diff options
Diffstat (limited to 'fs/nfs')
47 files changed, 2860 insertions, 2456 deletions
diff --git a/fs/nfs/Makefile b/fs/nfs/Makefile index 03192a66c14..4782e0840dc 100644 --- a/fs/nfs/Makefile +++ b/fs/nfs/Makefile @@ -29,8 +29,6 @@ nfsv4-$(CONFIG_NFS_USE_LEGACY_DNS) += cache_lib.o nfsv4-$(CONFIG_SYSCTL) += nfs4sysctl.o nfsv4-$(CONFIG_NFS_V4_1) += pnfs.o pnfs_dev.o -obj-$(CONFIG_PNFS_FILE_LAYOUT) += nfs_layout_nfsv41_files.o -nfs_layout_nfsv41_files-y := nfs4filelayout.o nfs4filelayoutdev.o - +obj-$(CONFIG_PNFS_FILE_LAYOUT) += filelayout/ obj-$(CONFIG_PNFS_OBJLAYOUT) += objlayout/ obj-$(CONFIG_PNFS_BLOCK) += blocklayout/ diff --git a/fs/nfs/blocklayout/blocklayout.c b/fs/nfs/blocklayout/blocklayout.c index e242bbf7297..9b431f44fad 100644 --- a/fs/nfs/blocklayout/blocklayout.c +++ b/fs/nfs/blocklayout/blocklayout.c @@ -134,8 +134,8 @@ bl_submit_bio(int rw, struct bio *bio) if (bio) { get_parallel(bio->bi_private); dprintk("%s submitting %s bio %u@%llu\n", __func__, - rw == READ ? "read" : "write", - bio->bi_size, (unsigned long long)bio->bi_sector); + rw == READ ? "read" : "write", bio->bi_iter.bi_size, + (unsigned long long)bio->bi_iter.bi_sector); submit_bio(rw, bio); } return NULL; @@ -156,7 +156,8 @@ static struct bio *bl_alloc_init_bio(int npg, sector_t isect, } if (bio) { - bio->bi_sector = isect - be->be_f_offset + be->be_v_offset; + bio->bi_iter.bi_sector = isect - be->be_f_offset + + be->be_v_offset; bio->bi_bdev = be->be_mdev; bio->bi_end_io = end_io; bio->bi_private = par; @@ -201,19 +202,15 @@ static struct bio *bl_add_page_to_bio(struct bio *bio, int npg, int rw, static void bl_end_io_read(struct bio *bio, int err) { struct parallel_io *par = bio->bi_private; - const int uptodate = test_bit(BIO_UPTODATE, &bio->bi_flags); - struct bio_vec *bvec = bio->bi_io_vec + bio->bi_vcnt - 1; + struct bio_vec *bvec; + int i; - do { - struct page *page = bvec->bv_page; + if (!err) + bio_for_each_segment_all(bvec, bio, i) + SetPageUptodate(bvec->bv_page); - if (--bvec >= bio->bi_io_vec) - prefetchw(&bvec->bv_page->flags); - if (uptodate) - SetPageUptodate(page); - } while (bvec >= bio->bi_io_vec); - if (!uptodate) { - struct nfs_read_data *rdata = par->data; + if (err) { + struct nfs_pgio_data *rdata = par->data; struct nfs_pgio_header *header = rdata->header; if (!header->pnfs_error) @@ -227,17 +224,17 @@ static void bl_end_io_read(struct bio *bio, int err) static void bl_read_cleanup(struct work_struct *work) { struct rpc_task *task; - struct nfs_read_data *rdata; + struct nfs_pgio_data *rdata; dprintk("%s enter\n", __func__); task = container_of(work, struct rpc_task, u.tk_work); - rdata = container_of(task, struct nfs_read_data, task); + rdata = container_of(task, struct nfs_pgio_data, task); pnfs_ld_read_done(rdata); } static void bl_end_par_io_read(void *data, int unused) { - struct nfs_read_data *rdata = data; + struct nfs_pgio_data *rdata = data; rdata->task.tk_status = rdata->header->pnfs_error; INIT_WORK(&rdata->task.u.tk_work, bl_read_cleanup); @@ -245,7 +242,7 @@ bl_end_par_io_read(void *data, int unused) } static enum pnfs_try_status -bl_read_pagelist(struct nfs_read_data *rdata) +bl_read_pagelist(struct nfs_pgio_data *rdata) { struct nfs_pgio_header *header = rdata->header; int i, hole; @@ -383,21 +380,17 @@ static void mark_extents_written(struct pnfs_block_layout *bl, static void bl_end_io_write_zero(struct bio *bio, int err) { struct parallel_io *par = bio->bi_private; - const int uptodate = test_bit(BIO_UPTODATE, &bio->bi_flags); - struct bio_vec *bvec = bio->bi_io_vec + bio->bi_vcnt - 1; - - do { - struct page *page = bvec->bv_page; + struct bio_vec *bvec; + int i; - if (--bvec >= bio->bi_io_vec) - prefetchw(&bvec->bv_page->flags); + bio_for_each_segment_all(bvec, bio, i) { /* This is the zeroing page we added */ - end_page_writeback(page); - page_cache_release(page); - } while (bvec >= bio->bi_io_vec); + end_page_writeback(bvec->bv_page); + page_cache_release(bvec->bv_page); + } - if (unlikely(!uptodate)) { - struct nfs_write_data *data = par->data; + if (unlikely(err)) { + struct nfs_pgio_data *data = par->data; struct nfs_pgio_header *header = data->header; if (!header->pnfs_error) @@ -412,7 +405,7 @@ static void bl_end_io_write(struct bio *bio, int err) { struct parallel_io *par = bio->bi_private; const int uptodate = test_bit(BIO_UPTODATE, &bio->bi_flags); - struct nfs_write_data *data = par->data; + struct nfs_pgio_data *data = par->data; struct nfs_pgio_header *header = data->header; if (!uptodate) { @@ -430,10 +423,10 @@ static void bl_end_io_write(struct bio *bio, int err) static void bl_write_cleanup(struct work_struct *work) { struct rpc_task *task; - struct nfs_write_data *wdata; + struct nfs_pgio_data *wdata; dprintk("%s enter\n", __func__); task = container_of(work, struct rpc_task, u.tk_work); - wdata = container_of(task, struct nfs_write_data, task); + wdata = container_of(task, struct nfs_pgio_data, task); if (likely(!wdata->header->pnfs_error)) { /* Marks for LAYOUTCOMMIT */ mark_extents_written(BLK_LSEG2EXT(wdata->header->lseg), @@ -445,7 +438,7 @@ static void bl_write_cleanup(struct work_struct *work) /* Called when last of bios associated with a bl_write_pagelist call finishes */ static void bl_end_par_io_write(void *data, int num_se) { - struct nfs_write_data *wdata = data; + struct nfs_pgio_data *wdata = data; if (unlikely(wdata->header->pnfs_error)) { bl_free_short_extents(&BLK_LSEG2EXT(wdata->header->lseg)->bl_inval, @@ -519,7 +512,7 @@ bl_do_readpage_sync(struct page *page, struct pnfs_block_extent *be, isect = (page->index << PAGE_CACHE_SECTOR_SHIFT) + (offset / SECTOR_SIZE); - bio->bi_sector = isect - be->be_f_offset + be->be_v_offset; + bio->bi_iter.bi_sector = isect - be->be_f_offset + be->be_v_offset; bio->bi_bdev = be->be_mdev; bio->bi_end_io = bl_read_single_end_io; @@ -680,7 +673,7 @@ check_page: } static enum pnfs_try_status -bl_write_pagelist(struct nfs_write_data *wdata, int sync) +bl_write_pagelist(struct nfs_pgio_data *wdata, int sync) { struct nfs_pgio_header *header = wdata->header; int i, ret, npg_zero, pg_index, last = 0; @@ -1196,13 +1189,17 @@ bl_pg_init_read(struct nfs_pageio_descriptor *pgio, struct nfs_page *req) pnfs_generic_pg_init_read(pgio, req); } -static bool +/* + * Return 0 if @req cannot be coalesced into @pgio, otherwise return the number + * of bytes (maximum @req->wb_bytes) that can be coalesced. + */ +static size_t bl_pg_test_read(struct nfs_pageio_descriptor *pgio, struct nfs_page *prev, struct nfs_page *req) { if (pgio->pg_dreq != NULL && !is_aligned_req(req, SECTOR_SIZE)) - return false; + return 0; return pnfs_generic_pg_test(pgio, prev, req); } @@ -1220,7 +1217,7 @@ static u64 pnfs_num_cont_bytes(struct inode *inode, pgoff_t idx) end = DIV_ROUND_UP(i_size_read(inode), PAGE_CACHE_SIZE); if (end != NFS_I(inode)->npages) { rcu_read_lock(); - end = radix_tree_next_hole(&mapping->page_tree, idx + 1, ULONG_MAX); + end = page_cache_next_hole(mapping, idx + 1, ULONG_MAX); rcu_read_unlock(); } @@ -1248,13 +1245,17 @@ bl_pg_init_write(struct nfs_pageio_descriptor *pgio, struct nfs_page *req) } } -static bool +/* + * Return 0 if @req cannot be coalesced into @pgio, otherwise return the number + * of bytes (maximum @req->wb_bytes) that can be coalesced. + */ +static size_t bl_pg_test_write(struct nfs_pageio_descriptor *pgio, struct nfs_page *prev, struct nfs_page *req) { if (pgio->pg_dreq != NULL && !is_aligned_req(req, PAGE_CACHE_SIZE)) - return false; + return 0; return pnfs_generic_pg_test(pgio, prev, req); } diff --git a/fs/nfs/blocklayout/blocklayout.h b/fs/nfs/blocklayout/blocklayout.h index 8485978993e..9838fb02047 100644 --- a/fs/nfs/blocklayout/blocklayout.h +++ b/fs/nfs/blocklayout/blocklayout.h @@ -36,6 +36,7 @@ #include <linux/nfs_fs.h> #include <linux/sunrpc/rpc_pipe_fs.h> +#include "../nfs4_fs.h" #include "../pnfs.h" #include "../netns.h" diff --git a/fs/nfs/blocklayout/extents.c b/fs/nfs/blocklayout/extents.c index 9c3e117c3ed..4d016144256 100644 --- a/fs/nfs/blocklayout/extents.c +++ b/fs/nfs/blocklayout/extents.c @@ -44,7 +44,7 @@ static inline sector_t normalize(sector_t s, int base) { sector_t tmp = s; /* Since do_div modifies its argument */ - return s - do_div(tmp, base); + return s - sector_div(tmp, base); } static inline sector_t normalize_up(sector_t s, int base) diff --git a/fs/nfs/callback_proc.c b/fs/nfs/callback_proc.c index ae2e87b9545..41db5258e7a 100644 --- a/fs/nfs/callback_proc.c +++ b/fs/nfs/callback_proc.c @@ -112,7 +112,8 @@ out: * TODO: keep track of all layouts (and delegations) in a hash table * hashed by filehandle. */ -static struct pnfs_layout_hdr * get_layout_by_fh_locked(struct nfs_client *clp, struct nfs_fh *fh) +static struct pnfs_layout_hdr * get_layout_by_fh_locked(struct nfs_client *clp, + struct nfs_fh *fh, nfs4_stateid *stateid) { struct nfs_server *server; struct inode *ino; @@ -120,17 +121,19 @@ static struct pnfs_layout_hdr * get_layout_by_fh_locked(struct nfs_client *clp, list_for_each_entry_rcu(server, &clp->cl_superblocks, client_link) { list_for_each_entry(lo, &server->layouts, plh_layouts) { + if (!nfs4_stateid_match_other(&lo->plh_stateid, stateid)) + continue; if (nfs_compare_fh(fh, &NFS_I(lo->plh_inode)->fh)) continue; ino = igrab(lo->plh_inode); if (!ino) - continue; + break; spin_lock(&ino->i_lock); /* Is this layout in the process of being freed? */ if (NFS_I(ino)->layout != lo) { spin_unlock(&ino->i_lock); iput(ino); - continue; + break; } pnfs_get_layout_hdr(lo); spin_unlock(&ino->i_lock); @@ -141,13 +144,14 @@ static struct pnfs_layout_hdr * get_layout_by_fh_locked(struct nfs_client *clp, return NULL; } -static struct pnfs_layout_hdr * get_layout_by_fh(struct nfs_client *clp, struct nfs_fh *fh) +static struct pnfs_layout_hdr * get_layout_by_fh(struct nfs_client *clp, + struct nfs_fh *fh, nfs4_stateid *stateid) { struct pnfs_layout_hdr *lo; spin_lock(&clp->cl_lock); rcu_read_lock(); - lo = get_layout_by_fh_locked(clp, fh); + lo = get_layout_by_fh_locked(clp, fh, stateid); rcu_read_unlock(); spin_unlock(&clp->cl_lock); @@ -162,9 +166,9 @@ static u32 initiate_file_draining(struct nfs_client *clp, u32 rv = NFS4ERR_NOMATCHING_LAYOUT; LIST_HEAD(free_me_list); - lo = get_layout_by_fh(clp, &args->cbl_fh); + lo = get_layout_by_fh(clp, &args->cbl_fh, &args->cbl_stateid); if (!lo) - return NFS4ERR_NOMATCHING_LAYOUT; + goto out; ino = lo->plh_inode; spin_lock(&ino->i_lock); @@ -179,6 +183,7 @@ static u32 initiate_file_draining(struct nfs_client *clp, pnfs_free_lseg_list(&free_me_list); pnfs_put_layout_hdr(lo); iput(ino); +out: return rv; } diff --git a/fs/nfs/delegation.c b/fs/nfs/delegation.c index ef792f29f83..5d8ccecf5f5 100644 --- a/fs/nfs/delegation.c +++ b/fs/nfs/delegation.c @@ -659,16 +659,19 @@ int nfs_async_inode_return_delegation(struct inode *inode, rcu_read_lock(); delegation = rcu_dereference(NFS_I(inode)->delegation); + if (delegation == NULL) + goto out_enoent; - if (!clp->cl_mvops->match_stateid(&delegation->stateid, stateid)) { - rcu_read_unlock(); - return -ENOENT; - } + if (!clp->cl_mvops->match_stateid(&delegation->stateid, stateid)) + goto out_enoent; nfs_mark_return_delegation(server, delegation); rcu_read_unlock(); nfs_delegation_run_state_manager(clp); return 0; +out_enoent: + rcu_read_unlock(); + return -ENOENT; } static struct inode * diff --git a/fs/nfs/dir.c b/fs/nfs/dir.c index 812154aff98..4a3d4ef7612 100644 --- a/fs/nfs/dir.c +++ b/fs/nfs/dir.c @@ -69,21 +69,28 @@ const struct address_space_operations nfs_dir_aops = { static struct nfs_open_dir_context *alloc_nfs_open_dir_context(struct inode *dir, struct rpc_cred *cred) { + struct nfs_inode *nfsi = NFS_I(dir); struct nfs_open_dir_context *ctx; ctx = kmalloc(sizeof(*ctx), GFP_KERNEL); if (ctx != NULL) { ctx->duped = 0; - ctx->attr_gencount = NFS_I(dir)->attr_gencount; + ctx->attr_gencount = nfsi->attr_gencount; ctx->dir_cookie = 0; ctx->dup_cookie = 0; ctx->cred = get_rpccred(cred); + spin_lock(&dir->i_lock); + list_add(&ctx->list, &nfsi->open_files); + spin_unlock(&dir->i_lock); return ctx; } return ERR_PTR(-ENOMEM); } -static void put_nfs_open_dir_context(struct nfs_open_dir_context *ctx) +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); + spin_unlock(&dir->i_lock); put_rpccred(ctx->cred); kfree(ctx); } @@ -126,7 +133,7 @@ out: static int nfs_closedir(struct inode *inode, struct file *filp) { - put_nfs_open_dir_context(filp->private_data); + put_nfs_open_dir_context(filp->f_path.dentry->d_inode, filp->private_data); return 0; } @@ -274,6 +281,15 @@ out_eof: return -EBADCOOKIE; } +static bool +nfs_readdir_inode_mapping_valid(struct nfs_inode *nfsi) +{ + if (nfsi->cache_validity & (NFS_INO_INVALID_ATTR|NFS_INO_INVALID_DATA)) + return false; + smp_rmb(); + return !test_bit(NFS_INO_INVALIDATING, &nfsi->flags); +} + static int nfs_readdir_search_for_cookie(struct nfs_cache_array *array, nfs_readdir_descriptor_t *desc) { @@ -287,8 +303,8 @@ int nfs_readdir_search_for_cookie(struct nfs_cache_array *array, nfs_readdir_des struct nfs_open_dir_context *ctx = desc->file->private_data; new_pos = desc->current_index + i; - if (ctx->attr_gencount != nfsi->attr_gencount - || (nfsi->cache_validity & (NFS_INO_INVALID_ATTR|NFS_INO_INVALID_DATA))) { + 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) { @@ -297,10 +313,9 @@ int nfs_readdir_search_for_cookie(struct nfs_cache_array *array, nfs_readdir_des 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.name, - *desc->dir_cookie); + "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; @@ -428,6 +443,22 @@ void nfs_advise_use_readdirplus(struct inode *dir) set_bit(NFS_INO_ADVISE_RDPLUS, &NFS_I(dir)->flags); } +/* + * 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) +{ + if (!list_empty(&NFS_I(dir)->open_files)) { + nfs_advise_use_readdirplus(dir); + nfs_zap_mapping(dir, dir->i_mapping); + } +} + static void nfs_prime_dcache(struct dentry *parent, struct nfs_entry *entry) { @@ -806,6 +837,17 @@ int uncached_readdir(nfs_readdir_descriptor_t *desc) goto out; } +static bool nfs_dir_mapping_need_revalidate(struct inode *dir) +{ + struct nfs_inode *nfsi = NFS_I(dir); + + if (nfs_attribute_cache_expired(dir)) + return true; + if (nfsi->cache_validity & NFS_INO_INVALID_DATA) + return true; + return false; +} + /* The file offset position represents the dirent entry number. A last cookie cache takes care of the common case of reading the whole directory. @@ -838,7 +880,7 @@ static int nfs_readdir(struct file *file, struct dir_context *ctx) desc->plus = nfs_use_readdirplus(inode, ctx) ? 1 : 0; nfs_block_sillyrename(dentry); - if (ctx->pos == 0 || nfs_attribute_cache_expired(inode)) + if (ctx->pos == 0 || nfs_dir_mapping_need_revalidate(inode)) res = nfs_revalidate_mapping(inode, file->f_mapping); if (res < 0) goto out; @@ -1404,7 +1446,7 @@ int nfs_atomic_open(struct inode *dir, struct dentry *dentry, /* Expect a negative dentry */ BUG_ON(dentry->d_inode); - dfprintk(VFS, "NFS: atomic_open(%s/%ld), %pd\n", + dfprintk(VFS, "NFS: atomic_open(%s/%lu), %pd\n", dir->i_sb->s_id, dir->i_ino, dentry); err = nfs_check_flags(open_flags); @@ -1594,7 +1636,7 @@ int nfs_create(struct inode *dir, struct dentry *dentry, int open_flags = excl ? O_CREAT | O_EXCL : O_CREAT; int error; - dfprintk(VFS, "NFS: create(%s/%ld), %pd\n", + dfprintk(VFS, "NFS: create(%s/%lu), %pd\n", dir->i_sb->s_id, dir->i_ino, dentry); attr.ia_mode = mode; @@ -1621,7 +1663,7 @@ nfs_mknod(struct inode *dir, struct dentry *dentry, umode_t mode, dev_t rdev) struct iattr attr; int status; - dfprintk(VFS, "NFS: mknod(%s/%ld), %pd\n", + dfprintk(VFS, "NFS: mknod(%s/%lu), %pd\n", dir->i_sb->s_id, dir->i_ino, dentry); if (!new_valid_dev(rdev)) @@ -1650,7 +1692,7 @@ int nfs_mkdir(struct inode *dir, struct dentry *dentry, umode_t mode) struct iattr attr; int error; - dfprintk(VFS, "NFS: mkdir(%s/%ld), %pd\n", + dfprintk(VFS, "NFS: mkdir(%s/%lu), %pd\n", dir->i_sb->s_id, dir->i_ino, dentry); attr.ia_valid = ATTR_MODE; @@ -1678,7 +1720,7 @@ int nfs_rmdir(struct inode *dir, struct dentry *dentry) { int error; - dfprintk(VFS, "NFS: rmdir(%s/%ld), %pd\n", + dfprintk(VFS, "NFS: rmdir(%s/%lu), %pd\n", dir->i_sb->s_id, dir->i_ino, dentry); trace_nfs_rmdir_enter(dir, dentry); @@ -1747,7 +1789,7 @@ int nfs_unlink(struct inode *dir, struct dentry *dentry) int error; int need_rehash = 0; - dfprintk(VFS, "NFS: unlink(%s/%ld, %pd)\n", dir->i_sb->s_id, + dfprintk(VFS, "NFS: unlink(%s/%lu, %pd)\n", dir->i_sb->s_id, dir->i_ino, dentry); trace_nfs_unlink_enter(dir, dentry); @@ -1798,7 +1840,7 @@ int nfs_symlink(struct inode *dir, struct dentry *dentry, const char *symname) unsigned int pathlen = strlen(symname); int error; - dfprintk(VFS, "NFS: symlink(%s/%ld, %pd, %s)\n", dir->i_sb->s_id, + dfprintk(VFS, "NFS: symlink(%s/%lu, %pd, %s)\n", dir->i_sb->s_id, dir->i_ino, dentry, symname); if (pathlen > PAGE_SIZE) @@ -1821,7 +1863,7 @@ int nfs_symlink(struct inode *dir, struct dentry *dentry, const char *symname) error = NFS_PROTO(dir)->symlink(dir, dentry, page, pathlen, &attr); trace_nfs_symlink_exit(dir, dentry, error); if (error != 0) { - dfprintk(VFS, "NFS: symlink(%s/%ld, %pd, %s) error %d\n", + 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); @@ -1837,6 +1879,11 @@ int nfs_symlink(struct inode *dir, struct dentry *dentry, const char *symname) 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. + */ + page_cache_release(page); } else __free_page(page); @@ -1897,6 +1944,7 @@ int nfs_rename(struct inode *old_dir, struct dentry *old_dentry, struct inode *old_inode = old_dentry->d_inode; struct inode *new_inode = new_dentry->d_inode; struct dentry *dentry = NULL, *rehash = NULL; + struct rpc_task *task; int error = -EBUSY; dfprintk(VFS, "NFS: rename(%pd2 -> %pd2, ct=%d)\n", @@ -1944,8 +1992,16 @@ int nfs_rename(struct inode *old_dir, struct dentry *old_dentry, if (new_inode != NULL) NFS_PROTO(new_inode)->return_delegation(new_inode); - error = NFS_PROTO(old_dir)->rename(old_dir, &old_dentry->d_name, - new_dir, &new_dentry->d_name); + task = nfs_async_rename(old_dir, new_dir, old_dentry, new_dentry, NULL); + if (IS_ERR(task)) { + error = PTR_ERR(task); + goto out; + } + + error = rpc_wait_for_completion_task(task); + if (error == 0) + error = task->tk_status; + rpc_put_task(task); nfs_mark_for_revalidate(old_inode); out: if (rehash) @@ -1976,9 +2032,9 @@ static void nfs_access_free_entry(struct nfs_access_entry *entry) { put_rpccred(entry->cred); kfree(entry); - smp_mb__before_atomic_dec(); + smp_mb__before_atomic(); atomic_long_dec(&nfs_access_nr_entries); - smp_mb__after_atomic_dec(); + smp_mb__after_atomic(); } static void nfs_access_free_list(struct list_head *head) @@ -2026,9 +2082,9 @@ nfs_access_cache_scan(struct shrinker *shrink, struct shrink_control *sc) else { remove_lru_entry: list_del_init(&nfsi->access_cache_inode_lru); - smp_mb__before_clear_bit(); + smp_mb__before_atomic(); clear_bit(NFS_INO_ACL_LRU_SET, &nfsi->flags); - smp_mb__after_clear_bit(); + smp_mb__after_atomic(); } spin_unlock(&inode->i_lock); } @@ -2176,9 +2232,9 @@ void nfs_access_add_cache(struct inode *inode, struct nfs_access_entry *set) nfs_access_add_rbtree(inode, cache); /* Update accounting */ - smp_mb__before_atomic_inc(); + smp_mb__before_atomic(); atomic_long_inc(&nfs_access_nr_entries); - smp_mb__after_atomic_inc(); + smp_mb__after_atomic(); /* Add inode to global LRU list */ if (!test_bit(NFS_INO_ACL_LRU_SET, &NFS_I(inode)->flags)) { @@ -2304,7 +2360,7 @@ out: if (!res && (mask & MAY_EXEC) && !execute_ok(inode)) res = -EACCES; - dfprintk(VFS, "NFS: permission(%s/%ld), mask=0x%x, res=%d\n", + dfprintk(VFS, "NFS: permission(%s/%lu), mask=0x%x, res=%d\n", inode->i_sb->s_id, inode->i_ino, mask, res); return res; out_notsup: diff --git a/fs/nfs/direct.c b/fs/nfs/direct.c index d71d66c9e0a..f11b9eed0de 100644 --- a/fs/nfs/direct.c +++ b/fs/nfs/direct.c @@ -108,6 +108,97 @@ static inline int put_dreq(struct nfs_direct_req *dreq) return atomic_dec_and_test(&dreq->io_count); } +/* + * nfs_direct_select_verf - select the right verifier + * @dreq - direct request possibly spanning multiple servers + * @ds_clp - nfs_client of data server or NULL if MDS / non-pnfs + * @ds_idx - index of data server in data server list, only valid if ds_clp set + * + * returns the correct verifier to use given the role of the server + */ +static struct nfs_writeverf * +nfs_direct_select_verf(struct nfs_direct_req *dreq, + struct nfs_client *ds_clp, + int ds_idx) +{ + struct nfs_writeverf *verfp = &dreq->verf; + +#ifdef CONFIG_NFS_V4_1 + if (ds_clp) { + /* pNFS is in use, use the DS verf */ + if (ds_idx >= 0 && ds_idx < dreq->ds_cinfo.nbuckets) + verfp = &dreq->ds_cinfo.buckets[ds_idx].direct_verf; + else + WARN_ON_ONCE(1); + } +#endif + return verfp; +} + + +/* + * nfs_direct_set_hdr_verf - set the write/commit verifier + * @dreq - direct request possibly spanning multiple servers + * @hdr - pageio header to validate against previously seen verfs + * + * Set the server's (MDS or DS) "seen" verifier + */ +static void nfs_direct_set_hdr_verf(struct nfs_direct_req *dreq, + struct nfs_pgio_header *hdr) +{ + struct nfs_writeverf *verfp; + + verfp = nfs_direct_select_verf(dreq, hdr->data->ds_clp, + hdr->data->ds_idx); + WARN_ON_ONCE(verfp->committed >= 0); + memcpy(verfp, &hdr->verf, sizeof(struct nfs_writeverf)); + WARN_ON_ONCE(verfp->committed < 0); +} + +/* + * nfs_direct_cmp_hdr_verf - compare verifier for pgio header + * @dreq - direct request possibly spanning multiple servers + * @hdr - pageio header to validate against previously seen verf + * + * set the server's "seen" verf if not initialized. + * returns result of comparison between @hdr->verf and the "seen" + * verf of the server used by @hdr (DS or MDS) + */ +static int nfs_direct_set_or_cmp_hdr_verf(struct nfs_direct_req *dreq, + struct nfs_pgio_header *hdr) +{ + struct nfs_writeverf *verfp; + + verfp = nfs_direct_select_verf(dreq, hdr->data->ds_clp, + hdr->data->ds_idx); + if (verfp->committed < 0) { + nfs_direct_set_hdr_verf(dreq, hdr); + return 0; + } + return memcmp(verfp, &hdr->verf, sizeof(struct nfs_writeverf)); +} + +#if IS_ENABLED(CONFIG_NFS_V3) || IS_ENABLED(CONFIG_NFS_V4) +/* + * nfs_direct_cmp_commit_data_verf - compare verifier for commit data + * @dreq - direct request possibly spanning multiple servers + * @data - commit data to validate against previously seen verf + * + * returns result of comparison between @data->verf and the verf of + * the server used by @data (DS or MDS) + */ +static int nfs_direct_cmp_commit_data_verf(struct nfs_direct_req *dreq, + struct nfs_commit_data *data) +{ + struct nfs_writeverf *verfp; + + verfp = nfs_direct_select_verf(dreq, data->ds_clp, + data->ds_commit_index); + WARN_ON_ONCE(verfp->committed < 0); + return memcmp(verfp, &data->verf, sizeof(struct nfs_writeverf)); +} +#endif + /** * nfs_direct_IO - NFS address space operation for direct I/O * @rw: direction (read or write) @@ -121,20 +212,20 @@ static inline int put_dreq(struct nfs_direct_req *dreq) * shunt off direct read and write requests before the VFS gets them, * so this method is only ever called for swap. */ -ssize_t nfs_direct_IO(int rw, struct kiocb *iocb, const struct iovec *iov, loff_t pos, unsigned long nr_segs) +ssize_t nfs_direct_IO(int rw, struct kiocb *iocb, struct iov_iter *iter, loff_t pos) { #ifndef CONFIG_NFS_SWAP dprintk("NFS: nfs_direct_IO (%pD) off/no(%Ld/%lu) EINVAL\n", - iocb->ki_filp, (long long) pos, nr_segs); + iocb->ki_filp, (long long) pos, iter->nr_segs); return -EINVAL; #else VM_BUG_ON(iocb->ki_nbytes != PAGE_SIZE); if (rw == READ || rw == KERNEL_READ) - return nfs_file_direct_read(iocb, iov, nr_segs, pos, + return nfs_file_direct_read(iocb, iter, pos, rw == READ ? true : false); - return nfs_file_direct_write(iocb, iov, nr_segs, pos, + return nfs_file_direct_write(iocb, iter, pos, rw == WRITE ? true : false); #endif /* CONFIG_NFS_SWAP */ } @@ -168,6 +259,7 @@ static inline struct nfs_direct_req *nfs_direct_req_alloc(void) kref_get(&dreq->kref); init_completion(&dreq->completion); INIT_LIST_HEAD(&dreq->mds_cinfo.list); + dreq->verf.committed = NFS_INVALID_STABLE_HOW; /* not set yet */ INIT_WORK(&dreq->work, nfs_direct_write_schedule_work); spin_lock_init(&dreq->lock); @@ -222,14 +314,31 @@ out: * Synchronous I/O uses a stack-allocated iocb. Thus we can't trust * the iocb is still valid here if this is a synchronous request. */ -static void nfs_direct_complete(struct nfs_direct_req *dreq) +static void nfs_direct_complete(struct nfs_direct_req *dreq, bool write) { + struct inode *inode = dreq->inode; + + if (dreq->iocb && write) { + loff_t pos = dreq->iocb->ki_pos + dreq->count; + + spin_lock(&inode->i_lock); + if (i_size_read(inode) < pos) + i_size_write(inode, pos); + spin_unlock(&inode->i_lock); + } + + if (write) + nfs_zap_mapping(inode, inode->i_mapping); + + inode_dio_done(inode); + if (dreq->iocb) { long res = (long) dreq->error; if (!res) res = (long) dreq->count; aio_complete(dreq->iocb, res, 0); } + complete_all(&dreq->completion); nfs_direct_req_release(dreq); @@ -237,9 +346,9 @@ static void nfs_direct_complete(struct nfs_direct_req *dreq) static void nfs_direct_readpage_release(struct nfs_page *req) { - dprintk("NFS: direct read done (%s/%lld %d@%lld)\n", + dprintk("NFS: direct read done (%s/%llu %d@%lld)\n", req->wb_context->dentry->d_inode->i_sb->s_id, - (long long)NFS_FILEID(req->wb_context->dentry->d_inode), + (unsigned long long)NFS_FILEID(req->wb_context->dentry->d_inode), req->wb_bytes, (long long)req_offset(req)); nfs_release_request(req); @@ -272,7 +381,7 @@ static void nfs_direct_read_completion(struct nfs_pgio_header *hdr) } out_put: if (put_dreq(dreq)) - nfs_direct_complete(dreq); + nfs_direct_complete(dreq, false); hdr->release(hdr); } @@ -305,66 +414,42 @@ static const struct nfs_pgio_completion_ops nfs_direct_read_completion_ops = { * handled automatically by nfs_direct_read_result(). Otherwise, if * no requests have been sent, just return an error. */ -static ssize_t nfs_direct_read_schedule_segment(struct nfs_pageio_descriptor *desc, - const struct iovec *iov, - loff_t pos, bool uio) -{ - struct nfs_direct_req *dreq = desc->pg_dreq; - struct nfs_open_context *ctx = dreq->ctx; - struct inode *inode = ctx->dentry->d_inode; - unsigned long user_addr = (unsigned long)iov->iov_base; - size_t count = iov->iov_len; - size_t rsize = NFS_SERVER(inode)->rsize; - unsigned int pgbase; - int result; - ssize_t started = 0; - struct page **pagevec = NULL; - unsigned int npages; - - do { - size_t bytes; - int i; - pgbase = user_addr & ~PAGE_MASK; - bytes = min(max_t(size_t, rsize, PAGE_SIZE), count); +static ssize_t nfs_direct_read_schedule_iovec(struct nfs_direct_req *dreq, + struct iov_iter *iter, + loff_t pos) +{ + struct nfs_pageio_descriptor desc; + struct inode *inode = dreq->inode; + ssize_t result = -EINVAL; + size_t requested_bytes = 0; + size_t rsize = max_t(size_t, NFS_SERVER(inode)->rsize, PAGE_SIZE); - result = -ENOMEM; - npages = nfs_page_array_len(pgbase, bytes); - if (!pagevec) - pagevec = kmalloc(npages * sizeof(struct page *), - GFP_KERNEL); - if (!pagevec) - break; - if (uio) { - down_read(¤t->mm->mmap_sem); - result = get_user_pages(current, current->mm, user_addr, - npages, 1, 0, pagevec, NULL); - up_read(¤t->mm->mmap_sem); - if (result < 0) - break; - } else { - WARN_ON(npages != 1); - result = get_kernel_page(user_addr, 1, pagevec); - if (WARN_ON(result != 1)) - break; - } + nfs_pageio_init_read(&desc, dreq->inode, false, + &nfs_direct_read_completion_ops); + get_dreq(dreq); + desc.pg_dreq = dreq; + atomic_inc(&inode->i_dio_count); - if ((unsigned)result < npages) { - bytes = result * PAGE_SIZE; - if (bytes <= pgbase) { - nfs_direct_release_pages(pagevec, result); - break; - } - bytes -= pgbase; - npages = result; - } + while (iov_iter_count(iter)) { + struct page **pagevec; + size_t bytes; + size_t pgbase; + unsigned npages, i; + result = iov_iter_get_pages_alloc(iter, &pagevec, + rsize, &pgbase); + if (result < 0) + break; + + bytes = result; + iov_iter_advance(iter, bytes); + npages = (result + pgbase + PAGE_SIZE - 1) / PAGE_SIZE; for (i = 0; i < npages; i++) { struct nfs_page *req; unsigned int req_len = min_t(size_t, bytes, PAGE_SIZE - pgbase); /* XXX do we need to do the eof zeroing found in async_filler? */ - req = nfs_create_request(dreq->ctx, dreq->inode, - pagevec[i], + req = nfs_create_request(dreq->ctx, pagevec[i], NULL, pgbase, req_len); if (IS_ERR(req)) { result = PTR_ERR(req); @@ -372,54 +457,21 @@ static ssize_t nfs_direct_read_schedule_segment(struct nfs_pageio_descriptor *de } req->wb_index = pos >> PAGE_SHIFT; req->wb_offset = pos & ~PAGE_MASK; - if (!nfs_pageio_add_request(desc, req)) { - result = desc->pg_error; + if (!nfs_pageio_add_request(&desc, req)) { + result = desc.pg_error; nfs_release_request(req); break; } pgbase = 0; bytes -= req_len; - started += req_len; - user_addr += req_len; + requested_bytes += req_len; pos += req_len; - count -= req_len; dreq->bytes_left -= req_len; } - /* The nfs_page now hold references to these pages */ nfs_direct_release_pages(pagevec, npages); - } while (count != 0 && result >= 0); - - kfree(pagevec); - - if (started) - return started; - return result < 0 ? (ssize_t) result : -EFAULT; -} - -static ssize_t nfs_direct_read_schedule_iovec(struct nfs_direct_req *dreq, - const struct iovec *iov, - unsigned long nr_segs, - loff_t pos, bool uio) -{ - struct nfs_pageio_descriptor desc; - ssize_t result = -EINVAL; - size_t requested_bytes = 0; - unsigned long seg; - - NFS_PROTO(dreq->inode)->read_pageio_init(&desc, dreq->inode, - &nfs_direct_read_completion_ops); - get_dreq(dreq); - desc.pg_dreq = dreq; - - for (seg = 0; seg < nr_segs; seg++) { - const struct iovec *vec = &iov[seg]; - result = nfs_direct_read_schedule_segment(&desc, vec, pos, uio); + kvfree(pagevec); if (result < 0) break; - requested_bytes += result; - if ((size_t)result < vec->iov_len) - break; - pos += vec->iov_len; } nfs_pageio_complete(&desc); @@ -429,29 +481,69 @@ static ssize_t nfs_direct_read_schedule_iovec(struct nfs_direct_req *dreq, * generic layer handle the completion. */ if (requested_bytes == 0) { + inode_dio_done(inode); nfs_direct_req_release(dreq); return result < 0 ? result : -EIO; } if (put_dreq(dreq)) - nfs_direct_complete(dreq); + nfs_direct_complete(dreq, false); return 0; } -static ssize_t nfs_direct_read(struct kiocb *iocb, const struct iovec *iov, - unsigned long nr_segs, loff_t pos, bool uio) +/** + * nfs_file_direct_read - file direct read operation for NFS files + * @iocb: target I/O control block + * @iter: vector of user buffers into which to read data + * @pos: byte offset in file where reading starts + * + * We use this function for direct reads instead of calling + * generic_file_aio_read() in order to avoid gfar's check to see if + * the request starts before the end of the file. For that check + * to work, we must generate a GETATTR before each direct read, and + * even then there is a window between the GETATTR and the subsequent + * READ where the file size could change. Our preference is simply + * to do all reads the application wants, and the server will take + * care of managing the end of file boundary. + * + * This function also eliminates unnecessarily updating the file's + * atime locally, as the NFS server sets the file's atime, and this + * client must read the updated atime from the server back into its + * cache. + */ +ssize_t nfs_file_direct_read(struct kiocb *iocb, struct iov_iter *iter, + loff_t pos, bool uio) { - ssize_t result = -ENOMEM; - struct inode *inode = iocb->ki_filp->f_mapping->host; + struct file *file = iocb->ki_filp; + struct address_space *mapping = file->f_mapping; + struct inode *inode = mapping->host; struct nfs_direct_req *dreq; struct nfs_lock_context *l_ctx; + ssize_t result = -EINVAL; + size_t count = iov_iter_count(iter); + nfs_add_stats(mapping->host, NFSIOS_DIRECTREADBYTES, count); + + dfprintk(FILE, "NFS: direct read(%pD2, %zd@%Ld)\n", + file, count, (long long) pos); + + result = 0; + if (!count) + goto out; + mutex_lock(&inode->i_mutex); + result = nfs_sync_mapping(mapping); + if (result) + goto out_unlock; + + task_io_account_read(count); + + result = -ENOMEM; dreq = nfs_direct_req_alloc(); if (dreq == NULL) - goto out; + goto out_unlock; dreq->inode = inode; - dreq->bytes_left = iov_length(iov, nr_segs); + dreq->bytes_left = count; dreq->ctx = get_nfs_open_context(nfs_file_open_context(iocb->ki_filp)); l_ctx = nfs_get_lock_context(dreq->ctx); if (IS_ERR(l_ctx)) { @@ -462,22 +554,28 @@ static ssize_t nfs_direct_read(struct kiocb *iocb, const struct iovec *iov, if (!is_sync_kiocb(iocb)) dreq->iocb = iocb; - NFS_I(inode)->read_io += iov_length(iov, nr_segs); - result = nfs_direct_read_schedule_iovec(dreq, iov, nr_segs, pos, uio); - if (!result) + NFS_I(inode)->read_io += count; + result = nfs_direct_read_schedule_iovec(dreq, iter, pos); + + mutex_unlock(&inode->i_mutex); + + if (!result) { result = nfs_direct_wait(dreq); + if (result > 0) + iocb->ki_pos = pos + result; + } + + nfs_direct_req_release(dreq); + return result; + out_release: nfs_direct_req_release(dreq); +out_unlock: + mutex_unlock(&inode->i_mutex); out: return result; } -static void nfs_inode_dio_write_done(struct inode *inode) -{ - nfs_zap_mapping(inode, inode->i_mapping); - inode_dio_done(inode); -} - #if IS_ENABLED(CONFIG_NFS_V3) || IS_ENABLED(CONFIG_NFS_V4) static void nfs_direct_write_reschedule(struct nfs_direct_req *dreq) { @@ -496,7 +594,7 @@ static void nfs_direct_write_reschedule(struct nfs_direct_req *dreq) dreq->count = 0; get_dreq(dreq); - NFS_PROTO(dreq->inode)->write_pageio_init(&desc, dreq->inode, FLUSH_STABLE, + nfs_pageio_init_write(&desc, dreq->inode, FLUSH_STABLE, false, &nfs_direct_write_completion_ops); desc.pg_dreq = dreq; @@ -535,7 +633,7 @@ static void nfs_direct_commit_complete(struct nfs_commit_data *data) dprintk("NFS: %5u commit failed with error %d.\n", data->task.tk_pid, status); dreq->flags = NFS_ODIRECT_RESCHED_WRITES; - } else if (memcmp(&dreq->verf, &data->verf, sizeof(data->verf))) { + } else if (nfs_direct_cmp_commit_data_verf(dreq, data)) { dprintk("NFS: %5u commit verify failed\n", data->task.tk_pid); dreq->flags = NFS_ODIRECT_RESCHED_WRITES; } @@ -593,8 +691,7 @@ static void nfs_direct_write_schedule_work(struct work_struct *work) nfs_direct_write_reschedule(dreq); break; default: - nfs_inode_dio_write_done(dreq->inode); - nfs_direct_complete(dreq); + nfs_direct_complete(dreq, true); } } @@ -610,114 +707,10 @@ static void nfs_direct_write_schedule_work(struct work_struct *work) static void nfs_direct_write_complete(struct nfs_direct_req *dreq, struct inode *inode) { - nfs_inode_dio_write_done(inode); - nfs_direct_complete(dreq); + nfs_direct_complete(dreq, true); } #endif -/* - * NB: Return the value of the first error return code. Subsequent - * errors after the first one are ignored. - */ -/* - * For each wsize'd chunk of the user's buffer, dispatch an NFS WRITE - * operation. If nfs_writedata_alloc() or get_user_pages() fails, - * bail and stop sending more writes. Write length accounting is - * handled automatically by nfs_direct_write_result(). Otherwise, if - * no requests have been sent, just return an error. - */ -static ssize_t nfs_direct_write_schedule_segment(struct nfs_pageio_descriptor *desc, - const struct iovec *iov, - loff_t pos, bool uio) -{ - struct nfs_direct_req *dreq = desc->pg_dreq; - struct nfs_open_context *ctx = dreq->ctx; - struct inode *inode = ctx->dentry->d_inode; - unsigned long user_addr = (unsigned long)iov->iov_base; - size_t count = iov->iov_len; - size_t wsize = NFS_SERVER(inode)->wsize; - unsigned int pgbase; - int result; - ssize_t started = 0; - struct page **pagevec = NULL; - unsigned int npages; - - do { - size_t bytes; - int i; - - pgbase = user_addr & ~PAGE_MASK; - bytes = min(max_t(size_t, wsize, PAGE_SIZE), count); - - result = -ENOMEM; - npages = nfs_page_array_len(pgbase, bytes); - if (!pagevec) - pagevec = kmalloc(npages * sizeof(struct page *), GFP_KERNEL); - if (!pagevec) - break; - - if (uio) { - down_read(¤t->mm->mmap_sem); - result = get_user_pages(current, current->mm, user_addr, - npages, 0, 0, pagevec, NULL); - up_read(¤t->mm->mmap_sem); - if (result < 0) - break; - } else { - WARN_ON(npages != 1); - result = get_kernel_page(user_addr, 0, pagevec); - if (WARN_ON(result != 1)) - break; - } - - if ((unsigned)result < npages) { - bytes = result * PAGE_SIZE; - if (bytes <= pgbase) { - nfs_direct_release_pages(pagevec, result); - break; - } - bytes -= pgbase; - npages = result; - } - - for (i = 0; i < npages; i++) { - struct nfs_page *req; - unsigned int req_len = min_t(size_t, bytes, PAGE_SIZE - pgbase); - - req = nfs_create_request(dreq->ctx, dreq->inode, - pagevec[i], - pgbase, req_len); - if (IS_ERR(req)) { - result = PTR_ERR(req); - break; - } - nfs_lock_request(req); - req->wb_index = pos >> PAGE_SHIFT; - req->wb_offset = pos & ~PAGE_MASK; - if (!nfs_pageio_add_request(desc, req)) { - result = desc->pg_error; - nfs_unlock_and_release_request(req); - break; - } - pgbase = 0; - bytes -= req_len; - started += req_len; - user_addr += req_len; - pos += req_len; - count -= req_len; - dreq->bytes_left -= req_len; - } - /* The nfs_page now hold references to these pages */ - nfs_direct_release_pages(pagevec, npages); - } while (count != 0 && result >= 0); - - kfree(pagevec); - - if (started) - return started; - return result < 0 ? (ssize_t) result : -EFAULT; -} - static void nfs_direct_write_completion(struct nfs_pgio_header *hdr) { struct nfs_direct_req *dreq = hdr->dreq; @@ -747,13 +740,13 @@ static void nfs_direct_write_completion(struct nfs_pgio_header *hdr) if (dreq->flags == NFS_ODIRECT_RESCHED_WRITES) bit = NFS_IOHDR_NEED_RESCHED; else if (dreq->flags == 0) { - memcpy(&dreq->verf, hdr->verf, - sizeof(dreq->verf)); + nfs_direct_set_hdr_verf(dreq, hdr); bit = NFS_IOHDR_NEED_COMMIT; dreq->flags = NFS_ODIRECT_DO_COMMIT; } else if (dreq->flags == NFS_ODIRECT_DO_COMMIT) { - if (memcmp(&dreq->verf, hdr->verf, sizeof(dreq->verf))) { - dreq->flags = NFS_ODIRECT_RESCHED_WRITES; + if (nfs_direct_set_or_cmp_hdr_verf(dreq, hdr)) { + dreq->flags = + NFS_ODIRECT_RESCHED_WRITES; bit = NFS_IOHDR_NEED_RESCHED; } else bit = NFS_IOHDR_NEED_COMMIT; @@ -763,6 +756,7 @@ static void nfs_direct_write_completion(struct nfs_pgio_header *hdr) spin_unlock(&dreq->lock); while (!list_empty(&hdr->pages)) { + req = nfs_list_entry(hdr->pages.next); nfs_list_remove_request(req); switch (bit) { @@ -797,33 +791,77 @@ static const struct nfs_pgio_completion_ops nfs_direct_write_completion_ops = { .completion = nfs_direct_write_completion, }; + +/* + * NB: Return the value of the first error return code. Subsequent + * errors after the first one are ignored. + */ +/* + * For each wsize'd chunk of the user's buffer, dispatch an NFS WRITE + * operation. If nfs_writedata_alloc() or get_user_pages() fails, + * bail and stop sending more writes. Write length accounting is + * handled automatically by nfs_direct_write_result(). Otherwise, if + * no requests have been sent, just return an error. + */ static ssize_t nfs_direct_write_schedule_iovec(struct nfs_direct_req *dreq, - const struct iovec *iov, - unsigned long nr_segs, - loff_t pos, bool uio) + struct iov_iter *iter, + loff_t pos) { struct nfs_pageio_descriptor desc; struct inode *inode = dreq->inode; ssize_t result = 0; size_t requested_bytes = 0; - unsigned long seg; + size_t wsize = max_t(size_t, NFS_SERVER(inode)->wsize, PAGE_SIZE); - NFS_PROTO(inode)->write_pageio_init(&desc, inode, FLUSH_COND_STABLE, + nfs_pageio_init_write(&desc, inode, FLUSH_COND_STABLE, false, &nfs_direct_write_completion_ops); desc.pg_dreq = dreq; get_dreq(dreq); atomic_inc(&inode->i_dio_count); - NFS_I(dreq->inode)->write_io += iov_length(iov, nr_segs); - for (seg = 0; seg < nr_segs; seg++) { - const struct iovec *vec = &iov[seg]; - result = nfs_direct_write_schedule_segment(&desc, vec, pos, uio); + NFS_I(inode)->write_io += iov_iter_count(iter); + while (iov_iter_count(iter)) { + struct page **pagevec; + size_t bytes; + size_t pgbase; + unsigned npages, i; + + result = iov_iter_get_pages_alloc(iter, &pagevec, + wsize, &pgbase); if (result < 0) break; - requested_bytes += result; - if ((size_t)result < vec->iov_len) + + bytes = result; + iov_iter_advance(iter, bytes); + npages = (result + pgbase + PAGE_SIZE - 1) / PAGE_SIZE; + for (i = 0; i < npages; i++) { + struct nfs_page *req; + unsigned int req_len = min_t(size_t, bytes, PAGE_SIZE - pgbase); + + req = nfs_create_request(dreq->ctx, pagevec[i], NULL, + pgbase, req_len); + if (IS_ERR(req)) { + result = PTR_ERR(req); + break; + } + nfs_lock_request(req); + req->wb_index = pos >> PAGE_SHIFT; + req->wb_offset = pos & ~PAGE_MASK; + if (!nfs_pageio_add_request(&desc, req)) { + result = desc.pg_error; + nfs_unlock_and_release_request(req); + break; + } + pgbase = 0; + bytes -= req_len; + requested_bytes += req_len; + pos += req_len; + dreq->bytes_left -= req_len; + } + nfs_direct_release_pages(pagevec, npages); + kvfree(pagevec); + if (result < 0) break; - pos += vec->iov_len; } nfs_pageio_complete(&desc); @@ -842,98 +880,10 @@ static ssize_t nfs_direct_write_schedule_iovec(struct nfs_direct_req *dreq, return 0; } -static ssize_t nfs_direct_write(struct kiocb *iocb, const struct iovec *iov, - unsigned long nr_segs, loff_t pos, - size_t count, bool uio) -{ - ssize_t result = -ENOMEM; - struct inode *inode = iocb->ki_filp->f_mapping->host; - struct nfs_direct_req *dreq; - struct nfs_lock_context *l_ctx; - - dreq = nfs_direct_req_alloc(); - if (!dreq) - goto out; - - dreq->inode = inode; - dreq->bytes_left = count; - dreq->ctx = get_nfs_open_context(nfs_file_open_context(iocb->ki_filp)); - l_ctx = nfs_get_lock_context(dreq->ctx); - if (IS_ERR(l_ctx)) { - result = PTR_ERR(l_ctx); - goto out_release; - } - dreq->l_ctx = l_ctx; - if (!is_sync_kiocb(iocb)) - dreq->iocb = iocb; - - result = nfs_direct_write_schedule_iovec(dreq, iov, nr_segs, pos, uio); - if (!result) - result = nfs_direct_wait(dreq); -out_release: - nfs_direct_req_release(dreq); -out: - return result; -} - -/** - * nfs_file_direct_read - file direct read operation for NFS files - * @iocb: target I/O control block - * @iov: vector of user buffers into which to read data - * @nr_segs: size of iov vector - * @pos: byte offset in file where reading starts - * - * We use this function for direct reads instead of calling - * generic_file_aio_read() in order to avoid gfar's check to see if - * the request starts before the end of the file. For that check - * to work, we must generate a GETATTR before each direct read, and - * even then there is a window between the GETATTR and the subsequent - * READ where the file size could change. Our preference is simply - * to do all reads the application wants, and the server will take - * care of managing the end of file boundary. - * - * This function also eliminates unnecessarily updating the file's - * atime locally, as the NFS server sets the file's atime, and this - * client must read the updated atime from the server back into its - * cache. - */ -ssize_t nfs_file_direct_read(struct kiocb *iocb, const struct iovec *iov, - unsigned long nr_segs, loff_t pos, bool uio) -{ - ssize_t retval = -EINVAL; - struct file *file = iocb->ki_filp; - struct address_space *mapping = file->f_mapping; - size_t count; - - count = iov_length(iov, nr_segs); - nfs_add_stats(mapping->host, NFSIOS_DIRECTREADBYTES, count); - - dfprintk(FILE, "NFS: direct read(%pD2, %zd@%Ld)\n", - file, count, (long long) pos); - - retval = 0; - if (!count) - goto out; - - retval = nfs_sync_mapping(mapping); - if (retval) - goto out; - - task_io_account_read(count); - - retval = nfs_direct_read(iocb, iov, nr_segs, pos, uio); - if (retval > 0) - iocb->ki_pos = pos + retval; - -out: - return retval; -} - /** * nfs_file_direct_write - file direct write operation for NFS files * @iocb: target I/O control block - * @iov: vector of user buffers from which to write data - * @nr_segs: size of iov vector + * @iter: vector of user buffers from which to write data * @pos: byte offset in file where writing starts * * We use this function for direct writes instead of calling @@ -951,49 +901,97 @@ out: * Note that O_APPEND is not supported for NFS direct writes, as there * is no atomic O_APPEND write facility in the NFS protocol. */ -ssize_t nfs_file_direct_write(struct kiocb *iocb, const struct iovec *iov, - unsigned long nr_segs, loff_t pos, bool uio) +ssize_t nfs_file_direct_write(struct kiocb *iocb, struct iov_iter *iter, + loff_t pos, bool uio) { - ssize_t retval = -EINVAL; + ssize_t result = -EINVAL; struct file *file = iocb->ki_filp; struct address_space *mapping = file->f_mapping; - size_t count; + struct inode *inode = mapping->host; + struct nfs_direct_req *dreq; + struct nfs_lock_context *l_ctx; + loff_t end; + size_t count = iov_iter_count(iter); + end = (pos + count - 1) >> PAGE_CACHE_SHIFT; - count = iov_length(iov, nr_segs); nfs_add_stats(mapping->host, NFSIOS_DIRECTWRITTENBYTES, count); dfprintk(FILE, "NFS: direct write(%pD2, %zd@%Ld)\n", file, count, (long long) pos); - retval = generic_write_checks(file, &pos, &count, 0); - if (retval) + result = generic_write_checks(file, &pos, &count, 0); + if (result) goto out; - retval = -EINVAL; + result = -EINVAL; if ((ssize_t) count < 0) goto out; - retval = 0; + result = 0; if (!count) goto out; - retval = nfs_sync_mapping(mapping); - if (retval) - goto out; + mutex_lock(&inode->i_mutex); + + result = nfs_sync_mapping(mapping); + if (result) + goto out_unlock; + + if (mapping->nrpages) { + result = invalidate_inode_pages2_range(mapping, + pos >> PAGE_CACHE_SHIFT, end); + if (result) + goto out_unlock; + } task_io_account_write(count); - retval = nfs_direct_write(iocb, iov, nr_segs, pos, count, uio); - if (retval > 0) { - struct inode *inode = mapping->host; + result = -ENOMEM; + dreq = nfs_direct_req_alloc(); + if (!dreq) + goto out_unlock; - iocb->ki_pos = pos + retval; - spin_lock(&inode->i_lock); - if (i_size_read(inode) < iocb->ki_pos) - i_size_write(inode, iocb->ki_pos); - spin_unlock(&inode->i_lock); + dreq->inode = inode; + dreq->bytes_left = count; + dreq->ctx = get_nfs_open_context(nfs_file_open_context(iocb->ki_filp)); + l_ctx = nfs_get_lock_context(dreq->ctx); + if (IS_ERR(l_ctx)) { + result = PTR_ERR(l_ctx); + goto out_release; + } + dreq->l_ctx = l_ctx; + if (!is_sync_kiocb(iocb)) + dreq->iocb = iocb; + + result = nfs_direct_write_schedule_iovec(dreq, iter, pos); + + if (mapping->nrpages) { + invalidate_inode_pages2_range(mapping, + pos >> PAGE_CACHE_SHIFT, end); } + + mutex_unlock(&inode->i_mutex); + + if (!result) { + result = nfs_direct_wait(dreq); + if (result > 0) { + struct inode *inode = mapping->host; + + iocb->ki_pos = pos + result; + spin_lock(&inode->i_lock); + if (i_size_read(inode) < iocb->ki_pos) + i_size_write(inode, iocb->ki_pos); + spin_unlock(&inode->i_lock); + } + } + nfs_direct_req_release(dreq); + return result; + +out_release: + nfs_direct_req_release(dreq); +out_unlock: + mutex_unlock(&inode->i_mutex); out: - return retval; + return result; } /** diff --git a/fs/nfs/dns_resolve.c b/fs/nfs/dns_resolve.c index fc0f95ec735..d25f10fb492 100644 --- a/fs/nfs/dns_resolve.c +++ b/fs/nfs/dns_resolve.c @@ -46,7 +46,9 @@ ssize_t nfs_dns_resolve_name(struct net *net, char *name, size_t namelen, #include <linux/sunrpc/cache.h> #include <linux/sunrpc/svcauth.h> #include <linux/sunrpc/rpc_pipe_fs.h> +#include <linux/nfs_fs.h> +#include "nfs4_fs.h" #include "dns_resolve.h" #include "cache_lib.h" #include "netns.h" diff --git a/fs/nfs/file.c b/fs/nfs/file.c index e2fcacf07de..4042ff58fe3 100644 --- a/fs/nfs/file.c +++ b/fs/nfs/file.c @@ -165,22 +165,21 @@ nfs_file_flush(struct file *file, fl_owner_t id) EXPORT_SYMBOL_GPL(nfs_file_flush); ssize_t -nfs_file_read(struct kiocb *iocb, const struct iovec *iov, - unsigned long nr_segs, loff_t pos) +nfs_file_read(struct kiocb *iocb, struct iov_iter *to) { struct inode *inode = file_inode(iocb->ki_filp); ssize_t result; if (iocb->ki_filp->f_flags & O_DIRECT) - return nfs_file_direct_read(iocb, iov, nr_segs, pos, true); + return nfs_file_direct_read(iocb, to, iocb->ki_pos, true); - dprintk("NFS: read(%pD2, %lu@%lu)\n", + dprintk("NFS: read(%pD2, %zu@%lu)\n", iocb->ki_filp, - (unsigned long) iov_length(iov, nr_segs), (unsigned long) pos); + iov_iter_count(to), (unsigned long) iocb->ki_pos); result = nfs_revalidate_mapping(inode, iocb->ki_filp->f_mapping); if (!result) { - result = generic_file_aio_read(iocb, iov, nr_segs, pos); + result = generic_file_read_iter(iocb, to); if (result > 0) nfs_add_stats(inode, NFSIOS_NORMALREADBYTES, result); } @@ -354,7 +353,7 @@ static int nfs_write_begin(struct file *file, struct address_space *mapping, struct page *page; int once_thru = 0; - dfprintk(PAGECACHE, "NFS: write_begin(%pD2(%ld), %u@%lld)\n", + dfprintk(PAGECACHE, "NFS: write_begin(%pD2(%lu), %u@%lld)\n", file, mapping->host->i_ino, len, (long long) pos); start: @@ -395,7 +394,7 @@ static int nfs_write_end(struct file *file, struct address_space *mapping, struct nfs_open_context *ctx = nfs_file_open_context(file); int status; - dfprintk(PAGECACHE, "NFS: write_end(%pD2(%ld), %u@%lld)\n", + dfprintk(PAGECACHE, "NFS: write_end(%pD2(%lu), %u@%lld)\n", file, mapping->host->i_ino, len, (long long) pos); /* @@ -585,7 +584,7 @@ static int nfs_vm_page_mkwrite(struct vm_area_struct *vma, struct vm_fault *vmf) int ret = VM_FAULT_NOPAGE; struct address_space *mapping; - dfprintk(PAGECACHE, "NFS: vm_page_mkwrite(%pD2(%ld), offset %lld)\n", + dfprintk(PAGECACHE, "NFS: vm_page_mkwrite(%pD2(%lu), offset %lld)\n", filp, filp->f_mapping->host->i_ino, (long long)page_offset(page)); @@ -617,6 +616,7 @@ out: static const struct vm_operations_struct nfs_file_vm_ops = { .fault = filemap_fault, + .map_pages = filemap_map_pages, .page_mkwrite = nfs_vm_page_mkwrite, .remap_pages = generic_file_remap_pages, }; @@ -634,24 +634,24 @@ static int nfs_need_sync_write(struct file *filp, struct inode *inode) return 0; } -ssize_t nfs_file_write(struct kiocb *iocb, const struct iovec *iov, - unsigned long nr_segs, loff_t pos) +ssize_t nfs_file_write(struct kiocb *iocb, struct iov_iter *from) { struct file *file = iocb->ki_filp; struct inode *inode = file_inode(file); unsigned long written = 0; ssize_t result; - size_t count = iov_length(iov, nr_segs); + size_t count = iov_iter_count(from); + loff_t pos = iocb->ki_pos; result = nfs_key_timeout_notify(file, inode); if (result) return result; if (file->f_flags & O_DIRECT) - return nfs_file_direct_write(iocb, iov, nr_segs, pos, true); + return nfs_file_direct_write(iocb, from, pos, true); - dprintk("NFS: write(%pD2, %lu@%Ld)\n", - file, (unsigned long) count, (long long) pos); + dprintk("NFS: write(%pD2, %zu@%Ld)\n", + file, count, (long long) pos); result = -EBUSY; if (IS_SWAPFILE(inode)) @@ -669,7 +669,7 @@ ssize_t nfs_file_write(struct kiocb *iocb, const struct iovec *iov, if (!count) goto out; - result = generic_file_aio_write(iocb, iov, nr_segs, pos); + result = generic_file_write_iter(iocb, from); if (result > 0) written = result; @@ -690,36 +690,6 @@ out_swapfile: } EXPORT_SYMBOL_GPL(nfs_file_write); -ssize_t nfs_file_splice_write(struct pipe_inode_info *pipe, - struct file *filp, loff_t *ppos, - size_t count, unsigned int flags) -{ - struct inode *inode = file_inode(filp); - unsigned long written = 0; - ssize_t ret; - - dprintk("NFS splice_write(%pD2, %lu@%llu)\n", - filp, (unsigned long) count, (unsigned long long) *ppos); - - /* - * The combination of splice and an O_APPEND destination is disallowed. - */ - - ret = generic_file_splice_write(pipe, filp, ppos, count, flags); - if (ret > 0) - written = ret; - - if (ret >= 0 && nfs_need_sync_write(filp, inode)) { - int err = vfs_fsync(filp, 0); - if (err < 0) - ret = err; - } - if (ret > 0) - nfs_add_stats(inode, NFSIOS_NORMALWRITTENBYTES, written); - return ret; -} -EXPORT_SYMBOL_GPL(nfs_file_splice_write); - static int do_getlk(struct file *filp, int cmd, struct file_lock *fl, int is_local) { @@ -915,10 +885,6 @@ int nfs_flock(struct file *filp, int cmd, struct file_lock *fl) is_local = 1; /* We're simulating flock() locks using posix locks on the server */ - fl->fl_owner = (fl_owner_t)filp; - fl->fl_start = 0; - fl->fl_end = OFFSET_MAX; - if (fl->fl_type == F_UNLCK) return do_unlk(filp, cmd, fl, is_local); return do_setlk(filp, cmd, fl, is_local); @@ -938,10 +904,10 @@ EXPORT_SYMBOL_GPL(nfs_setlease); const struct file_operations nfs_file_operations = { .llseek = nfs_file_llseek, - .read = do_sync_read, - .write = do_sync_write, - .aio_read = nfs_file_read, - .aio_write = nfs_file_write, + .read = new_sync_read, + .write = new_sync_write, + .read_iter = nfs_file_read, + .write_iter = nfs_file_write, .mmap = nfs_file_mmap, .open = nfs_file_open, .flush = nfs_file_flush, @@ -950,7 +916,7 @@ const struct file_operations nfs_file_operations = { .lock = nfs_lock, .flock = nfs_flock, .splice_read = nfs_file_splice_read, - .splice_write = nfs_file_splice_write, + .splice_write = iter_file_splice_write, .check_flags = nfs_check_flags, .setlease = nfs_setlease, }; diff --git a/fs/nfs/filelayout/Makefile b/fs/nfs/filelayout/Makefile new file mode 100644 index 00000000000..8516cdffb9e --- /dev/null +++ b/fs/nfs/filelayout/Makefile @@ -0,0 +1,5 @@ +# +# Makefile for the pNFS Files Layout Driver kernel module +# +obj-$(CONFIG_PNFS_FILE_LAYOUT) += nfs_layout_nfsv41_files.o +nfs_layout_nfsv41_files-y := filelayout.o filelayoutdev.o diff --git a/fs/nfs/nfs4filelayout.c b/fs/nfs/filelayout/filelayout.c index b86464ba25e..d2eba1c13b7 100644 --- a/fs/nfs/nfs4filelayout.c +++ b/fs/nfs/filelayout/filelayout.c @@ -35,11 +35,11 @@ #include <linux/sunrpc/metrics.h> -#include "nfs4session.h" -#include "internal.h" -#include "delegation.h" -#include "nfs4filelayout.h" -#include "nfs4trace.h" +#include "../nfs4session.h" +#include "../internal.h" +#include "../delegation.h" +#include "filelayout.h" +#include "../nfs4trace.h" #define NFSDBG_FACILITY NFSDBG_PNFS_LD @@ -84,17 +84,17 @@ filelayout_get_dserver_offset(struct pnfs_layout_segment *lseg, loff_t offset) BUG(); } -static void filelayout_reset_write(struct nfs_write_data *data) +static void filelayout_reset_write(struct nfs_pgio_data *data) { struct nfs_pgio_header *hdr = data->header; struct rpc_task *task = &data->task; if (!test_and_set_bit(NFS_IOHDR_REDO, &hdr->flags)) { dprintk("%s Reset task %5u for i/o through MDS " - "(req %s/%lld, %u bytes @ offset %llu)\n", __func__, + "(req %s/%llu, %u bytes @ offset %llu)\n", __func__, data->task.tk_pid, hdr->inode->i_sb->s_id, - (long long)NFS_FILEID(hdr->inode), + (unsigned long long)NFS_FILEID(hdr->inode), data->args.count, (unsigned long long)data->args.offset); @@ -105,17 +105,17 @@ static void filelayout_reset_write(struct nfs_write_data *data) } } -static void filelayout_reset_read(struct nfs_read_data *data) +static void filelayout_reset_read(struct nfs_pgio_data *data) { struct nfs_pgio_header *hdr = data->header; struct rpc_task *task = &data->task; if (!test_and_set_bit(NFS_IOHDR_REDO, &hdr->flags)) { dprintk("%s Reset task %5u for i/o through MDS " - "(req %s/%lld, %u bytes @ offset %llu)\n", __func__, + "(req %s/%llu, %u bytes @ offset %llu)\n", __func__, data->task.tk_pid, hdr->inode->i_sb->s_id, - (long long)NFS_FILEID(hdr->inode), + (unsigned long long)NFS_FILEID(hdr->inode), data->args.count, (unsigned long long)data->args.offset); @@ -243,7 +243,7 @@ wait_on_recovery: /* NFS_PROTO call done callback routines */ static int filelayout_read_done_cb(struct rpc_task *task, - struct nfs_read_data *data) + struct nfs_pgio_data *data) { struct nfs_pgio_header *hdr = data->header; int err; @@ -270,7 +270,7 @@ static int filelayout_read_done_cb(struct rpc_task *task, * rfc5661 is not clear about which credential should be used. */ static void -filelayout_set_layoutcommit(struct nfs_write_data *wdata) +filelayout_set_layoutcommit(struct nfs_pgio_data *wdata) { struct nfs_pgio_header *hdr = wdata->header; @@ -279,7 +279,7 @@ filelayout_set_layoutcommit(struct nfs_write_data *wdata) return; pnfs_set_layoutcommit(wdata); - dprintk("%s ionde %lu pls_end_pos %lu\n", __func__, hdr->inode->i_ino, + dprintk("%s inode %lu pls_end_pos %lu\n", __func__, hdr->inode->i_ino, (unsigned long) NFS_I(hdr->inode)->layout->plh_lwb); } @@ -305,7 +305,7 @@ filelayout_reset_to_mds(struct pnfs_layout_segment *lseg) */ static void filelayout_read_prepare(struct rpc_task *task, void *data) { - struct nfs_read_data *rdata = data; + struct nfs_pgio_data *rdata = data; if (unlikely(test_bit(NFS_CONTEXT_BAD, &rdata->args.context->flags))) { rpc_exit(task, -EIO); @@ -317,26 +317,29 @@ static void filelayout_read_prepare(struct rpc_task *task, void *data) rpc_exit(task, 0); return; } - rdata->read_done_cb = filelayout_read_done_cb; + rdata->pgio_done_cb = filelayout_read_done_cb; if (nfs41_setup_sequence(rdata->ds_clp->cl_session, &rdata->args.seq_args, &rdata->res.seq_res, task)) return; - nfs4_set_rw_stateid(&rdata->args.stateid, rdata->args.context, - rdata->args.lock_context, FMODE_READ); + if (nfs4_set_rw_stateid(&rdata->args.stateid, rdata->args.context, + rdata->args.lock_context, FMODE_READ) == -EIO) + rpc_exit(task, -EIO); /* lost lock, terminate I/O */ } static void filelayout_read_call_done(struct rpc_task *task, void *data) { - struct nfs_read_data *rdata = data; + struct nfs_pgio_data *rdata = data; dprintk("--> %s task->tk_status %d\n", __func__, task->tk_status); if (test_bit(NFS_IOHDR_REDO, &rdata->header->flags) && - task->tk_status == 0) + task->tk_status == 0) { + nfs41_sequence_done(task, &rdata->res.seq_res); return; + } /* Note this may cause RPC to be resent */ rdata->header->mds_ops->rpc_call_done(task, data); @@ -344,14 +347,14 @@ static void filelayout_read_call_done(struct rpc_task *task, void *data) static void filelayout_read_count_stats(struct rpc_task *task, void *data) { - struct nfs_read_data *rdata = data; + struct nfs_pgio_data *rdata = data; rpc_count_iostats(task, NFS_SERVER(rdata->header->inode)->client->cl_metrics); } static void filelayout_read_release(void *data) { - struct nfs_read_data *rdata = data; + struct nfs_pgio_data *rdata = data; struct pnfs_layout_hdr *lo = rdata->header->lseg->pls_layout; filelayout_fenceme(lo->plh_inode, lo); @@ -360,7 +363,7 @@ static void filelayout_read_release(void *data) } static int filelayout_write_done_cb(struct rpc_task *task, - struct nfs_write_data *data) + struct nfs_pgio_data *data) { struct nfs_pgio_header *hdr = data->header; int err; @@ -416,7 +419,7 @@ static int filelayout_commit_done_cb(struct rpc_task *task, static void filelayout_write_prepare(struct rpc_task *task, void *data) { - struct nfs_write_data *wdata = data; + struct nfs_pgio_data *wdata = data; if (unlikely(test_bit(NFS_CONTEXT_BAD, &wdata->args.context->flags))) { rpc_exit(task, -EIO); @@ -433,17 +436,20 @@ static void filelayout_write_prepare(struct rpc_task *task, void *data) &wdata->res.seq_res, task)) return; - nfs4_set_rw_stateid(&wdata->args.stateid, wdata->args.context, - wdata->args.lock_context, FMODE_WRITE); + if (nfs4_set_rw_stateid(&wdata->args.stateid, wdata->args.context, + wdata->args.lock_context, FMODE_WRITE) == -EIO) + rpc_exit(task, -EIO); /* lost lock, terminate I/O */ } static void filelayout_write_call_done(struct rpc_task *task, void *data) { - struct nfs_write_data *wdata = data; + struct nfs_pgio_data *wdata = data; if (test_bit(NFS_IOHDR_REDO, &wdata->header->flags) && - task->tk_status == 0) + task->tk_status == 0) { + nfs41_sequence_done(task, &wdata->res.seq_res); return; + } /* Note this may cause RPC to be resent */ wdata->header->mds_ops->rpc_call_done(task, data); @@ -451,14 +457,14 @@ static void filelayout_write_call_done(struct rpc_task *task, void *data) static void filelayout_write_count_stats(struct rpc_task *task, void *data) { - struct nfs_write_data *wdata = data; + struct nfs_pgio_data *wdata = data; rpc_count_iostats(task, NFS_SERVER(wdata->header->inode)->client->cl_metrics); } static void filelayout_write_release(void *data) { - struct nfs_write_data *wdata = data; + struct nfs_pgio_data *wdata = data; struct pnfs_layout_hdr *lo = wdata->header->lseg->pls_layout; filelayout_fenceme(lo->plh_inode, lo); @@ -523,7 +529,7 @@ static const struct rpc_call_ops filelayout_commit_call_ops = { }; static enum pnfs_try_status -filelayout_read_pagelist(struct nfs_read_data *data) +filelayout_read_pagelist(struct nfs_pgio_data *data) { struct nfs_pgio_header *hdr = data->header; struct pnfs_layout_segment *lseg = hdr->lseg; @@ -554,6 +560,7 @@ filelayout_read_pagelist(struct nfs_read_data *data) /* No multipath support. Use first DS */ atomic_inc(&ds->ds_clp->cl_count); data->ds_clp = ds->ds_clp; + data->ds_idx = idx; fh = nfs4_fl_select_ds_fh(lseg, j); if (fh) data->args.fh = fh; @@ -562,14 +569,14 @@ filelayout_read_pagelist(struct nfs_read_data *data) data->mds_offset = offset; /* Perform an asynchronous read to ds */ - nfs_initiate_read(ds_clnt, data, - &filelayout_read_call_ops, RPC_TASK_SOFTCONN); + nfs_initiate_pgio(ds_clnt, data, + &filelayout_read_call_ops, 0, RPC_TASK_SOFTCONN); return PNFS_ATTEMPTED; } /* Perform async writes. */ static enum pnfs_try_status -filelayout_write_pagelist(struct nfs_write_data *data, int sync) +filelayout_write_pagelist(struct nfs_pgio_data *data, int sync) { struct nfs_pgio_header *hdr = data->header; struct pnfs_layout_segment *lseg = hdr->lseg; @@ -594,20 +601,18 @@ filelayout_write_pagelist(struct nfs_write_data *data, int sync) __func__, hdr->inode->i_ino, sync, (size_t) data->args.count, offset, ds->ds_remotestr, atomic_read(&ds->ds_clp->cl_count)); - data->write_done_cb = filelayout_write_done_cb; + data->pgio_done_cb = filelayout_write_done_cb; atomic_inc(&ds->ds_clp->cl_count); data->ds_clp = ds->ds_clp; + data->ds_idx = idx; fh = nfs4_fl_select_ds_fh(lseg, j); if (fh) data->args.fh = fh; - /* - * Get the file offset on the dserver. Set the write offset to - * this offset and save the original offset. - */ + data->args.offset = filelayout_get_dserver_offset(lseg, offset); /* Perform an asynchronous write */ - nfs_initiate_write(ds_clnt, data, + nfs_initiate_pgio(ds_clnt, data, &filelayout_write_call_ops, sync, RPC_TASK_SOFTCONN); return PNFS_ATTEMPTED; @@ -631,7 +636,6 @@ filelayout_check_layout(struct pnfs_layout_hdr *lo, struct nfs4_deviceid_node *d; struct nfs4_file_layout_dsaddr *dsaddr; int status = -EINVAL; - struct nfs_server *nfss = NFS_SERVER(lo->plh_inode); dprintk("--> %s\n", __func__); @@ -649,7 +653,7 @@ filelayout_check_layout(struct pnfs_layout_hdr *lo, goto out; } - if (!fl->stripe_unit || fl->stripe_unit % PAGE_SIZE) { + if (!fl->stripe_unit) { dprintk("%s Invalid stripe unit (%u)\n", __func__, fl->stripe_unit); goto out; @@ -686,12 +690,6 @@ filelayout_check_layout(struct pnfs_layout_hdr *lo, goto out_put; } - if (fl->stripe_unit % nfss->rsize || fl->stripe_unit % nfss->wsize) { - dprintk("%s Stripe unit (%u) not aligned with rsize %u " - "wsize %u\n", __func__, fl->stripe_unit, nfss->rsize, - nfss->wsize); - } - status = 0; out: dprintk("--> %s returns %d\n", __func__, status); @@ -844,11 +842,15 @@ filelayout_alloc_commit_info(struct pnfs_layout_segment *lseg, { struct nfs4_filelayout_segment *fl = FILELAYOUT_LSEG(lseg); struct pnfs_commit_bucket *buckets; - int size; + int size, i; if (fl->commit_through_mds) return 0; - if (cinfo->ds->nbuckets != 0) { + + size = (fl->stripe_type == STRIPE_SPARSE) ? + fl->dsaddr->ds_num : fl->dsaddr->stripe_count; + + if (cinfo->ds->nbuckets >= size) { /* This assumes there is only one IOMODE_RW lseg. What * we really want to do is have a layout_hdr level * dictionary of <multipath_list4, fh> keys, each @@ -858,30 +860,36 @@ filelayout_alloc_commit_info(struct pnfs_layout_segment *lseg, return 0; } - size = (fl->stripe_type == STRIPE_SPARSE) ? - fl->dsaddr->ds_num : fl->dsaddr->stripe_count; - buckets = kcalloc(size, sizeof(struct pnfs_commit_bucket), gfp_flags); if (!buckets) return -ENOMEM; - else { - int i; + for (i = 0; i < size; i++) { + INIT_LIST_HEAD(&buckets[i].written); + INIT_LIST_HEAD(&buckets[i].committing); + /* mark direct verifier as unset */ + buckets[i].direct_verf.committed = NFS_INVALID_STABLE_HOW; + } - spin_lock(cinfo->lock); - if (cinfo->ds->nbuckets != 0) - kfree(buckets); - else { - cinfo->ds->buckets = buckets; - cinfo->ds->nbuckets = size; - for (i = 0; i < size; i++) { - INIT_LIST_HEAD(&buckets[i].written); - INIT_LIST_HEAD(&buckets[i].committing); - } - } - spin_unlock(cinfo->lock); - return 0; + spin_lock(cinfo->lock); + if (cinfo->ds->nbuckets >= size) + goto out; + for (i = 0; i < cinfo->ds->nbuckets; i++) { + list_splice(&cinfo->ds->buckets[i].written, + &buckets[i].written); + list_splice(&cinfo->ds->buckets[i].committing, + &buckets[i].committing); + buckets[i].direct_verf.committed = + cinfo->ds->buckets[i].direct_verf.committed; + buckets[i].wlseg = cinfo->ds->buckets[i].wlseg; + buckets[i].clseg = cinfo->ds->buckets[i].clseg; } + swap(cinfo->ds->buckets, buckets); + cinfo->ds->nbuckets = size; +out: + spin_unlock(cinfo->lock); + kfree(buckets); + return 0; } static struct pnfs_layout_segment * @@ -909,47 +917,51 @@ filelayout_alloc_lseg(struct pnfs_layout_hdr *layoutid, /* * filelayout_pg_test(). Called by nfs_can_coalesce_requests() * - * return true : coalesce page - * return false : don't coalesce page + * Return 0 if @req cannot be coalesced into @pgio, otherwise return the number + * of bytes (maximum @req->wb_bytes) that can be coalesced. */ -static bool +static size_t filelayout_pg_test(struct nfs_pageio_descriptor *pgio, struct nfs_page *prev, struct nfs_page *req) { + unsigned int size; u64 p_stripe, r_stripe; - u32 stripe_unit; + u32 stripe_offset; + u64 segment_offset = pgio->pg_lseg->pls_range.offset; + u32 stripe_unit = FILELAYOUT_LSEG(pgio->pg_lseg)->stripe_unit; - if (!pnfs_generic_pg_test(pgio, prev, req) || - !nfs_generic_pg_test(pgio, prev, req)) - return false; + /* calls nfs_generic_pg_test */ + size = pnfs_generic_pg_test(pgio, prev, req); + if (!size) + return 0; - p_stripe = (u64)req_offset(prev); - r_stripe = (u64)req_offset(req); - stripe_unit = FILELAYOUT_LSEG(pgio->pg_lseg)->stripe_unit; + /* see if req and prev are in the same stripe */ + if (prev) { + p_stripe = (u64)req_offset(prev) - segment_offset; + r_stripe = (u64)req_offset(req) - segment_offset; + do_div(p_stripe, stripe_unit); + do_div(r_stripe, stripe_unit); - do_div(p_stripe, stripe_unit); - do_div(r_stripe, stripe_unit); + if (p_stripe != r_stripe) + return 0; + } - return (p_stripe == r_stripe); + /* calculate remaining bytes in the current stripe */ + div_u64_rem((u64)req_offset(req) - segment_offset, + stripe_unit, + &stripe_offset); + WARN_ON_ONCE(stripe_offset > stripe_unit); + if (stripe_offset >= stripe_unit) + return 0; + return min(stripe_unit - (unsigned int)stripe_offset, size); } static void filelayout_pg_init_read(struct nfs_pageio_descriptor *pgio, struct nfs_page *req) { - WARN_ON_ONCE(pgio->pg_lseg != NULL); - - if (req->wb_offset != req->wb_pgbase) { - /* - * Handling unaligned pages is difficult, because have to - * somehow split a req in two in certain cases in the - * pg.test code. Avoid this by just not using pnfs - * in this case. - */ - nfs_pageio_reset_read_mds(pgio); - return; - } - pgio->pg_lseg = pnfs_update_layout(pgio->pg_inode, + if (!pgio->pg_lseg) + pgio->pg_lseg = pnfs_update_layout(pgio->pg_inode, req->wb_context, 0, NFS4_MAX_UINT64, @@ -967,11 +979,8 @@ filelayout_pg_init_write(struct nfs_pageio_descriptor *pgio, struct nfs_commit_info cinfo; int status; - WARN_ON_ONCE(pgio->pg_lseg != NULL); - - if (req->wb_offset != req->wb_pgbase) - goto out_mds; - pgio->pg_lseg = pnfs_update_layout(pgio->pg_inode, + if (!pgio->pg_lseg) + pgio->pg_lseg = pnfs_update_layout(pgio->pg_inode, req->wb_context, 0, NFS4_MAX_UINT64, @@ -1061,6 +1070,7 @@ filelayout_choose_commit_list(struct nfs_page *req, */ j = nfs4_fl_calc_j_index(lseg, req_offset(req)); i = select_bucket_index(fl, j); + spin_lock(cinfo->lock); buckets = cinfo->ds->buckets; list = &buckets[i].written; if (list_empty(list)) { @@ -1074,6 +1084,7 @@ filelayout_choose_commit_list(struct nfs_page *req, } set_bit(PG_COMMIT_TO_DS, &req->wb_flags); cinfo->ds->nwritten++; + spin_unlock(cinfo->lock); return list; } @@ -1170,6 +1181,7 @@ transfer_commit_list(struct list_head *src, struct list_head *dst, return ret; } +/* Note called with cinfo->lock held. */ static int filelayout_scan_ds_commit_list(struct pnfs_commit_bucket *bucket, struct nfs_commit_info *cinfo, @@ -1214,19 +1226,22 @@ static void filelayout_recover_commit_reqs(struct list_head *dst, struct nfs_commit_info *cinfo) { struct pnfs_commit_bucket *b; + struct pnfs_layout_segment *freeme; int i; - /* NOTE cinfo->lock is NOT held, relying on fact that this is - * only called on single thread per dreq. - * Can't take the lock because need to do pnfs_put_lseg - */ +restart: + spin_lock(cinfo->lock); for (i = 0, b = cinfo->ds->buckets; i < cinfo->ds->nbuckets; i++, b++) { if (transfer_commit_list(&b->written, dst, cinfo, 0)) { - pnfs_put_lseg(b->wlseg); + freeme = b->wlseg; b->wlseg = NULL; + spin_unlock(cinfo->lock); + pnfs_put_lseg(freeme); + goto restart; } } cinfo->ds->nwritten = 0; + spin_unlock(cinfo->lock); } static unsigned int @@ -1237,6 +1252,7 @@ alloc_ds_commits(struct nfs_commit_info *cinfo, struct list_head *list) struct nfs_commit_data *data; int i, j; unsigned int nreq = 0; + struct pnfs_layout_segment *freeme; fl_cinfo = cinfo->ds; bucket = fl_cinfo->buckets; @@ -1247,8 +1263,10 @@ alloc_ds_commits(struct nfs_commit_info *cinfo, struct list_head *list) if (!data) break; data->ds_commit_index = i; + spin_lock(cinfo->lock); data->lseg = bucket->clseg; bucket->clseg = NULL; + spin_unlock(cinfo->lock); list_add(&data->pages, list); nreq++; } @@ -1258,8 +1276,11 @@ alloc_ds_commits(struct nfs_commit_info *cinfo, struct list_head *list) if (list_empty(&bucket->committing)) continue; nfs_retry_commit(&bucket->committing, bucket->clseg, cinfo); - pnfs_put_lseg(bucket->clseg); + spin_lock(cinfo->lock); + freeme = bucket->clseg; bucket->clseg = NULL; + spin_unlock(cinfo->lock); + pnfs_put_lseg(freeme); } /* Caller will clean up entries put on list */ return nreq; @@ -1324,7 +1345,7 @@ filelayout_alloc_layout_hdr(struct inode *inode, gfp_t gfp_flags) struct nfs4_filelayout *flo; flo = kzalloc(sizeof(*flo), gfp_flags); - return &flo->generic_hdr; + return flo != NULL ? &flo->generic_hdr : NULL; } static void diff --git a/fs/nfs/nfs4filelayout.h b/fs/nfs/filelayout/filelayout.h index cebd20e7e92..ffbddf2219e 100644 --- a/fs/nfs/nfs4filelayout.h +++ b/fs/nfs/filelayout/filelayout.h @@ -30,7 +30,7 @@ #ifndef FS_NFS_NFS4FILELAYOUT_H #define FS_NFS_NFS4FILELAYOUT_H -#include "pnfs.h" +#include "../pnfs.h" /* * Default data server connection timeout and retrans vaules. diff --git a/fs/nfs/nfs4filelayoutdev.c b/fs/nfs/filelayout/filelayoutdev.c index c7c295e556e..44bf0140a4c 100644 --- a/fs/nfs/nfs4filelayoutdev.c +++ b/fs/nfs/filelayout/filelayoutdev.c @@ -33,9 +33,9 @@ #include <linux/module.h> #include <linux/sunrpc/addr.h> -#include "internal.h" -#include "nfs4session.h" -#include "nfs4filelayout.h" +#include "../internal.h" +#include "../nfs4session.h" +#include "filelayout.h" #define NFSDBG_FACILITY NFSDBG_PNFS_LD @@ -95,7 +95,7 @@ same_sockaddr(struct sockaddr *addr1, struct sockaddr *addr2) b6 = (struct sockaddr_in6 *)addr2; /* LINKLOCAL addresses must have matching scope_id */ - if (ipv6_addr_scope(&a6->sin6_addr) == + if (ipv6_addr_src_scope(&a6->sin6_addr) == IPV6_ADDR_SCOPE_LINKLOCAL && a6->sin6_scope_id != b6->sin6_scope_id) return false; @@ -789,9 +789,9 @@ static void nfs4_wait_ds_connect(struct nfs4_pnfs_ds *ds) static void nfs4_clear_ds_conn_bit(struct nfs4_pnfs_ds *ds) { - smp_mb__before_clear_bit(); + smp_mb__before_atomic(); clear_bit(NFS4DS_CONNECTING, &ds->ds_state); - smp_mb__after_clear_bit(); + smp_mb__after_atomic(); wake_up_bit(&ds->ds_state, NFS4DS_CONNECTING); } diff --git a/fs/nfs/getroot.c b/fs/nfs/getroot.c index 66984a9aafa..b94f80420a5 100644 --- a/fs/nfs/getroot.c +++ b/fs/nfs/getroot.c @@ -120,7 +120,8 @@ struct dentry *nfs_get_root(struct super_block *sb, struct nfs_fh *mntfh, security_d_instantiate(ret, inode); spin_lock(&ret->d_lock); - if (IS_ROOT(ret) && !(ret->d_flags & DCACHE_NFSFS_RENAMED)) { + if (IS_ROOT(ret) && !ret->d_fsdata && + !(ret->d_flags & DCACHE_NFSFS_RENAMED)) { ret->d_fsdata = name; name = NULL; } diff --git a/fs/nfs/inode.c b/fs/nfs/inode.c index 18ab2da4eeb..9927913c97c 100644 --- a/fs/nfs/inode.c +++ b/fs/nfs/inode.c @@ -128,7 +128,7 @@ EXPORT_SYMBOL_GPL(nfs_clear_inode); void nfs_evict_inode(struct inode *inode) { - truncate_inode_pages(&inode->i_data, 0); + truncate_inode_pages_final(&inode->i_data); clear_inode(inode); nfs_clear_inode(inode); } @@ -147,6 +147,17 @@ int nfs_sync_mapping(struct address_space *mapping) return ret; } +static void nfs_set_cache_invalid(struct inode *inode, unsigned long flags) +{ + struct nfs_inode *nfsi = NFS_I(inode); + + if (inode->i_mapping->nrpages == 0) + flags &= ~NFS_INO_INVALID_DATA; + nfsi->cache_validity |= flags; + if (flags & NFS_INO_INVALID_DATA) + nfs_fscache_invalidate(inode); +} + /* * Invalidate the local caches */ @@ -162,19 +173,17 @@ static void nfs_zap_caches_locked(struct inode *inode) memset(NFS_I(inode)->cookieverf, 0, sizeof(NFS_I(inode)->cookieverf)); if (S_ISREG(mode) || S_ISDIR(mode) || S_ISLNK(mode)) { - nfs_fscache_invalidate(inode); - nfsi->cache_validity |= NFS_INO_INVALID_ATTR - | NFS_INO_INVALID_LABEL + nfs_set_cache_invalid(inode, NFS_INO_INVALID_ATTR | NFS_INO_INVALID_DATA | NFS_INO_INVALID_ACCESS | NFS_INO_INVALID_ACL - | NFS_INO_REVAL_PAGECACHE; + | NFS_INO_REVAL_PAGECACHE); } else - nfsi->cache_validity |= NFS_INO_INVALID_ATTR - | NFS_INO_INVALID_LABEL + nfs_set_cache_invalid(inode, NFS_INO_INVALID_ATTR | NFS_INO_INVALID_ACCESS | NFS_INO_INVALID_ACL - | NFS_INO_REVAL_PAGECACHE; + | NFS_INO_REVAL_PAGECACHE); + nfs_zap_label_cache_locked(nfsi); } void nfs_zap_caches(struct inode *inode) @@ -188,8 +197,7 @@ void nfs_zap_mapping(struct inode *inode, struct address_space *mapping) { if (mapping->nrpages != 0) { spin_lock(&inode->i_lock); - NFS_I(inode)->cache_validity |= NFS_INO_INVALID_DATA; - nfs_fscache_invalidate(inode); + nfs_set_cache_invalid(inode, NFS_INO_INVALID_DATA); spin_unlock(&inode->i_lock); } } @@ -210,7 +218,7 @@ EXPORT_SYMBOL_GPL(nfs_zap_acl_cache); void nfs_invalidate_atime(struct inode *inode) { spin_lock(&inode->i_lock); - NFS_I(inode)->cache_validity |= NFS_INO_INVALID_ATIME; + nfs_set_cache_invalid(inode, NFS_INO_INVALID_ATIME); spin_unlock(&inode->i_lock); } EXPORT_SYMBOL_GPL(nfs_invalidate_atime); @@ -266,6 +274,13 @@ nfs_init_locked(struct inode *inode, void *opaque) } #ifdef CONFIG_NFS_V4_SECURITY_LABEL +static void nfs_clear_label_invalid(struct inode *inode) +{ + spin_lock(&inode->i_lock); + NFS_I(inode)->cache_validity &= ~NFS_INO_INVALID_LABEL; + spin_unlock(&inode->i_lock); +} + void nfs_setsecurity(struct inode *inode, struct nfs_fattr *fattr, struct nfs4_label *label) { @@ -283,6 +298,7 @@ void nfs_setsecurity(struct inode *inode, struct nfs_fattr *fattr, __func__, (char *)label->label, label->len, error); + nfs_clear_label_invalid(inode); } } @@ -312,7 +328,7 @@ struct nfs4_label *nfs4_label_alloc(struct nfs_server *server, gfp_t flags) } EXPORT_SYMBOL_GPL(nfs4_label_alloc); #else -void inline nfs_setsecurity(struct inode *inode, struct nfs_fattr *fattr, +void nfs_setsecurity(struct inode *inode, struct nfs_fattr *fattr, struct nfs4_label *label) { } @@ -362,7 +378,7 @@ nfs_fhget(struct super_block *sb, struct nfs_fh *fh, struct nfs_fattr *fattr, st inode->i_mode = fattr->mode; if ((fattr->valid & NFS_ATTR_FATTR_MODE) == 0 && nfs_server_capable(inode, NFS_CAP_MODE)) - nfsi->cache_validity |= NFS_INO_INVALID_ATTR; + nfs_set_cache_invalid(inode, NFS_INO_INVALID_ATTR); /* Why so? Because we want revalidate for devices/FIFOs, and * that's precisely what we have in nfs_file_inode_operations. */ @@ -408,36 +424,36 @@ nfs_fhget(struct super_block *sb, struct nfs_fh *fh, struct nfs_fattr *fattr, st if (fattr->valid & NFS_ATTR_FATTR_ATIME) inode->i_atime = fattr->atime; else if (nfs_server_capable(inode, NFS_CAP_ATIME)) - nfsi->cache_validity |= NFS_INO_INVALID_ATTR; + nfs_set_cache_invalid(inode, NFS_INO_INVALID_ATTR); if (fattr->valid & NFS_ATTR_FATTR_MTIME) inode->i_mtime = fattr->mtime; else if (nfs_server_capable(inode, NFS_CAP_MTIME)) - nfsi->cache_validity |= NFS_INO_INVALID_ATTR; + nfs_set_cache_invalid(inode, NFS_INO_INVALID_ATTR); if (fattr->valid & NFS_ATTR_FATTR_CTIME) inode->i_ctime = fattr->ctime; else if (nfs_server_capable(inode, NFS_CAP_CTIME)) - nfsi->cache_validity |= NFS_INO_INVALID_ATTR; + nfs_set_cache_invalid(inode, NFS_INO_INVALID_ATTR); if (fattr->valid & NFS_ATTR_FATTR_CHANGE) inode->i_version = fattr->change_attr; else if (nfs_server_capable(inode, NFS_CAP_CHANGE_ATTR)) - nfsi->cache_validity |= NFS_INO_INVALID_ATTR; + nfs_set_cache_invalid(inode, NFS_INO_INVALID_ATTR); if (fattr->valid & NFS_ATTR_FATTR_SIZE) inode->i_size = nfs_size_to_loff_t(fattr->size); else - nfsi->cache_validity |= NFS_INO_INVALID_ATTR - | NFS_INO_REVAL_PAGECACHE; + nfs_set_cache_invalid(inode, NFS_INO_INVALID_ATTR + | NFS_INO_REVAL_PAGECACHE); if (fattr->valid & NFS_ATTR_FATTR_NLINK) set_nlink(inode, fattr->nlink); else if (nfs_server_capable(inode, NFS_CAP_NLINK)) - nfsi->cache_validity |= NFS_INO_INVALID_ATTR; + nfs_set_cache_invalid(inode, NFS_INO_INVALID_ATTR); if (fattr->valid & NFS_ATTR_FATTR_OWNER) inode->i_uid = fattr->uid; else if (nfs_server_capable(inode, NFS_CAP_OWNER)) - nfsi->cache_validity |= NFS_INO_INVALID_ATTR; + nfs_set_cache_invalid(inode, NFS_INO_INVALID_ATTR); if (fattr->valid & NFS_ATTR_FATTR_GROUP) inode->i_gid = fattr->gid; else if (nfs_server_capable(inode, NFS_CAP_OWNER_GROUP)) - nfsi->cache_validity |= NFS_INO_INVALID_ATTR; + nfs_set_cache_invalid(inode, NFS_INO_INVALID_ATTR); if (fattr->valid & NFS_ATTR_FATTR_BLOCKS_USED) inode->i_blocks = fattr->du.nfs2.blocks; if (fattr->valid & NFS_ATTR_FATTR_SPACE_USED) { @@ -458,9 +474,9 @@ nfs_fhget(struct super_block *sb, struct nfs_fh *fh, struct nfs_fattr *fattr, st unlock_new_inode(inode); } else nfs_refresh_inode(inode, fattr); - dprintk("NFS: nfs_fhget(%s/%Ld fh_crc=0x%08x ct=%d)\n", + dprintk("NFS: nfs_fhget(%s/%Lu fh_crc=0x%08x ct=%d)\n", inode->i_sb->s_id, - (long long)NFS_FILEID(inode), + (unsigned long long)NFS_FILEID(inode), nfs_display_fhandle_hash(fh), atomic_read(&inode->i_count)); @@ -543,6 +559,9 @@ static int nfs_vmtruncate(struct inode * inode, loff_t offset) spin_lock(&inode->i_lock); i_size_write(inode, offset); + /* Optimisation */ + if (offset == 0) + NFS_I(inode)->cache_validity &= ~NFS_INO_INVALID_DATA; spin_unlock(&inode->i_lock); truncate_pagecache(inode, offset); @@ -571,7 +590,8 @@ void nfs_setattr_update_inode(struct inode *inode, struct iattr *attr) inode->i_uid = attr->ia_uid; if ((attr->ia_valid & ATTR_GID) != 0) inode->i_gid = attr->ia_gid; - NFS_I(inode)->cache_validity |= NFS_INO_INVALID_ACCESS|NFS_INO_INVALID_ACL; + nfs_set_cache_invalid(inode, NFS_INO_INVALID_ACCESS + | NFS_INO_INVALID_ACL); spin_unlock(&inode->i_lock); } if ((attr->ia_valid & ATTR_SIZE) != 0) { @@ -581,6 +601,25 @@ void nfs_setattr_update_inode(struct inode *inode, struct iattr *attr) } EXPORT_SYMBOL_GPL(nfs_setattr_update_inode); +static void nfs_request_parent_use_readdirplus(struct dentry *dentry) +{ + struct dentry *parent; + + parent = dget_parent(dentry); + nfs_force_use_readdirplus(parent->d_inode); + dput(parent); +} + +static bool nfs_need_revalidate_inode(struct inode *inode) +{ + if (NFS_I(inode)->cache_validity & + (NFS_INO_INVALID_ATTR|NFS_INO_INVALID_LABEL)) + return true; + if (nfs_attribute_cache_expired(inode)) + return true; + return false; +} + int nfs_getattr(struct vfsmount *mnt, struct dentry *dentry, struct kstat *stat) { struct inode *inode = dentry->d_inode; @@ -609,10 +648,13 @@ int nfs_getattr(struct vfsmount *mnt, struct dentry *dentry, struct kstat *stat) ((mnt->mnt_flags & MNT_NODIRATIME) && S_ISDIR(inode->i_mode))) need_atime = 0; - if (need_atime) - err = __nfs_revalidate_inode(NFS_SERVER(inode), inode); - else - err = nfs_revalidate_inode(NFS_SERVER(inode), inode); + if (need_atime || nfs_need_revalidate_inode(inode)) { + struct nfs_server *server = NFS_SERVER(inode); + + if (server->caps & NFS_CAP_READDIRPLUS) + nfs_request_parent_use_readdirplus(dentry); + err = __nfs_revalidate_inode(server, inode); + } if (!err) { generic_fillattr(inode, stat); stat->ino = nfs_compat_user_ino64(NFS_FILEID(inode)); @@ -870,8 +912,8 @@ __nfs_revalidate_inode(struct nfs_server *server, struct inode *inode) struct nfs_fattr *fattr = NULL; struct nfs_inode *nfsi = NFS_I(inode); - dfprintk(PAGECACHE, "NFS: revalidating (%s/%Ld)\n", - inode->i_sb->s_id, (long long)NFS_FILEID(inode)); + dfprintk(PAGECACHE, "NFS: revalidating (%s/%Lu)\n", + inode->i_sb->s_id, (unsigned long long)NFS_FILEID(inode)); trace_nfs_revalidate_inode_enter(inode); @@ -895,9 +937,9 @@ __nfs_revalidate_inode(struct nfs_server *server, struct inode *inode) status = NFS_PROTO(inode)->getattr(server, NFS_FH(inode), fattr, label); if (status != 0) { - dfprintk(PAGECACHE, "nfs_revalidate_inode: (%s/%Ld) getattr failed, error=%d\n", + dfprintk(PAGECACHE, "nfs_revalidate_inode: (%s/%Lu) getattr failed, error=%d\n", inode->i_sb->s_id, - (long long)NFS_FILEID(inode), status); + (unsigned long long)NFS_FILEID(inode), status); if (status == -ESTALE) { nfs_zap_caches(inode); if (!S_ISDIR(inode->i_mode)) @@ -908,9 +950,9 @@ __nfs_revalidate_inode(struct nfs_server *server, struct inode *inode) status = nfs_refresh_inode(inode, fattr); if (status) { - dfprintk(PAGECACHE, "nfs_revalidate_inode: (%s/%Ld) refresh failed, error=%d\n", + dfprintk(PAGECACHE, "nfs_revalidate_inode: (%s/%Lu) refresh failed, error=%d\n", inode->i_sb->s_id, - (long long)NFS_FILEID(inode), status); + (unsigned long long)NFS_FILEID(inode), status); goto err_out; } @@ -919,9 +961,9 @@ __nfs_revalidate_inode(struct nfs_server *server, struct inode *inode) nfs_setsecurity(inode, fattr, label); - dfprintk(PAGECACHE, "NFS: (%s/%Ld) revalidation complete\n", + dfprintk(PAGECACHE, "NFS: (%s/%Lu) revalidation complete\n", inode->i_sb->s_id, - (long long)NFS_FILEID(inode)); + (unsigned long long)NFS_FILEID(inode)); err_out: nfs4_label_free(label); @@ -954,9 +996,7 @@ int nfs_attribute_cache_expired(struct inode *inode) */ int nfs_revalidate_inode(struct nfs_server *server, struct inode *inode) { - if (!(NFS_I(inode)->cache_validity & - (NFS_INO_INVALID_ATTR|NFS_INO_INVALID_LABEL)) - && !nfs_attribute_cache_expired(inode)) + if (!nfs_need_revalidate_inode(inode)) return NFS_STALE(inode) ? -ESTALE : 0; return __nfs_revalidate_inode(server, inode); } @@ -977,16 +1017,17 @@ static int nfs_invalidate_mapping(struct inode *inode, struct address_space *map if (ret < 0) return ret; } - spin_lock(&inode->i_lock); - nfsi->cache_validity &= ~NFS_INO_INVALID_DATA; - if (S_ISDIR(inode->i_mode)) + if (S_ISDIR(inode->i_mode)) { + spin_lock(&inode->i_lock); memset(nfsi->cookieverf, 0, sizeof(nfsi->cookieverf)); - spin_unlock(&inode->i_lock); + spin_unlock(&inode->i_lock); + } nfs_inc_stats(inode, NFSIOS_DATAINVALIDATE); nfs_fscache_wait_on_invalidate(inode); - dfprintk(PAGECACHE, "NFS: (%s/%Ld) data cache invalidated\n", - inode->i_sb->s_id, (long long)NFS_FILEID(inode)); + dfprintk(PAGECACHE, "NFS: (%s/%Lu) data cache invalidated\n", + inode->i_sb->s_id, + (unsigned long long)NFS_FILEID(inode)); return 0; } @@ -1007,6 +1048,7 @@ static bool nfs_mapping_need_revalidate_inode(struct inode *inode) int nfs_revalidate_mapping(struct inode *inode, struct address_space *mapping) { struct nfs_inode *nfsi = NFS_I(inode); + unsigned long *bitlock = &nfsi->flags; int ret = 0; /* swapfiles are not supposed to be shared. */ @@ -1018,12 +1060,46 @@ int nfs_revalidate_mapping(struct inode *inode, struct address_space *mapping) if (ret < 0) goto out; } - if (nfsi->cache_validity & NFS_INO_INVALID_DATA) { - trace_nfs_invalidate_mapping_enter(inode); - ret = nfs_invalidate_mapping(inode, mapping); - trace_nfs_invalidate_mapping_exit(inode, ret); + + /* + * We must clear NFS_INO_INVALID_DATA first to ensure that + * invalidations that come in while we're shooting down the mappings + * are respected. But, that leaves a race window where one revalidator + * can clear the flag, and then another checks it before the mapping + * gets invalidated. Fix that by serializing access to this part of + * the function. + * + * At the same time, we need to allow other tasks to see whether we + * might be in the middle of invalidating the pages, so we only set + * the bit lock here if it looks like we're going to be doing that. + */ + for (;;) { + ret = wait_on_bit(bitlock, NFS_INO_INVALIDATING, + nfs_wait_bit_killable, TASK_KILLABLE); + if (ret) + goto out; + spin_lock(&inode->i_lock); + if (test_bit(NFS_INO_INVALIDATING, bitlock)) { + spin_unlock(&inode->i_lock); + continue; + } + if (nfsi->cache_validity & NFS_INO_INVALID_DATA) + break; + spin_unlock(&inode->i_lock); + goto out; } + set_bit(NFS_INO_INVALIDATING, bitlock); + smp_wmb(); + nfsi->cache_validity &= ~NFS_INO_INVALID_DATA; + spin_unlock(&inode->i_lock); + trace_nfs_invalidate_mapping_enter(inode); + ret = nfs_invalidate_mapping(inode, mapping); + trace_nfs_invalidate_mapping_exit(inode, ret); + + clear_bit_unlock(NFS_INO_INVALIDATING, bitlock); + smp_mb__after_atomic(); + wake_up_bit(bitlock, NFS_INO_INVALIDATING); out: return ret; } @@ -1038,7 +1114,7 @@ static unsigned long nfs_wcc_update_inode(struct inode *inode, struct nfs_fattr && inode->i_version == fattr->pre_change_attr) { inode->i_version = fattr->change_attr; if (S_ISDIR(inode->i_mode)) - nfsi->cache_validity |= NFS_INO_INVALID_DATA; + nfs_set_cache_invalid(inode, NFS_INO_INVALID_DATA); ret |= NFS_INO_INVALID_ATTR; } /* If we have atomic WCC data, we may update some attributes */ @@ -1054,7 +1130,7 @@ static unsigned long nfs_wcc_update_inode(struct inode *inode, struct nfs_fattr && timespec_equal(&inode->i_mtime, &fattr->pre_mtime)) { memcpy(&inode->i_mtime, &fattr->mtime, sizeof(inode->i_mtime)); if (S_ISDIR(inode->i_mode)) - nfsi->cache_validity |= NFS_INO_INVALID_DATA; + nfs_set_cache_invalid(inode, NFS_INO_INVALID_DATA); ret |= NFS_INO_INVALID_ATTR; } if ((fattr->valid & NFS_ATTR_FATTR_PRESIZE) @@ -1065,9 +1141,6 @@ static unsigned long nfs_wcc_update_inode(struct inode *inode, struct nfs_fattr ret |= NFS_INO_INVALID_ATTR; } - if (nfsi->cache_validity & NFS_INO_INVALID_DATA) - nfs_fscache_invalidate(inode); - return ret; } @@ -1126,7 +1199,7 @@ static int nfs_check_inode_attributes(struct inode *inode, struct nfs_fattr *fat invalid |= NFS_INO_INVALID_ATIME; if (invalid != 0) - nfsi->cache_validity |= invalid; + nfs_set_cache_invalid(inode, invalid); nfsi->read_cache_jiffies = fattr->time_start; return 0; @@ -1282,12 +1355,28 @@ static int nfs_inode_attrs_need_update(const struct inode *inode, const struct n ((long)nfsi->attr_gencount - (long)nfs_read_attr_generation_counter() > 0); } +/* + * Don't trust the change_attribute, mtime, ctime or size if + * a pnfs LAYOUTCOMMIT is outstanding + */ +static void nfs_inode_attrs_handle_layoutcommit(struct inode *inode, + struct nfs_fattr *fattr) +{ + if (pnfs_layoutcommit_outstanding(inode)) + fattr->valid &= ~(NFS_ATTR_FATTR_CHANGE | + NFS_ATTR_FATTR_MTIME | + NFS_ATTR_FATTR_CTIME | + NFS_ATTR_FATTR_SIZE); +} + static int nfs_refresh_inode_locked(struct inode *inode, struct nfs_fattr *fattr) { int ret; trace_nfs_refresh_inode_enter(inode); + nfs_inode_attrs_handle_layoutcommit(inode, fattr); + if (nfs_inode_attrs_need_update(inode, fattr)) ret = nfs_update_inode(inode, fattr); else @@ -1323,13 +1412,11 @@ EXPORT_SYMBOL_GPL(nfs_refresh_inode); static int nfs_post_op_update_inode_locked(struct inode *inode, struct nfs_fattr *fattr) { - struct nfs_inode *nfsi = NFS_I(inode); + unsigned long invalid = NFS_INO_INVALID_ATTR|NFS_INO_REVAL_PAGECACHE; - nfsi->cache_validity |= NFS_INO_INVALID_ATTR|NFS_INO_REVAL_PAGECACHE; - if (S_ISDIR(inode->i_mode)) { - nfsi->cache_validity |= NFS_INO_INVALID_DATA; - nfs_fscache_invalidate(inode); - } + if (S_ISDIR(inode->i_mode)) + invalid |= NFS_INO_INVALID_DATA; + nfs_set_cache_invalid(inode, invalid); if ((fattr->valid & NFS_ATTR_FATTR) == 0) return 0; return nfs_refresh_inode_locked(inode, fattr); @@ -1434,7 +1521,7 @@ static int nfs_update_inode(struct inode *inode, struct nfs_fattr *fattr) unsigned long now = jiffies; unsigned long save_cache_validity; - dfprintk(VFS, "NFS: %s(%s/%ld fh_crc=0x%08x ct=%d info=0x%x)\n", + dfprintk(VFS, "NFS: %s(%s/%lu fh_crc=0x%08x ct=%d info=0x%x)\n", __func__, inode->i_sb->s_id, inode->i_ino, nfs_display_fhandle_hash(NFS_FH(inode)), atomic_read(&inode->i_count), fattr->valid); @@ -1455,7 +1542,7 @@ static int nfs_update_inode(struct inode *inode, struct nfs_fattr *fattr) /* * Big trouble! The inode has become a different object. */ - printk(KERN_DEBUG "NFS: %s: inode %ld mode changed, %07o to %07o\n", + printk(KERN_DEBUG "NFS: %s: inode %lu mode changed, %07o to %07o\n", __func__, inode->i_ino, inode->i_mode, fattr->mode); goto out_err; } @@ -1496,18 +1583,20 @@ static int nfs_update_inode(struct inode *inode, struct nfs_fattr *fattr) inode->i_version = fattr->change_attr; } } else if (server->caps & NFS_CAP_CHANGE_ATTR) - invalid |= save_cache_validity; + nfsi->cache_validity |= save_cache_validity; if (fattr->valid & NFS_ATTR_FATTR_MTIME) { memcpy(&inode->i_mtime, &fattr->mtime, sizeof(inode->i_mtime)); } else if (server->caps & NFS_CAP_MTIME) - invalid |= save_cache_validity & (NFS_INO_INVALID_ATTR + nfsi->cache_validity |= save_cache_validity & + (NFS_INO_INVALID_ATTR | NFS_INO_REVAL_FORCED); if (fattr->valid & NFS_ATTR_FATTR_CTIME) { memcpy(&inode->i_ctime, &fattr->ctime, sizeof(inode->i_ctime)); } else if (server->caps & NFS_CAP_CTIME) - invalid |= save_cache_validity & (NFS_INO_INVALID_ATTR + nfsi->cache_validity |= save_cache_validity & + (NFS_INO_INVALID_ATTR | NFS_INO_REVAL_FORCED); /* Check if our cached file size is stale */ @@ -1517,10 +1606,10 @@ static int nfs_update_inode(struct inode *inode, struct nfs_fattr *fattr) if (new_isize != cur_isize) { /* Do we perhaps have any outstanding writes, or has * the file grown beyond our last write? */ - if ((nfsi->npages == 0 && !test_bit(NFS_INO_LAYOUTCOMMIT, &nfsi->flags)) || - new_isize > cur_isize) { + if ((nfsi->npages == 0) || new_isize > cur_isize) { i_size_write(inode, new_isize); invalid |= NFS_INO_INVALID_ATTR|NFS_INO_INVALID_DATA; + invalid &= ~NFS_INO_REVAL_PAGECACHE; } dprintk("NFS: isize change on server for file %s/%ld " "(%Ld to %Ld)\n", @@ -1530,7 +1619,8 @@ static int nfs_update_inode(struct inode *inode, struct nfs_fattr *fattr) (long long)new_isize); } } else - invalid |= save_cache_validity & (NFS_INO_INVALID_ATTR + nfsi->cache_validity |= save_cache_validity & + (NFS_INO_INVALID_ATTR | NFS_INO_REVAL_PAGECACHE | NFS_INO_REVAL_FORCED); @@ -1538,7 +1628,8 @@ static int nfs_update_inode(struct inode *inode, struct nfs_fattr *fattr) if (fattr->valid & NFS_ATTR_FATTR_ATIME) memcpy(&inode->i_atime, &fattr->atime, sizeof(inode->i_atime)); else if (server->caps & NFS_CAP_ATIME) - invalid |= save_cache_validity & (NFS_INO_INVALID_ATIME + nfsi->cache_validity |= save_cache_validity & + (NFS_INO_INVALID_ATIME | NFS_INO_REVAL_FORCED); if (fattr->valid & NFS_ATTR_FATTR_MODE) { @@ -1549,7 +1640,8 @@ static int nfs_update_inode(struct inode *inode, struct nfs_fattr *fattr) invalid |= NFS_INO_INVALID_ATTR|NFS_INO_INVALID_ACCESS|NFS_INO_INVALID_ACL; } } else if (server->caps & NFS_CAP_MODE) - invalid |= save_cache_validity & (NFS_INO_INVALID_ATTR + nfsi->cache_validity |= save_cache_validity & + (NFS_INO_INVALID_ATTR | NFS_INO_INVALID_ACCESS | NFS_INO_INVALID_ACL | NFS_INO_REVAL_FORCED); @@ -1560,7 +1652,8 @@ static int nfs_update_inode(struct inode *inode, struct nfs_fattr *fattr) inode->i_uid = fattr->uid; } } else if (server->caps & NFS_CAP_OWNER) - invalid |= save_cache_validity & (NFS_INO_INVALID_ATTR + nfsi->cache_validity |= save_cache_validity & + (NFS_INO_INVALID_ATTR | NFS_INO_INVALID_ACCESS | NFS_INO_INVALID_ACL | NFS_INO_REVAL_FORCED); @@ -1571,7 +1664,8 @@ static int nfs_update_inode(struct inode *inode, struct nfs_fattr *fattr) inode->i_gid = fattr->gid; } } else if (server->caps & NFS_CAP_OWNER_GROUP) - invalid |= save_cache_validity & (NFS_INO_INVALID_ATTR + nfsi->cache_validity |= save_cache_validity & + (NFS_INO_INVALID_ATTR | NFS_INO_INVALID_ACCESS | NFS_INO_INVALID_ACL | NFS_INO_REVAL_FORCED); @@ -1584,7 +1678,8 @@ static int nfs_update_inode(struct inode *inode, struct nfs_fattr *fattr) set_nlink(inode, fattr->nlink); } } else if (server->caps & NFS_CAP_NLINK) - invalid |= save_cache_validity & (NFS_INO_INVALID_ATTR + nfsi->cache_validity |= save_cache_validity & + (NFS_INO_INVALID_ATTR | NFS_INO_REVAL_FORCED); if (fattr->valid & NFS_ATTR_FATTR_SPACE_USED) { @@ -1597,7 +1692,7 @@ static int nfs_update_inode(struct inode *inode, struct nfs_fattr *fattr) inode->i_blocks = fattr->du.nfs2.blocks; /* Update attrtimeo value if we're out of the unstable period */ - if (invalid & (NFS_INO_INVALID_ATTR|NFS_INO_INVALID_LABEL)) { + if (invalid & NFS_INO_INVALID_ATTR) { nfs_inc_stats(inode, NFSIOS_ATTRINVALIDATE); nfsi->attrtimeo = NFS_MINATTRTIMEO(inode); nfsi->attrtimeo_timestamp = now; @@ -1610,17 +1705,13 @@ static int nfs_update_inode(struct inode *inode, struct nfs_fattr *fattr) } } invalid &= ~NFS_INO_INVALID_ATTR; - invalid &= ~NFS_INO_INVALID_LABEL; /* Don't invalidate the data if we were to blame */ if (!(S_ISREG(inode->i_mode) || S_ISDIR(inode->i_mode) || S_ISLNK(inode->i_mode))) invalid &= ~NFS_INO_INVALID_DATA; if (!NFS_PROTO(inode)->have_delegation(inode, FMODE_READ) || (save_cache_validity & NFS_INO_REVAL_FORCED)) - nfsi->cache_validity |= invalid; - - if (invalid & NFS_INO_INVALID_DATA) - nfs_fscache_invalidate(inode); + nfs_set_cache_invalid(inode, invalid); return 0; out_err: @@ -1641,10 +1732,6 @@ struct inode *nfs_alloc_inode(struct super_block *sb) return NULL; nfsi->flags = 0UL; nfsi->cache_validity = 0UL; -#ifdef CONFIG_NFS_V3_ACL - nfsi->acl_access = ERR_PTR(-EAGAIN); - nfsi->acl_default = ERR_PTR(-EAGAIN); -#endif #if IS_ENABLED(CONFIG_NFS_V4) nfsi->nfs4_acl = NULL; #endif /* CONFIG_NFS_V4 */ diff --git a/fs/nfs/internal.h b/fs/nfs/internal.h index bca6a3e3c49..f415cbf9f6c 100644 --- a/fs/nfs/internal.h +++ b/fs/nfs/internal.h @@ -176,7 +176,8 @@ extern struct nfs_server *nfs4_create_server( extern struct nfs_server *nfs4_create_referral_server(struct nfs_clone_mount *, struct nfs_fh *); extern int nfs4_update_server(struct nfs_server *server, const char *hostname, - struct sockaddr *sap, size_t salen); + struct sockaddr *sap, size_t salen, + struct net *net); extern void nfs_free_server(struct nfs_server *server); extern struct nfs_server *nfs_clone_server(struct nfs_server *, struct nfs_fh *, @@ -230,13 +231,21 @@ extern void nfs_destroy_writepagecache(void); extern int __init nfs_init_directcache(void); extern void nfs_destroy_directcache(void); -extern bool nfs_pgarray_set(struct nfs_page_array *p, unsigned int pagecount); extern void nfs_pgheader_init(struct nfs_pageio_descriptor *desc, struct nfs_pgio_header *hdr, void (*release)(struct nfs_pgio_header *hdr)); void nfs_set_pgio_error(struct nfs_pgio_header *hdr, int error, loff_t pos); int nfs_iocounter_wait(struct nfs_io_counter *c); +extern const struct nfs_pageio_ops nfs_pgio_rw_ops; +struct nfs_rw_header *nfs_rw_header_alloc(const struct nfs_rw_ops *); +void nfs_rw_header_free(struct nfs_pgio_header *); +void nfs_pgio_data_release(struct nfs_pgio_data *); +int nfs_generic_pgio(struct nfs_pageio_descriptor *, struct nfs_pgio_header *); +int nfs_initiate_pgio(struct rpc_clnt *, struct nfs_pgio_data *, + const struct rpc_call_ops *, int, int); +void nfs_free_request(struct nfs_page *req); + static inline void nfs_iocounter_init(struct nfs_io_counter *c) { c->flags = 0; @@ -269,6 +278,30 @@ extern const u32 nfs41_maxgetdevinfo_overhead; extern struct rpc_procinfo nfs4_procedures[]; #endif +#ifdef CONFIG_NFS_V4_SECURITY_LABEL +extern struct nfs4_label *nfs4_label_alloc(struct nfs_server *server, gfp_t flags); +static inline void nfs4_label_free(struct nfs4_label *label) +{ + if (label) { + kfree(label->label); + kfree(label); + } + return; +} + +static inline void nfs_zap_label_cache_locked(struct nfs_inode *nfsi) +{ + if (nfs_server_capable(&nfsi->vfs_inode, NFS_CAP_SECURITY_LABEL)) + nfsi->cache_validity |= NFS_INO_INVALID_LABEL; +} +#else +static inline struct nfs4_label *nfs4_label_alloc(struct nfs_server *server, gfp_t flags) { return NULL; } +static inline void nfs4_label_free(void *label) {} +static inline void nfs_zap_label_cache_locked(struct nfs_inode *nfsi) +{ +} +#endif /* CONFIG_NFS_V4_SECURITY_LABEL */ + /* proc.c */ void nfs_close_context(struct nfs_open_context *ctx, int is_sync); extern struct nfs_client *nfs_init_client(struct nfs_client *clp, @@ -276,6 +309,7 @@ extern struct nfs_client *nfs_init_client(struct nfs_client *clp, const char *ip_addr); /* dir.c */ +extern void nfs_force_use_readdirplus(struct inode *dir); extern unsigned long nfs_access_cache_count(struct shrinker *shrink, struct shrink_control *sc); extern unsigned long nfs_access_cache_scan(struct shrinker *shrink, @@ -294,16 +328,14 @@ int nfs_rename(struct inode *, struct dentry *, struct inode *, struct dentry *) int nfs_file_fsync_commit(struct file *, loff_t, loff_t, int); loff_t nfs_file_llseek(struct file *, loff_t, int); int nfs_file_flush(struct file *, fl_owner_t); -ssize_t nfs_file_read(struct kiocb *, const struct iovec *, unsigned long, loff_t); +ssize_t nfs_file_read(struct kiocb *, struct iov_iter *); ssize_t nfs_file_splice_read(struct file *, loff_t *, struct pipe_inode_info *, size_t, unsigned int); int nfs_file_mmap(struct file *, struct vm_area_struct *); -ssize_t nfs_file_write(struct kiocb *, const struct iovec *, unsigned long, loff_t); +ssize_t nfs_file_write(struct kiocb *, struct iov_iter *); int nfs_file_release(struct inode *, struct file *); int nfs_lock(struct file *, int, struct file_lock *); int nfs_flock(struct file *, int, struct file_lock *); -ssize_t nfs_file_splice_write(struct pipe_inode_info *, struct file *, loff_t *, - size_t, unsigned int); int nfs_check_flags(int); int nfs_setlease(struct file *, long, struct file_lock **); @@ -369,19 +401,11 @@ extern int nfs4_get_rootfh(struct nfs_server *server, struct nfs_fh *mntfh, bool struct nfs_pgio_completion_ops; /* read.c */ -extern struct nfs_read_header *nfs_readhdr_alloc(void); -extern void nfs_readhdr_free(struct nfs_pgio_header *hdr); extern void nfs_pageio_init_read(struct nfs_pageio_descriptor *pgio, - struct inode *inode, + struct inode *inode, bool force_mds, const struct nfs_pgio_completion_ops *compl_ops); -extern int nfs_initiate_read(struct rpc_clnt *clnt, - struct nfs_read_data *data, - const struct rpc_call_ops *call_ops, int flags); extern void nfs_read_prepare(struct rpc_task *task, void *calldata); -extern int nfs_generic_pagein(struct nfs_pageio_descriptor *desc, - struct nfs_pgio_header *hdr); extern void nfs_pageio_reset_read_mds(struct nfs_pageio_descriptor *pgio); -extern void nfs_readdata_release(struct nfs_read_data *rdata); /* super.c */ void nfs_clone_super(struct super_block *, struct nfs_mount_info *); @@ -396,19 +420,10 @@ int nfs_remount(struct super_block *sb, int *flags, char *raw_data); /* write.c */ extern void nfs_pageio_init_write(struct nfs_pageio_descriptor *pgio, - struct inode *inode, int ioflags, + struct inode *inode, int ioflags, bool force_mds, const struct nfs_pgio_completion_ops *compl_ops); -extern struct nfs_write_header *nfs_writehdr_alloc(void); -extern void nfs_writehdr_free(struct nfs_pgio_header *hdr); -extern int nfs_generic_flush(struct nfs_pageio_descriptor *desc, - struct nfs_pgio_header *hdr); extern void nfs_pageio_reset_write_mds(struct nfs_pageio_descriptor *pgio); -extern void nfs_writedata_release(struct nfs_write_data *wdata); extern void nfs_commit_free(struct nfs_commit_data *p); -extern int nfs_initiate_write(struct rpc_clnt *clnt, - struct nfs_write_data *data, - const struct rpc_call_ops *call_ops, - int how, int flags); extern void nfs_write_prepare(struct rpc_task *task, void *calldata); extern void nfs_commit_prepare(struct rpc_task *task, void *calldata); extern int nfs_initiate_commit(struct rpc_clnt *clnt, @@ -421,6 +436,7 @@ extern void nfs_init_commit(struct nfs_commit_data *data, struct nfs_commit_info *cinfo); int nfs_scan_commit_list(struct list_head *src, struct list_head *dst, struct nfs_commit_info *cinfo, int max); +unsigned long nfs_reqs_to_commit(struct nfs_commit_info *); int nfs_scan_commit(struct inode *inode, struct list_head *dst, struct nfs_commit_info *cinfo); void nfs_mark_request_commit(struct nfs_page *req, @@ -449,6 +465,13 @@ extern int nfs_migrate_page(struct address_space *, #define nfs_migrate_page NULL #endif +/* unlink.c */ +extern struct rpc_task * +nfs_async_rename(struct inode *old_dir, struct inode *new_dir, + struct dentry *old_dentry, struct dentry *new_dentry, + void (*complete)(struct rpc_task *, struct nfs_renamedata *)); +extern int nfs_sillyrename(struct inode *dir, struct dentry *dentry); + /* direct.c */ void nfs_init_cinfo_from_dreq(struct nfs_commit_info *cinfo, struct nfs_direct_req *dreq); @@ -459,7 +482,7 @@ static inline void nfs_inode_dio_wait(struct inode *inode) extern ssize_t nfs_dreq_bytes_left(struct nfs_direct_req *dreq); /* nfs4proc.c */ -extern void __nfs4_read_done_cb(struct nfs_read_data *); +extern void __nfs4_read_done_cb(struct nfs_pgio_data *); extern struct nfs_client *nfs4_init_client(struct nfs_client *clp, const struct rpc_timeout *timeparms, const char *ip_addr); diff --git a/fs/nfs/nfs2xdr.c b/fs/nfs/nfs2xdr.c index 62db136339e..5f61b83f4a1 100644 --- a/fs/nfs/nfs2xdr.c +++ b/fs/nfs/nfs2xdr.c @@ -103,7 +103,7 @@ static void print_overflow_msg(const char *func, const struct xdr_stream *xdr) /* * typedef opaque nfsdata<>; */ -static int decode_nfsdata(struct xdr_stream *xdr, struct nfs_readres *result) +static int decode_nfsdata(struct xdr_stream *xdr, struct nfs_pgio_res *result) { u32 recvd, count; __be32 *p; @@ -613,7 +613,7 @@ static void nfs2_xdr_enc_readlinkargs(struct rpc_rqst *req, * }; */ static void encode_readargs(struct xdr_stream *xdr, - const struct nfs_readargs *args) + const struct nfs_pgio_args *args) { u32 offset = args->offset; u32 count = args->count; @@ -629,7 +629,7 @@ static void encode_readargs(struct xdr_stream *xdr, static void nfs2_xdr_enc_readargs(struct rpc_rqst *req, struct xdr_stream *xdr, - const struct nfs_readargs *args) + const struct nfs_pgio_args *args) { encode_readargs(xdr, args); prepare_reply_buffer(req, args->pages, args->pgbase, @@ -649,7 +649,7 @@ static void nfs2_xdr_enc_readargs(struct rpc_rqst *req, * }; */ static void encode_writeargs(struct xdr_stream *xdr, - const struct nfs_writeargs *args) + const struct nfs_pgio_args *args) { u32 offset = args->offset; u32 count = args->count; @@ -669,7 +669,7 @@ static void encode_writeargs(struct xdr_stream *xdr, static void nfs2_xdr_enc_writeargs(struct rpc_rqst *req, struct xdr_stream *xdr, - const struct nfs_writeargs *args) + const struct nfs_pgio_args *args) { encode_writeargs(xdr, args); xdr->buf->flags |= XDRBUF_WRITE; @@ -857,7 +857,7 @@ out_default: * }; */ static int nfs2_xdr_dec_readres(struct rpc_rqst *req, struct xdr_stream *xdr, - struct nfs_readres *result) + struct nfs_pgio_res *result) { enum nfs_stat status; int error; @@ -878,7 +878,7 @@ out_default: } static int nfs2_xdr_dec_writeres(struct rpc_rqst *req, struct xdr_stream *xdr, - struct nfs_writeres *result) + struct nfs_pgio_res *result) { /* All NFSv2 writes are "file sync" writes */ result->verf->committed = NFS_FILE_SYNC; diff --git a/fs/nfs/nfs3acl.c b/fs/nfs/nfs3acl.c index 4a1aafba6a2..8f854dde415 100644 --- a/fs/nfs/nfs3acl.c +++ b/fs/nfs/nfs3acl.c @@ -10,179 +10,7 @@ #define NFSDBG_FACILITY NFSDBG_PROC -ssize_t nfs3_listxattr(struct dentry *dentry, char *buffer, size_t size) -{ - struct inode *inode = dentry->d_inode; - struct posix_acl *acl; - int pos=0, len=0; - -# define output(s) do { \ - if (pos + sizeof(s) <= size) { \ - memcpy(buffer + pos, s, sizeof(s)); \ - pos += sizeof(s); \ - } \ - len += sizeof(s); \ - } while(0) - - acl = nfs3_proc_getacl(inode, ACL_TYPE_ACCESS); - if (IS_ERR(acl)) - return PTR_ERR(acl); - if (acl) { - output("system.posix_acl_access"); - posix_acl_release(acl); - } - - if (S_ISDIR(inode->i_mode)) { - acl = nfs3_proc_getacl(inode, ACL_TYPE_DEFAULT); - if (IS_ERR(acl)) - return PTR_ERR(acl); - if (acl) { - output("system.posix_acl_default"); - posix_acl_release(acl); - } - } - -# undef output - - if (!buffer || len <= size) - return len; - return -ERANGE; -} - -ssize_t nfs3_getxattr(struct dentry *dentry, const char *name, - void *buffer, size_t size) -{ - struct inode *inode = dentry->d_inode; - struct posix_acl *acl; - int type, error = 0; - - if (strcmp(name, POSIX_ACL_XATTR_ACCESS) == 0) - type = ACL_TYPE_ACCESS; - else if (strcmp(name, POSIX_ACL_XATTR_DEFAULT) == 0) - type = ACL_TYPE_DEFAULT; - else - return -EOPNOTSUPP; - - acl = nfs3_proc_getacl(inode, type); - if (IS_ERR(acl)) - return PTR_ERR(acl); - else if (acl) { - if (type == ACL_TYPE_ACCESS && acl->a_count == 0) - error = -ENODATA; - else - error = posix_acl_to_xattr(&init_user_ns, acl, buffer, size); - posix_acl_release(acl); - } else - error = -ENODATA; - - return error; -} - -int nfs3_setxattr(struct dentry *dentry, const char *name, - const void *value, size_t size, int flags) -{ - struct inode *inode = dentry->d_inode; - struct posix_acl *acl; - int type, error; - - if (strcmp(name, POSIX_ACL_XATTR_ACCESS) == 0) - type = ACL_TYPE_ACCESS; - else if (strcmp(name, POSIX_ACL_XATTR_DEFAULT) == 0) - type = ACL_TYPE_DEFAULT; - else - return -EOPNOTSUPP; - - acl = posix_acl_from_xattr(&init_user_ns, value, size); - if (IS_ERR(acl)) - return PTR_ERR(acl); - error = nfs3_proc_setacl(inode, type, acl); - posix_acl_release(acl); - - return error; -} - -int nfs3_removexattr(struct dentry *dentry, const char *name) -{ - struct inode *inode = dentry->d_inode; - int type; - - if (strcmp(name, POSIX_ACL_XATTR_ACCESS) == 0) - type = ACL_TYPE_ACCESS; - else if (strcmp(name, POSIX_ACL_XATTR_DEFAULT) == 0) - type = ACL_TYPE_DEFAULT; - else - return -EOPNOTSUPP; - - return nfs3_proc_setacl(inode, type, NULL); -} - -static void __nfs3_forget_cached_acls(struct nfs_inode *nfsi) -{ - if (!IS_ERR(nfsi->acl_access)) { - posix_acl_release(nfsi->acl_access); - nfsi->acl_access = ERR_PTR(-EAGAIN); - } - if (!IS_ERR(nfsi->acl_default)) { - posix_acl_release(nfsi->acl_default); - nfsi->acl_default = ERR_PTR(-EAGAIN); - } -} - -void nfs3_forget_cached_acls(struct inode *inode) -{ - dprintk("NFS: nfs3_forget_cached_acls(%s/%ld)\n", inode->i_sb->s_id, - inode->i_ino); - spin_lock(&inode->i_lock); - __nfs3_forget_cached_acls(NFS_I(inode)); - spin_unlock(&inode->i_lock); -} - -static struct posix_acl *nfs3_get_cached_acl(struct inode *inode, int type) -{ - struct nfs_inode *nfsi = NFS_I(inode); - struct posix_acl *acl = ERR_PTR(-EINVAL); - - spin_lock(&inode->i_lock); - switch(type) { - case ACL_TYPE_ACCESS: - acl = nfsi->acl_access; - break; - - case ACL_TYPE_DEFAULT: - acl = nfsi->acl_default; - break; - - default: - goto out; - } - if (IS_ERR(acl)) - acl = ERR_PTR(-EAGAIN); - else - acl = posix_acl_dup(acl); -out: - spin_unlock(&inode->i_lock); - dprintk("NFS: nfs3_get_cached_acl(%s/%ld, %d) = %p\n", inode->i_sb->s_id, - inode->i_ino, type, acl); - return acl; -} - -static void nfs3_cache_acls(struct inode *inode, struct posix_acl *acl, - struct posix_acl *dfacl) -{ - struct nfs_inode *nfsi = NFS_I(inode); - - dprintk("nfs3_cache_acls(%s/%ld, %p, %p)\n", inode->i_sb->s_id, - inode->i_ino, acl, dfacl); - spin_lock(&inode->i_lock); - __nfs3_forget_cached_acls(NFS_I(inode)); - if (!IS_ERR(acl)) - nfsi->acl_access = posix_acl_dup(acl); - if (!IS_ERR(dfacl)) - nfsi->acl_default = posix_acl_dup(dfacl); - spin_unlock(&inode->i_lock); -} - -struct posix_acl *nfs3_proc_getacl(struct inode *inode, int type) +struct posix_acl *nfs3_get_acl(struct inode *inode, int type) { struct nfs_server *server = NFS_SERVER(inode); struct page *pages[NFSACL_MAXPAGES] = { }; @@ -198,7 +26,6 @@ struct posix_acl *nfs3_proc_getacl(struct inode *inode, int type) .rpc_argp = &args, .rpc_resp = &res, }; - struct posix_acl *acl; int status, count; if (!nfs_server_capable(inode, NFS_CAP_ACLS)) @@ -207,10 +34,6 @@ struct posix_acl *nfs3_proc_getacl(struct inode *inode, int type) status = nfs_revalidate_inode(server, inode); if (status < 0) return ERR_PTR(status); - acl = nfs3_get_cached_acl(inode, type); - if (acl != ERR_PTR(-EAGAIN)) - return acl; - acl = NULL; /* * Only get the access acl when explicitly requested: We don't @@ -257,40 +80,41 @@ struct posix_acl *nfs3_proc_getacl(struct inode *inode, int type) } if (res.acl_access != NULL) { - if (posix_acl_equiv_mode(res.acl_access, NULL) == 0) { + if ((posix_acl_equiv_mode(res.acl_access, NULL) == 0) || + res.acl_access->a_count == 0) { posix_acl_release(res.acl_access); res.acl_access = NULL; } } - nfs3_cache_acls(inode, - (res.mask & NFS_ACL) ? res.acl_access : ERR_PTR(-EINVAL), - (res.mask & NFS_DFACL) ? res.acl_default : ERR_PTR(-EINVAL)); - switch(type) { - case ACL_TYPE_ACCESS: - acl = res.acl_access; - res.acl_access = NULL; - break; + if (res.mask & NFS_ACL) + set_cached_acl(inode, ACL_TYPE_ACCESS, res.acl_access); + else + forget_cached_acl(inode, ACL_TYPE_ACCESS); - case ACL_TYPE_DEFAULT: - acl = res.acl_default; - res.acl_default = NULL; + if (res.mask & NFS_DFACL) + set_cached_acl(inode, ACL_TYPE_DEFAULT, res.acl_default); + else + forget_cached_acl(inode, ACL_TYPE_DEFAULT); + + nfs_free_fattr(res.fattr); + if (type == ACL_TYPE_ACCESS) { + posix_acl_release(res.acl_default); + return res.acl_access; + } else { + posix_acl_release(res.acl_access); + return res.acl_default; } getout: posix_acl_release(res.acl_access); posix_acl_release(res.acl_default); nfs_free_fattr(res.fattr); - - if (status != 0) { - posix_acl_release(acl); - acl = ERR_PTR(status); - } - return acl; + return ERR_PTR(status); } -static int nfs3_proc_setacls(struct inode *inode, struct posix_acl *acl, - struct posix_acl *dfacl) +static int __nfs3_proc_setacls(struct inode *inode, struct posix_acl *acl, + struct posix_acl *dfacl) { struct nfs_server *server = NFS_SERVER(inode); struct nfs_fattr *fattr; @@ -353,7 +177,8 @@ static int nfs3_proc_setacls(struct inode *inode, struct posix_acl *acl, switch (status) { case 0: status = nfs_refresh_inode(inode, fattr); - nfs3_cache_acls(inode, acl, dfacl); + set_cached_acl(inode, ACL_TYPE_ACCESS, acl); + set_cached_acl(inode, ACL_TYPE_DEFAULT, dfacl); break; case -EPFNOSUPPORT: case -EPROTONOSUPPORT: @@ -373,40 +198,43 @@ out: return status; } -int nfs3_proc_setacl(struct inode *inode, int type, struct posix_acl *acl) +int nfs3_proc_setacls(struct inode *inode, struct posix_acl *acl, + struct posix_acl *dfacl) +{ + int ret; + ret = __nfs3_proc_setacls(inode, acl, dfacl); + return (ret == -EOPNOTSUPP) ? 0 : ret; + +} + +int nfs3_set_acl(struct inode *inode, struct posix_acl *acl, int type) { struct posix_acl *alloc = NULL, *dfacl = NULL; int status; if (S_ISDIR(inode->i_mode)) { switch(type) { - case ACL_TYPE_ACCESS: - alloc = dfacl = nfs3_proc_getacl(inode, - ACL_TYPE_DEFAULT); - if (IS_ERR(alloc)) - goto fail; - break; - - case ACL_TYPE_DEFAULT: - dfacl = acl; - alloc = acl = nfs3_proc_getacl(inode, - ACL_TYPE_ACCESS); - if (IS_ERR(alloc)) - goto fail; - break; - - default: - return -EINVAL; + case ACL_TYPE_ACCESS: + alloc = dfacl = get_acl(inode, ACL_TYPE_DEFAULT); + if (IS_ERR(alloc)) + goto fail; + break; + + case ACL_TYPE_DEFAULT: + dfacl = acl; + alloc = acl = get_acl(inode, ACL_TYPE_ACCESS); + if (IS_ERR(alloc)) + goto fail; + break; } - } else if (type != ACL_TYPE_ACCESS) - return -EINVAL; + } if (acl == NULL) { alloc = acl = posix_acl_from_mode(inode->i_mode, GFP_KERNEL); if (IS_ERR(alloc)) goto fail; } - status = nfs3_proc_setacls(inode, acl, dfacl); + status = __nfs3_proc_setacls(inode, acl, dfacl); posix_acl_release(alloc); return status; @@ -414,27 +242,51 @@ fail: return PTR_ERR(alloc); } -int nfs3_proc_set_default_acl(struct inode *dir, struct inode *inode, - umode_t mode) +const struct xattr_handler *nfs3_xattr_handlers[] = { + &posix_acl_access_xattr_handler, + &posix_acl_default_xattr_handler, + NULL, +}; + +static int +nfs3_list_one_acl(struct inode *inode, int type, const char *name, void *data, + size_t size, ssize_t *result) { - struct posix_acl *dfacl, *acl; - int error = 0; + struct posix_acl *acl; + char *p = data + *result; - dfacl = nfs3_proc_getacl(dir, ACL_TYPE_DEFAULT); - if (IS_ERR(dfacl)) { - error = PTR_ERR(dfacl); - return (error == -EOPNOTSUPP) ? 0 : error; - } - if (!dfacl) + acl = get_acl(inode, type); + if (!acl) return 0; - acl = posix_acl_dup(dfacl); - error = posix_acl_create(&acl, GFP_KERNEL, &mode); - if (error < 0) - goto out_release_dfacl; - error = nfs3_proc_setacls(inode, acl, S_ISDIR(inode->i_mode) ? - dfacl : NULL); + posix_acl_release(acl); -out_release_dfacl: - posix_acl_release(dfacl); - return error; + + *result += strlen(name); + *result += 1; + if (!size) + return 0; + if (*result > size) + return -ERANGE; + + strcpy(p, name); + return 0; +} + +ssize_t +nfs3_listxattr(struct dentry *dentry, char *data, size_t size) +{ + struct inode *inode = dentry->d_inode; + ssize_t result = 0; + int error; + + error = nfs3_list_one_acl(inode, ACL_TYPE_ACCESS, + POSIX_ACL_XATTR_ACCESS, data, size, &result); + if (error) + return error; + + error = nfs3_list_one_acl(inode, ACL_TYPE_DEFAULT, + POSIX_ACL_XATTR_DEFAULT, data, size, &result); + if (error) + return error; + return result; } diff --git a/fs/nfs/nfs3proc.c b/fs/nfs/nfs3proc.c index 01b6f6a49d1..f0afa291fd5 100644 --- a/fs/nfs/nfs3proc.c +++ b/fs/nfs/nfs3proc.c @@ -18,6 +18,7 @@ #include <linux/lockd/bind.h> #include <linux/nfs_mount.h> #include <linux/freezer.h> +#include <linux/xattr.h> #include "iostat.h" #include "internal.h" @@ -317,8 +318,8 @@ static int nfs3_proc_create(struct inode *dir, struct dentry *dentry, struct iattr *sattr, int flags) { + struct posix_acl *default_acl, *acl; struct nfs3_createdata *data; - umode_t mode = sattr->ia_mode; int status = -ENOMEM; dprintk("NFS call create %pd\n", dentry); @@ -340,7 +341,9 @@ nfs3_proc_create(struct inode *dir, struct dentry *dentry, struct iattr *sattr, data->arg.create.verifier[1] = cpu_to_be32(current->pid); } - sattr->ia_mode &= ~current_umask(); + status = posix_acl_create(dir, &sattr->ia_mode, &default_acl, &acl); + if (status) + goto out; for (;;) { status = nfs3_do_create(dir, dentry, data); @@ -366,7 +369,7 @@ nfs3_proc_create(struct inode *dir, struct dentry *dentry, struct iattr *sattr, } if (status != 0) - goto out; + goto out_release_acls; /* When we created the file with exclusive semantics, make * sure we set the attributes afterwards. */ @@ -385,9 +388,14 @@ nfs3_proc_create(struct inode *dir, struct dentry *dentry, struct iattr *sattr, nfs_post_op_update_inode(dentry->d_inode, data->res.fattr); dprintk("NFS reply setattr (post-create): %d\n", status); if (status != 0) - goto out; + goto out_release_acls; } - status = nfs3_proc_set_default_acl(dir, dentry->d_inode, mode); + + status = nfs3_proc_setacls(dentry->d_inode, acl, default_acl); + +out_release_acls: + posix_acl_release(acl); + posix_acl_release(default_acl); out: nfs3_free_createdata(data); dprintk("NFS reply create: %d\n", status); @@ -471,41 +479,6 @@ nfs3_proc_rename_done(struct rpc_task *task, struct inode *old_dir, } static int -nfs3_proc_rename(struct inode *old_dir, struct qstr *old_name, - struct inode *new_dir, struct qstr *new_name) -{ - struct nfs_renameargs arg = { - .old_dir = NFS_FH(old_dir), - .old_name = old_name, - .new_dir = NFS_FH(new_dir), - .new_name = new_name, - }; - struct nfs_renameres res; - struct rpc_message msg = { - .rpc_proc = &nfs3_procedures[NFS3PROC_RENAME], - .rpc_argp = &arg, - .rpc_resp = &res, - }; - int status = -ENOMEM; - - dprintk("NFS call rename %s -> %s\n", old_name->name, new_name->name); - - res.old_fattr = nfs_alloc_fattr(); - res.new_fattr = nfs_alloc_fattr(); - if (res.old_fattr == NULL || res.new_fattr == NULL) - goto out; - - status = rpc_call_sync(NFS_CLIENT(old_dir), &msg, 0); - nfs_post_op_update_inode(old_dir, res.old_fattr); - nfs_post_op_update_inode(new_dir, res.new_fattr); -out: - nfs_free_fattr(res.old_fattr); - nfs_free_fattr(res.new_fattr); - dprintk("NFS reply rename: %d\n", status); - return status; -} - -static int nfs3_proc_link(struct inode *inode, struct inode *dir, struct qstr *name) { struct nfs3_linkargs arg = { @@ -572,18 +545,20 @@ out: static int nfs3_proc_mkdir(struct inode *dir, struct dentry *dentry, struct iattr *sattr) { + struct posix_acl *default_acl, *acl; struct nfs3_createdata *data; - umode_t mode = sattr->ia_mode; int status = -ENOMEM; dprintk("NFS call mkdir %pd\n", dentry); - sattr->ia_mode &= ~current_umask(); - data = nfs3_alloc_createdata(); if (data == NULL) goto out; + status = posix_acl_create(dir, &sattr->ia_mode, &default_acl, &acl); + if (status) + goto out; + data->msg.rpc_proc = &nfs3_procedures[NFS3PROC_MKDIR]; data->arg.mkdir.fh = NFS_FH(dir); data->arg.mkdir.name = dentry->d_name.name; @@ -592,9 +567,13 @@ nfs3_proc_mkdir(struct inode *dir, struct dentry *dentry, struct iattr *sattr) status = nfs3_do_create(dir, dentry, data); if (status != 0) - goto out; + goto out_release_acls; - status = nfs3_proc_set_default_acl(dir, dentry->d_inode, mode); + status = nfs3_proc_setacls(dentry->d_inode, acl, default_acl); + +out_release_acls: + posix_acl_release(acl); + posix_acl_release(default_acl); out: nfs3_free_createdata(data); dprintk("NFS reply mkdir: %d\n", status); @@ -691,19 +670,21 @@ static int nfs3_proc_mknod(struct inode *dir, struct dentry *dentry, struct iattr *sattr, dev_t rdev) { + struct posix_acl *default_acl, *acl; struct nfs3_createdata *data; - umode_t mode = sattr->ia_mode; int status = -ENOMEM; dprintk("NFS call mknod %pd %u:%u\n", dentry, MAJOR(rdev), MINOR(rdev)); - sattr->ia_mode &= ~current_umask(); - data = nfs3_alloc_createdata(); if (data == NULL) goto out; + status = posix_acl_create(dir, &sattr->ia_mode, &default_acl, &acl); + if (status) + goto out; + data->msg.rpc_proc = &nfs3_procedures[NFS3PROC_MKNOD]; data->arg.mknod.fh = NFS_FH(dir); data->arg.mknod.name = dentry->d_name.name; @@ -731,8 +712,13 @@ nfs3_proc_mknod(struct inode *dir, struct dentry *dentry, struct iattr *sattr, status = nfs3_do_create(dir, dentry, data); if (status != 0) - goto out; - status = nfs3_proc_set_default_acl(dir, dentry->d_inode, mode); + goto out_release_acls; + + status = nfs3_proc_setacls(dentry->d_inode, acl, default_acl); + +out_release_acls: + posix_acl_release(acl); + posix_acl_release(default_acl); out: nfs3_free_createdata(data); dprintk("NFS reply mknod: %d\n", status); @@ -809,7 +795,7 @@ nfs3_proc_pathconf(struct nfs_server *server, struct nfs_fh *fhandle, return status; } -static int nfs3_read_done(struct rpc_task *task, struct nfs_read_data *data) +static int nfs3_read_done(struct rpc_task *task, struct nfs_pgio_data *data) { struct inode *inode = data->header->inode; @@ -821,18 +807,18 @@ static int nfs3_read_done(struct rpc_task *task, struct nfs_read_data *data) return 0; } -static void nfs3_proc_read_setup(struct nfs_read_data *data, struct rpc_message *msg) +static void nfs3_proc_read_setup(struct nfs_pgio_data *data, struct rpc_message *msg) { msg->rpc_proc = &nfs3_procedures[NFS3PROC_READ]; } -static int nfs3_proc_read_rpc_prepare(struct rpc_task *task, struct nfs_read_data *data) +static int nfs3_proc_pgio_rpc_prepare(struct rpc_task *task, struct nfs_pgio_data *data) { rpc_call_start(task); return 0; } -static int nfs3_write_done(struct rpc_task *task, struct nfs_write_data *data) +static int nfs3_write_done(struct rpc_task *task, struct nfs_pgio_data *data) { struct inode *inode = data->header->inode; @@ -843,17 +829,11 @@ static int nfs3_write_done(struct rpc_task *task, struct nfs_write_data *data) return 0; } -static void nfs3_proc_write_setup(struct nfs_write_data *data, struct rpc_message *msg) +static void nfs3_proc_write_setup(struct nfs_pgio_data *data, struct rpc_message *msg) { msg->rpc_proc = &nfs3_procedures[NFS3PROC_WRITE]; } -static int nfs3_proc_write_rpc_prepare(struct rpc_task *task, struct nfs_write_data *data) -{ - rpc_call_start(task); - return 0; -} - static void nfs3_proc_commit_rpc_prepare(struct rpc_task *task, struct nfs_commit_data *data) { rpc_call_start(task); @@ -904,20 +884,28 @@ static const struct inode_operations nfs3_dir_inode_operations = { .permission = nfs_permission, .getattr = nfs_getattr, .setattr = nfs_setattr, +#ifdef CONFIG_NFS_V3_ACL .listxattr = nfs3_listxattr, - .getxattr = nfs3_getxattr, - .setxattr = nfs3_setxattr, - .removexattr = nfs3_removexattr, + .getxattr = generic_getxattr, + .setxattr = generic_setxattr, + .removexattr = generic_removexattr, + .get_acl = nfs3_get_acl, + .set_acl = nfs3_set_acl, +#endif }; static const struct inode_operations nfs3_file_inode_operations = { .permission = nfs_permission, .getattr = nfs_getattr, .setattr = nfs_setattr, +#ifdef CONFIG_NFS_V3_ACL .listxattr = nfs3_listxattr, - .getxattr = nfs3_getxattr, - .setxattr = nfs3_setxattr, - .removexattr = nfs3_removexattr, + .getxattr = generic_getxattr, + .setxattr = generic_setxattr, + .removexattr = generic_removexattr, + .get_acl = nfs3_get_acl, + .set_acl = nfs3_set_acl, +#endif }; const struct nfs_rpc_ops nfs_v3_clientops = { @@ -939,7 +927,6 @@ const struct nfs_rpc_ops nfs_v3_clientops = { .unlink_setup = nfs3_proc_unlink_setup, .unlink_rpc_prepare = nfs3_proc_unlink_rpc_prepare, .unlink_done = nfs3_proc_unlink_done, - .rename = nfs3_proc_rename, .rename_setup = nfs3_proc_rename_setup, .rename_rpc_prepare = nfs3_proc_rename_rpc_prepare, .rename_done = nfs3_proc_rename_done, @@ -953,19 +940,16 @@ const struct nfs_rpc_ops nfs_v3_clientops = { .fsinfo = nfs3_proc_fsinfo, .pathconf = nfs3_proc_pathconf, .decode_dirent = nfs3_decode_dirent, + .pgio_rpc_prepare = nfs3_proc_pgio_rpc_prepare, .read_setup = nfs3_proc_read_setup, - .read_pageio_init = nfs_pageio_init_read, - .read_rpc_prepare = nfs3_proc_read_rpc_prepare, .read_done = nfs3_read_done, .write_setup = nfs3_proc_write_setup, - .write_pageio_init = nfs_pageio_init_write, - .write_rpc_prepare = nfs3_proc_write_rpc_prepare, .write_done = nfs3_write_done, .commit_setup = nfs3_proc_commit_setup, .commit_rpc_prepare = nfs3_proc_commit_rpc_prepare, .commit_done = nfs3_commit_done, .lock = nfs3_proc_lock, - .clear_acl_cache = nfs3_forget_cached_acls, + .clear_acl_cache = forget_all_cached_acls, .close_context = nfs_close_context, .have_delegation = nfs3_have_delegation, .return_delegation = nfs3_return_delegation, diff --git a/fs/nfs/nfs3super.c b/fs/nfs/nfs3super.c index cc471c72523..d6a98949af1 100644 --- a/fs/nfs/nfs3super.c +++ b/fs/nfs/nfs3super.c @@ -12,6 +12,9 @@ static struct nfs_subversion nfs_v3 = { .rpc_vers = &nfs_version3, .rpc_ops = &nfs_v3_clientops, .sops = &nfs_sops, +#ifdef CONFIG_NFS_V3_ACL + .xattr = nfs3_xattr_handlers, +#endif }; static int __init init_nfs_v3(void) diff --git a/fs/nfs/nfs3xdr.c b/fs/nfs/nfs3xdr.c index fa6d72131c1..8f4cbe7f4aa 100644 --- a/fs/nfs/nfs3xdr.c +++ b/fs/nfs/nfs3xdr.c @@ -953,7 +953,7 @@ static void nfs3_xdr_enc_readlink3args(struct rpc_rqst *req, * }; */ static void encode_read3args(struct xdr_stream *xdr, - const struct nfs_readargs *args) + const struct nfs_pgio_args *args) { __be32 *p; @@ -966,7 +966,7 @@ static void encode_read3args(struct xdr_stream *xdr, static void nfs3_xdr_enc_read3args(struct rpc_rqst *req, struct xdr_stream *xdr, - const struct nfs_readargs *args) + const struct nfs_pgio_args *args) { encode_read3args(xdr, args); prepare_reply_buffer(req, args->pages, args->pgbase, @@ -992,7 +992,7 @@ static void nfs3_xdr_enc_read3args(struct rpc_rqst *req, * }; */ static void encode_write3args(struct xdr_stream *xdr, - const struct nfs_writeargs *args) + const struct nfs_pgio_args *args) { __be32 *p; @@ -1008,7 +1008,7 @@ static void encode_write3args(struct xdr_stream *xdr, static void nfs3_xdr_enc_write3args(struct rpc_rqst *req, struct xdr_stream *xdr, - const struct nfs_writeargs *args) + const struct nfs_pgio_args *args) { encode_write3args(xdr, args); xdr->buf->flags |= XDRBUF_WRITE; @@ -1589,7 +1589,7 @@ out_default: * }; */ static int decode_read3resok(struct xdr_stream *xdr, - struct nfs_readres *result) + struct nfs_pgio_res *result) { u32 eof, count, ocount, recvd; __be32 *p; @@ -1625,7 +1625,7 @@ out_overflow: } static int nfs3_xdr_dec_read3res(struct rpc_rqst *req, struct xdr_stream *xdr, - struct nfs_readres *result) + struct nfs_pgio_res *result) { enum nfs_stat status; int error; @@ -1673,7 +1673,7 @@ out_status: * }; */ static int decode_write3resok(struct xdr_stream *xdr, - struct nfs_writeres *result) + struct nfs_pgio_res *result) { __be32 *p; @@ -1697,7 +1697,7 @@ out_eio: } static int nfs3_xdr_dec_write3res(struct rpc_rqst *req, struct xdr_stream *xdr, - struct nfs_writeres *result) + struct nfs_pgio_res *result) { enum nfs_stat status; int error; diff --git a/fs/nfs/nfs4_fs.h b/fs/nfs/nfs4_fs.h index 3ce79b04522..ba2affa5194 100644 --- a/fs/nfs/nfs4_fs.h +++ b/fs/nfs/nfs4_fs.h @@ -9,6 +9,14 @@ #ifndef __LINUX_FS_NFS_NFS4_FS_H #define __LINUX_FS_NFS_NFS4_FS_H +#if defined(CONFIG_NFS_V4_2) +#define NFS4_MAX_MINOR_VERSION 2 +#elif defined(CONFIG_NFS_V4_1) +#define NFS4_MAX_MINOR_VERSION 1 +#else +#define NFS4_MAX_MINOR_VERSION 0 +#endif + #if IS_ENABLED(CONFIG_NFS_V4) #define NFS4_MAX_LOOP_ON_RECOVER (10) @@ -222,7 +230,7 @@ int nfs_atomic_open(struct inode *, struct dentry *, struct file *, extern struct file_system_type nfs4_fs_type; /* nfs4namespace.c */ -struct rpc_clnt *nfs4_create_sec_client(struct rpc_clnt *, struct inode *, struct qstr *); +struct rpc_clnt *nfs4_negotiate_security(struct rpc_clnt *, struct inode *, struct qstr *); struct vfsmount *nfs4_submount(struct nfs_server *, struct dentry *, struct nfs_fh *, struct nfs_fattr *); int nfs4_replace_transport(struct nfs_server *server, @@ -262,6 +270,7 @@ static inline struct nfs4_session *nfs4_get_session(const struct nfs_server *ser extern int nfs41_setup_sequence(struct nfs4_session *session, struct nfs4_sequence_args *args, struct nfs4_sequence_res *res, struct rpc_task *task); +extern int nfs41_sequence_done(struct rpc_task *, struct nfs4_sequence_res *); extern int nfs4_proc_create_session(struct nfs_client *, struct rpc_cred *); extern int nfs4_proc_destroy_session(struct nfs4_session *, struct rpc_cred *); extern int nfs4_proc_get_lease_time(struct nfs_client *clp, @@ -328,7 +337,7 @@ nfs4_state_protect(struct nfs_client *clp, unsigned long sp4_mode, */ static inline void nfs4_state_protect_write(struct nfs_client *clp, struct rpc_clnt **clntp, - struct rpc_message *msg, struct nfs_write_data *wdata) + struct rpc_message *msg, struct nfs_pgio_data *wdata) { if (_nfs4_state_protect(clp, NFS_SP4_MACH_CRED_WRITE, clntp, msg) && !test_bit(NFS_SP4_MACH_CRED_COMMIT, &clp->cl_sp4_flags)) @@ -360,7 +369,7 @@ nfs4_state_protect(struct nfs_client *clp, unsigned long sp4_flags, static inline void nfs4_state_protect_write(struct nfs_client *clp, struct rpc_clnt **clntp, - struct rpc_message *msg, struct nfs_write_data *wdata) + struct rpc_message *msg, struct nfs_pgio_data *wdata) { } #endif /* CONFIG_NFS_V4_1 */ @@ -418,6 +427,7 @@ extern void nfs4_close_sync(struct nfs4_state *, fmode_t); extern void nfs4_state_set_mode_locked(struct nfs4_state *, fmode_t); extern void nfs_inode_find_state_and_recover(struct inode *inode, const nfs4_stateid *stateid); +extern int nfs4_state_mark_reclaim_nograce(struct nfs_client *, struct nfs4_state *); extern void nfs4_schedule_lease_recovery(struct nfs_client *); extern int nfs4_wait_clnt_recover(struct nfs_client *clp); extern int nfs4_client_recover_expired_lease(struct nfs_client *clp); @@ -491,6 +501,16 @@ static inline bool nfs4_stateid_match(const nfs4_stateid *dst, const nfs4_statei return memcmp(dst, src, sizeof(*dst)) == 0; } +static inline bool nfs4_stateid_match_other(const nfs4_stateid *dst, const nfs4_stateid *src) +{ + return memcmp(dst->other, src->other, NFS4_STATEID_OTHER_SIZE) == 0; +} + +static inline bool nfs4_stateid_is_newer(const nfs4_stateid *s1, const nfs4_stateid *s2) +{ + return (s32)(be32_to_cpu(s1->seqid) - be32_to_cpu(s2->seqid)) > 0; +} + static inline bool nfs4_valid_open_stateid(const struct nfs4_state *state) { return test_bit(NFS_STATE_RECOVERY_FAILED, &state->flags) == 0; diff --git a/fs/nfs/nfs4client.c b/fs/nfs/nfs4client.c index b4a160a405c..aa9ef487604 100644 --- a/fs/nfs/nfs4client.c +++ b/fs/nfs/nfs4client.c @@ -10,6 +10,7 @@ #include <linux/sunrpc/auth.h> #include <linux/sunrpc/xprt.h> #include <linux/sunrpc/bc_xprt.h> +#include <linux/sunrpc/rpc_pipe_fs.h> #include "internal.h" #include "callback.h" #include "delegation.h" @@ -169,7 +170,7 @@ void nfs41_shutdown_client(struct nfs_client *clp) void nfs40_shutdown_client(struct nfs_client *clp) { if (clp->cl_slot_tbl) { - nfs4_release_slot_table(clp->cl_slot_tbl); + nfs4_shutdown_slot_table(clp->cl_slot_tbl); kfree(clp->cl_slot_tbl); } } @@ -370,6 +371,7 @@ struct nfs_client *nfs4_init_client(struct nfs_client *clp, __set_bit(NFS_CS_INFINITE_SLOTS, &clp->cl_flags); __set_bit(NFS_CS_DISCRTRY, &clp->cl_flags); __set_bit(NFS_CS_NO_RETRANS_TIMEOUT, &clp->cl_flags); + error = nfs_create_rpc_client(clp, timeparms, RPC_AUTH_GSS_KRB5I); if (error == -EINVAL) error = nfs_create_rpc_client(clp, timeparms, RPC_AUTH_UNIX); @@ -409,13 +411,11 @@ struct nfs_client *nfs4_init_client(struct nfs_client *clp, error = nfs4_discover_server_trunking(clp, &old); if (error < 0) goto error; - nfs_put_client(clp); - if (clp != old) { - clp->cl_preserve_clid = true; - clp = old; - } - return clp; + if (clp != old) + clp->cl_preserve_clid = true; + nfs_put_client(clp); + return old; error: nfs_mark_client_ready(clp, error); @@ -493,9 +493,10 @@ int nfs40_walk_client_list(struct nfs_client *new, prev = pos; status = nfs_wait_client_init_complete(pos); - spin_lock(&nn->nfs_client_lock); if (status < 0) - continue; + goto out; + status = -NFS4ERR_STALE_CLIENTID; + spin_lock(&nn->nfs_client_lock); } if (pos->cl_cons_state != NFS_CS_READY) continue; @@ -530,6 +531,13 @@ int nfs40_walk_client_list(struct nfs_client *new, *result = pos; dprintk("NFS: <-- %s using nfs_client = %p ({%d})\n", __func__, pos, atomic_read(&pos->cl_count)); + goto out; + case -ERESTARTSYS: + case -ETIMEDOUT: + /* The callback path may have been inadvertently + * changed. Schedule recovery! + */ + nfs4_schedule_path_down_recovery(pos); default: goto out; } @@ -633,7 +641,8 @@ int nfs41_walk_client_list(struct nfs_client *new, } spin_lock(&nn->nfs_client_lock); if (status < 0) - continue; + break; + status = -NFS4ERR_STALE_CLIENTID; } if (pos->cl_cons_state != NFS_CS_READY) continue; @@ -1133,6 +1142,7 @@ static int nfs_probe_destination(struct nfs_server *server) * @hostname: new end-point's hostname * @sap: new end-point's socket address * @salen: size of "sap" + * @net: net namespace * * The nfs_server must be quiescent before this function is invoked. * Either its session is drained (NFSv4.1+), or its transport is @@ -1141,13 +1151,13 @@ static int nfs_probe_destination(struct nfs_server *server) * Returns zero on success, or a negative errno value. */ int nfs4_update_server(struct nfs_server *server, const char *hostname, - struct sockaddr *sap, size_t salen) + struct sockaddr *sap, size_t salen, struct net *net) { struct nfs_client *clp = server->nfs_client; struct rpc_clnt *clnt = server->client; struct xprt_create xargs = { .ident = clp->cl_proto, - .net = &init_net, + .net = net, .dstaddr = sap, .addrlen = salen, .servername = hostname, @@ -1187,7 +1197,7 @@ int nfs4_update_server(struct nfs_server *server, const char *hostname, error = nfs4_set_client(server, hostname, sap, salen, buf, clp->cl_rpcclient->cl_auth->au_flavor, clp->cl_proto, clnt->cl_timeout, - clp->cl_minorversion, clp->cl_net); + clp->cl_minorversion, net); nfs_put_client(clp); if (error != 0) { nfs_server_insert_lists(server); diff --git a/fs/nfs/nfs4file.c b/fs/nfs/nfs4file.c index 8de3407e036..a816f0627a6 100644 --- a/fs/nfs/nfs4file.c +++ b/fs/nfs/nfs4file.c @@ -100,8 +100,7 @@ nfs4_file_fsync(struct file *file, loff_t start, loff_t end, int datasync) break; mutex_lock(&inode->i_mutex); ret = nfs_file_fsync_commit(file, start, end, datasync); - if (!ret && !datasync) - /* application has asked for meta-data sync */ + if (!ret) ret = pnfs_layoutcommit_inode(inode, true); mutex_unlock(&inode->i_mutex); /* @@ -118,10 +117,10 @@ nfs4_file_fsync(struct file *file, loff_t start, loff_t end, int datasync) const struct file_operations nfs4_file_operations = { .llseek = nfs_file_llseek, - .read = do_sync_read, - .write = do_sync_write, - .aio_read = nfs_file_read, - .aio_write = nfs_file_write, + .read = new_sync_read, + .write = new_sync_write, + .read_iter = nfs_file_read, + .write_iter = nfs_file_write, .mmap = nfs_file_mmap, .open = nfs4_file_open, .flush = nfs_file_flush, @@ -130,7 +129,7 @@ const struct file_operations nfs4_file_operations = { .lock = nfs_lock, .flock = nfs_flock, .splice_read = nfs_file_splice_read, - .splice_write = nfs_file_splice_write, + .splice_write = iter_file_splice_write, .check_flags = nfs_check_flags, .setlease = nfs_setlease, }; diff --git a/fs/nfs/nfs4namespace.c b/fs/nfs/nfs4namespace.c index 4e7f05d3e9d..3d83cb1fdc7 100644 --- a/fs/nfs/nfs4namespace.c +++ b/fs/nfs/nfs4namespace.c @@ -121,9 +121,8 @@ static int nfs4_validate_fspath(struct dentry *dentry, } static size_t nfs_parse_server_name(char *string, size_t len, - struct sockaddr *sa, size_t salen, struct nfs_server *server) + struct sockaddr *sa, size_t salen, struct net *net) { - struct net *net = rpc_net_ns(server->client); ssize_t ret; ret = rpc_pton(net, string, len, sa, salen); @@ -140,16 +139,22 @@ static size_t nfs_parse_server_name(char *string, size_t len, * @server: NFS server struct * @flavors: List of security tuples returned by SECINFO procedure * - * Return the pseudoflavor of the first security mechanism in - * "flavors" that is locally supported. Return RPC_AUTH_UNIX if - * no matching flavor is found in the array. The "flavors" array + * Return an rpc client that uses the first security mechanism in + * "flavors" that is locally supported. The "flavors" array * is searched in the order returned from the server, per RFC 3530 - * recommendation. + * recommendation and each flavor is checked for membership in the + * sec= mount option list if it exists. + * + * Return -EPERM if no matching flavor is found in the array. + * + * Please call rpc_shutdown_client() when you are done with this rpc client. + * */ -static rpc_authflavor_t nfs_find_best_sec(struct nfs_server *server, +static struct rpc_clnt *nfs_find_best_sec(struct rpc_clnt *clnt, + struct nfs_server *server, struct nfs4_secinfo_flavors *flavors) { - rpc_authflavor_t pseudoflavor; + rpc_authflavor_t pflavor; struct nfs4_secinfo4 *secinfo; unsigned int i; @@ -160,62 +165,73 @@ static rpc_authflavor_t nfs_find_best_sec(struct nfs_server *server, case RPC_AUTH_NULL: case RPC_AUTH_UNIX: case RPC_AUTH_GSS: - pseudoflavor = rpcauth_get_pseudoflavor(secinfo->flavor, + pflavor = rpcauth_get_pseudoflavor(secinfo->flavor, &secinfo->flavor_info); - /* make sure pseudoflavor matches sec= mount opt */ - if (pseudoflavor != RPC_AUTH_MAXFLAVOR && - nfs_auth_info_match(&server->auth_info, - pseudoflavor)) - return pseudoflavor; - break; + /* does the pseudoflavor match a sec= mount opt? */ + if (pflavor != RPC_AUTH_MAXFLAVOR && + nfs_auth_info_match(&server->auth_info, pflavor)) { + struct rpc_clnt *new; + struct rpc_cred *cred; + + /* Cloning creates an rpc_auth for the flavor */ + new = rpc_clone_client_set_auth(clnt, pflavor); + if (IS_ERR(new)) + continue; + /** + * Check that the user actually can use the + * flavor. This is mostly for RPC_AUTH_GSS + * where cr_init obtains a gss context + */ + cred = rpcauth_lookupcred(new->cl_auth, 0); + if (IS_ERR(cred)) { + rpc_shutdown_client(new); + continue; + } + put_rpccred(cred); + return new; + } } } - - /* if there were any sec= options then nothing matched */ - if (server->auth_info.flavor_len > 0) - return -EPERM; - - return RPC_AUTH_UNIX; + return ERR_PTR(-EPERM); } -static rpc_authflavor_t nfs4_negotiate_security(struct inode *inode, struct qstr *name) +/** + * nfs4_negotiate_security - in response to an NFS4ERR_WRONGSEC on lookup, + * return an rpc_clnt that uses the best available security flavor with + * respect to the secinfo flavor list and the sec= mount options. + * + * @clnt: RPC client to clone + * @inode: directory inode + * @name: lookup name + * + * Please call rpc_shutdown_client() when you are done with this rpc client. + */ +struct rpc_clnt * +nfs4_negotiate_security(struct rpc_clnt *clnt, struct inode *inode, + struct qstr *name) { struct page *page; struct nfs4_secinfo_flavors *flavors; - rpc_authflavor_t flavor; + struct rpc_clnt *new; int err; page = alloc_page(GFP_KERNEL); if (!page) - return -ENOMEM; + return ERR_PTR(-ENOMEM); + flavors = page_address(page); err = nfs4_proc_secinfo(inode, name, flavors); if (err < 0) { - flavor = err; + new = ERR_PTR(err); goto out; } - flavor = nfs_find_best_sec(NFS_SERVER(inode), flavors); + new = nfs_find_best_sec(clnt, NFS_SERVER(inode), flavors); out: put_page(page); - return flavor; -} - -/* - * Please call rpc_shutdown_client() when you are done with this client. - */ -struct rpc_clnt *nfs4_create_sec_client(struct rpc_clnt *clnt, struct inode *inode, - struct qstr *name) -{ - rpc_authflavor_t flavor; - - flavor = nfs4_negotiate_security(inode, name); - if ((int)flavor < 0) - return ERR_PTR((int)flavor); - - return rpc_clone_client_set_auth(clnt, flavor); + return new; } static struct vfsmount *try_location(struct nfs_clone_mount *mountdata, @@ -223,6 +239,7 @@ static struct vfsmount *try_location(struct nfs_clone_mount *mountdata, const struct nfs4_fs_location *location) { const size_t addr_bufsize = sizeof(struct sockaddr_storage); + struct net *net = rpc_net_ns(NFS_SB(mountdata->sb)->client); struct vfsmount *mnt = ERR_PTR(-ENOENT); char *mnt_path; unsigned int maxbuflen; @@ -248,8 +265,7 @@ static struct vfsmount *try_location(struct nfs_clone_mount *mountdata, continue; mountdata->addrlen = nfs_parse_server_name(buf->data, buf->len, - mountdata->addr, addr_bufsize, - NFS_SB(mountdata->sb)); + mountdata->addr, addr_bufsize, net); if (mountdata->addrlen == 0) continue; @@ -398,11 +414,6 @@ struct vfsmount *nfs4_submount(struct nfs_server *server, struct dentry *dentry, if (client->cl_auth->au_flavor != flavor) flavor = client->cl_auth->au_flavor; - else { - rpc_authflavor_t new = nfs4_negotiate_security(dir, name); - if ((int)new >= 0) - flavor = new; - } mnt = nfs_do_submount(dentry, fh, fattr, flavor); out: rpc_shutdown_client(client); @@ -419,6 +430,7 @@ static int nfs4_try_replacing_one_location(struct nfs_server *server, const struct nfs4_fs_location *location) { const size_t addr_bufsize = sizeof(struct sockaddr_storage); + struct net *net = rpc_net_ns(server->client); struct sockaddr *sap; unsigned int s; size_t salen; @@ -440,7 +452,7 @@ static int nfs4_try_replacing_one_location(struct nfs_server *server, continue; salen = nfs_parse_server_name(buf->data, buf->len, - sap, addr_bufsize, server); + sap, addr_bufsize, net); if (salen == 0) continue; rpc_set_port(sap, NFS_PORT); @@ -450,7 +462,7 @@ static int nfs4_try_replacing_one_location(struct nfs_server *server, if (hostname == NULL) break; - error = nfs4_update_server(server, hostname, sap, salen); + error = nfs4_update_server(server, hostname, sap, salen, net); kfree(hostname); if (error == 0) break; diff --git a/fs/nfs/nfs4proc.c b/fs/nfs/nfs4proc.c index 659990c0109..4bf3d97cc5a 100644 --- a/fs/nfs/nfs4proc.c +++ b/fs/nfs/nfs4proc.c @@ -539,7 +539,7 @@ static int nfs40_sequence_done(struct rpc_task *task, struct nfs4_slot *slot = res->sr_slot; struct nfs4_slot_table *tbl; - if (!RPC_WAS_SENT(task)) + if (slot == NULL) goto out; tbl = slot->table; @@ -559,15 +559,10 @@ static void nfs41_sequence_free_slot(struct nfs4_sequence_res *res) { struct nfs4_session *session; struct nfs4_slot_table *tbl; + struct nfs4_slot *slot = res->sr_slot; bool send_new_highest_used_slotid = false; - if (!res->sr_slot) { - /* just wake up the next guy waiting since - * we may have not consumed a slot after all */ - dprintk("%s: No slot\n", __func__); - return; - } - tbl = res->sr_slot->table; + tbl = slot->table; session = tbl->session; spin_lock(&tbl->slot_tbl_lock); @@ -577,11 +572,11 @@ static void nfs41_sequence_free_slot(struct nfs4_sequence_res *res) if (tbl->highest_used_slotid > tbl->target_highest_slotid) send_new_highest_used_slotid = true; - if (nfs41_wake_and_assign_slot(tbl, res->sr_slot)) { + if (nfs41_wake_and_assign_slot(tbl, slot)) { send_new_highest_used_slotid = false; goto out_unlock; } - nfs4_free_slot(tbl, res->sr_slot); + nfs4_free_slot(tbl, slot); if (tbl->highest_used_slotid != NFS4_NO_SLOT) send_new_highest_used_slotid = false; @@ -592,19 +587,20 @@ out_unlock: nfs41_server_notify_highest_slotid_update(session->clp); } -static int nfs41_sequence_done(struct rpc_task *task, struct nfs4_sequence_res *res) +int nfs41_sequence_done(struct rpc_task *task, struct nfs4_sequence_res *res) { struct nfs4_session *session; - struct nfs4_slot *slot; + struct nfs4_slot *slot = res->sr_slot; struct nfs_client *clp; bool interrupted = false; int ret = 1; + if (slot == NULL) + goto out_noaction; /* don't increment the sequence number if the task wasn't sent */ if (!RPC_WAS_SENT(task)) goto out; - slot = res->sr_slot; session = slot->table->session; if (slot->interrupted) { @@ -679,6 +675,7 @@ out: /* The session may be reset by one of the error handlers. */ dprintk("%s: Error %d free the slot \n", __func__, res->sr_status); nfs41_sequence_free_slot(res); +out_noaction: return ret; retry_nowait: if (rpc_restart_call_prepare(task)) { @@ -692,6 +689,7 @@ out_retry: rpc_delay(task, NFS4_POLL_RETRY_MAX); return 0; } +EXPORT_SYMBOL_GPL(nfs41_sequence_done); static int nfs4_sequence_done(struct rpc_task *task, struct nfs4_sequence_res *res) @@ -1070,6 +1068,7 @@ static void nfs4_opendata_free(struct kref *kref) dput(p->dentry); nfs_sb_deactive(sb); nfs_fattr_free_names(&p->f_attr); + kfree(p->f_attr.mdsthreshold); kfree(p); } @@ -1139,12 +1138,71 @@ static void update_open_stateflags(struct nfs4_state *state, fmode_t fmode) nfs4_state_set_mode_locked(state, state->state | fmode); } -static void nfs_set_open_stateid_locked(struct nfs4_state *state, nfs4_stateid *stateid, fmode_t fmode) +static void nfs_test_and_clear_all_open_stateid(struct nfs4_state *state) { + struct nfs_client *clp = state->owner->so_server->nfs_client; + bool need_recover = false; + + if (test_and_clear_bit(NFS_O_RDONLY_STATE, &state->flags) && state->n_rdonly) + need_recover = true; + if (test_and_clear_bit(NFS_O_WRONLY_STATE, &state->flags) && state->n_wronly) + need_recover = true; + if (test_and_clear_bit(NFS_O_RDWR_STATE, &state->flags) && state->n_rdwr) + need_recover = true; + if (need_recover) + nfs4_state_mark_reclaim_nograce(clp, state); +} + +static bool nfs_need_update_open_stateid(struct nfs4_state *state, + nfs4_stateid *stateid) +{ + if (test_and_set_bit(NFS_OPEN_STATE, &state->flags) == 0) + return true; + if (!nfs4_stateid_match_other(stateid, &state->open_stateid)) { + nfs_test_and_clear_all_open_stateid(state); + return true; + } + if (nfs4_stateid_is_newer(stateid, &state->open_stateid)) + return true; + return false; +} + +static void nfs_clear_open_stateid_locked(struct nfs4_state *state, + nfs4_stateid *stateid, fmode_t fmode) +{ + clear_bit(NFS_O_RDWR_STATE, &state->flags); + switch (fmode & (FMODE_READ|FMODE_WRITE)) { + case FMODE_WRITE: + clear_bit(NFS_O_RDONLY_STATE, &state->flags); + break; + case FMODE_READ: + clear_bit(NFS_O_WRONLY_STATE, &state->flags); + break; + case 0: + clear_bit(NFS_O_RDONLY_STATE, &state->flags); + clear_bit(NFS_O_WRONLY_STATE, &state->flags); + clear_bit(NFS_OPEN_STATE, &state->flags); + } + if (stateid == NULL) + return; + if (!nfs_need_update_open_stateid(state, stateid)) + return; if (test_bit(NFS_DELEGATED_STATE, &state->flags) == 0) nfs4_stateid_copy(&state->stateid, stateid); nfs4_stateid_copy(&state->open_stateid, stateid); - set_bit(NFS_OPEN_STATE, &state->flags); +} + +static void nfs_clear_open_stateid(struct nfs4_state *state, nfs4_stateid *stateid, fmode_t fmode) +{ + write_seqlock(&state->seqlock); + nfs_clear_open_stateid_locked(state, stateid, fmode); + write_sequnlock(&state->seqlock); + if (test_bit(NFS_STATE_RECLAIM_NOGRACE, &state->flags)) + nfs4_schedule_state_manager(state->owner->so_server->nfs_client); +} + +static void nfs_set_open_stateid_locked(struct nfs4_state *state, nfs4_stateid *stateid, fmode_t fmode) +{ switch (fmode) { case FMODE_READ: set_bit(NFS_O_RDONLY_STATE, &state->flags); @@ -1155,13 +1213,11 @@ static void nfs_set_open_stateid_locked(struct nfs4_state *state, nfs4_stateid * case FMODE_READ|FMODE_WRITE: set_bit(NFS_O_RDWR_STATE, &state->flags); } -} - -static void nfs_set_open_stateid(struct nfs4_state *state, nfs4_stateid *stateid, fmode_t fmode) -{ - write_seqlock(&state->seqlock); - nfs_set_open_stateid_locked(state, stateid, fmode); - write_sequnlock(&state->seqlock); + if (!nfs_need_update_open_stateid(state, stateid)) + return; + if (test_bit(NFS_DELEGATED_STATE, &state->flags) == 0) + nfs4_stateid_copy(&state->stateid, stateid); + nfs4_stateid_copy(&state->open_stateid, stateid); } static void __update_open_stateid(struct nfs4_state *state, nfs4_stateid *open_stateid, const nfs4_stateid *deleg_stateid, fmode_t fmode) @@ -1219,6 +1275,8 @@ no_delegation: __update_open_stateid(state, open_stateid, NULL, fmode); ret = 1; } + if (test_bit(NFS_STATE_RECLAIM_NOGRACE, &state->flags)) + nfs4_schedule_state_manager(state->owner->so_server->nfs_client); return ret; } @@ -1452,12 +1510,15 @@ static int nfs4_open_recover(struct nfs4_opendata *opendata, struct nfs4_state * struct nfs4_state *newstate; int ret; + /* Don't trigger recovery in nfs_test_and_clear_all_open_stateid */ + clear_bit(NFS_O_RDWR_STATE, &state->flags); + clear_bit(NFS_O_WRONLY_STATE, &state->flags); + clear_bit(NFS_O_RDONLY_STATE, &state->flags); /* memory barrier prior to reading state->n_* */ clear_bit(NFS_DELEGATED_STATE, &state->flags); clear_bit(NFS_OPEN_STATE, &state->flags); smp_rmb(); if (state->n_rdwr != 0) { - clear_bit(NFS_O_RDWR_STATE, &state->flags); ret = nfs4_open_recover_helper(opendata, FMODE_READ|FMODE_WRITE, &newstate); if (ret != 0) return ret; @@ -1465,7 +1526,6 @@ static int nfs4_open_recover(struct nfs4_opendata *opendata, struct nfs4_state * return -ESTALE; } if (state->n_wronly != 0) { - clear_bit(NFS_O_WRONLY_STATE, &state->flags); ret = nfs4_open_recover_helper(opendata, FMODE_WRITE, &newstate); if (ret != 0) return ret; @@ -1473,7 +1533,6 @@ static int nfs4_open_recover(struct nfs4_opendata *opendata, struct nfs4_state * return -ESTALE; } if (state->n_rdonly != 0) { - clear_bit(NFS_O_RDONLY_STATE, &state->flags); ret = nfs4_open_recover_helper(opendata, FMODE_READ, &newstate); if (ret != 0) return ret; @@ -1622,15 +1681,15 @@ static void nfs4_open_confirm_prepare(struct rpc_task *task, void *calldata) { struct nfs4_opendata *data = calldata; - nfs40_setup_sequence(data->o_arg.server, &data->o_arg.seq_args, - &data->o_res.seq_res, task); + nfs40_setup_sequence(data->o_arg.server, &data->c_arg.seq_args, + &data->c_res.seq_res, task); } static void nfs4_open_confirm_done(struct rpc_task *task, void *calldata) { struct nfs4_opendata *data = calldata; - nfs40_sequence_done(task, &data->o_res.seq_res); + nfs40_sequence_done(task, &data->c_res.seq_res); data->rpc_status = task->tk_status; if (data->rpc_status == 0) { @@ -1688,7 +1747,7 @@ static int _nfs4_proc_open_confirm(struct nfs4_opendata *data) }; int status; - nfs4_init_sequence(&data->o_arg.seq_args, &data->o_res.seq_res, 1); + nfs4_init_sequence(&data->c_arg.seq_args, &data->c_res.seq_res, 1); kref_get(&data->kref); data->rpc_done = 0; data->rpc_status = 0; @@ -1968,7 +2027,7 @@ static int _nfs4_proc_open(struct nfs4_opendata *data) return status; } if (!(o_res->f_attr->valid & NFS_ATTR_FATTR)) - _nfs4_proc_getattr(server, &o_res->fh, o_res->f_attr, o_res->f_label); + nfs4_proc_getattr(server, &o_res->fh, o_res->f_attr, o_res->f_label); return 0; } @@ -2246,10 +2305,12 @@ static int _nfs4_do_open(struct inode *dir, } } - if (ctx_th && server->attr_bitmask[2] & FATTR4_WORD2_MDSTHRESHOLD) { - opendata->f_attr.mdsthreshold = pnfs_mdsthreshold_alloc(); - if (!opendata->f_attr.mdsthreshold) - goto err_free_label; + if (server->attr_bitmask[2] & FATTR4_WORD2_MDSTHRESHOLD) { + if (!opendata->f_attr.mdsthreshold) { + opendata->f_attr.mdsthreshold = pnfs_mdsthreshold_alloc(); + if (!opendata->f_attr.mdsthreshold) + goto err_free_label; + } opendata->o_arg.open_bitmap = &nfs4_pnfs_open_bitmap[0]; } if (dentry->d_inode != NULL) @@ -2277,11 +2338,10 @@ static int _nfs4_do_open(struct inode *dir, if (opendata->file_created) *opened |= FILE_CREATED; - if (pnfs_use_threshold(ctx_th, opendata->f_attr.mdsthreshold, server)) + if (pnfs_use_threshold(ctx_th, opendata->f_attr.mdsthreshold, server)) { *ctx_th = opendata->f_attr.mdsthreshold; - else - kfree(opendata->f_attr.mdsthreshold); - opendata->f_attr.mdsthreshold = NULL; + opendata->f_attr.mdsthreshold = NULL; + } nfs4_label_free(olabel); @@ -2291,7 +2351,6 @@ static int _nfs4_do_open(struct inode *dir, err_free_label: nfs4_label_free(olabel); err_opendata_put: - kfree(opendata->f_attr.mdsthreshold); nfs4_opendata_put(opendata); err_put_state_owner: nfs4_put_state_owner(sp); @@ -2400,13 +2459,16 @@ static int _nfs4_do_setattr(struct inode *inode, struct rpc_cred *cred, if (nfs4_copy_delegation_stateid(&arg.stateid, inode, fmode)) { /* Use that stateid */ - } else if (truncate && state != NULL && nfs4_valid_open_stateid(state)) { + } else if (truncate && state != NULL) { struct nfs_lockowner lockowner = { .l_owner = current->files, .l_pid = current->tgid, }; - nfs4_select_rw_stateid(&arg.stateid, state, FMODE_WRITE, - &lockowner); + if (!nfs4_valid_open_stateid(state)) + return -EBADF; + if (nfs4_select_rw_stateid(&arg.stateid, state, FMODE_WRITE, + &lockowner) == -EIO) + return -EBADF; } else nfs4_stateid_copy(&arg.stateid, &zero_stateid); @@ -2478,26 +2540,6 @@ static void nfs4_free_closedata(void *data) kfree(calldata); } -static void nfs4_close_clear_stateid_flags(struct nfs4_state *state, - fmode_t fmode) -{ - spin_lock(&state->owner->so_lock); - clear_bit(NFS_O_RDWR_STATE, &state->flags); - switch (fmode & (FMODE_READ|FMODE_WRITE)) { - case FMODE_WRITE: - clear_bit(NFS_O_RDONLY_STATE, &state->flags); - break; - case FMODE_READ: - clear_bit(NFS_O_WRONLY_STATE, &state->flags); - break; - case 0: - clear_bit(NFS_O_RDONLY_STATE, &state->flags); - clear_bit(NFS_O_WRONLY_STATE, &state->flags); - clear_bit(NFS_OPEN_STATE, &state->flags); - } - spin_unlock(&state->owner->so_lock); -} - static void nfs4_close_done(struct rpc_task *task, void *data) { struct nfs4_closedata *calldata = data; @@ -2516,11 +2558,10 @@ static void nfs4_close_done(struct rpc_task *task, void *data) if (calldata->roc) pnfs_roc_set_barrier(state->inode, calldata->roc_barrier); - nfs_set_open_stateid(state, &calldata->res.stateid, 0); + nfs_clear_open_stateid(state, &calldata->res.stateid, 0); renew_lease(server, calldata->timestamp); - nfs4_close_clear_stateid_flags(state, - calldata->arg.fmode); - break; + goto out_release; + case -NFS4ERR_ADMIN_REVOKED: case -NFS4ERR_STALE_STATEID: case -NFS4ERR_OLD_STATEID: case -NFS4ERR_BAD_STATEID: @@ -2528,9 +2569,13 @@ static void nfs4_close_done(struct rpc_task *task, void *data) if (calldata->arg.fmode == 0) break; default: - if (nfs4_async_handle_error(task, server, state) == -EAGAIN) + if (nfs4_async_handle_error(task, server, state) == -EAGAIN) { rpc_restart_call_prepare(task); + goto out_release; + } } + nfs_clear_open_stateid(state, NULL, calldata->arg.fmode); +out_release: nfs_release_seqid(calldata->arg.seqid); nfs_refresh_inode(calldata->inode, calldata->res.fattr); dprintk("%s: done, ret = %d!\n", __func__, task->tk_status); @@ -2705,7 +2750,7 @@ static void nfs4_close_context(struct nfs_open_context *ctx, int is_sync) #define FATTR4_WORD1_NFS40_MASK (2*FATTR4_WORD1_MOUNTED_ON_FILEID - 1UL) #define FATTR4_WORD2_NFS41_MASK (2*FATTR4_WORD2_SUPPATTR_EXCLCREAT - 1UL) -#define FATTR4_WORD2_NFS42_MASK (2*FATTR4_WORD2_CHANGE_SECURITY_LABEL - 1UL) +#define FATTR4_WORD2_NFS42_MASK (2*FATTR4_WORD2_SECURITY_LABEL - 1UL) static int _nfs4_server_capabilities(struct nfs_server *server, struct nfs_fh *fhandle) { @@ -2741,7 +2786,8 @@ static int _nfs4_server_capabilities(struct nfs_server *server, struct nfs_fh *f NFS_CAP_OWNER_GROUP|NFS_CAP_ATIME| NFS_CAP_CTIME|NFS_CAP_MTIME| NFS_CAP_SECURITY_LABEL); - if (res.attr_bitmask[0] & FATTR4_WORD0_ACL) + if (res.attr_bitmask[0] & FATTR4_WORD0_ACL && + res.acl_bitmask & ACL4_SUPPORT_ALLOW_ACL) server->caps |= NFS_CAP_ACLS; if (res.has_links != 0) server->caps |= NFS_CAP_HARDLINKS; @@ -3201,7 +3247,7 @@ static int nfs4_proc_lookup_common(struct rpc_clnt **clnt, struct inode *dir, err = -EPERM; if (client != *clnt) goto out; - client = nfs4_create_sec_client(client, dir, name); + client = nfs4_negotiate_security(client, dir, name); if (IS_ERR(client)) return PTR_ERR(client); @@ -3502,49 +3548,6 @@ static int nfs4_proc_rename_done(struct rpc_task *task, struct inode *old_dir, return 1; } -static int _nfs4_proc_rename(struct inode *old_dir, struct qstr *old_name, - struct inode *new_dir, struct qstr *new_name) -{ - struct nfs_server *server = NFS_SERVER(old_dir); - struct nfs_renameargs arg = { - .old_dir = NFS_FH(old_dir), - .new_dir = NFS_FH(new_dir), - .old_name = old_name, - .new_name = new_name, - }; - struct nfs_renameres res = { - .server = server, - }; - struct rpc_message msg = { - .rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_RENAME], - .rpc_argp = &arg, - .rpc_resp = &res, - }; - int status = -ENOMEM; - - status = nfs4_call_sync(server->client, server, &msg, &arg.seq_args, &res.seq_res, 1); - if (!status) { - update_changeattr(old_dir, &res.old_cinfo); - update_changeattr(new_dir, &res.new_cinfo); - } - return status; -} - -static int nfs4_proc_rename(struct inode *old_dir, struct qstr *old_name, - struct inode *new_dir, struct qstr *new_name) -{ - struct nfs4_exception exception = { }; - int err; - do { - err = _nfs4_proc_rename(old_dir, old_name, - new_dir, new_name); - trace_nfs4_rename(old_dir, old_name, new_dir, new_name, err); - err = nfs4_handle_exception(NFS_SERVER(old_dir), err, - &exception); - } while (exception.retry); - return err; -} - static int _nfs4_proc_link(struct inode *inode, struct inode *dir, struct qstr *name) { struct nfs_server *server = NFS_SERVER(inode); @@ -4009,8 +4012,9 @@ static bool nfs4_stateid_is_current(nfs4_stateid *stateid, { nfs4_stateid current_stateid; - if (nfs4_set_rw_stateid(¤t_stateid, ctx, l_ctx, fmode)) - return false; + /* If the current stateid represents a lost lock, then exit */ + if (nfs4_set_rw_stateid(¤t_stateid, ctx, l_ctx, fmode) == -EIO) + return true; return nfs4_stateid_match(stateid, ¤t_stateid); } @@ -4029,12 +4033,12 @@ static bool nfs4_error_stateid_expired(int err) return false; } -void __nfs4_read_done_cb(struct nfs_read_data *data) +void __nfs4_read_done_cb(struct nfs_pgio_data *data) { nfs_invalidate_atime(data->header->inode); } -static int nfs4_read_done_cb(struct rpc_task *task, struct nfs_read_data *data) +static int nfs4_read_done_cb(struct rpc_task *task, struct nfs_pgio_data *data) { struct nfs_server *server = NFS_SERVER(data->header->inode); @@ -4051,7 +4055,7 @@ static int nfs4_read_done_cb(struct rpc_task *task, struct nfs_read_data *data) } static bool nfs4_read_stateid_changed(struct rpc_task *task, - struct nfs_readargs *args) + struct nfs_pgio_args *args) { if (!nfs4_error_stateid_expired(task->tk_status) || @@ -4064,7 +4068,7 @@ static bool nfs4_read_stateid_changed(struct rpc_task *task, return true; } -static int nfs4_read_done(struct rpc_task *task, struct nfs_read_data *data) +static int nfs4_read_done(struct rpc_task *task, struct nfs_pgio_data *data) { dprintk("--> %s\n", __func__); @@ -4073,19 +4077,19 @@ static int nfs4_read_done(struct rpc_task *task, struct nfs_read_data *data) return -EAGAIN; if (nfs4_read_stateid_changed(task, &data->args)) return -EAGAIN; - return data->read_done_cb ? data->read_done_cb(task, data) : + return data->pgio_done_cb ? data->pgio_done_cb(task, data) : nfs4_read_done_cb(task, data); } -static void nfs4_proc_read_setup(struct nfs_read_data *data, struct rpc_message *msg) +static void nfs4_proc_read_setup(struct nfs_pgio_data *data, struct rpc_message *msg) { data->timestamp = jiffies; - data->read_done_cb = nfs4_read_done_cb; + data->pgio_done_cb = nfs4_read_done_cb; msg->rpc_proc = &nfs4_procedures[NFSPROC4_CLNT_READ]; nfs4_init_sequence(&data->args.seq_args, &data->res.seq_res, 0); } -static int nfs4_proc_read_rpc_prepare(struct rpc_task *task, struct nfs_read_data *data) +static int nfs4_proc_pgio_rpc_prepare(struct rpc_task *task, struct nfs_pgio_data *data) { if (nfs4_setup_sequence(NFS_SERVER(data->header->inode), &data->args.seq_args, @@ -4093,14 +4097,14 @@ static int nfs4_proc_read_rpc_prepare(struct rpc_task *task, struct nfs_read_dat task)) return 0; if (nfs4_set_rw_stateid(&data->args.stateid, data->args.context, - data->args.lock_context, FMODE_READ) == -EIO) + data->args.lock_context, data->header->rw_ops->rw_mode) == -EIO) return -EIO; if (unlikely(test_bit(NFS_CONTEXT_BAD, &data->args.context->flags))) return -EIO; return 0; } -static int nfs4_write_done_cb(struct rpc_task *task, struct nfs_write_data *data) +static int nfs4_write_done_cb(struct rpc_task *task, struct nfs_pgio_data *data) { struct inode *inode = data->header->inode; @@ -4117,7 +4121,7 @@ static int nfs4_write_done_cb(struct rpc_task *task, struct nfs_write_data *data } static bool nfs4_write_stateid_changed(struct rpc_task *task, - struct nfs_writeargs *args) + struct nfs_pgio_args *args) { if (!nfs4_error_stateid_expired(task->tk_status) || @@ -4130,18 +4134,18 @@ static bool nfs4_write_stateid_changed(struct rpc_task *task, return true; } -static int nfs4_write_done(struct rpc_task *task, struct nfs_write_data *data) +static int nfs4_write_done(struct rpc_task *task, struct nfs_pgio_data *data) { if (!nfs4_sequence_done(task, &data->res.seq_res)) return -EAGAIN; if (nfs4_write_stateid_changed(task, &data->args)) return -EAGAIN; - return data->write_done_cb ? data->write_done_cb(task, data) : + return data->pgio_done_cb ? data->pgio_done_cb(task, data) : nfs4_write_done_cb(task, data); } static -bool nfs4_write_need_cache_consistency_data(const struct nfs_write_data *data) +bool nfs4_write_need_cache_consistency_data(const struct nfs_pgio_data *data) { const struct nfs_pgio_header *hdr = data->header; @@ -4154,7 +4158,7 @@ bool nfs4_write_need_cache_consistency_data(const struct nfs_write_data *data) return nfs4_have_delegation(hdr->inode, FMODE_READ) == 0; } -static void nfs4_proc_write_setup(struct nfs_write_data *data, struct rpc_message *msg) +static void nfs4_proc_write_setup(struct nfs_pgio_data *data, struct rpc_message *msg) { struct nfs_server *server = NFS_SERVER(data->header->inode); @@ -4164,8 +4168,8 @@ static void nfs4_proc_write_setup(struct nfs_write_data *data, struct rpc_messag } else data->args.bitmask = server->cache_consistency_bitmask; - if (!data->write_done_cb) - data->write_done_cb = nfs4_write_done_cb; + if (!data->pgio_done_cb) + data->pgio_done_cb = nfs4_write_done_cb; data->res.server = server; data->timestamp = jiffies; @@ -4173,21 +4177,6 @@ static void nfs4_proc_write_setup(struct nfs_write_data *data, struct rpc_messag nfs4_init_sequence(&data->args.seq_args, &data->res.seq_res, 1); } -static int nfs4_proc_write_rpc_prepare(struct rpc_task *task, struct nfs_write_data *data) -{ - if (nfs4_setup_sequence(NFS_SERVER(data->header->inode), - &data->args.seq_args, - &data->res.seq_res, - task)) - return 0; - if (nfs4_set_rw_stateid(&data->args.stateid, data->args.context, - data->args.lock_context, FMODE_WRITE) == -EIO) - return -EIO; - if (unlikely(test_bit(NFS_CONTEXT_BAD, &data->args.context->flags))) - return -EIO; - return 0; -} - static void nfs4_proc_commit_rpc_prepare(struct rpc_task *task, struct nfs_commit_data *data) { nfs4_setup_sequence(NFS_SERVER(data->inode), @@ -4318,9 +4307,7 @@ static int nfs4_proc_renew(struct nfs_client *clp, struct rpc_cred *cred) static inline int nfs4_server_supports_acls(struct nfs_server *server) { - return (server->caps & NFS_CAP_ACLS) - && (server->acl_bitmask & ACL4_SUPPORT_ALLOW_ACL) - && (server->acl_bitmask & ACL4_SUPPORT_DENY_ACL); + return server->caps & NFS_CAP_ACLS; } /* Assuming that XATTR_SIZE_MAX is a multiple of PAGE_SIZE, and that @@ -4802,7 +4789,7 @@ nfs4_async_handle_error(struct rpc_task *task, const struct nfs_server *server, dprintk("%s ERROR %d, Reset session\n", __func__, task->tk_status); nfs4_schedule_session_recovery(clp->cl_session, task->tk_status); - goto restart_call; + goto wait_on_recovery; #endif /* CONFIG_NFS_V4_1 */ case -NFS4ERR_DELAY: nfs_inc_server_stats(server, NFSIOS_DELAY); @@ -4880,6 +4867,20 @@ nfs4_init_uniform_client_string(const struct nfs_client *clp, nodename); } +/* + * nfs4_callback_up_net() starts only "tcp" and "tcp6" callback + * services. Advertise one based on the address family of the + * clientaddr. + */ +static unsigned int +nfs4_init_callback_netid(const struct nfs_client *clp, char *buf, size_t len) +{ + if (strchr(clp->cl_ipaddr, ':') != NULL) + return scnprintf(buf, len, "tcp6"); + else + return scnprintf(buf, len, "tcp"); +} + /** * nfs4_proc_setclientid - Negotiate client ID * @clp: state data structure @@ -4921,12 +4922,10 @@ int nfs4_proc_setclientid(struct nfs_client *clp, u32 program, setclientid.sc_name, sizeof(setclientid.sc_name)); /* cb_client4 */ - rcu_read_lock(); - setclientid.sc_netid_len = scnprintf(setclientid.sc_netid, - sizeof(setclientid.sc_netid), "%s", - rpc_peeraddr2str(clp->cl_rpcclient, - RPC_DISPLAY_NETID)); - rcu_read_unlock(); + setclientid.sc_netid_len = + nfs4_init_callback_netid(clp, + setclientid.sc_netid, + sizeof(setclientid.sc_netid)); setclientid.sc_uaddr_len = scnprintf(setclientid.sc_uaddr, sizeof(setclientid.sc_uaddr), "%s.%u.%u", clp->cl_ipaddr, port >> 8, port & 255); @@ -4987,11 +4986,17 @@ static void nfs4_delegreturn_done(struct rpc_task *task, void *calldata) trace_nfs4_delegreturn_exit(&data->args, &data->res, task->tk_status); switch (task->tk_status) { - case -NFS4ERR_STALE_STATEID: - case -NFS4ERR_EXPIRED: case 0: renew_lease(data->res.server, data->timestamp); break; + case -NFS4ERR_ADMIN_REVOKED: + case -NFS4ERR_DELEG_REVOKED: + case -NFS4ERR_BAD_STATEID: + case -NFS4ERR_OLD_STATEID: + case -NFS4ERR_STALE_STATEID: + case -NFS4ERR_EXPIRED: + task->tk_status = 0; + break; default: if (nfs4_async_handle_error(task, data->res.server, NULL) == -EAGAIN) { @@ -5822,8 +5827,7 @@ struct nfs_release_lockowner_data { struct nfs4_lock_state *lsp; struct nfs_server *server; struct nfs_release_lockowner_args args; - struct nfs4_sequence_args seq_args; - struct nfs4_sequence_res seq_res; + struct nfs_release_lockowner_res res; unsigned long timestamp; }; @@ -5831,7 +5835,7 @@ static void nfs4_release_lockowner_prepare(struct rpc_task *task, void *calldata { struct nfs_release_lockowner_data *data = calldata; nfs40_setup_sequence(data->server, - &data->seq_args, &data->seq_res, task); + &data->args.seq_args, &data->res.seq_res, task); data->timestamp = jiffies; } @@ -5840,7 +5844,7 @@ static void nfs4_release_lockowner_done(struct rpc_task *task, void *calldata) struct nfs_release_lockowner_data *data = calldata; struct nfs_server *server = data->server; - nfs40_sequence_done(task, &data->seq_res); + nfs40_sequence_done(task, &data->res.seq_res); switch (task->tk_status) { case 0: @@ -5881,7 +5885,6 @@ static int nfs4_release_lockowner(struct nfs_server *server, struct nfs4_lock_st data = kmalloc(sizeof(*data), GFP_NOFS); if (!data) return -ENOMEM; - nfs4_init_sequence(&data->seq_args, &data->seq_res, 0); data->lsp = lsp; data->server = server; data->args.lock_owner.clientid = server->nfs_client->cl_clientid; @@ -5889,6 +5892,8 @@ static int nfs4_release_lockowner(struct nfs_server *server, struct nfs4_lock_st data->args.lock_owner.s_dev = server->s_dev; msg.rpc_argp = &data->args; + msg.rpc_resp = &data->res; + nfs4_init_sequence(&data->args.seq_args, &data->res.seq_res, 0); rpc_call_async(server->client, &msg, 0, &nfs4_release_lockowner_ops, data); return 0; } @@ -7400,9 +7405,9 @@ static void nfs4_layoutget_done(struct rpc_task *task, void *calldata) struct nfs_server *server = NFS_SERVER(inode); struct pnfs_layout_hdr *lo; struct nfs4_state *state = NULL; - unsigned long timeo, giveup; + unsigned long timeo, now, giveup; - dprintk("--> %s\n", __func__); + dprintk("--> %s tk_status => %d\n", __func__, -task->tk_status); if (!nfs41_sequence_done(task, &lgp->res.seq_res)) goto out; @@ -7410,12 +7415,38 @@ static void nfs4_layoutget_done(struct rpc_task *task, void *calldata) switch (task->tk_status) { case 0: goto out; + /* + * NFS4ERR_LAYOUTTRYLATER is a conflict with another client + * (or clients) writing to the same RAID stripe + */ case -NFS4ERR_LAYOUTTRYLATER: + /* + * NFS4ERR_RECALLCONFLICT is when conflict with self (must recall + * existing layout before getting a new one). + */ case -NFS4ERR_RECALLCONFLICT: timeo = rpc_get_timeout(task->tk_client); giveup = lgp->args.timestamp + timeo; - if (time_after(giveup, jiffies)) - task->tk_status = -NFS4ERR_DELAY; + now = jiffies; + if (time_after(giveup, now)) { + unsigned long delay; + + /* Delay for: + * - Not less then NFS4_POLL_RETRY_MIN. + * - One last time a jiffie before we give up + * - exponential backoff (time_now minus start_attempt) + */ + delay = max_t(unsigned long, NFS4_POLL_RETRY_MIN, + min((giveup - now - 1), + now - lgp->args.timestamp)); + + dprintk("%s: NFS4ERR_RECALLCONFLICT waiting %lu\n", + __func__, delay); + rpc_delay(task, delay); + task->tk_status = 0; + rpc_restart_call_prepare(task); + goto out; /* Do not call nfs4_async_handle_error() */ + } break; case -NFS4ERR_EXPIRED: case -NFS4ERR_BAD_STATEID: @@ -7589,7 +7620,14 @@ static void nfs4_layoutreturn_done(struct rpc_task *task, void *calldata) return; server = NFS_SERVER(lrp->args.inode); - if (nfs4_async_handle_error(task, server, NULL) == -EAGAIN) { + switch (task->tk_status) { + default: + task->tk_status = 0; + case 0: + break; + case -NFS4ERR_DELAY: + if (nfs4_async_handle_error(task, server, NULL) != -EAGAIN) + break; rpc_restart_call_prepare(task); return; } @@ -7764,10 +7802,7 @@ nfs4_layoutcommit_done(struct rpc_task *task, void *calldata) case -NFS4ERR_BADLAYOUT: /* no layout */ case -NFS4ERR_GRACE: /* loca_recalim always false */ task->tk_status = 0; - break; case 0: - nfs_post_op_update_inode_force_wcc(data->args.inode, - data->res.fattr); break; default: if (nfs4_async_handle_error(task, server, NULL) == -EAGAIN) { @@ -7782,6 +7817,8 @@ static void nfs4_layoutcommit_release(void *calldata) struct nfs4_layoutcommit_data *data = calldata; pnfs_cleanup_layoutcommit(data); + nfs_post_op_update_inode_force_wcc(data->args.inode, + data->res.fattr); put_rpccred(data->cred); kfree(data); } @@ -7904,7 +7941,7 @@ nfs41_proc_secinfo_no_name(struct nfs_server *server, struct nfs_fh *fhandle, switch (err) { case 0: case -NFS4ERR_WRONGSEC: - case -NFS4ERR_NOTSUPP: + case -ENOTSUPP: goto out; default: err = nfs4_handle_exception(server, err, &exception); @@ -7938,7 +7975,7 @@ nfs41_find_root_sec(struct nfs_server *server, struct nfs_fh *fhandle, * Fall back on "guess and check" method if * the server doesn't support SECINFO_NO_NAME */ - if (err == -NFS4ERR_WRONGSEC || err == -NFS4ERR_NOTSUPP) { + if (err == -NFS4ERR_WRONGSEC || err == -ENOTSUPP) { err = nfs4_find_root_sec(server, fhandle, info); goto out_freepage; } @@ -8366,7 +8403,6 @@ const struct nfs_rpc_ops nfs_v4_clientops = { .unlink_setup = nfs4_proc_unlink_setup, .unlink_rpc_prepare = nfs4_proc_unlink_rpc_prepare, .unlink_done = nfs4_proc_unlink_done, - .rename = nfs4_proc_rename, .rename_setup = nfs4_proc_rename_setup, .rename_rpc_prepare = nfs4_proc_rename_rpc_prepare, .rename_done = nfs4_proc_rename_done, @@ -8381,13 +8417,10 @@ const struct nfs_rpc_ops nfs_v4_clientops = { .pathconf = nfs4_proc_pathconf, .set_capabilities = nfs4_server_capabilities, .decode_dirent = nfs4_decode_dirent, + .pgio_rpc_prepare = nfs4_proc_pgio_rpc_prepare, .read_setup = nfs4_proc_read_setup, - .read_pageio_init = pnfs_pageio_init_read, - .read_rpc_prepare = nfs4_proc_read_rpc_prepare, .read_done = nfs4_read_done, .write_setup = nfs4_proc_write_setup, - .write_pageio_init = pnfs_pageio_init_write, - .write_rpc_prepare = nfs4_proc_write_rpc_prepare, .write_done = nfs4_write_done, .commit_setup = nfs4_proc_commit_setup, .commit_rpc_prepare = nfs4_proc_commit_rpc_prepare, diff --git a/fs/nfs/nfs4session.c b/fs/nfs/nfs4session.c index cf883c7ae05..e799dc3c3b1 100644 --- a/fs/nfs/nfs4session.c +++ b/fs/nfs/nfs4session.c @@ -231,14 +231,23 @@ out: return ret; } +/* + * nfs4_release_slot_table - release all slot table entries + */ +static void nfs4_release_slot_table(struct nfs4_slot_table *tbl) +{ + nfs4_shrink_slot_table(tbl, 0); +} + /** - * nfs4_release_slot_table - release resources attached to a slot table + * nfs4_shutdown_slot_table - release resources attached to a slot table * @tbl: slot table to shut down * */ -void nfs4_release_slot_table(struct nfs4_slot_table *tbl) +void nfs4_shutdown_slot_table(struct nfs4_slot_table *tbl) { - nfs4_shrink_slot_table(tbl, 0); + nfs4_release_slot_table(tbl); + rpc_destroy_wait_queue(&tbl->slot_tbl_waitq); } /** @@ -422,7 +431,7 @@ void nfs41_update_target_slotid(struct nfs4_slot_table *tbl, spin_unlock(&tbl->slot_tbl_lock); } -static void nfs4_destroy_session_slot_tables(struct nfs4_session *session) +static void nfs4_release_session_slot_tables(struct nfs4_session *session) { nfs4_release_slot_table(&session->fc_slot_table); nfs4_release_slot_table(&session->bc_slot_table); @@ -450,7 +459,7 @@ int nfs4_setup_session_slot_tables(struct nfs4_session *ses) if (status && tbl->slots == NULL) /* Fore and back channel share a connection so get * both slot tables or neither */ - nfs4_destroy_session_slot_tables(ses); + nfs4_release_session_slot_tables(ses); return status; } @@ -470,6 +479,12 @@ struct nfs4_session *nfs4_alloc_session(struct nfs_client *clp) return session; } +static void nfs4_destroy_session_slot_tables(struct nfs4_session *session) +{ + nfs4_shutdown_slot_table(&session->fc_slot_table); + nfs4_shutdown_slot_table(&session->bc_slot_table); +} + void nfs4_destroy_session(struct nfs4_session *session) { struct rpc_xprt *xprt; diff --git a/fs/nfs/nfs4session.h b/fs/nfs/nfs4session.h index 23230610065..b34ada9bc6a 100644 --- a/fs/nfs/nfs4session.h +++ b/fs/nfs/nfs4session.h @@ -74,7 +74,7 @@ enum nfs4_session_state { extern int nfs4_setup_slot_table(struct nfs4_slot_table *tbl, unsigned int max_reqs, const char *queue); -extern void nfs4_release_slot_table(struct nfs4_slot_table *tbl); +extern void nfs4_shutdown_slot_table(struct nfs4_slot_table *tbl); extern struct nfs4_slot *nfs4_alloc_slot(struct nfs4_slot_table *tbl); extern void nfs4_free_slot(struct nfs4_slot_table *tbl, struct nfs4_slot *slot); extern void nfs4_slot_tbl_drain_complete(struct nfs4_slot_table *tbl); diff --git a/fs/nfs/nfs4state.c b/fs/nfs/nfs4state.c index 059c01b67a7..848f6853c59 100644 --- a/fs/nfs/nfs4state.c +++ b/fs/nfs/nfs4state.c @@ -974,9 +974,6 @@ static int nfs4_copy_lock_stateid(nfs4_stateid *dst, else if (lsp != NULL && test_bit(NFS_LOCK_INITIALIZED, &lsp->ls_flags) != 0) { nfs4_stateid_copy(dst, &lsp->ls_stateid); ret = 0; - smp_rmb(); - if (!list_empty(&lsp->ls_seqid.list)) - ret = -EWOULDBLOCK; } spin_unlock(&state->state_lock); nfs4_put_lock_state(lsp); @@ -984,10 +981,9 @@ out: return ret; } -static int nfs4_copy_open_stateid(nfs4_stateid *dst, struct nfs4_state *state) +static void nfs4_copy_open_stateid(nfs4_stateid *dst, struct nfs4_state *state) { const nfs4_stateid *src; - int ret; int seq; do { @@ -996,12 +992,7 @@ static int nfs4_copy_open_stateid(nfs4_stateid *dst, struct nfs4_state *state) if (test_bit(NFS_OPEN_STATE, &state->flags)) src = &state->open_stateid; nfs4_stateid_copy(dst, src); - ret = 0; - smp_rmb(); - if (!list_empty(&state->owner->so_seqid.list)) - ret = -EWOULDBLOCK; } while (read_seqretry(&state->seqlock, seq)); - return ret; } /* @@ -1015,15 +1006,19 @@ int nfs4_select_rw_stateid(nfs4_stateid *dst, struct nfs4_state *state, if (ret == -EIO) /* A lost lock - don't even consider delegations */ goto out; - if (nfs4_copy_delegation_stateid(dst, state->inode, fmode)) + /* returns true if delegation stateid found and copied */ + if (nfs4_copy_delegation_stateid(dst, state->inode, fmode)) { + ret = 0; goto out; + } if (ret != -ENOENT) /* nfs4_copy_delegation_stateid() didn't over-write * dst, so it still has the lock stateid which we now * choose to use. */ goto out; - ret = nfs4_copy_open_stateid(dst, state); + nfs4_copy_open_stateid(dst, state); + ret = 0; out: if (nfs_server_capable(state->inode, NFS_CAP_STATEID_NFSV41)) dst->seqid = 0; @@ -1071,7 +1066,7 @@ void nfs_free_seqid(struct nfs_seqid *seqid) /* * Increment the seqid if the OPEN/OPEN_DOWNGRADE/CLOSE succeeded, or * failed with a seqid incrementing error - - * see comments nfs_fs.h:seqid_mutating_error() + * see comments nfs4.h:seqid_mutating_error() */ static void nfs_increment_seqid(int status, struct nfs_seqid *seqid) { @@ -1116,7 +1111,7 @@ void nfs_increment_open_seqid(int status, struct nfs_seqid *seqid) /* * Increment the seqid if the LOCK/LOCKU succeeded, or * failed with a seqid incrementing error - - * see comments nfs_fs.h:seqid_mutating_error() + * see comments nfs4.h:seqid_mutating_error() */ void nfs_increment_lock_seqid(int status, struct nfs_seqid *seqid) { @@ -1145,9 +1140,9 @@ static int nfs4_run_state_manager(void *); static void nfs4_clear_state_manager_bit(struct nfs_client *clp) { - smp_mb__before_clear_bit(); + smp_mb__before_atomic(); clear_bit(NFS4CLNT_MANAGER_RUNNING, &clp->cl_state); - smp_mb__after_clear_bit(); + smp_mb__after_atomic(); wake_up_bit(&clp->cl_state, NFS4CLNT_MANAGER_RUNNING); rpc_wake_up(&clp->cl_rpcwaitq); } @@ -1321,7 +1316,7 @@ static int nfs4_state_mark_reclaim_reboot(struct nfs_client *clp, struct nfs4_st return 1; } -static int nfs4_state_mark_reclaim_nograce(struct nfs_client *clp, struct nfs4_state *state) +int nfs4_state_mark_reclaim_nograce(struct nfs_client *clp, struct nfs4_state *state) { set_bit(NFS_STATE_RECLAIM_NOGRACE, &state->flags); clear_bit(NFS_STATE_RECLAIM_REBOOT, &state->flags); @@ -1461,7 +1456,7 @@ static int nfs4_reclaim_open_state(struct nfs4_state_owner *sp, const struct nfs * server that doesn't support a grace period. */ spin_lock(&sp->so_lock); - write_seqcount_begin(&sp->so_reclaim_seqcount); + raw_write_seqcount_begin(&sp->so_reclaim_seqcount); restart: list_for_each_entry(state, &sp->so_states, open_states) { if (!test_and_clear_bit(ops->state_flag_bit, &state->flags)) @@ -1524,13 +1519,13 @@ restart: spin_lock(&sp->so_lock); goto restart; } - write_seqcount_end(&sp->so_reclaim_seqcount); + raw_write_seqcount_end(&sp->so_reclaim_seqcount); spin_unlock(&sp->so_lock); return 0; out_err: nfs4_put_open_state(state); spin_lock(&sp->so_lock); - write_seqcount_end(&sp->so_reclaim_seqcount); + raw_write_seqcount_end(&sp->so_reclaim_seqcount); spin_unlock(&sp->so_lock); return status; } @@ -2080,8 +2075,10 @@ again: switch (status) { case 0: break; - case -NFS4ERR_DELAY: case -ETIMEDOUT: + if (clnt->cl_softrtry) + break; + case -NFS4ERR_DELAY: case -EAGAIN: ssleep(1); case -NFS4ERR_STALE_CLIENTID: diff --git a/fs/nfs/nfs4super.c b/fs/nfs/nfs4super.c index 65ab0a0ca1c..6f340f02f2b 100644 --- a/fs/nfs/nfs4super.c +++ b/fs/nfs/nfs4super.c @@ -77,17 +77,9 @@ static int nfs4_write_inode(struct inode *inode, struct writeback_control *wbc) { int ret = nfs_write_inode(inode, wbc); - if (ret >= 0 && test_bit(NFS_INO_LAYOUTCOMMIT, &NFS_I(inode)->flags)) { - int status; - bool sync = true; - - if (wbc->sync_mode == WB_SYNC_NONE) - sync = false; - - status = pnfs_layoutcommit_inode(inode, sync); - if (status < 0) - return status; - } + if (ret == 0) + ret = pnfs_layoutcommit_inode(inode, + wbc->sync_mode == WB_SYNC_ALL); return ret; } @@ -98,7 +90,7 @@ static int nfs4_write_inode(struct inode *inode, struct writeback_control *wbc) */ static void nfs4_evict_inode(struct inode *inode) { - truncate_inode_pages(&inode->i_data, 0); + truncate_inode_pages_final(&inode->i_data); clear_inode(inode); pnfs_return_layout(inode); pnfs_destroy_layout(NFS_I(inode)); diff --git a/fs/nfs/nfs4sysctl.c b/fs/nfs/nfs4sysctl.c index 2628d921b7e..b6ebe7e445f 100644 --- a/fs/nfs/nfs4sysctl.c +++ b/fs/nfs/nfs4sysctl.c @@ -16,7 +16,7 @@ static const int nfs_set_port_min = 0; static const int nfs_set_port_max = 65535; static struct ctl_table_header *nfs4_callback_sysctl_table; -static ctl_table nfs4_cb_sysctls[] = { +static struct ctl_table nfs4_cb_sysctls[] = { { .procname = "nfs_callback_tcpport", .data = &nfs_callback_set_tcpport, @@ -36,7 +36,7 @@ static ctl_table nfs4_cb_sysctls[] = { { } }; -static ctl_table nfs4_cb_sysctl_dir[] = { +static struct ctl_table nfs4_cb_sysctl_dir[] = { { .procname = "nfs", .mode = 0555, @@ -45,7 +45,7 @@ static ctl_table nfs4_cb_sysctl_dir[] = { { } }; -static ctl_table nfs4_cb_sysctl_root[] = { +static struct ctl_table nfs4_cb_sysctl_root[] = { { .procname = "fs", .mode = 0555, diff --git a/fs/nfs/nfs4trace.h b/fs/nfs/nfs4trace.h index 849cf146db3..0a744f3a86f 100644 --- a/fs/nfs/nfs4trace.h +++ b/fs/nfs/nfs4trace.h @@ -932,7 +932,7 @@ DEFINE_NFS4_IDMAP_EVENT(nfs4_map_gid_to_group); DECLARE_EVENT_CLASS(nfs4_read_event, TP_PROTO( - const struct nfs_read_data *data, + const struct nfs_pgio_data *data, int error ), @@ -972,7 +972,7 @@ DECLARE_EVENT_CLASS(nfs4_read_event, #define DEFINE_NFS4_READ_EVENT(name) \ DEFINE_EVENT(nfs4_read_event, name, \ TP_PROTO( \ - const struct nfs_read_data *data, \ + const struct nfs_pgio_data *data, \ int error \ ), \ TP_ARGS(data, error)) @@ -983,7 +983,7 @@ DEFINE_NFS4_READ_EVENT(nfs4_pnfs_read); DECLARE_EVENT_CLASS(nfs4_write_event, TP_PROTO( - const struct nfs_write_data *data, + const struct nfs_pgio_data *data, int error ), @@ -1024,7 +1024,7 @@ DECLARE_EVENT_CLASS(nfs4_write_event, #define DEFINE_NFS4_WRITE_EVENT(name) \ DEFINE_EVENT(nfs4_write_event, name, \ TP_PROTO( \ - const struct nfs_write_data *data, \ + const struct nfs_pgio_data *data, \ int error \ ), \ TP_ARGS(data, error)) diff --git a/fs/nfs/nfs4xdr.c b/fs/nfs/nfs4xdr.c index 5be2868c02f..939ae606cfa 100644 --- a/fs/nfs/nfs4xdr.c +++ b/fs/nfs/nfs4xdr.c @@ -203,8 +203,7 @@ static int nfs4_stat_to_errno(int); 2 + encode_verifier_maxsz + 5 + \ nfs4_label_maxsz) #define decode_readdir_maxsz (op_decode_hdr_maxsz + \ - decode_verifier_maxsz + \ - nfs4_label_maxsz + nfs4_fattr_maxsz) + decode_verifier_maxsz) #define encode_readlink_maxsz (op_encode_hdr_maxsz) #define decode_readlink_maxsz (op_decode_hdr_maxsz + 1) #define encode_write_maxsz (op_encode_hdr_maxsz + \ @@ -1557,7 +1556,8 @@ static void encode_putrootfh(struct xdr_stream *xdr, struct compound_hdr *hdr) encode_op_hdr(xdr, OP_PUTROOTFH, decode_putrootfh_maxsz, hdr); } -static void encode_read(struct xdr_stream *xdr, const struct nfs_readargs *args, struct compound_hdr *hdr) +static void encode_read(struct xdr_stream *xdr, const struct nfs_pgio_args *args, + struct compound_hdr *hdr) { __be32 *p; @@ -1702,7 +1702,8 @@ static void encode_setclientid_confirm(struct xdr_stream *xdr, const struct nfs4 encode_nfs4_verifier(xdr, &arg->confirm); } -static void encode_write(struct xdr_stream *xdr, const struct nfs_writeargs *args, struct compound_hdr *hdr) +static void encode_write(struct xdr_stream *xdr, const struct nfs_pgio_args *args, + struct compound_hdr *hdr) { __be32 *p; @@ -2452,7 +2453,7 @@ static void nfs4_xdr_enc_readdir(struct rpc_rqst *req, struct xdr_stream *xdr, * Encode a READ request */ static void nfs4_xdr_enc_read(struct rpc_rqst *req, struct xdr_stream *xdr, - struct nfs_readargs *args) + struct nfs_pgio_args *args) { struct compound_hdr hdr = { .minorversion = nfs4_xdr_minorversion(&args->seq_args), @@ -2514,7 +2515,7 @@ static void nfs4_xdr_enc_getacl(struct rpc_rqst *req, struct xdr_stream *xdr, * Encode a WRITE request */ static void nfs4_xdr_enc_write(struct rpc_rqst *req, struct xdr_stream *xdr, - struct nfs_writeargs *args) + struct nfs_pgio_args *args) { struct compound_hdr hdr = { .minorversion = nfs4_xdr_minorversion(&args->seq_args), @@ -3097,7 +3098,8 @@ out_overflow: return -EIO; } -static int decode_op_hdr(struct xdr_stream *xdr, enum nfs_opnum4 expected) +static bool __decode_op_hdr(struct xdr_stream *xdr, enum nfs_opnum4 expected, + int *nfs_retval) { __be32 *p; uint32_t opnum; @@ -3107,19 +3109,32 @@ static int decode_op_hdr(struct xdr_stream *xdr, enum nfs_opnum4 expected) if (unlikely(!p)) goto out_overflow; opnum = be32_to_cpup(p++); - if (opnum != expected) { - dprintk("nfs: Server returned operation" - " %d but we issued a request for %d\n", - opnum, expected); - return -EIO; - } + if (unlikely(opnum != expected)) + goto out_bad_operation; nfserr = be32_to_cpup(p); - if (nfserr != NFS_OK) - return nfs4_stat_to_errno(nfserr); - return 0; + if (nfserr == NFS_OK) + *nfs_retval = 0; + else + *nfs_retval = nfs4_stat_to_errno(nfserr); + return true; +out_bad_operation: + dprintk("nfs: Server returned operation" + " %d but we issued a request for %d\n", + opnum, expected); + *nfs_retval = -EREMOTEIO; + return false; out_overflow: print_overflow_msg(__func__, xdr); - return -EIO; + *nfs_retval = -EIO; + return false; +} + +static int decode_op_hdr(struct xdr_stream *xdr, enum nfs_opnum4 expected) +{ + int retval; + + __decode_op_hdr(xdr, expected, &retval); + return retval; } /* Dummy routine */ @@ -3435,7 +3450,7 @@ static int decode_attr_aclsupport(struct xdr_stream *xdr, uint32_t *bitmap, uint { __be32 *p; - *res = ACL4_SUPPORT_ALLOW_ACL|ACL4_SUPPORT_DENY_ACL; + *res = 0; if (unlikely(bitmap[0] & (FATTR4_WORD0_ACLSUPPORT - 1U))) return -EIO; if (likely(bitmap[0] & FATTR4_WORD0_ACLSUPPORT)) { @@ -5001,11 +5016,12 @@ static int decode_open(struct xdr_stream *xdr, struct nfs_openres *res) uint32_t savewords, bmlen, i; int status; - status = decode_op_hdr(xdr, OP_OPEN); - if (status != -EIO) - nfs_increment_open_seqid(status, res->seqid); - if (!status) - status = decode_stateid(xdr, &res->stateid); + if (!__decode_op_hdr(xdr, OP_OPEN, &status)) + return status; + nfs_increment_open_seqid(status, res->seqid); + if (status) + return status; + status = decode_stateid(xdr, &res->stateid); if (unlikely(status)) return status; @@ -5071,7 +5087,8 @@ static int decode_putrootfh(struct xdr_stream *xdr) return decode_op_hdr(xdr, OP_PUTROOTFH); } -static int decode_read(struct xdr_stream *xdr, struct rpc_rqst *req, struct nfs_readres *res) +static int decode_read(struct xdr_stream *xdr, struct rpc_rqst *req, + struct nfs_pgio_res *res) { __be32 *p; uint32_t count, eof, recvd; @@ -5325,7 +5342,7 @@ static int decode_setclientid_confirm(struct xdr_stream *xdr) return decode_op_hdr(xdr, OP_SETCLIENTID_CONFIRM); } -static int decode_write(struct xdr_stream *xdr, struct nfs_writeres *res) +static int decode_write(struct xdr_stream *xdr, struct nfs_pgio_res *res) { __be32 *p; int status; @@ -6622,7 +6639,7 @@ out: * Decode Read response */ static int nfs4_xdr_dec_read(struct rpc_rqst *rqstp, struct xdr_stream *xdr, - struct nfs_readres *res) + struct nfs_pgio_res *res) { struct compound_hdr hdr; int status; @@ -6647,7 +6664,7 @@ out: * Decode WRITE response */ static int nfs4_xdr_dec_write(struct rpc_rqst *rqstp, struct xdr_stream *xdr, - struct nfs_writeres *res) + struct nfs_pgio_res *res) { struct compound_hdr hdr; int status; diff --git a/fs/nfs/nfstrace.h b/fs/nfs/nfstrace.h index 89fe741e58b..59f838cdc00 100644 --- a/fs/nfs/nfstrace.h +++ b/fs/nfs/nfstrace.h @@ -36,6 +36,7 @@ __print_flags(v, "|", \ { 1 << NFS_INO_ADVISE_RDPLUS, "ADVISE_RDPLUS" }, \ { 1 << NFS_INO_STALE, "STALE" }, \ + { 1 << NFS_INO_INVALIDATING, "INVALIDATING" }, \ { 1 << NFS_INO_FLUSHING, "FLUSHING" }, \ { 1 << NFS_INO_FSCACHE, "FSCACHE" }, \ { 1 << NFS_INO_COMMIT, "COMMIT" }, \ diff --git a/fs/nfs/objlayout/objio_osd.c b/fs/nfs/objlayout/objio_osd.c index 5457745dd4f..611320753db 100644 --- a/fs/nfs/objlayout/objio_osd.c +++ b/fs/nfs/objlayout/objio_osd.c @@ -439,7 +439,7 @@ static void _read_done(struct ore_io_state *ios, void *private) objlayout_read_done(&objios->oir, status, objios->sync); } -int objio_read_pagelist(struct nfs_read_data *rdata) +int objio_read_pagelist(struct nfs_pgio_data *rdata) { struct nfs_pgio_header *hdr = rdata->header; struct objio_state *objios; @@ -487,7 +487,7 @@ static void _write_done(struct ore_io_state *ios, void *private) static struct page *__r4w_get_page(void *priv, u64 offset, bool *uptodate) { struct objio_state *objios = priv; - struct nfs_write_data *wdata = objios->oir.rpcdata; + struct nfs_pgio_data *wdata = objios->oir.rpcdata; struct address_space *mapping = wdata->header->inode->i_mapping; pgoff_t index = offset / PAGE_SIZE; struct page *page; @@ -531,7 +531,7 @@ static const struct _ore_r4w_op _r4w_op = { .put_page = &__r4w_put_page, }; -int objio_write_pagelist(struct nfs_write_data *wdata, int how) +int objio_write_pagelist(struct nfs_pgio_data *wdata, int how) { struct nfs_pgio_header *hdr = wdata->header; struct objio_state *objios; @@ -564,14 +564,22 @@ int objio_write_pagelist(struct nfs_write_data *wdata, int how) return 0; } -static bool objio_pg_test(struct nfs_pageio_descriptor *pgio, +/* + * Return 0 if @req cannot be coalesced into @pgio, otherwise return the number + * of bytes (maximum @req->wb_bytes) that can be coalesced. + */ +static size_t objio_pg_test(struct nfs_pageio_descriptor *pgio, struct nfs_page *prev, struct nfs_page *req) { - if (!pnfs_generic_pg_test(pgio, prev, req)) - return false; + unsigned int size; + + size = pnfs_generic_pg_test(pgio, prev, req); + + if (!size || pgio->pg_count + req->wb_bytes > + (unsigned long)pgio->pg_layout_private) + return 0; - return pgio->pg_count + req->wb_bytes <= - (unsigned long)pgio->pg_layout_private; + return min(size, req->wb_bytes); } static void objio_init_read(struct nfs_pageio_descriptor *pgio, struct nfs_page *req) diff --git a/fs/nfs/objlayout/objlayout.c b/fs/nfs/objlayout/objlayout.c index e4f9cbfec67..765d3f54e98 100644 --- a/fs/nfs/objlayout/objlayout.c +++ b/fs/nfs/objlayout/objlayout.c @@ -53,10 +53,10 @@ objlayout_alloc_layout_hdr(struct inode *inode, gfp_t gfp_flags) struct objlayout *objlay; objlay = kzalloc(sizeof(struct objlayout), gfp_flags); - if (objlay) { - spin_lock_init(&objlay->lock); - INIT_LIST_HEAD(&objlay->err_list); - } + if (!objlay) + return NULL; + spin_lock_init(&objlay->lock); + INIT_LIST_HEAD(&objlay->err_list); dprintk("%s: Return %p\n", __func__, objlay); return &objlay->pnfs_layout; } @@ -229,11 +229,11 @@ objlayout_io_set_result(struct objlayout_io_res *oir, unsigned index, static void _rpc_read_complete(struct work_struct *work) { struct rpc_task *task; - struct nfs_read_data *rdata; + struct nfs_pgio_data *rdata; dprintk("%s enter\n", __func__); task = container_of(work, struct rpc_task, u.tk_work); - rdata = container_of(task, struct nfs_read_data, task); + rdata = container_of(task, struct nfs_pgio_data, task); pnfs_ld_read_done(rdata); } @@ -241,7 +241,7 @@ static void _rpc_read_complete(struct work_struct *work) void objlayout_read_done(struct objlayout_io_res *oir, ssize_t status, bool sync) { - struct nfs_read_data *rdata = oir->rpcdata; + struct nfs_pgio_data *rdata = oir->rpcdata; oir->status = rdata->task.tk_status = status; if (status >= 0) @@ -266,7 +266,7 @@ objlayout_read_done(struct objlayout_io_res *oir, ssize_t status, bool sync) * Perform sync or async reads. */ enum pnfs_try_status -objlayout_read_pagelist(struct nfs_read_data *rdata) +objlayout_read_pagelist(struct nfs_pgio_data *rdata) { struct nfs_pgio_header *hdr = rdata->header; struct inode *inode = hdr->inode; @@ -312,11 +312,11 @@ objlayout_read_pagelist(struct nfs_read_data *rdata) static void _rpc_write_complete(struct work_struct *work) { struct rpc_task *task; - struct nfs_write_data *wdata; + struct nfs_pgio_data *wdata; dprintk("%s enter\n", __func__); task = container_of(work, struct rpc_task, u.tk_work); - wdata = container_of(task, struct nfs_write_data, task); + wdata = container_of(task, struct nfs_pgio_data, task); pnfs_ld_write_done(wdata); } @@ -324,7 +324,7 @@ static void _rpc_write_complete(struct work_struct *work) void objlayout_write_done(struct objlayout_io_res *oir, ssize_t status, bool sync) { - struct nfs_write_data *wdata = oir->rpcdata; + struct nfs_pgio_data *wdata = oir->rpcdata; oir->status = wdata->task.tk_status = status; if (status >= 0) { @@ -351,7 +351,7 @@ objlayout_write_done(struct objlayout_io_res *oir, ssize_t status, bool sync) * Perform sync or async writes. */ enum pnfs_try_status -objlayout_write_pagelist(struct nfs_write_data *wdata, +objlayout_write_pagelist(struct nfs_pgio_data *wdata, int how) { struct nfs_pgio_header *hdr = wdata->header; diff --git a/fs/nfs/objlayout/objlayout.h b/fs/nfs/objlayout/objlayout.h index 87aa1dec612..01e041029a6 100644 --- a/fs/nfs/objlayout/objlayout.h +++ b/fs/nfs/objlayout/objlayout.h @@ -119,8 +119,8 @@ extern void objio_free_lseg(struct pnfs_layout_segment *lseg); */ extern void objio_free_result(struct objlayout_io_res *oir); -extern int objio_read_pagelist(struct nfs_read_data *rdata); -extern int objio_write_pagelist(struct nfs_write_data *wdata, int how); +extern int objio_read_pagelist(struct nfs_pgio_data *rdata); +extern int objio_write_pagelist(struct nfs_pgio_data *wdata, int how); /* * callback API @@ -168,10 +168,10 @@ extern struct pnfs_layout_segment *objlayout_alloc_lseg( extern void objlayout_free_lseg(struct pnfs_layout_segment *); extern enum pnfs_try_status objlayout_read_pagelist( - struct nfs_read_data *); + struct nfs_pgio_data *); extern enum pnfs_try_status objlayout_write_pagelist( - struct nfs_write_data *, + struct nfs_pgio_data *, int how); extern void objlayout_encode_layoutcommit( diff --git a/fs/nfs/pagelist.c b/fs/nfs/pagelist.c index 2ffebf2081c..17fab89f635 100644 --- a/fs/nfs/pagelist.c +++ b/fs/nfs/pagelist.c @@ -24,9 +24,12 @@ #include "internal.h" #include "pnfs.h" +#define NFSDBG_FACILITY NFSDBG_PAGECACHE + static struct kmem_cache *nfs_page_cachep; +static const struct rpc_call_ops nfs_pgio_common_ops; -bool nfs_pgarray_set(struct nfs_page_array *p, unsigned int pagecount) +static bool nfs_pgarray_set(struct nfs_page_array *p, unsigned int pagecount) { p->npages = pagecount; if (pagecount <= ARRAY_SIZE(p->page_array)) @@ -95,7 +98,7 @@ nfs_iocounter_dec(struct nfs_io_counter *c) { if (atomic_dec_and_test(&c->io_count)) { clear_bit(NFS_IO_INPROGRESS, &c->flags); - smp_mb__after_clear_bit(); + smp_mb__after_atomic(); wake_up_bit(&c->flags, NFS_IO_INPROGRESS); } } @@ -133,11 +136,168 @@ nfs_iocounter_wait(struct nfs_io_counter *c) return __nfs_iocounter_wait(c); } +static int nfs_wait_bit_uninterruptible(void *word) +{ + io_schedule(); + return 0; +} + +/* + * nfs_page_group_lock - lock the head of the page group + * @req - request in group that is to be locked + * + * this lock must be held if modifying the page group list + */ +void +nfs_page_group_lock(struct nfs_page *req) +{ + struct nfs_page *head = req->wb_head; + + WARN_ON_ONCE(head != head->wb_head); + + wait_on_bit_lock(&head->wb_flags, PG_HEADLOCK, + nfs_wait_bit_uninterruptible, + TASK_UNINTERRUPTIBLE); +} + +/* + * nfs_page_group_unlock - unlock the head of the page group + * @req - request in group that is to be unlocked + */ +void +nfs_page_group_unlock(struct nfs_page *req) +{ + struct nfs_page *head = req->wb_head; + + WARN_ON_ONCE(head != head->wb_head); + + smp_mb__before_atomic(); + clear_bit(PG_HEADLOCK, &head->wb_flags); + smp_mb__after_atomic(); + wake_up_bit(&head->wb_flags, PG_HEADLOCK); +} + +/* + * nfs_page_group_sync_on_bit_locked + * + * must be called with page group lock held + */ +static bool +nfs_page_group_sync_on_bit_locked(struct nfs_page *req, unsigned int bit) +{ + struct nfs_page *head = req->wb_head; + struct nfs_page *tmp; + + WARN_ON_ONCE(!test_bit(PG_HEADLOCK, &head->wb_flags)); + WARN_ON_ONCE(test_and_set_bit(bit, &req->wb_flags)); + + tmp = req->wb_this_page; + while (tmp != req) { + if (!test_bit(bit, &tmp->wb_flags)) + return false; + tmp = tmp->wb_this_page; + } + + /* true! reset all bits */ + tmp = req; + do { + clear_bit(bit, &tmp->wb_flags); + tmp = tmp->wb_this_page; + } while (tmp != req); + + return true; +} + +/* + * nfs_page_group_sync_on_bit - set bit on current request, but only + * return true if the bit is set for all requests in page group + * @req - request in page group + * @bit - PG_* bit that is used to sync page group + */ +bool nfs_page_group_sync_on_bit(struct nfs_page *req, unsigned int bit) +{ + bool ret; + + nfs_page_group_lock(req); + ret = nfs_page_group_sync_on_bit_locked(req, bit); + nfs_page_group_unlock(req); + + return ret; +} + +/* + * nfs_page_group_init - Initialize the page group linkage for @req + * @req - a new nfs request + * @prev - the previous request in page group, or NULL if @req is the first + * or only request in the group (the head). + */ +static inline void +nfs_page_group_init(struct nfs_page *req, struct nfs_page *prev) +{ + WARN_ON_ONCE(prev == req); + + if (!prev) { + /* a head request */ + req->wb_head = req; + req->wb_this_page = req; + } else { + /* a subrequest */ + WARN_ON_ONCE(prev->wb_this_page != prev->wb_head); + WARN_ON_ONCE(!test_bit(PG_HEADLOCK, &prev->wb_head->wb_flags)); + req->wb_head = prev->wb_head; + req->wb_this_page = prev->wb_this_page; + prev->wb_this_page = req; + + /* All subrequests take a ref on the head request until + * nfs_page_group_destroy is called */ + kref_get(&req->wb_head->wb_kref); + + /* grab extra ref if head request has extra ref from + * the write/commit path to handle handoff between write + * and commit lists */ + if (test_bit(PG_INODE_REF, &prev->wb_head->wb_flags)) { + set_bit(PG_INODE_REF, &req->wb_flags); + kref_get(&req->wb_kref); + } + } +} + +/* + * nfs_page_group_destroy - sync the destruction of page groups + * @req - request that no longer needs the page group + * + * releases the page group reference from each member once all + * members have called this function. + */ +static void +nfs_page_group_destroy(struct kref *kref) +{ + struct nfs_page *req = container_of(kref, struct nfs_page, wb_kref); + struct nfs_page *tmp, *next; + + /* subrequests must release the ref on the head request */ + if (req->wb_head != req) + nfs_release_request(req->wb_head); + + if (!nfs_page_group_sync_on_bit(req, PG_TEARDOWN)) + return; + + tmp = req; + do { + next = tmp->wb_this_page; + /* unlink and free */ + tmp->wb_this_page = tmp; + tmp->wb_head = tmp; + nfs_free_request(tmp); + tmp = next; + } while (tmp != req); +} + /** * nfs_create_request - Create an NFS read/write request. * @ctx: open context to use - * @inode: inode to which the request is attached * @page: page to write + * @last: last nfs request created for this page group or NULL if head * @offset: starting offset within the page for the write * @count: number of bytes to read/write * @@ -146,9 +306,9 @@ nfs_iocounter_wait(struct nfs_io_counter *c) * User should ensure it is safe to sleep in this function. */ struct nfs_page * -nfs_create_request(struct nfs_open_context *ctx, struct inode *inode, - struct page *page, - unsigned int offset, unsigned int count) +nfs_create_request(struct nfs_open_context *ctx, struct page *page, + struct nfs_page *last, unsigned int offset, + unsigned int count) { struct nfs_page *req; struct nfs_lock_context *l_ctx; @@ -180,6 +340,7 @@ nfs_create_request(struct nfs_open_context *ctx, struct inode *inode, req->wb_bytes = count; req->wb_context = get_nfs_open_context(ctx); kref_init(&req->wb_kref); + nfs_page_group_init(req, last); return req; } @@ -193,9 +354,9 @@ void nfs_unlock_request(struct nfs_page *req) printk(KERN_ERR "NFS: Invalid unlock attempted\n"); BUG(); } - smp_mb__before_clear_bit(); + smp_mb__before_atomic(); clear_bit(PG_BUSY, &req->wb_flags); - smp_mb__after_clear_bit(); + smp_mb__after_atomic(); wake_up_bit(&req->wb_flags, PG_BUSY); } @@ -237,16 +398,22 @@ static void nfs_clear_request(struct nfs_page *req) } } - /** * nfs_release_request - Release the count on an NFS read/write request * @req: request to release * * Note: Should never be called with the spinlock held! */ -static void nfs_free_request(struct kref *kref) +void nfs_free_request(struct nfs_page *req) { - struct nfs_page *req = container_of(kref, struct nfs_page, wb_kref); + WARN_ON_ONCE(req->wb_this_page != req); + + /* extra debug: make sure no sync bits are still set */ + WARN_ON_ONCE(test_bit(PG_TEARDOWN, &req->wb_flags)); + WARN_ON_ONCE(test_bit(PG_UNLOCKPAGE, &req->wb_flags)); + WARN_ON_ONCE(test_bit(PG_UPTODATE, &req->wb_flags)); + WARN_ON_ONCE(test_bit(PG_WB_END, &req->wb_flags)); + WARN_ON_ONCE(test_bit(PG_REMOVE, &req->wb_flags)); /* Release struct file and open context */ nfs_clear_request(req); @@ -255,13 +422,7 @@ static void nfs_free_request(struct kref *kref) void nfs_release_request(struct nfs_page *req) { - kref_put(&req->wb_kref, nfs_free_request); -} - -static int nfs_wait_bit_uninterruptible(void *word) -{ - io_schedule(); - return 0; + kref_put(&req->wb_kref, nfs_page_group_destroy); } /** @@ -279,22 +440,249 @@ nfs_wait_on_request(struct nfs_page *req) TASK_UNINTERRUPTIBLE); } -bool nfs_generic_pg_test(struct nfs_pageio_descriptor *desc, struct nfs_page *prev, struct nfs_page *req) +/* + * nfs_generic_pg_test - determine if requests can be coalesced + * @desc: pointer to descriptor + * @prev: previous request in desc, or NULL + * @req: this request + * + * Returns zero if @req can be coalesced into @desc, otherwise it returns + * the size of the request. + */ +size_t nfs_generic_pg_test(struct nfs_pageio_descriptor *desc, + struct nfs_page *prev, struct nfs_page *req) { - /* - * FIXME: ideally we should be able to coalesce all requests - * that are not block boundary aligned, but currently this - * is problematic for the case of bsize < PAGE_CACHE_SIZE, - * since nfs_flush_multi and nfs_pagein_multi assume you - * can have only one struct nfs_page. - */ - if (desc->pg_bsize < PAGE_SIZE) + if (desc->pg_count > desc->pg_bsize) { + /* should never happen */ + WARN_ON_ONCE(1); return 0; + } - return desc->pg_count + req->wb_bytes <= desc->pg_bsize; + return min(desc->pg_bsize - desc->pg_count, (size_t)req->wb_bytes); } EXPORT_SYMBOL_GPL(nfs_generic_pg_test); +static inline struct nfs_rw_header *NFS_RW_HEADER(struct nfs_pgio_header *hdr) +{ + return container_of(hdr, struct nfs_rw_header, header); +} + +/** + * nfs_rw_header_alloc - Allocate a header for a read or write + * @ops: Read or write function vector + */ +struct nfs_rw_header *nfs_rw_header_alloc(const struct nfs_rw_ops *ops) +{ + struct nfs_rw_header *header = ops->rw_alloc_header(); + + if (header) { + struct nfs_pgio_header *hdr = &header->header; + + INIT_LIST_HEAD(&hdr->pages); + spin_lock_init(&hdr->lock); + atomic_set(&hdr->refcnt, 0); + hdr->rw_ops = ops; + } + return header; +} +EXPORT_SYMBOL_GPL(nfs_rw_header_alloc); + +/* + * nfs_rw_header_free - Free a read or write header + * @hdr: The header to free + */ +void nfs_rw_header_free(struct nfs_pgio_header *hdr) +{ + hdr->rw_ops->rw_free_header(NFS_RW_HEADER(hdr)); +} +EXPORT_SYMBOL_GPL(nfs_rw_header_free); + +/** + * nfs_pgio_data_alloc - Allocate pageio data + * @hdr: The header making a request + * @pagecount: Number of pages to create + */ +static struct nfs_pgio_data *nfs_pgio_data_alloc(struct nfs_pgio_header *hdr, + unsigned int pagecount) +{ + struct nfs_pgio_data *data, *prealloc; + + prealloc = &NFS_RW_HEADER(hdr)->rpc_data; + if (prealloc->header == NULL) + data = prealloc; + else + data = kzalloc(sizeof(*data), GFP_KERNEL); + if (!data) + goto out; + + if (nfs_pgarray_set(&data->pages, pagecount)) { + data->header = hdr; + atomic_inc(&hdr->refcnt); + } else { + if (data != prealloc) + kfree(data); + data = NULL; + } +out: + return data; +} + +/** + * nfs_pgio_data_release - Properly free pageio data + * @data: The data to release + */ +void nfs_pgio_data_release(struct nfs_pgio_data *data) +{ + struct nfs_pgio_header *hdr = data->header; + struct nfs_rw_header *pageio_header = NFS_RW_HEADER(hdr); + + put_nfs_open_context(data->args.context); + if (data->pages.pagevec != data->pages.page_array) + kfree(data->pages.pagevec); + if (data == &pageio_header->rpc_data) { + data->header = NULL; + data = NULL; + } + if (atomic_dec_and_test(&hdr->refcnt)) + hdr->completion_ops->completion(hdr); + /* Note: we only free the rpc_task after callbacks are done. + * See the comment in rpc_free_task() for why + */ + kfree(data); +} +EXPORT_SYMBOL_GPL(nfs_pgio_data_release); + +/** + * nfs_pgio_rpcsetup - Set up arguments for a pageio call + * @data: The pageio data + * @count: Number of bytes to read + * @offset: Initial offset + * @how: How to commit data (writes only) + * @cinfo: Commit information for the call (writes only) + */ +static void nfs_pgio_rpcsetup(struct nfs_pgio_data *data, + unsigned int count, unsigned int offset, + int how, struct nfs_commit_info *cinfo) +{ + struct nfs_page *req = data->header->req; + + /* Set up the RPC argument and reply structs + * NB: take care not to mess about with data->commit et al. */ + + data->args.fh = NFS_FH(data->header->inode); + data->args.offset = req_offset(req) + offset; + /* pnfs_set_layoutcommit needs this */ + data->mds_offset = data->args.offset; + data->args.pgbase = req->wb_pgbase + offset; + data->args.pages = data->pages.pagevec; + data->args.count = count; + data->args.context = get_nfs_open_context(req->wb_context); + data->args.lock_context = req->wb_lock_context; + data->args.stable = NFS_UNSTABLE; + switch (how & (FLUSH_STABLE | FLUSH_COND_STABLE)) { + case 0: + break; + case FLUSH_COND_STABLE: + if (nfs_reqs_to_commit(cinfo)) + break; + default: + data->args.stable = NFS_FILE_SYNC; + } + + data->res.fattr = &data->fattr; + data->res.count = count; + data->res.eof = 0; + data->res.verf = &data->verf; + nfs_fattr_init(&data->fattr); +} + +/** + * nfs_pgio_prepare - Prepare pageio data to go over the wire + * @task: The current task + * @calldata: pageio data to prepare + */ +static void nfs_pgio_prepare(struct rpc_task *task, void *calldata) +{ + struct nfs_pgio_data *data = calldata; + int err; + err = NFS_PROTO(data->header->inode)->pgio_rpc_prepare(task, data); + if (err) + rpc_exit(task, err); +} + +int nfs_initiate_pgio(struct rpc_clnt *clnt, struct nfs_pgio_data *data, + const struct rpc_call_ops *call_ops, int how, int flags) +{ + struct rpc_task *task; + struct rpc_message msg = { + .rpc_argp = &data->args, + .rpc_resp = &data->res, + .rpc_cred = data->header->cred, + }; + struct rpc_task_setup task_setup_data = { + .rpc_client = clnt, + .task = &data->task, + .rpc_message = &msg, + .callback_ops = call_ops, + .callback_data = data, + .workqueue = nfsiod_workqueue, + .flags = RPC_TASK_ASYNC | flags, + }; + int ret = 0; + + data->header->rw_ops->rw_initiate(data, &msg, &task_setup_data, how); + + dprintk("NFS: %5u initiated pgio call " + "(req %s/%llu, %u bytes @ offset %llu)\n", + data->task.tk_pid, + data->header->inode->i_sb->s_id, + (unsigned long long)NFS_FILEID(data->header->inode), + data->args.count, + (unsigned long long)data->args.offset); + + task = rpc_run_task(&task_setup_data); + if (IS_ERR(task)) { + ret = PTR_ERR(task); + goto out; + } + if (how & FLUSH_SYNC) { + ret = rpc_wait_for_completion_task(task); + if (ret == 0) + ret = task->tk_status; + } + rpc_put_task(task); +out: + return ret; +} +EXPORT_SYMBOL_GPL(nfs_initiate_pgio); + +/** + * nfs_pgio_error - Clean up from a pageio error + * @desc: IO descriptor + * @hdr: pageio header + */ +static int nfs_pgio_error(struct nfs_pageio_descriptor *desc, + struct nfs_pgio_header *hdr) +{ + set_bit(NFS_IOHDR_REDO, &hdr->flags); + nfs_pgio_data_release(hdr->data); + hdr->data = NULL; + desc->pg_completion_ops->error_cleanup(&desc->pg_list); + return -ENOMEM; +} + +/** + * nfs_pgio_release - Release pageio data + * @calldata: The pageio data to release + */ +static void nfs_pgio_release(void *calldata) +{ + struct nfs_pgio_data *data = calldata; + if (data->header->rw_ops->rw_release) + data->header->rw_ops->rw_release(data); + nfs_pgio_data_release(data); +} + /** * nfs_pageio_init - initialise a page io descriptor * @desc: pointer to descriptor @@ -307,6 +695,7 @@ void nfs_pageio_init(struct nfs_pageio_descriptor *desc, struct inode *inode, const struct nfs_pageio_ops *pg_ops, const struct nfs_pgio_completion_ops *compl_ops, + const struct nfs_rw_ops *rw_ops, size_t bsize, int io_flags) { @@ -320,6 +709,7 @@ void nfs_pageio_init(struct nfs_pageio_descriptor *desc, desc->pg_inode = inode; desc->pg_ops = pg_ops; desc->pg_completion_ops = compl_ops; + desc->pg_rw_ops = rw_ops; desc->pg_ioflags = io_flags; desc->pg_error = 0; desc->pg_lseg = NULL; @@ -328,6 +718,94 @@ void nfs_pageio_init(struct nfs_pageio_descriptor *desc, } EXPORT_SYMBOL_GPL(nfs_pageio_init); +/** + * nfs_pgio_result - Basic pageio error handling + * @task: The task that ran + * @calldata: Pageio data to check + */ +static void nfs_pgio_result(struct rpc_task *task, void *calldata) +{ + struct nfs_pgio_data *data = calldata; + struct inode *inode = data->header->inode; + + dprintk("NFS: %s: %5u, (status %d)\n", __func__, + task->tk_pid, task->tk_status); + + if (data->header->rw_ops->rw_done(task, data, inode) != 0) + return; + if (task->tk_status < 0) + nfs_set_pgio_error(data->header, task->tk_status, data->args.offset); + else + data->header->rw_ops->rw_result(task, data); +} + +/* + * Create an RPC task for the given read or write request and kick it. + * The page must have been locked by the caller. + * + * It may happen that the page we're passed is not marked dirty. + * This is the case if nfs_updatepage detects a conflicting request + * that has been written but not committed. + */ +int nfs_generic_pgio(struct nfs_pageio_descriptor *desc, + struct nfs_pgio_header *hdr) +{ + struct nfs_page *req; + struct page **pages; + struct nfs_pgio_data *data; + struct list_head *head = &desc->pg_list; + struct nfs_commit_info cinfo; + + data = nfs_pgio_data_alloc(hdr, nfs_page_array_len(desc->pg_base, + desc->pg_count)); + if (!data) + return nfs_pgio_error(desc, hdr); + + nfs_init_cinfo(&cinfo, desc->pg_inode, desc->pg_dreq); + pages = data->pages.pagevec; + while (!list_empty(head)) { + req = nfs_list_entry(head->next); + nfs_list_remove_request(req); + nfs_list_add_request(req, &hdr->pages); + *pages++ = req->wb_page; + } + + if ((desc->pg_ioflags & FLUSH_COND_STABLE) && + (desc->pg_moreio || nfs_reqs_to_commit(&cinfo))) + desc->pg_ioflags &= ~FLUSH_COND_STABLE; + + /* Set up the argument struct */ + nfs_pgio_rpcsetup(data, desc->pg_count, 0, desc->pg_ioflags, &cinfo); + hdr->data = data; + desc->pg_rpc_callops = &nfs_pgio_common_ops; + return 0; +} +EXPORT_SYMBOL_GPL(nfs_generic_pgio); + +static int nfs_generic_pg_pgios(struct nfs_pageio_descriptor *desc) +{ + struct nfs_rw_header *rw_hdr; + struct nfs_pgio_header *hdr; + int ret; + + rw_hdr = nfs_rw_header_alloc(desc->pg_rw_ops); + if (!rw_hdr) { + desc->pg_completion_ops->error_cleanup(&desc->pg_list); + return -ENOMEM; + } + hdr = &rw_hdr->header; + nfs_pgheader_init(desc, hdr, nfs_rw_header_free); + atomic_inc(&hdr->refcnt); + ret = nfs_generic_pgio(desc, hdr); + if (ret == 0) + ret = nfs_initiate_pgio(NFS_CLIENT(hdr->inode), + hdr->data, desc->pg_rpc_callops, + desc->pg_ioflags, 0); + if (atomic_dec_and_test(&hdr->refcnt)) + hdr->completion_ops->completion(hdr); + return ret; +} + static bool nfs_match_open_context(const struct nfs_open_context *ctx1, const struct nfs_open_context *ctx2) { @@ -356,18 +834,23 @@ static bool nfs_can_coalesce_requests(struct nfs_page *prev, struct nfs_page *req, struct nfs_pageio_descriptor *pgio) { - if (!nfs_match_open_context(req->wb_context, prev->wb_context)) - return false; - if (req->wb_context->dentry->d_inode->i_flock != NULL && - !nfs_match_lock_context(req->wb_lock_context, prev->wb_lock_context)) - return false; - if (req->wb_pgbase != 0) - return false; - if (prev->wb_pgbase + prev->wb_bytes != PAGE_CACHE_SIZE) - return false; - if (req_offset(req) != req_offset(prev) + prev->wb_bytes) - return false; - return pgio->pg_ops->pg_test(pgio, prev, req); + size_t size; + + if (prev) { + if (!nfs_match_open_context(req->wb_context, prev->wb_context)) + return false; + if (req->wb_context->dentry->d_inode->i_flock != NULL && + !nfs_match_lock_context(req->wb_lock_context, + prev->wb_lock_context)) + return false; + if (req_offset(req) != req_offset(prev) + prev->wb_bytes) + return false; + } + size = pgio->pg_ops->pg_test(pgio, prev, req); + WARN_ON_ONCE(size > req->wb_bytes); + if (size && size < req->wb_bytes) + req->wb_bytes = size; + return size > 0; } /** @@ -381,17 +864,16 @@ static bool nfs_can_coalesce_requests(struct nfs_page *prev, static int nfs_pageio_do_add_request(struct nfs_pageio_descriptor *desc, struct nfs_page *req) { + struct nfs_page *prev = NULL; if (desc->pg_count != 0) { - struct nfs_page *prev; - prev = nfs_list_entry(desc->pg_list.prev); - if (!nfs_can_coalesce_requests(prev, req, desc)) - return 0; } else { if (desc->pg_ops->pg_init) desc->pg_ops->pg_init(desc, req); desc->pg_base = req->wb_pgbase; } + if (!nfs_can_coalesce_requests(prev, req, desc)) + return 0; nfs_list_remove_request(req); nfs_list_add_request(req, &desc->pg_list); desc->pg_count += req->wb_bytes; @@ -421,22 +903,72 @@ static void nfs_pageio_doio(struct nfs_pageio_descriptor *desc) * @desc: destination io descriptor * @req: request * + * This may split a request into subrequests which are all part of the + * same page group. + * * Returns true if the request 'req' was successfully coalesced into the * existing list of pages 'desc'. */ static int __nfs_pageio_add_request(struct nfs_pageio_descriptor *desc, struct nfs_page *req) { - while (!nfs_pageio_do_add_request(desc, req)) { - desc->pg_moreio = 1; - nfs_pageio_doio(desc); - if (desc->pg_error < 0) - return 0; - desc->pg_moreio = 0; - if (desc->pg_recoalesce) - return 0; - } + struct nfs_page *subreq; + unsigned int bytes_left = 0; + unsigned int offset, pgbase; + + nfs_page_group_lock(req); + + subreq = req; + bytes_left = subreq->wb_bytes; + offset = subreq->wb_offset; + pgbase = subreq->wb_pgbase; + + do { + if (!nfs_pageio_do_add_request(desc, subreq)) { + /* make sure pg_test call(s) did nothing */ + WARN_ON_ONCE(subreq->wb_bytes != bytes_left); + WARN_ON_ONCE(subreq->wb_offset != offset); + WARN_ON_ONCE(subreq->wb_pgbase != pgbase); + + nfs_page_group_unlock(req); + desc->pg_moreio = 1; + nfs_pageio_doio(desc); + if (desc->pg_error < 0) + return 0; + if (desc->pg_recoalesce) + return 0; + /* retry add_request for this subreq */ + nfs_page_group_lock(req); + continue; + } + + /* check for buggy pg_test call(s) */ + WARN_ON_ONCE(subreq->wb_bytes + subreq->wb_pgbase > PAGE_SIZE); + WARN_ON_ONCE(subreq->wb_bytes > bytes_left); + WARN_ON_ONCE(subreq->wb_bytes == 0); + + bytes_left -= subreq->wb_bytes; + offset += subreq->wb_bytes; + pgbase += subreq->wb_bytes; + + if (bytes_left) { + subreq = nfs_create_request(req->wb_context, + req->wb_page, + subreq, pgbase, bytes_left); + if (IS_ERR(subreq)) + goto err_ptr; + nfs_lock_request(subreq); + subreq->wb_offset = offset; + subreq->wb_index = req->wb_index; + } + } while (bytes_left > 0); + + nfs_page_group_unlock(req); return 1; +err_ptr: + desc->pg_error = PTR_ERR(subreq); + nfs_page_group_unlock(req); + return 0; } static int nfs_do_recoalesce(struct nfs_pageio_descriptor *desc) @@ -449,6 +981,7 @@ static int nfs_do_recoalesce(struct nfs_pageio_descriptor *desc) desc->pg_count = 0; desc->pg_base = 0; desc->pg_recoalesce = 0; + desc->pg_moreio = 0; while (!list_empty(&head)) { struct nfs_page *req; @@ -535,3 +1068,13 @@ void nfs_destroy_nfspagecache(void) kmem_cache_destroy(nfs_page_cachep); } +static const struct rpc_call_ops nfs_pgio_common_ops = { + .rpc_call_prepare = nfs_pgio_prepare, + .rpc_call_done = nfs_pgio_result, + .rpc_release = nfs_pgio_release, +}; + +const struct nfs_pageio_ops nfs_pgio_rw_ops = { + .pg_test = nfs_generic_pg_test, + .pg_doio = nfs_generic_pg_pgios, +}; diff --git a/fs/nfs/pnfs.c b/fs/nfs/pnfs.c index d75d938d36c..6fdcd233d6f 100644 --- a/fs/nfs/pnfs.c +++ b/fs/nfs/pnfs.c @@ -662,7 +662,18 @@ pnfs_destroy_all_layouts(struct nfs_client *clp) */ static bool pnfs_seqid_is_newer(u32 s1, u32 s2) { - return (s32)s1 - (s32)s2 > 0; + return (s32)(s1 - s2) > 0; +} + +static void +pnfs_verify_layout_stateid(struct pnfs_layout_hdr *lo, + const nfs4_stateid *new, + struct list_head *free_me_list) +{ + if (nfs4_stateid_match_other(&lo->plh_stateid, new)) + return; + /* Layout is new! Kill existing layout segments */ + pnfs_mark_matching_lsegs_invalid(lo, free_me_list, NULL); } /* update lo->plh_stateid with new if is more recent */ @@ -1315,6 +1326,7 @@ pnfs_layout_process(struct nfs4_layoutget *lgp) struct nfs4_layoutget_res *res = &lgp->res; struct pnfs_layout_segment *lseg; struct inode *ino = lo->plh_inode; + LIST_HEAD(free_me); int status = 0; /* Inject layout blob into I/O device driver */ @@ -1341,6 +1353,8 @@ pnfs_layout_process(struct nfs4_layoutget *lgp) goto out_forget_reply; } + /* Check that the new stateid matches the old stateid */ + pnfs_verify_layout_stateid(lo, &res->stateid, &free_me); /* Done processing layoutget. Set the layout stateid */ pnfs_set_layout_stateid(lo, &res->stateid, false); @@ -1355,6 +1369,7 @@ pnfs_layout_process(struct nfs4_layoutget *lgp) } spin_unlock(&ino->i_lock); + pnfs_free_lseg_list(&free_me); return lseg; out: return ERR_PTR(status); @@ -1373,11 +1388,6 @@ pnfs_generic_pg_init_read(struct nfs_pageio_descriptor *pgio, struct nfs_page *r WARN_ON_ONCE(pgio->pg_lseg != NULL); - if (req->wb_offset != req->wb_pgbase) { - nfs_pageio_reset_read_mds(pgio); - return; - } - if (pgio->pg_dreq == NULL) rd_size = i_size_read(pgio->pg_inode) - req_offset(req); else @@ -1402,11 +1412,6 @@ pnfs_generic_pg_init_write(struct nfs_pageio_descriptor *pgio, { WARN_ON_ONCE(pgio->pg_lseg != NULL); - if (req->wb_offset != req->wb_pgbase) { - nfs_pageio_reset_write_mds(pgio); - return; - } - pgio->pg_lseg = pnfs_update_layout(pgio->pg_inode, req->wb_context, req_offset(req), @@ -1419,56 +1424,49 @@ pnfs_generic_pg_init_write(struct nfs_pageio_descriptor *pgio, } EXPORT_SYMBOL_GPL(pnfs_generic_pg_init_write); -void -pnfs_pageio_init_read(struct nfs_pageio_descriptor *pgio, struct inode *inode, - const struct nfs_pgio_completion_ops *compl_ops) -{ - struct nfs_server *server = NFS_SERVER(inode); - struct pnfs_layoutdriver_type *ld = server->pnfs_curr_ld; - - if (ld == NULL) - nfs_pageio_init_read(pgio, inode, compl_ops); - else - nfs_pageio_init(pgio, inode, ld->pg_read_ops, compl_ops, server->rsize, 0); -} - -void -pnfs_pageio_init_write(struct nfs_pageio_descriptor *pgio, struct inode *inode, - int ioflags, - const struct nfs_pgio_completion_ops *compl_ops) -{ - struct nfs_server *server = NFS_SERVER(inode); - struct pnfs_layoutdriver_type *ld = server->pnfs_curr_ld; - - if (ld == NULL) - nfs_pageio_init_write(pgio, inode, ioflags, compl_ops); - else - nfs_pageio_init(pgio, inode, ld->pg_write_ops, compl_ops, server->wsize, ioflags); -} - -bool +/* + * Return 0 if @req cannot be coalesced into @pgio, otherwise return the number + * of bytes (maximum @req->wb_bytes) that can be coalesced. + */ +size_t pnfs_generic_pg_test(struct nfs_pageio_descriptor *pgio, struct nfs_page *prev, struct nfs_page *req) { - if (pgio->pg_lseg == NULL) - return nfs_generic_pg_test(pgio, prev, req); + unsigned int size; + u64 seg_end, req_start, seg_left; + + size = nfs_generic_pg_test(pgio, prev, req); + if (!size) + return 0; /* - * Test if a nfs_page is fully contained in the pnfs_layout_range. - * Note that this test makes several assumptions: - * - that the previous nfs_page in the struct nfs_pageio_descriptor - * is known to lie within the range. - * - that the nfs_page being tested is known to be contiguous with the - * previous nfs_page. - * - Layout ranges are page aligned, so we only have to test the - * start offset of the request. + * 'size' contains the number of bytes left in the current page (up + * to the original size asked for in @req->wb_bytes). + * + * Calculate how many bytes are left in the layout segment + * and if there are less bytes than 'size', return that instead. * * Please also note that 'end_offset' is actually the offset of the * first byte that lies outside the pnfs_layout_range. FIXME? * */ - return req_offset(req) < end_offset(pgio->pg_lseg->pls_range.offset, - pgio->pg_lseg->pls_range.length); + if (pgio->pg_lseg) { + seg_end = end_offset(pgio->pg_lseg->pls_range.offset, + pgio->pg_lseg->pls_range.length); + req_start = req_offset(req); + WARN_ON_ONCE(req_start > seg_end); + /* start of request is past the last byte of this segment */ + if (req_start >= seg_end) + return 0; + + /* adjust 'size' iff there are fewer bytes left in the + * segment than what nfs_generic_pg_test returned */ + seg_left = seg_end - req_start; + if (seg_left < size) + size = (unsigned int)seg_left; + } + + return size; } EXPORT_SYMBOL_GPL(pnfs_generic_pg_test); @@ -1481,7 +1479,7 @@ int pnfs_write_done_resend_to_mds(struct inode *inode, LIST_HEAD(failed); /* Resend all requests through the MDS */ - nfs_pageio_init_write(&pgio, inode, FLUSH_STABLE, compl_ops); + nfs_pageio_init_write(&pgio, inode, FLUSH_STABLE, true, compl_ops); pgio.pg_dreq = dreq; while (!list_empty(head)) { struct nfs_page *req = nfs_list_entry(head->next); @@ -1504,7 +1502,7 @@ int pnfs_write_done_resend_to_mds(struct inode *inode, } EXPORT_SYMBOL_GPL(pnfs_write_done_resend_to_mds); -static void pnfs_ld_handle_write_error(struct nfs_write_data *data) +static void pnfs_ld_handle_write_error(struct nfs_pgio_data *data) { struct nfs_pgio_header *hdr = data->header; @@ -1523,7 +1521,7 @@ static void pnfs_ld_handle_write_error(struct nfs_write_data *data) /* * Called by non rpc-based layout drivers */ -void pnfs_ld_write_done(struct nfs_write_data *data) +void pnfs_ld_write_done(struct nfs_pgio_data *data) { struct nfs_pgio_header *hdr = data->header; @@ -1539,7 +1537,7 @@ EXPORT_SYMBOL_GPL(pnfs_ld_write_done); static void pnfs_write_through_mds(struct nfs_pageio_descriptor *desc, - struct nfs_write_data *data) + struct nfs_pgio_data *data) { struct nfs_pgio_header *hdr = data->header; @@ -1548,11 +1546,11 @@ pnfs_write_through_mds(struct nfs_pageio_descriptor *desc, nfs_pageio_reset_write_mds(desc); desc->pg_recoalesce = 1; } - nfs_writedata_release(data); + nfs_pgio_data_release(data); } static enum pnfs_try_status -pnfs_try_to_write_data(struct nfs_write_data *wdata, +pnfs_try_to_write_data(struct nfs_pgio_data *wdata, const struct rpc_call_ops *call_ops, struct pnfs_layout_segment *lseg, int how) @@ -1574,41 +1572,36 @@ pnfs_try_to_write_data(struct nfs_write_data *wdata, } static void -pnfs_do_multiple_writes(struct nfs_pageio_descriptor *desc, struct list_head *head, int how) +pnfs_do_write(struct nfs_pageio_descriptor *desc, + struct nfs_pgio_header *hdr, int how) { - struct nfs_write_data *data; + struct nfs_pgio_data *data = hdr->data; const struct rpc_call_ops *call_ops = desc->pg_rpc_callops; struct pnfs_layout_segment *lseg = desc->pg_lseg; + enum pnfs_try_status trypnfs; desc->pg_lseg = NULL; - while (!list_empty(head)) { - enum pnfs_try_status trypnfs; - - data = list_first_entry(head, struct nfs_write_data, list); - list_del_init(&data->list); - - trypnfs = pnfs_try_to_write_data(data, call_ops, lseg, how); - if (trypnfs == PNFS_NOT_ATTEMPTED) - pnfs_write_through_mds(desc, data); - } + trypnfs = pnfs_try_to_write_data(data, call_ops, lseg, how); + if (trypnfs == PNFS_NOT_ATTEMPTED) + pnfs_write_through_mds(desc, data); pnfs_put_lseg(lseg); } static void pnfs_writehdr_free(struct nfs_pgio_header *hdr) { pnfs_put_lseg(hdr->lseg); - nfs_writehdr_free(hdr); + nfs_rw_header_free(hdr); } EXPORT_SYMBOL_GPL(pnfs_writehdr_free); int pnfs_generic_pg_writepages(struct nfs_pageio_descriptor *desc) { - struct nfs_write_header *whdr; + struct nfs_rw_header *whdr; struct nfs_pgio_header *hdr; int ret; - whdr = nfs_writehdr_alloc(); + whdr = nfs_rw_header_alloc(desc->pg_rw_ops); if (!whdr) { desc->pg_completion_ops->error_cleanup(&desc->pg_list); pnfs_put_lseg(desc->pg_lseg); @@ -1619,12 +1612,12 @@ pnfs_generic_pg_writepages(struct nfs_pageio_descriptor *desc) nfs_pgheader_init(desc, hdr, pnfs_writehdr_free); hdr->lseg = pnfs_get_lseg(desc->pg_lseg); atomic_inc(&hdr->refcnt); - ret = nfs_generic_flush(desc, hdr); + ret = nfs_generic_pgio(desc, hdr); if (ret != 0) { pnfs_put_lseg(desc->pg_lseg); desc->pg_lseg = NULL; } else - pnfs_do_multiple_writes(desc, &hdr->rpc_list, desc->pg_ioflags); + pnfs_do_write(desc, hdr, desc->pg_ioflags); if (atomic_dec_and_test(&hdr->refcnt)) hdr->completion_ops->completion(hdr); return ret; @@ -1640,7 +1633,7 @@ int pnfs_read_done_resend_to_mds(struct inode *inode, LIST_HEAD(failed); /* Resend all requests through the MDS */ - nfs_pageio_init_read(&pgio, inode, compl_ops); + nfs_pageio_init_read(&pgio, inode, true, compl_ops); pgio.pg_dreq = dreq; while (!list_empty(head)) { struct nfs_page *req = nfs_list_entry(head->next); @@ -1659,7 +1652,7 @@ int pnfs_read_done_resend_to_mds(struct inode *inode, } EXPORT_SYMBOL_GPL(pnfs_read_done_resend_to_mds); -static void pnfs_ld_handle_read_error(struct nfs_read_data *data) +static void pnfs_ld_handle_read_error(struct nfs_pgio_data *data) { struct nfs_pgio_header *hdr = data->header; @@ -1678,7 +1671,7 @@ static void pnfs_ld_handle_read_error(struct nfs_read_data *data) /* * Called by non rpc-based layout drivers */ -void pnfs_ld_read_done(struct nfs_read_data *data) +void pnfs_ld_read_done(struct nfs_pgio_data *data) { struct nfs_pgio_header *hdr = data->header; @@ -1694,7 +1687,7 @@ EXPORT_SYMBOL_GPL(pnfs_ld_read_done); static void pnfs_read_through_mds(struct nfs_pageio_descriptor *desc, - struct nfs_read_data *data) + struct nfs_pgio_data *data) { struct nfs_pgio_header *hdr = data->header; @@ -1703,14 +1696,14 @@ pnfs_read_through_mds(struct nfs_pageio_descriptor *desc, nfs_pageio_reset_read_mds(desc); desc->pg_recoalesce = 1; } - nfs_readdata_release(data); + nfs_pgio_data_release(data); } /* * Call the appropriate parallel I/O subsystem read function. */ static enum pnfs_try_status -pnfs_try_to_read_data(struct nfs_read_data *rdata, +pnfs_try_to_read_data(struct nfs_pgio_data *rdata, const struct rpc_call_ops *call_ops, struct pnfs_layout_segment *lseg) { @@ -1732,41 +1725,35 @@ pnfs_try_to_read_data(struct nfs_read_data *rdata, } static void -pnfs_do_multiple_reads(struct nfs_pageio_descriptor *desc, struct list_head *head) +pnfs_do_read(struct nfs_pageio_descriptor *desc, struct nfs_pgio_header *hdr) { - struct nfs_read_data *data; + struct nfs_pgio_data *data = hdr->data; const struct rpc_call_ops *call_ops = desc->pg_rpc_callops; struct pnfs_layout_segment *lseg = desc->pg_lseg; + enum pnfs_try_status trypnfs; desc->pg_lseg = NULL; - while (!list_empty(head)) { - enum pnfs_try_status trypnfs; - - data = list_first_entry(head, struct nfs_read_data, list); - list_del_init(&data->list); - - trypnfs = pnfs_try_to_read_data(data, call_ops, lseg); - if (trypnfs == PNFS_NOT_ATTEMPTED) - pnfs_read_through_mds(desc, data); - } + trypnfs = pnfs_try_to_read_data(data, call_ops, lseg); + if (trypnfs == PNFS_NOT_ATTEMPTED) + pnfs_read_through_mds(desc, data); pnfs_put_lseg(lseg); } static void pnfs_readhdr_free(struct nfs_pgio_header *hdr) { pnfs_put_lseg(hdr->lseg); - nfs_readhdr_free(hdr); + nfs_rw_header_free(hdr); } EXPORT_SYMBOL_GPL(pnfs_readhdr_free); int pnfs_generic_pg_readpages(struct nfs_pageio_descriptor *desc) { - struct nfs_read_header *rhdr; + struct nfs_rw_header *rhdr; struct nfs_pgio_header *hdr; int ret; - rhdr = nfs_readhdr_alloc(); + rhdr = nfs_rw_header_alloc(desc->pg_rw_ops); if (!rhdr) { desc->pg_completion_ops->error_cleanup(&desc->pg_list); ret = -ENOMEM; @@ -1778,18 +1765,27 @@ pnfs_generic_pg_readpages(struct nfs_pageio_descriptor *desc) nfs_pgheader_init(desc, hdr, pnfs_readhdr_free); hdr->lseg = pnfs_get_lseg(desc->pg_lseg); atomic_inc(&hdr->refcnt); - ret = nfs_generic_pagein(desc, hdr); + ret = nfs_generic_pgio(desc, hdr); if (ret != 0) { pnfs_put_lseg(desc->pg_lseg); desc->pg_lseg = NULL; } else - pnfs_do_multiple_reads(desc, &hdr->rpc_list); + pnfs_do_read(desc, hdr); if (atomic_dec_and_test(&hdr->refcnt)) hdr->completion_ops->completion(hdr); return ret; } EXPORT_SYMBOL_GPL(pnfs_generic_pg_readpages); +static void pnfs_clear_layoutcommitting(struct inode *inode) +{ + unsigned long *bitlock = &NFS_I(inode)->flags; + + clear_bit_unlock(NFS_INO_LAYOUTCOMMITTING, bitlock); + smp_mb__after_atomic(); + wake_up_bit(bitlock, NFS_INO_LAYOUTCOMMITTING); +} + /* * There can be multiple RW segments. */ @@ -1807,7 +1803,6 @@ static void pnfs_list_write_lseg(struct inode *inode, struct list_head *listp) static void pnfs_list_write_lseg_done(struct inode *inode, struct list_head *listp) { struct pnfs_layout_segment *lseg, *tmp; - unsigned long *bitlock = &NFS_I(inode)->flags; /* Matched by references in pnfs_set_layoutcommit */ list_for_each_entry_safe(lseg, tmp, listp, pls_lc_list) { @@ -1815,9 +1810,7 @@ static void pnfs_list_write_lseg_done(struct inode *inode, struct list_head *lis pnfs_put_lseg(lseg); } - clear_bit_unlock(NFS_INO_LAYOUTCOMMITTING, bitlock); - smp_mb__after_clear_bit(); - wake_up_bit(bitlock, NFS_INO_LAYOUTCOMMITTING); + pnfs_clear_layoutcommitting(inode); } void pnfs_set_lo_fail(struct pnfs_layout_segment *lseg) @@ -1827,7 +1820,7 @@ void pnfs_set_lo_fail(struct pnfs_layout_segment *lseg) EXPORT_SYMBOL_GPL(pnfs_set_lo_fail); void -pnfs_set_layoutcommit(struct nfs_write_data *wdata) +pnfs_set_layoutcommit(struct nfs_pgio_data *wdata) { struct nfs_pgio_header *hdr = wdata->header; struct inode *inode = hdr->inode; @@ -1881,43 +1874,37 @@ pnfs_layoutcommit_inode(struct inode *inode, bool sync) struct nfs4_layoutcommit_data *data; struct nfs_inode *nfsi = NFS_I(inode); loff_t end_pos; - int status = 0; - - dprintk("--> %s inode %lu\n", __func__, inode->i_ino); + int status; - if (!test_bit(NFS_INO_LAYOUTCOMMIT, &nfsi->flags)) + if (!pnfs_layoutcommit_outstanding(inode)) return 0; - /* Note kzalloc ensures data->res.seq_res.sr_slot == NULL */ - data = kzalloc(sizeof(*data), GFP_NOFS); - if (!data) { - status = -ENOMEM; - goto out; - } - - if (!test_bit(NFS_INO_LAYOUTCOMMIT, &nfsi->flags)) - goto out_free; + dprintk("--> %s inode %lu\n", __func__, inode->i_ino); + status = -EAGAIN; if (test_and_set_bit(NFS_INO_LAYOUTCOMMITTING, &nfsi->flags)) { - if (!sync) { - status = -EAGAIN; - goto out_free; - } - status = wait_on_bit_lock(&nfsi->flags, NFS_INO_LAYOUTCOMMITTING, - nfs_wait_bit_killable, TASK_KILLABLE); + if (!sync) + goto out; + status = wait_on_bit_lock(&nfsi->flags, + NFS_INO_LAYOUTCOMMITTING, + nfs_wait_bit_killable, + TASK_KILLABLE); if (status) - goto out_free; + goto out; } - INIT_LIST_HEAD(&data->lseg_list); + status = -ENOMEM; + /* Note kzalloc ensures data->res.seq_res.sr_slot == NULL */ + data = kzalloc(sizeof(*data), GFP_NOFS); + if (!data) + goto clear_layoutcommitting; + + status = 0; spin_lock(&inode->i_lock); - if (!test_and_clear_bit(NFS_INO_LAYOUTCOMMIT, &nfsi->flags)) { - clear_bit(NFS_INO_LAYOUTCOMMITTING, &nfsi->flags); - spin_unlock(&inode->i_lock); - wake_up_bit(&nfsi->flags, NFS_INO_LAYOUTCOMMITTING); - goto out_free; - } + if (!test_and_clear_bit(NFS_INO_LAYOUTCOMMIT, &nfsi->flags)) + goto out_unlock; + INIT_LIST_HEAD(&data->lseg_list); pnfs_list_write_lseg(inode, &data->lseg_list); end_pos = nfsi->layout->plh_lwb; @@ -1940,8 +1927,11 @@ out: mark_inode_dirty_sync(inode); dprintk("<-- %s status %d\n", __func__, status); return status; -out_free: +out_unlock: + spin_unlock(&inode->i_lock); kfree(data); +clear_layoutcommitting: + pnfs_clear_layoutcommitting(inode); goto out; } diff --git a/fs/nfs/pnfs.h b/fs/nfs/pnfs.h index a4f41810a7f..4fb309a2b4c 100644 --- a/fs/nfs/pnfs.h +++ b/fs/nfs/pnfs.h @@ -113,8 +113,8 @@ struct pnfs_layoutdriver_type { * Return PNFS_ATTEMPTED to indicate the layout code has attempted * I/O, else return PNFS_NOT_ATTEMPTED to fall back to normal NFS */ - enum pnfs_try_status (*read_pagelist) (struct nfs_read_data *nfs_data); - enum pnfs_try_status (*write_pagelist) (struct nfs_write_data *nfs_data, int how); + enum pnfs_try_status (*read_pagelist) (struct nfs_pgio_data *nfs_data); + enum pnfs_try_status (*write_pagelist) (struct nfs_pgio_data *nfs_data, int how); void (*free_deviceid_node) (struct nfs4_deviceid_node *); @@ -180,11 +180,6 @@ extern int nfs4_proc_layoutreturn(struct nfs4_layoutreturn *lrp); void pnfs_get_layout_hdr(struct pnfs_layout_hdr *lo); void pnfs_put_lseg(struct pnfs_layout_segment *lseg); -void pnfs_pageio_init_read(struct nfs_pageio_descriptor *, struct inode *, - const struct nfs_pgio_completion_ops *); -void pnfs_pageio_init_write(struct nfs_pageio_descriptor *, struct inode *, - int, const struct nfs_pgio_completion_ops *); - void set_pnfs_layoutdriver(struct nfs_server *, const struct nfs_fh *, u32); void unset_pnfs_layoutdriver(struct nfs_server *); void pnfs_generic_pg_init_read(struct nfs_pageio_descriptor *, struct nfs_page *); @@ -192,7 +187,8 @@ int pnfs_generic_pg_readpages(struct nfs_pageio_descriptor *desc); void pnfs_generic_pg_init_write(struct nfs_pageio_descriptor *pgio, struct nfs_page *req, u64 wb_size); int pnfs_generic_pg_writepages(struct nfs_pageio_descriptor *desc); -bool pnfs_generic_pg_test(struct nfs_pageio_descriptor *pgio, struct nfs_page *prev, struct nfs_page *req); +size_t pnfs_generic_pg_test(struct nfs_pageio_descriptor *pgio, + struct nfs_page *prev, struct nfs_page *req); void pnfs_set_lo_fail(struct pnfs_layout_segment *lseg); struct pnfs_layout_segment *pnfs_layout_process(struct nfs4_layoutget *lgp); void pnfs_free_lseg_list(struct list_head *tmp_list); @@ -217,13 +213,13 @@ bool pnfs_roc(struct inode *ino); void pnfs_roc_release(struct inode *ino); void pnfs_roc_set_barrier(struct inode *ino, u32 barrier); bool pnfs_roc_drain(struct inode *ino, u32 *barrier, struct rpc_task *task); -void pnfs_set_layoutcommit(struct nfs_write_data *wdata); +void pnfs_set_layoutcommit(struct nfs_pgio_data *wdata); void pnfs_cleanup_layoutcommit(struct nfs4_layoutcommit_data *data); int pnfs_layoutcommit_inode(struct inode *inode, bool sync); int _pnfs_return_layout(struct inode *); int pnfs_commit_and_return_layout(struct inode *); -void pnfs_ld_write_done(struct nfs_write_data *); -void pnfs_ld_read_done(struct nfs_read_data *); +void pnfs_ld_write_done(struct nfs_pgio_data *); +void pnfs_ld_read_done(struct nfs_pgio_data *); struct pnfs_layout_segment *pnfs_update_layout(struct inode *ino, struct nfs_open_context *ctx, loff_t pos, @@ -275,7 +271,7 @@ pnfs_get_lseg(struct pnfs_layout_segment *lseg) { if (lseg) { atomic_inc(&lseg->pls_refcount); - smp_mb__after_atomic_inc(); + smp_mb__after_atomic(); } return lseg; } @@ -359,6 +355,15 @@ pnfs_ld_layoutret_on_setattr(struct inode *inode) PNFS_LAYOUTRET_ON_SETATTR; } +static inline bool +pnfs_layoutcommit_outstanding(struct inode *inode) +{ + struct nfs_inode *nfsi = NFS_I(inode); + + return test_bit(NFS_INO_LAYOUTCOMMIT, &nfsi->flags) != 0 || + test_bit(NFS_INO_LAYOUTCOMMITTING, &nfsi->flags) != 0; +} + static inline int pnfs_return_layout(struct inode *ino) { struct nfs_inode *nfsi = NFS_I(ino); @@ -452,18 +457,6 @@ static inline void unset_pnfs_layoutdriver(struct nfs_server *s) { } -static inline void pnfs_pageio_init_read(struct nfs_pageio_descriptor *pgio, struct inode *inode, - const struct nfs_pgio_completion_ops *compl_ops) -{ - nfs_pageio_init_read(pgio, inode, compl_ops); -} - -static inline void pnfs_pageio_init_write(struct nfs_pageio_descriptor *pgio, struct inode *inode, int ioflags, - const struct nfs_pgio_completion_ops *compl_ops) -{ - nfs_pageio_init_write(pgio, inode, ioflags, compl_ops); -} - static inline int pnfs_commit_list(struct inode *inode, struct list_head *mds_pages, int how, struct nfs_commit_info *cinfo) @@ -515,6 +508,13 @@ pnfs_use_threshold(struct nfs4_threshold **dst, struct nfs4_threshold *src, return false; } +static inline bool +pnfs_layoutcommit_outstanding(struct inode *inode) +{ + return false; +} + + static inline struct nfs4_threshold *pnfs_mdsthreshold_alloc(void) { return NULL; diff --git a/fs/nfs/proc.c b/fs/nfs/proc.c index fddbba2d9ef..c171ce1a8a3 100644 --- a/fs/nfs/proc.c +++ b/fs/nfs/proc.c @@ -357,30 +357,6 @@ nfs_proc_rename_done(struct rpc_task *task, struct inode *old_dir, } static int -nfs_proc_rename(struct inode *old_dir, struct qstr *old_name, - struct inode *new_dir, struct qstr *new_name) -{ - struct nfs_renameargs arg = { - .old_dir = NFS_FH(old_dir), - .old_name = old_name, - .new_dir = NFS_FH(new_dir), - .new_name = new_name, - }; - struct rpc_message msg = { - .rpc_proc = &nfs_procedures[NFSPROC_RENAME], - .rpc_argp = &arg, - }; - int status; - - dprintk("NFS call rename %s -> %s\n", old_name->name, new_name->name); - status = rpc_call_sync(NFS_CLIENT(old_dir), &msg, 0); - nfs_mark_for_revalidate(old_dir); - nfs_mark_for_revalidate(new_dir); - dprintk("NFS reply rename: %d\n", status); - return status; -} - -static int nfs_proc_link(struct inode *inode, struct inode *dir, struct qstr *name) { struct nfs_linkargs arg = { @@ -602,7 +578,7 @@ nfs_proc_pathconf(struct nfs_server *server, struct nfs_fh *fhandle, return 0; } -static int nfs_read_done(struct rpc_task *task, struct nfs_read_data *data) +static int nfs_read_done(struct rpc_task *task, struct nfs_pgio_data *data) { struct inode *inode = data->header->inode; @@ -618,18 +594,18 @@ static int nfs_read_done(struct rpc_task *task, struct nfs_read_data *data) return 0; } -static void nfs_proc_read_setup(struct nfs_read_data *data, struct rpc_message *msg) +static void nfs_proc_read_setup(struct nfs_pgio_data *data, struct rpc_message *msg) { msg->rpc_proc = &nfs_procedures[NFSPROC_READ]; } -static int nfs_proc_read_rpc_prepare(struct rpc_task *task, struct nfs_read_data *data) +static int nfs_proc_pgio_rpc_prepare(struct rpc_task *task, struct nfs_pgio_data *data) { rpc_call_start(task); return 0; } -static int nfs_write_done(struct rpc_task *task, struct nfs_write_data *data) +static int nfs_write_done(struct rpc_task *task, struct nfs_pgio_data *data) { struct inode *inode = data->header->inode; @@ -638,19 +614,13 @@ static int nfs_write_done(struct rpc_task *task, struct nfs_write_data *data) return 0; } -static void nfs_proc_write_setup(struct nfs_write_data *data, struct rpc_message *msg) +static void nfs_proc_write_setup(struct nfs_pgio_data *data, struct rpc_message *msg) { /* Note: NFSv2 ignores @stable and always uses NFS_FILE_SYNC */ data->args.stable = NFS_FILE_SYNC; msg->rpc_proc = &nfs_procedures[NFSPROC_WRITE]; } -static int nfs_proc_write_rpc_prepare(struct rpc_task *task, struct nfs_write_data *data) -{ - rpc_call_start(task); - return 0; -} - static void nfs_proc_commit_rpc_prepare(struct rpc_task *task, struct nfs_commit_data *data) { BUG(); @@ -745,7 +715,6 @@ const struct nfs_rpc_ops nfs_v2_clientops = { .unlink_setup = nfs_proc_unlink_setup, .unlink_rpc_prepare = nfs_proc_unlink_rpc_prepare, .unlink_done = nfs_proc_unlink_done, - .rename = nfs_proc_rename, .rename_setup = nfs_proc_rename_setup, .rename_rpc_prepare = nfs_proc_rename_rpc_prepare, .rename_done = nfs_proc_rename_done, @@ -759,13 +728,10 @@ const struct nfs_rpc_ops nfs_v2_clientops = { .fsinfo = nfs_proc_fsinfo, .pathconf = nfs_proc_pathconf, .decode_dirent = nfs2_decode_dirent, + .pgio_rpc_prepare = nfs_proc_pgio_rpc_prepare, .read_setup = nfs_proc_read_setup, - .read_pageio_init = nfs_pageio_init_read, - .read_rpc_prepare = nfs_proc_read_rpc_prepare, .read_done = nfs_read_done, .write_setup = nfs_proc_write_setup, - .write_pageio_init = nfs_pageio_init_write, - .write_rpc_prepare = nfs_proc_write_rpc_prepare, .write_done = nfs_write_done, .commit_setup = nfs_proc_commit_setup, .commit_rpc_prepare = nfs_proc_commit_rpc_prepare, diff --git a/fs/nfs/read.c b/fs/nfs/read.c index 31db5c366b8..e818a475ca6 100644 --- a/fs/nfs/read.c +++ b/fs/nfs/read.c @@ -24,85 +24,24 @@ #include "internal.h" #include "iostat.h" #include "fscache.h" +#include "pnfs.h" #define NFSDBG_FACILITY NFSDBG_PAGECACHE -static const struct nfs_pageio_ops nfs_pageio_read_ops; -static const struct rpc_call_ops nfs_read_common_ops; static const struct nfs_pgio_completion_ops nfs_async_read_completion_ops; +static const struct nfs_rw_ops nfs_rw_read_ops; static struct kmem_cache *nfs_rdata_cachep; -struct nfs_read_header *nfs_readhdr_alloc(void) +static struct nfs_rw_header *nfs_readhdr_alloc(void) { - struct nfs_read_header *rhdr; - - rhdr = kmem_cache_zalloc(nfs_rdata_cachep, GFP_KERNEL); - if (rhdr) { - struct nfs_pgio_header *hdr = &rhdr->header; - - INIT_LIST_HEAD(&hdr->pages); - INIT_LIST_HEAD(&hdr->rpc_list); - spin_lock_init(&hdr->lock); - atomic_set(&hdr->refcnt, 0); - } - return rhdr; + return kmem_cache_zalloc(nfs_rdata_cachep, GFP_KERNEL); } -EXPORT_SYMBOL_GPL(nfs_readhdr_alloc); -static struct nfs_read_data *nfs_readdata_alloc(struct nfs_pgio_header *hdr, - unsigned int pagecount) +static void nfs_readhdr_free(struct nfs_rw_header *rhdr) { - struct nfs_read_data *data, *prealloc; - - prealloc = &container_of(hdr, struct nfs_read_header, header)->rpc_data; - if (prealloc->header == NULL) - data = prealloc; - else - data = kzalloc(sizeof(*data), GFP_KERNEL); - if (!data) - goto out; - - if (nfs_pgarray_set(&data->pages, pagecount)) { - data->header = hdr; - atomic_inc(&hdr->refcnt); - } else { - if (data != prealloc) - kfree(data); - data = NULL; - } -out: - return data; -} - -void nfs_readhdr_free(struct nfs_pgio_header *hdr) -{ - struct nfs_read_header *rhdr = container_of(hdr, struct nfs_read_header, header); - kmem_cache_free(nfs_rdata_cachep, rhdr); } -EXPORT_SYMBOL_GPL(nfs_readhdr_free); - -void nfs_readdata_release(struct nfs_read_data *rdata) -{ - struct nfs_pgio_header *hdr = rdata->header; - struct nfs_read_header *read_header = container_of(hdr, struct nfs_read_header, header); - - put_nfs_open_context(rdata->args.context); - if (rdata->pages.pagevec != rdata->pages.page_array) - kfree(rdata->pages.pagevec); - if (rdata == &read_header->rpc_data) { - rdata->header = NULL; - rdata = NULL; - } - if (atomic_dec_and_test(&hdr->refcnt)) - hdr->completion_ops->completion(hdr); - /* Note: we only free the rpc_task after callbacks are done. - * See the comment in rpc_free_task() for why - */ - kfree(rdata); -} -EXPORT_SYMBOL_GPL(nfs_readdata_release); static int nfs_return_empty_page(struct page *page) @@ -114,17 +53,24 @@ int nfs_return_empty_page(struct page *page) } void nfs_pageio_init_read(struct nfs_pageio_descriptor *pgio, - struct inode *inode, + struct inode *inode, bool force_mds, const struct nfs_pgio_completion_ops *compl_ops) { - nfs_pageio_init(pgio, inode, &nfs_pageio_read_ops, compl_ops, - NFS_SERVER(inode)->rsize, 0); + struct nfs_server *server = NFS_SERVER(inode); + const struct nfs_pageio_ops *pg_ops = &nfs_pgio_rw_ops; + +#ifdef CONFIG_NFS_V4_1 + if (server->pnfs_curr_ld && !force_mds) + pg_ops = server->pnfs_curr_ld->pg_read_ops; +#endif + nfs_pageio_init(pgio, inode, pg_ops, compl_ops, &nfs_rw_read_ops, + server->rsize, 0); } EXPORT_SYMBOL_GPL(nfs_pageio_init_read); void nfs_pageio_reset_read_mds(struct nfs_pageio_descriptor *pgio) { - pgio->pg_ops = &nfs_pageio_read_ops; + pgio->pg_ops = &nfs_pgio_rw_ops; pgio->pg_bsize = NFS_SERVER(pgio->pg_inode)->rsize; } EXPORT_SYMBOL_GPL(nfs_pageio_reset_read_mds); @@ -139,7 +85,7 @@ int nfs_readpage_async(struct nfs_open_context *ctx, struct inode *inode, len = nfs_page_length(page); if (len == 0) return nfs_return_empty_page(page); - new = nfs_create_request(ctx, inode, page, 0, len); + new = nfs_create_request(ctx, page, NULL, 0, len); if (IS_ERR(new)) { unlock_page(page); return PTR_ERR(new); @@ -147,7 +93,8 @@ int nfs_readpage_async(struct nfs_open_context *ctx, struct inode *inode, if (len < PAGE_CACHE_SIZE) zero_user_segment(page, len, PAGE_CACHE_SIZE); - NFS_PROTO(inode)->read_pageio_init(&pgio, inode, &nfs_async_read_completion_ops); + nfs_pageio_init_read(&pgio, inode, false, + &nfs_async_read_completion_ops); nfs_pageio_add_request(&pgio, new); nfs_pageio_complete(&pgio); NFS_I(inode)->read_io += pgio.pg_bytes_written; @@ -158,20 +105,31 @@ static void nfs_readpage_release(struct nfs_page *req) { struct inode *d_inode = req->wb_context->dentry->d_inode; - if (PageUptodate(req->wb_page)) - nfs_readpage_to_fscache(d_inode, req->wb_page, 0); + dprintk("NFS: read done (%s/%llu %d@%lld)\n", d_inode->i_sb->s_id, + (unsigned long long)NFS_FILEID(d_inode), req->wb_bytes, + (long long)req_offset(req)); - unlock_page(req->wb_page); + if (nfs_page_group_sync_on_bit(req, PG_UNLOCKPAGE)) { + if (PageUptodate(req->wb_page)) + nfs_readpage_to_fscache(d_inode, req->wb_page, 0); - dprintk("NFS: read done (%s/%Ld %d@%Ld)\n", + unlock_page(req->wb_page); + } + + dprintk("NFS: read done (%s/%Lu %d@%Ld)\n", req->wb_context->dentry->d_inode->i_sb->s_id, - (long long)NFS_FILEID(req->wb_context->dentry->d_inode), + (unsigned long long)NFS_FILEID(req->wb_context->dentry->d_inode), req->wb_bytes, (long long)req_offset(req)); nfs_release_request(req); } -/* Note io was page aligned */ +static void nfs_page_group_set_uptodate(struct nfs_page *req) +{ + if (nfs_page_group_sync_on_bit(req, PG_UPTODATE)) + SetPageUptodate(req->wb_page); +} + static void nfs_read_completion(struct nfs_pgio_header *hdr) { unsigned long bytes = 0; @@ -181,21 +139,32 @@ static void nfs_read_completion(struct nfs_pgio_header *hdr) while (!list_empty(&hdr->pages)) { struct nfs_page *req = nfs_list_entry(hdr->pages.next); struct page *page = req->wb_page; + unsigned long start = req->wb_pgbase; + unsigned long end = req->wb_pgbase + req->wb_bytes; if (test_bit(NFS_IOHDR_EOF, &hdr->flags)) { - if (bytes > hdr->good_bytes) - zero_user(page, 0, PAGE_SIZE); - else if (hdr->good_bytes - bytes < PAGE_SIZE) - zero_user_segment(page, - hdr->good_bytes & ~PAGE_MASK, - PAGE_SIZE); + /* note: regions of the page not covered by a + * request are zeroed in nfs_readpage_async / + * readpage_async_filler */ + if (bytes > hdr->good_bytes) { + /* nothing in this request was good, so zero + * the full extent of the request */ + zero_user_segment(page, start, end); + + } else if (hdr->good_bytes - bytes < req->wb_bytes) { + /* part of this request has good bytes, but + * not all. zero the bad bytes */ + start += hdr->good_bytes - bytes; + WARN_ON(start < req->wb_pgbase); + zero_user_segment(page, start, end); + } } bytes += req->wb_bytes; if (test_bit(NFS_IOHDR_ERROR, &hdr->flags)) { if (bytes <= hdr->good_bytes) - SetPageUptodate(page); + nfs_page_group_set_uptodate(req); } else - SetPageUptodate(page); + nfs_page_group_set_uptodate(req); nfs_list_remove_request(req); nfs_readpage_release(req); } @@ -203,95 +172,14 @@ out: hdr->release(hdr); } -int nfs_initiate_read(struct rpc_clnt *clnt, - struct nfs_read_data *data, - const struct rpc_call_ops *call_ops, int flags) +static void nfs_initiate_read(struct nfs_pgio_data *data, struct rpc_message *msg, + struct rpc_task_setup *task_setup_data, int how) { struct inode *inode = data->header->inode; int swap_flags = IS_SWAPFILE(inode) ? NFS_RPC_SWAPFLAGS : 0; - struct rpc_task *task; - struct rpc_message msg = { - .rpc_argp = &data->args, - .rpc_resp = &data->res, - .rpc_cred = data->header->cred, - }; - struct rpc_task_setup task_setup_data = { - .task = &data->task, - .rpc_client = clnt, - .rpc_message = &msg, - .callback_ops = call_ops, - .callback_data = data, - .workqueue = nfsiod_workqueue, - .flags = RPC_TASK_ASYNC | swap_flags | flags, - }; - - /* Set up the initial task struct. */ - NFS_PROTO(inode)->read_setup(data, &msg); - - dprintk("NFS: %5u initiated read call (req %s/%lld, %u bytes @ " - "offset %llu)\n", - data->task.tk_pid, - inode->i_sb->s_id, - (long long)NFS_FILEID(inode), - data->args.count, - (unsigned long long)data->args.offset); - - task = rpc_run_task(&task_setup_data); - if (IS_ERR(task)) - return PTR_ERR(task); - rpc_put_task(task); - return 0; -} -EXPORT_SYMBOL_GPL(nfs_initiate_read); -/* - * Set up the NFS read request struct - */ -static void nfs_read_rpcsetup(struct nfs_read_data *data, - unsigned int count, unsigned int offset) -{ - struct nfs_page *req = data->header->req; - - data->args.fh = NFS_FH(data->header->inode); - data->args.offset = req_offset(req) + offset; - data->args.pgbase = req->wb_pgbase + offset; - data->args.pages = data->pages.pagevec; - data->args.count = count; - data->args.context = get_nfs_open_context(req->wb_context); - data->args.lock_context = req->wb_lock_context; - - data->res.fattr = &data->fattr; - data->res.count = count; - data->res.eof = 0; - nfs_fattr_init(&data->fattr); -} - -static int nfs_do_read(struct nfs_read_data *data, - const struct rpc_call_ops *call_ops) -{ - struct inode *inode = data->header->inode; - - return nfs_initiate_read(NFS_CLIENT(inode), data, call_ops, 0); -} - -static int -nfs_do_multiple_reads(struct list_head *head, - const struct rpc_call_ops *call_ops) -{ - struct nfs_read_data *data; - int ret = 0; - - while (!list_empty(head)) { - int ret2; - - data = list_first_entry(head, struct nfs_read_data, list); - list_del_init(&data->list); - - ret2 = nfs_do_read(data, call_ops); - if (ret == 0) - ret = ret2; - } - return ret; + task_setup_data->flags |= swap_flags; + NFS_PROTO(inode)->read_setup(data, msg); } static void @@ -311,143 +199,14 @@ static const struct nfs_pgio_completion_ops nfs_async_read_completion_ops = { .completion = nfs_read_completion, }; -static void nfs_pagein_error(struct nfs_pageio_descriptor *desc, - struct nfs_pgio_header *hdr) -{ - set_bit(NFS_IOHDR_REDO, &hdr->flags); - while (!list_empty(&hdr->rpc_list)) { - struct nfs_read_data *data = list_first_entry(&hdr->rpc_list, - struct nfs_read_data, list); - list_del(&data->list); - nfs_readdata_release(data); - } - desc->pg_completion_ops->error_cleanup(&desc->pg_list); -} - -/* - * Generate multiple requests to fill a single page. - * - * We optimize to reduce the number of read operations on the wire. If we - * detect that we're reading a page, or an area of a page, that is past the - * end of file, we do not generate NFS read operations but just clear the - * parts of the page that would have come back zero from the server anyway. - * - * We rely on the cached value of i_size to make this determination; another - * client can fill pages on the server past our cached end-of-file, but we - * won't see the new data until our attribute cache is updated. This is more - * or less conventional NFS client behavior. - */ -static int nfs_pagein_multi(struct nfs_pageio_descriptor *desc, - struct nfs_pgio_header *hdr) -{ - struct nfs_page *req = hdr->req; - struct page *page = req->wb_page; - struct nfs_read_data *data; - size_t rsize = desc->pg_bsize, nbytes; - unsigned int offset; - - offset = 0; - nbytes = desc->pg_count; - do { - size_t len = min(nbytes,rsize); - - data = nfs_readdata_alloc(hdr, 1); - if (!data) { - nfs_pagein_error(desc, hdr); - return -ENOMEM; - } - data->pages.pagevec[0] = page; - nfs_read_rpcsetup(data, len, offset); - list_add(&data->list, &hdr->rpc_list); - nbytes -= len; - offset += len; - } while (nbytes != 0); - - nfs_list_remove_request(req); - nfs_list_add_request(req, &hdr->pages); - desc->pg_rpc_callops = &nfs_read_common_ops; - return 0; -} - -static int nfs_pagein_one(struct nfs_pageio_descriptor *desc, - struct nfs_pgio_header *hdr) -{ - struct nfs_page *req; - struct page **pages; - struct nfs_read_data *data; - struct list_head *head = &desc->pg_list; - - data = nfs_readdata_alloc(hdr, nfs_page_array_len(desc->pg_base, - desc->pg_count)); - if (!data) { - nfs_pagein_error(desc, hdr); - return -ENOMEM; - } - - pages = data->pages.pagevec; - while (!list_empty(head)) { - req = nfs_list_entry(head->next); - nfs_list_remove_request(req); - nfs_list_add_request(req, &hdr->pages); - *pages++ = req->wb_page; - } - - nfs_read_rpcsetup(data, desc->pg_count, 0); - list_add(&data->list, &hdr->rpc_list); - desc->pg_rpc_callops = &nfs_read_common_ops; - return 0; -} - -int nfs_generic_pagein(struct nfs_pageio_descriptor *desc, - struct nfs_pgio_header *hdr) -{ - if (desc->pg_bsize < PAGE_CACHE_SIZE) - return nfs_pagein_multi(desc, hdr); - return nfs_pagein_one(desc, hdr); -} -EXPORT_SYMBOL_GPL(nfs_generic_pagein); - -static int nfs_generic_pg_readpages(struct nfs_pageio_descriptor *desc) -{ - struct nfs_read_header *rhdr; - struct nfs_pgio_header *hdr; - int ret; - - rhdr = nfs_readhdr_alloc(); - if (!rhdr) { - desc->pg_completion_ops->error_cleanup(&desc->pg_list); - return -ENOMEM; - } - hdr = &rhdr->header; - nfs_pgheader_init(desc, hdr, nfs_readhdr_free); - atomic_inc(&hdr->refcnt); - ret = nfs_generic_pagein(desc, hdr); - if (ret == 0) - ret = nfs_do_multiple_reads(&hdr->rpc_list, - desc->pg_rpc_callops); - if (atomic_dec_and_test(&hdr->refcnt)) - hdr->completion_ops->completion(hdr); - return ret; -} - -static const struct nfs_pageio_ops nfs_pageio_read_ops = { - .pg_test = nfs_generic_pg_test, - .pg_doio = nfs_generic_pg_readpages, -}; - /* * This is the callback from RPC telling us whether a reply was * received or some error occurred (timeout or socket shutdown). */ -int nfs_readpage_result(struct rpc_task *task, struct nfs_read_data *data) +static int nfs_readpage_done(struct rpc_task *task, struct nfs_pgio_data *data, + struct inode *inode) { - struct inode *inode = data->header->inode; - int status; - - dprintk("NFS: %s: %5u, (status %d)\n", __func__, task->tk_pid, - task->tk_status); - - status = NFS_PROTO(inode)->read_done(task, data); + int status = NFS_PROTO(inode)->read_done(task, data); if (status != 0) return status; @@ -460,10 +219,10 @@ int nfs_readpage_result(struct rpc_task *task, struct nfs_read_data *data) return 0; } -static void nfs_readpage_retry(struct rpc_task *task, struct nfs_read_data *data) +static void nfs_readpage_retry(struct rpc_task *task, struct nfs_pgio_data *data) { - struct nfs_readargs *argp = &data->args; - struct nfs_readres *resp = &data->res; + struct nfs_pgio_args *argp = &data->args; + struct nfs_pgio_res *resp = &data->res; /* This is a short read! */ nfs_inc_stats(data->header->inode, NFSIOS_SHORTREAD); @@ -480,17 +239,11 @@ static void nfs_readpage_retry(struct rpc_task *task, struct nfs_read_data *data rpc_restart_call_prepare(task); } -static void nfs_readpage_result_common(struct rpc_task *task, void *calldata) +static void nfs_readpage_result(struct rpc_task *task, struct nfs_pgio_data *data) { - struct nfs_read_data *data = calldata; struct nfs_pgio_header *hdr = data->header; - /* Note the only returns of nfs_readpage_result are 0 and -EAGAIN */ - if (nfs_readpage_result(task, data) != 0) - return; - if (task->tk_status < 0) - nfs_set_pgio_error(hdr, task->tk_status, data->args.offset); - else if (data->res.eof) { + if (data->res.eof) { loff_t bound; bound = data->args.offset + data->res.count; @@ -505,26 +258,6 @@ static void nfs_readpage_result_common(struct rpc_task *task, void *calldata) nfs_readpage_retry(task, data); } -static void nfs_readpage_release_common(void *calldata) -{ - nfs_readdata_release(calldata); -} - -void nfs_read_prepare(struct rpc_task *task, void *calldata) -{ - struct nfs_read_data *data = calldata; - int err; - err = NFS_PROTO(data->header->inode)->read_rpc_prepare(task, data); - if (err) - rpc_exit(task, err); -} - -static const struct rpc_call_ops nfs_read_common_ops = { - .rpc_call_prepare = nfs_read_prepare, - .rpc_call_done = nfs_readpage_result_common, - .rpc_release = nfs_readpage_release_common, -}; - /* * Read a page over NFS. * We read the page synchronously in the following case: @@ -592,7 +325,6 @@ static int readpage_async_filler(void *data, struct page *page) { struct nfs_readdesc *desc = (struct nfs_readdesc *)data; - struct inode *inode = page_file_mapping(page)->host; struct nfs_page *new; unsigned int len; int error; @@ -601,7 +333,7 @@ readpage_async_filler(void *data, struct page *page) if (len == 0) return nfs_return_empty_page(page); - new = nfs_create_request(desc->ctx, inode, page, 0, len); + new = nfs_create_request(desc->ctx, page, NULL, 0, len); if (IS_ERR(new)) goto out_error; @@ -630,9 +362,9 @@ int nfs_readpages(struct file *filp, struct address_space *mapping, unsigned long npages; int ret = -ESTALE; - dprintk("NFS: nfs_readpages (%s/%Ld %d)\n", + dprintk("NFS: nfs_readpages (%s/%Lu %d)\n", inode->i_sb->s_id, - (long long)NFS_FILEID(inode), + (unsigned long long)NFS_FILEID(inode), nr_pages); nfs_inc_stats(inode, NFSIOS_VFSREADPAGES); @@ -654,7 +386,8 @@ int nfs_readpages(struct file *filp, struct address_space *mapping, if (ret == 0) goto read_complete; /* all pages were read */ - NFS_PROTO(inode)->read_pageio_init(&pgio, inode, &nfs_async_read_completion_ops); + nfs_pageio_init_read(&pgio, inode, false, + &nfs_async_read_completion_ops); ret = read_cache_pages(mapping, pages, readpage_async_filler, &desc); @@ -671,7 +404,7 @@ out: int __init nfs_init_readpagecache(void) { nfs_rdata_cachep = kmem_cache_create("nfs_read_data", - sizeof(struct nfs_read_header), + sizeof(struct nfs_rw_header), 0, SLAB_HWCACHE_ALIGN, NULL); if (nfs_rdata_cachep == NULL) @@ -684,3 +417,12 @@ void nfs_destroy_readpagecache(void) { kmem_cache_destroy(nfs_rdata_cachep); } + +static const struct nfs_rw_ops nfs_rw_read_ops = { + .rw_mode = FMODE_READ, + .rw_alloc_header = nfs_readhdr_alloc, + .rw_free_header = nfs_readhdr_free, + .rw_done = nfs_readpage_done, + .rw_result = nfs_readpage_result, + .rw_initiate = nfs_initiate_read, +}; diff --git a/fs/nfs/super.c b/fs/nfs/super.c index 910ed906eb8..084af1060d7 100644 --- a/fs/nfs/super.c +++ b/fs/nfs/super.c @@ -2180,11 +2180,23 @@ out_no_address: return -EINVAL; } +#define NFS_MOUNT_CMP_FLAGMASK ~(NFS_MOUNT_INTR \ + | NFS_MOUNT_SECURE \ + | NFS_MOUNT_TCP \ + | NFS_MOUNT_VER3 \ + | NFS_MOUNT_KERBEROS \ + | NFS_MOUNT_NONLM \ + | NFS_MOUNT_BROKEN_SUID \ + | NFS_MOUNT_STRICTLOCK \ + | NFS_MOUNT_UNSHARED \ + | NFS_MOUNT_NORESVPORT \ + | NFS_MOUNT_LEGACY_INTERFACE) + static int nfs_compare_remount_data(struct nfs_server *nfss, struct nfs_parsed_mount_data *data) { - if (data->flags != nfss->flags || + if ((data->flags ^ nfss->flags) & NFS_MOUNT_CMP_FLAGMASK || data->rsize != nfss->rsize || data->wsize != nfss->wsize || data->version != nfss->nfs_client->rpc_ops->version || @@ -2215,6 +2227,8 @@ nfs_remount(struct super_block *sb, int *flags, char *raw_data) struct nfs4_mount_data *options4 = (struct nfs4_mount_data *)raw_data; u32 nfsvers = nfss->nfs_client->rpc_ops->version; + sync_filesystem(sb); + /* * Userspace mount programs that send binary options generally send * them populated with default values. We have no way to know which @@ -2246,6 +2260,7 @@ nfs_remount(struct super_block *sb, int *flags, char *raw_data) data->nfs_server.addrlen = nfss->nfs_client->cl_addrlen; data->version = nfsvers; data->minorversion = nfss->nfs_client->cl_minorversion; + data->net = current->nsproxy->net_ns; memcpy(&data->nfs_server.address, &nfss->nfs_client->cl_addr, data->nfs_server.addrlen); @@ -2345,18 +2360,6 @@ void nfs_clone_super(struct super_block *sb, struct nfs_mount_info *mount_info) nfs_initialise_sb(sb); } -#define NFS_MOUNT_CMP_FLAGMASK ~(NFS_MOUNT_INTR \ - | NFS_MOUNT_SECURE \ - | NFS_MOUNT_TCP \ - | NFS_MOUNT_VER3 \ - | NFS_MOUNT_KERBEROS \ - | NFS_MOUNT_NONLM \ - | NFS_MOUNT_BROKEN_SUID \ - | NFS_MOUNT_STRICTLOCK \ - | NFS_MOUNT_UNSHARED \ - | NFS_MOUNT_NORESVPORT \ - | NFS_MOUNT_LEGACY_INTERFACE) - static int nfs_compare_mount_options(const struct super_block *s, const struct nfs_server *b, int flags) { const struct nfs_server *a = s->s_fs_info; diff --git a/fs/nfs/sysctl.c b/fs/nfs/sysctl.c index 6b3f2535a3e..bb6ed810fa6 100644 --- a/fs/nfs/sysctl.c +++ b/fs/nfs/sysctl.c @@ -13,7 +13,7 @@ static struct ctl_table_header *nfs_callback_sysctl_table; -static ctl_table nfs_cb_sysctls[] = { +static struct ctl_table nfs_cb_sysctls[] = { { .procname = "nfs_mountpoint_timeout", .data = &nfs_mountpoint_expiry_timeout, @@ -31,7 +31,7 @@ static ctl_table nfs_cb_sysctls[] = { { } }; -static ctl_table nfs_cb_sysctl_dir[] = { +static struct ctl_table nfs_cb_sysctl_dir[] = { { .procname = "nfs", .mode = 0555, @@ -40,7 +40,7 @@ static ctl_table nfs_cb_sysctl_dir[] = { { } }; -static ctl_table nfs_cb_sysctl_root[] = { +static struct ctl_table nfs_cb_sysctl_root[] = { { .procname = "fs", .mode = 0555, diff --git a/fs/nfs/unlink.c b/fs/nfs/unlink.c index 11d78944de7..de54129336c 100644 --- a/fs/nfs/unlink.c +++ b/fs/nfs/unlink.c @@ -14,6 +14,7 @@ #include <linux/sched.h> #include <linux/wait.h> #include <linux/namei.h> +#include <linux/fsnotify.h> #include "internal.h" #include "nfs4_fs.h" @@ -353,8 +354,8 @@ static void nfs_async_rename_done(struct rpc_task *task, void *calldata) return; } - if (task->tk_status != 0) - nfs_cancel_async_unlink(old_dentry); + if (data->complete) + data->complete(task, data); } /** @@ -399,9 +400,10 @@ static const struct rpc_call_ops nfs_rename_ops = { * * It's expected that valid references to the dentries and inodes are held */ -static struct rpc_task * +struct rpc_task * nfs_async_rename(struct inode *old_dir, struct inode *new_dir, - struct dentry *old_dentry, struct dentry *new_dentry) + struct dentry *old_dentry, struct dentry *new_dentry, + void (*complete)(struct rpc_task *, struct nfs_renamedata *)) { struct nfs_renamedata *data; struct rpc_message msg = { }; @@ -438,6 +440,7 @@ nfs_async_rename(struct inode *old_dir, struct inode *new_dir, data->new_dentry = dget(new_dentry); nfs_fattr_init(&data->old_fattr); nfs_fattr_init(&data->new_fattr); + data->complete = complete; /* set up nfs_renameargs */ data->args.old_dir = NFS_FH(old_dir); @@ -456,6 +459,27 @@ nfs_async_rename(struct inode *old_dir, struct inode *new_dir, return rpc_run_task(&task_setup_data); } +/* + * Perform tasks needed when a sillyrename is done such as cancelling the + * queued async unlink if it failed. + */ +static void +nfs_complete_sillyrename(struct rpc_task *task, struct nfs_renamedata *data) +{ + struct dentry *dentry = data->old_dentry; + + if (task->tk_status != 0) { + nfs_cancel_async_unlink(dentry); + return; + } + + /* + * vfs_unlink and the like do not issue this when a file is + * sillyrenamed, so do it here. + */ + fsnotify_nameremove(dentry, 0); +} + #define SILLYNAME_PREFIX ".nfs" #define SILLYNAME_PREFIX_LEN ((unsigned)sizeof(SILLYNAME_PREFIX) - 1) #define SILLYNAME_FILEID_LEN ((unsigned)sizeof(u64) << 1) @@ -548,7 +572,8 @@ nfs_sillyrename(struct inode *dir, struct dentry *dentry) } /* run the rename task, undo unlink if it fails */ - task = nfs_async_rename(dir, dir, dentry, sdentry); + task = nfs_async_rename(dir, dir, dentry, sdentry, + nfs_complete_sillyrename); if (IS_ERR(task)) { error = -EBUSY; nfs_cancel_async_unlink(dentry); diff --git a/fs/nfs/write.c b/fs/nfs/write.c index c1d548211c3..5e2f1030454 100644 --- a/fs/nfs/write.c +++ b/fs/nfs/write.c @@ -42,10 +42,11 @@ * Local function declarations */ static void nfs_redirty_request(struct nfs_page *req); -static const struct rpc_call_ops nfs_write_common_ops; static const struct rpc_call_ops nfs_commit_ops; static const struct nfs_pgio_completion_ops nfs_async_write_completion_ops; static const struct nfs_commit_completion_ops nfs_commit_completion_ops; +static const struct nfs_rw_ops nfs_rw_write_ops; +static void nfs_clear_request_commit(struct nfs_page *req); static struct kmem_cache *nfs_wdata_cachep; static mempool_t *nfs_wdata_mempool; @@ -70,76 +71,19 @@ void nfs_commit_free(struct nfs_commit_data *p) } EXPORT_SYMBOL_GPL(nfs_commit_free); -struct nfs_write_header *nfs_writehdr_alloc(void) +static struct nfs_rw_header *nfs_writehdr_alloc(void) { - struct nfs_write_header *p = mempool_alloc(nfs_wdata_mempool, GFP_NOIO); - - if (p) { - struct nfs_pgio_header *hdr = &p->header; + struct nfs_rw_header *p = mempool_alloc(nfs_wdata_mempool, GFP_NOIO); + if (p) memset(p, 0, sizeof(*p)); - INIT_LIST_HEAD(&hdr->pages); - INIT_LIST_HEAD(&hdr->rpc_list); - spin_lock_init(&hdr->lock); - atomic_set(&hdr->refcnt, 0); - hdr->verf = &p->verf; - } return p; } -EXPORT_SYMBOL_GPL(nfs_writehdr_alloc); -static struct nfs_write_data *nfs_writedata_alloc(struct nfs_pgio_header *hdr, - unsigned int pagecount) +static void nfs_writehdr_free(struct nfs_rw_header *whdr) { - struct nfs_write_data *data, *prealloc; - - prealloc = &container_of(hdr, struct nfs_write_header, header)->rpc_data; - if (prealloc->header == NULL) - data = prealloc; - else - data = kzalloc(sizeof(*data), GFP_KERNEL); - if (!data) - goto out; - - if (nfs_pgarray_set(&data->pages, pagecount)) { - data->header = hdr; - atomic_inc(&hdr->refcnt); - } else { - if (data != prealloc) - kfree(data); - data = NULL; - } -out: - return data; -} - -void nfs_writehdr_free(struct nfs_pgio_header *hdr) -{ - struct nfs_write_header *whdr = container_of(hdr, struct nfs_write_header, header); mempool_free(whdr, nfs_wdata_mempool); } -EXPORT_SYMBOL_GPL(nfs_writehdr_free); - -void nfs_writedata_release(struct nfs_write_data *wdata) -{ - struct nfs_pgio_header *hdr = wdata->header; - struct nfs_write_header *write_header = container_of(hdr, struct nfs_write_header, header); - - put_nfs_open_context(wdata->args.context); - if (wdata->pages.pagevec != wdata->pages.page_array) - kfree(wdata->pages.pagevec); - if (wdata == &write_header->rpc_data) { - wdata->header = NULL; - wdata = NULL; - } - if (atomic_dec_and_test(&hdr->refcnt)) - hdr->completion_ops->completion(hdr); - /* Note: we only free the rpc_task after callbacks are done. - * See the comment in rpc_free_task() for why - */ - kfree(wdata); -} -EXPORT_SYMBOL_GPL(nfs_writedata_release); static void nfs_context_set_write_error(struct nfs_open_context *ctx, int error) { @@ -148,8 +92,15 @@ static void nfs_context_set_write_error(struct nfs_open_context *ctx, int error) set_bit(NFS_CONTEXT_ERROR_WRITE, &ctx->flags); } +/* + * nfs_page_find_head_request_locked - find head request associated with @page + * + * must be called while holding the inode lock. + * + * returns matching head request with reference held, or NULL if not found. + */ static struct nfs_page * -nfs_page_find_request_locked(struct nfs_inode *nfsi, struct page *page) +nfs_page_find_head_request_locked(struct nfs_inode *nfsi, struct page *page) { struct nfs_page *req = NULL; @@ -161,25 +112,33 @@ nfs_page_find_request_locked(struct nfs_inode *nfsi, struct page *page) /* Linearly search the commit list for the correct req */ list_for_each_entry_safe(freq, t, &nfsi->commit_info.list, wb_list) { if (freq->wb_page == page) { - req = freq; + req = freq->wb_head; break; } } } - if (req) + if (req) { + WARN_ON_ONCE(req->wb_head != req); + kref_get(&req->wb_kref); + } return req; } -static struct nfs_page *nfs_page_find_request(struct page *page) +/* + * nfs_page_find_head_request - find head request associated with @page + * + * returns matching head request with reference held, or NULL if not found. + */ +static struct nfs_page *nfs_page_find_head_request(struct page *page) { struct inode *inode = page_file_mapping(page)->host; struct nfs_page *req = NULL; spin_lock(&inode->i_lock); - req = nfs_page_find_request_locked(NFS_I(inode), page); + req = nfs_page_find_head_request_locked(NFS_I(inode), page); spin_unlock(&inode->i_lock); return req; } @@ -211,18 +170,78 @@ static void nfs_set_pageerror(struct page *page) nfs_zap_mapping(page_file_mapping(page)->host, page_file_mapping(page)); } +/* + * nfs_page_group_search_locked + * @head - head request of page group + * @page_offset - offset into page + * + * Search page group with head @head to find a request that contains the + * page offset @page_offset. + * + * Returns a pointer to the first matching nfs request, or NULL if no + * match is found. + * + * Must be called with the page group lock held + */ +static struct nfs_page * +nfs_page_group_search_locked(struct nfs_page *head, unsigned int page_offset) +{ + struct nfs_page *req; + + WARN_ON_ONCE(head != head->wb_head); + WARN_ON_ONCE(!test_bit(PG_HEADLOCK, &head->wb_head->wb_flags)); + + req = head; + do { + if (page_offset >= req->wb_pgbase && + page_offset < (req->wb_pgbase + req->wb_bytes)) + return req; + + req = req->wb_this_page; + } while (req != head); + + return NULL; +} + +/* + * nfs_page_group_covers_page + * @head - head request of page group + * + * Return true if the page group with head @head covers the whole page, + * returns false otherwise + */ +static bool nfs_page_group_covers_page(struct nfs_page *req) +{ + struct nfs_page *tmp; + unsigned int pos = 0; + unsigned int len = nfs_page_length(req->wb_page); + + nfs_page_group_lock(req); + + do { + tmp = nfs_page_group_search_locked(req->wb_head, pos); + if (tmp) { + /* no way this should happen */ + WARN_ON_ONCE(tmp->wb_pgbase != pos); + pos += tmp->wb_bytes - (pos - tmp->wb_pgbase); + } + } while (tmp && pos < len); + + nfs_page_group_unlock(req); + WARN_ON_ONCE(pos > len); + return pos == len; +} + /* We can set the PG_uptodate flag if we see that a write request * covers the full page. */ -static void nfs_mark_uptodate(struct page *page, unsigned int base, unsigned int count) +static void nfs_mark_uptodate(struct nfs_page *req) { - if (PageUptodate(page)) - return; - if (base != 0) + if (PageUptodate(req->wb_page)) return; - if (count != nfs_page_length(page)) + if (!nfs_page_group_covers_page(req)) return; - SetPageUptodate(page); + SetPageUptodate(req->wb_page); } static int wb_priority(struct writeback_control *wbc) @@ -258,46 +277,259 @@ static void nfs_set_page_writeback(struct page *page) } } -static void nfs_end_page_writeback(struct page *page) +static void nfs_end_page_writeback(struct nfs_page *req) { - struct inode *inode = page_file_mapping(page)->host; + struct inode *inode = page_file_mapping(req->wb_page)->host; struct nfs_server *nfss = NFS_SERVER(inode); - end_page_writeback(page); + if (!nfs_page_group_sync_on_bit(req, PG_WB_END)) + return; + + end_page_writeback(req->wb_page); if (atomic_long_dec_return(&nfss->writeback) < NFS_CONGESTION_OFF_THRESH) clear_bdi_congested(&nfss->backing_dev_info, BLK_RW_ASYNC); } -static struct nfs_page *nfs_find_and_lock_request(struct page *page, bool nonblock) + +/* nfs_page_group_clear_bits + * @req - an nfs request + * clears all page group related bits from @req + */ +static void +nfs_page_group_clear_bits(struct nfs_page *req) +{ + clear_bit(PG_TEARDOWN, &req->wb_flags); + clear_bit(PG_UNLOCKPAGE, &req->wb_flags); + clear_bit(PG_UPTODATE, &req->wb_flags); + clear_bit(PG_WB_END, &req->wb_flags); + clear_bit(PG_REMOVE, &req->wb_flags); +} + + +/* + * nfs_unroll_locks_and_wait - unlock all newly locked reqs and wait on @req + * + * this is a helper function for nfs_lock_and_join_requests + * + * @inode - inode associated with request page group, must be holding inode lock + * @head - head request of page group, must be holding head lock + * @req - request that couldn't lock and needs to wait on the req bit lock + * @nonblock - if true, don't actually wait + * + * NOTE: this must be called holding page_group bit lock and inode spin lock + * and BOTH will be released before returning. + * + * returns 0 on success, < 0 on error. + */ +static int +nfs_unroll_locks_and_wait(struct inode *inode, struct nfs_page *head, + struct nfs_page *req, bool nonblock) + __releases(&inode->i_lock) +{ + struct nfs_page *tmp; + int ret; + + /* relinquish all the locks successfully grabbed this run */ + for (tmp = head ; tmp != req; tmp = tmp->wb_this_page) + nfs_unlock_request(tmp); + + WARN_ON_ONCE(test_bit(PG_TEARDOWN, &req->wb_flags)); + + /* grab a ref on the request that will be waited on */ + kref_get(&req->wb_kref); + + nfs_page_group_unlock(head); + spin_unlock(&inode->i_lock); + + /* release ref from nfs_page_find_head_request_locked */ + nfs_release_request(head); + + if (!nonblock) + ret = nfs_wait_on_request(req); + else + ret = -EAGAIN; + nfs_release_request(req); + + return ret; +} + +/* + * nfs_destroy_unlinked_subrequests - destroy recently unlinked subrequests + * + * @destroy_list - request list (using wb_this_page) terminated by @old_head + * @old_head - the old head of the list + * + * All subrequests must be locked and removed from all lists, so at this point + * they are only "active" in this function, and possibly in nfs_wait_on_request + * with a reference held by some other context. + */ +static void +nfs_destroy_unlinked_subrequests(struct nfs_page *destroy_list, + struct nfs_page *old_head) +{ + while (destroy_list) { + struct nfs_page *subreq = destroy_list; + + destroy_list = (subreq->wb_this_page == old_head) ? + NULL : subreq->wb_this_page; + + WARN_ON_ONCE(old_head != subreq->wb_head); + + /* make sure old group is not used */ + subreq->wb_head = subreq; + subreq->wb_this_page = subreq; + + nfs_clear_request_commit(subreq); + + /* subreq is now totally disconnected from page group or any + * write / commit lists. last chance to wake any waiters */ + nfs_unlock_request(subreq); + + if (!test_bit(PG_TEARDOWN, &subreq->wb_flags)) { + /* release ref on old head request */ + nfs_release_request(old_head); + + nfs_page_group_clear_bits(subreq); + + /* release the PG_INODE_REF reference */ + if (test_and_clear_bit(PG_INODE_REF, &subreq->wb_flags)) + nfs_release_request(subreq); + else + WARN_ON_ONCE(1); + } else { + WARN_ON_ONCE(test_bit(PG_CLEAN, &subreq->wb_flags)); + /* zombie requests have already released the last + * reference and were waiting on the rest of the + * group to complete. Since it's no longer part of a + * group, simply free the request */ + nfs_page_group_clear_bits(subreq); + nfs_free_request(subreq); + } + } +} + +/* + * nfs_lock_and_join_requests - join all subreqs to the head req and return + * a locked reference, cancelling any pending + * operations for this page. + * + * @page - the page used to lookup the "page group" of nfs_page structures + * @nonblock - if true, don't block waiting for request locks + * + * This function joins all sub requests to the head request by first + * locking all requests in the group, cancelling any pending operations + * and finally updating the head request to cover the whole range covered by + * the (former) group. All subrequests are removed from any write or commit + * lists, unlinked from the group and destroyed. + * + * Returns a locked, referenced pointer to the head request - which after + * this call is guaranteed to be the only request associated with the page. + * Returns NULL if no requests are found for @page, or a ERR_PTR if an + * error was encountered. + */ +static struct nfs_page * +nfs_lock_and_join_requests(struct page *page, bool nonblock) { struct inode *inode = page_file_mapping(page)->host; - struct nfs_page *req; + struct nfs_page *head, *subreq; + struct nfs_page *destroy_list = NULL; + unsigned int total_bytes; int ret; +try_again: + total_bytes = 0; + + WARN_ON_ONCE(destroy_list); + spin_lock(&inode->i_lock); - for (;;) { - req = nfs_page_find_request_locked(NFS_I(inode), page); - if (req == NULL) - break; - if (nfs_lock_request(req)) - break; - /* Note: If we hold the page lock, as is the case in nfs_writepage, - * then the call to nfs_lock_request() will always - * succeed provided that someone hasn't already marked the - * request as dirty (in which case we don't care). - */ + + /* + * A reference is taken only on the head request which acts as a + * reference to the whole page group - the group will not be destroyed + * until the head reference is released. + */ + head = nfs_page_find_head_request_locked(NFS_I(inode), page); + + if (!head) { spin_unlock(&inode->i_lock); - if (!nonblock) - ret = nfs_wait_on_request(req); - else - ret = -EAGAIN; - nfs_release_request(req); - if (ret != 0) + return NULL; + } + + /* lock each request in the page group */ + nfs_page_group_lock(head); + subreq = head; + do { + /* + * Subrequests are always contiguous, non overlapping + * and in order. If not, it's a programming error. + */ + WARN_ON_ONCE(subreq->wb_offset != + (head->wb_offset + total_bytes)); + + /* keep track of how many bytes this group covers */ + total_bytes += subreq->wb_bytes; + + if (!nfs_lock_request(subreq)) { + /* releases page group bit lock and + * inode spin lock and all references */ + ret = nfs_unroll_locks_and_wait(inode, head, + subreq, nonblock); + + if (ret == 0) + goto try_again; + return ERR_PTR(ret); - spin_lock(&inode->i_lock); + } + + subreq = subreq->wb_this_page; + } while (subreq != head); + + /* Now that all requests are locked, make sure they aren't on any list. + * Commit list removal accounting is done after locks are dropped */ + subreq = head; + do { + nfs_list_remove_request(subreq); + subreq = subreq->wb_this_page; + } while (subreq != head); + + /* unlink subrequests from head, destroy them later */ + if (head->wb_this_page != head) { + /* destroy list will be terminated by head */ + destroy_list = head->wb_this_page; + head->wb_this_page = head; + + /* change head request to cover whole range that + * the former page group covered */ + head->wb_bytes = total_bytes; } + + /* + * prepare head request to be added to new pgio descriptor + */ + nfs_page_group_clear_bits(head); + + /* + * some part of the group was still on the inode list - otherwise + * the group wouldn't be involved in async write. + * grab a reference for the head request, iff it needs one. + */ + if (!test_and_set_bit(PG_INODE_REF, &head->wb_flags)) + kref_get(&head->wb_kref); + + nfs_page_group_unlock(head); + + /* drop lock to clear_request_commit the head req and clean up + * requests on destroy list */ spin_unlock(&inode->i_lock); - return req; + + nfs_destroy_unlinked_subrequests(destroy_list, head); + + /* clean up commit list state */ + nfs_clear_request_commit(head); + + /* still holds ref on head from nfs_page_find_head_request_locked + * and still has lock on head from lock loop */ + return head; } /* @@ -310,7 +542,7 @@ static int nfs_page_async_flush(struct nfs_pageio_descriptor *pgio, struct nfs_page *req; int ret = 0; - req = nfs_find_and_lock_request(page, nonblock); + req = nfs_lock_and_join_requests(page, nonblock); if (!req) goto out; ret = PTR_ERR(req); @@ -354,10 +586,8 @@ static int nfs_writepage_locked(struct page *page, struct writeback_control *wbc struct nfs_pageio_descriptor pgio; int err; - NFS_PROTO(page_file_mapping(page)->host)->write_pageio_init(&pgio, - page->mapping->host, - wb_priority(wbc), - &nfs_async_write_completion_ops); + nfs_pageio_init_write(&pgio, page->mapping->host, wb_priority(wbc), + false, &nfs_async_write_completion_ops); err = nfs_do_writepage(page, wbc, &pgio); nfs_pageio_complete(&pgio); if (err < 0) @@ -400,12 +630,13 @@ int nfs_writepages(struct address_space *mapping, struct writeback_control *wbc) nfs_inc_stats(inode, NFSIOS_VFSWRITEPAGES); - NFS_PROTO(inode)->write_pageio_init(&pgio, inode, wb_priority(wbc), &nfs_async_write_completion_ops); + nfs_pageio_init_write(&pgio, inode, wb_priority(wbc), false, + &nfs_async_write_completion_ops); err = write_cache_pages(mapping, wbc, nfs_writepages_callback, &pgio); nfs_pageio_complete(&pgio); clear_bit_unlock(NFS_INO_FLUSHING, bitlock); - smp_mb__after_clear_bit(); + smp_mb__after_atomic(); wake_up_bit(bitlock, NFS_INO_FLUSHING); if (err < 0) @@ -425,6 +656,8 @@ static void nfs_inode_add_request(struct inode *inode, struct nfs_page *req) { struct nfs_inode *nfsi = NFS_I(inode); + WARN_ON_ONCE(req->wb_this_page != req); + /* Lock the request! */ nfs_lock_request(req); @@ -441,6 +674,9 @@ static void nfs_inode_add_request(struct inode *inode, struct nfs_page *req) set_page_private(req->wb_page, (unsigned long)req); } nfsi->npages++; + /* this a head request for a page group - mark it as having an + * extra reference so sub groups can follow suit */ + WARN_ON(test_and_set_bit(PG_INODE_REF, &req->wb_flags)); kref_get(&req->wb_kref); spin_unlock(&inode->i_lock); } @@ -452,16 +688,23 @@ static void nfs_inode_remove_request(struct nfs_page *req) { struct inode *inode = req->wb_context->dentry->d_inode; struct nfs_inode *nfsi = NFS_I(inode); + struct nfs_page *head; - spin_lock(&inode->i_lock); - if (likely(!PageSwapCache(req->wb_page))) { - set_page_private(req->wb_page, 0); - ClearPagePrivate(req->wb_page); - clear_bit(PG_MAPPED, &req->wb_flags); + if (nfs_page_group_sync_on_bit(req, PG_REMOVE)) { + head = req->wb_head; + + spin_lock(&inode->i_lock); + if (likely(!PageSwapCache(head->wb_page))) { + set_page_private(head->wb_page, 0); + ClearPagePrivate(head->wb_page); + clear_bit(PG_MAPPED, &head->wb_flags); + } + nfsi->npages--; + spin_unlock(&inode->i_lock); } - nfsi->npages--; - spin_unlock(&inode->i_lock); - nfs_release_request(req); + + if (test_and_clear_bit(PG_INODE_REF, &req->wb_flags)) + nfs_release_request(req); } static void @@ -583,7 +826,7 @@ nfs_clear_request_commit(struct nfs_page *req) } static inline -int nfs_write_need_commit(struct nfs_write_data *data) +int nfs_write_need_commit(struct nfs_pgio_data *data) { if (data->verf.committed == NFS_DATA_SYNC) return data->header->lseg == NULL; @@ -614,7 +857,7 @@ nfs_clear_request_commit(struct nfs_page *req) } static inline -int nfs_write_need_commit(struct nfs_write_data *data) +int nfs_write_need_commit(struct nfs_pgio_data *data) { return 0; } @@ -645,7 +888,7 @@ static void nfs_write_completion(struct nfs_pgio_header *hdr) goto next; } if (test_bit(NFS_IOHDR_NEED_COMMIT, &hdr->flags)) { - memcpy(&req->wb_verf, &hdr->verf->verifier, sizeof(req->wb_verf)); + memcpy(&req->wb_verf, &hdr->verf.verifier, sizeof(req->wb_verf)); nfs_mark_request_commit(req, hdr->lseg, &cinfo); goto next; } @@ -653,7 +896,7 @@ remove_req: nfs_inode_remove_request(req); next: nfs_unlock_request(req); - nfs_end_page_writeback(req->wb_page); + nfs_end_page_writeback(req); nfs_release_request(req); } out: @@ -661,7 +904,7 @@ out: } #if IS_ENABLED(CONFIG_NFS_V3) || IS_ENABLED(CONFIG_NFS_V4) -static unsigned long +unsigned long nfs_reqs_to_commit(struct nfs_commit_info *cinfo) { return cinfo->mds->ncommit; @@ -718,7 +961,7 @@ nfs_scan_commit(struct inode *inode, struct list_head *dst, } #else -static unsigned long nfs_reqs_to_commit(struct nfs_commit_info *cinfo) +unsigned long nfs_reqs_to_commit(struct nfs_commit_info *cinfo) { return 0; } @@ -754,10 +997,14 @@ static struct nfs_page *nfs_try_to_update_request(struct inode *inode, spin_lock(&inode->i_lock); for (;;) { - req = nfs_page_find_request_locked(NFS_I(inode), page); + req = nfs_page_find_head_request_locked(NFS_I(inode), page); if (req == NULL) goto out_unlock; + /* should be handled by nfs_flush_incompatible */ + WARN_ON_ONCE(req->wb_head != req); + WARN_ON_ONCE(req->wb_this_page != req); + rqend = req->wb_offset + req->wb_bytes; /* * Tell the caller to flush out the request if @@ -819,7 +1066,7 @@ static struct nfs_page * nfs_setup_write_request(struct nfs_open_context* ctx, req = nfs_try_to_update_request(inode, page, offset, bytes); if (req != NULL) goto out; - req = nfs_create_request(ctx, inode, page, offset, bytes); + req = nfs_create_request(ctx, page, NULL, offset, bytes); if (IS_ERR(req)) goto out; nfs_inode_add_request(inode, req); @@ -837,7 +1084,7 @@ static int nfs_writepage_setup(struct nfs_open_context *ctx, struct page *page, return PTR_ERR(req); /* Update file length */ nfs_grow_file(page, offset, count); - nfs_mark_uptodate(page, req->wb_pgbase, req->wb_bytes); + nfs_mark_uptodate(req); nfs_mark_request_dirty(req); nfs_unlock_and_release_request(req); return 0; @@ -858,11 +1105,13 @@ int nfs_flush_incompatible(struct file *file, struct page *page) * dropped page. */ do { - req = nfs_page_find_request(page); + req = nfs_page_find_head_request(page); if (req == NULL) return 0; l_ctx = req->wb_lock_context; do_flush = req->wb_page != page || req->wb_context != ctx; + /* for now, flush if more than 1 request in page_group */ + do_flush |= req->wb_this_page != req; if (l_ctx && ctx->dentry->d_inode->i_flock != NULL) { do_flush |= l_ctx->lockowner.l_owner != current->files || l_ctx->lockowner.l_pid != current->tgid; @@ -909,11 +1158,18 @@ bool nfs_ctx_key_to_expire(struct nfs_open_context *ctx) */ static bool nfs_write_pageuptodate(struct page *page, struct inode *inode) { + struct nfs_inode *nfsi = NFS_I(inode); + if (nfs_have_delegated_attributes(inode)) goto out; - if (NFS_I(inode)->cache_validity & (NFS_INO_INVALID_DATA|NFS_INO_REVAL_PAGECACHE)) + if (nfsi->cache_validity & NFS_INO_REVAL_PAGECACHE) + return false; + smp_rmb(); + if (test_bit(NFS_INO_INVALIDATING, &nfsi->flags)) return false; out: + if (nfsi->cache_validity & NFS_INO_INVALID_DATA) + return false; return PageUptodate(page) != 0; } @@ -922,19 +1178,20 @@ out: * extend the write to cover the entire page in order to avoid fragmentation * inefficiencies. * - * If the file is opened for synchronous writes or if we have a write delegation - * from the server then we can just skip the rest of the checks. + * If the file is opened for synchronous writes then we can just skip the rest + * of the checks. */ static int nfs_can_extend_write(struct file *file, struct page *page, struct inode *inode) { if (file->f_flags & O_DSYNC) return 0; + if (!nfs_write_pageuptodate(page, inode)) + return 0; if (NFS_PROTO(inode)->have_delegation(inode, FMODE_WRITE)) return 1; - if (nfs_write_pageuptodate(page, inode) && (inode->i_flock == NULL || - (inode->i_flock->fl_start == 0 && + if (inode->i_flock == NULL || (inode->i_flock->fl_start == 0 && inode->i_flock->fl_end == OFFSET_MAX && - inode->i_flock->fl_type != F_RDLCK))) + inode->i_flock->fl_type != F_RDLCK)) return 1; return 0; } @@ -984,126 +1241,17 @@ static int flush_task_priority(int how) return RPC_PRIORITY_NORMAL; } -int nfs_initiate_write(struct rpc_clnt *clnt, - struct nfs_write_data *data, - const struct rpc_call_ops *call_ops, - int how, int flags) +static void nfs_initiate_write(struct nfs_pgio_data *data, struct rpc_message *msg, + struct rpc_task_setup *task_setup_data, int how) { struct inode *inode = data->header->inode; int priority = flush_task_priority(how); - struct rpc_task *task; - struct rpc_message msg = { - .rpc_argp = &data->args, - .rpc_resp = &data->res, - .rpc_cred = data->header->cred, - }; - struct rpc_task_setup task_setup_data = { - .rpc_client = clnt, - .task = &data->task, - .rpc_message = &msg, - .callback_ops = call_ops, - .callback_data = data, - .workqueue = nfsiod_workqueue, - .flags = RPC_TASK_ASYNC | flags, - .priority = priority, - }; - int ret = 0; - - /* Set up the initial task struct. */ - NFS_PROTO(inode)->write_setup(data, &msg); - dprintk("NFS: %5u initiated write call " - "(req %s/%lld, %u bytes @ offset %llu)\n", - data->task.tk_pid, - inode->i_sb->s_id, - (long long)NFS_FILEID(inode), - data->args.count, - (unsigned long long)data->args.offset); + task_setup_data->priority = priority; + NFS_PROTO(inode)->write_setup(data, msg); nfs4_state_protect_write(NFS_SERVER(inode)->nfs_client, - &task_setup_data.rpc_client, &msg, data); - - task = rpc_run_task(&task_setup_data); - if (IS_ERR(task)) { - ret = PTR_ERR(task); - goto out; - } - if (how & FLUSH_SYNC) { - ret = rpc_wait_for_completion_task(task); - if (ret == 0) - ret = task->tk_status; - } - rpc_put_task(task); -out: - return ret; -} -EXPORT_SYMBOL_GPL(nfs_initiate_write); - -/* - * Set up the argument/result storage required for the RPC call. - */ -static void nfs_write_rpcsetup(struct nfs_write_data *data, - unsigned int count, unsigned int offset, - int how, struct nfs_commit_info *cinfo) -{ - struct nfs_page *req = data->header->req; - - /* Set up the RPC argument and reply structs - * NB: take care not to mess about with data->commit et al. */ - - data->args.fh = NFS_FH(data->header->inode); - data->args.offset = req_offset(req) + offset; - /* pnfs_set_layoutcommit needs this */ - data->mds_offset = data->args.offset; - data->args.pgbase = req->wb_pgbase + offset; - data->args.pages = data->pages.pagevec; - data->args.count = count; - data->args.context = get_nfs_open_context(req->wb_context); - data->args.lock_context = req->wb_lock_context; - data->args.stable = NFS_UNSTABLE; - switch (how & (FLUSH_STABLE | FLUSH_COND_STABLE)) { - case 0: - break; - case FLUSH_COND_STABLE: - if (nfs_reqs_to_commit(cinfo)) - break; - default: - data->args.stable = NFS_FILE_SYNC; - } - - data->res.fattr = &data->fattr; - data->res.count = count; - data->res.verf = &data->verf; - nfs_fattr_init(&data->fattr); -} - -static int nfs_do_write(struct nfs_write_data *data, - const struct rpc_call_ops *call_ops, - int how) -{ - struct inode *inode = data->header->inode; - - return nfs_initiate_write(NFS_CLIENT(inode), data, call_ops, how, 0); -} - -static int nfs_do_multiple_writes(struct list_head *head, - const struct rpc_call_ops *call_ops, - int how) -{ - struct nfs_write_data *data; - int ret = 0; - - while (!list_empty(head)) { - int ret2; - - data = list_first_entry(head, struct nfs_write_data, list); - list_del_init(&data->list); - - ret2 = nfs_do_write(data, call_ops, how); - if (ret == 0) - ret = ret2; - } - return ret; + &task_setup_data->rpc_client, msg, data); } /* If a nfs_flush_* function fails, it should remove reqs from @head and @@ -1114,7 +1262,7 @@ static void nfs_redirty_request(struct nfs_page *req) { nfs_mark_request_dirty(req); nfs_unlock_request(req); - nfs_end_page_writeback(req->wb_page); + nfs_end_page_writeback(req); nfs_release_request(req); } @@ -1134,173 +1282,30 @@ static const struct nfs_pgio_completion_ops nfs_async_write_completion_ops = { .completion = nfs_write_completion, }; -static void nfs_flush_error(struct nfs_pageio_descriptor *desc, - struct nfs_pgio_header *hdr) -{ - set_bit(NFS_IOHDR_REDO, &hdr->flags); - while (!list_empty(&hdr->rpc_list)) { - struct nfs_write_data *data = list_first_entry(&hdr->rpc_list, - struct nfs_write_data, list); - list_del(&data->list); - nfs_writedata_release(data); - } - desc->pg_completion_ops->error_cleanup(&desc->pg_list); -} - -/* - * Generate multiple small requests to write out a single - * contiguous dirty area on one page. - */ -static int nfs_flush_multi(struct nfs_pageio_descriptor *desc, - struct nfs_pgio_header *hdr) -{ - struct nfs_page *req = hdr->req; - struct page *page = req->wb_page; - struct nfs_write_data *data; - size_t wsize = desc->pg_bsize, nbytes; - unsigned int offset; - int requests = 0; - struct nfs_commit_info cinfo; - - nfs_init_cinfo(&cinfo, desc->pg_inode, desc->pg_dreq); - - if ((desc->pg_ioflags & FLUSH_COND_STABLE) && - (desc->pg_moreio || nfs_reqs_to_commit(&cinfo) || - desc->pg_count > wsize)) - desc->pg_ioflags &= ~FLUSH_COND_STABLE; - - - offset = 0; - nbytes = desc->pg_count; - do { - size_t len = min(nbytes, wsize); - - data = nfs_writedata_alloc(hdr, 1); - if (!data) { - nfs_flush_error(desc, hdr); - return -ENOMEM; - } - data->pages.pagevec[0] = page; - nfs_write_rpcsetup(data, len, offset, desc->pg_ioflags, &cinfo); - list_add(&data->list, &hdr->rpc_list); - requests++; - nbytes -= len; - offset += len; - } while (nbytes != 0); - nfs_list_remove_request(req); - nfs_list_add_request(req, &hdr->pages); - desc->pg_rpc_callops = &nfs_write_common_ops; - return 0; -} - -/* - * Create an RPC task for the given write request and kick it. - * The page must have been locked by the caller. - * - * It may happen that the page we're passed is not marked dirty. - * This is the case if nfs_updatepage detects a conflicting request - * that has been written but not committed. - */ -static int nfs_flush_one(struct nfs_pageio_descriptor *desc, - struct nfs_pgio_header *hdr) -{ - struct nfs_page *req; - struct page **pages; - struct nfs_write_data *data; - struct list_head *head = &desc->pg_list; - struct nfs_commit_info cinfo; - - data = nfs_writedata_alloc(hdr, nfs_page_array_len(desc->pg_base, - desc->pg_count)); - if (!data) { - nfs_flush_error(desc, hdr); - return -ENOMEM; - } - - nfs_init_cinfo(&cinfo, desc->pg_inode, desc->pg_dreq); - pages = data->pages.pagevec; - while (!list_empty(head)) { - req = nfs_list_entry(head->next); - nfs_list_remove_request(req); - nfs_list_add_request(req, &hdr->pages); - *pages++ = req->wb_page; - } - - if ((desc->pg_ioflags & FLUSH_COND_STABLE) && - (desc->pg_moreio || nfs_reqs_to_commit(&cinfo))) - desc->pg_ioflags &= ~FLUSH_COND_STABLE; - - /* Set up the argument struct */ - nfs_write_rpcsetup(data, desc->pg_count, 0, desc->pg_ioflags, &cinfo); - list_add(&data->list, &hdr->rpc_list); - desc->pg_rpc_callops = &nfs_write_common_ops; - return 0; -} - -int nfs_generic_flush(struct nfs_pageio_descriptor *desc, - struct nfs_pgio_header *hdr) -{ - if (desc->pg_bsize < PAGE_CACHE_SIZE) - return nfs_flush_multi(desc, hdr); - return nfs_flush_one(desc, hdr); -} -EXPORT_SYMBOL_GPL(nfs_generic_flush); - -static int nfs_generic_pg_writepages(struct nfs_pageio_descriptor *desc) -{ - struct nfs_write_header *whdr; - struct nfs_pgio_header *hdr; - int ret; - - whdr = nfs_writehdr_alloc(); - if (!whdr) { - desc->pg_completion_ops->error_cleanup(&desc->pg_list); - return -ENOMEM; - } - hdr = &whdr->header; - nfs_pgheader_init(desc, hdr, nfs_writehdr_free); - atomic_inc(&hdr->refcnt); - ret = nfs_generic_flush(desc, hdr); - if (ret == 0) - ret = nfs_do_multiple_writes(&hdr->rpc_list, - desc->pg_rpc_callops, - desc->pg_ioflags); - if (atomic_dec_and_test(&hdr->refcnt)) - hdr->completion_ops->completion(hdr); - return ret; -} - -static const struct nfs_pageio_ops nfs_pageio_write_ops = { - .pg_test = nfs_generic_pg_test, - .pg_doio = nfs_generic_pg_writepages, -}; - void nfs_pageio_init_write(struct nfs_pageio_descriptor *pgio, - struct inode *inode, int ioflags, + struct inode *inode, int ioflags, bool force_mds, const struct nfs_pgio_completion_ops *compl_ops) { - nfs_pageio_init(pgio, inode, &nfs_pageio_write_ops, compl_ops, - NFS_SERVER(inode)->wsize, ioflags); + struct nfs_server *server = NFS_SERVER(inode); + const struct nfs_pageio_ops *pg_ops = &nfs_pgio_rw_ops; + +#ifdef CONFIG_NFS_V4_1 + if (server->pnfs_curr_ld && !force_mds) + pg_ops = server->pnfs_curr_ld->pg_write_ops; +#endif + nfs_pageio_init(pgio, inode, pg_ops, compl_ops, &nfs_rw_write_ops, + server->wsize, ioflags); } EXPORT_SYMBOL_GPL(nfs_pageio_init_write); void nfs_pageio_reset_write_mds(struct nfs_pageio_descriptor *pgio) { - pgio->pg_ops = &nfs_pageio_write_ops; + pgio->pg_ops = &nfs_pgio_rw_ops; pgio->pg_bsize = NFS_SERVER(pgio->pg_inode)->wsize; } EXPORT_SYMBOL_GPL(nfs_pageio_reset_write_mds); -void nfs_write_prepare(struct rpc_task *task, void *calldata) -{ - struct nfs_write_data *data = calldata; - int err; - err = NFS_PROTO(data->header->inode)->write_rpc_prepare(task, data); - if (err) - rpc_exit(task, err); -} - void nfs_commit_prepare(struct rpc_task *task, void *calldata) { struct nfs_commit_data *data = calldata; @@ -1308,23 +1313,8 @@ void nfs_commit_prepare(struct rpc_task *task, void *calldata) NFS_PROTO(data->inode)->commit_rpc_prepare(task, data); } -/* - * Handle a write reply that flushes a whole page. - * - * FIXME: There is an inherent race with invalidate_inode_pages and - * writebacks since the page->count is kept > 1 for as long - * as the page has a write request pending. - */ -static void nfs_writeback_done_common(struct rpc_task *task, void *calldata) -{ - struct nfs_write_data *data = calldata; - - nfs_writeback_done(task, data); -} - -static void nfs_writeback_release_common(void *calldata) +static void nfs_writeback_release_common(struct nfs_pgio_data *data) { - struct nfs_write_data *data = calldata; struct nfs_pgio_header *hdr = data->header; int status = data->task.tk_status; @@ -1333,34 +1323,46 @@ static void nfs_writeback_release_common(void *calldata) if (test_bit(NFS_IOHDR_NEED_RESCHED, &hdr->flags)) ; /* Do nothing */ else if (!test_and_set_bit(NFS_IOHDR_NEED_COMMIT, &hdr->flags)) - memcpy(hdr->verf, &data->verf, sizeof(*hdr->verf)); - else if (memcmp(hdr->verf, &data->verf, sizeof(*hdr->verf))) + memcpy(&hdr->verf, &data->verf, sizeof(hdr->verf)); + else if (memcmp(&hdr->verf, &data->verf, sizeof(hdr->verf))) set_bit(NFS_IOHDR_NEED_RESCHED, &hdr->flags); spin_unlock(&hdr->lock); } - nfs_writedata_release(data); } -static const struct rpc_call_ops nfs_write_common_ops = { - .rpc_call_prepare = nfs_write_prepare, - .rpc_call_done = nfs_writeback_done_common, - .rpc_release = nfs_writeback_release_common, -}; +/* + * Special version of should_remove_suid() that ignores capabilities. + */ +static int nfs_should_remove_suid(const struct inode *inode) +{ + umode_t mode = inode->i_mode; + int kill = 0; + /* suid always must be killed */ + if (unlikely(mode & S_ISUID)) + kill = ATTR_KILL_SUID; + + /* + * sgid without any exec bits is just a mandatory locking mark; leave + * it alone. If some exec bits are set, it's a real sgid; kill it. + */ + if (unlikely((mode & S_ISGID) && (mode & S_IXGRP))) + kill |= ATTR_KILL_SGID; + + if (unlikely(kill && S_ISREG(mode))) + return kill; + + return 0; +} /* * This function is called when the WRITE call is complete. */ -void nfs_writeback_done(struct rpc_task *task, struct nfs_write_data *data) +static int nfs_writeback_done(struct rpc_task *task, struct nfs_pgio_data *data, + struct inode *inode) { - struct nfs_writeargs *argp = &data->args; - struct nfs_writeres *resp = &data->res; - struct inode *inode = data->header->inode; int status; - dprintk("NFS: %5u nfs_writeback_done (status %d)\n", - task->tk_pid, task->tk_status); - /* * ->write_done will attempt to use post-op attributes to detect * conflicting writes by other clients. A strict interpretation @@ -1370,11 +1372,11 @@ void nfs_writeback_done(struct rpc_task *task, struct nfs_write_data *data) */ status = NFS_PROTO(inode)->write_done(task, data); if (status != 0) - return; - nfs_add_stats(inode, NFSIOS_SERVERWRITTENBYTES, resp->count); + return status; + nfs_add_stats(inode, NFSIOS_SERVERWRITTENBYTES, data->res.count); #if IS_ENABLED(CONFIG_NFS_V3) || IS_ENABLED(CONFIG_NFS_V4) - if (resp->verf->committed < argp->stable && task->tk_status >= 0) { + if (data->res.verf->committed < data->args.stable && task->tk_status >= 0) { /* We tried a write call, but the server did not * commit data to stable storage even though we * requested it. @@ -1390,18 +1392,31 @@ void nfs_writeback_done(struct rpc_task *task, struct nfs_write_data *data) dprintk("NFS: faulty NFS server %s:" " (committed = %d) != (stable = %d)\n", NFS_SERVER(inode)->nfs_client->cl_hostname, - resp->verf->committed, argp->stable); + data->res.verf->committed, data->args.stable); complain = jiffies + 300 * HZ; } } #endif - if (task->tk_status < 0) - nfs_set_pgio_error(data->header, task->tk_status, argp->offset); - else if (resp->count < argp->count) { + + /* Deal with the suid/sgid bit corner case */ + if (nfs_should_remove_suid(inode)) + nfs_mark_for_revalidate(inode); + return 0; +} + +/* + * This function is called when the WRITE call is complete. + */ +static void nfs_writeback_result(struct rpc_task *task, struct nfs_pgio_data *data) +{ + struct nfs_pgio_args *argp = &data->args; + struct nfs_pgio_res *resp = &data->res; + + if (resp->count < argp->count) { static unsigned long complain; /* This a short write! */ - nfs_inc_stats(inode, NFSIOS_SHORTWRITE); + nfs_inc_stats(data->header->inode, NFSIOS_SHORTWRITE); /* Has the server at least made some progress? */ if (resp->count == 0) { @@ -1452,7 +1467,7 @@ static int nfs_commit_set_lock(struct nfs_inode *nfsi, int may_wait) static void nfs_commit_clear_lock(struct nfs_inode *nfsi) { clear_bit(NFS_INO_COMMIT, &nfsi->flags); - smp_mb__after_clear_bit(); + smp_mb__after_atomic(); wake_up_bit(&nfsi->flags, NFS_INO_COMMIT); } @@ -1606,9 +1621,9 @@ static void nfs_commit_release_pages(struct nfs_commit_data *data) nfs_list_remove_request(req); nfs_clear_page_commit(req->wb_page); - dprintk("NFS: commit (%s/%lld %d@%lld)", + dprintk("NFS: commit (%s/%llu %d@%lld)", req->wb_context->dentry->d_sb->s_id, - (long long)NFS_FILEID(req->wb_context->dentry->d_inode), + (unsigned long long)NFS_FILEID(req->wb_context->dentry->d_inode), req->wb_bytes, (long long)req_offset(req)); if (status < 0) { @@ -1782,27 +1797,28 @@ int nfs_wb_page_cancel(struct inode *inode, struct page *page) struct nfs_page *req; int ret = 0; - for (;;) { - wait_on_page_writeback(page); - req = nfs_page_find_request(page); - if (req == NULL) - break; - if (nfs_lock_request(req)) { - nfs_clear_request_commit(req); - nfs_inode_remove_request(req); - /* - * In case nfs_inode_remove_request has marked the - * page as being dirty - */ - cancel_dirty_page(page, PAGE_CACHE_SIZE); - nfs_unlock_and_release_request(req); - break; - } - ret = nfs_wait_on_request(req); - nfs_release_request(req); - if (ret < 0) - break; + wait_on_page_writeback(page); + + /* blocking call to cancel all requests and join to a single (head) + * request */ + req = nfs_lock_and_join_requests(page, false); + + if (IS_ERR(req)) { + ret = PTR_ERR(req); + } else if (req) { + /* all requests from this page have been cancelled by + * nfs_lock_and_join_requests, so just remove the head + * request from the inode / page_private pointer and + * release it */ + nfs_inode_remove_request(req); + /* + * In case nfs_inode_remove_request has marked the + * page as being dirty + */ + cancel_dirty_page(page, PAGE_CACHE_SIZE); + nfs_unlock_and_release_request(req); } + return ret; } @@ -1868,7 +1884,7 @@ int nfs_migrate_page(struct address_space *mapping, struct page *newpage, int __init nfs_init_writepagecache(void) { nfs_wdata_cachep = kmem_cache_create("nfs_write_data", - sizeof(struct nfs_write_header), + sizeof(struct nfs_rw_header), 0, SLAB_HWCACHE_ALIGN, NULL); if (nfs_wdata_cachep == NULL) @@ -1930,3 +1946,12 @@ void nfs_destroy_writepagecache(void) kmem_cache_destroy(nfs_wdata_cachep); } +static const struct nfs_rw_ops nfs_rw_write_ops = { + .rw_mode = FMODE_WRITE, + .rw_alloc_header = nfs_writehdr_alloc, + .rw_free_header = nfs_writehdr_free, + .rw_release = nfs_writeback_release_common, + .rw_done = nfs_writeback_done, + .rw_result = nfs_writeback_result, + .rw_initiate = nfs_initiate_write, +}; |