diff options
Diffstat (limited to 'fs')
110 files changed, 4206 insertions, 3313 deletions
diff --git a/fs/9p/vfs_addr.c b/fs/9p/vfs_addr.c index 055562c580b4..9ff073f4090a 100644 --- a/fs/9p/vfs_addr.c +++ b/fs/9p/vfs_addr.c @@ -148,13 +148,14 @@ static int v9fs_release_page(struct page *page, gfp_t gfp) * @offset: offset in the page */ -static void v9fs_invalidate_page(struct page *page, unsigned long offset) +static void v9fs_invalidate_page(struct page *page, unsigned int offset, + unsigned int length) { /* * If called with zero offset, we should release * the private state assocated with the page */ - if (offset == 0) + if (offset == 0 && length == PAGE_CACHE_SIZE) v9fs_fscache_invalidate_page(page); } diff --git a/fs/afs/file.c b/fs/afs/file.c index 8f6e9234d565..66d50fe2ee45 100644 --- a/fs/afs/file.c +++ b/fs/afs/file.c @@ -19,7 +19,8 @@ #include "internal.h" static int afs_readpage(struct file *file, struct page *page); -static void afs_invalidatepage(struct page *page, unsigned long offset); +static void afs_invalidatepage(struct page *page, unsigned int offset, + unsigned int length); static int afs_releasepage(struct page *page, gfp_t gfp_flags); static int afs_launder_page(struct page *page); @@ -310,16 +311,17 @@ static int afs_launder_page(struct page *page) * - release a page and clean up its private data if offset is 0 (indicating * the entire page) */ -static void afs_invalidatepage(struct page *page, unsigned long offset) +static void afs_invalidatepage(struct page *page, unsigned int offset, + unsigned int length) { struct afs_writeback *wb = (struct afs_writeback *) page_private(page); - _enter("{%lu},%lu", page->index, offset); + _enter("{%lu},%u,%u", page->index, offset, length); BUG_ON(!PageLocked(page)); /* we clean up only if the entire page is being invalidated */ - if (offset == 0) { + if (offset == 0 && length == PAGE_CACHE_SIZE) { #ifdef CONFIG_AFS_FSCACHE if (PageFsCache(page)) { struct afs_vnode *vnode = AFS_FS_I(page->mapping->host); diff --git a/fs/btrfs/disk-io.c b/fs/btrfs/disk-io.c index b8b60b660c8f..b0292b3ead54 100644 --- a/fs/btrfs/disk-io.c +++ b/fs/btrfs/disk-io.c @@ -1013,7 +1013,8 @@ static int btree_releasepage(struct page *page, gfp_t gfp_flags) return try_release_extent_buffer(page); } -static void btree_invalidatepage(struct page *page, unsigned long offset) +static void btree_invalidatepage(struct page *page, unsigned int offset, + unsigned int length) { struct extent_io_tree *tree; tree = &BTRFS_I(page->mapping->host)->io_tree; diff --git a/fs/btrfs/extent_io.c b/fs/btrfs/extent_io.c index e7e7afb4a872..6bca9472f313 100644 --- a/fs/btrfs/extent_io.c +++ b/fs/btrfs/extent_io.c @@ -2957,7 +2957,7 @@ static int __extent_writepage(struct page *page, struct writeback_control *wbc, pg_offset = i_size & (PAGE_CACHE_SIZE - 1); if (page->index > end_index || (page->index == end_index && !pg_offset)) { - page->mapping->a_ops->invalidatepage(page, 0); + page->mapping->a_ops->invalidatepage(page, 0, PAGE_CACHE_SIZE); unlock_page(page); return 0; } diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c index a46b656d08de..4f9d16b70d3d 100644 --- a/fs/btrfs/inode.c +++ b/fs/btrfs/inode.c @@ -7493,7 +7493,8 @@ static int btrfs_releasepage(struct page *page, gfp_t gfp_flags) return __btrfs_releasepage(page, gfp_flags & GFP_NOFS); } -static void btrfs_invalidatepage(struct page *page, unsigned long offset) +static void btrfs_invalidatepage(struct page *page, unsigned int offset, + unsigned int length) { struct inode *inode = page->mapping->host; struct extent_io_tree *tree; diff --git a/fs/buffer.c b/fs/buffer.c index d2a4d1bb2d57..f93392e2df12 100644 --- a/fs/buffer.c +++ b/fs/buffer.c @@ -1454,7 +1454,8 @@ static void discard_buffer(struct buffer_head * bh) * block_invalidatepage - invalidate part or all of a buffer-backed page * * @page: the page which is affected - * @offset: the index of the truncation point + * @offset: start of the range to invalidate + * @length: length of the range to invalidate * * block_invalidatepage() is called when all or part of the page has become * invalidated by a truncate operation. @@ -1465,15 +1466,22 @@ static void discard_buffer(struct buffer_head * bh) * point. Because the caller is about to free (and possibly reuse) those * blocks on-disk. */ -void block_invalidatepage(struct page *page, unsigned long offset) +void block_invalidatepage(struct page *page, unsigned int offset, + unsigned int length) { struct buffer_head *head, *bh, *next; unsigned int curr_off = 0; + unsigned int stop = length + offset; BUG_ON(!PageLocked(page)); if (!page_has_buffers(page)) goto out; + /* + * Check for overflow + */ + BUG_ON(stop > PAGE_CACHE_SIZE || stop < length); + head = page_buffers(page); bh = head; do { @@ -1481,6 +1489,12 @@ void block_invalidatepage(struct page *page, unsigned long offset) next = bh->b_this_page; /* + * Are we still fully in range ? + */ + if (next_off > stop) + goto out; + + /* * is this block fully invalidated? */ if (offset <= curr_off) @@ -1501,6 +1515,7 @@ out: } EXPORT_SYMBOL(block_invalidatepage); + /* * We attach and possibly dirty the buffers atomically wrt * __set_page_dirty_buffers() via private_lock. try_to_free_buffers @@ -2841,7 +2856,7 @@ int block_write_full_page_endio(struct page *page, get_block_t *get_block, * they may have been added in ext3_writepage(). Make them * freeable here, so the page does not leak. */ - do_invalidatepage(page, 0); + do_invalidatepage(page, 0, PAGE_CACHE_SIZE); unlock_page(page); return 0; /* don't care */ } diff --git a/fs/cachefiles/interface.c b/fs/cachefiles/interface.c index 746ce532e130..d4c1206af9fc 100644 --- a/fs/cachefiles/interface.c +++ b/fs/cachefiles/interface.c @@ -13,8 +13,6 @@ #include <linux/mount.h> #include "internal.h" -#define list_to_page(head) (list_entry((head)->prev, struct page, lru)) - struct cachefiles_lookup_data { struct cachefiles_xattr *auxdata; /* auxiliary data */ char *key; /* key path */ @@ -212,20 +210,29 @@ static void cachefiles_update_object(struct fscache_object *_object) object = container_of(_object, struct cachefiles_object, fscache); cache = container_of(object->fscache.cache, struct cachefiles_cache, cache); + + if (!fscache_use_cookie(_object)) { + _leave(" [relinq]"); + return; + } + cookie = object->fscache.cookie; if (!cookie->def->get_aux) { + fscache_unuse_cookie(_object); _leave(" [no aux]"); return; } auxdata = kmalloc(2 + 512 + 3, cachefiles_gfp); if (!auxdata) { + fscache_unuse_cookie(_object); _leave(" [nomem]"); return; } auxlen = cookie->def->get_aux(cookie->netfs_data, auxdata->data, 511); + fscache_unuse_cookie(_object); ASSERTCMP(auxlen, <, 511); auxdata->len = auxlen + 1; @@ -263,7 +270,7 @@ static void cachefiles_drop_object(struct fscache_object *_object) #endif /* delete retired objects */ - if (object->fscache.state == FSCACHE_OBJECT_RECYCLING && + if (test_bit(FSCACHE_COOKIE_RETIRED, &object->fscache.cookie->flags) && _object != cache->cache.fsdef ) { _debug("- retire object OBJ%x", object->fscache.debug_id); diff --git a/fs/cachefiles/namei.c b/fs/cachefiles/namei.c index 8c01c5fcdf75..25badd1aec5c 100644 --- a/fs/cachefiles/namei.c +++ b/fs/cachefiles/namei.c @@ -38,7 +38,7 @@ void __cachefiles_printk_object(struct cachefiles_object *object, printk(KERN_ERR "%sobject: OBJ%x\n", prefix, object->fscache.debug_id); printk(KERN_ERR "%sobjstate=%s fl=%lx wbusy=%x ev=%lx[%lx]\n", - prefix, fscache_object_states[object->fscache.state], + prefix, object->fscache.state->name, object->fscache.flags, work_busy(&object->fscache.work), object->fscache.events, object->fscache.event_mask); printk(KERN_ERR "%sops=%u inp=%u exc=%u\n", @@ -127,10 +127,10 @@ static void cachefiles_mark_object_buried(struct cachefiles_cache *cache, found_dentry: kdebug("preemptive burial: OBJ%x [%s] %p", object->fscache.debug_id, - fscache_object_states[object->fscache.state], + object->fscache.state->name, dentry); - if (object->fscache.state < FSCACHE_OBJECT_DYING) { + if (fscache_object_is_live(&object->fscache)) { printk(KERN_ERR "\n"); printk(KERN_ERR "CacheFiles: Error:" " Can't preemptively bury live object\n"); @@ -192,7 +192,7 @@ try_again: /* an old object from a previous incarnation is hogging the slot - we * need to wait for it to be destroyed */ wait_for_old_object: - if (xobject->fscache.state < FSCACHE_OBJECT_DYING) { + if (fscache_object_is_live(&object->fscache)) { printk(KERN_ERR "\n"); printk(KERN_ERR "CacheFiles: Error:" " Unexpected object collision\n"); @@ -836,7 +836,7 @@ static struct dentry *cachefiles_check_active(struct cachefiles_cache *cache, // dir->d_name.len, dir->d_name.len, dir->d_name.name, filename); /* look up the victim */ - mutex_lock_nested(&dir->d_inode->i_mutex, 1); + mutex_lock_nested(&dir->d_inode->i_mutex, I_MUTEX_PARENT); start = jiffies; victim = lookup_one_len(filename, dir, strlen(filename)); diff --git a/fs/cachefiles/xattr.c b/fs/cachefiles/xattr.c index 73b46288b54b..2476e5162609 100644 --- a/fs/cachefiles/xattr.c +++ b/fs/cachefiles/xattr.c @@ -109,13 +109,12 @@ int cachefiles_set_object_xattr(struct cachefiles_object *object, struct dentry *dentry = object->dentry; int ret; - ASSERT(object->fscache.cookie); ASSERT(dentry); _enter("%p,#%d", object, auxdata->len); /* attempt to install the cache metadata directly */ - _debug("SET %s #%u", object->fscache.cookie->def->name, auxdata->len); + _debug("SET #%u", auxdata->len); ret = vfs_setxattr(dentry, cachefiles_xattr_cache, &auxdata->type, auxdata->len, @@ -138,13 +137,12 @@ int cachefiles_update_object_xattr(struct cachefiles_object *object, struct dentry *dentry = object->dentry; int ret; - ASSERT(object->fscache.cookie); ASSERT(dentry); _enter("%p,#%d", object, auxdata->len); /* attempt to install the cache metadata directly */ - _debug("SET %s #%u", object->fscache.cookie->def->name, auxdata->len); + _debug("SET #%u", auxdata->len); ret = vfs_setxattr(dentry, cachefiles_xattr_cache, &auxdata->type, auxdata->len, diff --git a/fs/ceph/addr.c b/fs/ceph/addr.c index 3e68ac101040..38b5c1bc6776 100644 --- a/fs/ceph/addr.c +++ b/fs/ceph/addr.c @@ -143,7 +143,8 @@ static int ceph_set_page_dirty(struct page *page) * dirty page counters appropriately. Only called if there is private * data on the page. */ -static void ceph_invalidatepage(struct page *page, unsigned long offset) +static void ceph_invalidatepage(struct page *page, unsigned int offset, + unsigned int length) { struct inode *inode; struct ceph_inode_info *ci; @@ -163,20 +164,20 @@ static void ceph_invalidatepage(struct page *page, unsigned long offset) if (!PageDirty(page)) pr_err("%p invalidatepage %p page not dirty\n", inode, page); - if (offset == 0) + if (offset == 0 && length == PAGE_CACHE_SIZE) ClearPageChecked(page); ci = ceph_inode(inode); - if (offset == 0) { - dout("%p invalidatepage %p idx %lu full dirty page %lu\n", - inode, page, page->index, offset); + if (offset == 0 && length == PAGE_CACHE_SIZE) { + dout("%p invalidatepage %p idx %lu full dirty page\n", + inode, page, page->index); ceph_put_wrbuffer_cap_refs(ci, 1, snapc); ceph_put_snap_context(snapc); page->private = 0; ClearPagePrivate(page); } else { - dout("%p invalidatepage %p idx %lu partial dirty page\n", - inode, page, page->index); + dout("%p invalidatepage %p idx %lu partial dirty page %u(%u)\n", + inode, page, page->index, offset, length); } } diff --git a/fs/cifs/file.c b/fs/cifs/file.c index 0630710a9c3f..91d8629e69a2 100644 --- a/fs/cifs/file.c +++ b/fs/cifs/file.c @@ -3547,11 +3547,12 @@ static int cifs_release_page(struct page *page, gfp_t gfp) return cifs_fscache_release_page(page, gfp); } -static void cifs_invalidate_page(struct page *page, unsigned long offset) +static void cifs_invalidate_page(struct page *page, unsigned int offset, + unsigned int length) { struct cifsInodeInfo *cifsi = CIFS_I(page->mapping->host); - if (offset == 0) + if (offset == 0 && length == PAGE_CACHE_SIZE) cifs_fscache_invalidate_page(page, &cifsi->vfs_inode); } diff --git a/fs/dcache.c b/fs/dcache.c index 3199fe6863a8..87bdb5329c3c 100644 --- a/fs/dcache.c +++ b/fs/dcache.c @@ -1612,6 +1612,10 @@ EXPORT_SYMBOL(d_obtain_alias); * If a dentry was found and moved, then it is returned. Otherwise NULL * is returned. This matches the expected return value of ->lookup. * + * Cluster filesystems may call this function with a negative, hashed dentry. + * In that case, we know that the inode will be a regular file, and also this + * will only occur during atomic_open. So we need to check for the dentry + * being already hashed only in the final case. */ struct dentry *d_splice_alias(struct inode *inode, struct dentry *dentry) { @@ -1636,8 +1640,11 @@ struct dentry *d_splice_alias(struct inode *inode, struct dentry *dentry) security_d_instantiate(dentry, inode); d_rehash(dentry); } - } else - d_add(dentry, inode); + } else { + d_instantiate(dentry, inode); + if (d_unhashed(dentry)) + d_rehash(dentry); + } return new; } EXPORT_SYMBOL(d_splice_alias); diff --git a/fs/debugfs/file.c b/fs/debugfs/file.c index c5ca6ae5a30c..63146295153b 100644 --- a/fs/debugfs/file.c +++ b/fs/debugfs/file.c @@ -21,6 +21,7 @@ #include <linux/debugfs.h> #include <linux/io.h> #include <linux/slab.h> +#include <linux/atomic.h> static ssize_t default_read_file(struct file *file, char __user *buf, size_t count, loff_t *ppos) @@ -403,6 +404,47 @@ struct dentry *debugfs_create_size_t(const char *name, umode_t mode, } EXPORT_SYMBOL_GPL(debugfs_create_size_t); +static int debugfs_atomic_t_set(void *data, u64 val) +{ + atomic_set((atomic_t *)data, val); + return 0; +} +static int debugfs_atomic_t_get(void *data, u64 *val) +{ + *val = atomic_read((atomic_t *)data); + return 0; +} +DEFINE_SIMPLE_ATTRIBUTE(fops_atomic_t, debugfs_atomic_t_get, + debugfs_atomic_t_set, "%lld\n"); +DEFINE_SIMPLE_ATTRIBUTE(fops_atomic_t_ro, debugfs_atomic_t_get, NULL, "%lld\n"); +DEFINE_SIMPLE_ATTRIBUTE(fops_atomic_t_wo, NULL, debugfs_atomic_t_set, "%lld\n"); + +/** + * debugfs_create_atomic_t - create a debugfs file that is used to read and + * write an atomic_t value + * @name: a pointer to a string containing the name of the file to create. + * @mode: the permission that the file should have + * @parent: a pointer to the parent dentry for this file. This should be a + * directory dentry if set. If this parameter is %NULL, then the + * file will be created in the root of the debugfs filesystem. + * @value: a pointer to the variable that the file should read to and write + * from. + */ +struct dentry *debugfs_create_atomic_t(const char *name, umode_t mode, + struct dentry *parent, atomic_t *value) +{ + /* if there are no write bits set, make read only */ + if (!(mode & S_IWUGO)) + return debugfs_create_file(name, mode, parent, value, + &fops_atomic_t_ro); + /* if there are no read bits set, make write only */ + if (!(mode & S_IRUGO)) + return debugfs_create_file(name, mode, parent, value, + &fops_atomic_t_wo); + + return debugfs_create_file(name, mode, parent, value, &fops_atomic_t); +} +EXPORT_SYMBOL_GPL(debugfs_create_atomic_t); static ssize_t read_file_bool(struct file *file, char __user *user_buf, size_t count, loff_t *ppos) @@ -431,6 +473,7 @@ static ssize_t write_file_bool(struct file *file, const char __user *user_buf, if (copy_from_user(buf, user_buf, buf_size)) return -EFAULT; + buf[buf_size] = '\0'; if (strtobool(buf, &bv) == 0) *val = bv; diff --git a/fs/dlm/config.c b/fs/dlm/config.c index 7d58d5b112b5..76feb4b60fa6 100644 --- a/fs/dlm/config.c +++ b/fs/dlm/config.c @@ -138,8 +138,9 @@ static ssize_t cluster_cluster_name_read(struct dlm_cluster *cl, char *buf) static ssize_t cluster_cluster_name_write(struct dlm_cluster *cl, const char *buf, size_t len) { - strncpy(dlm_config.ci_cluster_name, buf, DLM_LOCKSPACE_LEN); - strncpy(cl->cl_cluster_name, buf, DLM_LOCKSPACE_LEN); + strlcpy(dlm_config.ci_cluster_name, buf, + sizeof(dlm_config.ci_cluster_name)); + strlcpy(cl->cl_cluster_name, buf, sizeof(cl->cl_cluster_name)); return len; } diff --git a/fs/dlm/lock.c b/fs/dlm/lock.c index 1b1146670c4b..e223a911a834 100644 --- a/fs/dlm/lock.c +++ b/fs/dlm/lock.c @@ -2038,8 +2038,8 @@ static void set_lvb_lock_pc(struct dlm_rsb *r, struct dlm_lkb *lkb, b = dlm_lvb_operations[lkb->lkb_grmode + 1][lkb->lkb_rqmode + 1]; if (b == 1) { int len = receive_extralen(ms); - if (len > DLM_RESNAME_MAXLEN) - len = DLM_RESNAME_MAXLEN; + if (len > r->res_ls->ls_lvblen) + len = r->res_ls->ls_lvblen; memcpy(lkb->lkb_lvbptr, ms->m_extra, len); lkb->lkb_lvbseq = ms->m_lvbseq; } @@ -3893,8 +3893,8 @@ static int receive_lvb(struct dlm_ls *ls, struct dlm_lkb *lkb, if (!lkb->lkb_lvbptr) return -ENOMEM; len = receive_extralen(ms); - if (len > DLM_RESNAME_MAXLEN) - len = DLM_RESNAME_MAXLEN; + if (len > ls->ls_lvblen) + len = ls->ls_lvblen; memcpy(lkb->lkb_lvbptr, ms->m_extra, len); } return 0; diff --git a/fs/dlm/lockspace.c b/fs/dlm/lockspace.c index 3ca79d3253b9..88556dc0458e 100644 --- a/fs/dlm/lockspace.c +++ b/fs/dlm/lockspace.c @@ -883,17 +883,24 @@ int dlm_release_lockspace(void *lockspace, int force) void dlm_stop_lockspaces(void) { struct dlm_ls *ls; + int count; restart: + count = 0; spin_lock(&lslist_lock); list_for_each_entry(ls, &lslist, ls_list) { - if (!test_bit(LSFL_RUNNING, &ls->ls_flags)) + if (!test_bit(LSFL_RUNNING, &ls->ls_flags)) { + count++; continue; + } spin_unlock(&lslist_lock); log_error(ls, "no userland control daemon, stopping lockspace"); dlm_ls_stop(ls); goto restart; } spin_unlock(&lslist_lock); + + if (count) + log_print("dlm user daemon left %d lockspaces", count); } diff --git a/fs/dlm/lowcomms.c b/fs/dlm/lowcomms.c index d0ccd2fd79eb..d90909ec6aa6 100644 --- a/fs/dlm/lowcomms.c +++ b/fs/dlm/lowcomms.c @@ -52,7 +52,6 @@ #include <linux/mutex.h> #include <linux/sctp.h> #include <linux/slab.h> -#include <linux/sctp.h> #include <net/sctp/sctp.h> #include <net/ipv6.h> @@ -126,6 +125,7 @@ struct connection { struct connection *othercon; struct work_struct rwork; /* Receive workqueue */ struct work_struct swork; /* Send workqueue */ + bool try_new_addr; }; #define sock2con(x) ((struct connection *)(x)->sk_user_data) @@ -144,6 +144,7 @@ struct dlm_node_addr { struct list_head list; int nodeid; int addr_count; + int curr_addr_index; struct sockaddr_storage *addr[DLM_MAX_ADDR_COUNT]; }; @@ -310,7 +311,7 @@ static int addr_compare(struct sockaddr_storage *x, struct sockaddr_storage *y) } static int nodeid_to_addr(int nodeid, struct sockaddr_storage *sas_out, - struct sockaddr *sa_out) + struct sockaddr *sa_out, bool try_new_addr) { struct sockaddr_storage sas; struct dlm_node_addr *na; @@ -320,8 +321,16 @@ static int nodeid_to_addr(int nodeid, struct sockaddr_storage *sas_out, spin_lock(&dlm_node_addrs_spin); na = find_node_addr(nodeid); - if (na && na->addr_count) - memcpy(&sas, na->addr[0], sizeof(struct sockaddr_storage)); + if (na && na->addr_count) { + if (try_new_addr) { + na->curr_addr_index++; + if (na->curr_addr_index == na->addr_count) + na->curr_addr_index = 0; + } + + memcpy(&sas, na->addr[na->curr_addr_index ], + sizeof(struct sockaddr_storage)); + } spin_unlock(&dlm_node_addrs_spin); if (!na) @@ -353,19 +362,22 @@ static int addr_to_nodeid(struct sockaddr_storage *addr, int *nodeid) { struct dlm_node_addr *na; int rv = -EEXIST; + int addr_i; spin_lock(&dlm_node_addrs_spin); list_for_each_entry(na, &dlm_node_addrs, list) { if (!na->addr_count) continue; - if (!addr_compare(na->addr[0], addr)) - continue; - - *nodeid = na->nodeid; - rv = 0; - break; + for (addr_i = 0; addr_i < na->addr_count; addr_i++) { + if (addr_compare(na->addr[addr_i], addr)) { + *nodeid = na->nodeid; + rv = 0; + goto unlock; + } + } } +unlock: spin_unlock(&dlm_node_addrs_spin); return rv; } @@ -561,8 +573,23 @@ static void sctp_send_shutdown(sctp_assoc_t associd) static void sctp_init_failed_foreach(struct connection *con) { + + /* + * Don't try to recover base con and handle race where the + * other node's assoc init creates a assoc and we get that + * notification, then we get a notification that our attempt + * failed due. This happens when we are still trying the primary + * address, but the other node has already tried secondary addrs + * and found one that worked. + */ + if (!con->nodeid || con->sctp_assoc) + return; + + log_print("Retrying SCTP association init for node %d\n", con->nodeid); + + con->try_new_addr = true; con->sctp_assoc = 0; - if (test_and_clear_bit(CF_CONNECT_PENDING, &con->flags)) { + if (test_and_clear_bit(CF_INIT_PENDING, &con->flags)) { if (!test_and_set_bit(CF_WRITE_PENDING, &con->flags)) queue_work(send_workqueue, &con->swork); } @@ -579,15 +606,56 @@ static void sctp_init_failed(void) mutex_unlock(&connections_lock); } +static void retry_failed_sctp_send(struct connection *recv_con, + struct sctp_send_failed *sn_send_failed, + char *buf) +{ + int len = sn_send_failed->ssf_length - sizeof(struct sctp_send_failed); + struct dlm_mhandle *mh; + struct connection *con; + char *retry_buf; + int nodeid = sn_send_failed->ssf_info.sinfo_ppid; + + log_print("Retry sending %d bytes to node id %d", len, nodeid); + + con = nodeid2con(nodeid, 0); + if (!con) { + log_print("Could not look up con for nodeid %d\n", + nodeid); + return; + } + + mh = dlm_lowcomms_get_buffer(nodeid, len, GFP_NOFS, &retry_buf); + if (!mh) { + log_print("Could not allocate buf for retry."); + return; + } + memcpy(retry_buf, buf + sizeof(struct sctp_send_failed), len); + dlm_lowcomms_commit_buffer(mh); + + /* + * If we got a assoc changed event before the send failed event then + * we only need to retry the send. + */ + if (con->sctp_assoc) { + if (!test_and_set_bit(CF_WRITE_PENDING, &con->flags)) + queue_work(send_workqueue, &con->swork); + } else + sctp_init_failed_foreach(con); +} + /* Something happened to an association */ static void process_sctp_notification(struct connection *con, struct msghdr *msg, char *buf) { union sctp_notification *sn = (union sctp_notification *)buf; - if (sn->sn_header.sn_type == SCTP_ASSOC_CHANGE) { + switch (sn->sn_header.sn_type) { + case SCTP_SEND_FAILED: + retry_failed_sctp_send(con, &sn->sn_send_failed, buf); + break; + case SCTP_ASSOC_CHANGE: switch (sn->sn_assoc_change.sac_state) { - case SCTP_COMM_UP: case SCTP_RESTART: { @@ -662,9 +730,11 @@ static void process_sctp_notification(struct connection *con, log_print("connecting to %d sctp association %d", nodeid, (int)sn->sn_assoc_change.sac_assoc_id); + new_con->sctp_assoc = sn->sn_assoc_change.sac_assoc_id; + new_con->try_new_addr = false; /* Send any pending writes */ clear_bit(CF_CONNECT_PENDING, &new_con->flags); - clear_bit(CF_INIT_PENDING, &con->flags); + clear_bit(CF_INIT_PENDING, &new_con->flags); if (!test_and_set_bit(CF_WRITE_PENDING, &new_con->flags)) { queue_work(send_workqueue, &new_con->swork); } @@ -683,14 +753,10 @@ static void process_sctp_notification(struct connection *con, } break; - /* We don't know which INIT failed, so clear the PENDING flags - * on them all. if assoc_id is zero then it will then try - * again */ - case SCTP_CANT_STR_ASSOC: { + /* Will retry init when we get the send failed notification */ log_print("Can't start SCTP association - retrying"); - sctp_init_failed(); } break; @@ -699,6 +765,8 @@ static void process_sctp_notification(struct connection *con, (int)sn->sn_assoc_change.sac_assoc_id, sn->sn_assoc_change.sac_state); } + default: + ; /* fall through */ } } @@ -958,6 +1026,24 @@ static void free_entry(struct writequeue_entry *e) kfree(e); } +/* + * writequeue_entry_complete - try to delete and free write queue entry + * @e: write queue entry to try to delete + * @completed: bytes completed + * + * writequeue_lock must be held. + */ +static void writequeue_entry_complete(struct writequeue_entry *e, int completed) +{ + e->offset += completed; + e->len -= completed; + + if (e->len == 0 && e->users == 0) { + list_del(&e->list); + free_entry(e); + } +} + /* Initiate an SCTP association. This is a special case of send_to_sock() in that we don't yet have a peeled-off socket for this association, so we use the listening socket @@ -977,15 +1063,14 @@ static void sctp_init_assoc(struct connection *con) int addrlen; struct kvec iov[1]; + mutex_lock(&con->sock_mutex); if (test_and_set_bit(CF_INIT_PENDING, &con->flags)) - return; - - if (con->retries++ > MAX_CONNECT_RETRIES) - return; + goto unlock; - if (nodeid_to_addr(con->nodeid, NULL, (struct sockaddr *)&rem_addr)) { + if (nodeid_to_addr(con->nodeid, NULL, (struct sockaddr *)&rem_addr, + con->try_new_addr)) { log_print("no address for nodeid %d", con->nodeid); - return; + goto unlock; } base_con = nodeid2con(0, 0); BUG_ON(base_con == NULL); @@ -1003,17 +1088,25 @@ static void sctp_init_assoc(struct connection *con) if (list_empty(&con->writequeue)) { spin_unlock(&con->writequeue_lock); log_print("writequeue empty for nodeid %d", con->nodeid); - return; + goto unlock; } e = list_first_entry(&con->writequeue, struct writequeue_entry, list); len = e->len; offset = e->offset; - spin_unlock(&con->writequeue_lock); /* Send the first block off the write queue */ iov[0].iov_base = page_address(e->page)+offset; iov[0].iov_len = len; + spin_unlock(&con->writequeue_lock); + + if (rem_addr.ss_family == AF_INET) { + struct sockaddr_in *sin = (struct sockaddr_in *)&rem_addr; + log_print("Trying to connect to %pI4", &sin->sin_addr.s_addr); + } else { + struct sockaddr_in6 *sin6 = (struct sockaddr_in6 *)&rem_addr; + log_print("Trying to connect to %pI6", &sin6->sin6_addr); + } cmsg = CMSG_FIRSTHDR(&outmessage); cmsg->cmsg_level = IPPROTO_SCTP; @@ -1021,8 +1114,9 @@ static void sctp_init_assoc(struct connection *con) cmsg->cmsg_len = CMSG_LEN(sizeof(struct sctp_sndrcvinfo)); sinfo = CMSG_DATA(cmsg); memset(sinfo, 0x00, sizeof(struct sctp_sndrcvinfo)); - sinfo->sinfo_ppid = cpu_to_le32(dlm_our_nodeid()); + sinfo->sinfo_ppid = cpu_to_le32(con->nodeid); outmessage.msg_controllen = cmsg->cmsg_len; + sinfo->sinfo_flags |= SCTP_ADDR_OVER; ret = kernel_sendmsg(base_con->sock, &outmessage, iov, 1, len); if (ret < 0) { @@ -1035,15 +1129,12 @@ static void sctp_init_assoc(struct connection *con) } else { spin_lock(&con->writequeue_lock); - e->offset += ret; - e->len -= ret; - - if (e->len == 0 && e->users == 0) { - list_del(&e->list); - free_entry(e); - } + writequeue_entry_complete(e, ret); spin_unlock(&con->writequeue_lock); } + +unlock: + mutex_unlock(&con->sock_mutex); } /* Connect a new socket to its peer */ @@ -1075,7 +1166,7 @@ static void tcp_connect_to_sock(struct connection *con) goto out_err; memset(&saddr, 0, sizeof(saddr)); - result = nodeid_to_addr(con->nodeid, &saddr, NULL); + result = nodeid_to_addr(con->nodeid, &saddr, NULL, false); if (result < 0) { log_print("no address for nodeid %d", con->nodeid); goto out_err; @@ -1254,6 +1345,7 @@ static int sctp_listen_for_all(void) int result = -EINVAL, num = 1, i, addr_len; struct connection *con = nodeid2con(0, GFP_NOFS); int bufsize = NEEDED_RMEM; + int one = 1; if (!con) return -ENOMEM; @@ -1288,6 +1380,11 @@ static int sctp_listen_for_all(void) goto create_delsock; } + result = kernel_setsockopt(sock, SOL_SCTP, SCTP_NODELAY, (char *)&one, + sizeof(one)); + if (result < 0) + log_print("Could not set SCTP NODELAY error %d\n", result); + /* Init con struct */ sock->sk->sk_user_data = con; con->sock = sock; @@ -1493,13 +1590,7 @@ static void send_to_sock(struct connection *con) } spin_lock(&con->writequeue_lock); - e->offset += ret; - e->len -= ret; - - if (e->len == 0 && e->users == 0) { - list_del(&e->list); - free_entry(e); - } + writequeue_entry_complete(e, ret); } spin_unlock(&con->writequeue_lock); out: diff --git a/fs/exec.c b/fs/exec.c index 0f6c96c57b2f..03b907cfd765 100644 --- a/fs/exec.c +++ b/fs/exec.c @@ -1137,13 +1137,6 @@ void setup_new_exec(struct linux_binprm * bprm) set_dumpable(current->mm, suid_dumpable); } - /* - * Flush performance counters when crossing a - * security domain: - */ - if (!get_dumpable(current->mm)) - perf_event_exit_task(current); - /* An exec changes our domain. We are no longer part of the thread group */ @@ -1207,6 +1200,15 @@ void install_exec_creds(struct linux_binprm *bprm) commit_creds(bprm->cred); bprm->cred = NULL; + + /* + * Disable monitoring for regular users + * when executing setuid binaries. Must + * wait until new credentials are committed + * by commit_creds() above + */ + if (get_dumpable(current->mm) != SUID_DUMP_USER) + perf_event_exit_task(current); /* * cred_guard_mutex must be held at least to this point to prevent * ptrace_attach() from altering our determination of the task's diff --git a/fs/exofs/inode.c b/fs/exofs/inode.c index d1f80abd8828..2ec8eb1ab269 100644 --- a/fs/exofs/inode.c +++ b/fs/exofs/inode.c @@ -953,9 +953,11 @@ static int exofs_releasepage(struct page *page, gfp_t gfp) return 0; } -static void exofs_invalidatepage(struct page *page, unsigned long offset) +static void exofs_invalidatepage(struct page *page, unsigned int offset, + unsigned int length) { - EXOFS_DBGMSG("page 0x%lx offset 0x%lx\n", page->index, offset); + EXOFS_DBGMSG("page 0x%lx offset 0x%x length 0x%x\n", + page->index, offset, length); WARN_ON(1); } diff --git a/fs/ext3/inode.c b/fs/ext3/inode.c index 23c712825640..f67668f724ba 100644 --- a/fs/ext3/inode.c +++ b/fs/ext3/inode.c @@ -1825,19 +1825,20 @@ ext3_readpages(struct file *file, struct address_space *mapping, return mpage_readpages(mapping, pages, nr_pages, ext3_get_block); } -static void ext3_invalidatepage(struct page *page, unsigned long offset) +static void ext3_invalidatepage(struct page *page, unsigned int offset, + unsigned int length) { journal_t *journal = EXT3_JOURNAL(page->mapping->host); - trace_ext3_invalidatepage(page, offset); + trace_ext3_invalidatepage(page, offset, length); /* * If it's a full truncate we just forget about the pending dirtying */ - if (offset == 0) + if (offset == 0 && length == PAGE_CACHE_SIZE) ClearPageChecked(page); - journal_invalidatepage(journal, page, offset); + journal_invalidatepage(journal, page, offset, length); } static int ext3_releasepage(struct page *page, gfp_t wait) diff --git a/fs/ext3/namei.c b/fs/ext3/namei.c index 7523c61f796c..998ea111e537 100644 --- a/fs/ext3/namei.c +++ b/fs/ext3/namei.c @@ -576,11 +576,8 @@ static int htree_dirblock_to_tree(struct file *dir_file, if (!ext3_check_dir_entry("htree_dirblock_to_tree", dir, de, bh, (block<<EXT3_BLOCK_SIZE_BITS(dir->i_sb)) +((char *)de - bh->b_data))) { - /* On error, skip the f_pos to the next block. */ - dir_file->f_pos = (dir_file->f_pos | - (dir->i_sb->s_blocksize - 1)) + 1; - brelse (bh); - return count; + /* silently ignore the rest of the block */ + break; } ext3fs_dirhash(de->name, de->name_len, hinfo); if ((hinfo->hash < start_hash) || diff --git a/fs/ext4/balloc.c b/fs/ext4/balloc.c index d0f13eada0ed..58339393fa6e 100644 --- a/fs/ext4/balloc.c +++ b/fs/ext4/balloc.c @@ -682,11 +682,15 @@ ext4_fsblk_t ext4_count_free_clusters(struct super_block *sb) static inline int test_root(ext4_group_t a, int b) { - int num = b; - - while (a > num) - num *= b; - return num == a; + while (1) { + if (a < b) + return 0; + if (a == b) + return 1; + if ((a % b) != 0) + return 0; + a = a / b; + } } static int ext4_group_sparse(ext4_group_t group) diff --git a/fs/ext4/ext4.h b/fs/ext4/ext4.h index 4af03ea84aa3..b577e45425b0 100644 --- a/fs/ext4/ext4.h +++ b/fs/ext4/ext4.h @@ -177,38 +177,28 @@ struct ext4_map_blocks { }; /* - * For delayed allocation tracking - */ -struct mpage_da_data { - struct inode *inode; - sector_t b_blocknr; /* start block number of extent */ - size_t b_size; /* size of extent */ - unsigned long b_state; /* state of the extent */ - unsigned long first_page, next_page; /* extent of pages */ - struct writeback_control *wbc; - int io_done; - int pages_written; - int retval; -}; - -/* * Flags for ext4_io_end->flags */ #define EXT4_IO_END_UNWRITTEN 0x0001 -#define EXT4_IO_END_ERROR 0x0002 -#define EXT4_IO_END_DIRECT 0x0004 +#define EXT4_IO_END_DIRECT 0x0002 /* - * For converting uninitialized extents on a work queue. + * For converting uninitialized extents on a work queue. 'handle' is used for + * buffered writeback. */ typedef struct ext4_io_end { struct list_head list; /* per-file finished IO list */ + handle_t *handle; /* handle reserved for extent + * conversion */ struct inode *inode; /* file being written to */ + struct bio *bio; /* Linked list of completed + * bios covering the extent */ unsigned int flag; /* unwritten or not */ loff_t offset; /* offset in the file */ ssize_t size; /* size of the extent */ struct kiocb *iocb; /* iocb struct for AIO */ int result; /* error value for AIO */ + atomic_t count; /* reference counter */ } ext4_io_end_t; struct ext4_io_submit { @@ -581,11 +571,6 @@ enum { #define EXT4_FREE_BLOCKS_NOFREE_LAST_CLUSTER 0x0020 /* - * Flags used by ext4_discard_partial_page_buffers - */ -#define EXT4_DISCARD_PARTIAL_PG_ZERO_UNMAPPED 0x0001 - -/* * ioctl commands */ #define EXT4_IOC_GETFLAGS FS_IOC_GETFLAGS @@ -879,6 +864,7 @@ struct ext4_inode_info { rwlock_t i_es_lock; struct list_head i_es_lru; unsigned int i_es_lru_nr; /* protected by i_es_lock */ + unsigned long i_touch_when; /* jiffies of last accessing */ /* ialloc */ ext4_group_t i_last_alloc_group; @@ -903,12 +889,22 @@ struct ext4_inode_info { qsize_t i_reserved_quota; #endif - /* completed IOs that might need unwritten extents handling */ - struct list_head i_completed_io_list; + /* Lock protecting lists below */ spinlock_t i_completed_io_lock; + /* + * Completed IOs that need unwritten extents handling and have + * transaction reserved + */ + struct list_head i_rsv_conversion_list; + /* + * Completed IOs that need unwritten extents handling and don't have + * transaction reserved + */ + struct list_head i_unrsv_conversion_list; atomic_t i_ioend_count; /* Number of outstanding io_end structs */ atomic_t i_unwritten; /* Nr. of inflight conversions pending */ - struct work_struct i_unwritten_work; /* deferred extent conversion */ + struct work_struct i_rsv_conversion_work; + struct work_struct i_unrsv_conversion_work; spinlock_t i_block_reservation_lock; @@ -1245,7 +1241,6 @@ struct ext4_sb_info { unsigned int s_mb_stats; unsigned int s_mb_order2_reqs; unsigned int s_mb_group_prealloc; - unsigned int s_max_writeback_mb_bump; unsigned int s_max_dir_size_kb; /* where last allocation was done - for stream allocation */ unsigned long s_mb_last_group; @@ -1281,8 +1276,10 @@ struct ext4_sb_info { struct flex_groups *s_flex_groups; ext4_group_t s_flex_groups_allocated; - /* workqueue for dio unwritten */ - struct workqueue_struct *dio_unwritten_wq; + /* workqueue for unreserved extent convertions (dio) */ + struct workqueue_struct *unrsv_conversion_wq; + /* workqueue for reserved extent conversions (buffered io) */ + struct workqueue_struct *rsv_conversion_wq; /* timer for periodic error stats printing */ struct timer_list s_err_report; @@ -1307,6 +1304,7 @@ struct ext4_sb_info { /* Reclaim extents from extent status tree */ struct shrinker s_es_shrinker; struct list_head s_es_lru; + unsigned long s_es_last_sorted; struct percpu_counter s_extent_cache_cnt; spinlock_t s_es_lru_lock ____cacheline_aligned_in_smp; }; @@ -1342,6 +1340,9 @@ static inline void ext4_set_io_unwritten_flag(struct inode *inode, struct ext4_io_end *io_end) { if (!(io_end->flag & EXT4_IO_END_UNWRITTEN)) { + /* Writeback has to have coversion transaction reserved */ + WARN_ON(EXT4_SB(inode->i_sb)->s_journal && !io_end->handle && + !(io_end->flag & EXT4_IO_END_DIRECT)); io_end->flag |= EXT4_IO_END_UNWRITTEN; atomic_inc(&EXT4_I(inode)->i_unwritten); } @@ -1999,7 +2000,6 @@ static inline unsigned char get_dtype(struct super_block *sb, int filetype) /* fsync.c */ extern int ext4_sync_file(struct file *, loff_t, loff_t, int); -extern int ext4_flush_unwritten_io(struct inode *); /* hash.c */ extern int ext4fs_dirhash(const char *name, int len, struct @@ -2088,7 +2088,7 @@ extern int ext4_change_inode_journal_flag(struct inode *, int); extern int ext4_get_inode_loc(struct inode *, struct ext4_iloc *); extern int ext4_can_truncate(struct inode *inode); extern void ext4_truncate(struct inode *); -extern int ext4_punch_hole(struct file *file, loff_t offset, loff_t length); +extern int ext4_punch_hole(struct inode *inode, loff_t offset, loff_t length); extern int ext4_truncate_restart_trans(handle_t *, struct inode *, int nblocks); extern void ext4_set_inode_flags(struct inode *); extern void ext4_get_inode_flags(struct ext4_inode_info *); @@ -2096,9 +2096,12 @@ extern int ext4_alloc_da_blocks(struct inode *inode); extern void ext4_set_aops(struct inode *inode); extern int ext4_writepage_trans_blocks(struct inode *); extern int ext4_chunk_trans_blocks(struct inode *, int nrblocks); -extern int ext4_discard_partial_page_buffers(handle_t *handle, - struct address_space *mapping, loff_t from, - loff_t length, int flags); +extern int ext4_block_truncate_page(handle_t *handle, + struct address_space *mapping, loff_t from); +extern int ext4_block_zero_page_range(handle_t *handle, + struct address_space *mapping, loff_t from, loff_t length); +extern int ext4_zero_partial_blocks(handle_t *handle, struct inode *inode, + loff_t lstart, loff_t lend); extern int ext4_page_mkwrite(struct vm_area_struct *vma, struct vm_fault *vmf); extern qsize_t *ext4_get_reserved_space(struct inode *inode); extern void ext4_da_update_reserve_space(struct inode *inode, @@ -2111,7 +2114,7 @@ extern ssize_t ext4_ind_direct_IO(int rw, struct kiocb *iocb, const struct iovec *iov, loff_t offset, unsigned long nr_segs); extern int ext4_ind_calc_metadata_amount(struct inode *inode, sector_t lblock); -extern int ext4_ind_trans_blocks(struct inode *inode, int nrblocks, int chunk); +extern int ext4_ind_trans_blocks(struct inode *inode, int nrblocks); extern void ext4_ind_truncate(handle_t *, struct inode *inode); extern int ext4_free_hole_blocks(handle_t *handle, struct inode *inode, ext4_lblk_t first, ext4_lblk_t stop); @@ -2166,42 +2169,96 @@ extern int ext4_alloc_flex_bg_array(struct super_block *sb, ext4_group_t ngroup); extern const char *ext4_decode_error(struct super_block *sb, int errno, char nbuf[16]); + extern __printf(4, 5) void __ext4_error(struct super_block *, const char *, unsigned int, const char *, ...); -#define ext4_error(sb, message...) __ext4_error(sb, __func__, \ - __LINE__, ## message) extern __printf(5, 6) -void ext4_error_inode(struct inode *, const char *, unsigned int, ext4_fsblk_t, +void __ext4_error_inode(struct inode *, const char *, unsigned int, ext4_fsblk_t, const char *, ...); extern __printf(5, 6) -void ext4_error_file(struct file *, const char *, unsigned int, ext4_fsblk_t, +void __ext4_error_file(struct file *, const char *, unsigned int, ext4_fsblk_t, const char *, ...); extern void __ext4_std_error(struct super_block *, const char *, unsigned int, int); extern __printf(4, 5) void __ext4_abort(struct super_block *, const char *, unsigned int, const char *, ...); -#define ext4_abort(sb, message...) __ext4_abort(sb, __func__, \ - __LINE__, ## message) extern __printf(4, 5) void __ext4_warning(struct super_block *, const char *, unsigned int, const char *, ...); -#define ext4_warning(sb, message...) __ext4_warning(sb, __func__, \ - __LINE__, ## message) extern __printf(3, 4) -void ext4_msg(struct super_block *, const char *, const char *, ...); +void __ext4_msg(struct super_block *, const char *, const char *, ...); extern void __dump_mmp_msg(struct super_block *, struct mmp_struct *mmp, const char *, unsigned int, const char *); -#define dump_mmp_msg(sb, mmp, msg) __dump_mmp_msg(sb, mmp, __func__, \ - __LINE__, msg) extern __printf(7, 8) void __ext4_grp_locked_error(const char *, unsigned int, struct super_block *, ext4_group_t, unsigned long, ext4_fsblk_t, const char *, ...); -#define ext4_grp_locked_error(sb, grp, message...) \ - __ext4_grp_locked_error(__func__, __LINE__, (sb), (grp), ## message) + +#ifdef CONFIG_PRINTK + +#define ext4_error_inode(inode, func, line, block, fmt, ...) \ + __ext4_error_inode(inode, func, line, block, fmt, ##__VA_ARGS__) +#define ext4_error_file(file, func, line, block, fmt, ...) \ + __ext4_error_file(file, func, line, block, fmt, ##__VA_ARGS__) +#define ext4_error(sb, fmt, ...) \ + __ext4_error(sb, __func__, __LINE__, fmt, ##__VA_ARGS__) +#define ext4_abort(sb, fmt, ...) \ + __ext4_abort(sb, __func__, __LINE__, fmt, ##__VA_ARGS__) +#define ext4_warning(sb, fmt, ...) \ + __ext4_warning(sb, __func__, __LINE__, fmt, ##__VA_ARGS__) +#define ext4_msg(sb, level, fmt, ...) \ + __ext4_msg(sb, level, fmt, ##__VA_ARGS__) +#define dump_mmp_msg(sb, mmp, msg) \ + __dump_mmp_msg(sb, mmp, __func__, __LINE__, msg) +#define ext4_grp_locked_error(sb, grp, ino, block, fmt, ...) \ + __ext4_grp_locked_error(__func__, __LINE__, sb, grp, ino, block, \ + fmt, ##__VA_ARGS__) + +#else + +#define ext4_error_inode(inode, func, line, block, fmt, ...) \ +do { \ + no_printk(fmt, ##__VA_ARGS__); \ + __ext4_error_inode(inode, "", 0, block, " "); \ +} while (0) +#define ext4_error_file(file, func, line, block, fmt, ...) \ +do { \ + no_printk(fmt, ##__VA_ARGS__); \ + __ext4_error_file(file, "", 0, block, " "); \ +} while (0) +#define ext4_error(sb, fmt, ...) \ +do { \ + no_printk(fmt, ##__VA_ARGS__); \ + __ext4_error(sb, "", 0, " "); \ +} while (0) +#define ext4_abort(sb, fmt, ...) \ +do { \ + no_printk(fmt, ##__VA_ARGS__); \ + __ext4_abort(sb, "", 0, " "); \ +} while (0) +#define ext4_warning(sb, fmt, ...) \ +do { \ + no_printk(fmt, ##__VA_ARGS__); \ + __ext4_warning(sb, "", 0, " "); \ +} while (0) +#define ext4_msg(sb, level, fmt, ...) \ +do { \ + no_printk(fmt, ##__VA_ARGS__); \ + __ext4_msg(sb, "", " "); \ +} while (0) +#define dump_mmp_msg(sb, mmp, msg) \ + __dump_mmp_msg(sb, mmp, "", 0, "") +#define ext4_grp_locked_error(sb, grp, ino, block, fmt, ...) \ +do { \ + no_printk(fmt, ##__VA_ARGS__); \ + __ext4_grp_locked_error("", 0, sb, grp, ino, block, " "); \ +} while (0) + +#endif + extern void ext4_update_dynamic_rev(struct super_block *sb); extern int ext4_update_compat_feature(handle_t *handle, struct super_block *sb, __u32 compat); @@ -2312,6 +2369,7 @@ struct ext4_group_info *ext4_get_group_info(struct super_block *sb, { struct ext4_group_info ***grp_info; long indexv, indexh; + BUG_ON(group >= EXT4_SB(sb)->s_groups_count); grp_info = EXT4_SB(sb)->s_group_info; indexv = group >> (EXT4_DESC_PER_BLOCK_BITS(sb)); indexh = group & ((EXT4_DESC_PER_BLOCK(sb)) - 1); @@ -2598,8 +2656,7 @@ struct ext4_extent; extern int ext4_ext_tree_init(handle_t *handle, struct inode *); extern int ext4_ext_writepage_trans_blocks(struct inode *, int); -extern int ext4_ext_index_trans_blocks(struct inode *inode, int nrblocks, - int chunk); +extern int ext4_ext_index_trans_blocks(struct inode *inode, int extents); extern int ext4_ext_map_blocks(handle_t *handle, struct inode *inode, struct ext4_map_blocks *map, int flags); extern void ext4_ext_truncate(handle_t *, struct inode *); @@ -2609,8 +2666,8 @@ extern void ext4_ext_init(struct super_block *); extern void ext4_ext_release(struct super_block *); extern long ext4_fallocate(struct file *file, int mode, loff_t offset, loff_t len); -extern int ext4_convert_unwritten_extents(struct inode *inode, loff_t offset, - ssize_t len); +extern int ext4_convert_unwritten_extents(handle_t *handle, struct inode *inode, + loff_t offset, ssize_t len); extern int ext4_map_blocks(handle_t *handle, struct inode *inode, struct ext4_map_blocks *map, int flags); extern int ext4_ext_calc_metadata_amount(struct inode *inode, @@ -2650,12 +2707,15 @@ extern int ext4_move_extents(struct file *o_filp, struct file *d_filp, /* page-io.c */ extern int __init ext4_init_pageio(void); -extern void ext4_add_complete_io(ext4_io_end_t *io_end); extern void ext4_exit_pageio(void); -extern void ext4_ioend_shutdown(struct inode *); -extern void ext4_free_io_end(ext4_io_end_t *io); extern ext4_io_end_t *ext4_init_io_end(struct inode *inode, gfp_t flags); -extern void ext4_end_io_work(struct work_struct *work); +extern ext4_io_end_t *ext4_get_io_end(ext4_io_end_t *io_end); +extern int ext4_put_io_end(ext4_io_end_t *io_end); +extern void ext4_put_io_end_defer(ext4_io_end_t *io_end); +extern void ext4_io_submit_init(struct ext4_io_submit *io, + struct writeback_control *wbc); +extern void ext4_end_io_rsv_work(struct work_struct *work); +extern void ext4_end_io_unrsv_work(struct work_struct *work); extern void ext4_io_submit(struct ext4_io_submit *io); extern int ext4_bio_write_page(struct ext4_io_submit *io, struct page *page, @@ -2668,20 +2728,17 @@ extern void ext4_mmp_csum_set(struct super_block *sb, struct mmp_struct *mmp); extern int ext4_mmp_csum_verify(struct super_block *sb, struct mmp_struct *mmp); -/* BH_Uninit flag: blocks are allocated but uninitialized on disk */ +/* + * Note that these flags will never ever appear in a buffer_head's state flag. + * See EXT4_MAP_... to see where this is used. + */ enum ext4_state_bits { BH_Uninit /* blocks are allocated but uninitialized on disk */ - = BH_JBDPrivateStart, + = BH_JBDPrivateStart, BH_AllocFromCluster, /* allocated blocks were part of already - * allocated cluster. Note that this flag will - * never, ever appear in a buffer_head's state - * flag. See EXT4_MAP_FROM_CLUSTER to see where - * this is used. */ + * allocated cluster. */ }; -BUFFER_FNS(Uninit, uninit) -TAS_BUFFER_FNS(Uninit, uninit) - /* * Add new method to test whether block and inode bitmaps are properly * initialized. With uninit_bg reading the block from disk is not enough diff --git a/fs/ext4/ext4_jbd2.c b/fs/ext4/ext4_jbd2.c index 451eb4045330..72a3600aedbd 100644 --- a/fs/ext4/ext4_jbd2.c +++ b/fs/ext4/ext4_jbd2.c @@ -38,31 +38,43 @@ static void ext4_put_nojournal(handle_t *handle) /* * Wrappers for jbd2_journal_start/end. */ -handle_t *__ext4_journal_start_sb(struct super_block *sb, unsigned int line, - int type, int nblocks) +static int ext4_journal_check_start(struct super_block *sb) { journal_t *journal; might_sleep(); - - trace_ext4_journal_start(sb, nblocks, _RET_IP_); if (sb->s_flags & MS_RDONLY) - return ERR_PTR(-EROFS); - + return -EROFS; WARN_ON(sb->s_writers.frozen == SB_FREEZE_COMPLETE); journal = EXT4_SB(sb)->s_journal; - if (!journal) - return ext4_get_nojournal(); /* * Special case here: if the journal has aborted behind our * backs (eg. EIO in the commit thread), then we still need to * take the FS itself readonly cleanly. */ - if (is_journal_aborted(journal)) { + if (journal && is_journal_aborted(journal)) { ext4_abort(sb, "Detected aborted journal"); - return ERR_PTR(-EROFS); + return -EROFS; } - return jbd2__journal_start(journal, nblocks, GFP_NOFS, type, line); + return 0; +} + +handle_t *__ext4_journal_start_sb(struct super_block *sb, unsigned int line, + int type, int blocks, int rsv_blocks) +{ + journal_t *journal; + int err; + + trace_ext4_journal_start(sb, blocks, rsv_blocks, _RET_IP_); + err = ext4_journal_check_start(sb); + if (err < 0) + return ERR_PTR(err); + + journal = EXT4_SB(sb)->s_journal; + if (!journal) + return ext4_get_nojournal(); + return jbd2__journal_start(journal, blocks, rsv_blocks, GFP_NOFS, + type, line); } int __ext4_journal_stop(const char *where, unsigned int line, handle_t *handle) @@ -86,6 +98,30 @@ int __ext4_journal_stop(const char *where, unsigned int line, handle_t *handle) return err; } +handle_t *__ext4_journal_start_reserved(handle_t *handle, unsigned int line, + int type) +{ + struct super_block *sb; + int err; + + if (!ext4_handle_valid(handle)) + return ext4_get_nojournal(); + + sb = handle->h_journal->j_private; + trace_ext4_journal_start_reserved(sb, handle->h_buffer_credits, + _RET_IP_); + err = ext4_journal_check_start(sb); + if (err < 0) { + jbd2_journal_free_reserved(handle); + return ERR_PTR(err); + } + + err = jbd2_journal_start_reserved(handle, type, line); + if (err < 0) + return ERR_PTR(err); + return handle; +} + void ext4_journal_abort_handle(const char *caller, unsigned int line, const char *err_fn, struct buffer_head *bh, handle_t *handle, int err) diff --git a/fs/ext4/ext4_jbd2.h b/fs/ext4/ext4_jbd2.h index c8c6885406db..2877258d9497 100644 --- a/fs/ext4/ext4_jbd2.h +++ b/fs/ext4/ext4_jbd2.h @@ -134,7 +134,8 @@ static inline int ext4_jbd2_credits_xattr(struct inode *inode) #define EXT4_HT_MIGRATE 8 #define EXT4_HT_MOVE_EXTENTS 9 #define EXT4_HT_XATTR 10 -#define EXT4_HT_MAX 11 +#define EXT4_HT_EXT_CONVERT 11 +#define EXT4_HT_MAX 12 /** * struct ext4_journal_cb_entry - Base structure for callback information. @@ -265,7 +266,7 @@ int __ext4_handle_dirty_super(const char *where, unsigned int line, __ext4_handle_dirty_super(__func__, __LINE__, (handle), (sb)) handle_t *__ext4_journal_start_sb(struct super_block *sb, unsigned int line, - int type, int nblocks); + int type, int blocks, int rsv_blocks); int __ext4_journal_stop(const char *where, unsigned int line, handle_t *handle); #define EXT4_NOJOURNAL_MAX_REF_COUNT ((unsigned long) 4096) @@ -300,21 +301,37 @@ static inline int ext4_handle_has_enough_credits(handle_t *handle, int needed) } #define ext4_journal_start_sb(sb, type, nblocks) \ - __ext4_journal_start_sb((sb), __LINE__, (type), (nblocks)) + __ext4_journal_start_sb((sb), __LINE__, (type), (nblocks), 0) #define ext4_journal_start(inode, type, nblocks) \ - __ext4_journal_start((inode), __LINE__, (type), (nblocks)) + __ext4_journal_start((inode), __LINE__, (type), (nblocks), 0) + +#define ext4_journal_start_with_reserve(inode, type, blocks, rsv_blocks) \ + __ext4_journal_start((inode), __LINE__, (type), (blocks), (rsv_blocks)) static inline handle_t *__ext4_journal_start(struct inode *inode, unsigned int line, int type, - int nblocks) + int blocks, int rsv_blocks) { - return __ext4_journal_start_sb(inode->i_sb, line, type, nblocks); + return __ext4_journal_start_sb(inode->i_sb, line, type, blocks, + rsv_blocks); } #define ext4_journal_stop(handle) \ __ext4_journal_stop(__func__, __LINE__, (handle)) +#define ext4_journal_start_reserved(handle, type) \ + __ext4_journal_start_reserved((handle), __LINE__, (type)) + +handle_t *__ext4_journal_start_reserved(handle_t *handle, unsigned int line, + int type); + +static inline void ext4_journal_free_reserved(handle_t *handle) +{ + if (ext4_handle_valid(handle)) + jbd2_journal_free_reserved(handle); +} + static inline handle_t *ext4_journal_current_handle(void) { return journal_current_handle(); diff --git a/fs/ext4/extents.c b/fs/ext4/extents.c index bc0f1910b9cf..7097b0f680e6 100644 --- a/fs/ext4/extents.c +++ b/fs/ext4/extents.c @@ -2125,7 +2125,8 @@ static int ext4_fill_fiemap_extents(struct inode *inode, next_del = ext4_find_delayed_extent(inode, &es); if (!exists && next_del) { exists = 1; - flags |= FIEMAP_EXTENT_DELALLOC; + flags |= (FIEMAP_EXTENT_DELALLOC | + FIEMAP_EXTENT_UNKNOWN); } up_read(&EXT4_I(inode)->i_data_sem); @@ -2328,17 +2329,15 @@ int ext4_ext_calc_credits_for_single_extent(struct inode *inode, int nrblocks, } /* - * How many index/leaf blocks need to change/allocate to modify nrblocks? + * How many index/leaf blocks need to change/allocate to add @extents extents? * - * if nrblocks are fit in a single extent (chunk flag is 1), then - * in the worse case, each tree level index/leaf need to be changed - * if the tree split due to insert a new extent, then the old tree - * index/leaf need to be updated too + * If we add a single extent, then in the worse case, each tree level + * index/leaf need to be changed in case of the tree split. * - * If the nrblocks are discontiguous, they could cause - * the whole tree split more than once, but this is really rare. + * If more extents are inserted, they could cause the whole tree split more + * than once, but this is really rare. */ -int ext4_ext_index_trans_blocks(struct inode *inode, int nrblocks, int chunk) +int ext4_ext_index_trans_blocks(struct inode *inode, int extents) { int index; int depth; @@ -2349,7 +2348,7 @@ int ext4_ext_index_trans_blocks(struct inode *inode, int nrblocks, int chunk) depth = ext_depth(inode); - if (chunk) + if (extents <= 1) index = depth * 2; else index = depth * 3; @@ -2357,20 +2356,24 @@ int ext4_ext_index_trans_blocks(struct inode *inode, int nrblocks, int chunk) return index; } +static inline int get_default_free_blocks_flags(struct inode *inode) +{ + if (S_ISDIR(inode->i_mode) || S_ISLNK(inode->i_mode)) + return EXT4_FREE_BLOCKS_METADATA | EXT4_FREE_BLOCKS_FORGET; + else if (ext4_should_journal_data(inode)) + return EXT4_FREE_BLOCKS_FORGET; + return 0; +} + static int ext4_remove_blocks(handle_t *handle, struct inode *inode, struct ext4_extent *ex, - ext4_fsblk_t *partial_cluster, + long long *partial_cluster, ext4_lblk_t from, ext4_lblk_t to) { struct ext4_sb_info *sbi = EXT4_SB(inode->i_sb); unsigned short ee_len = ext4_ext_get_actual_len(ex); ext4_fsblk_t pblk; - int flags = 0; - - if (S_ISDIR(inode->i_mode) || S_ISLNK(inode->i_mode)) - flags |= EXT4_FREE_BLOCKS_METADATA | EXT4_FREE_BLOCKS_FORGET; - else if (ext4_should_journal_data(inode)) - flags |= EXT4_FREE_BLOCKS_FORGET; + int flags = get_default_free_blocks_flags(inode); /* * For bigalloc file systems, we never free a partial cluster @@ -2388,7 +2391,8 @@ static int ext4_remove_blocks(handle_t *handle, struct inode *inode, * partial cluster here. */ pblk = ext4_ext_pblock(ex) + ee_len - 1; - if (*partial_cluster && (EXT4_B2C(sbi, pblk) != *partial_cluster)) { + if ((*partial_cluster > 0) && + (EXT4_B2C(sbi, pblk) != *partial_cluster)) { ext4_free_blocks(handle, inode, NULL, EXT4_C2B(sbi, *partial_cluster), sbi->s_cluster_ratio, flags); @@ -2414,41 +2418,46 @@ static int ext4_remove_blocks(handle_t *handle, struct inode *inode, && to == le32_to_cpu(ex->ee_block) + ee_len - 1) { /* tail removal */ ext4_lblk_t num; + unsigned int unaligned; num = le32_to_cpu(ex->ee_block) + ee_len - from; pblk = ext4_ext_pblock(ex) + ee_len - num; - ext_debug("free last %u blocks starting %llu\n", num, pblk); + /* + * Usually we want to free partial cluster at the end of the + * extent, except for the situation when the cluster is still + * used by any other extent (partial_cluster is negative). + */ + if (*partial_cluster < 0 && + -(*partial_cluster) == EXT4_B2C(sbi, pblk + num - 1)) + flags |= EXT4_FREE_BLOCKS_NOFREE_LAST_CLUSTER; + + ext_debug("free last %u blocks starting %llu partial %lld\n", + num, pblk, *partial_cluster); ext4_free_blocks(handle, inode, NULL, pblk, num, flags); /* * If the block range to be freed didn't start at the * beginning of a cluster, and we removed the entire - * extent, save the partial cluster here, since we - * might need to delete if we determine that the - * truncate operation has removed all of the blocks in - * the cluster. + * extent and the cluster is not used by any other extent, + * save the partial cluster here, since we might need to + * delete if we determine that the truncate operation has + * removed all of the blocks in the cluster. + * + * On the other hand, if we did not manage to free the whole + * extent, we have to mark the cluster as used (store negative + * cluster number in partial_cluster). */ - if (pblk & (sbi->s_cluster_ratio - 1) && - (ee_len == num)) + unaligned = pblk & (sbi->s_cluster_ratio - 1); + if (unaligned && (ee_len == num) && + (*partial_cluster != -((long long)EXT4_B2C(sbi, pblk)))) *partial_cluster = EXT4_B2C(sbi, pblk); - else + else if (unaligned) + *partial_cluster = -((long long)EXT4_B2C(sbi, pblk)); + else if (*partial_cluster > 0) *partial_cluster = 0; - } else if (from == le32_to_cpu(ex->ee_block) - && to <= le32_to_cpu(ex->ee_block) + ee_len - 1) { - /* head removal */ - ext4_lblk_t num; - ext4_fsblk_t start; - - num = to - from; - start = ext4_ext_pblock(ex); - - ext_debug("free first %u blocks starting %llu\n", num, start); - ext4_free_blocks(handle, inode, NULL, start, num, flags); - - } else { - printk(KERN_INFO "strange request: removal(2) " - "%u-%u from %u:%u\n", - from, to, le32_to_cpu(ex->ee_block), ee_len); - } + } else + ext4_error(sbi->s_sb, "strange request: removal(2) " + "%u-%u from %u:%u\n", + from, to, le32_to_cpu(ex->ee_block), ee_len); return 0; } @@ -2461,12 +2470,16 @@ static int ext4_remove_blocks(handle_t *handle, struct inode *inode, * @handle: The journal handle * @inode: The files inode * @path: The path to the leaf + * @partial_cluster: The cluster which we'll have to free if all extents + * has been released from it. It gets negative in case + * that the cluster is still used. * @start: The first block to remove * @end: The last block to remove */ static int ext4_ext_rm_leaf(handle_t *handle, struct inode *inode, - struct ext4_ext_path *path, ext4_fsblk_t *partial_cluster, + struct ext4_ext_path *path, + long long *partial_cluster, ext4_lblk_t start, ext4_lblk_t end) { struct ext4_sb_info *sbi = EXT4_SB(inode->i_sb); @@ -2479,6 +2492,7 @@ ext4_ext_rm_leaf(handle_t *handle, struct inode *inode, unsigned short ex_ee_len; unsigned uninitialized = 0; struct ext4_extent *ex; + ext4_fsblk_t pblk; /* the header must be checked already in ext4_ext_remove_space() */ ext_debug("truncate since %u in leaf to %u\n", start, end); @@ -2490,7 +2504,9 @@ ext4_ext_rm_leaf(handle_t *handle, struct inode *inode, return -EIO; } /* find where to start removing */ - ex = EXT_LAST_EXTENT(eh); + ex = path[depth].p_ext; + if (!ex) + ex = EXT_LAST_EXTENT(eh); ex_ee_block = le32_to_cpu(ex->ee_block); ex_ee_len = ext4_ext_get_actual_len(ex); @@ -2517,6 +2533,16 @@ ext4_ext_rm_leaf(handle_t *handle, struct inode *inode, /* If this extent is beyond the end of the hole, skip it */ if (end < ex_ee_block) { + /* + * We're going to skip this extent and move to another, + * so if this extent is not cluster aligned we have + * to mark the current cluster as used to avoid + * accidentally freeing it later on + */ + pblk = ext4_ext_pblock(ex); + if (pblk & (sbi->s_cluster_ratio - 1)) + *partial_cluster = + -((long long)EXT4_B2C(sbi, pblk)); ex--; ex_ee_block = le32_to_cpu(ex->ee_block); ex_ee_len = ext4_ext_get_actual_len(ex); @@ -2592,7 +2618,7 @@ ext4_ext_rm_leaf(handle_t *handle, struct inode *inode, sizeof(struct ext4_extent)); } le16_add_cpu(&eh->eh_entries, -1); - } else + } else if (*partial_cluster > 0) *partial_cluster = 0; err = ext4_ext_dirty(handle, inode, path + depth); @@ -2610,17 +2636,13 @@ ext4_ext_rm_leaf(handle_t *handle, struct inode *inode, err = ext4_ext_correct_indexes(handle, inode, path); /* - * If there is still a entry in the leaf node, check to see if - * it references the partial cluster. This is the only place - * where it could; if it doesn't, we can free the cluster. + * Free the partial cluster only if the current extent does not + * reference it. Otherwise we might free used cluster. */ - if (*partial_cluster && ex >= EXT_FIRST_EXTENT(eh) && + if (*partial_cluster > 0 && (EXT4_B2C(sbi, ext4_ext_pblock(ex) + ex_ee_len - 1) != *partial_cluster)) { - int flags = EXT4_FREE_BLOCKS_FORGET; - - if (S_ISDIR(inode->i_mode) || S_ISLNK(inode->i_mode)) - flags |= EXT4_FREE_BLOCKS_METADATA; + int flags = get_default_free_blocks_flags(inode); ext4_free_blocks(handle, inode, NULL, EXT4_C2B(sbi, *partial_cluster), @@ -2664,7 +2686,7 @@ int ext4_ext_remove_space(struct inode *inode, ext4_lblk_t start, struct super_block *sb = inode->i_sb; int depth = ext_depth(inode); struct ext4_ext_path *path = NULL; - ext4_fsblk_t partial_cluster = 0; + long long partial_cluster = 0; handle_t *handle; int i = 0, err = 0; @@ -2676,7 +2698,7 @@ int ext4_ext_remove_space(struct inode *inode, ext4_lblk_t start, return PTR_ERR(handle); again: - trace_ext4_ext_remove_space(inode, start, depth); + trace_ext4_ext_remove_space(inode, start, end, depth); /* * Check if we are removing extents inside the extent tree. If that @@ -2844,17 +2866,14 @@ again: } } - trace_ext4_ext_remove_space_done(inode, start, depth, partial_cluster, - path->p_hdr->eh_entries); + trace_ext4_ext_remove_space_done(inode, start, end, depth, + partial_cluster, path->p_hdr->eh_entries); /* If we still have something in the partial cluster and we have removed * even the first extent, then we should free the blocks in the partial * cluster as well. */ - if (partial_cluster && path->p_hdr->eh_entries == 0) { - int flags = EXT4_FREE_BLOCKS_FORGET; - - if (S_ISDIR(inode->i_mode) || S_ISLNK(inode->i_mode)) - flags |= EXT4_FREE_BLOCKS_METADATA; + if (partial_cluster > 0 && path->p_hdr->eh_entries == 0) { + int flags = get_default_free_blocks_flags(inode); ext4_free_blocks(handle, inode, NULL, EXT4_C2B(EXT4_SB(sb), partial_cluster), @@ -4363,7 +4382,7 @@ out2: } out3: - trace_ext4_ext_map_blocks_exit(inode, map, err ? err : allocated); + trace_ext4_ext_map_blocks_exit(inode, flags, map, err ? err : allocated); return err ? err : allocated; } @@ -4446,7 +4465,7 @@ long ext4_fallocate(struct file *file, int mode, loff_t offset, loff_t len) return -EOPNOTSUPP; if (mode & FALLOC_FL_PUNCH_HOLE) - return ext4_punch_hole(file, offset, len); + return ext4_punch_hole(inode, offset, len); ret = ext4_convert_inline_data(inode); if (ret) @@ -4548,10 +4567,9 @@ retry: * function, to convert the fallocated extents after IO is completed. * Returns 0 on success. */ -int ext4_convert_unwritten_extents(struct inode *inode, loff_t offset, - ssize_t len) +int ext4_convert_unwritten_extents(handle_t *handle, struct inode *inode, + loff_t offset, ssize_t len) { - handle_t *handle; unsigned int max_blocks; int ret = 0; int ret2 = 0; @@ -4566,16 +4584,32 @@ int ext4_convert_unwritten_extents(struct inode *inode, loff_t offset, max_blocks = ((EXT4_BLOCK_ALIGN(len + offset, blkbits) >> blkbits) - map.m_lblk); /* - * credits to insert 1 extent into extent tree + * This is somewhat ugly but the idea is clear: When transaction is + * reserved, everything goes into it. Otherwise we rather start several + * smaller transactions for conversion of each extent separately. */ - credits = ext4_chunk_trans_blocks(inode, max_blocks); + if (handle) { + handle = ext4_journal_start_reserved(handle, + EXT4_HT_EXT_CONVERT); + if (IS_ERR(handle)) + return PTR_ERR(handle); + credits = 0; + } else { + /* + * credits to insert 1 extent into extent tree + */ + credits = ext4_chunk_trans_blocks(inode, max_blocks); + } while (ret >= 0 && ret < max_blocks) { map.m_lblk += ret; map.m_len = (max_blocks -= ret); - handle = ext4_journal_start(inode, EXT4_HT_MAP_BLOCKS, credits); - if (IS_ERR(handle)) { - ret = PTR_ERR(handle); - break; + if (credits) { + handle = ext4_journal_start(inode, EXT4_HT_MAP_BLOCKS, + credits); + if (IS_ERR(handle)) { + ret = PTR_ERR(handle); + break; + } } ret = ext4_map_blocks(handle, inode, &map, EXT4_GET_BLOCKS_IO_CONVERT_EXT); @@ -4586,10 +4620,13 @@ int ext4_convert_unwritten_extents(struct inode *inode, loff_t offset, inode->i_ino, map.m_lblk, map.m_len, ret); ext4_mark_inode_dirty(handle, inode); - ret2 = ext4_journal_stop(handle); - if (ret <= 0 || ret2 ) + if (credits) + ret2 = ext4_journal_stop(handle); + if (ret <= 0 || ret2) break; } + if (!credits) + ret2 = ext4_journal_stop(handle); return ret > 0 ? ret2 : ret; } @@ -4659,7 +4696,7 @@ static int ext4_xattr_fiemap(struct inode *inode, error = ext4_get_inode_loc(inode, &iloc); if (error) return error; - physical = iloc.bh->b_blocknr << blockbits; + physical = (__u64)iloc.bh->b_blocknr << blockbits; offset = EXT4_GOOD_OLD_INODE_SIZE + EXT4_I(inode)->i_extra_isize; physical += offset; @@ -4667,7 +4704,7 @@ static int ext4_xattr_fiemap(struct inode *inode, flags |= FIEMAP_EXTENT_DATA_INLINE; brelse(iloc.bh); } else { /* external block */ - physical = EXT4_I(inode)->i_file_acl << blockbits; + physical = (__u64)EXT4_I(inode)->i_file_acl << blockbits; length = inode->i_sb->s_blocksize; } diff --git a/fs/ext4/extents_status.c b/fs/ext4/extents_status.c index e6941e622d31..ee018d5f397e 100644 --- a/fs/ext4/extents_status.c +++ b/fs/ext4/extents_status.c @@ -10,6 +10,7 @@ * Ext4 extents status tree core functions. */ #include <linux/rbtree.h> +#include <linux/list_sort.h> #include "ext4.h" #include "extents_status.h" #include "ext4_extents.h" @@ -291,7 +292,6 @@ out: read_unlock(&EXT4_I(inode)->i_es_lock); - ext4_es_lru_add(inode); trace_ext4_es_find_delayed_extent_range_exit(inode, es); } @@ -672,7 +672,6 @@ int ext4_es_insert_extent(struct inode *inode, ext4_lblk_t lblk, error: write_unlock(&EXT4_I(inode)->i_es_lock); - ext4_es_lru_add(inode); ext4_es_print_tree(inode); return err; @@ -734,7 +733,6 @@ out: read_unlock(&EXT4_I(inode)->i_es_lock); - ext4_es_lru_add(inode); trace_ext4_es_lookup_extent_exit(inode, es, found); return found; } @@ -878,12 +876,28 @@ int ext4_es_zeroout(struct inode *inode, struct ext4_extent *ex) EXTENT_STATUS_WRITTEN); } +static int ext4_inode_touch_time_cmp(void *priv, struct list_head *a, + struct list_head *b) +{ + struct ext4_inode_info *eia, *eib; + eia = list_entry(a, struct ext4_inode_info, i_es_lru); + eib = list_entry(b, struct ext4_inode_info, i_es_lru); + + if (eia->i_touch_when == eib->i_touch_when) + return 0; + if (time_after(eia->i_touch_when, eib->i_touch_when)) + return 1; + else + return -1; +} + static int ext4_es_shrink(struct shrinker *shrink, struct shrink_control *sc) { struct ext4_sb_info *sbi = container_of(shrink, struct ext4_sb_info, s_es_shrinker); struct ext4_inode_info *ei; - struct list_head *cur, *tmp, scanned; + struct list_head *cur, *tmp; + LIST_HEAD(skiped); int nr_to_scan = sc->nr_to_scan; int ret, nr_shrunk = 0; @@ -893,23 +907,41 @@ static int ext4_es_shrink(struct shrinker *shrink, struct shrink_control *sc) if (!nr_to_scan) return ret; - INIT_LIST_HEAD(&scanned); - spin_lock(&sbi->s_es_lru_lock); + + /* + * If the inode that is at the head of LRU list is newer than + * last_sorted time, that means that we need to sort this list. + */ + ei = list_first_entry(&sbi->s_es_lru, struct ext4_inode_info, i_es_lru); + if (sbi->s_es_last_sorted < ei->i_touch_when) { + list_sort(NULL, &sbi->s_es_lru, ext4_inode_touch_time_cmp); + sbi->s_es_last_sorted = jiffies; + } + list_for_each_safe(cur, tmp, &sbi->s_es_lru) { - list_move_tail(cur, &scanned); + /* + * If we have already reclaimed all extents from extent + * status tree, just stop the loop immediately. + */ + if (percpu_counter_read_positive(&sbi->s_extent_cache_cnt) == 0) + break; ei = list_entry(cur, struct ext4_inode_info, i_es_lru); - read_lock(&ei->i_es_lock); - if (ei->i_es_lru_nr == 0) { - read_unlock(&ei->i_es_lock); + /* Skip the inode that is newer than the last_sorted time */ + if (sbi->s_es_last_sorted < ei->i_touch_when) { + list_move_tail(cur, &skiped); continue; } - read_unlock(&ei->i_es_lock); + + if (ei->i_es_lru_nr == 0) + continue; write_lock(&ei->i_es_lock); ret = __es_try_to_reclaim_extents(ei, nr_to_scan); + if (ei->i_es_lru_nr == 0) + list_del_init(&ei->i_es_lru); write_unlock(&ei->i_es_lock); nr_shrunk += ret; @@ -917,7 +949,9 @@ static int ext4_es_shrink(struct shrinker *shrink, struct shrink_control *sc) if (nr_to_scan == 0) break; } - list_splice_tail(&scanned, &sbi->s_es_lru); + + /* Move the newer inodes into the tail of the LRU list. */ + list_splice_tail(&skiped, &sbi->s_es_lru); spin_unlock(&sbi->s_es_lru_lock); ret = percpu_counter_read_positive(&sbi->s_extent_cache_cnt); @@ -925,21 +959,19 @@ static int ext4_es_shrink(struct shrinker *shrink, struct shrink_control *sc) return ret; } -void ext4_es_register_shrinker(struct super_block *sb) +void ext4_es_register_shrinker(struct ext4_sb_info *sbi) { - struct ext4_sb_info *sbi; - - sbi = EXT4_SB(sb); INIT_LIST_HEAD(&sbi->s_es_lru); spin_lock_init(&sbi->s_es_lru_lock); + sbi->s_es_last_sorted = 0; sbi->s_es_shrinker.shrink = ext4_es_shrink; sbi->s_es_shrinker.seeks = DEFAULT_SEEKS; register_shrinker(&sbi->s_es_shrinker); } -void ext4_es_unregister_shrinker(struct super_block *sb) +void ext4_es_unregister_shrinker(struct ext4_sb_info *sbi) { - unregister_shrinker(&EXT4_SB(sb)->s_es_shrinker); + unregister_shrinker(&sbi->s_es_shrinker); } void ext4_es_lru_add(struct inode *inode) @@ -947,11 +979,14 @@ void ext4_es_lru_add(struct inode *inode) struct ext4_inode_info *ei = EXT4_I(inode); struct ext4_sb_info *sbi = EXT4_SB(inode->i_sb); + ei->i_touch_when = jiffies; + + if (!list_empty(&ei->i_es_lru)) + return; + spin_lock(&sbi->s_es_lru_lock); if (list_empty(&ei->i_es_lru)) list_add_tail(&ei->i_es_lru, &sbi->s_es_lru); - else - list_move_tail(&ei->i_es_lru, &sbi->s_es_lru); spin_unlock(&sbi->s_es_lru_lock); } diff --git a/fs/ext4/extents_status.h b/fs/ext4/extents_status.h index f740eb03b707..e936730cc5b0 100644 --- a/fs/ext4/extents_status.h +++ b/fs/ext4/extents_status.h @@ -39,6 +39,7 @@ EXTENT_STATUS_DELAYED | \ EXTENT_STATUS_HOLE) +struct ext4_sb_info; struct ext4_extent; struct extent_status { @@ -119,8 +120,8 @@ static inline void ext4_es_store_status(struct extent_status *es, es->es_pblk = block; } -extern void ext4_es_register_shrinker(struct super_block *sb); -extern void ext4_es_unregister_shrinker(struct super_block *sb); +extern void ext4_es_register_shrinker(struct ext4_sb_info *sbi); +extern void ext4_es_unregister_shrinker(struct ext4_sb_info *sbi); extern void ext4_es_lru_add(struct inode *inode); extern void ext4_es_lru_del(struct inode *inode); diff --git a/fs/ext4/file.c b/fs/ext4/file.c index 469361dbe619..6f4cc567c382 100644 --- a/fs/ext4/file.c +++ b/fs/ext4/file.c @@ -312,7 +312,7 @@ static int ext4_find_unwritten_pgoff(struct inode *inode, blkbits = inode->i_sb->s_blocksize_bits; startoff = *offset; lastoff = startoff; - endoff = (map->m_lblk + map->m_len) << blkbits; + endoff = (loff_t)(map->m_lblk + map->m_len) << blkbits; index = startoff >> PAGE_CACHE_SHIFT; end = endoff >> PAGE_CACHE_SHIFT; @@ -457,7 +457,7 @@ static loff_t ext4_seek_data(struct file *file, loff_t offset, loff_t maxsize) ret = ext4_map_blocks(NULL, inode, &map, 0); if (ret > 0 && !(map.m_flags & EXT4_MAP_UNWRITTEN)) { if (last != start) - dataoff = last << blkbits; + dataoff = (loff_t)last << blkbits; break; } @@ -468,7 +468,7 @@ static loff_t ext4_seek_data(struct file *file, loff_t offset, loff_t maxsize) ext4_es_find_delayed_extent_range(inode, last, last, &es); if (es.es_len != 0 && in_range(last, es.es_lblk, es.es_len)) { if (last != start) - dataoff = last << blkbits; + dataoff = (loff_t)last << blkbits; break; } @@ -486,7 +486,7 @@ static loff_t ext4_seek_data(struct file *file, loff_t offset, loff_t maxsize) } last++; - dataoff = last << blkbits; + dataoff = (loff_t)last << blkbits; } while (last <= end); mutex_unlock(&inode->i_mutex); @@ -530,7 +530,7 @@ static loff_t ext4_seek_hole(struct file *file, loff_t offset, loff_t maxsize) ret = ext4_map_blocks(NULL, inode, &map, 0); if (ret > 0 && !(map.m_flags & EXT4_MAP_UNWRITTEN)) { last += ret; - holeoff = last << blkbits; + holeoff = (loff_t)last << blkbits; continue; } @@ -541,7 +541,7 @@ static loff_t ext4_seek_hole(struct file *file, loff_t offset, loff_t maxsize) ext4_es_find_delayed_extent_range(inode, last, last, &es); if (es.es_len != 0 && in_range(last, es.es_lblk, es.es_len)) { last = es.es_lblk + es.es_len; - holeoff = last << blkbits; + holeoff = (loff_t)last << blkbits; continue; } @@ -556,7 +556,7 @@ static loff_t ext4_seek_hole(struct file *file, loff_t offset, loff_t maxsize) &map, &holeoff); if (!unwritten) { last += ret; - holeoff = last << blkbits; + holeoff = (loff_t)last << blkbits; continue; } } diff --git a/fs/ext4/fsync.c b/fs/ext4/fsync.c index e0ba8a408def..a8bc47f75fa0 100644 --- a/fs/ext4/fsync.c +++ b/fs/ext4/fsync.c @@ -73,32 +73,6 @@ static int ext4_sync_parent(struct inode *inode) return ret; } -/** - * __sync_file - generic_file_fsync without the locking and filemap_write - * @inode: inode to sync - * @datasync: only sync essential metadata if true - * - * This is just generic_file_fsync without the locking. This is needed for - * nojournal mode to make sure this inodes data/metadata makes it to disk - * properly. The i_mutex should be held already. - */ -static int __sync_inode(struct inode *inode, int datasync) -{ - int err; - int ret; - - ret = sync_mapping_buffers(inode->i_mapping); - if (!(inode->i_state & I_DIRTY)) - return ret; - if (datasync && !(inode->i_state & I_DIRTY_DATASYNC)) - return ret; - - err = sync_inode_metadata(inode, 1); - if (ret == 0) - ret = err; - return ret; -} - /* * akpm: A new design for ext4_sync_file(). * @@ -116,7 +90,7 @@ int ext4_sync_file(struct file *file, loff_t start, loff_t end, int datasync) struct inode *inode = file->f_mapping->host; struct ext4_inode_info *ei = EXT4_I(inode); journal_t *journal = EXT4_SB(inode->i_sb)->s_journal; - int ret, err; + int ret = 0, err; tid_t commit_tid; bool needs_barrier = false; @@ -124,25 +98,24 @@ int ext4_sync_file(struct file *file, loff_t start, loff_t end, int datasync) trace_ext4_sync_file_enter(file, datasync); - ret = filemap_write_and_wait_range(inode->i_mapping, start, end); - if (ret) - return ret; - mutex_lock(&inode->i_mutex); - - if (inode->i_sb->s_flags & MS_RDONLY) - goto out; - - ret = ext4_flush_unwritten_io(inode); - if (ret < 0) + if (inode->i_sb->s_flags & MS_RDONLY) { + /* Make sure that we read updated s_mount_flags value */ + smp_rmb(); + if (EXT4_SB(inode->i_sb)->s_mount_flags & EXT4_MF_FS_ABORTED) + ret = -EROFS; goto out; + } if (!journal) { - ret = __sync_inode(inode, datasync); + ret = generic_file_fsync(file, start, end, datasync); if (!ret && !hlist_empty(&inode->i_dentry)) ret = ext4_sync_parent(inode); goto out; } + ret = filemap_write_and_wait_range(inode->i_mapping, start, end); + if (ret) + return ret; /* * data=writeback,ordered: * The caller's filemap_fdatawrite()/wait will sync the data. @@ -172,8 +145,7 @@ int ext4_sync_file(struct file *file, loff_t start, loff_t end, int datasync) if (!ret) ret = err; } - out: - mutex_unlock(&inode->i_mutex); +out: trace_ext4_sync_file_exit(inode, ret); return ret; } diff --git a/fs/ext4/ialloc.c b/fs/ext4/ialloc.c index 00a818d67b54..f03598c6ffd3 100644 --- a/fs/ext4/ialloc.c +++ b/fs/ext4/ialloc.c @@ -747,7 +747,8 @@ repeat_in_this_group: if (!handle) { BUG_ON(nblocks <= 0); handle = __ext4_journal_start_sb(dir->i_sb, line_no, - handle_type, nblocks); + handle_type, nblocks, + 0); if (IS_ERR(handle)) { err = PTR_ERR(handle); ext4_std_error(sb, err); diff --git a/fs/ext4/indirect.c b/fs/ext4/indirect.c index b8d5d351e24f..87b30cd357e7 100644 --- a/fs/ext4/indirect.c +++ b/fs/ext4/indirect.c @@ -624,7 +624,7 @@ cleanup: partial--; } out: - trace_ext4_ind_map_blocks_exit(inode, map, err); + trace_ext4_ind_map_blocks_exit(inode, flags, map, err); return err; } @@ -675,11 +675,6 @@ ssize_t ext4_ind_direct_IO(int rw, struct kiocb *iocb, retry: if (rw == READ && ext4_should_dioread_nolock(inode)) { - if (unlikely(atomic_read(&EXT4_I(inode)->i_unwritten))) { - mutex_lock(&inode->i_mutex); - ext4_flush_unwritten_io(inode); - mutex_unlock(&inode->i_mutex); - } /* * Nolock dioread optimization may be dynamically disabled * via ext4_inode_block_unlocked_dio(). Check inode's state @@ -779,27 +774,18 @@ int ext4_ind_calc_metadata_amount(struct inode *inode, sector_t lblock) return (blk_bits / EXT4_ADDR_PER_BLOCK_BITS(inode->i_sb)) + 1; } -int ext4_ind_trans_blocks(struct inode *inode, int nrblocks, int chunk) +/* + * Calculate number of indirect blocks touched by mapping @nrblocks logically + * contiguous blocks + */ +int ext4_ind_trans_blocks(struct inode *inode, int nrblocks) { - int indirects; - - /* if nrblocks are contiguous */ - if (chunk) { - /* - * With N contiguous data blocks, we need at most - * N/EXT4_ADDR_PER_BLOCK(inode->i_sb) + 1 indirect blocks, - * 2 dindirect blocks, and 1 tindirect block - */ - return DIV_ROUND_UP(nrblocks, - EXT4_ADDR_PER_BLOCK(inode->i_sb)) + 4; - } /* - * if nrblocks are not contiguous, worse case, each block touch - * a indirect block, and each indirect block touch a double indirect - * block, plus a triple indirect block + * With N contiguous data blocks, we need at most + * N/EXT4_ADDR_PER_BLOCK(inode->i_sb) + 1 indirect blocks, + * 2 dindirect blocks, and 1 tindirect block */ - indirects = nrblocks * 2 + 1; - return indirects; + return DIV_ROUND_UP(nrblocks, EXT4_ADDR_PER_BLOCK(inode->i_sb)) + 4; } /* @@ -940,11 +926,13 @@ static int ext4_clear_blocks(handle_t *handle, struct inode *inode, __le32 *last) { __le32 *p; - int flags = EXT4_FREE_BLOCKS_FORGET | EXT4_FREE_BLOCKS_VALIDATED; + int flags = EXT4_FREE_BLOCKS_VALIDATED; int err; if (S_ISDIR(inode->i_mode) || S_ISLNK(inode->i_mode)) - flags |= EXT4_FREE_BLOCKS_METADATA; + flags |= EXT4_FREE_BLOCKS_FORGET | EXT4_FREE_BLOCKS_METADATA; + else if (ext4_should_journal_data(inode)) + flags |= EXT4_FREE_BLOCKS_FORGET; if (!ext4_data_block_valid(EXT4_SB(inode->i_sb), block_to_free, count)) { diff --git a/fs/ext4/inline.c b/fs/ext4/inline.c index 1a346a6bdc8f..d9ecbf1113a7 100644 --- a/fs/ext4/inline.c +++ b/fs/ext4/inline.c @@ -72,7 +72,7 @@ static int get_max_inline_xattr_value_size(struct inode *inode, entry = (struct ext4_xattr_entry *) ((void *)raw_inode + EXT4_I(inode)->i_inline_off); - free += le32_to_cpu(entry->e_value_size); + free += EXT4_XATTR_SIZE(le32_to_cpu(entry->e_value_size)); goto out; } @@ -1810,7 +1810,7 @@ int ext4_inline_data_fiemap(struct inode *inode, if (error) goto out; - physical = iloc.bh->b_blocknr << inode->i_sb->s_blocksize_bits; + physical = (__u64)iloc.bh->b_blocknr << inode->i_sb->s_blocksize_bits; physical += (char *)ext4_raw_inode(&iloc) - iloc.bh->b_data; physical += offsetof(struct ext4_inode, i_block); length = i_size_read(inode); diff --git a/fs/ext4/inode.c b/fs/ext4/inode.c index d6382b89ecbd..0188e65e1f58 100644 --- a/fs/ext4/inode.c +++ b/fs/ext4/inode.c @@ -132,12 +132,12 @@ static inline int ext4_begin_ordered_truncate(struct inode *inode, new_size); } -static void ext4_invalidatepage(struct page *page, unsigned long offset); +static void ext4_invalidatepage(struct page *page, unsigned int offset, + unsigned int length); static int __ext4_journalled_writepage(struct page *page, unsigned int len); static int ext4_bh_delay_or_unwritten(handle_t *handle, struct buffer_head *bh); -static int ext4_discard_partial_page_buffers_no_lock(handle_t *handle, - struct inode *inode, struct page *page, loff_t from, - loff_t length, int flags); +static int ext4_meta_trans_blocks(struct inode *inode, int lblocks, + int pextents); /* * Test whether an inode is a fast symlink. @@ -215,7 +215,8 @@ void ext4_evict_inode(struct inode *inode) filemap_write_and_wait(&inode->i_data); } truncate_inode_pages(&inode->i_data, 0); - ext4_ioend_shutdown(inode); + + WARN_ON(atomic_read(&EXT4_I(inode)->i_ioend_count)); goto no_delete; } @@ -225,8 +226,8 @@ void ext4_evict_inode(struct inode *inode) if (ext4_should_order_data(inode)) ext4_begin_ordered_truncate(inode, 0); truncate_inode_pages(&inode->i_data, 0); - ext4_ioend_shutdown(inode); + WARN_ON(atomic_read(&EXT4_I(inode)->i_ioend_count)); if (is_bad_inode(inode)) goto no_delete; @@ -423,66 +424,6 @@ static int __check_block_validity(struct inode *inode, const char *func, #define check_block_validity(inode, map) \ __check_block_validity((inode), __func__, __LINE__, (map)) -/* - * Return the number of contiguous dirty pages in a given inode - * starting at page frame idx. - */ -static pgoff_t ext4_num_dirty_pages(struct inode *inode, pgoff_t idx, - unsigned int max_pages) -{ - struct address_space *mapping = inode->i_mapping; - pgoff_t index; - struct pagevec pvec; - pgoff_t num = 0; - int i, nr_pages, done = 0; - - if (max_pages == 0) - return 0; - pagevec_init(&pvec, 0); - while (!done) { - index = idx; - nr_pages = pagevec_lookup_tag(&pvec, mapping, &index, - PAGECACHE_TAG_DIRTY, - (pgoff_t)PAGEVEC_SIZE); - if (nr_pages == 0) - break; - for (i = 0; i < nr_pages; i++) { - struct page *page = pvec.pages[i]; - struct buffer_head *bh, *head; - - lock_page(page); - if (unlikely(page->mapping != mapping) || - !PageDirty(page) || - PageWriteback(page) || - page->index != idx) { - done = 1; - unlock_page(page); - break; - } - if (page_has_buffers(page)) { - bh = head = page_buffers(page); - do { - if (!buffer_delay(bh) && - !buffer_unwritten(bh)) - done = 1; - bh = bh->b_this_page; - } while (!done && (bh != head)); - } - unlock_page(page); - if (done) - break; - idx++; - num++; - if (num >= max_pages) { - done = 1; - break; - } - } - pagevec_release(&pvec); - } - return num; -} - #ifdef ES_AGGRESSIVE_TEST static void ext4_map_blocks_es_recheck(handle_t *handle, struct inode *inode, @@ -573,6 +514,8 @@ int ext4_map_blocks(handle_t *handle, struct inode *inode, "logical block %lu\n", inode->i_ino, flags, map->m_len, (unsigned long) map->m_lblk); + ext4_es_lru_add(inode); + /* Lookup extent status tree firstly */ if (ext4_es_lookup_extent(inode, map->m_lblk, &es)) { if (ext4_es_is_written(&es) || ext4_es_is_unwritten(&es)) { @@ -1118,10 +1061,13 @@ static int ext4_write_end(struct file *file, } } - if (ext4_has_inline_data(inode)) - copied = ext4_write_inline_data_end(inode, pos, len, - copied, page); - else + if (ext4_has_inline_data(inode)) { + ret = ext4_write_inline_data_end(inode, pos, len, + copied, page); + if (ret < 0) + goto errout; + copied = ret; + } else copied = block_write_end(file, mapping, pos, len, copied, page, fsdata); @@ -1157,8 +1103,6 @@ static int ext4_write_end(struct file *file, if (i_size_changed) ext4_mark_inode_dirty(handle, inode); - if (copied < 0) - ret = copied; if (pos + len > inode->i_size && ext4_can_truncate(inode)) /* if we have allocated more blocks and copied * less. We will have blocks allocated outside @@ -1415,21 +1359,28 @@ static void ext4_da_release_space(struct inode *inode, int to_free) } static void ext4_da_page_release_reservation(struct page *page, - unsigned long offset) + unsigned int offset, + unsigned int length) { int to_release = 0; struct buffer_head *head, *bh; unsigned int curr_off = 0; struct inode *inode = page->mapping->host; struct ext4_sb_info *sbi = EXT4_SB(inode->i_sb); + unsigned int stop = offset + length; int num_clusters; ext4_fsblk_t lblk; + BUG_ON(stop > PAGE_CACHE_SIZE || stop < length); + head = page_buffers(page); bh = head; do { unsigned int next_off = curr_off + bh->b_size; + if (next_off > stop) + break; + if ((offset <= curr_off) && (buffer_delay(bh))) { to_release++; clear_buffer_delay(bh); @@ -1460,140 +1411,43 @@ static void ext4_da_page_release_reservation(struct page *page, * Delayed allocation stuff */ -/* - * mpage_da_submit_io - walks through extent of pages and try to write - * them with writepage() call back - * - * @mpd->inode: inode - * @mpd->first_page: first page of the extent - * @mpd->next_page: page after the last page of the extent - * - * By the time mpage_da_submit_io() is called we expect all blocks - * to be allocated. this may be wrong if allocation failed. - * - * As pages are already locked by write_cache_pages(), we can't use it - */ -static int mpage_da_submit_io(struct mpage_da_data *mpd, - struct ext4_map_blocks *map) -{ - struct pagevec pvec; - unsigned long index, end; - int ret = 0, err, nr_pages, i; - struct inode *inode = mpd->inode; - struct address_space *mapping = inode->i_mapping; - loff_t size = i_size_read(inode); - unsigned int len, block_start; - struct buffer_head *bh, *page_bufs = NULL; - sector_t pblock = 0, cur_logical = 0; - struct ext4_io_submit io_submit; +struct mpage_da_data { + struct inode *inode; + struct writeback_control *wbc; - BUG_ON(mpd->next_page <= mpd->first_page); - memset(&io_submit, 0, sizeof(io_submit)); + pgoff_t first_page; /* The first page to write */ + pgoff_t next_page; /* Current page to examine */ + pgoff_t last_page; /* Last page to examine */ /* - * We need to start from the first_page to the next_page - 1 - * to make sure we also write the mapped dirty buffer_heads. - * If we look at mpd->b_blocknr we would only be looking - * at the currently mapped buffer_heads. + * Extent to map - this can be after first_page because that can be + * fully mapped. We somewhat abuse m_flags to store whether the extent + * is delalloc or unwritten. */ - index = mpd->first_page; - end = mpd->next_page - 1; - - pagevec_init(&pvec, 0); - while (index <= end) { - nr_pages = pagevec_lookup(&pvec, mapping, index, PAGEVEC_SIZE); - if (nr_pages == 0) - break; - for (i = 0; i < nr_pages; i++) { - int skip_page = 0; - struct page *page = pvec.pages[i]; - - index = page->index; - if (index > end) - break; - - if (index == size >> PAGE_CACHE_SHIFT) - len = size & ~PAGE_CACHE_MASK; - else - len = PAGE_CACHE_SIZE; - if (map) { - cur_logical = index << (PAGE_CACHE_SHIFT - - inode->i_blkbits); - pblock = map->m_pblk + (cur_logical - - map->m_lblk); - } - index++; - - BUG_ON(!PageLocked(page)); - BUG_ON(PageWriteback(page)); - - bh = page_bufs = page_buffers(page); - block_start = 0; - do { - if (map && (cur_logical >= map->m_lblk) && - (cur_logical <= (map->m_lblk + - (map->m_len - 1)))) { - if (buffer_delay(bh)) { - clear_buffer_delay(bh); - bh->b_blocknr = pblock; - } - if (buffer_unwritten(bh) || - buffer_mapped(bh)) - BUG_ON(bh->b_blocknr != pblock); - if (map->m_flags & EXT4_MAP_UNINIT) - set_buffer_uninit(bh); - clear_buffer_unwritten(bh); - } - - /* - * skip page if block allocation undone and - * block is dirty - */ - if (ext4_bh_delay_or_unwritten(NULL, bh)) - skip_page = 1; - bh = bh->b_this_page; - block_start += bh->b_size; - cur_logical++; - pblock++; - } while (bh != page_bufs); - - if (skip_page) { - unlock_page(page); - continue; - } - - clear_page_dirty_for_io(page); - err = ext4_bio_write_page(&io_submit, page, len, - mpd->wbc); - if (!err) - mpd->pages_written++; - /* - * In error case, we have to continue because - * remaining pages are still locked - */ - if (ret == 0) - ret = err; - } - pagevec_release(&pvec); - } - ext4_io_submit(&io_submit); - return ret; -} + struct ext4_map_blocks map; + struct ext4_io_submit io_submit; /* IO submission data */ +}; -static void ext4_da_block_invalidatepages(struct mpage_da_data *mpd) +static void mpage_release_unused_pages(struct mpage_da_data *mpd, + bool invalidate) { int nr_pages, i; pgoff_t index, end; struct pagevec pvec; struct inode *inode = mpd->inode; struct address_space *mapping = inode->i_mapping; - ext4_lblk_t start, last; + + /* This is necessary when next_page == 0. */ + if (mpd->first_page >= mpd->next_page) + return; index = mpd->first_page; end = mpd->next_page - 1; - - start = index << (PAGE_CACHE_SHIFT - inode->i_blkbits); - last = end << (PAGE_CACHE_SHIFT - inode->i_blkbits); - ext4_es_remove_extent(inode, start, last - start + 1); + if (invalidate) { + ext4_lblk_t start, last; + start = index << (PAGE_CACHE_SHIFT - inode->i_blkbits); + last = end << (PAGE_CACHE_SHIFT - inode->i_blkbits); + ext4_es_remove_extent(inode, start, last - start + 1); + } pagevec_init(&pvec, 0); while (index <= end) { @@ -1606,14 +1460,15 @@ static void ext4_da_block_invalidatepages(struct mpage_da_data *mpd) break; BUG_ON(!PageLocked(page)); BUG_ON(PageWriteback(page)); - block_invalidatepage(page, 0); - ClearPageUptodate(page); + if (invalidate) { + block_invalidatepage(page, 0, PAGE_CACHE_SIZE); + ClearPageUptodate(page); + } unlock_page(page); } index = pvec.pages[nr_pages - 1]->index + 1; pagevec_release(&pvec); } - return; } static void ext4_print_free_blocks(struct inode *inode) @@ -1642,215 +1497,6 @@ static void ext4_print_free_blocks(struct inode *inode) return; } -/* - * mpage_da_map_and_submit - go through given space, map them - * if necessary, and then submit them for I/O - * - * @mpd - bh describing space - * - * The function skips space we know is already mapped to disk blocks. - * - */ -static void mpage_da_map_and_submit(struct mpage_da_data *mpd) -{ - int err, blks, get_blocks_flags; - struct ext4_map_blocks map, *mapp = NULL; - sector_t next = mpd->b_blocknr; - unsigned max_blocks = mpd->b_size >> mpd->inode->i_blkbits; - loff_t disksize = EXT4_I(mpd->inode)->i_disksize; - handle_t *handle = NULL; - - /* - * If the blocks are mapped already, or we couldn't accumulate - * any blocks, then proceed immediately to the submission stage. - */ - if ((mpd->b_size == 0) || - ((mpd->b_state & (1 << BH_Mapped)) && - !(mpd->b_state & (1 << BH_Delay)) && - !(mpd->b_state & (1 << BH_Unwritten)))) - goto submit_io; - - handle = ext4_journal_current_handle(); - BUG_ON(!handle); - - /* - * Call ext4_map_blocks() to allocate any delayed allocation - * blocks, or to convert an uninitialized extent to be - * initialized (in the case where we have written into - * one or more preallocated blocks). - * - * We pass in the magic EXT4_GET_BLOCKS_DELALLOC_RESERVE to - * indicate that we are on the delayed allocation path. This - * affects functions in many different parts of the allocation - * call path. This flag exists primarily because we don't - * want to change *many* call functions, so ext4_map_blocks() - * will set the EXT4_STATE_DELALLOC_RESERVED flag once the - * inode's allocation semaphore is taken. - * - * If the blocks in questions were delalloc blocks, set - * EXT4_GET_BLOCKS_DELALLOC_RESERVE so the delalloc accounting - * variables are updated after the blocks have been allocated. - */ - map.m_lblk = next; - map.m_len = max_blocks; - /* - * We're in delalloc path and it is possible that we're going to - * need more metadata blocks than previously reserved. However - * we must not fail because we're in writeback and there is - * nothing we can do about it so it might result in data loss. - * So use reserved blocks to allocate metadata if possible. - */ - get_blocks_flags = EXT4_GET_BLOCKS_CREATE | - EXT4_GET_BLOCKS_METADATA_NOFAIL; - if (ext4_should_dioread_nolock(mpd->inode)) - get_blocks_flags |= EXT4_GET_BLOCKS_IO_CREATE_EXT; - if (mpd->b_state & (1 << BH_Delay)) - get_blocks_flags |= EXT4_GET_BLOCKS_DELALLOC_RESERVE; - - - blks = ext4_map_blocks(handle, mpd->inode, &map, get_blocks_flags); - if (blks < 0) { - struct super_block *sb = mpd->inode->i_sb; - - err = blks; - /* - * If get block returns EAGAIN or ENOSPC and there - * appears to be free blocks we will just let - * mpage_da_submit_io() unlock all of the pages. - */ - if (err == -EAGAIN) - goto submit_io; - - if (err == -ENOSPC && ext4_count_free_clusters(sb)) { - mpd->retval = err; - goto submit_io; - } - - /* - * get block failure will cause us to loop in - * writepages, because a_ops->writepage won't be able - * to make progress. The page will be redirtied by - * writepage and writepages will again try to write - * the same. - */ - if (!(EXT4_SB(sb)->s_mount_flags & EXT4_MF_FS_ABORTED)) { - ext4_msg(sb, KERN_CRIT, - "delayed block allocation failed for inode %lu " - "at logical offset %llu with max blocks %zd " - "with error %d", mpd->inode->i_ino, - (unsigned long long) next, - mpd->b_size >> mpd->inode->i_blkbits, err); - ext4_msg(sb, KERN_CRIT, - "This should not happen!! Data will be lost"); - if (err == -ENOSPC) - ext4_print_free_blocks(mpd->inode); - } - /* invalidate all the pages */ - ext4_da_block_invalidatepages(mpd); - - /* Mark this page range as having been completed */ - mpd->io_done = 1; - return; - } - BUG_ON(blks == 0); - - mapp = ↦ - if (map.m_flags & EXT4_MAP_NEW) { - struct block_device *bdev = mpd->inode->i_sb->s_bdev; - int i; - - for (i = 0; i < map.m_len; i++) - unmap_underlying_metadata(bdev, map.m_pblk + i); - } - - /* - * Update on-disk size along with block allocation. - */ - disksize = ((loff_t) next + blks) << mpd->inode->i_blkbits; - if (disksize > i_size_read(mpd->inode)) - disksize = i_size_read(mpd->inode); - if (disksize > EXT4_I(mpd->inode)->i_disksize) { - ext4_update_i_disksize(mpd->inode, disksize); - err = ext4_mark_inode_dirty(handle, mpd->inode); - if (err) - ext4_error(mpd->inode->i_sb, - "Failed to mark inode %lu dirty", - mpd->inode->i_ino); - } - -submit_io: - mpage_da_submit_io(mpd, mapp); - mpd->io_done = 1; -} - -#define BH_FLAGS ((1 << BH_Uptodate) | (1 << BH_Mapped) | \ - (1 << BH_Delay) | (1 << BH_Unwritten)) - -/* - * mpage_add_bh_to_extent - try to add one more block to extent of blocks - * - * @mpd->lbh - extent of blocks - * @logical - logical number of the block in the file - * @b_state - b_state of the buffer head added - * - * the function is used to collect contig. blocks in same state - */ -static void mpage_add_bh_to_extent(struct mpage_da_data *mpd, sector_t logical, - unsigned long b_state) -{ - sector_t next; - int blkbits = mpd->inode->i_blkbits; - int nrblocks = mpd->b_size >> blkbits; - - /* - * XXX Don't go larger than mballoc is willing to allocate - * This is a stopgap solution. We eventually need to fold - * mpage_da_submit_io() into this function and then call - * ext4_map_blocks() multiple times in a loop - */ - if (nrblocks >= (8*1024*1024 >> blkbits)) - goto flush_it; - - /* check if the reserved journal credits might overflow */ - if (!ext4_test_inode_flag(mpd->inode, EXT4_INODE_EXTENTS)) { - if (nrblocks >= EXT4_MAX_TRANS_DATA) { - /* - * With non-extent format we are limited by the journal - * credit available. Total credit needed to insert - * nrblocks contiguous blocks is dependent on the - * nrblocks. So limit nrblocks. - */ - goto flush_it; - } - } - /* - * First block in the extent - */ - if (mpd->b_size == 0) { - mpd->b_blocknr = logical; - mpd->b_size = 1 << blkbits; - mpd->b_state = b_state & BH_FLAGS; - return; - } - - next = mpd->b_blocknr + nrblocks; - /* - * Can we merge the block to our big extent? - */ - if (logical == next && (b_state & BH_FLAGS) == mpd->b_state) { - mpd->b_size += 1 << blkbits; - return; - } - -flush_it: - /* - * We couldn't merge the block to our extent, so we - * need to flush current extent and start new one - */ - mpage_da_map_and_submit(mpd); - return; -} - static int ext4_bh_delay_or_unwritten(handle_t *handle, struct buffer_head *bh) { return (buffer_delay(bh) || buffer_unwritten(bh)) && buffer_dirty(bh); @@ -1883,6 +1529,8 @@ static int ext4_da_map_blocks(struct inode *inode, sector_t iblock, "logical block %lu\n", inode->i_ino, map->m_len, (unsigned long) map->m_lblk); + ext4_es_lru_add(inode); + /* Lookup extent status tree firstly */ if (ext4_es_lookup_extent(inode, iblock, &es)) { @@ -2156,7 +1804,7 @@ out: * lock so we have to do some magic. * * This function can get called via... - * - ext4_da_writepages after taking page lock (have journal handle) + * - ext4_writepages after taking page lock (have journal handle) * - journal_submit_inode_data_buffers (no journal handle) * - shrink_page_list via the kswapd/direct reclaim (no journal handle) * - grab_page_cache when doing write_begin (have journal handle) @@ -2234,76 +1882,405 @@ static int ext4_writepage(struct page *page, */ return __ext4_journalled_writepage(page, len); - memset(&io_submit, 0, sizeof(io_submit)); + ext4_io_submit_init(&io_submit, wbc); + io_submit.io_end = ext4_init_io_end(inode, GFP_NOFS); + if (!io_submit.io_end) { + redirty_page_for_writepage(wbc, page); + unlock_page(page); + return -ENOMEM; + } ret = ext4_bio_write_page(&io_submit, page, len, wbc); ext4_io_submit(&io_submit); + /* Drop io_end reference we got from init */ + ext4_put_io_end_defer(io_submit.io_end); return ret; } +#define BH_FLAGS ((1 << BH_Unwritten) | (1 << BH_Delay)) + /* - * This is called via ext4_da_writepages() to - * calculate the total number of credits to reserve to fit - * a single extent allocation into a single transaction, - * ext4_da_writpeages() will loop calling this before - * the block allocation. + * mballoc gives us at most this number of blocks... + * XXX: That seems to be only a limitation of ext4_mb_normalize_request(). + * The rest of mballoc seems to handle chunks upto full group size. */ +#define MAX_WRITEPAGES_EXTENT_LEN 2048 -static int ext4_da_writepages_trans_blocks(struct inode *inode) +/* + * mpage_add_bh_to_extent - try to add bh to extent of blocks to map + * + * @mpd - extent of blocks + * @lblk - logical number of the block in the file + * @b_state - b_state of the buffer head added + * + * the function is used to collect contig. blocks in same state + */ +static int mpage_add_bh_to_extent(struct mpage_da_data *mpd, ext4_lblk_t lblk, + unsigned long b_state) +{ + struct ext4_map_blocks *map = &mpd->map; + + /* Don't go larger than mballoc is willing to allocate */ + if (map->m_len >= MAX_WRITEPAGES_EXTENT_LEN) + return 0; + + /* First block in the extent? */ + if (map->m_len == 0) { + map->m_lblk = lblk; + map->m_len = 1; + map->m_flags = b_state & BH_FLAGS; + return 1; + } + + /* Can we merge the block to our big extent? */ + if (lblk == map->m_lblk + map->m_len && + (b_state & BH_FLAGS) == map->m_flags) { + map->m_len++; + return 1; + } + return 0; +} + +static bool add_page_bufs_to_extent(struct mpage_da_data *mpd, + struct buffer_head *head, + struct buffer_head *bh, + ext4_lblk_t lblk) +{ + struct inode *inode = mpd->inode; + ext4_lblk_t blocks = (i_size_read(inode) + (1 << inode->i_blkbits) - 1) + >> inode->i_blkbits; + + do { + BUG_ON(buffer_locked(bh)); + + if (!buffer_dirty(bh) || !buffer_mapped(bh) || + (!buffer_delay(bh) && !buffer_unwritten(bh)) || + lblk >= blocks) { + /* Found extent to map? */ + if (mpd->map.m_len) + return false; + if (lblk >= blocks) + return true; + continue; + } + if (!mpage_add_bh_to_extent(mpd, lblk, bh->b_state)) + return false; + } while (lblk++, (bh = bh->b_this_page) != head); + return true; +} + +static int mpage_submit_page(struct mpage_da_data *mpd, struct page *page) { - int max_blocks = EXT4_I(inode)->i_reserved_data_blocks; + int len; + loff_t size = i_size_read(mpd->inode); + int err; + + BUG_ON(page->index != mpd->first_page); + if (page->index == size >> PAGE_CACHE_SHIFT) + len = size & ~PAGE_CACHE_MASK; + else + len = PAGE_CACHE_SIZE; + clear_page_dirty_for_io(page); + err = ext4_bio_write_page(&mpd->io_submit, page, len, mpd->wbc); + if (!err) + mpd->wbc->nr_to_write--; + mpd->first_page++; + return err; +} + +/* + * mpage_map_buffers - update buffers corresponding to changed extent and + * submit fully mapped pages for IO + * + * @mpd - description of extent to map, on return next extent to map + * + * Scan buffers corresponding to changed extent (we expect corresponding pages + * to be already locked) and update buffer state according to new extent state. + * We map delalloc buffers to their physical location, clear unwritten bits, + * and mark buffers as uninit when we perform writes to uninitialized extents + * and do extent conversion after IO is finished. If the last page is not fully + * mapped, we update @map to the next extent in the last page that needs + * mapping. Otherwise we submit the page for IO. + */ +static int mpage_map_and_submit_buffers(struct mpage_da_data *mpd) +{ + struct pagevec pvec; + int nr_pages, i; + struct inode *inode = mpd->inode; + struct buffer_head *head, *bh; + int bpp_bits = PAGE_CACHE_SHIFT - inode->i_blkbits; + ext4_lblk_t blocks = (i_size_read(inode) + (1 << inode->i_blkbits) - 1) + >> inode->i_blkbits; + pgoff_t start, end; + ext4_lblk_t lblk; + sector_t pblock; + int err; + + start = mpd->map.m_lblk >> bpp_bits; + end = (mpd->map.m_lblk + mpd->map.m_len - 1) >> bpp_bits; + lblk = start << bpp_bits; + pblock = mpd->map.m_pblk; + + pagevec_init(&pvec, 0); + while (start <= end) { + nr_pages = pagevec_lookup(&pvec, inode->i_mapping, start, + PAGEVEC_SIZE); + if (nr_pages == 0) + break; + for (i = 0; i < nr_pages; i++) { + struct page *page = pvec.pages[i]; + + if (page->index > end) + break; + /* Upto 'end' pages must be contiguous */ + BUG_ON(page->index != start); + bh = head = page_buffers(page); + do { + if (lblk < mpd->map.m_lblk) + continue; + if (lblk >= mpd->map.m_lblk + mpd->map.m_len) { + /* + * Buffer after end of mapped extent. + * Find next buffer in the page to map. + */ + mpd->map.m_len = 0; + mpd->map.m_flags = 0; + add_page_bufs_to_extent(mpd, head, bh, + lblk); + pagevec_release(&pvec); + return 0; + } + if (buffer_delay(bh)) { + clear_buffer_delay(bh); + bh->b_blocknr = pblock++; + } + clear_buffer_unwritten(bh); + } while (++lblk < blocks && + (bh = bh->b_this_page) != head); + + /* + * FIXME: This is going to break if dioread_nolock + * supports blocksize < pagesize as we will try to + * convert potentially unmapped parts of inode. + */ + mpd->io_submit.io_end->size += PAGE_CACHE_SIZE; + /* Page fully mapped - let IO run! */ + err = mpage_submit_page(mpd, page); + if (err < 0) { + pagevec_release(&pvec); + return err; + } + start++; + } + pagevec_release(&pvec); + } + /* Extent fully mapped and matches with page boundary. We are done. */ + mpd->map.m_len = 0; + mpd->map.m_flags = 0; + return 0; +} + +static int mpage_map_one_extent(handle_t *handle, struct mpage_da_data *mpd) +{ + struct inode *inode = mpd->inode; + struct ext4_map_blocks *map = &mpd->map; + int get_blocks_flags; + int err; + + trace_ext4_da_write_pages_extent(inode, map); /* - * With non-extent format the journal credit needed to - * insert nrblocks contiguous block is dependent on - * number of contiguous block. So we will limit - * number of contiguous block to a sane value + * Call ext4_map_blocks() to allocate any delayed allocation blocks, or + * to convert an uninitialized extent to be initialized (in the case + * where we have written into one or more preallocated blocks). It is + * possible that we're going to need more metadata blocks than + * previously reserved. However we must not fail because we're in + * writeback and there is nothing we can do about it so it might result + * in data loss. So use reserved blocks to allocate metadata if + * possible. + * + * We pass in the magic EXT4_GET_BLOCKS_DELALLOC_RESERVE if the blocks + * in question are delalloc blocks. This affects functions in many + * different parts of the allocation call path. This flag exists + * primarily because we don't want to change *many* call functions, so + * ext4_map_blocks() will set the EXT4_STATE_DELALLOC_RESERVED flag + * once the inode's allocation semaphore is taken. */ - if (!(ext4_test_inode_flag(inode, EXT4_INODE_EXTENTS)) && - (max_blocks > EXT4_MAX_TRANS_DATA)) - max_blocks = EXT4_MAX_TRANS_DATA; + get_blocks_flags = EXT4_GET_BLOCKS_CREATE | + EXT4_GET_BLOCKS_METADATA_NOFAIL; + if (ext4_should_dioread_nolock(inode)) + get_blocks_flags |= EXT4_GET_BLOCKS_IO_CREATE_EXT; + if (map->m_flags & (1 << BH_Delay)) + get_blocks_flags |= EXT4_GET_BLOCKS_DELALLOC_RESERVE; - return ext4_chunk_trans_blocks(inode, max_blocks); + err = ext4_map_blocks(handle, inode, map, get_blocks_flags); + if (err < 0) + return err; + if (map->m_flags & EXT4_MAP_UNINIT) { + if (!mpd->io_submit.io_end->handle && + ext4_handle_valid(handle)) { + mpd->io_submit.io_end->handle = handle->h_rsv_handle; + handle->h_rsv_handle = NULL; + } + ext4_set_io_unwritten_flag(inode, mpd->io_submit.io_end); + } + + BUG_ON(map->m_len == 0); + if (map->m_flags & EXT4_MAP_NEW) { + struct block_device *bdev = inode->i_sb->s_bdev; + int i; + + for (i = 0; i < map->m_len; i++) + unmap_underlying_metadata(bdev, map->m_pblk + i); + } + return 0; } /* - * write_cache_pages_da - walk the list of dirty pages of the given - * address space and accumulate pages that need writing, and call - * mpage_da_map_and_submit to map a single contiguous memory region - * and then write them. + * mpage_map_and_submit_extent - map extent starting at mpd->lblk of length + * mpd->len and submit pages underlying it for IO + * + * @handle - handle for journal operations + * @mpd - extent to map + * + * The function maps extent starting at mpd->lblk of length mpd->len. If it is + * delayed, blocks are allocated, if it is unwritten, we may need to convert + * them to initialized or split the described range from larger unwritten + * extent. Note that we need not map all the described range since allocation + * can return less blocks or the range is covered by more unwritten extents. We + * cannot map more because we are limited by reserved transaction credits. On + * the other hand we always make sure that the last touched page is fully + * mapped so that it can be written out (and thus forward progress is + * guaranteed). After mapping we submit all mapped pages for IO. */ -static int write_cache_pages_da(handle_t *handle, - struct address_space *mapping, - struct writeback_control *wbc, - struct mpage_da_data *mpd, - pgoff_t *done_index) +static int mpage_map_and_submit_extent(handle_t *handle, + struct mpage_da_data *mpd, + bool *give_up_on_write) { - struct buffer_head *bh, *head; - struct inode *inode = mapping->host; - struct pagevec pvec; - unsigned int nr_pages; - sector_t logical; - pgoff_t index, end; - long nr_to_write = wbc->nr_to_write; - int i, tag, ret = 0; - - memset(mpd, 0, sizeof(struct mpage_da_data)); - mpd->wbc = wbc; - mpd->inode = inode; - pagevec_init(&pvec, 0); - index = wbc->range_start >> PAGE_CACHE_SHIFT; - end = wbc->range_end >> PAGE_CACHE_SHIFT; + struct inode *inode = mpd->inode; + struct ext4_map_blocks *map = &mpd->map; + int err; + loff_t disksize; - if (wbc->sync_mode == WB_SYNC_ALL || wbc->tagged_writepages) + mpd->io_submit.io_end->offset = + ((loff_t)map->m_lblk) << inode->i_blkbits; + while (map->m_len) { + err = mpage_map_one_extent(handle, mpd); + if (err < 0) { + struct super_block *sb = inode->i_sb; + + if (EXT4_SB(sb)->s_mount_flags & EXT4_MF_FS_ABORTED) + goto invalidate_dirty_pages; + /* + * Let the uper layers retry transient errors. + * In the case of ENOSPC, if ext4_count_free_blocks() + * is non-zero, a commit should free up blocks. + */ + if ((err == -ENOMEM) || + (err == -ENOSPC && ext4_count_free_clusters(sb))) + return err; + ext4_msg(sb, KERN_CRIT, + "Delayed block allocation failed for " + "inode %lu at logical offset %llu with" + " max blocks %u with error %d", + inode->i_ino, + (unsigned long long)map->m_lblk, + (unsigned)map->m_len, -err); + ext4_msg(sb, KERN_CRIT, + "This should not happen!! Data will " + "be lost\n"); + if (err == -ENOSPC) + ext4_print_free_blocks(inode); + invalidate_dirty_pages: + *give_up_on_write = true; + return err; + } + /* + * Update buffer state, submit mapped pages, and get us new + * extent to map + */ + err = mpage_map_and_submit_buffers(mpd); + if (err < 0) + return err; + } + + /* Update on-disk size after IO is submitted */ + disksize = ((loff_t)mpd->first_page) << PAGE_CACHE_SHIFT; + if (disksize > i_size_read(inode)) + disksize = i_size_read(inode); + if (disksize > EXT4_I(inode)->i_disksize) { + int err2; + + ext4_update_i_disksize(inode, disksize); + err2 = ext4_mark_inode_dirty(handle, inode); + if (err2) + ext4_error(inode->i_sb, + "Failed to mark inode %lu dirty", + inode->i_ino); + if (!err) + err = err2; + } + return err; +} + +/* + * Calculate the total number of credits to reserve for one writepages + * iteration. This is called from ext4_writepages(). We map an extent of + * upto MAX_WRITEPAGES_EXTENT_LEN blocks and then we go on and finish mapping + * the last partial page. So in total we can map MAX_WRITEPAGES_EXTENT_LEN + + * bpp - 1 blocks in bpp different extents. + */ +static int ext4_da_writepages_trans_blocks(struct inode *inode) +{ + int bpp = ext4_journal_blocks_per_page(inode); + + return ext4_meta_trans_blocks(inode, + MAX_WRITEPAGES_EXTENT_LEN + bpp - 1, bpp); +} + +/* + * mpage_prepare_extent_to_map - find & lock contiguous range of dirty pages + * and underlying extent to map + * + * @mpd - where to look for pages + * + * Walk dirty pages in the mapping. If they are fully mapped, submit them for + * IO immediately. When we find a page which isn't mapped we start accumulating + * extent of buffers underlying these pages that needs mapping (formed by + * either delayed or unwritten buffers). We also lock the pages containing + * these buffers. The extent found is returned in @mpd structure (starting at + * mpd->lblk with length mpd->len blocks). + * + * Note that this function can attach bios to one io_end structure which are + * neither logically nor physically contiguous. Although it may seem as an + * unnecessary complication, it is actually inevitable in blocksize < pagesize + * case as we need to track IO to all buffers underlying a page in one io_end. + */ +static int mpage_prepare_extent_to_map(struct mpage_da_data *mpd) +{ + struct address_space *mapping = mpd->inode->i_mapping; + struct pagevec pvec; + unsigned int nr_pages; + pgoff_t index = mpd->first_page; + pgoff_t end = mpd->last_page; + int tag; + int i, err = 0; + int blkbits = mpd->inode->i_blkbits; + ext4_lblk_t lblk; + struct buffer_head *head; + + if (mpd->wbc->sync_mode == WB_SYNC_ALL || mpd->wbc->tagged_writepages) tag = PAGECACHE_TAG_TOWRITE; else tag = PAGECACHE_TAG_DIRTY; - *done_index = index; + pagevec_init(&pvec, 0); + mpd->map.m_len = 0; + mpd->next_page = index; while (index <= end) { nr_pages = pagevec_lookup_tag(&pvec, mapping, &index, tag, min(end - index, (pgoff_t)PAGEVEC_SIZE-1) + 1); if (nr_pages == 0) - return 0; + goto out; for (i = 0; i < nr_pages; i++) { struct page *page = pvec.pages[i]; @@ -2318,31 +2295,21 @@ static int write_cache_pages_da(handle_t *handle, if (page->index > end) goto out; - *done_index = page->index + 1; - - /* - * If we can't merge this page, and we have - * accumulated an contiguous region, write it - */ - if ((mpd->next_page != page->index) && - (mpd->next_page != mpd->first_page)) { - mpage_da_map_and_submit(mpd); - goto ret_extent_tail; - } + /* If we can't merge this page, we are done. */ + if (mpd->map.m_len > 0 && mpd->next_page != page->index) + goto out; lock_page(page); - /* - * If the page is no longer dirty, or its - * mapping no longer corresponds to inode we - * are writing (which means it has been - * truncated or invalidated), or the page is - * already under writeback and we are not - * doing a data integrity writeback, skip the page + * If the page is no longer dirty, or its mapping no + * longer corresponds to inode we are writing (which + * means it has been truncated or invalidated), or the + * page is already under writeback and we are not doing + * a data integrity writeback, skip the page */ if (!PageDirty(page) || (PageWriteback(page) && - (wbc->sync_mode == WB_SYNC_NONE)) || + (mpd->wbc->sync_mode == WB_SYNC_NONE)) || unlikely(page->mapping != mapping)) { unlock_page(page); continue; @@ -2351,106 +2318,70 @@ static int write_cache_pages_da(handle_t *handle, wait_on_page_writeback(page); BUG_ON(PageWriteback(page)); - /* - * If we have inline data and arrive here, it means that - * we will soon create the block for the 1st page, so - * we'd better clear the inline data here. - */ - if (ext4_has_inline_data(inode)) { - BUG_ON(ext4_test_inode_state(inode, - EXT4_STATE_MAY_INLINE_DATA)); - ext4_destroy_inline_data(handle, inode); - } - - if (mpd->next_page != page->index) + if (mpd->map.m_len == 0) mpd->first_page = page->index; mpd->next_page = page->index + 1; - logical = (sector_t) page->index << - (PAGE_CACHE_SHIFT - inode->i_blkbits); - /* Add all dirty buffers to mpd */ + lblk = ((ext4_lblk_t)page->index) << + (PAGE_CACHE_SHIFT - blkbits); head = page_buffers(page); - bh = head; - do { - BUG_ON(buffer_locked(bh)); - /* - * We need to try to allocate unmapped blocks - * in the same page. Otherwise we won't make - * progress with the page in ext4_writepage - */ - if (ext4_bh_delay_or_unwritten(NULL, bh)) { - mpage_add_bh_to_extent(mpd, logical, - bh->b_state); - if (mpd->io_done) - goto ret_extent_tail; - } else if (buffer_dirty(bh) && - buffer_mapped(bh)) { - /* - * mapped dirty buffer. We need to - * update the b_state because we look - * at b_state in mpage_da_map_blocks. - * We don't update b_size because if we - * find an unmapped buffer_head later - * we need to use the b_state flag of - * that buffer_head. - */ - if (mpd->b_size == 0) - mpd->b_state = - bh->b_state & BH_FLAGS; - } - logical++; - } while ((bh = bh->b_this_page) != head); - - if (nr_to_write > 0) { - nr_to_write--; - if (nr_to_write == 0 && - wbc->sync_mode == WB_SYNC_NONE) - /* - * We stop writing back only if we are - * not doing integrity sync. In case of - * integrity sync we have to keep going - * because someone may be concurrently - * dirtying pages, and we might have - * synced a lot of newly appeared dirty - * pages, but have not synced all of the - * old dirty pages. - */ + if (!add_page_bufs_to_extent(mpd, head, head, lblk)) + goto out; + /* So far everything mapped? Submit the page for IO. */ + if (mpd->map.m_len == 0) { + err = mpage_submit_page(mpd, page); + if (err < 0) goto out; } + + /* + * Accumulated enough dirty pages? This doesn't apply + * to WB_SYNC_ALL mode. For integrity sync we have to + * keep going because someone may be concurrently + * dirtying pages, and we might have synced a lot of + * newly appeared dirty pages, but have not synced all + * of the old dirty pages. + */ + if (mpd->wbc->sync_mode == WB_SYNC_NONE && + mpd->next_page - mpd->first_page >= + mpd->wbc->nr_to_write) + goto out; } pagevec_release(&pvec); cond_resched(); } return 0; -ret_extent_tail: - ret = MPAGE_DA_EXTENT_TAIL; out: pagevec_release(&pvec); - cond_resched(); - return ret; + return err; } +static int __writepage(struct page *page, struct writeback_control *wbc, + void *data) +{ + struct address_space *mapping = data; + int ret = ext4_writepage(page, wbc); + mapping_set_error(mapping, ret); + return ret; +} -static int ext4_da_writepages(struct address_space *mapping, - struct writeback_control *wbc) +static int ext4_writepages(struct address_space *mapping, + struct writeback_control *wbc) { - pgoff_t index; + pgoff_t writeback_index = 0; + long nr_to_write = wbc->nr_to_write; int range_whole = 0; + int cycled = 1; handle_t *handle = NULL; struct mpage_da_data mpd; struct inode *inode = mapping->host; - int pages_written = 0; - unsigned int max_pages; - int range_cyclic, cycled = 1, io_done = 0; - int needed_blocks, ret = 0; - long desired_nr_to_write, nr_to_writebump = 0; - loff_t range_start = wbc->range_start; + int needed_blocks, rsv_blocks = 0, ret = 0; struct ext4_sb_info *sbi = EXT4_SB(mapping->host->i_sb); - pgoff_t done_index = 0; - pgoff_t end; + bool done; struct blk_plug plug; + bool give_up_on_write = false; - trace_ext4_da_writepages(inode, wbc); + trace_ext4_writepages(inode, wbc); /* * No pages to write? This is mainly a kludge to avoid starting @@ -2460,164 +2391,165 @@ static int ext4_da_writepages(struct address_space *mapping, if (!mapping->nrpages || !mapping_tagged(mapping, PAGECACHE_TAG_DIRTY)) return 0; + if (ext4_should_journal_data(inode)) { + struct blk_plug plug; + int ret; + + blk_start_plug(&plug); + ret = write_cache_pages(mapping, wbc, __writepage, mapping); + blk_finish_plug(&plug); + return ret; + } + /* * If the filesystem has aborted, it is read-only, so return * right away instead of dumping stack traces later on that * will obscure the real source of the problem. We test * EXT4_MF_FS_ABORTED instead of sb->s_flag's MS_RDONLY because * the latter could be true if the filesystem is mounted - * read-only, and in that case, ext4_da_writepages should + * read-only, and in that case, ext4_writepages should * *never* be called, so if that ever happens, we would want * the stack trace. */ if (unlikely(sbi->s_mount_flags & EXT4_MF_FS_ABORTED)) return -EROFS; + if (ext4_should_dioread_nolock(inode)) { + /* + * We may need to convert upto one extent per block in + * the page and we may dirty the inode. + */ + rsv_blocks = 1 + (PAGE_CACHE_SIZE >> inode->i_blkbits); + } + + /* + * If we have inline data and arrive here, it means that + * we will soon create the block for the 1st page, so + * we'd better clear the inline data here. + */ + if (ext4_has_inline_data(inode)) { + /* Just inode will be modified... */ + handle = ext4_journal_start(inode, EXT4_HT_INODE, 1); + if (IS_ERR(handle)) { + ret = PTR_ERR(handle); + goto out_writepages; + } + BUG_ON(ext4_test_inode_state(inode, + EXT4_STATE_MAY_INLINE_DATA)); + ext4_destroy_inline_data(handle, inode); + ext4_journal_stop(handle); + } + if (wbc->range_start == 0 && wbc->range_end == LLONG_MAX) range_whole = 1; - range_cyclic = wbc->range_cyclic; if (wbc->range_cyclic) { - index = mapping->writeback_index; - if (index) + writeback_index = mapping->writeback_index; + if (writeback_index) cycled = 0; - wbc->range_start = index << PAGE_CACHE_SHIFT; - wbc->range_end = LLONG_MAX; - wbc->range_cyclic = 0; - end = -1; + mpd.first_page = writeback_index; + mpd.last_page = -1; } else { - index = wbc->range_start >> PAGE_CACHE_SHIFT; - end = wbc->range_end >> PAGE_CACHE_SHIFT; - } - - /* - * This works around two forms of stupidity. The first is in - * the writeback code, which caps the maximum number of pages - * written to be 1024 pages. This is wrong on multiple - * levels; different architectues have a different page size, - * which changes the maximum amount of data which gets - * written. Secondly, 4 megabytes is way too small. XFS - * forces this value to be 16 megabytes by multiplying - * nr_to_write parameter by four, and then relies on its - * allocator to allocate larger extents to make them - * contiguous. Unfortunately this brings us to the second - * stupidity, which is that ext4's mballoc code only allocates - * at most 2048 blocks. So we force contiguous writes up to - * the number of dirty blocks in the inode, or - * sbi->max_writeback_mb_bump whichever is smaller. - */ - max_pages = sbi->s_max_writeback_mb_bump << (20 - PAGE_CACHE_SHIFT); - if (!range_cyclic && range_whole) { - if (wbc->nr_to_write == LONG_MAX) - desired_nr_to_write = wbc->nr_to_write; - else - desired_nr_to_write = wbc->nr_to_write * 8; - } else - desired_nr_to_write = ext4_num_dirty_pages(inode, index, - max_pages); - if (desired_nr_to_write > max_pages) - desired_nr_to_write = max_pages; - - if (wbc->nr_to_write < desired_nr_to_write) { - nr_to_writebump = desired_nr_to_write - wbc->nr_to_write; - wbc->nr_to_write = desired_nr_to_write; + mpd.first_page = wbc->range_start >> PAGE_CACHE_SHIFT; + mpd.last_page = wbc->range_end >> PAGE_CACHE_SHIFT; } + mpd.inode = inode; + mpd.wbc = wbc; + ext4_io_submit_init(&mpd.io_submit, wbc); retry: if (wbc->sync_mode == WB_SYNC_ALL || wbc->tagged_writepages) - tag_pages_for_writeback(mapping, index, end); - + tag_pages_for_writeback(mapping, mpd.first_page, mpd.last_page); + done = false; blk_start_plug(&plug); - while (!ret && wbc->nr_to_write > 0) { + while (!done && mpd.first_page <= mpd.last_page) { + /* For each extent of pages we use new io_end */ + mpd.io_submit.io_end = ext4_init_io_end(inode, GFP_KERNEL); + if (!mpd.io_submit.io_end) { + ret = -ENOMEM; + break; + } /* - * we insert one extent at a time. So we need - * credit needed for single extent allocation. - * journalled mode is currently not supported - * by delalloc + * We have two constraints: We find one extent to map and we + * must always write out whole page (makes a difference when + * blocksize < pagesize) so that we don't block on IO when we + * try to write out the rest of the page. Journalled mode is + * not supported by delalloc. */ BUG_ON(ext4_should_journal_data(inode)); needed_blocks = ext4_da_writepages_trans_blocks(inode); - /* start a new transaction*/ - handle = ext4_journal_start(inode, EXT4_HT_WRITE_PAGE, - needed_blocks); + /* start a new transaction */ + handle = ext4_journal_start_with_reserve(inode, + EXT4_HT_WRITE_PAGE, needed_blocks, rsv_blocks); if (IS_ERR(handle)) { ret = PTR_ERR(handle); ext4_msg(inode->i_sb, KERN_CRIT, "%s: jbd2_start: " "%ld pages, ino %lu; err %d", __func__, wbc->nr_to_write, inode->i_ino, ret); - blk_finish_plug(&plug); - goto out_writepages; + /* Release allocated io_end */ + ext4_put_io_end(mpd.io_submit.io_end); + break; } - /* - * Now call write_cache_pages_da() to find the next - * contiguous region of logical blocks that need - * blocks to be allocated by ext4 and submit them. - */ - ret = write_cache_pages_da(handle, mapping, - wbc, &mpd, &done_index); - /* - * If we have a contiguous extent of pages and we - * haven't done the I/O yet, map the blocks and submit - * them for I/O. - */ - if (!mpd.io_done && mpd.next_page != mpd.first_page) { - mpage_da_map_and_submit(&mpd); - ret = MPAGE_DA_EXTENT_TAIL; + trace_ext4_da_write_pages(inode, mpd.first_page, mpd.wbc); + ret = mpage_prepare_extent_to_map(&mpd); + if (!ret) { + if (mpd.map.m_len) + ret = mpage_map_and_submit_extent(handle, &mpd, + &give_up_on_write); + else { + /* + * We scanned the whole range (or exhausted + * nr_to_write), submitted what was mapped and + * didn't find anything needing mapping. We are + * done. + */ + done = true; + } } - trace_ext4_da_write_pages(inode, &mpd); - wbc->nr_to_write -= mpd.pages_written; - ext4_journal_stop(handle); - - if ((mpd.retval == -ENOSPC) && sbi->s_journal) { - /* commit the transaction which would + /* Submit prepared bio */ + ext4_io_submit(&mpd.io_submit); + /* Unlock pages we didn't use */ + mpage_release_unused_pages(&mpd, give_up_on_write); + /* Drop our io_end reference we got from init */ + ext4_put_io_end(mpd.io_submit.io_end); + + if (ret == -ENOSPC && sbi->s_journal) { + /* + * Commit the transaction which would * free blocks released in the transaction * and try again */ jbd2_journal_force_commit_nested(sbi->s_journal); ret = 0; - } else if (ret == MPAGE_DA_EXTENT_TAIL) { - /* - * Got one extent now try with rest of the pages. - * If mpd.retval is set -EIO, journal is aborted. - * So we don't need to write any more. - */ - pages_written += mpd.pages_written; - ret = mpd.retval; - io_done = 1; - } else if (wbc->nr_to_write) - /* - * There is no more writeout needed - * or we requested for a noblocking writeout - * and we found the device congested - */ + continue; + } + /* Fatal error - ENOMEM, EIO... */ + if (ret) break; } blk_finish_plug(&plug); - if (!io_done && !cycled) { + if (!ret && !cycled) { cycled = 1; - index = 0; - wbc->range_start = index << PAGE_CACHE_SHIFT; - wbc->range_end = mapping->writeback_index - 1; + mpd.last_page = writeback_index - 1; + mpd.first_page = 0; goto retry; } /* Update index */ - wbc->range_cyclic = range_cyclic; if (wbc->range_cyclic || (range_whole && wbc->nr_to_write > 0)) /* - * set the writeback_index so that range_cyclic + * Set the writeback_index so that range_cyclic * mode will write it back later */ - mapping->writeback_index = done_index; + mapping->writeback_index = mpd.first_page; out_writepages: - wbc->nr_to_write -= nr_to_writebump; - wbc->range_start = range_start; - trace_ext4_da_writepages_result(inode, wbc, ret, pages_written); + trace_ext4_writepages_result(inode, wbc, ret, + nr_to_write - wbc->nr_to_write); return ret; } @@ -2829,7 +2761,8 @@ static int ext4_da_write_end(struct file *file, return ret ? ret : copied; } -static void ext4_da_invalidatepage(struct page *page, unsigned long offset) +static void ext4_da_invalidatepage(struct page *page, unsigned int offset, + unsigned int length) { /* * Drop reserved blocks @@ -2838,10 +2771,10 @@ static void ext4_da_invalidatepage(struct page *page, unsigned long offset) if (!page_has_buffers(page)) goto out; - ext4_da_page_release_reservation(page, offset); + ext4_da_page_release_reservation(page, offset, length); out: - ext4_invalidatepage(page, offset); + ext4_invalidatepage(page, offset, length); return; } @@ -2864,7 +2797,7 @@ int ext4_alloc_da_blocks(struct inode *inode) * laptop_mode, not even desirable). However, to do otherwise * would require replicating code paths in: * - * ext4_da_writepages() -> + * ext4_writepages() -> * write_cache_pages() ---> (via passed in callback function) * __mpage_da_writepage() --> * mpage_add_bh_to_extent() @@ -2989,37 +2922,40 @@ ext4_readpages(struct file *file, struct address_space *mapping, return mpage_readpages(mapping, pages, nr_pages, ext4_get_block); } -static void ext4_invalidatepage(struct page *page, unsigned long offset) +static void ext4_invalidatepage(struct page *page, unsigned int offset, + unsigned int length) { - trace_ext4_invalidatepage(page, offset); + trace_ext4_invalidatepage(page, offset, length); /* No journalling happens on data buffers when this function is used */ WARN_ON(page_has_buffers(page) && buffer_jbd(page_buffers(page))); - block_invalidatepage(page, offset); + block_invalidatepage(page, offset, length); } static int __ext4_journalled_invalidatepage(struct page *page, - unsigned long offset) + unsigned int offset, + unsigned int length) { journal_t *journal = EXT4_JOURNAL(page->mapping->host); - trace_ext4_journalled_invalidatepage(page, offset); + trace_ext4_journalled_invalidatepage(page, offset, length); /* * If it's a full truncate we just forget about the pending dirtying */ - if (offset == 0) + if (offset == 0 && length == PAGE_CACHE_SIZE) ClearPageChecked(page); - return jbd2_journal_invalidatepage(journal, page, offset); + return jbd2_journal_invalidatepage(journal, page, offset, length); } /* Wrapper for aops... */ static void ext4_journalled_invalidatepage(struct page *page, - unsigned long offset) + unsigned int offset, + unsigned int length) { - WARN_ON(__ext4_journalled_invalidatepage(page, offset) < 0); + WARN_ON(__ext4_journalled_invalidatepage(page, offset, length) < 0); } static int ext4_releasepage(struct page *page, gfp_t wait) @@ -3067,9 +3003,13 @@ static void ext4_end_io_dio(struct kiocb *iocb, loff_t offset, struct inode *inode = file_inode(iocb->ki_filp); ext4_io_end_t *io_end = iocb->private; - /* if not async direct IO or dio with 0 bytes write, just return */ - if (!io_end || !size) - goto out; + /* if not async direct IO just return */ + if (!io_end) { + inode_dio_done(inode); + if (is_async) + aio_complete(iocb, ret, 0); + return; + } ext_debug("ext4_end_io_dio(): io_end 0x%p " "for inode %lu, iocb 0x%p, offset %llu, size %zd\n", @@ -3077,25 +3017,13 @@ static void ext4_end_io_dio(struct kiocb *iocb, loff_t offset, size); iocb->private = NULL; - - /* if not aio dio with unwritten extents, just free io and return */ - if (!(io_end->flag & EXT4_IO_END_UNWRITTEN)) { - ext4_free_io_end(io_end); -out: - inode_dio_done(inode); - if (is_async) - aio_complete(iocb, ret, 0); - return; - } - io_end->offset = offset; io_end->size = size; if (is_async) { io_end->iocb = iocb; io_end->result = ret; } - - ext4_add_complete_io(io_end); + ext4_put_io_end_defer(io_end); } /* @@ -3129,6 +3057,7 @@ static ssize_t ext4_ext_direct_IO(int rw, struct kiocb *iocb, get_block_t *get_block_func = NULL; int dio_flags = 0; loff_t final_size = offset + count; + ext4_io_end_t *io_end = NULL; /* Use the old path for reads and writes beyond i_size. */ if (rw != WRITE || final_size > inode->i_size) @@ -3136,11 +3065,18 @@ static ssize_t ext4_ext_direct_IO(int rw, struct kiocb *iocb, BUG_ON(iocb->private == NULL); + /* + * Make all waiters for direct IO properly wait also for extent + * conversion. This also disallows race between truncate() and + * overwrite DIO as i_dio_count needs to be incremented under i_mutex. + */ + if (rw == WRITE) + atomic_inc(&inode->i_dio_count); + /* If we do a overwrite dio, i_mutex locking can be released */ overwrite = *((int *)iocb->private); if (overwrite) { - atomic_inc(&inode->i_dio_count); down_read(&EXT4_I(inode)->i_data_sem); mutex_unlock(&inode->i_mutex); } @@ -3167,13 +3103,16 @@ static ssize_t ext4_ext_direct_IO(int rw, struct kiocb *iocb, iocb->private = NULL; ext4_inode_aio_set(inode, NULL); if (!is_sync_kiocb(iocb)) { - ext4_io_end_t *io_end = ext4_init_io_end(inode, GFP_NOFS); + io_end = ext4_init_io_end(inode, GFP_NOFS); if (!io_end) { ret = -ENOMEM; goto retake_lock; } io_end->flag |= EXT4_IO_END_DIRECT; - iocb->private = io_end; + /* + * Grab reference for DIO. Will be dropped in ext4_end_io_dio() + */ + iocb->private = ext4_get_io_end(io_end); /* * we save the io structure for current async direct * IO, so that later ext4_map_blocks() could flag the @@ -3197,33 +3136,42 @@ static ssize_t ext4_ext_direct_IO(int rw, struct kiocb *iocb, NULL, dio_flags); - if (iocb->private) - ext4_inode_aio_set(inode, NULL); /* - * The io_end structure takes a reference to the inode, that - * structure needs to be destroyed and the reference to the - * inode need to be dropped, when IO is complete, even with 0 - * byte write, or failed. - * - * In the successful AIO DIO case, the io_end structure will - * be destroyed and the reference to the inode will be dropped - * after the end_io call back function is called. - * - * In the case there is 0 byte write, or error case, since VFS - * direct IO won't invoke the end_io call back function, we - * need to free the end_io structure here. + * Put our reference to io_end. This can free the io_end structure e.g. + * in sync IO case or in case of error. It can even perform extent + * conversion if all bios we submitted finished before we got here. + * Note that in that case iocb->private can be already set to NULL + * here. */ - if (ret != -EIOCBQUEUED && ret <= 0 && iocb->private) { - ext4_free_io_end(iocb->private); - iocb->private = NULL; - } else if (ret > 0 && !overwrite && ext4_test_inode_state(inode, + if (io_end) { + ext4_inode_aio_set(inode, NULL); + ext4_put_io_end(io_end); + /* + * When no IO was submitted ext4_end_io_dio() was not + * called so we have to put iocb's reference. + */ + if (ret <= 0 && ret != -EIOCBQUEUED && iocb->private) { + WARN_ON(iocb->private != io_end); + WARN_ON(io_end->flag & EXT4_IO_END_UNWRITTEN); + WARN_ON(io_end->iocb); + /* + * Generic code already did inode_dio_done() so we + * have to clear EXT4_IO_END_DIRECT to not do it for + * the second time. + */ + io_end->flag = 0; + ext4_put_io_end(io_end); + iocb->private = NULL; + } + } + if (ret > 0 && !overwrite && ext4_test_inode_state(inode, EXT4_STATE_DIO_UNWRITTEN)) { int err; /* * for non AIO case, since the IO is already * completed, we could do the conversion right here */ - err = ext4_convert_unwritten_extents(inode, + err = ext4_convert_unwritten_extents(NULL, inode, offset, ret); if (err < 0) ret = err; @@ -3231,9 +3179,10 @@ static ssize_t ext4_ext_direct_IO(int rw, struct kiocb *iocb, } retake_lock: + if (rw == WRITE) + inode_dio_done(inode); /* take i_mutex locking again if we do a ovewrite dio */ if (overwrite) { - inode_dio_done(inode); up_read(&EXT4_I(inode)->i_data_sem); mutex_lock(&inode->i_mutex); } @@ -3292,6 +3241,7 @@ static const struct address_space_operations ext4_aops = { .readpage = ext4_readpage, .readpages = ext4_readpages, .writepage = ext4_writepage, + .writepages = ext4_writepages, .write_begin = ext4_write_begin, .write_end = ext4_write_end, .bmap = ext4_bmap, @@ -3307,6 +3257,7 @@ static const struct address_space_operations ext4_journalled_aops = { .readpage = ext4_readpage, .readpages = ext4_readpages, .writepage = ext4_writepage, + .writepages = ext4_writepages, .write_begin = ext4_write_begin, .write_end = ext4_journalled_write_end, .set_page_dirty = ext4_journalled_set_page_dirty, @@ -3322,7 +3273,7 @@ static const struct address_space_operations ext4_da_aops = { .readpage = ext4_readpage, .readpages = ext4_readpages, .writepage = ext4_writepage, - .writepages = ext4_da_writepages, + .writepages = ext4_writepages, .write_begin = ext4_da_write_begin, .write_end = ext4_da_write_end, .bmap = ext4_bmap, @@ -3355,89 +3306,56 @@ void ext4_set_aops(struct inode *inode) inode->i_mapping->a_ops = &ext4_aops; } - /* - * ext4_discard_partial_page_buffers() - * Wrapper function for ext4_discard_partial_page_buffers_no_lock. - * This function finds and locks the page containing the offset - * "from" and passes it to ext4_discard_partial_page_buffers_no_lock. - * Calling functions that already have the page locked should call - * ext4_discard_partial_page_buffers_no_lock directly. + * ext4_block_truncate_page() zeroes out a mapping from file offset `from' + * up to the end of the block which corresponds to `from'. + * This required during truncate. We need to physically zero the tail end + * of that block so it doesn't yield old data if the file is later grown. */ -int ext4_discard_partial_page_buffers(handle_t *handle, - struct address_space *mapping, loff_t from, - loff_t length, int flags) +int ext4_block_truncate_page(handle_t *handle, + struct address_space *mapping, loff_t from) { + unsigned offset = from & (PAGE_CACHE_SIZE-1); + unsigned length; + unsigned blocksize; struct inode *inode = mapping->host; - struct page *page; - int err = 0; - page = find_or_create_page(mapping, from >> PAGE_CACHE_SHIFT, - mapping_gfp_mask(mapping) & ~__GFP_FS); - if (!page) - return -ENOMEM; - - err = ext4_discard_partial_page_buffers_no_lock(handle, inode, page, - from, length, flags); + blocksize = inode->i_sb->s_blocksize; + length = blocksize - (offset & (blocksize - 1)); - unlock_page(page); - page_cache_release(page); - return err; + return ext4_block_zero_page_range(handle, mapping, from, length); } /* - * ext4_discard_partial_page_buffers_no_lock() - * Zeros a page range of length 'length' starting from offset 'from'. - * Buffer heads that correspond to the block aligned regions of the - * zeroed range will be unmapped. Unblock aligned regions - * will have the corresponding buffer head mapped if needed so that - * that region of the page can be updated with the partial zero out. - * - * This function assumes that the page has already been locked. The - * The range to be discarded must be contained with in the given page. - * If the specified range exceeds the end of the page it will be shortened - * to the end of the page that corresponds to 'from'. This function is - * appropriate for updating a page and it buffer heads to be unmapped and - * zeroed for blocks that have been either released, or are going to be - * released. - * - * handle: The journal handle - * inode: The files inode - * page: A locked page that contains the offset "from" - * from: The starting byte offset (from the beginning of the file) - * to begin discarding - * len: The length of bytes to discard - * flags: Optional flags that may be used: - * - * EXT4_DISCARD_PARTIAL_PG_ZERO_UNMAPPED - * Only zero the regions of the page whose buffer heads - * have already been unmapped. This flag is appropriate - * for updating the contents of a page whose blocks may - * have already been released, and we only want to zero - * out the regions that correspond to those released blocks. - * - * Returns zero on success or negative on failure. + * ext4_block_zero_page_range() zeros out a mapping of length 'length' + * starting from file offset 'from'. The range to be zero'd must + * be contained with in one block. If the specified range exceeds + * the end of the block it will be shortened to end of the block + * that cooresponds to 'from' */ -static int ext4_discard_partial_page_buffers_no_lock(handle_t *handle, - struct inode *inode, struct page *page, loff_t from, - loff_t length, int flags) +int ext4_block_zero_page_range(handle_t *handle, + struct address_space *mapping, loff_t from, loff_t length) { ext4_fsblk_t index = from >> PAGE_CACHE_SHIFT; - unsigned int offset = from & (PAGE_CACHE_SIZE-1); - unsigned int blocksize, max, pos; + unsigned offset = from & (PAGE_CACHE_SIZE-1); + unsigned blocksize, max, pos; ext4_lblk_t iblock; + struct inode *inode = mapping->host; struct buffer_head *bh; + struct page *page; int err = 0; - blocksize = inode->i_sb->s_blocksize; - max = PAGE_CACHE_SIZE - offset; + page = find_or_create_page(mapping, from >> PAGE_CACHE_SHIFT, + mapping_gfp_mask(mapping) & ~__GFP_FS); + if (!page) + return -ENOMEM; - if (index != page->index) - return -EINVAL; + blocksize = inode->i_sb->s_blocksize; + max = blocksize - (offset & (blocksize - 1)); /* * correct length if it does not fall between - * 'from' and the end of the page + * 'from' and the end of the block */ if (length > max || length < 0) length = max; @@ -3455,106 +3373,91 @@ static int ext4_discard_partial_page_buffers_no_lock(handle_t *handle, iblock++; pos += blocksize; } - - pos = offset; - while (pos < offset + length) { - unsigned int end_of_block, range_to_discard; - - err = 0; - - /* The length of space left to zero and unmap */ - range_to_discard = offset + length - pos; - - /* The length of space until the end of the block */ - end_of_block = blocksize - (pos & (blocksize-1)); - - /* - * Do not unmap or zero past end of block - * for this buffer head - */ - if (range_to_discard > end_of_block) - range_to_discard = end_of_block; - - - /* - * Skip this buffer head if we are only zeroing unampped - * regions of the page - */ - if (flags & EXT4_DISCARD_PARTIAL_PG_ZERO_UNMAPPED && - buffer_mapped(bh)) - goto next; - - /* If the range is block aligned, unmap */ - if (range_to_discard == blocksize) { - clear_buffer_dirty(bh); - bh->b_bdev = NULL; - clear_buffer_mapped(bh); - clear_buffer_req(bh); - clear_buffer_new(bh); - clear_buffer_delay(bh); - clear_buffer_unwritten(bh); - clear_buffer_uptodate(bh); - zero_user(page, pos, range_to_discard); - BUFFER_TRACE(bh, "Buffer discarded"); - goto next; - } - - /* - * If this block is not completely contained in the range - * to be discarded, then it is not going to be released. Because - * we need to keep this block, we need to make sure this part - * of the page is uptodate before we modify it by writeing - * partial zeros on it. - */ + if (buffer_freed(bh)) { + BUFFER_TRACE(bh, "freed: skip"); + goto unlock; + } + if (!buffer_mapped(bh)) { + BUFFER_TRACE(bh, "unmapped"); + ext4_get_block(inode, iblock, bh, 0); + /* unmapped? It's a hole - nothing to do */ if (!buffer_mapped(bh)) { - /* - * Buffer head must be mapped before we can read - * from the block - */ - BUFFER_TRACE(bh, "unmapped"); - ext4_get_block(inode, iblock, bh, 0); - /* unmapped? It's a hole - nothing to do */ - if (!buffer_mapped(bh)) { - BUFFER_TRACE(bh, "still unmapped"); - goto next; - } + BUFFER_TRACE(bh, "still unmapped"); + goto unlock; } + } - /* Ok, it's mapped. Make sure it's up-to-date */ - if (PageUptodate(page)) - set_buffer_uptodate(bh); + /* Ok, it's mapped. Make sure it's up-to-date */ + if (PageUptodate(page)) + set_buffer_uptodate(bh); - if (!buffer_uptodate(bh)) { - err = -EIO; - ll_rw_block(READ, 1, &bh); - wait_on_buffer(bh); - /* Uhhuh. Read error. Complain and punt.*/ - if (!buffer_uptodate(bh)) - goto next; - } + if (!buffer_uptodate(bh)) { + err = -EIO; + ll_rw_block(READ, 1, &bh); + wait_on_buffer(bh); + /* Uhhuh. Read error. Complain and punt. */ + if (!buffer_uptodate(bh)) + goto unlock; + } + if (ext4_should_journal_data(inode)) { + BUFFER_TRACE(bh, "get write access"); + err = ext4_journal_get_write_access(handle, bh); + if (err) + goto unlock; + } + zero_user(page, offset, length); + BUFFER_TRACE(bh, "zeroed end of block"); - if (ext4_should_journal_data(inode)) { - BUFFER_TRACE(bh, "get write access"); - err = ext4_journal_get_write_access(handle, bh); - if (err) - goto next; - } + if (ext4_should_journal_data(inode)) { + err = ext4_handle_dirty_metadata(handle, inode, bh); + } else { + err = 0; + mark_buffer_dirty(bh); + if (ext4_test_inode_state(inode, EXT4_STATE_ORDERED_MODE)) + err = ext4_jbd2_file_inode(handle, inode); + } + +unlock: + unlock_page(page); + page_cache_release(page); + return err; +} - zero_user(page, pos, range_to_discard); +int ext4_zero_partial_blocks(handle_t *handle, struct inode *inode, + loff_t lstart, loff_t length) +{ + struct super_block *sb = inode->i_sb; + struct address_space *mapping = inode->i_mapping; + unsigned partial_start, partial_end; + ext4_fsblk_t start, end; + loff_t byte_end = (lstart + length - 1); + int err = 0; - err = 0; - if (ext4_should_journal_data(inode)) { - err = ext4_handle_dirty_metadata(handle, inode, bh); - } else - mark_buffer_dirty(bh); + partial_start = lstart & (sb->s_blocksize - 1); + partial_end = byte_end & (sb->s_blocksize - 1); - BUFFER_TRACE(bh, "Partial buffer zeroed"); -next: - bh = bh->b_this_page; - iblock++; - pos += range_to_discard; - } + start = lstart >> sb->s_blocksize_bits; + end = byte_end >> sb->s_blocksize_bits; + /* Handle partial zero within the single block */ + if (start == end && + (partial_start || (partial_end != sb->s_blocksize - 1))) { + err = ext4_block_zero_page_range(handle, mapping, + lstart, length); + return err; + } + /* Handle partial zero out on the start of the range */ + if (partial_start) { + err = ext4_block_zero_page_range(handle, mapping, + lstart, sb->s_blocksize); + if (err) + return err; + } + /* Handle partial zero out on the end of the range */ + if (partial_end != sb->s_blocksize - 1) + err = ext4_block_zero_page_range(handle, mapping, + byte_end - partial_end, + partial_end + 1); return err; } @@ -3580,14 +3483,12 @@ int ext4_can_truncate(struct inode *inode) * Returns: 0 on success or negative on failure */ -int ext4_punch_hole(struct file *file, loff_t offset, loff_t length) +int ext4_punch_hole(struct inode *inode, loff_t offset, loff_t length) { - struct inode *inode = file_inode(file); struct super_block *sb = inode->i_sb; ext4_lblk_t first_block, stop_block; struct address_space *mapping = inode->i_mapping; - loff_t first_page, last_page, page_len; - loff_t first_page_offset, last_page_offset; + loff_t first_block_offset, last_block_offset; handle_t *handle; unsigned int credits; int ret = 0; @@ -3638,23 +3539,16 @@ int ext4_punch_hole(struct file *file, loff_t offset, loff_t length) offset; } - first_page = (offset + PAGE_CACHE_SIZE - 1) >> PAGE_CACHE_SHIFT; - last_page = (offset + length) >> PAGE_CACHE_SHIFT; + first_block_offset = round_up(offset, sb->s_blocksize); + last_block_offset = round_down((offset + length), sb->s_blocksize) - 1; - first_page_offset = first_page << PAGE_CACHE_SHIFT; - last_page_offset = last_page << PAGE_CACHE_SHIFT; - - /* Now release the pages */ - if (last_page_offset > first_page_offset) { - truncate_pagecache_range(inode, first_page_offset, - last_page_offset - 1); - } + /* Now release the pages and zero block aligned part of pages*/ + if (last_block_offset > first_block_offset) + truncate_pagecache_range(inode, first_block_offset, + last_block_offset); /* Wait all existing dio workers, newcomers will block on i_mutex */ ext4_inode_block_unlocked_dio(inode); - ret = ext4_flush_unwritten_io(inode); - if (ret) - goto out_dio; inode_dio_wait(inode); if (ext4_test_inode_flag(inode, EXT4_INODE_EXTENTS)) @@ -3668,66 +3562,10 @@ int ext4_punch_hole(struct file *file, loff_t offset, loff_t length) goto out_dio; } - /* - * Now we need to zero out the non-page-aligned data in the - * pages at the start and tail of the hole, and unmap the - * buffer heads for the block aligned regions of the page that - * were completely zeroed. - */ - if (first_page > last_page) { - /* - * If the file space being truncated is contained - * within a page just zero out and unmap the middle of - * that page - */ - ret = ext4_discard_partial_page_buffers(handle, - mapping, offset, length, 0); - - if (ret) - goto out_stop; - } else { - /* - * zero out and unmap the partial page that contains - * the start of the hole - */ - page_len = first_page_offset - offset; - if (page_len > 0) { - ret = ext4_discard_partial_page_buffers(handle, mapping, - offset, page_len, 0); - if (ret) - goto out_stop; - } - - /* - * zero out and unmap the partial page that contains - * the end of the hole - */ - page_len = offset + length - last_page_offset; - if (page_len > 0) { - ret = ext4_discard_partial_page_buffers(handle, mapping, - last_page_offset, page_len, 0); - if (ret) - goto out_stop; - } - } - - /* - * If i_size is contained in the last page, we need to - * unmap and zero the partial page after i_size - */ - if (inode->i_size >> PAGE_CACHE_SHIFT == last_page && - inode->i_size % PAGE_CACHE_SIZE != 0) { - page_len = PAGE_CACHE_SIZE - - (inode->i_size & (PAGE_CACHE_SIZE - 1)); - - if (page_len > 0) { - ret = ext4_discard_partial_page_buffers(handle, - mapping, inode->i_size, page_len, 0); - - if (ret) - goto out_stop; - } - } + ret = ext4_zero_partial_blocks(handle, inode, offset, + length); + if (ret) + goto out_stop; first_block = (offset + sb->s_blocksize - 1) >> EXT4_BLOCK_SIZE_BITS(sb); @@ -3803,7 +3641,6 @@ void ext4_truncate(struct inode *inode) unsigned int credits; handle_t *handle; struct address_space *mapping = inode->i_mapping; - loff_t page_len; /* * There is a possibility that we're either freeing the inode @@ -3830,12 +3667,6 @@ void ext4_truncate(struct inode *inode) return; } - /* - * finish any pending end_io work so we won't run the risk of - * converting any truncated blocks to initialized later - */ - ext4_flush_unwritten_io(inode); - if (ext4_test_inode_flag(inode, EXT4_INODE_EXTENTS)) credits = ext4_writepage_trans_blocks(inode); else @@ -3847,14 +3678,8 @@ void ext4_truncate(struct inode *inode) return; } - if (inode->i_size % PAGE_CACHE_SIZE != 0) { - page_len = PAGE_CACHE_SIZE - - (inode->i_size & (PAGE_CACHE_SIZE - 1)); - - if (ext4_discard_partial_page_buffers(handle, - mapping, inode->i_size, page_len, 0)) - goto out_stop; - } + if (inode->i_size & (inode->i_sb->s_blocksize - 1)) + ext4_block_truncate_page(handle, mapping, inode->i_size); /* * We add the inode to the orphan list, so that if this @@ -4623,7 +4448,8 @@ static void ext4_wait_for_tail_page_commit(struct inode *inode) inode->i_size >> PAGE_CACHE_SHIFT); if (!page) return; - ret = __ext4_journalled_invalidatepage(page, offset); + ret = __ext4_journalled_invalidatepage(page, offset, + PAGE_CACHE_SIZE - offset); unlock_page(page); page_cache_release(page); if (ret != -EBUSY) @@ -4805,7 +4631,7 @@ int ext4_getattr(struct vfsmount *mnt, struct dentry *dentry, struct kstat *stat) { struct inode *inode; - unsigned long delalloc_blocks; + unsigned long long delalloc_blocks; inode = dentry->d_inode; generic_fillattr(inode, stat); @@ -4823,15 +4649,16 @@ int ext4_getattr(struct vfsmount *mnt, struct dentry *dentry, delalloc_blocks = EXT4_C2B(EXT4_SB(inode->i_sb), EXT4_I(inode)->i_reserved_data_blocks); - stat->blocks += (delalloc_blocks << inode->i_sb->s_blocksize_bits)>>9; + stat->blocks += delalloc_blocks << (inode->i_sb->s_blocksize_bits-9); return 0; } -static int ext4_index_trans_blocks(struct inode *inode, int nrblocks, int chunk) +static int ext4_index_trans_blocks(struct inode *inode, int lblocks, + int pextents) { if (!(ext4_test_inode_flag(inode, EXT4_INODE_EXTENTS))) - return ext4_ind_trans_blocks(inode, nrblocks, chunk); - return ext4_ext_index_trans_blocks(inode, nrblocks, chunk); + return ext4_ind_trans_blocks(inode, lblocks); + return ext4_ext_index_trans_blocks(inode, pextents); } /* @@ -4845,7 +4672,8 @@ static int ext4_index_trans_blocks(struct inode *inode, int nrblocks, int chunk) * * Also account for superblock, inode, quota and xattr blocks */ -static int ext4_meta_trans_blocks(struct inode *inode, int nrblocks, int chunk) +static int ext4_meta_trans_blocks(struct inode *inode, int lblocks, + int pextents) { ext4_group_t groups, ngroups = ext4_get_groups_count(inode->i_sb); int gdpblocks; @@ -4853,14 +4681,10 @@ static int ext4_meta_trans_blocks(struct inode *inode, int nrblocks, int chunk) int ret = 0; /* - * How many index blocks need to touch to modify nrblocks? - * The "Chunk" flag indicating whether the nrblocks is - * physically contiguous on disk - * - * For Direct IO and fallocate, they calls get_block to allocate - * one single extent at a time, so they could set the "Chunk" flag + * How many index blocks need to touch to map @lblocks logical blocks + * to @pextents physical extents? */ - idxblocks = ext4_index_trans_blocks(inode, nrblocks, chunk); + idxblocks = ext4_index_trans_blocks(inode, lblocks, pextents); ret = idxblocks; @@ -4868,12 +4692,7 @@ static int ext4_meta_trans_blocks(struct inode *inode, int nrblocks, int chunk) * Now let's see how many group bitmaps and group descriptors need * to account */ - groups = idxblocks; - if (chunk) - groups += 1; - else - groups += nrblocks; - + groups = idxblocks + pextents; gdpblocks = groups; if (groups > ngroups) groups = ngroups; @@ -4904,7 +4723,7 @@ int ext4_writepage_trans_blocks(struct inode *inode) int bpp = ext4_journal_blocks_per_page(inode); int ret; - ret = ext4_meta_trans_blocks(inode, bpp, 0); + ret = ext4_meta_trans_blocks(inode, bpp, bpp); /* Account for data blocks for journalled mode */ if (ext4_should_journal_data(inode)) diff --git a/fs/ext4/mballoc.c b/fs/ext4/mballoc.c index def84082a9a9..a9ff5e5137ca 100644 --- a/fs/ext4/mballoc.c +++ b/fs/ext4/mballoc.c @@ -2105,6 +2105,7 @@ repeat: group = ac->ac_g_ex.fe_group; for (i = 0; i < ngroups; group++, i++) { + cond_resched(); /* * Artificially restricted ngroups for non-extent * files makes group > ngroups possible on first loop. @@ -4405,17 +4406,20 @@ ext4_fsblk_t ext4_mb_new_blocks(handle_t *handle, repeat: /* allocate space in core */ *errp = ext4_mb_regular_allocator(ac); - if (*errp) { - ext4_discard_allocated_blocks(ac); - goto errout; - } + if (*errp) + goto discard_and_exit; /* as we've just preallocated more space than - * user requested orinally, we store allocated + * user requested originally, we store allocated * space in a special descriptor */ if (ac->ac_status == AC_STATUS_FOUND && - ac->ac_o_ex.fe_len < ac->ac_b_ex.fe_len) - ext4_mb_new_preallocation(ac); + ac->ac_o_ex.fe_len < ac->ac_b_ex.fe_len) + *errp = ext4_mb_new_preallocation(ac); + if (*errp) { + discard_and_exit: + ext4_discard_allocated_blocks(ac); + goto errout; + } } if (likely(ac->ac_status == AC_STATUS_FOUND)) { *errp = ext4_mb_mark_diskspace_used(ac, handle, reserv_clstrs); @@ -4612,10 +4616,11 @@ void ext4_free_blocks(handle_t *handle, struct inode *inode, BUG_ON(bh && (count > 1)); for (i = 0; i < count; i++) { + cond_resched(); if (!bh) tbh = sb_find_get_block(inode->i_sb, block + i); - if (unlikely(!tbh)) + if (!tbh) continue; ext4_forget(handle, flags & EXT4_FREE_BLOCKS_METADATA, inode, tbh, block + i); diff --git a/fs/ext4/move_extent.c b/fs/ext4/move_extent.c index 3dcbf364022f..e86dddbd8296 100644 --- a/fs/ext4/move_extent.c +++ b/fs/ext4/move_extent.c @@ -912,7 +912,6 @@ move_extent_per_page(struct file *o_filp, struct inode *donor_inode, struct page *pagep[2] = {NULL, NULL}; handle_t *handle; ext4_lblk_t orig_blk_offset; - long long offs = orig_page_offset << PAGE_CACHE_SHIFT; unsigned long blocksize = orig_inode->i_sb->s_blocksize; unsigned int w_flags = 0; unsigned int tmp_data_size, data_size, replaced_size; @@ -940,8 +939,6 @@ again: orig_blk_offset = orig_page_offset * blocks_per_page + data_offset_in_page; - offs = (long long)orig_blk_offset << orig_inode->i_blkbits; - /* Calculate data_size */ if ((orig_blk_offset + block_len_in_page - 1) == ((orig_inode->i_size - 1) >> orig_inode->i_blkbits)) { diff --git a/fs/ext4/namei.c b/fs/ext4/namei.c index f91002f8c017..234b834d5a97 100644 --- a/fs/ext4/namei.c +++ b/fs/ext4/namei.c @@ -918,11 +918,8 @@ static int htree_dirblock_to_tree(struct file *dir_file, bh->b_data, bh->b_size, (block<<EXT4_BLOCK_SIZE_BITS(dir->i_sb)) + ((char *)de - bh->b_data))) { - /* On error, skip the f_pos to the next block. */ - dir_file->f_pos = (dir_file->f_pos | - (dir->i_sb->s_blocksize - 1)) + 1; - brelse(bh); - return count; + /* silently ignore the rest of the block */ + break; } ext4fs_dirhash(de->name, de->name_len, hinfo); if ((hinfo->hash < start_hash) || diff --git a/fs/ext4/page-io.c b/fs/ext4/page-io.c index 4acf1f78881b..48786cdb5e6c 100644 --- a/fs/ext4/page-io.c +++ b/fs/ext4/page-io.c @@ -46,46 +46,121 @@ void ext4_exit_pageio(void) } /* - * This function is called by ext4_evict_inode() to make sure there is - * no more pending I/O completion work left to do. + * Print an buffer I/O error compatible with the fs/buffer.c. This + * provides compatibility with dmesg scrapers that look for a specific + * buffer I/O error message. We really need a unified error reporting + * structure to userspace ala Digital Unix's uerf system, but it's + * probably not going to happen in my lifetime, due to LKML politics... */ -void ext4_ioend_shutdown(struct inode *inode) +static void buffer_io_error(struct buffer_head *bh) { - wait_queue_head_t *wq = ext4_ioend_wq(inode); + char b[BDEVNAME_SIZE]; + printk(KERN_ERR "Buffer I/O error on device %s, logical block %llu\n", + bdevname(bh->b_bdev, b), + (unsigned long long)bh->b_blocknr); +} - wait_event(*wq, (atomic_read(&EXT4_I(inode)->i_ioend_count) == 0)); - /* - * We need to make sure the work structure is finished being - * used before we let the inode get destroyed. - */ - if (work_pending(&EXT4_I(inode)->i_unwritten_work)) - cancel_work_sync(&EXT4_I(inode)->i_unwritten_work); +static void ext4_finish_bio(struct bio *bio) +{ + int i; + int error = !test_bit(BIO_UPTODATE, &bio->bi_flags); + + for (i = 0; i < bio->bi_vcnt; i++) { + struct bio_vec *bvec = &bio->bi_io_vec[i]; + struct page *page = bvec->bv_page; + struct buffer_head *bh, *head; + unsigned bio_start = bvec->bv_offset; + unsigned bio_end = bio_start + bvec->bv_len; + unsigned under_io = 0; + unsigned long flags; + + if (!page) + continue; + + if (error) { + SetPageError(page); + set_bit(AS_EIO, &page->mapping->flags); + } + bh = head = page_buffers(page); + /* + * We check all buffers in the page under BH_Uptodate_Lock + * to avoid races with other end io clearing async_write flags + */ + local_irq_save(flags); + bit_spin_lock(BH_Uptodate_Lock, &head->b_state); + do { + if (bh_offset(bh) < bio_start || + bh_offset(bh) + bh->b_size > bio_end) { + if (buffer_async_write(bh)) + under_io++; + continue; + } + clear_buffer_async_write(bh); + if (error) + buffer_io_error(bh); + } while ((bh = bh->b_this_page) != head); + bit_spin_unlock(BH_Uptodate_Lock, &head->b_state); + local_irq_restore(flags); + if (!under_io) + end_page_writeback(page); + } } -void ext4_free_io_end(ext4_io_end_t *io) +static void ext4_release_io_end(ext4_io_end_t *io_end) { - BUG_ON(!io); - BUG_ON(!list_empty(&io->list)); - BUG_ON(io->flag & EXT4_IO_END_UNWRITTEN); + struct bio *bio, *next_bio; + + BUG_ON(!list_empty(&io_end->list)); + BUG_ON(io_end->flag & EXT4_IO_END_UNWRITTEN); + WARN_ON(io_end->handle); - if (atomic_dec_and_test(&EXT4_I(io->inode)->i_ioend_count)) - wake_up_all(ext4_ioend_wq(io->inode)); - kmem_cache_free(io_end_cachep, io); + if (atomic_dec_and_test(&EXT4_I(io_end->inode)->i_ioend_count)) + wake_up_all(ext4_ioend_wq(io_end->inode)); + + for (bio = io_end->bio; bio; bio = next_bio) { + next_bio = bio->bi_private; + ext4_finish_bio(bio); + bio_put(bio); + } + if (io_end->flag & EXT4_IO_END_DIRECT) + inode_dio_done(io_end->inode); + if (io_end->iocb) + aio_complete(io_end->iocb, io_end->result, 0); + kmem_cache_free(io_end_cachep, io_end); } -/* check a range of space and convert unwritten extents to written. */ +static void ext4_clear_io_unwritten_flag(ext4_io_end_t *io_end) +{ + struct inode *inode = io_end->inode; + + io_end->flag &= ~EXT4_IO_END_UNWRITTEN; + /* Wake up anyone waiting on unwritten extent conversion */ + if (atomic_dec_and_test(&EXT4_I(inode)->i_unwritten)) + wake_up_all(ext4_ioend_wq(inode)); +} + +/* + * Check a range of space and convert unwritten extents to written. Note that + * we are protected from truncate touching same part of extent tree by the + * fact that truncate code waits for all DIO to finish (thus exclusion from + * direct IO is achieved) and also waits for PageWriteback bits. Thus we + * cannot get to ext4_ext_truncate() before all IOs overlapping that range are + * completed (happens from ext4_free_ioend()). + */ static int ext4_end_io(ext4_io_end_t *io) { struct inode *inode = io->inode; loff_t offset = io->offset; ssize_t size = io->size; + handle_t *handle = io->handle; int ret = 0; ext4_debug("ext4_end_io_nolock: io 0x%p from inode %lu,list->next 0x%p," "list->prev 0x%p\n", io, inode->i_ino, io->list.next, io->list.prev); - ret = ext4_convert_unwritten_extents(inode, offset, size); + io->handle = NULL; /* Following call will use up the handle */ + ret = ext4_convert_unwritten_extents(handle, inode, offset, size); if (ret < 0) { ext4_msg(inode->i_sb, KERN_EMERG, "failed to convert unwritten extents to written " @@ -93,30 +168,22 @@ static int ext4_end_io(ext4_io_end_t *io) "(inode %lu, offset %llu, size %zd, error %d)", inode->i_ino, offset, size, ret); } - /* Wake up anyone waiting on unwritten extent conversion */ - if (atomic_dec_and_test(&EXT4_I(inode)->i_unwritten)) - wake_up_all(ext4_ioend_wq(inode)); - if (io->flag & EXT4_IO_END_DIRECT) - inode_dio_done(inode); - if (io->iocb) - aio_complete(io->iocb, io->result, 0); + ext4_clear_io_unwritten_flag(io); + ext4_release_io_end(io); return ret; } -static void dump_completed_IO(struct inode *inode) +static void dump_completed_IO(struct inode *inode, struct list_head *head) { #ifdef EXT4FS_DEBUG struct list_head *cur, *before, *after; ext4_io_end_t *io, *io0, *io1; - if (list_empty(&EXT4_I(inode)->i_completed_io_list)) { - ext4_debug("inode %lu completed_io list is empty\n", - inode->i_ino); + if (list_empty(head)) return; - } - ext4_debug("Dump inode %lu completed_io list\n", inode->i_ino); - list_for_each_entry(io, &EXT4_I(inode)->i_completed_io_list, list) { + ext4_debug("Dump inode %lu completed io list\n", inode->i_ino); + list_for_each_entry(io, head, list) { cur = &io->list; before = cur->prev; io0 = container_of(before, ext4_io_end_t, list); @@ -130,23 +197,30 @@ static void dump_completed_IO(struct inode *inode) } /* Add the io_end to per-inode completed end_io list. */ -void ext4_add_complete_io(ext4_io_end_t *io_end) +static void ext4_add_complete_io(ext4_io_end_t *io_end) { struct ext4_inode_info *ei = EXT4_I(io_end->inode); struct workqueue_struct *wq; unsigned long flags; BUG_ON(!(io_end->flag & EXT4_IO_END_UNWRITTEN)); - wq = EXT4_SB(io_end->inode->i_sb)->dio_unwritten_wq; - spin_lock_irqsave(&ei->i_completed_io_lock, flags); - if (list_empty(&ei->i_completed_io_list)) - queue_work(wq, &ei->i_unwritten_work); - list_add_tail(&io_end->list, &ei->i_completed_io_list); + if (io_end->handle) { + wq = EXT4_SB(io_end->inode->i_sb)->rsv_conversion_wq; + if (list_empty(&ei->i_rsv_conversion_list)) + queue_work(wq, &ei->i_rsv_conversion_work); + list_add_tail(&io_end->list, &ei->i_rsv_conversion_list); + } else { + wq = EXT4_SB(io_end->inode->i_sb)->unrsv_conversion_wq; + if (list_empty(&ei->i_unrsv_conversion_list)) + queue_work(wq, &ei->i_unrsv_conversion_work); + list_add_tail(&io_end->list, &ei->i_unrsv_conversion_list); + } spin_unlock_irqrestore(&ei->i_completed_io_lock, flags); } -static int ext4_do_flush_completed_IO(struct inode *inode) +static int ext4_do_flush_completed_IO(struct inode *inode, + struct list_head *head) { ext4_io_end_t *io; struct list_head unwritten; @@ -155,8 +229,8 @@ static int ext4_do_flush_completed_IO(struct inode *inode) int err, ret = 0; spin_lock_irqsave(&ei->i_completed_io_lock, flags); - dump_completed_IO(inode); - list_replace_init(&ei->i_completed_io_list, &unwritten); + dump_completed_IO(inode, head); + list_replace_init(head, &unwritten); spin_unlock_irqrestore(&ei->i_completed_io_lock, flags); while (!list_empty(&unwritten)) { @@ -167,30 +241,25 @@ static int ext4_do_flush_completed_IO(struct inode *inode) err = ext4_end_io(io); if (unlikely(!ret && err)) ret = err; - io->flag &= ~EXT4_IO_END_UNWRITTEN; - ext4_free_io_end(io); } return ret; } /* - * work on completed aio dio IO, to convert unwritten extents to extents + * work on completed IO, to convert unwritten extents to extents */ -void ext4_end_io_work(struct work_struct *work) +void ext4_end_io_rsv_work(struct work_struct *work) { struct ext4_inode_info *ei = container_of(work, struct ext4_inode_info, - i_unwritten_work); - ext4_do_flush_completed_IO(&ei->vfs_inode); + i_rsv_conversion_work); + ext4_do_flush_completed_IO(&ei->vfs_inode, &ei->i_rsv_conversion_list); } -int ext4_flush_unwritten_io(struct inode *inode) +void ext4_end_io_unrsv_work(struct work_struct *work) { - int ret; - WARN_ON_ONCE(!mutex_is_locked(&inode->i_mutex) && - !(inode->i_state & I_FREEING)); - ret = ext4_do_flush_completed_IO(inode); - ext4_unwritten_wait(inode); - return ret; + struct ext4_inode_info *ei = container_of(work, struct ext4_inode_info, + i_unrsv_conversion_work); + ext4_do_flush_completed_IO(&ei->vfs_inode, &ei->i_unrsv_conversion_list); } ext4_io_end_t *ext4_init_io_end(struct inode *inode, gfp_t flags) @@ -200,83 +269,70 @@ ext4_io_end_t *ext4_init_io_end(struct inode *inode, gfp_t flags) atomic_inc(&EXT4_I(inode)->i_ioend_count); io->inode = inode; INIT_LIST_HEAD(&io->list); + atomic_set(&io->count, 1); } return io; } -/* - * Print an buffer I/O error compatible with the fs/buffer.c. This - * provides compatibility with dmesg scrapers that look for a specific - * buffer I/O error message. We really need a unified error reporting - * structure to userspace ala Digital Unix's uerf system, but it's - * probably not going to happen in my lifetime, due to LKML politics... - */ -static void buffer_io_error(struct buffer_head *bh) +void ext4_put_io_end_defer(ext4_io_end_t *io_end) { - char b[BDEVNAME_SIZE]; - printk(KERN_ERR "Buffer I/O error on device %s, logical block %llu\n", - bdevname(bh->b_bdev, b), - (unsigned long long)bh->b_blocknr); + if (atomic_dec_and_test(&io_end->count)) { + if (!(io_end->flag & EXT4_IO_END_UNWRITTEN) || !io_end->size) { + ext4_release_io_end(io_end); + return; + } + ext4_add_complete_io(io_end); + } +} + +int ext4_put_io_end(ext4_io_end_t *io_end) +{ + int err = 0; + + if (atomic_dec_and_test(&io_end->count)) { + if (io_end->flag & EXT4_IO_END_UNWRITTEN) { + err = ext4_convert_unwritten_extents(io_end->handle, + io_end->inode, io_end->offset, + io_end->size); + io_end->handle = NULL; + ext4_clear_io_unwritten_flag(io_end); + } + ext4_release_io_end(io_end); + } + return err; +} + +ext4_io_end_t *ext4_get_io_end(ext4_io_end_t *io_end) +{ + atomic_inc(&io_end->count); + return io_end; } static void ext4_end_bio(struct bio *bio, int error) { ext4_io_end_t *io_end = bio->bi_private; - struct inode *inode; - int i; - int blocksize; sector_t bi_sector = bio->bi_sector; BUG_ON(!io_end); - inode = io_end->inode; - blocksize = 1 << inode->i_blkbits; - bio->bi_private = NULL; bio->bi_end_io = NULL; if (test_bit(BIO_UPTODATE, &bio->bi_flags)) error = 0; - for (i = 0; i < bio->bi_vcnt; i++) { - struct bio_vec *bvec = &bio->bi_io_vec[i]; - struct page *page = bvec->bv_page; - struct buffer_head *bh, *head; - unsigned bio_start = bvec->bv_offset; - unsigned bio_end = bio_start + bvec->bv_len; - unsigned under_io = 0; - unsigned long flags; - if (!page) - continue; - - if (error) { - SetPageError(page); - set_bit(AS_EIO, &page->mapping->flags); - } - bh = head = page_buffers(page); + if (io_end->flag & EXT4_IO_END_UNWRITTEN) { /* - * We check all buffers in the page under BH_Uptodate_Lock - * to avoid races with other end io clearing async_write flags + * Link bio into list hanging from io_end. We have to do it + * atomically as bio completions can be racing against each + * other. */ - local_irq_save(flags); - bit_spin_lock(BH_Uptodate_Lock, &head->b_state); - do { - if (bh_offset(bh) < bio_start || - bh_offset(bh) + blocksize > bio_end) { - if (buffer_async_write(bh)) - under_io++; - continue; - } - clear_buffer_async_write(bh); - if (error) - buffer_io_error(bh); - } while ((bh = bh->b_this_page) != head); - bit_spin_unlock(BH_Uptodate_Lock, &head->b_state); - local_irq_restore(flags); - if (!under_io) - end_page_writeback(page); + bio->bi_private = xchg(&io_end->bio, bio); + } else { + ext4_finish_bio(bio); + bio_put(bio); } - bio_put(bio); if (error) { - io_end->flag |= EXT4_IO_END_ERROR; + struct inode *inode = io_end->inode; + ext4_warning(inode->i_sb, "I/O error writing to inode %lu " "(offset %llu size %ld starting block %llu)", inode->i_ino, @@ -285,13 +341,7 @@ static void ext4_end_bio(struct bio *bio, int error) (unsigned long long) bi_sector >> (inode->i_blkbits - 9)); } - - if (!(io_end->flag & EXT4_IO_END_UNWRITTEN)) { - ext4_free_io_end(io_end); - return; - } - - ext4_add_complete_io(io_end); + ext4_put_io_end_defer(io_end); } void ext4_io_submit(struct ext4_io_submit *io) @@ -305,43 +355,38 @@ void ext4_io_submit(struct ext4_io_submit *io) bio_put(io->io_bio); } io->io_bio = NULL; - io->io_op = 0; +} + +void ext4_io_submit_init(struct ext4_io_submit *io, + struct writeback_control *wbc) +{ + io->io_op = (wbc->sync_mode == WB_SYNC_ALL ? WRITE_SYNC : WRITE); + io->io_bio = NULL; io->io_end = NULL; } -static int io_submit_init(struct ext4_io_submit *io, - struct inode *inode, - struct writeback_control *wbc, - struct buffer_head *bh) +static int io_submit_init_bio(struct ext4_io_submit *io, + struct buffer_head *bh) { - ext4_io_end_t *io_end; - struct page *page = bh->b_page; int nvecs = bio_get_nr_vecs(bh->b_bdev); struct bio *bio; - io_end = ext4_init_io_end(inode, GFP_NOFS); - if (!io_end) - return -ENOMEM; bio = bio_alloc(GFP_NOIO, min(nvecs, BIO_MAX_PAGES)); + if (!bio) + return -ENOMEM; bio->bi_sector = bh->b_blocknr * (bh->b_size >> 9); bio->bi_bdev = bh->b_bdev; - bio->bi_private = io->io_end = io_end; bio->bi_end_io = ext4_end_bio; - - io_end->offset = (page->index << PAGE_CACHE_SHIFT) + bh_offset(bh); - + bio->bi_private = ext4_get_io_end(io->io_end); io->io_bio = bio; - io->io_op = (wbc->sync_mode == WB_SYNC_ALL ? WRITE_SYNC : WRITE); io->io_next_block = bh->b_blocknr; return 0; } static int io_submit_add_bh(struct ext4_io_submit *io, struct inode *inode, - struct writeback_control *wbc, struct buffer_head *bh) { - ext4_io_end_t *io_end; int ret; if (io->io_bio && bh->b_blocknr != io->io_next_block) { @@ -349,18 +394,14 @@ submit_and_retry: ext4_io_submit(io); } if (io->io_bio == NULL) { - ret = io_submit_init(io, inode, wbc, bh); + ret = io_submit_init_bio(io, bh); if (ret) return ret; } - io_end = io->io_end; - if (test_clear_buffer_uninit(bh)) - ext4_set_io_unwritten_flag(inode, io_end); - io->io_end->size += bh->b_size; - io->io_next_block++; ret = bio_add_page(io->io_bio, bh->b_page, bh->b_size, bh_offset(bh)); if (ret != bh->b_size) goto submit_and_retry; + io->io_next_block++; return 0; } @@ -432,7 +473,7 @@ int ext4_bio_write_page(struct ext4_io_submit *io, do { if (!buffer_async_write(bh)) continue; - ret = io_submit_add_bh(io, inode, wbc, bh); + ret = io_submit_add_bh(io, inode, bh); if (ret) { /* * We only get here on ENOMEM. Not much else diff --git a/fs/ext4/resize.c b/fs/ext4/resize.c index b27c96d01965..c5adbb318a90 100644 --- a/fs/ext4/resize.c +++ b/fs/ext4/resize.c @@ -79,12 +79,20 @@ static int verify_group_input(struct super_block *sb, ext4_fsblk_t end = start + input->blocks_count; ext4_group_t group = input->group; ext4_fsblk_t itend = input->inode_table + sbi->s_itb_per_group; - unsigned overhead = ext4_group_overhead_blocks(sb, group); - ext4_fsblk_t metaend = start + overhead; + unsigned overhead; + ext4_fsblk_t metaend; struct buffer_head *bh = NULL; ext4_grpblk_t free_blocks_count, offset; int err = -EINVAL; + if (group != sbi->s_groups_count) { + ext4_warning(sb, "Cannot add at group %u (only %u groups)", + input->group, sbi->s_groups_count); + return -EINVAL; + } + + overhead = ext4_group_overhead_blocks(sb, group); + metaend = start + overhead; input->free_blocks_count = free_blocks_count = input->blocks_count - 2 - overhead - sbi->s_itb_per_group; @@ -96,10 +104,7 @@ static int verify_group_input(struct super_block *sb, free_blocks_count, input->reserved_blocks); ext4_get_group_no_and_offset(sb, start, NULL, &offset); - if (group != sbi->s_groups_count) - ext4_warning(sb, "Cannot add at group %u (only %u groups)", - input->group, sbi->s_groups_count); - else if (offset != 0) + if (offset != 0) ext4_warning(sb, "Last group not full"); else if (input->reserved_blocks > input->blocks_count / 5) ext4_warning(sb, "Reserved blocks too high (%u)", @@ -1551,11 +1556,10 @@ int ext4_group_add(struct super_block *sb, struct ext4_new_group_data *input) int reserved_gdb = ext4_bg_has_super(sb, input->group) ? le16_to_cpu(es->s_reserved_gdt_blocks) : 0; struct inode *inode = NULL; - int gdb_off, gdb_num; + int gdb_off; int err; __u16 bg_flags = 0; - gdb_num = input->group / EXT4_DESC_PER_BLOCK(sb); gdb_off = input->group % EXT4_DESC_PER_BLOCK(sb); if (gdb_off == 0 && !EXT4_HAS_RO_COMPAT_FEATURE(sb, @@ -1656,12 +1660,10 @@ errout: err = err2; if (!err) { - ext4_fsblk_t first_block; - first_block = ext4_group_first_block_no(sb, 0); if (test_opt(sb, DEBUG)) printk(KERN_DEBUG "EXT4-fs: extended group to %llu " "blocks\n", ext4_blocks_count(es)); - update_backups(sb, EXT4_SB(sb)->s_sbh->b_blocknr - first_block, + update_backups(sb, EXT4_SB(sb)->s_sbh->b_blocknr, (char *)es, sizeof(struct ext4_super_block), 0); } return err; diff --git a/fs/ext4/super.c b/fs/ext4/super.c index 94cc84db7c9a..85b3dd60169b 100644 --- a/fs/ext4/super.c +++ b/fs/ext4/super.c @@ -69,6 +69,7 @@ static void ext4_mark_recovery_complete(struct super_block *sb, static void ext4_clear_journal_err(struct super_block *sb, struct ext4_super_block *es); static int ext4_sync_fs(struct super_block *sb, int wait); +static int ext4_sync_fs_nojournal(struct super_block *sb, int wait); static int ext4_remount(struct super_block *sb, int *flags, char *data); static int ext4_statfs(struct dentry *dentry, struct kstatfs *buf); static int ext4_unfreeze(struct super_block *sb); @@ -398,6 +399,11 @@ static void ext4_handle_error(struct super_block *sb) } if (test_opt(sb, ERRORS_RO)) { ext4_msg(sb, KERN_CRIT, "Remounting filesystem read-only"); + /* + * Make sure updated value of ->s_mount_flags will be visible + * before ->s_flags update + */ + smp_wmb(); sb->s_flags |= MS_RDONLY; } if (test_opt(sb, ERRORS_PANIC)) @@ -422,9 +428,9 @@ void __ext4_error(struct super_block *sb, const char *function, ext4_handle_error(sb); } -void ext4_error_inode(struct inode *inode, const char *function, - unsigned int line, ext4_fsblk_t block, - const char *fmt, ...) +void __ext4_error_inode(struct inode *inode, const char *function, + unsigned int line, ext4_fsblk_t block, + const char *fmt, ...) { va_list args; struct va_format vaf; @@ -451,9 +457,9 @@ void ext4_error_inode(struct inode *inode, const char *function, ext4_handle_error(inode->i_sb); } -void ext4_error_file(struct file *file, const char *function, - unsigned int line, ext4_fsblk_t block, - const char *fmt, ...) +void __ext4_error_file(struct file *file, const char *function, + unsigned int line, ext4_fsblk_t block, + const char *fmt, ...) { va_list args; struct va_format vaf; @@ -570,8 +576,13 @@ void __ext4_abort(struct super_block *sb, const char *function, if ((sb->s_flags & MS_RDONLY) == 0) { ext4_msg(sb, KERN_CRIT, "Remounting filesystem read-only"); - sb->s_flags |= MS_RDONLY; EXT4_SB(sb)->s_mount_flags |= EXT4_MF_FS_ABORTED; + /* + * Make sure updated value of ->s_mount_flags will be visible + * before ->s_flags update + */ + smp_wmb(); + sb->s_flags |= MS_RDONLY; if (EXT4_SB(sb)->s_journal) jbd2_journal_abort(EXT4_SB(sb)->s_journal, -EIO); save_error_info(sb, function, line); @@ -580,7 +591,8 @@ void __ext4_abort(struct super_block *sb, const char *function, panic("EXT4-fs panic from previous error\n"); } -void ext4_msg(struct super_block *sb, const char *prefix, const char *fmt, ...) +void __ext4_msg(struct super_block *sb, + const char *prefix, const char *fmt, ...) { struct va_format vaf; va_list args; @@ -750,8 +762,10 @@ static void ext4_put_super(struct super_block *sb) ext4_unregister_li_request(sb); dquot_disable(sb, -1, DQUOT_USAGE_ENABLED | DQUOT_LIMITS_ENABLED); - flush_workqueue(sbi->dio_unwritten_wq); - destroy_workqueue(sbi->dio_unwritten_wq); + flush_workqueue(sbi->unrsv_conversion_wq); + flush_workqueue(sbi->rsv_conversion_wq); + destroy_workqueue(sbi->unrsv_conversion_wq); + destroy_workqueue(sbi->rsv_conversion_wq); if (sbi->s_journal) { err = jbd2_journal_destroy(sbi->s_journal); @@ -760,7 +774,7 @@ static void ext4_put_super(struct super_block *sb) ext4_abort(sb, "Couldn't clean up the journal"); } - ext4_es_unregister_shrinker(sb); + ext4_es_unregister_shrinker(sbi); del_timer(&sbi->s_err_report); ext4_release_system_zone(sb); ext4_mb_release(sb); @@ -849,6 +863,7 @@ static struct inode *ext4_alloc_inode(struct super_block *sb) rwlock_init(&ei->i_es_lock); INIT_LIST_HEAD(&ei->i_es_lru); ei->i_es_lru_nr = 0; + ei->i_touch_when = 0; ei->i_reserved_data_blocks = 0; ei->i_reserved_meta_blocks = 0; ei->i_allocated_meta_blocks = 0; @@ -859,13 +874,15 @@ static struct inode *ext4_alloc_inode(struct super_block *sb) ei->i_reserved_quota = 0; #endif ei->jinode = NULL; - INIT_LIST_HEAD(&ei->i_completed_io_list); + INIT_LIST_HEAD(&ei->i_rsv_conversion_list); + INIT_LIST_HEAD(&ei->i_unrsv_conversion_list); spin_lock_init(&ei->i_completed_io_lock); ei->i_sync_tid = 0; ei->i_datasync_tid = 0; atomic_set(&ei->i_ioend_count, 0); atomic_set(&ei->i_unwritten, 0); - INIT_WORK(&ei->i_unwritten_work, ext4_end_io_work); + INIT_WORK(&ei->i_rsv_conversion_work, ext4_end_io_rsv_work); + INIT_WORK(&ei->i_unrsv_conversion_work, ext4_end_io_unrsv_work); return &ei->vfs_inode; } @@ -1093,6 +1110,7 @@ static const struct super_operations ext4_nojournal_sops = { .dirty_inode = ext4_dirty_inode, .drop_inode = ext4_drop_inode, .evict_inode = ext4_evict_inode, + .sync_fs = ext4_sync_fs_nojournal, .put_super = ext4_put_super, .statfs = ext4_statfs, .remount_fs = ext4_remount, @@ -1908,7 +1926,6 @@ static int ext4_fill_flex_info(struct super_block *sb) struct ext4_sb_info *sbi = EXT4_SB(sb); struct ext4_group_desc *gdp = NULL; ext4_group_t flex_group; - unsigned int groups_per_flex = 0; int i, err; sbi->s_log_groups_per_flex = sbi->s_es->s_log_groups_per_flex; @@ -1916,7 +1933,6 @@ static int ext4_fill_flex_info(struct super_block *sb) sbi->s_log_groups_per_flex = 0; return 1; } - groups_per_flex = 1U << sbi->s_log_groups_per_flex; err = ext4_alloc_flex_bg_array(sb, sbi->s_groups_count); if (err) @@ -2164,19 +2180,22 @@ static void ext4_orphan_cleanup(struct super_block *sb, list_add(&EXT4_I(inode)->i_orphan, &EXT4_SB(sb)->s_orphan); dquot_initialize(inode); if (inode->i_nlink) { - ext4_msg(sb, KERN_DEBUG, - "%s: truncating inode %lu to %lld bytes", - __func__, inode->i_ino, inode->i_size); + if (test_opt(sb, DEBUG)) + ext4_msg(sb, KERN_DEBUG, + "%s: truncating inode %lu to %lld bytes", + __func__, inode->i_ino, inode->i_size); jbd_debug(2, "truncating inode %lu to %lld bytes\n", inode->i_ino, inode->i_size); mutex_lock(&inode->i_mutex); + truncate_inode_pages(inode->i_mapping, inode->i_size); ext4_truncate(inode); mutex_unlock(&inode->i_mutex); nr_truncates++; } else { - ext4_msg(sb, KERN_DEBUG, - "%s: deleting unreferenced inode %lu", - __func__, inode->i_ino); + if (test_opt(sb, DEBUG)) + ext4_msg(sb, KERN_DEBUG, + "%s: deleting unreferenced inode %lu", + __func__, inode->i_ino); jbd_debug(2, "deleting unreferenced inode %lu\n", inode->i_ino); nr_orphans++; @@ -2377,7 +2396,10 @@ struct ext4_attr { ssize_t (*show)(struct ext4_attr *, struct ext4_sb_info *, char *); ssize_t (*store)(struct ext4_attr *, struct ext4_sb_info *, const char *, size_t); - int offset; + union { + int offset; + int deprecated_val; + } u; }; static int parse_strtoull(const char *buf, @@ -2446,7 +2468,7 @@ static ssize_t inode_readahead_blks_store(struct ext4_attr *a, static ssize_t sbi_ui_show(struct ext4_attr *a, struct ext4_sb_info *sbi, char *buf) { - unsigned int *ui = (unsigned int *) (((char *) sbi) + a->offset); + unsigned int *ui = (unsigned int *) (((char *) sbi) + a->u.offset); return snprintf(buf, PAGE_SIZE, "%u\n", *ui); } @@ -2455,7 +2477,7 @@ static ssize_t sbi_ui_store(struct ext4_attr *a, struct ext4_sb_info *sbi, const char *buf, size_t count) { - unsigned int *ui = (unsigned int *) (((char *) sbi) + a->offset); + unsigned int *ui = (unsigned int *) (((char *) sbi) + a->u.offset); unsigned long t; int ret; @@ -2504,12 +2526,20 @@ static ssize_t trigger_test_error(struct ext4_attr *a, return count; } +static ssize_t sbi_deprecated_show(struct ext4_attr *a, + struct ext4_sb_info *sbi, char *buf) +{ + return snprintf(buf, PAGE_SIZE, "%d\n", a->u.deprecated_val); +} + #define EXT4_ATTR_OFFSET(_name,_mode,_show,_store,_elname) \ static struct ext4_attr ext4_attr_##_name = { \ .attr = {.name = __stringify(_name), .mode = _mode }, \ .show = _show, \ .store = _store, \ - .offset = offsetof(struct ext4_sb_info, _elname), \ + .u = { \ + .offset = offsetof(struct ext4_sb_info, _elname),\ + }, \ } #define EXT4_ATTR(name, mode, show, store) \ static struct ext4_attr ext4_attr_##name = __ATTR(name, mode, show, store) @@ -2520,6 +2550,14 @@ static struct ext4_attr ext4_attr_##name = __ATTR(name, mode, show, store) #define EXT4_RW_ATTR_SBI_UI(name, elname) \ EXT4_ATTR_OFFSET(name, 0644, sbi_ui_show, sbi_ui_store, elname) #define ATTR_LIST(name) &ext4_attr_##name.attr +#define EXT4_DEPRECATED_ATTR(_name, _val) \ +static struct ext4_attr ext4_attr_##_name = { \ + .attr = {.name = __stringify(_name), .mode = 0444 }, \ + .show = sbi_deprecated_show, \ + .u = { \ + .deprecated_val = _val, \ + }, \ +} EXT4_RO_ATTR(delayed_allocation_blocks); EXT4_RO_ATTR(session_write_kbytes); @@ -2534,7 +2572,7 @@ EXT4_RW_ATTR_SBI_UI(mb_min_to_scan, s_mb_min_to_scan); EXT4_RW_ATTR_SBI_UI(mb_order2_req, s_mb_order2_reqs); EXT4_RW_ATTR_SBI_UI(mb_stream_req, s_mb_stream_request); EXT4_RW_ATTR_SBI_UI(mb_group_prealloc, s_mb_group_prealloc); -EXT4_RW_ATTR_SBI_UI(max_writeback_mb_bump, s_max_writeback_mb_bump); +EXT4_DEPRECATED_ATTR(max_writeback_mb_bump, 128); EXT4_RW_ATTR_SBI_UI(extent_max_zeroout_kb, s_extent_max_zeroout_kb); EXT4_ATTR(trigger_fs_error, 0200, NULL, trigger_test_error); @@ -3763,7 +3801,7 @@ static int ext4_fill_super(struct super_block *sb, void *data, int silent) sbi->s_err_report.data = (unsigned long) sb; /* Register extent status tree shrinker */ - ext4_es_register_shrinker(sb); + ext4_es_register_shrinker(sbi); err = percpu_counter_init(&sbi->s_freeclusters_counter, ext4_count_free_clusters(sb)); @@ -3787,7 +3825,6 @@ static int ext4_fill_super(struct super_block *sb, void *data, int silent) } sbi->s_stripe = ext4_get_stripe_size(sbi); - sbi->s_max_writeback_mb_bump = 128; sbi->s_extent_max_zeroout_kb = 32; /* @@ -3915,12 +3952,20 @@ no_journal: * The maximum number of concurrent works can be high and * concurrency isn't really necessary. Limit it to 1. */ - EXT4_SB(sb)->dio_unwritten_wq = - alloc_workqueue("ext4-dio-unwritten", WQ_MEM_RECLAIM | WQ_UNBOUND, 1); - if (!EXT4_SB(sb)->dio_unwritten_wq) { - printk(KERN_ERR "EXT4-fs: failed to create DIO workqueue\n"); + EXT4_SB(sb)->rsv_conversion_wq = + alloc_workqueue("ext4-rsv-conversion", WQ_MEM_RECLAIM | WQ_UNBOUND, 1); + if (!EXT4_SB(sb)->rsv_conversion_wq) { + printk(KERN_ERR "EXT4-fs: failed to create workqueue\n"); ret = -ENOMEM; - goto failed_mount_wq; + goto failed_mount4; + } + + EXT4_SB(sb)->unrsv_conversion_wq = + alloc_workqueue("ext4-unrsv-conversion", WQ_MEM_RECLAIM | WQ_UNBOUND, 1); + if (!EXT4_SB(sb)->unrsv_conversion_wq) { + printk(KERN_ERR "EXT4-fs: failed to create workqueue\n"); + ret = -ENOMEM; + goto failed_mount4; } /* @@ -4074,14 +4119,17 @@ failed_mount4a: sb->s_root = NULL; failed_mount4: ext4_msg(sb, KERN_ERR, "mount failed"); - destroy_workqueue(EXT4_SB(sb)->dio_unwritten_wq); + if (EXT4_SB(sb)->rsv_conversion_wq) + destroy_workqueue(EXT4_SB(sb)->rsv_conversion_wq); + if (EXT4_SB(sb)->unrsv_conversion_wq) + destroy_workqueue(EXT4_SB(sb)->unrsv_conversion_wq); failed_mount_wq: if (sbi->s_journal) { jbd2_journal_destroy(sbi->s_journal); sbi->s_journal = NULL; } failed_mount3: - ext4_es_unregister_shrinker(sb); + ext4_es_unregister_shrinker(sbi); del_timer(&sbi->s_err_report); if (sbi->s_flex_groups) ext4_kvfree(sbi->s_flex_groups); @@ -4517,19 +4565,52 @@ static int ext4_sync_fs(struct super_block *sb, int wait) { int ret = 0; tid_t target; + bool needs_barrier = false; struct ext4_sb_info *sbi = EXT4_SB(sb); trace_ext4_sync_fs(sb, wait); - flush_workqueue(sbi->dio_unwritten_wq); + flush_workqueue(sbi->rsv_conversion_wq); + flush_workqueue(sbi->unrsv_conversion_wq); /* * Writeback quota in non-journalled quota case - journalled quota has * no dirty dquots */ dquot_writeback_dquots(sb, -1); + /* + * Data writeback is possible w/o journal transaction, so barrier must + * being sent at the end of the function. But we can skip it if + * transaction_commit will do it for us. + */ + target = jbd2_get_latest_transaction(sbi->s_journal); + if (wait && sbi->s_journal->j_flags & JBD2_BARRIER && + !jbd2_trans_will_send_data_barrier(sbi->s_journal, target)) + needs_barrier = true; + if (jbd2_journal_start_commit(sbi->s_journal, &target)) { if (wait) - jbd2_log_wait_commit(sbi->s_journal, target); + ret = jbd2_log_wait_commit(sbi->s_journal, target); + } + if (needs_barrier) { + int err; + err = blkdev_issue_flush(sb->s_bdev, GFP_KERNEL, NULL); + if (!ret) + ret = err; } + + return ret; +} + +static int ext4_sync_fs_nojournal(struct super_block *sb, int wait) +{ + int ret = 0; + + trace_ext4_sync_fs(sb, wait); + flush_workqueue(EXT4_SB(sb)->rsv_conversion_wq); + flush_workqueue(EXT4_SB(sb)->unrsv_conversion_wq); + dquot_writeback_dquots(sb, -1); + if (wait && test_opt(sb, BARRIER)) + ret = blkdev_issue_flush(sb->s_bdev, GFP_KERNEL, NULL); + return ret; } diff --git a/fs/f2fs/Kconfig b/fs/f2fs/Kconfig index fd27e7e6326e..e06e0995e00f 100644 --- a/fs/f2fs/Kconfig +++ b/fs/f2fs/Kconfig @@ -51,3 +51,15 @@ config F2FS_FS_POSIX_ACL Linux website <http://acl.bestbits.at/>. If you don't know what Access Control Lists are, say N + +config F2FS_FS_SECURITY + bool "F2FS Security Labels" + depends on F2FS_FS_XATTR + help + Security labels provide an access control facility to support Linux + Security Models (LSMs) accepted by AppArmor, SELinux, Smack and TOMOYO + Linux. This option enables an extended attribute handler for file + security labels in the f2fs filesystem, so that it requires enabling + the extended attribute support in advance. + + If you are not using a security module, say N. diff --git a/fs/f2fs/acl.c b/fs/f2fs/acl.c index 44abc2f286e0..b7826ec1b470 100644 --- a/fs/f2fs/acl.c +++ b/fs/f2fs/acl.c @@ -250,7 +250,7 @@ static int f2fs_set_acl(struct inode *inode, int type, struct posix_acl *acl) } } - error = f2fs_setxattr(inode, name_index, "", value, size); + error = f2fs_setxattr(inode, name_index, "", value, size, NULL); kfree(value); if (!error) diff --git a/fs/f2fs/checkpoint.c b/fs/f2fs/checkpoint.c index b1de01da1a40..66a6b85a51d8 100644 --- a/fs/f2fs/checkpoint.c +++ b/fs/f2fs/checkpoint.c @@ -357,8 +357,8 @@ static struct page *validate_checkpoint(struct f2fs_sb_info *sbi, unsigned long blk_size = sbi->blocksize; struct f2fs_checkpoint *cp_block; unsigned long long cur_version = 0, pre_version = 0; - unsigned int crc = 0; size_t crc_offset; + __u32 crc = 0; /* Read the 1st cp block in this CP pack */ cp_page_1 = get_meta_page(sbi, cp_addr); @@ -369,7 +369,7 @@ static struct page *validate_checkpoint(struct f2fs_sb_info *sbi, if (crc_offset >= blk_size) goto invalid_cp1; - crc = *(unsigned int *)((unsigned char *)cp_block + crc_offset); + crc = le32_to_cpu(*((__u32 *)((unsigned char *)cp_block + crc_offset))); if (!f2fs_crc_valid(crc, cp_block, crc_offset)) goto invalid_cp1; @@ -384,7 +384,7 @@ static struct page *validate_checkpoint(struct f2fs_sb_info *sbi, if (crc_offset >= blk_size) goto invalid_cp2; - crc = *(unsigned int *)((unsigned char *)cp_block + crc_offset); + crc = le32_to_cpu(*((__u32 *)((unsigned char *)cp_block + crc_offset))); if (!f2fs_crc_valid(crc, cp_block, crc_offset)) goto invalid_cp2; @@ -450,13 +450,30 @@ fail_no_cp: return -EINVAL; } -void set_dirty_dir_page(struct inode *inode, struct page *page) +static int __add_dirty_inode(struct inode *inode, struct dir_inode_entry *new) { struct f2fs_sb_info *sbi = F2FS_SB(inode->i_sb); struct list_head *head = &sbi->dir_inode_list; - struct dir_inode_entry *new; struct list_head *this; + list_for_each(this, head) { + struct dir_inode_entry *entry; + entry = list_entry(this, struct dir_inode_entry, list); + if (entry->inode == inode) + return -EEXIST; + } + list_add_tail(&new->list, head); +#ifdef CONFIG_F2FS_STAT_FS + sbi->n_dirty_dirs++; +#endif + return 0; +} + +void set_dirty_dir_page(struct inode *inode, struct page *page) +{ + struct f2fs_sb_info *sbi = F2FS_SB(inode->i_sb); + struct dir_inode_entry *new; + if (!S_ISDIR(inode->i_mode)) return; retry: @@ -469,23 +486,31 @@ retry: INIT_LIST_HEAD(&new->list); spin_lock(&sbi->dir_inode_lock); - list_for_each(this, head) { - struct dir_inode_entry *entry; - entry = list_entry(this, struct dir_inode_entry, list); - if (entry->inode == inode) { - kmem_cache_free(inode_entry_slab, new); - goto out; - } - } - list_add_tail(&new->list, head); - sbi->n_dirty_dirs++; + if (__add_dirty_inode(inode, new)) + kmem_cache_free(inode_entry_slab, new); - BUG_ON(!S_ISDIR(inode->i_mode)); -out: inc_page_count(sbi, F2FS_DIRTY_DENTS); inode_inc_dirty_dents(inode); SetPagePrivate(page); + spin_unlock(&sbi->dir_inode_lock); +} +void add_dirty_dir_inode(struct inode *inode) +{ + struct f2fs_sb_info *sbi = F2FS_SB(inode->i_sb); + struct dir_inode_entry *new; +retry: + new = kmem_cache_alloc(inode_entry_slab, GFP_NOFS); + if (!new) { + cond_resched(); + goto retry; + } + new->inode = inode; + INIT_LIST_HEAD(&new->list); + + spin_lock(&sbi->dir_inode_lock); + if (__add_dirty_inode(inode, new)) + kmem_cache_free(inode_entry_slab, new); spin_unlock(&sbi->dir_inode_lock); } @@ -499,8 +524,10 @@ void remove_dirty_dir_inode(struct inode *inode) return; spin_lock(&sbi->dir_inode_lock); - if (atomic_read(&F2FS_I(inode)->dirty_dents)) - goto out; + if (atomic_read(&F2FS_I(inode)->dirty_dents)) { + spin_unlock(&sbi->dir_inode_lock); + return; + } list_for_each(this, head) { struct dir_inode_entry *entry; @@ -508,12 +535,38 @@ void remove_dirty_dir_inode(struct inode *inode) if (entry->inode == inode) { list_del(&entry->list); kmem_cache_free(inode_entry_slab, entry); +#ifdef CONFIG_F2FS_STAT_FS sbi->n_dirty_dirs--; +#endif + break; + } + } + spin_unlock(&sbi->dir_inode_lock); + + /* Only from the recovery routine */ + if (is_inode_flag_set(F2FS_I(inode), FI_DELAY_IPUT)) { + clear_inode_flag(F2FS_I(inode), FI_DELAY_IPUT); + iput(inode); + } +} + +struct inode *check_dirty_dir_inode(struct f2fs_sb_info *sbi, nid_t ino) +{ + struct list_head *head = &sbi->dir_inode_list; + struct list_head *this; + struct inode *inode = NULL; + + spin_lock(&sbi->dir_inode_lock); + list_for_each(this, head) { + struct dir_inode_entry *entry; + entry = list_entry(this, struct dir_inode_entry, list); + if (entry->inode->i_ino == ino) { + inode = entry->inode; break; } } -out: spin_unlock(&sbi->dir_inode_lock); + return inode; } void sync_dirty_dir_inodes(struct f2fs_sb_info *sbi) @@ -595,7 +648,7 @@ static void do_checkpoint(struct f2fs_sb_info *sbi, bool is_umount) block_t start_blk; struct page *cp_page; unsigned int data_sum_blocks, orphan_blocks; - unsigned int crc32 = 0; + __u32 crc32 = 0; void *kaddr; int i; @@ -664,8 +717,8 @@ static void do_checkpoint(struct f2fs_sb_info *sbi, bool is_umount) get_nat_bitmap(sbi, __bitmap_ptr(sbi, NAT_BITMAP)); crc32 = f2fs_crc32(ckpt, le32_to_cpu(ckpt->checksum_offset)); - *(__le32 *)((unsigned char *)ckpt + - le32_to_cpu(ckpt->checksum_offset)) + *((__le32 *)((unsigned char *)ckpt + + le32_to_cpu(ckpt->checksum_offset))) = cpu_to_le32(crc32); start_blk = __start_cp_addr(sbi); diff --git a/fs/f2fs/data.c b/fs/f2fs/data.c index 91ff93b0b0f4..035f9a345cdf 100644 --- a/fs/f2fs/data.c +++ b/fs/f2fs/data.c @@ -68,7 +68,9 @@ static int check_extent_cache(struct inode *inode, pgoff_t pgofs, struct buffer_head *bh_result) { struct f2fs_inode_info *fi = F2FS_I(inode); +#ifdef CONFIG_F2FS_STAT_FS struct f2fs_sb_info *sbi = F2FS_SB(inode->i_sb); +#endif pgoff_t start_fofs, end_fofs; block_t start_blkaddr; @@ -78,7 +80,9 @@ static int check_extent_cache(struct inode *inode, pgoff_t pgofs, return 0; } +#ifdef CONFIG_F2FS_STAT_FS sbi->total_hit_ext++; +#endif start_fofs = fi->ext.fofs; end_fofs = fi->ext.fofs + fi->ext.len - 1; start_blkaddr = fi->ext.blk_addr; @@ -96,7 +100,9 @@ static int check_extent_cache(struct inode *inode, pgoff_t pgofs, else bh_result->b_size = UINT_MAX; +#ifdef CONFIG_F2FS_STAT_FS sbi->read_hit_ext++; +#endif read_unlock(&fi->ext.ext_lock); return 1; } @@ -199,7 +205,7 @@ struct page *find_data_page(struct inode *inode, pgoff_t index, bool sync) if (dn.data_blkaddr == NEW_ADDR) return ERR_PTR(-EINVAL); - page = grab_cache_page(mapping, index); + page = grab_cache_page_write_begin(mapping, index, AOP_FLAG_NOFS); if (!page) return ERR_PTR(-ENOMEM); @@ -233,18 +239,23 @@ struct page *get_lock_data_page(struct inode *inode, pgoff_t index) struct page *page; int err; +repeat: + page = grab_cache_page_write_begin(mapping, index, AOP_FLAG_NOFS); + if (!page) + return ERR_PTR(-ENOMEM); + set_new_dnode(&dn, inode, NULL, NULL, 0); err = get_dnode_of_data(&dn, index, LOOKUP_NODE); - if (err) + if (err) { + f2fs_put_page(page, 1); return ERR_PTR(err); + } f2fs_put_dnode(&dn); - if (dn.data_blkaddr == NULL_ADDR) + if (dn.data_blkaddr == NULL_ADDR) { + f2fs_put_page(page, 1); return ERR_PTR(-ENOENT); -repeat: - page = grab_cache_page(mapping, index); - if (!page) - return ERR_PTR(-ENOMEM); + } if (PageUptodate(page)) return page; @@ -274,9 +285,10 @@ repeat: * * Also, caller should grab and release a mutex by calling mutex_lock_op() and * mutex_unlock_op(). + * Note that, npage is set only by make_empty_dir. */ -struct page *get_new_data_page(struct inode *inode, pgoff_t index, - bool new_i_size) +struct page *get_new_data_page(struct inode *inode, + struct page *npage, pgoff_t index, bool new_i_size) { struct f2fs_sb_info *sbi = F2FS_SB(inode->i_sb); struct address_space *mapping = inode->i_mapping; @@ -284,18 +296,20 @@ struct page *get_new_data_page(struct inode *inode, pgoff_t index, struct dnode_of_data dn; int err; - set_new_dnode(&dn, inode, NULL, NULL, 0); + set_new_dnode(&dn, inode, npage, npage, 0); err = get_dnode_of_data(&dn, index, ALLOC_NODE); if (err) return ERR_PTR(err); if (dn.data_blkaddr == NULL_ADDR) { if (reserve_new_block(&dn)) { - f2fs_put_dnode(&dn); + if (!npage) + f2fs_put_dnode(&dn); return ERR_PTR(-ENOSPC); } } - f2fs_put_dnode(&dn); + if (!npage) + f2fs_put_dnode(&dn); repeat: page = grab_cache_page(mapping, index); if (!page) @@ -325,6 +339,8 @@ repeat: if (new_i_size && i_size_read(inode) < ((index + 1) << PAGE_CACHE_SHIFT)) { i_size_write(inode, ((index + 1) << PAGE_CACHE_SHIFT)); + /* Only the directory inode sets new_i_size */ + set_inode_flag(F2FS_I(inode), FI_UPDATE_DIR); mark_inode_dirty_sync(inode); } return page; @@ -481,8 +497,9 @@ int do_write_data_page(struct page *page) * If current allocation needs SSR, * it had better in-place writes for updated data. */ - if (old_blk_addr != NEW_ADDR && !is_cold_data(page) && - need_inplace_update(inode)) { + if (unlikely(old_blk_addr != NEW_ADDR && + !is_cold_data(page) && + need_inplace_update(inode))) { rewrite_data_page(F2FS_SB(inode->i_sb), page, old_blk_addr); } else { @@ -684,6 +701,27 @@ err: return err; } +static int f2fs_write_end(struct file *file, + struct address_space *mapping, + loff_t pos, unsigned len, unsigned copied, + struct page *page, void *fsdata) +{ + struct inode *inode = page->mapping->host; + + SetPageUptodate(page); + set_page_dirty(page); + + if (pos + copied > i_size_read(inode)) { + i_size_write(inode, pos + copied); + mark_inode_dirty(inode); + update_inode_page(inode); + } + + unlock_page(page); + page_cache_release(page); + return copied; +} + static ssize_t f2fs_direct_IO(int rw, struct kiocb *iocb, const struct iovec *iov, loff_t offset, unsigned long nr_segs) { @@ -698,7 +736,8 @@ static ssize_t f2fs_direct_IO(int rw, struct kiocb *iocb, get_data_block_ro); } -static void f2fs_invalidate_data_page(struct page *page, unsigned long offset) +static void f2fs_invalidate_data_page(struct page *page, unsigned int offset, + unsigned int length) { struct inode *inode = page->mapping->host; struct f2fs_sb_info *sbi = F2FS_SB(inode->i_sb); @@ -740,7 +779,7 @@ const struct address_space_operations f2fs_dblock_aops = { .writepage = f2fs_write_data_page, .writepages = f2fs_write_data_pages, .write_begin = f2fs_write_begin, - .write_end = nobh_write_end, + .write_end = f2fs_write_end, .set_page_dirty = f2fs_set_data_page_dirty, .invalidatepage = f2fs_invalidate_data_page, .releasepage = f2fs_release_data_page, diff --git a/fs/f2fs/debug.c b/fs/f2fs/debug.c index 8d9943786c31..0d6c6aafb235 100644 --- a/fs/f2fs/debug.c +++ b/fs/f2fs/debug.c @@ -175,12 +175,12 @@ get_cache: static int stat_show(struct seq_file *s, void *v) { - struct f2fs_stat_info *si, *next; + struct f2fs_stat_info *si; int i = 0; int j; mutex_lock(&f2fs_stat_mutex); - list_for_each_entry_safe(si, next, &f2fs_stat_list, stat_list) { + list_for_each_entry(si, &f2fs_stat_list, stat_list) { char devname[BDEVNAME_SIZE]; update_general_status(si->sbi); diff --git a/fs/f2fs/dir.c b/fs/f2fs/dir.c index 600bb5efe603..9d1cd423450d 100644 --- a/fs/f2fs/dir.c +++ b/fs/f2fs/dir.c @@ -13,6 +13,7 @@ #include "f2fs.h" #include "node.h" #include "acl.h" +#include "xattr.h" static unsigned long dir_blocks(struct inode *inode) { @@ -215,9 +216,9 @@ struct f2fs_dir_entry *f2fs_find_entry(struct inode *dir, struct f2fs_dir_entry *f2fs_parent_dir(struct inode *dir, struct page **p) { - struct page *page = NULL; - struct f2fs_dir_entry *de = NULL; - struct f2fs_dentry_block *dentry_blk = NULL; + struct page *page; + struct f2fs_dir_entry *de; + struct f2fs_dentry_block *dentry_blk; page = get_lock_data_page(dir, 0); if (IS_ERR(page)) @@ -264,15 +265,10 @@ void f2fs_set_link(struct inode *dir, struct f2fs_dir_entry *de, f2fs_put_page(page, 1); } -void init_dent_inode(const struct qstr *name, struct page *ipage) +static void init_dent_inode(const struct qstr *name, struct page *ipage) { struct f2fs_node *rn; - if (IS_ERR(ipage)) - return; - - wait_on_page_writeback(ipage); - /* copy name info. to this inode page */ rn = (struct f2fs_node *)page_address(ipage); rn->i.i_namelen = cpu_to_le32(name->len); @@ -280,14 +276,15 @@ void init_dent_inode(const struct qstr *name, struct page *ipage) set_page_dirty(ipage); } -static int make_empty_dir(struct inode *inode, struct inode *parent) +static int make_empty_dir(struct inode *inode, + struct inode *parent, struct page *page) { struct page *dentry_page; struct f2fs_dentry_block *dentry_blk; struct f2fs_dir_entry *de; void *kaddr; - dentry_page = get_new_data_page(inode, 0, true); + dentry_page = get_new_data_page(inode, page, 0, true); if (IS_ERR(dentry_page)) return PTR_ERR(dentry_page); @@ -317,63 +314,76 @@ static int make_empty_dir(struct inode *inode, struct inode *parent) return 0; } -static int init_inode_metadata(struct inode *inode, +static struct page *init_inode_metadata(struct inode *inode, struct inode *dir, const struct qstr *name) { + struct page *page; + int err; + if (is_inode_flag_set(F2FS_I(inode), FI_NEW_INODE)) { - int err; - err = new_inode_page(inode, name); - if (err) - return err; + page = new_inode_page(inode, name); + if (IS_ERR(page)) + return page; if (S_ISDIR(inode->i_mode)) { - err = make_empty_dir(inode, dir); - if (err) { - remove_inode_page(inode); - return err; - } + err = make_empty_dir(inode, dir, page); + if (err) + goto error; } err = f2fs_init_acl(inode, dir); - if (err) { - remove_inode_page(inode); - return err; - } + if (err) + goto error; + + err = f2fs_init_security(inode, dir, name, page); + if (err) + goto error; + + wait_on_page_writeback(page); } else { - struct page *ipage; - ipage = get_node_page(F2FS_SB(dir->i_sb), inode->i_ino); - if (IS_ERR(ipage)) - return PTR_ERR(ipage); - set_cold_node(inode, ipage); - init_dent_inode(name, ipage); - f2fs_put_page(ipage, 1); + page = get_node_page(F2FS_SB(dir->i_sb), inode->i_ino); + if (IS_ERR(page)) + return page; + + wait_on_page_writeback(page); + set_cold_node(inode, page); } + + init_dent_inode(name, page); + + /* + * This file should be checkpointed during fsync. + * We lost i_pino from now on. + */ if (is_inode_flag_set(F2FS_I(inode), FI_INC_LINK)) { + file_lost_pino(inode); inc_nlink(inode); - update_inode_page(inode); } - return 0; + return page; + +error: + f2fs_put_page(page, 1); + remove_inode_page(inode); + return ERR_PTR(err); } static void update_parent_metadata(struct inode *dir, struct inode *inode, unsigned int current_depth) { - bool need_dir_update = false; - if (is_inode_flag_set(F2FS_I(inode), FI_NEW_INODE)) { if (S_ISDIR(inode->i_mode)) { inc_nlink(dir); - need_dir_update = true; + set_inode_flag(F2FS_I(dir), FI_UPDATE_DIR); } clear_inode_flag(F2FS_I(inode), FI_NEW_INODE); } dir->i_mtime = dir->i_ctime = CURRENT_TIME; if (F2FS_I(dir)->i_current_depth != current_depth) { F2FS_I(dir)->i_current_depth = current_depth; - need_dir_update = true; + set_inode_flag(F2FS_I(dir), FI_UPDATE_DIR); } - if (need_dir_update) + if (is_inode_flag_set(F2FS_I(dir), FI_UPDATE_DIR)) update_inode_page(dir); else mark_inode_dirty(dir); @@ -423,6 +433,7 @@ int __f2fs_add_link(struct inode *dir, const struct qstr *name, struct inode *in struct page *dentry_page = NULL; struct f2fs_dentry_block *dentry_blk = NULL; int slots = GET_DENTRY_SLOTS(namelen); + struct page *page; int err = 0; int i; @@ -448,7 +459,7 @@ start: bidx = dir_block_index(level, (le32_to_cpu(dentry_hash) % nbucket)); for (block = bidx; block <= (bidx + nblock - 1); block++) { - dentry_page = get_new_data_page(dir, block, true); + dentry_page = get_new_data_page(dir, NULL, block, true); if (IS_ERR(dentry_page)) return PTR_ERR(dentry_page); @@ -465,12 +476,13 @@ start: ++level; goto start; add_dentry: - err = init_inode_metadata(inode, dir, name); - if (err) - goto fail; - wait_on_page_writeback(dentry_page); + page = init_inode_metadata(inode, dir, name); + if (IS_ERR(page)) { + err = PTR_ERR(page); + goto fail; + } de = &dentry_blk->dentry[bit_pos]; de->hash_code = dentry_hash; de->name_len = cpu_to_le16(namelen); @@ -481,11 +493,14 @@ add_dentry: test_and_set_bit_le(bit_pos + i, &dentry_blk->dentry_bitmap); set_page_dirty(dentry_page); - update_parent_metadata(dir, inode, current_depth); - - /* update parent inode number before releasing dentry page */ + /* we don't need to mark_inode_dirty now */ F2FS_I(inode)->i_pino = dir->i_ino; + update_inode(inode, page); + f2fs_put_page(page, 1); + + update_parent_metadata(dir, inode, current_depth); fail: + clear_inode_flag(F2FS_I(dir), FI_UPDATE_DIR); kunmap(dentry_page); f2fs_put_page(dentry_page, 1); return err; diff --git a/fs/f2fs/f2fs.h b/fs/f2fs/f2fs.h index 20aab02f2a42..467d42d65c48 100644 --- a/fs/f2fs/f2fs.h +++ b/fs/f2fs/f2fs.h @@ -37,21 +37,35 @@ typecheck(unsigned long long, b) && \ ((long long)((a) - (b)) > 0)) -typedef u64 block_t; +typedef u32 block_t; /* + * should not change u32, since it is the on-disk block + * address format, __le32. + */ typedef u32 nid_t; struct f2fs_mount_info { unsigned int opt; }; -static inline __u32 f2fs_crc32(void *buff, size_t len) +#define CRCPOLY_LE 0xedb88320 + +static inline __u32 f2fs_crc32(void *buf, size_t len) { - return crc32_le(F2FS_SUPER_MAGIC, buff, len); + unsigned char *p = (unsigned char *)buf; + __u32 crc = F2FS_SUPER_MAGIC; + int i; + + while (len--) { + crc ^= *p++; + for (i = 0; i < 8; i++) + crc = (crc >> 1) ^ ((crc & 1) ? CRCPOLY_LE : 0); + } + return crc; } -static inline bool f2fs_crc_valid(__u32 blk_crc, void *buff, size_t buff_size) +static inline bool f2fs_crc_valid(__u32 blk_crc, void *buf, size_t buf_size) { - return f2fs_crc32(buff, buff_size) == blk_crc; + return f2fs_crc32(buf, buf_size) == blk_crc; } /* @@ -148,7 +162,7 @@ struct extent_info { * i_advise uses FADVISE_XXX_BIT. We can add additional hints later. */ #define FADVISE_COLD_BIT 0x01 -#define FADVISE_CP_BIT 0x02 +#define FADVISE_LOST_PINO_BIT 0x02 struct f2fs_inode_info { struct inode vfs_inode; /* serve a vfs inode */ @@ -369,7 +383,6 @@ struct f2fs_sb_info { /* for directory inode management */ struct list_head dir_inode_list; /* dir inode list */ spinlock_t dir_inode_lock; /* for dir inode list lock */ - unsigned int n_dirty_dirs; /* # of dir inodes */ /* basic file system units */ unsigned int log_sectors_per_block; /* log2 sectors per block */ @@ -406,12 +419,15 @@ struct f2fs_sb_info { * for stat information. * one is for the LFS mode, and the other is for the SSR mode. */ +#ifdef CONFIG_F2FS_STAT_FS struct f2fs_stat_info *stat_info; /* FS status information */ unsigned int segment_count[2]; /* # of allocated segments */ unsigned int block_count[2]; /* # of allocated blocks */ - unsigned int last_victim[2]; /* last victim segment # */ int total_hit_ext, read_hit_ext; /* extent cache hit ratio */ int bg_gc; /* background gc calls */ + unsigned int n_dirty_dirs; /* # of dir inodes */ +#endif + unsigned int last_victim[2]; /* last victim segment # */ spinlock_t stat_lock; /* lock for stat operations */ }; @@ -495,9 +511,17 @@ static inline void clear_ckpt_flags(struct f2fs_checkpoint *cp, unsigned int f) static inline void mutex_lock_all(struct f2fs_sb_info *sbi) { - int i = 0; - for (; i < NR_GLOBAL_LOCKS; i++) - mutex_lock(&sbi->fs_lock[i]); + int i; + + for (i = 0; i < NR_GLOBAL_LOCKS; i++) { + /* + * This is the only time we take multiple fs_lock[] + * instances; the order is immaterial since we + * always hold cp_mutex, which serializes multiple + * such operations. + */ + mutex_lock_nest_lock(&sbi->fs_lock[i], &sbi->cp_mutex); + } } static inline void mutex_unlock_all(struct f2fs_sb_info *sbi) @@ -843,9 +867,12 @@ static inline int f2fs_clear_bit(unsigned int nr, char *addr) /* used for f2fs_inode_info->flags */ enum { FI_NEW_INODE, /* indicate newly allocated inode */ + FI_DIRTY_INODE, /* indicate inode is dirty or not */ FI_INC_LINK, /* need to increment i_nlink */ FI_ACL_MODE, /* indicate acl mode */ FI_NO_ALLOC, /* should not allocate any blocks */ + FI_UPDATE_DIR, /* should update inode block for consistency */ + FI_DELAY_IPUT, /* used for the recovery */ }; static inline void set_inode_flag(struct f2fs_inode_info *fi, int flag) @@ -878,14 +905,21 @@ static inline int cond_clear_inode_flag(struct f2fs_inode_info *fi, int flag) return 0; } +static inline int f2fs_readonly(struct super_block *sb) +{ + return sb->s_flags & MS_RDONLY; +} + /* * file.c */ int f2fs_sync_file(struct file *, loff_t, loff_t, int); void truncate_data_blocks(struct dnode_of_data *); void f2fs_truncate(struct inode *); +int f2fs_getattr(struct vfsmount *, struct dentry *, struct kstat *); int f2fs_setattr(struct dentry *, struct iattr *); int truncate_hole(struct inode *, pgoff_t, pgoff_t); +int truncate_data_blocks_range(struct dnode_of_data *, int); long f2fs_ioctl(struct file *, unsigned int, unsigned long); long f2fs_compat_ioctl(struct file *, unsigned int, unsigned long); @@ -913,7 +947,6 @@ struct f2fs_dir_entry *f2fs_parent_dir(struct inode *, struct page **); ino_t f2fs_inode_by_name(struct inode *, struct qstr *); void f2fs_set_link(struct inode *, struct f2fs_dir_entry *, struct page *, struct inode *); -void init_dent_inode(const struct qstr *, struct page *); int __f2fs_add_link(struct inode *, const struct qstr *, struct inode *); void f2fs_delete_entry(struct f2fs_dir_entry *, struct page *, struct inode *); int f2fs_make_empty(struct inode *, struct inode *); @@ -948,8 +981,8 @@ void get_node_info(struct f2fs_sb_info *, nid_t, struct node_info *); int get_dnode_of_data(struct dnode_of_data *, pgoff_t, int); int truncate_inode_blocks(struct inode *, pgoff_t); int remove_inode_page(struct inode *); -int new_inode_page(struct inode *, const struct qstr *); -struct page *new_node_page(struct dnode_of_data *, unsigned int); +struct page *new_inode_page(struct inode *, const struct qstr *); +struct page *new_node_page(struct dnode_of_data *, unsigned int, struct page *); void ra_node_page(struct f2fs_sb_info *, nid_t); struct page *get_node_page(struct f2fs_sb_info *, pgoff_t); struct page *get_node_page_ra(struct page *, int); @@ -974,7 +1007,6 @@ void destroy_node_manager_caches(void); */ void f2fs_balance_fs(struct f2fs_sb_info *); void invalidate_blocks(struct f2fs_sb_info *, block_t); -void locate_dirty_segment(struct f2fs_sb_info *, unsigned int); void clear_prefree_segments(struct f2fs_sb_info *); int npages_for_summary_flush(struct f2fs_sb_info *); void allocate_new_segments(struct f2fs_sb_info *); @@ -1011,7 +1043,9 @@ void remove_orphan_inode(struct f2fs_sb_info *, nid_t); int recover_orphan_inodes(struct f2fs_sb_info *); int get_valid_checkpoint(struct f2fs_sb_info *); void set_dirty_dir_page(struct inode *, struct page *); +void add_dirty_dir_inode(struct inode *); void remove_dirty_dir_inode(struct inode *); +struct inode *check_dirty_dir_inode(struct f2fs_sb_info *, nid_t); void sync_dirty_dir_inodes(struct f2fs_sb_info *); void write_checkpoint(struct f2fs_sb_info *, bool); void init_orphan_info(struct f2fs_sb_info *); @@ -1025,7 +1059,7 @@ int reserve_new_block(struct dnode_of_data *); void update_extent_cache(block_t, struct dnode_of_data *); struct page *find_data_page(struct inode *, pgoff_t, bool); struct page *get_lock_data_page(struct inode *, pgoff_t); -struct page *get_new_data_page(struct inode *, pgoff_t, bool); +struct page *get_new_data_page(struct inode *, struct page *, pgoff_t, bool); int f2fs_readpage(struct f2fs_sb_info *, struct page *, block_t, int); int do_write_data_page(struct page *); diff --git a/fs/f2fs/file.c b/fs/f2fs/file.c index 1cae864f8dfc..d2d2b7dbdcc1 100644 --- a/fs/f2fs/file.c +++ b/fs/f2fs/file.c @@ -63,9 +63,10 @@ static int f2fs_vm_page_mkwrite(struct vm_area_struct *vma, f2fs_put_dnode(&dn); mutex_unlock_op(sbi, ilock); + file_update_time(vma->vm_file); lock_page(page); if (page->mapping != inode->i_mapping || - page_offset(page) >= i_size_read(inode) || + page_offset(page) > i_size_read(inode) || !PageUptodate(page)) { unlock_page(page); err = -EFAULT; @@ -76,10 +77,7 @@ static int f2fs_vm_page_mkwrite(struct vm_area_struct *vma, * check to see if the page is mapped already (no holes) */ if (PageMappedToDisk(page)) - goto out; - - /* fill the page */ - wait_on_page_writeback(page); + goto mapped; /* page is wholly or partially inside EOF */ if (((page->index + 1) << PAGE_CACHE_SHIFT) > i_size_read(inode)) { @@ -90,7 +88,9 @@ static int f2fs_vm_page_mkwrite(struct vm_area_struct *vma, set_page_dirty(page); SetPageUptodate(page); - file_update_time(vma->vm_file); +mapped: + /* fill the page */ + wait_on_page_writeback(page); out: sb_end_pagefault(inode->i_sb); return block_page_mkwrite_return(err); @@ -102,6 +102,24 @@ static const struct vm_operations_struct f2fs_file_vm_ops = { .remap_pages = generic_file_remap_pages, }; +static int get_parent_ino(struct inode *inode, nid_t *pino) +{ + struct dentry *dentry; + + inode = igrab(inode); + dentry = d_find_any_alias(inode); + iput(inode); + if (!dentry) + return 0; + + inode = igrab(dentry->d_parent->d_inode); + dput(dentry); + + *pino = inode->i_ino; + iput(inode); + return 1; +} + int f2fs_sync_file(struct file *file, loff_t start, loff_t end, int datasync) { struct inode *inode = file->f_mapping->host; @@ -114,7 +132,7 @@ int f2fs_sync_file(struct file *file, loff_t start, loff_t end, int datasync) .for_reclaim = 0, }; - if (inode->i_sb->s_flags & MS_RDONLY) + if (f2fs_readonly(inode->i_sb)) return 0; trace_f2fs_sync_file_enter(inode); @@ -134,7 +152,7 @@ int f2fs_sync_file(struct file *file, loff_t start, loff_t end, int datasync) if (!S_ISREG(inode->i_mode) || inode->i_nlink != 1) need_cp = true; - else if (is_cp_file(inode)) + else if (file_wrong_pino(inode)) need_cp = true; else if (!space_for_roll_forward(sbi)) need_cp = true; @@ -142,11 +160,23 @@ int f2fs_sync_file(struct file *file, loff_t start, loff_t end, int datasync) need_cp = true; if (need_cp) { + nid_t pino; + /* all the dirty node pages should be flushed for POR */ ret = f2fs_sync_fs(inode->i_sb, 1); + if (file_wrong_pino(inode) && inode->i_nlink == 1 && + get_parent_ino(inode, &pino)) { + F2FS_I(inode)->i_pino = pino; + file_got_pino(inode); + mark_inode_dirty_sync(inode); + ret = f2fs_write_inode(inode, NULL); + if (ret) + goto out; + } } else { /* if there is no written node page, write its inode page */ while (!sync_node_pages(sbi, inode->i_ino, &wbc)) { + mark_inode_dirty_sync(inode); ret = f2fs_write_inode(inode, NULL); if (ret) goto out; @@ -168,7 +198,7 @@ static int f2fs_file_mmap(struct file *file, struct vm_area_struct *vma) return 0; } -static int truncate_data_blocks_range(struct dnode_of_data *dn, int count) +int truncate_data_blocks_range(struct dnode_of_data *dn, int count) { int nr_free = 0, ofs = dn->ofs_in_node; struct f2fs_sb_info *sbi = F2FS_SB(dn->inode->i_sb); @@ -185,10 +215,10 @@ static int truncate_data_blocks_range(struct dnode_of_data *dn, int count) update_extent_cache(NULL_ADDR, dn); invalidate_blocks(sbi, blkaddr); - dec_valid_block_count(sbi, dn->inode, 1); nr_free++; } if (nr_free) { + dec_valid_block_count(sbi, dn->inode, nr_free); set_page_dirty(dn->node_page); sync_inode_page(dn); } @@ -291,7 +321,7 @@ void f2fs_truncate(struct inode *inode) } } -static int f2fs_getattr(struct vfsmount *mnt, +int f2fs_getattr(struct vfsmount *mnt, struct dentry *dentry, struct kstat *stat) { struct inode *inode = dentry->d_inode; @@ -387,7 +417,7 @@ static void fill_zero(struct inode *inode, pgoff_t index, f2fs_balance_fs(sbi); ilock = mutex_lock_op(sbi); - page = get_new_data_page(inode, index, false); + page = get_new_data_page(inode, NULL, index, false); mutex_unlock_op(sbi, ilock); if (!IS_ERR(page)) { @@ -575,10 +605,10 @@ long f2fs_ioctl(struct file *filp, unsigned int cmd, unsigned long arg) int ret; switch (cmd) { - case FS_IOC_GETFLAGS: + case F2FS_IOC_GETFLAGS: flags = fi->i_flags & FS_FL_USER_VISIBLE; return put_user(flags, (int __user *) arg); - case FS_IOC_SETFLAGS: + case F2FS_IOC_SETFLAGS: { unsigned int oldflags; diff --git a/fs/f2fs/gc.c b/fs/f2fs/gc.c index 14961593e93c..35f9b1a196aa 100644 --- a/fs/f2fs/gc.c +++ b/fs/f2fs/gc.c @@ -76,7 +76,9 @@ static int gc_thread_func(void *data) else wait_ms = increase_sleep_time(wait_ms); +#ifdef CONFIG_F2FS_STAT_FS sbi->bg_gc++; +#endif /* if return value is not zero, no victim was selected */ if (f2fs_gc(sbi)) @@ -89,23 +91,28 @@ int start_gc_thread(struct f2fs_sb_info *sbi) { struct f2fs_gc_kthread *gc_th; dev_t dev = sbi->sb->s_bdev->bd_dev; + int err = 0; if (!test_opt(sbi, BG_GC)) - return 0; + goto out; gc_th = kmalloc(sizeof(struct f2fs_gc_kthread), GFP_KERNEL); - if (!gc_th) - return -ENOMEM; + if (!gc_th) { + err = -ENOMEM; + goto out; + } sbi->gc_thread = gc_th; init_waitqueue_head(&sbi->gc_thread->gc_wait_queue_head); sbi->gc_thread->f2fs_gc_task = kthread_run(gc_thread_func, sbi, "f2fs_gc-%u:%u", MAJOR(dev), MINOR(dev)); if (IS_ERR(gc_th->f2fs_gc_task)) { + err = PTR_ERR(gc_th->f2fs_gc_task); kfree(gc_th); sbi->gc_thread = NULL; - return -ENOMEM; } - return 0; + +out: + return err; } void stop_gc_thread(struct f2fs_sb_info *sbi) @@ -234,14 +241,14 @@ static int get_victim_by_default(struct f2fs_sb_info *sbi, { struct dirty_seglist_info *dirty_i = DIRTY_I(sbi); struct victim_sel_policy p; - unsigned int secno; + unsigned int secno, max_cost; int nsearched = 0; p.alloc_mode = alloc_mode; select_policy(sbi, gc_type, type, &p); p.min_segno = NULL_SEGNO; - p.min_cost = get_max_cost(sbi, &p); + p.min_cost = max_cost = get_max_cost(sbi, &p); mutex_lock(&dirty_i->seglist_lock); @@ -280,7 +287,7 @@ static int get_victim_by_default(struct f2fs_sb_info *sbi, p.min_cost = cost; } - if (cost == get_max_cost(sbi, &p)) + if (cost == max_cost) continue; if (nsearched++ >= MAX_VICTIM_SEARCH) { @@ -288,8 +295,8 @@ static int get_victim_by_default(struct f2fs_sb_info *sbi, break; } } -got_it: if (p.min_segno != NULL_SEGNO) { +got_it: if (p.alloc_mode == LFS) { secno = GET_SECNO(sbi, p.min_segno); if (gc_type == FG_GC) @@ -314,28 +321,21 @@ static const struct victim_selection default_v_ops = { static struct inode *find_gc_inode(nid_t ino, struct list_head *ilist) { - struct list_head *this; struct inode_entry *ie; - list_for_each(this, ilist) { - ie = list_entry(this, struct inode_entry, list); + list_for_each_entry(ie, ilist, list) if (ie->inode->i_ino == ino) return ie->inode; - } return NULL; } static void add_gc_inode(struct inode *inode, struct list_head *ilist) { - struct list_head *this; - struct inode_entry *new_ie, *ie; + struct inode_entry *new_ie; - list_for_each(this, ilist) { - ie = list_entry(this, struct inode_entry, list); - if (ie->inode == inode) { - iput(inode); - return; - } + if (inode == find_gc_inode(inode->i_ino, ilist)) { + iput(inode); + return; } repeat: new_ie = kmem_cache_alloc(winode_slab, GFP_NOFS); diff --git a/fs/f2fs/inode.c b/fs/f2fs/inode.c index 91ac7f9d88ee..2b2d45d19e3e 100644 --- a/fs/f2fs/inode.c +++ b/fs/f2fs/inode.c @@ -109,12 +109,6 @@ struct inode *f2fs_iget(struct super_block *sb, unsigned long ino) ret = do_read_inode(inode); if (ret) goto bad_inode; - - if (!sbi->por_doing && inode->i_nlink == 0) { - ret = -ENOENT; - goto bad_inode; - } - make_now: if (ino == F2FS_NODE_INO(sbi)) { inode->i_mapping->a_ops = &f2fs_node_aops; @@ -130,8 +124,7 @@ make_now: inode->i_op = &f2fs_dir_inode_operations; inode->i_fop = &f2fs_dir_operations; inode->i_mapping->a_ops = &f2fs_dblock_aops; - mapping_set_gfp_mask(inode->i_mapping, GFP_HIGHUSER_MOVABLE | - __GFP_ZERO); + mapping_set_gfp_mask(inode->i_mapping, GFP_F2FS_ZERO); } else if (S_ISLNK(inode->i_mode)) { inode->i_op = &f2fs_symlink_inode_operations; inode->i_mapping->a_ops = &f2fs_dblock_aops; @@ -199,6 +192,7 @@ void update_inode(struct inode *inode, struct page *node_page) set_cold_node(inode, node_page); set_page_dirty(node_page); + clear_inode_flag(F2FS_I(inode), FI_DIRTY_INODE); } int update_inode_page(struct inode *inode) @@ -224,6 +218,9 @@ int f2fs_write_inode(struct inode *inode, struct writeback_control *wbc) inode->i_ino == F2FS_META_INO(sbi)) return 0; + if (!is_inode_flag_set(F2FS_I(inode), FI_DIRTY_INODE)) + return 0; + if (wbc) f2fs_balance_fs(sbi); diff --git a/fs/f2fs/namei.c b/fs/f2fs/namei.c index 47abc9722b17..64c07169df05 100644 --- a/fs/f2fs/namei.c +++ b/fs/f2fs/namei.c @@ -112,7 +112,7 @@ static inline void set_cold_files(struct f2fs_sb_info *sbi, struct inode *inode, int count = le32_to_cpu(sbi->raw_super->extension_count); for (i = 0; i < count; i++) { if (is_multimedia_file(name, extlist[i])) { - set_cold_file(inode); + file_set_cold(inode); break; } } @@ -149,8 +149,7 @@ static int f2fs_create(struct inode *dir, struct dentry *dentry, umode_t mode, alloc_nid_done(sbi, ino); - if (!sbi->por_doing) - d_instantiate(dentry, inode); + d_instantiate(dentry, inode); unlock_new_inode(inode); return 0; out: @@ -173,7 +172,7 @@ static int f2fs_link(struct dentry *old_dentry, struct inode *dir, f2fs_balance_fs(sbi); inode->i_ctime = CURRENT_TIME; - atomic_inc(&inode->i_count); + ihold(inode); set_inode_flag(F2FS_I(inode), FI_INC_LINK); ilock = mutex_lock_op(sbi); @@ -182,17 +181,10 @@ static int f2fs_link(struct dentry *old_dentry, struct inode *dir, if (err) goto out; - /* - * This file should be checkpointed during fsync. - * We lost i_pino from now on. - */ - set_cp_file(inode); - d_instantiate(dentry, inode); return 0; out: clear_inode_flag(F2FS_I(inode), FI_INC_LINK); - make_bad_inode(inode); iput(inode); return err; } @@ -498,6 +490,7 @@ const struct inode_operations f2fs_dir_inode_operations = { .rmdir = f2fs_rmdir, .mknod = f2fs_mknod, .rename = f2fs_rename, + .getattr = f2fs_getattr, .setattr = f2fs_setattr, .get_acl = f2fs_get_acl, #ifdef CONFIG_F2FS_FS_XATTR @@ -512,6 +505,7 @@ const struct inode_operations f2fs_symlink_inode_operations = { .readlink = generic_readlink, .follow_link = page_follow_link_light, .put_link = page_put_link, + .getattr = f2fs_getattr, .setattr = f2fs_setattr, #ifdef CONFIG_F2FS_FS_XATTR .setxattr = generic_setxattr, @@ -522,6 +516,7 @@ const struct inode_operations f2fs_symlink_inode_operations = { }; const struct inode_operations f2fs_special_inode_operations = { + .getattr = f2fs_getattr, .setattr = f2fs_setattr, .get_acl = f2fs_get_acl, #ifdef CONFIG_F2FS_FS_XATTR diff --git a/fs/f2fs/node.c b/fs/f2fs/node.c index 3df43b4efd89..b418aee09573 100644 --- a/fs/f2fs/node.c +++ b/fs/f2fs/node.c @@ -408,10 +408,13 @@ int get_dnode_of_data(struct dnode_of_data *dn, pgoff_t index, int mode) level = get_node_path(index, offset, noffset); nids[0] = dn->inode->i_ino; - npage[0] = get_node_page(sbi, nids[0]); - if (IS_ERR(npage[0])) - return PTR_ERR(npage[0]); + npage[0] = dn->inode_page; + if (!npage[0]) { + npage[0] = get_node_page(sbi, nids[0]); + if (IS_ERR(npage[0])) + return PTR_ERR(npage[0]); + } parent = npage[0]; if (level != 0) nids[1] = get_nid(parent, offset[0], true); @@ -430,7 +433,7 @@ int get_dnode_of_data(struct dnode_of_data *dn, pgoff_t index, int mode) } dn->nid = nids[i]; - npage[i] = new_node_page(dn, noffset[i]); + npage[i] = new_node_page(dn, noffset[i], NULL); if (IS_ERR(npage[i])) { alloc_nid_failed(sbi, nids[i]); err = PTR_ERR(npage[i]); @@ -803,22 +806,19 @@ int remove_inode_page(struct inode *inode) return 0; } -int new_inode_page(struct inode *inode, const struct qstr *name) +struct page *new_inode_page(struct inode *inode, const struct qstr *name) { - struct page *page; struct dnode_of_data dn; /* allocate inode page for new inode */ set_new_dnode(&dn, inode, NULL, NULL, inode->i_ino); - page = new_node_page(&dn, 0); - init_dent_inode(name, page); - if (IS_ERR(page)) - return PTR_ERR(page); - f2fs_put_page(page, 1); - return 0; + + /* caller should f2fs_put_page(page, 1); */ + return new_node_page(&dn, 0, NULL); } -struct page *new_node_page(struct dnode_of_data *dn, unsigned int ofs) +struct page *new_node_page(struct dnode_of_data *dn, + unsigned int ofs, struct page *ipage) { struct f2fs_sb_info *sbi = F2FS_SB(dn->inode->i_sb); struct address_space *mapping = sbi->node_inode->i_mapping; @@ -851,7 +851,10 @@ struct page *new_node_page(struct dnode_of_data *dn, unsigned int ofs) set_cold_node(dn->inode, page); dn->node_page = page; - sync_inode_page(dn); + if (ipage) + update_inode(dn->inode, ipage); + else + sync_inode_page(dn); set_page_dirty(page); if (ofs == 0) inc_valid_inode_count(sbi); @@ -1205,7 +1208,8 @@ static int f2fs_set_node_page_dirty(struct page *page) return 0; } -static void f2fs_invalidate_node_page(struct page *page, unsigned long offset) +static void f2fs_invalidate_node_page(struct page *page, unsigned int offset, + unsigned int length) { struct inode *inode = page->mapping->host; struct f2fs_sb_info *sbi = F2FS_SB(inode->i_sb); @@ -1492,9 +1496,10 @@ int recover_inode_page(struct f2fs_sb_info *sbi, struct page *page) new_ni = old_ni; new_ni.ino = ino; + if (!inc_valid_node_count(sbi, NULL, 1)) + WARN_ON(1); set_node_addr(sbi, &new_ni, NEW_ADDR); inc_valid_inode_count(sbi); - f2fs_put_page(ipage, 1); return 0; } diff --git a/fs/f2fs/node.h b/fs/f2fs/node.h index 0a2d72f0024d..c65fb4f4230f 100644 --- a/fs/f2fs/node.h +++ b/fs/f2fs/node.h @@ -275,25 +275,27 @@ static inline nid_t get_nid(struct page *p, int off, bool i) * - Mark cold node blocks in their node footer * - Mark cold data pages in page cache */ -static inline int is_cold_file(struct inode *inode) +static inline int is_file(struct inode *inode, int type) { - return F2FS_I(inode)->i_advise & FADVISE_COLD_BIT; + return F2FS_I(inode)->i_advise & type; } -static inline void set_cold_file(struct inode *inode) +static inline void set_file(struct inode *inode, int type) { - F2FS_I(inode)->i_advise |= FADVISE_COLD_BIT; + F2FS_I(inode)->i_advise |= type; } -static inline int is_cp_file(struct inode *inode) +static inline void clear_file(struct inode *inode, int type) { - return F2FS_I(inode)->i_advise & FADVISE_CP_BIT; + F2FS_I(inode)->i_advise &= ~type; } -static inline void set_cp_file(struct inode *inode) -{ - F2FS_I(inode)->i_advise |= FADVISE_CP_BIT; -} +#define file_is_cold(inode) is_file(inode, FADVISE_COLD_BIT) +#define file_wrong_pino(inode) is_file(inode, FADVISE_LOST_PINO_BIT) +#define file_set_cold(inode) set_file(inode, FADVISE_COLD_BIT) +#define file_lost_pino(inode) set_file(inode, FADVISE_LOST_PINO_BIT) +#define file_clear_cold(inode) clear_file(inode, FADVISE_COLD_BIT) +#define file_got_pino(inode) clear_file(inode, FADVISE_LOST_PINO_BIT) static inline int is_cold_data(struct page *page) { @@ -310,29 +312,16 @@ static inline void clear_cold_data(struct page *page) ClearPageChecked(page); } -static inline int is_cold_node(struct page *page) +static inline int is_node(struct page *page, int type) { void *kaddr = page_address(page); struct f2fs_node *rn = (struct f2fs_node *)kaddr; - unsigned int flag = le32_to_cpu(rn->footer.flag); - return flag & (0x1 << COLD_BIT_SHIFT); + return le32_to_cpu(rn->footer.flag) & (1 << type); } -static inline unsigned char is_fsync_dnode(struct page *page) -{ - void *kaddr = page_address(page); - struct f2fs_node *rn = (struct f2fs_node *)kaddr; - unsigned int flag = le32_to_cpu(rn->footer.flag); - return flag & (0x1 << FSYNC_BIT_SHIFT); -} - -static inline unsigned char is_dent_dnode(struct page *page) -{ - void *kaddr = page_address(page); - struct f2fs_node *rn = (struct f2fs_node *)kaddr; - unsigned int flag = le32_to_cpu(rn->footer.flag); - return flag & (0x1 << DENT_BIT_SHIFT); -} +#define is_cold_node(page) is_node(page, COLD_BIT_SHIFT) +#define is_fsync_dnode(page) is_node(page, FSYNC_BIT_SHIFT) +#define is_dent_dnode(page) is_node(page, DENT_BIT_SHIFT) static inline void set_cold_node(struct inode *inode, struct page *page) { @@ -346,26 +335,15 @@ static inline void set_cold_node(struct inode *inode, struct page *page) rn->footer.flag = cpu_to_le32(flag); } -static inline void set_fsync_mark(struct page *page, int mark) +static inline void set_mark(struct page *page, int mark, int type) { - void *kaddr = page_address(page); - struct f2fs_node *rn = (struct f2fs_node *)kaddr; - unsigned int flag = le32_to_cpu(rn->footer.flag); - if (mark) - flag |= (0x1 << FSYNC_BIT_SHIFT); - else - flag &= ~(0x1 << FSYNC_BIT_SHIFT); - rn->footer.flag = cpu_to_le32(flag); -} - -static inline void set_dentry_mark(struct page *page, int mark) -{ - void *kaddr = page_address(page); - struct f2fs_node *rn = (struct f2fs_node *)kaddr; + struct f2fs_node *rn = (struct f2fs_node *)page_address(page); unsigned int flag = le32_to_cpu(rn->footer.flag); if (mark) - flag |= (0x1 << DENT_BIT_SHIFT); + flag |= (0x1 << type); else - flag &= ~(0x1 << DENT_BIT_SHIFT); + flag &= ~(0x1 << type); rn->footer.flag = cpu_to_le32(flag); } +#define set_dentry_mark(page, mark) set_mark(page, mark, DENT_BIT_SHIFT) +#define set_fsync_mark(page, mark) set_mark(page, mark, FSYNC_BIT_SHIFT) diff --git a/fs/f2fs/recovery.c b/fs/f2fs/recovery.c index 60c8a5097058..d56d951c2253 100644 --- a/fs/f2fs/recovery.c +++ b/fs/f2fs/recovery.c @@ -40,36 +40,54 @@ static struct fsync_inode_entry *get_fsync_inode(struct list_head *head, static int recover_dentry(struct page *ipage, struct inode *inode) { - struct f2fs_node *raw_node = (struct f2fs_node *)kmap(ipage); + void *kaddr = page_address(ipage); + struct f2fs_node *raw_node = (struct f2fs_node *)kaddr; struct f2fs_inode *raw_inode = &(raw_node->i); - struct qstr name; + nid_t pino = le32_to_cpu(raw_inode->i_pino); struct f2fs_dir_entry *de; + struct qstr name; struct page *page; - struct inode *dir; + struct inode *dir, *einode; int err = 0; - if (!is_dent_dnode(ipage)) - goto out; - - dir = f2fs_iget(inode->i_sb, le32_to_cpu(raw_inode->i_pino)); - if (IS_ERR(dir)) { - err = PTR_ERR(dir); - goto out; + dir = check_dirty_dir_inode(F2FS_SB(inode->i_sb), pino); + if (!dir) { + dir = f2fs_iget(inode->i_sb, pino); + if (IS_ERR(dir)) { + err = PTR_ERR(dir); + goto out; + } + set_inode_flag(F2FS_I(dir), FI_DELAY_IPUT); + add_dirty_dir_inode(dir); } name.len = le32_to_cpu(raw_inode->i_namelen); name.name = raw_inode->i_name; - +retry: de = f2fs_find_entry(dir, &name, &page); - if (de) { + if (de && inode->i_ino == le32_to_cpu(de->ino)) { kunmap(page); f2fs_put_page(page, 0); - } else { - err = __f2fs_add_link(dir, &name, inode); + goto out; + } + if (de) { + einode = f2fs_iget(inode->i_sb, le32_to_cpu(de->ino)); + if (IS_ERR(einode)) { + WARN_ON(1); + if (PTR_ERR(einode) == -ENOENT) + err = -EEXIST; + goto out; + } + f2fs_delete_entry(de, page, einode); + iput(einode); + goto retry; } - iput(dir); + err = __f2fs_add_link(dir, &name, inode); out: - kunmap(ipage); + f2fs_msg(inode->i_sb, KERN_NOTICE, "recover_inode and its dentry: " + "ino = %x, name = %s, dir = %lx, err = %d", + ino_of_node(ipage), raw_inode->i_name, + IS_ERR(dir) ? 0 : dir->i_ino, err); return err; } @@ -79,6 +97,9 @@ static int recover_inode(struct inode *inode, struct page *node_page) struct f2fs_node *raw_node = (struct f2fs_node *)kaddr; struct f2fs_inode *raw_inode = &(raw_node->i); + if (!IS_INODE(node_page)) + return 0; + inode->i_mode = le16_to_cpu(raw_inode->i_mode); i_size_write(inode, le64_to_cpu(raw_inode->i_size)); inode->i_atime.tv_sec = le64_to_cpu(raw_inode->i_mtime); @@ -88,7 +109,12 @@ static int recover_inode(struct inode *inode, struct page *node_page) inode->i_ctime.tv_nsec = le32_to_cpu(raw_inode->i_ctime_nsec); inode->i_mtime.tv_nsec = le32_to_cpu(raw_inode->i_mtime_nsec); - return recover_dentry(node_page, inode); + if (is_dent_dnode(node_page)) + return recover_dentry(node_page, inode); + + f2fs_msg(inode->i_sb, KERN_NOTICE, "recover_inode: ino = %x, name = %s", + ino_of_node(node_page), raw_inode->i_name); + return 0; } static int find_fsync_dnodes(struct f2fs_sb_info *sbi, struct list_head *head) @@ -119,14 +145,13 @@ static int find_fsync_dnodes(struct f2fs_sb_info *sbi, struct list_head *head) lock_page(page); if (cp_ver != cpver_of_node(page)) - goto unlock_out; + break; if (!is_fsync_dnode(page)) goto next; entry = get_fsync_inode(head, ino_of_node(page)); if (entry) { - entry->blkaddr = blkaddr; if (IS_INODE(page) && is_dent_dnode(page)) set_inode_flag(F2FS_I(entry->inode), FI_INC_LINK); @@ -134,48 +159,40 @@ static int find_fsync_dnodes(struct f2fs_sb_info *sbi, struct list_head *head) if (IS_INODE(page) && is_dent_dnode(page)) { err = recover_inode_page(sbi, page); if (err) - goto unlock_out; + break; } /* add this fsync inode to the list */ entry = kmem_cache_alloc(fsync_entry_slab, GFP_NOFS); if (!entry) { err = -ENOMEM; - goto unlock_out; + break; } entry->inode = f2fs_iget(sbi->sb, ino_of_node(page)); if (IS_ERR(entry->inode)) { err = PTR_ERR(entry->inode); kmem_cache_free(fsync_entry_slab, entry); - goto unlock_out; + break; } - list_add_tail(&entry->list, head); - entry->blkaddr = blkaddr; - } - if (IS_INODE(page)) { - err = recover_inode(entry->inode, page); - if (err == -ENOENT) { - goto next; - } else if (err) { - err = -EINVAL; - goto unlock_out; - } } + entry->blkaddr = blkaddr; + + err = recover_inode(entry->inode, page); + if (err && err != -ENOENT) + break; next: /* check next segment */ blkaddr = next_blkaddr_of_node(page); } -unlock_out: unlock_page(page); out: __free_pages(page, 0); return err; } -static void destroy_fsync_dnodes(struct f2fs_sb_info *sbi, - struct list_head *head) +static void destroy_fsync_dnodes(struct list_head *head) { struct fsync_inode_entry *entry, *tmp; @@ -186,15 +203,15 @@ static void destroy_fsync_dnodes(struct f2fs_sb_info *sbi, } } -static void check_index_in_prev_nodes(struct f2fs_sb_info *sbi, - block_t blkaddr) +static int check_index_in_prev_nodes(struct f2fs_sb_info *sbi, + block_t blkaddr, struct dnode_of_data *dn) { struct seg_entry *sentry; unsigned int segno = GET_SEGNO(sbi, blkaddr); unsigned short blkoff = GET_SEGOFF_FROM_SEG0(sbi, blkaddr) & (sbi->blocks_per_seg - 1); struct f2fs_summary sum; - nid_t ino; + nid_t ino, nid; void *kaddr; struct inode *inode; struct page *node_page; @@ -203,7 +220,7 @@ static void check_index_in_prev_nodes(struct f2fs_sb_info *sbi, sentry = get_seg_entry(sbi, segno); if (!f2fs_test_bit(blkoff, sentry->cur_valid_map)) - return; + return 0; /* Get the previous summary */ for (i = CURSEG_WARM_DATA; i <= CURSEG_COLD_DATA; i++) { @@ -222,20 +239,39 @@ static void check_index_in_prev_nodes(struct f2fs_sb_info *sbi, f2fs_put_page(sum_page, 1); } + /* Use the locked dnode page and inode */ + nid = le32_to_cpu(sum.nid); + if (dn->inode->i_ino == nid) { + struct dnode_of_data tdn = *dn; + tdn.nid = nid; + tdn.node_page = dn->inode_page; + tdn.ofs_in_node = le16_to_cpu(sum.ofs_in_node); + truncate_data_blocks_range(&tdn, 1); + return 0; + } else if (dn->nid == nid) { + struct dnode_of_data tdn = *dn; + tdn.ofs_in_node = le16_to_cpu(sum.ofs_in_node); + truncate_data_blocks_range(&tdn, 1); + return 0; + } + /* Get the node page */ - node_page = get_node_page(sbi, le32_to_cpu(sum.nid)); + node_page = get_node_page(sbi, nid); + if (IS_ERR(node_page)) + return PTR_ERR(node_page); bidx = start_bidx_of_node(ofs_of_node(node_page)) + - le16_to_cpu(sum.ofs_in_node); + le16_to_cpu(sum.ofs_in_node); ino = ino_of_node(node_page); f2fs_put_page(node_page, 1); /* Deallocate previous index in the node page */ inode = f2fs_iget(sbi->sb, ino); if (IS_ERR(inode)) - return; + return PTR_ERR(inode); truncate_hole(inode, bidx, bidx + 1); iput(inode); + return 0; } static int do_recover_data(struct f2fs_sb_info *sbi, struct inode *inode, @@ -245,7 +281,7 @@ static int do_recover_data(struct f2fs_sb_info *sbi, struct inode *inode, struct dnode_of_data dn; struct f2fs_summary sum; struct node_info ni; - int err = 0; + int err = 0, recovered = 0; int ilock; start = start_bidx_of_node(ofs_of_node(page)); @@ -283,13 +319,16 @@ static int do_recover_data(struct f2fs_sb_info *sbi, struct inode *inode, } /* Check the previous node page having this index */ - check_index_in_prev_nodes(sbi, dest); + err = check_index_in_prev_nodes(sbi, dest, &dn); + if (err) + goto err; set_summary(&sum, dn.nid, dn.ofs_in_node, ni.version); /* write dummy data page */ recover_data_page(sbi, NULL, &sum, src, dest); update_extent_cache(dest, &dn); + recovered++; } dn.ofs_in_node++; } @@ -305,9 +344,14 @@ static int do_recover_data(struct f2fs_sb_info *sbi, struct inode *inode, set_page_dirty(dn.node_page); recover_node_page(sbi, dn.node_page, &sum, &ni, blkaddr); +err: f2fs_put_dnode(&dn); mutex_unlock_op(sbi, ilock); - return 0; + + f2fs_msg(sbi->sb, KERN_NOTICE, "recover_data: ino = %lx, " + "recovered_data = %d blocks, err = %d", + inode->i_ino, recovered, err); + return err; } static int recover_data(struct f2fs_sb_info *sbi, @@ -340,7 +384,7 @@ static int recover_data(struct f2fs_sb_info *sbi, lock_page(page); if (cp_ver != cpver_of_node(page)) - goto unlock_out; + break; entry = get_fsync_inode(head, ino_of_node(page)); if (!entry) @@ -348,7 +392,7 @@ static int recover_data(struct f2fs_sb_info *sbi, err = do_recover_data(sbi, entry->inode, page, blkaddr); if (err) - goto out; + break; if (entry->blkaddr == blkaddr) { iput(entry->inode); @@ -359,7 +403,6 @@ next: /* check next segment */ blkaddr = next_blkaddr_of_node(page); } -unlock_out: unlock_page(page); out: __free_pages(page, 0); @@ -382,6 +425,7 @@ int recover_fsync_data(struct f2fs_sb_info *sbi) INIT_LIST_HEAD(&inode_list); /* step #1: find fsynced inode numbers */ + sbi->por_doing = 1; err = find_fsync_dnodes(sbi, &inode_list); if (err) goto out; @@ -390,13 +434,13 @@ int recover_fsync_data(struct f2fs_sb_info *sbi) goto out; /* step #2: recover data */ - sbi->por_doing = 1; err = recover_data(sbi, &inode_list, CURSEG_WARM_NODE); - sbi->por_doing = 0; BUG_ON(!list_empty(&inode_list)); out: - destroy_fsync_dnodes(sbi, &inode_list); + destroy_fsync_dnodes(&inode_list); kmem_cache_destroy(fsync_entry_slab); - write_checkpoint(sbi, false); + sbi->por_doing = 0; + if (!err) + write_checkpoint(sbi, false); return err; } diff --git a/fs/f2fs/segment.c b/fs/f2fs/segment.c index d8e84e49a5c3..a86d125a9885 100644 --- a/fs/f2fs/segment.c +++ b/fs/f2fs/segment.c @@ -94,7 +94,7 @@ static void __remove_dirty_segment(struct f2fs_sb_info *sbi, unsigned int segno, * Adding dirty entry into seglist is not critical operation. * If a given segment is one of current working segments, it won't be added. */ -void locate_dirty_segment(struct f2fs_sb_info *sbi, unsigned int segno) +static void locate_dirty_segment(struct f2fs_sb_info *sbi, unsigned int segno) { struct dirty_seglist_info *dirty_i = DIRTY_I(sbi); unsigned short valid_blocks; @@ -126,17 +126,16 @@ void locate_dirty_segment(struct f2fs_sb_info *sbi, unsigned int segno) static void set_prefree_as_free_segments(struct f2fs_sb_info *sbi) { struct dirty_seglist_info *dirty_i = DIRTY_I(sbi); - unsigned int segno, offset = 0; + unsigned int segno = -1; unsigned int total_segs = TOTAL_SEGS(sbi); mutex_lock(&dirty_i->seglist_lock); while (1) { segno = find_next_bit(dirty_i->dirty_segmap[PRE], total_segs, - offset); + segno + 1); if (segno >= total_segs) break; __set_test_and_free(sbi, segno); - offset = segno + 1; } mutex_unlock(&dirty_i->seglist_lock); } @@ -144,17 +143,16 @@ static void set_prefree_as_free_segments(struct f2fs_sb_info *sbi) void clear_prefree_segments(struct f2fs_sb_info *sbi) { struct dirty_seglist_info *dirty_i = DIRTY_I(sbi); - unsigned int segno, offset = 0; + unsigned int segno = -1; unsigned int total_segs = TOTAL_SEGS(sbi); mutex_lock(&dirty_i->seglist_lock); while (1) { segno = find_next_bit(dirty_i->dirty_segmap[PRE], total_segs, - offset); + segno + 1); if (segno >= total_segs) break; - offset = segno + 1; if (test_and_clear_bit(segno, dirty_i->dirty_segmap[PRE])) dirty_i->nr_dirty[PRE]--; @@ -257,11 +255,11 @@ void invalidate_blocks(struct f2fs_sb_info *sbi, block_t addr) * This function should be resided under the curseg_mutex lock */ static void __add_sum_entry(struct f2fs_sb_info *sbi, int type, - struct f2fs_summary *sum, unsigned short offset) + struct f2fs_summary *sum) { struct curseg_info *curseg = CURSEG_I(sbi, type); void *addr = curseg->sum_blk; - addr += offset * sizeof(struct f2fs_summary); + addr += curseg->next_blkoff * sizeof(struct f2fs_summary); memcpy(addr, sum, sizeof(struct f2fs_summary)); return; } @@ -311,64 +309,14 @@ static void write_sum_page(struct f2fs_sb_info *sbi, f2fs_put_page(page, 1); } -static unsigned int check_prefree_segments(struct f2fs_sb_info *sbi, int type) -{ - struct dirty_seglist_info *dirty_i = DIRTY_I(sbi); - unsigned long *prefree_segmap = dirty_i->dirty_segmap[PRE]; - unsigned int segno; - unsigned int ofs = 0; - - /* - * If there is not enough reserved sections, - * we should not reuse prefree segments. - */ - if (has_not_enough_free_secs(sbi, 0)) - return NULL_SEGNO; - - /* - * NODE page should not reuse prefree segment, - * since those information is used for SPOR. - */ - if (IS_NODESEG(type)) - return NULL_SEGNO; -next: - segno = find_next_bit(prefree_segmap, TOTAL_SEGS(sbi), ofs); - ofs += sbi->segs_per_sec; - - if (segno < TOTAL_SEGS(sbi)) { - int i; - - /* skip intermediate segments in a section */ - if (segno % sbi->segs_per_sec) - goto next; - - /* skip if the section is currently used */ - if (sec_usage_check(sbi, GET_SECNO(sbi, segno))) - goto next; - - /* skip if whole section is not prefree */ - for (i = 1; i < sbi->segs_per_sec; i++) - if (!test_bit(segno + i, prefree_segmap)) - goto next; - - /* skip if whole section was not free at the last checkpoint */ - for (i = 0; i < sbi->segs_per_sec; i++) - if (get_seg_entry(sbi, segno + i)->ckpt_valid_blocks) - goto next; - - return segno; - } - return NULL_SEGNO; -} - static int is_next_segment_free(struct f2fs_sb_info *sbi, int type) { struct curseg_info *curseg = CURSEG_I(sbi, type); - unsigned int segno = curseg->segno; + unsigned int segno = curseg->segno + 1; struct free_segmap_info *free_i = FREE_I(sbi); - if (segno + 1 < TOTAL_SEGS(sbi) && (segno + 1) % sbi->segs_per_sec) - return !test_bit(segno + 1, free_i->free_segmap); + if (segno < TOTAL_SEGS(sbi) && segno % sbi->segs_per_sec) + return !test_bit(segno, free_i->free_segmap); return 0; } @@ -495,7 +443,7 @@ static void new_curseg(struct f2fs_sb_info *sbi, int type, bool new_sec) int dir = ALLOC_LEFT; write_sum_page(sbi, curseg->sum_blk, - GET_SUM_BLOCK(sbi, curseg->segno)); + GET_SUM_BLOCK(sbi, segno)); if (type == CURSEG_WARM_DATA || type == CURSEG_COLD_DATA) dir = ALLOC_RIGHT; @@ -599,11 +547,7 @@ static void allocate_segment_by_default(struct f2fs_sb_info *sbi, goto out; } - curseg->next_segno = check_prefree_segments(sbi, type); - - if (curseg->next_segno != NULL_SEGNO) - change_curseg(sbi, type, false); - else if (type == CURSEG_WARM_NODE) + if (type == CURSEG_WARM_NODE) new_curseg(sbi, type, false); else if (curseg->alloc_type == LFS && is_next_segment_free(sbi, type)) new_curseg(sbi, type, false); @@ -612,7 +556,10 @@ static void allocate_segment_by_default(struct f2fs_sb_info *sbi, else new_curseg(sbi, type, false); out: +#ifdef CONFIG_F2FS_STAT_FS sbi->segment_count[curseg->alloc_type]++; +#endif + return; } void allocate_new_segments(struct f2fs_sb_info *sbi) @@ -795,7 +742,7 @@ static int __get_segment_type_6(struct page *page, enum page_type p_type) if (S_ISDIR(inode->i_mode)) return CURSEG_HOT_DATA; - else if (is_cold_data(page) || is_cold_file(inode)) + else if (is_cold_data(page) || file_is_cold(inode)) return CURSEG_COLD_DATA; else return CURSEG_WARM_DATA; @@ -844,11 +791,13 @@ static void do_write_page(struct f2fs_sb_info *sbi, struct page *page, * because, this function updates a summary entry in the * current summary block. */ - __add_sum_entry(sbi, type, sum, curseg->next_blkoff); + __add_sum_entry(sbi, type, sum); mutex_lock(&sit_i->sentry_lock); __refresh_next_blkoff(sbi, curseg); +#ifdef CONFIG_F2FS_STAT_FS sbi->block_count[curseg->alloc_type]++; +#endif /* * SIT information should be updated before segment allocation, @@ -943,7 +892,7 @@ void recover_data_page(struct f2fs_sb_info *sbi, curseg->next_blkoff = GET_SEGOFF_FROM_SEG0(sbi, new_blkaddr) & (sbi->blocks_per_seg - 1); - __add_sum_entry(sbi, type, sum, curseg->next_blkoff); + __add_sum_entry(sbi, type, sum); refresh_sit_entry(sbi, old_blkaddr, new_blkaddr); @@ -980,7 +929,7 @@ void rewrite_node_page(struct f2fs_sb_info *sbi, } curseg->next_blkoff = GET_SEGOFF_FROM_SEG0(sbi, new_blkaddr) & (sbi->blocks_per_seg - 1); - __add_sum_entry(sbi, type, sum, curseg->next_blkoff); + __add_sum_entry(sbi, type, sum); /* change the current log to the next block addr in advance */ if (next_segno != segno) { @@ -1579,13 +1528,13 @@ static void init_dirty_segmap(struct f2fs_sb_info *sbi) { struct dirty_seglist_info *dirty_i = DIRTY_I(sbi); struct free_segmap_info *free_i = FREE_I(sbi); - unsigned int segno = 0, offset = 0; + unsigned int segno = 0, offset = 0, total_segs = TOTAL_SEGS(sbi); unsigned short valid_blocks; - while (segno < TOTAL_SEGS(sbi)) { + while (1) { /* find dirty segment based on free segmap */ - segno = find_next_inuse(free_i, TOTAL_SEGS(sbi), offset); - if (segno >= TOTAL_SEGS(sbi)) + segno = find_next_inuse(free_i, total_segs, offset); + if (segno >= total_segs) break; offset = segno + 1; valid_blocks = get_valid_blocks(sbi, segno, 0); diff --git a/fs/f2fs/super.c b/fs/f2fs/super.c index 8555f7df82c7..75c7dc363e92 100644 --- a/fs/f2fs/super.c +++ b/fs/f2fs/super.c @@ -34,7 +34,7 @@ static struct kmem_cache *f2fs_inode_cachep; enum { - Opt_gc_background_off, + Opt_gc_background, Opt_disable_roll_forward, Opt_discard, Opt_noheap, @@ -46,7 +46,7 @@ enum { }; static match_table_t f2fs_tokens = { - {Opt_gc_background_off, "background_gc_off"}, + {Opt_gc_background, "background_gc=%s"}, {Opt_disable_roll_forward, "disable_roll_forward"}, {Opt_discard, "discard"}, {Opt_noheap, "no_heap"}, @@ -76,6 +76,91 @@ static void init_once(void *foo) inode_init_once(&fi->vfs_inode); } +static int parse_options(struct super_block *sb, char *options) +{ + struct f2fs_sb_info *sbi = F2FS_SB(sb); + substring_t args[MAX_OPT_ARGS]; + char *p, *name; + int arg = 0; + + if (!options) + return 0; + + while ((p = strsep(&options, ",")) != NULL) { + int token; + if (!*p) + continue; + /* + * Initialize args struct so we know whether arg was + * found; some options take optional arguments. + */ + args[0].to = args[0].from = NULL; + token = match_token(p, f2fs_tokens, args); + + switch (token) { + case Opt_gc_background: + name = match_strdup(&args[0]); + + if (!name) + return -ENOMEM; + if (!strncmp(name, "on", 2)) + set_opt(sbi, BG_GC); + else if (!strncmp(name, "off", 3)) + clear_opt(sbi, BG_GC); + else { + kfree(name); + return -EINVAL; + } + kfree(name); + break; + case Opt_disable_roll_forward: + set_opt(sbi, DISABLE_ROLL_FORWARD); + break; + case Opt_discard: + set_opt(sbi, DISCARD); + break; + case Opt_noheap: + set_opt(sbi, NOHEAP); + break; +#ifdef CONFIG_F2FS_FS_XATTR + case Opt_nouser_xattr: + clear_opt(sbi, XATTR_USER); + break; +#else + case Opt_nouser_xattr: + f2fs_msg(sb, KERN_INFO, + "nouser_xattr options not supported"); + break; +#endif +#ifdef CONFIG_F2FS_FS_POSIX_ACL + case Opt_noacl: + clear_opt(sbi, POSIX_ACL); + break; +#else + case Opt_noacl: + f2fs_msg(sb, KERN_INFO, "noacl options not supported"); + break; +#endif + case Opt_active_logs: + if (args->from && match_int(args, &arg)) + return -EINVAL; + if (arg != 2 && arg != 4 && arg != NR_CURSEG_TYPE) + return -EINVAL; + sbi->active_logs = arg; + break; + case Opt_disable_ext_identify: + set_opt(sbi, DISABLE_EXT_IDENTIFY); + break; + default: + f2fs_msg(sb, KERN_ERR, + "Unrecognized mount option \"%s\" or missing value", + p); + return -EINVAL; + } + } + return 0; +} + static struct inode *f2fs_alloc_inode(struct super_block *sb) { struct f2fs_inode_info *fi; @@ -112,6 +197,17 @@ static int f2fs_drop_inode(struct inode *inode) return generic_drop_inode(inode); } +/* + * f2fs_dirty_inode() is called from __mark_inode_dirty() + * + * We should call set_dirty_inode to write the dirty inode through write_inode. + */ +static void f2fs_dirty_inode(struct inode *inode, int flags) +{ + set_inode_flag(F2FS_I(inode), FI_DIRTY_INODE); + return; +} + static void f2fs_i_callback(struct rcu_head *head) { struct inode *inode = container_of(head, struct inode, i_rcu); @@ -170,7 +266,7 @@ static int f2fs_freeze(struct super_block *sb) { int err; - if (sb->s_flags & MS_RDONLY) + if (f2fs_readonly(sb)) return 0; err = f2fs_sync_fs(sb, 1); @@ -214,10 +310,10 @@ static int f2fs_show_options(struct seq_file *seq, struct dentry *root) { struct f2fs_sb_info *sbi = F2FS_SB(root->d_sb); - if (test_opt(sbi, BG_GC)) - seq_puts(seq, ",background_gc_on"); + if (!(root->d_sb->s_flags & MS_RDONLY) && test_opt(sbi, BG_GC)) + seq_printf(seq, ",background_gc=%s", "on"); else - seq_puts(seq, ",background_gc_off"); + seq_printf(seq, ",background_gc=%s", "off"); if (test_opt(sbi, DISABLE_ROLL_FORWARD)) seq_puts(seq, ",disable_roll_forward"); if (test_opt(sbi, DISCARD)) @@ -244,11 +340,64 @@ static int f2fs_show_options(struct seq_file *seq, struct dentry *root) return 0; } +static int f2fs_remount(struct super_block *sb, int *flags, char *data) +{ + struct f2fs_sb_info *sbi = F2FS_SB(sb); + struct f2fs_mount_info org_mount_opt; + int err, active_logs; + + /* + * Save the old mount options in case we + * need to restore them. + */ + org_mount_opt = sbi->mount_opt; + active_logs = sbi->active_logs; + + /* parse mount options */ + err = parse_options(sb, data); + if (err) + goto restore_opts; + + /* + * Previous and new state of filesystem is RO, + * so no point in checking GC conditions. + */ + if ((sb->s_flags & MS_RDONLY) && (*flags & MS_RDONLY)) + goto skip; + + /* + * We stop the GC thread if FS is mounted as RO + * or if background_gc = off is passed in mount + * option. Also sync the filesystem. + */ + if ((*flags & MS_RDONLY) || !test_opt(sbi, BG_GC)) { + if (sbi->gc_thread) { + stop_gc_thread(sbi); + f2fs_sync_fs(sb, 1); + } + } else if (test_opt(sbi, BG_GC) && !sbi->gc_thread) { + err = start_gc_thread(sbi); + if (err) + goto restore_opts; + } +skip: + /* Update the POSIXACL Flag */ + sb->s_flags = (sb->s_flags & ~MS_POSIXACL) | + (test_opt(sbi, POSIX_ACL) ? MS_POSIXACL : 0); + return 0; + +restore_opts: + sbi->mount_opt = org_mount_opt; + sbi->active_logs = active_logs; + return err; +} + static struct super_operations f2fs_sops = { .alloc_inode = f2fs_alloc_inode, .drop_inode = f2fs_drop_inode, .destroy_inode = f2fs_destroy_inode, .write_inode = f2fs_write_inode, + .dirty_inode = f2fs_dirty_inode, .show_options = f2fs_show_options, .evict_inode = f2fs_evict_inode, .put_super = f2fs_put_super, @@ -256,6 +405,7 @@ static struct super_operations f2fs_sops = { .freeze_fs = f2fs_freeze, .unfreeze_fs = f2fs_unfreeze, .statfs = f2fs_statfs, + .remount_fs = f2fs_remount, }; static struct inode *f2fs_nfs_get_inode(struct super_block *sb, @@ -303,79 +453,6 @@ static const struct export_operations f2fs_export_ops = { .get_parent = f2fs_get_parent, }; -static int parse_options(struct super_block *sb, struct f2fs_sb_info *sbi, - char *options) -{ - substring_t args[MAX_OPT_ARGS]; - char *p; - int arg = 0; - - if (!options) - return 0; - - while ((p = strsep(&options, ",")) != NULL) { - int token; - if (!*p) - continue; - /* - * Initialize args struct so we know whether arg was - * found; some options take optional arguments. - */ - args[0].to = args[0].from = NULL; - token = match_token(p, f2fs_tokens, args); - - switch (token) { - case Opt_gc_background_off: - clear_opt(sbi, BG_GC); - break; - case Opt_disable_roll_forward: - set_opt(sbi, DISABLE_ROLL_FORWARD); - break; - case Opt_discard: - set_opt(sbi, DISCARD); - break; - case Opt_noheap: - set_opt(sbi, NOHEAP); - break; -#ifdef CONFIG_F2FS_FS_XATTR - case Opt_nouser_xattr: - clear_opt(sbi, XATTR_USER); - break; -#else - case Opt_nouser_xattr: - f2fs_msg(sb, KERN_INFO, - "nouser_xattr options not supported"); - break; -#endif -#ifdef CONFIG_F2FS_FS_POSIX_ACL - case Opt_noacl: - clear_opt(sbi, POSIX_ACL); - break; -#else - case Opt_noacl: - f2fs_msg(sb, KERN_INFO, "noacl options not supported"); - break; -#endif - case Opt_active_logs: - if (args->from && match_int(args, &arg)) - return -EINVAL; - if (arg != 2 && arg != 4 && arg != NR_CURSEG_TYPE) - return -EINVAL; - sbi->active_logs = arg; - break; - case Opt_disable_ext_identify: - set_opt(sbi, DISABLE_EXT_IDENTIFY); - break; - default: - f2fs_msg(sb, KERN_ERR, - "Unrecognized mount option \"%s\" or missing value", - p); - return -EINVAL; - } - } - return 0; -} - static loff_t max_file_size(unsigned bits) { loff_t result = ADDRS_PER_INODE; @@ -541,6 +618,7 @@ static int f2fs_fill_super(struct super_block *sb, void *data, int silent) if (err) goto free_sb_buf; } + sb->s_fs_info = sbi; /* init some FS parameters */ sbi->active_logs = NR_CURSEG_TYPE; @@ -553,7 +631,7 @@ static int f2fs_fill_super(struct super_block *sb, void *data, int silent) set_opt(sbi, POSIX_ACL); #endif /* parse mount options */ - err = parse_options(sb, sbi, (char *)data); + err = parse_options(sb, (char *)data); if (err) goto free_sb_buf; @@ -565,7 +643,6 @@ static int f2fs_fill_super(struct super_block *sb, void *data, int silent) sb->s_xattr = f2fs_xattr_handlers; sb->s_export_op = &f2fs_export_ops; sb->s_magic = F2FS_SUPER_MAGIC; - sb->s_fs_info = sbi; sb->s_time_gran = 1; sb->s_flags = (sb->s_flags & ~MS_POSIXACL) | (test_opt(sbi, POSIX_ACL) ? MS_POSIXACL : 0); @@ -674,10 +751,16 @@ static int f2fs_fill_super(struct super_block *sb, void *data, int silent) "Cannot recover all fsync data errno=%ld", err); } - /* After POR, we can run background GC thread */ - err = start_gc_thread(sbi); - if (err) - goto fail; + /* + * If filesystem is not mounted as read-only then + * do start the gc_thread. + */ + if (!(sb->s_flags & MS_RDONLY)) { + /* After POR, we can run background GC thread.*/ + err = start_gc_thread(sbi); + if (err) + goto fail; + } err = f2fs_build_stats(sbi); if (err) diff --git a/fs/f2fs/xattr.c b/fs/f2fs/xattr.c index 0b02dce31356..3ab07ecd86ca 100644 --- a/fs/f2fs/xattr.c +++ b/fs/f2fs/xattr.c @@ -20,6 +20,7 @@ */ #include <linux/rwsem.h> #include <linux/f2fs_fs.h> +#include <linux/security.h> #include "f2fs.h" #include "xattr.h" @@ -43,6 +44,10 @@ static size_t f2fs_xattr_generic_list(struct dentry *dentry, char *list, prefix = XATTR_TRUSTED_PREFIX; prefix_len = XATTR_TRUSTED_PREFIX_LEN; break; + case F2FS_XATTR_INDEX_SECURITY: + prefix = XATTR_SECURITY_PREFIX; + prefix_len = XATTR_SECURITY_PREFIX_LEN; + break; default: return -EINVAL; } @@ -50,7 +55,7 @@ static size_t f2fs_xattr_generic_list(struct dentry *dentry, char *list, total_len = prefix_len + name_len + 1; if (list && total_len <= list_size) { memcpy(list, prefix, prefix_len); - memcpy(list+prefix_len, name, name_len); + memcpy(list + prefix_len, name, name_len); list[prefix_len + name_len] = '\0'; } return total_len; @@ -70,13 +75,14 @@ static int f2fs_xattr_generic_get(struct dentry *dentry, const char *name, if (!capable(CAP_SYS_ADMIN)) return -EPERM; break; + case F2FS_XATTR_INDEX_SECURITY: + break; default: return -EINVAL; } if (strcmp(name, "") == 0) return -EINVAL; - return f2fs_getxattr(dentry->d_inode, type, name, - buffer, size); + return f2fs_getxattr(dentry->d_inode, type, name, buffer, size); } static int f2fs_xattr_generic_set(struct dentry *dentry, const char *name, @@ -93,13 +99,15 @@ static int f2fs_xattr_generic_set(struct dentry *dentry, const char *name, if (!capable(CAP_SYS_ADMIN)) return -EPERM; break; + case F2FS_XATTR_INDEX_SECURITY: + break; default: return -EINVAL; } if (strcmp(name, "") == 0) return -EINVAL; - return f2fs_setxattr(dentry->d_inode, type, name, value, size); + return f2fs_setxattr(dentry->d_inode, type, name, value, size, NULL); } static size_t f2fs_xattr_advise_list(struct dentry *dentry, char *list, @@ -145,6 +153,31 @@ static int f2fs_xattr_advise_set(struct dentry *dentry, const char *name, return 0; } +#ifdef CONFIG_F2FS_FS_SECURITY +static int f2fs_initxattrs(struct inode *inode, const struct xattr *xattr_array, + void *page) +{ + const struct xattr *xattr; + int err = 0; + + for (xattr = xattr_array; xattr->name != NULL; xattr++) { + err = f2fs_setxattr(inode, F2FS_XATTR_INDEX_SECURITY, + xattr->name, xattr->value, + xattr->value_len, (struct page *)page); + if (err < 0) + break; + } + return err; +} + +int f2fs_init_security(struct inode *inode, struct inode *dir, + const struct qstr *qstr, struct page *ipage) +{ + return security_inode_init_security(inode, dir, qstr, + &f2fs_initxattrs, ipage); +} +#endif + const struct xattr_handler f2fs_xattr_user_handler = { .prefix = XATTR_USER_PREFIX, .flags = F2FS_XATTR_INDEX_USER, @@ -169,6 +202,14 @@ const struct xattr_handler f2fs_xattr_advise_handler = { .set = f2fs_xattr_advise_set, }; +const struct xattr_handler f2fs_xattr_security_handler = { + .prefix = XATTR_SECURITY_PREFIX, + .flags = F2FS_XATTR_INDEX_SECURITY, + .list = f2fs_xattr_generic_list, + .get = f2fs_xattr_generic_get, + .set = f2fs_xattr_generic_set, +}; + static const struct xattr_handler *f2fs_xattr_handler_map[] = { [F2FS_XATTR_INDEX_USER] = &f2fs_xattr_user_handler, #ifdef CONFIG_F2FS_FS_POSIX_ACL @@ -176,6 +217,9 @@ static const struct xattr_handler *f2fs_xattr_handler_map[] = { [F2FS_XATTR_INDEX_POSIX_ACL_DEFAULT] = &f2fs_xattr_acl_default_handler, #endif [F2FS_XATTR_INDEX_TRUSTED] = &f2fs_xattr_trusted_handler, +#ifdef CONFIG_F2FS_FS_SECURITY + [F2FS_XATTR_INDEX_SECURITY] = &f2fs_xattr_security_handler, +#endif [F2FS_XATTR_INDEX_ADVISE] = &f2fs_xattr_advise_handler, }; @@ -186,6 +230,9 @@ const struct xattr_handler *f2fs_xattr_handlers[] = { &f2fs_xattr_acl_default_handler, #endif &f2fs_xattr_trusted_handler, +#ifdef CONFIG_F2FS_FS_SECURITY + &f2fs_xattr_security_handler, +#endif &f2fs_xattr_advise_handler, NULL, }; @@ -218,6 +265,8 @@ int f2fs_getxattr(struct inode *inode, int name_index, const char *name, return -ENODATA; page = get_node_page(sbi, fi->i_xattr_nid); + if (IS_ERR(page)) + return PTR_ERR(page); base_addr = page_address(page); list_for_each_xattr(entry, base_addr) { @@ -268,6 +317,8 @@ ssize_t f2fs_listxattr(struct dentry *dentry, char *buffer, size_t buffer_size) return 0; page = get_node_page(sbi, fi->i_xattr_nid); + if (IS_ERR(page)) + return PTR_ERR(page); base_addr = page_address(page); list_for_each_xattr(entry, base_addr) { @@ -296,7 +347,7 @@ cleanup: } int f2fs_setxattr(struct inode *inode, int name_index, const char *name, - const void *value, size_t value_len) + const void *value, size_t value_len, struct page *ipage) { struct f2fs_sb_info *sbi = F2FS_SB(inode->i_sb); struct f2fs_inode_info *fi = F2FS_I(inode); @@ -335,7 +386,7 @@ int f2fs_setxattr(struct inode *inode, int name_index, const char *name, set_new_dnode(&dn, inode, NULL, NULL, fi->i_xattr_nid); mark_inode_dirty(inode); - page = new_node_page(&dn, XATTR_NODE_OFFSET); + page = new_node_page(&dn, XATTR_NODE_OFFSET, ipage); if (IS_ERR(page)) { alloc_nid_failed(sbi, fi->i_xattr_nid); fi->i_xattr_nid = 0; @@ -435,7 +486,10 @@ int f2fs_setxattr(struct inode *inode, int name_index, const char *name, inode->i_ctime = CURRENT_TIME; clear_inode_flag(fi, FI_ACL_MODE); } - update_inode_page(inode); + if (ipage) + update_inode(inode, ipage); + else + update_inode_page(inode); mutex_unlock_op(sbi, ilock); return 0; diff --git a/fs/f2fs/xattr.h b/fs/f2fs/xattr.h index 49c9558305e3..3c0817bef25d 100644 --- a/fs/f2fs/xattr.h +++ b/fs/f2fs/xattr.h @@ -112,21 +112,19 @@ extern const struct xattr_handler f2fs_xattr_trusted_handler; extern const struct xattr_handler f2fs_xattr_acl_access_handler; extern const struct xattr_handler f2fs_xattr_acl_default_handler; extern const struct xattr_handler f2fs_xattr_advise_handler; +extern const struct xattr_handler f2fs_xattr_security_handler; extern const struct xattr_handler *f2fs_xattr_handlers[]; -extern int f2fs_setxattr(struct inode *inode, int name_index, const char *name, - const void *value, size_t value_len); -extern int f2fs_getxattr(struct inode *inode, int name_index, const char *name, - void *buffer, size_t buffer_size); -extern ssize_t f2fs_listxattr(struct dentry *dentry, char *buffer, - size_t buffer_size); - +extern int f2fs_setxattr(struct inode *, int, const char *, + const void *, size_t, struct page *); +extern int f2fs_getxattr(struct inode *, int, const char *, void *, size_t); +extern ssize_t f2fs_listxattr(struct dentry *, char *, size_t); #else #define f2fs_xattr_handlers NULL static inline int f2fs_setxattr(struct inode *inode, int name_index, - const char *name, const void *value, size_t value_len) + const char *name, const void *value, size_t value_len) { return -EOPNOTSUPP; } @@ -142,4 +140,14 @@ static inline ssize_t f2fs_listxattr(struct dentry *dentry, char *buffer, } #endif +#ifdef CONFIG_F2FS_FS_SECURITY +extern int f2fs_init_security(struct inode *, struct inode *, + const struct qstr *, struct page *); +#else +static inline int f2fs_init_security(struct inode *inode, struct inode *dir, + const struct qstr *qstr, struct page *ipage) +{ + return 0; +} +#endif #endif /* __F2FS_XATTR_H__ */ diff --git a/fs/fs-writeback.c b/fs/fs-writeback.c index 3be57189efd5..a85ac4e33436 100644 --- a/fs/fs-writeback.c +++ b/fs/fs-writeback.c @@ -45,6 +45,7 @@ struct wb_writeback_work { unsigned int for_kupdate:1; unsigned int range_cyclic:1; unsigned int for_background:1; + unsigned int for_sync:1; /* sync(2) WB_SYNC_ALL writeback */ enum wb_reason reason; /* why was writeback initiated? */ struct list_head list; /* pending work list */ @@ -443,9 +444,11 @@ __writeback_single_inode(struct inode *inode, struct writeback_control *wbc) /* * Make sure to wait on the data before writing out the metadata. * This is important for filesystems that modify metadata on data - * I/O completion. + * I/O completion. We don't do it for sync(2) writeback because it has a + * separate, external IO completion path and ->sync_fs for guaranteeing + * inode metadata is written back correctly. */ - if (wbc->sync_mode == WB_SYNC_ALL) { + if (wbc->sync_mode == WB_SYNC_ALL && !wbc->for_sync) { int err = filemap_fdatawait(mapping); if (ret == 0) ret = err; @@ -578,6 +581,7 @@ static long writeback_sb_inodes(struct super_block *sb, .tagged_writepages = work->tagged_writepages, .for_kupdate = work->for_kupdate, .for_background = work->for_background, + .for_sync = work->for_sync, .range_cyclic = work->range_cyclic, .range_start = 0, .range_end = LLONG_MAX, @@ -1362,6 +1366,7 @@ void sync_inodes_sb(struct super_block *sb) .range_cyclic = 0, .done = &done, .reason = WB_REASON_SYNC, + .for_sync = 1, }; /* Nothing to do? */ diff --git a/fs/fscache/cache.c b/fs/fscache/cache.c index b52aed1dca97..f7cff367db7f 100644 --- a/fs/fscache/cache.c +++ b/fs/fscache/cache.c @@ -115,7 +115,7 @@ struct fscache_cache *fscache_select_cache_for_object( struct fscache_object, cookie_link); cache = object->cache; - if (object->state >= FSCACHE_OBJECT_DYING || + if (fscache_object_is_dying(object) || test_bit(FSCACHE_IOERROR, &cache->flags)) cache = NULL; @@ -224,8 +224,10 @@ int fscache_add_cache(struct fscache_cache *cache, BUG_ON(!ifsdef); cache->flags = 0; - ifsdef->event_mask = ULONG_MAX & ~(1 << FSCACHE_OBJECT_EV_CLEARED); - ifsdef->state = FSCACHE_OBJECT_ACTIVE; + ifsdef->event_mask = + ((1 << NR_FSCACHE_OBJECT_EVENTS) - 1) & + ~(1 << FSCACHE_OBJECT_EV_CLEARED); + __set_bit(FSCACHE_OBJECT_IS_AVAILABLE, &ifsdef->flags); if (!tagname) tagname = cache->identifier; @@ -330,25 +332,25 @@ static void fscache_withdraw_all_objects(struct fscache_cache *cache, { struct fscache_object *object; - spin_lock(&cache->object_list_lock); - while (!list_empty(&cache->object_list)) { - object = list_entry(cache->object_list.next, - struct fscache_object, cache_link); - list_move_tail(&object->cache_link, dying_objects); + spin_lock(&cache->object_list_lock); - _debug("withdraw %p", object->cookie); + if (!list_empty(&cache->object_list)) { + object = list_entry(cache->object_list.next, + struct fscache_object, cache_link); + list_move_tail(&object->cache_link, dying_objects); - spin_lock(&object->lock); - spin_unlock(&cache->object_list_lock); - fscache_raise_event(object, FSCACHE_OBJECT_EV_WITHDRAW); - spin_unlock(&object->lock); + _debug("withdraw %p", object->cookie); + + /* This must be done under object_list_lock to prevent + * a race with fscache_drop_object(). + */ + fscache_raise_event(object, FSCACHE_OBJECT_EV_KILL); + } + spin_unlock(&cache->object_list_lock); cond_resched(); - spin_lock(&cache->object_list_lock); } - - spin_unlock(&cache->object_list_lock); } /** diff --git a/fs/fscache/cookie.c b/fs/fscache/cookie.c index e2cba1f60c21..0e91a3c9fdb2 100644 --- a/fs/fscache/cookie.c +++ b/fs/fscache/cookie.c @@ -95,6 +95,11 @@ struct fscache_cookie *__fscache_acquire_cookie( atomic_set(&cookie->usage, 1); atomic_set(&cookie->n_children, 0); + /* We keep the active count elevated until relinquishment to prevent an + * attempt to wake up every time the object operations queue quiesces. + */ + atomic_set(&cookie->n_active, 1); + atomic_inc(&parent->usage); atomic_inc(&parent->n_children); @@ -177,7 +182,6 @@ static int fscache_acquire_non_index_cookie(struct fscache_cookie *cookie) cookie->flags = (1 << FSCACHE_COOKIE_LOOKING_UP) | - (1 << FSCACHE_COOKIE_CREATING) | (1 << FSCACHE_COOKIE_NO_DATA_YET); /* ask the cache to allocate objects for this cookie and its parent @@ -205,7 +209,7 @@ static int fscache_acquire_non_index_cookie(struct fscache_cookie *cookie) /* initiate the process of looking up all the objects in the chain * (done by fscache_initialise_object()) */ - fscache_enqueue_object(object); + fscache_raise_event(object, FSCACHE_OBJECT_EV_NEW_CHILD); spin_unlock(&cookie->lock); @@ -285,7 +289,7 @@ static int fscache_alloc_object(struct fscache_cache *cache, object_already_extant: ret = -ENOBUFS; - if (object->state >= FSCACHE_OBJECT_DYING) { + if (fscache_object_is_dead(object)) { spin_unlock(&cookie->lock); goto error; } @@ -321,7 +325,7 @@ static int fscache_attach_object(struct fscache_cookie *cookie, ret = -EEXIST; hlist_for_each_entry(p, &cookie->backing_objects, cookie_link) { if (p->cache == object->cache) { - if (p->state >= FSCACHE_OBJECT_DYING) + if (fscache_object_is_dying(p)) ret = -ENOBUFS; goto cant_attach_object; } @@ -332,7 +336,7 @@ static int fscache_attach_object(struct fscache_cookie *cookie, hlist_for_each_entry(p, &cookie->parent->backing_objects, cookie_link) { if (p->cache == object->cache) { - if (p->state >= FSCACHE_OBJECT_DYING) { + if (fscache_object_is_dying(p)) { ret = -ENOBUFS; spin_unlock(&cookie->parent->lock); goto cant_attach_object; @@ -400,7 +404,7 @@ void __fscache_invalidate(struct fscache_cookie *cookie) object = hlist_entry(cookie->backing_objects.first, struct fscache_object, cookie_link); - if (object->state < FSCACHE_OBJECT_DYING) + if (fscache_object_is_live(object)) fscache_raise_event( object, FSCACHE_OBJECT_EV_INVALIDATE); } @@ -467,9 +471,7 @@ EXPORT_SYMBOL(__fscache_update_cookie); */ void __fscache_relinquish_cookie(struct fscache_cookie *cookie, int retire) { - struct fscache_cache *cache; struct fscache_object *object; - unsigned long event; fscache_stat(&fscache_n_relinquishes); if (retire) @@ -481,8 +483,11 @@ void __fscache_relinquish_cookie(struct fscache_cookie *cookie, int retire) return; } - _enter("%p{%s,%p},%d", - cookie, cookie->def->name, cookie->netfs_data, retire); + _enter("%p{%s,%p,%d},%d", + cookie, cookie->def->name, cookie->netfs_data, + atomic_read(&cookie->n_active), retire); + + ASSERTCMP(atomic_read(&cookie->n_active), >, 0); if (atomic_read(&cookie->n_children) != 0) { printk(KERN_ERR "FS-Cache: Cookie '%s' still has children\n", @@ -490,62 +495,28 @@ void __fscache_relinquish_cookie(struct fscache_cookie *cookie, int retire) BUG(); } - /* wait for the cookie to finish being instantiated (or to fail) */ - if (test_bit(FSCACHE_COOKIE_CREATING, &cookie->flags)) { - fscache_stat(&fscache_n_relinquishes_waitcrt); - wait_on_bit(&cookie->flags, FSCACHE_COOKIE_CREATING, - fscache_wait_bit, TASK_UNINTERRUPTIBLE); - } - - event = retire ? FSCACHE_OBJECT_EV_RETIRE : FSCACHE_OBJECT_EV_RELEASE; + /* No further netfs-accessing operations on this cookie permitted */ + set_bit(FSCACHE_COOKIE_RELINQUISHED, &cookie->flags); + if (retire) + set_bit(FSCACHE_COOKIE_RETIRED, &cookie->flags); -try_again: spin_lock(&cookie->lock); - - /* break links with all the active objects */ - while (!hlist_empty(&cookie->backing_objects)) { - int n_reads; - object = hlist_entry(cookie->backing_objects.first, - struct fscache_object, - cookie_link); - - _debug("RELEASE OBJ%x", object->debug_id); - - set_bit(FSCACHE_COOKIE_WAITING_ON_READS, &cookie->flags); - n_reads = atomic_read(&object->n_reads); - if (n_reads) { - int n_ops = object->n_ops; - int n_in_progress = object->n_in_progress; - spin_unlock(&cookie->lock); - printk(KERN_ERR "FS-Cache:" - " Cookie '%s' still has %d outstanding reads (%d,%d)\n", - cookie->def->name, - n_reads, n_ops, n_in_progress); - wait_on_bit(&cookie->flags, FSCACHE_COOKIE_WAITING_ON_READS, - fscache_wait_bit, TASK_UNINTERRUPTIBLE); - printk("Wait finished\n"); - goto try_again; - } - - /* detach each cache object from the object cookie */ - spin_lock(&object->lock); - hlist_del_init(&object->cookie_link); - - cache = object->cache; - object->cookie = NULL; - fscache_raise_event(object, event); - spin_unlock(&object->lock); - - if (atomic_dec_and_test(&cookie->usage)) - /* the cookie refcount shouldn't be reduced to 0 yet */ - BUG(); + hlist_for_each_entry(object, &cookie->backing_objects, cookie_link) { + fscache_raise_event(object, FSCACHE_OBJECT_EV_KILL); } + spin_unlock(&cookie->lock); - /* detach pointers back to the netfs */ + /* Wait for cessation of activity requiring access to the netfs (when + * n_active reaches 0). + */ + if (!atomic_dec_and_test(&cookie->n_active)) + wait_on_atomic_t(&cookie->n_active, fscache_wait_atomic_t, + TASK_UNINTERRUPTIBLE); + + /* Clear pointers back to the netfs */ cookie->netfs_data = NULL; cookie->def = NULL; - - spin_unlock(&cookie->lock); + BUG_ON(cookie->stores.rnode); if (cookie->parent) { ASSERTCMP(atomic_read(&cookie->parent->usage), >, 0); @@ -553,7 +524,7 @@ try_again: atomic_dec(&cookie->parent->n_children); } - /* finally dispose of the cookie */ + /* Dispose of the netfs's link to the cookie */ ASSERTCMP(atomic_read(&cookie->usage), >, 0); fscache_cookie_put(cookie); diff --git a/fs/fscache/fsdef.c b/fs/fscache/fsdef.c index f5b4baee7352..10a2ade0bdf8 100644 --- a/fs/fscache/fsdef.c +++ b/fs/fscache/fsdef.c @@ -55,6 +55,7 @@ static struct fscache_cookie_def fscache_fsdef_index_def = { struct fscache_cookie fscache_fsdef_index = { .usage = ATOMIC_INIT(1), + .n_active = ATOMIC_INIT(1), .lock = __SPIN_LOCK_UNLOCKED(fscache_fsdef_index.lock), .backing_objects = HLIST_HEAD_INIT, .def = &fscache_fsdef_index_def, diff --git a/fs/fscache/internal.h b/fs/fscache/internal.h index ee38fef4be51..12d505bedb5c 100644 --- a/fs/fscache/internal.h +++ b/fs/fscache/internal.h @@ -93,14 +93,11 @@ static inline bool fscache_object_congested(void) extern int fscache_wait_bit(void *); extern int fscache_wait_bit_interruptible(void *); +extern int fscache_wait_atomic_t(atomic_t *); /* * object.c */ -extern const char fscache_object_states_short[FSCACHE_OBJECT__NSTATES][5]; - -extern void fscache_withdrawing_object(struct fscache_cache *, - struct fscache_object *); extern void fscache_enqueue_object(struct fscache_object *); /* @@ -110,8 +107,10 @@ extern void fscache_enqueue_object(struct fscache_object *); extern const struct file_operations fscache_objlist_fops; extern void fscache_objlist_add(struct fscache_object *); +extern void fscache_objlist_remove(struct fscache_object *); #else #define fscache_objlist_add(object) do {} while(0) +#define fscache_objlist_remove(object) do {} while(0) #endif /* @@ -291,6 +290,10 @@ static inline void fscache_raise_event(struct fscache_object *object, unsigned event) { BUG_ON(event >= NR_FSCACHE_OBJECT_EVENTS); +#if 0 + printk("*** fscache_raise_event(OBJ%d{%lx},%x)\n", + object->debug_id, object->event_mask, (1 << event)); +#endif if (!test_and_set_bit(event, &object->events) && test_bit(event, &object->event_mask)) fscache_enqueue_object(object); diff --git a/fs/fscache/main.c b/fs/fscache/main.c index f9d856773f79..7c27907e650c 100644 --- a/fs/fscache/main.c +++ b/fs/fscache/main.c @@ -205,7 +205,6 @@ int fscache_wait_bit(void *flags) schedule(); return 0; } -EXPORT_SYMBOL(fscache_wait_bit); /* * wait_on_bit() sleep function for interruptible waiting @@ -215,4 +214,12 @@ int fscache_wait_bit_interruptible(void *flags) schedule(); return signal_pending(current); } -EXPORT_SYMBOL(fscache_wait_bit_interruptible); + +/* + * wait_on_atomic_t() sleep function for uninterruptible waiting + */ +int fscache_wait_atomic_t(atomic_t *p) +{ + schedule(); + return 0; +} diff --git a/fs/fscache/netfs.c b/fs/fscache/netfs.c index e028b8eb1c40..b1bb6117473a 100644 --- a/fs/fscache/netfs.c +++ b/fs/fscache/netfs.c @@ -40,6 +40,7 @@ int __fscache_register_netfs(struct fscache_netfs *netfs) /* initialise the primary index cookie */ atomic_set(&netfs->primary_index->usage, 1); atomic_set(&netfs->primary_index->n_children, 0); + atomic_set(&netfs->primary_index->n_active, 1); netfs->primary_index->def = &fscache_fsdef_netfs_def; netfs->primary_index->parent = &fscache_fsdef_index; diff --git a/fs/fscache/object-list.c b/fs/fscache/object-list.c index f27c89d17885..e1959efad64f 100644 --- a/fs/fscache/object-list.c +++ b/fs/fscache/object-list.c @@ -70,13 +70,10 @@ void fscache_objlist_add(struct fscache_object *obj) write_unlock(&fscache_object_list_lock); } -/** - * fscache_object_destroy - Note that a cache object is about to be destroyed - * @object: The object to be destroyed - * - * Note the imminent destruction and deallocation of a cache object record. +/* + * Remove an object from the object list. */ -void fscache_object_destroy(struct fscache_object *obj) +void fscache_objlist_remove(struct fscache_object *obj) { write_lock(&fscache_object_list_lock); @@ -85,7 +82,6 @@ void fscache_object_destroy(struct fscache_object *obj) write_unlock(&fscache_object_list_lock); } -EXPORT_SYMBOL(fscache_object_destroy); /* * find the object in the tree on or after the specified index @@ -166,15 +162,14 @@ static int fscache_objlist_show(struct seq_file *m, void *v) { struct fscache_objlist_data *data = m->private; struct fscache_object *obj = v; + struct fscache_cookie *cookie; unsigned long config = data->config; - uint16_t keylen, auxlen; char _type[3], *type; - bool no_cookie; u8 *buf = data->buf, *p; if ((unsigned long) v == 1) { seq_puts(m, "OBJECT PARENT STAT CHLDN OPS OOP IPR EX READS" - " EM EV F S" + " EM EV FL S" " | NETFS_COOKIE_DEF TY FL NETFS_DATA"); if (config & (FSCACHE_OBJLIST_CONFIG_KEY | FSCACHE_OBJLIST_CONFIG_AUX)) @@ -193,7 +188,7 @@ static int fscache_objlist_show(struct seq_file *m, void *v) if ((unsigned long) v == 2) { seq_puts(m, "======== ======== ==== ===== === === === == =====" - " == == = =" + " == == == =" " | ================ == == ================"); if (config & (FSCACHE_OBJLIST_CONFIG_KEY | FSCACHE_OBJLIST_CONFIG_AUX)) @@ -216,10 +211,11 @@ static int fscache_objlist_show(struct seq_file *m, void *v) } \ } while(0) + cookie = obj->cookie; if (~config) { - FILTER(obj->cookie, + FILTER(cookie->def, COOKIE, NOCOOKIE); - FILTER(obj->state != FSCACHE_OBJECT_ACTIVE || + FILTER(fscache_object_is_active(obj) || obj->n_ops != 0 || obj->n_obj_ops != 0 || obj->flags || @@ -235,10 +231,10 @@ static int fscache_objlist_show(struct seq_file *m, void *v) } seq_printf(m, - "%8x %8x %s %5u %3u %3u %3u %2u %5u %2lx %2lx %1lx %1x | ", + "%8x %8x %s %5u %3u %3u %3u %2u %5u %2lx %2lx %2lx %1x | ", obj->debug_id, obj->parent ? obj->parent->debug_id : -1, - fscache_object_states_short[obj->state], + obj->state->short_name, obj->n_children, obj->n_ops, obj->n_obj_ops, @@ -250,48 +246,40 @@ static int fscache_objlist_show(struct seq_file *m, void *v) obj->flags, work_busy(&obj->work)); - no_cookie = true; - keylen = auxlen = 0; - if (obj->cookie) { - spin_lock(&obj->lock); - if (obj->cookie) { - switch (obj->cookie->def->type) { - case 0: - type = "IX"; - break; - case 1: - type = "DT"; - break; - default: - sprintf(_type, "%02u", - obj->cookie->def->type); - type = _type; - break; - } + if (fscache_use_cookie(obj)) { + uint16_t keylen = 0, auxlen = 0; - seq_printf(m, "%-16s %s %2lx %16p", - obj->cookie->def->name, - type, - obj->cookie->flags, - obj->cookie->netfs_data); - - if (obj->cookie->def->get_key && - config & FSCACHE_OBJLIST_CONFIG_KEY) - keylen = obj->cookie->def->get_key( - obj->cookie->netfs_data, - buf, 400); - - if (obj->cookie->def->get_aux && - config & FSCACHE_OBJLIST_CONFIG_AUX) - auxlen = obj->cookie->def->get_aux( - obj->cookie->netfs_data, - buf + keylen, 512 - keylen); - - no_cookie = false; + switch (cookie->def->type) { + case 0: + type = "IX"; + break; + case 1: + type = "DT"; + break; + default: + sprintf(_type, "%02u", cookie->def->type); + type = _type; + break; } - spin_unlock(&obj->lock); - if (!no_cookie && (keylen > 0 || auxlen > 0)) { + seq_printf(m, "%-16s %s %2lx %16p", + cookie->def->name, + type, + cookie->flags, + cookie->netfs_data); + + if (cookie->def->get_key && + config & FSCACHE_OBJLIST_CONFIG_KEY) + keylen = cookie->def->get_key(cookie->netfs_data, + buf, 400); + + if (cookie->def->get_aux && + config & FSCACHE_OBJLIST_CONFIG_AUX) + auxlen = cookie->def->get_aux(cookie->netfs_data, + buf + keylen, 512 - keylen); + fscache_unuse_cookie(obj); + + if (keylen > 0 || auxlen > 0) { seq_printf(m, " "); for (p = buf; keylen > 0; keylen--) seq_printf(m, "%02x", *p++); @@ -302,12 +290,11 @@ static int fscache_objlist_show(struct seq_file *m, void *v) seq_printf(m, "%02x", *p++); } } - } - if (no_cookie) - seq_printf(m, "<no_cookie>\n"); - else seq_printf(m, "\n"); + } else { + seq_printf(m, "<no_netfs>\n"); + } return 0; } diff --git a/fs/fscache/object.c b/fs/fscache/object.c index 50d41c180211..86d75a60b20c 100644 --- a/fs/fscache/object.c +++ b/fs/fscache/object.c @@ -15,52 +15,131 @@ #define FSCACHE_DEBUG_LEVEL COOKIE #include <linux/module.h> #include <linux/slab.h> +#include <linux/prefetch.h> #include "internal.h" -const char *fscache_object_states[FSCACHE_OBJECT__NSTATES] = { - [FSCACHE_OBJECT_INIT] = "OBJECT_INIT", - [FSCACHE_OBJECT_LOOKING_UP] = "OBJECT_LOOKING_UP", - [FSCACHE_OBJECT_CREATING] = "OBJECT_CREATING", - [FSCACHE_OBJECT_AVAILABLE] = "OBJECT_AVAILABLE", - [FSCACHE_OBJECT_ACTIVE] = "OBJECT_ACTIVE", - [FSCACHE_OBJECT_INVALIDATING] = "OBJECT_INVALIDATING", - [FSCACHE_OBJECT_UPDATING] = "OBJECT_UPDATING", - [FSCACHE_OBJECT_DYING] = "OBJECT_DYING", - [FSCACHE_OBJECT_LC_DYING] = "OBJECT_LC_DYING", - [FSCACHE_OBJECT_ABORT_INIT] = "OBJECT_ABORT_INIT", - [FSCACHE_OBJECT_RELEASING] = "OBJECT_RELEASING", - [FSCACHE_OBJECT_RECYCLING] = "OBJECT_RECYCLING", - [FSCACHE_OBJECT_WITHDRAWING] = "OBJECT_WITHDRAWING", - [FSCACHE_OBJECT_DEAD] = "OBJECT_DEAD", +static const struct fscache_state *fscache_abort_initialisation(struct fscache_object *, int); +static const struct fscache_state *fscache_kill_dependents(struct fscache_object *, int); +static const struct fscache_state *fscache_drop_object(struct fscache_object *, int); +static const struct fscache_state *fscache_initialise_object(struct fscache_object *, int); +static const struct fscache_state *fscache_invalidate_object(struct fscache_object *, int); +static const struct fscache_state *fscache_jumpstart_dependents(struct fscache_object *, int); +static const struct fscache_state *fscache_kill_object(struct fscache_object *, int); +static const struct fscache_state *fscache_lookup_failure(struct fscache_object *, int); +static const struct fscache_state *fscache_look_up_object(struct fscache_object *, int); +static const struct fscache_state *fscache_object_available(struct fscache_object *, int); +static const struct fscache_state *fscache_parent_ready(struct fscache_object *, int); +static const struct fscache_state *fscache_update_object(struct fscache_object *, int); + +#define __STATE_NAME(n) fscache_osm_##n +#define STATE(n) (&__STATE_NAME(n)) + +/* + * Define a work state. Work states are execution states. No event processing + * is performed by them. The function attached to a work state returns a + * pointer indicating the next state to which the state machine should + * transition. Returning NO_TRANSIT repeats the current state, but goes back + * to the scheduler first. + */ +#define WORK_STATE(n, sn, f) \ + const struct fscache_state __STATE_NAME(n) = { \ + .name = #n, \ + .short_name = sn, \ + .work = f \ + } + +/* + * Returns from work states. + */ +#define transit_to(state) ({ prefetch(&STATE(state)->work); STATE(state); }) + +#define NO_TRANSIT ((struct fscache_state *)NULL) + +/* + * Define a wait state. Wait states are event processing states. No execution + * is performed by them. Wait states are just tables of "if event X occurs, + * clear it and transition to state Y". The dispatcher returns to the + * scheduler if none of the events in which the wait state has an interest are + * currently pending. + */ +#define WAIT_STATE(n, sn, ...) \ + const struct fscache_state __STATE_NAME(n) = { \ + .name = #n, \ + .short_name = sn, \ + .work = NULL, \ + .transitions = { __VA_ARGS__, { 0, NULL } } \ + } + +#define TRANSIT_TO(state, emask) \ + { .events = (emask), .transit_to = STATE(state) } + +/* + * The object state machine. + */ +static WORK_STATE(INIT_OBJECT, "INIT", fscache_initialise_object); +static WORK_STATE(PARENT_READY, "PRDY", fscache_parent_ready); +static WORK_STATE(ABORT_INIT, "ABRT", fscache_abort_initialisation); +static WORK_STATE(LOOK_UP_OBJECT, "LOOK", fscache_look_up_object); +static WORK_STATE(CREATE_OBJECT, "CRTO", fscache_look_up_object); +static WORK_STATE(OBJECT_AVAILABLE, "AVBL", fscache_object_available); +static WORK_STATE(JUMPSTART_DEPS, "JUMP", fscache_jumpstart_dependents); + +static WORK_STATE(INVALIDATE_OBJECT, "INVL", fscache_invalidate_object); +static WORK_STATE(UPDATE_OBJECT, "UPDT", fscache_update_object); + +static WORK_STATE(LOOKUP_FAILURE, "LCFL", fscache_lookup_failure); +static WORK_STATE(KILL_OBJECT, "KILL", fscache_kill_object); +static WORK_STATE(KILL_DEPENDENTS, "KDEP", fscache_kill_dependents); +static WORK_STATE(DROP_OBJECT, "DROP", fscache_drop_object); +static WORK_STATE(OBJECT_DEAD, "DEAD", (void*)2UL); + +static WAIT_STATE(WAIT_FOR_INIT, "?INI", + TRANSIT_TO(INIT_OBJECT, 1 << FSCACHE_OBJECT_EV_NEW_CHILD)); + +static WAIT_STATE(WAIT_FOR_PARENT, "?PRN", + TRANSIT_TO(PARENT_READY, 1 << FSCACHE_OBJECT_EV_PARENT_READY)); + +static WAIT_STATE(WAIT_FOR_CMD, "?CMD", + TRANSIT_TO(INVALIDATE_OBJECT, 1 << FSCACHE_OBJECT_EV_INVALIDATE), + TRANSIT_TO(UPDATE_OBJECT, 1 << FSCACHE_OBJECT_EV_UPDATE), + TRANSIT_TO(JUMPSTART_DEPS, 1 << FSCACHE_OBJECT_EV_NEW_CHILD)); + +static WAIT_STATE(WAIT_FOR_CLEARANCE, "?CLR", + TRANSIT_TO(KILL_OBJECT, 1 << FSCACHE_OBJECT_EV_CLEARED)); + +/* + * Out-of-band event transition tables. These are for handling unexpected + * events, such as an I/O error. If an OOB event occurs, the state machine + * clears and disables the event and forces a transition to the nominated work + * state (acurrently executing work states will complete first). + * + * In such a situation, object->state remembers the state the machine should + * have been in/gone to and returning NO_TRANSIT returns to that. + */ +static const struct fscache_transition fscache_osm_init_oob[] = { + TRANSIT_TO(ABORT_INIT, + (1 << FSCACHE_OBJECT_EV_ERROR) | + (1 << FSCACHE_OBJECT_EV_KILL)), + { 0, NULL } +}; + +static const struct fscache_transition fscache_osm_lookup_oob[] = { + TRANSIT_TO(LOOKUP_FAILURE, + (1 << FSCACHE_OBJECT_EV_ERROR) | + (1 << FSCACHE_OBJECT_EV_KILL)), + { 0, NULL } }; -EXPORT_SYMBOL(fscache_object_states); - -const char fscache_object_states_short[FSCACHE_OBJECT__NSTATES][5] = { - [FSCACHE_OBJECT_INIT] = "INIT", - [FSCACHE_OBJECT_LOOKING_UP] = "LOOK", - [FSCACHE_OBJECT_CREATING] = "CRTN", - [FSCACHE_OBJECT_AVAILABLE] = "AVBL", - [FSCACHE_OBJECT_ACTIVE] = "ACTV", - [FSCACHE_OBJECT_INVALIDATING] = "INVL", - [FSCACHE_OBJECT_UPDATING] = "UPDT", - [FSCACHE_OBJECT_DYING] = "DYNG", - [FSCACHE_OBJECT_LC_DYING] = "LCDY", - [FSCACHE_OBJECT_ABORT_INIT] = "ABTI", - [FSCACHE_OBJECT_RELEASING] = "RELS", - [FSCACHE_OBJECT_RECYCLING] = "RCYC", - [FSCACHE_OBJECT_WITHDRAWING] = "WTHD", - [FSCACHE_OBJECT_DEAD] = "DEAD", + +static const struct fscache_transition fscache_osm_run_oob[] = { + TRANSIT_TO(KILL_OBJECT, + (1 << FSCACHE_OBJECT_EV_ERROR) | + (1 << FSCACHE_OBJECT_EV_KILL)), + { 0, NULL } }; static int fscache_get_object(struct fscache_object *); static void fscache_put_object(struct fscache_object *); -static void fscache_initialise_object(struct fscache_object *); -static void fscache_lookup_object(struct fscache_object *); -static void fscache_object_available(struct fscache_object *); -static void fscache_invalidate_object(struct fscache_object *); -static void fscache_release_object(struct fscache_object *); -static void fscache_withdraw_object(struct fscache_object *); -static void fscache_enqueue_dependents(struct fscache_object *); +static bool fscache_enqueue_dependents(struct fscache_object *, int); static void fscache_dequeue_object(struct fscache_object *); /* @@ -75,295 +154,116 @@ static inline void fscache_done_parent_op(struct fscache_object *object) object->debug_id, parent->debug_id, parent->n_ops); spin_lock_nested(&parent->lock, 1); - parent->n_ops--; parent->n_obj_ops--; + parent->n_ops--; if (parent->n_ops == 0) fscache_raise_event(parent, FSCACHE_OBJECT_EV_CLEARED); spin_unlock(&parent->lock); } /* - * Notify netfs of invalidation completion. + * Object state machine dispatcher. */ -static inline void fscache_invalidation_complete(struct fscache_cookie *cookie) +static void fscache_object_sm_dispatcher(struct fscache_object *object) { - if (test_and_clear_bit(FSCACHE_COOKIE_INVALIDATING, &cookie->flags)) - wake_up_bit(&cookie->flags, FSCACHE_COOKIE_INVALIDATING); -} - -/* - * process events that have been sent to an object's state machine - * - initiates parent lookup - * - does object lookup - * - does object creation - * - does object recycling and retirement - * - does object withdrawal - */ -static void fscache_object_state_machine(struct fscache_object *object) -{ - enum fscache_object_state new_state; - struct fscache_cookie *cookie; - int event; + const struct fscache_transition *t; + const struct fscache_state *state, *new_state; + unsigned long events, event_mask; + int event = -1; ASSERT(object != NULL); _enter("{OBJ%x,%s,%lx}", - object->debug_id, fscache_object_states[object->state], - object->events); - - switch (object->state) { - /* wait for the parent object to become ready */ - case FSCACHE_OBJECT_INIT: - object->event_mask = - FSCACHE_OBJECT_EVENTS_MASK & - ~(1 << FSCACHE_OBJECT_EV_CLEARED); - fscache_initialise_object(object); - goto done; - - /* look up the object metadata on disk */ - case FSCACHE_OBJECT_LOOKING_UP: - fscache_lookup_object(object); - goto lookup_transit; - - /* create the object metadata on disk */ - case FSCACHE_OBJECT_CREATING: - fscache_lookup_object(object); - goto lookup_transit; - - /* handle an object becoming available; start pending - * operations and queue dependent operations for processing */ - case FSCACHE_OBJECT_AVAILABLE: - fscache_object_available(object); - goto active_transit; - - /* normal running state */ - case FSCACHE_OBJECT_ACTIVE: - goto active_transit; - - /* Invalidate an object on disk */ - case FSCACHE_OBJECT_INVALIDATING: - clear_bit(FSCACHE_OBJECT_EV_INVALIDATE, &object->events); - fscache_stat(&fscache_n_invalidates_run); - fscache_stat(&fscache_n_cop_invalidate_object); - fscache_invalidate_object(object); - fscache_stat_d(&fscache_n_cop_invalidate_object); - fscache_raise_event(object, FSCACHE_OBJECT_EV_UPDATE); - goto active_transit; - - /* update the object metadata on disk */ - case FSCACHE_OBJECT_UPDATING: - clear_bit(FSCACHE_OBJECT_EV_UPDATE, &object->events); - fscache_stat(&fscache_n_updates_run); - fscache_stat(&fscache_n_cop_update_object); - object->cache->ops->update_object(object); - fscache_stat_d(&fscache_n_cop_update_object); - goto active_transit; - - /* handle an object dying during lookup or creation */ - case FSCACHE_OBJECT_LC_DYING: - object->event_mask &= ~(1 << FSCACHE_OBJECT_EV_UPDATE); - fscache_stat(&fscache_n_cop_lookup_complete); - object->cache->ops->lookup_complete(object); - fscache_stat_d(&fscache_n_cop_lookup_complete); - - spin_lock(&object->lock); - object->state = FSCACHE_OBJECT_DYING; - cookie = object->cookie; - if (cookie) { - if (test_and_clear_bit(FSCACHE_COOKIE_LOOKING_UP, - &cookie->flags)) - wake_up_bit(&cookie->flags, - FSCACHE_COOKIE_LOOKING_UP); - if (test_and_clear_bit(FSCACHE_COOKIE_CREATING, - &cookie->flags)) - wake_up_bit(&cookie->flags, - FSCACHE_COOKIE_CREATING); + object->debug_id, object->state->name, object->events); + + event_mask = object->event_mask; +restart: + object->event_mask = 0; /* Mask normal event handling */ + state = object->state; +restart_masked: + events = object->events; + + /* Handle any out-of-band events (typically an error) */ + if (events & object->oob_event_mask) { + _debug("{OBJ%x} oob %lx", + object->debug_id, events & object->oob_event_mask); + for (t = object->oob_table; t->events; t++) { + if (events & t->events) { + state = t->transit_to; + ASSERT(state->work != NULL); + event = fls(events & t->events) - 1; + __clear_bit(event, &object->oob_event_mask); + clear_bit(event, &object->events); + goto execute_work_state; + } } - spin_unlock(&object->lock); + } - fscache_done_parent_op(object); + /* Wait states are just transition tables */ + if (!state->work) { + if (events & event_mask) { + for (t = state->transitions; t->events; t++) { + if (events & t->events) { + new_state = t->transit_to; + event = fls(events & t->events) - 1; + clear_bit(event, &object->events); + _debug("{OBJ%x} ev %d: %s -> %s", + object->debug_id, event, + state->name, new_state->name); + object->state = state = new_state; + goto execute_work_state; + } + } - /* wait for completion of all active operations on this object - * and the death of all child objects of this object */ - case FSCACHE_OBJECT_DYING: - dying: - clear_bit(FSCACHE_OBJECT_EV_CLEARED, &object->events); - spin_lock(&object->lock); - _debug("dying OBJ%x {%d,%d}", - object->debug_id, object->n_ops, object->n_children); - if (object->n_ops == 0 && object->n_children == 0) { - object->event_mask &= - ~(1 << FSCACHE_OBJECT_EV_CLEARED); - object->event_mask |= - (1 << FSCACHE_OBJECT_EV_WITHDRAW) | - (1 << FSCACHE_OBJECT_EV_RETIRE) | - (1 << FSCACHE_OBJECT_EV_RELEASE) | - (1 << FSCACHE_OBJECT_EV_ERROR); - } else { - object->event_mask &= - ~((1 << FSCACHE_OBJECT_EV_WITHDRAW) | - (1 << FSCACHE_OBJECT_EV_RETIRE) | - (1 << FSCACHE_OBJECT_EV_RELEASE) | - (1 << FSCACHE_OBJECT_EV_ERROR)); - object->event_mask |= - 1 << FSCACHE_OBJECT_EV_CLEARED; + /* The event mask didn't include all the tabled bits */ + BUG(); } - spin_unlock(&object->lock); - fscache_enqueue_dependents(object); - fscache_start_operations(object); - goto terminal_transit; - - /* handle an abort during initialisation */ - case FSCACHE_OBJECT_ABORT_INIT: - _debug("handle abort init %lx", object->events); - object->event_mask &= ~(1 << FSCACHE_OBJECT_EV_UPDATE); - - spin_lock(&object->lock); - fscache_dequeue_object(object); - - object->state = FSCACHE_OBJECT_DYING; - if (test_and_clear_bit(FSCACHE_COOKIE_CREATING, - &object->cookie->flags)) - wake_up_bit(&object->cookie->flags, - FSCACHE_COOKIE_CREATING); - spin_unlock(&object->lock); - goto dying; - - /* handle the netfs releasing an object and possibly marking it - * obsolete too */ - case FSCACHE_OBJECT_RELEASING: - case FSCACHE_OBJECT_RECYCLING: - object->event_mask &= - ~((1 << FSCACHE_OBJECT_EV_WITHDRAW) | - (1 << FSCACHE_OBJECT_EV_RETIRE) | - (1 << FSCACHE_OBJECT_EV_RELEASE) | - (1 << FSCACHE_OBJECT_EV_ERROR)); - fscache_release_object(object); - spin_lock(&object->lock); - object->state = FSCACHE_OBJECT_DEAD; - spin_unlock(&object->lock); - fscache_stat(&fscache_n_object_dead); - goto terminal_transit; - - /* handle the parent cache of this object being withdrawn from - * active service */ - case FSCACHE_OBJECT_WITHDRAWING: - object->event_mask &= - ~((1 << FSCACHE_OBJECT_EV_WITHDRAW) | - (1 << FSCACHE_OBJECT_EV_RETIRE) | - (1 << FSCACHE_OBJECT_EV_RELEASE) | - (1 << FSCACHE_OBJECT_EV_ERROR)); - fscache_withdraw_object(object); - spin_lock(&object->lock); - object->state = FSCACHE_OBJECT_DEAD; - spin_unlock(&object->lock); - fscache_stat(&fscache_n_object_dead); - goto terminal_transit; - - /* complain about the object being woken up once it is - * deceased */ - case FSCACHE_OBJECT_DEAD: - printk(KERN_ERR "FS-Cache:" - " Unexpected event in dead state %lx\n", - object->events & object->event_mask); - BUG(); - - default: - printk(KERN_ERR "FS-Cache: Unknown object state %u\n", - object->state); - BUG(); - } - - /* determine the transition from a lookup state */ -lookup_transit: - event = fls(object->events & object->event_mask) - 1; - switch (event) { - case FSCACHE_OBJECT_EV_WITHDRAW: - case FSCACHE_OBJECT_EV_RETIRE: - case FSCACHE_OBJECT_EV_RELEASE: - case FSCACHE_OBJECT_EV_ERROR: - new_state = FSCACHE_OBJECT_LC_DYING; - goto change_state; - case FSCACHE_OBJECT_EV_INVALIDATE: - new_state = FSCACHE_OBJECT_INVALIDATING; - goto change_state; - case FSCACHE_OBJECT_EV_REQUEUE: - goto done; - case -1: - goto done; /* sleep until event */ - default: - goto unsupported_event; + /* Randomly woke up */ + goto unmask_events; } - /* determine the transition from an active state */ -active_transit: - event = fls(object->events & object->event_mask) - 1; - switch (event) { - case FSCACHE_OBJECT_EV_WITHDRAW: - case FSCACHE_OBJECT_EV_RETIRE: - case FSCACHE_OBJECT_EV_RELEASE: - case FSCACHE_OBJECT_EV_ERROR: - new_state = FSCACHE_OBJECT_DYING; - goto change_state; - case FSCACHE_OBJECT_EV_INVALIDATE: - new_state = FSCACHE_OBJECT_INVALIDATING; - goto change_state; - case FSCACHE_OBJECT_EV_UPDATE: - new_state = FSCACHE_OBJECT_UPDATING; - goto change_state; - case -1: - new_state = FSCACHE_OBJECT_ACTIVE; - goto change_state; /* sleep until event */ - default: - goto unsupported_event; - } +execute_work_state: + _debug("{OBJ%x} exec %s", object->debug_id, state->name); - /* determine the transition from a terminal state */ -terminal_transit: - event = fls(object->events & object->event_mask) - 1; - switch (event) { - case FSCACHE_OBJECT_EV_WITHDRAW: - new_state = FSCACHE_OBJECT_WITHDRAWING; - goto change_state; - case FSCACHE_OBJECT_EV_RETIRE: - new_state = FSCACHE_OBJECT_RECYCLING; - goto change_state; - case FSCACHE_OBJECT_EV_RELEASE: - new_state = FSCACHE_OBJECT_RELEASING; - goto change_state; - case FSCACHE_OBJECT_EV_ERROR: - new_state = FSCACHE_OBJECT_WITHDRAWING; - goto change_state; - case FSCACHE_OBJECT_EV_CLEARED: - new_state = FSCACHE_OBJECT_DYING; - goto change_state; - case -1: - goto done; /* sleep until event */ - default: - goto unsupported_event; + new_state = state->work(object, event); + event = -1; + if (new_state == NO_TRANSIT) { + _debug("{OBJ%x} %s notrans", object->debug_id, state->name); + fscache_enqueue_object(object); + event_mask = object->oob_event_mask; + goto unmask_events; } -change_state: - spin_lock(&object->lock); - object->state = new_state; - spin_unlock(&object->lock); + _debug("{OBJ%x} %s -> %s", + object->debug_id, state->name, new_state->name); + object->state = state = new_state; -done: - _leave(" [->%s]", fscache_object_states[object->state]); - return; + if (state->work) { + if (unlikely(state->work == ((void *)2UL))) { + _leave(" [dead]"); + return; + } + goto restart_masked; + } -unsupported_event: - printk(KERN_ERR "FS-Cache:" - " Unsupported event %d [%lx/%lx] in state %s\n", - event, object->events, object->event_mask, - fscache_object_states[object->state]); - BUG(); + /* Transited to wait state */ + event_mask = object->oob_event_mask; + for (t = state->transitions; t->events; t++) + event_mask |= t->events; + +unmask_events: + object->event_mask = event_mask; + smp_mb(); + events = object->events; + if (events & event_mask) + goto restart; + _leave(" [msk %lx]", event_mask); } /* * execute an object */ -void fscache_object_work_func(struct work_struct *work) +static void fscache_object_work_func(struct work_struct *work) { struct fscache_object *object = container_of(work, struct fscache_object, work); @@ -372,14 +272,70 @@ void fscache_object_work_func(struct work_struct *work) _enter("{OBJ%x}", object->debug_id); start = jiffies; - fscache_object_state_machine(object); + fscache_object_sm_dispatcher(object); fscache_hist(fscache_objs_histogram, start); - if (object->events & object->event_mask) - fscache_enqueue_object(object); - clear_bit(FSCACHE_OBJECT_EV_REQUEUE, &object->events); fscache_put_object(object); } -EXPORT_SYMBOL(fscache_object_work_func); + +/** + * fscache_object_init - Initialise a cache object description + * @object: Object description + * @cookie: Cookie object will be attached to + * @cache: Cache in which backing object will be found + * + * Initialise a cache object description to its basic values. + * + * See Documentation/filesystems/caching/backend-api.txt for a complete + * description. + */ +void fscache_object_init(struct fscache_object *object, + struct fscache_cookie *cookie, + struct fscache_cache *cache) +{ + const struct fscache_transition *t; + + atomic_inc(&cache->object_count); + + object->state = STATE(WAIT_FOR_INIT); + object->oob_table = fscache_osm_init_oob; + object->flags = 1 << FSCACHE_OBJECT_IS_LIVE; + spin_lock_init(&object->lock); + INIT_LIST_HEAD(&object->cache_link); + INIT_HLIST_NODE(&object->cookie_link); + INIT_WORK(&object->work, fscache_object_work_func); + INIT_LIST_HEAD(&object->dependents); + INIT_LIST_HEAD(&object->dep_link); + INIT_LIST_HEAD(&object->pending_ops); + object->n_children = 0; + object->n_ops = object->n_in_progress = object->n_exclusive = 0; + object->events = 0; + object->store_limit = 0; + object->store_limit_l = 0; + object->cache = cache; + object->cookie = cookie; + object->parent = NULL; + + object->oob_event_mask = 0; + for (t = object->oob_table; t->events; t++) + object->oob_event_mask |= t->events; + object->event_mask = object->oob_event_mask; + for (t = object->state->transitions; t->events; t++) + object->event_mask |= t->events; +} +EXPORT_SYMBOL(fscache_object_init); + +/* + * Abort object initialisation before we start it. + */ +static const struct fscache_state *fscache_abort_initialisation(struct fscache_object *object, + int event) +{ + _enter("{OBJ%x},%d", object->debug_id, event); + + object->oob_event_mask = 0; + fscache_dequeue_object(object); + return transit_to(KILL_OBJECT); +} /* * initialise an object @@ -387,130 +343,136 @@ EXPORT_SYMBOL(fscache_object_work_func); * immediately to do a creation * - we may need to start the process of creating a parent and we need to wait * for the parent's lookup and creation to complete if it's not there yet - * - an object's cookie is pinned until we clear FSCACHE_COOKIE_CREATING on the - * leaf-most cookies of the object and all its children */ -static void fscache_initialise_object(struct fscache_object *object) +static const struct fscache_state *fscache_initialise_object(struct fscache_object *object, + int event) { struct fscache_object *parent; + bool success; - _enter(""); - ASSERT(object->cookie != NULL); - ASSERT(object->cookie->parent != NULL); - - if (object->events & ((1 << FSCACHE_OBJECT_EV_ERROR) | - (1 << FSCACHE_OBJECT_EV_RELEASE) | - (1 << FSCACHE_OBJECT_EV_RETIRE) | - (1 << FSCACHE_OBJECT_EV_WITHDRAW))) { - _debug("abort init %lx", object->events); - spin_lock(&object->lock); - object->state = FSCACHE_OBJECT_ABORT_INIT; - spin_unlock(&object->lock); - return; - } + _enter("{OBJ%x},%d", object->debug_id, event); - spin_lock(&object->cookie->lock); - spin_lock_nested(&object->cookie->parent->lock, 1); + ASSERT(list_empty(&object->dep_link)); parent = object->parent; if (!parent) { - _debug("no parent"); - set_bit(FSCACHE_OBJECT_EV_WITHDRAW, &object->events); - } else { - spin_lock(&object->lock); - spin_lock_nested(&parent->lock, 1); - _debug("parent %s", fscache_object_states[parent->state]); - - if (parent->state >= FSCACHE_OBJECT_DYING) { - _debug("bad parent"); - set_bit(FSCACHE_OBJECT_EV_WITHDRAW, &object->events); - } else if (parent->state < FSCACHE_OBJECT_AVAILABLE) { - _debug("wait"); - - /* we may get woken up in this state by child objects - * binding on to us, so we need to make sure we don't - * add ourself to the list multiple times */ - if (list_empty(&object->dep_link)) { - fscache_stat(&fscache_n_cop_grab_object); - object->cache->ops->grab_object(object); - fscache_stat_d(&fscache_n_cop_grab_object); - list_add(&object->dep_link, - &parent->dependents); - - /* fscache_acquire_non_index_cookie() uses this - * to wake the chain up */ - if (parent->state == FSCACHE_OBJECT_INIT) - fscache_enqueue_object(parent); - } - } else { - _debug("go"); - parent->n_ops++; - parent->n_obj_ops++; - object->lookup_jif = jiffies; - object->state = FSCACHE_OBJECT_LOOKING_UP; - set_bit(FSCACHE_OBJECT_EV_REQUEUE, &object->events); - } + _leave(" [no parent]"); + return transit_to(DROP_OBJECT); + } - spin_unlock(&parent->lock); - spin_unlock(&object->lock); + _debug("parent: %s of:%lx", parent->state->name, parent->flags); + + if (fscache_object_is_dying(parent)) { + _leave(" [bad parent]"); + return transit_to(DROP_OBJECT); } - spin_unlock(&object->cookie->parent->lock); - spin_unlock(&object->cookie->lock); + if (fscache_object_is_available(parent)) { + _leave(" [ready]"); + return transit_to(PARENT_READY); + } + + _debug("wait"); + + spin_lock(&parent->lock); + fscache_stat(&fscache_n_cop_grab_object); + success = false; + if (fscache_object_is_live(parent) && + object->cache->ops->grab_object(object)) { + list_add(&object->dep_link, &parent->dependents); + success = true; + } + fscache_stat_d(&fscache_n_cop_grab_object); + spin_unlock(&parent->lock); + if (!success) { + _leave(" [grab failed]"); + return transit_to(DROP_OBJECT); + } + + /* fscache_acquire_non_index_cookie() uses this + * to wake the chain up */ + fscache_raise_event(parent, FSCACHE_OBJECT_EV_NEW_CHILD); + _leave(" [wait]"); + return transit_to(WAIT_FOR_PARENT); +} + +/* + * Once the parent object is ready, we should kick off our lookup op. + */ +static const struct fscache_state *fscache_parent_ready(struct fscache_object *object, + int event) +{ + struct fscache_object *parent = object->parent; + + _enter("{OBJ%x},%d", object->debug_id, event); + + ASSERT(parent != NULL); + + spin_lock(&parent->lock); + parent->n_ops++; + parent->n_obj_ops++; + object->lookup_jif = jiffies; + spin_unlock(&parent->lock); + _leave(""); + return transit_to(LOOK_UP_OBJECT); } /* * look an object up in the cache from which it was allocated * - we hold an "access lock" on the parent object, so the parent object cannot * be withdrawn by either party till we've finished - * - an object's cookie is pinned until we clear FSCACHE_COOKIE_CREATING on the - * leaf-most cookies of the object and all its children */ -static void fscache_lookup_object(struct fscache_object *object) +static const struct fscache_state *fscache_look_up_object(struct fscache_object *object, + int event) { struct fscache_cookie *cookie = object->cookie; - struct fscache_object *parent; + struct fscache_object *parent = object->parent; int ret; - _enter(""); + _enter("{OBJ%x},%d", object->debug_id, event); + + object->oob_table = fscache_osm_lookup_oob; - parent = object->parent; ASSERT(parent != NULL); ASSERTCMP(parent->n_ops, >, 0); ASSERTCMP(parent->n_obj_ops, >, 0); /* make sure the parent is still available */ - ASSERTCMP(parent->state, >=, FSCACHE_OBJECT_AVAILABLE); - - if (parent->state >= FSCACHE_OBJECT_DYING || - test_bit(FSCACHE_IOERROR, &object->cache->flags)) { - _debug("unavailable"); - set_bit(FSCACHE_OBJECT_EV_WITHDRAW, &object->events); - _leave(""); - return; + ASSERT(fscache_object_is_available(parent)); + + if (fscache_object_is_dying(parent) || + test_bit(FSCACHE_IOERROR, &object->cache->flags) || + !fscache_use_cookie(object)) { + _leave(" [unavailable]"); + return transit_to(LOOKUP_FAILURE); } - _debug("LOOKUP \"%s/%s\" in \"%s\"", - parent->cookie->def->name, cookie->def->name, - object->cache->tag->name); + _debug("LOOKUP \"%s\" in \"%s\"", + cookie->def->name, object->cache->tag->name); fscache_stat(&fscache_n_object_lookups); fscache_stat(&fscache_n_cop_lookup_object); ret = object->cache->ops->lookup_object(object); fscache_stat_d(&fscache_n_cop_lookup_object); - if (test_bit(FSCACHE_OBJECT_EV_ERROR, &object->events)) - set_bit(FSCACHE_COOKIE_UNAVAILABLE, &cookie->flags); + fscache_unuse_cookie(object); if (ret == -ETIMEDOUT) { /* probably stuck behind another object, so move this one to * the back of the queue */ fscache_stat(&fscache_n_object_lookups_timed_out); - set_bit(FSCACHE_OBJECT_EV_REQUEUE, &object->events); + _leave(" [timeout]"); + return NO_TRANSIT; } - _leave(""); + if (ret < 0) { + _leave(" [error]"); + return transit_to(LOOKUP_FAILURE); + } + + _leave(" [ok]"); + return transit_to(OBJECT_AVAILABLE); } /** @@ -524,32 +486,20 @@ void fscache_object_lookup_negative(struct fscache_object *object) { struct fscache_cookie *cookie = object->cookie; - _enter("{OBJ%x,%s}", - object->debug_id, fscache_object_states[object->state]); + _enter("{OBJ%x,%s}", object->debug_id, object->state->name); - spin_lock(&object->lock); - if (object->state == FSCACHE_OBJECT_LOOKING_UP) { + if (!test_and_set_bit(FSCACHE_OBJECT_IS_LOOKED_UP, &object->flags)) { fscache_stat(&fscache_n_object_lookups_negative); - /* transit here to allow write requests to begin stacking up - * and read requests to begin returning ENODATA */ - object->state = FSCACHE_OBJECT_CREATING; - spin_unlock(&object->lock); - - set_bit(FSCACHE_COOKIE_PENDING_FILL, &cookie->flags); + /* Allow write requests to begin stacking up and read requests to begin + * returning ENODATA. + */ set_bit(FSCACHE_COOKIE_NO_DATA_YET, &cookie->flags); _debug("wake up lookup %p", &cookie->flags); - smp_mb__before_clear_bit(); - clear_bit(FSCACHE_COOKIE_LOOKING_UP, &cookie->flags); - smp_mb__after_clear_bit(); + clear_bit_unlock(FSCACHE_COOKIE_LOOKING_UP, &cookie->flags); wake_up_bit(&cookie->flags, FSCACHE_COOKIE_LOOKING_UP); - set_bit(FSCACHE_OBJECT_EV_REQUEUE, &object->events); - } else { - ASSERTCMP(object->state, ==, FSCACHE_OBJECT_CREATING); - spin_unlock(&object->lock); } - _leave(""); } EXPORT_SYMBOL(fscache_object_lookup_negative); @@ -568,38 +518,26 @@ void fscache_obtained_object(struct fscache_object *object) { struct fscache_cookie *cookie = object->cookie; - _enter("{OBJ%x,%s}", - object->debug_id, fscache_object_states[object->state]); + _enter("{OBJ%x,%s}", object->debug_id, object->state->name); /* if we were still looking up, then we must have a positive lookup * result, in which case there may be data available */ - spin_lock(&object->lock); - if (object->state == FSCACHE_OBJECT_LOOKING_UP) { + if (!test_and_set_bit(FSCACHE_OBJECT_IS_LOOKED_UP, &object->flags)) { fscache_stat(&fscache_n_object_lookups_positive); - clear_bit(FSCACHE_COOKIE_NO_DATA_YET, &cookie->flags); + /* We do (presumably) have data */ + clear_bit_unlock(FSCACHE_COOKIE_NO_DATA_YET, &cookie->flags); - object->state = FSCACHE_OBJECT_AVAILABLE; - spin_unlock(&object->lock); - - smp_mb__before_clear_bit(); - clear_bit(FSCACHE_COOKIE_LOOKING_UP, &cookie->flags); - smp_mb__after_clear_bit(); + /* Allow write requests to begin stacking up and read requests + * to begin shovelling data. + */ + clear_bit_unlock(FSCACHE_COOKIE_LOOKING_UP, &cookie->flags); wake_up_bit(&cookie->flags, FSCACHE_COOKIE_LOOKING_UP); - set_bit(FSCACHE_OBJECT_EV_REQUEUE, &object->events); } else { - ASSERTCMP(object->state, ==, FSCACHE_OBJECT_CREATING); fscache_stat(&fscache_n_object_created); - - object->state = FSCACHE_OBJECT_AVAILABLE; - spin_unlock(&object->lock); - set_bit(FSCACHE_OBJECT_EV_REQUEUE, &object->events); - smp_wmb(); } - if (test_and_clear_bit(FSCACHE_COOKIE_CREATING, &cookie->flags)) - wake_up_bit(&cookie->flags, FSCACHE_COOKIE_CREATING); - + set_bit(FSCACHE_OBJECT_IS_AVAILABLE, &object->flags); _leave(""); } EXPORT_SYMBOL(fscache_obtained_object); @@ -607,15 +545,14 @@ EXPORT_SYMBOL(fscache_obtained_object); /* * handle an object that has just become available */ -static void fscache_object_available(struct fscache_object *object) +static const struct fscache_state *fscache_object_available(struct fscache_object *object, + int event) { - _enter("{OBJ%x}", object->debug_id); + _enter("{OBJ%x},%d", object->debug_id, event); - spin_lock(&object->lock); + object->oob_table = fscache_osm_run_oob; - if (object->cookie && - test_and_clear_bit(FSCACHE_COOKIE_CREATING, &object->cookie->flags)) - wake_up_bit(&object->cookie->flags, FSCACHE_COOKIE_CREATING); + spin_lock(&object->lock); fscache_done_parent_op(object); if (object->n_in_progress == 0) { @@ -631,130 +568,158 @@ static void fscache_object_available(struct fscache_object *object) fscache_stat(&fscache_n_cop_lookup_complete); object->cache->ops->lookup_complete(object); fscache_stat_d(&fscache_n_cop_lookup_complete); - fscache_enqueue_dependents(object); fscache_hist(fscache_obj_instantiate_histogram, object->lookup_jif); fscache_stat(&fscache_n_object_avail); _leave(""); + return transit_to(JUMPSTART_DEPS); } /* - * drop an object's attachments + * Wake up this object's dependent objects now that we've become available. */ -static void fscache_drop_object(struct fscache_object *object) +static const struct fscache_state *fscache_jumpstart_dependents(struct fscache_object *object, + int event) { - struct fscache_object *parent = object->parent; - struct fscache_cache *cache = object->cache; + _enter("{OBJ%x},%d", object->debug_id, event); - _enter("{OBJ%x,%d}", object->debug_id, object->n_children); + if (!fscache_enqueue_dependents(object, FSCACHE_OBJECT_EV_PARENT_READY)) + return NO_TRANSIT; /* Not finished; requeue */ + return transit_to(WAIT_FOR_CMD); +} - ASSERTCMP(object->cookie, ==, NULL); - ASSERT(hlist_unhashed(&object->cookie_link)); +/* + * Handle lookup or creation failute. + */ +static const struct fscache_state *fscache_lookup_failure(struct fscache_object *object, + int event) +{ + struct fscache_cookie *cookie; - spin_lock(&cache->object_list_lock); - list_del_init(&object->cache_link); - spin_unlock(&cache->object_list_lock); + _enter("{OBJ%x},%d", object->debug_id, event); - fscache_stat(&fscache_n_cop_drop_object); - cache->ops->drop_object(object); - fscache_stat_d(&fscache_n_cop_drop_object); + object->oob_event_mask = 0; - if (parent) { - _debug("release parent OBJ%x {%d}", - parent->debug_id, parent->n_children); + fscache_stat(&fscache_n_cop_lookup_complete); + object->cache->ops->lookup_complete(object); + fscache_stat_d(&fscache_n_cop_lookup_complete); - spin_lock(&parent->lock); - parent->n_children--; - if (parent->n_children == 0) - fscache_raise_event(parent, FSCACHE_OBJECT_EV_CLEARED); - spin_unlock(&parent->lock); - object->parent = NULL; + cookie = object->cookie; + set_bit(FSCACHE_COOKIE_UNAVAILABLE, &cookie->flags); + if (test_and_clear_bit(FSCACHE_COOKIE_LOOKING_UP, &cookie->flags)) + wake_up_bit(&cookie->flags, FSCACHE_COOKIE_LOOKING_UP); + + fscache_done_parent_op(object); + return transit_to(KILL_OBJECT); +} + +/* + * Wait for completion of all active operations on this object and the death of + * all child objects of this object. + */ +static const struct fscache_state *fscache_kill_object(struct fscache_object *object, + int event) +{ + _enter("{OBJ%x,%d,%d},%d", + object->debug_id, object->n_ops, object->n_children, event); + + clear_bit(FSCACHE_OBJECT_IS_LIVE, &object->flags); + object->oob_event_mask = 0; + + if (list_empty(&object->dependents) && + object->n_ops == 0 && + object->n_children == 0) + return transit_to(DROP_OBJECT); + + if (object->n_in_progress == 0) { + spin_lock(&object->lock); + if (object->n_ops > 0 && object->n_in_progress == 0) + fscache_start_operations(object); + spin_unlock(&object->lock); } - /* this just shifts the object release to the work processor */ - fscache_put_object(object); + if (!list_empty(&object->dependents)) + return transit_to(KILL_DEPENDENTS); - _leave(""); + return transit_to(WAIT_FOR_CLEARANCE); } /* - * release or recycle an object that the netfs has discarded + * Kill dependent objects. */ -static void fscache_release_object(struct fscache_object *object) +static const struct fscache_state *fscache_kill_dependents(struct fscache_object *object, + int event) { - _enter(""); + _enter("{OBJ%x},%d", object->debug_id, event); - fscache_drop_object(object); + if (!fscache_enqueue_dependents(object, FSCACHE_OBJECT_EV_KILL)) + return NO_TRANSIT; /* Not finished */ + return transit_to(WAIT_FOR_CLEARANCE); } /* - * withdraw an object from active service + * Drop an object's attachments */ -static void fscache_withdraw_object(struct fscache_object *object) +static const struct fscache_state *fscache_drop_object(struct fscache_object *object, + int event) { - struct fscache_cookie *cookie; - bool detached; + struct fscache_object *parent = object->parent; + struct fscache_cookie *cookie = object->cookie; + struct fscache_cache *cache = object->cache; + bool awaken = false; - _enter(""); + _enter("{OBJ%x,%d},%d", object->debug_id, object->n_children, event); - spin_lock(&object->lock); - cookie = object->cookie; - if (cookie) { - /* need to get the cookie lock before the object lock, starting - * from the object pointer */ - atomic_inc(&cookie->usage); - spin_unlock(&object->lock); + ASSERT(cookie != NULL); + ASSERT(!hlist_unhashed(&object->cookie_link)); - detached = false; - spin_lock(&cookie->lock); - spin_lock(&object->lock); + /* Make sure the cookie no longer points here and that the netfs isn't + * waiting for us. + */ + spin_lock(&cookie->lock); + hlist_del_init(&object->cookie_link); + if (test_and_clear_bit(FSCACHE_COOKIE_INVALIDATING, &cookie->flags)) + awaken = true; + spin_unlock(&cookie->lock); - if (object->cookie == cookie) { - hlist_del_init(&object->cookie_link); - object->cookie = NULL; - fscache_invalidation_complete(cookie); - detached = true; - } - spin_unlock(&cookie->lock); - fscache_cookie_put(cookie); - if (detached) - fscache_cookie_put(cookie); - } + if (awaken) + wake_up_bit(&cookie->flags, FSCACHE_COOKIE_INVALIDATING); + /* Prevent a race with our last child, which has to signal EV_CLEARED + * before dropping our spinlock. + */ + spin_lock(&object->lock); spin_unlock(&object->lock); - fscache_drop_object(object); -} + /* Discard from the cache's collection of objects */ + spin_lock(&cache->object_list_lock); + list_del_init(&object->cache_link); + spin_unlock(&cache->object_list_lock); -/* - * withdraw an object from active service at the behest of the cache - * - need break the links to a cached object cookie - * - called under two situations: - * (1) recycler decides to reclaim an in-use object - * (2) a cache is unmounted - * - have to take care as the cookie can be being relinquished by the netfs - * simultaneously - * - the object is pinned by the caller holding a refcount on it - */ -void fscache_withdrawing_object(struct fscache_cache *cache, - struct fscache_object *object) -{ - bool enqueue = false; + fscache_stat(&fscache_n_cop_drop_object); + cache->ops->drop_object(object); + fscache_stat_d(&fscache_n_cop_drop_object); - _enter(",OBJ%x", object->debug_id); + /* The parent object wants to know when all it dependents have gone */ + if (parent) { + _debug("release parent OBJ%x {%d}", + parent->debug_id, parent->n_children); - spin_lock(&object->lock); - if (object->state < FSCACHE_OBJECT_WITHDRAWING) { - object->state = FSCACHE_OBJECT_WITHDRAWING; - enqueue = true; + spin_lock(&parent->lock); + parent->n_children--; + if (parent->n_children == 0) + fscache_raise_event(parent, FSCACHE_OBJECT_EV_CLEARED); + spin_unlock(&parent->lock); + object->parent = NULL; } - spin_unlock(&object->lock); - if (enqueue) - fscache_enqueue_object(object); + /* this just shifts the object release to the work processor */ + fscache_put_object(object); + fscache_stat(&fscache_n_object_dead); _leave(""); + return transit_to(OBJECT_DEAD); } /* @@ -771,7 +736,7 @@ static int fscache_get_object(struct fscache_object *object) } /* - * discard a ref on a work item + * Discard a ref on an object */ static void fscache_put_object(struct fscache_object *object) { @@ -780,6 +745,22 @@ static void fscache_put_object(struct fscache_object *object) fscache_stat_d(&fscache_n_cop_put_object); } +/** + * fscache_object_destroy - Note that a cache object is about to be destroyed + * @object: The object to be destroyed + * + * Note the imminent destruction and deallocation of a cache object record. + */ +void fscache_object_destroy(struct fscache_object *object) +{ + fscache_objlist_remove(object); + + /* We can get rid of the cookie now */ + fscache_cookie_put(object->cookie); + object->cookie = NULL; +} +EXPORT_SYMBOL(fscache_object_destroy); + /* * enqueue an object for metadata-type processing */ @@ -803,7 +784,7 @@ void fscache_enqueue_object(struct fscache_object *object) /** * fscache_object_sleep_till_congested - Sleep until object wq is congested - * @timoutp: Scheduler sleep timeout + * @timeoutp: Scheduler sleep timeout * * Allow an object handler to sleep until the object workqueue is congested. * @@ -831,18 +812,21 @@ bool fscache_object_sleep_till_congested(signed long *timeoutp) EXPORT_SYMBOL_GPL(fscache_object_sleep_till_congested); /* - * enqueue the dependents of an object for metadata-type processing - * - the caller must hold the object's lock - * - this may cause an already locked object to wind up being processed again + * Enqueue the dependents of an object for metadata-type processing. + * + * If we don't manage to finish the list before the scheduler wants to run + * again then return false immediately. We return true if the list was + * cleared. */ -static void fscache_enqueue_dependents(struct fscache_object *object) +static bool fscache_enqueue_dependents(struct fscache_object *object, int event) { struct fscache_object *dep; + bool ret = true; _enter("{OBJ%x}", object->debug_id); if (list_empty(&object->dependents)) - return; + return true; spin_lock(&object->lock); @@ -851,23 +835,23 @@ static void fscache_enqueue_dependents(struct fscache_object *object) struct fscache_object, dep_link); list_del_init(&dep->dep_link); - - /* sort onto appropriate lists */ - fscache_enqueue_object(dep); + fscache_raise_event(dep, event); fscache_put_object(dep); - if (!list_empty(&object->dependents)) - cond_resched_lock(&object->lock); + if (!list_empty(&object->dependents) && need_resched()) { + ret = false; + break; + } } spin_unlock(&object->lock); + return ret; } /* * remove an object from whatever queue it's waiting on - * - the caller must hold object->lock */ -void fscache_dequeue_object(struct fscache_object *object) +static void fscache_dequeue_object(struct fscache_object *object) { _enter("{OBJ%x}", object->debug_id); @@ -886,7 +870,10 @@ void fscache_dequeue_object(struct fscache_object *object) * @data: The auxiliary data for the object * @datalen: The size of the auxiliary data * - * This function consults the netfs about the coherency state of an object + * This function consults the netfs about the coherency state of an object. + * The caller must be holding a ref on cookie->n_active (held by + * fscache_look_up_object() on behalf of the cache backend during object lookup + * and creation). */ enum fscache_checkaux fscache_check_aux(struct fscache_object *object, const void *data, uint16_t datalen) @@ -927,12 +914,23 @@ EXPORT_SYMBOL(fscache_check_aux); /* * Asynchronously invalidate an object. */ -static void fscache_invalidate_object(struct fscache_object *object) +static const struct fscache_state *_fscache_invalidate_object(struct fscache_object *object, + int event) { struct fscache_operation *op; struct fscache_cookie *cookie = object->cookie; - _enter("{OBJ%x}", object->debug_id); + _enter("{OBJ%x},%d", object->debug_id, event); + + /* We're going to need the cookie. If the cookie is not available then + * retire the object instead. + */ + if (!fscache_use_cookie(object)) { + ASSERT(object->cookie->stores.rnode == NULL); + set_bit(FSCACHE_COOKIE_RETIRED, &cookie->flags); + _leave(" [no cookie]"); + return transit_to(KILL_OBJECT); + } /* Reject any new read/write ops and abort any that are pending. */ fscache_invalidate_writes(cookie); @@ -941,14 +939,13 @@ static void fscache_invalidate_object(struct fscache_object *object) /* Now we have to wait for in-progress reads and writes */ op = kzalloc(sizeof(*op), GFP_KERNEL); - if (!op) { - fscache_raise_event(object, FSCACHE_OBJECT_EV_ERROR); - _leave(" [ENOMEM]"); - return; - } + if (!op) + goto nomem; fscache_operation_init(op, object->cache->ops->invalidate_object, NULL); - op->flags = FSCACHE_OP_ASYNC | (1 << FSCACHE_OP_EXCLUSIVE); + op->flags = FSCACHE_OP_ASYNC | + (1 << FSCACHE_OP_EXCLUSIVE) | + (1 << FSCACHE_OP_UNUSE_COOKIE); spin_lock(&cookie->lock); if (fscache_submit_exclusive_op(object, op) < 0) @@ -965,13 +962,50 @@ static void fscache_invalidate_object(struct fscache_object *object) /* We can allow read and write requests to come in once again. They'll * queue up behind our exclusive invalidation operation. */ - fscache_invalidation_complete(cookie); - _leave(""); - return; + if (test_and_clear_bit(FSCACHE_COOKIE_INVALIDATING, &cookie->flags)) + wake_up_bit(&cookie->flags, FSCACHE_COOKIE_INVALIDATING); + _leave(" [ok]"); + return transit_to(UPDATE_OBJECT); + +nomem: + clear_bit(FSCACHE_OBJECT_IS_LIVE, &object->flags); + fscache_unuse_cookie(object); + _leave(" [ENOMEM]"); + return transit_to(KILL_OBJECT); submit_op_failed: + clear_bit(FSCACHE_OBJECT_IS_LIVE, &object->flags); spin_unlock(&cookie->lock); kfree(op); - fscache_raise_event(object, FSCACHE_OBJECT_EV_ERROR); _leave(" [EIO]"); + return transit_to(KILL_OBJECT); +} + +static const struct fscache_state *fscache_invalidate_object(struct fscache_object *object, + int event) +{ + const struct fscache_state *s; + + fscache_stat(&fscache_n_invalidates_run); + fscache_stat(&fscache_n_cop_invalidate_object); + s = _fscache_invalidate_object(object, event); + fscache_stat_d(&fscache_n_cop_invalidate_object); + return s; +} + +/* + * Asynchronously update an object. + */ +static const struct fscache_state *fscache_update_object(struct fscache_object *object, + int event) +{ + _enter("{OBJ%x},%d", object->debug_id, event); + + fscache_stat(&fscache_n_updates_run); + fscache_stat(&fscache_n_cop_update_object); + object->cache->ops->update_object(object); + fscache_stat_d(&fscache_n_cop_update_object); + + _leave(""); + return transit_to(WAIT_FOR_CMD); } diff --git a/fs/fscache/operation.c b/fs/fscache/operation.c index 762a9ec4ffa4..318071aca217 100644 --- a/fs/fscache/operation.c +++ b/fs/fscache/operation.c @@ -35,7 +35,7 @@ void fscache_enqueue_operation(struct fscache_operation *op) ASSERT(list_empty(&op->pend_link)); ASSERT(op->processor != NULL); - ASSERTCMP(op->object->state, >=, FSCACHE_OBJECT_AVAILABLE); + ASSERT(fscache_object_is_available(op->object)); ASSERTCMP(atomic_read(&op->usage), >, 0); ASSERTCMP(op->state, ==, FSCACHE_OP_ST_IN_PROGRESS); @@ -119,7 +119,7 @@ int fscache_submit_exclusive_op(struct fscache_object *object, /* need to issue a new write op after this */ clear_bit(FSCACHE_OBJECT_PENDING_WRITE, &object->flags); ret = 0; - } else if (object->state == FSCACHE_OBJECT_CREATING) { + } else if (test_bit(FSCACHE_OBJECT_IS_LOOKED_UP, &object->flags)) { op->object = object; object->n_ops++; object->n_exclusive++; /* reads and writes must wait */ @@ -144,7 +144,7 @@ int fscache_submit_exclusive_op(struct fscache_object *object, */ static void fscache_report_unexpected_submission(struct fscache_object *object, struct fscache_operation *op, - unsigned long ostate) + const struct fscache_state *ostate) { static bool once_only; struct fscache_operation *p; @@ -155,11 +155,8 @@ static void fscache_report_unexpected_submission(struct fscache_object *object, once_only = true; kdebug("unexpected submission OP%x [OBJ%x %s]", - op->debug_id, object->debug_id, - fscache_object_states[object->state]); - kdebug("objstate=%s [%s]", - fscache_object_states[object->state], - fscache_object_states[ostate]); + op->debug_id, object->debug_id, object->state->name); + kdebug("objstate=%s [%s]", object->state->name, ostate->name); kdebug("objflags=%lx", object->flags); kdebug("objevent=%lx [%lx]", object->events, object->event_mask); kdebug("ops=%u inp=%u exc=%u", @@ -190,7 +187,7 @@ static void fscache_report_unexpected_submission(struct fscache_object *object, int fscache_submit_op(struct fscache_object *object, struct fscache_operation *op) { - unsigned long ostate; + const struct fscache_state *ostate; int ret; _enter("{OBJ%x OP%x},{%u}", @@ -226,16 +223,14 @@ int fscache_submit_op(struct fscache_object *object, fscache_run_op(object, op); } ret = 0; - } else if (object->state == FSCACHE_OBJECT_CREATING) { + } else if (test_bit(FSCACHE_OBJECT_IS_LOOKED_UP, &object->flags)) { op->object = object; object->n_ops++; atomic_inc(&op->usage); list_add_tail(&op->pend_link, &object->pending_ops); fscache_stat(&fscache_n_op_pend); ret = 0; - } else if (object->state == FSCACHE_OBJECT_DYING || - object->state == FSCACHE_OBJECT_LC_DYING || - object->state == FSCACHE_OBJECT_WITHDRAWING) { + } else if (fscache_object_is_dying(object)) { fscache_stat(&fscache_n_op_rejected); op->state = FSCACHE_OP_ST_CANCELLED; ret = -ENOBUFS; @@ -265,8 +260,8 @@ void fscache_abort_object(struct fscache_object *object) } /* - * jump start the operation processing on an object - * - caller must hold object->lock + * Jump start the operation processing on an object. The caller must hold + * object->lock. */ void fscache_start_operations(struct fscache_object *object) { @@ -428,14 +423,10 @@ void fscache_put_operation(struct fscache_operation *op) object = op->object; - if (test_bit(FSCACHE_OP_DEC_READ_CNT, &op->flags)) { - if (atomic_dec_and_test(&object->n_reads)) { - clear_bit(FSCACHE_COOKIE_WAITING_ON_READS, - &object->cookie->flags); - wake_up_bit(&object->cookie->flags, - FSCACHE_COOKIE_WAITING_ON_READS); - } - } + if (test_bit(FSCACHE_OP_DEC_READ_CNT, &op->flags)) + atomic_dec(&object->n_reads); + if (test_bit(FSCACHE_OP_UNUSE_COOKIE, &op->flags)) + fscache_unuse_cookie(object); /* now... we may get called with the object spinlock held, so we * complete the cleanup here only if we can immediately acquire the diff --git a/fs/fscache/page.c b/fs/fscache/page.c index ff000e52072d..d479ab3c63e4 100644 --- a/fs/fscache/page.c +++ b/fs/fscache/page.c @@ -109,7 +109,7 @@ page_busy: * allocator as the work threads writing to the cache may all end up * sleeping on memory allocation, so we may need to impose a timeout * too. */ - if (!(gfp & __GFP_WAIT)) { + if (!(gfp & __GFP_WAIT) || !(gfp & __GFP_FS)) { fscache_stat(&fscache_n_store_vmscan_busy); return false; } @@ -163,10 +163,12 @@ static void fscache_attr_changed_op(struct fscache_operation *op) fscache_stat(&fscache_n_attr_changed_calls); - if (fscache_object_is_active(object)) { + if (fscache_object_is_active(object) && + fscache_use_cookie(object)) { fscache_stat(&fscache_n_cop_attr_changed); ret = object->cache->ops->attr_changed(object); fscache_stat_d(&fscache_n_cop_attr_changed); + fscache_unuse_cookie(object); if (ret < 0) fscache_abort_object(object); } @@ -233,7 +235,7 @@ static void fscache_release_retrieval_op(struct fscache_operation *_op) _enter("{OP%x}", op->op.debug_id); - ASSERTCMP(op->n_pages, ==, 0); + ASSERTCMP(atomic_read(&op->n_pages), ==, 0); fscache_hist(fscache_retrieval_histogram, op->start_time); if (op->context) @@ -246,6 +248,7 @@ static void fscache_release_retrieval_op(struct fscache_operation *_op) * allocate a retrieval op */ static struct fscache_retrieval *fscache_alloc_retrieval( + struct fscache_cookie *cookie, struct address_space *mapping, fscache_rw_complete_t end_io_func, void *context) @@ -260,7 +263,10 @@ static struct fscache_retrieval *fscache_alloc_retrieval( } fscache_operation_init(&op->op, NULL, fscache_release_retrieval_op); - op->op.flags = FSCACHE_OP_MYTHREAD | (1 << FSCACHE_OP_WAITING); + atomic_inc(&cookie->n_active); + op->op.flags = FSCACHE_OP_MYTHREAD | + (1UL << FSCACHE_OP_WAITING) | + (1UL << FSCACHE_OP_UNUSE_COOKIE); op->mapping = mapping; op->end_io_func = end_io_func; op->context = context; @@ -310,7 +316,7 @@ static void fscache_do_cancel_retrieval(struct fscache_operation *_op) struct fscache_retrieval *op = container_of(_op, struct fscache_retrieval, op); - op->n_pages = 0; + atomic_set(&op->n_pages, 0); } /* @@ -394,12 +400,13 @@ int __fscache_read_or_alloc_page(struct fscache_cookie *cookie, if (fscache_wait_for_deferred_lookup(cookie) < 0) return -ERESTARTSYS; - op = fscache_alloc_retrieval(page->mapping, end_io_func, context); + op = fscache_alloc_retrieval(cookie, page->mapping, + end_io_func,context); if (!op) { _leave(" = -ENOMEM"); return -ENOMEM; } - op->n_pages = 1; + atomic_set(&op->n_pages, 1); spin_lock(&cookie->lock); @@ -408,7 +415,7 @@ int __fscache_read_or_alloc_page(struct fscache_cookie *cookie, object = hlist_entry(cookie->backing_objects.first, struct fscache_object, cookie_link); - ASSERTCMP(object->state, >, FSCACHE_OBJECT_LOOKING_UP); + ASSERT(test_bit(FSCACHE_OBJECT_IS_LOOKED_UP, &object->flags)); atomic_inc(&object->n_reads); __set_bit(FSCACHE_OP_DEC_READ_CNT, &op->op.flags); @@ -465,6 +472,7 @@ nobufs_unlock_dec: atomic_dec(&object->n_reads); nobufs_unlock: spin_unlock(&cookie->lock); + atomic_dec(&cookie->n_active); kfree(op); nobufs: fscache_stat(&fscache_n_retrievals_nobufs); @@ -522,10 +530,10 @@ int __fscache_read_or_alloc_pages(struct fscache_cookie *cookie, if (fscache_wait_for_deferred_lookup(cookie) < 0) return -ERESTARTSYS; - op = fscache_alloc_retrieval(mapping, end_io_func, context); + op = fscache_alloc_retrieval(cookie, mapping, end_io_func, context); if (!op) return -ENOMEM; - op->n_pages = *nr_pages; + atomic_set(&op->n_pages, *nr_pages); spin_lock(&cookie->lock); @@ -589,6 +597,7 @@ nobufs_unlock_dec: atomic_dec(&object->n_reads); nobufs_unlock: spin_unlock(&cookie->lock); + atomic_dec(&cookie->n_active); kfree(op); nobufs: fscache_stat(&fscache_n_retrievals_nobufs); @@ -631,10 +640,10 @@ int __fscache_alloc_page(struct fscache_cookie *cookie, if (fscache_wait_for_deferred_lookup(cookie) < 0) return -ERESTARTSYS; - op = fscache_alloc_retrieval(page->mapping, NULL, NULL); + op = fscache_alloc_retrieval(cookie, page->mapping, NULL, NULL); if (!op) return -ENOMEM; - op->n_pages = 1; + atomic_set(&op->n_pages, 1); spin_lock(&cookie->lock); @@ -675,6 +684,7 @@ error: nobufs_unlock: spin_unlock(&cookie->lock); + atomic_dec(&cookie->n_active); kfree(op); nobufs: fscache_stat(&fscache_n_allocs_nobufs); @@ -729,8 +739,9 @@ static void fscache_write_op(struct fscache_operation *_op) */ spin_unlock(&object->lock); fscache_op_complete(&op->op, false); - _leave(" [cancel] op{f=%lx s=%u} obj{s=%u f=%lx}", - _op->flags, _op->state, object->state, object->flags); + _leave(" [cancel] op{f=%lx s=%u} obj{s=%s f=%lx}", + _op->flags, _op->state, object->state->short_name, + object->flags); return; } @@ -796,11 +807,16 @@ void fscache_invalidate_writes(struct fscache_cookie *cookie) _enter(""); - while (spin_lock(&cookie->stores_lock), - n = radix_tree_gang_lookup_tag(&cookie->stores, results, 0, - ARRAY_SIZE(results), - FSCACHE_COOKIE_PENDING_TAG), - n > 0) { + for (;;) { + spin_lock(&cookie->stores_lock); + n = radix_tree_gang_lookup_tag(&cookie->stores, results, 0, + ARRAY_SIZE(results), + FSCACHE_COOKIE_PENDING_TAG); + if (n == 0) { + spin_unlock(&cookie->stores_lock); + break; + } + for (i = n - 1; i >= 0; i--) { page = results[i]; radix_tree_delete(&cookie->stores, page->index); @@ -812,7 +828,6 @@ void fscache_invalidate_writes(struct fscache_cookie *cookie) page_cache_release(results[i]); } - spin_unlock(&cookie->stores_lock); _leave(""); } @@ -829,14 +844,12 @@ void fscache_invalidate_writes(struct fscache_cookie *cookie) * (1) negative lookup, object not yet created (FSCACHE_COOKIE_CREATING is * set) * - * (a) no writes yet (set FSCACHE_COOKIE_PENDING_FILL and queue deferred - * fill op) + * (a) no writes yet * * (b) writes deferred till post-creation (mark page for writing and * return immediately) * * (2) negative lookup, object created, initial fill being made from netfs - * (FSCACHE_COOKIE_INITIAL_FILL is set) * * (a) fill point not yet reached this page (mark page for writing and * return) @@ -873,7 +886,9 @@ int __fscache_write_page(struct fscache_cookie *cookie, fscache_operation_init(&op->op, fscache_write_op, fscache_release_write_op); - op->op.flags = FSCACHE_OP_ASYNC | (1 << FSCACHE_OP_WAITING); + op->op.flags = FSCACHE_OP_ASYNC | + (1 << FSCACHE_OP_WAITING) | + (1 << FSCACHE_OP_UNUSE_COOKIE); ret = radix_tree_preload(gfp & ~__GFP_HIGHMEM); if (ret < 0) @@ -919,6 +934,7 @@ int __fscache_write_page(struct fscache_cookie *cookie, op->op.debug_id = atomic_inc_return(&fscache_op_debug_id); op->store_limit = object->store_limit; + atomic_inc(&cookie->n_active); if (fscache_submit_op(object, &op->op) < 0) goto submit_failed; @@ -945,6 +961,7 @@ already_pending: return 0; submit_failed: + atomic_dec(&cookie->n_active); spin_lock(&cookie->stores_lock); radix_tree_delete(&cookie->stores, page->index); spin_unlock(&cookie->stores_lock); diff --git a/fs/fuse/file.c b/fs/fuse/file.c index bfb20a8642c3..5c121fe19c5f 100644 --- a/fs/fuse/file.c +++ b/fs/fuse/file.c @@ -2469,13 +2469,16 @@ static long fuse_file_fallocate(struct file *file, int mode, loff_t offset, .mode = mode }; int err; + bool lock_inode = !(mode & FALLOC_FL_KEEP_SIZE) || + (mode & FALLOC_FL_PUNCH_HOLE); if (fc->no_fallocate) return -EOPNOTSUPP; - if (mode & FALLOC_FL_PUNCH_HOLE) { + if (lock_inode) { mutex_lock(&inode->i_mutex); - fuse_set_nowrite(inode); + if (mode & FALLOC_FL_PUNCH_HOLE) + fuse_set_nowrite(inode); } req = fuse_get_req_nopages(fc); @@ -2510,8 +2513,9 @@ static long fuse_file_fallocate(struct file *file, int mode, loff_t offset, fuse_invalidate_attr(inode); out: - if (mode & FALLOC_FL_PUNCH_HOLE) { - fuse_release_nowrite(inode); + if (lock_inode) { + if (mode & FALLOC_FL_PUNCH_HOLE) + fuse_release_nowrite(inode); mutex_unlock(&inode->i_mutex); } diff --git a/fs/gfs2/Kconfig b/fs/gfs2/Kconfig index 5a376ab81feb..90c6a8faaecb 100644 --- a/fs/gfs2/Kconfig +++ b/fs/gfs2/Kconfig @@ -20,13 +20,12 @@ config GFS2_FS be found here: http://sources.redhat.com/cluster The "nolock" lock module is now built in to GFS2 by default. If - you want to use the DLM, be sure to enable HOTPLUG and IPv4/6 - networking. + you want to use the DLM, be sure to enable IPv4/6 networking. config GFS2_FS_LOCKING_DLM bool "GFS2 DLM locking" depends on (GFS2_FS!=n) && NET && INET && (IPV6 || IPV6=n) && \ - HOTPLUG && CONFIGFS_FS && SYSFS && (DLM=y || DLM=GFS2_FS) + CONFIGFS_FS && SYSFS && (DLM=y || DLM=GFS2_FS) help Multiple node locking module for GFS2 diff --git a/fs/gfs2/aops.c b/fs/gfs2/aops.c index 0bad69ed6336..ee48ad37d9c0 100644 --- a/fs/gfs2/aops.c +++ b/fs/gfs2/aops.c @@ -110,7 +110,7 @@ static int gfs2_writepage_common(struct page *page, /* Is the page fully outside i_size? (truncate in progress) */ offset = i_size & (PAGE_CACHE_SIZE-1); if (page->index > end_index || (page->index == end_index && !offset)) { - page->mapping->a_ops->invalidatepage(page, 0); + page->mapping->a_ops->invalidatepage(page, 0, PAGE_CACHE_SIZE); goto out; } return 1; @@ -299,7 +299,8 @@ static int gfs2_write_jdata_pagevec(struct address_space *mapping, /* Is the page fully outside i_size? (truncate in progress) */ if (page->index > end_index || (page->index == end_index && !offset)) { - page->mapping->a_ops->invalidatepage(page, 0); + page->mapping->a_ops->invalidatepage(page, 0, + PAGE_CACHE_SIZE); unlock_page(page); continue; } @@ -943,27 +944,33 @@ static void gfs2_discard(struct gfs2_sbd *sdp, struct buffer_head *bh) unlock_buffer(bh); } -static void gfs2_invalidatepage(struct page *page, unsigned long offset) +static void gfs2_invalidatepage(struct page *page, unsigned int offset, + unsigned int length) { struct gfs2_sbd *sdp = GFS2_SB(page->mapping->host); + unsigned int stop = offset + length; + int partial_page = (offset || length < PAGE_CACHE_SIZE); struct buffer_head *bh, *head; unsigned long pos = 0; BUG_ON(!PageLocked(page)); - if (offset == 0) + if (!partial_page) ClearPageChecked(page); if (!page_has_buffers(page)) goto out; bh = head = page_buffers(page); do { + if (pos + bh->b_size > stop) + return; + if (offset <= pos) gfs2_discard(sdp, bh); pos += bh->b_size; bh = bh->b_this_page; } while (bh != head); out: - if (offset == 0) + if (!partial_page) try_to_release_page(page, 0); } diff --git a/fs/gfs2/bmap.c b/fs/gfs2/bmap.c index 93b5809c20bb..5e2f56fccf6b 100644 --- a/fs/gfs2/bmap.c +++ b/fs/gfs2/bmap.c @@ -1232,7 +1232,9 @@ static int do_grow(struct inode *inode, u64 size) unstuff = 1; } - error = gfs2_trans_begin(sdp, RES_DINODE + RES_STATFS + RES_RG_BIT, 0); + error = gfs2_trans_begin(sdp, RES_DINODE + RES_STATFS + RES_RG_BIT + + (sdp->sd_args.ar_quota == GFS2_QUOTA_OFF ? + 0 : RES_QUOTA), 0); if (error) goto do_grow_release; diff --git a/fs/gfs2/dir.c b/fs/gfs2/dir.c index e0449c10286a..0cb4c1557f20 100644 --- a/fs/gfs2/dir.c +++ b/fs/gfs2/dir.c @@ -1125,13 +1125,14 @@ static int dir_double_exhash(struct gfs2_inode *dip) if (IS_ERR(hc)) return PTR_ERR(hc); - h = hc2 = kmalloc(hsize_bytes * 2, GFP_NOFS | __GFP_NOWARN); + hc2 = kmalloc(hsize_bytes * 2, GFP_NOFS | __GFP_NOWARN); if (hc2 == NULL) hc2 = __vmalloc(hsize_bytes * 2, GFP_NOFS, PAGE_KERNEL); if (!hc2) return -ENOMEM; + h = hc2; error = gfs2_meta_inode_buffer(dip, &dibh); if (error) goto out_kfree; @@ -1547,9 +1548,9 @@ out: /** * gfs2_dir_search - Search a directory - * @dip: The GFS2 inode - * @filename: - * @inode: + * @dip: The GFS2 dir inode + * @name: The name we are looking up + * @fail_on_exist: Fail if the name exists rather than looking it up * * This routine searches a directory for a file or another directory. * Assumes a glock is held on dip. @@ -1557,22 +1558,25 @@ out: * Returns: errno */ -struct inode *gfs2_dir_search(struct inode *dir, const struct qstr *name) +struct inode *gfs2_dir_search(struct inode *dir, const struct qstr *name, + bool fail_on_exist) { struct buffer_head *bh; struct gfs2_dirent *dent; - struct inode *inode; + u64 addr, formal_ino; + u16 dtype; dent = gfs2_dirent_search(dir, name, gfs2_dirent_find, &bh); if (dent) { if (IS_ERR(dent)) return ERR_CAST(dent); - inode = gfs2_inode_lookup(dir->i_sb, - be16_to_cpu(dent->de_type), - be64_to_cpu(dent->de_inum.no_addr), - be64_to_cpu(dent->de_inum.no_formal_ino), 0); + dtype = be16_to_cpu(dent->de_type); + addr = be64_to_cpu(dent->de_inum.no_addr); + formal_ino = be64_to_cpu(dent->de_inum.no_formal_ino); brelse(bh); - return inode; + if (fail_on_exist) + return ERR_PTR(-EEXIST); + return gfs2_inode_lookup(dir->i_sb, dtype, addr, formal_ino, 0); } return ERR_PTR(-ENOENT); } diff --git a/fs/gfs2/dir.h b/fs/gfs2/dir.h index ba9000bc1397..4f03bbd1873f 100644 --- a/fs/gfs2/dir.h +++ b/fs/gfs2/dir.h @@ -18,7 +18,8 @@ struct gfs2_inode; struct gfs2_inum; extern struct inode *gfs2_dir_search(struct inode *dir, - const struct qstr *filename); + const struct qstr *filename, + bool fail_on_exist); extern int gfs2_dir_check(struct inode *dir, const struct qstr *filename, const struct gfs2_inode *ip); extern int gfs2_dir_add(struct inode *inode, const struct qstr *filename, diff --git a/fs/gfs2/file.c b/fs/gfs2/file.c index cebfd404c1d9..72c3866a7320 100644 --- a/fs/gfs2/file.c +++ b/fs/gfs2/file.c @@ -531,21 +531,30 @@ static int gfs2_mmap(struct file *file, struct vm_area_struct *vma) } /** - * gfs2_open - open a file - * @inode: the inode to open - * @file: the struct file for this opening + * gfs2_open_common - This is common to open and atomic_open + * @inode: The inode being opened + * @file: The file being opened * - * Returns: errno + * This maybe called under a glock or not depending upon how it has + * been called. We must always be called under a glock for regular + * files, however. For other file types, it does not matter whether + * we hold the glock or not. + * + * Returns: Error code or 0 for success */ -static int gfs2_open(struct inode *inode, struct file *file) +int gfs2_open_common(struct inode *inode, struct file *file) { - struct gfs2_inode *ip = GFS2_I(inode); - struct gfs2_holder i_gh; struct gfs2_file *fp; - int error; + int ret; - fp = kzalloc(sizeof(struct gfs2_file), GFP_KERNEL); + if (S_ISREG(inode->i_mode)) { + ret = generic_file_open(inode, file); + if (ret) + return ret; + } + + fp = kzalloc(sizeof(struct gfs2_file), GFP_NOFS); if (!fp) return -ENOMEM; @@ -553,29 +562,43 @@ static int gfs2_open(struct inode *inode, struct file *file) gfs2_assert_warn(GFS2_SB(inode), !file->private_data); file->private_data = fp; + return 0; +} + +/** + * gfs2_open - open a file + * @inode: the inode to open + * @file: the struct file for this opening + * + * After atomic_open, this function is only used for opening files + * which are already cached. We must still get the glock for regular + * files to ensure that we have the file size uptodate for the large + * file check which is in the common code. That is only an issue for + * regular files though. + * + * Returns: errno + */ + +static int gfs2_open(struct inode *inode, struct file *file) +{ + struct gfs2_inode *ip = GFS2_I(inode); + struct gfs2_holder i_gh; + int error; + bool need_unlock = false; if (S_ISREG(ip->i_inode.i_mode)) { error = gfs2_glock_nq_init(ip->i_gl, LM_ST_SHARED, LM_FLAG_ANY, &i_gh); if (error) - goto fail; + return error; + need_unlock = true; + } - if (!(file->f_flags & O_LARGEFILE) && - i_size_read(inode) > MAX_NON_LFS) { - error = -EOVERFLOW; - goto fail_gunlock; - } + error = gfs2_open_common(inode, file); + if (need_unlock) gfs2_glock_dq_uninit(&i_gh); - } - - return 0; -fail_gunlock: - gfs2_glock_dq_uninit(&i_gh); -fail: - file->private_data = NULL; - kfree(fp); return error; } diff --git a/fs/gfs2/glops.c b/fs/gfs2/glops.c index c66e99c97571..5f2e5224c51c 100644 --- a/fs/gfs2/glops.c +++ b/fs/gfs2/glops.c @@ -54,7 +54,6 @@ static void __gfs2_ail_flush(struct gfs2_glock *gl, bool fsync) struct gfs2_bufdata *bd, *tmp; struct buffer_head *bh; const unsigned long b_state = (1UL << BH_Dirty)|(1UL << BH_Pinned)|(1UL << BH_Lock); - sector_t blocknr; gfs2_log_lock(sdp); spin_lock(&sdp->sd_ail_lock); @@ -65,13 +64,6 @@ static void __gfs2_ail_flush(struct gfs2_glock *gl, bool fsync) continue; gfs2_ail_error(gl, bh); } - blocknr = bh->b_blocknr; - bh->b_private = NULL; - gfs2_remove_from_ail(bd); /* drops ref on bh */ - - bd->bd_bh = NULL; - bd->bd_blkno = blocknr; - gfs2_trans_add_revoke(sdp, bd); } GLOCK_BUG_ON(gl, !fsync && atomic_read(&gl->gl_ail_count)); diff --git a/fs/gfs2/inode.c b/fs/gfs2/inode.c index 62b484e4a9e4..bbb2715171cd 100644 --- a/fs/gfs2/inode.c +++ b/fs/gfs2/inode.c @@ -313,7 +313,7 @@ struct inode *gfs2_lookupi(struct inode *dir, const struct qstr *name, goto out; } - inode = gfs2_dir_search(dir, name); + inode = gfs2_dir_search(dir, name, false); if (IS_ERR(inode)) error = PTR_ERR(inode); out: @@ -346,17 +346,6 @@ static int create_ok(struct gfs2_inode *dip, const struct qstr *name, if (!dip->i_inode.i_nlink) return -ENOENT; - error = gfs2_dir_check(&dip->i_inode, name, NULL); - switch (error) { - case -ENOENT: - error = 0; - break; - case 0: - return -EEXIST; - default: - return error; - } - if (dip->i_entries == (u32)-1) return -EFBIG; if (S_ISDIR(mode) && dip->i_inode.i_nlink == (u32)-1) @@ -546,6 +535,7 @@ static int gfs2_security_init(struct gfs2_inode *dip, struct gfs2_inode *ip, * gfs2_create_inode - Create a new inode * @dir: The parent directory * @dentry: The new dentry + * @file: If non-NULL, the file which is being opened * @mode: The permissions on the new inode * @dev: For device nodes, this is the device number * @symname: For symlinks, this is the link destination @@ -555,8 +545,9 @@ static int gfs2_security_init(struct gfs2_inode *dip, struct gfs2_inode *ip, */ static int gfs2_create_inode(struct inode *dir, struct dentry *dentry, + struct file *file, umode_t mode, dev_t dev, const char *symname, - unsigned int size, int excl) + unsigned int size, int excl, int *opened) { const struct qstr *name = &dentry->d_name; struct gfs2_holder ghs[2]; @@ -564,6 +555,7 @@ static int gfs2_create_inode(struct inode *dir, struct dentry *dentry, struct gfs2_inode *dip = GFS2_I(dir), *ip; struct gfs2_sbd *sdp = GFS2_SB(&dip->i_inode); struct gfs2_glock *io_gl; + struct dentry *d; int error; u32 aflags = 0; int arq; @@ -584,15 +576,30 @@ static int gfs2_create_inode(struct inode *dir, struct dentry *dentry, goto fail; error = create_ok(dip, name, mode); - if ((error == -EEXIST) && S_ISREG(mode) && !excl) { - inode = gfs2_lookupi(dir, &dentry->d_name, 0); - gfs2_glock_dq_uninit(ghs); - d_instantiate(dentry, inode); - return IS_ERR(inode) ? PTR_ERR(inode) : 0; - } if (error) goto fail_gunlock; + inode = gfs2_dir_search(dir, &dentry->d_name, !S_ISREG(mode) || excl); + error = PTR_ERR(inode); + if (!IS_ERR(inode)) { + d = d_splice_alias(inode, dentry); + error = 0; + if (file && !IS_ERR(d)) { + if (d == NULL) + d = dentry; + if (S_ISREG(inode->i_mode)) + error = finish_open(file, d, gfs2_open_common, opened); + else + error = finish_no_open(file, d); + } + gfs2_glock_dq_uninit(ghs); + if (IS_ERR(d)) + return PTR_RET(d); + return error; + } else if (error != -ENOENT) { + goto fail_gunlock; + } + arq = error = gfs2_diradd_alloc_required(dir, name); if (error < 0) goto fail_gunlock; @@ -686,10 +693,12 @@ static int gfs2_create_inode(struct inode *dir, struct dentry *dentry, goto fail_gunlock3; mark_inode_dirty(inode); + d_instantiate(dentry, inode); + if (file) + error = finish_open(file, dentry, gfs2_open_common, opened); gfs2_glock_dq_uninit(ghs); gfs2_glock_dq_uninit(ghs + 1); - d_instantiate(dentry, inode); - return 0; + return error; fail_gunlock3: gfs2_glock_dq_uninit(ghs + 1); @@ -729,36 +738,56 @@ fail: static int gfs2_create(struct inode *dir, struct dentry *dentry, umode_t mode, bool excl) { - return gfs2_create_inode(dir, dentry, S_IFREG | mode, 0, NULL, 0, excl); + return gfs2_create_inode(dir, dentry, NULL, S_IFREG | mode, 0, NULL, 0, excl, NULL); } /** - * gfs2_lookup - Look up a filename in a directory and return its inode + * __gfs2_lookup - Look up a filename in a directory and return its inode * @dir: The directory inode * @dentry: The dentry of the new inode - * @nd: passed from Linux VFS, ignored by us + * @file: File to be opened + * @opened: atomic_open flags * - * Called by the VFS layer. Lock dir and call gfs2_lookupi() * * Returns: errno */ -static struct dentry *gfs2_lookup(struct inode *dir, struct dentry *dentry, - unsigned int flags) +static struct dentry *__gfs2_lookup(struct inode *dir, struct dentry *dentry, + struct file *file, int *opened) { - struct inode *inode = gfs2_lookupi(dir, &dentry->d_name, 0); - if (inode && !IS_ERR(inode)) { - struct gfs2_glock *gl = GFS2_I(inode)->i_gl; - struct gfs2_holder gh; - int error; - error = gfs2_glock_nq_init(gl, LM_ST_SHARED, LM_FLAG_ANY, &gh); - if (error) { - iput(inode); - return ERR_PTR(error); - } - gfs2_glock_dq_uninit(&gh); + struct inode *inode; + struct dentry *d; + struct gfs2_holder gh; + struct gfs2_glock *gl; + int error; + + inode = gfs2_lookupi(dir, &dentry->d_name, 0); + if (!inode) + return NULL; + if (IS_ERR(inode)) + return ERR_CAST(inode); + + gl = GFS2_I(inode)->i_gl; + error = gfs2_glock_nq_init(gl, LM_ST_SHARED, LM_FLAG_ANY, &gh); + if (error) { + iput(inode); + return ERR_PTR(error); } - return d_splice_alias(inode, dentry); + + d = d_splice_alias(inode, dentry); + if (file && S_ISREG(inode->i_mode)) + error = finish_open(file, dentry, gfs2_open_common, opened); + + gfs2_glock_dq_uninit(&gh); + if (error) + return ERR_PTR(error); + return d; +} + +static struct dentry *gfs2_lookup(struct inode *dir, struct dentry *dentry, + unsigned flags) +{ + return __gfs2_lookup(dir, dentry, NULL, NULL); } /** @@ -1076,7 +1105,7 @@ static int gfs2_symlink(struct inode *dir, struct dentry *dentry, if (size > sdp->sd_sb.sb_bsize - sizeof(struct gfs2_dinode) - 1) return -ENAMETOOLONG; - return gfs2_create_inode(dir, dentry, S_IFLNK | S_IRWXUGO, 0, symname, size, 0); + return gfs2_create_inode(dir, dentry, NULL, S_IFLNK | S_IRWXUGO, 0, symname, size, 0, NULL); } /** @@ -1092,7 +1121,7 @@ static int gfs2_mkdir(struct inode *dir, struct dentry *dentry, umode_t mode) { struct gfs2_sbd *sdp = GFS2_SB(dir); unsigned dsize = sdp->sd_sb.sb_bsize - sizeof(struct gfs2_dinode); - return gfs2_create_inode(dir, dentry, S_IFDIR | mode, 0, NULL, dsize, 0); + return gfs2_create_inode(dir, dentry, NULL, S_IFDIR | mode, 0, NULL, dsize, 0, NULL); } /** @@ -1107,7 +1136,43 @@ static int gfs2_mkdir(struct inode *dir, struct dentry *dentry, umode_t mode) static int gfs2_mknod(struct inode *dir, struct dentry *dentry, umode_t mode, dev_t dev) { - return gfs2_create_inode(dir, dentry, mode, dev, NULL, 0, 0); + return gfs2_create_inode(dir, dentry, NULL, mode, dev, NULL, 0, 0, NULL); +} + +/** + * gfs2_atomic_open - Atomically open a file + * @dir: The directory + * @dentry: The proposed new entry + * @file: The proposed new struct file + * @flags: open flags + * @mode: File mode + * @opened: Flag to say whether the file has been opened or not + * + * Returns: error code or 0 for success + */ + +static int gfs2_atomic_open(struct inode *dir, struct dentry *dentry, + struct file *file, unsigned flags, + umode_t mode, int *opened) +{ + struct dentry *d; + bool excl = !!(flags & O_EXCL); + + d = __gfs2_lookup(dir, dentry, file, opened); + if (IS_ERR(d)) + return PTR_ERR(d); + if (d == NULL) + d = dentry; + if (d->d_inode) { + if (!(*opened & FILE_OPENED)) + return finish_no_open(file, d); + return 0; + } + + if (!(flags & O_CREAT)) + return -ENOENT; + + return gfs2_create_inode(dir, dentry, file, S_IFREG | mode, 0, NULL, 0, excl, opened); } /* @@ -1787,6 +1852,7 @@ const struct inode_operations gfs2_dir_iops = { .removexattr = gfs2_removexattr, .fiemap = gfs2_fiemap, .get_acl = gfs2_get_acl, + .atomic_open = gfs2_atomic_open, }; const struct inode_operations gfs2_symlink_iops = { diff --git a/fs/gfs2/inode.h b/fs/gfs2/inode.h index c53c7477f6da..ba4d9492d422 100644 --- a/fs/gfs2/inode.h +++ b/fs/gfs2/inode.h @@ -109,6 +109,7 @@ extern int gfs2_permission(struct inode *inode, int mask); extern int gfs2_setattr_simple(struct inode *inode, struct iattr *attr); extern struct inode *gfs2_lookup_simple(struct inode *dip, const char *name); extern void gfs2_dinode_out(const struct gfs2_inode *ip, void *buf); +extern int gfs2_open_common(struct inode *inode, struct file *file); extern const struct inode_operations gfs2_file_iops; extern const struct inode_operations gfs2_dir_iops; diff --git a/fs/gfs2/log.c b/fs/gfs2/log.c index b404f4853034..610613fb65b5 100644 --- a/fs/gfs2/log.c +++ b/fs/gfs2/log.c @@ -211,15 +211,16 @@ static void gfs2_ail1_empty_one(struct gfs2_sbd *sdp, struct gfs2_trans *tr) static int gfs2_ail1_empty(struct gfs2_sbd *sdp) { struct gfs2_trans *tr, *s; + int oldest_tr = 1; int ret; spin_lock(&sdp->sd_ail_lock); list_for_each_entry_safe_reverse(tr, s, &sdp->sd_ail1_list, tr_list) { gfs2_ail1_empty_one(sdp, tr); - if (list_empty(&tr->tr_ail1_list)) + if (list_empty(&tr->tr_ail1_list) && oldest_tr) list_move(&tr->tr_list, &sdp->sd_ail2_list); else - break; + oldest_tr = 0; } ret = list_empty(&sdp->sd_ail1_list); spin_unlock(&sdp->sd_ail_lock); @@ -317,7 +318,7 @@ static void ail2_empty(struct gfs2_sbd *sdp, unsigned int new_tail) int gfs2_log_reserve(struct gfs2_sbd *sdp, unsigned int blks) { - unsigned reserved_blks = 6 * (4096 / sdp->sd_vfs->s_blocksize); + unsigned reserved_blks = 7 * (4096 / sdp->sd_vfs->s_blocksize); unsigned wanted = blks + reserved_blks; DEFINE_WAIT(wait); int did_wait = 0; @@ -545,6 +546,76 @@ void gfs2_ordered_del_inode(struct gfs2_inode *ip) spin_unlock(&sdp->sd_ordered_lock); } +void gfs2_add_revoke(struct gfs2_sbd *sdp, struct gfs2_bufdata *bd) +{ + struct buffer_head *bh = bd->bd_bh; + struct gfs2_glock *gl = bd->bd_gl; + + gfs2_remove_from_ail(bd); + bd->bd_bh = NULL; + bh->b_private = NULL; + bd->bd_blkno = bh->b_blocknr; + bd->bd_ops = &gfs2_revoke_lops; + sdp->sd_log_num_revoke++; + atomic_inc(&gl->gl_revokes); + set_bit(GLF_LFLUSH, &gl->gl_flags); + list_add(&bd->bd_list, &sdp->sd_log_le_revoke); +} + +void gfs2_write_revokes(struct gfs2_sbd *sdp) +{ + struct gfs2_trans *tr; + struct gfs2_bufdata *bd, *tmp; + int have_revokes = 0; + int max_revokes = (sdp->sd_sb.sb_bsize - sizeof(struct gfs2_log_descriptor)) / sizeof(u64); + + gfs2_ail1_empty(sdp); + spin_lock(&sdp->sd_ail_lock); + list_for_each_entry(tr, &sdp->sd_ail1_list, tr_list) { + list_for_each_entry(bd, &tr->tr_ail2_list, bd_ail_st_list) { + if (list_empty(&bd->bd_list)) { + have_revokes = 1; + goto done; + } + } + } +done: + spin_unlock(&sdp->sd_ail_lock); + if (have_revokes == 0) + return; + while (sdp->sd_log_num_revoke > max_revokes) + max_revokes += (sdp->sd_sb.sb_bsize - sizeof(struct gfs2_meta_header)) / sizeof(u64); + max_revokes -= sdp->sd_log_num_revoke; + if (!sdp->sd_log_num_revoke) { + atomic_dec(&sdp->sd_log_blks_free); + /* If no blocks have been reserved, we need to also + * reserve a block for the header */ + if (!sdp->sd_log_blks_reserved) + atomic_dec(&sdp->sd_log_blks_free); + } + gfs2_log_lock(sdp); + spin_lock(&sdp->sd_ail_lock); + list_for_each_entry(tr, &sdp->sd_ail1_list, tr_list) { + list_for_each_entry_safe(bd, tmp, &tr->tr_ail2_list, bd_ail_st_list) { + if (max_revokes == 0) + goto out_of_blocks; + if (!list_empty(&bd->bd_list)) + continue; + gfs2_add_revoke(sdp, bd); + max_revokes--; + } + } +out_of_blocks: + spin_unlock(&sdp->sd_ail_lock); + gfs2_log_unlock(sdp); + + if (!sdp->sd_log_num_revoke) { + atomic_inc(&sdp->sd_log_blks_free); + if (!sdp->sd_log_blks_reserved) + atomic_inc(&sdp->sd_log_blks_free); + } +} + /** * log_write_header - Get and initialize a journal header buffer * @sdp: The GFS2 superblock @@ -562,7 +633,6 @@ static void log_write_header(struct gfs2_sbd *sdp, u32 flags) lh = page_address(page); clear_page(lh); - gfs2_ail1_empty(sdp); tail = current_tail(sdp); lh->lh_header.mh_magic = cpu_to_be32(GFS2_MAGIC); diff --git a/fs/gfs2/log.h b/fs/gfs2/log.h index 3566f35915e0..37216634f0aa 100644 --- a/fs/gfs2/log.h +++ b/fs/gfs2/log.h @@ -72,5 +72,7 @@ extern void gfs2_ail1_flush(struct gfs2_sbd *sdp, struct writeback_control *wbc) extern void gfs2_log_shutdown(struct gfs2_sbd *sdp); extern void gfs2_meta_syncfs(struct gfs2_sbd *sdp); extern int gfs2_logd(void *data); +extern void gfs2_add_revoke(struct gfs2_sbd *sdp, struct gfs2_bufdata *bd); +extern void gfs2_write_revokes(struct gfs2_sbd *sdp); #endif /* __LOG_DOT_H__ */ diff --git a/fs/gfs2/lops.c b/fs/gfs2/lops.c index 6c33d7b6e0c4..17c5b5d7dc88 100644 --- a/fs/gfs2/lops.c +++ b/fs/gfs2/lops.c @@ -16,6 +16,7 @@ #include <linux/gfs2_ondisk.h> #include <linux/bio.h> #include <linux/fs.h> +#include <linux/list_sort.h> #include "gfs2.h" #include "incore.h" @@ -401,6 +402,20 @@ static void gfs2_check_magic(struct buffer_head *bh) kunmap_atomic(kaddr); } +static int blocknr_cmp(void *priv, struct list_head *a, struct list_head *b) +{ + struct gfs2_bufdata *bda, *bdb; + + bda = list_entry(a, struct gfs2_bufdata, bd_list); + bdb = list_entry(b, struct gfs2_bufdata, bd_list); + + if (bda->bd_bh->b_blocknr < bdb->bd_bh->b_blocknr) + return -1; + if (bda->bd_bh->b_blocknr > bdb->bd_bh->b_blocknr) + return 1; + return 0; +} + static void gfs2_before_commit(struct gfs2_sbd *sdp, unsigned int limit, unsigned int total, struct list_head *blist, bool is_databuf) @@ -413,6 +428,7 @@ static void gfs2_before_commit(struct gfs2_sbd *sdp, unsigned int limit, __be64 *ptr; gfs2_log_lock(sdp); + list_sort(NULL, blist, blocknr_cmp); bd1 = bd2 = list_prepare_entry(bd1, blist, bd_list); while(total) { num = total; @@ -590,6 +606,7 @@ static void revoke_lo_before_commit(struct gfs2_sbd *sdp) struct page *page; unsigned int length; + gfs2_write_revokes(sdp); if (!sdp->sd_log_num_revoke) return; @@ -836,10 +853,6 @@ const struct gfs2_log_operations gfs2_revoke_lops = { .lo_name = "revoke", }; -const struct gfs2_log_operations gfs2_rg_lops = { - .lo_name = "rg", -}; - const struct gfs2_log_operations gfs2_databuf_lops = { .lo_before_commit = databuf_lo_before_commit, .lo_after_commit = databuf_lo_after_commit, @@ -851,7 +864,6 @@ const struct gfs2_log_operations gfs2_databuf_lops = { const struct gfs2_log_operations *gfs2_log_ops[] = { &gfs2_databuf_lops, &gfs2_buf_lops, - &gfs2_rg_lops, &gfs2_revoke_lops, NULL, }; diff --git a/fs/gfs2/lops.h b/fs/gfs2/lops.h index 87e062e05c92..9ca2e6438419 100644 --- a/fs/gfs2/lops.h +++ b/fs/gfs2/lops.h @@ -23,7 +23,6 @@ extern const struct gfs2_log_operations gfs2_glock_lops; extern const struct gfs2_log_operations gfs2_buf_lops; extern const struct gfs2_log_operations gfs2_revoke_lops; -extern const struct gfs2_log_operations gfs2_rg_lops; extern const struct gfs2_log_operations gfs2_databuf_lops; extern const struct gfs2_log_operations *gfs2_log_ops[]; diff --git a/fs/gfs2/meta_io.c b/fs/gfs2/meta_io.c index 1a89afb68472..0da390686c08 100644 --- a/fs/gfs2/meta_io.c +++ b/fs/gfs2/meta_io.c @@ -296,10 +296,6 @@ void gfs2_remove_from_journal(struct buffer_head *bh, struct gfs2_trans *tr, int if (bd) { spin_lock(&sdp->sd_ail_lock); if (bd->bd_tr) { - gfs2_remove_from_ail(bd); - bh->b_private = NULL; - bd->bd_bh = NULL; - bd->bd_blkno = bh->b_blocknr; gfs2_trans_add_revoke(sdp, bd); } spin_unlock(&sdp->sd_ail_lock); diff --git a/fs/gfs2/ops_fstype.c b/fs/gfs2/ops_fstype.c index 60ede2a0f43f..0262c190b6f9 100644 --- a/fs/gfs2/ops_fstype.c +++ b/fs/gfs2/ops_fstype.c @@ -916,16 +916,16 @@ static int init_threads(struct gfs2_sbd *sdp, int undo) goto fail_quotad; p = kthread_run(gfs2_logd, sdp, "gfs2_logd"); - error = IS_ERR(p); - if (error) { + if (IS_ERR(p)) { + error = PTR_ERR(p); fs_err(sdp, "can't start logd thread: %d\n", error); return error; } sdp->sd_logd_process = p; p = kthread_run(gfs2_quotad, sdp, "gfs2_quotad"); - error = IS_ERR(p); - if (error) { + if (IS_ERR(p)) { + error = PTR_ERR(p); fs_err(sdp, "can't start quotad thread: %d\n", error); goto fail; } diff --git a/fs/gfs2/quota.c b/fs/gfs2/quota.c index c253b13722e8..3768c2f40e43 100644 --- a/fs/gfs2/quota.c +++ b/fs/gfs2/quota.c @@ -1154,11 +1154,6 @@ int gfs2_quota_sync(struct super_block *sb, int type) return error; } -static int gfs2_quota_sync_timeo(struct super_block *sb, int type) -{ - return gfs2_quota_sync(sb, type); -} - int gfs2_quota_refresh(struct gfs2_sbd *sdp, struct kqid qid) { struct gfs2_quota_data *qd; @@ -1414,7 +1409,7 @@ int gfs2_quotad(void *data) &tune->gt_statfs_quantum); /* Update quota file */ - quotad_check_timeo(sdp, "sync", gfs2_quota_sync_timeo, t, + quotad_check_timeo(sdp, "sync", gfs2_quota_sync, t, "ad_timeo, &tune->gt_quota_quantum); /* Check for & recover partially truncated inodes */ diff --git a/fs/gfs2/rgrp.c b/fs/gfs2/rgrp.c index 9809156e3d04..69317435faa7 100644 --- a/fs/gfs2/rgrp.c +++ b/fs/gfs2/rgrp.c @@ -1288,13 +1288,15 @@ int gfs2_fitrim(struct file *filp, void __user *argp) minlen = max_t(u64, r.minlen, q->limits.discard_granularity) >> bs_shift; + if (end <= start || minlen > sdp->sd_max_rg_data) + return -EINVAL; + rgd = gfs2_blk2rgrpd(sdp, start, 0); - rgd_end = gfs2_blk2rgrpd(sdp, end - 1, 0); + rgd_end = gfs2_blk2rgrpd(sdp, end, 0); - if (end <= start || - minlen > sdp->sd_max_rg_data || - start > rgd_end->rd_data0 + rgd_end->rd_data) - return -EINVAL; + if ((gfs2_rgrpd_get_first(sdp) == gfs2_rgrpd_get_next(rgd_end)) + && (start > rgd_end->rd_data0 + rgd_end->rd_data)) + return -EINVAL; /* start is beyond the end of the fs */ while (1) { @@ -1336,7 +1338,7 @@ int gfs2_fitrim(struct file *filp, void __user *argp) } out: - r.len = trimmed << 9; + r.len = trimmed << bs_shift; if (copy_to_user(argp, &r, sizeof(r))) return -EFAULT; diff --git a/fs/gfs2/trans.c b/fs/gfs2/trans.c index 7374907742a8..2b20d7046bf3 100644 --- a/fs/gfs2/trans.c +++ b/fs/gfs2/trans.c @@ -270,19 +270,12 @@ void gfs2_trans_add_meta(struct gfs2_glock *gl, struct buffer_head *bh) void gfs2_trans_add_revoke(struct gfs2_sbd *sdp, struct gfs2_bufdata *bd) { - struct gfs2_glock *gl = bd->bd_gl; struct gfs2_trans *tr = current->journal_info; BUG_ON(!list_empty(&bd->bd_list)); - BUG_ON(!list_empty(&bd->bd_ail_st_list)); - BUG_ON(!list_empty(&bd->bd_ail_gl_list)); - bd->bd_ops = &gfs2_revoke_lops; + gfs2_add_revoke(sdp, bd); tr->tr_touched = 1; tr->tr_num_revoke++; - sdp->sd_log_num_revoke++; - atomic_inc(&gl->gl_revokes); - set_bit(GLF_LFLUSH, &gl->gl_flags); - list_add(&bd->bd_list, &sdp->sd_log_le_revoke); } void gfs2_trans_add_unrevoke(struct gfs2_sbd *sdp, u64 blkno, unsigned int len) diff --git a/fs/jbd/transaction.c b/fs/jbd/transaction.c index e3e255c0a509..be0c39b66fe0 100644 --- a/fs/jbd/transaction.c +++ b/fs/jbd/transaction.c @@ -2019,16 +2019,20 @@ zap_buffer_unlocked: * void journal_invalidatepage() - invalidate a journal page * @journal: journal to use for flush * @page: page to flush - * @offset: length of page to invalidate. + * @offset: offset of the range to invalidate + * @length: length of the range to invalidate * - * Reap page buffers containing data after offset in page. + * Reap page buffers containing data in specified range in page. */ void journal_invalidatepage(journal_t *journal, struct page *page, - unsigned long offset) + unsigned int offset, + unsigned int length) { struct buffer_head *head, *bh, *next; + unsigned int stop = offset + length; unsigned int curr_off = 0; + int partial_page = (offset || length < PAGE_CACHE_SIZE); int may_free = 1; if (!PageLocked(page)) @@ -2036,6 +2040,8 @@ void journal_invalidatepage(journal_t *journal, if (!page_has_buffers(page)) return; + BUG_ON(stop > PAGE_CACHE_SIZE || stop < length); + /* We will potentially be playing with lists other than just the * data lists (especially for journaled data mode), so be * cautious in our locking. */ @@ -2045,11 +2051,14 @@ void journal_invalidatepage(journal_t *journal, unsigned int next_off = curr_off + bh->b_size; next = bh->b_this_page; + if (next_off > stop) + return; + if (offset <= curr_off) { /* This block is wholly outside the truncation point */ lock_buffer(bh); may_free &= journal_unmap_buffer(journal, bh, - offset > 0); + partial_page); unlock_buffer(bh); } curr_off = next_off; @@ -2057,7 +2066,7 @@ void journal_invalidatepage(journal_t *journal, } while (bh != head); - if (!offset) { + if (!partial_page) { if (may_free && try_to_free_buffers(page)) J_ASSERT(!page_has_buffers(page)); } diff --git a/fs/jbd2/Kconfig b/fs/jbd2/Kconfig index 69a48c2944da..5a9f5534d57b 100644 --- a/fs/jbd2/Kconfig +++ b/fs/jbd2/Kconfig @@ -20,7 +20,7 @@ config JBD2 config JBD2_DEBUG bool "JBD2 (ext4) debugging support" - depends on JBD2 && DEBUG_FS + depends on JBD2 help If you are using the ext4 journaled file system (or potentially any other filesystem/device using JBD2), this option @@ -29,7 +29,7 @@ config JBD2_DEBUG By default, the debugging output will be turned off. If you select Y here, then you will be able to turn on debugging - with "echo N > /sys/kernel/debug/jbd2/jbd2-debug", where N is a + with "echo N > /sys/module/jbd2/parameters/jbd2_debug", where N is a number between 1 and 5. The higher the number, the more debugging output is generated. To turn debugging off again, do - "echo 0 > /sys/kernel/debug/jbd2/jbd2-debug". + "echo 0 > /sys/module/jbd2/parameters/jbd2_debug". diff --git a/fs/jbd2/checkpoint.c b/fs/jbd2/checkpoint.c index c78841ee81cf..7f34f4716165 100644 --- a/fs/jbd2/checkpoint.c +++ b/fs/jbd2/checkpoint.c @@ -120,8 +120,8 @@ void __jbd2_log_wait_for_space(journal_t *journal) int nblocks, space_left; /* assert_spin_locked(&journal->j_state_lock); */ - nblocks = jbd_space_needed(journal); - while (__jbd2_log_space_left(journal) < nblocks) { + nblocks = jbd2_space_needed(journal); + while (jbd2_log_space_left(journal) < nblocks) { if (journal->j_flags & JBD2_ABORT) return; write_unlock(&journal->j_state_lock); @@ -140,8 +140,8 @@ void __jbd2_log_wait_for_space(journal_t *journal) */ write_lock(&journal->j_state_lock); spin_lock(&journal->j_list_lock); - nblocks = jbd_space_needed(journal); - space_left = __jbd2_log_space_left(journal); + nblocks = jbd2_space_needed(journal); + space_left = jbd2_log_space_left(journal); if (space_left < nblocks) { int chkpt = journal->j_checkpoint_transactions != NULL; tid_t tid = 0; @@ -156,7 +156,15 @@ void __jbd2_log_wait_for_space(journal_t *journal) /* We were able to recover space; yay! */ ; } else if (tid) { + /* + * jbd2_journal_commit_transaction() may want + * to take the checkpoint_mutex if JBD2_FLUSHED + * is set. So we need to temporarily drop it. + */ + mutex_unlock(&journal->j_checkpoint_mutex); jbd2_log_wait_commit(journal, tid); + write_lock(&journal->j_state_lock); + continue; } else { printk(KERN_ERR "%s: needed %d blocks and " "only had %d space available\n", @@ -625,10 +633,6 @@ int __jbd2_journal_remove_checkpoint(struct journal_head *jh) __jbd2_journal_drop_transaction(journal, transaction); jbd2_journal_free_transaction(transaction); - - /* Just in case anybody was waiting for more transactions to be - checkpointed... */ - wake_up(&journal->j_wait_logspace); ret = 1; out: return ret; @@ -690,9 +694,7 @@ void __jbd2_journal_drop_transaction(journal_t *journal, transaction_t *transact J_ASSERT(transaction->t_state == T_FINISHED); J_ASSERT(transaction->t_buffers == NULL); J_ASSERT(transaction->t_forget == NULL); - J_ASSERT(transaction->t_iobuf_list == NULL); J_ASSERT(transaction->t_shadow_list == NULL); - J_ASSERT(transaction->t_log_list == NULL); J_ASSERT(transaction->t_checkpoint_list == NULL); J_ASSERT(transaction->t_checkpoint_io_list == NULL); J_ASSERT(atomic_read(&transaction->t_updates) == 0); diff --git a/fs/jbd2/commit.c b/fs/jbd2/commit.c index 0f53946f13c1..559bec1a37b4 100644 --- a/fs/jbd2/commit.c +++ b/fs/jbd2/commit.c @@ -30,15 +30,22 @@ #include <trace/events/jbd2.h> /* - * Default IO end handler for temporary BJ_IO buffer_heads. + * IO end handler for temporary buffer_heads handling writes to the journal. */ static void journal_end_buffer_io_sync(struct buffer_head *bh, int uptodate) { + struct buffer_head *orig_bh = bh->b_private; + BUFFER_TRACE(bh, ""); if (uptodate) set_buffer_uptodate(bh); else clear_buffer_uptodate(bh); + if (orig_bh) { + clear_bit_unlock(BH_Shadow, &orig_bh->b_state); + smp_mb__after_clear_bit(); + wake_up_bit(&orig_bh->b_state, BH_Shadow); + } unlock_buffer(bh); } @@ -85,8 +92,7 @@ nope: __brelse(bh); } -static void jbd2_commit_block_csum_set(journal_t *j, - struct journal_head *descriptor) +static void jbd2_commit_block_csum_set(journal_t *j, struct buffer_head *bh) { struct commit_header *h; __u32 csum; @@ -94,12 +100,11 @@ static void jbd2_commit_block_csum_set(journal_t *j, if (!JBD2_HAS_INCOMPAT_FEATURE(j, JBD2_FEATURE_INCOMPAT_CSUM_V2)) return; - h = (struct commit_header *)(jh2bh(descriptor)->b_data); + h = (struct commit_header *)(bh->b_data); h->h_chksum_type = 0; h->h_chksum_size = 0; h->h_chksum[0] = 0; - csum = jbd2_chksum(j, j->j_csum_seed, jh2bh(descriptor)->b_data, - j->j_blocksize); + csum = jbd2_chksum(j, j->j_csum_seed, bh->b_data, j->j_blocksize); h->h_chksum[0] = cpu_to_be32(csum); } @@ -116,7 +121,6 @@ static int journal_submit_commit_record(journal_t *journal, struct buffer_head **cbh, __u32 crc32_sum) { - struct journal_head *descriptor; struct commit_header *tmp; struct buffer_head *bh; int ret; @@ -127,12 +131,10 @@ static int journal_submit_commit_record(journal_t *journal, if (is_journal_aborted(journal)) return 0; - descriptor = jbd2_journal_get_descriptor_buffer(journal); - if (!descriptor) + bh = jbd2_journal_get_descriptor_buffer(journal); + if (!bh) return 1; - bh = jh2bh(descriptor); - tmp = (struct commit_header *)bh->b_data; tmp->h_magic = cpu_to_be32(JBD2_MAGIC_NUMBER); tmp->h_blocktype = cpu_to_be32(JBD2_COMMIT_BLOCK); @@ -146,9 +148,9 @@ static int journal_submit_commit_record(journal_t *journal, tmp->h_chksum_size = JBD2_CRC32_CHKSUM_SIZE; tmp->h_chksum[0] = cpu_to_be32(crc32_sum); } - jbd2_commit_block_csum_set(journal, descriptor); + jbd2_commit_block_csum_set(journal, bh); - JBUFFER_TRACE(descriptor, "submit commit block"); + BUFFER_TRACE(bh, "submit commit block"); lock_buffer(bh); clear_buffer_dirty(bh); set_buffer_uptodate(bh); @@ -180,7 +182,6 @@ static int journal_wait_on_commit_record(journal_t *journal, if (unlikely(!buffer_uptodate(bh))) ret = -EIO; put_bh(bh); /* One for getblk() */ - jbd2_journal_put_journal_head(bh2jh(bh)); return ret; } @@ -321,7 +322,7 @@ static void write_tag_block(int tag_bytes, journal_block_tag_t *tag, } static void jbd2_descr_block_csum_set(journal_t *j, - struct journal_head *descriptor) + struct buffer_head *bh) { struct jbd2_journal_block_tail *tail; __u32 csum; @@ -329,12 +330,10 @@ static void jbd2_descr_block_csum_set(journal_t *j, if (!JBD2_HAS_INCOMPAT_FEATURE(j, JBD2_FEATURE_INCOMPAT_CSUM_V2)) return; - tail = (struct jbd2_journal_block_tail *) - (jh2bh(descriptor)->b_data + j->j_blocksize - + tail = (struct jbd2_journal_block_tail *)(bh->b_data + j->j_blocksize - sizeof(struct jbd2_journal_block_tail)); tail->t_checksum = 0; - csum = jbd2_chksum(j, j->j_csum_seed, jh2bh(descriptor)->b_data, - j->j_blocksize); + csum = jbd2_chksum(j, j->j_csum_seed, bh->b_data, j->j_blocksize); tail->t_checksum = cpu_to_be32(csum); } @@ -343,20 +342,21 @@ static void jbd2_block_tag_csum_set(journal_t *j, journal_block_tag_t *tag, { struct page *page = bh->b_page; __u8 *addr; - __u32 csum; + __u32 csum32; if (!JBD2_HAS_INCOMPAT_FEATURE(j, JBD2_FEATURE_INCOMPAT_CSUM_V2)) return; sequence = cpu_to_be32(sequence); addr = kmap_atomic(page); - csum = jbd2_chksum(j, j->j_csum_seed, (__u8 *)&sequence, - sizeof(sequence)); - csum = jbd2_chksum(j, csum, addr + offset_in_page(bh->b_data), - bh->b_size); + csum32 = jbd2_chksum(j, j->j_csum_seed, (__u8 *)&sequence, + sizeof(sequence)); + csum32 = jbd2_chksum(j, csum32, addr + offset_in_page(bh->b_data), + bh->b_size); kunmap_atomic(addr); - tag->t_checksum = cpu_to_be32(csum); + /* We only have space to store the lower 16 bits of the crc32c. */ + tag->t_checksum = cpu_to_be16(csum32); } /* * jbd2_journal_commit_transaction @@ -368,7 +368,8 @@ void jbd2_journal_commit_transaction(journal_t *journal) { struct transaction_stats_s stats; transaction_t *commit_transaction; - struct journal_head *jh, *new_jh, *descriptor; + struct journal_head *jh; + struct buffer_head *descriptor; struct buffer_head **wbuf = journal->j_wbuf; int bufs; int flags; @@ -392,6 +393,8 @@ void jbd2_journal_commit_transaction(journal_t *journal) tid_t first_tid; int update_tail; int csum_size = 0; + LIST_HEAD(io_bufs); + LIST_HEAD(log_bufs); if (JBD2_HAS_INCOMPAT_FEATURE(journal, JBD2_FEATURE_INCOMPAT_CSUM_V2)) csum_size = sizeof(struct jbd2_journal_block_tail); @@ -424,13 +427,13 @@ void jbd2_journal_commit_transaction(journal_t *journal) J_ASSERT(journal->j_committing_transaction == NULL); commit_transaction = journal->j_running_transaction; - J_ASSERT(commit_transaction->t_state == T_RUNNING); trace_jbd2_start_commit(journal, commit_transaction); jbd_debug(1, "JBD2: starting commit of transaction %d\n", commit_transaction->t_tid); write_lock(&journal->j_state_lock); + J_ASSERT(commit_transaction->t_state == T_RUNNING); commit_transaction->t_state = T_LOCKED; trace_jbd2_commit_locking(journal, commit_transaction); @@ -520,6 +523,12 @@ void jbd2_journal_commit_transaction(journal_t *journal) */ jbd2_journal_switch_revoke_table(journal); + /* + * Reserved credits cannot be claimed anymore, free them + */ + atomic_sub(atomic_read(&journal->j_reserved_credits), + &commit_transaction->t_outstanding_credits); + trace_jbd2_commit_flushing(journal, commit_transaction); stats.run.rs_flushing = jiffies; stats.run.rs_locked = jbd2_time_diff(stats.run.rs_locked, @@ -533,7 +542,7 @@ void jbd2_journal_commit_transaction(journal_t *journal) wake_up(&journal->j_wait_transaction_locked); write_unlock(&journal->j_state_lock); - jbd_debug(3, "JBD2: commit phase 2\n"); + jbd_debug(3, "JBD2: commit phase 2a\n"); /* * Now start flushing things to disk, in the order they appear @@ -545,10 +554,10 @@ void jbd2_journal_commit_transaction(journal_t *journal) blk_start_plug(&plug); jbd2_journal_write_revoke_records(journal, commit_transaction, - WRITE_SYNC); + &log_bufs, WRITE_SYNC); blk_finish_plug(&plug); - jbd_debug(3, "JBD2: commit phase 2\n"); + jbd_debug(3, "JBD2: commit phase 2b\n"); /* * Way to go: we have now written out all of the data for a @@ -571,8 +580,8 @@ void jbd2_journal_commit_transaction(journal_t *journal) atomic_read(&commit_transaction->t_outstanding_credits)); err = 0; - descriptor = NULL; bufs = 0; + descriptor = NULL; blk_start_plug(&plug); while (commit_transaction->t_buffers) { @@ -604,8 +613,6 @@ void jbd2_journal_commit_transaction(journal_t *journal) record the metadata buffer. */ if (!descriptor) { - struct buffer_head *bh; - J_ASSERT (bufs == 0); jbd_debug(4, "JBD2: get descriptor\n"); @@ -616,26 +623,26 @@ void jbd2_journal_commit_transaction(journal_t *journal) continue; } - bh = jh2bh(descriptor); jbd_debug(4, "JBD2: got buffer %llu (%p)\n", - (unsigned long long)bh->b_blocknr, bh->b_data); - header = (journal_header_t *)&bh->b_data[0]; + (unsigned long long)descriptor->b_blocknr, + descriptor->b_data); + header = (journal_header_t *)descriptor->b_data; header->h_magic = cpu_to_be32(JBD2_MAGIC_NUMBER); header->h_blocktype = cpu_to_be32(JBD2_DESCRIPTOR_BLOCK); header->h_sequence = cpu_to_be32(commit_transaction->t_tid); - tagp = &bh->b_data[sizeof(journal_header_t)]; - space_left = bh->b_size - sizeof(journal_header_t); + tagp = &descriptor->b_data[sizeof(journal_header_t)]; + space_left = descriptor->b_size - + sizeof(journal_header_t); first_tag = 1; - set_buffer_jwrite(bh); - set_buffer_dirty(bh); - wbuf[bufs++] = bh; + set_buffer_jwrite(descriptor); + set_buffer_dirty(descriptor); + wbuf[bufs++] = descriptor; /* Record it so that we can wait for IO completion later */ - BUFFER_TRACE(bh, "ph3: file as descriptor"); - jbd2_journal_file_buffer(descriptor, commit_transaction, - BJ_LogCtl); + BUFFER_TRACE(descriptor, "ph3: file as descriptor"); + jbd2_file_log_bh(&log_bufs, descriptor); } /* Where is the buffer to be written? */ @@ -658,29 +665,22 @@ void jbd2_journal_commit_transaction(journal_t *journal) /* Bump b_count to prevent truncate from stumbling over the shadowed buffer! @@@ This can go if we ever get - rid of the BJ_IO/BJ_Shadow pairing of buffers. */ + rid of the shadow pairing of buffers. */ atomic_inc(&jh2bh(jh)->b_count); - /* Make a temporary IO buffer with which to write it out - (this will requeue both the metadata buffer and the - temporary IO buffer). new_bh goes on BJ_IO*/ - - set_bit(BH_JWrite, &jh2bh(jh)->b_state); /* - * akpm: jbd2_journal_write_metadata_buffer() sets - * new_bh->b_transaction to commit_transaction. - * We need to clean this up before we release new_bh - * (which is of type BJ_IO) + * Make a temporary IO buffer with which to write it out + * (this will requeue the metadata buffer to BJ_Shadow). */ + set_bit(BH_JWrite, &jh2bh(jh)->b_state); JBUFFER_TRACE(jh, "ph3: write metadata"); flags = jbd2_journal_write_metadata_buffer(commit_transaction, - jh, &new_jh, blocknr); + jh, &wbuf[bufs], blocknr); if (flags < 0) { jbd2_journal_abort(journal, flags); continue; } - set_bit(BH_JWrite, &jh2bh(new_jh)->b_state); - wbuf[bufs++] = jh2bh(new_jh); + jbd2_file_log_bh(&io_bufs, wbuf[bufs]); /* Record the new block's tag in the current descriptor buffer */ @@ -694,10 +694,11 @@ void jbd2_journal_commit_transaction(journal_t *journal) tag = (journal_block_tag_t *) tagp; write_tag_block(tag_bytes, tag, jh2bh(jh)->b_blocknr); tag->t_flags = cpu_to_be16(tag_flag); - jbd2_block_tag_csum_set(journal, tag, jh2bh(new_jh), + jbd2_block_tag_csum_set(journal, tag, wbuf[bufs], commit_transaction->t_tid); tagp += tag_bytes; space_left -= tag_bytes; + bufs++; if (first_tag) { memcpy (tagp, journal->j_uuid, 16); @@ -809,7 +810,7 @@ start_journal_io: the log. Before we can commit it, wait for the IO so far to complete. Control buffers being written are on the transaction's t_log_list queue, and metadata buffers are on - the t_iobuf_list queue. + the io_bufs list. Wait for the buffers in reverse order. That way we are less likely to be woken up until all IOs have completed, and @@ -818,47 +819,33 @@ start_journal_io: jbd_debug(3, "JBD2: commit phase 3\n"); - /* - * akpm: these are BJ_IO, and j_list_lock is not needed. - * See __journal_try_to_free_buffer. - */ -wait_for_iobuf: - while (commit_transaction->t_iobuf_list != NULL) { - struct buffer_head *bh; + while (!list_empty(&io_bufs)) { + struct buffer_head *bh = list_entry(io_bufs.prev, + struct buffer_head, + b_assoc_buffers); - jh = commit_transaction->t_iobuf_list->b_tprev; - bh = jh2bh(jh); - if (buffer_locked(bh)) { - wait_on_buffer(bh); - goto wait_for_iobuf; - } - if (cond_resched()) - goto wait_for_iobuf; + wait_on_buffer(bh); + cond_resched(); if (unlikely(!buffer_uptodate(bh))) err = -EIO; - - clear_buffer_jwrite(bh); - - JBUFFER_TRACE(jh, "ph4: unfile after journal write"); - jbd2_journal_unfile_buffer(journal, jh); + jbd2_unfile_log_bh(bh); /* - * ->t_iobuf_list should contain only dummy buffer_heads - * which were created by jbd2_journal_write_metadata_buffer(). + * The list contains temporary buffer heads created by + * jbd2_journal_write_metadata_buffer(). */ BUFFER_TRACE(bh, "dumping temporary bh"); - jbd2_journal_put_journal_head(jh); __brelse(bh); J_ASSERT_BH(bh, atomic_read(&bh->b_count) == 0); free_buffer_head(bh); - /* We also have to unlock and free the corresponding - shadowed buffer */ + /* We also have to refile the corresponding shadowed buffer */ jh = commit_transaction->t_shadow_list->b_tprev; bh = jh2bh(jh); - clear_bit(BH_JWrite, &bh->b_state); + clear_buffer_jwrite(bh); J_ASSERT_BH(bh, buffer_jbddirty(bh)); + J_ASSERT_BH(bh, !buffer_shadow(bh)); /* The metadata is now released for reuse, but we need to remember it against this transaction so that when @@ -866,14 +853,6 @@ wait_for_iobuf: required. */ JBUFFER_TRACE(jh, "file as BJ_Forget"); jbd2_journal_file_buffer(jh, commit_transaction, BJ_Forget); - /* - * Wake up any transactions which were waiting for this IO to - * complete. The barrier must be here so that changes by - * jbd2_journal_file_buffer() take effect before wake_up_bit() - * does the waitqueue check. - */ - smp_mb(); - wake_up_bit(&bh->b_state, BH_Unshadow); JBUFFER_TRACE(jh, "brelse shadowed buffer"); __brelse(bh); } @@ -883,26 +862,19 @@ wait_for_iobuf: jbd_debug(3, "JBD2: commit phase 4\n"); /* Here we wait for the revoke record and descriptor record buffers */ - wait_for_ctlbuf: - while (commit_transaction->t_log_list != NULL) { + while (!list_empty(&log_bufs)) { struct buffer_head *bh; - jh = commit_transaction->t_log_list->b_tprev; - bh = jh2bh(jh); - if (buffer_locked(bh)) { - wait_on_buffer(bh); - goto wait_for_ctlbuf; - } - if (cond_resched()) - goto wait_for_ctlbuf; + bh = list_entry(log_bufs.prev, struct buffer_head, b_assoc_buffers); + wait_on_buffer(bh); + cond_resched(); if (unlikely(!buffer_uptodate(bh))) err = -EIO; BUFFER_TRACE(bh, "ph5: control buffer writeout done: unfile"); clear_buffer_jwrite(bh); - jbd2_journal_unfile_buffer(journal, jh); - jbd2_journal_put_journal_head(jh); + jbd2_unfile_log_bh(bh); __brelse(bh); /* One for getblk */ /* AKPM: bforget here */ } @@ -952,9 +924,7 @@ wait_for_iobuf: J_ASSERT(list_empty(&commit_transaction->t_inode_list)); J_ASSERT(commit_transaction->t_buffers == NULL); J_ASSERT(commit_transaction->t_checkpoint_list == NULL); - J_ASSERT(commit_transaction->t_iobuf_list == NULL); J_ASSERT(commit_transaction->t_shadow_list == NULL); - J_ASSERT(commit_transaction->t_log_list == NULL); restart_loop: /* diff --git a/fs/jbd2/journal.c b/fs/jbd2/journal.c index 95457576e434..02c7ad9d7a41 100644 --- a/fs/jbd2/journal.c +++ b/fs/jbd2/journal.c @@ -103,6 +103,24 @@ EXPORT_SYMBOL(jbd2_inode_cache); static void __journal_abort_soft (journal_t *journal, int errno); static int jbd2_journal_create_slab(size_t slab_size); +#ifdef CONFIG_JBD2_DEBUG +void __jbd2_debug(int level, const char *file, const char *func, + unsigned int line, const char *fmt, ...) +{ + struct va_format vaf; + va_list args; + + if (level > jbd2_journal_enable_debug) + return; + va_start(args, fmt); + vaf.fmt = fmt; + vaf.va = &args; + printk(KERN_DEBUG "%s: (%s, %u): %pV\n", file, func, line, &vaf); + va_end(args); +} +EXPORT_SYMBOL(__jbd2_debug); +#endif + /* Checksumming functions */ int jbd2_verify_csum_type(journal_t *j, journal_superblock_t *sb) { @@ -310,14 +328,12 @@ static void journal_kill_thread(journal_t *journal) * * If the source buffer has already been modified by a new transaction * since we took the last commit snapshot, we use the frozen copy of - * that data for IO. If we end up using the existing buffer_head's data - * for the write, then we *have* to lock the buffer to prevent anyone - * else from using and possibly modifying it while the IO is in - * progress. + * that data for IO. If we end up using the existing buffer_head's data + * for the write, then we have to make sure nobody modifies it while the + * IO is in progress. do_get_write_access() handles this. * - * The function returns a pointer to the buffer_heads to be used for IO. - * - * We assume that the journal has already been locked in this function. + * The function returns a pointer to the buffer_head to be used for IO. + * * * Return value: * <0: Error @@ -330,15 +346,14 @@ static void journal_kill_thread(journal_t *journal) int jbd2_journal_write_metadata_buffer(transaction_t *transaction, struct journal_head *jh_in, - struct journal_head **jh_out, - unsigned long long blocknr) + struct buffer_head **bh_out, + sector_t blocknr) { int need_copy_out = 0; int done_copy_out = 0; int do_escape = 0; char *mapped_data; struct buffer_head *new_bh; - struct journal_head *new_jh; struct page *new_page; unsigned int new_offset; struct buffer_head *bh_in = jh2bh(jh_in); @@ -368,14 +383,13 @@ retry_alloc: /* keep subsequent assertions sane */ atomic_set(&new_bh->b_count, 1); - new_jh = jbd2_journal_add_journal_head(new_bh); /* This sleeps */ + jbd_lock_bh_state(bh_in); +repeat: /* * If a new transaction has already done a buffer copy-out, then * we use that version of the data for the commit. */ - jbd_lock_bh_state(bh_in); -repeat: if (jh_in->b_frozen_data) { done_copy_out = 1; new_page = virt_to_page(jh_in->b_frozen_data); @@ -415,7 +429,7 @@ repeat: jbd_unlock_bh_state(bh_in); tmp = jbd2_alloc(bh_in->b_size, GFP_NOFS); if (!tmp) { - jbd2_journal_put_journal_head(new_jh); + brelse(new_bh); return -ENOMEM; } jbd_lock_bh_state(bh_in); @@ -426,7 +440,7 @@ repeat: jh_in->b_frozen_data = tmp; mapped_data = kmap_atomic(new_page); - memcpy(tmp, mapped_data + new_offset, jh2bh(jh_in)->b_size); + memcpy(tmp, mapped_data + new_offset, bh_in->b_size); kunmap_atomic(mapped_data); new_page = virt_to_page(tmp); @@ -452,14 +466,14 @@ repeat: } set_bh_page(new_bh, new_page, new_offset); - new_jh->b_transaction = NULL; - new_bh->b_size = jh2bh(jh_in)->b_size; - new_bh->b_bdev = transaction->t_journal->j_dev; + new_bh->b_size = bh_in->b_size; + new_bh->b_bdev = journal->j_dev; new_bh->b_blocknr = blocknr; + new_bh->b_private = bh_in; set_buffer_mapped(new_bh); set_buffer_dirty(new_bh); - *jh_out = new_jh; + *bh_out = new_bh; /* * The to-be-written buffer needs to get moved to the io queue, @@ -470,11 +484,9 @@ repeat: spin_lock(&journal->j_list_lock); __jbd2_journal_file_buffer(jh_in, transaction, BJ_Shadow); spin_unlock(&journal->j_list_lock); + set_buffer_shadow(bh_in); jbd_unlock_bh_state(bh_in); - JBUFFER_TRACE(new_jh, "file as BJ_IO"); - jbd2_journal_file_buffer(new_jh, transaction, BJ_IO); - return do_escape | (done_copy_out << 1); } @@ -484,35 +496,6 @@ repeat: */ /* - * __jbd2_log_space_left: Return the number of free blocks left in the journal. - * - * Called with the journal already locked. - * - * Called under j_state_lock - */ - -int __jbd2_log_space_left(journal_t *journal) -{ - int left = journal->j_free; - - /* assert_spin_locked(&journal->j_state_lock); */ - - /* - * Be pessimistic here about the number of those free blocks which - * might be required for log descriptor control blocks. - */ - -#define MIN_LOG_RESERVED_BLOCKS 32 /* Allow for rounding errors */ - - left -= MIN_LOG_RESERVED_BLOCKS; - - if (left <= 0) - return 0; - left -= (left >> 3); - return left; -} - -/* * Called with j_state_lock locked for writing. * Returns true if a transaction commit was started. */ @@ -564,20 +547,17 @@ int jbd2_log_start_commit(journal_t *journal, tid_t tid) } /* - * Force and wait upon a commit if the calling process is not within - * transaction. This is used for forcing out undo-protected data which contains - * bitmaps, when the fs is running out of space. - * - * We can only force the running transaction if we don't have an active handle; - * otherwise, we will deadlock. - * - * Returns true if a transaction was started. + * Force and wait any uncommitted transactions. We can only force the running + * transaction if we don't have an active handle, otherwise, we will deadlock. + * Returns: <0 in case of error, + * 0 if nothing to commit, + * 1 if transaction was successfully committed. */ -int jbd2_journal_force_commit_nested(journal_t *journal) +static int __jbd2_journal_force_commit(journal_t *journal) { transaction_t *transaction = NULL; tid_t tid; - int need_to_start = 0; + int need_to_start = 0, ret = 0; read_lock(&journal->j_state_lock); if (journal->j_running_transaction && !current->journal_info) { @@ -588,16 +568,53 @@ int jbd2_journal_force_commit_nested(journal_t *journal) transaction = journal->j_committing_transaction; if (!transaction) { + /* Nothing to commit */ read_unlock(&journal->j_state_lock); - return 0; /* Nothing to retry */ + return 0; } - tid = transaction->t_tid; read_unlock(&journal->j_state_lock); if (need_to_start) jbd2_log_start_commit(journal, tid); - jbd2_log_wait_commit(journal, tid); - return 1; + ret = jbd2_log_wait_commit(journal, tid); + if (!ret) + ret = 1; + + return ret; +} + +/** + * Force and wait upon a commit if the calling process is not within + * transaction. This is used for forcing out undo-protected data which contains + * bitmaps, when the fs is running out of space. + * + * @journal: journal to force + * Returns true if progress was made. + */ +int jbd2_journal_force_commit_nested(journal_t *journal) +{ + int ret; + + ret = __jbd2_journal_force_commit(journal); + return ret > 0; +} + +/** + * int journal_force_commit() - force any uncommitted transactions + * @journal: journal to force + * + * Caller want unconditional commit. We can only force the running transaction + * if we don't have an active handle, otherwise, we will deadlock. + */ +int jbd2_journal_force_commit(journal_t *journal) +{ + int ret; + + J_ASSERT(!current->journal_info); + ret = __jbd2_journal_force_commit(journal); + if (ret > 0) + ret = 0; + return ret; } /* @@ -798,7 +815,7 @@ int jbd2_journal_bmap(journal_t *journal, unsigned long blocknr, * But we don't bother doing that, so there will be coherency problems with * mmaps of blockdevs which hold live JBD-controlled filesystems. */ -struct journal_head *jbd2_journal_get_descriptor_buffer(journal_t *journal) +struct buffer_head *jbd2_journal_get_descriptor_buffer(journal_t *journal) { struct buffer_head *bh; unsigned long long blocknr; @@ -817,7 +834,7 @@ struct journal_head *jbd2_journal_get_descriptor_buffer(journal_t *journal) set_buffer_uptodate(bh); unlock_buffer(bh); BUFFER_TRACE(bh, "return this buffer"); - return jbd2_journal_add_journal_head(bh); + return bh; } /* @@ -1062,11 +1079,10 @@ static journal_t * journal_init_common (void) return NULL; init_waitqueue_head(&journal->j_wait_transaction_locked); - init_waitqueue_head(&journal->j_wait_logspace); init_waitqueue_head(&journal->j_wait_done_commit); - init_waitqueue_head(&journal->j_wait_checkpoint); init_waitqueue_head(&journal->j_wait_commit); init_waitqueue_head(&journal->j_wait_updates); + init_waitqueue_head(&journal->j_wait_reserved); mutex_init(&journal->j_barrier); mutex_init(&journal->j_checkpoint_mutex); spin_lock_init(&journal->j_revoke_lock); @@ -1076,6 +1092,7 @@ static journal_t * journal_init_common (void) journal->j_commit_interval = (HZ * JBD2_DEFAULT_MAX_COMMIT_AGE); journal->j_min_batch_time = 0; journal->j_max_batch_time = 15000; /* 15ms */ + atomic_set(&journal->j_reserved_credits, 0); /* The journal is marked for error until we succeed with recovery! */ journal->j_flags = JBD2_ABORT; @@ -1318,6 +1335,7 @@ static int journal_reset(journal_t *journal) static void jbd2_write_superblock(journal_t *journal, int write_op) { struct buffer_head *bh = journal->j_sb_buffer; + journal_superblock_t *sb = journal->j_superblock; int ret; trace_jbd2_write_superblock(journal, write_op); @@ -1339,6 +1357,7 @@ static void jbd2_write_superblock(journal_t *journal, int write_op) clear_buffer_write_io_error(bh); set_buffer_uptodate(bh); } + jbd2_superblock_csum_set(journal, sb); get_bh(bh); bh->b_end_io = end_buffer_write_sync; ret = submit_bh(write_op, bh); @@ -1435,7 +1454,6 @@ void jbd2_journal_update_sb_errno(journal_t *journal) jbd_debug(1, "JBD2: updating superblock error (errno %d)\n", journal->j_errno); sb->s_errno = cpu_to_be32(journal->j_errno); - jbd2_superblock_csum_set(journal, sb); read_unlock(&journal->j_state_lock); jbd2_write_superblock(journal, WRITE_SYNC); @@ -2325,13 +2343,13 @@ static struct journal_head *journal_alloc_journal_head(void) #ifdef CONFIG_JBD2_DEBUG atomic_inc(&nr_journal_heads); #endif - ret = kmem_cache_alloc(jbd2_journal_head_cache, GFP_NOFS); + ret = kmem_cache_zalloc(jbd2_journal_head_cache, GFP_NOFS); if (!ret) { jbd_debug(1, "out of memory for journal_head\n"); pr_notice_ratelimited("ENOMEM in %s, retrying.\n", __func__); while (!ret) { yield(); - ret = kmem_cache_alloc(jbd2_journal_head_cache, GFP_NOFS); + ret = kmem_cache_zalloc(jbd2_journal_head_cache, GFP_NOFS); } } return ret; @@ -2393,10 +2411,8 @@ struct journal_head *jbd2_journal_add_journal_head(struct buffer_head *bh) struct journal_head *new_jh = NULL; repeat: - if (!buffer_jbd(bh)) { + if (!buffer_jbd(bh)) new_jh = journal_alloc_journal_head(); - memset(new_jh, 0, sizeof(*new_jh)); - } jbd_lock_bh_journal_head(bh); if (buffer_jbd(bh)) { diff --git a/fs/jbd2/recovery.c b/fs/jbd2/recovery.c index 626846bac32f..d4851464b57e 100644 --- a/fs/jbd2/recovery.c +++ b/fs/jbd2/recovery.c @@ -399,18 +399,17 @@ static int jbd2_commit_block_csum_verify(journal_t *j, void *buf) static int jbd2_block_tag_csum_verify(journal_t *j, journal_block_tag_t *tag, void *buf, __u32 sequence) { - __u32 provided, calculated; + __u32 csum32; if (!JBD2_HAS_INCOMPAT_FEATURE(j, JBD2_FEATURE_INCOMPAT_CSUM_V2)) return 1; sequence = cpu_to_be32(sequence); - calculated = jbd2_chksum(j, j->j_csum_seed, (__u8 *)&sequence, - sizeof(sequence)); - calculated = jbd2_chksum(j, calculated, buf, j->j_blocksize); - provided = be32_to_cpu(tag->t_checksum); + csum32 = jbd2_chksum(j, j->j_csum_seed, (__u8 *)&sequence, + sizeof(sequence)); + csum32 = jbd2_chksum(j, csum32, buf, j->j_blocksize); - return provided == cpu_to_be32(calculated); + return tag->t_checksum == cpu_to_be16(csum32); } static int do_one_pass(journal_t *journal, diff --git a/fs/jbd2/revoke.c b/fs/jbd2/revoke.c index f30b80b4ce8b..198c9c10276d 100644 --- a/fs/jbd2/revoke.c +++ b/fs/jbd2/revoke.c @@ -122,9 +122,10 @@ struct jbd2_revoke_table_s #ifdef __KERNEL__ static void write_one_revoke_record(journal_t *, transaction_t *, - struct journal_head **, int *, + struct list_head *, + struct buffer_head **, int *, struct jbd2_revoke_record_s *, int); -static void flush_descriptor(journal_t *, struct journal_head *, int, int); +static void flush_descriptor(journal_t *, struct buffer_head *, int, int); #endif /* Utility functions to maintain the revoke table */ @@ -531,9 +532,10 @@ void jbd2_journal_switch_revoke_table(journal_t *journal) */ void jbd2_journal_write_revoke_records(journal_t *journal, transaction_t *transaction, + struct list_head *log_bufs, int write_op) { - struct journal_head *descriptor; + struct buffer_head *descriptor; struct jbd2_revoke_record_s *record; struct jbd2_revoke_table_s *revoke; struct list_head *hash_list; @@ -553,7 +555,7 @@ void jbd2_journal_write_revoke_records(journal_t *journal, while (!list_empty(hash_list)) { record = (struct jbd2_revoke_record_s *) hash_list->next; - write_one_revoke_record(journal, transaction, + write_one_revoke_record(journal, transaction, log_bufs, &descriptor, &offset, record, write_op); count++; @@ -573,13 +575,14 @@ void jbd2_journal_write_revoke_records(journal_t *journal, static void write_one_revoke_record(journal_t *journal, transaction_t *transaction, - struct journal_head **descriptorp, + struct list_head *log_bufs, + struct buffer_head **descriptorp, int *offsetp, struct jbd2_revoke_record_s *record, int write_op) { int csum_size = 0; - struct journal_head *descriptor; + struct buffer_head *descriptor; int offset; journal_header_t *header; @@ -609,26 +612,26 @@ static void write_one_revoke_record(journal_t *journal, descriptor = jbd2_journal_get_descriptor_buffer(journal); if (!descriptor) return; - header = (journal_header_t *) &jh2bh(descriptor)->b_data[0]; + header = (journal_header_t *)descriptor->b_data; header->h_magic = cpu_to_be32(JBD2_MAGIC_NUMBER); header->h_blocktype = cpu_to_be32(JBD2_REVOKE_BLOCK); header->h_sequence = cpu_to_be32(transaction->t_tid); /* Record it so that we can wait for IO completion later */ - JBUFFER_TRACE(descriptor, "file as BJ_LogCtl"); - jbd2_journal_file_buffer(descriptor, transaction, BJ_LogCtl); + BUFFER_TRACE(descriptor, "file in log_bufs"); + jbd2_file_log_bh(log_bufs, descriptor); offset = sizeof(jbd2_journal_revoke_header_t); *descriptorp = descriptor; } if (JBD2_HAS_INCOMPAT_FEATURE(journal, JBD2_FEATURE_INCOMPAT_64BIT)) { - * ((__be64 *)(&jh2bh(descriptor)->b_data[offset])) = + * ((__be64 *)(&descriptor->b_data[offset])) = cpu_to_be64(record->blocknr); offset += 8; } else { - * ((__be32 *)(&jh2bh(descriptor)->b_data[offset])) = + * ((__be32 *)(&descriptor->b_data[offset])) = cpu_to_be32(record->blocknr); offset += 4; } @@ -636,8 +639,7 @@ static void write_one_revoke_record(journal_t *journal, *offsetp = offset; } -static void jbd2_revoke_csum_set(journal_t *j, - struct journal_head *descriptor) +static void jbd2_revoke_csum_set(journal_t *j, struct buffer_head *bh) { struct jbd2_journal_revoke_tail *tail; __u32 csum; @@ -645,12 +647,10 @@ static void jbd2_revoke_csum_set(journal_t *j, if (!JBD2_HAS_INCOMPAT_FEATURE(j, JBD2_FEATURE_INCOMPAT_CSUM_V2)) return; - tail = (struct jbd2_journal_revoke_tail *) - (jh2bh(descriptor)->b_data + j->j_blocksize - + tail = (struct jbd2_journal_revoke_tail *)(bh->b_data + j->j_blocksize - sizeof(struct jbd2_journal_revoke_tail)); tail->r_checksum = 0; - csum = jbd2_chksum(j, j->j_csum_seed, jh2bh(descriptor)->b_data, - j->j_blocksize); + csum = jbd2_chksum(j, j->j_csum_seed, bh->b_data, j->j_blocksize); tail->r_checksum = cpu_to_be32(csum); } @@ -662,25 +662,24 @@ static void jbd2_revoke_csum_set(journal_t *j, */ static void flush_descriptor(journal_t *journal, - struct journal_head *descriptor, + struct buffer_head *descriptor, int offset, int write_op) { jbd2_journal_revoke_header_t *header; - struct buffer_head *bh = jh2bh(descriptor); if (is_journal_aborted(journal)) { - put_bh(bh); + put_bh(descriptor); return; } - header = (jbd2_journal_revoke_header_t *) jh2bh(descriptor)->b_data; + header = (jbd2_journal_revoke_header_t *)descriptor->b_data; header->r_count = cpu_to_be32(offset); jbd2_revoke_csum_set(journal, descriptor); - set_buffer_jwrite(bh); - BUFFER_TRACE(bh, "write"); - set_buffer_dirty(bh); - write_dirty_buffer(bh, write_op); + set_buffer_jwrite(descriptor); + BUFFER_TRACE(descriptor, "write"); + set_buffer_dirty(descriptor); + write_dirty_buffer(descriptor, write_op); } #endif diff --git a/fs/jbd2/transaction.c b/fs/jbd2/transaction.c index 10f524c59ea8..7aa9a32573bb 100644 --- a/fs/jbd2/transaction.c +++ b/fs/jbd2/transaction.c @@ -89,7 +89,8 @@ jbd2_get_transaction(journal_t *journal, transaction_t *transaction) transaction->t_expires = jiffies + journal->j_commit_interval; spin_lock_init(&transaction->t_handle_lock); atomic_set(&transaction->t_updates, 0); - atomic_set(&transaction->t_outstanding_credits, 0); + atomic_set(&transaction->t_outstanding_credits, + atomic_read(&journal->j_reserved_credits)); atomic_set(&transaction->t_handle_count, 0); INIT_LIST_HEAD(&transaction->t_inode_list); INIT_LIST_HEAD(&transaction->t_private_list); @@ -141,6 +142,112 @@ static inline void update_t_max_wait(transaction_t *transaction, } /* + * Wait until running transaction passes T_LOCKED state. Also starts the commit + * if needed. The function expects running transaction to exist and releases + * j_state_lock. + */ +static void wait_transaction_locked(journal_t *journal) + __releases(journal->j_state_lock) +{ + DEFINE_WAIT(wait); + int need_to_start; + tid_t tid = journal->j_running_transaction->t_tid; + + prepare_to_wait(&journal->j_wait_transaction_locked, &wait, + TASK_UNINTERRUPTIBLE); + need_to_start = !tid_geq(journal->j_commit_request, tid); + read_unlock(&journal->j_state_lock); + if (need_to_start) + jbd2_log_start_commit(journal, tid); + schedule(); + finish_wait(&journal->j_wait_transaction_locked, &wait); +} + +static void sub_reserved_credits(journal_t *journal, int blocks) +{ + atomic_sub(blocks, &journal->j_reserved_credits); + wake_up(&journal->j_wait_reserved); +} + +/* + * Wait until we can add credits for handle to the running transaction. Called + * with j_state_lock held for reading. Returns 0 if handle joined the running + * transaction. Returns 1 if we had to wait, j_state_lock is dropped, and + * caller must retry. + */ +static int add_transaction_credits(journal_t *journal, int blocks, + int rsv_blocks) +{ + transaction_t *t = journal->j_running_transaction; + int needed; + int total = blocks + rsv_blocks; + + /* + * If the current transaction is locked down for commit, wait + * for the lock to be released. + */ + if (t->t_state == T_LOCKED) { + wait_transaction_locked(journal); + return 1; + } + + /* + * If there is not enough space left in the log to write all + * potential buffers requested by this operation, we need to + * stall pending a log checkpoint to free some more log space. + */ + needed = atomic_add_return(total, &t->t_outstanding_credits); + if (needed > journal->j_max_transaction_buffers) { + /* + * If the current transaction is already too large, + * then start to commit it: we can then go back and + * attach this handle to a new transaction. + */ + atomic_sub(total, &t->t_outstanding_credits); + wait_transaction_locked(journal); + return 1; + } + + /* + * The commit code assumes that it can get enough log space + * without forcing a checkpoint. This is *critical* for + * correctness: a checkpoint of a buffer which is also + * associated with a committing transaction creates a deadlock, + * so commit simply cannot force through checkpoints. + * + * We must therefore ensure the necessary space in the journal + * *before* starting to dirty potentially checkpointed buffers + * in the new transaction. + */ + if (jbd2_log_space_left(journal) < jbd2_space_needed(journal)) { + atomic_sub(total, &t->t_outstanding_credits); + read_unlock(&journal->j_state_lock); + write_lock(&journal->j_state_lock); + if (jbd2_log_space_left(journal) < jbd2_space_needed(journal)) + __jbd2_log_wait_for_space(journal); + write_unlock(&journal->j_state_lock); + return 1; + } + + /* No reservation? We are done... */ + if (!rsv_blocks) + return 0; + + needed = atomic_add_return(rsv_blocks, &journal->j_reserved_credits); + /* We allow at most half of a transaction to be reserved */ + if (needed > journal->j_max_transaction_buffers / 2) { + sub_reserved_credits(journal, rsv_blocks); + atomic_sub(total, &t->t_outstanding_credits); + read_unlock(&journal->j_state_lock); + wait_event(journal->j_wait_reserved, + atomic_read(&journal->j_reserved_credits) + rsv_blocks + <= journal->j_max_transaction_buffers / 2); + return 1; + } + return 0; +} + +/* * start_this_handle: Given a handle, deal with any locking or stalling * needed to make sure that there is enough journal space for the handle * to begin. Attach the handle to a transaction and set up the @@ -151,18 +258,24 @@ static int start_this_handle(journal_t *journal, handle_t *handle, gfp_t gfp_mask) { transaction_t *transaction, *new_transaction = NULL; - tid_t tid; - int needed, need_to_start; - int nblocks = handle->h_buffer_credits; + int blocks = handle->h_buffer_credits; + int rsv_blocks = 0; unsigned long ts = jiffies; - if (nblocks > journal->j_max_transaction_buffers) { + /* + * 1/2 of transaction can be reserved so we can practically handle + * only 1/2 of maximum transaction size per operation + */ + if (WARN_ON(blocks > journal->j_max_transaction_buffers / 2)) { printk(KERN_ERR "JBD2: %s wants too many credits (%d > %d)\n", - current->comm, nblocks, - journal->j_max_transaction_buffers); + current->comm, blocks, + journal->j_max_transaction_buffers / 2); return -ENOSPC; } + if (handle->h_rsv_handle) + rsv_blocks = handle->h_rsv_handle->h_buffer_credits; + alloc_transaction: if (!journal->j_running_transaction) { new_transaction = kmem_cache_zalloc(transaction_cache, @@ -199,8 +312,12 @@ repeat: return -EROFS; } - /* Wait on the journal's transaction barrier if necessary */ - if (journal->j_barrier_count) { + /* + * Wait on the journal's transaction barrier if necessary. Specifically + * we allow reserved handles to proceed because otherwise commit could + * deadlock on page writeback not being able to complete. + */ + if (!handle->h_reserved && journal->j_barrier_count) { read_unlock(&journal->j_state_lock); wait_event(journal->j_wait_transaction_locked, journal->j_barrier_count == 0); @@ -213,7 +330,7 @@ repeat: goto alloc_transaction; write_lock(&journal->j_state_lock); if (!journal->j_running_transaction && - !journal->j_barrier_count) { + (handle->h_reserved || !journal->j_barrier_count)) { jbd2_get_transaction(journal, new_transaction); new_transaction = NULL; } @@ -223,85 +340,18 @@ repeat: transaction = journal->j_running_transaction; - /* - * If the current transaction is locked down for commit, wait for the - * lock to be released. - */ - if (transaction->t_state == T_LOCKED) { - DEFINE_WAIT(wait); - - prepare_to_wait(&journal->j_wait_transaction_locked, - &wait, TASK_UNINTERRUPTIBLE); - read_unlock(&journal->j_state_lock); - schedule(); - finish_wait(&journal->j_wait_transaction_locked, &wait); - goto repeat; - } - - /* - * If there is not enough space left in the log to write all potential - * buffers requested by this operation, we need to stall pending a log - * checkpoint to free some more log space. - */ - needed = atomic_add_return(nblocks, - &transaction->t_outstanding_credits); - - if (needed > journal->j_max_transaction_buffers) { + if (!handle->h_reserved) { + /* We may have dropped j_state_lock - restart in that case */ + if (add_transaction_credits(journal, blocks, rsv_blocks)) + goto repeat; + } else { /* - * If the current transaction is already too large, then start - * to commit it: we can then go back and attach this handle to - * a new transaction. + * We have handle reserved so we are allowed to join T_LOCKED + * transaction and we don't have to check for transaction size + * and journal space. */ - DEFINE_WAIT(wait); - - jbd_debug(2, "Handle %p starting new commit...\n", handle); - atomic_sub(nblocks, &transaction->t_outstanding_credits); - prepare_to_wait(&journal->j_wait_transaction_locked, &wait, - TASK_UNINTERRUPTIBLE); - tid = transaction->t_tid; - need_to_start = !tid_geq(journal->j_commit_request, tid); - read_unlock(&journal->j_state_lock); - if (need_to_start) - jbd2_log_start_commit(journal, tid); - schedule(); - finish_wait(&journal->j_wait_transaction_locked, &wait); - goto repeat; - } - - /* - * The commit code assumes that it can get enough log space - * without forcing a checkpoint. This is *critical* for - * correctness: a checkpoint of a buffer which is also - * associated with a committing transaction creates a deadlock, - * so commit simply cannot force through checkpoints. - * - * We must therefore ensure the necessary space in the journal - * *before* starting to dirty potentially checkpointed buffers - * in the new transaction. - * - * The worst part is, any transaction currently committing can - * reduce the free space arbitrarily. Be careful to account for - * those buffers when checkpointing. - */ - - /* - * @@@ AKPM: This seems rather over-defensive. We're giving commit - * a _lot_ of headroom: 1/4 of the journal plus the size of - * the committing transaction. Really, we only need to give it - * committing_transaction->t_outstanding_credits plus "enough" for - * the log control blocks. - * Also, this test is inconsistent with the matching one in - * jbd2_journal_extend(). - */ - if (__jbd2_log_space_left(journal) < jbd_space_needed(journal)) { - jbd_debug(2, "Handle %p waiting for checkpoint...\n", handle); - atomic_sub(nblocks, &transaction->t_outstanding_credits); - read_unlock(&journal->j_state_lock); - write_lock(&journal->j_state_lock); - if (__jbd2_log_space_left(journal) < jbd_space_needed(journal)) - __jbd2_log_wait_for_space(journal); - write_unlock(&journal->j_state_lock); - goto repeat; + sub_reserved_credits(journal, blocks); + handle->h_reserved = 0; } /* OK, account for the buffers that this operation expects to @@ -309,15 +359,16 @@ repeat: */ update_t_max_wait(transaction, ts); handle->h_transaction = transaction; - handle->h_requested_credits = nblocks; + handle->h_requested_credits = blocks; handle->h_start_jiffies = jiffies; atomic_inc(&transaction->t_updates); atomic_inc(&transaction->t_handle_count); - jbd_debug(4, "Handle %p given %d credits (total %d, free %d)\n", - handle, nblocks, + jbd_debug(4, "Handle %p given %d credits (total %d, free %lu)\n", + handle, blocks, atomic_read(&transaction->t_outstanding_credits), - __jbd2_log_space_left(journal)); + jbd2_log_space_left(journal)); read_unlock(&journal->j_state_lock); + current->journal_info = handle; lock_map_acquire(&handle->h_lockdep_map); jbd2_journal_free_transaction(new_transaction); @@ -348,16 +399,21 @@ static handle_t *new_handle(int nblocks) * * We make sure that the transaction can guarantee at least nblocks of * modified buffers in the log. We block until the log can guarantee - * that much space. - * - * This function is visible to journal users (like ext3fs), so is not - * called with the journal already locked. + * that much space. Additionally, if rsv_blocks > 0, we also create another + * handle with rsv_blocks reserved blocks in the journal. This handle is + * is stored in h_rsv_handle. It is not attached to any particular transaction + * and thus doesn't block transaction commit. If the caller uses this reserved + * handle, it has to set h_rsv_handle to NULL as otherwise jbd2_journal_stop() + * on the parent handle will dispose the reserved one. Reserved handle has to + * be converted to a normal handle using jbd2_journal_start_reserved() before + * it can be used. * * Return a pointer to a newly allocated handle, or an ERR_PTR() value * on failure. */ -handle_t *jbd2__journal_start(journal_t *journal, int nblocks, gfp_t gfp_mask, - unsigned int type, unsigned int line_no) +handle_t *jbd2__journal_start(journal_t *journal, int nblocks, int rsv_blocks, + gfp_t gfp_mask, unsigned int type, + unsigned int line_no) { handle_t *handle = journal_current_handle(); int err; @@ -374,13 +430,24 @@ handle_t *jbd2__journal_start(journal_t *journal, int nblocks, gfp_t gfp_mask, handle = new_handle(nblocks); if (!handle) return ERR_PTR(-ENOMEM); + if (rsv_blocks) { + handle_t *rsv_handle; - current->journal_info = handle; + rsv_handle = new_handle(rsv_blocks); + if (!rsv_handle) { + jbd2_free_handle(handle); + return ERR_PTR(-ENOMEM); + } + rsv_handle->h_reserved = 1; + rsv_handle->h_journal = journal; + handle->h_rsv_handle = rsv_handle; + } err = start_this_handle(journal, handle, gfp_mask); if (err < 0) { + if (handle->h_rsv_handle) + jbd2_free_handle(handle->h_rsv_handle); jbd2_free_handle(handle); - current->journal_info = NULL; return ERR_PTR(err); } handle->h_type = type; @@ -395,10 +462,65 @@ EXPORT_SYMBOL(jbd2__journal_start); handle_t *jbd2_journal_start(journal_t *journal, int nblocks) { - return jbd2__journal_start(journal, nblocks, GFP_NOFS, 0, 0); + return jbd2__journal_start(journal, nblocks, 0, GFP_NOFS, 0, 0); } EXPORT_SYMBOL(jbd2_journal_start); +void jbd2_journal_free_reserved(handle_t *handle) +{ + journal_t *journal = handle->h_journal; + + WARN_ON(!handle->h_reserved); + sub_reserved_credits(journal, handle->h_buffer_credits); + jbd2_free_handle(handle); +} +EXPORT_SYMBOL(jbd2_journal_free_reserved); + +/** + * int jbd2_journal_start_reserved(handle_t *handle) - start reserved handle + * @handle: handle to start + * + * Start handle that has been previously reserved with jbd2_journal_reserve(). + * This attaches @handle to the running transaction (or creates one if there's + * not transaction running). Unlike jbd2_journal_start() this function cannot + * block on journal commit, checkpointing, or similar stuff. It can block on + * memory allocation or frozen journal though. + * + * Return 0 on success, non-zero on error - handle is freed in that case. + */ +int jbd2_journal_start_reserved(handle_t *handle, unsigned int type, + unsigned int line_no) +{ + journal_t *journal = handle->h_journal; + int ret = -EIO; + + if (WARN_ON(!handle->h_reserved)) { + /* Someone passed in normal handle? Just stop it. */ + jbd2_journal_stop(handle); + return ret; + } + /* + * Usefulness of mixing of reserved and unreserved handles is + * questionable. So far nobody seems to need it so just error out. + */ + if (WARN_ON(current->journal_info)) { + jbd2_journal_free_reserved(handle); + return ret; + } + + handle->h_journal = NULL; + /* + * GFP_NOFS is here because callers are likely from writeback or + * similarly constrained call sites + */ + ret = start_this_handle(journal, handle, GFP_NOFS); + if (ret < 0) + jbd2_journal_free_reserved(handle); + handle->h_type = type; + handle->h_line_no = line_no; + return ret; +} +EXPORT_SYMBOL(jbd2_journal_start_reserved); /** * int jbd2_journal_extend() - extend buffer credits. @@ -423,49 +545,53 @@ EXPORT_SYMBOL(jbd2_journal_start); int jbd2_journal_extend(handle_t *handle, int nblocks) { transaction_t *transaction = handle->h_transaction; - journal_t *journal = transaction->t_journal; + journal_t *journal; int result; int wanted; - result = -EIO; + WARN_ON(!transaction); if (is_handle_aborted(handle)) - goto out; + return -EROFS; + journal = transaction->t_journal; result = 1; read_lock(&journal->j_state_lock); /* Don't extend a locked-down transaction! */ - if (handle->h_transaction->t_state != T_RUNNING) { + if (transaction->t_state != T_RUNNING) { jbd_debug(3, "denied handle %p %d blocks: " "transaction not running\n", handle, nblocks); goto error_out; } spin_lock(&transaction->t_handle_lock); - wanted = atomic_read(&transaction->t_outstanding_credits) + nblocks; + wanted = atomic_add_return(nblocks, + &transaction->t_outstanding_credits); if (wanted > journal->j_max_transaction_buffers) { jbd_debug(3, "denied handle %p %d blocks: " "transaction too large\n", handle, nblocks); + atomic_sub(nblocks, &transaction->t_outstanding_credits); goto unlock; } - if (wanted > __jbd2_log_space_left(journal)) { + if (wanted + (wanted >> JBD2_CONTROL_BLOCKS_SHIFT) > + jbd2_log_space_left(journal)) { jbd_debug(3, "denied handle %p %d blocks: " "insufficient log space\n", handle, nblocks); + atomic_sub(nblocks, &transaction->t_outstanding_credits); goto unlock; } trace_jbd2_handle_extend(journal->j_fs_dev->bd_dev, - handle->h_transaction->t_tid, + transaction->t_tid, handle->h_type, handle->h_line_no, handle->h_buffer_credits, nblocks); handle->h_buffer_credits += nblocks; handle->h_requested_credits += nblocks; - atomic_add(nblocks, &transaction->t_outstanding_credits); result = 0; jbd_debug(3, "extended handle %p by %d\n", handle, nblocks); @@ -473,7 +599,6 @@ unlock: spin_unlock(&transaction->t_handle_lock); error_out: read_unlock(&journal->j_state_lock); -out: return result; } @@ -490,19 +615,22 @@ out: * to a running handle, a call to jbd2_journal_restart will commit the * handle's transaction so far and reattach the handle to a new * transaction capabable of guaranteeing the requested number of - * credits. + * credits. We preserve reserved handle if there's any attached to the + * passed in handle. */ int jbd2__journal_restart(handle_t *handle, int nblocks, gfp_t gfp_mask) { transaction_t *transaction = handle->h_transaction; - journal_t *journal = transaction->t_journal; + journal_t *journal; tid_t tid; int need_to_start, ret; + WARN_ON(!transaction); /* If we've had an abort of any type, don't even think about * actually doing the restart! */ if (is_handle_aborted(handle)) return 0; + journal = transaction->t_journal; /* * First unlink the handle from its current transaction, and start the @@ -515,12 +643,18 @@ int jbd2__journal_restart(handle_t *handle, int nblocks, gfp_t gfp_mask) spin_lock(&transaction->t_handle_lock); atomic_sub(handle->h_buffer_credits, &transaction->t_outstanding_credits); + if (handle->h_rsv_handle) { + sub_reserved_credits(journal, + handle->h_rsv_handle->h_buffer_credits); + } if (atomic_dec_and_test(&transaction->t_updates)) wake_up(&journal->j_wait_updates); + tid = transaction->t_tid; spin_unlock(&transaction->t_handle_lock); + handle->h_transaction = NULL; + current->journal_info = NULL; jbd_debug(2, "restarting handle %p\n", handle); - tid = transaction->t_tid; need_to_start = !tid_geq(journal->j_commit_request, tid); read_unlock(&journal->j_state_lock); if (need_to_start) @@ -557,6 +691,14 @@ void jbd2_journal_lock_updates(journal_t *journal) write_lock(&journal->j_state_lock); ++journal->j_barrier_count; + /* Wait until there are no reserved handles */ + if (atomic_read(&journal->j_reserved_credits)) { + write_unlock(&journal->j_state_lock); + wait_event(journal->j_wait_reserved, + atomic_read(&journal->j_reserved_credits) == 0); + write_lock(&journal->j_state_lock); + } + /* Wait until there are no running updates */ while (1) { transaction_t *transaction = journal->j_running_transaction; @@ -619,6 +761,12 @@ static void warn_dirty_buffer(struct buffer_head *bh) bdevname(bh->b_bdev, b), (unsigned long long)bh->b_blocknr); } +static int sleep_on_shadow_bh(void *word) +{ + io_schedule(); + return 0; +} + /* * If the buffer is already part of the current transaction, then there * is nothing we need to do. If it is already part of a prior @@ -634,17 +782,16 @@ do_get_write_access(handle_t *handle, struct journal_head *jh, int force_copy) { struct buffer_head *bh; - transaction_t *transaction; + transaction_t *transaction = handle->h_transaction; journal_t *journal; int error; char *frozen_buffer = NULL; int need_copy = 0; unsigned long start_lock, time_lock; + WARN_ON(!transaction); if (is_handle_aborted(handle)) return -EROFS; - - transaction = handle->h_transaction; journal = transaction->t_journal; jbd_debug(5, "journal_head %p, force_copy %d\n", jh, force_copy); @@ -754,41 +901,29 @@ repeat: * journaled. If the primary copy is already going to * disk then we cannot do copy-out here. */ - if (jh->b_jlist == BJ_Shadow) { - DEFINE_WAIT_BIT(wait, &bh->b_state, BH_Unshadow); - wait_queue_head_t *wqh; - - wqh = bit_waitqueue(&bh->b_state, BH_Unshadow); - + if (buffer_shadow(bh)) { JBUFFER_TRACE(jh, "on shadow: sleep"); jbd_unlock_bh_state(bh); - /* commit wakes up all shadow buffers after IO */ - for ( ; ; ) { - prepare_to_wait(wqh, &wait.wait, - TASK_UNINTERRUPTIBLE); - if (jh->b_jlist != BJ_Shadow) - break; - schedule(); - } - finish_wait(wqh, &wait.wait); + wait_on_bit(&bh->b_state, BH_Shadow, + sleep_on_shadow_bh, TASK_UNINTERRUPTIBLE); goto repeat; } - /* Only do the copy if the currently-owning transaction - * still needs it. If it is on the Forget list, the - * committing transaction is past that stage. The - * buffer had better remain locked during the kmalloc, - * but that should be true --- we hold the journal lock - * still and the buffer is already on the BUF_JOURNAL - * list so won't be flushed. + /* + * Only do the copy if the currently-owning transaction still + * needs it. If buffer isn't on BJ_Metadata list, the + * committing transaction is past that stage (here we use the + * fact that BH_Shadow is set under bh_state lock together with + * refiling to BJ_Shadow list and at this point we know the + * buffer doesn't have BH_Shadow set). * * Subtle point, though: if this is a get_undo_access, * then we will be relying on the frozen_data to contain * the new value of the committed_data record after the * transaction, so we HAVE to force the frozen_data copy - * in that case. */ - - if (jh->b_jlist != BJ_Forget || force_copy) { + * in that case. + */ + if (jh->b_jlist == BJ_Metadata || force_copy) { JBUFFER_TRACE(jh, "generate frozen data"); if (!frozen_buffer) { JBUFFER_TRACE(jh, "allocate memory for buffer"); @@ -915,14 +1050,16 @@ int jbd2_journal_get_write_access(handle_t *handle, struct buffer_head *bh) int jbd2_journal_get_create_access(handle_t *handle, struct buffer_head *bh) { transaction_t *transaction = handle->h_transaction; - journal_t *journal = transaction->t_journal; + journal_t *journal; struct journal_head *jh = jbd2_journal_add_journal_head(bh); int err; jbd_debug(5, "journal_head %p\n", jh); + WARN_ON(!transaction); err = -EROFS; if (is_handle_aborted(handle)) goto out; + journal = transaction->t_journal; err = 0; JBUFFER_TRACE(jh, "entry"); @@ -1128,12 +1265,14 @@ void jbd2_buffer_abort_trigger(struct journal_head *jh, int jbd2_journal_dirty_metadata(handle_t *handle, struct buffer_head *bh) { transaction_t *transaction = handle->h_transaction; - journal_t *journal = transaction->t_journal; + journal_t *journal; struct journal_head *jh; int ret = 0; + WARN_ON(!transaction); if (is_handle_aborted(handle)) - goto out; + return -EROFS; + journal = transaction->t_journal; jh = jbd2_journal_grab_journal_head(bh); if (!jh) { ret = -EUCLEAN; @@ -1227,7 +1366,7 @@ int jbd2_journal_dirty_metadata(handle_t *handle, struct buffer_head *bh) JBUFFER_TRACE(jh, "file as BJ_Metadata"); spin_lock(&journal->j_list_lock); - __jbd2_journal_file_buffer(jh, handle->h_transaction, BJ_Metadata); + __jbd2_journal_file_buffer(jh, transaction, BJ_Metadata); spin_unlock(&journal->j_list_lock); out_unlock_bh: jbd_unlock_bh_state(bh); @@ -1258,12 +1397,17 @@ out: int jbd2_journal_forget (handle_t *handle, struct buffer_head *bh) { transaction_t *transaction = handle->h_transaction; - journal_t *journal = transaction->t_journal; + journal_t *journal; struct journal_head *jh; int drop_reserve = 0; int err = 0; int was_modified = 0; + WARN_ON(!transaction); + if (is_handle_aborted(handle)) + return -EROFS; + journal = transaction->t_journal; + BUFFER_TRACE(bh, "entry"); jbd_lock_bh_state(bh); @@ -1290,7 +1434,7 @@ int jbd2_journal_forget (handle_t *handle, struct buffer_head *bh) */ jh->b_modified = 0; - if (jh->b_transaction == handle->h_transaction) { + if (jh->b_transaction == transaction) { J_ASSERT_JH(jh, !jh->b_frozen_data); /* If we are forgetting a buffer which is already part @@ -1385,19 +1529,21 @@ drop: int jbd2_journal_stop(handle_t *handle) { transaction_t *transaction = handle->h_transaction; - journal_t *journal = transaction->t_journal; - int err, wait_for_commit = 0; + journal_t *journal; + int err = 0, wait_for_commit = 0; tid_t tid; pid_t pid; + if (!transaction) + goto free_and_exit; + journal = transaction->t_journal; + J_ASSERT(journal_current_handle() == handle); if (is_handle_aborted(handle)) err = -EIO; - else { + else J_ASSERT(atomic_read(&transaction->t_updates) > 0); - err = 0; - } if (--handle->h_ref > 0) { jbd_debug(4, "h_ref %d -> %d\n", handle->h_ref + 1, @@ -1407,7 +1553,7 @@ int jbd2_journal_stop(handle_t *handle) jbd_debug(4, "Handle %p going down\n", handle); trace_jbd2_handle_stats(journal->j_fs_dev->bd_dev, - handle->h_transaction->t_tid, + transaction->t_tid, handle->h_type, handle->h_line_no, jiffies - handle->h_start_jiffies, handle->h_sync, handle->h_requested_credits, @@ -1518,33 +1664,13 @@ int jbd2_journal_stop(handle_t *handle) lock_map_release(&handle->h_lockdep_map); + if (handle->h_rsv_handle) + jbd2_journal_free_reserved(handle->h_rsv_handle); +free_and_exit: jbd2_free_handle(handle); return err; } -/** - * int jbd2_journal_force_commit() - force any uncommitted transactions - * @journal: journal to force - * - * For synchronous operations: force any uncommitted transactions - * to disk. May seem kludgy, but it reuses all the handle batching - * code in a very simple manner. - */ -int jbd2_journal_force_commit(journal_t *journal) -{ - handle_t *handle; - int ret; - - handle = jbd2_journal_start(journal, 1); - if (IS_ERR(handle)) { - ret = PTR_ERR(handle); - } else { - handle->h_sync = 1; - ret = jbd2_journal_stop(handle); - } - return ret; -} - /* * * List management code snippets: various functions for manipulating the @@ -1601,10 +1727,10 @@ __blist_del_buffer(struct journal_head **list, struct journal_head *jh) * Remove a buffer from the appropriate transaction list. * * Note that this function can *change* the value of - * bh->b_transaction->t_buffers, t_forget, t_iobuf_list, t_shadow_list, - * t_log_list or t_reserved_list. If the caller is holding onto a copy of one - * of these pointers, it could go bad. Generally the caller needs to re-read - * the pointer from the transaction_t. + * bh->b_transaction->t_buffers, t_forget, t_shadow_list, t_log_list or + * t_reserved_list. If the caller is holding onto a copy of one of these + * pointers, it could go bad. Generally the caller needs to re-read the + * pointer from the transaction_t. * * Called under j_list_lock. */ @@ -1634,15 +1760,9 @@ static void __jbd2_journal_temp_unlink_buffer(struct journal_head *jh) case BJ_Forget: list = &transaction->t_forget; break; - case BJ_IO: - list = &transaction->t_iobuf_list; - break; case BJ_Shadow: list = &transaction->t_shadow_list; break; - case BJ_LogCtl: - list = &transaction->t_log_list; - break; case BJ_Reserved: list = &transaction->t_reserved_list; break; @@ -2034,18 +2154,23 @@ zap_buffer_unlocked: * void jbd2_journal_invalidatepage() * @journal: journal to use for flush... * @page: page to flush - * @offset: length of page to invalidate. + * @offset: start of the range to invalidate + * @length: length of the range to invalidate * - * Reap page buffers containing data after offset in page. Can return -EBUSY - * if buffers are part of the committing transaction and the page is straddling - * i_size. Caller then has to wait for current commit and try again. + * Reap page buffers containing data after in the specified range in page. + * Can return -EBUSY if buffers are part of the committing transaction and + * the page is straddling i_size. Caller then has to wait for current commit + * and try again. */ int jbd2_journal_invalidatepage(journal_t *journal, struct page *page, - unsigned long offset) + unsigned int offset, + unsigned int length) { struct buffer_head *head, *bh, *next; + unsigned int stop = offset + length; unsigned int curr_off = 0; + int partial_page = (offset || length < PAGE_CACHE_SIZE); int may_free = 1; int ret = 0; @@ -2054,6 +2179,8 @@ int jbd2_journal_invalidatepage(journal_t *journal, if (!page_has_buffers(page)) return 0; + BUG_ON(stop > PAGE_CACHE_SIZE || stop < length); + /* We will potentially be playing with lists other than just the * data lists (especially for journaled data mode), so be * cautious in our locking. */ @@ -2063,10 +2190,13 @@ int jbd2_journal_invalidatepage(journal_t *journal, unsigned int next_off = curr_off + bh->b_size; next = bh->b_this_page; + if (next_off > stop) + return 0; + if (offset <= curr_off) { /* This block is wholly outside the truncation point */ lock_buffer(bh); - ret = journal_unmap_buffer(journal, bh, offset > 0); + ret = journal_unmap_buffer(journal, bh, partial_page); unlock_buffer(bh); if (ret < 0) return ret; @@ -2077,7 +2207,7 @@ int jbd2_journal_invalidatepage(journal_t *journal, } while (bh != head); - if (!offset) { + if (!partial_page) { if (may_free && try_to_free_buffers(page)) J_ASSERT(!page_has_buffers(page)); } @@ -2138,15 +2268,9 @@ void __jbd2_journal_file_buffer(struct journal_head *jh, case BJ_Forget: list = &transaction->t_forget; break; - case BJ_IO: - list = &transaction->t_iobuf_list; - break; case BJ_Shadow: list = &transaction->t_shadow_list; break; - case BJ_LogCtl: - list = &transaction->t_log_list; - break; case BJ_Reserved: list = &transaction->t_reserved_list; break; @@ -2248,10 +2372,12 @@ void jbd2_journal_refile_buffer(journal_t *journal, struct journal_head *jh) int jbd2_journal_file_inode(handle_t *handle, struct jbd2_inode *jinode) { transaction_t *transaction = handle->h_transaction; - journal_t *journal = transaction->t_journal; + journal_t *journal; + WARN_ON(!transaction); if (is_handle_aborted(handle)) - return -EIO; + return -EROFS; + journal = transaction->t_journal; jbd_debug(4, "Adding inode %lu, tid:%d\n", jinode->i_vfs_inode->i_ino, transaction->t_tid); diff --git a/fs/jfs/jfs_metapage.c b/fs/jfs/jfs_metapage.c index 6740d34cd82b..9e3aaff11f89 100644 --- a/fs/jfs/jfs_metapage.c +++ b/fs/jfs/jfs_metapage.c @@ -571,9 +571,10 @@ static int metapage_releasepage(struct page *page, gfp_t gfp_mask) return ret; } -static void metapage_invalidatepage(struct page *page, unsigned long offset) +static void metapage_invalidatepage(struct page *page, unsigned int offset, + unsigned int length) { - BUG_ON(offset); + BUG_ON(offset || length < PAGE_CACHE_SIZE); BUG_ON(PageWriteback(page)); diff --git a/fs/logfs/file.c b/fs/logfs/file.c index c2219a6dd3c8..57914fc32b62 100644 --- a/fs/logfs/file.c +++ b/fs/logfs/file.c @@ -159,7 +159,8 @@ static int logfs_writepage(struct page *page, struct writeback_control *wbc) return __logfs_writepage(page); } -static void logfs_invalidatepage(struct page *page, unsigned long offset) +static void logfs_invalidatepage(struct page *page, unsigned int offset, + unsigned int length) { struct logfs_block *block = logfs_block(page); diff --git a/fs/logfs/segment.c b/fs/logfs/segment.c index 038da0991794..d448a777166b 100644 --- a/fs/logfs/segment.c +++ b/fs/logfs/segment.c @@ -884,7 +884,8 @@ static struct logfs_area *alloc_area(struct super_block *sb) return area; } -static void map_invalidatepage(struct page *page, unsigned long l) +static void map_invalidatepage(struct page *page, unsigned int o, + unsigned int l) { return; } diff --git a/fs/nfs/file.c b/fs/nfs/file.c index a87a44f84113..6b4a79f4ad1d 100644 --- a/fs/nfs/file.c +++ b/fs/nfs/file.c @@ -451,11 +451,13 @@ static int nfs_write_end(struct file *file, struct address_space *mapping, * - Called if either PG_private or PG_fscache is set on the page * - Caller holds page lock */ -static void nfs_invalidate_page(struct page *page, unsigned long offset) +static void nfs_invalidate_page(struct page *page, unsigned int offset, + unsigned int length) { - dfprintk(PAGECACHE, "NFS: invalidate_page(%p, %lu)\n", page, offset); + dfprintk(PAGECACHE, "NFS: invalidate_page(%p, %u, %u)\n", + page, offset, length); - if (offset != 0) + if (offset != 0 || length < PAGE_CACHE_SIZE) return; /* Cancel any unstarted writes on this page */ nfs_wb_page_cancel(page_file_mapping(page)->host, page); diff --git a/fs/ntfs/aops.c b/fs/ntfs/aops.c index fa9c05f97af4..d267ea6aa1a0 100644 --- a/fs/ntfs/aops.c +++ b/fs/ntfs/aops.c @@ -1372,7 +1372,7 @@ retry_writepage: * The page may have dirty, unmapped buffers. Make them * freeable here, so the page does not leak. */ - block_invalidatepage(page, 0); + block_invalidatepage(page, 0, PAGE_CACHE_SIZE); unlock_page(page); ntfs_debug("Write outside i_size - truncated?"); return 0; diff --git a/fs/ocfs2/aops.c b/fs/ocfs2/aops.c index 20dfec72e903..79736a28d84f 100644 --- a/fs/ocfs2/aops.c +++ b/fs/ocfs2/aops.c @@ -603,11 +603,12 @@ static void ocfs2_dio_end_io(struct kiocb *iocb, * from ext3. PageChecked() bits have been removed as OCFS2 does not * do journalled data. */ -static void ocfs2_invalidatepage(struct page *page, unsigned long offset) +static void ocfs2_invalidatepage(struct page *page, unsigned int offset, + unsigned int length) { journal_t *journal = OCFS2_SB(page->mapping->host->i_sb)->journal->j_journal; - jbd2_journal_invalidatepage(journal, page, offset); + jbd2_journal_invalidatepage(journal, page, offset, length); } static int ocfs2_releasepage(struct page *page, gfp_t wait) diff --git a/fs/reiserfs/inode.c b/fs/reiserfs/inode.c index f844533792ee..0048cc16a6a8 100644 --- a/fs/reiserfs/inode.c +++ b/fs/reiserfs/inode.c @@ -2975,16 +2975,19 @@ static int invalidatepage_can_drop(struct inode *inode, struct buffer_head *bh) } /* clm -- taken from fs/buffer.c:block_invalidate_page */ -static void reiserfs_invalidatepage(struct page *page, unsigned long offset) +static void reiserfs_invalidatepage(struct page *page, unsigned int offset, + unsigned int length) { struct buffer_head *head, *bh, *next; struct inode *inode = page->mapping->host; unsigned int curr_off = 0; + unsigned int stop = offset + length; + int partial_page = (offset || length < PAGE_CACHE_SIZE); int ret = 1; BUG_ON(!PageLocked(page)); - if (offset == 0) + if (!partial_page) ClearPageChecked(page); if (!page_has_buffers(page)) @@ -2996,6 +2999,9 @@ static void reiserfs_invalidatepage(struct page *page, unsigned long offset) unsigned int next_off = curr_off + bh->b_size; next = bh->b_this_page; + if (next_off > stop) + goto out; + /* * is this block fully invalidated? */ @@ -3014,7 +3020,7 @@ static void reiserfs_invalidatepage(struct page *page, unsigned long offset) * The get_block cached value has been unconditionally invalidated, * so real IO is not possible anymore. */ - if (!offset && ret) { + if (!partial_page && ret) { ret = try_to_release_page(page, 0); /* maybe should BUG_ON(!ret); - neilb */ } diff --git a/fs/splice.c b/fs/splice.c index cc53bd04be8f..3b7ee656f3aa 100644 --- a/fs/splice.c +++ b/fs/splice.c @@ -1269,6 +1269,7 @@ static int direct_splice_actor(struct pipe_inode_info *pipe, * @in: file to splice from * @ppos: input file offset * @out: file to splice to + * @opos: output file offset * @len: number of bytes to splice * @flags: splice modifier flags * diff --git a/fs/sysfs/dir.c b/fs/sysfs/dir.c index 4cfd742d260d..e068e744dbdd 100644 --- a/fs/sysfs/dir.c +++ b/fs/sysfs/dir.c @@ -74,7 +74,7 @@ static int sysfs_sd_compare(const struct sysfs_dirent *left, } /** - * sysfs_link_subling - link sysfs_dirent into sibling rbtree + * sysfs_link_sibling - link sysfs_dirent into sibling rbtree * @sd: sysfs_dirent of interest * * Link @sd into its sibling rbtree which starts from diff --git a/fs/sysfs/file.c b/fs/sysfs/file.c index 602f56db0442..d2bb7ed8fa74 100644 --- a/fs/sysfs/file.c +++ b/fs/sysfs/file.c @@ -449,10 +449,12 @@ void sysfs_notify_dirent(struct sysfs_dirent *sd) spin_lock_irqsave(&sysfs_open_dirent_lock, flags); - od = sd->s_attr.open; - if (od) { - atomic_inc(&od->event); - wake_up_interruptible(&od->poll); + if (!WARN_ON(sysfs_type(sd) != SYSFS_KOBJ_ATTR)) { + od = sd->s_attr.open; + if (od) { + atomic_inc(&od->event); + wake_up_interruptible(&od->poll); + } } spin_unlock_irqrestore(&sysfs_open_dirent_lock, flags); diff --git a/fs/sysfs/inode.c b/fs/sysfs/inode.c index 0ce3ccf7f401..3e2837a633ed 100644 --- a/fs/sysfs/inode.c +++ b/fs/sysfs/inode.c @@ -24,8 +24,6 @@ #include <linux/security.h> #include "sysfs.h" -extern struct super_block * sysfs_sb; - static const struct address_space_operations sysfs_aops = { .readpage = simple_readpage, .write_begin = simple_write_begin, diff --git a/fs/ubifs/file.c b/fs/ubifs/file.c index 14374530784c..123c79b7261e 100644 --- a/fs/ubifs/file.c +++ b/fs/ubifs/file.c @@ -1277,13 +1277,14 @@ int ubifs_setattr(struct dentry *dentry, struct iattr *attr) return err; } -static void ubifs_invalidatepage(struct page *page, unsigned long offset) +static void ubifs_invalidatepage(struct page *page, unsigned int offset, + unsigned int length) { struct inode *inode = page->mapping->host; struct ubifs_info *c = inode->i_sb->s_fs_info; ubifs_assert(PagePrivate(page)); - if (offset) + if (offset || length < PAGE_CACHE_SIZE) /* Partial page remains dirty */ return; diff --git a/fs/xfs/xfs_aops.c b/fs/xfs/xfs_aops.c index 41a695048be7..596ec71da00e 100644 --- a/fs/xfs/xfs_aops.c +++ b/fs/xfs/xfs_aops.c @@ -843,10 +843,12 @@ xfs_cluster_write( STATIC void xfs_vm_invalidatepage( struct page *page, - unsigned long offset) + unsigned int offset, + unsigned int length) { - trace_xfs_invalidatepage(page->mapping->host, page, offset); - block_invalidatepage(page, offset); + trace_xfs_invalidatepage(page->mapping->host, page, offset, + length); + block_invalidatepage(page, offset, length); } /* @@ -910,7 +912,7 @@ next_buffer: xfs_iunlock(ip, XFS_ILOCK_EXCL); out_invalidate: - xfs_vm_invalidatepage(page, 0); + xfs_vm_invalidatepage(page, 0, PAGE_CACHE_SIZE); return; } @@ -940,7 +942,7 @@ xfs_vm_writepage( int count = 0; int nonblocking = 0; - trace_xfs_writepage(inode, page, 0); + trace_xfs_writepage(inode, page, 0, 0); ASSERT(page_has_buffers(page)); @@ -1171,7 +1173,7 @@ xfs_vm_releasepage( { int delalloc, unwritten; - trace_xfs_releasepage(page->mapping->host, page, 0); + trace_xfs_releasepage(page->mapping->host, page, 0, 0); xfs_count_page_state(page, &delalloc, &unwritten); diff --git a/fs/xfs/xfs_trace.h b/fs/xfs/xfs_trace.h index aa4db3307d36..a04701de6bbd 100644 --- a/fs/xfs/xfs_trace.h +++ b/fs/xfs/xfs_trace.h @@ -974,14 +974,16 @@ DEFINE_RW_EVENT(xfs_file_splice_read); DEFINE_RW_EVENT(xfs_file_splice_write); DECLARE_EVENT_CLASS(xfs_page_class, - TP_PROTO(struct inode *inode, struct page *page, unsigned long off), - TP_ARGS(inode, page, off), + TP_PROTO(struct inode *inode, struct page *page, unsigned long off, + unsigned int len), + TP_ARGS(inode, page, off, len), TP_STRUCT__entry( __field(dev_t, dev) __field(xfs_ino_t, ino) __field(pgoff_t, pgoff) __field(loff_t, size) __field(unsigned long, offset) + __field(unsigned int, length) __field(int, delalloc) __field(int, unwritten) ), @@ -995,24 +997,27 @@ DECLARE_EVENT_CLASS(xfs_page_class, __entry->pgoff = page_offset(page); __entry->size = i_size_read(inode); __entry->offset = off; + __entry->length = len; __entry->delalloc = delalloc; __entry->unwritten = unwritten; ), TP_printk("dev %d:%d ino 0x%llx pgoff 0x%lx size 0x%llx offset %lx " - "delalloc %d unwritten %d", + "length %x delalloc %d unwritten %d", MAJOR(__entry->dev), MINOR(__entry->dev), __entry->ino, __entry->pgoff, __entry->size, __entry->offset, + __entry->length, __entry->delalloc, __entry->unwritten) ) #define DEFINE_PAGE_EVENT(name) \ DEFINE_EVENT(xfs_page_class, name, \ - TP_PROTO(struct inode *inode, struct page *page, unsigned long off), \ - TP_ARGS(inode, page, off)) + TP_PROTO(struct inode *inode, struct page *page, unsigned long off, \ + unsigned int len), \ + TP_ARGS(inode, page, off, len)) DEFINE_PAGE_EVENT(xfs_writepage); DEFINE_PAGE_EVENT(xfs_releasepage); DEFINE_PAGE_EVENT(xfs_invalidatepage); |