandle; + struct nfs_fattr *fattr; + unsigned long dir_verifier; + int ret; + + trace_nfs_lookup_revalidate_enter(dir, dentry, flags); + + ret = -ENOMEM; + fhandle = nfs_alloc_fhandle(); + fattr = nfs_alloc_fattr_with_label(NFS_SERVER(inode)); + if (fhandle == NULL || fattr == NULL) + goto out; + + dir_verifier = nfs_save_change_attribute(dir); + ret = NFS_PROTO(dir)->lookup(dir, dentry, name, fhandle, fattr); + if (ret < 0) + goto out; + + /* Request help from readdirplus */ + nfs_lookup_advise_force_readdirplus(dir, flags); + + ret = 0; + if (nfs_compare_fh(NFS_FH(inode), fhandle)) + goto out; + if (nfs_refresh_inode(inode, fattr) < 0) + goto out; + + nfs_setsecurity(inode, fattr); + nfs_set_verifier(dentry, dir_verifier); + + ret = 1; +out: + nfs_free_fattr(fattr); + nfs_free_fhandle(fhandle); + + /* + * If the lookup failed despite the dentry change attribute being + * a match, then we should revalidate the directory cache. + */ + if (!ret && nfs_dentry_verify_change(dir, dentry)) + nfs_mark_dir_for_revalidate(dir); + return nfs_lookup_revalidate_done(dir, dentry, inode, ret); +} + /* * This is called every time the dcache has a lookup hit, * and we should check whether we can really trust that @@ -1070,60 +1734,41 @@ int nfs_neg_need_reval(struct inode *dir, struct dentry *dentry, * If the parent directory is seen to have changed, we throw out the * cached dentry and do a new lookup. */ -static int nfs_lookup_revalidate(struct dentry *dentry, unsigned int flags) +static int +nfs_do_lookup_revalidate(struct inode *dir, const struct qstr *name, + struct dentry *dentry, unsigned int flags) { - struct inode *dir; struct inode *inode; - struct dentry *parent; - struct nfs_fh *fhandle = NULL; - struct nfs_fattr *fattr = NULL; - struct nfs4_label *label = NULL; - int error; + int error = 0; - if (flags & LOOKUP_RCU) { - parent = ACCESS_ONCE(dentry->d_parent); - dir = d_inode_rcu(parent); - if (!dir) - return -ECHILD; - } else { - parent = dget_parent(dentry); - dir = d_inode(parent); - } nfs_inc_stats(dir, NFSIOS_DENTRYREVALIDATE); inode = d_inode(dentry); - if (!inode) { - if (nfs_neg_need_reval(dir, dentry, flags)) { - if (flags & LOOKUP_RCU) - return -ECHILD; - goto out_bad; - } - goto out_valid; - } + if (!inode) + return nfs_lookup_revalidate_negative(dir, dentry, flags); if (is_bad_inode(inode)) { - if (flags & LOOKUP_RCU) - return -ECHILD; dfprintk(LOOKUPCACHE, "%s: %pd2 has dud inode\n", __func__, dentry); goto out_bad; } - if (NFS_PROTO(dir)->have_delegation(inode, FMODE_READ)) - goto out_set_verifier; + if ((flags & LOOKUP_RENAME_TARGET) && d_count(dentry) < 2 && + nfs_server_capable(dir, NFS_CAP_CASE_INSENSITIVE)) + goto out_bad; + + if (nfs_verifier_is_delegated(dentry)) + return nfs_lookup_revalidate_delegated(dir, dentry, inode); /* Force a full look up iff the parent directory has changed */ - if (!nfs_is_exclusive_create(dir, flags) && + if (!(flags & (LOOKUP_EXCL | LOOKUP_REVAL)) && nfs_check_verifier(dir, dentry, flags & LOOKUP_RCU)) { error = nfs_lookup_verify_inode(inode, flags); if (error) { - if (flags & LOOKUP_RCU) - return -ECHILD; if (error == -ESTALE) - goto out_zap_parent; - goto out_error; + nfs_mark_dir_for_revalidate(dir); + goto out_bad; } - nfs_advise_use_readdirplus(dir); goto out_valid; } @@ -1133,81 +1778,57 @@ static int nfs_lookup_revalidate(struct dentry *dentry, unsigned int flags) if (NFS_STALE(inode)) goto out_bad; - error = -ENOMEM; - fhandle = nfs_alloc_fhandle(); - fattr = nfs_alloc_fattr(); - if (fhandle == NULL || fattr == NULL) - goto out_error; + return nfs_lookup_revalidate_dentry(dir, name, dentry, inode, flags); +out_valid: + return nfs_lookup_revalidate_done(dir, dentry, inode, 1); +out_bad: + if (flags & LOOKUP_RCU) + return -ECHILD; + return nfs_lookup_revalidate_done(dir, dentry, inode, error); +} - label = nfs4_label_alloc(NFS_SERVER(inode), GFP_NOWAIT); - if (IS_ERR(label)) - goto out_error; +static int +__nfs_lookup_revalidate(struct dentry *dentry, unsigned int flags) +{ + if (flags & LOOKUP_RCU) { + if (dentry->d_fsdata == NFS_FSDATA_BLOCKED) + return -ECHILD; + } else { + /* Wait for unlink to complete - see unblock_revalidate() */ + wait_var_event(&dentry->d_fsdata, + smp_load_acquire(&dentry->d_fsdata) + != NFS_FSDATA_BLOCKED); + } + return 0; +} - trace_nfs_lookup_revalidate_enter(dir, dentry, flags); - error = NFS_PROTO(dir)->lookup(dir, &dentry->d_name, fhandle, fattr, label); - trace_nfs_lookup_revalidate_exit(dir, dentry, flags, error); - if (error == -ESTALE || error == -ENOENT) - goto out_bad; - if (error) - goto out_error; - if (nfs_compare_fh(NFS_FH(inode), fhandle)) - goto out_bad; - if ((error = nfs_refresh_inode(inode, fattr)) != 0) - goto out_bad; +static int nfs_lookup_revalidate(struct inode *dir, const struct qstr *name, + struct dentry *dentry, unsigned int flags) +{ + if (__nfs_lookup_revalidate(dentry, flags)) + return -ECHILD; + return nfs_do_lookup_revalidate(dir, name, dentry, flags); +} - nfs_setsecurity(inode, fattr, label); +static void block_revalidate(struct dentry *dentry) +{ + /* old devname - just in case */ + kfree(dentry->d_fsdata); - nfs_free_fattr(fattr); - nfs_free_fhandle(fhandle); - nfs4_label_free(label); + /* Any new reference that could lead to an open + * will take ->d_lock in lookup_open() -> d_lookup(). + * Holding this lock ensures we cannot race with + * __nfs_lookup_revalidate() and removes and need + * for further barriers. + */ + lockdep_assert_held(&dentry->d_lock); - /* set a readdirplus hint that we had a cache miss */ - nfs_force_use_readdirplus(dir); + dentry->d_fsdata = NFS_FSDATA_BLOCKED; +} -out_set_verifier: - nfs_set_verifier(dentry, nfs_save_change_attribute(dir)); - out_valid: - if (flags & LOOKUP_RCU) { - if (parent != ACCESS_ONCE(dentry->d_parent)) - return -ECHILD; - } else - dput(parent); - dfprintk(LOOKUPCACHE, "NFS: %s(%pd2) is valid\n", - __func__, dentry); - return 1; -out_zap_parent: - nfs_zap_caches(dir); - out_bad: - WARN_ON(flags & LOOKUP_RCU); - nfs_free_fattr(fattr); - nfs_free_fhandle(fhandle); - nfs4_label_free(label); - nfs_mark_for_revalidate(dir); - if (inode && S_ISDIR(inode->i_mode)) { - /* Purge readdir caches. */ - nfs_zap_caches(inode); - /* - * We can't d_drop the root of a disconnected tree: - * its d_hash is on the s_anon list and d_drop() would hide - * it from shrink_dcache_for_unmount(), leading to busy - * inodes on unmount and further oopses. - */ - if (IS_ROOT(dentry)) - goto out_valid; - } - dput(parent); - dfprintk(LOOKUPCACHE, "NFS: %s(%pd2) is invalid\n", - __func__, dentry); - return 0; -out_error: - WARN_ON(flags & LOOKUP_RCU); - nfs_free_fattr(fattr); - nfs_free_fhandle(fhandle); - nfs4_label_free(label); - dput(parent); - dfprintk(LOOKUPCACHE, "NFS: %s(%pd2) lookup returned error %d\n", - __func__, dentry, error); - return error; +static void unblock_revalidate(struct dentry *dentry) +{ + store_release_wake_up(&dentry->d_fsdata, NULL); } /* @@ -1241,8 +1862,7 @@ static int nfs_weak_revalidate(struct dentry *dentry, unsigned int flags) return 0; } - if (nfs_mapping_need_revalidate_inode(inode)) - error = __nfs_revalidate_inode(NFS_SERVER(inode), inode); + error = nfs_lookup_verify_inode(inode, flags); dfprintk(LOOKUPCACHE, "NFS: %s: inode %lu is %s\n", __func__, inode->i_ino, error ? "invalid" : "valid"); return !error; @@ -1264,7 +1884,7 @@ static int nfs_dentry_delete(const struct dentry *dentry) /* Unhash it, so that ->d_iput() would be called */ return 1; } - if (!(dentry->d_sb->s_flags & MS_ACTIVE)) { + if (!(dentry->d_sb->s_flags & SB_ACTIVE)) { /* Unhash it, so that ancestors of killed async unlink * files will be cleaned up during umount */ return 1; @@ -1278,9 +1898,12 @@ static void nfs_drop_nlink(struct inode *inode) { spin_lock(&inode->i_lock); /* drop the inode if we're reasonably sure this is the last link */ - if (inode->i_nlink == 1) - clear_nlink(inode); - NFS_I(inode)->cache_validity |= NFS_INO_INVALID_ATTR; + if (inode->i_nlink > 0) + drop_nlink(inode); + NFS_I(inode)->attr_gencount = nfs_inc_attr_generation_counter(); + nfs_set_cache_invalid( + inode, NFS_INO_INVALID_CHANGE | NFS_INO_INVALID_CTIME | + NFS_INO_INVALID_NLINK); spin_unlock(&inode->i_lock); } @@ -1290,10 +1913,6 @@ static void nfs_drop_nlink(struct inode *inode) */ static void nfs_dentry_iput(struct dentry *dentry, struct inode *inode) { - if (S_ISDIR(inode->i_mode)) - /* drop any readdir cache as it could easily be old */ - NFS_I(inode)->cache_validity |= NFS_INO_INVALID_DATA; - if (dentry->d_flags & DCACHE_NFSFS_RENAMED) { nfs_complete_unlink(dentry, inode); nfs_drop_nlink(inode); @@ -1328,7 +1947,7 @@ struct dentry *nfs_lookup(struct inode *dir, struct dentry * dentry, unsigned in struct inode *inode = NULL; struct nfs_fh *fhandle = NULL; struct nfs_fattr *fattr = NULL; - struct nfs4_label *label = NULL; + unsigned long dir_verifier; int error; dfprintk(VFS, "NFS: lookup(%pd2)\n", dentry); @@ -1341,58 +1960,67 @@ struct dentry *nfs_lookup(struct inode *dir, struct dentry * dentry, unsigned in * If we're doing an exclusive create, optimize away the lookup * but don't hash the dentry. */ - if (nfs_is_exclusive_create(dir, flags)) + if (nfs_is_exclusive_create(dir, flags) || flags & LOOKUP_RENAME_TARGET) return NULL; res = ERR_PTR(-ENOMEM); fhandle = nfs_alloc_fhandle(); - fattr = nfs_alloc_fattr(); + fattr = nfs_alloc_fattr_with_label(NFS_SERVER(dir)); if (fhandle == NULL || fattr == NULL) goto out; - label = nfs4_label_alloc(NFS_SERVER(dir), GFP_NOWAIT); - if (IS_ERR(label)) - goto out; - + dir_verifier = nfs_save_change_attribute(dir); trace_nfs_lookup_enter(dir, dentry, flags); - error = NFS_PROTO(dir)->lookup(dir, &dentry->d_name, fhandle, fattr, label); - if (error == -ENOENT) + error = NFS_PROTO(dir)->lookup(dir, dentry, &dentry->d_name, + fhandle, fattr); + if (error == -ENOENT) { + if (nfs_server_capable(dir, NFS_CAP_CASE_INSENSITIVE)) + dir_verifier = inode_peek_iversion_raw(dir); goto no_entry; + } if (error < 0) { res = ERR_PTR(error); - goto out_label; + goto out; } - inode = nfs_fhget(dentry->d_sb, fhandle, fattr, label); + inode = nfs_fhget(dentry->d_sb, fhandle, fattr); res = ERR_CAST(inode); if (IS_ERR(res)) - goto out_label; + goto out; /* Notify readdir to use READDIRPLUS */ - nfs_force_use_readdirplus(dir); + nfs_lookup_advise_force_readdirplus(dir, flags); no_entry: res = d_splice_alias(inode, dentry); if (res != NULL) { if (IS_ERR(res)) - goto out_label; + goto out; dentry = res; } - nfs_set_verifier(dentry, nfs_save_change_attribute(dir)); -out_label: - trace_nfs_lookup_exit(dir, dentry, flags, error); - nfs4_label_free(label); + nfs_set_verifier(dentry, dir_verifier); out: + trace_nfs_lookup_exit(dir, dentry, flags, PTR_ERR_OR_ZERO(res)); nfs_free_fattr(fattr); nfs_free_fhandle(fhandle); return res; } EXPORT_SYMBOL_GPL(nfs_lookup); +void nfs_d_prune_case_insensitive_aliases(struct inode *inode) +{ + /* Case insensitive server? Revalidate dentries */ + if (inode && nfs_server_capable(inode, NFS_CAP_CASE_INSENSITIVE)) + d_prune_aliases(inode); +} +EXPORT_SYMBOL_GPL(nfs_d_prune_case_insensitive_aliases); + #if IS_ENABLED(CONFIG_NFS_V4) -static int nfs4_lookup_revalidate(struct dentry *, unsigned int); +static int nfs4_lookup_revalidate(struct inode *, const struct qstr *, + struct dentry *, unsigned int); const struct dentry_operations nfs4_dentry_operations = { .d_revalidate = nfs4_lookup_revalidate, + .d_weak_revalidate = nfs_weak_revalidate, .d_delete = nfs_dentry_delete, .d_iput = nfs_dentry_iput, .d_automount = nfs_d_automount, @@ -1400,16 +2028,6 @@ const struct dentry_operations nfs4_dentry_operations = { }; EXPORT_SYMBOL_GPL(nfs4_dentry_operations); -static fmode_t flags_to_mode(int flags) -{ - fmode_t res = (__force fmode_t)flags & FMODE_EXEC; - if ((flags & O_ACCMODE) != O_WRONLY) - res |= FMODE_READ; - if ((flags & O_ACCMODE) != O_RDONLY) - res |= FMODE_WRITE; - return res; -} - static struct nfs_open_context *create_nfs_open_context(struct dentry *dentry, int open_flags, struct file *filp) { return alloc_nfs_open_context(dentry, flags_to_mode(open_flags), filp); @@ -1423,25 +2041,24 @@ static int do_open(struct inode *inode, struct file *filp) static int nfs_finish_open(struct nfs_open_context *ctx, struct dentry *dentry, - struct file *file, unsigned open_flags, - int *opened) + struct file *file, unsigned open_flags) { int err; - err = finish_open(file, dentry, do_open, opened); + err = finish_open(file, dentry, do_open); if (err) goto out; - if (S_ISREG(file->f_path.dentry->d_inode->i_mode)) + if (S_ISREG(file_inode(file)->i_mode)) nfs_file_set_open_context(file, ctx); else - err = -ESTALE; + err = -EOPENSTALE; out: return err; } int nfs_atomic_open(struct inode *dir, struct dentry *dentry, struct file *file, unsigned open_flags, - umode_t mode, int *opened) + umode_t mode) { DECLARE_WAIT_QUEUE_HEAD_ONSTACK(wq); struct nfs_open_context *ctx; @@ -1449,7 +2066,9 @@ int nfs_atomic_open(struct inode *dir, struct dentry *dentry, struct iattr attr = { .ia_valid = ATTR_OPEN }; struct inode *inode; unsigned int lookup_flags = 0; + unsigned long dir_verifier; bool switched = false; + int created = 0; int err; /* Expect a negative dentry */ @@ -1510,7 +2129,9 @@ int nfs_atomic_open(struct inode *dir, struct dentry *dentry, goto out; trace_nfs_atomic_open_enter(dir, ctx, open_flags); - inode = NFS_PROTO(dir)->open_context(dir, ctx, open_flags, &attr, opened); + inode = NFS_PROTO(dir)->open_context(dir, ctx, open_flags, &attr, &created); + if (created) + file->f_mode |= FMODE_CREATED; if (IS_ERR(inode)) { err = PTR_ERR(inode); trace_nfs_atomic_open_exit(dir, ctx, open_flags, err); @@ -1519,7 +2140,11 @@ int nfs_atomic_open(struct inode *dir, struct dentry *dentry, switch (err) { case -ENOENT: d_splice_alias(NULL, dentry); - nfs_set_verifier(dentry, nfs_save_change_attribute(dir)); + if (nfs_server_capable(dir, NFS_CAP_CASE_INSENSITIVE)) + dir_verifier = inode_peek_iversion_raw(dir); + else + dir_verifier = nfs_save_change_attribute(dir); + nfs_set_verifier(dentry, dir_verifier); break; case -EISDIR: case -ENOTDIR: @@ -1534,8 +2159,9 @@ 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, opened); + err = nfs_finish_open(ctx, ctx->dentry, file, open_flags); trace_nfs_atomic_open_exit(dir, ctx, open_flags, err); put_nfs_open_context(ctx); out: @@ -1547,6 +2173,24 @@ out: no_open: res = nfs_lookup(dir, dentry, lookup_flags); + if (!res) { + inode = d_inode(dentry); + if ((lookup_flags & LOOKUP_DIRECTORY) && inode && + !(S_ISDIR(inode->i_mode) || S_ISLNK(inode->i_mode))) + res = ERR_PTR(-ENOTDIR); + else if (inode && S_ISREG(inode->i_mode)) + res = ERR_PTR(-EOPENSTALE); + } else if (!IS_ERR(res)) { + inode = d_inode(res); + if ((lookup_flags & LOOKUP_DIRECTORY) && inode && + !(S_ISDIR(inode->i_mode) || S_ISLNK(inode->i_mode))) { + dput(res); + res = ERR_PTR(-ENOTDIR); + } else if (inode && S_ISREG(inode->i_mode)) { + dput(res); + res = ERR_PTR(-EOPENSTALE); + } + } if (switched) { d_lookup_done(dentry); if (!res) @@ -1554,113 +2198,147 @@ no_open: else dput(dentry); } - if (IS_ERR(res)) - return PTR_ERR(res); return finish_no_open(file, res); } EXPORT_SYMBOL_GPL(nfs_atomic_open); -static int nfs4_lookup_revalidate(struct dentry *dentry, unsigned int flags) +static int +nfs4_lookup_revalidate(struct inode *dir, const struct qstr *name, + struct dentry *dentry, unsigned int flags) { struct inode *inode; - int ret = 0; + + if (__nfs_lookup_revalidate(dentry, flags)) + return -ECHILD; + + trace_nfs_lookup_revalidate_enter(dir, dentry, flags); if (!(flags & LOOKUP_OPEN) || (flags & LOOKUP_DIRECTORY)) - goto no_open; + goto full_reval; if (d_mountpoint(dentry)) - goto no_open; - if (NFS_SB(dentry->d_sb)->caps & NFS_CAP_ATOMIC_OPEN_V1) - goto no_open; + goto full_reval; inode = d_inode(dentry); /* We can't create new files in nfs_open_revalidate(), so we * optimize away revalidation of negative dentries. */ - if (inode == NULL) { - struct dentry *parent; - struct inode *dir; - - if (flags & LOOKUP_RCU) { - parent = ACCESS_ONCE(dentry->d_parent); - dir = d_inode_rcu(parent); - if (!dir) - return -ECHILD; - } else { - parent = dget_parent(dentry); - dir = d_inode(parent); - } - if (!nfs_neg_need_reval(dir, dentry, flags)) - ret = 1; - else if (flags & LOOKUP_RCU) - ret = -ECHILD; - if (!(flags & LOOKUP_RCU)) - dput(parent); - else if (parent != ACCESS_ONCE(dentry->d_parent)) - return -ECHILD; - goto out; - } + if (inode == NULL) + goto full_reval; + + if (nfs_verifier_is_delegated(dentry)) + return nfs_lookup_revalidate_delegated(dir, dentry, inode); /* NFS only supports OPEN on regular files */ if (!S_ISREG(inode->i_mode)) - goto no_open; + goto full_reval; + /* We cannot do exclusive creation on a positive dentry */ - if (flags & LOOKUP_EXCL) - goto no_open; + if (flags & (LOOKUP_EXCL | LOOKUP_REVAL)) + goto reval_dentry; - /* Let f_op->open() actually open (and revalidate) the file */ - ret = 1; + /* Check if the directory changed */ + if (!nfs_check_verifier(dir, dentry, flags & LOOKUP_RCU)) + goto reval_dentry; -out: - return ret; + /* Let f_op->open() actually open (and revalidate) the file */ + return 1; +reval_dentry: + if (flags & LOOKUP_RCU) + return -ECHILD; + return nfs_lookup_revalidate_dentry(dir, name, dentry, inode, flags); -no_open: - return nfs_lookup_revalidate(dentry, flags); +full_reval: + return nfs_do_lookup_revalidate(dir, name, dentry, flags); } #endif /* CONFIG_NFSV4 */ -/* - * Code common to create, mkdir, and mknod. - */ -int nfs_instantiate(struct dentry *dentry, struct nfs_fh *fhandle, - struct nfs_fattr *fattr, - struct nfs4_label *label) +int nfs_atomic_open_v23(struct inode *dir, struct dentry *dentry, + struct file *file, unsigned int open_flags, + umode_t mode) +{ + struct dentry *res = NULL; + /* Same as look+open from lookup_open(), but with different O_TRUNC + * handling. + */ + int error = 0; + + if (dentry->d_name.len > NFS_SERVER(dir)->namelen) + return -ENAMETOOLONG; + + if (open_flags & O_CREAT) { + error = nfs_do_create(dir, dentry, mode, open_flags); + if (!error) { + file->f_mode |= FMODE_CREATED; + return finish_open(file, dentry, NULL); + } else if (error != -EEXIST || open_flags & O_EXCL) + return error; + } + if (d_in_lookup(dentry)) { + /* The only flags nfs_lookup considers are + * LOOKUP_EXCL and LOOKUP_RENAME_TARGET, and + * we want those to be zero so the lookup isn't skipped. + */ + res = nfs_lookup(dir, dentry, 0); + } + return finish_no_open(file, res); + +} +EXPORT_SYMBOL_GPL(nfs_atomic_open_v23); + +struct dentry * +nfs_add_or_obtain(struct dentry *dentry, struct nfs_fh *fhandle, + struct nfs_fattr *fattr) { struct dentry *parent = dget_parent(dentry); struct inode *dir = d_inode(parent); struct inode *inode; - int error = -EACCES; + struct dentry *d; + int error; d_drop(dentry); - /* We may have been initialized further down */ - if (d_really_is_positive(dentry)) - goto out; if (fhandle->size == 0) { - error = NFS_PROTO(dir)->lookup(dir, &dentry->d_name, fhandle, fattr, NULL); + error = NFS_PROTO(dir)->lookup(dir, dentry, &dentry->d_name, + fhandle, fattr); if (error) goto out_error; } nfs_set_verifier(dentry, nfs_save_change_attribute(dir)); if (!(fattr->valid & NFS_ATTR_FATTR)) { struct nfs_server *server = NFS_SB(dentry->d_sb); - error = server->nfs_client->rpc_ops->getattr(server, fhandle, fattr, NULL); + error = server->nfs_client->rpc_ops->getattr(server, fhandle, + fattr, NULL); if (error < 0) goto out_error; } - inode = nfs_fhget(dentry->d_sb, fhandle, fattr, label); - error = PTR_ERR(inode); - if (IS_ERR(inode)) - goto out_error; - d_add(dentry, inode); + inode = nfs_fhget(dentry->d_sb, fhandle, fattr); + d = d_splice_alias(inode, dentry); out: dput(parent); - return 0; + return d; out_error: - nfs_mark_for_revalidate(dir); - dput(parent); - return error; + d = ERR_PTR(error); + goto out; +} +EXPORT_SYMBOL_GPL(nfs_add_or_obtain); + +/* + * Code common to create, mkdir, and mknod. + */ +int nfs_instantiate(struct dentry *dentry, struct nfs_fh *fhandle, + struct nfs_fattr *fattr) +{ + struct dentry *d; + + d = nfs_add_or_obtain(dentry, fhandle, fattr); + if (IS_ERR(d)) + return PTR_ERR(d); + + /* Callers don't care */ + dput(d); + return 0; } EXPORT_SYMBOL_GPL(nfs_instantiate); @@ -1670,18 +2348,23 @@ EXPORT_SYMBOL_GPL(nfs_instantiate); * that the operation succeeded on the server, but an error in the * reply path made it appear to have failed. */ -int nfs_create(struct inode *dir, struct dentry *dentry, - umode_t mode, bool excl) +static int nfs_do_create(struct inode *dir, struct dentry *dentry, + umode_t mode, int open_flags) { struct iattr attr; - int open_flags = excl ? O_CREAT | O_EXCL : O_CREAT; int error; + open_flags |= O_CREAT; + dfprintk(VFS, "NFS: create(%s/%lu), %pd\n", dir->i_sb->s_id, dir->i_ino, dentry); attr.ia_mode = mode; attr.ia_valid = ATTR_MODE; + if (open_flags & O_TRUNC) { + attr.ia_size = 0; + attr.ia_valid |= ATTR_SIZE; + } trace_nfs_create_enter(dir, dentry, open_flags); error = NFS_PROTO(dir)->create(dir, dentry, &attr, open_flags); @@ -1693,13 +2376,20 @@ out_err: d_drop(dentry); return error; } + +int nfs_create(struct mnt_idmap *idmap, struct inode *dir, + struct dentry *dentry, umode_t mode, bool excl) +{ + return nfs_do_create(dir, dentry, mode, excl ? O_EXCL : 0); +} EXPORT_SYMBOL_GPL(nfs_create); /* * See comments for nfs_proc_create regarding failed operations. */ int -nfs_mknod(struct inode *dir, struct dentry *dentry, umode_t mode, dev_t rdev) +nfs_mknod(struct mnt_idmap *idmap, struct inode *dir, + struct dentry *dentry, umode_t mode, dev_t rdev) { struct iattr attr; int status; @@ -1725,10 +2415,11 @@ EXPORT_SYMBOL_GPL(nfs_mknod); /* * See comments for nfs_proc_create regarding failed operations. */ -int nfs_mkdir(struct inode *dir, struct dentry *dentry, umode_t mode) +struct dentry *nfs_mkdir(struct mnt_idmap *idmap, struct inode *dir, + struct dentry *dentry, umode_t mode) { struct iattr attr; - int error; + struct dentry *ret; dfprintk(VFS, "NFS: mkdir(%s/%lu), %pd\n", dir->i_sb->s_id, dir->i_ino, dentry); @@ -1737,14 +2428,9 @@ int nfs_mkdir(struct inode *dir, struct dentry *dentry, umode_t mode) attr.ia_mode = mode | S_IFDIR; trace_nfs_mkdir_enter(dir, dentry); - error = NFS_PROTO(dir)->mkdir(dir, dentry, &attr); - trace_nfs_mkdir_exit(dir, dentry, error); - if (error != 0) - goto out_err; - return 0; -out_err: - d_drop(dentry); - return error; + ret = NFS_PROTO(dir)->mkdir(dir, dentry, &attr); + trace_nfs_mkdir_exit(dir, dentry, PTR_ERR_OR_ZERO(ret)); + return ret; } EXPORT_SYMBOL_GPL(nfs_mkdir); @@ -1754,6 +2440,21 @@ static void nfs_dentry_handle_enoent(struct dentry *dentry) d_delete(dentry); } +static void nfs_dentry_remove_handle_error(struct inode *dir, + struct dentry *dentry, int error) +{ + switch (error) { + case -ENOENT: + if (d_really_is_positive(dentry)) + d_delete(dentry); + nfs_set_verifier(dentry, nfs_save_change_attribute(dir)); + break; + case 0: + nfs_d_prune_case_insensitive_aliases(d_inode(dentry)); + nfs_set_verifier(dentry, nfs_save_change_attribute(dir)); + } +} + int nfs_rmdir(struct inode *dir, struct dentry *dentry) { int error; @@ -1776,6 +2477,7 @@ int nfs_rmdir(struct inode *dir, struct dentry *dentry) up_write(&NFS_I(d_inode(dentry))->rmdir_sem); } else error = NFS_PROTO(dir)->rmdir(dir, &dentry->d_name); + nfs_dentry_remove_handle_error(dir, dentry, error); trace_nfs_rmdir_exit(dir, dentry, error); return error; @@ -1805,12 +2507,11 @@ static int nfs_safe_remove(struct dentry *dentry) trace_nfs_remove_enter(dir, dentry); if (inode != NULL) { - NFS_PROTO(inode)->return_delegation(inode); - error = NFS_PROTO(dir)->remove(dir, &dentry->d_name); + error = NFS_PROTO(dir)->remove(dir, dentry); if (error == 0) nfs_drop_nlink(inode); } else - error = NFS_PROTO(dir)->remove(dir, &dentry->d_name); + error = NFS_PROTO(dir)->remove(dir, dentry); if (error == -ENOENT) nfs_dentry_handle_enoent(dentry); trace_nfs_remove_exit(dir, dentry, error); @@ -1826,30 +2527,37 @@ out: int nfs_unlink(struct inode *dir, struct dentry *dentry) { int error; - int need_rehash = 0; dfprintk(VFS, "NFS: unlink(%s/%lu, %pd)\n", dir->i_sb->s_id, dir->i_ino, dentry); trace_nfs_unlink_enter(dir, dentry); spin_lock(&dentry->d_lock); - if (d_count(dentry) > 1) { + if (d_count(dentry) > 1 && !test_bit(NFS_INO_PRESERVE_UNLINKED, + &NFS_I(d_inode(dentry))->flags)) { spin_unlock(&dentry->d_lock); /* Start asynchronous writeout of the inode */ write_inode_now(d_inode(dentry), 0); error = nfs_sillyrename(dir, dentry); goto out; } - if (!d_unhashed(dentry)) { - __d_drop(dentry); - need_rehash = 1; + /* We must prevent any concurrent open until the unlink + * completes. ->d_revalidate will wait for ->d_fsdata + * to clear. We set it here to ensure no lookup succeeds until + * the unlink is complete on the server. + */ + error = -ETXTBSY; + if (WARN_ON(dentry->d_flags & DCACHE_NFSFS_RENAMED) || + WARN_ON(dentry->d_fsdata == NFS_FSDATA_BLOCKED)) { + spin_unlock(&dentry->d_lock); + goto out; } + block_revalidate(dentry); + spin_unlock(&dentry->d_lock); error = nfs_safe_remove(dentry); - if (!error || error == -ENOENT) { - nfs_set_verifier(dentry, nfs_save_change_attribute(dir)); - } else if (need_rehash) - d_rehash(dentry); + nfs_dentry_remove_handle_error(dir, dentry, error); + unblock_revalidate(dentry); out: trace_nfs_unlink_exit(dir, dentry, error); return error; @@ -1871,9 +2579,10 @@ EXPORT_SYMBOL_GPL(nfs_unlink); * now have a new file handle and can instantiate an in-core NFS inode * and move the raw page into its mapping. */ -int nfs_symlink(struct inode *dir, struct dentry *dentry, const char *symname) +int nfs_symlink(struct mnt_idmap *idmap, struct inode *dir, + struct dentry *dentry, const char *symname) { - struct page *page; + struct folio *folio; char *kaddr; struct iattr attr; unsigned int pathlen = strlen(symname); @@ -1888,43 +2597,40 @@ int nfs_symlink(struct inode *dir, struct dentry *dentry, const char *symname) attr.ia_mode = S_IFLNK | S_IRWXUGO; attr.ia_valid = ATTR_MODE; - page = alloc_page(GFP_USER); - if (!page) + folio = folio_alloc(GFP_USER, 0); + if (!folio) return -ENOMEM; - kaddr = page_address(page); + kaddr = folio_address(folio); memcpy(kaddr, symname, pathlen); if (pathlen < PAGE_SIZE) memset(kaddr + pathlen, 0, PAGE_SIZE - pathlen); trace_nfs_symlink_enter(dir, dentry); - error = NFS_PROTO(dir)->symlink(dir, dentry, page, pathlen, &attr); + error = NFS_PROTO(dir)->symlink(dir, dentry, folio, pathlen, &attr); trace_nfs_symlink_exit(dir, dentry, error); if (error != 0) { dfprintk(VFS, "NFS: symlink(%s/%lu, %pd, %s) error %d\n", dir->i_sb->s_id, dir->i_ino, dentry, symname, error); d_drop(dentry); - __free_page(page); + folio_put(folio); return error; } + nfs_set_verifier(dentry, nfs_save_change_attribute(dir)); + /* * No big deal if we can't add this page to the page cache here. * READLINK will get the missing page from the server if needed. */ - if (!add_to_page_cache_lru(page, d_inode(dentry)->i_mapping, 0, - GFP_KERNEL)) { - SetPageUptodate(page); - unlock_page(page); - /* - * add_to_page_cache_lru() grabs an extra page refcount. - * Drop it here to avoid leaking this page later. - */ - put_page(page); - } else - __free_page(page); + if (filemap_add_folio(d_inode(dentry)->i_mapping, folio, 0, + GFP_KERNEL) == 0) { + folio_mark_uptodate(folio); + folio_unlock(folio); + } + folio_put(folio); return 0; } EXPORT_SYMBOL_GPL(nfs_symlink); @@ -1939,11 +2645,12 @@ nfs_link(struct dentry *old_dentry, struct inode *dir, struct dentry *dentry) old_dentry, dentry); trace_nfs_link_enter(inode, dir, dentry); - NFS_PROTO(inode)->return_delegation(inode); - d_drop(dentry); + if (S_ISREG(inode->i_mode)) + nfs_sync_inode(inode); error = NFS_PROTO(dir)->link(inode, dir, &dentry->d_name); if (error == 0) { + nfs_set_verifier(dentry, nfs_save_change_attribute(dir)); ihold(inode); d_add(dentry, inode); } @@ -1952,6 +2659,26 @@ nfs_link(struct dentry *old_dentry, struct inode *dir, struct dentry *dentry) } EXPORT_SYMBOL_GPL(nfs_link); +static void +nfs_unblock_rename(struct rpc_task *task, struct nfs_renamedata *data) +{ + struct dentry *new_dentry = data->new_dentry; + + unblock_revalidate(new_dentry); +} + +static bool nfs_rename_is_unsafe_cross_dir(struct dentry *old_dentry, + struct dentry *new_dentry) +{ + struct nfs_server *server = NFS_SB(old_dentry->d_sb); + + if (old_dentry->d_parent != new_dentry->d_parent) + return false; + if (server->fh_expire_type & NFS_FH_RENAME_UNSAFE) + return !(server->fh_expire_type & NFS_FH_NOEXPIRE_WITH_OPEN); + return true; +} + /* * RENAME * FIXME: Some nfsds, like the Linux user space nfsd, may generate a @@ -1976,14 +2703,15 @@ EXPORT_SYMBOL_GPL(nfs_link); * If these conditions are met, we can drop the dentries before doing * the rename. */ -int nfs_rename(struct inode *old_dir, struct dentry *old_dentry, - struct inode *new_dir, struct dentry *new_dentry, - unsigned int flags) +int nfs_rename(struct mnt_idmap *idmap, struct inode *old_dir, + struct dentry *old_dentry, struct inode *new_dir, + struct dentry *new_dentry, unsigned int flags) { struct inode *old_inode = d_inode(old_dentry); struct inode *new_inode = d_inode(new_dentry); - struct dentry *dentry = NULL, *rehash = NULL; + struct dentry *dentry = NULL; struct rpc_task *task; + bool must_unblock = false; int error = -EBUSY; if (flags) @@ -2001,18 +2729,22 @@ int nfs_rename(struct inode *old_dir, struct dentry *old_dentry, * the new target. */ if (new_inode && !S_ISDIR(new_inode->i_mode)) { - /* - * To prevent any new references to the target during the - * rename, we unhash the dentry in advance. + /* We must prevent any concurrent open until the unlink + * completes. ->d_revalidate will wait for ->d_fsdata + * to clear. We set it here to ensure no lookup succeeds until + * the unlink is complete on the server. */ - if (!d_unhashed(new_dentry)) { - d_drop(new_dentry); - rehash = new_dentry; - } + error = -ETXTBSY; + if (WARN_ON(new_dentry->d_flags & DCACHE_NFSFS_RENAMED) || + WARN_ON(new_dentry->d_fsdata == NFS_FSDATA_BLOCKED)) + goto out; + spin_lock(&new_dentry->d_lock); if (d_count(new_dentry) > 2) { int err; + spin_unlock(&new_dentry->d_lock); + /* copy the target dentry's name */ dentry = d_alloc(new_dentry->d_parent, &new_dentry->d_name); @@ -2025,17 +2757,23 @@ int nfs_rename(struct inode *old_dir, struct dentry *old_dentry, goto out; new_dentry = dentry; - rehash = NULL; new_inode = NULL; + } else { + block_revalidate(new_dentry); + must_unblock = true; + spin_unlock(&new_dentry->d_lock); } - } - NFS_PROTO(old_inode)->return_delegation(old_inode); - if (new_inode != NULL) - NFS_PROTO(new_inode)->return_delegation(new_inode); + } - task = nfs_async_rename(old_dir, new_dir, old_dentry, new_dentry, NULL); + if (S_ISREG(old_inode->i_mode) && + nfs_rename_is_unsafe_cross_dir(old_dentry, new_dentry)) + nfs_sync_inode(old_inode); + task = nfs_async_rename(old_dir, new_dir, old_dentry, new_dentry, + must_unblock ? nfs_unblock_rename : NULL); if (IS_ERR(task)) { + if (must_unblock) + unblock_revalidate(new_dentry); error = PTR_ERR(task); goto out; } @@ -2048,10 +2786,16 @@ int nfs_rename(struct inode *old_dir, struct dentry *old_dentry, } else error = task->tk_status; rpc_put_task(task); - nfs_mark_for_revalidate(old_inode); + /* Ensure the inode attributes are revalidated */ + if (error == 0) { + spin_lock(&old_inode->i_lock); + NFS_I(old_inode)->attr_gencount = nfs_inc_attr_generation_counter(); + nfs_set_cache_invalid(old_inode, NFS_INO_INVALID_CHANGE | + NFS_INO_INVALID_CTIME | + NFS_INO_REVAL_FORCED); + spin_unlock(&old_inode->i_lock); + } out: - if (rehash) - d_rehash(rehash); trace_nfs_rename_exit(old_dir, old_dentry, new_dir, new_dentry, error); if (!error) { @@ -2064,7 +2808,7 @@ out: * should mark the directories for revalidation. */ d_move(old_dentry, new_dentry); - nfs_set_verifier(new_dentry, + nfs_set_verifier(old_dentry, nfs_save_change_attribute(new_dir)); } else if (error == -ENOENT) nfs_dentry_handle_enoent(old_dentry); @@ -2080,13 +2824,13 @@ static DEFINE_SPINLOCK(nfs_access_lru_lock); static LIST_HEAD(nfs_access_lru_list); static atomic_long_t nfs_access_nr_entries; -static unsigned long nfs_access_max_cachesize = ULONG_MAX; +static unsigned long nfs_access_max_cachesize = 4*1024*1024; module_param(nfs_access_max_cachesize, ulong, 0644); MODULE_PARM_DESC(nfs_access_max_cachesize, "NFS access maximum total cache length"); static void nfs_access_free_entry(struct nfs_access_entry *entry) { - put_rpccred(entry->cred); + put_group_info(entry->group_info); kfree_rcu(entry, rcu_head); smp_mb__before_atomic(); atomic_long_dec(&nfs_access_nr_entries); @@ -2212,17 +2956,55 @@ void nfs_access_zap_cache(struct inode *inode) } EXPORT_SYMBOL_GPL(nfs_access_zap_cache); -static struct nfs_access_entry *nfs_access_search_rbtree(struct inode *inode, struct rpc_cred *cred) +static int access_cmp(const struct cred *a, const struct nfs_access_entry *b) +{ + struct group_info *ga, *gb; + int g; + + if (uid_lt(a->fsuid, b->fsuid)) + return -1; + if (uid_gt(a->fsuid, b->fsuid)) + return 1; + + if (gid_lt(a->fsgid, b->fsgid)) + return -1; + if (gid_gt(a->fsgid, b->fsgid)) + return 1; + + ga = a->group_info; + gb = b->group_info; + if (ga == gb) + return 0; + if (ga == NULL) + return -1; + if (gb == NULL) + return 1; + if (ga->ngroups < gb->ngroups) + return -1; + if (ga->ngroups > gb->ngroups) + return 1; + + for (g = 0; g < ga->ngroups; g++) { + if (gid_lt(ga->gid[g], gb->gid[g])) + return -1; + if (gid_gt(ga->gid[g], gb->gid[g])) + return 1; + } + return 0; +} + +static struct nfs_access_entry *nfs_access_search_rbtree(struct inode *inode, const struct cred *cred) { struct rb_node *n = NFS_I(inode)->access_cache.rb_node; - struct nfs_access_entry *entry; while (n != NULL) { - entry = rb_entry(n, struct nfs_access_entry, rb_node); + struct nfs_access_entry *entry = + rb_entry(n, struct nfs_access_entry, rb_node); + int cmp = access_cmp(cred, entry); - if (cred < entry->cred) + if (cmp < 0) n = n->rb_left; - else if (cred > entry->cred) + else if (cmp > 0) n = n->rb_right; else return entry; @@ -2230,9 +3012,30 @@ static struct nfs_access_entry *nfs_access_search_rbtree(struct inode *inode, st return NULL; } -static int nfs_access_get_cached(struct inode *inode, struct rpc_cred *cred, struct nfs_access_entry *res, bool may_block) +static u64 nfs_access_login_time(const struct task_struct *task, + const struct cred *cred) +{ + const struct task_struct *parent; + const struct cred *pcred; + u64 ret; + + rcu_read_lock(); + for (;;) { + parent = rcu_dereference(task->real_parent); + pcred = __task_cred(parent); + if (parent == task || cred_fscmp(pcred, cred) != 0) + break; + task = parent; + } + ret = task->start_time; + rcu_read_unlock(); + return ret; +} + +static int nfs_access_get_cached_locked(struct inode *inode, const struct cred *cred, u32 *mask, bool may_block) { struct nfs_inode *nfsi = NFS_I(inode); + u64 login_time = nfs_access_login_time(current, cred); struct nfs_access_entry *cache; bool retry = true; int err; @@ -2248,11 +3051,11 @@ static int nfs_access_get_cached(struct inode *inode, struct rpc_cred *cred, str /* Found an entry, is our attribute cache valid? */ if (!nfs_check_cache_invalid(inode, NFS_INO_INVALID_ACCESS)) break; + if (!retry) + break; err = -ECHILD; if (!may_block) goto out; - if (!retry) - goto out_zap; spin_unlock(&inode->i_lock); err = __nfs_revalidate_inode(NFS_SERVER(inode), inode); if (err) @@ -2260,9 +3063,10 @@ static int nfs_access_get_cached(struct inode *inode, struct rpc_cred *cred, str spin_lock(&inode->i_lock); retry = false; } - res->jiffies = cache->jiffies; - res->cred = cache->cred; - res->mask = cache->mask; + err = -ENOENT; + if ((s64)(login_time - cache->timestamp) > 0) + goto out; + *mask = cache->mask; list_move_tail(&cache->lru, &nfsi->access_cache_entry_lru); err = 0; out: @@ -2274,12 +3078,13 @@ out_zap: return -ENOENT; } -static int nfs_access_get_cached_rcu(struct inode *inode, struct rpc_cred *cred, struct nfs_access_entry *res) +static int nfs_access_get_cached_rcu(struct inode *inode, const struct cred *cred, u32 *mask) { /* Only check the most recently returned cache entry, * but do it without locking. */ struct nfs_inode *nfsi = NFS_I(inode); + u64 login_time = nfs_access_login_time(current, cred); struct nfs_access_entry *cache; int err = -ECHILD; struct list_head *lh; @@ -2287,40 +3092,58 @@ static int nfs_access_get_cached_rcu(struct inode *inode, struct rpc_cred *cred, rcu_read_lock(); if (nfsi->cache_validity & NFS_INO_INVALID_ACCESS) goto out; - lh = rcu_dereference(nfsi->access_cache_entry_lru.prev); + lh = rcu_dereference(list_tail_rcu(&nfsi->access_cache_entry_lru)); cache = list_entry(lh, struct nfs_access_entry, lru); if (lh == &nfsi->access_cache_entry_lru || - cred != cache->cred) + access_cmp(cred, cache) != 0) cache = NULL; if (cache == NULL) goto out; + if ((s64)(login_time - cache->timestamp) > 0) + goto out; if (nfs_check_cache_invalid(inode, NFS_INO_INVALID_ACCESS)) goto out; - res->jiffies = cache->jiffies; - res->cred = cache->cred; - res->mask = cache->mask; + *mask = cache->mask; err = 0; out: rcu_read_unlock(); return err; } -static void nfs_access_add_rbtree(struct inode *inode, struct nfs_access_entry *set) +int nfs_access_get_cached(struct inode *inode, const struct cred *cred, + u32 *mask, bool may_block) +{ + int status; + + status = nfs_access_get_cached_rcu(inode, cred, mask); + if (status != 0) + status = nfs_access_get_cached_locked(inode, cred, mask, + may_block); + + return status; +} +EXPORT_SYMBOL_GPL(nfs_access_get_cached); + +static void nfs_access_add_rbtree(struct inode *inode, + struct nfs_access_entry *set, + const struct cred *cred) { struct nfs_inode *nfsi = NFS_I(inode); struct rb_root *root_node = &nfsi->access_cache; struct rb_node **p = &root_node->rb_node; struct rb_node *parent = NULL; struct nfs_access_entry *entry; + int cmp; spin_lock(&inode->i_lock); while (*p != NULL) { parent = *p; entry = rb_entry(parent, struct nfs_access_entry, rb_node); + cmp = access_cmp(cred, entry); - if (set->cred < entry->cred) + if (cmp < 0) p = &parent->rb_left; - else if (set->cred > entry->cred) + else if (cmp > 0) p = &parent->rb_right; else goto found; @@ -2338,22 +3161,25 @@ found: nfs_access_free_entry(entry); } -void nfs_access_add_cache(struct inode *inode, struct nfs_access_entry *set) +void nfs_access_add_cache(struct inode *inode, struct nfs_access_entry *set, + const struct cred *cred) { struct nfs_access_entry *cache = kmalloc(sizeof(*cache), GFP_KERNEL); if (cache == NULL) return; RB_CLEAR_NODE(&cache->rb_node); - cache->jiffies = set->jiffies; - cache->cred = get_rpccred(set->cred); + cache->fsuid = cred->fsuid; + cache->fsgid = cred->fsgid; + cache->group_info = get_group_info(cred->group_info); cache->mask = set->mask; + cache->timestamp = ktime_get_ns(); /* The above field assignments must be visible * before this item appears on the lru. We cannot easily * use rcu_assign_pointer, so just force the memory barrier. */ smp_wmb(); - nfs_access_add_rbtree(inode, cache); + nfs_access_add_rbtree(inode, cache, cred); /* Update accounting */ smp_mb__before_atomic(); @@ -2372,15 +3198,15 @@ void nfs_access_add_cache(struct inode *inode, struct nfs_access_entry *set) } EXPORT_SYMBOL_GPL(nfs_access_add_cache); -#define NFS_MAY_READ (NFS4_ACCESS_READ) -#define NFS_MAY_WRITE (NFS4_ACCESS_MODIFY | \ - NFS4_ACCESS_EXTEND | \ - NFS4_ACCESS_DELETE) -#define NFS_FILE_MAY_WRITE (NFS4_ACCESS_MODIFY | \ - NFS4_ACCESS_EXTEND) +#define NFS_MAY_READ (NFS_ACCESS_READ) +#define NFS_MAY_WRITE (NFS_ACCESS_MODIFY | \ + NFS_ACCESS_EXTEND | \ + NFS_ACCESS_DELETE) +#define NFS_FILE_MAY_WRITE (NFS_ACCESS_MODIFY | \ + NFS_ACCESS_EXTEND) #define NFS_DIR_MAY_WRITE NFS_MAY_WRITE -#define NFS_MAY_LOOKUP (NFS4_ACCESS_LOOKUP) -#define NFS_MAY_EXECUTE (NFS4_ACCESS_EXECUTE) +#define NFS_MAY_LOOKUP (NFS_ACCESS_LOOKUP) +#define NFS_MAY_EXECUTE (NFS_ACCESS_EXECUTE) static int nfs_access_calc_mask(u32 access_result, umode_t umode) { @@ -2409,18 +3235,16 @@ void nfs_access_set_mask(struct nfs_access_entry *entry, u32 access_result) } EXPORT_SYMBOL_GPL(nfs_access_set_mask); -static int nfs_do_access(struct inode *inode, struct rpc_cred *cred, int mask) +static int nfs_do_access(struct inode *inode, const struct cred *cred, int mask) { struct nfs_access_entry cache; bool may_block = (mask & MAY_NOT_BLOCK) == 0; - int cache_mask; + int cache_mask = -1; int status; trace_nfs_access_enter(inode); - status = nfs_access_get_cached_rcu(inode, cred, &cache); - if (status != 0) - status = nfs_access_get_cached(inode, cred, &cache, may_block); + status = nfs_access_get_cached(inode, cred, &cache.mask, may_block); if (status == 0) goto out_cached; @@ -2428,27 +3252,32 @@ static int nfs_do_access(struct inode *inode, struct rpc_cred *cred, int mask) if (!may_block) goto out; - /* Be clever: ask server to check for all possible rights */ - cache.mask = NFS_MAY_LOOKUP | NFS_MAY_EXECUTE - | NFS_MAY_WRITE | NFS_MAY_READ; - cache.cred = cred; - cache.jiffies = jiffies; - status = NFS_PROTO(inode)->access(inode, &cache); + /* + * Determine which access bits we want to ask for... + */ + cache.mask = NFS_ACCESS_READ | NFS_ACCESS_MODIFY | NFS_ACCESS_EXTEND | + nfs_access_xattr_mask(NFS_SERVER(inode)); + if (S_ISDIR(inode->i_mode)) + cache.mask |= NFS_ACCESS_DELETE | NFS_ACCESS_LOOKUP; + else + cache.mask |= NFS_ACCESS_EXECUTE; + status = NFS_PROTO(inode)->access(inode, &cache, cred); if (status != 0) { if (status == -ESTALE) { - nfs_zap_caches(inode); if (!S_ISDIR(inode->i_mode)) - set_bit(NFS_INO_STALE, &NFS_I(inode)->flags); + nfs_set_inode_stale(inode); + else + nfs_zap_caches(inode); } goto out; } - nfs_access_add_cache(inode, &cache); + nfs_access_add_cache(inode, &cache, cred); out_cached: cache_mask = nfs_access_calc_mask(cache.mask, inode->i_mode); if ((mask & ~cache_mask & (MAY_READ | MAY_WRITE | MAY_EXEC)) != 0) status = -EACCES; out: - trace_nfs_access_exit(inode, status); + trace_nfs_access_exit(inode, mask, cache_mask, status); return status; } @@ -2469,7 +3298,7 @@ static int nfs_open_permission_mask(int openflags) return mask; } -int nfs_may_open(struct inode *inode, struct rpc_cred *cred, int openflags) +int nfs_may_open(struct inode *inode, const struct cred *cred, int openflags) { return nfs_do_access(inode, cred, nfs_open_permission_mask(openflags)); } @@ -2480,7 +3309,9 @@ static int nfs_execute_ok(struct inode *inode, int mask) struct nfs_server *server = NFS_SERVER(inode); int ret = 0; - if (nfs_check_cache_invalid(inode, NFS_INO_INVALID_ACCESS)) { + if (S_ISDIR(inode->i_mode)) + return 0; + if (nfs_check_cache_invalid(inode, NFS_INO_INVALID_MODE)) { if (mask & MAY_NOT_BLOCK) return -ECHILD; ret = __nfs_revalidate_inode(server, inode); @@ -2490,9 +3321,11 @@ static int nfs_execute_ok(struct inode *inode, int mask) return ret; } -int nfs_permission(struct inode *inode, int mask) +int nfs_permission(struct mnt_idmap *idmap, + struct inode *inode, + int mask) { - struct rpc_cred *cred; + const struct cred *cred = current_cred(); int res = 0; nfs_inc_stats(inode, NFSIOS_VFSACCESS); @@ -2524,23 +3357,7 @@ force_lookup: if (!NFS_PROTO(inode)->access) goto out_notsup; - /* Always try fast lookups first */ - rcu_read_lock(); - cred = rpc_lookup_cred_nonblock(); - if (!IS_ERR(cred)) - res = nfs_do_access(inode, cred, mask|MAY_NOT_BLOCK); - else - res = PTR_ERR(cred); - rcu_read_unlock(); - if (res == -ECHILD && !(mask & MAY_NOT_BLOCK)) { - /* Fast lookup failed, try the slow way */ - cred = rpc_lookup_cred(); - if (!IS_ERR(cred)) { - res = nfs_do_access(inode, cred, mask); - put_rpccred(cred); - } else - res = PTR_ERR(cred); - } + res = nfs_do_access(inode, cred, mask); out: if (!res && (mask & MAY_EXEC)) res = nfs_execute_ok(inode, mask); @@ -2552,16 +3369,10 @@ out_notsup: if (mask & MAY_NOT_BLOCK) return -ECHILD; - res = nfs_revalidate_inode(NFS_SERVER(inode), inode); + res = nfs_revalidate_inode(inode, NFS_INO_INVALID_MODE | + NFS_INO_INVALID_OTHER); if (res == 0) - res = generic_permission(inode, mask); + res = generic_permission(&nop_mnt_idmap, inode, mask); goto out; } EXPORT_SYMBOL_GPL(nfs_permission); - -/* - * Local variables: - * version-control: t - * kept-new-versions: 5 - * End: - */ |