diff options
Diffstat (limited to 'fs')
49 files changed, 336 insertions, 380 deletions
diff --git a/fs/Kconfig b/fs/Kconfig index d7312825592b..c509123bea49 100644 --- a/fs/Kconfig +++ b/fs/Kconfig @@ -1744,10 +1744,10 @@ config ROOT_NFS If you want your Linux box to mount its whole root file system (the one containing the directory /) from some other computer over the net via NFS (presumably because your box doesn't have a hard disk), - say Y. Read <file:Documentation/nfsroot.txt> for details. It is - likely that in this case, you also want to say Y to "Kernel level IP - autoconfiguration" so that your box can discover its network address - at boot time. + say Y. Read <file:Documentation/filesystems/nfsroot.txt> for + details. It is likely that in this case, you also want to say Y to + "Kernel level IP autoconfiguration" so that your box can discover + its network address at boot time. Most people say N here. diff --git a/fs/afs/cell.c b/fs/afs/cell.c index 970d38f30565..584bb0f9c36a 100644 --- a/fs/afs/cell.c +++ b/fs/afs/cell.c @@ -127,14 +127,21 @@ struct afs_cell *afs_cell_create(const char *name, char *vllist) _enter("%s,%s", name, vllist); + down_write(&afs_cells_sem); + read_lock(&afs_cells_lock); + list_for_each_entry(cell, &afs_cells, link) { + if (strcasecmp(cell->name, name) == 0) + goto duplicate_name; + } + read_unlock(&afs_cells_lock); + cell = afs_cell_alloc(name, vllist); if (IS_ERR(cell)) { _leave(" = %ld", PTR_ERR(cell)); + up_write(&afs_cells_sem); return cell; } - down_write(&afs_cells_sem); - /* add a proc directory for this cell */ ret = afs_proc_cell_setup(cell); if (ret < 0) @@ -167,6 +174,11 @@ error: kfree(cell); _leave(" = %d", ret); return ERR_PTR(ret); + +duplicate_name: + read_unlock(&afs_cells_lock); + up_write(&afs_cells_sem); + return ERR_PTR(-EEXIST); } /* diff --git a/fs/afs/internal.h b/fs/afs/internal.h index 5ca3625cd39e..eec41c76de72 100644 --- a/fs/afs/internal.h +++ b/fs/afs/internal.h @@ -573,7 +573,6 @@ extern const struct file_operations afs_mntpt_file_operations; extern int afs_mntpt_check_symlink(struct afs_vnode *, struct key *); extern void afs_mntpt_kill_timer(void); -extern void afs_umount_begin(struct vfsmount *, int); /* * proc.c @@ -750,7 +749,7 @@ extern int afs_fsync(struct file *, struct dentry *, int); extern unsigned afs_debug; #define dbgprintk(FMT,...) \ - printk("[%x%-6.6s] "FMT"\n", smp_processor_id(), current->comm ,##__VA_ARGS__) + printk("[%-6.6s] "FMT"\n", current->comm ,##__VA_ARGS__) /* make sure we maintain the format strings, even when debugging is disabled */ static inline __attribute__((format(printf,1,2))) diff --git a/fs/afs/mntpt.c b/fs/afs/mntpt.c index a3510b8ba3e7..2f5503902c37 100644 --- a/fs/afs/mntpt.c +++ b/fs/afs/mntpt.c @@ -283,11 +283,3 @@ void afs_mntpt_kill_timer(void) cancel_delayed_work(&afs_mntpt_expiry_timer); flush_scheduled_work(); } - -/* - * begin unmount by attempting to remove all automounted mountpoints we added - */ -void afs_umount_begin(struct vfsmount *vfsmnt, int flags) -{ - shrink_submounts(vfsmnt, &afs_vfsmounts); -} diff --git a/fs/afs/super.c b/fs/afs/super.c index 36bbce45f44b..4b572b801d8d 100644 --- a/fs/afs/super.c +++ b/fs/afs/super.c @@ -50,7 +50,6 @@ static const struct super_operations afs_super_ops = { .write_inode = afs_write_inode, .destroy_inode = afs_destroy_inode, .clear_inode = afs_clear_inode, - .umount_begin = afs_umount_begin, .put_super = afs_put_super, .show_options = generic_show_options, }; @@ -936,14 +936,6 @@ int aio_complete(struct kiocb *iocb, long res, long res2) return 1; } - /* - * Check if the user asked us to deliver the result through an - * eventfd. The eventfd_signal() function is safe to be called - * from IRQ context. - */ - if (!IS_ERR(iocb->ki_eventfd)) - eventfd_signal(iocb->ki_eventfd, 1); - info = &ctx->ring_info; /* add a completion event to the ring buffer. @@ -992,6 +984,15 @@ int aio_complete(struct kiocb *iocb, long res, long res2) kunmap_atomic(ring, KM_IRQ1); pr_debug("added to ring %p at [%lu]\n", iocb, tail); + + /* + * Check if the user asked us to deliver the result through an + * eventfd. The eventfd_signal() function is safe to be called + * from IRQ context. + */ + if (!IS_ERR(iocb->ki_eventfd)) + eventfd_signal(iocb->ki_eventfd, 1); + put_rq: /* everything turned out well, dispose of the aiocb. */ ret = __aio_put_req(ctx, iocb); @@ -1790,6 +1791,7 @@ asmlinkage long sys_io_getevents(aio_context_t ctx_id, put_ioctx(ioctx); } + asmlinkage_protect(5, ret, ctx_id, min_nr, nr, events, timeout); return ret; } diff --git a/fs/buffer.c b/fs/buffer.c index 7ba58386beee..39ff14403d13 100644 --- a/fs/buffer.c +++ b/fs/buffer.c @@ -1181,7 +1181,20 @@ __getblk_slow(struct block_device *bdev, sector_t block, int size) void mark_buffer_dirty(struct buffer_head *bh) { WARN_ON_ONCE(!buffer_uptodate(bh)); - if (!buffer_dirty(bh) && !test_set_buffer_dirty(bh)) + + /* + * Very *carefully* optimize the it-is-already-dirty case. + * + * Don't let the final "is it dirty" escape to before we + * perhaps modified the buffer. + */ + if (buffer_dirty(bh)) { + smp_mb(); + if (buffer_dirty(bh)) + return; + } + + if (!test_set_buffer_dirty(bh)) __set_page_dirty(bh->b_page, page_mapping(bh->b_page), 0); } @@ -2564,14 +2577,13 @@ int nobh_write_end(struct file *file, struct address_space *mapping, struct inode *inode = page->mapping->host; struct buffer_head *head = fsdata; struct buffer_head *bh; + BUG_ON(fsdata != NULL && page_has_buffers(page)); - if (!PageMappedToDisk(page)) { - if (unlikely(copied < len) && !page_has_buffers(page)) - attach_nobh_buffers(page, head); - if (page_has_buffers(page)) - return generic_write_end(file, mapping, pos, len, - copied, page, fsdata); - } + if (unlikely(copied < len) && !page_has_buffers(page)) + attach_nobh_buffers(page, head); + if (page_has_buffers(page)) + return generic_write_end(file, mapping, pos, len, + copied, page, fsdata); SetPageUptodate(page); set_page_dirty(page); diff --git a/fs/cifs/cifs_dfs_ref.c b/fs/cifs/cifs_dfs_ref.c index a1a95b027136..56c924033b78 100644 --- a/fs/cifs/cifs_dfs_ref.c +++ b/fs/cifs/cifs_dfs_ref.c @@ -33,7 +33,6 @@ void dfs_shrink_umount_helper(struct vfsmount *vfsmnt) { mark_mounts_for_expiry(&cifs_dfs_automount_list); mark_mounts_for_expiry(&cifs_dfs_automount_list); - shrink_submounts(vfsmnt, &cifs_dfs_automount_list); } /** diff --git a/fs/cifs/cifsacl.c b/fs/cifs/cifsacl.c index 1f5a4289b848..1cb5b0a9f2ac 100644 --- a/fs/cifs/cifsacl.c +++ b/fs/cifs/cifsacl.c @@ -261,7 +261,7 @@ static void mode_to_access_flags(umode_t mode, umode_t bits_to_use, return; } -static __le16 fill_ace_for_sid(struct cifs_ace *pntace, +static __u16 fill_ace_for_sid(struct cifs_ace *pntace, const struct cifs_sid *psid, __u64 nmode, umode_t bits) { int i; @@ -411,7 +411,7 @@ static void parse_dacl(struct cifs_acl *pdacl, char *end_of_acl, static int set_chmod_dacl(struct cifs_acl *pndacl, struct cifs_sid *pownersid, struct cifs_sid *pgrpsid, __u64 nmode) { - __le16 size = 0; + u16 size = 0; struct cifs_acl *pnndacl; pnndacl = (struct cifs_acl *)((char *)pndacl + sizeof(struct cifs_acl)); diff --git a/fs/efs/inode.c b/fs/efs/inode.c index 79e19e5958e1..a8e7797b9477 100644 --- a/fs/efs/inode.c +++ b/fs/efs/inode.c @@ -140,7 +140,7 @@ struct inode *efs_iget(struct super_block *super, unsigned long ino) brelse(bh); #ifdef DEBUG - printk(KERN_DEBUG "EFS: read_inode(): inode %lu, extents %d, mode %o\n", + printk(KERN_DEBUG "EFS: efs_iget(): inode %lu, extents %d, mode %o\n", inode->i_ino, in->numextents, inode->i_mode); #endif diff --git a/fs/hfsplus/dir.c b/fs/hfsplus/dir.c index 29683645fa0a..5f4023678251 100644 --- a/fs/hfsplus/dir.c +++ b/fs/hfsplus/dir.c @@ -340,16 +340,23 @@ static int hfsplus_unlink(struct inode *dir, struct dentry *dentry) if (inode->i_nlink > 0) drop_nlink(inode); - hfsplus_delete_inode(inode); - if (inode->i_ino != cnid && !inode->i_nlink) { - if (!atomic_read(&HFSPLUS_I(inode).opencnt)) { - res = hfsplus_delete_cat(inode->i_ino, HFSPLUS_SB(sb).hidden_dir, NULL); - if (!res) - hfsplus_delete_inode(inode); + if (inode->i_ino == cnid) + clear_nlink(inode); + if (!inode->i_nlink) { + if (inode->i_ino != cnid) { + HFSPLUS_SB(sb).file_count--; + if (!atomic_read(&HFSPLUS_I(inode).opencnt)) { + res = hfsplus_delete_cat(inode->i_ino, + HFSPLUS_SB(sb).hidden_dir, + NULL); + if (!res) + hfsplus_delete_inode(inode); + } else + inode->i_flags |= S_DEAD; } else - inode->i_flags |= S_DEAD; + hfsplus_delete_inode(inode); } else - clear_nlink(inode); + HFSPLUS_SB(sb).file_count--; inode->i_ctime = CURRENT_TIME_SEC; mark_inode_dirty(inode); diff --git a/fs/jbd/journal.c b/fs/jbd/journal.c index 9816293442ab..0e081d5f32e8 100644 --- a/fs/jbd/journal.c +++ b/fs/jbd/journal.c @@ -1620,14 +1620,14 @@ static int journal_init_journal_head_cache(void) { int retval; - J_ASSERT(journal_head_cache == 0); + J_ASSERT(journal_head_cache == NULL); journal_head_cache = kmem_cache_create("journal_head", sizeof(struct journal_head), 0, /* offset */ SLAB_TEMPORARY, /* flags */ NULL); /* ctor */ retval = 0; - if (journal_head_cache == 0) { + if (!journal_head_cache) { retval = -ENOMEM; printk(KERN_EMERG "JBD: no memory for journal_head cache\n"); } diff --git a/fs/jbd/revoke.c b/fs/jbd/revoke.c index ad2eacf570c6..d5f8eee7c88c 100644 --- a/fs/jbd/revoke.c +++ b/fs/jbd/revoke.c @@ -173,13 +173,13 @@ int __init journal_init_revoke_caches(void) 0, SLAB_HWCACHE_ALIGN|SLAB_TEMPORARY, NULL); - if (revoke_record_cache == 0) + if (!revoke_record_cache) return -ENOMEM; revoke_table_cache = kmem_cache_create("revoke_table", sizeof(struct jbd_revoke_table_s), 0, SLAB_TEMPORARY, NULL); - if (revoke_table_cache == 0) { + if (!revoke_table_cache) { kmem_cache_destroy(revoke_record_cache); revoke_record_cache = NULL; return -ENOMEM; diff --git a/fs/jbd2/journal.c b/fs/jbd2/journal.c index 96ba846992e9..954cff001df6 100644 --- a/fs/jbd2/journal.c +++ b/fs/jbd2/journal.c @@ -219,7 +219,7 @@ static int jbd2_journal_start_thread(journal_t *journal) if (IS_ERR(t)) return PTR_ERR(t); - wait_event(journal->j_wait_done_commit, journal->j_task != 0); + wait_event(journal->j_wait_done_commit, journal->j_task != NULL); return 0; } @@ -231,7 +231,7 @@ static void journal_kill_thread(journal_t *journal) while (journal->j_task) { wake_up(&journal->j_wait_commit); spin_unlock(&journal->j_state_lock); - wait_event(journal->j_wait_done_commit, journal->j_task == 0); + wait_event(journal->j_wait_done_commit, journal->j_task == NULL); spin_lock(&journal->j_state_lock); } spin_unlock(&journal->j_state_lock); @@ -1969,14 +1969,14 @@ static int journal_init_jbd2_journal_head_cache(void) { int retval; - J_ASSERT(jbd2_journal_head_cache == 0); + J_ASSERT(jbd2_journal_head_cache == NULL); jbd2_journal_head_cache = kmem_cache_create("jbd2_journal_head", sizeof(struct journal_head), 0, /* offset */ SLAB_TEMPORARY, /* flags */ NULL); /* ctor */ retval = 0; - if (jbd2_journal_head_cache == 0) { + if (!jbd2_journal_head_cache) { retval = -ENOMEM; printk(KERN_EMERG "JBD: no memory for journal_head cache\n"); } @@ -2002,14 +2002,14 @@ static struct journal_head *journal_alloc_journal_head(void) atomic_inc(&nr_journal_heads); #endif ret = kmem_cache_alloc(jbd2_journal_head_cache, GFP_NOFS); - if (ret == 0) { + if (!ret) { jbd_debug(1, "out of memory for journal_head\n"); if (time_after(jiffies, last_warning + 5*HZ)) { printk(KERN_NOTICE "ENOMEM in %s, retrying.\n", __FUNCTION__); last_warning = jiffies; } - while (ret == 0) { + while (!ret) { yield(); ret = kmem_cache_alloc(jbd2_journal_head_cache, GFP_NOFS); } diff --git a/fs/jbd2/revoke.c b/fs/jbd2/revoke.c index df36f42e19e1..2e1453a5e998 100644 --- a/fs/jbd2/revoke.c +++ b/fs/jbd2/revoke.c @@ -174,13 +174,13 @@ int __init jbd2_journal_init_revoke_caches(void) 0, SLAB_HWCACHE_ALIGN|SLAB_TEMPORARY, NULL); - if (jbd2_revoke_record_cache == 0) + if (!jbd2_revoke_record_cache) return -ENOMEM; jbd2_revoke_table_cache = kmem_cache_create("jbd2_revoke_table", sizeof(struct jbd2_revoke_table_s), 0, SLAB_TEMPORARY, NULL); - if (jbd2_revoke_table_cache == 0) { + if (!jbd2_revoke_table_cache) { kmem_cache_destroy(jbd2_revoke_record_cache); jbd2_revoke_record_cache = NULL; return -ENOMEM; diff --git a/fs/namespace.c b/fs/namespace.c index 7953c96a2071..94f026ec990a 100644 --- a/fs/namespace.c +++ b/fs/namespace.c @@ -155,15 +155,15 @@ static void __touch_mnt_namespace(struct mnt_namespace *ns) } } -static void detach_mnt(struct vfsmount *mnt, struct nameidata *old_nd) +static void detach_mnt(struct vfsmount *mnt, struct path *old_path) { - old_nd->path.dentry = mnt->mnt_mountpoint; - old_nd->path.mnt = mnt->mnt_parent; + old_path->dentry = mnt->mnt_mountpoint; + old_path->mnt = mnt->mnt_parent; mnt->mnt_parent = mnt; mnt->mnt_mountpoint = mnt->mnt_root; list_del_init(&mnt->mnt_child); list_del_init(&mnt->mnt_hash); - old_nd->path.dentry->d_mounted--; + old_path->dentry->d_mounted--; } void mnt_set_mountpoint(struct vfsmount *mnt, struct dentry *dentry, @@ -174,12 +174,12 @@ void mnt_set_mountpoint(struct vfsmount *mnt, struct dentry *dentry, dentry->d_mounted++; } -static void attach_mnt(struct vfsmount *mnt, struct nameidata *nd) +static void attach_mnt(struct vfsmount *mnt, struct path *path) { - mnt_set_mountpoint(nd->path.mnt, nd->path.dentry, mnt); + mnt_set_mountpoint(path->mnt, path->dentry, mnt); list_add_tail(&mnt->mnt_hash, mount_hashtable + - hash(nd->path.mnt, nd->path.dentry)); - list_add_tail(&mnt->mnt_child, &nd->path.mnt->mnt_mounts); + hash(path->mnt, path->dentry)); + list_add_tail(&mnt->mnt_child, &path->mnt->mnt_mounts); } /* @@ -262,10 +262,8 @@ static struct vfsmount *clone_mnt(struct vfsmount *old, struct dentry *root, /* stick the duplicate mount on the same expiry list * as the original if that was on one */ if (flag & CL_EXPIRE) { - spin_lock(&vfsmount_lock); if (!list_empty(&old->mnt_expire)) list_add(&mnt->mnt_expire, &old->mnt_expire); - spin_unlock(&vfsmount_lock); } } return mnt; @@ -548,6 +546,7 @@ void release_mounts(struct list_head *head) m = mnt->mnt_parent; mnt->mnt_mountpoint = mnt->mnt_root; mnt->mnt_parent = mnt; + m->mnt_ghosts--; spin_unlock(&vfsmount_lock); dput(dentry); mntput(m); @@ -572,12 +571,16 @@ void umount_tree(struct vfsmount *mnt, int propagate, struct list_head *kill) __touch_mnt_namespace(p->mnt_ns); p->mnt_ns = NULL; list_del_init(&p->mnt_child); - if (p->mnt_parent != p) + if (p->mnt_parent != p) { + p->mnt_parent->mnt_ghosts++; p->mnt_mountpoint->d_mounted--; + } change_mnt_propagation(p, MS_PRIVATE); } } +static void shrink_submounts(struct vfsmount *mnt, struct list_head *umounts); + static int do_umount(struct vfsmount *mnt, int flags) { struct super_block *sb = mnt->mnt_sb; @@ -650,6 +653,9 @@ static int do_umount(struct vfsmount *mnt, int flags) spin_lock(&vfsmount_lock); event++; + if (!(flags & MNT_DETACH)) + shrink_submounts(mnt, &umount_list); + retval = -EBUSY; if (flags & MNT_DETACH || !propagate_mount_busy(mnt, 2)) { if (!list_empty(&mnt->mnt_list)) @@ -744,7 +750,7 @@ struct vfsmount *copy_tree(struct vfsmount *mnt, struct dentry *dentry, int flag) { struct vfsmount *res, *p, *q, *r, *s; - struct nameidata nd; + struct path path; if (!(flag & CL_COPY_ALL) && IS_MNT_UNBINDABLE(mnt)) return NULL; @@ -769,14 +775,14 @@ struct vfsmount *copy_tree(struct vfsmount *mnt, struct dentry *dentry, q = q->mnt_parent; } p = s; - nd.path.mnt = q; - nd.path.dentry = p->mnt_mountpoint; + path.mnt = q; + path.dentry = p->mnt_mountpoint; q = clone_mnt(p, p->mnt_root, flag); if (!q) goto Enomem; spin_lock(&vfsmount_lock); list_add_tail(&q->mnt_list, &res->mnt_list); - attach_mnt(q, &nd); + attach_mnt(q, &path); spin_unlock(&vfsmount_lock); } } @@ -876,11 +882,11 @@ void drop_collected_mounts(struct vfsmount *mnt) * in allocations. */ static int attach_recursive_mnt(struct vfsmount *source_mnt, - struct nameidata *nd, struct nameidata *parent_nd) + struct path *path, struct path *parent_path) { LIST_HEAD(tree_list); - struct vfsmount *dest_mnt = nd->path.mnt; - struct dentry *dest_dentry = nd->path.dentry; + struct vfsmount *dest_mnt = path->mnt; + struct dentry *dest_dentry = path->dentry; struct vfsmount *child, *p; if (propagate_mnt(dest_mnt, dest_dentry, source_mnt, &tree_list)) @@ -892,9 +898,9 @@ static int attach_recursive_mnt(struct vfsmount *source_mnt, } spin_lock(&vfsmount_lock); - if (parent_nd) { - detach_mnt(source_mnt, parent_nd); - attach_mnt(source_mnt, nd); + if (parent_path) { + detach_mnt(source_mnt, parent_path); + attach_mnt(source_mnt, path); touch_mnt_namespace(current->nsproxy->mnt_ns); } else { mnt_set_mountpoint(dest_mnt, dest_dentry, source_mnt); @@ -930,7 +936,7 @@ static int graft_tree(struct vfsmount *mnt, struct nameidata *nd) err = -ENOENT; if (IS_ROOT(nd->path.dentry) || !d_unhashed(nd->path.dentry)) - err = attach_recursive_mnt(mnt, nd, NULL); + err = attach_recursive_mnt(mnt, &nd->path, NULL); out_unlock: mutex_unlock(&nd->path.dentry->d_inode->i_mutex); if (!err) @@ -1059,7 +1065,8 @@ static inline int tree_contains_unbindable(struct vfsmount *mnt) */ static noinline int do_move_mount(struct nameidata *nd, char *old_name) { - struct nameidata old_nd, parent_nd; + struct nameidata old_nd; + struct path parent_path; struct vfsmount *p; int err = 0; if (!capable(CAP_SYS_ADMIN)) @@ -1114,21 +1121,19 @@ static noinline int do_move_mount(struct nameidata *nd, char *old_name) if (p == old_nd.path.mnt) goto out1; - err = attach_recursive_mnt(old_nd.path.mnt, nd, &parent_nd); + err = attach_recursive_mnt(old_nd.path.mnt, &nd->path, &parent_path); if (err) goto out1; - spin_lock(&vfsmount_lock); /* if the mount is moved, it should no longer be expire * automatically */ list_del_init(&old_nd.path.mnt->mnt_expire); - spin_unlock(&vfsmount_lock); out1: mutex_unlock(&nd->path.dentry->d_inode->i_mutex); out: up_write(&namespace_sem); if (!err) - path_put(&parent_nd.path); + path_put(&parent_path); path_put(&old_nd.path); return err; } @@ -1189,12 +1194,9 @@ int do_add_mount(struct vfsmount *newmnt, struct nameidata *nd, if ((err = graft_tree(newmnt, nd))) goto unlock; - if (fslist) { - /* add to the specified expiration list */ - spin_lock(&vfsmount_lock); + if (fslist) /* add to the specified expiration list */ list_add_tail(&newmnt->mnt_expire, fslist); - spin_unlock(&vfsmount_lock); - } + up_write(&namespace_sem); return 0; @@ -1206,75 +1208,6 @@ unlock: EXPORT_SYMBOL_GPL(do_add_mount); -static void expire_mount(struct vfsmount *mnt, struct list_head *mounts, - struct list_head *umounts) -{ - spin_lock(&vfsmount_lock); - - /* - * Check if mount is still attached, if not, let whoever holds it deal - * with the sucker - */ - if (mnt->mnt_parent == mnt) { - spin_unlock(&vfsmount_lock); - return; - } - - /* - * Check that it is still dead: the count should now be 2 - as - * contributed by the vfsmount parent and the mntget above - */ - if (!propagate_mount_busy(mnt, 2)) { - /* delete from the namespace */ - touch_mnt_namespace(mnt->mnt_ns); - list_del_init(&mnt->mnt_list); - mnt->mnt_ns = NULL; - umount_tree(mnt, 1, umounts); - spin_unlock(&vfsmount_lock); - } else { - /* - * Someone brought it back to life whilst we didn't have any - * locks held so return it to the expiration list - */ - list_add_tail(&mnt->mnt_expire, mounts); - spin_unlock(&vfsmount_lock); - } -} - -/* - * go through the vfsmounts we've just consigned to the graveyard to - * - check that they're still dead - * - delete the vfsmount from the appropriate namespace under lock - * - dispose of the corpse - */ -static void expire_mount_list(struct list_head *graveyard, struct list_head *mounts) -{ - struct mnt_namespace *ns; - struct vfsmount *mnt; - - while (!list_empty(graveyard)) { - LIST_HEAD(umounts); - mnt = list_first_entry(graveyard, struct vfsmount, mnt_expire); - list_del_init(&mnt->mnt_expire); - - /* don't do anything if the namespace is dead - all the - * vfsmounts from it are going away anyway */ - ns = mnt->mnt_ns; - if (!ns || !ns->root) - continue; - get_mnt_ns(ns); - - spin_unlock(&vfsmount_lock); - down_write(&namespace_sem); - expire_mount(mnt, mounts, &umounts); - up_write(&namespace_sem); - release_mounts(&umounts); - mntput(mnt); - put_mnt_ns(ns); - spin_lock(&vfsmount_lock); - } -} - /* * process a list of expirable mountpoints with the intent of discarding any * mountpoints that aren't in use and haven't been touched since last we came @@ -1284,10 +1217,12 @@ void mark_mounts_for_expiry(struct list_head *mounts) { struct vfsmount *mnt, *next; LIST_HEAD(graveyard); + LIST_HEAD(umounts); if (list_empty(mounts)) return; + down_write(&namespace_sem); spin_lock(&vfsmount_lock); /* extract from the expiration list every vfsmount that matches the @@ -1298,16 +1233,19 @@ void mark_mounts_for_expiry(struct list_head *mounts) */ list_for_each_entry_safe(mnt, next, mounts, mnt_expire) { if (!xchg(&mnt->mnt_expiry_mark, 1) || - atomic_read(&mnt->mnt_count) != 1) + propagate_mount_busy(mnt, 1)) continue; - - mntget(mnt); list_move(&mnt->mnt_expire, &graveyard); } - - expire_mount_list(&graveyard, mounts); - + while (!list_empty(&graveyard)) { + mnt = list_first_entry(&graveyard, struct vfsmount, mnt_expire); + touch_mnt_namespace(mnt->mnt_ns); + umount_tree(mnt, 1, &umounts); + } spin_unlock(&vfsmount_lock); + up_write(&namespace_sem); + + release_mounts(&umounts); } EXPORT_SYMBOL_GPL(mark_mounts_for_expiry); @@ -1343,7 +1281,6 @@ resume: } if (!propagate_mount_busy(mnt, 1)) { - mntget(mnt); list_move_tail(&mnt->mnt_expire, graveyard); found++; } @@ -1363,22 +1300,22 @@ resume: * process a list of expirable mountpoints with the intent of discarding any * submounts of a specific parent mountpoint */ -void shrink_submounts(struct vfsmount *mountpoint, struct list_head *mounts) +static void shrink_submounts(struct vfsmount *mnt, struct list_head *umounts) { LIST_HEAD(graveyard); - int found; - - spin_lock(&vfsmount_lock); + struct vfsmount *m; /* extract submounts of 'mountpoint' from the expiration list */ - while ((found = select_submounts(mountpoint, &graveyard)) != 0) - expire_mount_list(&graveyard, mounts); - - spin_unlock(&vfsmount_lock); + while (select_submounts(mnt, &graveyard)) { + while (!list_empty(&graveyard)) { + m = list_first_entry(&graveyard, struct vfsmount, + mnt_expire); + touch_mnt_namespace(mnt->mnt_ns); + umount_tree(mnt, 1, umounts); + } + } } -EXPORT_SYMBOL_GPL(shrink_submounts); - /* * Some copy_from_user() implementations do not return the exact number of * bytes remaining to copy on a fault. But copy_mount_options() requires that. @@ -1683,7 +1620,7 @@ void set_fs_pwd(struct fs_struct *fs, struct path *path) path_put(&old_pwd); } -static void chroot_fs_refs(struct nameidata *old_nd, struct nameidata *new_nd) +static void chroot_fs_refs(struct path *old_root, struct path *new_root) { struct task_struct *g, *p; struct fs_struct *fs; @@ -1695,12 +1632,12 @@ static void chroot_fs_refs(struct nameidata *old_nd, struct nameidata *new_nd) if (fs) { atomic_inc(&fs->count); task_unlock(p); - if (fs->root.dentry == old_nd->path.dentry - && fs->root.mnt == old_nd->path.mnt) - set_fs_root(fs, &new_nd->path); - if (fs->pwd.dentry == old_nd->path.dentry - && fs->pwd.mnt == old_nd->path.mnt) - set_fs_pwd(fs, &new_nd->path); + if (fs->root.dentry == old_root->dentry + && fs->root.mnt == old_root->mnt) + set_fs_root(fs, new_root); + if (fs->pwd.dentry == old_root->dentry + && fs->pwd.mnt == old_root->mnt) + set_fs_pwd(fs, new_root); put_fs_struct(fs); } else task_unlock(p); @@ -1737,7 +1674,8 @@ asmlinkage long sys_pivot_root(const char __user * new_root, const char __user * put_old) { struct vfsmount *tmp; - struct nameidata new_nd, old_nd, parent_nd, root_parent, user_nd; + struct nameidata new_nd, old_nd, user_nd; + struct path parent_path, root_parent; int error; if (!capable(CAP_SYS_ADMIN)) @@ -1811,19 +1749,19 @@ asmlinkage long sys_pivot_root(const char __user * new_root, goto out3; } else if (!is_subdir(old_nd.path.dentry, new_nd.path.dentry)) goto out3; - detach_mnt(new_nd.path.mnt, &parent_nd); + detach_mnt(new_nd.path.mnt, &parent_path); detach_mnt(user_nd.path.mnt, &root_parent); /* mount old root on put_old */ - attach_mnt(user_nd.path.mnt, &old_nd); + attach_mnt(user_nd.path.mnt, &old_nd.path); /* mount new_root on / */ attach_mnt(new_nd.path.mnt, &root_parent); touch_mnt_namespace(current->nsproxy->mnt_ns); spin_unlock(&vfsmount_lock); - chroot_fs_refs(&user_nd, &new_nd); + chroot_fs_refs(&user_nd.path, &new_nd.path); security_sb_post_pivotroot(&user_nd, &new_nd); error = 0; - path_put(&root_parent.path); - path_put(&parent_nd.path); + path_put(&root_parent); + path_put(&parent_path); out2: mutex_unlock(&old_nd.path.dentry->d_inode->i_mutex); up_write(&namespace_sem); diff --git a/fs/nfs/file.c b/fs/nfs/file.c index ef57a5ae5904..5d2e9d9a4e28 100644 --- a/fs/nfs/file.c +++ b/fs/nfs/file.c @@ -64,7 +64,11 @@ const struct file_operations nfs_file_operations = { .write = do_sync_write, .aio_read = nfs_file_read, .aio_write = nfs_file_write, +#ifdef CONFIG_MMU .mmap = nfs_file_mmap, +#else + .mmap = generic_file_mmap, +#endif .open = nfs_file_open, .flush = nfs_file_flush, .release = nfs_file_release, diff --git a/fs/nfs/inode.c b/fs/nfs/inode.c index a4c7cf2bff3a..6f88d7c77ac9 100644 --- a/fs/nfs/inode.c +++ b/fs/nfs/inode.c @@ -506,6 +506,7 @@ static struct nfs_open_context *alloc_nfs_open_context(struct vfsmount *mnt, str ctx->cred = get_rpccred(cred); ctx->state = NULL; ctx->lockowner = current->files; + ctx->flags = 0; ctx->error = 0; ctx->dir_cookie = 0; atomic_set(&ctx->count, 1); diff --git a/fs/nfs/super.c b/fs/nfs/super.c index dd4dfcd632ec..f9219024f31a 100644 --- a/fs/nfs/super.c +++ b/fs/nfs/super.c @@ -589,8 +589,6 @@ static void nfs_umount_begin(struct vfsmount *vfsmnt, int flags) struct nfs_server *server = NFS_SB(vfsmnt->mnt_sb); struct rpc_clnt *rpc; - shrink_submounts(vfsmnt, &nfs_automount_list); - if (!(flags & MNT_FORCE)) return; /* -EIO all pending I/O */ diff --git a/fs/open.c b/fs/open.c index a4b12022edaa..3fa4e4ffce4c 100644 --- a/fs/open.c +++ b/fs/open.c @@ -335,7 +335,7 @@ asmlinkage long sys_ftruncate(unsigned int fd, unsigned long length) { long ret = do_sys_ftruncate(fd, length, 1); /* avoid REGPARM breakage on x86: */ - prevent_tail_call(ret); + asmlinkage_protect(2, ret, fd, length); return ret; } @@ -350,7 +350,7 @@ asmlinkage long sys_ftruncate64(unsigned int fd, loff_t length) { long ret = do_sys_ftruncate(fd, length, 0); /* avoid REGPARM breakage on x86: */ - prevent_tail_call(ret); + asmlinkage_protect(2, ret, fd, length); return ret; } #endif @@ -1067,7 +1067,7 @@ asmlinkage long sys_open(const char __user *filename, int flags, int mode) ret = do_sys_open(AT_FDCWD, filename, flags, mode); /* avoid REGPARM breakage on x86: */ - prevent_tail_call(ret); + asmlinkage_protect(3, ret, filename, flags, mode); return ret; } @@ -1081,7 +1081,7 @@ asmlinkage long sys_openat(int dfd, const char __user *filename, int flags, ret = do_sys_open(dfd, filename, flags, mode); /* avoid REGPARM breakage on x86: */ - prevent_tail_call(ret); + asmlinkage_protect(4, ret, dfd, filename, flags, mode); return ret; } diff --git a/fs/pnode.c b/fs/pnode.c index 05ba692bc540..1d8f5447f3f7 100644 --- a/fs/pnode.c +++ b/fs/pnode.c @@ -225,7 +225,7 @@ out: */ static inline int do_refcount_check(struct vfsmount *mnt, int count) { - int mycount = atomic_read(&mnt->mnt_count); + int mycount = atomic_read(&mnt->mnt_count) - mnt->mnt_ghosts; return (mycount > count); } diff --git a/fs/reiserfs/do_balan.c b/fs/reiserfs/do_balan.c index f85c5cf4934c..7ee4208793b6 100644 --- a/fs/reiserfs/do_balan.c +++ b/fs/reiserfs/do_balan.c @@ -283,7 +283,7 @@ static int balance_leaf(struct tree_balance *tb, struct item_head *ih, /* item h return balance_leaf_when_delete(tb, flag); zeros_num = 0; - if (flag == M_INSERT && body == 0) + if (flag == M_INSERT && !body) zeros_num = ih_item_len(ih); pos_in_item = tb->tb_path->pos_in_item; @@ -1728,7 +1728,7 @@ struct buffer_head *get_FEB(struct tree_balance *tb) struct buffer_info bi; for (i = 0; i < MAX_FEB_SIZE; i++) - if (tb->FEB[i] != 0) + if (tb->FEB[i] != NULL) break; if (i == MAX_FEB_SIZE) @@ -1827,7 +1827,7 @@ int get_left_neighbor_position(struct tree_balance *tb, int h) { int Sh_position = PATH_H_POSITION(tb->tb_path, h + 1); - RFALSE(PATH_H_PPARENT(tb->tb_path, h) == 0 || tb->FL[h] == 0, + RFALSE(PATH_H_PPARENT(tb->tb_path, h) == NULL || tb->FL[h] == NULL, "vs-12325: FL[%d](%p) or F[%d](%p) does not exist", h, tb->FL[h], h, PATH_H_PPARENT(tb->tb_path, h)); @@ -1841,7 +1841,7 @@ int get_right_neighbor_position(struct tree_balance *tb, int h) { int Sh_position = PATH_H_POSITION(tb->tb_path, h + 1); - RFALSE(PATH_H_PPARENT(tb->tb_path, h) == 0 || tb->FR[h] == 0, + RFALSE(PATH_H_PPARENT(tb->tb_path, h) == NULL || tb->FR[h] == NULL, "vs-12330: F[%d](%p) or FR[%d](%p) does not exist", h, PATH_H_PPARENT(tb->tb_path, h), h, tb->FR[h]); diff --git a/fs/reiserfs/fix_node.c b/fs/reiserfs/fix_node.c index 0ee35c6c9b72..07d05e0842b7 100644 --- a/fs/reiserfs/fix_node.c +++ b/fs/reiserfs/fix_node.c @@ -153,7 +153,7 @@ static void create_virtual_node(struct tree_balance *tb, int h) if (vn->vn_mode == M_INSERT) { struct virtual_item *vi = vn->vn_vi + vn->vn_affected_item_num; - RFALSE(vn->vn_ins_ih == 0, + RFALSE(vn->vn_ins_ih == NULL, "vs-8040: item header of inserted item is not specified"); vi->vi_item_len = tb->insert_size[0]; vi->vi_ih = vn->vn_ins_ih; @@ -857,7 +857,8 @@ static int get_lfree(struct tree_balance *tb, int h) struct buffer_head *l, *f; int order; - if ((f = PATH_H_PPARENT(tb->tb_path, h)) == 0 || (l = tb->FL[h]) == 0) + if ((f = PATH_H_PPARENT(tb->tb_path, h)) == NULL || + (l = tb->FL[h]) == NULL) return 0; if (f == l) @@ -878,7 +879,8 @@ static int get_rfree(struct tree_balance *tb, int h) struct buffer_head *r, *f; int order; - if ((f = PATH_H_PPARENT(tb->tb_path, h)) == 0 || (r = tb->FR[h]) == 0) + if ((f = PATH_H_PPARENT(tb->tb_path, h)) == NULL || + (r = tb->FR[h]) == NULL) return 0; if (f == r) diff --git a/fs/reiserfs/lbalance.c b/fs/reiserfs/lbalance.c index 281f8061ac58..6de060a6aa7f 100644 --- a/fs/reiserfs/lbalance.c +++ b/fs/reiserfs/lbalance.c @@ -626,7 +626,7 @@ static void leaf_define_dest_src_infos(int shift_mode, struct tree_balance *tb, "vs-10250: leaf_define_dest_src_infos: shift type is unknown (%d)", shift_mode); } - RFALSE(src_bi->bi_bh == 0 || dest_bi->bi_bh == 0, + RFALSE(!src_bi->bi_bh || !dest_bi->bi_bh, "vs-10260: mode==%d, source (%p) or dest (%p) buffer is initialized incorrectly", shift_mode, src_bi->bi_bh, dest_bi->bi_bh); } diff --git a/fs/reiserfs/namei.c b/fs/reiserfs/namei.c index b378eea332ca..8867533cb727 100644 --- a/fs/reiserfs/namei.c +++ b/fs/reiserfs/namei.c @@ -452,7 +452,7 @@ static int reiserfs_add_entry(struct reiserfs_transaction_handle *th, buflen = DEH_SIZE + ROUND_UP(namelen); if (buflen > sizeof(small_buf)) { buffer = kmalloc(buflen, GFP_NOFS); - if (buffer == 0) + if (!buffer) return -ENOMEM; } else buffer = small_buf; diff --git a/fs/signalfd.c b/fs/signalfd.c index cb2b63ae0bf4..8ead0db35933 100644 --- a/fs/signalfd.c +++ b/fs/signalfd.c @@ -111,9 +111,14 @@ static int signalfd_copyinfo(struct signalfd_siginfo __user *uinfo, err |= __put_user(kinfo->si_uid, &uinfo->ssi_uid); err |= __put_user((long) kinfo->si_ptr, &uinfo->ssi_ptr); break; - default: /* this is just in case for now ... */ + default: + /* + * This case catches also the signals queued by sigqueue(). + */ err |= __put_user(kinfo->si_pid, &uinfo->ssi_pid); err |= __put_user(kinfo->si_uid, &uinfo->ssi_uid); + err |= __put_user((long) kinfo->si_ptr, &uinfo->ssi_ptr); + err |= __put_user(kinfo->si_int, &uinfo->ssi_int); break; } diff --git a/fs/smbfs/smbiod.c b/fs/smbfs/smbiod.c index fae8e85af0ed..6bd9b691a463 100644 --- a/fs/smbfs/smbiod.c +++ b/fs/smbfs/smbiod.c @@ -206,7 +206,7 @@ int smbiod_retry(struct smb_sb_info *server) smb_close_socket(server); - if (pid == 0) { + if (!pid) { /* FIXME: this is fatal, umount? */ printk(KERN_ERR "smb_retry: no connection process\n"); server->state = CONN_RETRIED; diff --git a/fs/splice.c b/fs/splice.c index 0670c915cd35..eeb1a86a7014 100644 --- a/fs/splice.c +++ b/fs/splice.c @@ -320,7 +320,7 @@ __generic_file_splice_read(struct file *in, loff_t *ppos, break; error = add_to_page_cache_lru(page, mapping, index, - GFP_KERNEL); + mapping_gfp_mask(mapping)); if (unlikely(error)) { page_cache_release(page); if (error == -EEXIST) @@ -370,8 +370,10 @@ __generic_file_splice_read(struct file *in, loff_t *ppos, * for an in-flight io page */ if (flags & SPLICE_F_NONBLOCK) { - if (TestSetPageLocked(page)) + if (TestSetPageLocked(page)) { + error = -EAGAIN; break; + } } else lock_page(page); @@ -479,9 +481,8 @@ ssize_t generic_file_splice_read(struct file *in, loff_t *ppos, struct pipe_inode_info *pipe, size_t len, unsigned int flags) { - ssize_t spliced; - int ret; loff_t isize, left; + int ret; isize = i_size_read(in->f_mapping->host); if (unlikely(*ppos >= isize)) @@ -491,29 +492,9 @@ ssize_t generic_file_splice_read(struct file *in, loff_t *ppos, if (unlikely(left < len)) len = left; - ret = 0; - spliced = 0; - while (len && !spliced) { - ret = __generic_file_splice_read(in, ppos, pipe, len, flags); - - if (ret < 0) - break; - else if (!ret) { - if (spliced) - break; - if (flags & SPLICE_F_NONBLOCK) { - ret = -EAGAIN; - break; - } - } - + ret = __generic_file_splice_read(in, ppos, pipe, len, flags); + if (ret > 0) *ppos += ret; - len -= ret; - spliced += ret; - } - - if (spliced) - return spliced; return ret; } diff --git a/fs/xfs/linux-2.6/xfs_ioctl.c b/fs/xfs/linux-2.6/xfs_ioctl.c index a9952e490ac9..f34bd010eb51 100644 --- a/fs/xfs/linux-2.6/xfs_ioctl.c +++ b/fs/xfs/linux-2.6/xfs_ioctl.c @@ -732,7 +732,7 @@ xfs_ioctl( * Only allow the sys admin to reserve space unless * unwritten extents are enabled. */ - if (!XFS_SB_VERSION_HASEXTFLGBIT(&mp->m_sb) && + if (!xfs_sb_version_hasextflgbit(&mp->m_sb) && !capable(CAP_SYS_ADMIN)) return -EPERM; diff --git a/fs/xfs/quota/xfs_qm.c b/fs/xfs/quota/xfs_qm.c index 1f3da5b8657b..8e9c5ae6504d 100644 --- a/fs/xfs/quota/xfs_qm.c +++ b/fs/xfs/quota/xfs_qm.c @@ -1405,13 +1405,13 @@ xfs_qm_qino_alloc( #if defined(DEBUG) && defined(XFS_LOUD_RECOVERY) unsigned oldv = mp->m_sb.sb_versionnum; #endif - ASSERT(!XFS_SB_VERSION_HASQUOTA(&mp->m_sb)); + ASSERT(!xfs_sb_version_hasquota(&mp->m_sb)); ASSERT((sbfields & (XFS_SB_VERSIONNUM | XFS_SB_UQUOTINO | XFS_SB_GQUOTINO | XFS_SB_QFLAGS)) == (XFS_SB_VERSIONNUM | XFS_SB_UQUOTINO | XFS_SB_GQUOTINO | XFS_SB_QFLAGS)); - XFS_SB_VERSION_ADDQUOTA(&mp->m_sb); + xfs_sb_version_addquota(&mp->m_sb); mp->m_sb.sb_uquotino = NULLFSINO; mp->m_sb.sb_gquotino = NULLFSINO; @@ -1954,7 +1954,7 @@ xfs_qm_init_quotainos( /* * Get the uquota and gquota inodes */ - if (XFS_SB_VERSION_HASQUOTA(&mp->m_sb)) { + if (xfs_sb_version_hasquota(&mp->m_sb)) { if (XFS_IS_UQUOTA_ON(mp) && mp->m_sb.sb_uquotino != NULLFSINO) { ASSERT(mp->m_sb.sb_uquotino > 0); diff --git a/fs/xfs/quota/xfs_qm_bhv.c b/fs/xfs/quota/xfs_qm_bhv.c index 97bb32937585..f4f6c4c861d7 100644 --- a/fs/xfs/quota/xfs_qm_bhv.c +++ b/fs/xfs/quota/xfs_qm_bhv.c @@ -118,7 +118,7 @@ xfs_qm_newmount( *quotaflags = 0; *needquotamount = B_FALSE; - quotaondisk = XFS_SB_VERSION_HASQUOTA(&mp->m_sb) && + quotaondisk = xfs_sb_version_hasquota(&mp->m_sb) && (mp->m_sb.sb_qflags & XFS_ALL_QUOTA_ACCT); if (quotaondisk) { diff --git a/fs/xfs/quota/xfs_qm_syscalls.c b/fs/xfs/quota/xfs_qm_syscalls.c index 2cc5886cfe85..d2b8be7e75f9 100644 --- a/fs/xfs/quota/xfs_qm_syscalls.c +++ b/fs/xfs/quota/xfs_qm_syscalls.c @@ -377,7 +377,7 @@ xfs_qm_scall_trunc_qfiles( if (!capable(CAP_SYS_ADMIN)) return XFS_ERROR(EPERM); error = 0; - if (!XFS_SB_VERSION_HASQUOTA(&mp->m_sb) || flags == 0) { + if (!xfs_sb_version_hasquota(&mp->m_sb) || flags == 0) { qdprintk("qtrunc flags=%x m_qflags=%x\n", flags, mp->m_qflags); return XFS_ERROR(EINVAL); } @@ -522,7 +522,7 @@ xfs_qm_scall_getqstat( memset(out, 0, sizeof(fs_quota_stat_t)); out->qs_version = FS_QSTAT_VERSION; - if (! XFS_SB_VERSION_HASQUOTA(&mp->m_sb)) { + if (!xfs_sb_version_hasquota(&mp->m_sb)) { out->qs_uquota.qfs_ino = NULLFSINO; out->qs_gquota.qfs_ino = NULLFSINO; return (0); diff --git a/fs/xfs/xfs_attr_leaf.c b/fs/xfs/xfs_attr_leaf.c index b08e2a2a8add..96ba6aa4ed8c 100644 --- a/fs/xfs/xfs_attr_leaf.c +++ b/fs/xfs/xfs_attr_leaf.c @@ -227,10 +227,10 @@ STATIC void xfs_sbversion_add_attr2(xfs_mount_t *mp, xfs_trans_t *tp) { if ((mp->m_flags & XFS_MOUNT_ATTR2) && - !(XFS_SB_VERSION_HASATTR2(&mp->m_sb))) { + !(xfs_sb_version_hasattr2(&mp->m_sb))) { spin_lock(&mp->m_sb_lock); - if (!XFS_SB_VERSION_HASATTR2(&mp->m_sb)) { - XFS_SB_VERSION_ADDATTR2(&mp->m_sb); + if (!xfs_sb_version_hasattr2(&mp->m_sb)) { + xfs_sb_version_addattr2(&mp->m_sb); spin_unlock(&mp->m_sb_lock); xfs_mod_sb(tp, XFS_SB_VERSIONNUM | XFS_SB_FEATURES2); } else diff --git a/fs/xfs/xfs_bmap.c b/fs/xfs/xfs_bmap.c index 1c0a5a585a82..2def273855a2 100644 --- a/fs/xfs/xfs_bmap.c +++ b/fs/xfs/xfs_bmap.c @@ -4047,17 +4047,17 @@ xfs_bmap_add_attrfork( xfs_trans_log_inode(tp, ip, logflags); if (error) goto error2; - if (!XFS_SB_VERSION_HASATTR(&mp->m_sb) || - (!XFS_SB_VERSION_HASATTR2(&mp->m_sb) && version == 2)) { + if (!xfs_sb_version_hasattr(&mp->m_sb) || + (!xfs_sb_version_hasattr2(&mp->m_sb) && version == 2)) { __int64_t sbfields = 0; spin_lock(&mp->m_sb_lock); - if (!XFS_SB_VERSION_HASATTR(&mp->m_sb)) { - XFS_SB_VERSION_ADDATTR(&mp->m_sb); + if (!xfs_sb_version_hasattr(&mp->m_sb)) { + xfs_sb_version_addattr(&mp->m_sb); sbfields |= XFS_SB_VERSIONNUM; } - if (!XFS_SB_VERSION_HASATTR2(&mp->m_sb) && version == 2) { - XFS_SB_VERSION_ADDATTR2(&mp->m_sb); + if (!xfs_sb_version_hasattr2(&mp->m_sb) && version == 2) { + xfs_sb_version_addattr2(&mp->m_sb); sbfields |= (XFS_SB_VERSIONNUM | XFS_SB_FEATURES2); } if (sbfields) { @@ -5043,7 +5043,7 @@ xfs_bmapi( * A wasdelay extent has been initialized, so * shouldn't be flagged as unwritten. */ - if (wr && XFS_SB_VERSION_HASEXTFLGBIT(&mp->m_sb)) { + if (wr && xfs_sb_version_hasextflgbit(&mp->m_sb)) { if (!wasdelay && (flags & XFS_BMAPI_PREALLOC)) got.br_state = XFS_EXT_UNWRITTEN; } @@ -5483,7 +5483,7 @@ xfs_bunmapi( * get rid of part of a realtime extent. */ if (del.br_state == XFS_EXT_UNWRITTEN || - !XFS_SB_VERSION_HASEXTFLGBIT(&mp->m_sb)) { + !xfs_sb_version_hasextflgbit(&mp->m_sb)) { /* * This piece is unwritten, or we're not * using unwritten extents. Skip over it. @@ -5535,7 +5535,7 @@ xfs_bunmapi( } else if ((del.br_startoff == start && (del.br_state == XFS_EXT_UNWRITTEN || xfs_trans_get_block_res(tp) == 0)) || - !XFS_SB_VERSION_HASEXTFLGBIT(&mp->m_sb)) { + !xfs_sb_version_hasextflgbit(&mp->m_sb)) { /* * Can't make it unwritten. There isn't * a full extent here so just skip it. diff --git a/fs/xfs/xfs_bmap_btree.h b/fs/xfs/xfs_bmap_btree.h index 2d950e975918..cd0d4b4bb816 100644 --- a/fs/xfs/xfs_bmap_btree.h +++ b/fs/xfs/xfs_bmap_btree.h @@ -120,7 +120,7 @@ typedef enum { * Extent state and extent format macros. */ #define XFS_EXTFMT_INODE(x) \ - (XFS_SB_VERSION_HASEXTFLGBIT(&((x)->i_mount->m_sb)) ? \ + (xfs_sb_version_hasextflgbit(&((x)->i_mount->m_sb)) ? \ XFS_EXTFMT_HASSTATE : XFS_EXTFMT_NOSTATE) #define ISUNWRITTEN(x) ((x)->br_state == XFS_EXT_UNWRITTEN) diff --git a/fs/xfs/xfs_dir2.c b/fs/xfs/xfs_dir2.c index be7c4251fa61..e92e73f0e6af 100644 --- a/fs/xfs/xfs_dir2.c +++ b/fs/xfs/xfs_dir2.c @@ -49,7 +49,7 @@ void xfs_dir_mount( xfs_mount_t *mp) { - ASSERT(XFS_SB_VERSION_HASDIRV2(&mp->m_sb)); + ASSERT(xfs_sb_version_hasdirv2(&mp->m_sb)); ASSERT((1 << (mp->m_sb.sb_blocklog + mp->m_sb.sb_dirblklog)) <= XFS_MAX_BLOCKSIZE); mp->m_dirblksize = 1 << (mp->m_sb.sb_blocklog + mp->m_sb.sb_dirblklog); diff --git a/fs/xfs/xfs_fsops.c b/fs/xfs/xfs_fsops.c index eadc1591c795..d3a0f538d6a6 100644 --- a/fs/xfs/xfs_fsops.c +++ b/fs/xfs/xfs_fsops.c @@ -77,36 +77,36 @@ xfs_fs_geometry( if (new_version >= 3) { geo->version = XFS_FSOP_GEOM_VERSION; geo->flags = - (XFS_SB_VERSION_HASATTR(&mp->m_sb) ? + (xfs_sb_version_hasattr(&mp->m_sb) ? XFS_FSOP_GEOM_FLAGS_ATTR : 0) | - (XFS_SB_VERSION_HASNLINK(&mp->m_sb) ? + (xfs_sb_version_hasnlink(&mp->m_sb) ? XFS_FSOP_GEOM_FLAGS_NLINK : 0) | - (XFS_SB_VERSION_HASQUOTA(&mp->m_sb) ? + (xfs_sb_version_hasquota(&mp->m_sb) ? XFS_FSOP_GEOM_FLAGS_QUOTA : 0) | - (XFS_SB_VERSION_HASALIGN(&mp->m_sb) ? + (xfs_sb_version_hasalign(&mp->m_sb) ? XFS_FSOP_GEOM_FLAGS_IALIGN : 0) | - (XFS_SB_VERSION_HASDALIGN(&mp->m_sb) ? + (xfs_sb_version_hasdalign(&mp->m_sb) ? XFS_FSOP_GEOM_FLAGS_DALIGN : 0) | - (XFS_SB_VERSION_HASSHARED(&mp->m_sb) ? + (xfs_sb_version_hasshared(&mp->m_sb) ? XFS_FSOP_GEOM_FLAGS_SHARED : 0) | - (XFS_SB_VERSION_HASEXTFLGBIT(&mp->m_sb) ? + (xfs_sb_version_hasextflgbit(&mp->m_sb) ? XFS_FSOP_GEOM_FLAGS_EXTFLG : 0) | - (XFS_SB_VERSION_HASDIRV2(&mp->m_sb) ? + (xfs_sb_version_hasdirv2(&mp->m_sb) ? XFS_FSOP_GEOM_FLAGS_DIRV2 : 0) | - (XFS_SB_VERSION_HASSECTOR(&mp->m_sb) ? + (xfs_sb_version_hassector(&mp->m_sb) ? XFS_FSOP_GEOM_FLAGS_SECTOR : 0) | (xfs_sb_version_haslazysbcount(&mp->m_sb) ? XFS_FSOP_GEOM_FLAGS_LAZYSB : 0) | - (XFS_SB_VERSION_HASATTR2(&mp->m_sb) ? + (xfs_sb_version_hasattr2(&mp->m_sb) ? XFS_FSOP_GEOM_FLAGS_ATTR2 : 0); - geo->logsectsize = XFS_SB_VERSION_HASSECTOR(&mp->m_sb) ? + geo->logsectsize = xfs_sb_version_hassector(&mp->m_sb) ? mp->m_sb.sb_logsectsize : BBSIZE; geo->rtsectsize = mp->m_sb.sb_blocksize; geo->dirblocksize = mp->m_dirblksize; } if (new_version >= 4) { geo->flags |= - (XFS_SB_VERSION_HASLOGV2(&mp->m_sb) ? + (xfs_sb_version_haslogv2(&mp->m_sb) ? XFS_FSOP_GEOM_FLAGS_LOGV2 : 0); geo->logsunit = mp->m_sb.sb_logsunit; } diff --git a/fs/xfs/xfs_ialloc.c b/fs/xfs/xfs_ialloc.c index db9d5fa600af..5a146cb22980 100644 --- a/fs/xfs/xfs_ialloc.c +++ b/fs/xfs/xfs_ialloc.c @@ -191,7 +191,7 @@ xfs_ialloc_ag_alloc( ASSERT(!(args.mp->m_flags & XFS_MOUNT_NOALIGN)); args.alignment = args.mp->m_dalign; isaligned = 1; - } else if (XFS_SB_VERSION_HASALIGN(&args.mp->m_sb) && + } else if (xfs_sb_version_hasalign(&args.mp->m_sb) && args.mp->m_sb.sb_inoalignmt >= XFS_B_TO_FSBT(args.mp, XFS_INODE_CLUSTER_SIZE(args.mp))) @@ -230,7 +230,7 @@ xfs_ialloc_ag_alloc( args.agbno = be32_to_cpu(agi->agi_root); args.fsbno = XFS_AGB_TO_FSB(args.mp, be32_to_cpu(agi->agi_seqno), args.agbno); - if (XFS_SB_VERSION_HASALIGN(&args.mp->m_sb) && + if (xfs_sb_version_hasalign(&args.mp->m_sb) && args.mp->m_sb.sb_inoalignmt >= XFS_B_TO_FSBT(args.mp, XFS_INODE_CLUSTER_SIZE(args.mp))) args.alignment = args.mp->m_sb.sb_inoalignmt; @@ -271,7 +271,7 @@ xfs_ialloc_ag_alloc( * use the old version so that old kernels will continue to be * able to use the file system. */ - if (XFS_SB_VERSION_HASNLINK(&args.mp->m_sb)) + if (xfs_sb_version_hasnlink(&args.mp->m_sb)) version = XFS_DINODE_VERSION_2; else version = XFS_DINODE_VERSION_1; diff --git a/fs/xfs/xfs_inode.c b/fs/xfs/xfs_inode.c index a550546a7083..f43a6e01d68f 100644 --- a/fs/xfs/xfs_inode.c +++ b/fs/xfs/xfs_inode.c @@ -1147,7 +1147,7 @@ xfs_ialloc( * the inode version number now. This way we only do the conversion * here rather than here and in the flush/logging code. */ - if (XFS_SB_VERSION_HASNLINK(&tp->t_mountp->m_sb) && + if (xfs_sb_version_hasnlink(&tp->t_mountp->m_sb) && ip->i_d.di_version == XFS_DINODE_VERSION_1) { ip->i_d.di_version = XFS_DINODE_VERSION_2; /* @@ -3434,9 +3434,9 @@ xfs_iflush_int( * has been updated, then make the conversion permanent. */ ASSERT(ip->i_d.di_version == XFS_DINODE_VERSION_1 || - XFS_SB_VERSION_HASNLINK(&mp->m_sb)); + xfs_sb_version_hasnlink(&mp->m_sb)); if (ip->i_d.di_version == XFS_DINODE_VERSION_1) { - if (!XFS_SB_VERSION_HASNLINK(&mp->m_sb)) { + if (!xfs_sb_version_hasnlink(&mp->m_sb)) { /* * Convert it back. */ diff --git a/fs/xfs/xfs_inode_item.c b/fs/xfs/xfs_inode_item.c index 034ca7202295..2c775b4ae9e6 100644 --- a/fs/xfs/xfs_inode_item.c +++ b/fs/xfs/xfs_inode_item.c @@ -296,9 +296,9 @@ xfs_inode_item_format( */ mp = ip->i_mount; ASSERT(ip->i_d.di_version == XFS_DINODE_VERSION_1 || - XFS_SB_VERSION_HASNLINK(&mp->m_sb)); + xfs_sb_version_hasnlink(&mp->m_sb)); if (ip->i_d.di_version == XFS_DINODE_VERSION_1) { - if (!XFS_SB_VERSION_HASNLINK(&mp->m_sb)) { + if (!xfs_sb_version_hasnlink(&mp->m_sb)) { /* * Convert it back. */ diff --git a/fs/xfs/xfs_itable.c b/fs/xfs/xfs_itable.c index 658aab6b1bbf..f615e04364f4 100644 --- a/fs/xfs/xfs_itable.c +++ b/fs/xfs/xfs_itable.c @@ -45,7 +45,7 @@ xfs_internal_inum( xfs_ino_t ino) { return (ino == mp->m_sb.sb_rbmino || ino == mp->m_sb.sb_rsumino || - (XFS_SB_VERSION_HASQUOTA(&mp->m_sb) && + (xfs_sb_version_hasquota(&mp->m_sb) && (ino == mp->m_sb.sb_uquotino || ino == mp->m_sb.sb_gquotino))); } diff --git a/fs/xfs/xfs_log.c b/fs/xfs/xfs_log.c index a75edca1860f..31f2b04f2c97 100644 --- a/fs/xfs/xfs_log.c +++ b/fs/xfs/xfs_log.c @@ -1090,7 +1090,7 @@ xlog_get_iclog_buffer_size(xfs_mount_t *mp, size >>= 1; } - if (XFS_SB_VERSION_HASLOGV2(&mp->m_sb)) { + if (xfs_sb_version_haslogv2(&mp->m_sb)) { /* # headers = size / 32K * one header holds cycles from 32K of data */ @@ -1186,13 +1186,13 @@ xlog_alloc_log(xfs_mount_t *mp, log->l_grant_reserve_cycle = 1; log->l_grant_write_cycle = 1; - if (XFS_SB_VERSION_HASSECTOR(&mp->m_sb)) { + if (xfs_sb_version_hassector(&mp->m_sb)) { log->l_sectbb_log = mp->m_sb.sb_logsectlog - BBSHIFT; ASSERT(log->l_sectbb_log <= mp->m_sectbb_log); /* for larger sector sizes, must have v2 or external log */ ASSERT(log->l_sectbb_log == 0 || log->l_logBBstart == 0 || - XFS_SB_VERSION_HASLOGV2(&mp->m_sb)); + xfs_sb_version_haslogv2(&mp->m_sb)); ASSERT(mp->m_sb.sb_logsectlog >= BBSHIFT); } log->l_sectbb_mask = (1 << log->l_sectbb_log) - 1; @@ -1247,7 +1247,7 @@ xlog_alloc_log(xfs_mount_t *mp, memset(head, 0, sizeof(xlog_rec_header_t)); head->h_magicno = cpu_to_be32(XLOG_HEADER_MAGIC_NUM); head->h_version = cpu_to_be32( - XFS_SB_VERSION_HASLOGV2(&log->l_mp->m_sb) ? 2 : 1); + xfs_sb_version_haslogv2(&log->l_mp->m_sb) ? 2 : 1); head->h_size = cpu_to_be32(log->l_iclog_size); /* new fields */ head->h_fmt = cpu_to_be32(XLOG_FMT); @@ -1402,7 +1402,7 @@ xlog_sync(xlog_t *log, int roundoff; /* roundoff to BB or stripe */ int split = 0; /* split write into two regions */ int error; - int v2 = XFS_SB_VERSION_HASLOGV2(&log->l_mp->m_sb); + int v2 = xfs_sb_version_haslogv2(&log->l_mp->m_sb); XFS_STATS_INC(xs_log_writes); ASSERT(iclog->ic_refcnt == 0); @@ -2881,7 +2881,7 @@ xlog_state_switch_iclogs(xlog_t *log, log->l_curr_block += BTOBB(eventual_size)+BTOBB(log->l_iclog_hsize); /* Round up to next log-sunit */ - if (XFS_SB_VERSION_HASLOGV2(&log->l_mp->m_sb) && + if (xfs_sb_version_haslogv2(&log->l_mp->m_sb) && log->l_mp->m_sb.sb_logsunit > 1) { __uint32_t sunit_bb = BTOBB(log->l_mp->m_sb.sb_logsunit); log->l_curr_block = roundup(log->l_curr_block, sunit_bb); @@ -3334,7 +3334,7 @@ xlog_ticket_get(xlog_t *log, unit_bytes += sizeof(xlog_op_header_t) * num_headers; /* for roundoff padding for transaction data and one for commit record */ - if (XFS_SB_VERSION_HASLOGV2(&log->l_mp->m_sb) && + if (xfs_sb_version_haslogv2(&log->l_mp->m_sb) && log->l_mp->m_sb.sb_logsunit > 1) { /* log su roundoff */ unit_bytes += 2*log->l_mp->m_sb.sb_logsunit; diff --git a/fs/xfs/xfs_log_priv.h b/fs/xfs/xfs_log_priv.h index e008233ee249..c6244cc733c0 100644 --- a/fs/xfs/xfs_log_priv.h +++ b/fs/xfs/xfs_log_priv.h @@ -49,10 +49,10 @@ struct xfs_mount; #define XLOG_HEADER_SIZE 512 #define XLOG_REC_SHIFT(log) \ - BTOBB(1 << (XFS_SB_VERSION_HASLOGV2(&log->l_mp->m_sb) ? \ + BTOBB(1 << (xfs_sb_version_haslogv2(&log->l_mp->m_sb) ? \ XLOG_MAX_RECORD_BSHIFT : XLOG_BIG_RECORD_BSHIFT)) #define XLOG_TOTAL_REC_SHIFT(log) \ - BTOBB(XLOG_MAX_ICLOGS << (XFS_SB_VERSION_HASLOGV2(&log->l_mp->m_sb) ? \ + BTOBB(XLOG_MAX_ICLOGS << (xfs_sb_version_haslogv2(&log->l_mp->m_sb) ? \ XLOG_MAX_RECORD_BSHIFT : XLOG_BIG_RECORD_BSHIFT)) diff --git a/fs/xfs/xfs_log_recover.c b/fs/xfs/xfs_log_recover.c index b82d5d4d2462..b2b70eba282c 100644 --- a/fs/xfs/xfs_log_recover.c +++ b/fs/xfs/xfs_log_recover.c @@ -478,7 +478,7 @@ xlog_find_verify_log_record( * reset last_blk. Only when last_blk points in the middle of a log * record do we update last_blk. */ - if (XFS_SB_VERSION_HASLOGV2(&log->l_mp->m_sb)) { + if (xfs_sb_version_haslogv2(&log->l_mp->m_sb)) { uint h_size = be32_to_cpu(head->h_size); xhdrs = h_size / XLOG_HEADER_CYCLE_SIZE; @@ -888,7 +888,7 @@ xlog_find_tail( * unmount record if there is one, so we pass the lsn of the * unmount record rather than the block after it. */ - if (XFS_SB_VERSION_HASLOGV2(&log->l_mp->m_sb)) { + if (xfs_sb_version_haslogv2(&log->l_mp->m_sb)) { int h_size = be32_to_cpu(rhead->h_size); int h_version = be32_to_cpu(rhead->h_version); @@ -1101,7 +1101,7 @@ xlog_add_record( recp->h_magicno = cpu_to_be32(XLOG_HEADER_MAGIC_NUM); recp->h_cycle = cpu_to_be32(cycle); recp->h_version = cpu_to_be32( - XFS_SB_VERSION_HASLOGV2(&log->l_mp->m_sb) ? 2 : 1); + xfs_sb_version_haslogv2(&log->l_mp->m_sb) ? 2 : 1); recp->h_lsn = cpu_to_be64(xlog_assign_lsn(cycle, block)); recp->h_tail_lsn = cpu_to_be64(xlog_assign_lsn(tail_cycle, tail_block)); recp->h_fmt = cpu_to_be32(XLOG_FMT); @@ -3348,7 +3348,7 @@ xlog_pack_data( dp += BBSIZE; } - if (XFS_SB_VERSION_HASLOGV2(&log->l_mp->m_sb)) { + if (xfs_sb_version_haslogv2(&log->l_mp->m_sb)) { xhdr = (xlog_in_core_2_t *)&iclog->ic_header; for ( ; i < BTOBB(size); i++) { j = i / (XLOG_HEADER_CYCLE_SIZE / BBSIZE); @@ -3388,7 +3388,7 @@ xlog_unpack_data_checksum( be32_to_cpu(rhead->h_chksum), chksum); cmn_err(CE_DEBUG, "XFS: Disregard message if filesystem was created with non-DEBUG kernel"); - if (XFS_SB_VERSION_HASLOGV2(&log->l_mp->m_sb)) { + if (xfs_sb_version_haslogv2(&log->l_mp->m_sb)) { cmn_err(CE_DEBUG, "XFS: LogR this is a LogV2 filesystem\n"); } @@ -3415,7 +3415,7 @@ xlog_unpack_data( dp += BBSIZE; } - if (XFS_SB_VERSION_HASLOGV2(&log->l_mp->m_sb)) { + if (xfs_sb_version_haslogv2(&log->l_mp->m_sb)) { xhdr = (xlog_in_core_2_t *)rhead; for ( ; i < BTOBB(be32_to_cpu(rhead->h_len)); i++) { j = i / (XLOG_HEADER_CYCLE_SIZE / BBSIZE); @@ -3494,7 +3494,7 @@ xlog_do_recovery_pass( * Read the header of the tail block and get the iclog buffer size from * h_size. Use this to tell how many sectors make up the log header. */ - if (XFS_SB_VERSION_HASLOGV2(&log->l_mp->m_sb)) { + if (xfs_sb_version_haslogv2(&log->l_mp->m_sb)) { /* * When using variable length iclogs, read first sector of * iclog header and extract the header size from it. Get a @@ -3838,7 +3838,7 @@ xlog_do_recover( sbp = &log->l_mp->m_sb; xfs_sb_from_disk(sbp, XFS_BUF_TO_SBP(bp)); ASSERT(sbp->sb_magicnum == XFS_SB_MAGIC); - ASSERT(XFS_SB_GOOD_VERSION(sbp)); + ASSERT(xfs_sb_good_version(sbp)); xfs_buf_relse(bp); /* We've re-read the superblock so re-initialize per-cpu counters */ diff --git a/fs/xfs/xfs_mount.c b/fs/xfs/xfs_mount.c index 6409b3762995..8ed164eb9544 100644 --- a/fs/xfs/xfs_mount.c +++ b/fs/xfs/xfs_mount.c @@ -44,7 +44,7 @@ #include "xfs_quota.h" #include "xfs_fsops.h" -STATIC void xfs_mount_log_sbunit(xfs_mount_t *, __int64_t); +STATIC void xfs_mount_log_sb(xfs_mount_t *, __int64_t); STATIC int xfs_uuid_mount(xfs_mount_t *); STATIC void xfs_uuid_unmount(xfs_mount_t *mp); STATIC void xfs_unmountfs_wait(xfs_mount_t *); @@ -119,6 +119,7 @@ static const struct { { offsetof(xfs_sb_t, sb_logsectsize),0 }, { offsetof(xfs_sb_t, sb_logsunit), 0 }, { offsetof(xfs_sb_t, sb_features2), 0 }, + { offsetof(xfs_sb_t, sb_bad_features2), 0 }, { sizeof(xfs_sb_t), 0 } }; @@ -225,7 +226,7 @@ xfs_mount_validate_sb( return XFS_ERROR(EWRONGFS); } - if (!XFS_SB_GOOD_VERSION(sbp)) { + if (!xfs_sb_good_version(sbp)) { xfs_fs_mount_cmn_err(flags, "bad version"); return XFS_ERROR(EWRONGFS); } @@ -300,7 +301,7 @@ xfs_mount_validate_sb( /* * Version 1 directory format has never worked on Linux. */ - if (unlikely(!XFS_SB_VERSION_HASDIRV2(sbp))) { + if (unlikely(!xfs_sb_version_hasdirv2(sbp))) { xfs_fs_mount_cmn_err(flags, "file system using version 1 directory format"); return XFS_ERROR(ENOSYS); @@ -449,6 +450,7 @@ xfs_sb_from_disk( to->sb_logsectsize = be16_to_cpu(from->sb_logsectsize); to->sb_logsunit = be32_to_cpu(from->sb_logsunit); to->sb_features2 = be32_to_cpu(from->sb_features2); + to->sb_bad_features2 = be32_to_cpu(from->sb_bad_features2); } /* @@ -781,7 +783,7 @@ xfs_update_alignment(xfs_mount_t *mp, int mfsi_flags, __uint64_t *update_flags) * Update superblock with new values * and log changes */ - if (XFS_SB_VERSION_HASDALIGN(sbp)) { + if (xfs_sb_version_hasdalign(sbp)) { if (sbp->sb_unit != mp->m_dalign) { sbp->sb_unit = mp->m_dalign; *update_flags |= XFS_SB_UNIT; @@ -792,7 +794,7 @@ xfs_update_alignment(xfs_mount_t *mp, int mfsi_flags, __uint64_t *update_flags) } } } else if ((mp->m_flags & XFS_MOUNT_NOALIGN) != XFS_MOUNT_NOALIGN && - XFS_SB_VERSION_HASDALIGN(&mp->m_sb)) { + xfs_sb_version_hasdalign(&mp->m_sb)) { mp->m_dalign = sbp->sb_unit; mp->m_swidth = sbp->sb_width; } @@ -869,7 +871,7 @@ xfs_set_rw_sizes(xfs_mount_t *mp) STATIC void xfs_set_inoalignment(xfs_mount_t *mp) { - if (XFS_SB_VERSION_HASALIGN(&mp->m_sb) && + if (xfs_sb_version_hasalign(&mp->m_sb) && mp->m_sb.sb_inoalignmt >= XFS_B_TO_FSBT(mp, mp->m_inode_cluster_size)) mp->m_inoalign_mask = mp->m_sb.sb_inoalignmt - 1; @@ -970,6 +972,38 @@ xfs_mountfs( xfs_mount_common(mp, sbp); /* + * Check for a mismatched features2 values. Older kernels + * read & wrote into the wrong sb offset for sb_features2 + * on some platforms due to xfs_sb_t not being 64bit size aligned + * when sb_features2 was added, which made older superblock + * reading/writing routines swap it as a 64-bit value. + * + * For backwards compatibility, we make both slots equal. + * + * If we detect a mismatched field, we OR the set bits into the + * existing features2 field in case it has already been modified; we + * don't want to lose any features. We then update the bad location + * with the ORed value so that older kernels will see any features2 + * flags, and mark the two fields as needing updates once the + * transaction subsystem is online. + */ + if (xfs_sb_has_mismatched_features2(sbp)) { + cmn_err(CE_WARN, + "XFS: correcting sb_features alignment problem"); + sbp->sb_features2 |= sbp->sb_bad_features2; + sbp->sb_bad_features2 = sbp->sb_features2; + update_flags |= XFS_SB_FEATURES2 | XFS_SB_BAD_FEATURES2; + + /* + * Re-check for ATTR2 in case it was found in bad_features2 + * slot. + */ + if (xfs_sb_version_hasattr2(&mp->m_sb)) + mp->m_flags |= XFS_MOUNT_ATTR2; + + } + + /* * Check if sb_agblocks is aligned at stripe boundary * If sb_agblocks is NOT aligned turn off m_dalign since * allocator alignment is within an ag, therefore ag has @@ -1159,11 +1193,10 @@ xfs_mountfs( } /* - * If fs is not mounted readonly, then update the superblock - * unit and width changes. + * If fs is not mounted readonly, then update the superblock changes. */ if (update_flags && !(mp->m_flags & XFS_MOUNT_RDONLY)) - xfs_mount_log_sbunit(mp, update_flags); + xfs_mount_log_sb(mp, update_flags); /* * Initialise the XFS quota management subsystem for this mount @@ -1875,16 +1908,18 @@ xfs_uuid_unmount( /* * Used to log changes to the superblock unit and width fields which could - * be altered by the mount options. Only the first superblock is updated. + * be altered by the mount options, as well as any potential sb_features2 + * fixup. Only the first superblock is updated. */ STATIC void -xfs_mount_log_sbunit( +xfs_mount_log_sb( xfs_mount_t *mp, __int64_t fields) { xfs_trans_t *tp; - ASSERT(fields & (XFS_SB_UNIT|XFS_SB_WIDTH|XFS_SB_UUID)); + ASSERT(fields & (XFS_SB_UNIT | XFS_SB_WIDTH | XFS_SB_UUID | + XFS_SB_FEATURES2 | XFS_SB_BAD_FEATURES2)); tp = xfs_trans_alloc(mp, XFS_TRANS_SB_UNIT); if (xfs_trans_reserve(tp, 0, mp->m_sb.sb_sectsize + 128, 0, 0, diff --git a/fs/xfs/xfs_sb.h b/fs/xfs/xfs_sb.h index 94660b1a6ccc..d904efe7f871 100644 --- a/fs/xfs/xfs_sb.h +++ b/fs/xfs/xfs_sb.h @@ -89,6 +89,7 @@ struct xfs_mount; /* * Superblock - in core version. Must match the ondisk version below. + * Must be padded to 64 bit alignment. */ typedef struct xfs_sb { __uint32_t sb_magicnum; /* magic number == XFS_SB_MAGIC */ @@ -145,10 +146,21 @@ typedef struct xfs_sb { __uint16_t sb_logsectsize; /* sector size for the log, bytes */ __uint32_t sb_logsunit; /* stripe unit size for the log */ __uint32_t sb_features2; /* additional feature bits */ + + /* + * bad features2 field as a result of failing to pad the sb + * structure to 64 bits. Some machines will be using this field + * for features2 bits. Easiest just to mark it bad and not use + * it for anything else. + */ + __uint32_t sb_bad_features2; + + /* must be padded to 64 bit alignment */ } xfs_sb_t; /* - * Superblock - on disk version. Must match the in core version below. + * Superblock - on disk version. Must match the in core version above. + * Must be padded to 64 bit alignment. */ typedef struct xfs_dsb { __be32 sb_magicnum; /* magic number == XFS_SB_MAGIC */ @@ -205,6 +217,15 @@ typedef struct xfs_dsb { __be16 sb_logsectsize; /* sector size for the log, bytes */ __be32 sb_logsunit; /* stripe unit size for the log */ __be32 sb_features2; /* additional feature bits */ + /* + * bad features2 field as a result of failing to pad the sb + * structure to 64 bits. Some machines will be using this field + * for features2 bits. Easiest just to mark it bad and not use + * it for anything else. + */ + __be32 sb_bad_features2; + + /* must be padded to 64 bit alignment */ } xfs_dsb_t; /* @@ -223,7 +244,7 @@ typedef enum { XFS_SBS_GQUOTINO, XFS_SBS_QFLAGS, XFS_SBS_FLAGS, XFS_SBS_SHARED_VN, XFS_SBS_INOALIGNMT, XFS_SBS_UNIT, XFS_SBS_WIDTH, XFS_SBS_DIRBLKLOG, XFS_SBS_LOGSECTLOG, XFS_SBS_LOGSECTSIZE, XFS_SBS_LOGSUNIT, - XFS_SBS_FEATURES2, + XFS_SBS_FEATURES2, XFS_SBS_BAD_FEATURES2, XFS_SBS_FIELDCOUNT } xfs_sb_field_t; @@ -248,13 +269,15 @@ typedef enum { #define XFS_SB_IFREE XFS_SB_MVAL(IFREE) #define XFS_SB_FDBLOCKS XFS_SB_MVAL(FDBLOCKS) #define XFS_SB_FEATURES2 XFS_SB_MVAL(FEATURES2) +#define XFS_SB_BAD_FEATURES2 XFS_SB_MVAL(BAD_FEATURES2) #define XFS_SB_NUM_BITS ((int)XFS_SBS_FIELDCOUNT) #define XFS_SB_ALL_BITS ((1LL << XFS_SB_NUM_BITS) - 1) #define XFS_SB_MOD_BITS \ (XFS_SB_UUID | XFS_SB_ROOTINO | XFS_SB_RBMINO | XFS_SB_RSUMINO | \ XFS_SB_VERSIONNUM | XFS_SB_UQUOTINO | XFS_SB_GQUOTINO | \ XFS_SB_QFLAGS | XFS_SB_SHARED_VN | XFS_SB_UNIT | XFS_SB_WIDTH | \ - XFS_SB_ICOUNT | XFS_SB_IFREE | XFS_SB_FDBLOCKS | XFS_SB_FEATURES2) + XFS_SB_ICOUNT | XFS_SB_IFREE | XFS_SB_FDBLOCKS | XFS_SB_FEATURES2 | \ + XFS_SB_BAD_FEATURES2) /* @@ -271,7 +294,6 @@ typedef enum { #define XFS_SB_VERSION_NUM(sbp) ((sbp)->sb_versionnum & XFS_SB_VERSION_NUMBITS) -#define XFS_SB_GOOD_VERSION(sbp) xfs_sb_good_version(sbp) #ifdef __KERNEL__ static inline int xfs_sb_good_version(xfs_sb_t *sbp) { @@ -297,7 +319,15 @@ static inline int xfs_sb_good_version(xfs_sb_t *sbp) } #endif /* __KERNEL__ */ -#define XFS_SB_VERSION_TONEW(v) xfs_sb_version_tonew(v) +/* + * Detect a mismatched features2 field. Older kernels read/wrote + * this into the wrong slot, so to be safe we keep them in sync. + */ +static inline int xfs_sb_has_mismatched_features2(xfs_sb_t *sbp) +{ + return (sbp->sb_bad_features2 != sbp->sb_features2); +} + static inline unsigned xfs_sb_version_tonew(unsigned v) { return ((((v) == XFS_SB_VERSION_1) ? \ @@ -308,7 +338,6 @@ static inline unsigned xfs_sb_version_tonew(unsigned v) XFS_SB_VERSION_4); } -#define XFS_SB_VERSION_TOOLD(v) xfs_sb_version_toold(v) static inline unsigned xfs_sb_version_toold(unsigned v) { return (((v) & (XFS_SB_VERSION_QUOTABIT | XFS_SB_VERSION_ALIGNBIT)) ? \ @@ -320,7 +349,6 @@ static inline unsigned xfs_sb_version_toold(unsigned v) XFS_SB_VERSION_1))); } -#define XFS_SB_VERSION_HASATTR(sbp) xfs_sb_version_hasattr(sbp) static inline int xfs_sb_version_hasattr(xfs_sb_t *sbp) { return ((sbp)->sb_versionnum == XFS_SB_VERSION_2) || \ @@ -329,7 +357,6 @@ static inline int xfs_sb_version_hasattr(xfs_sb_t *sbp) ((sbp)->sb_versionnum & XFS_SB_VERSION_ATTRBIT)); } -#define XFS_SB_VERSION_ADDATTR(sbp) xfs_sb_version_addattr(sbp) static inline void xfs_sb_version_addattr(xfs_sb_t *sbp) { (sbp)->sb_versionnum = (((sbp)->sb_versionnum == XFS_SB_VERSION_1) ? \ @@ -339,7 +366,6 @@ static inline void xfs_sb_version_addattr(xfs_sb_t *sbp) (XFS_SB_VERSION_4 | XFS_SB_VERSION_ATTRBIT))); } -#define XFS_SB_VERSION_HASNLINK(sbp) xfs_sb_version_hasnlink(sbp) static inline int xfs_sb_version_hasnlink(xfs_sb_t *sbp) { return ((sbp)->sb_versionnum == XFS_SB_VERSION_3) || \ @@ -347,7 +373,6 @@ static inline int xfs_sb_version_hasnlink(xfs_sb_t *sbp) ((sbp)->sb_versionnum & XFS_SB_VERSION_NLINKBIT)); } -#define XFS_SB_VERSION_ADDNLINK(sbp) xfs_sb_version_addnlink(sbp) static inline void xfs_sb_version_addnlink(xfs_sb_t *sbp) { (sbp)->sb_versionnum = ((sbp)->sb_versionnum <= XFS_SB_VERSION_2 ? \ @@ -355,115 +380,63 @@ static inline void xfs_sb_version_addnlink(xfs_sb_t *sbp) ((sbp)->sb_versionnum | XFS_SB_VERSION_NLINKBIT)); } -#define XFS_SB_VERSION_HASQUOTA(sbp) xfs_sb_version_hasquota(sbp) static inline int xfs_sb_version_hasquota(xfs_sb_t *sbp) { return (XFS_SB_VERSION_NUM(sbp) == XFS_SB_VERSION_4) && \ ((sbp)->sb_versionnum & XFS_SB_VERSION_QUOTABIT); } -#define XFS_SB_VERSION_ADDQUOTA(sbp) xfs_sb_version_addquota(sbp) static inline void xfs_sb_version_addquota(xfs_sb_t *sbp) { (sbp)->sb_versionnum = \ (XFS_SB_VERSION_NUM(sbp) == XFS_SB_VERSION_4 ? \ ((sbp)->sb_versionnum | XFS_SB_VERSION_QUOTABIT) : \ - (XFS_SB_VERSION_TONEW((sbp)->sb_versionnum) | \ + (xfs_sb_version_tonew((sbp)->sb_versionnum) | \ XFS_SB_VERSION_QUOTABIT)); } -#define XFS_SB_VERSION_HASALIGN(sbp) xfs_sb_version_hasalign(sbp) static inline int xfs_sb_version_hasalign(xfs_sb_t *sbp) { return (XFS_SB_VERSION_NUM(sbp) == XFS_SB_VERSION_4) && \ ((sbp)->sb_versionnum & XFS_SB_VERSION_ALIGNBIT); } -#define XFS_SB_VERSION_SUBALIGN(sbp) xfs_sb_version_subalign(sbp) -static inline void xfs_sb_version_subalign(xfs_sb_t *sbp) -{ - (sbp)->sb_versionnum = \ - XFS_SB_VERSION_TOOLD((sbp)->sb_versionnum & ~XFS_SB_VERSION_ALIGNBIT); -} - -#define XFS_SB_VERSION_HASDALIGN(sbp) xfs_sb_version_hasdalign(sbp) static inline int xfs_sb_version_hasdalign(xfs_sb_t *sbp) { return (XFS_SB_VERSION_NUM(sbp) == XFS_SB_VERSION_4) && \ ((sbp)->sb_versionnum & XFS_SB_VERSION_DALIGNBIT); } -#define XFS_SB_VERSION_ADDDALIGN(sbp) xfs_sb_version_adddalign(sbp) -static inline int xfs_sb_version_adddalign(xfs_sb_t *sbp) -{ - return (sbp)->sb_versionnum = \ - ((sbp)->sb_versionnum | XFS_SB_VERSION_DALIGNBIT); -} - -#define XFS_SB_VERSION_HASSHARED(sbp) xfs_sb_version_hasshared(sbp) static inline int xfs_sb_version_hasshared(xfs_sb_t *sbp) { return (XFS_SB_VERSION_NUM(sbp) == XFS_SB_VERSION_4) && \ ((sbp)->sb_versionnum & XFS_SB_VERSION_SHAREDBIT); } -#define XFS_SB_VERSION_ADDSHARED(sbp) xfs_sb_version_addshared(sbp) -static inline int xfs_sb_version_addshared(xfs_sb_t *sbp) -{ - return (sbp)->sb_versionnum = \ - ((sbp)->sb_versionnum | XFS_SB_VERSION_SHAREDBIT); -} - -#define XFS_SB_VERSION_SUBSHARED(sbp) xfs_sb_version_subshared(sbp) -static inline int xfs_sb_version_subshared(xfs_sb_t *sbp) -{ - return (sbp)->sb_versionnum = \ - ((sbp)->sb_versionnum & ~XFS_SB_VERSION_SHAREDBIT); -} - -#define XFS_SB_VERSION_HASDIRV2(sbp) xfs_sb_version_hasdirv2(sbp) static inline int xfs_sb_version_hasdirv2(xfs_sb_t *sbp) { return (XFS_SB_VERSION_NUM(sbp) == XFS_SB_VERSION_4) && \ ((sbp)->sb_versionnum & XFS_SB_VERSION_DIRV2BIT); } -#define XFS_SB_VERSION_HASLOGV2(sbp) xfs_sb_version_haslogv2(sbp) static inline int xfs_sb_version_haslogv2(xfs_sb_t *sbp) { return (XFS_SB_VERSION_NUM(sbp) == XFS_SB_VERSION_4) && \ ((sbp)->sb_versionnum & XFS_SB_VERSION_LOGV2BIT); } -#define XFS_SB_VERSION_HASEXTFLGBIT(sbp) xfs_sb_version_hasextflgbit(sbp) static inline int xfs_sb_version_hasextflgbit(xfs_sb_t *sbp) { return (XFS_SB_VERSION_NUM(sbp) == XFS_SB_VERSION_4) && \ ((sbp)->sb_versionnum & XFS_SB_VERSION_EXTFLGBIT); } -#define XFS_SB_VERSION_ADDEXTFLGBIT(sbp) xfs_sb_version_addextflgbit(sbp) -static inline int xfs_sb_version_addextflgbit(xfs_sb_t *sbp) -{ - return (sbp)->sb_versionnum = \ - ((sbp)->sb_versionnum | XFS_SB_VERSION_EXTFLGBIT); -} - -#define XFS_SB_VERSION_SUBEXTFLGBIT(sbp) xfs_sb_version_subextflgbit(sbp) -static inline int xfs_sb_version_subextflgbit(xfs_sb_t *sbp) -{ - return (sbp)->sb_versionnum = \ - ((sbp)->sb_versionnum & ~XFS_SB_VERSION_EXTFLGBIT); -} - -#define XFS_SB_VERSION_HASSECTOR(sbp) xfs_sb_version_hassector(sbp) static inline int xfs_sb_version_hassector(xfs_sb_t *sbp) { return (XFS_SB_VERSION_NUM(sbp) == XFS_SB_VERSION_4) && \ ((sbp)->sb_versionnum & XFS_SB_VERSION_SECTORBIT); } -#define XFS_SB_VERSION_HASMOREBITS(sbp) xfs_sb_version_hasmorebits(sbp) static inline int xfs_sb_version_hasmorebits(xfs_sb_t *sbp) { return (XFS_SB_VERSION_NUM(sbp) == XFS_SB_VERSION_4) && \ @@ -476,24 +449,22 @@ static inline int xfs_sb_version_hasmorebits(xfs_sb_t *sbp) * For example, for a bit defined as XFS_SB_VERSION2_FUNBIT, has a macro: * * SB_VERSION_HASFUNBIT(xfs_sb_t *sbp) - * ((XFS_SB_VERSION_HASMOREBITS(sbp) && + * ((xfs_sb_version_hasmorebits(sbp) && * ((sbp)->sb_features2 & XFS_SB_VERSION2_FUNBIT) */ static inline int xfs_sb_version_haslazysbcount(xfs_sb_t *sbp) { - return (XFS_SB_VERSION_HASMOREBITS(sbp) && \ + return (xfs_sb_version_hasmorebits(sbp) && \ ((sbp)->sb_features2 & XFS_SB_VERSION2_LAZYSBCOUNTBIT)); } -#define XFS_SB_VERSION_HASATTR2(sbp) xfs_sb_version_hasattr2(sbp) static inline int xfs_sb_version_hasattr2(xfs_sb_t *sbp) { - return (XFS_SB_VERSION_HASMOREBITS(sbp)) && \ + return (xfs_sb_version_hasmorebits(sbp)) && \ ((sbp)->sb_features2 & XFS_SB_VERSION2_ATTR2BIT); } -#define XFS_SB_VERSION_ADDATTR2(sbp) xfs_sb_version_addattr2(sbp) static inline void xfs_sb_version_addattr2(xfs_sb_t *sbp) { ((sbp)->sb_versionnum = \ diff --git a/fs/xfs/xfs_utils.c b/fs/xfs/xfs_utils.c index 45d740df53b7..18a85e746680 100644 --- a/fs/xfs/xfs_utils.c +++ b/fs/xfs/xfs_utils.c @@ -339,10 +339,10 @@ xfs_bump_ino_vers2( ip->i_d.di_onlink = 0; memset(&(ip->i_d.di_pad[0]), 0, sizeof(ip->i_d.di_pad)); mp = tp->t_mountp; - if (!XFS_SB_VERSION_HASNLINK(&mp->m_sb)) { + if (!xfs_sb_version_hasnlink(&mp->m_sb)) { spin_lock(&mp->m_sb_lock); - if (!XFS_SB_VERSION_HASNLINK(&mp->m_sb)) { - XFS_SB_VERSION_ADDNLINK(&mp->m_sb); + if (!xfs_sb_version_hasnlink(&mp->m_sb)) { + xfs_sb_version_addnlink(&mp->m_sb); spin_unlock(&mp->m_sb_lock); xfs_mod_sb(tp, XFS_SB_VERSIONNUM); } else { diff --git a/fs/xfs/xfs_vfsops.c b/fs/xfs/xfs_vfsops.c index 7321304a69cc..7094caff13cf 100644 --- a/fs/xfs/xfs_vfsops.c +++ b/fs/xfs/xfs_vfsops.c @@ -330,7 +330,7 @@ xfs_finish_flags( int ronly = (mp->m_flags & XFS_MOUNT_RDONLY); /* Fail a mount where the logbuf is smaller then the log stripe */ - if (XFS_SB_VERSION_HASLOGV2(&mp->m_sb)) { + if (xfs_sb_version_haslogv2(&mp->m_sb)) { if ((ap->logbufsize <= 0) && (mp->m_sb.sb_logsunit > XLOG_BIG_RECORD_BSIZE)) { mp->m_logbsize = mp->m_sb.sb_logsunit; @@ -349,9 +349,8 @@ xfs_finish_flags( } } - if (XFS_SB_VERSION_HASATTR2(&mp->m_sb)) { + if (xfs_sb_version_hasattr2(&mp->m_sb)) mp->m_flags |= XFS_MOUNT_ATTR2; - } /* * prohibit r/w mounts of read-only filesystems @@ -366,7 +365,7 @@ xfs_finish_flags( * check for shared mount. */ if (ap->flags & XFSMNT_SHARED) { - if (!XFS_SB_VERSION_HASSHARED(&mp->m_sb)) + if (!xfs_sb_version_hasshared(&mp->m_sb)) return XFS_ERROR(EINVAL); /* @@ -512,7 +511,7 @@ xfs_mount( if (!error && logdev && logdev != ddev) { unsigned int log_sector_size = BBSIZE; - if (XFS_SB_VERSION_HASSECTOR(&mp->m_sb)) + if (xfs_sb_version_hassector(&mp->m_sb)) log_sector_size = mp->m_sb.sb_logsectsize; error = xfs_setsize_buftarg(mp->m_logdev_targp, mp->m_sb.sb_blocksize, diff --git a/fs/xfs/xfs_vnodeops.c b/fs/xfs/xfs_vnodeops.c index 51305242ff8c..64c5953feca4 100644 --- a/fs/xfs/xfs_vnodeops.c +++ b/fs/xfs/xfs_vnodeops.c @@ -4132,7 +4132,7 @@ xfs_free_file_space( * actually need to zero the extent edges. Otherwise xfs_bunmapi * will take care of it for us. */ - if (rt && !XFS_SB_VERSION_HASEXTFLGBIT(&mp->m_sb)) { + if (rt && !xfs_sb_version_hasextflgbit(&mp->m_sb)) { nimap = 1; error = xfs_bmapi(NULL, ip, startoffset_fsb, 1, 0, NULL, 0, &imap, &nimap, NULL, NULL); |