diff options
Diffstat (limited to 'fs/afs/dir.c')
| -rw-r--r-- | fs/afs/dir.c | 2307 |
1 files changed, 1430 insertions, 877 deletions
diff --git a/fs/afs/dir.c b/fs/afs/dir.c index 8a2562e3a316..f4e9e12373ac 100644 --- a/fs/afs/dir.c +++ b/fs/afs/dir.c @@ -1,12 +1,8 @@ +// SPDX-License-Identifier: GPL-2.0-or-later /* dir.c: AFS filesystem directory handling * * Copyright (C) 2002, 2018 Red Hat, Inc. All Rights Reserved. * Written by David Howells (dhowells@redhat.com) - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public License - * as published by the Free Software Foundation; either version - * 2 of the License, or (at your option) any later version. */ #include <linux/kernel.h> @@ -16,40 +12,38 @@ #include <linux/swap.h> #include <linux/ctype.h> #include <linux/sched.h> +#include <linux/iversion.h> +#include <linux/iov_iter.h> #include <linux/task_io_accounting_ops.h> #include "internal.h" +#include "afs_fs.h" #include "xdr_fs.h" static struct dentry *afs_lookup(struct inode *dir, struct dentry *dentry, unsigned int flags); static int afs_dir_open(struct inode *inode, struct file *file); static int afs_readdir(struct file *file, struct dir_context *ctx); -static int afs_d_revalidate(struct dentry *dentry, unsigned int flags); +static int afs_d_revalidate(struct inode *dir, const struct qstr *name, + struct dentry *dentry, unsigned int flags); static int afs_d_delete(const struct dentry *dentry); -static int afs_lookup_one_filldir(struct dir_context *ctx, const char *name, int nlen, +static void afs_d_iput(struct dentry *dentry, struct inode *inode); +static bool afs_lookup_one_filldir(struct dir_context *ctx, const char *name, int nlen, loff_t fpos, u64 ino, unsigned dtype); -static int afs_lookup_filldir(struct dir_context *ctx, const char *name, int nlen, +static bool afs_lookup_filldir(struct dir_context *ctx, const char *name, int nlen, loff_t fpos, u64 ino, unsigned dtype); -static int afs_create(struct inode *dir, struct dentry *dentry, umode_t mode, - bool excl); -static int afs_mkdir(struct inode *dir, struct dentry *dentry, umode_t mode); +static int afs_create(struct mnt_idmap *idmap, struct inode *dir, + struct dentry *dentry, umode_t mode, bool excl); +static struct dentry *afs_mkdir(struct mnt_idmap *idmap, struct inode *dir, + struct dentry *dentry, umode_t mode); static int afs_rmdir(struct inode *dir, struct dentry *dentry); static int afs_unlink(struct inode *dir, struct dentry *dentry); static int afs_link(struct dentry *from, struct inode *dir, struct dentry *dentry); -static int afs_symlink(struct inode *dir, struct dentry *dentry, - const char *content); -static int afs_rename(struct inode *old_dir, struct dentry *old_dentry, - struct inode *new_dir, struct dentry *new_dentry, - unsigned int flags); -static int afs_dir_releasepage(struct page *page, gfp_t gfp_flags); -static void afs_dir_invalidatepage(struct page *page, unsigned int offset, - unsigned int length); - -static int afs_dir_set_page_dirty(struct page *page) -{ - BUG(); /* This should never happen. */ -} +static int afs_symlink(struct mnt_idmap *idmap, struct inode *dir, + struct dentry *dentry, const char *content); +static int afs_rename(struct mnt_idmap *idmap, struct inode *old_dir, + struct dentry *old_dentry, struct inode *new_dir, + struct dentry *new_dentry, unsigned int flags); const struct file_operations afs_dir_file_operations = { .open = afs_dir_open, @@ -71,13 +65,10 @@ const struct inode_operations afs_dir_inode_operations = { .permission = afs_permission, .getattr = afs_getattr, .setattr = afs_setattr, - .listxattr = afs_listxattr, }; const struct address_space_operations afs_dir_aops = { - .set_page_dirty = afs_dir_set_page_dirty, - .releasepage = afs_dir_releasepage, - .invalidatepage = afs_dir_invalidatepage, + .writepages = afs_single_writepages, }; const struct dentry_operations afs_fs_dentry_operations = { @@ -85,6 +76,7 @@ const struct dentry_operations afs_fs_dentry_operations = { .d_delete = afs_d_delete, .d_release = afs_d_release, .d_automount = afs_d_automount, + .d_iput = afs_d_iput, }; struct afs_lookup_one_cookie { @@ -97,66 +89,124 @@ struct afs_lookup_one_cookie { struct afs_lookup_cookie { struct dir_context ctx; struct qstr name; - bool found; - bool one_only; unsigned short nr_fids; - struct afs_file_status *statuses; - struct afs_callback *callbacks; struct afs_fid fids[50]; }; +static void afs_dir_unuse_cookie(struct afs_vnode *dvnode, int ret) +{ + if (ret == 0) { + struct afs_vnode_cache_aux aux; + loff_t i_size = i_size_read(&dvnode->netfs.inode); + + afs_set_cache_aux(dvnode, &aux); + fscache_unuse_cookie(afs_vnode_cache(dvnode), &aux, &i_size); + } else { + fscache_unuse_cookie(afs_vnode_cache(dvnode), NULL, NULL); + } +} + /* - * check that a directory page is valid + * Iterate through a kmapped directory segment, dumping a summary of + * the contents. */ -static bool afs_dir_check_page(struct afs_vnode *dvnode, struct page *page, - loff_t i_size) +static size_t afs_dir_dump_step(void *iter_base, size_t progress, size_t len, + void *priv, void *priv2) { - struct afs_xdr_dir_page *dbuf; - loff_t latter, off; - int tmp, qty; + do { + union afs_xdr_dir_block *block = iter_base; - /* Determine how many magic numbers there should be in this page, but - * we must take care because the directory may change size under us. - */ - off = page_offset(page); - if (i_size <= off) - goto checked; + pr_warn("[%05zx] %32phN\n", progress, block); + iter_base += AFS_DIR_BLOCK_SIZE; + progress += AFS_DIR_BLOCK_SIZE; + len -= AFS_DIR_BLOCK_SIZE; + } while (len > 0); - latter = i_size - off; - if (latter >= PAGE_SIZE) - qty = PAGE_SIZE; - else - qty = latter; - qty /= sizeof(union afs_xdr_dir_block); - - /* check them */ - dbuf = kmap(page); - for (tmp = 0; tmp < qty; tmp++) { - if (dbuf->blocks[tmp].hdr.magic != AFS_DIR_MAGIC) { - printk("kAFS: %s(%lx): bad magic %d/%d is %04hx\n", - __func__, dvnode->vfs_inode.i_ino, tmp, qty, - ntohs(dbuf->blocks[tmp].hdr.magic)); - trace_afs_dir_check_failed(dvnode, off, i_size); - kunmap(page); - trace_afs_file_error(dvnode, -EIO, afs_file_error_dir_bad_magic); - goto error; - } + return len; +} - /* Make sure each block is NUL terminated so we can reasonably - * use string functions on it. The filenames in the page - * *should* be NUL-terminated anyway. - */ - ((u8 *)&dbuf->blocks[tmp])[AFS_DIR_BLOCK_SIZE - 1] = 0; - } +/* + * Dump the contents of a directory. + */ +static void afs_dir_dump(struct afs_vnode *dvnode) +{ + struct iov_iter iter; + unsigned long long i_size = i_size_read(&dvnode->netfs.inode); + + pr_warn("DIR %llx:%llx is=%llx\n", + dvnode->fid.vid, dvnode->fid.vnode, i_size); - kunmap(page); + iov_iter_folio_queue(&iter, ITER_SOURCE, dvnode->directory, 0, 0, i_size); + iterate_folioq(&iter, iov_iter_count(&iter), NULL, NULL, + afs_dir_dump_step); +} + +/* + * check that a directory folio is valid + */ +static bool afs_dir_check_block(struct afs_vnode *dvnode, size_t progress, + union afs_xdr_dir_block *block) +{ + if (block->hdr.magic != AFS_DIR_MAGIC) { + pr_warn("%s(%lx): [%zx] bad magic %04x\n", + __func__, dvnode->netfs.inode.i_ino, + progress, ntohs(block->hdr.magic)); + trace_afs_dir_check_failed(dvnode, progress); + trace_afs_file_error(dvnode, -EIO, afs_file_error_dir_bad_magic); + return false; + } -checked: + /* Make sure each block is NUL terminated so we can reasonably + * use string functions on it. The filenames in the folio + * *should* be NUL-terminated anyway. + */ + ((u8 *)block)[AFS_DIR_BLOCK_SIZE - 1] = 0; afs_stat_v(dvnode, n_read_dir); return true; +} -error: - return false; +/* + * Iterate through a kmapped directory segment, checking the content. + */ +static size_t afs_dir_check_step(void *iter_base, size_t progress, size_t len, + void *priv, void *priv2) +{ + struct afs_vnode *dvnode = priv; + + if (WARN_ON_ONCE(progress % AFS_DIR_BLOCK_SIZE || + len % AFS_DIR_BLOCK_SIZE)) + return len; + + do { + if (!afs_dir_check_block(dvnode, progress, iter_base)) + break; + iter_base += AFS_DIR_BLOCK_SIZE; + len -= AFS_DIR_BLOCK_SIZE; + } while (len > 0); + + return len; +} + +/* + * Check all the blocks in a directory. + */ +static int afs_dir_check(struct afs_vnode *dvnode) +{ + struct iov_iter iter; + unsigned long long i_size = i_size_read(&dvnode->netfs.inode); + size_t checked = 0; + + if (unlikely(!i_size)) + return 0; + + iov_iter_folio_queue(&iter, ITER_SOURCE, dvnode->directory, 0, 0, i_size); + checked = iterate_folioq(&iter, iov_iter_count(&iter), dvnode, NULL, + afs_dir_check_step); + if (checked != i_size) { + afs_dir_dump(dvnode); + return -EIO; + } + return 0; } /* @@ -176,143 +226,140 @@ static int afs_dir_open(struct inode *inode, struct file *file) } /* - * Read the directory into the pagecache in one go, scrubbing the previous - * contents. The list of pages is returned, pinning them so that they don't - * get reclaimed during the iteration. + * Read a file in a single download. */ -static struct afs_read *afs_read_dir(struct afs_vnode *dvnode, struct key *key) - __acquires(&dvnode->validate_lock) +static ssize_t afs_do_read_single(struct afs_vnode *dvnode, struct file *file) { - struct afs_read *req; + struct iov_iter iter; + ssize_t ret; loff_t i_size; - int nr_pages, nr_inline, i, n; - int ret = -ENOMEM; - -retry: - i_size = i_size_read(&dvnode->vfs_inode); - if (i_size < 2048) - return ERR_PTR(afs_bad(dvnode, afs_file_error_dir_small)); - if (i_size > 2048 * 1024) { - trace_afs_file_error(dvnode, -EFBIG, afs_file_error_dir_big); - return ERR_PTR(-EFBIG); + bool is_dir = (S_ISDIR(dvnode->netfs.inode.i_mode) && + !test_bit(AFS_VNODE_MOUNTPOINT, &dvnode->flags)); + + i_size = i_size_read(&dvnode->netfs.inode); + if (is_dir) { + if (i_size < AFS_DIR_BLOCK_SIZE) + return afs_bad(dvnode, afs_file_error_dir_small); + if (i_size > AFS_DIR_BLOCK_SIZE * 1024) { + trace_afs_file_error(dvnode, -EFBIG, afs_file_error_dir_big); + return -EFBIG; + } + } else { + if (i_size > AFSPATHMAX) { + trace_afs_file_error(dvnode, -EFBIG, afs_file_error_dir_big); + return -EFBIG; + } } - _enter("%llu", i_size); + /* Expand the storage. TODO: Shrink the storage too. */ + if (dvnode->directory_size < i_size) { + size_t cur_size = dvnode->directory_size; - /* Get a request record to hold the page list. We want to hold it - * inline if we can, but we don't want to make an order 1 allocation. - */ - nr_pages = (i_size + PAGE_SIZE - 1) / PAGE_SIZE; - nr_inline = nr_pages; - if (nr_inline > (PAGE_SIZE - sizeof(*req)) / sizeof(struct page *)) - nr_inline = 0; - - req = kzalloc(sizeof(*req) + sizeof(struct page *) * nr_inline, - GFP_KERNEL); - if (!req) - return ERR_PTR(-ENOMEM); - - refcount_set(&req->usage, 1); - req->nr_pages = nr_pages; - req->actual_len = i_size; /* May change */ - req->len = nr_pages * PAGE_SIZE; /* We can ask for more than there is */ - req->data_version = dvnode->status.data_version; /* May change */ - if (nr_inline > 0) { - req->pages = req->array; - } else { - req->pages = kcalloc(nr_pages, sizeof(struct page *), - GFP_KERNEL); - if (!req->pages) - goto error; + ret = netfs_alloc_folioq_buffer(NULL, + &dvnode->directory, &cur_size, i_size, + mapping_gfp_mask(dvnode->netfs.inode.i_mapping)); + dvnode->directory_size = cur_size; + if (ret < 0) + return ret; } - /* Get a list of all the pages that hold or will hold the directory - * content. We need to fill in any gaps that we might find where the - * memory reclaimer has been at work. If there are any gaps, we will - * need to reread the entire directory contents. + iov_iter_folio_queue(&iter, ITER_DEST, dvnode->directory, 0, 0, dvnode->directory_size); + + /* AFS requires us to perform the read of a directory synchronously as + * a single unit to avoid issues with the directory contents being + * changed between reads. */ - i = 0; - do { - n = find_get_pages_contig(dvnode->vfs_inode.i_mapping, i, - req->nr_pages - i, - req->pages + i); - _debug("find %u at %u/%u", n, i, req->nr_pages); - if (n == 0) { - gfp_t gfp = dvnode->vfs_inode.i_mapping->gfp_mask; - - if (test_and_clear_bit(AFS_VNODE_DIR_VALID, &dvnode->flags)) - afs_stat_v(dvnode, n_inval); - - ret = -ENOMEM; - req->pages[i] = __page_cache_alloc(gfp); - if (!req->pages[i]) - goto error; - ret = add_to_page_cache_lru(req->pages[i], - dvnode->vfs_inode.i_mapping, - i, gfp); - if (ret < 0) - goto error; - - set_page_private(req->pages[i], 1); - SetPagePrivate(req->pages[i]); - unlock_page(req->pages[i]); - i++; - } else { - i += n; + ret = netfs_read_single(&dvnode->netfs.inode, file, &iter); + if (ret >= 0) { + i_size = i_size_read(&dvnode->netfs.inode); + if (i_size > ret) { + /* The content has grown, so we need to expand the + * buffer. + */ + ret = -ESTALE; + } else if (is_dir) { + int ret2 = afs_dir_check(dvnode); + + if (ret2 < 0) + ret = ret2; + } else if (i_size < folioq_folio_size(dvnode->directory, 0)) { + /* NUL-terminate a symlink. */ + char *symlink = kmap_local_folio(folioq_folio(dvnode->directory, 0), 0); + + symlink[i_size] = 0; + kunmap_local(symlink); } - } while (i < req->nr_pages); + } + + return ret; +} + +ssize_t afs_read_single(struct afs_vnode *dvnode, struct file *file) +{ + ssize_t ret; + + fscache_use_cookie(afs_vnode_cache(dvnode), false); + ret = afs_do_read_single(dvnode, file); + fscache_unuse_cookie(afs_vnode_cache(dvnode), NULL, NULL); + return ret; +} + +/* + * Read the directory into a folio_queue buffer in one go, scrubbing the + * previous contents. We return -ESTALE if the caller needs to call us again. + */ +ssize_t afs_read_dir(struct afs_vnode *dvnode, struct file *file) + __acquires(&dvnode->validate_lock) +{ + ssize_t ret; + loff_t i_size; + + i_size = i_size_read(&dvnode->netfs.inode); - /* If we're going to reload, we need to lock all the pages to prevent - * races. - */ ret = -ERESTARTSYS; if (down_read_killable(&dvnode->validate_lock) < 0) goto error; - if (test_bit(AFS_VNODE_DIR_VALID, &dvnode->flags)) - goto success; + /* We only need to reread the data if it became invalid - or if we + * haven't read it yet. + */ + if (test_bit(AFS_VNODE_DIR_VALID, &dvnode->flags) && + test_bit(AFS_VNODE_DIR_READ, &dvnode->flags)) { + ret = i_size; + goto valid; + } up_read(&dvnode->validate_lock); if (down_write_killable(&dvnode->validate_lock) < 0) goto error; - if (!test_bit(AFS_VNODE_DIR_VALID, &dvnode->flags)) { - ret = afs_fetch_data(dvnode, key, req); + if (!test_bit(AFS_VNODE_DIR_VALID, &dvnode->flags)) + afs_invalidate_cache(dvnode, 0); + + if (!test_bit(AFS_VNODE_DIR_VALID, &dvnode->flags) || + !test_bit(AFS_VNODE_DIR_READ, &dvnode->flags)) { + trace_afs_reload_dir(dvnode); + ret = afs_read_single(dvnode, file); if (ret < 0) goto error_unlock; - task_io_account_read(PAGE_SIZE * req->nr_pages); - - if (req->len < req->file_size) - goto content_has_grown; - - /* Validate the data we just read. */ - ret = -EIO; - for (i = 0; i < req->nr_pages; i++) - if (!afs_dir_check_page(dvnode, req->pages[i], - req->actual_len)) - goto error_unlock; - // TODO: Trim excess pages set_bit(AFS_VNODE_DIR_VALID, &dvnode->flags); + set_bit(AFS_VNODE_DIR_READ, &dvnode->flags); + } else { + ret = i_size; } downgrade_write(&dvnode->validate_lock); -success: - return req; +valid: + return ret; error_unlock: up_write(&dvnode->validate_lock); error: - afs_put_read(req); - _leave(" = %d", ret); - return ERR_PTR(ret); - -content_has_grown: - up_write(&dvnode->validate_lock); - afs_put_read(req); - goto retry; + _leave(" = %zd", ret); + return ret; } /* @@ -320,75 +367,71 @@ content_has_grown: */ static int afs_dir_iterate_block(struct afs_vnode *dvnode, struct dir_context *ctx, - union afs_xdr_dir_block *block, - unsigned blkoff) + union afs_xdr_dir_block *block) { union afs_xdr_dirent *dire; - unsigned offset, next, curr; + unsigned int blknum, base, hdr, pos, next, nr_slots; size_t nlen; int tmp; - _enter("%u,%x,%p,,",(unsigned)ctx->pos,blkoff,block); + blknum = ctx->pos / AFS_DIR_BLOCK_SIZE; + base = blknum * AFS_DIR_SLOTS_PER_BLOCK; + hdr = (blknum == 0 ? AFS_DIR_RESV_BLOCKS0 : AFS_DIR_RESV_BLOCKS); + pos = DIV_ROUND_UP(ctx->pos, AFS_DIR_DIRENT_SIZE) - base; - curr = (ctx->pos - blkoff) / sizeof(union afs_xdr_dirent); + _enter("%llx,%x", ctx->pos, blknum); /* walk through the block, an entry at a time */ - for (offset = (blkoff == 0 ? AFS_DIR_RESV_BLOCKS0 : AFS_DIR_RESV_BLOCKS); - offset < AFS_DIR_SLOTS_PER_BLOCK; - offset = next - ) { - next = offset + 1; - + for (unsigned int slot = hdr; slot < AFS_DIR_SLOTS_PER_BLOCK; slot = next) { /* skip entries marked unused in the bitmap */ - if (!(block->hdr.bitmap[offset / 8] & - (1 << (offset % 8)))) { - _debug("ENT[%zu.%u]: unused", - blkoff / sizeof(union afs_xdr_dir_block), offset); - if (offset >= curr) - ctx->pos = blkoff + - next * sizeof(union afs_xdr_dirent); + if (!(block->hdr.bitmap[slot / 8] & + (1 << (slot % 8)))) { + _debug("ENT[%x]: Unused", base + slot); + next = slot + 1; + if (next >= pos) + ctx->pos = (base + next) * sizeof(union afs_xdr_dirent); continue; } /* got a valid entry */ - dire = &block->dirents[offset]; + dire = &block->dirents[slot]; nlen = strnlen(dire->u.name, - sizeof(*block) - - offset * sizeof(union afs_xdr_dirent)); + (unsigned long)(block + 1) - (unsigned long)dire->u.name - 1); + if (nlen > AFSNAMEMAX - 1) { + _debug("ENT[%x]: Name too long (len %zx)", + base + slot, nlen); + return afs_bad(dvnode, afs_file_error_dir_name_too_long); + } - _debug("ENT[%zu.%u]: %s %zu \"%s\"", - blkoff / sizeof(union afs_xdr_dir_block), offset, - (offset < curr ? "skip" : "fill"), + _debug("ENT[%x]: %s %zx \"%s\"", + base + slot, (slot < pos ? "skip" : "fill"), nlen, dire->u.name); - /* work out where the next possible entry is */ - for (tmp = nlen; tmp > 15; tmp -= sizeof(union afs_xdr_dirent)) { - if (next >= AFS_DIR_SLOTS_PER_BLOCK) { - _debug("ENT[%zu.%u]:" - " %u travelled beyond end dir block" - " (len %u/%zu)", - blkoff / sizeof(union afs_xdr_dir_block), - offset, next, tmp, nlen); - return afs_bad(dvnode, afs_file_error_dir_over_end); - } - if (!(block->hdr.bitmap[next / 8] & - (1 << (next % 8)))) { - _debug("ENT[%zu.%u]:" - " %u unmarked extension (len %u/%zu)", - blkoff / sizeof(union afs_xdr_dir_block), - offset, next, tmp, nlen); + nr_slots = afs_dir_calc_slots(nlen); + next = slot + nr_slots; + if (next > AFS_DIR_SLOTS_PER_BLOCK) { + _debug("ENT[%x]: extends beyond end dir block (len %zx)", + base + slot, nlen); + return afs_bad(dvnode, afs_file_error_dir_over_end); + } + + /* Check that the name-extension dirents are all allocated */ + for (tmp = 1; tmp < nr_slots; tmp++) { + unsigned int xslot = slot + tmp; + + if (!(block->hdr.bitmap[xslot / 8] & (1 << (xslot % 8)))) { + _debug("ENT[%x]: Unmarked extension (%x/%x)", + base + slot, tmp, nr_slots); return afs_bad(dvnode, afs_file_error_dir_unmarked_ext); } - - _debug("ENT[%zu.%u]: ext %u/%zu", - blkoff / sizeof(union afs_xdr_dir_block), - next, tmp, nlen); - next++; } /* skip if starts before the current position */ - if (offset < curr) + if (slot < pos) { + if (next > pos) + ctx->pos = (base + next) * sizeof(union afs_xdr_dirent); continue; + } /* found the next entry */ if (!dir_emit(ctx, dire->u.name, nlen, @@ -400,82 +443,110 @@ static int afs_dir_iterate_block(struct afs_vnode *dvnode, return 0; } - ctx->pos = blkoff + next * sizeof(union afs_xdr_dirent); + ctx->pos = (base + next) * sizeof(union afs_xdr_dirent); } _leave(" = 1 [more]"); return 1; } +struct afs_dir_iteration_ctx { + struct dir_context *dir_ctx; + int error; +}; + /* - * iterate through the data blob that lists the contents of an AFS directory + * Iterate through a kmapped directory segment. */ -static int afs_dir_iterate(struct inode *dir, struct dir_context *ctx, - struct key *key) +static size_t afs_dir_iterate_step(void *iter_base, size_t progress, size_t len, + void *priv, void *priv2) { - struct afs_vnode *dvnode = AFS_FS_I(dir); - struct afs_xdr_dir_page *dbuf; - union afs_xdr_dir_block *dblock; - struct afs_read *req; - struct page *page; - unsigned blkoff, limit; + struct afs_dir_iteration_ctx *ctx = priv2; + struct afs_vnode *dvnode = priv; int ret; - _enter("{%lu},%u,,", dir->i_ino, (unsigned)ctx->pos); - - if (test_bit(AFS_VNODE_DELETED, &AFS_FS_I(dir)->flags)) { - _leave(" = -ESTALE"); - return -ESTALE; + if (WARN_ON_ONCE(progress % AFS_DIR_BLOCK_SIZE || + len % AFS_DIR_BLOCK_SIZE)) { + pr_err("Mis-iteration prog=%zx len=%zx\n", + progress % AFS_DIR_BLOCK_SIZE, + len % AFS_DIR_BLOCK_SIZE); + return len; } - req = afs_read_dir(dvnode, key); - if (IS_ERR(req)) - return PTR_ERR(req); + do { + ret = afs_dir_iterate_block(dvnode, ctx->dir_ctx, iter_base); + if (ret != 1) + break; - /* round the file position up to the next entry boundary */ - ctx->pos += sizeof(union afs_xdr_dirent) - 1; - ctx->pos &= ~(sizeof(union afs_xdr_dirent) - 1); + ctx->dir_ctx->pos = round_up(ctx->dir_ctx->pos, AFS_DIR_BLOCK_SIZE); + iter_base += AFS_DIR_BLOCK_SIZE; + len -= AFS_DIR_BLOCK_SIZE; + } while (len > 0); - /* walk through the blocks in sequence */ - ret = 0; - while (ctx->pos < req->actual_len) { - blkoff = ctx->pos & ~(sizeof(union afs_xdr_dir_block) - 1); + return len; +} - /* Fetch the appropriate page from the directory and re-add it - * to the LRU. - */ - page = req->pages[blkoff / PAGE_SIZE]; - if (!page) { - ret = afs_bad(dvnode, afs_file_error_dir_missing_page); - break; - } - mark_page_accessed(page); +/* + * Iterate through the directory folios. + */ +static int afs_dir_iterate_contents(struct inode *dir, struct dir_context *dir_ctx) +{ + struct afs_dir_iteration_ctx ctx = { .dir_ctx = dir_ctx }; + struct afs_vnode *dvnode = AFS_FS_I(dir); + struct iov_iter iter; + unsigned long long i_size = i_size_read(dir); - limit = blkoff & ~(PAGE_SIZE - 1); + /* Round the file position up to the next entry boundary */ + dir_ctx->pos = round_up(dir_ctx->pos, sizeof(union afs_xdr_dirent)); - dbuf = kmap(page); + if (i_size <= 0 || dir_ctx->pos >= i_size) + return 0; - /* deal with the individual blocks stashed on this page */ - do { - dblock = &dbuf->blocks[(blkoff % PAGE_SIZE) / - sizeof(union afs_xdr_dir_block)]; - ret = afs_dir_iterate_block(dvnode, ctx, dblock, blkoff); - if (ret != 1) { - kunmap(page); - goto out; - } + iov_iter_folio_queue(&iter, ITER_SOURCE, dvnode->directory, 0, 0, i_size); + iov_iter_advance(&iter, round_down(dir_ctx->pos, AFS_DIR_BLOCK_SIZE)); - blkoff += sizeof(union afs_xdr_dir_block); + iterate_folioq(&iter, iov_iter_count(&iter), dvnode, &ctx, + afs_dir_iterate_step); - } while (ctx->pos < dir->i_size && blkoff < limit); + if (ctx.error == -ESTALE) + afs_invalidate_dir(dvnode, afs_dir_invalid_iter_stale); + return ctx.error; +} - kunmap(page); - ret = 0; - } +/* + * iterate through the data blob that lists the contents of an AFS directory + */ +static int afs_dir_iterate(struct inode *dir, struct dir_context *ctx, + struct file *file, afs_dataversion_t *_dir_version) +{ + struct afs_vnode *dvnode = AFS_FS_I(dir); + int retry_limit = 100; + int ret; + + _enter("{%lu},%llx,,", dir->i_ino, ctx->pos); + + do { + if (--retry_limit < 0) { + pr_warn("afs_read_dir(): Too many retries\n"); + ret = -ESTALE; + break; + } + ret = afs_read_dir(dvnode, file); + if (ret < 0) { + if (ret != -ESTALE) + break; + if (test_bit(AFS_VNODE_DELETED, &AFS_FS_I(dir)->flags)) { + ret = -ESTALE; + break; + } + continue; + } + *_dir_version = inode_peek_iversion_raw(dir); + + ret = afs_dir_iterate_contents(dir, ctx); + up_read(&dvnode->validate_lock); + } while (ret == -ESTALE); -out: - up_read(&dvnode->validate_lock); - afs_put_read(req); _leave(" = %d", ret); return ret; } @@ -485,7 +556,9 @@ out: */ static int afs_readdir(struct file *file, struct dir_context *ctx) { - return afs_dir_iterate(file_inode(file), ctx, afs_file_key(file)); + afs_dataversion_t dir_version; + + return afs_dir_iterate(file_inode(file), ctx, file, &dir_version); } /* @@ -493,7 +566,7 @@ static int afs_readdir(struct file *file, struct dir_context *ctx) * - if afs_dir_iterate_block() spots this function, it'll pass the FID * uniquifier through dtype */ -static int afs_lookup_one_filldir(struct dir_context *ctx, const char *name, +static bool afs_lookup_one_filldir(struct dir_context *ctx, const char *name, int nlen, loff_t fpos, u64 ino, unsigned dtype) { struct afs_lookup_one_cookie *cookie = @@ -509,43 +582,43 @@ static int afs_lookup_one_filldir(struct dir_context *ctx, const char *name, if (cookie->name.len != nlen || memcmp(cookie->name.name, name, nlen) != 0) { - _leave(" = 0 [no]"); - return 0; + _leave(" = true [keep looking]"); + return true; } cookie->fid.vnode = ino; cookie->fid.unique = dtype; cookie->found = 1; - _leave(" = -1 [found]"); - return -1; + _leave(" = false [found]"); + return false; } /* * Do a lookup of a single name in a directory * - just returns the FID the dentry name maps to if found */ -static int afs_do_lookup_one(struct inode *dir, struct dentry *dentry, - struct afs_fid *fid, struct key *key) +static int afs_do_lookup_one(struct inode *dir, const struct qstr *name, + struct afs_fid *fid, + afs_dataversion_t *_dir_version) { struct afs_super_info *as = dir->i_sb->s_fs_info; struct afs_lookup_one_cookie cookie = { .ctx.actor = afs_lookup_one_filldir, - .name = dentry->d_name, + .name = *name, .fid.vid = as->volume->vid }; int ret; - _enter("{%lu},%p{%pd},", dir->i_ino, dentry, dentry); + _enter("{%lu},{%.*s},", dir->i_ino, name->len, name->name); /* search the directory */ - ret = afs_dir_iterate(dir, &cookie.ctx, key); + ret = afs_dir_iterate(dir, &cookie.ctx, NULL, _dir_version); if (ret < 0) { _leave(" = %d [iter]", ret); return ret; } - ret = -ENOENT; if (!cookie.found) { _leave(" = -ENOENT [not found]"); return -ENOENT; @@ -561,12 +634,11 @@ static int afs_do_lookup_one(struct inode *dir, struct dentry *dentry, * - if afs_dir_iterate_block() spots this function, it'll pass the FID * uniquifier through dtype */ -static int afs_lookup_filldir(struct dir_context *ctx, const char *name, +static bool afs_lookup_filldir(struct dir_context *ctx, const char *name, int nlen, loff_t fpos, u64 ino, unsigned dtype) { struct afs_lookup_cookie *cookie = container_of(ctx, struct afs_lookup_cookie, ctx); - int ret; _enter("{%s,%u},%s,%u,,%llu,%u", cookie->name.name, cookie->name.len, name, nlen, @@ -576,23 +648,121 @@ static int afs_lookup_filldir(struct dir_context *ctx, const char *name, BUILD_BUG_ON(sizeof(union afs_xdr_dir_block) != 2048); BUILD_BUG_ON(sizeof(union afs_xdr_dirent) != 32); - if (cookie->found) { - if (cookie->nr_fids < 50) { - cookie->fids[cookie->nr_fids].vnode = ino; - cookie->fids[cookie->nr_fids].unique = dtype; - cookie->nr_fids++; + if (cookie->nr_fids < 50) { + cookie->fids[cookie->nr_fids].vnode = ino; + cookie->fids[cookie->nr_fids].unique = dtype; + cookie->nr_fids++; + } + + return cookie->nr_fids < 50; +} + +/* + * Deal with the result of a successful lookup operation. Turn all the files + * into inodes and save the first one - which is the one we actually want. + */ +static void afs_do_lookup_success(struct afs_operation *op) +{ + struct afs_vnode_param *vp; + struct afs_vnode *vnode; + struct inode *inode; + u32 abort_code; + int i; + + _enter(""); + + for (i = 0; i < op->nr_files; i++) { + switch (i) { + case 0: + vp = &op->file[0]; + abort_code = vp->scb.status.abort_code; + if (abort_code != 0) { + op->call_abort_code = abort_code; + afs_op_set_error(op, afs_abort_to_error(abort_code)); + op->cumul_error.abort_code = abort_code; + } + break; + + case 1: + vp = &op->file[1]; + break; + + default: + vp = &op->more_files[i - 2]; + break; + } + + if (vp->scb.status.abort_code) + trace_afs_bulkstat_error(op, &vp->fid, i, vp->scb.status.abort_code); + if (!vp->scb.have_status && !vp->scb.have_error) + continue; + + _debug("do [%u]", i); + if (vp->vnode) { + if (!test_bit(AFS_VNODE_UNSET, &vp->vnode->flags)) + afs_vnode_commit_status(op, vp); + } else if (vp->scb.status.abort_code == 0) { + inode = afs_iget(op, vp); + if (!IS_ERR(inode)) { + vnode = AFS_FS_I(inode); + afs_cache_permit(vnode, op->key, + 0 /* Assume vnode->cb_break is 0 */ + + op->cb_v_break, + &vp->scb); + vp->vnode = vnode; + vp->put_vnode = true; + } + } else { + _debug("- abort %d %llx:%llx.%x", + vp->scb.status.abort_code, + vp->fid.vid, vp->fid.vnode, vp->fid.unique); } - } else if (cookie->name.len == nlen && - memcmp(cookie->name.name, name, nlen) == 0) { - cookie->fids[0].vnode = ino; - cookie->fids[0].unique = dtype; - cookie->found = 1; - if (cookie->one_only) - return -1; } - ret = cookie->nr_fids >= 50 ? -1 : 0; - _leave(" = %d", ret); + _leave(""); +} + +static const struct afs_operation_ops afs_inline_bulk_status_operation = { + .issue_afs_rpc = afs_fs_inline_bulk_status, + .issue_yfs_rpc = yfs_fs_inline_bulk_status, + .success = afs_do_lookup_success, +}; + +static const struct afs_operation_ops afs_lookup_fetch_status_operation = { + .issue_afs_rpc = afs_fs_fetch_status, + .issue_yfs_rpc = yfs_fs_fetch_status, + .success = afs_do_lookup_success, + .aborted = afs_check_for_remote_deletion, +}; + +/* + * See if we know that the server we expect to use doesn't support + * FS.InlineBulkStatus. + */ +static bool afs_server_supports_ibulk(struct afs_vnode *dvnode) +{ + struct afs_server_list *slist; + struct afs_volume *volume = dvnode->volume; + struct afs_server *server; + bool ret = true; + int i; + + if (!test_bit(AFS_VOLUME_MAYBE_NO_IBULK, &volume->flags)) + return true; + + rcu_read_lock(); + slist = rcu_dereference(volume->servers); + + for (i = 0; i < slist->nr_servers; i++) { + server = slist->servers[i].server; + if (server == dvnode->cb_server) { + if (test_bit(AFS_SERVER_FL_NO_IBULK, &server->flags)) + ret = false; + break; + } + } + + rcu_read_unlock(); return ret; } @@ -601,17 +771,17 @@ static int afs_lookup_filldir(struct dir_context *ctx, const char *name, * files in one go and create inodes for them. The inode of the file we were * asked for is returned. */ -static struct inode *afs_do_lookup(struct inode *dir, struct dentry *dentry, - struct key *key) +static struct inode *afs_do_lookup(struct inode *dir, struct dentry *dentry) { struct afs_lookup_cookie *cookie; - struct afs_cb_interest *cbi = NULL; - struct afs_super_info *as = dir->i_sb->s_fs_info; - struct afs_iget_data data; - struct afs_fs_cursor fc; - struct afs_vnode *dvnode = AFS_FS_I(dir); - struct inode *inode = NULL; - int ret, i; + struct afs_vnode_param *vp; + struct afs_operation *op; + struct afs_vnode *dvnode = AFS_FS_I(dir), *vnode; + struct inode *inode = NULL, *ti; + afs_dataversion_t data_version = READ_ONCE(dvnode->status.data_version); + bool supports_ibulk, isnew; + long ret; + int i; _enter("{%lu},%p{%pd},", dir->i_ino, dentry, dentry); @@ -619,154 +789,127 @@ static struct inode *afs_do_lookup(struct inode *dir, struct dentry *dentry, if (!cookie) return ERR_PTR(-ENOMEM); + for (i = 0; i < ARRAY_SIZE(cookie->fids); i++) + cookie->fids[i].vid = dvnode->fid.vid; cookie->ctx.actor = afs_lookup_filldir; cookie->name = dentry->d_name; - cookie->nr_fids = 1; /* slot 0 is saved for the fid we actually want */ - - read_seqlock_excl(&dvnode->cb_lock); - if (dvnode->cb_interest && - dvnode->cb_interest->server && - test_bit(AFS_SERVER_FL_NO_IBULK, &dvnode->cb_interest->server->flags)) - cookie->one_only = true; - read_sequnlock_excl(&dvnode->cb_lock); - - for (i = 0; i < 50; i++) - cookie->fids[i].vid = as->volume->vid; + cookie->nr_fids = 2; /* slot 1 is saved for the fid we actually want + * and slot 0 for the directory */ - /* search the directory */ - ret = afs_dir_iterate(dir, &cookie->ctx, key); - if (ret < 0) { - inode = ERR_PTR(ret); + /* Search the directory for the named entry using the hash table... */ + ret = afs_dir_search(dvnode, &dentry->d_name, &cookie->fids[1], &data_version); + if (ret < 0) goto out; + + supports_ibulk = afs_server_supports_ibulk(dvnode); + if (supports_ibulk) { + /* ...then scan linearly from that point for entries to lookup-ahead. */ + cookie->ctx.pos = (ret + 1) * AFS_DIR_DIRENT_SIZE; + afs_dir_iterate(dir, &cookie->ctx, NULL, &data_version); } - inode = ERR_PTR(-ENOENT); - if (!cookie->found) - goto out; + dentry->d_fsdata = (void *)(unsigned long)data_version; /* Check to see if we already have an inode for the primary fid. */ - data.volume = dvnode->volume; - data.fid = cookie->fids[0]; - inode = ilookup5(dir->i_sb, cookie->fids[0].vnode, afs_iget5_test, &data); + inode = ilookup5(dir->i_sb, cookie->fids[1].vnode, + afs_ilookup5_test_by_fid, &cookie->fids[1]); if (inode) + goto out; /* We do */ + + /* Okay, we didn't find it. We need to query the server - and whilst + * we're doing that, we're going to attempt to look up a bunch of other + * vnodes also. + */ + op = afs_alloc_operation(NULL, dvnode->volume); + if (IS_ERR(op)) { + ret = PTR_ERR(op); goto out; + } + + afs_op_set_vnode(op, 0, dvnode); + afs_op_set_fid(op, 1, &cookie->fids[1]); + + op->nr_files = cookie->nr_fids; + _debug("nr_files %u", op->nr_files); /* Need space for examining all the selected files */ - inode = ERR_PTR(-ENOMEM); - cookie->statuses = kcalloc(cookie->nr_fids, sizeof(struct afs_file_status), - GFP_KERNEL); - if (!cookie->statuses) - goto out; + if (op->nr_files > 2) { + op->more_files = kvcalloc(op->nr_files - 2, + sizeof(struct afs_vnode_param), + GFP_KERNEL); + if (!op->more_files) { + afs_op_nomem(op); + goto out_op; + } - cookie->callbacks = kcalloc(cookie->nr_fids, sizeof(struct afs_callback), - GFP_KERNEL); - if (!cookie->callbacks) - goto out_s; + for (i = 2; i < op->nr_files; i++) { + vp = &op->more_files[i - 2]; + vp->fid = cookie->fids[i]; + + /* Find any inodes that already exist and get their + * callback counters. + */ + ti = ilookup5_nowait(dir->i_sb, vp->fid.vnode, + afs_ilookup5_test_by_fid, &vp->fid, &isnew); + if (!IS_ERR_OR_NULL(ti)) { + vnode = AFS_FS_I(ti); + vp->dv_before = vnode->status.data_version; + vp->cb_break_before = afs_calc_vnode_cb_break(vnode); + vp->vnode = vnode; + vp->put_vnode = true; + vp->speculative = true; /* vnode not locked */ + } + } + } /* Try FS.InlineBulkStatus first. Abort codes for the individual * lookups contained therein are stored in the reply without aborting * the whole operation. */ - if (cookie->one_only) - goto no_inline_bulk_status; - - inode = ERR_PTR(-ERESTARTSYS); - if (afs_begin_vnode_operation(&fc, dvnode, key)) { - while (afs_select_fileserver(&fc)) { - if (test_bit(AFS_SERVER_FL_NO_IBULK, - &fc.cbi->server->flags)) { - fc.ac.abort_code = RX_INVALID_OPERATION; - fc.ac.error = -ECONNABORTED; - break; - } - afs_fs_inline_bulk_status(&fc, - afs_v2net(dvnode), - cookie->fids, - cookie->statuses, - cookie->callbacks, - cookie->nr_fids, NULL); - } - - if (fc.ac.error == 0) - cbi = afs_get_cb_interest(fc.cbi); - if (fc.ac.abort_code == RX_INVALID_OPERATION) - set_bit(AFS_SERVER_FL_NO_IBULK, &fc.cbi->server->flags); - inode = ERR_PTR(afs_end_vnode_operation(&fc)); + afs_op_set_error(op, -ENOTSUPP); + if (supports_ibulk) { + op->ops = &afs_inline_bulk_status_operation; + afs_begin_vnode_operation(op); + afs_wait_for_operation(op); } - if (!IS_ERR(inode)) - goto success; - if (fc.ac.abort_code != RX_INVALID_OPERATION) - goto out_c; - -no_inline_bulk_status: - /* We could try FS.BulkStatus next, but this aborts the entire op if - * any of the lookups fails - so, for the moment, revert to - * FS.FetchStatus for just the primary fid. - */ - cookie->nr_fids = 1; - inode = ERR_PTR(-ERESTARTSYS); - if (afs_begin_vnode_operation(&fc, dvnode, key)) { - while (afs_select_fileserver(&fc)) { - afs_fs_fetch_status(&fc, - afs_v2net(dvnode), - cookie->fids, - cookie->statuses, - cookie->callbacks, - NULL); - } - - if (fc.ac.error == 0) - cbi = afs_get_cb_interest(fc.cbi); - inode = ERR_PTR(afs_end_vnode_operation(&fc)); + if (afs_op_error(op) == -ENOTSUPP) { + /* We could try FS.BulkStatus next, but this aborts the entire + * op if any of the lookups fails - so, for the moment, revert + * to FS.FetchStatus for op->file[1]. + */ + op->fetch_status.which = 1; + op->ops = &afs_lookup_fetch_status_operation; + afs_begin_vnode_operation(op); + afs_wait_for_operation(op); } - if (IS_ERR(inode)) - goto out_c; - - for (i = 0; i < cookie->nr_fids; i++) - cookie->statuses[i].abort_code = 0; - -success: - /* Turn all the files into inodes and save the first one - which is the - * one we actually want. - */ - if (cookie->statuses[0].abort_code != 0) - inode = ERR_PTR(afs_abort_to_error(cookie->statuses[0].abort_code)); - - for (i = 0; i < cookie->nr_fids; i++) { - struct inode *ti; - - if (cookie->statuses[i].abort_code != 0) - continue; - - ti = afs_iget(dir->i_sb, key, &cookie->fids[i], - &cookie->statuses[i], - &cookie->callbacks[i], - cbi); - if (i == 0) { - inode = ti; +out_op: + if (!afs_op_error(op)) { + if (op->file[1].scb.status.abort_code) { + afs_op_accumulate_error(op, -ECONNABORTED, + op->file[1].scb.status.abort_code); } else { - if (!IS_ERR(ti)) - iput(ti); + inode = &op->file[1].vnode->netfs.inode; + op->file[1].vnode = NULL; } } -out_c: - afs_put_cb_interest(afs_v2net(dvnode), cbi); - kfree(cookie->callbacks); -out_s: - kfree(cookie->statuses); + if (op->file[0].scb.have_status) + dentry->d_fsdata = (void *)(unsigned long)op->file[0].scb.status.data_version; + else + dentry->d_fsdata = (void *)(unsigned long)op->file[0].dv_before; + ret = afs_put_operation(op); out: kfree(cookie); - return inode; + _leave(""); + return inode ?: ERR_PTR(ret); } /* * Look up an entry in a directory with @sys substitution. */ -static struct dentry *afs_lookup_atsys(struct inode *dir, struct dentry *dentry, - struct key *key) +static struct dentry *afs_lookup_atsys(struct inode *dir, struct dentry *dentry) { struct afs_sysnames *subs; struct afs_net *net = afs_i2net(dir); @@ -800,7 +943,7 @@ static struct dentry *afs_lookup_atsys(struct inode *dir, struct dentry *dentry, } strcpy(p, name); - ret = lookup_one_len(buf, dentry->d_parent, len); + ret = lookup_noperm(&QSTR(buf), dentry->d_parent); if (IS_ERR(ret) || d_is_positive(ret)) goto out_s; dput(ret); @@ -814,7 +957,6 @@ out_s: afs_put_sysnames(subs); kfree(buf); out_p: - key_put(key); return ret; } @@ -825,9 +967,9 @@ static struct dentry *afs_lookup(struct inode *dir, struct dentry *dentry, unsigned int flags) { struct afs_vnode *dvnode = AFS_FS_I(dir); + struct afs_fid fid = {}; struct inode *inode; struct dentry *d; - struct key *key; int ret; _enter("{%llx:%llu},%p{%pd},", @@ -845,15 +987,9 @@ static struct dentry *afs_lookup(struct inode *dir, struct dentry *dentry, return ERR_PTR(-ESTALE); } - key = afs_request_key(dvnode->volume->cell); - if (IS_ERR(key)) { - _leave(" = %ld [key]", PTR_ERR(key)); - return ERR_CAST(key); - } - - ret = afs_validate(dvnode, key); + ret = afs_validate(dvnode, NULL); if (ret < 0) { - key_put(key); + afs_dir_unuse_cookie(dvnode, ret); _leave(" = %d [val]", ret); return ERR_PTR(ret); } @@ -863,40 +999,76 @@ static struct dentry *afs_lookup(struct inode *dir, struct dentry *dentry, dentry->d_name.name[dentry->d_name.len - 3] == 's' && dentry->d_name.name[dentry->d_name.len - 2] == 'y' && dentry->d_name.name[dentry->d_name.len - 1] == 's') - return afs_lookup_atsys(dir, dentry, key); + return afs_lookup_atsys(dir, dentry); afs_stat_v(dvnode, n_lookup); - inode = afs_do_lookup(dir, dentry, key); - key_put(key); - if (inode == ERR_PTR(-ENOENT)) { - inode = afs_try_auto_mntpt(dentry, dir); - } else { - dentry->d_fsdata = - (void *)(unsigned long)dvnode->status.data_version; - } + inode = afs_do_lookup(dir, dentry); + if (inode == ERR_PTR(-ENOENT)) + inode = NULL; + else if (!IS_ERR_OR_NULL(inode)) + fid = AFS_FS_I(inode)->fid; + + _debug("splice %p", dentry->d_inode); d = d_splice_alias(inode, dentry); - if (!IS_ERR_OR_NULL(d)) + if (!IS_ERR_OR_NULL(d)) { d->d_fsdata = dentry->d_fsdata; + trace_afs_lookup(dvnode, &d->d_name, &fid); + } else { + trace_afs_lookup(dvnode, &dentry->d_name, &fid); + } + _leave(""); return d; } /* + * Check the validity of a dentry under RCU conditions. + */ +static int afs_d_revalidate_rcu(struct afs_vnode *dvnode, struct dentry *dentry) +{ + long dir_version, de_version; + + _enter("%p", dentry); + + if (test_bit(AFS_VNODE_DELETED, &dvnode->flags)) + return -ECHILD; + + if (!afs_check_validity(dvnode)) + return -ECHILD; + + /* We only need to invalidate a dentry if the server's copy changed + * behind our back. If we made the change, it's no problem. Note that + * on a 32-bit system, we only have 32 bits in the dentry to store the + * version. + */ + dir_version = (long)READ_ONCE(dvnode->status.data_version); + de_version = (long)READ_ONCE(dentry->d_fsdata); + if (de_version != dir_version) { + dir_version = (long)READ_ONCE(dvnode->invalid_before); + if (de_version - dir_version < 0) + return -ECHILD; + } + + return 1; /* Still valid */ +} + +/* * check that a dentry lookup hit has found a valid entry * - NOTE! the hit can be a negative hit too, so we can't assume we have an * inode */ -static int afs_d_revalidate(struct dentry *dentry, unsigned int flags) +static int afs_d_revalidate(struct inode *parent_dir, const struct qstr *name, + struct dentry *dentry, unsigned int flags) { - struct afs_vnode *vnode, *dir; - struct afs_fid uninitialized_var(fid); - struct dentry *parent; + struct afs_vnode *vnode, *dir = AFS_FS_I(parent_dir); + struct afs_fid fid; struct inode *inode; struct key *key; - long dir_version, de_version; + afs_dataversion_t dir_version, invalid_before; + long de_version; int ret; if (flags & LOOKUP_RCU) - return -ECHILD; + return afs_d_revalidate_rcu(dir, dentry); if (d_really_is_positive(dentry)) { vnode = AFS_FS_I(d_inode(dentry)); @@ -911,26 +1083,16 @@ static int afs_d_revalidate(struct dentry *dentry, unsigned int flags) if (IS_ERR(key)) key = NULL; - if (d_really_is_positive(dentry)) { - inode = d_inode(dentry); - if (inode) { - vnode = AFS_FS_I(inode); - afs_validate(vnode, key); - if (test_bit(AFS_VNODE_DELETED, &vnode->flags)) - goto out_bad; - } - } - - /* lock down the parent dentry so we can peer at it */ - parent = dget_parent(dentry); - dir = AFS_FS_I(d_inode(parent)); - /* validate the parent directory */ - afs_validate(dir, key); + ret = afs_validate(dir, key); + if (ret == -ERESTARTSYS) { + key_put(key); + return ret; + } if (test_bit(AFS_VNODE_DELETED, &dir->flags)) { _debug("%pd: parent dir deleted", dentry); - goto out_bad_parent; + goto not_found; } /* We only need to invalidate a dentry if the server's copy changed @@ -938,30 +1100,30 @@ static int afs_d_revalidate(struct dentry *dentry, unsigned int flags) * on a 32-bit system, we only have 32 bits in the dentry to store the * version. */ - dir_version = (long)dir->status.data_version; + dir_version = dir->status.data_version; de_version = (long)dentry->d_fsdata; - if (de_version == dir_version) - goto out_valid; + if (de_version == (long)dir_version) + goto out_valid_noupdate; - dir_version = (long)dir->invalid_before; - if (de_version - dir_version >= 0) + invalid_before = dir->invalid_before; + if (de_version - (long)invalid_before >= 0) goto out_valid; _debug("dir modified"); afs_stat_v(dir, n_reval); /* search the directory for this vnode */ - ret = afs_do_lookup_one(&dir->vfs_inode, dentry, &fid, key); + ret = afs_do_lookup_one(&dir->netfs.inode, name, &fid, &dir_version); switch (ret) { case 0: /* the filename maps to something */ if (d_really_is_negative(dentry)) - goto out_bad_parent; + goto not_found; inode = d_inode(dentry); if (is_bad_inode(inode)) { printk("kAFS: afs_d_revalidate: %pd2 has bad inode\n", dentry); - goto out_bad_parent; + goto not_found; } vnode = AFS_FS_I(inode); @@ -982,10 +1144,7 @@ static int afs_d_revalidate(struct dentry *dentry, unsigned int flags) _debug("%pd: file deleted (uq %u -> %u I:%u)", dentry, fid.unique, vnode->fid.unique, - vnode->vfs_inode.i_generation); - write_seqlock(&vnode->cb_lock); - set_bit(AFS_VNODE_DELETED, &vnode->flags); - write_sequnlock(&vnode->cb_lock); + vnode->netfs.inode.i_generation); goto not_found; } goto out_valid; @@ -998,28 +1157,19 @@ static int afs_d_revalidate(struct dentry *dentry, unsigned int flags) goto out_valid; default: - _debug("failed to iterate dir %pd: %d", - parent, ret); - goto out_bad_parent; + _debug("failed to iterate parent %pd2: %d", dentry, ret); + goto not_found; } out_valid: - dentry->d_fsdata = (void *)dir_version; - dput(parent); + dentry->d_fsdata = (void *)(unsigned long)dir_version; +out_valid_noupdate: key_put(key); _leave(" = 1 [valid]"); return 1; - /* the dirent, if it exists, now points to a different vnode */ not_found: - spin_lock(&dentry->d_lock); - dentry->d_flags |= DCACHE_NFSFS_RENAMED; - spin_unlock(&dentry->d_lock); - -out_bad_parent: _debug("dropping dentry %pd2", dentry); - dput(parent); -out_bad: key_put(key); _leave(" = 0 [bad]"); @@ -1053,6 +1203,16 @@ zap: } /* + * Clean up sillyrename files on dentry removal. + */ +static void afs_d_iput(struct dentry *dentry, struct inode *inode) +{ + if (dentry->d_flags & DCACHE_NFSFS_RENAMED) + afs_silly_iput(dentry, inode); + iput(inode); +} + +/* * handle dentry release */ void afs_d_release(struct dentry *dentry) @@ -1060,95 +1220,130 @@ void afs_d_release(struct dentry *dentry) _enter("%pd", dentry); } +void afs_check_for_remote_deletion(struct afs_operation *op) +{ + struct afs_vnode *vnode = op->file[0].vnode; + + switch (afs_op_abort_code(op)) { + case VNOVNODE: + set_bit(AFS_VNODE_DELETED, &vnode->flags); + clear_nlink(&vnode->netfs.inode); + afs_break_callback(vnode, afs_cb_break_for_deleted); + } +} + /* * Create a new inode for create/mkdir/symlink */ -static void afs_vnode_new_inode(struct afs_fs_cursor *fc, - struct dentry *new_dentry, - struct afs_fid *newfid, - struct afs_file_status *newstatus, - struct afs_callback *newcb) +static void afs_vnode_new_inode(struct afs_operation *op) { + struct afs_vnode_param *dvp = &op->file[0]; + struct afs_vnode_param *vp = &op->file[1]; struct afs_vnode *vnode; struct inode *inode; - if (fc->ac.error < 0) - return; + _enter(""); - inode = afs_iget(fc->vnode->vfs_inode.i_sb, fc->key, - newfid, newstatus, newcb, fc->cbi); + ASSERTCMP(afs_op_error(op), ==, 0); + + inode = afs_iget(op, vp); if (IS_ERR(inode)) { /* ENOMEM or EINTR at a really inconvenient time - just abandon * the new directory on the server. */ - fc->ac.error = PTR_ERR(inode); + afs_op_accumulate_error(op, PTR_ERR(inode), 0); return; } vnode = AFS_FS_I(inode); set_bit(AFS_VNODE_NEW_CONTENT, &vnode->flags); - afs_vnode_commit_status(fc, vnode, 0); - d_instantiate(new_dentry, inode); + if (S_ISDIR(inode->i_mode)) + afs_mkdir_init_dir(vnode, dvp->vnode); + else if (S_ISLNK(inode->i_mode)) + afs_init_new_symlink(vnode, op); + if (!afs_op_error(op)) + afs_cache_permit(vnode, op->key, vnode->cb_break, &vp->scb); + d_instantiate(op->dentry, inode); +} + +static void afs_create_success(struct afs_operation *op) +{ + _enter("op=%08x", op->debug_id); + op->ctime = op->file[0].scb.status.mtime_client; + afs_vnode_commit_status(op, &op->file[0]); + afs_update_dentry_version(op, &op->file[0], op->dentry); + afs_vnode_new_inode(op); } +static void afs_create_edit_dir(struct afs_operation *op) +{ + struct netfs_cache_resources cres = {}; + struct afs_vnode_param *dvp = &op->file[0]; + struct afs_vnode_param *vp = &op->file[1]; + struct afs_vnode *dvnode = dvp->vnode; + + _enter("op=%08x", op->debug_id); + + fscache_begin_write_operation(&cres, afs_vnode_cache(dvnode)); + down_write(&dvnode->validate_lock); + if (test_bit(AFS_VNODE_DIR_VALID, &dvnode->flags) && + dvnode->status.data_version == dvp->dv_before + dvp->dv_delta) + afs_edit_dir_add(dvnode, &op->dentry->d_name, &vp->fid, + op->create.reason); + up_write(&dvnode->validate_lock); + fscache_end_operation(&cres); +} + +static void afs_create_put(struct afs_operation *op) +{ + _enter("op=%08x", op->debug_id); + + if (afs_op_error(op)) + d_drop(op->dentry); +} + +static const struct afs_operation_ops afs_mkdir_operation = { + .issue_afs_rpc = afs_fs_make_dir, + .issue_yfs_rpc = yfs_fs_make_dir, + .success = afs_create_success, + .aborted = afs_check_for_remote_deletion, + .edit_dir = afs_create_edit_dir, + .put = afs_create_put, +}; + /* * create a directory on an AFS filesystem */ -static int afs_mkdir(struct inode *dir, struct dentry *dentry, umode_t mode) +static struct dentry *afs_mkdir(struct mnt_idmap *idmap, struct inode *dir, + struct dentry *dentry, umode_t mode) { - struct afs_file_status newstatus; - struct afs_fs_cursor fc; - struct afs_callback newcb; + struct afs_operation *op; struct afs_vnode *dvnode = AFS_FS_I(dir); - struct afs_fid newfid; - struct key *key; - u64 data_version = dvnode->status.data_version; int ret; - mode |= S_IFDIR; - _enter("{%llx:%llu},{%pd},%ho", dvnode->fid.vid, dvnode->fid.vnode, dentry, mode); - key = afs_request_key(dvnode->volume->cell); - if (IS_ERR(key)) { - ret = PTR_ERR(key); - goto error; + op = afs_alloc_operation(NULL, dvnode->volume); + if (IS_ERR(op)) { + d_drop(dentry); + return ERR_CAST(op); } - ret = -ERESTARTSYS; - if (afs_begin_vnode_operation(&fc, dvnode, key)) { - while (afs_select_fileserver(&fc)) { - fc.cb_break = afs_calc_vnode_cb_break(dvnode); - afs_fs_create(&fc, dentry->d_name.name, mode, data_version, - &newfid, &newstatus, &newcb); - } - - afs_check_for_remote_deletion(&fc, fc.vnode); - afs_vnode_commit_status(&fc, dvnode, fc.cb_break); - afs_vnode_new_inode(&fc, dentry, &newfid, &newstatus, &newcb); - ret = afs_end_vnode_operation(&fc); - if (ret < 0) - goto error_key; - } else { - goto error_key; - } - - if (ret == 0 && - test_bit(AFS_VNODE_DIR_VALID, &dvnode->flags)) - afs_edit_dir_add(dvnode, &dentry->d_name, &newfid, - afs_edit_dir_for_create); - - key_put(key); - _leave(" = 0"); - return 0; - -error_key: - key_put(key); -error: - d_drop(dentry); - _leave(" = %d", ret); - return ret; + fscache_use_cookie(afs_vnode_cache(dvnode), true); + + afs_op_set_vnode(op, 0, dvnode); + op->file[0].dv_delta = 1; + op->file[0].modification = true; + op->file[0].update_ctime = true; + op->dentry = dentry; + op->create.mode = S_IFDIR | mode; + op->create.reason = afs_edit_dir_for_mkdir; + op->mtime = current_time(dir); + op->ops = &afs_mkdir_operation; + ret = afs_do_sync_operation(op); + afs_dir_unuse_cookie(dvnode, ret); + return ERR_PTR(ret); } /* @@ -1159,63 +1354,109 @@ static void afs_dir_remove_subdir(struct dentry *dentry) if (d_really_is_positive(dentry)) { struct afs_vnode *vnode = AFS_FS_I(d_inode(dentry)); - clear_nlink(&vnode->vfs_inode); + clear_nlink(&vnode->netfs.inode); set_bit(AFS_VNODE_DELETED, &vnode->flags); - clear_bit(AFS_VNODE_CB_PROMISED, &vnode->flags); - clear_bit(AFS_VNODE_DIR_VALID, &vnode->flags); + afs_clear_cb_promise(vnode, afs_cb_promise_clear_rmdir); + afs_invalidate_dir(vnode, afs_dir_invalid_subdir_removed); } } +static void afs_rmdir_success(struct afs_operation *op) +{ + _enter("op=%08x", op->debug_id); + op->ctime = op->file[0].scb.status.mtime_client; + afs_vnode_commit_status(op, &op->file[0]); + afs_update_dentry_version(op, &op->file[0], op->dentry); +} + +static void afs_rmdir_edit_dir(struct afs_operation *op) +{ + struct netfs_cache_resources cres = {}; + struct afs_vnode_param *dvp = &op->file[0]; + struct afs_vnode *dvnode = dvp->vnode; + + _enter("op=%08x", op->debug_id); + afs_dir_remove_subdir(op->dentry); + + fscache_begin_write_operation(&cres, afs_vnode_cache(dvnode)); + down_write(&dvnode->validate_lock); + if (test_bit(AFS_VNODE_DIR_VALID, &dvnode->flags) && + dvnode->status.data_version == dvp->dv_before + dvp->dv_delta) + afs_edit_dir_remove(dvnode, &op->dentry->d_name, + afs_edit_dir_for_rmdir); + up_write(&dvnode->validate_lock); + fscache_end_operation(&cres); +} + +static void afs_rmdir_put(struct afs_operation *op) +{ + _enter("op=%08x", op->debug_id); + if (op->file[1].vnode) + up_write(&op->file[1].vnode->rmdir_lock); +} + +static const struct afs_operation_ops afs_rmdir_operation = { + .issue_afs_rpc = afs_fs_remove_dir, + .issue_yfs_rpc = yfs_fs_remove_dir, + .success = afs_rmdir_success, + .aborted = afs_check_for_remote_deletion, + .edit_dir = afs_rmdir_edit_dir, + .put = afs_rmdir_put, +}; + /* * remove a directory from an AFS filesystem */ static int afs_rmdir(struct inode *dir, struct dentry *dentry) { - struct afs_fs_cursor fc; + struct afs_operation *op; struct afs_vnode *dvnode = AFS_FS_I(dir), *vnode = NULL; - struct key *key; - u64 data_version = dvnode->status.data_version; int ret; _enter("{%llx:%llu},{%pd}", dvnode->fid.vid, dvnode->fid.vnode, dentry); - key = afs_request_key(dvnode->volume->cell); - if (IS_ERR(key)) { - ret = PTR_ERR(key); - goto error; - } + op = afs_alloc_operation(NULL, dvnode->volume); + if (IS_ERR(op)) + return PTR_ERR(op); + + fscache_use_cookie(afs_vnode_cache(dvnode), true); + + afs_op_set_vnode(op, 0, dvnode); + op->file[0].dv_delta = 1; + op->file[0].modification = true; + op->file[0].update_ctime = true; + + op->dentry = dentry; + op->ops = &afs_rmdir_operation; /* Try to make sure we have a callback promise on the victim. */ if (d_really_is_positive(dentry)) { vnode = AFS_FS_I(d_inode(dentry)); - ret = afs_validate(vnode, key); + ret = afs_validate(vnode, op->key); if (ret < 0) - goto error_key; + goto error; } - ret = -ERESTARTSYS; - if (afs_begin_vnode_operation(&fc, dvnode, key)) { - while (afs_select_fileserver(&fc)) { - fc.cb_break = afs_calc_vnode_cb_break(dvnode); - afs_fs_remove(&fc, vnode, dentry->d_name.name, true, - data_version); - } - - afs_vnode_commit_status(&fc, dvnode, fc.cb_break); - ret = afs_end_vnode_operation(&fc); - if (ret == 0) { - afs_dir_remove_subdir(dentry); - if (test_bit(AFS_VNODE_DIR_VALID, &dvnode->flags)) - afs_edit_dir_remove(dvnode, &dentry->d_name, - afs_edit_dir_for_rmdir); - } + if (vnode) { + ret = down_write_killable(&vnode->rmdir_lock); + if (ret < 0) + goto error; + op->file[1].vnode = vnode; } -error_key: - key_put(key); -error: + ret = afs_do_sync_operation(op); + + /* Not all systems that can host afs servers have ENOTEMPTY. */ + if (ret == -EEXIST) + ret = -ENOTEMPTY; +out: + afs_dir_unuse_cookie(dvnode, ret); return ret; + +error: + ret = afs_put_operation(op); + goto out; } /* @@ -1228,56 +1469,95 @@ error: * However, if we didn't have a callback promise outstanding, or it was * outstanding on a different server, then it won't break it either... */ -static int afs_dir_remove_link(struct dentry *dentry, struct key *key, - unsigned long d_version_before, - unsigned long d_version_after) +static void afs_dir_remove_link(struct afs_operation *op) { - bool dir_valid; - int ret = 0; - - /* There were no intervening changes on the server if the version - * number we got back was incremented by exactly 1. - */ - dir_valid = (d_version_after == d_version_before + 1); + struct afs_vnode *dvnode = op->file[0].vnode; + struct afs_vnode *vnode = op->file[1].vnode; + struct dentry *dentry = op->dentry; + int ret; - if (d_really_is_positive(dentry)) { - struct afs_vnode *vnode = AFS_FS_I(d_inode(dentry)); + if (afs_op_error(op) || + (op->file[1].scb.have_status && op->file[1].scb.have_error)) + return; + if (d_really_is_positive(dentry)) + return; - if (test_bit(AFS_VNODE_DELETED, &vnode->flags)) { - /* Already done */ - } else if (dir_valid) { - drop_nlink(&vnode->vfs_inode); - if (vnode->vfs_inode.i_nlink == 0) { - set_bit(AFS_VNODE_DELETED, &vnode->flags); - clear_bit(AFS_VNODE_CB_PROMISED, &vnode->flags); - } - ret = 0; - } else { - clear_bit(AFS_VNODE_CB_PROMISED, &vnode->flags); + if (test_bit(AFS_VNODE_DELETED, &vnode->flags)) { + /* Already done */ + } else if (test_bit(AFS_VNODE_DIR_VALID, &dvnode->flags)) { + write_seqlock(&vnode->cb_lock); + drop_nlink(&vnode->netfs.inode); + if (vnode->netfs.inode.i_nlink == 0) { + set_bit(AFS_VNODE_DELETED, &vnode->flags); + __afs_break_callback(vnode, afs_cb_break_for_unlink); + } + write_sequnlock(&vnode->cb_lock); + } else { + afs_break_callback(vnode, afs_cb_break_for_unlink); - if (test_bit(AFS_VNODE_DELETED, &vnode->flags)) - kdebug("AFS_VNODE_DELETED"); + if (test_bit(AFS_VNODE_DELETED, &vnode->flags)) + _debug("AFS_VNODE_DELETED"); - ret = afs_validate(vnode, key); - if (ret == -ESTALE) - ret = 0; - } - _debug("nlink %d [val %d]", vnode->vfs_inode.i_nlink, ret); + ret = afs_validate(vnode, op->key); + if (ret != -ESTALE) + afs_op_set_error(op, ret); } - return ret; + _debug("nlink %d [val %d]", vnode->netfs.inode.i_nlink, afs_op_error(op)); +} + +static void afs_unlink_success(struct afs_operation *op) +{ + _enter("op=%08x", op->debug_id); + op->ctime = op->file[0].scb.status.mtime_client; + afs_check_dir_conflict(op, &op->file[0]); + afs_vnode_commit_status(op, &op->file[0]); + afs_vnode_commit_status(op, &op->file[1]); + afs_update_dentry_version(op, &op->file[0], op->dentry); + afs_dir_remove_link(op); } +static void afs_unlink_edit_dir(struct afs_operation *op) +{ + struct netfs_cache_resources cres = {}; + struct afs_vnode_param *dvp = &op->file[0]; + struct afs_vnode *dvnode = dvp->vnode; + + _enter("op=%08x", op->debug_id); + fscache_begin_write_operation(&cres, afs_vnode_cache(dvnode)); + down_write(&dvnode->validate_lock); + if (test_bit(AFS_VNODE_DIR_VALID, &dvnode->flags) && + dvnode->status.data_version == dvp->dv_before + dvp->dv_delta) + afs_edit_dir_remove(dvnode, &op->dentry->d_name, + afs_edit_dir_for_unlink); + up_write(&dvnode->validate_lock); + fscache_end_operation(&cres); +} + +static void afs_unlink_put(struct afs_operation *op) +{ + _enter("op=%08x", op->debug_id); + if (op->unlink.need_rehash && afs_op_error(op) < 0 && afs_op_error(op) != -ENOENT) + d_rehash(op->dentry); +} + +static const struct afs_operation_ops afs_unlink_operation = { + .issue_afs_rpc = afs_fs_remove_file, + .issue_yfs_rpc = yfs_fs_remove_file, + .success = afs_unlink_success, + .aborted = afs_check_for_remote_deletion, + .edit_dir = afs_unlink_edit_dir, + .put = afs_unlink_put, +}; + /* * Remove a file or symlink from an AFS filesystem. */ static int afs_unlink(struct inode *dir, struct dentry *dentry) { - struct afs_fs_cursor fc; - struct afs_vnode *dvnode = AFS_FS_I(dir), *vnode = NULL; - struct key *key; - unsigned long d_version = (unsigned long)dentry->d_fsdata; - u64 data_version = dvnode->status.data_version; + struct afs_operation *op; + struct afs_vnode *dvnode = AFS_FS_I(dir); + struct afs_vnode *vnode = AFS_FS_I(d_inode(dentry)); int ret; _enter("{%llx:%llu},{%pd}", @@ -1286,207 +1566,221 @@ static int afs_unlink(struct inode *dir, struct dentry *dentry) if (dentry->d_name.len >= AFSNAMEMAX) return -ENAMETOOLONG; - key = afs_request_key(dvnode->volume->cell); - if (IS_ERR(key)) { - ret = PTR_ERR(key); - goto error; - } + op = afs_alloc_operation(NULL, dvnode->volume); + if (IS_ERR(op)) + return PTR_ERR(op); + + fscache_use_cookie(afs_vnode_cache(dvnode), true); + + afs_op_set_vnode(op, 0, dvnode); + op->file[0].dv_delta = 1; + op->file[0].modification = true; + op->file[0].update_ctime = true; /* Try to make sure we have a callback promise on the victim. */ - if (d_really_is_positive(dentry)) { - vnode = AFS_FS_I(d_inode(dentry)); - ret = afs_validate(vnode, key); - if (ret < 0) - goto error_key; + ret = afs_validate(vnode, op->key); + if (ret < 0) { + afs_op_set_error(op, ret); + goto error; } - ret = -ERESTARTSYS; - if (afs_begin_vnode_operation(&fc, dvnode, key)) { - while (afs_select_fileserver(&fc)) { - fc.cb_break = afs_calc_vnode_cb_break(dvnode); - - if (test_bit(AFS_SERVER_FL_IS_YFS, &fc.cbi->server->flags) && - !test_bit(AFS_SERVER_FL_NO_RM2, &fc.cbi->server->flags)) { - yfs_fs_remove_file2(&fc, vnode, dentry->d_name.name, - data_version); - if (fc.ac.error != -ECONNABORTED || - fc.ac.abort_code != RXGEN_OPCODE) - continue; - set_bit(AFS_SERVER_FL_NO_RM2, &fc.cbi->server->flags); - } + spin_lock(&dentry->d_lock); + if (d_count(dentry) > 1) { + spin_unlock(&dentry->d_lock); + /* Start asynchronous writeout of the inode */ + write_inode_now(d_inode(dentry), 0); + afs_op_set_error(op, afs_sillyrename(dvnode, vnode, dentry, op->key)); + goto error; + } + if (!d_unhashed(dentry)) { + /* Prevent a race with RCU lookup. */ + __d_drop(dentry); + op->unlink.need_rehash = true; + } + spin_unlock(&dentry->d_lock); - afs_fs_remove(&fc, vnode, dentry->d_name.name, false, - data_version); - } + op->file[1].vnode = vnode; + op->file[1].update_ctime = true; + op->file[1].op_unlinked = true; + op->dentry = dentry; + op->ops = &afs_unlink_operation; + afs_begin_vnode_operation(op); + afs_wait_for_operation(op); - afs_vnode_commit_status(&fc, dvnode, fc.cb_break); - ret = afs_end_vnode_operation(&fc); - if (ret == 0) - ret = afs_dir_remove_link( - dentry, key, d_version, - (unsigned long)dvnode->status.data_version); - if (ret == 0 && - test_bit(AFS_VNODE_DIR_VALID, &dvnode->flags)) - afs_edit_dir_remove(dvnode, &dentry->d_name, - afs_edit_dir_for_unlink); + /* If there was a conflict with a third party, check the status of the + * unlinked vnode. + */ + if (afs_op_error(op) == 0 && (op->flags & AFS_OPERATION_DIR_CONFLICT)) { + op->file[1].update_ctime = false; + op->fetch_status.which = 1; + op->ops = &afs_fetch_status_operation; + afs_begin_vnode_operation(op); + afs_wait_for_operation(op); } -error_key: - key_put(key); error: - _leave(" = %d", ret); + ret = afs_put_operation(op); + afs_dir_unuse_cookie(dvnode, ret); return ret; } +static const struct afs_operation_ops afs_create_operation = { + .issue_afs_rpc = afs_fs_create_file, + .issue_yfs_rpc = yfs_fs_create_file, + .success = afs_create_success, + .aborted = afs_check_for_remote_deletion, + .edit_dir = afs_create_edit_dir, + .put = afs_create_put, +}; + /* * create a regular file on an AFS filesystem */ -static int afs_create(struct inode *dir, struct dentry *dentry, umode_t mode, - bool excl) +static int afs_create(struct mnt_idmap *idmap, struct inode *dir, + struct dentry *dentry, umode_t mode, bool excl) { - struct afs_fs_cursor fc; - struct afs_file_status newstatus; - struct afs_callback newcb; + struct afs_operation *op; struct afs_vnode *dvnode = AFS_FS_I(dir); - struct afs_fid newfid; - struct key *key; - u64 data_version = dvnode->status.data_version; - int ret; - - mode |= S_IFREG; + int ret = -ENAMETOOLONG; - _enter("{%llx:%llu},{%pd},%ho,", + _enter("{%llx:%llu},{%pd},%ho", dvnode->fid.vid, dvnode->fid.vnode, dentry, mode); - ret = -ENAMETOOLONG; if (dentry->d_name.len >= AFSNAMEMAX) goto error; - key = afs_request_key(dvnode->volume->cell); - if (IS_ERR(key)) { - ret = PTR_ERR(key); + op = afs_alloc_operation(NULL, dvnode->volume); + if (IS_ERR(op)) { + ret = PTR_ERR(op); goto error; } - ret = -ERESTARTSYS; - if (afs_begin_vnode_operation(&fc, dvnode, key)) { - while (afs_select_fileserver(&fc)) { - fc.cb_break = afs_calc_vnode_cb_break(dvnode); - afs_fs_create(&fc, dentry->d_name.name, mode, data_version, - &newfid, &newstatus, &newcb); - } + fscache_use_cookie(afs_vnode_cache(dvnode), true); - afs_check_for_remote_deletion(&fc, fc.vnode); - afs_vnode_commit_status(&fc, dvnode, fc.cb_break); - afs_vnode_new_inode(&fc, dentry, &newfid, &newstatus, &newcb); - ret = afs_end_vnode_operation(&fc); - if (ret < 0) - goto error_key; - } else { - goto error_key; - } + afs_op_set_vnode(op, 0, dvnode); + op->file[0].dv_delta = 1; + op->file[0].modification = true; + op->file[0].update_ctime = true; - if (test_bit(AFS_VNODE_DIR_VALID, &dvnode->flags)) - afs_edit_dir_add(dvnode, &dentry->d_name, &newfid, - afs_edit_dir_for_create); - - key_put(key); - _leave(" = 0"); - return 0; + op->dentry = dentry; + op->create.mode = S_IFREG | mode; + op->create.reason = afs_edit_dir_for_create; + op->mtime = current_time(dir); + op->ops = &afs_create_operation; + ret = afs_do_sync_operation(op); + afs_dir_unuse_cookie(dvnode, ret); + return ret; -error_key: - key_put(key); error: d_drop(dentry); _leave(" = %d", ret); return ret; } +static void afs_link_success(struct afs_operation *op) +{ + struct afs_vnode_param *dvp = &op->file[0]; + struct afs_vnode_param *vp = &op->file[1]; + + _enter("op=%08x", op->debug_id); + op->ctime = dvp->scb.status.mtime_client; + afs_vnode_commit_status(op, dvp); + afs_vnode_commit_status(op, vp); + afs_update_dentry_version(op, dvp, op->dentry); + if (op->dentry_2->d_parent == op->dentry->d_parent) + afs_update_dentry_version(op, dvp, op->dentry_2); + ihold(&vp->vnode->netfs.inode); + d_instantiate(op->dentry, &vp->vnode->netfs.inode); +} + +static void afs_link_put(struct afs_operation *op) +{ + _enter("op=%08x", op->debug_id); + if (afs_op_error(op)) + d_drop(op->dentry); +} + +static const struct afs_operation_ops afs_link_operation = { + .issue_afs_rpc = afs_fs_link, + .issue_yfs_rpc = yfs_fs_link, + .success = afs_link_success, + .aborted = afs_check_for_remote_deletion, + .edit_dir = afs_create_edit_dir, + .put = afs_link_put, +}; + /* * create a hard link between files in an AFS filesystem */ static int afs_link(struct dentry *from, struct inode *dir, struct dentry *dentry) { - struct afs_fs_cursor fc; - struct afs_vnode *dvnode, *vnode; - struct key *key; - u64 data_version; - int ret; - - vnode = AFS_FS_I(d_inode(from)); - dvnode = AFS_FS_I(dir); - data_version = dvnode->status.data_version; + struct afs_operation *op; + struct afs_vnode *dvnode = AFS_FS_I(dir); + struct afs_vnode *vnode = AFS_FS_I(d_inode(from)); + int ret = -ENAMETOOLONG; _enter("{%llx:%llu},{%llx:%llu},{%pd}", vnode->fid.vid, vnode->fid.vnode, dvnode->fid.vid, dvnode->fid.vnode, dentry); - ret = -ENAMETOOLONG; if (dentry->d_name.len >= AFSNAMEMAX) goto error; - key = afs_request_key(dvnode->volume->cell); - if (IS_ERR(key)) { - ret = PTR_ERR(key); + op = afs_alloc_operation(NULL, dvnode->volume); + if (IS_ERR(op)) { + ret = PTR_ERR(op); goto error; } - ret = -ERESTARTSYS; - if (afs_begin_vnode_operation(&fc, dvnode, key)) { - if (mutex_lock_interruptible_nested(&vnode->io_lock, 1) < 0) { - afs_end_vnode_operation(&fc); - goto error_key; - } - - while (afs_select_fileserver(&fc)) { - fc.cb_break = afs_calc_vnode_cb_break(dvnode); - fc.cb_break_2 = afs_calc_vnode_cb_break(vnode); - afs_fs_link(&fc, vnode, dentry->d_name.name, data_version); - } - - afs_vnode_commit_status(&fc, dvnode, fc.cb_break); - afs_vnode_commit_status(&fc, vnode, fc.cb_break_2); - ihold(&vnode->vfs_inode); - d_instantiate(dentry, &vnode->vfs_inode); - - mutex_unlock(&vnode->io_lock); - ret = afs_end_vnode_operation(&fc); - if (ret < 0) - goto error_key; - } else { - goto error_key; - } - - if (test_bit(AFS_VNODE_DIR_VALID, &dvnode->flags)) - afs_edit_dir_add(dvnode, &dentry->d_name, &vnode->fid, - afs_edit_dir_for_link); - - key_put(key); - _leave(" = 0"); - return 0; + fscache_use_cookie(afs_vnode_cache(dvnode), true); + + ret = afs_validate(vnode, op->key); + if (ret < 0) + goto error_op; + + afs_op_set_vnode(op, 0, dvnode); + afs_op_set_vnode(op, 1, vnode); + op->file[0].dv_delta = 1; + op->file[0].modification = true; + op->file[0].update_ctime = true; + op->file[1].update_ctime = true; + + op->dentry = dentry; + op->dentry_2 = from; + op->ops = &afs_link_operation; + op->create.reason = afs_edit_dir_for_link; + ret = afs_do_sync_operation(op); + afs_dir_unuse_cookie(dvnode, ret); + return ret; -error_key: - key_put(key); +error_op: + afs_put_operation(op); + afs_dir_unuse_cookie(dvnode, ret); error: d_drop(dentry); _leave(" = %d", ret); return ret; } +static const struct afs_operation_ops afs_symlink_operation = { + .issue_afs_rpc = afs_fs_symlink, + .issue_yfs_rpc = yfs_fs_symlink, + .success = afs_create_success, + .aborted = afs_check_for_remote_deletion, + .edit_dir = afs_create_edit_dir, + .put = afs_create_put, +}; + /* * create a symlink in an AFS filesystem */ -static int afs_symlink(struct inode *dir, struct dentry *dentry, - const char *content) +static int afs_symlink(struct mnt_idmap *idmap, struct inode *dir, + struct dentry *dentry, const char *content) { - struct afs_fs_cursor fc; - struct afs_file_status newstatus; + struct afs_operation *op; struct afs_vnode *dvnode = AFS_FS_I(dir); - struct afs_fid newfid; - struct key *key; - u64 data_version = dvnode->status.data_version; int ret; _enter("{%llx:%llu},{%pd},%s", @@ -1501,69 +1795,285 @@ static int afs_symlink(struct inode *dir, struct dentry *dentry, if (strlen(content) >= AFSPATHMAX) goto error; - key = afs_request_key(dvnode->volume->cell); - if (IS_ERR(key)) { - ret = PTR_ERR(key); + op = afs_alloc_operation(NULL, dvnode->volume); + if (IS_ERR(op)) { + ret = PTR_ERR(op); goto error; } - ret = -ERESTARTSYS; - if (afs_begin_vnode_operation(&fc, dvnode, key)) { - while (afs_select_fileserver(&fc)) { - fc.cb_break = afs_calc_vnode_cb_break(dvnode); - afs_fs_symlink(&fc, dentry->d_name.name, - content, data_version, - &newfid, &newstatus); - } - - afs_check_for_remote_deletion(&fc, fc.vnode); - afs_vnode_commit_status(&fc, dvnode, fc.cb_break); - afs_vnode_new_inode(&fc, dentry, &newfid, &newstatus, NULL); - ret = afs_end_vnode_operation(&fc); - if (ret < 0) - goto error_key; - } else { - goto error_key; - } + fscache_use_cookie(afs_vnode_cache(dvnode), true); - if (test_bit(AFS_VNODE_DIR_VALID, &dvnode->flags)) - afs_edit_dir_add(dvnode, &dentry->d_name, &newfid, - afs_edit_dir_for_symlink); + afs_op_set_vnode(op, 0, dvnode); + op->file[0].dv_delta = 1; - key_put(key); - _leave(" = 0"); - return 0; + op->dentry = dentry; + op->ops = &afs_symlink_operation; + op->create.reason = afs_edit_dir_for_symlink; + op->create.symlink = content; + op->mtime = current_time(dir); + ret = afs_do_sync_operation(op); + afs_dir_unuse_cookie(dvnode, ret); + return ret; -error_key: - key_put(key); error: d_drop(dentry); _leave(" = %d", ret); return ret; } +static void afs_rename_success(struct afs_operation *op) +{ + struct afs_vnode *vnode = op->more_files[0].vnode; + struct afs_vnode *new_vnode = op->more_files[1].vnode; + + _enter("op=%08x", op->debug_id); + + op->ctime = op->file[0].scb.status.mtime_client; + afs_check_dir_conflict(op, &op->file[1]); + afs_vnode_commit_status(op, &op->file[0]); + if (op->file[1].vnode != op->file[0].vnode) { + op->ctime = op->file[1].scb.status.mtime_client; + afs_vnode_commit_status(op, &op->file[1]); + } + if (op->more_files[0].scb.have_status) + afs_vnode_commit_status(op, &op->more_files[0]); + if (op->more_files[1].scb.have_status) + afs_vnode_commit_status(op, &op->more_files[1]); + + /* If we're moving a subdir between dirs, we need to update + * its DV counter too as the ".." will be altered. + */ + if (op->file[0].vnode != op->file[1].vnode) { + if (S_ISDIR(vnode->netfs.inode.i_mode)) { + u64 new_dv; + + write_seqlock(&vnode->cb_lock); + + new_dv = vnode->status.data_version + 1; + trace_afs_set_dv(vnode, new_dv); + vnode->status.data_version = new_dv; + inode_set_iversion_raw(&vnode->netfs.inode, new_dv); + + write_sequnlock(&vnode->cb_lock); + } + + if ((op->rename.rename_flags & RENAME_EXCHANGE) && + S_ISDIR(new_vnode->netfs.inode.i_mode)) { + u64 new_dv; + + write_seqlock(&new_vnode->cb_lock); + + new_dv = new_vnode->status.data_version + 1; + new_vnode->status.data_version = new_dv; + inode_set_iversion_raw(&new_vnode->netfs.inode, new_dv); + + write_sequnlock(&new_vnode->cb_lock); + } + } +} + +static void afs_rename_edit_dir(struct afs_operation *op) +{ + struct netfs_cache_resources orig_cres = {}, new_cres = {}; + struct afs_vnode_param *orig_dvp = &op->file[0]; + struct afs_vnode_param *new_dvp = &op->file[1]; + struct afs_vnode *orig_dvnode = orig_dvp->vnode; + struct afs_vnode *new_dvnode = new_dvp->vnode; + struct afs_vnode *vnode = AFS_FS_I(d_inode(op->dentry)); + struct dentry *old_dentry = op->dentry; + struct dentry *new_dentry = op->dentry_2; + struct inode *new_inode; + + _enter("op=%08x", op->debug_id); + + if (op->rename.rehash) { + d_rehash(op->rename.rehash); + op->rename.rehash = NULL; + } + + fscache_begin_write_operation(&orig_cres, afs_vnode_cache(orig_dvnode)); + if (new_dvnode != orig_dvnode) + fscache_begin_write_operation(&new_cres, afs_vnode_cache(new_dvnode)); + + down_write(&orig_dvnode->validate_lock); + if (test_bit(AFS_VNODE_DIR_VALID, &orig_dvnode->flags) && + orig_dvnode->status.data_version == orig_dvp->dv_before + orig_dvp->dv_delta) + afs_edit_dir_remove(orig_dvnode, &old_dentry->d_name, + afs_edit_dir_for_rename_0); + + if (new_dvnode != orig_dvnode) { + up_write(&orig_dvnode->validate_lock); + down_write(&new_dvnode->validate_lock); + } + + if (test_bit(AFS_VNODE_DIR_VALID, &new_dvnode->flags) && + new_dvnode->status.data_version == new_dvp->dv_before + new_dvp->dv_delta) { + if (!op->rename.new_negative) + afs_edit_dir_remove(new_dvnode, &new_dentry->d_name, + afs_edit_dir_for_rename_1); + + afs_edit_dir_add(new_dvnode, &new_dentry->d_name, + &vnode->fid, afs_edit_dir_for_rename_2); + } + + if (S_ISDIR(vnode->netfs.inode.i_mode) && + new_dvnode != orig_dvnode && + test_bit(AFS_VNODE_DIR_VALID, &vnode->flags)) + afs_edit_dir_update(vnode, &dotdot_name, new_dvnode, + afs_edit_dir_for_rename_sub); + + new_inode = d_inode(new_dentry); + if (new_inode) { + spin_lock(&new_inode->i_lock); + if (S_ISDIR(new_inode->i_mode)) + clear_nlink(new_inode); + else if (new_inode->i_nlink > 0) + drop_nlink(new_inode); + spin_unlock(&new_inode->i_lock); + } + + /* Now we can update d_fsdata on the dentries to reflect their + * new parent's data_version. + */ + afs_update_dentry_version(op, new_dvp, op->dentry); + afs_update_dentry_version(op, new_dvp, op->dentry_2); + + d_move(old_dentry, new_dentry); + + up_write(&new_dvnode->validate_lock); + fscache_end_operation(&orig_cres); + if (new_dvnode != orig_dvnode) + fscache_end_operation(&new_cres); +} + +static void afs_rename_exchange_edit_dir(struct afs_operation *op) +{ + struct afs_vnode_param *orig_dvp = &op->file[0]; + struct afs_vnode_param *new_dvp = &op->file[1]; + struct afs_vnode *orig_dvnode = orig_dvp->vnode; + struct afs_vnode *new_dvnode = new_dvp->vnode; + struct afs_vnode *old_vnode = op->more_files[0].vnode; + struct afs_vnode *new_vnode = op->more_files[1].vnode; + struct dentry *old_dentry = op->dentry; + struct dentry *new_dentry = op->dentry_2; + + _enter("op=%08x", op->debug_id); + + if (new_dvnode == orig_dvnode) { + down_write(&orig_dvnode->validate_lock); + if (test_bit(AFS_VNODE_DIR_VALID, &orig_dvnode->flags) && + orig_dvnode->status.data_version == orig_dvp->dv_before + orig_dvp->dv_delta) { + afs_edit_dir_update(orig_dvnode, &old_dentry->d_name, + new_vnode, afs_edit_dir_for_rename_0); + afs_edit_dir_update(orig_dvnode, &new_dentry->d_name, + old_vnode, afs_edit_dir_for_rename_1); + } + + d_exchange(old_dentry, new_dentry); + up_write(&orig_dvnode->validate_lock); + } else { + down_write(&orig_dvnode->validate_lock); + if (test_bit(AFS_VNODE_DIR_VALID, &orig_dvnode->flags) && + orig_dvnode->status.data_version == orig_dvp->dv_before + orig_dvp->dv_delta) + afs_edit_dir_update(orig_dvnode, &old_dentry->d_name, + new_vnode, afs_edit_dir_for_rename_0); + + up_write(&orig_dvnode->validate_lock); + down_write(&new_dvnode->validate_lock); + + if (test_bit(AFS_VNODE_DIR_VALID, &new_dvnode->flags) && + new_dvnode->status.data_version == new_dvp->dv_before + new_dvp->dv_delta) + afs_edit_dir_update(new_dvnode, &new_dentry->d_name, + old_vnode, afs_edit_dir_for_rename_1); + + if (S_ISDIR(old_vnode->netfs.inode.i_mode) && + test_bit(AFS_VNODE_DIR_VALID, &old_vnode->flags)) + afs_edit_dir_update(old_vnode, &dotdot_name, new_dvnode, + afs_edit_dir_for_rename_sub); + + if (S_ISDIR(new_vnode->netfs.inode.i_mode) && + test_bit(AFS_VNODE_DIR_VALID, &new_vnode->flags)) + afs_edit_dir_update(new_vnode, &dotdot_name, orig_dvnode, + afs_edit_dir_for_rename_sub); + + /* Now we can update d_fsdata on the dentries to reflect their + * new parents' data_version. + */ + afs_update_dentry_version(op, new_dvp, old_dentry); + afs_update_dentry_version(op, orig_dvp, new_dentry); + + d_exchange(old_dentry, new_dentry); + up_write(&new_dvnode->validate_lock); + } +} + +static void afs_rename_put(struct afs_operation *op) +{ + _enter("op=%08x", op->debug_id); + if (op->rename.rehash) + d_rehash(op->rename.rehash); + dput(op->rename.tmp); + if (afs_op_error(op)) + d_rehash(op->dentry); +} + +static const struct afs_operation_ops afs_rename_operation = { + .issue_afs_rpc = afs_fs_rename, + .issue_yfs_rpc = yfs_fs_rename, + .success = afs_rename_success, + .edit_dir = afs_rename_edit_dir, + .put = afs_rename_put, +}; + +#if 0 /* Autoswitched in yfs_fs_rename_replace(). */ +static const struct afs_operation_ops afs_rename_replace_operation = { + .issue_afs_rpc = NULL, + .issue_yfs_rpc = yfs_fs_rename_replace, + .success = afs_rename_success, + .edit_dir = afs_rename_edit_dir, + .put = afs_rename_put, +}; +#endif + +static const struct afs_operation_ops afs_rename_noreplace_operation = { + .issue_afs_rpc = NULL, + .issue_yfs_rpc = yfs_fs_rename_noreplace, + .success = afs_rename_success, + .edit_dir = afs_rename_edit_dir, + .put = afs_rename_put, +}; + +static const struct afs_operation_ops afs_rename_exchange_operation = { + .issue_afs_rpc = NULL, + .issue_yfs_rpc = yfs_fs_rename_exchange, + .success = afs_rename_success, + .edit_dir = afs_rename_exchange_edit_dir, + .put = afs_rename_put, +}; + /* * rename a file in an AFS filesystem and/or move it between directories */ -static int afs_rename(struct inode *old_dir, struct dentry *old_dentry, - struct inode *new_dir, struct dentry *new_dentry, - unsigned int flags) +static int afs_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 afs_fs_cursor fc; - struct afs_vnode *orig_dvnode, *new_dvnode, *vnode; - struct key *key; - u64 orig_data_version, new_data_version; - bool new_negative = d_is_negative(new_dentry); + struct afs_operation *op; + struct afs_vnode *orig_dvnode, *new_dvnode, *vnode, *new_vnode = NULL; int ret; - if (flags) + if (flags & ~(RENAME_NOREPLACE | RENAME_EXCHANGE)) + return -EINVAL; + + /* Don't allow silly-rename files be moved around. */ + if (old_dentry->d_flags & DCACHE_NFSFS_RENAMED) return -EINVAL; vnode = AFS_FS_I(d_inode(old_dentry)); orig_dvnode = AFS_FS_I(old_dir); new_dvnode = AFS_FS_I(new_dir); - orig_data_version = orig_dvnode->status.data_version; - new_data_version = new_dvnode->status.data_version; + if (d_is_positive(new_dentry)) + new_vnode = AFS_FS_I(d_inode(new_dentry)); _enter("{%llx:%llu},{%llx:%llu},{%llx:%llu},{%pd}", orig_dvnode->fid.vid, orig_dvnode->fid.vnode, @@ -1571,99 +2081,142 @@ static int afs_rename(struct inode *old_dir, struct dentry *old_dentry, new_dvnode->fid.vid, new_dvnode->fid.vnode, new_dentry); - key = afs_request_key(orig_dvnode->volume->cell); - if (IS_ERR(key)) { - ret = PTR_ERR(key); - goto error; - } - - ret = -ERESTARTSYS; - if (afs_begin_vnode_operation(&fc, orig_dvnode, key)) { - if (orig_dvnode != new_dvnode) { - if (mutex_lock_interruptible_nested(&new_dvnode->io_lock, 1) < 0) { - afs_end_vnode_operation(&fc); - goto error_key; - } - } - while (afs_select_fileserver(&fc)) { - fc.cb_break = afs_calc_vnode_cb_break(orig_dvnode); - fc.cb_break_2 = afs_calc_vnode_cb_break(new_dvnode); - afs_fs_rename(&fc, old_dentry->d_name.name, - new_dvnode, new_dentry->d_name.name, - orig_data_version, new_data_version); - } + op = afs_alloc_operation(NULL, orig_dvnode->volume); + if (IS_ERR(op)) + return PTR_ERR(op); - afs_vnode_commit_status(&fc, orig_dvnode, fc.cb_break); - afs_vnode_commit_status(&fc, new_dvnode, fc.cb_break_2); - if (orig_dvnode != new_dvnode) - mutex_unlock(&new_dvnode->io_lock); - ret = afs_end_vnode_operation(&fc); - if (ret < 0) - goto error_key; - } + fscache_use_cookie(afs_vnode_cache(orig_dvnode), true); + if (new_dvnode != orig_dvnode) + fscache_use_cookie(afs_vnode_cache(new_dvnode), true); - if (ret == 0) { - if (test_bit(AFS_VNODE_DIR_VALID, &orig_dvnode->flags)) - afs_edit_dir_remove(orig_dvnode, &old_dentry->d_name, - afs_edit_dir_for_rename); + ret = afs_validate(vnode, op->key); + afs_op_set_error(op, ret); + if (ret < 0) + goto error; - if (!new_negative && - test_bit(AFS_VNODE_DIR_VALID, &new_dvnode->flags)) - afs_edit_dir_remove(new_dvnode, &new_dentry->d_name, - afs_edit_dir_for_rename); + ret = -ENOMEM; + op->more_files = kvcalloc(2, sizeof(struct afs_vnode_param), GFP_KERNEL); + if (!op->more_files) + goto error; - if (test_bit(AFS_VNODE_DIR_VALID, &new_dvnode->flags)) - afs_edit_dir_add(new_dvnode, &new_dentry->d_name, - &vnode->fid, afs_edit_dir_for_rename); - } + afs_op_set_vnode(op, 0, orig_dvnode); + afs_op_set_vnode(op, 1, new_dvnode); /* May be same as orig_dvnode */ + op->file[0].dv_delta = 1; + op->file[1].dv_delta = 1; + op->file[0].modification = true; + op->file[1].modification = true; + op->file[0].update_ctime = true; + op->file[1].update_ctime = true; + op->more_files[0].vnode = vnode; + op->more_files[0].speculative = true; + op->more_files[1].vnode = new_vnode; + op->more_files[1].speculative = true; + op->nr_files = 4; + + op->dentry = old_dentry; + op->dentry_2 = new_dentry; + op->rename.rename_flags = flags; + op->rename.new_negative = d_is_negative(new_dentry); + + if (flags & RENAME_NOREPLACE) { + op->ops = &afs_rename_noreplace_operation; + } else if (flags & RENAME_EXCHANGE) { + op->ops = &afs_rename_exchange_operation; + d_drop(new_dentry); + } else { + /* If we might displace the target, we might need to do silly + * rename. + */ + op->ops = &afs_rename_operation; -error_key: - key_put(key); -error: - _leave(" = %d", ret); - return ret; -} + /* For non-directories, check whether the target is busy and if + * so, make a copy of the dentry and then do a silly-rename. + * If the silly-rename succeeds, the copied dentry is hashed + * and becomes the new target. + */ + if (d_is_positive(new_dentry) && !d_is_dir(new_dentry)) { + /* To prevent any new references to the target during + * the rename, we unhash the dentry in advance. + */ + if (!d_unhashed(new_dentry)) { + d_drop(new_dentry); + op->rename.rehash = new_dentry; + } -/* - * Release a directory page and clean up its private state if it's not busy - * - return true if the page can now be released, false if not - */ -static int afs_dir_releasepage(struct page *page, gfp_t gfp_flags) -{ - struct afs_vnode *dvnode = AFS_FS_I(page->mapping->host); + if (d_count(new_dentry) > 2) { + /* copy the target dentry's name */ + op->rename.tmp = d_alloc(new_dentry->d_parent, + &new_dentry->d_name); + if (!op->rename.tmp) { + afs_op_nomem(op); + goto error; + } + + ret = afs_sillyrename(new_dvnode, + AFS_FS_I(d_inode(new_dentry)), + new_dentry, op->key); + if (ret) { + afs_op_set_error(op, ret); + goto error; + } + + op->dentry_2 = op->rename.tmp; + op->rename.rehash = NULL; + op->rename.new_negative = true; + } + } + } - _enter("{{%llx:%llu}[%lu]}", dvnode->fid.vid, dvnode->fid.vnode, page->index); + /* This bit is potentially nasty as there's a potential race with + * afs_d_revalidate{,_rcu}(). We have to change d_fsdata on the dentry + * to reflect it's new parent's new data_version after the op, but + * d_revalidate may see old_dentry between the op having taken place + * and the version being updated. + * + * So drop the old_dentry for now to make other threads go through + * lookup instead - which we hold a lock against. + */ + d_drop(old_dentry); - set_page_private(page, 0); - ClearPagePrivate(page); + ret = afs_do_sync_operation(op); + if (ret == -ENOTSUPP) + ret = -EINVAL; +out: + afs_dir_unuse_cookie(orig_dvnode, ret); + if (new_dvnode != orig_dvnode) + afs_dir_unuse_cookie(new_dvnode, ret); + return ret; - /* The directory will need reloading. */ - if (test_and_clear_bit(AFS_VNODE_DIR_VALID, &dvnode->flags)) - afs_stat_v(dvnode, n_relpg); - return 1; +error: + ret = afs_put_operation(op); + goto out; } /* - * invalidate part or all of a page - * - release a page and clean up its private data if offset is 0 (indicating - * the entire page) + * Write the file contents to the cache as a single blob. */ -static void afs_dir_invalidatepage(struct page *page, unsigned int offset, - unsigned int length) +int afs_single_writepages(struct address_space *mapping, + struct writeback_control *wbc) { - struct afs_vnode *dvnode = AFS_FS_I(page->mapping->host); - - _enter("{%lu},%u,%u", page->index, offset, length); - - BUG_ON(!PageLocked(page)); - - /* The directory will need reloading. */ - if (test_and_clear_bit(AFS_VNODE_DIR_VALID, &dvnode->flags)) - afs_stat_v(dvnode, n_inval); + struct afs_vnode *dvnode = AFS_FS_I(mapping->host); + struct iov_iter iter; + bool is_dir = (S_ISDIR(dvnode->netfs.inode.i_mode) && + !test_bit(AFS_VNODE_MOUNTPOINT, &dvnode->flags)); + int ret = 0; - /* we clean up only if the entire page is being invalidated */ - if (offset == 0 && length == PAGE_SIZE) { - set_page_private(page, 0); - ClearPagePrivate(page); + /* Need to lock to prevent the folio queue and folios from being thrown + * away. + */ + down_read(&dvnode->validate_lock); + + if (is_dir ? + test_bit(AFS_VNODE_DIR_VALID, &dvnode->flags) : + atomic64_read(&dvnode->cb_expires_at) != AFS_NO_CB_PROMISE) { + iov_iter_folio_queue(&iter, ITER_SOURCE, dvnode->directory, 0, 0, + i_size_read(&dvnode->netfs.inode)); + ret = netfs_writeback_single(mapping, wbc, &iter); } + + up_read(&dvnode->validate_lock); + return ret; } |