diff options
Diffstat (limited to 'fs/nfs/write.c')
| -rw-r--r-- | fs/nfs/write.c | 1555 |
1 files changed, 780 insertions, 775 deletions
diff --git a/fs/nfs/write.c b/fs/nfs/write.c index b1af5dee5e0a..336c510f3750 100644 --- a/fs/nfs/write.c +++ b/fs/nfs/write.c @@ -1,3 +1,4 @@ +// SPDX-License-Identifier: GPL-2.0-only /* * linux/fs/nfs/write.c * @@ -23,8 +24,11 @@ #include <linux/export.h> #include <linux/freezer.h> #include <linux/wait.h> +#include <linux/iversion.h> +#include <linux/filelock.h> #include <linux/uaccess.h> +#include <linux/sched/mm.h> #include "delegation.h" #include "internal.h" @@ -54,39 +58,28 @@ 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 void nfs_inode_remove_request(struct nfs_page *req); +static void nfs_clear_request_commit(struct nfs_commit_info *cinfo, + struct nfs_page *req); static void nfs_init_cinfo_from_inode(struct nfs_commit_info *cinfo, struct inode *inode); -static struct nfs_page * -nfs_page_search_commits_for_head_request_locked(struct nfs_inode *nfsi, - struct page *page); static struct kmem_cache *nfs_wdata_cachep; static mempool_t *nfs_wdata_mempool; static struct kmem_cache *nfs_cdata_cachep; static mempool_t *nfs_commit_mempool; -struct nfs_commit_data *nfs_commitdata_alloc(bool never_fail) +struct nfs_commit_data *nfs_commitdata_alloc(void) { struct nfs_commit_data *p; - if (never_fail) - p = mempool_alloc(nfs_commit_mempool, GFP_NOIO); - else { - /* It is OK to do some reclaim, not no safe to wait - * for anything to be returned to the pool. - * mempool_alloc() cannot handle that particular combination, - * so we need two separate attempts. - */ + p = kmem_cache_zalloc(nfs_cdata_cachep, nfs_io_gfp_mask()); + if (!p) { p = mempool_alloc(nfs_commit_mempool, GFP_NOWAIT); if (!p) - p = kmem_cache_alloc(nfs_cdata_cachep, GFP_NOIO | - __GFP_NOWARN | __GFP_NORETRY); - if (!p) return NULL; + memset(p, 0, sizeof(*p)); } - - memset(p, 0, sizeof(*p)); INIT_LIST_HEAD(&p->pages); return p; } @@ -100,12 +93,16 @@ EXPORT_SYMBOL_GPL(nfs_commit_free); static struct nfs_pgio_header *nfs_writehdr_alloc(void) { - struct nfs_pgio_header *p = mempool_alloc(nfs_wdata_mempool, GFP_NOIO); + struct nfs_pgio_header *p; - if (p) { + p = kmem_cache_zalloc(nfs_wdata_cachep, nfs_io_gfp_mask()); + if (!p) { + p = mempool_alloc(nfs_wdata_mempool, GFP_NOWAIT); + if (!p) + return NULL; memset(p, 0, sizeof(*p)); - p->rw_mode = FMODE_WRITE; } + p->rw_mode = FMODE_WRITE; return p; } @@ -147,142 +144,110 @@ static void nfs_io_completion_put(struct nfs_io_completion *ioc) kref_put(&ioc->refcount, nfs_io_completion_release); } -static void nfs_context_set_write_error(struct nfs_open_context *ctx, int error) +static void +nfs_page_set_inode_ref(struct nfs_page *req, struct inode *inode) +{ + if (!test_and_set_bit(PG_INODE_REF, &req->wb_flags)) { + kref_get(&req->wb_kref); + atomic_long_inc(&NFS_I(inode)->nrequests); + } +} + +static void nfs_cancel_remove_inode(struct nfs_page *req, struct inode *inode) { - ctx->error = error; - smp_wmb(); - set_bit(NFS_CONTEXT_ERROR_WRITE, &ctx->flags); + if (test_and_clear_bit(PG_REMOVE, &req->wb_flags)) + nfs_page_set_inode_ref(req, inode); } -/* - * nfs_page_find_head_request_locked - find head request associated with @page +/** + * nfs_folio_find_head_request - find head request associated with a folio + * @folio: pointer to folio * * 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_head_request_locked(struct nfs_inode *nfsi, struct page *page) +static struct nfs_page *nfs_folio_find_head_request(struct folio *folio) { - struct nfs_page *req = NULL; - - if (PagePrivate(page)) - req = (struct nfs_page *)page_private(page); - else if (unlikely(PageSwapCache(page))) - req = nfs_page_search_commits_for_head_request_locked(nfsi, - page); + struct address_space *mapping = folio->mapping; + struct nfs_page *req; + if (!folio_test_private(folio)) + return NULL; + spin_lock(&mapping->i_private_lock); + req = folio->private; if (req) { WARN_ON_ONCE(req->wb_head != req); kref_get(&req->wb_kref); } - - return req; -} - -/* - * 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_head_request_locked(NFS_I(inode), page); - spin_unlock(&inode->i_lock); + spin_unlock(&mapping->i_private_lock); return req; } /* Adjust the file length if we're writing beyond the end */ -static void nfs_grow_file(struct page *page, unsigned int offset, unsigned int count) +static void nfs_grow_file(struct folio *folio, unsigned int offset, + unsigned int count) { - struct inode *inode = page_file_mapping(page)->host; + struct inode *inode = folio->mapping->host; loff_t end, i_size; pgoff_t end_index; spin_lock(&inode->i_lock); i_size = i_size_read(inode); - end_index = (i_size - 1) >> PAGE_SHIFT; - if (i_size > 0 && page_index(page) < end_index) + end_index = ((i_size - 1) >> folio_shift(folio)) << folio_order(folio); + if (i_size > 0 && folio->index < end_index) goto out; - end = page_file_offset(page) + ((loff_t)offset+count); + end = folio_pos(folio) + (loff_t)offset + (loff_t)count; if (i_size >= end) goto out; + trace_nfs_size_grow(inode, end); i_size_write(inode, end); + NFS_I(inode)->cache_validity &= ~NFS_INO_INVALID_SIZE; nfs_inc_stats(inode, NFSIOS_EXTENDWRITE); out: + /* Atomically update timestamps if they are delegated to us. */ + nfs_update_delegated_mtime_locked(inode); spin_unlock(&inode->i_lock); + nfs_fscache_invalidate(inode, 0); } /* A writeback failed: mark the page as bad, and invalidate the page cache */ -static void nfs_set_pageerror(struct page *page) +static void nfs_set_pageerror(struct address_space *mapping) { - nfs_zap_mapping(page_file_mapping(page)->host, page_file_mapping(page)); + struct inode *inode = mapping->host; + + nfs_zap_mapping(mapping->host, mapping); + /* Force file size revalidation */ + spin_lock(&inode->i_lock); + nfs_set_cache_invalid(inode, NFS_INO_REVAL_FORCED | + NFS_INO_INVALID_CHANGE | + NFS_INO_INVALID_SIZE); + spin_unlock(&inode->i_lock); } -/* - * 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) +static void nfs_mapping_set_error(struct folio *folio, int error) { - 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; + struct address_space *mapping = folio->mapping; - req = req->wb_this_page; - } while (req != head); - - return NULL; + filemap_set_wb_err(mapping, error); + if (mapping->host) + errseq_set(&mapping->host->i_sb->s_wb_err, + error == -ENOSPC ? -ENOSPC : -EIO); + nfs_set_pageerror(mapping); } /* - * nfs_page_group_covers_page - * @head - head request of page group + * nfs_page_covers_folio + * @req: struct nfs_page * - * Return true if the page group with head @head covers the whole page, - * returns false otherwise + * Return true if the request covers the whole folio. + * Note that the caller should ensure all subrequests have been joined */ 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); + unsigned int len = nfs_folio_length(nfs_page_to_folio(req)); - nfs_page_group_lock(req, false); - - 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; + return req->wb_pgbase == 0 && req->wb_bytes == len; } /* We can set the PG_uptodate flag if we see that a write request @@ -290,11 +255,13 @@ static bool nfs_page_group_covers_page(struct nfs_page *req) */ static void nfs_mark_uptodate(struct nfs_page *req) { - if (PageUptodate(req->wb_page)) + struct folio *folio = nfs_page_to_folio(req); + + if (folio_test_uptodate(folio)) return; if (!nfs_page_group_covers_page(req)) return; - SetPageUptodate(req->wb_page); + folio_mark_uptodate(folio); } static int wb_priority(struct writeback_control *wbc) @@ -316,93 +283,34 @@ int nfs_congestion_kb; #define NFS_CONGESTION_OFF_THRESH \ (NFS_CONGESTION_ON_THRESH - (NFS_CONGESTION_ON_THRESH >> 2)) -static void nfs_set_page_writeback(struct page *page) +static void nfs_folio_set_writeback(struct folio *folio) { - struct inode *inode = page_file_mapping(page)->host; - struct nfs_server *nfss = NFS_SERVER(inode); - int ret = test_set_page_writeback(page); + struct nfs_server *nfss = NFS_SERVER(folio->mapping->host); - WARN_ON_ONCE(ret != 0); - - if (atomic_long_inc_return(&nfss->writeback) > - NFS_CONGESTION_ON_THRESH) - set_bdi_congested(inode_to_bdi(inode), BLK_RW_ASYNC); + folio_start_writeback(folio); + if (atomic_long_inc_return(&nfss->writeback) > NFS_CONGESTION_ON_THRESH) + nfss->write_congested = 1; } -static void nfs_end_page_writeback(struct nfs_page *req) +static void nfs_folio_end_writeback(struct folio *folio) { - struct inode *inode = page_file_mapping(req->wb_page)->host; - struct nfs_server *nfss = NFS_SERVER(inode); - - 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(inode_to_bdi(inode), BLK_RW_ASYNC); -} - + struct nfs_server *nfss = NFS_SERVER(folio->mapping->host); -/* 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); + folio_end_writeback_no_dropbehind(folio); + if (atomic_long_dec_return(&nfss->writeback) < + NFS_CONGESTION_OFF_THRESH) { + nfss->write_congested = 0; + wake_up_all(&nfss->write_congestion_wait); + } } - -/* - * 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) +static void nfs_page_end_writeback(struct nfs_page *req) { - 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; + if (nfs_page_group_sync_on_bit(req, PG_WB_END)) { + nfs_unlock_request(req); + nfs_folio_end_writeback(nfs_page_to_folio(req)); + } else + nfs_unlock_request(req); } /* @@ -417,7 +325,8 @@ nfs_unroll_locks_and_wait(struct inode *inode, struct nfs_page *head, */ static void nfs_destroy_unlinked_subrequests(struct nfs_page *destroy_list, - struct nfs_page *old_head) + struct nfs_page *old_head, + struct inode *inode) { while (destroy_list) { struct nfs_page *subreq = destroy_list; @@ -425,140 +334,84 @@ nfs_destroy_unlinked_subrequests(struct nfs_page *destroy_list, destroy_list = (subreq->wb_this_page == old_head) ? NULL : subreq->wb_this_page; + /* Note: lock subreq in order to change subreq->wb_head */ + nfs_page_set_headlock(subreq); WARN_ON_ONCE(old_head != subreq->wb_head); /* make sure old group is not used */ - subreq->wb_head = subreq; subreq->wb_this_page = subreq; + subreq->wb_head = subreq; - /* subreq is now totally disconnected from page group or any - * write / commit lists. last chance to wake any waiters */ - nfs_unlock_request(subreq); + clear_bit(PG_REMOVE, &subreq->wb_flags); - if (!test_bit(PG_TEARDOWN, &subreq->wb_flags)) { - /* release ref on old head request */ - nfs_release_request(old_head); + /* Note: races with nfs_page_group_destroy() */ + if (!kref_read(&subreq->wb_kref)) { + /* Check if we raced with nfs_page_group_destroy() */ + if (test_and_clear_bit(PG_TEARDOWN, &subreq->wb_flags)) { + nfs_page_clear_headlock(subreq); + nfs_free_request(subreq); + } else + nfs_page_clear_headlock(subreq); + continue; + } + nfs_page_clear_headlock(subreq); - nfs_page_group_clear_bits(subreq); + nfs_release_request(old_head); - /* 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); + if (test_and_clear_bit(PG_INODE_REF, &subreq->wb_flags)) { + nfs_release_request(subreq); + atomic_long_dec(&NFS_I(inode)->nrequests); } + + /* subreq is now totally disconnected from page group or any + * write / commit lists. last chance to wake any waiters */ + nfs_unlock_and_release_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 + * nfs_join_page_group - destroy subrequests of the head req + * @head: the page used to lookup the "page group" of nfs_page structures + * @inode: Inode to which the request belongs. * * 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) +void nfs_join_page_group(struct nfs_page *head, struct nfs_commit_info *cinfo, + struct inode *inode) { - struct inode *inode = page_file_mapping(page)->host; - struct nfs_page *head, *subreq; + struct nfs_page *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); - - /* - * 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); - return NULL; - } - - /* holding inode lock, so always make a non-blocking call to try the - * page group lock */ - ret = nfs_page_group_lock(head, true); - if (ret < 0) { - spin_unlock(&inode->i_lock); - - if (!nonblock && ret == -EAGAIN) { - nfs_page_group_lock_wait(head); - nfs_release_request(head); - goto try_again; + unsigned int pgbase, off, bytes; + + pgbase = head->wb_pgbase; + bytes = head->wb_bytes; + off = head->wb_offset; + for (subreq = head->wb_this_page; subreq != head; + subreq = subreq->wb_this_page) { + /* Subrequests should always form a contiguous range */ + if (pgbase > subreq->wb_pgbase) { + off -= pgbase - subreq->wb_pgbase; + bytes += pgbase - subreq->wb_pgbase; + pgbase = subreq->wb_pgbase; } - - nfs_release_request(head); - return ERR_PTR(ret); + bytes = max(subreq->wb_pgbase + subreq->wb_bytes + - pgbase, bytes); } - /* lock each request in the page group */ - subreq = head; - do { - /* - * Subrequests are always contiguous, non overlapping - * and in order - but may be repeated (mirrored writes). - */ - if (subreq->wb_offset == (head->wb_offset + total_bytes)) { - /* keep track of how many bytes this group covers */ - total_bytes += subreq->wb_bytes; - } else if (WARN_ON_ONCE(subreq->wb_offset < head->wb_offset || - ((subreq->wb_offset + subreq->wb_bytes) > - (head->wb_offset + total_bytes)))) { - nfs_page_group_unlock(head); - spin_unlock(&inode->i_lock); - return ERR_PTR(-EIO); - } - - 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); - } - - subreq = subreq->wb_this_page; - } while (subreq != head); + /* Set the head request's range to cover the former page group */ + head->wb_pgbase = pgbase; + head->wb_bytes = bytes; + head->wb_offset = off; /* 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_clear_request_commit(subreq); + nfs_clear_request_commit(cinfo, subreq); subreq = subreq->wb_this_page; } while (subreq != head); @@ -567,81 +420,186 @@ try_again: /* destroy list will be terminated by head */ destroy_list = head->wb_this_page; head->wb_this_page = head; + } + + nfs_destroy_unlinked_subrequests(destroy_list, head, inode); +} + +/** + * nfs_wait_on_request - Wait for a request to complete. + * @req: request to wait upon. + * + * Interruptible by fatal signals only. + * The user is responsible for holding a count on the request. + */ +static int nfs_wait_on_request(struct nfs_page *req) +{ + if (!test_bit(PG_BUSY, &req->wb_flags)) + return 0; + set_bit(PG_CONTENDED2, &req->wb_flags); + smp_mb__after_atomic(); + return wait_on_bit_io(&req->wb_flags, PG_BUSY, + TASK_UNINTERRUPTIBLE); +} + +/* + * nfs_unroll_locks - unlock all newly locked reqs and wait on @req + * @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 + * + * This is a helper function for nfs_lock_and_join_requests + * returns 0 on success, < 0 on error. + */ +static void +nfs_unroll_locks(struct nfs_page *head, struct nfs_page *req) +{ + struct nfs_page *tmp; - /* change head request to cover whole range that - * the former page group covered */ - head->wb_bytes = total_bytes; + /* relinquish all the locks successfully grabbed this run */ + for (tmp = head->wb_this_page ; tmp != req; tmp = tmp->wb_this_page) { + if (!kref_read(&tmp->wb_kref)) + continue; + nfs_unlock_and_release_request(tmp); } +} - /* - * prepare head request to be added to new pgio descriptor - */ - nfs_page_group_clear_bits(head); +/* + * nfs_page_group_lock_subreq - try to lock a subrequest + * @head: head request of page group + * @subreq: request to lock + * + * This is a helper function for nfs_lock_and_join_requests which + * must be called with the head request and page group both locked. + * On error, it returns with the page group unlocked. + */ +static int +nfs_page_group_lock_subreq(struct nfs_page *head, struct nfs_page *subreq) +{ + int ret; + + if (!kref_get_unless_zero(&subreq->wb_kref)) + return 0; + while (!nfs_lock_request(subreq)) { + nfs_page_group_unlock(head); + ret = nfs_wait_on_request(subreq); + if (!ret) + ret = nfs_page_group_lock(head); + if (ret < 0) { + nfs_unroll_locks(head, subreq); + nfs_release_request(subreq); + return ret; + } + } + return 0; +} + +/* + * nfs_lock_and_join_requests - join all subreqs to the head req + * @folio: the folio used to lookup the "page group" of nfs_page structures + * + * 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 @folio, or a ERR_PTR if an + * error was encountered. + */ +static struct nfs_page *nfs_lock_and_join_requests(struct folio *folio) +{ + struct inode *inode = folio->mapping->host; + struct nfs_page *head, *subreq; + struct nfs_commit_info cinfo; + int ret; /* - * 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. + * 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. */ - if (!test_and_set_bit(PG_INODE_REF, &head->wb_flags)) - kref_get(&head->wb_kref); +retry: + head = nfs_folio_find_head_request(folio); + if (!head) + return NULL; - nfs_page_group_unlock(head); + while (!nfs_lock_request(head)) { + ret = nfs_wait_on_request(head); + if (ret < 0) { + nfs_release_request(head); + return ERR_PTR(ret); + } + } - /* drop lock to clean uprequests on destroy list */ - spin_unlock(&inode->i_lock); + ret = nfs_page_group_lock(head); + if (ret < 0) + goto out_unlock; + + /* Ensure that nobody removed the request before we locked it */ + if (head != folio->private) { + nfs_page_group_unlock(head); + nfs_unlock_and_release_request(head); + goto retry; + } - nfs_destroy_unlinked_subrequests(destroy_list, head); + nfs_cancel_remove_inode(head, inode); - /* still holds ref on head from nfs_page_find_head_request_locked - * and still has lock on head from lock loop */ + /* lock each request in the page group */ + for (subreq = head->wb_this_page; + subreq != head; + subreq = subreq->wb_this_page) { + ret = nfs_page_group_lock_subreq(head, subreq); + if (ret < 0) + goto out_unlock; + } + + nfs_page_group_unlock(head); + + nfs_init_cinfo_from_inode(&cinfo, inode); + nfs_join_page_group(head, &cinfo, inode); return head; -} -static void nfs_write_error_remove_page(struct nfs_page *req) -{ - nfs_unlock_request(req); - nfs_end_page_writeback(req); - generic_error_remove_page(page_file_mapping(req->wb_page), - req->wb_page); - nfs_release_request(req); +out_unlock: + nfs_unlock_and_release_request(head); + return ERR_PTR(ret); } -static bool -nfs_error_is_fatal_on_server(int err) +static void nfs_write_error(struct nfs_page *req, int error) { - switch (err) { - case 0: - case -ERESTARTSYS: - case -EINTR: - return false; - } - return nfs_error_is_fatal(err); + trace_nfs_write_error(nfs_page_to_inode(req), req, error); + nfs_mapping_set_error(nfs_page_to_folio(req), error); + nfs_inode_remove_request(req); + nfs_page_end_writeback(req); + nfs_release_request(req); } /* * Find an associated nfs write request, and prepare to flush it out * May return an error if the user signalled nfs_wait_on_request(). */ -static int nfs_page_async_flush(struct nfs_pageio_descriptor *pgio, - struct page *page, bool nonblock) +static int nfs_do_writepage(struct folio *folio, struct writeback_control *wbc, + struct nfs_pageio_descriptor *pgio) { struct nfs_page *req; - int ret = 0; + int ret; + + nfs_pageio_cond_complete(pgio, folio->index); - req = nfs_lock_and_join_requests(page, nonblock); + req = nfs_lock_and_join_requests(folio); if (!req) - goto out; - ret = PTR_ERR(req); + return 0; if (IS_ERR(req)) - goto out; + return PTR_ERR(req); - nfs_set_page_writeback(page); + trace_nfs_do_writepage(req); + nfs_folio_set_writeback(folio); WARN_ON_ONCE(test_bit(PG_CLEAN, &req->wb_flags)); - ret = 0; /* If there is a fatal error that covers this write, just exit */ - if (nfs_error_is_fatal_on_server(req->wb_context->error)) + ret = pgio->pg_error; + if (nfs_error_is_fatal_on_server(ret)) goto out_launder; if (!nfs_pageio_add_request(pgio, req)) { @@ -649,75 +607,39 @@ static int nfs_page_async_flush(struct nfs_pageio_descriptor *pgio, /* * Remove the problematic req upon fatal errors on the server */ - if (nfs_error_is_fatal(ret)) { - nfs_context_set_write_error(req->wb_context, ret); - if (nfs_error_is_fatal_on_server(ret)) - goto out_launder; - } + if (nfs_error_is_fatal_on_server(ret)) + goto out_launder; + folio_redirty_for_writepage(wbc, folio); nfs_redirty_request(req); - ret = -EAGAIN; - } else - nfs_add_stats(page_file_mapping(page)->host, - NFSIOS_WRITEPAGES, 1); -out: - return ret; -out_launder: - nfs_write_error_remove_page(req); - return ret; -} + pgio->pg_error = 0; + return ret; + } -static int nfs_do_writepage(struct page *page, struct writeback_control *wbc, - struct nfs_pageio_descriptor *pgio) -{ - int ret; + nfs_add_stats(folio->mapping->host, NFSIOS_WRITEPAGES, 1); + return 0; - nfs_pageio_cond_complete(pgio, page_index(page)); - ret = nfs_page_async_flush(pgio, page, wbc->sync_mode == WB_SYNC_NONE); - if (ret == -EAGAIN) { - redirty_page_for_writepage(wbc, page); - ret = 0; - } - return ret; +out_launder: + nfs_write_error(req, ret); + return 0; } /* * Write an mmapped page to the server. */ -static int nfs_writepage_locked(struct page *page, +static int nfs_writepage_locked(struct folio *folio, struct writeback_control *wbc) { struct nfs_pageio_descriptor pgio; - struct inode *inode = page_file_mapping(page)->host; + struct inode *inode = folio->mapping->host; int err; nfs_inc_stats(inode, NFSIOS_VFSWRITEPAGE); - nfs_pageio_init_write(&pgio, inode, 0, - false, &nfs_async_write_completion_ops); - err = nfs_do_writepage(page, wbc, &pgio); + nfs_pageio_init_write(&pgio, inode, 0, false, + &nfs_async_write_completion_ops); + err = nfs_do_writepage(folio, wbc, &pgio); + pgio.pg_error = 0; nfs_pageio_complete(&pgio); - if (err < 0) - return err; - if (pgio.pg_error < 0) - return pgio.pg_error; - return 0; -} - -int nfs_writepage(struct page *page, struct writeback_control *wbc) -{ - int ret; - - ret = nfs_writepage_locked(page, wbc); - unlock_page(page); - return ret; -} - -static int nfs_writepages_callback(struct page *page, struct writeback_control *wbc, void *data) -{ - int ret; - - ret = nfs_do_writepage(page, wbc, data); - unlock_page(page); - return ret; + return err; } static void nfs_io_completion_commit(void *inode) @@ -729,64 +651,82 @@ int nfs_writepages(struct address_space *mapping, struct writeback_control *wbc) { struct inode *inode = mapping->host; struct nfs_pageio_descriptor pgio; - struct nfs_io_completion *ioc = nfs_io_completion_alloc(GFP_NOFS); + struct nfs_io_completion *ioc = NULL; + unsigned int mntflags = NFS_SERVER(inode)->flags; + struct nfs_server *nfss = NFS_SERVER(inode); + int priority = 0; int err; + trace_nfs_writepages(inode, wbc->range_start, wbc->range_end - wbc->range_start); + + /* Wait with writeback until write congestion eases */ + if (wbc->sync_mode == WB_SYNC_NONE && nfss->write_congested) { + err = wait_event_killable(nfss->write_congestion_wait, + nfss->write_congested == 0); + if (err) + goto out_err; + } + nfs_inc_stats(inode, NFSIOS_VFSWRITEPAGES); - if (ioc) - nfs_io_completion_init(ioc, nfs_io_completion_commit, inode); + if (!(mntflags & NFS_MOUNT_WRITE_EAGER) || wbc->for_kupdate || + wbc->for_background || wbc->for_sync) { + ioc = nfs_io_completion_alloc(GFP_KERNEL); + if (ioc) + nfs_io_completion_init(ioc, nfs_io_completion_commit, + inode); + priority = wb_priority(wbc); + } - nfs_pageio_init_write(&pgio, inode, wb_priority(wbc), false, - &nfs_async_write_completion_ops); - pgio.pg_io_completion = ioc; - err = write_cache_pages(mapping, wbc, nfs_writepages_callback, &pgio); - nfs_pageio_complete(&pgio); + do { + struct folio *folio = NULL; + + nfs_pageio_init_write(&pgio, inode, priority, false, + &nfs_async_write_completion_ops); + pgio.pg_io_completion = ioc; + while ((folio = writeback_iter(mapping, wbc, folio, &err))) { + err = nfs_do_writepage(folio, wbc, &pgio); + folio_unlock(folio); + } + pgio.pg_error = 0; + nfs_pageio_complete(&pgio); + if (err == -EAGAIN && mntflags & NFS_MOUNT_SOFTERR) + break; + } while (err < 0 && !nfs_error_is_fatal(err)); nfs_io_completion_put(ioc); - if (err < 0) - goto out_err; - err = pgio.pg_error; - if (err < 0) - goto out_err; - return 0; + if (err > 0) + err = 0; out_err: + trace_nfs_writepages_done(inode, wbc->range_start, wbc->range_end - wbc->range_start, err); return err; } /* * Insert a write request into an inode */ -static void nfs_inode_add_request(struct inode *inode, struct nfs_page *req) +static void nfs_inode_add_request(struct nfs_page *req) { - struct nfs_inode *nfsi = NFS_I(inode); + struct folio *folio = nfs_page_to_folio(req); + struct address_space *mapping = folio->mapping; + struct nfs_inode *nfsi = NFS_I(mapping->host); WARN_ON_ONCE(req->wb_this_page != req); /* Lock the request! */ nfs_lock_request(req); - - spin_lock(&inode->i_lock); - if (!nfsi->nrequests && - NFS_PROTO(inode)->have_delegation(inode, FMODE_WRITE)) - inode->i_version++; - /* - * Swap-space should not get truncated. Hence no need to plug the race - * with invalidate/truncate. - */ - if (likely(!PageSwapCache(req->wb_page))) { - set_bit(PG_MAPPED, &req->wb_flags); - SetPagePrivate(req->wb_page); - set_page_private(req->wb_page, (unsigned long)req); - } - nfsi->nrequests++; + spin_lock(&mapping->i_private_lock); + set_bit(PG_MAPPED, &req->wb_flags); + folio_set_private(folio); + folio->private = req; + spin_unlock(&mapping->i_private_lock); + atomic_long_inc(&nfsi->nrequests); /* this a head request for a page group - mark it as having an * extra reference so sub groups can follow suit. * This flag also informs pgio layer when to bump nrequests when * adding subrequests. */ WARN_ON(test_and_set_bit(PG_INODE_REF, &req->wb_flags)); kref_get(&req->wb_kref); - spin_unlock(&inode->i_lock); } /* @@ -794,68 +734,36 @@ static void nfs_inode_add_request(struct inode *inode, struct nfs_page *req) */ static void nfs_inode_remove_request(struct nfs_page *req) { - struct inode *inode = d_inode(req->wb_context->dentry); - struct nfs_inode *nfsi = NFS_I(inode); - struct nfs_page *head; + struct nfs_inode *nfsi = NFS_I(nfs_page_to_inode(req)); - if (nfs_page_group_sync_on_bit(req, PG_REMOVE)) { - head = req->wb_head; + nfs_page_group_lock(req); + if (nfs_page_group_sync_on_bit_locked(req, PG_REMOVE)) { + struct folio *folio = nfs_page_to_folio(req->wb_head); + struct address_space *mapping = folio->mapping; - spin_lock(&inode->i_lock); - if (likely(head->wb_page && !PageSwapCache(head->wb_page))) { - set_page_private(head->wb_page, 0); - ClearPagePrivate(head->wb_page); - clear_bit(PG_MAPPED, &head->wb_flags); + spin_lock(&mapping->i_private_lock); + if (likely(folio)) { + folio->private = NULL; + folio_clear_private(folio); + clear_bit(PG_MAPPED, &req->wb_head->wb_flags); } - nfsi->nrequests--; - spin_unlock(&inode->i_lock); - } else { - spin_lock(&inode->i_lock); - nfsi->nrequests--; - spin_unlock(&inode->i_lock); + spin_unlock(&mapping->i_private_lock); + + folio_end_dropbehind(folio); } + nfs_page_group_unlock(req); - if (test_and_clear_bit(PG_INODE_REF, &req->wb_flags)) + if (test_and_clear_bit(PG_INODE_REF, &req->wb_flags)) { + atomic_long_dec(&nfsi->nrequests); nfs_release_request(req); + } } -static void -nfs_mark_request_dirty(struct nfs_page *req) -{ - if (req->wb_page) - __set_page_dirty_nobuffers(req->wb_page); -} - -/* - * nfs_page_search_commits_for_head_request_locked - * - * Search through commit lists on @inode for the head request for @page. - * Must be called while holding the inode (which is cinfo) lock. - * - * Returns the head request if found, or NULL if not found. - */ -static struct nfs_page * -nfs_page_search_commits_for_head_request_locked(struct nfs_inode *nfsi, - struct page *page) +static void nfs_mark_request_dirty(struct nfs_page *req) { - struct nfs_page *freq, *t; - struct nfs_commit_info cinfo; - struct inode *inode = &nfsi->vfs_inode; - - nfs_init_cinfo_from_inode(&cinfo, inode); - - /* search through pnfs commit lists */ - freq = pnfs_search_commit_reqs(inode, &cinfo, page); - if (freq) - return freq->wb_head; - - /* Linearly search the commit list for the correct request */ - list_for_each_entry_safe(freq, t, &cinfo.mds->list, wb_list) { - if (freq->wb_page == page) - return freq->wb_head; - } - - return NULL; + struct folio *folio = nfs_page_to_folio(req); + if (folio) + filemap_dirty_folio(folio_mapping(folio), folio); } /** @@ -868,7 +776,8 @@ nfs_page_search_commits_for_head_request_locked(struct nfs_inode *nfsi, * number of outstanding requests requiring a commit as well as * the MM page stats. * - * The caller must hold cinfo->inode->i_lock, and the nfs_page lock. + * The caller must hold NFS_I(cinfo->inode)->commit_mutex, and the + * nfs_page lock. */ void nfs_request_add_commit_list_locked(struct nfs_page *req, struct list_head *dst, @@ -876,14 +785,13 @@ nfs_request_add_commit_list_locked(struct nfs_page *req, struct list_head *dst, { set_bit(PG_CLEAN, &req->wb_flags); nfs_list_add_request(req, dst); - cinfo->mds->ncommit++; + atomic_long_inc(&cinfo->mds->ncommit); } EXPORT_SYMBOL_GPL(nfs_request_add_commit_list_locked); /** * nfs_request_add_commit_list - add request to a commit list * @req: pointer to a struct nfs_page - * @dst: commit list head * @cinfo: holds list lock and accounting info * * This sets the PG_CLEAN bit, updates the cinfo count of @@ -896,11 +804,10 @@ EXPORT_SYMBOL_GPL(nfs_request_add_commit_list_locked); void nfs_request_add_commit_list(struct nfs_page *req, struct nfs_commit_info *cinfo) { - spin_lock(&cinfo->inode->i_lock); + mutex_lock(&NFS_I(cinfo->inode)->commit_mutex); nfs_request_add_commit_list_locked(req, &cinfo->mds->list, cinfo); - spin_unlock(&cinfo->inode->i_lock); - if (req->wb_page) - nfs_mark_page_unstable(req->wb_page, cinfo); + mutex_unlock(&NFS_I(cinfo->inode)->commit_mutex); + nfs_folio_mark_unstable(nfs_page_to_folio(req), cinfo); } EXPORT_SYMBOL_GPL(nfs_request_add_commit_list); @@ -922,7 +829,7 @@ nfs_request_remove_commit_list(struct nfs_page *req, if (!test_and_clear_bit(PG_CLEAN, &(req)->wb_flags)) return; nfs_list_remove_request(req); - cinfo->mds->ncommit--; + atomic_long_dec(&cinfo->mds->ncommit); } EXPORT_SYMBOL_GPL(nfs_request_remove_commit_list); @@ -959,27 +866,31 @@ nfs_mark_request_commit(struct nfs_page *req, struct pnfs_layout_segment *lseg, nfs_request_add_commit_list(req, cinfo); } -static void -nfs_clear_page_commit(struct page *page) +static void nfs_folio_clear_commit(struct folio *folio) { - dec_node_page_state(page, NR_UNSTABLE_NFS); - dec_wb_stat(&inode_to_bdi(page_file_mapping(page)->host)->wb, - WB_RECLAIMABLE); + if (folio) { + long nr = folio_nr_pages(folio); + + node_stat_mod_folio(folio, NR_WRITEBACK, -nr); + wb_stat_mod(&inode_to_bdi(folio->mapping->host)->wb, + WB_WRITEBACK, -nr); + } } -/* Called holding inode (/cinfo) lock */ -static void -nfs_clear_request_commit(struct nfs_page *req) +/* Called holding the request lock on @req */ +static void nfs_clear_request_commit(struct nfs_commit_info *cinfo, + struct nfs_page *req) { if (test_bit(PG_CLEAN, &req->wb_flags)) { - struct inode *inode = d_inode(req->wb_context->dentry); - struct nfs_commit_info cinfo; + struct nfs_open_context *ctx = nfs_req_openctx(req); + struct inode *inode = d_inode(ctx->dentry); - nfs_init_cinfo_from_inode(&cinfo, inode); - if (!pnfs_clear_request_commit(req, &cinfo)) { - nfs_request_remove_commit_list(req, &cinfo); + mutex_lock(&NFS_I(inode)->commit_mutex); + if (!pnfs_clear_request_commit(req, cinfo)) { + nfs_request_remove_commit_list(req, cinfo); } - nfs_clear_page_commit(req->wb_page); + mutex_unlock(&NFS_I(inode)->commit_mutex); + nfs_folio_clear_commit(nfs_page_to_folio(req)); } } @@ -1010,21 +921,23 @@ static void nfs_write_completion(struct nfs_pgio_header *hdr) nfs_list_remove_request(req); if (test_bit(NFS_IOHDR_ERROR, &hdr->flags) && (hdr->good_bytes < bytes)) { - nfs_set_pageerror(req->wb_page); - nfs_context_set_write_error(req->wb_context, hdr->error); + trace_nfs_comp_error(hdr->inode, req, hdr->error); + nfs_mapping_set_error(nfs_page_to_folio(req), + hdr->error); goto remove_req; } if (nfs_write_need_commit(hdr)) { + /* Reset wb_nio, since the write was successful. */ + req->wb_nio = 0; memcpy(&req->wb_verf, &hdr->verf.verifier, sizeof(req->wb_verf)); nfs_mark_request_commit(req, hdr->lseg, &cinfo, - hdr->pgio_mirror_idx); + hdr->ds_commit_idx); goto next; } remove_req: nfs_inode_remove_request(req); next: - nfs_unlock_request(req); - nfs_end_page_writeback(req); + nfs_page_end_writeback(req); nfs_release_request(req); } out: @@ -1035,10 +948,10 @@ out: unsigned long nfs_reqs_to_commit(struct nfs_commit_info *cinfo) { - return cinfo->mds->ncommit; + return atomic_long_read(&cinfo->mds->ncommit); } -/* cinfo->inode->i_lock held by caller */ +/* NFS_I(cinfo->inode)->commit_mutex held by caller */ int nfs_scan_commit_list(struct list_head *src, struct list_head *dst, struct nfs_commit_info *cinfo, int max) @@ -1047,19 +960,22 @@ nfs_scan_commit_list(struct list_head *src, struct list_head *dst, int ret = 0; list_for_each_entry_safe(req, tmp, src, wb_list) { - if (!nfs_lock_request(req)) - continue; kref_get(&req->wb_kref); - if (cond_resched_lock(&cinfo->inode->i_lock)) - list_safe_reset_next(req, tmp, wb_list); + if (!nfs_lock_request(req)) { + nfs_release_request(req); + continue; + } nfs_request_remove_commit_list(req, cinfo); + clear_bit(PG_COMMIT_TO_DS, &req->wb_flags); nfs_list_add_request(req, dst); ret++; if ((ret == max) && !cinfo->dreq) break; + cond_resched(); } return ret; } +EXPORT_SYMBOL_GPL(nfs_scan_commit_list); /* * nfs_scan_commit - Scan an inode for commit requests @@ -1076,15 +992,17 @@ nfs_scan_commit(struct inode *inode, struct list_head *dst, { int ret = 0; - spin_lock(&cinfo->inode->i_lock); - if (cinfo->mds->ncommit > 0) { + if (!atomic_long_read(&cinfo->mds->ncommit)) + return 0; + mutex_lock(&NFS_I(cinfo->inode)->commit_mutex); + if (atomic_long_read(&cinfo->mds->ncommit) > 0) { const int max = INT_MAX; ret = nfs_scan_commit_list(&cinfo->mds->list, dst, cinfo, max); ret += pnfs_scan_commit_lists(inode, cinfo, max - ret); } - spin_unlock(&cinfo->inode->i_lock); + mutex_unlock(&NFS_I(cinfo->inode)->commit_mutex); return ret; } @@ -1095,53 +1013,31 @@ nfs_scan_commit(struct inode *inode, struct list_head *dst, * If the attempt fails, then the existing request is flushed out * to disk. */ -static struct nfs_page *nfs_try_to_update_request(struct inode *inode, - struct page *page, - unsigned int offset, - unsigned int bytes) +static struct nfs_page *nfs_try_to_update_request(struct folio *folio, + unsigned int offset, + unsigned int bytes) { struct nfs_page *req; unsigned int rqend; unsigned int end; int error; - if (!PagePrivate(page)) - return NULL; - + trace_nfs_try_to_update_request(folio_inode(folio), offset, bytes); end = offset + bytes; - spin_lock(&inode->i_lock); - - for (;;) { - 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 - * the offsets are non-contiguous. - * Note: nfs_flush_incompatible() will already - * have flushed out requests having wrong owners. - */ - if (offset > rqend - || end < req->wb_offset) - goto out_flushme; - - if (nfs_lock_request(req)) - break; + req = nfs_lock_and_join_requests(folio); + if (IS_ERR_OR_NULL(req)) + goto out; - /* The request is locked, so wait and then retry */ - spin_unlock(&inode->i_lock); - error = nfs_wait_on_request(req); - nfs_release_request(req); - if (error != 0) - goto out_err; - spin_lock(&inode->i_lock); - } + rqend = req->wb_offset + req->wb_bytes; + /* + * Tell the caller to flush out the request if + * the offsets are non-contiguous. + * Note: nfs_flush_incompatible() will already + * have flushed out requests having wrong owners. + */ + if (offset > rqend || end < req->wb_offset) + goto out_flushme; /* Okay, the request matches. Update the region */ if (offset < req->wb_offset) { @@ -1152,17 +1048,22 @@ static struct nfs_page *nfs_try_to_update_request(struct inode *inode, req->wb_bytes = end - req->wb_offset; else req->wb_bytes = rqend - req->wb_offset; -out_unlock: - if (req) - nfs_clear_request_commit(req); - spin_unlock(&inode->i_lock); + req->wb_nio = 0; +out: + trace_nfs_try_to_update_request_done(folio_inode(folio), offset, bytes, + PTR_ERR_OR_ZERO(req)); return req; out_flushme: - spin_unlock(&inode->i_lock); - nfs_release_request(req); - error = nfs_wb_page(inode, page); -out_err: - return ERR_PTR(error); + /* + * Note: we mark the request dirty here because + * nfs_lock_and_join_requests() cannot preserve + * commit flags, so we have to replay the write. + */ + nfs_mark_request_dirty(req); + nfs_unlock_and_release_request(req); + error = nfs_wb_folio(folio->mapping->host, folio); + trace_nfs_try_to_update_request_done(folio_inode(folio), offset, bytes, error); + return (error < 0) ? ERR_PTR(error) : NULL; } /* @@ -1172,44 +1073,47 @@ out_err: * if we have to add a new request. Also assumes that the caller has * already called nfs_flush_incompatible() if necessary. */ -static struct nfs_page * nfs_setup_write_request(struct nfs_open_context* ctx, - struct page *page, unsigned int offset, unsigned int bytes) +static struct nfs_page *nfs_setup_write_request(struct nfs_open_context *ctx, + struct folio *folio, + unsigned int offset, + unsigned int bytes) { - struct inode *inode = page_file_mapping(page)->host; - struct nfs_page *req; + struct nfs_page *req; - req = nfs_try_to_update_request(inode, page, offset, bytes); + req = nfs_try_to_update_request(folio, offset, bytes); if (req != NULL) goto out; - req = nfs_create_request(ctx, page, NULL, offset, bytes); + req = nfs_page_create_from_folio(ctx, folio, offset, bytes); if (IS_ERR(req)) goto out; - nfs_inode_add_request(inode, req); + nfs_inode_add_request(req); out: return req; } -static int nfs_writepage_setup(struct nfs_open_context *ctx, struct page *page, - unsigned int offset, unsigned int count) +static int nfs_writepage_setup(struct nfs_open_context *ctx, + struct folio *folio, unsigned int offset, + unsigned int count) { - struct nfs_page *req; + struct nfs_page *req; - req = nfs_setup_write_request(ctx, page, offset, count); + req = nfs_setup_write_request(ctx, folio, offset, count); if (IS_ERR(req)) return PTR_ERR(req); + trace_nfs_writepage_setup(req); /* Update file length */ - nfs_grow_file(page, offset, count); + nfs_grow_file(folio, offset, count); nfs_mark_uptodate(req); nfs_mark_request_dirty(req); nfs_unlock_and_release_request(req); return 0; } -int nfs_flush_incompatible(struct file *file, struct page *page) +int nfs_flush_incompatible(struct file *file, struct folio *folio) { struct nfs_open_context *ctx = nfs_file_open_context(file); struct nfs_lock_context *l_ctx; - struct file_lock_context *flctx = file_inode(file)->i_flctx; + struct file_lock_context *flctx = locks_inode_context(file_inode(file)); struct nfs_page *req; int do_flush, status; /* @@ -1221,14 +1125,12 @@ int nfs_flush_incompatible(struct file *file, struct page *page) * dropped page. */ do { - req = nfs_page_find_head_request(page); + req = nfs_folio_find_head_request(folio); if (req == NULL) return 0; l_ctx = req->wb_lock_context; - do_flush = req->wb_page != page || - !nfs_match_open_context(req->wb_context, ctx); - /* for now, flush if more than 1 request in page_group */ - do_flush |= req->wb_this_page != req; + do_flush = nfs_page_to_folio(req) != folio || + !nfs_match_open_context(nfs_req_openctx(req), ctx); if (l_ctx && flctx && !(list_empty_careful(&flctx->flc_posix) && list_empty_careful(&flctx->flc_flock))) { @@ -1237,7 +1139,7 @@ int nfs_flush_incompatible(struct file *file, struct page *page) nfs_release_request(req); if (!do_flush) return 0; - status = nfs_wb_page(page_file_mapping(page)->host, page); + status = nfs_wb_folio(folio->mapping->host, folio); } while (status == 0); return status; } @@ -1256,9 +1158,12 @@ int nfs_key_timeout_notify(struct file *filp, struct inode *inode) { struct nfs_open_context *ctx = nfs_file_open_context(filp); - struct rpc_auth *auth = NFS_SERVER(inode)->client->cl_auth; - return rpcauth_key_timeout_notify(auth, ctx->cred); + if (nfs_ctx_key_to_expire(ctx, inode) && + !rcu_access_pointer(ctx->ll_cred)) + /* Already expired! */ + return -EACCES; + return 0; } /* @@ -1267,8 +1172,38 @@ nfs_key_timeout_notify(struct file *filp, struct inode *inode) bool nfs_ctx_key_to_expire(struct nfs_open_context *ctx, struct inode *inode) { struct rpc_auth *auth = NFS_SERVER(inode)->client->cl_auth; + struct rpc_cred *cred, *new, *old = NULL; + struct auth_cred acred = { + .cred = ctx->cred, + }; + bool ret = false; + + rcu_read_lock(); + cred = rcu_dereference(ctx->ll_cred); + if (cred && !(cred->cr_ops->crkey_timeout && + cred->cr_ops->crkey_timeout(cred))) + goto out; + rcu_read_unlock(); - return rpcauth_cred_key_to_expire(auth, ctx->cred); + new = auth->au_ops->lookup_cred(auth, &acred, 0); + if (new == cred) { + put_rpccred(new); + return true; + } + if (IS_ERR_OR_NULL(new)) { + new = NULL; + ret = true; + } else if (new->cr_ops->crkey_timeout && + new->cr_ops->crkey_timeout(new)) + ret = true; + + rcu_read_lock(); + old = rcu_dereference_protected(xchg(&ctx->ll_cred, + RCU_INITIALIZER(new)), 1); +out: + rcu_read_unlock(); + put_rpccred(old); + return ret; } /* @@ -1276,28 +1211,30 @@ bool nfs_ctx_key_to_expire(struct nfs_open_context *ctx, struct inode *inode) * the PageUptodate() flag. In this case, we will need to turn off * write optimisations that depend on the page contents being correct. */ -static bool nfs_write_pageuptodate(struct page *page, struct inode *inode) +static bool nfs_folio_write_uptodate(struct folio *folio, unsigned int pagelen) { + struct inode *inode = folio->mapping->host; struct nfs_inode *nfsi = NFS_I(inode); if (nfs_have_delegated_attributes(inode)) goto out; - if (nfsi->cache_validity & NFS_INO_REVAL_PAGECACHE) + if (nfsi->cache_validity & + (NFS_INO_INVALID_CHANGE | NFS_INO_INVALID_SIZE)) return false; smp_rmb(); - if (test_bit(NFS_INO_INVALIDATING, &nfsi->flags)) + if (test_bit(NFS_INO_INVALIDATING, &nfsi->flags) && pagelen != 0) return false; out: - if (nfsi->cache_validity & NFS_INO_INVALID_DATA) + if (nfsi->cache_validity & NFS_INO_INVALID_DATA && pagelen != 0) return false; - return PageUptodate(page) != 0; + return folio_test_uptodate(folio) != 0; } static bool is_whole_file_wrlock(struct file_lock *fl) { return fl->fl_start == 0 && fl->fl_end == OFFSET_MAX && - fl->fl_type == F_WRLCK; + lock_is_write(fl); } /* If we know the page is up to date, and we're not using byte range locks (or @@ -1308,17 +1245,22 @@ is_whole_file_wrlock(struct file_lock *fl) * 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) +static int nfs_can_extend_write(struct file *file, struct folio *folio, + unsigned int pagelen) { - int ret; - struct file_lock_context *flctx = inode->i_flctx; + struct inode *inode = file_inode(file); + struct file_lock_context *flctx = locks_inode_context(inode); struct file_lock *fl; + int ret; + unsigned int mntflags = NFS_SERVER(inode)->flags; + if (mntflags & NFS_MOUNT_NO_ALIGNWRITE) + return 0; if (file->f_flags & O_DSYNC) return 0; - if (!nfs_write_pageuptodate(page, inode)) + if (!nfs_folio_write_uptodate(folio, pagelen)) return 0; - if (NFS_PROTO(inode)->have_delegation(inode, FMODE_WRITE)) + if (nfs_have_write_delegation(inode)) return 1; if (!flctx || (list_empty_careful(&flctx->flc_flock) && list_empty_careful(&flctx->flc_posix))) @@ -1329,13 +1271,13 @@ static int nfs_can_extend_write(struct file *file, struct page *page, struct ino spin_lock(&flctx->flc_lock); if (!list_empty(&flctx->flc_posix)) { fl = list_first_entry(&flctx->flc_posix, struct file_lock, - fl_list); + c.flc_list); if (is_whole_file_wrlock(fl)) ret = 1; } else if (!list_empty(&flctx->flc_flock)) { fl = list_first_entry(&flctx->flc_flock, struct file_lock, - fl_list); - if (fl->fl_type == F_WRLCK) + c.flc_list); + if (lock_is_write(fl)) ret = 1; } spin_unlock(&flctx->flc_lock); @@ -1348,33 +1290,40 @@ static int nfs_can_extend_write(struct file *file, struct page *page, struct ino * XXX: Keep an eye on generic_file_read to make sure it doesn't do bad * things with a page scheduled for an RPC call (e.g. invalidate it). */ -int nfs_updatepage(struct file *file, struct page *page, - unsigned int offset, unsigned int count) +int nfs_update_folio(struct file *file, struct folio *folio, + unsigned int offset, unsigned int count) { struct nfs_open_context *ctx = nfs_file_open_context(file); - struct inode *inode = page_file_mapping(page)->host; + struct address_space *mapping = folio->mapping; + struct inode *inode = mapping->host; + unsigned int pagelen = nfs_folio_length(folio); int status = 0; nfs_inc_stats(inode, NFSIOS_VFSUPDATEPAGE); - dprintk("NFS: nfs_updatepage(%pD2 %d@%lld)\n", - file, count, (long long)(page_file_offset(page) + offset)); + trace_nfs_update_folio(inode, offset, count); + + dprintk("NFS: nfs_update_folio(%pD2 %d@%lld)\n", file, count, + (long long)(folio_pos(folio) + offset)); if (!count) goto out; - if (nfs_can_extend_write(file, page, inode)) { - count = max(count + offset, nfs_page_length(page)); - offset = 0; + if (nfs_can_extend_write(file, folio, pagelen)) { + unsigned int end = count + offset; + + offset = round_down(offset, PAGE_SIZE); + if (end < pagelen) + end = min(round_up(end, PAGE_SIZE), pagelen); + count = end - offset; } - status = nfs_writepage_setup(ctx, page, offset, count); + status = nfs_writepage_setup(ctx, folio, offset, count); if (status < 0) - nfs_set_pageerror(page); - else - __set_page_dirty_nobuffers(page); + nfs_set_pageerror(mapping); out: - dprintk("NFS: nfs_updatepage returns %d (isize %lld)\n", + trace_nfs_update_folio_done(inode, offset, count, status); + dprintk("NFS: nfs_update_folio returns %d (isize %lld)\n", status, (long long)i_size_read(inode)); return status; } @@ -1397,11 +1346,11 @@ static void nfs_initiate_write(struct nfs_pgio_header *hdr, { int priority = flush_task_priority(how); + if (IS_SWAPFILE(hdr->inode)) + task_setup_data->flags |= RPC_TASK_SWAPPER; task_setup_data->priority = priority; - rpc_ops->write_setup(hdr, msg); - - nfs4_state_protect_write(NFS_SERVER(hdr->inode)->nfs_client, - &task_setup_data->rpc_client, msg, hdr); + rpc_ops->write_setup(hdr, msg, &task_setup_data->rpc_client); + trace_nfs_initiate_write(hdr); } /* If a nfs_flush_* function fails, it should remove reqs from @head and @@ -1410,27 +1359,33 @@ static void nfs_initiate_write(struct nfs_pgio_header *hdr, */ static void nfs_redirty_request(struct nfs_page *req) { + struct nfs_inode *nfsi = NFS_I(nfs_page_to_inode(req)); + + /* Bump the transmission count */ + req->wb_nio++; nfs_mark_request_dirty(req); - set_bit(NFS_CONTEXT_RESEND_WRITES, &req->wb_context->flags); - nfs_unlock_request(req); - nfs_end_page_writeback(req); + atomic_long_inc(&nfsi->redirtied_pages); + nfs_page_end_writeback(req); nfs_release_request(req); } -static void nfs_async_write_error(struct list_head *head) +static void nfs_async_write_error(struct list_head *head, int error) { struct nfs_page *req; while (!list_empty(head)) { req = nfs_list_entry(head->next); nfs_list_remove_request(req); - nfs_redirty_request(req); + if (nfs_error_is_fatal_on_server(error)) + nfs_write_error(req, error); + else + nfs_redirty_request(req); } } static void nfs_async_write_reschedule_io(struct nfs_pgio_header *hdr) { - nfs_async_write_error(&hdr->pages); + nfs_async_write_error(&hdr->pages, 0); } static const struct nfs_pgio_completion_ops nfs_async_write_completion_ops = { @@ -1452,7 +1407,7 @@ void nfs_pageio_init_write(struct nfs_pageio_descriptor *pgio, 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, GFP_NOIO); + server->wsize, ioflags); } EXPORT_SYMBOL_GPL(nfs_pageio_init_write); @@ -1480,31 +1435,6 @@ void nfs_commit_prepare(struct rpc_task *task, void *calldata) NFS_PROTO(data->inode)->commit_rpc_prepare(task, data); } -/* - * 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; -} - static void nfs_writeback_check_extend(struct nfs_pgio_header *hdr, struct nfs_fattr *fattr) { @@ -1531,6 +1461,13 @@ void nfs_writeback_update_inode(struct nfs_pgio_header *hdr) struct nfs_fattr *fattr = &hdr->fattr; struct inode *inode = hdr->inode; + if (nfs_have_delegated_mtime(inode)) { + spin_lock(&inode->i_lock); + nfs_set_cache_invalid(inode, NFS_INO_INVALID_BLOCKS); + spin_unlock(&inode->i_lock); + return; + } + spin_lock(&inode->i_lock); nfs_writeback_check_extend(hdr, fattr); nfs_post_op_update_inode_force_wcc_locked(inode, fattr); @@ -1557,33 +1494,51 @@ static int nfs_writeback_done(struct rpc_task *task, status = NFS_PROTO(inode)->write_done(task, hdr); if (status != 0) return status; + nfs_add_stats(inode, NFSIOS_SERVERWRITTENBYTES, hdr->res.count); + trace_nfs_writeback_done(task, hdr); - if (hdr->res.verf->committed < hdr->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. - * Note: There is a known bug in Tru64 < 5.0 in which - * the server reports NFS_DATA_SYNC, but performs - * NFS_FILE_SYNC. We therefore implement this checking - * as a dprintk() in order to avoid filling syslog. - */ - static unsigned long complain; + if (task->tk_status >= 0) { + enum nfs3_stable_how committed = hdr->res.verf->committed; - /* Note this will print the MDS for a DS write */ - if (time_before(complain, jiffies)) { - dprintk("NFS: faulty NFS server %s:" - " (committed = %d) != (stable = %d)\n", - NFS_SERVER(inode)->nfs_client->cl_hostname, - hdr->res.verf->committed, hdr->args.stable); - complain = jiffies + 300 * HZ; + if (committed == NFS_UNSTABLE) { + /* + * We have some uncommitted data on the server at + * this point, so ensure that we keep track of that + * fact irrespective of what later writes do. + */ + set_bit(NFS_IOHDR_UNSTABLE_WRITES, &hdr->flags); + } + + if (committed < hdr->args.stable) { + /* We tried a write call, but the server did not + * commit data to stable storage even though we + * requested it. + * Note: There is a known bug in Tru64 < 5.0 in which + * the server reports NFS_DATA_SYNC, but performs + * NFS_FILE_SYNC. We therefore implement this checking + * as a dprintk() in order to avoid filling syslog. + */ + static unsigned long complain; + + /* Note this will print the MDS for a DS write */ + if (time_before(complain, jiffies)) { + dprintk("NFS: faulty NFS server %s:" + " (committed = %d) != (stable = %d)\n", + NFS_SERVER(inode)->nfs_client->cl_hostname, + committed, hdr->args.stable); + complain = jiffies + 300 * HZ; + } } } /* Deal with the suid/sgid bit corner case */ - if (nfs_should_remove_suid(inode)) - nfs_mark_for_revalidate(inode); + if (nfs_should_remove_suid(inode)) { + spin_lock(&inode->i_lock); + nfs_set_cache_invalid(inode, NFS_INO_INVALID_MODE + | NFS_INO_REVAL_FORCED); + spin_unlock(&inode->i_lock); + } return 0; } @@ -1634,25 +1589,30 @@ static void nfs_writeback_result(struct rpc_task *task, */ argp->stable = NFS_FILE_SYNC; } + resp->count = 0; + resp->verf->committed = 0; rpc_restart_call_prepare(task); } } static int wait_on_commit(struct nfs_mds_commit_info *cinfo) { - return wait_on_atomic_t(&cinfo->rpcs_out, - nfs_wait_atomic_killable, TASK_KILLABLE); + return wait_var_event_killable(&cinfo->rpcs_out, + !atomic_read(&cinfo->rpcs_out)); } -static void nfs_commit_begin(struct nfs_mds_commit_info *cinfo) +void nfs_commit_begin(struct nfs_mds_commit_info *cinfo) { atomic_inc(&cinfo->rpcs_out); } -static void nfs_commit_end(struct nfs_mds_commit_info *cinfo) +bool nfs_commit_end(struct nfs_mds_commit_info *cinfo) { - if (atomic_dec_and_test(&cinfo->rpcs_out)) - wake_up_atomic_t(&cinfo->rpcs_out); + if (atomic_dec_and_test(&cinfo->rpcs_out)) { + wake_up_var(&cinfo->rpcs_out); + return true; + } + return false; } void nfs_commitdata_release(struct nfs_commit_data *data) @@ -1665,7 +1625,8 @@ EXPORT_SYMBOL_GPL(nfs_commitdata_release); int nfs_initiate_commit(struct rpc_clnt *clnt, struct nfs_commit_data *data, const struct nfs_rpc_ops *nfs_ops, const struct rpc_call_ops *call_ops, - int how, int flags) + int how, int flags, + struct nfsd_file *localio) { struct rpc_task *task; int priority = flush_task_priority(how); @@ -1684,13 +1645,18 @@ int nfs_initiate_commit(struct rpc_clnt *clnt, struct nfs_commit_data *data, .flags = RPC_TASK_ASYNC | flags, .priority = priority, }; + + if (nfs_server_capable(data->inode, NFS_CAP_MOVEABLE)) + task_setup_data.flags |= RPC_TASK_MOVEABLE; + /* Set up the initial task struct. */ - nfs_ops->commit_setup(data, &msg); + nfs_ops->commit_setup(data, &msg, &task_setup_data.rpc_client); + trace_nfs_initiate_commit(data); dprintk("NFS: initiated commit call\n"); - nfs4_state_protect(NFS_SERVER(data->inode)->nfs_client, - NFS_SP4_MACH_CRED_COMMIT, &task_setup_data.rpc_client, &msg); + if (localio) + return nfs_local_commit(localio, data, call_ops, how); task = rpc_run_task(&task_setup_data); if (IS_ERR(task)) @@ -1722,16 +1688,22 @@ void nfs_init_commit(struct nfs_commit_data *data, struct pnfs_layout_segment *lseg, struct nfs_commit_info *cinfo) { - struct nfs_page *first = nfs_list_entry(head->next); - struct inode *inode = d_inode(first->wb_context->dentry); + struct nfs_page *first; + struct nfs_open_context *ctx; + struct inode *inode; /* Set up the RPC argument and reply structs * NB: take care not to mess about with data->commit et al. */ - list_splice_init(head, &data->pages); + if (head) + list_splice_init(head, &data->pages); + + first = nfs_list_entry(data->pages.next); + ctx = nfs_req_openctx(first); + inode = d_inode(ctx->dentry); data->inode = inode; - data->cred = first->wb_context->cred; + data->cred = ctx->cred; data->lseg = lseg; /* reference transferred */ /* only set lwb for pnfs commit */ if (lseg) @@ -1744,10 +1716,11 @@ void nfs_init_commit(struct nfs_commit_data *data, /* Note: we always request a commit of the entire inode */ data->args.offset = 0; data->args.count = 0; - data->context = get_nfs_open_context(first->wb_context); + data->context = get_nfs_open_context(ctx); data->res.fattr = &data->fattr; data->res.verf = &data->verf; nfs_fattr_init(&data->fattr); + nfs_commit_begin(cinfo->mds); } EXPORT_SYMBOL_GPL(nfs_init_commit); @@ -1762,18 +1735,18 @@ void nfs_retry_commit(struct list_head *page_list, req = nfs_list_entry(page_list->next); nfs_list_remove_request(req); nfs_mark_request_commit(req, lseg, cinfo, ds_commit_idx); - if (!cinfo->dreq) - nfs_clear_page_commit(req->wb_page); + nfs_folio_clear_commit(nfs_page_to_folio(req)); nfs_unlock_and_release_request(req); } } EXPORT_SYMBOL_GPL(nfs_retry_commit); -static void -nfs_commit_resched_write(struct nfs_commit_info *cinfo, - struct nfs_page *req) +static void nfs_commit_resched_write(struct nfs_commit_info *cinfo, + struct nfs_page *req) { - __set_page_dirty_nobuffers(req->wb_page); + struct folio *folio = nfs_page_to_folio(req); + + filemap_dirty_folio(folio_mapping(folio), folio); } /* @@ -1784,18 +1757,30 @@ nfs_commit_list(struct inode *inode, struct list_head *head, int how, struct nfs_commit_info *cinfo) { struct nfs_commit_data *data; + struct nfsd_file *localio; + unsigned short task_flags = 0; /* another commit raced with us */ if (list_empty(head)) return 0; - data = nfs_commitdata_alloc(true); + data = nfs_commitdata_alloc(); + if (!data) { + nfs_retry_commit(head, NULL, cinfo, -1); + return -ENOMEM; + } /* Set up the argument struct */ nfs_init_commit(data, head, NULL, cinfo); - atomic_inc(&cinfo->mds->rpcs_out); + if (NFS_SERVER(inode)->nfs_client->cl_minorversion) + task_flags = RPC_TASK_MOVEABLE; + + localio = nfs_local_open_fh(NFS_SERVER(inode)->nfs_client, data->cred, + data->args.fh, &data->context->nfl, + data->context->mode); return nfs_initiate_commit(NFS_CLIENT(inode), data, NFS_PROTO(inode), - data->mds_ops, how, 0); + data->mds_ops, how, + RPC_TASK_CRED_NOREF | task_flags, localio); } /* @@ -1805,58 +1790,59 @@ static void nfs_commit_done(struct rpc_task *task, void *calldata) { struct nfs_commit_data *data = calldata; - dprintk("NFS: %5u nfs_commit_done (status %d)\n", - task->tk_pid, task->tk_status); - /* Call the NFS version-specific code */ NFS_PROTO(data->inode)->commit_done(task, data); + trace_nfs_commit_done(task, data); } static void nfs_commit_release_pages(struct nfs_commit_data *data) { + const struct nfs_writeverf *verf = data->res.verf; struct nfs_page *req; int status = data->task.tk_status; struct nfs_commit_info cinfo; - struct nfs_server *nfss; + struct folio *folio; while (!list_empty(&data->pages)) { req = nfs_list_entry(data->pages.next); nfs_list_remove_request(req); - if (req->wb_page) - nfs_clear_page_commit(req->wb_page); + folio = nfs_page_to_folio(req); + nfs_folio_clear_commit(folio); dprintk("NFS: commit (%s/%llu %d@%lld)", - req->wb_context->dentry->d_sb->s_id, - (unsigned long long)NFS_FILEID(d_inode(req->wb_context->dentry)), + nfs_req_openctx(req)->dentry->d_sb->s_id, + (unsigned long long)NFS_FILEID(d_inode(nfs_req_openctx(req)->dentry)), req->wb_bytes, (long long)req_offset(req)); if (status < 0) { - nfs_context_set_write_error(req->wb_context, status); - if (req->wb_page) + if (folio) { + trace_nfs_commit_error(data->inode, req, + status); + nfs_mapping_set_error(folio, status); nfs_inode_remove_request(req); - dprintk_cont(", error = %d\n", status); + } + dprintk(", error = %d\n", status); goto next; } /* Okay, COMMIT succeeded, apparently. Check the verifier * returned by the server against all stored verfs. */ - if (!nfs_write_verifier_cmp(&req->wb_verf, &data->verf.verifier)) { + if (nfs_write_match_verf(verf, req)) { /* We have a match */ - if (req->wb_page) + if (folio) nfs_inode_remove_request(req); - dprintk_cont(" OK\n"); + dprintk(" OK\n"); goto next; } /* We have a mismatch. Write the page again */ - dprintk_cont(" mismatch\n"); + dprintk(" mismatch\n"); nfs_mark_request_dirty(req); - set_bit(NFS_CONTEXT_RESEND_WRITES, &req->wb_context->flags); + atomic_long_inc(&NFS_I(data->inode)->redirtied_pages); next: nfs_unlock_and_release_request(req); + /* Latency breaker */ + cond_resched(); } - nfss = NFS_SERVER(data->inode); - if (atomic_long_read(&nfss->writeback) < NFS_CONGESTION_OFF_THRESH) - clear_bdi_congested(inode_to_bdi(data->inode), BLK_RW_ASYNC); nfs_init_cinfo(&cinfo, data->inode, data->dreq); nfs_commit_end(cinfo.mds); @@ -1892,38 +1878,44 @@ int nfs_generic_commit_list(struct inode *inode, struct list_head *head, return status; } -int nfs_commit_inode(struct inode *inode, int how) +static int __nfs_commit_inode(struct inode *inode, int how, + struct writeback_control *wbc) { LIST_HEAD(head); struct nfs_commit_info cinfo; int may_wait = how & FLUSH_SYNC; - int error = 0; - int res; + int ret, nscan; + how &= ~FLUSH_SYNC; nfs_init_cinfo_from_inode(&cinfo, inode); nfs_commit_begin(cinfo.mds); - res = nfs_scan_commit(inode, &head, &cinfo); - if (res) - error = nfs_generic_commit_list(inode, &head, how, &cinfo); + for (;;) { + ret = nscan = nfs_scan_commit(inode, &head, &cinfo); + if (ret <= 0) + break; + ret = nfs_generic_commit_list(inode, &head, how, &cinfo); + if (ret < 0) + break; + ret = 0; + if (wbc && wbc->sync_mode == WB_SYNC_NONE) { + if (nscan < wbc->nr_to_write) + wbc->nr_to_write -= nscan; + else + wbc->nr_to_write = 0; + } + if (nscan < INT_MAX) + break; + cond_resched(); + } nfs_commit_end(cinfo.mds); - if (error < 0) - goto out_error; - if (!may_wait) - goto out_mark_dirty; - error = wait_on_commit(cinfo.mds); - if (error < 0) - return error; - return res; -out_error: - res = error; - /* Note: If we exit without ensuring that the commit is complete, - * we must mark the inode as dirty. Otherwise, future calls to - * sync_inode() with the WB_SYNC_ALL flag set will fail to ensure - * that the data is on the disk. - */ -out_mark_dirty: - __mark_inode_dirty(inode, I_DIRTY_DATASYNC); - return res; + if (ret || !may_wait) + return ret; + return wait_on_commit(cinfo.mds); +} + +int nfs_commit_inode(struct inode *inode, int how) +{ + return __nfs_commit_inode(inode, how, NULL); } EXPORT_SYMBOL_GPL(nfs_commit_inode); @@ -1933,11 +1925,11 @@ int nfs_write_inode(struct inode *inode, struct writeback_control *wbc) int flags = FLUSH_SYNC; int ret = 0; - /* no commits means nothing needs to be done */ - if (!nfsi->commit_info.ncommit) - return ret; - if (wbc->sync_mode == WB_SYNC_NONE) { + /* no commits means nothing needs to be done */ + if (!atomic_long_read(&nfsi->commit_info.ncommit)) + goto check_requests_outstanding; + /* Don't commit yet if this is a non-blocking flush and there * are a lot of outstanding writes for this mapping. */ @@ -1948,16 +1940,16 @@ int nfs_write_inode(struct inode *inode, struct writeback_control *wbc) flags = 0; } - ret = nfs_commit_inode(inode, flags); - if (ret >= 0) { - if (wbc->sync_mode == WB_SYNC_NONE) { - if (ret < wbc->nr_to_write) - wbc->nr_to_write -= ret; - else - wbc->nr_to_write = 0; - } - return 0; - } + ret = __nfs_commit_inode(inode, flags, wbc); + if (!ret) { + if (flags & FLUSH_SYNC) + return 0; + } else if (atomic_long_read(&nfsi->commit_info.ncommit)) + goto out_mark_dirty; + +check_requests_outstanding: + if (!atomic_read(&nfsi->commit_info.rpcs_out)) + return ret; out_mark_dirty: __mark_inode_dirty(inode, I_DIRTY_DATASYNC); return ret; @@ -2006,87 +1998,100 @@ out: } EXPORT_SYMBOL_GPL(nfs_wb_all); -int nfs_wb_page_cancel(struct inode *inode, struct page *page) +int nfs_wb_folio_cancel(struct inode *inode, struct folio *folio) { struct nfs_page *req; int ret = 0; - wait_on_page_writeback(page); + folio_wait_writeback(folio); /* blocking call to cancel all requests and join to a single (head) * request */ - req = nfs_lock_and_join_requests(page, false); + req = nfs_lock_and_join_requests(folio); if (IS_ERR(req)) { ret = PTR_ERR(req); } else if (req) { - /* all requests from this page have been cancelled by + /* all requests from this folio 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); nfs_unlock_and_release_request(req); + folio_cancel_dirty(folio); } return ret; } -/* - * Write back all requests on one page - we do this before reading it. +/** + * nfs_wb_folio - Write back all requests on one page + * @inode: pointer to page + * @folio: pointer to folio + * + * Assumes that the folio has been locked by the caller, and will + * not unlock it. */ -int nfs_wb_page(struct inode *inode, struct page *page) +int nfs_wb_folio(struct inode *inode, struct folio *folio) { - loff_t range_start = page_file_offset(page); - loff_t range_end = range_start + (loff_t)(PAGE_SIZE - 1); + loff_t range_start = folio_pos(folio); + size_t len = folio_size(folio); struct writeback_control wbc = { .sync_mode = WB_SYNC_ALL, .nr_to_write = 0, .range_start = range_start, - .range_end = range_end, + .range_end = range_start + len - 1, }; int ret; - trace_nfs_writeback_page_enter(inode); + trace_nfs_writeback_folio(inode, range_start, len); for (;;) { - wait_on_page_writeback(page); - if (clear_page_dirty_for_io(page)) { - ret = nfs_writepage_locked(page, &wbc); + folio_wait_writeback(folio); + if (folio_clear_dirty_for_io(folio)) { + ret = nfs_writepage_locked(folio, &wbc); if (ret < 0) goto out_error; continue; } ret = 0; - if (!PagePrivate(page)) + if (!folio_test_private(folio)) break; ret = nfs_commit_inode(inode, FLUSH_SYNC); if (ret < 0) goto out_error; } out_error: - trace_nfs_writeback_page_exit(inode, ret); + trace_nfs_writeback_folio_done(inode, range_start, len, ret); return ret; } #ifdef CONFIG_MIGRATION -int nfs_migrate_page(struct address_space *mapping, struct page *newpage, - struct page *page, enum migrate_mode mode) +int nfs_migrate_folio(struct address_space *mapping, struct folio *dst, + struct folio *src, enum migrate_mode mode) { /* - * If PagePrivate is set, then the page is currently associated with + * If the private flag is set, the folio is currently associated with * an in-progress read or write request. Don't try to migrate it. * * FIXME: we could do this in principle, but we'll need a way to ensure * that we can safely release the inode reference while holding - * the page lock. + * the folio lock. */ - if (PagePrivate(page)) - return -EBUSY; + if (folio_test_private(src)) { + if (mode == MIGRATE_SYNC) + nfs_wb_folio(src->mapping->host, src); + if (folio_test_private(src)) + return -EBUSY; + } - if (!nfs_fscache_release_page(page, GFP_KERNEL)) - return -EBUSY; + if (folio_test_private_2(src)) { /* [DEPRECATED] */ + if (mode == MIGRATE_ASYNC) + return -EBUSY; + folio_wait_private_2(src); + } - return migrate_page(mapping, newpage, page, mode); + return migrate_folio(mapping, dst, src, mode); } #endif @@ -2132,7 +2137,7 @@ int __init nfs_init_writepagecache(void) * This allows larger machines to have larger/more transfers. * Limit the default to 256M */ - nfs_congestion_kb = (16*int_sqrt(totalram_pages)) << (PAGE_SHIFT-10); + nfs_congestion_kb = (16*int_sqrt(totalram_pages())) << (PAGE_SHIFT-10); if (nfs_congestion_kb > 256*1024) nfs_congestion_kb = 256*1024; |